Biology of Metabolism in Growing Animals

Biology of Metabolismin Growing Animals

Edited by

D.G. BurrinUSDA /ARS Children’s Nutrition Research Center, Department of Pediatrics,

Baylor College of Medicine, Houston, Texas, USA

H.J. MersmannUSDA/ARS Children’s Nutrition Research Center, Department of Pediatrics,

Baylor College of Medicine, Houston, Texas, USA

Technical Editor

E. SalekThe Kielanowski Institute of Animal Physiology and Nutrition,

Polish Academy of Sciences, Jablonna n/Warsaw, Poland

Edinburg London New York Oxford PhiladelphiaSt. Louis Sydney Toronto 2005

Elsevier Limited

© 2005 Elsevier Limited. All rights reserved.

No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form orby any means, electronic, mechanical, photocopying, recording or otherwise, without either the prior per-mission of the publishers or a licence permitting restricted copying in the United Kingdom issued by theCopyright Licensing Agency, 90 Tottenham Court Road, London W1T 4LP. Permissions may be soughtdirectly from Elsevier’s Health Sciences Rights Department in Philadelphia, USA: phone: (+1) 215 2387869, fax: (+1) 215 238 2239, e-mail: [email protected]. You may also complete yourrequest on-line via the Elsevier homepage (http://www.elsevier.com), by selecting ‘Customer Support’and then ‘Obtaining Permissions’.

First published 2005

ISBN 0 444 510133

British Library Cataloguing in Publication DataA catalogue record for this book is available from the British Library

Library of Congress Cataloging in Publication DataA catalog record for this book is available from the Library of Congress

Notice Veterinary knowledge and best practice in this field are constantly changing. As new research and experi-ence broaden our knowledge, changes in practice, treatment and drug therapy may become necessary orappropriate. Readers are advised to check the most current information provided (i) on procedures fea-tured or (ii) by the manufacturer of each product to be administered, to verify the recommended dose orformula, the method and duration of administration, and contraindications. It is the responsiblity of thepractitioner, relying on their own experience and knowldege of the patient, to make diagnoses, to deter-mine dosages and the best treatment for each individual patient, and to take all appropriate safety precau-tions. To the fullest extent of the law, neither the publisher nor the editors assumes any liability for anyinjury and/or damage.

The Publisher

Printed in China The Publisher’s

policy is to usepaper manufactured

from sustainable forests

v

Keynotes

Progress in life sciences is unbelievably quick and usually unpredictable. The amount ofresearch results communicated each minute, every day of the week makes it impossible to beup-to-date even in a very narrow scientific field. The situation as regards the transfer of theseachievements to lecture halls and their integration with current “practical” scientific knowl-edge is even worse. The gap between the latest developments in life sciences announced by theworld’s leading labs and the possibilities of their verification in medicine, biomedicine, andanimal production seems to be expanding at a geometrical rate. At the same time “more andless” is known. It appears that the professional scientific world has run into difficulties in inte-grating what the scientific world knows. Soon, the old Scandinavian adage “the top consultantsknow everything about nothing” will be a truism.

This series of books prepared by leading professionals will try to fill the gap between practi-cal and basic knowledge in life sciences. We believe that the authors and their selections of theinformation presented in their chapters will still leave room for young animals to grow.

Stepan Pierzynowski, ProfSeries Editor

INSTITUTIONS PROVIDING PATRONAGE AND FINANCIAL SUPPORT

USDA/ARS Children’s NutritionResearch Center, Department ofPediatrics, Baylor College of Medicine,Houston, Texas, USA

MS Milk Specialties Company

CIL Cambridge Isotope Laboratories, Inc.

ISOTEC Member of the SIGMA-ALDRICH Family

Lund University, Sweden

The Kielanowski Institute of AnimalPhysiology and Nutrition, PolishAcademy of Sciences, Poland

SGP Consulting, Lund Sweden

Gramineer International AB, Lund,Sweden

vii

Preface

This book Biology of Metabolism in Growing Animals is the third volume in the Elsevierbook series entitled Biology of Growing Animals. This book is intended to provide in-depthreviews of the major areas of metabolism in growing domestic animals. The authors areleading, internationally recognized experts in the fields of nutrition, metabolism, and physi-ology and highlight some of the most recent advances in the field of metabolism. The chap-ters cover important new developments in interorgan, tissue-specific, and cell-specificmetabolism of protein and amino acids, lipids and fatty acids and carbohydrates in mono-gastric and ruminant species, including humans. The study of metabolism represents a nexusof biological phenomena that integrates the nutrition, physiology, endocrinology, immunol-ogy, biochemistry and cell biology in an organism. The development of new methodologicaltechniques and experimental approaches has provided scientists with a greater understandingof how key nutrients or substrates are metabolized at the cellular, organ and whole animallevel. The book describes the impact of specific biochemical pathways and expression ofcritical enzymes, routes of nutrient or substrate input and anatomical or structural influenceson the rates of metabolism in a given tissue or cell type. Major substrates/fuels for oxidativemetabolism, key endocrine signaling pathways and intracellular molecules that regulate themajor metabolic processes are described. Also discussed is the influence of ontogeny, stageof differentiation and major changes in diet, or the environment, on metabolism of growinganimals. The concepts and specific findings in each area are discussed in the context of theirimpact on the nutrient requirements, growth, environmental impact, health and well-being of animals.

Acknowledgements

The editors wish to thank all of the authors for their outstanding contributions to the book.We also thank Ewa Salek for her assistance with technical editing and Jane Schoppe foradministrative support. Thanks also go to the Series Editors, Stefan Pierzynowski andRomuald Zabielski, for the invitation and opportunity to put together this book. We sin-cerely thank the sponsors for their financial support, including USDA/ARS, Milk SpecialtiesCompany, Cambridge Isotope Laboratories, and Sigma-Aldrich-Isotec Inc.

D.G. Burrin and H.J. MersmannEditors

ix

Dedication

The editors and many contributing authors of the book wish to dedicate this book to the mem-ory of Dr. Peter Reeds. Peter Reeds was a close colleague, friend and mentor to many of thecontributing authors of this book. Peter Reeds was born in England in 1945 and completed hisPh.D. in nutritional biochemistry at the University of Southampton, in 1971. His doctoralresearch focused on the interactions between insulin and growth hormone in the regulation ofmuscle protein synthesis and demonstrated the synergy between their separate mechanisms ofaction. Peter Reeds went on to complete postdoctoral training at the Tropical MetabolismResearch Unit in Jamaica under the mentorship of Professor John Waterlow. His early yearsof training provided a foundation in key areas that would be central themes in his career,namely protein metabolism, isotope kinetics and growth regulation. In 1976, Peter Reedsmoved to the Rowett Research Institute in Aberdeen, Scotland, to work under the guidance ofthe Director, Sir Kenneth Blaxter. During his years at the Rowett, Peter Reeds establishedhimself as a leader in the science of growth regulation, protein metabolism and the nutrientrequirements of farm livestock. In 1987, Peter Reeds moved to the Children’s NutritionResearch Center in the Department of Pediatrics at Baylor College of Medicine, where heresumed his longstanding interests in human pediatric nutrition and developmental aspects ofgrowth. In 2001, Peter Reeds left the Children’s Nutrition Research Center to assume a posi-tion as Professor of Animal Sciences in the Faculty Excellence Program at the University ofIllinois at Urbana-Champaign.

During his career, Peter Reeds made many seminal contributions to our understanding ofprotein and amino acid metabolism and the biology of growth regulation. His intellectual brilliance was evident in the breadth and volume of his work. More importantly, however,

Peter J. Reeds

Peter Reeds was a wonderful human being with an irrepressible wit and sense of humor. Hissense of humor was reflected in his exuberance and excitement for science, which was infec-tious to those with whom he worked. Peter Reeds died on August 13, 2002, from complica-tions of Legionnaire’s disease. His legacy to the science of nutrition and metabolism will belong remembered by his countless friends, colleagues and members of the nutrition sciencecommunity.

Dedicationx

xi

Contributors

Ball R.O. – Department of Agricultural, Food and Nutritional Sciences, University of Alberta, Edmonton, Alberta, Canada T6G 2P5; The Research Institute, TheHospital for Sick Children, Toronto, Department of Nutritional Sciences, Universityof Toronto, Toronto, Ontoria, Canada

Baracos V.E. – Department of Oncology, University of Alberta, Edmonton, Alberta,Canada T6G1Z2

Bell A.W. – Department of Animal Science, Cornell University, Ithaca, NY14853–4801, USA

Bertolo R.F.P. – Department of Biochemistry, Memorial University of Newfoundland,St. John’s, Newfoundland, Canada, A1B 3X9

Burrin D.G. – USDA/ARS Children’s Nutrition Research Center, Department ofPediatrics, Baylor College of Medicine, Houston, TX 77030, USA

Carstens G.E. – Department of Animal Science, Texas A&M University, CollegeStation, TX 77483–2471, USA

Damon M. – INRA, Joint Research Unit for Calf and Pig Production, 35590 SaintGilles, France

Davis T.A. – United States Department of Agriculture/Agricultural ResearchService, Children's Nutrition Research Center, Department of Pediatrics, BaylorCollege of Medicine, Houston, TX 77030, USA

Donkin S.S. – Department of Animal Sciences, Purdue University, West Lafayette,IN 47907, USA

Drackley J.K. – Department of Animal Sciences, University of Illinois, Urbana, IL 61801, USA

Ehrhardt R.A. – Department of Animal Science, Cornell University, Ithaca, NY14853–4801, USA

Escobar J. – Department of Animal Sciences, University of Illinois, Urbana,IL61801, USA

Fiorotto M.L. – United States Department of Agriculture/Agricultural ResearchService, Children's Nutrition Research Center, Department of Pediatrics, BaylorCollege of Medicine, Houston, TX 77030, USA

Flynn N.E. – Department of Chemistry and Biochemistry, Angelo State University,San Angelo, TX 76909, USA

Greenwood P.L. – NSW Agriculture Beef Industry Centre, University of NewEngland, Armidale, NSW 2351, Australia

Guan X. – USDA/ARS Children’s Nutrition Research Center, Department ofPediatrics, Baylor College of Medicine, Houston, TX 77030, USA

Hammon H. – Research Institute for Biology of Farm Animals (Oskar KellnerInstitute), 18196 Dummerstorf, Germany

Harmon D.L. – Department of Animal Sciences, University of Kentucky,Lexington, KY 40546-0215, USA

Herpin P. – INRA, Joint Research Unit for Calf and Pig Production, 35590, Saint-Gilles, France

Huntington G.B. – Department of Animal Science, North Carolina StateUniversity, Raleigh, NC 27695-7621, USA

Innis S.M. – Department of Paediatrics, University of British Columbia, Vancouver,British Columbia, Canada, V5Z 4H4

Jesse B.W. – Department of Animal Science, Rutgers, The State University of New Jersey, New Brunswick, NJ 08901-8525, USA

Johnson R.W. – Department of Animal Sciences, University of Illinois, Urbana, IL61801, USA

Knabe D.A. – Department of Animal Science and Faculty of Nutrition, Texas A & MUniversity, College Station, TX 77843-2471, USA

Kristensen N.B. – Department of Animal Nutrition and Physiology, DanishInstitute of Agricultural Sciences, DK-8830 Tjele, Denmark

Le Dividich J. – INRA, Joint Research Unit for Calf and Pig Production, 35590Saint-Gilles, France

Lin X. – Department of Animal Science, North Carolina State University, Raleigh,NC 27695-7621, USA

Louveau I. – INRA, Joint Research Unit for Calf and Pig Production, 35590 Saint-Gilles, France

Lyvers-Peffer P. – Department of Animal Science, North Carolina State University,Raleigh, NC 27695-7621, USA

Mersmann H.J. – USDA/ARS Children’s Nutrition Research Center, Departmentof Pediatrics, Baylor College of Medicine, Houston, TX 77030, USA.

Odle J. – Department of Animal Science, North Carolina State University, Raleigh,NC 27695-7621, USA

Pencharz P.B. – Department of Paediatrics, University of Toronto, Toronto, Ontario,Canada M5G 1X8; The Research Institute, The Hospital for Sick Children, Toronto,Department of Nutritional Sciences, University of Toronto, Toronto, Ontaria, Canada

Reynolds C.K. – Department of Animal Sciences, The Ohio State University,OARDC, 1680 Madison Avenue, Wooster, OH 44691-4096, USA

Smith S.B. – Department of Animal Science, Texas A & M University, CollegeStation, TX 77843-2471, USA

Stoll B. – USDA/ARS Children’s Nutrition Research Center, Department ofPediatrics, Baylor College of Medicine, Houston, TX 77030, USA

Wu G. – Department of Animal Science and Faculty of Nutrition, Texas A & MUniversity, College Station, TX 77843-2471, USA

Contributorsxii

3

1 Regulation of metabolism and growthduring prenatal life

A. W. Bella, P. L. Greenwoodb, and R. A. Ehrhardta

aDepartment of Animal Science, Cornell University, Ithaca, NY 14853-4801, USAbNSW Agriculture Beef Industry Centre, University of New England, Armidale,NSW 2351, Australia

Fetal energy and nitrogen requirements are met mostly by placental transfer of glucose andamino acids; fatty acids may contribute additional energy in some species. Placental metab-olism accounts for much of the total net consumption of oxygen and macronutrients by theconceptus, and alters the composition of nutrients delivered to the fetus. The molecular basisfor the facilitated transport of glucose by the placenta is well described; molecular character-ization of the more complex systems for the active transport of most amino acids is underway. Maternal and placental macronutrient supply is a powerful regulator of fetal metabolismand growth, especially in late gestation. Endocrine mediation of these responses maturesas gestation advances, adding to the influences of locally expressed regulators throughoutgestation. Insulin, thyroid hormones, and, near term, corticosteroids, are especially influentialin the direct and indirect control of fetal nutrient disposal and tissue growth. Prenatal growthretardation does not necessarily constrain the rate of neonatal growth, but at any given post-natal body weight, low-birth-weight lambs are fatter and have smaller muscles. Experimentalevidence is accumulating for longer-term influences of prenatal nutrition through fetal pro-gramming of propensity for mature-onset diseases such as hypertension and type II diabetes.

1. INTRODUCTION

The coordination of nutrient supply with tissue metabolism and growth during prenatal life inplacental mammals is complex due to the varying influences of maternal nutrition and meta-bolic adaptations to the state of pregnancy, placental function, and gestational maturationof fetal endocrine and local regulatory systems. It is important to understand the separateand interdependent mechanisms by which these factors exert their effects on fetal growthand development, for several reasons. Increased neonatal mortality and morbidity in low-birth-weight offspring remain major problems in some human and livestock populations, despitedecades of study on the multifaceted etiology of intrauterine growth retardation (IUGR).

Biology of Metabolism in Growing AnimalsD.G. Burrin and H. Mersmann (Eds.)


Fetal overgrowth due to maternal nutrition or diseases, such as diabetes, also increases perina-tal mortality and incidence of postnatal problems. More intriguing and, possibly, with majorramifications for long-term health and productivity of humans and other animals, is theemerging evidence that fetal metabolic disturbance can lead to “programming” of increasedpredisposition to various disease syndromes during later postnatal life.

This chapter will summarize briefly the quite detailed state of knowledge of quantitativemetabolism of macronutrients in individual tissues and whole body of the fetus, and in theplacenta, with emphasis on data obtained in vivo. The current understanding of placentaltransport of macronutrients and its implications for fetal nutrition and growth will be treatedsimilarly. These topics will be a prelude to the major theme of regulation and coordination ofmetabolism and growth in the conceptus. Finally, the influence of prenatal experienceon postnatal performance will be considered, with brief reference to recent experimentalevidence for the concept of “fetal programming”.

2. MAJOR FEATURES OF CONCEPTUS METABOLISM AND GROWTH

2.1. Patterns of prenatal growth

Early embryonic development, including organogenesis and initiation of placentation, is beyondthe scope of this review. The morphology of embryo development in domestic animal specieshas been described by Noden and deLahunta (1985). Patterns of fetal and placental growth inthe normal and growth-retarded sheep conceptus are illustrated in fig. 1. In this species, as in

A. W. Bell et al.4

Fig. 1. Patterns of fetal and placental growth in the normal (——) and growth-retarded (---) sheep conceptus.Adapted from the data of Ehrhardt and Bell (1995) and Greenwood et al. (2000).

other placental mammals, postembryonic growth becomes quantitatively significant onlyafter mid-gestation. However, this is preceded by rapid hyperplastic growth of the placenta,which attains all or most of its mass of dry tissue, protein, and DNA by mid-gestation(Ehrhardt and Bell, 1995). Fetal growth then follows its familiar, flattened sigmoid patternduring the latter half of gestation as it proceeds from an early exponential phase through arapid, linear phase, and then, as term approaches, begins to diminish in rate. In most species,there is little or no increase in placental weight during this period; the ovine placenta actuallydiminishes in weight, mostly due to loss of extracellular water (Ehrhardt and Bell, 1995).However, the placenta undergoes extensive tissue remodeling after mid-gestation, includingmajor proliferative growth of the umbilical vasculature (Teasdale, 1976), which is associatedwith a progressive increase in its functional capacity. Relations between placental size andfunction, and implications for fetal growth, are discussed in the next section.

2.2. Fetal requirements and metabolism of macronutrients

Numerous studies on pregnant ewes have described fetal macronutrient requirements andmetabolism in terms of umbilical exchanges of oxygen, nutrients, and metabolites, and ofrates of net accretion of nutrients in growing tissues (see Battaglia and Meschia, 1988; Bell,1993). These and similar data from pregnant cows (Comline and Silver, 1976; Reynolds et al.,1986; Ferrell, 1991) are summarized in table 1.

During late pregnancy in these species, 35–40% of fetal energy is taken up as glucose and itsfetal-placental metabolite, lactate, and a further 55% is taken up as free amino acids. In contrastto its importance as an energy source in the maternal ruminant, umbilical uptake of acetate couldaccount for only 5–10% of fetal energy consumption. Placental capacity for transfer of long-chain, nonesterified fatty acids (NEFA) and keto-acids is even more limited (see Bell, 1993),making these maternal substrates trivial contributors to fetal metabolism. Almost all of thenitrogen acquired by the fetus is in the form of amino acids, but a small net umbilical uptakeof ammonia is derived from placental deamination of amino acids during the latter half of

Regulation of metabolism and growth during prenatal life 5

Table 1

Fetal sources and disposal of energy and nitrogen in ewes and cows during late pregnancy

Energy (kJ/kg·d) Nitrogen (g/kg·d)

Ewe Cow Ewe Cow

SourcesGlucose + lactate 217a 114f — —Amino acids 177a 156g 1.19a 1.09g

Acetate 20b 30h — —NH3 — — 0.05e NDTotal 414 300 1.24 1.09

DisposalAccretion 133c 72i 0.79c 0.34i

Heat 240a 192g — —Urea 16d 15g 0.36d 0.66g

Glutamate + serine efflux 16a ND 0.11a NDTotal 405 279 1.26 1.00

a Chung et al. (1998), b Char and Creasy (1976), c McNeill et al. (1997), d Lemons and Schreiner (1983),e Holzman et al. (1977), f Reynolds et al. (1986), g Ferrell (1991), h Comline and Silver (1976), i Ferrell et al. (1976).

gestation (Holzman et al., 1977; Bell et al., 1989). About 60% of these amino acids are used fortissue protein synthesis, which accounts for ~18% of fetal energy expenditure (Kennaugh et al.,1987). The remaining 40% are rapidly catabolized, accounting for at least 30% of the oxidativerequirements in the well-nourished sheep fetus (Faichney and White, 1987), or, in the case ofglutamate and serine, taken up and metabolized by the placenta (Battaglia and Regnault, 2001).

Less comprehensive studies of the fetal pig (Fowden et al., 1997) and horse (Fowden andSilver, 1995) suggest that in these species during late pregnancy, glucose is an even moreimportant energy substrate than in fetal ruminants. The fetal horse, at least, appears to make less extensive use of amino acids as a source of energy (Silver et al., 1994; Fowden et al., 2000a).

In all species studied, the fetal liver and, to a lesser extent, kidneys, develop the enzymaticcapacity for gluconeogenesis during late gestation (see Fowden, 1997). In the well-fed,unstressed sheep fetus, endogenous glucose synthesis is negligible (Hay et al., 1984; Leuryet al., 1990a). However, significant endogenous synthesis of glucose can be induced bymaternal starvation or chronic undernutrition, presumably due to hepatic gluconeogenesisfrom amino acids (Hay et al., 1984; Leury et al., 1990a). Acute hypoxia and other stressorsalso increase net hepatic release of glucose due to increased rates of gluconeogenesis and/orglycogenolysis in fetal sheep (Rudolph et al., 1989; Townsend et al., 1991).

2.3. Metabolism of nonfetal conceptus tissues

2.3.1. Glucose metabolism

The major contribution of the nonfetal components of the gravid uterus, especially the placenta,to oxygen and nutrient requirements of the conceptus is sometimes ignored. However, theserequirements greatly affect the partitioning of nutrients within the gravid uterus and add sub-stantially to the nutrient demands upon the dam. In late-pregnant ewes and cows, the aggregateweight of placentomes, consisting of fetal (cotyledonary) and maternal (caruncular) tissues, isless than 15% that of the attached fetus. However, the weight-specific metabolic rate of the pla-centa is so great that the uteroplacental tissues (placentomes, endometrium, myometrium)consume 35–50% of the oxygen and 60–70% of the glucose taken up by the uterus in ewes(Meschia et al., 1980) and cows (Reynolds et al., 1986). The weight-specific consumption ofglucose by the diffuse placental tissues of the horse and pig is even greater than that of theepitheliochorial ruminant placenta, accounting for 80–90% of uterine glucose uptake during lategestation (Fowden, 1997).

In all species, a considerable fraction of the glucose consumed by uteroplacental tissues isconverted to lactate. Rates of lactate production and disposal into maternal and fetal circula-tions vary with species and gestational age. For example, production is relatively high anddistributed mostly into the uterine circulation during late pregnancy in the mare, whereas thelower production in ruminants is mostly released into the umbilical circulation (Fowden,1997). In ruminants, horse, and pig, a further, smaller fraction of glucose consumed by utero-placental tissues is converted to fructose which is released into the fetal circulation and slowlymetabolized by fetal tissues (Meznarich et al., 1987).

2.3.2 Amino acid metabolism

Net uteroplacental consumption of amino acids, as a fraction of uterine uptake, is lower than that of glucose, presumably related to the negligible or small growth of the placenta

A. W. Bell et al.6

and uterine tissues in sheep (Ehrhardt and Bell, 1995) and cattle (Bell et al., 1995) during late pregnancy. Nevertheless, net removal by the uteroplacental tissues has been estimated toaccount for 24% of uterine uptake of amino acid nitrogen in well-fed ewes during late preg-nancy (Chung et al., 1998).

2.4. Gestational development of conceptus metabolism

The many-fold increase in fetal mass from mid- to late gestation is, not unexpectedly, accom-panied by increased absolute rates of uterine and umbilical uptake of oxygen and nutrientsand of urea export by conceptus tissues, and of fetal whole-body protein synthesis in sheepand cattle (Bell et al., 1986, 1989; Reynolds et al., 1986; Kennaugh et al., 1987; Ferrell, 1991).However, when expressed on a weight-specific basis these rates are considerably greater inmid than in late gestation, concomitant with greater rates of relative growth in the immaturefetus. More recent studies of fetal and uteroplacental metabolic ontogeny in the horse haveshown a qualitatively similar pattern (Fowden et al., 2000a,b). The apparent absence of adecrease in weight-specific fetal oxygen consumption between mid- and late gestation in thisspecies (Fowden et al., 2000a) may be related to its slower relative rates of fetal growth andthe failure to account for the greater tissue hydration of the mid-gestation fetus. In sheep, thegestational decline in weight-specific fetal whole-body metabolic rates is associated withchanges in the allometric growth of metabolically active vital organs, such as the liver, versusthat of less active skeletal tissues (Bell et al., 1987a), as well as a decline in the weight-specificrate of fetal hepatic oxygen consumption (Vatnick and Bell, 1992).

3. PLACENTAL TRANSPORT OF MACRONUTRIENTS

3.1. Molecular and physiological mechanisms

3.1.1. Glucose

Glucose is transported from the maternal to the fetal circulation by carrier-mediated, facilitateddiffusion (Widdas, 1952; Simmons et al., 1979). This process is strongly dependent on thematernal–fetal plasma glucose concentration gradient (Simmons et al., 1979; DiGiacomo andHay, 1990a). The predominant glucose transporter protein isoforms in the sheep placenta areGLUT-1 and GLUT-3 (Ehrhardt and Bell, 1997; Das et al., 1998), the mRNA and proteinabundance of which increase with gestational age, especially for GLUT-3 (Currie et al., 1997;Ehrhardt and Bell, 1997). This, together with its low Km and localization at the apical, maternal-facing layer of the trophoblastic cell layer (Das et al., 2000), suggests that ontogenic changesin GLUT-3 expression and activity may account for much of the 5-fold increase in glucosetransport capacity of the sheep placenta in vivo between mid- and late gestation (Molinaet al., 1991). Other factors must include remodeling and expansion of the placenta’s effec-tive exchange surface (Stegeman, 1974) and the increasing maternal–fetal plasmaconcentration gradient (Molina et al., 1991). Similar developmental patterns in placentalGLUT expression have been observed in the rat (Zhou and Bondy, 1993) but not in the human(see Illsley, 2000) or horse placenta (Wooding et al., 2000), in which gestational changeswere small or absent. These species differences may be due to the considerably slower ratesof fetal growth and glucose demand in humans and horses, and, possibly, their greaterdependence on changes in placental morphology to permit increased fetal access to glucoseduring late pregnancy.


3.1.2. Amino acids

Most amino acids taken up by the placenta are transported against a fetal–maternal concen-tration gradient, implying the use of energy-dependent, active transport processes (Young andMcFadyen, 1973). Studies of isolated human and rodent placental vesicles have confirmed thatthe transport systems in the placenta are similar to those described for plasma membranes ofother tissues (see Battaglia and Regnault, 2001). These include at least six sodium-dependentand five sodium-independent systems that have been classified systematically on the basis oftheir affinity for neutral, acidic, or basic amino acids, and their intracellular location (Battagliaand Regnault, 2001). Recent results from in vivo studies on sheep suggest that rapid placentaltransport of neutral amino acids requires both sodium-dependent transport at the maternalepithelial surface and affinity for highly reversible, sodium-independent transporters locatedat the fetal surface (Jozwik et al., 1998; Paolini et al., 2001). These researchers also demon-strated major differences in placental clearance among the essential amino acids, with themore rapidly transported branched-chain acids, plus methionine and phenylalanine, apparentlysharing the same rate-limiting transport system (Paolini et al., 2001).

3.1.3. Fatty acids

Placental capacity for maternal–fetal transport of short- and long-chain fatty acids and theirketo-acid derivatives varies widely among species, associated with variations in placentalstructure (see Bell and Ehrhardt, 2002). Thus, the epitheliochorial placentae of ruminantsand the diffuse placentae of pigs and horses allow only meager fetal access to maternalfatty acids and ketones, whereas the hemochorial placentae of rodents, lagomorphs, and, byinference, humans, are more permeable to these substrates. Molecular mechanisms for pla-cental transport of fatty acids have yet to be defined but may involve a placenta-specificfatty-acid binding protein that has been identified in sheep (Campbell et al., 1994) andhumans (Campbell et al., 1995).

3.2. Influence of placental metabolism on maternal–fetal nutrient transfer

3.2.1. Glucose metabolism

Glucose entry into the gravid uterus and its component tissues is determined by maternal arte-rial glucose concentration (Hay and Meznarich, 1988; Leury et al., 1990b), while glucosetransport to the fetus is determined by the transplacental (maternal–fetal) concentration gra-dient (Hay et al., 1984). In turn, the transplacental gradient is directly related to both placentaland fetal glucose consumption, which are dependent on fetal arterial glucose concentration(Hay et al., 1990). Thus, as fetal glucose concentration changes relative to that of the mother,thereby changing the transplacental gradient, placental transfer of glucose to the fetus variesreciprocally with placental glucose consumption.

In addition to its quantitative impact on placental transfer of glucose, placental glucosemetabolism has a major qualitative influence on the pattern of carbohydrate metabolitesdelivered to the fetus. Rapid metabolism to lactate (~35%), fructose (~4%), and CO2 (~17%)accounted for about 56% of uteroplacental glucose consumption in late-pregnant ewes, andwas directly related to placental glucose supply (Aldoretta and Hay, 1999). The fate of theremaining 44% of glucose metabolized by the placenta must include synthesis of alanine and othernonessential amino acids (Timmerman et al., 1998), directly or via lactate (Carter et al., 1995).

A. W. Bell et al.8

Umbilical uptake and fetal oxidation of placentally derived lactate (Sparks et al., 1982; Hayet al., 1983) and fructose (Meznarich et al., 1987) are estimated to contribute approximately20% and 5%, respectively, to fetal CO2 production, in addition to the 30% contributed by therapid oxidation of glucose (Hay et al., 1983).

3.2.2. Amino acid metabolism

Placental metabolism also affects the quantity and composition of amino acids delivered tothe fetus. The significant net consumption by uteroplacental tissues of glutamate, serine, andthe branched-chain amino acids (Liechty et al., 1991b; Chung et al., 1998) implies catabolismor transamination of these acids. An additional, small fraction of this net loss of amino acidswill be in the form of secreted peptides.

The ovine placenta has very little enzymatic capacity for urea synthesis, but produces con-siderable amounts of ammonia, much of which is released into maternal and, to a lesserextent, fetal circulations (Holzman et al., 1977; Bell et al., 1989). This is consistent withextensive placental deamination of branched-chain amino acids to their respective keto-acids,which are released into fetal and maternal bloodstreams (Smeaton et al., 1989; Loy et al.,1990), and with rapid rates of glutamate oxidation in the placenta (Moores et al., 1994).Transamination of branched-chain amino acids accounts for some of the net glutamate acqui-sition by the placenta, the remainder of which is taken up from the umbilical circulation(Moores et al., 1994). That which is not quickly oxidized combines with ammonia to synthe-size glutamine, which is then released back into the umbilical bloodstream (Chung et al.,1998). Some of this glutamine is converted back to glutamate by the fetal liver, which pro-duces most of the glutamate consumed by the placenta (Vaughn et al., 1995). This establishesa glutamate–glutamine shuttle which promotes placental oxidation of glutamate and fetalhepatic utilization of the amide group of glutamine.

Similarly, the placenta almost quantitatively converts serine, mostly taken up from maternalblood, to glycine (Chung et al., 1998), reconciling the discrepancy between the negligible netuptake of glycine by the uterus and substantial net release of this amino acid into the umbilicalcirculation (see Hay, 1998).

The complexity of interrelations among placental uptake, metabolism, and transport ofamino acids was further illustrated by a study of alanine metabolism in ewes during latepregnancy (Timmerman et al., 1998). Application of tracer methodology showed that negli-gible net placental consumption or production of alanine masks an appreciable metabolism ofmaternal alanine entering the placenta which exchanges with endogenously produced alanine.Thus, most of the alanine delivered to the fetus is of placental origin, derived from placentalprotein turnover and transamination.

3.2.3. Fatty acid metabolism

The relative abundance of polyunsaturated C20 and C22 derivatives of the essential C18 fattyacids in fetal tissues has been attributed largely to the placenta’s capacity for hydrolyzingesterified lipids (Clegg, 1981) and for desaturation and chain elongation of the resulting freepolyunsaturated C18 acids (Noble et al., 1985). Thus, placental metabolism ensures anadequate fetal supply of the longer-chain ω6 and ω3 metabolites of the C18 essential fattyacids, which are the forms ultimately required by tissues, despite the poor placental transportof the parent molecules


3.3. Factors affecting placental transport capacity

3.3.1. Placental size

Placental capacity for glucose transport was substantially reduced, as were uteroplacentalglucose consumption rate and fetal glycemia, in carunclectomized (Owens et al., 1987a) andheat-treated ewes (Bell et al., 1987b; Thureen et al., 1992). At least part of the absolute reduc-tion in glucose transport capacity is presumed to be due to reduction in exchange surface areaof the trophoblastic membrane, as shown in carunclectomized ewes (Robinson et al., 1995).In previously heat-treated ewes (Thureen et al., 1992), placental weight-specific glucosetransport capacity was also reduced. This implies that chronic heat stress, which reduces aver-age weight but not total number of placentomes, additionally reduces the number and/oractivity of specific glucose transport proteins at maternal and/or fetal exchange surfaces. Incontrast, carunclectomy, which reduces placentome number but may stimulate a compensa-tory increase in average weight of individual placentomes, caused a modest increase in theplacental weight-specific clearance of the nonmetabolizable glucose analog, 3-O-methyl glucose (Owens et al., 1987b). This implies that glucose transporter expression was preservedor increased in the remaining placentomes.

Placental insufficiency in heat-treated ewes also extends to impaired capacity for aminoacid transport, including major reductions in placental uptake and fetal transfer of leucine(Ross et al., 1996) and threonine (Anderson et al., 1997). The normally extensive placentalcatabolism of leucine was also greatly reduced (Ross et al., 1996).

3.3.2. Maternal nutrition

Recent evidence indicates that the activity of placental transport mechanisms can be modu-lated by maternal nutrition, independent of more chronic effects on placental size. Forexample, moderate undernutrition of ditocous ewes during late pregnancy caused a 50%increase in capacity for maternal–fetal glucose transport in vivo (Ehrhardt et al., 1996) whichwas at least partly explained by a 20% increase in total GLUT abundance, associated with asimilar increase in GLUT-3 protein abundance (Ehrhardt et al., 1998). These responses helpexplain how placental glucose transfer remained sufficient to sustain normal fetal growth,despite chronic maternal hypoglycemia and a 26% decrease in the maternal–fetal gradient inarterial plasma glucose concentration (Bell et al., 1999).

During more severe, chronic undernutrition or starvation for several days, the developmentof profound fetal hypoglycemia helps to sustain the maternal–fetal gradient in glucose con-centration by restricting the reverse transfer of glucose to the placenta, and reducing placentalglucose consumption (see Hay, 1995). Under these more stringent conditions, fetal gluco-neogenesis is induced (Leury et al., 1990a), with amino acids being the presumed majorsubstrate, consistent with increased fetal urea synthesis (Lemons and Schreiner, 1983;Faichney and White, 1987). The ultimate consequence is reduced fetal tissue protein synthesis(Krishnamurti and Schaefer, 1984) and slowing of fetal growth to a rate that can be sustainedby the reduced placental nutrient supply.

Effects of energy and/or protein supply on placental capacity for amino acid transport have beenlittle studied. Fasting late-pregnant ewes for 5 days had an insignificant effect on umbilical netuptake of amino acids despite appreciable decreases in maternal arterial plasma concentrations ofmany amino acids (Lemons and Schreiner, 1983). This suggests that during short-term energy/protein deprivation, placental mechanisms for active transport of amino acids are unimpaired andmay even be upregulated. Under similar fasting conditions, the uteroplacental deamination of

A. W. Bell et al.10

branched-chain amino acids appeared to be increased, judging from a 3-fold increase in the effluxof α-ketoisocaproate, the keto-acid derivative of leucine, into uterine and umbilical circulations(Liechty et al., 1991a). This suggests that increased amino acid catabolism may partly compen-sate for the likely reduction in placental glucose oxidation under these conditions.

Placental transport and metabolism of amino acids have not been studied during more pro-longed restriction of energy or protein. However, in ewes fed adequate energy but insufficientprotein during the last month of pregnancy, fetal growth and protein deposition over thisperiod were decreased by 18% (McNeill et al., 1997). It is also notable that in chronicallyhyperglycemic ewes with secondary hyperinsulinemia and hypoaminoacidemia, placentaland fetal uptakes of several amino acids were reduced, and fetal total nitrogen uptake declinedby 60% (Thureen et al., 2001).

3.4. Consequences of placental insufficiency

Placental weight and associated capacity for maternal–fetal nutrient transfer are powerfuldeterminants of fetal growth during late gestation in all species studied. This has been mostpersuasively demonstrated by controlled manipulation of placental size and/or functionalcapacity using premating carunclectomy (Alexander, 1964), heat-induced placental stunting(Alexander and Williams, 1971), or uteroplacental vascular embolization (Creasy et al., 1972).Natural variations in fetal weight due to varying litter size in prolific ewes are strongly corre-lated with placental mass per fetus (Rhind et al., 1980; Greenwood et al., 2000). Recently, the quite profound growth retardation of fetuses in overfed, primiparous ewes also has beenattributed to a primary reduction in placental growth (Wallace et al., 2000).

The probably common etiology of IUGR in experimentally induced and natural cases ofplacental insufficiency is illustrated by the similar patterns of association between fetaland placental weights in pregnant ewes with varying conceptus weights due to carunclectomy,heat stress, litter size, and overfeeding of primiparous dams (fig. 2). In each case, severegrowth retardation was associated with chronic fetal hypoxemia and hypoglycemia during lategestation (Creasy et al., 1972; Harding et al., 1985; Bell et al., 1987b; Wallace et al., 2002).

4. REGULATION OF CONCEPTUS METABOLISM

4.1. General features

The extracellular and local regulation of fetal metabolism and its relation to tissue growth hasseveral distinctive characteristics. First, placental nutrient supply has a powerful, limitinginfluence on nutrient disposal, especially in late gestation when fetal demands are greatest.Second, the fetal endocrine system is largely independent from the direct influence of mater-nal hormones because the placenta is impermeable to most of the important metabolicregulatory peptide and steroid hormones. Thus, reported effects of maternal hormones onfetal growth must be mediated indirectly by changes in maternal metabolism and/or utero-placental tissue growth and resulting changes in fetal nutrient supply. Examples include theeffects of maternal treatment with growth hormone (GH) during early pregnancy on fetalgrowth in pigs (Sterle et al., 1995; Rehfeldt et al., 2001) and of maternal immunizationagainst placental lactogen (PL) on fetal growth in sheep (Leibovich et al., 2000). Third, whilemost fetal endocrine organs develop the capacity to synthesize and secrete hormones early ingestation, target tissue receptor and neuroendocrine feedback systems are variably immatureuntil late pregnancy. As a result, there is a much greater reliance on paracrine and autocrineregulation by locally expressed factors, especially in early and mid-pregnancy.


4.2. Nutrient supply

4.2.1. Glucose

The Km for saturable glucose transport by the sheep placenta is ~3.9 mM (Simmons et al.,1979), which is within the physiological range of glycemia in well-fed, adult sheep. Thus,uterine uptake, placental metabolism and transfer, and fetal metabolism of glucose are verysensitive to maternal arterial glucose concentration in sheep (fig. 3; Hay and Meznarich,1988). In sheep, cows, and horses fetal utilization of glucose is highly correlated with fetalplasma glucose concentration, which, in turn is correlated with maternal glycemia (seeFowden, 1997). In contrast, fetal glucose metabolism was not related to fetal glycemia inpigs, possibly because in this species, fetal glycemia is influenced by individual relative tototal fetal mass, as well as maternal nutrition (Fowden et al., 1997).

It is well established that in sheep, the maternal and fetal hypoglycemia caused by starva-tion or chronic undernutrition is associated with increased fetal urea synthesis (Hodgson et al.,1982; Lemons and Schreiner, 1983; Faichney and White, 1987) due to increased amino aciddeamination (Krishnamurti and Schaefer, 1984; Van Veen et al., 1987). Conversely, fetalhyperglycemia appears to cause substitution of glucose for amino acids as an oxidative fuelbecause under these conditions, increased glucose oxidation (Hay and Meznarich, 1986) isassociated with decreased leucine oxidation (Liechty et al., 1991a). Interestingly, the latterresponse occurred independently of glucose-induced changes in fetal insulin concentration(Liechty et al., 1993).

Fetal glucose supply also influences fetal endogenous glucose production, presumably dueto hepatic gluconeogenesis. In addition to the association of increased endogenous production

A. W. Bell et al.12

Fig. 2. Relation between fetal and placental weights in ewes representing different models of placentalinsufficiency during late pregnancy. Variation in placental weight was achieved by premating carunclectomy(●; Owens et al., 1986), chronic heat treatment (�; Bell et al., 1987b), natural variation in litter size(▲; Greenwood et al., 2000), and overfeeding of adolescent ewes (�; Wallace et al., 2000). Reproduced withpermission from the Society for Reproduction and Fertility (Greenwood and Bell, 2003).


Fig. 3. Relations between maternal arterial blood glucose concentration and (A) uterine, (B) fetal, and(C) uteroplacental net uptakes of glucose in ewes during late pregnancy. Reproduced with permission fromthe Society for Experimental Biology and Medicine (Hay and Meznarich, 1988).

with fetal hypoglycemia in starved or undernourished ewes (Hay et al., 1984; Leury et al., 1990a),progressive fetal hypoglycemia induced by different levels of maternal insulin infusioncaused fetal endogenous glucose production to increase linearly (DiGiacomo and Hay, 1990b).A mediating role for fetal insulin was suggested by the incomplete suppression of endoge-nous glucogenesis by fetal infusion with insulin while maintaining basal fetal glycemia(DiGiacomo and Hay, 1990b).

4.2.2. Amino acids

Effects of amino acid supply on fetal metabolism have not been studied systematically.Decreased maternal plasma concentrations of essential amino acids in fasted ewes were notassociated with a significant decrease in umbilical uptake of these acids (Lemons andSchreiner, 1983). In contrast, maternal hyperglycemia with secondary hyperinsulinemia andhypoaminoacidemia caused substantial reductions in uterine, uteroplacental, and fetal uptakesof several amino acids, particularly the branched-chain acids, and a 60% reduction in totalfetal uptake of nitrogen (Thureen et al., 2000, 2001). Correction of maternal amino acid con-centrations by appropriate exogenous infusion restored uterine and umbilical exchanges tonormal levels (Thureen et al., 2000). Maternal hyperaminoacidemia, caused by infusion ofamino acids, had little effect on the umbilical uptake of most amino acids, except forincreased uptake of the branched-chain acids, and did not affect fetal total nitrogen supply(Jozwik et al., 1999). However, uteroplacental production and fetal concentrations of ammo-nia increased moderately, implying some increase in placental deamination of amino acids.

4.3. Fetal hormones and growth factors

4.3.1. Pancreatic hormones

Insulin protein and preproinsulin mRNA are detectable from early gestation in the fetal pan-creas of all species studied (D’Agostino et al., 1985). In the sheep fetus, gestational increasesin pancreatic and basal plasma concentrations of insulin (Alexander et al., 1968) are accom-panied by a steady increase in glucose- and arginine-stimulated insulin secretion during thelatter half of gestation (Aldoretta et al., 1998). Euglycemic, hyperinsulinemic clamp studieshave demonstrated that fetal insulin and glucose have independent, positive effects on fetalwhole-body glucose utilization (Hay et al., 1988). These observations are consistent withtissue-specific responses that vary between insulin-responsive tissues, such as skeletal muscle(Wilkening et al., 1987; Anderson et al., 2001b), and tissues unresponsive to insulin, such asthe brain (Anderson et al., 2001a).

Neither fetal (Jodarski et al., 1985) nor maternal (Rankin et al., 1986) plasma insulinconcentration has a direct effect on placental transport of glucose, consistent with our failureto detect significant concentrations of the insulin-responsive glucose transport protein,GLUT-4, in the ovine placenta (Ehrhardt and Bell, 1997). However, fetal hyperinsulinemiaindirectly promotes placental transfer and umbilical uptake of glucose through its influenceon fetal glycemia and the maternal–fetal glucose concentration gradient (see Hay, 1995).

In addition to its promotion of glucose uptake and metabolism in fetal tissues, a physio-logical increase in fetal plasma insulin stimulated umbilical uptake and whole-bodyutilization of amino acids when fetal glycemia and aminoacidemia were carefully controlled(Thureen et al., 2000). The specific metabolic fates of amino acids were not measured, but itis likely that protein anabolism was increased by both reduction of proteolysis (Milley, 1994)

A. W. Bell et al.14

and stimulation of protein synthesis (Horn et al., 1983). This anabolic effect may have beenreinforced indirectly by the effect of increased glucose utilization in reducing amino aciddeamination (Liechty et al., 1993).

Independently of its metabolic effects, insulin may influence fetal tissue growth throughmodulation of the expression and activity of other growth regulators such as the insulin-likegrowth factor (IGF) system. For example, when fetal plasma glucose and amino acid concen-trations were clamped, fetal insulin infusion caused an increase in plasma concentration of IGFbinding protein (BP)-3 and a decrease in hepatic expression of IGFBP-1 (Shen et al., 2001).The latter response is consistent with the opposite effects of hypoinsulinemia in the under-nourished sheep fetus (Osborn et al., 1992). Ovine fetal hyperinsulinemia also increased thefarnesylation of p21 Ras in ovine fetal liver, skeletal muscle, adipose tissue, and white bloodcells (Stephens et al., 2001). This is significant because the Ras pathway is an important intra-cellular signaling element in the mitogenic actions of insulin and other growth factors,including the IGFs, and greater availability of farnesylated Ras augments mitogenic cellularresponsiveness to IGF-1 and other growth factors in isolated systems (Goalstone et al., 1998).

The fetal pancreas synthesizes glucagon from early in gestation, but the regulation andmetabolic role of this peptide in fetal life remain unclear. Secretory responses to hypo-glycemia and other metabolic stimuli in fetal sheep are small and sluggish during lategestation (Alexander et al., 1976), but birth is accompanied by a major surge in secretion ofglucagon (Grajwer et al., 1977). Exogenous administration of glucagon to mimic fetal plasmaconcentrations observed during maternal fasting (Schreiner et al., 1980) caused hyper-glycemia in the fetal sheep (Philipps et al., 1983), implying a possible role in regulation ofhepatic glycogenolysis and/or gluconeogenesis.

4.3.2. Growth hormone and the IGF system

During postnatal life, growth hormone (GH) is a powerful homeorhetic regulator of metabolismand growth through its direct actions on some tissues, such as adipose tissue, and its indirectactions on most lean tissues, mediated by the IGF system (see Etherton and Bauman, 1998).Notable among its pleiotropic effects are inhibition of lipogenesis and enhancement ofresponses to lipolytic stimuli in adipose tissue, and potent effects on cell cycle activity andprotein turnover in muscle and other tissues via regulation of the expression of IGF-1 and itsbinding proteins in multiple tissues, including the liver. In general, these effects are greatlymuted during fetal life, which is characterized by persistently high plasma levels of GH (Bassettet al., 1970; Gluckman et al., 1979) and low plasma levels of IGF-1 (Van Vliet et al., 1983). Theapparent uncoupling of the GH/IGF-1 axis is consistent with low hepatic expression of the GHreceptor, IGF-1, IGFBP-3, and the acid-labile subunit (ALS) (Klempt et al., 1993; Rhoads et al.,2000a). Thus, although pituitary secretion of GH is active through much of gestation (de Zegheret al., 1989), maturation of the endocrine IGF-1 system is retarded by hepatic unresponsivenessto GH, which, in postnatal life, strongly regulates expression of all three components of theternary binding complex (IGF-1, IGFBP-3, ALS) that accounts for most circulating IGF-1(Boisclair et al., 2001). Therefore, it is not surprising that infusion of normal sheep fetuses withGH for 10 days had no effect on fetal plasma IGF-1 levels (Bauer et al., 2000).

It is possible that some direct metabolic effects of GH develop before engagement of theGH/IGF-1 system. For example, Bauer et al. (2000) reported a decrease in glucose uptakeand, presumably, utilization, with no change in plasma insulin in GH-infused fetal sheep,consistent with an earlier report of apparent insulin resistance in GH-treated fetuses (Parkesand Bassett, 1985). Also, hypophysectomy of fetal lambs causes increased fat deposition that


can be reversed by GH administration (Stevens and Alexander, 1986), implying the existenceof functional GH receptors in adipose tissue during late gestation. This could account for thesubstantial decline in capacity for adipose tissue lipogenesis in fetal sheep during the lastmonth of gestation (Vernon et al., 1981).

Immaturity of the fetal GH/IGF-1 system raises the possibility that fetal protein anabolismand tissue growth may be limited by low levels of circulating IGF-1, despite the generallyaccepted notion that, during fetal life, the metabolic and trophic influences of locallyexpressed IGF are more important than those of systemic IGF (see Jones and Clemmons,1995). It is therefore notable that infusion of IGF-1 into fetal sheep decreased proteolysis andamino acid catabolism (Harding et al., 1994; Liechty et al., 1996). Conversely, increasedamino acid catabolism in the undernourished sheep fetus (Hodgson et al., 1982; Lemons andSchreiner, 1983) is associated with decreased plasma IGF-1 levels, whether due to maternalnutrient deprivation (Bassett et al., 1990) or placental insufficiency (Owens et al., 1994).

In all species studied, fetal tissue expression and plasma levels of IGF-2 exceed those ofIGF-1 (Han et al., 1988; Mesiano et al., 1989; Lee et al., 1991; Delhanty and Han, 1993).A special role for IGF-2 in the regulation of prenatal growth was demonstrated by initial geneknockout studies in the mouse (DeChiara et al., 1991). Recently, tissue-specific gene inacti-vation has been used to show that the IGF-2 gene is paternally imprinted in the placenta andacts to promote placental growth and functional capacity, thereby influencing fetal nutrientsupply and growth in late gestation (fig. 4; Constancia et al., 2002). Lack of IGF-2 alsoreduced fetal hepatic glycogen storage and glycemia, associated with decreased activity butnot mRNA abundance of glycogen synthase, and impaired the ability of newborn IGF-2knockout mice to survive fasting for 12h (Lopez et al., 1999).

4.3.3. Placental lactogen

Placental lactogen (PL; also known as chorionic somatomammotropin) is a major, unique pro-tein product of the placentae of ruminants, humans, rodents, and some other species. Themolecular identity and interspecies homology of these molecules, as well as their lactogenic andsomatogenic effects through their ability to bind to both GH and prolactin receptors, has beenreviewed recently (Gertler and Djiane, 2002). Ovine and bovine fetal plasma contains PLthroughout gestation (Anthony et al., 1995) and the effective half-life of circulating PL in fetalsheep is similar to that of GH (Schoknecht et al., 1992). The physiological roles of this putativeregulator of fetal metabolism and growth remain to be established definitively. Glycogensynthesis in isolated fetal hepatocytes was promoted by PL treatment in sheep (Freemark andHandwerger, 1986) and rats (Freemark and Handwerger, 1984), and we observed a 56% increasein hepatic glycogen accumulation in fetal sheep infused i.v. with native ovine PL for 14 days(table 2; Schoknecht et al., 1996). In the latter study, PL treatment caused modest increases infetal plasma IGF-1 concentration and the relative weights of some visceral organs but did notsignificantly affect fetal weight.

4.3.4. Glucocorticoids

In all species studied, there is a major increase in the circulating glucocorticoid concentrationin the fetus toward term, mostly due to a pronounced surge in fetal adrenal cortisol secretion.The vital, pleiotropic influences of fetal cortisol on the structural and biochemical maturationof multiple fetal tissues to prepare them for postnatal functions have been reviewed byFowden et al. (1998). Less is known about the effects of glucocorticoids on the regulation

A. W. Bell et al.16


Table 2

Effect of i.v. infusion of ovine placental lactogen for 14 days on liver glycogen concentration,content in fetal sheep at day 136 of gestation (from Schoknecht et al., 1996)a

Parameter Controlb Placental lactogenc

Liver weight, g 115.8 ± 9.2 124.8 ± 9.9Glycogen concentration, mg/g 79.3 ± 6.9 105.0 ± 5.6*Glycogen content, g 8.4 ± 0.7 13.1 ± 1.7*

a Values are means ± SEM, n = 5.b Infused with saline containing ovine plasma (15 ml/l), days 122 to 136 of gestation.c Infused with ovine placental lactogen (1.2 mg/d), days 122 to 136 of gestation; caused a 4-fold increase infetal plasma concentration of placental lactogen.* Treatment effect was significant at P<0.05.

of fetal metabolism in relation to growth and development. In general, fetal cortisol appearsto promote the availability of glucose to the neonatal animal by stimulating both hepaticglycogen synthesis (Barnes et al., 1978) and maturation of the capacity for hepatic glucoseproduction (Townsend et al., 1991; Barbera et al., 1997) in the near-term sheep fetus. Duringlate gestation, treatment with glucocorticoids reduced umbilical glucose uptake (Milley, 1996;

Fig. 4. Placental and fetal growth in mutant mice lacking paternal expression of the IGF-2 gene inlabyrinthine trophoblastic tissue of the placenta, and in their wild-type littermates. Significant differencesbetween wild-type and mutant mice are indicated: * P< 0.05; *** P<0.001. Adapted from the data ofConstancia et al. (2002).

Barbera et al., 1997) and placental uptake of fetal glutamate (Barbera et al., 1997; Timmermanet al., 2000). The latter response was associated with decreased hepatic output of glutamateapparently due to decreased fetal hepatic uptake of glucogenic amino acids, includingglutamine, and diversion of hepatic glutamine to metabolism in the TCA cycle rather thanglutamate synthesis (Timmerman et al., 2000).

Other growth-related effects of the prepartum increase in fetal cortisol include inductionof the hepatic GH receptor and hepatic synthesis of IGF-1 (Li et al., 1996) but suppressionof IGF-1 expression in muscle independently of any change in GH receptor gene expression(Li et al., 2002). Cortisol also suppresses IGF-2 expression in liver, muscle, and adrenalglands (Li et al., 1993), stimulates the deiodination of thyroxine (T4) to triiodothyronine (T3)(Fowden et al., 1998), and appears to downregulate the production of PL by binucleate cellsin the ovine placenta (Ward et al., 2002).

4.3.5. Thyroid hormones

The fetal thyroid secretes T4 from early in gestation, and thyroidectomy of fetal sheep in mid-gestation causes generalized growth retardation and delayed maturation of the skin, skeleton,and pulmonary and neuromuscular systems (Hopkins and Thorburn, 1972). Fetal sheep madehypothyroid by thyroidectomy or hypophysectomy suffered a 20–30% decrease in umbilicaloxygen uptake that was restored to normal by exogenous T4 administration (Fowden andSilver, 1995). This reduction in oxygen consumption was accompanied by abnormal blood-gasstatus and reductions in rate of glucose oxidation and the fraction of oxygen consumptionused for glucose oxidation, all of which also were normalized by T4 replacement (Fowden andSilver, 1995). Interestingly, plasma T3 levels remained low and were unchanged by thyroid orpituitary ablation or exogenous T4, suggesting that at least before maturation of the enzymaticcapacity for T4 deiodination near term (Polk et al., 1988), thyroid hormone effects may bemediated directly by T4. However, it should be noted that administration of T3 alone, albeitin supraphysiological doses, caused an increase in oxygen consumption of fetal sheep (Lorijnet al., 1980).

In addition to its negative effects on glucose utilization, thyroid deficiency impaired theability of fetal sheep to increase hepatic glucogenesis in response to fasting (Fowden et al.,2001). Recent evidence also suggests that the cortisol-induced increase in deiodination ofT4 to T3, and the consequent prenatal surge in fetal plasma T3, at least partly mediates thematurational effects of cortisol on the hepatic somatotropic axis (Forhead et al., 2000).

4.3.6. Catecholamines

Prenatal maturation of the sympathoadrenomedullary system is vital to enable the perinatalanimal to respond to the stresses of parturition and adaptation to the extrauterine environment.In precocial species such as the sheep, central nervous (splanchnic) control of the adrenalmedulla develops relatively early; in other species, functional innervation is not apparent untilafter birth (see Slotkin and Seidler, 1988). During late gestation, the fetal sheep responds toacute hypoxia (Cohen et al., 1982) and hypoglycemia (Harwell et al., 1990) with pronouncedincreases in adrenomedullary secretion of epinephrine and norepinephrine. Metabolic conse-quences include rapid stimulation of hepatic glucose production, presumably through increasedglycogenolysis (Jones et al., 1983), and mobilization of NEFA (Harwell et al., 1990), associatedwith reduced pancreatic secretion and plasma concentrations of insulin (Bassett and Hanson,1998), and attenuated action of IGF-1 (Hooper et al., 1994). Restoration of normal insulinemia

A. W. Bell et al.18

by insulin infusion abolished most of the metabolic and growth-inhibitory effects of prolongedcatecholamine infusion in the sheep fetus (Bassett and Hanson, 2000). This suggests thatestablishment and maintenance of hypoinsulinemia is a necessary mediating factor for theadverse effects of elevated circulating catecholamines on fetal growth.

4.3.7. Leptin

The peptide hormone leptin is synthesized and secreted primarily by adipose tissue in postnatalanimals and is considered to play an important role in the regulation of energy balance (Ahimaand Flier, 2000). Leptin has been detected in ovine fetal plasma as early as day 40 of gestation(Ehrhardt et al., 2002) and its concentration increases moderately throughout gestation, espe-cially during the last 2 weeks (Ehrhardt et al., 2002; Forhead et al., 2002). The gestationalincrease in plasma concentration is accompanied by increased abundance of leptin mRNA inperirenal adipose tissue in late pregnancy (Yuen et al., 1999). Our results indicate that before100 days of gestation, tissues other than adipose tissue, such as brain and liver, are the primarysource of circulating leptin, and that this role is assumed by brown adipose tissue only after thistissue develops appreciably during the last one-third of gestation (fig. 5; Ehrhardt et al., 2002).

Regulation of tissue expression and biological actions of leptin during fetal life have yet tobe studied systematically. The increase in plasma leptin during late gestation in fetal sheep wasassociated with the prepartum surge in fetal cortisol and abolished by fetal adrenalectomy(Forhead et al., 2002). It also appears that expression of leptin mRNA in perirenal brown adi-pose tissue in the sheep fetus responds positively to hyperinsulinemia but not hyperglycemia(Devaskar et al., 2002). The functional significance of fetal leptin is unclear. Leptin signalingapparently is not essential during prenatal life because leptin-deficient ob/ob mice are bornrelatively normal (Mounzih et al., 1998). Also, fetal plasma leptin was unaffected by changes inmaternal nutrition sufficient to change fetal glycemia and insulinemia in late-pregnant ewes(Ehrhardt et al., 2002; Mühlhäusler et al., 2002; Yuen et al., 2002). Fetal plasma leptin wascorrelated with body fatness as represented by the relative mass of unilocular cells in perirenaland interscapular brown adipose tissue (Mühlhäusler et al., 2002). Infusion of leptin for severaldays into the sheep fetus caused decreases in relative abundance of leptin mRNA and the pro-portion of unilocular cells in perirenal adipose tissue, suggesting a feedback effect on adiposetissue function (Yuen et al., 2003). However, the relevance of this observation is unclear becauseof the unphysiologically high levels of plasma leptin in treated fetuses. A potential role for fetaland/or maternal leptin in the regulation of placental function is suggested by the abundantexpression of the physiologically relevant long (Ob-Rb) form of the leptin receptor by the ovineplacenta (Ehrhardt et al., 1999; Thomas et al., 2001).

4.4. Coordination of fetal metabolism and growth

The mechanisms relating nutrient supply to expression of endocrine and local regulatoryfactors and, thence, tissue metabolism and growth, can be illustrated by synthesis of the presentknowledge on IUGR, whether caused by placental insufficiency, maternal undernutrition, orinsulin-induced maternal hypoglycemia. Effects on the local expression of trophic factors andthe cellular growth of skeletal muscle will serve as an example of tissue responses toan altered extracellular milieu. The putative relationships discussed below are schematicallyrepresented in fig. 6.

Placental insufficiency during late gestation is generally characterized by fetal hypoxemiaand hypoglycemia, whether caused by surgical reduction (carunclectomy; Harding et al., 1985),


placental embolization (Creasy et al., 1972), maternal heat stress (Bell et al., 1987b), or over-feeding of adolescent ewes (Wallace et al., 2002). Associated endocrine changes includedecreased fetal plasma concentrations of insulin (Robinson et al., 1980) and IGF-1 and -2(Owens et al., 1994), and increased concentrations of cortisol (Phillips et al., 1996). All ofthese changes can be elicited by maternal undernutrition or insulin-induced hypoglycemia,implicating fetal glycemia as an important primary signal (Mellor et al., 1977; Osgerby et al.,2002). However, it must be recognized that hypoxemia may reinforce these responses throughits stimulation of fetal adrenal secretion of cortisol and catecholamines, and the inhibitoryinfluence of the latter on fetal insulin secretion.

It seems likely that hypoinsulinemia is a primary, coordinating mediator of the numerousmetabolic and trophic consequences of reduced fetal nutrient supply. Disruption of fetal pan-creatic insulin secretion has a potent, negative effect on fetal growth (Fowden et al., 1995),

A. W. Bell et al.20

Fig. 5. Relative abundance of leptin mRNA in ovine fetal tissues at different stages of gestation. Adaptedfrom the data of Ehrhardt et al. (2002).

associated with decreased fetal tissue uptake and metabolism of glucose (Fowden and Hay,1988), decreased uptake of amino acids and increased proteolysis (Carver et al., 1997), andreduced circulating levels of IGF-1 (Gluckman et al., 1987).

However, although circulating IGF-1 may be of increasing importance during late gesta-tion, it is likely that local tissue expression and actions of this and other growth factors aremore significant mediators of tissue growth responses to altered nutrient supply. For exam-ple, fetal muscle strongly expresses IGF-1 throughout gestation (Dickson et al., 1991; Leeet al., 1993) and disruption of the IGF-1 gene causes lethal abnormalities in muscle develop-ment (Liu et al., 1993), consistent with the extensive evidence for the role of IGF-1 inregulation of myogenesis (Florini et al., 1996). It therefore seems likely that the reducedmitotic activity of myosatellite cells and growth of skeletal muscle in acutely undernourishedor placentally growth-retarded sheep fetuses (Greenwood et al., 1999) was mediated, at leastpartly, by reduced local expression of IGF-1, possibly caused by elevated plasma levels ofcortisol (Li et al., 2002).

Finally, although this section has focused on IUGR to illustrate aspects of the coordinationof nutrient supply with growth in the fetus, it should be recognized that even in optimally fed,healthy animals, fetal growth is constrained by placental capacity for nutrient transfer duringlate pregnancy. This phenomenon ensures that the unborn animal’s demands upon itsdam’s nutrient reserves are not excessive, and reduces the possibility of birth injury to itself


Fig. 6. Schematic outline of some important factors linking maternal undernutrition and placentalinsufficiency to intrauterine growth retardation.

and its mother. The capacity for increased growth in response to increased nutrient supply wasdemonstrated by the almost 20% increase in birth weight of singleton lambs that had beeninfused directly with glucose for the last 30 days of gestation, in ewes that were extremelywell fed (Stevens et al., 1990).

5. INFLUENCE OF PRENATAL METABOLISM AND GROWTHON POSTNATAL PERFORMANCE AND HEALTH

5.1. Postpartum metabolism and growth

We recently have reported some of the metabolic characteristics of naturally growth-retardedlambs from prolific ewes immediately after birth and during neonatal growth to a nominal liveweight (LW) of 20 kg (Greenwood et al., 2002; Greenwood and Bell, 2003). At birth, theselambs tended to be hypoglycemic and had elevated plasma urea nitrogen levels. More strik-ing was the apparent immaturity of their hepatic GH/IGF system as represented by greatlyelevated plasma concentrations of GH and low concentrations of IGF-1. This blood picturewas associated with reduced hepatic expression of the GH receptor and the GH-dependentALS necessary to form the ternary binding complex which contains most circulating IGF-1in postnatal life (Rhoads et al., 2000a,b).

Postnatal changes in superficial indices of carbohydrate and protein metabolism were littleaffected by birth weight in small and normal lambs that were artificially reared with ad libitumaccess to milk replacer. The very high concentrations of plasma GH in small, newborn lambsdecreased markedly within 2 days of birth but remained significantly higher than concentra-tions in normal lambs for about 2 weeks. During the same period, plasma IGF-1 increasedsteadily in both groups but remained significantly lower in the small lambs (Greenwood et al.,2002). These observations suggest that the apparent immaturity of the GH/IGF axis in growth-retarded newborn lambs persists for several weeks after birth. Interestingly, only during thisearly postnatal phase did the absolute growth rates of low-birth-weight lambs (248 g/d) lagsignificantly behind those of normal birth weight lambs (353 g/d) (Greenwood et al., 1998).Thereafter, during rapid growth from about 2 weeks of age to slaughter at 20 kg (attained at6.5–8 weeks of age), plasma IGF-1 concentrations were persistently higher but GH concen-trations were not different in low-versus normal-birth-weight lambs, perhaps related to thehigher relative energy intakes and plasma insulin concentrations (see below) of the small lambs.This study did not examine the consequences of low birth weight after weaning. However,plasma GH concentrations tended to be higher during adolescence (~132 days of age) andadulthood (~378 days of age) in low-birth-weight male lambs from carunclectomized ewescompared to lambs of normal birth weight and were negatively correlated with indices ofbirth size (Gatford et al., 2002).

Plasma insulin concentrations increased rapidly during the early postnatal period in smalllambs feeding ad libitum, consistent with their very high levels of energy intake. Then, fromabout 2 weeks of age until slaughter at 20 kg, plasma insulin concentrations were persistentlyhigher in low-compared with normal-birth-weight lambs. We speculate that this relativehyperinsulinemia may be due to the predisposition of growth-retarded neonates to developinsulin resistance (Hales et al., 1996).

Plasma leptin concentrations were somewhat higher in rapidly fattening, low-birth-weightlambs during the first week post partum, but not thereafter (Ehrhardt et al., 2003), despite thefact that at any subsequent body weight up to 20 kg LW, these lambs were significantly fatterthan their normal-birth-weight counterparts (Greenwood et al., 1998).

A. W. Bell et al.22

Additional aspects of whole-body and tissue growth and development in lambs sufferingIUGR are discussed in another, recent review (Greenwood and Bell, 2003).

5.2. Fetal programming of postnatal pathophysiology

The human epidemiological evidence for fetal programming has implicated IUGR as animportant risk factor for mature onset of diseases including hypertension and type II diabetes(Barker, 1998). Although the methodology and interpretation of aspects of this work recentlyhave been challenged (Huxley et al., 2002), these epidemiological associations have beenreplicated experimentally in rodents (Langley-Evans, 2001) and various models of IUGR insheep (McMillen et al., 2001; Greenwood and Bell, 2003). For example, low-birth-weightoffspring born to protein-deprived rats (Langley and Jackson, 1994) and placentally insuffi-cient ewes (McMillen et al., 2001) display hypertension during postweaning growth andadulthood. The rat model also has been used to demonstrate a relation between prenatal nutri-tion and the later development of insulin resistance (Langley et al., 1994), and similarevidence is emerging from sheep studies (Greenwood and Bell, 2003).

Mechanisms linking prenatal nutrition, organ and tissue development, and the programming oflater pathophysiology are unclear. However, excessive fetal exposure to glucocorticoids is a con-sistent feature of most animal studies involving prenatal nutrient deprivation, especially duringlate gestation. Also, treatment of pregnant rats and sheep with glucocorticoids during late preg-nancy can replicate some of the programming effects of fetal undernutrition on later developmentof hypertension and insulin resistance (Langley-Evans, 2001; Greenwood and Bell, 2003).

Growing evidence from studies on sheep and other species indicates that fetal programmingcan involve long-term sequelae to changes in the early prenatal environment that do not neces-sarily cause changes in fetal gross morphology. For example, modest undernutrition of ewesduring the first half of pregnancy had no effect on growth of lambs during fetal or postnatal lifebut caused relative hypertension and increased activity of the hypothalamic–pituitary–adrenal(HPA) axis in lambs aged 12–13 weeks (Hawkins et al., 2000). Consistent with these responses,maternal undernutrition between early and mid-gestation caused increased expression of theglucocorticoid receptor in adrenals, kidney, liver, lungs, and perirenal adipose tissue of the fetusat term (~145 days) (Whorwood et al., 2001). At the same time, there were marked changes inthe enzymatic capacity of several fetal tissues to deactivate cortisol, which may have led toexcessive fetal exposure to this hormone during late gestation. Some of these tissue-specificfetal responses were evident as early as day 77 of gestation.

A central role for corticosteroids in the mediation of fetal programming was further impli-cated by the remarkable finding that exposure of ewes to high doses of dexamethasone foronly 2 days in early pregnancy resulted in hypertensive offspring at 3–4 months of age (Dodicet al., 1998). This hypertension amplified with age to beyond 3 years and was associated withincreased cardiac output (Dodic et al., 1999) but no change in responsiveness of the HPA axis(Dodic et al., 2002). Glucose metabolic responses to insulin were unaltered, but the ability ofinsulin to suppress net fatty acid release from adipose tissue (plasma nonesterified fatty acidconcentration) was moderately enhanced (Gatford et al., 2000).

6. FUTURE PERSPECTIVES

The development almost four decades ago of novel techniques for studying fetal and placentalphysiology and metabolism in utero has led to considerable understanding of the regulation


of the metabolic and developmental processes that culminate in the birth of a healthy, well-grown neonate. Nevertheless, unexplained dysfunctions of conceptus growth remain,associated with unacceptable incidence of perinatal morbidity and mortality in many humanand domestic animal populations. There also is a new awareness of the possible longer-termeffects of nutritional and other environmental insults during fetal life, some of which may bequite subtle and without influence on gross morphology. Unraveling the mechanisms under-lying such effects will be the major challenge of prenatal biology for the foreseeable futureand should lead to a greater understanding of both human mature-onset pathologies andunexplained variation in the productivity and disease resistance of domestic animals.

REFERENCES

Ahima, R.S., Flier, J.S., 2000. Leptin. Annu. Rev. Physiol. 62, 413–437.

Aldoretta, P.W., Carver, T.D., Hay, W.W. Jr., 1998. Maturation of glucose-stimulated insulin secretion in

fetal sheep. Biol. Neonate 73, 375–396.

Aldoretta, P.W., Hay, W.W. Jr., 1999. Effect of glucose supply on ovine uteroplacental glucose metabo-

lism. Amer. J. Physiol. 277, R947–R958.

Alexander, D.P., Assan, R., Britton, H.G., Fenton, E., Redstone, D., 1976. Glucagon release in the sheep

fetus I. Effect of hypo- and hyperglycemia and arginine. Biol. Neonate 30, 1–10.

Alexander, D.P, Britton, H.G., Cohen, N.M., Nixon, D.A., Parker, R.A., 1968. Insulin concentrations in

the foetal plasma and foetal fluids of the sheep. J. Endocrinol. 40, 389–390.

Alexander, G., 1964. Studies on the placenta of the sheep (Ovis aries L.): effect of reduction in the

number of caruncles. J. Reprod. Fertil. 7, 307–322.

Alexander, G., Williams, D., 1971. Heat stress and development of the conceptus in domestic sheep.

J. Agr. Sci. 76, 53–72.

Anderson, A.H., Fennessey, P.V., Meschia, G., Wilkening, R.B., Battaglia, F.C., 1997. Placental transport

of threonine and its utilization in the normal and growth-restricted fetus. Amer. J. Physiol. 272,

E892–E900.

Anderson, M.S., Flowers-Ziegler, J., Das, U.G., Hay, W.W. Jr., Devaskar, S.U., 2001a. Glucose trans-

porter protein responses to selective hyperglycemia or hyperinsulinemia in fetal sheep. Amer.

J. Physiol. 281, R1545–R1552.

Anderson, M.S., He, J., Flowers-Ziegler, J., Devaskar, S.U., Hay, W.W. Jr., 2001b. Effects of selective

hyperglycemia and hyperinsulinemia on glucose transporters in fetal ovine skeletal muscle. Amer.

J. Physiol. 281, R1256–R1263.

Anthony, R.V., Pratt, S.L., Liang, R., Holland, M.D., 1995. Placental-fetal hormonal interactions: impact

on fetal growth. J. Anim. Sci. 73, 1861–1871.

Barbera, A., Wilkening, R.B., Teng, C., Battaglia, F.C., Meschia, G., 1997. Metabolic alterations in the

fetal hepatic and umbilical circulations during glucocorticoid-induced parturition in sheep. Pediat.

Res. 41, 242–248.

Barker, D.J.P., 1998. Mothers, Babies and Health in Later Life, 2nd Edition. Churchill Livingstone,

Edinburgh.

Barnes, R.J., Comline, R.S., Silver, M., 1978. Effect of cortisol on liver glycogen concentrations in

hypophysectomized, adrenalectomized and normal foetal lambs during late or prolonged gestation.

J. Physiol. 275, 567–579.

Bassett, J.M., Hanson, C., 1998. Catecholamines inhibit growth in fetal sheep in the absence of hypox-

emia. Amer. J. Physiol. 274, R1536–R1545.

Bassett, J.M., Hanson, C., 2000. Prevention of hypoinsulinemia modifies catecholamine effects in fetal

sheep. Amer. J. Physiol. 278, R1171–R1181.

A. W. Bell et al.24

Bassett, J.M., Thorburn, G.D., Wallace, A.L.C., 1970. The plasma growth hormone concentration of the

foetal lamb. J. Endocrinol. 48, 251–263.

Bassett, N.S., Oliver, M.H., Breier, B.H., Gluckman, P.D., 1990. The effect of maternal starvation on plasma

insulin-like growth factor I concentrations in the late gestation ovine fetus. Pediat. Res. 27, 401–404.

Battaglia, F.C., Meschia, G., 1988. Fetal nutrition. Annu. Rev. Nutr. 8, 43–61.

Battaglia, F.C., Regnault, T.R.H., 2001. Placental transport and metabolism of amino acids. Placenta 22,

145–161.

Bauer, M.K., Harding, J.E., Breier, B.H., Gluckman, P.D., 2000. Exogenous GH infusion to late-gestational

fetal sheep does not alter fetal growth and metabolism. J. Endocrinol. 166, 591–597.

Bell, A.W., 1993. Pregnancy and fetal metabolism. In: Forbes, J.M., France, J. (Eds.), Quantitative

Aspects of Ruminant Digestion and Metabolism. CAB International, Oxford, pp. 406–431.

Bell, A.W., Battaglia, F.C., Meschia, G., 1987a. Relation between metabolic rate and body size in the

ovine fetus. J. Nutr. 117, 1181–1186.

Bell, A.W., Ehrhardt, R.A., 2002. Regulation of placental nutrient transport and implications for fetal

growth. Nutr. Res. Rev. 15, 211–230.Bell, A.W., Hay, W.W. Jr., Ehrhardt, R.A., 1999. Placental transport of nutrients and its implications for

fetal growth. J. Reprod. Fertil., Suppl. 54, 401–410.

Bell, A.W., Kennaugh, J.M., Battaglia, F.C., Makowski, E.L., Meschia, G., 1986. Metabolic and circula-

tory studies of fetal lamb at midgestation. Amer. J. Physiol. 250, E538–E544.

Bell, A.W., Kennaugh, J.M., Battaglia, F.C., Meschia, G., 1989. Uptake of amino acids and ammonia at

mid-gestation by the fetal lamb. Quart. J. Exp. Physiol. 74, 635–643.

Bell, A.W., Slepetis, R., Ehrhardt, R.A., 1995. Growth and accretion of energy and protein in the gravid

uterus during late pregnancy in Holstein cows. J. Dairy Sci. 78, 1954–1961.

Bell, A.W., Wilkening, R.B., Meschia, G., 1987b. Some aspects of placental function in chronically heat-

stressed ewes. J. Dev. Physiol. 9, 17–29.

Boisclair, Y.R., Rhoads, R.P., Ueki, I., Wang, J., Ooi, G.T., 2001. The acid-labile subunit (ALS) of the

150 kDa IGF-binding protein complex: an important but forgotten component of the circulating IGF

system. J. Endocrinol. 170, 63–70.

Campbell, F.M., Gordon, M.J., Dutta-Roy, A.K., 1994. Plasma membrane fatty acid-binding protein

(FABPpm) from the sheep placenta. Biochim. Biophys. Acta 1214, 187–192.

Campbell, F.M., Taffesse, S., Gordon, M.J., Dutta-Roy, A.K., 1995. Plasma membrane fatty acid-binding

protein from human placenta: identification and characterisation. Biochem. Biophys. Res. Comm.

209, 1011–1017.

Carter, B.S., Moores, R.R., Jr., Teng, C., Meschia, G., Battaglia, F.C., 1995. Main routes of plasma lac-

tate carbon disposal in the midgestation fetal lamb. Biol. Neonate 67, 295–300.

Carver, T.D., Quick, A.A., Teng, C., Pike, A.W., Fennessey, P.V., Hay, W.W. Jr., 1997. Leucine metabo-

lism in chronically hypoglycemic hypoinsulinemic growth-restricted fetal sheep. Amer. J. Physiol.

272, E107–E117.

Char, V.C., Creasy, R.K., 1976. Acetate as a metabolic substrate in the fetal lamb. Amer. J. Physiol. 230,

357–361.

Chung, M., Teng, C., Timmerman, M., Meschia, G., Battaglia, F.C., 1998. Production and utilization of

amino acids by ovine placenta in vivo. Amer. J. Physiol. 274, E13–E22.

Clegg, R.A., 1981. Placental lipoprotein lipase activity in the rabbit, rat and sheep. Comp. Biochem.

Physiol. 69B, 585–591.

Cohen, W.R., Piasecki, G.J., Jackson, B.T., 1982. Plasma catecholamines during hypoxemia in fetal lamb.

Amer. J. Physiol. 243, R520–R525.

Comline, R.S., Silver, M., 1976. Some aspects of foetal and utero-placental metabolism in cows with

indwelling umbilical and uterine vascular catheters. J. Physiol. 260, 571–586.


Constancia, M., Hemberger, M., Hughes, J., Dean, W., Ferguson-Smith, A., Fundeles, R., Stewart, F.,

Kelsey, G., Fowden, A., Sibley, C., Reik, W., 2002. Placental-specific IGF-II is a major modulator of

placental and fetal growth. Nature 417, 945–948.

Creasy, R.K., Barrett, C.T., de Swiet, M., Kahanpää, K.V., Rudolph, A.M., 1972. Experimental intrauterine

growth retardation in the sheep. Amer. J. Obstet. Gynecol. 112, 566–573.

Currie, M.J., Bassett, N.S., Gluckman, P.D., 1997. Ovine glucose transporter-1 and -3: cDNA partial

sequences and developmental gene expression in the placenta. Placenta 18, 393–401.

D’Agostino, J., Field, J.B., Frazier, M.L., 1985. Ontogeny of immunoreactive insulin in the fetal bovine

pancreas. Endocrinology 116, 1108–1116.

Das, U.G., He, J., Ehrhardt, R.A., Hay, W.W. Jr., Devaskar, S.U., 2000. Time-dependent physiological reg-

ulation of ovine placental GLUT-3 glucose transporter protein. Amer. J. Physiol. 279, R2252–R2261.

Das, U.G., Sadiq, H.F., Soares, M.J., Hay, W.W. Jr., Devaskar, S.U., 1998. Time-dependent physiological

regulation of rodent and ovine placental glucose transporter (GLUT-1) protein. Amer. J. Physiol. 274,

R339–R347.

DeChiara, T.M., Robertson, E.J., Efstratiadis, A., 1991. Parental imprinting of the mouse insulin-like

growth factor II gene. Cell 64, 849–859.

Delhanty, P.J., Han, V.K., 1993. The expression of insulin-like growth factor (IGF)-binding protein-2 and

IGF-II genes in the tissues of the developing ovine fetus. Endocrinology 132, 41–52.

Devaskar, S.U., Anthony, R., Hay, W. Jr., 2002. Ontogeny and insulin regulation of fetal ovine white

adipose tissue leptin expression. Amer. J. Physiol. 282, R431–R438.

de Zegher, F., Styne, D.M., Daaboul, J., Bettendorf, M., Kaplan, S.L., Grumbach, M.M., 1989. Hormone

ontogeny in the ovine fetus and neonatal lamb. XX. Effect of age, breeding season, and twinning on

the growth hormone (GH) response to GH-releasing factor: evidence for a homeostatic role of fetal

GH. Endocrinology 124, 124–128.

Dickson, M.C., Saunders, J.C., Gilmour, R.S., 1991. The ovine insulin-like growth factor-I gene:

characterization, expression and identification of a putative promoter. J. Mol. Endocrinol. 6, 17–31.

DiGiacomo, J.E., Hay, W.W. Jr., 1990a. Placental-fetal glucose exchange and placental glucose

consumption in pregnant sheep. Amer. J. Physiol. 258, E360–E367.

DiGiacomo, J.E., Hay, W.W. Jr., 1990b. Fetal glucose metabolism and oxygen consumption during

sustained hypoglycemia. Metabolism 39, 193–202.

Dodic, M., May, C.N., Wintour, E.M., Coghlan, J.P., 1998. An early prenatal exposure to excess gluco-

corticoid leads to hypertensive offspring in sheep. Clin. Sci. 94, 149–155.

Dodic, M., Peers, A., Coghlan, J.P., May, C.N., Lumbers, E., Yu, Z.-Y., Wintour, E.M., 1999. Altered car-

diovascular haemodynamics and baroreceptor-heart rate reflex in adult sheep after prenatal exposure

to dexamethasone. Clin. Sci. 97, 103–109.

Dodic, M., Peers, A., Moritz, K., Hantzis, V., Wintour, E.M., 2002. No evidence for HPA reset in adult

sheep with high blood pressure due to short prenatal exposure to dexamethasone. Amer. J. Physiol.

282, R343–R350.

Ehrhardt, R.A., Bell, A.W., 1995. Growth and metabolism of the ovine placenta during mid-gestation.

Placenta 16, 727–741.

Ehrhardt, R.A., Bell, A.W., 1997. Developmental increases in glucose transporter concentration in the

sheep placenta. Amer. J. Physiol. 273, R1132–R1141.

Ehrhardt, R.A., Bell, A.W., Boisclair, Y.R., 1999. Analysis of leptin and leptin receptor mRNA expres-

sion in the sheep placenta. J. Anim. Sci., 77, Suppl. 1, 159.

Ehrhardt, R.A., Bell, A.W., Boisclair, Y.R., 2002. Spatial and developmental regulation of leptin in fetal

sheep. Amer. J. Physiol. 282, R1628–R1635.

Ehrhardt, R.A., Boston, R.C., Slepetis, R.M., Bell, A.W., 1996. Moderate maternal undernutrition

elevates placental glucose transport capacity in sheep. J. Anim. Sci. 74, Suppl. 1, 152.

A. W. Bell et al.26

Ehrhardt, R.A., Greenwood, P.L., Bell, A.W., Boisclair, Y.R. 2003. Plasma leptin is regulated predomi-

nantly by nutrition in preruminant lambs. J. Nutr. 133, 4196–4201.

Ehrhardt, R.A., Slepetis, R.M., Bell, A.W., 1998. Moderate maternal undernutrition alters glucose

transporter levels in maternal insulin-responsive tissues and in the placenta. J. Anim. Sci. 76,

Suppl. 1, 130.

Etherton, T.D., Bauman, D.E., 1998. Biology of somatotropin in growth and lactation of domestic

animals. Physiol. Rev. 78, 745–761.

Faichney, G.J., White, G.A., 1987. Effects of maternal nutritional status on fetal and placental growth and

on fetal urea synthesis in sheep. Aust. J. Biol. Sci. 40, 365–377.

Ferrell, C.L., 1991. Maternal and fetal influences on uterine and conceptus development in the cow: II.

Blood flow and nutrient flux. J. Anim. Sci. 69, 1954–1965.

Ferrell, C.L., Garrett, W.N., Hinman, N., 1976. Growth, development and composition of the udder and

gravid uterus of beef heifers during pregnancy. J. Anim. Sci. 42, 1477–1489.

Florini, J.R., Ewton, D.Z., Coolican, S.A., 1996. Growth hormone and insulin-like growth factor system

in myogenesis. Endocr. Rev. 17, 481–517.

Forhead, A.J., Li, J., Saunders, J.C., Dauncey, M.J., Gilmour, R.S., Fowden, A.L., 2000. Control of ovine

hepatic growth hormone receptor and insulin-like growth factor I by thyroid hormones in utero. Amer.

J. Physiol. 278, E1166–E1174.

Forhead, A.J., Thomas, L., Crabtree, J., Hoggard, N., Gardner, D.S., Giussani, D.A., Fowden, A.L., 2002.

Plasma leptin concentration in fetal sheep during late gestation: ontogeny and effect of glucocorti-

coids. Endocrinology 143, 1166–1173.

Fowden, A.L., 1997. Comparative aspects of fetal carbohydrate metabolism. Equine Vet. J. 24, 19–25.

Fowden, A.L., Apatu, R.S.K., Silver, M., 1995. The glucogenic capacity of the fetal pig: developmental

regulation by cortisol. Exp. Physiol. 80, 457–467.

Fowden, A.L., Forhead, A.J., Silver, M., Macdonald, A.A., 1997. Glucose, lactate and oxygen metabo-

lism in the fetal pig during late gestation. Exp. Physiol. 82, 171–182.

Fowden, A.L., Forhead, White, K.L., Taylor, P.M., 2000b. Equine uteroplacental metabolism at mid- and

late gestation. Exp. Physiol. 85, 539–545.

Fowden, A.L., Hay, W.W. Jr. 1988. The effects of pancreatectomy on the rates of glucose utilization,

oxidation and production in the sheep fetus. Quart. J. Exp. Physiol. 73, 973–984.

Fowden, A.L., Li, J., Forhead, A.L., 1998. Glucocorticoids and the preparation for life after birth: are

there long-term consequences of the life insurance? Proc. Nutr. Soc. 57, 113–122.

Fowden, A.L., Mapstone, J., Forhead, A.J., 2001. Regulation of glucogenesis by thyroid hormones in

fetal sheep during late gestation. J. Endocrinol. 170, 461–469.

Fowden, A.L., Silver, M., 1995. The effect of thyroid hormones on oxygen and glucose metabolism in

the sheep fetus during late gestation. J. Physiol. 482, 203–213.

Fowden, A.L., Taylor, P.M., White, K.L., Forhead, A.J., 2000a. Ontogenic and nutritionally induced

changes in fetal metabolism in the horse. J. Physiol. 528, 209–219.

Freemark, M., Handwerger, S., 1984. Ovine placental lactogen stimulates glycogen synthesis in fetal rat

hepatocytes. Amer. J. Physiol. 246, E21–E24.

Freemark, M., Handwerger, S., 1986. The glycogenic effects of placental lactogen and growth hormone

in ovine fetal liver are mediated through binding to specific fetal ovine placental lactogen receptors.

Endocrinology 118, 613–618.

Gatford, K.L., Clarke, I.J., De Blasio, M.J., McMillen, I.C., Robinson, J.S., Owens, J.A., 2002. Perinatal

growth and plasma GH profiles in adolescent and adult sheep. J. Endocrinol. 173, 151–159.

Gatford, K.L., Wintour, E.M., De Blasio, M.J., Owens, J.A., Dodic, M., 2000. Differential timing for pro-

gramming of glucose homeostasis, sensitivity to insulin and blood pressure by in utero exposure to

dexamethasone in sheep. Clin. Sci. 98, 553–560.


Gertler, A., Djiane, J., 2002. Mechanism of ruminant placental lactogen action: molecular and in vivo

studies. Mol. Genet. Metab. 75, 189–201.

Gluckman, P.D., Butler, J.H., Comline, R., Fowden, A., 1987. The effects of pancreatectomy on the plasma

concentrations of insulin-like growth factors 1 and 2 in the sheep fetus. J. Dev. Physiol. 9, 79–88.

Gluckman, P.D., Mueller, P.L., Kaplan, S.L., Rudolph, A.M., Grumbach, M.M., 1979. Hormone

ontogeny in the ovine fetus. I. Circulating growth hormone in mid and late gestation. Endocrinology

104, 162–168.

Goalstone, M.L., Leitner, J.W., Wall, K., Dolgonos, L., Rothen, K.I., Accili, D., Draznin, B., 1998. Effect

of insulin on farnesyltransferase: specificity of insulin action and potentiation of nuclear effects of

insulin-like growth factor-1, epidermal growth factor, and platelet-derived growth factor. J. Biol.

Chem. 273, 23892–23896.

Grajwer, L.A., Sperling, M.A., Sack, J., Fisher, D.A., 1977. Possible mechanisms and significance of the

neonatal surge in glucagon secretion: studies in newborn lambs. Pediat. Res. 11, 833–836.

Greenwood, P.L, Bell, A.W., 2003. Consequences of intrauterine growth retardation for postnatal growth,

metabolism and pathophysiology. Reproduction, Suppl. 61, 195–206.

Greenwood, P.L., Hunt A.S., Hermanson, J.W., Bell, A.W., 1998. Effects of birth weight and postnatal

nutrition on neonatal sheep: I. Body growth and composition, and some aspects of energetic

efficiency. J. Anim. Sci. 76, 2354–2367.

Greenwood, P.L., Hunt, A.S., Slepetis, R.M., Finnerty, K.D., Alston, C., Beermann, D.H., Bell, A.W.,

2002. Effects of birth weight and postnatal nutrition on neonatal sheep: III. Regulation of energy

metabolism. J. Anim. Sci. 80, 2850–2861.

Greenwood, P.L., Slepetis, R.M., Bell, A.W., 2000. Influences on fetal and placental weights during mid to

late gestation in prolific ewes well nourished throughout pregnancy. Reprod. Fertil. Dev. 12, 149–156.

Greenwood, P.L., Slepetis, R.M., Hermanson, J.W., Bell, A.W., 1999. Intrauterine growth retardation is

associated with reduced cell cycle activity, but not myofibre number, in ovine fetal muscle. Reprod.

Fertil. Dev. 11, 281–291.

Hales, C.N., Desai, M., Ozanne, S.E., Crowther, N.J., 1996. Fishing in the stream of diabetes: from meas-

uring insulin to the control of fetal organogenesis. Biochem. Soc. Trans. 24, 341–350.

Han, V.K., Lund, P.K., Lee, D.C., D’Ercole, A.J., 1988. Expression of somatomedin/insulin-like growth

factor messenger ribonucleic acids in the human fetus: identification, characterization, and tissue

distribution. J. Clin. Endocrinol. Metab. 66, 422–429.

Harding, J.E., Jones, C.T., Robinson, J.S., 1985. Studies on experimental growth retardation in sheep: the

effects of a small placenta in restricting transport to and growth of the fetus. J. Dev. Physiol. 7, 427–442.

Harding, J.E., Liu, L., Evans, P.C., Gluckman, P.D., 1994. Insulin-like growth factor I alters feto-placental

protein and carbohydrate metabolism in fetal sheep. Endocrinology 134, 1509–1514.

Harwell, C.M., Padbury, J.F., Anand, R.S., Martinez, A.M., Ipp, E., Thio, L., Burnell, E.E., 1990. Fetal

catecholamine responses to maternal hypoglycemia. Amer. J. Physiol. 259, R1126–R1130.

Hawkins, P., Steyn, C., McGarrigle, H.H.G., Calder, N.A., Saito, T., Stratford, L.L., Noakes, D.E.,

Hanson, M.A., 2000. Cardiovascular and hypothalamic-pituitary-adrenal axis development in late

gestation fetal sheep and young lambs following modest maternal nutrient restriction in early gestation.

Reprod. Fertil. Dev. 12, 443–456.

Hay, W.W. Jr., 1995. Regulation of placental metabolism by glucose supply. Reprod. Fertil. Dev. 7, 365–375.

Hay, W.W., 1998. Fetal requirements, placental transfer of nitrogenous compounds. In: Polin, R.A., Fox,

W.W (Eds.), Fetal, Neonatal Physiology, 2nd Edition. Saunders, Philadelphia, pp. 619–634.

Hay, W.W. Jr., Meznarich, H.K., 1986. The effect of hyperinsulinaemia on glucose utilization and oxidation

and on oxygen consumption in the fetal lamb. Quart. J. Exp. Physiol. 71, 689–698.

Hay, W.W. Jr., Meznarich, H.K., 1988. Effect of maternal glucose concentration on uteroplacental glucose

consumption and transfer in pregnant sheep. Proc. Soc. Exp. Biol. Med. 190, 63–69.

A. W. Bell et al.28

Hay, W.W. Jr., Meznarich, H.K., DiGiacomo, J.E., Hirst, K., Zerbe, G., 1988. Effects of insulin and glucose

concentrations on glucose utilization in fetal sheep. Pediat. Res. 23, 381–387.

Hay, W.W. Jr., Molina, R.D., DiGiacomo, J.E., Meschia, G., 1990. Model of placental glucose consump-

tion and glucose transfer. Amer. J. Physiol. 258, R569–R577.

Hay, W.W. Jr., Myers, S.A., Sparks, J.W., Wilkening, R.B., Meschia, G., Battaglia, F.C., 1983. Glucose

and lactate oxidation rates in the fetal lamb. Proc. Soc. Exp. Biol. Med. 173, 553–563.

Hay, W.W. Jr., Sparks, J.W., Wilkening, R.B., Battaglia, F.C., Meschia, G., 1984. Fetal glucose

uptake and utilization as functions of maternal glucose concentration. Amer. J. Physiol. 246,

E237–EE242.

Hodgson, J.C., Mellor, D.J., Field, A.C., 1982. Foetal and maternal rates of urea production and disposal

in well-nourished and undernourished sheep. Brit. J. Nutr. 48, 49–58.

Holzman, I.R., Lemons, J.A., Meschia, G., Battaglia, F.C., 1977. Ammonia production by the pregnant

uterus. Proc. Soc. Exp. Biol. Med. 156, 27–30.

Hooper, S.B., Bocking, A.D., White, S.E., Fraher, L.J., McDonald, T.J., Han, V.K., 1994. Catecholamines

stimulate the synthesis and release of insulin-like growth factor binding protein-1 (IGFBP-1) by fetal

sheep liver in vivo. Endocrinology 134, 1104–1112.

Hopkins, P.S., Thorburn, G.D., 1972. The effects of fetal thyroidectomy on the development of the ovine

fetus. J. Endocrinol. 54, 55–66.

Horn, J., Stern, M.D.R., Young, M., Noakes, D.E., 1983. Influence of insulin and substrate concentration

on protein synthetic rate in fetal tissues. Res. Vet. Sci. 35, 35–41.

Huxley, R., Neil, A., Collins, R., 2002. Unravelling the fetal origins hypothesis: is there really an inverse

association between birthweight and subsequent blood pressure? Lancet 360, 659–665.

Illsley, N.P., 2000. Glucose transporters in the human placenta. Placenta 21, 14–22.

Jodarski, G.D., Shanahan, M.F., Rankin, J.H.G., 1985. Fetal insulin and placental 3-O-methyl glucose

clearance in near-term sheep. J. Dev. Physiol. 7, 251–258.

Jones, C.T., Ritchie, J.W., Walker, D., 1983. The effects of hypoxia on glucose turnover in the fetal sheep.

J. Dev. Physiol. 5, 223–235.

Jones, J.I., Clemmons, D.R., 1995. Insulin-like growth factors and their binding proteins: biological

actions. Endocr. Rev. 16, 3–34.

Jozwik, M., Teng, C., Battaglia, F.C., Meschia, G., 1999. Fetal supply of amino acids and amino

nitrogen after maternal infusion of amino acids in pregnant sheep. Amer. J. Obstet. Gynecol. 180,

447–453.

Jozwik, M., Teng, C., Timmerman, M., Chung, M., Meschia, G., Battaglia, F.C., 1998. Uptake and trans-

port by the ovine placenta of neutral nonmetabolizable amino acids with different transport system

affinities. Placenta 19, 531–538.

Kennaugh, J.M., Bell, A.W., Teng, C., Meschia, G., Battaglia, F.C., 1987. Ontogenetic changes in the

rates of protein synthesis and leucine oxidation during fetal life. Pediat. Res. 22, 688–692.

Klempt, M., Bingham, B., Breier, B.H., Baumbach, W.R., Gluckman, P.D., 1993. Tissue distribution

and ontogeny of growth hormone receptor messenger ribonucleic acid and ligand binding to hepatic

tissue in the midgestation sheep fetus. Endocrinology 132, 1071–1077.

Krishnamurti, C.R., Schaefer, A.L., 1984. Effect of acute maternal starvation on tyrosine metabolism and

protein synthesis in fetal sheep. Growth 48, 391–403.

Langley, S.C., Browne, R.F., Jackson, A.A., 1994. Altered glucose tolerance in rats exposed to maternal

low protein diets in utero. Comp. Biochem. Physiol. 109A, 223–229.

Langley, S.C., Jackson, A.A., 1994. Increased systolic blood pressure in adult rats induced by fetal

exposure to maternal low protein diet. Clin. Sci. 86, 217–222.

Langley-Evans, S.C., 2001. Fetal programming of cardiovascular function through exposure to maternal

undernutrition. Proc. Nutr. Soc. 60, 505–513.


Lee, C.Y., Bazer, F.W., Etherton, T.D., Simmen, F.H., 1991. Ontogeny of insulin-like growth factors

(IGF-I and IGF-II) and IGF-binding proteins in porcine serum during fetal and postnatal develop-

ment. Endocrinology 128, 2336–2344.

Lee, C.Y., Chung, C.S., Simmen, F.A., 1993. Ontogeny of the porcine insulin-like growth factor system.

Mol. Cell. Endocrinol. 93, 71–80.

Leibovich, H., Gertler, A., Bazer, F.W., Gootwine, E., 2000. Active immunization of ewes against ovine

placental lactogen increases birth weight of lambs and milk production with no adverse effect on

conception rate. Anim. Reprod. Sci. 64, 33–47.

Lemons, J.A., Schreiner, R.L., 1983. Amino acid metabolismin the ovine fetus. Amer. J. Physiol. 244,

E459–E466.

Leury, B.J., Bird, A.R., Chandler, K.D., Bell, A.W., 1990a. Effects of maternal undernutrition and

exercise on glucose kinetics in fetal sheep. Brit. J. Nutr. 64, 463–472.

Leury, B.J., Bird, A.R., Chandler, K.D., Bell, A.W., 1990b. Glucose partitioning in the pregnant ewe:

effects of undernutrition and exercise. Brit. J. Nutr. 64, 449–462.

Li, J., Forhead, A.J., Dauncey, M.J., Gilmour, R.S., Fowden, A.L., 2002. Control of growth hormone

receptor and insulin-like growth factor-I expression by cortisol in ovine fetal skeletal muscle.

J. Physiol. 541, 581–589.

Li, J., Owens, J.A., Owens, P.C., Saunders, J.C., Fowden, A.L., Gilmour, R.S., 1996. The ontogeny of

hepatic growth hormone receptor and insulin-like growth factor I gene expression in the sheep fetus

during late gestation: developmental regulation by cortisol. Endocrinology 137, 1650–1657.

Li, J., Saunders, J.C., Gilmour, R.S., Silver, M., Fowden, A.L., 1993. Insulin-like growth factor-II mes-

senger ribonucleic acid expression in fetal tissues of the sheep during late gestation: effects of

cortisol. Endocrinology 132, 2083–2089.

Liechty, E.A., Boyle, D.W., Moorehead, H., Lee, W.-H., Bowsher, R.R., Denne, S.C., 1996. Effects of cir-

culating IGF-I on glucose and amino acid kinetics in the ovine fetus. Amer. J. Physiol. 271, E177–E185.

Liechty, E.A., Boyle, D.W., Moorehead, H., Liu, Y.M., Denne, S.C., 1993. Increased fetal glucose

concentration decreases ovine fetal leucine oxidation independent of insulin. Amer. J. Physiol. 265,

E617–E623.

Liechty, E.A., Denne, S., Lemons, J.A., Klein, C.L., 1991a. Effects of glucose infusion on leucine

transamination and oxidation in the ovine fetus. Pediat. Res. 30, 423–429.

Liechty, E.A., Kelley, J., Lemons, J.A., 1991b. Effect of fasting on uteroplacental amino acid metabolism

in the pregnant sheep. Biol. Neonate 60, 207–214.

Liu, J.-L., Baker, J., Perkins, A.S., Roberson, E.J., Efstratiadis, A., 1993. Mice carrying null mutations of

the genes encoding insulin-like growth factor-I and type 1 receptor. Cell 75, 59–72.

Lopez, M.F., Dikkes, P., Zurakowski, D., Villa-Komaroff, L., Majzoub, J.A., 1999. Regulation of hepatic

glycogen in the insulin-like growth factor II-deficient mouse. Endocrinology 140, 1442–1448.

Lorijn, R.H.W., Nelson, J.C., Longo, L.D., 1980. Induced fetal hyperthyroidism: cardiac output and

oxygen consumption. Amer. J. Physiol. 235, H302–H307.

Loy, G.L., Quick, A.N. Jr., Hay, W.W. Jr., Meschia, G., Battaglia, F.W., Fennessey, P.V., 1990.

Fetoplacental deamination and decarboxylation of leucine. Amer. J. Physiol. 259, E492–E497.

McMillen, I.C., Adams, M.B., Ross, J.T., Coulter, C.L., Simonetta, G., Owens, J.A., Robinson, J.S., Edwards,

L.J., 2001. Fetal growth restriction: adaptations and consequences. Reproduction 122, 195–204.

McNeill, D.M., Slepetis, R., Ehrhardt, R.A., Smith, D.M., Bell, A.W., 1997. Protein requirements of

sheep in late pregnancy: partitioning of nitrogen between gravid uterus and maternal tissues. J. Anim.

Sci. 75, 809–816.

Mellor, D.J., Matheson, I.C., Small, J., 1977. Some changes in the composition of maternal and fetal

plasma from chronically catheterised sheep during short periods of reduced feed intake in late

pregnancy. Res. Vet. Sci. 23, 119–121.

A. W. Bell et al.30

Meschia, G., Battaglia, F.C., Hay, W.W. Jr., Sparks, J.W., 1980. Utilization of substrates by the ovine

placenta in vivo. Fed. Proc. 39, 245–249.

Mesiano, S., Young, I.R., Hey, A.W., Browne, C.A., Thorburn, G.D., 1989. Hypophysectomy of the fetal

lamb leads to a fall in the plasma concentration of insulin-like growth factor I (IGF-I), but not

IGF-II. Endocrinology 124, 1485–1491.

Meznarich, H.K., Hay, W.W. Jr., Sparks, J.W., Meschia, G., Battaglia, F.C., 1987. Fructose disposal and

oxidation rates in the ovine fetus. Quart. J. Exp. Physiol. 72, 617–625.

Milley, J.R., 1994. Effects of insulin on ovine fetal leucine kinetics and protein metabolism. J. Clin.

Invest. 93, 1616–1624.

Milley, J.R., 1996. Fetal substrate uptake during increased ovine fetal cortisol concentration. Amer.

J. Physiol. 271, E186–E191.

Molina, R.D., Meschia, G., Battaglia, F.C., Hay, W.W. Jr., 1991. Gestational maturation of placental

glucose transfer capacity in sheep. Amer. J. Physiol. 261, R697–R704.

Moores, R.R. Jr., Vaughn, P.R., Battaglia, F.C., Fennessey, P.V., Wilkening, R.B., Meschia, G., 1994.

Glutamate metabolism in fetus and placenta of late-gestation sheep. Amer. J. Physiol. 267, R89–R96.

Mounzih, K, Qiu, J., Ewart-Toland, A., Chehab, F.F., 1998. Leptin is not necessary for gestation and par-

turition but regulates maternal nutrition via a leptin resistance state. Endocrinology 139, 5259–5262.

Mühlhäusler, B.S., Roberts, C.T., McFarlane, J.R., Kauter, K.G., McMillen, I.C., 2002. Fetal leptin is a

signal of fat mass independent of maternal nutrition in ewes fed at or above maintenance energy

requirements. Biol. Reprod. 67, 493–499.

Noble, R.C., Shand, J.H., Christie, W.W., 1985. Synthesis of C20 and C22 polyunsaturated fatty acids by

the placenta of the sheep. Biol. Neonate 47, 333–338.

Noden, D.M., deLahunta, A., 1985. The Embryology of Domestic Animals: Developmental Mechanisms,

Malformations. Williams & Wilkins, Baltimore.

Osborn, B.H., Fowlkes, J., Han, V.K.M., Freemark, M., 1992. Nutritional regulation of insulin-like

growth factor-binding protein gene expression in the ovine fetus and pregnant ewe. Endocrinology

131, 1743–1750.

Osgerby, J.C., Wathes, D.C., Howard, D., Gadd, T.S., 2002. The effect of maternal undernutrition on

ovine fetal growth. J. Endocrinol. 173, 131–141.

Owens, J. A., Falconer, J., Robinson, J. S., 1986. Effect of restriction of placental growth on umbilical

and uterine blood flows. Amer. J. Physiol. 250, R427–R434.

Owens, J.A., Falconer, J., Robinson, J.S., 1987a. Effect of restriction of placental growth on fetal and

utero-placental metabolism. J. Dev. Physiol. 9, 225–238.

Owens, J.A., Falconer, J., Robinson, J.S., 1987b. Restriction of placental size in sheep enhances

efficiency of placental transfer of antipyrine, 3-O-methyl-D-glucose but not urea. J. Dev. Physiol. 9,

457–464.

Owens, J.A., Kind, K.L., Carbone, F., Robinson, J.S., Owens, P.C., 1994. Circulating insulin-like growth

factors-I and -II and substrates in fetal sheep following restriction of placental growth. J. Endocrinol.

140, 5–13.

Paolini, C.L., Meschia, G., Fennessey, P.V., Pike, A.W., Teng, C., Battaglia, F.C., Wilkening, R.B., 2001.

An in vivo study of ovine placental transport of essential amino acids. Amer. J. Physiol. 280, E31–E39.

Parkes, M.J., Bassett, J.M., 1985. Antagonism by growth hormone of insulin action in fetal sheep.

J. Endocrinol. 105, 379–382.

Philipps, A.F., Dubin, J.W., Matty, P.J., Raye, J.R., 1983. Influence of exogenous glucagon on fetal glucose

metabolism and ketone production. Pediat. Res. 17, 51–56.

Phillips, I.D., Simonetta, G., Owens, J.A., Robinson, J.S., Clarke, I.J., McMillen, I.C., 1996. Placental

restriction alters the functional development of the pituitary-adrenal axis in the sheep fetus during late

gestation. Pediat. Res. 40, 861–866.


Polk, D.H., Wu, S.Y., Wright, C., Reviczky, A.L., Fisher, D.A., 1988. Ontogeny of thyroid hormone effect

on tissue 5′-monoiodinase activity in fetal sheep. Amer. J. Physiol. 254, E337–E341.

Rankin, J.H.G., Jodarski, G., Shanahan, M.R., 1986. Maternal insulin and placental 3-O-methyl glucose

transport. J. Dev. Physiol. 8, 247–253.

Rehfeldt, C., Kuhn, G., Nürnberg, G., Kanitz, E., Schneiders, F., Beyer, M., Nürnberg, K., Ender, K.,

2001. Effects of exogenous somatotropin during early gestation on maternal performance, fetal

growth, and composition traits in sheep. J. Anim. Sci. 79, 1789–1799.

Reynolds, L.P., Ferrell, C.L., Robertson, D.A., Ford, S.P., 1986. Metabolism of the gravid uterus, foetus

and utero-placenta at several stages of gestation in cows. J. Agr. Sci. 106, 437–444.

Rhind, S.M., Robinson, J.J., McDonald, I., 1980. Relationships among uterine and placental factors in

prolific ewes and their relevance to variations in foetal weight. Anim. Prod. 30, 115–124.

Rhoads, R.P., Greenwood, P.L., Bell, A.W., Boisclair, Y.R., 2000a. Organization and regulation of the

gene encoding the sheep acid-labile subunit of the 150-kilodalton insulin-like growth factor-binding

protein complex. Endocrinology 141, 1425–1433.

Rhoads, R.P., Greenwood, P.L., Bell, A.W., Boisclair, Y.R., 2000b. Nutritional regulation of the genes

encoding the acid-labile subunit and other components of the circulating insulin-like growth factor

system in the sheep. J. Anim. Sci. 78, 2681–2689.

Robinson, J.S., Chidzanja, S., Kind, K., Lok, F., Owens, P., Owens, J., 1995. Placental control of fetal

growth. Reprod. Fertil. Dev. 7, 333–344.

Robinson, J.S., Hart, I.C., Kingston, E.J., Jones, C.T., Thorburn, G.D., 1980. Studies on the growth of the

fetal sheep: the effects of reduction of placental size on hormone concentration in fetal plasma. J. Dev.

Physiol. 2, 239–248.

Ross, J.C., Fennessey, P.V., Wilkening, R.B., Battaglia, F.C., Meschia, G., 1996. Placental transport and

fetal utilization of leucine in a model of fetal growth retardation. Amer. J. Physiol. 270, E491–E503.

Rudolph, C.D., Roman, C., Rudolph, A.M., 1989. Effect of cortisol on hepatic gluconeogenesis in the

fetal sheep. J. Dev. Physiol. 11, 219–223.

Schoknecht, P.A., Currie, W.B., Bell, A.W., 1992. Kinetics of placental lactogen in mid- and late-gestation

ovine fetuses. J. Endocrinol. 188, 95–100.

Schoknecht, P.A., McGuire, M.A., Cohick, W.S., Currie, W.B., Bell, A.W., 1996. Effect of chronic

infusion of placental lactogen on ovine fetal growth in late gestation. Domest. Anim. Endocrinol.

13, 519–528.

Schreiner, R.L., Nolen, P.A., Bonderman, P.W., Moorehead, H.C., Gresham, E.L., Lemons, J.A.,

Escobedo, M.B., 1980. Fetal and maternal hormonal response to starvation in the ewe. Pediat. Res. 14,

103–108.

Shen, W.-H., Yang, X., Boyle, D.W., Lee, W.H., Liechty, E.A., 2001. Effects of intravenous insulin-like

growth factor-I and insulin administration on insulin-like growth factor binding proteins in the ovine

fetus. J. Endocrinol. 171, 143–151.

Silver, M., Fowden, A.L., Taylor, P.M., Knox, J., Hill, C.M., 1994. Blood amino acids in the pregnant

mare and fetus: the effects of maternal fasting and intrafetal insulin. Exp. Physiol. 79, 423–433.

Simmons, M.A., Battaglia, F.C., Meschia, G., 1979. Placental transfer of glucose. J. Dev. Physiol. 1, 227–243.

Slotkin, T.A., Seidler, F.J., 1988. Adrenomedullary catecholamine release in the fetus and newborn:

secretory mechanisms and their role in stress and survival. J. Dev. Physiol. 10, 1–16.

Smeaton, T.C., Owens, J.A., Kind, K.L., Robinson, J.S., 1989. The placenta releases branched-chain amino

acids into the umbilical and uterine circulations in the pregnant sheep. J. Dev. Physiol. 12, 95–99.

Sparks, J.W., Hay, W.W. Jr., Bonds, D., Meschia, G., Battaglia, F.C., 1982. Simultaneous measurements

of lactate turnover rate and umbilical lactate uptake in the fetal lamb. J. Clin. Invest. 70, 179–192.

Stegeman, J.H.G., 1974. Placental development in the sheep and its relation to fetal development. Bjd. Dierk.

44, 1–72.

A. W. Bell et al.32

Stephens, E., Thureen, P.J., Goalstone, M.L., Serson, M., Leitner, J.W., Hay, W.W. Jr., Draznin, B., 2001.

Fetal hyperinsulinemia increases farnesylation of p21 Ras in fetal tissues. Amer. J. Physiol. 281,

E217–E223.

Sterle, J.A., Cantley, T.C., Lamberson, W.R., Lucey, M.C., Gerrard, D.E., Matteri, R.L., Day, B.N., 1995.

Effects of recombinant porcine somatotropin on placental size, fetal growth, and IGF-I and IGF-II

concentrations in pigs. J. Anim. Sci. 73, 2980–2985.

Stevens, D., Alexander, G., 1986. Lipid deposition after hypophysectomy and growth hormone treatment

in the sheep fetus. J. Dev. Physiol. 8, 139–145.

Stevens, D., Alexander, G., Bell, A.W., 1990. Effect of prolonged glucose infusion into fetal sheep on

body growth, fat deposition and gestation length. J. Dev. Physiol. 13, 277–281.

Teasdale, F., 1976. Numerical density of nuclei in the sheep placenta. Anat. Rec. 185, 187–196.

Thomas, L., Wallace, J.M., Aitken, R.P., Mercer, J.G., Trayhurn, P., 2001. Circulating leptin during ovine

pregnancy in relation to maternal nutrition, body composition and pregnancy outcome. J. Endocrinol.

169, 465–476.

Thureen, P.J., Merson, S., Hay, W.W. Jr., 2000. Regulation of uterine and umbilical amino acid uptakes

by maternal amino acid concentrations. Amer. J. Physiol. 279, R849–R859.

Thureen, P.J., Padbury, J.F., Hay, W.W. Jr., 2001. The effect of maternal hypoaminoacidaemia on

placental uptake and transport of amino acids in pregnant sheep. Placenta 22, 162–170.

Thureen, P., Trembler, K.A., Meschia, G., Makowski, E.L., Wilkening, R.B., 1992. Placental glucose

transport in heat-induced fetal growth retardation. Amer. J. Physiol. 263, R578–R585.

Timmerman, M., Chung, M., Wilkening, R.B., Fennessey, P.V., Battaglia, F.C., Meschia, G., 1998.

Relationship of fetal alanine uptake and placental alanine metabolism to maternal plasma alanine

concentration. Amer. J. Physiol. 275, E942–E950.

Timmerman, M., Teng, C., Wilkening, R.B., Fennessey, P., Battaglia, F.C., Meschia, G., 2000. Effect of

dexamethasone on fetal hepatic glutamine-glutamate exchange. Amer. J. Physiol. 278, E839–E845.

Townsend, S.F., Rudolph, C.D., Rudolph, A.M., 1991. Cortisol induces perinatal hepatic gluconeogene-

sis in the lamb. J. Dev. Physiol. 16, 71–79.

Van Veen, L.C.P., Teng, C., Hay, W.W., Jr., Meschia, G., Battaglia, F.C., 1987. Leucine disposal and

oxidation rates in the fetal lamb. Metabolism 36, 48–53.

Van Vliet, G., Styne, D.M., Kaplan, S.L., Grumbach, M.M., 1983. Hormone ontogeny in the ovine fetus.

XVI. Plasma immunoreactive somatomedin C/insulin-like growth factor I in the fetal and neonatal

lamb and in the pregnant ewe. Endocrinology 113, 1716–1720.

Vatnick, I., Bell, A.W., 1992. Ontogeny of fetal hepatic and placental growth and metabolism in sheep.

Amer. J. Physiol. 263, R619–R623.

Vaughn, P.R., Lobo, C., Battaglia, F.C., Fennessey, P.V., Wilkening, R.B., Meschia, G., 1995. Glutamine-

glutamate exchange between placenta and fetal liver. Amer. J. Physiol. 268, E705–E711.

Vernon, R.G., Robertson, J.P., Clegg, R.A., Flint, D.J., 1981. Aspects of adipose-tissue metabolism in

foetal lambs. Biochem. J. 196, 819–824.

Wallace, J.M., Bourke D.A., Aitken, R.P., Leitch, N., Hay, W.W. Jr., 2002. Blood flows and nutrient

uptakes in growth-restricted pregnancies induced by overnourishing adolescent sheep. Amer.

J. Physiol. 282, R1027–R1036.

Wallace, J.M., Bourke, D.A., Aitken, R.P., Palmer, R.M., Da Silva, P., Cruickshank, M.A., 2000.

Relationship between nutritionally-mediated placental growth restriction and fetal growth, body com-

position and endocrine status during late gestation in adolescent sheep. Placenta 21, 100–108.

Ward, J.W., Wooding, F.B.P., Fowden, A.L., 2002. The effects of cortisol on the binucleate cell popula-

tion in the ovine placenta during late gestation. Placenta 23, 451–458.

Whorwood, C.B., Firth, K.M., Budge, H., Symonds, M.E., 2001. Maternal undernutrition during early to

midgestation programs tissue-specific alterations in the expression of the glucocorticoid receptor,


11β-hydroxysteroid dehydrogenase isoforms, and type 1 angiotensin II receptor in neonatal sheep.


Widdas, W.F., 1952. Inability of diffusion to account for placental glucose transfer in the sheep and con-

sideration of the kinetics of a possible carrier transfer. J. Physiol. 118, 23–39.

Wilkening, R.B., Molina, R.D., Battaglia, F.C., Meschia, G., 1987. Effect of insulin on glucose/oxygen

and lactate/oxygen quotients across the hindlimb of fetal lambs. Biol. Neonate 51, 18–23.

Wooding, F.B.P., Morgan, G., Fowden, A.L., Allen, W.R., 2000. Separate sites and mechanisms for

placental transport of calcium, iron and glucose in the equine placenta. Placenta 21, 635–645.

Young, M., McFadyen, I.R., 1973. Placental transfer and fetal uptake of amino acids in the pregnant ewe.

J. Perinatal Med. 1, 174–182.

Yuen, B.S.J., McMillen, I.C., Symonds, M.E., Owens, P.C., 1999. Abundance of leptin mRNA in fetal

adipose tissue is related to fetal body weight. J. Endocrinol. 163, R11–R14.

Yuen, B.S.J., Owens, P.C., McFarlane, J.R., Symonds, M.E., Edwards, L.J., Kauter, K.G., McMillen, I.C.,

2002. Circulating leptin concentrations are positively related to leptin messenger RNA expression in

adipose tissue of fetal sheep in the pregnant ewe fed at or below maintenance energy requirements

during late gestation. Biol. Reprod. 67, 911–916.

Yuen, B.S.J., Owens, P.C., Mühlhäusler, B.S., Roberts, C.T., Symonds, M.E., Keisler, D.H., McFarlane,

J.R., Kauter, K.G., Evens, Y., McMillen, I.C., 2003. Leptin alters the structural and functional char-

acteristics of adipose tissue before birth. FASEB J. Express article 10.1096/fj.02–0756fje. Published

online, April 22, 2003.

Zhou, J., Bondy, C.A., 1993. Placental glucose transporter gene expression and metabolism in the rat.

J. Clin. Invest. 91, 845–852.

A. W. Bell et al.34

PART IIProtein metabolism

37

2 Regulation of skeletal muscle proteinmetabolism in growing animals

T. A. Davis and M. L. Fiorotto

United States Department of Agriculture/Agricultural Research Service,Children’s Nutrition Research Center, Department of Pediatrics,Baylor College of Medicine, Houston, TX 77030, USA

The skeletal musculature is not only of great significance to the physiological function andlong-term well-being of the growing animal, but, by virtue of its large mass, has tremendousimpact on the overall rate of protein metabolism in the whole body. Protein deposition is veryrapid during early life and this is largely driven by the high fractional rate of protein synthesisin skeletal muscle. A number of factors regulate the growth, development, and metabolicactivity of the skeletal musculature, and these include the intrinsic or genetic factors thatinfluence muscle differentiation as well as the extrinsic factors such as nutrients, hormones,and activity that influence muscle hypertrophy.

1. INTRODUCTION

In the mature adult, the skeletal musculature constitutes the largest single protein pool in thebody, and comprises approximately 60% of the body’s metabolically active mass. Thus,despite its relatively low basal rate of metabolism, skeletal muscle mass is such that changesto its composition and/or its size have implications for the overall metabolism of the body.Until relatively recently, the primary interests in skeletal muscle metabolism were related to itsfunctional role in dictating locomotor performance, specifically speed, strength, and endurance,and its influence on the quantity and quality of meat products that constitute a primary sourceof protein and micronutrients in the human diet. More recently, however, a renewed interestin the contribution of the muscle metabolism to the overall health of the human individual hasemerged. This interest, together with the developments in our understanding of the regulationof gene expression and cellular signalling, have spurred substantial amounts of research toadvance our understanding of how muscle mass, metabolism, and function respond to nutrients,hormones, growth factors, activity, and other anabolic agents.

The fully differentiated skeletal muscle is made up of multinucleated myofibres. The post-natal growth rate of muscle mass is a function of the total number of fibres, and the growth



rate of each fibre. Thus, understanding the regulation of myofibre formation during prenatal life,and the rate of muscle protein accretion in postnatal life, are critical for evaluating the animal’smaximal capacity for muscle growth. The mature myofibre is composed primarily of the threebasic systems required for muscle contraction: the myofibrillar proteins composed primarily ofthe contractile elements and the associated structural proteins; an extensive membrane systemthat regulates the release and uptake of ions in response to the neural inputs; and the mitochon-drial and cytoplasmic system of enzymes involved in the generation of the ATP required to drivefunctional processes. These components vary in their relative abundance, as well as in their levelof activity, thereby giving rise to substantial diversity in fibre function and size.

The combination of functional and metabolic properties of fibres is used as the basis forthe standard classification of muscle fibres in the adult. Muscle fibres can be divided intotwo main categories on the basis of their twitch characteristics, that is, slow-twitch (S) orfast-twitch (F), which correspond with the Type I and Type II nomenclature, respectively. Thecontractile property of the myofibrils is largely determined by the ATPase activity of themyosin heavy chain (MHC) isoform expressed within each fibre (Schiaffino and Reggiani,1996). Fibres are also identified in a variety of ways by their metabolic properties: slow-twitchfibres, and a subcategory of fast-twitch fibres, generate ATP predominantly by oxidativemetabolism (O). All fast-twitch fibres also can generate limited amounts of ATP by glycolysisand, hence, they are categorized as glycolytic (FG or Type IIB) or fast-twitch, oxidative-glycolytic (FOG or Type IIA). The capacity for oxidative metabolism is supported by a rela-tively high abundance of mitochondria, enzymes for fatty acid oxidation, and oxygen deliveryand uptake. The latter is effected by a high capillary density, and the presence of large amountsof myoglobin in the sarcoplasm, both of which impart a red tint to the muscle, so that SO andFOG fibres also are referred to as red muscle. Because these metabolic properties render tothe fibres a greater degree of fatigue-resistance, they constitute primarily those muscles thatundergo prolonged periods of sustained slow isometric contraction, such as postural muscles(e.g. soleus and rhomboides), or muscles that are required for continuous episodes of isotoniccontraction (e.g. diaphragm and jaw muscles). In contrast, FG fibres have a paucity of thesecomponents and thus, muscles where these fibres predominate (e.g. longissimus dorsi) aremuch lighter in colour and are referred to as white muscles.

The origin of myofibre heterogeneity is complex and not entirely understood. There isevidence that the various fibre types are derived from distinct myoblast lineages. However,within an organism, the relative abundance of different fibre types varies between musclesaccording to their physiological function. There is also variability between individuals of thesame breed, between different breeds of the same species, and between different species(Rehfeldt et al., 1999). Fibre-type proportions are of relevance with respect to livestock, as theyare a key determinant of meat quality (Koohmaraie, 2003). Fibre diversity has its origins in theembryo, is amplified during the process of differentiation during fetal and early postnatal life,and thereafter is fine-tuned in response to the functional demands placed on the muscle. By andlarge, however, fibre-type composition is a genetically determined, inherited trait. Muscle fibrehypertrophy, on the other hand, occurs after the fibres have differentiated. Hypertrophy islargely a postnatal event and metabolically is dominated by the accretion of muscle-specificproteins. Unlike the early processes of determination, commitment, and differentiation that arenormally orchestrated by signalling molecules and intracellular pathways inherent to the devel-oping organism and are primarily under genetic control, muscle hypertrophy is highlyresponsive to external cues, such as nutrient availability, muscle use, and various hormones.

The aim of this review is to consider the factors that influence muscle protein metabolismin the growing organism. We shall address the “heritable/congenital” component, i.e. the

T. A. Davis and M. L. Fiorotto38

developmental aspects of muscle growth that principally determine fibre number and fibre-type diversity, and the role of external influences that modulate muscle hypertrophy. Owingto space limitations, this review cannot cover all aspects of muscle metabolism in equal depth.Rather, we have selected those areas in which substantial progress has been made recently,and have focused largely on early life, when the most marked changes occur.

2. MUSCLE DIFFERENTIATION

2.1. From myoblast to myofibre

Much of our understanding of early myogenesis is derived from studies on the chick.However, the overall pattern is similar in mammals as indicated by the extensive study of themouse in which the advent of transgenic technology has permitted the function of individualgenes to be identified. Myogenesis begins in early embryonic life and, with the exception ofcraniofacial muscles, the skeletal musculature develops from mesodermal progenitor cells inthe somites (Perry and Rudnicki, 2000; Buckingham, 2001). The somites develop in pairsfrom aggregates of epithelial cells on either side of the neural tube and notochord and maturein a rostro-caudal direction under the regulation of positive signals and negative regulators, inthe form of diffusible molecules, produced by the tissues adjacent to the somites (Buckingham,2001; Buckingham et al., 2003; Francis-West et al., 2003) (fig. 1). Cells of the dorsal surface ofthe somite are compartmentalized into the dermomyotome and are specified to form myogenicand dermal progenitors, whereas signals to cells on the ventral aspect of the somite specifythe formation of the sclerotome which gives rise to the ribs and axial skeleton (Buckingham,2001). Myogenic precursor cells originate from the dorso-medial (epaxial) and ventro-lateral(hypaxial) edges of the dermomyotome. The epaxial precursors delaminate and translocateventrolaterally to form the myotome, under the dermomyotome, and eventually they expandand differentiate to form the deep back muscles. The hypaxial myogenic precursors of thedermomyotome migrate ventrally to form the ventral body wall muscles, tongue, and diaphragm,or delaminate and migrate into the limb buds to give rise to the limb musculature.

Specification of myogenic cells occurs upon the activation of myogenic regulatory factors(MRF), specifically those that encode the basic helix–loop–helix transcription factors Myf5,MyoD, MRF4, and myogenin (Molkentin and Olson, 1996; Perry and Rudwicki, 2000).

Regulation of skeletal muscle protein metabolism 39

Fig. 1. Schematic representation of somite structure and key molecules responsible for myogenic specifica-tion, determination, delamination, and migration.

Myf5 and MyoD are the earliest MRFs to be expressed in the dermomyotome, and their pat-terns of expression appear to be spatially and temporally regulated. Wnt-1 produced in thedorsal neural tubes induces Myf5 expression in the epaxial region, whilst Wnts produced bythe dorsal ectoderm induce MyoD expression by the dermomyotome. Sonic hedgehog (Shh),produced from the notochord and floor plate of the neural tube, also appears to play a role inthe early activation of Myf5 in the determination of the epaxial dermomyotome (Buckinghamet al., 2003). Cells committed to the muscle lineage express the homeobox gene, Pax-3, and itsexpression is most likely activated by Wnts and Shh. During early myogenesis, however, thedifferentiation of cells that express Pax-3 and are committed to the muscle lineage is blockedby the expression of inhibitory factors. These include the growth factor, bone morphogenicprotein-4 (BMP-4), and fibroblast growth factors, both of which are produced by the lateralplate mesoderm. Inhibition of differentiation is critical as it enables the cells to continueproliferation, delaminate, and migrate. In the regions of the limbs, migrating cells do notexpress MRFs until they reach the limbs where, after some delay during which they undergoseveral rounds of replication, they give rise to the appendicular muscles (Buckingham, 2001;Buckingham et al., 2003; Francis-West et al., 2003).

Delamination and migration of muscle precursor cells are crucial steps in myogenesis(Birchmeier and Brohmann, 2000; Francis-West et al., 2003) and require the activation of thec-met receptor by hepatocyte growth factor (HGF, also known as scatter factor) (Scaal et al.,1999; Birchmeier and Brohmann, 2000). Transcription of the c-met gene is activated by Pax-3,and in its absence no limb muscles form, whereas when a constitutively active form isexpressed, there is an overproduction of hypaxial muscles (Epstein et al., 1996). Lbx1 is anothertranscription factor required for migration of the somitic cells, with particular importance forthose cells that give rise to the dorsal muscle masses of the limbs (Brohmann et al., 2000).

Once commitment of cells to the myogenic lineage has occurred, the cells are preventedfrom differentiating further by a variety of gene products produced by the cells themselves, aswell as the cells’ matrix, and a variety of mitogens that promote proliferation (Perry andRudnicki, 2000; Fuchtbauer, 2002). The maintenance of the committed, but not fully differ-entiated, state is a critical determinant of myofibre number and size because it permitsmyoblast proliferation to continue and, hence, to expand the population of cells that can giverise to myofibres (Fuchtbauer, 2002). Indeed, all instances of muscle hypertrophy that areassociated with increased myofibre number have their origin in embryonic and fetal life, andby birth, fibre number is fixed (Rehfeldt et al., 1999). The degree of myoblast proliferationin vivo is the product of the balance between activities of stimulatory and inhibitory factors(Fuchtbauer, 2002). The former include Msx1, basic fibroblast growth factor (bFGF), HGF,and the insulin-like growth factors (IGF)-I and -II. The transforming growth factor (TGF-β)super-family of peptides exert a variety of effects which appear to be species-dependent, butby and large, they inhibit terminal differentiation with either little effect or even inhibitionof proliferation (Fuchtbauer, 2002). Members of the TGF-β family that have been shown toregulate myogenesis include TGF-β itself, activin, BMPs, and growth differentiation factor 8,also known as myostatin (McPherron et al., 1997; Fuchtbauer, 2002). Their precise role andthe incurred responses vary between muscle beds and according to the net balance betweenpositive and negative regulators.

The in vivo significance of these factors in regulating myoblast hyperplasia and fibreformation has been demonstrated in studies with an extensive variety of transgenic animals.Notable among these has been the development of transgenic mice in which the myostatingene is inactivated (McPherron et al., 1997). Myostatin is expressed beginning early inembryonic life and inhibits cell cycle progression, thereby limiting myoblast proliferation


(Lee and McPherron, 1999). Myostatin also inhibits differentiation by down-regulatingMyoD and myogenin expression and activity (Langley et al., 2002). Hence, in its absence,there is greater myoblast proliferation and this ultimately results in the formation of a largernumber of muscle fibres. The characterization of myostatin and its mechanism of actionpointed to the identification of the genetic basis for the muscle hypertrophy in some breedsof cattle, commonly known as “double-muscling” (Grobet et al., 1997). In vivo, myostatin isregulated by follistatin, and follistatin overproduction increases muscle mass, whereasimpaired production reduces muscle mass.

The IGFs are unique in their actions in that they can stimulate both proliferation and dif-ferentiation by activating the type 1 IGF receptor; these effects, however, are temporallyseparated. The switch from proliferative to myogenic effects is associated with changes in theintracellular signalling pathways from primary activation of the mitogen-activated proteinkinase (MAPK) pathway during the proliferative phase, to signalling through the phos-phatidylinositol 3-kinase (PI 3-kinase) pathway for differentiation (Coolican et al., 1997). Incell culture, the proliferative phase is associated with IGF-I-stimulated expression of cellcycle markers and cell proliferation and reduced expression of myogenic markers, whereasduring differentiation IGF-I promotes expression of myogenin and muscle-specific geneexpression (Engert et al., 1996). The exact trigger responsible for switching the response fromproliferation to differentiation is unclear, especially in vivo. Studies in the chick embryo havedemonstrated that increased local expression of IGF-I in the limb at an early stage of devel-opment before cells have differentiated increases myoblast number with the formation oflarger muscles containing an increased number of myofibres (Mitchell et al., 2002), in muchthe same way as, although independently of, decreased myostatin expression. Similarly,administration of growth hormone (GH) to pregnant sows in early gestation (10–24 days)indirectly stimulates fetal IGF secretion and results in greater myofibre number at birth(Rehfeldt et al., 1993). This response contrasts with the response to over-expression of IGF-Iin the differentiated muscle, where muscle hypertrophy is not associated with increased fibrenumber (Musaro et al., 2001; Fiorotto et al., 2003).

The stimulation of muscle differentiation by the MRFs entails not only up-regulation oftheir expression within presumptive muscle cells, but also the co-ordinated orchestration ofa series of events that enables them to be transcriptionally active. These include: the down-regulation of Id proteins which mitigate the binding of MRFs to their E-box consensussequence (Benezra et al., 1990); the association with the myogenic enhancer factor 2 (MEF2)family of transcription factors that bind to both their own DNA site and form protein–proteininteractions with the MRFs (Molkentin and Olson, 1996); and the dissociation of histonedeacetylases from the transcription factors and subsequent recruitment of histone acetylasesto E-boxes associated with muscle-specific genes. The resulting acetylation of histonesproduces the chromatin relaxation necessary to permit transcription of the underlying gene(McKinsey et al., 2001, 2002). This differentiation step is associated concomitantly withinhibition of cell cycling, and the alignment and subsequent fusion of the adhered myoblaststo form myotubes.

Fibres form in two waves: the first wave occurs during early embryogenesis and resultsin the formation of primary myotubes that shape and position the orientation of individualmuscles (Ontell and Dunn, 1978; Ontell, 1982). These primary fibres are originally in clusters,but progressively become separated by the basal lamina. Subsequently, a secondary populationof myoblasts located under the basal lamina of the primary fibres begins to proliferate usingthe plasma membrane of the primary fibres as scaffolding, but without fusing to them. Thesethen fuse among themselves to form secondary myotubes, and gradually separate from the


primaries by forming their own basal lamina. There is a third population of myoblasts, thesatellite cells, that remain quiescent, sandwiched between the plasma membrane and basallamina. The expression of the homeobox protein, Pax-7, distinguishes satellite cells fromother lineages of myoblast, and appears to be essential for the formation of satellite cells(Seale et al., 2000). Satellite cells proliferate to enlarge the myonuclear population present inthe myofibres to which they are attached, and contribute only to myofibre hypertrophy, notnew fibres (except to repair muscle damage).

To this point in the process of skeletal muscle development, the primary effect of pertur-bations in the growth process are likely to manifest themselves as variations in the number ofmyofibres that form. The timing of the perturbation will determine if primary or secondaryfibres are affected. It is unclear if impairment of satellite cell replication during the terminalphase of muscle formation can permanently impact the size of a muscle’s reserve of satellitecells and, hence, postnatal muscle growth potential.

2.2. Compositional development

Once myoblasts have fused, the expression of muscle-specific proteins dominates musclegrowth. The cells undergo a complex set of transformations to create the highly structuredmyofibrils which have the capacity to perform contractile work. This process of maturationis critical for the developing offspring as it is essential for postnatal survival: skeletal musclesare a prerequisite for independent breathing and suckling. Indeed, the offspring die at birth inall transgenic animals in which normal skeletal muscle development is impaired by targeteddisruption of key regulatory molecules (e.g. Venuti et al., 1995). However, there is a widevariation in the level of muscle maturity at birth among species, and between muscle groupswithin the same individual. From these observations, it is evident that birth occurs at differentstages of muscle development across species, and that within an individual, maturationproceeds in a rostral to caudal direction, and from proximal to distal in the appendages.Hence, altricial species, like rabbits, rodents, and most carnivores, are born with functionalhead and thoracic muscles, whereas their lower abdominal and limb muscles, especially thoseof the hindlimbs, are still immature. In contrast, precocial species, such as ungulates, haverelatively longer gestations and the newborn muscle is at a fairly advanced stage of maturity.Indeed, locomotor function is attained very soon after birth.

The maturation of myotubes into fully functional myofibres involves the co-ordinateddevelopment of the metabolic machinery of the cells, the ion transport membrane system, andthe contractile elements. The complex sarcoplasmic reticulum and T-tubular system respon-sible for coupling excitation and contraction followed by muscle relaxation develop in aco-ordinated fashion and attain their mature configuration at approximately 2 weeks of age inthe rat. At this point, the membrane system constitutes approximately 40% of the non-myofibrillar compartment in the muscle fibre (Schiaffino and Margreth, 1969). Contractileproteins, not present in the myotube at fusion, comprise 55−65% of total muscle protein by2 weeks of age in the rat (Yates and Greaser, 1983; Fiorotto et al., 2000a). The accretion ofmyofibrillar proteins, therefore, is a major determinant of whole-body protein accretionduring this developmental period.

The diversity in fibre-type characteristics results from the combination of the inherentproperties of the myoblast lineage from which the fibres were derived and extrinsic signalsfrom the organism. Thus, in the development of a myofibre from a myotube, in addition tothe first-order general pattern of compositional development that occurs (described above),


significant changes also occur in the isoform composition of muscle proteins and thesupporting metabolic machinery. These result in a second order of compositional changes thatproduce the development of the adult phenotype of individual fibre types (Schiaffino andReggiani, 1996). A significant proportion of the proteins that constitute the thick, intermediate,and thin filaments initially are expressed as immature isoforms and subsequently are replacedby the adult isoforms during maturation (table 1). In the thick filaments, all primary fibresinitially express a combination of embryonic and slow MHC isoforms, or neonatal andslow MHC isoforms. The associated myosin light chains (MLC) also differ according to theMHC isoform with which they are associated and, therefore, their expression changes duringmaturation. Secondary fibres, in contrast, initially express the embryonic and neonatal MHCisoforms, which are replaced usually by any of the adult fast MHC isoforms according to thefunctional characteristics of the individual muscle. This process of terminal maturation isamenable to regulation by hormones, activity pattern, and neural inputs, and contrasts with theearlier differentiation of myoblast lineage which proceeds independently of extrinsic factors.

A fundamental conundrum regarding the differentiation among fibre type relates to themechanism that co-ordinates the appropriate expression of the plethora of proteins responsiblefor the metabolic and contractile characteristics of muscle. Recent studies have focused on the


Table 1

Contractile protein isoforms expressed in the developing and mature muscles of the rat

Developing muscles Adult muscles

Embryonic Neonatal Fast Slow

Myosin heavy chainsembryonic neonatal 2B β-slowβ-slow (embryonic) 2X

2AMyosin light chains (MLC)

MLC-1embryonic MLC-1fast MLC-1fast MLC-1slow/ventricularMLC-1slow-α (MLC-3fast) MLC-3fast (MLC-1slow-α)MLC-1fastMLC-2fast MLC-2fast MLC-2fast MLC-2slow

Actinα-cardiac α-skeletal α-skeletal α-skeletalα-skeletal (α-cardiac)

Troponins (Tn)TnC-fast TnC-fast TnC-fast TnC-slow/cardiacTnC-slow/cardiacTnI-slow TnI-fast TnI-fast TnI-slowTnT-cardiac TnT-fast, fetal TnT-fast, adult TnT-slow

isoforms isoformsTnT-slow

Tropomyosins (TM)TM-β TM-β TM-αfast TM-αslowTM-αfast TM-αfast (TM-β) TM-αfastTM-αslow TM-β

Minor isoforms are indicated in parentheses. Isoform profile indicated for neonatal developing muscles is that ofthe majority of hindlimb muscle fibres, which are secondary generation fibres destined to become fast-type fibres.Adapted from Schiaffino and Reggiano (1996).

role of calcineurin (CaN), a calcium and calmodulin-dependent serine/threonine phosphatase,in the regulation of expression of those genes responsible for the slow muscle phenotype(Schiaffino and Serrano, 2002; Spangenburg and Booth, 2003). CaN is activated when thereare high intracellular steady-state levels of Ca2+, typical of slow fibres that are subjected tochronic, low-frequency nerve stimulation. Once activated, CaN dephosphorylates the tran-scription factors, nuclear factor of activated T-cells (NFAT) and MEF-2, so that they can thenbe translocated into the nucleus where, in conjunction with MRFs, they effect changes in geneexpression (McKinsey et al., 2001, 2002). These factors bind to their respective DNA consen-sus sequences which form a characteristic motif, the slow upstream element (SURE), presentin the promoter of a variety of slow muscle genes such as slow troponin I and myoglobin(Calvo et al., 2001). Despite the ability of CaN to sense and transduce changes in cell calciumlevels into changes in gene expression, it is by no means a global regulator of the SO fibrephenotype as was initially proposed. For example, it has been observed that expression ofMHC-IIa, the predominant MHC isoform in FOG muscles, is also highly responsive to CaNactivation (Allen and Leinwand, 2002).

Considerable progress has been made in identification of the mechanisms that co-ordinatethe contractile and metabolic characteristics of muscle. Again, the sustained elevation ofintracellular calcium appears to be a central factor in signalling not only the fast-to-slow shiftin muscle gene expression (Allen and Leinwand, 2002), but also an increase in mitochondrialbiogenesis (Ojuka et al., 2003). In addition to CaN, calcium activates calcium/calmodulin-dependent protein kinases (CaMK) which catalyse a series of reactions that result in thetranscription of a coactivator of nuclear receptors, peroxisome proliferator-activated receptorcoactivator-1α (PGC-1α) (Handschin et al., 2003). PGC-1α plays a pivotal role in glucosemetabolism, mitochondrial biogenesis, and adaptive thermogenesis by activating various tran-scription factors. Specifically, in muscle PGC-1α has been shown to stimulate mitochondrialDNA replication, mitochondrial abundance, cytochrome c and cytochrome oxidase levels,GLUT4 expression, and uncoupling protein expression. PGC-1α also enhances its own tran-scription (Handschin et al., 2003). Consequently, once activated, an autoregulatory loop is setup which sustains PGC-1α expression and its downstream effects, and thereby maintainsstable expression of the oxidative phenotype.

In addition to calcium, PGC-1α expression is regulated by thyroid hormone and AMP-activated protein kinase, an enzyme that is activated by chronic reductions in the cellularATP/AMP ratio, for example, with energy deprivation (Irrcher et al., 2003; Ojuka et al., 2003;Spangenburg and Booth, 2003). These mechanisms that co-ordinate the metabolic properties ofa muscle with its contractile properties and pattern of use, however, are pertinent primarily to thedevelopment of slow-twitch and/or oxidative properties, presumably in muscles where thesecharacteristics are not present. This suggests that fast-twitch, glycolytic properties are the defaultphenotype of skeletal muscle and there is, indeed, some evidence to support this suggestion.During terminal maturation, the loss of polyinnervation and acquisition of single innervationfrom a nerve with a low-frequency firing pattern is necessary for the development and mainte-nance of slow-twitch characteristics. Moreover, if the soleus is denervated at birth in the rat, slowmyosin isoenzymes are gradually replaced by fast myosins (Gambke et al., 1983). Thus, thereplacement of the immature isoforms of myosin specifically by adult slow myosin occurs onlywith the appropriate neural input. In the mature muscle, cross-innervation of a mature fast-twitchmuscle with the nerve from a slow-twitch muscle gradually transforms the entire contractilephenotype and metabolic properties to those of a slow muscle (Barany and Close, 1971).

Thyroid hormone also plays a critical role in the maturation of skeletal muscle. Moreover,because thyroid hormone is sensitive to changes in energy balance, it may serve as a signal


to the muscle to produce adaptive changes in muscle metabolism. The importance of thyroidhormone has been studied extensively with respect to the regulation of MHC expressionwhere it is required for down-regulation of the neonatal isoform of MHC in muscles destinedto become either fast or slow (Gambke et al., 1983; Adams et al., 1999). Furthermore, in theabsence of thyroid hormone, the accumulation of slow MHC is accelerated, whereas thatof IIA MHC is down-regulated. Thus, variations in energy balance that produce changes inthyroid hormone might be anticipated to alter the postnatal pattern of muscle maturation. Ourstudies in suckling rats suggest that the changes in thyroid hormone have to be relativelysevere in order to effect changes at the level of gene expression.

However, changes in muscle use and protein turnover also occur as a consequence of alter-ations in food intake and growth rate. These changes serve to mitigate the effect of alteredgene expression and, consequently, the maturation of muscle phenotype is preserved (Fiorottoand Davis, 1997; Fiorotto et al., 2000a). The suckling pig responds to mild hypothyroidismduring the suckling period by increasing slow MHC expression, although the effect is some-what mitigated by increases in nuclear thyroid hormone receptor (Harrison et al., 1996).Relatively severe energy restriction in post-weaned pigs has similar effects, increasing theabundance of slow MHC at both the protein and mRNA level, and increasing the oxidativeproperties of the muscles, but with substantial muscle-to-muscle variability (White et al.,2000). Although the increase in slow MHC expression is compatible with the known effectsof hypothyroidism on MHC expression, the enhanced oxidative properties are the opposite ofthose anticipated on the basis of PGC-1α regulation by thyroid hormone. However, they arecompatible with a change in AMP kinase activity, which increases with a chronic deficit inenergy intake, and thereby promotes mitochondrial biogenesis and fatty acid oxidation.Overall, these responses to a chronic deficit in energy intake represent beneficial adaptationsby the muscle to enhance its metabolic efficiency: a slow muscle requires less energy than afast-twitch muscle to generate the same amount of tension, and it is able to derive more of itsenergy by fatty acid oxidation and oxidative phosphorylation (Henriksson, 1990).

The regulation of the fast-twitch, glycolytic phenotype of muscles is much less clearlyunderstood than that of slow-twitch muscle. Some genes expressed in fast fibres containa characteristic binding motif, the fast intronic regulatory element (FIRE), analogous to theSURE motif in slow fibre genes (Nakayama et al., 1996). Myoblast lineage established duringfetal life appears to be a primary determinant of whether a fibre matures into a fast fibre.As noted previously, gene mutations that promote secondary myoblast proliferation resultprimarily in fast fibre hypertrophy. During terminal maturation, thyroid hormone is requiredfor the down-regulation of neonatal MHC and, if present at high levels, thyroid hormonetends to drive the expression of fast MHC in muscle fibres that normally would be slow(Nakayama et al., 1996). The role of innervation also appears to be less critical in the devel-opment of fast-twitch fibres than for slow fibres. In both rodents and chickens, denervationdelays the elimination of immature MHC isoforms, but does not prevent the development ofthe fast phenotype. Inactivity also tends to promote the fast phenotype, although this is attrib-utable in part to the preferential atrophy of slow fibres. In post-weaned pigs (Katsumata et al.,2000), but possibly not neonatal pigs (Louveau and Le Dividich, 2002), mild undernutritionup-regulates the expression of the growth hormone receptor (GHR) on FOG and FG fibreswhich normally express the lowest level of GHR. Together with the reduction in thyroidhormone expression, the resulting changes in hormone responsiveness may be responsible forthe metabolic shift that occurs in muscle during undernutrition and which enables it to derivemore of its energy from lipid oxidation. A broader implication of these findings, however,relates to the anti-insulin effects of GH in the undernourished animal which serve to divert


nutrients from muscle towards the visceral organs. This is demonstrated by the differentialresponse in rates of tissue growth in protein-malnourished piglets where body protein parti-tioned into gastrointestinal tissue is preserved, while that of skeletal muscle is reduced (Ebneret al., 1994).

2.3. Role of protein synthesis and degradation in the regulationof compositional development

As the above discussion of the developmental changes in muscle composition indicates, pro-tein synthetic rates in the immature muscle must sustain not only the de novo accretion ofmyofibrillar proteins and membrane structures, but also their continuous and co-ordinatedreplacement as the tissue develops its mature compositional and functional characteristics.The faster accumulation of myofibrillar proteins compared to sarcoplasmic proteins isexplained almost entirely by their higher fractional synthesis rate compared to other proteincomponents (Fiorotto et al., 2000a). As compositional maturity is attained, the synthesis rateof myofibrillar proteins decreases to a greater extent than sarcoplasmic proteins, and in themature muscle, sarcoplasmic protein synthesis rates are higher than for myofibrillar proteins.However, once the mature composition is attained, the rates of degradation also differ inparallel and results in the maintenance of constant composition. In altricial mammals such asrodents and rabbits, the differential regulation of protein synthesis in the different proteinpools occurs in the immediate postnatal period. In precocial animals, the full complement ofmyofibrillar proteins in fibres is mostly completed by birth, although they still undergo somelimited, second-order compositional maturation postnatally. Nonetheless, the intrauterinepattern of development and mechanisms of regulation at comparable stages of maturity arelikely to be similar across species.

In the mature muscle, there are fibre-type differences in the rate of protein turnover thatreflect their compositional differences; slow fibres have higher rates of protein turnover thanfast-twitch fibres (Bark et al., 1998; Dardevet et al., 2002), and this diversity emerges onlyupon maturation (Davis et al., 1989). These phenotypic differences are attributable to thedifferences in the synthesis rate of muscle proteins in combination with the variation in theirrelative abundances among muscles. In skeletal muscles from mature pigs, the averagesynthesis rate of mitochondrial proteins is higher than for sarcoplasmic proteins and this, in turn,tends to be slightly higher than for the myofibrillar proteins (Balagopal et al., 1997; Fiorottoet al., 2000a). Although the ratio of myofibrillar to sarcoplasmic proteins tends to be greaterin slower muscles (Hemel-Grooten et al., 1995), the difference in synthesis rates is substan-tially lower than that of mitochondrial proteins, which are more abundant in the slower,oxidative muscles. The greater overall protein synthetic activity of the slow muscles is sup-ported by a higher ribosomal abundance and entails minimal changes in protein syntheticefficiency. In addition to the inherent variation in synthesis rates, the myofibre protein com-ponents can also differ in their responses to extrinsic stimuli. For example, in adult porcinemuscle, stimulation of protein synthesis by insulin appears to be limited to the mitochondrialproteins (Boirie et al., 2001); the developmental decline in muscle protein synthesis rates isdominated by myofibrillar proteins (Fiorotto et al., 2000a). Clearly, these differences amongmuscle protein components have to be factored into our understanding of the overall regula-tion of skeletal muscle protein metabolism.

In the newly differentiated muscle, the high ribosomal abundance is the principal factorthat enables high rates of protein synthesis to be attained, and its reduction with maturationis one mechanism that may underlie the general reduction in fractional synthesis rates


observed for all muscle proteins. However, this cannot explain the differences in compositionof proteins synthesized, and some regulation must occur at the level of gene transcription.During the transition from myoblast to myotube, genes encoding non-muscle proteins arerepressed, while those specific for muscle proteins are induced in a co-ordinated manner.There is then a commensurate change in the composition of proteins expressed (Devlinand Emerson, 1978, 1979; Shani et al., 1981). In vivo, it has been demonstrated that thestoichiometry of the total mRNAs encoding all isoforms within a protein family is maintainedaccurately, and that production of individual myofibrillar proteins in appropriate stoichio-metric amounts, therefore, is regulated at the message level (Wade et al., 1990). However,such changes would not explain the differential responses of sarcoplasmic and myofibrillarproteins even if the decrease in ribosomal abundance were accompanied by a reduction in theproportion of myofibrillar mRNAs. Such a change in mRNA composition would increase thetranslational efficiency of sarcoplasmic proteins, but decrease the translational efficiency ofmyofibrillar proteins. Translational efficiency, however, increases for all proteins in theimmature muscle. Clearly, therefore, although differences in mRNA abundances are involved,as we have demonstrated in newborn pigs (Fiorotto et al., 2000b), this also cannot be entirelyresponsible for the compositional differences in protein synthesis, suggesting that there mustalso be regulation of mRNA at the translational and post-translational levels.

3. POSTNATAL MUSCLE GROWTH

3.1. Satellite cells and hyperplasia

Postnatal growth of skeletal muscle is driven by hypertrophy of the existing fibres. Thisrequires both an increase in myonuclear content, and the accretion of muscle proteins.Myonuclei are postmitotic and, thus, satellite cells are entirely responsible for the postnatalincrease in muscle fibre DNA. This is clearly demonstrated in mice in which the expressionof Pax-7 is abolished and consequently no satellite cells form (Seale et al., 2000). Thesemuscles contain both primary and secondary fibres but they fail to hypertrophy postnatally.Indeed, in a variety of circumstances normally associated with accelerated postnatal musclegrowth, the inhibition of satellite cell replication will prevent the growth response (Rosenblattand Parry, 1992).

Recent evidence suggests that subpopulations of satellite cells may exist that can be dis-tinguished by their proliferative potential (Perry and Rudnicki, 2000; Seale and Rudnicki,2000). There is a reserve population of quiescent, non-differentiated cells that retains itsmitogenic potential and has the capacity for self-renewal. Under the appropriate stimulation,these satellite cells become activated, migrate as necessary, and undergo a limited number ofreplications before they terminally differentiate. These cells can no longer divide and undergofusion into the myofibre. In the rat, satellite cells comprise approximately 32% of musclenuclei at birth and decrease to 10% at 4 weeks of age and less than 5% at sexual maturitywhen the cells are largely mitotically quiescent (Allbrook et al., 1971). A similar pattern isseen in the pig, in which satellite cells constitute approximately 20% of total muscle nucleiat birth, and 4% at 64 weeks of age (Campion et al., 1981; Mesires and Doumit, 2002). Thesevalues vary according to the metabolic properties of the muscle. The significance of myonu-clear number in the context of muscle growth relates to the observation that in skeletalmuscle, “myonuclear domain” size, i.e. the quantity of cytoplasm regulated by a singlemyonucleus (and reflected by the protein:DNA ratio), is tightly regulated (Allen et al., 1999).This implies that the amount of protein that can be deposited without further addition of


myonuclei is limited. Nuclear domain size under “steady state” conditions appears to varyaccording to the metabolic activity of the fibre. It is smaller for oxidative than glycolyticfibres, and for any given fibre type, values increase with age. Although satellite cells becomequiescent as growth rate plateaus, their proliferation can be reactivated in response to muscleinjury, denervation, or increased muscle stretch and it is essential for muscle repair and hyper-trophy (Bischoff, 1994).

The close similarity between the developmental changes in satellite cell replication andprotein synthesis strongly suggests that these processes may be linked. This is further sup-ported by the differential response of immature and mature skeletal muscles to suboptimalnutrient intakes. Alterations in food intake (greater or less than average) in the neonatal animal,provided they are not severe, alter DNA and protein accretion proportionally as indicated bythe maintenance of relatively normal, age-appropriate protein:DNA ratios despite a wide rangeof growth rates (Fiorotto and Davis, 1997). In transgenic animals, sustained over-expressionof IGF-I in the skeletal muscle promotes satellite cell replication and transiently increases theaccretion of total muscle DNA; this increase precedes the enhanced accumulation of muscleprotein and results in protein:DNA ratios that temporarily are lower than normal (Fiorottoet al., 2003). These ultimately increase to age-appropriate values, but never surpass those inwild-type control animals.

A potential link between satellite cell replication and the capacity for protein synthesis isthrough the regulation of ribosomal production, ribosomal abundance being the primarydeterminant of a cell’s maximal capacity for protein synthesis. Regulation of ribosomebiogenesis is achieved in the majority of cells by altering the rate of rRNA synthesis by rDNAtranscription (Zahradka et al., 1991), the regulation of which is coupled to cell cycling via theretinoblastoma gene product, pRb (Hannan et al., 2000). We have demonstrated that theenhanced replicative capacity of satellite cells from muscles that overexpress IGF-I is associ-ated with increased phosphorylation of pRb upon mitogen stimulation (Chakravarthy et al.,2001). Thus, when rates of satellite cell division are high, pRB is phosphorylated, and in thisform it enables a key transcription factor for rDNA transcription, UBF, to transactivate rDNAgenes to promote rRNA synthesis. Thus, accretion of ribosomes is necessarily correlated tothe rate of cell division.

3.2. Role of protein synthesis in the regulation of muscle growth

A rapid increase in the absolute rate of growth occurs during early postnatal life and a majorityof this growth is comprised of skeletal muscle protein (Young, 1970). The more rapid accre-tion of muscle protein than other tissue proteins results in an increase in the proportion of thebody protein pool that is represented by muscle protein from ~30% in the newborn to ~50%in the adult (fig. 2). However, the fractional rate of growth, i.e. the amount of weight gainedin relation to the existing mass, is extremely high at birth and decreases with development,with the most rapid change in the fractional rate of growth occurring during the neonatalperiod. This developmental decline in the fractional rate of growth is largely explained by adevelopmental decline in the fractional rate of protein deposition in skeletal muscle (Shieldset al., 1983; Mitchell et al., 2001).

Changes in the rate of protein deposition are driven by changes in the rates of protein syn-thesis or protein degradation such that a decline in the fractional rate of protein deposition canbe due to a decline in the fractional rate of protein synthesis, an increase in the fractional rateof protein degradation, or both. The developmental decline in protein deposition in skeletalmuscle is due to a developmental decline in the fractional rate of muscle protein synthesis


(Kelly et al., 1984; Denne and Kalhan, 1987; Davis et al., 1989) (fig. 3). In fact, the fractionalrate of muscle protein synthesis in the pig and rat is about 3-fold higher in the newborn than atweaning, and the rate of decline is attenuated as development proceeds (Kelly et al., 1984; Daviset al., 1989, 1996; Baillie and Garlick, 1992; Fiorotto et al., 2000a). This developmental declinein skeletal muscle protein synthesis is more profound in muscles containing predominatelyFG fibres than in those containing primarily SO fibres (Davis et al., 1989). By contrast, fractionalprotein degradation rates in skeletal muscle decline modestly with development.

The rate of protein synthesis is determined by the abundance of ribosomes, the efficiencyof the translational process, and potentially, the concentration of translatable mRNA (Kimballand Jefferson, 1988). Because the majority of RNA in tissues is rRNA, ribosomal abundancecan be estimated from the RNA to protein ratio, or can be measured more precisely from theamount of 18S rRNA expressed per unit protein. The efficiency of the translation process can


Fig. 2. Relative changes in the proportion of whole-body protein mass attributable to skeletal muscle in therat between birth and weaning (Fiorotto et al., unpublished observations).

Fig. 3. Relationship between the postnatal decline in the rate of muscle protein accretion and the fractionalsynthesis rate of skeletal muscle proteins in the hindlimbs of rats. (Data compiled from Davis et al., 1989;Fiorotto et al., 2000a.)

be calculated from the amount of protein synthesized per unit RNA and reflects how well theprotein synthetic machinery is functioning. Chronic changes in protein synthesis are thoughtto be a result of a change in ribosome number. Thus, the high rate of protein synthesis inimmature muscle and its overall decline with development are driven largely by an elevatednumber of ribosomes at birth and a developmental decline in ribosome concentration as themusculature matures (Kelly et al., 1984; Davis et al., 2001). Rapid changes in the rate of pro-tein synthesis, including those due to food ingestion, are generally regulated by changes in theefficiency of translation process secondary to modulation of the rate of translation initiation(Harmon et al., 1984; Kimball and Jefferson, 1988; Kimball et al., 1994), the rate-limitingstep in protein synthesis.

One of the best characterized steps involved in the regulation of translation initiation,depicted in fig. 4, is the binding of initiator methionyl-tRNA (met-tRNA) to the 40S ribosomalsubunit to form the 43S preinitiation complex via mediation of eukaryotic initiation factor(eIF) 2 (Pain, 1996; Kimball et al., 1997; Webb and Proud, 1997). The eIF2-mediated met-tRNAbinding to the 40S subunit is further regulated by the activity of eIF2B, which exchangesGDP for GTP on eIF2 (Kimball et al., 1996). A second well-characterized step in translationinitiation, shown in fig. 4, is the binding of mRNA to the 43S preinitiation complex via medi-ation of the assembly of the eIF4F complex of proteins (Lin et al., 1994; Rhoads et al., 1994;Sonenberg, 1994). The three proteins comprising the eIF4F complex are eIF4A, an RNAhelicase, eIF4E, the protein that binds to the m7GTP cap structure at the 5′-end of the mRNA,and eIF4G, a scaffolding protein that binds to the 40S ribosomal subunit. Thus, mRNA bindsto the 40S ribosomal subunit through the association of eIF4E with eIF4G. The availabilityof eIF4E for binding to eIF4G is regulated by its association with 4E-BP1, a repressor pro-tein that competes with eIF4G for binding to eIF4E (Pause et al., 1994). Upon stimulation byan anabolic agent, such as insulin, 4E-BP1 becomes phosphorylated, resulting in reducedaffinity of eIF4E for 4E-BP1, release of eIF4E, and enhanced binding of eIF4E to eIF4G toform the active eIF4E:eIF4G complex (Gingras et al., 1999).

Activation of translation initiation is mediated through a signal transduction pathwayinvolving a protein kinase referred to as the mammalian target of rapamycin (mTOR) which,


Fig. 4. Regulation of translation initiation. Abbreviations: eIF, eukaryotic initiation factor; 4EBP1, eIF4Ebinding protein; Met-tRNA, initiator methionyl tRNA; S6K, 70 kDa ribosomal protein S6 kinase 1; 43S, 43Sribosomal subunit; 48S, 48S ribosomal subunit.

in addition to phosphorylating 4E-BP1, also phosphorylates and activates the 70 kDa ribosomalprotein S6 kinase, S6K1 (Jefferies et al., 1994; von Manteuffel et al., 1997). These phospho-rylation events lead to an increase in the rate at which most proteins are synthesized and, inaddition, the preferential increase in translation of mRNAs encoding elements of the transla-tional apparatus, including ribosomal proteins and elongation factors.

Recent studies in growing pigs suggest that the overall developmental decline in theresponse of skeletal muscle protein synthesis to feeding involves regulation by eIF2B (Daviset al., 2000). Availability of eIF4E for 48S ribosomal complex formation follows a similarpattern. This response is primarily modulated by the developmental change in the feeding-induced activation of the factors involved in the binding of mRNA to the 43S preinitiationcomplex.

3.3. Role of protein degradation in the regulation of muscle growth

Less is known about the mechanisms that regulate protein degradation than those that regu-late protein synthesis. It is known that there are multiple pathways in mammalian tissues forthe degradation of proteins and that these pathways are highly controlled and selectivelydegrade specific protein substrates. These pathways include the lysosomal–autophagicsystem, the calpain–calpastatin system, and the ubiquitin–proteasome system (Goll et al.,1989; Attaix et al., 1999). The lysosomal–autophagic systems involve primarily cathepsins.Most evidence suggests that this pathway of degradation is unselective and may be of specialimportance under conditions in which cellular proteolysis is maximally activated. Thecalpain–calpastatin system is the major calcium-activated pathway of protein degradation.At least two main calpain isoforms, μ−calpain and m-calpain, have been identified and thesystem is subject to inhibition by the protein, calpastatin. The proteases play an important rolein muscle myofibrillar protein turnover by catalysing initial disruption of the structure viaproteolysis at the Z-disc. Released myofilaments can then be degraded into amino acids bythe proteasome and/or lysosomal enzymes (Goll et al., 1992). The ubiquitin–proteasomepathway is widely distributed among tissues and has a relatively broad protein specificity.It consists of a recognition system involving the protein ubiquitin, which is responsible fortargeting the protein substrates towards degradation by forming a polyubiquitin complex, anda multifunctional protease, referred to as the proteasome, which degrades the proteins. Therole of these proteolytic pathways in the regulation of muscle growth and development remainsto be explored.

4. REGULATORS OF PROTEIN SYNTHESIS

4.1. Feeding

Dietary protein is utilized very efficiently for the deposition of whole-body protein duringearly postnatal life (Pellett and Kaba, 1972; McCracken et al., 1980; Fiorotto et al., 1991;Davis et al., 1993a). The accumulated evidence suggests that young animals utilize theirdietary amino acids more efficiently for growth because they are capable of a greater increasein muscle protein synthesis in response to feeding than older animals (Davis et al., 1991,1996). Feeding stimulates protein synthesis in the whole body of the newborn human (Denneet al., 1991); in skeletal muscle of the suckling lamb (Oddy et al., 1987; Wester et al., 2000);and in skeletal muscle of the post-weaned, but still growing, rat (Garlick et al., 1983).However, the stimulation of muscle protein synthesis by feeding is blunted or absent in adultmammals (Melville et al., 1989; Baillie and Garlick, 1992; Tessari et al., 1996).


In the suckling pig (Davis et al., 1996, 1997; Burrin et al., 1997a) and rat (Davis et al.,1991, 1993b), protein synthesis in skeletal muscle is maximally stimulated after eating.Figure 5 shows that the postprandial rise in protein synthesis in skeletal muscle of the neona-tal pig declines sharply during the first 4 weeks of life. Although feeding stimulatesprotein synthesis in all tissues of the neonatal animal, the magnitude and the developmentaldecline in the response to feeding are most pronounced in skeletal muscle (Burrin et al., 1991,1995, 1997a; Davis et al., 1991, 1993b, 1996). This enhanced ability of skeletal muscle proteinsynthesis to respond to the provision of nutrients in young growing animals should not besurprising, because the rate of protein deposition during the postprandial period must behigher than the rate of protein loss during the postabsorptive period to permit growth of skeletalmuscle.

Recent studies have examined the developmental changes in the expression and activationof factors that regulate the feeding-induced stimulation of protein synthesis in skeletal muscleof young, growing pigs. The results show that eIF2B activity, which regulates the binding ofmet-tRNA to the 40S ribosomal subunit, is unaffected by feeding but decreases with devel-opment. The stimulation of muscle protein synthesis by feeding, and the developmentaldecline in this response, involve regulation by the eIF4F complex (Davis et al., 2000; Kimballet al., 2002). In skeletal muscle of the neonatal pig, feeding increases the phosphorylationof 4E-BP1, resulting in dissociation of the inactive 4E-BP1 . eIF4E complex, and increasedassociation of the active eIF4E .eIF4G complex. This response leads to a global increase inthe rate of muscle protein synthesis. These feeding-induced changes in the activity of factorsthat regulate eIF4F formation decrease with development in parallel with the developmentalchange in the feeding-induced stimulation of muscle protein synthesis. A response to feedinghas been observed in “teenage” rats (Yoshizawa et al., 1997). However, the magnitude of theresponse is smaller than that in neonatal pigs, thus further supporting a developmental declinein the feeding-induced formation of the eIF4F complex.

The developmental changes in the feeding-induced eIF4F activation occur in parallel withincreased phosphorylation of S6K1, which is involved in the translation of mRNAs encodingspecific proteins that regulate translation initiation (Davis et al., 2000; Kimball et al., 2002).An increased phosphorylation of both 4E-BP1 and S6K1 suggests involvement of the mTORsignalling pathway in this process. Furthermore, rapamycin, a specific inhibitor of mTOR,


Fig. 5. Stimulation of protein synthesis by feeding decreases with development in neonatal pigs.

strongly attenuates the feeding-induced assembly of both eIF4F and S6K1 activation(Kimball et al., 2000). Thus, the enhanced activation of the eIF4F complex following foodconsumption likely plays an important role in the postprandial stimulation of muscle proteinsynthesis in growing animals and the efficient use of dietary amino acids for muscle proteindeposition in the neonate.

4.2. Insulin

Studies performed in incubated muscles and in perfused hindlimbs of growing animals clearlydemonstrate that insulin stimulates protein synthesis (Jefferson et al., 1977; Davis et al., 1987;Kimball et al., 1994). The infusion of physiological concentrations of insulin in fasted,weaned rats stimulates muscle protein synthesis in vivo to rates similar to those found in thefed state (Garlick et al., 1983). This response to feeding can be blocked by co-administrationof anti-insulin serum (Preedy and Garlick, 1986). Furthermore, insulin has been shown tostimulate whole-body amino acid utilization and protein synthesis in the fetal sheep (Liechtyet al., 1992; Thureen et al., 2000), protein synthesis in hindlimb of the young lamb (Westeret al., 2000), and skeletal muscle protein synthesis in the weaned rat (Garlick et al., 1983). Inmarked contrast to studies conducted in growing animals, most studies in adult animals(Baillie and Garlick, 1992; McNulty et al., 1993) and humans (Gelfand et al., 1987; Heslinet al., 1992; Louard et al., 1992) show little, if any, response of muscle protein synthesis tophysiological increases in insulin. This suggests that the response of muscle protein synthe-sis to insulin is developmentally regulated.

Insulin plays a key role in the increased response of skeletal muscle protein synthesis to feed-ing, and thus the increased rate of protein deposition, during the early postnatal period. In fastedand fed neonatal pigs, there is a positive curvilinear relationship between the postprandialincrease in fractional muscle protein synthesis rates and circulating insulin concentrations(Davis et al., 1997). Studies using a hyperinsulinemic–euglycemic–euaminoacidemic clamptechnique show that when amino acids and glucose are maintained at fasting levels, insulininfusion increases amino acid disposal, and that the insulin sensitivity and responsiveness ofamino acid disposal decrease with development (Wray-Cahen et al., 1997). This responsesuggests that the developmental change in the insulin sensitivity of whole-body amino aciddisposal may underlie the developmental change in the efficiency of utilization of dietaryamino acids for protein deposition. Furthermore, raising insulin concentrations in the neonatalpig to levels typical of the fed state increases the rate of skeletal muscle protein synthesis towithin the range normally present in the fed state, even when amino acids and glucose aremaintained at fasting levels (Wray-Cahen et al., 1998). This response to insulin, like theresponse to feeding, is attenuated with development and is greater in muscles that are com-posed primarily of FG fibres, and is not specific to myofibrillar proteins (Davis et al., 2001).

The insulin signalling cascade (fig. 6) leading to the stimulation of protein synthesis isinitiated by insulin binding to its receptor. This leads to autophosphorylation of the receptor,the activation of insulin receptor tyrosine kinase, and the subsequent phosphorylation ofseveral cytosolic substrates including insulin receptor substrate (IRS)-1 and -2 (Sun et al.,1991; White and Kahn, 1994). IRS-1 and -2 serve as “docking proteins”, transmitting insulinsignals to several proteins that contain Src-homology 2 (SH2) domains (Backer et al., 1992;Sun et al., 1993) including phosphatidylinositol (PI) 3-kinase, which catalyses the phosphory-lation of PI. The activation of PI 3-kinase triggers the activation of components of the insulinsignalling pathway leading to translation initiation, i.e. protein kinase B (Akt) and mTOR.


Studies focusing on the developmental changes in the insulin signalling pathway that leadsto translation initiation have shown that in the pig the abundance of insulin receptor proteinin muscle during the early suckling period is 2-fold higher than at weaning (Suryawan et al.,2001). Although the abundance of IRS-1 and IRS-2 does not change with development, theabundance of the downstream signalling proteins, protein kinase B and mTOR, decreaseswith development (Kimball et al., 2002). This developmental decline in the abundance ofinsulin receptor, protein kinase B, and mTOR in skeletal muscle likely contributes to the over-all decline in the responsiveness of muscle protein synthesis to feeding that occurs over thecourse of development.

Because insulin mediates the postprandial elevation in skeletal muscle protein synthesisand this response decreases with development (Wray-Cahen et al., 1998; Davis et al., 2001),it is not surprising that the feeding-induced activation of the insulin signalling pathway thatregulates protein synthesis decreases with development. Thus, the feeding-induced activationof the insulin receptor, IRS-1, IRS-2, PI 3-kinase, and protein kinase B in skeletal muscledecreases with development (Suryawan et al., 2001; Kimball et al., 2002), in parallel withthe developmental decline in the feeding-induced activation of translation initiation factorsand protein synthesis (Davis et al., 1996). This suggests that the developmental decline inthe postprandial stimulation of protein synthesis in skeletal muscle results from a reductionin the capacity of the intracellular insulin signalling pathway to transduce to the translationalapparatus the stimulus provided by the feeding-induced rise in insulin and/or amino acidconcentrations.

A number of studies performed in cell culture, in the perfused hindlimb, and in intact growingrats have demonstrated that the stimulation of protein synthesis by insulin involves increasedphosphorylation of the translational repressor protein, 4E-BP1, reduced interaction witheIF4E, and increased assembly of the mRNA cap-binding complex, eIF4G:eIF4E (Kimball et al.,1994, 1997). Furthermore, insulin increases phosphorylation of S6K1, thereby increasing thetranslation of specific proteins involved in the regulation of translation. Recent in vivo studiesperformed in neonatal pigs support these findings and further show that the insulin-induced


Fig. 6. Insulin signalling pathway leading to translation initiation.

changes in factors regulating translation initiation as well as the upstream components of theinsulin signalling pathway occur in a dose-response manner within the physiological range(Suryawan et al., 2001; O’Connor et al., 2003). Recently, however, studies in adult ratssuggest that while insulin increases the phosphorylation of S6K1, insulin does not alter4E-BP1 phosphorylation (Long et al., 2000). This lack of effect of insulin on 4E-BP1 phos-phorylation and, by inference, eIF4F formation, is not surprising as physiologicalhyperinsulinaemia has no effect on muscle protein synthesis in adults. Thus, insulin plays animportant role in the regulation of protein synthesis in muscle of growing animals, but itsimportance during adulthood is less apparent.

4.3. Amino acids

Although amino acids are the precursors for the synthesis of proteins, they also play a keyrole as nutritional signals in the regulation of muscle protein synthesis. Amino acids have thecapability to stimulate muscle protein synthesis throughout a substantial part of the life cycle,in contrast to the developmental decline and loss of the capability of insulin to stimulatemuscle protein synthesis with age. In weaned but still growing rats (Preedy and Garlick,1986), adult humans and rats (Bennet et al., 1990; McNulty et al., 1993; Vary et al., 1999),and elderly people (Volpi et al., 1998), acute amino acid infusion, either alone or concurrentwith insulin infusion, stimulates protein synthesis in skeletal muscle. Recent studies suggest,however, that the magnitude of the stimulation of muscle protein synthesis by amino acidsmay decrease in the early postnatal period (Davis et al., 2002a).

When a balanced amino acid mixture is infused into fasted, growing pigs, muscle proteinsynthesis increases and this response to amino acid infusion decreases with development,in parallel with the developmental decline in the feeding-induced stimulation of skeletalmuscle protein synthesis (Davis et al., 2002a). In young pigs, the stimulation of skeletalmuscle protein synthesis by amino acids is greater in muscles that contain predominatelyFG muscle fibres than in those that contain primarily SO fibres, and is similar for myofibril-lar and sarcoplasmic proteins. The response to amino acid infusion occurs when insulin levelseither remain at the fasting level or are raised to the fed level by infusion (O’Connor et al.,2003). Indeed, the magnitude of the increase in muscle protein synthesis with amino acidstimulation is similar to that which occurs with insulin stimulation alone, implying thatinsulin and amino acids may be interacting with the same signalling pathway within skeletalmuscle.

Studies performed in cell culture have shown that amino acid availability modulates pro-tein synthesis by regulating both the met-tRNA and mRNA binding steps of translationinitiation (Fox et al., 1998; Hara et al., 1998; Kimball et al., 1998; Patti et al., 1998; Jeffersonand Kimball, 2001). In vivo studies in mature, food-deprived rats in which a large oral doseof leucine was administered suggest that in muscle, leucine promotes the binding of eIF4G toeIF4E, increases the phosphorylation of 4E-BP1, and represses the association of eIF4E with4E-BP1 (Anthony et al., 2000). In neonatal pigs, raising amino acids from the fasting to thefed levels in the presence of insulin produced a similar response (O’Connor et al., 2003).In the absence of insulin, amino acids do not affect either the phosphorylation of S6K1 and4E-BP1, or the association of eIF4E with 4E-BP1 and eIF4G, even though they stimulatemuscle protein synthesis. This suggests that amino acids stimulate muscle protein synthesisin growing animals by modulating the availability of eIF4E for 48S ribosomal complex for-mation, and by processes that do not require enhanced assembly of the mRNA cap-bindingcomplex.


4.4. Insulin-like growth factors

Many (Douglas et al., 1991; Fryburg et al., 1995; Bark et al., 1998; Vary et al., 2000; Daviset al., 2002b), although not all (Oddy and Owens, 1996; Boyle et al., 1998), studies havedemonstrated an anabolic effect of IGF-I on protein synthesis in skeletal muscle. However, insome studies reductions in circulating concentrations of amino acids, insulin, and/or glucoseduring the administration of IGF-I may have limited the ability of IGF-I to stimulate proteinsynthesis. Thus, when amino acids, glucose, and insulin are maintained at fasting levels, infu-sion of IGF-I to the level seen in the fed state stimulates muscle protein synthesis in growingswine (Davis et al., 2002b).

IGF-I, however, is unlikely to play a role in the feeding-induced stimulation of muscleprotein synthesis. First, in contrast to insulin, the rise in circulating IGF-I after feedingis not immediate (Buonomo and Baile, 1991; Goldstein et al., 1991; Davis et al., 1993b, 1996;Svanberg et al., 1996). Second, the postprandial changes in muscle protein synthesis inyoung animals are positively correlated with changes in circulating insulin, but not IGF-I,concentrations (Davis et al., 1997, 1998). Third, with development, circulating IGF-I levelsincrease, whereas skeletal muscle protein synthesis rates decrease (Davis et al., 1996).Although circulating IGF-I is unlikely to be a physiologically significant regulator ofthe feeding-induced stimulation of skeletal muscle protein synthesis, this does not negatethe potential role of IGF-I as a long-term regulator of growth, as has been suggested byothers (Buonomo and Baile, 1991; Donovan et al., 1991; VandeHaar et al., 1991), or thepotential usefulness of IGF-I as an anabolic agent to enhance protein deposition as discussedpreviously.

IGF-I likely stimulates protein synthesis in skeletal muscle by acting on the same signallingpathway as insulin that leads to translation initiation (Dardevet et al., 1996; Vary et al., 2000).The receptors for both IGF-I and insulin share considerable homology of structure and func-tion (Ullrich et al., 1986; Cheatham and Kahn, 1995; LeRoith et al., 1995) and both hormonesact on some of the same intracellular signalling pathways (Dardevet et al., 1996; Suryawanet al., 2001). Furthermore, both insulin and IGF-I stimulate protein synthesis by increasingthe formation of the active eIF4E . eIF4G complex that regulates the binding of mRNA to theribosome (Kimball et al., 1997; Vary et al., 2000).

4.5. Growth hormone

Growth hormone treatment increases protein deposition, improves nitrogen retention, andenhances the efficiency with which dietary protein is utilized for growth (Campbell et al.,1990; Caperna et al., 1991; Vann et al., 2000a). Furthermore, GH treatment profoundlydecreases the synthesis and excretion of urea, and the oxidation of amino acids. Whole-bodyprotein balance is improved in response to GH treatment due to the minimization of proteinloss during fasting, and maximization of protein gain during meal absorption (Vann et al.,2000b). GH treatment in GH-deficient (Bier, 1991; Russell-Jones et al., 1998) and normal,mature animals and adult humans (Eisemann et al., 1989; Pell et al., 1990; Fryburg et al.,1991; Bell et al., 1998) increases protein deposition by stimulating whole-body and skeletalmuscle protein synthesis. Chronic GH treatment in cattle and swine increases amino aciduptake by the hindquarter (Boisclair et al., 1994; Bush et al., 2003a) and protein synthesis inmuscle (Eisemann et al., 1989; Seve et al., 1993) with no change in protein degradation acrossthe hindlimb (Bush et al., 2003a).


In young, growing swine, GH treatment increases skeletal muscle protein synthesis in thepostprandial state, but not in the fasting condition. This increase is due to modulation of trans-lational efficiency by GH and not by ribosome number (Bush et al., 2003b). The GH-inducedincrease in translation initiation is attributable to modulation of the factors associated with thebinding of both mRNA and met-tRNA to the ribosomal complex, that is, the phosphorylationof 4E-BP1, association of eIF4E with eIF4G, and eIF2B activity. Because GH increasescirculating IGF-I and insulin concentrations and this increase is greater in the fed than in thefasting state, the GH-induced increase in protein synthesis may involve mediation by IGF-Iand/or insulin, or may be due to a direct effect of GH. In fact, GH indirectly activates someof the same signalling components as insulin and IGF-I, i.e. IRS-1 and -2, PI 3-kinase, proteinkinase B, and S6K1 (Anderson, 1993; Yenush and White, 1997). In addition, the increasedsubstrate availability, i.e. amino acids, provided in the fed condition may be permissive forthe GH-induced increase in muscle protein synthesis.

4.6. Colostrum

Colostrum provides a rich source of nutrients for the newborn mammal that supports the rapidgrowth and accretion of body protein during the first few days of postnatal life (Burrin et al.,1997b). In addition to nutrients, colostrum also contains maternal immunoglobulins that formany species are essential for passive immunity, and a variety of bioactive components thatinclude insulin, IGF-I, IGF-II, and epidermal growth factor. Although the benefits of the con-sumption of nutrients and immune factors are readily apparent, the functional significance ofthe numerous hormones and growth factors present in colostrum is unclear. Studies that havecompared the growth of newborns have demonstrated an enhanced anabolic response in asso-ciation with the feeding of colostrum, especially of the visceral organs (Widdowson andCrabb, 1976; Widdowson et al., 1976). Given their mitogenic and anabolic properties, thisresponse was often attributed to the presence of trophic factors in the colostrum. However, itmust also be considered that the consumption of colostrum entails the ingestion of a largerquantity of nutrients than that typically provided by mature milk or, indeed, many formulas.

Studies designed to distinguish between the trophic effects of macronutrient intake andthose due to factors in colostrum (Burrin et al., 1995; Fiorotto et al., 2000b) showed that innewborn pigs, feeding stimulates protein synthesis in all tissues, but the stimulation of pro-tein synthesis in skeletal muscle is greater when colostrum, as opposed to a nutrient-matchedformula or mature sow’s milk, is fed. This suggests that the enhanced stimulation of skeletalmuscle protein synthesis in newborn pigs fed colostrum, as opposed to other feeds, is not duesolely to the provision of macronutrients. Furthermore, the stimulation of protein synthesisby colostrum feeding was restricted specifically to the myofibrillar proteins, unlike thegeneral stimulation of protein synthesis by feeding which incurred a proportional stimulationof the synthesis of both sarcoplasmic and myofibrillar proteins (Fiorotto et al., 2000b).Feeding also resulted in a general increase in muscle mRNA concentration, but in thecolostrum-fed piglets the enhanced synthesis rate of myofibrillar proteins was associated witha disproportionate increase in the abundance of myofibrillar mRNA, as exemplified by totalMHC mRNAs. Additionally, colostrum augmented the effect of feeding on proteinsynthesis by promoting a greater accretion of ribosomes.

Thus, feeding colostrum has both quantitative consequences for the anabolic process in theskeletal musculature of the newborn animal and qualitative consequences, with potentialimplications for the development of muscle function. Improvement of skeletal muscle


function is advantageous insofar as it is critical for the development of the newborn’s abilityto survive independently from its mother. The effects observed are likely attributable to non-nutritive factors present in colostrum, although these have not yet been identified. However,a number of potential factors, including insulin, IGF-I, thyroid hormone, and growth hor-mone, have been excluded. Identification of the mechanisms underlying this phenomenonwill be critical for advancing our understanding of the biological role of early mammarysecretions in the regulation of neonatal growth and in establishing how diet contributes to theregulation of skeletal muscle growth in early postnatal life.


There are numerous issues concerning the regulation of skeletal muscle growth and meta-bolism that need to be explored further. From the point of view of agriculture, the relativesignificance of these is determined by the economic benefits to be gained. The ultimate aimis to enhance feed efficiency. This, however, needs to be accomplished without compromis-ing meat quality, especially tenderness and fat content. However, it is becoming increasinglyevident that consumers are becoming more resistant to the use by the livestock industry ofanabolic agents, growth promoters, and antibiotics, and frequently are prepared to pay a pre-mium for products in which they have not been used. Although one may question the validityof these concerns, it must be acknowledged that they are widespread and, therefore, shouldnot be ignored. In this regard, the application of genomics and proteomics to select for breed-ing stock with desirable traits, and improved husbandry practices to reduce mortality andmorbidity in the birth to weaning period, are likely to be the most productive approaches.Given the large increase in fish consumption, research on the growth and composition ofmuscle of different fish varieties deserves substantially more attention.

Enhancement of muscle growth can be accomplished either by increasing the number ofmyofibres, or by promoting myofibre hypertrophy. As should be evident, the former is aprenatal event and is dictated by maternal and genetic factors. Thus, continued research onthe regulation of cell cycle progression and withdrawal of individual myoblast lineages, aswell as the factors that control terminal differentiation, are likely to yield relevant information,especially when this can be merged with genomic trait analysis. Because mechanistic studiesare difficult to conduct in vivo, in normal animals, much of the basic research on these mech-anisms must be performed in cell and tissue culture. However, the widespread use ofgenetically engineered mice has been most productive and helpful in this regard because,although far removed from livestock animals, transgenic mice provide an important and appo-site tool with which to assess the relevance of specific cellular events in the context of thewhole animal. “The Myostatin Knockout Story” presents an excellent example of the useful-ness of this approach.

For some mammalian species, the ability to increase muscle fibre number has limitations,however, because the enhanced fetal growth may increase maternal morbidity and/or com-promise maternal lactational capacity. In principle, therefore, enhanced postnatal musclehypertrophy would be preferable. As we have presented, postnatal hypertrophy is dictated bytwo key factors, satellite cell number and protein accretion. Satellite cell number is the bal-ance between the continued division of these cells acquired during the third phase of myoblastdetermination, and their loss, either by terminal differentiation and fusion into the myofibre,or by apoptosis. Although substantial progress has been made in understanding the factorsthat regulate satellite cell division, there are still many unanswered questions. Satellite cellshave their origins in the latter part of fetal life and, therefore, are likely to be influenced


by maternal variables; however, relatively little is known about the influence of maternalphysiology and metabolism on the satellite cells of the progeny. The factors, especially envi-ronmental factors, that dictate terminal differentiation and apoptosis of satellite cells, and theextent to which these processes can be manipulated through husbandry practices and diet, aremuch less clearly understood, and warrant a closer examination as they are likely to havelong-term consequences.

More recently it is has been demonstrated that under certain in vitro conditions, satellitecells can change lineage and form adipocytes. Clearly this has significant consequences notonly for overall muscle growth potential, but also for the composition of meat. The extent towhich this occurs in vivo, and the conditions that would favour such a change, merit furtherattention.

The rate of protein accretion is the balance between protein synthesis and degradation. Wehave demonstrated that in young animals the rate of protein synthesis is the principal regula-tory factor. This unique feature is attributable to the ability of the immature muscle tomarkedly increase translation when food is available. The latter is critical because it enablesamino acids to be diverted towards protein synthesis rather than to be oxidized. Thus, dietaryprotein can be used with greater efficiency, provided the composition of amino acids andenergy intake are optimal. Thus, from the nutritional standpoint, the ability to meter theamino acid composition of dietary protein to meet the needs for growth versus maintenanceduring development can enable maximal exploitation of this high synthetic capacity of theimmature muscle. The characteristics of the immature muscle that enable this syntheticresponse are primarily its high ribosomal content and an enhanced sensitivity and respon-siveness of muscle protein synthesis to insulin. Clearly, therefore, further understanding ofthose factors that are responsible for these unique features of the immature muscle, and theirdown-regulation with maturation, would be warranted. Moreover, it is equally important thatthe impact of environmental variables such as infection, temperature, activity (duration, type,and intensity), and dietary nutrients other than protein and energy (e.g. micronutrients,modified lipids, and various non-nutritive factors present in foodstuffs) on protein synthesisduring this anabolic phase of growth be investigated.

Our emphasis on protein synthesis rather than degradation does not negate the importanceof the latter in the regulation of protein accretion. Indeed, the regulation of protein degrada-tion potentially represents a much more energetically efficient approach for improving theefficiency of muscle protein deposition (Goll et al., 1989) beyond the early postnatal period.However, much less is understood about the in vivo regulation of protein degradation, espe-cially the factors that regulate myofibrillar breakdown, and the variability in these mechanismsbetween muscles and among different species. In addition to the consequences for proteindeposition, protein degradation has consequences for meat quality because those enzymesthat are responsible for the degradation of muscle myofibrillar proteins are also importantdeterminants of post-mortem meat tenderisation. The evidence would suggest that in domesticanimals muscle hypertrophy resulting from suppression of protein degradation in vivo cancompromise meat tenderness. Examples of this negative consequence of suppressing proteindegradation is the callipyge lamb in which the degree of hypertrophy of certain muscles ispositively correlated to calpastatin expression, and increased toughness. The effects of someβ-adrenergic agonists in certain species is similar to the effect of the callipyge gene. Thus, aclear understanding of the interplay between the structural characteristics of a muscle, the rel-ative contribution of protein synthesis versus degradation to its overall growth, the variationamong species, and how these aspects of muscle structure and metabolism are influenced byenvironmental factors and husbandry practices, are subjects that merit further study.


Recently there has been much effort refocused on skeletal muscle metabolism in humanswith the recognition that there is an inevitable depletion of skeletal muscle with ageing(sarcopenia). The consequence of this loss is quite severe as it results in a loss of strength,flexibility, and overall mobility, which thereby compromises the individual’s quality of life.The resulting decrease in activity not only exacerbates the muscle loss, but also decreasesbasal and activity-related energy expenditure, which therefore enhances the propensity forexcessive fat deposition and glucose intolerance. The causes of sarcopenia appear to be exten-sive and include the loss in the replicative capacity of satellite cells, age-related increases infactors that are antagonistic to muscle growth, such as myostatin and Id factors, and a loss inthe body’s capacity to produce anabolic agents such as growth hormone and testosterone. Therelative importance of these, however, is far from clear. Additionally, or possibly in conse-quence to these changes, skeletal muscle loses its regenerative capacity with ageing. As theaverage life expectancy of humans increases, understanding the causes of sarcopenia and thedevelopment of therapies and modalities to mitigate its occurrence has enormous economicimplications. Importantly, from the metabolic perspective, a better appreciation of the nutri-ent needs and dietary regimens that are required to sustain optimal muscle metabolism arewarranted.

REFERENCES

Adams, G.R., McCue, S.A., Zeng, M., Baldwin, K.M., 1999. Time course of myosin heavy chain transi-tions in neonatal rats: importance of innervation and thyroid state. Amer. J. Physiol. 276, R954–R961.

Allbrook, D.B., Han, M.F., Hellmuth, A.E., 1971. Population of muscle satellite cells in relation to ageand mitotic activity. Pathology 3, 233−243.

Allen, D.L., Leinwand, L.A., 2002. Intracellular calcium and myosin isoform transitions: calcineurin andcalcium-calmodulin kinase pathways regulate preferential activation of the IIa myosin heavy chainpromoter. J. Biol. Chem. 277, 45323−45330.

Allen, D.L., Roy, R.R., Edgerton, V.R., 1999. Myonuclear domains in muscle adaptation and disease.Muscle Nerve 22, 1350−1360.

Anderson, N.G., 1993. Simultaneous activation of p90rsk and p70s6k S6 kinases by growth hormone in3T3-F442A preadipocytes. Biochem. Biophys. Res. Commun. 193, 284−290.

Anthony, J.C., Anthony, T.G., Kimball, S.R., Vary, T.C., Jefferson, L.S., 2000. Orally administeredleucine stimulates protein synthesis in skeletal muscle of postabsorptive rats in association withincreased eIF4F formation. J. Nutr. 130, 139−145.

Attaix, D., Combaret, L., Taillandier, D., 1999. Mechanisms and regulation in protein degradation.In: Lobley, G.E., White, A., MacRae, J.C. (Eds.), Protein Metabolism and Nutrition. WageningenPress, Wageningen, The Netherlands, pp. 51−67.

Backer, J.M., Myers, M.G., Jr., Shoelson, S.E., Chin, D.J., Sun, X.J., Miralpeix, M., Hu, P., Margolis, B.,Skolnik, E.Y., Schlessinger, J., 1992. Phosphatidylinositol 3′-kinase is activated by association withIRS-1 during insulin stimulation. EMBO J. 11, 3469−3479.

Baillie, A.G., Garlick, P.J., 1992. Attenuated responses of muscle protein synthesis to fasting and insulinin adult female rats. Amer. J. Physiol. 262, E1−E5.

Balagopal, P., Rooyackers, O.E., Adey, D.B., Ades, P.A., Nair, K.S., 1997. Effects of aging on in vivosynthesis of skeletal muscle myosin heavy-chain and sarcoplasmic protein in humans. Amer. J.Physiol. 273, E790−E800.

Barany, M., Close, R.I., 1971. The transformation of myosin in cross-innervated rat muscles. J. Physiol.213, 455−474.

Bark, T.H., McNurlan, M.A., Lang, C.H., Garlick, P.J., 1998. Increased protein synthesis after acute IGF-Ior insulin infusion is localized to muscle in mice. Amer. J. Physiol. 275, E118−E123.

Bell, A.W., Bauman, D.E., Beermann, D.H., Harrell, R.J., 1998. Nutrition, development and efficacy ofgrowth modifiers in livestock species. J. Nutr. 128, 360S−363S.

Benezra, R., Davis, R.L., Lockshon, D., Turner, D.L., Weintraub, H., 1990. The protein Id: a negativeregulator of helix-loop-helix DNA binding proteins. Cell 61, 49−59.


Bennet, W.M., Connacher, A.A., Scrimgeour, C.M., Rennie, M.J., 1990. The effect of amino acid infu-sion on leg protein turnover assessed by L-[15N]phenylalanine and L-[1-13C]leucine exchange. Eur. J.Clin. Invest. 20, 41−50.

Bier, D.M., 1991. Growth hormone and insulin-like growth factor I: nutritional pathophysiology and ther-apeutic potential. Acta Paediatr. Scand., Suppl. 374, 119−128.

Birchmeier, C., Brohmann, H., 2000. Genes that control the development of migrating muscle precursorcells. Curr. Opin. Cell Biol. 12, 725−730.

Bischoff, R., 1994. The satellite cell and muscle regeneration. In: Engel, A.G., Franzini-Armstrong, C. (Eds.),Myology, McGraw-Hill, New York, Vol. 1, pp. 97−118.

Boirie, Y., Short, K.R., Ahlman, B., Charlton, M., Nair, K.S., 2001. Tissue-specific regulation ofmitochondrial and cytoplasmic protein synthesis rates by insulin. Diabetes 50, 2652−2658.

Boisclair, Y.R., Bauman, D.E., Bell, A.W., Dunshea, F.R., Harkins, M., 1994. Nutrient utilizationand protein turnover in the hindlimb of cattle treated with bovine somatotropin. J. Nutr. 124,664−673.

Boyle, D.W., Denne, S.C., Moorehead, H., Lee, W.H., Bowsher, R.R., Liechty, E.A., 1998. Effect ofrhIGF-I infusion on whole fetal and fetal skeletal muscle protein metabolism in sheep. Amer.J. Physiol. 275, E1082−E1091.

Brohmann, H., Jagla, K., Birchmeier, C., 2000. The role of Lbx1 in migration of muscle precursor cells.Development 127, 437−445.

Buckingham, M., 2001. Skeletal muscle formation in vertebrates. Curr. Opin. Genet. Dev. 11, 440−448.Buckingham, M., Bajard, L., Chang, T., Daubas, P., Hadchouel, J., Meilhac, S., Montarras, D.,

Rocancourt, D., Relaix, F., 2003. The formation of skeletal muscle: from somite to limb. J. Anat. 202,59−68.

Buonomo, F.C., Baile, C.A., 1991. Influence of nutritional deprivation on insulin-like growth factor I,somatotropin, and metabolic hormones in swine. J. Anim. Sci. 69, 755−760.

Burrin, D.G., Davis, T.A., Ebner, S., Schoknecht, P.A., Fiorotto, M.L., Reeds, P.J., 1997a. Colostrumenhances the nutritional stimulation of vital organ protein synthesis in neonatal pigs. J. Nutr. 127,1284−1289.

Burrin, D.G., Davis, T.A., Ebner, S., Schoknecht, P.A., Fiorotto, M.L., Reeds, P.J., McAvoy, S., 1995.Nutrient-independent and nutrient-dependent factors stimulate protein synthesis in colostrum-fednewborn pigs. Pediat. Res. 37, 593−599.

Burrin, D.G., Davis, T.A., Fiorotto, M.L., Reeds, P.J., 1991. Stage of development and fasting affectprotein synthetic activity in the gastrointestinal tissues of suckling rats. J. Nutr. 121, 1099−1108.

Burrin, D.G., Davis, T.A., Fiorotto, M.L., Reeds, P.J., 1997b. Role of milk-borne vs endogenous insulin-like growth factor I in neonatal growth. J. Anim. Sci. 75, 2739−2743.

Bush, J.A., Burrin, D.G., Suryawan, A., O’Connor, P.M., Nguyen, H.V., Reeds, P.J., Steele, N.C.,van Goudoever, J.B., Davis, T.A., 2003a. Somatotropin-induced protein anabolism in hindquartersand portal-drained viscera of growing pigs. Amer. J. Physiol. Endocrinol. Metab. 284, E302−E312.

Bush, J.A., Kimball, S.R., O’Connor, P.M., Suryawan, A., Orellana, R.A., Nguyen, H.V., Jefferson, L.S.,Davis, T.A., 2003b. Translational control of protein synthesis in muscle and liver of growth hormone-treated pigs. Endocrinology 144, 1273−1283.

Calvo, S., Vullhorst, D., Venepally, P., Cheng, J., Karavanova, I., Buonanno, A., 2001. Molecular dissec-tion of DNA sequences and factors involved in slow muscle-specific transcription. Mol. Cell Biol. 21,8490−8503.

Campbell, R.G., Johnson, R.J., King, R.H., Taverner, M.R., Meisinger, D.J., 1990. Interaction of dietaryprotein content and exogenous porcine growth hormone administration on protein and lipid accretionrates in growing pigs. J. Anim. Sci. 68, 3217−3225.

Campion, D.R., Richardson, R.L., Reagan, J.O., Kraeling, R.R., 1981. Changes in the satellite cell pop-ulation during postnatal growth of pig skeletal muscle. J. Anim Sci. 52, 1014−1018.

Caperna, T.J., Komarek, D.R., Gavelek, D., Steele, N.C., 1991. Influence of dietary protein and recom-binant porcine somatotropin administration in young pigs: II. Accretion rates of protein, collagen, andfat. J. Anim. Sci. 69, 4019−4029.

Chakravarthy, M.V., Fiorotto, M.L., Schwartz, R.J., Booth, F.W., 2001. Long-term insulin-like growthfactor-I expression in skeletal muscles attenuates the enhanced in vitro proliferation ability of theresident satellite cells in transgenic mice. Mech. Ageing Dev. 122, 1303−1320.

Cheatham, B., Kahn, C.R., 1995. Insulin action and the insulin signaling network. Endocr. Rev. 16,117−142.


Coolican, S.A., Samuel, D.S., Ewton, D.Z., McWade, F.J., Florini, J.R., 1997. The mitogenic and myo-genic actions of insulin-like growth factors utilize distinct signaling pathways. J. Biol. Chem. 272,6653−6662.

Dardevet, D., Sornet, C., Bayle, G., Prugnaud, J., Pouyet, C., Grizard, J., 2002. Postprandial stimulationof muscle protein synthesis in old rats can be restored by a leucine-supplemented meal. J. Nutr. 132,95−100.

Dardevet, D., Sornet, C., Vary, T., Grizard, J., 1996. Phosphatidylinositol 3-kinase and p70 S6 kinase par-ticipate in the regulation of protein turnover in skeletal muscle by insulin and insulin-like growthfactor I. Endocrinology 137, 4087−4094.

Davis, T.A., Burrin, D.G., Fiorotto, M.L., Nguyen, H.V., 1996. Protein synthesis in skeletal muscleand jejunum is more responsive to feeding in 7- than 26-day-old pigs. Amer. J. Physiol. 270,E802−E809.

Davis, T.A., Burrin, D.G., Fiorotto, M.L., Reeds, P.J., Jahoor, F., 1998. Roles of insulin and amino acidsin the regulation of protein synthesis in the neonate. J. Nutr. 128, 347S−350S.

Davis, T.A., Fiorotto, M.L., Beckett, P.R., Burrin, D.G., Reeds, P.J., Wray-Cahen, D., Nguyen, H.V.,2001. Differential effects of insulin on peripheral and visceral tissue protein synthesis in neonatalpigs. Amer. J. Physiol. Endocrinol. Metab. 280, E770−E779.

Davis, T.A., Fiorotto, M.L., Burrin, D.G., Pond, W.G., Nguyen, H.V., 1997. Intrauterine growth restric-tion does not alter response of protein synthesis to feeding in newborn pigs. Amer. J. Physiol. 272,E877−E884.

Davis, T.A., Fiorotto, M.L., Burrin, D.G., Reeds, P.J., Nguyen, H.V., Beckett, P.R., Vann, R.C.,O’Connor, P.M., 2002a. Stimulation of protein synthesis by both insulin and amino acids is unique toskeletal muscle in neonatal pigs. Amer. J. Physiol. Endocrinol. Metab. 282, E880−E890.

Davis, T.A., Fiorotto, M.L., Burrin, D.G., Vann, R.C., Reeds, P.J., Nguyen, H.V., Beckett, P.R., Bush, J.A.,2002b. Acute IGF-I infusion stimulates protein synthesis in skeletal muscle and other tissues ofneonatal pigs. Amer. J. Physiol. Endocrinol. Metab. 283, E638−E647.

Davis, T.A., Fiorotto, M.L., Nguyen, H.V., Burrin, D.G., Reeds, P.J., 1991. Response of muscle proteinsynthesis to fasting in suckling and weaned rats. Amer. J. Physiol. 261, R1373−R1380.

Davis, T.A., Fiorotto, M.L., Nguyen, H.V., Reeds, P.J., 1989. Protein turnover in skeletal muscle of sucklingrats. Amer. J. Physiol. 257, R1141−R1146.

Davis, T.A., Fiorotto, M.L., Nguyen, H.V., Reeds, P.J., 1993b. Enhanced response of muscle proteinsynthesis and plasma insulin to food intake in suckled rats. Amer. J. Physiol. 265, R334−R340.

Davis, T.A., Fiorotto, M.L., Reeds, P.J., 1993a. Amino acid compositions of body and milk proteinchange during the suckling period in rats. J. Nutr. 123, 947−956.

Davis, T.A., Klahr, S., Karl, I.E., 1987. Insulin-stimulated protein metabolism in chronic azotemia andexercise. Amer. J. Physiol. 253, F164−F169.

Davis, T.A., Nguyen, H.V., Suryawan, A., Bush, J.A., Jefferson, L.S., Kimball, S.R., 2000.Developmental changes in the feeding-induced stimulation of translation initiation in muscle ofneonatal pigs. Amer. J. Physiol. Endocrinol. Metab. 279, E1226−E1234.

Denne, S.C., Kalhan, S.C., 1987. Leucine metabolism in human newborns. Amer. J. Physiol. 253,E608−E615.

Denne, S.C., Rossi, E.M., Kalhan, S.C., 1991. Leucine kinetics during feeding in normal newborns.Pediat. Res. 30, 23−27.

Devlin, R.B., Emerson, C.P. Jr., 1978. Coordinate regulation of contractile protein synthesis duringmyoblast differentiation. Cell 13, 599−611.

Devlin, R.B., Emerson, C.P. Jr., 1979. Coordinate accumulation of contractile protein mRNAs duringmyoblast differentiation. Dev. Biol. 69, 202−216.

Donovan, S.M., Atilano, L.C., Hintz, R.L., Wilson, D.M., Rosenfeld, R.G., 1991. Differential regulationof the insulin-like growth factors (IGF-I and -II) and IGF binding proteins during malnutrition in theneonatal rat. Endocrinology 129, 149−157.

Douglas, R.G., Gluckman, P.D., Ball, K., Breier, B., Shaw, J.H., 1991. The effects of infusion of insulin-like growth factor (IGF) I, IGF-II, and insulin on glucose and protein metabolism in fasted lambs.J. Clin. Invest. 88, 614−622.

Ebner, S., Schoknecht, P., Reeds, P., Burrin, D., 1994. Growth and metabolism of gastrointestinal andskeletal muscle tissues in protein-malnourished neonatal pigs. Amer. J. Physiol. 266, R1736−R1743.

Eisemann, J.H., Hammond, A.C., Rumsey, T.S., 1989. Tissue protein synthesis and nucleic acid concen-trations in steers treated with somatotropin. Brit. J. Nutr. 62, 657−671.


Engert, J.C., Berglund, E.B., Rosenthal, N., 1996. Proliferation precedes differentiation in IGF-I-stimulated myogenesis. J. Cell Biol. 135, 431−440.

Epstein, J.A., Shapiro, D.N., Cheng, J., Lam, P.Y., Maas, R.L., 1996. Pax3 modulates expression of thec-Met receptor during limb muscle development. Proc. Natl. Acad. Sci. USA 93, 4213−4218.

Fiorotto, M.L., Davis, T.A., 1997. Food intake alters muscle protein gain with little effect on Na+–K+-ATPase and myosin isoforms in suckled rats. Amer. J. Physiol. 272, R1461−R1471.

Fiorotto, M.L., Burrin, D.G., Perez, M., Reeds, P.J., 1991. Intake and use of milk nutrients by rat pupssuckled in small, medium, or large litters. Amer. J. Physiol. 260, R1104−R1113.

Fiorotto, M.L., Davis, T.A., Reeds, P.J., 2000a. Regulation of myofibrillar protein turnover during matu-ration in normal and undernourished rat pups. Amer. J. Physiol. 278, 845−854.

Fiorotto, M.L., Davis, T.A., Reeds, P.J., Burrin, D.G., 2000b. Nonnutritive factors in colostrum enhancemyofibrillar protein synthesis in the newborn pig. Pediat. Res. 48, 1−7.

Fiorotto, M.L., Schwartz, R.J., Delaughter, M.C., 2003. Persistent IGF-I overexpression in skeletalmuscle transiently enhances DNA accretion and growth. FASEB J. 17, 59−60.

Fox, H.L., Pham, P.T., Kimball, S.R., Jefferson, L.S., Lynch, C.J., 1998. Amino acid effects on transla-tional repressor 4E-BP1 are mediated primarily by L-leucine in isolated adipocytes. Amer. J. Physiol.275, C1232−C1238.

Francis-West, P.H., Antoni, L., Anakwe, K., 2003. Regulation of myogenic differentiation in the devel-oping limb bud. J. Anat. 202, 69−81.

Fryburg, D.A., Gelfand, R.A., Barrett, E.J., 1991. Growth hormone acutely stimulates forearm muscleprotein synthesis in normal humans. Amer. J. Physiol. 260, E499−E504.

Fryburg, D.A., Jahn, L.A., Hill, S.A., Oliveras, D.M., Barrett, E.J., 1995. Insulin and insulin-like growthfactor-I enhance human skeletal muscle protein anabolism during hyperaminoacidemia by differentmechanisms. J. Clin. Invest. 96, 1722−1729.

Fuchtbauer, E.M., 2002. Inhibition of skeletal muscle development: less differentiation gives moremuscle. Results Probl. Cell Differ. 38, 143−161.

Gambke, B., Lyons, G.E., Haselgrove, J., Kelly, A.M., Rubinstein, N.A., 1983. Thyroidal and neuralcontrol of myosin transitions during development of rat fast and slow muscles. FEBS Lett. 156,335−339.

Garlick, P.J., Fern, M., Preedy, V.R., 1983. The effect of insulin infusion and food intake on muscle pro-tein synthesis in postabsorptive rats. Biochem. J. 210, 669−676.

Gelfand, R.A., Barrett, E.J., 1987. Effect of physiologic hyperinsulinemia on skeletal muscle protein syn-thesis and breakdown in man. J. Clin. Invest. 80, 1−6.

Gingras, A.C., Raught, B., Sonenberg, N., 1999. eIF4 initiation factors: effectors of mRNA recruitmentto ribosomes and regulators of translation. Annu. Rev. Biochem. 68, 913−963.

Goldstein, S., Harp, J.B., Phillips, L.S., 1991. Nutrition and somatomedin. XXII: Molecular regulationof insulin-like growth factor-I during fasting and refeeding in rats. J. Mol. Endocrinol. 6, 33−43.

Goll, D.E., Kleese, W.C., Szapchenko, A., 1989. Skeletal muscle proteases and protein turnover. In:Campion, D.R., Hausman, G.J., Martin, R.J. (Eds.), Animal Growth Regulation. Plenum PublishingCompany, New York, pp. 141−182.

Goll, D.E., Thompson, V.F., Taylor, R.G., Christiansen, J.A., 1992. Role of the calpain system in musclegrowth. Biochimie 74, 225−237.

Grobet, L., Martin, L.J., Poncelet, D., Pirottin, D., Brouwers, B., Riquet, J., Schoeberlein, A., Dunner, S.,Menissier, F., Massabanda, J., Fries, R., Hanset, R., Georges, M., 1997. A deletion in the bovine myo-statin gene causes the double-muscled phenotype in cattle. Nat. Genet. 17, 71−74.

Handschin, C., Rhee, J., Lin, J., Tarr, P.T., Spiegelman, B.M., 2003. An autoregulatory loop controls per-oxisome proliferator-activated receptor gamma coactivator 1α expression in muscle. Proc. Natl. Acad.Sci. USA 100, 7111−7116.

Hannan, K.M., Kennedy, B.K., Cavanaugh, A.H., Hannan, R.D., Hirschler-Laszkiewicz, I., Jefferson, L.S.,Rothblum, L.I., 2000. RNA polymerase I transcription in confluent cells: Rb downregulates rDNAtranscription during confluence-induced cell cycle arrest. Oncogene 19, 3487−3497.

Hara, K., Yonezawa, K., Weng, Q.P., Kozlowski, M.T., Belham, C., Avruch, J., 1998. Amino acid suffi-ciency and mTOR regulate p70 S6 kinase and eIF-4E BP1 through a common effector mechanism.J. Biol. Chem. 273, 14484−14494.

Harmon, C.S., Proud, C.G., Pain, V.M., 1984. Effects of starvation, diabetes and acute insulintreatment on the regulation of polypeptide-chain initiation in rat skeletal muscle. Biochem. J. 223,687−696.


Harrison, A.P., Tivey, D.R., Clausen, T., Duchamp, C., Dauncey, M.J., 1996. Role of thyroid hormonesin early postnatal development of skeletal muscle and its implications for undernutrition. Brit. J. Nutr.76, 841−855.

Hemel-Grooten, H.N., Koohmaraie, M., Yen, J.T., Arbona, J.R., Rathmacher, J.A., Nissen, S.L.,Fiorotto, M.L., Garssen, G.J., Verstegen, M.W., 1995. Comparison between 3-methylhistidineproduction and proteinase activity as measures of skeletal muscle breakdown in protein-deficientgrowing barrows. J. Anim. Sci. 73, 2272−2281.

Henriksson, J., 1990. The possible role of skeletal muscle in the adaptation to periods of energydeficiency. Eur. J. Clin. Nutr. 44, Suppl. 1, 55−64.

Heslin, M.J., Newman, E., Wolf, R.F., Pisters, P.W., Brennan, M.F., 1992. Effect of hyperinsulinemiaon whole body and skeletal muscle leucine carbon kinetics in humans. Amer. J. Physiol. 262,E911−E918.

Irrcher, I., Adhihetty, P.J., Sheehan, T., Joseph, A.M., Hood, D.A., 2003. PPAR-γ coactivator-1α expres-sion during thyroid hormone- and contractile activity-induced mitochondrial adaptations. Amer.J. Physiol. Cell Physiol. 284, C1669−C1677.

Jefferies, H.B., Reinhard, C., Kozma, S.C., Thomas, G., 1994. Rapamycin selectively represses transla-tion of the “polypyrimidine tract” mRNA family. Proc. Natl. Acad. Sci. USA 91, 4441−4445.

Jefferson, L.S., Kimball, S.R., 2001. Amino acid regulation of gene expression. J. Nutr. 131,2460S−2466S.

Jefferson, L.S., Li, J.B., Rannels, S.R., 1977. Regulation by insulin of amino acid release and proteinturnover in the perfused rat hemicorpus. J. Biol. Chem. 252, 1476−1483.

Katsumata, M., Cattaneo, D., White, P., Burton, K.A., Dauncey, M.J., 2000. Growth hormone receptorgene expression in porcine skeletal and cardiac muscles is selectively regulated by postnatal under-nutrition. J. Nutr. 130, 2482−2488.

Kelly, F.J., Lewis, S.E., Anderson, P., Goldspink, D.F., 1984. Pre- and postnatal growth and proteinturnover in four muscles of the rat. Muscle Nerve 7, 235−242.

Kimball, S.R., Jefferson, L.S., 1988. Cellular mechanisms involved in the action of insulin on proteinsynthesis. Diabetes Metab. Rev. 4, 773−787.

Kimball, S.R., Farrell, P.A., Nguyen, H.V., Jefferson, L.S., Davis, T.A., 2002. Developmental decline incomponents of signal transduction pathways regulating protein synthesis in pig muscle. Amer.J. Physiol. Endocrinol. Metab. 282, E585−E592.

Kimball, S.R., Horetsky, R.L., Jefferson, L.S., 1998. Signal transduction pathways involved in the regu-lation of protein synthesis by insulin in L6 myoblasts. Amer. J. Physiol. 274, C221−C228.

Kimball, S.R., Jefferson, L.S., Nguyen, H.V., Suryawan, A., Bush, J.A., Davis, T.A., 2000. Feeding stim-ulates protein synthesis in muscle and liver of neonatal pigs through an mTOR-dependent process.Amer. J. Physiol. Endocrinol. Metab. 279, E1080−E1087.

Kimball, S.R., Jurasinski, C.V., Lawrence, J.C. Jr., Jefferson, L.S., 1997. Insulin stimulates proteinsynthesis in skeletal muscle by enhancing the association of eIF-4E and eIF-4G. Amer. J. Physiol.272, C754−C759.

Kimball, S.R., Mellor, H., Flowers, K.M., Jefferson, L.S., 1996. Role of translation initiation factoreIF-2B in the regulation of protein synthesis in mammalian cells. Prog. Nucleic Acid Res. Mol. Biol.54, 165−196.

Kimball, S.R., Vary, T.C., Jefferson, L.S., 1994. Regulation of protein synthesis by insulin. Annu. Rev.Physiol. 56, 321−348.

Koohmaraie, M., 2003. Meat tenderness and muscle growth: is there relationship? Meat Sci. 62,345−352.

Langley, B., Thomas, M., Bishop, A., Sharma, M., Gilmour, S., Kambadur, R., 2002. Myostatininhibits myoblast differentiation by down-regulating MyoD expression. J. Biol. Chem. 277,49831−49840.

Lee, S.J., McPherron, A.C., 1999. Myostatin and the control of skeletal muscle mass. Curr. Opin. Genet.Dev. 9, 604−607.

LeRoith, D., Werner, H., Beitner-Johnson, D., Roberts, C.T. Jr., 1995. Molecular and cellular aspects ofthe insulin-like growth factor I receptor. Endocr. Rev. 16, 143−163.

Liechty, E.A., Boyle, D.W., Moorehead, H., Liu, Y.M., Denne, S.C., 1992. Effect of hyperinsuline-mia on ovine fetal leucine kinetics during prolonged maternal fasting. Amer. J. Physiol. 263,E696−E702.


Lin, T.A., Kong, X., Haystead, T.A., Pause, A., Belsham, G., Sonenberg, N., Lawrence, J.C. Jr., 1994.PHAS-I as a link between mitogen-activated protein kinase and translation initiation. Science 266,653−656.

Long, W., Saffer, L., Wei, L., Barrett, E.J., 2000. Amino acids regulate skeletal muscle PHAS-I and p70S6-kinase phosphorylation independently of insulin. Amer. J. Physiol. Endocrinol. Metab. 279,E301−E306.

Louard, R.J., Fryburg, D.A., Gelfand, R.A., Barrett, E.J., 1992. Insulin sensitivity of protein and glucosemetabolism in human forearm skeletal muscle. J. Clin. Invest. 90, 2348−2354.

Louveau, I., Le Dividich, J., 2002. GH and IGF-I binding in adipose tissue, liver, and skeletal muscle inresponse to milk intake level in piglets. Gen. Comp. Endocrinol. 126, 310−317.

McCracken, K.J., Eddie, S.M., Stevenson, W.G., 1980. Energy and protein nutrition of early-weanedpigs. 1. Effect of energy intake and energy:protein on growth, efficiency and nitrogen utilization ofpigs between 8–32 days. Brit. J. Nutr. 43, 289−304.

McKinsey, T.A., Zhang, C.L., Olson, E.N., 2001. Control of muscle development by dueling HATs andHDACs. Curr. Opin. Genet. Dev. 11, 497−504.

McKinsey, T.A., Zhang, C.L., Olson, E.N., 2002. Signaling chromatin to make muscle. Curr. Opin. CellBiol. 14, 763−772.

McNulty, P.H., Young, L.H., Barrett, E.J., 1993. Response of rat heart and skeletal muscle protein in vivoto insulin and amino acid infusion. Amer. J. Physiol. 264, E958−E965.

McPherron, A.C., Lawler, A.M., Lee, S.J., 1997. Regulation of skeletal muscle mass in mice by a newTGF-β superfamily member. Nature 387, 83−90.

Melville, S., McNurlan, M.A., McHardy, K.C., Broom, J., Milne, E., Calder, A.G., Garlick, P.J.,1989. The role of degradation in the acute control of protein balance in adult man: failure offeeding to stimulate protein synthesis as assessed by L-[1-13C]leucine infusion. Metabolism 38,248−255.

Mesires, N.T., Doumit, M.E., 2002. Satellite cell proliferation and differentiation during postnatal growthof porcine skeletal muscle. Amer. J. Physiol. Cell Physiol. 282, C899−C906.

Mitchell, A.D., Scholz, A.M., Mersmann, H.J., 2001. Growth and body composition. In: Pond, W.G.,Mersmann, H.J. (Eds.), Biology of the Domestic Pig. Comstock Publishing Association, Ithaca, NY,pp. 225−308.

Mitchell, P.J., Johnson, S.E., Hannon, K., 2002. Insulin-like growth factor I stimulates myoblastexpansion and myofiber development in the limb. Dev. Dyn. 223, 12−23.

Molkentin, J.D., Olson, E.N., 1996. Combinatorial control of muscle development by basichelix-loop-helix and MADS-box transcription factors. Proc. Natl. Acad. Sci. USA 93,9366−9373.

Musaro, A., McCullagh, K., Paul, A., Houghton, L., Dobrowolny, G., Molinaro, M., Barton, E.R.,Sweeney, H.L., Rosenthal, N., 2001. Localized Igf-1 transgene expression sustains hypertrophy andregeneration in senescent skeletal muscle. Nat. Genet. 27, 195–200.

Nakayama, M., Stauffer, J., Cheng, J., Banerjee-Basu, S., Wawrousek, E., Buonanno, A., 1996. Commoncore sequences are found in skeletal muscle slow- and fast-fiber-type-specific regulatory elements.Mol. Cell Biol. 16, 2408−2417.

O’Connor, P.M., Bush, J.A., Suryawan, A., Nguyen, H.V., Davis, T.A., 2003. Insulin and amino acidsindependently stimulate skeletal muscle protein synthesis in neonatal pigs. Amer. J. Physiol.Endocrinol. Metab. 284, E110−E119.

Oddy, V.H., Owens, P.C., 1996. Insulin-like growth factor I inhibits degradation and improves retentionof protein in hindlimb muscle of lambs. Amer. J. Physiol. 271, E973−E982.

Oddy, V.H., Lindsay, D.B., Barker, P.J., Northrop, A.J., 1987. Effect of insulin on hind-limb and whole-body leucine and protein metabolism in fed and fasted lambs. Brit. J. Nutr. 58, 437−452.

Ojuka, E.O., Jones, T.E., Han, D.H., Chen, M., Holloszy, J.O., 2003. Raising Ca2+ in L6 myotubesmimics effects of exercise on mitochondrial biogenesis in muscle. FASEB J. 17, 675−681.

Ontell, M., 1982. The growth and metabolism of developing muscle. In: Jones, C.T. (Ed.),Biochemical Development of the Fetus and Neonate. Elsevier Biomedical Press, Amsterdam,pp. 213−247.

Ontell, M., Dunn, R.F., 1978. Neonatal muscle growth: a quantitative study. Amer. J. Anat. 152,539−556.

Pain, V.M., 1996. Initiation of protein synthesis in eukaryotic cells. Eur. J. Biochem. 236, 747−771.


Patti, M.E., Brambilla, E., Luzi, L., Landaker, E.J., Kahn, C.R., 1998. Bidirectional modulation of insulinaction by amino acids. J. Clin. Invest 101, 1519−1529.

Pause, A., Belsham, G.J., Gingras, A.C., Donze, O., Lin, T.A., Lawrence, J.C. Jr., Sonenberg, N.,1994. Insulin-dependent stimulation of protein synthesis by phosphorylation of a regulator of 5′-capfunction. Nature 371, 762−767.

Pell, J.M., Elcock, C., Harding, R.L., Morrell, D.J., Simmonds, A.D., Wallis, M., 1990. Growth, bodycomposition, hormonal and metabolic status in lambs treated long-term with growth hormone. Brit.J. Nutr. 63, 431−445.

Pellett, P.L., Kaba, H., 1972. Carcass amino acids of the rat under conditions of determination of netprotein utilization. J. Nutr. 102, 61−68.

Perry, R.L., Rudnicki, M.A., 2000. Molecular mechanisms regulating myogenic determination anddifferentiation. Front Biosci. 5, D750−D767.

Preedy, V.R., Garlick, P.J., 1986. The response of muscle protein synthesis to nutrient intake in postab-sorptive rats: the role of insulin and amino acids. Biosci. Rep. 6, 177−183.

Rehfeldt, C., Fiedler, I., Weikard, R., Kanitz, E., Ender, K., 1993. It is possible to increase skeletal musclefibre number in utero. Biosci. Rep. 13, 213−220.

Rehfeldt, C., Stickland, N.C., Fiedler, I., Wegner, J., 1999. Environmental and genetic factors as sourcesof variation in skeletal muscle fibre number. Basic Appl. Myol. 9, 235−253.

Rhoads, R.E., Joshi, B., Minich, W.B., 1994. Participation of initiation factors in the recruitment ofmRNA to ribosomes. Biochimie 76, 831−838.

Rosenblatt, J.D., Parry, D.J., 1992. Gamma irradiation prevents compensatory hypertrophy of overloadedmouse extensor digitorum longus muscle. J. Appl. Physiol. 73, 2538−2543.

Russell-Jones, D.L., Bowes, S.B., Rees, S.E., Jackson, N.C., Weissberger, A.J., Hovorka, R., Sonksen, P.H.,Umpleby, A.M., 1998. Effect of growth hormone treatment on postprandial protein metabolism ingrowth hormone-deficient adults. Amer. J. Physiol. 274, E1050−E1056.

Scaal, M., Bonafede, A., Dathe, V., Sachs, M., Cann, G., Christ, B., Brand-Saberi, B., 1999. SF/HGF isa mediator between limb patterning and muscle development. Development 126, 4885−4893.

Schiaffino, S., Margreth, A., 1969. Coordinated development of the sarcoplasmic reticulum and T systemduring postnatal differentiation of rat skeletal muscle. J. Cell Biol. 41, 855−875.

Schiaffino, S., Reggiani, C., 1996. Molecular diversity of myofibrillar proteins: gene regulation andfunctional significance. Physiol. Rev. 76, 371−423.

Schiaffino, S., Serrano, A., 2002. Calcineurin signaling and neural control of skeletal muscle fiber typeand size. Trends Pharmacol. Sci. 23, 569−575.

Seale, P., Rudnicki, M.A., 2000. A new look at the origin, function, and “stem-cell” status of musclesatellite cells. Dev. Biol. 218, 115−124.

Seale, P., Sabourin, L.A., Girgis-Gabardo, A., Mansouri, A., Gruss, P., Rudnicki, M.A., 2000. Pax7 isrequired for the specification of myogenic satellite cells. Cell 102, 777−786.

Seve, B., Ballevre, O., Ganier, P., Noblet, J., Prugnaud, J., Obled, C., 1993. Recombinant porcinesomatotropin and dietary protein enhance protein synthesis in growing pigs. J. Nutr. 123,529−540.

Shani, M., Zevin-Sonkin, D., Saxel, O., Carmon, Y., Katcoff, D., Nudel, U., Yaffe, D., 1981. Thecorrelation between the synthesis of skeletal muscle actin, myosin heavy chain, and myosin lightchain and the accumulation of corresponding mRNA sequences during myogenesis. Dev. Biol. 86,483−492.

Shields, R.G. Jr., Mahan, D.C., Graham, P.L., 1983. Changes in swine body composition from birth to145 kg. J. Anim. Sci. 57, 43−54.

Sonenberg, N., 1994. Regulation of translation and cell growth by eIF-4E. Biochimie 76, 839−846.Spangenburg, E.E., Booth, F.W., 2003. Molecular regulation of individual skeletal muscle fibre types.

Acta Physiol. Scand. 178, 413−424.Sun, X.J., Crimmins, D.L., Myers, M.G. Jr., Miralpeix, M., White, M.F., 1993. Pleiotropic insulin signals

are engaged by multisite phosphorylation of IRS-1. Mol. Cell Biol. 13, 7418−7428.Sun, X.J., Rothenberg, P., Kahn, C.R., Backer, J.M., Araki, E., Wilden, P.A., Cahill, D.A., Goldstein, B.J.,

White, M.F., 1991. Structure of the insulin receptor substrate IRS-1 defines a unique signal transduc-tion protein. Nature 352, 73−77.

Suryawan, A., Nguyen, H.V., Bush, J.A., Davis, T.A., 2001. Developmental changes in the feeding-induced activation of the insulin-signaling pathway in neonatal pigs. Amer. J. Physiol. Endocrinol.Metab. 281, E908−E915.


Svanberg, E., Zachrisson, H., Ohlsson, C., Iresjo, B.M., Lundholm, K.G., 1996. Role of insulin and IGF-Iin activation of muscle protein synthesis after oral feeding. Amer. J. Physiol. 270, E614−E620.

Tessari, P., Zanetti, M., Barazzoni, R., Vettore, M., Michielan, F., 1996. Mechanisms of postprandialprotein accretion in human skeletal muscle: insight from leucine and phenylalanine forearm kinetics.J. Clin. Invest. 98, 1361−1372.

Thureen, P.J., Scheer, B., Anderson, S.M., Tooze, J.A., Young, D.A., Hay, W.W. Jr., 2000. Effect of hyper-insulinemia on amino acid utilization in the ovine fetus. Amer. J. Physiol. Endocrinol. Metab. 279,E1294−E1304.

Ullrich, A., Gray, A., Tam, A.W., Yang-Feng, T., Tsubokawa, M., Collins, C., Henzel, W., Le Bon, T.,Kathuria, S., Chen, E., 1986. Insulin-like growth factor I receptor primary structure: comparisonwith insulin receptor suggests structural determinants that define functional specificity. EMBO J. 5,2503−2512.

VandeHaar, M.J., Moats-Staats, B.M., Davenport, M.L., Walker, J.L., Ketelslegers, J.M., Sharma, B.K.,Underwood, L.E., 1991. Reduced serum concentrations of insulin-like growth factor-I (IGF-I) inprotein-restricted growing rats are accompanied by reduced IGF-I mRNA levels in liver and skeletalmuscle. J. Endocrinol. 130, 305−312.

Vann, R.C., Nguyen, H.V., Reeds, P.J., Burrin, D.G., Fiorotto, M.L., Steele, N.C., Deaver, D.R., Davis, T.A.,2000a. Somatotropin increases protein balance by lowering body protein degradation in fed, growingpigs. Amer. J. Physiol. Endocrinol. Metab. 278, E477−E483.

Vann, R.C., Nguyen, H.V., Reeds, P.J., Steele, N.C., Deaver, D.R., Davis, T.A., 2000b. Somatotropinincreases protein balance independent of insulin’s effects on protein metabolism in growing pigs.Amer. J. Physiol. Endocrinol. Metab. 279, E1−E10.

Vary, T.C., Jefferson, L.S., Kimball, S.R., 1999. Amino acid-induced stimulation of translation initiationin rat skeletal muscle. Amer. J. Physiol. 277, E1077−E1086.

Vary, T.C., Jefferson, L.S., Kimball, S.R., 2000. Role of eIF4E in stimulation of protein synthesis by IGF-Iin perfused rat skeletal muscle. Amer. J. Physiol. Endocrinol. Metab. 278, E58−E64.

Venuti, J.M., Morris, J.H., Vivian, J.L., Olson, E.N., Klein, W.H., 1995. Myogenin is required for late butnot early aspects of myogenesis during mouse development. J. Cell Biol. 128, 563−576.

Volpi, E., Ferrando, A.A., Yeckel, C.W., Tipton, K.D., Wolfe, R.R., 1998. Exogenous amino acidsstimulate net muscle protein synthesis in the elderly. J. Clin. Invest. 101, 2000−2007.

von Manteuffel, S.R., Dennis, P.B., Pullen, N., Gingras, A.C., Sonenberg, N., Thomas, G., 1997. Theinsulin-induced signalling pathway leading to S6 and initiation factor 4E binding protein 1 phospho-rylation bifurcates at a rapamycin-sensitive point immediately upstream of P70S6K. Mol. Cell Biol.17, 5426−5436.

Wade, R., Sutherland, C., Gahlmann, R., Kedes, L., Hardeman, E., Gunning, P., 1990. Regulation ofcontractile protein gene family mRNA pool sizes during myogenesis. Dev. Biol. 142, 270−282.

Webb, B.L., Proud, C.G., 1997. Eukaryotic initiation factor 2B (eIF2B). Int. J. Biochem. Cell Biol. 29,1127−1131.

Wester, T.J., Lobley, G.E., Birnie, L.M., Lomax, M.A., 2000. Insulin stimulates phenylalanine uptakeacross the hind limb in fed lambs. J. Nutr. 130, 608−611.

White, M.F., Kahn, C.R., 1994. The insulin signaling system. J. Biol. Chem. 269, 1−4.White, P., Cattaneo, D., Dauncey, M.J., 2000. Postnatal regulation of myosin heavy chain isoform expres-

sion and metabolic enzyme activity by nutrition. Brit. J. Nutr. 84, 185−194.Widdowson, E.M., Crabb, D.E., 1976. Changes in the organs of pigs in response to feeding for the first

24h after birth. I. The internal organs and muscles. Biol. Neonate 28, 261−271.Widdowson, E.M., Colombo, V.E., Artavanis, C.A., 1976. Changes in the organs of pigs in response to

feeding for the first 24h after birth. II. The digestive tract. Biol. Neonate 28, 272−281.Wray-Cahen, D., Beckett, P.R., Nguyen, H.V., Davis, T.A., 1997. Insulin-stimulated amino acid utiliza-

tion during glucose and amino acid clamps decreases with development. Amer. J. Physiol. 273,E305−E314.

Wray-Cahen, D., Nguyen, H.V., Burrin, D.G., Beckett, P.R., Fiorotto, M.L., Reeds, P.J., Wester, T.J.,Davis, T.A., 1998. Response of skeletal muscle protein synthesis to insulin in suckling pigs decreaseswith development. Amer. J. Physiol. 275, E602−E609.

Yates, L.D., Greaser, M.L., 1983. Quantitative determination of myosin and actin in rabbit skeletalmuscle. J. Mol. Biol. 168, 123−141.

Yenush, L., White, M.F., 1997. The IRS-signalling system during insulin and cytokine action. Bioessays19, 491−500.


Yoshizawa, F., Kimball, S.R., Jefferson, L.S., 1997. Modulation of translation initiation in rat skeletalmuscle and liver in response to food intake. Biochem. Biophys. Res. Commun. 240, 825−831.

Young, V.R., 1970. The role of skeletal and cardiac muscle in the regulation of protein metabolism. In:Munro, H.M. (Ed.), Mammalian Protein Metabolism. Academic Press, New York, pp. 585−674.

Zahradka, P., Larson, D.E., Sells, B.H., 1991. Regulation of ribosome biogenesis in differentiated ratmyotubes. Mol. Cell Biochem. 104, 189−194.


69

3 Whole animal and tissue proteolysisin growing animals

V. E. Baracos

Department of Oncology, University of Alberta, 11560 University Avenue,Edmonton, Alberta, Canada T6G 1Z2

While it is convenient to conceptualize protein synthesis as being associated with growth and pro-tein degradation as being associated only with atrophy or senescence, both processes proceedcontinuously in all tissues at all stages of life. Since the highest rates of protein degradation occurduring the most rapid growth, protein deposition is inefficient. Our current understanding of pro-teolytic processes comes in large part from studies of skeletal muscle, including methods andapproaches for its determination, contributing proteases and regulators, physiological controls,and post-mortem proteolytic events contributing to meat quality. Because of the contribution ofprotein catabolism to deposition of marketable muscle tissue, to its energetic cost, and to productquality, there is interest in different strategies increasing or modulating the rate of animal growth,especially the relative rates of protein synthesis and catabolism in skeletal muscle.

1. PROTEIN DEGRADATION: A KEY DETERMINANT OF GROWTH,METABOLIC RATE, AND GROWTH EFFICIENCY

This review covers a variety of topics pertaining to protein degradation and regulation of growthin animals. Numerous excellent reviews are cited, and I have not attempted to cover in detaildomains for which recent synthesis articles are available. Not all of the work relevant to thetopic of growth and protein degradation has been conducted in domestic animal species. One ofthe reasons for this is that there are many difficulties associated with measures of protein degra-dation and these approaches are generally more difficult to implement in domestic animalsbecause of their cost, invasiveness, and need for the use of stable or radioactive isotopes andrelated analytical equipment not routinely available in Animal Science Departments, such asisotope ratio mass spectrometry (Patterson et al., 1997). The reader is thus strongly encouragedto be open-minded to the broader scope of the protein degradation literature, in all species.

A fundamental concept in metabolism is that all proteins are in a continual state of turnover.While protein synthesis is associated with growth and protein degradation is a dominant fea-ture of atrophy or senescence, both processes proceed continuously. In fact the highest rates



of protein degradation occur during the most rapid growth and protein deposition. This meansthat a large fraction of proteins synthesized in any period of growth are broken back down.The amount of protein formed and broken down again over any period of time is dependenton developmental stage, species, and organ. For example, protein breakdown rates in indi-vidual tissues of animals and poultry undergoing high rates of growth vary from <10%/dayto >80%/day. This means that in a matter of 10 days, an animal at this stage will have brokendown and resynthesized a slowly turning over tissue once and a rapidly turning over tissueeight times! The vast majority of animal protein formed, thus never goes to market.

The results of Lapierre et al. (1999) further illustrate the impact of protein degradationacross different growth rates. These authors evaluated whole-body protein metabolism in rela-tion to intake (0.6, 1.0, and 1.6 × maintenance requirements) in growing beef steers. Proteinretention in the whole body increased with intake, as a result of a greater increase in proteinsynthesis compared with protein degradation. Protein breakdown had a major impact, as 65%of the protein synthesized was degraded when intake varied from 1.0 to 1.6 times maintenance.

Protein degradation is sometimes determined on a tissue- or organ-specific basis (i.e. Bioloet al., 1994; Samuels and Baracos, 1996; Zhang et al., 1996a,b; Lapierre et al., 1999), althoughrelatively few tissue-specific determinations have been done in domestic species. Where this hasbeen looked at, there is considerable emphasis on the skeletal muscles, since this organ is themain product of meat animal agriculture. Splanchnic tissues drained by the portal and hepaticveins are organ sites amenable to determination of protein degradation by tracer techniques (i.e.Lapierre et al., 1999). These organ systems are also a considerable focus, since because of theirhigh turnover rates they constitute a major fraction of whole-body catabolism. In spite of its rel-atively small size, the liver in a growing monogastric can comprise 25% of whole-body proteincatabolism. In growing cattle, the total splanchnic tissues inclusive of the liver accounted for 44%of whole-body turnover (25% from the portal-drained viscera and 19% from the liver) (Lapierreet al., 1999). There is increased protein synthesis in gut epithelium of cattle in response to feed-ing (Kelly et al., 1995), and this has implications for energy expenditure.

When the degradation of protein is considered, the production of animal protein seemsstartlingly inefficient. This apparently wasteful metabolism, however, performs essentialfunctions. Protein turnover is metabolically costly but is thought to convey flexibility inprotein and amino acid metabolism. Protein breakdown is a particular feature of remodelling,such as during involution or metamorphosis when entire structures or organs are removedor replaced. Protein breakdown provides a means to be able to make a rapid change in themetabolic mass of any protein or group of proteins. Proteins not needed, non-functional, ordamaged may be rapidly removed by activating their catabolism. Proteins may be rapidlyinduced, by activation of their synthesis and simultaneous suppression of their catabolism.Amino acids can be mobilized by degradation of existing proteins, and used for the synthesisof other proteins or other purposes such as gluconeogenesis. This is an essential function,which is capable of providing a continuous source of essential and non-essential amino acids,in a fashion independent of dietary intake.

Continuous turnover of proteins incurs a considerable cost in ATP (Mitch and Goldberg, 1996).This cost in ATP may be largely associated with the energetic costs of protein synthesis; however,it is now recognized that at least some elements of the process of protein degradation also requireATP (Mitch and Goldberg, 1996). Energy economy could be improved by reducing flow throughcyclical metabolic pathways that use ATP, such as protein turnover (Gill et al., 1989).

Because of the contribution of protein catabolism to deposition of marketable muscle tissue andto its energetic cost, there is interest in different strategies increasing or modulating the rate ofanimal growth, especially the relative rates of protein synthesis and catabolism in skeletal muscle.

V. E. Baracos70

2. MECHANISMS OF DEGRADATION: PROTEOLYTIC SYSTEMS

Intracellular proteolytic systems are extensively characterized in a wide variety of cells, tissues,and species, including domestic livestock. The intracellular proteolytic systems of muscle thatdegrade the myofibrillar proteins are now well characterized, and are the main focus of thissection. Three intracellular proteolytic systems are known to have the potential to contributeto myofibrillar protein degradation: the lysosomal system, the calcium-dependent proteolyticsystem (Tan et al., 1988), and the ATP-ubiquitin–proteasome-dependent proteolytic system(Attaix et al., 1998). There is a low level of lysosomal proteolytic activity in skeletal muscleand its overall contribution to catabolism in this tissue is small, except in the case of tissueinjury (Farges et al., 2002).

Cytosolic Ca2+-activated proteases (calpains) and their inhibitors have been extensivelystudied in muscle of domestic animals, because of their putative relationship with post-mortem proteolysis and hence of meat quality (see below). The calpains constitute a largefamily comprising ubiquitous, tissue-specific, and atypical calpains (reviewed by Sorimachiet al., 1997; Kinbara et al., 1998). The calpains are cysteine proteases with a Ca2+ requirementfor activation, in either the millimolar or micromolar concentration range. There are alsoatypical calpains, such as p94 (also called calpain 3), a mammalian calpain homologue pre-dominantly expressed in skeletal muscle, which has been shown to be responsible for a formof limb-girdle muscular dystrophy. Calpastatin is a specific inhibitor of the calpains and theisolation of this protein from animal species such as cattle, cloning of its complementaryDNA, and nucleotide sequencing have been completed (Killefer and Koomaraie, 1994). Thecontribution of this system to overall muscle protein breakdown in vivo is difficult to esti-mate; however, in incubated muscles inhibition of this system decreases protein degradationby less than 10% in most studies. Although some authors suggest that calpains are rate-limitingfor release of filaments from the myofibrillar superstructure, if this were rate-limiting formyofibrillar proteins to be degraded, inhibition of this system would be expected to block allmyofibrillar proteolysis and this is clearly not so (Attaix et al., 1998, 2001).

Muscle protein catabolism appears primarily mediated by the ATP-dependent ubiquitin–proteasome system, which is responsible for degrading the bulk of intracellular proteinsincluding myofibrillar proteins. This conclusion is based on the use of specific proteasomeinhibitors, which are able to block upwards of 60% of total myofibrillar protein catabolism(Attaix et al., 1998, 2001, 2002). This has been well established in a wide range of animalmodels (reviewed by Mitch and Goldberg, 1996; Attaix et al., 2002). The ubiquitination/deubiquitination system is a complex machine responsible for the specific tagging and proof-reading of substrates degraded by the proteasome. Polyubiquitination of substrates targetsthem for degradation by the proteasome, a multiprotein complex conserved from archaebacteriato humans. Ubiquitin is an evolutionarily highly conserved 76-amino-acid polypeptide that isabundant in all eukaryotic cells. The initial step in the ubiquitin pathway is ATP-dependentand involves the linkage of ubiquitin to a ubiquitin-activating enzyme, or E1, in a high-energythioester bond. Ubiquitin is then transferred in a second thioester linkage to a ubiquitin-conjugating enzyme, which in turn catalyses the transfer of ubiquitin to the substrate proteinin a covalent bond. In some cases, substrate polyubiquitination requires another enzyme, theubiquitin ligase (Bodine et al., 2001; Gomes et al., 2001). The ubiquitin ligase can participatein the hierarchic transfer of ubiquitin into the substrate, or can function as an adaptor to facil-itate positioning and transfer of ubiquitin from the ubiquitin-conjugating enzyme directlyonto the substrate. The substrates tagged by ubiquitin are then recognized by the proteasomeand degraded into peptides. How this proteolytic pathway degrades muscle proteins, and

Whole animal and tissue proteolysis 71

more particularly contractile proteins, remains largely unknown. Information on the ubiquitin-conjugating enzymes and ubiquitin ligases that operate in muscle is still scarce. Similarly,neither the signals that target myofibrillar proteins for breakdown nor the precise substratesof the pathway have been identified. Finally, the possible relationships between the ubiquitinproteasome pathway and the lysosomal cathepsins and calpains are not well understood.

The matrix metalloproteinases (MMP) represent a family of enzymes responsible for con-nective tissue catabolism. Extensive studies in a variety of tissues suggest that the regulationof MMP activities is complex. MMP are secreted in a latent form as zymogens and activatedsequentially in a cascade initiated by other proteases including plasmin or membrane-typeMMP (MT-MMP). A third level of regulation involves local production of polypeptide tissueinhibitors of metalloproteinases. The matrix metalloproteinase system involved in intramus-cular connective tissue degradation has effectively just been described (Balcerzak et al.,2001). The genetic and physiological modulation of this system is barely characterized.

3. DETERMINATION OF PROTEIN CATABOLISM: WHOLE BODY,TISSUES, INDIVIDUAL PROTEINS

Measurement of protein catabolism is technically and conceptually difficult. It is not theintent of this chapter to cover all of the methodological considerations; however, it is impor-tant to understand thoroughly the inherent limitations of the method used, in the interpretationof any given set of results. The clearest picture is based on multiple independent approachesgiving the same overall conclusion. A physiologically relevant alteration in rates of proteincatabolism may be small, and a major problem is the size of these changes relative to the errorterm of the measurement.

3.1. Degradation by difference: protein synthesis + net protein accretion (loss)

It should be noted at the outset that if the experimental system can be shown to be in a steadystate (i.e. no net protein accretion or loss), rates of protein synthesis and degradation are bydefinition identical. In this case protein synthesis is a useful surrogate for measures of proteindegradation. Under non-steady-state conditions, protein degradation may be estimated as thedifference between protein synthesis and net protein gain or loss (i.e. Samuels and Baracos,1995; Wheeler et al., 2000). Protein gain or loss over time is determined by serial slaughterof groups of animals on the experimental treatments and protein synthesis is determined ineach group immediately before animals are killed, often using the “flooding dose” approach.Using this method, the degradative rates of many organs and tissues can be estimated. Thismethod is applicable to small, inexpensive animals such as chicks or lambs and has beenextensively used in laboratory rodents. Calculated degradation rates include the summederrors inherent in the directly measured variables, and may be considerable.

3.1.1. Urinary excretion of 3-methylhistidine

Post-translationally modified amino acids released on protein catabolism are not re-incorporatedinto proteins and provide an index of catabolism of the proteins of which they are character-istic. Measurement of urinary 3-methylhistidine (3-MH) excretion is used to estimatemyofibrillar protein breakdown. Similarly, urinary OH− proline reflects the appearance of this amino acid from the catabolism of connective tissue proteins, mainly collagen (Funabaet al., 1996). This approach requires quantitative collection of urine and is based on the

V. E. Baracos72

assumption that no metabolism of 3-MH occurs once it is released from actin and myosin.This is true in most species, but in sheep and swine a proportion is retained in muscle as adipeptide, balenine. In neither of these species does urinary 3-MH yield any data on proteinbreakdown. Rathmacher and Nissen (1998) proposed a compartmental model of 3-MH that isapplicable in domestic animals and does not involve the collection of urine. In this approach,3-MH metabolism in cattle, swine, and sheep was defined from a single bolus infusion of astable isotope, 3-[methyl-2H3]-methylhistidine. Following the bolus dose of the stable isotopetracer, serial blood and urine samples are collected. At least three exponentials were requiredto describe the plasma decay curve adequately. A simple three-compartment model describedthe plasma kinetics of 3-[methyl-2H3]-MH/3-MH for cattle with one urinary exit from theplasma compartment. The de novo production of 3-MH as calculated by the compartmentalmodel in cattle was not different when compared to total urinary 3-MH production. A plasma-urinary kinetic three-compartment model with two exits was used for sheep with a urinaryexit out of the plasma compartment and a balenine exit out of a tissue compartment. A plasmathree-compartment model was used in swine with an exit out of a tissue compartment. Thekinetic parameters reflect the differences in known physiology of 3-MH metabolism of therespective species. Steady-state model calculations define masses and fluxes of 3-MHbetween three compartments and, importantly, the de novo production of 3-MH.

3.2. Isotopic tracer approaches for whole-body and tissue catabolism in vivo

A primed, constant infusion of an isotopically labelled amino acid such as leucine or phe-nylalanine may be used to estimate whole protein degradation (i.e. Lapierre et al., 1999; Vannet al., 2000). Splanchnic tissues drained by the portal and hepatic veins (i.e. Lapierre et al.,1999) and the hindlimb drained by the femoral vein (i.e. Savary et al., 2001) are organ sitesamenable to determination of protein degradation by techniques based on arterio-venousdifferences combined with radioactive or stable isotope tracers.

Wolfe and co-workers have produced a steady stream of methodological advances in thisarea (Biolo et al., 1994; Ferrando et al., 1995; Zhang et al., 1996a,b; Patterson et al., 1997).Zhang et al. (1996a) developed an attractive method to measure the fractional breakdown rateof muscle protein. This method involves infusing labelled amino acid to reach an isotopicequilibrium and then observing its decay in the arterial blood and muscle intracellular pool.The calculation of fractional breakdown rate is based on the rate at which tracer released frombreakdown dilutes the intracellular enrichment using a modified precursor-product equation.The measured fractional breakdown rates were in agreement with the results from the arterio-venous balance method. This provides a feasible approach for measurement of muscle proteincatabolism. This method can be combined with the tracer incorporation method to measureboth breakdown and synthesis in the same infusion study.

One limitation of tracer/arterio-venous balance approaches is that the degradative rates of theindividual organs and tissues within the organ system(s) within the studied vascular bed cannot bedescriminated. While limb protein metabolism is often interpreted as equivalent to skeletal muscleprotein metabolism, skin protein synthesis and degradation accounted for approximately 10−15%of the total leg protein kinetics in different species (Baracos et al., 1991; Biolo et al., 1994).

3.3. Protease gene expression

Protein degradation is clearly regulated, at least to some extent, at the level of gene expres-sion, and in a wide variety of physiological and pathological states expression of various


elements of proteolytic systems varies with measured overall degradative rates. Regulationof protease gene expression in muscle has been the subject of several elegant studies andcomprehensive review articles by Attaix and co-workers (1997, 1998, 2001, 2002; Larbaudet al., 2001).

While assessment of gene expression is not a primary measure of degradation, thisapproach has applications where it is presently not possible to make direct determinations. Asmall amount of data on local protease gene expression is emerging in tissues of the gas-trointestinal tract, which are suggestive of regulation of proteolysis at this level. For example,Samuels et al. (1996) measured mRNA levels for components of the lysosomal (cathepsins Band D), Ca2+-activated (m-calpain), and ubiquitin-dependent (ubiquitin, 14 kDa ubiquitin-conjugating enzyme E2, and C8 and C9 proteasome subunits) proteolytic pathways, in thesmall intestine of rats during food deprivation. mRNA levels for most of these componentsincreased during fasting, suggesting that a co-ordinated activation of multiple proteolyticsystems contributed to intestinal protein wasting. Adegoke et al. (1999) tested the effects of aluminal infusion of an amino acid mixture on protease mRNA in jejeunal mucosa of pigletsafter overnight food deprivation. Amino acids acutely suppressed mucosal levels of mRNAencoding ubiquitin, 14 kDa ubiquitin-conjugating enzyme, and the C9 subunit of the prote-asome by 20–30%, demonstrating the sensitivity of components of the ATP-ubiquitinproteolytic pathway to acute regulation by nutrients.

3.3.1. In vitro techniques

Aside from poultry (i.e. Baracos et al., 1989), the in vitro incubation techniques developed byA.L. Goldberg and used widely in muscles of laboratory rodents are not applicable in domes-tic animal species. A major attribute of this system has been the ability to study thedifferential regulation of the lysosomal, Ca2+-dependent and ubiquitin/proteasome-dependentproteolytic pathways using inhibitors. (i.e. Larbaud et al., 2001).

4. PROTEIN DEGRADATION IN RELATION TO GENETIC MAKE-UP

Protein turnover rates are subject to genetic variation, and differences in protein turnover mayexplain part of the inherent differences in efficiency and growth of different animal breeds(Reeds et al., 1998). For example, Wheeler et al. (2000) evaluated the effect of the callipygephenotype in lambs on protein kinetics. These authors studied callipyge and normal lambs at5, 8, and 11 weeks of age. The synthesis rates of proteins in various tissues were measuredusing a primed, continuous infusion of [2H5]phenylalanine. Rates of protein degradation wereestimated by difference between protein synthesis and net protein accretion. Enhancedmuscle growth seems to be maintained in callipyge lambs by reduced protein degradation.Consistent with this observation, Koohmaraie et al. (1995) reported that the activity ofcalpastatin is about 80% higher in the callipyge phenotype.

Unpublished work from our group suggests that different breeds of cattle may exhibit char-acteristic differences in protein degradation. When animals were fed identical amounts ofmetabolizable energy and protein/kg BW·75, Brahman × Angus cross cattle showed a lowerlevel of urinary 3-MH excretion (1.82 mg/d/kg BW·75) than Charolais (3.06 mg/d/kg BW·75)(SE = 0.25; P < 0.009). Lower protein degradation, together with a tendency towards lowermetabolic rate, could be responsible for increased protein deposition and consequentlya higher growth rate observed in the Brahman × Angus cattle.

V. E. Baracos74

5. REGULATION OF DEGRADATION: ENDOCRINEAND AUTOCRINE CONTROLS

Rates of protein degradation are precisely regulated and there are multiple sites of hormonaland metabolic controls. In general, circulating hormones have an anabolic or catabolic effecton a tissue, affecting rates of protein synthesis, degradation, or both. Tissues may be in a steadystate, enter into a catabolic state (wasting) or grow, in response to a concert of hormones andfactors in a given physiological or pathological state. Note that these factors are diverse andmay be hormones, growth factors, substrates, and metabolites. Tissue-specific factors, suchas contractile activity and stretch, also greatly influence muscle protein turnover. The list offactors affecting the process of protein catabolism is most fully understood for skeletalmuscle (table 1). This symphony of signals act collectively to inform muscle protein catabo-lism in three information subsets:1. Contractile activity. Two components of muscular activity, active contraction and passive

stretch, are perceived as anabolic signals. Lack of activity and shortening of muscleresult in activation of the degradation of contractile proteins. This regulation allows forthe maintenance of a muscle mass appropriate to the level of work.

2. Nutritional status/glycemia. Since muscle protein comprises the principal gluconeogenicprecursor, the need for muscle protein mobilization is conveyed through the factors


Table 1

Regulatory factors in muscle protein synthesis and degradation

Synthesis Degradation Overall promotes

Factors related to level of contractile workContractile activity ↑ ↓ Protein depositionStretch ↑ → Protein depositionDisuse (inactivity) ↓ ↑ AtrophyGonadal steroids ↑ ↓ Protein deposition

Factors related to nutritional status/glycemiaInsulin ↑ ↓ Protein depositionInsulin-like growth factor I ↑ ↓ Protein depositionGrowth hormone ↑ → Protein depositionGlucose → ↓ Protein depositionKetone bodies → ↓ Protein depositionGlutamine ↑ ↓ Protein depositionBranched-chain amino acids ↑ ↓ Protein depositionGlucagon ↓ → AtrophyGlucocorticoids → ↑ Atrophyβ-Adrenergic agonists ↑ ↓ Protein depositionThyroid hormones (normal) ↑ ↑ Protein depositionThyroid (excess) ↑ ↑↑ Atrophy

Factors related to the presence of injury/inflammationProstaglandin E2 → ↑ AtrophyProstaglandin F2α ↑ → Protein depositionInterleukin-1β ↓ ↑ AtrophyInterleukin-6 ↓ ↑ AtrophyTumor necrosis factor α ↓ ↑ AtrophyInterferon γ ↓ ↑ Atrophy

regulating glycemia and gluconeogenesis. Since muscle protein comprises a reserveof amino acids and protein in case of food deprivation, muscle catabolism is sensitiveto indices of food intake, which indicate a state of plenty where growth may occur, ora state of fasting or nutritional deprivation when net catabolism is required.

3. Stress, infection, and injury. Disease or injury impose additional metabolic demandswhen food intake may also be low or zero. Factors arising in the context of thestress response and immune and inflammatory responses, such as cytokinesand prostaglandins, communicate to muscle the need for extra protein catabolism(Castaneda, 2002).All of the above factors are simultaneously at play. When muscle is catabolized, several

metabolic changes (reduced food intake, impaired mobility, and perturbations in the productionor responsiveness of catabolic and anabolic hormones, cytokines, and/or proteolysis-inducingfactors) act in concert. In the context of this complex regulatory system, it is often difficult toidentify the primary and secondary factors, and a complete understanding of their interactionsremains to be developed.

6. PROTEIN DEGRADATION, NUTRITIONAL STATUS, AND GROWTH

6.1. Feeding and diet

The rapid loss of skeletal muscle protein during acute starvation occurs primarily throughincreased rates of protein breakdown and activation of the ubiquitin–proteasome-dependentproteolytic process (Wing et al., 1995). The levels of ubiquitin-conjugated proteins increased50−250% after food deprivation in various muscles. Like rates of proteolysis, the amount ofubiquitin–protein conjugates and the fraction of ubiquitin conjugated to proteins increasedprogressively during food deprivation and returned to normal within 1 day of refeeding.Larbaud et al. (1996) showed that euglycemic hyperinsulinemia and hyperaminoacidemiadecrease skeletal muscle ubiquitin mRNA in goats, suggesting insulin and amino acids aspossible mediators of this effect.

Restricted feeding during growth constitutes a stress to energy metabolism, and energyutilization is curtailed in this circumstance by reduction in growth. In sheep and cattle, proteinturnover is positively related to plane of nutrition (Lobley et al., 1992; Reecy et al., 1996).

The turnover of actomyosin, the major myofibril constituent, is modulated in animalstested on various planes of nutrition. Lobley et al. (2000) demonstrated that both fractionaldegradation rate and fractional synthesis rate were lower in muscles of steers with a lowgrowth rate (1 kg/d) on restricted feeding in comparison with high growth rate (1.4 kg/d). Thisand other studies (i.e. Boisclair et al., 1993) emphasized myofibrillar turnover and muscle;however, the connective tissue protein catabolism is similarly affected. Lambs with a lowgrowth rate present less active matrix metalloproteinase-2, suggesting a decrease in collagencatabolism (Sylvestre et al., 2002). Other studies suggested lower collagen turnover and reduceddeposition of neo-synthesized, immature, and non-cross-linked collagen related with a lowgrowth rate (Aberle et al., 1981; Crouse et al., 1985; Miller et al., 1987; McCormick, 1994).

The quality of dietary protein has an impact on protein degradation. Branched-chain aminoacids appear to have a specific regulatory effect on protein degradation and decrease the rateof this process (Ferrando et al., 1995). Diets with protein of inferior quality may increase pro-tein breakdown in skeletal muscle (Lohrke et al., 2001). These authors studied the activationof skeletal muscle protein breakdown in pigs fed isoenergetic and isonitrogenous diets basedon soy protein isolate compared with casein.

V. E. Baracos76

6.2. Role of the somatotrophic axis

The somatotrophic axis plays a key role in the co-ordination of protein metabolism duringpostnatal growth (reviewed by Breier, 1999). Acute growth hormone treatment improves thepartitioning of nutrients by increasing protein synthesis and decreasing protein degradation.Short-term infusion of IGF-1 also reduces whole-body protein breakdown and increasesprotein synthesis, and Boyle et al. (1998) have also shown this to be true in fetal sheep duringlate gestation. More recently, Vann et al. (2000) suggest that growth hormone increasesprotein balance by lowering body protein degradation in fed, growing pigs. Pair-fed, weight-matched growing swine were treated with porcine growth hormone (150 μg/kg/d) or vehiclefor a week. Growth hormone treatment increased the efficiency with which the diet was usedfor growth, but did not alter protein synthesis in skeletal muscles, liver, or jejunum. In theabsence of any changes in protein synthesis at these sites, the results suggest that in the fedstate, growth hormone treatment of growing swine increases protein deposition primarilythrough a suppression of protein degradation.

6.3. Stress

Domestic animals are exposed to behavioural, environmental, and infectious stressors. One ofthe metabolic hallmarks of these stresses is the catabolic response in skeletal muscle, mainlyreflecting increased protein breakdown, in particular myofibrillar protein breakdown(reviewed by Hasselgren, 2002). Among different intracellular proteolytic pathways, theenergy-ubiquitin-dependent pathway is particularly important for the regulation of muscleprotein breakdown during acute infection. The gene expression of ubiquitin-conjugatingenzyme E214k, ubiquitin ligase E3α, and several components of the proteasome is up-regulatedand the activity of the proteasome is increased in muscle during infection. An increasedunderstanding of the molecular regulation of muscle wasting in stress may help in the futureto mitigate or prevent catabolic losses in animal production.

7. PROMOTION OF THE EFFICIENCY AND RATE OF GROWTHBY MANIPULATION OF PROTEIN DEGRADATION

It is clearly important that proteins must be broken down; however, the physiological rangeof protein catabolism rates is quite broad. It is of particular interest that some breeds of ani-mals and physiological situations are associated with low breakdown rates. The breeds ofanimals with lower rates of protein degradation appear to be those breeds adapted to harshenvironments and a low availability or quality of forage. Animal selection and breeding basedupon efficiency of protein deposition may potentially be used to develop this trait.

It has been known for about a decade that β-adrenergic agonists act in part throughsuppression of protein degradation (Bardsley et al., 1992; Parr et al., 1992; Mills, 2002).β-Adrenergic agonists are structurally similar to the catecholamines epinephrine and norepi-nephrine and bind with high affinity to β-adrenergic receptors in adipose and muscle tissue.This class of compound includes agents such as clenbuterol, cimaterol, and ractopamine.Ractopamine was the first β-adrenergic receptor ligand to be cleared for use in pigs in theUSA, about 4 years ago. Ractopamine consistently increases muscle protein accretion in pigsand while the mechanism responsible for increased protein accretion is not clear, cumulativeevidence points to a direct effect, possibly on both protein synthesis and degradation. Onereason why the role of protein catabolism is not entirely clear is that while β-agonists cause


large differences in protein accretion, these are manifest over long periods of time, and themarginal difference in degradation rates required to result in these changes would be quitesmall. β-Agonist treatment does cause large increases in protease inhibitor activity and geneexpression, and these changes are suggestive. For example, cimaterol treatment of Friesiansteers (Parr et al., 1992) caused significant increases in muscle mass (+37%) and calpastatinspecific activity (+76%). Total RNA was unchanged, but there was a 96% overall increase incalpastatin mRNA in muscle from treated animals.

If protein breakdown could be minimized, this would be an attractive way of promotingprotein deposition as well as a means of lowering the metabolic cost of maintaining any givenprotein mass. Thus protein degradation would be an attractive target for growth promotants,which should properly be called “anticatabolic” rather than “anabolic” factors. As we obtainmore details regarding the proteolytic processes and their regulation, the possibility of iden-tifying molecular targets for anticatabolic agents seems tangible. A likely site of such targetswould be the ubiquitin–proteasome system and in particular the recently identified class ofmuscle-specific ubiquitin ligases (Bodine et al., 2001). Two unique ubiquitin ligases, MuRF1,a RING finger protein, and MAFbx (Bodine et al., 2001), also called Atrogin-1 (Gomes et al.,2001), of the SCF family, have been reported to play a role in muscle atrophy. Unlike otherknown ubiquitin ligases found in many tissues, these enzymes appear to be expressed mainlyin muscle cells, especially skeletal muscle. Multiple ubiquitin ligases may operate in skeletalmuscle, possibly to connect protein catabolism to different classes of external stimuli. This issuggested by the reported findings (Bodine et al., 2001) that null mutation of either MAFbxor MuRF1 in mice led to resistance to denervation-induced muscle atrophy.

The ubiquitin–proteasome system appears to be central in muscle protein degradation,regardless of the humoral signal for the system’s activation. The intracellular signal trans-duction from multiple factors converges upon a common proteolytic pathway, of whichubiquitin ligases are likely to be a critical element. Further studies are required to betterunderstand the importance of the ubiquitin ligase family, including identifying the physio-logical substrates for these enzymes in skeletal muscle, elucidating signalling events thatregulate their activity, and analysing the effects of specific inhibition through gene ablationand/or the design of selective small molecule inhibitors. Ubiquitin ligases may be attractivemolecular targets for manipulation of proteolysis since there are isoforms specific to muscle.These features may potentially allow for local suppression of muscle catabolism withoutaffecting the basal proteolytic processes in non-muscle tissues or associated with essentialfunctions.

8. PROTEIN DEGRADATION AND POST-MORTEM PROTEOLYSIS

Catabolic processes are modified, but not interrupted, at death. While the activity of ATP-dependent processes (and proteases) would cease, the vast majority of proteolytic enzymescan continue to express activity. This autolysis is prefaced by the pre-mortem level of proteo-lytic activities, but thereafter evolves in a manner dissimilar to in vivo events, because ofchanges in tissue temperature, pH, and the loss of structural integrity. The identity of all ofthe active enzymes and the substrates to which they have access is only partly understood(Ho et al., 1994).

It has long been believed that proteases play a key role in post-mortem tenderizationof meat (reviewed by Koohmaraie, 1992), and this concept is supported by the observationthat low levels of pre-mortem proteolysis are associated with reduced degradation duringmeat maturation. Tissue growth made to be more efficient by reducing proteolytic activity

V. E. Baracos78

(i.e. β-adrenergic agonist-induced muscle hypertrophy) manifested lowered rates of post-mortem proteolysis. Koohmaraie et al. (1991) showed that the pattern of post-mortemproteolysis was altered by β-adrenergic agonists. In β-agonist-treated lambs, post-mortemstorage was not associated with increased myofibril fragmentation index or degradation ofdesmin and troponin-T. These results indicate that the ability of the muscle to undergo post-mortem proteolysis has been dramatically reduced with β-adrenergic agonist feeding.Similarly, enhanced muscle growth seems to be maintained in callipyge lambs by reducedprotein degradation, and Koohmaraie et al. (1995) suggested a causal relationship betweenthis effect and increased shear force in meat of calipyge lambs.

The relationship between pre-mortem proteolysis, post-mortem proteolysis, and meat qual-ity is far from being fully explored. This would be a worthy target for future experimentation,since an ability to modulate the structural integrity of tissue elements that confer toughnessto meat would have considerable value.


A means of reducing protein degradation to its physiological minimum during growth holdsthe potential to increase the efficiency of animal production by a large factor. A means ofactivating protein degradation in the peri-mortem period holds the potential to increase thequality of meat. Both of these outcomes could have a large economic impact. Given thepotential impact for animal growth and production, it is perhaps surprising that the researchmomentum on proteolysis is not emanating from the agricultural research community.A pharmaceutical industry strongly motivated to produce therapies for inappropriate degra-dation, muscle atrophy, and wasting syndromes (i.e. Bodine et al., 2001) is presently makinglarge financial commitments in this area. Whatever the source, new developments in ourunderstanding of proteolysis are showing the way towards targets for intervention in theseprocesses, and animal agriculture may benefit greatly if it is poised to capture the relevantinformation.

REFERENCES

Aberle, E.D., Reeves, E.S., Judge, M.D., Hunsley, R.E., Perry, T.W., 1981. Palatability and musclecharacteristics of cattle with controlled weight gain: time on a high energy diet. J. Anim. Sci. 52,757–763.

Adegoke, O.A., McBurney, M.I., Samuels, S.E., Baracos, V.E., 1999. Luminal amino acids acutelydecrease intestinal mucosal protein synthesis and protease mRNA in piglets. J. Nutr. 129, 1871−1878.

Attaix, D., Aurousseau, E., Combaret, L., Kee, A., Larbaud, D., Rallière, C., Souweine, B., Taillandier, D.,Tilignac, T., 1998. Ubiquitin-proteasome-dependent proteolysis in skeletal muscle. Reprod. Nutr. Dev.38, 153−165.

Attaix, D., Combaret, L., Pouch, M.N., Taillandier, D., 2001. Regulation of proteolysis. (Review),(49 refs). Curr. Opin. Clin. Nutr. Metab. Care 4, 45−49.

Attaix, D., Combaret, L., Pouch, M.N., Taillandier, D., 2002. Cellular control of ubiquitin-proteasome-dependent proteolysis. J. Anim. Sci. 80, Suppl. 2, E56−E63.

Attaix, D., Taillandier, D., Combaret, L., Ralliere, C., Larbaud, D., Aurousseau, E., Tanaka, K., 1997.Expression of subunits of the 19S complex and of the PA28 activator in rat skeletal muscle (Review).Mol. Biol. Rep. 24, 95−98.

Balcerzak, D., Querengesser, L., Dixon, W.T., Baracos, V.E., 2001. Coordinate expression of matrix-degrading proteinases and their activators and inhibitors in bovine skeletal muscle. J. Anim. Sci.79, 94−107.

Baracos, V.E., Brun-Bellut, J., Marie, M., 1991. Tissue protein-synthesis in lactating and dry goats.Brit. J. Nutr. 66, 451−465.


Baracos, V.E., Langman, M., Mak, A., 1989. An in vitro preparation of the extensor digitorum communismuscle from the chick (Gallus domesticus) for studies of protein turnover. Comp. Biochem. Physiol.A Comp. Physiol. 92, 555–563.

Bardsley, R.G., Allcock, S.M., Dawson, J.M., Dumelow, N.W., Higgins, J.A., Lasslett, Y.V., Lockley, A.K.,Parr, T., Buttery, P.J., 1992. Effect of beta-agonists on expression of calpain and calpastatin activityin skeletal muscle. Biochimie 74, 267−273.

Biolo, G., Gastaldelli, A., Zhang, X.J., Wolfe, R.R., 1994. Protein synthesis and breakdown in skin andmuscle: a leg model of amino acid kinetics. Amer. J. Physiol. 267, E467−E474.

Bodine, S.C., Latres, E., Baumhueter, S., Lai, V.K., Nunez, L., Clarke, B.A., Poueymirou, W.T., Panaro, F.J.,Na, E., Dharmarajan, K., Pan, Z.Q., Valenzuela, D.M., DeChiara, T.M., Stitt, T.N., Yancopoulos, G.D.,Glass, D.J., 2001. Identification of ubiquitin ligases required for skeletal muscle atrophy. Science294, 1704−1708.

Boisclair, Y.R., Bell, A.W., Dunshea, F.R., Harkins, M., Bauman, D.E., 1993. Evaluation of the arteri-ovenous difference technique to simultaneously estimate protein synthesis and degradation in thehindlimb of fed and chronically underfed steers. J. Nutr. 123, 1076−1088.

Boyle, D.W., Denne, S.C., Moorehead, H., Lee, W.H., Bowsher, R.R., Liechty, E.A., 1998. Effect ofrhIGF-I infusion on whole fetal and fetal skeletal muscle protein metabolism in sheep.Amer. J. Physiol. 275, E1082−E1091.

Breier, B.H., 1999. Regulation of protein and energy metabolism by the somatotropic axis. (Review, 28 refs.)Domest. Anim. Endocrinol. 17, 209−218.

Castaneda, C., 2002. Muscle wasting and protein metabolism. J. Anim. Sci. 80, Suppl. 2,E50−E55.

Crouse, J.D., Cross, H.R., Seideman, S.C., 1985. Effects of sex condition, genotype, diet and carcasselectrical stimulation on the collagen content and palatability of two bovine muscles. J. Anim. Sci. 60,1228−1234.

Farges, M.C., Balcerzak, D., Fisher, B.D., Attaix, D., Bechet, D., Ferrara, M., Baracos, V.E., 2002.Increased muscle proteolysis after local trauma mainly reflects macrophage-associated lysosomalproteolysis. Amer. J. Physiol. 282, E326−E335.

Ferrando, A.A., Williams, B.D., Stuart, C.A., Lane, H.W., Wolfe, R.R., 1995. Oral branched-chain aminoacids decrease whole-body proteolysis. J. Parent. Enter. Nutr. 19, 47−54.

Funaba, M., Saito, S., Kagiyama, K., Iriki, T., Abe, M., 1996. Bone growth rather than myofibrillarprotein turnover is strongly affected by nutritional restriction at early weaning of calves. J. Nutr.126, 898−905.

Gill, M., France, J., Summers, M., McBride, B.W., Milligan, L.P., 1989. Simulation of the energycosts associated with protein turnover and Na+, K+-transport in growing lambs. J. Nutr. 119,1287−1299.

Gomes, M.D., Lecker, S.H., Jagoe, R.T., Navon, A., Goldberg, A.L., 2001. Atrogin-1, a muscle-specific F-box protein highly expressed during muscle atrophy. Proc. Natl. Acad. Sci. USA 98,14440−14445.

Hasselgren, P.-O., 2002. Stress and muscle wasting J. Anim. Sci. 80, Suppl. 2, E98−E105.Ho, C.Y., Stromer, M.H., Robson, R.M., 1994. Identification of the 30 kDa polypeptide in post mortem

skeletal muscle as a degradation product of troponin-T. Biochimie 76, 369−375.Kelly, J.M., Vaage, A.S., Milligan, L.P., McBride, B.W.T., 1995. In vitro ouabain-sensitive respiration and

protein synthesis in rumen epithelial papillae of Hereford steers fed either timothy hay or timothy haysupplemented with cracked corn once daily. J. Anim. Sci. 73, 3775−3784.

Killefer, J., Koohmaraie, M., 1994. Bovine skeletal muscle calpastatin: cloning, sequence analysis, andsteady-state mRNA expression. J. Anim. Sci. 72, 606−614.

Kinbara, K., Sorimachi, H., Ishiura, S., Suzuki, K., 1998. Skeletal muscle-specific calpain, p94: structureand physiological function (Review). Biochem. Pharmacol. 56, 415−420.

Koohmaraie, M., 1992. The role of Ca2+-dependent proteases (calpains) in post mortem proteolysis andmeat tenderness (Review). Biochimie 74, 239−245.

Koohmaraie, M., Shackelford, S.D., Muggli-Cockett, N.E., Stone, R.T., 1991. Effect of the beta-adrenergicagonist L644,969 on muscle growth, endogenous proteinase activities, and postmortem proteolysis inwether lambs. J. Anim. Sci. 69, 4823−4835.

Koohmaraie, M., Shackelford, S.D., Wheeler, T.L., Lonergan, S.M., Doumit, M.E., 1995. A musclehypertrophy condition in lamb (callipyge): characterization of effects on muscle growth and meatquality traits. J. Anim. Sci. 73, 3596−3607.

V. E. Baracos80

Lapierre, H., Bernier, J.F., Dubreuil, P., Reynolds, C.K., Farmer, C., Ouellet, D.R., Lobley, G.E., 1999.The effect of intake on protein metabolism across splanchnic tissues in growing beef steers. Brit. J. Nutr.81, 457−466.

Larbaud, D., Balage, M., Taillandier, D., Combaret, L., Grizard, J., Attaix, D., 2001. Differentialregulation of the lysosomal, Ca2+-dependent and ubiquitin/proteasome-dependent proteolyticpathways in fast-twitch and slow-twitch rat muscle following hyperinsulinaemia. Clin. Sci. 101,551−558.

Larbaud, D., Debras, E., Taillandier, D., Samuels, S.E., Temparis, S., Champredon, C., Grizard, J., Attaix, D.,1996. Euglycemic hyperinsulinemia and hyperaminoacidemia decrease skeletal muscle ubiquitinmRNA in goats. Amer. J. Physiol. 271, E505−E512.

Lobley, G.E., Harris, P.M., Skene, P.A., Brown, D., Milne, E., Calder, A.G., Anderson, S.E., Garlick, P.J.,Nevison, I., Connell, A., 1992. Response in tissue protein-synthesis to submaintenance and supra-maintenance intake in young growing sheep: comparison of large-dose and continuous-infusiontechniques. Brit. J. Nutr. 68, 373−388.

Lobley, G.E., Sinclair, K.D., Grant, C.M., Miller, L., Mantle, D., Calder, A.G., Warkup, C.C.,Maltin, C.A., 2000. The effects of breed and level of nutrition on whole-body and muscleprotein metabolism in pure-bred Aberdeen Angus and Charolais beef steers. Brit. J. Nutr. 84,275−284.

Lohrke, B., Saggau, E., Schadereit, R., Beyer, M., Bellmann, O., Kuhla, S., Hagemeister, H., 2001.Activation of skeletal muscle protein breakdown following consumption of soyabean protein in pigs.Brit. J. Nutr. 85, 447−457.

McCormick, R.J., 1994. The flexibility of the collagen compartment of muscle. Meat Sci. 36, 79−91.Miller, M.F., Cross, H.R., Crouse, J.D., Jenkins, T.G., 1987. Effect of feed energy intake on collagen

characteristics and muscle quality of mature cows. Meat Sci. 21, 287−294.Mills, S.E., 2002. Biological basis of the ractopamine response. J. Anim. Sci. 80, Suppl. 2, E28−E32.Mitch, W.E., Goldberg, A.L., 1996. Mechanisms of muscle wasting: the role of the ubiquitin-proteasome

pathway. N. Engl. J. Med. 335, 1897−1905.Parr, T., Bardsley, R.G., Gilmour, R.S., Buttery, P.J., 1992. Changes in calpain and calpastatin

mRNA induced by beta-adrenergic stimulation of bovine skeletal muscle. Eur. J. Biochem. 208,333−339.

Patterson, B.W., Zhang, X.J., Chen, Y., Klein, S., Wolfe, R.R., 1997. Measurement of very low stable isotopeenrichments by gas chromatography/mass spectrometry: application to measurement of muscle proteinsynthesis. Metab. Clin. Exp. 46, 943−948.

Rathmacher, J.A., Nissen, S.L., 1998. Development and application of a compartmental model of3-methylhistidine metabolism in humans and domestic animals. Adv. Exp. Med. Biol. 445, 303−324.

Reecy, J.M., Williams, J.E., Kerley, M.S., MacDonald, R.S., Thornton, W.H., Davis, J.L., 1996. Theeffect of postruminal amino acid flow on muscle cell proliferation and protein turnover. J. Anim. Sci.74, 2158−2169.

Reeds, P.J., Burrin, D.G., Davis, T.A., Stoll, B., 1998. Amino acid metabolism and the energetics ofgrowth. Arch. Anim. Nutr. 51, 187−197.

Samuels, S.E., Baracos, V.E., 1995. Tissue protein-turnover is altered during catch-up growth followingEscherichia coli infection in weaning rats. J. Nutr. 125, 520−530.

Samuels, S.E., Taillandier, D., Aurousseau, E., Cherel, Y., Le Maho, Y., Arnal, M., Attaix, D., 1996.Gastrointestinal tract protein synthesis and mRNA levels for proteolytic systems in adult fasted rats.Amer. J. Physiol. 271, E232−E238.

Savary, I.C., Hoskin, S.O., Dennison, N., Lobley, G.E., 2001. Lysine metabolism across the hindquartersof sheep: effect of intake on transfers from plasma and red blood cells. Brit. J. Nutr. 85, 565−573.

Sorimachi, H., Ishiura, S., Suzuki, K., 1997. Structure and physiological function of calpains (Review).Biochem. J. 328, 721−732.

Sylvestre, M.N., Balcerzak, D., Feidt, C., Baracos, V.E., Brun-Bellut, J., 2002. Elevated rate of collagensolubilization and post-mortem degradation in muscles of lambs with high growth rates: possiblerelationship with activity of matrix metalloproteinases. J. Anim. Sci. 80, 1871–1878.

Tan, F.C., Goll, D.E., Otsuka, Y., 1988. Some properties of the millimolar Ca2+-dependent proteinasefrom bovine cardiac muscle. J. Mol. Cell. Cardiol. 20, 983–997.

Vann, R.C., Nguyen, H.V., Reeds, P.J., Burrin, D.G., Fiorotto, M.L., Steele, N.C., Deaver, D.R., Davis, T.A.,2000. Somatotropin increases protein balance by lowering body protein degradation in fed, growingpigs. Amer. J. Physiol. 278, E477−E783.


Wheeler, T.L., Savell, J.W., Fiorotto, M.L., 2000. Protein kinetics in callipyge lambs. J. Anim. Sci.78, 78−87.

Wing, S.S., Haas, A.L., Goldberg, A.L., 1995. Increase in ubiquitin-protein conjugates concomitant withthe increase in proteolysis in rat skeletal muscle during starvation and atrophy denervation. Biochem. J.307, 639−645.

Zhang, X.J., Chinkes, D.L., Sakurai, Y., Wolfe, R.R., 1996a. An isotopic method for measurement ofmuscle protein fractional breakdown rate in vivo. Amer. J. Physiol. 270, E759−E767.

Zhang, X.J., Sakurai, Y., Wolfe, R.R., 1996b. An animal model for measurement of protein metabolismin the skin. Surgery 119, 326−332.

V. E. Baracos82

83

4 Cytokine regulation of protein accretionin growing animals

R. W. Johnson and J. Escobar

Department of Animal Sciences, University of Illinois, Urbana, IL 61801, USA

Inflammatory cytokines secreted by activated leukocytes are the critical molecules that enablethe immune system to influence disparate physiological systems that are important for deter-mining protein accretion in growing animals. The inflammatory cytokines, interleukin-1,interleukin-6, and tumor necrosis factor α, reduce feed intake, interfere with the somatotropicaxis, reduce skeletal muscle protein synthesis, and enhance skeletal muscle protein degradation.

The purpose of this chapter is to discuss how the immune system regulates feed intakeand arbitrates the balance between skeletal muscle protein synthesis and degradation, so as toprovide a biological explanation for why sick animals do not grow well.

1. INTRODUCTION

Skeletal muscle protein accretion is the net result of both protein synthesis and degradation.Both events occur constantly in normal skeletal muscle, but the mechanisms regulatingprotein synthesis and degradation are distinct and therefore can be influenced independently.It is now evident that the mechanisms that control skeletal muscle protein synthesis anddegradation are subject to regulation by the immune system (Johnson, 1997). Infectiouspathogens stimulate the immune system, and the immune system in turn actively suppressesfeed intake and skeletal muscle protein accretion. The general notion is that nutrients thatwere allocated to support skeletal muscle protein accretion are reassigned to metabolicprocesses that support the immune system, which at the time is a higher biological priority(Klasing, 1988). This places the immune system at the interface of environmental pathogensand animal growth (Broussard et al., 2001).

Inflammatory cytokines secreted by activated leukocytes are the critical molecules thatenable the immune system to regulate feed intake and nutrient allocation. Because inflam-matory cytokines reduce voluntary feed intake, and thus the nutrients available to supportprotein accretion, this issue will be briefly discussed. Feed intake alone, however, cannotaccount for the decreased protein accretion witnessed in sick animals because inflammatorycytokines also affect protein metabolism by several tissues, including skeletal muscle.



Therefore, the effects of inflammatory cytokines on skeletal muscle protein synthesis will bediscussed. Some attention will be given to the somatotropic axis because inflammatorycytokines regulate animals’ capacity to accrete skeletal muscle protein in part by reducing theamount of growth hormone (GH) and insulin-like growth factor-I (IGF-I) available to skeletalmuscle and by reducing the sensitivity of receptors for GH-releasing hormone (GHRH), GH,and IGF-I. Because the collective actions of cytokines lead to inhibition of mRNA translationinitiation – an obvious prerequisite for skeletal muscle protein accretion – this issue will bebriefly covered. Finally, the inflammatory cytokines that inhibit protein synthesis concomi-tantly enhance skeletal muscle protein degradation. Therefore, the effects of infection andinflammatory cytokines on the ATP-ubiquitin-dependent, calcium-dependent (calpains), andlysosomal (cathepsins) proteolytic pathways will be discussed as well. The purpose of thischapter is to discuss how the immune system regulates feed intake and arbitrates the balancebetween skeletal muscle protein synthesis and degradation, so as to provide a biologicalexplanation for why sick animals do not grow well.

2. CYTOKINES ORCHESTRATE ANIMALS’ RESPONSESTO INFECTIONS

Agricultural animals live surrounded by pathogens and routinely become infected, butbecause of a well-developed defense system only occasionally do they show clinical signs ofillness. Still, they are constantly challenged by pathogens and must contend with subclinicalinfections on a daily basis. This never-ending mêlée between the animal’s immune systemand pathogens is costly because there is a negative relationship between animal productivityand the pathogenic environment: as pathogens in the environment increase, animal productivitydecreases (fig. 1). The animal’s immune system “senses” (Blalock, 1984) the pathogenic environ-ment and biological functions, including feed intake and growth, are adjusted accordingly. Toappreciate how the pathogenic environment impinges upon skeletal muscle protein accretion,a superficial understanding of the animal’s primary and secondary defenses is obligatory.

The body surfaces are made up of epithelial cells that provide a physical barrier betweenthe internal milieu and the external pathogen-containing environment. Epithelial cells form theouter layer of skin and line gastrointestinal, respiratory, and genitourinary tracts. For infectionto occur, primary pathogens must penetrate one of these barriers. Surface epithelia providemechanical, chemical, and microbiological protection against infectious pathogens (table 1).

The immune system provides a secondary defense that deals with organisms once theyhave entered the body proper. The innate immune response is the first secondary defense tobe mounted. For example, complement and certain acute-phase proteins bind and help destroysome pathogens, and macrophages trap, engulf, and destroy others. Activated macrophagesalso secrete cytokines that cause inflammation, which among other things attracts otherphagocytic cells (e.g. neutrophils and monocytes). These cytokines are collectively calledinflammatory cytokines. They include interleukin-1α/β (IL-1), interleukin-6 (IL-6), and tumor

R. W. Johnson and J. Escobar84

Fig. 1. As pathogens in the environmentincrease, animal productivity decreases.

necrosis factor α (TNFα). Virus-infected cells produce interferon (IFN) α and γ cytokines thatinterfere with viral replication. The interferons increase expression of major histocompability(MHC) class I molecules on the surface of virally infected cells, thereby flagging these infectedcells for killing by cytotoxic T cells. They also activate natural killer (NK) cells that recognizeand kill virally infected cells. Often this set of responses suffices to eliminate or at least containthe infection. If the infection cannot be contained it will spread to the lymphatic system wheremacrophages and other specialized antigen-presenting cells (e.g. dendritic cells) present the anti-gen to lymphocytes so they can initiate the second secondary defense – adaptive immunity.

The adaptive immune system is more efficient at eliminating pathogens from the host body ascompared to the innate immune system. When macrophages ingest and degrade pathogenicmicroorganisms they process antigen from the pathogen and present it to lymphocytes. This is akey step for proliferation of the B cells that in turn differentiate into plasma cells that will produceantigen-specific antibody as well as forming memory B cells that provide long-lasting immunity.This component of adaptive immunity is called humoral or antibody-mediated immunity.

An antigen also stimulates T cells to form cytotoxic T cells and memory T cells. This compo-nent of adaptive immunity – cell-mediated immunity – is most effective against intracellularpathogens such as viruses. Antigen binds a specific receptor on a T cell, stimulating that cellto differentiate and proliferate. The result is formation of cytotoxic T cells and memoryT cells with receptors appropriate for the subject antigen.

Inflammatory cytokines have critical roles in orchestrating both innate and adaptiveimmune responses (table 2). For example, macrophages secrete IL-1β and TNFα to causeinflammation in order to facilitate movement of other effector cells to the infection site. IL-6produced by macrophages stimulates hepatocytes to synthesize and secrete acute-phaseproteins, which bind and help remove certain bacteria. IL-6 also stimulates B cells to differ-entiate into antibody-producing plasma cells. IFNα and γ inhibit virus replication and activateNK cells which hunt down and kill virally infected cells. And IL-1β stimulates T lymphocytesto express IL-2 and its receptor – a critical step for T-cell proliferation. A surprising findingin the late 1970s and early 1980s was that cytokines produced by activated leukocytes affectdisparate physiological systems and orchestrate a systemic response that also helps protectthe host animal. The systemic response initiated by inflammatory cytokines has profoundeffects on animal metabolism (table 2). How these cytokines might influence skeletal muscleprotein accretion in young animals is discussed herein.

3. CYTOKINES INHIBIT ANIMAL GROWTH

Animals with infections have reduced appetites, reduced growth rates, and convert feed toproduct in an inefficient manner. Indeed, feed intake and growth are usually inversely related

Cytokine regulation of protein accretion 85

Table 1

Epithelial barriers to infection

Mechanical Epithelial cells joined by tight junctionsCiliated epithelial cells and mucin that trap and remove pathogens

Chemical Bactericidal enzymes in saliva, sweat, tears, and gutLow pH in stomachAntibacterial peptides

Microbiological Normal flora in gastrointestinal tract produce antibacterial substances and compete against pathogenic microorganisms

to the level of interaction between the host immune system and pathogens (fig. 1). This is whyanimals kept in poorly sanitized environments that afford a high degree of host–pathogeninteraction eat less and grow more slowly than their counterparts kept in cleaner environ-ments. This indicates that the immune system “senses” the pathogenic environment andinteracts with the brain and other disparate physiological systems to regulate feed intake andgrowth. Initial studies showed that activation of the hypothalamic–pituitary–adrenal (HPA)axis, fever, and behavioral signs of illness (e.g. hypersomnia) could be induced by injectinganimals with cell-free supernatants collected from activated leukocytes (reviewed by Hart,1988). The biologically active molecule in the conditioned supernatants was subsequentlydetermined to be endogenous pyrogen – a protein eventually renamed IL-1. Thus, the immunesystem conveys its message to other physiological systems via inflammatory cytokines. It is nowdogma that inflammatory cytokines inhibit animal growth. Administration of inflammatorystimuli that increase circulating levels of TNFα, IL-1, and IL-6 (e.g. lipopolysaccharide, LPS),or injection of recombinant TNFα, IL-1, and IL-6, decrease skeletal muscle protein accretion.These cytokines can reduce skeletal muscle protein accretion in several ways.

3.1. Inflammatory cytokines decrease appetite

A prolonged reduction in feed intake depletes protein and fat reserves. In AIDS and certainneoplastic diseases, loss of lean body mass is correlated with increased morbidity and mor-tality (Dewys et al., 1980; Delmore, 1997; Roubenoff, 2000). One thought is that decreasednutrient intake reduces growth rate, or in adult animals perpetuates loss of skeletal musclemass, by limiting the supply of amino acids for protein synthesis. When decreased voluntaryfeed intake induced by an inflammatory challenge is accounted for by covariate analysis orby including a pair-fed control treatment, approximately 20–30% of the decreased growth canbe attributed to a reduction in nutrient intake (Ballinger et al., 2000). The inflammatory


Table 2

Immunological and metabolic effects of cytokines produced by macrophages

Cytokine Major immunological effects Major metabolic effects

Interleukin-1 Inflammation Muscle protein degradationActivates lymphocytes Reduced muscle protein synthesisT-cell proliferation Fever

AnorexiaHypoferremiaHypozincemiaHypercupremia

Interleukin-6 Activates lymphocytes Muscle protein degradationB-cell differentiation Reduced muscle protein synthesisAntibody production FeverAcute-phase protein synthesis Acute-phase protein synthesis

Tumor necrosis factor α Inflammation Muscle protein degradationReduced muscle protein synthesisFeverAnorexiaLipolysis

Interferon α/γ Activates natural killer cells Not generally considered to have Inhibits virus replication significant metabolic effects

cytokines produced by activated mononuclear phagocytic cells, IL-1, IL-6, and TNFα, reduceappetite and feed intake. The fact that IL-1, IL-6, and TNFα all reduce appetite illustrates animportant characteristic of this group of cytokines – redundancy. Indeed, blocking any one ortwo of the three cytokines did not prevent LPS-induced anorexia (Swiergiel and Dunn, 1999).Only when all three were antagonized simultaneously was LPS-induced anorexia prevented(Swiergiel and Dunn, 1999). Still, based on dose-response studies, IL-1 appears to be mostpotent at reducing appetite. Recombinant IL-1 injected peripherally (i.v., i.p., or s.c.) reducesanimals’ feed intake, similar to what occurs during an acute infection (Plata-Salaman et al.,1988). The decrease in feed intake is due to a decrease in meal frequency and size (Langhanset al., 1993). Cytokines can directly change the activity of hypothalamic neurons that medi-ate feed intake, or affect neurochemicals and neuropeptides that are implicated in the controlof feed intake. For example, peripheral injection of IL-1 caused anorexia and increased thesteady-state level of corticotropin-releasing hormone (CRH) mRNA in the hypothalamus(Suda et al., 1990). A CRH antagonist administered intracerebroventricularly (ICV) partiallyblocked IL-1β-induced anorexia (Uehara et al., 1989). IL-1 also decreases hypothalamicneuropeptide Y – a potent appetite-stimulating factor (Gayle et al., 1997). And LPS stimulatesthe release of α-melanocyte-stimulating hormone, which has been shown to enhanceLPS-induced anorexia (Huang et al., 1999).

3.1.1. Reduced feed intake is an adaptive response to infection

Many explanations for why sick animals reduce feed intake are based on teleology. For exam-ple, wild animals may expend considerable energy foraging or hunting for feed. Thus, feedingbehavior would enhance heat loss and thwart the beneficial fever response, place weakened, vulnerable animals in harm’s way of predators, and perhaps facilitate disease transmissionwithin a group. However, there is tangible evidence that the loss of appetite benefits sick ani-mals, too. In one study, researchers experimentally infected mice with Listeria monocytogenes(LD50) and let some consume feed ad libitum, while others were intubated and force-fed to thelevel of free-feeding, noninfected controls (Murray and Murray, 1979). Mice allowed to con-sume feed ad libitum ate 58% of the controls and were much more likely to survive than thoseforce-fed: nearly 100% of infected, force-fed mice died, whereas only about 50% of infected,ad libitum-fed mice died. Furthermore, there was a positive relationship between weight lossand survival for the infected mice with ad libitum access to feed. In some cases,survival appears to be positively related to anorexia and weight loss, provided it does notpersist too long. In general, the behavioral and metabolic responses to acute infection are ben-eficial because they inhibit the pathogen and enhance animals’ immunological defenses (fig. 2).

Animals seem to employ their nutritional wisdom and simply eat what they can use. Inother words, feed intake in a growing animal might be determined by its capacity to accreteprotein. When rats were injected with LPS or IL-1 and allowed to self-select betweenmacronutrients during a 4 h meal period, they decreased total caloric intake by about 50% butingested relatively less protein and more carbohydrate; relative fat intake was unchanged(Aubert et al., 1995). The fact that animals disproportionately reduced protein intake com-pared to other macronutrients during an inflammatory challenge suggests a shift in metabolicpriorities and nutrient needs. Furthermore, increasing the diet concentration of limiting aminoacids to account for decreased appetite of chicks and pigs under immunological stress is noteffective for increasing whole-body protein accretion (Williams et al., 1997a,b,c; Webel et al.,1998). Cytokines apparently reduce the animal’s capacity to accrete protein and feed intakeis adjusted accordingly.


3.1.2. How do cytokines affect feed intake regulatory centers?

Cytokines produced in the periphery can interact directly with central feed intake regulatorycenters by entering the circulatory system and moving from the blood into the brain (fig. 3).Recombinant inflammatory cytokines administered directly into the brain via an indwellingICV cannula, for example, induce anorexia, suggesting that cytokines act centrally to reducefeed intake (Plata-Salaman, 1988). Moreover infusing the IL-1 receptor antagonist ICV inorder to block IL-1 receptors in the brain inhibited anorexia caused by inflammation in theperiphery (Kent et al., 1992; McHugh et al., 1994). Because inflammatory cytokine proteinsare 17–26 kD in size, they are ordinarily too large to diffuse passively from the blood, acrossthe blood–brain barrier, into the brain. However, pathogens or cytokines might promotepassive movement of cytokine from the blood into the brain by increasing the permeability ofthe blood–brain barrier (de Vries et al., 1996). There is also evidence that cytokines areactively transported from the blood into the brain (fig. 3). For example, Banks and colleaguesinjected radiolabeled cytokine (e.g. IL-1, IL-6, and TNFα) intravenously and were able torecover from the brain a portion of what was injected (Banks et al., 1991, 1994a,b; Gutierrezet al., 1993). The transport mechanism for each cytokine was saturable and the transport ofradiolabeled cytokine could be competitively blocked by intravenous injection of unlabeledcytokine.

Peripheral cytokines may also access the brain through circumventricular organs, whichare devoid of blood–brain barrier (fig. 3). Here, peripherally produced cytokines diffuseinto the brain or stimulate glial cells, causing them to produce inflammatory molecules(e.g. prostaglandins and cytokines), which diffuse into the brain. Based on extensive temporaland spatial mapping of Fos expression (a marker for neural activity) and IL-1, Konsman et al.(1999) proposed that cytokines produced in the periphery affect the brain according to theprinciples of volume transmission. In this model, IL-1 or other inflammatory mediators inthe periphery induce IL-1 production in the choroid plexus and circumventricular organs. Thecytokine then slowly diffuses into the brain by volume transmission, along the way activatingneurons and neural pathways that result in anorexia (Konsman and Dantzer, 2001). Consistentwith this hypothesis, inflammatory stimuli in the periphery (e.g. LPS and inflammatory


Fig. 2. Mononuclear phagocytic cells produce inflammatory cytokines when activated by pathogens.The cytokine molecules act in the brain to reorganize the animal’s behavioral priorities. The sickness behaviorsyndrome that results is an adaptive response that enhances the animal’s immunological defenses and inhibitsproliferation of the pathogen. Thus, sickness behavior enhances disease resistance and promotes recovery(Johnson, 2002).

cytokines) induce de novo synthesis of IL-1, IL-6, and TNFα in the brain of the mouse andrat (Ban et al., 1992; Laye et al., 1994). For example, inflammatory stimuli in the peripheryinduce perivascular microglial cells to express cytokines (van Dam et al., 1992). Moreover,anorectic rats bearing prostate adenocarcinoma tumor cells had increased IL-1 mRNA in thecerebellum, cortex, and hypothalamus.

Cytokines in the periphery can also convey a message to the brain via the vagus nerve (fig. 3).After i.p. LPS challenge, dendritic cells and macrophages that are closely associated with theabdominal vagus express IL-1 protein (Goehler et al., 1999). IL-1 binding sites are evident inseveral regions of the vagus as well (Goehler et al., 1997). When activated by peripheralcytokines the vagus can activate specific neural pathways that are involved in sickness behavior.Activation of the vagus also appears to stimulate microglia in the brain to produce cytokines.If the vagus nerve is severed just below the diaphragm in rats, the expression of cytokines inthe brain and the sickness behavior that normally occurs after intraperitoneal injection of LPSis inhibited (Laye et al., 1995). Plasma levels of cytokines are elevated in LPS-injected vago-tomized rats, indicating that the neural signal is needed for the induction of sickness. Theneural signal may be necessary for the induction of cytokines in the brain, or may sensitizethe brain to cytokines produced in the periphery. The neural pathways activated in the brainby the vagus nerve for rapid immune-to-brain signaling have been recently described in somedetail (Dantzer, 2001a,b). These pathways appear to be responsible for activating the HPAaxis and depressing behavior in response to infection.

Cytokines originating in the periphery act on other peripheral targets as well, which in turnreduce appetite (fig. 3). For instance, leptin is a 16 kD protein secreted by adipocytes. Byacting in the hypothalamus to reduce appetite and increase energy expenditure, leptin playsan important role in long-term energy balance. Mice have increased circulating levels of


Fig. 3. Cytokines produced in the periphery can convey a message to the brain in several ways. Peripheralcytokines may cross the blood–brain barrier by diffusion or active transport. In addition, peripheral cytokinesmay activate the vagus nerve, which in turn induces cells in the brain (e.g. microglia) to produce cytokines.Finally, peripheral cytokines may stimulate the release of hormones that are able to cross the blood–brainbarrier. Adapted from Johnson (2002).

leptin when the immune system is stimulated with LPS (see Johnson and Finck, 2001). Thiseffect of LPS is cytokine-dependent because mice with a mutated toll-like receptor 4 gene –a defect that prevents them from secreting cytokines in response to LPS – do not haveincreased leptin when challenged with LPS. However, when mice with the mutation areinjected with recombinant TNFα, circulating leptin increases (Finck et al., 1998). TNFα actsdirectly on adipocytes via the p55 TNF receptor to induce expression of leptin (Finck andJohnson, 2000). It is reasonable to postulate that a cytokine-induced elevation in circulatingleptin is involved in the cytokine-induced anorexia. Accordingly, the anorectic response toLPS is attenuated in mice lacking the leptin receptor (Faggioni et al., 1997). However, micewith a mutated leptin gene reduce feed intake similar to mice with a fully functional leptingene after LPS injection (Faggioni et al., 1997), so this issue is not fully resolved.

3.2. Inflammatory cytokines decrease protein accretion in growing animals

Inflammatory cytokines are pleiotropic molecules that can either increase or decrease proteinsynthesis, depending on the target tissue. In general, protein synthesis is decreased in skeletalmuscle and is increased in liver, lung, and heart. In the liver, for example, inflammatorycytokines induce a marked increase in acute-phase protein synthesis. The weight of liver andtotal liver protein is increased in animals chronically infused with IL-1, TNFα, or the twocytokines together. However, animals infused with cytokines lose body weight becausemuscle protein accretion is reduced due to a decrease in protein synthesis and an increase inprotein degradation. The liver represents roughly 3% of animals’ total body mass whereasskeletal muscle represents 40–45%. Thus, the net effect of increased circulating cytokines ina growing animal is a decrease in whole-body protein accretion.

Administration of IL-1 and TNFα – alone or in combination – results in increased urinarynitrogen excretion and skeletal muscle catabolism accompanied by weight loss in rats (Floreset al., 1989; Ling et al., 1997). The effects of cytokines on protein kinetics are vastly differentfrom those induced by fasting or feed restriction, where peripheral proteins are spared andvisceral proteins are degraded. Chronic treatment with either TNFα or IL-1 results in a redis-tribution of body protein. Rats that were injected twice daily for 7 days with LPS, TNFα, orIL-1 lost a comparable amount of weight to respective pair-fed animals. However, the LPSand cytokine-treated animals had accelerated skeletal muscle protein degradation but pre-served liver protein content, which was not the case for pair-fed animals (Fong et al., 1989).The decrease in skeletal muscle protein under inflammatory conditions is associated withdecreases in steady-state levels of muscle mRNA for myofibrillar proteins myosin heavychain, myosin light chain, actin, and in 18S and 28S subunits of ribosomal RNA. Similarly,the body weight of transgenic mice that overexpress IL-6 is comparable to that of wild-typecontrols at 16 weeks of age. However, the transgenic mice have reduced gastrocnemiusmuscle weights and suffer from severe muscle atrophy, which is prevented by treatment withan antagonistic anti-mouse IL-6 receptor antibody (Tsujinaka et al., 1996). Furthermore,implantation of a TNFα-secreting tumor in the hind leg muscles of nude mice led within50 days to profound fat and protein loss (Tracey et al., 1990). In sepsis, protein synthesisand translational efficiency are reduced in gastrocnemius muscle, and prior treatment withTNF-binding protein (TNFBP) prevented these effects (Cooney et al., 1999). Prior treatmentwith IL-1 receptor antagonist also prevented the inhibitory effects of sepsis on protein synthe-sis (Vary et al., 1996). The release of one cytokine often initiates a cascade of cytokine synthesisand release. For example, animals challenged with E. coli or endotoxin and treated with TNFBPhad reduced plasma levels of IL-1 and IL-6 (Roth et al., 1998; Solorzano et al., 1998).


Consistent with these results, when pigs were challenged with LPS there was a markedincrease in circulating TNFα and IL-6 that preceded a 3-fold increase in plasma ureanitrogen (PUN). Because pigs were fasted, the increase in PUN was interpreted to suggest anincrease in skeletal muscle protein degradation (Webel et al., 1997). The Porcine Reproductiveand Respiratory Syndrome Virus (PRRSV) preferentially infects and replicates withinmononuclear phagocytic cells (i.e. macrophages). Mononuclear phagocytic cells infectedby PRRSV produce copious amounts of inflammatory cytokines (van Reeth et al., 1999;van Reeth and Nauwynck, 2000). Whole-body protein accretion is markedly reduced in nurserypigs infected with the PRRSV. There is a high negative correlation between protein accretionand circulating IL-1 and IL-6 (Escobar et al., 2002).

The influence of cytokines on protein synthesis and degradation seems to be dependent onskeletal muscle fiber type. Slow-twitch, or Type I muscle fibers are designed to work repeti-tively and generally use oxygen to fuel metabolic processes. Fast-twitch, or Type II musclefibers contract at a high rate of speed and work well in the absence of oxygen. Vary andKimball (1992) demonstrated that muscles consisting of fast-twitch fibers (gastrocnemius andpsoas) were subject to breakdown during sepsis whereas protein kinetics was unaffected bysepsis in muscles consisting of slow-twitch fibers (soleus and heart). The specific effects ofinflammatory cytokines on fast-twitch and slow-twitch fibers may be important in domesticfood-producing animals. For example, the longissimus muscle in domestic pigs containsfewer slow-twitch fibers and more fast-twitch fibers than wild boars of the same age (Essen-Gustavsson and Lindholm, 1984). Pigs and chickens intended for meat production have beenselected for maximal lean growth rate and increased breast-meat yield, respectively. Becausepigs selected for maximal lean growth rate have a greater proportion of muscles containingfast-twitch vs slow-twitch muscles (Rahelic and Puac, 1981), and chickens selected formaximal breast-meat yield likewise have higher levels of fast-twitch fibers, the effects ofcytokines on muscle tissue growth are potentially more deleterious in leaner, more moderngenotypes.

It appears that a portion of the amino acids released by skeletal muscle as a result ofprotein degradation are taken up by leukocytes to support cell proliferation and by the liverto support acute-phase protein synthesis (fig. 4). For example, the inflammatory cytokines,IL-1, IL-6, and TNFα, increase the rate of hepatic amino acid uptake (Argiles et al., 1989;Argiles and Lopez-Soriano, 1990) and protein synthesis (Klasing and Austic, 1984; Geigeret al., 1988; Ballmer et al., 1991). Reeds et al. (1994) proposed that a significant portion ofnitrogen excreted during an inflammatory response was the result of excessive demands forthe aromatic amino acids, phenylalanine, tyrosine, and tryptophan. Their conclusions werebased on a comparison of the amino acid profiles of the major acute-phase proteins producedby humans and the amino acid profile of mixed muscle protein. Analysis of the acute-phaseproteins indicated that four of the six proteins contained high levels of phenylalanine, fiveof the proteins were rich in tryptophan, and three contained high levels of tyrosine. By cal-culating the quantity of amino acids incorporated into a typical acute-phase protein mixture(850 mg/kg BW), they calculated that 1980 mg of muscle protein per kg body weight wouldneed to be liberated to supply an adequate quantity of phenylalanine for the increased hepaticprotein synthesis. The amino acids that are released in excess of the need for acute-phase pro-tein production (1980–850 mg) are catabolized because they cannot be used for proteinresynthesis due to the phenylalanine limitation, with the end result being an excessive excre-tion of nitrogen. Assuming that animals have a similar pattern and quantity of acute-phaseproteins, it is apparent that an infectious insult could result in a substantial amount of skeletalmuscle protein degradation and nitrogen excretion. For example, for a 100 kg pig there would


be roughly 200 g of protein broken down to supply amino acids for acute-phase protein syn-thesis, and approximately 13 g nitrogen would be excreted.

Glutamine is another example of how amino acids are repartitioned by cytokines (fig. 4).Skeletal muscle is the major repository of glutamine. During infection, there is a 2-foldincrease in glutamine release from skeletal muscle. Despite a significant increase in endoge-nous glutamine biosynthesis in skeletal muscle, intracellular glutamine becomes depleted.At the same time there is an 8- to 10-fold increase in hepatic glutamine uptake where it canbe used for (1) biosynthesis of nonessential amino acids; (2) gluconeogensesis; (3) energy;and (4) biosynthesis of urea, which is ultimately excreted.

4. CYTOKINES AND MUSCLE PROTEIN SYNTHESISAND DEGRADATION

When protein degradation remains constant, decreased protein accretion can occur when sub-strates necessary for protein synthesis are limiting or when signals that promote proteinsynthesis are thwarted. When protein synthesis remains constant, decreased protein accretioncan occur when signals that promote protein degradation are enhanced. Protein synthesis anddegradation can be influenced independently, so protein accretion is most profoundly affectedwhen there is a decrease in muscle protein synthesis and a concomitant increase in muscleprotein degradation. If the amount of degradation exceeds that of synthesis, muscle wastingoccurs. Inflammatory cytokines are uniquely qualified to manipulate protein accretionbecause they have the ability to simultaneously influence both protein synthesis and degra-dation. Thus, the immune system, by production of inflammatory cytokines, is able to adjustanimal growth according to the level of immunological challenge.

4.1. Inflammatory cytokines inhibit skeletal muscle protein synthesis

4.1.1. Inflammatory cytokines inhibit the GH–IGF-I axis

One way cytokines inhibit skeletal muscle protein synthesis is by adversely affecting thesomatotropic axis (fig. 5). Growth hormone induces IGF-I secretion, a potent growth factorthat is responsible for a wide range of anabolic processes. Insulin-like growth factor-Iincreases skeletal muscle mass by binding the type I IGF-I receptor and initiating a cascadeof intracellular signaling events that ultimately initiate protein synthesis. For a detailed


Fig. 4. Inflammatory cytokines inhibit skeletal muscle protein synthesis and enhance its degradation.A portion of the freed amino acids (AA; e.g. glutamine) is taken up by leukocytes to support cell proliferationand by the liver to support acute-phase protein (APP) synthesis and other metabolic processes.

description of IGF-I receptor signaling, the reader is referred to recent reviews (LeRoith,2000; Broussard et al., 2001; Nakae et al., 2001). In short, the IGF-I receptor belongs to thefamily of tyrosine kinase receptors. It is composed of two ligand-binding extracellular α subunitsand two transmembrane-spanning β subunits that have tyrosine kinase activity. Followingbinding of the IGF-I to the extracellular receptor α subunit, the receptor dimerizes, forms aheterotetramer (βααβ), and the tyrosine residues in the kinase domain of the β chains areautophosphorylated. The tyrosine phosphorylated IGF-I receptor causes tyrosine phosphory-lation of insulin receptor substrate (IRS)-1 and IRS-2 – docking molecules that can recruitand bind the p85 regulatory subunit of phosphatidylinositol 3′-kinase (PI 3-kinase). The IRSdocking molecules can lead to a sustained activation of PI 3-kinase, which is key in connect-ing several intracellular pathways that promote cell survival, differentiation, and proteinsynthesis.

In general, inflammatory stimuli reduce IGF-I levels and reduce sensitivity of receptors forGHRH, GH, and IGF-I. Receptors for IL-1 are present in the anterior pituitary and are localizedexclusively to somatotrophs – cells that produce GH (French et al., 1996). Stimulation of theimmune system with LPS causes species-specific changes in circulating GH levels. In humansand sheep, GH is increased, but in other animals including cattle, chickens, and rats it isdecreased. The effects of LPS and cytokines on GH are discussed elsewhere and the generalconclusion is that the effects are variable and not consistent among species (Broussard et al.,2001). Of seemingly greater importance is IGF-I, which is consistently depressed in immuno-logically challenged animals. Pigs injected with LPS or infected with Salmonella typhimuriumshowed a marked decrease in serum IGF-I levels, but little or no change in circulating GH(Balaji et al., 2000; Wright et al., 2000). An uncoupling of GH and IGF-I secretion has beenreported in a number of species and is due to the impaired ability of GH to induce hepaticIGF-I synthesis. TNFα and IL-1 decrease hepatic GH receptors and may inhibit post-receptorsignaling events necessary for IGF-I synthesis and release. Interleukin-6 also profoundlyaffects IGF-I. Transgenic mice that overexpressed IL-6 and wild-type mice injected withrecombinant IL-6 had decreased IGF-I levels and stunted growth (De Benedetti et al., 1997).


Fig. 5. Inflammatory cytokines can inhibit skeletal muscle protein synthesis by interfering with the secretionof growth hormone (GH) and insulin-like growth factor (IGF)-I. Cytokines also lead to GH and IGF-I receptorresistance.

Mice injected with IL-6 had decreased IGF-I levels even when feed intake was not depressed,indicating that the decrease in IGF-I in sick animals is not necessarily due to reduced feedintake.

Inflammatory cytokines can also act directly on skeletal muscle and induce IGF-I receptorresistance. Bona fide receptors for IL-1, IL-6, TNFα, and IFNγ are present in skeletal muscle(Zhang et al., 2000; Alvarez et al., 2002a) and a model for how cytokines might interfere withIGF-I receptor signaling has been proposed (Broussard et al., 2001). TNFα completelyinhibits the IGF-I-induced increase in protein synthesis of human myoblasts (Frost et al.,1997). Apparently, TNFα impairs the ability of IGF-I receptors to exert their biological effectwhen IGF-I ligand binds. TNFα reduces IGF-I-induced tyrosine phosphorylation of the IGF-Ireceptor and IRS (Venters et al., 1999). This decreases PI 3-kinase activity and thus inhibitsthe ability of IGF-I to promote protein synthesis. The IGF-I receptor resistance partiallyexplains why administration of recombinant IGF-I to rats with colitis failed to restore lineargrowth to that of control rats (Ballinger et al., 2000). However, most circulating IGF-I isbound to IGF-binding protein 3 (IGFBP3). A recent study showed that administration ofIGF-I/IGFBP3 binary complex enhanced protein synthesis in septic rats, suggesting that thedecreased responsiveness of muscle to exogenous IGF-I may be due to both an induction of IGF-I receptor resistance and decreased circulating levels of important IGF-I-binding proteins (Svanberg et al., 2000).

4.1.2. Inflammatory cytokines inhibit the initiation of mRNA translation

The synthesis of new protein – an obvious prerequisite to protein accretion – begins with theinitiation of mRNA translation (i.e. translation initiation; fig. 6). In eukaryotic cells, transla-tion initiation involves more than a dozen proteins referred to as eukaryotic initiation factors(eIF). For a complete up-to-date description of translation initiation and how the process is


Fig. 6. A truncated schematic diagram of the translation initiation process. See text for details on howcytokines inhibit the initiation of protein synthesis (eukaryotic initiation factor, eIF; binding protein, BP).

influenced by nutrients and hormones, the reader is referred to excellent reviews by Kimball(2002) and Shah et al. (2000). In short, translation initiation consists of four major steps(Cooney et al., 1997): (1) dissociation of the 80S ribosomal complex into the 40S and 60Sribosomal subunits; (2) binding of met-tRNAi (methyonil-tRNA initiator) to the 40S ribosomalsubunit to form the 43S pre-initiation complex; (3) binding of mRNA to the 43S pre-initiationcomplex: and (4) association of the 60S ribosomal subunit to form the active ribosome(fig. 6). Several recent reports indicate that steps 2 and 3 are affected most during infection,so these are discussed below.

4.1.2.1. Formation of the 43S pre-initiation complex The primary function of eIF2 is tobind met-tRNAi to the 40S ribosomal subunit. Thus, eIF2 can regulate translation initiationin two ways. First, a decrease in the amount of intracellular eIF2 or its activity will reducetranslation initiation. However, to our best knowledge a reduction in eIF2 during infectionwhen skeletal muscle protein synthesis is reduced has not been reported. Second, eIF2 activityis controlled by eIF2B, whose function is to exchange GDP for GTP in eIF2 (i.e., eIF2 . GDP+ eIF2B + GTP → eIF2 . GTP + eIF2B + GDP). Only the eIF2 .GTP complex is able to bindmRNA. Therefore, a reduction in eIF2 .GTP complex will decrease the translation initiationprocess. Indeed, a decrease in eIF2B was accompanied by a decrease in protein synthesis andtranslational efficiency in the gastrocnemius muscle of septic rats (Vary et al., 1994; Voisinet al., 1996b). In the same study, treatment of septic rats with IL-1 receptor antagonistreturned eIF2B and protein synthesis and translational efficiency in the gastrocnemius muscleto control levels. In a separate but similar study, injection of LPS did not affect eIF2B levelsin the gastrocnemius muscle, but eIF2B activity was decreased as was the rate of protein syn-thesis and translational efficiency (Lang et al., 2000). Thus, a change in either the amount oractivity of eIF2B is sufficient to influence skeletal muscle protein synthesis. Similar resultswere reported when rats were infused with TNFα (Lang et al., 2002). Administration of theTNFα antagonist, TNFBP, restored the translational efficiency and protein synthesis in septicrats to control levels (Cooney et al., 1999). Furthermore, TNFBP prevented the decrease ineIF2B in gastrocnemius muscle of septic rats.

4.1.2.2. Binding of mRNA to the 43S pre-initiation complex The primary function ofeIF4E is to bind mRNA to form the eIF4E . mRNA complex, which binds to eIF4G andeIF4A to form the active eIF4F (Gingras et al., 1999). The 43S pre-initiation complex directlybinds to eIF4F. The eIF4E is bound to its repressors 4E-binding protein (4E-BP1), 4E-BP2,and 4E-BP3, where 4E-BP1 is the predominant form in skeletal muscle (Vary and Kimball,2000). Kinases phosphorylate 4E-BP1 and free eIF4E. Thus, phosphorylation of 4E-BP1 andthe level and activity of eIF4E are important factors that can influence protein synthesis.Interference of this signaling cascade by cytokines might partially explain the decreased pro-tein synthesis in sick animals. However, it is not clear if the phosphorylation state of 4E-BP1and the level and activity of eIF4E are affected during infection (Vary and Kimball, 2000;Vary et al., 2001). For example, on the one hand, IGF-I stimulates protein synthesis in skele-tal muscle by inducing an intracellular signaling cascade (reviewed by Broussard et al., 2001)that ultimately results in the phosphorylation of 4E-BP1 and release of eIF4E. In septic rats,phosphorylation of 4E-BP1 is markedly reduced in the gastrocnemius muscle so there ismore 4E-BP1 associated with eIF4E. Accordingly, both protein synthesis and translationalefficiency are reduced (Svanberg et al., 2000). Administration of IGF-I/IGFBP3 binarycomplex to septic rats restored translational efficiency without affecting the phosphorylationstate of 4E-BP1 or the association of eIF4E with 4E-BP1 (Svanberg et al., 2000). Thus,how IGF-I/IGFBP3 increases protein synthesis during sepsis is not yet known. On the


other hand, insulin induces hyperphosphorylation of 4E-BP1 during sepsis, which causes thepredicted dissociation of eIF4E . 4E-BP1 complex (Vary et al., 2001). However, formation ofeIF4E . eIF4G complex, which is necessary for protein synthesis, was drastically reduced inseptic rats despite the availability of eIF4E (Vary et al., 2001). The reduced binding of eIF4Eto eIF4G may be important for inhibition of protein synthesis during infection.

In summary, anti-cytokine treatment during sepsis appears to prevent the decrease in eIF2Band enhance protein synthesis and translational efficiency. In contrast, administration ofIGF-I/IGFBP3 binary complex enhances protein synthesis and translational efficiency duringinfection even though the amount of eIF2B is reduced. Thus, regulation of eIF2B duringsepsis might be independent of IGF-I but under direct cytokine control.

4.2. Inflammatory cytokines enhance skeletal muscle protein degradation

4.2.1. ATP-ubiquitin-dependent proteolytic pathway

The majority of proteolysis in skeletal muscle, including short-lived and long-lived myofib-rillar proteins, is ATP-dependent and involves the cofactor ubiquitin (Ub) and the 26Sproteasome complex (Solomon and Goldberg, 1996; Hasselgren and Fischer, 2001; fig. 7).The functionality, structural features, regulation, and modes of action of the 26S proteasomehave been extensively reviewed (Gorbea et al., 1999; Tanahashi et al., 1999; Voges et al., 1999).In short, an enzyme (E1) activates Ub in the presence of ATP to produce Ub-adenylate, whichsubsequently is transferred to one of several Ub-carrier enzymes (E2). Ubiquitin is attachedto the ε-amino group in a lysine residue of a target protein directly by Ub-carrier enzymes orthrough the action of Ub-ligases (E3). Attachment of more than one Ub (poly-Ub) is generallyobserved in E3-dependent reactions. Poly-Ub is the preferred signal for protein degradationwithin the 26S proteasome. Protein degradation by the 26S proteasome results in many small


Fig. 7. A schematic diagram of the ATP–ubiquitin-dependent proteolytic pathway. See text for details onhow cytokines influence this proteolytic pathway (ubiquitin, Ub; enzyme, E).

peptide chains that can be further degraded to free amino acids. During or after proteasomedegradation, Ub is released from the remnant of the target protein and is recycled for later use.Not all Ub-tagged proteins are degraded because Ub can be freed by the action of specificisopeptidases. The specificity of the ubiquitination process is thought to be primarily due tothe many different enzyme isoforms.

Because infection is associated with proteolysis, the role of the ATP-Ub-dependent pathwayin skeletal muscle has been investigated. The increase in skeletal muscle protein degradationin septic or tumor-bearing rats is paralleled by an increase in ATP-Ub-dependent proteolysis(Temparis et al., 1994; Tiao et al., 1994). For example, protein degradation was increased inextensor digitorum longus muscle isolated from septic rats compare to controls (Tawa et al.,1997; Hobler et al., 1998). Protein degradation was inhibited when the muscle was incubatedin the presence of specific proteasome inhibitors, MG101 (N-acetyl-Leu-Leu-norleucinal;Hobler et al., 1998), lactacystin (Hobler et al., 1998), or MG132 (Tawa et al., 1997). Proteolysisincreases progressively in denervated muscle. Proteolysis in denervated soleus muscle wasinhibited 68% by the proteasome inhibitor, MG132 (Tawa et al., 1997). These results stronglyindicate the crucial role of the ATP-Ub-dependent pathway in skeletal muscle proteolysis.

As indicated earlier, skeletal muscle expresses receptors for IL-1, IL-6, TNFα, and IFNγ(Zhang et al., 2000; Alvarez et al., 2002a). Inflammatory cytokines appear to play a key rolein activating the ATP-Ub-dependent proteolytic pathway (Llovera et al., 1998a); however, thespecific role of each cytokine is not clear. Tumor necrosis factor α but not IL-1 increased UbmRNA in gastrocnemius muscle of rats 3 h after bolus injection (Garcia-Martinez et al.,1995). However, when examined 6 h after bolus injection, Ub mRNA in gastrocnemiusmuscle was increased by TNFα, IL-1, and IFNγ, but not IL-6 (Llovera et al., 1998a). Tumornecrosis factor α has been studied for its role in protein degradation more than any othercytokine, having been initially referred to as cachectin. For example, acute treatment of ratswith TNFα increased protein ubiquitination (Garcia-Martinez et al., 1993) and enhancedthe degradation of both total and myofibrillar proteins (Zamir et al., 1992; Fischer et al.,2001). And incubation of C2C12 murine myotubes with TNFα reduced total cellular proteincontent (Li et al., 1998; Li and Reid, 2000; Alvarez et al., 2002b). The protein degradationinduced by TNFα is partially due to activation of the ATP-Ub-dependent proteolytic pathway.During tumor growth, muscle wasting is associated with the activation of the ATP-Ub-dependentproteolytic pathway, which is mediated via cytokines. Immunoneutralization of TNFα witha polyclonal anti-TNF antibody blocked the increase in steady-state levels of Ub mRNA ingastrocnemius muscle of tumor-bearing rats (Llovera et al., 1996). Xanthine derivativessuch as pentoxifylline have been used to inhibit TNFα production. A new xanthine derivative,torbafylline, decreased plasma concentration of TNFα in rats injected with LPS and inYoshida sarcoma-bearing rats (Combaret et al., 2002). The decrease in circulating TNFαwas paralleled by decreases in muscle Ub mRNA, Ub-conjugated myofibrillar proteins,proteasome mRNA (C2 subunit of 20S), and proteasome-dependent proteolysis (Combaretet al., 2002).

The soluble TNF receptor can serve as a decoy for biologically active TNFα. Binding ofTNFα to the soluble receptor prevents the cytokine from binding cell membrane-boundreceptors, thus effectively eliminating the cytokine’s bioactivity. Implantation of Lewislung carcinoma cells caused a decrease in gastrocnemius muscle protein accumulation inboth wild-type mice and transgenic mice that overexpressed soluble TNF receptor (Lloveraet al., 1998b). However, the reduction in protein accumulation was substantially greaterin wild-type mice compared to transgenic mice. In both cases, when protein accumulationwas decreased, Ub mRNA in gastrocnemius muscle was increased. Tumor necrosis factor


receptor type I knockout mice implanted with Lewis lung carcinoma cells had reduced gas-trocnemius muscle wasting compared to wild-type controls (Llovera et al., 1998c).

Further evidence that IL-6 does not activate the ATP-Ub-dependent proteolytic pathwaycomes from studies with IL-6 knockout mice. Sepsis induced by cecal ligation and punctureinduced a similar increase in total and myofibrillar protein degradation in both wild-type andIL-6 knockout mice, suggesting that IL-6 was not necessary for protein degradation (Williamset al., 1998). Interestingly, in wild-type mice sepsis was accompanied by a marked increasein muscle Ub mRNA, whereas Ub mRNA was only moderately increased in muscle of IL-6knockout mice. In contrast, administration of IL-6 to rats reportedly increased total andmyofibrillar protein degradation (Goodman, 1994). The expression of Ub mRNA, however,is not always indicative of the proteolytic activity of the ATP-Ub-dependent pathway(Hasselgren, 2000). In contrast, transgenic mice that overexpressed IL-6 had severe muscleatrophy and increased levels of Ub mRNA in gastrocnemius muscle. The increase in UbmRNA levels was reduced when mice were treated with an anti-IL-6 antibody, which alsorestored muscle mass (Tsujinaka et al., 1996). These contradictory results exemplify thepotential difficulties of interpreting results from genetically altered animals. For example,TNFα, which induces proteolysis, was reportedly 3-fold higher in septic IL-6 knockout micecompared to wild-type controls (Fattori et al., 1994). Manipulating a single cytokine oftendisrupts the normal cytokine cascade. In addition, the ability of the ATP-Ub-dependent pathwayto degrade protein may depend on the activity of other proteolytic systems (e.g. Ca2+-dependent).Thus, it might be important to examine the effects of multiple cytokines and multiple proteolyticsystems simultaneously.

4.2.2. Calcium (Ca2+)-dependent proteolytic pathway

The Ca2+-dependent proteolytic pathway involves the proteases μ-calpain, m-calpain, andcalpain 3 (muscle-specific calpain p94). μ-Calpain and m-calpain are so named because theyare activated by micro- and millimolar concentrations of Ca2+, respectively. A concentrationof 1–20 μM of Ca2+ is required to activate μ-calpain, which exceeds the normal physiologicalintracellular concentration. Therefore, the calpains are generally inactive, so their role in normalcellular function is largely unknown (Stracher, 1999). Mitochondria and the sarcoplasmicreticulum release large amounts of Ca2+ postmortem and activate the calpains. The calpainsare most noted for the proteolytic changes in postmortem muscle, which are important forimproving tenderness (Koohmaraie, 1992).

The overall contribution of calpains to skeletal muscle protein degradation during infectionwhen protein accretion is decreased is relatively small compared to ATP-Ub-dependentproteolysis. The calpains, however, are important for degradation of certain muscle proteinsincluding the sarcomeric proteins (Huang and Forsberg, 1998). Even though the total muscleprotein degraded by calpains is relatively small, it is important because degradation ofsarcomeric proteins may facilitate degradation of myofilaments. In other words, the activityof the Ca2+-dependent pathway may facilitate the activity of the ATP-Ub-dependent pathway.It has been proposed that the Ca2+-dependent release of myofilaments from the sarcomereof myofibrils is the rate-limiting event for ATP-Ub-dependent proteasome degradation ofthe myofilaments (Hasselgren et al., 2002). Indeed, it appears that the activity of the Ca2+-dependent proteolytic pathway is increased during infection and that blockade of thispathway markedly reduces total skeletal muscle protein degradation.

Skeletal muscle m-calpain was increased in septic rats experiencing accelerated muscleprotein degradation (Voisin et al., 1996a). In a recent study, steady-state mRNA levels


of μ-calpain, m-calpain, and calpain 3 were increased and the Z-disks disrupted in extensordigitorum longus muscle of septic rats (Williams et al., 1999). The addition of dantrolene, aninhibitor of the release of Ca2+ from intracellular stores to the cytoplasm, significantlyreduced myofilament release from the sarcomere in septic rats, suggesting involvement of aCa2+-dependent proteolytic pathway – probably calpain (Williams et al., 1999). In support ofthis, dantrolene prevented the sepsis-induced increase in muscle Ca2+ levels, mRNA levels form-calpain, μ-calpain, and calpain 3, the increase in release of myofilaments, and total proteindegradation in extensor digitorum longus muscle (Fischer et al., 2001). Interestingly, dantro-lene reduced serum TNFα as well, suggesting an important but yet to be defined relationshipbetween cytokines, Ca2+-dependent proteolysis, and muscle wasting.

4.2.3. Lysosomal proteolytic pathway

Skeletal muscle contains relatively few lysosomes, but the involvement of the lysosomalproteolytic pathway in skeletal muscle protein degradation during disease has been investi-gated nonetheless. The main lysosomal proteases are cathepsins B, H, L, and D, which playa major role in the degradation of long-lived, soluble, and integral membrane proteins.Cathepsins, however, are unable to degrade myofibrillar proteins (Furuno et al., 1990), so thelysosomal proteolytic pathway’s contribution to disease-induced muscle wasting is consid-ered to be relatively small. Inconsistent findings have been reported for the activity of thelysosomal proteolytic pathway in skeletal muscle of wasting animals (Llovera et al., 1994,1995; Temparis et al., 1994; Baracos et al., 1995; Voisin et al., 1996a). For example, musclelevels of cathepsin B were increased in septic rats compared to pair-fed controls (Voisin et al.,1996a), whereas cathepsin B and B + L activities and cathepsin B mRNA were unchanged inwasting tumor-bearing rats (Temparis et al., 1994). Transgenic mice that overexpressed IL-6suffered from muscle atrophy and had increased steady-state mRNA levels of cathepsins Band L as well as ubiquitin. Injecting IL-6 transgenic mice with an IL-6 receptor-blocking anti-body decreased cathepsin (B and L) and Ub mRNA. In injured muscle, lysosomal proteolysisappears to be mediated by enzymes produced primarily by infiltrating macrophages (Fargeset al., 2002). Enzymes from macrophages may not be a prerequisite, however. Incubation ofC2C12 myotubes with IL-6 reduced the half-life of long-lived proteins and increased cathep-sin B + L activity (Ebisui et al., 1995). It is possible that the lysosomal pathway participatesin muscle wasting indirectly by controlling degradation of intracellular regulatory proteins.


Most of what is known about the in vivo effects of cytokines on skeletal muscle proteinsynthesis and degradation has been derived from adult animal models of sepsis or cancer –diseases that are often characterized by severe muscle wasting. These are important modelsfor human medicine and have served to amplify the host’s responses to acute immune systemactivation, which is useful for elucidating the underlying mechanisms involved. It must benoted, however, that these diseases are only marginally significant to growing domestic food-producing animals. In general, the most significant diseases in growing agricultural animalsdo not cause muscle wasting, but instead cause a significant and chronic decrease in skeletalmuscle protein accretion. Thus, despite the presence of infectious disease, farm animals oftencontinue to grow. In any case, it is reasonable to postulate that the same machinery that resultsin muscle wasting in sepsis, for example, is involved in decreasing protein accretion ininfected slow-growing animals. Therefore, results of studies on muscle wasting might be


germane to increasing protein synthesis and reducing protein degradation in slow-growinganimals. However, it is not clear if the “markers” of protein synthesis and degradation evidentin animals undergoing severe muscle wasting are detectable in farm animals that are merelyaccreting skeletal muscle protein at a less than maximal rate due to infection.

To defend the host animal from infectious pathogens, the immune system and liver usesome of the amino acids made available by the effects of cytokines on skeletal muscle. Thus,in growing animals we believe that the repartitioning of nutrients is a constructive adaptationthat enables the animal to contend against the pathogen and continue to accrete protein, albeitat a decreased rate. It should be possible to develop novel nutrition programs that better matchthe animal’s metabolic state(s) during an infection, so that protein accretion is maintainedwhile the needs of the defense systems are met. However, the nutrient requirements foragriculture animals are, for the most part, based on experiments conducted in laboratorysituations where exposure to infectious pathogens and other stresses is minimized. Therefore,the estimated nutrient requirements for animals have been established to maximize productionof healthy animals. If the nutritional requirements of slow-growing infected animals can beprecisely defined, it will be possible to formulate cost-effective diets that maximize proteinaccretion under the given circumstance.

Of course the goal should be to minimize infectious disease in animal production systems.However, because many infectious pathogens are endemic, it will be necessary to understandhow and why the immune system regulates protein accretion in growing animals. This isparticularly important because even animals with subclinical infections have reduced growth.How cytokines simultaneously influence systems involved in protein synthesis and degradationin growing animals is needed because certainly the whole is greater than the sum of their effects.

REFERENCES

Alvarez, B., Quinn, L.S., Busquets, S., Lopez-Soriano, F.J., Argiles, J.M., 2002a. TNF-α modulatescytokine and cytokine receptors in C2C12 myotubes. Cancer Lett. 175, 181–185.

Alvarez, B., Quinn, L.S., Busquets, S., Quiles, M.T., Lopez-Soriano, F.J., Argiles, J.M., 2002b. Tumornecrosis factor-α exerts interleukin-6-dependent and -independent effects on cultured skeletal musclecells. Biochim. Biophys. Acta 1542, 66–72.

Argiles, J.M., Lopez-Soriano, F.J., 1990. The effects of tumour necrosis factor-α (cachectin) and tumourgrowth on hepatic amino acid utilization in the rat. Biochem. J. 266, 123–126.

Argiles, J.M., Lopez-Soriano, F.J., Wiggins, D., Williamson, D.H., 1989. Comparative effects of tumournecrosis factor-α (cachectin), interleukin-1β and tumour growth on amino acid metabolism in the ratin vivo: absorption and tissue uptake of α-amino[1-14C]isobutyrate. Biochem. J. 261, 357–362.

Aubert, A., Goodall, G., Dantzer, R., 1995. Compared effects of cold ambient temperature and cytokineson macronutrient intake in rats. Physiol. Behav. 57, 869–873.

Balaji, R., Wright, K.J., Hill, C.M., Dritz, S.S., Knoppel, E.L., Minton, J.E., 2000. Acute phase responsesof pigs challenged orally with Salmonella typhimurium. J. Anim. Sci. 78, 1885–1891.

Ballinger, A.B., Azooz, O., El-Haj, T., Poole, S., Farthing, M.J., 2000. Growth failure occurs through adecrease in insulin-like growth factor 1 which is independent of undernutrition in a rat model of colitis.Gut 46, 694–700.

Ballmer, P.E., McNurlan, M.A., Southorn, B.G., Grant, I., Garlick, P.J., 1991. Effects of human recombi-nant interleukin-1β on protein synthesis in rat tissues compared with a classical acute-phase reactioninduced by turpentine: rapid response of muscle to interleukin-1β. Biochem. J. 279, 683–688.

Ban, E., Haour, F., Lenstra, R., 1992. Brain interleukin 1 gene expression induced by peripherallipopolysaccharide administration. Cytokine 4, 48–54.

Banks, W.A., Kastin, A.J., Ehrensing, C.A., 1994a. Blood-borne interleukin-1α is transported across theendothelial blood-spinal cord barrier of mice. J. Physiol. 479, 257–264.

Banks, W.A., Kastin, A.J., Gutierrez, E.G., 1994b. Penetration of interleukin-6 across the murineblood-brain barrier. Neurosci. Lett. 179, 53–56.


Banks, W.A., Ortiz, L., Plotkin, S.R., Kastin, A.J., 1991. Human interleukin (IL) 1α, murine IL-1α andmurine IL-1β are transported from blood to brain in the mouse by a shared saturable mechanism.J. Pharmacol. Exp. Ther. 259, 988–996.

Baracos, V.E., DeVivo, C., Hoyle, D.H., Goldberg, A.L., 1995. Activation of the ATP-ubiquitin-proteasome pathway in skeletal muscle of cachectic rats bearing a hepatoma. Amer. J. Physiol. 268,E996–E1006.

Blalock, J.E., 1984. The immune system as a sensory organ. J. Immunol. 132, 1067–1070.Broussard, S., Zhou, J.H., Venters, H.D., Bluthé, R.M., Johnson, R.W., Dantzer, R., Kelley, K.W., 2001.

At the interface of environment-immune interactions: cytokines and growth factor receptors. J. Anim. Sci.79, E268–E284.

Combaret, L., Tilignac, T., Claustre, A., Voisin, L., Taillandier, D., Obled, C., Tanaka, K., Attaix, D.,2002. Torbafylline (HWA 448) inhibits enhanced skeletal muscle ubiquitin-proteasome-dependentproteolysis in cancer and septic rats. Biochem. J. 361, 185–192.

Cooney, R., Kimball, S.R., Eckman, R., Maish, G. 3rd, Shumate, M., Vary, T.C., 1999. TNF-bindingprotein ameliorates inhibition of skeletal muscle protein synthesis during sepsis. Amer. J. Physiol.276, E611–E619.

Cooney, R.N., Kimball, S.R., Vary, T.C., 1997. Regulation of skeletal muscle protein turnover duringsepsis: mechanisms and mediators. Shock 7, 1–16.

Dantzer, R., 2001a. Cytokine-induced sickness behavior: mechanisms and implications. Ann. N.Y. Acad.Sci. 933, 222–234.

Dantzer, R., 2001b. Cytokine-induced sickness behavior: where do we stand? Brain Behav. Immun.15, 7–24.

De Benedetti, F., Alonzi, T., Moretta, A., Lazzaro, D., Costa, P., Poli, V., Martini, A., Ciliberto, G., Fattori, E.,1997. Interleukin 6 causes growth impairment in transgenic mice through a decrease in insulin-likegrowth factor-I: a model for stunted growth in children with chronic inflammation. J. Clin. Invest. 99,643–650.

Delmore, G., 1997. Assessment of nutritional status in cancer patients: widely neglected? Support. CareCancer. 5, 376–380.

de Vries, H.E., Blom-Roosemalen, M.C., van Oosten, M., de Boer, A.G., van Berkel, T.J., Breimer, D.D.,Kuiper, J., 1996. The influence of cytokines on the integrity of the blood-brain barrier in vitro.J. Neuroimmunol. 64, 37–43.

Dewys, W.D., Begg, C., Lavin, P.T., Band, P.R., Bennett, J.M., Bertino, J.R., Cohen, M.H., Douglass, H.O.Jr., Engstrom, P.F., Ezdinli, E.Z., Horton, J., Johnson, G.J., Moertel, C.G., Oken, M.M., Perlia, C.,Rosenbaum, C., Silverstein, M.N., Skeel, R.T., Sponzo, R.W., Tormey, D.C., 1980. Prognostic effectof weight loss prior to chemotherapy in cancer patients. Eastern Cooperative Oncology Group. Amer.J. Med. 69, 491–497.

Ebisui, C., Tsujinaka, T., Morimoto, T., Kan, K., Iijima, S., Yano, M., Kominami, E., Tanaka, K., Monden, M.,1995. Interleukin-6 induces proteolysis by activating intracellular proteases (cathepsins B and L,proteasome) in C2C12 myotubes. Clin. Sci. (Colch.) 89, 431–439.

Escobar, J., Toepfer, T.L., Van Alstine, W.G., Baker, D.H., Johnson, R.W., 2002. Porcine Reproductiveand Respiratory Syndrome Virus (PRRSV) but not Mycoplasma hyopneumoniae (Mh) decreasesprotein accretion and markedly increases skeletal muscle myostatin (MSTN) gene expression inyoung pigs. FASEB J. 16, A618–A619.

Essen-Gustavsson, B., Lindholm, A., 1984. Fiber types and metabolic characteristics in muscles ofwild boars, normal and halothane sensitive Swedish landrace pigs. Comp. Biochem. Physiol. A.78, 67–71.

Faggioni, R., Fuller, J., Moser, A., Feingold, K.R., Grunfeld, C., 1997. LPS-induced anorexia inleptin-deficient (ob/ob) and leptin receptor-deficient (db/db) mice. Amer. J. Physiol. 273,R181–R186.

Farges, M.C., Balcerzak, D., Fisher, B.D., Attaix, D., Bechet, D., Ferrara, M., Baracos, V.E., 2002.Increased muscle proteolysis after local trauma mainly reflects macrophage-associated lysosomalproteolysis. Amer. J. Physiol. Endocrinol. Metab. 282, E326–E335.

Fattori, E., Cappelletti, M., Costa, P., Sellitto, C., Cantoni, L., Carelli, M., Faggioni, R., Fantuzzi, G.,Ghezzi, P., Poli, V., 1994. Defective inflammatory response in interleukin 6-deficient mice. J. Exp.Med. 180, 1243–1250.

Finck, B.N., Johnson, R.W., 2000. Tumor necrosis factor (TNF)-α induces leptin production through thep55 TNF receptor. Amer. J. Physiol. Regul. Integr. Comp. Physiol. 278, R537–R543.


Finck, B.N., Kelley, K.W., Dantzer, R., Johnson, R.W., 1998. In vivo and in vitro evidence for theinvolvement of tumor necrosis factor-α in the induction of leptin by lipopolysaccharide.Endocrinology 139, 2278–2283.

Fischer, D.R., Sun, X., Williams, A.B., Gang, G., Pritts, T.A., James, J.H., Molloy, M., Fischer, J.E., Paul, R.J.,Hasselgren, P.O., 2001. Dantrolene reduces serum TNFα and corticosterone levels and musclecalcium, calpain gene expression, and protein breakdown in septic rats. Shock 15, 200–207.

Flores, E.A., Bistrian, B.R., Pomposelli, J.J., Dinarello, C.A., Blackburn, G.L., Istfan, N.W., 1989.Infusion of tumor necrosis factor/cachectin promotes muscle catabolism in the rat: a synergistic effectwith interleukin 1. J. Clin. Invest. 83, 1614–1622.

Fong, Y., Moldawer, L.L., Marano, M., Wei, H., Barber, A., Manogue, K., Tracey, K.J., Kuo, G.,Fischman, D.A., Cerami, A., et al., 1989. Cachectin/TNF or IL-1α induces cachexia with redistribu-tion of body proteins. Amer. J. Physiol. 256, R659–R665.

French, R.A., Zachary, J.F., Dantzer, R., Frawley, L.S., Chizzonite, R., Parnet, P., Kelley, K.W., 1996.Dual expression of p80 type I and p68 type II interleukin-I receptors on anterior pituitary cellssynthesizing growth hormone. Endocrinology 137, 4027–4036.

Frost, R.A., Lang, C.H., Gelato, M.C., 1997. Transient exposure of human myoblasts to tumor necrosisfactor-α inhibits serum and insulin-like growth factor-I stimulated protein synthesis. Endocrinology138, 4153–4159.

Furuno, K., Goodman, M.N., Goldberg, A.L., 1990. Role of different proteolytic systems in thedegradation of muscle proteins during denervation atrophy. J. Biol. Chem. 265, 8550–8557.

Garcia-Martinez, C., Agell, N., Llovera, M., Lopez-Soriano, F.J., Argiles, J.M., 1993. Tumour necrosisfactor-α increases the ubiquitinization of rat skeletal muscle proteins. FEBS Lett. 323, 211–214.

Garcia-Martinez, C., Llovera, M., Agell, N., Lopez-Soriano, F.J., Argiles, J.M., 1995. Ubiquitin geneexpression in skeletal muscle is increased during sepsis: involvement of TNF-α but not IL-1.Biochem. Biophys. Res. Commun. 217, 839–844.

Gayle, D., Ilyin, S.E., Plata-Salaman, C.R., 1997. Central nervous system IL-1β system and neuropep-tide Y mRNAs during IL-1β-induced anorexia in rats. Brain Res. Bull. 44, 311–317.

Geiger, T., Andus, T., Klapproth, J., Hirano, T., Kishimoto, T., Heinrich, P.C., 1988. Induction of ratacute-phase proteins by interleukin 6 in vivo. Eur. J. Immunol. 18, 717–721.

Gingras, A.C., Raught, B., Sonenberg, N., 1999. eIF4 initiation factors: effectors of mRNA recruitmentto ribosomes and regulators of translation. Annu. Rev. Biochem. 68, 913–963.

Goehler, L.E., Gaykema, R.P., Nguyen, K.T., Lee, J.E., Tilders, F.J., Maier, S.F., Watkins, L.R., 1999.Interleukin-1β in immune cells of the abdominal vagus nerve: a link between the immune andnervous systems? J. Neurosci. 19, 2799–2806.

Goehler, L.E., Relton, J.K., Dripps, D., Kiechle, R., Tartaglia, N., Maier, S.F., Watkins, L.R., 1997. Vagalparaganglia bind biotinylated interleukin-1 receptor antagonist: a possible mechanism for immune-to-brain communication. Brain Res. Bull. 43, 357–364.

Goodman, M.N., 1994. Interleukin-6 induces skeletal muscle protein breakdown in rats. Proc. Soc.Exp. Biol. Med. 205, 182–185.

Gorbea, C., Taillandier, D., Rechsteiner, M., 1999. Assembly of the regulatory complex of the 26Sproteasome. Mol. Biol. Rep. 26, 15–19.

Gutierrez, E.G., Banks, W.A., Kastin, A.J., 1993. Murine tumor necrosis factor α is transported fromblood to brain in the mouse. J. Neuroimmunol. 47, 169–176.

Hart, B.L., 1988. Biological basis of the behavior of sick animals. Neurosci. Biobehav. Rev. 12,123–137.

Hasselgren, P.O., 2000. Catabolic response to stress and injury: implications for regulation. World J. Surg.24, 1452–1459.

Hasselgren, P.O., Fischer, J.E., 2001. Muscle cachexia: current concepts of intracellular mechanisms andmolecular regulation. Ann. Surg. 233, 9–17.

Hasselgren, P.O., Wray, C., Mammen, J., 2002. Molecular regulation of muscle cachexia: it may be morethan the proteasome. Biochem. Biophys. Res. Commun. 290, 1–10.

Hobler, S.C., Tiao, G., Fischer, J.E., Monaco, J., Hasselgren, P.O., 1998. Sepsis-induced increase inmuscle proteolysis is blocked by specific proteasome inhibitors. Amer. J. Physiol. 274, R30–R37.

Huang, J., Forsberg, N.E., 1998. Role of calpain in skeletal-muscle protein degradation. Proc. Natl. Acad.Sci. USA 95, 12100–12105.

Huang, Q.H., Hruby, V.J., Tatro, J.B., 1999. Role of central melanocortins in endotoxin-induced anorexia.Amer. J. Physiol. 276, R864–R871.


Johnson, R.W., 1997. Inhibition of growth by pro-inflammatory cytokines: an integrated view. J. Anim.Sci. 75, 1244–1255.

Johnson, R.W., 2002. The concept of sickness behavior: a brief chronological account of four keydiscoveries. Vet. Immunol. Immunopathol. 87, 443–450.

Johnson, R.W., Finck, B.N., 2001. Tumor necrosis factor α and leptin: two players in animals’ metabolicand immunologic responses to infection. J. Anim. Sci. 79, E118–E127.

Kent, S., Bluthe, R.M., Kelley, K.W., Dantzer, R., 1992. Sickness behavior as a new target for drug devel-opment. Trends Pharmacol. Sci. 13, 24–28.

Kimball, S.R., 2002. Regulation of global and specific mRNA translation by amino acids. J. Nutr.132, 883–886.

Klasing, K.C., 1988. Nutritional aspects of leukocytic cytokines. J. Nutr. 118, 1436–1446.Klasing, K.C., Austic, R.E., 1984. Changes in protein synthesis due to an inflammatory challenge. Proc.

Soc. Exp. Biol. Med. 176, 285–291.Konsman, J.P., Dantzer, R., 2001. How the immune and nervous systems interact during disease-associated

anorexia. Nutrition 17, 664–668.Konsman, J.P., Kelley, K., Dantzer, R., 1999. Temporal and spatial relationships between lipopolysaccharide-

induced expression of Fos, interleukin-1β and inducible nitric oxide synthase in rat brain.Neuroscience 89, 535–548.

Koohmaraie, M., 1992. The role of Ca2+-dependent proteases (calpains) in post mortem proteolysis andmeat tenderness. Biochimie. 74, 239–245.

Lang, C.H., Frost, R.A., Jefferson, L.S., Kimball, S.R., Vary, T.C., 2000. Endotoxin-induced decrease inmuscle protein synthesis is associated with changes in eIF2B, eIF4E, and IGF-I. Amer. J. Physiol.Endocrinol. Metab. 278, E1133–E1143.

Lang, C.H., Frost, R.A., Nairn, A.C., MacLean, D.A., Vary, T.C., 2002. TNF-α impairs heart and skeletalmuscle protein synthesis by altering translation initiation. Amer. J. Physiol. Endocrinol. Metab. 282,E336–E347.

Langhans, W., Savoldelli, D., Weingarten, S., 1993. Comparison of the feeding responses to bacteriallipopolysaccharide and interleukin-1 β. Physiol. Behav. 53, 643–649.

Laye, S., Bluthe, R.M., Kent, S., Combe, C., Medina, C., Parnet, P., Kelley, K., Dantzer, R., 1995.Subdiaphragmatic vagotomy blocks induction of IL-1β mRNA in mice brain in response to peripheralLPS. Amer. J. Physiol. 268, R1327–R1331.

Laye, S., Parnet, P., Goujon, E., Dantzer, R., 1994. Peripheral administration of lipopolysaccharide inducesthe expression of cytokine transcripts in the brain and pituitary of mice. Brain Res. Mol. Brain Res.27, 157–162.

LeRoith, D., 2000. Insulin-like growth factor I receptor signaling: overlapping or redundant pathways?Endocrinology 141, 1287–1288.

Li, Y.P., Reid, M.B., 2000. NF-κB mediates the protein loss induced by TNF-α in differentiated skeletalmuscle myotubes. Amer. J. Physiol. Regul. Integr. Comp. Physiol. 279, R1165–R1170.

Li, Y.P., Schwartz, R.J., Waddell, I.D., Holloway, B.R., Reid, M.B., 1998. Skeletal muscle myocytesundergo protein loss and reactive oxygen-mediated NF-κB activation in response to tumor necrosisfactor α. FASEB J. 12, 871–880.

Ling, P.R., Schwartz, J.H., Bistrian, B.R., 1997. Mechanisms of host wasting induced by administrationof cytokines in rats. Amer. J. Physiol. 272, E333–E339.

Llovera, M., Carbo, N., Garcia-Martinez, C., Costelli, P., Tessitore, L., Baccino, F.M., Agell, N.,Bagby, G.J., Lopez-Soriano, F.J., Argiles, J.M., 1996. Anti-TNF treatment reverts increased muscleubiquitin gene expression in tumour-bearing rats. Biochem. Biophys. Res. Commun. 221, 653–655.

Llovera, M., Carbo, N., Lopez-Soriano, J., Garcia-Martinez, C., Busquets, S., Alvarez, B., Agell, N.,Costelli, P., Lopez-Soriano, F.J., Celada, A., Argiles, J.M., 1998a. Different cytokines modulate ubiq-uitin gene expression in rat skeletal muscle. Cancer Lett. 133, 83–87.

Llovera, M., Garcia-Martinez, C., Agell, N., Lopez-Soriano, F.J., Argiles, J.M., 1995. Muscle wastingassociated with cancer cachexia is linked to an important activation of the ATP-dependent ubiquitin-mediated proteolysis. Int. J. Cancer. 61, 138–141.

Llovera, M., Garcia-Martinez, C., Agell, N., Marzabal, M., Lopez-Soriano, F.J., Argiles, J.M., 1994. Ubiquitingene expression is increased in skeletal muscle of tumour-bearing rats. FEBS Lett. 338, 311–318.

Llovera, M., Garcia-Martinez, C., Lopez-Soriano, J., Agell, N., Lopez-Soriano, F.J., Garcia, I., Argiles, J.M.,1998b. Protein turnover in skeletal muscle of tumour-bearing transgenic mice overexpressing the sol-uble TNF receptor-1. Cancer Lett. 130, 19–27.


Llovera, M., Garcia-Martinez, C., Lopez-Soriano, J., Carbo, N., Agell, N., Lopez-Soriano, F.J., Argiles, J.M.,1998c. Role of TNF receptor 1 in protein turnover during cancer cachexia using gene knockout mice.Mol. Cell. Endocrinol. 142, 183–189.

McHugh, K.J., Collins, S.M., Weingarten, H.P., 1994. Central interleukin-1 receptors contribute tosuppression of feeding after acute colitis in the rat. Amer. J. Physiol. 266, R1659–R1663.

Murray, M.J., Murray, A.B., 1979. Anorexia of infection as a mechanism of host defense. Amer. J. Clin.Nutr. 32, 593–596.

Nakae, J., Kido, Y., Accili, D., 2001. Distinct and overlapping functions of insulin and IGF-I receptors.Endocr. Rev. 22, 818–835.

Plata-Salaman, C.R., 1988. Food intake suppression by growth factors and platelet peptides by directaction in the central nervous system. Neurosci. Lett. 94, 161–166.

Plata-Salaman, C.R., Oomura, Y., Kai, Y., 1988. Tumor necrosis factor and interleukin-1β: suppressionof food intake by direct action in the central nervous system. Brain Res. 448, 106–114.

Rahelic, S., Puac, S., 1981. Fiber types in longissimus dorsi from wild and highly selected pig breeds.Meat Sci. 5, 439.

Reeds, P.J., Fjeld, C.R., Jahoor, F., 1994. Do the differences between the amino acid compositionsof acute-phase and muscle proteins have a bearing on nitrogen loss in traumatic states? J. Nutr.124, 906–910.

Roth, J., Martin, D., Storr, B., Zeisberger, E., 1998. Neutralization of pyrogen-induced tumour necrosisfactor by its type 1 soluble receptor in guinea-pigs: effects on fever and interleukin-6 release.J. Physiol. 509, 267–275.

Roubenoff, R., 2000. Acquired immunodeficiency syndrome wasting, functional performance, and qualityof life. Amer. J. Manag. Care 6, 1003–1016.

Shah, O.J., Anthony, J.C., Kimball, S.R., Jefferson, L.S., 2000. 4E-BP1 and S6K1: translational integra-tion sites for nutritional and hormonal information in muscle. Amer. J. Physiol. Endocrinol. Metab.279, E715–E729.

Solomon, V., Goldberg, A.L., 1996. Importance of the ATP-ubiquitin-proteasome pathway in the degra-dation of soluble and myofibrillar proteins in rabbit muscle extracts. J. Biol. Chem. 271,26690–26697.

Solorzano, C.C., Kaibara, A., Hess, P.J., Edwards, P.D., Ksontini, R., Abouhamze, A., McDaniel, S.,Frazier, J., Trujillo, D., Kieft, G., Seely, J., Kohno, T., Cosenza, M.E., Clare-Salzler, M., MacKay, S.L.,Martin, S.W., Moldawer, L.L., Edwards, C.K. 3rd, 1998. Pharmacokinetics, immunogenicity, andefficacy of dimeric TNFR binding proteins in healthy and bacteremic baboon. J. Appl. Physiol.84, 1119–1130.

Stracher, A., 1999. Calpain inhibitors as therapeutic agents in nerve and muscle degeneration. Ann.N.Y. Acad. Sci. 884, 52–59.

Suda, T., Tozawa, F., Ushiyama, T., Sumitomo, T., Yamada, M., Demura, H., 1990. Interleukin-1stimulates corticotropin-releasing factor gene expression in rat hypothalamus. Endocrinology126, 1223–1228.

Svanberg, E., Frost, R.A., Lang, C.H., Isgaard, J., Jefferson, L.S., Kimball, S.R., Vary, T.C., 2000.IGF-I/IGFBP-3 binary complex modulates sepsis-induced inhibition of protein synthesis in skeletalmuscle. Amer. J. Physiol. Endocrinol. Metab. 279, E1145–E1158.

Swiergiel, A.H., Dunn, A.J., 1999. The roles of IL-1, IL-6, and TNFα in the feeding responses toendotoxin and influenza virus infection in mice. Brain Behav. Immun. 13, 252–265.

Tanahashi, N., Kawahara, H., Murakami, Y., Tanaka, K., 1999. The proteasome-dependent proteolyticsystem. Mol. Biol. Rep. 26, 3–9.

Tawa, N.E. Jr., Odessey, R., Goldberg, A.L., 1997. Inhibitors of the proteasome reduce the acceleratedproteolysis in atrophying rat skeletal muscles. J. Clin. Invest. 100, 197–203.

Temparis, S., Asensi, M., Taillandier, D., Aurousseau, E., Larbaud, D., Obled, A., Bechet, D., Ferrara, M.,Estrela, J.M., Attaix, D., 1994. Increased ATP-ubiquitin-dependent proteolysis in skeletal muscles oftumor-bearing rats. Cancer Res. 54, 5568–5573.

Tiao, G., Fagan, J.M., Samuels, N., James, J.H., Hudson, K., Lieberman, M., Fischer, J.E., Hasselgren, P.O.,1994. Sepsis stimulates nonlysosomal, energy-dependent proteolysis and increases ubiquitin mRNAlevels in rat skeletal muscle. J. Clin. Invest. 94, 2255–2264.

Tracey, K.J., Morgello, S., Koplin, B., Fahey, T.J. 3rd, Fox, J., Aledo, A., Manogue, K.R., Cerami, A.,1990. Metabolic effects of cachectin/tumor necrosis factor are modified by site of production:cachectin/tumor necrosis factor-secreting tumor in skeletal muscle induces chronic cachexia,


while implantation in brain induces predominantly acute anorexia. J. Clin. Invest. 86,2014–2024.

Tsujinaka, T., Fujita, J., Ebisui, C., Yano, M., Kominami, E., Suzuki, K., Tanaka, K., Katsume, A.,Ohsugi, Y., Shiozaki, H., Monden, M., 1996. Interleukin 6 receptor antibody inhibits muscle atrophyand modulates proteolytic systems in interleukin 6 transgenic mice. J. Clin. Invest. 97, 244–249.

Uehara, A., Sekiya, C., Takasugi, Y., Namiki, M., Arimura, A., 1989. Anorexia induced by interleukin 1:involvement of corticotropin-releasing factor. Amer. J. Physiol. 257, R613–R617.

van Dam, A.M., Brouns, M., Louisse, S., Berkenbosch, F., 1992. Appearance of interleukin-1 inmacrophages and in ramified microglia in the brain of endotoxin-treated rats: a pathway for the induc-tion of non-specific symptoms of sickness? Brain Res. 588, 291–296.

van Reeth, K., Nauwynck, H., 2000. Proinflammatory cytokines and viral respiratory disease in pigs.Vet. Res. 31, 187–213.

van Reeth, K., Labarque, G., Nauwynck, H., Pensaert, M., 1999. Differential production of proinflam-matory cytokines in the pig lung during different respiratory virus infections: correlations withpathogenicity. Res. Vet. Sci. 67, 47–52.

Vary, T.C., Kimball, S.R., 1992. Sepsis-induced changes in protein synthesis: differential effects onfast- and slow-twitch muscles. Amer. J. Physiol. 262, C1513–C1519.

Vary, T.C., Kimball, S.R., 2000. Effect of sepsis on eIE4E availability in skeletal muscle. Amer.J. Physiol. Endocrinol. Metab. 279, E1178–E1184.

Vary, T.C., Jefferson, L.S., Kimball, S.R., 2001. Insulin fails to stimulate muscle protein synthesis insepsis despite unimpaired signaling to 4E-BP1 and S6K1. Amer. J. Physiol. Endocrinol. Metab. 281,E1045–E1053.

Vary, T.C., Jurasinski, C.V., Karinch, A.M., Kimball, S.R., 1994. Regulation of eukaryotic initiationfactor-2 expression during sepsis. Amer. J. Physiol. 266, E193–E201.

Vary, T.C., Owens, E.L., Beers, J.K., Verner, K., Cooney, R.N., 1996. Sepsis inhibits synthesis ofmyofibrillar and sarcoplasmic proteins: modulation by interleukin-1 receptor antagonist. Shock6, 13–18.

Venters, H.D., Tang, Q., Liu, Q., VanHoy, R.W., Dantzer, R., Kelley, K.W., 1999. A new mechanism ofneurodegeneration: a proinflammatory cytokine inhibits receptor signaling by a survival peptide.Proc. Natl. Acad. Sci. USA 96, 9879–9884.

Voges, D., Zwickl, P., Baumeister, W., 1999. The 26S proteasome: a molecular machine designed forcontrolled proteolysis. Annu. Rev. Biochem. 68, 1015–1068.

Voisin, L., Breuille, D., Combaret, L., Pouyet, C., Taillandier, D., Aurousseau, E., Obled, C. Attaix, D.,1996a. Muscle wasting in a rat model of long-lasting sepsis results from the activation of lysosomal,Ca2+-activated, and ubiquitin-proteasome proteolytic pathways. J. Clin. Invest. 97, 1610–1617.

Voisin, L., Gray, K., Flowers, K.M., Kimball, S.R., Jefferson, L.S., Vary, T.C., 1996b. Altered expres-sion of eukaryotic initiation factor 2B in skeletal muscle during sepsis. Amer. J. Physiol. 270,E43–E50.

Webel, D.M., Finck, B.N., Baker, D.H., Johnson, R.W., 1997. Time course of increased plasma cytokines,cortisol, and urea nitrogen in pigs following intraperitoneal injection of lipopolysaccharide. J. Anim. Sci.75, 1514–1520.

Webel, D.M., Johnson, R.W., Baker, D.H., 1998. Lipopolysaccharide-induced reductions in food intakedo not decrease the efficiency of lysine and threonine utilization for protein accretion in chickens.J. Nutr. 128, 1760–1766.

Williams, A., Wang, J.J., Wang, L., Sun, X., Fischer, J.E., Hasselgren, P.O., 1998. Sepsis in mice stimu-lates muscle proteolysis in the absence of IL-6. Amer. J. Physiol. 275, R1983–R1991.

Williams, A.B., Decourten-Myers, G.M., Fischer, J.E., Luo, G., Sun, X., Hasselgren, P.O., 1999. Sepsisstimulates release of myofilaments in skeletal muscle by a calcium-dependent mechanism. FASEB J.13, 1435–1443.

Williams, N.H., Stahly, T.S., Zimmerman, D.R., 1997a. Effect of chronic immune system activation onbody nitrogen retention, partial efficiency of lysine utilization, and lysine needs of pigs. J. Anim. Sci.75, 2472–2480.

Williams, N.H., Stahly, T.S., Zimmerman, D.R., 1997b. Effect of chronic immune system activation onthe rate, efficiency, and composition of growth and lysine needs of pigs fed from 6 to 27 kg. J. Anim. Sci.75, 2463–2471.

Williams, N.H., Stahly, T.S., Zimmerman, D.R., 1997c. Effect of level of chronic immune system activationon the growth and dietary lysine needs of pigs fed from 6 to 112 kg. J. Anim. Sci. 75, 2481–2496.


Wright, K.J., Balaji, R., Hill, C.M., Dritz, S.S., Knoppel, E.L., Minton, J.E., 2000. Integrated adrenal,somatotropic, and immune responses of growing pigs to treatment with lipopolysaccharide. J. Anim.Sci. 78, 1892–1899.

Zamir, O., Hasselgren, P.O., Kunkel, S.L., Frederick, J., Higashiguchi, T., Fischer, J.E., 1992. Evidencethat tumor necrosis factor participates in the regulation of muscle proteolysis during sepsis. Arch.Surg. 127, 170–174.

Zhang, Y., Pilon, G., Marette, A., Baracos, V.E., 2000. Cytokines and endotoxin induce cytokine recep-tors in skeletal muscle. Amer. J. Physiol. Endocrinol. Metab. 279, E196–E205.


107

5 Amino acid metabolism in thesmall intestine: biochemical basesand nutritional significance1

G. Wua, D. A. Knabea, and N. E. Flynnb

aDepartment of Animal Science and Faculty of Nutrition,Texas A & M University, College Station, Texas, TX 77843-2471, USAbDepartment of Chemistry and Biochemistry, Angelo State University,San Angelo, TX 76909, USA

The small intestine is a highly differentiated and complex organ, which is not only responsi-ble for the terminal digestion and absorption of nutrients, but also plays an important role inamino acid metabolism. Most of glutamine and almost all of glutamate and aspartate in thediet are catabolized by the small intestinal mucosa in the first pass. The small intestinalmucosa also degrades enteral arginine, ornithine, proline, branched-chain amino acids andlysine, and perhaps enteral methionine, phenylalanine, threonine, glycine, and serine, suchthat 30−50% of these dietary amino acids do not enter the portal circulation. In the post-absorptive state, the small intestine actively takes up arterial glutamine and releases ammonia,alanine, citrulline, and proline as the major nitrogenous products. The intestine-derived citrullineis effectively utilized for arginine synthesis by extrahepatic cells and organs (e.g. the kidneys).This is of nutritional significance, particularly for suckling neonates because the milk of mostspecies, including the pig, cattle, sheep, rat, and human, is remarkably deficient in arginine.In addition to hepatic gluconeogenesis, the alanine released by the small intestine plays a keyrole in the extensive recycling of nitrogen between the liver and the gut. Because dietaryamino acids are major fuels for the small intestinal mucosa, and are essential precursors forintestinal synthesis of proteins, glutathione, polyamines, nitric oxide, purine, and pyrimidinenucleotides, intestinal amino acid metabolism is obligatory for maintaining intestinal mucosalmass, function, and integrity. However, the extensive catabolism of enteral amino acids by the



1We thank our students, technicians, and collaborators who have contributed to the work cited here. Our research onintestinal amino acid metabolism was supported, in part, by USDA National Research Initiative competitive grantsNo. 92-37206-8004, No. 94-37206-1100, No. 97-35206-5096, No. 2001-35203-11247 and No. 2003-35206-13694(GW), by Hatch projects No. 8200 (GW) and No. 6601 (DAK) from the Texas Agricultural Experiment Station, bygrants from the Houston Livestock Show and Rodeo (GW and DAK), and by a Texas A & M University FacultyFellowship (GW). This paper is dedicated to the memory of Dr. Peter J. Reeds, our dear friend and mentor.

small intestine substantially reduces their availability to extraintestinal tissues and selectivelyalters the patterns of amino acids that enter the systemic circulation. This has important practicalimplications for the utilization efficiency and recommended requirements of dietary proteinand amino acids by animals, including humans.

1. INTRODUCTION

The small intestine is a highly differentiated and complex organ, which is responsible for theterminal digestion and absorption of dietary nutrients and therefore is essential to health,growth, development, reproduction, and sustaining life of the organism (Madara, 1991).Enterocytes (epithelial absorptive cells of the small intestine) constitute >80% of the mucosalepithelial cell population (Cheng and Leblond, 1974; Klein and McKenzie, 1983) and havehigh rates of intracellular protein turnover and cell proliferation (Smith and Jarvis, 1978;Burrin and Reeds, 1997). Interestingly, the apical and basolateral membranes of each entero-cyte are chemically, biochemically, and physically distinct (Madara, 1991). Such a polarorganization of the enterocyte allows it to selectively receive nutrients from two sources: thearterial blood across its basolateral membrane and the intestinal lumen across its brush bordermembrane. This has important practical implications for choosing the route of feeding (e.g.enteral vs parenteral) for nutrient delivery to animals.

The gut is also the barrier separating the internal milieu of the organism from the externalenvironment, therefore excluding food-borne pathogens and preventing the translocation ofluminal microorganisms into the circulation. As the largest lymphoid organ in the body, thesmall intestine participates in immune surveillance of the intestinal epithelial layer and regu-lation of the mucosal response to foreign antigens (Mowat, 1987). The pioneering studies ofWindmueller and coworkers in the 1970s have demonstrated extensive intestinal catabolismof glutamine, glutamate, and aspartate (see Windmueller, 1982, for review). In recent years,there has been growing recognition that the small intestinal mucosa also degrades enteralarginine, ornithine, proline, branched-chain amino acids (BCAA), and lysine, and perhapsenteral methionine, phenylalanine, threonine, glycine, and serine, such that 30−50% of thesedietary amino acids do not enter the portal circulation (Wu, 1998a; Reeds and Burrin, 2001).The major objective of this chapter is to review recent work on intestinal amino acid metab-olism, with an emphasis on its biochemical bases and nutritional significance.

2. AMINO ACID METABOLISM IN THE SMALL INTESTINE

Amino acids can be classified into four major groups on the basis of their metabolic fates inthe small intestinal mucosa of rats, pigs, and ruminants (sheep and cattle): (1) amino acidsthat are neither synthesized nor degraded; (2) amino acids that are synthesized but notdegraded; (3) amino acids that are degraded but not synthesized; and (4) amino acids that areboth degraded and synthesized (table 1). Note that there are species differences in intestinalamino acid metabolism (Wu, 1998b).

2.1. Amino acids that are neither synthesized nor degraded by the intestinal mucosa

2.1.1. Asparagine

There is no synthesis of asparagine from glutamine and aspartate in enterocytes of rats, pigs(Wu et al., 1995b; Wu, 1998b), and ruminants (sheep and cattle) (Wu, unpublished data).

G. Wu et al.108

Asparagine is not degraded by the rat small intestine, because of the absence of asparaginase;thus in rats all of the asparagine absorbed by enterocytes from the intestinal lumen appears inthe intestinal venous blood intact (Windmueller, 1982). There is no release of asparagine bythe postabsorptive small intestine of pigs (Wu et al., 1994a) and rats (Brosnan et al., 1983),indicating the absence of asparagine synthesis by the gut. Interestingly, asparaginase activityis detectable in the canine small intestine, and asparagine is hydrolyzed to aspartate plusammonia in the guinea pig small intestine (Windmueller, 1982), indicating species differ-ences in intestinal asparagine metabolism. However, in both dogs and guinea pigs, intestinalmucosal asparagine catabolism is quantitatively low.

2.1.2. Cysteine, tryptophan, and histidine

Cysteine, tryptophan, and histidine are neither synthesized nor degraded by enterocytes ofrats, pigs, sheep, and cattle (Wu, unpublished data). However, cysteine is used for glutathionesynthesis in enterocytes (Reeds et al., 1997). Because there are mixed cell populations in thesmall intestine, the infiltrating mast cells can decarboxylate histidine to produce histamine inresponse to immunological activation (Wu, 1998b), thereby contributing to intestinal histidineutilization by intestinal mucosal cells and the portal-drained viscera (PDV).

2.2. Amino acids that are synthesized but not degraded by the intestinal mucosa

2.2.1. Tyrosine

Tyrosine is synthesized from phenylalanine by phenylalanine hydroxylase, a tetrahydro-biopterin-dependent enzyme. This enzyme is restricted primarily to the liver but is alsoexpressed in the kidney and pancreas (Tourian et al., 1969). Phenylalanine hydroxylase waspreviously found to be absent from the small intestine (Tourian et al., 1969). However, inthese earlier studies, protease inhibitors were not used for preparing intestinal extracts or theenzyme assay, and thus intestinal activity of phenylalanine hydroxylase should be reexam-ined. We have recently found that phenylalanine was converted into tyrosine in enterocytes ofpigs, rats, sheep, and cattle (Wu, unpublished data), indicating the presence of intestinalphenylalanine hydroxylation. In support of this view, there is significant output of tyrosine(167% of dietary intake) by the PDV of the milk protein-fed pig (Stoll et al., 1998) and oftyrosine (28%) by the small intestine of sheep fed a 20% crude protein diet (Tagari andBergman, 1978).

Amino acid metablism in the small intestine 109

Table 1

Metabolic fates of amino acid in the small intestine

Metabolic fates Amino acids

Neither synthesized nor degraded Asparagine, cysteine, histidine, tryptophanSynthesized but not degraded TyrosineDegraded but not synthesized Branched-chain amino acids (isoleucine, leucine, valine),

lysine, methionine, phenylalanine, threonineBoth degraded and synthesized Alanine, arginine, aspartate, citrulline, glutamate, glutamine,

glycine, ornithine, proline, serine

2.3. Amino acids that are degraded but not synthesized by the intestinal mucosa

2.3.1. BCAA

Studies from a number of species have documented BCAA catabolism by the small intestineof nonruminant animals. For example, 30% of the total ingested dietary leucine is extractedby the dog small intestine in the first pass, 55% of which enters the transamination reaction(Yu et al., 1990). In adult humans, 20−30% of enterally delivered leucine is taken up in thefirst pass within the splanchnic region (mainly the small intestine) during the postabsorptivestate or feeding (Hoerr et al., 1993). Similarly, in growing pigs fed a milk protein-based diet,40% of leucine, 30% of isoleucine, and 40% of valine in the diet are sequestered by the PDVin the first pass, and <20% of the sequestered BCAA are utilized for mucosal protein syn-thesis (Stoll et al., 1998). These results suggest substantial catabolism of dietary BCAA bythe small intestine of humans, dogs, and pigs. In more recent studies, we found that in pigenterocytes, most of the transaminated BCAA were released as branched-chain α-ketoacid(BCKA) (Wu, unpublished data), indicating a low rate of oxidative decarboxylation of BCKAby the small intestinal mucosa. In contrast to nonruminant animals, BCAA catabolism isnegligible in the small intestinal mucosa of fed or fasted sheep (Pell et al., 1986; Cappelliet al., 1997).

Both BCAA transaminase, which initiates BCAA degradation, and BCKA dehydrogenase,which oxidatively decarboxylates BCKA, are present in the nonruminant small intestinalmucosa (Khatra et al., 1977; Harper et al., 1984). The specific activities of BCAA transaminaseand BCKA dehydrogenase are relatively low in the small intestine, compared with skeletalmuscle and liver, respectively. However, this should not negate a quantitatively important roleof the small intestine in BCAA catabolism in the whole animal, partly due to a relatively largemass of the gut. On the basis of the intestinal activities of BCAA transaminase and BCKAdehydrogenase, most of the BCKA produced by enterocytes likely enters the portal circulationand are then utilized by the liver for complete oxidation and/or gluconeogenesis.

2.3.2. Lysine, methionine, phenylalanine, and threonine

The degradation of these four amino acids has recently been demonstrated in the small intes-tine, such that ~50% of dietary lysine and methionine, 45% of dietary phenylalanine, and60% of dietary threonine are extracted in the first pass by the PDV of milk protein-fed pigs(Stoll et al., 1998). Only <20% of the extracted amino acids are utilized for mucosal proteinsynthesis, whereas one-third of the extracted amino acids are catabolized by the small intes-tinal mucosa such that intestinal metabolism dominates the splanchnic extraction of lysine,methionine, phenylalanine, and threonine in pigs (Stoll et al., 1997, 1998). Using stable iso-topes, van Goudoever et al. (2000) showed that intestinal oxidation of enteral lysinecontributed one-third of total body lysine oxidation in growing pigs fed a high-protein diet,but was virtually absent in pigs fed a low-protein diet. These results indicate adaptive regula-tion of intestinal lysine metabolism. In support of intestinal lysine oxidation, Pink et al.(2002) recently reported the production of CO2 from [14C]lysine in mitochondria of pig ente-rocytes. In adult humans, 30% and 58% of enterally delivered lysine and phenylalanine areextracted in the first pass, respectively, within the splanchnic bed (Biolo et al., 1992; Hoerret al., 1993). Remarkably, in sheep fed 16−20% crude protein diets, 30−40% of lysine,37−43% of phenylalanine, 55−73% of threonine, and 69−71% of methionine that disappearedfrom the small intestinal lumen did not enter the portal circulation (Tagari and Bergman,1978), suggesting extensive catabolism of these amino acids by the ovine small intestine.

G. Wu et al.110

There is evidence for the presence of key enzymes responsible for the degradation oflysine, methionine, phenylalanine, and threonine in the intestinal mucosa. For example, Pinket al. (2002) recently detected lysine α-ketoglutarate reductase activity in the pig small intes-tinal mucosa when the enzyme assay was conducted in the presence of protease inhibitors.Also, both methionine transamination (Mitchell and Benevenga, 1978) and the transsulfura-tion pathway (Luk et al., 1980) are present in the rat small intestinal mucosa. In addition,glutamine transaminase K, whose major substrates include glutamine, phenylalanine, andmethionine, is widespread in mammalian tissues, including the small intestine of rats (Cooperand Meister, 1977) and pigs (Wu, unpublished data). Furthermore, there is an inherent phenyl-alanine transaminase activity in porcine aspartate transaminase isoenzymes (Shrawder andMartinez-Carrion, 1972). More recently, we were able to detect threonine dehydrogenaseactivity in mitochondria of pig enterocytes when protease inhibitors were used for preparingcell extracts and the enzyme assay (Wu, unpublished data).

2.4. Amino acids that are both degraded and synthesized by the intestinal mucosa

2.4.1. Glutamine, glutamate, aspartate, and alanine

The pioneering studies of Windmueller and coworkers have clearly demonstrated that the smallintestine extensively catabolizes enteral glutamine, glutamate, and aspartate as well as arterialglutamine, and releases large amounts of alanine, citrulline, and proline (Windmueller, 1982). Forexample, the small intestine of postabsorptive rats extracts 30% of arterial glutamine in a singlepass, which accounts for 30% of whole-body glutamine utilization (Windmueller and Spaeth,1975). Interestingly, intestinal utilization of arterial glutamine appears to be lower in ruminants,compared with nonruminants (Gate et al., 1999). Approximately 55%, 66%, and almost 100%of enterally delivered glutamine are sequestered in the first pass by the small intestine of adulthumans (Matthews et al., 1993), adult rats (Windmueller and Spaeth, 1975), and growing pigs(Stoll et al., 1998), respectively. In contrast, there is no significant uptake of arterial glutamateand aspartate by the small intestine. Remarkably, 98% and >99% of luminal glutamate andaspartate (6 mM) are catabolized in a single pass by the rat jejunum, respectively (Windmueller,1982). Similarly, 96% and 95% of enterally delivered glutamate is extracted in the first pass bythe human splanchnic bed (Battezzati et al., 1995) and by the porcine PDV (Reeds et al., 1996),respectively. Likewise, there is negligible appearance of intra-abomasum infused glutamate inthe portal circulation of sheep (Tagari and Bergman, 1978). Furthermore, the PDV of growing beefsteers extracts virtually all diet- and rumen-derived glutamate, glutamine, and aspartate (Lapierreet al., 2000). Thus, most of glutamine and almost all glutamate and aspartate in the diet do notenter portal circulation in both ruminant and nonruminant animals.

The metabolic fate of glutamine, glutamate, and aspartate has been quantified in the smallintestine. Ammonia, citrulline, alanine, and proline released by the rat jejunum account for38%, 28%, 24%, and 7% of the metabolized glutamine nitrogen, respectively. In postabsorp-tive pigs, the small intestine takes up arterial glutamine and releases not only ammonia,citrulline, alanine, and proline but also arginine, glutamate, and aspartate (Wu et al., 1994a).In vitro studies have also shown that pig enterocytes extensively utilize glutamine andproduce ammonia, glutamate, alanine, aspartate, CO2, ornithine, citrulline, arginine, andproline, and rates of glutamine catabolism are greater in cells from newborn pigs compared withsuckling and postweanling pigs (Wu et al., 1995b). Ammonia, glutamate, alanine, aspartate, CO2,ornithine, citrulline, arginine, and proline are also produced from glutamine in enterocytes ofsheep and cattle (Wu, unpublished data). Interestingly, there is no production of ammonia


from glutamate and aspartate in the small intestine (Windmueller, 1982), suggesting a dominantrole of transamination in their catabolism. In the rat small intestine, CO2, lactate, alanine, andglucose account for 56−64%, 16−20%, 4−8%, and 2−10% of the total catabolized carbons ofluminal glutamine, glutamate, and aspartate, respectively. Under conditions similar to a meal,oxidation of arterial glutamine, luminal glutamine plus glutamate plus aspartate, and luminalglucose contribute 38%, 39%, and 6% of the CO2 produced by the rat small intestine, respec-tively (Windmueller, 1982). Similarly, the oxidation of enteral glutamate accounts for 36% ofthe total CO2 production by the PDV of fully fed growing pigs, and is the most important singlecontributor to mucosal oxidative ATP production (Stoll et al., 1999b). Approximately 80% and65% of enterally delivered glutamate is oxidized to CO2 in the first pass by the human splanch-nic bed (Battezzati et al., 1995) and by the porcine PDV (Reeds et al., 1996), respectively,indicating that oxidation dominates intestinal glutamate metabolism. These results demonstratethat amino acids, rather than glucose, are major fuels for the small intestinal mucosa.

Syntheses of purine and pyrimidine nucleotides and of glutathione represent physiologicallyimportant pathways for intestinal utilization of glutamine and aspartate and of glutamate,respectively (Burrin and Reeds, 1997; Wu, 1998b). Reeds et al. (1997) have shown that lumi-nal glutamate, rather than the glutamate derived from intracellular glutamine degradation, isthe preferential source of glutamate for glutathione synthesis in the porcine intestinal mucosa,suggesting that intestinal glutamate utilization is highly compartmentalized.

Enzymes for intestinal glutamine degradation have been identified, which include phosphate-dependent glutaminase (PDG), carbamoylphosphate synthase II (glutamine) (CPS-II),glutamate-oxaloacetate transaminase (GOT), glutamate-pyruvate transaminase (GPT), pyrroline-5-carboxylate (P5C) synthase, ornithine aminotransferase (OAT), P5C reductase, ornithinecarbamoyltransferase (OCT), carbamoylphosphate synthase I (ammonia) (CPS-I), arginino-succinate synthase (ASS), argininosuccinate lyase (ASL), ornithine decarboxylase (ODC), andKrebs cycle enzymes (Wu and Morris, 1998; Bush et al., 2002; Morris, 2002). In animals, P5Csynthase is expressed primarily in enterocytes, indicating a unique role of the small intestine incitrulline production and thus endogenous synthesis of arginine. The intestinal mucosa also con-tains N-acetylglutamate (NAG) synthase, which synthesizes NAG (an allosteric activator ofCPS-I) from glutamate and acetyl-CoA (Wakabayashi et al., 1991). All of the glutamine-metabolicenzymes, except for CPS II, P5C reductase, ASS, and ASL (cytosolic enzymes), are located inmitochondria, whereas GPT and GOT are expressed in both the cytosol and mitochondria of theintestinal mucosa.

Glutamine can be synthesized from glutamate and ammonia by glutamine synthetase inavian and mammalian small intestines. For example, chick enterocytes are capable of syn-thesizing glutamine in the presence of glutamate and ammonia (Porteous, 1980), which mayexplain the net release of glutamine by the chick small intestine in vivo (Windmueller, 1982).The intestinal synthesis of glutamine, coupled with a low rate of intestinal glutamine degra-dation, helps explain a relatively high plasma concentration of free glutamine (1.1 mM) inchicks (Wu et al., 1995a). In addition, a small amount of [14C]glutamine (1.2% of infused[14C]glutamate) appeared in the sheep portal circulation when abomasum was infused with[14C]glutamate (Tagari and Bergman, 1978). Both immunological and in situ hybridizationstudies have shown that glutamine synthetase protein and mRNA are located primarily in theintestinal crypt (Roig et al., 1995). Using the IEC-6 cell line (a well-characterized rat smallintestinal epithelial cell line), DeMarco et al. (1999) have shown that an inhibition of glutaminesynthetase reduces cell proliferation. This result suggests that when extracellular glutamine isabsent, the cytosolic synthesis of glutamine from glutamate and NH+

4 may play a role inendogenous provision of glutamine for supporting DNA and protein synthesis. However, the

G. Wu et al.112

nutritional significance of such an observation is not entirely clear because the small intestineconstantly receives a supply of arterial glutamine. Moreover, the activity of glutamine syn-thetase in the small intestine is generally very low, compared with PDG (Wu et al., 1994b,1995b), and some of the glutamine synthesized in the cytosol likely enters mitochondria forextensive catabolism. Thus, although there may be an intracellular (between the cytosol andmitochondrion) or perhaps an intercellular glutamine–glutamate cycle (between crypt andvillus cells) in the small intestine, it is unlikely that net synthesis of glutamine in a nutrition-ally significant quantity occurs in the mammalian gut in vivo.

2.4.2. Arginine, citrulline, ornithine, and proline

These four closely related amino acids are interconverted in the small intestinal mucosa.Studies over the past 25 years have established that the production of citrulline by enterocytesof the small intestine plays a crucial role in the endogenous synthesis of arginine (Wu andMorris, 1998). Glutamine/glutamate and proline (abundant amino acids in milk) are the majorprecursors for citrulline and arginine synthesis in incubated enterocytes from pigs (Wu et al.,1994a; Wu, 1997), sheep, and cattle (Wu, unpublished data). In vivo studies have also demon-strated citrulline and arginine synthesis from enteral proline and glutamate in pigs (Murphyet al., 1996; Brunton et al., 1999) and the release of citrulline by the ruminant small intestine(Bergman and Heitmann, 1978). The citrulline released by the small intestine is not taken upby the liver, and is utilized for arginine synthesis primarily in kidneys (Dhanakoti et al.,1990). Also, uptake of physiological concentrations of arginine by the liver is low due to alow activity of the amino acid transport system y+ in hepatocytes (Wu and Morris, 1998).Importantly, almost all extrahepatic cells are capable of synthesizing arginine from citrulline(Wu and Morris, 1998). Thus, intestine-derived citrulline and arginine are equally effective assources of arginine for the whole body. The release of citrulline into the portal circulation bythe small intestine and the uptake of arterial citrulline by the kidneys for arginine productionis referred to as the intestinal–renal axis for endogenous synthesis of arginine.

Recent studies with pigs have demonstrated marked developmental changes in intestinalarginine metabolism. First, in 1- to 7-day-old pigs, most of the citrulline synthesized fromglutamine and proline in enterocytes is converted locally into arginine because of high activ-ities of both ASS and ASL (Wu and Knabe, 1995). However, in older piglets (14- to21-day-old), enterocytes release most of the synthesized citrulline due to a low ASS activity(Wu et al., 1994a). Thus, the small intestine shifts from the major site of net arginine synthe-sis in 1-week-old pigs to the major site of net citrulline synthesis in 2- to 3-week-old pigs.

Second, intestinal synthesis of citrulline and arginine decreases by 60−75% in 7-day-old suck-ling pigs compared with newborn pigs, and declines further in 14- to 21-day-old suckling pigs(Wu, 1997). The metabolic basis for the marked decrease in citrulline and arginine synthesis byenterocytes of 7- to 21-day-old pigs is not known. Because the ratios of small intestinal weight(or mucosal protein weight) to body weight do not change substantially in newborn and sucklingpiglets (28−32 g small intestine per kg body wt from 1 to 21 days of age), intestinal synthesis ofcitrulline and arginine per kg body weight remains strikingly low in 7- to 21-day-old piglets com-pared with 1- to 3-day-old piglets. Consequently, plasma concentrations of arginine and itsimmediate precursors (ornithine and citrulline) decrease progressively by 20−41%, as the age ofsow-reared piglets increases from 3 to 14 days (Flynn et al., 2000). It should be borne in mindthat plasma arginine concentration is regulated by a number of factors, including arginine syn-thesis and degradation, dietary arginine intake, degradation of plasma proteins and peptides, andintracellular protein turnover (protein synthesis and degradation) (Young and El-Khoury, 1995).


Thus, in suckling piglets, plasma arginine concentration would not be expected to decrease to thesame extent as the intestinal synthesis of citrulline and arginine.

In contrast to pigs, there are no postnatal decreases in plasma citrulline and arginine concen-trations or in rates of intestinal citrulline and arginine synthesis from glutamine between 2- and7-day-old suckling calves (Wu, unpublished data). For examples, rates of citrulline productionfrom 2 mM glutamine are 149 ±13 and 132 ± 16 nmol/mg DNA per 30 min (means ± SEM,n = 5), respectively, in jejunal enterocytes of 2- and 7-day-old suckling calves. In addition, ratesof arginine synthesis from 2 mM glutamine are 185 ± 20 and 169 ± 22 nmol/mg DNA per 30 min(means ± SEM, n = 5), respectively, in jejunal enterocytes of 2- and 7-day-old suckling calves.The marked postnatal decline in intestinal citrulline and arginine synthesis represents an intrigu-ing, and perhaps unique, aspect of amino acid metabolism in neonatal pigs.

Third, arginine catabolism in pig enterocytes is limited at birth and during the sucklingperiod due to a negligible arginase activity, but is markedly enhanced at weaning (Wu et al.,1996a) due to the cortisol-mediated induction of arginase (Flynn and Wu, 1997a,b).Consequently, urea is synthesized from extracellular and intramitochondrially generatedammonia and from arginine in enterocytes of weaned pigs (Wu, 1995). This novel findingchallenges the textbook view that ureagenesis occurs only in the mammalian liver. The induc-tion of arginase also makes possible the synthesis of proline and polyamines from arginine inenterocytes of postweaning pigs (Wu, 1995; Wu et al., 2000a,b). Because of a relatively highactivity of arginase in the small intestinal mucosa of postweaning mammals, 40% of thearginine absorbed by enterocytes from the intestinal lumen is degraded in a single pass in adultrats (Windmueller and Spaeth, 1976). Similarly, in adult humans, 38% of dietary arginine isremoved in the first pass within the splanchnic region, and most of the arginine uptake isaccounted for by the small intestine (Castillo et al., 1993a). In sheep fed 16−20% crude proteindiets, 42−68% of arginine that disappears from the small intestinal lumen does not enter theportal circulation (Tagari and Bergman, 1978).

In addition to ornithine production, there is nitric oxide (NO) synthesis from arginine byNO synthase in enterocytes (Wu et al., 1996a) or the small intestine (Alican and Kubes,1996). This pathway, however, is quantitatively low in enterocytes of newborn, suckling, andweaned pigs (Wu et al., 1996a). Similarly, in healthy adult humans, only 0.34% of the dietaryarginine taken up in the first pass within the splanchnic region is utilized for NO production(Castillo et al., 1993b). Despite the presence of arginine decarboxylase in a number of animaltissues (e.g., brain, liver, and kidney) for agmatine synthesis from arginine, this enzyme isabsent from pig enterocytes (Wu et al., 1996a).

The findings that proline is actively catabolized by porcine enterocytes to produce ornithine,citrulline, arginine, and polyamines (Wu, 1997; Wu et al., 2000a,b) are some of the most excit-ing developments in intestinal amino acid metabolism in recent years. In pigs, the activities ofproline oxidase and OAT are greatest in the small intestinal mucosa (Wu, 1997). Proline oxidaseactivity is also present in enterocyte mitochondria of rats (Wu, 1997), sheep, and cattle (Wu,unpublished data). Considering the relatively large mass of the small intestine as compared withthe liver and kidneys, the intestinal mucosa likely plays a major role in initiating proline degra-dation in the body. Consistent with this suggestion, Stoll et al. (1998) demonstrated that 38% ofdietary proline was extracted in the first pass by the PDV of milk protein-fed piglets. In sheepfed 16−20% crude protein diets, 54−71% of proline that disappeared from the small intestinallumen did not enter the portal circulation (Tagari and Bergman, 1978).

By bridging the urea cycle with the Krebs cycle, arginine and proline metabolism,and polyamine synthesis, OAT plays a central role in intestinal nitrogen and carbonmetabolism (fig. 1). We have recently suggested that the intestinal OAT reaction proceeds in

G. Wu et al.114

Fig

. 1.

Inte

stin

al m

ucos

al a

min

o ac

id m

etab

olis

m. E

nzym

es th

at c

atal

yze

the

indi

cate

d re

actio

ns a

re: (

1) p

hosp

hate

-dep

ende

nt g

luta

min

ase;

(2)

glu

tam

ine

synt

heta

se; (

3) p

yrro

line-

5-ca

rbox

ylat

e sy

ntha

se; (

4) o

rnith

ine

amin

otra

nsfe

rase

; (5)

orn

ithin

e ca

rbam

oyltr

ansf

eras

e; (

6) a

rgin

inos

ucci

nate

syn

thas

e; (

7) a

rgin

inos

ucci

nate

lyas

e; (

8) a

rgin

ase;

(9)

orn

ithin

e de

carb

oxy-

lase

; (10

) sp

erm

idin

e sy

ntha

se; (

11)

sper

min

e sy

ntha

se; (

12)

S-ad

enos

ylm

ethi

onin

e sy

ntha

se; (

13)

S-ad

enos

ylm

ethi

onin

e de

carb

oxyl

ase;

(14

) ni

tric

oxi

de s

ynth

ase;

(15

) N

-ace

tylg

luta

mat

esy

ntha

se; (

16)

carb

amoy

lpho

spha

te s

ynth

ase-

I ; (

17)

prol

ine

oxid

ase;

(18

) py

rrol

ine-

5-ca

rbox

ylat

e re

duct

ase;

(19

) br

anch

ed-c

hain

am

ino

acid

tran

sam

inas

e; (

20)

bran

ched

-cha

in α

-ket

oaci

dde

hydr

ogen

ase;

(21

) al

anin

e tr

ansa

min

ase;

(22

) as

part

ate

tran

sam

inas

e; (

23)

puri

ne-

and

pyri

mid

ine-

synt

hesi

zing

enz

ymes

; (24

) as

part

ate

deca

rbox

ylas

e; (

25)

hist

idin

e de

carb

oxyl

ase;

(26)

α-k

etog

luta

rate

deh

ydro

gena

se; (

27)

aspa

ragi

nase

; (28

) vi

a as

part

ate

tran

sam

inas

e an

d K

rebs

cyc

le e

nzym

es; (

29)

poss

ibly

via

NA

DP-

linke

d m

alic

enz

yme,

pho

spho

enol

pyru

vate

carb

oxyk

inas

e/py

ruva

te k

inas

e, a

nd o

xalo

acet

ate

deca

rbox

ylas

e; (

30)

pyru

vate

deh

ydro

gena

se; (

31)

via

Kre

bs c

ycle

enz

ymes

; (32

) ph

enyl

alan

ine

hydr

oxyl

ase;

(33

) ph

enyl

alan

ine

tran

sam

inas

e;(3

4) ly

sine

: α-k

etog

luta

rate

red

ucta

se; (

35)

seri

ne tr

ansa

min

ase;

(36

) se

rine

deh

ydra

tase

; (37

) vi

a en

zym

es o

f gl

ucon

eoge

nesi

s; (

38)

seri

ne h

ydro

xym

ethy

ltran

sfer

ase;

(39

) gl

utat

hion

e-sy

nthe

sizi

ng e

nzym

es; a

nd (

40)

glyc

ine

clea

vage

sys

tem

. The

sym

bol “

?” d

enot

es u

nkno

wn

reac

tions

in in

test

inal

muc

osa.

Abb

revi

atio

ns: A

la, a

lani

ne; A

sp, a

spar

tate

; BC

AA

, bra

nche

d-ch

ain

amin

o ac

ids;

BC

KA

, bra

nche

d-ch

ain

α-ke

toac

id; C

oA, c

oenz

yme

A; C

P, c

arba

moy

l pho

spha

te; D

CA

M, d

ecar

boxy

late

d S-

aden

osyl

met

hion

ine;

α-K

G, α

-ket

oglu

tara

te; M

et, m

ethi

onin

e;M

TA, m

ethy

lthio

aden

osin

e; M

TF,

N5 ,

N10

-met

hyle

nete

trah

ydro

fola

te; N

AG

, N-a

cety

lglu

tam

ate;

OA

A, o

xalo

acet

ate;

PPY

R, p

heny

lpyr

uvat

e; P

yr, p

yruv

ate;

TF,

tetr

ahyd

rofo

late

.

G. Wu et al.116

the direction of net synthesis of either ornithine or P5C, depending on intramitochondrial con-centrations of ornithine and P5C (Wu, 1998b; Dekaney et al., 2000). For example, when thereis a high concentration of P5C in mitochondria due to the oxidation of large amounts of prolineby proline oxidase, the synthesis of ornithine and therefore of citrulline from P5C is favored inenterocytes (Wu, 1997). On the other hand, when there is a high concentration of mitochondr-ial ornithine due to the hydrolysis of large amounts of arginine by arginase II, the synthesis ofP5C and therefore of proline from ornithine is favored in enterocytes (Wu et al., 1996a).

Arginine-metabolic enzymes have been identified in the small intestine, and arginase is themajor enzyme initiating arginine catabolism in enterocytes (Wu et al., 1996a). Both enzymolog-ical and metabolic evidence have established that P5C synthase is a key regulatory enzyme inintestinal synthesis of citrulline from glutamine/glutamate (Wakabayashi et al., 1991; Wu andMorris, 1998). By synthesizing NAG, NAG synthase may be another key regulatory enzyme inintestinal synthesis of citrulline and arginine from glutamine/glutamate and proline. Given thehigh activity of OCT in the small intestine, it is surprising that the major product of the degrada-tion of extracellular ornithine in enterocytes is proline but not citrulline (Wu et al., 1996a). Thismay be explained by the following reasons. First, enterocytes have an exceedingly high activityof mitochondrial OAT, but a relatively low activity of mitochondrial CPS I for yielding carbamoylphosphate (a substrate for OCT) from NH3, HCO3

−, and ATP. Thus, intramitochondrial ornithineis preferentially utilized by OAT to form P5C instead of citrulline. Second, enterocytes have ahigh activity of cytosolic P5C reductase. Therefore, when P5C enters the cytosol from mito-chondria, P5C is readily converted into proline by P5C reductase. Thus, dietary or arterial bloodornithine is a poor precursor for intestinal synthesis of citrulline, and does not contribute signif-icantly to maintaining arginine homeostasis in animals (Wu and Morris, 1998).

There are marked species differences in intestinal citrulline and proline synthesis, both of whichare NADPH-dependent (Wu, 1996). For example, P5C synthase is absent and OAT activity is verylow in chick enterocytes (Wu et al., 1995a). This provides the metabolic basis for explaining theabsence of citrulline synthesis from glutamate and glutamine in the avian small intestine and thusthe nutritional essentiality of arginine for birds. Similarly, the near absence of P5C synthase in theintestinal mucosa of the cat explains the lack of intestinal production of citrulline and thus littleendogenous synthesis of arginine in this species (Rogers and Phang, 1985). Strikingly, the rate ofconversion of arginine or ornithine into proline in chick enterocytes is only about 4% of that in pigenterocytes owing to low activities of both arginase and OAT (Wu et al., 1995a). These data helpexplain why ornithine is an ineffective replacement of proline in the chicken diet and why prolineis an essential amino acid for the chick (Graber and Baker, 1971, 1973).

2.4.3. Serine and glycine

Although the small intestine was not traditionally considered as a major organ for the catab-olism of serine and glycine, Stoll et al. (1998) have recently reported that 40% and 50% ofdietary serine and glycine are extracted in the first pass by the PDV of the milk protein-fedpig, respectively. Interestingly, <20% of the extracted serine and glycine are utilized for intes-tinal protein synthesis, and the majority of the extracted amino acids are catabolized (Stollet al., 1998). In sheep fed a 20% crude protein diet, 28% of serine and 36% of glycinethat disappear from the small intestinal lumen do not enter the portal circulation (Tagari andBergman, 1978). These results suggest that dietary serine and glycine may be substantiallycatabolized by the small intestine of both monogastric animals and ruminants.

There is some published evidence for the presence of key enzymes for intestinal metabo-lism of serine and glycine. For example, the rat small intestine has been reported to contain

the activities of serine dehydratase, serine aminotransferase, and serine hydroxymethyltrans-ferase (Kikuchi et al., 1980). The latter interconverts serine into glycine and generatesN5,N10-methylenetetrahydrofolate for purine and pyrimidine synthesis. This reaction isimportant for supporting the high rates of protein synthesis and epithelial cell proliferation inthe small intestinal mucosa. Glutathione synthesis also represents a physiologically importantpathway for glycine utilization by the small intestinal mucosa (Reeds et al., 1997). Furthermore,the transsulfuration pathway occurs in the small intestinal mucosa (Luk et al., 1980), which cancontribute to serine metabolism. Although previous studies could not show the presence of theglycine cleavage system in the rat small intestine (Kikuchi et al., 1980), likely due to the lackof protease inhibitors in the enzyme assay system, we recently detected the activity of theglycine cleavage system for the tetrahydrofolate-dependent production of ammonia and CO2

from glycine in enterocytes of rats, pigs, sheep, and cattle (Wu, unpublished data).

3. SIGNIFICANCE OF INTESTINAL AMINO ACID METABOLISM

3.1. Intestinal integrity and function

A characteristic of the small intestine physiology is high rates of intracellular protein turnoverand protein secretion by mucosal epithelial cells (Burrin and Reeds, 1997). Although proteinin the small intestine accounts for only 5% of whole-body protein, the amount of protein syn-thesized by the small intestine daily contributes 15−20% of the whole-body protein synthesis(McNurlan and Garlick, 1980; Reeds et al., 1993). This necessitates an adequate supply ofboth amino acids and energy to the small intestinal mucosa. In support of this notion, theavailable evidence shows that enteral feeding is the primary source of amino acids for theintestinal mucosa because uptake of amino acids other than glutamine from arterial blood iseither low or insignificant (Windmueller, 1982; Wu et al., 1994a).

Amino acid metabolism is crucial for intestinal integrity and function through the followingmechanisms. First, dietary glutamine, glutamate and aspartate, and arterial blood glutamine aremajor fuels for the small intestinal mucosa, and are responsible for providing energy required forintestinal ATP-dependent metabolic processes, including active nutrient transport, intracellularprotein turnover, as well as epithelial cell proliferation and migration (Burrin and Reeds, 1997;van der Schoor et al., 2001). On the basis of both experimental and clinical evidence, the impor-tance of glutamine for supporting the metabolic function of intestinal mucosa has now generallybeen accepted (Reeds and Burrin, 2001). Second, ornithine (a product of arginine, glutamine, andproline metabolism) is the immediate precursor for the synthesis of polyamines in the enterocyte(Wu et al., 2000a,b), which are essential to DNA and protein synthesis, as well as to the prolif-eration, differentiation, and repair of intestinal epithelial cells (Luk et al., 1980; Wu, 1998b). Inaddition, glutamine, asparagine, and glycine are potent stimulators of intestinal ODC (Kandil et al., 1995), thereby enhancing polyamine synthesis from arginine- and proline-derived ornithine.Third, arginine is the physiological precursor of NO, which plays an important role in regulatingintestinal blood flow, integrity, secretion, and epithelial cell migration (Alican and Kubes, 1996),the relaxation of gastrointestinal smooth muscle cells, and feed intake (Morley and Flood, 1991).Fourth, glutamate, glycine, and cysteine are precursors for the synthesis of glutathione, a tripep-tide critical for defending the intestinal mucosa against toxic and oxidative damage (Reeds et al.,1997). Because there is compartmentation of glutathione synthesis in enterocytes, and becausethere is no significant uptake of arterial glutamate, glycine, and cysteine by the small intestine,adequate dietary supply of these amino acids plays a vital role in glutathione synthesis by theintestinal mucosa (Reeds et al., 2000a; Reeds and Burrin, 2001). Fifth, villus enterocytes deriveamino acids for protein synthesis and cell proliferation preferentially from the intestinal lumen


rather than from the arterial blood (Alpers, 1972; Stoll et al., 1999a, 2000; Burrin et al., 2000).This helps explain why parenteral nutrition selectively decreases protein synthesis in the smallintestinal mucosa and results in intestinal atrophy (Dudley et al., 1998; Stoll et al., 2000), and fur-ther supports the view that enteral feeding of amino acids is obligatory for maintaining intestinalmucosal mass and integrity (Wu, 1998a; Reeds et al., 2000a,b).

3.2. Endogenous synthesis of amino acids

3.2.1. Citrulline and arginine

Although arginine is formed via the urea cycle in the mammalian liver, there is no net syn-thesis of arginine by this organ due to an exceedingly high activity of arginase in hepatocytes(Wu and Morris, 1998). A unique, important aspect of intestinal amino acid metabolism is thesynthesis of citrulline and arginine from glutamate, glutamine, and proline in ruminant andnonruminant animals (Bergman and Heitmann, 1978; Wu, 1998a). In both neonates andadults, enterocytes are almost the exclusive source of citrulline for endogenous synthesis ofarginine. The near absence of arginase from neonatal enterocytes helps maximize the outputof arginine by the small intestine. This is of nutritional significance because arginine isremarkably deficient in the milk of most mammals studied, including primates (human, chim-panzee, gorilla, and rhesus), ruminants (cow, goat, sheep), and other nonprimates (elephant,llama, pig, and rat) (Davis et al., 1994; Wu and Knabe, 1994; Reeds et al., 2000a), owing toextensive catabolism of arginine by lactating mammary tissue (O’Quinn et al., 2002). Forexample, concentrations of amino acids in sow’s whole milk (containing 18.6% dry matter)on day 7 to day 28 of lactation are as follows (g/L): alanine, 1.89; arginine, 1.43; aspartate plusasparagine, 5.02; cysteine, 0.72; glutamate plus glutamine, 9.42; glycine, 1.16; histidine, 0.92;isoleucine, 2.12; leucine, 4.23; lysine, 3.87; methionine, 0.98; phenylalanine, 1.95; proline, 5.56;serine, 2.30; threonine, 2.08; tryptophan, 0.64; tyrosine, 1.92; and valine, 2.40 (N = 10 sows;Wu and Knabe, unpublished data). Note that the ratio of arginine/lysine in the sow’s milk (0.37)is only 40% of that in the pig body (0.92) (Davis et al., 1993).

Arginine requirements by young mammals are particularly high (Rogers et al., 1970; Wuet al., 2000a). The relative contribution of milk vs endogenous synthesis to meeting argininerequirements by the suckling neonate can be estimated on the basis of arginine intake andarginine accretion plus catabolism in the body. For example, for a 7-day-old pig (2.5 kg)which gains 200 g body weight (27.2 g crude protein or 1.88 g arginine) per day (Flynn et al.,2000), catabolizes 0.65 g arginine daily via arginase and NOS pathways (Murch et al., 1996),and utilizes 0.17 g arginine daily for creatine synthesis [calculated on the basis of urinary cre-atinine excretion (0.38 mmol/kg body wt/day)] (Weiler et al., 1997), the arginine requirementis at least 2.7 g/day (table 2). On the basis of milk consumption by the suckling 7-day-old pig(0.78 L milk/day), arginine content in sow’s milk (1.43 g/L of whole milk) (Wu and Knabe,1994), and digestibility of arginine in milk protein (90.4%) (Mavromichalis et al., 2001), theintake of sow’s milk provides only 1.01 g arginine/day, or at most 37% of the daily argininerequirement (table 2). Thus, endogenous synthesis of arginine must provide at least 63% ofarginine for the suckling piglet.

The crucial role of the small intestine in endogenous arginine synthesis is further supported bythe following lines of direct evidence. First, an inhibition of intestinal synthesis of citrullineretards the growth of young rats fed an arginine-deficient diet (Hoogenraad et al., 1985). Second,resection of the rat small intestine (removal of 80% of the gut) results in (1) marked deficienciesof citrulline and arginine in plasma and of arginine in skeletal muscle, (2) reduced animalgrowth, (3) negative nitrogen balance, and (4) hypertension due to a deficiency of arginine for

G. Wu et al.118

NO synthesis (Wakabayashi et al., 1994a,b). Third, hypoornithinemia, hypocitrullinemia,hypoargininemia, and hyperammonemia occur in patients with short bowel syndrome andchronic renal failure (Yokoyama et al., 1996), due to reduced intestinal production of citrullineand impaired synthesis of arginine from citrulline in kidneys. Fourth, a deficiency of intestinalOAT results in hypoornithinemia, hypocitrullinemia, hypoargininemia, hyperammonemia, andeven death in mice and humans during the neonatal period (Wang et al., 1995). Fifth, a recentlyrecognized deficiency of intestinal P5C synthase in humans causes hypoornithinemia, hypo-citrullinemia, hypoargininemia, and hyperammonemia, retarded mental development, and death(Kamoun et al., 1998). Sixth, in the sparse-fur mutant mouse, the inborn X-linked deficiencyof OCT limits intestinal citrulline synthesis, leads to impaired maturation of intestinal epithe-lial cells (Malo et al., 1986), and causes retarded postnatal growth and death (DeMars et al.,1976). Finally, an inhibition of intestinal citrulline synthesis for 12 h results in decreasedplasma concentrations of ornithine, citrulline, and arginine by 59%, 52%, and 76%, respec-tively, in 4-day-old neonatal pigs nursed by sows (Flynn and Wu, 1996).

On the basis of decreases in plasma concentrations of arginine, ornithine, and citrulline aswell as nitrite and nitrate (stable end products of NO oxidation), and a concomitant increasein plasma ammonia concentration, we have suggested that arginine is deficient in 7- to 21-day-old sow-reared piglets (Flynn et al., 2000). Although sow-reared piglets continue togrow during the 21-day suckling period, this does not necessarily mean that arginine supplyfrom milk plus endogenous synthesis meets arginine requirements for maximal growth, asexemplified by submaximal growth and impaired NO synthesis in arginine-deficient youngrats (Wu et al., 1999). Indeed, recent artificial rearing data show that the biological potentialfor neonatal pig growth (from birth to day 21 of age) is at least 74% greater than that for sow-reared piglets (Boyd et al., 1995). Both metabolic and growth data indicate that argininedeficiency represents a major obstacle to maximal growth in milk-fed piglets (Kim et al.,2004). In view of reduced arginine supply from endogenous synthesis in suckling comparedwith newborn piglets and a great potential of arginine to enhance neonatal pig growth, it is ofcrucial importance to identify an effective means for enhancing intestinal synthesis of citrulline, thereby improving arginine nutrition and growth of sow-reared piglets.

Endogenous synthesis of arginine also plays an important role in maintaining argininehomeostasis in postweaning growing animals. In 75-day-old pigs fed a conventional diet con-taining 0.98% arginine (2.5 times the recommended National Research Council requirement fordietary arginine), an inhibition of intestinal citrulline synthesis decreases plasma concentrations


Table 2

Relative contributions of milk vs endogenous synthesis of arginine to argininerequirements by the 7-day-old (2.5 kg) sow-reared piglet

Arginine requirements and sources Amounts of arginine (g/day)

Arginine requirements ≥2.7Body weight gain (200 g/day; 27.2 g crude protein)a 1.88Arginine catabolism via arginase and NO synthaseb 0.65Creatine synthesis (0.38 mmol/kg body wt/day)c 0.17

Arginine supply from sow’s milk ≤1.01Milk consumption (0.78 L/day; 1.43 g/L whole milk)d 1.12Undigestible arginine in sow’s milk (9.6%)e −0.11

Arginine supply from endogenous synthesis ≥1.69

a Wu et al. (2000c), b Murch et al. (1996), c Weiler et al. (1997), d Wu and Knabe (1994), e Mavromichalis et al. (2001).

of citrulline and arginine by 26% and 22%, respectively (Wu et al., 1997). On the basis ofdietary arginine intake, ileal digestibility of dietary arginine, tissue protein accretion, andoxidation of plasma arginine, we have estimated that endogenous synthesis of arginine provides~50% of the total daily arginine requirement in the postweaning growing pig (Wu et al., 1997).

3.2.2. Proline

Proline is synthesized from arginine, ornithine, glutamine, and glutamate in enterocytes ofmost mammals (Wu, 1998a). The finding that there is no synthesis of proline from intra-venously infused glutamate in both pigs and humans (Matthews et al., 1993; Murphy et al.,1996) not only suggests the complex compartmentation of intestinal glutamate metabolism andproline synthesis, but also underscores the essential role for the small intestine in synthesizingproline from enteral glutamate. The dietary essentiality of proline for birds results from (1) alow rate of endogenous synthesis of proline from arginine in the small intestine because of alow activity of intestinal OAT in birds, and (2) the lack of synthesis of proline from glutamateor glutamine in the small intestine because of the absence of P5C synthase (Wu et al., 1995a).The synthesis of proline from arginine and glutamine is low in enterocytes of suckling pigsbecause of a negligible activity of arginase and P5C synthase, but is markedly increased incells from postweaning pigs owing to the induction of both enzymes (Wu et al., 1996a). Thismay also explain, in part, (1) why proline is an essential amino acid for neonatal pigs (2.5 kgof body weight) (Ball et al., 1986), and (2) why proline is a nonessential amino acid for post-weaning pigs (5−15 kg of body weight) (Chung and Baker, 1993).

3.2.3. Alanine

In nonruminant animals, alanine is an important nitrogenous product of the intestinal catabolismof glutamate, aspartate, and BCAA (Windmueller, 1982; Brosnan et al., 1983), and its carbonskeleton (pyruvate) is derived partially from enteral glucose (Stoll et al., 1999b). Large amountsof alanine are also released by the ruminant small intestine (Bergman and Heitmann, 1978).Thus, alanine transaminase functions primarily for alanine synthesis in the small intestinalmucosa of both monogastric animals and ruminants. Because alanine is a major amino acid forhepatic gluconeogenesis, the intestine-derived alanine plays an important role in maintaining glu-cose homeostasis. Alanine released by the small intestine also helps transport the nitrogen andperhaps the carbons of dietary amino acids and arterial glutamine from the small intestine to theliver. Because the liver actively takes up alanine and ammonia from portal and arterial blood andreleases glutamine, and because the small intestine substantially utilizes enteral and arterial bloodglutamine and releases large amounts of alanine and ammonia during both the postabsorptivestate and feeding, there appears to be a “glutamine–alanine cycle” involving the small intestineand the liver. This cycle seems to be analogous to the glucose–lactate cycle (the Cori cycle) whichspans the skeletal muscle and liver. However, it should be pointed out that the carbons and nitro-gen of the alanine released by the small intestine are utilized preferentially for the synthesis ofglucose and urea, respectively, rather than for glutamine synthesis, in the liver. Thus, the splanch-nic “glutamine–alanine” cycle is indeed not a true metabolic cycle, but illustrates a key role ofalanine in the extensive recycling of nitrogen between the liver and the gut (Wu, 1998a).

3.3. Availability of dietary amino acids to extraintestinal tissues

A theme that has emerged from this review is that intestinal mucosal amino acid catabo-lism plays an important role in regulating the availability of dietary amino acids to

G. Wu et al.120

extraintestinal tissues. This novel concept has important implications for protein and aminoacid nutrition in animals. First, the extensive catabolism of dietary essential amino acids bythe small intestine results in a decrease in their nutritional efficiency. For example, the par-ticularly high requirement of dietary arginine by the postweaning growing animal results, inpart, from the extensive hydrolysis of the absorbed dietary arginine by enterocytes (Wu et al.,1997). Also, the extensive degradation of dietary glutamine limits the role for enteral gluta-mine feeding to increase its plasma concentrations in many monogastric mammals, includingthe rat and pig (Wu et al., 1996b; Wu, 1998b). In addition, infection with Trichostrongylus col-ubriformis in 6- to 9-month-old lambs increases the catabolism of luminal leucine by thegastrointestinal tract and reduces the availability of diet- and rumen-derived leucine for othertissues, which provides a metabolic basis for the decreases in nitrogen retention and growthrates under conditions of subclinical nematode infection (Yu et al., 2000). In support of theconcept of first-pass intestinal catabolism of essential amino acids, recent in vivo studies haveshown that methionine and threonine requirements are 35% and 45% higher, respectively,during oral compared with parenteral feeding in neonatal pigs (Bertolo et al., 1998; Shovelleret al., 2000). Most recently, Elango et al. (2002) reported that the parenteral requirement fortotal BCAA is only 56% of the enteral requirement in neonatal pigs, indicating that 44% oftotal BCAA is extracted by the first-pass metabolism in the gut.

There is a positive correlation between first-pass intestinal catabolism of dietary aminoacids and mucosal mass (Stoll et al., 1998). Thus, factors that affect intestinal mass (e.g.antibiotics, growth hormone, insulin-like growth factor-I, and diabetes) may have an impor-tant impact on the requirements of dietary amino acids (Wu, 1998b). For example, in pigs fedantimicrobial agents (antibiotics and chemotherapeutics), a decrease in the small intestinalmucosal mass is associated with an increase in whole-body growth rate (Yen et al., 1985;Cromwell, 2001). This raises a possibility that reduced catabolism of dietary amino acids by thesmall intestine is a mechanism responsible for the growth-promoting effect of antimicrobialagents.

Second, intestinal amino acid metabolism modulates the entry of absorbed dietary aminoacids into portal circulation. Therefore, the pattern of amino acids in the diet differs remark-ably from that in the intestinal tissue, portal circulation, and the extraintestinal organs (LeFloc’h and Seve, 2000; Daenzer et al., 2001). Also, the pattern of amino acids in tissue pro-teins or animal products (e.g. egg, milk, wool, and meat) is not necessarily similar to the idealpattern of dietary amino acid requirements in animals. In addition, there are profound differ-ences in organ or plasma amino acid concentrations between enteral and parenteral feedingin neonates, including infants and piglets (Bertolo et al., 2000; Wu et al., 2001).

Third, there are developmental changes, disease-associated alterations, and species differ-ences in intestinal amino acid catabolism. Thus, these factors should be taken intoconsideration in recommending dietary amino acid requirements and in refining in vivomodels of amino acid and protein metabolism. This is graphically demonstrated, for example,by the dynamic changes in dietary requirements of arginine and proline by developmental pigs(Wu, 1998b; Wu et al., 2000c), by increased requirements of arginine and proline for woundhealing in burned patients (Young and El-Khoury, 1995), and by the nutritional essentialityof arginine and proline for birds (Wu and Morris, 1998).


Much has been learned in recent years regarding intestinal amino acid metabolism. However,there are many important and yet challenging questions in this rapidly growing and fruitful


area of investigation. First, recent intriguing findings of the extensive first-pass extraction ofdietary essential amino acids and cysteine by the porcine PDV and human splanchnic bed raisean important question regarding their catabolism in the small intestinal mucosa. This novelconcept should be firmly established by biochemical studies with isolated enterocytes. In addi-tion, microorganisms in the intestinal lumen may substantially contribute to the catabolism ofenteral essential amino acids and cysteine and such microbial pathways should be quantified.Second, in view of the recently recognized deficiency of arginine (an essential amino acid forneonates) in sow-reared piglets (Flynn et al., 2000), further studies are necessary to elucidatethe mechanisms responsible for the marked decline in intestinal synthesis of citrulline and argi-nine in suckling piglets. This new knowledge will undoubtedly help design new, effectivemeans to enhance arginine supply to the piglets and therefore improve their postnatal growth.Third, much work is required to define hormonal and nutritional regulation of intestinal aminoacid metabolism at molecular, cellular, and whole-body levels. This will be facilitated by therecent availability of porcine small intestinal epithelial cells (Lu et al., 2002) and of mam-malian cDNAs for key regulatory enzymes [e.g. arginase II (Morris, 2002), P5C synthase (Aralet al., 1996), and NAG synthase (Caldovic et al., 2002)], and by recent biochemical and molec-ular biology techniques (e.g. proteomics, metabolomics, and microarrays) (Phelps et al., 2002).We predict that exciting new knowledge on the regulation of intestinal amino acid metabolismwill be discovered in the coming years, which will help design new means to improve the effi-ciency of protein and amino acid utilization by animals, including humans.

REFERENCES

Alican, I., Kubes, P., 1996. A critical role for nitric oxide in intestinal barrier function and dysfunction.Amer. J. Physiol. 270, G225−G237.

Alpers, D.H., 1972. Protein synthesis in intestinal mucosal. J. Clin. Invest. 51, 167−173.Aral, B., Schlenzig, J.-S., Liu, G., Kamoun, P., 1996. Database cloning human Δ1-pyrroline-5-carboxylate

synthetase (P5CS) cDNA: a bifunctional enzyme catalyzing the first 2 steps in proline biosynthesis.C.R. Acad. Sci. Paris 319, 171−178.

Ball, R.O., Atkinson, J.L., Bayley, S.H., 1986. Proline as an essential amino acid for the young pig. Brit.J. Nutr. 55, 659−668.

Battezzati, A., Brillon, D.J., Matthews, D.E., 1995. Oxidation of glutamic acid by the splanchnic bed inhumans. Amer. J. Physiol. 269, E269−E276.

Bergman, E.N., Heitmann, R.N., 1978. Metabolism of amino acids by the gut, liver, kidneys, and periph-eral tissues. Fed. Proc. 37, 1228−1232.

Bertolo, R.F.P., Chen, C.Z.L., Law, G., Pencharz, P.B., Ball, R.O., 1998. Threonine requirement of neonatalpiglets receiving total parenteral nutrition is considerably lower than that of piglets receiving an iden-tical diet intragastrically. J. Nutr. 128, 1752−1759.

Bertolo, R.F.P., Pencharz, P.B., Ball, R.O., 2000. Organ and plasma amino acid concentrations areprofoundly different in piglets fed identical diets via gastric, central venous or portal venous route.J. Nutr. 130, 1261−1266.

Biolo, G., Tessari, P., Inchiostro, S., Bruttomesso, D., Fongher, C., Sabadin, L., Fratton, M.G., Valerio, A.,Tiengo, A., 1992. Leucine and phenylalanine kinetics during mixed meal ingestion: a multiple tracerapproach. Amer. J. Physiol. 262, E455−E463.

Boyd, R.D., Kensinger, R.S., Harrell, R.J., Bauman, D.E., 1995. Nutrient uptake and endocrine regula-tion of milk synthesis by mammary tissue of lactating sows. J. Anim. Sci. 73, Suppl. 2, 36−56.

Brosnan, J.T., Man, K.C., Hall, D.E., Colbourne, S.A., Brosnan, M.E., 1983. Interorgan metabolism ofamino acids in streptozotocin-diabetic ketoacidotic rat. Amer. J. Physiol. 244, E151−E158.

Brunton, J.A., Bertolo, R.F.P., Pencharz, P.B., Ball, R.O., 1999. Proline ameliorates arginine deficiencyduring enteral but not parenteral feeding in neonatal pigs. Amer. J. Physiol. 277, E223−E231.

Burrin, D.G., Reeds, P.J., 1997. Alternative fuels in the gastrointestinal tract. Curr. Opin. Gastroenterol.13, 165−170.

G. Wu et al.122

Burrin, D.G., Stoll, B., Jiang, R.H., Chang, X.Y., Hartmann, B., Holst, J.J., Greeley, G.H., Reeds, P.J.,2000. Minimal enteral nutrient requirements for intestinal growth in neonatal piglets: how much isenough? Amer. J. Clin. Nutr. 71, 1603−1610.

Bush, J.A., Wu, G., Suryawan, A., Nguyen, H.V., Davis, T.A., 2002. Somatotropin-induced amino acidconservation in pigs involves differential regulation of liver and gut urea cycle enzyme activity.J. Nutr. 132, 59−67.

Caldovic, L., Morizono, H., Yu, X., Thompson, M., Shi, D., Gallegos, R., Allewell, N.M., Malamy, M.H.,Tuchman, M., 2002. Identification, cloning and expression of the mouse N-acetylglutamate synthasegene. Biochem. J. 364, 825−831.

Cappelli, F.P., Seal, C.J., Parker, D.S., 1997. Glucose and [13C]leucine metabolism by the portal-drainedviscera of sheep fed on dried grass with acute intravenous and intraduodenal infusion of glucose.Brit. J. Nutr. 78, 931−946.

Castillo, L., Chapman, T.E., Yu, Y.M., Ajami, A., Burke, J.F., Young, V.R., 1993a. Dietary arginine uptakeby the splanchnic region in adult humans. Amer. J. Physiol. 265, E532−E539.

Castillo, L., DeRojas, T.C., Chapman, T.E., Vogt, J., Burke, J.F., Tannenbaum, S.R., Young, V.R., 1993b.Splanchnic metabolism of dietary arginine in relation to nitric oxide synthesis in normal adult man.Proc. Natl. Acad. Sci. USA 90, 193−197.

Cheng, H., Leblond, C.P., 1974. Origin, differentiation and renewal of the four main epithelial cell typesin the mouse small intestine. V. Unitarian theory of the origin of the four epithelial cell types.Amer. J. Anat. 141, 537−548.

Chung, T.K., Baker, D.H., 1993. A note on the dispensability of proline for weanling pigs. Anim. Prod.56, 407−408.

Cooper, A.J., Meister, A., 1977. The glutamine transaminase–ω-amidase pathway. CRC Crit. Rev.Biochem. 4, 281−303.

Cromwell, G.L., 2001. Antimicrobial and promicrobial agents. In: Lewis, A.J., Southern, L.L. (Eds.),Swine Nutrition. CRC Press, New York, pp. 401−426.

Daenzer, M., Petzke, K.J., Bequette, B.J., Metges, C.C., 2001. Whole-body nitrogen and splanchnicamino acid metabolism differ in rats fed mixed diets containing casein or its corresponding aminoacid mixture. J. Nutr. 131, 1965−1972.

Davis, T.A., Fiorotto, M.L., Reeds, P.J., 1993. Amino acid compositions of body and milk protein changeduring the suckling period in rats. J. Nutr. 123, 947−956.

Davis, T.A., Nguyen, H.V., Garcia-Bravo, R., Fiorotto, M.L., Jackson, E.M., Lewis, D.S., Lee, D.R.,Reeds, P.J., 1994. Amino acid composition of human milk is not unique. J. Nutr. 124, 1126−1132.

Dekaney, C.M., Wu, G., Jaeger, L.A., 2000. Regulation and function of ornithine aminotransferase inanimals. Trends Comp. Biochem. Physiol. 6, 175−183.

DeMarco, V., Dyess, K., Strauss, D., West, C.M., Neu, J., 1999. Inhibition of glutamine synthetasedecreases proliferation of cultured rat intestinal epithelial cells. J. Nutr. 129, 57−62.

DeMars, R., LeVan, S.L., Trend, B.L., Russell, L.B., 1976. Abnormal ornithine carbamoyltransferase inmice having the sparse-fur mutation. Proc. Natl. Acad. Sci. USA 73, 1693−1697.

Dhanakoti, S.N., Brosnan, J.T., Herzberg, G.R., Brosnan, M.E., 1990. Renal arginine synthesis: studiesin vitro and in vivo. Amer. J. Physiol. 259, E437−E442.

Dudley, M.A., Wykes, L.J., Dudley, A.W. Jr., Burrin, D.G., Nichols, B.L., Rosenberger, J., Jahoor, F.,Heird, W.C., Reeds, P.J., 1998. Parenteral nutrition selectively decreases protein synthesis in the smallintestine. Amer. J. Physiol. 274, G131−G137.

Elango, R., Pencharz, P.B., Ball, R.O., 2002. The branched-chain amino acid requirement of parenterallyfed neonatal piglets is less than the enteral requirement. J. Nutr. 132, 3123−3129.

Flynn, N.E., Wu, G., 1996. An important role for endogenous synthesis of arginine in maintainingarginine homeostasis in neonatal pigs. Amer. J. Physiol. 271, R1149−R1155.

Flynn, N.E., Wu, G., 1997a. Glucocorticoids play an important role in mediating the enhanced meta-bolism of arginine and glutamine in enterocytes of postweaning pigs. J. Nutr. 127, 732−737.

Flynn, N.E., Wu, G., 1997b. Enhanced metabolism of arginine and glutamine in enterocytes of cortisol-treated pigs. Amer. J. Physiol. 272, G474−G480.

Flynn, N.E., Knabe, D.A., Mallick, B.K., Wu, G., 2000. Postnatal changes of plasma amino acids in sucklingpigs. J. Anim. Sci. 78, 2369−2375.

Gate, J.J., Parker, D.S., Lobley, G.E., 1999. The metabolic fate of the amino-N group of glutamine in thetissues of the gastrointestinal tract in 24 h-fasted sheep. Brit. J. Nutr. 81, 297−306.

Graber, G., Baker, D.H., 1971. Ornithine utilization by the chick. Proc. Soc. Exp. Biol. Med. 138, 585−588.


Graber, G., Baker, D.H., 1973. The essential nature of glycine and proline for growing chickens. PoultrySci. 52, 892−896.

Harper, A.E., Miller, R.H., Block, K.P., 1984. Branched-chain amino acid metabolism. Annu. Rev. Nutr.4, 409−454.

Hoerr, R.A., Matthews, D.E., Bier, D.M., Young, V.R., 1993. Effects of protein restriction and acuterefeeding on leucine and lysine kinetics in young men. Amer. J. Physiol. 264, E567−E575.

Hoogenraad, N., Totino, N., Elmer, H., Wraight, C., Alewood, P., Johns, R.B., 1985. Inhibition of intes-tinal citrulline synthesis causes growth retardation in rats. Amer. J. Physiol. 249, G792−G799.

Kamoun, P., Aral, B., Saudubray, J.M., 1998. A new inherited metabolic disease: pyrroline-5-carboxylatesynthetase deficiency. Bull. Acad. Natle. Med. Paris 182, 131−139.

Kandil, H.M., Argenzio, R.A., Chen, W., Berschneider, H.M., Stiles, A.D., Westwick, J.K., Rippe, R.A.,Brenner, D.A., Rhoads, J.M., 1995. L-Glutamine and L-asparagine stimulate ODC activity andproliferation in a porcine jejunal enterocytes line. Amer. J. Physiol. 269, G591−G599.

Khatra, B.S., Chawla, R.K., Sewell, C.W., Rudman, D., 1977. Distribution of branched-chain α-keto aciddehydrogenase in primate tissues. J. Clin. Invest. 59, 558−564.

Kikuchi, G., Hiraga, K., Yoshida, T., 1980. Role of the glycine-cleavage system in glycine and serinemetabolism in various organs. Biochem. Soc. Trans. 8, 504−506.

Kim, S.W., McPherson, R.L., Wu, G., 2004. Dietary arginine supplementation enhances the growth ofmilk-fed young pigs. J. Nutr. 134, 625–630.

Klein, R.M., McKenzie, J.C., 1983. The role of cell renewal in the ontogeny of the intestine. I. Cellproliferation patterns in adult, fetal, and neonatal intestine. J. Pediat. Gastroenterol. Nutr. 2, 10−43.

Lapierre, H., Bernier, J.F., Dubreuil, P., Reynolds, C.K., Farmer, C., Ouellet, D.R., Lobley, G.E., 2000. Theeffect of feed intake level on splanchnic metabolism in growing beef steers. J. Anim. Sci. 78, 1084−1099.

Le Floc’h, N., Seve, B., 2000. Protein and amino acid metabolism in the intestine of the pig: from digestionto appearance in the portal vein. Prod. Anim. 13, 303−314.

Lu, S., Yao, Y., Meng, S., Cheng, X., Black, D.D., 2002. Overexpression of apolipoprotein A-IV enhanceslipid transport in newborn swine intestinal epithelial cells. J. Biol. Chem. 277, 31929−31937.

Luk, G.D., Marton, L.J., Baylin, S.B., 1980. Ornithine decarboxylase is important in intestinal mucosalmaturation and recovery from injury in rats. Science 210, 195−198.

Madara, J.L., 1991. Functional morphology of epithelium of the small intestine. In: Shultz, S.G. (Ed.),Handbook of Physiology: The Gastrointestinal System. American Physiological Society, Bethesda,MD, pp. 83−120.

Malo, C., Qureshi, I.A., Letarte, J., 1986. Postnatal maturation of enterocytes in sparse-fur mutant mice.Amer. J. Physiol. 250, G177−G184.

Matthews, D.E., Marano, M.A., Campbell, R.G., 1993. Splanchnic bed utilization of glutamine andglutamic acid in humans. Amer. J. Physiol. 264, E848−E854.

Mavromichalis, I., Parr, T.M., Gabert, V.M., Baker, D.H., 2001. True ileal digestibility of amino acids insow’s milk for 17-day-old pigs. J. Anim. Sci. 79, 707−713.

McNurlan, M.A., Garlick, P.J., 1980. Contribution of rat liver and gastrointestinal tract to whole bodyprotein synthesis in the rat. Biochem. J. 186, 381−383.

Mitchell, A.D., Benevenga, N.J., 1978. The role of transamination in methionine oxidation in the rat.J. Nutr. 108, 67−78.

Morley, J.E., Flood, J.F., 1991. Evidence that nitric oxide modulates food intake in mice. Life Sci. 49, 707−711.Morris, S.M. Jr., 2002. Regulation of enzymes of the urea cycle and arginine metabolism. Annu. Rev.

Nutr. 22, 87−105.Mowat, A.M., 1987. The cellular basis of gastrointestinal immunity. In: Marsh, M.N. (Ed.),

Immunopathology of the Small Intestine. John Wiley, London, pp. 41−72.Murch, S.J., Wilson, R.L., Murphy, J.M., Ball, R.O., 1996. Proline is synthesized from intravenously

infused arginine by piglets consuming low proline diets. Can. J. Anim. Sci. 76, 435−441.Murphy, J.M., Murch, S.J., Ball, R.O., 1996. Proline is synthesized from glutamate during intragastric

infusion but not during intravenous infusion in neonatal piglets. J. Nutr. 126, 878−886.O’Quinn, P.R., Knabe, D.A., Wu, G., 2002. Arginine catabolism in lactating porcine mammary tissue.

J. Anim. Sci. 80, 467−474.Pell, J.M., Caldarone, E.M., Bergman, E.N., 1986. Leucine and α-ketoisocaproate metabolism and

interconversion in fed and fasted sheep. Metabolism 35, 1005−1016.Phelps, T.J., Palumbo, A.V., Beliaev, A.S., 2002. Metabolomics and microarrays for improved under-

standing of phenotypic characteristics controlled by both genomics and environmental constraints.Curr. Opin. Biotechnol. 13, 20−24.

G. Wu et al.124

Pink, D.B.S., Dixon, W.T., Ball, R.O., 2002. Lysine catabolism in swine: an enzymatic approach. FASEB J.16, A258.

Porteous, J.W., 1980. Glutamate, glutamine, aspartate, asparagines, glucose and ketone-body metabolismin chick intestinal brush-border cells. Biochem. J. 188, 619−632.

Reeds, P.J., Burrin, D.G., 2001. Glutamine and the bowel. J. Nutr. 131, 2505S−2508S.Reeds, P.J., Burrin, D.G., Davis, T.A., Fiorotto, M.L., 1993. Postnatal growth of gut and muscle: com-

petitors or collaborators. Proc. Nutr. Soc. 52, 57−67.Reeds, P.J., Burrin, D.G., Davis, T.A., Fiorotto, M.L., Stoll, B., van Goudoever, J.B., 2000a. Protein nutrition

in the neonate. Proc. Nutr. Soc. 59, 87−97.Reeds, P.J., Burrin, D.G., Jahoor, F., Wykes, L., Henry, J., Frazer, E.M., 1996. Enteral glutamate is almost

completely metabolized in first pass by the gastrointestinal tract of infant pigs. Amer. J. Physiol. 270,E413−E418.

Reeds, P.J., Burrin, D.G., Stoll, B., Jahoor, F., 2000b. Intestinal glutamate metabolism. J. Nutr. 130, 978S−982S.Reeds, P.J., Burrin, D.G., Stoll, B., Jahoor, F., Wykes, L., Henry, J., Frazer, M.E., 1997. Enteral glutamate is

the preferential source for mucosal glutathione synthesis in pigs. Amer. J. Physiol. 273, E408−E415.Roig, J.C., Shenoy, V.B., Chakrabarti, R., Lau, J.Y.N., Neu, J., 1995. Localization of rat small intestine gluta-

mine synthetase using immunofluorescence and in situ hybridization. J. Parent. Enter. Nutr. 19, 179−181.Rogers, Q.R., Chen, D.M., Harper, A.E., 1970. The importance of dispensable amino acids for maximal

growth in rats. Proc. Soc. Exp. Biol. Med. 134, 517−522.Rogers, Q.R., Phang, J.M., 1985. Deficiency of pyrroline-5-carboxylate synthase in the intestinal mucosa

of the cat. J. Nutr. 115, 146−150.Shoveller, A.K., Ball, R.O., Pencharz, P.B., 2000. Methionine requirement is 35% lower during parenteral

versus oral feeding in neonatal piglets. FASEB J. 14, A558.Shrawder, E., Martinez-Carrion, M., 1972. Evidence of phenylalanine transaminase activity in the isoen-

zymes of aspartate transaminase. J. Biol. Chem. 247, 2486−2492.Smith, M.W., Jarvis, L.G., 1978. Growth and cell replacement in the new-born pig intestine. Proc. R. Soc.

Lond. B 203, 69−89.Stoll, B., Burrin, D.G., Henry, J., Yu, H., Jahoor, F., Reeds, P.J., 1997. Phenylalanine utilization by the gut and

liver measured with intravenous and intragastric tracers in pigs. Amer. J. Physiol. 273, G1208−G1217.Stoll, B., Burrin, D.G., Henry, J., Yu, H., Jahoor, F., Reeds, P.J., 1999a. Utilization of enteral and arterial pheny-

lalanine for mucosal and hepatic constitutive protein synthesis in pigs. Amer. J. Physiol. 276, G49−G57.Stoll, B., Burrin, D.G., Henry, J., Yu, H., Jahoor, F., Reeds, P.J., 1999b. Substrate oxidation by the portal

drained viscera of fed pigs. Amer. J. Physiol. 277, E168−E175.Stoll, B., Chang, X.Y., Fan, M.Z., Reeds, P.J., Burrin, D.G., 2000. Enteral nutrient intake level determines

intestinal protein synthesis and accretion rates in neonatal pigs. Amer. J. Physiol. 279, G288−G294.Stoll, B., Henry, J., Reeds, P.J., Yu, H., Jahoor, F., Burrin, D.G., 1998. Catabolism dominates the first-pass

intestinal metabolism of dietary essential amino acids in milk-protein-fed piglets. J. Nutr. 128, 606−614.Tagari, H., Bergman, E.N., 1978. Intestinal disappearance and portal blood appearance of amino acids in

sheep. J. Nutr. 108, 790−803.Tourian, A., Goddard, J., Puck, T.T., 1969. Phenylalanine hydroxylase activity in mammalian cells.

J. Cell. Physiol. 73, 59−170.van der Schoor, S.R.D., van Goudoever, J.B., Stoll, B., Henry, J.F., Rosenberger, J.R., Burrin, D.G.,

Reeds, P.J., 2001. The pattern of intestinal substrate oxidation is altered by protein restriction in pigs.Gastroenterology 121, 1167−1175.

van Goudoever, J.B., Stoll, B., Henry, J.F., Burrin, D.G., Reeds, P.J., 2000. Adaptive regulation of intes-tinal lysine metabolism. Proc. Natl Acad. Sci. USA 97, 11620−11625.

Wakabayashi, Y., Yamada, E., Hasegawa, T., Yamada, R., 1991. Enzymological evidence for the indispens-ability of small intestine in the synthesis of arginine from glutamate. Arch. Biochem. Biophys. 291, 1−8.

Wakabayashi, Y., Yamada, E., Yoshida, T., Takahashi, H., 1994a. Arginine becomes an essential aminoacid after massive resection of rat small intestine. J. Biol. Chem. 269, 32667−32671.

Wakabayashi, Y., Yamada, E., Yoshida, T., Takahashi, H., 1994b. Deficiency of endogenous argininesynthesis provokes hypertension by exhausting substrate arginine for nitric oxide synthesis. Biochem.Biophys. Res. Commun. 205, 1391−1398.

Wang, T., Lawler, A.M., Steel, G., Sipik, I., Milam, A.H., Valle, D., 1995. Mice lacking ornithine-δ-aminotransferase have paradoxical neonatal hypo-ornithinaemia and retinal degeneration. NatureGenet. 11, 185−190.

Weiler, H.A., Wang, Z., Atkinson, S.A., 1997. Whole body lean mass is altered by dexamethasonetreatment through reduction in protein and energy utilization in piglets. Biol. Neonate 71, 53−59.


Windmueller, H.G., 1982. Glutamine utilization by the small intestine. Adv. Enzymol. 53, 201−237.Windmueller, H.G., Spaeth, A.E., 1975. Intestinal metabolism of glutamine and glutamate from the

lumen as compared to glutamine from blood. Arch. Biochem. Biophys. 171, 662−672.Windmueller, H.G., Spaeth, A.E., 1976. Metabolism of absorbed aspartate, asparagine, and arginine by

rat small intestine in vivo. Arch. Biochem. Biophys. 175, 670−676.Wu, G., 1995. Urea synthesis in enterocytes of developing pigs. Biochem. J. 312, 717−723.Wu, G., 1996. An important role for pentose cycle in the synthesis of citrulline and proline from glutamine

in porcine enterocytes. Arch. Biochem. Biophys. 336, 224−230.Wu, G., 1997. Synthesis of citrulline and arginine from proline in enterocytes of postnatal pigs. Amer.

J. Physiol. 272, G1382−G1390.Wu, G., 1998a. Intestinal mucosal amino acid catabolism. J. Nutr. 128, 1249−1252.Wu, G., 1998b. Amino acid metabolism in the small intestine. Trends Comp. Biochem. Physiol. 4, 39−74.Wu, G., Becker, R.M., Bose, C.L., Rhoads, J.M., 2001. Serum amino acid concentrations in preterm

infants. J. Pediat. 139, 334−337.Wu, G., Borbolla, A.G., Knabe, D.A., 1994a. The uptake of glutamine and release of arginine, citrulline

and proline by the small intestine of developing pigs. J. Nutr. 124, 2437−2444.Wu, G., Davis, P.K., Flynn, N.E., Knabe, D.A., Davidson, J.T., 1997. Endogenous synthesis of arginine

plays an important role in maintaining arginine homeostasis in postweaning growing pigs. J. Nutr.127, 2342−2349.

Wu, G., Flynn, N.E., Flynn, S.P., Jolly, C.A., Davis, P.K., 1999. Dietary protein or arginine deficiencyimpairs constitutive and inducible nitric oxide synthesis in young rats. J. Nutr. 129, 1347−1354.

Wu, G., Flynn, N.E., Knabe, D.A., 2000a. Enhanced intestinal synthesis of polyamines from proline incortisol-treated pigs. Amer. J. Physiol. 279, E395−E402.

Wu, G., Flynn, N.E., Knabe, D.A., Jaeger, L.A., 2000b. A cortisol surge mediates the enhancedpolyamine synthesis in porcine enterocytes during weaning. Amer. J. Physiol. 279, R554−R559.

Wu, G., Flynn, N.E., Yan, W., Barstow, D.G. Jr., 1995a. Glutamine metabolism in chick enterocytes:absence of pyrroline-5-carboxylate synthase and citrulline synthesis. Biochem. J. 306, 717−721.

Wu, G., Knabe, D.A., 1994. Free and protein-bound amino acids in sow’s colostrums and milk. J. Nutr.124, 415−424.

Wu, G., Knabe, D.A., 1995. Arginine synthesis in enterocytes of neonatal pigs. Amer. J. Physiol. 269,R621−R629.

Wu, G., Knabe, D.A., Flynn, N.E., 1994b. Synthesis of citrulline from glutamine in pig enterocytes.Biochem. J. 299, 115−121.

Wu, G., Knabe, D.A., Flynn, N.E., Yan, W., Flynn, S.P., 1996a. Arginine degradation in developingporcine enterocytes. Amer. J. Physiol. 271, G913−G919.

Wu, G., Knabe, D.A., Yan, W., Flynn, N.E., 1995b. Glutamine and glucose metabolism in enterocytes ofthe neonatal pig. Amer. J. Physiol. 268, R334−R342.

Wu, G., Meier, S.A., Knabe, D.A., 1996b. Dietary glutamine supplementation prevents jejunal atrophy inweaned pigs. J. Nutr. 126, 2578−2584.

Wu, G., Meininger, C.J., Knabe, D.A., Bazer, F.W., Rhoads, J.M., 2000c. Arginine nutrition in develop-ment, health and disease. Curr. Opin. Clin. Nutr. Metab. Care 3, 59−66.

Wu, G., Morris, S.M. Jr., 1998. Arginine metabolism: nitric oxide and beyond. Biochem. J. 336, 1−17.Yen, J.T., Nienaber, J.A., Pond, W.G., Varel, V.H., 1985. Effect of carbadox on growth, fasting metabo-

lism, thyroid function and gastrointestinal tract in young pigs. J. Nutr. 115, 970−979.Yokoyama, K., Ogura, Y., Kawabata, M., Hinoshita, F., Susuki, Y., Hara, S., Yamada, A., Mimura, N.,

Nakayama, M., Kawaguchi, Y., Sakai, O., 1996. Hyperammonemia in a patient with short bowelsyndrome and chronic renal failure. Nephron 72, 693−695.

Young, V.R., El-Khoury, A.E., 1995. The notion of the nutritional essentiality of amino acids, revisited,with a note on the indispensable amino acid requirements in adults. In: Cynober, L.A. (Ed.), AminoAcid Metabolism and Therapy in Health and Nutritional Disease, CRC Press, New York, pp. 191−232.

Yu, F., Bruce, L.A., Calder, A.G., Milne, E., Coop, R.L., Jackson, F., Horgan, G.W., MacRae, J.C., 2000.Subclinical infection with the nematode Trichostrongylus colubriformis increases gastrointestinal tractleucine metabolism and reduces availability of leucine for other tissues. J. Anim. Sci. 78, 380−390.

Yu, Y.-M., Wagner, D.A., Tredget, E.E., Walaszewski, J.A., Burke, J.F., Young, V.R., 1990. Quantitativerole of splanchnic region in leucine metabolism: L-[1-13C,15N]leucine and substrate balance studies.Amer. J. Physiol. 259, E36−E51.

G. Wu et al.126

127

6 Role of intestinal first-pass metabolismon whole-body amino acid requirements

R. F. P. Bertoloa, P. B. Pencharzc,d and R. O. Ballb,d

aDepartment of Biochemistry, Memorial University of Newfoundland,St. John’s, NewfoundLand, Canada A1B 3X9bDepartment of Agricultural, Food and Nutritional Science, University of Alberta,Edmonton, Alberta, Canada T6G 2P5cDepartment of Paediatrics, University of Toronto, Toronto, Ontario,Canada M5G 1X8dThe Research Institute, The Hospital for Sick Children, Toronto,Department of Nutritional Sciences, University of Toronto,Toronto, Ontario, Canada

The small intestine utilizes a different profile of amino acids compared to whole-bodyrequirements. Quantifying the gut requirements for amino acids is critical to understand thelimiting availability of these amino acids during periods of rapid growth in animals. Manymethods have been employed to determine amino acid requirements in man and animalsincluding growth assays, nitrogen balance and amino acid oxidation methods. The most versa-tile approach is the indicator amino acid oxidation technique which can be safely employedin many vulnerable populations. The amino acid requirements of the gut have been estimatedusing this technique in the parenterally fed piglet, which is a model of a gut-deficient animal,and comparing requirements to enterally fed controls. The gut’s requirement for threonine isproportionately greatest of the amino acids tested due to its role in mucin synthesis. The sulphurand branched-chain amino acids are also significantly utilized by the gut. Tryptophan, lysine,phenylalanine and tyrosine utilization by the gut is not significant. The availability of threonineand sulphur amino acids may be limiting for growth in situations of gut stress or disease due tothe higher maintenance requirements during such gut challenges.

1. INTRODUCTION

The small intestine has classically been regarded as a digestive organ responsible for theabsorption of nutrients from foods. Only recently has the gastrointestinal tract, whose metab-olism is dominated by the small intestine, been studied as a significant metabolic tissue with



great impact on whole-body metabolism. Through co-ordinated inter-organ pathways, the gas-trointestinal tract is involved in the synthesis, conversion and catabolism of amino acids to beused by other tissues in the body. In addition to this critical role in whole-body nutriture, the gutalso requires vast amounts of particular amino acids for maintenance and growth. The profile ofthese amino acid requirements does not seem to parallel those for growth and maintenance ofthe rest of the body. Rather, because of the gut’s specific functions in digestion, absorption andimmunity, the gut requires a different profile of amino acids. Quantifying these requirementshas become an important goal in understanding the role of the gut in amino acid availability forthe rest of the body. In particular, this availability becomes of paramount concern in situationsof gut disease or stress where increased maintenance requirements can limit the availability ofcertain amino acids for whole-body growth and physiological functions.

2. METHODS TO MEASURE AMINO ACID REQUIREMENTS

Many methods have been employed in humans and animals to determine amino acid require-ments. The advantages and disadvantages of many of these techniques have been extensivelyreviewed by others (Lewis, 1992; Fuller and Garlick, 1994; Young and el-Khoury, 1995; Zelloet al., 1995; Waterlow, 1999). Many of these methods were originally developed in animals andthen modified for humans. However, it is important to note that most animal research on aminoacid requirements has primarily focused on the growing phase for economic reasons, whereashuman research has almost exclusively focused on the adult phase which comprises most of thelifespan. The basic strategy employed by almost all studies involved with the determination ofamino acid requirements includes the feeding of graded levels of the test amino acid and themeasurement of a specific biological response. The choice of biological response depends onmany factors including species, age, health status, sample availability as well as ethicalconsiderations, analytical equipment availability, financial constraints and practicality.

In choosing the biological response, the most important aspect of amino acid metabolismto consider is the need for an amino acid for incorporation into protein. If all other essentialnutrients, especially energy and other amino acids, are at or above requirement levels, thenwhole-body protein synthesis will occur at a level determined by the intake of the mostlimiting amino acid (i.e. the test amino acid). If the intake of this test amino acid is below itsrequirement, then protein synthesis will be reduced and the intake of all other essential aminoacids will be in relative excess; because amino acids cannot be stored, this excess must becatabolized by the body and excreted as bicarbonate and ammonia via carbon dioxide andurea, respectively. Increasing the intake of the test amino acid will result in greater proteinsynthesis and the concomitant reduction in excess amino acid catabolism indicated by lowercarbon dioxide and urea excretion. At intakes above its requirement, the test amino acid willno longer be the first limiting one and additional intakes will not result in greater proteinsynthesis. At these intakes, the test amino acid itself is in relative excess and must be catab-olized to carbon dioxide and urea. This general scheme has been used to develop almost alltechniques employed to determine amino acid requirements, including growth assays, serialslaughter, nitrogen balance, plasma amino acids, plasma urea, direct amino acid oxidation,amino acid balance, and indicator amino acid oxidation (fig. 1).

It is also important to note that the chosen biological response should be amenable to sta-tistical modelling techniques so that an objective estimate for an amino acid requirement canbe determined, preferably with an estimate of population variance. Because the pattern of thebiological response is rarely predictable over deficient to excess intakes of the test aminoacid, several statistical models have been proposed and employed. The overall response is

R. F. P. Bertolo et al.128

often argued, from a biological standpoint, to be a quadratic model; however, a two-phaselinear regression breakpoint model sometimes fits better. In practice, when both models fitwell, the breakpoint estimate is usually similar between techniques (Baker et al., 2002). Usingeither model, the requirement can be determined within individuals and then averaged for arequirement estimate with population variance. Or the model can be applied to a completedata set of many animals over many different intakes and the error of the fit could be used topredict the population variance. In any event, it is obvious that the more data are available forstatistical manipulation, the more versatile the modelling can be. This issue of statisticalmanipulation is not a trivial one. Indeed, it has been recently demonstrated that re-analysis ofthe classic nitrogen balance studies of Rose and Jones yielded very different conclusionsabout the amino acid requirements in humans using the exact same data (Rand and Young,1999; Di Buono et al., 2001a). These studies have clearly demonstrated that the importanceof the chosen statistical model is almost as important as the data.

2.1. Growth assays

Because the primary role of an amino acid is its incorporation into protein, the measurementof protein synthesis itself, during varying intakes of the limiting amino acid, can be consid-ered to be the most direct of approaches. As such, growth assays, and more specifically theserial slaughter technique, have often been considered as the “gold standard” of techniques inthe determination of amino acid requirements in animals. As the test amino acid intake isincreased towards its requirement, then more protein is synthesized which leads to increasedlean tissue deposition. In young animals with minimal fat deposition, growth is almostdirectly proportional to lean tissue deposition and the growth assay is appropriate. Obviously,as animals approach maturity, more fat is deposited and this proportional relationshipbetween lean and body growth is not constant. To resolve this discrepancy, the serial slaugh-ter technique employs body composition analysis to accurately determine lean tissue content.Using a reference group of animals analysed at the starting body weight, lean tissue accretioncan be determined. However, the main drawback to this approach is the necessity of using

Role of intestinal first-pass metabolism 129

Fig. 1. Graphical representation of various response curves to increasing intakes of the test amino acid.The “breakpoint” requirement is usually determined using two-phase linear regression.

different animals at different time points. Also, body tissue analysis for amino acids is notvery accurate given the heterogeneity of ground body samples and the problems associatedwith protein-bound amino acid analysis. So although these techniques are simple and direct,they are limited to fast-growing lean animals with low and constant fat deposition or mustinvolve large numbers of slaughtered growing animals of similar genetic background to minimizethe inter-animal variation. These techniques are not useful in adult, non-growing animals orin animals with special conditions (i.e. gestation, lactation, egg production, disease, etc.).

2.2. Nitrogen balance

When protein synthesis is limited, excess amino acids are catabolized to their metabolic end-products which for all amino acids include bicarbonate and ammonia. The measurement ofthese excreted biological products provides an indirect and inverse measurement of changesin protein synthesis and thus can be used to determine amino acid requirements. Ammoniaenters the nitrogen pool of the body and is excreted primarily as urea in mammals and as uricacid in birds. Because such excreta are relatively easy to collect and analyse for total nitrogen,the nitrogen balance method was one of the first to be developed for assessment of amino acidrequirements in humans and animals. The amino acid requirement can be determined fromeither balance calculations or over a range of test amino acid intakes. The balance approachregresses nitrogen balance on test amino acid intake and defines the requirement as the intakelevel at which optimum balance is achieved (i.e. zero or positive balance in adults). A consid-erable drawback with balance calculations is the need to make an assumption of unmeasurablelosses (sweat, skin, nails, hair, etc.). Indeed, some of the original landmark experiments byRose and Jones to determine human amino acid requirements did not include such an assump-tion; when these data were later corrected for an estimate of these losses, amino acidrequirement estimates more than doubled (Young and Marchini, 1990; Rand and Young, 1999).And because two large numbers are being subtracted (nitrogen intake and nitrogen output),the difference is relatively very small and this assumption becomes extremely important. Analternative approach not involving this correction is to measure the qualitative change in nitro-gen balance over a range of test amino acid intakes. Nitrogen output is high when proteinsynthesis is limited by a single amino acid because the other amino acids are in relative excessand are catabolized. As the test amino acid is added incrementally to the diet, nitrogen outputdecreases until the requirement is met and further increments will not stimulate proteinsynthesis and nitrogen balance will remain constant (assuming isonitrogenous diets).

The many technical advantages and disadvantages of the nitrogen balance technique havebeen extensively reviewed over the years (Waterlow, 1999; Tome and Bos, 2000) and will notbe discussed here. However, several important points need to be mentioned in the presentchapter. An important advantage of the nitrogen balance technique is that unlike growthassays, this method can be successfully applied to adult as well as young species. The majordisadvantage of the technique is that because of the very large urea pool in most species,its response to dietary manipulation is rather slow and thus adaptations of a week or greaterare generally required. If one is studying 6 or 7 test amino acid levels, this long adaptationresults in a lengthy experimental period which is not useful in special physiological situationssuch as gestation, lactation or disease progression. In addition, this technique is also inap-propriate in vulnerable populations where long-term feeding of deficient diets is not ethical.The nitrogen balance technique has been extensively applied for all indispensable aminoacids in many species.


2.3. Direct amino acid oxidation technique

As opposed to the nitrogen balance technique, oxidation methods monitor the excretion ofcarbon dioxide, the other obligatory end-product of amino acid catabolism. However, thebasic principle is similar in concept to other methods to determine amino acid requirements.At deficient test amino acid intakes, the test amino acid is efficiently utilized for proteinsynthesis and its oxidation is low and constant. At intakes above requirement, protein syn-thesis is maximized and excess test amino acid is preferentially oxidized to its end-products.Instead of measuring changes in nitrogen excretion as in the nitrogen balance technique,amino acid oxidation methods measure changes in carbon dioxide excretion in breath.

The use of isotopically labelled amino acids allows for an extremely sensitive means ofmeasuring small changes in amino acid oxidation in response to changes in intake. Althoughall carbons of amino acid skeletons are eventually oxidized, the most sensitive approach is tomonitor the expiration of the cleaved carboxyl group at the 1-carbon position. Thus, oxida-tion of uniformly labelled amino acids results in distribution of the label among manymetabolites which makes interpretation difficult and somewhat less sensitive. Alternatively,the oxidation of carboxyl-labelled amino acids is more direct and easier to interpret providedthe decarboxylation step is irreversible. However, because one is measuring carbon dioxidein breath, the carboxyl group must also enter general bicarbonate pools which equilibratereadily with carbon dioxide expiration at the lungs. For example, experiments with labelledthreonine and methionine have found that non-linear responses are typical with infusion ofthese amino acids over varying intakes as a result of their more complex degradative pathways(Chavez and Bayley, 1976; Zhao et al., 1986; Storch et al., 1988; Ballevre et al., 1990). Thus, assummarized by Zello et al. (1995), there are general criteria for choosing appropriate carboxyl-labelled amino acids for oxidation studies: (1) the amino acid must be indispensable; (2) it mustbe primarily partitioned between oxidation to carbon dioxide and protein incorporation; and(3) the labelled carboxyl group must be irreversibly oxidized and sufficiently equilibratedwith labelled carbon dioxide in breath. These restrictions adequately apply for some indis-pensable amino acids such as phenylalanine (provided excess dietary tyrosine is fed), lysineand the branched-chain amino acids; as expected, these amino acids have been used exten-sively in direct oxidation studies. Although the carboxyl group of methionine is irreversiblyoxidized, methionine can equilibrate reversibly with homocysteine prior to the irreversibleoxidative pathway, thus complicating interpretation of oxidation data.

Brookes et al. (1972) were the first to use the oxidation of isotopically labelled amino acidsto determine amino acid requirements in rats. Since then, the direct oxidation technique hasbeen used to determine the requirements of several amino acids in growing rats (Kang-Leeand Harper, 1977, 1978; Harper and Benjamin, 1984), adult rats (Simon et al., 1978), youngpigs (Kim et al., 1983a,b; Ball and Bayley, 1984, 1986; House et al., 1997a,b), infants(Roberts et al., 2001a) and adult humans (Meguid et al., 1986a,b; Meredith et al., 1986; Zhaoet al., 1986; Zello et al., 1990). For animals, the direct oxidation technique has yielded simi-lar or slightly lower amino acid requirements compared to “classical” techniques such asnitrogen balance and growth assays; this finding validates the approach to a certain extent. Incontrast, direct oxidation studies in adult humans have yielded much higher (i.e. 2- to 3-fold)requirements for all tested amino acids compared to those proposed by the FAO/WHO/UNU(1985) based upon nitrogen balance studies. This latter discrepancy has been extensivelydebated and reviewed (Young and Borgonha, 2000). However, one cannot ignore the prob-lems demonstrated in the interpretation of the small amount of original balance data (Randand Young, 1999). Indeed, the human lysine requirement estimated from the re-analysed


original nitrogen balance data (Rand and Young, 1999) actually agrees very well with the esti-mates derived from various techniques using direct oxidation (Meredith et al., 1986), indicatoramino acid oxidation (Zello et al., 1993; Duncan et al., 1996), 24 h oxidation (el-Khoury et al.,2000) and indicator amino acid balance (Kurpad et al., 2001). In addition, animal studiesemploying nitrogen balance techniques, which are numerous and well controlled, result inmore reliable and agreeable results compared to those of more recent kinetic techniques. Inany event, the direct oxidation technique provides as biologically valid an approach as thenitrogen balance technique and generally agrees with the growth assays performed to date inanimals.

A very important advantage of the oxidation method compared to the nitrogen balancetechnique is the more rapid adaptation of the biological response to test amino acid intakechanges. Initial studies in direct amino acid oxidation fed particular test amino acid levels for7–10 days prior to oxidation measurement, analogous to nitrogen balance studies. However,more recently it has been demonstrated that prior adaptation to amino acid intake (Zello et al.,1990; Motil et al., 1994) does not affect the breakpoint estimate of its requirement using thedirect oxidation approach. Therefore, only hours of adaptation to a deficient or excess levelof test amino acid seems to be necessary to measure changes in oxidation, and thus to deter-mine requirement. To our knowledge, there have been no studies in animals designed toaddress this adaptation issue using the direct oxidation technique. However, we have suc-cessfully employed the direct oxidation technique to measure phenylalanine requirement inparenterally fed piglets using only 16 h of adaptation prior to oxidation measurement (Houseet al., 1997a). The adaptation issue has been carefully addressed in indicator amino acidoxidation studies in animals and will be discussed below.

It is important to note that this issue of sufficient adaptation is the subject of considerabledebate and has been reviewed (Young and Marchini, 1990). With long-term adaptation to adeficient diet, the subject will “accommodate” to this situation and possibly become moreefficient in its metabolism. The question is: does this accommodation come at the cost ofother unmeasured metabolic functions? If such costs were incurred, then it violatesWaterlow’s (1985) reasonable definition of adaptation: the process that permits the organismto respond to a dietary change without adverse consequences. We also need to consider thedefinition of amino acid requirement which has been proposed by Young and Borgonha(2000) as the minimal intake level needed to maintain a specific nutritional criterion such asgrowth, body composition, body amino acid balance, organ or system function, etc: thechoice of nutritional criterion then becomes the subject of debate. In spite of this ongoingdebate, the studies in humans have methodically shown that the direct oxidation method hasthe distinct advantage over the nitrogen balance technique of very short adaptation periodsresulting in more time-efficient and cost-effective studies.

Although the direct oxidation technique has provided a more sensitive approach to deter-mination of amino acid requirements, its disadvantages have limited its widespread use inmany species. Most importantly, as mentioned above, not all indispensable amino acids canbe easily used for direct oxidation measurements, limiting its general application. Morespecifically, of the indispensable amino acids, only phenylalanine (with excess dietary tyro-sine), methionine, lysine and the branched-chain amino acids undergo irreversible oxidationof the carboxyl carbon so that amino acid oxidation can be calculated from expired breathcarbon dioxide. However, Young and colleagues have attempted to predict the requirementsof the other amino acids using previously determined tracer techniques, composition of bodyproteins and assumed obligatory oxidative amino acid losses (Young and el-Khoury, 1995);these predictions were subsequently validated (Raguso et al., 1999). An additional criticism


from a kinetic standpoint is that the feeding of deficient to excess amounts of test amino acidchanges that amino acid’s pool size dramatically, thereby diluting the tracer being measured.This variable dilution of the tracer in the pool increases variability and reduces sensitivity ofoxidation measurements and hence requirement estimates.

2.4. Indicator amino acid oxidation technique

The indicator amino acid oxidation (IAAO) technique is an extrapolation of the earlier workwith the direct oxidation technique. Both require the accurate measurement of amino acidoxidation by collecting isotopically labelled carbon dioxide in breath. As with the directoxidation method, the IAAO method is based on the hypothesis that the partitioning of aminoacid metabolism between incorporation into protein and catabolism via oxidation is deter-mined by the most limiting amino acid in the diet. However, the difference between thetechniques is that instead of measuring oxidation of the test amino acid, the IAAO methodmeasures the oxidation of one of the other amino acids that is also responding to changes inprotein synthesis. When the test amino acid is deficient, protein synthesis is limited and otheramino acids are in excess. Indispensable amino acids in excess must be catabolized at a levelinversely reflecting the rate of protein synthesis which is dictated by the test amino acidintake. By monitoring the oxidation of one of these “indicator” amino acids over a range oftest amino acid intakes, one can estimate the test amino acid requirement for protein synthesis.As intake of the test amino acid increases towards requirement, protein synthesis increaseswhich utilizes more of the indicator amino acid resulting in a smaller excess and lower oxida-tion. Once the test amino acid intake equals requirement, then greater intakes of this amino acidwill not lead to greater protein synthesis and therefore indicator amino acid oxidation remainsconstant.

The choice of indicator amino acid depends on its metabolic characteristics. Phenylalanine(see below), lysine (Ball and Bayley, 1984; Roberts et al., 2001b) and leucine (Kurpad et al.,2001) have been used with success to determine various amino acid requirements. Methionine(Brookes et al., 1972) has been employed unsuccessfully due to its complicated metabolicpathways, as mentioned previously. When the requirement of a test amino acid has beendetermined using more than one of the indicators, the requirement estimates were very similar(Ball and Bayley, 1984; Zello et al., 1993; Kurpad et al., 2001). Phenylalanine has been usedmost often as the indicator with successful determinations of the requirements for lysine (Kimet al., 1983a; House et al., 1998a), histidine (Kim et al., 1983b), threonine (Kim et al., 1983a;Bertolo et al., 1998), tryptophan (Cvitkovic et al., 2000), methionine and total sulphur aminoacids (Kim and Bayley, 1983; Shoveller et al., 2001), branched-chain amino acids (Elangoet al., 2002a), arginine (Ball et al., 1986), proline (Ball et al., 1986) and total protein (Ball andBayley, 1986) in piglets, tryptophan in trout (Were, 1989), lysine in chickens (Coleman et al.,2002), lysine in growing pigs (Bertolo et al., 2001), and in humans, lysine (Zello et al., 1993;Duncan et al., 1996; Kriengsinyos et al., 2002), threonine (Wilson et al., 2000), tryptophan(Lazaris-Brunner et al., 1998), methionine (Di Buono et al., 2001b), total sulphur amino acids(Di Buono et al., 2001a), branched-chain amino acids (Mager et al., 2001, 2002; Riazi et al.,2001) and tyrosine (Bross et al., 2000; Roberts et al., 2001b).

The IAAO technique to determine amino acid requirements was first developed in theneonatal piglet by Bayley and colleagues. Following observations that amino acid catabolismdepends on the balance of other amino acids (Brookes et al., 1972; Newport et al., 1976),these researchers successfully demonstrated that the IAAO technique can be used to deter-mine the amino acid requirements in piglets. In particular, Kim et al. (1983b) showed that the


estimate for histidine requirement in piglets was similar using both the direct oxidation andIAAO techniques with phenylalanine as the indicator. In addition, Ball and Bayley (1984)found that either phenylalanine or lysine could be used as an indicator because the oxidationof both responded to varying tryptophan intakes similarly. To further validate the theoreticalconcept, Ball and Bayley (1984) also demonstrated that liver protein synthesis was inverselycorrelated with phenylalanine oxidation. In addition, this research group demonstrated thatthe amino acid requirements for piglets determined using the IAAO technique, namely histi-dine (Kim et al., 1983b), sulphur amino acids (Kim and Bayley, 1983), lysine and threonine(Kim et al., 1983a), tryptophan (Ball and Bayley, 1984), proline and arginine (Ball et al.,1986) and total protein (Ball and Bayley, 1986), agreed very closely with those determinedby classical techniques (NRC, 1979, 1998).

From the initial studies of Bayley and colleagues, the IAAO method has been subsequentlyrefined and expanded from the original approach. All indispensable amino acids have beentested using this method; this aspect is the main advantage of the IAAO method over thedirect oxidation technique. In addition, the aforementioned criticism of the direct oxidationtechnique regarding amino acid pool size does not apply to the IAAO approach. Indeed,because the indicator amino acid is fed at the same level over varying test amino acid intakes,its pool size is not permitted to change and therefore the tracer is not variably diluted. Thisunchanging pool size is probably the main reason why the IAAO approach tends to giverequirement estimates with less variability compared to direct oxidation estimates.

As with the direct oxidation technique, the adaptation period required to a particular testamino acid intake has been shown to be minimal. Indeed, in recent human studies, the lysinerequirement was similar whether hours (Zello et al., 1993; Duncan et al., 1996; Kriengsinyoset al., 2002), 7 days or 21 days of adaptation (Kurpad et al., 2002b) were employed. This findingis profound in context with the aforementioned ongoing debate about adaptation versusaccommodation. Furthermore, this adaptation seems to be relatively insensitive to body sizeor growth. We have also recently found that phenylalanine oxidation after 1.5 days of adap-tation (the shortest adaptation tested) to a high or deficient lysine diet was not different up to8 days of adaptation in both 25 kg growing pigs and 250 kg sows (Bertolo et al., 2001). Thisdistinct advantage of short adaptation over the classical nitrogen balance or growth assaysallows great versatility in the application of oxidation techniques, especially to vulnerablepopulations (Brunton et al., 1998). Indeed, we have recently determined the branched-chainamino acid requirement of children with the inherited genetic disorder, maple syrup urinedisease (unpublished), as well as the tyrosine requirements of parenterally fed infants (Robertset al., 2001a) and the phenylalanine (unpublished) and tyrosine (Bross et al., 2000) requirementsof children with phenylketonuria.

Because an estimate of population variance is critical to recommending amino acidrequirements, the more data in the breakpoint model, the better. Because of the short adapta-tion time associated with the IAAO method, it is possible to measure amino acid requirementsin individuals. To accurately determine a breakpoint in a two-phase linear regression model,at least six test amino acid intakes should be included (i.e. three oxidation measurements perregression). In the nitrogen balance technique, this would require at least 6 weeks of experi-mentation assuming 7 days of adaptation per diet. Because we have shown that only 1.5 daysof adaptation were necessary in the IAAO method (Bertolo et al., 2001), we recently deter-mined the lysine requirement of individual growing pigs in 2 weeks by changing diets andmeasuring indicator oxidation every other day (Moehn et al., 2001). Similarly, we havedemonstrated that individual amino acid requirements of chickens can be determined in lessthan 3 weeks (Coleman et al., 2002). With the determination of enough individual amino acid


requirements, an accurate population variance can be calculated; for animal species, this tech-nique allows such a calculation for the first time. Indeed, more accurate diet formulation andanimal performance can be achieved with knowledge of such an error within a genetic popu-lation of animals. In addition, long-term genetic improvement can also be achieved if animalswith low amino acid requirements, or more efficient utilization of dietary amino acids, couldbe selected and subsequently bred. The obvious potential for such a technique to be exploitedby various animal production groups is enormous.

2.5. 24 hour oxidation/indicator amino acid balance

The balance technique is based on the principle that in non-growing adults, protein synthesisis balanced with protein breakdown and thus protein or nitrogen intake is balanced with nitro-gen excretion. These two balances are connected through the free amino acid pool which isrelatively tightly regulated and represents a minute proportion of total body nitrogen. Thisrelationship between the two balances is often expressed by the steady-state flux equationproposed by Waterlow et al. (1978): flux (Q) = synthesis (S) + oxidation/excretion (O) = break-down (B) + input (I). This equation is rearranged so that protein balance (S − B) = input/outputbalance (I − O). This latter equation is the basis of the nitrogen balance calculation techniqueas well as the amino acid balance technique.

A recent adaptation of the oxidation techniques incorporates the balance concept describedby Waterlow et al. (1978). Young and colleagues have developed a new method involving a24 h infusion of amino acid tracer and the measurement of labelled carbon dioxide output inadult humans (el-Khoury et al., 1994a,b, 1995). These data are then used to calculate carbonbalance at different levels of test amino acid intake. The requirement is taken as the minimalamino acid intake necessary to maintain balance. This amino acid balance is the differencebetween the intake of the test amino acid and whole-body oxidation of that amino acid. Thisapproach was first employed to determine the leucine requirement in adult humans whichcompared very well with their previous direct oxidation experiments (Young et al., 1989).Subsequently, the technique was successfully used to verify direct oxidation determinationsof aromatic amino acid (Basile-Filho et al., 1997, 1998; Sanchez et al., 1995, 1996) and lysine(el-Khoury et al., 1998, 2000) requirements. These experiments also demonstrated that whenthe test amino acids were fed at the FAO/WHO/UNU (1985) requirement levels, subjectswere in significantly negative amino acid balance, indicating that the present acceptedrequirements are too low (Young and Borgonha, 2000). This approach was advanced by thesame group by applying the IAAO technique. Using 24 h labelled leucine infusions, lysine(Kurpad et al., 2001, 2002b) and threonine (Kurpad et al., 2002a) requirements weredetermined by measuring leucine oxidation and balance over a range of test amino acidintakes. These new techniques are particularly suited for non-growing adults and account foramino acid metabolism in both the fasting and fed states. However, fasting-state amino acidkinetics are not relevant to young suckling animals (Bertolo et al., 2000a). Furthermore, thisapproach has only been used in humans and may not be widely applicable to fast-growingmeat-producing animals that are in continuous positive nitrogen balance due to high rates oflean tissue deposition.

As with the nitrogen balance calculation technique, the most important criticism of theamino acid balance techniques is the reliance on absolute calculations based on variousassumptions. With amino acid kinetic calculations, the most important assumption is that theprecursor amino acid enrichment in readily accessible body pools (i.e. plasma) is representa-tive of the enrichment of the true intracellular precursors for protein synthesis (i.e. tRNA)


and oxidation; this assumption is almost always untrue in constant infusion experiments(Wolfe, 1992). Intracellular exchange of labelled amino acids is incomplete and variableamong tissues. Indeed, plasma free amino acid enrichments have been found to be 2 to 3 timesgreater than those for corresponding amino acyl-tRNA enrichments for leucine, lysine andphenylalanine (Caso et al., 2001). Because α-ketoisocaproate (KIC) is synthesized fromintracellular leucine only and is released easily into plasma, KIC enrichment has often beenproposed as a suitable representation of intracellular leucine enrichments. However, severalstudies have shown that KIC enrichment is not in equilibrium with the entire intracellularleucine pool (Cobelli et al., 1991; Chinkes et al., 1993) or leucine-tRNA in pigs (Baumannet al., 1994) or rats (Watt et al., 1991). Alternatively, others have used enrichments of aminoacids that have been incorporated into apolipoprotein B-100 (Reeds et al., 1992; Cayol et al.,1996; Stoll et al., 1999), fibrinogen (Bennet and Haymond, 1991; Stoll et al., 1999) or albumin(Cayol et al., 1996). These very rapidly synthesized plasma proteins of hepatic origin canreflect isotopic steady state within hours. These enrichments are much lower than KIC enrich-ments but similar to tRNA measurements. Another issue is the possibility that precursors foroxidation may not equilibrate with tRNA pools, either intracellularly or between tissues, sothat different enrichments may need to be measured to accurately calculate balance. Theamino acid balance techniques will need to address this precursor enrichments issue.

An advantage of the “relative” techniques comparing biological outcomes across dietaryintakes is the avoidance of absolute assumptions. Indeed, in direct oxidation and IAAO analy-sis of amino acid requirements, the most reliable estimate of requirement with the lowesterror is when percent dose oxidized is used as the biological outcome (House et al., 1997a,b,1998a; Bertolo et al., 1998; Lazaris-Brunner et al., 1998; Bross et al., 2000; Roberts et al.,2001a,b). This outcome is in contrast to equivalent measurements of oxidation rate, whichemploy flux calculations that also use assumptions about the precursor pool. Ultimately, the“black box” approach of total labelled carbons in and total labelled carbons out provides themost reliable requirement estimates. Despite these methodological comparisons, it is impor-tant to reiterate that the amino acid balance studies to date have calculated amino acidrequirements that are very close to those determined by the more qualitative direct oxidationand IAAO techniques. Therefore, the error associated with these kinetic assumptions andcalculations may not be as significant as some have proposed.

3. RECENT DEVELOPMENTS IN THE INDICATOR AMINO ACIDOXIDATION TECHNIQUE

Considering that all of the techniques used to determine amino acid requirements generallyagree in their final estimate, one must choose the most versatile method available for widerapplication. Given the short adaptation time, applicability to all indispensable amino acids,ease of biological sampling (i.e. breath) and applicability to most populations, the IAAOtechnique has been successfully adapted for use in a variety of situations and purposes. Animportant advance in the method was its adaptation for use in parenterally fed piglets asmodels for parenterally fed infants (Wykes et al., 1993; Bertolo et al., 1998; House et al.,1998a). The original work by Bayley in piglets employed one or two oral bolus feeds whichincluded the isotope dose; the total labelled carbon dioxide excretion was not kineticallyquantified, but rather relatively compared over different test amino acid intakes. With par-enterally fed piglets, primed constant intravenous infusions of labelled indicator amino acidwere introduced to acquire added information through kinetic calculations. With these studies,it was shown that the indicator (phenylalanine) flux rate was unchanging across test amino


acid intakes (Bertolo et al., 1998; House et al., 1998a). In addition, it was demonstrated that,statistically, percent dose oxidized was the most reliable estimate in terms of variability(House et al., 1997a,b, 1998a; Bertolo et al., 1998); this finding has been confirmed in humanstudies (Lazaris-Brunner et al., 1998; Bross et al., 2000; Roberts et al., 2001a,b). Recently, wehave also directly compared amino acid requirements using either intravenous or oral/gastricinfusion of labelled phenylalanine as the indicator amino acid. The breakpoint estimates ofthe lysine requirement in adult humans were the same whether the indicator was deliveredintravenously or orally (Kriengsinyos et al., 2002); similarly, the tryptophan requirements ingastrically fed piglets were similar with intravenous or intragastric infusion of the indicator(Cvitkovic et al., 2000). These subsequent methodological developments were critical inadapting the IAAO technique for vulnerable populations.

In order to use the IAAO technique in infants and children, the dietary interventions mustbe short and the sample collections must be non-invasive. We have recently validated the useof oral isotope dosing and collection of urinary amino acids to measure enrichment as repre-sentative of plasma enrichment (Bross et al., 1998). Study diets were fed hourly for 4 h priorto dosing and subsequent half-hourly meals with isotope led to enrichment plateaux within 2 h.This simple non-invasive protocol can be used to study many different vulnerable popula-tions, provided that dietary ingestion and breath collection are feasible. Similarly, this oraldosing protocol has been shown to be very effective in determining amino acid requirementsin large pigs where the implantation and maintenance of catheters can be problematic (unpub-lished data). In addition, such a simplified protocol allows for a broader application of thetechnique to other experimental models. Such models include neonatal and adult animals,gestating or lactating animals, as well as compromised populations which include disease orsurgical interventions.

The IAAO technique can also be applied to research investigating other aspects of proteinmetabolism where the goal is not simply to determine amino acid requirements. Because indi-cator amino acid oxidation responds to protein synthesis, intracellular changes in test aminoacid availability are reflected in changes in indicator oxidation. Recently we have exploitedthis principle by adapting the technique to determine true metabolic availability of lysinefrom feedstuffs in growing pigs (Ball et al., 2001). We designed a low-lysine diet with allother indispensable amino acids above requirements. With incremental additions of syntheticlysine, which is assumed to be 100% available, phenylalanine oxidation declined linearlyuntil the requirement was met. Within this deficient range of lysine intakes, we used the linearresponse equation to predict true availability of lysine from added feedstuffs (fig. 2). Whenpeas were added to the low (50% of requirement) lysine diet so that the total true availablelysine content was 90% of lysine requirement, the phenylalanine oxidation corresponded tothe availability predicted by ileal digestibility estimates (i.e. the true available amount accord-ing to NRC, 1998). When peas were heated to render some lysine unavailable via Maillardproducts, the phenylalanine oxidation increased and corresponded to an availability of 50%of unheated peas. When synthetic lysine was added to the heated peas, phenylalanine oxida-tion decreased below that determined with 90% of requirement, demonstrating that theincrease in oxidation due to heating was due solely to loss of available lysine. This techniqueis a new, rapid approach to determining true metabolic availability of lysine that does notrequire a series of tenuous assumptions about endogenous losses, which are problematic tothe true ileal digestibility method. This novel application of IAAO principles to improve esti-mates of amino acid availability has demonstrated the technique’s versatility.

We have also used the basic qualitative principle of the IAAO technique to determinewhether amino acid formulations are inadequate. Recently, by adding amino acids suspected


of being deficient to TPN solutions and monitoring lysine (the indicator) oxidation, we havesystematically demonstrated that the amino acid profile of some commercial TPN solutionsare inadequate (unpublished data). This approach can be further adapted to simulate the strategypioneered by Baker and colleagues to determine amino acid requirements by dietary aminoacid supplementation and deletion by employing the ideal ratios to lysine (Mavromichaliset al., 1998). The important advantage of the IAAO approach is that multiple modificationscan be made to the diets of individual animals because short oxidation measurements areused, as opposed to the more time-consuming growth assays. Because of the relatively simpleand direct biological response in the IAAO technique, many more extrapolations and adaptationsof the original technique will probably be developed in the future.

4. INTESTINAL FIRST-PASS METABOLISM

In addition to being responsible for the digestion and absorption of nutrients, the gut is alsoa major metabolic organ in the body, responsible for the synthesis, conversion and degrada-tion of amino acids. The gut has a very high metabolic activity and extracts a significantproportion of the absorbed dietary and endogenous amino acids before transport to the portalcirculation and the rest of the body. Indeed, although the portal-drained viscera (PDV) (intes-tines, pancreas, spleen, stomach) represents only ~5−7% of body mass, these tissuesdisproportionately account for 20–35% of whole-body energy expenditure and protein syn-thesis (Lobley et al., 1980; McNurlan and Garlick, 1980; Burrin et al., 1990). This significantextraction of dietary amino acids by splanchnic tissues has been demonstrated by comparingamino acid kinetics when isotopes are delivered intravenously or orally. The “splanchnicdisappearance” of labelled amino acids when given orally has led many researchers to spec-ulate on the fate of this irreversible loss of label in kinetics experiments. From these types ofstudies, several researchers have determined that in humans and animals, the splanchnictissues metabolize between 20% and 50% of dietary essential amino acids (leucine, lysine,phenylalanine) on first-pass (Yu et al., 1990, 1995; Hoerr et al., 1991, 1993; Biolo et al., 1992;Matthews et al., 1993; van Goudoever et al., 2000). In addition, with low-protein diets, thisextraction is as high as 70% of lysine intake (van Goudoever et al., 2000). Albeit the data arevery impressive, one problem with this approach is that it is difficult to separate the metabo-lism of the liver from the PDV, although this can be overcome by measuring amino acid


Fig. 2. Representation of novel technique to determine metabolic amino acid availability by measuringindicator amino acid oxidation.

enrichments and flow in the portal vein (van Goudoever et al., 2000). An additional problemwith the technique is that these types of studies vastly underestimate arterial extraction ofrecirculating enteral isotope and hence overestimate first-pass extraction. Indeed, the elegantstudy by van Goudoever et al. (2000) demonstrated that a 46% portal mass balance extractioncorresponded to a 22% isotopic extraction by the PDV, which when corrected for arterialrecirculation amounted to insignificant utilization of dietary lysine on first-pass.

The relative importance of nutrient processing by the small intestine (the predominant PDVorgan) versus the liver has only recently been elucidated. Indeed, we have demonstrated thatthe small intestine is more important than the liver in modifying nitrogen utilization whenpigs are chronically fed by central vein, portal vein or stomach (Bertolo et al., 1999). In addi-tion, other researchers have demonstrated that intestinal metabolism dominates splanchnicmetabolism of phenylalanine (Stoll et al., 1997) and lysine (van Goudoever et al., 2000) inpigs and leucine in dogs (Yu et al., 1990). In addition to dietary extraction of amino acids, the gutalso transports an enormous amount of amino acids from the arterial circulation, especiallyduring the post-absorptive state. Isotopic data describing the incorporation of amino acidsfrom both arterial and dietary sources have demonstrated that both sources of precursor aminoacids are critical and that the partitioning between them is dependent on specific amino acidand dietary protein level (MacRae et al., 1997; Stoll et al., 1999; van Goudoever et al., 2000).Recent work by van Goudoever et al. (2000) has determined that with high-protein feeding,almost all of the lysine utilized by the PDV was of arterial origin, whereas with low-proteinfeeding, approximately half of the lysine utilized was from both arterial and dietary sources.It is important to note that level of protein feeding did not affect total lysine use by the PDV,demonstrating an enormous obligatory utilization of amino acids by the gut for normal func-tion and growth.

Because of this high obligatory protein turnover in the gut, it follows that with gut chal-lenges (i.e. tissue damage, increased growth, pathogen exposure, dietary anti-nutritionalfactors, etc.) the amino acid requirements of the gut increase. Indeed, first-pass intestinalextraction of amino acids is proportional to mucosal mass. Stoll et al. (1997, 1999) demon-strated that phenylalanine splanchnic extraction was 50% higher in pigs raised outdoors,where rooting and pathogen challenges are greater, compared to pigs raised in a relativelyclean indoor research facility; this higher extraction correlated with measured mucosal mass.Infestation of pathogens is known to affect growth rate as well as gut function. The stimula-tion of whole-body and gut immune systems must impart a protein synthetic cost to theinfected animal. Indeed, sepsis in rats has been shown to stimulate intestinal protein synthe-sis (von Allmen et al., 1992; Higashiguchi et al., 1994b); in particular, sepsis stimulates thesynthesis of endogenous and secretory proteins, including certain gut peptides, in small intes-tine mucosa (Higashiguchi et al., 1994a). More recently, Yu et al. (2000) have shown thatsubclinical nematode infection in sheep increased total gastrointestinal tract leucine seques-tration by 24% and gastrointestinal tract oxidative losses of leucine by 22–41%. In anotherstudy in pigs, the infusion of endotoxin led to enhanced intestinal catabolism of aminoacids (Bruins et al., 2000). These intriguing studies suggest a possible mechanism for thegrowth-stimulating effect of feed-grade antibiotics; increased protein synthesis duringinfection by the already metabolically demanding intestinal tissues limits the availability ofamino acids for extraintestinal lean tissue growth. With the impending discontinuation ofprophylactic feeding of antibiotics in animal production, the prevention of subclinical infec-tions or challenges needs to be a priority in the development of alternative strategies in thenear future.


The primary fate of indispensable amino acids is presumably to protein synthesis; however,recent intriguing data have demonstrated that catabolism dominates the first-pass utilizationof these amino acids by the gut (Stoll et al., 1998; Wu, 1998; van Goudoever et al., 2000).This seemingly wasteful oxidation of indispensable amino acids amounts to a small but signif-icant proportion of dietary intake (Yu et al., 1992, 2000; Cappelli et al., 1997; van Goudoeveret al., 2000), but a large proportion of whole-body amino acid oxidation (van Goudoever et al.,2000). Indeed, we have recently demonstrated that phenylalanine oxidation is significantlygreater when labelled phenylalanine is delivered orally, as opposed to intravenously in theIAAO technique (Cvitkovic et al., 2000; Kriengsinyos et al., 2002). This increased oxidationduring feeding of adequate diets demonstrates that first-pass catabolism of phenylalanine bythe gut is significant.

Whatever the fate of indispensable amino acids extracted by the gut, the evidence clearlysuggests that this tissue plays a significant role in modulating the profile of amino acids deliveredto the rest of the body (Stoll et al., 1998; Wu, 1998; Bertolo et al., 2000b). Much of this roleresults in a net loss of amino acids to extraintestinal tissues. Presumably, when gut metabolicactivity is increased by growth or stress, so is the loss of dietary amino acids from whole-bodyfunctions. This aspect of whole-body amino acid requirements has not been fully explored.In addition to gut metabolic activity, the quality of dietary protein (Deutz et al., 1998; Gaudichonet al., 1999; Mariotti et al., 1999) and type of dietary carbohydrate (van der Meulen et al.,1997) also influence first-pass extraction of amino acids. So the actual availability of dietaryamino acids for muscular protein synthesis is highly dependent on the metabolic activity ofintestinal tissues. It is important to note that this concept is accommodated by our recentadaptation of the IAAO technique mentioned above for determination of true metabolic avail-ability of dietary amino acids from heat-treated feedstuffs.

Given the significant demand of the gut for dietary and arterial amino acids, it is obviousthat the maintenance and growth of this organ already constitutes a significant proportion ofwhole-body amino acid requirements. Furthermore, it follows that in certain situations thatincrease the metabolic activity of the gut, this proportion will increase at the expense ofwhole-body growth. Indeed, this hypothesis is supported by the abovementioned studies insubclinical nematode infection and the reduced growth rate commonly observed in gastro-intestinal disease. If the availability of an indispensable amino acid is already limiting in ananimal’s diet, then an unobservable subclinical challenge to the gut could feasibly limitanimal growth further. Although this concept seems intuitive, it is very difficult to demonstrateexperimentally.

5. IMPACT OF INTESTINAL METABOLISM ON AMINOACID REQUIREMENTS

Many of the approaches recently employed to demonstrate changes in amino acid utilizationby the gut or PDV tissues could be adapted to quantify amino acid requirements of thesetissues relative to whole-body requirements. It has been well demonstrated that the propor-tion of dietary amino acids extracted by the gut changes with dietary composition; but asystematic approach to dietary ingredient requirements for the gut has yet to be published. Byemploying graded intakes of indispensable amino acids, it is possible to determine the mini-mum amount required by the gut for normal function and growth. To date, our body ofliterature regarding amino acid requirements during intravenous or intragastric feeding inpiglets provides the only attempt, to the authors’ knowledge, to quantify requirements withand without first-pass splanchnic metabolism.


5.1. The parenterally fed piglet as a model of splanchnic metabolism

The emerging evidence that the gut utilizes a significant proportion of dietary amino acids ledus to speculate that in situations of gut bypass or stress, the whole-body amino acid require-ments must be different. We proposed that total parenteral nutrition (TPN), which bypassesfirst-pass metabolism by the small intestine and liver and can cause gut atrophy (Johnsonet al., 1975; Goldstein et al., 1985; Alverdy, 1995; Bertolo et al., 1999; Burrin et al., 2000),would lead to changes in amino acid requirements that can be measured. Indeed, we (Duffyand Pencharz, 1986; Bertolo et al., 1999) and others (Sim et al., 1979; Lanza-Jacoby et al.,1982; Jeevanandam et al., 1987) have shown that TPN feeding alters whole-body nitrogenmetabolism compared to oral feeding. The different nitrogen utilization as a consequence ofparenteral feeding may be due to reduced gastrointestinal metabolism associated with gutatrophy and/or due to lack of hepatic first-pass metabolism. With the development of the TPN-fed piglet model (Wykes et al., 1993), we subsequently demonstrated using three infusionroutes that intestinal atrophy has a greater impact on nitrogen metabolism than liver bypass(Bertolo et al., 1999). Whatever the fate of indispensable amino acids in the gut, an atrophiedgut will utilize fewer amino acids and thereby affect whole-body requirements. Therefore, wehave proposed that the TPN-fed piglet is a “gut-deficient” model. So the amino acid require-ments of TPN-fed piglets approximate the requirements for extraintestinal tissues. Whencompared to a piglet gastrically fed identical diets, the amino acid requirements for the intactgut could at least be estimated (table 1).

However, TPN feeding is only one of many relevant clinical scenarios that lead to com-promised gut metabolic capacity. Gut dysfunction can also be caused by malnutrition,diarrhoea, chemotherapy, gastrointestinal surgery and gastrointestinal diseases. Furthermore,with relevance to the animal industry, gut stress is often associated with weaning, especially


Table 1

Effects of gut atrophy and bypass on whole-body amino acid requirements as determined inparenterally and enterally fed piglets using the indicator amino acid oxidation technique

Oral requirement Parenteral requirement Gut bypass effecta

Amino acid (g/kg/d) (g/kg/d) (%)

Threonineb 0.42 0.19 55Methioninec 0.25 0.18 28Total sulphursc 0.42 0.29 31Branched-chaind 2.64 1.53 42Tryptophane 0.13 and 0.11 0.14 0Lysinef 0.85 0.79 7Phenylalanineg 0.50 0.45 10Tyrosineh 0.30 0.35 0

a This effect includes atrophy from 7 d on TPN as well as bypass of first-pass gut metabolism.b Bertolo et al. (1998).c Shoveller et al. (2001). Methionine requirement was determined with excess dietary cysteine; total sulphurs referto the methionine requirement determined with no dietary cysteine.d Elango et al. (2002a). Leucine, isoleucine and valine dietary ratio (NRC, 1998) was maintained across intakes.e Cvitkovic et al. (2000). Tryptophan requirement using intravenous (0.13) or oral isotope (0.11).f House et al. (1998a). Oral requirement for lysine was estimated from NRC (1998).g House et al. (1997b). Parenteral phenylalanine requirement was determined using direct oxidation technique;oral phenylalanine requirement was estimated from NRC (1998).h House et al. (1997a). Oral tyrosine requirement was estimated from NRC (1998).

with the increasingly popular early weaning practice in swine production. Any clinical situa-tion which diminishes gut capacity would affect the metabolism of many amino acids; forexample, increased amino acid requirements discovered during TPN feeding should also beapplied to the treatment of any of the above conditions. Thus, our research has relevance tomany conditions in addition to the TPN-fed neonate. However, we chose to use the TPN-fedpiglet as a model for diminished small intestinal capacity because of demonstrated gutatrophy, reproducibility and ease of development compared to gut disease models. With thisclinically relevant model, we could extrapolate our results to any situation involving gutdysfunction.

5.2. Threonine

Perhaps the most impressive effect of reduced gut metabolism on whole-body amino acidrequirement was demonstrated for threonine. Using the TPN-fed versus gastrically fed piglet,we have demonstrated that the threonine requirement is reduced by 55% when the gut isbypassed and atrophied (Bertolo et al., 1998). These data are supported by Stoll et al. (1998),who also observed that the PDV tissues extract 60% of dietary threonine measured by bothnet portal balance and labelled threonine extraction. In addition, van Goudoever et al. (2000)showed that when pigs were fed the high-protein diet, 84% of threonine was retained by thegut, and on the low-protein diet, all of the threonine was retained. This enormous demand forthreonine by the gut is probably reflected by its role in mucin synthesis for maintenance ofthe luminal mucus layer (Lamont, 1992). Intestinal mucins are continuously secreted by theintestines and are critical in the defence of the mucosa from mechanical and pathogenicinsults. The core protein of mucins contains a disproportionate amount of threonine, prolineand cysteine (Specian and Oliver, 1991). With parenteral feeding and gut atrophy, mucin syn-thesis is reduced and so is the gut’s requirement for threonine. Recently we havedemonstrated that feeding threonine-deficient diets to gastrically fed piglets reduces gutgrowth and goblet cell numbers and alters the mucin profile of intestinal mucus; in addition,parenteral threonine cannot completely restore normal gut function and histology comparedto enteral threonine (Ball et al., 1999; table 2).

This profound impact of gut metabolism on whole-body threonine requirement must also beconsidered as a minimum effect. Because the parenteral threonine requirement (0.19 g/kg/d)was so much lower than NRC (1998) recommendations (0.53 g/kg/d on a true ileal digestiblebasis), we introduced the gastrically infused control group which received identical diets toverify NRC estimates for piglets. The requirement for these control pigs (0.42 g/kg/d) waslower than that recommended by NRC, but still substantially higher than the parenteralrequirement. This discrepancy with NRC values is not surprising given that the requirementsrecommended by NRC (1998) include digestibility estimates for corn–soybean diets and adjustthese values for endogenous loss estimates, whereas our diet was completely elemental andavailable. In subsequent IAAO experiments, we have demonstrated that these NRC estimatesare proportionately closer to our gastrically fed estimates for tryptophan (0.15 vs 0.15 g/kg/dfor NRC), methionine (0.25 vs 0.23 g/kg/d for NRC) and methionine plus cysteine (0.42 vs0.48 g/kg/d for NRC). The discrepancy in threonine requirements between that recommendedby NRC and our oral estimates could be due to the increased sensitivity of the threoninerequirement to endogenous losses. Because the mucin protein core is resistant to digestionand is almost completely recovered in ileal digesta (Mantle and Allen, 1981), a major com-ponent of endogenous losses at the ileum is mucins which are rich in threonine, proline,serine and cysteine (Specian and Oliver, 1991). Mucin secretion and hence losses are known


to be sensitive to dietary composition as well as presence of fibre and anti-nutritional factors(More et al., 1987; Sharma and Schumacher, 1995). Indeed, in a very recent experiment, wehave demonstrated that dietary supplementation of wheat bran, a stimulant of mucin synthe-sis, leads to increased ileal losses of threonine which may affect whole-body availability ofdietary threonine (Myrie et al., 2002). Therefore, this is probably the reason why the threoninerequirement of pigs fed a fibre-free elemental nutrition solution (i.e. our gastrically fed controlpigs) was lower than the estimated requirement of pigs fed a corn–soybean meal diet (i.e. NRCrecommendations). Indeed, as a percentage of the NRC recommendation, the threoninerequirement in parenterally fed pigs was 36% (instead of 45%), which translates to a gututilization of 64% of dietary threonine. Therefore, the requirement for threonine by the gutversus the whole body depends on dietary composition.

5.3. Sulphur amino acids

Although methionine is an indispensable amino acid, cysteine is not because it can be syn-thesized from methionine. However, increased metabolism of methionine to meet cysteineneeds could limit methionine availability for protein synthesis and growth. As a result, dietarycysteine has a “sparing effect” on the amount of methionine required. In a series of experi-ments, we have recently determined the methionine requirements of piglets fed orally andintravenously, with and without dietary cysteine, using the IAAO technique (Shoveller et al.,2001; table 1). With excess or without dietary cysteine, the methionine requirements in par-enterally fed piglets were 72% or 69% of the respective requirements in orally fed piglets. Inother words, approximately 30% of dietary methionine is utilized by the gut whether dietarycysteine is present or not. These data are supported by those of Stoll et al. (1998),


Table 2

Goblet cell parameters in piglets fed adequate oral threonine (IG-A), deficient oral threonine(IG-D) or deficient oral threonine with adequate parenteral threonine (IV-A)a

PAS/AB 2.5b AB 2.5b AB 1.0b

Cellsc Staind Cells Stain Cells Stain

Duodenum IG-A 18.8a 39 17.6a 44a 16.3a 27IG-D 15.2b 41 5.5c 6b 7.1b 12IV-A 14.9b 39 12.6b 25a,b 13.5a,b 31SDpooled 2.7 13 4.4 12 4.4 15

Ileum IG-A 20.9 79 30.1a 63a 24.6a 30IG-D 22.6 68 13.9b 14b 13.3b 13IV-A 27.5 82 19.9a,b 31a,b 20.7a,b 31SDpooled 5.9 18 7.8 18 6.5 11

a Ball et al. (1999). Data are means for n = 7 piglets. For data with letter superscripts within a row, those notsharing a letter are different (P < 0.05, LSD comparisons).b PAS/AB 2.5: staining is combination of Alcian blue/periodic acid (5 min)-Schiff base (15 min) (PAS) reactionallowing unsubstituted α-glycol-rich neutral mucins (pink) and acidic mucins (blue) to be differentiated; AB 2.5:1% Alcian blue (AB, pH 2.5, 1 h) for the localization of carboxylated and/or sulphated acidic mucins; AB 1.0:1% Alcian blue (AB, pH 1.0, 1 h) for the selective identification of sulphomucins.c Goblet cells in the mucosa stained with PAS/AB 2.5, AB 2.5 or AB 1.0 were counted in 10 well-orientedcrypt-villus units ~25 μm in each animal.d Semi-quantitative staining intensities based upon a scale ranging from 0 (unreactive) to 3 (intensely stained)were multiplied by total number of goblet cells.

who showed that approximately 30% of dietary methionine disappears in first-pass meta-bolism by the PDV. Furthermore, the sparing effect of cysteine was similar in pigs fed eitherintravenously or orally (i.e. excess cysteine reduced the respective methionine requirementsby 40% regardless of feeding route) (Shoveller et al., 2001). In addition to demands forprotein turnover, this relatively high demand by the gut for both sulphur amino acids canalso be attributed to other metabolic functions specific to methionine and cysteine. It ispossible that the high nucleic acid turnover of intestinal cells requires a significant amountof methionine, an important methyl donor. Furthermore, cysteine, as a product of methioninemetabolism, is incorporated to a large extent into intestinal mucins and the tripeptide anti-oxidant glutathione; both of these products are critical for the maintenance of the mucosaltissue and protection against pathogens (Martensson et al., 1990). Indeed, the impact of apathogenic challenge on the methionine requirement may therefore be significant, but has yetto be studied.

5.4. Branched-chain amino acids

The aforementioned data regarding substantial first-pass splanchnic metabolism of leucinesuggest that the splanchnic tissues extract a surprisingly large amount (i.e. 20−50%) ofbranched-chain amino acids (BCAA). Recently, using a diet with a fixed ratio of BCAA(1:1.8:1.2, isoleucine:leucine:valine), we have determined that the BCAA requirement inintravenously fed piglets was 56% of that in intragastrically fed piglets (Elango et al., 2002a).The apparent uptake of 44% of enterally fed BCAA by the splanchnic tissues is a significantfinding because it is generally accepted that the BCAA are predominantly metabolized by theextrahepatic tissues due to the higher activity of branched-chain aminotransferase (BCAT),the first enzyme in the catabolic pathway of the BCAA, in skeletal muscle compared to theliver. In addition, the pattern of BCAA in the plasma of enterally fed piglets, when comparedwith parenterally fed piglets, clearly demonstrates that the gut has a high demand for leucineand a clear preference for leucine compared to isoleucine or valine. The observation, duringenteral feeding, that plasma valine and isoleucine concentrations increased while leucineconcentrations remained low indicates that leucine is being extracted by the gut and there-fore may be limiting protein synthesis in the rest of the body. Valine and isoleucine do notappear to be utilized by the gut to the same extent and are therefore being passed to thesystemic circulation, but because protein synthesis is limited by leucine, these two aminoacids, as well as most of the other indispensable amino acids, increase in concentration inthe plasma.

The difference in BCAA requirements between routes of feeding is supported by the datain humans (Gelfand et al., 1988; Hoerr et al., 1991, 1993; Biolo et al., 1992; Matthews et al.,1993) and dogs (Yu et al., 1990, 1995), regarding first-pass splanchnic extraction of leucinemeasured by isotope infusions. In addition, Stoll et al. (1998) reported that the pig PDVextracted 43% of leucine, 39% of valine and 31% of isoleucine. Altogether, the first-passextraction data compare well with the 44% lower BCAA requirement in parenterally fedpiglets observed in our recent study. In a subsequent study, we adapted the IAAO techniqueto systematically determine which of the BCAA was most limiting (Elango et al., 2002b).In both orally and intravenously fed piglets, diets moderately deficient in BCAA (75% ofrespective requirement) were fed and indicator amino acid oxidation determined. Pigletswere then randomly assigned to receive one of three test diets containing either isoleucine,leucine or valine to meet 100% of requirement, with the remaining two amino acids at 75%.


The difference in phenylalanine oxidation between unsupplemented and supplemented dietswas used as an indicator of BCAA adequacy (fig. 3). In orally fed piglets, the difference in per-cent dose oxidized was not significant for any supplemented amino acid. However, inparenterally fed piglets, isoleucine and valine supplementation decreased phenylalanine oxi-dation; isoleucine had the greatest effect and was first limiting (i.e. oxidation decreased fromapproximately 22% to 10%) and valine was second limiting (22% to 15%). Leucine, whichis the preferred amino acid by the gut according to our previous data, had no effect when sup-plemented to gut-atrophied, parenterally fed pigs. The optimal ratio of BCAA for orally fedpigs is adequately predicted by requirement estimates (NRC, 1998). However, because thegut does not utilize the BCAA in this proportion, the ideal ratio of BCAA for maintenance ofthe gut is not the same as that for the whole body and has not yet been determined.

5.5. Tryptophan and lysine

The tryptophan requirements of parenterally and orally fed piglets were not different whenidentical diets were employed (Cvitkovic et al., 2000). This result suggests that the gut’srequirement of tryptophan for protein synthesis and/or for oxidation does not significantlyimpact whole-body requirements, possibly due to either efficient recycling by an atrophiedgut or to this amino acid’s low proportion in protein.

We have also determined the lysine requirement of parenterally fed piglets but did notemploy a gastrically fed control group (House et al., 1998a). Because this was the first aminoacid for which we determined the requirement during TPN feeding, we did not appreciate thelarge impact of the gut. Because NRC estimates were proportionately similar to gastricallyfed control pigs for other amino acids, we therefore compared the parenteral lysine require-ment of 0.79 g/kg/d to the NRC (1998) estimate of 0.85 g/kg/d, providing a 7% differencedue to gut metabolism. It is important to note that this comparison may not be valid given thelack of empirical data for amino acid requirements in young piglets (NRC, 1998). The netlysine utilization by the gut may still be significant as in previous studies regarding the


Fig. 3. The change in phenylalanine oxidation in parenterally (IV) or enterally (IG) fed pigs aftersupplementation of individual amino acids (Elango et al., 2002b). Diets were formulated to be slightlydeficient in all branched-chain amino acids and indicator oxidation was performed before and afterisoleucine, leucine or valine supplementation. * indicates oxidation change was different than zero.

substantial splanchnic extraction of lysine (Hoerr et al., 1993; van Goudoever et al., 2000);however, it appears that in piglets, this utilization is not nearly as profound as that for threo-nine, the sulphur amino acids or the branched-chain amino acids. The impact of parenteralfeeding (i.e. with gut atrophy and lack of first-pass metabolism) on whole-body lysinerequirements is probably accounted for by the reduced general protein turnover in the atro-phied gut; in other words, there seems to be no disproportionate requirement for lysine by thegut versus the whole body.

5.6. Phenylalanine and tyrosine

By employing the direct oxidation approach, we found only moderate differences in thephenylalanine (House et al., 1997b) and tyrosine (House et al., 1997a) requirements ofparenterally fed piglets compared to the NRC (1998) estimates of oral requirements. Theparenteral tyrosine requirement was estimated at 0.35 g/kg/d; in addition, the parenteralphenylalanine requirement (with excess dietary tyrosine) was only 10% lower than NRCestimates (0.45 vs 0.50 g/kg/d). Combined, the phenylalanine plus tyrosine requirementof parenterally fed piglets (0.80 g/kg/d) was equal to that estimated by NRC (0.80 g/kg/d).Furthermore, we have recently been able to compare these results in piglets to the parenteraltyrosine requirement in parenterally fed human infants (Roberts et al., 2001a); the tyrosinerequirement determined for parenterally fed infants was similar to the broad range recom-mended for orally fed neonates (Snyderman, 1971). Again, this minor effect of gutmetabolism on whole-body requirements is in contrast to dual-isotope infusion studies (Bioloet al., 1992; Matthews et al., 1993) where 29–58% of dietary phenylalanine was extracted bysplanchnic tissues in adult humans; however, arterial recirculation (van Goudoever et al.,2000) was not estimated in these studies. In addition, we have shown that when the indicatoramino acid is delivered orally or intravenously, basal phenylalanine oxidation is significantlyincreased when first-pass splanchnic metabolism is maintained in adults (Kriengsinyos et al.,2002) or piglets (Cvitkovic et al., 2000).

However, an increase in phenylalanine oxidation with first-pass metabolism by the gut doesnot necessarily translate to a substantial extraction of dietary phenylalanine, as demonstratedwith lysine by van Goudoever et al. (2000). Indeed, although phenylalanine oxidationincreased by 70% when infused orally versus intravenously in humans, the percent extractiondetermined by flux rates was only increased by 30% (Kriengsinyos et al., 2002). Furthermore,phenylalanine oxidation amounted to less than 15% of intake during either route of infusion.In piglets, when the indicator phenylalanine was fed at the requirement, oxidation rates wereincreased from 0.6% to 0.8% of dose, which translated to 0.9% and 1.5% of phenylalanineintake or requirement (Cvitkovic et al., 2000). These latter data also provide a reason whyphenylalanine is a good choice for the indicator amino acid. Even if there are large differencesbetween diets or individuals in the level of phenylalanine oxidation, the total amount oxidizedis still rather insignificant relative to the quantity used for protein synthesis which is drivingthe whole-body requirement for the amino acid.

As with lysine and tryptophan, it appears that dietary phenylalanine is primarily utilized fornon-specific protein synthesis, which does not appear to disproportionately impact whole-body requirements. It has been suggested that the gut may hydroxylate a substantial amountof phenylalanine to tyrosine (Stoll et al., 1998); however, the impact of this capacity onwhole-body hydroxylation or requirements has yet to be explored. Estimates of phenylalaninehydroxylation to determine tyrosine requirements have been shown to be unreliable (Houseet al., 1998b; Thorpe et al., 2000; Roberts et al., 2001a).


5.7. Arginine and proline

Although for most species studied arginine is considered dispensable, arginine has beenfound to be an indispensable amino acid in some species such as cats (Morris and Rogers,1978), chicks (Tamir and Ratner, 1963) and ferrets (Deshmukh and Shope, 1983). Furthermore,arginine may be conditionally indispensable in young mammals including the dog (Visek,1984), rat (Borman et al., 1946) and piglet (Mertz et al., 1952; Ball et al., 1986; Brunton et al.,1999), which means arginine can be synthesized de novo, but not at sufficient rates to maintainrequired functions (i.e. syntheses of protein, urea cycle intermediates, creatine, nitric oxide, etc.).The neonatal small intestine has been suggested to be the major site of arginine synthesis (Wuet al., 1994; Stoll et al., 1998) and the ontogeny of the necessary enzymes in enterocytes hasbeen well described (Wu, 1998).

We initially planned to use the IAAO technique to determine the arginine requirement ofpiglets during parenteral feeding. An initial pilot experiment was conducted whereby par-enterally fed piglets were fed arginine-free diets so that growth and nitrogen balance datacould be assessed. Both pigs experienced severe hyperammonemia after only ~16 h withoutdietary arginine; one pig died and the other was comatose. Because plasma ammonia con-centration was found to be a sensitive indicator of arginine deficiency, we used this biologicaloutcome to determine if the arginine synthesis rate of the piglet gut was sufficient to main-tain the urea cycle and whether proline, the primary precursor of arginine synthesis in the gut,must be available to maintain synthesis rates. The subsequent study successfully demon-strated that parenterally fed piglets could not synthesize sufficient arginine to maintain theurea cycle, let alone to maintain growth, whether or not proline was present in the diet. Thisstudy also demonstrated that orally fed piglets could not synthesize arginine and proline atrates sufficient to maintain plasma concentrations or to prevent hyperammonemia. However,unlike the gut-atrophied parenterally fed piglets, the gut-intact orally fed piglets experiencedless severe hyperammonemia when proline was provided in the diet. These data suggestedthat the conversion of proline to arginine occurs in the piglet, but only during oral feeding.We therefore hypothesized that this conversion occurs almost exclusively in the gut and thatparenterally fed piglets could not use proline for arginine synthesis because of gut atrophyand/or gut bypass during feeding. This experiment convincingly demonstrated the essentialityof arginine and proline in continuously fed piglets; we predicted that with voluntary feeding,animals would refuse feed if severely deficient, especially if elevated plasma ammonia levelsdevelop. Indeed, vomiting is a symptom of hyperammonemia in pigs, which functions tolessen the ammonia load by expelling potentially toxic amino acids.

This clear demonstration of arginine indispensability in piglets was followed by a multi-isotope, dual-route infusion study whereby labelled proline, ornithine and arginine wereinfused intragastrically or into the portal vein to isolate the in vivo effects of small intestinalfirst-pass metabolism. This experiment demonstrated that the conversion from proline to argi-nine is completely dependent on the small intestine, confirming the conclusions from ourprevious experiment assessing hyperammonemia. Thus, in situations where gut metabolismis bypassed or compromised, such as during TPN feeding or gut disease, arginine synthesisis diminished, increasing the overall requirement compared to normal oral feeding. In addi-tion, this study demonstrated that proline synthesis from arginine is also dependent on gutmetabolism and its requirement also would be higher when gut metabolism is bypassed.Although the cumulative evidence clearly indicates that arginine synthesis is dependent onthe neonatal small intestine, the impact of this first-pass metabolism on whole-body argininerequirements can only be estimated.


Most of these data support the hypothesis that the arginine requirement is much higherduring parenteral feeding, for both maintenance and growth, due to lower synthetic capacityof an atrophied gut. Indeed, we speculate that for arginine, a separate maintenance require-ment can be distinguishable from the growth requirement. Such a hypothesis has enormousimplications in neonatal populations in which small intestinal first-pass metabolism isbypassed or compromised by gastrointestinal disease or stress. Indeed, these implicationshave been demonstrated in parenterally fed infants (Heird et al., 1972) and adult rats withsmall intestinal resection (Wakabayashi et al., 1995). This latter study is interesting consider-ing that adult rats normally can synthesize adequate amounts of arginine in the kidney, but thecitrulline precursor originates in the small intestine (Morris, 1992). So although net intestinalarginine synthesis declines during late suckling as renal synthesis increases (Wu, 1998), theimportance of intestinal metabolism in the inter-organ synthesis of arginine is still potentiallycritical in adult species that normally do not require arginine.

Proline has been suggested to be an indispensable amino acid for the piglet (Ball et al.,1986), but subsequent studies indicated that proline indispensability is dependent on avail-ability of precursors such as glutamate (Murphy et al., 1996; Wu, 1998) and arginine (Bruntonet al., 1999). The extensive gut metabolism of glutamate and glutamine (Windmueller andSpaeth, 1980; Stoll et al., 1999; Reeds et al., 2000) may limit arginine and proline synthesisin certain conditions. Furthermore, the importance of proline for collagen synthesis probablyincreases in situations of injury and stress. The obvious interdependence of arginine and pro-line requirement on gut health as well as availability of precursors makes the quantificationof such requirements very complicated. However, there is enough evidence to date to suggestthat the impact of gut first-pass metabolism on whole-body requirements must be significantand warrants future investigation.


Albeit the parenterally fed piglet model has proved useful in estimating the impact of first-pass gut metabolism on whole-body requirements, a more direct determination of gutrequirements for amino acids has yet to be developed. With a more direct technique,researchers could then attempt to quantify the effects of gastrointestinal stress, injury, diseaseor dysfunction on whole-body requirements. In particular, the recent intriguing work onleucine extraction and nematode infection in sheep provides preliminary evidence for theimportance of this type of investigation. In addition, Klasing and Calvert (1999) have pro-vided an important advance in this area by estimating that the percent of lysine intakeconsumed by the chicken immune system increases from 1.2% to 6.7% with an injectedimmune challenge. This “cost” of an immune challenge must be even more profound with agastrointestinal pathogenic challenge where the additional costs of gut secretion stimulation,gut tissue repair and compromised dietary absorption of lysine must be considered.Furthermore, this cost mostly relates to lysine requirements for non-specific protein syntheticprocesses. The costs for threonine, the branched-chain amino acids, methionine and cysteinewould be much greater due to their more specialized roles in gut maintenance. In addition,arginine and proline requirements would also be increased due to reduced synthesis as wellas increased utilization. The nutritional consequences of gastrointestinal disease and stressrequire further investigation considering that its importance in the treatment of such condi-tions cannot be underestimated.

Because emerging evidence is demonstrating a substantial impact of gut first-pass metab-olism on amino acid requirements, the accommodation of this metabolism will undoubtedly


generate interest, especially among researchers in animal production and clinical treatment ofgut diseases. The role of the native microbial population and subclinical gut infections has animportant impact on whole-body requirements; indeed, the higher amino acid utilization effi-ciency with growth-promoting feed-grade antibiotics may partially be explained by theminimization of subclinical challenges. Considering that the small intestinal capacity todigest and absorb protein and amino acids is substantially greater than possible dietary inputs(Burrin et al., 1999), one is tempted to consider that much of this organ’s demand for aminoacids for maintenance may be an unnecessary burden. Considering that the protein compo-nent, and especially synthetic amino acids, is the most expensive component of animals feeds,the cost of maintaining the surplus capacity of the gut becomes significant. In addition, recentevidence has demonstrated that a significant proportion of whole-body amino acid catabolismoccurs in the gut, presumably for energy. Alternative fuels may be sought to replace this per-haps unnecessary, expensive source of energy. Another compelling solution would be to selectanimals for lower intestinal metabolism without compromising gut absorptive capacity orprotective functions. These animals would have a substantially greater amino acid utilizationefficiency.

The IAAO technique has proven very useful and versatile in determining amino acidrequirements of animals. Its adaptation for vulnerable populations is especially advantageousif amino acid requirements during pathogenic challenges or disease states are to be explored.Furthermore, the development of the technique to determine metabolic availability of aminoacids is an important advance in the field of amino acid digestibility, which is of criticalimportance when providing dietary amino acids to meet requirements. Because the IAAOtechnique relies on relative differences, its dependence on questionable kinetics assumptionsis minimal. This is particularly important given the accumulating evidence that luminal andarterial amino acids are channelled differently intracellularly; the amino acid pools for oxi-dation and protein synthesis are probably separated to some extent so that respectiveprecursor enrichments may be very different, rendering kinetic equations irrelevant. Furtheradaptation of the IAAO technique to answer many of these questions is eagerly sought.

REFERENCES

Alverdy, J.C., 1995. Amino acids to support gut function and morphology. In: Cynober, L.A. (Ed.), AminoAcid Metabolism and Therapy in Health and Nutritional Disease. CRC Press, New York, pp. 435−440.

Baker, D.H., Batal, A.B., Parr, T.M., Augspurger, N.R., Parsons, C.M., 2002. Ideal ratio (relative tolysine) of tryptophan, threonine, isoleucine, and valine for chicks during the second and third weeksposthatch. Poultry Sci. 81, 485−494.

Ball, R.O., Atkinson, J.L., Bayley, H.S., 1986. Proline as an essential amino acid for the young pig.Brit. J. Nutr. 55, 659−668.

Ball, R.O., Bayley, H.S., 1984. Tryptophan requirement of the 2.5-kg piglet determined by the oxidationof an indicator amino acid. J. Nutr. 114, 1741−1746.

Ball, R.O., Bayley, H.S., 1986. Influence of dietary protein concentration on the oxidation of phenylala-nine by the young pig. Brit. J. Nutr. 55, 651−658.

Ball, R.O., Bertolo, R.F.P., Pencharz, P.B., Moehn, S., 2001. A rapid method to determine “true metabolicavailability” of amino acids in feedstuffs for pigs. J. Anim. Sci. 79, Suppl. 1, 66.

Ball, R.O., Law, G., Bertolo, R.F.P., Pencharz, P.B., 1999. Adequate oral threonine is critical for mucinproduction and mucosal growth by the neonatal piglet gut. In: Lobley, G.E., White, A., MacRae, J.C.(Eds.), Protein Metabolism and Nutrition: Book of Abstracts of the VIIIth International Symposium.Wageningen Press, Wageningen, The Netherlands, p. 31.

Ballevre, O., Cadenhead, A., Calder, A.G., Rees, W.D., Lobley, G.E., Fuller, M.F., Garlick, P.J., 1990.Quantitative partition of threonine oxidation in pigs: effect of dietary threonine. Amer. J. Physiol. 259,E483−E491.


Basile-Filho, A., Beaumier, L., el-Khoury, A.E., Yu, Y.M., Kenneway, M., Gleason, R.E., Young, V.R.,1998. Twenty-four-hour L-[1-13C]tyrosine and L-[3,3-2H2]phenylalanine oral tracer studies at generous,intermediate, and low phenylalanine intakes to estimate aromatic amino acid requirements in adults.Amer. J. Clin. Nutr. 67, 640−659.

Basile-Filho, A., el-Khoury, A.E., Beaumier, L., Wang, S.Y., Young, V.R., 1997. Continuous 24-h L-[1-13C]phenylalanine and L-[3,3-2H2]tyrosine oral-tracer studies at an intermediate phenylalanine intake toestimate requirements in adults. Amer. J. Clin. Nutr. 65, 473−488.

Baumann, P.Q., Stirewalt, W.S., O’Rourke, B.D., Howard, D., Nair, K.S., 1994. Precursor pools of pro-tein synthesis: a stable isotope study in a swine model. Amer. J. Physiol. 267, E203−E209.

Bennet, W.M., Haymond, M.W., 1991. Plasma pool source for fibrinogen synthesis in postabsorptiveconscious dogs. Amer. J. Physiol. 260, E581−E587.

Bertolo, R.F.P., Brunton, J.A., Pencharz, P.B., Ball, R.O., 2000a. Steady state is not achieved for mostplasma amino acids during 12 hours of fasting in the neonatal piglet. Pediat. Res. 48, 701−707.

Bertolo, R.F.P., Chen, C.Z.L., Law, G., Pencharz, P.B., Ball, R.O., 1998. Threonine requirement of neonatalpiglets receiving total parenteral nutrition is considerably lower than that of piglets receiving an iden-tical diet intragastrically. J. Nutr. 128, 1752−1759.

Bertolo, R.F.P., Chen, C.Z.L., Pencharz, P.B., Ball, R.O., 1999. Intestinal atrophy has a greater impact onnitrogen metabolism than liver by-pass in piglets fed identical diets via gastric, central venous orportal venous routes. J. Nutr. 129, 1045−1052.

Bertolo, R.F.P., Moehn, S., Pencharz, P.B., Ball, R.O., 2001. Adaptation of indicator amino acid oxidationoccurs within 2 days in 25 or 250 kg pigs fed varying levels of lysine or protein. FASEB J. 15, A266.

Bertolo, R.F.P., Pencharz, P.B., Ball, R.O., 2000b. Organ and plasma amino acid concentrations areprofoundly different in piglets fed identical diets via gastric, central venous or portal venous routes.J. Nutr. 130, 1261−1266.

Biolo, G., Tessari, P., Inchiostro, S., Bruttomesso, D., Fongher, C., Sabadin, L., Fratton, M.G., Valerio, A.,Tiengo, A., 1992. Leucine and phenylalanine kinetics during mixed meal ingestion: a multiple tracerapproach. Amer. J. Physiol. 262, E455−E463.

Borman, A., Wood, T.R., Black, H.C., Anderson, E.G., Osterling, M.J., Womack, M., Rose, W.C., 1946.The role of arginine in growth with some observations on the effect of argininic acid. J. Biol. Chem.166, 585−594.

Brookes, I.M., Owens, F.N., Garrigus, U.S., 1972. Influence of amino acid level in the diet upon aminoacid oxidation by the rat. J. Nutr. 102, 27−35.

Bross, R., Ball, R.O., Clarke, J.T., Pencharz, P.B., 2000. Tyrosine requirements in children with classical PKUdetermined by indicator amino acid oxidation. Amer. J. Physiol. Endocrinol. Metab. 278, E195−E201.

Bross, R., Ball, R.O., Pencharz, P.B., 1998. Development of a minimally invasive protocol for thedetermination of phenylalanine and lysine kinetics in humans during the fed state. J. Nutr. 128,1913−1919.

Bruins, M.J., Soeters, P.B., Deutz, N.E., 2000. Endotoxemia affects organ protein metabolism differentlyduring prolonged feeding in pigs. J. Nutr. 130, 3003−3013.

Brunton, J.A., Ball, R.O., Pencharz, P.B., 1998. Determination of amino acid requirements byindicator amino acid oxidation: applications in health and disease. Curr. Opin. Clin. Nutr. Metab. Care1, 449−453.

Brunton, J.A., Bertolo, R.F.P., Pencharz, P.B., Ball, R.O., 1999. Proline ameliorates arginine deficiencyduring enteral but not parenteral feeding in neonatal piglets. Amer. J. Physiol. 277, E223−E231.

Burrin, D.G., Ferrell, C.L., Britton, R.A., Bauer, M., 1990. Level of nutrition and visceral organ size andmetabolic activity in sheep. Brit. J. Nutr. 64, 439−448.

Burrin, D.G., Stoll, B., Chang, X., Yu, H., Reeds, P.J., 1999. Total parenteral nutrition does not compro-mise digestion or absorption of dietary protein in neonatal pigs. FASEB J. 13, A1024.

Burrin, D.G., Stoll, B., Jiang, R., Chang, X., Hartmann, B., Holst, J.J., Greeley, G.H. Jr., Reeds, P.J., 2000.Minimal enteral nutrient requirements for intestinal growth in neonatal piglets: how much is enough?Amer. J. Clin. Nutr. 71, 1603−1610.

Cappelli, F.P., Seal, C.J., Parker, D.S., 1997. Glucose and [13C]leucine metabolism by the portal-drainedviscera of sheep fed on dried grass with acute intravenous and intraduodenal infusions of glucose.Brit. J. Nutr. 78, 931−946.

Caso, G., Ford, G.C., Nair, K.S., Vosswinkel, J.A., Garlick, P.J., McNurlan, M.A., 2001. Increased con-centration of tracee affects estimates of muscle protein synthesis. Amer. J. Physiol. Endocrinol.Metab. 280, E937−E946.


Cayol, M., Boirie, Y., Prugnaud, J., Gachon, P., Beaufrere, B., Obled, C., 1996. Precursor pool for hepaticprotein synthesis in humans: effects of tracer route infusion and dietary proteins. Amer. J. Physiol.270, E980−E987.

Chavez, E.R., Bayley, H.S., 1976. The oxidation in vivo of uniformly and carboxyl-14C-labelled methio-nine by young pigs. Proc. Nutr. Soc. 35, 150A−151A.

Chinkes, D.L., Rosenblatt, J., Wolfe, R.R., 1993. Assessment of the mathematical issues involved inmeasuring the fractional synthesis rate of protein using the flooding dose technique. Clin. Sci. (Lond.)84, 177−183.

Cobelli, C., Saccomani, M.P., Tessari, P., Biolo, G., Luzi, L., Matthews, D.E., 1991. Compartmentalmodel of leucine kinetics in humans. Amer. J. Physiol. 261, E539−E550.

Coleman, R.A., Korver, D.R., Bertolo, R.F., Ball, R.O., 2002. Lysine requirements of pre-lay broilerbreeders as determined by the indicator amino acid oxidation method. Presented at Poultry ScienceAssociation 91st Annual Meeting, Newark, Delaware.

Cvitkovic, S., Pencharz, P.B., Ball, R.O., 2000. Comparison of oral and intravenous isotopic tracers indetermining tryptophan requirements of piglets. FASEB J. 14, A745.

Deshmukh, D.R., Shope, T.C., 1983. Arginine requirement and ammonia toxicity in ferrets. J. Nutr. 113,1664−1667.

Deutz, N.E., Bruins, M.J., Soeters, P.B., 1998. Infusion of soy and casein protein meals affects inter-organ amino acid metabolism and urea kinetics differently in pigs. J. Nutr. 128, 2435−2445.

Di Buono, M., Wykes, L.J., Ball, R.O., Pencharz, P.B., 2001a. Total sulfur amino acid requirement inyoung men as determined by indicator amino acid oxidation with L-[1-13C]phenylalanine. Amer. J. Clin.Nutr. 74, 756−760.

Di Buono, M., Wykes, L.J., Ball, R.O., Pencharz, P.B., 2001b. Dietary cysteine reduces the methioninerequirement in men. Amer. J. Clin. Nutr. 74, 761−766.

Duffy, B., Pencharz, P.B., 1986. The effect of feeding route (IV or oral) on the protein metabolism of theneonate. Amer. J. Clin. Nutr. 43, 108−111.

Duncan, A.M., Ball, R.O., Pencharz, P.B., 1996. Lysine requirement of adult males is not affected bydecreasing dietary protein. Amer. J. Clin. Nutr. 64, 718−725.

Elango, R., Pencharz, P.B., Ball, R.O., 2002a. The branched-chain amino acid requirement of parenter-ally fed neonatal piglets is less than the enteral requirement. J. Nutr. 132, 3123−3129.

Elango, R., Pencharz, P.B., Ball, R.O., 2002b. The optimum ratio of branched-chain amino acids (BCAA)differs between intragastric and intravenous routes of feeding in the neonatal piglet. FASEB J.16, A746.

el-Khoury, A.E., Basile, A., Beaumier, L., Wang, S.Y., Al-Amiri, H.A., Selvaraj, A., Wong, S., Atkinson, A.,Ajami, A.M., Young, V.R., 1998. Twenty-four-hour intravenous and oral tracer studies withL-[1-13C]-2-aminoadipic acid and L-[1-13C]lysine as tracers at generous nitrogen and lysine intakes inhealthy adults. Amer. J. Clin. Nutr. 68, 827−839.

el-Khoury, A.E., Fukagawa, N.K., Sanchez, M., Tsay, R.H., Gleason, R.E., Chapman, T.E., Young, V.R.,1994a. Validation of the tracer-balance concept with reference to leucine: 24-h intravenous tracerstudies with L-[1-13C]leucine and [15N-15N]urea. Amer. J. Clin. Nutr. 59, 1000−1011.

el-Khoury, A.E., Fukagawa, N.K., Sanchez, M., Tsay, R.H., Gleason, R.E., Chapman, T.E., Young, V.R.,1994b. The 24-h pattern and rate of leucine oxidation, with particular reference to tracer estimates ofleucine requirements in healthy adults. Amer. J. Clin. Nutr. 59, 1012−1020.

el-Khoury, A.E., Pereira, P.C., Borgonha, S., Basile-Filho, A., Beaumier, L., Wang, S.Y., Metges, C.C.,Ajami, A.M., Young, V.R., 2000. Twenty-four-hour oral tracer studies with L-[1-13C]lysine at a low(15 mg kg−1.d−1) and intermediate (29 mg kg−1.d−1)) lysine intake in healthy adults. Amer. J. Clin. Nutr.72, 122−130.

el-Khoury, A.E., Sanchez, M., Fukagawa, N.K., Gleason, R.E., Tsay, R.H., Young, V.R., 1995. The 24-hkinetics of leucine oxidation in healthy adults receiving a generous leucine intake via three discretemeals. Amer. J. Clin. Nutr. 62, 579−590.

FAO/WHO/UNU Expert Consultation, 1985. Energy and protein requirements. World HealthOrganization, Geneva. Tech. Rep. Ser. No. 724.

Fuller, M.F., Garlick, P.J., 1994. Human amino acid requirements: can the controversy be resolved?Annu. Rev. Nutr. 14, 217−241.

Gaudichon, C., Mahe, S., Benamouzig, R., Luengo, C., Fouillet, H., Dare, S., Van Oycke, M., Ferriere, F.,Rautureau, J., Tome, D., 1999. Net postprandial utilization of [15N]-labeled milk protein nitrogen isinfluenced by diet composition in humans. J. Nutr. 129, 890−895.


Gelfand, R.A., Glickman, M.G., Castellino, P., Louard, R.J., DeFronzo, R.A., 1988. Measurement ofL-[1-14C]leucine kinetics in splanchnic and leg tissues in humans: effect of amino acid infusion.Diabetes 37, 1365−1372.

Goldstein, R.M., Hebiguchi, T., Luk, G.D., Taqi, F., Guilarte, T.R., Franklin, F.A., Niemiec, P.W.,Dudgeon, D.L., 1985. The effects of total parenteral nutrition on gastrointestinal growth and devel-opment. J. Pediat. Surg. 20, 785−791.

Harper, A.E., Benjamin, E., 1984. Relationship between intake and rate of oxidation of leucine andα-ketoisocaproate in vivo in the rat. J. Nutr. 114, 431-440.

Heird, W.C., Nicholson, J.F., Driscoll, J.M., Schullinger, J.N., Winters, R.W., 1972. Hyperammonemiaresulting from intravenous alimentation using a mixture of synthetic L-amino acids: a preliminaryreport. J. Pediat. 81, 162−165.

Higashiguchi, T., Noguchi, Y., Noffsinger, A., Fischer, J.E., Hasselgren, P.O., 1994a. Sepsis increasesproduction of total secreted proteins, vasoactive intestinal peptide, and peptide YY in isolated rat ente-rocytes. Amer. J. Surg. 168, 251−256.

Higashiguchi, T., Noguchi, Y., O’Brien, W., Wagner, K., Fischer, J.E., Hasselgren, P.O., 1994b. Effect ofsepsis on mucosal protein synthesis in different parts of the gastrointestinal tract in rats. Clin. Sci.(Lond.) 87, 207−211.

Hoerr, R.A., Matthews, D.E., Bier, D.M., Young, V.R., 1991. Leucine kinetics from [2H3]- and[13C]leucine infused simultaneously by gut and vein. Amer. J. Physiol. 260, E111−E117.

Hoerr, R.A., Matthews, D.E., Bier, D.M., Young, V.R., 1993. Effects of protein restriction and acuterefeeding on leucine and lysine kinetics in young men. Amer. J. Physiol. 264, E567−E575.

House, J.D., Pencharz, P.B., Ball, R.O., 1997a. Tyrosine kinetics and requirements during total parenteralnutrition in the neonatal piglet: the effect of glycyl-L-tyrosine supplementation. Pediat. Res. 41, 575−583.

House, J.D., Pencharz, P.B., Ball, R.O., 1997b. Phenylalanine requirements determined by usingL-[1-14C]phenylalanine in neonatal piglets receiving total parenteral nutrition supplemented with tyro-sine. Amer. J. Clin. Nutr. 65, 984−993.

House, J.D., Pencharz, P.B., Ball, R.O., 1998a. Lysine requirement of neonatal piglets receiving total par-enteral nutrition as determined by oxidation of the indicator amino acid L-[1-14C]phenylalanine.Amer. J. Clin. Nutr. 67, 67−73.

House, J.D., Thorpe, J.M., Wykes, L.J., Pencharz, P.B., Ball, R.O., 1998b. Evidence that phenylalaninehydroxylation rates are overestimated in neonatal subjects receiving total parenteral nutrition with ahigh phenylalanine content. Pediat. Res. 43, 461−466.

Jeevanandam, M., Lowry, S.F., Brennan, M.F., 1987. Effect of the route of nutrient administration onwhole-body protein kinetics in man. Metabolism 36, 968−973.

Johnson, L.R., Copeland, E.M., Dudrick, S.J., Lichtenberger, L.M., Castro, G.A., 1975. Structural and hor-monal alterations in the gastrointestinal tract of parenterally fed rats. Gastroenterology 68, 1177−1183.

Kang-Lee, Y.A., Harper, A.E., 1977. Effect of histidine intake and hepatic histidase activity on the metab-olism of histidine in vivo. J. Nutr. 107, 1427−1443.

Kang-Lee, Y.A., Harper, A.E., 1978. Threonine metabolism in vivo: effect of threonine intake and priorinduction of threonine dehydratase in rats. J. Nutr. 108, 163−175.

Kim, K.I., Bayley, H.S., 1983. Amino acid oxidation by young pigs receiving diets with varying levels ofsulphur amino acids. Brit. J. Nutr. 50, 383−390.

Kim, K.I., Elliott, J.I., Bayley, H.S., 1983a. Oxidation of an indicator amino acid by young pigs receiv-ing diets with varying levels of lysine or threonine, and an assessment of amino acid requirements.Brit. J. Nutr. 50, 391−399.

Kim, K.I., McMillan, I., Bayley, H.S., 1983b. Determination of amino acid requirements of young pigsusing an indicator amino acid. Brit. J. Nutr. 50, 369−382.

Klasing, K.C., Calvert, C.C., 1999. The care and feeding of an immune system: an analysis of lysineneeds. In: Lobley, G.E., White, A., MacRae, J.C. (Eds.), Protein Metabolism and Nutrition:Proceedings of the VIIIth International Symposium. Wageningen Press, Wageningen, TheNetherlands, pp. 253−264.

Kriengsinyos, W., Wykes, L.J., Ball, R.O., Pencharz, P.B., 2002. Effect of oral and intravenous tracer proto-cols on the estimate of lysine requirement in healthy adult males L-[1-13C]phenylalanine as an indicatoramino acid. J. Nutr. 132, 2251–2257.

Kurpad, A.V., Raj, T., el-Khoury, A., Beaumier, L., Kuriyan, R., Srivatsa, A., Borgonha, S., Selvaraj, A.,Regan, M.M., Young, V.R., 2001. Lysine requirements of healthy adult Indian subjects, measured byan indicator amino acid balance technique. Amer. J. Clin. Nutr. 73, 900−907.


Kurpad, A.V., Raj, T., Regan, M.M., Vasudevan, J., Caszo, B., Nazareth, D., Gnanou, J., Young, V.R.,2002a. Threonine requirements of healthy Indian men, measured by a 24-h indicator amino acidoxidation and balance technique. Amer. J. Clin. Nutr. 76, 789−797.

Kurpad, A.V., Regan, M.M., Raj, T., el-Khoury, A., Kuriyan, R., Vaz, M., Chandakudlu, D.,Venkataswamy, V.G., Borgonha, S., Young, V.R., 2002b. Lysine requirements of healthy adult Indiansubjects receiving long-term feeding, measured with a 24-h indicator amino acid oxidation andbalance technique. Amer. J. Clin. Nutr. 76, 404−412.

Lamont, J., 1992. Mucus: the front line of intestinal mucosal defense. Ann. N.Y. Acad. Sci. 664, 190−201.Lanza-Jacoby, S., Sitren, H.S., Stevenson, N.R., Rosato, F.E., 1982. Changes in circadian rhythmicity of

liver and serum parameters in rats fed a total parenteral nutrition solution by continuous and discon-tinuous intravenous or intragastric infusion. J. Parent. Enter. Nutr. 6, 496−502.

Lazaris-Brunner, G., Rafii, M., Ball, R.O., Pencharz, P.B., 1998. Tryptophan requirement in young adultwomen as determined by indicator amino acid oxidation with L-[13C]phenylalanine. Amer. J. Clin.Nutr. 68, 303−310.

Lewis, A.J., 1992. Determination of the amino acid requirements of animals. In: Nissen, S. (Ed.), ModernMethods in Protein Nutrition and Metabolism. Academic Press, San Diego, pp. 67−86.

Lobley, G.E., Milne, V., Lovie, J.M., Reeds, P.J., Pennie, K., 1980. whole-body and tissue proteinsynthesis in cattle. Brit. J. Nutr. 43, 491−502.

MacRae, J.C., Bruce, L.A., Brown, D.S., Calder, A.G., 1997. Amino acid use by the gastrointestinal tractof sheep given lucerne forage. Amer. J. Physiol. 273, G1200−G1207.

Mager, D.R., Ball, R.O., Pencharz, P.B., 2002. The effect of orthotopic liver transplantation (OLT) ontotal branched-chain amino acid requirements in children as determined by indicator amino acidoxidation (IAAO). FASEB J. 16, A259.

Mager, D.R., Pencharz, P.B., Ball, R.O., 2001. Branchedchain amino acid requirements in school agedchildren determined by indicator amino acid oxidation (IAAO). FASEB J. 15, A266.

Mantle, M., Allen, A., 1981. Isolation and characterization of the native glycoprotein from pig small-intestinal mucus. Biochem J. 195, 267−275.

Mariotti, F., Mahe, S., Benamouzig, R., Luengo, C., Dare, S., Gaudichon, C., Tome, D., 1999. Nutritionalvalue of [15N]-soy protein isolate assessed from ileal digestibility and postprandial protein utilizationin humans. J. Nutr. 129, 1992−1997.

Martensson, J., Jain, A., Meister, A., 1990. Glutathione is required for intestinal function. Proc. Natl.Acad. Sci. USA 87, 1715−1719.

Matthews, D.E., Marano, M.A., Campbell, R.G., 1993. Splanchnic bed utilization of leucine and phe-nylalanine in humans. Amer. J. Physiol. 264, E109−E118.

Mavromichalis, I., Webel, D.M., Emmert, J.L., Moser, R.L., Baker, D.H., 1998. Limiting order of aminoacids in a low-protein corn-soybean meal-whey-based diet for nursery pigs. J. Anim. Sci. 76, 2833−2837.

McNurlan, M.A., Garlick, P.J., 1980. Contribution of rat liver and gastrointestinal tract to whole-bodyprotein synthesis in the rat. Biochem. J. 186, 381−383.

Meguid, M.M., Matthews, D.E., Bier, D.M., Meredith, C.N., Soeldner, J.S., Young, V.R., 1986a. Leucinekinetics at graded leucine intakes in young men. Amer. J. Clin. Nutr. 43, 770−780.

Meguid, M.M., Matthews, D.E., Bier, D.M., Meredith, C.N., Young, V.R., 1986b. Valine kinetics atgraded valine intakes in young men. Amer. J. Clin. Nutr. 43, 781−786.

Meredith, C.N., Wen, Z.M., Bier, D.M., Matthews, D.E., Young, V.R., 1986. Lysine kinetics at gradedlysine intakes in young men. Amer. J. Clin. Nutr. 43, 787−794.

Mertz, E.T., Beeson, W.M., Jackson, H.D., 1952. Classification of essential amino acids for the weanlingpigs. Arch. Biochem. Biophys. 38, 121−128.

Moehn, S., Bertolo, R.F.P., Ball, R.O., 2001. A method to measure the amino acid requirement of indi-vidual pigs. J. Anim. Sci. 79, Suppl. 1, 66.

More, J., Fioramonti, J., Benazet, F., Bueno, L., 1987. Histochemical characterization of glycoproteinspresent in jejunal and colonic goblet cells of pigs on different diets. Histochemistry 87, 189−194.

Morris, J.G., Rogers, Q.R., 1978. Arginine: an essential amino acid for the cat. J. Nutr. 108, 1944−1953.Morris, S.M., 1992. Regulation of enzymes of urea and arginine synthesis. Annu. Rev. Nutr. 12, 81−101.Motil, K.J., Opekun, A.R., Montandon, C.M., Berthold, H.K., Davis, T.A., Klein, P.D., Reeds, P.J., 1994.

Leucine oxidation changes rapidly after dietary protein intake is altered in adult women but lysineflux is unchanged as is lysine incorporation into VLDL-apolipoprotein B-100. J. Nutr. 124, 41−51.

Murphy, J.M., Murch, S.J., Ball, R.O., 1996. Proline is synthesized from glutamate during intragastricinfusion but not during intravenous infusion in neonatal piglets. J. Nutr. 126, 878−886.


Myrie, S.B., Bertolo, R.F.P., Sauer, W.C., Ball, R.O., 2002. Will anti-nutritive factors in foods stimulateincreased intestinal mucin secretion and increase threonine requirement? FASEB J. 16, A260.

NRC (National Research Council), 1979. Nutrient Requirements of Swine (8th Edition). NationalAcademy Press, Washington, DC.

NRC (National Research Council), 1998. Nutrient Requirements of Swine (10th Edition). NationalAcademy Press, Washington, DC.

Newport, M.J., Chavez, E.R., Horney, F.D., Bayley, H.S., 1976. Amino acid metabolism in the piglet:influence of level of protein and of methionine in the diet on tissue uptake and in vivo oxidation.Brit. J. Nutr. 36, 87−99.

Raguso, C.A., Pereira, P., Young, V.R., 1999. A tracer investigation of obligatory oxidative amino acidlosses in healthy, young adults. Amer. J. Clin. Nutr. 70, 474−483.

Rand, W.M., Young, V.R., 1999. Statistical analysis of nitrogen balance data with reference to the lysinerequirement in adults. J. Nutr. 129, 1920−1926.

Reeds, P.J., Burrin, D.G., Stoll, B., Jahoor, F., 2000. Intestinal glutamate metabolism. J. Nutr. 130,978S−982S.

Reeds, P.J., Hachey, D.L., Patterson, B.W., Motil, K.J., Klein, P.D., 1992. VLDL apolipoprotein B-100,a potential indicator of the isotopic labeling of the hepatic protein synthetic precursor pool in humans:studies with multiple stable isotopically labeled amino acids. J. Nutr. 122, 457−466.

Riazi, R., Ball, R.O., Pencharz, P.B., 2001. Total branched-chain amino acid requirement in younghealthy adult men determined by indicator amino acid oxidation using L-[1-13C]phenylalanine.FASEB J. 15, A266.

Roberts, S.A., Ball, R.O., Moore, A.M., Filler, R.M., Pencharz, P.B., 2001a. The effect of graded intakeof glycyl-L-tyrosine on phenylalanine and tyrosine metabolism in parenterally fed neonates with anestimation of tyrosine requirement. Pediat. Res. 49, 111−119.

Roberts, S.A., Thorpe, J.M., Ball, R.O., Pencharz, P.B., 2001b. Tyrosine requirement of healthymen receiving a fixed phenylalanine intake determined by using indicator amino acid oxidation.Amer. J. Clin. Nutr. 73, 276−282.

Sanchez, M., el-Khoury, A.E., Castillo, L., Chapman, T.E., Filho, A.B., Beaumier, L., Young, V.R.,1996. Twenty-four-hour intravenous and oral tracer studies with L-[1-13C]phenylalanine andL-[3,3-2H2]tyrosine at a tyrosine-free, generous phenylalanine intake in adults. Amer. J. Clin. Nutr. 63,532−545.

Sanchez, M., el-Khoury, A.E., Castillo, L., Chapman, T.E., Young, V.R., 1995. Phenylalanine andtyrosine kinetics in young men throughout a continuous 24-h period, at a low phenylalanine intake.Amer. J. Clin. Nutr. 61, 555−570.

Sharma, R., Schumacher, U., 1995. The influence of diets and gut microflora on lectin binding patternsof intestinal mucins in rats. Lab. Invest. 73, 558−564.

Shoveller, A.K., Brunton, J.A., Bertolo, R.F.P., Pencharz, P.B., Ball, R.O., 2001. Intestinal metabolism ofmethionine significantly affects requirement and proportion spared by cysteine. In: Lindberg, J.E.,Ogle, B. (Eds.), Digestive Physiology of Pigs: Proceedings of the 8th Symposium. CABI Publishing,New York, pp. 89−91.

Sim, A.J., Wolfe, B.M., Young, V.R., Clarke, D., Moore, F.D., 1979. Glucose promotes whole-bodyprotein synthesis from infused aminoacids in fasting man: isotopic demonstration. Lancet 1(8107), 68−72.

Simon, O., Adam, K., Bergner, H., 1978. [Metabolism-oriented lysine requirement of mature rats basedon the catabolism rate for 14C- and 15N-labelled lysine]. Archiv. Tierernähr. 28, 609−617.

Snyderman, S.E., 1971. The protein and amino acid requirements of the premature infant. In:Jonxis, J.H.P., Visser, H.K.A., Troelstra, J.A. (Eds.), Metabolic Processes in the Foetus and Newborn.Stenfert Kroese, Leiden, The Netherlands, pp. 128−141.

Specian, R.D., Oliver, M.G., 1991. Functional biology of intestinal goblet cells. Amer. J. Physiol. 260,C183−C193.

Stoll, B., Burrin, D.G., Henry, J., Jahoor, F., Reeds, P.J., 1997. Phenylalanine utilization by the gut andliver measured with intravenous and intragastric tracers in pigs. Amer. J. Physiol. 273, G1208−G1217.

Stoll, B., Burrin, D.G., Henry, J., Yu, H., Jahoor, F., Reeds, P.J., 1999. Substrate oxidation by the portaldrained viscera of fed piglets. Amer. J. Physiol. 277, E168−E175.

Stoll, B., Henry, J., Reeds, P.J., Yu, H., Jahoor, F., Burrin, D.G., 1998. Catabolism dominates the first-pass intestinal metabolism of dietary essential amino acids in milk protein-fed piglets. J. Nutr. 128,606−614.


Storch, K.J., Wagner, D.A., Burke, J.F., Young, V.R., 1988. Quantitative study in vivo of methionine cyclein humans using [methyl-2H3]- and [1-13C]methionine. Amer. J. Physiol. 255, E322−E331.

Tamir, H., Ratner, S., 1963. A study of ornithine, citrulline and arginine synthesis in growing chicks.Arch. Biochem. Biophys. 102, 259−269.

Thorpe, J.M., Roberts, S.A., Ball, R.O., Pencharz, P.B., 2000. Effect of tyrosine intake on the rate ofphenylalanine hydroxylation in adult males. Metabolism 49, 444−449.

Tome, D., Bos, C., 2000. Dietary protein and nitrogen utilization. J. Nutr. 130, 1868S−1873S.van der Meulen, J., Bakker, J.G., Smits, B., de Visser, H., 1997. Effects of source of starch on net portal

flux of glucose, lactate, volatile fatty acids and amino acids in the pig. Brit. J. Nutr. 78, 533−544.van Goudoever, J.B., Stoll, B., Henry, J.F., Burrin, D.G., Reeds, P.J., 2000. Adaptive regulation of intes-

tinal lysine metabolism. Proc. Natl. Acad. Sci. USA 97, 11620−11625.Visek, W.J., 1984. An update of concepts of essential amino acids. Annu. Rev. Nutr. 4, 137−155.von Allmen, D., Hasselgren, P.O., Higashiguchi, T., Frederick, J., Zamir, O., Fischer, J.E., 1992.

Increased intestinal protein synthesis during sepsis and following the administration of tumour necrosisfactor α or interleukin-1α. Biochem. J. 286, 585−589.

Wakabayashi, Y., Yamada, E., Yoshida, T., Takahashi, N., 1995. Effect of intestinal resection and arginine-free diet on rat physiology. Amer. J. Physiol. 269, G313−G318.

Waterlow, J.C., 1985. What do we mean by adaptation? In: Blaxter, K.L., Waterlow, J.C. (Eds.),Nutritional Adaptation in Man. John Libbey, London, pp. 1−10.

Waterlow, J.C., 1999. The mysteries of nitrogen balance. Nutr. Res. Rev. 12, 25−54.Waterlow, J.C., Garlick, P.J., Millward, D.J., 1978. Protein Turnover in Mammalian Tissues and in the

Whole Body. Elsevier/North-Holland Biomedical Press, Amsterdam.Watt, P.W., Lindsay, Y., Scrimgeour, C.M., Chien, P.A., Gibson, J.N., Taylor, D.J., Rennie, M.J., 1991.

Isolation of aminoacyl-tRNA and its labeling with stable-isotope tracers: use in studies of humantissue protein synthesis. Proc. Natl. Acad. Sci. USA 88, 5892−5896.

Were, K.E., 1989. Determination of the tryptophan requirement of the juvenile rainbow trout using theindicator oxidation technique. M.Sc. Thesis, University of Guelph, Guelph, Canada.

Wilson, D.C., Rafii, M., Ball, R.O., Pencharz, P.B., 2000. Threonine requirement of young mendetermined by indicator amino acid oxidation with use of L-[1-13C]phenylalanine. Amer. J. Clin. Nutr.71, 757−764.

Windmueller, H.G., Spaeth, A.E., 1980. Respiratory fuels and nitrogen metabolism in vivo in small intes-tine of fed rats: quantitative importance of glutamine, glutamate, and aspartate. J. Biol. Chem. 255,107−112.

Wolfe, R.R., 1992. Radioactive and Stable Isotope Tracers in Biomedicine: Principles and Practice ofKinetic Analysis. Wiley-Liss, Toronto, pp. 357−376.

Wu, G., 1998. Amino acid metabolism in the small intestine. Trends Comp. Biochem. Physiol. 4, 39−74.Wu, G., Borbolla, A.G., Knabe, D.A., 1994. The uptake of glutamine and release of arginine, citrulline

and proline by the small intestine of developing pigs. J. Nutr. 124, 2437−2444.Wykes, L.J., Ball, R.O., Pencharz, P.B., 1993. Development and validation of a total parenteral nutrition

model in the neonatal piglet. J. Nutr. 123, 1248−1259.Young, V.R., Bier, D.M., Pellett, P.L., 1989. A theoretical basis for increasing current estimates of the

amino acid requirements in adult man, with experimental support. Amer. J. Clin. Nutr. 50, 80−92.Young, V.R., Borgonha, S., 2000. Nitrogen and amino acid requirements: the Massachusetts Institute of

Technology amino acid requirement pattern. J. Nutr. 130, 1841S−1849S.Young, V.R., el-Khoury, A.E., 1995. Can amino acid requirements for nutritional maintenance in adult

humans be approximated from the amino acid composition of body mixed proteins? Proc. Natl. Acad.Sci. USA. 92, 300−304.

Young, V.R., Marchini, J.S., 1990. Mechanisms and nutritional significance of metabolic responses toaltered intakes of protein and amino acids, with reference to nutritional adaptation in humans.Amer. J. Clin. Nutr. 51, 270−289.

Yu, F., Bruce, L.A., Calder, A.G., Milne, E., Coop, R.L., Jackson, F., Horgan, G.W., MacRae, J.C.,2000. Subclinical infection with the nematode Trichostrongylus colubriformis increases gastroin-testinal tract leucine metabolism and reduces availability of leucine for other tissues. J. Anim. Sci.78, 380−390.

Yu, Y.M., Burke, J.F., Vogt, J.A., Chambers, L., Young, V.R., 1992. Splanchnic and whole bodyL-[1-13C,15N]leucine kinetics in relation to enteral and parenteral amino acid supply. Amer. J. Physiol.262, E687−E694.


Yu, Y.M., Wagner, D.A., Tredget, E.E., Walaszewski, J.A., Burke, J.F., Young, V.R., 1990. Quantitativerole of splanchnic region in leucine metabolism: L-[1-13C,15N]leucine and substrate balance studies.Amer. J. Physiol. 259, E36−E51.

Yu, Y.M., Young, V.R., Tompkins, R.G., Burke, J.F., 1995. Comparative evaluation of the quantitativeutilization of parenterally and enterally administered leucine and L-[1-13C,15N]leucine within thewhole-body and the splanchnic region. J. Parent. Enter. Nutr. 19, 209−215.

Zello, G.A., Pencharz, P.B., Ball, R.O., 1990. Phenylalanine flux, oxidation, and conversion to tyrosinein humans studied with L-[1-13C]phenylalanine. Amer. J. Physiol. 259, E835−E843.

Zello, G.A., Pencharz, P.B., Ball, R.O., 1993. Dietary lysine requirement of young adult males deter-mined by oxidation of L-[1-13C]phenylalanine. Amer. J. Physiol. 264, E677−E685.

Zello, G.A., Wykes, L.J., Ball, R.O., Pencharz, P.B., 1995. Recent advances in methods of assessingdietary amino acid requirements for adult humans. J. Nutr. 125, 2907−2915.

Zhao, X.H., Wen, Z.M., Meredith, C.N., Matthews, D.E., Bier, D.M., Young, V.R., 1986. Threonine kineticsat graded threonine intakes in young men. Amer. J. Clin. Nutr. 43, 795−802.


157

7 Splanchnic protein and amino acidmetabolism in growing animals1

D. G. Burrin and B. Stoll

USDA/ARS Children’s Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine, 1100 Bates Street, Houston, TX 77030, USA

The splanchnic tissues, namely liver and gut, play a major role in the regulation of whole-body protein and amino acid metabolism. Given their anatomical design for assimilation offood by the host, these tissues metabolize in first-pass a significant proportion of the dietaryamino acids via protein synthesis and oxidation and thereby limit the quantity and alter thepattern of amino acids for systemic availability. This substantial “metabolic cost” incurred bysplanchnic tissues is in large part related to the numerous critical physiological functions theyperform for the mammalian host, such as digestion, ureagenesis, gluconeogensis, and acute-phase protein synthesis. Splanchnic tissues also play a key regulatory role by transmittingendocrine, immune, and neural signals in response to the diet and environment, which in turndetermine the rates of peripheral tissue protein metabolism and growth. Splanchnic tissueprotein metabolism is regulated by specific amino acids and hormones. Amino acids functionnot only as substrates, but also as extracellular signals that influence cell functions, such asprotein turnover, proliferation, apoptosis, cell volume, and redox status. Many of the intra-cellular signaling and biochemical pathways involved in amino acid metabolism have beendescribed in the liver, but less is known about the gut tissues. The beneficial health effects ofkey immunonutrients are mediated by improved cell functions in different splanchnic tissues.

1. INTRODUCTION

Numerous studies with growing mammals, ruminant and nonruminant, have establishedthat the splanchnic tissues have a substantial impact on whole-body protein and amino



1 The authors thank Jane Schoppe for her assistance in the preparation of this manuscript. This work is a publication ofthe USDA/ARS Children’s Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine and TexasChildren’s Hospital, Houston, Texas. The work was supported in part by federal funds from the U.S. Department ofAgriculture Agricultural Research Service, Cooperative Agreement No. 58-6258-6001, and by the National Institutes ofHealth R01 HD33920. The contents of this publication do not necessarily reflect the views or policies of the U.S.Department of Agriculture, nor does mention of trade names, commercial products, or organizations imply endorsementby the U.S. Government.

acid metabolism. The liver and portal-drained visceral (PDV) tissues, composed mainly ofgastrointestinal tissues, contribute roughly 10% of body protein, yet they account for 30–35%of whole-body protein turnover and energy expenditure (Burrin et al., 1989; Yen et al., 1989;Nieto and Lobley, 1999; Stoll et al., 1999a). Relative to their mass, the splanchnic tissuesexert a disproportionate impact on whole-body metabolism due to their relatively high frac-tional rates of protein synthesis and oxygen consumption. Studies with domestic animals havedemonstrated that the fractional protein synthesis rates in the liver and intestinal tissues areseveral-fold higher than that in peripheral tissues, such as muscle (Lobley et al., 1980, 1992;Attaix et al., 1986; Burrin et al., 1992; Davis et al., 1996).

The inherently high metabolic rate of the splanchnic tissues is a function of several factorsrelated to the multitude of biological functions performed by these tissues and the relativelyhigh rate of cell turnover, particularly in the intestine. The gastrointestinal tract, including thestomach, small and large intestine, and pancreas, has a primary function to digest and absorbnutrients from the diet. In these processes, these tissues consume substantial amounts of aminoacids for synthesis of structural and secretory proteins and for oxidative energy necessary toactively transport nutrients and replenish a continual loss of epithelial cells and digestiveenzymes. However, in addition, these tissues function as a physical and immunological barrierto environmental pathogens and noxious substances, representing one of the largest lymphoidtissues in the body. Compromise of this critical gut barrier can not only lead to a suppressionof growth, but jeopardize survival of the organism. The GI tract collectively is also a majorendocrine organ secreting dozens of peptide hormones that provide key signals for the metab-olism and growth of the organism as a whole. The gastrointestinal (GI) tract is also extensivelyenervated with its own intrinsic (enteric nervous system) as well as an extrinsic neural network,which allows it to function autonomously or in concert with the central nervous system.

Some important examples of how these GI functions affect the host include the regulationof food intake (e.g. peptide YY, ghrelin, glucagon-like peptide 1) and substrate homeostasis(e.g. insulin and glucagon). From a protein metabolic perspective, the GI tract functions toassimilate dietary protein in a chemical form (e.g. amino acids) that can be readily used byall somatic cells and simultaneously communicate the availability of nutrients (e.g. insulin)to enhance their utilization by peripheral tissues (e.g. skeletal muscle). The metabolic func-tion of the liver is closely linked to the GI tract by acting as a buffer and scavenger of thedietary nutrients and byproducts absorbed into the portal venous circulation, such as aminoacids, ammonia, and bile acids. The liver also plays a major role in the metabolism of dietaryamino acids that serve as carbon precursors for gluconeogenesis and the transfer of nitrogenreleased from skeletal muscle amino acid catabolism. The liver is a central organ involved inthe production of acute-phase proteins in response to inflammation and infection. The intentof this review is to provide an overview of protein and amino acid metabolism in splanchnictissues and highlight some of the recent advancements in our understanding of how thesetissues function in growing mammals.

2. METABOLIC FATE OF AMINO ACIDS

2.1. Anatomical and morphological considerations

There have been significant advancements in our understanding of splanchnic amino acidmetabolism in the last thirty years. In the early studies, much was learned about amino acidmetabolism from in vivo measurements of splanchnic organ balance. These pioneering studieswere originally derived based on measurements of the net difference in the concentration of

D. G. Burrin and B. Stoll158

amino acids in arterial input and venous drainage and of blood flow (Elwyn et al., 1968; Wolffet al., 1972; Felig, 1975). More detailed kinetic information of unidirectional amino acid fluxeswas subsequently obtained by adapting this approach to the use of radioactive, isotopic aminoacid tracers (Heitmann and Bergman, 1978) and more recently with stable isotopes of aminoacids, in humans, pigs, dogs, mice, sheep, and cattle (Hoerr et al., 1991; Yu et al., 1995; Lobleyet al., 1996a; Stoll et al., 1998; Lapierre et al., 1999; Hallemeesch et al., 2001). Other importantexperimental approaches include in situ organ perfusion, which has been extensively used forliver and intestinal amino acid metabolism (Windmueller and Spaeth, 1980; Haussinger, 1990).However, because there is considerable cellular heterogeneity in the liver and gut tissues, studieswith isolated cells have also provided critical evidence for the cellular basis of amino acidmetabolism (Haussinger, 1990, 1996; Wu, 1998; van Sluijters et al., 2000; Meijer, 2003).

The metabolic fate of amino acids in the splanchnic tissues differs not only between theliver and gastrointestinal tissues, but also among cells within each tissue bed. Even within theliver or gut, the metabolic fate of amino acids is significantly affected by how they are pre-sented to the tissue. For the purposes of this review, the starting point in the metabolic fate ofamino acids will be in the intestinal epithelial cell after transport from the gut lumen.However, it is important to recognize that within the gastrointestinal tissues (e.g. PDV),amino acids are presented via both the lumen and arterial circulation. From a quantitative per-spective, the rate of input of most amino acids from the arterial circulation is substantiallygreater (3–5-fold) than that from the diet (fig. 1). However, the fractional PDV utilization (i.e.uptake/input) of dietary amino acids (ranging from 95% to 20%) is generally much greaterthan amino acids derived from the arterial blood, ranging from approximately 5% to 15%.

In the liver, the metabolic fate of amino acids may vary depending on whether the input isfrom the portal vein or hepatic artery. Studies in piglets suggest that after feeding, portalrather than arterial phenylalanine is preferentially used for the synthesis of constitutiveand secretory hepatic proteins (Stoll et al., 1999b). Within a tissue, the morphological local-ization of a cell can dictate the metabolic fate of amino acids. For example, in the gut, theextent to which epithelial cells derive their amino acids from the luminal or vascular input isaffected by their stage of differentiation and physical location along the crypt–villus axis.Studies showed that crypt cells are more highly labeled with isotopic tracers derived from theblood, whereas villus cells are more highly labeled with tracers given luminally (Alpers, 1972).

Splanchnic protein and amino acid metabolism 159

Fig. 1. Rates of amino acid input into the PDV tissues from the diet and arterial circulation in young piglets.Dietary inputs were based on intake of sow’s milk replacer fed at 12 g protein/kg/day. Arterial inputs werecalculated from measurements of arterial amino acid concentration and portal blood flow rate; this assumedtotal arterial and portal blood flow to be equal. Adapted from Stoll et al. (1998).

These results implied that crypt cells derive their nutrients predominantly from the arterialcirculation, whereas villus cells rely on nutrients absorbed luminally from the diet. Likewise,in the liver, studies have also shown a cellular zonation in the acinus, in which periportalhepatocytes scavenge excess portal ammonia by ureagenesis, whereas perivenous hepatocytessequester ammonia via glutamine synthetase (Haussinger, 1990).

Once taken up by the splanchnic tissues, amino acids have three possible metabolic fates:(1) incorporation into protein; (2) conversion via transamination or deamination into otheramino acids, metabolic substrates, and biosynthetic intermediates; and (3) complete oxidationto CO2. In the case of intestinal epithelial cells, a fourth metabolic fate is transport intothe portal blood stream. In the first two pathways, amino acids can be deposited and recycledby the body for purposes of growth or other biological functions. This recycling process maybe affected by whether an amino acid is incorporated into a constitutive protein or secretedprotein. However, from a nutritional perspective, if essential amino acids are irreversiblymetabolized or completely oxidized to CO2, this represents a nutritional loss to the animal.

If we first consider the metabolic fate of dietary amino acids in the gut, there are some gen-eral observations that can be made from estimates of the net portal balance expressed as aproportion of intake (table 1). The net portal balance represents the quantity of dietary aminoacid absorption into the portal blood expressed as a percent of the intake. Many of the values


Table 1

Summary of portal amino acid balance estimates in young pigs fed liquid milk-replacer underdifferent feeding conditionsa

Gastric hourly Continuous Single oral Continuousbolusb (6 h) duodenalc (6 h) bolusd (8 h) duodenale (24 h)

Lysine 54 51 49 64Threonine 38 67 62 33Leucine 60 55 74 66Isoleucine 70 84 78 30Valine 61 74 72 51Methionine 48 82 70 75Phenylalanine 61 81 61 49Histidine – – 70 –Arginine 138 108 149 155Proline 62 43 88 43Tyrosine 167 78 96 168Cysteine – – 17 –Alanine 205 110 190 112Serine 58 69 84 51Glycine 52 65 69 89Glutamate 7 5 29 10Glutamine –8 –10 –29 –18Aspartate 4 7 24 6

a Values represent net portal balance expressed as percentage of dietary intake. Pigs ranged from 6 to 10 kg bodyweight and diets were fed to supply the estimated daily NRC protein requirement.b Pigs fed via intragastric hourly boluses for 6 h and average portal balance measured between 4 and 6 h(Stoll et al., 1998).c Pigs fed continuously via intraduodenal infusion for 6 h and average portal balance measured between4 and 6 h (Burrin et al., 2003a).d Pigs fed single bolus meal orally and cumulative portal balance measured for 8 h (Bos et al., 2003).e Pigs fed continuously via intraduodenal infusion for 12 h and fasted for 12 h; cumulative portal balancemeasured for total 24 h period (van der Schoor et al., 2002).

are less than 100%, which implies that a portion of the amino acid is utilized by the gut; thisis because the diet fed in these studies was based on milk protein and essentially 95–100%digestible. It is also important to note that the mode of enteral feeding in these four studiesvaried considerably, between bolus versus continuous and gastric versus duodenal adminis-tration of the exact same diet. The first observation is that the net balances of glutamate,glutamine, and aspartate are nearly zero. In other words, the net utilization of these aminoacids by the PDV is approximately equal to the dietary intake. In some cases, the net balanceof glutamine is negative even in the fed state due to the high rate of metabolism in gut tissues.As will be discussed below, the high fractional metabolism of these amino acids is due to theirintegral role as oxidative fuels. A second remarkable observation is that the net balance ofarginine, alanine, and in some cases tyrosine and proline is greater than 100%, which sug-gests a net production of these amino acids by the gut. The last observation is that net balanceof many essential amino acids is significantly less than 100%, and in some cases less than50% of the dietary intake. These results indicate that gut metabolism of the dietary aminoacids significantly alters both the amount and pattern of amino acids absorbed into the portalcirculation.

In contrast to the PDV, there are few reports in the literature describing the net amino acidbalance by the liver in nonruminants, particularly in the fed state (Elwyn et al., 1968; Barrettet al., 1986; Rerat et al., 1992; de Blaauw et al., 1996). There are numerous reports of hepaticamino acid balance in ruminants (Wolff et al., 1972; Lobley et al., 1996a; Lapierre et al.,1999; Blouin et al., 2002). Studies in pigs and dogs fed enterally demonstrated that the nethepatic uptake of glycine and alanine is substantially greater (150–250%) than the dietaryintake. The only amino acids that were significantly released were glutamate and aspartate;the net balance of glutamine was essentially zero. Among the remaining amino acids, hepaticuptake was approximately 50–60% of the dietary intake. Interestingly, the net uptake of thebranched-chain amino acids was lower (35–43% of intake) than the other essentials; thislatter observation translates into a greater net splanchnic output of BCAA compared to otheressential amino acids. In addition to the animals where gut and liver amino acid metabolismhave been measured separately, there also have been numerous reports of total splanchnicamino acid uptake in adult humans using stable isotopes (Castillo et al., 1993a; Matthewset al., 1993; Battezzati et al., 1995, 1999; Haisch et al., 2000). These studies demonstrated thesubstantial first-pass splanchnic extraction (percent enteral input) of amino acids, includingglutamate (96%), glutamine (64%), alanine (69%), arginine (38%), leucine (21%), andphenylalanine (29%). A recent report in humans describing the kinetics of ingested 15N-labeledsoy protein showed that the splanchnic bed extracted nearly 60% of the dietary N, of which40% was channeled into protein anabolism and 20% was deaminated (Fouillet et al., 2003).Similar studies examining leucine kinetics found that first-pass splanchnic uptake in preterminfants and elderly men was approximately 2-fold higher than in young adult men (Beaufrereet al., 1992; Boirie et al., 1997).

2.2. Protein synthesis

A major metabolic fate of amino acids taken up by the gut and liver is incorporation into cel-lular protein. Numerous studies have measured the rates of protein synthesis in various tissuesof the gastrointestinal tract and the liver. Among the literature reports, various methods havebeen used to measure tissue protein synthesis in vivo; however, the best established and vali-dated approach has been the flooding dose method, first reported by Garlick and others (seediscussion in Chapter 18; Garlick et al., 1980, 1994). Review of these studies reveals some


general characteristics of protein synthesis (table 2), the first of which is relative rates of pro-tein synthesis within the splanchnic tissue bed. In growing animals, the fractional proteinsynthesis rates (FSR) are generally highest in the pancreas and progressively decline in smallintestine, stomach, and large intestine. In a weanling, 28-day-old rat, FSRs (%/day) in thepancreas, small intestine, and stomach are 440%, 150%, and 140%, respectively (Burrin etal., 1991). In the preruminant, one-week-old lamb, the FSRs also vary considerably amongthe regions of the gut, including the rumen (30%), abomasum (56%), small intestine (88%),cecum (45%), and colon (38%) (Attaix and Arnal, 1987; Attaix et al., 1992). The pattern ofdeclining proximal to distal gradient in intestinal FSR is also observed in neonatal piglets(Stoll et al., 2000a) and mice (Bark et al., 1998). In comparison to the gastrointestinal tissues,the FSR in the liver is relatively high, being similar to that of the proximal small intestine, butlower than the pancreas.

In comparison to the splanchnic tissues, mainly small intestine and liver, the protein syn-thesis rates in the other visceral organs and peripheral tissues are substantially lower. It isnotable that the skeletal muscle generally has the lowest protein synthesis rate among alltissues, yet comprises the largest proportion of whole-body protein mass. If we consider the issueof age or stage of development, the fractional rates of whole-body protein metabolism arehighest during fetal life, and decline progressively with advancing postconceptual age, com-mensurate with fractional growth rates (fig. 2). (Goldspink and Kelly, 1984; Goldspink et al.,1984.) This is also evident in the splanchnic tissues, in which the fractional protein synthesisrates in the small and large intestine (107% to 61%) and in the liver (134% to 48%) declineduring the lifespan in rats. Interestingly, in the intestine, after weaning, the decline in intes-tinal FSR with age is largely due to a decreased synthesis in the muscularis and serosal layers,whereas the mucosal FSR remains constant (Merry et al., 1992). However, in striking contrastto the fractional protein synthesis rate, the protein mass increases approximately 750-fold


Table 2

Summary of tissue fractional protein synthesis rates in various mammalian speciesa

Weaned ratb Weaned mousec Neonatal pigletd Preruminant lambe

Stomach 140 51 55 –Reticulo-rumen – 67 – 30Abomasum – 55 – 56Duodenum – 84 – 86Jejunum 150 78 124 93Ileum – 79 60 84Cecum – – – 45Colon 58 43 – 38Pancreas 440 – 143 –Liver 125 106 85 115Spleen 38 70 55 –Kidney 44 38 45 –Brain 12 12 20 –Skeletal muscle 7 3 30 21

a Values expressed as %/day.b Goldspink and Kelly (1984); Goldspink et al. (1984); Preedy et al. (1988); Burrin et al. (1991, 1992).c Bark et al. (1998); Burrin et al. (1999).d Burrin et al. (1995, 1997); Stoll et al. (2000a).e Attaix and Arnal (1987); Attaix et al. (1992).

(2.5 to 1893 mg) in the intestine and 600-fold (17 to 3252 mg) in the liver between 18 daysgestational age and 105 weeks postpartum age (Goldspink and Kelly, 1984; Goldspink et al.,1984). Despite the overall age-related decline, studies in neonatal rats and mice indicate thatthe FSRs of the stomach, small intestine, and pancreas increase significantly after weaning(Burrin et al., 1991, 1999a). In domestic animals, there are few, if any, estimates of gastroin-testinal FSRs before birth and beyond pubertal ages, yet the changes between birth andweaning in pigs and sheep tend to parallel those in rodents (Seve et al., 1986; Attaix et al.,1992; Davis et al., 1996). In milk-fed animals, the intestinal FSR declines during the neona-tal period, but increases markedly (40–50%) after weaning. The sharp increase in gut proteinsynthesis after weaning in pigs is likely due to the substantial change in the composition ofthe diet, gut microflora, and resultant stimulation of mucosal cellularity and proliferation(Attaix and Meslin, 1991; Jiang et al., 2000).

2.3. Protein degradation

The kinetics of protein degradation are technically difficult to quantify, especially in vivo, andthus there is a limited understanding of how certain metabolic factors affect this in the liverand particularly in gut tissues (see discussion in Chapter 4). With regard to the gut, somereports based on indirect estimates suggest that the fractional rates of protein degradation arequantitatively similar to rates of protein synthesis, because the net balance between these twoopposing phenomena, i.e. the fractional protein accretion rate, is relatively low, at least in guttissues (Burrin et al., 1999b; Stoll et al., 2000a). There is a considerably larger literature onthe factors that influence qualitative aspects of protein degradation, based on studies withperfused livers and cultured hepatocytes and colon carcinoma cells (HT-29) (Mortimore et al.,1989; Kadowaki and Kanazawa, 2003; Ogier-Denis and Codogno, 2003).

The major class of protein degradation in the liver is lysosomal autophagy via the cathepsinproteases, but also includes the ubiquitin–proteosomal system. In the intestine, three proteolyticsystems have been identified – lysosomal–autophagic cathepsins, ubiquitin–proteosomal, andcalpains – yet the relative significance of these systems to overall protein degradation isunknown (Baracos et al., 2000). Hepatic autophagy has been shown to be potently inhibited


Fig. 2. Ontogeny of whole-body and splanchnic tissue protein synthesis rates in rats. Adapted fromGoldspink and Kelly (1984) and Goldspink et al. (1984).

by several anabolic factors, including amino acids and insulin, and activated by glucagon(Mortimore and Poso, 1987; van Sluijters et al., 2000). The regulatory actions of amino acidson hepatic autophagy are specific for certain amino acids, particularly leucine. Moreover, theprocess of autophagy has been linked to cell volume regulation and apoptotic cell death(Schliess and Haussinger, 2002). These extracellular and intracellular signaling pathways willbe discussed in more detail in subsequent sections. Studies with perfused rat liver and isolatedhepatocytes have shown that autophagy increases to a peak at 6 months of age and thendeclines with age by a process that can be prevented with calorie restriction (Del Roso et al.,2003). As mentioned, the quantitative significance of protein degradation in intestinalmucosal growth is essentially unknown. However, recent reports have shown that degradationis upregulated by nutrient deprivation and exercise (de Blaauw et al., 1996; Samuels et al.,1996; Halseth et al., 1997), but suppressed by glucagon-like peptide 2 (Burrin et al., 2000a).In the study of de Blaauw et al. (1996), it is of interest to note that the rate of proteolysisderived from 3H-phenylalanine kinetics was markedly higher (5-fold) in the PDV than in theliver of the fasted rat.

2.4. Endogenous protein secretion and amino acid recycling

Another aspect of protein metabolism in the gut involves the metabolic fate of endogenousprotein secreted into the gut lumen. Endogenous proteins include secretions arising from thesaliva, gastric mucosa, bile, pancreas, and the exfoliation of epithelial cells. The biochemicalcharacteristics of endogenous proteins include enzymes (amylases, pepsinogen, trypsinogen),glycocalix constituents (mucins), growth factors (epidermal growth factor, insulin-like growthfactor), and antioxidants (glutathione). Another significant source of protein in terminal ileumis of microbial origin. Some reports indicate that as much as 25–50% of the nitrogen appearingin ileal output is of microbial origin (Stein and Nyachoti, 2003). Given the long list ofendogenous proteins, it is perhaps not surprising that, collectively, these proteins represent aquantitatively significant amount of amino acid released into the gut lumen. Estimated ilealendogenous protein losses range from approximately 10–40 g protein/kg dry matter intakeand represent up to 10–25% of the dietary protein intake and 5–10% of the whole-body proteinturnover (Fuller and Reeds, 1998; Reeds et al., 1999). Recent studies have also shown thatendogenous ileal protein loss increases with protein intake (Hodgkinson et al., 2000). A criticalnutritional and metabolic question with respect to endogenous proteins secreted into the gutlumen centers around the extent to which they are recycled and absorbed by the host. It isimportant to note that measurements of ileal protein losses represent a minimal estimate oftotal upper gut endogenous protein secretion, because evidence suggests that most (70–80%)of the proteins secreted are hydrolyzed and reabsorbed within the small intestine (Stein andNyachoti, 2003). A recent study in young pigs demonstrated that, over a 24 h period, 52% ofthe dietary amino acid intake was absorbed into the portal circulation and one-third of thiswas derived from recycled intestinal secretions (van der Schoor et al., 2002).

In is generally held that the amino acids that pass from the terminal ileum into the cecumand large intestine are catabolized by microbial fermentation. The assumption that theendogenous amino acids are fermented and lost in the large intestine is based on early reportsthat colonic absorption of amino acid is limited and occurs only during early postnataldevelopment (Fuller and Reeds, 1998). The carbon from colonic microbial amino acid catab-olism can be lost as CO2 or reabsorbed into the portal blood in the form of short-chain fattyacids. The nitrogen released by microbial amino acid catabolism may be in the form ofammonia, which can be absorbed and recycled into the body amino acid and urea pools.


Historically, the colonic microbial catabolism of essential amino acids has been considered anutritional loss, because by definition, if an amino acid is essential, then this usually meansthat it cannot be synthesized by mammalian cells. This concept has recently been challenged bya number of elegant studies in pigs and humans, which demonstrated the microbial synthesisof some essential amino acids, particularly lysine, based on labeling with 15N-labeled ammo-nia and urea (fig. 3) (Metges, 2000; Torrallardona et al., 2003a,b). These studies have revealedtwo critical phenomena: (1) that microbial lysine synthesis occurs in the upper gastrointesti-nal tract and thus can be absorbed in the small intestine; and (2) the synthesis of severalessential amino acids may be nutritionally significant. These studies raise several intriguingquestions, including, (1) What are the carbon and nitrogen precursors for microbial amino acidsynthesis?, (2) How does the gut microbial load affect amino acid synthesis?, and (3) How isthis phenomenon regulated by dietary nutrient intake?

2.5. Amino acids as oxidative fuels

In recent years, it has become increasingly apparent that the splanchnic bed derives a majorityof its oxidative energy from the catabolism of amino acids, rather than glucose or fatty acids.The liver is clearly the major site of amino acid metabolism in mammals and has been his-torically considered a major site of catabolism and oxidation. However, since the classicstudies of Windmueller and Spaeth (1974, 1975, 1976, 1978, 1980), it has become readilyapparent that the gut, particularly the intestine, is also a major site of catabolism of severalamino acids. An important distinction to be made, however, is that while amino acids arecatabolized in both the liver and gut tissues, the extent to which they are completely oxidizedto CO2 may vary. In a brilliant review of amino acid oxidation, Jungas et al. (1992) system-atically characterized the metabolic fate of dietary amino acids in the gut, liver, kidney, andmuscle and derived two important conclusions. First, they estimated that a primary metabolicfate of amino acid carbon in the liver is conversion to glucose. They make the argument that


Fig. 3. Model illustrating gut microbial amino acid synthesis. Model depicts microbial lysine synthesisderived from luminal ammonia nitrogen and blood urea nitrogen.

sufficient ATP is generated from the partial oxidation of dietary amino acids to generateroughly half of the liver’s energy needs, an amount that approximates the energy required forthe synthesis of glucose. Thus, while amino acids are consumed in oxidative metabolic path-ways in the liver, the complete oxidation of amino acids would far exceed the liver’s energyneeds and capacity to handle the end-products. A second important observation from theiranalysis is that the hepatic oxidation of amino acids to glucose makes nearly two-thirds of thetotal energy content of dietary amino acids available to peripheral tissues (as glucose). Asa result, there is no need for peripheral tissues to synthesize a complex array of enzymes tooxidize amino acids and synthesize urea.

As discussed below, the gut is a willing participant in this process by releasing some of thecarbon derived from nonessential amino acid catabolism into the portal vein as alanine andlactate, both of which are key precursors for hepatic gluconeogenesis. Glutamate is a keyamino acid linking hepatic amino acid catabolism and gluconeogenesis (Brosnan, 2000,2003). Thus, it is not surprising that the gut releases proline, arginine, and ornithine into theportal vein, all amino acids whose metabolism converges at glutamate in the liver. Therefore,the splanchnic tissues, by design, are situated anatomically and metabolically within themammalian organism to regulate the flow of dietary amino acids, in such manner as to meettheir oxidative energy needs first, while at the same time ensuring the delivery of the primaloxidative fuel for peripheral tissues, namely glucose.

The seminal studies of Windmueller and Spaeth (1974, 1975, 1976, 1978, 1980) were thefirst to show evidence of extensive metabolism of glutamine, glutamate, and aspartate inin situ intestinal perfusions in fasted, anesthetized rats. Results from young piglets fed a high-protein, milk-based formula indicated that more than 95% of the dietary glutamine,glutamate, and aspartate is utilized by the gut (Stoll et al., 1998). The studies of Windmuellerand Spaeth focused attention on the role of glutamine as the major oxidative fuel in the gut.However, it is important to note that both glutamate and aspartate are of perhaps equal impor-tance as intestinal oxidative fuels. Recent studies in young pigs and humans confirm theextensive intestinal oxidation of dietary 13C-labeled glutamate and glutamine (Battezati et al.,1995; Stoll et al., 1999a; Haisch et al., 2000).

The metabolism of glutamine is accomplished first by the catalysis via phosphate-dependentglutaminase and subsequently by glutamate dehydrogenase (GDH) enzymes, both of whichare present in the stomach, small intestine, and colon of the young pig (Madej et al., 1999).Interestingly, the activity of GDH is increased approximately 3-fold in the small intestineafter weaning. The resulting ketoacid product of GDH is α-ketoglutarate, which is thenmetabolized yielding CO2 via the tricarboxylic acid cycle. It is important to note that,although there is extensive uptake and metabolism of these three amino acids, their carbonskeletons are not completely oxidized to CO2 and they do not account for all of the CO2

released by the gut. The in situ studies with perfused rat intestine and those in vivo withpiglets and humans indicate that most of the glutamine (55–70%), glutamate (52–64%), andaspartate (52%) are oxidized to CO2 (Windmueller and Spaeth, 1976, 1978; Stoll et al.,1999a). The remaining carbon atoms from these three substrates, which are not oxidized toCO2, are converted to lactate, alanine, proline, citrulline, ornithine, and arginine and thenreleased into the portal circulation (Windmueller and Spaeth, 1975; Stoll et al., 1999a). Themetabolic fate of nitrogen from these amino acids is not fully understood. However, evidencesuggests that a portion of the nitrogen derived from glutamine and glutamate metabolism istransferred to ammonia and other amino acids, including citrulline, ornithine, proline, andarginine; much of the nitrogen from these products is converted to urea in the liver.


The rate of glutamine oxidation in isolated enterocytes decreases by roughly 90% in thefirst 3 weeks of life (Darcy-Vrillon et al., 1994; Wu et al., 1995). Also, after weaning, in intra-epithelial lymphocytes, glutamine is mainly metabolized to glutamate and ammonia (92%),with minimal oxidation (4%). The decline in glutamine oxidation with age is paralleled byincreased activities of glutamine synthetase, glutaminase, and glutamate dehydrogenase(Hahn et al., 1988; Shenoy et al., 1996; Madej et al., 1999). Studies with isolated enterocytesalso have shown that, along with glutamine, glucose is an important intestinal oxidative fuel(Darcy-Vrillon et al., 1994; Kight and Fleming, 1995; Wu et al., 1995). However, glutamineeffectively suppresses glucose oxidation in enterocytes, whereas glucose has little impact onglutamine oxidation (Kight and Fleming, 1995; Wu et al., 1995). The oxidation of glutamineand its suppression of glucose oxidation was also found to be nearly twice as high (60%) inthe proximal compared to the distal (31%) small intestine (Kight and Fleming, 1995). Therelationship is consistent with in vivo studies in piglets demonstrating that, although glucoserepresents an important oxidative fuel (29%), the proportion of glucose oxidized completelyto CO2 is substantially less than that of either glutamate or glutamine. The implication is thatglutamine and glutamate are preferentially channeled toward mitochondrial oxidation, whilemost of the glucose is utilized for other metabolic or biosynthetic purposes.

Recent studies based on isotopic PDV tracer kinetics have shown that dietary essentialamino acids are also oxidized within the gut. Studies in young pigs showed that intestinaloxidation of dietary lysine accounted for about one-third of whole-body lysine oxidationand was completely suppressed by feeding a low-protein diet (van Goudoever et al., 2000).Interestingly, although arterial lysine was taken up by the PDV, none of this was oxidized,suggesting a preferential oxidation of dietary lysine (van Goudoever et al., 2000). As withlysine, there is significant leucine metabolism by the gut via both transamination to ketoiso-caproic acid (KIC) and complete oxidation to CO2. Studies in young pigs and dogs havedemonstrated that approximately 5–10% of whole-body leucine flux is oxidized by the PDV(Yu et al., 1995; van der Schoor et al., 2001). Although approximately 40% of the leucinetaken up by the gut was converted to KIC, nearly all of this is transaminated back to leucine;thus, the net KIC release is negligible (Yu et al., 1995). Studies in young, grower pigs (15–20 kg)suggest that approximately 40% of the whole-body phenylalanine oxidation occurred in thePDV tissues (Bush et al., 2003a). The oxidation of phenylalanine implies that phenylalaninehydroxylation occurs in the gut and is consistent with previous observations suggesting netportal tyrosine production in excess of dietary intake. Given that hydroxylation rather thancomplete oxidation to CO2 represents the point of irreversible loss of phenylalanine, furtherstudies are warranted to quantify the proportion of whole-body phenylalanine flux metabolizedto tyrosine by the gut.

The reports of essential amino acid oxidation by the gut have raised the question of whetherthis is due to mucosal metabolism or microbial fermentation. The recent evidence of de novolysine synthesis in the proximal intestine in pigs implicates a metabolically significant microbialflora, which could also catabolize dietary amino acids. To address this issue, emerging studiesare beginning to identify the localization of essential amino acid catabolic enzymes within thedifferent mucosal cell phenotypes, i.e. enterocytes and lymphoid cells. Two reports have char-acterized BCAA and lysine catabolic enzymes in enterocytes isolated from piglets at 0, 3 and7 days of age (Elango et al., 2003; Pink et al., 2003). Branched-chain amino acid transferaseactivity (BCAT) was detected in enterocytes and liver of 7-day-old piglets at a level that was18% and 9%, respectively, of muscle mitochondrial BCAT activity. Enterocyte BCAT andbranched-chain dehydrogenase (BCKD) activity also increased between 0 and 7 days-of age.


BCAT and BCKD activity also was present in small intestinal mucosal tissue in piglets(Burrin et al., 2003a). In the case of lysine, the first catabolic enzyme in the pathway, lysineketoglutarate reductase, was found in mitochondria from freshly isolated enterocytes at anactivity level about 50–60% as high as that measured in liver, whereas the activity of saccha-ropine dehydrogenase was low. These findings are important for two reasons, the first ofwhich is that they demonstrate the amino acid catabolic capacity of the major mucosal celltype relative to liver. Secondly, the results suggest that the lysine and leucine oxidationreported in vivo in piglets is mediated partially by mucosal metabolism; however, the relativesignificance of microbial catabolism remains to be determined.

2.6. Amino acids as biosynthetic precursors

Besides their incorporation into the bulk protein pool and use as oxidative fuels, amino acidsare metabolized by splanchnic tissues into a variety of end-products, which serve a variety ofkey functions for the cell, specifically, and the host in general (table 3). An especially impor-tant phenomenon is the interconversion of arginine, ornithine, and proline and their respectiveroles in the biosynthesis of polyamines and nitric oxide in the intestinal mucosa. Studies haveshown that neonatal small intestine is an important site of arginine and proline synthesis andthis interconversion depends on first-pass metabolism (Murphy et al., 1996; Wu, 1998; Stollet al., 1999a; Bertolo et al., 2003). The milk of humans, pigs, rats, and many other mammalsis relatively deficient in arginine (Davis et al., 1994). Moreover, in the suckling pig, the intes-tinal synthesis of arginine provides only about half of the animal’s needs for growth. Thus,supplementation of dietary arginine is considered to be essential for maximal growth in youngpiglets. The intestinal synthesis of arginine declines, while arginase activity increases sub-stantially during the late suckling period (Wu and Morris, 1998). Also during this time, thesynthesis of citrulline and ornithine increases with age, particularly after weaning, and prolineand arginine are major precursors for their synthesis (Wu, 1997). Thus, in adult rats and weaningpigs, the intestinal conversion of glutamine, glutamate, and proline to citrulline provides acritical precursor for arginine synthesis in the kidney (Windmueller and Spaeth, 1980; Duganet al., 1995). The dependence on intestinal first-pass metabolism for synthesis of either arginine(in neonates) or its immediate renal precursor, citrulline (in adults), results in arginine defi-ciency when intestinal metabolism is either bypassed during TPN (Brunton et al., 1999) orremoved surgically by resection (Wakabayashi et al., 1995). The dependence on the intestinefor citrulline synthesis has led to development of plasma citrulline as a marker for enterocytemass in conditions of disease (Crenn et al., 2003).

Studies in cultured intestinal cells have shown that ornithine derived from arginine metab-olism is converted to polyamines (Blachier et al., 1995). Polyamines (putrescine, spermidine,


Table 3

Amino acid precursors and functional end-products produced in splanchnic tissues

Precursors • Arginine • Glutamine • Methionine • Threonine• Proline • Glutamate • Cysteine • Serine

• AspartateProducts • Nitric oxide • Nucleotides • Cysteine • Mucins

• Polyamines • Glutathione • Taurine• Creatinine • Glucosamine • Glutathione

• Thioredoxin

spermine, cadaverine) are ubiquitous cationic amines involved in cell proliferation anddifferentiation in many tissues, including the gastrointestinal tract. Ornithine decarboxylase(ODC) and S-adenosylmethionine decarboxylase (SAMDC), converting ornithine to putrescineand putrescine to spermine, respectively, are the rate-limiting enzymes in polyamine synthesis.The synthesis of polyamines from arginine is negligible in enterocytes of newborn and suck-ling animals (Blachier et al., 1991, 1992), but polyamines are present in mammalian milk(Pollack et al., 1992; Buts et al., 1995). As piglets progress from suckling to weaning, majorend-products of intestinal enterocyte arginine metabolism are proline and ornithine. Thus,when the ingestion of milk-borne polyamines by the neonate ceases after weaning, the inductionof intestinal polyamine synthesis from ornithine, arginine, and proline becomes physiologi-cally significant for the maintenance of normal intestinal growth and function (Wu et al.,2000a,b). Furthermore, the induction of intestinal polyamine synthesis is dependent on theweaning-induced cortisol surge.

Another important end-product of arginine metabolism is nitric oxide. Nitric oxide is amajor physiological regulator in the body, particularly by enhancing vascular function, tissueperfusion, and immune function. For this reason, arginine is a key component supplementedto enteral formulas designed for trauma and surgically stressed patients (Huang et al., 2003;McCowen and Bistrian, 2003) and may become limiting under conditions of increased NOproduction (Hallemeesch et al., 2002). Nitric oxide is produced by conversion of arginine tocitrulline by the enzyme nitric oxide synthase (NOS), which is expressed in three isoforms,all of which are found in gastrointestinal and liver tissue. In the whole body, the proportionof arginine that is converted to NO is relatively low (1–10% arginine flux) (Castillo et al.,1996), but is increased in response to stress and trauma (Argaman et al., 2003). However, thefirst-pass splanchnic utilization of enteral arginine is about 40%, of which metabolism to NOrepresents 16% of the whole-body nitrate production (Castillo et al., 1993a,b). The rates ofliver and PDV NO production have been measured using a similar isotopic approach basedon kinetic conversion of 15N-arginine to 15N-citrulline (Luiking and Deutz, 2003). These studiesshowed that as much as 35% of the arginine utilization is converted to NO in the liver and gutin endotoxemic pigs; NO synthesis also is increased markedly with supplemental arginine(Bruins et al., 2002a,b).

The nonessential amino acids, aspartate, glutamine, and glycine, are key precursors for thesynthesis of nucleotides. This fact is quantitatively important in the intestinal mucosa giventhe high rate of cell proliferation coupled with the fact that most of the nucleotides, at leastribonucleotides, are synthesized de novo (Boza et al., 1996). The amide nitrogen from gluta-mine serves as the nitrogen donor for the synthesis of both purines and pyrimidines, whereasmost of the carbon skeleton of nucleotides is derived from aspartate and glycine. Thebiochemical mechanism whereby glutamine affects intestinal function also may be related toits conversion to glucosamine, which reduces the cellular NADPH and suppresses nitric oxidesynthesis (Wu et al., 2001). In immune cells, glutamine and glutamate also provide an impor-tant source of NADPH via conversion of malate to pyruvate; NADPH is critical in these cellsfor production of superoxide and NO and for glutathione reductase activity (Newsholmeet al., 2003).

Amino acids also serve as precursors for synthesis of compounds involved in support ofinnate immunity in the gut and antioxidant function in both liver and gut. Dietary threonineand cysteine are considered to be important for mucin synthesis by goblet cells within thestomach and intestinal mucosa. The secretory mucins play a key role in the innate immunedefense of the mucosa, and the core protein of the major intestinal mucins contains a largeamount of threonine and cysteine (van Klinken et al., 1997; Faure et al., 2002). Studies in pigs


indicated that as much as 60% of dietary threonine is utilized by the gut in first-pass (Stollet al., 1998). Consistent with these findings, other studies in piglets (Bertolo et al., 1998)demonstrated that the threonine requirement of piglets maintained by parenteral nutrition wasnearly 60% lower than that of piglets receiving enteral feedings. A subsequent report foundthat feeding threonine-deficient diets to piglets significantly reduces intestinal mass andgoblet cell numbers, and this suppression of intestinal growth cannot be fully restored by pro-viding threonine parenterally (Ball et al., 1999). The mucosal synthesis and secretion ofmucins are likely to be quantitatively significant, and thus the needs for dietary threonine andcysteine may be increased under conditions of gut hypersensitivity or inflammation. A recentreport in rats, using a novel approach to purify mucins, demonstrated that the synthesis rateof mucin glycoproteins was relatively constant along the length of the intestine (range112–138%/day), but substantially higher than the total mucosal protein synthesis rate, espe-cially in the distal bowel: values for total protein were 77% and 44%/day in the ileum andcolon, respectively (Faure et al., 2002).

Besides incorporation into mucin and other mucosal proteins, the extent to which threonineis further metabolized within the gut tissues is poorly understood. There is some debate as towhich metabolic pathway is most important in the catabolism of threonine in mammals(House et al., 2001). The two predominant pathways of threonine catabolism in the pig arebelieved to be catalyzed by either threonine aldolase/dehydrogenase or threonine dehydratase(Ballevre et al., 1990; Le Floc’h et al., 1996, 1997). Studies in pigs suggest that conversionto glycine via threonine aldolase/dehydrogenase is the predominant pathway of irreversiblethreonine catabolism. Moreover, threonine dehydrogenase activity was localized in boththe liver and pancreas, but not other gut tissues, implicating the PDV as a possible site ofthreonine catabolism.

In addition to mucins, methionine and cysteine serve as biosynthetic precursors for numer-ous functional end-products, including glutathione, polyamines, and taurine. In numerouscells within the body, methionine is metabolized via transmethylation to homocysteine and inthe process produces S-adenosylmethionine, which donates an aminopropyl moiety in the for-mation of the polyamines, spermidine and spermine (fig. 4) (Finkelstein, 2000). Homocysteineis converted to cysteine via transsulfuration. Cysteine is one of three constituent amino acidsof glutathione (GSH), along with glutamate and glycine. Moreover, cysteine is metabolizedto form taurine. Glutathione is a major cellular antioxidant in cells found in the intestinalmucosa and liver (Lash et al., 1986; Martensson et al., 1991). However, cysteine and taurinecan also function as cellular antioxidants (Santangelo, 2002; Zafarullah et al., 2003). Thequantitative significance of these functional end-products to splanchnic methionine, cysteine,and glutamate utilization is unknown. Early studies in humans suggested that splanchnic tis-sues are an important site of transsulfuration (Stegink and Den Besten, 1972). Studies inpiglets indicate that first-pass utilization of dietary methionine ranged from 30% to 40%(Rerat et al., 1992; Stoll et al., 1998; Bos et al., 2003). A recent study in neonatal pigletsshowed that the methionine requirement (g/kg/day) was 0.42 and 0.29 in enteral and parenter-ally fed piglets, respectively, which implies that first-pass splanchnic metabolism accounts for30% of the dietary methionine requirement (Shoveller et al., 2003).

The original studies by Finkelstein (2000) demonstrated that gastrointestinal tissuespossess the enzymes necessary to metabolize methionine to cysteine, albeit at significantlylower activities than the liver. However, there are few reports describing the kinetics ofmethionine metabolism in the gut, either in vivo or in vitro with isolated enterocytes. Recentstudies based on enzyme assay and in vivo isotopic tracers in ruminants imply that methioninetransmethylation occurs in the ruminant gut and that the activities are comparable to the liver


(Lobley et al., 1996b, 2003; Lambert et al., 2002). In the case of dietary cysteine, studies inpigs indicate that rate of appearance into the portal blood is very limited (less than 20%dietary, intake), suggesting extensive intestinal utilization of cysteine in first-pass (Rerat et al.,1992; Stoll et al., 1998; Bos et al., 2003). The first step in cysteine catabolism is conversionto cysteinesulfinate via the enzyme cysteine dioxygenase. Cysteinesulfinate is then convertedto taurine via cysteinesulfinate decarboxylase or pyruvate via aspartate aminotransferase.Rodent studies with 1-14C-labeled cysteine demonstrated significantly higher oxidation whengiven via the intragastric (70%) than intraperitoneal (41%) route, suggesting that nearly halfof the whole-body cysteine oxidation occurs in splanchnic tissues (Stipanuk and Rotter,1984). The increased oxidation of intragastric versus systemic cysteine was largely attributedto increased oxidation to pyruvate rather than to taurine. Subsequent work demonstrated thatintestinal enterocytes extensively metabolize cysteine via cysteine dioxygenase to cysteine-sulfanate (Coloso and Stipanuk, 1989). In vivo rodent studies with intravenous infusion ofisotopically labeled 15N-cysteine indicate that an important metabolic fate of cysteine in thegut is incorporation into glutathione (GSH) (Malmezat et al., 2000a). With respect to gluta-mate, studies in piglets demonstrated that enteral glutamate is preferentially incorporated intomucosal GSH (Reeds et al., 1997).


Fig. 4. Schematic overview of sulfur amino acid metabolism. Abbreviations: SAM, S-adenosylmethionine;SAH, S-adenosylhomocysteine.

3. FACTORS REGULATING PROTEIN AND AMINO ACID METABOLISM

3.1. Nutrition

Oral feeding is probably the most potent stimulus of splanchnic protein and amino acidmetabolism in growing animals (McNurlan et al., 1979; Burrin et al., 1991, 1992, 1995).Prolonged fasting leads to markedly reduced protein mass in gut and liver tissues via sup-pressed protein synthesis and increased protein degradation, especially in the small intestine(Samuels et al., 1996). Recent studies have shown that the stimulatory effect of nutrient intakeon gut protein metabolism is dependent on enteral administration of nutrients (Dudley et al.,1998; Burrin et al., 2000b, Stoll et al., 2000a). These studies showed that the fractional proteinsynthesis rate and the net balance between intestinal protein loss and accretion is directly andpositively determined by the level of enteral intake. An enteral nutrient intake of 20% of totalwas found to maintain intestinal protein balance. Similarly, a recent study in preterm infantsreported that minimal enteral feeding of approximately 10–15% of total increased splanchnicleucine uptake (Saenz de Pipaon et al., 2003). In contrast to the gut, however, the rate ofhepatic protein synthesis decreased with the level of enteral nutrition in neonatal piglets,which is consistent with an enlarged liver mass associated with TPN (fig. 5).

The composition of dietary nutrients can also substantially affect splanchnic protein andamino acid metabolism. An important consideration in the neonate is the stimulation of liverand gut function in response to the onset of suckling colostrum at birth (Burrin et al., 1995).Studies in neonatal piglets showed that both liver and gut protein synthesis is rapidly upreg-ulated with the first feeding of colostrum. The relatively high concentrations of colostralgrowth factors are thought to provide key signals for intestinal and liver development.However, studies comparing colostrum to macronutrient matched formula and those withenteral supplementation of recombinant growth factors (IGF-I) suggest that milk-bornegrowth factors have limited trophic effects on the intestine (Burrin et al., 1999a, 2001). Theimpact of these milk-borne growth factors is probably more important for stimulation of thegut immune system. Thus, it appears that the macronutrient intake is the major component ofthe diet affecting splanchnic tissue protein metabolism in developing mammals.

In humans, first-pass splanchnic leucine uptake is nearly 2-fold lower in subjects fed aprotein-free diet compared to controls fed complete protein-containing diet (Cayol et al., 1997).


Fig. 5. Role of enteral and parenteral nutrition on intestinal and liver protein synthesis in neonatal piglets.Adapted from Stoll et al. (2000a) and unpublished results.

Some studies have shown that restriction of dietary protein has limited effects on intestinalprotein synthesis (Seve et al., 1986), whereas others have reported a decrease in protein syn-thesis in both liver and gut (McNurlan and Garlick, 1981; Wykes et al., 1996). Studies inneonatal pigs demonstrated that protein malnutrition significantly reduces whole-bodygrowth and amino acid absorption, but does not affect gut tissue growth (Ebner et al., 1994;van Goudoever et al., 2000). Consistent with this, studies in pigs fed low-protein diets showedthat PDV metabolized 75% of enteral 13C-lysine intake, compared to 45% in high-protein-fedpigs (van Goudoever et al., 2000). Protein malnutrition also suppressed the gut oxidation oflysine, leucine, and glutamate, whereas glucose oxidation increased (van Goudoever et al.,2000; van der Schoor et al., 2001). These piglet studies suggest that when dietary proteinintake is reduced, the gut amino acid utilization for protein synthesis is maintained, but aminoacid oxidation is suppressed.

Studies focusing on the impact of dietary macronutrients found that enteral amino acids,but not carbohydrate or lipid, stimulated intestinal protein synthesis, whereas each of thesestimulated gut protein accretion, suggesting that carbohydrate and lipid suppressed gutproteolysis (Stoll et al., 2000b). In contrast, in other studies, enteral amino acids rapidly decreasedintestinal protein synthesis and proteolysis, whereas parenteral or luminal glucose infusionincreased intestinal protein synthesis (Weber et al., 1989; Adegoke et al., 1999, 2003). Aminoacids also potently inhibit hepatic autophagic and proteasomal proteolysis (Mortimore andPoso, 1987; Hamel et al., 2003). Feeding a high-fat versus high-carbohydrate diet to youngpigs stimulated protein synthesis in the intestine, but not the liver (Ponter et al., 1994).Numerous studies have examined the effect of dietary supplementation with amino acids onintestinal and liver protein metabolism. In young pigs, amino acid deprivation suppresseswhereas supplementing amino acids parenterally stimulates hepatic protein and albumin syn-thesis, but not intestinal protein synthesis (Davis et al., 2002a; Hellstern et al., 2002). Similarly,in humans, parenteral amino acid infusion stimulated splanchnic tissue protein synthesis andsuppressed protein degradation (Nygren and Nair, 2003).

Among the specific amino acids, glutamine is most extensively studied. Supplementingglutamine has been shown to stimulate protein synthesis and reduce proteolysis in the gut insome cases (Coeffier et al., 2003), but not in others (Garcia-Arumi et al., 1995; Marchini et al.,1999; Bouteloup-Demange et al., 2000). Glutamine also has been shown to stimulate proteinsynthesis and suppress protein degradation in cultured enterocytes and hepatocytes(Higashiguchi et al., 1993; Le Bacquer et al., 2001, 2003; Haussinger et al., 2001). Studieshave shown positive effects of glutamine-supplemented TPN in preventing atrophy, stimulat-ing protein anabolism, and maintaining intestinal permeability (Tamada et al., 1992; Inoueet al., 1993; Platell et al., 1993; Haque et al., 1996; Naka et al., 1997; Khan et al., 1999).In contrast, other studies have shown no effect on gut growth or protein metabolism in ani-mals receiving glutamine supplementation parenterally (Spaeth et al., 1993; Burrin et al.,1994; Marchini et al., 1999; Humbert et al., 2001). Furthermore, parenteral infusion of lipidhas been shown to stimulate intestinal protein synthesis more than either glucose or glutamine(Stein et al., 1994). Thus, the intestinal effects of glutamine-supplemented TPN are evidentunder conditions of compromised gut function, such as sepsis, inflammation, and small-bowelresection, while in healthy animals glutamine has limited effect.

Dietary leucine stimulates liver protein synthesis (Kimball and Jefferson, 2001; Lynch et al.,2002) and suppresses proteolysis (Mortimore and Poso, 1987). Dietary tryptophan increased(Ponter et al., 1994) whereas arginine supplementation reduced (Bruins et al., 2002b) hepaticprotein synthesis in pigs. Other dietary components that stimulate mucosal proliferation andcell turnover also tend to increase protein synthesis. Feeding fiber and lectins has been shown


to stimulate intestinal protein synthesis in some cases (Southon et al., 1985; Palmer et al.,1987), but not in others (Nyachoti et al., 2000). Studies show that neonatal pigs fed elemen-tal diets have higher rates of protein synthesis than those fed a polymeric, cow’s milk formula(Stoll et al., 2000c).

3.2. Anabolic and catabolic hormones

3.2.1. Insulin

Insulin is generally considered to be one of the most anabolic hormones in the body, affect-ing multiple metabolic pathways, including those involving protein and amino acidmetabolism. However, with respect to protein and amino acid metabolism, the effects ofinsulin on splanchnic tissues (i.e. liver and gut) are somewhat different from other majortissues, such as muscle. Early studies in vivo and with perfused livers and isolated hepatocytesfrom normal and streptozotocin-induced diabetic rats indicated that insulin increased proteinsynthesis and suppressed proteolysis (McNurlan and Garlick, 1981; Mortimore andPoso, 1987; Kimball and Jefferson, 1994). The effects of insulin on hepatic protein synthesisare most pronounced for albumin. In the intestine, however, diabetes has no effect on proteinsynthesis despite the induction of significant hyperphagia. More recently, several in vivostudies in pigs, mice, and humans suggest that insulin has either no effect or suppresseshepatic protein synthesis (Mosoni et al., 1993; Nair et al., 1995; Bark et al., 1998; Meeket al., 1998; Ahlman et al., 2001, Boirie et al., 2001; Davis et al., 2001, 2002a; Nygren andNair, 2003). Similarly, studies in pigs reported no effect of insulin on intestinal protein syn-thesis (Davis et al., 2001, 2002a), whereas insulin administration to adult, type I diabeticsubjects only modestly (~15%) increased intestinal protein synthesis (Charlton et al., 2000).Thus, with respect to protein synthesis, the splanchnic tissues appear to be relatively insulin-insensitive. This finding is consistent with the idea that protein synthesis in the splanchnictissues is more tightly regulated by nutrient availability, especially amino acids, than byhormones.

The early observations that insulin suppresses proteolysis in perfused liver and hepatocyteshave been supported by recent in vivo studies in humans (Nair et al., 1995; Nygren and Nair,2003). Moreover, the studies have established several cellular signaling pathways that appearto mediate the insulin-induced suppression of hepatic proteolysis (Kadowaki and Kanazawa,2003; Schliess and Haussinger, 2003). In isolated hepatocytes, insulin activates its membranereceptor, which triggers activation of downstream factors including PI3-kinase, protein kinase B,and p70S6-kinase (fig. 6) (Krause et al., 2002). The key amino acid-sensing and insulin-inducible signaling component involved in cell protein metabolism is mTOR, yet it is unclearwhether mTOR mediates the insulin-induced suppression of proteolysis in hepatocytes. Withrespect to protein metabolism, the explanation for the observed insulin-insensitivity of intes-tinal tissue is largely unexplored, yet intestinal epithelial cells express insulin receptors andexpress all of the signaling pathways known to be responsive to insulin. Cell volume or hydra-tion state is another key cellular signaling process found to be involved in insulin-mediatedsuppression of hepatic proteolysis. Studies in hepatocytes indicate that increased cell volumeis critical for many of the insulin-mediated effects. Insulin produces cell swelling by induc-tion of sodium and potassium accumulation. The signaling mechanisms involved withinsulin-induced cell swelling are not fully established, although cell swelling has been shownto activate three major mitogen-activated protein (MAP)-kinase pathways (ERK, JNK,p38MAPK) (Haussinger and Schliess, 1999).


3.2.2. Growth hormone–insulin-like growth factor I axis

Another major anabolic influence on protein metabolism in developing mammals is the soma-totropic axis, which involves pituitary growth hormone (GH) secretion and local expressionand secretion of insulin-like growth factors (IGF), mainly IGF-I. The somatomedin theoryoriginally proposed was based on the idea that GH-induced secretion of IGF-I in the liver isa key endocrine signal for somatic growth and metabolism during postnatal development(Butler et al., 2002). However, recent use of tissue-specific gene deletion of hepatic IGF-I haschallenged this idea and suggestes that the endocrine role of circulating IGF-I is not essentialfor normal postnatal growth. Both GH and IGF-I appear to have trophic effects in splanchnictissues that appear to be mediated by stimulation of protein and amino acid metabolism; how-ever, there is considerably less information than for skeletal muscle. The receptors for bothGH and the type I IGF receptor are expressed in liver and intestine tissues. Systemic admin-istration of either GH or IGF-I has been shown to stimulate liver and intestinal growth ingrowing animals, yet the response is dependent on stage of development, being less respon-sive during early postnatal life (Etherton and Bauman, 1998; Wester et al., 1998). Thediminished effect of GH in neonatal animals is due to lower expression of the GH receptor.

Early studies with rats in vivo and perfused livers indicated that hepatic protein synthesis,particularly albumin, and amino acid transport were suppressed by hypophysectomy andrestored with growth hormone treatment (Jefferson et al., 1975; Feldhoff et al., 1977). Thesefindings have been confirmed by more recent studies in pigs, rats, and humans showingGH-mediated increases in liver protein synthesis (Pell and Bates, 1992; Wester et al., 1998; Barleet al., 1999, 2001; O’Leary et al., 2003; Bush et al., 2003b). In pigs, the stimulation of liver pro-tein synthesis occurred in both fasted and fed animals and was associated with increasedribosome number rather than translational mechanisms (Bush et al., 2003a). Evidence of GH


Fig. 6. Intracellular signaling pathways for selected hormones and nutrients that alter tissue protein metabo-lism. Abbreviations: cAMP, cyclic AMP; PKA, protein kinase A; AMPK, adenosine monophosphate-activatedprotein kinase; ATP, adenosine triphosphate; GLP-2, glucagon-like peptide 2: GCN2, general control non-depressing kinase-2 amino acid-regulated eukaryotic initiation factor kinase; mTOR, mammalian target ofrapamycin; eIF2B, eukaryotic initiation factor (eIF) 2B; PKB, protein kinase B; IRS, insulin receptor substrate;GSK3, glycogen synthase kinase-3; 4E-BP1, eukaryotic initiation factor 4E binding protein-1; p70S6K, 70 kDaribosomal protein S6 protein kinase; PI3K, phosphatidylinositol 3-kinase; IGF-I, insulin-like growth factor I.

action in the gut is limited, with two studies reporting a stimulation of protein synthesis inintestinal (Lo and Ney, 1996) and PDV tissues (Bush et al., 2003b).

An important consideration with respect to the mechanism of GH action is whether theeffects are mediated directly via the GH receptor or indirectly via increased local expressionof IGF-I. Interpretation of this issue has been complicated by studies in which IGF-I infusionwas found to stimulate hepatic (Douglas et al., 1991; Koea et al., 1992) and intestinal (Lo andNey, 1996; Tashiro et al., 1999) protein synthesis in some cases, but not in others (Pell andBates, 1992; Ling et al., 1995; Lo and Ney, 1996; Bark et al., 1998; Davis et al., 2002b).However, enteral administration of IGF-I in neonatal piglets and mice has no effect on liver orintestinal protein synthesis (Burrin et al., 1999a, 2001). The observation of consistentresponses to GH, but not IGF-I infusion alone, suggests a direct effect of GH on hepaticprotein synthesis. However, there is virtually no information describing the intracellularsignaling pathways, which link GH receptor signals (i.e. Janus family kinases [JAK] and signaltransducers and activators of transcription [STAT]) to the cellular protein synthesis machinery.

To the extent that the anabolic effects of GH are driven by IGF-I production, it is of interestthat studies in vivo and in cultured hepatocytes indicate that the hepatic response to GH isreduced by limiting dietary protein intake or amino acid availability (Harp et al., 1991;Brameld, 1997; Brameld et al., 1999). Recent studies in hepatocytes suggest that the avail-ability of amino acids and glucose directly suppresses IGF-I and GH receptor expression,respectively (Brameld et al., 1999; Stubbs et al., 2002). Moreover, some specific amino acids(methionine, lysine, leucine, tryptophan) appear to be essential for the GH-mediated induc-tion of IGF-I expression in hepatocytes. A further contributing factor to the positiveinteraction between dietary protein and GH-induced IGF-I expression is insulin. Studies indi-cate that insulin stimulates hepatic IGF-I expression, thus increased insulin secretion inresponse to higher dietary protein may contribute to the stimulation of IGF-I expression(Boni-Schnetzler et al., 1991; Brameld, 1997). Other hormones, namely thyroxine and gluco-corticoids, have been shown to increase hepatic GH receptor and IGF-I expression when givenin combination with GH, in some cases (Brameld, 1997) but not others (Beauloye et al., 1999).

The suppression of hepatic amino acid catabolism and urea synthesis also contributes to theprotein anabolic effect of GH. A number of studies in rodents, pigs, and humans have shownthat GH reduces the in vivo synthesis of urea, activity of urea cycle enzymes, and catabolismof essential amino acids (Dahms et al., 1989; Blemings et al., 1996; Grofte et al., 1997; Gahlet al., 1998; Bush et al., 2002). GH has also been reported to increase the expression ofhepatic glutamine synthetase (Nolan et al., 1990). Whether these effects are mediated directlyvia GH receptor signaling or indirectly via IGF-I expression is unknown and warrants furtherstudy. In addition, it is unknown how the intracellular signaling pathways associated with GHand IGF-I are linked to cytosolic and mitochondrial amino acid catabolic and urea cycleenzymes. Thus, the general metabolic effect of GH is to reduce amino acid catabolism andincrease scavenging of ammonia in the liver, thereby channeling amino acids into proteinsynthesis in both the liver and their release to peripheral tissues, such as muscle.

3.2.3. Glucocorticoids and glucagon

Glucocorticoids play a critical role in the induction of differentiation and development ofmany organ systems in mammals. However, glucocorticoids appear to have tissue-specificactions on protein metabolism, being catabolic in skeletal muscle and intestine and anabolicin the liver. Pharmacological doses of glucocorticoids (e.g. dexamethasone) usually increaseliver protein mass, thus its classification as a catabolic hormone is not strictly correct.


Early studies in rodents in vivo and in cultured hepatocytes showed that corticosterone ordexamethasone treatment induced a significant stimulation of constitutive and secretory hepaticprotein synthesis (Odedra et al., 1983; Hutson et al., 1987; Southorn et al., 1990).Glucocorticoids also stimulate proteolysis in hepatocytes isolated from weanling but not fetal orneonatal rats, which may in part explain the generally higher state of liver protein anabolism inearly development (Hopgood et al., 1981; Blommaart et al., 1993). However, under conditionswhere plasma glucocorticoids are physiologically elevated, as in starvation and stress, hepaticprotein mass declines and gluconeogenic enzyme activity is increased. Consistent with this,dexamethasone has been shown to increase the hepatic activity of amino acid catabolic enzymesinvolved in transsulfuration and branched-chain amino acid metabolism (Huang and Chuang,1999; Ratnam et al., 2002). These findings in the liver contrast with the suppression of proteinsynthesis by glucocorticoids in intestine (Burrin et al., 1999c; Boza et al., 2001) and muscle(Southorn et al., 1990). Dexamethasone results in a catabolic suppression of intestinal growth, aswell as a stimulation of the transport and metabolism of amino acids, especially glutamine(Souba et al., 1985; Salleh et al., 1988; Iannoli et al., 1998). The dexamethasone induction ofglutamine metabolism is associated with an increase in glutaminase activity and gene expression.

Glucagon is generally considered as a catabolic hormone associated with starvation, butalso is significantly increased in the circulation during high-protein feeding. The availableliterature describing the effects of glucagon on splanchnic tissue protein metabolism is con-fined largely to the liver. Studies in vivo and in isolated hepatocytes showed that glucagonstimulates hepatic amino acid catabolism, urea cycle enzyme activities, and urea synthesis(Morris, 2002). Most of the effects of glucagon are considered to be mediated by inductionof cellular cAMP production. These studies have reported a 2-fold stimulation of methionineuptake and a 5-fold stimulation of transsulfuration via cystathionine β-synthase in glucagon-treated hepatocytes (Jacobs et al., 2001). Glucagon treatment also increased threonine uptakeand oxidation via activation of threonine dehydratase activity in hepatocytes (House et al.,2001). Oxidation of arginine, but not ornithine, in hepatocytes is increased by glucagon treat-ment (O’Sullivan et al., 2000). Studies with perfused livers have shown an increasedincorporation of glutamine nitrogen into urea in association with increased glutaminase activ-ity after glucagon treatment (Brosnan et al., 1995; Nissim et al., 1999). With respect to proteinturnover, studies with hepatocytes and perfused livers indicate that glucagon stimulates bothalbumin synthesis (Kimball et al., 1995) and autophagic proteolysis (Mortimore and Poso,1987). There is little known about the effect of glucagon, per se, on the intestinal proteinmetabolism. However, there are numerous studies that have examined the effects of glucagon-like peptides (GLP-1 and GLP-2) on the intestine and pancreas (Drucker, 2002; Burrin et al.,2003b); yet few of these have examined aspects of protein and amino acid metabolism. GLP-1and GLP-2 have anabolic effects on the pancreatic beta cells and small intestinal mucosa,respectively. Recent studies in TPN-fed neonatal piglets have demonstrated that chronic GLP-2treatment prevents mucosal atrophy by suppressing apoptosis and proteolysis (Burrin et al.,2000a), whereas acute GLP-2 infusion upregulates intestinal amino acid uptake and proteinsynthesis (Guan et al., 2003).

3.3. Infection, inflammation, and commensal microflora

3.3.1. Pathogenic infection

In the past twenty years, it has become increasingly evident that the relationship between thehost and the ecology of resident microbes plays an integral role in the normal development,metabolism, and survival of mammalian species (Hooper et al., 2002). The infestation with


pathogenic microbial and viral organisms, or exposure to the toxins they produce, has a potentprotein anabolic effect on splanchnic tissues, whereas protein catabolism is increased inskeletal muscle in response to infection and inflammation (MacRae, 1993; Grimble, 2001).Proinflammatory cytokines, (i.e. tumor necrosis factor, interleukin-1, and interleukin-6) arethe key signals that transmit the presence of pathogenic insult to the organism resulting inactivation of the immune system and acute-phase response and suppression of feeding behavior(Johnson, 1997, 2002). Splanchnic tissues play a central role in the proinflammatory responsesince the liver, spleen, and intestine are the major lymphoid tissues and primary sites of acute-phase protein synthesis in the body. The acute-phase response functions to mobilizeendogenous amino acids from muscle, which are used for support of acute-phase protein syn-thesis and cell-mediated immune response. Numerous studies have demonstrated thattreatment with proinflammatory stimuli, including live bacteria, enteric parasites, endotoxin,and specific cytokines, significantly increases acute-phase protein synthesis in visceral tis-sues, especially the liver, gut, and spleen (Jepson et al., 1986; Vary and Kimball, 1992;von Allmen et al., 1992; Higashiguchi et al., 1994; Breuille et al., 1998; Wang et al., 1998;Breuille et al., 1999; Mack et al., 1999). Conversely, these factors increase catabolism and netloss of muscle protein mass. In growing animals, the amino acids required to maintain theproinflammatory, acute-phase protein synthesis in splanchnic tissues impart a metabolic cost,which results in suppression of muscle protein synthesis and increased catabolism and loss ofskeletal muscle mass. In addition to suppressed growth, the loss of lean body mass coupledwith disruption of organ function contributes to the increased morbidity and mortality associatedwith infection.

3.3.2. Commensal microflora

The phenomenon described above illustrates an acute mechanism whereby the pathogenicmicroflora activate the immune system, suppress the growth rate, induce fever and diarrhea,and in some cases cause death in domestic animals. In the long term, however, over the life-span of all mammals beginning at birth, animals are naturally colonized with commensalmicrobes and these mostly bacterial species coexist with the host organism. This symbiosisserves to activate the maturation of the immune system and enables nonruminant animals toutilize dietary carbohydrate fermented to short-chain fatty acids from otherwise indigestibleplant polysaccharides and recycle body nitrogen when dietary protein availability is limited(Fuller and Reeds, 1998; Hooper et al., 2002). Despite the general mutually beneficial rela-tionship between commensal microbes and the host, there remains to a lesser degree someactivation of the immune system. The manifestations of commensal microbes have beendemonstrated most clearly in animals reared under germ-free compared to conventional envi-ronments. These studies indicate that commensal microbes modestly reduced food intake,increased metabolic rate, and increased mass and metabolic activity of gastrointestinal tissues(Gaskins et al., 2002). The increase in gastrointestinal mass is associated with increasednumbers and activity of lymphoid cells and increased proliferation of epithelial crypt cells.As with more severe, pathogenic infection, the presence of commensal bacteria results insuppression of growth, albeit of lesser magnitude.

In the past fifty years, antimicrobial compounds have been fed to domestic animals in orderto suppress the activity of the gut microflora and enhance growth. Despite the widespread useand success of antimicrobials, however, their exact mechanism of action remains poorly defined.It has been shown that by suppressing microbial activity, antimicrobials reduce the luminalconcentration and associated toxic insult of ammonia, and thereby diminish the thickness and


mass of the intestinal mucosa and associated lymphoid tissue (Visek, 1978). Studies in pigsand chickens have shown that feeding antimicrobial compounds significantly reduces liverand small intestinal mass, cell proliferation, protein synthesis, and intestinal ammonia absorp-tion (Muramatsu et al., 1983; Yen et al., 1987; Yen and Pond, 1990; Krinke and Jamroz,1996). Additional evidence indicates that much of the luminal ammonia originates frombacterial hydrolysis of urea and deamination of dietary amino acids. Thus, it appears thatpart of the protein anabolic response of antimicrobials is associated with three phenomena:(1) reduced microbial degradation of dietary essential amino acids, (2) increased intestinalabsorption of dietary amino acids, and (3) reduced utilization of dietary amino acids bysplanchnic tissues for maintenance of immune-associated lymphoid cells and acute-phaseprotein synthesis.

3.3.3. Immunonutrients

During the proinflammatory, acute-phase response, the metabolic basis for increased splanch-nic amino acid utilization has been linked to several cellular immune functions, includingsynthesis of specific acute-phase proteins, glutathione synthesis and antioxidant function,lymphocyte and mucosal crypt cell proliferation, and nitric oxide production. Researchershave determined that a particular group of nutrients, termed “immunonutrients”, play a keysupportive role in these immune-related processes, serving to either act as biosynthetic sub-strates or alter cellular function (Grimble, 2001; Huang et al., 2003). Among those consideredas immunonutrients are the sulfur amino acids, glutamine, arginine, nucleotides, and omega-3 fatty acids. There are numerous acute-phase proteins synthesized by the liver, whose plasmaconcentration increases significantly during infection and trauma, including C-reactive pro-tein, serum amyloid A, fibrinogen, and haptoglobin. Cysteine is considered to be one of thelimiting amino acids for acute-phase protein synthesis, based on the estimated balance ofamino acids released from muscle proteolysis (Reeds et al., 1994). Cysteine is also a con-stituent amino acid of glutathione and thioredoxin, both major cellular antioxidants (seesection below). Studies in infected and stressed rats indicate that cysteine utilization and GSHsynthesis and concentration in splanchnic tissues increased markedly (Malmezat et al., 1998,2000a; Mercier et al., 2002). In protein-malnourished children and pigs, GSH concentrationsand synthesis rates in response to infection and stress are compromised, but can be restoredwith cysteine supplementation (Jahoor et al., 1995; Reid et al., 2000; Badaloo et al., 2002).The increased demand for cysteine during infection also markedly stimulated methionineutilization via the transsulfuration pathway (Malmezat et al., 2000b). Recent studies provideevidence that the proinflammatory cytokine, tumor necrosis factor-α, directly activatescystathionine β-synthase activity, the enzyme catalyzing transsulfuration (Zou and Banerjee,2003). In addition, methionine and cysteine are precursors for taurine, which also functionsas an important antioxidant in phagocytic cells (e.g. neutrophils) via stable neutralization ofintracellular hypochlorous acid (Santangelo, 2002). Hepatic taurine synthesis also increased3-fold in infected rats (Malmezat et al., 2000a).

Glutamine and arginine also are considered immunonutrients because of their ability tostimulate lymphoid and epithelial cell function and serve as precursors for nitric oxide, glu-tathione, and nucleotide synthesis. Glutamine increased proliferation of lymphocytes andintestinal crypt cells, increased macrophage phagocytic activity, reduced mucosal inflamma-tory cytokine production, improved intestinal epithelial tight junction function, and increasedsurvival in bacterial-infected mice (Wilmore and Shabert, 1998; Huang et al., 2003).Glutamine can be readily deaminated to glutamate in gut, liver, and lymphoid cells and can


also serve as a precursor for glutathione. Glutamine and glutamate nitrogen moieties are usedfor nucleotide synthesis and nucleotides have been shown to enhance lymphoid cell function.Sepsis has been shown to significantly increase uptake of glutamine by the liver, but not theintestine (Karinch et al., 2001). Arginine deficiency induced by overexpression of intestinalarginase I compromised the development of B-cell-related gut-associated lymphoid tissue(de Jonge et al., 2002). Studies in endotoxemic pigs indicate that the liver and gut nitric oxideproduction is increased in parallel with arginine utilization, whereas arginine is released frommuscle (Bruins et al., 2002a).

4. AMINO ACIDS AS EXTRACELLULAR SIGNALS

4.1. Glutamine

There is a clear recognition that extracellular amino acid availability has profound effects onmany aspects of cell function, including the control of cell signaling, gene expression, cellvolume, cell proliferation, apoptosis, and protein turnover. The precise cellular mechanismsby which amino acids are able to elicit control over such diverse processes have become thefocus of intense investigation recently (fig. 6) (McDaniel et al., 2002; Averous et al., 2003;Jefferson and Kimball, 2003; Kadowaki and Kanazawa, 2003; Meijer, 2003). Many of thenew developments in this area stem from the discovery of nutrient responsive genes and theirfunction in Drosophila and yeast systems. The effects of glutamine have been of particularinterest, since it has been shown to have pluripotent actions, including stimulation of cell pro-liferation, protein synthesis, differentiation, ornithine decarboxylase (ODC) and immediateearly gene (c-jun) expression, and polyamine synthesis (Higashiguchi et al., 1993; Kandilet al., 1995; Wang et al., 1996). In a recent review, Rhoads (1999) postulated that glutaminestimulates cell proliferation by a signaling mechanism that involves the activation of tworelated, but distinct, classes of mitogen-activated protein kinases (MAPK). Glutamine alsosuppresses apoptosis and protein degradation in intestinal and liver tissue, which suggests thatit may be an important survival factor (Mortimore and Poso, 1987; Papaconstantinou et al.,1998; Coeffier et al., 2003). Glutamine has been shown to suppress proteolysis in hepatocytesin association with increased cell hydration and swelling, a process that involves the MAPKpathways, extracellular-regulated kinases (ERK), and p38MAPK (Haussinger et al., 2001).Recent studies also indicate that the antiapoptotic effect of glutamine involves an inhibitoryinteraction between glutaminyl-tRNA synthetase and apoptosis signal-regulating kinase 1(Ko et al., 2001).

4.2. Leucine

Leucine is another amino acid that has been shown to have important cell regulatory effectsin various tissues, including the liver and pancreatic beta-cells, yet there is limited informa-tion of its effects in intestinal cells. Studies in perfused livers and hepatocytes demonstratedthat leucine was capable of suppressing proteolysis and stimulating protein synthesis(Mortimore and Poso et al., 1987; Anthony et al., 2001a). In vivo studies also have shown thatfeeding a leucine-deficient diet or leucine alone affects the expression of ribosomal proteinmRNA and activation of intracellular signaling molecules involved in protein synthesis(Anthony et al., 2001b). In vivo studies in diabetic rats indicate that the effects of leucineappear to be mediated, in part, via an insulin-independent pathway. The specific actionof leucine may be mediated by an interaction with a cell-membrane binding protein, since


cell-impermeant and nonmetabolizable analogs of leucine also suppress hepatic proteolysis andstimulate protein synthesis (van Sluijters et al., 2000; Lynch, 2001; Lynch et al., 2002). Othershave suggested that leucine metabolism is important for cellular activation, based on reports thatBCAA ketoacids activate downstream signaling elements involved in protein synthesis.

The intracellular signaling pathway that mediates the effects of leucine on protein turnoverhas been studied intensively in recent years. Amino acid deprivation, including leucine, acti-vates the general control nonderepressing kinase 2 (GCN2), resulting in phosphorylation ofeukaryotic initiation factor-2 (eIF2), which leads to suppression of global protein synthesis(Jefferson and Kimball, 2003). Specific deprivation of BCAA in cultured lymphocytes acti-vates the branched-chain α-ketoacid dehydrogenase (BCKD) kinase, which translates intoreduced BCKD activity and BCAA catabolism (Doering and Danner, 2000). Other players inthe amino acid signaling pathway include those involved in assembly of initiation factors(eIF4) and phosphorylation of ribosomal S6. A key intermediate in this pathway is the mam-malian target of rapamycin (mTOR), which is a serine–threonine kinase involved incoordinating nutrient availability with cell growth and proliferation. mTOR represents a pointof convergence in the pathways that mediate the stimulation of protein synthesis by aminoacids and insulin. Studies in hepatocytes show that leucine, glutamine, and insulin activatep70 ribosomal S6 kinase, which is downstream of mTOR (Krause et al., 2002). However,insulin also activates signals upstream from mTOR, namely phosphatidylinositol 3-kinase(PI3K) and protein kinase B (PKB), whereas leucine and glutamine do not affect theseintermediates.

The factors that transduce the signal between leucine (or other effector amino acids) andmTOR is the focus of intense interest. One such mechanism may involve adenosinemonophosphate-activated protein kinase (AMPK), which senses cellular AMP levels and isactivated under conditions of nutrient deprivation and ATP depletion. Activation of AMPKleads to suppression of protein synthesis and mTOR phosphorylation and stimulation ofautophagic proteolysis (Kimura et al., 2003; Meijer, 2003). The role of AMPK is intriguingin light of evidence that mTOR seems to function as a cellular ATP-sensing molecule (Denniset al., 2001). Another potential upstream activator of mTOR may be phosphatidic acid, whichis produced by the action of phospholipase D (Chen and Fang, 2002). This model of mTORactivation by phosphatidic acid could form a link to other cellular signals such as the Rhofamily of G proteins, protein kinase C, and intracellular calcium. Intracellular aminoacyl-tRNAsynthetases also have been implicated in the regulation of mTOR.

4.3. Sulfur amino acids

Sulfur amino acids (SAA), especially methionine and cysteine, play a key role in antioxidantstatus and cellular function (Sen, 1998; Aw, 1999; Deplancke and Gaskins, 2002).Glutathione and thioredoxin are the most important cellular antioxidants in mammals andhave a critical function in reacting with reactive oxygen species and maintaining cellularredox status. Reduced glutathione (GSH) is an ubiquitous tripeptide (Glu-Cys-Gly) presentthroughout the body at relatively high intracellular concentrations, especially in the smallintestine. Cellular GSH homeostasis is maintained through de novo synthesis from precursorSAA methionine and cysteine, regeneration from its oxidized form glutathione disulfide(GSSG), and uptake of extracellular intact GSH. Thioredoxin is another major cellular anti-oxidant; it is a larger protein (12 kDa) than GSH, but also contains key cysteine residues in thecatalytic active site. Mediating oxidant stress and maintaining normal redox status is espe-cially important in intestinal and liver cells. Studies with intestinal epithelial cells indicate


that increased oxidant stress and redox imbalance suppress cell proliferation and induce apop-tosis and that this is closely correlated with a higher oxidized glutathione state, as measuredby the ratio of GSH : GSSG (Noda et al., 2001; Jonas et al., 2002; Pias and Aw, 2002). Culturestudies also show that cells grown in cysteine-deficient media have suppressed GSH concen-trations and cell proliferation rates, both of which are stimulated with increased cysteinesupplementation (Miller et al., 2002; Noda et al., 2002). Other studies with human colonicepithelial cells (Caco-2) indicate that, as differentiation proceeds, cell GSH concentration andproliferation rate decrease, whereas apoptosis rate increases (Nkabyo et al., 2002).Collectively, these studies suggest that cysteine availability and local GSH concentration havea direct influence on epithelial cell proliferation and survival and are inversely proportionalto cellular differentiation state.

Given evidence that cysteine availability is important for maintenance of epithelial cellGSH level and cell redox status, the question has been raised as to whether methionine can,via transsulfuration, affect the cysteine availability. Evidence in support of this idea is thefinding that, in HepG2 cells, oxidant stress increased transsulfuration measured by cystathio-nine synthesis and 35S-methionine incorporation into glutathione (Mosharov et al., 2000).Moreover, studies in HepG2 cells also show that cystathionine synthase activity is coordi-nately regulated with proliferation via a redox-sensitive mechanism (MacLean et al., 2002).These results imply that cells exposed to oxidant stress may meet the increased cysteinerequirement for GSH synthesis via activation of methionine transsulfuration. A broader impli-cation of these results is that methionine availability and its conversion to cysteine and GSHvia transsulfuration may be important for maintenance of normal intestinal and hepatic cellproliferation and survival.


The splanchnic tissues represent a quantitatively and functionally significant component ofwhole-body protein and amino acid metabolism of growing animals. The gastrointestinal andliver tissues consume a substantial fraction of the dietary amino acid intake for the purposes ofoxidative metabolism, protein synthesis, and gluconeogenesis. Relatively few nonessentialamino acids (glutamine, glutamate, aspartate) appear to be major oxidative fuels in the intes-tine and this metabolism occurs in both epithelial and lymphoid cells. Although ammoniaappears to be the main fate of nitrogen derived from glutamine and glutamate catabolism,incorporation of these nitrogen moieties into nucleotides and glucosamines also may be func-tionally significant. However, some essential amino acids are also oxidized in the intestine, yetthe biochemical and cellular bases for this metabolism are poorly understood. Preliminarystudies suggest that essential amino acid catabolism occurs in epithelial cells, but further stud-ies are warranted to examine the impact of the gut microflora on this process, especially in thesmall intestine. Whether this phenomenon is a central mechanism explaining the growth-promoting action for antimicrobials has yet to be proven. However, there is a compelling needto address this question given the increasing concern about the impact of antimicrobials on theenvironment and human health. Among the factors that regulate splanchnic tissue protein andamino acid metabolism in healthy animals, nutrition plays a major role and it is apparent thatthese tissues may be more responsive to extracellular nutrient availability than endocrine sig-nals. However, under conditions of infection or stress, proinflammatory cytokines are keyactivators of hepatic acute-phase protein synthesis, but stimulate protein catabolism in muscle.This phenomenon serves to shuttle amino acids from muscle into hepatic protein synthesis, yetit is not clear how the same cytokines differentially affect protein turnover in these two tissues.


The answers to many of these questions will require a greater understanding of the intra-cellular signaling pathways that link both extracellular nutrient availability and endocrinesignals to cellular protein and amino acid metabolism. Despite the explosion of knowledge ofcellular signaling mechanisms in the past twenty years, there have been few efforts to exploretheir relevance to protein and amino acid metabolism in splanchnic tissues, especially theintestine. In the past, animal science has legitimately focused considerable attention on thegrowth and metabolism of skeletal muscle and adipose tissue, given its relevance to leantissue growth. Yet, the splanchnic tissues act as key metabolic regulators of lean tissue growth,serving to alter the peripheral availability of dietary nutrients and transmit nutrient-dependentendocrine signals that determine peripheral tissue metabolism. Previous studies of physiologyand whole animal metabolism combined with clinical trials have shown that some nutrientsappear to be unique in their ability to ameliorate the metabolic effects of diseases. Moreover,these nutrients seem to affect splanchnic tissue function, especially the immune system. Thus,further study with these immunonutrients is necessary to establish their mechanism of actionand whether they can be used to improve growth and the health of domestic animals. Anexpanding range of analytical tools, from the molecular to the whole-organ level, is nowbecoming available to explore these issues in growing animals. The use of genomic and pro-teomic analysis of individual cells and tissues coupled with genetic manipulation of animalswill aid in determining the essentiality and function of specific regulators of protein andamino acid metabolism. In addition, the novel application of techniques such as germ-freeenvironments, mass isotopomer analysis, laser capture microdissection, and transorgan bal-ance approaches should provide useful information on the localization of amino acidmetabolism in specific tissues and cells and how this is altered by the host environment.

REFERENCES

Adegoke, O.A., McBurney, M.I., Samuels, S.E., Baracos, V.E., 1999. Luminal amino acids acutelydecrease intestinal mucosal protein synthesis and protease mRNA in piglets. J. Nutr. 129, 1871–1878.

Adegoke, O.A., McBurney, M.I., Samuels, S.E., Baracos, V.E., 2003. Modulation of intestinal proteinsynthesis and protease mRNA by luminal and systemic nutrients. Amer. J. Physiol. 284, G1017–G1026.

Ahlman, B., Charlton, M., Fu, A., Berg, C., O’Brien, P., Nair, K.S., 2001. Insulin’s effect on synthesisrates of liver proteins: a swine model comparing various precursors of protein synthesis. Diabetes 50,947–954.

Alpers, D.H., 1972. Protein synthesis in intestinal mucosa: the effect of route of administration of pre-cursor amino acids. J. Clin. Invest. 51, 167–173.

Anthony, T.G., Anthony, J.C., Yoshizawa, F., Kimball, S.R., Jefferson, L.S., 2001a. Oral administrationof leucine stimulates ribosomal protein mRNA translation but not global rates of protein synthesis inthe liver of rats. J. Nutr. 131, 1171–1176.

Anthony, T.G., Reiter, A.K., Anthony, J.C., Kimball, S.R., Jefferson, L.S., 2001b. Deficiency of dietaryEAA preferentially inhibits mRNA translation of ribosomal proteins in liver of meal-fed rats. Amer.J. Physiol. 281, E430–E439.

Argaman, Z., Young, V.R., Noviski, N., Castillo-Rosas, L., Lu, X.M., Zurakowski, D., Cooper, M.,Davison, C., Tharakan, J.F., Ajami, A., Castillo, L., 2003. Arginine and nitric oxide metabolism incritically ill septic pediatric patients. Crit. Care Med. 31, 591–597.

Attaix, D., Arnal, M., 1987. Protein synthesis and growth in the gastrointestinal tract of the youngpreruminant lamb. Brit. J. Nutr. 58, 159–169.

Attaix, D., Meslin, J.C., 1991. Changes in small intestinal mucosa morphology and cell renewal insuckling, prolonged-suckling, and weaned lambs. Amer. J. Physiol. 261, R811–R818.

Attaix, D., Manghebati, A., Grizard, J., Arnal, M., 1986. Assessment of in vivo protein synthesis in lambtissues with [3H]valine flooding doses. Biochim. Biophys. Acta 882, 389–397.

Attaix, D., Aurousseau, E., Rosolowska-Huszcz, D., Bayle, G., Arnal, M., 1992. In vivo longitudinal vari-ations in protein synthesis in developing ovine intestines. Amer. J. Physiol. 263, R1318–R1323.


Averous, J., Bruhat, A., Mordier, S., Fafournoux, P., 2003. Recent advances in the understanding ofamino acid regulation of gene expression. J. Nutr. 133, 2040S–2045S.

Aw, T.Y., 1999. Molecular and cellular responses to oxidative stress and changes in oxidation-reductionimbalance in the intestine. Amer. J. Clin. Nutr. 70, 557–565.

Badaloo, A., Reid, M., Forrester, T., Heird, W.C., Jahoor, F., 2002. Cysteine supplementation improvesthe erythrocyte glutathione synthesis rate in children with severe edematous malnutrition. Amer. J. Clin.Nutr. 76, 646–652.

Ball, R.O., Law, G., Bertolo, R.F.P., Pencharz, P.B., 1999. Adequate oral threonine is critical for mucinproduction and mucosal growth by the neonatal piglet gut. VIIIth International Symposium on ProteinMetabolism and Nutrition, Aberdeen, UK, p. 31. (Abstr.)

Ballevre, O., Cadenhead, A., Calder, A.G., Rees, W.D., Lobley, G.E., Fuller, M.F., Garlick, P.J., 1990.Quantitative partition of threonine oxidation in pigs: effect of dietary threonine. Amer. J. Physiol. 259,E483–E491.

Baracos, V.E., Samuels, S.E., Adegoke, O.A., 2000. Anabolic and catabolic mediators of intestinalprotein turnover: a new experimental approach. Curr. Opin. Clin. Nutr. Metab. Care 3, 183–189.

Bark, T.H., McNurlan, M.A., Lang, C.H., Garlick, P.J., 1998. Increased protein synthesis after acuteIGF-I or insulin infusion is localized to muscle in mice. Amer. J. Physiol. 275, E118–E123.

Barle, H., Essen, P., Nyberg, B., Olivecrona, H., Tally, M., McNurlan, M.A., Wernerman, J., Garlick, P.J.,1999. Depression of liver protein synthesis during surgery is prevented by growth hormone. Amer.J. Physiol. 276, E620–Ε627.

Barle, H., Rahlen, L., Essen, P., McNurlan, M.A., Garlick, P.J., Holgersson, J., Wernerman, J., 2001.Stimulation of human albumin synthesis and gene expression by growth hormone treatment. Clin.Nutr. 20, 59–67.

Barrett, E.J., Gusberg, R., Ferrannini, E., Tepler, J., Felig, P., Jacob, R., Smith, D., DeFronzo, R.A., 1986.Amino acid and glucose metabolism in the postabsorptive state and following amino acid ingestionin the dog. Metabolism 35, 709–717.

Battezzati, A., Brillon, D.J., Matthews, D.E., 1995. Oxidation of glutamic acid by the splanchnic bed inhumans. Amer. J. Physiol. 269, E269–E276.

Battezzati, A., Haisch, M., Brillon, D.J., Matthews, D.E., 1999. Splanchnic utilization of enteral alaninein humans. Metabolism 48, 915–921.

Beaufrere, B., Fournier, V., Salle, B., Putet, G., 1992. Leucine kinetics in fed low-birth-weight infants:importance of splanchnic tissues. Amer. J. Physiol. 263, E214–E220.

Beauloye, V., Ketelslegers, J.M., Moreau, B., Thissen, J.P., 1999. Dexamethasone inhibits both growthhormone (GH)-induction of insulin-like growth factor-I (IGF-I) mRNA and GH receptor (GHR)mRNA levels in rat primary cultured hepatocytes. Growth Horm. IGF Res. 9, 205–211.

Bertolo, R.F., Chen, C.Z., Law, G., Pencharz, P.B., Ball, R.O., 1998. Threonine requirement of neonatalpiglets receiving total parenteral nutrition is considerably lower than that of piglets receiving an iden-tical diet intragastrically. J. Nutr. 128, 1752–1759.

Bertolo, R.F., Brunton, J.A., Pencharz, P.B., Ball, R.O., 2003. Arginine, ornithine, and proline intercon-version is dependent on small intestinal metabolism in neonatal pigs. Amer. J. Physiol. Endocrinol.Metab. 284, E915–E922.

Blachier, F., Darcy-Vrillon, B., Sener, A., Duee, P.H., Malaisse, W.J., 1991. Arginine metabolism in ratenterocytes. Biochim. Biophys. Acta. 1092, 304–310.

Blachier, F., M’Rabet-Touil, H., Posho, L., Morel, M.T., Bernard, F., Darcy-Vrillon, B., Duee, P.H.,1992. Polyamine metabolism in enterocytes isolated from newborn pigs. Biochim. Biophys. Acta1175, 21–26.

Blachier, F., Selamnia, M., Robert, V., M’Rabet-Touil, H., Duee, P.H., 1995. Metabolism of L-argininethrough polyamine and nitric oxide synthase pathways in proliferative or differentiated human coloncarcinoma cells. Biochim. Biophys. Acta 1268, 255.

Blemings, K.P., Gahl, M.J., Crenshaw, T.D., Benevenga, N.J., 1996. Recombinant bovine somatotropindecreases hepatic amino acid catabolism in female rats. J. Nutr. 126, 1657–1661.

Blommaart, P.J., Zonneveld, D., Meijer, A.J., Lamers, W.H., 1993. Effects of intracellular amino acidconcentrations, cyclic AMP, and dexamethasone on lysosomal proteolysis in primary cultures of peri-natal rat hepatocytes. J. Biol. Chem. 268, 1610–1617.

Blouin, J.P., Bernier, J.F., Reynolds, C.K., Lobley, G.E., Dubreuil, P., Lapierre, H., 2002. Effect of supplyof metabolizable protein on splanchnic fluxes of nutrients and hormones in lactating dairy cows.J. Dairy Sci. 85, 2618–2630.


Boirie, Y., Gachon, P., Beaufrere, B., 1997. Splanchnic and whole-body leucine kinetics in young andelderly men. Amer. J. Clin. Nutr. 65, 489–495.

Boirie, Y., Short, K.R., Ahlman, B., Charlton, M., Nair, K.S., 2001. Tissue-specific regulation of mito-chondrial and cytoplasmic protein synthesis rates by insulin. Diabetes 50, 2652–2658.

Boni-Schnetzler, M., Schmid, C., Meier, P.J., Froesch, E.R., 1991. Insulin regulates insulin-like growthfactor I mRNA in rat hepatocytes. Amer. J. Physiol. 260, E846–E851.

Bos, C., Stoll, B., Fouillet, H., Gaudichon, C., Guan, X., Grusak, M.A., Reeds, P.J., Tome, D., Burrin, D.G.,2003. Intestinal lysine metabolism is driven by the enteral availability of dietary lysine in piglets feda bolus meal. Amer. J. Physiol. Endocrinol. Metab. 285, E1246–E1257.

Bouteloup-Demange, C., Claeyssens, S., Maillot, C., Lavoinne, A., Lerebours E., Dechelotte P., 2000.Effects of enteral glutamine on gut mucosal protein synthesis in healthy humans receiving glucocor-ticoids. Amer. J. Physiol. 278, G677–G681.

Boza, J.J., Jahoor, F., Reeds, P.J., 1996. Ribonucleic acid nucleotides in maternal and fetal tissues derivealmost exclusively from synthesis de novo in pregnant mice. J. Nutr. 126, 1749–1758.

Boza, J.J., Turini, M., Moennoz, D., Montigon, F., Vuichoud, J., Gueissaz, N., Gremaud, G., Pouteau, E.,Piguet-Welsch, C., Finot, P.A., Ballevre, O., 2001. Effect of glutamine supplementation of the diet ontissue protein synthesis rate of glucocorticoid-treated rats. Nutrition 17, 35–40.

Brameld, J.M., 1997. Molecular mechanisms involved in the nutritional and hormonal regulation ofgrowth in pigs. Proc. Nutr. Soc. 56, 607–619.

Brameld, J.M., Gilmour, R.S., Buttery, P.J., 1999. Glucose and amino acids interact with hormones tocontrol expression of insulin-like growth factor-I and growth hormone receptor mRNA in cultured pighepatocytes. J. Nutr. 129, 1298–1306.

Breuille, D., Arnal, M., Rambourdin, F., Bayle, G., Levieux, D., Obled, C., 1998. Sustained modifications ofprotein metabolism in various tissues in a rat model of long-lasting sepsis. Clin. Sci. 94, 413–423.

Breuille, D., Voisin, L., Contrepois, M., Arnal, M., Rose, F., Obled, C., 1999. A sustained rat model forstudying the long-lasting catabolic state of sepsis. Infect. Immun. 67, 1079–1085.

Brosnan, J.T., 2000. Glutamate, at the interface between amino acid and carbohydrate metabolism.J. Nutr. 130, 988S–990S.

Brosnan, J.T., 2003. Interorgan amino acid transport and its regulation. J. Nutr. 133, 2068S–2072S.Brosnan, J.T., Ewart, H.S., Squires, S.A., 1995. Hormonal control of hepatic glutaminase. Adv. Enzyme

Regul. 35, 131–146.Bruins, M.J., Lamers, W.H., Meijer, A.J., Soeters, P.B., Deutz, N.E., 2002a. In vivo measurement of nitric

oxide production in porcine gut, liver and muscle during hyperdynamic endotoxaemia.Brit. J. Pharmacol. 137, 1225–1236.

Bruins, M.J., Soeters, P.B., Lamers, W.H., Deutz, N.E., 2002b. L-arginine supplementation in pigsdecreases liver protein turnover and increases hindquarter protein turnover both during and afterendotoxemia. Am. J. Clin. Nutr. 75, 1031–1044.

Brunton, J.A., Bertolo, R.F.P., Pencharz, P.B., Ball, R.O., 1999. L-Arginine supplementation in pigsdecreases liver protein turnover and increases hindquarter protein turnover both during and afterendotoxemia. Amer. J. Physiol. 277, E223–E231.

Burrin, D.G., Ferrell, C.L., Eisemann, J., Britton, R.A., Nienaber, J.A., 1989. Effect of level of nutritionon splanchnic blood flow and oxygen consumption in sheep. Brit. J. Nutr. 62, 23–34.

Burrin, D.G., Davis, T.A., Fiorotto, M.L., Reeds, P.J., 1991. Stage of development and fasting affectprotein synthetic activity in the gastrointestinal tissues of suckling pigs. J. Nutr. 121, 1099–1108.

Burrin, D.G., Shulman, R.J., Reeds, P.J., Davis, T.A., Gravitt, K.R., 1992. Porcine colostrum and milkstimulate visceral organ and skeletal muscle synthesis in neonatal piglets. J. Nutr. 122, 1205–1213.

Burrin, D.G., Shulman, R.J., Langston, C., Storm, M.C., 1994. Supplemental alanylglutamine, organgrowth and nitrogen metabolism in neonatal pigs fed by total parenteral nutrition. J. Parent. Enter.Nutr. 18, 313–319.

Burrin, D.G., Davis, T.A., Ebner, S., Schoknecht, P.A., Fiorotto, M.L., Reeds, P.J., McAvoy, S., 1995.Nutrient-independent and nutrient-dependent factors stimulate protein synthesis in colostrum-fednewborn pigs. Pediat. Res. 37, 593–599.

Burrin, D.G., Davis, T.A., Ebner, S., Schoknecht, P.A., Fiorotto, M.L., Reeds, P.J., 1997. Colostrum enhancesthe nutritional stimulation of vital organ protein synthesis in neonatal pigs. J. Nutr. 127, 1284–1289.

Burrin, D.G., Fiorotto, M.L., Hadsell, D.L, 1999a. Transgenic hypersection of des(1–3) human insulin-like growth factor I in mouse milk has limited effects on the gastrointestinal tract in suckling pups.J. Nutr. 129, 51–56.


Burrin, D.G., Stoll, B., Chang, X., Yu, H., Reeds, P.J., 1999b. Total parenteral nutrition does not com-promise digestion or absorption of dietary protein in neonatal pigs. FASEB J. 13, A1024.

Burrin, D.G., Wester, T.J., Davis, T.A., Fiorotto, M.L., Chang, X., 1999c. Dexamethasone inhibits smallintestinal growth via increased protein catabolism in neonatal pigs. Amer. J. Physiol. 276, E269–E277.

Burrin, D.G., Stoll, B., Jiang, R., Petersen, Y., Elnif, J., Buddington, R.K., Schmidt, M., Holst, J.J.,Hartmann, B., Sangild, P.T., 2000a. GLP-2 stimulates intestinal growth in premature TPN-fed pigs bysuppressing proteolysis and apoptosis. Amer. J. Physiol. 279, G1249–G1256.

Burrin, D.G., Stoll, B., Jiang, R., Chang, X., Hartmann, B., Holst, J.J., Greeley, G.H. Jr., Reeds, P.J.,2000b. Minimal enteral nutrient requirements for intestinal growth in neonatal piglets: how much isenough? Amer. J. Clin. Nutr. 71, 1603–1610.

Burrin, D.G., Stoll, B., Fan, M.Z., Dudley, M.A., Donovan, S.M., Reeds, P.J., 2001. Oral IGF-I alters theposttranslational processing but not the activity of lactase-phlorizin hydrolase in formula-fed neonatalpigs. J. Nutr. 131, 2235–2241.

Burrin, D.G., Stoll, B., Chang, X., van Goudoever, J.B., Fujii, H., Hutson, S., Reeds, P.J., 2003a.Parenteral nutrition results in impaired lactose digestion and hexose absorption when enteral feedingis initiated in infant pigs. Amer. J. Clin. Nutr. 78, 461–470.

Burrin, D.G., Stoll, B., Guan, X., 2003b. Glucagon-like peptide 2 function in domestic animals. Domest.Anim. Endocrinol. 24, 103–122.

Bush, J.A., Wu, G., Suryawan, A., Nguyen, H.V., Davis, T.A., 2002. Somatotropin-induced amino acidconservation in pigs involves differential regulation of liver and gut urea cycle enzyme activity.J. Nutr. 132, 59–67.

Bush, J.A., Burrin, D.G., Suryawan, A., O’Connor, P.M., Nguyen, H.V., Reeds, P.J., Steele, N.C.,van Goudoever, J.B., Davis, T.A., 2003a. Somatotropin-induced protein anabolism in hindquartersand portal-drained viscera of growing pigs. Amer. J. Physiol. 284, E302–E312.

Bush, J.A., Kimball, S.R., O’Connor, P.M., Suryawan, A., Orellana, R.A., Nguyen, H.V., Jefferson, L.S.,Davis, T.A., 2003b. Translational control of protein synthesis in muscle and liver of growth hormone-treated pigs. Endocrinology 144, 1273–1283.

Butler, A.A., Yakar, S., LeRoith, D., 2002. Insulin-like growth factor-I: compartmentalization within thesomatotropic axis? News Physiol. Sci. 17, 82–85.

Buts, J.P., De Keyser, N., De Raedemaeker, L., Collette, E., Sokal, E.M., 1995. Polyamine profiles in humanmilk, infant artificial formulas, and semi-elemental diets. J. Pediat. Gastroenterol. Nutr. 21, 44–49.

Castillo, L., Chapman, T.E., Yu, Y.M., Ajami, A., Burke, J.F., Young, V.R., 1993a. Dietary arginine uptakeby the splanchnic region in adult humans. Amer. J. Physiol. 265, E532–E539.

Castillo, L., de Rojas, T.C., Chapman, T.E., Vogt, J., Burke, J.F., Tannenbaum, S.R., Young, V.R., 1993b.Splanchnic metabolism of dietary arginine in relation to nitric oxide synthesis in normal adult man.Proc. Natl. Acad. Sci. USA 90, 193–197.

Castillo, L., Beaumier, L., Ajami, A.M., Young, V.R., 1996. Whole body nitric oxide synthesis in healthy mendetermined from 15N arginine-to-15N citrulline labeling. Proc. Natl. Acad. Sci. USA 93, 11460–11465.

Cayol, M., Boirie, Y., Rambourdin, F., Prugnaud, J., Gachon, P., Beaufrere, B., Obled, C., 1997. Influence of pro-tein intake on whole body and splanchnic leucine kinetics in humans. Amer. J. Physiol. 272, E584–E591.

Charlton, M., Ahlman, B., Nair, K.S., 2000. The effect of insulin on human small intestinal mucosal protein synthesis. Gastroenterology 118, 299–306.

Chen, J., Fang, Y., 2002. A novel pathway regulating the mammalian target of rapamycin (mTOR) signaling. Biochem. Pharmacol. 64, 1071.

Coeffier, M., Claeyssens, S., Hecketsweiler, B., Lavoinne, A., Ducrotte, P., Dechelotte, P., 2003. Enteralglutamine stimulates protein synthesis and decreases ubiquitin mRNA level in human gut mucosa.Amer. J. Physiol. 285, G266–G273.

Coloso, R.M., Stipanuk, M.H., 1989. Metabolism of cyst(e)ine in rat enterocytes. J. Nutr. 119, 1914–24.Crenn, P., Vahedi, K., Lavergne-Slove, A., Cynober, L., Matuchansky, C., Messing, B., 2003. Plasma cit-

rulline: a marker of enterocyte mass in villous atrophy-associated small bowel disease.Gastroenterology 124, 1210–1219.

Dahms, W.T., Owens, R.P., Kalhan, S.C., Kerr, D.S., Danish, R.K., 1989. Urea synthesis, nitrogen bal-ance, and glucose turnover in growth-hormone-deficient children before and after growth hormoneadministration. Metabolism 38, 197–203.

Darcy-Vrillon, B., Posho, L., Morel, M.-T., Bernard, F., Blachier, F., Meslin, J.-C., Duee, P.-H., 1994.Glucose, galactose, and glutamine metabolism in pig isolated enterocytes during development. Pediat.Res. 36, 175–181.


Davis, T.A., Nguyen, H.V., Garcia-Bravo, R., Fiorotto, M.L., Jackson, E.M., Lewis, D.S., Lee, D.R.,Reeds, P.J., 1994. Amino acid composition of human milk is not unique. J. Nutr. 124, 1126–1132.

Davis, T.A., Burrin, D.G., Fiorotto, M.L., Nguyen, H.V., 1996. Protein synthesis in skeletal muscleand jejunum is more responsive to feeding in 7- than in 26-day-old pigs. Amer. J. Physiol. 270,E802–E809.

Davis, T.A., Fiorotto, M.L., Beckett, P.R., Burrin, D.G., Reeds, P.J., Wray-Cahen, D., Nguyen, H.V.,2001. Differential effects of insulin on peripheral and visceral tissue protein synthesis in neonatalpigs. Amer. J. Physiol. 280, E770–E779.

Davis, T.A., Fiorotto, M.L., Burrin, D.G., Reeds, P.J., Nguyen, H.V., Beckett, P.R., Vann, R.C.,O’Connor, P.M., 2002a. Stimulation of protein synthesis by both insulin and amino acids is unique toskeletal muscle in neonatal pigs. Amer. J. Physiol. 282, E880–E890.

Davis, T.A., Fiorotto, M.L., Burrin, D.G., Vann, R.C., Reeds, P.J., Nguyen, H.V., Beckett, P.R., Bush, J.A.,2002b. Acute IGF-I infusion stimulates protein synthesis in skeletal muscle and other tissues ofneonatal pigs. Amer. J. Physiol. 283, E638–E647.

de Blaauw, I., Deutz, N.E., von Meyenfeldt, M.F., 1996. In vivo amino acid metabolism of gut and liverduring short and prolonged starvation. Amer. J. Physiol. 270, G298–G306.

de Jonge, W.J., Kwikkers, K.L., te Velde, A.A., van Deventer, S.J., Nolte, M.A., Mebius, R.E., Ruijter, J.M.,Lamers, M.C., Lamers, W.H., 2002. Arginine deficiency affects early B cell maturation and lymphoidorgan development in transgenic mice. J. Clin. Invest. 110, 1539–1548.

Del Roso, A., Vittorini, S., Cavallini, G., Donati, A., Gori, Z., Masini, M., Pollera, M., Bergamini, E.,2003. Ageing-related changes in the in vivo function of rat liver macroautophagy and proteolysis.Exp. Gerontol. 38, 519–527.

Dennis, P.B., Jaeschke, A., Saitoh, M., Fowler, B., Kozma, S.C., Thomas, G., 2001. Mammalian TOR:a homeostatic ATP sensor. Science 294, 1102–1105.

Deplancke, B., Gaskins, H.R., 2002. Redox control of the transsulfuration and glutathione biosynthesispathways. Curr. Opin. Clin. Nutr. Metab. Care 5, 85–92.

Doering, C.B., Danner, D.J., 2000. Amino acid deprivation induces translation of branched-chain α-ketoacid dehydrogenase kinase. Amer. J. Physiol. 279, C1587–C1594.

Douglas, R.G., Gluckman, P.D., Ball, K., Breier, B., Shaw, J.H., 1991. The effects of infusion of insulin-like growth factor (IGF) I, IGF-II, and insulin on glucose and protein metabolism in fasted lambs.J. Clin. Invest. 88, 614–622.

Drucker, D.J., 2002. Biological actions and therapeutic potential of the glucagon-like peptides.Gastroenterology 122, 531–544.

Dudley, M.A., Burrin, D.G., Wykes, L.J., Toffolo, G., Cobelli, C., Nichols, B.L., Rosenberger, J., Jahoor, F.,Reeds, P.J., 1998. Protein kinetics determined in vivo with a multiple-tracer, single-sample protocol:application to lactase synthesis. Amer. J. Physiol. 274, G591–G598.

Dugan, M.E.R., Knabe, D.A., Wu, G., 1995. The induction of citrulline synthesis from glutamine inenterocytes of weaned pigs is not due primarily to age or change in diet. J. Nutr. 125, 2388–2393.

Ebner, S., Schoknecht, P., Reeds, P.J., Burrin, D.G., 1994. Growth and metabolism of gastrointestinaland skeletal muscle tissues in protein malnourished neonatal pigs. Amer. J. Physiol. 266,R1736–R1743.

Elango, R., Pink, D., Pencharz, P.B., Ball, R.O., 2003. Branched chain amino acid catabolism in theneonatal piglet gut. In: Ball, R.O. (Ed.), Proceedings of the 9th International Symposium on DigestivePhysiology in Pigs. University of Alberta, Department of Agriculture, Food, and Nutritional Science,Edmonton, Alberta, Vol. 2, p. 207–209.

Elwyn, D.H., Parikh, H.C., Shoemaker, W.C., 1968. Amino acid movements between gut, liver, andperiphery in unanesthetized dogs. Amer. J. Physiol. 215, 1260–1275.

Etherton, T.D., Bauman, D.E., 1998. Biology of somatotropin in growth and lactation of domestic animals.Physiol. Rev. 78, 745–761.

Faure, M., Moennoz, D., Montigon, F., Fay, L.B., Breuille, D., Finot, P.A., Ballevre, O., Boza, J., 2002.Development of a rapid and convenient method to purify mucins and determine their in vivo synthe-sis rate in rats. Anal. Biochem. 307, 244–251.

Feldhoff, R.C., Taylor, J.M., Jefferson, L.S., 1977. Synthesis and secretion of rat albumin in vivo, in per-fused liver, and in isolated hepatocytes. J. Biol. Chem. 252, 3611–3616.

Felig, P., 1975. Amino acid metabolism in man. Annu. Rev. Biochem. 44, 933.Finkelstein, J.D., 2000. Pathways and regulation of homocysteine metabolism in mammals. Semin.

Thromb. Hemost. 26, 219–225.


Fouillet, H., Gaudichon, C., Bos, C. Mariotti, F., Tome, D., 2003. Contribution of plasma proteins tosplanchnic and total anabolic utilization of dietary nitrogen in humans. Amer. J. Physiol. EndocrinolMetab. 285, E88–E97.

Fuller, M.F., Reeds, P.J., 1998. Nitrogen cycling in the gut. Annu. Rev. Nutr. 18, 385–411.Gahl, M.J., Benevenga, N.J., Crenshaw, T.D., 1998. Rates of lysine catabolism are inversely related to

rates of protein synthesis when measured concurrently in adult female rats induced to grow at differ-ent rates. J. Nutr. 128, 1503–1511.

Garcia-Arumi, E., Schwartz, S., Lopez-Hellin, J., Arbos, M.A., Andreu, A.L., Farriol, M., 1995. Additionof glutamine does not improve protein synthesis and jejunal mucosa morphology in non-hypercatabolicstress. Physiol. Res. 44, 233–239.

Garlick, P.J., McNurlan, M.A., Preedy, V.R., 1980. A rapid and convenient technique for measuring therate of protein synthesis in tissues by injection of 3H-phenylalanine. Biochem. J. 192, 719–723.

Garlick, P.J., McNurlan, M.A., Essen, P., Wernerman, J., 1994. Measurement of tissue protein synthesisrates in vivo: a critical analysis of contrasting methods. Amer. J. Physiol. 266, E287–E297.

Gaskins, H.R., Collier, C.T., Anderson, D.B., 2002. Antibiotics as growth promotants: mode of action.Anim. Biotechnol. 13, 29–42.

Goldspink, D.F., Kelly, F.J., 1984. Protein turnover and growth in the whole body, liver and kidney of therat from the foetus to senility. Biochem. J. 217, 507–516.

Goldspink, D.F., Lewis, S.E.M., Kelly, F.J., 1984. Protein synthesis during the developmental growth ofthe small and large intestine of the rat. Biochem. J. 217, 527–534.

Grimble, R.F., 2001. Nutritional modulation of immune function. Proc. Nutr. Soc. 60, 389.Grofte, T., Wolthers, T., Jensen, S.A., Moller, N., Jorgensen, J.O., Tygstrup, N., Orskov, H., Vilstrup, H.,

1997. Effects of growth hormone and insulin-like growth factor-I singly and in combination on in vivocapacity of urea synthesis, gene expression of urea cycle enzymes, and organ nitrogen contents in rats.Hepatology 25, 964–969.

Guan, X., Stoll, B., Lu, X., Tappenden, K.A., Holst, J.J., Hartmann, B., Burrin, D.G., 2003. GLP-2-mediatedup-regulation of intestinal blood flow and glucose uptake is NO-dependent in TPN-fed piglets.Gastroenterology 125, 136–147.

Hahn, P., Taller, M., Chan, H., 1988. Pyruvate carboxylase, phosphate-dependent glutaminase and gluta-mate dehydrogenase in the developing rat small intestinal mucosa. Biol. Neonate 53, 362–366.

Haisch, M., Fukagawa, N.K., Matthews, D.E., 2000. Oxidation of glutamine by the splanchinic bed inhumans. Amer. J. Physiol. Endocrinol. Metab. 278, E593.

Hallemeesch, M.M., ten Have, G.A., Deutz, N.E., 2001. Metabolic flux measurements across portaldrained viscera, liver, kidney and hindquarter in mice. Lab. Anim. 35, 101–110.

Hallemeesch, M.M., Lamers, W.H., Deutz, N.E., 2002. Reduced arginine availability and nitric oxideproduction. Clin. Nutr. 21, 273–279.

Halseth, A.E., Flakoll, P.J., Reed, E.K., Messina, A.B., Krishna, M.G., Lacy, D.B., Williams, P.E.,Wasserman, D.H., 1997. Effect of physical activity and fasting on gut and liver proteolysis in the dog.Amer. J. Physiol. 273, E1073–E1082.

Hamel, F.G., Upward, J.L., Siford, G.L., Duckworth, W.C., 2003. Inhibition of proteasome activity byselected amino acids. Metabolism 52, 810–814.

Haque, S.M., Chen, K., Usui, N., Iiboshi, Y., Okuyama, H., Masunari, A., Cui, L., Nezu, R., Takagi, Y.,Okada, A., 1996. Alanyl-glutamine dipeptide-supplemented parenteral nutrition improves intestinalmetabolism and prevents increased permeability in rats. Ann. Surg. 223, 334–341.

Harp, J.B., Goldstein, S., Phillips, L.S., 1991. Nutrition and somatomedin. XXIII. Molecular regulationof IGF-I by amino acid availability in cultured hepatocytes. Diabetes 40, 95–101.

Haussinger, D., 1990. Nitrogen metabolism in liver: structural and functional organization and physio-logical relevance. Biochem. J. 267, 281–290.

Haussinger, D., 1996. The role of cellular hydration in the regulation of cell function. Biochem. J. 313,697–710.

Haussinger, D., Schliess, F., 1999. Osmotic induction of signaling cascades: role in regulation of cellfunction. Biochem. Biophys. Res. Commun. 255, 551–555.

Haussinger, D., Graf, D., Weiergraber, O.H., 2001. Glutamine and cell signaling in liver. J. Nutr. 131, 2509S.Heitmann, R.N., Bergman, E.N., 1978. Glutamine metabolism, interorgan transport and glucogenicity in

the sheep. Amer. J. Physiol. 234, E197–E203.Hellstern, G., Kaempf-Rotzoll, D.D., Linderkamp, O.O., Langhans, K.D., Rating, D., 2002. Parenteral

amino acids increase albumin and skeletal muscle protein fractional synthetic rates in prematurenewborn minipigs. J. Pediat. Gastroenterol. Nutr. 35, 270–274.


Higashiguchi, T., Hasselgren, P.O., Wagner, K., Fischer, J.E., 1993. Effect of glutamine on protein syn-thesis in isolated intestinal epithelial cells. J. Parent. Enter. Nutr. 17, 307–314.

Higashiguchi, T., Noguchi, Y., O’Brien, W., Wagner, K., Fischer, J.E., Hasselgren, P.O., 1994. Effect of sepsison mucosal protein synthesis in different parts of the gastrointestinal tract in rats. Clin. Sci. 87, 207.

Hodgkinson, S.M., Moughan, P.J., Reynolds, G.W., James, K.A., 2000. The effect of dietary peptideconcentration on endogenous ileal amino acid loss in the growing pig. Brit. J. Nutr. 83, 421–430.

Hoerr, R.A., Matthews, D.E., Bier, D.M., Young, V.R., 1991. Leucine kinetics from 2H3- and 13C-leucineinfused simultaneously by gut and vein. Amer. J. Physiol. 260, E111–E117.

Hooper, L.V., Midtvedt, T., Gordon. J.I., 2002. How host-microbial interactions shape the nutrient envi-ronment of the mammalian intestine. Annu. Rev. Nutr. 22, 283–307.

Hopgood, M.F., Clark, M.G., Ballard, F.J., 1981. Stimulation by glucocorticoids of protein degradationin hepatocyte monolayers. Biochem. J. 196, 33–40.

House, J.D., Hall, B.N., Brosnan, J.T., 2001. Threonine metabolism in isolated rat hepatocytes. Amer.J. Physiol. 281, E1300–E1307.

Huang, Y.S., Chuang, D.T., 1999. Down-regulation of rat mitochondrial branched-chain 2-oxoaciddehydrogenase kinase gene expression by glucocorticoids. Biochem. J. 339, 503–510.

Huang, Y., Shao, X.M., Neu, J., 2003. Immunonutrients and neonates. Eur. J. Pediat. 162, 122–128.Humbert, B., Le Bacquer, O., Nguyen, P., Dumon, H., Darmaun, D., 2001. Protein restriction and

dexamethasone as a model of protein hypercatabolism in dogs: effect of glutamine on leucineturnover. Metabolism 50, 293–298.

Hutson, S.M., Stinson-Fisher, C., Shiman, R., Jefferson, L.S., 1987. Regulation of albumin synthesis byhormones and amino acids in primary cultures of rat hepatocytes. Amer. J. Physiol. 252, E291–E298.

Iannoli, P., Miller, J.H., Ryan, C.K., Sax, H.C., 1998. Glucocorticoids upregulate intestinal nutrient trans-port in a time-dependent and substrate-specific fashion. J. Gastrointest. Surg. 2, 449–457.

Inoue, Y., Grant, J.P., Snyder, P.J., 1993. Effect of glutamine-supplemented total parenteral nutrition onrecovery of the small intestine after starvation atrophy. J. Parent. Enter. Nutr. 17, 165–170.

Jacobs, R.L., Stead, L.M., Brosnan, M.E., Brosnan, J.T., 2001. Hyperglucagonemia in rats results indecreased plasma homocysteine and increased flux through the transsulfuration pathway in liver.J. Biol. Chem. 276, 43740–43747.

Jahoor, F., Wykes, L.J., Reeds, P.J., Henry, J.F., del Rosario, M.P., Frazer, M.E., 1995. Protein-deficientpigs cannot maintain reduced glutathione homeostasis when subjected to the stress of inflammation.J. Nutr. 125, 1462–1472.

Jefferson, L.S., Kimball, S.R., 2003. Amino acids as regulators of gene expression at the level of mRNAtranslation. J. Nutr. 133, 2046S–2051S.

Jefferson, L.S., Schworer, C.M., Tolman, E.L., 1975. Growth hormone stimulation of amino acid trans-port and utilization by the perfused rat liver. J. Biol. Chem. 250, 197–204.

Jepson, M.M., Pell, J.M., Bates, P.C., Millward, D.J., 1986. The effects of endotoxaemia on proteinmetabolism in skeletal muscle and liver of fed and fasted rats. Biochem. J. 235, 329–336.

Jiang, R., Chang, X., Stoll, B., Ellis, K.J., Shypailo, R.J., Weaver, E., Campbell, J., Burrin, D.G., 2000.Dietary plasma protein is used more efficiently than extruded soy protein for lean tissue growth inearly-weaned pigs. J. Nutr. 130, 2016–2019.

Johnson, R.W., 1997. Inhibition of growth by pro-inflammatory cytokines: an integrated view. J. Anim.Sci. 75, 1244–1255.

Johnson, R.W., 2002. The concept of sickness behavior: a brief chronological account of four keydiscoveries. Vet. Immunol. Immunopathol. 87, 443–450.

Jonas, C.R., Ziegler, T.R., Gu, L.H., Jones, D.P., 2002. Extracellular thiol/disulfide redox state affects pro-liferation rate in a human colon carcinoma (Caco2) cell line. Free Radical Biol. Med. 33, 1499–1506.

Jungas, R.L., Halperin, M.L., Brosnan, J.T., 1992. Quantitative analysis of amino acid oxidation andrelated gluconeogenesis in humans. Physiol. Rev. 72, 419.

Kadowaki, M., Kanazawa, T., 2003. Amino acids as regulators of proteolysis. J. Nutr. 133, 2052S–2056S.Kandil, H.M., Argenzio, R.A., Chen, W., Berschneider, H.M., Stiles, A.D., Westwick, J.K., Rippe, R.A.,

Brenner, D.A., Rhoads, J.M., 1995. L-Glutamine and L-asparagine stimulate ODC activity and prolif-eration in a porcine jejunal enterocyte line. Amer. J. Physiol. 269, G591–G599.

Karinch, A.M., Pan, M., Lin, C.M., Strange, R., Souba, W.W., 2001. Glutamine metabolism in sepsis andinfection. J. Nutr. 131, 2535S–2538S.

Khan, J., Iiboshi, Y., Cui, L., Wasa, M., Sando, K., Takagi, Y., Okada, A., 1999. Alanyl-glutamine-supplemented parenteral nutrition increases luminal mucus gel and decreases permeability in the ratsmall intestine. J. Parent. Enter. Nutr. 23, 24–31.


Kight, C.E., Fleming, S.E., 1995. Oxidation of glucose carbon entering the TCA cycle is reduced by glu-tamine in small intestine epithelial cells. Amer. J. Physiol. 268, G879–G888.

Kimball, S.R., Jefferson, L.S., 1994, Mechanisms of translational control in liver and skeletal muscle.Biochimie 76, 729–736.

Kimball, S.R., Jefferson, L.S., 2001. Regulation of protein synthesis by branched-chain amino acids.Curr. Opin. Clin. Nutr. Metab. Care 4, 39–43.

Kimball, S.R., Horetsky, R.L., Jefferson, L.S., 1995. Hormonal regulation of albumin gene expression inprimary cultures of rat hepatocytes. Amer. J. Physiol. 268, E6–E14.

Kimura, N., Tokunaga, C., Dalal, S., Richardson, C., Yoshino, K., Hara, K., Kemp, B.E., Witters, L.A.,Mimura, O., Yonezawa, K., 2003. A possible linkage between AMP-activated protein kinase (AMPK)and mammalian target of rapamycin (mTOR) signalling pathway. Genes Cells 8, 65–79.

Ko, Y.G., Kim, E.Y., Kim, T., Park, H., Park, H.S., Choi, E.J., Kim, S., 2001. Glutamine-dependent anti-apoptotic interaction of human glutaminyl-tRNA synthetase with apoptosis signal-regulating kinase 1.J. Biol. Chem. 276, 6030–6036.

Koea, J.B., Douglas, R.G., Breier, B.H., Shaw, J.H., Gluckman, P.D., 1992. Synergistic effect of insulin-like growth factor-I administration on the protein-sparing effects of total parenteral nutrition in fastedlambs. Endocrinology 131, 643–648.

Krause, U., Bertrand, L., Maisin, L., Rosa, M., Hue, L., 2002. Signalling pathways and combinatoryeffects of insulin and amino acids in isolated rat hepatocytes. Eur. J. Biochem. 269, 3742–3750.

Krinke, A.L., Jamroz, D., 1996. Signalling pathways and combinatory effects of insulin and amino acidsin isolated rat hepatocytes. Poultry Sci. 75, 705–710.

Lambert, B.D., Titgemeyer, E.C., Stokka, G.L., DeBey, B.M., Loest, C.A., 2002. Methionine supply togrowing steers affects hepatic activities of methionine synthase and betaine-homocysteine methyl-transferase, but not cystathionine synthase. J. Nutr. 132, 2004–2009.

Lapierre, H., Bernier, J.F., Dubreuil, P., Reynolds, C.K., Farmer, C., Ouellet, D.R., Lobley, G.E., 1999.The effect of intake on protein metabolism across splanchnic tissues in growing beef steers. Brit. J. Nutr.81, 457–466.

Lash, L.H., Hagen, T.M., Jones, D.P., 1986. Exogenous glutathione protects intestinal epithelial cellsfrom oxidative injury. Proc. Natl. Acad. Sci. USA 83, 4641–4645.

Le Bacquer, O., Nazih, H., Blottiere, H., Meynial-Denis, D., Laboisse, C., Darmaun, D., 2001. Effects ofglutamine deprivation on protein synthesis in a model of human enterocytes in culture. Amer.J. Physiol. 281, G1340–G1347.

Le Bacquer, O., Laboisse, C., Darmaun, D., 2003. Glutamine preserves protein synthesis and paracellu-lar permeability in Caco-2 cells submitted to “luminal fasting”. Amer. J. Physiol. 285, G128–G136.

Le Floc’h, N.L., Obled, C., Seve, B., 1996. In vivo threonine oxidation in growing pigs fed on diets withgraded levels of threonine. Brit. J. Nutr. 75, 825–837.

Le Floc’h, N., Thibault, J.N., Seve, B., 1997. Tissue localization of threonine oxidation in pigs. Brit.J. Nutr. 77, 593–603.

Ling, P.R., Gollaher, C., Colon, E., Istfan, N., Bistrian, B.R., 1995. IGF-I alters energy expenditure andprotein metabolism during parenteral feeding in rats. Amer. J. Clin. Nutr. 61, 116–120.

Lo, H.C., Ney, D.M., 1996. GH and IGF-I differentially increase protein synthesis in skeletal muscle andjejunum of parenterally fed rats. Amer. J. Physiol. 271, E872–E878.

Lobley, G.E., Milne, V., Lovie, J.M., Reeds, P.J., Pennie, K., 1980. Whole body and tissue proteinsynthesis in cattle. Brit. J. Nutr. 43, 491–502.

Lobley, G.E., Harris, P.M., Skene, P.A., Brown, D., Milne, E., Calder, A.G., Anderson, S.E., Garlick, P.J.,Nevison, I., Connell, A., 1992. Responses in tissue protein synthesis to sub- and supra-maintenanceintake in young growing sheep: comparison of large-dose and continuous-infusion techniques. Brit.J. Nutr. 68, 373–388.

Lobley, G.E., Connell, A., Revell, D.K., Bequette, B.J., Brown, D.S., Calder, A.G., 1996a. Splanchnic-bed transfers of amino acids in sheep blood and plasma, as monitored through use of a multipleU-13C-labelled amino acid mixture. Brit. J. Nutr. 75, 217–235.

Lobley, G.E., Connell, A., Revell, D., 1996b. The importance of transmethylation reactions to methionine metabolism in sheep: effects of supplementation with creatine and choline. Brit. J. Nutr.75, 47–56.

Lobley, G.E., Shen, X., Le, G., Bremner, D.M., Milne, E., Calder, A.G., Anderson, S.E., Dennison, N.,2003. Oxidation of essential amino acids by the ovine gastrointestinal tract. Brit. J. Nutr. 89, 617–630.


Luiking, Y.C., Deutz, N.E., 2003. Isotopic investigation of nitric oxide metabolism in disease. Curr. Opin.Clin. Nutr. Metab. Care 6, 103–108.

Lynch, C.J., 2001. Role of leucine in the regulation of mTOR by amino acids: revelations from structure-activity studies. J. Nutr. 131, 861S–865S.

Lynch, C.J., Hutson, S.M., Patson, B.J., Vaval, A., Vary, T.C., 2002. Tissue-specific effects of chronic dietaryleucine and norleucine supplementation on protein synthesis in rats. Amer. J. Physiol. 283, E824–E835.

Mack, D.R., Michail, S., Wei, S., McDougall, L., Hollingsworth, M.A., 1999. Probiotics inhibitenteropathogenic E. coli adherence in vitro by inducing intestinal mucin gene expression. Amer.J. Physiol. 39, G941–G950.

MacLean, K.N., Janosik, M., Kraus, E., Kozich, V., Allen, R.H., Raab, B.K., 2002. Cystathionineβ-synthase is coordinately regulated with proliferation through a redox-sensitive mechanism incultured human cells and Saccharomyces cerevisiae. J. Cell. Physiol. 192, 81–92.

MacRae, J.C., 1993. Metabolic consequences of intestinal parasitism. Proc. Nutr. Soc. 52, 121–130.Madej, M., Lundh, T., Lindberg, J.E., 1999. Activities of enzymes involved in glutamine metabolism in

connection with energy production in the gastrointestinal tract epithelium of newborn, suckling andweaned piglets. Biol. Neonate 75, 250–258.

Malmezat, T., Breuille, D., Pouyet, C., Mirand, P.P., Obled, C., 1998. Metabolism of cysteine is modifiedduring the acute phase of sepsis in rats. J. Nutr. 128, 97–105.

Malmezat, T., Breuille, D., Capitan, P., Mirand, P.P., Obled, C., 2000a. Glutathione turnover is increasedduring the acute phase of sepsis in rats. J. Nutr. 130, 1239–1246.

Malmezat, T., Breuille, D., Pouyet, C., Buffiere, C., Denis, P., Mirand, P.P., Obled, C., 2000b. Methioninetranssulfuration is increased during sepsis in rats. Amer. J. Physiol. 279, E1391–E1397.

Marchini, J.S., Nguyen, P., Deschamps, J.-Y., Maugere, P., Krempf, M., Darmaun, D., 1999. Effectof intratvenous glutamine on duodenal mucosa protein synthesis in healthy growing dogs. Amer.J. Physiol. 276, E747–E753.

Martensson, J., Jain, A., Meister, A., 1991. Glutathione is required for intestinal function. Proc. Natl.Acad. Sci. USA, 87, 1715–1719.

Matthews, D.E., Marano, M.A., Campbell, R.G., 1993. Splanchnic bed utilization of leucine and pheny-lalanine in humans. Amer. J. Physiol. 264, E109–E118.

McCowen, K.C., Bistrian, B.R., 2003. Immunonutrition: problematic or problem solving? Amer. J. Clin.Nutr. 77, 764–770.

McDaniel, M.L., Marshall, C.A., Pappan, K.L., Kwon, G., 2002. Metabolic and autocrine regulation ofthe mammalian target of rapamycin by pancreatic beta-cells. Diabetes 51, 2877–2885.

McNurlan, M.A., Garlick, P.J., 1981. Protein synthesis in liver and small intestine in protein deprivationand diabetes. Amer. J. Physiol. 241, E238–E245.

McNurlan, M.A., Tomkins, A.M., Garlick, P.J., 1979. The effect of starvation on the rate of protein-synthesis in rat liver and small intestine. Biochem. J. 178, 373–379.

Meek, S.E., Persson, M., Ford, G.C., Nair, K.A., 1998. Differential regulation of amino acid exchangeand protein dynamics across splanchnic and skeletal muscle beds by insulin in healthy humansubjects. Diabetes 47, 1824–1835.

Meijer, A.J., 2003. Amino acids as regulators and components of nonproteinogenic pathways. J. Nutr.133, 2057S–2062S.

Mercier, S., Breuille, D., Mosoni, L., Obled, C., Patureau Mirand, P., 2002. Chronic inflammation altersprotein metabolism in several organs of adult rats. J. Nutr. 132, 1921–1928.

Merry, B.J., Lewis, S.E., Goldspink, D.F., 1992. The influence of age and chronic restricted feeding onprotein synthesis in the small intestine of the rat. Exp. Gerontol. 27, 191–200.

Metges, C.C., 2000. Contribution of microbial amino acids to amino acid homeostasis of the host. J. Nutr.130, 1857S–1864S.

Miller, L.T., Watson, W.H., Kirlin, W.G., Ziegler, T.R., Jones, D.P., 2002. Oxidation of the glutathionedisulfide redox state is induced by cysteine deficiency in human colon carcinoma HT29 cells. J. Nutr.132, 2303–2306.

Morris, S.M., 2002. Regulation of enzymes of the urea cycle and arginine metabolism. Annu. Rev. Nutr.22, 87–105.

Mortimore, G.E., Poso, A.R., 1987. Intracellular protein catabolism and its control during nutrientdeprivation and supply. Annu. Rev. Nutr. 7, 539–564.

Mortimore, G.E., Poso, A.R., Lardeux, B.R., 1989. Mechanism and regulation of protein degradation inliver. Diabetes Metab. Rev. 5, 49–70.


Mosharov, E., Cranford, M.R., Banerjee, R., 2000. The quantitatively important relationship betweenhomocysteine metabolism and glutathione synthesis by the transsulfuration pathway and its regula-tion by redox changes. Biochemistry 39, 13005–13011.

Mosoni, L., Houlier, M.L., Mirand, P.P., Bayle, G., Grizard, J., 1993. Effect of amino acids alone or withinsulin on muscle and liver protein synthesis in adult and old rats. Amer. J. Physiol. 264, E614–E620.

Muramatsu, T., Coates, M.E., Hewitt, D., Salter, D.N., Garlick, P.J., 1983. The influence of the gutmicroflora on protein synthesis in liver and jejunal mucosa in chicks. Brit. J. Nutr. 49, 453–462.

Murphy, J.M., Murch, S.J., Ball, R.O., 1996. Proline is synthesized from glutamate during intragastricinfusion but not during intravenous infusion in neonatal piglets. J. Nutr. 26, 878–886.

Nair, K.S., Ford, G.C., Ekberg, K., Fernqvist-Forbes, E., Wahren, J., 1995. Protein dynamics in wholebody and in splanchnic and leg tissues in type I diabetic patients. J. Clin. Invest. 95, 2926–2937.

Naka, S., Saito, H., Hashiguchi, Y., Lin, M.T., Furukawa, S., Inaba, T., Fukushima, R., Wada, N., Muto, T.,1997. Alanylglutamine-enriched total parenteral nutrition improves protein metabolism more thanbranched chain amino acid-enriched total parenteral nutrition in protracted peritonitis. J. Trauma 42,183–190.

Newsholme, P., Procopio, J., Lima, M.M., Pithon-Curi, T.C., Curi, R., 2003. Glutamine and glutamate:their central role in cell metabolism and function. Cell Biochem. Funct. 21, 1–9.

Nieto, R., Lobley, G.E., 1999. Integration of protein metabolism within the whole body and betweenorgans. In: Lobley, G.E., White, A., MacRae, J.C. (Eds.), Proceedings of the 8th InternationalSymposium on Protein Metabolism and Nutrition. EAAP Publication No. 96, p. 69.

Nissim, I., Brosnan, M.E., Yudkoff, M., Brosnan, J.T., 1999. Studies of hepatic glutamine metabolism inthe perfused rat liver with 15N-labeled glutamine. J. Biol. Chem. 274, 28958–28965.

Nkabyo, Y.S., Ziegler, T.R., Li, H.G., Watson, W.H., Jones, D.P., 2002. Glutathione and thioredoxin redoxduring differentiation in human colon epithelial (Caco-2) cells. Amer. J. Physiol. 283, G1352–G1359.

Noda, T., Iwakiri, R., Fujimoto, K., Aw, T.Y., 2001. Induction of mild intracellular redox imbalanceinhibits proliferation of CaCo-2 cells. FASEB J. 15, 2131–2139.

Noda, T., Iwakiri, R., Fujimoto, K., Rhoads, C.A., Aw, T.Y., 2002. Exogenous cysteine and cystine pro-mote cell proliferation in CaCo-2 cells. Cell Prolif. 35, 117–129.

Nolan, E.M., Masters, J.N., Dunn, A., 1990. Growth hormone regulation of hepatic glutamine synthetasemRNA levels in rats. Mol. Cell Endocrinol. 69, 101–110.

Nyachoti, C.M., de Lange, C.F., McBride, B.W., Leeson, S., Gabert, V.M., 2000. Endogenous gut nitro-gen losses in growing pigs are not caused by increased protein synthesis rates in the small intestine.J. Nutr. 130, 566–572.

Nygren, J., Nair, K.S., 2003. Differential regulation of protein dynamics in splanchnic and skeletalmuscle beds by insulin and amino acids in healthy human subjects. Diabetes 52, 1377–1385.

Odedra, B.R., Bates, P.C., Millward, D.J., 1983. Time course of the effect of catabolic doses of corticos-terone on protein turnover in rat skeletal muscle and liver. Biochem. J. 214, 617–627.

Ogier-Denis, E., Codogno, P., 2003. Autophagy: a barrier or an adaptive response to cancer. Biochim.Biophys. Acta 1603, 113–128.

O’Leary, M.J., Koll, M., Ferguson, C.N., Coakley, J.H., Hinds, C.J., Preedy, V.R., Garlick, P.J., 2003.Liver albumin synthesis in sepsis in the rat: influence of parenteral nutrition, glutamine and growthhormone. Clin. Sci. (Lond.) 105, 691–698.

O’Sullivan, D., Brosnan, J.T., Brosnan, M.E., 2000. Catabolism of arginine and ornithine in the perfusedrat liver: effect of dietary protein and of glucagon. Amer. J. Physiol. 278, E516–E521.

Palmer, R.M., Pusztai, A., Bain, P., Grant, G., 1987. Changes in rates of tissue protein synthesis in ratsinduced in vivo by consumption of kidney bean lectins. Comp. Biochem. Physiol. 88C, 179.

Papaconstantinou, H.T., Hwang, K.O., Rajaraman, S., Hellmich, M.R., Townsend, C.M. Jr., Ko, T.C.,1998. Glutamine deprivation induces apoptosis in intestinal epithelial cells. Surgery 124,152–160.

Pell, J.M., Bates, P.C., 1992. Differential actions of growth hormone and insulin-like growth factor-I ontissue protein metabolism in dwarf mice. Endocrinology 130, 1942–1950.

Pias, E.K., Aw, T.Y., 2002. Apoptosis in mitotic competent undifferentiated cells is induced by cellularredox imbalance independent of reactive oxygen species production. FASEB J. 16, 781–790.

Pink, D., Elango, R., Dixon, W.T., Ball, R.O., 2003. Intestinal catabolism of lysine in swine.In: Ball, R.O. (Ed.), Proceedings of the 9th International Symposium on Digestive Physiology in Pigs.University of Alberta, Department of Agriculture, Food and Nutritional Science, Edmonton, Alberta,Vol. 2, pp. 131–133.


Platell, C., McCauley, R., McCulloch, R., Hall, J., 1993. The influence of parenteral glutamine andbranched-chain amino acids on total parenteral nutrition-induced atrophy of the gut. J. Parent. Enter.Nutr. 17, 348–354.

Pollack, P.F., Koldovsky, O., Nishioka, K., 1992. The influence of parenteral glutamine and branched-chainamino acids on total parenteral nutrition-induced atrophy of the gut. Amer. J. Clin. Nutr. 56, 371–375.

Ponter, A.A., Cortamira, N.O., Seve, B., Salter, D.N., Morgan, L.M., 1994. The effects of energy sourceand tryptophan on the rate of protein synthesis and on hormones of the entero-insular axis in thepiglet. Brit. J. Nutr. 71, 661–674.

Preedy, V.R., Paska, L., Sugden, P.H., Schofield, P.S., Sugden, M.C., 1988. The effects of surgical stressand short-term fasting on protein synthesis in vivo in diverse tissues of the mature rat. Biochem.J. 250, 179–188.

Ratnam, S., Maclean, K.N., Jacobs, R.L., Brosnan, M.E., Kraus, J.P., Brosnan, J.T., 2002. Hormonalregulation of cystathionine β-synthase expression in liver. J. Biol. Chem. 277, 42912–42918.

Reeds, P.J., Fjeld, C.R., Jahoor, F., 1994. Do the differences between the amino acid compositionsof acute-phase and muscle proteins have a bearing on nitrogen loss in traumatic states? J. Nutr.124, 906–910.

Reeds, P.J., Burrin, D.G., Stoll, B., Jahoor, F., Wykes, L., Henry, J., Frazer, M.E., 1997. Enteral glutamateis the preferential source for mucosal glutathione synthesis in fed piglets. Amer. J. Physiol.273, E408–E415.

Reeds, P.J., Burrin, D.G., Stoll, B., van Goudoever, J.B., 1999. Consequences and regulation of gutmetabolism. In: Lobley, G.E., White, A., MacRae, J.C. (Eds.), Protein Metabolism and Nutrition.EAAP Publication No. 96, p. 127.

Reid, M., Badaloo, A., Forrester, T., Morlese, J.F., Frazer, M., Heird, W.C., Jahoor, F., 2000. In vivo ratesof erythrocyte glutathione synthesis in children with severe protein-energy malnutrition. Amer.J. Physiol. 278, E405–E412.

Rerat, A., Simoes-Nunes, C., Mendy, F., Vaissade, P., Vaugelade, P., 1992. Splanchnic fluxes of aminoacids after duodenal infusion of carbohydrate solutions containing free amino acids or oligopeptidesin the non-anaesthetized pig. Brit. J. Nutr. 68, 111–138.

Rhoads, M., 1999. Glutamine signaling in intestinal cells. J. Parent. Enter. Nutr. 23, S38–S40.Saenz de Pipaon, M., VanBeek, R.H., Quero, J., Perez, J., Wattimena, D.J., Sauer, P.J., 2003. Effect of

minimal enteral feeding on splanchnic uptake of leucine in the postabsorptive state in preterm infants.Pediat. Res. 53, 281–287.

Salleh, M., Ardawi, M., Majzoub, M.F., Newsholme, E.A., 1988. Effect of glucocorticoid treatmenton glucose and glutamine metabolism by the small intestine of the rat. Clin. Sci. (Lond.) 75,93–100.

Samuels, S.E., Taillandier, D., Aurousseau, E., Cherel, Y., Maho, L., Arnal, M., Attaix, D., 1996.Gastrointestinal tract protein synthesis and mRNA levels for proteolytic systems in adult fasted rats.Amer. J. Physiol. 271, E232–E238.

Santangelo, F., 2002. The regulation of sulphurated amino acid junctions: fact or fiction in the field ofinflammation? Amino Acids 23, 359–365.

Schliess, F., Haussinger, D., 2002. The cellular hydration state: a critical determinant for cell death andsurvival. Biol. Chem. 383, 577–583.

Schliess, F., Haussinger, D., 2003. Cell volume and insulin signaling. Int. Rev. Cytol. 225, 187–228.Sen, C.K., 1998. Redox signaling and the emerging therapeutic potential of thiol antioxidants. Biochem.

Pharmacol. 55, 1747–1758.Seve, I., Reeds, P.J., Fuller, M.S., Cadenhead, A., Hay, S.M., 1986. Protein synthesis and retention in

some tissues of the young pig as influenced by dietary protein intake after early-weaning: possibleconnection to the energy metabiolism. Reprod. Nutr. Dev. 26, 849–861.

Shenoy, V., Roig, J.C., Chakrabarti, R., Kubilis, P., Neu, J., 1996. Ontogeny of glutamine synthetase inrat small intestine. Pediat. Res. 39, 643–648.

Shoveller, A.K., Brunton, J.A., Pencharz, P.B., Ball, R.O., 2003. The methionine requirement is lower inneonatal piglets fed parenterally than in those fed enterally. J. Nutr. 133, 1390–1397.

Souba, W.W., Smith, R.J., Wilmore, D.W., 1985. Effects of glucocorticoids on glutamine metabolism invisceral organs. Metabolism 34, 450–456.

Southon, S., Livesey, G., Gee, J.M., Johnson, I.T., 1985. Differences in intestinal protein synthesisand cellular proliferation in well-nourished rats consuming conventional laboratory diets. Brit. J. Nutr.53, 87–95.


Southorn, B.G., Palmer, R.M., Garlick, P.J., 1990. Acute effects of corticosterone on tissue proteinsynthesis and insulin-sensitivity in rats in vivo. Biochem. J. 272, 187–191.

Spaeth, G., Gottwald, T., Haas, W., Holmer, M., 1993. Glutamine peptide does not improve gutbarrier function and mucosal immunity in total parenteral nutrition. J. Parent. Enter. Nutr. 17,317–323.

Stegink, L.D., Den Besten, L., 1972. Synthesis of cysteine from methionine in normal adult subjects:effect of route of alimentation. Science 178, 514–516.

Stein, H.H., Nyachoti, M., 2003. Animal effects on ileal amino acid digestibility. In: Ball, R. (Ed.),Proceedings of the 9th International Symposium on Digestive Physiology in Pigs. Universityof Alberta, Department of Agriculture, Food, and Nutritional Science, Edmonton, Alberta. Vol. 2,p. 223.

Stein, T.P., Yoshida, S., Schluter, M.D., Drews, D., Assimon, S.A., Leskiw, M.D., 1994. Comparison ofintravenous nutrients on gut mucosal protein synthesis. J. Parent. Enter. Nutr. 18, 447–452.

Stipanuk, M.H., Rotter, M.A., 1984. Metabolism of cysteine, cysteinesulfinate and cysteinesulfonate inrats fed adequate and excess levels of sulfur-containing amino acids. J. Nutr. 114, 1426–1437.

Stoll, B., Henry, J., Reeds, P.J., Yu, H., Jahoor, F., Burrin, D.G., 1998. Catabolism dominates thefirst-pass intestinal metabolism of dietary essential amino acids in milk protein-fed piglets. J. Nutr.128, 606–614.

Stoll, B., Burrin, D.G., Henry, J., Yu, H., Jahoor, F., Reeds, P.J., 1999a. Substrate oxidation by the portaldrained viscera of fed piglets. Amer. J. Physiol. 277, E168–E175.

Stoll, B., Burrin, D.G., Henry, J., Jahoor, F., Reeds, P.J., 1999b. Dietary and systemic phenylalanineutilization for mucosal and hepatic constitutive protein synthesis in pigs. Amer. J. Physiol. 276,G49–G57.

Stoll, B., Chang, X., Fan, M.Z., Reeds, P.J., Burrin, D.G., 2000a. Enteral nutrient intake leveldetermines intestinal protein synthesis and accretion rates in neonatal pigs. Amer. J. Physiol. 279,G288–G294.

Stoll, B., Chang, X., Jiang, R., van Goudoever, J.B., Reeds, P.J., Burrin, D.G., 2000b. Enteral carbohy-drate and lipid inhibit small intestinal proteolysis in neonatal pigs. FASEB J. 14, A558.

Stoll, B., Price, P., Reeds, P.J., Henry, J., Burrin, D.G., 2000c. Feeding an elemental diet versus a milk-based formula does not decrease intestinal mucosal growth in infant pigs. J. Pediat. Gastroenterol.Nutr. 31, S169.

Stubbs, A.K., Wheelhouse, N.M., Lomax, M.A., Hazlerigg, D.G., 2002. Nutrient-hormone interaction inthe ovine liver: methionine supply selectively modulates growth hormone-induced IGF-I gene expres-sion. J. Endocrinol. 174, 335–341.

Tamada, H., Nezu, R., Imamura, I., Matsuo, Y., Takagi, Y., Kamata, S., Okada, A., 1992. The dipeptidealanyl-glutamine prevents intestinal mucosal atrophy in parenterally fed rats. J. Parent. Enter. Nutr.16, 110–116.

Tashiro, T., Sugiura, T., Morishima, Y., Shimoda, N., Yamamori, H., Takagi, K., Nakajima, N.,1999. Effect of IGF-1 on protein metabolism in burned rats. J. Parent. Enter. Nutr. 23, 5 Suppl., S93.

Torrallardona, D., Harris, C.I., Fuller, M.F., 2003a. Lysine synthesized by the gastrointestinal microfloraof pigs is absorbed, mostly in the small intestine. Amer. J. Physiol. 284, E1177–E1180.

Torrallardona, D., Harris, C.I., Fuller, M.F., 2003b. Pigs’ gastrointestinal microflora provide them withessential amino acids. J. Nutr. 133, 1127–1131.

van der Schoor, S.R.D., van Goudeoever, J.B., Stoll, B., Henry, J.F., Rosenberger, J.R., Burrin, D.G.,Reeds, P.J., 2001. The pattern of intestinal substrate oxidation is altered by protein restriction in pigs.Gastroenterology 121, 1167–1175.

van der Schoor, S.R., Reeds, P.J., Stoll, B., Henry, J.F., Rosenberger, J.R., Burrin, D.G., van Goudoever,J.B., 2002. The high metabolic cost of a functional gut. Gastroenterology 123, 1931–1940.

van Goudoever, J.B., Stoll, B., Henry, J.F., Burrin, D.G., Reeds, P.J., 2000. Adaptive regulation of intes-tinal lysine metabolism. Proc. Natl. Acad. Sci. 97, 11620–11625.

van Klinken, B.J., Dekker, J., Büller, H.A., de Bols, C., Einerhand, A.W., 1997. Biosynthesis of mucins(MUC2-6) along the longitudinal axis of the human gastrointestinal tract. Amer. J. Physiol. 273,G296–G302.

van Sluijters, D.A., Dubbelhuis, P.F., Blommaart, E.F., Meijer, A.J., 2000. Amino-acid-dependent signaltransduction. Biochem. J. 351, 545–550.

Vary, T.C., Kimball, S.R., 1992. Regulation of hepatic protein synthesis in chronic inflammation andsepsis. Amer. J. Physiol. 262, C445–C452.


Visek, W.J., 1978. The mode of growth promotion by antibiotics. J. Anim. Sci. 46, 1447.von Allmen, D., Hasselgren, P.-O., Higashiguchi, T., Frederick, J., Zamir, O., Fischer, J.E., 1992.

Increased intestinal protein synthesis during sepsis and following the administration of tumour necro-sis factor α or interleukin-1α. Biochem. J. 286, 585–589.

Wakabayashi, Y., Yamada, E., Yoshida, T., Takahashi, N., 1995. Effect of intestinal resection and arginine-free diet on rat physiology. Amer. J. Physiol. 269, G313–G318.

Wang, J.-Y., Viar, M.J., Blanner, P.M., Johnson, L.R., 1996. Expression of the ornithine decar-boxylase gene in response to asparagine in intestinal epithelial cells. Amer. J. Physiol. 271,G164–G171.

Wang, Q., Meyer, T.A., Boyce, S.T., Wang, J.J., Sun, X., Tiao, G., Fischer, J.E., Hasselgren, P.-O., 1998.Endotoxemia in mice stimulates production of complement C3 and serum amyloid A in mucosa ofsmall intestine. Amer. J. Physiol. 275, R1584–R1592.

Weber, F.L. Jr., Fresard, K.M., Veach, G.L., 1989. Stimulation of jejunal mucosal protein synthesis byluminal glucose: effects with luminal and vascular leucine in fed and fasted rats. Gastroenterology 96,935–937.

Wester, T.J., Davis, T.A., Fiorotto, M.L., Burrin, D.G., 1998. Exogenous growth hormone stimulatessomatotropic axis function and growth in neonatal pigs. Amer. J. Physiol. 274, E29–E37.

Wilmore, D.W., Shabert, J.K., 1998. Role of glutamine in immunologic responses. Nutrition 14,618–626.

Windmueller, H.G., Spaeth, A.E., 1974. Uptake and metabolism of plasma glutamine by the small intes-tine. J. Biol. Chem. 249, 5070–5079.

Windmueller, H.G., Spaeth, A.E., 1975. Intestinal metabolism of glutamine and glutamate from thelumen as compared to glutamine from blood. Arch. Biochem. Biophys. 171, 662–672.

Windmueller, H.G., Spaeth, A.E., 1976. Metabolism of absorbed aspartate, asparagine, and arginine byrat small intestine in vivo. Arch. Biochem. Biophys. 175, 670–676.

Windmueller, H.G., Spaeth, A.E., 1978. Identification of ketone bodies and glutamine as the majorrespiratory fuels in vivo for postabsorptive rat small intestine. J. Biol. Chem. 253, 69–76.

Windmueller, H.G., Spaeth, A.E., 1980. Respiratory fuels and nitrogen metabolism in vivo in smallintestine of fed rats: quantitative importance of glutamine, glutamate, and aspartate. J. Biol. Chem.255, 107–112.

Wolff, J.E., Bergman, E.N., Williams, H.H., 1972. Net metabolism of plasma amino acids by liver andportal-drained viscera of fed sheep. Amer. J. Physiol. 223, 438–446.

Wu, G., 1997. Synthesis of citrulline and arginine from proline in enterocytes of postnatal pigs. Amer.J. Physiol. 272, G1382–G1390.

Wu, G., 1998. Intestinal mucosal amino acid catabolism. J. Nutr. 128, 1249–1252.Wu, G., Morris, S.M. Jr. 1998. Arginine metabolism: nitric oxide and beyond. Biochem. J. 336,

1–17.Wu, G., Knabe, D.A., Yan, W., Flynn, N.E., 1995. Glutamine and glucose metabolism in enterocytes of

the neonatal pig. Amer. J. Physiol. 268, R334–R342.Wu, G., Flynn, N.E., Knabe, D.A., 2000a. Enhanced intestinal synthesis of polyamines from proline in

cortisol-treated piglets. Amer. J. Physiol. 279, E395–E402.Wu, G., Flynn, N.E., Knabe, D.A., Jaeger, L.A., 2000b. A cortisol surge mediates the enhanced

polyamine synthesis in procine enterocytes during weaning. Amer. J. Physiol. 279, R554–R559.Wu, G., Haynes, T.E., Li, H., Yan, W., Meininger, C.J., 2001. Glutamine metabolism to glucosamine

is necessary for glutamine inhibition of endothelial nitric oxide synthesis. Biochem. J. 353,245–252.

Wykes, L.J., Fiorotto, M., Burrin, D.G., Del Rosario, M., Frazer, M.E., Pond, W.G., Jahoor, F., 1996.Chronic low protein intake reduces tissue protein synthesis in a pig model of protein malnutrition.J. Nutr. 126, 1481–1488.

Yen, J.T., Pond, W.G., 1990. Effect of carbadox on net absorption of ammonia and glucose into hepaticportal vein of growing pigs. J. Anim. Sci. 68, 4236–4242.

Yen, J.T., Nienaber, J.A., Pond, W.G., 1987. Effect of neomycin, carbadox and length of adaptation tocalorimeter on performance, fasting metabolism and gastrointestinal tract of young pigs. J. Anim. Sci.65, 1243–1248.

Yen, J.T., Nienaber, J.A., Hill, D.A., Pond, W.G., 1989. Oxygen consumption by portal vein-drained organs and by whole animal in conscious growing swine. Proc. Soc. Exp. Biol. Med. 190,393–398.


Yu, Y.M., Young, V.R., Tompkins, R.G., Burke, J.F., 1995. Comparative evaluation of the quantitativeutilization of parenterally and enterally administered leucine and L-[1-13C,15N]leucine within thewhole body and the splanchnic region. J. Parent. Enter. Nutr. 19, 209–215.

Zafarullah, M., Li, W.Q., Sylvester, J., Ahmad, M., 2003. Molecular mechanisms of N-acetylcysteineactions. Cell. Mol. Life Sci. 60, 6–20.

Zou, C.G., Banerjee, R., 2003. Tumor necrosis factor-α-induced targeted proteolysis of cystathionineβ-synthase modulates redox homeostasis. J. Biol. Chem. 278, 16802–16808.


197

8 Nitrogen metabolism by splanchnic tissuesof ruminants

C. K. Reynolds

Department of Animal Sciences, The Ohio State University, OARDC,1680 Madison Avenue, Wooster, OH 44691-4096, USA

Ruminant amino acid metabolism is differentiated from the nonruminant by the extensivedevelopment of the stomach and dietary fermentation that occurs there. These tissues areextremely active metabolically, and their metabolism reduces the net availability of absorbedamino acids. However, for essential amino acids, this metabolism largely reflects the seques-tration of arterial amino acids, and not the utilization of amino acids during their absorption,which represents a relatively small portion of total use by the PDV. The use of essential aminoacids by the PDV largely represents synthesis of constitutive and secreted proteins, andendogenous losses to the gut, as well as oxidation of branched-chain amino acids. Absorptiveuse is more extensive for nonessential amino acids, particularly those used as oxidative fuels.Fermentation in the rumen results in a substantial recycling of urea to the gut lumen and reab-sorption as ammonia, which is subsequently converted to urea in the liver. The costs of ureasynthesis from ammonia appear to be relatively small in terms of oxidative metabolism, andcontrary to earlier suggestions it does not appear to require additional amino acid nitrogen. Incontrast, consumption of excess protein can increase heat production in both the PDV andliver, perhaps as a consequence of surplus amino acid oxidation. Liver removal of amino acidsreflects liver requirements and supply relative to body requirements, but in ruminants a netrelease of branched-chain amino acids is often observed, and leucine is oxidized in the PDVand other peripheral tissues. Finally, amino acids also make important contributions to liverglucose production in ruminants, but apart from alanine, this appears to reflect the availabilityof excess amino acid carbon, and not a metabolic priority, even in very early lactation.

1. INTRODUCTION

The ability of ruminants to derive their nutritional requirements from forages and byproductfeeds is a consequence of their extensive fore-stomach development and fermentative capacity.This pregastric microbial symbiosis characterizes ruminant digestion and has unique effects ontheir protein and amino acid metabolism compared to nonruminants. Microbial fermentation



provides essential amino acids, even in the absence of protein from their diet, allows the uti-lization of nonprotein nitrogen from blood urea for microbial amino acid synthesis andabsorption by the host, and thus enables extensive recycling of nitrogenous compoundsbetween the gut lumen, blood, and body tissues (Lapierre and Lobley, 2001). This enables theruminant to survive, and reproduce, under conditions where forage quality and thus food proteinare scarce. However, the extensive development, metabolism, and fermentation within thestomach are not without consequences for the ruminant animal. Microbial protein digestionand urea catabolism give rise to extensive absorption of ammonia, which must be detoxifiedthrough urea synthesis and amination reactions, largely occurring in the liver. In addition, theextensive development of the musculature and epithelium of the reticulo-rumen and omasum,and their functional activity, give rise to a high rate of protein turnover and amino acidrequirement (MacRae et al., 1997a).

Other considerations for the nitrogen economy of the ruminant include the effects of thefermentative process on the form and pattern in which energy is absorbed. Extensive carbohydratefermentation in the rumen provides a more continuous pattern of energy absorption, primarily inthe form of volatile fatty acids, as typically little starch escapes rumen fermentation and is avail-able for digestion to glucose absorbed from the small intestine. As a consequence, the ruminantrelies primarily on liver gluconeogensis to meet glucose requirements, and glucose is continuallyreleased from the liver, with little apparent diurnal fluctuation in liver glycogen flux (Bergman,1975). The constant requirement for glucose precursors is met to a large extent by propionateabsorbed from the rumen and hindgut, as well as amino acid carbon. In addition, the flow of pro-tein to the small intestine, and amino acid absorption, exhibit less variation than in nonruminants.

Most current feeding standards and models of nutrient utilization used for rationing proteinin beef and dairy cattle are based on estimates of dietary protein degradation in the rumen, thesynthesis of microbial protein using available energy and nitrogenous compounds, and the netflow of feed and microbial and endogenous protein to the abomasum. After reaching the smallintestine, protein digestion and amino acid absorption proceed, much as in nonruminants.While far from perfect, dynamic, mechanistic models of rumen digestion are available for theprediction of metabolizable protein flow to the small intestine (Russell et al., 1992; Dijkstraet al., 2002). However, most of these approaches rely on empirical relationships between esti-mates of total or individual amino acid flow to the small intestine, assumptions about the basalflow of endogenous amino acid secretions into the small intestine, and a fixed efficiency oftransfer from the lumen of the small intestine to their appearance in a product (Sniffen et al.,1992; NRC, 2001). Extensive metabolism of absorbed amino acids occurs in the tissues of thegut and liver, and the extent to which this metabolism determines the quantity and pattern ofamino acids reaching peripheral tissues, and the effects of diet composition and intake level onfractional amino acid utilization, are not certain. Therefore, a greater understanding of theseprocesses, and the contributions of endogenous recycling to amino acid supply, is needed torefine current empirical models of the efficiency of utilization of amino acids reaching thesmall intestine in ruminants, and the development of more enlightened, mechanistic predic-tions of nutrient utilization for production. Ultimately, these models will enable more preciserationing of amino acids for productive purposes and thus reductions in the total amounts ofprotein fed, and subsequent environmental losses arising from the production of beef and milk.

2. SPLANCHNIC TISSUE METABOLISM

While the word “splanchnic” generally applies to the viscera, the term “total splanchnic” hashistorically been used by physiologists as a collective term for the tissues of the portal-drained

C. K. Reynolds198

viscera (PDV) and liver combined (e.g. Elwyn, 1970). The PDV are those tissues drained by thehepatic portal vein, and in ruminants include the gastrointestinal tract (reticulo-rumen, omasum,abomasum, small intestine, cecum, and large intestine), pancreas, spleen, and associatedomental, mesenteric, and other adipose tissue. The PDV and liver are tissues in vascular series,which are integrated anatomically and functionally through their vascular and neural connec-tions. This integrated metabolism determines the flow of nutrients from food to other bodytissues, through the gastrointestinal tract’s role in digestion and absorption, and the liver’smetabolic role as integrator of nutrient supply with requirement. In addition, the splanchnictissues are extremely active metabolically, accounting for as much as 50% of total bodyoxygen consumption, but a considerably lower proportion of body mass (Reynolds, 1995).

This high rate of metabolism requires a high rate of blood flow, and PDV and liver bloodflow (fig. 1) and oxygen consumption (Reynolds, 2002) in cattle increase with diet dry matterintake (DMI). The high rate of oxygen consumption by the splanchnic tissues is in part a con-sequence of a high rate of protein turnover and secretion, and PDV and liver mass, and thusamino acid requirements, increase with greater DMI and metabolizable energy (ME) supply(Burrin et al., 1990; Lobley, 1994). In cattle, daily PDV and liver heat production, estimatedfrom oxygen consumption, increases with greater ME both on a total (fig. 2) or metabolic bodysize (body weight0.75) basis (fig. 3). As much as 38% of cardiac output is distributed to theliver (Huntington et al., 1990), thus while the splanchnic tissues produce a disproportionateamount of the body’s oxidative metabolism and heat, they have access to a disproportionateamount of available nutrients, both during absorption into blood and transfer to the vena cava,as well as from the arterial blood pool (Reynolds, 2002).

Nitrogen metabolism by splanchnic tissues of ruminants 199

Fig. 1. Portal vein and liver blood flow in growing, lactating, and dry mature cattle. Data are averages ofhourly measurements for a given animal and treatment (n = 335). Data sources described by Reynolds (1995,2002), Maltby et al. (1993), and Benson et al. (2002).

C. K. Reynolds200

Fig. 2. Portal-drained visceral (PDV) and liver blood heat production and metabolizable energy (ME) intakein growing, lactating, and dry mature cattle. Data are daily averages of hourly measurements for a givenanimal and treatment (n = 335). Data sources described by Reynolds (1995, 2002), Maltby et al. (1993), andBenson et al. (2002).

Fig. 3. Portal-drained visceral (PDV) and liver blood heat production and metabolizable energy (ME) intakein growing, lactating, and dry mature cattle. Data are daily averages of hourly measurements for a givenanimal and treatment (n = 335) scaled to metabolic body size (body weight0.75). Data sources described byReynolds (1995, 2002), Maltby et al. (1993), and Benson et al. (2002).

2.1. Measurement of net splanchnic metabolism

The use of multicatheterization procedures to measure the quantitative metabolism of nitrogenouscompounds by the splanchnic tissues of ruminants in vivo has been previously described indetail (Katz and Bergman, 1969; Bergman, 1975; Huntington et al., 1989; Seal and Reynolds,1993; Reynolds, 1995). Based on initial studies in dogs (Shoemaker, 1964), pioneering work insheep by E.N. Bergman of Cornell University elegantly described basic aspects of the inter-organ flow of amino acids and other metabolites between the ruminant PDV, liver, kidneys, andhindlimbs (Bergman and Heitmann, 1980; Bergman, 1986). The basic procedure is relativelysimple: permanent indwelling catheters are surgically established in appropriate blood vesselswhich enable the measurement of venous–arterial concentration difference (VA) and bloodflow across the tissue of interest, which are multiplied mathematically to obtain a measurementof net nutrient appearance in venous blood (a positive VA) or removal from arterial blood(a negative VA). Blood flow is typically obtained by measuring the downstream dilution of a dye(usually ρ-aminohippurate) infused into a distal mesenteric vein, but can also be measuredelectronically using electromagnetic or ultrasound probes. Historically, the usefulness of elec-tronic probes for measurement of portal vein blood flow was limited by their longevity andrequirement for a round vessel (i.e. an artery). The more recent development of “transit-time”ultrasound probes (Transonics®, Ithaca, NY) has enabled measurement of total blood flow inblood vessels with irregular shape using long-term (months) in vivo preparations. However,until the recent development of a slimmer probe, the bulky size of the probe body andanatomy of the portal vein have limited the usefulness of transit-time probes for measuringportal vein blood flow in cattle (Lindsay and Reynolds, 2003).

While simple in concept, the obtaining of valid, statistically sound measurements of dietaryeffects or physiological state on splanchnic nutrient flux goes far beyond the maintenance ofcatheter patency. Considerations include laminar flow in the portal vein, heterogeneous portalblood distribution to the liver, the frequency of sampling, postprandial fluctuations, the measure-ment of small VA coupled with large blood flow rates, etc. The PDV represents a diverse,heterogeneous, collection of tissue types, and the small intestinal enterocytes where amino acidabsorption occurs represent a small fraction of the total PDV tissues. Therefore, as a consequenceof their anatomical location, the majority of the PDV tissues must to a large extent rely on thesupply of amino acids in arterial blood for their requirements (MacRae et al., 1997a; Reynolds,2002). Measurements of nutrient flux across sections of the PDV can also be obtained by plac-ing catheters in vessels draining specific tissue beds, such as the mesenteric-drained viscera(MDV), but depending on the species of interest and the location of the sampling tip relative tothe ileocecal vein, the measurement may or may not include the cecum and large intestine (Sealand Reynolds, 1993). As cattle, but not sheep, have a collateral branch of the mesenteric veindraining the ileum (Habel, 1992), measurements of total small intestinal flux, excluding the con-tributions of the hindgut, are exceedingly difficult to obtain in cattle without ligating the collateralbranch. Precise placement of the mesenteric sampling catheter between the convergence of thecollateral and jejunal branches and the ileocecal branch is required, and complete mixing of PAHand venous blood in such a short distance makes measurements of blood flow difficult.

2.2. Isotopic labeling approaches

Measurements of net amino acid flux across the PDV reflect the mathematical summation ofa number of metabolic processes, including absorption from the lumen of the small intestineinto the mesenteric veins, utilization during absorption by the enterocytes, release fromprotein turnover, or synthesis, by a variety of PDV tissues, and removal from arterial blood.


Similarly, liver is not a homogeneous tissue, and simultaneously removes individual amino acidsfrom the portal vein and hepatic artery and releases them into the hepatic veins. To obtain thetrue or “gross” rate of metabolite release and removal by a tissue (often called “unidirectionalflux”), isotopic labeling of the arterial blood pool is used to obtain an estimate of gross nutrientremoval from blood (fractional extraction times total arterial supply, or “arterial use”), whichis added to net flux to calculate the gross release into venous blood (Bergman, 1975). Thisstill underestimates utilization during absorption from the gut lumen, before reaching venousblood. This “absorptive use” can be measured by infusing labeled metabolite into the smallintestine (usually the duodenum or jejunum) and measuring its quantitative appearance in theportal (or mesenteric) vein (MacRae et al., 1997a; Yu et al., 2000; Lindsay and Reynolds, 2003).However, this measurement must be corrected for the extraction of absorbed label from arterialblood, which is accomplished by simultaneously, or on a separate occasion, labeling the arterialblood pool (MacRae et al., 1997a). Otherwise, measurements of absorptive use or “first-pass”metabolism by the small intestine will be overestimated.

Isotopic labeling of the carbon in individual amino acids also allows the measurement of oxi-dation by specific tissues (Lobley et al., 1995, 2003; Lapierre et al., 1999, 2002), depending on theamino acid labeled, the specific carbon labeled, and the metabolic route of oxidation. However,when 13C labeling is used, consideration must be given to the amount of tracer required to labelCO2 relative to tracee turnover, especially under nutritional conditions that reduce oxidation. Inaddition, the PDV release of fermentative and salivary CO2, as well as the removal of arterialblood CO2 (Lobley et al., 2003), complicate the interpretation of CO2 flux across the PDV(Lindsay and Reynolds, 2003). In addition to measurements of amino acid oxidation, the use ofmultiple labeled amino acids, on separate occasions or differentially labeled, allows estimationof the transfer of carbon among individual amino acids and metabolites within specific tissues(Wolff and Bergman, 1972a; Bergman, 1975; Lindsay and Reynolds, 2003).

2.3. Interorgan amino acid exchanges in maintenance-fed sheep

Using a combination of multicatheterization and isotopic labeling procedures, as well as multi-ple gut cannulation, Bergman and his colleagues conducted a detailed series of studiesdescribing the metabolism of amino acids by the visceral tissues and hind limbs of sheep (forreviews see Bergman and Heitmann, 1980; Bergman, 1986). The data provide a picture of thebasic patterns of interorgan amino acid exchange in maintenance-fed or 3-day fasted sheep fedalfalfa pellets, and provided a basis for subsequent studies of the effects of diet composition andphysiological state on amino acid metabolism in sheep and cattle. However, the data were oftenobtained from a limited number of animals sampled multiple times to increase replication (seeWolff et al., 1972), and measurements appear to have been made with limited time for recoveryfrom surgery. Regardless, these concerns do not diminish the contribution that this researchrepresents. The data represent a pioneering effort, and have stood the test of time. As emphasizedin their reviews, the work of Bergman and his colleagues highlighted a number of basic concepts,which will be explored in the sections that follow in light of more recent observations.

3. AMINO ACID UTILIZATION BY THE PORTAL-DRAINED VISCERA

3.1. Absorptive use of amino acids

On a net basis, the disappearance of many amino acids from the lumen of the small intestineof 2 sheep was considerably greater than their simultaneous net appearance in the portal vein

C. K. Reynolds202

(table 1; Tagari and Bergman, 1978). This was especially true for some of the nonessentialamino acids, such as glutamate and aspartate, which are present in large amounts in duodenaldigesta, as well as some of the essential amino acids. The data suggested that there was a con-siderable absorptive use of amino acids by the enterocytes of the small intestine, as observedpreviously for volatile fatty acid absorption by rumen epithelia (Bergman, 1975). This inter-pretation was supported by the work of Windmeuller and Spaeth (1980), which showedthat glutamate, glutamine, and aspartate are important energy substrates for the small intestinalenterocyte of rats. More recent studies in pigs have confirmed that there is an extensive use ofglutamate during its absorption from the small intestine (Stoll et al., 1999). For alanine and serine,the apparent recovery in the portal vein was much higher, but this in part reflects synthesis by thePDV from products of glycolysis. Alanine, which is synthesized by the PDV and peripheralmuscle from pyruvate, is often the amino acid released by the PDV in the largest amount, acrossa variety of nutritional and physiological states (Wolff et al., 1972; Bergman and Heitmann,1980). This in part reflects a transfer of N from the catabolism of other amino acids in thePDV to the liver for ureagenesis, and alanine is typically removed by the liver at rates equal tonet PDV release or greater, on a net basis (Bergman, 1986; Reynolds, 1995; Lobley et al., 2001).


Table 1

Apparent disappearance of amino acids from the small intestine (SI, g/d) in ruminants andtheir net appearance across the portal-drained viscera (PDV) or mesenteric-drained viscera(MDV) as a fraction of disappearance from the SIa

Sheep (n = 2)b Sheep (n = 3)c Lactating dairy cowsd

Amino acid SI, g/d PDV/SI SI, g/d PDV/SI MDV/SI SI, g/d PDV/SI MDV/SI

Leucine 5.02 0.21 7.77 0.65 1.02 127.7 0.62 0.92Valine 3.73 0.33 5.47 0.57 0.95 77.9 0.51 1.11Lysine 3.60 0.66 5.78 0.64 1.09 107.9 0.55 0.76Threonine 2.82 0.38 4.67 0.73 0.99 65.2 0.43 1.15Isoleucine 3.38 0.43 5.08 0.58 1.13 81.7 0.62 1.02Phenylalanine 3.22 0.64 5.06 0.76 1.11 69.6 0.76 1.00Histidine 1.49 0.11 23.6 0.95 1.27Methionine 1.44 0.60 36.3 0.67 1.01Arginine 2.84 0.48 79.6 0.63 1.03Alanine 3.59 0.85 88 0.80 1.16Aspartate 5.97 0.02 153.4 0.08 0.03Asparagine – 0.12 – 0.19 0.37Cysteine 0.61 0.48 15.1 (1.62) 0.28Glutamate 6.23 (0.01) 191.1 0.08 0.11Glutamine – (0.68) – (0.19) 0.20Glycine 3.61 0.52 88.8 0.42 0.57Proline 2.10 0.39 64.5 0.09 0.49Serine 2.10 0.83 58.2 0.75 1.23

a Disappearance calculated as duodenum minus ileal flow, uncorrected for endogenous amino acid flow in theileum. Recoveries of asparagine and glutamine calculated relative to aspartate and glutamate disappearance,respectively. Negative recoveries (net removal from arterial blood) given in parentheses.b Tagari and Bergman (1978). Averages for measurements in 2 sheep fed high- and medium-protein diets.c MacRae et al. (1997b). Averages for measurements in 3 sheep fed 800 or 1200 g alfalfa pellets per day.d Berthiaume et al. (2001). Unlike the data for sheep, measurements of SI, PDV, and MDV were not obtainedsimultaneously in the same animals. Data for SI are from 2 cows, whilst data for PDV are apparently from 3 separate cows, 2 of which were also sampled for MDV.

The extensive absorptive use of amino acids suggested by Tagari and Bergman (1978), andother estimates of “first-pass absorptive use” of amino acids, are in part attributable toanatomical and technical considerations for the measurements reported. As discussed previ-ously (Reynolds, 2002), the high rates of glutamate and aspartate disappearance from thesmall intestine in part reflects the deamination of glutamine and asparagine during the acidhydrolysis of digesta prior to amino acid analysis, and resulting overestimate of their flow tothe duodenum. For other amino acids, their low net recovery in the portal vein largely reflectstheir arterial use by other PDV tissues, such as the stomach. In sheep (MacRae et al., 1997b)and cattle (Berthiaume et al., 2001), measurements of net amino acid release by the MDVwere considerably higher (30–50%) than their net release by the PDV (table 1). This is a con-sequence of the utilization of amino acids from arterial blood by those tissues not drained bythe mesenteric vein. These data strongly suggests that the low recovery of absorbed aminoacids across the PDV, on a net basis, is due to sequestration of amino acids from the arterialblood pool. In these studies (MacRae et al., 1997b; Berthiaume et al., 2001), comparison ofsmall intestinal disappearance and net MDV appearance of essential amino acids showednearly equal, or lower, rates of disappearance from the small intestine compared to appear-ance in the mesenteric vein, on a net basis, but these estimates underestimate true rates ofdisappearance to the extent that endogenous secretions appear in the ileum. For nonessentialamino acids (Berthiaume et al., 2001), the same comparisons of net intestinal disappearanceand MDV appearance (table 1) compare more favorably with the findings of Tagari andBergman (1978), with a low net recovery of aspartate, glutamate, cysteine, proline, andglycine, and a greater appearance of alanine and serine suggesting the synthesis of the lattertwo amino acids by tissues of the MDV.

3.2. Comparison of absorptive and arterial essential amino acid use

Direct measurements of the absorptive use of a mixture of essential amino acids wereobtained by MacRae et al. (1997a) in sheep equipped with multiple intestinal cannulas andcatheters enabling measurement of small intestinal disappearance and PDV flux of aminoacids, along with dual-site isotope infusions to allow simultaneous (consecutive samplings)measurements of absorptive and arterial use of a mixture of 13C-labeled essential amino acids.The measurements confirmed that across the total PDV, arterial use of most essential aminoacids (leucine, valine, lysine, threonine, isoleucine, and histidine) accounts for the majority(75–87%) of total PDV use or “sequestration”. For phenylalanine, a greater proportion (51%)of total PDV utilization was accounted for by absorptive use. Using a similar approach tostudy both PDV and MDV utilization of dual-labeled leucine in sheep (Yu et al., 2000), theMDV accounted for only a small proportion (12%) of total arterial use of leucine by the PDV.Of the leucine sequestered during absorption, oxidation accounted for only 2%, thus mostleucine sequestered during absorption is used for synthesis of constitutive and secretedproteins. Using a similar approach, measurements of arterial and absorptive use of leucineand phenylalanine were obtained in late-lactation dairy cows fed the same ration at two levelsof DMI (16 or 20 kg/day) and two different levels of DMI in the subsequent dry period (8 or12 kg/day; Reynolds et al., 2001a). These data confirmed that arterial use of these amino acidsaccounted for the majority of total PDV use, but absorptive use accounted for a greaterproportion of total PDV phenylalanine use than measured for leucine. Absorptive use wasnegligible at lower DMI within each physiological state, and increased with greater intake,perhaps reflecting an increase in gut mass relative to body requirements. A low absorptive useof phenylalanine was also evident in other studies in lactating dairy cows at restricted intakes

C. K. Reynolds204

(Reynolds et al., 2000), where the increase in PDV release of phenylalanine at the end of a6-day abomasal infusion of a mixture of essential amino acids equivalent to 800 g of milkprotein was 77% of the phenylalanine infused on a net basis, and 99% after correction forarterial use.

Taken together, these data confirm that there is substantial utilization of amino acids by thetissues of the PDV. For many nonessential amino acids, with some notable exceptions, theirutilization during absorption is substantial, and they make important contributions to the energyand synthetic process of the enterocytes of the small intestine. Those contributions, in thenonruminant small intestine, are discussed in more detail in other chapters. For most of theessential amino acids, absorptive use accounts for 25%, or less, of total PDV sequestration.Therefore, metabolism by the PDV tissues largely reflects use of amino acids from the arterialpool, and absorptive use has less effect on the quantity and profile of amino acids reaching theportal vein during the “first pass” of absorption than suggested by simple net flux measurements.For phenylalanine, absorptive use accounted for a greater proportion of total PDV sequestration,but this reflected less quantitative arterial use relative to the other essential amino acids, andnot more absorptive use per se (MacRae et al., 1997a; Reynolds et al., 2001a).

3.3. Arterial use of essential amino acids by the PDV

In sheep, total PDV sequestration of essential amino acids varied from 32% (histidine) to 67%(valine) of whole-body irreversible loss (IRL), indicating a substantial contribution of PDVtissues to whole-body protein synthesis (MacRae et al., 1997a). In dairy cows, total PDVsequestration of leucine (27–40%) and phenylalanine (22–37%) represented a smaller pro-portion of body IRL (table 2), but the proportion was increased by greater DMI in both dryand lactating cows, yet was not affected by stage of lactation and associated differences inaverage DMI (Reynolds et al., 2001a). This suggests that PDV sequestration of these aminoacids, as a fraction of total body IRL, is affected by their supply from the diet relative torequirement. Although these measurements were not corrected for oxidation, or release of4-methyl-2-oxopentanoate (MOP), it is likely that a large portion of the leucine sequesteredby the PDV is used for protein synthesis. Numerous studies have shown that the PDV ofsheep and cattle accounts for a major portion of body protein synthesis (24–35%), dependingon the technology used and interpretation of the results obtained (Lobley, 1994; Lobley et al.,1995, 1996; Lapierre et al., 1999, 2002). Therefore, the tissues of the PDV have a substantialrequirement for essential amino acids, which must to a large extent be met by extraction fromthe arterial supply.

The relationship between PDV protein turnover and essential amino acid use is illustratedby the close agreement between the proportions of essential amino acids sequestered by thePDV and their relative concentration in constitutive proteins (MacRae et al., 1997a). In addi-tion, branched-chain amino acids are to a large extent oxidized by extrahepatic tissues, but itis apparent that leucine, and perhaps other branched-chain amino acids, are oxidized by thePDV (Lapierre et al., 1999, 2002; Lobley et al., 2003). This oxidation appears to be sensitiveto leucine supply relative to requirement (van der Schoor et al., 2001). In dairy cows, PDVoxidation of leucine tended to increase when a higher protein diet was fed (Lapierre et al.,2002), but the effect of protein status on PDV oxidation of other amino acids has not beendetermined in ruminants, to my knowledge, and deserves further research. For phenylalanine,little oxidation occurs in the PDV (Reynolds et al., 2000; Lobley et al., 2003), but liver clear-ance and hydroxylation is extensive relative to other essential amino acids (Reynolds et al.,2000; Lobley et al, 2001). This in part reflects the conversion of phenylalanine to tyrosine, as


well as its use for protein synthesis, but largely represents oxidative metabolism. In lactatingdairy cows in mid-lactation, gross liver removal was equal to 70–80% of total body IRL(Reynolds et al., 2000). Of the phenylalanine removed, 33–57% was oxidized, while only4–8% was converted to tyrosine which was subsequently released into blood.

3.4. Endogenous amino acid secretions

In addition to constitutive protein synthesis and oxidation, essential amino acid sequestrationby the PDV must also support the synthesis of protein appearing in the lumen of the gut asendogenous secretions (or sloughage of constitutive proteins). Depending on their site ofentry into the gut lumen, these endogenous amino acids are reabsorbed or lost in the faeces.Tagari and Bergman (1978) did not account for endogenous amino acid flow in the ileum intheir study, and thus underestimated total amino acid disappearance. Measurements ofendogenous amino acid flow within the lumen of the gut are difficult to obtain and are oftenassumed to be a fixed, basal amount from the regression of amino acid flow on intake(MacRae et al., 1997a). Other approaches for estimating endogenous protein or amino acidflow in the gut are based on statistical interpretation of digesta amino acid composition(Larsen et al., 2000), or the isotopic labeling of precursor pools and gut contents, often using15N- or 13C-labeled leucine (see Ouellet et al., 2002).

Using long-term intravenous 15N-labeled leucine infusion to estimate total endogenous N flowinto specific components of duodenal and fecal nitrogen flow, and mathematical modeling ofN exchanges, Ouellet et al. (2002) recently estimated that total endogenous N in the duodenum,arising from urea and amino acid N transfer to bacterial N, or appearing directly as free endoge-nous N, accounted for 24% of total N flow into the duodenum, and 4% of total N loss in thefeces. Using 13C-labeled leucine infused into the jugular vein for 7 days, endogenous leucinerepresented 8% of duodenal leucine flow to the duodenum (table 2; Reynolds et al., 2001a),

C. K. Reynolds206

Table 2

Effects of intake level during lactation and the subsequent dry period on leucine andphenylalanine kinetics and endogenous leucine flow in the duodenum and ileum of 3 dairycows (Reynolds et al., 2001b)

Dry Lactating P<

Low High Low High SEM Stage DMI Intera

DMI, kg/d 8.0 12.0 14.9 19.4 0.4 – – –Milk yield, kg/d – – 14.3 16.5 2.9 – 0.200 –Leucine metabolism, mmol/h

Body irreversible loss rate 51.5 63.2 73.0 94.4 5.0 0.905 0.015 0.393Total PDV sequestrationb 14.8 29.3 25.3 40.3 3.2 0.153 0.021 0.884Duodenal flow 35.4 46.6 82.0 106.9 3.3 0.030 0.010 0.204Endogenous duodenal flow 3.2 3.4 6.4 8.9 0.5 0.040 0.025 0.042Ileal flow 9.8 15.2 21.1 26.2 1.1 0.307 0.003 0.611Endogenous ileal flow 1.5 2.2 3.0 3.6 0.5 0.693 0.129 0.848

Phenylalanine metabolism, mmol/hBody irreversible loss rate 20.4 25.4 28.5 33.6 1.5 0.898 0.013 0.832Total PDV sequestrationb 6.4 10.4 6.3 17.7 1.7 0.070 0.017 0.163

a Interaction of DMI and stage of lactation.b Sum of arterial and absorptive use.

a number that is in agreement with measurements based on 14C-labeled leucine in sheep(Marsden et al., 1988). Endogenous leucine flow in the ileum accounted for 14.5% of totalleucine flow, and the fraction of duodenal and ileal leucine flow of endogenous origin wasunaffected by stage of lactation or DMI, thus endogenous leucine flow increased linearly withincreased DMI (table 2). Larsen et al. (2000) also estimated an increase in endogenous aminoacid flow into the duodenum with increased DMI. These data show that, rather than comprisinga basal amount equal to flow in fasted animals, endogenous amino acid flow increases withintake and digesta flow. While the values obtained through the use of 13C- or 14C-leucine label-ing are considerably lower than estimates of the total endogenous contribution to N flow in theduodenum of dairy cows (Ouellet et al., 2002), they are similar to the estimate of the contri-bution of free endogenous sources (Ouellet et al., 2002), and thus the differences surely reflecttechnical differences between the two methods for estimating endogenous amino acid flow.

The appearance of endogenous amino acids in the rumen and small intestine represents aroute of N recycling, which has been estimated to account for 30–40% of N absorption asamino acids (Lapierre and Lobley, 2001). Based on their models, Ouellet et al. (2002) esti-mated that endogenous secretions account for as much as 30% of total PDV sequestration ofamino acids. In our study (table 2), endogenous leucine flow in the duodenum and ileumaccounted for 19–37% of total leucine sequestration by the PDV of dry and lactating dairycows. While the reabsorption of endogenous N represents an opportunity for reuse of N, italso incurs a cost in terms of the rate of protein synthesis in gut tissues (Reeds et al., 1999;van der Schoor et al., 2002). In ruminants, the efficiency of utilization of absorbed essentialamino acids for growth (largely muscle accretion) is typically lower (50–55%) than in non-ruminants (70–75%), and this may well be a consequence of the disproportionately high ratesof body protein turnover occurring in the gut compared to nonruminants (MacRae et al.,1997a). These high rates of protein turnover and endogenous secretion appear to be one cost ofsupporting the extensive development of metabolically active rumen tissues and the consumptionof high-fiber diets that increase gut fill.

4. AMMONIA ABSORPTION AND UREA SYNTHESIS

Initial studies of net amino acid absorption across the PDV in sheep fed alfalfa pellets (Wolffet al., 1972) also highlighted the extensive absorption of ammonia N compared to amino acid N.Across the PDV, net ammonia absorption accounted for a greater fraction of dietary N intake(48%) than did the net absorption of N as amino acids (26%). This in part reflects the diet fed,as the alfalfa pellets fed were relatively high in protein (20%), which is highly rumen-degradable.However, across a range of dietary treatments, net PDV absorption of ammonia N is highlycorrelated with N or digestible N intake, and typically exceeds the absorption of total aminoacid N (Seal and Reynolds, 1993; Reynolds, 1995; Lapierre and Lobley, 2001; Lindsay andReynolds, 2003).

In a recent summary of published data from cattle (Lindsay and Reynolds, 2003), increasednet PDV release of ammonia N accounted for 41% of incremental N intake, whilst simulta-neous increases in α-amino N or total amino acid absorption accounted for 31% of theincrements in N intake. The ammonia absorbed is derived from microbial degradation ofnitrogenous compounds, including feed protein and endogenous urea and proteins, as well asany ammonia arising from metabolism within PDV tissues. As emphasized by Lapierre andLobley (2001), a substantial portion of ammonia absorbed is derived from blood urea. Theyestimated that, on average, roughly two-thirds of urea synthesized is transferred to the lumenof the gut via saliva and direct blood transfer, which may involve a specific transporter


(Waterlow, 1999). Of the urea N transferred to the gut, 40% was recycled as absorbed ammonia,whilst 50% was absorbed as amino acids synthesized in the rumen. The remainder was lostin the feces, presumably in the form of microbial protein synthesized in the hindgut. Asmicrobial protein synthesis is a major route of urea and ammonia N utilization in the gutlumen, dietary N supply and digestion is not the only determinant of ammonia absorption,and the supply of fermentable energy and other factors affecting microbial growth alsoimpacts net ammonia appearance in the portal vein. For example, starch infusion into eitherthe rumen or abomasum of lactating dairy cows decreased net PDV release of ammonia(Reynolds et al., 1998), and abomasal starch infusion increased fecal excretion of N, presumablyas a consequence of incomplete starch digestion in the small intestine and increased micro-bial protein synthesis in the hindgut (Reynolds et al., 2001b).

4.1. Costs of excess protein digestion and urea synthesis in the liver

Ammonia absorbed into the portal vein is efficiently cleared by the liver and detoxified byincorporation into urea and to a lesser extent other nitrogenous compounds such as glutamine(Lobley et al., 2000). When considered in isolation, the urea cycle requires four phosphatebonds per mole of urea synthesized, and on this basis ureagenesis in the liver appears toaccount for a substantial portion of liver oxygen use (Reynolds et al., 1991b; Lobley et al.,1995; Reynolds, 1995; Milano et al., 2000). This accounting is supported by calorimetry studiesin which increased heat production was measured in sheep infused into the abomasum withammonia or urea, or intravenously with urea (Martin and Blaxter, 1965). These studies sug-gested a greater systemic cost of urea synthesis than the theoretical cost of convertingammonia to urea, which was attributed to increased cycling of urea and ammonia betweenthe blood and gut lumen (Martin and Blaxter, 1965). In addition, calorimetry studies withdairy cattle suggested an energetic cost of consuming digestible protein in excess of require-ments equivalent to 30 kJ ME per g N (Tyrrell et al., 1970). However, these increases in heatproduction (i.e. oxygen consumption) were measured on a whole-body basis.

In multicatheterized cattle fed isonitrogenous diets differing in forage:concentrate ratio atequal ME intakes, digested N, ammonia absorption and liver removal, and liver urea synthesiswere greater when the high-forage diet was fed, but liver oxygen consumption was not affectedsignificantly (table 3; Reynolds et al., 1991a,b). Liver urea release and oxygen consumptionincreased with intake of both rations, but this was associated with an increase in ME, glucosesynthesis, and presumably liver mass. In other work, adding urea to a high-protein diet fed atmaintenance to beef cattle caused a large increase in ammonia absorption and liver urearelease, but again had no significant effect on liver oxygen consumption (table 3; Maltby et al.,1993). Subsequent studies in sheep in which ammonia delivery to the liver and ureagenesiswere increased by ammonia infusion (table 3) have also failed to show a statistically signifi-cant increase in liver oxygen consumption (Lobley et al., 1995, 1996), although numericaltrends for increased oxygen uptake were observed in one study when ammonia absorptionwas increased by 165% (Milano et al., 2000). These observations suggest that increases inammonia absorption and subsequently liver urea synthesis do not require increased oxidativemetabolism in the liver. This may reflect shifts in other metabolic processes within the liver,to provide the ATP required for ureagenesis without increasing oxidative metabolism in total.An alternative explanation is that the net cost of urea synthesis is lower when the ATP gainof fumarate metabolism is included in the balance sheet, which reduces the energy cost of ureaformation by 75% (Reynolds et al., 1991b). This accounting does not explain the observedincrease in body oxygen consumption reported in sheep infused with ammonium bicarbonate

C. K. Reynolds208


Table 3

Effects of diet intake and composition on net portal-drained visceral ammonia absorption(PDV NH3N) and net liver release of urea N and use of oxygen in beef cattle

kg/d mmol/h

Animal and diet DMI N intake PDV NH3N Liver urea N Liver O2 use

Growing heifersa

Low intake75% alfalfa 4.75 0.133 186 354 71975% corn:SBMb 3.60 0.098 143 235 621

High intake75% alfalfa 7.78 0.209 340 593 120575% corn:SBM 6.51 0.174 250 491 1192

Diet effect, P < 0.001 0.001 0.067 0.080 0.460Mature steersc

75% alfalfa 5.67 0.153 253 363 92675% alfalfa + urea 5.80 0.209 398 537 947Diet effect, P < 0.001 0.001 0.001 0.004 0.457

Growing steersd

Medium intake75% corn 5.10 0.096 121 199 82175% corn:SBM 4.98 0.131 212 323 865

High intake75% corn 6.87 0.128 165 289 109175% corn:SBM 6.96 0.179 261 473 1180

Diet effect, P < 0.718 0.001 0.001 0.001 0.051Mature sheep–mesenteric vein NH3N infusione

Alfalfa pellets 0.71 0.018 27.1 37.7 92.4+ 12.6 mmol/h NH4Cl 0.71 0.018 39.5 61.9 110.4

NH3 effect, P < 0.001 0.071 0.535Mature sheep–mesenteric vein NH3N infusionf

Grass pellets 1.09 0.035 37.0 69.0 151.2+ 7.5 mmol/h N4CO3 1.09 0.035 47.8 84.9 194.4

Grass pellets–barley 1.08 0.031 42.6 71.2 173.4+ 7.5 mmol/h N4CO3 1.08 0.031 57.7 85.8 188.4

NH3 effect, P < 0.001 0.001 0.252

(Martin and Blaxter, 1965), but differences in the relative amount of ammonia infused maybe a factor (Milano et al., 2000).

In contrast to the studies cited in the preceding discussion, increases in dietary proteinlevel, via the addition of soybean meal to a corn-based diet fed to growing beef steers, sig-nificantly increased body heat production (Reynolds et al., 1992a), in line with the results ofTyrrell et al. (1970). In this case, increases in body heat production with higher dietary pro-tein intake were associated with increases in ammonia absorption (table 3), as well asincreased absorption of some amino acids (Reynolds et al., 1995a), and the increases in bodyoxygen consumption were attributable to relatively small, but significant, increases in bothliver and PDV oxygen consumption. The significant increase in liver oxygen use in this laterstudy may in part reflect the larger number of animals sampled, and low variation observed.However, in sheep fed low-protein concentrate-based diets supplemented with urea or twosources of protein, liver oxygen consumption was increased when the supplemental protein

Continued

was fed, but not urea (Ferrell et al., 2001; table 3). This was also associated with a trend forgreater net PDV release of α-amino N when the protein-based supplements were fed.Therefore, increases in oxygen consumption observed in animals fed protein in excess ofrequirement may reflect an increased oxidation of amino acids, rather than an energy cost ofurea synthesis from ammonia per se. As emphasized by Bergman and Heitmann (1980),interorgan cycles involving a number of amino acids shuttle N and carbon between the liverand peripheral tissues, and provide spatial separation of the urea cycle. With the exception ofbranched-chain amino acids, oxidation of essential amino acids occurs to a large extent in theliver, but oxidation also occurs in peripheral tissues such as muscle, which is the primary siteof branched-chain amino acid catabolism (Layman, 2003). It is apparent that PDV tissues par-ticipate in the oxidative metabolism of leucine and other amino acids (Lapierre et al., 1999,2002; Lobley et al., 2001, 2003).

Urea synthesis and excretion essentially represents the catabolism of amino acids availablein excess of their requirement for anabolic and metabolic functions, and the fraction ofammonia N absorbed that is not recycled to the gut. In the hepatocyte, one N in urea arisesdirectly from ammonia, whilst the other is derived from aspartate, which is derived from glu-tamate and thus indirectly from ammonia. Previous studies in multicatheterized cattleobserved an increase in liver amino acid removal under conditions of increased ammoniaabsorption, suggesting that the synthesis of aspartate from ammonia via glutamate might belimiting urea synthesis, thus increasing liver catabolism of amino acids to support ureagene-sis (Reynolds, 1992; Parker et al., 1995). However, variables other than ammonia supply to

C. K. Reynolds210

Table 3—Cont’d

Effects of diet intake and composition on net portal-drained visceral ammonia absorption(PDV NH3N) and net liver release of urea N and use of oxygen in beef cattle

kg/d mmol/h

Animal and diet DMI N intake PDV NH3N Liver urea N Liver O2 use

Mature sheep–mesenteric vein NH3N infusiong

Grass pellets 0.78 0.017 20.6 42.6 96.6+ 9 mmol/h NH4CO3 0.78 0.017 35.3 54.7 96.0+ 24 mmol/h NH4CO3 0.78 0.017 54.5 81.4 120.2

NH3 effect, P < 0.001 0.002 0.130Sheep–95% concentrateh

Control 1.10 0.012 6.5 37.4 178+ Urea 1.13 0.021 12.8 52.3 185+ SBM 1.19 0.021 11.7 55.2 229+ BFM 1.01 0.018 11.8 68.3 239

Protein effect,i P < 0.57 0.43 0.40 0.18 0.01

a Reynolds et al. (1991a,b).b Soybean meal.c Maltby et al. (1993).d Reynolds et al. (1992a).e Lobley et al. (1995). Sheep received 5-day mesenteric vein infusions of low (1.5 mmol/h) or high(14.1 mmol/h) NH4Cl.f Lobley et al. (1996). Sheep received 4-day mesenteric vein infusions of low (1.5 mmol/h NH4Cl)or high (1.5 mmol/h NH4Cl plus 7.5 mmol/h NH4CO3) ammonium salt.g Milano et al. (2000). Sheep received 4-day mesenteric vein infusions of 0, 9, or 24 mmol/h NH4CO3.h Ferrell et al. (2001). Sheep were fed a low-protein control concentrate (6.6% crude protein) supplemented(to 11.2% crude protein) with urea, SBM, or blood and feather meal (BFM).i Comparison of urea with SBM and BFM protein supplements.

the liver were also altered in the studies forming the basis of this hypothesis (e.g. Reynoldset al., 1991b). In the intervening period a detailed, systematic series of studies has beenconducted by G.E. Lobley and his colleagues at the Rowett Research Institute to explorethe effects of increased ammonia supply on liver metabolism in sheep, under experimental con-ditions that sought to control variation in other metabolic variables affecting liver metabolism,and incorporating novel isotopic labelling experimentation and mathematical modeling (Lobleyet al., 1995, 1996; Milano et al., 2000; Milano and Lobley, 2001). Initial indications suggestedan effect of mesenteric vein NH4Cl infusion on leucine oxidation (Lobley et al., 1995).However, more recent studies in which shifts in acid–base status were avoided by infusingNH4HCO3 (Lobley and Milano, 1997; Milano et al., 2000; Milano and Lobley, 2001) foundno indication that increased amino acid deamination is required for the synthesis of urea underconditions of increased ammonia load, at least in maintenance-fed sheep.

4.2. Regulation of urea synthesis

Since its description seventy years ago, the regulation of the ornithine cycle and liver ureasynthesis has been the subject of numerous reviews and much discussion (e.g. Waterlow,1999; Lobley et al., 2000), and will not be considered in detail in the present chapter.Generally, liver removal of amino acids is determined by the functional requirements of theliver (e.g. constitutive and export protein synthesis) and the availability of amino acids in theblood pool relative to body requirements. The extent to which ureagenesis in the liver isdirectly regulated by factors other than plasma amino acid concentrations has been the sub-ject of debate, but after decades of research, current thinking is that “it is very unlikely thatthe signal (integrating protein intake and urea cycle activity) is an alteration in the plasmaconcentration either of total amino-N or any single amino acid” (Waterlow, 1999). In growingsteers treated with growth hormone-releasing factor, increases in body N retention, at equalN intake, were accompanied by decreases in liver removal of α-amino and ammonia N andrelease of urea, but with no change observed in blood α-amino N concentration (Lapierre et al., 1992; Reynolds et al., 1992b). Similar responses were observed in growing steers treatedwith growth hormone (Bruckenthal et al., 1997). In nonruminants, growth hormone treatmentcaused a decrease in the activity of urea cycle enzymes in the liver, but the effects may havebeen an indirect consequence of increased protein retention in vivo (McLean and Gurney,1963; Palekar et al., 1981). Abomasal infusion of glucose decreased liver urea production andincreased N retention in maintenance-fed sheep (Obitsu et al., 2000), presumably througheffects of insulin, but the effects on urea production may also reflect indirect responses toincreased amino acid retention. Similarly, abomasal infusion of starch for 2 weeks markedlyincreased tissue N retention in late-lactation, pregnant dairy cows (Reynolds et al., 2001b).

5. LIVER AMINO ACID METABOLISM

In maintenance-fed sheep (Wolff et al., 1972; Bergman and Heitmann, 1980), liver removalof a number of amino acids accounted for substantial portions of their net PDV absorp-tion and release, but as emphasized by Bergman (1986), the animals were not in positivetissue N balance. Regardless, these studies identified a number of interorgan cycles involvingspecific groups of amino acids. For the glucose precursors alanine, serine, and glycine, theirnet removal by the liver was greater than their net absorption, such that their net total splanchnicflux was negative. A negative total splanchnic flux for these amino acids represents thecontributions of peripheral tissues to their liver metabolism, and the hind limbs released them


C. K. Reynolds212

on a net basis, reflecting the catabolism of glucose and other amino acids, such as leucine(Layman, 2003). The branched-chain amino acids were removed by the liver, but in the lowestamount relative to net PDV release, such that their net total splanchnic release was more positivethan for other amino acids, and they were removed by the hindlimbs. Glutamate was releasedby the liver, whilst glutamine was removed, and the opposite occurred in the hindlimbs, repre-senting a means of transporting ammonia N from peripheral tissues to the liver for ureasynthesis. A similar interorgan “cycle” was apparent for arginine and ornithine plus citrulline,which shuttled N from the kidney and hindlimbs to the liver as arginine. These data suggestthat some of this arginine was synthesized via the action of arginine synthetase in peripheraltissues, using ornithine and citrulline which were released by the liver and subsequentlyremoved by the kidney and hind limbs (Bergman and Heitmann, 1980). Again, this spatialseparation of the urea cycle provides opportunity for metabolic flexibility within and amongtissues of the body. On average, the liver removed more amino acid N than appeared acrossthe PDV (Wolff et al., 1972), but this reflects the level of feed intake and N status of the ani-mals, and the fact that net absorption underestimates true rates of absorption to the extent thatamino acids are utilized by the PDV.

With greater intake, the relationship between liver removal of amino acids and their true ornet absorption across the PDV will vary, depending in part on the energy status of the animal,which determines protein requirement in growing animals. Similarly, relationships betweenPDV and liver flux of amino acids in dairy cattle will be determined by stage of lactation andrelative milk protein production. In attempting to ascertain the fractional clearance of aminoacids during their absorption and passage through the liver, information needed for the devel-opment of mechanistic models of amino acid utilization, amino acids mixtures (or proteins)have been infused into the abomasum, to increase absorption, or mesenteric veins, to mimicincreased absorption, over short (hours) or long (days) periods (e.g. Reynolds and Tyrrell,1991; Reynolds et al., 1995b, 1999, 2000; Bruckenthal et al., 1997; Wray-Cahen et al., 1997;Lobley et al., 2001). The interpretation of these results must consider a variety of factors, suchas the amounts infused relative to requirements, the time allowed for adaptation to the infusedamino acids, the form in which the amino acids are provided (protein vs free amino acids),and the balance of amino acids provided (see Wolfe and Miller, 2002). Essential amino acidsare not required in isolation, thus provision of a limiting amino acid often creates a deficiencyof a second limiting amino acid. Therefore, the efficiency of utilization of one amino acid,and thus liver clearance, is dependent on a balanced supply of other amino acids relative tobody requirements.

The inverse of this concept of a balanced amino acid supply is illustrated by a model forthe creation of a deficiency of a specific amino acid through the feeding of a low-protein dietto lactating goats, and the infusion of a mixture of essential amino acids into the abomasumwhich is balanced relative to milk protein composition but for the absence of a specific aminoacid, such as histidine (Bequette et al., 2000). Under these circumstances, the metabolic adap-tations of the mammary gland to maximize histidine supply for milk protein synthesis(increased blood flow and mammary histidine extraction) are nothing short of amazing.Similarly, the metabolic and production responses of the gut and liver to mesenteric vein infusionof mixtures of amino acids based on milk protein composition were affected by the presenceor absence of nonessential amino acids (Reynolds et al., 1995b). In other studies, the responseof net PDV amino acid release and liver removal of essential amino acids (Reynolds et al.,1999), and PDV sequestration of leucine and phenylalanine (Caton et al., 2001), differedwhen the same quantity of essential amino acids was provided as a component of casein,which also provided supplemental nonessential amino acids, or a mixture of free essential


amino acids. Differences in whole-body and splanchnic leucine metabolism were alsoobserved in rats fed supplemental free amino acids compared to the same mixture of aminoacids provided as casein (Daenzer et al., 2001).

Despite these factors, some general patterns of liver metabolism are apparent across studiesin cattle and sheep (Lobley et al., 1996, 2001; Lobley and Milano, 1997; Wray-Cahen et al.,1997; Lapierre et al., 1999; Reynolds et al., 1999, 2000, 2001a; Blouin et al., 2002). As a pro-portion of net PDV absorption, liver removal of alanine, serine, glycine, phenylalanine, andhistidine is typically high. This in part reflects the role of the first three as glucose precursors(Bergman and Heitmann, 1980), the role of glycine in detoxification processes, and the highconcentration of phenylalanine and histidine in plasma protein synthesized by the liver(Lobley et al., 2001). In contrast, net liver removal of the branched-chain amino acids is typicallylow relative to their rate of absorption, as observed in maintenance-fed sheep (Bergman, 1986).In cattle, the liver typically releases leucine and often valine and isoleucine (Blouin et al., 2002).Their net removal was increased when their supply to the liver was increased by abomasalcasein infusion, but not a mixture of free essential amino acids, which caused an increase intheir net liver release (Reynolds et al., 1999).

Similar responses were observed when mixtures of free amino acids, based on the compo-sition of milk protein, were infused into the mesenteric vein of early-lactation dairy cows feda low-protein diet for 3 days (Reynolds et al., 1995b). Infusion of the mixture of essentialamino acids increased net liver release of leucine and tended to increase the release of valine,while the same amino acid mixture infused along with nonessential amino acids had no effecton their net flux across the liver. On a net basis, the liver releases leucine in dogs as well, butthis could be attributable in part to liver removal and transamination of MOP (Abumrad et al.,1982). In dairy cows, net liver removal of MOP is insufficient to account for the leucinereleased (Lapierre et al., 2002), suggesting other sources such as peptides or blood-borneproteins arising in other tissues and degraded in the liver (Elwyn, 1970). i-Valerate is a productof leucine catabolism in the liver, and the synthesis of branched amino acids from branched-chain volatile fatty acids in the ruminant liver has been proposed (van der Walt, 1993). Inlactating dairy cows, the absorption and liver removal of i-valerate cows (e.g. Reynolds et al.,2003) is typically much greater than net leucine release (e.g. Lapierre et al., 2002). Leucinedelivery to peripheral tissues is important not only as an essential amino acid for proteinsynthesis, but also as a regulator of protein synthesis and other anabolic processes (Abumradet al., 1982), in part through modification of insulin signalling (Layman, 2003).

5.1. Gluconeogenesis

In sheep, Wolff and Bergman (1972b) estimated that amino acid carbon accounts for from11% to 30% of liver glucose synthesis, based on measured transfers of alanine, glutamate,aspartate, glycine, and serine 14C to glucose (11% of glucose production), or the maximalpotential net contribution of all the plasma amino acids removed by the liver (30% of glucoseproduction). There is no question that the carbon from these glucogenic amino acids makesan important contribution to liver glucose synthesis in ruminants. Net liver removal of otherprecursors (propionate, lactate, glycerol, i-butyrate, and n-valerate) is seldom adequate toaccount for all of the glucose released by the liver, but the extent to which amino acid supplylimits glucose production is not certain.

Glucose production by the liver is regulated by requirement, thus the provision of addi-tional precursor that is efficiently removed by the liver, such as propionate or alanine,typically decreases liver lactate removal, without affecting glucose production in fed animals

(Reynolds, 1995). However, in early-lactation dairy cows, glucose production is often con-sidered to be limited by precursor supply from the diet, as occurs in fasting, and increases inpyruvate carboxylase activity have suggested a greater contribution of alanine, and by con-jecture other amino acids, to glucose production in early lactation (Drackley et al., 2001).However, in early-lactation cows fed low-protein diets, mesenteric vein infusion of a mixtureof nonessential amino acids for 3 days had no significant effect on liver glucose production,and significantly reduced milk yield (Reynolds et al., 1995b). In contrast, abomasal infusionsof casein, or an equivalent mixture of essential amino acids, increased liver glucose produc-tion (Reynolds et al., 1999). This suggests that the response to casein was not due to theprovision of nonessential amino acids, but perhaps an effect of essential amino acid supplyon liver metabolism or glucose requirement. Even in very early lactation, the net balance ofliver glucose release and precursor removal does not support the concept of an obligatoryincrease in nonessential amino acid contributions to liver glucose synthesis, apart from anincreased potential contribution of alanine (Reynolds et al., 2003). Between 9 days beforecalving and 11 days after calving, the potential contribution of alanine to liver glucose releasedoubled (from 2.5% to 5%), but increases in net liver removal of propionate, lactate, alanine,and glycerol were sufficient to account for all of the increase in liver glucose release. Thissuggests that the amino acid contributions to liver glucose synthesis in early lactation arisefrom the obligatory deamination of amino acids in the liver, rather than being a metabolicrequirement of early lactation.

6. CONCLUSIONS

The extensive development of the ruminant forestomach distinguishes their N economy fromthat of nonruminants. The microbial fermentation that occurs there markedly alters the pro-file and form of protein and amino acids presented to the small intestine for digestionand absorption, and provides opportunities for extensive recycling of N between the body andlumen of the gut. This recycling occurs via exchanges of ammonia and urea with blood andluminal pools, endogenous gut and secretory N entry to the gut lumen, microbial protein syn-thesis, and the subsequent digestion and absorption of microbial and endogenous aminoacids. The costs of this exquisite microbial symbiosis include the costs of urea synthesis,which may be less then hypothesized at the level of the liver, and an extensive utilization ofabsorbed amino acids from arterial blood, which masks their net appearance from the lumenof the gut into the portal vein. Liver metabolism of amino acids includes substantial require-ments for liver functions and the integration of the supply of nitrogenous compounds fromthe diet with body requirements. A more detailed understanding of these processes within thesplanchnic tissues and their response to changes in diet composition and intake, relative tonutrient requirements for production, is needed to improve current empirical models of theefficiency of amino acid utilization for production. Such data are forthcoming; however, theprediction of productive responses to changes in amino acid supply from the small intestinecan not be based simply on estimates of nutrient supply. The genetic and environmentalpropensity of the animal for the productive use of their nutrient supply will ultimately deter-mine their N economy.


The ability of current feed-rationing and nutrient requirement systems to predict productiveresponse of ruminants to changes in the supply of protein to the small intestine is currently

C. K. Reynolds214

limited by an oversimplification of postabsorptive metabolism. Future approaches will needto address the quantitative metabolism of individual amino acids, which is highly integrated,complex, flexible, and adaptive. Ultimately, the efficiency with which nitrogenous com-pounds are utilized will be determined by the propensity of the animal fed for productivenutrient use. However, the exact mechanisms by which nutrient absorption and tissue require-ments are integrated are poorly understood. For example, the production of urea from aminoacids is largely a consequence of the availability of amino acids in excess of requirement, yetthe signal by which this oversupply is communicated to the urea cycle remains a mystery.Although the urea cycle was the first metabolic cycle delineated, and despite years of researchon its regulation, it remains a potentially fruitful area for future research. Unraveling the costsof excess protein intake in terms of the integration of amino acid catabolism and urea gene-sis is particularly relevant. The processes of amino acid catabolism do not occur solely in theliver for all amino acids, yet the response of amino acid metabolism in the gut and other extra-hepatic tissues to increased protein intake are still poorly described. The regulation ofbranched-chain amino acid metabolism and their role in regulating metabolic processesdeserves particular attention in this regard. In addition, the metabolic source of the branched-chain amino acids released by the liver of ruminants should be determined. As restrictions onN losses from animal production facilities increase, so to will the need to more precisely formu-late diets to meet the requirements for specific amino acids. This will only be achieved througha greater understanding of the metabolic integration of dietary supply with tissue demand.

REFERENCES

Abumrad, N.N., Wise, K.L., Williams, P.E., Abumrad, N.A., Lacy, W.W., 1982. Disposal of α-ketoiso-caproate: roles of liver, gut and kidneys. Amer. J. Physiol. 243, E123–E131.

Benson, J.A., Reynolds, C.K., Aikman, P.C., Lupoli, B., Beever, D.E., 2002. Effects of abomasal longchain fatty acid infusion on splanchnic nutrient metabolism in lactating diary cows. J. Dairy Sci. 85,1804–1814.

Bequette, B.J., Hanigan, M.D., Calder, A.G., Reynolds, C.K., Lobley, G.E., MacRae, J.C., 2000. Kineticsof amino acid transport in the mammary gland of lactating goats when histidine supply is limiting formilk production. J. Dairy Sci. 83, 765–775.

Bergman, E.N., 1975. Production and utilisation of metabolites by the alimentary tract as measured inportal and hepatic blood. In: McDonald, I.W., Warner, A.C.I. (Eds.), Digestion and Metabolism in theRuminant. The University of New England, Armidale, Australia, p. 292.

Bergman, E.N., 1986. Splanchnic and peripheral uptake of amino acids in relation to the gut. Fed. Proc.45, 2277–2282.

Bergman, E.N., Heitmann, R.N., 1980. Integration of whole-body amino acid metabolism. In: Buttery,P.J., Lindsay, D.B. (Eds.), Protein Deposition in Animals. Butterworth, London, pp. 69–84.

Berthiaume, R., Dubreuil, P., Stevenson, M., McBride, B.W., Lapierre, H., 2001. Intestinal disappearanceand mesenteric and portal appearance of amino acids in dairy cows fed ruminally protected methion-ine. J. Dairy Sci. 84, 194–203.

Blouin, J.P., Bernier, J.F., Reynolds, C.K., Lobely, G.E., Dubreuil, P., Lapierre, H., 2002. Effect of diet qual-ity on splanchnic fluxes of nutrients and hormones in lactating diary cows. J. Dairy Sci. 85, 2618–2630.

Bruckenthal, I., Huntington, G.B., Baer, C.K., Erdman, R.A., 1997. The effect of abomasal infusion ofcasein and recombinant somatotropin hormone injection on nitrogen balance and amino acid fluxesin portal-drained viscera and net hepatic and total splanchnic blood in Holstein steers. J. Anim. Sci.75, 1119–1129.

Burrin, D.G., Ferrell, C.L., Britton, R.A., Bauer, M., 1990. Level of nutrition and visceral organ size andmetabolic activity in sheep. Brit. J. Nutr. 64, 439–448.

Caton, J.S., Reynolds, C.K., Bequette, B.J., Lupoli, B., Aikman, P.C., Humphries, D.J., 2001. Effects ofabomasal casein or essential amino acid infusions on splanchnic leucine and phenylalanine metabo-lism in lactating dairy cows. J. Dairy Sci. 84, Suppl. 1, 363.


Daenzer, M., Petzke, K.J., Bequette, B.J., Metges, C.C., 2001. Whole-body and splanchnic amino acidmetabolism differ in rats fed mixed diets containing casein or its corresponding amino acid mixture.J. Nutr. 131, 1965–1972.

Dijkstra, J., Mills, J.A.N., France, J., 2002. The role of dynamic modelling in understanding the micro-bial contribution to rumen function. Nutr. Res. Rev. 15, 67–90.

Drackley, J.K., Overton, T.R., Douglas, G.N., 2001. Adaptations of glucose and long-chain fatty acidmetabolism in liver of dairy cows during the periparturient period. J. Dairy Sci. 84, E. Suppl.E100–E112.

Elwyn, D.H., 1970. The role of the liver in regulation of amino acid and protein metabolism. In:Munro, H.N. (Ed.), Mammalian Protein Metabolism. Academic Press, New York, p. 523.

Ferrell, C.L., Freetly, H.C., Goetsch, A.L., Kreikemeier, K.K., 2001. The effect of dietary nitrogen andprotein on feed intake, nutrient digestibility, and nitrogen flux across the portal-drained viscera andliver of sheep consuming high-concentrate diets ad libitum. J. Anim. Sci. 79, 1322–1328.

Habel, R.E., 1992. Guide to the Dissection of Domestic Ruminants. Habel, R.E., Ithaca, NY.Huntington, G.B., Reynolds, C.K., Stroud, B., 1989. Techniques for measuring blood flow in the splanch-

nic tissues of cattle. J. Dairy Sci. 72, 1583–1595.Huntington, G.B., Eisemann, J.H., Whitt, J.M., 1990. Portal blood flow in beef steers: comparison of tech-

niques and relation to hepatic blood flow, cardiac output and oxygen uptake. J. Anim. Sci. 68, 1666–1673.Katz, M.L., Bergman, E.N., 1969. Simultaneous measurements of hepatic and portal venous blood flow

in the sheep and dog. Amer. J. Physiol. 216, 946–952.Lapierre, H., Lobley, G.E., 2001. Nitrogen recycling in the ruminant: a review. J. Dairy Sci. 84, E. Suppl.

E223–E236.Lapierre, H., Reynolds, C.K., Elsasser, T.H., Gaudreau, P., Brazeau, P., Tyrrell, H.F., 1992. Effects of

growth hormone-releasing factor and intake on energy metabolism in growing beef steers: nethormone metabolism by portal-drained viscera and liver. J. Anim. Sci. 70, 742–751.

Lapierre, H., Bernier, J.F., Dubreuil, P., Reynolds, C.K., Farmer, C., Ouellet, D.R., Lobley, G.E., 1999.The effect of intake level on splanchnic protein metabolism in growing beef steers. Brit. J. Nutr.81, 457–466.

Lapierre, H., Blouin, J.P., Bernier, J.F., Reynolds, C.K., Dubreuil, P., Lobley, G.E., 2002. Effect of diet qual-ity on leucine metabolism across splanchnic tissues in lactating dairy cows. J. Dairy Sci. 85, 2631–2641.

Larsen, M., Madsen, T.G., Weisbjerg, M.R., Hvelplund, T., Madsen, J., 2000. Endogenous amino acidflow in the duodenum of dairy cows. Acta Agric. Scand. Sect. A. Anim. Sci. 50, 161–173.

Layman, D.K., 2003. The role of leucine in weight loss diets and glucose homeostasis. J. Nutr. 133,261S–267S.

Lindsay, D.B., Reynolds, C.K., 2003. Metabolism of the portal-drained viscera. In: Dyjkstra, J., France, J.,Forbes, J.M. (Eds.), Quantitative Aspects of Ruminant Digestion and Metabolism. CABI,Wallingford, UK. Submitted.

Lobley, G.E., 1994. Amino acid and protein metabolism in the whole body and individual tissues of rumi-nants. In: Asplund, J.M. (Ed.), Principles of Protein Nutrition in Ruminants. CRC Press, London, p. 147.

Lobley, G.E., Bremner, D.M., Brown, D.S., 2001. Response in hepatic removal of amino acids by thesheep to short-term infusions of varied amounts of an amino acid mixture into the mesenteric vein.Brit. J. Nutr. 85, 689–698.

Lobley, G.E., Connell, A., Lomax, M.A., Brown, D.S., Milne, E., Calder, A.G., Faringham, D.A.H., 1995.Hepatic detoxification of ammonia in the ovine liver: possible consequences for amino acid catabo-lism. Brit. J. Nutr. 75, 217–235.

Lobley, G.E., Milano, G.D., 1997. Regulation of hepatic nitrogen metabolism in ruminants. Proc. Nutr.Soc. 56, 547–563.

Lobley, G.E., Milano, G.D., van der Walt, J.G., 2000. The liver: integrator of nitrogen metabolism. In:Cronje, P.B. (Ed.), Ruminant Physiology: Digestion, Metabolism, Growth and Reproduction. CABInternational, Wallingford, UK, pp. 149−168.

Lobley, G.E., Shen, X., Le, G., Bremmer, D.M., Milne, E., Calder, A.G., Anderson, S.E., Dennison, N.,2003. Oxidation of essential amino acids by the ovine gastro-intestinal tract. Brit. J. Nutr. 89, 617–629.

Lobley, G.E., Weijs, P.J.M., Connell, A., Calder, A.G., Brown, D.S., Milne, E., 1996. The fate of absorbedand exogenous ammonia as influenced by forage or forage-concentrate diets in growing sheep. Brit.J. Nutr. 76, 231–248.

MacRae, J.C., Bruce, L.A., Brown, D.S., Calder, A.G., 1997a. Amino acid use by the gastrointestinal tractof sheep given lucerne forage. Amer. J. Physiol. 273, G1200–G1207.

C. K. Reynolds216

MacRae, J.C., Bruce, L.A., Brown, D.S., Farningham, D.A.H., Franklin, M., 1997b. Absorption of aminoacids from the intestine and their net flux across the mesenteric- and portal-drained viscera of lambs.J. Anim. Sci. 75, 3307–3314.

Maltby, S.A., Reynolds, C.K., Lomax, M.A., Beever, D.E., 1993. The effect of increased absorption ofammonia and arginine on splanchnic metabolism of beef steers. Anim. Prod. 56, 462.

Marsden, M., Bruce, C.I., Bartram, C.G., Buttery, P.J., 1988. Initial studies on leucine metabolism in therumen of sheep. Brit. J. Nutr. 60, 161–171.

Martin, A.K., Blaxter, K.L., 1965. The energy cost of urea synthesis in sheep. In: Blaxter, K.L. (Ed.),Energy Metabolism of Farm Animals. EAAP Publ. No. 11. Academic Press, London, pp. 83–91.

McLean, P., Gurney, M.W., 1963. Effect of adrenalectomy and of growth hormone on enzymes concernedwith urea synthesis in rat liver. Biochem. J. 87, 96–104.

Milano, G.D., Hotston-Moore, A., Lobley, G.E., 2000. Influence of hepatic ammonia removal on ure-agenesis, amino acid utilization and energy metabolism in the ovine liver. Brit. J. Nutr. 83, 307–315.

Milano, G.D., Lobley, G.E., 2001. Liver nitrogen movements during short-term infusion of high levels ofammonia into the mesenteric vein of sheep. Brit. J. Nutr. 86, 507–513.

NRC, 2001. Nutrient Requirements of Dairy Cattle, 7th Revised Edition. National Academy Press,Washington, DC.

Obitsu, T., Bremmer, D., Milne, E., Lobley, G.E., 2000. Effect of abomasal glucose infusion on alaninemetabolism and urea production in sheep. Brit. J. Nutr. 84, 157–163.

Ouellet, D.R., Demers, M., Zuur, G., Lobley, G.E., Seoane, J.R., Nolan, J.V., Lapierre, H., 2002. Effectof dietary fiber on endogenous nitrogen flows in lactating dairy cows. J. Dairy Sci. 85, 3013–3025.

Palekar, A.G., Collipp, P.J., Maddaiah, V.T., 1981. Growth hormone and rat liver mitochondria: Effectson urea cycle enzymes. Biochem. Biophys. Res. Commun. 100, 1604–1610.

Parker, D.S., Lomax, M.A., Seal, C.J., Wilton, J.C., 1995. Metabolic implications of ammonia produc-tion in the ruminant. Proc. Nutr. Soc. 54, 549–563.

Reeds, P.J., Burrin, D.G., Stoll, B., van Goudoever, J.B., 1999. Consequences and regulation of gutmetabolism. Proceedings of the VIIIth International Symposium on Protein Metabolism andNutrition. EAAP Publ. No. 96, Wageningen Press, Wageningen, The Netherlands, pp. 127–153.

Reynolds, C.K., 1992. Nitrogen metabolism by ruminant liver. J. Nutr. 122, 850–854.Reynolds, C.K., 1995. Quantitative aspects of liver metabolism in ruminants. In: engelhardt, W.,

Leonhard-Marek, S., Breves, G., Giesecke, D. (Eds.), Ruminant Physiology: Digestion, Metabolism,Growth and Reproduction: Proceedings of the 8th International Symposium on Ruminant Physiology.Ferdinand Enke Verlag, Stuttgart, Germany, p. 351.

Reynolds, C.K., 2002. Economics of visceral tissue metabolism in ruminants: toll keeping or internal rev-enue service? J. Anim. Sci. 80, E. Suppl., E74–E84.

Reynolds, C.K., Aikman, P.C., Lupoli, B., Humphries, D.J., Beever, D.E., 2003. Splanchnic metabolismof dairy cows during the transition from late gestation through early lactation. J. Dairy Sci. 86,1201–1217.

Reynolds, C.K., Bequette, B.J., Caton, J.S., Humphries, D.J., Aikman, P.C., Lupoli, B., Sutton, J.D.,2001a. Effects of intake and lactation on absorption and metabolism of leucine and phenylalanine bysplanchnic tissues of dairy cows. J. Dairy Sci. 84, Suppl. 1, 362.

Reynolds, C.K., Cammell, S.B., Humphries, D.J., Beever, D.E., Sutton, J.D., Newbold, J.R., 2001b.Effects of post-rumen starch infusion on milk production and energy metabolism in dairy cows.J. Dairy Sci. 84, 2250–2259.

Reynolds, C.K., Casper, D.P., Harmon, D.L., Milton, C.T., 1992a. Effect of CP and ME intake on vis-ceral nutrient metabolism in beef steers. J. Anim. Sci. 70, Suppl. 1, 315.

Reynolds, C.K., Crompton, L.A., Firth, K., Beever, D., Sutton, J., Lomax, M., Wray-Cahen, D., Metcalf, J.,Chettle, E., Bequette, B., Backwell, C., Lobley G., MacRae, J., 1995b. Splanchnic and milk proteinresponses to mesenteric vein infusion of 3 mixtures of amino acids in lactating dairy cows. J. Anim. Sci.73, Suppl. 1, 274.

Reynolds, C.K., Harmon, D.L., Prior, R.L., Casper, D.P., Milton, C.T., 1995a. Splanchnic metabolism of aminoacids in beef steers fed diets differing in CP content at two ME intakes. J. Anim. Sci. 73, Suppl. 1, 270.

Reynolds, C.K., Humphries, D.J., Cammell, S.B., Benson, J., Sutton, J.D., Beever, D.E., 1998. Effects of abomasal wheat starch infusion on splanchnic metabolism and energy balance of lactating dairycows. In: McCracken, K.J., Unsworth, E.F., Wylie, A.R.G. (Eds.), Energy Metabolism of FarmAnimals: Proceedings of the 14th Symposium on Energy Metabolism, CAB International,Wallingford, UK, p. 39.


C. K. Reynolds218

Reynolds, C.K., Lapierre, H., Tyrrell, H.F., Elsasser, T.H., Staples, R.C., Gaudreau, P., Brazeau, P.,1992b. Effects of growth hormone-releasing factor and intake on energy metabolism in growing beefsteers: net nutrient metabolism by portal-drained viscera and liver. J. Anim. Sci. 70, 752–763.

Reynolds, C.K., Lupoli, B., Aikman, P.C., Benson, J.A., Humphries, D.J., Crompton, L.A., France, J.,Beever, D.E., MacRae, J.C., 2000. Effects of diet protein level and abomasal amino acid infusions onsplanchnic metabolism in lactating dairy cows. J. Dairy Sci. 83, Suppl. 1, 299.

Reynolds, C.K., Lupoli, B., Aikman, P.C., Humphries, D.J., Crompton, L.A., Sutton, J.D., France, J.,Beever, D.E., MacRae, J.C., 1999. Effects of abomasal casein or essential amino acid infusions onsplanchnic metabolism in lactating dairy cows. J. Anim. Sci. 77, Suppl. 1, 266.

Reynolds, C.K., Tyrrell, H.F., 1991. Effects of mesenteric vein L-alanine infusion on liver metabolism inheifers fed on diets differing in forage-to-concentrate ratio. Brit. J. Nutr. 66, 437–450.

Reynolds, C.K., Tyrrell, H.F., Reynolds, P.J., 1991a. Effects of diet forage-to-concentrate ratio and intakeon energy metabolism in growing beef heifers: whole body energy and nitrogen balance and visceralheat production. J. Nutr. 121, 994–1003.

Reynolds, C.K., Tyrrell, H.F., Reynolds, P.J., 1991b. Effects of diet forage-to-concentrate ratio and intakeon energy metabolism in growing beef heifers: net nutrient metabolism by visceral tissues. J. Nutr.121, 1004–1015.

Russell, J.B., O’Connor, J.D., Fox, D.G., Van Soest, P.J., Sniffen, C.J., 1992. A net carbohydrate andprotein system for evaluating cattle diets: I. Ruminal fermentation. J. Anim. Sci. 70, 3551–3561.

Seal, C.J., Reynolds, C.K., 1993. Nutritional implications of gastrointestinal and liver metabolism inruminants. Nutr. Res. Rev. 6, 185–208.

Shoemaker, W.C., 1964. Methods and techniques for measurement of hepatic physiology and metabolism.In: Rouiller, C. (Ed.), The Liver: Morphology, Biochemistry, Physiology. Academic Press, New York,pp. 243–266.

Sniffen, C.J., O’Connor, J.D., Van Soest, P.J., Fox, D.G., Russell, J.B., 1992. A net carbohydrate and proteinsystem for evaluating cattle diets: II. Carbohydrate and protein availability. J. Anim. Sci. 70, 3562–3577.

Stoll, B., Burrin, D.G., Henry, J., Yu, H., Jahoor, F., Reeds, P.J., 1999. Substrate oxidation by the portaldrained viscera of fed piglets. Amer. J. Physiol. 277, E168–E175.

Tagari, H., Bergman, E.N., 1978. Intestinal disappearance and portal blood appearance of amino acids insheep. J. Nutr. 108, 790–803.

Tyrrell, H.F., Moe, P.W., Flatt, W.P., 1970. Influence of excess protein intake on energy metabolism ofthe dairy cow. In: Schurch, A., Wenks, C. (Eds.), Energy Metabolism of Farm Animals. EAAP Publ.No. 13. Juris Verlag, Zürich, Switzerland, pp. 69–72.

van der Schoor, S.R.D., van Goudoever, J.B., Stoll, B., Henry, J.F., Rosenberger, J.R., Burrin, D.G.,Reeds, P.J., 2001. The pattern of intestinal substrate oxidation is altered by protein restriction in pigs.Gastroenterology 121, 1167–1175.

van der Schoor, S.R.D., Reeds, P.J., Stoll, B., Henry, J.F., Rosenberger, J.R., Burrin, D.G., van Goudoever, J.B.,2002. The high metabolic cost of a functional gut. Gastroenterology 123, 1931–1940.

van der Walt, J.G., 1993. Nitrogen metabolism of the ruminant liver. Aust. J. Agric. Res. 44, 381–403.Waterlow, J.C., 1999. The mysteries of nitrogen balance. Nutr. Res. Rev. 12, 25–54.Windmueller, H.G., Spaeth, A.E., 1980. Respiratory fuels and nitrogen metabolism in vivo in small intes-

tine of rats: quantitative importance of glutamine, glutamate and aspartate. J. Biol. Chem. 255, 107–112.Wolfe, R.R., Miller, S.L., 2002. Protein metabolism in response to ingestion pattern and composition of

proteins: introduction. J. Nutr. 132, 3207S.Wolff, J.E., Bergman, E.N., 1972a. Metabolism and interconversions of five plasma amino acids by tissues

of the sheep. Amer. J. Physiol. 223, 447–454.Wolff, J.E., Bergman, E.N., 1972b. Gluconeogenesis from plasma amino acids in fed sheep. Amer. J. Physiol.

223, 455–460.Wolff, J.E., Bergman, E.N., Williams, H.H., 1972. Net metabolism of plasma amino acids by liver and

portal-drained viscera of fed sheep. Amer. J. Physiol. 223, 438–446.Wray-Cahen, D., Metcalf, J.A., Backwell, F.R.C., Bequette, B.J., Brown, D.S., Sutton, J.D., Lobley, G.E.,

1997. Hepatic response to increased exogenous supply of plasma amino acids by infusion into themesenteric vein of Holstein-Friesian cows in late gestation. Brit. J. Nutr. 78, 913–930.

Yu, F., Bruce L.A., Calder, A.G., Milne, E., Coop, R.L., Jackson, F., Horgan, G.W., MacRae, J.C., 2000.Subclinical infection with the nematode Trichostrongylus colubriformis increases gastrointestinal tractleucine metabolism and reduces availability of leucine for other tissues. J. Anim. Sci. 78, 380–390.

PART IIILipid metabolism

221

9 Hepatic fatty acid oxidationand ketogenesis in young pigs1

J. Odle, P. Lyvers-Peffer, and X. Lin

Department of Animal Science, North Carolina State University,Raleigh, NC 27695, USA

A primary limitation to efficient pork production is morbidity and mortality during the peri-natal period. Because pigs are born with low energy reserves, their survival hinges on timelyconsumption of milk. In contrast to carbohydrate-based fetal metabolism, the transition to amilk-based diet necessitates rapid biochemical adaptations to accommodate the oxidation offatty acids that comprise more than 60% of milk energy. From research reported to date, thedegree to which neonatal pigs make these adaptations is questionable. In stark contrast toother mammalian neonates, piglets do not demonstrate elevated ketogenesis despite highmilk-fat intake. Ketone bodies play a pivotal role in the transition from carbohydrate-basedmetabolism to fat-based metabolism, providing an important alternative fuel for glucose-dependent tissues. Impaired adaptation limits the piglets’ ability to oxidize fat which likelycontributes to the etiology of mortality. Therefore, this review considers the developmentalaspects of lipid oxidation in the young pig. The key regulatory enzymes previously elucidatedin rodents are reviewed, with inclusion of the limited knowledge available in pigs. Furtherresearch in this area will hopefully assist in development of strategies (via nutritional and/orexogenous hormonal manipulation) to enhance development of fatty acid oxidation andultimately improve piglet survival and growth.

1. INTRODUCTION

Impaired growth and high mortality of neonatal pigs pose significant challenges to the swineindustry. Postnatal mortality varies among production units, but has been recently estimatedby the Agricultural Statistics Service of the United States Department of Agriculture to averageapproximately 12% of live births, and has shown only modest improvement over the past twentyyears (USDA, 2002). In addition, it is estimated that prenatal (embryo and fetal) mortality in



1 Supported in part by a grant from the USDA-NRI, No. 98-35206-6645.

swine may be as high as 25% so that, collectively, these data imply that the number of pigletsweaned per litter currently may be less than 65% of true potential. Associated problems ofslow growth (morbidity) further add to the inefficiency. The combined economic impact ofthese losses is enormous. The cost is ultimately carried by the consumer in the price paid forpork. For this reason, there is great impetus for identifying and studying the stressors respon-sible for the high postnatal mortality. The etiology is complex, as a number of factors maycontribute, including nutritional deficiency, low immunocompetence and disease resistance,hypothermia, and crushing by the dam. Many retrospective survey studies have attempted todetermine the relative importance of these stressors, but epidemiological/survey approacheshave contributed little useful information because the cause of death is difficult to determineprecisely, and interactions among the factors complicate interpretation. Consequently, ifprogress is to be made, controlled experimentation is needed to better understand the devel-oping piglet’s nutritional, immunological, and behavioral responses to its environment.

This review addresses a metabolic component of this multifactorial etiology, examining thebiochemical competency of piglets to oxidize fatty acids during early postnatal life. In par-ticular, the ontogeny and regulation of hepatic fatty acid oxidation is highlighted owing to thedramatically low level of ketogenesis expressed in neonatal pigs compared to other species.

2. THE NEED FOR RAPID DEVELOPMENTOF FATTY ACID OXIDATION

Prior to birth, the fetus oxidizes predominantly glucose, lactate, and amino acids (Battagliaand Meschia, 1978). At parturition, the newborn must elicit the behavioral responses neces-sary to acquire milk from the dam. This requires effective competition among littermates andoccurs in a thermal environment that may be more than 10ºC below the animal’s critical tem-perature (Stanier et al., 1984). Rohde Parfet and Gonyou (1988) have shown that >30 minmay lapse before the first milk is consumed and considerably longer time is required beforepositive energy balance is regained. Owing to limited body reserves at birth, negative energybalance quickly becomes life-threatening to the piglet. Survival therefore hinges on the timelyconsumption of the dam’s milk which provides 60% of its calories as fat (Ferre et al., 1986).These events necessitate rapid metabolic adaptations to shift from carbohydrate-based fetalmetabolism to fat-based postnatal metabolism.

Other mammalian species, faced with a similar challenge, demonstrate elevated ketogene-sis during this transition (Girard et al., 1992). For example, ketogenesis measured inhepatocytes from newborn rats increases 6-fold between 0 and 16 h of age (Ferre et al., 1983),and blood ketone body concentrations may exceed 1.5 mM (Foster and Bailey, 1976) in thesuckling rat. Thus, neonatal hyperketonemia plays a significant role in the energy economyof the neonate (Girard et al., 1992), sparing glucose and providing carbon for lipogenesis inneural tissue. In contrast, piglets do not display hyperketonemia (Bengtsson et al., 1969;Pegorier et al., 1981) (i.e. less than 0.2 mM) and this may further compromise their survival.Available data indicate that piglets apparently digest and absorb milk fat with high efficiency(digestion coefficients >95% at 2 days of age) and that a major portion of absorbed fat isdeposited in adipose tissues prior to weaning (e.g. pigs are <2% fat at birth and ~15% atweaning). However, relatively little is known regarding the oxidative fate of lipid in the pigletand even less is known regarding the regulation of lipid oxidation in this species. Beforereviewing the available literature addressing development of fatty acid oxidation in piglets,the major regulatory features of fatty acid oxidation (elucidated in other species) aredescribed below.

J. Odle et al.222

3. REGULATION OF FATTY ACID OXIDATION AND KETOGENESIS

The major metabolic pathways involved in hepatic fatty acid metabolism are summarizedschematically in fig. 1. In animal cells there are two fatty acid β-oxidation systems, onelocated in the mitochondria and the second in the peroxisome. The mitochondrion is consid-ered the primary site for fatty acid β-oxidation, while the peroxisome is considered analternative pathway. Under conditions that enhance peroxisome proliferation, the relativecontribution of the peroxisome to total fatty acid oxidation in the liver may be as high as 30%in the rat (Kondrup and Lazarow, 1985) and 47% in the neonatal pig (Yu et al., 1997a).

Hepatic fatty acid oxidation and ketogenesis in young pigs 223

Fig. 1. Pathways of hepatic lipid metabolism with emphasis on oxidative metabolism. Enzymes/pathwaysare numbered as follows: (1) long-chain acyl-CoA synthetase, (2) acetyl-CoA carboxylase, (3) variousacyl-CoA transferases, (4) carnitine shuttle consisting of CPT I, translocase, and CPT II, (5) medium-chainacyl-CoA synthetase, (6) mitochondrial hydroxymethylglutaryl-CoA synthase, (7) acyl-CoA dehydrogenase,(8) long-chain acyl-CoA synthetase, (9) very long-chain acyl-CoA synthetase, (10) acyl-CoA oxidase,(11) carnitine octanoyltransferase.

The mitochondria and peroxisome are often found in close proximity to lipid droplets andare believed to work in concert (Latruffe et al., 2001). The β-oxidation reactions of the twosystems are similar beginning with an initial dehydrogenation followed by a hydration, asecond dehydrogenation, and finally thiolytic cleavage to produce acetyl-CoA and an acyl-CoA shortened by two carbons (see Reddy and Mannaerts, 1994, for review). Although thereactions are similar, the actual proteins involved differ between the β-oxidation systems.

3.1. Biochemical dogma

Long-chain fatty acids (LCFA) are activated to their CoA thioesters via synthetases (fig. 1,enzyme 1; EC 6.2.1.3) located in the ER and in the outer membrane of the mitochondria (Aas,1971). To date, five genetic variants of the long-chain acyl-CoA synthetases have been clonedfrom rodent species (Oikawa et al., 1998) and their differential regulation may influence themetabolic fate of the activated fatty acids (Lewin et al., 2001). While preference is shown forLCFA substrates, those of medium-chain length (MCFA, i.e. C6–C10) may be activated aswell. The fatty acyl-CoA (FA-CoA) may then be esterified via acyl-CoA transferases (fig. 1,enzyme 3; EC 2.3.1.15) located in the ER, forming various triglycerides, cholesterol esters,phospholipids, etc., which may be exported as lipoproteins (VLDL; Coleman et al., 2000).These transferases have low affinity for MCFA-CoAs such that MCFA are obligate fuels (Bachand Babayan, 1982). The FA-CoA may alternatively be transported into the mitochondria viathe coordinated activities of three membrane proteins: carnitine acyltransferase I (CAT, fig. 1,enzyme 4; EC 2.3.1.21) catalyzing the formation of FA-carnitine from FA-CoA outside of themitochondrial matrix, translocase catalyzing the exchange/diffusion (antiport) of FA-carnitinefor free carnitine across the inner mitochondrial membrane, and carnitine acyltransferase II(CAT II), similar to CAT I except residing on the matrix side of the inner membrane.

The CAT I and II activities are catalyzed by a family of acyltransferase enzymes. Proteins havebeen identified with optimum activity toward C2, C8, and C16 FA-CoAs in various tissues. The latter, referred to as carnitine palmitoyltransferase (CPT; discussed below) is likely of great-est importance in the young pig given the predominantly long-chain fatty acid composition ofsow’s milk. Within the mitochondrial matrix, the FA-CoA are subjected to oxidation at the β-carbon-yielding acetyl-CoA, which may be further oxidized to CO2 in the TCA cycle.Alternatively, it may exit as acetylcarnitine (using the CAT system in reverse), may produceketone bodies (acetoacetate, β-OH-butyrate, and acetone), or may be hydrolyzed to free acetateby acetyl-CoA hydrolase. Hydroxymethylglutaryl-CoA (HMG-CoA) synthase (fig. 1, enzyme 6;EC 4.1.3.5) is presumably the rate-limiting enzyme in the ketogenic pathway (Williamson et al.,1968). Cytosolic acetyl-CoA (derived from the mitochondria via CAT or citrate lyase or from theperoxisome) may be carboxylated to malonyl-CoA by acetyl-CoA carboxylase (ACC, fig. 1,enzyme 2; EC 6.4.1.2) within the lipogenic pathway wherein FA-CoA is synthesized de novo.

Very long-chain and long-chain fatty acids are also activated to their CoA thioesters pre-ceding catabolism in the peroxisome. The peroxisomal membrane contains two synthetases,a long-chain fatty acid synthetase positioned on the cytosolic side of the membrane (fig. 1,enzyme 8), and a very long-chain fatty acid (VLCFA) synthetase positioned toward the per-oxisomal matrix (fig. 1, enzyme 9). The location of the VLCFA synthetase is responsible forthe substrate specificity between the peroxisome and the mitochondrion. While the mito-chondria can oxidize long, medium, and short-chain fatty acids, VLCFA are either poorlyoxidized by this organelle or not at all (Lazo et al., 1990).

Following CoA activation, the initial dehydrogenation step in the peroxisome is catalyzedby multiple acyl-CoA oxidases (fig. 1, enzyme 10); two have been identified in the human

J. Odle et al.224

(Wanders et al., 2001) and three in the rat (Van Veldhoven et al., 2001). While acyl-CoAdehydrogenase, the first enzyme in mitochondrial β-oxidation, produces two ATP as electronsare donated directly to coenzyme Q of the electron transport chain (ETC), the first step inperoxisomal β-oxidation catalyzed by acyl-CoA oxidase results in the production of H2O2 aselectrons are passed directly to molecular oxygen. As a result, the peroxisome is approxi-mately half as efficient as the mitochondrion in producing energy from the β-oxidation offatty acids. Acyl-CoA oxidase shows very little affinity for medium- and short-chain fattyacids (see Reddy and Mannaerts, 1994, for review); as a result, fatty acids are only chain-shortened in the peroxisome. In addition, the peroxisome, lacking TCA cycle enzymes,cannot metabolize the acetyl-CoA to CO2, nor can it produce ketone bodies because theenzymes of ketogenesis are also absent. Therefore, the end-products of peroxisomal fatty acidoxidation include acetyl-CoA and chain-shortened acyl-CoA.

The transport of fatty acids and the subsequent end-products of their oxidation across theperoxisomal membrane is a subject of much debate. It was originally believed that the per-oxisomal membrane was highly permeable. However, isolation of the peroxisome results in aloss of the structural integrity of the membrane (Wanders et al., 2001); therefore, many ear-lier studies involving peroxisome permeability may have been misleading. Studies involvingS. cerevisiae provide evidence for the involvement of transport proteins (see Hettema andTabak, 2000, for review); in addition, peroxisomal half transporters have been identified inhumans, although their definitive role in the transport of fatty acids has not been fully eluci-dated (Wanders and Tager, 1998).

Although a CAT protein has been identified in the peroxisome, it is not membrane-boundand is therefore not implicated in the transport of FA across the membrane. The CAT identi-fied in the peroxisome, or carnitine octanoyltransferase (COT, fig. 1, enzyme 11) as it is oftenreferred to in literature, has optimum activity toward fatty acids of medium chain length. It isspeculated that the peroxisomal COT catalyzes the conversion of the end-products of perox-isomal β-oxidation to their carnitine esters. Subsequently, these carnitine esters may exit theperoxisome and be directed toward the mitochondria where they may undergo completeoxidation to CO2. Studies have shown that 4,8-dimethylnonanoyl-CoA derived from theincomplete oxidation of pristanic acid in the peroxisome is indeed translocated to the mito-chondria for complete oxidation (Verhoeven et al., 1998). Furthermore, the concerted action ofthe two β-oxidation systems is further supported by the observation that natural and syntheticligands (i.e. ligands of the peroxisome proliferator-activated receptor or PPAR), whichincrease peroxisome proliferation and peroxisomal enzymes, also increase mitochondrialenzymes involved in fatty acid metabolism (i.e. CPT I and HMG-CoA synthase).

3.2. Regulation of carnitine palmitoyltransferase I (CPT I)

Using fed, fasted, and alloxan-diabetic rats, McGarry and Foster (1980) have reported con-siderable evidence establishing the allosteric control of CPT I by malonyl-CoA. Duringphysiological states in which lipogenesis is occurring, ACC is activated and the associatedhigh level of malonyl-CoA serves to inhibit CPT I and thereby prevent the simultaneous andfutile oxidation of fatty acids by preventing their entry into the mitochondria. As such, regu-lation at CPT I is thought to function in directing FA-CoA between esterification andoxidative fates. Beyond changes in malonyl-CoA concentration, changes in the sensitivity ofCPT I to malonyl-CoA inhibition have also been reported in various physiological statesincluding the perinatal period in rabbits (Prip-Buus et al., 1990). Furthermore, low levels ofcarnitine in tissues of colostrum-deprived neonates (Borum, 1983) could limit transport and


thus oxidation of fatty acids. Milk, however, is high in carnitine (Kerner et al., 1984), andsuckling results in elevated hepatic carnitine postnatally (Robles-Valdes et al., 1976).Medium-chain fatty acids also are a valuable probe in studying regulation at CPT I becausethey can diffuse across the mitochondrial membranes and be activated by an alternative acyl-CoA synthetase (fig. 1, enzyme 5; EC 6.2.1.2) located in the matrix (Groot et al., 1976).Thus, medium-chain fatty acids may bypass, in part, regulation via CPT I and be oxidizedindependently of carnitine. Medium-chain triglyceride utilization by neonatal pigs has beenstudied extensively in our laboratory and has been previously reviewed (Odle, 1997, 1998).

Recent advances in regulation of CPT have accompanied cloning of the genes for CPT Iand II (see McGarry and Brown, 1997, for review). Two isoforms of CPT I exist: the L-form(liver), which possesses relatively low sensitivity to malonyl-CoA inhibition, and the M-form(muscle), possessing very high sensitivity to malonyl-CoA. Both forms show high inter-species homology (>80%). In rats, hepatic concentrations of CPT I mRNA have been shownto increase markedly (up to 5-fold) within 24 h after birth, while CPT II expression was con-stitutive (Thumelin et al., 1994; Asins et al., 1995). Studying hepatocytes cultured fromperinatal rats, Chatelain et al. (1996) have shown that mRNA levels for CPT I (L) respondmarkedly and rapidly to in vitro supplementation with clofibrate, linoleate, and dibutyryl-cAMP, presumably through interaction with respective cis-acting elements and transcriptionfactors (e.g. PPAR-RXR, FFAR, and CREB, respectively).

3.3. Regulation of acetyl-CoA carboxylase (ACC)

Regulation of ACC plays a central role in controlling carbon flux through both anabolic andcatabolic pathways of fatty acid metabolism. As the rate-limiting step in de novo fatty acidbiosynthesis, and because of the allosteric influence of the product (malonyl-CoA) of thisenzyme on CPT I (described above), its regulation takes many forms including rapidallosteric and phosphorylation/dephosporylation mechanisms as well as longer-term mecha-nisms at the level of gene expression (see Kim, 1997, for review). Hormonal stimulation(e.g. glucagon) results in increased intracellular cAMP and causes rapid inactivation of ACCvia phosphorylation at multiple sites. Recent findings have suggested that different isozymes(designated α and β) of ACC, encoded for by different genes, may differentially regulate ana-bolic and catabolic carbon flux. Indeed, the recently cloned β form (Ha et al., 1996) has anadditional 150 amino acids at the N-terminus that may direct it to insertion into the mito-chondrial membrane where it may play a direct role in regulating CPT I. This may be ofparticular importance in tissues (e.g. piglet liver) in which fatty acid synthesis is negligibleand yet CPT I is highly sensitive to malonyl-CoA inhibition.

3.4. Regulation of mitochondiral HMG-CoA synthase (mHMGCS)

Since the establishment of the CPT I control theory, a growing body of evidence hasaccumulated suggesting other possible intramitochondrial regulatory sites, particularly inneonates (Pegorier et al., 1983; Escriva et al., 1986; Decaux et al., 1988). Much of the earlyevidence was indirect and speculative. However, researchers at Cambridge (Lowe and Tubbs,1985a,b,c; Quant et al., 1989, 1990, 1991, 1993) reported compelling evidence that control ofmHMGCS activity is likely. Specifically, they have shown that its activity is regulated by asuccinylation–desuccinylation mechanism. In the first step of its normal catalytic cycle,mHMGCS becomes acetylated at the active site by reaction with acetyl-CoA (its first sub-strate). They have shown that succinyl-CoA (at physiological concentrations) may also react

J. Odle et al.226

leading to competitive inhibition of the enzyme (Lowe and Tubbs, 1985c). Furthermore, var-ious in vivo treatments which stimulate ketogenesis (fasting, glucagon or mannoheptuloseinjection, alloxan diabetes, high-fat feeding, etc.) all increased mHMGCS activity by decreas-ing its degree of succinylation (Quant et al., 1989). The enzyme was 40−50% succinylated(and inactive) in the livers of normally fed rats and could be rapidly (within minutes) acti-vated in vitro. This led the authors to speculate that the succinylation-control mechanismcould allow for rapid changes in ketogenic flux rate in vivo.

More recently (Quant et al., 1991), the control of ketogenesis in the neonatal rat has beenshown to be mediated, in part, by changes in the amount and activity of mHMGCS, presumablyat the level of gene expression (Casals et al., 1992; see Hegardt, 1999, for review). Indeed,Thumelin et al. (1993) showed that hepatic mRNA concentrations for mHMGCS increased byabout 3-fold within 24 h of birth in rats, remained constant throughout suckling, and thenrapidly declined when animals were weaned onto a low-fat diet. Furthermore, mRNA incultured fetal hepatocytes increased by 4-fold within 4 h after exposure to glucagon. Additionalresearch (Ayte et al., 1993) has suggested that expression may be regulated in part bymethylation/demethylation of the 5′ flanking region of the gene and has identified CRE andC-EBP as potential cis regulatory elements (Brady et al., 1993; Gomez et al., 1993) which couldmediate the glucagon effects. Subsequent research also identified the PPAR-RXR diad as animportant regulator of expression induced by clofibrate and fatty acids (Rodriguez et al., 1994).

3.5. Hormonal support of fatty acid oxidation and ketogenesis

Regardless of the underlying biochemical regulatory mechanism(s), the major hormonalinfluence is most likely mediated by the insulin/glucagon ratio (McGarry and Foster, 1977).When the ratio is low, as observed during the neonatal period, fasting, or diabetes, ketogenesisis stimulated. The hormonal effect may be mediated through regulation of acetyl-CoA car-boxylase (Borthwick et al., 1986), thus affecting malonyl-CoA levels and CPT I activityand/or by decreasing succinyl-CoA levels and thereby activating mHMGCS (Quant et al.,1989) and/or by affecting levels of TCA cycle intermediates. Similarly, hormonal alterationof intracellular cAMP concentrations (Pegorier et al., 1989) may directly impact gene tran-scription (as described previously) via interaction with cAMP response elements.Furthermore, in vivo and in vitro exposure of rodent tissues to the peroxisome-proliferatinghypolipidemic drugs (e.g. clofibrate) and dehydroepiandrosterone (Brady et al., 1991) hasbeen shown to upregulate fatty acid oxidation and/or ketogenesis by increasing transcriptionof CPT and/or mHMGCS genes.

3.6. Hepatic lipid metabolism in the piglet

Clearly, a fundamental understanding of the developmental aspects of lipid metabolism isessential in order to optimize postnatal fat utilization by the piglet. Putative changes in meta-bolic capacity within the first week of life are of greatest concern because 75% of mortalityoccurs during this time period (USDA, 1991). Unfortunately, relatively little research hasbeen focused on these animals. The following review examines pig-specific studies.

3.7. Development of fatty acid oxidation in piglets

Postnatal increases in fatty acid oxidation have been reported (Wolfe et al., 1978), butmust be interpreted in light of the general increase in metabolic rate that occurs after birth


(Odle et al., 1991b). For example, the oxidation of [U-14C]palmitate to CO2 and acid-solubleproducts (considered to represent ketone bodies and/or TCA cycle intermediates) by liverhomogenates was reported to increase 4-fold between 0 and 7 days of age (Mersmann andPhinney, 1973). However, marked increases in mitochondrial respiration from several TCAcycle intermediates has likewise been reported (Mersmann et al., 1972) wherein oxygen con-sumption increased by 5-fold between 6 and 12 h postpartum. Subsequent histological worksuggested rapid mitochondrial proliferation during this early neonatal period. Odle et al.(1991b) have likewise observed developmental increases in hepatic fatty acid oxidation insmall and normal-birth-weight pigs during the first 48 h of life which could be largelyexplained by increases in oxygen consumption. Thus, increases in fatty acid oxidation post-natally may be accounted for, in part, by increased oxidative metabolism in general, and maynot necessarily infer an increased reliance upon fat as a fuel (Adams et al., 1997a).

3.8. Lack of ketogenesis in piglets

While piglets appear to display a hormonal profile (low insulin/glucagon) that would supportketogenesis (Pegorier et al., 1981), and have ample substrate from the fat present in milk, theydo not display hyperketonemia (Bengtsson et al., 1969; Pegorier et al., 1981), despite elevatedplasma nonesterified fatty acids (Adams and Odle, 1993b). This starkly contrasts with othermammalian species (e.g. rats, rabbits, etc.) which show pronounced hyperketonemia duringsuckling (Foster and Bailey, 1976) as well as ruminant species which under extreme lacta-tional stress can die from ketoacidosis. Because ketone bodies provide importantglucose-sparing carbon, aiding otherwise glucose-dependent tissues (e.g. neural tissues), theirabsence may be detrimental to the survival of the piglet, which is keenly susceptible to hypo-glycemia (Swiatek et al., 1968). Furthermore, insofar as fatty acid oxidation also is requiredto support active gluconeogenesis, impaired fat oxidation also could contribute indirectly tohypoglycemia (Lepine et al., 1993; Duee et al., 1996). In theory, low ketone concentrationscould be due to a low production rate (i.e. ketogenesis) and/or a high rate of utilization.

Using continuous-infusion isotope kinetics, we have observed limited β-OH-butyrateoxidation rates in vivo (Tetrick et al., 1995). Piglets were arterially infused with 14C-β-OH-butyrate at rates sufficient to supply 15%, 30%, 45%, and 60% of their estimated ATPturnover. Michaelis–Menten analysis of measured oxidation rates versus plasma concentra-tions showed that β-OH-butyrate could supply a maximum of 32% of the piglet’s total bodyATP turnover, but at physiological concentrations would supply <5% of the animal’s energyneed. More likely, an impaired rate of hepatic ketogenesis (Pegorier et al., 1983; Duee et al.,1994) is the cause. Comparative in vivo research (Adams and Odle, 1993b) showed that therelative ketogenic capacity (measured by regression of plasma β-OH-butyrate on plasmaoctanoate concentrations following intraperitoneal injection of octanoate) of neonatal pigs isgreatly attenuated (by up to 1−2 orders of magnitude) compared to weanling or mature swineand to neonatal or mature rabbits. Furthermore, using radio-HPLC to characterize products ofradiolabeled fatty acid oxidation, we have observed trivial accumulation of isotope in ketonebodies from piglet liver compared with neonatal rat liver preparations (Adams et al., 1997a).Concurrent with these findings, measurements of hepatic mHMGCS have shown 70% loweractivities in neonatal pigs than in neonatal rabbits (Adams and Odle, 1993a) or adult rats(Duee et al., 1994).

Following the cloning of the pig gene (Adams et al., 1997b), most recent findings con-firmed the attenuated expression of this enzyme (compared with rats) during suckling, butshowed that starvation led to increased mRNA concentrations. However, substantial increases

J. Odle et al.228

in mRNA did not occur until 2−3 weeks of age, indicating a developmental lag in inductioncompared with the rapid postnatal rise observed in the suckling rat (Thumelin et al., 1993).Subsequent examination of the upstream regulatory region of the pig gene has failed to iden-tify any idiosyncrasies that might impair expression in that it contains the anticipated PPARresponse element (Ortiz et al., 1999). Despite increased mRNA concentrations, enzyme activ-ity still remained low. Current research findings indicate that pig mRNA codes for acatalytically active protein, and low synthase activity is a result of attenuated translation.Furthermore, the decrease in mRNA translation was not a result of alteration of thepolyadenylate tail which can influence both mRNA stability and subsequent translation intoprotein (Barrero et al., 2001).

3.9. Regulation of fatty acid oxidation and ketogenesis in piglets

The degree to which the regulatory features of ketogenesis (reviewed previously) describedin other species extrapolates to the neonatal pig is not known. A putative limitation of fattyacid oxidation at the level of CPT I was suggested by data from Honeyfield and Froseth(1991), who reported a >10-fold increase in oxygen consumption when palmitoylcarnitinewas compared to palmitoyl-CoA in heart and liver homogenates from newborn pigs.Likewise, using hepatocytes isolated from piglets at birth and from piglets fed or fasted 24 hfrom birth, we (Odle et al., 1995) have obtained evidence consistent with the hypothesis thatCPT I is a potential regulatory site: (1) The molar oxidation rate of octanoate was 4 timeshigher than that of palmitate. This is 2-fold higher than can be explained by the molar energydifference between these fatty acids, and implies a putative limitation in the oxidation ofpalmitate compared to octanoate, given the measured oxygen consumption rates were similarin cells incubated with each fatty acid. While differences in the relative activities of themedium- and long-chain acyl-CoA synthetases cannot be excluded, this observation could beexplained by a limitation of palmitoyl-CoA transport into the mitochondrion via CPT.(2) Carnitine (a co-substrate for CPT) increased oxidation and decreased esterification ofpalmitate, but had no effect on octanoate metabolism. (3) TDGA (an irreversible inhibitor ofCPT I) reduced oxidation and increased esterification of palmitate, but again had no effect onoctanoate metabolism.

Developmental changes in liver and muscle CPT activity during the neonatal period havebeen reported (Bieber et al., 1973; Lin and Odle, 1995; Schmidt and Herpin, 1998). The activ-ity doubled in the liver of pigs between 0 and 1 day of age but then plateaued to equal theactivity in mitochondria from 24-day-old animals. Interestingly, palmitoyl-CoA oxidationcontinued to increase, doubling between 1 and 2 days postpartum. This suggests that some-thing other than CPT activity was limiting oxidation as the animals aged. Perhaps decreasesin the sensitivity of CPT to malonyl-CoA (Duee et al., 1994; Lin and Odle, 1995; Schmidtand Herpin, 1998) are responsible, as has been reported in rabbits (Prip-Buus et al., 1990).Indeed, Pegorier et al. (1983) challenged the idea of control mediated via CPT I. In contrastto our findings (Odle et al., 1991a,b, 1995; Lin et al., 1996), they reported low oxidation ratesfor octanoate (vs oleate). In addition, they were unable to increase oleate oxidation by iso-lated piglet hepatocytes incubated with glucagon (which should have decreased malonyl-CoAlevels). This may not be surprising because the liver is not a major site of lipogenesis in thepig (Mersmann et al., 1973). In general, malonyl-CoA concentrations are higher in lipogenictissues as the synthesis of malonyl-CoA is the first committed step of lipogenesis. The role ofmalonyl-CoA in nonlipogenic tissues, such as skeletal muscle, is regulation of CPT I activity.Thus, in nonlipogenic tissues the concentration of malonyl-CoA is much less than what is


observed in lipogenic tissues and CPT I exhibits greater sensitivity to malonyl-CoA. Recentcloning of the pig M- and L-CPT I isoforms has revealed that L-CPT I is a natural chimeraof rat M- and L-CPT I isoforms (Nicot et al., 2001). Pig L-CPT I shows similar kineticstoward carnitine as rat L-CPT I; however, IC50 for malonyl-CoA is similar to rat M-CPT I.Thus, pig L-CPT I has a greater sensitivity to malonyl-CoA than rat L-CPT I.

In addition to the potential control by CPT I, regulation may also shift to an intramito-chondrial site (e.g. mHMGCS). Research reported by Duee et al. (1994) verified the low rateof ketogenesis in mitochondria from 2-day-old piglets, and implicated an intramitochondrialconstraint; namely, when the mitochondria were treated with malonate (a blocker of succinatedehydrogenase) to decrease TCA cycle flux, the major end-product of palmitoylcarnitineoxidation was presumed to be acetoacetate. In this case, the rate of palmitoylcarnitine utiliza-tion was only 2.5 nmol/min/mg protein. However, when malonate was added, the end-productwas citrate and the rate of palmitoylcarnitine utilization increased by 3-fold.

3.10. Peroxisomal β-oxidation in piglets

We have characterized the postnatal development and tissue distribution of peroxisomalfatty acid oxidation in piglets (Yu et al., 1997a,b, 1998). In general, activity measured in liver,kidney, and heart as either antimycin/rotenone-insensitive palmitate oxidation or palmitoyl-CoA dependent KCN-insensitive reduction of NAD, was high (i.e. 40−50% of total fatβ-oxidation) compared with published rat values (e.g. 25%) and developed rapidly after birth.Hepatic activity of peroxisomal oxidation was increased when suckling-aged piglets werefasted, presumably due to elevated plasma free fatty acids and/or glucagon. Most recent datafrom our laboratory (Yu et al., 2001) showed that hepatic activities of fatty acid oxidase (anenzyme unique to peroxisomes) as well as CPT were dramatically induced in piglets fed milkreplacer containing clofibrate. Clofibrate is a hypolipidemic drug that is known to induceperoxisome (and possibly mitochondrial) biogenesis in rodents (Brady et al., 1991), throughinteraction with the peroxisome proliferator-activated receptor (PPAR). We hypothesize thatthe extra thermogenesis associated with peroxisomal β-oxidation may be important in thesuckling piglet’s maintenance of homeothermy.


Research findings to date indicate impairment in ketone body synthesis for the neonatal pigwhen compared to other species. This reduction in ketone bodies during a period when fattyacids are the main source of energy implies that the young pig is inefficient in utilizing dietaryfat, and thus may impact piglet survival and performance. In order to optimize nutrition forthe neonatal pig, an understanding of the underlying mechanisms of lipid metabolism mustbe elucidated. Further insight into the regulation of key lipid enzymes, including rate of syn-thesis and degradation of mRNA and protein, and subsequent protein activity is required.Recent research on CPT I and HMG-CoA synthase has shown key differences between theseenzymes in piglets when compared to the rat, a popular model in lipid research. These dif-ferences highlight the need for more research in the neonatal pig; only then will themechanisms responsible for lipid oxidation be defined so that improvements in lipid metab-olism can be made at a cellular level.

Research in the arena of molecular mechanisms of lipid oxidation will not provide overnightinsight into improving current practices for piglet survival and performance. However, asswine production becomes increasingly specialized, the demand for the formulation of milk

J. Odle et al.230

replacers that optimize early growth will increase also. The use of such milk replacers willallow dietary manipulation that cannot be achieved through nutritional management of thedam. Control over the piglet’s diet would allow for manipulation of the relative percent ofenergy contributed by carbohydrates versus fat and could have a profound impact on pigletmorbidity and mortality. The specific fatty acids added to the diet also could be altered toincrease the amount of MCT (for example), which serve as obligatory fuels. Furthermore,current research in the use of peroxisome proliferators indicate that their supplementation intothe diet may provide a useful tool in improving fatty acid utilization by the pig. Furtherresearch in these areas will hopefully assist in development of strategies to enhance fatty acidoxidation and ultimately improve piglet survival and growth.

REFERENCES

Aas, M., 1971. Organ and subcellular distribution of fatty acid activating enzymes in the rat. Biochim.Biophys. Acta 23, 32–47.

Adams, S.H., Odle, J., 1993a. Does HMG-CoA synthase play a role in limitation of ketogenesis in neona-tal pigs? FASEB J. 7, A379.

Adams, S.H., Odle, J., 1993b. Plasma β-hydroxybutyrate after octanoate challenge: attenuated ketogeniccapacity in neonatal swine. Amer. J. Physiol. 265, R761−R765.

Adams, S.H., Lin, X., Yu, X.X., Odle, J., Drackley, J.K., 1997a. Hepatic fatty acid metabolism in pigsand rats: major differences in end-products, O2 uptake and β-oxidation. Amer. J. Physiol. 272,R1641−R1646.

Adams, S.H., Alho, S.C., Asins, G., Hegardt, F.G., Marrero, P.F., 1997b. Gene expression of mitochon-drial 3-hydroxy-3-methylglutaryl-CoA synthase in a poorly ketogenic mammal: effect of starvationduring the neonatal period of the piglet. Biochem. J. 324, 65−73.

Asins, G., Serra, D., Arias, G., Hegardt, F.G., 1995. Developmental changes in carnitine palmitoyltrans-ferases I and II gene expression in intestine and liver of suckling rats. Biochem. J. 306, 379−384.

Ayte, J., Gomez, G.G., Hegardt, F.G., 1993. Structural characterization of the 3′ noncoding region of the geneencoding rat mitochondrial 3-hydroxy-3-methylglutaryl coenzyme A synthase. Gene 123, 267−270.

Bach, A.C., Babayan, V.K., 1982. Medium-chain triglycerides: an update. Amer. J. Clin. Nutr. 36,950−962.

Barrero, M.J., Alho, C.S., Oritz, J.A., Hegardt, F.G., Haro, D., Marrero, P.F., 2001. Low activity of mito-chondrial HMG-CoA synthase in liver of starved piglets is due to low levels of protein despite highmRNA levels. Arch. Biochem. Biophys. 385, 364−371.

Battaglia, F.C., Meschia, G., 1978. Principal substrates of fetal metabolism. Physiol. Rev. 58, 499.Bengtsson, N.J., Gentz, J., Hakkarainen, J., Hellstrom, R., Persson, B., 1969. Plasma levels of FFA, glycerol,

β-hydroxybutyrate and blood glucose during the postnatal development of the pig. J. Nutr. 97, 311−315.Bieber, L.L., Markwell, M.A.K., Blair, M., Helmrath, T.A., 1973. Studies on the development of carni-

tine palmitoyltransferase and fatty acid oxidation in liver mitochondria of neonatal pigs. Biochim.Biophys. Acta 326, 145−154.

Borthwick, A.C., Edgell, N.J., Denton, R.M., 1986. Mechanisms involved in the short-term regulation ofacetyl-CoA carboxylase by insulin. Biochem. Soc. Trans. 14, 563−565.

Borum, P.R., 1983. Carnitine. Annu. Rev. Nutr. 3, 233−259.Brady, L.J., Ramsay, R.R., Brady, P.S., 1991. Regulation of carnitine acyltransferase synthesis in lean and

obese zucker rats by dehydroepiandrosterone and clofibrate. J. Nutr. 121, 525−531.Brady, P.S., Barke, R.A., Brady, L.J., 1993. Regulation of the 68-kDa hepatic carnitine palmitoy-

ltransferase. In: Berdanier, C.D. (Ed.), Nutrition and Gene Expression. CRC Press, Boca Raton, FL,pp. 319−334.

Casals, N., Roca, N., Guerrero, M., Gil-Gomez, G., Ayte, J., Ciudad, C.J., Hegardt, F.G., 1992. Regulationof the expression of the mitochondrial 3-hydroxy-3-methylglutaryl-CoA synthase gene: its role in thecontrol of ketogenesis. Biochem. J. 283, 261−264.

Chatelain, F., Kohl, C., Esser, V., McGarry, J.D., Girard, J., Pegorier, J.P., 1996. Cyclic AMP and fattyacids increase carnitine palmitoyltransferase I gene transcription in cultured fetal rat hepatocytes.Eur. J. Biochem.235, 789−798.


Coleman, R.A., Lewin, T.M., Muoio, D.M., 2000. Physiological and nutritional regulation of enzymes oftriacylglycerol synthesis. Annu. Rev. Nutr. 20, 77−103.

Decaux, J.F., Robin, D., Robin, P., Ferre, P., Girard, J., 1988. Intramitochondrial factors controllinghepatic fatty acid oxidation at weaning in the rat. FEBS Lett. 232, 156−158.

Duee, P.-H., Pegorier, J.P., Quant, P.A., Herbin, C., Kohl, C., Girard, J., 1994. Hepatic ketogenesis innewborn pigs is limited by low mitochondrial 3-hydroxy-3-methylglutaryl-CoA synthase activity.Biochem. J. 298, 207−212.

Duee, P.-H., Pegorier, J.P., Darcy-Vrillon, B., Girard, J., 1996. Glucose and fatty acid metabolism in thenewborn pig. In: Tumbleson, M.E., Schook, L.B. (Eds.), Advances in Swine in Biomedical ResearchPlenum Press,New York, pp. 865−884.

Escriva, F., Ferre, P., Robin, D., Decaux, J.F., Girard, J., 1986. Evidence that the development of hepaticfatty acid oxidation at birth in the rat is concomitant with an increased intramitochondrial CoA con-centration. Eur. J. Biochem. 156, 603−607.

Ferre, P., Satabin, P., Decaux, J.F., Escriva, F., Girard, J., 1983. Development and regulation of ketogen-esis in hepatocytes isolated from newborn rats. Biochem. J. 214, 937−942.

Ferre, P., Decaux, J.F., Issad, T., Girard, J., 1986. Changes in energy metabolism during the suckling andweaning period in the newborn. Reprod. Nutr. Dev. 26, 619−631.

Foster, P.C., Bailey, E., 1976. Changes in hepatic fatty acid degradation and blood lipid and ketone bodycontent during development in the rat. Enzyme 21, 397−407.

Girard, J., Ferre, P., Pegorier, J.P., Duee, P.H., 1992. Adaptations of glucose and fatty acid metabolismduring the perinatal period and suckling-weaning transition. Physiol. Rev. 72, 507−562.

Gomez, G.G., Ayte, J., Hegardt, F.G., 1993. The rat mitochondrial 3-hydroxy-3-methylglutaryl-coenzyme-A-synthase gene contains elements that mediate its multihormonal regulation and tissuespecificity. Eur. J. Biochem. 213, 773−779.

Groot, P.H.E., Scholte, H.R., Hulsmann, W.C., 1976. Fatty acid activation: specificity, localization andfunction. Adv. Lipid Res. 14, 75−126.

Ha, J., Lee, J.K., Kim, D.S., Witters, L.A., Kim, K.H., 1996. Cloning of human acetyl-CoA carboxylase-βand its unique features. Proc. Natl. Acad. Sci. USA 93, 11466−11470.

Hegardt, F.G., 1999. Mitochondrial 3-hydroxy-3-methylglutaryl-CoA synthase: a control enzyme inketogenesis. Biochem. J. 338, 569−582.

Hettema, E.H., Tabak, H.F., 2000. Transport of fatty acids and metabolite across the peroxisomal mem-brane. Biochim. Biophys. Acta 1486, 18−27.

Honeyfield, D.C., Froseth, J.A., 1991. Evaluation of energy sources with and without carnitine in new-born pig heart and liver. J. Nutr. 121, 1117−1122.

Kerner, J., Froseth, J.A., Miller, E.R., Bieber, L.L., 1984. A study of the acylcarnitine content of sows’colostrum, milk and serum and newborn piglet tissues: demonstration of high amounts of isovaleryl-carnitine in colostrum and milk. J. Nutr. 114, 854−861.

Kim, K.H., 1997. Regulation of mammalian acetyl-coenzyme A carboxylase. Annu. Rev. Nutr. 17, 77−99.Kondrup, J., Lazarow, P.B., 1985. Flux of palmitate through the peroxisomal and mitochondrial β-oxidation

systems in isolated rat hepatocytes. Biochim. Biophys. Acta 835, 147−153.Latruffe, N., Cherkaoui Malki, M., Nicolas-Frances, V., Jannin, B., Clemencet, M.-C., Hansmannel, F,

Passilly-Degrace, P., Berlot, J.-P., 2001. Peroxisome-proliferator-activated receptors as physiological sen-sors of fatty acid metabolism: molecular regulation in peroxisomes. Biochem. Soc. Trans. 29, 305−309.

Lazo, O., Contreras, M., Singh, I., 1990. Topographical localization of peroxisomal acyl-CoA ligases: dif-ferential localization of palmitoyl-CoA and lignoceroyl-CoA ligases. Biochemistry 29, 3981−3986.

Lepine, A.J., Watford, M., Boyd, R.D., Ross, D.A., Whitehead, D.M., 1993. Relationship betweenhepatic fatty acid oxidation and gluconeogenesis in the fasting neonatal pig. Brit. J. Nutr. 70, 81−91.

Lewin, T.M., Kim, J.H., Granger, D.A., Vance, J.E., Coleman, R.A., 2001. Acyl-CoA synthetase isoforms1, 4, and 5 are present in different subcellular membranes in rat liver and can be inhibited independently.J. Biol. Chem. 276, 24674−24679.

Lin, X., Odle, J., 1995. Regulation of fatty acid oxidation by hepatic mitochondria from neonatal pigs viacarnitine palmitoyltransferase I. J. Anim. Sci. 73, Suppl. 1, 77.

Lin, X., Adams, S.H., Odle, J., 1996. Acetate represents a major product of heptanoate and octanoateβ-oxidation in hepatocytes isolated from neonatal piglets. Biochem. J. 318, 235−240.

Lowe, D.M., Tubbs, P.K., 1985a. 3-Hydroxy-3-methylglutaryl-coenzyme A synthase from ox liver:purification, molecular and catalytic properties. Biochem. J. 227, 591−599.

Lowe, D.M., Tubbs, P.K., 1985b. 3-Hydroxy-3-methylglutaryl-coenzyme A synthase from ox liver:properties of its acetyl derivative. Biochem. J. 227, 601−607.

J. Odle et al.232

Lowe, D.M., Tubbs, P.K., 1985c. Succinylation and inactivation of 3-hydroxy-3-methylglutaryl-CoA synthaseby succinyl-CoA and its possible relevance to the control of ketogenesis. Biochem. J. 232, 37−42.

McGarry, J.D., Brown, N.F., 1997. The mitochondrial carnitine palmitoyltransferase system: from conceptto molecular analysis. Eur. J. Biochem. 244, 1−14.

McGarry, J.D., Foster, D.W., 1977. Hormonal control of ketogenesis. Arch. Intern. Med. 137, 495−501.McGarry, J.D., Foster, D.W., 1980. Regulation of hepatic fatty acid oxidation and ketone body production.

Annu. Rev. Biochem. 49, 395−420.Mersmann, H.J., Phinney, G., 1973. In vitro fatty acid oxidation in liver and heart from neonatal swine

(Sus domesticus). Comp. Biochem. Physiol. 44B, 219−223.Mersmann, H.J., Goodman, J., Houk, J.M., Anderson, S., 1972. Studies on the biochemistry of mito-

chondrial and cell morphology in the neonatal swine hepatocyte. J. Cell. Biol. 53D, 335−347.Mersmann, H.J., Phinney, G., Sanguinetti, M.C., Houk, J.M., 1973. Lipogenic capacity of liver from peri-

natal swine (Sus domesticus). Comp. Biochem. Physiol. 46B, 493−497.Nicot, C., Hegardt, F.G., Woldegiorgis, G., Haro, D., Marrero, P.F., 2001. Pig liver palmitoyltransferase I,

with low Km for carnitine and high sensitivity to malonyl-CoA inhibition, is a natural chimera of ratliver and muscle enzymes. Biochemistry 40, 2260−2266.

Odle, J., 1997. New insights into medium-chain triglyceride utilization by the neonate: observations froma piglet model. J. Nutr. 127, 1061−1067.

Odle, J., 1998. Medium-chain triglycerides: a unique energy source for neonatal pigs. Pig News 20, 25−32.Odle, J., Benevenga, N.J., Crenshaw, T.D., 1991a. Utilization of medium-chain triglycerides by neonatal

piglets: chain length of even- and odd-carbon fatty acids and apparent digestion/absorption andhepatic metabolism. J. Nutr. 121, 605−614.

Odle, J., Benevenga, N.J., Crenshaw, T.D., 1991b. Postnatal age and the metabolism of medium- andlong-chain fatty acids by isolated hepatocytes from small-for-gestational-age and appropriate-for-gestational-age piglets. J. Nutr. 121, 615−621.

Odle, J., Lin, X., van Kempen, T.A.T.G., Drackley, J.K., Adams, S.H., 1995. Carnitine palmitoyltrans-ferase modulation of hepatic fatty acid metabolism and radio-HPLC evidence for low ketogenesis inneonatal pigs. J. Nutr. 125, 2541−2549.

Oikawa, E., Iijima, H., Suzuki, T., Sasano, H., Sato, H., Kamataki, A., Nagura, H., Kang, M., Fujino, T.,Suzuki, H., Yamamoto, T.T., 1998. A novel acyl-CoA synthetase, ACS5, expressed in intestinalepithelial cells and proliferating preadipocytes. J. Biochem. 124, 679−685.

Ortiz, J.A., Mallolas, J., Nicot, C., Bofarull, J., Rodriguez, J.C., Hegardt, F.G., Haro, D., Marrero, P.F.,1999. Isolation of pig mitochondrial 3-hydroxy-3-methylglutaryl-CoA synthase gene promoter: char-acterization of a peroxisome proliferator-responsive element. Biochem. J. 337, 329−335.

Pegorier, J.P., Duee, P.H., Peret, J., Girard, J., 1981. Changes in circulating fuels, pancreatic hormonesand liver glycogen concentration in fasting or suckling newborn pigs. J. Dev. Physiol. 3, 203−217.

Pegorier, J.P., Duee, P.H., Girard, J., Peret, J., 1983. Metabolic fate of non-esterified fatty acids inisolated hepatocytes from newborn and young pigs: evidence for a limited capacity for oxidation andincreased capacity for esterification. Biochem. J. 212, 93−97.

Pegorier, J.P., Garcia-Garcia, M.-V., Prip-Buus, C., Duee, P.-H., Kohl, C., Girard., J., 1989. Induction ofketogenesis and fatty acid oxidation by glucagon and cyclic AMP in cultured hepatocytes from rabbitfetuses. Biochem. J. 264, 93−100.

Prip-Buus, C., Pegorier, J., Duee, P., Kohl, C., Girard, J., 1990. Evidence that the sensitivity of carnitinepalmitoyltransferase I to inhibition by malonyl-CoA is an important site of regulation of hepatic fattyacid oxidation in the fetal and newborn rabbit. Biochem. J. 269, 409−415.

Quant, P.A., Tubbs, P.K., Brand, M.D., 1989. Treatment of rats with glucagon or mannoheptuloseincreases mitochondrial 3-hydroxy-3-methylglutaryl-CoA synthase activity and decreases succinyl-CoA content in liver. Biochem. J. 262, 159−164.

Quant, P.A., Tubbs, P.K., Brand, M.D., 1990. Glucagon activates mitochondrial 3-hydroxy-3-methylglu-taryl-CoA synthase in vivo by decreasing the extent of succinylation of the enzyme. Eur. J. Biochem.187, 169−174.

Quant, P.A., Robin, D., Robin, P., Ferre, P., Brand, M.D., Girard, J., 1991. Control of hepatic mitochon-drial 3-hydroxy-3-methylglutaryl-CoA synthase during the foetal/neonatal transition, suckling andweaning in the rat. Eur. J. Biochem. 195, 449−454.

Quant, P.A., Robin, D., Robin, P., Girard J., Brand, M.D., 1993. A top-down control analysis in isolatedrat liver mitochondria: can the 3-hydroxy-3-methylglutaryl-CoA pathway be rate-controlling for keto-genesis? Biochim. Biophys. Acta 1156, 135−143.

Reddy, J.K., Mannaerts, G.P., 1994. Peroxisomal lipid metabolism. Annu. Rev. Nutr. 14, 343−370.


Robles-Valdes, C., McGarry, J.D., Foster, D.W., 1976. Maternal-fetal carnitine relationships and neona-tal ketosis in the rat. J. Biol. Chem. 251, 6007−6012.

Rodríguez, J.C., Gomez, G.G., Hegardt, F.G., Haro, D., 1994. Peroxisome proliferator-activated receptormediates induction of the mitochondrial 3-hydroxy-3-methylglutaryl-CoA synthase gene by fattyacids. J. Biol. Chem. 269, 18767−18772.

Rohde Parfet, K.A., Gonyou, H.W., 1988. Effect of creep partitions on teat-seeking behavior of newbornpiglets. J. Anim. Sci. 66, 2165.

Schmidt, I., Herpin, P., 1998. Carnitine palmitoyltransferase I (CPTI) activity and its regulation by mal-onyl-CoA are modulated by age and cold exposure in skeletal muscle mitochondria from newbornpigs. J. Nutr. 128, 886−893.

Stanier, M.W., Mount, L.E., Bligh, J., 1984. Energy Balance and Temperature Regulation. CambridgeUniversity Press, New York.

Swiatek, K.R., Kipnis, D.M., Mason, G., Chao, K., Cornblath, M., 1968. Starvation hypoglycemia innewborn pigs. Amer. J. Physiol. 214, 400−405.

Tetrick, M.A., Adams, S.H., Odle, J., Benevenga, N.J., 1995. Contribution of D-(−)-3-hydroxybutyrate tothe energy expenditure of neonatal pigs. J. Nutr. 125, 264−272.

Thumelin, S., Forestier, M., Girard, J., Pegorier, J.P., 1993. Developmental changes in mitochondrial3-hydroxy-3-methylglutaryl-CoA synthase gene expression in rat liver, intestine and kidney. Biochem. J.292, 493−496.

Thumelin, S., Esser, V., Charvy, D., Kolodziej, M.P., Zammit, V.A., McGarry, J.D., Girard, J., Pegorier, J.P.,1994. Expression of liver carnitine palmitoyltransferase I and II genes during development in the rat.Biochem. J. 300, 583−587.

USDA, 1991. National swine survey: morbidity/mortality and health management of swine in the U.S.National Animal Health Monitoring System APHIS/U.S. Department of Agriculture, Washington, DC.

USDA, 2002. Reference of swine health and environmental management in the United States, 2000, Part III.U.S. Department of Agriculture/APHIS/VS/CEAH, National, Fort Collins, CO, No. N361.0902.

Van Veldhoven, P.P., Casteels, M., Mannaerts, G.P., Baes, M., 2001. Further insights into peroxisomallipid breakdown via α- and β-oxidation. Biochem. Soc. Trans. 29, 292−297.

Verhoeven, N.M., Roe, D.S., Kok, R.M., Wanders, R.J., Jakobs, C., Roe, C.R., 1998. Phytanic acid andpristanic acid are oxidized by sequential peroxisomal and mitochondrial reactions in cultured fibro-blasts. J. Lipid Res. 39, 66−74.

Wanders, R.J., Tager, J.M., 1998. Lipid metabolism in peroxisomes in relation to human disease. Mol.Aspects Med. 19, 69−154.

Wanders, R.J.A., Vreken, P., Ferdinandusse, S., Jansen, G.A., Waterham, H.R., van Roermund, C.W.T.,van Grunsven, E.G., 2001. Peroxisomal fatty acid α- and β-oxidation in humans: enzymology,peroxisomal metabolite transporters and peroxisomal diseases. Biochem. Soc. Trans. 29, 250−267.

Williamson, D.H., Bates, M.W., Krebs, H.A., 1968. Activity and intracellular distribution of enzymes ofketone-body metabolism in rat liver. Biochem. J. 108, 353−361.

Wolfe, R.G., Maxwell, C.V., Nelson, E.C., 1978. Effect of age and dietary fat level on fatty acid oxida-tion in the neonatal pig. J. Nutr. 108, 1621−1634.

Yu, X.X., Drackley, J.K., Odle, J., Lin, X., 1997a. Response of hepatic mitochondrial and peroxisomalβ-oxidation to increasing palmitate concentrations in piglets. Biol. Neonate 72, 284−292.

Yu, X.X., Drackley, J.K., Odle, J., 1997b. Rates of mitochondrial and peroxisomal β-oxidation of palmitatechange during postnatal development and feed deprivation in liver, kidney and heart of pigs. J. Nutr.127, 1814−1821.

Yu, X.X., Drackley, J.K., Odle, J., 1998. Food deprivation changes peroxisomal β-oxidation activity butnot catalase activity during postnatal development in pig tissues. J. Nutr. 128, 1114−1121.

Yu, X.X., Drackley, J.K., Odle, J., 2001. Differential induction of peroxisomal β-oxidation enzymes byclofibric acid and aspirin in piglet tissues. Amer. J. Physiol. Regul. Integr. Comp. Physiol. 281,R1553−R1561.

J. Odle et al.234

235

10 Essential fatty acid metabolism duringearly development

S. M. Innis

Department of Paediatrics, University of British Columbia, Vancouver,British Columbia, Canada V5Z 4H4

The n-6 and n-3 polyunsaturated fatty acids are essential nutrients that are required for growthand normal cell function. These fatty acids are present in cells as the acyl moieties of phos-pholipids which make up the structural matrix of cell and subcellular membranes, andfunction directly, or as precursors to other molecules that modulate cell growth, metabolism,inter- and intracellular communication and gene expression. The n-3 fatty acid docosahexaenoicacid (22:6n-3) is accumulated in the retina and brain grey matter during development, but iscontinually turned over, recycled and replenished by uptake from plasma during the dynamicprocesses of signal transduction in the retina and neuronal membranes. Depletion of 22:6n-3from retinal and neural membranes results in reduced visual function, behavioural abnormal-ities, and alterations in the metabolism of neurotransmitters, and in membrane proteins,receptors and ion channel activities. Large gaps still exist in understanding of dietary require-mements for n-3 fatty acids at different stages of the life cycle, species differences in essentialfatty acid metabolism, and the process that controls the partitioning of n-3 fatty acids forgeneration of energy and further metabolism for incorporation into membrane lipids.

1. OVERVIEW OF ESSENTIALITY OF n-6 AND n-3 FATTY ACIDS

Studies in animals over 70 years ago provided the first demonstration that dietary fat containscomponents that are essential to normal growth and development. The signs of deficiency,which included scaly skin, growth retardation, reproductive failure and histological abnor-malities (table 1), were ascribed to the absence of n-6 fatty acids, and were reversed orprevented by feeding the n-6 fatty acid linoleic acid (18:2n-6) (Innis, 1991). The importanceof a second class of polyunsaturated fatty acids, the n-3 fatty acids, did not emerge until the1970s when altered electroretinograph (ERG) recordings and behaviour were found in ratsfed diets deficient in n-3 fatty acids (Benolken et al., 1973; Wheeler et al., 1975; Lamptey andWalker, 1976). Despite the knowledge that the n-3 fatty acid docosahexaenoic acid (22:6n-3)



represents a major proportion of the polyunsaturated fatty acids in brain and retina (Flieslerand Anderson, 1983; Sastry, 1985; Giusto et al., 2000), the acceptance of the essential role ofn-3 fatty acids, particularly in human nutrition, grew only slowly, perhaps because of theabsence of overt effects of n-3 fatty acid deficiency on growth or health. The formation andextension of neural and retinal membranes, however, requires a large amount of 22:6n-3, allof which must be derived from n-3 fatty acids in the diet. It is well appreciated that deficiencyof one or more key nutrients during brain development can, depending on the timing, sever-ity and duration, decrease cell division, dendritic arborization and myelination with resultantlong-term effects on cognitive and behavioural functions (Dobbing et al., 1971; Dobbing,1972; Wiggins, 1986; Morgane et al., 1992; Levitsky and Strupp, 1995). Because of this,much of the recent interest on essential fatty acids in human growth and development hasfocused on the requirements for n-3 fatty acids for visual and neural function. This chapterreviews current research on polyunsaturated fatty acid metabolism in development, the supplyof n-6 and n-3 fatty acids before and after birth, and the role of n-3 fatty acids in the devel-oping brain and retina.

2. ESSENTIAL FATTY ACID STRUCTURE AND METABOLISM

2.1. Structure and nomenclature

Fatty acids are identified using a systematic nomenclature which identifies the fatty acid bythe number of carbon atoms, and the number and position of any unsaturated double bondsrelative to the carboxyl group (table 2, fig. 1). More commonly, fatty acids are referredto using a structural designation, or trivial name. The structural designation describes afatty acid by the number of carbons, the number of double bonds, and the position of thefirst double bond from the methyl terminal carbon which is designated by the Greek letteromega (ω) or “n”. The delta (Δ) notation is used to designate the position of a carbon fromthe carboxyl terminus and is used to denote the site of action of the fatty acid desaturaseenzymes. Polyunsaturated fatty acids, which are often referred to as PUFA, are grouped intofamilies based on the position of the first methylene interrupted double bond from the methyl(ω or n) end.

S. M. Innis236

Table 1

Signs of n-6 and n-3 fatty acid deficiency

Essential fatty acid/n-6 fatty acid deficiency n-3 fatty acid deficiency

Decreased growth Reduced electroretinogram A and/or B wave recordingScaly dermatoses Reduced visual acuityInflamed skin Reduced discrimination learningDecreased skin pigmentation Reduced exploratory behaviourIncreased transepidermal water loss Increased stereotyped behaviourImpaired wound healingAlopeciaTail necrosisFatty liverKidney degenerationReproductive failure

2.2. Metabolism

The Δ12- and Δ15-desaturase enzymes necessary to form n-6 and n-3 fatty acids, respectively,are present in plant but not in animal cells. Consequently, animals require a dietary source ofn-6 and n-3 fatty acids, all of which are ultimately derived from plants either directly, or aftertransfer up the food chain. The parent 18-carbon n-6 and n-3 fatty acids are linoleic acid(18:2n-6) and α-linolenic acid (18:3n-3). These fatty acids are considered essential dietarynutrients for most animals.

Linoleic acid (18:2n-6) and 18:3n-3 can be further metabolized by desaturation and elon-gation, as well as chain shortening, but cannot be interconverted (Innis, 1991) (fig. 2). Themajor site of desaturation and elongation is the liver, although desaturase activity is also presentin some other cells. Desaturation and elongation proceed alternately, by Δ6-desaturation,elongation and Δ5-desaturation, to yield arachidonic acid (20:4n-6) and eicosapentaenoic acid(20:5n-3) from 18:2n-6 and 18:3n-3, respectively (fig. 2). These metabolites of 18:2n-6 and18:3n-3 with 20 or more carbon atoms are often referred to as long-chain polyunsaturatedfatty acids, and abbreviated as LC-PUFA. For many years it was believed that docosapentaenoicacid (22:5n-3) was desaturated by a microsomal Δ4-desaturase to form docosahexaenoic acid(22:6n-3). Such a Δ4-desaturase, however, has not been found in animal cells (Moore et al., 1991;Voss et al., 1991; Mohammed et al., 1995, 1997; Luthria et al., 1996). The pathway for

Essential fatty acid metabolism during early development 237

Table 2

Systematic names and abbreviations for essential n-6 and n-3 fatty acids

Systematic name Abbreviation Trivial name

n-6 family9,12-octadecadienoic 18:2n-6 Linoleic6,9,12-octadecatroenic 18:3n-6 γ-Linolenic8,11,14-eicosatrienoic 20:3n-6 Dihomo-γ-linolenic5,8,11,14-eicosatetraenoic 20:4n-6 Arachidonic7,10,13,16-docosatetraenoic 22:4n-6 Adrenic4,7,10,13,16-docosapentaenoic 22:5n-6

n-3 family9,12,15-octadecatrienoic 18:3n-3 α-Linolenic6,9,12,15-octadecatetraenoic 18:4n-3 Stearidonic5,8,11,14,17-eicosapentaenoic 20:5n-34,7,10,13,16,19-docosahexaenoic 22:6n-36,9,12,15,18,21-tetracosahexanenoic 24:6n-3 Nisinic

Fig. 1. Schematic representation of a fatty acid.

synthesis of 22:6n-3 is now known to involve synthesis of a 24-carbon n-3 fatty acid by twosuccessive elongations of 20:5n-3 to form 24:5n-3, which is then desaturated at position 6 toform tetracosahexaenoic acid (24:6n-3), and translocated to the peroxisomes where one cycleof β-oxidation leads to formation of 22:6n-3 (Sprecher et al., 1995, 1999; Ferdinandusse et al.,2001). The 22:6n-3 thus formed is then shuttled back to the endoplasmic reticulum, which isthe major site of phospholipid biosynthesis (Vance, 1990; Choy, 1997). Synthesis of then-6 docosapentaenoic acid (22:5n-6) from 20:4n-6 occurs in an analogous pathway (Sprecheret al., 1995, 1999). The regulation and steps involved in the intracellular trafficking of fattyacids between the endoplasmic reticulum and peroxisomes is still incompletely understood.

The desaturase enzymes necessary for synthesis of 20:4n-6 and 22:6n-3 are present inanimal but not in plant cells. Thus, omnivores and carnivores obtain 20:4n-6 and 22:6n-3from the flesh of other animals, while herbivores consume only the precursors 18:2n-6 and18:3n-3. The activity of the essential fatty acid desaturation and elongation pathway appearsto vary markedly among species, being higher in rodents and low in pigs and humans(Pooviah et al., 1976; Yamazaki et al., 1992; Emken et al., 1994; Li et al., 2000; Pawlosky et al.,2001). Rats, for example, have very high amounts of 22:6n-3 in liver phosphatidylethanolamine(PE) when compared to humans and other primates (table 3). The felinae, which are obligatecarnivores, have very low activity of Δ6-desaturase, and thus require a dietary source of

S. M. Innis238

Fig. 2. Schematic representation of the major pathways of essential fatty acid desaturation and elongation.

Tabl

e 3

Ess

enti

al n

-6 a

nd n

-3 f

atty

aci

ds in

bra

in g

rey

mat

ter

and

liver

pho

spha

tidy

leth

anol

amin

e of

dif

fere

nt s

peci

es

Bra

inL

iver

18:2

n-6

20:4

n-6

22:4

n-6

22:5

n-6

22:6

n-3

18:2

n-6

20:4

n-6

22:4

n-6

22:5

n-6

22:6

n-3

Rat

0.3

115.

10.

621

6.0

210.

60.

422

Gui

nea

pig

1.6

177.

92.

419

2311

1.3

0.9

21H

amst

er1.

59.

64.

20.

222

1116

0.3

0.5

10L

emm

ing

1.3

104.

31.

529

167.

60.

30.

923

Bat

0.9

155.

22.

521

811

0.6

0.2

3.9

Vol

e0.

511

1.1

0.5

207.

25.

50.

20.

126

War

thog

0.9

126.

31.

223

2212

0.6

0.2

1.6

Hor

se1.

012

7.1

1.3

1917

6.2

0.6

0.4

3.6

Zeb

ra0.

110

4.9

1847

4.2

0.1

0.1

0.2

Ele

phan

t1.

111

7.2

2.1

2510

100.

40.

10.

6H

arte

bees

t1.

511

7.2

0.0

1711

140.

70.

30.

9R

ed d

eer

0.8

106.

91.

225

5.1

110.

10.

10.

3O

x1.

211

6.3

0.8

224.

016

2.3

0.6

0.3

Buf

falo

2.3

126.

51.

116

9.7

130.

30.

12.

3E

land

1.7

147.

01.

221

8.5

140.

60.

21.

8G

iraf

fe1.

414

6.2

0.9

2412

121.

10.

80.

8C

at0.

913

6.7

1.1

275.

518

1.3

2.0

22C

ivet

1.2

145.

91.

326

3.4

153.

61.

220

Leo

pard

0.7

147.

11.

022

3.8

152.

40.

718

Mar

mos

et0.

811

4.7

1.6

2910

230.

71.

38.

2V

erve

t0.

16.

85.

31.

221

9.2

170.

30.

45.

9M

an0.

314

6.0

2.2

236.

316

3.5

2.3

7.7

Dol

phin

0.7

6.9

3.6

0.2

271.

730

0.9

0.9

11

Ada

pted

fro

m C

raw

ford

et a

l. (1

976)

.

20:4n-6 and 22:6n-3 (Rivers et al., 1975, 1976; MacDonald et al., 1983; Pawlosky et al.,1997), which is consumed as part of their natural meat diet.

Many of the essential roles of n-6 and n-3 fatty acids are fulfilled by 20:4n-6 and 20:5n-3and 22:6n-3, rather than their precursors 18:2n-6 and 18:3n-3, respectively. Linoleic acid(18:2n-6) and metabolites in the pathway to 20:4n-6, however, have important roles in cho-lesterol metabolism and in the skin, and 20:3n-6 is the precursor for synthesis of series 1eicosanoids (Wertz et al., 1987; Ziboh and Chapkin, 1988; Innis, 1996). α-Linolenic acid, onthe other hand, is not known to have any essential functions other than as a precursor for syn-thesis of 20:5n-3 and 22:6n-3. Acetyl-CoA derived from β-oxidation of 18:3n-3 is extensivelyrecycled for synthesis of saturated and monounsaturated fatty acids and cholesterol synthesis indeveloping tissues (Cunnane et al., 1994, 1999; Sheaff-Greiner et al., 1996; Menard et al., 1998).

Desaturation of 18:2n-6 and 18:3n-3 is believed to involve the same enzymes. In vitro, theΔ6-desaturase enzyme shows clear preference in order 18:3n-3 > 18:2n-6 > 18:1n-9 (Brennerand Peluffo, 1966, 1969; Brenner et al., 1969). Owing to the abundance of 18:2n-6 in com-monly used vegetable oils, such as safflower, corn and soybean oil (Chow, 2000), human dietsin many Western countries have much higher 18:2n-6 than 18:3n-3 (Simopoulos, 1999). Thehigh proportion of 18:2n-6 to 18:3n-3 in the diet has implications for reducing synthesis of22:6n-3, which is important for brain and visual development, and for increasing the risk ofhealth problems associated with increased production of 20:4n-6-derived eicosanoids.Synthesis of 22:6n-3 and 20:4n-6 is also reduced by products of the same and opposing seriesof fatty acids. For example, high intakes of 20:5n-3 or 22:6n-3 from fish or fish oil reducestissue 20:4n-6, decreases the synthesis of n-6 fatty acid-derived eicosanoids, and increases thesynthesis of eicosanoids derived from 20:5n-3 (Fischer et al., 1989; Broughton and Morgan,1994; Ferretti et al., 1998; Broughton and Wade, 2002). Changes in the balance of n-6 andn-3 fatty acid-derived eicosanoids can have important effects on inflammation and immunity,hemostatic and endothelial function, and reproductive functions including ovulation rate,progesterone production by the corpus luteum, timing of luteolysis and gestational length(von Schacky and Weber, 1985; Rogers et al., 1987; Kristensen et al., 1989; Tremoli et al.,1995; Calder, 1998, 2001; Abayasekara and Wathes, 1999).

β-Oxidation of the 18-carbon essential fatty acids in the mitochondria depends on carnitine-dependent translocation, and leads to generation of acetyl-CoA which then enters the tricar-boxylic acid cycle. The first and rate-limiting step of β-oxidation of the longer-chain n-6 andn-3 fatty acids in the peroxisomes is catalysed by straight-chain fatty acyl-CoA oxidase andgenerates hydrogen peroxide (Wanders et al., 2001). In a similar process, 22:6n-3 and 22:4n-6can be retroconverted to 20:5n-3 and 20:4n-6, respectively, thereby maintaining tissue poolsof these fatty acids.

In the absence of a dietary supply of n-6 and n-3 fatty acids, oleic acid (18:1n-9) derivedfrom the diet or synthesized de novo from acetyl-CoA, undergoes Δ6- and Δ5-desaturationand elongation to form eicosatrienoic acid (20:3n-9), and concentrations of 20:4n-6 decreasedue to the absence of 18:2n-6 (Innis, 1991, 1996). The usual biochemical method for estab-lishing essential fatty acid deficiency is to calculate the ratio of 20:3n-9 to 20:4n-6. This iscommonly referred to as the triene to tetraene ratio. An increase in the ratio of plasma 20:3n 9to 20:4n-6 to >0.2 is considered to indicate essential fatty acid deficiency in humans (Holmanet al., 1991; Jeppensen et al., 1998). Dietary deficiency of n-3 fatty acids results in decreased22:6n-3 and increased desaturation of n-6 series fatty acids, leading to increased 22:4n-6 and22:5n-3 in brain and retinal membranes (Bourre et al., 1984; Innis, 1991). Refeeding deficientdeveloping and adult animals with 18:3n-3 results in recovery of neural cell membrane 22:6n-3, although the rate of recovery may be slower in the central nervous system than in

S. M. Innis240

other organs (Youyou et al., 1986; Bourre et al., 1989a; Connor et al., 1990; Weisinger et al.,1999; Moriguchi et al., 2001). Desaturation and elongation of n-9 fatty acids and the triene/tetraene ratio is normal unless there is a concomitant deficiency of n-6 fatty acids. The com-pensatory increased desaturation and elongation of n-6 fatty acids in n-3 fatty acid-deficientanimals results in maintenance of the normal total n-6 + n-3 polyunsaturated fatty acids in thebrain (Galli et al., 1971; Neuringer et al., 1986; Hrboticky et al., 1990), although this may notbe so for all neural membrane phospholipids (Murthy et al., 2002).

3. FUNCTIONAL ROLES OF ESSENTIAL FATTY ACIDS

Long-chain polyunsaturated fatty acids (those with 20 or more carbon chains and 3 or moredouble bonds) are found predominately in phospholipids in which they form the hydrophobiccore of all cell and subcellular membranes. Linoleic acid (18:2n-6) is also present in mem-brane phospholipids, adipose tissue triglycerides, plasma cholesterol esters, and in thespecialized lipids of the skin. Reviews of the role of n-6 fatty acids in maintaining the normalepithelial cell–water barrier are available (Ziboh and Chapkin, 1988). The n-6 and n-3 fattyacids, 20:3n-6, 20:4n-6 and 20:5n-3 are precursors for prostaglandins, hydroxy fatty acids,leukotrienes and lipoxins, often collectively referred to as eicosanoids. These oxygenatedmetabolites, which are formed via cyclo-oxygenase and lipoxygenase, are synthesized fol-lowing a stimulus and act locally as autocoids, often initiating a cascade of events. In general,metabolites formed from 20:5n-3 have weaker or opposing effects than the metabolitesformed from 20:4n-6. Several reviews on eicosanoid metabolism have been published(Fischer, 1989; Kinsella and Lokesh, 1990; Funk, 2001).

The importance of n-6 and n-3 fatty acids in metabolic and physiological processes can besummarized into three general mechanisms: the fatty acid moieties of membrane phospho-lipids contribute to the physical properties of the membrane bilayer, with secondary effectson the activity of membrane-associated proteins, receptors and ion channels; n-6 and n-3 fattyacids are precursors for generation of membrane-derived signal molecules, as well aseicosanoids; and n-6 and n-3 fatty acids have direct effects on gene expression. Docosahexaenoicacid, (22:6n-3), unlike the n-6 fatty acids, has a highly specific distribution in tissues andphospholipids. Concentrations of 22:6n-3 are particularly high in the amino phospholipids,phosphatidylserine (PS) and phosphatidylethanolamine (PE) of neural grey matter, and in theouter segments of rod and cone photoreceptors in the retina (Fliesler and Anderson, 1983;Sastry, 1985; Giusto et al., 2000). Large amounts of 22:6n-3 are also present in specific phos-pholipids in the heart and sarcoplasmic reticulum of skeletal muscle sarcolemma, and insperm (Fiehn and Pewter, 1971; Poulos et al., 1973; Gudbjarnason et al., 1978; Charnock et al.,1983; Ollero et al., 2000). Severe restriction of n-3 fatty acids throughout gestation, lactationand postnatal development, when there is a need for new tissue synthesis, results in reducedtissue 22:6n-3, decreased visual function, decreased performance on discrimination learningtasks, and increased stereotyped behaviour in rodents and non-human primates (Benolken et al.,1973; Wheeler et al., 1975; Neuringer et al., 1984, 1986; Yamamoto et al., 1988; Bourre et al.,1989b; Reisbick et al., 1990, 1994; Frances et al., 1996a,b; Okada et al., 1996; Gamoh et al., 1999;Moriguchi et al., 2000; Greiner et al., 2001) (table 1). In adult animals, 22:6n-3 is aggres-sively retained, even during longstanding and severe dietary n-3 fatty acid restriction (Tinocoet al., 1979; Tinoco, 1982). Some species, including many fish, insects and pulmonates, requirea dietary source of 18:3n-3 for normal growth and feed efficiency (Tinoco et al., 1979; Tinoco,1982). n-3 fatty acids do not appear to be essential for growth and feed efficiency in mammals,although 22:6n-3 is involved in energy metabolism and calcium ion channel activity in the


heart (Kang and Leaf, 1996; Leaf et al., 1999; Leifert et al., 1999; McLennan, 2001; Ferrieret al., 2002), which may explain the relation of high dietary intakes of n-3 fatty acids withreduced heart rate variability and sudden death (Christensen et al., 1996, 1997; Nair et al.,1997; Albert et al., 1998).

When released by phospholipases, membrane phospholipid n-6 and n-3 fatty acids becomeavailable as unesterified fatty acids and function as important signal molecules, in addition toserving as substrates for eicosanoid synthesis. Their functions include regulation of the activityof protein kinases, G-proteins, adenylate and guanylate cyclases, phospholipases, ion channelsand multiple other proteins and receptors (Bernsohn and Spitz, 1974; Bourre et al., 1989b; Parkand Ahmed, 1992; Gerbi et al., 1994; Poling et al., 1995; Kang and Leaf, 1996; Litman andMitchell, 1996; Koenig et al., 1997; Huster et al., 1998; Leifert et al., 1999; Bonin and Khan,2000; Litman et al., 2001). In addition, polyunsaturated fatty acids regulate the expressionof genes for regulatory proteins of lipid and carbohydrate metabolism through peroxisomeproliferator-activated receptor (PPAR)-dependent and independent mechanisms (Duplus et al.,2000; Clarke, 2001; Jump, 2002), and influence leptin gene expression (Reseland et al., 2001).New information also suggests that n-6 fatty acids are involved in adipogenesis during devel-opment by pathways involving PPAR γ2, although polyunsaturated fatty acids also suppressgenes related to lipogenesis (Ntambi et al., 1988; Clarke et al., 1990; Reginato et al., 1998).Recently, it has become clear that n-3 fatty acids alter the expression of genes related to endo-cytosis, signal transduction, synaptic vesicle recycling and formation, lipid metabolism,nuclear ligand-activated transcription factor receptors in brain, retinoic acid receptor (R × R)(Khair-El-Din et al., 1996; Mata de Urquiza et al., 2000; Kitajka et al., 2002), and intestinalnutrient absorption (Lampen et al., 2001).

3.1. n-3 essential fatty acids and visual function

The retina is an integral part of the central nervous system, which is composed of six celltypes – the photoreceptor cells, horizontal, bipolar, amacrine, interplexiform and ganglion cells –and communicates directly with the brain via ganglion cells passing through the optic nerve.The two photoreceptor cell types are the rods and cones. Rods are elongate and cylindricaland function as dim light receptors, while cones are shorter and usually cylindrical, mediatecolour vision and function at relatively higher light intensities (Giusto et al., 2000). Rods andcones are highly specialized differentiated neurons that contain a stack of photosensitvemembranes at the distal end (known as the outer segments), a central region containing mito-chondria, golgi and nucleus, and a synaptic terminal. The outer segments are made up ofdensely stacked disks, each of which is a double layer of infolded plasma membrane whichis highly enriched in 22 :6n-3. Vertebrate retina photoreceptor cells contain 50% protein and50% lipid, with 90–95% of the lipid present as phospholipid and 4–6% as cholesterol (Giustoet al., 2000). The major phospholipid species are PE, PC and PS. Within the outer segmentdisks, as much as 80% of the polyunsaturated fatty acids are 22:6n-3, with species of PE, PSand PC in which both fatty acids are present as 22:6n-3 (Fliesler and Anderson, 1983;Aveldano, 1987; Giusto et al., 2000). In bovine rod outer segment membranes, about 30% ofthe PC, 20% of the PE and 50% of the PS have a long-chain polyunsaturated fatty acid ester-ified at both the sn-1 and the sn-2 position (Aveldano et al., 1983; Aveldano, 1987; Aveldanoand Sprecher, 1987). It is notable that this unusual and characteristic membrane enrichmentof 22:6n-3 is present even in the bovine retina, a herbivore species that obtains no dietary22:6n-3. Specific pathways allow efficient recycling of 22:6n-3 from photoreceptor cellsduring shedding (turnover) from the disk tips. This involves phagocytosis by retinal pigment

S. M. Innis242

epithelium cells, then recycling of 22:6n-3 for reuse in synthesis of new disk membranes(Rodriguez de Turco et al., 1999).

About 80–90% of the retina rod and cone outer segment protein is the visual pigmentrhodopsin (opsin plus the carotenoid 11-cis-retinal), which functions as a photon receptorcoupled to regulatory G-proteins (Giusto et al., 2000). A light-induced change in the confor-mation of rhodopsin triggers a cascade of reactions that result in increased phosphodiesteraseactivity and decreased cyclic GMP, which leads to closure of the photoreceptor membranesodium channels and hyperpolarization, followed by depolarization. These events correspondto the positive A and negative B waves, respectively, of the electroretinograph (ERG). Recentstudies have provided evidence that 22:6n-3 may influence photoreceptor signal transduction byinfluencing the ability of photons to transform rhodopsin from metarhodopsin I to the activatedmetarhodopsin II state (Mitchell et al., 1992; Mitchell and Litman, 1998; Litman et al., 2001).Retina concentrations of 22:6n-3 increase during gestation and reach adult concentrations bythe time of term birth in humans (Martinez, 1992). Dietary deficiency of n-3 fatty acids duringdevelopment results in reduced 22:6n-3 in the retina of animals (Hrboticky et al., 1991). Inaddition to 22:6n-3, n-3 fatty acids with up to 36 carbons are present in small amounts in theretina (Aveldano et al., 1983; Aveldano, 1987; Aveldano and Sprecher, 1987), and these arereduced in rats fed a diet deficient in n-3 fatty acids throughout development (Suh et al., 1996,2000). The role of these fatty acids in retinal function has not yet been elucidated, althoughthey may be related to rhodopsin kinetics.

Several studies have addressed the effect of decreased retina 22:6n-3 on retinal and visualfunction in animals and human infants. Early studies in this field reported increased A and Bwave amplitudes in ERG responses of rats fed a fat-free diet (Benolken et al., 1973; Wheeleret al., 1975; Anderson et al., 1976). Later, Neuringer et al. (1984, 1986) reported prolongedrecovery times of dark-adapted A and B wave responses, and reduced rod and cone A waveresponses in full-field ERG of monkeys fed a severely n-3 fatty acid-restricted diet throughfetal and neonatal development. The role of n-3 fatty acids in retinal and visual developmentin human infants is discussed in section 8.

3.2. n-3 essential fatty acids and brain function

The brain contains the second highest concentration of lipid in the body, after adipose tissue,with 50% lipid on a dry weight basis, 10% lipid on a wet weight basis (Sastry, 1985). Unlikeadipose, however, the brain contains minimal amounts of triglyceride, the lipids of brainbeing almost entirely composed of the membrane structural components. About half of thelipid is phospholipid, with about 20% cholesterol, 15–20% cerebrosides and smaller amountsof sulphatides and gangliosides. The phospholipids of brain grey matter contain largeamounts of 20 : 4n-6 and 22: 4n-6, particularly in PI and PE, and high amounts of 22:6n-3 ingrey matter PE and PS (Sastry, 1985). Myelin, on the other hand, contains mainly saturatedand monounsaturated fatty acids (O’Brien and Sampson, 1965). The brain of herbivores, likethe visual photoreceptor cells, is enriched in 22:6n-3 despite the absence of a dietary intakeof preformed 22 : 6n-3, and only low amounts of 22:6n-3 in liver phospholipids (table 3).Only small amounts of 18 : 2n-6 are present in neural phospholipids, usually less than 1% ofall the fatty acids, and concentrations of 18:3n-3 are even lower. This unusual characteristicfeature of brain suggests the presence of specific pathways for selective uptake of 20:4n-6 and 22:6n-3 from plasma against a considerable concentration gradient, and that 20:4n-6 and 22:6n-3 are important to normal neural metabolism. The enrichment of 22:6n-3 in mammalian brain grey matter, together with the inability of animals to form n-3 fatty acids


de novo, has focused attention on the need to elucidate the role of 22:6n-3 in neural functionand requirements for n-3 fatty acids during brain growth and development.

The behaviour changes in developing animals fed diets deficient in n-3 fatty acids includereduced performance in maze tasks, habituation, exploratory activity in novel environmentsand brightness and olfactory-based discrimination learning (Yamamoto et al., 1988; Bourreet al., 1989b; Enslen et al., 1991; Innis, 1991; Yonekubo et al., 1993; Frances et al., 1996a,b;Greiner et al., 1997, 2001; Wainwright et al., 1998). Polydypsia, increased stereotypic (loco-motor) activity and altered performance on a task of recognition looking memory have alsobeen reported for monkeys fed a diet very low in n-3 fatty acids (<0.1% dietary energy 18:3n-3)through gestation and postnatal development (Reisbick et al., 1990, 1994). In addition to thedecrease in 22:6n-3, n-3 fatty acid deprivation results in increased n-6 fatty acids, including20:4n-6, 22:4n-6 and 22:5n-6, in the brain of animals (Galli et al., 1971; Bourre et al., 1984;Hrboticky et al., 1989, 1990; Favrelière et al., 1998) and human infants (Farquharson et al.,1995). It needs to be considered whether or not functional and biochemical changes associ-ated with n-3 fatty acid deficiency are due to a specific requirement for 22:6n-3, the increasein n-6 fatty acids, or a combination of effects.

Although considerable information has been published to describe the changes in brain andother tissues of animals fed diets varying in essential fatty acids, less is known on how changesin n-3 fatty acids alter behaviour. Recent research has established that 22:6n-3 alters metab-olism of several neurotransmitters, including dopamine, serotonin (5-HT), epinephrine,acetylcholine, GABA and N-methyl-D-aspartate (NMDA) channel activity (Delion et al., 1994,1996, 1997; Nishikawa et al., 1994; Hamano et al., 1996; Jones et al., 1997; Minami et al.,1997; Young et al., 1998; Zimmer et al., 1998, 1999, 2000a,b, 2002; de la Presa Owens andInnis, 1999a,b; McGahon et al., 1999; Itokazu et al., 2000). The cortical dopaminergic systemis important in modulation of learning, attention and motivation, and in the visual pathways(Gava and McKean, 1977; Brozoski et al., 1979; Le Moal and Simon, 1991; Antal et al., 1997;Basmak et al., 1999) and has been the focus of several studies with n-3 fatty acid-deficientanimals. Usually, these studies involve feeding a diet severely restricted in n-3 fatty acidsthrough several generations in order to deplete adipose tissue n-3 fatty acids that can be trans-ported from mother to offspring during gestation and suckling.

Newer research has shown that the effects of n-3 fatty acid deficiency on the brain are complex and region specific. The effects described include changes in dopamine concentra-tion, vesicular monoamine transporter 2, dopamine D2 receptor, tyrosine hydroxylase (the rate-limiting enzyme in dopamine synthesis), and dopamine storage pools (Delion et al., 1994, 1996,1997; Yoshida et al., 1997; Zimmer et al., 1998, 1999, 2000a,b, 2002). Newer techniques ofdual-probe microdialysis have shown that although dopamine is decreased in frontal cortexof n-3 fatty acid-deficient animals, dopamine may be increased in the nucleus accumbens(Zimmer et al., 2002). This could suggest that the mesocorticolimbic area dopaminergicsystem functions more, but the mesocortical pathway is less active in n-3 fatty acid-deprivedanimals. Newer studies also suggest that the effects of n-3 fatty acid deficiency may be moreapparent after a learning task, or administration of drugs that deplete endogenous dopaminestorage pools (Yoshida et al., 1997; Zimmer et al., 1998, 2000a,b).

Dietary deficiency of n-3 fatty acids leading to decreased 22:6n-3 in the developing brainis also associated with reduced serotonin and serotonin receptor binding in frontal cortex(Delion et al., 1996; de la Presa Owens and Innis, 1999a,b). This is of interest because 5-HT2receptors are believed to play an important role in behaviour (Leysen and Pauwels, 1990).Cross-sectional studies in humans have described a negative relation between 5-hydroxyindolaceteic acid (5-HIAA, which is the metabolite of serotonin) in cerebrospinal fluid and

S. M. Innis244

plasma 22:6n-3 among adults with violent behaviours (Hibbeln et al., 1998). Other aggres-sive and depressive affective disorders have also been postulated to be linked to altered n-3fatty acid metabolism (Adams et al., 1995; Hamazaki et al., 1996; Edwards et al., 1998;Hibbeln, 1998; Maes et al., 1999; Stoll et al., 1999). In vascular smooth muscle cells, enrich-ment with 22:6n-3 results in failure to respond to serotonin (Pakala et al., 2000). In othertissues, 20:5n-3 and 22:6n-3 both block the effect of growth factors that act through sig-nalling pathways involving receptor tyrosine kinase (such as platelet-derived growth factor,fibroblast growth factor, epidermal growth factor and insulin-like growth factor) and G-proteincoupled receptors (such as bomberin, bradykinin, vasopressin, thrombin, serotonin and throm-boxane A2) (Kaminski et al., 1993). These findings have potential implications for linking theeffects of 22:6n-3 to altered behaviour, and for a wide spectrum of physiological processes indeveloping animals.

The abundance of 22:6n-3 in synaptic terminals and cortical growth cone membranes sug-gests n-3 fatty acids could be important for brain structural growth. Morphological analysisof brain from rats fed a n-3 fatty acid-deficient diet through three generations have describeda lower cell body size in CA1 pyrimidal neurons at the septal location, although no differ-ences were found in other hippocampal regions, or in neuron volume, density or number(Ahmad et al., 2002a). Consistent with this, others have shown reduced nerve growth factorconcentrations in hippocampus of n-3 fatty acid-deficient rats (Ikemoto et al., 2000). Lowerneuron size in hypothalamus and parietal cortex of weanling rats, and in periform cortex ofmature rats with a 90% decrease in brain 22:6n-3, has also been reported (Ahmad et al.,2002b). Non-randomized studies in children with disorders of peroxisomal biogenesis haveprovided evidence that supplemental 22:6n-3 can lead to improvements in the marked deficitsin cognitive and visual function, and impaired myelination in these disorders (Martinez andVazquez, 1998; Martinez et al., 2000). In these disorders, the absence of peroxisomes resultsin the inability to form 22:6n-3. A mechanism through which n-3 fatty acids could stimulateoligodendrocyte metabolism and the synthesis of myelin has not been described.

Information to suggest that n-3 fatty acids influence synthesis and turnover of brain phos-pholipids has also been published. Elucidation of this possibility is important becausedecreased phospholipid synthesis or turnover could have profound effects on membranephospholipid-dependent signalling pathways. Studies with F2 generation 8-week-old ratsfound decreased PS in olfactory bulb, brain cortex and brain mitochondria, while PS in liverand adrenal were not affected (Hamilton et al., 2000). In other studies, PS was reduced by28% and PC was increased in hippocampus phospholipids of rats fed a n-3 fatty acid-deficientdiet (Murthy et al., 2002). Further evidence for a role of 22:6n-3 in PS metabolism has comefrom studies showing that intra-amniotic injection of ethyl-22:6n-3 increased both 22:6n-3and PS in brain of fetal rats (Green and Yavin, 1995). PS synthesis was also increased in C6glioma cells cultured with 22:6n-3 (Garcia et al., 1998). The significance of these findingsrelates to the role of PS in signal transduction through a regulatory role in protein kinase C(PKC) activation (Bell, 1986; Bell and Burns, 1991; Casamenti et al., 1991; Borghese et al.,1993; Mosier and Newton, 1998). Some evidence to suggest that dietary essential fatty acidsaffect the activity of enzymes involved in PC biosynthesis in brain synaptic membranes hasalso been published (Hargreaves and Clandinin, 1987). Studies using radiolabelled tracershave also provided evidence of decreased docosahexaenoyl-CoA, and phospholipid synthesisand turnover in vivo in n-3 fatty acid-deficient animals (Gazzah et al., 1995; Contreras et al.,2001). Chronic n-3 fatty acid deficiency, however, does not affect 20:4n-6 recycling in the brain,which suggests that phospholipases involved in release of 20:4n-6 and 22:6n-3 may be regu-lated independently (Contreras et al., 2001). Decreased turnover of membrane phospholipids


enriched in 22:6n-3, reflecting adaptation by the brain to conserve important membranepolyunsaturated fatty acids during limited dietary availability, has important implications foroptimal signal transduction.

4. ESSENTIAL FATTY ACID TRANSFER IN GESTATION

All of the n-6 and n-3 fatty acids accumulated in the fetus must ultimately be derived fromthe mother by placental transfer. Placental tissue contains Δ6- and Δ5-desaturases. The activityof Δ6-desaturase in ovine placenta increases towards term (Shand and Noble, 1981; Choy et al.,1997). Synthesis of 22:6n-3 by the placenta, however, has not been demonstrated. Whetherplacental desaturase activity contributes 20:4n-6 to the foetus does not appear to be known.Concentrations of 20:4n-6 and 22:6n-3 are much higher, whereas 18:2n-6 is lower in foetalthan maternal plasma lipids (Crawford et al., 1981; Elias and Innis, 2001) (fig. 3). Currentinformation suggests that placental transfer of 20:4n-6 and 22:6n-3 involves a multi-stepprocess of cell uptake and intracellular translocation that is facilitated by several membrane-associated and cytosolic fatty acid binding proteins. Specific membrane binding of n-6 andn-3 fatty acids, favouring 20:4n-6 and 22:6n-3 over 18:2n-6 or 18:3n-3, and preferential trans-fer of n-6 and n-3 fatty acid compared to non-essential fatty acids by human placenta havebeen reported (Campbell et al., 1996, 1998a,b; Haggarty et al., 1997; Dutta-Roy, 2000).

Despite the presence of pathways to facilitate the transfer of n-6 and n-3 fatty acids acrossthe placenta, the maternal intake of essential fatty acids during pregnancy can have a markedeffect on n-6 and n-3 fatty acid accretion in developing foetal tissues. Dietary deficiency ofn-3 fatty acids in gestation results in decreased 22:6n-3 and increased 22:4n-6 and 22:5n-6 ingrowth cones, the amoeboid leading edge of the growing neurite, in the fetal rat brain whilehigh intakes of 20:5n-3 and 22:6n-3 result in increased 22:6n-3 and decreased n-6 fatty acids(Innis and de la Presa Owens, 2001) (fig. 4). In the latter studies, differences in maternal

S. M. Innis246

Fig. 3. The relative enrichment of 18:2n-6, 18:3n-3, 20:4n-6 and 22:6n-3 in fetal compared to maternalplasma was calculated for each mother–fetal cord plasma pair as the difference in the given fatty acid in thematernal compared to fetal plasma/maternal plasma × 100%. Values shown are means ± SEM, n=55. Adaptedfrom Elias and Innis (2001).

Fig

. 4.

Eff

ect o

f m

ater

nal e

ssen

tial f

atty

aci

d in

take

on

n-6

and

n-3

fatty

aci

ds in

fet

al r

at b

rain

neu

rona

l gro

wth

con

e ph

osph

atid

ylch

olin

e (P

C),

pho

spha

tidyl

etha

nola

min

e (P

E),

pho

s-ph

atid

ylse

rine

(PS

) an

d ph

osph

atid

ylin

iosi

tol (

PI).

Rat

s w

ere

fed

saff

low

er, s

oybe

an o

r fi

sh o

il th

roug

hout

pre

gnan

cy a

nd f

etal

bra

in w

as a

naly

sed

at b

irth

. The

bar

s re

pres

ent m

eans

+SE

M,

valu

es w

ith a

dif

fere

nt s

uper

scri

pt a

re d

iffe

rent

(P

<0.

05).

Adp

ated

fro

m I

nnis

and

de

la P

resa

Ow

ens

(200

1).

dietary fat intake also resulted in altered dopamine in the fetal rat brain. Levels of 20:4n-6 and22:6n-3 in maternal plasma are also significantly correlated with the concentration of thesame fatty acid in newborn human infant plasma (Elias and Innis, 2001), and infants bornwith higher blood levels of 22:6n-3 and 20:4n-6 maintain this advantage for several weeksafter birth (Foreman-van Drangelen et al., 1995; Guesnet et al., 1999). This adipose tissue22:6n-3 and 20:4n-6 accumulated during foetal development contributes to the postnatal bloodpool of 22:6n-3 and 20:4n-6. Several studies have shown that supplementation of pregnantwomen with 22:6n-3 increases 22:6n-3 in plasma and red blood cell lipids of infants at birth(van Houwelingen et al., 1995; Connor et al., 1996; Helland et al., 2001), providing convincingevidence that the maternal intake of 22:6n-3 is an important factor in the placental transfer ofthis fatty acid.

Research is now starting to address the functional significance of the supply of 20:4n-6 and22:6n-3 to the developing foetus. A positive relation between 20:4n-6 and birth weight, despitehigh 18:2n-6, has been found in several studies (Koletzko and Braun, 1991; Elias and Innis,2001). Although this is consistent with the essentiality of 20:4n-6 in growth (Mohrhauer andHolman, 1963), the biological explanation for this is not known. Electroencephalography(EEG) at 2 days, as a measure of infant CNS maturity, indicated infants with a more matureEEG pattern have significantly higher 22:6n-3 in cord plasma phospholipids than infants witha less mature EEG pattern (Helland et al., 2001). Other recent studies have found an inverserelationship between maternal plasma 22:6n-3 and active sleep and sleep–wake transitions,and a positive association with wakefulness in 2-day-old infants (Ceruku et al., 2002), alsosuggesting that lower 22:6n-3 during gestation may be associated with delayed CNS maturation.Evidence of long-term sequelae, should these be present, due to early differences in 22:6n-3status at birth is still needed, as are intervention studies to rule out the possibility that theseassociations are explained by differences in other dietary and lifestyle variables that accom-pany differences in 22:6n-3 intake.

Long-chain polyunsaturated n-3 fatty acids, and the balance of n-6 to n-3 fatty acids in thediet, also influences gestation length in animals and humans (Olsen et al., 1992, 1995;Abayasekara and Wathes, 1999). In humans, high habitual intakes of 20:5n-3 and 22:6n-3,and supplementation with 2.7 g/day fish oil have been associated with longer gestation, pos-sibly due to suppression of n-6 fatty acid-derived eicosanoids that are involved in initiation ofparturition (Olsen et al., 1991, 1992). However, no association was found between length ofgestation and the intake of n-3 fatty acids among women for whom the 95% range of intakewas 0 to 0.75 g/day (Olsen et al., 1995). Typical intakes of long-chain n-3 fatty acids inNorth America are in the range of 100–200 mg/person/day (Innis and Elias, 2003). Whetherthe small increase in length of gestation in humans (of about 3–4 days) found at very highintakes of 20:5n-3 + 22:6n-3 is of functional benefit to the infant is not known.

5. ESSENTIAL FATTY ACID TRANSFER IN MILK

The n-6 and n-3 fatty acid composition of milk varies considerably among different species,probably reflecting differences in diet and lipid metabolism. The milks of ruminants, such ascattle and goats, have low concentrations of all the n-6 and n-3 fatty acids, while concentra-tion of 20:5n-3 and 22:6n-3 are particularly high in the milk of marine mammals (table 4).In addition to the effects of the diet, species differences in the desaturation and elongation, orβ-oxidation, of 18:2n-6 and 18:3n-3 also influence the amount of 20:4n-6 and 22:6n-3 inmilk. For example, milk from rats fed a diet containing 18:2n-6 and 18:3n-3, but no 20:4n-6 or22:6n-3, had 0.8% 20:4n-6 and 0.5% 22:6n-3. Milk from pigs fed the same fat source, on the

S. M. Innis248

other hand, had 0.6% 20:4n-6 and only 0.1% 22:6n-3. Human milk also typically has about0.5–0.7% 20:4n-6 and 0.2–0.4% 22:6n-3, with the amount of 22:6n-3 decreasing to 0.1%fatty acids among women following vegan diets with no preformed 22:6n-3 (Sanders et al.,1978; Innis, 1992; Innis and King, 1999).

Studies in multiple species have shown that the amounts of 18:2n-6, 18:3n-3, 20:5n-3 and22:6n-3 secreted in milk depends on the amount of the same fatty acid in the maternal diet.As for other dietary situations, this can be expected to influence the amount of essential fattyacids deposited in the tissues of the growing milk-fed neonate. Studies in animals, includingcattle, have shown that increasing the dietary intake of 18:2n-6 results in increased secretionof 18:2n-6 in milk (Bayourthe et al., 2000; Morales et al., 2000). Similarly, supplementationof diets with fish oil results in increased secretion of 20:5n-3 and 22:6n-3 in milk (Arbuckleand Innis, 1993). The increased 20:5n-3 and 22:6n-3 in the milk of lactating sows fed 1%weight diet as fish oil results in marked increases in 20:5n-3 and 22:6n-3 in plasma, red bloodcells and liver, and a smaller increase in 22:6n-3 in brain of the suckling piglets (table 5). Theimplications of the polyunsaturated fatty acid supply in milk to the growth, behaviour andhealth of livestock is worthy of consideration.

Autposy analysis has confirmed that, as in animals, the essential fatty acid compositionof the diet does influence the accretion of n-6 and n-3 fatty acids in developing human infanttissues (Makrides et al., 1994; Farquharson et al., 1995; Jamieson et al., 1999). Because of this,it is important to understand the effects of diet on the essential fatty acid composition ofhuman milk. The essential fatty acid composition of human milk shows considerable vari-ability both among and within populations (Jensen, 1989, 1999; Innis, 1992). This variabilityappears to be explained by differences in the maternal essential fatty acid intake. Human milkin North America and Europe currently has 12–16% total fat as 18:2n-6, representing about6–8% of the infant’s energy intake (Innis, 1992; Innis and King, 1999). Reports from the1950s found 7% 18:2n-6 in human milk lipids, and clearly showed that a diet high in 18:2n-6results in a marked increase in 18:2n-6 in the milk of lactating women (Insull and Ahrens,1959; Insull et al., 1959). Whether the 2-fold increase in 18:2n-6 in human milk is explainedby changes in dietary fat over the last half century is unclear.


Table 4

Fatty acid composition of selected animal and human milk lipids

Cowa Goatb Pigc Catd Dogd Rate Whalee Dolphine Sea lione Humanf

<10:0 10.8 22.4 1.0 0.4 0.2 14.8 0.612:0 3.0 6.4 4.1 0.9 0.7 10.7 0.1 0.2 0.1 4.114:0 10.6 12.4 3.1 5.4 4.9 11.6 6.1 5.6 3.7 6.116:0 33.7 33.7 28.0 26.8 26.5 23.5 12.8 12.9 18.0 19.418:0 12.6 3.8 5.1 10.1 3.7 4.8 5.0 4.8 2.1 7.216:1 1.8 9.4 5.1 7.8 1.3 7.3 8.1 7.6 2.518:1 21.4 11.3 13.2 40.3 10.9 17.8 25.6 38.2 40.1 35.718:2n-6 2.9 2.7 5.8 9.2 10.4 0.3 1.1 1.5 12.120:4n-6 0.2 1.1 0.8 0.8 0.5 1.1 0.418:3n-3 0.3 0.3 0.6 1.3 2.1 0.8 0.8 1.1 1.5 1.420:5n-3 0.2 0.4 6.5 4.2 5.5 0.122:6n-3 0.1 0.5 9.6 6.0 7.1 0.2

Adapted from: a Jensen (2002); b Le Doux et al. (2002); c Foote et al. (1990); d Parodi (1982); e Unpublished data;f Innis and King (1999).

Tabl

e 5

Pol

yuns

atur

ated

fat

ty a

cids

in m

ilk li

pids

and

suc

klin

g pi

glet

s fr

om s

ows

fed

vege

tabl

e an

d fi

sh o

ils

Sow

milk

Pigl

et p

lasm

aL

iver

PE

Cer

ebru

m P

ER

etin

a

Veg

. oil

+Fis

h oi

lV

eg. o

il+F

ish

oil

Veg

. oil

+Fis

h oi

lV

eg. o

il+F

ish

oil

Veg

. oil

+Fi

sh o

ill

16:0

27.6

28.6

23.7

23.8

10.4

10.2

10.4

8.5

8.1

9.4

18:0

4.4

4.5

20.5

22.6

*31

.826

.5*

28.7

23.9

25.6

27.9

18:1

37.5

28.6

14.9

9.7*

9.8

5.8*

14.7

18.4

9.1

9.8

18:2

n-6

11.0

20.7

20.8

19.7

8.7

10.0

0.9

0.7

1.3

0.9

20:4

n-6

0.6

0.6

11.1

9.2*

24.0

19.8

*15

.415

.714

.812

.822

:5n-

60.

10.

2*0.

50.

32.

42.

01.

60.

718

:3n-

31.

12.

60.

10.

5*0.

10.

10.

10.

10.

10.

120

:5n-

30.

10.

40.

41.

10.

41.

4*0.

20.

10.

10.

122

:5n-

30.

20.

41.

81.

22.

81.

9*0.

70.

51.

61.

022

:6n-

30.

11.

52.

67.

9*7.

321

.1*

13.1

16.6

*31

.432

.3

Ada

pted

fro

m A

rbuc

kle

et a

l. (1

993)

Sow

s w

ere

fed

diet

s w

ith 4

% b

y w

eigh

t fat

as

vege

tabl

e oi

l with

35%

18:

2n-6

and

5.3

% 1

8:3n

-3, o

r ve

geta

ble

plus

fis

h oi

l with

47.

7% 1

8:2n

-6, 7

.2%

18:

3n-3

,0.

7% 2

0:5n

-3, 0

.7%

20:

5n-3

and

0.2

% 2

2:6n

-3 f

rom

5 d

ays

befo

re p

artu

ritio

n th

roug

h la

ctat

ion.

The

pig

let t

issu

e fa

tty a

cids

wer

e de

term

ined

at 1

5 da

ys o

f ag

e. P

E, p

hosp

hatid

ylet

hano

lam

ine.

* P<

0.05

bet

wee

n gr

oups

.

Human milk concentrations of 22:6n-3 also show considerable variability, by over an orderof magnitude among populations and individuals. The average amount of 22:6n-3 in humanmilk has been reported to be 2.8% in Zhangzi, China, 1.4% fatty acids in Canadian Inuit,1.0% in Japan, 0.8–0.9% on the Malay Peninsula, 0.5% in Norway, 0.2–0.4% in Canada,Europe and the USA (Kneebone et al., 1985; Innis and Kuhnlein, 1988; Innis, 1992; Chuleiet al., 1995; Helland et al., 2001), and 0.1% among women in the UK following vegan diets(Sanders et al., 1978). While 18:2n-6 has increased, the amount of 22:6n-3 in human milk inWestern countries has decreased. Analyses conducted over the last 15 years in Australia andCanada show human milk 22:6n-3 has declined by 25–50% (Makrides et al., 1995b; Innis,2002). Supplementation of lactating women with fish or fish oil increases the secretion of20:5n-3 and 22:6n-3 into human milk (Harris et al., 1984; Henderson et al., 1992; Helland et al.,2001). Supplementation of lactating women with 22:6n-3 from single-cell oils with >40%22:6n-3 and <0.1% 20:5n-3, on the other hand, increases 22:6n-3 in human milk withoutincreasing 20:5n-3 (Makrides et al., 1996; Gibson et al., 1997; Fidler et al., 2000). Higheramounts of 22:6n-3 in milk result in higher intakes and higher blood lipid levels of 22:6n-3in breast-fed infants (Henderson et al., 1992; Gibson et al., 1997; Innis and King, 1999;Jensen et al., 2000). The amount of 20:4n-6 in milk appears to be much more tightly regu-lated, and does not seem to be reduced even with high intakes of 20:5n-3 and 22:6n-3 (Innisand Kuhnlein, 1988; Makrides et al., 1996; Fidler et al., 2000; Jensen et al., 2000).

Recent studies using stable isotope tracer methodologies have estimated that 20–25% of22:6n-3, 33% of 18:2n-6 and 12% of 20:4n-6 secreted in human milk are derived from the dietaryintake of the previous 48 hours (del Prado et al., 2000; Fidler et al., 2000). This may suggest thatadipose tissue provides a significant portion of the fatty acids secreted in milk. Consequently, theessential fatty acid content of the diet fed during gestation may also be important to the essentialfatty acid quality of the milk in later lactation. The importance of the supply of essential fattyacids in milk or milk substitutes to infant growth and development is discussed in section 8.

6. ESSENTIAL FATTY ACID ACCRETION IN THE BRAIN AND RETINA

Arachidonic acid (20:4n-6) and 22:6n-3 are accumulated in large amounts in the brain andretina during brain growth, particularly the brain growth spurt when the relative increase inbrain weight is at its highest (Dobbing and Sands, 1979). Animals that are born with a rela-tively mature brain, such as the monkey and guinea pig, known as precocial animals, have thehighest requirement for 20:4n-6 and 22:6n-3 for brain growth in utero. Altricial animals, suchas rats, which are born immature, have the largest period of brain growth after birth. Theincrease in rat brain 20:4n-6 and 22:6n-3 is most rapid between days 10 and 15 after birth. Atbirth, rat brain contains about 0.6 mg 20:4n-6 and 0.6 mg 22:6n-3. This increases to 2.46 mg20:4n-6 and 2.48 mg 22:6n-3 by 10 days of age, then almost doubles over the next 5 days to4.21 mg 20:4n-6 and 4.8 mg 22:6n-3 at 15 days of age, while the adult brain has 6.3 mg20:4n-6 and 10.2 mg 22:6n-3 (Sinclair and Crawford, 1972). In the human and pig, braingrowth is intermediate about the time of birth. The human brain weighs about 100 g at thebeginning of the third trimester of gestation, 370 g at term birth, and 2000 g at 2 years of age.The pig brain weighs about 5.8 g at 70 days of gestation, 10.7 g at 80 days, then increasesanother 3-fold to 31 g at 110 days of gestation, and weighs about 46 g by 4 weeks after birth(Pond et al., 2000). However, the rate of brain growth does not reflect the rate of maturationwithin individual regions of the brain, or the sensitivity of particular developing anatomical,biochemical or functional systems to the supply of essential fatty acids. For example, in thehuman, the growth of dendritic arbors and peak formation of synapses extends from about


34 weeks of gestation through 24 months after birth, during which time new connections format a rate of up to almost 40,000 synapses/sec (Huttenlocher et al., 1982; Borgeois, 1997;Huttenlocher and Dabholkar, 1997; Levitt, 2003).

Autopsy analyses of fetal and infant tissue from 22 weeks gestation to 17 weeks after birthhas been used to estimate the amounts of n-6 and n-3 fatty acids accumulated in developinghuman tissues (Clandinin et al., 1980a,b, 1981). Estimates of intrauterine accretion of essen-tial fatty acids indicate deposition of 552 mg/day n-6 fatty acids and 67 mg/day n-3 fatty acidsduring the last trimester of gestation. Most of this is accumulated in white adipose tissue(368 mg n-6 and 52 mg n-3 fatty acids/day). Accretion in the human fetal brain amounts to5.8 mg n-6 and 3.1 mg n-3 fatty acids/day, representing about 1.1% and 4.65% of total bodyaccretion, respectively. It is not known if the brain is protected, for example at the expense ofadipose tissue during limited dietary availability of essential fatty acids.

The extent to which the developing brain and retina depend on plasma 22:6n-3, or are ableto convert 18:3n-3 to 22:6n-3, is important to understand the dietary requirements for n-3fatty acids and the interpretation of differences in plasma 22:6n-3. Several studies have shownthat brain is able to take up and convert 14C-labelled 18:2n-6 and 18:3n-3 into longer-chainmetabolites (Sinclair and Crawford, 1972; Dhopeshwarkar and Subramanian, 1975a,b, 1976;Cohen and Bernsohn, 1978; Anderson and Connor, 1988; Washizaki et al., 1994; Chang et al.,1997). Conversion of 18:3n-3 and 22:5n-3 to 22:6n-3 in the retina and retinal pigment epithe-lium has also been demonstrated (Wang and Anderson, 1993; Alvarez et al., 1994). In vitro,brain cerebroendothelial cells take up 18:2n-6 and 18:3n-3 and can form 20:4n-6 and 22:4n-6,and 20:5n-3 and 22:5n-3 (Moore et al., 1991; Moore, 1994) as well as 22:6n-3 in a pathwayinvolving carbon chain 24 intermediates (Delton-Vandenbroucke et al., 1997). However, cerebraland cerebellar neuronal cells do not form 22:6n-3 (Moore, 2001). Astrocytes from sucklingrat brain, however, can form 22:6n-3, which after release can be taken up by neurons (Mooreet al., 1991; Moore, 2001). Although the pathway for 22:6n-3 formation is present, the majorproduct of 18:3n-3 metabolism in neonatal brain astrocytes is 22:5n-3, not 22:6n-3 (Innis andDyer, 2002, Williard et al., 2002); in vivo, brain has only minor amounts of 22:5n-3 (Sastry,1985). In the absence of n-3 fatty acids, neonatal brain astrocytes cultured with 18:2n-6 accu-mulate 20:4n-6 and 22:4n-6, rather than 22:5n-6 as occurs in n-3 fatty acid- deficient animals(Innis and Dyer, 2002). These findings indicate that while astrocytes have the ability to form22:6n-3 uptake of 22:6n-3 from plasma derived from synthesis in the liver, or from placentaltransfer before birth or provided preformed in the diet after birth is likely to be quantitativelymore important for brain 22:6n-3 accretion.

The pathways of transfer of 22:6n-3 from plasma to the developing brain are still incom-pletely understood. Several studies have shown that albumin-bound (unesterified) 20:4n-6and 22:6n-3 is taken up by the brain (Washizaki et al., 1994; Jones et al., 1997). Other stud-ies have shown that 18:2n-6 and 20:4n-6, but not 16:0, from physiological amounts ofradiolabelled 2-arachidonyl-lysophosphatidylcholine, 2-palmitoyl-lysophosphatidylcholine,and 2-linoleoyl-lysophosphatidylcholine, rapidly appear in the brain, and that uptake of18:2n-6, 20:4n-6 and 22:6n-3 occurs more readily from lysophospholipids than from unes-terified fatty acids (Thies et al., 1992, 1994; Bernoud et al., 1999; Lagarde et al., 2001).Lysophospholipids represent a significant (5–20%) proportion of total phospholipids inmammalian plasma, and thus could play an important role in the transfer of 20:4n-6 and22:6n-3 to extrahepatic tissues. Other recent studies have shown that HDL PE may deliverlong-chain polyunsaturated fatty acids to the brain via the sequential methylation of PE to PCat the blood–brain barrier (Magret et al., 1996). Further elucidation of the role of differentplasma lipids in transporting essential fatty acids to the developing brain is important in order

S. M. Innis252

to better identify the dietary and metabolic conditions that place developing infants at risk ofdelayed neural and retinal development.

7. ESSENTIAL FATTY ACID METABOLISM IN DEVELOPMENT

Prior to birth and during suckling, all of the n-6 and n-3 fatty acids accumulated by thedeveloping foetus and neonate must be derived by transfer from the mother, either by pla-cental transfer or from milk. The extent to which the foetus and young animal are able to use18:2n-6 and 18:3n-3 or depend on a supply of preformed 20:4n-6 and 22:6n-3 is important inunderstanding dietary essential fatty acid requirements during development. Both Δ6- andΔ5-desaturase activities are present in liver and brain from early in development. Desaturaseactivities are relatively low in fetal rat liver and higher in fetal brain (Sanders and Rana,1987). In mouse, brain Δ6-desaturase activity decreased about 4-fold from 3 days before birthto weaning, and activity in liver increased about 7-fold until 11 days after birth (Bourre andPiciotti, 1992). Linolenic acid (18:3n-3) deficiency increased, and feeding with 20:5n-3 +22:6n-3 decreased, activity of the Δ6-desaturase (Dinh et al., 1993). Similarly, pig liver andbrain at 63 days of gestation, 3 days postpartum (term birth, about 115 days) and at 12 weeksof age are able to convert [1-14C]18:2n-6 and 20:3n-6 to tetraenes (20:4n-6, 22:4n-6) andpentaenes (22:5n-6), but the rate of conversion in the fetal piglet liver and brain is 3–5-foldlower than in the animals after birth, and also much lower in brain than liver (Clandinin et al.,1985a,b). Li et al. (2000) found no synthesis of 22:6n-3 from [U-13C]18:3n-3 in liver of foetalpiglets at 70–72 or 110–112 days gestation. Instead, desaturation of 18:3n-3 was limited at20:5n-3, suggesting that the final steps of elongation and peroxisomal chain shortening maybe limiting. Biosynthesis of 22:6n-3, however, increased rapidly over the first 14 days afterbirth (Li et al., 2000). This is compatible with placental supply of 22:6n-3 before birth, andthe low amount of 22:6n-3 in sow milk. Stable isotope tracer methodology has been used to show that foetal baboons can form 18:3n-3 from a dose of intravenous [U-13C]18:3n-3 (Su et al., 2001). However, only about 0.6% of the 18:3n-3 administered was recovered in brain22:6n-3. Again, this is consistent with low desaturation of n-3 fatty acids during prenataldevelopment when placental transfer facilitates high concentrations of 22:6n-3 in foetalplasma. Activity of Δ6- and Δ5-desaturase has also been shown in human foetal liver micro-somes from as early as 17 weeks of gestation (Chambaz et al., 1985; Poisson et al., 1993). Theactivity of the pathway to 22:6n-3 in human foetal liver prior to birth, however, is not known.

Studies over a decade ago established that blood lipid 20:4n-6 and 22:6n-3 are lower ininfants fed formula with 18:2n-6 and 18:3n-3 as the only n-6 and n-3 fatty acids than inbreast-fed infants (Putnam et al., 1982; Ponder et al., 1992). Higher amounts of 18:3n-3 in thediet do not result in higher plasma or red blood cell 22:6n-3 in infants (Ponder et al., 1992).Studies in this field interpreted these findings as evidence of “immature” activity of a putativeΔ4-desaturase believed to be responsible for synthesis of 22:6n-3 from 22:5n-3 (Putnam et al.,1982; Carlson et al., 1993). However, 20:4n-6, which is formed by Δ6- and Δ5-desaturationof 18:2n-6, is also lower in infants fed formula than in breast-fed infants. Advances in stableisotope tracer technologies have now allowed the demonstration that conversion of isotopi-cally labelled 18:3n-3 to 22:6n-3 and of 18:2n-6 to 20:4n-6 is as high or higher in preterm asin term infants, and higher in infants not receiving preformed 20:4n-6 and 22:6n-3 fromhuman milk (Carnielli et al., 1996; Salem et al., 1996; Sauerwald et al., 1997; Uauy et al.,2000; Pawlosky et al., 2001). The conversion of 18:3n-3 to 22:6n-3, however, appears to behighly variable among individuals and could be as low as <1 to 4% 18:3n-3 converted to22:6n-3 in humans (Emken et al., 1994; Pawlosky et al., 2001). These tracer methodologies


have not yet been able to provide quantitative estimates of the contribution of 22:6n-3 syn-thesis to accumulation at the level of the tissues. Data from autopsy tissue also providesinformation on essential fatty acid metabolism in human infants. Infants fed formula con-taining 0.4% or 1.5% 18:3n-3 but no long-chain polyunsaturated fatty acids had lower brain22:6n-3 and higher 22:4n-6 and 22:5n-6 in cerebral cortex PE and PS than infants who werebreast-fed (Farquharson et al., 1995). Brain 22:5n-6 was also higher and the 22:6n-3 to 22:5n-3ratio lower in infants fed the formula with 0.4% 18:3n-3 rather than 1.5% 18:3n-3 (table 6).The increase in 22:5n-6 is consistent with deficiency of 18:3n-3, and also provides evidencethat the final steps of elongation, Δ6-desaturation and peroxisomal partial chain shortening ofessential fatty acids are active, even in young infants. Stable isotope studies have also shownthat 22:6n-3 synthesis is higher in infants with higher intakes of 18:3n-3 (Sauerwald et al.,1996). The findings, however, do not address whether a higher intake of 18:3n-3 can supportsufficient endogenous synthesis of 22:6n-3 to meet the needs of the developing brain.

8. LONG-CHAIN FATTY ACIDS IN HUMAN INFANT NUTRITION

Several studies have reported that groups of infants who are breast-fed perform better on testsof neurodevelopment than bottle-fed infants (Anderson et al., 1999). This advantage appearsto remain even when many factors, such as socio-demographic variables, maternal eductionand birth order, are controlled, although further work still needs to be done with respect toclearly identifying and controlling extent and duration of breast-feeding in these comparativestudies (Drane and Logemann, 2000). Although milk contains numerous biologically activecomponents not present in infant formulas, the presence of 22:6n-3 in human milk and thecritical role of this fatty acid in normal retina and brain development has led to intense inves-tigation of the essential fatty acid needs of the human infant for growth and development. Thecentral question is whether the rate of conversion of 18:3n-3 to 22:6n-3 in human infants issufficient to provide enough 22:6n-3 for optimal brain and retinal function. These studies arecomplicated because the relationship between plasma or red blood cell and brain 22:6n-3 iscurvilinear, rather than linear (Arbuckle et al., 1991; Rioux et al., 1997; Ward et al., 1998).At intakes above requirement, progressive increases in plasma and red blood cell 22:6n-3with increasng dietary intake are not accompanied by similar increases in 22:6n-3 in the brain

S. M. Innis254

Table 6

Cerebral cortex long-chain polyunsaturated fatty acids in human infants fed human milkor infant formula

PE PS

Formula 18:3n-3 Formula 18:3n-3

Fatty acid Human milk 1.5% 0.4% Human milk 1.5% 0.4%

20:4n-6 17.6 20.1* 19.6∗∗ 7.9 9.0 9.022:4n-6 12.0 12.6 14.3* 7.9 8.5 9.322:5n-6 3.2 4.8* 7.0* 5.3 7.7* 10.4**

22:6n-3 17.7 13.4* 11.6* 23.5 19.3* 14.4*

22 :5n-6/22:6n-3 0.18 0.36 0.60 0.22 0.40 0.72

*P<0.01, **P< 0.5 compared to human milk. The formula contained 1.5% 18:3n-3 and 16.0% 18:2n-6 or 0.4%18:3n-3 and 14.5% 18:2n-6. Adapted from Farquharson et al. (1995). PE, phosphatidylethanolamine; PS, phosphatidylserine.

and retina. As might be expected, increasing the dietary intake of 20:4n-6 and 22:6n-3 in younginfants results in an increase in these fatty acids in blood lipids, with the increase dependenton the amount of 20:4n-6 and 22:6n-3 fed, and also the concurrent intake of 18:2n-6 (Carlsonet al., 1993b; Makrides et al., 1995a; Innis et al., 1996). Because of this, the assessment ofdietary n-3 fatty acid requirements requires an approach combining measurement of growth,functions dependent on neural or other tissue 22:6n-3, and blood lipid 22:6n-3.

Premature infants are considered particularly vulnerable to nutritional deficiency becauseof their limited adipose tissue mass and immaturity in many metabolic and physiologicalpathways at birth. Following birth, there is a rapid decline in the high 20:4n-6 and 22:6n-3characterisitic of foetal plasma, and a large increase in 18:2n-6 (Innis, 1991). The increase inbrain weight relative to body weight is also more rapid during the last trimester of gestationthan at later ages. Numerous studies have documented an increased incidence of develop-mental problems in prematurely born children at school age (Bhutta et al., 2002), which islikely to reflect in part failure to provide optimal essential nutrients to support third trimesterbrain growth and development ex utero.

Several options are available for including 20:4n-6 and 22:6n-3 in infant formulas. For20:4n-6, these include egg total lipids, egg-derived triglycerides and egg phospholipids, ortriglycerides from the single-cell fungus Mortierella alpina (which are highly enriched in20:4n-6). Sources of 22:6n-3 include fish oils, either with high 20:5n-3 and 22:6n-3, fish oilssuch as tuna oils specifically selected to be low in 20:5n-3 and high in 22:6n-3, egg lipids, oroils from the single-cell micro-algae Crypthecodinium cohnii which contains >40% 22:6n-3and virtually no 20:5n-3. Early studies by Birch et al. reported a higher rod threshold andlower maximal amplitude values in the B wave in ERG recordings of infants fed formulacontaining corn oil with only 0.5% 18:3n-3 (about 0.25% dietary energy) (Birch et al., 1992;Uauy et al., 1992). The higher threshold and lower maximum amplitude suggest that greaterluminescence was needed to elicit a response, and that signal transduction was reduced,respectively. Subsequent studies found that supplementation of preterm infant formula con-taining 1.2% energy as 18:3n-3 (about 2.4% fatty acids) with 0.12% or more energy as22:6n-3 and 0.23% or more energy as 20:4n-6 resulted in higher visual acuity when measuredeither by VEP techniques or with behavioural measures based on the ability of the infant todemonstrate a looking response to black and white gratings (stripes) of varying size (Carlsonet al., 1993b, 1996a; Faldella et al., 1996; San Giovanni et al., 2000; O’Connor et al., 2001).

Recently, a large multicenter trial found an advantage in Bayley mental developmentalinventories and the MacArthur communicative inventories in preterm infants of birth weightless than 1250 g fed formula supplemented with 20:4n-6 and 22:6n-3 (O’Connor et al., 2001).The long-term significance of these early changes in visual and neural development is not yetknown. The findings, however, show that 1.2% dietary energy as 18:3n-3 does not meet then-3 fatty acid requirements of preterm infants, and that visual and some aspects of neuraldevelopment are increased by small dietary intakes of 22:6n-3. Several studies, however, havefound evidence of reduced growth in preterm infants fed formulas containing 22:6n-3 fromfish oils (Carlson et al., 1992, 1996a; Ryan et al., 1999), and a positive relation between20:4n-6 and growth has been described (Carlson et al., 1993a). Recent clinical studies havealso provided evidence of higher growth in preterm infants fed formulas supplemented with0.27% energy 20:4n-6 and 0.14% 22:6n-3 from single-cell triglycerides (Innis et al., 2002).While it is not clear if this is related to the role of 20:4n-6 in regulating early aspects of adipo-genesis (Reginato et al., 1998), bone growth (Weiler and Fitzpatrick-Wong, 2002) or othermechanisms, it is evident that both the n-6 and the n-3 long-chain polyunsaturated fatty acidsplay an important role in the growth and development of preterm infants. Following these


studies, preterm infant formulas containing both 20:4n-6 and 22:6n-3 are now available in manycountries. Some evidence also suggests that inclusion of 20:4n-6 and 22:6n-3 in formula fed topreterm infants may increase immune system development, as assessed by ex vivo mitogenstimulation of peripheral blood lymphocytes (Field et al., 2001). Much more work on the roleof essential fatty acids and their role in the developing immune and other systems is needed.

Although many studies have addressed the role of dietary 22:6n-3 and 20:4n-6 in thegrowth and development of term infants, current findings are not consistent. Several earlystudies reported evidence of higher looking acuity and higher VEP acuity in term gestationinfants fed formula supplemented with 22:6n-3 in amounts similar to that in human milk(Makrides et al., 1995c; Carlson et al., 1996b). Recent large multicenter trials, however, havenot found a benefit of adding 0.12% to up to 0.36% 22:6n-3 (about 0.06–0.18% energy) ofthe formula fat on looking acuity, VEP acuity, or scores on tests of mental and motor skilldevelopment (Auestad et al., 1997, 2001; Lucas et al., 1999; Makrides et al., 2000). Birch et al.(1998, 2000), however, found higher VEP acuity during the the first year of life, and higherscores on the Bayley scales of infant development at 18 months of age among term infantsfed formula containing a fat blend with 0.36% 22:6n-3 compared with no 22:6n-3 and 1.49%18:3n-3. VEP acuity was also higher in 12-month-old infants who were weaned from breast-feeding to formula with 22:6n-3 rather than unsupplemented formula at about 5 months ofage (Birch et al., 2002). Higher scores on a three-step problem-solving task have also beenreported for term infants fed a formula containing fat with 0.15–0.25% 22:6n-3 and 0.7%18:3n-3 (Willatts et al., 1998). No evidence of altered growth was found in any of the latterstudies with term infants fed formulas containing 22:6n-3 and 20:4n-6. The discrepancyamong studies with term infants could involve the low 18:3n-3 content and thus dependenceon 22:6n-3 in some of the formulas tested, addition of inadequate amounts of 22:6n-3 to seea positive effect, differences in the formula balance of 20:4n-6 to 22:6n-3, variability in the22:6n-3 status of the infants at birth, age of the infants at testing, and lack of test sensititivityto detect biologically important differences in development.

Recent studies have also addressed whether the variability in 22:6n-3 in human milk, andas a result in the diet and blood lipids of the breast-fed infant, is of physiological significanceto infant development. Infants in the lowest tertile of RBC PE 22:6n-3 who received milk with0.17% fat as 22:6n-3 had significantly lower visual acuity at 2 and 12 months of age thaninfants in the highest tertile of RBC PE 22:6n-3 who received milk with 0.31% 22:6n-3 (Inniset al., 2001; fig. 5). No relation was found between the infants’ 20:4n-6 or 22:6n-3 status andscores on the Bayley II mental or motor developmental indices, novelty preference assessedusing the Fagan test, or on a standardized object search task (Piaget’s A not B). However, theinfants’ 22:6n-3 status at 2 months of age was significantly related to the ability to discrimi-nate a non-native (Hindi) retroflex and dental phonetic contrast at 9 months of age, and tolanguage production and comprehension assessed with the CDI at 14 and 17 months of age,after adjusting for confounding variables (Innis et al., 2001; Innis, 2002). A significant asso-ciation between sweep VEP acuity and human milk 22:6n-3 was also recently reported in across-sectional study of breast-fed infants (Jorgensen et al., 2001). These associationsbetween 22:6n-3 and visual and neural development in breast-fed infants, while consistentwith the essential role of 22:6n-3 in retina and brain function, cannot be interpreted as ademonstration of causality. This requires dietary intervention that modifies the maternalintake of 22:6n-3 but not other nutrients. However, newer research to show dependence of thefetal and infant 22:6n-3 on the maternal intake of 22:6n-3 raises important questions aboutthe n-3 fatty acid requirements of pregnant and lactating women with respect to supportingoptimal visual and neural development in the infant.

S. M. Innis256

9. DIETARY ESSENTIAL FATTY ACID REQUIREMENTS

The physiological effects of essential fatty acid deficiency have been well described, partic-ularly in rats fed a fat-free diet and replenished with varying amounts of n-6 fatty acids(Mohrhauer and Holman, 1963; Holman, 1991; Innis, 1992; table 1). Studies in rodents haveshown that a dietary intake of as low as 1–2% energy 18:2n-6 is sufficient to avoid anyadverse effects on reproduction, gestation, perinatal mortality or growth. Careful dose-response studies have shown that the minimum intake of 18:2n-6 required to sustain aconstant level of 20:4n-6 in all organs (including brain, liver, lung, heart, kidney, muscle andadipose) is 2.4% of dietary energy (Bourre et al., 1990). A similar effect, with maintenanceof normal growth, is achieved with a much smaller amount of 20:4n-6, of about 0.2% dietaryenergy (Mohrhauer and Holman, 1963). Similar studies on the requirement for n-3 fatty acidshave shown that a dietary intake of 0.26% energy 18:3n-3 in adults and 0.4% energy 18:3n-3in developing animals maintains 22:6n-3 in brain, liver, heart and other organs (Bourre et al.,1989a, 1993). There is no evidence that rodents require a dietary source of 22:6n-3 if fed adiet containing sufficient 18:3n-3. The requirement for n-3 fatty acids, however, can be metby small amounts of 22:6n-3 rather than conversion from 18:3n-3.

Studies in young piglets have shown that an intake of <0.7% energy from 18:3n-3 is inad-equate to support accretion of 22:6n-3 in neural tissues (Hrboticky et al., 1989, 1990, 1991).The accretion of 22:6n-3 was reduced when the diet of the neonatal piglet contained a ratioof 18:2n-6 to 18:3n-3 of 16:1 or higher, rather than 8:1 or lower (Arbuckle et al., 1992, 1994).Further understanding of the importance of the concurrent intake of 18:2n-6 to tissue accre-tion of 20:4n-6 and 22:6n-3 is needed. Higher intakes of 2% energy 18:3n-3 resulted in higher22:6n-3 in the developing piglet brain (Arbuckle et al., 1992, 1994), compatible with studiesshowing postnatal desaturation and elongation of 18:3n-3 to 22:6n-3 in this species (Li et al.,2000). However, studies in young piglets also provide ample evidence that 22:6n-3, either insows’ milk or in a milk replacer diet, is much more efficacious than 18:3n-3 in increasing22:6n-3 in blood, liver, and brain and retina 22:6n-3 (Arbuckle et al., 1991; Arbuckle andInnis 1992, 1993; de la Presa Owens and Innis, 1999a,b). A dietary intake of as little as 0.15%energy as 22:6n-3 supports similar or higher brain and retina 22:6n-3 than a diet with 2%energy as 18:3n-3.


Fig. 5. Relation of milk and infant red blood cell phosphatidylethanolamine (PE) 22:6n-3 at 2 months ofage to visual acuity at 2 and 12 months of age in healthy term gestation infants. The bars represent mean visualacuity ± SD; * significantly different from infants in the lowest tertile of 22:6n-3. Adapted from Innis et al. (2001).

Stable isotope tracer studies in pregnant and fetal baboons have also provided convincingevidence that conversion of 18:3n-3 to 22:6n-3 is much less efficient in supplying 22:6n-3 forincorporation into developing brain and other organs than preformed 22:6n-3 (Greiner et al.,1997; Su et al., 1999, 2001). The available data suggest that 90% and perhaps more of dietary18:3n-3 undergoes β-oxidation, with the acetyl-CoA either further oxidized to generate ATP,or recycled for de novo synthesis of other lipids. Although the increase in brain and retina22:6n-3 with increasing intakes of 22:6n-3 is much lower than the increase in 22:6n-3 in liveror plasma, high intakes of 20:5n-3 and 22:6n-3 have been shown to decrease visual functionin guinea pigs, and to decrease auditory brainstem evoked potentials in neonatal rats(Weisinger et al., 1995; Stockard et al., 2000). Although the mechanism of these effects is notyet known, these findings suggest the need for dose-response studies to determine the safeupper limit of 22:6n-3 intake in developing animals.

Studies in the 1960s identified skin lesions and growth failure among human infants fedskimmed cows’ milk, which provides minimal amounts of n-6 fatty acids (Hansen et al., 1958).Clinical and biochemical signs of n-6 fatty acid deficiency, many of which are related to the roleof n-6 fatty acids in the skin, are not seen in human infants fed diets providing 18:2n-6 in amountsabove recommended intakes of 3–4% energy as 18:2n-6 (FAO/WHO, 1993). Expert groups havesuggested an acceptable range of intake of 18:2n-6 for preterm and term human infants of3.2–12.8% energy (LSRO, 1998, 2001). The absence of skin signs, normal growth, and atriene/tetraene ratio of <0.2 in plasma lipids, however, ensures neither optimal 20:4n-6 nor balance between n-6 and n-3 fatty acids in developing tissues. Definitive information on n-6 fattyacid requirements that are based on functional endpoint indicators related to 20:4n-6 metabolismare largely lacking. Similarly, the n-3 fatty acid requirements of human infants are still unclear.The suggested acceptable range of 18:3n-3 intake for preterm and term infants is 0.7–2.1% ofdietary energy (LSRO, 1998, 2001). The requirement for 18:3n-3, however, depends on theamount of 22:6n-3 provided to the infant. Recent clinical studies in premature infants suggest thatin the absence of a dietary intake of 22:6n-3, an intake of 1.2% energy as 18:3n-3 does not meetthe needs of the developing brain and retina (O’Connor et al., 2001).

Another approach is to estimate a dietary intake likely to meet the needs for accretion,based on knowledge gained from autopsy tissue analyses. These analyses suggest that about67 mg n-3 fatty acids (mostly 22:6n-3) and 552 mg n-6 fatty acids are accumulated per dayin fetal tissue during the last trimester of gestation (Clandinin et al., 1981). Thus, an infantfed human milk or a milk substitute that provides the only dietary source of polyunsaturatedfatty acids would need to recieve about 0.23% fat as 22:6n-3 and 2% fat as n-6 fatty acids,assuming an intake of 780 ml/day. Definitive data on the amount of dietary 18:3n-3 convertedto 22:6n-3 during development are not available. Assuming this to be 10%, then the infant dietwould need to contain at least 2.3% 18:3n-3 in order to meet the needs for n-3 fatty acids.

Some expert groups have suggested a dietary intake of essential polyunsaturated fatty acidsbased on the amounts present in human milk. This is problematic because the essential fattyacid content of milk is greatly influenced by the amounts of n-6 and n-3 fatty acids in thematernal diet. The competition between 18:2n-6 and 18:3n-3 for desaturation has also led torecommendations for the n-6 to n-3 fatty acid in infant diets that are based on the composition ofhuman milk. The total n-6 to total n-3 fatty acid ratio of human milk is generally in the range of4:1 to 10:1 (Neuringer and Connor, 1986), but this ratio does not consider the differences in tissuehandling and biological activity of 18:2n-6 and 20:4n-6, and 18:3n-3, 20:5n-3 and 22:6n-3.Dietary 22:6n-3 results in increased blood lipid 22:6n-3 in human infants, while the increasingdietary intake of 18:3n-3 has little effect. For example, the plasma phospholipids of infants fedformula with about 0.12% or 0% energy 22:6n-3 had 5.2 ± 0.2% and 2.0 ± 0.1% 22:2n-3,

S. M. Innis258

respectively, but infants fed formula with 0.4% or 2.4% 18:3n-3 and 0% 22:6n-3 had 2.3 ± 0.2%and 2.2 ± 0.3% 22:6n-3, respectively (Ponder et al., 1992; Innis et al., 1996). The increase inred blood cell and plasma phospholipid 22:6n-3 was lower in infants fed formula with 32%rather than 20% 18:2n-6 (Innis et al., 1996). The Food and Agriculture/World HealthOrganization recommend that term infants receive, per kg body weight, 600 mg 18:2n-6, 50 mg18:3n-3, 40 mg 20:4n-6 and 20 mg 22:6n-3 per day, based on the amounts in human milk(FAO/WHO, 1993). Similarly, the Institute of Medicine of the National Academies of Medicineprovides an adequate intake (AI) for infants of 0–12 months of 4.4–4.6 g/day of n-6 fatty acidsand 0.5 g/day n-3 fatty acids, based on the intake of breast-fed infants (IOM, 2002). However,there is no evidence from clinical studies to indicate that these amounts exceed or meet theneeds to maximize potential neural development in young infants (FAO/WHO, 1993).

10. CONCLUSIONS AND FUTURE PERSPECTIVES

It is clear that essential fatty acid deficiency can lead to profound problems in growth, andfunctional disturbances in many organs including the developing central nervous system.Large gaps still exist in understanding the requirements, metabolism and functions of essentialfatty acids, particularly the long-chain polyunsaturated fatty acids 20:4n-6 and 22:6n-3,during development. In particular, pathways and regulatory mechanisms involved in the trans-fer of 20:4n-6 and 22:6n-3 across the placenta and in milk, and the effect of the concurrentintake of 18:2n-6 in modulating fatty acid accretion in developing tissues, is incompletelyunderstood. Likewise, much is yet to be learned regarding the role of n-6 and n-3 fatty acidsin the molecular, biochemical and histological development of the central nervous system,and their relation to cognitive and behavioural impairments in developing animals andinfants. Little information is available on the role of polyunsaturated fatty acids in the molec-ular and functional development of the immune system, intestine, bone, adipose tissueand many other organs; these areas afford considerable oportunities for new research. Futureexpert groups seeking to establish polyunsaturated fatty acid requirements in developmentwill benefit from a greater understanding of species differences in essential fatty acid metabo-lism, and dose-response studies to elucidate the safe and adequate range and tolerableupper limit of intake of individual n-6 and n-3 fatty acids, and their appropriate balance inthe diet.

REFERENCES

Abayasekara, D.R., Wathes, D.C., 1999. Effects of altering dietary fatty acid composition on prostaglandinsynthesis and fertility. Prostagland. Leuk. Essent. Fatty Acids 61, 275–287.

Adams, P.B., Lawson, S., Sanigorski, A., Sinclair, A.J., 1995. Arachidonic to eicosapentaenoic acid ratioin blood correlates positively with clinical symptoms of depression. Lipids 31, S157–S161.

Ahmad, A., Moriguchi, T., Salem, N., 2002a. Decrease in neuron size in docosahexaenoic acid-deficientbrain. Pediatr. Neurol. 26, 210–218.

Ahmad, A., Murthy, M., Greiner, R.S., Moriguchi, T., Salem, N. Jr., 2002b. A decrease in cell sizeaccompanies loss of docosahexaenoate in rat hippocampus. Nutr. Neurosci. 5, 103–113.

Albert, C.M., Hennekens, C.H., O’Donnell, C.J., Ajani, U.A., Carey, V.J., Willett, W.C., Ruskin, J.N.,Manson, J.E., 1998. Fish consumption and risk of sudden cardiac death. J. Amer. Med. Assoc. 279, 23–28.

Alvarez, R.A., Aguire, G.D., Acland, G.M., Anderson, R.E., 1994. Docosapentaenoic acid is convertedto docosahexaenoic acid in the retinas of normal and prcd-affected miniature poodle dogs. Invest.Ophthalmol. Vis. Sci. 35, 402–408.

Anderson, G.J., Connor, W.E., 1988. Uptake of fatty acids by the developing rat brain. Lipids 23,286–290.


Anderson, J.W., Johnstone, B.M., Remley, D.T., 1999. Breast-feeding and cognitive development: a meta-analysis. Amer. J. Clin. Nutr. 70, 525–535.

Anderson, R.E., Landis, D.J., Dudley, P.A., 1976. Essential fatty acid deficiency and renewal of rod outersegments in albino rat. Invest. Ophthalmol. 15, 232–236.

Antal, A., Keri, S., Bodis-Wollner, J., 1997. Dopamine D2 receptor blockade alters the primary and cog-nitive components of visual and evoked potentials in the monkey, Macaca fasciliaris. Neurosci. Lett.232, 179–181.

Arbuckle, L.D., Innis, S.M., 1992. Docosahexaenoic acid in developing brain and retina of piglets fedhigh or low α-linolenate formula with and without fish oil. Lipids 27, 89–93.

Arbuckle, L.D., Innis, S.M., 1993. Docosahexaenoic acid is transferred through maternal diet to milk andto tissues of natural milk-fed piglets. J. Nutr. 123, 1668–1675.

Arbuckle, L.D., MacKinnon, M.J., Innis, S.M., 1994. Formula 18:2(n-6) and 18:3(n-3) content and ratioinfluence long-chain polyunsaturated fatty acids in the developing piglet liver and central nervoussystem. J. Nutr. 124, 289–298.

Arbuckle, L.D., Rioux, F.M., Mackinnon, M.J., Hrboticky, N., Innis, S.M., 1991. Response of (n-3) and(n-6) fatty acids in brain, liver and plasma of piglets fed formula to increasing, but low, levels of fishsupplementation. J. Nutr. 121, 1536–1547.

Arbuckle, L.D., Rioux, F.M., MacKinnon, M.J., Innis, S.M., 1992. Formula α-linoleic (18:3(n-6) andlinoleic (18:2(n-6)) acid influence neonatal piglet liver and brain saturated fatty acids, as well asdocosahexaenoic acid (22:6(n-3)). Biochim. Biophys. Acta 1125, 262–267.

Auestad, N., Halter, R., Hall, R., Blatter, M., Bogle, M.L., Burks, W., Erickson, J.R., Fitzgerald, K.M.,Dobson, V., Innis, S.M., Singer, L.T., Montalto, M.G., Jacobs, J.R., Qiu, W., Bornstein, M.H., 2001.Growth and development in term infants fed long-chain polyunsaturated fatty acids, a double-masked,randomized, parallel, prospective, multivariate study. J. Pediatr. 108, 372–381.

Auestad, N., Montalto, M.B., Hall, R.T., Fitzgerald, K.M., Wheeler, R.E., Connor, W.E., Neuringer, M.,Connor, S.L., Taylor, J.A., Hartmann, E.E., 1997. Visual acuity, erythrocyte fatty acid composition,and growth in term infants fed formulas with long chain polyunsaturated fatty acids for one year.Pediatr. Res. 41, 1–10.

Aveldano, M.I., 1987. A novel group of very long chain polyenoic fatty acids in dipolyunsaturated phos-phatidylcholines from vertebrate retina. J. Biol. Chem. 262, 1172–1179.

Aveldano, M.I., Sprecher, H., 1987. Very long chain (C24–C36) polyenoic fatty acids of the n-3 andn-6 series in dipolyunsaturated phosphatidylcholines from bovine retina. J. Biol. Chem. 262,1180–1186.

Aveldano, M.I., Pasquare de Garcia, S.J., Bazan, N.G., 1983. Biosynthesis of molecular species of inos-itol, choline, serine, and ethanolamine glycerophospholipids in the bovine retina. J. Lipid Res. 24,628–639.

Basmak, H., Yildirim, N., Erdinc, O., Ywdakul, S., Ozdemir, G., 1999. Effect of levodopa therapy onvisual evoked potentials and visual acuity in amblyopia. Ophthalmologica 213, 110–113.

Bayourthe, C., Enjalbert, F., Moncoulon, R., 2000. Effects of differing forms of canola oil fatty acids pluscanola meal on milk composition and physical properties of butter. J. Dairy Sci. 83, 690–695.

Bell, R.M., 1986. Protein kinase C activation by diacylglycerol second messengers. Cell 45, 631–632.Bell, R.M., Burns, D.J., 1991. Lipid activation of protein kinase C. J. Biol. Chem. 266, 4661–4664.Benolken, R.M., Anderson, R.E., Wheeler, I.G., 1973. Membrane fatty acids associated with the electri-

cal response in visual excitation. Science 182, 1253.Bernoud, N., Fenart, L., Moliere, P., Dehouk, M.P., Lagarde, M., Cecchelli, R., Lecerf, J., 1999.

Preferential transfer of 2-docosahexaenoyl-1-lysophosphatidylcholine through an in vitro blood-brainbarrier over unesterified docosahexaenoic acid. J. Neurochem. 72, 338–345.

Bernsohn, J., Spitz, F.J., 1974. Linoleic and linolenic acid dependency of some brain membrane-boundenzymes after lipid deprivation in rats. Biochem. Biophys. Res. Commun. 57, 293–298.

Bhutta, A.T., Cleves, M.A., Casey, P.H., Cradock, M.M., Anand, K.J.S., 2002. Cognitive and behaviouraloutcomes of school-aged children who were born preterm. J. Amer. Med. Assoc. 288, 728–737.

Birch, D.G., Birch, E.E., Hoffman, D.R., Uauy, R.D., 1992. Retinal development in very-low-birth-weightinfants fed diets differing in omega-3 fatty acids. Invest. Ophthalmol. Vis. Sci. 33, 2365–2376.

Birch, E.E., Garfield, S., Hoffman, D.R., Uauy, R., Birch, D.G., 2000. A randomized controlled trial ofearly dietary supply of long-chain polyunsaturated fatty acids and mental development in terminfants. Dev. Med. Child. Neurol. 42, 174–181.

S. M. Innis260

Birch, E.E., Hoffman, D.R., Castaneda, Y.S., Fawcett, S.L., Birch, D., Uauy, R.D., 2002. A randomizedcontrolled trial of long-chain polyunsaturated fatty acid supplementation of formula in term infantsafter weaning at 6 wk of age. Amer. J. Clin. Nutr. 75, 570–580.

Birch, E.E., Hoffman, D.R., Uauy, R., Birch, D.G., Prestidge, C., 1998. Visual acuity and the essential-ity of docosahexaenoic acid and arachidonic acid in the diet of term infants. Pediatr. Res. 44,201–209.

Bonin, A., Khan, N.A., 2000. Regulation of calcium signaling by docosahexaenoic acid in human T-cells:implication of CRAC channels. J. Lipid. Res. 41, 277–284.

Borgeois, J.P., 1997. Synaptogenesis, heterochrony and epigenesis in the mammalian cerebral cortex.Acta Paediatr. Suppl. 422, 27–33.

Borghese, C.M., Gomez, R.A., Ramirez, O.A., 1993. Phosphatidylserine increases hippocampal synap-tic efficacy. Brain. Res. Bull. 31, 697–700.

Bourre, J.M., Dumont, O., Pascal, G., Durand, G., 1993. Dietary α-linolenic acid at 1.3 g/kg maintainsmaximal docosahexaenoic acid concentration in brain, heart and liver of adult rats. J. Nutr. 123,1313–1319.

Bourre, J.M., Durand, G., Pascal, G., Youyou, A., 1989a. Brain cell and tissue recovery in rats made defi-cient in n-3 fatty acids by alteration of dietary fat. J. Nutr. 119, 15–22.

Bourre, J.M., François, M., Youyou, A., Dumont, O., Piciotti, M., Pascal, G., Durand, G., 1989b. Theeffects of dietary α-linolenic acid on the composition of nerve membranes, enzymatic activity, ampli-tude of electrophysiological parameters, resistance to poisons and performance of learning tasks inrats. J. Nutr. 119, 1880–1892.

Bourre, J.M., Pascal, G., Masson, M., Dumont, O., Piciotti, M., 1984. Alterations in the fatty acidcomposition of rat brain cells (neurons, astrocytes and oligodendrocytes) and of subcellularfractions (myelin and synaptosomes) induced by a diet devoid of n-3 fatty acids. J. Neurochem. 43,342–348.

Bourre, J.M., Piciotti, M., 1992. Δ-6 desaturation of α-linolenic acid in brain and liver during develop-ment and aging in the mouse. J. Neurosci. 6, 65–68.

Bourre, J.M., Piciotti, M., Dumont, O., Pascal, G., Durand, G., 1990. Dietary linoleic acid and polyun-saturated fatty acid acids in rat brain and other organs: minimal requirements of linoleic acid. Lipids25, 465–472.

Brenner, R.R., Peluffo, R.O., 1966. Effect of saturated and unsaturated fatty acids on the desaturation invitro of palmitic, stearic, oleic, linoleic, and linolenic acids. J. Biol. Chem. 241, 5213–5219.

Brenner, R.R., Peluffo, R.O., 1969. Regulation of unsaturated fatty acids biosynthesis. 1. Effect of unsat-urated fatty acids of 18 carbons on the microsomal desaturation of linoleic acid into γ-linolenic acid.Biochim. Biophys. Acta 176, 471–479.

Brenner, R.R., Peluffo, R.O., Nervi, A.M., De Tomas, M.E., 1969. Competitive effect of α- andγ-linolenyl-CoA in linolenyl-CoA desaturation to γ-linolenyl-CoA. Biochim. Biophys. Acta 176,420–422.

Broughton, K.S., Morgan, L.J., 1994. Frequency of (n-3) polyunsaturated fatty acid consumption inducesalterations in tissue lipid composition and eicosanoid synthesis in CD-1 mice. J. Nutr. 124,1104–1111.

Broughton, K.S., Wade, J.W., 2002. Total fat and (n-3, n-6) fat ratios influence eicosanoid production inmice. J. Nutr. 132, 88–94.

Brozoski, T.J., Brown, R.M., Rosvold, H.E., Goldman, P.S. 1979. Cognitive deficit caused by regionaldepletion of dopamine in prefrontal cortex of rhesus monkey. Science 205, 929–932.

Calder, P.C., 1998. Dietary fatty acids and the immune system. Nutr. Rev. 56, 570–583.Calder, P.C., 2001. Polyunsaturated fatty acids, inflammation, and immunity. Lipids 36, 1007–1024.Campbell, F.M., Bush, P.G., Veerkamp, J.H., Dutta-Roy, A.K., 1998a. Detection and cellular localization

of plasma membrane-associated and cytoplasmic fatty acid-binding proteins in human placenta.Placenta 19, 409–415.

Campbell, F.M., Gordon, M.J., Dutta-Roy, A.K., 1996. Preferential uptake of long chain polyunsaturatedfatty acids by isolated placental membranes. Mol. Cell. Biochem. 155, 77–83.

Campbell, F.M., Gordon, M.J., Dutta-Roy, A.K., 1998b. Placental membrane fatty acid-binding proteinpreferentially binds arachidonic and docosahexaenoic acids. Life Sci. 63, 235–240.

Carlson, S.E., Cooke, R.J., Werkman, S.H., Tolley, E.A., 1992. First year growth of preterm infants fedstandard compared to marine-oil n-3 supplemented formula. Lipids 27, 901–907.


Carlson, S.E., Ford, A.J., Werkman, S.H., Peeples, J.M., Koo, W.W.K., 1996b. Visual acuity and fatty acidstatus of term infants fed human milk and formulas with and without docosahexaenoate and arachi-donate from egg yolk lecithin. Pediatr. Res. 39, 882–888.

Carlson, S.E., Werkman, S.H., Peeples, J.M., Cooke, R.J., Tolley, E.A., 1993b. Arachidonic acid statuscorrelates with first year growth in preterm infants. Proc. Natl. Acad. Sci. USA 90, 1073–1077.

Carlson, S.E., Werkman, S.H., Rhodes, P.G., Tolley, E.A., 1993a. Visual-acuity development in healthypreterm infants, effect of marine-oil supplementation. Amer. J. Clin. Nutr. 58, 35–42.

Carlson, S.E., Werkman, S.H., Tolley, E.A., 1996a. Effect of long-chain n-3 fatty acid supplementationon visual acuity and growth of preterm infants with and without bronchopulmonary dysplasia. Amer.J. Clin. Nutr. 63, 687–697.

Carnielli, V.P., Wattimena, D.J., Luijendijk, I.H., Boerlage, A., Degenhart, H.J., Sauer, P.J., 1996. Thevery low birth weight premature infant is capable of synthesizing arachidonic and docosahexaenoicacids from linoleic and linolenic acids. Pediatr. Res. 40, 169–174.

Casamenti, F., Scali, C., Pepeu, G., 1991. Phosphatidylserine reverses the age-dependent decrease incortical acetylcholine release, a microdialysis study. Eur. J. Pharmacol. 194, 11–16.

Ceruku, S.R., Montegomery-Downs, H.E., Farkas, S.L., Thoman, E.B., Lammi-Keefe, C.J., 2002. Highermaternal plasma docosahexaenoic acid during pregnancy is associated with more mature neonatalsleep-state patterning. Amer. J. Clin. Nutr. 76, 608–613.

Chambaz, J., Ravel, D., Manier, M.C., Pepin, D., Mulliez, N., Bereziat, G., 1985. Essential fatty acidsinterconversion in the human fetal liver. Biol. Neonate 47, 136–140.

Chang, M.C.J., Arai, T., Freed, L.M., Wakabayashi, S., Channing, M.A., Dunn, B.B., Der, M.G.,Bell, J.M., Sasaki, T., Herscovitch, P., Eckelman, W.C., Rapoport, S.I., 1997. Brain incorporation of[1-11C]-arachidonate in normocapnic and hypercapnic monkeys, measured with positron emissiontomography. Brain Res. 755, 74–83.

Charnock, J.S., Dryden, W.F., McMurchie, E.J., Abeywardena, M.Y., Russell, G.R., 1983. Differences inthe fatty acid composition of atrial and ventricular phospholipids of rat heart following standard andlipid supplemented diets. Comp. Biochem. Physiol. 75, 47–52.

Chow, C.K. (Ed.), 2000. Fatty Acids in Foods and Their Health Implications, 2nd Edition. MarcelDekker, New York.

Choy, P.C., Tran, K., Hatch, G.M., Kroeger, E.A., 1997. Phospholipid metabolism in the mammalianheart. Prog. Lipid Res. 36, 85–101.

Christensen, J.H., Gustenhoff, P., Korup, E., Aaroe, J., Toft, E., Moller, J., Rasmussen, K., Dyerberg, J.,Schmidt, E.B., 1996. Effect of fish oil on heart rate variability in survivors of myocardial infarction,a double blind randomised controlled trial. Brit. Med. J. 312, 677–678.

Christensen, J.H., Korup, E., Aaroe, J., Toft, E., Moller, J., Rasmussen, K., Dyerberg, J., Schmidt, E.B.,1997. Fish consumption, n-3 fatty acids in cell membranes, and heart rate variability in survivors ofmyocardial infarction with left ventricular dysfunction. Amer. J. Cardiol. 79, 1670–1673.

Chulei, R., Xiafang, I., Hongseng, M., Xiulan, M., Guizheng, I., Gianhong, D., DeFrancesco, C.A.,Connor, W.E., 1995. Milk composition in women from five different regions of China, the great diver-sity of milk fatty acids. J. Nutr. 125, 2993–2998.

Clandinin, M.T., Chappell, J.E., Heim, T., Swyer, P.R., Chance, G.W., 1981. Fatty acid utilization in peri-natal de novo synthesis of tissues. Early Human Dev. 5, 355–366.

Clandinin, M.T., Chappell, J.E., Leong, S., Heim, T., Sawyer, P.R., Chance, G.W., 1980a. Intrauterinefatty acid accretion rates in human brain, implication for fatty acid requirements. Early Human Dev.4, 121–129.

Clandinin, M.T., Chappell, J.E., Leung, S., Heim, T., Swyer, P.R., Chance, G.W., 1980b. Extrauterinefatty acid accretion in infant brain, implications for fatty acid requirements. Early Human Dev. 4,131–138.

Clandinin, M.T., Wong, K., Hacker, R.R., 1985a. Synthesis of chain elongation-desaturation productsof linoleic acid by liver and brain microsomes during development of the pig. Biochem. J. 226,305–309.

Clandinin, M.T., Wong, K., Hacker, R.R., 1985b. Δ5-desaturase activity in liver and brain microsomesduring development of the pig. Biochem. J. 227, 1021–1023.

Clarke, S.D., 2001. Polyunsaturated fatty acid regulation of gene transcription, a molecular mechanismto improve the metabolic syndrome. J. Nutr. 131, 1129–1132.

Clarke, S.D., Armstrong, M.K., Jump, D.B., 1990. Dietary polyunsaturated fats uniquely suppress ratliver fatty acid synthase and s14 mRNA content. J. Nutr. 120, 225–231.

S. M. Innis262

Cohen, S.R., Bernsohn, J., 1978. The in vivo incorporation of linolenic acid into neuronal and glial cellsand myelin. J. Neurochem. 30, 661–669.

Connor, W.E., Lowensohn, R., Hacher, L., 1996. Increased docosahexaenoic acid levels in newborninfants by administration of sardines and fish oil during pregnancy. Lipids 31, S183–S187.

Connor, W.E., Neuringer, M., Lin, D.S., 1990. Dietary effects on brain fatty acid composition, thereversibility of n-3 fatty acid deficiency and turnover of docosahexaenoic acid in brain, erythrocytes,and plasma of rhesus monkeys. J. Lipid Res. 31, 237–247.

Contreras, M.S., Chang, M.C., Rosenberger, T.A., Greiner, R.S., Myers, C.S., Salem, N. Jr., Rapoport, S.I.,2001. Chronic nutritional deprivation of n-3 linolenic acid does not affect n-6 arachidonic acid recyclingwith brain phospholipids in awake rats. J. Neurochem. 79, 1090–1099.

Crawford, M.A., Casperd, N.M., Sinclair, A.J., 1976. The long chain metabolites of linoleic and linolenicacids in liver and brain in herbivores and carnivores. Comp. Biochem. Physiol. 54B, 395–401.

Crawford, M.A., Hassam, A.G., Stevens, P.A., 1981. Essential fatty acid requirements in pregnancy andlactation with special reference to brain development. Prog. Lipid Res. 20, 31–40.

Cunnane, S.C., Menard, C.R., Likhodii, S.S., Brenna, J.T., Crawford, M.C., 1999. Carbon recycling intode novo lipogenesis is a major pathway in neonatal metabolism of linoleate and α-linoleate.Prostagland. Leuk. Essent. Fatty Acids 60, 387–392.

Cunnane, S.C., Williams, S.C., Bell, J.D., Brookes, S., Craig, K., Iles, R.A., Crawford, M.A., 1994.Utilization of uniformly labeled 13C-polyunsaturated fatty acids in the synthesis of long-chain fattyacids and cholesterol accumulating in the neonatal rat brain. J. Neurochem. 62, 2429–2436.

de la Presa Owens, S., Innis, S.M., 1999a. Docosahexaenoic and arachidonic acid reverse changes indopaminergic and serotoninergic neurotransmitters in piglets frontal cortex caused by a linoleic andα-linolenic acid deficient diet. J. Nutr. 129, 2088–2093.

de la Presa Owens, S., Innis, S.M., 1999b. Diverse, region-specific effects of the addition of arachidonicand docosahexaenoic acid to formula with low or adequate linoleic and α-linolenic acid on pigletbrain monoaminergic neurotransmitters. Pediatr. Res. 48, 125–130.

Delion, S., Chalon, S., Guilloteau, D., Besnard, J.-C., Dwand, G., 1996. α-Linolenic acid dietary defi-ciency alters age-related changes of dopaminergic and serotoninergic neurotransmission in the ratfrontal cortex. J. Neurochem. 66, 1582–1591.

Delion, S., Chalon, S., Guilloteau, D., Lejeune, B., Besnard, J.C., Durand, G., 1997. Age-relatedchanges in phospholipid fatty acid composition and monoaminergic neurotransmission in the hip-pocampus of rats fed a balanced or an n-3 polyunsaturated fatty acid-deficient diet. J. Lipid Res. 38,680–689.

Delion, S., Chalon, S., Herault, J., Guilloteau, D., Besnard, J.C., Durand, G., 1994. Chronic dietaryα-linoleic acid deficiency alters dopaminergic and serotoninergic neurotransmitters in rats. J. Nutr.24, 2466–2476.

Del Prado, M., Villalpando, S., Lance, A., Alfonso, E., Demmelmair, H., Koletzko, B., 2000. Contributionof dietary and newly formed arachidonic acid to milk secretion in women on low fat diets. Adv. Exp.Med. Biol. 478, 407–408.

Delton-Vandenbroucke, I., Gammas, P., Anderson, R.E., 1997. Polyunsaturated fatty acid metabolism inretinal and cerebral microvascular endothelial cells. J. Lipid. Res. 38, 147–159.

Dhopeshwarkar, G.A., Subramanian, C., 1975a. Metabolism of linoleic acid in developing brain:I. Incorporation of radioactivity from 1-14C linoleic acid into brain fatty acids. Lipids 10, 238–241.

Dhopeshwarkar, G.A., Subramanian, C., 1975b. Metabolism of 1-14C linolenic acid in developing brain:II. Incorporation of radioactivity from 1-14C linolenate into brain lipids. Lipids 10, 242–245.

Dhopeshwarkar, G.A., Subramanian, C., 1976. Biosynthesis of polyunsaturated fatty acids in thedeveloping brain: I. Metabolic transformation of intracranially administered 1-14C linolenic acid.Lipids 11, 67–71.

Dinh, T.K., Bourre, J.M., Durand, G., 1993. Effect of age and α-linolenic acid deficiency in Δ6 desaturaseactivity and liver lipids in rats. Lipids 28, 517–523.

Dobbing, J., 1972. Vulnerable periods of brain development. In Eliott, K., Kights, J. (Eds.), Lipids,Malnutrition and Developing Brain. A Ciba Foundation Symposium. Excerpta Medica, Amsterdam,pp. 9–20.

Dobbing, J., Sands, J., 1979. Comparative aspects of the brain growth spurt. Early Human Dev. 3, 79–83.Dobbing, J., Hopewell, J.W., Lynch, A., 1971. Vulnerability of developing brain. VII. Permanent deficit

of neurons in cerebral and cerebellar cortex following early mild undernutrition. Exp. Neurol. 32,439–447.


Drane, D.L., Logemann, J.A., 2000. A critical evaluation of the evidence on the assocaition between typeof infant feeding and cognitive development. Paediatr. Perinat. Epidemiol. 14, 349–356.

Duplus, E., Glorian, M., Forest, C., 2000. Fatty acid regulation of gene transcription. J. Biol. Chem. 275,30749–30752.

Dutta-Roy, A.K., 2000. Transport mechanisms for long-chain polyunsaturated fatty acids in the humanaplacenta. Amer. J. Clin. Nutr. 71, 315S–332S.

Edwards, R., Peet, M., Shay, J., Horrobin, D., 1998. Depletion of docosahexaenoic acid in red blood cellmembranes of depressive patients. Biochem. Soc. Trans. 26, S142.

Elias, S.L., Innis, S.M., 2001. Newborn infant plasma trans conjugated linoleic, n-6 and n-3 fatty acidsare related to maternal plasma fatty acids, length of gestation and birth weight and length. Amer.J. Clin. Nutr. 73, 807–814.

Emken, E.A., Adolf, R.O., Gulley, R.M., 1994. Dietary linoleic acid influences desaturation andacylation of deuterium-labeled linoleic and linolenic acids in young adult males. Biochim. Biophys.Acta 1213, 277–288.

Enslen, M., Milon, H., Malnoe, A., 1991. Effect of low intake of n-3 fatty acids during developmenton brain phospholipid fatty acid composition and exploratory behaviour in rats. Lipids 26,203–207.

Faldella, G., Govoni, M., Alessandroni, R., Marchiani, E., Salvioloi, G.P., Biagi, P.L., Spano, C., 1996.Visual evoked potentials and dietary long chain polyunsaturated fatty acids in preterm infats. Arch.Dis. Child. 75, F108–F112.

FAO/WHO (Food and Agriculture Organization, World Health Organization), 1993. Fats and Oilsin Human Nutrition: Report of a Joint Expert Consultation. Food and Nutrition Paper 57,FAO, Rome.

Farquharson, J., Cockburn, F., Patrick, W.A., Jamieson, E.C., Logan, R.W., 1995. Effect of diet on thefatty acid composition of the major phospholipids of infant cerebral cortex. Arch. Dis. Child. 72,198–203.

Favrelière, S., Barrier, L., Darand, G., Chalon, S., Tallrieau, C., 1998. Chronic dietary n-3 polyunsatu-rated fatty acids deficiency affects the fatty acid composition of plasmenylethanolamine andphosphatidylethanolamine differently in rat frontal cortex, striatum, and cerebellum. Lipids 33,401–407.

Ferdinandusse, S., Denis, S., Mooijer, P.A.W., Zhang, Z., Reddy, J.K., Spector, A.A., Wanders, R.J.A.,2001. Identification of the peroxisomal β-oxidation enzymes involved in the biosynthesis of docosa-hexaenoic acid. J. Lipid Res. 42, 1987–1995.

Ferretti, A., Nelson, G.J., Schmidt, P.S., Bartolini, G.L., Kelly, D.S., Flanagan, V.P., 1998. Dietarydocosahexaenoic acid reduces the thromboxane/prostacyclin synthesis ratio in humans. J. Nutr.Biochem. 32, 79–82.

Ferrier, G.R., Redondo, I., Zhu, J., Murphy, M.G., 2002. Differential effects of docosaehexaenoicacid on contractions and L type Ca2+ current in adult cardiomyocytes. Cardiovasc. Res. 54,601–610.

Fidler, N., Sauerwald, T., Pohl, A., Demmelmair, H., Koletzko, B., 2000. Docosahexaenoic acid transferinto human milk after dietary supplementation: a randomized clinical trial. J. Lipid Res. 41,1376–1383.

Fiehn, W., Pewter, J.B., 1971. Lipids and fatty acids of sarcolemma, sarcoplasmic reticulum, and mito-chondria from rat skeletal muscle. J. Biol. Chem. 246, 5617.

Field, C., Clandinin, M.T., van Aerde, J.E., 2001. Polyunsaturated fatty acids and T-cell function: impli-cations for the neonate. Lipids 36, 1025–1032.

Fischer, S., 1989. Dietary polyunsaturated fatty acids and eicosanoid formation in humans. Adv. LipidRes. 23, 169–198.

Fliesler, S.J., Anderson, R.E., 1983. Chemistry and metabolism of lipids in the vertebrate retina. Prog.Lipid Res. 22, 79–131.

Foote, K.D., Hrboticky, N., MacKinnon, M.J., Innis, S.M., 1990. Brain synaptosomal, liver, plasma, andred blood cell lipids in piglets fed exclusively on a vegetable-oil-containing formula with and with-out fish-oil supplements. Amer. J. Clin. Nutr. 51, 1001–1006.

Foreman-van Drangelen, M., van Houwelingen, A.-C., Kester, A.D.M., Hassart, T.H.M., Blanco, C.E.,Hornstra, G., 1995. Long-chain polyunsaturated fatty acids in preterm infants, status at birth and itsfatty acids in preterm infants, status at birth and its influence on postnatal levels. J. Pediatr. 126,611–618.

S. M. Innis264

Frances, H., Coudereau, J.P., Sandouk, P., Clement, M., Monier, C., Bourre, J.M., 1996a. Influence of adietary α-linolenic acid deficiency on learning in the Morris water maze and on the effects of mor-phine. Eur. J. Pharmacol. 298, 217–225.

Frances, H., Monier, C., Clement, M., Lecorsier, A., Debray, M., Bourre, J.M., 1996b. Effect of dietaryα-linolenic acid acid deficiency on habituation. Life Sci. 58, 1805–1816.

Funk, C.D., 2001. Prostaglandins and leukotrienes, advances in eicosanoid biology. Science 294,1871–1875.

Galli, C., Trzeciak, H.I., Paoletti, R., 1971. Effects of dietary fatty acids on the fatty acid composition ofbrain ethanolamine phosphoglyceride, reciprocal replacement of ω6 and ω3 polyunsaturated fattyacids. Biochim. Biophys. Acta 248, 449–454.

Gamoh, S., Hashimoto, M., Sugioka, K., Hossain, M.S., Hata, N., Misawa, Y., Matsumura, S., 1999.Chronic administration of docosahexaenoic acid improves reference memory-related learning abilityin young rats. Neuroscience 93, 237–241

Garcia, M.C., Ward, G., Ma, Y.C., Salem, N. Jr., Kim, H.Y., 1998. Effect of docosahexaenoic acid onthe synthesis of phosphatidylserine in rat brain microsomes and C6 glioma cells. J. Neurochem.70, 24–30.

Gava, V.K., McKean, C.M., 1977. Role of 5-hydroxytryptamine in the modulation of acoustic brainstem(far-field) potentials. Neuropharmacology 16, 447–449.

Gazzah, N., Gharib, A., Craset, M., Bobillier, P., Lagarde, M., Sarda, N., 1995. Decrease in brainphospholipid synthesis in free-moving n-3 fatty acid deficient rats. J. Neurochem. 64, 908–918.

Gerbi, A., Zerouga, M., Debray, A., Durand, A., Chanez, C., Bourre, J.M., 1994. Effect of fish oil on fattyacid compositon of phospholipids of brain membranes and on kinetic properties of Na+,K+-ATPaseisoenzymes of weaned and adult rats. J. Neurochem. 62, 1560–1569.

Gibson, R.A., Neuman, M.A., Makrides, M., 1997. Effect of increasing breast milk docosahexaenoic acidon plasma and erythrocyte phospholipid fatty acids and neural indices of exclusively breast-fedinfants. Eur. J. Clin. Nutr. 51, 578–584.

Giusto, N.M., Pasquare, S.J., Salvador, G.A., Castagnet, P.I., Roque, M.E., Ilincheta de Boschero, M.G.,2000. Lipid metabolism in vertebrate retinal rod outer segments. Prog. Lipid Res. 39, 315–391.

Green, P., Yavin, E., 1995. Modulation of fetal rat brain and liver phospholipid content by intraamnioticethyl docosahexaenoate administration. J. Neurochem. 65, 2555–2560.

Greiner, R.S., Moriguchi, T., Slotnick, B.M., Hutton, A., Salem, N., 2001. Olfactory discriminationdeficits in n-3 fatty acid-deficient rats. Physiol. Behav. 72, 379–385.

Greiner, R.S., Winter, J., Nathanielsz, P.W., Brenna, J.T., 1997. Brain docosahexaenoate accretion in fetalbaboons, bioequivalence of dietary α-linolenic and docosahexaenoate acids. Pediatr. Res. 42,826–834.

Gudbjarnason, S., Doell, B., Oskarsdottir, G., 1978. Docosahexaenoic acid in cardiac metabolism andfunction. Acta Biol. Med. Germ. 37, 777–784.

Guesnet, P., Pugo-Gunsam, P., Maurage, C., Pinault, M., Giraudeau, B., Alessandri, J.-M., Durand, G.,Antoine, J.-M., Couet, C., 1999. Blood lipid concentrations of docosahexaenoic and arachidonic acidsat birth determine their relative postnatal changes in term infants fed breast milk or formula. Amer.J. Clin. Nutr. 70, 292–298.

Haggarty, P., Page, K., Abramovich, D.R., Ashton, J., Brown, D., 1997. Long chain polyunsaturated fattyacid transport across the human placenta. Placenta 18, 635–642.

Hamano, H., Nabekura, J., Nishikawa, M., Ogawa, T., 1996. Docosahexaenoic acid reduces GABAresponse in substantia nigra neuron of rat. J. Neurophysiol. 75, 1264–1270.

Hamazaki, T., Sawazaki, S., Itomura, M., Asaoka, E., Nagao, Y., Nishimura, N., 1996. The effect ofdocosahexaenoic acid on aggression in young adults: a placebo-controlled double blind study. J. Clin.Invest. 97, 1129–1133.

Hamilton, L., Greiner, R., Salem, N. Jr., Kim, H.Y., 2000. n-3 fatty acid deficiency decreases phos-phatidylserine accumulation selectively in neuronal tissues. Lipids 35, 863–869.

Hansen, A.E., Haggard, M.E., Boelsche, A.N., Adam, D.J.D., Wiese, H.F., 1958. Essential fatty acids ininfant nutrition. III. Clinical manifestations of linoleic acid deficiency. J. Nutr. 66, 565–576.

Hargreaves, K.M., Clandinin, M.T., 1987. Phosphatidlylethanolamine methyltransferase, evidence forinfluence of diet fat on selectivity of substrate for methylation in rat brain synaptic plasma mem-branes. Biochim. Biophys. Acta 918, 97–105.

Harris, W.S., Connor, W.E., Lindsay, S., 1984. Will dietary omega-3 fatty acids change the compositionof human milk? Amer. J. Clin. Nutr. 40, 780–785.


Helland, I.B., Saugstad, O.D., Smith, L., Saarem, K., Soluoll, R., Ganes, T., Drevon, C.A., 2001. Similareffects on infants of n-3 and n-6 fatty acids supplementation to pregnant and lactating women.Pediatrics 108, E82–E92.

Henderson, R.A., Jenson, R.G., Lammi-Keefe, C.J., Ferris, A.M., Dardick, K.R., 1992. Effect of fish oilon the fatty acid composition of human milk and maternal and infant erythrocytes. Lipids 27,863–869.

Hibbeln, J.R. 1998. Fish consumption and major depression. Lancet 351, 1213.Hibbeln, J.R., Umhau, J.C., Linnoila, M., George, D.T., Ragan, P.W., Shoaf, S.E., Vaughan, M.R.,

Rawlings, R., Salem, N. Jr., 1998. A replication study of violent and non-violent subjects: cere-brospinal fluid metabolites of serotonin and dopamine are predicted by plasma essential fatty acids.Biol. Psychiatry 44, 243–249.

Holman, R.T., Johnson, S.B., Ogburn, P.L., 1991. Deficiency of essential fatty acids and membrane flu-idity during pregnancy and lactation. Proc. Natl. Acad. Sci. USA 88, 4835–4839.

Hrboticky, N., MacKinnon, M.J., Innis, S.M., 1990. Effect of vegetable oil formula rich in linoleic acidon tissue fatty acid accretion in brain, liver, plasma and erythrocytes of infant piglets. Amer. J. Clin.Nutr. 51, 173–182.

Hrboticky, N., MacKinnon, M.J., Innis, S.M., 1991. Retina fatty acid composition of piglets fed frombirth with a linoleic acid-rich vegetable-oil formula for infants. Amer. J. Clin. Nutr. 53, 483–490.

Hrboticky, N., MacKinnon, M.J., Puterman, M.L., Innis, S.M., 1989. Effect of a linoleic acid infantformula feeding on brain synaptosomal lipid accretion and enzyme thermotropic behavior in thepiglet. J. Lipid Res. 30, 1173–1184.

Huster, D., Arnold, K., Gawrisch, K., 1998. Influence of docosahexaenoic acid and cholesterol on laterallipid organization in phospholipid mixtures. Biochemistry 49, 17299–17308.

Huttenlocher, P.R., Dabholkar, A.S., 1997. Regional differences in synpatogenesis in human cerebralcortex. J. Comp. Neurol. 20, 167–178.

Huttenlocher, P.R., de Courten, C., Garey, L.J., Van der Loos, H., 1982. Synaptogenesis in humanvisual cortex: evidence for synapse elimination during normal development. Neurosci. Lett. 33,247–253.

Ikemoto, A., Nitta, A., Furukawa, S., Ohishi, M., Nakamura, A., Fuji, Y., Okuyama, H., 2000. Dietaryn-3 fatty acid deficiency decreases nerve growth factor content in rat hippocampus. Neurosci. Lett.285, 99–102.

Innis, S.M., 1991. Essential fatty acids in growth and development. Prog. Lipid. Res. 30, 39–103.Innis, S.M., 1992. Human milk and formula fatty acids. J. Pediatr. 120, S56–S61.Innis, S.M., 1996. Essential dietary lipids. In: Ziegler, E.E., Filer, L.J. (Eds.), Present Knowledge in

Nutrtion, 7th Edition. ILSI Press, Washington, DC, pp. 58–66.Innis, S.M., 2003. Perinatal biochemistry and physiology of long chain polyunsaturated fatty acids.

J. Pediatr. 143, 4 Suppl., S1–S8.Innis, S.M., Adamkim, D.H., Hall, R.T., Kalhan, S.C., Lair, C., Lim, M., Stevens, D.C., Twist, P.F.,

Diersen-Schade, D.A., Harris, C.L., Merkel, K.L., Hansen, J.W., 2002. Docosahexaenoic acid andarachidonic acid from single cell triglycerides enhance growth with no adverse effects in preterminfants fed formula. J. Pediatr. 140, 547–554.

Innis, S.M., Auestad, N., Siegman, J.S., 1996. Blood lipid docosahexaenoic acid in term gestation infantsfed formulas with high docosahexaenoic acid, low eicosapentaenoic acid fish oil. Lipids 31, 617–625.

Innis, S.M., de La Presa Owens, S., 2001. Dietary fatty acid composition in pregnancy alters neuritemembrane fatty acids and dopamine in newborn rat brain. J. Nutr. 131, 118–122.

Innis, S.M., Dyer, R.A., 2002. Brain astrocyte synthesis of docosahexaenoic acid from n-3 fatty acids islimited at the elongation of docosapentaenoic acid. J. Lipid Res. 43, 1529–1536.

Innis, S.M., Elias, S.L., 2003. Intakes of n-6 and n-3 fatty acids among pregnant Canadian women. Amer.J. Clin. Nutr. 77, 473–478.

Innis, S.M., Gilley, J., Werker, J., 2001. Are human-milk long-chain polyunsaturated fatty acids relatedto visual and neural development in breast-fed infants? J. Pediatr. 139, 532–538.

Innis, S.M., King, D.J., 1999. Trans fatty acids in human milk are inversely associated with levels ofessential all-cis n-6 and n-3 fatty acids, and determine trans, but not n-6 and n-3 fatty acids in plasmaof breast-fed infants. Amer. J. Clin. Nutr. 70, 383–390.

Innis, S.M., Kuhnlein, H.V., 1988. Long chain n-3 fatty acids in breast milk of Inuit consuming tradi-tional foods. Early Human Dev. 18, 185–189.

S. M. Innis266

Insull, W., Ahrens, E.H., 1959. The fatty acids of human milk. I. Ad libitum maternal diets. J. Biochem.72, 27–33.

Insull, W., Hirsch, J., James, J., Ahrens, E.H., 1959. The fatty acids of human milk. II. Alterationsproduced by manipluation of calorie balance and exchange of dietary fats. J. Clin. Invest. 38,443–450.

IOM (Institute of Medicine of the National Academies), 2002. Dietary Reference Intakes for Energy,Carbohydrate, Fibre, Fat, Fatty Acids, Cholesterol, Protein and Amino Acids. The NationalAcademies Press, Washington, DC.

Itokazu, N., Ikegaya, Y., Nishikawa, M., Matsuki, N., 2000. Bidirectional actions of docosahexaenoicacid on hippocampal neurotransmissions in vivo. Brain Res. 862, 211–216.

Jamieson, E.C., Farquharson, J., Logan, R.W., Howatson, A.G., Patrick, W.J.A., Weaver, L.T., Cockburn, F.,1999. Infant cerebellar gray and white matter fatty acids in relation to age and diet. Lipids 34,1065–1071.

Jensen, C.L., Maude, M., Anderson, R.E., Heird, W.C., 2000. Effect of docosahexaenoic acid supple-mentation of lactating women on the fatty acid composition of breast milk lipids and maternal andinfant plasma phospholipids. Amer. J. Clin. Nutr. 71, 292S–299S.

Jensen, R.G., 1989. The Lipids of Human Milk. CRC Press, Boca Raton, FL.Jensen, R.G., 1999. Lipids in human milk. Lipids 34, 1243–1247.Jensen, R.G., 2002. The composition of bovine milk lipids, January 1995 to December 2000. J. Dairy Sci.

85, 295–350.Jeppensen, P.B., Hoy, C.-E., Mortensen, P.B., 1998. Essential fatty acid deficiency in patients receiving

home parenteral nutrition. Amer. J. Clin. Nutr. 68, 126–133.Jones, C.R., Arai, T., Rapoport, S.I., 1997. Evidence for the involvement of docosahexaenoic acid in

cholinergic stimulated signal transduction at the synapse. Neurochem. Res. 22, 663–670.Jorgensen, M.H., Hernell, O., Hughes, E.L., Michaelsen, K.F., 2001. Is there a relation between docosa-

hexaenoic acid concentration in mothers’ milk and visual development in term infants? J. Pediatr.Gastroenterol. Nutr. 32, 293–296.

Jump, D.B., 2002. Dietary polyunsaturated fatty acids and regulation of gene transcription. Curr. Lipidol.13, 155–164.

Kaminski, W.E., Jendraschak, E., Kieft, R., von Schacky, C., 1993. Dietary omega-3 fatty acidslower levels of platelet derived growth factor mRNA in human mononuclear cells. Blood 81,1871–1879.

Kang, J.X., Leaf, A., 1996. Evidence that free polyunsaturated fatty acids modify Na+ channels bydirectly binding to the channel proteins. Proc. Natl. Acad. Sci. USA 93, 3542–3546.

Khair-El-Din, T., Sicher, S.C., Vasquez, M.A., Chung, G.W., Stallworth, K.A., Kitamura, K., Miller, R.T.,Lu, C.Y., 1996. Transcription of the murine iNOS gene is inhibited by docosahexaenoic acid, a majorconstituent of fetal and neonatal sera as well as fish oil. J. Exp. Med. 183, 1241–1246.

Kinsella, J.E., Lokesh, B., 1990. Dietary lipids, eicosanoids and the immune system. Crit. Care Med. 18,S95–S113.

Kitajka, K., Puskas, L.G., Zvara, A., Hackler, L. Jr., Barcelo-Coblijn, G., Yeo, Y.K., Farkas, T., 2002. Therole of n-3 polyunsaturated fatty acids in brain, modulation of rat brain gene expression by dietaryn-3 fatty acids. Proc. Natl. Acad. Sci. USA 99, 2619–2624.

Kneebone, G.M., Kneebone, R., Gibson, R.A., 1985. Fatty acid composition of breast milk from threeracial groups from Penang, Malaysia. Amer. J. Clin. Nutr. 41, 765–769.

Koenig, B.W., Strey, H.H., Gawrisch, K., 1997. Membrane lateral compressibility determined by NMRand x-ray diffraction, effect of acyl chain polyunsaturation. Biophys. J. 73, 1954–1966

Koletzko, B., Braun, M., 1991. Arachidonic acid and early human growth, is there a relation? Ann. Nutr.Metab. 35, 128–131.

Kristensen, S.D., Schmidt, E.B., Dyerberg, J., 1989 Dietary supplementation with n-3 polyunsaturatedfatty acids and human platelet function with particular emphasis on implications for cardiovasculardisease. J. Intern. Med., Suppl. 225, 141–150.

Lagarde, M., Bernoud, N., Lamaitre-Delaunay, D., Thies, F., Croset, M., Lecerf, J., 2001.Lysophosphatidylcholine as a preferred carrier form of docosahexaenoic acid to the brain. J. Mol.Neurosci. 16, 201–204.

Lampen, A., Meyer, S., Nau, H., 2001. Phytanic acid and docosahexaenoic acid increase the metabolismof all-trans-retinoic acid and CYP 26 gene expression in intestinal cells. Biochim. Biophys. Acta1521, 97–106.


Lamptey, M.S., Walker, B.L., 1976. A possible essential role for dietary linolenic acid in the developmentof the young rat. J. Nutr. 106, 86–93.

Leaf, A., Khang, J.X., Xiao, Y.-F., Billman, G.E., Voskuyl, R.A., 1999. The antiarrhythmic and anticon-vulsant effects of dietary n-3 fatty acids. J. Membrane Biol. 172, 1–11.

Le Doux, M., Rouzeau, A., Bas, P., Sauvant, D., 2002. Occurrence of trans-C18, 1 fatty acid isomers ingoat milk, effect of two dietary regimens. J. Dairy Sci. 85, 190–197.

Leifert, W.R., McMurchie, E.J., Saint, D.A., 1999. Inhibition of cardiac sodium currents in adult ratmyocytes by n-3 polyunsaturated fatty acids. J. Physiol. 520, 571–679.

Le Moal, M., Simon, H., 1991. Mesocorticolimbic dopaminergic networks. Physiol. Rev. 71, 155–234.Levitsky, D.A., Strupp, B.J., 1995. Malnutrtion and the developing brain, changing concepts, changing

concerns. J. Nutr. 125, 2212S–2220S.Levitt, P., 2003. Structural and functional maturation of the developing primate brain. J. Pediatr. 143,

S35–S45.Leysen, J.E., Pauwels, P.J., 1990. 5-HT2 receptors, roles and regulation. Ann. N.Y. Acad. Sci. 600,

183–193.Li, Z., Kaplan, M.L., Hachey, D., 2000. Hepatic microsomal and peroxisomal docosahexaenoate biosyn-

thesis during piglet development. Lipids 35, 1325–1333.Litman, B.J., Mitchell, D.C., 1996. A role for phospholipid polyunsaturation in modulating membrane

protein function. Lipids 31, S193–S197.Litman, B.J., Niu, S.L., Polozova, A., Mitchell, D.C., 2001. The role of docosahexaenoic acid containing

phospholipids in modulating G protein-coupled signaling pathways, visual transduction. J. Mol.Neurosci. 16, 237–242.

LSRO (Life Sciences Research Office), 1998. Fat. In: Raiten, D.J., Talbot, J.M., Waters, J.H. (Eds.),Assessment of Nutrient Requirements for Infant Formulas. LSRO, Bethesda, MD.

LSRO (Life Sciences Research Office), 2001. Fat. In: Klein, C. (Ed.), Assessment of NutrientRequirements for Preterm Infant Formulas. LSRO, Bethesda, MD.

Lucas, A., Stafford, M., Morley, R., Abbott, R., Stephenson, T., MacFayden, U., Elias-Jones, A.,Clements, H., 1999. Efficacy and safety of long-chain polyunsaturated fatty acid supplementation ofinfant-formula milk, a randomized trial. Lancet 354, 1948–1954.

Luthria, D.L., Mohammed, B.S., Sprecher, H., 1996. Regulation of the biosynthesis of 4,7,10,13,16-docosapentaenoic acid. J. Biol. Chem. 271, 16020–16025.

MacDonald, M.L., Rogers, Q.R., Moris, J.G., 1983. Role of linoleate as an essential fatty acid for catindependent of arachidonate synthesis. J. Nutr. 113, 1422–1433.

Maes, M., Christophe, A., Delanghe, J., Altamura, C., Neels, H., Meltzer, H.Y., 1999. Lowered omega-3polyunsaturated fatty acids in serum phospholipids and cholesteryl esters of depressed patients.Psychiatry Res. 85, 275–291.

Magret, V., Elkhalil, L., Nazih-Sanderson, F., Martin, F., Bourre, J.M., Fruchart, J.C., Delbart, C., 1996.Entry of polyunsaturated fatty acids into the brain: evidence that high-density lipoprotein-inducedmethylation of phosphatidylethanolamine and phospholipase A2 are involved. Biochem. J. 316,805–811.

Makrides, M., Neumann, M.A., Byard, R.W., Simmer, K., Gibson, R.A., 1994. Fatty acid composition ofbrain, retina, erythrocytes in breast- and formula-fed infants. Amer. J. Clin. Nutr. 60, 189–194.

Makrides, M., Neumann, M.A., Gibson, R.A., 1996. Effect of maternal docosahexaenoic acid (DHA)supplementation on breast milk composition. Eur. J. Clin. Nutr. 50, 352–357.

Makrides, M., Neumann, M.A., Simmer, K., Gibson, R.A., 1995a. Erythrocyte fatty acids of term infantsfed either breast milk, standard formula, or formula supplemented with long-chain polyunsaturates.Lipids 30, 941–948.

Makrides, M., Neumann, M.A., Simmer, K., Gibson, R.A., 2000. A critical appraisal of the role of dietarylong-chain polyunsaturated fatty acids on neural indices of term infants, a randomized, controlledtrial. Pediatrics 105, 32–38.

Makrides, M., Neumann, M., Simmer, K., Pater, J., Gibson, R., 1995c. Are long-chain polyunsaturatedfatty acids essential nutrients in infancy? Lancet 345, 1463–1468.

Makrides, M., Simmer, K., Neumann, M., Gibson, R., 1995b. Changes in the polyunsaturated fatty acidsof breast milk from mothers of full-term infants over 30 wk of lactation. Amer. J. Clin. Nutr. 61,1231–1233.

Martinez, M., 1992. Tissue levels of polyunsaturated fatty acids in early human development. J. Pediatr.120, 129–138.

S. M. Innis268

Martinez, M., Vazquez, E., 1998. MRI evidence that docosahexaenoic acid ethyl ester improves myeli-nation in generalized peroxisomal disorders. Neurology 51, 26–32.

Martinez, M., Vazquez, E., Garcia de Silva, M.T., Manzanares, J., Bertran, J.M., Castello, F., 2000.Therapeutic effects of docosahexaenoic acid ethyl ester in patients with generalized peroxisomaldisorders. Amer. J. Clin. Nutr. 71, 376S–385S.

Mata de Urquiza, A., Liu, S., Sjoberg, M., Zetterstrom, R.H., Griffiths, W., Sjovall, J., Perlmann, T.,2000. Docosahexaenoic acid, a ligand for the retinoid X receptor in mouse brain. Science 290,2140–2144.

McGahon, B.M., Martin, D.S., Horrobin, D.F., Lynch, M.A., 1999. Age-related changes in synaptic func-tion, analysis of the effect of dietary supplementation with omega-3 fatty acids. Neuroscience 94,305–314.

McLennan, P.L., 2001. Myocardial membrane fatty acids and the antiarrhythmic actions of dietary fishoil in animal models. Lipids 36, S111–S114.

Menard, C.R., Goodman, K.J., Corso, T.N., Brenna, J.T., Cunnane, S.C., 1998. Recycling of carbon intolipids synthesized de novo is a quantitatively important pathway of α-[U-13C] linolenate utilization indeveloping rat brain. J. Neurochem. 71, 2151–2158.

Minami, M., Kimura, S., Endo, T., Hamaue, N., Hirafugi, M., Togashi, H., Matsumoto, M., Yoshioka, M.,Saito, H., Watanabe, S., Kobayashi, T., Okuyama, H., 1997. Dietary docosahexaenoic acid increasescerebral acetylcholine levels and improves passive avoidance performance in stroke-prone sponta-neously hypertensive rats. Pharmacol. Biochem. Behav. 58, 1123–1129.

Mitchell, D.C., Litman, B.J., 1998. Molecular order and dynamics in highly polyunsaturated phospho-lipid bilayers. Biophys. J. 75, 896–908.

Mitchell, D.C., Straume, M., Litman, B.J., 1992. Role of sn-1-saturated, sn-2-polyunsaturated phospho-lipids in control of membrane receptor conformational equilibrium, effects of cholesterol and acylchain unsaturation on the metarhodopsin I in equilibrium with metarhodopsin II equilibrium.Biochemistry 31, 662–670.

Mohammad, B.S., Luthria, D.L., Bakousheva, S.P., Sprecher, H., 1997. Regulation of the biosynthesis of4,7,10,13,16-docosapentaenoic acid. J. Biol. Chem. 326, 425–430.

Mohammed, B.S., Sankarappa, S., Geiger, M., Sprecher, H., 1995. Re-evaluation of the pathway for themetabolism of 7,10,13,16-docosatetraenoic acid to 4,7,10,13,16-docosapentaenoic acid in rat liver.Acta Biochem. Biophys. 317, 179–184.

Mohrhauer, H., Holman, R.T., 1963. The effect of dose level of essential fatty acids upon fatty acidcomposition of the rat liver. J. Lipid Res. 4, 153–159.

Moore, S.A., 1994. Local synthesis and targeting of essential fatty acids at the cellular interface betweenblood and brain: a role for cerebral endothelium and astrocytes in the accretion of CNS docosa-hexaenoic acid. World Rev. Nutr. Diet. 75, 128–133.

Moore, S.A., 2001. Polyunasaturated fatty acid synthesis and release by brain-derived cells in vitro.J. Mol. Neurosci. 16, 195–200.

Moore, S.A., Yoder, E., Murphy, S., Dutton, G.R., Spector, A.A., 1991. Astrocytes, not neurons, producedocosahexaenoic acid (22:6n-3) and arachidonic acid (20:4n-6). J. Neurochem. 56, 518–524.

Morales, M.S., Palmquist, D.L., Weiss, W.P., 2000. Effects of fat source and copper on unsatura-tion of milk and blood triacylglycerol fatty acids in Holstein and Jersey cows. J. Dairy Sci. 83,2105–2111.

Morgane, P.J., Autin-LaFrance, R.J., Bronzinio, J.D., Galler, J.R., 1992. Malnutrition and the developingcentral nervous system. In: Issacson, R.L., Jensen, K.F. (Eds.), The Vulnerable Brain andEnvironmental Risks, Vol. 1: Malnutrition and Hazard Assessment. Plenum Press, New York, pp. 3–44.

Moriguchi, T., Greiner, S.R., Salem, N. Jr., 2000. Behavioural deficits associated with dietary inductionof decreased brain docosahexaenoic acid concentration. J. Neurochem. 75, 2563–2573.

Moriguchi, T., Loewke, T., Garrison, M., Catalan, J., Salem, N. Jr., 2001. Reversal of docosahexaenoicacid deficiency in the rat brain, retina, liver and serum. J. Lipid Res. 42, 419–427.

Mosier, M., Newton, A.C., 1998. Mechanisms of apparent cooperativity in the interaction of proteinkinase C with phosphatidylserine. Biochem. J. 37, 17271–17279.

Murthy, M., Hamilton, J., Greiner, R.S., Moriguchi, T., Salem, N. Jr., Kim, H.Y., 2002. Differentialeffects of n-3 fatty acids on phospholipid molecular species composition in the rat hippocampus.J. Lipid Res. 43, 611–617.

Nair, S.S., Leitch, J.W., Falconer, J., Garg, M.L., 1997. Prevention of cardiac arrhythmia by dietary(n-3) polyunsaturated fatty acids and their mechanism of action. J. Nutr. 127, 383–393.


Neuringer, M., Connor, W.E., 1986. n-3 fatty acids in the brain and retina, evidence for their essentiality.Nutr. Rev. 44, 285–294.

Neuringer, M., Connor, W.E., Lin, D.S., Barstad, L., Luck, S. 1986. Biochemical and functional effectsof prenatal and postnatal ω-3 fatty acid deficiency on retina and brain in rhesus monkeys. Proc.Natl. Acad. Sci. USA 83, 4021–4025.

Neuringer, M., Connor, W.E., Van Petten, C., Barstad, L., 1984. Dietary ω-3 fatty acid deficiency andvisual loss in infant rhesus monkeys. J. Clin. Invest. 73, 272–276.

Nishikawa, M., Kimura, S., Akaike, N., 1994. Facilitatory effect of docosahexaenoic acid onN-methyl-D-aspartate response in pyramidal neurons of rat cerebral cortex. J. Physiol. 475, 83–93.

Ntambi, J.M., Buhrow, S.A., Kaestner, K.H., Christy, R.J., Sibley, E., Kelley, T.J., Lane, M.D., 1988.Differentiation and induced gene expression in 3T3-L1 preadipocytes: characterization of a differen-tially expressed gene encoding stearoyl-CoA desaturase. J. Biol. Chem. 263, 17291–17300.

O’Brien, J.S., Sampson, E.L., 1965. Fatty acid and fatty aldehyde composition of the major brain lipidsin normal human gray matter, white matter, and myelin. J. Lipid Res. 6, 545–551.

O’Connor, D.L., Auestad, N., Jacobs, J., 2001. Growth and development in preterm infants fedlong-chain polyunsaturated fatty acids, a prospective randomized controlled trial. Pediatrics 108,359–372.

Okada, M., Amamoto, T., Tomonaga, M., Kawachi, A., Yazawa, K., Mine, K., Fujiwara, M., 1996. Thechronic administration of docosahexaenoic acid reduces the spatial cognitive deficit followingtransient forebrain ischemia in rats. Neuroscience 71, 17–25.

Ollero, M., Powers, R.D., Alvarez, J.G., 2000. Variation of docosahexaenoic acid content in subsets ofhuman spermatozoa at different stages of maturation: implications for sperm lipoperoxidativedamage. Mol. Reprod. Dev. 55, 326–334.

Olsen, S.F., Hansen, H.S., Secher, N.J., Jensen, B., Sandstrom, B., 1995. Gestation length and birthweight in relation to the intake of marine n-3 fatty acids. Brit. J. Nutr. 73, 397–404.

Olsen, S.F., Hansen, H.S., Sommer, S., Jensen, B., Sorensen, T.I.A., Secher, N.J., Zachariassen, P., 1991.Gestational age in relation to marine n-3 fatty acids in maternal erythocytes, a study of women inFaroe Islands and Denmark. Amer. J. Obstet. Gynecol. 164, 1203–1209.

Olsen, S.F., Sorensen, J.D., Secher, N.J., Hedegaard, M., Henriksen, T.B., Hansen, H., Grant, A., 1992.Randomized controlled trial of effect of fish oil supplementation on pregnancy duration. Lancet 339,1003–1007.

Pakala, R., Sheng, W.L., Benedict, C.R., 2000. Vascular smooth muscle cells preloaded with eicosapen-taenoic acid and docosahexaenoic acid fail to respond to serotonin stimulation. Atherosclerosis 153,47–57.

Park, C.C., Ahmed, Z., 1992. Alterations of plasma membrane fatty acid composition modify the kineticsof Na+ current in cultured rat diencephalic neurons. Brain Res. 570, 75–84.

Parodi, P.W., 1982. Positional distribution of fatty acids in triglycerides from milk of several species ofmammals. Lipids 17, 437–442.

Pawlosky, R.J., Denkins, Y., Ward, G., Salem, N. Jr., 1997. Retinal and brain accretion of long-chainpolyunsaturated fatty acids in developing felines, the effects of corn oil-based maternal diets. Amer.J. Clin. Nutr. 65, 465–472.

Pawlosky, R.J., Hibbeln, J.R., Novotny, J.A., Salem, N. Jr., 2001. Physiological compartmental analysisof α-linolenic acid metabolism in adult humans. J. Lipid Res. 42, 1257–1265.

Poisson, J.P., Dupuy, R.P., Sarda, P., Descomps, B., Narce, M., Rieu, D., Crastes de Paulet, A., 1993.Evidence that liver microsomes of human neonates desaturate essential fatty acids. Biochim. Biophys.Acta 1167, 109–113.

Poling, J.S., Karanian, J.W., Salem, N. Jr., Vincini, S., 1995. Time- and voltage-dependent block of potas-sium channels by docosahexaenoic acid. Mol. Pharmacol. 47, 381–390.

Pond, W.G., Boleman, S.L., Firotto, M.L., Ho, H., Knabe, D.A., Mersmann, H.J., Savell, J.W., Su, D.R.,2000. Perinatal ontogeny of brain growth in the domestic pig. Proc. Soc. Exp. Biol. Med. 223,102–108.

Ponder, D.L., Innis, S.M., Benson, J.D., Siegman, J., 1992. Docosahexaenoic acid status of term infantsfed breast milk or infant formula containing soy oil or corn oil. Pediatr. Res. 32, 683–688.

Pooviah, B.P., Tinoco, J., Lyman, R.L., 1976. Influence of diet on the conversion of 14C-linolenic acid todocosahexaenoic acid in the rat. Lipids 1, 194–202.

Poulos, A., Darin-Bennett, A., White, I.G., 1973. The phospholipid bound fatty acids and aldehydes ofmammalian spermatozoa. Comp. Biochem. Physiol. 46, 541.

S. M. Innis270

Putnam, J.C., Carlson, S.E., Phillip, W., DeVoe, M.D., Barness, L.A., 1982. The effect of variationsin dietary fatty acids on the fatty acid composition of erythrocyte phosphatidylcholine andphosphatidylethanolamine in human infants. Amer. J. Clin. Nutr. 36, 106–114.

Reginato, M.J., Krakow, S.L., Bailey, S.T., Lazar, M.A., 1998. Prostaglandins promote and block adipo-genesis through opposing effects on peroxisome prolierator-activated receptor gamma. J. Biol. Chem.273, 1855–1858.

Reisbick, S., Neuringer, M., Hasnain, R., Connor, W.E., 1990. Polydipsia in rhesus monkeys deficient inomega-3 fatty acids. Physiol. Behav. 47, 315–323.

Reisbick, S., Neuringer, M., Hasnain, R., Connor, W.E., 1994. Home cage behavior of rhesus monkeyswith long-term deficiency of omega-3 fatty acids. Physiol. Behav. 55, 231–239.

Reseland, J.E., Haugen, F., Hollung, K., Solvoll, K., Halvorsen, B., Brude, I.R., Nenseter, M.S.,Christiansen, E.N., Drevon, C.A., 2001. Reduction of leptin gene expression by dietary polyunsatu-rated fatty acids. J. Lipid Res. 42, 743–750.

Rioux, F.M., Innis, S.M., Dyer, R., MacKinnon, M., 1997. Diet-induced changes in liver and bile but notbrain fatty acids can be predicted from differences in plasma phospholipid fatty acids in milk and for-mula fed piglets. J. Nutr. 127, 370–377.

Rivers, J.P.W., Mausan, A.G., Crawford, M.A., Brantell, M.R., 1976. The inability of the lion, Pantheroleo L., to desaturate linoleic acid. FEBS Lett. 67, 269–270.

Rivers, J.P.W., Sinclair, A.J., Crawford, M.A., 1975. Inability of the cat to desaturate essential fatty acids.Nature 258, 171–173.

Rodriguez de Turco, E.B., Parkins, N., Ershov, A.V., Bazan, N.G., 1999. Selective retinal pigment epithe-lial cell lipid metabolism and remodelling conserves photoreceptor docosahexaenoic acid followingphagocytosis. J. Neurosci. Res. 57, 479–486.

Rogers, S., James, K.S., Butland, B.K., 1987. Effects of a fish oil supplement on serum lipids, bloodpressure, bleeding time, hemostatic and rheological variables: a double blind randomized controlledtrial in healthy volunteers. Atherosclerosis 63, 137–143.

Ryan, A.S., Montalto, M.B., Groh-Wargo, S., Mimouni, F., Sentipal-Walerius, J., Doyle, J., Siegman, J.S.,Thomas, A.J., 1999. Effect of DHA-containing formula on growth of preterm infants to 59 weekspostmenstrual age. Amer. J. Human Biol. 11, 457–467.

Salem Jr., N., Vegher, B., Mena, P., Uauy, P., 1996. Arachidonic and docosahexaenoic acids are biosyn-thesized from their 18-carbon precursors in human infants. Proc. Natl. Acad. Sci. USA 93, 45–54.

Sanders, T.A.B., Ellis, F.R., Dickerson, J.W.T., 1978. Studies of vegans, the fatty acid composition ofplasma choline phosphoglycerides, erythrocytes, adipose tissue, and breast milk, and some indicatorsof susceptibility to ischemic heart disease in vegans and omnivore controls. Amer. J. Clin. Nutr. 31,805–813.

Sanders, T.A.B., Rana, S.K., 1987. Comparison of the metabolism of linoleic and linolenic acids in thefetal rat. Ann. Nutr. Metab. 31, 349–353.

San Giovanni, J.P., Parra-Cabrera, S., Colditz, G.A., Berkey, C.S., Dwyer, J.T., 2000. Meta-analysis ofdietary essential fatty acids and long-chain polyunsaturated fatty acids as they relate to visual resolu-tion acuity in healthy preterm infants. Pediatrics 105, 1292–1298.

Sastry, P.S., 1985. Lipids of nervous tissue, composition and metabolism. Prog. Lipid Res. 24, 169–176.Sauerwald, T.U., Hachey, D.L., Jensen, C.L., Chen, H., Anderson, R.E., Heird, W.C., 1996. Effect of

dietary α-linolenic acid intake on incorporation of docosahexaenoic and arachidonic acids intoplasma phospholipids of term infants. Lipids 31, S131–S135.

Sauerwald, T.U., Hachey, D.L., Jensen, C.L., Chen, H., Anderson, R.E., Heird, W.C., 1997. Intermediatesin endogenous synthesis of C22:6ω3 and C20:4ω6 by term and preterm infants. Pediatr. Res.41,183–187.

Shand, J.H., Noble, R.C., 1981. The metabolism of 18:0 and 18:2(n-6) by the ovine placenta at 120 and150 days of gestation. Lipids 16, 68–71.

Sheaff-Greiner, R.C., Zhang, Q., Goodman, K.J., Giussani, D.A., Nathanielsz, P.W., Brenna, J.T., 1996.Linoleate, α-linoleate, and docosahexaenoate recycling into saturated and monounsaturated fattyacids is a major pathway in pregnant or lactating adults and fetal or infant rhesus monkeys. J. LipidRes. 37, 2675–2686.

Simopoulos, A.P., 1999. Evolutionary aspects of omega-3 fatty acids in the food supply. Prostagland.Leuk. Essent. Fatty Acids. 60, 421–429.

Sinclair, A.J., Crawford, M.A., 1972. The incorporation of linolenic acid and docosahexaenoic acid intoliver and brain lipids of developing rats. FEBS Lett. 26, 127–129.


Sinclair, A.J., Crawford, M.A., 1972. The accumulation of arachidonate and docosahexaenoate in thedeveloping rat brain. J. Neurochem. 99, 1753–1758.

Sprecher, H., Chen, Q., Yin, F.Q., 1999. Regulation of the biosynthesis of 22:5n-6 and 22:6n-3, acomplex intracellular process. Lipids 34, S153–S156.

Sprecher, H., Luthria, D.L., Mohammed, B.S., Baykousheva, S.P., 1995. Reevaluation of the pathwaysfor biosynthesis of polyunsaturated fatty acids. J. Lipid Res. 36, 2471–2477.

Stockard, J.E., Saste, M.D., Benford, V.J., Barness, L., Auestad, J.D., 2000. Effect of docosahexaenoicacid content of maternal diet on auditory brainstem conduction times in rat pups. Dev. Neurosci. 22,494–499.

Stoll, A.L., Severus, W.E., Freeman, M.P., Rueter, S., Zboyan, H.A., Diamond, E.A., 1999. Omega-3 fattyacids in bipolar disorder, a preliminary double-blind, placebo-controlled trial. Arch. Gen. Psychiatry56, 407–412.

Su, H.M., Bernardo, M., Mirmiran, X.H., Ma, T.N., Corso, P.W., Nathanielsz, J.T., Brenna, J.T.,1999. Bioequivalence of dietary α-linolenate and docosahexaenoate acids as possible sourcesof docosahexaenoate accretion in brain and associated organs of neonatal baboons. Pediatr. Res.45, 87–93.

Su, H.-M., Huan, M.-C., Saad, N.M.R., Nathanielsz, P.W., Brenna, J.T., 2001. Fetal baboons convert18:3n-3 to 22:6n-3 in vivo, a stable isotope tracer study. J. Lipid. Res. 42, 581–586.

Suh, M., Wierzbicki, A.A., Lien, E., Clandinin, M.T., 1996. Relationship between dietary supply oflong-chain fatty acids and membrane composition of long- and very long chain essential fatty acidsin developing rat photoreceptors. Lipids 313, 61–64.

Suh, M., Wierzbicki, A.A., Lien, E., Clandinin, M.T., 2000. Dietary 20:4n-6 and 22:6n-3 modulates theprofile of long- and very-long-chain essential fatty acids, rhodopsin content, and kinetics in develop-ing photoreceptor cells. Pediatr. Res. 48, 524–530.

Thies, F., Delachambre, M.C., Bentejac, M., Lagarde, M., Lecerf, J., 1992. Unsaturated fatty acidsesterified in 2-acyl-1-lysophosphatidylcholine bound to albumin are more efficiently taken up by theyoung rat brain than the unesterified from. J. Neurochem. 59, 1110–1116.

Thies, F., Pillon, C., Moliere, P., Lagarde, M., Lecerf, J., 1994. Preferential incorporation of sn-2 lysoPCDHA over unesterified DHA in young rat brain. Amer. J. Physiol. 267, R1273–1279.

Tinoco, J., 1982. Dietary requirements and functions of α-linolenic acid in animals. Prog. Lipid Res.21, 1–45.

Tinoco, J., Babcock, R., Hincenbergs, I., Medwadowski, B., Jiljanich, P., Williams, M.A., 1979.Linolenic acid deficiency. Lipids 14, 166–173.

Tremoli, E., Maderna, P., Marangoni, F., Colli, S., Eligini, S., Catalano, I., Angeli, M.T., Pazzucconi, F.,Gianfranceschi, G., Davi, G., 1995. Prolonged inhibition of platelet aggregation after n-3 fatty acidethyl ester ingestion by healthy volunteers. Amer. J. Clin. Nutr. 61, 607–613.

Uauy, R., Birch, D.G., Birch, E.E., 1992. Retinal development in very low birthweight infants fed dietsdiffering in omega-3 fatty acids. Pediatr. Res. 28, 485–492.

Uauy, R., Mena, P., Wegher, B., Nieto, S., Salem, N. Jr., 2000. Long chain polyunsaturated fatty acidformation in neonates, effect of gestational age and intrauterine growth. Pediatr. Res. 47, 127–135.

Vance, D.E., 1990. Phosphatidylcholine metabolism, masochistic enzymology, metabolic regulation, andlipoprotein assembly. Biochem. Cell. Biol. 68, 1151–1165.

van Houwelingen, A.C., Sorensen, J.D., Hornstra, G., Simonis, M.M., Boris, J., Olsen, S.F., Secher, N.J.,1995. Essential fatty acid status in neonates after fish-oil supplementation during late pregnancy.Brit. J. Nutr. 74, 723–731.

von Schacky, C., Weber, P.C., 1985. Metabolism and effects on platelet function of the purifiedeicosapentaenoic and docosahexaenoic acids in humans. J. Clin. Invest. 76, 2446–2450.

Voss, A., Reinhart, M., Sankarappa, S., Sprecher, H., 1991. The metabolism of 7,10,13,16,19-docosapentaenoic acid to 4,7,10,13,16,19-docosahexaenoic acid in rat liver is independent of a4-desaturase. J. Biol. Chem. 226, 19995–20000.

Wainwright, P.E., Xing, H.C., Girard, T., Parker, L., Ward, G.R., 1998. Effects of dietary n-3 fatty aciddeficiency on Morris water-maze performance and amphetamine conditioned place preference in rats.Nutr. Neurosci. 1, 281–293.

Wanders, R.J., Vreken, P., Ferdinandusse, S., Jansen, G.A., Waterham, C.W., Van Grunsven, E.G., 2001.Peroxisomal fatty acid α- and β-oxidation in human enzymology, peroxisomal metabolite trans-porters and peroxisomal diseases. Biochem. Soc. Trans. 29, 250–267.

S. M. Innis272

Wang, N., Anderson, R.E., 1993. Synthesis of docosahexaenoic acid by retina and retina pigment epithe-lium. Biochemistry 32, 13703–13709.

Ward, G.R., Huang, Y.S., Bobik, E., Xing, H.C, Mutsaers, L., Auestad, N., Montalto, M., Wainwright, P.,1998. Long-chain polyunsaturated fatty acid levels in formulae influence deposition of docosa-hexaenoic acid and arachidonic acid in brain and red blood cells of artificially reared neonatal rats.J. Nutr. 128, 2473–2487.

Washizaki, K., Smith, Q.R., Rapoport, S.I., Purdon, A.D., 1994. Brain arachidonic acid incorporation andprecursor pool specific activity during intravenous infusion of unesterified [3H]arachidonate in theanesthetized rat. J. Neurochem. 63, 727–736.

Weiler, H.A., Fitzpatrick-Wong, S., 2002. Dietary long-chain polyunsaturated fatty acids minimizedexamethasone-induced reductions in arachidonic acid status but not bone mineral content in piglets.Pediatr. Res. 51, 282–289.

Weisinger, H.S., Vinigrys, A.J., Bui, B.V., Sinclair, A.J., 1999. Effects of dietary n-3 fatty acid deficiencyand repletion in the guinea pig retina. Invest. Ophthalmol. Vis. Sci. 40, 327–328.

Weisinger, H.S., Vingrys, A.J., Sinclair, J.J., 1995. The effect of docosahexaenoic acid on theelectroretinogram of the guinea pig. Lipids 31, 65–70.

Wertz, P.W., Swartzendruber, D.C., Abraham, W., Madison, K.C., Downing, D.T., 1987 Essential fattyacids and epidermal integrity. Arch. Dermatol. 123, 1381–1384.

Wheeler, T.G., Benolken, R.M., Anderson, R.E., 1975. Visual membranes, specificity of fatty acidprecursors for the electrical response to illumination. Science 188, 1312–1314.

Wiggins, R.C., 1986. Myelination, a critical stage in development. Neurotoxicology 7, 103–120.Willatts, P., Forsyth, P., Forsyth, J.S., DiModugno, M.K., Varna, S., Colvin, M., 1998. Effect of long-

chain polyunsaturated fatty acids in formula on infant problem solving at 10 months of age. Lancet352, 688–691.

Williard, D., Harmon, S., Kaduce, T., Spector, A., 2002. Comparison of 20-, 22-, and 24-carbon n-3 andn-6 polyunsaturated fatty acid utilization in differentiated rat brain astrocytes. Prostagland. Leuk.Essent. Fatty Acids 67, 99–104.

Yamamoto, N., Hashimoto, A., Takemoto, Y., Okuyama, H., Nomura, M., Kitajima, R., Togashi, T.,Tamai, Y., 1988 Effects of the dietary α-linoleate/linoleate balance on lipid compositions and learn-ing ability of rats. II. Discrimination process, extinction process, and glycolipid compositions.J. Lipid Res. 29, 1013–1021.

Yamazaki, K., Fujikawa, W., Hamazaki, T., Yano, S., Shono, T., 1992. Comparison of the conversion ratesof α-linolenic acid (18,3(n - 3)) and stearidonic acid (18,4(n - 3)) to longer polyunsaturated fatty acidsin rats. Biochim. Biophys. Acta 1123, 18–26.

Yonekubo, A., Honda, S., Okano, M., Yamamoto, Y., 1993. Effects of dietary safflower oil or soybean oilon the milk composition of the maternal rat, and tissue fatty acid composition and learning ability ofpostnatal rats. Biosci. Biotech. Biochem. 57, 253–259.

Yoshida, S., Yasuda, A., Kawazato, K., Sakai, K., Shimada, T., Takeshita, Y., Yuasa, S., Kobayashi, T.,Watanabe, S., Okuyama, H., 1997. Synaptic vesicle ultrastructural changes in the rat hippocampusinduced by a combination of α-linolenate deficiency and a learning task. J. Neurochem. 68,1261–1268.

Young, C., Gean, P., Wu, S., Lin, C., Shen, Y., 1998. Cancellation of low-frequency stimulation-induced long-term depression by docosahexaenoic acid in the rat hippocampus. Neurosci. Lett. 247,198–200.

Youyou, A., Durand, G., Pascal, G., Piciotti, M., Dumont, O., Bourre, J.M., 1986. Recovery of alteredfatty acid composition induced by a diet devoid of n-3 fatty acids in myelin, synaptosomes, mito-chondria and microsomes of developing rat brain. J. Neurochem. 46, 224–228.

Ziboh, V.A., Chapkin, R.S., 1988. Metabolism and function of skin lipids. Prog. Lipid Res. 27,81–105.

Zimmer, L., Breton, P., Durand, G., Guilloteau, D., Besnard, J.C., Chalon, S., 1999. Prominent role ofn-3 polyunsaturated fatty acids in cortical dopamine metabolism. Nutr. Neurosci. 2, 257–265.

Zimmer, L., Delion-Vancassel, S., Durand, G., Guilloteau, D., Bodard, S., Besnard, J.C., Chalon, S.,2000a. Modification of dopamine neurotransmission in the nucleus accumbens of rats deficient inn-3 polyunsaturated fatty acids. J. Lipid Res. 41, 32–40.

Zimmer, L., Delpal, S., Guilloteau, D., Aioun, J., Durand, G., Chalon, S., 2000b. Chronic n-3 polyunsat-urated fatty acid deficiency alters dopamine vesicle density in the rat frontal cortex. Neurosci. Lett.284, 25–28.


Zimmer, L., Hembert, S., Durand, G., Breton, P., Guilloteau, D., Besnard, J.-C., Chalon, S., 1998.Chronic n-3 polyunsaturated fatty acid diet-deficiency acts on dopamine metabolism in the rat frontalcortex, a microdialysis study. Neurosci. Lett. 240, 177–181.

Zimmer, L., Vancassel, S., Contagrel, S., Breton, P., Delmanche, S., Guilloteau, D., Durand, G., 2002.The dopamine mesocorticolimbic pathway is affected by deficiency in n-3 polyunsaturated fattyacids. Amer. J. Clin. Nutr. 75, 662–667.

S. M. Innis274

275

11 Development of white adipose tissuelipid metabolism1

H. J. Mersmanna and S. B. Smithb

a USDA/ARS Children’s Nutrition Research Center, Department of Pediatrics,Baylor College of Medicine, Houston, TX 77030, USAb Department of Animal Science, Texas A & M University, College Station,TX 77843-2471, USA

Most mammals are born with little white adipose tissue; however, the guinea pig and humanare exceptions. Limited adipose tissue places the newborn at risk when confronted withenvironmental challenges such as cold temperatures or limited milk supply. White adiposetissue develops rapidly after birth in those species with limited depots. Adipose tissue growthis a combination of cell proliferation coupled with differentiation and cell hypertrophy.Proliferation is a property of the undifferentiated preadipocyte and is predominant in neonataldevelopment. It continues at a lower rate to accommodate growth and replace cells, but canbe activated, even in adults, when caloric intake is excessive and continuous. Preadipocytesdifferentiate into adipocytes with subsequent growth of the adipocyte, increase in mass of thetissue being the result of accumulation of triacylglycerol in a large central intracellular lipiddroplet. Glucose is the carbon precursor of fatty acid synthesis in adipocytes from nonrumi-nant mammals, whereas acetate is the carbon precursor in ruminant species. The human haslittle or no capacity for adipocyte de novo fatty acid synthesis. Because milk is a high-fatfood, there is little fatty acid synthesis during the suckling period. De novo fatty acid synthesisis primarily a process of importance in the postweaning mammal. Triacylglycerol is carriedin lipoproteins and these are cleaved by lipoprotein lipase to yield fatty acids that are readilyabsorbed by adipocytes. Fatty acids are esterified in the adipocyte to triacylglycerol. Thelipoprotein lipase and triacylglycerol biosynthetic activities increase rapidly after birth toallow uptake and esterification of fatty acids. The newborn mammal also needs fatty acids



1 This work is a publication of the USDA/ARS Children’s Nutrition Research Center, Department of Pediatrics, BaylorCollege of Medicine, Houston, Texas. This project has been funded in part with federal funds from the USDA/ARSunder Cooperative Agreement No. 58-6250-6001. The contents of this publication do not necessarily reflect the viewsor policies of the USDA, nor does mention of trade names, commercial products, or organizations imply endorsementby the U.S. Government.

for an oxidative fuel. The source of these fatty acids is from the diet and immediately fromcirculating lipoproteins coupled with fatty acids released from the adipocyte by lipolysis, thebreakdown of adipocyte triacylglycerol.

1. INTRODUCTION

There are two types of adipose tissue in mammals, white and brown. The brown adipocyte isa specialized cell for generation of heat. It has multiple lipid droplets, i.e. stores of triacyl-glycerol (TG), but also a large number of mitochondria to oxidize long-chain fatty acids (FAs)mobilized from TG. Brown adipose tissue is present in most mammalian newborns. It isstrategically located in the body, i.e. in the thorax and near the kidneys, to provide heat toessential organs, by FA mobilization and inefficient oxidation. Brown adipose tissue properties,developmental patterns, and functions are discussed by Smith and Carsten (Chapter 12). Thenewborn pig is different from most newborn mammals in that it has no brown adipose tissue.

The white adipocyte is characterized by a large central lipid droplet that is the repositoryfor storage of energy in the form of TG containing esterified FAs. Most mammalian speciesare born with little white adipose tissue (<5% of the body weight at birth). Most mammalianfetuses have a limited capacity to synthesize fatty acids de novo and the passage of FAs acrossthe placenta is highly restricted (Noble, 1981; Martin et al., 1985; Hamosh, 1998; Kimura,1998; Herrera, 2002). However, the guinea pig and human are born with 10% and 15% bodyfat (Bonnet, 1981; Widdowson and Lister, 1991); in these species, de novo FA synthesisincreases markedly during the last trimester of pregnancy. White adipose tissue has multiplefunctions, but in the neonate, the two most important are for insulation, primarily provided bythe subcutaneous fat depots, and as a repository for energy storage to be mobilized when thenutrient intake does not provide the required energy. The uterine environment provides shelterin the way of temperature control and provision of nutrients, both of which are lost at partu-rition. Mammalian newborns are generally in a precarious position if they are challenged bythe environment, e.g. cold exposure, or if they cannot obtain sufficient nutrients by way of thediet, i.e. by suckling. The newborn pig is at a particular disadvantage because it has <2% bodyfat limiting the insulation and energy supplies, the FA mobilization and oxidation capacity isreduced, there is very little hair (for insulation), and there is no brown adipose tissue(Mersmann, 1974). The neonatal period is characterized by a multitude of metabolic changesas the organism adapts to the many challenges of the environment. The exact timing of a par-ticular adaptation varies among species, but these changes occur during the window of timebetween the last days of gestation and the first days or weeks postpartum. More extensive dis-cussions of metabolism in the neonatal pig are found in Chapter 14 by Herpin et al. andChapter 9 by Odle et al.

The distribution of adipose tissue is different in different species, as are the growth ratesfor the individual depots (Berg and Walters, 1983; Trenkle and Marple, 1983). For example,the predominant adipose tissue depot in the pig is the subcutaneous depot, with lesser fatdeposition at the perirenal, mesenteric/omental, and intermuscular sites (Walstra, 1980; Kauffmanet al., 1986; Kouba et al., 1999; Mitchell et al., 2001). In sheep, the subcutaneous adipose tissuedepot is large, but the intermuscular depot is almost as large and the omental depot is about 50%of the subcutaneous depot (Moloney et al., 2002). In cattle, the subcutaneous depot is large,particularly in breeds that are selected for muscle production and that fatten readily, whereas inbreeds with accentuated lactation rates, the internal fat depots are more extensively developed(Truscott et al., 1983). The metabolic activity and the differentiation of individual adipose tissuedepots may be markedly different (Adams et al., 1997; Wajchenberg, 2000).

H. J. Mersmann and S. B. Smith276

We describe the development of white adipose tissue, and the key metabolic pathways thatprovide the primary functions of adipose tissue, the storage and mobilization of energy. Thereare several major reviews of the literature regarding these subjects in cattle, pigs, and sheep(Allen et al., 1976; Vernon, 1980, 1981; Mersmann, 1986; Smith and Smith, 1995).

2. DEVELOPMENT OF WHITE ADIPOSE TISSUE

A single large central lipid droplet occupying much of the volume of the cell characterizeswhite adipocytes. The functional cytoplasm and nucleoplasm are pushed to the periphery;thus, the relative space occupied by the cytoplasm and nucleus shrinks as the cell stores morelipids. Adipocytes can become extremely large at full development, reaching sizes greaterthan almost any cell in the body. The diameter of completely expanded adipocytes is in excessof 100 μm in most mammalian species and can be 200–500 μm in pigs and cattle.

2.1. Hyperplasia

The expansion of adipose tissue depots involves an increase in cell number coupled with anincrease in the mass of individual adipocytes. The increase in cell number is a major factor inthe increase of mammalian adipose tissue in young mammals (Allen, 1976). Ultimately, thecell number sets limits on the absolute mass of adipose tissue because differentiated adipocytesdo not divide. The hyperplastic process is characteristic of the preadipocyte, the precursor cellthat has not differentiated and begun to fill with TG; it is controlled by numerous factorsincluding age, depot site, sex, and endocrine and growth factors (Hausman et al., 2001). Inmost adult mammals, the direct contribution of hyperplasia to an increase in the mass of adiposetissue is limited. A low rate of hyperplasia and differentiation is required for cell replacementduring the entire life of the organism. However, after the major increase in cell number earlyin the life of a mammal, substantial hyperplastic rates are only measured, in most species,when the organism is exposed to excessive caloric intake.

The increase in the size of an individual adipocyte by accumulation of lipid is limited.When some percentage of the adipocytes reach the size limits for that species, hyperplasia isincreased to provide additional preadipocytes that can differentiate and fill with lipid toexpand the depot. This has clearly been demonstrated in rodents where several adipose tissuedepots (perirenal, perigonadal, inguinal) can be dissected in toto to enable determination ofthe entire number of cells per depot. Also, DNA synthesis can be determined readily in thesesmall animals in individual depots. If the animals are fed excess energy for an extendedperiod, hyperplasia is increased and is readily demonstrated (DiGirolamo and Mendlinger,1971; Greenwood and Hirsch, 1974; Miller et al., 1984; Shillabeer and Lau, 1994). Hyperplasiaand/or DNA synthesis are not readily quantified in larger mammals where the expense ofmeasuring DNA synthesis is considerable and the individual depots are extremely difficult toremove quantitatively, limiting the capability for determination of cell number in the entiredepot (Gurr and Kirtland, 1978).

Attempts to measure total adipocyte number have been reported for depots in cattle andsheep (e.g. Hood and Allen, 1973b; Robelin, 1981; Vernon, 1986) and pigs (e.g. Anderson et al.,1972; Enser et al., 1976; Hood and Allen, 1977; Desnoyers et al., 1980; Hauser et al., 1997),as well as subcutaneous adipose tissue from specific portions of thoracic rib sections in pigs(Demaree et al., 2002) and cattle (Schiavetta et al., 1990). Also, DNA synthesis was demon-strated in young pigs (Gurr et al., 1977; Hausman and Kauffman, 1986a), and in subcutaneousadipose tissue explants from mature cattle (May et al., 1994). Measurement of cell number in

Development of white adipose tissue lipid metabolism 277

a portion of an adipose tissue depot may not reflect changes in cell number in the entire depotbecause cell number in a small sample from a large depot, with expression of number per g tissue,may reflect changes in cell size and not hyperplasia. Likewise, changes in DNA synthesisin explants may not reflect hyperplasia in the entire depot.

2.2. Differentiation

Adipocytes develop from precursor cells, preadipocytes. Preadipocytes can be prepared fromdigested adipose tissue and are obtainable from both young and older mammals. However,the concentration of preadipocytes is greater in neonatal mammals because, at this time, thereis rapid cell division to provide preadipocytes to differentiate and fill with lipids to provideexpansion of the adipose tissue depots. The stromal–vascular cell fraction, isolated fromdigested adipose tissue, contains numerous cell types, e.g. fibroblasts, reticulocytes, endothe-lial cells, blood cells, and preadipocytes. When plated with the appropriate medium, most ofthe attached cells appear as elongated, fibroblastic-like cells. Cell replication is rapid in theproper culture medium with serum being an important component. There are no morphologicalindications that any of these cells are preadipocytes. However, there are proteins characteristicof the preadipocyte, e.g. Pref 1 (Gregoire et al., 1998; Gregoire, 2001), and antigenic materialsdetectable in the preadipocyte that are specific to the adipocyte (Lee et al., 1986; Wright andHausman, 1990; Cryer et al., 1992; Yu et al., 1997). When the appropriate growth factors andhormones are provided, many of the cells in the stromal–vascular fraction differentiate intoadipocytes, i.e. they begin to deposit large amounts of lipid. The amount of differentiationseems to depend on the exact culture conditions. For example, if porcine stromal–vascularcells are differentiated in serum, the extent of differentiation (the total number of differentiatedcells) is limited and differentiation occurs in clusters of cells. If the medium does not containserum, differentiation is more uniform across the culture plate, i.e. there are no clusters, andthe extent of differentiation is greater than in the presence of serum. In clonal lines ofpreadipocytes, derived from rodents, the extent of differentiation is extensive (80% to >90%),even when serum is present. Probably, there are species differences, as well as differencesbetween the clonal cells and the primary preadipocytes directly derived from the adiposetissue stromal–vascular fraction.

Most of the concepts about adipocyte development come from study of clonal cells coupledwith a few studies of primary preadipocytes isolated from rodents. Primary preadipocytes thatdifferentiate in culture under the appropriate conditions have been prepared from many species(Novakofski and Hu, 1987; Suryawan and Hu, 1995), including cattle (Plaas and Cryer, 1980;Cryer et al., 1984; Aso et al., 1995; Ohyama et al., 1998; Torii et al., 1998; Peixing et al., 2000;Wu et al., 2000), pigs (Hausman et al., 1984; Suryawan and Hu, 1993; Boone et al., 2000), rats(Bjorntorp et al., 1980), and sheep (Broad and Ham, 1983; Vierck et al., 1996; Soret et al.,1999; Arana et al., 2002), as well as humans (Hauner et al., 1989). The most extensive studiesusing preadipocytes isolated from a domestic species are with porcine preadipocytes with anoverwhelming contribution by Hausman and coworkers.

The distinguishing morphological feature of a cell beginning to differentiate into an adipocyteis the deposition of small lipid droplets. Seldom do normal cells deposit more than a very fewsmall lipid droplets. In areas where adipose tissue is developing, elongated, fibroblastic-likecells can be observed with multiple small lipid droplets. As the cells accumulate more lipids,the number of droplets increases and droplets fuse to form larger droplets. The adipocytes withmultiple lipid droplets are termed multilocular. Eventually the cell will contain a few largelipid droplets and as lipid deposition continues, fusion of lipid droplets will lead to a cell with


a very large central lipid droplet, a unilocular adipocyte. Unilocular adipocytes also containmany small lipid droplets confined to the peripheral cytoplasmic space. These stages ofadipocyte development are readily observed in the neonatal period, either before or afterbirth, depending on the species (Napolitano, 1963; Slavin, 1985; Cinti, 2001); the pig hasbeen extensively studied (Mersmann et al., 1975; Hausman and Richardson, 1982; Hausmanand Kauffman, 1986b).

Coupling of cell culture systems and molecular biology techniques has led to a model fordifferentiation of preadipocytes to adipocytes (fig. 1). The events that characterize thedifferentiation from the totipotent stem cell to the multipotent mesenchymal precursor cell tothe committed preadipocyte remain largely unknown. However, given a cell that is committedto become an adipocyte, i.e. a preadipocyte, a relatively clear series of sequential events seemsto occur. The exact timing of the events may vary with the species from which the preadipocytewas derived and with the exact conditions used for cell culture. Preadipocytes grown in serumwithout the appropriate adipogenic factors continue to multiply with essentially no differentia-tion. When preadipocytes are presented with the proper stimuli, i.e. an appropriate combinationof growth factors and hormones, the cells begin to differentiate. The factors needed to initiatedifferentiation vary with the cell type and/or the species, e.g. the clonal 3T3-L1 cells requireinsulin, a glucocorticoid, and a cAMP-phosphodiesterase inhibitor, whereas the related clonalcell, 3T3-F442A, requires only insulin. All cells require insulin, most require a glucocorti-coid, many require a phosphodiesterase inhibitor, some require thyroid hormone, etc.(Hausman et al, 1989, 1993; Ramsay et al., 1989; Cryer et al., 1992; Jump and MacDougald,1993). Porcine primary preadipocytes do not require thyroid hormone or a phosphodiesteraseinhibitor; serum impedes the extent of differentiation but does not stop it (Hentges andHausman, 1989; Suryawan and Hu, 1993). Porcine preadipocyte differentiation is suppressedby serum, but if oleic acid is added to the serum-containing medium, almost every cell in theculture differentiates (McNeel and Mersmann, unpublished data). It is important to recognizethat the exact culture conditions dictate the extent of differentiation observed.

An initial differentiation event is the decreased expression of several genes characteristicof the preadipocyte. This is followed by an increase in transcription factors that guide thedevelopment of the adipocyte phenotype (fig. 1). There are multiple reviews of adipocyte


Fig. 1. Simplified model for adipocyte differentiation.

differentiation (e.g. Smas and Sul, 1995; Brun et al., 1996; Loftus and Lane, 1997; Fajaset al., 1998; Morrison and Farmer, 2000; Ntambi and Young-Cheul, 2000; Rangwala andLazar, 2000; Gregoire, 2001). Early in differentiation, there is an increase in the transcriptionfactors CCAAT-enhancer binding proteins beta and delta (C/EBPβ and C/EBPδ). This isfollowed by an increase in another transcription factor, peroxisome proliferator-activatedreceptor gamma (PPARγ), and finally by an increase in C/EBPα. The active form of thePPARγ transcription factor is as a heterodimer with retinoid x receptor alpha (RXRα). Thisheterodimer must be activated by the binding of an appropriate ligand. Fatty acids are potentialligands that stimulate differentiation (Schoonjans et al., 1996; Kliewer et al., 1997). The tran-scription factor, adipocyte determination and differentiation-dependent factor 1 (ADD1),plays a key role in inducing both PPARγ and fatty acid synthase. The provision of FAs by synthase provides potential ligand for activation of PPARγ-RXRα (Kim and Spiegelman,1996). The process of transformation to the adipocyte phenotype, a cell that is primarilygeared to synthesize and mobilize lipid, is guided primarily by PPARγ and C/EBPα (Castilloet al., 1999; Lane et al., 1999; Lazar, 1999).

A number of genes that characterize the adipocyte phenotype have response elements intheir promoter region that bind either PPARγ or C/EBPα. Binding of the appropriate tran-scription factor activates the transcription of that gene (fig. 1). In most cases, an increase inthe transcript for a gene, i.e. the mRNA, results in activation of the translation process with aresultant increase in the protein. The proteins that characterize the adipocyte each have aunique chronological pattern during differentiation, e.g. lipoprotein lipase (LPL) appearsearly in differentiation, whereas adipocyte fatty acid binding protein (aP2) or glucose trans-porter 4 (Glut 4) appear later (fig. 2). Clonal cells, which have been used to develop thismodel, appear to differentiate more synchronously than primary cells so that the chronologicalpatterns may be more clearly observed with clonal cells. Also, there is evidence that primarypreadipocytes are somewhat further along in development at isolation than the clonalpreadipocytes. For example, porcine primary adipocytes have substantial concentrations ofboth the mRNA (Ding et al., 1999) and the protein (Kim et al., 2000) for PPARγ, whereas this


Fig. 2. Development of adipocyte transcripts in dorsal subcutaneous adipose tissue obtained from the neckregion of milk-fed neonatal pigs. Data adapted from Ding et al. (1999) and McNeel et al. (2000).

transcription factor is undetectable or present at very low concentration in clonal preadipocytes(Chawla et al., 1994; Tontonoz et al., 1994). The pattern of development of the mRNAs fornumerous adipocyte transcription factors and adipocyte-characteristic genes has been docu-mented in porcine primary preadipocytes differentiating in vitro (Ding et al., 1999; McNeelet al., 2000). Also, the pattern for development of the proteins for the key transcriptionfactors, PPARγ (Kim et al., 2000) and the C/EBPs (Lee et al., 1998; Yu and Hausman, 1998;Chen et al., 1999), have been described for the differentiating porcine preadipocyte. It isimportant to emphasize that the culture conditions may modify the chronological patternobserved, even with clonal cells.

It is difficult to study adipocyte differentiation in vivo because a tissue sample containscells in many stages of differentiation. The pig offers a modestly reasonable model in vivobecause at birth, there are many preadipocytes present, essentially all adipocytes are multi-locular, and there is rapid transformation of many cells to the unilocular stage over a very fewdays. The tissue is undoubtedly more heterogeneous than clonal preadipocytes, or even primarypreadipocytes differentiating in culture. However, development of several transcripts thatcharacterize adipocyte differentiation or the adipocyte phenotype (Ding et al., 1999; McNeelet al., 2000), follow a pattern that is approximately the same as differentiation of porcinepreadipocytes in culture (fig. 2).

2.3. Hypertrophy

After differentiation of the adipocyte, most of the growth of the cell and, consequently, thetissue is by hypertrophy, which is the lipid filling process. Thus, the average adipocyte size ina given depot increases as the mammal grows, as indicated for ruminants (Hood and Allen,1973b; Allen, 1976; Hood and Thornton, 1979; Hood, 1982; Robelin, 1986; Vernon, 1986)and pigs (Anderson and Kauffman, 1973; Mersmann et al., 1973c; Hood and Allen, 1977;Desnoyers et al., 1980; Hausman, 1985; Hauser et al., 1997). Adipocytes have flattened sidesin vivo, but the shape approximates a sphere. Thus, the diameter is a reasonable expression ofthe cell size. It must be recognized that as the diameter doubles (e.g. from 20 to 40 μm), thevolume increases 8 times (from 4200 μm3 to 33,500 μm3). Measurement of average cell sizewith extrapolation to calculate the number of cells packaged in a defined depot, e.g. in arodent depot, is not unreasonable. However, extrapolation of the average cell size from asmall sample obtained from a large mammal is probably not very valid because it is difficultto quantitatively dissect the depot. This is particularly applicable to the generally contiguoussubcutaneous depot and the mesenteric depot, scattered within the entire gut mesentery.Furthermore, a single size determination on a small sample from a large depot is probablyinvalid because the cell size is not expected to be uniform over the entire depot.

The composition of adipose tissue changes with growth (fig. 3). Most of the changes areattributable to the extensive accumulation of intracellular TG resulting in an increase inadipocyte size. Consequently, the TG expressed per g tissue increases with adipose tissuegrowth. In species of mammals with a marked increase in adipocyte size, the TG concentrationeasily reaches 700 mg per g tissue and may approach 900 mg per g tissue. Because the indi-vidual cells are increasing in size, the number of cells per g tissue is concomitantly decreasing.Consequently, the DNA and protein expressed per g tissue also decreases. The exact chrono-logical pattern for these changes will depend on the species, the capacity of the adipocyte to fillwith lipid, the nutritional status of the animals, and the particular tissue components measured.

To a large extent, adipose tissue growth is governed by the caloric intake of the animal. Excessenergy is deposited as fat. Thus, limits on energy intake limit the growth of adipose tissue.


This is primarily the result of limits on adipocyte hypertrophy, but as indicated earlier, whencaloric intake is elevated chronically, there is usually an increase in cell hyperplasia.

Recent study of adipocyte biology indicates that the adipocyte is an endocrine cell export-ing numerous peptides that have the potential to modify the biological function of theadipocyte and other tissues (Hwang et al., 1997; Kim and Moustaid-Moussa, 2000; Fruhbecket al., 2001; Trayhurn and Beattie, 2001). The first factor discovered was leptin, a cytokine-like peptide that is made and secreted by the adipocyte. There are leptin receptors in thehypothalamus that regulate feeding behavior. Thus, as adipocytes grow and the amount of adi-pose tissue increases, the amount of leptin secreted to the blood increases. The increasedleptin signals the brain to decrease feed intake. There are leptin receptors in other tissues,including adipocytes themselves. Leptin decreases adipocyte lipid anabolism and increaseslipid catabolism in porcine adipocytes (Ramsay, 2001), but not in ovine adipocytes (Newbyet al., 2001).

3. DEVELOPMENT OF LIPID SYNTHESIS

The overall pathways for adipocyte lipid synthesis and degradation were understood decadesago (Bauman, 1976); a simplified scheme is indicated in fig. 4. Glucose or acetate areobtained from the plasma, transformed to acetyl-CoA, then synthesized to FAs. The degrada-tion of lipoproteins by lipoprotein lipase (LPL) provides an additional source of FAs. The FAsare esterified to TG, the major storage lipid. Degradation of TG occurs by lipolysis, yielding3 FAs and glycerol. The overall development of these pathways (fig. 5) indicates that LPLactivity and TG synthesis develop rapidly after birth in the pig during the suckling periodwhen fat deposition is rapid using milk fat as a source of FAs. The degradative activity alsodevelops rapidly after birth to provide energy during times of environmental stress or infre-quent feeding. De novo synthesis of FAs does not substantially increase until after weaning.


Fig. 3. Composition of porcine dorsal subcutaneous adipose tissue obtained from the neck region of suck-ling pigs until day 20, and from pigs weaned at day 21 and fed a low-fat grain-based diet. The 100% valuesare: triacylglycerol (TG) = 900 mg/g tissue; DNA = 1.9 ng DNA phosphorus/g tissue; protein = 71.6 mg/gtissue; collagen = 35.8 mg/g tissue. Data adapted from Mersmann et al. (1973c).

Fig

. 4.

Sim

plif

ied

depi

ctio

n of

adi

pocy

te li

pid

anab

olic

and

cat

abol

ic li

pid

met

abol

ism

. Abb

revi

atio

ns u

sed

are:

AcC

C =

acet

yl-C

oA c

arbo

xyla

se; β

-oxi

datio

n =

mito

chon

dria

l oxi

datio

nof

fat

ty a

cids

; FA

=lo

ng-c

hain

fat

ty a

cid;

FA

S =

fatty

aci

d sy

ntha

se; l

ipol

ysis

=se

quen

tial d

egra

datio

n of

TG

to 3

FA

s +

glyc

erol

; LPL

=lip

opro

tein

lipa

se; O

AA

=ox

aloa

ceta

te; P

=ph

osph

ate;

TC

A c

ycle

=m

itoch

ondr

ial t

rica

rbox

ylic

aci

d cy

cle;

TG

=tr

iacy

lgly

cero

l.

Lipid and glucose metabolism in various tissues undergo multiple adaptations during the peri-natal, suckling, and weaning periods (Girard et al., 1992), as well as during postweaning growth(Saggerson, 1985; Vernon, 1992). There are extensive reviews of the early literature about rumi-nant (Vernon, 1980, 1981; Noble, 1981) and porcine (Mersmann, 1986; Farnsworth and Kramer,1987) adipocyte lipid metabolism, including developmental aspects. Adipose tissue lipid metab-olism, including influences such as genetics, temperature, or diet on its development, continuesto be of interest (Bass et al., 1990; Wood, 1990; Chilliard, 1993) with emphasis on pigs (Le Dividich et al., 1994; Mourot et al., 1995, 1996; Camara et al., 1996; Boone et al., 1999;Gerfault et al., 2000; McNeel and Mersmann, 2000; Robert et al., 2000), cattle (Mendizabal et al., 1999), goats (Bas, 1992), and lambs (e.g. Vernon, 1982; Mendizabal et al., 1997; Purroy et al., 1997; Soret et al., 1998; Payne, 1999; Greathead et al., 2001).

3.1. Sources of fatty acids

The supply of FAs to the organism is from the diet and from de novo FA synthesis. The directsource of FAs for the adipocyte is lipoproteins circulating in the blood plasma, and in somespecies, from de novo synthesis in the adipocyte. Developmental aspects of plasma lipoproteinsin cattle and pigs have been reviewed (Chapman and Forgez, 1985). The major lipoproteinsources of FAs for the adipocyte are chylomicrons, the very large, heavily lipid-laden particlesproduced in the intestine after ingestion of a fatty meal and the very low-density lipoproteins(VLDL). Ruminant species do not produce chylomicrons per se due to their low level of fatintake. Rather, they export VLDL (associated with apolipoprotein B48) from the intestinalmucosal cells (see Chapter 13 by Drackley).


Fig. 5. Ontogeny of porcine dorsal subcutaneous adipose tissue lipid metabolism. Weaning was at 21 dayspostpartum with pigs fed a low-fat diet after weaning. The 100% data are: LPL (lipoprotein lipase) = 6.6 μmolfatty acid released/60 min/106 cells; FA (de novo fatty acid synthesis from glucose) = 661 nmol glucose incor-porated into total lipid/60 min/106 cells; TG (triacylglycerol + diacylglycerol + phospholipid synthesis fromglycerol 3-phosphate) = 1.9 μmol glycerophosphate incorporated/60 min/106 cells; lipolysis = 285 μmol fattyacid released/60 min/106cells. Data adapted from: LPL (Steffen et al., 1978); FA (Mersmann et al., 1973c);TG (Steffen et al., 1979); lipolysis (Mersmann et al., 1976).

Chylomicrons and VLDL have a high concentration of TG, but the adipocyte does notdirectly absorb TG. The TG is cleaved at the endothelial cell surface by the adipocyte-produced enzyme, LPL, and the products of this reaction, the FAs and 2-monoacylglycerols,can be moved to the adipocyte and absorbed. Fatty acids are quickly processed in the cell bya variety of mechanisms because the nonesterified and unbound FA is a detergent and poten-tially quite toxic to the cell. Formation of the coenzyme A thioester is one pathway for furtherprocessing of FAs. Once the CoA derivative is formed (acyl-CoA), the FA can be oxidized orused for various biosynthetic purposes, including synthesis of phospholipids, cholesterolesters, and TG. Nonesterified FAs are bound to several proteins that have specific bindingsites for FAs. In the blood plasma, almost all of the nonesterified FA is bound to albumin,whereas in the cell, there are specific FA binding proteins, among which is adipocyte fattyacid binding protein (aP2). Fatty acid binding has been demonstrated in adipose tissues ofpigs, sheep, and cattle (St. John et al., 1987; Coleman et al., 1988; Miller et al., 1988). Incattle and sheep, depression of de novo fatty acid synthesis is accompanied by depressions infatty acid binding protein activity (Coleman et al., 1988; Miller et al., 1988).

3.2. Synthesis of FA (fig. 4)

Synthesis of FAs from nonlipid sources occurs in the adipocyte of many, but not all, mam-malian species (Vernon and Clegg, 1985; Vernon and Taylor, 1986). Glucose or closelyrelated sugars derived from carbohydrates are the usual precursors of FA carbon in nonrumi-nant species. Lactate and the carbon skeletons derived from nonessential amino acids bytransamination may each be important lipogenic precursors, under some circumstances. Theultimate donor molecule for FA synthesis is the two-carbon donor, acetyl-CoA. Glycolyticmetabolism of glucose to pyruvate is followed by pyruvate entry into the mitochondrion withsubsequent decarboxylation to acetyl-CoA. The mitochondrial acetyl-CoA is not used for FAsynthesis because the enzymatic machinery for FA synthesis is located in the cytosol. Theacetyl-CoA does not traverse the mitochondrial membrane, but is coupled with the four-carbon product of the tricarboxylic acid cycle, oxaloacetate, to yield the six-carbon, citrate.Citrate can traverse the mitochondrial membrane and in the cytosol is converted back toacetyl-CoA plus oxaloacetate by an enzyme, ATP-citrate lyase (citrate cleavage enzyme in theolder literature).

Ruminant species generally use glucose sparingly as a FA precursor because metabolismin the rumen limits the glucose supply to the animal. Ruminants use acetate as the primarycarbon precursor for FA synthesis. De novo FA synthesis is initiated by the carboxylation ofcytosolic acetyl-CoA by acetyl-CoA carboxylase, the tightly regulated and many times rate-limiting enzyme for FA synthesis. The three-carbon product of this reaction, malonyl-CoA, isthen the substrate for fatty acid synthase, the enzyme complex that polymerizes two carbonmoieties into long-chain FAs. One carbon is removed from malonyl-CoA during the poly-merization process. Acetyl-CoA carboxylase limits the rate of de novo fatty acid biosynthesisfrom glucose in porcine adipose tissue and from acetate in bovine adipose tissue. However,glucose incorporation into fatty acids in bovine adipose tissue apparently is limited by theglycolytic pathway, probably at 6-phosphofructokinase (Smith, 1984). Ruminant adipocytesubstrate utilization has been reviewed (Bauman, 1976; Smith, 1995).

The major product of de novo FA synthesis is the 16-carbon saturated FA, palmitic acid(C16:0). (Fatty acids are designated as the number of carbons in the FA chain; C16 = 16 carbons,with the number of double bonds indicated after the colon. This simple nomenclature doesnot indicate the location of the double bonds.) The adipocyte can extend C16:0 by adding


two-carbon moieties in a chain-elongation reaction, for example, to yield the 18-carbonsaturated FA, stearic acid (C18:0). Double bonds can be inserted into the saturated FAs to pro-duce a series of unsaturated FAs. The positions for desaturation are limited in mammals sothat certain FAs are essential and must be obtained from the diet. The enzyme stearoyl-CoAdesaturase uses C18:0 as a substrate to produce the monounsaturated FA, oleic acid (C18:1),with one double bond between the 9 and 10 carbon atoms. This enzyme can also transformother saturated fatty acids to their 9 monounsaturated counterparts (Yang et al., 1999).

One of the essential FAs, obtained from the diet and ultimately from plant products, is the18-carbon FA with two double bonds (cis-9, cis-12), linoleic acid (C18:2). The cis-12 doublebond is at the number 12 carbon counting from the carboxyl group, but at the 6 carbon count-ing from the methyl group of the FA. Thus, C18:2 is the primary member of a series of FAscalled n-6 or omega-6 FAs. Chain elongation and further desaturation of C18:2 gives rise toarachidonic acid (C20:4). The C20:4 is an important constituent of some membranes and aprecursor for many eicosanoid molecules, including prostaglandins, leukotrienes, and throm-boxanes. The other essential FA is the 18-carbon FA with three double bonds, α-linolenic acid(C18:3). This FA is the primary member of the n-3 series of FAs; it has double bonds at the9, 12, and 15 carbons (counting from the carboxyl group) or n-3, 6, and 9 counting from theomega or methyl carbon. Chain elongation and desaturation of this FA produces severaleicosanoid molecules plus eicosapentaenoic acid (C20:5) and docosahexaenoic acid (C22:6),both of which are important FA constituents of some membranes, particularly in the mam-malian central nervous system. Most of these fatty acids are constituents of animal, plant, andmicroorganism lipids; thus the diet, regardless of its composition, is a major source of long-chain FAs. Individual dietary fats are specifically enriched with individual FAs: C16:0 andC18:0 in mammalian meat and milk products, C18:1 in olive oil or some canola oils, C18:2in corn oil or safflower oil, C18:3 in flaxseed or linseed oil, and C20:5 and C22:6 in certainoils from fatty fishes.

3.3. Regulation of FA synthesis

The development of adipocyte de novo FA synthesis has been documented in ruminant andnonruminant mammals. De novo FA synthesis is reciprocally regulated to accommodate thedietary supply of FAs. The measured rate of placental transport of FAs is limited (Leat andHarrison, 1980; Thulin et al., 1989). However, because transport is a continuous process overthe extended period of gestation and fetal de novo FA synthesis rates are low, in utero, FAsare probably supplied primarily by the dam. After birth, the milk supplied to the sucklingmammal has a relatively high fat concentration so that de novo FA synthesis is again notneeded. After weaning, the extent of de novo FA synthesis is dictated by the synthetic andoxidative requirements for FAs coupled with the FA supply from the diet. Most mammalsraised for meat production or for biomedical research are fed a relatively low-fat postwean-ing diet so that de novo FA synthesis is increased at this time. Decreased feed intake andfasting markedly decrease de novo FA synthesis (Mersmann et al., 1981); the timing of thedecrease during fasting is probably dependent on the gut transit time.

Within days of transformation from the high-fat suckling diet to the low-fat postweaningdiet, porcine adipocytes develop increased capacity to synthesize FAs de novo (Mersmann et al., 1973a,b). Enzyme activities associated with de novo FA synthesis, e.g. ATP citratelyase, acetyl-CoA carboxylase, fatty acid synthase, and cytosolic enzymes that producereducing equivalents for the biosynthetic process, e.g. glucose-6 phosphate dehydrogenase,NADP-linked malic enzyme, and NADP-linked isocitrate dehydrogenase, are all increased in


the adipocyte. In cattle, there can be a considerable delay between weaning and the expres-sion of ATP-citrate lyase and acetyl-CoA carboxylase (Smith et al., 1984; Martin et al., 1999).

The manner in which the carbon flux data or the enzyme activities are expressed is impor-tant. As the adipocyte increases in size, there is increased capacity for de novo synthesis ofFAs. Thus, depending on the exact chronology of the increase in adipocyte size, expressionof these metabolic activities on a per g tissue basis can be misleading because the number ofcells per g tissue is decreasing. At some point, the metabolic activity per g tissue willdecrease, when in actuality the activity per adipocyte is increasing (fig. 6). It is important toexpress adipocyte metabolic activity per adipocyte to understand the biology of the adipocyte.To understand the biology of the animal, it would be desirable to know the activity for theentire adipose tissue depot and ultimately for the entire animal. Total depot activity is mostreadily obtained from small laboratory mammals and for selected depots, e.g. the peri-gonadal, the perirenal, or the inguinal fat depots. Each of these depots can be dissected in totoand because the size is not great, measurement of metabolic activity in a sample of the depotcan be extrapolated to the entire depot. Extrapolation is difficult in larger mammals becausedissection of an entire depot is usually difficult, and because of the size of the depot,adipocyte development is not uniform across the depot. This is particularly true for the sub-cutaneous depot that is prevalent in most agricultural species, and for the mesenteric depotthat is prominent in ruminant species. Modes of expression of adipose tissue metabolic dataare discussed for development of porcine adipose tissue (Hood and Allen, 1973a; Mersmannand Brown, 1973).

Many mammals have both a hepatic and an adipocyte capacity for de novo FA biosynthesis,e.g. rats, cattle, and sheep. Because there is little depot fat in the newborn of most mammalianspecies, the predominant site of FA synthesis is initially the liver. As the organism develops,


Fig. 6. Expression of adipocyte enzyme or metabolic data. Lipoprotein lipase activity (LPL) and glyc-erophosphate acyltransferase activity (GPA) expressed per g adipose tissue or per adipocyte. The maximalactivity is indicated as 100%. Data are adapted from: LPL (Steffen et al., 1978) and GPA (Steffen et al., 1979).

with a continuous increase in adipose tissue mass, the adipocytes become more importantfor total body FA synthesis. In the pig, there is little or no capacity for hepatic de novo FAsynthesis at any age so that the adipocyte is the primary tissue source of FAs. In the human(and the chicken) the reverse is true, i.e. there is little adipocyte capacity for de novo synthesisof long-chain FAs.

The extent of de novo FA synthesis is inversely regulated by the dietary fat concentration,so that as the fat is increased in the diet, there is a progressive decrease in de novo FA syn-thesis. Thus, when nonruminant mammals are fed a particular fat source in the diet, the FAcomposition of the fat depots will more or less reflect the composition of the diet. In rumi-nants, the FAs emerging from the rumen are highly saturated and it is difficult to substantiallyincrease the unsaturated FA concentration of adipose tissue depots (Rule et al., 1995).

3.4. Lipoprotein lipase

The alternative source of adipocyte FAs is from hydrolysis of plasma lipoprotein TG.Lipoprotein lipase is synthesized by the adipocyte, and migrates to the surface of the adjacentcapillary endothelial cell where it hydrolyzes lipoprotein to release FAs. Triacylglycerolcannot be transported into the adipocyte, but the FAs can. Adipocyte LPL is regulated suchthat it is active in the fed animal to supply the adipocyte with FAs; the LPL activity decreasesin the fasted animal. Insulin is the primary mediator of this regulation of adipocyte LPL. Inthe newborn pig there is limited adipocyte LPL activity, but this increases rapidly so that afterseveral days there is adequate LPL to hydrolyze the lipoproteins synthesized from the lipidsprovided by the high-fat milk diet (fig. 5). There is a tendency for the LPL activity to begreater in adipose tissue from animals fed high-fat diets compared to low-fat diets, so that inpigs fed grain diets after weaning, the LPL activity is lower than before weaning. Skeletalmuscle LPL generally is regulated in an opposite direction to adipocyte LPL; the muscleenzyme is increased during fasting to increase the supply of FAs to the muscle for oxidation.Measurement of LPL activity, LPL protein, and LPL mRNA in the same animal allows inter-pretation of the regulatory events controlling LPL function (Tavangar et al., 1992). This typeof study has not been done in agricultural species, but enzyme activity or the mRNA havebeen documented in cattle (de la Hoz and Vernon, 1996; Hocquette et al., 2001), pigs (Steffenet al., 1978; McNeel and Mersmann, 2000), and sheep (Andersen et al., 1996; Bonnet et al.,1998, 2000).

3.5. Triacylglycerol synthesis (fig. 4)

As previously indicated, cells cannot accumulate large amounts of nonesterified FAs. Theadipocyte may be considered a sink or reservoir for FA storage with the TG molecule beingthe major storage molecule. Synthesis of TG is a sequential process leading to esterificationof long-chain FAs to each of the three hydroxyl groups of glycerol. The glycerol molecule,serving as the backbone for triacylglycerol synthesis, as well as synthesis of phospholipidmolecules, is derived from glucose via glycolysis. The six-carbon sugar derivative is split intotwo three-carbon moieties, with one of these, dihydroxyacetone phosphate, providing theinitial source of carbons for the glyceride–glycerol moiety. Dihydroxyacetone phosphateis hydrogenated to glycerol-3 phosphate by the enzyme glycerophosphate dehydrogenase.(The activity or mRNA concentration of this enzyme is often used as an indicator foradipocyte differentiation.) Two acyl groups are sequentially added to the glycerophosphate by


the enzyme, α-glycerophosphate acyltransferase. The resulting product is phosphatidic acid.The phosphatidic acid is dephosphorylated by the enzyme, phosphatidate phosphohydrolase,to yield diacylglycerol. Finally, the diacylglycerol is acylated to yield the triacylglycerolmolecule representing the major storage molecule in the adipocyte. Both phosphatidic acidand diacylglycerol are also precursors of phospholipids, so the regulation of TG synthesis ispartly regulated by the demand for the intermediates. Phosphatidate phosphohydrolase isgenerally considered to be a key regulatory enzyme for these pathways, but diacylglycerolacyltransferase has recently emerged as an enzymatic site for regulation.

If FAs are supplied to the adipocyte, there must be adequate TG synthesis capacity to esterifythe FAs. In a mammal like the pig, with an extremely rapid postnatal increase in adipocytehypertrophy, there is an early postnatal demand for esterification of FAs (fig. 5). Enzymes orother aspects of this pathway have seldom been measured in agricultural species (Rule, 1995);however, there are reports for cattle (Lin et al., 1992; Wilson et al., 1992; Smith et al., 1998),pigs (see Mersmann, 1986; Rule et al., 1988a,b, 1989), and lambs (Andersen et al., 1996).

3.6. Endocrine regulation of anabolic metabolism

Regulation of mammalian adipocyte lipid anabolic processes is primarily via adrenergic andinsulin receptors (Etherton and Walton, 1986; Mersmann, 1991; table 1). There are two majortypes of adrenergic receptors, the α- and β-adrenergic receptors (αAR and βAR, respec-tively). In many situations, stimulation of αAR produces actions opposing stimulation ofβAR. The insulin receptor tends to work in opposition to the βAR to provide the majoradipocyte regulatory system. Thus, insulin stimulates the adipocyte anabolic lipid metabolismpathways and βAR agonists inhibit these same pathways. Rodent adipocytes are particularlysensitive to insulin stimulation of anabolism in vitro, with rates of de novo FA synthesis beingincreased 10 or more times by insulin. Anabolic processes in adipocytes from many othermammalian species, including pigs (Romsos et al., 1971; Etherton and Chung, 1981; Waltonand Etherton, 1986, 1987; Mersmann and Hu, 1987; Mersmann, 1989a; Budd et al., 1994; Mills,1999), are stimulated by insulin, but the magnitude is much less than in rodent adipocytes.The bovine adipocyte is relatively insensitive to insulin (Smith et al., 1983; Vasilatos et al.,1983). After a meal, when insulin is elevated, it stimulates lipid synthesis, whereas after fastingthe insulin concentrations are lowered to cause decreased lipid synthesis. The βAR agonistsdecrease anabolic adipocyte lipid metabolism (Rule et al., 1987; Coleman et al., 1988;Mersmann, 1989b, 1995, 1998, 2002a; Miller et al., 1989; Mills et al., 1990; Etherton and Smith,1991; Budd et al., 1994; Moody et al., 2000; Bergen, 2001).

Adipocyte anabolism may also be controlled by other hormones. Decreased thyroid func-tion leads to increased fat accumulation, as does increased adrenocorticoid production. Incontrast, increased somatotropin leads to decreased adipocyte anabolic lipid metabolism(Vernon, 1991; Vernon et al., 1991; Etherton and Louveau, 1992; Harris et al., 1993; Ethertonet al., 1995 Bergen, 2001). Estrogens and androgens have effects on adipocyte function as well,leading to the increased fat deposition in females compared to males, and in some species, sex-dependent differential sites of fat deposition. Most of these additional endocrine regulations donot enter into the day-to-day control of adipocyte lipid metabolism. Rather they provide toniccontrol or long-term modulation. It should be noted that exogenous sex steroids (Hancocket al., 1991), somatotropin (Sejrsen et al., 1989; Beermann and DeVol, 1991; Beermann, 1994),and selected βAR agonists (Moloney et al., 1991) have been used to modify animal growth andcarcass composition (NRC, 1994; Steele, 1991; Steele et al., 1994).


3.7. Conjugated linoleic acid

There is recent evidence that administration of exogenous conjugated linoleic acid can reducecarcass fat deposition in mice, rats, and pigs. The mechanism(s) is not clear, but there is evidencefor increased energy expenditure, increased fat oxidation, decreased preadipocyte prolifera-tion and differentiation, reduced FA synthesis, reduced LPL activity, increased lipolysis(triacylglycerol degradation), and reduced monounsaturated FA production because ofreduced stearoyl-CoA reductase activity (Mersmann, 2002b).

4. DEVELOPMENT OF LIPID DEGRADATION

The process of adipocyte lipid degradation, or lipolysis, is initiated by stimulation of a mem-brane-bound receptor (Belfrage, 1985). Typically this would be a β-adrenergic receptor(βAR) that is coupled to a Gs protein that is coupled to adenylyl cyclase (table 1). Thus, thesequential activation of the receptor, the Gs protein, and adenylyl cyclase yields increasedsynthesis of cAMP from ATP. An increase in intracellular cAMP leads to activation of pro-tein kinase A by attachment of cAMP to the regulatory subunit of the kinase with subsequentcleavage of the catalytic subunit. The kinase catalytic subunit, freed from the regulatory sub-unit, then phosphorylates hormone-sensitive lipase to activate it. Hormone-sensitive lipase isthe rate-limiting enzyme in the process of lipolysis and for the most part is in the nonphos-phorylated state until activated by protein kinase A. Hormone-sensitive lipase cleaves the firsttwo FAs from the TG substrate, whereas the last FA is cleaved by another enzyme, mono-acylglycerol lipase. The latter lipase is continuously active and does not participate in theregulation of lipolysis. The products of lipolysis are the three FAs that were esterified to TGplus glycerol. The FAs may be transported to the plasma or re-enter the adipocyte intra-cellular FA pool where they may be utilized for oxidation or esterification to form complexlipid esters, including the synthesis of TG (fig. 4). Glycerol is not recycled in the adipocytebecause glycerol kinase, the enzyme responsible for phosphorylation of glycerol and itsre-entry into the pathway for synthesis of TG, is present at extremely low concentration inadipocytes.

Typically the lipolytic rate in nonstimulated adipocytes is low, but after maximal stimula-tion with a βAR agonist, the rate is increased several-fold. The extent of stimulation dependson the species, the age of the animal, and the nutritional status. Many of these aspects ofthe regulation lipolysis have been reviewed (Mersmann, 1990, 1991; Chilliard et al., 2000).


Table 1

Major endocrine effects on adipocyte lipid metabolisma

Anabolicb Catabolic

Hormone FA LPL TG Lipolysis

Insulin � � � �β-adrenergic � � � �GH � � � �Glucocorticoids � � � �

a These effects are more or less evident, depending on the species. Broken arrows indicate effects are marginal orreports are mixed regarding the effect.b Abbreviations: FA = de novo fatty acid synthesis; LPL = lipoprotein lipase; TG = triacyglycerol synthesis.

For example, in adipocytes from young rats, lipolysis is stimulated as much as 6 to 10 times,whereas in adipocytes from older rats the increase may be only 3 to 6 times. The degree oflipolytic stimulation by a βAR agonist tends to be greater in rat adipocytes than in adipocytesfrom cattle, chickens, or pigs, wherein the maximal rate is usually only 2 to 5 times that ofthe unstimulated rate. The lipolytic rate is increased in the fasted compared to the fed state inmost species to provide substrate, i.e. FAs for oxidative metabolism.

There are three subtypes of βAR: β1AR, β2AR, and β3AR. These are three distinct proteins,coded by three different genes. Within a mammalian species the homology of the three sub-types is approximately 50%, whereas across species the homology for a single subtypeis ≥75%. Species-specific differences in the protein structure of a βAR subtype can lead tomajor differences in the function of the receptor and in the potency and efficacy of agonistsand antagonists for the receptor. Thus, an antagonist that is specific for the rat β2AR, ICI118,551, is not specific for the porcine β2AR, and a specific agonist for the rat β3AR, BRL37,344, is not specific for the porcine β3AR, but is specific for the porcine β2AR, where it actsas an antagonist (Liang and Mills, 2002). Propranolol is an antagonist for the cloned mouseβ3AR, but a partial agonist for the cloned bovine and human β3AR (Piétri-Rouxel et al.,1995). Thus, it cannot be assumed that the βAR subtype specificity of an agonist or antago-nist in one species will be maintained in another species; the specificity must be tested usingcloned receptors from the individual species of interest (Mills and Mersmann, 1995;Mersmann, 2002a).

The βAR subtypes are differentially distributed in the various tissues of a particularspecies, e.g. rat heart has >90% β1AR, rat lung has >85% β2AR, and rat adipocyteshave >90% β3AR. Receptor distribution in a particular tissue also varies across species, e.g.there are approximately 65% β1AR in porcine heart compared to the 90% in rat heart andthere are approximately 75% β1AR in porcine adipocytes compared to the 90% β3AR in ratadipocytes (McNeel and Mersmann, 1999; Liang and Mills, 2002). In a few cases, there isevidence for a change in adrenergic receptors in a tissue during development. For example,in rat adipocytes the α2AR increased 4-fold between 6 and 20 weeks of age, whereas the βARdecreased 25% in the same time span (Kobatake et al., 1991). In human adipocytes isolatedfrom infants <2 months old, there was more α2AR activity than in adipocytes isolated fromadults (Marcus et al., 1987). Also, the rodent preadipocyte has few or no β3AR with a shiftto ≥90% β3AR during differentiation to adipocytes (Feve et al., 1991). Ultimately the responseof the adipocyte to adrenergic stimulation, in a particular species, will depend on the βARsubtypes present, the ratio of α2AR to βAR on the adipocyte (stimulation of α2ARs inhibitslipolysis), the stage of development of the adipocyte (possibly leading to shifts in receptorsubtypes), and the concentration of norepinephrine and epinephrine, the physiologicalagonists for βAR, to which the adipocyte is exposed (Mersmann, 1998, 2002a).

Depending on the species, there are other membrane-bound receptors that may regulatehormone-sensitive lipase through the cAMP-protein kinase A system. Rat adipocytes havereceptors for adrenocorticotropin, glucagon, somatotropin, thyrotropin, etc. Stimulation ofany of these receptors increases the lipolytic rate. For the most part, these receptors appear tobe minor factors in the regulation of lipolysis in most nonrodent species (Mersmann, 1990,1991; Lanna and Bauman, 1999).

Lipolysis is regulated, not only by receptors that stimulate the process, but also byinhibitory receptors. Stimulation of the α2-adrenergic receptor (α2AR) inhibits lipolysisbecause the α2AR is coupled to the Gi protein, an inhibitory factor. The physiological adrener-gic hormone, epinephrine, has both βAR and αAR activity. Thus, the effect of epinephrine onadipocyte lipolysis depends on the relative populations of α2AR and βAR on the adipocyte.


There are few α2ARs on adipocytes from some species, e.g. rats and pigs, so that this mech-anism of inhibition is not operative (Mersmann, 1990). Perhaps the most importantphysiological negative modulator for lipolysis is insulin (table 1). Increased insulin inhibitslipolysis (Mersmann and Hu, 1987; Mersmann, 1990). Thus, after a meal when insulin con-centration is elevated, lipolysis is decreased. At least part of the mechanism for insulininhibition of lipolysis is stimulation of the activity of cAMP-phosphodiesterase to decreasethe concentration of intracellular cAMP, and consequently decrease the activation of lipolysisthrough protein kinase A and hormone-sensitive lipase. Another negative control of lipolysisis provided by the adenosine receptor (A1R). This receptor also couples to Gi proteins toinhibit lipolysis. Adenosine, the agonist, is produced extracellularly by adipocytes fromcAMP. The process is very active and to measure substantial rates of lipolysis with adipocytesisolated from many species, it is necessary to either inhibit the A1R or to destroy the adenosineby addition of adenosine deaminase. The extent of adenosine formation in vivo is not yetclear, but for many adipocyte preparations in vitro, inhibition of adenosine formation leads toa considerable enhancement of the lipolytic rate (Carey, 1995).

Although subject to several qualifications, the lipolytic activity may be approximatedin vivo by changes in the plasma FA and/or glycerol concentration. The assumption is that thesource of most of the nonesterified FA and glycerol in the plasma is adipocyte lipolysis. Thesetypes of measurements can estimate the extent of stimulation or inhibition by an acutelyinfused compound, but do not reflect the actual rate. They also can be used to estimate adose-response for the compound upon infusion in vivo. Likewise, the nonesterified FA con-centration in the plasma can be used to assess the effects of chronic treatment with a particularcompound on lipolytic activity in vivo. Such approaches have been used to estimate theresponse of adipose tissue to various βAR and αAR agonists and antagonists, and other meta-bolic hormones in cattle, pigs, and sheep (Mersmann, 1987, 1989c, 1995, 1998, 2002a).

Function of the lipolytic process in adipocytes from newborn or young mammals isimpaired relative to function in mature adipocytes. Thus, the βAR-stimulated rate increasesas adipocytes increase in size. Lipolysis is impaired in adipocytes isolated from newborn pigs,but increases several-fold within the first few days of postnatal life (fig. 5; Mersmann, 1986).The ability to mobilize FAs immediately after birth is species-specific and depends on theamount of stored adipose tissue, the chronological development of the enzymatic machineryfor lipolysis, the establishment of the appropriate receptor populations, and the developmentof coupled receptor-driven metabolism.

The development of both anabolic and catabolic adipocyte lipid metabolism primarilyoccurs after birth in many mammalian species, including cattle, pigs, and sheep. The minimalcapacity to mobilize FAs after birth is critical in individuals exposed to cold or that do nothave sufficient milk intake.


Although the goal in modern meat animal production is to produce animal protein with min-imal fat content, the animal raised under conditions where the environment is not optimal andconstant requires fat depots for insulation, to provide oxidative substrates, and to produceselected endocrine materials and growth factors. Mechanisms to decrease adipose tissue lipidmetabolism anabolic processes and to increase catabolic processes have been sought.Somatotropin and βAR agonists are two exogenous compounds that mechanistically operatein this fashion. However, in the neonate, the goal cannot be to decrease the anabolic activityor increase the catabolic activity because the neonate is in a precarious metabolic state with


limited capacity to mobilize fatty acids for oxidative fuels or to synthesize fatty acids foresterification to triacylglycerol. In fact, it might be appropriate to seek to temporarily increasethe rate of fatty acid biosynthesis and lipolysis in neonates. Ideally, if selected genes can beregulated, i.e., turned on and off, it might be possible to enhance selected gene expression atparticular stages of growth and to diminish expression at others.

REFERENCES

Adams, M., Montague, C.T., Prins, J.B., Holder, J.C., Smith, S.A., Sanders, L., Digby, J.E., Sewter, C.P.,Lazar, M.A., Chatterjee, V.K.K., O’Rahilly, S., 1997. Activators of peroxisome proliferator-activatedreceptor have depot-specific effects on human preadipocyte differentiation. J. Clin. Invest. 100,3149–3153.

Allen, C.E., 1976. Cellularity of adipose tissue in meat animals. Fed. Proc. 35, 2302–2307.Allen, C.E., Beitz, D.C., Cramer, D.A., Kauffman, R.G., 1976. Biology of fat in meat animals. North

Central Regional Research Publication No. 234, Research Division, College of Agricultural and LifeSciences, University of Wisconsin-Madison.

Andersen, M.K., Bailey, J.W., Wilken, C., Rule, D.C., 1996. Lipoprotein lipase and glycerophosphateacyltransferase in ovine tissues are influenced by growth and energy intake regimen. J. Nutr.Biochem. 7, 610–616.

Anderson, D.B., Kauffman, R.G., 1973. Cellular and enzymatic changes in porcine adipose tissue duringgrowth. J. Lipid Res. 14, 160–168.

Anderson, D.B., Kauffman, R.G., Kastenschmidt, L.L., 1972. Lipogenic enzyme activities and cellular-ity of porcine adipose tissue from various anatomical locations. J. Lipid. Res. 13, 593–599.

Arana, A., Vernon, R.G., Eguinoa, P., Soret, B., Mendizabal, A., Purroy, A., 2002. Differentiation in vitroof omental and subcutaneous pre-adipocytes from Spanish Lacha and Rasa Aragonesa sheep. Anim.Sci. 74, 469–476.

Aso, H., Abe, H., Nakajima, I., Ozutsumi, K., Yamaguchi, T., Takamori, Y., Kodama, A., Hoshino, F.B.,Takano, S., 1995. A preadipocyte clonal line from bovine intramuscular adipose tissue: nonexpressionof glut-4 protein during adipocyte differentiation. Biochem. Biophys. Res. Commun. 213, 369–375.

Bauman, D.E., 1976. Intermediary metabolism of adipose tissue. Fed. Proc. 35, 2308–2313.Bas, P., 1992. Changes in activities of lipogenic enzymes in adipose tissue and liver of growing goats.

J. Anim. Sci. 70, 3857–3866.Bass, J.J., Butler-Hogg, B.W., Kirton, A.H., 1990. Practical methods of controlling fatness in farm

animals. In: Wood, J.D., Fisher, A.V. (Eds.), Reducing Fat in Meat Animals. Elsevier Applied Science,London, pp. 145–200.

Beermann, D.H., 1994. Carcass composition of animals given partitioning agents. In: Hafs, H.D.,Zimbelman, R.G. (Eds.), Low-Fat Meats Design Strategies and Human Implications. Academic Press,New York, pp. 203–232.

Beermann, D.H., DeVol, D.L., 1991. Effects of somatotropin, somatotropin releasing factor and somato-statin on growth. In: Pearson, A.M., Dutson, T.R. (Eds.), Growth Regulation in Farm Animals.Elsevier Applied Science, London, pp. 373–426.

Belfrage, P., 1985. Hormonal control of lipid degradation. In: Cryer, A., Van, R.L.R. (Eds.),New Perspectives in Adipose Tissue: Structure, Function and Development. Butterworths, London,pp. 121–144.

Berg, R.T., Walters, L.E., 1983. The meat animal: changes and challenges. J. Anim. Sci. 57, Suppl. 2,133–146.

Bergen, W.G., 2001. The role of cyclic AMP elevating agents and somatotropin in pre and posttransla-tional regulation of lipogenesis and lipolysis in Bos taurus and Sus scrofa. Recent Res. Dev. Lipids,5, 47–59.

Bjorntorp, P., Karlsson, M., Pettersson, P., Sypniewska, G., 1980. Differentiation and function of ratadipocyte precursor cells in primary culture. J. Lipid Res. 21, 714–723.

Bonnet, F.P., 1981. Adipose Tissue in Childhood. CRC Press, Boca Raton, FL.Bonnet, M., Faulconnier, Y., Flechet, J., Hocquette, J.F., Leroux, C., Langin, D., Martin, P., Chilliard, Y.,

1998. Messenger RNAs encoding lipoprotein lipase, fatty acid synthase and hormone-sensitive lipasein the adipose tissue of underfed-refed ewes and cows. Reprod. Nutr. Dev. 38, 297–307.


Bonnet, M., Leroux, C., Faulconnier, Y., Hocquette, J.F., Bocquier, F., Martin, P., Chilliard, Y., 2000.Lipoprotein lipase activity and mRNA are up-regulated by refeeding in adipose tissue and cardiacmuscle of sheep. J. Nutr. 130, 749–756.

Boone, C., Gregoire, F., Remacle, C., 1999. Regulation of porcine adipogenesis in vitro, as comparedwith other species. Domest. Anim. Endocrinol. 17, 257–267.

Boone, C., Gregoire, F., Remacle, C., 2000. Culture of porcine stromal-vascular cells in serum-freemedium: differential action of various hormonal agents on adipose conversion. J. Anim. Sci. 78,885–895.

Broad, T.E., Ham, R.G., 1983. Growth and adipose differentiation of sheep preadipocyte fibroblasts inserum-free medium. Eur. J. Biochem. 135, 33–39.

Brun, R.P., Kim, J.B., Hu, E., Altiok, S., Spiegelman, B.M., 1996. Adipocyte differentiation: A tran-scriptional regulatory cascade. Curr. Opin. Cell. Biol. 8, 826–832.

Budd, T.J., Atkinson, J.L., Buttery, P.J., Salter, A.M., Wiseman, J., 1994. Effect of insulin and isopro-terenol on lipid metabolism in porcine adipose tissue from different depots. Comp. Biochem. Physiol.108C, 137–143.

Camara, M., Mourot, J., Fevrier, C., 1996. Influence of two dairy fats on lipid syntheis in the pig: com-parative study of liver, muscle and the two backfat layers. Ann. Nutr. Metab. 40, 287–295.

Carey, G.B., 1995. Adenosine regulation of adipose tissue metabolism. In: Smith, S.B., Smith, D.R.(Eds.), The Biology of Fat in Meat Animals: Current Advances. Amer. Soc. Anim. Sci., Champaign, IL,pp. 35–52.

Castillo, G., Hauser, S., Rosenfield, J.K., Spiegelman, B.M., 1999. Role and regulation of PPARγ duringadipogenesis. J. Anim. Sci. 77, Suppl. 3, 9–15.

Chapman, M.J., Forgez, P., 1985. Lipid transport systems: some recent aspects in swine, cattle and troutduring development. Reprod. Nutr. Dev. 25, 217–226.

Chawla, A., Schwarz, E.J., Dimaculangan, D.D., Lazar, M.A., 1994. Peroxisome proliferator-activatedreceptor (PPAR): adipose-predominant expression and induction early in adipocyte differentiation.Endocrinology 135, 798–800.

Chen, X.L., Dean, R.G., Hausman, G.J., 1999. Expression of leptin MRNA and CCAAT-enhancerbinding proteins in response to insulin deprivation during preadipocyte differentiation in primarycultures of porcine stromal-vascular cells. Domest. Anim. Endocrinol. 17, 389–401.

Chilliard, Y., 1993. Dietary fat and adipose tissue metabolism in ruminants, pigs, and rodents: a review.J. Dairy Sci. 76, 3897–3931.

Chilliard, Y., Ferlay, A., Faulconnier, Y., Bonnet, M., Rouel, J., Bocquier, F., 2000. Adipose tissue metab-olism and its role in adaptations to undernutrition in ruminants. Proc. Nutr. Soc. 59, 127–134.

Cinti, S., 2001. The adipose organ: morphological perspectives of adipose tissues. Proc. Nutr. Soc. 60,319–328.

Coleman, M.E., Ekeren, P.A., Smith, S.B., 1988. Lipid synthesis and adipocyte growth in adipose tissuefrom sheep chronically fed a beta-adrenergic agent. J. Anim. Sci. 66, 372–378.

Cryer, A., Gray, B.R., Woodhead, J.S., 1984. Studies on the characterization of bovine adipocyte precursorcells and their differentiation in vitro, using an indirect-labeled-second-antibody cellular immunoassay.J. Dev. Physiol. 6, 159–176.

Cryer, A., Williams, S.E., Cryer, J., 1992. Dietary and other factors involved in the proliferation, deter-mination and differentiation of adipocyte precursor cells. Proc. Nutr. Soc. 51, 379–385.

de la Hoz, L., Vernon, R.G., 1996. Regulation of lipoprotein lipase activity by growth hormone in differ-entiated adipocytes from sheep. Horm. Metab. Res. 28, 202–204.

Demaree, S.R., Gilbert, C.D., Mersmann, H.J., Smith, S.B., 2002. Conjugated linoleic acid differentiallymodifies fatty acid composition in subcellular fractions of muscle and adipose tissue but not adiposityof postweanling pigs. J. Nutr. 132, 3272–3279.

Desnoyers, F., Pascal, G., Etienne, M., Vodovar, N., 1980. Cellularity of adipose tissue in fetal pig.J. Lipid Res. 21, 301–308.

DiGirolamo, M., Mendlinger, S., 1971. Role of fat cell size and number in enlargement of epididymal fatpads in three species. Amer. J. Physiol. 221, 859–864.

Ding, S.T., McNeel, R.L., Mersmann, H.J., 1999. Expression of porcine adipocyte transcripts: tissuedistribution and differentiation in vitro and in vivo. Comp. Biochem. Physiol. 123B, 307–318.

Enser, M.B., Wood, J.D., Restall, D.J., MacFie, H.J.H., 1976. The cellularity of adipose tissue from pigsof different weights. J. Agr. Sci. Camb. 86, 633–638.


Etherton, T.D., Chung, C.S., 1981. Preparation, characterization, and insulin sensitivity of isolated swineadipocytes: comparison with adipose tissue slices. J. Lipid Res. 22, 1053–1059.

Etherton, T.D., Louveau, I., 1992. Manipulation of adiposity by somatotropin and β-adrenergic agonists:a comparison of their mechanisms of action. Proc. Nutr. Soc. 51, 419–431.

Etherton, T.D., Smith, S.B., 1991. Somatotropin and β-adrenergic agonists: their efficacy and mecha-nisms of action. J. Anim. Sci. 69, 2–26.

Etherton, T.D., Walton, P.E., 1986. Hormonal and metabolic regulation of lipid metabolism in domesticlivestock. J. Anim. Sci. 63, Suppl. 2, 76–88.

Etherton, T.D., Donkin, S.S., Bauman, D.E., 1995. Mechanisms by which porcine somatotropin(pST) decreases adipose tissue growth in growing pigs. In: Smith, S.B., Smith, D.R. (Eds.). TheBiology of Fat in Meat Animals: Current Advances, Amer. Soc. Anim. Sci., Champaign, IL,pp. 53–69.

Fajas, L., Fruchart, J.C., Auwerx, J., 1998. Transcriptional control of adipogenesis. Curr. Opin. Cell. Biol.10, 165–173.

Farnsworth, E.R., Kramer, J.K.G., 1987. Fat metabolism in growing swine: a review. Can. J. Anim. Sci.67, 301–318.

Feve, B., Emorine, L.J., Lasnier, F., Blin, N., Baude, B., Nahmias, C., Strosberg, A.D., Pairault, J., 1991.Atypical β-adrenergic receptor in 3T3-F442A adipocytes. J. Biol. Chem. 266, 20329–20336.

Fruhbeck, G., Gomez-Ambrosi, J., Muruzabal, F.J., Burrell, M.A., 2001. The adipocyte: a model for inte-gration of endocrine and metabolic signaling in energy metabolism regulation. Amer. J. Physiol.Endocrinol. Metab. 280, E827–E847.

Gerfault, V., Louveau, I., Mourot, J., Le Dividich, J., 2000. Lipogenic enzyme activities in subcutaneousadipose tissue and skeletal muscle from neonatal pigs consuming maternal or formula milk. Reprod.Nutr. Dev. 40, 103–112.

Girard, J., Ferre, P., Pegorier, J.P., Duee, P.H., 1992. Adaptations of glucose and fatty acid metabolismduring perinatal period and suckling-weaning transition. Physiol. Rev. 72, 507–562.

Greathead, H.M., Dawson, J.M., Scollan, N.D., Buttery, P.J., 2001. In vivo measurement of lipogenesisin ruminants using [1-14C]acetate. Brit. J. Nutr. 86, 37–44.

Greenwood, M.R., Hirsch, J., 1974. Postnatal development of adipocyte cellularity in the normal rat.J. Lipid Res. 15, 474–483.

Gregoire, F.M., 2001. Adipocyte differentiation: from fibroblast to endocrine cell. Exp. Biol. Med. 226,997–1002.

Gregoire, F.M., Smas, C.M., Sul, H.S., 1998. Understanding adipocyte differentiation. Physiol. Rev.78, 783–809.

Gurr, M.I., Kirtland, J., 1978. Adipose tissue cellularity: a review. I. Techniques for studying cellularity.Int. J. Obes. 2, 401–427.

Gurr, M.I., Kirtland, J., Phillip, M., Robinson, M.P., 1977. The consequences of early overnutrition forfat cell size and number: the pig as an experimental model for human obesity. Int. J. Obes. 1, 151–170.

Hamosh, M., 1998. Neonatal lipid metabolism. In: Cowett, R.M. (Ed.), Principles of Perinatal-NeonatalMetabolism. Springer, New York, pp. 821–846.

Hancock, D.L., Wagner, J.F., Anderson, D.B., 1991. Effects of estrogens and androgens on animalgrowth. In: Pearson, A.M., Dutson, T.R. (Eds.), Growth Regulation in Farm Animals. ElsevierApplied Science, London, pp. 255–297.

Harris, D.M., Dunshea, F.R., Bauman, D.E., Boyd, R.D., Wang, S.Y., Johnson, P.A., Clarke, S.D., 1993.Effect of in vivo somatotropin treatment of growing pigs on adipose tissue lipogenesis. J. Anim. Sci.71, 3293–3300.

Hauner, H., Entenmann, G., Wabitsch, M., Gaillard, D., Ailhaud, G., Negrel, R., Pfeiffer, E.F., 1989.Promoting effect of glucocorticoids on the differentiation of human adipocyte precursor cells culturedin a chemically defined medium. J. Clin. Invest. 84, 1663–1670.

Hauser, N., Mourot, J., De Clercq, L., Genart, C., Remacle, C., 1997. The cellularity of developingadipose tissues in Pietrain and Meishan pigs. Reprod. Nutr. Dev. 37, 617–625.

Hausman, D.B., DiGirolamo, M., Bartness, T.J., Hausman, G.J., Martin, R.J., 2001. The biology of whiteadipocyte proliferation. Obes. Rev. 2, 239–254.

Hausman, G.J., 1985. The comparative anatomy of adipose tissue. In: Cryer, A., Van, R.L.R. (Eds.),New Perspectives in Adipose Tissue: Structure, Function and Development. Butterworths, London,pp. 1–21.


Hausman, G.J., Kauffman, R.G., 1986a. Mitotic activity in fetal and early postnatal porcine adiposetissue. J. Anim. Sci. 63, 659–673.

Hausman, G.J., Kauffman, R.G., 1986b. The histology of developing porcine adipose tissue. J. Anim. Sci.63, 642–658.

Hausman, G.J., Richardson, L.R., 1982. Histochemical and ultrastructural analysis of developingadipocytes in the fetal pig. Acta Anat. 114, 228–247.

Hausman, G.J., Jewell, D.E., Hentges, E.J., 1989. Endocrine regulation of adipogenesis. In: Champion, D.R.,Hausman, G.J., Martin, R.J. (Eds.), Animal Growth Regulation. Plenum Press, New York,pp. 49–68.

Hausman, G.J., Novakofski, J.E., Martin, R.J., Thomas, G.B., 1984. The development of adipocytesin primary stromal-vascular culture of fetal pig adipose tissue. Cell Tissue Res. 236, 459–464.

Hausman, G.J., Wright, J.T., Dean, R., Richardson, R.L., 1993. Cellular and molecular aspects of theregulation of adipogenesis. J. Anim. Sci. 71, Suppl. 2, 33–55.

Hentges, E.J., Hausman, G.J., 1989. Primary cultures of stromal-vascular cells from pig adipose tissue:the influence of glucocorticoids and insulin as inducers of adipocyte differentiation. Domest. Anim.Endocrinol. 6, 275–285.

Herrera, E., 2002. Implications of dietary fatty acids during pregnancy on placental, fetal and postnataldevelopment: a review. Placenta 23, S9–S19.

Hocquette, J.F., Graulet, B., Vermorel, M., Bauchart, D., 2001. Weaning affects lipoprotein lipaseactivity and gene expression in adipose tissues and in masseter but not in other muscles of the calf.Brit. J. Nutr. 86, 433–441.

Hood, R.L., 1982. Relationships among growth, adipose cell size, and lipid metabolism in ruminantadipose tissue. Fed. Proc. 41, 2555–2561.

Hood, R.L., Allen, C.E., 1973a. Comparative methods for the expression of enzyme data in porcineadipose tissue. Comp. Biochem. Physiol. 44B, 677–686.

Hood, R.L., Allen, C.E., 1973b. Cellularity of bovine adipose tissue. J. Lipid Res. 14, 605–610.Hood, R.L., Allen, C.E., 1977. Cellularity of porcine adipose tissue: effects of growth and adiposity.

J. Lipid Res. 18, 275–284.Hood, R.L., Thornton, R.F., 1979. The cellularity of ovine adipose tissue. Aust. J. Agric. Res. 30,

153–161.Hwang, C.S., Loftus, T.M., Mandrup, S., Lane, M.D., 1997. Adipocyte differentiation and leptin expression.

Annu. Rev. Cell. Dev. Biol. 13, 231–259.Jump, D.B., MacDougald, O.A., 1993. Hormonal regulation of gene expression in cultured adipocytes.

J. Anim. Sci. 71, Suppl. 2, 56–64.Kauffman, R.G., Crenshaw, T.D., Rutledge, J.J., Hull, D.H., Grisdale, B.S., Panalba, J., 1986.

Porcine growth: postnatal development of major body components in the boar. In: Universityof Wisconsin-Madison, College of Agricultural and Life Sciences Research Report R3355,pp. 1–25.

Kim, H.S., Hausman, G.J., Hausman, D.B., Martin, R.J., Dean, R.G., 2000. The expression of peroxi-some proliferator-activated receptor gamma in pig fetal tissue and primary stromal-vascular cultures.Obes. Res. 8, 83–88.

Kim, J.B., Spiegelman, B.M., 1996. ADD1/SREBP1 promotes adipocyte differentiation and gene expres-sion linked to fatty acid metabolism. Genes Dev. 10, 1096–1107.

Kim, S., Moustaid-Moussa, N., 2000. Secretory, endocrine and autocrine / paracrine function of theadipocyte. J. Nutr. 130, 3110S–3115S.

Kimura, R.E., 1998. Lipid metabolism in the fetal-placental unit. In: Cowett, R.M. (Ed.), Principles ofPerinatal-Neonatal Metabolism. Springer, New York, pp. 389–402.

Kliewer, S.A., Sundseth, S.S., Jones, S.A., Brown, P.J., Wisely, G.B., Koble, C.S., Devchand, P., Wahli, W.,Willson, T.M., Lenhard, J.M., Lehmann, J.M., 1997. Fatty acids and eicosanoids regulate geneexpression through direct interactions with peroxisome proliferator-activated receptors. Proc. Natl.Acad. Sci. USA 94, 4318–4323.

Kobatake, T., Watanabe, Y., Matsuzawa, Y., Tokunaga, K., Fujioka, S., Kawamoto, T., Keno, Y., Tarui, S.,Yoshida, H., 1991. Age-related changes in adrenergic alpha 1, alpha 2, and beta receptors of rat whitefat cell membranes: an analysis using [3H]bunazosin as a novel ligand for the alpha 1 adrenoceptor.J. Lipid Res. 32, 191–196.

Kouba, M., Bonneau, M., Noble, J., 1999. Relative development of subcutaneous, intermuscular, andkidney fat in growing pigs with different body compositions. J. Anim. Sci. 77, 622–629.


Lane, M.D., Tang, Q.Q., Jiang, M.S., 1999. The adipocyte differentiation program: repression/depressionof the C/EBPα gene. J. Anim. Sci. 77, Suppl. 3, 23–32.

Lanna, D.P.D., Bauman, D.E., 1999. Effect of somatotropin, insulin, and glucocorticoid on lipolysis inchronic cultures of adipose tissue from lactating cows. J. Dairy Sci. 82, 60–68.

Lazar, M.A., 1999. PPARγ in adipocyte differentiation. J. Anim. Sci. 77, Suppl. 3, 16–22.Leat, W.M.F., Harrison, F.A., 1980. Transfer of long-chain fatty acids to the fetal and neonatal lamb.

J. Dev. Physiol. 2, 257–274.Le Dividich, J., Herpin, P., Mourot, J., Colin, A.-P., 1994. Effect of low-fat colostrums on fat accretion and

lipogenic enzyme activities in adipose tissue in the 1-day-old pig. Comp. Biochem. Physiol. 108A,663–671.

Lee, K., Hausman, G.J., Dean, R.G., 1998. Expression of C/EBP alpha, beta and delta in fetal and post-natal subcutaneous adipose tissue. Mol. Cell. Biochem. 178, 269–274.

Lee, S.R., Tume, R.K., Cryer, J., Cryer, A., 1986. Studies on the expression of adipocyte-specific cellsurface antigens during the differentiation of adipocyte precursor cells in vitro. J. Dev. Physiol.8, 207–226.

Liang, W., Mills, S.E., 2002. Quantitative analysis of beta-adrenergic receptor subtypes in pig tissues.J. Anim. Sci. 80, 963–970.

Lin, K.C., Cross, H.R., Smith, S.B., 1992. Esterification of fatty acids by bovine intramuscular andsubcutaneous adipose tissues. Lipids 27, 111–116.

Loftus, T.M., Lane, M.D., 1997. Modulating the transcriptional control of adipogenesis. Curr. Opin.Genet. Dev. 7, 603–608.

Marcus, C., Karpe, B., Bolme, P., Sonnenfeld, T., Arner, P., 1987. Changes in catecholamine inducedlipolysis in isolated human fat cells during the first year of life. J. Clin. Invest. 79, 1812–1818.

Martin, G.S., Lunt, D.K., Britain, K.G., Smith, S.B., 1999. Postnatal development of stearoyl coenzymeA desaturase gene expression and adiposity in bovine subcutaneous adipose tissue. J. Anim. Sci. 77,630–636.

Martin, R.J., Kasser, T.R., Ramsay, T.G., Hausman, G.J., 1985. Regulation of adipose tissue developmentin utero. In: Cryer, A., Van, R.L.R. (Eds.), New Perspectives in Adipose Tissue: Structure, Functionand Development. Butterworths, London, pp. 303–317.

May, S.G., Savell, J.W., Lunt, D.K., Wilson, J.J., Laurenz, J.C., Smith, S.B., 1994. Evidence forpreadipocyte proliferation during culture of subcutaneous and intramuscular adipose tissues fromAngus and Wagyu crossbred steers. J. Anim. Sci. 72, 3110–3117.

McNeel, R.L., Mersmann, H.J., 1999. Distribution and quantification of beta1-, beta2-, and beta3-adrenergic receptor subtype transcripts in porcine tissues. J. Anim. Sci. 77, 611–621.

McNeel, R.L., Mersmann, H.J., 2000. Nutritional deprivation reduces the transcripts for transcriptionfactors and adipocyte-characteristic proteins in porcine adipocytes. J. Nutr. Biochem. 11,139–146.

McNeel, R.L., Ding, S., O’Brian, Smith, E., Mersmann, H.J., 2000. Expression of porcine adipocytetranscripts during differentiation in vitro and in vivo. Comp. Biochem. Physiol. 126B, 291–302.

Mendizabal, J.A., Alberti, P., Eguinoa, P., Arana, A., Soret, B., Purroy, A., 1999. Adipocyte size andenzyme lipogenic activities in different adipose tissue depots in steers of local Spanish breeds. Anim.Sci. 69, 115–121.

Mendizabal, J.A., Soret, B., Purroy, A., Arana, A., Horcada, A., 1997. Infuence of sex on cellularity andon lipogenic enzymes of Spanish lamb breeds (Lacha and Rasa Aragonesa). Anim. Sci. 64, 283–289.

Mersmann, H.J., 1974. Metabolic patterns in the neonatal swine. J. Anim. Sci. 38, 1022–1030.Mersmann, H.J., 1986. Lipid metabolism in swine. In: Stanton, H.C., Mersmann, H.J. (Eds.), Swine in

Cardiovascular Research. CRC Press, Boca Raton, FL, Vol. 1, pp. 75–103.Mersmann, H.J., 1987. Acute metabolic effects of adrenergic agents in swine. Amer. J. Physiol. 252,

E85–E95.Mersmann, H.J., 1989a. The effect of insulin on porcine adipose tissue lipogenesis. Comp. Biochem.

Physiol. 94B, 709–713.Mersmann, H.J., 1989b. Inhibition of porcine adipose tissue lipogenesis by beta-adrenergic agonists.

Comp. Biochem. Physiol. 94C, 619–623.Mersmann, H.J., 1989c. Influence of infused β-adrenergic agonists on porcine blood metabolites and

catecholamines. J. Anim. Sci. 67, 2633–2645.Mersmann, H.J., 1990. Metabolic and endocrine control of adipose tissue accretion. In: Wood, J.D.,

Fisher, A.V. (Eds.), Reducing Fat in Meat Animals. Elsevier Applied Science, London, pp. 101–144.


Mersmann, H.J., 1991. Regulation of adipose tissue metabolism and accretion in mammals raised formeat production. In: Pearson, A.M., Dutson, T.R. (Eds.), Growth Regulation in Farm Animals.Advances in Meat Research, Vol. 7. Elsevier Applied Science, London. pp. 135–168.

Mersmann, H.J., 1995. Species variation in mechanisms for modulation of growth by beta-adrenergicreceptors. J. Nutr. 125, 1777S–1782S.

Mersmann, H.J., 1998. Overview of the effects of beta-adrenergic receptor agonists on animal growthincluding mechanisms of action. J. Anim. Sci. 76, 160–172.

Mersmann, H.J., 2002a. Beta-adrenergic receptor modulation of adipocyte metabolism and growth.J. Anim. Sci. 80, E24–E29.

Mersmann, H.J., 2002b. Mechanisms for conjugated linoleic acid-mediated reduction in fat deposition.J. Anim. Sci. 80, E126–E134.

Mersmann, H.J., Brown, L.J., 1973. Adaptation of swine (Sus domesticus) adipose tissue to increasedlipogenesis: expression of data. Int. J. Biochem. 4, 503–510.

Mersmann, H.J., Hu, C.Y., 1987. Factors affecting measurements of glucose metabolism and lipolyticrates in porcine adipose tissue slices in vitro. J. Anim. Sci. 64, 148–164.

Mersmann, H.J., Allen, C.D., Chai, E.Y., Brown, L.J., Fogg, T.J., 1981. Factors influencing the lipogenicrate in swine adipose tissue. J. Anim. Sci. 52, 1298–1305.

Mersmann, H.J., Brown, L.J., Beuving, R.M., Arakelian, M.C., 1976. Lipolytic activity of swineadipocytes. Amer. J. Physiol. 230, 1439–1443.

Mersmann, H.J., Goodman, J.R., Brown, L.J., 1975. Development of swine adipose tissue: morphologyand chemical composition. J. Lipid. Res. 16, 269–279.

Mersmann, H.J., Houk, J.M., Phinney, G., Underwood, M.C., 1973a. Effect of diet and weaning age onin vitro lipogenesis in young swine. J. Nutr. 103, 821–828.

Mersmann, H.J., Houk, J.M., Phinney, G., Underwood, M.C., Brown, L.J., 1973b. Lipogenesis by in vitroliver and adipose tissue preparations from neonatal swine. Amer. J. Physiol. 224, 1123–1129.

Mersmann, H.J., Underwood, M.C., Brown, L.J., Houk, J.M., 1973c. Adipose tissue composition andlipogenic capacity in developing swine. Amer. J. Physiol. 224, 1130–1135.

Miller, M.F., Cross, H.R., Wilson, J.J., Smith, S.B., 1989. Acute and long-term lipogenic response toinsulin and clenbuterol in bovine intramuscular and subcutaneous adipose tissues. J. Anim. Sci. 67,928–933.

Miller, M.F., Garcia, D.K., Coleman, M.E., Ekeren, P.A., Lunt, D.K., Wagner, K.A., Procknor, M.,Welsh, T.H. Jr., Smith, S.B., 1988. Adipose tissue, longissimus muscle and anterior pituitary growthand function in clenbuterol-fed heifers. J. Anim. Sci. 66, 12–20.

Miller, W.H. Jr., Faust, I.M., Hirsch, J., 1984. Demonstration of de novo production of adipocytes in adultrats by biochemical and radioautographic techniques. J. Lipid. Res. 25, 336–347.

Mills, S.E., 1999. Regulation of porcine adipocyte metabolism by insulin and adenosine. J. Anim. Sci.77, 3201–3207.

Mills, S., Mersmann, H.J., 1995. Beta-adrenergic agonists, their receptors, and growth: special referenceto the peculiarities in pigs. In: Smith, S.B., Smith, D.R. (Eds.), The Biology of Fat in Meat Animals:Current Advances. Amer. Soc. Anim. Sci., Champaign, IL. pp. 1–34.

Mills, S.E., Liu, C.Y., Gu, Y., Schinckel, A.P., 1990. Effects of ractopamine on adipose tissue metabolismand insulin binding in finishing hogs: interaction with genotype and slaughter weight. Domest. Anim.Endocrinol. 7, 251–263.

Mitchell, A.D., Scholz, A.M., Mersmann, H.J., 2001. Growth and body composition. In: Pond, W.G.,Mersmann, H.J. (Eds.), Biology of the Domestic Pig. Cornell University Press, Ithaca, NY,pp. 225–308.

Moloney, A.P., Allen, P., Enright, W.J., 2002. Body composition and adipose tissue accretion in lambspassively immunised against adipose tissue. Livest. Prod. Sci, 74, 165–174.

Moloney, A., Allen, P., Joseph, R., Tarrant, V., 1991. Influence of beta-adrenergic agonists and similarcompounds in growth. In: Pearson, A.M., Dutson, T.R. (Eds.), Growth Regulation in Farm Animals.Elsevier Applied Science, London, pp. 455–513.

Moody, D.E., Hancock, D.L., Anderson, D.B., 2000. Phenethanolamine repartitioning agents. In:D’Mello, J.P.F. (Ed.), Farm Animal Metabolism and Nutrition. CABI Publishing, New Yorkpp. 65–96.

Morrison, R.F., Farmer, S.R., 2000. Hormonal signaling and transcriptional control of adipocyte differ-entiation. J. Nutr. 130, 3116S–3121S.


Mourot, J., Kouba, M., Bonneau, M., 1996. Comparative study of in vitro lipogenesis in various adiposetissues in the growing Meishan pig: comparison with the large white pig (Sus domesticus). Comp.Biochem. Physiol. 115B, 383–388.

Mourot, J., Kouba, M., Peiniau, P., 1995. Comparative study of in vitro lipogenesis in variousadipose tissues in the growing domestic pig (Sus domesticus). Comp. Biochem. Physiol. 111B,379–384.

Napolitano, L.M., 1963. The differentiation of white adipose cells: an electron microscope study. J. CellBiol. 18, 663–679.

Newby, D., Gertler, A., Vernon, R.G., 2001. Effects of recombinant ovine leptin on in vitro lipolysis andlipogenesis in subcutaneous adipose tissue from lactating and nonlactating sheep. J. Anim. Sci. 79,445–452.

Noble, R.C., 1981. Lipid metabolism in the neonatal ruminant. In: Christie, W.W. (Ed.), LipidMetabolism in Ruminant Animals. Pergamon Press, Oxford, pp. 411–448.

Novakofski, J., Hu, C.Y., 1987. Culture of isolated adipose tissue cells. J. Anim. Sci. 65, Suppl. 2, 12–24.NRC, 1994. Metabolic Modifiers: Effects on the Nutrient Requirements of Food-Producing Animals.

National Research Council, National Academy Press, Washington, DC.Ntambi, J.M., Young-Cheul, K., 2000. Adipocyte differentiation and gene expression. J. Nutr. 130,

3122S–3126S.Ohyama, M., Matsuda, K., Torii, S., Matsui, T., Yano, H., Kawada, T., Ishihara, T., 1998. The interaction

between vitamin A and thiazolidinedione on bovine adipocyte differentiation in primary culture.J. Anim. Sci. 76, 61–65.

Payne, E., 1999. Fatty acid synthesis in lambs in vivo as affected by nutrition and hormone administra-tion. Aust. J. Agric. Res., 50, 163–174.

Peixing, W.U., Sato, K., Suzuta, F., Hikasa, Y., Kagota, K., 2000. Effects of lipid-related factors onadipocyte differentiation of bovine stromal-vascular cells in primary culture. J. Vet. Med. Sci. 62,933–939.

Piétri-Rouxel, F., Lenzen, G., Kapoor, A., Drumare, M.-F., Archimbault, P., Strosberg, A.D.,Manning, B.S.J., 1995. Molecular cloning and pharmacological characterization of the bovineβ3-adrenergic receptor. Eur. J. Biochem. 230, 350–358.

Plaas, H.A., Cryer, A., 1980. The isolation and characterization of a proposed adipocyte precursor celltype from bovine subcutaneous white adipose tissue. J. Dev. Physiol. 2, 275–289.

Purroy, A., Mendizabal, J.A., Soret, B., Arana, A., Mendizabal, F.J., 1997. Changes in cell number andsize and in lipogenic enzyme activity in adipose tissues during growth and fattening of Lacha(Manech) lambs. Ann. Zootech. 46, 309–319.

Ramsay, T.G., 2001. Porcine leptin alters insulin inhibition of lipolysis in porcine adipocytes in vitro.J. Anim. Sci. 79, 653–657.

Ramsay, T.G., White, M.E., Wolverton, C.K., 1989. Glucocorticoids and the differentiation of porcinepreadipocytes. J. Anim. Sci. 67, 2222–2229.

Rangwala, S.M., Lazar, M.A., 2000. Transcriptional control of adipogenesis. Annu. Rev. Nutr. 20,535–559.

Robelin, J., 1981. Cellularity of bovine adipose tissues: developmental changes from 15 to 65 percentmature weight. J. Lipid Res. 22, 452–457.

Robelin, J., 1986. Growth of adipose tissues in cattle: partitioning between depots, chemical compositionand cellularity. A review. Livest. Prod. Sci, 14, 349–364.

Robert, C., Palin, M.-F., Silversides, F.G., McKay, R.M., Pelletier, G., 2000. Messenger RNA levels ofgrowth factors, ligands, receptors, and proteins affecting lipid metabolism in pigs. Can. J. Anim. Sci.80, 559–567.

Romsos, D.R., Leveille, G.A., Allee, G.L., 1971. Alloxan diabetes in the pig (Sus domesticus): responeto glucose, tolbutamide and insulin administration. Comp. Biochem. Physiol. 40A, 557–568.

Rule, D.C., 1995. Adipose tissue glycerolipid biosynthesis. In: Smith, S.B., Smith, D.R. (Eds.),The Biology of Fat in Meat Animals: Current Advances. Amer. Soc. Anim. Sci., Champaign, IL,pp. 129–143.

Rule, D.C., Smith, S.B., Mersmann, H.J., 1987. Effects of adrenergic agonists and insulin on porcineadipose tissue lipid metabolism in vitro. J. Anim. Sci. 65, 136–149.

Rule, D.C., Smith, S.B., Mersmann, H.J., 1988a. Glycerolipid biosynthesis in porcine adipose tissuein vitro. II. Synthesis by various types of cellular preparations. J. Anim. Sci. 66, 1665–1675.


Rule, D.C., Smith, S.B., Mersmann, H.J., 1988b. Glycerolipid biosynthesis in porcine adipose tissuein vitro. I. Assay conditions for homogenates. J. Anim. Sci. 66, 1656–1664.

Rule, D.C., Smith, S.B., Mersmann, H.J., 1989. Glycerolipid biosynthesis in porcine adipose tissuein vitro: effect of adiposity and depot site. J. Anim. Sci. 67, 364–373.

Rule, D.C., Smith, S.B., Romans, J.R., 1995. Fatty acid composition of muscle and adipose tissue of meatanimals. In: Smith, S.B., Smith, D.R. (Eds.), The Biology of Fat in Meat Animals: Current Advances.Amer. Soc. Anim. Sci., Champaign, IL, pp. 144–165.

Saggerson, E.D., 1985. Hormonal regulation of biosynthetic activities in white adipose tissue. In: Cryer, A.,Van, R.L.R. (Eds.), New Perspectives in Adipose Tissue: Structure, Function and Development.Butterworths, London, pp. 87–120.

Schiavetta, A.M., Miller, M.F., Lunt, D.K., Davis, S.K., Smith, S.B., 1990. Adipose tissue cellularity andmuscle growth in young steers fed the beta-adrenergic agonist clenbuterol for 50 days and after78 days of withdrawal. J. Anim. Sci. 68, 3614–3623.

Schoonjans, K., Staels, B., Auwerx, J., 1996. Role of the peroxisome proliferator-activated receptor (PPAR)in mediating the effect of fibrates and fatty acids on gene expression. J. Lipid Res. 37, 907–925.

Sejrsen, K., Vestergaard, M., Neimann-Sorensen, A. (Eds.), 1989. Use of Somatotropin in LivestockProduction. Elsevier Applied Science, London.

Shillabeer, G., Lau, D.C., 1994. Regulation of new fat cell formation in rats: the role of dietary fats.J. Lipid Res. 35, 592–600.

Slavin, B.G., 1985. The morphology of adipose tissue. In: Cryer, A., Van, R.L.R. (Eds.), NewPerspectives in Adipose Tissue: Structure, Function and Development. Butterworths, London,pp. 23–43.

Smas, C.M., Sul, H.S., 1995. Control of adipocyte differentiation. Biochem. J. 309, 697–710.Smith, S.B., 1984. Evidence that phosphofructokinase limits glucose utilization in bovine adipose tissue.

J. Anim. Sci., 58, 1198–1204.Smith, S.B., 1995. Substrate utilization in ruminant adipose tissues. In: Smith, S.B., Smith, D.R. (Eds.),

The Biology of Fat in Meat Animals: Current Advances. Amer. Soc. Anim. Sci., Champaign, IL,pp. 166–188.

Smith, S.B., Smith, D.R. (Eds.), 1995. The Biology of Fat in Meat Animals: Current Advances. Amer.Soc. Anim. Sci., Champaign, IL.

Smith, S.B., Lin, K.C., Wilson, J.J., Lunt, D.K., Cross, H.R., 1998. Starvation depresses acylglycerolbiosynthesis in bovine subcutaneous but not intramuscular adipose tissue homogenates. Comp.Biochem. Physiol. 120B, 165–174.

Smith, S.B., Prior, R.L., Ferrell, C.L., Mersmann, H.J., 1984. Interrelationships among diet, age, fatdeposition and lipid metabolism in growing steers. J. Nutr. 114, 153–162.

Smith, S.B., Prior, R.L., Mersmann, H.J., 1983. Interrelationships between insulin and lipid metabolismin normal and alloxan-diabetic cattle. J. Nutr. 113, 1002–1015.

Soret, B., Lee, H.J., Finley, E., Lee, S.C., Vernon, R.G., 1999. Regulation of differentiation of sheepsubcutaneous and abdominal preadipocytes in culture. J. Endocrinol. 161, 517–524.

Soret, B., Mendizabal, J.A., Arana, A., Purroy, A., Eguinoa, P., 1998. Breed effects on cellularity andlipogenic enzymes in growing Spanish lambs. Small Ruminant Res. 29, 103–112.

Steele, N.C. (Ed.), 1991. Emerging technologies to alter fat:lean ratio in animal products. J. Anim. Sci.69, Suppl. 2, pp. 1–124.

Steele, N.C., Boyd, R.D., Campbell, R.G., 1994. Growth, metabolic modifiers, and nutrient considera-tions. In: Hafs, H.D., Zimbelman, R.G. (Eds.), Low-Fat Meats Design Strategies and HumanImplications. Academic Press, New York, pp. 167–189.

Steffen, D.G., Brown, L.J., Mersmann, H.J., 1978. Ontogenic development of swine (Sus domesticus)adipose tissue lipases. Comp. Biochem. Physiol. 59, 195–198.

Steffen, D.G., Phinney, G., Brown, L.J., Mersmann, H.J., 1979. Ontogeny of glycerolipid biosyntheticenzymes in swine liver and adipose tissue. J. Lipid Res. 20, 246–253.

St. John, L.C., Rule, D.C., Knabe, D.A., Mersmann, H.J., Smith, S.B., 1987. Fatty acid-binding proteinactivity in tissues from pigs fed diets containing 0 and 20% high oleate oil. J. Nutr. 117, 2021–2026.

Suryawan, A., Hu, C.Y., 1993. Effect of serum on differentiation of porcine adipose stromal-vascularcells in primary culture. Comp. Biochem. Physiol. 105A, 485–492.

Suryawan, A., Hu, C.Y., 1995. The primary cell culture system for preadipocytes. In: Smith, S.B.,Smith, D.R. (Eds.), The Biology of Fat in Meat Animals: Current Advances. Amer. Soc. Anim. Sci.,Champaign, IL, pp. 78–92.


Tavangar, K., Murata, Y., Patel, S., Kalinyak, J.E., Pedersen, M.E., Goers, J.F., Hoffman, A.R.,Kraemer, F.B., 1992. Developmental regulation of lipoprotein lipase in rats. Amer. J. Physiol. 262,E330–E337.

Thulin, A.J., Allee, G.L., Harmon, D.L., Davis, D.L., 1989. Utero-placental transfer of octanoic, palmiticand linoleic acids during late gestation in gilts. J. Anim. Sci. 67, 738–745.

Tontonoz, P., Hu, E., Graves, R.A., Budavari, A.I., Spiegelman, B.M., 1994. mPPAR gamma 2: tissue-specific regulator of an adipocyte enhancer. Genes Dev. 8, 1224–1234.

Torii, S.-I., Kawada, T., Matsuda, K., Matsui, T., Ishihara, T., Yano, H., 1998. Thiazolidinedione inducesthe adipose differentiation of fibroblast-like cells resident within bovine skeletal muscle. Cell Biol.Int. 22, 421–427.

Trayhurn, P., Beattie, J.H., 2001. Physiological role of adipose tissue: white adipose tissue as anendocrine and secretory organ. Proc. Nutr. Soc. 60, 329–339.

Trenkle, A., Marple, D.N., 1983. Growth and development of meat animals. J. Anim. Sci. 57, Suppl. 2,273–283.

Truscott, T.G., Wood, J.D., Macfie, H.J.H., 1983. Fat deposition in Hereford and Friesian steers. J. Agr.Sci. Camb. 100, 257–270.

Vasilatos, R., Etherton, T.D., Wangsness, P.J., 1983. Preparation of isolated bovine adipocytes: validationof use for studies characterizing insulin sensitivity and binding. Endocrinology 112, 1667–1673.

Vernon, R.G., 1980. Lipid metabolism in the adipose tissue of ruminant animals. Prog. Lipid Res. 19,23–106.

Vernon, R.G., 1981. Lipid metabolism in the adipose tissue of ruminant animals. In: Christie, W.W. (Ed.),Lipid Metabolism in Ruminant Animals. Pergamon Press, Oxford, pp. 279–362.

Vernon, R.G., 1982. Development in vitro of fatty acid synthesis in perirenal adipose tissue fromnew-born lambs. Biol. Neonate 42, 275–278.

Vernon, R.G., 1986. The growth and metabolism of adipocytes. In: Buttery, P.J., Lindsay, D.B.,Haynes, N.B. (Eds.), Control and Manipulation of Animal Growth. Butterworths, London, pp. 67–84.

Vernon, R.G., 1991. Depot specific endocrine control of fatty acid synthesis in adipose tissues of foetallambs. Domest. Anim. Endocrinol. 8, 161–164.

Vernon, R.G., 1992. Control of lipogenesis and lipolysis. In: Boorman, K.N., Buttery, P.J., Lindsay, D.B.(Eds.), The Control of Fat and Lean Deposition. Butterworth-Heinemann, Oxford, pp. 59–81.

Vernon, R.G., Clegg, R.A., 1985. The metabolism of white adipose tissue in vivo and in vitro. In: Cryer, A.,Van, R.L.R. (Eds.), New Perspectives in Adipose Tissue: Structure, Function and Development.Buttersworths. London, pp. 65–86.

Vernon, R.G., Taylor, E., 1986. Acetyl-CoA carboxylase of sheep adipose tissue: problems of the assayand adaptation during fetal development. J. Anim. Sci. 63, 1119–1125.

Vernon, R.G., Barber, M.C., Finley, E., 1991. Modulation of the activity of acetyl-CoA carboxylase andother lipogenic enzymes by growth hormone, insulin and dexamethasone in sheep adipose tissue andrelationship to adaptations to lactation. Biochem. J. 274, 543–548.

Vierck, J.L., McNamara, J.P., Dodson, M.V., 1996. Proliferation and differentiation of progeny of ovineunilocular fat cells (adipofibroblasts). In Vitro Cell Dev. Biol. Anim. 32, 564–572.

Wajchenberg, B.L., 2000. Subcutaneous and visceral adipose tissue: their relation to the metabolicsyndrome. Endocr. Rev. 21, 697–738.

Walstra, P., 1980. Growth and Carcass Composition from Birth to Maturity in Relation to FeedingLevel and Sex in Dutch Landrace Pigs. Mededelingen Landbouwhogeschool, Wageningen, TheNetherlands.

Walton, P.E., Etherton, T.D., 1986. Stimulation of lipogenesis by insulin in swine adipose tissue: antag-onism by porcine growth hormone. J. Anim. Sci. 62, 1584–1595.

Walton, P.E., Etherton, T.D., 1987. The culture of adipose tissue explants in serum-free medium. J. Anim.Sci. 65, Suppl. 2, 25–30.

Widdowson, E.M., Lister, D., 1991. Nutritional control on growth. In: Pearson, A.M., Dutson, T.R. (Eds.),Growth Regulation in Farm Animals. Elsevier Applied Science, London, pp. 67–101.

Wilson, J.J., Young, C.R., Smith, S.B., 1992. Triacylglycerol biosynthesis in bovine liver and subcuta-neous adipose tissue. Comp. Biochem. Physiol. 103B, 511–516.

Wood, J.D., 1990. Consequences for meat quality of reducing carcass fatness. In: Wood, J.D., Fisher, A.V.(Eds.), Reducing Fat in Meat Animals. Elsevier Applied Science, London, pp. 344–397.

Wright, J.T., Hausman, G.J., 1990. Adipose tissue development in the fetal pig examined using mono-clonal antibodies. J. Anim. Sci. 68, 1170–1175.


Wu, P., Sato, K., Suzuta, F., Hikasa, Y., Kagota, K., 2000. Effects of lipid-related factors on adipocytedifferentiation of bovine stromal-vascular cells in primary culture. J. Vet. Med. Sci. 62, 933–939.

Yang, A., Larsen, T.W., Smith, S.B., Tume, R.K., 1999. Delta 9 desaturase activity in bovine subcutaneousadipose tissue of different fatty acid composition. Lipids 34, 971–978.

Yu, Z.K., Hausman, G.J., 1998. Expression of CCAAT/enhancer binding proteins during porcinepreadipocyte differentiation. Exp. Cell. Res. 245, 343–349.

Yu, Z.K., Wright, J.T., Hausman, G.J., 1997. Preadipocyte recruitment in stromal vascular cultures afterdepletion of committed preadipocytes by immunocytotoxicity. Obes. Res. 5, 9–15.


303

12 Ontogeny and metabolism of brownadipose tissue in livestock species

S. B. Smith and G. E. Carstens

Department of Animal Science, Texas A & M University, College Station,TX 77483-2971, USA

We have reported several aspects of brown adipose tissue (BAT) development and responseto environmental stimuli. The metabolism of BAT resembles that from mature ruminant whiteadipose tissue, in that acetate is the primary precursor for lipogenesis. There is a precipitousdecline in rates of de novo lipogenesis during the last trimester, indicating that the contribu-tion of fatty acid biosynthesis to lipid filling is more important earlier in fetal development.The mixture of brown and white adipocytes in subcutaneous adipose tissues from newborncalves suggests that brown adipocytes may involute into white adipocytes in this species.However, there is growing evidence in rodent species that brown and white adipocytesdifferentiate and develop independently. Clearly, brown adipose tissue from sheep and cattlerapidly dedifferentiates and/or is lost via apoptosis early postnatally. This phenomenon isespecially rapid in warm ambient temperatures, but may be delayed by feeding dams dietshigh in polyunsaturated fatty acids. In Wagyu × Angus crossbred calves, brown adipose tissuefunction is remarkably refractory to profound reductions in dietary protein intake in dams,indicating the high priority animals place in ensuring adequate thermogenic capacity in theirnewborn.

1. INTRODUCTION

Adverse climatic conditions during the early postnatal period can disrupt thermal balance innewborn lambs and calves leading to hypothermia and/or death. Numerous reports havedemonstrated that calf mortality increases during inclement weather. In an epidemiologicalstudy involving more than 87,000 Bos taurus calves, Azzam et al. (1993) found that calf mor-tality increased progressively as ambient temperature decreased or as precipitation amount onthe day of birth increased. Moreover, the stress of maintaining homeothermy during severecold exposure for extended periods may interact with other etiological factors associated withneonatal calf mortality and morbidity through depletion of energy reserves, induction ofphysical weakness, and/or delay in absorption of immunoglobulins.



A major thermoregulatory mechanism for survival of neonatal ruminants during cold stressis heat production by brown adipose tissue (BAT). Maintenance of homeothermy during theearly postnatal period necessitates an acute and sustained thermogenic response by the new-born calf. Maximal thermogenic response to cold (i.e. summit metabolism), which includesboth shivering and nonshivering thermogenesis, is 4- to 5-fold higher than thermoneutralmetabolism in neonatal lambs and 3- to 4-fold higher in neonatal calves (Stott and Slee, 1985;Okamoto et al., 1986; Robinson and Young, 1988a,b).

Approximately half of the cold-induced summit metabolism in newborn lambs is derivedfrom nonshivering thermogenesis (Stott and Slee, 1985). Therefore, newborn lambs andcalves must possess highly active BAT during the early postnatal period when the demand forthermogenesis is greatest. Most of the adipose tissue in the newborn ruminant species is BAT,although small amounts of white adipose tissue (WAT) are present (Alexander et al., 1975;Alexander, 1978; Martin et al., 1997, 1999). Alexander et al. (1975) estimated that the quantityof BAT in newborn calves was ~1.5–2.0% of body weight.

This review will focus primarily on BAT in lambs and calves. Little mention will be madeof piglets because distinctive brown adipocytes do not appear to be present in porcine adiposetissue.

2. MORPHOLOGY OF BROWN ADIPOSE TISSUE IN NEWBORNLIVESTOCK SPECIES

Brown adipocytes from newborn calves do not display the typical multilocular feature that ischaracteristic of brown adipose tissue in other species. Instead, bovine brown adipocytes con-tain a large central lipid vacuole with few peripheral lipid inclusions. This is consistent withAlexander et al. (1975), who examined perirenal brown adipocytes in newborn calves at lowermagnification and described adipocytes as dominated by a large lipid vacuole with smallerlipid inclusions in the marginal cytoplasm of some of the cells. Napolitano (1963) stated thatthe major criterion used to characterize brown adipocytes morphologically should be theappearance and differentiation of mitochondria rather than the occurrence of multilocularlipid droplets.

Subcutaneous adipose tissue overlying the sternum from Wagyu × Angus crossbred calvescontains unilocular adipocytes with few cytoplasmic inclusions (Martin et al., 1997). Unlikeperirenal adipocytes from these same calves, only a small number of mitochondria withpoorly developed cristae were present. Thus, sternum adipose tissue represents a WAT depotin Wagyu × Angus calves. These results are similar to those of Alexander et al. (1975) fornewborn calves, although the earlier work described adipocytes containing a few small lipiddroplets in addition to the large central vacuole.

3. SPECIFIC GENE EXPRESSION IN BROWN ADIPOSE TISSUE

3.1. Uncoupling protein-1

Brown adipose tissue’s thermogenic capacity is attributed to uncoupling protein-1 (UCP1)located in the inner mitochondrial membrane. The principle function of UCP1 is to dissipatethe proton gradient created by mitochondrial respiration, which uncouples mitochondrial res-piration from synthesis of ATP and allows energy to be dissipated as heat. The concentrationof UCP1 is a key biochemical marker of the thermogenic capacity of BAT (Himms-Hagen,1986). Brander et al. (1993) demonstrated that UCP1 mRNA was present in intrascapular

S. B. Smith and G. E. Carstens304

BAT, but was not present in rat omental, liver, kidney, skeletal muscle, or heart. Casteillaet al. (1989) examined perirenal, pericardial, peritoneal, and subcutaneous adipose tissuedepots in newborn calves, and found UCP1 mRNA in all tissues except subcutaneous adiposetissue. UCP1 mRNA was highest in perirenal adipose tissue, followed by pericardial tissue(~50% of perirenal) and peritoneal tissue (~15% of perirenal). More recently, Landis et al.(2002) reported the presence of UCP1 mRNA in s.c. adipose tissue of Brahman and Angusfetal calves (last trimester), indicating that s.c. adipose tissue contained a population of brownadipocytes during late fetal development.

3.2. β-Adrenergic receptors

Expression of the β3-adrenergic receptor (β3-AR) gene is especially high in bovine perirenalBAT during fetal and early postnatal development (Casteilla et al., 1989, 1994). β3-Adrenergicreceptors have been documented in porcine (white) adipocytes (Mersmann, 1996), indicating thatthis receptor type is not unique to BAT. However, β3-AR gene expression may be prerequisiteto BAT differentiation in fetal calves (Casteilla et al., 1994).

3.3. Brown unknown gene

Moulin et al. (2001a,b) reported the existence of a gene that is expressed early in the differ-entiation of stromal–vascular cells from rat brown adipose tissue. The gene, which theytermed BUG (brown unknown gene), is expressed at high levels in rat interscapular adiposetissue, cardiac muscle, brain, and kidney, but at low levels in white inguinal adipose tissue,muscle, liver, and spleen. It also is expressed at higher levels in interscapular adipose tissuefrom obese rats, which exhibits depressed UCP1 gene expression (Moulin et al., 2001a). Theauthors concluded that BUG gene expression must be depressed to obtain high rates of UCP1gene expression. Their results also indicate that stromal–vascular preadipocytes from inter-scapular (brown) adipocytes represent a cell line that is distinct from preadipocytes frominguinal (white) adipose tissue. To our knowledge, the expression of BUG in BAT from ruminantspecies has not been described.

4. ONTOGENIC DEVELOPMENT OF BROWN ADIPOSE TISSUE

Detectable quantities of perirenal BAT in lambs first appear at 70 days of gestation(Alexander, 1978). Alexander (1978) also reported that the mass of perirenal adipose tissuein fetal Merino sheep increased by approximately 34% over the last 3 weeks of gestation.Vernon et al. (1981) documented a 40% increase in adipocyte volume of fetal lambs in thelast 4 weeks of pregnancy, indicating that most, if not all, of the increase in perirenal BATmass in fetal lambs was due to adipocyte hypertrophy. In fetal calves, some portion of BAThypertrophy is due to increased adipocyte volume (Landis et al., 2002), at least during theperiod between 96 and 48 days before parturition (fig. 1). During this period there is a largenumerical increase in brown adipocyte volume. However, subsequent to 48 days before birth,total apparent perirenal brown adipocyte number more than doubles (Landis et al., 2002).Thus, hyperplastic growth of adipocytes contributes substantially to the fetal growth ofbovine perirenal BAT.

Growth of BAT in lambs, relative to fetal weight, is allometric (100 mg/kg of fetal weight) from70 to 120 days of gestation, and isometric (6 mg/kg) thereafter until term (ovine gestation

Ontogeny and metabolism of brown adipose tissue 305

length is 150 days) (Alexander, 1978). Lipid locules were first identified in perirenal adipocyteson day 70 of gestation, and by day 80 to 90, mitochondria began to proliferate and take on themorphological features of brown adipocytes. Rapid accumulation of lipid in perirenaladipocytes occurs during the allometric growth phase, whereas increased mitochondrialbiogenesis and sympathetic innervation of adipocytes occur during the isometric growth phase.Therefore, even though prenatal growth of BAT is most rapid during mid-gestation, functionaldevelopment of brown adipocytes does not occur until late gestation.

4.1. Morphological changes during ontogeny

Based on histological examinations of bovine BAT during late gestation, fetal brown adipocytescontain fewer mitochondria that are also less developed compared to brown adipocytes fromnewborn calves (Landis et al., 2002; fig. 2). Mitochondria from fetuses sampled as late asmidway through the last trimester are spherical, with cristae that do not traverse the entiremitochondrion, whereas brown adipocytes from late term and newborn BAT are nearlyunilocular, with small lipid inclusions peripheral to larger central lipid vacuoles, and themitochondria are more abundant and more fully differentiated (fig. 3). These observationscorroborate Nedergaard et al. (1986), who described similar morphological changes in brownadipocytes during fetal development of rats.

At 96 days before birth, some BAT cells have accumulated very little lipid, whereas othersalready are nearly unilocular (figs. 2A and D). Brown adipocytes gradually accumulate lipid,with the multilocular appearance persisting in Brahman BAT at 24 days before birth (fig. 2E).


Fig. 1. Adipocyte density and mean volume of perirenal brown adipose tissue. Each data point representsthe mean for 3 fetal calves per breed type. The overall SEM for each breed type is affixed to the symbols. TheSEM are not large enough to be visible for adipocytes/g. Adapted from Figure 4 of Landis et al. (2002).

At birth, brown adipocytes become essentially unilocular, with only a few small lipidvacuoles set apart from the large, central vacuole (figs. 2C and F).

At the earliest sampling period, BAT mitochondria are smaller, spherical, and quite dense(figs. 3A and D). By 24 days before birth, the mitochondria become quite large, and cristae,although distinct, are extensive (figs. 3B and E). Within 7 days of parturition, the morphologyof the mitochondria changes from primarily spherical to markedly elongated, with extensivelydifferentiated cristae. This morphology is apparent at birth (figs. 3C and F).


Fig. 2. Transmission electron micrographs of perirenal brown adipocytes. Perirenal brown adipose tissuewas obtained by cesarean section at 96 and 24 days before birth, and at birth, from Angus (left column) andBrahman (right column) fetuses. Thick arrows indicate lipid vacuoles and thin arrows point to mitochondria.A nucleus is indicated at the bottom left. Scale bar at the bottom right indicates magnification. Adapted fromFigure 2 of Landis et al. (2002).

4.2. Gene expression during ontogeny

Fetal bovine perirenal UCP1 mRNA is not detectable until day 211 of gestation, and only insmall quantities thereafter until day 259, when levels increase markedly (Casteilla et al.,1989). We recently documented a depression and then recovery in UCP1 gene expressionduring the last 14 days of gestation (fig. 4; Landis et al., 2002). The decline in UCP1 geneexpression coincided with the change in mitochondrial morphology from spherical to elongated,and therefore may have biological significance.


Fig. 3. Transmission electron micrographs of mitochondria from perirenal brown adipocytes. Perirenal brownadipose tissue was obtained by cesarean section at 96 and 24 days before birth, and at birth, from Angus(left column) and Brahman (right column) fetuses. Arrows point to cristae and mitochondria are labeled M insome panels. Scale bar at lower right indicates magnification. Adapted from Figure 3 of Landis et al. (2002).

The activity of 5′-deiodinase first appears in fetal bovine perirenal adipose tissue at2 months of gestation, increases rapidly until activity peaks at 7 months, and declines there-after until birth (Giralt et al., 1989). The activity of 5′-deiodinase is responsible fordeiodination of T4 to T3, which is thought to be required for optimal synthesis of UCP1 incold-adapted rats (Bianco and Silva, 1988). Giralt et al. (1989) noted that 5′-deiodinase peakedat approximately the same time that Casteilla et al. (1989) first detected UCP1 mRNA, andhypothesized that endogenous production of T3 may be involved in prenatal induction ofUCP1 expression in BAT.

Trayhurn et al. (1993) demonstrated small amounts of UCP1 protein in goat s.c. adiposetissue at 2.5 days of age. Consistent with this, our laboratory (Martin et al., 1999) demon-strated that s.c. adipose tissue from newborn Brahman and Angus calves contained adipocyteswith distinctive brown adipocyte morphology. More recent results (Landis et al., 2002)demonstrated UCP1 mRNA in s.c. adipose tissue during the last trimester of gestation,although the abundance of UCP1 mRNA dropped precipitously during the last 30 days priorto parturition (Fig. 4).

Casteilla et al. (1994) examined the expression of adrenergic receptor (β1- and β3-AR)genes in bovine perirenal BAT during fetal and early postnatal development. The β3-ARmRNA was first measurable during mid-gestation, and its concentration increased dramati-cally between 6 months of fetal life and birth. In contrast, β1-AR mRNA was expressed at lowlevels throughout fetal life. The appearance of β3-AR mRNA preceded the expression ofUCP1 mRNA (Casteilla et al., 1989), suggesting that β3-AR gene expression may be prereq-uisite to BAT differentiation (Casteilla et al., 1994). Postnatally, β3-AR gene expression


Fig. 4. UCP1 gene expression in perirenal and tailhead s.c. adipose tissues. Total RNA was extractedfrom the perirenal (brown adipose tissue, BAT) and tailhead s.c. adipose tissue (white adipose tissue, WAT)samples for slot blot analysis of uncoupling protein-1 (UCP1). Each data point represents the mean for 3 fetalcalves per breed. The overall SEM for each breed type is affixed to the symbols. Relative concentrations ofUCP1 mRNA are expressed as the ratio of UCP1:18S ribosomal RNA × 100. Adapted from Figure 5 ofLandis et al. (2002).

declined after approximately 3 months of age; the decline was much slower than the morerapid postnatal decline of UCP1 mRNA (Nougues et al., 1993; Trayhurn et al., 1993).

Response to β3-adrenergic stimulus apparently persists after birth. Acute treatment of ratswith a selective β3-agonist increased BAT GDP-binding of mitochondria (a measure of UCP1activity) after only 60 min of treatment (Milner et al., 1988). In adult dogs, treatment with aβ3-agonist increased UCP1 concentration and the expression of UCP1 mRNA (Champignyet al., 1991). Administration of the β3-agonist ICI-D7114 to newborn lambs during the firstseveral weeks of life delayed the apparent involution of BAT to WAT (Nougues et al., 1993).The β3-agonist-treated lambs continued to express UCP1 mRNA in perirenal and pericardialfat depots at 25 days of age, whereas control lambs of the same age did not.

4.3. Fatty acid metabolism during ontogeny

The incorporation of acetate, glucose, and palmitate into glycerolipids of perirenal adiposetissue decreases markedly during late gestation, especially in Brahman BAT (Landis et al.,2002) (fig. 5). Vernon et al. (1981) demonstrated reductions in fatty acid synthesis fromacetate and glucose in fetal ovine perirenal BAT that were similar in magnitude to those


Fig. 5. Lipogenesis from acetate, glucose, and palmitate in perirenal brown adipose tissue as a function ofaverage fetal age. The incorporation rate is expressed as nmol substrate incorporated/106 cells/h. Each datapoint represents the mean for 3 fetal calves per breed type. The rate of palmitate esterification in Brahmanfetuses at 96 days before birth was 1,570 nmol/106 cells/h (off scale in the figure). The overall SEM for eachbreed type is affixed to the symbols. The SEM for glucose incorporation are not large enough to be visible.There was a significant age effect for acetate incorporation into glycerolipids. There was a significant age ×breed type effect for palmitate and glucose incorporation into glycerolipids. Adapted from Figure 6 ofLandis et al. (2002).

observed in fetal calves. Although de novo fatty acid biosynthesis is barely detectable byparturition in bovine BAT, glycerolipid synthesis from palmitate remains elevated at that time.Thus, palmitate esterification accounts for 98% of total glycerolipid synthesis in vitro in thenewborn calves, and is a primary contributor to lipid filling throughout gestation.

Acetate has been well documented as the principal source of carbon for de novo fatty acidbiosynthesis in adipose tissue of young and adult ruminant species (Ballard et al., 1972; Ingleet al., 1972; Smith and Prior, 1986). Both acetate and glucose contribute significantly tode novo lipogenesis in newborn calves (Martin et al., 1999). At the beginning of the thirdtrimester in fetal calves, the rate of lipogenesis from acetate is approximately 10-fold greaterthan lipogenesis from glucose. Some portion of the glucose carbon would have been recov-ered as glyceride–glycerol (consistent with the high rates of palmitate incorporation intolipids), so that actual rates of de novo fatty acid biosynthesis from glucose would be very lowrelative to fatty acid biosynthesis from acetate. Thus, fetal bovine adipose tissue preferentiallyuses acetate as the carbon source for de novo fatty acid biosynthesis early in the thirdtrimester of gestation.

5. ENVIRONMENTAL EFFECTS ON BROWN ADIPOSE TISSUE

5.1. Metabolism and thermogenesis during cold exposure

Cold-induced secretion of norepinephrine (NE) increases the specific thermogenic activity ofBAT. In addition to its role in acute activation of BAT thermogenesis, NE is also involved inlong-term modulation of BAT growth and development during cold stress by enhancing dif-ferentiation of BAT precursor cells, mitochondrial proliferation, and transcription of UCP1 viaβ- and α1-AR pathways (Géloën et al., 1988). Type II thyroxine 5′-deiodinase, which is animportant enzyme regulating the thermogenic capacity of BAT, is increased in BAT duringcold exposure (Puig-Domingo et al., 1989).

Previous studies suggested that the improved survival of lambs born to cold-exposed eweswas due to an enhanced rate of BAT thermogenesis. Stott and Slee (1985) found that lambsborn to cold-exposed ewes exhibited significantly higher NE-induced thermogenic rates (in vivoassessment of BAT thermogenesis) than lambs from warm-exposed ewes. Likewise, Symondset al. (1992) found that lambs from cold-exposed ewes were 15% heavier at birth, and pos-sessed 21% more perirenal BAT that was also 40% more active thermogenically compared tolambs from control ewes. Newborn lambs from cold-exposed ewes were clearly more cold-tolerant as thermogenic rates were 16% greater in a warm (28°C) environment and 40% greaterin a cold (14°C) environment, relative to lambs from control ewes.

We investigated the postnatal changes in perirenal BAT in neonatal calves, and found thatcytochrome c oxidase activity of perirenal BAT was highest at birth, and decreased substan-tially after 7 days of warm exposure (Carstens, 1994). Mitochondrial protein concentrationsin warm-exposed calves were only 20% of that found in newborn calves (Carstens, 1994), andbovine perirenal UCP1 was reduced markedly in warm-exposed calves (Martin et al., 1997).Trayhurn et al. (1993) reported a similar postnatal decline in cytochrome c oxidase activityin neonatal goats (325, 50, and 3 μmol oxidized/min/g of BAT in newborn, 7-day-old, and21-day-old goats, respectively). Cytochrome c oxidase activity was also lower in cold-exposedcalves than newborn calves, but was still 2.7-fold higher than in warm-exposed calves at7 days of age (Carstens, 1994). Thus, cold exposure during the early postnatal period delaysthe apparent involution of BAT to WAT, resulting in higher BAT thermogenic rates during theneonatal period.


5.2. Involution during warm exposure

Postnatally, in neonatal ruminants BAT is involuted into WAT within 2 to 3 weeks if theneonate is not exposed to cold (Alexander et al., 1975). In bovine perirenal BAT, there is noapparent atrophy or degeneration of brown adipocytes postnatally (Alexander et al., 1975).Instead, there appears to be a progressive accumulation of lipid, accompanied by the dedif-ferentiation and loss of mitochondria, as perirenal adipocytes acquire the morphology ofwhite adipocytes. More recently, Trayhurn et al. (1993) demonstrated the presence of UCPprotein (measured by Western analysis) in s.c. (hindlimb and neck) and internal (perirenal,pericardial, and omental) adipose tissues of goats. Trayhurn et al. (1993) detected UCPprotein in s.c. adipose tissue of 7-day-old goats, and in perirenal BAT of 21-day-old goats.Uncoupling protein is more persistent than its mRNA; uncoupling protein mRNA was notdetectable by day 2.5 in goat perirenal BAT, and was undetectable in other adipose tissuedepots at birth.

We examined the effects of postnatal cold exposure on NE-induced BAT thermogenesisin Holstein calves in vivo (table 1). Peak metabolic (PM) rates were 26% lower in warm-exposed calves than in newborn calves, yet PM rates were similar between newborn andcold-exposed calves, demonstrating that cold exposure delayed the apparent postnatal invo-lution of BAT. This observation is supported by the fact that cytochrome c oxidase activityand total mitochondrial protein in BAT were 2.7- and 2.5-fold higher in cold-exposed than inwarm-exposed calves.

We measured UCP1 mRNA in a small number of BAT samples from newborn calves thathad been exposed to 4°C for 7 days postnatally. As observed for cytochrome c oxidase activ-ity and mitochondrial protein, there was a dramatic loss of UCP1 mRNA postnatally evenduring cold exposure (fig. 6). We also recently examined the effects of postnatal cold expo-sure on BAT thermogenesis in newborn lambs (table 2). Brown adipose tissue mass tended tobe greater in the lambs held for 48 h at 28°C than in lambs held at 6°C. Similarly, total andmitochondrial BAT protein concentrations were increased significantly by cold exposure.This observation is supported by the fact that cytochrome c oxidase activity and UCP1 mRNAwere approximately 3-fold and 5-fold higher, respectively, in BAT from cold-exposed lambsthan in warm-exposed lambs (table 2).


Table 1

Effects of cold exposure on postnatal changes in thermoneutral (TM) and NE-induced peakmetabolic rates (PM), and BAT cytochrome c oxidase activity and mitochondrial protein innewborn Holstein, and 7-day-old calves exposed to 4°C (cold-exposed) or 22°C (warm-exposed)temperatures from birth to 7 days of age

The reduction in BAT mass, and concomitant elevation in BAT protein content, were theresult of cold-induced mobilization of BAT lipid stores, which caused a remarkable delipida-tion of BAT in response to cold exposure (fig. 7). Mitochondria from warm-exposed lambsunderwent extensive dedifferentiation, and by this stage were structurally similar to mito-chondria from fetal calves (fig. 3), i.e. they were larger with less extensive cristae. There alsowas some apparent dedifferentiation of mitochondria in the cold-exposed lambs. The loss ofmitochondrial integrity could have been the result of dedifferentiation (involution).Alternatively, brown adipocytes in warm-exposed lambs may have been undergoing apoptosis,to be replaced subsequently with white adipocytes.

6. APOPTOSIS OF BROWN ADIPOCYTES

It is not clear if BAT from ruminant species undergoes dedifferentiation and direct conversionto WAT, or whether brown adipocytes experience apoptosis, to be replaced by newly differ-entiated white adipocytes. In bovine perirenal BAT, there is no apparent atrophy ordegeneration of brown adipocytes postnatally (Alexander et al., 1975). Instead, there appearsto be a progressive accumulation of lipid, accompanied by the dedifferentiation and loss of


Fig. 6. Uncoupling protein-1 (UCP1) mRNA innewborn and 7-day cold-adapted calves. Lane 1,bovine longissimus muscle RNA. Lanes 2 and 3,RNA from BAT of newborn calves. Lanes 4 and 5,RNA from BAT of calves that had been subjected to4°C for 7 days postnatally. No bands are visible inlane 5. Lane 6, DNA from a PCR reaction that usedthe calf UCP1 cDNA as template. The upper bandis the UCP-PCR probe. The lower band correspondsto unincorporated primers. Lane 7, Eco RI-excisedcalf UCP partial cDNA (1.4 kb). (S.B. Smith andG.E. Carstens, unpublished data.)

Table 2

Effects of cold exposure (28°C vs 6°C) on perirenal adipose tissue composition, cytochrome coxidase activity, mitochondrial protein, and UCP mRNA in 48-hour-old lambs

Warm-exposed Cold-exposed P <

Birth weight, kg 4.18 ± 0.17 4.06 ± 0.15 NSBAT, g/kg body weight 4.40 ± 0.27 2.69 ± 0.14 0.10BAT protein, mg/g 97.6 ± 2.8 131.2 ± 4.0 0.001Cytochrome c oxidase activity, μmol/g/min 57.7 ± 5.0 153.6 ± 7.5 0.001Mitochondrial protein, mg/g BAT 13.06 ± 1.4 26.65 ± 1.72 0.001UCP mRNA:28S rRNA ratio 0.10 ± 0.05 0.64 ± 0.11 0.05

n = 20 lambs/treatment.NS = not statistically different (P > 0.05).UCP:28S ratio = (Laser densitometer area for UCP mRNA: Laser densitometer area for 28S rRNA) × 100. Areaswere calculated from slot blots of 5 μg total RNA, hybridized either to a 32P-labeled 300-bp PCR-generated UCPprobe or a 32P-labeled cDNA for the rat 28S ribosomal subunit. (S. B. Smith and G. E. Carstens, unpublished data.)

mitochondria, as perirenal adipocytes acquired the morphology of white adipocytes. Thesedata provided evidence to suggest that, prenatally, adipocytes from all depots may initiallydifferentiate as BAT, and that they subsequently involute to acquire WAT morphological char-acteristics. Our histological examination of Angus and Brahman s.c. adipose tissue (Martinet al., 1999; fig. 8) tends to support this postulate. Although most adipocytes have alreadyacquired WAT characteristics by parturition, we located several adipocytes with distinctly


Fig. 7. Brown adipose tissue from newborn lambs exposed to 28°C (left) or 6°C (right) for 48 h postnatally.Thick arrows indicate lipid droplets and thin arrows indicate mitochondria. Adipocytes from warm-exposedlambs are multilocular, whereas adipocytes from cold-exposed lambs are nearly devoid of lipid. Mitochondriaare less numerous in BAT from warm-exposed lambs. Cristae structure is becoming disrupted in BAT fromwarm-exposed lambs, indicating dedifferentiation of brown adipocytes. Scale bars are indicated for the lowestand highest magnifications. (S.B. Smith and G.E. Carstens, unpublished data.)

brown adipocyte morphology, i.e. extensive, highly differentiated mitochondria surroundinga smaller lipid vacuole. This could be interpreted to mean that s.c. adipose tissue initially dif-ferentiated as BAT, and we are observing the involution of BAT to WAT initiated prenatally.However, we cannot rule out the possibility that brown adipocytes were lost to the total population of adipocytes via apoptosis, a process not detectable in transmission electron photomicrographs.

Lindquist and Rehnmark (1998) demonstrated that transferring cold-adapted mice to 28°Ccaused a rapid increase in the rate of apoptosis in interscapular BAT. Murine brown


Fig. 8. Subcutaneous adipose tissue from Angus (left column) and Brahman (right column) newborn calves.Cells with brown adipocyte morphology (arrows) and white adipocyte morphology (arrowheads) are apparent.A capillary endothelial cell (CE) is centrally located in the Brahman sample. Mitochondrial size and cristaedensity (bottom panels) are indicative of brown adipocytes, with no apparent difference between breed types.Scale bars are indicated for the lowest and highest magnifications. Adapted from Figure 5 of Martin et al.(1999).

adipocytes exposed to NE in culture exhibited a 50% decline in DNA fragmentation (i.e.apoptosis) (Lindquist and Rehnmark, 1998). The latter results demonstrate the involvementof sympathetic innervation of BAT in maintaining the differentiated state. These data alsosuggest that apoptosis in BAT may have occurred in our investigation of lambs held at warmvs cold temperatures (Smith et al., 2004). The maintenance of the BAT differentiated state byNE has been demonstrated previously. Norepinephrine induces the expression of UCP1 andmay increase of number of brown adipocytes (i.e. be mitogenic; Nedergaard et al., 1995).Conversely, cold exposure maintains the BAT viability by reducing the rate of DNA andprotein degradation (Desautels and Heal, 1999).

7. NUTRITION AND BROWN ADIPOSE TISSUE THERMOGENESIS

7.1. Dietary protein restriction

Malnutrition of the dam during late gestation has been shown to reduce neonatal calf survival(Hight, 1966; Corah et al., 1975). The inability of the neonate to maximize thermogenesis inresponse to cold stress during the early postnatal period may be caused by prepartum proteinand/or energy malnutrition. Previous studies have demonstrated that prepartum protein(Carstens et al., 1987) and energy (Ridder et al., 1991) restriction of nulliparous heifersreduced thermoneutral metabolism in newborn calves. Carstens et al. (1987) reported thatprepartum protein restriction reduced thermoneutral metabolic rates by 11.4%, even thoughbirth weights were unaffected by prepartum protein treatment.

Unlike previous reports, the thermoneutral metabolic rate of Wagyu × Angus newborncalves was not affected by prepartum protein restriction (Martin et al., 1997). Consistent withthe lack of a treatment effect on peak metabolic rates, prepartum protein restriction did notaffect perirenal adipose tissue mass or composition. Nor did prepartum protein restrictionalter UCP1 gene expression in BAT. Alexander (1978) fed high- and low-energy diets topregnant ewes, beginning on day 90 of gestation, and found that prepartum energy restrictionreduced the proportional weight of perirenal adipose tissue (the primary BAT depot innewborn ruminants) by 17% in single and 24% in twin fetuses at 125 days of gestation.Furthermore, Tyzbir (1984) demonstrated that prepartum protein restriction of rats reducedBAT mass by 40–50% as well as BAT mitochondrial thermogenic capacity in newbornrat pups, even though birth weights were not affected by prepartum protein restriction. Incontrast, the NE-induced peak metabolic rate was the same in calves born to adequate- andrestricted-protein heifers in the investigation of Martin et al. (1997).

The results of Martin et al. (1997) may have been confounded by the unusual breed typeof calves used in that study (Wagyu × Angus crossbred calves). Wagyu calves have lowerbirth weights than Angus calves (Smith et al., 1992) and, unlike Angus or Brahman purebredcalves, subcutaneous adipose tissue of Wagyu crossbred calves contains no detectable brownadipocytes at birth (Martin et al., 1997, 1999). Wagyu calves represent a distinct genetic lineresulting from crossbreeding of native Japanese cattle in the mid-nineteenth century (Smithet al., 2001). Thus, the lack of effect of protein malnutrition on fetal calf thermogenesisreported by Martin et al. (1997) should not be considered typical for this species.

7.2. Essential fatty acid supplementation

Several studies have shown that brown fat thermogenic rates were higher and sympathetic activ-ity of brown fat increased (higher norepinephrine turnover rates) in rats fed diets containing


safflower oil (high in polyunsaturated fatty acids; PUFA) compared to rats fed diets containingbeef tallow (high in saturated fatty acids; SFA) (Takeuchi et al., 1995a; Matsuo et al., 1995).Additionally, Takeuchi et al. (1995b) found that serum T3 levels were higher in rats fed asafflower-oil diet compared to those fed a beef tallow-fat diet. Lammoglia et al. (1999) demon-strated that prenatal supplementation of cracked safflower seeds to pregnant cows affected coldtolerance of newborn calves (fig. 9). Calves were exposed to cold ambient temperatures startingat 4 h of age. Rectal temperatures in the cold were significantly higher in calves born to cows sup-plemented with safflower seeds than calves born to cows fed the control diet containing no addedfat. Taken together, these observations suggest that maternal supplementation of a bypass sourceof PUFA to pregnant ewes may have the potential to enhance fetal BAT development.

Additionally, studies using fish oils have indicated that the n-3 PUFA eicosapentaenoicacid (20:5n-3; EPA) and docosahexaenoic acid (22:6n-3; DHA) also are effective in stimu-lating BAT thermogenesis (Sadurskis et al., 1995; Oudart et al., 1997; Saha et al., 1998;Kawada et al., 1998). Polyunsaturated fatty acids, and especially n-3 PUFA, stimulate non-shivering thermogenesis in rodents, and they may do so by increasing NE turnover rate inBAT. Exposure of BAT to NE reduces the extent of apoptosis in response to warm exposure,which in turn should cause elevated thermogenesis relative to animals fed beef tallow or noadded n-3 PUFA. Therefore, n-3 PUFA may increase thermogenesis by depressing apoptosis,or even stimulating brown adipocyte differentiation. This is in contrast to results with spleniclymphocytes (Avula et al., 1999) and tumor cells (Das, 1999), in which n-3 PUFA promotedapoptosis, and indicates a tissue-specific effect of n-3 PUFA on BAT.

We recently fed pregnant ewes 2%, 4%, or 8% rumen-protected fat. The fat sources werehigh in either saturated/monounsaturated fatty acids or n-3 PUFA (formaldehyde-protectedsoy/linseed lipid). The PUFA-fed ewes had higher plasma concentrations of 18:2, 18:3n-3,and EPA, and lower concentrations of 16:0, 16:1, and 18:1, than ewes fed the saturated/monounsaturated fatty acid diet. The BAT of lambs born to PUFA-fed ewes had higher


Fig. 9. Effects of feeding supplemental fat (safflower seeds) during late gestation on cold tolerance ofnewborn calves. Adapted from Figure 1 of Lammoglia et al. (1999).

concentrations of 18:2, EPA, and DHA than lambs born to ewes fed the saturated/monoun-saturated fatty acid diet. However, BAT mass, cytochrome c oxidase activity, and GDP bindingwere not affected by level or source of dietary fat. Cold-induced rectal temperature responsesof lambs were not affected by source of prenatal fat. Therefore, unlike results with calves(Lammoglia et al., 1999), prenatal PUFA supplementation did not affect BAT thermogenicactivity or cold tolerance of newborn lambs. The effects of DHA or EPA on BAT thermogen-esis in lambs cannot be tested until a rumen-bypass source of these fatty acids is developed.

7.3. Copper supplementation

Although few studies have investigated the impact of prenatal dietary copper on thermo-metabolism, clinical evidence of a link between copper deficiency and cold intolerance existsin lambs. Copper plays an essential role in several copper-dependent enzyme systems thatregulate thermometabolism, including cytochrome c oxidase and dopamine-β-hydroxylase.Cytochrome c oxidase is the terminal enzyme in the electron transport system linkingsubstrate oxidation to oxidative phosphorylation (ATP synthesis) in mitochondria. Anothercopper-dependent enzyme, dopamine-β-hydroxylase, regulates the synthesis of NE fromdopamine in the sympathetic nervous system.

We examined the effects of prenatal dietary copper level on thermometabolism in lambs(Carstens et al., 1999). Twin-bearing ewes were assigned to low- or high-copper treatmentsduring the last trimester of gestation. Even though liver copper concentrations in newborn lambswere reduced 57% by the low-copper treatment (132 vs. 306 ppm copper DM), theselambs would not be classified as being copper-deficient. Despite the fact that the low-copperlambs were not copper-deficient, their rectal temperatures at 2 h of age were 3.3°F lowerthan lambs born to high-copper ewes (fig. 10). We subsequently found that NE turnoverrates in BAT of lambs at 12 h of age were decreased by the low-copper treatment (0.16 vs0.3 ng NE/mg BAT/h). This suggests that the low-copper treatment decreased dopamine-β-hydroxylase enzyme activity which impaired the thermogenic function of BAT. Additionalevidence to support this idea is the finding that low-copper lambs also had lower plasma T3

levels compared to high-copper lambs (fig. 10), even though plasma T4 levels were unaffectedby prenatal copper treatment. An important regulatory aspect of BAT thermogenesis is theactivation of 5′-deiodinase by NE release from the sympathetic nervous system in responseto cold stress. Locally synthesized T3 is a potent regulator of uncoupling protein gene expres-sion in BAT. Because more than 60% of circulating T3 levels in newborn lambs are derivedfrom the conversion of T4 to T3 in peripheral tissues by 5′-deiodinase enzyme (Klein et al.,1980), the fact that plasma T3 levels were depressed in low-copper lambs suggests that5′-deiodinase activity may have been impaired by the low-copper treatment indirectly througha reduction in NE stimulation.


Research continues to describe the development and metabolism of brown adipose tissuein lambs and calves. This is important in respect to both the welfare of the newborn animalsand the economic impact of neonatal mortality to the livestock industry. Ideally, productionstrategies such as supplemental n-3 PUFA or copper would improve BAT functionality andthereby increase newborn lamb or calf survival. Conversely, strategies to increase BAT massin the neonate may lead to increased adiposity in the mature animal if brown adipocytesdedifferentiate into white adipocytes during development.


Fig

. 10.

Eff

ect o

f pr

enat

al c

oppe

r tr

eatm

ent o

n re

ctal

tem

pera

ture

(le

ft p

anel

) an

d pl

asm

a tr

iiodo

thyr

onin

e co

ncen

trat

ions

(ri

ght p

anel

) in

new

born

lam

bs. F

rom

Fig

ure

1 of

Car

sten

s et

al.

(199

9).

REFERENCES

Alexander, G., 1978. Quantitative development of adipose tissue in foetal sheep. Aust. J. Biol. Sci. 31,489–503.

Alexander, G., Bennett, J.W., Gemmell, R.T., 1975. Brown adipose tissue in the new-born calf(Bos taurus). J. Physiol. 244, 223–234.

Avula, C.P., Zaman, A.K., Lawrence, R., Fernandes, G., 1999. Induction of apoptosis and apoptoticmediators in Balb/C splenic lymphocytes by dietary n-3 and n-6 fatty acids. Lipids 34, 921–927.

Azzam, S.M., Kinder, J.E., Nielsen, M.K, 1993. Environmental effects on neonatal mortality of beefcalves. J. Anim. Sci. 71, 282–290.

Ballard, F.J., Filsell, O.H., Jarrett, I.G., 1972. Effects of carbohydrate availability on lipogenesis in sheep.Biochem. J. 226, 193–200.

Bianco, A.C., Silva, J.E., 1988. Cold exposure rapidly induces virtual saturation of brown adipose tissuenuclear T3 receptors. Amer. J. Physiol. 255, E496–E503.

Brander, F., Keith, J.S., Trayhurn, P., 1993. A 27-mer oligonucleotide probe for the detection and meas-urement of the mRNA for uncoupling protein in brown adipose tissue of different species. Comp.Biochem. Physiol. 104B, 125–131.

Carstens, G.E., 1994. Cold thermoregulation in newborn calves. Vet. Clin. N. Amer.-Food Anim. Pr. 10,69–106.

Carstens, G.E., Eckert, J.C., Greene, L.W., Smith, S.B., 1999. Effects of prenatal dietary copper onendocrine control of brown fat thermogenesis in newborn lambs. In: Proceedings of IX SymposiumRuminant Physiology. S. Afr. J. Anim. Sci. 29, 429–430.

Carstens, G.E., Johnson, D.E., Holland, M.D., Odde, K.G., 1987. Effects of prepartum protein nutritionand birth weight on basal metabolism in bovine neonates. J. Anim. Sci. 65, 745–751.

Carstens, G.E., Mostyn, P.C., Chapman, S.A., Randel, R.D., 1995. Prenatal and postnatal changes inbrown adipose tissue thermogenesis in the neonatal calf. In: Energy Metabolism of Farm Animals.EAAP Publ. No. 76. Pudoc, Wageningen, The Netherlands, pp. 101–104.

Casteilla, L., Champigny, O., Bouillaud, F., Robelin, J., Ricquier, D., 1989. Sequential changes in theexpression of mitochondrial protein mRNA during the development of brown adipose tissue in bovineand ovine species: sudden occurrence of uncoupling protein mRNA during embryogenesis and its dis-appearance after birth. Biochem. J. 257, 665–671.

Casteilla, L., Muzzin, P., Revelli, J.P., Ricquier, D., Giacobino, J.P., 1994. Expression of β1- andβ3-adrenergic-receptor messages and adenylate cyclase β-adrenergic response in bovine perirenaladipose tissue during its transformation from brown to white fat. Biochem. J. 297, 93–97.

Champigny, O., Ricquier, D., Blondel, O., Mayers, R.M., Briscoe, M.G., Holloway, B.R., 1991. β3 adren-ergic receptor stimulation restores message and expression of brown-bat mitochondrial uncoupling inadult dogs. Proc. Natl. Acad. Sci. USA 88, 1074–1077.

Corah, L.R., Dunn, T.G., Kaltenbach, C.C., 1975. Influence of prepartum nutrition on the reproductiveperformance of beef females and the performance of their progeny. J. Anim. Sci. 41, 819−824.

Das, U.N., 1999. Essential fatty acids, lipid peroxidation and apoptosis. Prostagland. Leuk. Essent. FattyAcids 61, 157–163.

Desautels, M., Heal, S., 1999. Differentiation-dependent inhibition of proteolysis by norepinephrine inbrown adipocytes. Amer. J. Physiol. 277, E215–Ε222.

Géloën, A., Collet, A.J., Guay, G., Bukowiecke, L.J., 1988. β-adrenergic stimulation of brown adipocyteproliferation. Am. J. Physiol. 254, C175–C182.

Giralt, M., Casteilla, L., Viñas, O., Mampel, T., Iglesias, R., Robelin, J., Villarroya, R., 1989. Iodothyronine5′-deiodinase activity as an early event of prenatal brown-fat differentiation in bovine development.Biochem. J. 259, 555–559.

Hight, G.K., 1966. The effects of undernutrition in late pregnancy on beef cattle production. NZ J. Agr. Res.9, 479–490.

Himms-Hagen, J., 1986. Brown adipose tissue and cold-acclimation. In: Trayhurn, T., Nicholls, D.G. (Eds.),Brown Adipose Tissue. Edward Arnold, Baltimore, MD, pp. 214–268.

Ingle, D.L., Bauman, D.E., Garrigus, U.S., 1972. Lipogenesis in the ruminant: in vivo site of fatty acidsynthesis in sheep. J. Nutr. 102, 617–624.

Kawada, T., Kayahashi, S., Hida, Y., Koga, K., Nadachi, Y., Fushiki, T., 1998. Fish (Bonito) oil supple-mentation enhances the expression of uncoupling protein in brown adipose tissue. J. Agr. Food Chem.46, 1225–1227.


Klein, A.H., Oddie, T.H., Fisher, D.A., 1980. Iodothyronine kinetic studies in the newborn lamb. J. Dev.Physiol. 2, 29–36.

Lammoglia, M.A., Bellows, R.A., Grings, E.E., Bergman, J.W., 1999. Effects of prepartum supplemen-tary fat and muscle hypertrophy genotype on cold tolerance in newborn calves. J. Anim. Sci. 77,2227–2233.

Landis, M.D., Carstens, G.E., McPhail, E.G., Randel, R.D., Green, K.K., Slay, L., Smith, S.B., 2002.Ontogenic development of brown adipose tissue in Angus and Brahman fetal calves. J. Anim. Sci. 80,591–601.

Lindquist, J.M., Rehnmark, S., 1998. Ambient temperature regulation of apoptosis in brown adiposetissue. J. Biol. Chem. 273, 30147–30156.

Martin, G.S., Carstens, G.E., Taylor, T.L., Eli, A.G., Tarrant, M., Britain, K., Smith, S.B., 1999.Metabolism and morphology of brown adipose tissue from Brahman and Angus newborn calves.J. Anim. Sci. 77, 388–399.

Martin, G.S., Carstens, G.E., Taylor, T.L., Sweatt, C.R., Eli, A.G., Lunt, D.K., Smith, S.B., 1997.Prepartum protein restriction does not alter norepinephrine-induced thermogenesis or brown adiposetissue function in newborn calves. J. Nutr. 127, 1929–1937.

Matsuo, T., Shimomura, Y., Saitoh, S., Tokuyama, K., Takeuchi, H., Suzuki, M., 1995. Sympatheticactivity is lower in rats fed a beef tallow diet than in rats fed a safflower diet. Metabolism 44,934–939.

Mersmann, H.J., 1996. Evidence of classic beta β3-adrenergic receptors in porcine adipocytes. J. Anim.Sci. 74, 984–992.

Milner, R.E., Wilson, S., Arch, J.R.S., Trayhurn, P., 1988. Acute effects of a β-adrenoceptor agonist (BRL26830A) on rat brown adipose tissue mitochondria: increased GDP binding and GDP-sensitiveproton conductance without changes in the concentration of uncoupling protein. Biochem. J. 249,759–763.

Moulin, K., Arnaud, E., Nibbelink, M., Viguerie-Bascands, N., Pénicaud, L., Casteilla, L., 2001a.Cloning of BUG demonstrates the existence of a brown adipocyte distinct from a white one. Int.J. Obes. 25, 1413–1441.

Moulin, K., Truel, N., André, M., Arnaud, E., Nibbelink, M., Cousin, B., Dani, C., Pénicaud, L, Casteilla, L.,2001b. Emergence during development of the white-adipocyte cell phenotype is independent of thebrown-adipocyte phenotype. Biochem. J. 356, 659–664.

Napolitano, L., 1963. The differentiation of white adipose cells: an electron microscopy study. J. CellBiol. 18, 663–679.

Nedergaard, J., Connolly, E., Cannon, B., 1986. Brown adipose tissue in the mammalian neonate.In: Trayhurn, P., Nicholls, D.G. (Eds.), Brown Adipose Tissue. Edward Arnold, London, pp. 152–213.

Nedergaard, J., Herron, D., Jacobsson, A., Rehnmark, S., Cannon, B., 1995. Norepinephrine as a morphogen?Its unique interaction with brown adipose tissue. Int. J. Dev. Biol. 39, 827–837.

Nougues, J., Reyne, Y., Champigny, O., Holloway, B., Casteilla, L., Ricquier, D., 1993. The β3-adrenoceptoragonist ICI-D7114 is not as efficient on reinduction of uncoupling protein mRNA in sheep as it isdogs and smaller species. J. Anim. Sci. 71, 2388–2394.

Okamoto, M., Robinson, J.B., Christopherson, R.J., Young, B.A., 1986. Summit metabolism of newborncalves with and without colostrum feeding. Can. J. Anim. Sci. 66, 937–944.

Oudart, H., Groscolas, R., Calgari, C., Nibbelink, M., Leray, C., Le Mayo, Y., Malan, A., 1997. Brownfat thermogenesis in rats fed high-fat diets enriched with n-3 polyunsaturated fatty acids. Int. J. Obes.21, 955–962.

Puig-Domingo, M., Guerrero, J.M., Vaughan, M.K., Little, J.C., Reiter, R.J., 1989. Activation of cere-brocortical type II 5′-deiodinase activity in Syrian hamsters kept under short photoperiod and reducedambient temperature. Brain Res. Bull. 22, 975–979.

Ridder, T.A., Young, J.W., Anderson, K.A., Lodman, D.W., Odde, K.G., Johnson, D.E., 1991. Effects ofprepartum energy nutrition and body condition on birthweight and basal metabolism in bovineneonates. J. Anim. Sci. 69, Suppl. 1, 450.

Robinson, J.B., Young, B.A., 1988a. Metabolic heat production of neonatal calves during hypothermiaand recovery. J. Anim. Sci. 66, 2538–2544.

Robinson, J.B., Young, B.A., 1988b. Recovery of neonatal lambs from hypothermia with thermal assis-tance. Can. J. Anim. Sci. 68, 183–190.

Sadurskis, I., Dicker, A., Cannon, B., Nedergaard, J., 1995. Polyunsaturated fatty acids recruit brownadipose tissue: increased UCP content and NST capacity. Amer. J. Physiol. 269, E351–E360.


Saha, S.K., Ohinata, H., Ohno, T., Kuroshima, A., 1998. Thermogenesis and fatty acid composition ofbrown adipose tissue in rats rendered hyperthyroid and hypothyroid – with special reference todocosahexanoic acid. Jpn. J. Physiol. 48, 355–364.

Smith, S.B., Carstens, G.E., Randel, R.D., Mersmann, H.J., Lunt, D.K., 2004. Brown adipose tissuedevelopment and metabolism in ruminants. J. Anim. Sci. 82, 942–954.

Smith, S.B., Prior, R.L., 1986. Comparison of lipogenesis and glucose metabolism between ovine andbovine adipose tissues. J. Nutr. 116, 1279–1286.

Smith, S.B., Sanders, J.O., Lunt, D.K., 1992. Evaluation of birth and weaning characteristics of halfbloodand three-quarter blood Wagyu-Angus calves. McGregor Field Day Report, Texas Agric. Exp. StationTech. Rep. 92-1, pp. 60–64.

Smith, S.B., Zembayashi, M., Lunt, D.K., Sanders, J.O., Gilbert, C.D., 2001. Carcass traits andmicrosatellite distributions in offspring of sires from three geographical regions of Japan. J. Anim.Sci. 79, 3041–3051.

Stott, A.W., Slee, J., 1985. The effect of environmental temperature during pregnancy on thermoregula-tion in the newborn lamb. Anim. Prod. 41, 341–347.

Symonds, M.E., Bryant, M.J., Clarke, L., Darby, C.J., Lomax, M.A., 1992. Effect of maternal cold expo-sure on brown adipose tissue and thermogenesis in the neonatal lamb. J. Physiol. 455, 487–502.

Takeuchi, J., Matsuo, T., Tokuyama, K., Shimomura, Y., Suzuki, M., 1995a. Diet-induced thermogenesisis lower in rats fed a lard diet than in those fed a high oleic sunflower oil diet, a safflower oil diet ora linseed oil diet. J. Nutr. 125, 920–925.

Takeuchi, J., Matsuo, T., Tokuyama, K., Suzuki, M., 1995b. Serum triiodothyronine concentration andNa+,K+-ATPase activity in liver and skeletal muscle are influenced by dietary fat type in rats. J. Nutr.125, 2364–2369.

Trayhurn, P., Thomas, M.E.A., Keith, J.S., 1993. Postnatal development of uncoupling protein, uncou-pling protein mRNA and GLUT4 in adipose tissues of goats. Amer. J. Physiol. 265, R676–R682.

Tyzbir, R.S., 1984. Altered brown adipose tissue mitochondrial function in neonates born to rats overfedfoods of various protein contents. J. Nutr. 114, 234–237.

Vernon, R.G., Robertson, J.P., Clegg, R.A., Flint, D.J., 1981. Aspects of adipose-tissue metabolism infoetal lambs. Biochem. J. 196, 819–824.


323

13 Interorgan lipid and fatty acid metabolismin growing ruminants

J. K. Drackley

Department of Animal Sciences, University of Illinois, Urbana, IL 61801, USA

Lipid metabolism is a dynamic and critically important function in growing ruminants.Preruminants consume a high-fat diet and deposit much of the dietary lipid in adipose tissue.During postweaning growth, dietary fatty acid intake is low and long-chain fatty acid syn-thesis increases in adipose tissue depots as the animal approaches physiological maturity.Deposition of long-chain fatty acids synthesized within adipose or taken up from bloodlipoproteins is extensive in near-mature fattening ruminants. Lipoprotein lipase is a key regu-latory enzyme in determining interorgan disposition of long-chain fatty acids from circulatingchylomicrons and very low-density lipoproteins in growing ruminants. Considerable researchin recent years has characterized the regulation of fatty acid esterification in adipose tissue.The role of molecules such as leptin and tumor necrosis factor α, which are synthesized andsecreted by adipose tissue, is only beginning to be elucidated in growing ruminants. The typeof dietary fat affects lipid metabolism in liver of preruminants, but surprisingly little is knownabout the developmental changes in hepatic metabolism of fatty acids in growing ruminants.Although much recent progress has been made in understanding regulation of interorgan lipidmetabolism in growing ruminants, many fundamental questions remain.

1. INTRODUCTION

Lipid metabolism plays a dynamic role during growth in ruminant animals. Lambs and calvesare born with minimal body lipid, accrete body lipid rapidly during suckling of fat-rich milk,undergo minimal fat deposition during the growth phase after weaning, and then again changeto fat deposition as the animal approaches physiological maturity. During the suckling ormilk-feeding period, the young ruminant (preruminant) functions as a monogastric animaland largely deposits the long-chain fatty acids (LCFA) from milk fat into adipose tissue forstorage. As skeletal and muscle growth near completion, adipose storage of triacylglycerol (TG)increases from a combination of both de novo lipogenesis (primarily from ruminally derivedacetate) and deposition of absorbed LCFA. Use of LCFA for fuel in well-fed ruminants is



relatively minimal, except in heart tissue and in skeletal muscle during exercise. Interorgantransport of LCFA is accomplished both by circulation of free or nonesterifed fatty acids(NEFA) and by the various classes of plasma lipoproteins.

Metabolism of LCFA and other lipids assumes obvious importance to the growing ruminantas a source of membrane components, signaling molecules, and a reserve of readily availableenergy. Lipid metabolism occupies a central position in the determination of energetic effi-ciency of growth and as a result has major impact on the profitability of meat animalproduction. Moreover, content and composition of carcass lipid has become an increasinglyimportant consideration for consumers of ruminant animal products. Consequently, ruminantlipid metabolism remains of major importance.

This chapter highlights some aspects where recent research has improved our understand-ing of lipid metabolism in growing ruminants. The major focus is on the interorganrelationships of lipid metabolism during growth, and is not intended to be a comprehensiveand exhaustive review of the literature. A number of excellent comprehensive and authorita-tive reviews on various aspects of lipid metabolism in ruminants are available (e.g. Noble,1978; Bell, 1980; Vernon, 1980; Noble and Shand, 1982; Bauchart, 1993; Chilliard, 1993;Jenkins, 1993) and key reviews are cited where applicable. For a general discussion of lipidmetabolism in domestic animals, see Drackley (2000).

2. DIGESTION AND ABSORPTION OF DIETARY LIPIDS

2.1. Preruminants

Preruminants fed milk or milk replacer consume a relatively high-fat diet. Bovine milk con-tains 29–31% fat on a dry solids basis; milk of goats and sheep may contain about 34% and40% of the solids as fat, respectively (Jenness, 1985). Most commercial milk replacers con-tain between 12% and 20% fat on a dry solids basis (Davis and Drackley, 1998). Dietary fatin milk or milk replacers consists primarily of TG, which are digested in the small intestineand packaged into chylomicrons for distribution throughout the body. The LCFA of dietaryorigin are delivered to tissues for oxidative use (primarily heart and skeletal muscle) or fordeposition in adipose tissue. The remaining components of the chylomicron particle (choles-terol, cholesterol esters, phospholipids, and apoproteins) participate in additional cycles oflipoprotein metabolism to distribute cholesterol and essential fatty acids (EFA) throughoutthe body for cell membrane and steroid hormone biosynthesis.

Preruminants digest the lipids in milk with high efficiency (typically greater than 97%;Toullec and Mathieu, 1969). A number of other fat sources in milk replacers can be welldigested (>90%) by preruminants if properly emulsified, including tallow, lard, and coconutoil (Toullec and Mathieu, 1969; Davis and Drackley, 1998). The milk fat of ruminants con-tains a relatively large proportion of short- and medium-chain fatty acids, probably as a strategyto maintain fluidity of the fat in the face of the mostly saturated LCFA that reach the maternalduodenum as a result of ruminal biohydrogenation of polyunsaturated LCFA consumed fromthe herbivorous diet.

Milk fat is entrapped in the casein coagulum in the abomasum, and is released as the caseinclot undergoes initial digestion. Within the abomasum, fat digestion is initiated by action ofan acid lipase secreted into the saliva by glands in the glossoepiglottic or pharyngeal area,which is highly homologous to the gastric lipases of many other species (Gargouri et al.,1989). About one-third of fatty acids in milk TG are hydrolyzed from the glycerol backbonein the abomasum (Edwards-Webb and Thompson, 1978). Activity of the acid lipase has been

J. K. Drackley324

thought to be greater toward the short- and medium-chain fatty acids (Edwards-Webb andThompson, 1978), although it has been shown recently that the acid lipase has selectivity bothfor short-chain fatty acids and for fatty acids on the sn-3 position of milk TG (Villenueve et al.,1996). Furthermore, specificity of the enzyme in these regards appears to vary inversely withthe same characteristics in mothers’ milk; i.e. enzyme specificity is highest in goat kids, inwhich the content of short-chain fatty acids is lower than in cows’ milk, and enzyme speci-ficity is lower in the calf where cows’ milk is rich in short-chain fatty acids (Jenness, 1985).

The metabolic significance of this preduodenal digestion is not certain but several possibil-ities exist. Short-chain fatty acids are absorbed into the portal vein and are extensively clearedby the liver. In this regard they are considered “obligate fuels” for the liver (see Chapter 9 byOdle et al., this volume). In preruminants consuming mothers’ milk, a firm coagulum is formedin the abomasum, which slows the flow of casein and long-chain fatty acids from the abomasum.However, lactose and the whey proteins are expelled from the coagulum, and are available forintestinal digestion much more quickly after a meal. Perhaps the initial digestion of milk fat,particularly the hydrolysis of short-chain fatty acids, provides a readily available fuel for theliver to spare amino acids derived from digestion of whey proteins for protein synthesis, and toprovide the ATP necessary to drive that protein synthesis. Moreover, the initial hydrolysis ofshort-chain fatty acids from milk fat may increase emulsification of fat droplets in the intestine(Armand et al., 1994) and may increase subsequent hydrolysis by pancreatic lipases in the smallintestine (Borel et al., 1994). While the role of gastric or preduodenal lipases in total fat digestionhas been shown to be larger than previously considered (Gooden, 1973), this may be especiallytrue in the young preruminant in which pancreatic lipase activity is immature. The whey proteinβ-lactoglobulin may bind and remove fatty acids from the acid lipase to prevent end-productinhibition (Perez et al., 1992), although the extent to which this occurs is questionable giventhat β-lactoglobulin is not retained in the abomasal coagulum.

While intestinal hydrolysis of the partially hydrolyzed dietary lipids traditionally has beenconsidered to occur by action of the colipase-dependent pancreatic lipase, recent evidence hasdemonstrated that the pancreas of most mammalian species also secretes a bile salt-activatedlipase (Wang and Hartsuck, 1993). This enzyme recently has been purified, characterized, andsequenced from bovine pancreas (Tanaka et al., 1999). The enzyme exerts considerable cat-alytic activity toward TG in the presence of bile salts, and has basal activity toward shorter-chainTG even in the absence of bile salts. The enzyme also is active in catalysis of phospholipids;indeed, the enzyme activity was first identified in the bovine as a lysophosholipase (van denBosch et al., 1993). In addition, the enzyme also hydrolyzes cholesterol esters, phosphatidyl-choline, and fatty acid esters of vitamins A and E (Wang and Hartsuck, 1993). The latter rolesmay be the more physiologically relevant in preruminants; alternatively, the apparent redun-dancy of having two distinct pancreatic enzymes active toward TG may be yet anotherexample of ensuring that almost all of the dietary TG presented to the young animal ishydrolyzed for absorption.

A number of fat sources are used in milk replacers (milk substitutes) fed to young preruminants. Commonly used sources include tallow, lard, and coconut oil (Davis andDrackley, 1998). Vegetable oils such as corn oil, soybean oil, and sunflower oil are not widelyused because of early research demonstrating that calves grew poorly and developed diarrhea(scouring) when fed milk replacers containing vegetable oils (Toullec and Mathieu, 1969;Jenkins et al., 1985, 1986). Canola oil did not seem to cause scouring (Jenkins et al., 1986).Subsequent research by the same group suggested that the scouring in earlier studies mayhave been a result of improper emulsification so that lipid particle size was too large (Jenkins,1988). Consequently, the degree to which animal-derived fats could be replaced with more

Interorgan lipid and fatty acid metabolism 325

highly unsaturated vegetable sources, if properly emulsified, without compromising calf healthor performance would appear to be an unresolved issue.

Free fatty acids also are less acceptable as lipid sources for young preruminants than TGof the same fatty acid composition (Jenkins et al., 1985). Although emulsification and dropletsize may have been concerns in that study, free fatty acids clearly inhibit feed intake in youngcalves (Spanski et al., 1997) as well as in adult cows (Drackley et al., 1992; Christensenet al., 1994; Bremmer et al., 1998). This effect is specific for unsaturated rather than saturatedfatty acids (Drackley et al., 1992; Christensen et al., 1994; Bremmer et al., 1998), and is dose-dependent (Overton et al., 1998; Drackley et al., 2000). Presence of the unsaturated free fattyacids in the duodenum interacts with receptors that in turn signal through glucagon-likepeptide-1 and perhaps cholecystokinin (Drackley et al., 2000; Benson and Reynolds, 2001)to decrease feed intake. Because dietary TG are not hydrolyzed until more distal regions ofthe jejunum, past the site of greatest density of these neuroendocrine cells, unsaturated TG donot inhibit feed intake as potently as do the same LCFA provided as free acids (Bremmer et al., 1998). This explains why decreases of dry matter intake have been more pronouncedwith postruminal administration of much smaller quantities of free fatty acids than whenmuch larger quantities of unsaturated TG were administered postruminally (e.g. Gagliostroand Chilliard, 1991; Drackley et al., 2000). Digestibility of the free fatty acids is high(Bremmer et al., 1998) and similar to that of TG in young calves, at least when some TG ispresent to furnish 2-monoglycerides in the intestine (Spanski et al., 1997).

2.2. Ruminants

Digestion and absorption of LCFA in ruminants have been addressed in a number of reviews(Noble, 1978; Moore and Christie, 1984; Bauchart, 1993; Jenkins, 1993; Doreau and Chilliard,1997) and will be discussed only briefly here. Lipid digestion in ruminants begins in the rumen(see Jenkins, 1993, for review). A number of ruminal microorganisms are actively lipolytic,resulting in extensive hydrolysis of most dietary complex lipids. The resulting free fatty acidsundergo varying degrees of biohydrogenation, with production of the saturated stearic acid,as well as smaller quantities of trans-unsaturated monoenes and dienes that leave the rumenfor absorption. The microbial population also synthesizes a variety of fatty acids, principallyodd-chain and branched-chain acids of 15 to 17 carbons, by elongating shorter-chain fattyacids (<14 carbons) from the diet. The microbes also synthesize phospholipids.

Lipids in digesta reaching the postruminal tract consist mainly of mostly saturated freefatty acids adsorbed to the surface of feed particles and bacteria, with the remainder beingpredominantly phospholipids and sterol esters as components of microbial cells. Because ofthe low pH in the abomasum and duodenum of ruminants (pH 2.0–2.5), the free fatty acidsexist in the protonated state, which facilitates their adsorption to the surface of feed particles.The strong detergent properties of bile salts secreted in the upper duodenum serve to desorbLCFA from particulate matter, with formation of a liquid crystalline phase. Subsequent for-mation of lysophosphatidylcholine (lysolecithin) from phosphatidylcholine (lecithin, frombile or from acid-mediated disruption of rumen microbial cells) by pancreatic phospholipase A2

promotes formation of micelles. Stable micelles formed from LCFA, lysolecithin, and bilesalts function to move lipid across the unstirred water layer of the small intestinal epithelium,where absorption of free LCFA and lysolecithin can occur by diffusion.

Stearic acid is the predominant lipid to reach the small intestine. Ruminants are able toabsorb saturated LCFA such as palmitic acid and stearic acid with substantially greater effi-ciency than nonruminants (Moore and Christie, 1984). One reason for this adaptation is the

J. K. Drackley326

reliance on lysolecithin as the major micelle stabilizer. Swelling amphiphiles are substances thatcan expand the volume of bile salt micelles and their hydrophobic interior in the aqueous envi-ronment of the small intestinal lumen (Small, 1968). Of all the naturally occurring swellingamphiphiles (monoglycerides, medium-chain fatty acids, long-chain unsaturated fatty acids,phospholipids, and lysophospholipids), lysolecithin is the most efficient at increasing the solu-bility of stearic acid. For example, lysolecithin increases the partitioning of stearic acid into themicelle by 115%, compared with only 36% for 1-monolein (Freeman, 1984). A second factorcontributing to greater absorption of long-chain saturated fatty acids in ruminants is the lowerpH in the upper small intestine, ranging from about 3.0 in the duodenum to about 6.0 in themid-jejunum. The relatively low pH helps minimize formation of calcium soaps of palmitateand stearate, which has long been known to decrease absorption of these fatty acids in non-ruminants (Cheng et al., 1949). In ruminants, the major bile salt is taurocholate rather thanglycocholate; the lower pKa of taurocholate (2.0) is likely an advantage since it is less likelyto become insoluble in the acid conditions of the ruminant small intestine than is glycocholate(pKa 4.7; Harrison and Leat, 1975).

True digestibility of LCFA generally is quite high in ruminants (Moore and Christie, 1984).Based on results of Palmquist (1991), true digestibility of LCFA may decline with increasingLCFA intake. As discussed by Bauchart (1993), this decrease suggests that pancreatic phos-pholipase activity and bile lipids (phospholipids and bile salts) may become limiting forabsorption of large dietary loads of LCFA.

2.3. Changes during the weaning transition from preruminant to ruminant

During the weaning transition, the young animal changes to a relatively low-fat diet, in whichdietary lipids may constitute only 2–6% of dry matter. These lipids typically consist of galacto-lipids and phospholipids from forages and TG from cereals and oilseeds (Noble, 1978). Duringthe weaning transition, the dietary supply of glucose largely ceases as rumen microbial fermen-tation of dietary carbohydrates is established. The resultant short-chain or volatile fatty acids(VFA), particularly acetate, become the major oxidative fuel for most tissues of the ruminant.Acetate also becomes the major precursor for lipogenesis. As the young ruminant begins to con-sume solid food, the ruminal tissue function, capacity, and microbial activity increases (seeDavis and Drackley, 1998), leading eventually to the pattern discussed for ruminants.

During the transition period when the young animal is receiving both a liquid diet and dryfeed, lipids in the liquid (milk) diet continue to reach the abomasum and small intestine via clo-sure of the esophageal groove. Few data are available that quantify LCFA absorption during thetransition from preruminant to ruminant. Spanski et al. (1997) found that apparent total tractdigestibility of LCFA by calves fed a control (starter) diet averaged about 82% and did not differamong measurements at 6, 8, and 10 weeks of age. Digestibility of LCFA when a liquid sup-plement was provided that contained either lard TG or a mixture of lard TG and free fatty acidsfrom lard was not appreciably different from LCFA digestibility of the control diet.

An enigma exists for transitioning ruminants, as well as for pigs, in that digestibility andutilization of LCFA is markedly less after weaning to dry feed than for LCFA use from theliquid diet before weaning. Indeed, these changes seem to occur as abruptly as the weaningprocess itself. Before weaning preruminants consume a diet in which a major portion ofdietary lipid is provided in milk or milk replacer, which may contain from 15% to 30% fat ona dry solids basis. After weaning, dry starter diets may contain only 3% to 6% dietary fat, yetdigestibility and use are much lower even with the lower fat content. For example, Fallon et al. (1986) showed that average daily gain (ADG) of body weight decreased as a fat


supplement (calcium soaps of palm oil fatty acids) was increased from 0 to 20% of the starterformulation. What factors suddenly become limiting for use of fat?

A portion of the decreased performance might be attributed to interference by fat withruminal fermentation; however, differences persist even when fats that should be relativelyinert in the young rumen are fed (Fallon et al., 1986). Increasing fat content of milk replaceror starter decreases development of starter intake (Kuehn et al., 1994). In contrast, increasingfat in a liquid diet (as in veal production) increases ADG, whereas increasing fat in starterdecreases ADG (Doppenberg and Palmquist, 1991). Because of ruminal hydrolysis of dietarylipids, substantial quantities of free fatty acids reach the upper small intestine when youngruminants consume starter feeds; the presence of free fatty acids in the duodenum may exertnegative feedback on appetite as described earlier.

Likely of major importance is the nature of the fat in the feed, i.e. a highly emulsified milkfat or fat in milk replacer vs. fat incorporated into a solid feed matrix. A limiting factor maybe that the compounds needed to emulsify lipid and form micelles in the small intestinallumen may be secreted in insufficient amounts in the young ruminant. For example, as dis-cussed earlier, when TG are no longer being hydrolyzed to 2-monoglycerides in the intestine,lysophosphatidylcholine becomes the predominant stabilizing compound of mixed micelles(Freeman, 1984). Whether biliary secretion of phospholipids to provide substrate for phospholipase-mediated production of lysophospholipids is limiting, or whether phospholi-pase activity itself might be limiting during this transition, has not been investigated. Gooden(1973) determined that conversion of lecithin to lysolecithin occurred much more rapidly inintestinal contents of ruminating calves compared with 1- to 2-week old calves, althoughmuch of the lower rate of conversion in milk-fed calves may have been from the presence ofmilk TG. The possibility that hepatic synthesis and secretion of bile salts might be inadequateto disperse lipids for micelle formation also does not appear to have been investigated inyoung ruminants.

3. INTESTINAL METABOLISM AND TRANSPORT OF ABSORBED DIETARY LIPIDS

General aspects of intestinal absorption of LCFA, re-synthesis of TG, and synthesis oflipoproteins have been reviewed recently by Phan and Tso (2001). Function of theseprocesses in preruminants and ruminants also has been reviewed (Noble and Shand, 1982;Bauchart, 1993; Doreau and Chilliard, 1997). Absorption of LCFA into intestinal epithelialcells generally has been assumed to occur by simple diffusion into and across the lipid-bilayermembrane, down the concentration gradient maintained by intracellular binding and metabolism.Recently, several putative transporter proteins for LCFA have been identified in other species(Chen et al., 2001). These proteins may facilitate uptake across cell membranes, especially atlow extracellular concentrations, and may be subject to metabolic regulation depending onthe physiological state of the animal. To date, however, no data are available on the presenceand role of such proteins in ruminant tissues.

Cytosolic fatty acid binding proteins (FABP) have been identified in most tissues ofrodents and other species that have been examined (see Glatz and van der Vusse, 1996, forreview). Intracellular FABP activity has been identified in the small intestine of the prerumi-nant calf (Jenkins, 1986). Binding activity in calf intestinal tissue was associated with aprotein fraction with molecular weight of approximately 12 kD, which is in the range reportedfor FABP of other species. Functions of intestinal FABP are not entirely clear, but it is intu-itive that FABP binds LCFA as they desorb from the plasma membrane, to keep intracellular

J. K. Drackley328

concentrations of potentially toxic free fatty acids low. In pigs, FABP is induced in the smallintestine by colostrum feeding after birth (Reinhart et al., 1992). Evidence has been presentedin other tissues and species for possible roles of FABP in directing LCFA to appropriate intra-cellular sites of metabolism, such as mitochondria or endoplasmic reticulum (see Glatz andvan der Vusse, 1996) or nucleus (Wolfrum et al., 2001).

Absorbed LCFA are activated by acyl-CoA synthetase located on the outer leaflet of theendoplasmic reticulum (Moore and Christie, 1984). After activation of LCFA to their acyl-CoAesters, they undergo re-synthesis to TG by one of two pathways (fig. 1). In the preruminant thatabsorbs both free LCFA and 2-monoglycerides, fatty acid esterification proceeds by themonoglyceride pathway, in which the remaining two positions on the glyceride molecule areesterified with acyl-CoA to form TG. In functioning ruminants, no 2-monoglyceride isabsorbed; consequently, TG synthesis in the intestine proceeds by the phosphatidate(Kennedy) pathway from glycerol-3-phosphate. While few data are available for ruminants,these pathways are not believed to be major sites for metabolic regulation. Rather, the path-ways function to repackage the amount of LCFA presented to the intestine into TG fordistribution throughout the body. Teleologically, one can argue that it is prudent for the animalto efficiently take up all LCFA available from the diet at any time, given the possibility thatdietary energy sources might not be available at some point in the future. Thus, regulationoccurs at disposition and storage sites within the body rather than at the site of assimilation.

Triacylglycerol synthesis occurs in concert with synthesis of the TG-rich lipoproteins, chylo-microns and very low-density lipoproteins (VLDL). The general scheme for synthesis andsecretion of these lipoproteins has been well described (Tso and Balint, 1986; Davidson andShelness, 2000; Phan and Tso, 2001), although specific details in ruminants are lacking. Theprimary structural apolipoprotein (apoprotein) of intestinal TG-rich lipoproteins is apoprotein-(apo)-B48, whereas the liver synthesizes primarily apo-B100. Ruminants synthesize apo-B


Fig. 1. Major pathways of esterification of fatty acids to glycerolipids in growing ruminants. Key enzymesinvolved: (1) glycerophosphate acyltransferase (GPAT), which may be mitochondrial or microsomal; (2) lysophosphatidate acyltransferase; (3) phosphatidate phosphohydrolase; (4) diacylglycerol acyltransferase;and (5) monoacylglycerol acyltransferase. Reactions 1 to 4 constitute the phosphatidate (Kennedy) pathwayand are found in adipose tissue, liver, and muscle. Activity of reaction 5 (monoacylglycerol acyltransferase) is largely confined to intestinal enterocytes during digestion of milk triglycerides in preruminants. Pi, inorganicphosphate. Adapted from Drackley (2000) based on Rule (1995).

proteins similar to those of other species. Chylomicrons and VLDL found in intestinal lymphof ruminants contain only the protein analogous to apo-B48, whereas the VLDL found inplasma contain both apo-B100- and apo-B48-like proteins (Laplaud et al., 1990, 1991). Bothforms of the protein are translated from the mRNA produced by a single gene. The shorterform (apo-B48) is produced as a result of a unique post-transcriptional mRNA editing mech-anism in intestinal cells (Davidson and Shelness, 2000). The apo-B mRNA of bovine intestineis almost completely edited to the shorter form (95%), whereas less editing (40%) occurs insheep intestine (Greeve et al., 1993). Apo-B synthesized and secreted by ruminant liver in theform of VLDL contains only the apo-B100-like protein (Laplaud et al., 1991).

Whether ruminants synthesize and secrete chylomicrons or just VLDL has been the sub-ject of considerable debate, but clear evidence for the existence of chylomicrons is availablein preruminants (Laplaud et al., 1990; Bennis et al., 1992). In functioning ruminants fed typ-ical diets, even those supplemented with fats, it appears that the intestinal cells mainly secreteVLDL (Bauchart, 1993). However, if polyunsaturated LCFA are infused postruminally inlarge amounts, more of the absorbed LCFA will be transported as TG in chylomicrons ratherthan VLDL (Harrison et al., 1974). Of interest also is the observation that ruminants, espe-cially calves fed high-fat milk diets, appear to secrete considerable amounts of VLDL into theportal vein rather than into the lymphatics (Bauchart et al., 1989; Laplaud et al., 1990).

The ontogeny of ruminant lipoproteins has been characterized by several researchers. NoVLDL were detected in plasma of fetal calves near term (Forte et al., 1981). Both low-densitylipoproteins (LDL) and high-density lipoproteins (HDL) were present in plasma of fetalcalves near term, with LDL being the more abundant lipoprotein class (Forte et al., 1981).Marcos et al. (1991) found that plasma TG decreased steadily from about day 115 to day 265of gestation in cattle; TG concentrations near term were very low, while apo-B concentrationshad decreased less, suggesting that VLDL decreased from mid- to late gestation but that LDLremained unchanged or decreased less. Declining concentrations of LDL were also reportedduring gestation in fetal lambs (Turley et al., 1996) so that HDL became the major lipopro-tein fraction at term (Noble and Shand, 1983). After birth, concentrations of VLDL increasewith consumption of colostrum and milk, while LDL concentrations decrease in calves(Jenkins et al., 1988; Marcos et al., 1991). During the suckling period, concentrations of LDLappear to increase in sheep (Turley et al., 1996) and goats (Bennis et al., 1992). With the onsetof suckling, HDL increases rapidly and become the predominant lipoprotein class in calves,lambs, and kids (Forte et al., 1981; Noble and Shand, 1983; Jenkins et al., 1988; Marcos et al.,1991; Bennis et al., 1992).

4. TISSUE UTILIZATION OF CIRCULATING TRIGLYCERIDES

Triglycerides in circulating chylomicrons and VLDL are hydrolyzed by the enzyme lipopro-tein lipase (LPL) found in peripheral tissues. Products of the LPL reaction are free fatty acids(i.e. NEFA) and monoglycerides, which are hydrolyzed by nonspecific lipases associatedwith peripheral tissues. Activity of LPL creates a locally increased NEFA concentration thatincreases the likelihood for NEFA uptake by cells, although not all NEFA are taken up by thetissue and NEFA concentration may increase in venous blood draining the tissue bed. Theunusual LPL enzyme is a glycosylated protein produced by tissue parenchymal cells (e.g.adipocytes), which is secreted from those cells and becomes active when anchored viaheparin sulfate proteoglycans on the inner surface of capillary endothelial cells (seeOlivecrona and Olivecrona, 1999, for review). In growing ruminants LPL is found predomi-nantly in adipose tissue, skeletal muscle, and heart. Very low activities of LPL were found in

J. K. Drackley330

small intestine, kidney, spleen, adrenals, ovary, and lung in bovines; the LPL mRNA also wasbarely detectable in these tissues and was undetectable in brain (Hocquette et al., 1998a).Only trace amounts of LPL activity were found in bovine liver, and the absence of detectablemRNA by Northern analysis indicates that this probably represented LPL from blood retainedin the tissue (Hocquette et al., 1998a). Activity and mRNA abundance were higher in heartand masseter muscle than in other oxidative and glycolytic muscles of calves, but LPL activ-ity was lower for bovine muscles in general than for rat muscles in parallel experiments(Hocquette et al., 1998a). Internal adipose depots (perirenal and omental) had greater LPLactivity and mRNA abundance than subcutaneous adipose tissue from fattened calves (Hocquetteet al., 1998a). Activity of LPL in perirenal adipose tissue was similar in cows and ewes, and LPLis transcribed from similar 3.4 and 3.8 kb mRNA in both species (Bonnet et al., 1998).

Extensive research has been conducted on LPL in laboratory animals, farm animals, andhumans over the last several decades. Despite this intense scrutiny, details of the regulationof LPL still are not resolved (Olivecrona and Olivecrona, 1999). Because LPL is synthesizedby parenchymal cells of the tissue and then transported across the capillary endothelial cells,where the enzyme is anchored on the capillary luminal surface to be active, its synthesis andregulation are understandably complex. Dogma developed on the basis of studies in ratsassumes that LPL in adipose tissue is upregulated during positive energy balance, decreasesduring fasting, and increases on refeeding after a fast, whereas in heart and skeletal musclethe enzyme activity changes less but in opposite direction to that in adipose, i.e. increasingduring fasting and decreasing with refeeding (e.g. Sugden et al., 1993; Cortright et al., 1997).In this way, heart and skeletal muscles would receive metabolic priority for use of TG circu-lating in VLDL derived from liver repackaging of NEFA mobilized from adipose tissue.

In both ewes and cows, underfeeding (20% of maintenance requirements) results in decreasedLPL activity and mRNA abundance in perirenal adipose tissue. Contrary to the dogma from lab-oratory animals, however, LPL activity in heart and skeletal muscle also decreases duringunderfeeding, and increases with refeeding (Bonnet et al., 2000; Faulconnier et al., 2001). Similarchanges occur in pigs (Enser, 1973). Bonnet et al. (2000) proposed a linear relationship betweenthe change in LPL activity in cardiac muscle and the ability of liver slices to secrete TG amongsix species (pig, sheep, guinea pig, rabbit, rat, and chicken), which makes teleological sense inthat circulating TG concentrations are maintained during fasting or feed restriction in species thatactively secrete VLDL from liver (e.g. chicken, rat, rabbit) but fall in species that do not (e.g. pig,sheep). In ruminants, therefore, during feed restriction the energy needs of heart and skeletalmuscle are met more via NEFA and ketone bodies and less from TG.

In contrast to the results for ruminants subjected to underfeeding, altering the plane of nutri-tion above maintenance does not produce the same changes in LPL among tissues. Andersenet al. (1996) fed groups of ewe lambs a diet in amounts to support daily gains of 0.15 or 0.25 kg.Adipose tissue LPL activity was greatest for lambs fed at the highest plane of nutrition,whereas skeletal muscle LPL was highest for lambs grown at the slower rate. For cardiacmuscle LPL, rates were greatest for lambs grown at the slowest rate and for a group of lambsfed for compensatory gain (0.33 kg/d) after a period of growth at rates equal to the slowest rateof gain. These data suggest that the concept of reciprocal regulation between adipose andmuscle LPL is valid when comparing differences in energy balance above maintenance, butnot in more extreme situations where animals are in negative energy balance.

The molecular basis for the differences in response of LPL among tissues within species,and across species, is not known. The cDNA for bovine LPL has been cloned (Senda et al.,1987) and the mRNA has been characterized in both sheep and cattle (Hocquette et al., 1998a;Bonnet et al., 1998). In sheep, the 3.4 kb mRNA was predominantly expressed in adipose


tissue, whereas the 3.8 kb mRNA was predominant in cardiac muscle, with no differentialregulation according to nutritional state (Bonnet et al., 2000). Large differences in LPLexpression among different muscles (e.g. heart and masseter vs. rectus abdominis) were par-alleled by changes in mRNA abundance, suggesting that transcriptional regulation is a majorfactor in differences among tissues.

Activity of LPL and the presence of its mRNA were detected in heart, masseter muscle, andperirenal adipose tissue of bovine fetuses (230–260 days in gestation), but activities andmRNA were lower than in growing calves (Hocquette et al., 1998a). Changes in LPL activityand mRNA in adipose tissues and muscles over the transition from preruminant to ruminantwere quantified by Hocquette et al. (2001). Activity of LPL was 2-fold lower in adipose tissuefrom weaned calves than from milk-fed calves; mRNA abundance was correspondingly lowerfor weaned calves as well. In contrast, neither LPL activity nor mRNA abundance were dif-ferent between weaned and milk-fed calves in any of seven skeletal muscles studied, with thenotable exception of the masseter muscle. Both LPL activity and mRNA abundance were morethan doubled in masseter of weaned calves. Contraction of the masseter, located in the cheek,would increase greatly in weaned calves with the increased frequency and workload of chew-ing; consequently, LCFA use as oxidative fuel might be expected. Activity, but not mRNAabundance, of LPL tended to decrease in heart from weaned calves.

Together, the available data implicate a role for LPL in controlling use of TG-FA by grow-ing ruminants. Indeed, for well-fed growing ruminants, Pethick and Dunshea (1993)calculated that most of the NEFA flux was derived from hydrolysis of lipoprotein TG by LPL.The potential role of LPL in determining fat deposition, especially for intramuscular fat(marbling) in fattening ruminants, is still an unanswered question.

5. LIPOPROTEIN METABOLISM

Lipoprotein metabolism in ruminants differs significantly from the more-studied laboratoryspecies such as rats, mice, and guinea pigs. An excellent comprehensive review (Bauchart,1993) is available to which the interested reader is referred and in which can be found refer-ences to other, older reviews. A schematic is presented in fig. 2 to describe the generalpatterns of lipoprotein metabolism in ruminants.

The TG-rich lipoproteins (chylomicrons and VLDL) function to deliver LCFA absorbedfrom the intestine to peripheral tissues. As discussed earlier (see section 3), intestinally syn-thesized VLDL predominate in ruminants, whereas chylomicrons would be more importantin the milk-fed preruminant. Following secretion into the lymphatics and entry into thevenous blood system, chylomicrons and intestinal VLDL acquire additional surface-coatcompounds, including phospholipids and apo-C proteins, from circulating HDL. Apo-CI andapo-CII are important for lipid binding and activation of LPL activity, respectively.Chylomicrons and VLDL thus activated by apo-CI interact with LPL in the capillaries ofperipheral tissue to catalyze “unloading” of TG in those tissues. Following LPL action, chy-lomicrons and VLDL are converted to chylomicron remnants and intermediate-densitylipoproteins (IDL), respectively. Chylomicron remnants and IDL containing apo-B48, i.e.those of intestinal origin, are found in extremely low concentrations in ruminants, suggestingthat they may be more rapidly cleared by the liver. As hydrolysis of TG occurs, the particlesize decreases and thus surface components (phospholipid, apo-A and apo-C) become exces-sive and are transferred back to HDL.

Most LDL are formed in circulation from IDL in the ruminant. The LDL make up asmall proportion of total lipoproteins (about 10%) in adult ruminants (Bauchart, 1993).

J. K. Drackley332

Low concentrations have been attributed to the low activities of cholesterol ester transferprotein (Ha and Barter, 1982) and hepatic lipase in ruminants (Cordle et al., 1986), which areinvolved in the production of LDL in other species. Plasma concentrations of LDL in rumi-nants are also controlled by expression and activity of tissue LDL receptors. The major sitesfor receptor-mediated removal of LDL are bone, intestine, and liver (Rudling and Peterson,1985). High concentrations of LDL receptors are also found in bovine adrenals and corporalutea (Rudling and Peterson, 1985), which indicates that LDL may be important sources ofcholesterol for steroid hormone biosynthesis in these tissues. Because LDL are rich in cho-lesterol esters and phospholipids containing linoleic acid, tissue uptake of LDL also served todistribute this EFA for membrane formation.

Bovine HDL are the main lipoprotein class in ruminants, constituting over 80% of total plasmalipoproteins in both preruminants and ruminants (Bauchart, 1993). The synthesis of HDLremains an enigma even in more well-studied species (Fielding and Fielding, 2001), but precur-sor particles (apo-AI) are synthesized by liver and small intestine (Fielding and Fielding, 1995).


Fig. 2. Schematic diagram showing major aspects of lipoprotein metabolism in growing ruminants.Chylomicrons (mainly in preruminants) or very low-density lipoproteins (VLDL) secreted from the intestineor liver acquire apoproteins- (apo)-CI and apo-CII from circulating high-density lipoproteins (HDL).Triacylglycerols in chylomicrons or VLDL are hydrolyzed by lipoprotein lipase (LPL) in peripheral tissues, which is activated by apo-CII and allows fatty acid uptake by tissues. Excess surface components(phospholipids, PL; apoproteins C and A; free cholesterol, chol) that arise as chylomicrons and VLDLdecrease in size during LPL catalysis of triacylglycerols are transferred to HDL. The chylomicron remnants are cleared by the liver. Remnants of VLDL are called intermediate density lipoproteins (IDL) andare either cleared by the liver or undergo further triacylglycerol hydrolysis to produce low-density lipoproteins(LDL). LDL are degraded in liver or after receptor-mediated uptake in peripheral tissues. HDL take up excesscholesterol (chol) from peripheral tissues and convert it to cholesterol esters by action of lecithin cholesterolacyltransferase (LCAT); lysolecithin is released into plasma and cholesterol esters enter the core of the HDL.Action of LCAT to produce cholesterol esters, and the uptake of excess surface components from chylomi-crons and VLDL, results in increasing size and decreasing buoyant density of lipid-poor heavy HDL (HDLH)to lipid-rich light HDL (HDLL). HDL can deliver cholesterol and essential fatty acids to tissues or return cholesterol to liver for conversion to bile salts. Many tissues also possess an HDL receptor that results inclearance of HDL particles. Adapted from Drackley (2000).

Genesis of pre-HDL particles appears to occur in circulation, either from chylomicrons andVLDL or within the interstitial spaces (Fielding and Fielding, 1995). Initial complexes ofapo-AI and phospholipids (pre-β-HDL) acquire cholesterol from tissue membranes by as yetundefined mechanisms that may involve some combination of simple diffusion, facilitated dif-fusion, active transport, or other mechanisms (see Fielding and Fielding, 2001, for review). Thelecithin-cholesterol acyltransferase (LCAT) reaction then functions to catalyze transfer of anunsaturated fatty acid (primarily linoleic) from the 2-position of phosphatidylcholine (lecithin)to cholesterol, both polar surface-coat lipids of HDL. The more nonpolar cholesterol estermigrates to the core of the HDL, while the lysolecithin is shed to serum albumin. Continuedtransfer of cholesterol esters to the core of the HDL causes their expansion in size, forming“light” HDL. The lack of cholesterol ester transfer protein, which would transfer cholesterolesters to LDL or VLDL, in ruminants and the high turnover of surface-coat lipids from VLDLresult in the large size of HDL (Bauchart, 1993). The presence of the very large and less dense(i.e. high lipid content) HDL in ruminants results in overlapping density ranges for LDL andHDL, thus making separation of LDL and HDL by density gradient ultracentrifugation proce-dures impossible in ruminants. Alternative approaches have been developed that rely oncombinations of affinity chromatography and centrifugal separations (see Bauchart, 1993 forreview).

The HDL of ruminants possesses apo-AIV, which is secreted by intestinal cells in TG-richlipoproteins and then transferred to HDL in lymph and plasma (Bauchart et al., 1989).Originally described as an activator of peripheral LPL, apo-AIV has been implicated in recentyears as a regulator of food intake in humans and laboratory animals (see Tso et al., 2001, forreview). An intestinal fat load stimulates synthesis and secretion of apo-AIV, which in turnacts to suppress additional feed intake. The potential roles of apo-AIV in regulation of feedintake in ruminants, or its response to supplemental fat, have not been investigated.

Because of their predominance in ruminants, HDL function as the main distribution vehi-cle for cholesterol and EFA. Specific receptors for HDL have been identified andcharacterized (Graham and Oram, 1987). This receptor recognizes apo-AI but not LDL orapo-E; the apo-E protein is not expressed to any extent in ruminants (Bauchart, 1993). Uptakeof HDL by bovine liver, and probably other tissues, is regulated by the density of the HDLreceptors, which are almost always occupied with HDL, rather than by concentration of HDLin plasma (Mendel et al., 1986). Cholesterol uptake by tissues is used for membrane synthe-sis, steroidogenesis, or, in the liver, bile salt synthesis. Uptake of cholesterol from peripheraltissues by HDL followed by clearance of HDL by the liver constitutes a cycle that has beenreferred to as “reverse cholesterol transport” in other species and is one component of theregulation of cholesterol homeostasis in the body (Fielding and Fielding, 1995).

Relatively little is known about cholesterol synthesis and homeostasis in ruminants. Themajor site of cholesterol synthesis appears to be the small intestine (Nestel et al., 1978), whichis in keeping with the low rate of production of VLDL by the ruminant liver (Kleppe et al.,1988; Graulet et al., 1998; Gruffat-Mouty et al., 1999), although the ruminant liver does syn-thesize some cholesterol. Transport of cholesterol from intestine and liver to other tissues byLDL and HDL furnishes the needs of those tissues for membrane structure and steroid hormone biosynthesis.

Although beyond the scope of this chapter, metabolism of the EFA, i.e. those LCFA pro-duced from elongation and desaturation of linoleic acid (18:2n-6) and linolenic acid(18 : 3n-3), is a particularly fascinating topic in ruminants (Noble, 1984). Little transfer ofLCFA occurs across the ruminant placenta, so the fetuses and young are born with whatwould be considered severely deficient status for EFA in other mammalian species (see Noble

J. K. Drackley334

and Shand, 1982, for review). Linoleic and linolenic acids in plasma phospholipids and cho-lesterol esters are extremely low at birth, but then increase substantially by 3 weeks with littleadditional change after weaning (Jenkins et al., 1988). After birth, the young receive someEFA from colostrum and milk, but again intakes are insufficient relative to other speciesbecause of the low EFA content in ruminant milk. Because functional deficiencies of EFA donot seem to occur in preruminants or ruminants, the ruminant animal must have developedextremely efficient mechanisms for capturing and maintaining EFA within the body.

Mechanisms for the conservation and concentration of EFA include lower rates of oxida-tion compared with more abundant saturated LCFA (Lindsay and Leat, 1977), which may bemediated via the low affinity of mitochondrial dehydrogenases for EFA (Reid and Husbands,1985). Other mechanisms include the high affinity of phospholipid esterification enzymes forincorporation of EFA (Lindsay and Leat, 1977), the high affinity of lecithin cholesterol acyl-transferase for linoleic acid (Noble et al., 1972), and the slow turnover of the phospholipidand cholesterol ester pools in plasma (Palmquist, 1976).

Interorgan transport and metabolic conversions of the lipoproteins serve as a vehicle for dis-tribution of EFA to cells throughout the body for incorporation into cell membranes. Othermethods by which tissues acquire EFA include uptake of EFA as free fatty acids, uptake of theEFA-acyl group from lysophosphatidylcholine, and local desaturation and elongation of freelinoleic acid (see review by Zhou and Nilsson, 2001). The extent to which peripheral tissuesof ruminants acquire EFA by uptake or transfer of phospholipids from lipoproteins to cellmembranes vs. receptor-mediated uptake of LDL or HDL is not well characterized. Zhou et al.(2002) demonstrated that plasma free arachidonic acid was the major source of arachidonicacid taken up by extrahepatic tissues in rats. Regardless, increasing dietary intake of fats andoils rich in n-6 or n-3 LCFA results in corresponding increases in contents of these LCFA inphosphatidylcholine and cholesterol esters in plasma of preruminants, and correspondingincreases in the n-6 and n-3 LCFA in phophatidylcholine and phosphatidylethanolamine inliver and muscle (Jenkins and Kramer, 1990). Similar changes occur in functioning ruminantsif dietary sources of EFA are protected from ruminal biohydrogenation (Ashes et al., 1995). Itappears that phosphatidylethanolamine in muscle, primarily found in the inner leaflet ofplasma membranes, may be particularly important for concentration of n-3 LCFA (Jenkins andKramer, 1990; Ashes et al., 1995).

While a considerable body of research has appeared in the last decade describing inter-organ lipoprotein metabolism in growing ruminants, these studies mostly were conducted oncalves that were maintained in the preruminant state on milk diets long after what would beconventional in North American production systems. Research on lipoprotein metabolism ingrowing ruminants is lacking. The significance and role in growth, therefore, is still largelyas speculative as it was over twenty years ago (Kris-Etherton and Etherton, 1982).

6. TISSUE FATTY ACID METABOLISM IN GROWING RUMINANTS

Excellent comprehensive reviews of lipid metabolism in adipose tissue (Vernon, 1980; Chilliard,1993) and liver, muscle, and other tissues (Bell, 1980) are available. Discussion here will mainlyconsider recent research on lipid metabolism in key organs relative to growth of ruminants.

6.1. Skeletal muscle and heart

During growth of ruminants, muscle protein accretion is afforded a higher metabolic prioritythan is lipid deposition. For example, Smith et al. (1992) fed groups of ovariectomized


Angus–Hereford heifers different amounts of the same diet to supply 0.76, 1.43, 1.74, and2.05 times the estimated requirements for metabolizable energy for a 128-day feeding period.Retention of dietary N increased linearly with increasing intake, whereas measurements ofadipose lipid synthesis increased only for groups fed 1.74 and 2.05 times maintenance. Rates ofprotein deposition in muscle (and whole carcass) decrease as the animal approaches maturity(Campbell, 1988).

Skeletal muscles contain differing proportions of fibers that are primarily oxidative (redmuscle), primarily glycolytic (white muscle), or a mixture of both. Metabolic properties mayvary even within the same muscle (Brandstetter et al., 1997). So-called red muscle is character-ized by a higher density of mitochondria than is found in white muscle. Furthermore, at leasttwo distinct subpopulations of mitochondria exist in ruminant muscle (Piot et al., 2000). Thesubsarcolemmal mitochondria are located just beneath the muscle membrane or sarcolemma,whereas the intermyofibrillar mitochondria are found inserted into the myofibrillar matrix.Piot et al. (2000) found that enzymes of oxidative metabolism and LCFA oxidation were pres-ent in greater specific activity in intermyofibrillar mitochondria than in subsarcolemmalmitochondria. Capacity of heart tissue for LCFA oxidation was much greater than that ofskeletal muscles (rectus abdominis and longissimus thoracis) from preruminant calves, due to higher density of both intermyofibrillar and subsarcolemmal mitochondria and to greaterspecific activities of oxidative enzymes within the mitochondria from heart.

In fed, growing ruminants, glucose, acetate, and β-hydroxybutyrate (BHBA) are the majorfuels for heart and skeletal muscle, potentially accounting for 31–57%, 15–29%, and 18%,respectively, of fuel for muscle oxidation at rest (Hocquette et al., 1998b; Hocquette andBauchart, 1999). In contrast, oxidation of NEFA at rest accounts for only about 5% of oxida-tive needs. Uptake of NEFA is linearly related to arterial concentration up to about 1 mM(Bell and Thompson, 1979). Muscle uptake of NEFA likely is facilitated by muscle-typeFABP (Moore et al., 1993). Uptake exceeds the amount oxidized immediately for energy,with the excess stored in an intracellular TG pool (Bell and Thompson, 1979). Although notinvestigated in ruminants, hormone-sensitive lipase is present in muscle from other speciesand responds to catecholamine stimulation to initiate intracellular hydrolysis of stored TG foroxidation within the muscle (Langfort et al., 1998).

Rate-limiting control of NEFA oxidation lies at the enzyme carnitine palmitoyltransferase-1(CPT-1), which governs entrance of fatty acyl-CoA into mitochondria. Malonyl-CoA is pro-duced by the muscle form of acetyl-CoA carboxylase and inhibits muscle CPT-1 (Winder,2001). While not yet characterized in ruminants, in laboratory rodents and humans malonyl-CoA is produced by the muscle form of acetyl-CoA carboxylase (Chien et al., 2000) anddegraded by malonyl-CoA decarboxylase (Young et al., 2001). Regulation of malonyl-CoAconcentration in muscle thereby represents an elegant control system that coordinates utiliza-tion of glucose and LCFA depending on substrate availability and muscle energy needs (seeWinder, 2001, for review).

Oxidative capacity per unit of tissue is lower in muscle from cattle than in muscle from rats(Ottemann-Abbamonte et al., 1998; Piot et al., 1998). Oxidation of palmitate by isolatedstrips of muscle in vitro appeared to be saturated at a palmitate concentration of 0.5 mM formuscle from adult cows, but was not yet saturated at 2.0 mM for rat muscle (Ottemann-Abbamonte et al., 1998). Piot et al. (1998) attributed the greater oxidative capacity of ratmuscle to a greater mitochondrial density; peroxisomal oxidative capacity was not differentbetween rats and preruminant calves but was lower for 15-month-old growing bulls. In thatsame study, total oxidation capacity of muscle homogenates was about 1.7-fold greater forpreruminant calves than for growing bulls. Piot et al. (1998) proposed that the difference

J. K. Drackley336

between preruminant and older ruminant muscle was related both to decreased relative quan-tity of mitochondria as well as to changes in the oxidative properties of muscle mitochondria.

Bartelds et al. (1998) studied the in vivo flux of energy supplying nutrients to the heart infetal, newborn (1 to 4 days), and 7-week-old lambs. In fetal lambs, NEFA contributed no energyto the heart. In newborn lambs, the supply of NEFA to the heart increased 10-fold, but heart stilldid not oxidize NEFA for energy. Glucose was the major energy source in both fetal and new-born lambs, accounting for 89% and 69% of oxygen consumption, respectively. By 7 weeks, theflux of NEFA through heart was increased 3-fold, and the supply and use of ketone bodies like-wise was increased. Similar developmental data for skeletal muscle are lacking.

6.2. Adipose tissue

Adipose tissue represents a complex assortment of anatomical depots, including internal (e.g. perirenal, omental), intermuscular, subcutaneous, and intramuscular. In growing rumi-nants, adipose tissue functions to accrete excess energy in the form of TG (see Chapter 10 byMersmann and Smith, this volume). The LCFA that are incorporated into adipose tissue TGduring growth and fattening can derive from de novo lipogenesis within the adipose tissue orvia uptake from blood. From a biological perspective, adipose tissue functions as a reserve ofenergy for situations of dietary shortage. Recent research has demonstrated that adipose is notan inert storage vessel, however, but also communicates with other organs and systems of thebody by synthesizing and secreting a variety of mediators such as leptin, tumor-necrosisfactor α (TNFα), and adipsin (see Vernon et al., 1999; Chilliard et al., 2000). Given its impor-tance in determining eating quality of ruminant meat, and in the production economics ofruminant agriculture, it is not surprising that lipid metabolism in adipose tissue has beenwidely studied.

Lipogenesis is low during the milk-feeding period (Pothoven et al., 1975), then increasesduring growth as rates of lean tissue growth decline (Pothoven and Beitz, 1973; Smith et al.,1984, 1987). Internal adipose tissues have greater rates of de novo lipogenesis in younger ani-mals, whereas in more mature animals subcutaneous depots have the greater activity (Ingle et al., 1972; Pothoven and Beitz, 1973; Hansen et al., 1995). Increasing feed intake increaseslipogenesis in subcutaneous adipose (Mills et al., 1989; Smith et al., 1992); conversely, feedrestriction during growth decreases lipogenic rates (Pothoven et al., 1975). The intermediarymetabolism of lipogenesis in adipose tissue appears to differ between sheep and cattle in sev-eral ways (Smith and Prior, 1986); regulation in sheep and cattle adipose tissue is discussedin detail by Smith (1995).

Differential activity of lipogenesis among adipose depots and with stage of growth mayrelate to adipocyte size. Hood and Allen (1978) found that lipogenic enzyme activities werepositively correlated with the cell volume of adipocytes across several anatomical sites insheep. More recently, Barber et al. (2000) showed that expression of mRNA for LPL andacetyl-CoA carboxylase-α (ACCα), the rate-limiting and regulated step in LCFA synthesis,per 106 cells was highly correlated with the size of adipocytes isolated from seven subcuta-neous and internal depots from wethers at slaughter. These data agree with other observationsthat lipogenesis is lower in adipose tissue with smaller adipose cell diameters, such as intra-muscular adipose or in adipose tissue from young animals (Smith, 1995). However, factorsother than adipocyte size also must impact metabolic activity among different adipose tissuedepots. For example, in fattened sheep lipogenesis from acetate was about 10-fold lower inperirenal adipose than in subcutaneous adipose tissue, but mean cell size and numbers of cellsper gram of adipose tissue differed by only 10% and 24%, respectively (Hansen et al., 1995).


Classic experiments by Hanson and Ballard (1967) demonstrated that acetate and not glu-cose was the major carbon source for LCFA synthesis in adipose tissues of ruminants. Apossible exception is intramuscular adipose depots in which glucose may account for 50–75%of LCFA carbon (Smith and Crouse, 1984). However, overall rates of de novo LCFA synthe-sis are low in intramuscular adipose, with rates of incorporation of acetate into LCFA only10–50% of those in subcutaneous depots (Smith and Crouse, 1984; Miller et al., 1991).Although rates of esterification of LCFA to glycerol-3-phosphate are lower in intramuscularadipose than in subcutaneous, differences in LCFA synthesis are much more pronounced(Smith et al., 1998). This suggests that uptake of preformed LCFA from blood TG by actionof LPL may be more important for lipid deposition in intramuscular adipocytes.

Uptake of preformed LCFA into adipose tissue can be quite substantial in preruminantsconsuming high-fat milk-based diets. For example, it can be calculated from data of Tikofsyet al. (2001), in which preruminant calves were isocalorically fed milk replacers containing14.8%, 21.6%, or 30.6% fat, that about 45% of the additional fat intake was deposited in thebody. The amount of LCFA originating from the diet in functioning ruminants is much morelimited but not insignificant. For example, typical forage- and grain-based diets for growingruminants contain only 3–4% LCFA, which would supply about 6–10% of the digestibleenergy intake. Even with near-maximal incorporation of supplemental fat into diets for rumi-nants, LCFA will usually supply less than 20% of the energy. However, the amount of LCFAsupplied may still contribute a substantial portion of fat deposited in adipose tissues.

An example of the estimated contribution of dietary LCFA supply to adipose fat depositionis shown in table 1, based on data from Zinn (1992). In this example, cattle fed basal dietsbased on corn or wheat had digestible LCFA intakes of about 144 g/d, whereas cattle fedcorn- or wheat-based diets supplemented with 6% yellow grease would have digestible LCFA

J. K. Drackley338

Table 1

Estimation of contribution of dietary fat to adipose lipid deposition and suppression of de novo fatty acid synthesis in fattening beef cattlea

Diet

Basal Basal plusVariable (no supplemental fat) 6% yellow grease

Dry matter (DM) intake, kg/d 7.82 7.42Dietary fatty acids (FA)b, % of DM 2.45 7.45FA intake, g/d 192 553Estimated digestible FA intakec, g/d 144 415Estimated FA deposited in adiposed, g/d 118 340Measured fat gain, g/d 480 550Estimated FA gaine, g/d 432 495Dietary FA as percentage of FA gain 27 69Estimated de novo FA synthesis f, g/d 314 155% Suppression of de novo synthesis – 51

aData for intake, diet composition, and fat gain are means from Zinn (1992).bReported dietary ether extract concentration multiplied by assumed 85% fatty acid content.cAssumes dietary fatty acids are 75% digestible (Zinn, 1989).dDigestible energy of fat equals metabolizable energy; assumes that efficiency of use of metabolizable energy forfat deposition (i.e. NEg) is 82% (Zinn, 1989).eAssumes adipose tissue triacylglycerol is 90% fatty acids.fEstimated fatty acid gain minus dietary fatty acids deposited.

intakes of about 415 g/d. Assuming that absorbed LCFA are used in fattening cattle with anefficiency of 82% (Zinn, 1989), cattle fed basal and fat-supplemented diets would deposit 118and 340 g/d of dietary LCFA. Whole-body LCFA gains averaged 432 and 495 g/d, respec-tively, which indicates that 27% and 69% of the total LCFA deposited originated from dietaryLCFA for control and fat-supplemented diets, respectively. By difference, whole-bodyde novo synthesis of LCFA would need to be decreased by about 51% when the fat-supplementeddiet was fed. Page et al. (1997) found that de novo synthesis of LCFA from acetate was decreasedby 29% in subcutaneous adipose tissue from steers fed a diet containing 30% whole cottonseed(which supplied an additional 6% lipid to the diet) compared with adipose tissue from controlsteers, suggesting that decreased lipogenesis of the magnitude estimated here (table 1) is notbiologically unrealistic.

Other research suggests that dietary lipid supplements also may decrease lipogenesis insheep. Hood et al. (1980) found that rumen-protected safflower oil, but not rumen-protectedtallow or palm oil, decreased lipogenesis from acetate by 43% in perirenal adipose tissue. Insubcutaneous adipose tissue, protected safflower oil decreased in vitro lipogenesis by 75%;differences due to protected tallow (−42%) and protected palm oil (−32%) were not statisti-cally significant (Hood et al., 1980). Lipogenesis, measured both in vitro and in vivo, wasdecreased in sheep fed calcium salts of palm oil (Moibi et al., 2000a). Decreased lipogenesisby dietary lipid was accompanied by a 30% decrease in fatty acid synthase activity in sub-cutaneous, mesenteric, and perirenal adipose tissues; in contrast, activity of acetyl-CoAcarboxylase actually was increased by supplemental fat (Moibi et al., 2000b). On the otherhand, lipogenesis in isolated adipocytes was not different between control sheep and those feda diet containing 5% soybean oil (Jenkins et al., 1994). Although the data are not conclusive,together these observations suggest that increased dietary LCFA, even if not protected fromruminal biohydrogenation, may suppress de novo lipogenesis from acetate in ruminant adi-pose tissue. This differs from the situation in rodents, in which only unsaturated LCFA havebeen shown to be effective at suppressing lipogenesis (see Clarke, 2001, for review).

Regardless of the source of LCFA available to adipose tissue, the ultimate pathway foraccretion of lipid stores is the series of enzymatic steps involved with attachment of LCFA toglycerol-3-phosphate (fig. 1). Considerable progress has been made in the last decade in char-acterizing the esterification pathway (see review by Rule, 1995, and references therein).Esterification activity is greater in bovine subcutaneous adipose tissue than in liver (Wilsonet al., 1992). Evidence indicates that phosphatidate phosphohydrolase may be the rate-regulatingstep in adipose but that glycerol-3-phosphate acyltransferase may be more likely to controlesterification in liver (Wilson et al., 1992; Smith et al., 1998). Changes in palmitate esterifi-cation and activity of glycerol-3-phosphate acyltransferase parallel changes in fat depositionand carcass fat thickness with different rates of gain and degrees of maturity in both bovineand ovine adipose tissue (Bouyekhf et al., 1992; West et al., 1994; Andersen et al., 1996).Palmitate esterification in vitro was decreased by >48% by the catecholamines clenbuterol,norepinephrine, and isoproterenol as well as by cAMP in subcutaneous adipose tissue from ewesfed at maintenance; however, for tissue from ewes fed on a high-energy diet for 6 weeks, onlyisoproterenol and cAMP inhibited esterification but to a lesser degree than for maintenance-fedlambs (Bouyekhf et al., 1993). A period of 72 h of starvation decreased LCFA esterificationand phosphatidate phosphohydrolase activity by about 50% in bovine subcutaneous adiposetissue, but not in intramuscular adipose tissue (Smith et al., 1998).

Changes in adipose esterification rates in parallel with positive energy balance and duringfattening suggest a major control by insulin. However, insulin has only modest effects on esterification activity in vitro (Jacobi and Miner, 2002), suggesting that removal of


antagonistic effects during positive energy balance may be the more dominant regulatoryinfluence. Other factors may be involved in stimulating esterification. For example, acylation-stimulating protein is an 8.9 kD protein generated within adipose tissue from components ofthe alternative complement pathway (see Cianflone et al., 1999 for review). This protein actsin an autocrine fashion to promote uptake and esterification of LCFA by adipocytes fromhumans and laboratory rodents as well as cell lines. Recently, Jacobi and Miner (2002)demonstrated that human acylation-stimulating protein increased acetate incorporation intoLCFA by 15–30% and oleate incorporation into TG by 10–25% in bovine adipose tissue, andthat this stimulation was not affected by feed restriction. It will be of interest to isolate thebovine counterpart of acylation-stimulating protein and to determine its role in growth underdifferent dietary regimes. For example, in other species, a factor carried by chylomicrons aftera meal is important in activating the synthesis of acylation-stimulating protein (Cianflone et al., 1999); does increased synthesis of acylation-stimulating protein during fat feeding inruminants contribute to the general increase in body fat deposition under those conditions?

In growing ruminants consuming adequate feed, the primary state is lipid accretion so thatlipid mobilization is a less important process. This contrasts with lactating ruminants, whichregularly undergo periods of intense lipid mobilization after parturition to support lactation(McNamara, 1994). Situations do occur with growing ruminants that demand lipid mobiliza-tion from adipose tissues to meet energy needs, for example, animals grazing poor-qualitypasture (due to environmental conditions or poor management). Agents that increase intra-cellular cAMP concentrations, such as epinephrine and norepinephrine, are the primary stimulifor increased lipolysis in sheep and cattle (Etherton et al., 1977) and exert lipolytic activity viabinding to β-adrenergic receptors (Houseknecht et al., 1996). The adrenergic receptors, throughprotein kinase A, lead to phosphorylation and activation of hormone-sensitive lipase (HSL). In laboratory animals, proteins called perilipins also are phosphorylated by β-adrenergic acti-vation (Clifford et al., 2000). When phosphorylated, perilipins appear to increase access ofactivated HSL to the lipid droplet, and thereby promote lipolysis.

Stimulatory effects of catecholamines on lipolysis in ruminants are enhanced by fasting(DiMarco et al., 1991). Glucagon does not stimulate lipolysis in adipose tissue from eithersheep or cattle (Etherton et al., 1977). Somatotropin increases responsiveness of adiposetissue to catecholamine-stimulated lipolysis (for review see Etherton and Bauman, 1998), butsomatotropin, insulin-like growth factors I and II, prolactin, and placental lactogen are with-out effect on lipolysis in ruminants (Houseknecht et al., 1996). Prostaglandins (specificallyPGE2) also are not involved with regulation of lipolysis in bovine adipose tissue, as demon-strated by the lack of effect of both exogenous PGE2 and indomethacin, which blocksprostaglandin synthetase, on basal and epinephrine-stimulated lipolysis (DiMarco et al.,1991). Although insulin is known to suppress lipolysis, direct effects of insulin to inhibit basalor stimulated lipolysis have been difficult to demonstrate in vitro (DiMarco et al., 1991).

Basal and stimulated lipolytic rates of adipose tissues increase with growth and fattening(Smith et al., 1984; Rule et al., 1992) but are not affected by diet (high forage vs. high grain) atequal energy intake (Smith et al., 1984). Lipolytic rates generally are higher in subcutaneousadipose than in internal depots (Etherton et al., 1977; Rule et al., 1992), although differencesamong depots seem to diminish with increased age or body size (Rule et al., 1992). As animalsgrow and fatten, lipolysis becomes less complete, resulting in greater release of NEFA than ofglycerol (Smith et al., 1984); this may indicate greater re-esterification activity withinadipocytes that contributes to increased fat deposition. Hansen et al. (1995) found that glycerolrelease was greater from subcutaneous adipose than from perirenal adipose from sheep at market weight, but that fatty acid release and tissue fatty acid pool size were greater in

J. K. Drackley340

perirenal than in subcutaneous adipose. Consequently, lipolysis was less complete in the inter-nal depot than in the subcutaneous depot.

An exciting and expanding field concerns the impact of various factors produced byadipocytes that may act in autocrine, paracrine, or endocrine fashion to alter metabolism(Vernon et al., 1999; Chilliard et al., 2000). One of these, leptin, has received tremendousattention in both the biomedical arena (for review see Reidy and Weber, 2000) and in animalagriculture (for reviews see Hossner, 1998; Houseknecht et al., 1998; Chilliard et al., 2001;Ingvartsen and Boisclair, 2001). As adipocytes fill with lipid, leptin synthesis and secretionincreases. Leptin acts centrally to decrease feed intake and increase thermogenesis, and actsdirectly to increase lipolysis in adipose tissue and NEFA oxidation in peripheral tissues.Another autocrine or paracrine factor produced by adipocytes is TNFα, which acts to increaselipolysis and suppress LPL (Gasic et al., 1999). Expression of TNFα mRNA appears to changein bovine subcutaneous adipose tissue with degree of fatness (Drackley et al., unpublishedobservations) as in other species. The role of these adipose-derived regulatory factors incounteracting or decreasing the efficiency of fattening in ruminants will be an area of consider-able interest in the future.

6.3. Liver

Because the liver of ruminant animals is not a major site of de novo synthesis of fatty acids(Hanson and Ballard, 1967), its role lies in processing and redistributing exogenous andendogenous LCFA taken up from the diet and from plasma lipoproteins, as well as in metab-olism of NEFA mobilized from adipose tissue. In growing ruminants, the latter usually is aminor process. Glycolytic and lipogenic enzymes are present in the milk-fed calf, but specificactivities decrease sharply after weaning (Pearce and Unsworth, 1980). The rate of VLDLsynthesis and secretion in liver of ruminants is lower than in rats (Kleppe et al., 1988; Grauletet al., 1998), which is associated with the low rate of hepatic de novo fatty acid synthesis asobserved in other species (Pullen et al., 1990).

In fasted preruminant calves, considerable uptake of VLDL by the liver has been noted(Bauchart et al., 1989). Furthermore, heavy, lipid-poor HDL were also taken up by the liverof fasting calves, with concomitant production and release of lighter, lipid-enriched (prima-rily with cholesterol esters) HDL particles (Bauchart et al., 1989); no appreciable flux of LDLwas noted. Uptake of VLDL by the liver changed to secretion as dietary tallow intakeincreased in calves (Auboiron et al., 1995). Addition of supplemental methionine, whichmight be postulated to be limiting for hepatic synthesis of apo-B, slightly increased secretionof VLDL (Auboiron et al., 1995). Secretion of VLDL by calf liver slices occurred at 6- to 18-foldlower rates than in rat liver slices (Graulet et al., 1998), despite similar rates of synthesisof apoB in liver slices (Gruffat-Mouty et al., 1999). This lower rate of VLDL synthesis orsecretion may be attributable to a low rate of synthesis of TG in the microsomal compartmentresponsible for VLDL assembly, with no limitation in the ability to synthesize and storecytosolic TG (Graulet et al., 1998).

Hepatic lipid metabolism in preruminants may be sensitive to the LCFA profile of the diet,as replacement of tallow with hydrogenated coconut oil (rich in medium-chain saturatedfatty acids and deficient in unsaturated LCFA) or soybean oil (high in polyunsaturatedLCFA) caused accumulation of TG in the liver of calves (Jenkins and Kramer, 1986;Leplaix-Charlat et al., 1996; Piot et al., 1999). This suggests that changes in the profile ofdietary LCFA that are markedly different from that of bovine milk fat cause alterations inhepatic lipid metabolism.


Accumulation of TG in liver of preruminant calves fed milk replacer containing soybeanoil appeared to result from secretion of VLDL that were enriched with cholesterol estersrather than TG, which consequently decreased hepatic secretion of TG (Leplaix-Charlat et al.,1996). Because of the low oxidation of linoleic acid (Lindsay and Leat, 1977; Reid andHusbands, 1985), high dietary linoleate supply could have decreased hepatic LCFA oxidationand so contributed to TG accumulation. Liver uptake of very large and light HDL enrichedwith cholesterol esters was observed, and appeared to be increased by including cholesterolin the diet with soybean oil. This corresponded with simultaneous secretion of heavy HDL.

Liver slices from calves fed coconut oil-enriched milk replacers had lower rates of LCFAoxidation and greater rates of LCFA esterification than did liver slices from calves fed anequal amount of tallow (Graulet et al., 2000). Liver from coconut oil-fed calves also haddecreased rates of apo-B synthesis (Gruffat-Mouty et al., 2001); consequently, greater ratesof TG synthesis coupled with lower rates of VLDL export likely explain the accumulation ofTG in liver of calves fed coconut oil. Since milk fat contains more short- and medium-chainfatty acids, which are extensively metabolized by the liver, than does tallow, and less PUFAthan does soybean oil, it would be of interest to compare the effects caused by coconut oil andsoybean oil with milk-fat controls rather than tallow controls to discern more about the nativemetabolism in liver of preruminant calves.

Regulation of hepatic LCFA metabolism in ruminants has been reviewed extensively (Bell,1980; Bauchart, 1993; Grummer, 1993; Bauchart et al., 1996; Hocquette and Bauchart, 1999;Drackley et al., 2001). Liver removal of NEFA increases as arterial concentration of NEFAincreases due to changing intake in growing beef steers (Lapierre et al., 2000). Increaseduptake of NEFA mobilized from adipose tissue may lead to increases in TG formation andketone body synthesis, although these changes are much smaller in magnitude than thoseobserved during the negative energy balance of early lactation in dairy cows. For example, a9-day starvation period in growing steers only increased liver TG concentration from 0.47% to1.38% of wet weight, whereas plasma NEFA concentration increased to 1.03 mM (Lyle et al.,1984). Ketone body concentrations in that study were only modestly elevated (BHBA =6.9 mg/dl). Consequently, consideration of hepatic TG accumulation or excessive ketonebody production is not of major concern in growing ruminants.

Oxidation of LCFA in liver of preruminants and ruminants occurs in mitochondria andperoxisomes. Peroxisomes are subcellular organelles that possess a pathway for β-oxidationof fatty acids in which the first oxidation step is not coupled to ATP production but whichresults in the dissipation of heat (see Reddy and Hashimoto, 2001, for review). Peroxisomalβ-oxidation is believed to function to oxidize fatty acids that are poor substrates for mito-chondrial oxidative enzymes and to help process fatty acids when cellular uptake greatlyexceeds the cells’ need for energy. Peroxisomal β-oxidation in liver of ruminants was firststudied by Grum et al. (1994), who found that hepatic tissue capacity for peroxisomalβ-oxidation constituted about 50% of the total β-oxidation capacity in liver from mature dairycows, which was much greater than capacity measured in retired female breeder rats (26%).Subsequent studies showed that peroxisomal β-oxidation constituted from 44% to 48% oftotal β-oxidation in liver from sheep at market weight (Hansen et al., 1995). Absolute rates ofperoxisomal β-oxidation and the contribution to total β-oxidation capacity in isolated livertissue were lower in preruminant calves than in 15-month-old bulls (Piot et al., 1998). Thepotential role and importance of hepatic peroxisomal β-oxidation during growth is uncertain.

Surprisingly little is known about the development of hepatic tissue capacity for metabo-lism of LCFA during the transition from preruminant through weaning to functioningruminant. Ontogenic studies of key regulatory enzymes and pathways for fatty acid oxidation

J. K. Drackley342

and esterification, as well as for VLDL assembly and secretion, would seem to be of interestand importance to a full understanding of the metabolism of ruminant growth.


Regulation of interorgan lipid metabolism during growth will continue to be an importanttopic for biological scientists studying ruminant animals. Although much has been learnedabout metabolism of LCFA and other lipids during the last decade, there are numerous areasthat lack fundamental understanding. Notable among these is the inability to alter rates ofdevelopment and filling of intramuscular adipocytes, which is critically important to eatingqualities of meat. An understanding of genetic and physiological mechanisms that controladipose development among different depots would be of enormous benefit in developmentof strategies to minimize lipid accumulation in internal and subcutaneous depots while max-imizing marbling. The role of exogenous (dietary) LCFA, delivered by intestinally derivedchylomicrons and VLDL, as substrates for TG synthesis in intramuscular adipose also remainsof interest given the apparently low rate of de novo lipogenesis in this tissue. Another areawhere information is unexpectedly incomplete is the developmental changes in lipid meta-bolism in key organs, especially the liver, as ruminants make the transition from preruminantthrough weaning to functioning ruminant.

Additional progress will certainly be made in the near future as techniques of molecular biol-ogy are increasingly applied to questions concerning ruminant growth. The increasing ease ofmeasuring specific mRNA concentrations, as well as transcription rates, will likely lead to amore complete understanding of how lipid metabolism is regulated through transcriptional andpost-transcriptional means. Furthermore, high-throughput techniques such as microarray analy-sis of global changes in gene expression can be applied strategically, both to test hypotheses andto search for novel genes that are regulated differently between physiological states, geneticbackgrounds, disease states, or applications of nutritional or pharmacological treatments.

Finally, impacts of the external animal environment on lipid metabolism during growth is anarea that should see additional exploration. Increasing knowledge of the multiway communi-cation among the neuroendocrine system, the central nervous system, the immune system, andintermediary metabolism of numerous organs (see e.g. Johnson, 1997; Kelley, 2001; andChapter 4 by Johnson and Escobar, this volume) offers exciting possibilities to improve bothefficiency of ruminant animal production and animal well-being. Such efforts are critical to thelong-term sustainability of ruminant animal production.

REFERENCES

Andersen, M.K., Bailey, J.W., Wilken, C., Rule, D.C., 1996. Lipoprotein lipase and glycerophosphateacyltransferase in ovine tissues are influenced by growth and energy intake regimen. J. Nutr. Biochem.7, 610–616.

Armand, M., Borel, P., Dubois, C., Senft, M., Peyrot, J., Salducci, J., Lafont, H., Lairon, D., 1994.Characterization of emulsions and lipolysis of dietary lipids in the human stomach. Amer. J. Physiol.266, G372–G381.

Ashes, J.R., Fleck, E., Scott, T.W., 1995. Dietary manipulation of membrane lipids and its implicationsfor their role in the production of second messengers. In: Von Englehardt, W., Leonhard-Marek, S.,Breves, G., Giesecke, D. (Eds.), Ruminant Physiology: Digestion, Metabolism, Growth andReproduction. Ferdinand Enke Verlag, Stuttgart, Germany, pp. 373–386.

Auboiron, S., Durand, D., Robert, J.C., Chapman, M.J., Bauchart, D., 1995. Effects of dietary fat and L-methionine on the hepatic metabolism of very low density lipoproteins in the preruminant calf, Bosspp. Reprod. Nutr. Dev. 35, 167–178.


Barber, M.C., Ward, R.J., Richards, S.E., Salter, A.M., Buttery, P.J., Vernon, R.G., Travers, M.T., 2000.Ovine adipose tissue monounsaturated fat content is correlated to depot-specific expression of thestearoyl-CoA desaturase gene. J. Anim. Sci. 78, 62–68.

Bartelds, B., Gratama, J.W., Knoester, H., Takens, J., Smid, G.B., Aarnoudse, J.G., Heymans, H.S.,Kuipers, J.R., 1998. Perinatal changes in myocardial supply and flux of fatty acids, carbohydrates,and ketone bodies in lambs. Amer. J. Physiol. 274, H1962–H1969.

Bauchart, D., 1993. Lipid absorption and transport in ruminants. J. Dairy Sci. 76, 3864–3881.Bauchart, D., Durand, D., Laplaud, P.M., Forgez, P., Goulinet, S., Chapman, M.J., 1989. Plasma lipopro-

teins and apolipoproteins in the preruminant calf, Bos spp.: density distribution, physicochemicalproperties, and the in vivo evaluation of the contribution of the liver to lipoprotein homeostasis. J. Lipid Res. 30, 1499–1514.

Bauchart, D., Gruffat, D., Durand, D., 1996. Lipid absorption and hepatic metabolism in ruminants. Proc.Nutr. Soc. 55, 39–47.

Bell, A.W., 1980. Lipid metabolism in liver and selected tissues and in the whole body of ruminant animals. Prog. Lipid Res. 18, 117–164.

Bell, A.W., Thompson, G.E., 1979. Free fatty acid oxidation in bovine muscle in vivo: effects of coldexposure and feeding. Amer. J. Physiol. 237, E309–E315.

Bennis, A., de La Farge, F., Bézille, P., Valdiguié, P., Rico, A.G., Braun, J.P., 1992. Effects of age ofnewborn and delivery by female goats on plasma lipids and lipoproteins. Small Ruminant Res. 9,243–253.

Benson, J.A., Reynolds, C.K., 2001. Effects of abomasal infusion of long-chain fatty acids on splanch-nic metabolism of pancreatic and gut hormones in lactating dairy cows. J. Dairy Sci. 84, 1488–1500.

Bonnet, M., Faulconnier, Y., Fléchet, J., Hocquette, J.F., Leroux, C., Langin, D., Martin, P., Chilliard, Y.,1998. Messenger RNAs encoding lipoprotein lipase, fatty acid synthase and hormone-sensitive lipasein the adipose tissue of underfed-refed ewes and cows. Reprod. Nutr. Dev. 38, 297–307.

Bonnet, M., Leroux, C., Faulconnier, Y., Hocquette, J.F., Bocquier, F., Martin, P., Chilliard, Y., 2000.Lipoprotein lipase activity and mRNA are upregulated by refeeding in adipose tissue and cardiacmuscle of sheep. J. Nutr. 130, 749–756.

Borel, P., Armand, M., Ythier, P., Dutot, G., Melin, C., Senft, M., Lafont, H., Lairon, D., 1994. Hydrolysisof emulsions with different triglycerides and droplet sizes by gastric lipases in vitro: effect on pancreatic lipase activity. J. Nutr. Biochem. 5, 124–133.

Bouyekhf, M., Rule, D.C., Hu, C.Y., 1992. Glycerolipid biosynthesis in adipose tissue of the bovineduring growth. Comp. Biochem. Physiol. B 103, 101–104.

Bouyekhf, M., Rule, D.C., Hu, C.Y., 1993. Effect of catecholamines on lipolysis and esterification invitro in adipose tissue of sheep fed low and high energy diets. J. Nutr. Biochem. 4, 80–85.

Brandstetter, A., Picard, B., Geay, Y., 1997. Regional variations of muscle fiber characteristics inm. semitendinosus of growing cattle. J. Muscle Res. Cell Motil. 18, 57–62.

Bremmer, D.R., Ruppert, L.D., Clark, J.H., Drackley, J.K., 1998. Effects of chain length and unsatura-tion of fatty acid mixtures infused into the abomasum of lactating dairy cows. J. Dairy Sci. 81,176–188.

Campbell, R.G., 1988. Nutritional constraints to lean tissue accretion in farm animals. Nutr. Res. Rev.1, 233–253.

Chen, M., Yang, Y., Braunstein, E., Georgeson, K.E., Harmon, C.M., 2001. Gut expression and regula-tion of FAT/CD36: possible role in fatty acid transport in rat enterocytes. Amer. J. Physiol. 281,E916–E923.

Cheng, A.L.S., Morehouse, M.G., Duell, J.J., 1949. The effect of the level of dietary calcium and mag-nesium on the digestibility of fatty acids, simple triglycerides and some natural and hydrogenated fats.J. Nutr. 37, 237–250.

Chien, D., Dean, D., Saha, A.K., Flatt, J.P., Ruderman, N.B., 2000. Malonyl-CoA content and fatty acidoxidation in rat muscle and liver in vivo. Amer. J. Physiol. 279, E259–E265.

Chilliard, Y., 1993. Dietary fat and adipose tissue metabolism in ruminants, pigs, and rodents: a review.J. Dairy Sci. 76, 3897–3931.

Chilliard, Y., Bonnet, M., Delavaud, C., Faulconnier, Y., Leroux, C., Djiane, J., Bocquier, F., 2001. Leptinin ruminants: gene expression in adipose tissue and mammary gland, and regulation of plasmconcentration. Domest. Anim. Endocrinol. 21, 271–295.

Chilliard, Y., Ferlay, A., Faulconnier, Y., Bonnet, M., Rouel, J., Bocquier, F., 2000. Adipose tissue metab-olism and its role in adaptations to undernutrition in ruminants. Proc. Nutr. Soc. 59, 127–134.

J. K. Drackley344

Christensen, R.A., Drackley, J.K., LaCount, D.W., Clark, J.H., 1994. Infusion of four long-chain fattyacid mixtures into the abomasum of lactating dairy cows. J. Dairy Sci. 77, 1052–1069.

Cianflone, K., Maslowska, M., Sniderman, A.D., 1999. Acylation stimulating protein (ASP), anadipocyte autocrine: new directions. Sem. Cell Dev. Biol. 10, 31−41.

Clarke, S.D., 2001. Polyunsaturated fatty acid regulation of gene transcription: a molecular mechanismto improve the metabolic syndrome. J. Nutr. 131, 1129–1132.

Clifford, G.M., Londos, C., Kraemer, F.B., Vernon, R.G., Yeaman, S.J., 2000. Translocation of hormone-sensitive lipase and perilipin upon lipolytic stimulation of rat adipocytes. J. Biol. Chem. 275,5011–5015.

Cordle, S.R., Kroper, W.J., Yeaman, S.J., 1986. Isolation and characterization of bovine lipoproteins:incubation with hepatic lipase and isolated adrenal zona fasiculata cells. Biochem. Soc. Trans. 13,879–882.

Cortright, R.N., Muoio, D.M., Dohm, G.L., 1997. Skeletal muscle lipid metabolism: a frontier for newinsights into fuel homeostasis. J. Nutr. Biochem. 8, 228–245.

Davidson, N.O., Shelness, G.S., 2000. Apolipoprotein B: mRNA editing, lipoprotein assembly, and presecretory degradation. Annu. Rev. Nutr. 20, 169–193.

Davis, C.L., Drackley, J.K., 1998. The Development, Nutrition, and Management of the Young Calf. IowaState University Press, Ames, IA.

DiMarco, N.M., Rule, D.C., Whitehurst, G.B., Beitz, D.C., 1991. Effect of indomethacin, epinephrine,prostaglandin E2 and insulin on lipolysis in bovine adipose tissue in vitro. Int. J. Biochem. 23, 1231−1235.

Doppenberg, J., Palmquist, D.L., 1991. Effect of dietary fat level on feed intake, growth, plasma metabo-lites and hormones of calves fed dry or liquid diets. Livest. Prod. Sci. 29, 151–166.

Doreau, M., Chilliard, Y., 1997. Digestion and metabolism of dietary fat in farm animals. Brit. J. Nutr.78, Suppl. 1, S15–S35.

Drackley, J.K., 2000. Lipid metabolism. In: D’Mello, J.P.F. (Ed.), Farm Animal Metabolism andNutrition: Critical Reviews. CAB International, Wallingford, UK, pp. 97–119.

Drackley, J.K., Klusmeyer, T.H., Trusk, A.M., Clark, J.H., 1992. Infusion of long-chain fatty acids varyingin saturation and chain length into the abomasum of lactating dairy cows. J. Dairy Sci. 75, 1517–1526.

Drackley, J.K., Overton, T.R., Douglas, G.N., 2001. Adaptations of glucose and long-chain fatty acid metab-olism in liver of dairy cows during the periparturient period. J. Dairy Sci. 84, Suppl. E, E100–E112.

Drackley, J.K., Thire, S., Litherland, N.B., Beaulieu, A.D., 2000. Dry matter intake is decreased more byabomasal infusion of unsaturated free fatty acids than by unsaturated triglycerides. J. Dairy Sci. 83,Suppl. 1, 286 (Abstr.).

Edwards-Webb, J.D., Thompson, S.Y., 1978. Studies on lipid digestion in the preruminant calf. 3. Theaction of salivary lipase on milk fat in the abomasum. Brit. J. Nutr. 40, 125–131.

Enser, M., 1973. Clearing-factor lipase in muscle and adipose tissue of pigs. Biochem. J. 136, 381–385.Etherton, T.D., Bauman, D.E., 1998. Biology of somatotropin in growth and lactation of domestic animals.

Phyiol. Rev. 78, 745–761.Etherton, T.D., Bauman, D.E., Romans, J.R., 1977. Lipolysis in subcutaneous and perirenal adipose

tissue from sheep and dairy steers. J. Anim. Sci. 44, 1100–1106.Fallon, R.J., Williams, P.E.V., Innes, G.M., 1986. The effects on feed intake, growth and digestibility of

nutrients of including calcium soaps of fat in diets for young calves. Anim. Feed Sci. Tech. 14,103–115.

Faulconnier, Y., Bonnet, M., Bocquier, F., Leroux, C., Chilliard, Y., 2001. Effects of photoperiod andfeeding level on adipose tissue and muscle lipoprotein lipase activity and mRNA level in dry non-pregnant sheep. Brit. J. Nutr. 85, 299–306.

Fielding, C.J., Fielding, P.E., 1995. Molecular physiology of reverse cholesterol transport. J. Lipid Res.36, 211–228.

Fielding, C.J., Fielding, P.E., 2001. Cellular cholesterol efflux. Biochim. Biophys. Acta 1533, 175–189.Forte, T.M., Bell-Quint, J.J., Cheng, F., 1981. Lipoproteins of fetal and newborn calves and adult steers:

a study of developmental changes. Lipids 16, 240–245.Freeman, C.P., 1984. The digestion, absorption and transport of fats–non-ruminants. In: Wiseman, J. (Ed.),

Fats in Animal Nutrition. Butterworths, London, pp. 105–122.Gagliostro, G., Chilliard, Y., 1991. Duodenal rapeseed oil infusion in early and midlactation cows. 2.

Voluntary intake, milk production, and composition. J. Dairy Sci. 74, 499–509.Gargouri, Y., Moreau, H., Verger, R., 1989. Gastric lipases: biochemical and physiological studies.

Biochim. Biophys. Acta 1006, 255–271.


Gasic, S., Tian, B., Green, A., 1999. Tumor necrosis factor α stimulates lipolysis in adipocytes by decreas-ing Gi protein concentrations. J. Biol. Chem. 274, 6770–6775.

Glatz, J.F., van der Vusse, G.J., 1996. Cellular fatty acid-binding proteins: their function and physiologicalsignificance. Prog. Lipid Res. 35, 243–282.

Gooden, J.M., 1973. The importance of lipolytic enzymes in milk-fed and ruminating calves. Austr. J. Biol.Sci. 26, 1189–1199.

Graham, D.L., Oram, J.F., 1987. Identification and characterization of a high density lipoprotein-bindingprotein in cell membranes by ligand blotting. J. Biol. Chem. 262, 7439–7442.

Graulet, B., Gruffat, D., Durand, D., Bauchart, D., 1998. Fatty acid metabolism and very low densitylipoprotein secretion in liver slices from rats and preruminant calves. J. Biochem. 124, 1212–1219.

Graulet, B., Gruffat-Mouty, D., Durand, D., Bauchart, D., 2000. Effects of milk diets containing beeftallow or coconut oil on the fatty acid metabolism of liver slices from preruminant calves. Brit. J.Nutr. 84, 309–318.

Greeve, J., Altkemper, I., Dieterich, J.H., Greten, H., Windler, E., 1993. Apolipoprotein mRNA editing in 12 different mammalian species: hepatic expression is reflected in low concentrations of apo B-containing plasma lipoproteins. J. Lipid Res. 34, 1367–1383.

Gruffat-Mouty, D., Graulet, B., Durand, D., Samson-Bouma, M.E., Bauchart, D., 1999. ApolipoproteinB production and very low density lipoprotein secretion by calf liver slices. J. Biochem. 126,188–193.

Gruffat-Mouty, D., Graulet, B., Durand, D., Samson-Bouma, M.E., Bauchart, D., 2001. Effects of dietarycoconut oil on apolipoprotein synthesis and VLDL secretion by calf liver slices. Brit. J. Nutr. 86,13–19.

Grum, D.E., Hansen, L.R., Drackley, J.K., 1994. Peroxisomal β-oxidation of fatty acids in bovine and ratliver. Comp. Biochem. Physiol. B 109, 281–292.

Grummer, R.R., 1993. Etiology of lipid-related metabolic disorders in periparturient dairy cows. J. DairySci. 76, 3882–3896.

Ha, Y.C., Barter, P.J., 1982. Differences in plasma cholesteryl ester transfer activity in sixteen vertebratespecies. Comp. Biochem. Physiol. B 71, 265–269.

Hansen, L.R., Drackley, J.K., Berger, L.L., Grum, D.E., Cremin, J.D. Jr., Lin, X., Odle, J., 1995. Prenatalandrogenization of lambs: II. Metabolism in adipose tissue and liver. J. Anim. Sci. 73, 1701–1712.

Hanson, R.W., Ballard, F.J., 1967. The relative significance of acetate and glucose as precursors for lipidsynthesis in liver and adipose tissue from ruminants. Biochem. J. 105, 529–536.

Harrison, F.A., Leat, W.M.F., 1975. Digestion and absorption of lipids in non-ruminant and ruminant animals: a comparison. Proc. Nutr. Soc. 34, 203–210.

Harrison, F.A., Leat, W.M.F., Foster, A., 1974. Absorption of maize oil infused into the duodenum of thesheep. Proc. Nutr. Soc. 33, 103 (Abstr.).

Hocquette, J.F., Bauchart, D., 1999. Intestinal absorption, blood transport and hepatic and muscle metab-olism of fatty acids in preruminant and ruminant animals. Reprod. Nutr. Dev. 39, 27–48.

Hocquette, J.F., Graulet, B., Olivecrona, T., 1998a. Lipoprotein lipase activity and mRNA levels inbovine tissues. Comp. Biochem. Physiol. B121, 201–212.

Hocquette, J.F., Graulet, B., Vermorel, M., Bauchart, D., 2001. Weaning affects lipoprotein lipase activity and gene expression in adipose tissues and in masseter but not in other muscles of the calf.Brit. J. Nutr. 86, 433–441.

Hocquette, J.F., Ortigues-Marty, I., Pethick, D., Herpin, P., Fernandez, X., 1998b. Nutritional and hor-monal regulation of energy metabolism in skeletal muscles of meat-producing animals. Livest. Prod.Sci. 56, 115–143.

Hood, R.L., Allen, C.E., 1978. Lipogenesis in isolated intramuscular adipose tissue from four bovinemuscles. J. Anim. Sci. 46, 1626–1633.

Hood, R.L., Cook, L.J., Mills, S.C., Scott, T.W., 1980. Effect of feeding protected lipids on fatty acid synthesis in ovine tissues. Lipids 15, 644–650.

Hossner, K.L., 1998. Cellular, molecular and physiological aspects of leptin: potential application inanimal production. Can. J. Anim. Sci. 78, 463–472.

Houseknecht, K.L., Baile, C.A., Matteri, R.L., Spurlock, M.E., 1998. The biology of leptin: a review. J. Anim. Sci. 76, 1405–1420.

Houseknecht, K.L., Bauman, D.E., Vernon, R.G., Byatt, J.C., Collier, R.J., 1996. Insulin-like growth factors-I and -II, somatotropin, prolactin, and placental lactogen are not acute effectors of lipolysis in ruminants. Domest. Anim. Endocrinol. 13, 239–249.

J. K. Drackley346

Ingle, D.L., Bauman, D.E., Garrigus, U.S., 1972. Lipogenesis in the ruminant: in vitro study of tissuesites, carbon source and reducing equivalent generation for fatty acid synthesis. J. Nutr. 102, 609–616.

Ingvartsen, K.L., Boisclair, Y.R., 2001. Leptin and the regulation of food intake, energy homeostasis andimmunity with special focus on periparturient ruminants. Domest. Anim. Endocrinol. 21, 215–250.

Jacobi, S.K., Miner, J.L., 2002. Human acylation-stimulating protein and lipid biosynthesis in bovineadipose tissue explants. J. Anim. Sci. 80, 751–756.

Jenkins, K.J., 1986. Fatty acid-binding protein in liver and small intestine of the preruminant calf. J. Dairy Sci. 69, 155–159.

Jenkins, K.J., 1988. Factors affecting poor performance and scours in preruminant calves fed corn oil. J. Dairy Sci. 71, 3013–3020.

Jenkins, K.J., Kramer, J.K.G., 1986. Influence of low linoleic and linolenic acids in milk replacer on calfperformance and lipids in blood plasma, heart, and liver. J. Dairy Sci. 69, 1374–1386.

Jenkins, K.J., Kramer, J.K.G., 1990. Effects of dietary corn oil and fish oil concentrate on lipid compo-sition of calf tissues. J. Dairy Sci. 73, 2940−2951.

Jenkins, K.J., Griffith, G., Kramer, J.K.G., 1988. Plasma lipoproteins in neonatal, preruminant, andweaned calf. J. Dairy Sci. 71, 3003–3012.

Jenkins, K.J., Kramer, J.K.G., Emmons, D.B., 1986. Effect of lipids in milk replacers on calf perform-ance and lipids in blood plasma, liver, and perirenal fat. J. Dairy Sci. 69, 447–459.

Jenkins, K.J., Kramer, J.K.G., Sauer, F.D., Emmons, D.B., 1985. Influence of triglycerides and free fattyacids in milk replacers on calf performance, blood plasma, and adipose lipids. J. Dairy Sci. 68,669–680.

Jenkins, T.C., 1993. Lipid metabolism in the rumen. J. Dairy Sci. 76, 3851–3863.Jenkins, T.C., Thies, E.J., Fotouhi, N., 1994. Dietary soybean oil changes lipolytic rate and composition

of fatty acids in plasma membranes of ovine adipocytes. J. Nutr. 124, 566–570.Jenness, R., 1985. Biochemical and nutritional aspects of milk and colostrum. In: Larson, B.L. (Ed.),

Lactation. Iowa State University Press, Ames, IA, pp. 164–197.Johnson, R.W., 1997. Inhibition of growth by pro-inflammatory cytokines: an integrated view. J. Anim.

Sci. 75, 1244–1255.Kelley, K.W., 2001. It’s time for psychoneuroimmunology. Brain Behav. Immun. 15, 1–6.Kleppe, B.B., Aiello, R.J., Grummer, R.R., Armentano, L.E., 1988. Trigyceride accumulation and very

low density lipoprotein secretion by rat and goat hepatocytes in vitro. J. Dairy Sci. 71, 1813–1822.Kris-Etherton, P.M., Etherton, T.D., 1982. The role of lipoproteins in lipid metabolism of meat animals.

J. Anim. Sci. 55, 804–817.Kuehn, C.S., Otterby, D.E., Linn, J.G., Olson, W.G., Chester-Jones, H., Marx, G.D., Barmore, J.A., 1994.

The effect of dietary energy concentration on calf performance. J. Dairy Sci. 77, 2621–2629.Langfort, J., Ploug, T., Ihlemann, J., Enevoldsen, L.H., Stallknecht, B., Saldo, M., Kjaer, M., Hom, C.,

Galbo, H., 1998. Hormone-sensitive lipase (HSL) expression and regulation in skeletal muscle. Adv.Exp. Med. Biol. 441, 219–228.

Lapierre, H., Bernier, J.F., Dubreuil, P., Reynolds, C.K., Farmer, C., Ouellet, D.R., Lobley, G.E., 2000.The effect of feed intake level on splanchnic metabolism in growing beef steers. J. Anim. Sci. 78,1084–1099.

Laplaud, P.M., Bauchart, D., Durand, D., Beaubatie, L., Chapman, M.J., 1991. Intestinal lymph and plasmalipoproteins in the preruminant calf: partial resolution of particle heterogeneity in the 1.040−1.090 g/mLinterval. J. Lipid Res. 32, 1429–1439.

Laplaud, P.M., Bauchart, D., Durand, D., Chapman, M.J., 1990. Lipoproteins and apolipoproteins inintestinal lymph of the preruminant calf, Bos spp., at peak lipid absorption. J. Lipid Res. 31,1781–1792.

Leplaix-Charlat, L., Durand, D., Bauchart, D., 1996. Effects of diets containing tallow and soybean oilwith and without cholesterol on hepatic metabolism of lipids and lipoproteins in the preruminant calf.J. Dairy Sci. 79, 1826–1835.

Lindsay, D.B., Leat, W.M.F., 1977. Oxidation and metabolism of linoleic acid in fed and fasted sheep. J. Agr. Sci. 89, 215–221.

Lyle, R.R., DeBoer, G., Harrison, R.O., Young, J.W., 1984. Plasma and liver metabolites and glucosekinetics as affected by prolonged ketonemia-glucosuria and fasting in steers. J. Dairy Sci. 67,2274–2282.

Marcos, E., Mazur, A., Cardot, P., Coxam, V., Rayssiguier, Y., 1991. Developmental changes in plasmaapolipoproteins B and A-I in fetal bovines. Biol. Neonate 59, 22–29.


McNamara, J.P., 1994. Lipid metabolism in adipose tissue during lactation: a model of a metabolic con-trol system. J. Nutr. 124, 1383S–1391S.

Mendel, C.M., Kunitake, S.T., Kane, J.P., 1986. Discrimination between subclasses of human high-density lipoproteins by the HDL binding sites of bovine liver. Biochim. Biophys. Acta 875, 59–68.

Miller, M.F., Cross, H.R., Lunt, D.K., Smith, S.B., 1991. Lipogenesis in acute and 48-hour cultures ofbovine intramuscular and subcutaneous adipose tissue explants. J. Anim. Sci. 69, 162–170.

Mills, S.E., Lemenager, R.P., Horstman, L.A., 1989. Adipose tissue lipogenesis in growing steers adaptedto different levels of feed intake. J. Anim. Sci. 67, 3011–3017.

Moibi, J.A., Christopherson, R.J., Okine, E.K., 2000a. In vivo and in vitro lipogenesis and aspects of metabolism in ovines: effect of environmental temperature and dietary lipid supplementation. Can.J. Anim. Sci. 80, 59–67.

Moibi, J.A., Christopherson, R.J., Okine, E.K., 2000b. Effect of environmental temperature and dietarylipid supplement on activity and protein abundance of acetyl-CoA carboxylase and fatty acid synthasein skeletal muscle and adipose tissues of sheep. Can. J. Anim. Sci. 80, 69–77.

Moore, J.H., Christie, W.W., 1984. Digestion, absorption and transport of fats in ruminant animals. In: Wiseman, J. (Ed.), Fats in Animal Nutrition. Butterworths, London, pp. 123–149.

Moore, K.K., Cameron, P.J., Ekeren, P.A., Smith, S.B., 1993. Fatty acid binding protein in bovine longissimusdorsi muscle. Comp. Biochem. Physiol. B 104, 259–266.

Nestel, P.J., Poyser, A., Hood, R.L., Mills, S.C., Willis, M.R., Cook, L.J., Scott, T.W., 1978. The effect ofdietary fat supplements on cholesterol metabolism in ruminants. J. Lipid Res. 19, 899–909.

Noble, R.C., 1978. Digestion, absorption and transport of lipids in ruminant animals. Prog. Lipid Res.17, 55–91.

Noble, R.C., 1984. Essential fatty acids in the ruminant. In: Wiseman, J. (Ed.), Fats in Animal Nutrition.Butterworths, London, pp. 185–200.

Noble, R.C., Shand, J.H., 1982. Fatty acid metabolism in the neonatal ruminant. In: Draper, H.H. (Ed.),Advances in Nutritional Research, Vol. 4. Plenum Press, New York, pp. 287–337.

Noble, R.C., O’Kelly, J.C., Moore, J.H., 1972. Observations on the lecithin:cholesterol acyl transferasesystem in bovine plasma. Biochim. Biophys. Acta 270, 519–528.

Noble, R.C., Shand, J.H., 1983. A comparative study of the distribution and fatty acid composition of thelipoproteins in the fetal and maternal plasma of sheep. Biol. Neonate 44, 10–20.

Olivecrona, T., Olivecrona, G., 1999. Lipoprotein and hepatic lipases in lipoprotein metabolism. In: Betteridge, D.J., Illingworth, D.R., Shepherd, J. (Eds.), Lipoproteins in Health and Disease.Edward Arnold, London, pp. 223–246.

Ottemann-Abbamonte, C.J., Drackley, J.K., Beaulieu, A.D., Overton, T.R., Douglas, G.N., Palmquist, D.L.,1998. Effect of acetate on palmitate oxidation in bovine and rat skeletal muscle tissue. FASEB J.12, A835 (Abstr.).

Overton, T.R., Beaulieu, A.D., Zhu, J., Ortiz-Gonzalez, G., Drackley, J.K., Barbano, D.M., 1998. Doseresponse to increasing amounts of high-oleic sunflower fatty acids infused into the abomasum of lactating dairy cows. J. Dairy Sci. 81, Suppl. 1, 352 (Abstr.).

Page, A.M., Sturdivant, C.A., Lunt, D.K., Smith, S.B., 1997. Dietary whole cottonseed depresses lipoge-nesis but has no effect on stearoyl-coenzyme desaturase activity in bovine subcutaneous adiposetissue. Comp. Biochem. Physiol. B 118, 79–84.

Palmquist, D.L., 1976. A kinetic concept of lipid transport in ruminants. J. Dairy Sci. 59, 355–363.Palmquist, D.L., 1991. Influence of source and amount of dietary fat on digestibility in lactating cows.

J. Dairy Sci. 74, 1354–1360.Pearce, J., Unsworth, E.F., 1980. The effects of diet on some hepatic enzyme activities in the pre-ruminant

and ruminating calf. J. Nutr. 110, 255–261.Perez, M.D., Sanchez, L., Aranda, P., Ena, J.M., Oria, R., Calvo, M., 1992. Effect of β-lactoglobulin on

the activity of pregastric lipase: a possible role for this protein in ruminant milk. Biochim. Biophys.Acta 1123, 151–155.

Pethick, D.W., Dunshea, F.R., 1993. Fat metabolism and turnover. In: Forbes, J.M., France, J. (Eds.),Quantitative Aspects of Ruminant Digestion and Metabolism. CAB International, Wallingford, UK,pp. 291−311.

Phan, C.T., Tso, P., 2001. Intestinal lipid absorption and transport. Frontiers Biosci. 6, 299–319.Piot, C., Hocquette, J.F., Veerkamp, J.H., Durand, D., Bauchart, D., 1999. Effects of dietary coconut oil

on fatty acid oxidative capacity of the liver, the heart and skeletal muscles in the preruminant calf.Brit. J. Nutr. 82, 299–308.

J. K. Drackley348

Piot, C., Hocquette, J.F., Herpin, P., Veerkamp, J.H., Bauchart, D., 2000. Dietary coconut oil affects morelipoprotein lipase activity than the mitochondria oxidative capacities in muscles of preruminantcalves. J. Nutr. Biochem. 11, 231–238.

Piot, C., Veerkamp, J.H., Bauchart, D., Hocquette, J.F., 1998. Contribution of mitochondria and peroxi-somes to palmitate oxidation in rat and bovine tissues. Comp. Biochem. Phyiol. B 121, 185–194.

Pothoven, M.A., Beitz, D.C., 1973. Effect of adipose tissue site, animal weight, and long-term fasting onlipogenesis in the bovine. J. Nutr. 103, 468–475.

Pothoven, M.A., Beitz, D.C., Thornton, J.H., 1975. Lipogenesis and lipolysis in adipose tissue of ad libitumand restricted-fed beef cattle during growth. J. Anim. Sci. 40, 957–962.

Pullen, D.L., Liesman, J.S., Emery, R.S., 1990. A species comparison of liver slice synthesis and secretionof triacylglycerol from non-esterified fatty acids in media. J. Anim. Sci. 68, 1395–1399.

Reddy, J.K., Hashimoto, T., 2001. Peroxisomal β-oxidation and peroxisome proliferator-activated recep-tor: an adaptive metabolic system. Annu. Rev. Nutr. 21, 193–230.

Reid, J.C.W., Husbands, D.R., 1985. Oxidative metabolism of long-chain fatty acids in mitochondriafrom sheep and rat liver: evidence that sheep conserve linoleate by limiting its oxidation. Biochem. J. 225, 233–237.

Reidy, S.P., Weber, J.M., 2000. Leptin: an essential regulator of lipid metabolism. Comp. Biochem.Phyisiol. A 125, 285–297.

Reinhart, G.A., Simmen, F.A., Mahan, D.C., White, M.E., Roehrig, K.L., 1992. Intestinal development andfatty acid binding protein activity of newborn pigs fed colostrum or milk. Biol. Neonate 62, 155–163.

Rudling, M.J., Peterson, C.O., 1985. LDL receptors in bovine tissues assayed as the heparin-sensitivebinding or 125I-labeled LDL in homogenates: relation between liver LDL receptors and serum cholesterol in the fetus and post-term. Biochim. Biophys. Acta 836, 96–104.

Rule, D.C., 1995. Adipose tissue glycerolipid biosynthesis. In: Smith, S.B., Smith, D.R. (Eds.), TheBiology of Fat in Meat Animals. American Society of Animal Science, Champaign, IL, pp. 129–143.

Rule, D.C., Thornton, J.H., McGilliard, A.D., Beitz, D.C., 1992. Effect of adipose tissue site, animal size,and fasting on lipolysis in bovine adipose tissue in vitro. Int. J. Biochem. 24, 789–793.

Senda, M., Oka, K., Brown, W.V., Qasba, P.K., Furuichi, Y., 1987. Molecular cloning and sequence of acDNA coding for bovine lipoprotein lipase. Proc. Natl. Acad. Sci. USA 84, 4369–4373.

Small, D.M., 1968. A classification of biological lipids based upon their interaction in aqueous systems.J. Amer. Oil Chem. Soc. 45, 108–119.

Smith, S.B., 1995. Substrate utilization in ruminant adipose tissues. In: Smith, S.B., Smith, D.R. (Eds.),The Biology of Fat in Meat Animals. American Society of Animal Science, Champaign, IL, pp. 166–188.

Smith, S.B., Crouse, J.D., 1984. Relative contributions of acetate, lactate and glucose to lipogenesis inbovine intramuscular and subcutaneous adipose tissue. J. Nutr. 114, 792–800.

Smith, S.B., Prior, R.L., 1986. Comparisons of lipogenesis and glucose metabolism between ovine andbovine adipose tissues. J. Nutr. 116, 1279–1286.

Smith, S.B., Jenkins, T., Prior, R.L., 1987. Carcass composition and adipose tissue metabolism in grow-ing sheep. J. Anim. Sci. 65, 1525–1530.

Smith, S.B., Lin, K.C., Wilson, J.J., Lunt, D.K., Cross, H.R., 1998. Starvation depresses acylglycerolbiosynthesis in bovine subcutaneous but not intramuscular adipose tissue homogenates. Comp.Biochem. Physiol. B 120, 165–174.

Smith, S.B., Prior, R.L., Ferrell, C.L., Mersmann, H.J., 1984. Interrelationships among diet, age, fat deposition and lipid metabolism in growing steers. J. Nutr. 114, 153–162.

Smith, S.B., Prior, R.L., Koong, L.J., Mersmann, H.J., 1992. Nitrogen and lipid metabolism in heifers fedat increasing levels of intake. J. Anim. Sci. 70, 152–160.

Spanski, N.A., Drackley, J.K., Davis, C.L., Jaster, E.H., 1997. Utilization of supplemental triglyceridesor free fatty acids by calves from 4 to 10 weeks of age. J. Dairy Sci. 80, 573–585.

Sugden, M.C., Holness, M.J., Howard, R.M., 1993. Changes in lipoprotein lipase activities in adiposetissue, heart and skeletal muscle during continuous or interrupted feeding. Biochem. J. 292, 113–119.

Tanaka, H., Mierau, I., Ito, F., 1999. Purification and characterization of bovine pancreatic bile salt-activated lipase. J. Biochem. 125, 883–890.

Tikofsky, J.N., Van Amburgh, M.E., Ross, D.A., 2001. Effect of varying carbohydrate and fat content ofmilk replacer on body composition of Holstein bull calves. J. Anim. Sci. 79, 2260–2267.

Toullec, R., Mathieu, C.M., 1969. Digestive utilization of fats and their principal fatty acids by calves:influence on body composition. Ann. Biol. Anim. Biochem. Biophys. 9, 139–160 (in French).


Tso, P., Balint, J.A., 1986. Formation and transport of chylomicrons by enterocytes to the lymphatics.Amer. J. Phyiol. 250, G715–G726.

Tso, P., Liu, M., Kalogeris, T.J., Thomson, A.B.R., 2001. The role of apolipoprotein A-IV in the regulationof food intake. Annu. Rev. Nutr. 21, 231–254.

Turley, S.D., Burns, D.K., Rosenfeld, C.R., Dietschy, J.M., 1996. Brain does not utilize low densitylipoprotein-cholesterol during fetal and neonatal development in the sheep. J. Lipid Res. 37,1953–1961.

van den Bosch H., Aarsman, A.J., de Jong, J.G., van Deenem, L.L., 1993. Studies on lysophospholipases. I.Purification and some properties of a lysophospholipase from beef pancreas. Biochim. Biophys. Acta296, 94–104.

Vernon, R.G., 1980. Lipid metabolism in the adipose tissue of ruminant animals. Prog. Lipid Res. 19,23–106.

Vernon, R.G., Barber, M.C., Travers, M.T., 1999. Present and future studies on lipogenesis in animals andhuman subjects. Proc. Nutr. Soc. 58, 541–549.

Villeneuve, P., Pina, M., Graille, J., 1996. Determination of pregastric lipase specificity in young rumi-nants. Chem. Phys. Lipids 83, 161–168.

Wang, C.S., Hartsuck, J.A., 1993. Bile salt-activated lipase: a multiple function lipolytic enzyme.Biochim. Biophys. Acta 1166, 1–19.

West, T.R., Riley, M.L., Rule, D.C., 1994. Palmitate esterification and glycerophosphate acyltransferaseactivity in adipose tissue of growing lambs. J. Anim. Sci. 72, 81–86.

Wilson, J.J., Young, C.R., Smith, S.B., 1992. Triacylglycerol biosynthesis in bovine liver and subcuta-neous adipose tissue. Comp. Biochem. Physiol. B 103, 511–516.

Winder, W.W., 2001. Energy-sensing and signaling by AMP-activated protein kinase in skeletal muscle.J. Appl. Physiol. 91, 1017–1028.

Wolfrum, C., Borrmann, C.M., Borchers, T., Spener, F., 2001. Fatty acids and hypolipidemic drugs regu-late peroxisome proliferator-activated receptors alpha- and gamma-mediated gene expression via liverfatty acid binding protein: a signaling path to the nucleus. Proc. Natl. Acad. Sci. USA 98, 2323–2328.

Young, M.E., Goodwin, G.W., Ying, J., Guthrie, P., Wilson, C.R., Laws, F.A., Taegtmeyer, H., 2001.Regulation of cardiac and skeletal muscle malonyl-CoA decarboxylase by fatty acids. Amer. J. Physiol.280, E471–E479.

Zhou, L., Nilsson, A., 2001. Sources of eicosanoid precursor fatty acid pools in tissues. J. Lipid Res. 42,1521–1542.

Zhou, L., Vessby, B., Nilsson, A., 2002. Quantitative role of plasma free fatty acids in the supply ofarachidonic acid to extrahepatic tissues in rats. J. Nutr. 132, 2626–2631.

Zinn, R.A., 1989. Influence of level and source of dietary fat on its comparative feeding value in finish-ing diets for feedlot steers: metabolism. J. Anim. Sci. 67, 1038–1049.

Zinn, R.A., 1992. Comparative feeding value of supplemental fat in steam-flaked corn- and steam-flakedwheat-based finishing diets for feedlot steers. J. Anim. Sci. 70, 2959–2969.

J. K. Drackley350

PART IVCarbohydrate andenergy metabolism

353

14 Environmental and hormonal regulation of energy metabolism in early development of the pig

P. Herpin, I. Louveau, M. Damon and J. Le Dividich

INRA, Joint Research Unit for Calf and Pig Production, 35590 Saint-Gilles, France

This chapter is concerned with the development and the environmental regulation of the energymetabolism in the neonatal pig, a species devoid of brown adipose tissue. The newborn pig ispoorly insulated and maintenance of its homeothermic balance is essentially related to its abilityto produce heat. The skeletal muscle is the major contributor to the energy metabolism, with shiv-ering being the main mechanism of cold-induced thermogenesis. The piglet does not expressUCP1. UCP2 and UCP3 are both expressed in skeletal muscle, but there is no evidence that theyplay a role in the cold-induced thermogenesis. Major factors involved in the postnatal improve-ment in the cold-induced thermogenesis include changes in muscle structure and cardiovascular,biochemical and hormonal adjustments. Muscle maturation is suggested by the marked post-natal increase in myofibrillar mass and the transitory expression of α-cardiac heavy chain myosinand triad proliferation. Moreover, muscle mitochondria are functional at birth, and both theincrease in mitochondrial mass and their ultrastructural change account for the increased oxida-tive potential of the muscle. Cardiovascular adjustments include the redistribution of the cardiacoutput towards the skeletal muscle at the expense of the skin, liver and intestine. The effects of theshift in the energy source from carbohydrate to fat at birth on the development of metabolic path-ways including gluconeogenesis, lipogenesis and fatty acid oxidation are also examined.Emphasis is given to the key role of CPT-1 in the regulation of fatty acid oxidation. Finally, theactions of thyroid, HPA and, to a lesser extent, somatotropic axes on the regulation of the energymetabolism are considered. It is concluded that the newborn pig is immature at birth in severalrespects. As a whole, factors involved in the improvement of its postnatal thermogenic capacityare all suggestive of the enhancement of this maturity within the first postnatal days.

1. INTRODUCTION

The neonatal period is attended by important modifications in several physiological functionsassociated with dramatic changes in energy metabolism and nutrition. Before birth, the maternal



organism provides an appropriate environment so that the foetus does not have to actively reg-ulate body temperature. At birth the newborn pig experiences a sudden and dramatic cold stresssince the gap between the dam’s uterus and the ambient temperature (Ta) may be as high as10–12°C. The foetus receives a continuous supply of substrates for growth and metabolismamong which glucose is the main energy substrate (Père et al., 1995). At birth, the piglet isabruptly switched to intermittent feeding of high-fat, low-carbohydrate colostrum and milk.Because the ability to alter metabolic rate is a prerequisite for survival, successful adaptationto this critical period implies that the piglet is able to activate its thermoregulatory mechanismsand to provide energy to its heat-producing tissues. This, in turn, requires profound changes inmajor metabolic pathways including oxidative, lipogenic and gluconeogenic pathways associ-ated with changes in nutrition and hormonal status.

During the past years, several reviews have been devoted to development of the metabolicresponses to cold (Mount, 1968; Curtis, 1974) and the metabolic patterns in the neonatal pig(Mersmann, 1974). Data presented in this chapter rely mostly on recent results which providenew insights on the adaptation of the energy, fatty acids (oxidation, storage) and glucosemetabolism during the neonatal period and their environmental and hormonal control. Theactions of the thyroid, hypothalamic–pituitary–adrenal (HPA) and, to a lesser extent, soma-totropic axis will be developed even though other hormones like insulin play an importantrole in the control of metabolism.

2. OVERVIEW OF THE ENERGY METABOLISM IN THE NEONATAL PIG

2.1. General aspects

Maintenance of homeothermia results from a dynamic balance between heat loss and heatproduction. The newborn pig is poorly insulated, being virtually hairless and devoid of sub-cutaneous fat, and although cold exposure induces some vasoconstriction, it does not reducethe cardiac output to the skin (Lossec et al., 1999). This poor ability of the piglet to conserveheat is reflected by the fact that each 1°C coldness is associated with a 2 kJ/h/kg BW increasein heat production, which is 2.6-fold higher than at the time of weaning (Le Dividich et al.,1998). In fact, the weight-specific requirement for energy of the neonatal pig is maximal atbirth. This is due to the high rate of heat production associated with thermoregulation, phys-ical activity related to the establishment of the nursing order, and a high potential for growth.At birth, the typical ambient temperature (Ta) of 24–20°C is 10–12°C below the lower criti-cal temperature of the piglet and is close to the temperature of 18°C at which the metabolicrate is maximal (Berthon et al., 1993). This period of cold stress is commonly associated witha temporary fall in rectal temperature, the extent of which and the time taken for recoverybeing dependent on body weight and Ta. In practice, this period of hypothermia may be acause of mortality by weakening the piglet and predisposing it to crushing by the sow andstarvation that is illustrated by the usual high level of mortality in the first 48 h after birth.However, cold resistance improves markedly in the early postnatal period in fed piglets butnot in those consuming little or no colostrum (Le Dividich et al., 1991a; Herpin et al., 1994).

Because thermal insulation does not change substantially during the first postnatal days(Berthon, 1994), maintenance of the homeothermic balance is largely dependent on the abilityof the newborn to produce heat. The newborn pig responds quickly and vigorously to cold stressas exemplified by the 30% increase in metabolic rate at 18°C compared with 31°C within thefirst 20 min after birth (Noblet and Le Dividich, 1981), with the difference increasing to 100%

P. Herpin et al.354

in the first 90 min (Herpin et al., 1994). Also (table 1), maximal and minimal metabolic rateare increased by 56% and 28% respectively, during the first 48 h of life (Berthon, 1994).Together, these indicate that mechanisms responsible for heat production are active soon afterbirth. However, maintenance of a high metabolic rate during cold stress is closely dependenton both the availability of energy substrates and on the ability of the piglets to utilize these asan energy source. Once the piglet displays an efficient thermoregulatory behaviour and oncea regular milk intake is established, cold stress is usually of minor importance during theremaining suckling period. The energy metabolism is then regulated essentially by theamount of milk intake (Marion and Le Dividich, 1999).

2.2. Evidence for the major importance of muscle in energy metabolism

To cope with the neonatal cold challenge, skeletal muscle plays a central role in neonatalenergy metabolism (fig. 1). Indeed, the calculated contribution of skeletal muscle to total

Energy metabolism in early development of the pig 355

Table 1

Minimal (mmR) and maximal (MmR) metabolic rate, lower critical temperature (LCT) and ambient temperature at which maximal metabolic rate (TMmR) is reached in pigs aged from 2–48 h

Age, h

2 24 48

mmR, kJ/h/kg BW 12.9 16.9 20.2MmR, kJ/h/kg BW 36.6 43.3 >46.8LCT, °C 34.2 33.1 30.2TMmR, °C 17.8 12.8 <10

Adapted from Berthon et al. (1993, 1994).

Fig. 1. Total body O2 consumption and contribution of muscle to total body O2 consumption in relation toage and ambient temperature (TN, thermoneutral; C, cold). Adapted from Lossec et al. (1998b).

oxygen consumption averages 34–40% at thermal neutrality and 50–64% in the cold. Theskeletal muscle is the major contributor to regulatory thermogenesis, accounting for 97% ofthe cold-induced increase in heat production at 5 days of life (Lossec et al., 1998b). In addi-tion, this stimulation of muscle energy metabolism is accompanied by a redistribution ofcardiac output towards skeletal muscles at the expense of digestive tract and liver blood flows(Lossec et al., 1999) which, however, might be detrimental to the digestive processes throughthe possible shortening of energy supply in the cold.

3. MECHANISMS OF HEAT PRODUCTION

3.1. Overview of heat production mechanisms

During cold exposure, maintenance of the homeothermic balance is achieved through thedevelopment of specific heat-producing mechanisms. Two main mechanisms are usuallyreported: shivering and non-shivering thermogenesis (NST). NST has been defined as “heat-producing mechanisms due to processes that do not involve muscular contractions, such asthose involved in ion pumping or mitochondrial loose coupling mediated by uncouplingprotein-1 (UCP1) in brown fat”.

Shivering is defined as an involuntary rhythmic contraction of skeletal muscle myofibrilsinvolving no voluntary movements or external work. Heat production during shivering involvesbiochemical mechanisms close to those associated with the contraction of the skeletal musclemyofibril: heat is produced during both the hydrolysis of ATP and the associated processes ofATP re-synthesis, and all energy substrates can support shivering. Tremendous amounts of heatcan be produced during shivering but this mechanism is not very efficient because it occurs atthe periphery of the body, and therefore enhances thermolysis, and it impairs physical move-ments. In addition, heat produced during physical activity and meal consumption can contributesignificantly to the extra thermoregulatory heat produced in the cold.

3.2. Contribution of feeding to extra heat production

A major role of colostrum is the provision of immunoglobulins. However, owing to the lowbody energy stores at birth, it is obvious that colostrum is of utmost importance in the provi-sion of energy in the first day of life. This is evidenced by the fact that in cold conditions bothbody temperature and heat production are positively related to the amount of colostrum intake(Noblet and Le Dividich, 1981). Moreover, ingestion of colostrum results in metabolic heatproduction (Gentz et al., 1970). This metabolic response, referred to as postprandial thermo-genesis, represents the energy cost associated with digestion, absorption and processing ofnutrients. However, its contribution to thermoregulation is marginal, accounting for about10% of the extra heat produced in the cold (Herpin et al., 1994), probably because of the high(0.91) efficiency of colostral metabolizable energy (ME) for growth (Le Dividich et al.,1994). In contrast, the efficiency of milk ME for growth is lower (0.73; Marion and Le Dividich, 1999), suggesting a possible higher contribution of the thermal effect of milk tothe extra heat produced in the cold.

3.3. Shivering or non-shivering thermogenesis

The nature of heat production mechanisms in cold-exposed newborn mammals has long beenan open question. It has been known for years that the newborn pig shivers vigorously from

P. Herpin et al.356

birth (Mount, 1968) and is at variance with most other newborn mammals in that it containsvery little adipose tissue of any type at birth, and appears to have no brown fat. The unlikelypresence of brown fat and existence of NST are suggested by (i) the absence of calorigenicresponse to the injection of noradrenaline during the first week of life (LeBlanc and Mount,1968; Brück et al., 1969) and (ii) the immunoblotting studies of Trayhurn et al. (1989) showingthat piglets do not express UCP1 in various tissues. However, some doubts subsist, becausesmall quantities of adipose tissue resembling brown adipose tissue have been detected in3-month-old pigs (Dauncey et al., 1981). In skeletal muscle, the situation is even more intriguingbecause loose-coupled mitochondria have been detected in skeletal muscle of 2-month-oldcold-adapted pigs (Herpin and Barré, 1989). Interestingly, UCP homologues have been recentlyidentified in numerous mammalian tissues including skeletal muscle (review in Ricquier andBouillaud, 2000), which reinforces interest in the molecular mechanisms underlying cellularthermogenesis.

These proteins (UCP) allow the dissipation of part of the proton electrochemical gradientgenerated by the electron transfer chain across the inner mitochondrial membrane and canthus increase heat production by uncoupling respiration from ATP synthesis (Boss et al.,1998b). This role is probably significant since the proton leak, in part sustained by UCPs,contributes up to 50% to skeletal muscle basal respiration rate and nearly 30% to standardmetabolic rate in the rat (Brand et al., 1994). UCP1 is exclusively expressed in brown adiposetissue and plays a vital role in protection against cold. The UCP homologues, UCP2 andUCP3, are respectively 59% and 57% identical to UCP1 in their amino acid sequences. UCP2is widely expressed in human and rat tissues whereas UCP3, which shares 73% identity withUCP2, is highly expressed in muscle (Fleury et al., 1997). Both proteins are able to uncouplerespiration when they are recombinantly expressed in yeast (Fleury et al., 1997; Gong et al.,1997) and in myoblasts (Boss et al., 1998a).

3.3.1. Search for uncoupling proteins in piglet skeletal muscle

Despite the absence of UCP1 in pigs, UCP2 and UCP3 are expressed in skeletal muscle(Damon et al., 2000a). However, in agreement with most recent studies, cold stress has noeffect on UCP3 expression in pig skeletal muscle (Damon, unpublished observation).Results that do not support a classic thermogenic uncoupling role for UCP2 or UCP3 includethe absence of an abnormality in thermoregulation in UCP3 or UCP2 knockout mice(Arsenijevic et al., 2000; Gong et al., 2000) and a rise in their skeletal muscle transcriptlevels during fasting, food restriction or chronic exercise. Thus, the physiological impor-tance of UCP2 and UCP3 in regulatory cold-induced thermogenesis is still a matter ofdebate in the pig, as in other species, and the presence of UCP2 and UCP3 in pig muscle isnot conclusive of the existence of NST. As far as we know, the most recent investigationssuggest that UCP3 has the potential to act as a molecular determinant in the regulation ofresting metabolic rate by 3,5,3′-triiodothyronine (T3) (De Lange et al., 2001), and to play arole in the export of fatty acids from mitochondria (Himms-Hagen and Harper, 2001). Inpiglet muscle, the up-regulation of UCP3 mRNA by acute T3 treatment is less marked thanin rats (Damon et al., 2000a).

3.3.2. Search for the existence of non-shivering thermogenesis in the piglet

To establish the possible existence of non-shivering thermogenesis, it is necessary to measuresimultaneously the magnitude of shivering and the level of heat production at temperatures


ranging from thermoneutrality to cold (Barré et al., 1985). As illustrated in fig. 2, this allowscomparison of the lower critical temperature (LCT) with the temperature threshold for shiv-ering (STT). The lack of delay between the cold-induced increase in heat production (LCT)and the onset of shivering (STT), and the strict linearity of the relationship between meta-bolic rate and shivering intensity within the range of ambient temperature studied (fig. 2),confirm the absence of NST and the main role of shivering in neonatal thermogenesis(Berthon et al., 1994). This result is observed at 2 h as well as at 5 days of life and is notmodified after 2 days of cold exposure (Berthon et al., 1994). One interesting point in fig. 2is the reduction of shivering intensity between 2 h and 5 days of age. In the absence of reg-ulatory non-shivering thermogenesis, this reduction of shivering intensity with age for agiven metabolic rate suggests that the thermogenic efficiency of shivering, i.e. the heatpower of shivering, increases with postnatal age in the cold. In other words, more heat isproduced per unit increase in the electrical activity of shivering at 5 days than at 2 h of life(Berthon et al., 1994), which should definitely contribute to the improvement of pigletthermostability after birth.

In pigs aged 5–6 weeks and exposed to cold for 3 weeks, shivering is progressivelyreplaced by NST. In the absence of brown fat, skeletal muscles are potential candidates forsuch adaptations, especially when one considers that skeletal muscle mass contributes sub-stantially to total body weight. Two mechanisms that are likely to contribute to heat productionhave been identified in skeletal muscles from cold-acclimated weaned piglets. First, it is nowwell accepted that cellular O2 consumption and ion pumping are intimately linked (Gregg andMilligan, 1982), and it has been shown that Na+/K+-ATPase-dependent respiration representsabout 20% of whole-muscle O2 consumption. The activity of this pump is increased by 75%in the cold and this increase accounts for 70% of the total increase in muscle respiration(Herpin et al., 1987a; Harrison et al., 1994), which is probably associated with a similarincrease in heat production. Second, heat can also be directly produced at the level of ATPsynthesis. Indeed, a loose coupling between oxidation and phosphorylation is observed insubsarcolemmal mitochondria from rhomboideus muscle (Herpin et al., 1987b). This muscleis located between the shoulders, i.e. at the same place that brown fat, with its mitochondrialuncoupling, occurs in cold-acclimated rats. However, the biological significance and impor-tance of those mechanisms is difficult to assess and the existence of non-shivering thermogenesishas not been confirmed in vivo.

P. Herpin et al.358

Fig. 2. Changes in heat production (�) and shivering intensity (integrated EMG, �) with ambient temperatureat 2 h and 5 days of age. Adapted from Berthon et al. (1994).

3.4. Mechanisms involved in the postnatal improvement of shivering efficiency

All the components of skeletal muscle energy metabolism can be potentially involved in thepostnatal increase in the capacity to produce heat during shivering. Indeed, to produce heat,muscle fibres need to exhibit both an optimal coupling between excitation and contraction, anoptimal power of contraction, an adequate supply of substrates and oxygen, and finally anefficient ATP synthesis (fig. 3).

3.4.1. Mechanisms and structures involved in muscle contraction

During shivering, repetitive contractions of muscle fibres produce heat by myosin cross-bridgecycling, Ca2+ cycling and Na+/K+ transport. Obviously, the power of contraction of musclefibres is directly related to myofibril mass and, unfortunately, piglet fibres definitely lackmyofibrils at birth (Bradley et al., 1980; Handel and Stickland, 1987; Herpin et al., 2002).Recent results show that myofibril volume density is quite low in longissimus lumborum(LL, 32%) and rhomboideus (RH, 40%) muscles at birth, and increases markedly within 5 days(fig. 4). The high rate of muscle protein synthesis in the newborn is reported to be restrictedentirely to the myofibrillar protein compartment (Fiorotto et al., 2000). This marked postnatalincrease in myofibril mass is probably one of the key events in the development of muscle func-tion and should contribute to the enhancement of muscle contraction potential during shivering.

With regards to muscle fibre types, endurance and high aerobic capacities are importantfeatures for thermogenesis because of the continuous nature of shivering. Structures enhanc-ing endurance are mitochondria and their fuel source, lipids, and they are mostly present inslow-oxidative (type I) and fast-oxidoglycolytic (type IIa) fibres. Moreover, concerning theeconomy of contraction, fast muscles consume more energy than slow when working for thesame time against the same load (Crow and Kushmerick, 1982) and fast-cycling muscles have


Fig. 3. Overview of the structures and metabolic pathways involved in muscle contraction and heat productionduring shivering.

relatively uneconomical force development owing to their high cross-bridge cycling rate andshort contraction time (Alpert and Mulieri, 1986). As a whole, fast-contracting high-endurance fibers (type IIa fibres) may produce more heat per unit time than other fibres whenused in shivering. Interestingly, the proportion of type IIa fibres is very high in piglet musclesat birth (Lefaucheur and Vigneron, 1986). In addition, transitory expression of α-cardiacmyosin heavy chain (MHC), with contractile properties intermediate between type I andtype IIa MHC, has been recently demonstrated in muscle fibres of piglets (Lefaucheur et al.,1997). Moreover, the expression of this myosin isoform is markedly increased within the first5 days of life and is stimulated by chronic cold exposure (Lefaucheur et al., 2002). Althoughour knowledge of α-cardiac MHC properties is very limited, its marked up-regulation in cold-exposed shivering piglets points out its potential role in the enhancement of shiveringefficiency after birth.

Also, the excitation–contraction coupling apparatus appears to be immature at birth in pigs,as already shown in birds (Eppley and Russell, 1995). Triads correspond to the junctionalassociation of transverse tubules with sarcoplasmic reticulum terminal cisternae in matureskeletal muscle (fig. 3) and thereby play a crucial role in calcium release in excitation–contraction coupling (Flucher, 1992). Electron microscopic examination showed that theyproliferate rapidly in LL and RH muscles of cold-exposed piglets after birth (+70% to +90%within 5 days) (fig. 5), suggesting that the efficiency of excitation–contraction coupling

P. Herpin et al.360

Fig. 5. Number of triad profiles per unit fibre area (Ntriad) in longissimus lumborum (LL) and rhomboideus(RH) muscles of newborn (open bars) and five-day old piglets exposed to thermoneutral (striped bars) or cold(solid bars) conditions. Means sharing a common superscript did not differ significantly (P > 0.05). Adaptedfrom Herpin et al. (2002).

Fig. 4. (A) Fibre (Vfibre) and (B) myofibril (Vmyofibril) volume density in longissimus lumborum (LL) andrhomboideus (RH) muscles of newborn (open bars) and 5-day-old piglets exposed to thermoneutral (stripedbars) or cold (solid bars) conditions. Means sharing a common superscript did not differ significantly (P > 0.05). The effect of muscle type on Vfibre is significant (P < 0.05). Adapted from Herpin et al. (2002).

increased in the cold (Herpin et al., 2002). In other words, for a given electrical stimulationmore of the sarcoplasmic reticulum is likely activated during contraction and calcium releaseand cycling are higher, thereby promoting an increase in ATP hydrolysis and heat productionthrough cross-bridge cycling, calcium reuptake and myofibril contraction (Block, 1994).Therefore, triad proliferation is probably of utmost importance in the adaptive and improvedresponse of skeletal muscle to sustained shivering during the early neonatal period.

3.4.2. Muscle blood supply

Blood flow through muscles is highly adaptable and represents an important determinant ofoxygen and nutrient supply and consumption in all species (Hoppeler et al., 1981). It shouldbe sufficient to supply oxygen and nutrient during shivering activity and to support heat pro-duction mechanisms. When comparing muscle blood flow in 1- and 5-day-old piglets, itappears clearly that muscle cardiovascular adjustments in the cold are limited in the newbornpig (table 2).

Measurement of blood flow in a representative muscle compartment, i.e. the hindquarters,in conscious piglets shows that muscle blood flow increases with age and short-term coldexposure. However, changes in blood flow in response to a similar cold challenge were 3times higher in 5-day-old (+65%) than in 1-day-old (+25%) piglets, suggesting that bloodsupply to the shivering muscle was considerably improved with age (Lossec et al., 1998b).Measurement of blood flow in individual muscles using coloured microspheres confirmsthese results (Lossec et al., 1999). Interestingly, a preferential redistribution of cardiac outputtowards skeletal muscle was only observed at 5 days of life, at the expense of the small intes-tine, the liver and the skin; this cardiovascular response was more pronounced in the mostoxidative skeletal muscles studied (RH vs. LL). This should favour, and probably potentiate,the efficiency of shivering. Indeed, a redistribution of cardiac output to the most thermogenic


Table 2

Cardiac output and its fractional distribution Fco to selected tissues or whole organs in 1- and5-day-old pigs exposed to thermoneutral (TN) or cold (C) environment

1-day-old 5-day-old

TN C TN C

Cardiac output, mL/min/kg 436 535*,a 448 550*Fco for 10 g tissue (%)

Longissimus thoracis 0.42 0.55* 0.42 0.61*Rhomboideus 0.47 0.64* 0.49 0.87*Subcutaneous adipose tissue 0.25 0.22 0.21 0.22Skin 0.25 0.25 0.35 0.27*

Fco to whole organs, %Heart 1.87 2.45* 3.24 4.61*Liver 3.97 3.75 6.44 5.18*Small intestine 13.6 14.0 13.7 10.8*Brain 2.72 2.86 3.27 3.12Adrenals 0.14 0.14 0.13 0.15Thyroid 0.08 0.09 0.08 0.11*

aAt a given age. *denotes significant effect (P < 0.05) of cold.Adapted from Lossec et al. (1999).

organs has already been reported in rats (Foster and Frydman, 1979), pigs (Mayfield et al.,1986) and ducklings (Duchamp and Barré, 1993). Additional investigations using electronmicroscopy have shown no increase in skeletal muscle capillary bed with age or cold expo-sure (Herpin et al., 2002). This suggests that the observed postnatal improvement ofcardiovascular adjustments required for shivering thermogenesis may be accommodated byexisting capillaries. Therefore, present changes are more likely related to the postnatal matu-ration of the nervous and hormonal regulation of muscle blood flow.

3.4.3. Activity and ultrastructure of muscle mitochondria

The plasticity of mitochondrial density and activity according to age or cold acclimation iswell documented. During shivering, optimal supply of ATP via oxidative phosphorylation isnecessary when glycogen stores are exhausted. In contrast to liver mitochondria (Mersmannet al., 1972), muscle mitochondria are functional from birth in pigs and are changing prima-rily quantitatively during the first 5 days of life (Berthon et al., 1996a; Schmidt and Herpin,1997). The oxidative potential of pig muscle increases gradually after birth, but no consistentchanges in mitochondrial respiration, respiratory control and phosphorus:oxygen ratio areevidenced during this period and after short-term cold exposure (Schmidt and Herpin, 1997).However, biochemical characteristics of intermyofibrillar (IMF) and subsarcolemmal (SS)mitochondria differ from birth. The higher respiration rate and higher respiratory control ratioshown by IMF compared with those shown by SS mitochondria are principally due to thehigher activities involved in substrate oxidation because there is no difference in the protonleak between both populations (Lombardi et al., 2000).

Thus, the actual event responsible for the postnatal increase in skeletal muscle oxidativepotential is the enhancement of mitochondrial mass (Schmidt and Herpin, 1997), as alreadyreported in various tissues and species during the neonatal period (Mersmann et al., 1972;Eppley and Russell, 1995). Between birth and 5 days of life, mitochondrial mass increasedby 49% in glycolytic LL muscle and by 93% in oxidative RH muscle. In LL muscle thisincrease was only supported by the proliferation of IMF mitochondria, whereas both typesof mitochondria (IMF and SS) proliferate in RH muscle. The mechanisms underlying thesechanges have been elucidated by electron microscopic examination. Within 5 days (fig. 6),there is an increase in both the number of mitochondria and the surface of the inner mem-brane and cristae of each mitochondrion (Herpin et al., 2002). Indeed, the number ofrespiratory chain and F1-ATPase units is directly related to this parameter (Hoppeler, 1986).Interestingly, this postnatal change in the surface of the inner membrane is more marked in RH than in LL muscle, and is further enhanced when piglets are exposed to cold for 5 days. As a whole, this should contribute to the enhancement of muscle endurance duringcontractile activity associated with shivering, and to the postnatal acquisition of musclemetabolic type.

4. SOURCES OF ENERGY

The requirement for energy in the newborn pig is met by body energy reserves, colostrum andmilk. During the neonatal period the protein accretion is very high, and the potential for pro-tein deposition is probably beyond that allowed by milk intake (Le Dividich and Sève, 2001).The rate of amino acid catabolism is very low during this period and is not enhanced in coldconditions (Herpin et al., 1992). The contribution of energy derived from amino acid catabo-lism is therefore of marginal importance and will not be discussed.

P. Herpin et al.362

At birth, body energy stores are present as glycogen and fat (fig. 7). Because FFA arepoorly transferred across the swine placenta (Thulin et al., 1989), the amount of fat reservesin the newborn pig is low, ranging from 15 to 20 g/kg body weight (BW). Most (45%) of thisstored fat is structural fat and is not available for mobilization. Total body glycogen storesrange from 30–38 g/kg BW with the major part (~90%) being located in muscle. From an


Fig. 6. Stereological parameters of mitochondria in longissimus lumborum (LL) and rhomboideus (RH)muscles of newborn (open bars) and 5-day-old piglets exposed to thermoneutral (striped bars) or cold (solidbars) conditions.(A) Number of mitochondria per unit fiber area (Nmitochondria); (B) Mitochondria volumedensity (Vmitochondria); (C) Surface of outer mitochondrial membranes per unit tissue volume(Somm/Vtissue); (D) Surface of inner mitochondrial membrane and cristae per unit mitochondrial volume(Si+c/Vmitochondria). Means sharing a common superscript did not differ significantly (P > 0.05). The effectof muscle type on Si+c/Vmitochondria is significant (P < 0.05). Adpated from Herpin et al. (2002).

Fig. 7. Available energy stores at birth and cumulative available energy derived from ingested colostrums.Adapted from Mellor and Cockburn (1986) and Le Dividich et al. (1997).

energy point of view, glycogen represents >90% of the available stored energy. Nevertheless,available energy derived from body stores is low, accounting for <10% of that of the newborninfant (Mellor and Cockburn, 1986).

Soon after birth the pig is fed at intervals with colostrum for 24–36 h and thereafter milk.The first suckling occurs in the 20–30 min following birth and colostrum intake can be veryhigh immediately after birth since the first three sucklings account for ~25% of the totalcolostrum ingested during the first day (Fraser and Rushen, 1992; Le Dividich et al., 1997).Although highly variable, total consumption of colostrum in the first day of life is in the rangeof 310–340 g/kg BW and may be as high as 460 g/kg BW (Le Dividich et al., 1997). Fataccounts for 35–50% and 55–65% of total energy of colostrum and milk, respectively, andrepresents the main source of energy. Colostrum and milk fats are highly digestible (~100%).They are composed of long-chain fatty acids (>C14), but are devoid of medium-chain fattyacid (MCFA). Both the fatty acid profile and amount of colostrum and milk fat can be manip-ulated by the source of dietary fat provided to the dam in late gestation and throughoutlactation (Averette et al., 1999), but there is no evidence for mammary MCFA transfer(Newcomb et al., 1991). Moreover, lactose is the predominant carbohydrate in colostrum andmilk, but its content is lower in colostrum than in milk (3.1–3.9% vs. 4.8–5.5%).

This abrupt shift in the source of energy substrates from mainly glucose to colostral andmilk fat implies that the piglet is rapidly capable of (i) oxidizing fat to provide energy for heatproduction, (ii) depositing fat for thermal insulation and energy through its mobilization atweaning, and (iii) providing glucose to its glucose-dependent tissues.

5. EFFECTS OF THE SHIFT IN THE ENERGY SOURCE ON LIPIDOXIDATION, LIPOGENESIS AND GLUCONEOGENESIS

5.1. Lipid oxidation

During the first postnatal hours, piglets rely almost entirely on carbohydrate to meet theirthermoregulatory needs (Mount, 1968). However, in usual environmental conditions at birth,75% of liver glycogen and 41% of muscle glycogen is mobilized by 12 h postpartum (Elliotand Lodge, 1977). Cold exposure hastens the depletion in both tissues (Herpin et al., 1992),and increases the rate of glucose turnover (Lossec et al., 1998a) and of peripheral glucoseuptake (Duée et al., 1988; Lossec et al., 1998a). However, colostrum intake increases avail-ability of lipids. The ensuing increase in plasma NEFA (Le Dividich et al., 1991b) and glycerol(Bengtsson et al., 1969) secondary to the increased activity of the adipose tissue hormone-sensitive lipase (Horn et al., 1973; Steffen et al., 1978) is associated with a progressivedecline in respiratory quotient during the first postnatal day in both thermoneutral and coldenvironments (fig. 8), providing evidence for an early involvement of lipids as an energysource (Noblet and Le Dividich, 1981; Berthon et al., 1993). At 48 h of age, fatty acidsaccount for ~60% of the energy metabolism, increasing to 90% in the 7-day-old pig fed atmaintenance (Marion and Le Dividich, 1999). Biochemical adjustments associated with theimproved ability of the piglet to oxidize fat in the skeletal muscle include:1. An increase in muscle lipid content within the first postnatal days. For example, muscle

lipid content nearly doubled within 5 days in LL and RH muscle, in agreement with theincrease in the number of lipid droplets per unit tissue area (Herpin et al., 2002). At birththese lipid droplets are scarce, but at 5 days they are wedged between the myofibrils andthe IMF mitochondria, a position that is ideal for optimizing the provision of energy foroxidative metabolism and sustained shivering.

P. Herpin et al.364

2. An increase in NEFA uptake by the shivering muscle at 5 days of life whereas thisuptake is negligible in 1-day-old piglets (Lossec et al., 1998b).

3. An increase in mitochondrial (+100%) and peroxisomal (+160%) β-oxidation potentialin LL and RH muscle homogenates between birth and 5 days of life (Herpin, unpublishedobservation). Interestingly, a similar increase in mitochondrial and peroxisomal β-oxidationpotential has been previously reported in liver, kidney and heart of young pigs (Yu et al., 1997).Deeper investigations into the molecular and biochemical regulation of fatty acid oxidation

in piglet skeletal muscle showed that oleic, linoleic and palmitic acids are readily oxidizedfrom birth by isolated skeletal muscle mitochondria (Schmidt and Herpin, 1998). MCFA(C8–C10), which are now being introduced in colostrum and milk substitutes, are readily oxi-dized by the liver (for review, see Odle, 1997). However, in vivo studies in respirationchambers (Léon et al., 1998) provide evidence that substitution of MCFA for long-chain fattyacids in colostrum does not improve the energy status of the newborn, even in cold conditions(fig. 9). It is suggested that MCFA are poorly oxidized by skeletal muscle as indicated byin vitro studies using isolated muscle mitochondria (Schmidt and Herpin, 1997). Surprisingly,the mitochondrial potential is not increased with age, which suggests that the enhancement offatty acid oxidation potential with age is mostly supported by the above-mentioned postnatalproliferation of muscle mitochondria.

Complex changes in the expression and activity of carnitine palmitoyltransferase I (CPT I),which is the limiting enzyme of fatty acid β-oxidation, have also been observed (Odle et al.,1995; McGarry and Brown, 1997). In piglets, CPT I activity increases postnatally in SSmuscle mitochondria and is modulated by malonyl-CoA in IMF mitochondria (Schmidt andHerpin, 1998). Indeed, between birth and 5 days of life, both the sensitivity of CPT I to mal-onyl-CoA inhibition and the tissue level of malonyl-CoA decreased, which could partlyrelieve CPT I inhibition and enhances fatty acid utilization. Further, during cold stress, thedecrease in the tissue levels of malonyl-CoA is even more marked in the most oxidativemuscle. A surprising result is the difference in CPT I sensitivity to malonyl-CoA betweenpiglets and rats: in piglets, sensitivity to malonyl-CoA is much lower in muscle than in the


Fig. 8. Respiratory quotient of the neonatal pig in relation to age and ambient temperature. Adapted fromBerthon et al. (1993) and Schmidt and Herpin (1997).

liver (Schmidt and Herpin, 1998; Nicot et al., 2001) whereas the opposite has been obtainedin rats. At the molecular level, two isoforms with tissue-specific expression and sensitivity tomalonyl-CoA inhibition are usually described.

Recently, partial liver (CPT1-L) and muscle (CPT1-M) cDNA sequences have been suc-cessfully cloned and the co-expression of these two isoforms in skeletal muscle of neonatalpiglets has been demonstrated (Damon et al., 2000b). The expression of CPT1-L in pig skele-tal muscle could provide a partial answer to the difference of sensitivity to malonyl-CoAbetween rat and pig muscles (Schmidt and Herpin, 1998). However, co-expression of bothisoforms should result in an intermediate and not a reverse sensitivity to malonyl-CoA inhi-bition. Recent data support another seductive hypothesis. In fact, in yeast expressing pigCPT1-L, kinetic characteristics (Km’s for carnitine and palmitoyl-CoA) were similar to thoseof human and rat CPT1-L whereas sensitivity to malonyl-CoA inhibition was found to becloser to that of rat and human CPT1-M isoforms (Nicot et al., 2001). It then appears that pigCPT1-L possesses specific biochemical properties despite its high degree of homology withCPT1-L from other mammals. Thus pig CPT1-M could also behave as CPT1-L of other mam-mals in terms of malonyl-CoA sensitivity. It has been speculated that CPT1-L was prone tohormonal regulation whereas CPT1-M was more regulated by nutritional factors (Cook et al.,2001). Thus co-expression of both isoforms in pig skeletal muscle could allow a fine tuningof lipid utilization.

5.2. Lipogenesis

The neonatal pig has a remarkable capacity to deposit large amounts of fat soon after birth.Depending on the colostrum fat content, carcass fat content increases by 25–100% during thefirst day of life (Le Dividich et al., 1997). During the suckling phase, fat accretion occurs ata mean rate of 30–35 g/day, depending mainly on the amount of ingested milk (Marion andLe Dividich, 1999) and on the milk fat content (Jones et al., 1999). De novo lipogenesis ismarginal, albeit being limited not by enzyme activities or insulin-regulated glucose trans-porter (GLUT 4), but by the substrate availability (Gerfault et al., 2000). Colostrum and milkare the major routes for lipid acquisition (Sarkar et al., 1985). Adipose tissue lipoprotein lipase,

P. Herpin et al.366

Fig. 9. Effect of substitution of medium- (MCFA) for long- (LCFA) chain fatty acids in colostrum on heatproduction (a) and respiratory quotient (b) in the newborn pig in relation to environmental temperature. Dataadjusted to a common ME intake of 960 kJ/kg BW per 26 h. Adapted from Léon et al. (1998).

an enzyme playing a key role in fat storage, contributes to the ability of the suckling pig todeposit large amounts of fat. Its activity, already high at birth, is increased 3–4-fold in the earlyneonatal period (Steffen et al., 1978; Le Dividich et al., 1997). It is suggested that sow’s milk isdesigned to promote fat accretion in the young pig. Stored fat is mostly subcutaneous fat thatprovides thermal insulation to the young pig and energy through its mobilization during theperiod of low feed intake following weaning (Le Dividich and Sève, 2001).

5.3. Gluconeogenesis

The glucose requirement of the neonatal pig is very high, reaching ~15 g/kg BW per day(Flecknell et al., 1980; Pégorier et al., 1984). It is 50% higher than in lambs and humaninfants. Moreover, the glucose requirement is enhanced ~30% in the cold (Duée et al., 1988;Lossec et al., 1998a). Glucose requirements are met by (i) liver glycogenolysis, since onlyliver glycogen is able to release glucose into the blood, (ii) colostrum and milk and (iii) glu-coneogenesis. Provision of glucose by the first two sources meets 50–60% of therequirements during the first day of life, which underlines the importance of the gluco-neogenic pathway in the glucose homeostasis of the piglet.

Gluconeogenesis is the process by which glucose is synthesized from various precursors.The liver is the main site of gluconeogenesis. The developmental pattern of hepatic gluco-neogenesis has been the subject of several extensive reviews (Girard, 1986; Girard et al.,1992) and will not be discussed in detail in this chapter. In brief, key enzymes involved in thepathway, i.e. pyruvate carboxylase, phosphoenolpyruvate carboxykinase, fructose-1,6-diphosphatase and glucose-6-phosphatase (G6Pase), have a substantial activity (35-105% ofadult values) at birth. In both fed and unfed pigs, enzyme activity increases markedly duringthe first postnatal day. Also, the insulin:glucagon molar ratio decreases after birth in both fedand unfed piglets, thus providing an appropriate environment for an active gluconeogenesis.However, the level of plasma NEFA is much higher in the fed pigs. In fact, colostrum ingestionis essential to sustain a high rate of hepatic gluconeogenesis. Based on the amount of glucoseavailable from lactose digestion (glucose + galactose), colostrum provides at least 40–45% ofthe glucose requirement and ~80–90% on the assumption that all galactose is converted intoglucose by the G6Pase. However, piglets fasted from birth (Goodwin, 1957) or fed a low-fatcolostrum (Herpin et al., 1992) are unable to sustain normal glycaemia. It is suggested that fattyacid oxidation plays a major role in glucose homeostasis (Girard, 1986) through the supply ofATP and co-factors (NADH and acetyl-CoA) for catalysing key reactions. This is convincinglydemonstrated by the in vivo and in vitro studies of Pégorier et al. (1985) and Duée et al. (1985).In contrast, Lepine et al. (1991) failed to find any stimulatory effects of fatty acid oxidation onthe rate of glucose production by isolated hepatocytes from piglets.

6. REGULATION OF ENERGY METABOLISM DURING EARLY DEVELOPMENT

6.1. The thyroid axis

Thyroid hormones (TH) are known to play a major role in the regulation of metabolic adap-tations and growth. They exert their effects primarily through interactions with nuclear THreceptors (TR) which occur as a series of isoforms controlling the transcription of thyroidhormone-responsive genes (Lazar, 1993; Wrutniak-Cabello et al., 2001). The ontogenic


profile of the thyroid system suggests that TH metabolism is fully developed at birth. Plasmaconcentrations of both total and free TH, thyroid gland weights and hepatic 5′-deiodinaseactivity all increase during late gestation (Berthon et al., 1993). It is relevant to notice thatreceptors are already present in skeletal muscle, but not in the liver, at 80 days of gestation,suggesting that porcine muscle can potentially respond to TH much earlier in developmentthan the liver (Duchamp et al., 1994). During the first 6 h after birth, there is a surge in T3,free T3 and T4 plasma concentrations and, apart from a transient decline at 12 h, TH concen-trations remain elevated during the first 2 days and then decline slightly over the next 2 weeks(Slebodzinski et al., 1981; Berthon et al., 1993, 1996b).

The finding that the postnatal surge in plasma TH levels precedes the physiological rise inheat production in the newborn suggests a close relationship between perinatal thyroid statusand neonatal thermogenic capacity (Berthon et al., 1993). This is exemplified by the findingsthat (i) hypothyroidism at birth is associated with depressed thermoregulatory capabilities ofthe newborn (Berthon et al., 1993) and (ii) a single injection of T4 induces an increase inmetabolic rate (Slebodzinski, 1979). TH control the oxidative capacities in the newbornthrough a short-term regulation of mitochondial respiration (Herpin et al., 1996) and a long-term regulation of mitochondriogenesis (Mutvei et al., 1989). The thyroid axis is alsoregulated by nutrition and TH actions might be complementary to catecholamine actionsduring cold-induced thermogenesis. Cold-exposed newborn pigs fed a limited amount of milkexhibit high catecholamines but low T3 levels whereas the opposite is observed in piglets feda high milk intake (Herpin et al., 1995; Berthon et al., 1996b). These adaptations are assumedto optimize the utilization of either body stores (low intake) or exogenous substrates (highintake). The marked effects of food intake on the thyroid axis are also observed during thewhole suckling period: a low intake reduces thyroid gland activity, circulating TH concentra-tions and nuclear TR abundance in muscle (Dauncey, 1990; Morovat and Dauncey, 1995).

6.2. HPA axis

Circulating levels of glucorticoids and catecholamines are very high at the time of birth anddramatically decrease thereafter (Kaciuba-Uscilko, 1972; Randall, 1983). Cortisol and cate-cholamines are potent stimulators of catabolism and one can speculate that these high levelsinduce mobilization of glycogen stores after birth. However, response of catechalomine tocold exposure during the first 5 days of life is variable, with both no change (Lossec, 1998)and a marked increase (Duée et al., 1988; Le Dividich et al., 1991a) being reported.Moreover, the response to cold is found to be impaired in moderately hypothermic piglets(Mayfield et al., 1989) or, as mentioned above, to be dependent on the level of milk intake(Herpin et al., 1995). A lipolytic response is only detected at 2–4 days of age (Curtis and Rogler,1970; Persson et al., 1971). Neither norepinephrine nor epinephrine administration elicits athermogenic response in the neonatal pig (LeBlanc and Mount, 1968; Persson et al., 1971).Clearly, these observations indicate that the actual role of catecholamines in the neonatalthermogenesis requires further investigation.

6.3. Somatotropic axis

Even though the regulation of energy metabolism by the somatotropic axis is well docu-mented in the growing pig (Louveau and Bonneau, 2001), there is no evidence from theliterature that the somatotropic axis contributes to the cold-induced thermogenesis. However,because of a high potential for protein synthesis and growth of the neonatal pig (Le Dividich

P. Herpin et al.368

and Sève, 2001), one might expect that the energy metabolism is to some extent regulated bythe somatotropic axis. Plasma GH concentrations are very high at birth and decrease sharplyduring the next 2–3 days (Scanes et al., 1987; Carroll et al., 1998). Although the significanceof these high levels of plasma GH is not completely understood, GH could contribute to themaintenance of protein accretion in the newborn pig, even in negative energy balance (Herpinet al., 1992). Plasma IGF-I concentrations increase significantly during the first 3 weeks afterbirth (Lee et al., 1991, 1993; Louveau et al., 1996). After 24 h of feeding, IGFBP profilechanges with the abundance of plasma IGFBP-3 predominating (Lee et al., 1991). Changesin GH and IGF-I receptor levels are also observed during this period, with GH receptorincreasing over the first 10 days of life in liver and IGF-I receptor decreasing in skeletalmuscle and other tissues (Breier et al., 1989; Lee et al., 1993; Louveau et al., 1996;Schnoebelen-Combes et al., 1996). These profiles are modulated by thyroid status (Duchampet al., 1996). In addition, the somatotropic axis appears to be functional and responsive to GHadministration in neonatal pigs, although the responsiveness is reduced compared to olderpigs (Harrell et al., 1999). The administration of GH at a dose that is commonly used in olderpigs has little or no effect on growth rate or plasma IGF-I or IGFBP-3 (Harrell et al., 1999;Dunshea et al., 2001). Perhaps the lack of response in the growth rate is not surprising owingto the already high rate of protein synthesis.

Changes in nutritional status during the neonatal period are associated with several changesin the GH–IGF-I axis. Both moderate and severe feed restriction (Dauncey et al., 1994;Louveau and Le Dividich, 2002) in the suckling period decrease plasma IGF-I and IGFBP-3levels. These data indicate that circulating IGF-I is directly related to energy intake in neonatalpigs as observed in older animals. Even though the regulation of receptors may represent animportant mechanism of control within the GH–IGF-I axis, the few available data indicatethat the regulation of IGF-I and GH receptors is tissue-specific and dependent on the type ofundernutrition during the suckling period (Louveau and Le Dividich, 2002).

7. CONCLUDING COMMENTS AND FUTURE PERSPECTIVES

This chapter provides new insights on the development of the energy metabolism in a speciesdevoid of brown fat. Key factors involved in the poor abilities of the newborn pig to withstandcold stress include mainly the relative immaturity of the newborn pig and the availability ofenergy substrates. Improvement of its thermogenic capacities within the first postnatal daysparallels maturation of the skeletal muscle metabolism and function and of the cellularmachinery.

In the future, in the light of improving survival, it should be relevant to select piglets onphysiological traits related to maturity. This is convincingly attested by the first findings(Leenhouvers, 2001) that selection of pigs with different genetic merit for survival leads topiglets with a higher maturity at birth. Attempts made to improve the energy available at birthresulted only in moderate increase in energy stores at birth. However, effects of sow nutritionduring pregnancy on fetal muscle development during the critical stages of fetal developmentwarrant future investigation. In addition, we suggest that more research should be focused onfactors initiating and controlling quantity and quality of colostrum and milk produced by thesow. However, during the past decades, selection for lean tissue growth has led to less maturepigs at birth (Herpin et al., 1993). Selection of sows for higher litter size has resulted in prob-lems of increased intrapartum deaths, proportion of weak piglets and competition at the udder(Quiniou et al., 2002). Therefore, our efforts will be in vain if the survival of the piglet continues to be challenged unwisely by the pig industry.


REFERENCES

Alpert, N.R., Mulieri, L.A., 1986. Functional consequences of altered cardiac myosin isoforms. Med. Sci.Sport Exerc. 18, 309–313.

Arsenijevic, D., Onuma, H., Pecqueur, C., Raimbault, S., Manning, B.S., Miroux, B., Couplan, E., Alves-Guerra, M.C., Goubern, M., Surwit, R., Bouillaud, F., Richard, D., Collins, S., Ricquier, D., 2000.Disruption of the uncoupling protein-2 gene in mice reveals a role in immunity and reactive oxygenspecies production. Nat. Genet. 26, 435–439.

Averette, L.A., Odle, J., Monaco, M.H., Donavan, S.M., 1999. Dietary fat during pregnancy and lacta-tion increases milk fat and insulin-like growth factor 1 concentrations and improves neonatal growthrates in swine. J. Nutr. 129, 2123–2129.

Barré, H., Geloen, A., Chatonnet, J., Dittmar, A., Rouanet, J.L., 1985. Potentiated muscular thermogen-esis in cold-acclimated Muscovy duckling. Amer. J. Physiol. 249, R533–R538.

Bengsston, G., Gentz, J., Hallarainen, J., Hellstrom, R., Persson, B., 1969. Plasma levels of FFA, glycerol,β-hydroybutyrate and blood glucose during the postnatal development of the pig. J. Nutr. 97, 311–315.

Berthon, D., 1994. Thermogenesis in the neonatal pig. Ph.D. Thesis, Université Claude Bernard, Lyon,122. pp.

Berthon, D., Herpin, P., Duchamp, C., Dauncey, M.J., Le Dividich, J., 1993. Modification of thermogeniccapacity in neonatal pig by changes in thyroid status during late gestation. J. Dev. Physiol. 19, 253–261.

Berthon, D., Herpin, P., Bertin, R., De Marco, F., Le Dividich, J., 1996a. Metabolic changes associated withsustained 48-hr shivering thermogenesis in the newborn pig. Comp. Biochem. Physiol. 114, 327–335.

Berthon, D., Herpin, P., Le Dividich, J., 1994. Shivering thermogenesis in the neonatal pig. J. ThermalBiol. 19, 413–418.

Berthon, D., Herpin, P., Le Dividich, J., Dauncey, M.J., 1996b. Interactive effects of thermal environmentand energy intake on thyroid hormone and metabolism in newborn pigs. Biol. Neonate 69, 51–59.

Block, B.A., 1994. Thermogenesis in muscle. Annu. Rev. Physiol. 56, 535–577.Boss, O., Muzzin, P., Giacobino, J.P., 1998b. The uncoupling proteins, a review. Eur. J. Endocrinol. 139, 1–9.Boss, O., Samec, S., Kuhne, F., Bijlenga, P., Assimacopoulos-Jeannet, F., Seydoux, J., Giacobino, J.P.,

Muzzin, P., 1998a. Uncoupling protein-3 expression in rodent skeletal muscle is modulated by foodintake but not by changes in environmental temperature. J. Biol. Chem. 273, 5–8.

Bradley, R., Ward, P.S., Bailey, J., 1980. The ultrastructural morphology of the skeletal muscles of normalpigs and pigs with splayleg from birth to one week of age. J. Comp. Pathol. 90, 433–446.

Brand, M.D., Chien, L.F., Ainscow, E.K., Rolfe, D.F., Porter, R.K., 1994. The causes and functions ofmitochondrial proton leak. Biochim. Biophys. Acta 1187, 132–139.

Breier, B.H., Gluckman, P.D., Blair, H.T., McCutcheon, S.N., 1989. Somatotrophic receptors in hepatictissue of the developing pig. J. Endocrinol. 123, 25–31.

Brück, K., Wünnenberg, W., Zeisberger, E., 1969. Comparison of cold adaptive metabolic modificationsin different species with special reference to the miniature pig. Fed. Proc. 28, 1035–1041.

Carroll, J.A., Veum, T.L., Matteri, R.L., 1998. Endocrine responses to weaning and changes in post-weaning diet in the young pig. Domest. Anim. Endocrinol. 15, 183–194.

Cook, G.A., Edwards, T.L., Jansen, M.S., Bahouth, S.W., Wilcox, H.G., Park, E.A., 2001. Differentialregulation of carnitine palmitoyltransferase-I gene isoforms (CPT-I alpha and CPT-I beta) in the ratheart. J. Mol. Cell. Cardiol. 33, 317–329.

Crow, M.T., Kushmerick, M.J., 1982. Chemical energy of slow- and fast-twitch muscles of the mouse. J. Gen. Physiol. 79, 147–166.

Curtis, S.E., 1974. Responses of the piglet to perinatal stressors. J. Anim. Sci. 38, 1031–1035.Curtis, S., Rogler, J.C., 1970. Thermoregulatory ontogeny in piglets: sympathetic and adipokinetic

response to cold. Amer. J. Physiol. 218, 149–152.Damon, M., Vincent, A., Lombardi, A., Herpin, P., 2000a. First evidence of uncoupling protein-2

(UCP-2) and -3 (UCP-3) gene expression in piglet skeletal muscle and adipose tissue. Gene 246,133–141.

Damon, M., Vincent, A., Herpin, P., 2000b. Characterization and expression of skeletal and hepatic car-nitine palmitoyl transferase I genes in piglet skeletal muscle. Nutr. Clin. Metabol. 14, Suppl. 2, 158S.

Dauncey, M.J., 1990. Thyroid hormones and thermogenesis. Proc. Nutr. Soc. 49, 203–215.Dauncey, M.J., Burton, K.A., Tivey, D.R., 1994. Nutritional modulation of insulin-like growth factor-I

expression in early postnatal piglets. Pediat. Res. 36, 77–84.Dauncey, M.J., Wooding, F.B., Ingram, D.L., 1981. Evidence for the presence of brown adipose tissue in

the pig. Res. Vet. Sci. 31, 76–81.

P. Herpin et al.370

De Lange, P., Lanni, A., Beneduce, L., Moreno, M., Lombardi, A., Silvestri, E., Goglia, F., 2001.Uncoupling protein-3 is a molecular determinant for the regulation of resting metabolic rate bythyroid hormone. Endocrinology 142, 3414–3420.

Duchamp, C., Barré, H., 1993. Skeletal muscle as the major site of nonshivering thermogenesis in cold-acclimated ducklings. Amer. J. Physiol. 265, R1076–R1083.

Duchamp, C., Burton, K.A., Herpin, P., Dauncey, M.J., 1994. Perinatal ontogeny of porcine nuclear thyroid hormone receptors and its modulation by thyroid status. Amer. J. Physiol. 267, E687–E693.

Duchamp, C., Burton, K.A., Herpin, P., Dauncey, M.J., 1996. Perinatal ontogeny of porcine growth hormone receptor gene expression is modulated by thyroid status. Eur. J. Endocrinol. 134, 524–532.

Duée, P.H., Pégorier, J.P., Le Dividich, J., Girard, J., 1988. Metabolic and hormonal response to acutecold exposure in newborn pig. J. Dev. Physiol. 10, 371–381.

Duée, P.H., Pégorier, J.P., Péret, J., Girard, J., 1985. Separate effects of fatty acid oxidation and glucagonon gluconeogenesis in isolated hepatocytes from newborn pigs. Biol. Neonate 47, 77–83.

Dunshea, F.R., Suster, D., Kerton, D.J., Leury, B.J., 2001. Porcine somatotropin treatment of neonatalpigs alters lifetime fat but not lean accretion. In: Cranwell, P.D. (Ed.), Manipulating Pig ProductionVIII. Australasian Pig Science Association, Werribee, Victoria, Australia, p. 42.

Elliot, J.I., Lodge, G.A., 1977. Body composition and glycogen, reserves in the neonatal pig during thefirst 96 hours post-partum. Can. J. Anim. Sci. 57, 141–150.

Eppley, Z.A., Russell, B., 1995. Perinatal changes in avian muscle: implications from ultrastructure forthe development of endothermy. J. Morphol. 225, 357–367.

Flecknell, P.A., Wootton, R., John, M., 1980. Total body glucose metabolism in the conscious, unre-strained piglet and its relation to body and organ weight. Brit. J. Nutr. 44, 193–203.

Fleury, C., Neverova, M., Collins, S., Raimbault, S., Champigny, O., Levi-Meyrueis, C., Bouillaud, F.,Seldin, M.F., Surwit, R.S., Ricquier, D., Warden, C.H., 1997. Uncoupling protein-2: a novel genelinked to obesity and hyperinsulinemia. Nat. Genet. 15, 269–272.

Fiorotto, M.L., Davis, T.A., Reeds, P., Burrin, D., 2000. Nonnutritive factors in colostrum enhancemyofibrillar protein synthesis in the newborn pig. Pediat. Res. 48, 511–517.

Flucher, B.E., 1992. Structural analysis of muscle development: transverse tubules, sarcoplasmic reticulum,and the triad. Dev. Biol. 154, 245–260.

Foster, D.O., Frydman, M.L., 1979. Tissue distribution of cold-induced thermogenesis in consciouswarm- or cold-acclimated rats reevaluated from changes in tissue blood flow: the dominant role ofbrown adipose tissue in the replacement of shivering by non-shivering thermogenesis. Can. J. Physiol.Pharmacol. 57, 257–270.

Fraser, D., Rushen, J., 1992. Colostrum intake by the newborn piglet. Can. J. Anim. Sci. 72, 1–13.Gentz, J., Bengtsson, G., Hakkarainen, J., Hellström, R., Persson, B., 1970. Factors influencing oxygen

consumption in the newborn pig with special reference to feeding. Biol. Neonate 16, 328–341.Gerfault, V., Louveau, I., Mourot, J., Le Dividich, J., 2000. Lipogenic enzyme activity in subcutaneous

adipose tissue and skeletal muscle from neonatal pig consuming maternal or formula milk. Reprod.Nutr. Dev. 40, 103–111.

Girard, J., 1986. Gluconeogenesis in late and early life. Biol. Neonate 59, 257–267.Girard, J., Férré, P., Pégorier, J.P., Duée, P.H., 1992. Adaptations of glucose and fatty acid metabolism

during perinatal period and suckling-weaning transition. Physiol. Rev. 72, 501–562.Gong, D.W., He, Y., Karas, M., Reitman, M., 1997. Uncoupling protein-3 is a mediator of thermogenesis

regulated by thyroid hormone, beta3-adrenergic agonists, and leptin. J. Biol. Chem. 272, 24129–24132.Gong, D.W., Monemdjou, S., Gavrilova, O., Leon, L.R., Marcus-Samuels, B., Chou, C.J., Everett, C.,

Kozak, L.P., Li, C., Deng, C., Harper, M.E., Reitman, M.L., 2000. Lack of obesity and normalresponse to fasting and thyroid hormone in mice lacking uncoupling protein-3. J. Biol. Chem. 275,16251–16257.

Goodwin, R.F.W., 1957. The relationship between the concentration of blood sugar and some vital func-tions in the newborn pig. J. Physiol. (Lond.) 136, 208–217.

Gregg, V.A., Milligan, L.P., 1982. Role of Na+/K+-ATPase in muscular energy expenditure of warm- andcold-exposed sheep. Can. J. Anim. Sci. 62, 123–132.

Handel, S.E., Stickland, N.C., 1987. The effect of low birthweight on the ultrastructural development oftwo myofiber types in the pig. J. Anat. 150, 129–143.

Harrell, R.J., Thomas, M.J., Boyd, R.D., Czerwinski, S.M., Steele, N.C., Bauman, D.E., 1999. Ontogenicmaturation of the somatotropin/insulin-like growth factor axis. J. Anim. Sci. 77, 2934–2941.

Harrison, A.P., Clausen, T., Duchamp, C., Dauncey, M.J., 1994. Roles of skeletal muscle morphology andactivity in determining Na+–K+-ATPase concentration in young pigs. Amer. J. Physiol. 266, R102–R111.


Herpin, P., Barré, H., 1989. Loose-coupled subsarcolemmal mitochondria from muscle rhomboideus incold-acclimated piglets. Comp. Biochem. Physiol. B 92, 59–65.

Herpin, P., McBride, B.W., Bayley, H.S., 1987a. Effects of cold exposure on energy metabolism in theyoung pig. Can. J. Physiol. Pharmacol. 65, 236–245.

Herpin, P., Bertin, R., Le Dividich, J., Portet, R., 1987b. Some regulatory aspects of thermogenesis incold-exposed piglets. Comp. Biochem. Physiol. 87, 1073–1081.

Herpin, P., Le Dividich, J., van Os, M., 1992. Contribution of colostral fat to thermogenesis and glucosehomeostasis in the newborn pig. J. Dev. Physiol. 17, 133–141.

Herpin, P., Le Dividich, J., Amaral, N., 1993. Effect of selection for lean tissue growth on body compo-sition and physiological state of pig at birth. J. Anim. Sci. 71, 2645–2653.

Herpin, P., Le Dividich, J., Berthon, D., Hulin, J.C., 1994. Assessment of thermoregulatory and post-prandial thermogenesis over the first 24 hours after birth in pigs. Exp. Physiol. 79, 1011–1019.

Herpin, P., Berthon, D., Bertin, R., De Marco, F., Dauncey, M.J., Le Dividich, J., 1995. Cold-inducedchanges in circulating levels of catecholamines and thyroid hormones are modulated by energy intakein newborn pigs. Exp. Physiol. 80, 877–880.

Herpin, P., Berthon, D., Duchamp, C., Dauncey, M.J., Le Dividich, J., 1996. Effect of thyroid status inthe perinatal period on oxidative capacities and mitochondrial respiration in porcine liver and skele-tal muscle. Reprod. Fertil. Dev. 8, 147–155.

Herpin, P., Lossec, G., Schmidt, I., Cohen-Adad, F., Duchamp, C., Lefaucheur, L., Goglia, F., Lanni, A.,2002. Effect of age and cold exposure on morphofunctional characteristics of skeletal muscle inneonatal pigs. Pflügers Arch. Eur. J. Physiol. 444, 610–618.

Himms-Hagen, J., Harper, M.-E., 2001. Physiological role of UCP3 may be export of fatty acids frommitochondria when fatty acid oxidation predominates: an hypothesis. Exp. Biol. Med. 226, 78–84.

Hoppeler, H., 1986. Exercise-induced ultrastructural changes in skeletal muscle. Int. J. Sports Med. 7,187–204.

Hoppeler, H., Mathieu, O., Krauer, R., Claassen, H., Armstrong, R.B., Weiber, E.R., 1981. Design of themammalian respiratory system. VI. Distribution of mitochondria and capillaries in various muscles.Resp. Physiol. 44, 87–111.

Horn, G.W., Foley, C.W., Seerley, R.W., Munnel, J.F., 1973. Role of adipose tissue lipolysis in postnatalamelioration of thermogenesis in domestic and wild piglets. J. Anim. Sci. 37, 1356–1361.

Jones, G., Edwards, S.A., Traver, S., Jagger, S., Hoste, S., 1999. Body composition changes in piglets atweaning to nutritional modification of sow milk composition and effects on post-weaning perform-ance. Paper presented at the 50th Annual Meeting of the EAAP, Zürich, 5 pp.

Kaciuba-Uscilko, H., 1972. Hormonal regulation of thermogenesis in the new-born pig. Biol. Neonate21, 245–258.

Lazar, M.A., 1993. Thyroid hormone receptors: multiple forms, multiple possibilities. Endocr. Rev. 14,184–193.

LeBlanc, J., Mount, L.E., 1968. Effects of noradrenaline and adrenaline on oxygen consumption rate andarterial blood pressure in the newborn pig. Nature 217, 77–78.

Le Dividich, J., Esnault, Th., Lynch, B., Hoo-Paris, R., Castex, Ch., Peiniau, J., 1991b. Effect of colostralfat on fat deposition and plasma metabolites in newborn pig. J. Anim. Sci. 69, 2480–2488.

Le Dividich, J., Herpin, P., Paul, E., Strullu, F., 1997. Effect of fat content in colostrum on voluntarycolostrum intake and fat utilization in the newborn pig. J. Anim. Sci. 75, 707–713.

Le Dividich, J., Herpin, P., Rosario-Ludovino, R.M., 1994. Utilization of colostrum energy by the new-born pig. J. Anim. Sci. 72, 2082–2089.

Le Dividich, J., Mormède, P., Catheline, M., Caritez, J.C., 1991a. Body composition and cold resistance of theneonatal pig from European (Large White) and Chinese (Meishan) breeds. Biol. Neonate 59, 268–277.

Le Dividich, J., Noblet, J., Herpin, P.H., van Milgen, J., Quiniou, N., 1998. Thermoregulation. In:Wiseman, J., Varley, M.A., Charlick, J.P. (Eds.), Progress in Pig Science. Nottingham UniversityPress, Nottingham, UK, pp. 229–263.

Le Dividich, J., Sève, B., 2001. Energy requirement of the young pig. In: Varley, M.A., Wiseman, J. (Eds.), The Weaner Pig. CAB International, Wallingford, UK, pp. 17–44.

Lee, C.Y., Bazer, F.W., Etherton, T.D., Simmen, F.A., 1991. Ontogeny of insulin-like growth factors(IGF-I and IGF-II) and IGF-binding proteins in porcine serum during fetal and postnatal develop-ment. Endocrinology 128, 2336–2344.

Lee, C.Y., Chung, C.S., Simmen, F.A., 1993. Ontogeny of the porcine insulin-like growth factor system.Mol. Cell. Endocrinol. 63, 71–80.

P. Herpin et al.372

Leenhouvers, J.I., 2001. Biological aspects of genetic differences in piglet survival. Ph.D. dissertation,Wageningen University, Wageningen, The Netherlands.

Lefaucheur, L., Ecolan, P., Lossec, G., Gabillard, J.C., Butler-Browne, G.S., Herpin, P., 2002. Influenceof early postnatal cold-exposure on myofiber maturation in pig skeletal muscle. J. Muscle Res. Cell.Motil. 22, 439–452.

Lefaucheur, L., Hoffman, R., Okamura, C., Gerrard, D., Leger, J.J., Rubinstein, N., Kelly, A., 1997.Transitory expression of alpha cardiac myosin heavy chain in a subpopulation of secondary genera-tion muscle fibers in the pig. Dev. Dyn. 210, 106–116.

Lefaucheur, L., Vigneron, P., 1986. Post-natal changes in some histochemical and enzymatic character-istics of three pig muscles. Meat Sci. 16, 199–216.

Léon, A., Schmodt, I., Strullu, F., Fillaut, M., Gauthier, J., Hulin, J.C., Lebreton, Y., Herpin, P., LeDividich, J., 1998. Effects of substitution of medium- (MCFA) for long- (LCFA) chain fatty acids incolostrum on energy metabolism of the newborn pig in relation to envirionmental temperature.Journée Rech. Porcine en France 30, 275–280.

Lepine, A.J., Boyd, R.D., Whitehead, D.M., 1991. Effect of colostrum intake on hepatic gluconeogene-sis and fatty acid oxidation in the neonatal pig. J. Anim. Sci. 69, 1966–1974.

Lombardi, A., Damon, M., Vincent, A., Goglia, F., Herpin, P., 2000. Characterisation of oxidative phos-phorylation in skeletal muscle mitochondria subpopulations in pig: a study using top-down elasticityanalysis. FEBS Lett. 475, 84–88.

Lossec, G., 1998. Muscle energy metabolism during cold-induced shivering in the newborn pig. Ph.D.Thesis, Paris University, 210 pp.

Lossec, G., Duchamp, C., Lebreton Y., Herpin, P., 1999. Postnatal changes in regional blood flow duringcold-induced shivering in sow-reared piglets. Can. J. Physiol. Pharmacol. 77, 1–8.

Lossec, G., Herpin, P., Le Dividich, J., 1998a. Thermoregulatory responses of the newborn pig duringexperimentally induced hypothermia and rewarming. Exp. Physiol. 83, 667–678.

Lossec, G., Lebreton, Y., Hulin, J.C., Fillaut, M., Herpin, P., 1998b. Age-related changes in oxygen andnutrient uptake by hindquarters in newborn pigs during cold-induced shivering. Exp. Physiol. 83,793–807.

Louveau, I., Bonneau, M., 2001. Biology and actions of somatotropin in the pig. In: Renaville, R., Burny, A.(Eds.), Biotechnology and Animal Husbandry, Kluwer Academic Press, Dordrecht, The Netherlands,pp. 111–131.

Louveau, I., Combes, S., Cochard, A., Bonneau, M., 1996. Developmental changes in insulin-like growthfactor-I (IGF-I) receptor levels and plasma IGF-I concentrations in Large White and Meishan pigs.Gen. Comp. Endocrinol. 104, 29–36.

Louveau, I., Le Dividich, J., 2002. GH and IGF-I binding in adipose tissue, liver and skeletal muscle inresponse to milk intake level in piglets. Gen. Comp. Endocrinol. 126, 310–317.

Marion, J., Le Dividich, J., 1999. Utilization of the energy in sow’s milk by the piglet. In: Cranwell, P.D.(Ed.), Manipulating Pig Production VII. Australasian Pig Science Association, Werribee, Victoria,Australia, p. 254.

Mayfield, S.R., Stonestreet, B.S., Brubakk, A.M., Shaul, P.W., Susa, J., Oh, W., 1986. Regional bloodflow in newborn piglets during environmental cold stress. Amer. J. Physiol. 251, G308–G313.

Mayfield, S.R., Stonestreet, B.S., Shaul, P.W., Brubakk, A.M., Susa, J., Oh, W., 1989. Plasma catecholamines concentrations of newborn piglets in thermoneutral and cold environments. J. Dev.Physiol. 11, 331–334.

McGarry, J.D., Brown, N.F., 1997. The mitochondrial carnitine palmitoyltransferase system: from con-cept to molecular analysis. Eur. J. Biochem. 244, 1–14.

Mellor, D.J., Cockburn, F., 1986. A comparison of energy metabolism in the newborn infant, piglet andlamb. Q. J. Exp. Physiol. 71, 361–371.

Mersmann, H.J., 1974. Metabolic patterns in the neonatal pig. J. Anim. Sci. 38, 1022–1030.Mersmann, H.J., Goodman, J., Houk, J.M., Anderson, S., 1972. Studies on the biochemistry of mito-

chondria and cell morphology in the neonatal swine hepatocyte. J. Cell. Biol. 53, 335–347.Morovat, A., Dauncey, M.J., 1995. Regulation of porcine skeletal muscle nuclear 3,5,3′-triiodothyronine

receptor binding capacity by thyroid hormones: modification by energy balance. J. Endocrinol. 144,233–242.

Mount, L., 1968. The Climatic Physiology in the Pig. Edward Arnold, London.Mutvei, A., Husman, B., Anderson, G., Nelson, B.D., 1989. Thyroid hormone and not growth hormone

is the principal regulator of mammalian mitochondrial biogenesis. Acta Endocrinol. 121, 223–228.


Newcomb, M.D., Harmon, D.L., Nelssen, J.L., Thulin, A.J., Allee, G.L., 1991. Effect of energy sourcefed to sows during late gestation on neonatal blood metabolite homeostasis, energy stores and com-position. J. Anim. Sci. 69, 230–236.

Nicot, C., Hegardt, F.G., Woldegiorgis, G., Haro, D., Marrero, P.F., 2001. Pig liver carnitine palmitoyl-transferase I, with low Km for carnitine and high sensitivity to malonyl-CoA inhibition, is a naturalchimera of rat liver and muscle enzymes. Biochemistry 40, 2260–2266.

Noblet, J., Le Dividich, J., 1981. Energy metabolism of the newborn pig during the first 24 h after birth.Biol. Neonate 40, 175–182.

Odle, J., 1997. New insights into the utilization of medium-chain triglycerides by the neonate: observa-tions from a piglet model. J. Nutr. 127, 1061–1067.

Odle, J., Lin, X., van Kempen, T.A.T.G., Drackley, J.K., Adams, S.H., 1995. Carnitine palmitoyltrans-ferase modulation of hepatic fatty acid metabolism and radio-HPLC evidence for low ketogenesis inneonatal pigs. J. Nutr. 125, 2541–2549.

Pégorier, J.P., Duée, P.H., Simoes-Nunes, C., Peret, J. Girard, J., 1984. Glucose turnover and recycling inunrestrained and unaesthetized 48-h-old fasting or post-absorptive newborn pigs. Brit. J. Nutr. 52,277–287.

Pégorier, J.P., Simoes-Nunes, C., Duée, P.H.,. Peret, J. Girard, J., 1985. Effect of intragastric triglyceridesadministration on glucose homeostasis in newborn pigs. Amer. J. Physiol. 249, E268–E275.

Père, M.C., 1995. Maternal and fetal blood levels of glucose, lactate, fructose and insulin in the consciouspig. J. Anim. Sci. 73, 2994–2999.

Persson, B., Gentz, J.C.H., Hakkarainen, J., Kellum, M., 1971. Catecholamine-induced lipolysis and itsrelation to oxygen consumption in the newborn piglet. Pediat. Res. 5, 435–445.

Quiniou, N., Dagorn, J., Gaudré, D., 2002. Variation of piglets’ birth weight and consequences on sub-sequent performance. Livest. Prod. Sci., Special Issue 78, 63–70.

Randall, G.C., 1983. Changes in the concentrations of cortocosteroids in the blood of fetal pigs and theirdams during late gestation and labor. Biol. Reprod. 29, 1077–1084.

Ricquier, D., Bouillaud, F., 2000. Mitochondrial uncoupling proteins: from mitochondria to the regula-tion of energy balance. J. Physiol. 529, 3–10.

Sarkar, N.K., Kramer, J.K.G., Wolynetz, M.S., Elliot, J.I., 1985. Influence of dietary fat on growth andbody composition of the neonatal pig. Can. J. Anim. Sci. 65, 175–184.

Scanes, C.G., Lazarus, D., Bowen, S., Buonomo, F.C., Gilbreath, R.L., 1987. Postnatal changes in cir-culating concentrations of growth hormone, somatomedin C and thyroid hormones in pigs. Domest.Anim. Endocrinol. 4, 252–257.

Schmidt, I., Herpin, P., 1998. Carnitine palmitoyltransferase I (CPT I) activity and its regulation by mal-onyl-CoA are modulated by age and cold exposure in skeletal muscle mitochondria from newbornpigs. J. Nutr. 128, 886–893.

Schmidt, I., Herpin, P., 1997. Postnatal changes in mitochondrial protein mass and respiration in skeletalmuscle from the newborn pig. Comp. Biochem. Physiol. B118 639–647.

Schnoebelen-Combes, S., Louveau, I., Postel-Vinay, M.-C., Bonneau, M., 1996. Ontogeny of GH recep-tor and GH-binding protein in the pig. J. Endocrinol. 148, 249–255.

Slebodzinski, A.B., 1979. Metabolic responses to thyroxine in the newborn pig. Biol. Neonate 36, 198–205.Slebodzinski, A.B., Nowak, G., Zamyslowska, H., 1981. Sequential observation of changes in thyroxine,

triiodothyronine and reverse triiodothyronine during the postnatal adaptation of the pig. Biol. Neonate39, 191–199.

Steffen, D.G., Brown, L.J., Mersmann, H.J., 1978. Ontogenic development of swine (Sus domesticus)adipose tissue lipases. Comp. Biochem. Physiol. B 59, 195–198.

Trayhurn, P., Temple, N.J., Van Aerde, J., 1989. Evidence from immunoblotting studies on uncoupling proteinthat brown adipose tissue is not present in the domestic pig. Can. J. Physiol. Pharmacol. 67, 1480–1485.

Thulin, A.J., Allee, G.L., Harmon, D.L., Davis, D.L., 1989. Utero-placental transfer of octanoic, palmiticand linoleic acids during late gestation in gilts. J. Anim. Sci. 67, 738–745.

Wrutniak-Cabello, C., Casas, F., Cabello, G., 2001. Thyroid hormone action in mitochondria. J. Mol.Endocrinol. 26, 67–77.

Yu, X.X., Drackley, J.K., Odle, J., 1997. Rates of mitochondrial and peroxisomal β-oxidation of palmi-tate change during postnatal development and food deprivation in liver, kidney and heart of pigs. J. Nutr. 127, 1814–1821.

P. Herpin et al.374

375

15 Hepatic gluconeogenesis in developingruminants

S. S. Donkina and H. Hammonb

aDepartment of Animal Sciences, Purdue University, West Lafayette, IN 47907, USAbResearch Institute for Biology of Farm Animals Nutrition Physiology(Oskar Kellner Institute), 18196 Dummerstorf, Germany.

The transition from preruminating to ruminating status represents one of the most dramaticchanges in glucose metabolism in mammals. Within 5 weeks of birth, ruminants mustundergo the anatomical and physiological adaptations necessary to permit extensive fermen-tation of plant materials in the rumen and postabsorptive utilization of the end-products.Several well-characterized metabolic adaptations have been documented that act to spare glucoseoxidation with the onset of rumination; however, the endocrine and molecular factors thatmodulate changes in glucose synthesis and metabolism during this transition are not yet fullycharacterized. This review focuses on the endocrine and metabolic state of the ruminant fetusat term, the development of metabolic competence in the neonatal ruminant, and changes thatoccur during the transition to ruminating status.

1. GLUCONEOGENIC SUBSTRATES AND METABOLISM

Propionate, lactate, and amino acids furnish most of the carbon used for gluconeogenesis infed ruminants and glycerol provides some gluconeogenic carbon during feed restriction(Huntington, 1990). In neonatal and developing ruminants, milk lactose supplies approxi-mately 25% of the daily glucose needs (Girard, 1990). In the absence of a functional rumen,amino acids, lactate, and to a limited extent, glycerol from milk are used for gluconeogenesis.Development of the fermentation capacity of the rumen is accompanied by changes in thetype of carbohydrates ingested, reductions in the amount of fat in the diet, a decrease in avail-ability of dietary carbohydrate to the developing ruminant, and an increased supply ofpropionate as a gluconeogenic precursor. In both the preruminant and ruminant states theneed for active gluconeogenesis to maintain glucose homeostasis is apparent.

Pyruvate is a common entry point in the gluconeogenic pathway for lactate, alanine, andother gluconeogenic amino acids. Pyruvate formed from lactate and amino acids is transportedinto the mitochondria and carboxylated to oxaloacetate by pyruvate carboxylase (PC) (fig. 1).



Fig

. 1.

Met

abol

ism

of

prop

iona

te, l

acta

te, p

yruv

ate,

and

ala

nine

to g

luco

se in

bov

ine

liver

. Sub

stra

te a

bbre

viat

ions

are

giv

en in

the

key.

Rea

ctio

ns c

atal

yzed

by

the

key

enzy

mes

dis

cuss

edin

the

text

, cyt

osol

pho

spho

enol

pyru

vate

car

boxy

kina

se (

PEPC

K-C

), m

itoch

ondr

ial p

hosp

hoen

olpy

ruva

te c

arbo

xyki

nase

(PE

PCK

-M),

pyr

uvat

e ca

rbox

ylas

e (P

C),

and

pyr

uvat

e ki

nase

(PK

),ar

e in

dica

ted

by th

e sh

aded

bac

kgro

unds

.

Propionate, in contrast, is metabolized through part of the TCA cycle to oxaloacetate follow-ing activation to propionyl-CoA and metabolism through propionyl-CoA carboxylase,methylmalonyl-CoA racemase, and methylmalonyl-CoA mutase. Oxaloacetate can be metab-olized to phosphoenolpyruvate (PEP) by phosphoenolpyruvate carboxykinase (PEPCK) ormetabolized in the TCA cycle. In turn, PEP carbon can be metabolized to glucose or recycledto pyruvate via pyruvate kinase (PK). In order for lactate carbon to be metabolized to glucose,the flux through PEPCK and PC must exceed the PK flux, whereas net flux of propionatecarbon requires only a greater flux through PEPCK relative to PK. Therefore, an increase inPEPCK activity in the absence of changes in PC activity would favor the use of propionatefor gluconeogenesis.

The presence of PEPCK activity in the cytosol (PEPCK-C) and mitochondria (PEPCK-M)is one of the important features of gluconeogenesis that permits compartmentalization of thepathway and results in the characteristic pattern of regulation and use of lactate and pyruvate.The distribution of this activity is uniquely species-dependent and most mammals displayboth a mitochondrial and a cytosolic form of the PEPCK enzyme. Rodents express primarilyPEPCK-C and both forms are found in liver of the developing chicken, yet only the mitochon-drial form is found in liver from the adult chicken. There are approximately equal activitiesof PEPCK-M and PEPCK-C in the ruminant (Taylor et al., 1971) and human (Hod et al.,1987) liver. Bovine PEPCK-C and PEPCK-M have been recently cloned and characterized(Agca et al., 2002) and the ratio of mRNA indicates a 10-fold greater expression of PEPCK-Cthan PEPCK-M in lactating cows. Similar data are not yet available for developing bovine.

The stoichiometry of gluconeogenesis dictates that the formation of phosphoenolpyruvatefrom propionate, pyruvate, and some amino acids requires the independent synthesis ofNADH in the cytosol for the subsequent reduction of 1,3-diphosphoglycerate in gluconeo-genesis. It has been proposed that PEPCK-C is required for gluconeogenesis from aminoacids and PEPCK-M is more suited to gluconeogenesis from lactate (Watford et al., 1981).Pyruvate and amino acids are metabolized to oxaloacetate in mitochondria and are shuttled tothe cytosol as malate from which NADH and oxaloacetate are regenerated followed by PEPformation that is catalyzed by PEPCK-C. Lactate can also be metabolized to PEP in mito-chondria of species that possess appreciable PEPCK-M activity and subsequently shuttled tothe cytosol (Holcomb et al., 1995).

2. GLUCOSE RELEASE FROM HEPATOCYTES

Glucose-6-phosphatase is a membrane-bound enzyme that is located on the internal mem-brane of the endoplasmic reticulum and is involved in the terminal step of gluconeogenesisas well as glycogenolysis. The enzyme catalyzes the conversion of glucose-6-phosphate toglucose to enable release from the cell. In nonruminants the enzyme is expressed in liver,kidney cortex, and jejunum, but only the liver form of the enzyme is upregulated at birth andto weaning in rodents (Chatelain et al., 1998; Kalhan and Parimi, 2000). The hepatocyte glucosetransporter, GLUT2, and glucose-6-phosphatase act in concert to control the release of glucosefrom liver. The symmetry of GLUT2 enables the transport of glucose into or out of the hepato-cyte and the directionality depends only on the concentration differential between intracellularfree glucose and blood glucose (Burchell, 1994).

Glucose-6-phosphate (G-6-P), formed through gluconeogenesis or glycogenolysis, must bedephosphorylated through the action of glucose-6-phosphatase (G-6-Pase), an enzyme that iscontained within the endoplasmic reticulum in order to release glucose from the hepatocyte.An endoplasmic glucose transporter GLUT7 was initially proposed that would facilitate the

Hepatic gluconeogenesis in developing ruminants 377

transport of G-6-P to the endoplasmic reticulum where G-6-Pase acts to release free glucoseinto the cytoplasm (Burchell, 1994), but has since been retracted (Burchell, 1998). It is nowthought that G-6-Pase acts in combination with a specific G-6-P translocase to channel G-6-Pinto the endoplasmic reticulum where G-6-Pase is compartmentalized (Van Schaftingen andGerin, 2002). Recently a second transporter protein for G-6-P has been identified (Hosokawaand Thorens, 2002) which complements the activity of the specific G-6-P translocase.

When G-6-P is overexpressed in hepatocytes there is a marked increase in glucose releaseand a decline in intracellular G-6-P and glycogen concentrations (Seoane et al., 1997; Aistonet al., 1999). Measures of G-6-Pase activity at term indicate that the capacity of the enzymeis fully developed at birth in ruminants (Edwards et al., 1975; Stevenson et al., 1976;Narkewicz et al., 1993). Unfortunately data are not yet available for ruminants describingdevelopmental changes in expression of G-6-P translocases.

3. GLUCONEOGENESIS IN FETAL RUMINANTS

The contribution of fetal gluconeogenesis to the glucose needs of the developing ruminantconceptus is equivocal. A portion of the discrepancies regarding the contribution of gluco-neogenesis to the glucose economy of the developing fetus lies in recycling errors that areinherent to measuring glucose entry rates using isotope dilution (Kalhan and Parimi, 2000).Available data indicate that uterine glucose requirements during the last trimester of preg-nancy account for 20–70% of the glucose needs of pregnant ewes (Prior, 1982), that fetalglucose uptake is reduced when ewes are deprived of feed (Tsoulos et al., 1971; Boyd et al.,1973; Chandler et al., 1985; Leury et al., 1990), and that the rate of fetal glucose utilizationduring maternal feed restriction is less affected than glucose removed via the umbilical artery(Hay et al., 1984). These observations imply that the hypoglycemic ovine fetus is capable ofsignificant endogenous glucose release and is subject to activation in utero in response tomaternal nutrition.

Amino acids and lactate are the major gluconeogenic substrates in the fetus, and urea excre-tion rates indicate that 25% of oxygen consumption by fetal ovine liver is due to amino acidcatabolism (Gresham, 1972). Efficient extraction of propionate by maternal liver precludesappreciable propionate supply to the developing ruminant fetus for gluconeogenesis. In someexperiments, gluconeogenesis from lactate accounts for 22% of lactate turnover and supplies49% of fetal glucose (Prior, 1980), but in other experiments fetal gluconeogenesis fromlactate was undetectable (Warnes et al., 1977). These contradictory observations reflect vari-ations in maternal nutrition immediately prior to the experimental period (Girard et al., 1992).Experimental evidence suggests that the rate of gluconeogenesis is increased in fetal lambliver in response to inadequate nutrition of the dam (Leury et al., 1990; Apatu and Barnes,1991a). Increased fetal urea production during nutritional insufficiency in pregnant ewes isconsistent with an increase in fetal gluconeogenesis from amino acids (Hodgson et al., 1982).Therefore gluconeogenesis is active and adaptable in fetal ruminants, unlike rodents, andplays a critical role in the glucose economy of the maternal–fetal unit.

Enzyme activities have been used to characterize developmental changes in liver metabo-lism and provide an estimate of the maximum flux through a single step in the gluconeogenicpathway. The activities of glucose-6-phosphatase, fructose-1,6,-bisphosphatase, pyruvatecarboxylase, and PEPCK, key enzymes for gluconeogenesis, are similar in term-fetal, neonatal,and adult sheep (Warnes et al., 1977; Narkewicz et al., 1993). There appears to be sequentialdevelopment of gluconeogenic enzymes in caprine (Dhanotiya and Bhardwaj, 1988) and ovinefetuses (Stevenson et al., 1976). However, the enzymes of the gluconeogenic path are present

S. S. Donkin and H. Hammon378

in fetal liver by 100−128 days of gestation (Warnes et al., 1977; Prior, 1980). There is greateractivity of alanine aminotransferase in fetal than in neonatal liver, which may reflect a greatercapacity for amino acid metabolism to glucose in utero. Changes in hepatic enzyme activityduring the period of rumen development are modest and changes in glucose metabolismreflect decreased glycolytic activity in both muscle and liver (Howarth et al., 1968; Pearceand Unsworth, 1980). The activity of lactate dehydrogenase and alanine aminotransferase arelower in adult ewes compared to neonatal sheep or 3-month-old lambs (Edwards et al., 1975),suggesting decreased capacity to metabolize alanine to glucose with development in neonatalruminants.

4. NUTRITIONAL CHANGES AT BIRTH

At birth the neonate must cope with the loss of umbilical glucose supply and survive a briefperiod of starvation before receiving colostrum and milk. Liver glycogen, at birth, is approx-imately 4–6% of liver wet weight. This energy reserve is depleted within a few hours(Hamada and Matsumoto, 1984; Girard, 1990) and supplies glucose for erythrocytes, brain,and kidney medulla. Glucose supplied by milk lactose accounts for approximately 25% ofglucose utilization of the neonatal lamb (Girard, 1986), therefore gluconeogenesis is necessaryto maintain neonatal glucose homeostasis.

Development of supporting pathways, production of cofactors, and substrate supply mayaffect the rates of gluconeogenesis in utero and during postnatal development. For examplethe inability to oxidize fatty acids at birth has been characterized in detail and stems from alack of activity of fatty acyl-CoA synthases, carnitine palmitoyltransferase I (CPT-I), enoyl-CoA hydratase, 3-hydroxyacyl-CoA dehydrogenase, and oxoacyl-CoA thiolase (Girard et al.,1992). The gluconeogenic promoting effects of fatty acids have been recognized for sometime in nonruminants (Williamson et al., 1966). A similar regulation is likely in the bovinebased on data that indicate that specific long-chain fatty acids promote gluconeogenesis inhepatocytes from ruminating calves (Mashek et al., 2002). Likewise, inhibition of CPT-I activitydecreases gluconeogenesis in sheep hepatocytes (Chow and Jesse, 1992). Information is lackingon initiation of fatty acid oxidation in neonatal ruminants, but if a parallel can be drawn fromrodent data, the induction of gluconeogenic capacity may be linked to the induction of fattyacid oxidation.

5. HORMONAL CHANGES: INSULIN, GLUCAGON,GLUCOCORTICOIDS

Nutrient supply during the prenatal period consists primarily of a carbohydrate-rich energysupply (glucose, lactate), yet during the neonatal period a switch is made to a high-fat, low-carbohydrate diet (Aynsley-Green, 1988). Newborns develop marked hypoglycemia afterbirth because glucose derived from lactose in colostrum does not meet postnatal glucosedemands (Girard, 1986). Therefore, glycogenolysis and gluconeogenesis increase rapidlyin the liver of newborns; however, there are species differences in prenatal development of glu-coneogenesis. In the developing rat and pig fetus the gluconeogenic pathway does not maturein utero (Ballard and Oliver, 1963; Swiatek, 1971), whereas in the bovine fetus there is gluco-neogenic activity measurable from day 80 of gestation (Prior and Scott, 1977). This mightindicate, as discussed above, that the bovine fetus is less dependent on maternal glucosesupply than the rat and pig fetus; however, newborn calves experience hypoglycemia as doother species (Aynsley-Green, 1988; Egli and Blum, 1988; Hadorn et al., 1997; Hammon and


Blum, 1998). In neonatal ruminants and nonruminants, glucose is mostly produced by gluco-neogenesis using amino acids (alanine, glycine, glutamine), glycerol, and lactate. In humannewborns, research using stable isotopes demonstrates that gluconeogenesis from lactate,glycerol, and alanine occurs at a significant rate within the first 8 h of life and is critical forneonatal survival (Ferre et al., 1986).

Insulin and glucagon are integral to normal fetal development in ruminants (reviewed inBlum and Hammon, 1999a). The change in glucose supply that accompanies loss of umbili-cal nutrient supply is reflected by an increase in glucagon and decreased insulin concentrationduring the immediate postnatal period (Girard, 1990). Blood profiles in newborn calves arecharacterized by hypoglycemia, high nonesterified fatty acids, and low triglyceride, phospho-lipids, and cholesterol (Blum and Hammon, 1999b). Neonatal calves respond to nutritionalchallenges by increasing glucagon and decreasing insulin in a manner similar to adult animals;however, the glucose–insulin relationship is less developed in neonates. The lack of glucoseclearance in response to insulin may prevent hypoglycemia and serve to protect the neonatal calf.

Cortisol plays an important role in enhancing fetal capacity for glucose production andglycogen storage (Fowden, 1995; Barnes, 1997). Changes in plasma insulin and glucagonmay be related to the stress associated with birth and the concomitant rise in serum cortisol,fetal hypoxia, or both (Girard, 1990). Key genes for gluconeogenesis are also responsiveto thyroid hormones (Park et al., 1997); a rise in thyroxine during the first 24 h of life inneonatal sheep (Fisher et al., 1977) may play a role in induction of metabolic competence.Recent data indicate that fetal thyroid hormone production is essential to the development ofgluconeogenesis and is especially critical under adverse conditions such as undernutrition(Fowden et al., 2001).

Plasma glucagon concentrations rise during the immediate postnatal period due, in part, toa drop in blood glucose that occurs within a few hours of birth. An infusion of somatostatinin lambs induces hypoglycemia and infusion of glucagon reverses the effects of somatostatin(Sperling et al., 1977). The rate of glucose output by fetal, neonatal, and adult ovine liver wasincreased similarly during glucagon infusions (Apatu and Barnes, 1991b); however, the effectivedose of glucagon necessary to stimulate gluconeogenesis is greater in fetal liver (Girard andSperling, 1983). Postnatal increases in glucagon receptor numbers and full development ofintracellular signal transduction pathways along with a decrease in insulin receptor numbersfavor regulation of gluconeogenesis in the neonate that is more sensitive to changes in glucagonconcentrations (Girard and Sperling, 1983).

Hepatic glucagon receptor numbers are lower in fetal and newborn ruminants than inadults. The effective dose of glucagon required to stimulate gluconeogenesis in adult sheepliver is not effective in fetal sheep liver (Girard and Sperling, 1983). In 21-day-old rats thenumber of liver glucagon receptors was only 40% of the receptor number for adult liver(Ganguli et al., 1983). Insulin receptor number and affinity are higher in fetal than in adultliver in rats and humans (Neufeld et al., 1980). High insulin and low glucagon receptor activityin utero favors glucose oxidation, whereas the coupling of glucagon receptor to cAMP synthesiscombined with an increase in glucagon receptor numbers in early postnatal life favorsgluconeogenesis (Girard and Sperling, 1983).

6. GLUCONEOGENESIS IN NEONATALAND DEVELOPING RUMINANTS

Gluconeogenesis from lactate is similar between fetal and maternal liver in the bovine (Priorand Scott, 1977). The rates of [2-14C]propionate and [2-14C]lactate incorporation to glucose


and glycogen obtained from liver slices from adult sheep and newborn lambs are similar(Ballard and Oliver, 1963). In contrast, the rate of gluconeogenesis from [2-14C]pyruvate isgreater for liver slices from neonatal lambs compared with adult tissue. The rate of pyruvatemetabolism to glucose appears to peak at about 2–4 weeks of age in lambs (Ballard andOliver, 1963). Likewise the rates of metabolism of lactate are markedly reduced in weanedlambs (Savan et al., 1986) and calves (Donkin and Armentano, 1995).

In developing ruminants there is a marked decline in the capacity to metabolize lactate toglucose coupled with a reduced sensitivity to the effects of insulin (Donkin and Armentano,1995). Radioisotope tracer data indicate that there is almost exclusive flux of lactate throughpyruvate carboxylase (PC) to glucose in neonatal calf liver and very little isotope exchangewith carbon of the TCA cycle (Donkin and Armentano, 1994). The substantial loss in lactatemetabolism to glucose during the preruminant to ruminant transition (Donkin andArmentano, 1995), and similar use of propionate for gluconeogenesis between the two groups(fig. 2), suggests a loss in capacity to draw lactate into the gluconeogenic pathway. These resultsare perplexing in light of the extensive use of lactate for gluconeogenesis in nonruminants,but agree with the 10-fold lower rate of glucose recycling in vivo in adult versus neonatal (5- or21-day old) sheep (Muramatsu et al., 1974). These changes suggest developmentally regulateddifferences in gluconeogenesis that are unique to lactate.

Lactate is equilibrated rapidly with pyruvate in liver. The rates of [1-14C]lactate and[1-14C]pyruvate metabolism to glucose are not different for hepatocytes obtained from preru-minating calves (Donkin and Armentano, 1994). This measurement has not been madedirectly in hepatocytes from ruminating calves, but is not likely the limiting step in gluco-neogenesis from lactate based on a lack of control of gluconeogenesis in response toalterations in cytosolic redox state (Aiello and Armentano, 1987). As indicated above, pyru-vate formed from lactate is carboxylated to oxaloacetate by pyruvate carboxylase (PC) andoxaloacetate is either metabolized to phosphoenolpyruvate (PEP) or metabolized in the TCAcycle. The similar ratios of [14C]glucose:14CO2 from [2-14C]propionate and carbons 2 and 3 oflactate support a common oxaloacetate pool for the metabolism of propionate and lactatein bovine hepatocytes (Donkin and Armentano, 1994). Therefore it is likely that decreased


Fig. 2. Effect of developmental state on gluconeogenesis from propionate and lactate in calves. Hepatocyteswere isolated from preruminating (n = 4) and ruminating calves (n = 3) and cultured for 48 h. The rate of glu-coneogenesis from [2-14C]propionate or [U-14C]lactate was determined during the last 3 h of incubation.Adapted from Donkin and Armentano (1995).

flux through PC is the cause of decreased gluconeogenesis from lactate during postnataldevelopment in calves.

Chronic exposure of neonatal bovine hepatocytes to insulin results in decreased gluconeo-genesis from lactate (Donkin et al., 1997), which is consistent with data suggesting that aportion of the reduction in gluconeogenesis from lactate in milk-fed calves may be due tochronically elevated insulin concentrations (Breier et al., 1988). However, the direct actionsof insulin do not fully explain the reduction in gluconeogenesis from lactate that is observedin calves during the transition from the preruminant to the ruminant state. Comparing the rate ofgluconeogenesis from lactate relative to propionate metabolism suggests additional changesin hepatic lactate metabolism. Gluconeogenesis from lactate is reduced to 32% of the rate ofpropionate conversion to glucose following chronic exposure to insulin (Donkin et al., 1997),but the developmental change in lactate metabolism reduces gluconeogenesis from lactate toonly 10% of the rate of gluconeogenesis from propionate (Donkin and Armentano, 1995).

The data described above point to PC as a primary control point for gluconeogenesis indeveloping ruminants and is supported by data from adult ruminants suggesting that PC maybe a control point for gluconeogenesis. In cattle and sheep the activity of PC is responsive tonutritional and physiological states that impose the greatest demands for endogenous glucoseproduction such as lactation and feed deprivation (Greenfield et al., 2000; Velez and Donkin,2000). In contrast, the activity of PEPCK is relatively invariant between different nutritionaland physiological states in ruminants, diabetes being the exception (Filsell et al., 1969; Tayloret al., 1971). When both PEPCK and PC activity are examined in response to physiologicalstate or nutrient supply, the ratio of their activities suggests that an increase in capacity forlactate metabolism is primarily responsible for increases in hepatic gluconeogenesis.

The dramatic reduction in basal rate of gluconeogenesis from lactate appears to be a dueto a reduction in PC activity and gene expression. Data, from sheep, examining the relationshipbetween prenatal development of gluconeogenic enzymes and activities found in maternalliver fail to reveal any striking differences in activity of PC, PEPCK, or PK (Edwards et al.,1975; Stevenson et al., 1976). Analysis of PC mRNA in liver biopsy samples from 7 through84 days of age indicates a decline in expression of PC mRNA (Donkin et al., 1998) andsuggests a decrease in capacity for lactate metabolism. A decline in PC mRNA expressionwas observed in both milk-fed calves and ruminanting calves by 84 days of age that mirrorsa reduction in gluconeogenesis from lactate (Donkin et al., 1998). Taken together, these datasuggest a developmental decrease in PC expression that is likely reflected as a decrease in lactaterecycling (Muramatsu et al., 1974) and reduced lactate metabolism to glucose in the weanedcalf (Donkin and Armentano, 1995). Data from lactating cows indicate that PC activity andmRNA expression are induced when demands for gluconeogenesis are elevated at calving(Greenfield et al., 2000) and during restricted feed intake (Velez and Donkin, 2000).

The onset of rumen development is marked by the production and absorption of volatilefatty acids (VFA). Acetate and propionate form the bulk of VFA produced by rumen fermen-tation. Acute exposure to propionate decreases gluconeogenesis from lactate equally inhepatocytes from preruminating and ruminating calves (Donkin and Armentano, 1995).Prolonged exposure of hepatocytes from preruminating calves to valerate (which can bemetabolized to acetate and propionate) had no effect on subsequent capacity for gluconeo-genesis from propionate (Donkin and Armentano, 1993). However, an intermediate of propionatemetabolism, methyl malonyl-CoA, can directly inhibit lactate metabolism (Blair et al., 1973) andis thought to be responsible for the acute effects of propionate in limiting gluconeogenesis fromlactate in bovine hepatocytes (Donkin and Armentano, 1994). At present the nature of thedevelopmental suppression of PC activity and gene expression is unknown.


Long-term regulation of gluconeogenesis in nonruminants has been characterized bychanges in the expression of genes encoding glucoregulatory enzymes (Pilkis and Claus,1991). It is well established that insulin represses and glucagon (or cAMP) and glucocorti-coids induce the activity of the PEPCK enzyme by directly regulating expression of thePEPCK gene (reviewed in O’Brien and Granner, 1990). From control strength studies in rats,gluconeogenesis from lactate is distributed between pyruvate kinase and the reactions involvingPC and PEPCK (Sistare and Haynes, 1985). Glucocorticoids have little effect on flux through thePK-catalyzed reaction; therefore an increase in gluconeogenesis from lactate in glucocorticoid-treated rats or hepatocytes is mainly due to the combined increases in flux through reactionscatalyzed by PC and PEPCK (Jones et al., 1993).

7. GLUCOCORTICOIDS AND POSTNATAL DEVELOPMENT

Uncomplicated neonatal growth depends on maturation of vital organs and critical metabolicpathways including lung and cardiac development, thyroid axis, somatotropic axis, initiationof thermogenesis, and control of glucose homoeostasis. In altricial species, such as rodents,a rise in cortisol at birth is necessary to initiate neonatal maturation of many of the criticalmetabolic pathways including gluconeogenesis (Dalle et al., 1985; Gluckman et al., 1999). Inruminants, the concentration of fetal cortisol usually exceeds maternal cortisol concentra-tions; therefore caution should be exercised when extending data on the effects ofglucocorticoids from rodent studies to the biology of liver metabolism in neonatal ruminants.

The central role of glucocorticoids in regulation of expression of PEPCK, G-6-P, PC, andCPT-I (Jitrapakdee and Wallace, 1999; Van Schaftingen and Gerin, 2002) is established andthere are indications in nonruminants that these effects are mediated through peroxisomeproliferator-activated receptor γ coactivator-1 (Louet et al., 2002). Cortisol injected intodeveloping sheep fetuses induced activity of hepatic G-6-Pase, fructose-6-phosphatase, PC,and PEPCK by 2- to 3-fold (Fowden et al., 1993). Glucocorticoids may also play a more gen-eral role in switching the fetal physiological state to a postnatal state (Liggins, 1977; Fowden,1995). For example, gastrointestinal tract developmental, gastrin secretion, and intestinalabsorption of immunoglobulins are stimulated by cortisol in neonatal piglets and play a rolein maturation of the fetal exocrine pancreas of pigs and lambs (Sangild, 2001).

Glucocorticoids are important regulators of the glucose status after birth in the immatureneonatal calf. Cortisol concentrations decreased after birth in neonates (Baumrucker andBlum, 1994; Hadorn et al., 1997; Hammon and Blum, 1998). Importantly, plasma cortisolconcentrations depend on the level and source of nourishment (milk or colostrum) after birth(Hammon and Blum, 1998). Calves fed milk replacer from birth were characterized by higherplasma cortisol concentrations and lower plasma glucagon concentrations than calves fedcolostrum (Hammon and Blum, 1998).

The prepartum cortisol surge may play an important role in initiating the perinatal switch ofthe somatotropic axis from the fetal to the postnatal status and function (Breier et al., 2000).Glucocorticoids stimulate gluconeogenesis in vivo by increasing plasma glucagon concentrationsas well as augmenting the effects of glucagon to stimulate gluconeogenesis (Marco et al., 1973;Wise et al., 1973; Lecavalier et al., 1990). Furthermore, glucocorticoid treatment induces insulinresistance in late gestation in sheep (Challis et al., 2001) and postnatally in humans (Weinsteinet al., 1995; Dirlewanger et al., 2000) and in dairy cows (Maciel et al., 2001). The interaction ofglucocorticoids and growth hormone has not been fully characterized for neonatal ruminants, butpostnatal growth is characterized by changes from a substrate-limited prenatal growth to enteralfeeding with the somatotropic axis becoming the dominant endocrine regulatory system.


Glucocorticoids enhance the maturation of the somatotropic axis and the prepartum corti-sol surge may play an important role in initiating the perinatal switch of the somatotropic axisfrom the fetal to the postnatal status and function (Breier et al., 2000; Carroll et al., 2000).Cortisol acts to stimulates hepatic growth hormone receptor (GHR) numbers and IGF-I mRNAlevels in the sheep fetus (Li et al., 1996). In vivo studies using porcine hepatocytes indicatethat IGF-I mRNA expression is more responsive to GH in the presence of dexamethasone andthyroxine (Brameld et al., 1999). Therefore, it might be speculated that in precocious speciessuch as ruminants, elevated cortisol levels at birth serve to enhance postnatal maturation ofthe somatotropic axis. The somatotropic axis in neonatal calves is functional, but immature atbirth (Hammon and Blum, 1997) owing partly to reduced GH-binding capacity of the liver inneonatal calves (Breier et al., 1994). Little is known about the ontogeny of the growth hormonereceptor in the neonatal bovine or its coordination with other hepatic functions including glucosemetabolism. Growth hormone as well as the GH receptor are present in the bovine fetus butgrowth hormone does not affect IGF-I production in the liver (Gluckman et al., 1999), perhapsowing to GHR numbers, activity of receptors, or both (Fowden, 1995; Freemark, 1999).

8. GLUCONEOGENESIS AND REGULATION OF GENE EXPRESSION

Most of the enzymes for gluconeogenesis, including PC, PEPCK-M, fructose 1,6-bisphosphatase,and G-6-Pase, have substantial activity in near-term fetuses of ruminants (Edwards et al.,1975; Stevenson et al., 1976; Narkewicz et al., 1993). The classic work of Ballard and Hanson(1967) established PEPCK-C as the limiting step in development of gluconeogenesis in rats.These data have been substantiated for rabbit and other species (Girard et al., 1992), but thereis no limitation in development of PEPCK-C activity in ruminant liver (Edwards et al., 1975;Stevenson et al., 1976; Narkewicz et al., 1993). In rodents the rapid increase in PEPCK-C islinked to the process of birth rather than fetal age (Girard et al., 1992) and is related to the latefetal appearance of developmentally regulated transcription factors such as CCAAT/enhancer-binding protein (Cassuto et al., 1999). Therefore it would follow that these tran-scription factors or their functional homologs are likely to be present in utero in liver of thedeveloping ruminants.

The expression of PC is tissue-specific with the highest catalytic activity of the enzymefound in liver, kidney, adipose tissue, brain, adrenal gland, and lactating mammary tissue.Changes in PC abundance, through alteration in rate of synthesis, constitute long-term regula-tion of pyruvate metabolism for gluconeogenesis and lipogenesis (Barritt, 1985). Short-termallosteric regulation of PC activity by acetyl-CoA is well noted; however, sustained changes inthe activity of the PC enzyme require parallel increases in PC mRNA (Zhang et al., 1993).

Northern analysis of total RNA indicates the presence of a single 4.2 kb mRNA for rat andhuman PC that is the product of a single copy gene (Jitrapakdee et al., 1996). However, selec-tive amplification of the 5′ untranslated region (UTR) of PC cDNA indicates the presence offive alternative forms of PC cDNA that are generated through differential splicing of RNAtranscripts and use of two tissue-specific promoters (Jitrapakdee et al., 1996). Transcriptsgenerated from the proximal promoter are restricted to gluconeogenic and lipogenic tissuewhereas those generated from the distal promoter are expressed in several tissues. These5′ UTR isomers of the PC primary transcript share the same open reading frame and result inone PC protein.

The liver expresses the C, D, and E forms of PC transcript, although the C and D formspredominate. During the suckling to weaning transition the abundance of the C isoformdecreases as does PC mRNA and enzyme activity (Jitrapakdee et al., 1998). The fact that


the C transcript is more functionally potent (2× greater than D) in translation reactionssuggests that a small increase in the C transcript may result in proportionately greater increasein PC activity than changes in the D form. Furthermore an increase in C transcript and anoffsetting decrease in the D transcript would result in no net change in PC mRNA abundanceby Northern analysis but would increase PC translation and maximal PC activity.

The entire coding sequence of bovine PC has been cloned (Agca et al., 2000) and thecoding sequence contains 3075 bases with 85% identity to human PC. Furthermore, bovinePC is expressed as six 5′ UTR variants of different lengths (Agca and Donkin, 2001).Experiments are ongoing to test the functional significance of bovine PC variants relative togluconeogenesis and neonatal development in cattle.

Regulation of PEPCK expression has been extensively studied in liver of rodents as wellas rat and human hepatoma cell lines. Glucagon acting in the presence of dexamethasone isone of the primary stimulators of PEPCK gene expression. The activity of PEPCK-C is deter-mined by the rate of transcription of the PEPCK-C gene and the rate of turnover of its mRNAwhereas the activity of PEPCK-M appears to be constitutive (Hanson and Reshef, 1997). Thecoding sequence for bovine PEPCK-C and a fragment of bovine PEPCK-M have been clonedrecently (Agca et al., 2002). Unlike PEPCK-C the expression of PEPCK-M mRNA is notresponsive to changes in physiological state (Greenfield et al., 2000; Agca et al., 2002).

Control of PEPCK-C activity is largely exerted through transcription of the gene throughactivation of basal, tissue-specific, and hormone-dependent promoter elements within the5′ region of the PEPCK-C gene (Hanson and Reshef, 1997). Crucial liver control elementsare located within −460 to +73 of the promoter and six primary protein-binding sites havebeen characterized by DNAse I footprinting; these six sites contain docking sites for at least15 separate transcription factors (Hanson and Reshef, 1997). The cAMP response element I(CRE-I) acting synergistically with protein-binding sites 3 and 4 is primarily responsible forthe cAMP-mediated increase in PEPCK-C transcription (Hanson and Reshef, 1997). Insulincounteracts the effects of cAMP by repressing the promoter, perhaps by blocking the abilityof glucocorticoids to promote activity of accessory factor-2 (O’Brien and Granner, 1990).Although the PEPCK-C gene is generally thought to be transcriptionally controlled, there isregulation through stability of the PEPCK-C mRNA which is mediated through cAMP actionon a 3′ noncoding sequence (Lemaigre and Rousseau, 1994). There is some indication thatPEPCK-C expression may be inhibited directly by glucose as is the case with other insulin-responsive genes. The lack of appreciable glucokinase activity in ruminant liver leads toquestions regarding a similar control in ruminants.


There is no question that gluconeogenesis is critical to the survival and normal developmentof fetal, neonatal, and postnatal ruminants. The precocious development of gluconeogenicmachinery in the liver of the ruminant fetus provides a number of advantages for survival atbirth. There are many developmental aspects of gluconeogenesis that have been described indetail for nonruminants that are applicable to the developing ruminant, but several processesare species-specific. Information is lacking on the basic biology that accompanies the onsetof metabolic competence in the developing ruminant, including processes that may modulategluconeogenesis, and in many cases parallels must be drawn from rodent models. Issues asso-ciated with the initiation of expression of key genes for gluconeogenesis including PC andPEPCK in ruminants and the molecular cues that initiate development of gluconeogenesisremain to be clarified. Several aspects of gluconeogenesis in developing ruminants have been


identified, but a more complete characterization of fetal and neonatal gluconeogenesis isneeded to identify unique regulatory controls including the molecular and biochemical eventsthat accompany the postnatal reduction in gluconeogenesis from lactate. Conversely the bio-chemical anomalies identified for ruminants, such as the inherent lack of hepatic G-6-Paseactivity, could provide unique opportunities to study glucose trafficking in liver in order tobetter understand metabolic diseases of humans.

REFERENCES

Agca, C., Bidwell, C.A., Donkin, S.S., 2000. Cloning of two alternately spliced transcripts of bovinehepatic pyruvate carboxylase. FASEB J. 14, A515.

Agca, C., Donkin, S.S., 2001. Pyruvate carboxylase mRNA 5′UTR variants are heterogeneously expressedin bovine tissues except two variants specific to adipose tissue and liver. J. Dairy Sci. 84, Suppl. 1, 386.

Agca, C., Greenfield, R.B., Hartwell, J.R., Donkin, S.S., 2002. Cloning and characterization of bovinecytosolic and mitochondrial phosphoenolpyruvate carboxykinase and characterization during thetransition to lactation. Physiol. Genomics 11, 53–63.

Aiello, R.J., Armentano, L.E., 1987. Gluconeogenesis in goat hepatocytes is affected by calcium, ammo-nia and other key metabolites but not primarily through cytosolic redox state. Comp. Biochem.Physiol. B. 88, 193–201.

Aiston, S., Trinh, K.Y., Lange, A.J., Newgard, C.B., Agius, L., 1999. Glucose-6-phosphatase overexpres-sion lowers glucose 6-phosphate and inhibits glycogen synthesis and glycolysis in hepatocytes withoutaffecting glucokinase translocation: evidence against feedback inhibition of glucokinase. J. Biol.Chem. 274, 24559–24566.

Apatu, R.S., Barnes, R.J., 1991a. Release of glucose from the liver of fetal and postnatal sheep by portalvein infusion of catecholamines or glucagon. J. Physiol. 436, 449–468.

Apatu, R.S., Barnes, R.J., 1991b. Blood flow to and the metabolism of glucose and lactate by the liverin vivo in fetal, newborn and adult sheep. J. Physiol. 436, 431–447.

Aynsley-Green, A., 1988. Metabolic and endocrine interrelationships in the human fetus and neonate: anoverview of the control of the adaptation to postnatal nutrition, In: Lindblad, B.S. (Ed.), PerinatalNutrition. Academic Press, New York, pp. 161–191.

Ballard, F.J., Hanson, R.W., 1967. Phosphoenolpyruvate carboxykinase and pyruvate carboxylase indeveloping rat liver. Biochem J. 104, 866–871.

Ballard, F.J., Oliver, I.T., 1963. Glycogen metabolism in embryonic chick and neonatal rat liver. Biochim.Biophys. Acta 71, 578–588.

Barnes, R.J., 1997. Perinatal carbohydrate metabolism and the blood flow of the fetal liver. Equine Vet.J., Suppl., 24, 26–31.

Barritt, G.J., 1985. Regulation of enzymatic activity. In: Keech, D.B., Wallace, J.C. (Eds.), PyruvateCarboxylase. CRC Press, Boca Raton, FL, pp. 179–248.

Baumrucker, C.R., Blum, J.W., 1994. Effects of dietary recombinant human insulin-like growth factor onconcentrations of hormones and growth factors in the blood of newborn calves. J. Endocrinol. 140, 15–21.

Blair, J.B., Cook, D.E., Lardy, H.A., 1973. Interaction of propionate and lactate in perfused rat liver.J. Biol. Chem. 248, 3608–3615.

Blum, J.W., Hammon, H., 1999a. Pancreatic hormones (insulin and glucagon) in calves: ontogeneticchanges and nutritional effects. In: Pierzynowski, S.G., Zabielski, R. (Eds.), Biology of the Pancreasin Growing Animals. Elsevier, Amsterdam, pp. 27–43.

Blum, J.W., Hammon, H., 1999b. Endocrine and metabolic aspects in milk-fed calves. Domest. Anim.Endocrinol. 17, 219–230.

Boyd, R.D., Morris, F.H. Jr., Meschia, G., Makowski, E.L., Battaglia, F.C., 1973. Growth of glucose andoxygen uptakes by fetuses of fed and starved ewes. Amer. J. Physiol. 225, 897–902.

Brameld, J.M., Gilmour, R.S., Buttery, P.J., 1999. Glucose and amino acids interact with hormones tocontrol expression of insulin-like growth factor-I and growth hormone receptor mRNA in cultured pighepatocytes. J. Nutr. 129, 1298–1306.

Breier, B.H., Ambler, G.R., Sauerwein, H., Surus, A., Gluckman, P.D., 1994. The induction of hepaticsomatotrophic receptors after birth in sheep is dependent on parturition-associated mechanisms.J. Endocrinol. 141, 101–108.


Breier, B.H., Gluckman, P.D., Bass, J.J., 1988. Plasma concentrations of insulin-like growth factor-I andinsulin in the infant calf: ontogeny and influence of altered nutrition. J. Endocrinol. 119, 43–50.

Breier, B.H., Oliver, M.H., Gallaher, B.W., 2000. Regulation of growth and metabolism during postnataldevelopment. In: Cronjé, P.B. (Ed.), Ruminant Physiology: Digestion, Metabolism, Growth andReproduction. CABI Publishing, New York, pp. 187–204.

Burchell, A., 1994. Hepatic microsomal glucose transport. Biochem. Soc. Trans. 22, 658–663.Burchell, A., 1998. A re−evaluation of GLUT 7. Biochem. J. 331, 973.Carroll, J.A., Daniel, J.A., Keisler, D.H., Matteri, R.L., 2000. Postnatal function of the somatotrophic

axis in pigs born naturally or by caesarian section. Domest. Anim. Endocrinol. 19, 39–52.Cassuto, H., Aran, A., Cohen, H., Eisenberger, C.L., Reshef, L., 1999. Repression and activation of

transcription of phosphoenolpyruvate carboxykinase gene during liver development. FEBS Lett. 457,441–444.

Challis, J.R.G., Sloboda, D., Matthews, S.G., Holloway, A., Alfaidy, N., Patel, F.A., Whittle, W., Fraser, M.,Moss, T.J.M., Newnham, J., 2001. The fetal placental hypothalamic-pituitary-adrenal (HPA) axis,parturition and post natal health. Mol. Cell. Endocrinol. 185, 135–144.

Chandler, K.D., Leury, B.J., Bird, A.R., Bell, A.W., 1985. Effects of undernutrition and exercise duringlate pregnancy on uterine, fetal and uteroplacental metabolism in the ewe. Brit. J. Nutr. 53, 625–635.

Chatelain, F., Pegorier, J.P., Minassian, C., Bruni, N., Tarpin, S., Girard, J., Mithieux, G., 1998.Development and regulation of glucose-6-phosphatase gene expression in rat liver, intestine, andkidney: in vivo and in vitro studies in cultured fetal hepatocytes. Diabetes 47, 882–889.

Chow, J.C., Jesse, B.W., 1992. Interactions between gluconeogenesis and fatty acid oxidation in isolatedsheep hepatocytes. J. Dairy Sci. 75, 2142–2148.

Dalle, M., Pradier, P., Delost, P., 1985. The regulation of glucocorticosteroid secretion during theperinatal period. Reprod. Nutr. Dev. 25, 977–991.

Dhanotiya, R.S., Bhardwaj, R., 1988. Sequential development of enzymes of gluconeogenesis andglucose synthesis in fetal goat liver. Biomed. Biochim. Acta 47, 805–808.

Dirlewanger, M., Schneiter, P., Paquot, N., Jequier, E., Rey, V., Tappy, L., 2000. Effects of glucocorticoidson hepatic sensitivity to insulin and glucagon in man. Clin. Nutr. 19, 29–34.

Donkin, S.S., Armentano, L.E., 1993. Preparation of extended in vitro cultures of bovine hepatocytes thatare hormonally responsive. J. Anim. Sci. 71, 2218–2227.

Donkin, S.S., Armentano, L.E., 1994. Regulation of gluconeogenesis by insulin and glucagon in theneonatal bovine. Amer. J. Physiol. 266, R1229–R1237.

Donkin, S.S., Armentano, L.E., 1995. Insulin and glucagon regulation of gluconeogenesis in prerumi-nating and ruminating bovine. J. Anim. Sci. 73, 546–551.

Donkin, S.S., Bertics, S.J., Armentano, L.E., 1997. Chronic and transitional regulation of gluconeo-genesis and glyconeogenesis by insulin and glucagon in neonatal calf hepatocytes. J. Anim. Sci. 75,3082–3087.

Donkin, S.S., Black, D.S., Greenfield, R.B., Agca, C., 1998. Effects of age, gluconeogenic substrates, anddiet on hepatic gluconeogenesis and pyruvate carboxylase expression. J. Dairy Sci. 81, Suppl. 1, 137.

Edwards, E.M., Dhand, U.K., Jeacock, M.K., Shepard, D.A.L., 1975. Activities of enzymes concernedwith pyruvate and oxaloacetate metabolism in the heart and liver of developing sheep. Biochim.Biophys. Acta 399, 217–227.

Egli, C.P., Blum, J.W., 1998. Clinical, haematological, metabolic and endocrine traits during the first threemonths of life of suckling Simmentaler calves held in a cow-calf operation. J. Vet. Med. 45, 99–118.

Ferre, P, Decaux, J.F., Issad, T., Girard, J., 1986. Changes in energy metabolism during the suckling andweaning period in the newborn. Reprod. Nutr. Dev. 26, 619–631.

Filsell, O.H., Jarrett, I.G., Taylor, P.H., Keech, D.B., 1969. Effects of fasting, diabetes and glucocorti-coids on gluconeogenic enzymes in the sheep. Biochim. Biophys. Acta 184, 54–63.

Fisher, D.A., Dussault, J.H., Sack, J., Chapra, I.J., 1977. Ontogenesis of hypothalamic-pituitary-thyroidfunction and metabolism in man, sheep, and rat. Recent Prog. Horm. Res. 33, 59–116.

Fowden, A.L., 1995. Endocrine regulation of fetal growth. Reprod. Fertil. Dev. 7, 351–363.Fowden, A.L., Mapstone, J., Forhead, A.J., 2001. Regulation of glucogenesis by thyroid hormones in

fetal sheep during late gestation. J. Endocrinol. 170, 461–469.Fowden, A.L., Mijovic, J., Silver, M., 1993. The effects of cortisol on hepatic and renal gluconeogenic

enzyme activities in the sheep fetus during late gestation. J. Endocrinol. 137, 213–222.Freemark, M., 1999. The fetal adrenal and the maturation of the growth hormone and prolactin axis.



Ganguli, S., Sinha, M.K., Sterman, B., Harris, P., Sperling, M.A., 1983. Ontogeny of hepatic insulin andglucagon receptors and adenylate cyclase in rabbit. Amer. J. Physiol. 244, E624–E631.

Girard, J., 1986. Gluconeogenesis in late fetal and early neonatal life. Biol. Neonate 50, 237–258.Girard, J., 1990. Metabolic adaptations to change of nutrition at birth Biol. Neonate 58, Suppl. 1, 3–15.Girard, J., Sperling, M., 1983. Glucagon in the fetus and the newborn. In: Lefebre, P.J. (Ed.), Glucagon:

Handbook of Experimental Pharmacology. Springer-Verlag, New York, p. 66.Girard, J., Ferré, P., Pégorier, J.P., Duée, P.H., 1992. Adaptions of glucose and fatty acid metabolism

during perinatal period and suckling-weaning transition. Physiol. Rev. 72, 507–562.Gluckman, P.D., Sizonenko, S.V., Bassett, N.S., 1999. The transition from fetus to neonate – an endocrine

perspective. Acta Pædiat. 428, 7–11.Greenfield, R.B., Cecava, M.J., Donkin, S.S., 2000. Changes in mRNA expression for gluconeogenic

enzymes in liver of dairy cattle during the transition to lactation. J. Dairy Sci. 83, 1228–1236.Gresham, E.L., James, E.J., Raye, J.R., Battaglia, F.C., Makowski, E.L., Meschia, G., 1972. Production

and excretion of urea by the fetal lamb. Pediatrics 50, 372–379.Hadorn, U., Hammon, H., Bruckmaier, R.M., Blum, J.W., 1997. Delaying colostrum intake by one day

has important effects on metabolic traits and on gastrointestinal and metabolic hormones in neonatalcalves. J. Nutr. 127, 2011–2023.

Hamada, T., Matsumoto, M., 1984. Effects of nutrition and ontogeny on liver cytosolic and mitochondr-ial phosphoenolpyruvate carboxykinase activity of the rat, hamster, guinea-pig, pig, kid, calf andchick. Comp. Biochem. Physiol. B. 77, 547–550.

Hammon, H., Blum, J.W., 1997. The somatotropic axis in neonatal calves can be modulated by nutrition,growth hormone, and long-R3-IGF-I. Amer. J. Physiol. 273, E130–E138.

Hammon, H.M., Blum, J.W., 1998. Metabolic and endocrine traits of neonatal calves are influenced byfeeding colostrum for different durations or only milk replacer. J. Nutr. 128, 624–632.

Hanson, R.W., Reshef, L., 1997. Regulation of phosphoenolpyruvate carboxykinase (GTP) gene expres-sion. Annu. Rev. Biochem. 66, 581–611.

Hay, W.W. Jr., Sparks, J.W., Wilkening, R.B., Battaglia, F.C., Meschia, G., 1984. Partition of maternalglucose production between conceptus and maternal tissues in sheep. Amer. J. Physiol. 245,E347–E350.

Hod, Y., Cook, J.S., Weldon, S.L., Short, J.M., Wynshaw-Boris, A., Hanson, R.W., 1987. Differentialexpression of the genes for the mitochondrial and cytosolic forms of phosphoenolpyruvate carboxyk-inase. Ann. N.Y. Acad. Sci. 478, 31–35.

Hodgson, J.C., Mellor, D.J., Field, A.C., 1982. Foetal and maternal rates of urea production and disposalin well-nourished and undernourished sheep. Brit. J. Nutr. 48, 49–58.

Holcomb, T., Curthoys, N.P., Gstraunthaler, G., 1995. Subcellular localization of PEPCK and metabolismof gluconeogenic substrains of renal cell lines. Amer. J. Physiol. 268, C449–C457.

Hosokawa, M., Thorens, B., 2002. Glucose release from GLUT2-null hepatocytes: characterization ofa major and a minor pathway. Amer. J. Physiol. Endocrinol. Metab. 282, E794–E801.

Howarth, R.E., Baldwin, R.L., Ronning, M., 1968. Enzyme activities in liver, muscle, and adipose tissueof calves and steers. J. Dairy Sci. 51, 1270–1274.

Huntington, G.B., 1990. Energy metabolism in the digestive tract and liver of cattle: influence ofphysiological state and nutrition. Reprod. Nutr. Dev. 30, 35–47.

Jitrapakdee, S., Booker, G.W., Cassady, A.I., Wallace, J.C., 1996. Cloning, sequencing and expression ofrat liver pyruvate carboxylase. Biochem. J. 316, 631–637.

Jitrapakdee, S., Wallace, J.W., 1999. Structure and regulation of pyruvate carboxylase. Biochem. J.340, 1–16.

Jitrapakdee, S., Gong, Q., MacDonald, M.J., Wallace, J.C., 1998. The rat pyruvate carboxylase genestructure: alternate promoters generate multiple transcripts with the 5’-end heterogeneity. J. Biol.Chem. 272, 20522–20530.

Jones, C.G., Hothi, S.K., Titheradge, M.A., 1993. Effect of dexamethasone on gluconeogenesis, pyruvatekinase, pyruvate carboxylase and pyruvate dehydrogenase flux in isolated hepatocytes. Biochem. J.289, 821–830.

Kalhan, S., Parimi, P., 2000. Gluconeogenesis in the fetus and neonate. Semin. Perinatol. 24, 94–106.Lecavalier, L., Bolli, G., Gerich, G., 1990. Glucagon-cortisol interactions on glucose turnover and lactate

gluconeogenesis in normal humans. Amer. J. Physiol. 258, E569–E575.Lemaigre, F.P., Rousseau, G.G., 1994. Transcriptional control of genes that regulate glycolysis and glu-

coneogenesis in adult liver. Biochem. J. 303, 1–14.


Leury, B.J., Bird, A.R., Chandler, K.D., Bell A.W., 1990. Glucose partitioning in the pregnant ewe:effects of undernutrition and exercise. Brit. J. Nutr. 64, 449–462.

Li, J., Owens, J.A., Owens, P.C., Saunders, J.C., Fowden, A.L., Gilmour, R.S., 1996. The ontogeny ofhepatic growth hormone receptor and insulin-like growth factor I gene expression in the sheep fetusduring late gestation: developmental regulation by cortisol. Endocrinology 137, 1650–1657.

Liggins, G.C., 1977. The role of cortisol in preparing the fetus for birth. Reprod. Fertil. Dev. 6, 141–150.Louet, J.F., Hayhurst, G., Gonzalez, F.J., Girard, J., Decaux, J.F., 2002. The coactivator PGC-1 is

involved in the regulation of the liver carnitine palmitoyltransferase I gene expression by cAMP incombination with HNF4 alpha and cAMP-response element-binding protein (CREB). J. Biol. Chem.277, 37991–38000.

Maciel, S.M., Chamberlain, C.S., Wettemann, R.P., Spicer, L.J., 2001. Dexamethasone influencesendocrine and ovarian function in dairy cattle. J. Dairy Sci. 84, 1998–2009.

Marco, J., Calle, C., Román, D., Díaz-Fierros, M., Villanueva, M.L., Valverde, I., 1973. Hyperglucagonisminduced by glucocorticoid treatment in man. N. Engl. J. Med. 288, 128–131.

Mashek, D.G., Bertics, S.J., Grummer, R.R., 2002. Metabolic fate of long-chain unsaturated fatty acidsand their effects on palmitic acid metabolism and gluconeogenesis in bovine hepatocytes. J. DairySci. 85, 2283–2289.

Muramatsu, M., Sugawara, M., Tsuda, T., 1974. Changes in rates of recycling and turnover of glucoseduring development of sheep. Agr. Biol. Chem. 38, 259–267.

Narkewicz, M.R., Carver, T.D., Hay, W.W. Jr., 1993. Induction of cytosolic phosphoenolpyruvate car-boxykinase in the ovine fetal liver by chronic fetal hypoglycemia and hypoinsulinemia. Pediat. Res.33, 493–496.

Neufeld, N.D., Scott, M., Kaplan, S.A., 1980. Ontogeny of the mammalian insulin receptor: studies ofhuman and rat fetal liver plasma membranes. Dev. Biol. 78, 151–160.

O’Brien, R.M., Granner, D.K., 1990. PEPCK gene as model of inhibitory effects of insulin on gene tran-scription. Diabetes Care 13, 327–339.

Park, E.A., Song, S., Olive, M., Roesler, W.J., 1997. CCAAT-enhancer-binding protein alpha (C/EBPalpha) is required for the thyroid hormone but not the retinoic acid induction of phosphoenolpyruvatecarboxykinase (PEPCK) gene transcription. Biochem. J. 322, 343–349.

Pearce, J., Unsworth, E.F., 1980. The effects of diet on some hepatic enzyme activities in the pre-ruminantand ruminating calf. J. Nutr. 110, 255–261.

Pilkis, S.J., Claus, T.H., 1991. Hepatic gluconeogenesis/glycolysis: regulation and structure/functionrelationships of substrate cycle enzymes. Annu. Rev. Nutr. 11, 465–515.

Prior, R.L., 1980. Glucose and lactate metabolism in vivo in ovine fetus. Amer. J. Physiol. 239, E208–E214.Prior, R.L., 1982. Gluconeogenesis in the ruminant fetus: evaluation of conflicting evidence from radio-

tracer and other experimental techniques. Fed. Proc. 41, 117–122.Prior, R.L., Scott, R.A., 1977. Ontogeny of gluconeogenesis in the bovine fetus: influence of maternal

dietary energy. Dev. Biol. 58, 384–393.Sangild, P.T., 2001. Transitions in the life of the gut at birth, In: Lindberg, J.E., Ogle, B. (Eds.), Digestive

Physiology of Pigs. CABI Publishing, New York, pp. 3–17.Savan, P.M., Jeacock, M.K., Shepherd, D.A., 1986. Gluconeogenesis in foetal, suckling and weaned

lambs. J. Agr. Sci. 106, 259–268.Seoane, J., Trinh, K., O’Doherty, R.M., Gomez-Foix, A.M., Lange, A.J., Newgard, C.B., Guinovart, J.J.,

1997. Metabolic impact of adenovirus-mediated overexpression of the glucose-6-phosphatasecatalytic subunit in hepatocytes. J. Biol. Chem. 272, 26972–26977.

Sistare, F.D., Haynes, R.C., 1985. The interaction between the cytosolic pyridine nucleotide redox potentialand gluconeogenesis from lactate/pyruvate in isolated rat hepatocytes: implications for investigationsof hormone action. J. Biol. Chem. 260, 12748–12753.

Sperling, M.A., Grajwer, L., Leake, R.D., Fisher, D.A., 1977. Effects of somatostatin (SRIF) infusion on glu-cose homeostasis in newborn lambs: evidence for a significant role of glucagon. Pediat. Res. 11, 962–967.

Stevenson, R.E., Morris, F.H., Jr., Adcock, E.W. III, Howell, R.R., 1976. Development of gluconeogenicenzymes in fetal sheep liver and kidney. Dev. Biol. 52, 167–172.

Swiatek, K.R., 1971. Development of gluconeogenesis in pig liver slices. Biochim. Biophys. Acta 252,274–279.

Taylor, P.H., Wallace, J.C., Keech, D.B., 1971. Gluconeogenic enzymes in sheep liver: intracellular local-ization of pyruvate carboxylase and phosphoenolpyruvate carboxykinase in normal, fasted anddiabetic sheep. Biochim. Biophys. Acta 237, 179–191.


Tsoulos, N.G., Colwill, J.R., Battaglia, F.C., Makowski, E.L., Meschia, G., 1971. Comparison of glucose,fructose, and O2 uptakes by fetuses of fed and starved ewes. Amer. J. Physiol. 221, 234–237.

Van Schaftingen, E., Gerin, I., 2002. The glucose-6-phosphatase system. Biochem. J. 362, 513–532.Velez, J.C., Donkin, S.S., 2000. Administration of bST elevates phosphoenolpyruvate carboxykinase

mRNA in lactating dairy cows. J. Anim. Sci. 78, Suppl. 1, 144.Warnes, D.M., Seamark, R.F., Ballard, F.J., 1977. Metabolism of glucose, fructose and lactate in vivo in

chronically cannulated foetuses and in suckling lambs. Biochem. J. 162, 617–626.Watford, M., Hod, Y., Chiao, Y.B., Utter, M.F., Hanson, R.W., 1981. The unique role of the kidney in

gluconeogenesis in the chicken: the significance of a cytosolic form of phosphoenolpyruvate carboxyki-nase. J. Biol. Chem. 256, 10023–10027.

Weinstein, S.P., Paquin, T., Pritsker, A., Haber, R.S., 1995. Glucocorticoid-induced insulin resistance:dexamethasone inhibits the activation of glucose transport in rat skeletal muscle by both insulin- andnon-insulin-related stimuli. Diabetes 44, 441–445.

Williamson, J.R., Kreisberg, R.A., Felts, P.W., 1966. Mechanism for the stimulation of gluconeogenesisby fatty acids in perfused rat liver. Proc. Natl. Acad. Sci. USA 56, 247–254.

Wise, J.K., Hendler, R., Felig, P., 1973. Influence of glucocorticoids on glucagon secretion and plasmaamino acid concentrations in man. J. Clin. Invest. 52, 2774–2782.

Zhang, J., Xia, W., Brew, K., Ahmad, F., 1993. Adipose pyruvate carboxylase: amino acid sequence anddomain structure deduced from cDNA sequencing. Proc. Natl. Acad. Sci. USA 90, 1766–1770.


391

16 Energy metabolism in the developingrumen epithelium

B. W. Jesse

Department of Animal Science, Rutgers, The State University of New Jersey,84 Lipman Drive, New Brunswick, NJ 08901-8525, USA

The physical changes occurring during rumen epithelial development have been extensivelycharacterized. However, relatively little information is available concerning development ofenergy metabolism in rumen epithelium. Available data indicate that both ontogenic andphysiological/dietary factors are necessary for complete rumen epithelial metabolic develop-ment. Changes in the expression of specific genes, e.g. those for ketone body production, inresponse to ontogenic and physiological/dietary factors appear to be responsible for thechanges in energy metabolism in developing rumen epithelium. Future research efforts willneed to identify the mechanisms regulating gene expression within the developing rumenepithelium to obtain a better understanding of this process.

1. INTRODUCTION

Energy metabolism in the rumen epithelium of mature sheep and cattle has been extensivelycharacterized over the years. The specific oxidizable substrates required for energy produc-tion in the rumen epithelium in neonates of these species have received some attention, whilethe establishment of the rumen fermentation, and the physical changes occurring to the rumenepithelium during development, have been extensively researched. However, the changes inenergy metabolism that occur during neonatal rumen epithelial metabolic development, andmost importantly the timing and control mechanisms regulating those changes, have receivedrelatively little attention. This review will provide a historical overview of the state of ourknowledge in this area, and will discuss in more depth recent evidence that examines metabolicdevelopment in the neonatal rumen epithelium. It will become apparent that relatively little isknown concerning the mechanisms driving rumen metabolic development, and that this is theresult of relatively little research having been conducted in this area. The vast bulk of theresearch literature examining rumen development has focused on the rumen fermentationitself, or on physical changes manifested by changes in rumen epithelial morphology andblood chemistry in the growing ruminant. While the focus of the review will be on rumen



epithelium, it should be noted that developmental changes will also be occurring in the reticularand omasal epithelia. This review will conclude with a brief discussion of key areas for futureresearch into the topic of rumen epithelial metabolic development.

2. Historical Perspective

To understand the changes in energy metabolism occurring in developing rumen epithelium,it is necessary to first define the points that mark the limits of these changes, that is, the energymetabolism present at the start (in the neonatal rumen) and at the endpoint (in the maturerumen) of rumen development. For the purpose of this review this time period will generallycoincide with the 8-week period following birth, the time period in which the bulk of thephysical and metabolic changes occur in the developing rumen epithelium.

2.1. Energy metabolism in mature rumen epithelium

2.1.1. Substrates and their metabolism

The primary compounds that have been investigated as potential energy-yielding substrateswithin the mature rumen epithelium are fatty acids (both short- and long-chain), glucose, andglutamine (Weigand et al., 1975; White and Leng, 1980; Harmon, 1986; Harmon et al., 1991;Jesse et al., 1992; Britton and Krehbiel, 1993; Remond et al., 1995; Baldwin and McLeod,2000). Of these, the short-chain or volatile fatty acids (VFA) are quantitatively the mostimportant energy sources for the ruminal epithelium under most circumstances.

The VFA, predominantly acetate, propionate, and butyrate, are the products of the rumenfermentation, and are absorbed by the rumen epithelium for release into the portal circulation.Prior to release into the portal circulation, the VFA may undergo metabolism within the rumenepithelium. Depending upon the specific VFA considered, a variable amount of metabolismoccurs (Remond et al., 1995). VFA metabolism may include either oxidation, or conversioninto other intermediates (e.g. lactate, β-hydroxybutyrate [BHBA], acetoacetate [AcAc]), forrelease into the portal circulation. The activities of numerous enzymes in the glycolytic pathway,the citric acid cycle, the ketogenic pathway, the acyl-CoA synthetases for activation of VFA,and the various enzymes involved in VFA uptake have been determined (Young et al., 1969; Bushand Milligan, 1971; Ash and Baird, 1973; Nocek et al., 1980; Scaife and Tichivangana, 1980;Bush, 1982; Leighton et al., 1983; Harmon et al., 1991). These assays have generally beenconducted under saturating substrate concentrations to yield maximal activities of the assayedenzymes. Consequently, relatively little information is available concerning the kinetic prop-erties of rumen epithelial enzymes. Activation of VFA to the coenzyme A thioester has beenproposed as the key regulatory point for rumen epithelial VFA metabolism (Ash and Baird,1973). However, others have indicated that knowledge of both the kinetic properties of theseenzymes as well as the tissue substrate and inhibitor concentrations within the rumen epitheliumis needed to fully justify that statement (Britton and Krehbiel, 1993).

While various researchers have reported on the effects of dietary changes (composition,level of intake) on VFA metabolism and activities of specific enzymes within the rumenepithelium, no consensus has yet emerged from these studies. Harmon et al. (1991) noted anoverall increase in ruminal epithelial metabolism in cattle fed at twice maintenance require-ments than in cattle fed at maintenance. These authors also noted some increase in acyl-CoAsynthetase activities of rumen epithelium from cattle fed a high-forage diet. Some researchersreport no change in activity for a number of enzymes in rumen epithelium from cattle fed

B. W. Jesse392

a high-grain versus hay diet (e.g. Young et al., 1969). Others have noted differences inVFA transport across the rumen epithelium as well as changes in some enzyme activities(propionyl-CoA synthetase, glutamate dehydrogenase, and aspartate aminotransferase) as aresult of changes in ration physical form and level of rumen-degradable nitrogen (Nocek et al., 1980). A recent paper suggests that changes in rumen epithelial energy metabolism inresponse to dietary energy intake and composition is due in part to changes in tissue massrather than to changes in metabolism per unit epithelial mass (McLeod and Baldwin, 2000).This is in agreement with other studies noting increases in rumen epithelial mass and papil-lae length in response to increased dietary energy intake (Liebich et al., 1987).

Of the three major VFA absorbed by rumen epithelium, the proportion of absorbed acetatethat is metabolized is lower than any of the other VFA (18–30%; Remond et al., 1995). However,since significantly more acetate is absorbed than propionate and butyrate, the absolute amountof acetate metabolized can be relatively large. The literature indicates that, compared tobutyrate, relatively little acetate is converted to ketone bodies (BHBA and AcAc; Remond et al.,1995). Acetate undergoes primarily oxidation to carbon dioxide by the rumen epithelium,thereby contributing to the energy needs of the rumen epithelium (Britton and Krehbiel, 1993).

The situation with propionate metabolism in the ruminal epithelium is the least clear of theVFA. Propionate is not used for the synthesis of ketone bodies, but is converted primarily tolactate, with some complete oxidation to carbon dioxide, some pyruvate formation, and sometransamination of pyruvate to alanine (Remond et al., 1995). Some studies have estimated thatas much as 70% of absorbed propionate is converted to lactate prior to release into the portalcirculation, although more recent data suggest that the proportion is much less than that (30%;Remond et al., 1995). Propionate oxidation to carbon dioxide by ruminal epithelium is min-imal at physiological concentrations of propionate, presumably to spare propionate and itsmetabolites for hepatic gluconeogenesis (Remond et al., 1995), although at high propionateconcentrations rumen epithelium in vitro can oxidize propionate at relatively high rates(Harmon et al., 1991). Butyrate has been noted to inhibit propionate activation to propionyl-CoA,thereby minimizing propionate metabolism and further sparing propionate for release into theportal circulation (Harmon et al., 1991).

Butyrate has long been known to be the VFA most extensively metabolized by the rumenepithelium, undergoing both oxidation to carbon dioxide and conversion to BHBA and AcAc(Bergman, 1990; Remond et al., 1995). Various researchers have estimated that as much as90% of the absorbed butyrate undergoes metabolism by the ruminal epithelium (Bergman,1990; Remond et al., 1995). Generally, a higher proportion of butyrate is converted to ketonebodies, and a lower proportion to carbon dioxide, than occurs with acetate, although theabsolute rates of carbon dioxide production from acetate and butyrate are comparable(Harmon et al., 1991). This may simply be a reflection of the relative rate of activation ofthese two VFA by their respective acyl-CoA synthetases (Harmon et al., 1991). Acetyl-CoAsynthetase activity is significantly lower than either propionyl- or butyryl-CoA synthetaseactivities (Harmon et al., 1991). A lower rate of acetate activation may provide sufficientAc-CoA for use in the TCA cycle, but not a sufficiently high concentration for use of Ac-CoAas a ketogenic substrate, at least when acetate is the sole substrate in vitro. The kinetic param-eters of the ruminal enzymes responsible for use of acetyl-CoA in these pathways (i.e. citratesynthase and AcAc-CoA thiolase) are not known, however. Butyrate can inhibit acetate andpropionate activation to their respective coenzyme A thioesters, while acetate and propionatehave relatively little effect on butyrate activation (Harmon et al., 1991). These data are con-sistent with observations of Scaife and Tichivangana (1980), who isolated a partially purifiedshort-chain acyl-CoA synthetase from sheep rumen epithelium. The kinetic properties of this

Energy metabolism in the developing rumen epithelium 393

fraction suggested the existence of two distinct enzyme activities, one specific for butyrateactivation and the other capable of activating acetate, propionate, or butyrate (Scaife andTichivangana, 1980).

The long-chain fatty acid palmitate may be oxidized or used in the synthesis of ketonebodies by rumen epithelium in much the same manner as in liver (Jesse et al., 1992). Isolatedrumen epithelial cells oxidized palmitate at one-quarter the rate of butyrate, and convertedpalmitate to ketone bodies at one-half the rate of butyrate (Jesse et al., 1992). Propionate,butyrate, and ammonia inhibited ketogenesis from palmitate, but only butyrate and ammoniainhibited palmitate oxidation (Jesse et al., 1992). These data suggest that during feed restric-tion, or possibly when consuming a high-fat diet, mature rumen epithelium would be capableof using long-chain fatty acids as a major energy source.

Both glucose and glutamine can undergo oxidation to carbon dioxide, and conversion tolactate in the case of glucose, or to glutamate and alanine in the case of glutamine, within theruminal epithelium (Remond et al., 1995). Glucose may be a major source of the lactateproduced by rumen epithelium (Remond et al., 1995). Rates of glucose oxidation to carbondioxide are comparable to lactate production rates from glucose, but glucose oxidation occursat a lower rate than either acetate or butyrate oxidation (Harmon et al., 1991). Glutamineoxidation rates by ruminal epithelium reportedly were 7 times lower than glucose oxidation(Harmon et al., 1991), suggesting that glutamine is not a major energy source for ruminalepithelium, in contrast to the importance of glutamine as an energy source to other tissues ofthe digestive tract (Britton and Krehbiel, 1993). More recent data indicate that glutamine in vitro can be oxidized by rumen epithelial cells at rates faster than butyrate, if present atsufficiently high concentrations (50 mM; Baldwin and McLeod, 2000). However, the gluta-mine concentration required for half-maximal oxidation rates by rumen epithelial cells(6 mM; Baldwin and McLeod, 2000) is about 30 times greater than the glutamine concentra-tion found in vivo (0.20 mM; Alio et al., 2000; Noziere et al., 2000; Hanigan et al., 2001).This suggests that little glutamine oxidation would be expected to occur in vivo (Baldwin andMcLeod, 2000), as was noted by Harmon et al. (1991).

2.1.2. Substrate uptake

Prior to activation and metabolism within the rumen epithelium, energy substrates must betransported into the epithelium. For the rumen VFA this presents a unique challenge, as at thepH typical of rumen fluid (5.6–6.2), VFA exist predominantly in the ionized form. IonizedVFA would be unable to diffuse through the plasma membranes of the rumen epithelial cells.Consequently, some mechanism must exist for the movement of VFA across the epithelium.While early data supported a transcellular rather than a paracellular mechanism, the exactmechanism was not known (Remond et al., 1995). Early research suggested the importanceof carbonic anhydrase in the absorption of VFA by rumen epithelium (Aafjes, 1967;Bergman, 1990). The proposed mechanism involved production of HCO3

− and H+ within therumen epithelium, movement of the protons and bicarbonate across the rumen mucosa andinto the rumen fluid, and neutralization of the VFA followed by passive diffusion into therumen mucosa down a concentration gradient (Bergman, 1990). Recent data suggest theexistence of both a carrier-mediated transport mechanism and a passive diffusion mechanism(Sehested et al., 1999a), with the mediated transport mechanism coupled with sodium, chloride,and bicarbonate (Sehested et al., 1999b). These results are summarized in fig. 1. While the situ-ation for acetate and propionate is unknown, butyrate transport appears to be energy-dependent,as inhibition of ATP synthesis in rumen epithelium blocks butyrate uptake (Gabel et al., 2001).

B. W. Jesse394

This suggests that ATP is either directly involved in butyrate transport, or that energy-dependent metabolism of butyrate is necessary for butyrate uptake by rumen epithelium.

Until recently, the mechanism of glucose transport into rumen epithelium had not beenexamined. The general concept was that glucose utilized by the rumen epithelium was derivedfrom the blood and absorbed at the serosal side of the epithelium. A recent report indicatedthe presence of GLUT5 (the basolateral facilitative glucose transporter) mRNA in sheeprumen epithelium (Zhao et al., 1998), which would perform the uptake of blood glucose byrumen epithelium. Surprisingly, mRNA for the Na+-dependent glucose transporter (SGLT1)was also detected. Functional analysis of 3-O-methylglucose transport by sheep rumenepithelium in vitro demonstrated the presence of SGLT1, which was subsequently confirmedby cloning a cDNA from rumen epithelium with 100% identity to the sheep intestinal SGLT1(Aschenbach et al., 2000b). In vivo experiments also demonstrated the sodium-dependentabsorption of physiological concentrations of glucose by sheep rumen epithelium(Aschenbach et al., 2000a). The authors suggested that this could be an important route ofglucose absorption in ruminants consuming high-concentrate diets, especially as a mecha-nism to minimize the effects of rumen acidosis, as previously suggested by Ganter et al.(1993). This hypothesis was supported by the observation that sheep rumen epithelial uptakeof glucose by SGLT-1 can be stimulated by β2-adrenoceptors, since increased sympatheticactivity has been noted in acidotic ruminants (Aschenbach et al., 2002). Glucose uptake bymature rumen epithelium is thus more complex than previously believed.

No studies appear to have examined the mechanism for palmitate or glutamine specific uptakeby mature rumen epithelium. Alternatively, glutamine could enter the rumen epithelium as


Fig. 1. Diagram of volatile fatty acid (VFA) transport into rumen epithelial cells. Protons and bicarbonateare generated within the rumen epithelial cells by the action of carbonic anhydrase. Circles represent thepresence of specific transport molecules. Based on Sehested et al. (1999a).

a component of peptides rather than as the amino acid (Webb et al., 1992). A specific peptidetransporter has been detected in ruminal epithelium that could move glutamine-containingpeptides into the ruminal epithelium by an electrogenic mechanism (Chen et al., 1999).However, these results are not universally accepted (Martens et al., 2001), indicating that therole of peptide transport in moving glutamine into the ruminal epithelium has yet to be fullyresolved.

2.2. Energy metabolism in neonatal rumen epithelium

2.2.1. Substrates and their metabolism

Much of the early work on rumen epithelial development was concerned with factors thatwould promote anatomical development of the rumen. For example, the classical work ofWarner et al. (1956) was the first detailed study to examine the effect of different dietary treat-ments on both ruminal size and papillary development. This was the first study to postulatethe importance of VFA from the microbial fermentation as inducers of rumen epithelial devel-opment. Subsequent research (Tamate et al., 1962; Hamada et al., 1976; Klein et al., 1987)confirmed the importance of VFA for the stimulation of papillary growth, and noted thatincreased rumen volume and musculature were dependent on bulk fill of the rumen.Relatively little research, however, examined the energy metabolism of the developing rumenepithelium. The first study examining energy metabolism in undeveloped rumen epitheliumnoted that, prior to papillary development, metabolism of VFA by rumen epithelium was low(Sutton et al., 1963). Blood glucose was subsequently identified as the primary energy substrateof neonatal calf rumen epithelium (Juhasz et al., 1976).

Giesecke et al. (1979) performed the first systematic analysis of the changes in rumen epithe-lial metabolism that occur during rumen development. Using slices of rumen epitheliumisolated from weaned and unweaned lambs of various ages, these researchers measured oxygenconsumption and ketone body production by the rumen epithelial slices in vitro in the presenceof glucose, lactate, butyrate, or propionate. The importance of a number of observations thatwere made by these authors has repeatedly been demonstrated in the intervening years. Oxygenconsumption by rumen epithelium decreased with age independently of dietary changes andstage of rumen epithelial development. Glucose, lactate, and butyrate stimulated oxygen con-sumption by rumen epithelial slices from both 2-week-old and 6-month-old lambs. However,the ability of glucose and lactate to stimulate oxygen consumption in rumen epithelium from 6-month-old lambs was significantly less than in that from 2-week-old lambs, whereas butyratestimulated oxygen consumption equally well in ruminal epithelium from lambs of either age.This was interpreted as a shift in substrate preference from glucose to VFA by the developingrumen epithelium, and was supported by a decrease in glucose uptake by the rumen epitheliumfrom the older lambs. These results also implied that the rate of butyrate oxidation was notdependent on the stage of rumen epithelial development.

Ketogenesis at different developmental stages of the rumen epithelium was also examined.In rumen epithelial slices from 9–10-week-old lambs either maintained on milk (undevelopedepithelium) or weaned to solid feed (developed epithelium), total ketone body production(BHBA + AcAc) was nearly 1.60-fold greater in the developed than the undevelopedepithelium, and was comparable to ketogenic rates observed in older lambs. There was alsoa shift in the BHBA:AcAc ratio from about 2.7 to 6.2 in the undeveloped and developed lambrumen epithelium, respectively, indicating a shift in the redox potential of the rumen epithe-lium with age. These data indicated the importance of solid feed intake in promoting rumen

B. W. Jesse396

metabolic development. Glucose addition to these in vitro incubations stimulated ketogenesisin a synergistic manner. Perhaps the most interesting observation, however, was the change inketogenic capacity with age of rumen epithelium from milk-fed lambs. Total ketone bodyproduction (μmoles/(g tissue dry weight × hour)) by rumen epithelial slices from milk-fedlambs increased from 19.6 at 1 week of age, to 25.5 at 3–4 weeks of age, to 71.9 at 9–10 weeksof age. This was the first indication that changes in energy metabolism within the rumenepithelium could occur in the absence of solid feed intake and the consequent microbial production of VFA.

The observations of Giesecke et al. (1979) concerning ruminal ketogenesis were subse-quently confirmed and extended to rumen epithelium from milk-fed and normally reared(milk-fed to 28 days of age; starter and hay available after 10 days of age) calves by Bush(1988). This author examined rumen epithelium from 3-, 12-, 19-, 30-, and 60-day-old calves,providing a more complete time-course of the changes in ketogenesis occurring in develop-ing rumen epithelium. Total ketone body production from butyrate by rumen epithelium fromnormally reared calves was detectable at 3 days of age, and slowly increased through 19 daysof age. By 30 days of age ketogenic rates had jumped to about 40% of that observed in maturerumen epithelium, and by 60 days of age were similar to ketogenic rates in mature rumenepithelium. Ketogenesis from butyrate also increased with age in rumen epithelium frommilk-fed calves, although the difference in total ketone body production rate by rumen epithe-lium from milk-fed and conventionally reared calves at 60 days of age was about 4.8-fold(Bush, 1988), in contrast to the 1.6-fold difference observed by Giesecke et al. (1979).Acetate conversion rate to ketone bodies was nearly 12.5-fold less than was observed withbutyrate as substrate in rumen epithelium from 60-day-old conventionally reared calves,again similar to that in rumen epithelium from older animals (Bush, 1988).

Two important observations were made in both of these studies (Giesecke et al., 1979;Bush, 1988). First, some changes in rumen epithelial energy metabolism occur in an onto-genic manner even in the absence of solid feed intake and the associated rumen fermentation.Second, complete rumen epithelial metabolic development requires solid feed intake, and ismediated presumably by the VFA from the resultant feed fermentation.

2.2.2. Substrate uptake

The underlying mechanisms(s) responsible for the changes in glucose, butyrate, and ketonebody metabolism in the developing rumen epithelium were in general not identified by Bush(1988) or Giesecke et al. (1979). Altered substrate uptake or metabolism, or a combination ofboth, could be responsible for the observed changes in energy metabolism during rumenepithelial development. Giesecke et al. (1979) did note a nearly 10-fold decrease in glucoseuptake in rumen epithelium from 6-month-old lambs compared to that observed in 2-week-oldlambs. (In contrast, lactate uptake by rumen epithelium from 6-month-old lambs were morethan doubled.) Whether this decrease in glucose uptake was due to a decline in glucose trans-porter activity, or to a decrease in glucose metabolic capacity, during rumen epithelialdevelopment is not known. Similarly, changes in butyrate metabolism by developing rumenepithelium could be the result of changes in the activity of the appropriate transporter activity,or in the activity of butyrate metabolizing enzymes within the ruminal epithelium. Prior to1992, no reports had been made concerning activity changes of metabolic enzymes in devel-oping rumen epithelium that would provide an explanation for the major metabolic changesoccurring during rumen epithelial development. There apparently was also no research intoVFA absorption by the developing rumen epithelium. Similarly, although glucose oxidation


by mature rumen epithelium had been observed at that time, no research had been conductedinto the mechanism of glucose absorption by either mature or developing rumen epithelium.

3. RECENT DATA ON THE ENERGY METABOLISM OF DEVELOPINGRUMEN EPITHELIUM

Beginning in 1992, a series of papers appeared that attempted to define more finely the timingof the energy metabolism changes occurring in developing rumen epithelium, as well as toattempt identification of the mechanisms responsible for those changes.

3.1. Substrate oxidation

In 1992, Baldwin and Jesse reported on the developmental changes in glucose and butyratemetabolism by rumen epithelial cells isolated from conventionally reared lambs of differentages (0, 4, 7, 14, 28, 42, and 56 [weaned] days of age). In this study, glucose oxidation (basedon 14CO2 production from [1-14C]glucose by isolated rumen epithelial cells) increased frombirth to 14 days of age, remained elevated until 42 days of age, and decreased by weaning at56 days to rates lower than those observed at birth, but comparable to those observed in maturesheep (Baldwin and Jesse, 1992). Maximum glucose oxidation rates coincided with the timeperiod of allometric rumen growth, suggesting the importance of glucose oxidation for energygeneration during this time of rapid rumen tissue accretion. Surprisingly in this study, butyrateoxidation rates (based on 14CO2 production from [1-14C]butyrate by isolated rumen epithelialcells) were maximal at 4 days of age (nearly 7-fold greater than those observed at birth),clearly indicating the ability of undeveloped rumen epithelium to absorb and metabolize VFA.Butyrate oxidation rates decreased gradually until weaning at 56 days, and were comparableto those observed in older lambs. In rumen epithelial cells isolated from 28-day-old andyounger lambs, addition of unlabeled glucose or butyrate decreased 14CO2 production from theother labeled substrate. The data presented could not distinguish between actual inhibition ofoxidation of the alternative labeled substrate by addition of unlabeled substrate, or simple dilu-tion of the specific activity within the acetyl-CoA pool from the labeled substrate by theunlabeled substrate prior to complete oxidation in the citric acid cycle (Baldwin and Jesse,1992). Either explanation, however, is consistent with the ability of neonatal rumen epitheliumto absorb and oxidize butyrate. What is not clear is why neonatal rumen epithelium should pos-sess that ability in such a magnitude at a time when little if any butyrate is present within therumen. The changes in rumen epithelial metabolism are summarized in fig. 2.

3.2. Ketogenesis

Baldwin and Jesse (1992) also found that ketogenesis from butyrate, as measured by BHBAproduction rate, was undetectable at birth, but increased to a low, relatively steady rate

B. W. Jesse398

Fig. 2. Summary of the metabolic changes occurring during the development of lamb rumen epithelium.Based on Baldwin and Jesse (1992).

from 4 through 42 days of age. After 42 days of age ketogenic rates increased markedly (about8-fold), so that at weaning at 56 days of age butyrate conversion to ketone bodies was occur-ring at nearly adult rates (Giesecke et al., 1979; Bush, 1988; Harmon et al., 1991; Baldwinand Jesse, 1992). Based on these findings, lamb rumen epithelial development was suggestedto occur in stages, with the two most prominent stages being the period of rapid rumen growthand keratinization occurring between about 28 and 42 days of age, followed by the onset ofmetabolic maturity as indicated by the onset of high rates of ketogenesis from butyrate(between 42 and 56 days of age) at weaning (Baldwin and Jesse, 1992). The suggestion wasmade that VFA from the rumen fermentation may be acting to promote rumen epithelial meta-bolic development in the same way that VFA had been noted to promote rumen papillarydevelopment (Warner et al., 1956). However, two confounding variables existed in this study,namely the change in diet (leading to physiological adaptation) and the increase in age of thelambs that would be associated with ontogeny of rumen epithelial development (Shirazi-Beechey et al., 1991a; Baldwin and Jesse, 1992).

To distinguish between these two possibilities, a study was conducted to determine theability of VFA to stimulate rumen metabolic development (Lane and Jesse, 1997). Milk-fedlambs received either continuous intraruminal infusions of a physiological mixture of VFA(acetate, propionate, butyrate) or saline, or no intraruminal infusions, for 7–10 weeks. No sig-nificant differences in rumen epithelial parameters were found in this study, but several trendswere noted. Papillae length tended to be longer in the VFA-infused lambs, suggesting that theVFA were acting to stimulate papillae growth as expected (Warner et al., 1956). Glucoseoxidation tended to be lower, and AcAc production from butyrate higher, in the VFA-infusedlambs than in the saline-infused or uninfused controls. No other metabolic differences wereobserved among the three infusion treatments. Both glucose oxidation and BHBA productionfrom butyrate were similar between the various infusion treatments and conventionally rearedlambs (Baldwin and Jesse, 1992; Lane and Jesse, 1997). The results of VFA infusion on stim-ulating rumen epithelial development in this study were inconclusive, in view of the lack ofsignificant treatment differences. The minimal effect of VFA infusion on papillae develop-ment suggests that insufficient amounts of VFA may have been administered during thisexperiment to stimulate maximal rumen metabolic development. The similarity in glucoseoxidation and BHBA production from butyrate between the various infusion treatments andconventionally reared lambs, however, provided additional support to the concept that onto-genic factors play a prominent role in rumen epithelial metabolic development.

3.3. Ontogeny of rumen metabolic development

A subsequent study then addressed the role of ontogenic development of rumen metabolismby separating out the effects of age and diet on rumen epithelial metabolic development (Laneet al., 2000). Lambs were either maintained on a milk diet before slaughter (0, 4, 7, 14, 28,42, 49, 56, or 84 days of age), or at 49 days of age were weaned onto solid feed and slaugh-tered at 84 days of age. Glucose oxidation by rumen epithelial cells isolated from the milk-fedlambs followed the same general pattern as observed with conventionally reared lambs(Baldwin and Jesse, 1992; Lane et al., 2000). Glucose oxidation rates by isolated rumenepithelial cells were not different among the 84-day-old milk-fed lambs, 84-day-old lambsweaned at 49 days of age (Lane et al., 2000), or conventionally reared lambs weaned at 56 daysof age (Baldwin and Jesse, 1992). These results are similar to those of Giesecke et al. (1979),who observed no difference in glucose oxidation by rumen epithelial pieces from 8–12-week-oldmilk-fed or conventionally reared lambs.


In contrast to that observed with normally reared lambs (Baldwin and Jesse, 1992),butyrate oxidation by isolated rumen epithelial cells was undetectable in lambs 7 days old orless (Lane et al., 2000). No explanation was given for this difference, especially sinceconventionally reared lambs at this age would not yet have begun to consume solid feed andwould not differ physiologically from milk-fed lambs in that study (Lane et al., 2000).Different breeds of sheep were used in the two studies, however, which may have had someeffect on the results. Subsequent to that age (7 days), little difference in butyrate oxidationby isolated rumen epithelial cells was observed through 94 days. Similar to that for glucoseoxidation, butyrate oxidation was not different between rumen epithelial cells isolated from84-day-old milk-fed lambs and 84-day-old lambs weaned at 49 days of age (Lane et al.,2000). Ruminal butyrate oxidation by the 84-day-old milk-fed lambs and 84-day-old lambsweaned at 49 days of age was ~75% of the rate observed with conventionally reared lambsweaned at 56 days of age (Baldwin and Jesse, 1992; Lane et al., 2000). The study of Lane et al.(2000) confirms the findings of Giesecke et al. (1979), and demonstrates that changes in glucoseand butyrate oxidation in developing rumen epithelium can occur independently of diet.

Prior to 42 days of age, ketogenesis from butyrate by isolated rumen epithelial cells, asmeasured by BHBA production, was relatively low in rumen epithelial cells from the milk-fedlambs, but increased thereafter to rates comparable to conventionally reared lambs (Baldwinand Jesse, 1992; Lane et al., 2000). Similar ketogenic rates were observed in rumen epithelialcells from the 84-day-old lambs weaned at 49 days of age. These results are consistent withthose obtained by Giesecke et al. (1979), who observed increasing ketogenic rates frombutyrate by rumen epithelial pieces from 1-week, 3–4-week, and 9–10-week-old lambs. Incontrast to Lane et al. (2000), ketogenesis by the rumen epithelial pieces from 9–10-week-oldmilk-fed lambs was about 75% of the rate found in lambs of the same age that had been rearedand weaned conventionally. Bush (1988) also noted increased rates of ketogenesis by rumenepithelial tissue with age from milk-fed calves. However, in that study ketogenesis by rumenepithelial tissue from conventionally reared and weaned calves was nearly 8-fold greater thanin milk-fed calves of the same age. Thus, all three of these studies indicate that ketogeniccapacity of rumen epithelium increases with age regardless of the diet consumed, although themagnitude of the reported increase did differ, perhaps due to species differences (Gieseckeet al., 1979; Bush, 1988; Lane et al., 2000). This again suggests that rumen epithelial metabolicdevelopment can occur in the absence of the rumen fermentation and VFA production, althoughdietary responses may modulate that ontogenic process.

3.4. Differential gene expression in rumen metabolic development

The mechanism responsible for the observed increase in ketogenesis by the rumen epitheliumfrom milk-fed lambs may be an increase in expression of the genes encoding ketogenicenzymes (Lane et al., 2002). Northern blots of total rumen epithelial RNA isolated fromconventionally reared and milk-fed lambs of different ages were probed with cDNA probesagainst AcAc-CoA thiolase and HMG-CoA synthase, the two enzymes that are generallyregarded as regulating ruminal ketogenesis (Leighton et al., 1983). In conventionally rearedlambs the relative abundances of AcAc-CoA thiolase and HMG-CoA synthase mRNA inrumen epithelium increased gradually (Lane et al., 2002), and generally paralleled thereported changes in ketogenesis in rumen epithelium from conventionally reared lambs(Baldwin and Jesse, 1992). Relative abundance of HMG-CoA synthase mRNA followed asimilar pattern to that of AcAc-CoA thiolase mRNA, but exhibited a sharper increase between42 and 49 days of age. Because this was the time when ketogenesis markedly increased in

B. W. Jesse400

rumen epithelium from conventionally reared lambs (Baldwin and Jesse, 1992), these datasuggest that expression of the HMG-CoA synthase gene may be the factor controlling rate ofketogenesis in the developing rumen epithelium.

In milk-fed lambs, changes with age of relative AcAc-CoA thiolase mRNA abundancewere similar to those observed in conventionally reared lambs. However, relative abundanceof HMG-CoA synthase mRNA in rumen epithelium from milk-fed lambs differed from thatobserved in conventionally reared lambs. The relative abundance of HMG-CoA synthasemRNA in rumen epithelium from milk-fed lambs remained relatively low through 42 days ofage, then exhibited an almost quantum jump between 42 and 49 days of age, remaining rela-tively high thereafter (Lane et al., 2002). Again, this time corresponds generally with theonset of marked ketogenesis in rumen epithelium from the milk-fed lambs. These datasupport the concept that rumen epithelial metabolic development can occur in the absence ofrumen fermentation (i.e. the dramatic jump in relative abundance of HMG-CoA synthasemRNA after 42 days of age in milk-fed lambs). Nevertheless, solid feed intake and the con-comitant production of rumen fermentation products (i.e. VFA) can modulate that process, asshown by the difference in change in relative abundance of HMG-CoA synthase mRNA inrumen epithelium between the conventionally reared and the milk-fed lambs. These data arealso consistent with the findings of Giesecke et al. (1979) and Bush (1988).

These results in rumen epithelium are consistent with the findings of other researchers inthe small intestine, where both ontogenic development and dietary induction of variousenzymes have been reported. For example, lactase activity decreases and dipeptidylpeptidaseIV activity increases in the lamb small intestine regardless of dietary treatment (conventionalrearing or maintenance on a milk diet), indicating ontogenic control of these enzymes(Shirazi-Beechey et al., 1991b). On the other hand, activity of the sodium-dependent glucosecotransporter in lamb intestine does change in response to dietary treatment (Shirazi-Beecheyet al., 1991a). Harmon et al. (1991) have reported increased acyl-CoA synthase activites foracetate, propionate, and butyrate in adult bovine rumen epithelium in response to increaseddietary energy intake. The response of ketogenic gene expression to solid feed intake inconventionally reared lambs may be the result of a mechanism similar to that resulting inincreased acyl-CoA synthetase activity in adult rumen epithelium, acting in conjunction withontogenic factors. The unique aspect of epithelial metabolic development in the neonatal rumen,especially ketogenesis, is the association of both ontogenic and physiological factors that appar-ently affect metabolic development by altering expression of the genes encoding ketogenic, andperhaps other metabolic, enzymes within the rumen epithelium (Lane et al., 2002).

A recent report found similar ontogenic and physiological effects on sodium and chloridetransport in developing calf rumen epithelium (Breves et al., 2002). Sodium transport byrumen epithelium increased with age of the calves, independently of dietary treatment (milk-fed or weaned onto solid feed). Chloride transport by rumen epithelium also increased withage of the calf, but exhibited a greater increase in those calves weaned onto solid feed than inthose maintained on a milk diet. Increased sodium and chloride transport by the developingrumen epithelium could reflect an increase in the VFA absorptive capacity by the rumenepithelium (Sehested et al., 1999b). These data provide further evidence of the importance ofontogenic events in the metabolic development of the rumen epithelium.


From the above discussion it should be clear that there are many unanswered questionsconcerning metabolic development in the neonatal rumen epithelium. Certainly the area of


substrate uptake in developing rumen epithelium needs to be addressed. To date, no informa-tion is available on specific transporters for glucose or VFA in developing rumen epithelium,and how the activity of those transporters changes during development. The recent discoveryof the SGLT1 glucose transport protein in mature rumen epithelium leads to the question ofwhen that transporter appears during rumen epithelial development (Zhao et al., 1998;Aschenbach et al., 2000a,b, 2002), and the role the transporter plays in glucose metabolismby the developing rumen epithelium. Similarly, given the importance of sodium and chloridein VFA transport by rumen epithelium (Sehested et al., 1999b), the recent discovery thatsodium and chloride transport increase during rumen epithelial development (Breves et al.,2002) is suggestive that VFA transport capacity may also increase during development. Theseissues should be addressed in the future. Related to VFA uptake is the issue of VFA activa-tion, which has been suggested to be the rate-limiting factor in VFA metabolism (Ash andBaird, 1973). No information is available to indicate when the acyl-CoA synthetases appearduring rumen epithelial development. Given the ability of neonatal rumen epithelial cells toutilize butyrate (Baldwin and Jesse, 1992), the acyl-CoA synthetases are likely to be presentsoon after, if not at, birth, but that needs to be determined.

The role of ontogeny in rumen epithelial metabolic development is only now being recog-nized for the importance it plays in this process. The question then arises as to how thatprocess is controlled. Further research into this area will certainly require the isolation ofgenomic clones encoding proteins that respond to dietary changes, e.g. structural proteinssuch as the small proline-rich proteins (Wang et al., 1996), as well as those that exhibit onto-genic patterns of development, such as HMG-CoA synthase (Lane et al., 2002). A comparisonof the regulatory regions of these genes should provide information about the transcriptionfactors potentially involved in regulating the expression of these genes. That information inturn could lead to identification of the signal transduction pathways that ultimately lead to theactivation of these genes. Various reports have noted the importance of agents such asbutyrate, insulin, and epidermal growth factor in stimulating the proliferation of rumenepithelial cells (Sakata et al., 1980; Galfi et al., 1991; Baldwin, 1999; Galfi and Neogrady,2001). Since the signal transduction pathways of some of these agents have been identified,this information should be helpful in establishing the mechanisms regulating gene expressionwithin the developing ruminal epithelium, and the interplay between physiological/dietaryfactors and ontogenic factors that result in complete rumen epithelial metabolic development.Ultimately a more complete characterization of the processes involved in rumen epithelialmetabolic development should lead to more effective management techniques in rearingyoung ruminants.

REFERENCES

Aafjes, J.H., 1967. Carbonic anhydrase in the wall of the forestomachs of cows. Brit. Vet. J. 123, 252–256.Alio, A., Theurer, C.B., Lozano, O., Huber, J.T., Swingle, R.S., Delgado-Elorduy, A., Cuneo, P.,

Deyoung, D., Webb, K.E. Jr., 2000. Splanchnic nitrogen metabolism by growing beef steers fed dietscontaining sorghum grain flaked at different densities. J. Anim. Sci. 78, 1355–1363.

Aschenbach, J.R., Bhatia, S.K., Pfannkuche, H., Gabel, G., 2000a. Glucose is absorbed in a sodium-dependent manner from forestomach contents of sheep. J. Nutr. 130, 2797–2801.

Aschenbach, J.R., Wehning, H., Kurze, M., Schaberg, E., Nieper, H., Burckhardt, G., Gabel, G., 2000b.Functional and molecular biological evidence of SGLT-1 in the ruminal epithelium of sheep. Amer.J. Physiol. Gastrointest. Liver Physiol. 279, G20–G27.

Aschenbach, J.R., Borau, T., Gabel, G., 2002. Glucose uptake via SGLT-1 is stimulated by beta2-adrenocep-tors in the ruminal epithelium of sheep. J. Nutr. 132, 1254–1257.

Ash, R., Baird, G.D., 1973. Activation of volatile fatty acids in bovine liver and rumen epithelium: evidence for control by autoregulation. Biochem. J. 136, 311–319.

B. W. Jesse402

Baldwin, R.L. VI, 1999. The proliferative actions of insulin, insulin-like growth factor-I, epidermalgrowth factor, butyrate and propionate on ruminal epithelial cells in vitro. Small Ruminant Res. 32,261–268.

Baldwin, R.L. VI, Jesse, B.W., 1992. Developmental changes in glucose and butyrate metabolism by isolated sheep ruminal cells. J. Nutr. 122, 1149–1153.

Baldwin, R.L. VI, McLeod, K.R., 2000. Effects of diet forage:concentrate ratio and metabolizable energyintake on isolated rumen epithelial cell metabolism in vitro. J. Anim. Sci. 78, 771–783.

Bergman, E.N., 1990. Energy contributions of volatile fatty acids from the gastrointestinal tract in various species. Physiol. Rev. 70, 567–590.

Breves, G., Zitnan, R., Schroder, B., Winckler, C., Hagemeister, H., Failing, K., Voigt, J., 2002. Postnataldevelopment of electrolyte transport in calf rumen as affected by weaning time. Arch. Tierernähr. 56,371–377.

Britton, R., Krehbiel, C., 1993. Nutrient metabolism by gut tissues. J. Dairy Sci. 76, 2125–2131.Bush, R.S., 1982. Extraction of enzymes and assessment of metabolism in bovine rumen epithelium.

Can. J. Anim. Sci. 62, 429–438.Bush, R.S., 1988. Effect of age and diet on in vitro metabolism in rumen epithelium from holstein calves.

Can. J. Anim. Sci. 68, 1245–1251.Bush, R.S., Milligan, L.P., 1971. Enzymes of ketogenesis in bovine rumen epithelium. Can. J. Anim. Sci.

51, 129–133.Chen, H., Wong, E.A., Webb, K.E. Jr., 1999. Tissue distribution of a peptide transporter mRNA in sheep,

dairy cows, pigs, and chickens. J. Anim. Sci. 77, 1277–1283.Gabel, G., Muller, F., Pfannkuche, H., Aschenbach, J.R., 2001. Influence of isoform and DNP on butyrate

transport across the sheep ruminal epithelium. J. Comp. Physiol. B 171, 215–221.Galfi, P., Neogrady, S., 2001. The pH-dependent inhibitory action of N-butyrate on gastrointestinal

epithelial cell division. Food Res. Int. 34, 581–586.Galfi, P., Neogrady, S., Sakata, T., 1991. Effects of volatile fatty acids on the epithelial cell proliferation

of the digestive tract and its hormonal mediation. In: Tsuda, T., Sasaki, Y., Kawashima, R. (Eds.),Physiological Aspects of Digestion And Metabolism In Ruminants: Proceedings of the SeventhInternational Symposium on Ruminant Physiology. Academic Press, New York, pp. 49–59.

Ganter, M., Bickhardt, K., Winicker, M., Schwert, B., 1993. Experimental studies of the pathogenesis ofrumen acidosis in sheep. Zbl. Veterinärmedizin A 40, 731–740.

Giesecke, D., Beck, U., Wiesmayr, S., Stangassinger, M., 1979. The effect of rumen epithelial develop-ment on metabolic activities and ketogenesis by the tissue in vitro. Comp. Biochem. Physiol. B 62,459–463.

Hamada, T., Maeda, S., Kameoka, K., 1976. Factors influencing growth of rumen, liver, and other organsin kids weaned from milk replacers to solid foods. J. Dairy Sci. 59, 1110–1118.

Hanigan, M.D., Crompton, L.A., Metcalf, J.A., France, J., 2001. Modelling mammary metabolism in thedairy cow to predict milk constituent yield, with emphasis on amino acid metabolism and milk pro-tein production: model construction. J. Theor. Biol. 213, 223–239.

Harmon, D.L., 1986. Influence of dietary energy intake and substrate addition on the in vitro metabolismof glucose and glutamine in rumen epithelial tissue. Comp. Biochem. Physiol. B 85, 643–647.

Harmon, D.L., Gross, K.L., Krehbiel, C.R., Kreikemeier, K.K., Bauer, M.L., Britton, R.A., 1991.Influence of dietary forage and energy intake on metabolism and acyl-CoA synthetase activity inbovine ruminal epithelial tissue. J. Anim. Sci. 69, 4117–4127.

Jesse, B.W., Solomon, R.K., Baldwin, R.L. VI, 1992. Palmitate metabolism by isolated sheep rumenepithelial cells. J. Anim. Sci. 70, 2235–2242.

Juhasz, B., Szegedi, B., Keresztes, M., 1976. The abomasal digestion in calves during development of theforestomachs. Acta Vet. Acad. Sci. Hung. 26, 281–295.

Klein, R.D., Kincaid, R.L., Hodgson, A.S., Harrison, J.H., Hillers, J.K., Cronrath, J.D., 1987. Dietaryfiber and early weaning on growth and rumen development of calves. J. Dairy Sci. 70, 2095–2104.

Lane, M.A., Jesse, B.W., 1997. Effect of volatile fatty acid infusion on development of the rumen epitheliumin neonatal sheep. J. Dairy Sci. 80, 740–746.

Lane, M.A., Baldwin, R.L. VI, Jesse, B.W., 2000. Sheep rumen metabolic development in response toage and dietary treatments. J. Anim. Sci. 78, 1990–1996.

Lane, M.A., Baldwin, R.L. VI, Jesse, B.W., 2002. Developmental changes in ketogenic enzyme geneexpression during sheep rumen development. J. Anim. Sci. 80, 1538–1544.

Leighton, B., Nicholas, A.R., Pogson, C.I., 1983. The pathway of ketogenesis in rumen epithelium of thesheep. Biochem J. 216, 769–772.


Liebich, H.G., Dirksen, G., Arbel, A., Dori, S., Mayer, E., 1987. Feed-dependent changes in the rumenmucosa of high-producing cows from the dry period to eight weeks post partum. Zbl.Veterinärmedizin A 34, 661–672.

Martens, H., Kudritzki, J., Wolf, K., Schweigel, M., 2001. No Evidence for active peptide transport inforestomach epithelia of sheep. J. Anim. Physiol. Anim. Nutr. (Berlin) 85, 314–324.

McLeod, K.R., Baldwin, R.L. VI, 2000. Effects of diet forage:concentrate ratio and metabolizable energyintake on visceral organ growth and in vitro oxidative capacity of gut tissues in sheep. J. Anim. Sci.78, 760–770.

Nocek, J.E., Herbein, J.H., Polan, C.E., 1980. Influence of ration physical form, ruminal degradablenitrogen and age on rumen epithelial propionate and acetate transport and some enzymatic activities.J. Nutr. 110, 2355–2364.

Noziere, P., Remond, D., Bernard, L., Doreau, M., 2000. Effect of underfeeding on metabolism of portal-drained viscera in ewes. Brit. J. Nutr. 84, 821–828.

Remond, D., Ortigues, I., Jouany, J.P., 1995. Energy substrates for the rumen epithelium. Proc. Nutr. Soc.54, 95–105.

Sakata, T., Hikosaka, K., Shiomura, Y., Tamate, H., 1980. Stimulatory effect of insulin on ruminal epithe-lium cell mitosis in adult sheep. Brit. J. Nutr. 44, 325–331.

Scaife, J.R., Tichivangana, J.Z., 1980. Short chain acyl-CoA synthetases in ovine rumen epithelium.Biochim. Biophys. Acta 619, 445–450.

Sehested, J., Diernaes, L., Moller, P.D., Skadhauge, E., 1999a. Ruminal transport and metabolism ofshort-chain fatty acids (SCFA) in vitro: effect of SCFA chain length and pH. Comp. Biochem.Physiol. A Mol. Integr. Physiol. 123, 359–368.

Sehested, J., Diernaes, L., Moller, P.D., Skadhauge, E., 1999b. Transport of butyrate across the isolatedbovine rumen epithelium: interaction with sodium, chloride and bicarbonate. Comp. Biochem.Physiol. A Mol. Integr. Physiol. 123, 399–408.

Shirazi-Beechey, S.P., Hirayama, B.A., Wang, Y., Scott, D., Smith, M.W., Wright, E.M., 1991a.Ontogenic development of lamb intestinal sodium-glucose co-transporter is regulated by diet. J. Physiol. 437, 699–708.

Shirazi-Beechey, S.P., Smith, M.W., Wang, Y., James, P.S., 1991b. Postnatal development of lamb intes-tinal digestive enzymes is not regulated by diet. J. Physiol. 437, 691–698.

Sutton, J.D., McGilliard, A.D., Richard, M., Jacobson, N.L., 1963. Functional development of rumenmucosa. II. Metabolic activity. J. Dairy Sci. 46, 530–537.

Tamate, H., McGilliard, A.D., Jacobson, N.L., Getty, R., 1962. Effect of various dietaries on the anatom-ical development of the stomach in the calf. J. Dairy Sci. 45, 408–420.

Wang, L., Baldwin, R.L. VI, Jesse, B.W., 1996. Identification of two cDNA clones encoding small proline-rich proteins expressed in sheep ruminal epithelium. Biochem. J. 317, 225–233.

Warner, R.G., Flatt, W.P., Loosli, J.K., 1956. Dietary factors influencing the development of the ruminantstomach. J. Agr. Food Chem. 4, 788–792.

Webb, K.E. Jr., Matthews, J.C., Dirienzo, D.B., 1992. Peptide absorption: a review of current conceptsand future perspectives. J. Anim. Sci. 70, 3248–3257.

Weigand, E., Young, J.W., McGilliard, A.D., 1975. Volatile fatty acid metabolism by rumen mucosa fromcattle fed hay or grain. J. Dairy Sci. 58, 1294–1300.

White, R.G., Leng, R.A., 1980. Glucose metabolism in feeding and postabsorptive lambs and maturesheep. Comp. Biochem. Physiol. A67, 223–229.

Young, J.W., Thorp, S.L., Delumen, H.Z., 1969. Activity of selected gluconeogenic and lipogenicenzymes in bovine rumen mucosa, liver and adipose tissue. Biochem. J. 114, 83–88.

Zhao, F.Q., Okine, E.K., Cheeseman, C.I., Shirazi-Beechey, S.P., Kennelly, J.J., 1998. Glucose transportergene expression in lactating bovine gastrointestinal tract. J. Anim. Sci. 76, 2921–2929.

B. W. Jesse404

405

17 Splanchnic carbohydrate and energymetabolism in growing ruminants1

N. B. Kristensena, G. B. Huntingtonb, and D. L. Harmonc

aDepartment of Animal Nutrition and Physiology, Danish Institute ofAgricultural Sciences, DK-8830 Tjele, DenmarkbDepartment of Animal Science, North Carolina State University, Raleigh,NC 27695-7621, USAcDepartment of Animal Sciences, University of Kentucky, Lexington,KY 40546-0215, USA

Ruminal fermentation precludes a simple description of nutrient availability based on nutrientintake. Thus, we must strive to understand the nutrient needs of the microflora and gut and thenevaluate nutrient availability after these needs have been met. Glucose is extensively metabolizedby gut tissues such that the net supply to the liver is often zero or negative. Despite this extensivemetabolism, small intestinal digestion can significantly increase glucose availability and metab-olism. Lactate is derived from the diet, from ruminal bacterial metabolism and from endogenousmetabolism. Because of its ubiquitous nature, lactate production from the gastrointestinal tractand viscera varies widely. However, lactate is a major glucose precursor in ruminants, supplying9–35% of hepatic glucose carbon. Short-chain fatty acids are the major currency of ruminantenergy metabolism, accounting for 45% of digestible energy intake. Significant quantities ofshort-chain fatty acids are metabolized by ruminal epithelium; however, it appears that in the fedruminant this epithelial metabolism is limited to butyrate and longer short-chain fatty acids.Estimates indicate that 5% of ruminally supplied propionate is metabolized by the rumen epithe-lium and 30% of arterially supplied acetate is metabolized by the portal-drained viscera. Thesefindings allow estimates of ruminal short-chain fatty acid production to be obtained from portalappearance of short-chain fatty acids corrected for portal-drained visceral metabolism of arterialshort-chain fatty acids and ruminal epithelial metabolism of butyrate.

1. INTRODUCTION

Compared with other mammals, ruminants could seem less efficient in capturing energy inthe form of body tissue, fetus, or milk. For example, a young pig on a nutritionally adequate



1Approved as publication No. 02-07-97 by the Kentucky Agricultural Experiment Station.

diet captures 76% of ME (metabolizable energy) as tissue gain (Dunkin and Campell, 1982)whereas a growing steer on a forage diet captures 46% or less of ME as tissue gain (Vargaet al., 1990). Does this indicate that ruminants are energetically inefficient? The answer to thisquestion is not as straightforward as might be indicated by its simplicity. First of all, the relativegrowth rates of the pig and the steer will affect the energetic efficiency, i.e. lower relative growthrate means that a relatively higher proportion of the total energy is used for maintenance.A quite different aspect is that interchanging the diets would have disastrous consequencesfor the performance of the young pig whereas the steer might do well though likely to sufferfrom overfeeding. The pig will not be able to obtain sufficient amounts of nutrients from aforage diet to obtain its potential growth rate.

These findings and the fact that ruminants utilize intravenously infused glucose as effi-ciently as nonruminants (Reid et al., 1980) indicate that the key to understanding bothpossibilities and limitations in ruminant nutrition and efficiency is related to the digestivestrategy of ruminants. The forestomach fermentation in ruminants implicates that there isonly an indirect relationship between the molecular composition of the feed and the actualnutrients available for absorption. The fermentation has a major influence on digestion andmetabolism of all organic dietary components, i.e. carbohydrate, protein, fat, and vitamins.Carbohydrates make up the largest fraction of almost any diet for functional ruminants andthe utilization of carbohydrate will therefore be of importance to both efficiency and per-formance. However, a number of controversies still exist connected to the availability ofcarbohydrate (starch) for postruminal digestion and absorption as well as quantitative rela-tionships between carbohydrate fermentation and end-product (short-chain fatty acid; SCFA)availability to the animal. The purpose of this chapter is to detail some of the unique aspectsof ruminant energy metabolism. Primarily, we aim to focus on the supply of glucose, lactate,and SCFA as sources of energy and their availability to body tissues. Only through a thoroughunderstanding of these interrelationships can we hope to predict and explain growth responsesbased on dietary inputs.

2. GLUCOSE

Because of pregastric fermentation much of the dietary carbohydrate is fermented to SCFA. Thisfermentation leaves little dietary carbohydrate available for absorption in the small intestine.Only when high-concentrate diets are fed are significant quantities presented to the smallintestine for absorption (Huntington, 1997). Thus, pregastric fermentation necessitates a con-tinual need for very high rates of gluconeogenesis (Bergman, 1973) to meet the glucose needsof the ruminant. The fermentation of dietary carbohydrates necessitates unique adaptations inruminant glucose metabolism and many of these adaptations have been detailed in someexcellent reviews (Leng, 1970; Bergman, 1973; Young, 1976); gluconeogenesis is also dis-cussed in a separate chapter within this book (see Chapter 15 by Donkin and Hammon). Weshall focus on how dietary influences affect the glucose economy of growing ruminants andon current information on the interorgan metabolism of glucose in ruminants.

3. SOURCES OF GLUCOSE

Blood glucose concentrations are typically 4–6 mM in most mammals; however; ruminantconcentrations tend to be lower, at 2–5 mM (Bergman, 1973). Despite low blood glucoseconcentrations and continual gluconeogenesis, ruminant blood glucose concentrationsare very responsive to intestinal carbohydrate digestion and absorption (Larson et al., 1956).

N. B. Kristensen et al.406

The major carbohydrate that ruminants consume in early life is lactose from milk. However,by 1–3 weeks of age ruminal fermentation is active and only through suckling will animalsachieve closure of the esophageal groove and bypass significant quantities of materials to theabomasum for gastric digestion (Orskov et al., 1970). With the increase in ruminal fermenta-tion there is a decline in the ability to digest lactose in the small intestine. Intestinal lactase(St. Jean et al., 1989) and glucose transport (Shirazi-Beechey et al., 1989) activities in the smallintestine decline after weaning.

Postweaning dietary carbohydrates contributing directly to glucose supplies include thevarious forms of α-linked glucose available in plants. Russell and Gahr (2000) described theclassification of food carbohydrates as occurring in four forms: (1) free (not associated withthe cellular structure), such as lactose in milk or fructose in honey; (2) intracellular, whichincludes soluble sugars and storage polysaccharides such as starch and fructans; (3) cell wallcomponents including cellulose, hemicellulose, pectins, and gums; and (4) chitin, a componentof the exoskeleton. For the functioning ruminant, only the intracellular storage polysaccharide,starch, contributes significantly to absorbed glucose. The remaining forms of food carbohydrateare first fermented to SCFA.

Huntington (1997) summarized numerous digestion experiments with starch intakes rangingfrom 1.5 to 10.6 kg/d. In these experiments, ruminal starch digestibility ranged from 94% to50%. The net result is that starch flow to the small intestine ranged from 90 to over 5000 g/d.These data demonstrate that starch intake can make a sizable contribution to the glucose needsof growing ruminants. However, to determine the contribution of starch intake to glucose avail-ability, the efficiency of small intestinal digestion must be known.

4. IMPACT OF INTESTINAL DIGESTION ON GLUCOSE SUPPLY

Several experiments have used animals fitted with hepatic portal vein and hepatic veincatheters to measure the quantity of glucose exiting the portal-drained viscera (PDV) andentering the liver (Huntington et al., 1989). This measurement provides a means of deter-mining the net glucose contributions to the liver or peripheral tissues and measures the sumof glucose absorption and metabolism. Across a wide range of experiments encompassingvaried diets, intakes, and physiological states, net glucose absorption is almost always zero ornegative (Reynolds et al., 1994). This is not to say that glucose is not being absorbed, butrather that very large amounts of glucose from the arterial supply are being metabolized suchthat the “net” result from absorption and metabolism is zero or negative. In a study designedto quantitate intestinal contributions to portal glucose supply, Huntington and Reynolds(1986) abomasally infused glucose and corn starch into heifers. Overall, they recovered anaverage of 65% of the glucose and 35% of the starch as glucose in portal blood. No differ-ences were observed for the amounts of glucose recovered from animals fed alfalfa hay or ahigh-concentrate diet at two intakes, suggesting little effect of adaptation for carbohydrateassimilation. Kreikemeier et al. (1991) fed steers alfalfa hay to minimize intestinal carbohy-drate supply and abomasally infused them with glucose, corn starch, or corn dextrins at20, 40, and 60 g/h. Infusions all lasted 10 h, with samples taken during the final 6 h. Glucoseinfusion resulted in 90% recovery of intestinal glucose disappearance in portal blood whereasonly 19% and 32% of the dextrin and starch intestinal disappearance were recovered in portalblood, respectively.

Factors such as microbial fermentation and gut tissue metabolism must certainly makea large contribution to small intestinal carbohydrate disappearance and emphasize the needfor measures of tissue metabolism and intestinal disappearance to more accurately describe

Splanchnic carbohydrate and energy metabolism 407

processes of digestion and absorption. The very high metabolic activity of the PDV tissueshas been shown to be a major factor in the apparently low net rates of glucose absorption(Reynolds and Huntington, 1988a). These authors (Reynolds and Huntington, 1988a,b) meas-ured directly the contribution of stomach and intestinal tissues to nutrient absorption in beefsteers. When steers were fed a concentrate diet, comparatively large amounts of glucose wereabsorbed from the intestines; however, the amounts utilized by ruminal and other stomachtissues were so great that the overall net PDV absorption was negative. Attempts were madein previous studies to account for this negative net glucose absorption and thereby obtain abetter estimate of net glucose absorption by including control (water) infusions (Kreikemeieret al., 1991). However, more recent work has shown that increasing the peripheral supply ofcarbohydrate, either through intraduodenal or intrajugular infusion of glucose, increases themetabolism of arterially supplied glucose by the PDV (Balcells et al., 1995), making thesecorrections tenuous at best.

5. DIET EFFECTS ON GLUCOSE METABOLISM

In previous sections we have attempted to define relationships of intestinal supply andglucose availability. However, it needs to be clearly pointed out that the major determinant ofglucose supply is dietary energy intake (Herbein et al., 1978). Experiments assessing whole-body glucose metabolism have clearly shown that glucose irreversible loss, a measure of theflow of glucose through the body pool never to return, and thus, at steady state, an indicatorof glucose production, is a function of digestible energy intake. Schmidt and Keith (1983)tested this hypothesis using steers fed 70% corn vs. 70% alfalfa diets fed at equal energyintakes. They demonstrated that when steers were fed at equal energy intakes, glucoseirreversible loss was equal. When dry matter intakes were equalized, glucose irreversible losswas greater for the 70% corn diet because of the greater energy intake with the corn. In arelated study (Russell et al., 1986) it was demonstrated that glucose irreversible loss wasdirectly related to energy intake independent of body size in steers ranging in weight from136 to 470 kg.

These relationships depend on the tight control between digestible energy intake and glu-coneogenesis. Organic matter fermented in the rumen will supply glucose precursors,primarily propionate, to meet the glucose needs of the host. These relations are borne out inthe work of Van Maanen et al. (1978), who determined ruminal propionate production andglucose irreversible loss in steers fed forage and grain-based diets with the propionate-enhancing antibiotic, monensin. Monensin increased ruminal propionate production by 49%on the forage diet and by 76% on the grain diet. Associated with these increases in propionatewere increases in glucose irreversible loss of 7% and 16% for the forage and grain diets,respectively. This study shows that increasing propionate supply can increase glucose irreversibleloss, but not in direct proportion. These results were similarly borne out by Seal and Parker(1994) using intraruminal infusion of propionate in calves. Only at their highest propionateinfusion (1 mol/d) was glucose irreversible loss increased. Interestingly, ruminal propionateinfusion decreased PDV glucose use from 28% to 11% of glucose irreversible loss.

The relationships between dietary energy intake and glucose irreversible loss depend upontwo related assumptions: (1) ruminants have a very tight control of hepatic glucose production,and (2) digestion and absorption of starch in the small intestine contributes little to glucoseirreversible loss in these studies because they are dependent on glucose derived from theproducts of ruminal fermentation. Bauer et al. (1995) infused phlorizin, a potent inhibitor ofSGLT1, into the abomasum of steers and sheep and demonstrated that when glucose active


transport was inhibited, hepatic glucose production increased resulting in no change in totalsplanchnic glucose output. This demonstrates that glucose production is well coordinatedbetween the PDV and the liver.

A different approach was used in the study by Harmon et al. (2001). They infused a partiallyhydrolyzed starch solution either ruminally or abomasally in growing steers. Shifting the siteof starch digestion from the rumen to the small intestine increased glucose utilization by PDVtissues (132%), PDV glucose flux (310%), and irreversible loss of glucose (59%). Abomasalinfusion resulted in greater total energy availability (28%) from the total splanchnic tissues.Thus, shifting starch digestion to the small intestine increases PDV glucose uptake and uti-lization without a corresponding decrease in hepatic glucose production. This shift results ingreater glucose supplies to the periphery. This would seem in contrast to the results ofHerbein et al. (1978), who related glucose irreversible loss solely to energy intake. These rela-tionships may not hold if significant quantities of starch are digested and absorbed in thesmall intestine. Balcells et al. (1995) infused sheep jugularly with glucose and found that glu-cose irreversible loss increased over 2-fold. Accompanying this increase in systemic glucoseavailability was an increased utilization of glucose by the PDV. However, in their experiment,the fraction of whole-body glucose used by the PDV remained constant (30% of whole-bodyglucose irreversible loss) despite the increase in glucose irreversible loss. These results arein agreement with their later work (Cappelli et al., 1997) where sheep received exogenousglucose either intrajugularly or intraduodenally. Supplying glucose by either route increasedwhole-body glucose irreversible loss and portal glucose utilization, and again, portal glucoseutilization was approximately 30% of glucose irreversible loss.

These results suggest that the fraction of whole-body glucose irreversible loss used bythe PDV is relatively constant. However, both of these studies were relatively short-term,lasting 6 to 8 h. They do not answer whether or not long-term exposure causes tissues to adaptand use more or less of the available glucose. In the study by Harmon et al. (2001) theyinfused a partially hydrolyzed starch solution either ruminally or abomasally in growingsteers for 7 days. In their study, portal glucose utilization was 23% of whole-body glucoseirreversible loss with the ruminal infusion and this increased to 34% when the carbohydratewas infused abomasally. Thus, despite a 58% increase in glucose irreversible loss, therewas a concomitant increase in the fraction of glucose metabolized by PDV tissues. It is notknown if this increase in metabolism was the result of tissue adaptation or simply differencesin cattle and sheep. With the ruminal infusion an increase in metabolism could reflect moreenergy available as SCFA resulting in less PDV glucose use, as was seen with the ruminalpropionate infusions of Seal and Parker (1994) described above. A decrease in net PDVglucose use has also been reported for steers fed 450 g/d sodium propionate (Harmon andAvery, 1987).

McLeod et al. (2001) used the ruminal/postruminal infusion of carbohydrate modeldescribed above (Harmon et al., 2001) to study energy balance in growing steers. Theyreported that abomasal infusion of carbohydrate increased retained energy; however, based oncalorimetric data, the energy retained was retained solely as fat. When combined, these resultssuggest that an increased availability of glucose increases the energetic efficiency and PDVmetabolism of glucose, but this may also result in greater fat deposition. One could speculatethat increased circulating glucose results in increased insulin and increased fat deposition.Others have suggested that there are specific effects of glucose on lipogenesis in ruminants.Pearce and Piperova (1984) compared duodenal infusions of glucose and dextrins in sheepand found that glucose infusion increased in vitro lipogenesis from acetate nearly 7-fold insubcutaneous adipose tissue as compared with control (noninfused) sheep.


6. DIETARY AND DIGESTIVE SOURCES OF LACTATE

Lactate entering portal blood of the gastrointestinal tract of a ruminant can come from thediet, can be a product of rumen fermentation, or can be a product of tissue metabolism.Dietary sources generally include lactate from fermented feeds, e.g., a product of lactobacilliin silages. Lactate is produced in fermented feeds by homo- or hetero-fermenting lactobacillithat vary in substrate (sugar) preferences and isomer of lactic acid produced. Most bacteriacan produce either D(−)- or L(+)-lactic acid by virtue of isomer-specific lactate dehydrogenaseand lactate racemase enzyme activity (Counotte and Prins, 1981; McDonald et al., 1991).

Lactic acid concentrations in most silages prepared by adequate or competent techniquesrange from 3 to 12 g/kg DM. Treatments that limit fermentative activity, e.g. treatment withmineral acids, formic acid, or formalin, or wilting before ensiling, can reduce lactic acid con-centration by one-half or more. Treatments that induce or enhance fermentative activity in thesilo, e.g. inoculation with bacteria, addition of sugars or propionic acid, decreased particlesize by precision chopping of herbage before ensiling, in general increase lactic acid concen-tration 1.5–2.0-fold (McDonald et al., 1991; Sheperd et al., 1995; Kung et al., 2000; Kung andRanjit, 2001). The isomeric proportions of lactic acid in these feedstuffs have not been studiedextensively; available reports indicate that L(+):D(−) ratios range from 0.3:1 to 1:1 (Schaadt,1968; Hull, 1996; Kung et al., 2000). McDonald et al. (1991) suggested that as time of ensilingincreases, the L(+):D(−) ratio approaches 1:1 because of racemase activity of lactobacilli.

Lactate is both produced and used by ruminal microbes. Numbers (and activity) of lactateproducers and users respond rapidly to readily fermentable substrate (Counotte and Prins,1981; Goad et al., 1998), which means that ruminal lactate concentrations usually are verylow (1–3 mM) to nondetectable. Calculations of lactate production in the rumen are in asimilar range, 1–3 mmol/h (Counotte and Prins, 1981). In cases of abrupt changes in intakeof readily available carbohydrates there can be a rapid increase in ruminal lactate concentra-tions, indicating that production can exceed use or removal from the rumen. For example,Harmon et al. (1985) dosed beef steers intraruminally with 12 g of glucose per kg of bodyweight and measured peak concentrations of L(+)- lactate and D(−)-lactate of 77 and 40 mM,respectively, 30 h after the dosing. As a result of rapid fermentation of the carbohydrates, theproportion of L(+):D(−)-lactate may change from predominantly L(+) to predominantly D(−).The change in isomeric ratio is more a function of increased production than differences inuse rates, because both isomers are used by ruminal microbes at similar rates. The rapid pro-duction and accumulation causes a ruminal acidosis that is lethal to many ruminal protozoa,and also causes a systemic acidosis in the host ruminant (Dunlop, 1972; Counotte and Prins,1981; Goad et al., 1998). Ruminal concentrations and isomeric proportions of lactate are theproduct of the effects of ruminal production, use, absorption from the rumen, and passagewith digesta to more distal portions of the gastrointestinal tract.

7. ABSORPTION OF LACTATE FROM THEGASTROINTESTINAL TRACT

L(+)-lactate (and presumably D(−)-lactate) are transported across cell membranes by a familyof monocarboxylate transporters (Price et al., 1998). These transporters also transportketones, pyruvate, and acetate. Because lactate can be a product of tissue metabolism, a sub-strate for tissue metabolism, and the subject of transport across the plasma membrane ofepithelial cells, it is difficult to discern the relative importance of, or interactions among, theseprocesses on the rate of lactate appearance in portal blood draining the gastrointestinal tract.


Further, dietary and endogenous factors that alter blood flow can negate or amplify in vivochanges in concentration differences in blood supplying and draining the PDV. The few dataavailable for absorption of D(−)-lactate suggest that factors that promote production ofD(−)-lactate in the rumen also promote its absorption and appearance in hepatic portal blood(Huntington et al., 1980, 1981; Harmon et al., 1985). For the remainder of this discussion oflactate absorption and metabolism, “lactate” and “L(+)-lactate” will be used synonymouslyunless otherwise indicated.

The studies summarized in table 1 are representative of published literature that quantifiesnet flux of lactate across splanchnic tissues. The studies show that lactate absorption in sheepand cattle ranges from approximately 2 to 200 mmol/h. Increased intake of a given dietincreases net absorption (Reynolds et al., 1991; table 1), as does increased body mass (usuallyaccompanied by increased intake), albeit at a nonlinear rate (Eisemann et al., 1996; table 1).The data from Taniguchi et al. (1995; table 1) exemplify the positive relationship betweenincreased ruminal fermentation and lactate absorption (alfalfa vs. alfalfa and ruminal starchinfusion in table 1), and also indicate that increased intestinal appearance of glucose resultsin increased portal appearance of lactate, ostensibly as a result of postruminal gut tissuemetabolism (ruminal vs. abomasal infusion of starch in table 1). The postruminal digestivetract accounted for about one-third of lactate absorption in beef steers fed alfalfa hay or ahigh-concentrate diet (Reynolds and Huntington, 1988b). The lactating dairy cows in thestudies in table 1 had similar daily dry matter intakes (data not shown), but the cows eatingthe grass diet absorbed less lactate than the cows eating corn silage and supplement (Reynoldset al., 1991; De Visser et al., 1997; table 1). McLeod et al., (1997) (table 1) found that infusionof somatostatin decreased blood flow through PDV of sheep, but increased venoarterial dif-ference of lactate (data not shown), resulting in increased net absorption of lactate. The studyof Bauer et al. (1995; table 1) included intragastric infusion of phlorizin, which decreased netabsorption of glucose (data not shown) but had no statistically significant effect on lactateflux. Other examples of lack of effects of metabolic regulators include similar net absorptionof lactate in control beef steers vs. steers fed a β-adrenergic agonist (Eisemann and Huntington,1993) or control steers vs. hyperinsulemic, euglycemic beef steers receiving intravenous infusionof insulin and glucose (Eisemann and Huntington, 1994).

Lactate makes a small but measurable contribution to the overall energy supply for rumi-nants. Lactate accounted for approximately 4.3% of the sum of energy absorbed as SCFA andlactate by lactating dairy cows consuming all-forage diets (De Visser et al., 1997; table 1), 8%by lactating dairy cows consuming a 60:40 corn silage:supplement diet (Reynolds et al., 1991;table 1), 9% by steers consuming all-forage diets (Huntington et al., 1988), and 16% by heifersconsuming a diet containing 780 g corn grain/kg of DM (Huntington and Prior, 1983).

8. HEPATIC METABOLISM OF LACTATE

The metabolic importance of lactate for ruminants centers on its role as a glucose precursorin the liver; net lactate removal by the liver often exceeds portal supply (table 1) and can the-oretically account for 9–35% of net hepatic glucose release (data not shown) in studies withbovines listed in table 1. Studies with infusions of radiolabeled glucose and lactate into lambsand steers indicate that from 5% to 11% of glucose carbon comes from L(+)-lactate, and lessthan 1% comes from D(−)-lactate (Huntington et al., 1980, 1981; Harmon et al., 1983).Recycling of carbon through lactate and glucose would cause underestimations from isotopeinfusions, and calculations from net fluxes likely overestimated the true conversion of lactateto glucose. In the sheep studies of McLeod et al. (1997; table 1) net lactate removal could


Tabl

e 1

Sele

cted

stu

dies

of

L(+

)-la

ctat

e fl

uxa

acro

ss p

orta

l-dr

aine

d vi

scer

a (P

DV

), li

ver,

and

tot

al s

plan

chni

c (T

SP)

tiss

ues

of s

heep

and

cat

tle

Net

flu

x, m

mol

/h

Spec

ies

BW

, kg

Die

t des

crip

tion

PDV

Liv

erT

SPR

efer

ence

Shee

p w

ethe

rs36

Alf

alfa

hay

, duo

dena

l sta

rch

and

case

in in

fusi

on4.

8−9

.8−5

McL

eod

et a

l. (1

997)

Shee

p w

ethe

rs36

Alf

alfa

hay

, duo

dena

l sta

rch

and

case

in in

fusi

on,

6.6

−10.

6−4

McL

eod

et a

l. (1

997)

som

atos

tatin

inje

ctio

nSh

eep

wet

hers

40A

lfal

fa h

ay, s

tarc

h in

fusi

on2.

23.

25.

4B

auer

et a

l. (1

995)

Bee

f he

ifer

s32

1A

lfal

fa:c

once

ntra

te, l

ow in

take

45−1

728

Rey

nold

s et

al.

(199

1)B

eef

heif

ers

321

Alf

alfa

:con

cent

rate

, hig

h in

take

82−2

854

Rey

nold

s et

al.

(199

1)B

eef

stee

rs23

6B

rom

egra

ss h

ay:c

once

ntra

te 6

0:40

47−8

1−3

4E

isem

ann

et a

l. (1

996)

Bee

f st

eers

438

Bro

meg

rass

hay

:con

cent

rate

60:

4067

−101

−34

Eis

eman

n et

al.

(199

6)B

eef

stee

rs52

2B

rom

egra

ss h

ay:c

once

ntra

te 6

0:40

63−8

5−2

2E

isem

ann

et a

l. (1

996)

Bee

f st

eers

253

Alf

alfa

hay

39−6

3−2

4Ta

nigu

chi e

t al.

(199

5)B

eef

stee

rs25

3A

lfal

fa h

ay, r

umin

al s

tarc

h in

fusi

on50

−77

−27

Tani

guch

i et a

l. (1

995)

Bee

f st

eers

253

Alf

alfa

hay

, abo

mas

al s

tarc

h in

fusi

on75

−68

−7Ta

nigu

chi e

t al.

(199

5)L

acta

ting

dair

y co

ws

645

Cor

n si

lage

:sup

plem

ent 6

0:40

216

−249

−33

Rey

nold

s an

d H

untin

gton

(19

88c)

Lac

tatin

g da

iry

cow

s50

0Fr

esh

ryeg

rass

121

−144

−23

De

Vis

ser

et a

l. (1

997)

a Pos

itive

num

bers

indi

cate

net

abs

orpt

ion

or r

elea

se, n

egat

ive

num

bers

indi

cate

upt

ake

or r

emov

al.

maximally account for 41–62% of net hepatic glucose production. Lactate contribution is notcalculated for the data of Bauer et al. (1995; table 1) because in some of their treatments theymeasured net hepatic output of lactate.

The range of these potential hepatic fluxes and potential contribution to hepatic gluconeo-genesis attest to the flexibility and versatility of lactate to participate in postabsorptivemetabolism. The complete data of Reynolds et al. (1991; table 1) for beef heifers showed aninteraction between intake level and percentage of dietary concentrate; net hepatic lactateremoval and potential contribution of lactate to gluconeogenesis increased when the heifers’intake of a high-forage diet increased. However, lactate removal and potential contribution togluconeogenesis decreased when the heifers’ intake of a high-concentrate diet increased.A mesenteric vein infusion of alanine in the same heifers (Reynolds and Tyrrell, 1991) increasednet alanine removal and reduced net lactate removal by the liver, but did not affect net hepaticglucose output. These results indicate a replacement of lactate by alanine as a glucoseprecursor. The complete data of Eisemann et al. (1996; table 1) predict decreased net hepaticremoval or extraction of lactate, and increased net hepatic removal of amino acids to supportincreased hepatic glucose production in beef steers as they grow from 235 to 525 kg of bodyweight. The somatostatin injection that increased net portal absorption of lactate in sheep alsoincreased net hepatic removal of lactate and increased glucose output by the liver (McLeodet al., 1997; table 1). Steers fed a β-adrenergic agonist had an acute surge in lactate removal bythe liver that could account for up to 63% of liver glucose output on the first day of treatment.Hepatic removal and potential contribution to gluconeogensis subsided after 7 days of treatment(Eisemann and Huntington, 1993).

9. PERIPHERAL METABOLISM OF LACTATE

Circulating concentrations of L(+)-lactate range from 0.2 to 1.0 mM, and concentrations ofD(−)-lactate are 0.10 to 0.50 of concentrations of L(+)-lactate (Huntington et al., 1980, 1981;Harmon et al., 1983); these studies are cited in table 1. Whole-body lactate turnover in beefcattle and sheep ranges from approximately 5 to 10 times net portal absorption (Huntingtonet al., 1980, 1981), indicating the importance of the Cori cycle in movement of carbon throughlactate and glucose between the liver and peripheral tissues, mostly muscle. Excitement oragitation of animals can cause a rapid rise in blood lactate levels as a result of heightenedmuscle activity. The major fate of D(−)-lactate is oxidation, which accounted for essentiallyall D(−)-lactate turnover in steers (Harmon et al., 1983). In vitro studies with bovine tissuesshow significant potential for oxidation of D(−)-lactate, with the greatest activity in kidneycortex followed by heart and liver, the lowest activity being detected in muscle tissue(Harmon et al., 1984).

Net flux of L(+)-lactate across hindlimbs of cattle varies in response to physiological stateof the animal and physiological interventions by researchers. As stated previously, lactateinteracts with glucose through the Cori cycle, but lactate also is used as a substrate for lipidsynthesis. Therefore, depending on the contribution of fat to tissue makeup, the hindlimbsmay be net users or net releasers of lactate (Prior et al., 1984; Eisemann et al., 1996). Theacute response of beef steers to an orally administered β-adrenergic agonist was a dramaticincrease in lactate production by hindlimbs which was not evident after 7 days of treatment(Eisemann and Huntington, 1993). Establishment of hyperinsulemia with euglycemia in steersenhanced glucose uptake by hindquarters, but did not significantly change lactate flux acrossthose tissues (Eisemann and Huntington, 1994).


10. SHORT-CHAIN FATTY ACIDS OVERVIEW

Short-chain fatty acids are simple aliphatic carboxylic acids with straight or methyl-branchedhydrocarbon chains of 2 to 5 carbons. The SCFA anions with 2 (acetate), 3 (propionate), and4 (butyrate) carbons are the most prevalent SCFA in the rumen and colon (Bergman, 1990)and their production is closely related to the energy metabolism of rumen microbes (Russelland Wallace, 1988). The existence of acetate in the rumen was observed more than a hundredyears ago; however, not until the 1940s was it discovered that SCFA are absorbed from theforestomachs and make a significant contribution to ruminant metabolism (Barcroft et al.,1944). About 67% of ruminal SCFA are absorbed across the rumen epithelium or taken up bythe rumen microbes and about 33% are carried out of the rumen by liquid passage (Peterset al., 1990). Short-chain fatty acids leaving the rumen with liquid outflow are absorbed mainlyin the omasum and abomasum (Masson and Phillipson, 1952; Rupp et al., 1994).

In ruminants, as in other animals, a mixture of undigested feed and organic matter ofendogenous origin enters the hindgut and is fermented into gasses, SCFA, and microbialorganic matter. Fermentation in the hindgut is of little quantitative nutritional importance tothe animal compared to the forestomach, mostly because microbial protein and other non-SCFA products of fermentation are not readily absorbed. The SCFA production in the hindgutcan be estimated as 6–13% of the total gut production based on the propionate appearanceacross mesenteric drained tissues compared to the total PDV net appearance (Reynolds andHuntington, 1988b). Studies based on isotopic dilution in the rumen and cecum have yieldedsimilar relative production rates (12%) between forestomach and hindgut (Siciliano-Jonesand Murphy, 1989). Therefore, forestomach fermentation is quantitatively the most importantfermentation in ruminants, and most focus is given to forestomach physiology. However, itmust be kept in mind that total gut production of SCFA does contain a hindgut component.

11. TRANSPORT BY NONIONIC DIFFUSION

The rumen is lined with a keratinized stratified squamous epithelium. The epithelium is aheterogeneous structure with a physical barrier formed by keratinized cells facing the lumen.The chemical barrier of the epithelium is below the keratinized cells. The majority of metabolicactivity is located in the basal cells as indicated by their high concentration of mitochondria(Steven and Marshall, 1970; Henrikson and Stacy, 1971).

Weak electrolytes, a group to which SCFA belong, can pass biological membranes via non-ionic diffusion; the resulting unidirectional flux is a function of concentration (activity) andsolubility in the membrane (Rechkemmer, 1991). In accordance with this theory, it has beenshown in vivo (Thorlacius and Lodge, 1973) as well as in vitro (Sehested et al., 1999b) thatthe unidirectional flux rate of butyrate across rumen epithelium increases with decreasing pH.However, the lack of proportionality between concentration of protonized acids and acetateand propionate fluxes as well as a relatively high permeability of these acids comparedto longer-chain fatty acids has been seen as a challenge for the absorption theory based onnonionic diffusion. Nevertheless, a generally observed phenomenon is that SCFA have a relativelyhigh permeability to biological membranes relative to longer-chain fatty acids (Dietschy,1978). This means that the membranes behave as rather polar structures toward small solutessuch as SCFA. The relative absorption rates of SCFA from experiments with washed reticulo-rumens show that absorption rates of fatty acids longer than butyrate increase with increasedchain length (pH 7), and that methyl-branched SCFA (isobutyrate and isovalerate) have lowerabsorption rates than their corresponding straight-chain fatty acids (Oshio and Tahata, 1984;


Kristensen et al., 2000a). Although the membranes of the rumen epithelium apparently havea relatively high permeability to acetate, propionate, and butyrate, it still makes sense todescribe their absorption as regulated by mass action (as long as we consider unidirectionalmembrane fluxes of SCFA). Recent work has suggested that anion exchangers may contributeto apical SCFA fluxes in rumen epithelium even though the quantitative importance is unknown(Kramer et al., 1996). So far, the data available on SCFA absorption from the forestomachseem to indicate that absorption of SCFA by diffusion can account for the quantitatively mostimportant SCFA absorption.

12. CARRIER-MEDIATED TRANSPORT IN RUMEN EPITHELIUM

Rumen epithelium mounted in Ussing chambers has consistently shown a remarkable differencein the net transport of butyrate compared with the net transport of acetate and propionate (Stevensand Stettler, 1967; Sehested et al., 1999a). While the rumen epithelium shows a small net secre-tion (net transport from blood to lumen side of the isolated epithelium) of acetate and propionatewhen epithelia are incubated without an electrochemical gradient of SCFA, a relatively large netabsorption of butyrate carbon usually occurs. The secretion of acetate and propionate by theepithelium at first seems to argue against the concept of nonionic diffusion. However, most esti-mates of SCFA flux in vitro have been based on 14C-labeled acids, implying that release of anysubstance carrying carbon from SCFA will be interpreted as SCFA flux. A small proportion ofacetate and propionate transported across the epithelium will be oxidized under these conditionsand the epithelium has been shown to primarily excrete the CO2 on the luminal side, explainingat least partly the net excretion of these acids (Sehested et al., 1999a).

The rumen epithelium has long been known to be capable of metabolizing SCFA and, inparticular, to have high affinity and capacity for metabolism of butyrate (Pennington, 1952).This in fact is the key to explaining the differences in the epithelial transport of butyratecompared with acetate and propionate. The metabolism of butyrate into acetoacetate and3-hydroxybutyrate and the subsequent release of these compounds across the basolateralmembrane would be in agreement with the apparent normal metabolic activity of the epithe-lium and would also explain why [14C]-butyrate was transported differently from acetate andpropionate. It is likely that the products of butyrate metabolism are transported to the serosal(blood side) buffer carrying the label from butyrate. Keto- and hydroxyacids such as acetoac-etate, 3-hydroxybutyrate, and lactate are more polar than SCFA because of their hydrophilic,secondary functional group, and consequently these acids have a lower permeability in bio-logical membranes. In skeletal muscle a monocarboxylate transporter which co-transportslactate and protons solves an analogous transport problem for lactate across the cell membrane(Juel, 1997). The missing piece of the puzzle would therefore be to find monocarboxylatetransporters in the epithelium that enable polarized transport of acetoacetate and 3-hydroxy-butyrate. Recently, this transporter was shown to be present in rumen epithelium which agreeswith this sequence of events (Müller et al., 2001). It has also been shown that blocking cellularmetabolism abolishes the active component of butyrate absorption in vitro (Gäbel et al., 2001),confirming that it is the ketone bodies formed from butyrate that are selectively transportedto the serosal side of the epithelium and not butyrate itself.

13. RUMEN EPITHELIAL METABOLISM

One of the central observations on SCFA metabolism in ruminants has been the apparentlyextensive metabolism of ruminally produced SCFA by the rumen epithelium. However, this


has been among the most difficult features of SCFA metabolism to understand. Numerousreviews are available discussing SCFA metabolism (Bergman, 1990; Britton and Krehbiel,1993; Seal and Reynolds, 1993; Rémond et al., 1995; Kristensen et al., 1998; Seal and Parker,2000). Recent studies have challenged the view that the rumen epithelium has a dominant rolein the metabolism of acetate and propionate absorbed from the rumen.

The classic attempt to determine the quantitative relationship between SCFA production inthe gut and SCFA absorption was the work by Bergman and Wolff (1971). The production ofSCFA in the rumen based on isotopic dilution was compared with portal appearance of SCFAcorrected for PDV uptake of arterial acetate. It was concluded that large amounts not only ofbutyrate, but also of acetate and propionate, were metabolized by gut epithelia. In support ofthis conclusion, rumen epithelium also seemed to metabolize a large fraction of SCFA trans-ported in vitro (Stevens, 1970). Nevertheless, these figures have long been doubted whenconsidering the large amounts of SCFA apparently being absorbed in high producing rumi-nants (Sutton, 1985). These figures also lead to the paradoxical conclusion that the rumenepithelium of a lactating cow should have oxidative needs comparable to the entire fastingheat metabolism of the animal (Kristensen and Danfær, 2001).

Studies on rumen epithelial metabolism of absorbed SCFA may have overestimated themetabolism by the epithelium because the actual estimation is the mixed effect of rumenmicrobial and rumen epithelial metabolism. Studies on SCFA absorption under washed reticulo-ruminal conditions that minimize bacterial activity have shown that the portal appearance ofacetate, propionate, and isobutyrate could account for the entire disappearance of these acidsfrom the rumen when the PDV uptake of arterial acetate is taken into account and 5% ofthe propionate is assumed metabolized into lactate by the rumen epithelium (Kristensenet al., 2000a). Butyrate was also extensively metabolized by the rumen epithelium underwashed reticulo-rumen conditions and no more than 23% of the butyrate disappearance fromthe rumen could be accounted for by portal appearance of butyrate. It has previously beenobserved that there is increasing portal recovery of butyrate with increasing disappearancerates of butyrate from the rumen of sheep (Kristensen et al., 1996b, 2000b; Nozière et al.,2000). This effect is in agreement with a saturable metabolic capacity of the epithelium.

To what extent there is interspecies differences in the metabolic capacity of butyrate in therumen epithelium is not yet clear, but in a study with steers, the portal recovery of butyratedid not increase with increasing ruminal infusion rates of butyrate (Krehbiel et al., 1992).The recovery was relatively high at all infusion levels in the steers (25%), and was equivalentto the highest recovery level obtained in the sheep experiments. In sheep, increasing ruminalbutyrate infusion not only leads to increasing portal recovery of butyrate, but also to increas-ing portal recovery of ruminal valerate (Kristensen et al., 2000b). These results point to a redefinition of the role of the rumen epithelium in SCFA metabolism and suggest that therumen epithelium is not metabolizing large amounts of acetate and propionate as previouslyassumed.

14. IS BUTYRATE OXIDIZED TO CARBON DIOXIDEDURING ABSORPTION?

In vitro studies have shown that rumen epithelium is able to oxidize all of the three quantita-tively most important SCFA (Baldwin and McLeod, 2000); however, the epithelial productionof 3-hydroxybutyrate and acetoacetate imply that butyrate oxidation is far lower than itsdisappearance across the epithelium. Studies comparing net portal appearance of butyrateand butyrate infusion into the rumen have indicated that major parts of the butyrate were


oxidized (lost), because net portal appearance of butyrate, 3-hydroxybutyrate, and acetoac-etate accounted only for 25–45% of ruminal butyrate infusion (Krehbiel et al., 1992;Kristensen et al., 1996b). However, the PDV has been shown to utilize 3-hydroxybutyratefrom arterial blood equivalent to 32–42% of the whole-body flux in sheep and thereby maskthe true production rate by the gut epithelia (Kristensen et al., 2000c). Intraruminal microbialpathways might also utilize part of the infused butyrate and thereby contribute to what couldbe interpreted as epithelial oxidation. This latter effect has been indicated by relatively highrecoveries (as compared to expected recovery in the fed animal) of butyrate when infused intoanimals maintained under total intragastric nutrition (Gross et al., 1990a) or temporarilywashed reticulo-rumen conditions (Kristensen et al., 2000a). In conclusion, the rumen epithe-lium has oxidative needs and butyrate is likely the most important carbon source. Themajority of the butyrate absorbed is released as butyrate, acetoacetate, and most importantly,3-hydroxybutyrate to the portal blood.

15. WHY DO EPITHELIA METABOLIZE BUTYRATE?

Butyrate is generally considered a special metabolite for gut epithelial function (Topping andClifton, 2001). One way to explain the special behavior of gut epithelia toward butyrate com-pared with acetate and propionate is that butyrate is important as an energy source forepithelial cells (Bugaut, 1987). However, the rumen epithelium has a range of other metabo-lites available, e.g. acetate and propionate absorbed from the rumen as well as arteriallysupplied glucose. One might speculate that butyrate’s role as an important substrate forepithelial energy metabolism might have evolved secondary to the basic need of havingbutyrate removed before it enters the blood stream. Butyrate metabolism by rumen andhindgut epithelia could therefore be seen as a protective mechanism that has two disposalpathways, oxidation and ketogenesis. It is obvious that butyrate is handled differently fromacetate and propionate by the epithelia (Pennington, 1952), but another question remains tobe answered: is butyrate a unique metabolite? Valerate, for example, is also efficiently metab-olized by the rumen epithelium (Kristensen et al., 2000a,b) and it has been shown that theepithelium have the capacity to metabolize medium-chain (Hird et al., 1966) as well as long-chain fatty acids (Jesse et al., 1992).

Butyrate is an important substrate for gut epithelia compared with acetate and propionate,but it is apparently not a unique nutrient. Acetate, propionate, and isobutyrate are all metabo-lites of endogenous pathways in the organism. Acetate has the lowest membrane permeability,is utilized from peripheral arterial blood in major extrahepatic tissues (Pethick and Lindsay,1982), and is a universal metabolite in the body in the form of acetyl-CoA. Propionate is themain donor of 3-carbon units for gluconeogenesis in the ruminant liver and is efficiently takenup by the liver (Leng and Annison, 1963). The endogenous sources of propionate includedegradation of uneven chained fatty acids and some amino acids (methionine, threonine,isoleucine, and valine). Isobutyrate (an intermediate from catabolism of valine) appears inrelatively low concentrations in the rumen, but is efficiently taken up by the liver for gluco-neogenesis (Stangassinger and Giesecke, 1979). These SCFA are not only well tolerated inhepatic and peripheral tissues, but are key metabolites (especially acetate and propionate) inthese tissues, and this agrees with a limited uptake of these SCFA in the gut epithelia.

Butyrate, valerate, and probably longer, medium-chain fatty acids (MCFA) are less polarand will have a relatively high permeability in cell membranes. One way of controlling per-meability is partial oxidation of these SCFA into acetoacetate and 3-hydroxybutyrate in thegut epithelia. When butyrate appears in the systemic circulation or is added to cell cultures,


it has been shown to have a number of adverse effects: inhibition of growth and induction ofmorphological changes in cultured cells of different origins including ruminal epithelial celllines (Prasad and Sinha, 1976; Gálfi et al., 1991); being an insulin secretagogue (Manns andBoda, 1967); inhibition of gastrointestinal motility by stimulation of epithelial receptors(Crichlow, 1988) and/or via systemic effects (Le Bars et al., 1954); stimulation of rumenepithelial development (Sander et al., 1959); or killing (2.5 mmol butyrate/kg BW in lambs)the animal (Manns and Boda, 1967). The epithelia of the gut have apparently evolved toperform gatekeeping functions by controlling the entry of butyrate and longer-chain fermen-tation acids into the peripheral circulation. It is tempting to speculate that the side effects of thegatekeeper function are that these metabolites also become quantitatively important oxidativesubstrates.

16. ACYL-CoA SYNTHETASES

Activation of SCFA by an acyl-CoA synthetase (also named CoA ligase or thiokinase) is thefirst step in the metabolism of any SCFA in the cells of the gut epithelium, liver, or peripheraltissues (Groot et al., 1976). The acyl-CoA synthetases are therefore believed to be keyenzymes in different tissues’ selectivity to metabolism of different SCFA. There exista number of distinct acyl-CoA synthetases: acetyl-CoA, propionyl-CoA, butyryl-CoA, medium-chain fatty acid, and long-chain fatty acid-CoA synthetases.

The acetyl-CoA synthetase (EC 6.2.1.1) has a high affinity for acetate, and some affinityfor propionate (Campagnari and Webster, 1963; Groot et al., 1976; Ricks and Cook, 1981b).However, it is noteworthy that the activity of this enzyme has been found to be low in therumen epithelium and liver of ruminants (Cook et al., 1969; Ash and Baird, 1973). Theseobservations are in line with a limited role of the rumen epithelium and the liver in metabo-lism of absorbed acetate.

The ruminant liver has a relatively high propionyl-CoA synthetase (EC 6.2.1.17) activity(Ash and Baird, 1973) and there exist a number of indications that propionyl-CoA synthetaseis a distinct enzyme (Ricks and Cook, 1981a,c). Among the interesting features of thisenzyme is that it is not present in rumen epithelium. This is not the same as denying anypossible activation of propionate in rumen epithelium, which obviously can occur (Weekes,1974), but it has been shown that the propionyl-CoA synthetase activity in the liver is almostinsensitive to the presence of butyrate whereas the activity in the rumen epithelium is almostcompletely inhibited by the presence of butyrate (Ash and Baird, 1973; Harmon et al., 1991).As is the case with acetyl-CoA synthetase in rumen epithelium, the lack of propionyl-CoAsynthetase activity is in agreement with in vivo observations showing a very limited uptake ofpropionate by the rumen epithelium.

As described above, one of the most striking features of rumen epithelial metabolism is ahigh affinity and capacity for metabolism of butyrate. This feature is reflected in the butyryl-CoA synthetase activity of the epithelium (Ash and Baird, 1973). The relative importance ofthe liver and the rumen epithelium in the metabolism of propionate and butyrate, respectively,is directly reflected in the acyl-CoA synthetase activities. Moreover, as butyrate was found tohave an insignificant effect on propionate activation in the liver, propionate had no effect onbutyrate activation in the rumen epithelium, but decreased the butyrate activation in the liver(Ash and Baird, 1973). A distinct butyryl-CoA synthetase (EC 6.2.1.2) was first purified frombovine heart mitochondria and this enzyme showed a high affinity for valerate and caproate(Webster et al., 1965). In ruminants, butyrate affinity is also found in xenobiotic/medium-chainfatty acid-CoA synthetases. These acyl-CoA synthetases activate a broad spectrum of


straight-chain fatty acids: butyrate, longer SCFA, branched-chain fatty acids, and a numberof xenobiotic (of foreign origin) carboxylic acids; others include benzoate and phenylacetate(Aas, 1971; Vessy et al., 1999). Indirect evidence from PDV flux studies indicates cross-specificity for valerate activation, which agrees with both types of butyrate activatingsystems. So far, no specific information seems to be available on the interaction of SCFAactivation with longer-chain fatty acids or xenobiotic compounds absorbed from the rumen(Cremin et al., 1995); however, the fact that the isolated rumen epithelium or isolated rumenepithelial cells are able to use a wide range of fatty acids from SCFA to palmitate indicatesthe presence of some activity for activating medium- as well as long-chain fatty acids by theepithelium (Jesse et al., 1992; Hird et al., 1966).

17. HOW IS BUTYRATE METABOLIZED BY GUT EPITHELIA?

Rumen epithelial ketogenesis is remarkable compared to hepatic ketogenesis by virtue of thefact that rumen epithelial ketogenesis is a main pathway in the fed state, and not a pathwayturned on at fasting or when the organism is facing a high “metabolic drain”. This feature isobviously connected to the constant fueling of rumen epithelial ketogenesis via butyrateabsorption in combination with an apparent need for removal of butyrate before entering theblood stream.

The oxidation of butyryl-CoA to acetoacetyl-CoA in rumen epithelium (fig. 1) is, from achemical point of view, identical to the initial steps of long-chain fatty acid β-oxidation. (Fora review on this subject, see Eaton et al., 1996.) The first 3-hydroxybutyrate intermediate ofthis pathway is the L-(S)-isomer, which is not released to the peripheral circulation. Oxidationof L-3-hydroxybutyrate-CoA yields acetoacetyl-CoA. Acetoacetyl-CoA is a branching pointbetween acetyl-CoA formation and ketone release because of the acetoacetyl-CoA thiolase(EC 2.3.1.9) catalyzed equilibrium between acetoacetyl-CoA and acetyl-CoA (fig. 2). Theequilibrium constant of the reaction (6 × 10−6; Williamson et al., 1968) is strongly favoringacetyl-CoA and this means that the concentration of acetoacetyl-CoA probably will be rela-tively low in the mitochondrion. The production of ketone bodies from acetate (Harmon et al.,1991) or valerate (Weigand et al., 1975) in rumen epithelium confirms that acetoacetyl-CoAthiolase is present in the rumen epithelium, an observation also confirmed by assays onepithelial cell extracts (Baird et al., 1970). The main function of acetoacetyl-CoA thiolase isprobably not to mediate ketogenesis from absorbed acetate, although this mediation is possible.


Fig. 1. Initial oxidation of butyryl-CoA to acetoacetyl-CoA in rumen epithelium proceeds via a pathwaysimilar to the initial steps in β-oxidation. A number of isoenzymes are known for both acyl-CoA dehydroge-nases (first dehydrogenase of the pathway) and 3-hydroxyacyl-CoA dehydrogenases (Eaton et al., 1996).However, the isoenzymes with specificity for short-chain acyl-CoA are likely to predominate in the rumenepithelium. The hydratase in the pathway is likely crotonase (EC 4.2.1.17), also an enzyme with the highestspecificity toward short-chain acyl groups.


The fact that the rumen epithelium contains low amounts of acetyl-CoA synthetase, and thelow affinity of the butyryl-CoA synthetase for acetate when butyrate is present, points to theconclusion that the main function of the acetoacetyl-CoA thiolase is feeding acetyl-CoA fromthe acetoacetyl-CoA pool to the TCA cycle. Therefore, even though butyrate metabolism inthe epithelium cannot be explained from the point of the energy needs of the epithelium,butyrate metabolism is ensured to be the main oxidative substrate under in vivo conditions.

Contrary to the consensus about the initial steps of butyrate metabolism, there has been morediscussion of the subsequent metabolism of acetoacetyl-CoA. This compound can be deacylateddirectly (acetoacetyl-CoA deacylase; EC 3.1.2.11) or deacylated via the 3-hydroxy-3-methyl-glutaryl-CoA pathway (3-HMG pathway; 3-hydroxy-3-methylglutaryl-CoA synthetase andlyase; EC 4.1.3.5 and EC 4.1.3.4); however, other alternative pathways have been suggestedand will be discussed briefly. The presence of the 3-HMG pathway (fig. 3) in rumen epitheliumis supported by the fact that the enzymes of the pathway (3-hydroxy-3-methylglutaryl-CoAsynthetase, and lyase) have been shown to be present in the epithelium in significant amounts(Baird et al., 1970; Leighton et al., 1983).

However, isotopomer studies have had a dominant role in the arguments about ketogenicpathways in the epithelium. Hird and Symons (1961) investigated isolated ruminal andomasal epithelial metabolism of [1-14C]butyrate and [3-14C]butyrate into acetoacetate. Theisotopomers of acetoacetate could be partly identified by measuring the label in position 1(CO2 from decarboxylation of acetoacetate) and in the label in the acetone fraction afterdecarboxylation (interpreted as position 3). When the epithelium was incubated with[1-14C]butyrate, 80% of the label in acetoacetate was found in position 1 and 20% of the labelwas found in position 3. When the substrate was [3-14C]butyrate, 37% of the label in aceto-acetate was found in position 1 and only 63% in position 3. The probable explanation for the1 to 3 shifts in labeling is the thiolase-catalyzed equilibrium between acetoacetyl-CoA andacetyl-CoA (fig. 2). The labeling pattern also gives an indication of the relative importance ofthe pathway. The fact that 20% of the label in acetoacetate was found in position 3 could lead

Fig. 2. The acetoacetyl-CoA thiolase (EC 2.3.1.9) is catalyzing the reversible thiolytic cleavage of ace-toacetyl-CoA into two acetyl-CoAs. The equilibrium between acetoacetyl-CoA and acetyl-CoA is favoringacetyl-CoA and as a result the acetoacetyl-CoA concentration in the mitochondrion will usually be low.

to the conclusion that 80% of acetoacetate was generated without degradation to acetyl-CoA.However, that is a false conclusion because if we assume that the labeling does not cause frac-tionation, then the acetoacetyl-CoA generated from acetyl-CoA (acetyl-CoA will be labeledin position 1) will be evenly distributed among carbon 1 and carbon 3 of acetoacetate. Thismeans that at least 40% of the acetoacetate must have been equilibrating with acetyl-CoA toexplain 20% of the total activity in position 3.

The fact that the large majority of label from [1-14C]butyrate ends up in C1 of acetoacetatehas been used as an argument against the function of the 3-HMG pathway in the epithelium.However, this argument might not be justified because this pathway will conserve C1 label inposition C1, especially if the thiolase activity is relatively low compared to the flux throughthe 3-HMG pathway. The fact that Hird and Symons (1961) found a larger 3 to 1 shift in label-ing of acetoacetate from [3-14C]butyrate is therefore in agreement with the 3-HMG pathwaynot only working on acetoacetyl-CoA derived from acetyl-CoA, but also on acetoacetyl-CoAfrom the initial-oxidation steps on butyrate. It seems puzzling that only 37% of the label inacetoacetate generated from [3-14C]butyrate was found in position 1, especially if the majorityof the acetoacetate production is through the 3-HMG pathway. However, the relative enrichmentof the acetyl-CoA pool and the acetoacetyl-CoA pool will have a major impact on the results.It is likely that the metabolism of [3-14C]butyrate will be accompanied by a lower specificactivity of the acetyl-CoA pool compared with the [1-14C]butyrate because the [3-14C]butyratewill be less likely to deliver a labeled acetyl-CoA to the acetyl-CoA pool compared with[1-14C]butyrate as substrate. The only labeling of the acetyl-CoA pool from [3-14C]butyratewill be through the thiolase-catalyzed acetyl-CoA/acetoacetyl-CoA equilibrium. This impliesthat the 3 to 1 shift observed with the [3-14C] butyrate incubation indicates a far higher impor-tance of the 3-HMG-CoA pathway than that apparently shown by the 37% of [3-14C]butyratefound in position 1 of acetoacetate simply because the specific activity of acetyl-CoA will belower under these conditions.

Though acetoacetate is the product of rumen epithelial ketogenesis, it is not the primarycirculating ketone in plasma. A large proportion of acetoacetate is reduced to D-3-hydroxy-butyrate (fig. 4) before leaving the epithelial cells catalyzed by 3-hydroxybutyratedehydrogenase (EC 1.1.1.30). Data on rumen epithelial enzyme activity and isotopomerdistribution in acetoacetate suggest that the 3-HMG pathway is as quantitatively important inthis tissue as it is in liver. Earlier denials (Annison et al., 1963) are partly correct in pointingout that butyrate is not completely degraded to acetyl-CoA before incorporation into ketonebodies.


Fig. 3. The 3-hydroxy-3-methylglutaryl-CoA pathway (3-HMG pathway) ensures that acetoacetyl-CoA,despite its low concentration, can be “trapped” and deacylated. These steps of ruminal ketogenesis are similarto hepatic ketogenesis.

18. ALTERNATIVE KETOGENIC PATHWAYS

Although the 3-HMG pathway is active in rumen epithelium, it might not be the only keto-genic pathway. One obvious alternative would be acetate release by the epithelium.Endogenous acetate production has been observed in vitro when the epithelium is incubatedwithout substrate (Sehested et al., 1999a). The hydrolysis of acetyl-CoA into acetate and CoAis catalyzed by acetyl-CoA deacylase (EC 3.1.2.1; Grigat et al., 1979). We might wonder whythe ruminant produces ketone bodies at all when it seems much simpler just to use acetate asa carrier of acetyl units. One of the reasons for the production of ketone bodies could be thecost of reactivation in recipient tissues because they would have to pay double the price foractivation when acetate is the substrate compared with acetoacetate. Nevertheless, endoge-nous acetate production can be observed in vitro by rumen epithelium and acetate would bean obvious candidate for interorgan acetyl transfer. However, we have only limited and indi-rect evidence of endogenous acetate production by rumen epithelium in vivo (Kristensen et al., 2000a). It is unknown to what extent endogenous acetate from the PDV has a role ininterorgan acetyl exchange (i.e. acetyl carbon originally absorbed in fatty acids other thanacetate itself).

Not only acetyl-CoA, but also acetoacetyl-CoA, might be directly deacylated (acetoacetyl-CoA deacylase; EC 3.1.2.11), and thereby lead to 3-HMG-CoA-independent acetoacetatesynthesis. The acetoacetyl-CoA deacylase has been found in rumen epithelium though onlyat a low activity (Baird et al., 1970). One of reasons why direct deacylation of acetoacetyl-CoAmight be of limited importance is the low acetoacetyl-CoA concentration in the mitochondrion.The low affinity of the acetoacetyl-CoA deacylase present in rat liver was quantitatively notimportant, although it was functional under in vitro conditions with high acetoacetyl-CoAconcentrations (Williamson et al., 1968).

A number of alternative pathways have been suggested to explain various parts of ketonebody formation in rumen epithelium: succinyl-CoA:3-ketoacid CoA-transferase (Bush andMilligan, 1971); a L-3-hydroxybutyrate pathway not involving acetoacetate formation(Emmanuel et al., 1982); and a butyrate:acetoacetyl-CoA transferase pathway (Emmanueland Milligan, 1983). These pathways all suggest metabolism of butyrate to 3-hydroxybutyrateas one unbroken C4 unit. The two latter pathways appear to be closely related to cytosolicpathways in tissues utilizing acetoacetate in de novo synthesis of fatty acids (Robinson andWilliamson, 1980). However, it is difficult to determine the quantitative importance of non-3-HMG-CoA pathways in the rumen epithelium from the limited data available.


Fig. 4. D-3-Hydroxybutyrate is the dominating “ketone” in plasma due to the 3-hydroxybutyrate dehydroge-nase (EC 1.1.1.30) catalyzed and NADH-dependent reduction of acetoacetate.

The situation for the succinyl-CoA:3-ketoacid CoA-transferase (EC 2.8.3.5; SCOT) isdifferent because this enzyme is indeed anticipated to be a key enzyme in ketone body metab-olism; however, its function is opposite to the function proposed in the rumen epithelium.This enzyme is a key enzyme in the activation of ketone bodies in peripheral tissues and therare deficiency of this enzyme in human infants leads to severe ketoacidosis (Synderman et al.,1998). Succinyl-CoA:3-ketoacid transferase has been assumed to contribute to the acetoacetyl-CoA hydrolysis in rumen epithelium because addition of succinate was followed by increaseddisappearance of acetoacetyl-CoA (Bush and Milligan, 1971). If SCOT was important for thehydrolysis of acetoacetyl-CoA, it would suggest that the concentration of either acetoacetyl-CoAor succinate was higher in rumen epithelium compared with other tissues. However, owing tothe true reversibility (Stern et al., 1956) of the reaction catalyzed by SCOT (acetoacetyl-CoA +succinate ↔ acetoacetate + succinyl-CoA), it might be suggested that this activity in the rumenepithelium was connected to the specific incubation conditions in vitro and not necessarily thepathway of acetoacetyl-CoA metabolism in vivo.

19. METABOLITE INTERACTIONS IN RUMEN EPITHELIALKETOGENESIS

If the rumen epithelium works in its usual position in a ruminant, or is maintained for a shortperiod under in vitro conditions as epithelial slices or isolated cells, it will have an obligaterequirement for chemical energy to maintain Na+, Ca2+, and K+ ion concentration gradientsand other vital cell functions. Considering a situation with a relatively constant workload ofthe epithelium, it would then be expected that tissue supplied with small amounts of butyratewould oxidize a large fraction to CO2 simply to fulfill the basic needs of ATP and sustainbasic cell functions. This relationship has been confirmed in vitro when different butyrateconcentrations were compared. Increasing the supply of butyrate was followed by the oxida-tion of a decreasing fraction and an increasing fraction metabolized into ketone bodies (Becket al., 1984).

From a whole animal perspective, glucose is antiketogenic (Hamada et al., 1982) andinitially it was surprising that glucose had the opposite effect on rumen epithelial ketogenesis,i.e. ketogenesis was stimulated by glucose (Stangassinger et al., 1979). A number of gluco-genic substrates have been shown to impose a similar effect on epithelial metabolism. Somevariability in the response concerning the uptake of butyrate and the proportion of butyrateoxidized has been observed, but generally a shift toward the more reduced “ketone body”,3-hydroxybutyrate, compared with acetoacetate has been observed with the addition of aglucogenic substrate (Goosen, 1976; Beck et al., 1984; Giesecke et al., 1985; Baldwin andJesse, 1996). Although the rumen epithelium is able to take up a broad range of metabolitesincluding glucose, glutamine, and glutamate and oxidize them (Harmon, 1986; Baldwin andMcLeod, 2000), this does not mean that glucose is the oxidative substrate that caused the shiftin ketone body production. In fact, we would surmise from the discussion of butyrate metab-olism (see above) that the epithelium had a source of acetyl-CoA from butyrate that would beable to fulfill any oxidative need. The reason might be that epithelium incubated without aglucogenic source will become depleted of TCA cycle intermediates and subsequently havedifficulty maintaining ATP, NADH, and NADPH potentials. A very elegant example of thiseffect is the comparison between metabolite production from butyrate and valerate in rumenepithelium incubated in vitro (Weigand et al., 1975). When rumen epithelium was incubatedwith butyrate, 0.67 of the ketone bodies produced were acetoacetate; however, when incubatedwith valerate only 3-hydroxybutyrate was produced. This production was accompanied by


lactate produced from the 3-carbon fraction of valerate. Therefore, there seems to be noreason to believe that glucose or any other glucogenic substrates play a particular role as reg-ulators of rumen epithelial ketogenesis, but the results point to the general conclusion thatrumen epithelium has a range of nutritional requirements for proper function.

20. THE IN VIVO/IN VITRO PROPIONATE ENIGMA

In vitro, rumen epithelium metabolizes propionate into lactate (Weigand et al., 1975). In vivo,however, it has not been possible to demonstrate any major propionate metabolism intolactate using ruminal infusion of 14C or 13C labeled propionate (Weigand et al., 1972; Weekesand Webster, 1975; Kristensen et al., 2001). These matters have been even further confusedby the fact that in vivo experiments on portal recovery of ruminal propionate indicate that alarge proportion of propionate was metabolized by the epithelium (Bergman and Wolff,1971). Because of the large capacity of the liver to metabolize propionate in vivo (Berthelotet al., 2002) it is difficult to explain why the rumen epithelium should limit the propionatesupply to the liver. The reason for the large activation of propionate under in vitro conditionsis probably the cross-specificity of the butyryl-CoA synthetase. In vivo, propionyl-CoA couldbe generated by thiolysis of 3-oxo-valeryl-CoA (from valerate). This latter source might bethe explanation for the high capacity of propionyl-CoA-utilizing pathways in rumen epithe-lium. The usual metabolism of propionate via propionate carboxylation to methylmalonicacid followed by the TCA intermediate succinate will lead to the buildup of TCA intermedi-ates. In the liver the main pathway to export surplus TCA intermediates is gluconeogenesis.Other tissues use nonessential amino acids (e.g. alanine and glutamine synthesis in musclesand other tissues) to control excess TCA intermediates. In rumen epithelium, it is apparentlythe malic enzyme (EC 1.1.1.40) catalyzed decarboxylation of malate into pyruvate (coupledto reduction of NADP) and the subsequent reduction of pyruvate to lactate that removes thesurplus of propionyl-CoA from the rumen epithelial cells (Young et al., 1969). By thesemechanisms we are able to explain the differing in vitro and in vivo observations on rumenepithelial metabolism.

21. FITTING THE CARBON BALANCE OF FERMENTATION IN THE GUT

Although there is no doubt that SCFA are important in ruminant metabolism, no feed evalu-ation system has been able integrate knowledge of SCFA production, absorption, andmetabolism in ruminants under production conditions. Simulation models constructed todescribe fermentation and SCFA absorption, as well as other nutrients, need to improve inorder to predict SCFA proportions in the rumen (Baker and Dijkstra, 1999). The problem hasalso been what to do with the apparently huge metabolic activity of the rumen epithelium. Nomodel has been able to incorporate this metabolism, and this review attempts a possibleexplanation.

Simulation models of ruminal fermentation and metabolism developed to date have beenconstructed and validated mainly against duodenal nutrient flows. The re-evaluation of the roleof the gut epithelia in metabolism of SCFA has enabled an alternative method of model com-parison. If the rumen and other gut epithelia do not metabolize significant amounts of acetateand utilize only a small percentage of the propionate flux, then the net rumen productionof these acids would be predictable from PDV fluxes. Major corrections to be consideredare, however, uptake of arterial acetate by PDV tissues and epithelial butyrate metabolism.


The percentage of arterial acetate uptake by PDV tissues has been shown to be stable (about32% of arterial flux) when evaluated across different rations and with relatively little post-prandial variation in meal-fed sheep (Bergman and Wolff, 1971; Kristensen et al., 1996a). Asdiscussed earlier, the portal recovery of butyrate has been shown to be a more complex func-tion of its availability, and an increased portal recovery of butyrate with increasing ruminalproduction rates might be expected.

In an attempt to compare data from studies on PDV fluxes with model predictions of gas-trointestinal fermentation, Kristensen and Danfær (unpublished data) compared portal fluxesof SCFA from 36 studies in which a total of 58 different diets were fed to sheep and cattleunder different physiological conditions (growing/maintenance, nonpregnant/pregnant,dry/lactating). The model used to predict fermentation and digestion of the diets was Karoline(Danfær, 1990) (version 8a, a dynamic simulation model of a lactating cow mainly validatedagainst duodenal flow data; Danfær et al., unpublished). In the model, all diets were com-pared at a fixed dry matter intake of 20 kg/d and predicted carbon output in moles SCFAcarbon per kg dry matter intake was compared to the observed/recalculated PDV fluxes in thestudies. The experimentally observed PDV fluxes were corrected for acetate uptake in PDVtissues (assumed 32% of the arterial flux), propionate uptake by epithelial tissues (assumingthat portal flux was equal to 95% of true absorption), and butyrate recovery [assuming portalrecovery of gut butyrate production = 0.35 × P/(P + 0.05) where P = portal net appearancemmol × h−1 × (kg BW0.75)−1]. The portal recovery of butyrate is deliberately set to a higherlevel than those typically found following butyrate infusion into the normally functioningrumen. This recovery agrees with observations with sheep maintained on intragastric nutri-tion or short-term washed reticulo-rumen (Gross et al., 1990a,b; Kristensen et al., 2000a).

The calculated SCFA production in the 36 experiments using the correction factors above was11.9 ± 0.4 moles C in SCFA/kg dry matter intake. The simulated value was 12.3 ± 0.2 moles Cin SCFA/kg dry matter intake and the mean bias was 0.4 moles C/kg DMI [Σ(predicted −observed)/number of observations; see Kohn et al. (1998)]. However, the root mean squareprediction error (RMSPE) was 2.6 moles C in SCFA/kg DMI [(Σ(predicted − observed)2/number of observations); see Kohn et al. (1998)]. On average, the model and the correctedexperimental data are in good agreement. However, there is still a need for better models topredict net SCFA output. The corrected, experimentally determined SCFA production was, onaverage, 45 ± 2% of the simulated digestible energy. However, estimates based on intragas-tric tracer dilution, as discussed above, seem to overestimate the SCFA production andindirect evidence also supports these figures. In fact, the SCFA production accounting for45% of digestible energy implies that 65% of total digested carbon is found in fermentationgases and SCFA. However, if the true relationship between portal absorption and gut produc-tion of SCFA is similar to the relationship described by Bergman and Wolff (1971), then theproduction of SCFA would need 116 ± 5% of the digested carbon to account for SCFA andfermentation gases. This would seem to be impossible. The good news is that values of portalabsorption of SCFA actually make sense in terms of animal energy metabolism. It mustbe emphasized, however, that models of ruminal and hindgut fermentation still have a lot togain in terms of precision of SCFA production, especially in the prediction of ruminal SCFAcomposition.

22. CONCLUSIONS

Glucose is a major metabolic fuel for ruminant tissues, similar to most mammals. The pre-gastric fermentation dictates that gluconeogenesis serves to supply the glucose needs under


most feeding situations. Lactate derived from the diet, ruminal bacterial metabolism and fromendogenous metabolism is a major glucose precursor in ruminants, supplying 9–35% ofhepatic glucose carbon, and is a key carbon intermediate in growing ruminants. Ruminantshave been shown experimentally to be capable of contributing significant quantities of glucosethrough intestinal digestion and glucose absorption. This additional glucose does impact growthand retention of body tissues.

The gut epithelia have a central function as gatekeepers for butyrate and longer-chainSCFA and MCFA. These acids have also become the main energy substrates of gut epithelia.There is no evidence suggesting that the rumen epithelium should have excessive requirementsfor energy metabolism, but rather intraruminal (luminal) isotopic dilution techniques overesti-mate net SCFA production because of microbial metabolism. Data on portal appearance ofSCFA corrected for PDV metabolism of arterial metabolites is therefore the best direct measureof SCFA availability in ruminants. The average absorption of SCFA in ruminants is equivalentto about 45% of the digestible energy intake.


Current feeding systems often fail to meet today’s demand for accurate prediction of thenutrient needs of animals consuming a large menu of feedstuffs under a wide array ofenvironmental conditions. To accomplish this task we need to understand all phases fromdigestion and nutrient assimilation to the subsequent use of nutrients by various tissues. Thus,successful models in the future will span concepts from commodity to animal product byincorporating the metabolic transformations in between. This chapter has reviewed recentfindings on the impact of intestinal digestion on glucose availability, the contributions oflactate to meeting the glucose needs, and how our understanding of SCFA metabolism hasbeen revised from long-accepted concepts. Developing models for predicting animal biolog-ical response are dependent on findings such as these to supply quantitative information todescribe animal systems. Data are still greatly lacking for concepts as fundamental as SCFAproduction and glucose absorption, and their metabolism at various stages of growth andproduction. The near-global de-emphasis on agricultural production research may make thesepieces of the puzzle long in coming.

REFERENCES

Aas, M., 1971. Organ and subcellular distribution of fatty acid activating enzymes in the rat. Biochim.Biophys. Acta 231, 32–47.

Annison, E.F., Leng, R.A., Lindsay, D.B., White, B.A., 1963. The metabolism of acetic acid, propionicacid and butyric acid in sheep. Biochem. J. 88, 248–252.

Ash, R., Baird, G.D., 1973. Activation of volatile fatty acids in bovine liver and rumen epithelium:evidence for control by autoregulation. Biochem. J. 136, 311–319.

Baird, G.D., Hibbitt, K.G., Lee, J., 1970. Enzymes involved in acetoacetate formation in varous bovinetissues. Biochem. J. 117, 703–709.

Baker, S.K., Dijkstra, J., 1999. Dynamic aspects of the microbial ecosystem of the reticulo-rumen.In: Jung, H.-J.G., Fahey, G.C. Jr. (Eds.), Nutritional Ecology of Herbivores. American Society ofAnimal Science, Savoy, IL, pp. 261–311.

Balcells, J., Seal, C.J., Parker, D.S., 1995. Effect of intravenous glucose infusion on metabolism of portal-drained viscera in sheep fed a cereal/straw-based diet. J. Anim. Sci. 73, 2146−2155.

Baldwin, R.L., Jesse, B.W., 1996. Propionate modulation of ruminal ketogenesis. J. Anim. Sci. 74,1694–1700.

Baldwin, R.L., McLeod, K.R., 2000. Effects of diet forage:concentrate ratio and metabolizable energyintake on isolated rumen epithelial cell metabolism in vitro. J. Anim. Sci. 78, 771–783.


Barcroft, J., McAnally, R.A., Phillipson, A.T., 1944. Absorption of volatile acids from the alimentarytract of the sheep and other animals. J. Exp. Biol. 20, 120–129.

Bauer, M.L., Harmon, D.L., McLeod, K.R., Huntington, G.B., 1995. Adaptation to small intestinal starchassimilation and glucose transport in ruminants. J. Anim. Sci. 73, 1828–1838.

Beck, U., Emmanuel, B., Giesecke, D., 1984. The ketogenic effect of glucose in rumen epithelium ofovine (Ovis aries) and bovine (Bos taurus) origin. Comp. Biochem. Physiol. B 77, 517–521.

Bergman, E.N., 1973. Glucose metabolism in ruminants as related to hypoglycemia and ketosis. CornellVet. 63, 341–382.

Bergman, E.N., 1990. Energy contributions of volatile fatty acids from the gastrointestinal tract in variousspecies. Physiol. Rev. 70, 567–590.

Bergman, E.N., Wolff, J.E., 1971. Metabolism of volatile fatty acids by liver and portal-drained viscerain sheep. Amer. J. Physiol. 221, 586–592.

Berthelot, V., Pierzynowski, S.G., Sauvant, D., Kristensen, N.B., 2002. Hepatic metabolism of propionateand methylmalonate in growing lambs. Livest. Prod. Sci. 74, 33–43.

Britton, R., Krehbiel, C., 1993. Nutrient metabolism by gut tissues. J. Dairy Sci. 76, 2125–2131.Bugaut, M., 1987. Occurrence, absorption and metabolism of short chain fatty acids in the digestive tract

of mammals. Comp. Biochem. Physiol. B 86, 439–472.Bush, R.S., Milligan, L.P., 1971. Enzymes of ketogenesis in bovine rumen epithelium. Can. J. Anim. Sci.

51, 129–133.Campagnari, F., Webster, L.T. Jr., 1963. Purification and properties of acetyl coenzyme A synthetase from

bovine heart mitochondria. J. Biol. Chem. 238, 1628–1633.Cappelli, P.F., Seal, C.J., Parker, D.S., 1997. Glucose and [13C]leucine metabolism by the portal-drained

viscera of sheep fed on dried grass with acute intravenous and intraduodenal infusions of glucose.Brit. J. Nutr. 78, 931–946.

Cook, R.M., Liu, S.-C.C., Quraishi, S., 1969. Utilization of volatile fatty acids in ruminants. III.Comparison of mitochondrial acyl coenzyme A synthetase activity and substrate specificity in differ-ent tissues. Biochemistry 8, 2966–2969.

Counotte, G.H., Prins, R.A., 1981. Regulation of lactate metabolism in the rumen. Vet. Res. Commun. 5,101–115.

Cremin, J.D. Jr., McLeod, K.R., Harmon, D.L., Goetsch, A.L., Bourquin, L.D., Fahey, G.C. Jr., 1995. Portaland hepatic fluxes in sheep and concentrations in cattle ruminal fluid of 3-(4-hydroxyphenyl)propionic,benzoic, 3-phenylpropionic, and trans-cinnamic acids. J. Anim. Sci. 73, 1766–1775.

Crichlow, E.C., 1988. Ruminal lactic acidosis: forestomach epithelial receptor activation by undissoci-ated volatile fatty acids and rumen fluids collected during loss of reticuloruminal motility. Res. Vet.Sci. 45, 364–368.

Danfær, A., 1990. A Dynamic Model of Nutrient Digestion and Metabolism in Lactating Dairy Cows.Beretning fra Statens Husdyrbrugsforsøg, Frederiksberg Bogtrykkeri, Denmark.

De Visser, H., Valk, H., Klop, A., Van derMeulen, M.J., Bakker, J.G., Huntington, G.B., 1997. Nutrientfluxes in splanchnic tissue of dairy cows: influence of grass quality. J. Dairy Sci. 80, 1666–1673.

Dietschy, J.M., 1978. General principles govering movement of lipids across biological membranes. In:Dietschy, J.M., Gotto, A.M.J., Ontko, J.A. (Eds.), Disturbances in Lipid and Lipoprotein Metabolism.American Physiological Society, Baltimore, MD, pp. 1–28.

Dunkin, A.C., Campbell, R.G., 1982. Some aspects of the partition of metabolizable energy in the yougpig. In: Ekern, A., Sundstol, F. (Eds.), Proceedings of the 9th symposium on Energy Metabolism ofFarm Animals, Lillehammer, Norway. EAAP Publ. No. 29, pp. 198–201.

Dunlop, R.H., 1972. Pathogenesis of ruminant lactic acidosis. Adv. Vet. Sci. Comp. Med. 16, 259–302.Eaton, S., Bartlett, K., Pourfarzam, M., 1996. Mammalian mitochondrial β-oxidation. Biochem. J. 320,

345–357.Eisemann, J.H., Huntington, G.B., 1993. Effects of dietary clenbuterol on net flux across the portal-

drained viscera, liver and hindquarters of steers (Bos taurus). Comp. Biochem. Physiol. C 104,401–406.

Eisemann, J.H., Huntington, G.B., 1994. Metabolite flux across portal-drained viscera, liver, andhindquarters of hyperinsulinemic, euglycemic beef steers. J. Anim. Sci. 72, 2919–2929.

Eisemann, J.H., Huntington, G.B., Catherman, D.R., 1996. Patterns of nutrient interchange and oxygenuse among portal-drained viscera, liver, and hindquarters of beef steers from 235 to 525 kg bodyweight. J. Anim. Sci. 74, 1812–1831.

Emmanuel, B., Milligan, L.P., 1983. Butyrate:acetoacetyl-CoA transferase activity in bovine rumenepithelium. Can. J. Anim. Sci. 63, 355–360.


Emmanuel, B., Stangassinger, M., Giesecke, D., 1982. Production of D(−)-3-hydroxybutyrate by an alter-nate mechanism in rumen epithelium of ovine (Ovis aries) and bovine (Bos taurus). Comp. Biochem.Physiol. B 72, 415–419.

Gäbel, G., Müller, F., Pfannkuche, H., Aschenbach, J.R., 2001. Influence of isoform and DNP on butyratetransport across the sheep ruminal epithelium. J. Comp. Physiol. B 171, 215–221.

Gálfi, P., Neogrády, S., Sakata, T., 1991. Effects of volatile fatty acids on the epithelial cell proliferationof the digestive tract and its hormonal mediation. In: Tsuda, T., Sasaki, Y., Kawashima, R. (Eds.),Physiological Aspects of Digestion and Metabolism in Ruminants. Academic Press, San Diego, CA,pp. 49–59.

Giesecke, D., Beck, U., Emmanuel, B., 1985. Ketogenic regulation by certain metabolites in rumenepithelium. Comp. Biochem. Physiol. B 81, 863–867.

Goad, D.W., Goad, C.L., Nagaraja, T.G., 1998. Ruminal microbial and fermentative changes associatedwith experimentally induced subacute acidosis in steers. J. Anim. Sci. 76, 234–241.

Goosen, P.C.M., 1976. Metabolism in rumen epithelium oxidation of substrates and formation of ketonebodies by pieces of rumen epithelium. Z. Tierphysiol. Tierernähr. Futtermittelkde. 37, 14–25.

Grigat, K.-P., Koppe, K., Seufert, C.-D., Söling, H.-D., 1979. Acetyl-coenzyme A deacylase activity inliver is not an artifact: subcellular distribution and substrate specificity of acetyl-coenzyme A deacy-lase activities in rat liver. Biochem. J. 177, 71–79.

Groot, P.H.E., Scholte, H.R., Hülsmann, W.C., 1976. Fatty acid activation: specificity, localization, andfunction. Adv. Lipid Res. 14, 75–126.

Gross, K.L., Harmon, D.L., Avery, T.B., 1990a. Portal-drained visceral flux of nutrients in lambs fedalfalfa or maintained by total intragastric infusion. J. Anim. Sci. 68, 214–221.

Gross, K.L., Harmon, D.L., Minton, J.E., Avery, T.B., 1990b. Effects of isoenergetic infusions of propi-onate and glucose on portal-drained visceral nutrient flux and concentrations of hormones in lambsmaintained by total intragastric infusion. J. Anim. Sci. 68, 2566–2574.

Hamada, T., Ishii, T., Taguchi, S., 1982. Blood changes of spontaneously ketotic cows before and fourhours after administration of glucose, xylitol, 1,2-propanediol, or magnesium propionate. J. DairySci. 65, 1509–1513.

Harmon, D.L., 1986. Influence of dietary energy intake and substrate addition on the in vitro metabolismof glucose and glutamine in rumen epithelial tissue. Comp. Biochem. Physiol. B 85, 643–647.

Harmon, D.L., Avery, T.B., 1987. Effects of dietary monensin and sodium propionate on net nutrient fluxin steers fed a high-concentrate diet. J. Anim. Sci. 65, 1610–1616.

Harmon, D.L., Britton, R.A., Prior, R.L., 1983. Influence of diet on glucose turnover and rates of gluco-neogenesis, oxidation and turnover of D-(−)-lactate in the bovine. J. Nutr. 113, 1842–1850.

Harmon, D.L., Britton, R.A., Prior, R.L., 1984. In vitro rates of oxidation and gluconeogenesis fromL(+)- and D(−)lactate in bovine tissues. Comp. Biochem. Physiol. B 77, 365–368.

Harmon, D.L., Britton, R.A., Prior, R.L., Stock, R.A., 1985. Net portal absorption of lactate and volatilefatty acids in steers experiencing glucose-induced acidosis or fed a 70% concentrate diet ad libitum.J. Anim. Sci. 60, 560–569.

Harmon, D.L., Gross, K.L., Krehbiel, C.R., Kreikemeier, K.K., Bauer, M.L., Britton, R.A., 1991.Influence of dietary forage and energy intake on metabolism and acyl-CoA synthetase activity inbovine ruminal epithelial tissue. J. Anim. Sci. 69, 4117–4127.

Harmon, D.L., Richards, C.J., Swanson, K.C., Howell, J.A., Matthews, J.C., True, A.D., Huntington, G.,Gahr, S.A., Russell, R.W., 2001. Influence of ruminal and postruminal starch infusion on visceral glu-cose metabolism in steers. In: Chwalibog, A., Jakobsen, K. (Eds.), Energy Metabolism in Animals:Proceedings of the 15th Symposium on Energy Metabolism in Animals. Wageningen Press,Wageningen, The Netherlands, pp. 273–276.

Henrikson, R.C., Stacy, B.D., 1971. The barrier to diffusion across ruminal epithelium: a study by electronmicroscopy using horseradish peroxidase, lanthanum and ferritin. J. Ultrastruct. Res. 34, 72–82.

Herbein, J.H., Van Maanen, R.W., McGilliard, A.D., Young, J.W., 1978. Rumen propionate and bloodglucose kinetics in growing cattle fed isoenergetic diets. J. Nutr. 108, 994–1001.

Hird, F.J.R., Jackson, R.B., Weidemann, M.J., 1966. Transport and metabolism of fatty acids by isolatedrumen epithelium. Biochem. J. 98, 394–400.

Hird, F.J.R., Symons, R.H., 1961. The mode of formation of ketone bodies from butyrate by tissue fromthe rumen and omasum of the sheep. Biochim. Biophys. Acta 46, 457–467.

Hull, S.R., 1996. Composition of corn steep water during steeping. J Agr. Food Chem. 44, 1857–1863.Huntington, G.B., 1997. Starch utilization by ruminants: from basics to the bunk. J. Anim. Sci. 75, 852–867.


Huntington, G.B., Prior, R.L., 1983. Digestion and absorption of nutrients by beef heifers fed a high con-centrate diet. J. Nutr. 113, 2280–2288.

Huntington, G.B., Reynolds, P.J., 1986. Net absorption of glucose, L-lactate, volatile fatty acids, andnitrogenous compounds by bovine given abomasal infusions of starch or glucose. J. Dairy Sci. 69,2428–2436.

Huntington, G.B., Prior, R.L., Britton, R.A., 1980. Glucose and lactate absorption and metabolic interre-lationships in lambs switched from low to high concentrate diets. J. Nutr. 110, 1904–1913.

Huntington, G.B., Prior, R.L., Britton, R.A., 1981. Glucose and lactate absorption and metabolic interre-lationships in steers changed from low to high concentrate diets. J. Nutr. 111, 1164–1172.

Huntington, G.B., Reynolds, C.K., Stroud, B.H., 1989. Techniques for measuring blood flow in splanch-nic tissues of cattle. J. Dairy Sci. 72, 1583–1595.

Huntington, G.B., Varga, G.A., Glenn, B.P., Waldo, D.R., 1988. Net absorption and oxygen consumptionby Holstein steers fed alfalfa or orchardgrass silage at two equalized intakes. J. Anim. Sci. 66,1292–1302.

Jesse, B.W., Solomon, R.K., Baldwin, R.L., 1992. Palmitate metabolism by isolated sheep rumen epithe-lial cells. J. Anim. Sci. 70, 2235–2242.

Juel, C., 1997. Lactate-proton cotransport in skeletal muscle. Physiol. Rev. 77, 321–358.Kohn, R.A., Kalscheur, K.F., Hanigan, M., 1998. Evaluation of models for balancing the protein require-

ments of dairy cows. J. Dairy Sci. 81, 3402–3414.Kramer, T., Michelberger, T., Gürtler, H., Gäbel, G., 1996. Absorption of short-chain fatty acids across

ruminal epithelium of sheep. J. Comp. Physiol. B 166, 262–269.Krehbiel, C.R., Harmon, D.L., Schneider, J.E., 1992. Effect of increasing ruminal butyrate on portal and

hepatic nutrient flux in steers. J. Anim. Sci. 70, 904–914.Kreikemeier, K.K., Harmon, D.L., Brandt, R.T. Jr., Avery, T.B., Johnson, D.E., 1991. Small intestinal

starch digestion in steers: effect of various levels of abomasal glucose, corn starch and corn dextrininfusion on small intestinal disappearance and net glucose absorption. J. Anim. Sci. 69, 328–338.

Kristensen, N.B., Danfær, A., 2001. The relationship between gastrointestinal production and portalabsorption of short-chain fatty acids in ruminants. In: Chwalibog, A., Jakobsen, K. (Eds.), EnergyMetabolism in Animals: Proceedings of the 15th Symposium on Energy Metabolism in Animals.Wageningen Press, Wageningen, The Netherlands, pp. 277–280.

Kristensen, N.B., Danfær, A., Agergaard, N., 1996a. Diurnal patterns of ruminal concentrations andportal appearance rates of short-chain fatty acids in sheep fed a hay or a concentrate/straw diet in twomeals daily. Acta Agr. Scand. Sect. A 46, 227–238.

Kristensen, N.B., Danfær, A., Agergaard, N., 1998. Absorption and metabolism of short-chain fatty acidsin ruminants. Arch. Tierernähr. 51, 165–175.

Kristensen, N.B., Danfær, A., Tetens, V., Agergaard, N., 1996b. Portal recovery of intraruminally infusedshort-chain fatty acids in sheep. Acta Agr. Scand. Sect. A 46, 26–38.

Kristensen, N.B., Gäbel, G., Pierzynowski, S.G., Danfær, A., 2000a. Portal recovery of short-chainfatty acids infused into the temporarily-isolated and washed reticulo-rumen of sheep. Brit. J. Nutr.84, 477–482.

Kristensen, N.B., Pierzynowski, S.G., Danfær, A., 2000b. Net portal appearance of volatile fatty acids insheep intraruminally infused with mixtures of acetate, propionate, isobutyrate, butyrate, and valerate.J. Anim. Sci. 78, 1372–1379.

Kristensen, N.B., Pierzynowski, S.G., Danfær, A., 2000c. Portal-drained visceral metabolism of3-hydroxybutyrate in sheep. J. Anim. Sci. 78, 2223–2228.

Kristensen, N.B., Steensen, T.H., Pierzynowski, S.G., Danfær, A., 2001. Metabolism of 2-13C-propionatein the rumen epithelium of sheep. J. Anim. Sci. 79, Suppl. 1, 287.

Kung, L. Jr., Ranjit, N.K., 2001. The effect of Lactobacillus buchneri and other additives on thefermentation and aerobic stability of barley silage. J. Dairy Sci. 84, 1149–1155.

Kung, L. Jr., Robinson, J.R., Ranjit, N.K., Chen, J.H., Golt, C.M., Pesek, J.D., 2000. Microbial popula-tions, fermentation end-products, and aerobic stability of corn silage treated with ammonia or apropionic acid-based preservative. J. Dairy Sci. 83, 1479–1486.

Larson, H.J., Jacobson, N.L., McGilliard, A.D., Allen, R.S., 1956. Utilization of carbohydrates introduceddirectly into the omaso-abomasal area of the stomach of cattle of various ages. J. Anim. Sci. 44, 321–330.

Le Bars, H., Lebrument, J., Nitescu, R., Simonnet, H., 1954. Recherches sur la motricité du rumen chezles petits ruminants. IV. Action de l’injection intraveineuse d’acides gras à courte chaîne. Bull. Acad.Vet. 27, 53–67.


Leighton, B., Nicholas, A.R., Pogson, C.I., 1983. The pathway of ketogenesis in rumen epithelium of thesheep. Biochem. J. 216, 769–772.

Leng, R.A., 1970. Glucose synthesis in ruminants. Adv. Vet. Sci. Comp. Med. 14, 209–260.Leng, R.A., Annison, E.F., 1963. Metabolism of acetate, propionate and butyrate by sheep-liver slices.

Biochem. J. 86, 319–327.Manns, J.G., Boda, J.M., 1967. Insulin release by acetate, propionate, butyrate, and glucose in lambs and

adult sheep. Amer. J. Physiol. 212, 747–755.Masson, M.J., Phillipson, A.T., 1952. The composition of the digesta leaving the abomasum of sheep.

J. Physiol. 116, 98–111.McDonald, P., Henderson, A.R., Heron, S.J.E., 1991. The Biochemistry of Silage, 2nd Edition

Chalcombe Publications, Bucks, UK.McLeod, K.R., Baldwin, R.L., Harmon, D.L., Richards, C.J., Rumpler, W.V., 2001. Influence of ruminal

and postruminal starch infusion on energy balance in growing steers. In: Chwalibog, A., Jakobsen, K.(Eds.), Energy Metabolism in Animals: Proceedings of the 15th symposium on Energy Metabolismin Animals. Wageningen Press, Wageningen, The Netherlands, pp. 385–388.

McLeod, K.R., Bauer, M.L., Harmon, D.L., Reynolds, C.K., Mitchell, G.E. Jr., 1997. Effects of exoge-nous somatostatin and cysteamine on net nutrient flux across the portal-drained viscera and liver ofsheep during intraduodenal infusion of starch hydrolysate and casein. J. Anim. Sci. 75, 3026–3037.

Müller, F., Huber, K., Pfannkuche, H., Aschenbach, J.R., Breves, G., Gäbel, G., 2001. Functional char-acterization of the sheep ruminal monocarboxylate transporter (MCT1) using cultured ruminalepithelial cells. Eur. J. Physiol. 441, Suppl., R248.

Nozière, P., Martin, C., Rémond, D., Kristensen, N.B., Bernard, R., Doreau, M., 2000. Effect of compo-sition of ruminally-infused short-chain fatty acids on net fluxes of nutrients across portal-drainedviscera in underfed ewes. Brit. J. Nutr. 83, 521–531.

Orskov, E.R., Benzie, D., Kay, R.N., 1970. The effects of feeding procedure on closure of theoesophageal groove in young sheep. Brit. J. Nutr. 24, 785–795.

Oshio, S., Tahata, I., 1984. Absorption of dissociated volatile fatty acids through the rumen wall of sheep.Can. J. Anim. Sci. 64, Suppl., 167–168.

Pearce, J., Piperova, L.S., 1984. The effects of duodenal glucose and dextrin infusion on adipose tissuemetabolism in sheep. Comp. Biochem. Physiol. B 78, 565–567.

Pennington, R.J., 1952. The metabolism of short-chain fatty acids in the sheep 1. Fatty acid utilizationand ketone body production by rumen epithelium and other tissues. Biochem. J. 51, 251–258.

Peters, J.P., Shen, R.Y.W., Robinson, J.A., Chester, S.T., 1990. Disappearance and passage of propionicacid from the rumen of the beef steer. J. Anim. Sci. 68, 3337–3349.

Pethick, D.W., Lindsay, D.B., 1982. Acetate metabolism in lactating sheep. Brit. J. Nutr. 48, 319–328.Prasad, K.N., Sinha, P., 1976. Effect of sodium butyrate on mammalian cells in culture: a review. In Vitro

12, 125–132.Price, N.T., Jackson, V.N., Halestrap, A.P., 1998. Cloning and sequencing of four new mammalian mono-

carboxylate transporter (MCT) homologues confirms the existence of a transporter family with anancient past. Biochem. J. 329, 321–328.

Prior, R.L., Huntington, G.B., Reynolds, P.J., 1984. Role of insulin and glucose on metabolite uptake bythe hind half of beef steers. J. Anim. Sci. 58, 1446–1453.

Rechkemmer, G., 1991. Transport of weak electrolytes. In: Schultz, S.T., Field, M., Frizzell, R.A. (Eds.),Handbook of Physiology: The Gastrointestinal System IV. American Physiological Society, Bethesda,MD, pp. 371–388.

Reid, J.T., White, O.D., Anrique, R., Fortin, A., 1980. Nutritional energetics of livestock: some presentboundaries of knowledge and future research needs. J. Anim. Sci. 51, 1393–1415.

Rémond, D., Ortiques, I., Jouany, J.P., 1995. Energy substrates for the rumen epithelium. Proc. Nutr. Soc.54, 95–105.

Reynolds, C.K., Huntington, G.B., 1988a. Partition of portal-drained visceral net flux in beef steers. 1.Blood flow and net flux of oxygen, glucose and nitrogenous compounds across stomach and post-stomach tissues. Brit. J. Nutr. 60, 539–551.

Reynolds, C.K., Huntington, G.B., 1988b. Partition of portal-drained visceral net flux in beef steers. 2.Net flux of volatile fatty acids, D-beta-hydroxybutyrate and L-lactate across stomach and post-stomach tissues. Brit. J. Nutr. 60, 553–562.

Reynolds, P.J., Huntington, G.B., 1988c. Net portal absorption of volatile fatty acids and l(+)-lactate bylactating Hostein cows. J. Dairy Sci. 71, 124–133.


Reynolds, C.K., Harmon, D.L., Cecava, M.J., 1994. Absorption and delivery of nutrients for milk proteinsynthesis by portal-drained viscera. J. Dairy Sci. 77, 2787–2808.

Reynolds, C.K., Tyrrell, H.F., 1991. Effects of mesenteric vein L-alanine infusion on liver metabolism inbeef heifers fed on diets differing in forage:concentrate ratio. Brit. J. Nutr. 66, 437–450.

Reynolds, C.K., Tyrrell, H.F., Reynolds, P.J., 1991. Effects of diet forage-to-concentrate ratio and intakeon energy metabolism in growing beef heifers: net nutrient metabolism by visceral tissues. J. Nutr.121, 1004–1015.

Ricks, C.A., Cook, R.M., 1981a. Partial purification of enzymes of bovine kidney mitochondria activat-ing volatile fatty acids. J. Dairy Sci. 64, 2344–2349.

Ricks, C.A., Cook, R.M., 1981b. Regulation of volatile fatty acid uptake by mitochondrial acyl-CoAsynthetases of bovine heart. J. Dairy Sci. 64, 2336–2343.

Ricks, C.A., Cook, R.M., 1981c. Regulation of volatile fatty acid uptake by mitochondrial acyl-CoAsynthetases of bovine liver. J. Dairy Sci. 64, 2324–2335.

Robinson, A.M., Williamson, D.H., 1980. Physiological roles of ketone bodies as substrates and signalsin mammalian tissues. Physiol. Rev. 60, 143–187.

Rupp, G.P., Kreikemeier, K.K., Perino, L.J., Ross, G.S., 1994. Measurement of volatile fatty acid disap-pearance and fluid flux across the abomasum of cattle, using an improved omasal cannulationtechnique. Amer. J. Vet. Res. 55, 522–529.

Russell, J.B., Wallace, R.J., 1988. Energy yielding and consuming reactions. In: Hobson, P.N. (Ed.), TheRumen Microbial Ecosystem. Elsevier Applied Science, London, pp. 185–215.

Russell, R.W., Gahr, S.A., 2000. Glucose Availability and Associated Metabolism: Farm AnimalMetabolism and Nutrition. CABI Publishing, Edinburgh, UK, pp. 121–147.

Russell, R.W., Moss, L., Schmidt, S.P., Young, J.W., 1986. Effects of body size on kinetics of glucosemetabolism and on nitrogen balance in growing cattle. J. Nutr. 116, 2229–2243.

Sander, E.G., Warner, R.G., Harrison, H.N., Loosli, J.K., 1959. The stimulatory effect of sodium butyrate andsodium propionate on the development of rumen mucosa in the young calf. J. Dairy Sci. 42, 1600–1605.

Schaadt, H., 1968. Effects of maturity, fermentation time and urea treatment on D(−) and L(+) lactate incorn silage. J. Dairy Sci. 51, 802–805.

Schmidt, S.P., Keith, R.K., 1983. Effects of diet and energy intake on kinetics of glucose metabolism insteers. J. Nutr. 113, 2155–2163.

Seal, C.J., Parker, D.S., 1994. Effect of intraruminal propionic acid infusion on metabolism of mesen-teric- and portal-drained viscera in growing steers fed a forage diet. I. Volatile fatty acids, glucose,and lactate. J. Anim. Sci. 72, 1325–1334.

Seal, C.J., Parker, D.S., 2000. Influence of gastrointestinal metabolism on substrate supply to the liver.In: Cronjé, P.B. (Ed.), Ruminant Physiology: Digestion, Metabolism, Growth and Reproduction.CABI Publishing, Wallingford, UK, pp. 131–148.

Seal, C.J., Reynolds, C.K., 1993. Nutritional implication of gastrointestinal and liver metabolism in rumi-nants. Nutr. Res. Rev. 6, 185–208.

Sehested, J., Diernæs, L., Møller, P.D., Skadhauge, E., 1999a. Ruminal transport and metabolism ofshort-chain fatty acids (SCFA) in vitro: effect of SCFA chain length and pH. Comp. Biochem.Physiol. A 123, 359–368.

Sehested, J., Diernæs, L., Møller, P.D., Skadhauge, E., 1999b. Transport of butyrate across the isolatedbovine rumen epithelium: interaction with sodium, chloride and bicarbonate. Comp. Biochem.Physiol. A 123, 399–408.

Sheperd, A.C., Maslanka, M., Quinn, D., Kung, L. Jr., 1995. Additives containing bacteria and enzymesfor alfalfa silage. J. Dairy Sci. 78, 565–572.

Shirazi-Beechey, S.P., Kemp, R.B., Dyer, J., Beechey, R.B., 1989. Changes in the functions of the intes-tinal brush border membrane during the development of the ruminant habit in lambs. Comp.Biochem. Physiol. B 94, 801–806.

Siciliano-Jones, J., Murphy, M.R., 1989. Production of volatile fatty acids in the rumen and cecum-colonof steers as affected by forage:concentrate and forage physical form. J. Dairy Sci. 72, 485–492.

Stangassinger, M., Beck, U., Emmanuel, B., 1979. Is glucose ketogenic in rumen epithelium? Ann. Rech.Vét. 10, 413–416.

Stangassinger, M., Giesecke, D., 1979. Quantitative measurement of gluconeogenesis from isobutyratein sheep. Arch. Int. Physiol. Biochim. 87, 265–274.

Stern, J.R., Coon, M.J., Del Campillo, A., Schneider, M.C., 1956. Enzymes of fatty acid metabolism. IV.Preparation and properties of coenzyme A transferase. J. Biol. Chem. 221, 15–31.


Steven, D.H., Marshall, A.B., 1970. Organization of the rumen epithelium. In: Phillipson, A.T. (Ed.),Physiology of Digestion and Metabolism in the Ruminant. Oriel Press, Newcastle-upon-Tyne, UK,pp. 80–100.

Stevens, C.E., 1970. Fatty acid tranport through the rumen epithelium. In: Phillipson, A.T. (Ed.),Physiology of Digestion and Metabolism in the Ruminant. Oriel Press, Newcastle upon Tyne, UK,pp. 101–112.

Stevens, C.E., Stettler, B.K., 1967. Evidence for active transport of acetate across bovine rumen epithe-lium. Amer. J. Physiol. 213, 1335–1339.

St. Jean, G.D., Rings, D.M., Schmall, L.M., Hoffsis, G.F., Hull, B.L., 1989. Jejunal mucosal lactaseactivity from birth to 3 weeks in conventionally raised calves. Amer. J. Vet. Res. 50, 1496–1498.

Sutton, J.D., 1985. Digestion and absorption of energy substrates in the lactating cow. J. Dairy Sci. 68,3376–3393.

Synderman, S.E., Sansariq, C., Middleton, B., 1998. Succinyl-CoA:3-ketoacid CoA-transferase deficiency.Pediatrics 101, 709–711.

Taniguchi, K., Huntington, G.B., Glenn, B.P., 1995. Net nutrient flux by visceral tissues of beef steersgiven abomasal and ruminal infusions of casein and starch. J. Anim. Sci. 73, 236–249.

Thorlacius, S.O., Lodge, G.A., 1973. Absorption of steam-volatile fatty acids from the rumen of the cowas influenced by diet, buffers and pH. Can. J. Anim. Sci. 53, 279–288.

Topping, D.L., Clifton, P.M., 2001. Short-chain fatty acids and human colonic function: roles of resist-ant starch and nonstarch polysaccharides. Physiol. Rev. 81, 1031–1064.

Van Maanen, R.W., Herbein, J.H., McGilliard, A.D., Young, J.W., 1978. Effects of monensin on in vivorumen propionate production and blood glucose kinetics in cattle. J. Nutr. 108, 1002–1007.

Varga, G.A., Tyrrell, H.F., Huntington, G.B., Waldo, D.R., Glenn, B.P., 1990. Utilization of nitrogen andenergy by Holstein steers fed formaldehyde- and formic acid-treated alfalfa or orchardgrass silage attwo intakes. J. Anim. Sci. 68, 3780–3791.

Vessy, D.A., Kelly, M., Warren, R.S., 1999. Characterization of the CoA ligases of human liver mito-chondria catalyzing the activation of short- and medium-chain fatty acids and xenobiotic carboxylicacids. Biochim. Biophys. Acta 1428, 455–462.

Webster, L.T. Jr., Gerowin, L.D., Rakita, L., 1965. Purification and characteristics of a butyryl coenzyme Asynthetase from bovine heart mitochondria. J. Biol. Chem. 240, 29–33.

Weekes, T.E.C., 1974. The in vitro metabolism of propionate and glucose by the rumen epithelium.Comp. Biochem. Physiol. B 49, 393–406.

Weekes, T.E.C., Webster, A.J.F., 1975. Metabolism of propionate in the tissues of the sheep gut. Brit. J. Nutr.33, 425–438.

Weigand, E., Young, J.W., McGilliard, A.D., 1972. Extent of propionate metabolism during absorptionfrom the bovine ruminoreticulum. Biochem. J. 126, 201–209.

Weigand, E., Young, J.W., McGilliard, A.D., 1975. Volatile fatty acid metabolism by rumen mucosa fromcattle fed hay or grain. J. Dairy Sci. 58, 1294–1300.

Williamson, D.H., Bates, M.W., Krebs, H.A., 1968. Activity and intracellular distribution of enzymes ofketone-body metabolism in rat liver. Biochem. J. 108, 353–361.

Young, J.W., 1976. Gluconeogenesis in cattle: significance and methodology. J. Dairy Sci. 60, 1–15.Young, J.W., Thorp, S.L., De Lumen, H.Z., 1969. Activity of selected gluconeogenic and lipogenic

enzymes in bovine rumen mucosa, liver and adipose tissue. Biochem. J. 114, 83–88.


PART VMethodology

435

18 Methodological approaches to metabolism research

X. Guan and D. G. Burrin

USDA/ARS Children’s Nutrition Research Center, Department of Pediatrics,Baylor College of Medicine, Houston, TX 77030, USA

Advances in molecular biology and stable isotope techniques during the last decade have ledto an explosion of research aimed at understanding the biological basis of metabolomics fromthe level of systemic physiology, to intermediary metabolism, to molecular regulation of crit-ical proteins, and on down to genomic expression. We shall highlight principles, approaches,and applications of these cutting-edge molecular and metabolic techniques.

1. GENETICALLY ENGINEERED ANIMAL MODELS FOR METABOLISM RESEARCH

With the development of transgenic technologies (gene overexpression, knockout, and condi-tional expression), one is able to explore physiological roles and metabolic functions ofspecific genes and to identify individual proteins involved in the control of specific aspects ofmetabolism.

1.1. Transgenic techniques

Conventional transgenic technologies (gene overexpression and knockout) are invaluable formodeling genetic disorders and addressing developmental questions. However, this “all ornothing” mode is inflexible and cannot be used to fully answer subtle metabolic questions. Inorder to obtain precise information about the roles of a specific gene in a specific cell type ata critical stage of disease or development, conditional transgenic techniques that allow flexi-ble spatial and temporal control of gene deletion or expression in transgenic animals must beused (Ryding et al., 2001). In these systems, the switching “on” or “off” of the expression ofa particular gene is conditional upon exposure to a specific stimulus (Ryding et al., 2001).

Three approaches have been used to inhibit specific gene expression in mammalian systems.First, the most common approach is specific gene ablation by homologous recombination inembryonic stem (ES) cells (Bronson and Smithies, 1994) and then reproduction of animals



without expression of the specific gene. Targeted gene disruption by the homologous recom-bination technique takes advantage of the fact that pluripotent ES cells derived from mouseblastocysts can be cultured in vitro and remain viable for differentiation after injection into adifferent embryo. The most commonly used ES cells are those derived from mice that have anagouti coat color. These ES cells can be microinjected into embryos obtained from mice thathave a black coat color. Offspring with a high degree of agouti coat color, indicating the trans-mission of ES cell-derived genes, can then be crossbred to obtain mice with a geneticbackground identical to that of the ES cells. Therefore, mice with specific gene modificationscan be obtained by manipulation of the ES cell genome. Modification of specific genes in theES cell genome depends on the ability of transfected DNA to recombine with the homologousgene in the chromosome. Isogenic DNA for the targeting construct is used to maximizehybridization of the targeting DNA to the proper gene locus in the chromosome. A selectablegene marker, such as the neomycin-resistant gene, is inserted into an exon to disrupt the codingsequence of the gene of interest. The chimeric targeting gene construct is used to transfect EScells. Homologous recombination of the transfected DNA with chromosomal DNA at thetarget locus results in the replacement of a portion of the endogenous gene with the targetingconstruct, thus disrupting the coding sequence and inactivation of the endogenous gene. Theuse of the selectable gene marker allows the selection for cells that have taken up and expressedthe transfected DNA. Growth of the ES cells in the presence of antibiotic indicates the inte-gration of the transfected DNA into the ES cell genome. However, there are two limitations inthis approach: the low rate of homologous recombination in mammalian cells and the high rate of random (nontargeted) integration of the vector DNA. Chimeric nucleases and triplex-forming oligonucleotides may increase homologous recombination and decrease randomintegration in cells (Vasquez et al., 2001).

Secondly, anti-sense DNA or RNA that inhibits gene expression by complementation tosingle-stranded mRNA (Izant and Weintraub, 1985), and trans-splicing ribozymes (Kohler et al., 1999) that catalyze RNA hydrolysis in a sequence-specific manner, have been used successfully to abolish gene expression in mammalian cells. Anti-sense RNA is useful forsuppressing the expression of specific genes in vivo. The anti-sense plasmid construct can beintroduced into eukaryotic cells by transfection or microinjection. Anti-sense transcripts com-plementary to 5′ untranslated target gene mRNA specifically suppress gene activity or directagainst the protein-coding domain alone. Recently, trans-splicing ribozymes have beenemployed to repair mutant mRNAs in vivo. These trans-splicing ribozymes contain catalyticsequences derived from a self-splicing group I intron, which have been adapted to a chosentarget mRNA by fusion of a region of extended complementarity to the target RNA and pre-cise alteration of the guide sequences required for substrate recognition. The improvedtrans-splicing ribozymes may be tailored for virtually any target RNA, and provide a new toolfor triggering gene expression in specific cell types.

Thirdly, RNA interference (RNAi), an evolutionarily conserved pathway, uses these smallinterfering RNAs to degrade mRNAs before translation. Recently, RNAi has emerged as aspecific and efficient method to silence gene expression in mammalian cells and to probegene function on a whole-genome scale either by transfection of short interfering RNAs or bytranscription of short hairpin RNAs (Hammond et al., 2001; Hannon, 2002; McCaffrey et al.,2002). Short interfering RNAs typically consist of two 21-nucleotide (nt) single-strandedRNAs that form a 19-bp duplex with 2-nt 3′ overhangs. Its antisense strand is used by anRNAi silencing complex to guide mRNA cleavage, so promoting mRNA degradation. It iscertain that the ability of RNAi technology to silence specific genes will transform futurestudies of cellular systems and biology in mammalian cells (McManus and Sharp, 2002).

X. Guan and D. G. Burrin436

1.2. Applications of genetically engineered animal models

The use of nitric oxide synthase (NOS) gene knockout animals has helped elucidate the roles ofdifferent NOS isoforms in the synthesis and function of nitric oxide. Nitric oxide from neuronalNOS is a major inhibitory neurotransmitter; nitric oxide from endothelial NOS regulates bloodflow under physiological conditions; and nitric oxide from inducible NOS causes hypotensionduring severe inflammatory conditions. Moreover, nitric oxide from each isoform has uniqueroles in tissue injury and inflammation. The neuronal NOS-deficient mice develop gastricdilatation and stasis; the endothelial NOS-deficient mice develop hypotension and lackvasodilatory responses to injury; and inducible NOS-deficient mice are more susceptible toinflammatory damage but more resistant to septic shock (Mashimo and Goyal, 1999). Thisexample clearly demonstrates the enormous potential of genetically engineered mice lackingspecific genes in elucidation of mechanisms specific in physiology and pathology.

Another example is that gene silencing by siRNAs has provided insights into insulin regu-lation of glucose uptake and glycogen synthesis. The serine/threonine protein kinase Akt hasbeen proposed to mediate insulin signaling in several processes. However, it is unclear if Aktis involved in insulin-stimulated glucose uptake, and which isoforms of Akt are responsiblefor each insulin action. Recently, experiments with isoform-specific siRNA have revealed thatAkt2, and Akt1 to a lesser extent, has an essential role in insulin-stimulated glucose trans-porter-4 translocation and 2-deoxyglucose uptake in both Chinese hamster ovary cells and3T3-L1 adipocytes, while Akt1 and Akt2 contribute equally to insulin-stimulated glycogensynthesis. These data suggest a prerequisite role of Akt in insulin-stimulated glucose uptakeand distinct functions among Akt isoforms (Katome et al., 2003).

2. GENE EXPRESSION TECHNIQUES FOR METABOLISM RESEARCH

The transcription of genomic DNA to produce mRNA is the first step in the process of pro-tein synthesis, and differences in gene expression are indicative of cellular responses toenvironmental stimuli and perturbations and are responsible for both morphological and phe-notypic differences between tissues and stages of development. Knowing when, where, andto what extent a gene is expressed is central to understanding the activity and biological rolesof its encoded protein. In addition, changes in the multigene patterns of expression can pro-vide clues about regulatory mechanisms and broader cellular functions and biochemicalpathways (Lockhart and Winzeler, 2000).

2.1. Gene expression techniques

There are many techniques for measuring gene expression. Both conventional methods(including Northern blots, RNase protection assay, in situ hybridization, and RT-PCR) andDNA microarrays have been employed to measure expression levels of specific genes, tocharacterize global expression profiles, and to screen for differences in mRNA abundance.These conventional methods are simply used in a more targeted fashion to follow up on thespecific genes, pathways, and mechanisms implicated by the microarrays.

2.1.1. DNA microarray analysis for screening global gene expression profile

DNA microarrays are a miniaturized, ordered arrangement of nucleic acid fragments from indi-vidual genes located at defined positions on a solid support, enabling the expression analysis of

Methodological approaches to metabolism research 437

thousands of genes in parallel by specific hybridization (Arcellana-Panlilio and Robbins,2002). This technology is a powerful tool for rapid, comprehensive, and quantitative analysisof global gene expression profiles of normal/disease statuses and developmental processes(Bednar, 2000; Lockhart and Winzeler, 2000). In general, there are two kinds of DNAmicroarrays: cDNA arrays and oligonucleotide arrays. For cDNA arrays, the nucleic acidfragments are spotted robotically onto a glass slide. The cDNA used for spotting are usuallyderived by PCR amplification of cDNA libraries. For oligonucleotide arrays, oligonucleotidesare synthesized in situ by photolithography. Gene expression analysis using DNA micro-arrays is based on the competitive hybridization of differently labeled populations of cDNAs.Fluorescent dyes, usually Cy3 and Cy5, are used to label cDNA pools reverse transcribedfrom different mRNA samples (prepared from tissues or cells). The labeled cDNAs areapplied to the microarray and allowed to simultaneously hybridize under conditions analo-gous to those established for Southern blotting. After washing off nonspecific hybridization,the slide is read in a confocal laser scanner that can differentiate between Cy3 and Cy5 signals.Because hybridization is governed by the recognition rules, the signal intensity at each posi-tion gives not only a measurement of the number of molecules bound, but also the likelyidentity of the molecules. Thus, the relative intensity of Cy5/Cy3 signal for each gene is usedto assess the relative abundance of a specific mRNA. It should be noted that the extent ofhybridization on a DNA microarray is influenced by time, concentration of solution-phasecDNA probes, and length of the arrayed DNA sequences (Stillman and Tonkinson, 2001).

2.1.2. mRNA quantitative techniques for measuring specific gene expression

2.1.2.1. Northern blotting analysis The Northern blotting analysis separates RNAspecies on the basis of size by denaturing gel electrophoresis followed by transfer of the RNAonto a nylon membrane by capillary, vacuum, pressure, or electrical-assisted blotting. TheRNA is then irreversibly bound to the membrane by exposure to short-wave ultraviolet lightor by heating at 80°C in a vacuum oven. The RNA sequences of interest are detected on themembrane by hybridization to a specific labeled probe. Probes for Northern blot detectiongenerally contain full or partial cDNA sequences and may be labeled by enzymatic incorpo-ration of radiolabeled (32P) nucleotides or with nucleotides conjugated to haptens such asbiotin or digoxigenin. After washing off the unbound and nonspecifically bound probe, thehybridization signal is generally revealed by autoradiography or immunological detectionafter antibody incubation. Autoradiograph band intensities may be quantified by densitometry,by direct measurement of hybridized radiolabeled probe via storage phosphor imaging, or byscintillation counting of excised bands (for the technique in detail, see Rapley and Walker,1998). The band identified by the probe indicates the size of the mRNA, and the intensity ofthe band corresponds to the relative abundance.

The Northern blotting analysis can detect the steady-state level of a specific mRNAsequence in the sample. Association of the mRNA expression and the metabolic/physiologicalstate provides important clues regarding gene regulation, developmental characterization, andresponsiveness to stimuli. The abundance of mRNA is controlled by three major factors: genetranscription, mRNA processing and transport, and mRNA stability. More sensitive methodscan be used for the analysis of rare transcripts including RT-PCR and RNase protection assay.However, the Northern blotting analysis is the only method that can determine mRNA size.In general, this method is semi-quantitative if a standard is used, and is suitable for deter-mining relative abundances of mRNA species. To compare the relative abundances, eachsample on a membrane must be hybridized with a probe for the specific mRNA of interest and


a probe for an endogenous internal control. Constitutively expressed “housekeeping” genes,such as β-actin, cyclophilin, or glyceraldehyde-3-phosphate dehydrogenase (GAPDH), and aconstant level of 18S rRNA are used as the internal control.

Variations of the Northern blotting technology, such as dot blots, slot blots, and fast blots,have been developed to simplify blot preparation and improve quantitative analysis. Thesetechniques involve applying an RNA sample (dot and slot blots) or cell extract (fast blot)directly to the membrane without prior size fractionation on a gel. It is imperative that theprobe used for dot/slot blot analyses is specific for the target mRNA without cross-hybridizationor nonspecific hybridization to other sequences.

2.1.2.2. RNase protection assay Ribonuclease (RNase) protection assay is a techniqueused for detection and quantitative analysis of specific RNAs. In principle, cellular mRNA ishybridized with a gene-specific labeled single-stranded complementary RNA (labeled with32P). After the hybridization, all unbound single-stranded RNA molecules are degraded bysingle-stranded-specific ribonuclease. The protected double-stranded (i.e. hybridized) frag-ments are separated by denaturing polyacrylamide gel electrophoresis, detected by exposureon x-ray film, and quantified by densitometry, or quantified by excising and scintillationcounting the region of the gel that contain the protected fragments. The size of each protectedfragment may be derived from the standard marker and the intensity of the bands directly corresponds with the absolute concentration of the specific mRNA. Unlike Northern blots, thesize of product by the RNase protection assay does not depend on the size of the targetmRNA, but on the size of the probe used in the assay. This assay can determine absoluteabundance of mRNA at relatively high sensitivity (Reue, 1998).

2.1.2.3. In situ hybridization In situ hybridization (ISH) technique allows specificnucleic acid sequences to be detected in morphologically preserved chromosomes, cells, ortissue sections. In combination with immunocytochemistry, this technique can relate micro-scopic topological information to gene activity at the DNA, mRNA, and protein level.Localization of gene expression at the mRNA level is particular important to confirm theidentity of cells expressing soluble or secreted proteins. Currently, nonradioactive labeledcRNA probes have become more feasible for detecting target mRNA in tissue sections. Forexample, digoxigenin (DIG)-labeled nucleotides may be incorporated at a defined densityinto nucleic acid probes by DNA polymerases, RNA polymerases, or terminal transferase.Usually, cRNA probes are generated by in vitro transcription from a linearized DNA tem-plate. Hybridized DIG-labeled probes may be detected with high-affinity anti-DIG antibodiesthat are conjugated to alkaline phosphatase (AP) or horseradish peroxidase. The antibodiesconjugated to AP can be visualized with colorimetric or fluorescent AP substrates. The adventof the tyramide signal amplification (TSA) method has dramatically increased the sensitivityof nonradioactive ISH detection. Tyramide signal amplification is based on the horseradishperoxidase-catalyzed deposition of labeled tyramide molecules at sites of probe binding. Incontrast, typical AP substrates precipitate diffusely at sites of AP activity. Dual fluorescentISH and immunohistochemistry using TSA has provided a rapid and sensitive method tocompare mRNA and protein localization (Zaidi et al., 2000), which offers the ability to dis-tinguish between the cells responsible for production of the protein and its target cells.

2.1.2.4. Quantitative RT-PCR The reverse transcription polymerase chain reaction (RT-PCR) is the most sensitive method for the detection of low abundance of steady-statemRNA (Wang and Brown, 1999). The RT-PCR is an in vitro method for enzymatically ampli-fying defined sequences of RNA and permits the analysis of different samples from as little


as one cell. There are four types of RT-PCR: relative, competitive, comparative, and real-timeRT-PCR.

The first step in the RT-PCR is the reverse transcription of the RNA template into cDNA, fol-lowed by its exponential amplification in a PCR. The reverse transcription step can be primedusing specific primers, random hexamers, or oligo-dT primers. In theory, PCR should detect thecDNA derived from a single mRNA molecule; but in practice, ten or more mRNA copies arerequired because of the relative inefficiency of the reverse transcriptase reaction required to con-vert mRNAto cDNA for subsequent amplification. Typically, a small amount of total RNA (1 μgor less) is used for reverse transcription, and a fraction (1/20 to 1/50) of the resulting cDNA isused in the PCR. The cDNA is amplified exponentially via cycles of denaturation, annealing,and extension. Amplification products initially appear at undetectable levels, then accumulate ata nonlinear rate within an exponential phase, and eventually reach similar levels irrespective ofinitial template concentration. Thus, quantitative comparisons must be made during the expo-nential phase. One strategy to ensure that PCR products are analyzed within the exponentialphase of the amplification reaction is to examine products at progressive cycles during the reac-tion. This may be accomplished by the use of real-time quantitative RT-PCR, wherein the wholereaction is monitored rather than just the end product. Real-time RT-PCR employs a fluorescentsignal to report formation of PCR product as each cycle of the amplification proceeds, coupledwith an automated PCR/fluorescent detection system (Heid et al., 1996). For absolute quantita-tive analysis of a target mRNA, an internal control template and corresponding control probewith a unique reporter fluorescent dye is included in each reaction tube (Gibson et al., 1996; fora review, see Bustin, 2000). It should be noted that real-time RT-PCR quantifies steady-statemRNA levels, which tells the researcher nothing about either transcription levels or mRNA stability (Bustin, 2002).

2.1.2.5. The method of choice The choice of mRNA quantitative analysis method isdependent on the study of interest. (1) The Northern blotting analysis is the first step in thecharacterization of mRNA expression as it allows visualization of intact mRNA. That is the only method providing information about the mRNA size, alternative splicing, and theintegrity of the RNA. That also allows great flexibility, as the probe used for hybridizationdoes not require preparation with specific cloning vectors or primers. (2) The RNase protec-tion assay is the most useful for mapping transcript initiation and termination sites andintron/exon boundaries, and for discriminating between related mRNAs of similar size, whichwould migrate at similar positions on a Northern blot. (3) In situ hybridization is the mostcomplex method of all, but is the only one that allows localization of transcripts to specificcells within a tissue. (4) In term of sensitive, specific, and reproducible quantification ofmRNA, real-time RT-PCR is the method of choice (Bustin, 2000). The RNase protectionassay and real-time RT-PCR are most readily applied to the analysis of mRNAs that havebeen previously characterized and sequenced, as they require production of specific vectorsand primers for probe and control template preparation (Reue, 1998).

2.2. DNA binding assays for assessing DNA–protein interactions

It has been known that, at the simplest level, transcription of genes into mRNAs is governedby transcription factors, which bind to cis-regulatory regions of the DNA in the vicinity of thetarget gene. The current challenges, however, include an understanding of (1) which specificcis-acting DNA sequence elements and which trans-acting factors (transcription factors) arerequired for the expression of a given gene; (2) how a given set of DNA–protein interactions


regulates the expression of a tissue-specific gene; and (3) how these interactions are integratedinto the overall regulation of gene expression during development (Yang, 1998). The highlyspecific interaction between a given transcription factor and its recognized binding sequence(DNA) forms the basis for the biochemical characterization, and provides insight into theoverall molecular mechanisms controlling gene expression.

2.2.1. Electrophoretic mobility shift assay (EMSA)

This assay consists of three steps. (1) A DNA-binding protein (present in a nuclear extract) ismixed with a 32P-labeled DNA fragment (probe). (2) The DNA–protein complexes migratemore slowly than free, unbound DNA during electrophoresis. (3) The two bands containingradiolabeled DNA are detected by autoradiography or a phosphor screen. In general, eachadditional protein binding to a DNA–protein complex alters its electrophoretic mobility andresults in an additional retarded band. The EMSA is sufficiently sensitive for the binding ofa monoclonal antibody to the protein–DNA complex to cause a supershift band, which canconfirm the presence of a particular protein in the complex. Though it provides a quantitativemeasurement of the amount of a particular DNA binding activity, the EMSA does not give a direct readout of the DNA nucleotides that the protein recognizes.

2.2.2. DNase I protection (footprinting) assay

A specific binding protein (in a nuclear extract) binds to a specific region within a singly end-labeled DNA fragment (probe). After digestion by DNase I, the DNA products are elec-trophoresed in a denaturing polyacrylamide gel. In the absence of any binding protein, thebands appear as a ladder without any interruption. However, in the presence of the specificbinding protein, some bands disappear because DNase I cannot digest the region of DNAbound by the protein. This assay allows the determination of a short stretch of a protein-bindingsite within a relatively large DNA fragment. The exact nucleotide sequence in the protectedregion can readily be determined by concurrently running sequencing reactions of the sameDNA fragment alongside the DNase I digestion products.

2.2.3. Chromatin immunoprecipitation (ChIP) assay

An issue in gene transcription is the in vivo relevance of transcription factor binding sites thathave been identified in vitro. The ChIP assay is being successfully exploited to confirm in vivo binding sites of specific transcription factors. In this assay, an antibody to a specificDNA-binding protein is used to immunoprecipitate cross-linked protein–DNA complexes.Then, the DNA is experimentally released from the complexes and detected by DNA footprinting (Lee Kang et al., 2002) or DNA microarray (Weinmann et al., 2002). In combi-nation with the DNA microarray, the ChIP assay is used to probe the genome-wide pattern of DNA binding sites for specific transcription factors (Weinmann et al., 2002). Moreover,this technique can distinguish the direct targets of the transcription factors from indirectdownstream effects (Shannon and Rao, 2002).

2.3. Applications: nutrient regulation of gene expression

Effects of nutrition can be exerted at many stages between transcription of the geneticsequence and translation of a functional protein. Nutrients can influence gene expression


through control of the regulatory signals in the untranslated regions of the gene (Hesketh et al., 1998). In the research of nutritional control of gene expression, it is important not onlyto focus on regulation through gene promoter regions but also to consider the possibility ofpost-transcriptional control (Hesketh et al., 1998).

An example of nutrient regulation of gene expression is that the polyunsaturated fatty acid(PUFA) upregulates the expression of genes encoding proteins involved in fatty acid oxida-tion while simultaneously downregulating the expression of genes encoding proteinsinvolved in lipid synthesis. The PUFAs appear to regulate gene transcription by modifyingthe DNA binding activity and/or the nuclear abundance of the transcription factors (Clarke,2001). Furthermore, PUFAs govern the expression of enzymes in lipid oxidation and lipidsynthesis by two independent mechanisms: activating peroxisome proliferator-activatedreceptor α (Clarke, 2001) and suppressing sterol regulatory element binding protein-1 (Xu et al., 1999, 2001). Duplus et al. (2000) have postulated multiple mechanisms for fattyacid control of gene transcription. One of them is that the fatty acid itself or its derivativeacts as a ligand for a transcription factor, which then can bind to DNA at a fatty acidresponse element in the fatty acid-responsive gene and activate or repress transcription(Duplus et al., 2000).

Another example is that amino acid availability regulates the expression of genes encod-ing proteins in the control of growth (Fafournoux et al., 2000). Limitation of several aminoacids greatly increases the expression of a specific gene encoding the CHOP protein, a stress-inducible nuclear protein that dimerizes with members of the CCAAT/enhancer-bindingprotein (C/EBP) family of transcription factors (Bruhat et al., 1997; Fafournoux et al., 2000).Elevated abundance of CHOP mRNA results from an increased rate of its transcription andan increased stability of its mRNA (Bruhat et al., 1997, 1999). The C/EBP family is involvedin the regulation of processes relevant to gene expression, energy metabolism, cellular proliferation, and differentiation (Roesler, 2001). By forming heterodimers with members ofthe C/EBP family, the CHOP protein either as an inhibitor or an activator can influenceexpression of cell type-specific genes (Ubeda et al., 1996; Sok et al., 1999). In the promoterregion of the CHOP gene, an amino acid response element (AARE) is found to bind in vitrothe activating transcription factor 2, which is essential for leucine-induced transcriptionalactivation of the CHOP gene (Bruhat et al., 2000). Further work will be necessary to char-acterize the molecular steps by which the cellular amino acid availability can regulate geneexpression, particularly to determine (1) the pattern of the AARE in the regulated genes; (2) the nature of the protein complexes bound to these elements; (3) the identity of the intra-cellular metabolites that mediate transcriptional activation by amino acid limitation; and (4) the signaling pathways involved in the control of translation by amino acids (Fafournouxet al., 2000). These studies will eventually provide insight into the role of amino acids in the regulation of cellular functions such as protein synthesis and proteolysis (Bruhat et al., 2002).

3. PROTEIN ABUNDANCE, ACTIVITY, AND LOCALIZATION

Molecular mechanisms that govern cellular function and metabolism are controlled largely bythe structure and function of genetically encoded products, the proteins. Post-transcriptionalprocessing of mRNA and co-/post-translational processing of proteins lead to a fair degree ofdiscordance between the open reading frames predicting protein structure and the actual func-tional product (Witzmann and Li, 2002). Consequently, it is necessary to quantitatively


measure differential expression at the protein level. Moreover, many protein-mediated cellular functions are managed and regulated through mechanisms that do not even involvequantitative changes in protein expression. Instead, they are the consequences of proteininteractions and chemical modifications of existing proteins (e.g. phosphorylation and glyco-sylation). It is essential to characterize these changes of the proteins using functional andstructural proteomics. Finally, the localization of gene products, which is often difficult topredict from the DNA sequence, can be determined experimentally only at the protein level(Pandey and Mann, 2000).

3.1. Two-dimensional gel electrophoresis for screening protein expression profiles

Two-dimensional (2D) gel electrophoresis/mass spectrometry can be used to visualize differ-ential protein expression. In the 2D electrophoresis, proteins are subjected to orthogonalseparation methods, the first based on protein charge via isoelectric focusing and then bymass in sodium dodecyl sulfate. The final product of the 2D electrophoresis separation isessentially an in-gel array of proteins, each assuming a coordinate position corresponding tothe unique combination of isoelectric point and mass. Protein expression patterns are visual-ized by a number of staining methods such as fluorescent staining image analysis. Finally, the identity of the protein(s) in each spot is characterized by liquid chromatography–massspectrometry (fig. 1).


Fig. 1. A schematic showing the two-dimensional gel approach. Cells (or tissue) derived from two different conditions, A and B, are harvested and the proteins solubilized. The crude protein mixture is then applied to a “first dimension” gel strip that separates the proteins based on their isoelectric points. After this step, thestrip is subjected to reduction and alkylation and applied to a “second dimension” SDS–PAGE gel where proteins are denatured and separated on the basis of size. The gels are then fixed and the proteins visualizedby staining methods. After staining, the resulting protein spots are recorded and quantified. The spots of interest are then excised and subjected to mass spectrometric analysis; Reproduced with permission fromPandey and Mann (2000).

3.2. Protein microarrays for screening protein expression profiles

Protein microarrays are being developed for high-throughput analysis of protein expression.There are two types of protein microarrays. One type of protein array is termed a proteinfunction array, and consists of thousands of native proteins, recombinant proteins, or theirdomains immobilized in a defined pattern, which can be used to examine protein function(e.g. enzymatic activity or binding property). This type of array is incubated with a cell lysatecontaining putative interaction partners. After washing away unbound material, the boundproteins are eluted and then identified by mass spectrometry (Pandey and Mann, 2000). Theother type of protein array is termed a protein-detecting array and consists of large numbersof protein-binding agents, which allows for screening protein expression profiles under vari-ous physiological stimuli (Kodadek, 2001). For example, based on antibody–antigeninteractions, proteins isolated from cells in a particular physiological state are bound to anarray containing specific antigens or antibodies. The extent of the specific binding is thendetected by the fluorescence assay (Haab et al., 2001) or by the enhanced chemiluminescenceassay (Huang, 2001).

3.3. Western blot analysis for measuring specific protein expression

In Western blotting, a complex protein fraction is separated by electrophoresis and the pro-teins are transferred to a PVDF or nitrocellulose membrane, which is then hybridized with a primary antibody and visualized using a secondary antibody conjugated with horseradishperoxidase or alkaline phosphatase (Rapley and Walker, 1998).

3.4. Enzyme-linked immunosorbent assay (ELISA) for measuring specific protein activity

In the ELISA, the antigen to be detected, being passively attached to the plastic surface ofmicroplate wells, binds specifically to an antibody conjugated with an enzyme used for detec-tion (e.g. horseradish peroxidase or alkaline phosphatase). The antigen–enzyme linkedantibody complex is then reacted with a substrate/chromophore. The rate of color change,resulting from substrate metabolism by the enzyme, is proportional to the amount of enzymein the complex. Many modified ELISAs have been developed to detect and quantify specificproteins (Rapley and Walker, 1998).

3.5. Confocal laser scanning microscopy (CLSM) for visualizing specific protein location

The CLSM captures only the light coming immediately from the object point in focus andobstructs the light coming from out-of-focus areas of the sample. A laser beam is concen-trated on a very small spot and then scans the sample in the X–Y direction. As a result, thepart corresponding to the eliminated light is darkened in the image, making it possible tooptically slice a thick tissue sample. It detects the fluorescence or transmits light from thesample, and displays an image on the monitor. The CLSM has high contrast and superiorresolution in the light axis direction because of the use of confocal optics. In combinationwith immunohistochemistry, the CLSM provides specific information about protein expres-sion patterns at the single-cell level and may indicate molecular changes relevant tometabolism.


4. DNA AND RNA LABELING TECHNIQUES FOR MEASURING CELL PROLIFERATION AND APOPTOSIS IN VIVO

4.1. Cell proliferation

4.1.1. Bromodeoxyuridine (BrdU) labeling assay

During cell proliferation, the DNA has to be replicated before the cell is divided into two daugh-ter cells (Sawada et al., 1995). Because of the positive relation between fractional rate of DNAsynthesis and proportion of new cells by counting (Macallan et al., 1998), the measurement ofDNA synthesis is very attractive for assessing cell proliferation. Therefore, cell proliferation hasbeen assayed by measuring incorporation of radiolabeled nucleosides (e.g. [3H]thymidine) intoDNA. The amount of [3H]thymidine incorporated into the DNA is quantified by liquid scintil-lation counting. To avoid radioactivity hazards, a method of nonradioactive labeling of the DNAwith 5-bromo-2-deoxyuridine (BrdU, a thymidine analogue) has been developed for measur-ing cell proliferation. It has been shown that the BrdU, like thymidine, is incorporated intocellular DNA. The incorporated BrdU during DNA synthesis could be detected by an enzymeimmunoassay using monoclonal antibodies directed against BrdU, and used to quantify cellproliferation (Maghni et al., 1999).

4.1.2. Stable isotopic tracer incorporation methods

DNA synthesis and breakdown have been measured by labeling DNA with pyrimidinedeoxyribonucleosides (e.g. [3H]thymidine or BrdU); these techniques can be confounded byphysiological factors other than the rates of cell proliferation and death per se (Hellerstein,1999) and cannot be used safely in humans (Neese et al., 2002). Macallan et al. (1998) andMartini et al. (2002) have developed a stable isotopic tracer incorporation method for meas-uring DNA synthesis by labeling the deoxyribose moiety of purine deoxyribonucleotidesthrough the de novo nucleotide synthesis pathway using [2H]glucose or [U-13C6]glucose or2H2O (Macallan et al., 1998; Martini et al., 2002; Neese et al., 2002). It allows measurementof stable isotope incorporation into DNA and calculation of cell proliferation and death ratesin vivo in humans and animals (Hellerstein et al., 1999; Neese et al., 2001, 2002). This methodcounts cell divisions by measuring the proportion of labeled DNA strands present assumingthat each cell division in the presence of label generates two labeled DNA strands (one in eachdaughter cell) (Hellerstein et al., 1999).

Compared to BrdU or [3H]thymidine labeling techniques, there are three differences(Macallan et al., 1998): (1) This method labels deoxyribonucleotides in DNA through the de novo nucleotide synthesis pathway instead of the nucleoside salvage pathway. The pathwaysfor labeling of DNA are illustrated in fig. 2. The efficiency of de novo contribution to purinenucleosides is predictable and high in dividing cells. The activity of the de novo pathway forpurine nucleosides is relatively unaffected by extracellular nucleoside concentrations andderives almost entirely from extracellular glucose, so that the precursor–product relationshipcan be used in a predictable way across cell types (Macallan et al., 1998; Hellerstein, 1999). (2) This method measures labeling in purine deoxyribonucleosides instead of pyrimidines (e.g. from [3H]thymidine or BrdU) (Macallan et al., 1998). (3) BrdU is a pyrimidine nucleosidethat is used by the nucleoside salvage pathway and incorporated into DNA as a thymidine ana-logue. The efficiency of the pyrimidine nucleoside salvage pathway is variable and influencedby availability of extracellular nucleosides (Hellerstein, 1999). Moreover, BrdU does not truly quantify mitotic events, but rather labels descendants of dividing cells (Hellerstein, 1999).


In particular, the incorporation of 2H2O into the deoxyribose moiety in newly synthesized DNAallows safe, convenient, reproducible, and inexpensive measurement of in vivo proliferationrates of slow-turnover cells in humans (Neese et al., 2002).

The fractional production rate of dividing cells (k, per day) can be calculated based on theprecursor–product relationship provided that blood glucose reproducibly provides about 65%of the deoxyribose present in purine nucleotides recovered from DNA of various dividingcells (Macallan et al., 1998; Hellerstein et al., 1999). Therefore,

k = −In (1 − (IEd2-dA/IEd2-glucose × 0.65))/t

where IEd2-dA and IEd2-glucose stand for isotopic enrichment of [2H2]deoxyadenosine and[2H2]glucose, respectively

The absolute production rate of the specific-type cells can be derived by multiplying the kwith their pool size (cells/μl). The half-life (survival time) is indicated by dividing 0.63 by k.

4.2. Cell apoptosis

4.2.1. TUNEL method

The fragmentation of nuclear DNA is one of the endpoints in apoptotic pathways. DNA frag-mentation can be determined by electrophoresis. However, an in situ labeling DNA method hasbeen developed to quantify the DNA fragmentation on the basis of the terminal deoxy-nucleotidyl transferase enzyme reaction after adding deoxynucleotides labeled with biotin ordigoxigenin to free 3′-ends of DNA fragments. Therefore, the formation of a DNA strand breakearly in apoptosis is detected by enzymatic labeling of the 3′-OH termini with modifiednucleotides, which is visualized with streptavidin or anti-digoxigenin antibodies. This methodis called terminal deoxynucleotidyl transferase nucleic acid end labeling (TUNEL). However,it must be noted that the TUNEL will also stain necrotic cells due to extensive DNA degrada-tion (Walker and Quirke, 2001), and thus is a marker of the apoptotic process rather than acritical component of the death process itself. False positive staining in the TUNEL method to


Fig. 2. Labeling pathways for measuring DNA synthesis and thus cell proliferation (adapted from Neese et al., 2002). GNG, gluconeogenesis; G6P, glucose-6-phosphate; R5P, ribose 5-phosphate; PRPP, phosphori-bose pyrophosphate; NDP, nucleoside diphosphate; DNNS, de novo nucleotide synthesis pathway; DNPS, de novo purine/pyrimidine synthesis pathway; RR, ribonucleotide reductase; dNTP, deoxyribonucleosidetriphosphate; dN, deoxyribonucleosides; dT, thymidine deoxyribonucleoside; BrdU, bromodeoxyuridine.

detect apoptosis in the liver and intestine is caused by endogenous endonucleases and is inhib-ited by diethyl pyrocarbonate (Stahelin et al., 1998).

4.2.2. Caspase method

The cysteine–aspartic acid specific proteases (caspases) are activated in response to differentinducers of apoptosis. The process of their activation is considered to be the key event ofapoptosis (Shi, 2002). Caspases recognize a four-amino-acid sequence on their substrate pro-teins and target the carboxyl end of aspartic acid within the sequence. Several methods havebeen developed to detect the activation of caspases. After the pro-caspases are cleaved, theirproducts can be revealed electrophoretically and identified on immunoblots using caspase-specific antibodies. For example, immunostaining of active caspase-3 is a reliable indicatorof apoptotic rate (Marshman et al., 2001). Moreover, activation of caspases in situ can bemeasured by immunocytochemical detection of the epitope that is characteristic of the activeform of caspases or by immunocytochemical identification of the specific cleavage products.In addition, fluorochrome-labeled inhibitors or substrates of caspases have also been used formeasuring the activation of caspases with fluorescence microscopy and flow or laser scanningcytometry (Darzynkiewicz et al., 2002; Smolewski et al., 2002).

5. LASER CAPTURE MICRODISSECTION TO PROCURE PURE CELLS IN VIVO FOR MOLECULAR ANALYSIS

Laser capture microdissection (LCM) is a powerful method to procure pure populations oftargeted cells from specific microscopic regions of heterogeneous tissue sections (Emmert-Buck et al., 1996; Bonner et al., 1997). In this technique, a transparent thermoplastic film(ethylene vinyl acetate polymer) is applied to the surface of a tissue section mounted on aglass slide. While the film is activated through pulsing a laser beam, it becomes focally adhe-sive and fuses to the cells of interest. When the film is removed from the tissue section, theselected cells remain adherent to the film. The film is then placed directly into the isolationbuffer in a microfuge tube for the DNA, RNA, or protein analysis. For the LCM, individualcells can be identified based on histological morphology, immunophenotype, function-relatedantigen expression (Fend et al., 1999), or electronic images from serial sections (Wong et al.,2000). This technique allows in vivo analysis of tissue-, cell-, and function-specific molecu-lar analysis. In combination with high-density oligonucleotide microarray, LCM-procuredcells have been used to obtain gene expression profiles from a discrete cell population (Luzziet al., 2001). This technique can be further coupled with real-time quantitative RT-PCR toquantify mRNA abundance (Betsuyaku et al., 2001), proteomic-based approaches (e.g. 2D-PAGE) to analyze protein expression (Craven and Banks, 2002; Craven et al., 2002), andbiochemical assays to measure cellular metabolite concentrations and enzyme activities(Simone et al., 2000a,b; Stappenbeck et al., 2002).

6. STABLE ISOTOPIC TRACER TECHNIQUES FOR MEASURINGPROTEIN SYNTHESIS AND BREAKDOWN IN VIVO

Stable isotopically labeled tracer techniques have been used in the research of protein (aminoacid, AA), lipid, and carbohydrate metabolism. The principles and practice of stable isotopetracer methodology have been introduced in detail (Wolfe, 1992). New developments andtechniques will be highlighted in this section.


6.1. Whole-body protein kinetics

Assessment of whole-body protein turnover relies on the measurement of the dilution oftracer amino acids in plasma or whole blood, i.e. the rate of appearance (the flux) of the traceeamino acid assuming that the blood pool and tissue free amino acid pools are homogeneouslymixed. Provided that a carbon-labeled indispensable amino acid is infused intravenously at aconstant rate until isotopic equilibrium is attained in the plasma, the amino acid kinetic ratescan be converted to whole-body protein kinetics rates using the average fractional contents ofindividual amino acids in body protein (Waterlow and Stephen, 1967). The equation can beexpressed as follows:

Whole-body flux of AA (Q) = Intake of dietary AA + Release of AA from protein breakdown= Utilization of AA for protein synthesis (nonoxidative dis-

posal) + Oxidation of AA

Whole-body flux of AA (Q) can be estimated by the isotopic enrichment (IE) of plasmaAA at isotopic equilibrium, i.e.

Q = I × [(IEi/IEp) – 1]

where I is the infusion rate of the tracer AA, and IEi and IEp represent the isotopic enrichmentof the labeled AA in infusate and that of plasma AA at plateau, respectively. Both protein syn-thesis and protein degradation can be solved from the equation. Because protein synthesis,breakdown, and amino acid oxidation are intracellular events, it is necessary to measure theisotopic enrichment of the intracellular free amino acid pool rather than its plasma isotopicenrichment to calculate these kinetics.

There are three technique issues. The first one is how to assess the isotopic enrichment ofthe intracellular true precursor (i.e. amino acyl-tRNA) for protein synthesis (see section6.2.1). The second one is how to assess the rate of oxidation. When a primed constant infu-sion of 13C-labeled tracer (e.g. 1-13C-leucine or 1-13C-phenylalanine) is employed to estimaterates of whole-body protein synthesis, one has to determine the rate of oxidation of the traceeby measuring the appearance rate of 13CO2. However, the labeling position of 13C traceraffects recovery of the 13CO2. For example, the recovery of the 2-13C label in breath CO2 is58% relative to the 1-13C label, suggesting that a significant percentage (~42%) is retained inthe body although a majority of the 2-13C label of leucine is recovered in the breath CO2, presumably by transferring to other compounds via the tricarboxylic acid cycle (Toth et al.,2001). Not all of the 1-13C liberated from oxidative disposal appears in the breath CO2. Ring-2H5-phenylalanine (Phe) and 1-13C-tyrosine (Tyr) are infused simultaneously to estimatephenylalanine irreversible hydroxylation (Clarke and Bier, 1982). The hydroxylation rate ofPhe into Tyr can be derived from the equation (Short et al., 1999):

QPhe−Tyr = QTyr × IEd4−Tyr/IEd5− Phe

where QTyr is whole-body flux of plasma Tyr that is estimated from 1-13C-Tyr infusion, IEd4-Tyr

and IEd5−Phe are the isotopic enrichments of plasma L-ring-2H4-Tyr and L-ring-2H5-Phe, respec-tively. This approach has been employed to estimate the in vivo hydroxylation rate of Phe to Tyrin patients with phenlketonuria (van Spronsen et al., 1998). Other combinations (e.g. 15N-Pheand ring-2H4-Tyr or ring-2H5-Phe and ring-2H2-Tyr) have also been used to estimate whole-bodyhydroxylation (Meek et al., 1998; Short et al., 1999). The advantage of the Phe hydroxylationmodel is the rapid assessment of whole-body protein turnover from plasma samples alone with-out measurement of breath 13CO2 production (Clarke and Bier, 1982; Thompson et al., 1989).


The third technique issue is how to model experimental data. Recently, a three-compartment model has been developed to assess whole-body protein synthesis and break-down with a 15N,13C-Leu tracer (fig. 3) (Gowrie et al., 1999). This three-compartment modelthat represents 15N,13C-Leu tracer kinetics can be described by a set of differential equations.Fractional rate constants (fractional rates of protein synthesis and breakdown indicated by k32

and k23, respectively) can be solved using the SAAM II program.

6.2. Precursor method for measuring fractional synthesis rate of tissue protein

6.2.1. Constant infusion method

The fractional synthesis rate (FSR) of protein has been evaluated by a direct precursor–product relationship. The constant infusion method has been used to measure both whole-body protein turnover and tissue protein synthesis. This method involves the infusion of atracer amino acid at a constant rate until steady-state isotopic labeling of the precursor aminoacyl-tRNA pool is reached. Specifically, when a precursor tracer (e.g. a labeled amino acid) is provided as a primed constant infusion into a system, the isotopic enrichment of a homo-geneous product pool will increase as a monoexponential function of time (IEt), i.e. IEt =IEp (1 − e−kt), where IEp is the enrichment of the precursor pool. Therefore,

FSR = k = [(IEt2− IEt1

)/(t2 − t1)]/IEp

The FSR is determined by dividing the initial rate of change in the product isotopic enrich-ment by the precursor isotopic enrichment at the steady state (Patterson, 1997). For example,the FSR of human small intestinal mucosal protein is calculated by a primed constant infu-sion of 1-13C-leucine using this equation, in which IEt2

is the isotopic enrichment of mucosalprotein-bound leucine (from the mucosal biopsy at time 2), IEt1

is the isotopic enrichment ofmucosal protein-bound leucine (from the mucosal biopsy at time 1) or the isotopic enrich-ment of plasma protein-bound leucine at time 1, and IEp is the isotopic enrichment of theprecursor pool (e.g. tissue-free fluid 13C-leucine or plasma 13C-ketoisocaproate (Bouteloup-Demange et al., 1998; Charlton et al., 2000). In order to avoid multiple tissue samples, an


Fig. 3. A three-compartment model for deriving whole-body protein turnover using 15N,13C-Leu tracer(adapted from Gowrie et al., 1999). The model describes the kinetics events during an intravenous infusion of15N,13C-Leu tracer. In brief, infused 15N,13C-Leu tracer enters the plasma free Leu pool (compartment 1) andfurther enters the intracellular AA free pool (compartment 2) where it may be irreversibly deaminated (indi-cated by irreversible loss of 15NH2, k02) or incorporated into the intracellular protein pool (compartment 3,indicated by the fractional transfer rate from compartment 2 to 3, i.e. k32). The isotopic enrichment of theintracellular free Leu pool can be diluted by unlabeled tracee Leu from release of protein breakdown (indi-cated by the fractional transfer rate from compartment 3 to 2, i.e. k23).

overlapping (i.e. staggered) infusion of multiple stable amino acid isotopomers has beendeveloped to measure in vivo FSR when only a single tissue sample can be obtained (Dudleyet al., 1998). Labeled Phe has been also used to assess the FSR of intestinal mucosal andmuscle proteins (Stoll et al., 1997; Biolo et al., 1999). Furthermore, the isotopic enrichmentof plasma VLDL ApoB-100-bound Phe has been used to represent that of the intracellularfree Phe to calculate the FSR of hepatic protein synthesis (Stoll et al., 1997).

The accuracy of this precursor method depends on the measurement of the isotopic enrich-ment of the intracellular true precursor pool. By definition, the tissue tRNA-bound amino acidis the immediate precursor used for protein synthesis (Watt et al., 1991). However, it is diffi-cult to measure the isotopic enrichment of tRNA-bound amino acid, specifically when theprecursor pool is not accessible. There are four alternative solutions to measuring the isotopicenrichment of the intracellular true precursor pool. The first is to measure isotopic enrichmentof tissue free amino acids. During a constant infusion of labeled Leu, there is a quite closeisotopic equilibrium between muscle-free and tRNA-bound leucine pools (Watt et al., 1991;Reeds and Davis, 1999). The isotopic enrichment of tissue fluid Leu in human skeletal musclehas been proved a valid surrogate measurement of the isotopic enrichment for intracellularleucyl-tRNA (Ljungqvist et al., 1997). The isotopic enrichment of tissue free Leu has alsobeen used to estimate the FSR of intestinal mucosal protein (Stoll et al., 2000).

The second approach is to measure the isotopic enrichment of a plasma metabolite that isexclusively derived from the intracellular metabolism of the precursor, e.g. measuring the iso-topic enrichment of plasma α-ketoisocaproate (KIC) as an index of the isotopic enrichmentof the intracellular free leucine to calculate the FSR of muscle and hepatic proteins (Mansooret al., 1997). The KIC is formed intracellularly from leucine and is released, in part, into thesystemic circulation. Thus, the isotopic enrichment of plasma KIC can be used to representthe isotopic enrichment of the intracellular free leucine pool (Matthews et al., 1982).However, isotopic enrichments of plasma KIC and leucine have been shown to be consis-tently higher than those of tissue leucyl-tRNA and tissue fluid leucine (Chinkes et al., 1996a).Therefore, using the isotopic enrichment of plasma KIC as a surrogate measurement of the iso-topic enrichment for leucyl-tRNA will underestimate the FSR of muscle protein, whereas theisotopic enrichment of tissue fluid leucine is a valid surrogate measurement (Watt et al., 1991;Ljungqvist et al., 1997). In a reversal of this approach, constant infusion of α-[1-13C]KIC ismore accurate than labeled leucine to determine the FSR of muscle protein (Chinkes et al.,1996a). When labeled KIC is infused, the isotopic enrichment of intramuscular free Leu is thesame level as that of arterial Leu (Chinkes et al., 1996a).

The third approach is to use the isotopic enrichment of newly synthesized protein-boundamino acid to represent the isotopic enrichment of the true precursor, e.g. using the isotopicenrichment of very-low-density lipoprotein apolioprotein B (VLDL ApoB)-100-boundamino acid as an index of the isotopic enrichment of the hepatic amino acid pool (Reeds et al., 1992). The VLDL ApoB-100 is made in the liver and has a very short half-life in thecirculation. The isotopic enrichment of VLDL ApoB-100-bound amino acid rapidly rises tothe same level as that of the precursor pool in the liver. Because of the heterogeneous com-position of the hepatic intracellular precursor pool, the isotopic enrichment of VLDLApoB-100-bound amino acid may provide a more valid measurement of the isotopic enrich-ment of the hepatic protein synthetic precursor than the hepatic free amino acid pool does(Stoll et al., 1997, 1999b).

However, there are discrepancies in the literature. Isotopic enrichments of different pre-cursors for liver protein synthesis have been compared with that of amino acyl-tRNA using1-13C-Leu and 15N-Phe as tracers in miniature swine (Ahlman et al., 2001). It is shown in fig. 4


that isotopic enrichment ratios of 13C-Leu and 15N-Phe in liver tissue fluid and 13C-KIC inplasma to those of respective amino acyl-tRNA are close to 1.0, indicating that isotopicenrichments of tissue fluid amino acid and plasma 13C-KIC are the best predictors of the iso-topic enrichments of tissue amino acyl-tRNA in the liver and skeletal muscle under differentphysiological conditions (Barazzoni et al., 1999; Ahlman et al., 2001). In contrast, isotopicenrichments of plasma Leu and Phe are substantially higher, whereas that of plasma VLDLApoB-100-bound amino acid is lower than that of the respective amino acyl-tRNA (Ahlmanet al., 2001). Consequently, the FSR of the liver protein derived from isotopic enrichments ofplasma 13C-Leu or plasma VLDL ApoB-100-bound amino acid would be underestimated oroverestimated, respectively (Ahlman et al., 2001). Recently, the FSR of slow-turnover proteinhas been assessed by orally or intravenously administrating 2H2O to label nonessential aminoacids (Hellerstein et al., 2002; Previs, 2002). This method takes advantage of the fact thatthrough transamination reactions the α-hydrogen of nonessential amino acids (e.g. alanineand glutamine) equilibrates rapidly and completely with the 2H of body water. Thus, the FSRof tissue protein can be estimated by measuring the incorporation of 2H-alanine and/or 2H-glutamine into protein. However, at this time, there has been no demonstration of theequivalence of the isotopic enrichment of plasma alanine and glutamine and that of theirtissue free pools.

The fourth approach is to derive the isotopic enrichment of the intracellular true precursorpool from mass isotopomer distribution analysis (see section 9). In conclusion, using the con-stant infusion method for assessing FSR of tissue protein, the best estimate of the isotopicenrichment of intracellular true precursor (amino acyl-tRNA) pool seems to be the tissue freeamino acid in muscle (Davis and Reeds, 2001).

6.2.2. Flooding dose method

To avoid the problem in measuring isotopic enrichment of the intracellular true precursor(amino acyl-tRNA), the flooding dose method has been developed for measuring tissue protein synthesis. This approach involves giving a bolus injection of labeled amino acid with


Fig. 4. Ratios of other precursors to amino acyl-tRNA in porcine liver (data from Ahlman et al., 2001).

a large bolus dose of unlabeled amino acid (e.g. 5 to 10 times the endogenous flux of thetracee) to rapidly create a similar isotopic enrichment in the extra- and intracellular compart-ments. The FSR of liver protein is determined by flooding dose method with 2H5-Phe usingthe equation (Barle et al., 1999):

FSR = IEp × 100/AUC

where IEp is the isotopic enrichment of liver protein-bound Phe at the time of the biopsy andAUC is the area under the curve for the isotopic enrichment of plasma free Phe versus time.The flooding dose method assumes rapid equilibration of the isotopic labeling among theamino acyl-tRNA pool, the tissue free amino acid pool, and the blood free amino acid pool.This assumption has been recently validated by the fact that ratios of specific radioactivity areclose to 1.0 for the tissue free Phe pool versus the phenylalanyl-tRNA pool in either skeletalmuscle or liver (Davis et al., 1999). Under different nutritional and hormonal conditions, theisotopic enrichments of the tissue free Phe pool may be considered satisfactory for assessingthe FSR of skeletal muscle and liver proteins when a flooding dose of Phe is administered(Davis et al., 1999). The short period of measurement with this method is especially valuable,as it allows the determination of acute changes in tissue protein synthesis within 30 min (Garlick et al., 1994). However, it has been observed that large doses of leucine mightstimulate protein synthesis in muscle tissue (Ballmer et al., 1990). The FSR of muscle pro-tein by the flooding dose method is higher than that measured by the constant infusionmethod when 13C-leucine is used (Garlick et al., 1994; Rennie et al., 1994). To study luminalversus basolateral modulation of protein metabolism in small intestinal mucosa, a local(luminal) flooding dose method has been used to determine the fractional rate of protein synthesis in intestinal mucosa (Adegoke et al., 1999a,b).

6.3. Tracee release method for measuring fractional breakdown rate of tissue protein

To measure the fractional breakdown rate (FBR) of muscle protein, the tracee release methodhas been developed on the basis of the precursor–product principle (Zhang et al., 1996). Thismethod involves infusing isotope tracer (e.g. ring-2H5-Phe or ring-13C6-Phe) until isotopicequilibrium is reached. The assessment of the rate of protein breakdown is achieved by meas-uring isotopic enrichment decay curves of the arterial and tissue free amino acid pools afterthe tracer infusion is stopped. Because there is no de novo synthesis of Phe, its appearance inthe tissue free amino acid pool is solely attributed to transport from blood and release by prote-olysis. At isotopic equilibrium, the isotopic enrichment in the tissue free amino acid pool isalways lower than that in the arterial blood because the former is diluted by intracellular unla-beled amino acid released from protein breakdown. Once the isotopic infusion is stopped, theenrichment decay in the tissue free amino acid pool depends on the isotopic enrichment decay inthe arterial blood, which provides tracer and a part of the tracee (i.e. Phe), and on the proteinbreakdown, which provides another part of the tracee. The calculation of FBR is based on therate at which tracee is released from protein breakdown to dilute the isotopic enrichment of thetissue free amino acid pool using a modified precursor–product equation (Zhang et al., 1996), i.e.

FBR = ( ) ( ) ( / )m m m

a m

1

2

1

2

IE t IE t Q T

P IE t dt P IE t dtt

t

t

t

2 1

1

−[ ] ×

− +⎡

⎣⎢⎢

⎤

⎦⎥⎥

∫∫ ( ) ( ) ( )


where P = IEm/(IEa − IEm) represents the ratio of fractional tracee from artery versus the frac-tional tracee from protein breakdown, IEa and IEm being the isotopic enrichments at plateauin the artery pool and muscle intracellular free pool, respectively. The IEm(t2) − IEm(t1) is thechange of the isotopic enrichment in muscle intracellular free pool from time 1 (t1) to time 2(t2) after stopping the isotopic infusion,

are areas under the decay curves of the isotopic enrichments in arterial and muscle intracel-lular free pools, respectively, from t1 to t2. Qm/T is the ratio of the intracellular free traceemass versus protein-bound tracee mass in the muscle. The FBR assessed by the tracee releasemethod is in agreement with that derived from the arterio-venous tracer balance method (seesection 6.4). The tracee release method is the complement of the tracer incorporation method.These two methods can be combined to measure both muscle protein synthesis and break-down in one infusion study and can be applied to other tissues (e.g. skin) if a few biopsies canbe obtained (Zhang et al., 1996; Volpi et al., 2000). The tracee release method has beenrecently improved by using a pulse tracer injection (Zhang et al., 2002). This new approachdoes not require an isotopic steady state, and it can be completed within an hour and usingone or two muscle biopsies.

6.4. Arterio-venous tracer amino acid balance method for measuring tissueamino acid transport, protein synthesis, and breakdown

Since phenylalanine is neither synthesized nor degraded by muscle tissue, the measuredremoval of tracer and dilution of its isotopic enrichment across the hindlimb can be used toestimate rates of phenylalanine incorporation into and release from tissue protein. This meas-urement, coupled with an estimate of tissue blood flow, can provide a readily nondestructivemethod for estimation of protein turnover in specific muscle beds in vivo. Measurements canbe made repeatedly over time in a single experiment, allowing the study of acute regulationof protein turnover (Barrett et al., 1987). This conventional arterio-venous tracer amino acidbalance approach has been improved by measuring the isotopic enrichment of the intracellu-lar free amino acid pool using muscle biopsy to calculate the relative proportions ofintracellular amino acid derived directly from the blood (labeled) or from tissue proteinbreakdown (unlabeled) (Biolo et al., 1995a). Thus, tissue (e.g. muscle) protein synthesis andbreakdown and transmembrane transport of the amino acid can be determined simultane-ously. The arterio-venous tracer amino acid balance method (the A-V method) can bedescribed in a three-compartment model. This model is based on an anatomic compartmen-tation of an indispensable amino acid (e.g. Leu or Phe) into three compartments: the arterial,the intracellular free, and the venous compartments (fig. 5).

In this compartmental model, no interstitial free pool is assumed, i.e. isotopic tracers areassumed to enter their intracellular free compartments at the arterial values and leave fromtheir intracellular free compartments at the venous values. It is also assumed that there is norecycling of isotopic tracers released from protein breakdown into the intracellular free compartment. Isotopic enrichments of the intracellular free compartment (in the tissue fluid)may also be represented by measurement of the isotopic enrichments of other compounds. Forexample, the isotopic enrichment of the local venous plasma 13C-KIC is used to represent thatof the intracellular free 13C-Leu. The isotopic enrichments of liver-synthesized protein VLDL

IE t dt IE t dtt

t

t

t

a mand( ) ( )1

2

1

2

∫∫


ApoB-100-bound or mammary-synthesized casein-bound amino acids are also used to repre-sent those of the intracellular precursor for hepatic or mammary tissue protein synthesis (Reedset al., 1992; Bequette et al., 2000; Guan et al., 2002). Below, we use this three-compartmentmodel to derive the rates of protein synthesis and breakdown and the transmembrane transportof amino acids across an organ of interest.

6.4.1. Protein breakdown

Protein breakdown (PB) can be derived from appearance rate (Ra). Since Ra = PB + arterialinflux, thus PB = Ra − arterial influx, i.e. PB = (Ca × IEa × BF)/IEi – Ca × BF, where IEa and IEi

are isotopic enrichments of Phe in artery and tissue fluid (intracellular free pool), respectively,Ca is the arterial concentration of Phe, and BF is blood flow rate across the organ. Therefore,

Protein breakdown (PB) = Ca × BF × [(IEa/IEi) − 1] (1a)

Assuming that the isotopic enrichment of Phe in vein (IEv) can represent IEi, therefore,

Protein breakdown (PB) = Ca × BF × [(IEa/IEv) − 1] (1b)

In fact, eq. (1) can also be derived from irreversible loss of tracee (IL). Since IL − PB = Netmass balance (NB), thus, PB = IL − NB, i.e. PB = Tracer uptake/IEi − NB, therefore, PB =[(Ca × IEa – Cv × IEv) × BF]/IEi − (Ca − Cv) × BF; when simplified, its equation is identicalto eq. (1a) . It is noted that IEv usually overestimates IEi, thus protein breakdown from eq. (1b)may be underestimated. Alternatively, PB can be derived from unidirectional influx (UI).Since NB = UI − PB, thus PB = UI − NB, where UI = Tracer fractional extraction


Fig. 5. A three-compartment model of amino acid kinetics across the porcine mammary gland during lactation (adapted from Guan et al., 2002). Free amino acid compartments in artery (A), main mammary vein (V),and mammary gland (MG) are connected by arrows indicating unidirectional fluxes of free amino acidsbetween each compartment. Amino acids enter the MG via the mammary artery (Fa,o) and leave the MG viathe main mammary vein (Fo,v). Other fluxes are designated as follows: Fv,a, direct flow of amino acids fromartery to vein without entering the intracellular pool (by the arterial shunt); Fmg,a and Fv,mg, inward and outward transmembrane transport of amino acids from artery to the MG and from the MG to vein, respectively;Fmg,o, the rate of intracellular amino acid appearance from endogenous sources (i.e. release from proteinbreakdown (PB) and de novo synthesis (DS), if any); and Fo,mg, the rate of the intracellular amino’ acids disappearance (i.e. the rate of utilization of intracellular amino acids for protein synthesis (PS), oxidation(OX), and other metabolic fates (OM), if any).

rate × Arterial influx = [(Ca × IEa – Cv × IEv)/(Ca × IEa)] × (Ca × BF), and NB = (Ca − Cv) ×BF, through rearrangement of the equation, therefore,

Protein breakdown (PB) = Cv × BF × [1 − (IEv/IEa)] (2)

6.4.2. Protein synthesis

Protein synthesis (PS) can be derived from irreversible loss of tracee (IL). Since IL = PS +Oxidation (or hydroxylation of Phe to Tyr in liver or kidney), thus

Protein synthesis (PS) = Irreversible loss − Oxidation (or Hydroxylation) (3)

where irreversible loss (IL) = Tracer uptake/Ei, i.e.

Irreversible loss (IL) = (Ca × IEa – Cv × IEv) × BF/IEi (4)

In eq. (4), IEv may be used to replace IEi assuming that the isotopic enrichment of Phe in vein isproximate to that of its intracellular free pool. If 1-13C-Leu is infused, the isotopic enrichment ofvenous plasma 13C-KIC may represent IEi. In fact, IL can be derived from the difference betweennet mass balance and protein breakdown. Since NB = IL − PB, i.e. IL = NB + PB, thus

Irreversible loss (IL) = (Ca − Cv) × BF + PB

Oxidation rate across the organ can be determined by labeled CO2 production. If 1-13C-Leuis infused, oxidation of tracee = 13CO2 production/IEi, i.e.

Oxidation = (CCO2v × IECO2v− CCO2a × IECO2a) × BF/IEi

The isotopic enrichment of the intracellular free Leu may be indicated by the isotopic enrich-ment of venous plasma KIC. If different 13C labeling position or multiple 13C labeling ofleucine is infused, oxidation estimated from eq. (5) should be adjusted by a correction factor.Instead of estimating oxidation, hydroxylation of phenylalanine to tyrosine across the organ


Table 1

Applications of the arterio-venous stable isotopic tracer amino acid balance method

Regional bed Inflow Outflow Blood flow Reference

Kidney Femoral Renal vein The Fick method Moller et al. (2000) artery (paraaminohippurate) Tessari et al. (1996)

Liver Hepatic artery Hepatic vein Doppler flow probe Tessari et al. (1996; Portal vein The Fick method Halseth et al. (1997)

(indocyanine green)Mammary Carotid artery Mammary The Fick method Guan et al. (2002)gland vein (internal Phe + Tyr)Placenta Maternal Umbilical The Fick method Paolini et al. (2001)

femoral artery vein (3H2O dilution)Portal-drained Carotid artery Portal vein Transit-time Guan et al. (2003)viscera ultrasound flow meterSkeletal muscle Femoral Femoral The Fick method Meek et al. (1998)

artery vein (indocyanine green) Tessari et al. (1996)Splanchnic bed Femoral Hepatic vein The Fick method Meek et al. (1998)

artery (indocyanine green) Moller et al. (2000) Tessari et al. (1996)

can be determined (Moller et al., 2000). Note that if this hydroxylation is used to correct irre-versible loss to obtain protein synthesis on the basis of eq. (3), oxidation should not besubtracted any more because oxidation is part of hydroxylation.

This A-V method has been widely used to estimate amino acid kinetics across a particularorgan (table 1). For example, in the fasting state, healthy human skeletal muscle is in a catabolicstate to provide amino acids for protein synthesis required by the splanchnic bed. Net aminoacid balance between splanchnic and skeletal muscle beds is achieved through differential reg-ulations of protein metabolism in these tissues by insulin (Meek et al., 1998). In the fasting stateof the dog, the splanchnic bed contributes about 40% to the whole-body protein breakdown andthe gut and liver each contribute about 50% to the splanchnic bed (Halseth et al., 1997), indicating the equal significance of gut and hepatic proteolysis to whole-body proteolysis.

It is important to accurately measure regional blood flow rate when using this A-V method.Three methods are used for measuring regional blood flow rate: fluorescent microsphere,external unmetabolized marker (e.g. indocyanine green and paraaminohippurate), and ultra-sonic flow probe. If there are multiple entrance or exit vessels, blood flow rate may beappropriately measured by the Fick method based on the conservation of mass. For example,blood flow rates across the splanchnic bed and mammary gland have been estimated by theFick method (Meek et al., 1998; Guan et al., 2002). It has been shown that mammary bloodflow rates estimated by the ultrasonic method are comparable to those estimated by the Fickmethod (Trottier et al., 1997; Renaudeau et al., 2002).

6.4.3. Transmembrane transport of amino acids

The same three-compartment model has been employed to assess amino acid inward and out-ward transport (Biolo et al., 1992, 1995a). Using the porcine mammary gland as an example, thecalculations based on references are (Reeds et al., 1992; Bequette et al., 2000; Guan et al., 2002):

Fa,o = Ca × BF

Fo,v = Cv × BF

Net mass balance = (Ca − Cv) ⋅ BF

Based on the net mass balance of AA across the mammary gland (MG),

Fa,o = Fmg,a + Fv,a

Fo,v = Fv,mg + Fv,a

Thus,

(Ca − Cv) × BF = Fmg,a − Fv,mg

Based on the tracer balance of AA across the MG,

(Ca × IEa − Cv × IEv) × BF = Fmg,a × IEa − Fv,mg × IEi

Therefore,

Fmg,a ={[(IEi − IEv)/(IEa − IEi)] × Cv + Ca} × BF (6)

Fv,mg ={[(IEi − IEv)/(IEa − IEi)] × Cv + Cv} × BF (7)

The only source of tracer appearing in the mammary intracellular free AA compartment istransported inward from plasma. Thus, the isotopic enrichment of tracer AA in the intracellular


free AA is diluted by endogenous sources (Fmg,o) (e.g. protein breakdown and de novo syn-thesis, if any). Therefore,

Ra × IEi = Fmg,a × IEa,

i.e. Ra = (Fmg,a × IEa)/IEi (8)

where Ra is the sum of inward transmembrane transport (Fmg,a) and the appearance rate ofintracellular free AA from the endogenous sources (Fmg,o):

Ra = Fmg,a + Fmg,o

Thus,Fmg,o = Fmg,a × (IEa/IEi − 1) (9)

At steady state, the total fluxes into the mammary intracellular free AA compartment areequal to the total fluxes out of this compartment, i.e.

Fmg,a + Fmg,o = Fv,mg + Fo,mg

Thus,

Fo,mg = Fmg,o + NB (10)

The disappearance rate (Fo,mg, i.e. utilization rate of the intracellular free AA) of intracellu-lar free AA could also be directly calculated as the tracer balance divided by the precursorenrichment (IEi):

Fo,mg = (Ca × IEa − Cv × IEv) × BF/IEi (11)

It has been shown using this three-compartment model that increased net protein synthesisin human muscle (Volpi et al., 1998) and the porcine mammary gland (Guan et al., 2002) byintake of dietary indispensable amino acids is attributed to increased inward transmembranetransport of these amino acids into the respective organs, and that the net flow of amino acidsfrom muscle to the gut in the fasting state is attributed to differences in their transmembranetransport rates (Biolo et al., 1995b).

6.4.4. In vivo nitric oxide synthase activity

This arterio-venous tracer amino acid balance method can be used to assess in vivo nitricoxide synthase (NOS) activity across an organ. Guanidine-15N2-arginine is converted toureido-15N-citrulline and 15NO through NOS reaction (Palmer et al., 1988), and used to quan-tify in vivo NOS activity across organs (Bruins et al., 2002). If guanidine-15N2-arginine andureido-13C-5,5,2H2-citrulline are infused, unidirectional flux of guanidine-15N2-arginine toureido-15N-citrulline (QArg→Cit) can be estimated on the basis of ureido-15N-citrulline tracerbalance (see fig. 6), i.e. based on ureido-15N-citrulline tracer balance:

QIN + QNOS = QOUT + QM

where

QIN = Ca × IE15N-Cit,a × BF

QNOS = QArg→Cit × IE15N-Arg,i

QOUT = Cv × IE15N-Cit,v × BF


QM = Irreversible loss × IE15N-Cit,i

= [(Ca × IEd-Cit,a × BF − Cv × IEd-Cit,v × BF)/IEd-Cit,i] × IE15N-Cit,i

where BF is the blood flow across the organ; Ca and Cv are concentrations of free citrulline inartery and vein, respectively; IE15N-Cit,a (IEd-Cit,a), IE15N-Cit,i (IEd-Cit,i), and IE15N-Cit,v (IEd-Cit,v) areisotopic enrichments of ureido-15N-citrulline (ureido-13C-5,5,2H2-citrulline) in artery, the intra-cellular compartment, and vein, respectively; and IE15N-Arg,i and IE15N-Arg,v are isotopicenrichments of guanidine-15N2-arginine in artery and vein, respectively. Assuming that isotopic enrichment of the tracer in vein is proximate to that in the intracellular compartment, i.e., IE15N-Arg,i = k1 × IE15N-Arg,v; IE 5N-Cit,i = k2 × IE15N-Cit,v; IE d-Cit,i = k3 × IE d-Cit,v; and k1 ≈ k2 ≈k3 ≈ 1, therefore,

QArg→Cit = [(IE d-Cit,a/IEd-Cit,v) × (IE15N-Cit,v/IE15N-Arg,v) − (IE15N-Cit,a /IE15N-Arg,v)] × Ca × BF (2)

This unidirectional flux (QArg→Cit) indicates in vivo NOS activity across the organ (e.g. PDV).The principle of this method has been also applicable to assessment of the local conversion (e.g.hydroxylation of phenylalanine to tyrosine across the liver or kidney) (Moller et al., 2000).

6.4.5. First-pass utilization

In amino acid tracer kinetic studies, ingested amino acid is taken up during its initial transitthrough the splanchnic bed and thus not all absorbed amino acids enter the systemic com-partment. The amount of enterally delivered tracer (or tracee) sequestered by the splanchnicbed can be estimated by simultaneously administrating a labeled tracer intravenously (iv) andintraduodenally (id) (or intragastrically, ig) (Matthews et al., 1993a,b). To assess proteinmetabolism in the splanchnic bed, the infusion of tracer amino acid into the gastrointestinaltract should ideally avoid gastric emptying in the postabsorptive state. An intraduodenaladministration of AA tracers is recommended to obtain plasma isotopic enrichments at steadystate (Crenn et al., 2000). This method allows measurement of splanchnic extraction and


Fig. 6. A compartmental model of arginine and citrulline kinetics across the portal-drained viscera (PDV).QIN, QOUT, QM, and QArg→Cit represent unidirectional fluxes of ureido-15N-citrulline tracer entered from theartery, exited to the vein, metabolized in the organ, and de novo synthesized from arginine through the NOSreaction, respectively. Dotted lines and solid lines indicate fluxes of tracer and tracee, respectively.

indirect assessment of splanchnic protein metabolism under fasting and feeding conditions(Crenn et al., 2000). The splanchnic extraction coefficient of leucine is defined (Stoll et al.,1997, 1999b; Basile-Filho et al., 1998, 1999b) as:

Splanchnic extraction = [Qid − Qiv]/Qid = 1 − Qiv/Qid

in which Qid or Qiv (whole-body flux) = Rate of tracer infusion · [(IEi/IEp) − 1]. If whole-bodyflux is not corrected by the amount of tracer infused, then splanchnic extraction is simplifiedas (Crenn et al., 2000):

Splanchnic extraction = 1 − [(IEp,id/Iid) / (IEp,iv/Iiv)]

where IEp,id and IEp,iv are isotopic enrichments (tracer-to-tracee ratio, TTR) of plasma aminoacids at plateau after a stable isotopic tracer (e.g. 2H3-Leu) is infused intraduodenally andintravenously, respectively. Iid and Iiv are infusion rates of the tracer via intraduodenal andintravenous routes, respectively. Fractional splanchnic oxidation (of whole-body oxidation)and fractional splanchnic hydroxylation of Phe to Tyr (of whole-body hydroxylation) havebeen estimated using this method (Basile-Filho et al., 1998). Because the rate of intragastri-cally infused tracer (13C-Phe) appearing in the nonsplanchnic pool is the product of the tracerinfusion rate (Iig) and nonsplanchnic extraction (Fnsp,extraction = 1 − splanchnic extraction),which is handled in first pass in a manner similar to that for intravenously infused tracer, thenthe fractional nonsplanchnic oxidation (Fnsp,oxidation, of whole-body total oxidation) can becalculated as follows (Basile-Filho et al., 1998):

Fnsp,oxidation = Fnsp,extraction × Riv,oxidation/Rig,oxidation

and Fsp, oxidation = 1 − Fnsp,oxidation

where Fsp,oxidation is the fractional splanchnic oxidation (of whole-body oxidation), and Riv,oxidation

and Rig,oxidation are whole-body oxidation rates calculated by intravenously and intragastricallyinfused tracer, respectively. Similarly, fractional splanchnic hydroxylation (of whole-bodyhydroxylation) can be obtained with an intravenous infusion of ring-2H4-Tyr (see details inBasile-Filho et al., 1998).

We have combined this method and the portal tracer amino acid balance to further assessamino acid metabolism in the gut and liver (Stoll et al., 1997; van Goudoever et al., 2000).Hepatic extraction is defined by the difference between splanchnic extraction (derived fromabove) and portal extraction (derived from the portal tracer amino acid balance). We havefound that splanchnic extraction of Phe is attributed to fractional extraction of the gut and theliver by 75% and 25%, respectively (Stoll et al., 1997), and one-third of the dietary aminoacids is metabolized in the gut (Stoll et al., 1998, 1999a; van Goudoever et al., 2000). It isimportant to assess amino acid metabolism and protein turnover in the splanchnic bed(including the gut and liver) in order to predict the post-splanchnic availability of absorbeddietary amino acids and to understand tissue protein metabolism under different nutritional,physiological, and pathological conditions.

7. STABLE ISOTOPIC TRACER TECHNIQUES FOR STUDYING LIPID METABOLISM

Regulation of lipid metabolism is not only related to growth and fattening of animals, but alsoto the development of cardiovascular disease, insulin resistance, diabetes, and obesity inhumans. The measurement of dynamic fluxes of lipids (biosynthesis, oxidation, and lipoly-sis) poses difficult challenges. Two fundamental advances have recently been made for


measuring lipid biosynthesis, namely deuterium-labeled water incorporation method andmass isotopomer distribution analysis (MIDA). These techniques have resolved the centralmethodological problem in measuring the isotopic enrichment of the intracellular trueprecursor pool. In the 2H2O incorporation method, rates of deuterium incorporation into fattyacids and cholesterol are used to assess de novo lipogenesis and cholesterol synthesis, respec-tively (Diraison et al., 1997; Guo et al., 2000; McDevitt et al., 2001; Bassilian et al., 2002),in which 2H2O (tracer) equilibrates well among the intracellular precursors NADPHand water (Di Buono et al., 2000). Besides, the MIDA is also used to assess the FSR of theVLDL by measuring incorporation of repeating subunits of acetyl-CoA into the newlysynthesized triglyceride (TG) after a constant infusion of [1-13C]acetate (Chinkes et al.,1996b). Labeled glycerol is also used in the study of lipid metabolism (Siler et al., 1998;Lemieux et al., 1999).

7.1. Whole-body lipolysis

Whole-body lipolysis can be determined by the dilution method (e.g. at the infusion of[1,2,3,4-13C4]palmitate and [2H5]glycerol or [2-13C]glycerol) (Horowitz et al., 1999; Siler et al.,1999; Wang et al., 2000; Bergeron et al., 2001). With intravenous infusion of [2H5]glycerol,appearance rate (Ra) of plasma glycerol represents the rate of glycerol released into plasmafrom hormone-sensitive lipase hydrolysis of adipose tissue and intramuscular TG and the rate ofglycerol released into plasma during lipoprotein lipase hydrolysis of VLDL-TG (Mittendorferet al., 2001). However, it does not include the rate of glycerol released during lipolysis ofintra-abdominal adipose tissue TG, which is cleared by the liver (Mittendorfer et al., 2001).Moreover, lipolysis is underestimated by the extent to which glycerol released by lipolysisdoes not enter the systemic circulation, as occurs when lipolysis takes places in the nonhepatictissue of the splanchnic bed (Landau, 1999a). Thus, the glycerol Ra is used to calculate thelower limit for whole-body lipolysis (Aarsland et al., 1996). The rate of appearance of fattyacids (FA) in plasma (Ra) is determined by the equation:

Ra = I × [(IEi/IEp) − 1]

where I is the infusion rate of fatty acid tracer, and IEi and IEp are isotopic enrichments of thefatty acid in the infusate and in the plasma at plateau. However, FA may be re-esterified toTG in most tissues.

7.2. Fatty acid kinetics

The constant infusion of U-13C-labeled fatty acids is used to determine the effects of hyper-glycemia–hyperinsulinemia on whole-body, splanchnic, and leg fatty acid metabolism inhumans (Sidossis et al., 1999). It has been demonstrated that an increase in glucose avail-ability inhibits fatty acid oxidation across the leg and the splanchnic region under the constantavailability of fatty acids (Sidossis et al., 1998, 1999). The fatty acid kinetic parameters forthe leg and the splanchnic region are derived in the same manner as in section 6.4. In brief,

Net rate of NEFA uptake or release = (Ca − Chv) × hepatic (or leg) plasma flow

where Ca, Chv, and Cfv are arterial, hepatic venous, and femoral venous concentrations of non-esterified fatty acids (NEFA), respectively.

Fractional extraction of labeled NEFA = [(IEa × Ca − IEhv × Chv)]/(IEa × Ca);


or = [(IEa × Ca − IEfv × Cfv)]/(IEa × Ca)

where IEa, IEhv, and IEfv are the isotopic enrichments of NEFA in the artery, hepatic vein, andfemoral vein, respectively.

Absolute rate of uptake of NEFA = Fractional extraction × Ca

× Hepatic (or femoral) plasma flow.

Uptake of NEFA that is released as CO2 (%) = Regional 13CO2 production/regional uptake of labeled NEFA

= (IECO2,hv × CCO2,hv − IECO2,a × CCO2,a)/(IEa × Ca − IEhv × Chv)

where IECO2,a and IECO2,hv are the isotopic enrichments of CO2 in the artery and hepatic vein,respectively, and CCO2,a and CCO2,hv are the concentrations of CO2 in the artery and hepaticvein, respectively.

Absolute rate of oxidation of NEFA = Absolute rate of uptake of NEFA× % of NEFA uptake that is released as CO2/

the acetate correction factor.

The acetate correction factor accounts for label fixation that might occur at any step betweenthe entrance of labeled acetyl-CoA into the tricarboxylic acid cycle until the recovery of labelCO2 in breath (Sidossis et al., 1995a). Because label fixation occurs not only via the bicar-bonate pool, but also via isotopic exchange reactions in the tricarboxylic acid cycle (Sidossiset al., 1995b), bicarbonate cannot fully correct the label fixation.

7.3. Muscle triglyceride synthesis

On the basis of the precursor–product relationship and the assumption that the intramuscularNEFA are the synthetic precursors during the infusion of [U-13C]palmitate (Guo and Jensen,1998):

Fractional synthesis rate (FSR) of intramuscular TG = (IEt2,TG-palmitate– IEt1,TG-palmitate)/[Averaged IENEFA-palmitate × Time]

where the numerator is the increment in 13C enrichment of muscle TG palmitate duringa 2–4 h interval, and the denominator is the average 13C enrichment of intramuscular non-esterified palmitate over the same time interval. This measurement is across a particularmuscle bed.

7.4. Hepatic de novo lipogenesis

The rate at which de novo synthesized palmitate is secreted as VLDL-TG is assessed with aconstant infusion of [1,2-13C]acetate using the MIDA. To calculate the fractional synthesisrate of VLDL-bound palmitate (FSR), the following formula is used (Aarsland et al., 1996):

FSR = [(IE(t2) – IE(t1))/(t2 − t1)]/[8p(1−p)7]

where t2 and t1 are the times when samples are taken, IE(t) is the doubly labeled enrichment attime t, and p is the MIDA-derived enrichment of the intrahepatic precursor pool (hepaticacetyl-CoA) for fatty acid synthesis. Here, the factor of 8 accounts for the fact that it requires


eight acetate molecules to form one palmitate molecule (i.e. the principal product ofmammalian de novo fatty acid synthesis). The factor of (1−p)7 accounts for the probabilitythat seven unlabeled acetate molecules will be incorporated into a palmitate molecule(Chinkes et al., 1996b). The FSR is defined as the fraction of plasma VLDL-bound palmitatepool, per unit of time, which is newly synthesized. The absolute synthesis rate is then calcu-lated by multiplying the FSR and the pool size of VLDL-bound palmitate.

8. MASS ISOTOPOMER DISTRIBUTION ANALYSIS

Mass isotopomer distribution analysis (MIDA) is a new technique for quantifying synthesisrates of polymeric biomolecules from 15N-, 13C-, or 2H-labeled monomeric units in the pres-ence of unlabeled polymer. Mass isotopomer distribution is analyzed according to acombinatorial probability model. The isotopomers of a given type of molecule are the vari-ous combinations of positions of labeled atoms. For example, when the 12C isotope can bereplaced by 13C independently at each position in glucose, there are 64 (26 = 64) different iso-topomers. The MIDA allows the isotopic enrichment of the monomeric precursor to bederived indirectly from the isotopic enrichment of the polymer (product). This derived pre-cursor enrichment presumably represents the steady-state enrichment of the precursor. TheMIDA has been used to measure fractional rates of cholesterol biosynthesis (Lindenthal et al.,2002), gluconeogenesis (Trimmer et al., 2002), lipogenesis, and protein synthesis.

Monomer subunits are randomly selected from the precursor pool and incorporated into apolymer (the product). Theoretical distribution of newly formed product molecules can be pre-dicted by binomial or multinomial expansion. The probabilities of incorporating a given numberof labeled precursors into the product are determined by the isotopic enrichment of the precursorpool on the basis of a multinomial distribution (Hellerstein and Neese, 1999):

where σ is the number of labeled subunits present in the variable moiety of the polymer, z isthe maximum number of monomer subunits that can be labeled in the variable moiety of thepolymer, and p is the fraction of ΔAx

∞/ΔAy∞ isotopically labeled subunits in the subunit precursor

pool. The value of p is calculated from the best-fit polynomial regression equation of p againstthe ratio of in an appropriate reference table (Hellerstein and Neese, 1999). Here, ΔAx

∞ and ΔAy∞

are defined as the change in fractional abundance (i.e. excess mass isotopomer abundance) inthe newly synthesized or isotopically perturbed polymers only (e.g. ratio of doubly to singlylabeled product, or triply to doubly labeled product). The precursor enrichment (p) is deter-mined from the measured ratio of ΔAx

∞/ΔAy∞ using this equation. Based on the

precursor–product relationship, the fractional synthesis (f, the proportion of newly synthesizedmolecules present in the mixture) can be calculated using ΔAx

∞ at the value of p (derived fromthe ratio of ΔAx

∞/ΔAy∞), i.e.

f = ΔAx (mixture)/ΔAx∞

where ΔAx (mixture) is the change of the fractional abundance of a mass isotopomer Mx in themixture (measured), and ΔAx

∞ is the enrichment at plateau, i.e. the precursor enrichment in aone-source biosynthetic system (calculated from the regression equation of ΔAx

∞ against p, rep-resenting the asymptotic value of ΔAx

∞). When all the ions in the mass isotopomer spectrum arenot monitored, a correction equation is used for calculating f (Papageorgopoulos et al., 1999).

d z pz

zp Pz

z( , , )

!

( )! !( )( )σ

σ σσ σ

σ=

−− −

=1

0U


Therefore, the fractional synthetic rate constant (ks) is calculated as follows (Papageorgopouloset al., 1999):

ks = −In(1 − f)/t

8.1. Protein synthesis

It is difficult to assess protein synthesis using the conventional precursor–product methodbecause it not easy to accurately measure the isotopic enrichment of the precursor pool (intra-cellular amino acyl-tRNA). This difficulty has been conquered with the MIDA. Using theMIDA to assess protein synthesis has become technically feasible and practical in vivo usingproteolytically derived peptides (Papageorgopoulos et al., 1999). To obtain mass isotopomerdistribution, a small peptide that contains repeats of a selected amino acid is generated from awhole molecule of protein. The kinetics of the peptide component presumably represents thekinetics of the intact protein. For example, [5,5,5-2H3]leucine is intravenously infused into rats,and then a leucine-rich peptide is isolated and purified from trypsin-digested rat serum albumin.Theoretic abundances and excess abundances of mass isotopomers are calculated and measured.Biosynthetic rates of rat serum albumin are estimated by the MIDA, which are similar to previ-ously published values (Papageorgopoulos et al., 1999). Based on the exchange of 2H2O withα-hydrogen of non-essential amino acids (e.g. alanine and glutamine), the MIDA can be usedfor measurement of synthesis rates of slow-turnover proteins (Hellerstein et al., 2002).

8.2. Lipogenesis

If [1-13C]acetate is infused in vivo, VLDL-bound palmitate enrichment can be measured bythe tracer-to-tracee ratio (TTR). The precursor (acetyl-CoA) enrichment p is derived from theMIDA (Chinkes et al., 1996b):

p = [2 × TTR(M + 2)/TTR (M + 1)]/[(n − 1) + 2 × TTR(M + 2)/TTR(M + 1)]

This precursor enrichment is expressed in terms of acetate units, and is converted to units ofsingle labeled palmitate (IEp) using the binomial equation:

IEp = np(1−p)n−1

The FSR of VLDL-palmitate is calculated on the basis of the precursor-product relationshipas follows:

FSR = [(IE(t2) − IE(t1))/(t2 − t1)]/[np (1 − p)n−1]

where IE(t) is the singly labeled product enrichment at time t, i.e. singly labeled VLDL-palmitate(M + 2) enrichment. If doubly labeled acetate ([1,2-13C]acetate) is infused rather than singlylabeled acetate, palmitate will appear at the peaks M + 2 and M + 4 rather than M + 1 and M + 2.In the calculation of the precursor enrichment, TTR(M + 4)/TTR(M + 2) is used in place ofTTR(M + 2)/TTR(M + 1). Recently, 2H2O has been used to label the glycerol moiety of triglyc-eride to simultaneously measure in vivo TG synthesis and de novo lipogenesis in adipose tissue(Antelo et al., 2002; Turner et al., 2002).

8.3. Gluconeogenesis (GNG)

The rate of glucose production is the sum of rates of glycogenolysis and gluconeogenesis.The rate of glycogenolysis is the rate at which glucose is formed from glycogen, which can


be determined by the decline in liver glycogen content measured by 13C nuclear magnetic res-onance spectroscopy (Rothman et al., 1991). The rate of gluconeogenesis is the rate ofglucose synthesis via glucose-6-phosphate from gluconeogenic precursors (e.g. lactate, ala-nine, pyruvate, and glycerol). Gluconeogenesis can be determined directly by twoapproaches. The first one is to assess the fractional contribution of gluconeogenesis by meas-uring the ratio of the 2H enrichment of the hydrogen bound to C-5 and to that of C-2 of bloodglucose at steady state after oral intake of 2H2O (Landau et al., 1996; Chandramouli et al., 1997;Petersen et al., 1999). That is because a hydrogen atom from body water is bound to C-5 ofevery molecule of glucose formed via gluconeogenesis and none via glycogenolysis, while ahydrogen atom from body water is added at C-2 of glucose formed via both gluconeogenesisand glycogenolysis; the ratio of enrichment at C-5 to that at C-2 also provides a measure of that fraction (Chandramouli et al., 1997). The rate of gluconeogenesis is calculated by multiplying that ratio by the rate of glucose production, i.e. the rate of appearance of glucose.Gluconeogenesis determined by the ratio of the 2H enrichments will be overestimated by thedegree of cycling between glucose-6-phosphate and triose phosphate, and/or loss of label viatransaldolase exchange reactions that are part of the pentose cycle (Ackermans et al., 2001),for the contribution of the cycling between glucose-6-phosphate and triose phosphate resultsin an increase in the labeling of C-5 and, thus, in an overestimation of gluconeogenesis, i.e.the conversion of glycogen to triose phosphates (then used for glucose synthesis) is includedin the estimate of the contribution of gluconeogenesis rather than glycogenolysis(Chandramouli et al., 1997).

The second approach is to assess the fractional contribution of gluconeogenesis using theMIDA. Glucose can be considered as a dimer made of two triose subunits. The MIDA of glu-cose labeled from [2-13C]glycerol, [U-13C3]glycerol, [3-13C]lactate, or [U-13C3]lactate can beused for estimating the contribution of gluconeogenesis to glucose production (Neese et al.,1995). The MIDA of glucose is more precise with uniformly labeled than singly labeled 13Csubstrates (Previs et al., 1995). In the latter case, ratios of glucose molecules labeled with two13C atoms (M2) versus with one 13C atom (M1) are very sensitive to a small error in the fairlyhigh background correction at M2. Moreover, the contribution of gluconeogenesis to glucoseproduction is artifactually underestimated by loss of [2-13C]glycerol carbon via the pentosecycle when [2-13C]glycerol is infused (Previs et al., 1995; Kurland et al., 2000). It is also pos-sible that a proportion of glucose is formed from glycerol and from amino acids not convertedto glucose via pyruvate (Landau, 1999c). Thus, [U-13C3]lactate appears to be a suitable tracerfor the MIDA of gluconeogenesis in vivo (Previs et al., 1995), especially for tracing low ormoderate rates of gluconeogenesis (Previs et al., 1998).

The MIDA of plasma glucose and lactate can be carried out during an infusion of [U-13C6]glu-cose. During an infusion of [U-13C6]glucose (M6 glucose), glycolysis leads to the productionof labeled lactate (m3 lactate). When 13C carbon atoms are recycled in gluconeogenesis, glucosemolecules with one, two, or three 13C substitutions (M1, M2, and M3 glucose) are produced. Theappearance of mass isotopomers M1, M2, and M3 of glucose provides a measurement of therate of gluconeogenesis. Because the chance of two labeled triose phosphates combining toform glucose is negligible, M6 glucose behaves as a nonrecyclable tracer, and the steady-stateenrichment of M6 glucose in plasma allows the determination of the hepatic glucose produc-tion rate. Thus the infusion of [U-13C6]glucose has the advantage of being able to estimatesimultaneously hepatic glucose output and fractional gluconeogenesis from the MIDA ofplasma glucose and lactate and has been used to estimate gluconeogenesis by Tayek and Katz(Tayek and Katz, 1996, 1997; Katz and Tayek, 1999). However, different equations have beenused to calculate the contribution of gluconeogenesis to glucose production.


The dilution of the labeled lactate molecules by endogenous unlabeled lactate molecules(D) is calculated by the equation (Landau et al., 1998; Landau, 1999b; Radziuk and Lee,1999):

D = [0.5(M1 + M2 + M3) + M6]/(m1 + m2 + m3)

where M1, M2, M3, and M6 are, respectively, the percentages of blood glucose molecules withone, two, three, and six 13C atoms, i.e. isotopomers M1, M2, M3, and M6. Correspondingly, m1,m2, and m3 are the percentages for blood lactate of isotopomers m1, m2, and m3, respectively.

The fraction of glucose molecules in the blood that recycled (F), i.e. via the Cori cycle, iscalculated by the equation (Landau et al., 1998; Landau, 1999b; Radziuk and Lee, 1999):

F = 0.5(M1 + M2 + M3)/[0.5(M1 + M2 + M3) + M6]

The product of the Cori cycle and the dilution of glycolysis by endogenous lactate representsthe contribution of gluconeogenesis to the Ra glucose (Katz and Tayek, 1999). Thus, the frac-tional gluconeogenesis (% of glucose production) can be calculated by the following equation(Landau, 1999b; Radziuk and Lee, 1999; Mao et al., 2002), assuming that there is no loss oflabeled molecules via the tricarboxylic acid cycle because when mi → mj, i ≥ j, labeled mol-ecule is still counted (Kelleher, 1999; Radziuk and Lee, 1999):

Fractional gluconeogenesis (% of glucose production) = (M1 + M2 + M3)/[2(m1 + m2 + m3)]

Fractional gluconeogenesis can be derived directly by a binominal expansion approach(Kelleher, 1999; Radziuk and Lee, 1999):

Gluconeogenesis (% of glucose production) = (M1 + M2 + M3)/[2 ⋅ m0 ⋅ (m1 + m2 + m3)]

where m0 is approximate to 1. Fractional gluconeogenesis calculated from these equations isunderestimated (Landau et al., 1998; Kelleher, 1999; Landau, 1999b; Radziuk and Lee, 1999;Mao et al., 2002), which results from the lack of isotope equilibrium in both the lactate (m3)and glucose (M3) compartments and the tracer dilution by other unlabeled gluconeogenic substrates (Mao et al., 2002).

Finally,

Rate of gluconeogenesis = Ra glucose × D × F

Equations for D and F are applicable only when the rate of glucose infused is small relativeto glucose production, which will result in relatively low enrichments and with negligible formation of M4 and M5 as well as M6 isotopomers.

9. NUCLEAR MAGNETIC RESONANCE SPECTROSCOPY

Nuclear magnetic resonance (NMR) spectroscopy now provides a noninvasive means tomonitor metabolic flux and intracellular metabolite concentrations continuously. The basicprinciples of in vivo NMR spectroscopy have been described in detail (Roden and Shulman,1999). In brief, some atomic nuclei (e.g. 1H, 13C, and 31P) possess magnetic properties, i.e. themagnetic moment or “spin”. Under experimental conditions, resonant waves (resonance)from various nuclei/compounds can be translated into a display of peak intensities vs. fre-quencies. The frequency of a peak is the characteristic of a certain nucleus/compound and thearea under that peak corresponds to the concentration of that nucleus/compound. The abilityto distinguish between different molecules containing the same nucleus relies on the “chemicalshift”, given in parts per million (ppm). The nuclei of different molecules thereby experience


an altered static magnetic field and in turn resonate at an altered frequency, i.e. chemical shift,which is typical for the respective molecule.

Several measures are used to improve the signal-to-noise ratio of NMR spectroscopy.Increasing the field strength of the static magnetic field improves the signal-to-noise ratio andthereby the sensitivity of the technique. Studies in humans are routinely performed at 1.5–4.7Tesla. To examine a defined small volume of tissue, surface coils are placed tightly over theregion of interest to ensure homogeneous tissue filling in that region. The pulse angle and shapecan be selected to suppress signals from other tissues such as the subcutaneous fat layer. In vivoNMR spectroscopy can measure the concentrations and synthesis rates of individual biologi-cal molecules such as glycogen and neurotransmitters within precisely defined areas of specificorgans such as brain, liver, and muscle (Shulman and Rothman, 2001). Most studies to date have used 1H, 31P, and 13C to determine skeletal muscle glucose and glycogen metabolism.The suitability of a nucleus for NMR spectroscopy depends on its relative magnetic sensitiv-ity, the tissue concentration range of the metabolite, and the chemical shift range.

9.1. 1H NMR spectroscopy

Protons (1H) have a natural abundance close to 100% and overall offer the highest sensitivityfor NMR spectroscopy. However, the relatively low concentration of metabolites (comparedto the proton concentration in water) and the low chemical shift range (10 ppm) have limitedthe use of 1H for NMR spectroscopy. Measurement of intracellular triglyceride (TG) contentin vivo at 1.5 Tesla by 1H NMR spectroscopy has been validated biochemically by liverbiopsy (Szczepaniak et al., 1999). Furthermore, utilization of intramyocellular lipid in humanmuscle is measured by 1H NMR spectroscopy (Szczepaniak et al., 1999; Krssak et al., 2000).

9.2. 31P NMR spectroscopy

Phosphors (31P) occur 100% in nature and allow quantification of intramuscular concentra-tions of adenosine triphosphate (ATP), adenosine diphosphate, inorganic phosphate,phosphocreatine, and glucose-6-phosphate (G6P) (Krebs et al., 2001). The concentrations ofmetabolites are determined by comparing the spectral areas to the area of the β-ATP resonance,which is used as an internal concentration standard (Bloch et al., 1993). Measurement of mus-cular G6P concentrations by 31P NMR spectroscopy has been validated by a chemical assayof its concentration in rat muscle frozen in situ (Bloch et al., 1993). Glucose-6-phosphate isan intermediate in the muscle glycogen synthesis pathway, and its concentration depends onthe relative activities of muscle glycogen synthase enzyme and glucose transport into muscle.In addition, 31P NMR spectroscopy has been used to measure mitochondrial unidirectionalATP synthesis flux in vivo in rat skeletal muscle (Jucker et al., 2000a,b) and to measure G6Pconcentration in human muscle (Rothman et al., 1995).

9.3. 13C NMR spectroscopy

In contrast to 1H and 31P, 13C has a natural abundance of 1.1% and therefore a relatively lowsensitivity. Nevertheless, 13C NMR spectroscopy has been used to measure hepatic glycogenconcentrations and thus estimate rates of net hepatic glycogen synthesis and glycogenolysis in vivo. Since the resonance of 13C in the C-1 position of glycogen is clearly resolved at 100.5 ppmand all 13C signals from glycogen are detected by 13C NMR spectroscopy, it can be used to meas-ure 13C incorporation into glycogen during an infusion of [1-13C]glucose, which increases the


sensitivity of the method by up to 100-fold. Measurement of tissue glycogen content by 13CNMR spectroscopy has been validated for skeletal muscle and liver by comparison with muscle(Gruetter et al., 1991) and liver (Gruetter et al., 1994) biopsies (Taylor et al., 1992; Krssak et al.,2000). Furthermore, using a 13C-glucose pulse–12C-glucose chase experiment, rates of hepaticglycogen synthesis and glycogenolysis can be assessed (Magnusson et al., 1994; Roden et al.,1996; Petersen et al., 1998). The peak intensity of the C-1 resonance of the glycosyl units ofglycogen is monitored with 13C NMR spectroscopy during [1-13C]glucose infusion followed byunlabeled glucose infusion. Increment in the C-1 peak intensity during the [1-13C]glucose infu-sion represents glycogen synthesis, while decline in the C-1 peak intensity during unlabeledglucose infusion reflects glycogenolysis (Magnusson et al., 1994). Increments in muscle glyco-gen concentration can be calculated from the change in [1-13C]glycogen concentration and theisotopic enrichment of plasma [1-13C]glucose (Shulman et al., 1990).

In human and rat brains 13C NMR measurements of the in vivo flux of 13C label from [1-13C]glucose into glutamate and glutamine simultaneously determine the rate of glucose oxidation(tricarboxylic acid cycle rate) and glutamate/glutamine neurotransmitter cycling betweenastroglia and neurons (Sibson et al., 1998, 2001; Shen et al., 1999; Shulman et al., 2001). Theglutamate/glutamine neurotransmitter cycling, measured by 13C NMR spectroscopy, is themajor pathway for neuronal glutamate repletion (Lebon et al., 2002), which accounts for 80%of glucose oxidation in the resting state (Shen et al., 1999). 1H-decoupled 13C NMR spectrayields sufficient signal-to-noise resonance at C-4 glutamate and C-4 glutamine in the rat brainin vivo at 7.0 Tesla (Sibson et al., 1998). It is possible to detect 13C labeling of glutamate andglutamine in liver by 13C NMR spectroscopy. Additionally, the in vivo 13C labeling kinetics ofglutamate and glutamine in liver and glutamine in blood can be used to calculate the liver tri-carboxylic acid cycle flux (Jucker et al., 1998). 13C NMR and 31P NMR can be combined toquantify glycogen synthesis rate and glucose-6-phosphate concentration in rat gastrocnemiusmuscle (Chase et al., 2001). The concentration of glycogen is calculated from the increment inthe 13C spectra and the isotopic enrichment of [1-13C]glucose (Bloch et al., 1994).

10. FUTURE PERSPECTVES

In this chapter, we have discussed some new approaches aimed at understanding the biolog-ical basis of metabolomics from systemic physiology, to intermediary metabolism, and tomolecular regulation of critical gene and protein expression. Metabolomics has recently beendeveloped as a platform for the quantitative measurement of the dynamic multiparametricmetabolic response of living systems to genetic modification, developmental state, patho-physiological process, or environmental stimulus, which promises to identify gene function,evaluate drug efficacy and toxicity, and define in vivo metabolic profiling (of all the metabo-lites in an intact tissue, organ, or biofluid) (Raamsdonk et al., 2001; Brindle et al., 2002;Nicholson et al., 2002; Watkins et al., 2002). Metabolomics is becoming feasible directly incrude biological extracts with advances in nuclear magnetic resonance spectroscopy, massspectrometry coupled with bioinformatics techniques, and multivariate statistical analyses. Infact, the metabolic status of an integrated biological system can be defined by its spectralmetabolic profile. Because of metabolic dynamics caused by coordinated biochemical andmolecular events, metabolic profiles are spatial-specific and temporal-dependent in responseto developmental state and environmental stimuli, which may mirror tissue-specific and time-related changes in transcriptomic and proteomic patterns, thus limiting any physiologicalrelevance of single-time-point measurements of gene expression and protein abundance.Moreover, metabolomics may provide the most direct linkage between genetic function,


metabolic pathway, and physiological process to decipher metabolic networks (e.g. control ofglycolysis).

In molecular regulation, identifying changes in gene expression using cDNA microarraysis just the start of a long journey from tissue to cell. At this step, the principal aim is to assem-ble microarray hits into groups for particular metabolic pathways and/or functional processesthat provide an intelligible story of a cell’s state, or its metabolic responses to stimuli. Then,it is usual to select a subset of these genes to independently validate changes in their expres-sion. Combination of laser capture microdissection with real-time quantitative RT-PCR is ahelpful follow-up step that allows expression of selected genes to be quantified in a pure pop-ulation of defined individual cells. The voyage from chip to single cell can be completedusing sensitive new in situ hybridization and immunohistochemical methods based on tyramide signal amplification to identify cells that express mRNAs and proteins of interest(Mills et al., 2001). Finally, RNA interference can be used as a specific and efficient methodto silence gene expression in mammalian cells and to confirm gene function on a whole-genome scale (McManus and Sharp, 2002).

In intermediary metabolism, stable isotopic tracer methodology has become the most pow-erful tool to quantify metabolic fluxes both in the whole body and across an organ. Forexample, the arterio-venous tracer balance approach and mass isotopomer distribution analy-sis have been widely used to estimate in vivo enzyme activity (e.g. NOS activity) and nutrientmetabolism (e.g. protein synthesis and breakdown, lipogenesis and lipolysis, and gluconeoge-nesis). In the future, it will be possible to integrate data from transcriptomics, proteomics, andmetabolomics to provide an in vivo holistic picture of gene function and metabolic control(Nicholson et al., 2002; Fiehn and Weckwerth, 2003).

REFERENCES

Aarsland, A., Chinkes, D., Wolfe, R.R., 1996. Contributions of de novo synthesis of fatty acids to totalVLDL-triglyceride secretion during prolonged hyperglycemia/hyperinsulinemia in normal man. J. Clin. Invest. 98, 2008–2017.

Ackermans, M.T., Pereira Arias, A.M., Bisschop, P.H., Endert, E., Sauerwein, H.P., Romijn, J.A., 2001.The quantification of gluconeogenesis in healthy men by 2H2O and [2-13C]glycerol yields differentresults: rates of gluconeogenesis in healthy men measured with 2H2O are higher than those measuredwith [2-13C]glycerol. J. Clin. Endocrinol. Metab. 86, 2220–2226.

Adegoke, O.A., McBurney, M.I., Baracos, V.E., 1999a. Jejunal mucosal protein synthesis: validation ofluminal flooding dose method and effect of luminal osmolarity. Amer. J. Physiol. 276, G14–G20.

Adegoke, O.A., McBurney, M.I., Samuels, S.E., Baracos, V.E., 1999b. Luminal amino acids acutelydecrease intestinal mucosal protein synthesis and protease mRNA in piglets. J. Nutr. 129, 1871–1878.

Ahlman, B., Charlton, M., Fu, A., Berg, C., O’Brien, P., Nair, K.S., 2001. Insulin’s effect on synthesis ratesof liver proteins: a swine model comparing various precursors of protein synthesis. Diabetes 50, 947–954.

Antelo, F., Strawford, A., Neese, R.A., Christiansen, M., Hellerstein, M., 2002. Adipose triglyceride (TG)turnover and de novo lipogenesis (DNL) in humans: measurement by long-term 2H2O labeling andmass isotopomer distribution analysis (MIDA). FASEB J. 16, A400.

Arcellana-Panlilio, M., Robbins, S.M., 2002. Cutting-edge technology. I. Global gene expression profilingusing DNA microarrays. Amer. J. Physiol. Gastrointest. Liver Physiol. 282, G397–G402.

Ballmer, P.E., McNurlan, M.A., Milne, E., Heys, S.D., Buchan, V., Calder, A.G., Garlick, P.J., 1990.Measurement of albumin synthesis in humans: a new approach employing stable isotopes. Amer. J. Physiol. 259, E797–E803.

Barazzoni, R., Meek, S.E., Ekberg, K., Wahren, J., Nair, K.S., 1999. Arterial KIC as marker of liver and muscleintracellular leucine pools in healthy and type 1 diabetic humans. Amer. J. Physiol. 277, E238–E244.

Barle, H., Essen, P., Nyberg, B., Olivecrona, H., Tally, M., McNurlan, M.A., Wernerman, J., Garlick, P.J.,1999. Depression of liver protein synthesis during surgery is prevented by growth hormone. Amer.J. Physiol. 276, E620–E627.


Barrett, E.J., Revkin, J.H., Young, L.H., Zaret, B.L., Jacob, R., Gelfand, R.A., 1987. An isotopic methodfor measurement of muscle protein synthesis and degradation in vivo. Biochem. J. 245, 223–228.

Basile-Filho, A., Beaumier, L., El-Khoury, A.E., Yu, Y.M., Kenneway, M., Gleason, R.E., Young, V.R.,1998. Twenty-four-hour L-[1-13C]tyrosine and L-[3,3-2H2]phenylalanine oral tracer studies at generous,intermediate, and low phenylalanine intakes to estimate aromatic amino acid requirements in adults.Amer. J. Clin. Nutr. 67, 640–659.

Bassilian, S., Ahmed, S., Lim, S.K., Boros, L.G., Mao, C.S., Lee, W.N., 2002. Loss of regulation of lipoge-nesis in the Zucker diabetic rat. II. Changes in stearate and oleate synthesis. Amer. J. Physiol. Endocrinol.Metab. 282, E507–E513.

Bednar, M., 2000. DNA microarray technology and application. Med. Sci. Monit. 6, 796–800.Bequette, B.J., Hanigan, M.D., Calder, A.G., Reynolds, C.K., Lobley, G.E., MacRae, J.C., 2000. Amino

acid exchange by the mammary gland of lactating goats when histidine limits milk production. J. Dairy Sci. 83, 765–775.

Bergeron, R., Previs, S.F., Cline, G.W., Perret, P., Russell, R.R. 3rd, Young, L.H., Shulman, G.I., 2001.Effect of 5-aminoimidazole-4-carboxamide-1-beta-D-ribofuranoside infusion on in vivo glucose andlipid metabolism in lean and obese Zucker rats. Diabetes 50, 1076–1082.

Betsuyaku, T., Griffin, G.L., Watson, M.A., Senior, R.M., 2001. Laser capture microdissection and real-time reverse transcriptase/polymerase chain reaction of bronchiolar epithelium after bleomycin.Amer. J. Respir. Cell Mol. Biol. 25, 278–284.

Biolo, G., Chinkes, D., Zhang, X.J., Wolfe, R.R., 1992. Harry M. Vars Research Award. A new model todetermine in vivo the relationship between amino acid transmembrane transport and protein kineticsin muscle. J. Parenter. Enteral Nutr. 16, 305–315.

Biolo, G., Fleming, R.Y., Maggi, S.P., Wolfe, R.R., 1995a. Transmembrane transport and intracellularkinetics of amino acids in human skeletal muscle. Amer. J. Physiol. 268, E75–E84.

Biolo, G., Williams, B.D., Fleming, R.Y., Wolfe, R.R., 1999. Insulin action on muscle protein kineticsand amino acid transport during recovery after resistance exercise. Diabetes 48, 949–957.

Biolo, G., Zhang, X.J., Wolfe, R.R., 1995b. Role of membrane transport in interorgan amino acid flowbetween muscle and small intestine. Metabolism 44, 719–724.

Bloch, G., Chase, J.R., Avison, M.J., Shulman, R.G., 1993. In vivo 31P NMR measurement of glucose-6-phosphate in the rat muscle after exercise. Magnet. Reson. Med. 30, 347–350.

Bloch, G., Chase, J.R., Meyer, D.B., Avison, M.J., Shulman, G.I., Shulman, R.G., 1994. In vivo regula-tion of rat muscle glycogen resynthesis after intense exercise. Amer. J. Physiol. 266, E85–E91.

Bonner, R.F., Emmert-Buck, M., Cole, K., Pohida, T., Chuaqui, R., Goldstein, S., Liotta, L.A., 1997.Laser capture microdissection: molecular analysis of tissue. Science 278, 1481–1483.

Bouteloup-Demange, C., Boirie, Y., Dechelotte, P., Gachon, P., Beaufrere, B., 1998. Gut mucosal proteinsynthesis in fed and fasted humans. Amer. J. Physiol. 274, E541–E546.

Brindle, J.T., Antti, H., Holmes, E., Tranter, G., Nicholson, J.K., Bethell, H.W., Clarke, S., Schofield, P.M.,McKilligin, E., Mosedale, D.E., Grainger, D.J., 2002. Rapid and noninvasive diagnosis of thepresence and severity of coronary heart disease using 1H-NMR-based metabonomics. Nat. Med. 8,1439–1444.

Bronson, S.K., Smithies, O., 1994. Altering mice by homologous recombination using embryonic stemcells. J. Biol. Chem. 269, 27155–27158.

Bruhat, A., Averous, J., Carraro, V., Zhong, C., Reimold, A.M., Kilberg, M.S., Fafournoux, P., 2002.Differences in the molecular mechanisms involved in the transcriptional activation of the CHOP andasparagine synthetase genes in response to amino acid deprivation or activation of the unfolded protein response. J. Biol. Chem. 277, 48107–48114.

Bruhat, A., Jousse, C., Carraro, V., Reimold, A.M., Ferrara, M., Fafournoux, P., 2000. Amino acids con-trol mammalian gene transcription: activating transcription factor 2 is essential for the amino acidresponsiveness of the CHOP promoter. Mol. Cell Biol. 20, 7192–7204.

Bruhat, A., Jousse, C., Fafournoux, P., 1999. Amino acid limitation regulates gene expression. Proc. Nutr.Soc. 58, 625–632.

Bruhat, A., Jousse, C., Wang, X.Z., Ron, D., Ferrara, M., Fafournoux, P., 1997. Amino acid limitationinduces expression of CHOP, a CCAAT/enhancer binding protein-related gene, at both transcriptionaland post-transcriptional levels. J. Biol. Chem. 272, 17588–17593.

Bruins, M.J., Lamers, W.H., Meijer, A.J., Soeters, P.B., Deutz, N.E., 2002. In vivo measurement of nitricoxide production in porcine gut, liver and muscle during hyperdynamic endotoxaemia. Brit. J. Pharmacol.137, 1225–1236.


Bustin, S.A., 2000. Absolute quantification of mRNA using real-time reverse transcription polymerasechain reaction assays. J. Mol. Endocrinol. 25, 169–193.

Bustin, S.A., 2002. Quantification of mRNA using real-time reverse transcription PCR (RT-PCR): trendsand problems. J. Mol. Endocrinol. 29, 23–39.

Chandramouli, V., Ekberg, K., Schumann, W.C., Kalhan, S.C., Wahren, J., Landau, B.R., 1997.Quantifying gluconeogenesis during fasting. Amer. J. Physiol. 273, E1209–E1215.

Charlton, M., Ahlman, B., Nair, K.S., 2000. The effect of insulin on human small intestinal mucosal protein synthesis. Gastroenterology 118, 299–306.

Chase, J.R., Rothman, D.L., Shulman, R.G., 2001. Flux control in the rat gastrocnemius glycogen synthesis pathway by in vivo 13C/31P NMR spectroscopy. Amer. J. Physiol. 280, E598–E607.

Chinkes, D., Klein, S., Zhang, X.J., Wolfe, R.R., 1996a. Infusion of labeled KIC is more accurate thanlabeled leucine to determine human muscle protein synthesis. Amer. J. Physiol. 270, E67–E71.

Chinkes, D.L., Aarsland, A., Rosenblatt, J., Wolfe, R.R., 1996b. Comparison of mass isotopomer dilutionmethods used to compute VLDL production in vivo. Amer. J. Physiol. 271, E373–E383.

Clarke, J.T., Bier, D.M., 1982. The conversion of phenylalanine to tyrosine in man: direct measurementby continuous intravenous tracer infusions of L-[ring-2H5]phenylalanine and L-[1-13C] tyrosine in thepostabsorptive state. Metabolism 31, 999–1005.

Clarke, S.D., 2001. Polyunsaturated fatty acid regulation of gene transcription: a molecular mechanismto improve the metabolic syndrome. J. Nutr. 131, 1129–1132.

Craven, R.A., Banks, R.E., 2002. Use of laser capture microdissection to selectively obtain distinct populations of cells for proteomic analysis. Methods Enzymol. 356, 33–49.

Craven, R.A., Totty, N., Harnden, P., Selby, P.J., Banks, R.E., 2002. Laser capture microdissection andtwo-dimensional polyacrylamide gel electrophoresis: evaluation of tissue preparation and sample limitations. Amer. J. Pathol. 160, 815–822.

Crenn, P., Thuillier, F., Rakatoambinina, B., Rongier, M., Darmaun, D., Messing, B., 2000. Duodenal vs.gastric administration of labeled leucine for the study of splanchnic metabolism in humans. J. Appl.Physiol. 89, 573–580.

Darzynkiewicz, Z., Bedner, E., Smolewski, P., Lee, B.W., Johnson, G.L., 2002. Detection of caspasesactivation in situ by fluorochrome-labeled inhibitors of caspases (FLICA). Methods Mol. Biol. 203,289–299.

Davis, T.A., Fiorotto, M.L., Nguyen, H.V., Burrin, D.G., 1999. Aminoacyl-tRNA and tissue free aminoacid pools are equilibrated after a flooding dose of phenylalanine. Amer. J. Physiol. 277, E103–E109.

Davis, T.A., Reeds, P.J., 2001. Of flux and flooding: the advantages and problems of different isotopicmethods for quantifying protein turnover in vivo II. Methods based on the incorporation of a tracer.Curr. Opin. Clin. Nutr. Metab. Care 4, 51–56.

Di Buono, M., Jones, P.J., Beaumier, L., Wykes, L.J., 2000. Comparison of deuterium incorporation andmass isotopomer distribution analysis for measurement of human cholesterol biosynthesis. J. LipidRes. 41, 1516–1523.

Diraison, F., Pachiaudi, C., Beylot, M., 1997. Measuring lipogenesis and cholesterol synthesis in humanswith deuterated water: use of simple gas chromatographic/mass spectrometric techniques. J. MassSpectrom. 32, 81–86.

Dudley, M.A., Burrin, D.G., Wykes, L.J., Toffolo, G., Cobelli, C., Nichols, B.L., Rosenberger, J., Jahoor, F.,Reeds, P.J., 1998. Protein kinetics determined in vivo with a multiple-tracer, single-sample protocol:application to lactase synthesis. Amer. J. Physiol. 274, G591–G598.

Duplus, E., Glorian, M., Forest, C., 2000. Fatty acid regulation of gene transcription. J. Biol. Chem. 275,30749–30752.

Emmert-Buck, M.R., Bonner, R.F., Smith, P.D., Chuaqui, R.F., Zhuang, Z., Goldstein, S.R., Weiss, R.A.,Liotta, L.A., 1996. Laser capture microdissection. Science 274, 998–1001.

Fafournoux, P., Bruhat, A., Jousse, C., 2000. Amino acid regulation of gene expression. Biochem. J. 351, 1–12.Fend, F., Emmert-Buck, M.R., Chuaqui, R., Cole, K., Lee, J., Liotta, L.A., Raffeld, M., 1999. Immuno-

LCM: laser capture microdissection of immunostained frozen sections for mRNA analysis. Amer.J. Pathol. 154, 61–66.

Fiehn, O., Weckwerth, W., 2003. Deciphering metabolic networks. Eur. J. Biochem. 270, 579–588.Garlick, P.J., McNurlan, M.A., Essen, P., Wernerman, J., 1994. Measurement of tissue protein synthesis

rates in vivo: a critical analysis of contrasting methods. Amer. J. Physiol. 266, E287–E297.Gibson, U.E., Heid, C.A., Williams, P.M., 1996. A novel method for real time quantitative RT-PCR.

Genome Res. 6, 995–1001.


Gowrie, I.J., Roudsari, A.V., Umpleby, A.M., Hovorka, R., 1999. Estimating protein turnover with a[13N,15C]leucine tracer: a study using simulated data. J. Theor. Biol. 198, 165–172.

Gruetter, R., Magnusson, I., Rothman, D.L., Avison, M.J., Shulman, R.G., Shulman, G.I., 1994.Validation of 13C NMR measurements of liver glycogen in vivo. Magn. Reson. Med. 31, 583–588.

Gruetter, R., Prolla, T.A., Shulman, R.G., 1991. 13C NMR visibility of rabbit muscle glycogen in vivo.Magn. Reson. Med. 20, 327–332.

Guan, X., Bequette, B.J., Calder, G., Ku, P.K., Ames, K.N., Trottier, N.L., 2002. Amino acid availabilityaffects amino acid flux and protein metabolism in the porcine mammary gland. J. Nutr. 132, 1224–1234.

Guan, X., Stoll, B., Lu, X., Tappenden, K.A., Holst, J.J., Hartmann, B., Burrin, D.G., 2003. GLP-2-mediated up-regulation of intestinal blood flow and glucose uptake is nitric oxide-dependent inTPN-fed piglets. Gastroenterology 125, 136–147.

Guo, Z., Jensen, M.D., 1998. Intramuscular fatty acid metabolism evaluated with stable isotopic tracers.J. Appl. Physiol. 84, 1674–1679.

Guo, Z.K., Cella, L.K., Baum, C., Ravussin, E., Schoeller, D.A., 2000. De novo lipogenesis in adiposetissue of lean and obese women: application of deuterated water and isotope ratio mass spectrometry.Int. J. Obes. Relat. Metab. Disord. 24, 932–937.

Haab, B.B., Dunham, M.J., Brown, P.O., 2001. Protein microarrays for highly parallel detection andquantitation of specific proteins and antibodies in complex solutions. Genome Biol. 2, 1–13.

Halseth, A.E., Flakoll, P.J., Reed, E.K., Messina, A.B., Krishna, M.G., Lacy, D.B., Williams, P.E.,Wasserman, D.H., 1997. Effect of physical activity and fasting on gut and liver proteolysis in the dog.Amer. J. Physiol. 273, E1073–E1082.

Hammond, S.M., Caudy, A.A., Hannon, G.J., 2001. Post-transcriptional gene silencing by double-stranded RNA. Nat. Rev. Genet. 2, 110–119.

Hannon, G.J., 2002. RNA interference. Nature 418, 244–251.Heid, C.A., Stevens, J., Livak, K.J., Williams, P.M., 1996. Real time quantitative PCR. Genome Res. 6,

986–994.Hellerstein, M.K., 1999. Measurement of T-cell kinetics: recent methodologic advances. Immunol. Today

20, 438–441.Hellerstein, M., Hanley, M.B., Cesar, D., Siler, S., Papageorgopoulos, C., Wieder, E., Schmidt, D., Hoh, R.,

Neese, R., Macallan, D., Deeks, S., McCune, J.M., 1999. Directly measured kinetics of circulatingT lymphocytes in normal and HIV-1-infected humans. Nat. Med. 5, 83–89.

Hellerstein, M.K., Neese, R.A., 1999. Mass isotopomer distribution analysis at eight years: theoretical,analytic, and experimental considerations. Amer. J. Physiol. 276, E1146–E1170.

Hellerstein, M.K., Neese, R.A., Kim, Y.K., Valerie, S.S., Michelle, C., 2002. Measurement of synthesisrates of slow-turnover proteins from 2H2O incorporation into non-essential amino acids (NEAA) andapplication of mass isotopomer distribution analysis (MIDA). FASEB J. 16, A256.

Hesketh, J.E., Vasconcelos, M.H., Bermano, G., 1998. Regulatory signals in messenger RNA: determi-nants of nutrient-gene interaction and metabolic compartmentation. Brit. J. Nutr. 80, 307–321.

Horowitz, J.F., Coppack, S.W., Paramore, D., Cryer, P.E., Zhao, G., Klein, S., 1999. Effect of short-termfasting on lipid kinetics in lean and obese women. Amer. J. Physiol. 276, E278–E284.

Huang, R.P., 2001. Detection of multiple proteins in an antibody-based protein microarray system. J. Immunol. Methods 255, 1–13.

Izant, J.G., Weintraub, H., 1985. Constitutive and conditional suppression of exogenous and endogenousgenes by anti-sense RNA. Science 229, 345–352.

Jucker, B.M., Dufour, S., Ren, J., Cao, X., Previs, S.F., Underhill, B., Cadman, K.S., Shulman, G.I.,2000a. Assessment of mitochondrial energy coupling in vivo by 13C/31P NMR. Proc. Natl. Acad. Sci.USA 97, 6880–6884.

Jucker, B.M., Lee, J.Y., Shulman, R.G., 1998. In vivo 13C NMR measurements of hepatocellular tricar-boxylic acid cycle flux. J. Biol. Chem. 273, 12187–12194.

Jucker, B.M., Ren, J., Dufour, S., Cao, X., Previs, S.F., Cadman, K.S., Shulman, G.I., 2000b. 13C/31PNMR assessment of mitochondrial energy coupling in skeletal muscle of awake fed and fasted rats:relationship with uncoupling protein 3 expression. J. Biol. Chem. 275, 39279–39286.

Katome, T., Obata, T., Matsushima, R., Masuyama, N., Cantley, L.C., Gotoh, Y., Kishi, K., Shiota, H.,Ebina, Y., 2003. Use of RNA-interference-mediated gene silencing and adenoviral overexpression toelucidate the roles of AKT/PKB-isoforms in insulin actions. J. Biol. Chem., 278, 28312–28323.

Katz, J., Tayek, J.A., 1999. Recycling of glucose and determination of the Cori cycle and gluconeogenesis.Amer. J. Physiol. 277, E401–E407.


Kelleher, J.K., 1999. Estimating gluconeogenesis with [U-13C]glucose: molecular condensation requiresa molecular approach. Amer. J. Physiol. 277, E395–E400.

Kodadek, T., 2001. Protein microarrays: prospects and problems. Chem. Biol. 8, 105–115.Kohler, U., Ayre, B.G., Goodman, H.M., Haseloff, J., 1999. Trans-splicing ribozymes for targeted gene

delivery. J. Mol. Biol. 285, 1935–1950.Krebs, M., Krssak, M., Nowotny, P., Weghuber, D., Gruber, S., Mlynarik, V., Bischof, M., Stingl, H.,

Furnsinn, C., Waldhausl, W., Roden, M., 2001. Free fatty acids inhibit the glucose-stimulated increase ofintramuscular glucose-6-phosphate concentration in humans. J. Clin. Endocrinol. Metab. 86, 2153–2160.

Krssak, M., Petersen, K.F., Bergeron, R., Price, T., Laurent, D., Rothman, D.L., Roden, M., Shulman,G.I., 2000. Intramuscular glycogen and intramyocellular lipid utilization during prolonged exerciseand recovery in man: a 13C and 1H nuclear magnetic resonance spectroscopy study. J. Clin.Endocrinol. Metab. 85, 748–754.

Kurland, I.J., Alcivar, A., Bassilian, S., Lee, W.N., 2000. Loss of [13C]glycerol carbon via the pentosecycle: implications for gluconeogenesis measurement by mass isotoper distribution analysis. J. Biol.Chem. 275, 36787–36793.

Landau, B.R., 1999a. Glycerol production and utilization measured using stable isotopes. Proc. Nutr.Soc. 58, 973–978.

Landau, B.R., 1999b. Limitations in the use of [U-13C6]glucose to estimate gluconeogenesis. Amer.J. Physiol. 277, E408–E413.

Landau, B.R., 1999c. Quantifying the contribution of gluconeogenesis to glucose production in fastedhuman subjects using stable isotopes. Proc. Nutr. Soc. 58, 963–972.

Landau, B.R., Wahren, J., Chandramouli, V., Schumann, W.C., Ekberg, K., Kalhan, S.C., 1996.Contributions of gluconeogenesis to glucose production in the fasted state. J. Clin. Invest. 98, 378–385.

Landau, B.R., Wahren, J., Ekberg, K., Previs, S.F., Yang, D., Brunengraber, H., 1998. Limitations in esti-mating gluconeogenesis and Cori cycling from mass isotopomer distributions using [U-13C6]glucose.Amer. J. Physiol. 274, E954–E961.

Lebon, V., Petersen, K.F., Cline, G.W., Shen, J., Mason, G.F., Dufour, S., Behar, K.L., Shulman, G.I.,Rothman, D.L., 2002. Astroglial contribution to brain energy metabolism in humans revealed by 13Cnuclear magnetic resonance spectroscopy: elucidation of the dominant pathway for neurotransmitterglutamate repletion and measurement of astrocytic oxidative metabolism. J. Neurosci. 22, 1523–1531.

Lee Kang, S.H., Vieira, K., Bungert, J., 2002. Combining chromatin immunoprecipitation and DNA foot-printing: a novel method to analyze protein-DNA interactions in vivo. Nucleic Acids Res. 30, e44.

Lemieux, S., Patterson, B.W., Carpentier, A., Lewis, G.F., Steiner, G., 1999. A stable isotope methodusing a [2H5]glycerol bolus to measure very low density lipoprotein triglyceride kinetics in humans.J. Lipid Res. 40, 2111–2117.

Lindenthal, B., Aldaghlas, T.A., Holleran, A.L., Sudhop, T., Berthold, H.K., Von Bergmann, K., Kelleher,J.K., 2002. Isotopomer spectral analysis of intermediates of cholesterol synthesis in human subjectsand hepatic cells. Amer. J. Physiol. 282, E1222–E1230.

Ljungqvist, O.H., Persson, M., Ford, G.C., Nair, K.S., 1997. Functional heterogeneity of leucine pools inhuman skeletal muscle. Amer. J. Physiol. 273, E564–E570.

Lockhart, D.J., Winzeler, E.A., 2000. Genomics, gene expression and DNA arrays. Nature 405, 827–836.Luzzi, V., Holtschlag, V., Watson, M.A., 2001. Expression profiling of ductal carcinoma in situ by laser

capture microdissection and high-density oligonucleotide arrays. Amer. J. Pathol. 158, 2005–2010.Macallan, D.C., Fullerton, C.A., Neese, R.A., Haddock, K., Park, S.S., Hellerstein, M.K., 1998.

Measurement of cell proliferation by labeling of DNA with stable isotope-labeled glucose: studies invitro, in animals, and in humans. Proc. Natl. Acad. Sci. USA 95, 708–713.

Maghni, K., Nicolescu, O.M., Martin, J.G., 1999. Suitability of cell metabolic colorimetric assays forassessment of CD4+ T cell proliferation: comparison to 5-bromo-2-deoxyuridine (BrdU) ELISA. J. Immunol. Methods 223, 185–194.

Magnusson, I., Rothman, D.L., Jucker, B., Cline, G.W., Shulman, R.G., Shulman, G.I., 1994. Liverglycogen turnover in fed and fasted humans. Amer. J. Physiol. 266, E796–E803.

Mansoor, O., Cayol, M., Gachon, P., Boirie, Y., Schoeffler, P., Obled, C., Beaufrere, B., 1997. Albuminand fibrinogen syntheses increase while muscle protein synthesis decreases in head-injured patients.Amer. J. Physiol. 273, E898–E902.

Mao, C.S., Bassilian, S., Lim, S.K., Lee, W.N., 2002. Underestimation of gluconeogenesis by the [U-13C6]glucose method: effect of lack of isotope equilibrium. Amer. J. Physiol. Endocrinol. Metab.282, E376–E385.


Marshman, E., Ottewell, P.D., Potten, C.S., Watson, A.J., 2001. Caspase activation during spontaneousand radiation-induced apoptosis in the murine intestine. J. Pathol. 195, 285–292.

Martini, W.Z., Chinkes, D.L., Wolfe, R.R., 2002. DNA and protein turnover in human fibroblasts andmyocytes. FASEB J. 16, A613.

Mashimo, H., Goyal, R.K., 1999. Lessons from genetically engineered animal models. IV. Nitric oxidesynthase gene knockout mice. Amer. J. Physiol. 277, G745–G750.

Matthews, D.E., Marano, M.A., Campbell, R.G., 1993a. Splanchnic bed utilization of glutamine and glutamic acid in humans. Amer. J. Physiol. 264, E848–E854.

Matthews, D.E., Marano, M.A., Campbell, R.G., 1993b. Splanchnic bed utilization of leucine and phenylala-nine in humans. Amer. J. Physiol. 264, E109–E118.

Matthews, D.E., Schwarz, H.P., Yang, R.D., Motil, K.J., Young, V.R., Bier, D.M., 1982. Relationship ofplasma leucine and alpha-ketoisocaproate during a L-[1-13C]leucine infusion in man: a method formeasuring human intracellular leucine tracer enrichment. Metabolism 31, 1105–1112.

McCaffrey, A.P., Meuse, L., Pham, T.T., Conklin, D.S., Hannon, G.J., Kay, M.A., 2002. RNA interfer-ence in adult mice. Nature 418, 38–39.

McDevitt, R.M., Bott, S.J., Harding, M., Coward, W.A., Bluck, L.J., Prentice, A.M., 2001. De novo lipo-genesis during controlled overfeeding with sucrose or glucose in lean and obese women. Amer. J. Clin. Nutr. 74, 737–746.

McManus, M.T., Sharp, P.A., 2002. Gene silencing in mammals by small interfering RNAs. Nat. Rev.Genet. 3, 737–747.

Meek, S.E., Persson, M., Ford, G.C., Nair, K.S., 1998. Differential regulation of amino acid exchangeand protein dynamics across splanchnic and skeletal muscle beds by insulin in healthy human subjects. Diabetes 47, 1824–1835.

Mills, J.C., Roth, K.A., Cagan, R.L., Gordon, J.I., 2001. DNA microarrays and beyond: completing thejourney from tissue to cell. Nat. Cell Biol. 3, E175–E178.

Mittendorfer, B., Horowitz, J.F., Klein, S., 2001. Gender differences in lipid and glucose kinetics duringshort-term fasting. Amer. J. Physiol. 281, E1333–E1339.

Moller, N., Meek, S., Bigelow, M., Andrews, J., Nair, K.S., 2000. The kidney is an important site for invivo phenylalanine-to-tyrosine conversion in adult humans: a metabolic role of the kidney. Proc. Natl.Acad. Sci. USA 97, 1242–1246.

Neese, R.A., Misell, L.M., Turner, S., Chu, A., Kim, J., Cesar, D., Hoh, R., Antelo, F., Strawford, A.,McCune, J.M., Christiansen, M., Hellerstein, M.K., 2002. Measurement in vivo of proliferation ratesof slow turnover cells by 2H2O labeling of the deoxyribose moiety of DNA. Proc. Natl. Acad. Sci.USA 99, 15345–15350.

Neese, R.A., Schwarz, J.M., Faix, D., Turner, S., Letscher, A., Vu, D., Hellerstein, M.K., 1995.Gluconeogenesis and intrahepatic triose phosphate flux in response to fasting or substrate loads:application of the mass isotopomer distribution analysis technique with testing of assumptions andpotential problems. J. Biol. Chem. 270, 14452–14466.

Neese, R.A., Siler, S.Q., Cesar, D., Antelo, F., Lee, D., Misell, L., Patel, K., Tehrani, S., Shah, P.,Hellerstein, M.K., 2001. Advances in the stable isotope-mass spectrometric measurement of DNAsynthesis and cell proliferation. Anal. Biochem. 298, 189–195.

Nicholson, J.K., Connelly, J., Lindon, J.C., Holmes, E., 2002. Metabonomics: a platform for studyingdrug toxicity and gene function. Nat. Rev. Drug Discov. 1, 153–161.

Palmer, R.M., Ashton, D.S., Moncada, S., 1988. Vascular endothelial cells synthesize nitric oxide fromL-arginine. Nature 333, 664–666.

Pandey, A., Mann, M., 2000. Proteomics to study genes and genomes. Nature 405, 837–846.Paolini, C.L., Meschia, G., Fennessey, P.V., Pike, A.W., Teng, C., Battaglia, F.C., Wilkening, R.B., 2001.

An in vivo study of ovine placental transport of essential amino acids. Amer. J. Physiol. 280,E31–E39.

Papageorgopoulos, C., Caldwell, K., Shackleton, C., Schweingrubber, H., Hellerstein, M.K., 1999.Measuring protein synthesis by mass isotopomer distribution analysis (MIDA). Anal. Biochem.267, 1–16.

Patterson, B.W., 1997. Use of stable isotopically labeled tracers for studies of metabolic kinetics: anoverview. Metabolism 46, 322–329.

Petersen, K.F., Krssak, M., Navarro, V., Chandramouli, V., Hundal, R., Schumann, W.C., Landau, B.R.,Shulman, G.I., 1999. Contributions of net hepatic glycogenolysis and gluconeogenesis to glucose production in cirrhosis. Amer. J. Physiol. 276, E529–E535.


Petersen, K.F., Laurent, D., Rothman, D.L., Cline, G.W., Shulman, G.I., 1998. Mechanism by which glucose and insulin inhibit net hepatic glycogenolysis in humans. J. Clin. Invest. 101, 1203–1209.

Previs, S., 2002. Application of 2H2O for estimating protein synthesis (PS) and protein breakdown (PB)in vivo. FASEB J. 16, A788.

Previs, S.F., Fernandez, C.A., Yang, D., Soloviev, M.V., David, F., Brunengraber, H., 1995. Limitationsof the mass isotopomer distribution analysis of glucose to study gluconeogenesis: substrate cyclingbetween glycerol and triose phosphates in liver. J. Biol. Chem. 270, 19806–19815.

Previs, S.F., Hallowell, P.T., Neimanis, K.D., David, F., Brunengraber, H., 1998. Limitations of the massisotopomer distribution analysis of glucose to study gluconeogenesis: heterogeneity of glucose label-ing in incubated hepatocytes. J. Biol. Chem. 273, 16853–16859.

Raamsdonk, L.M., Teusink, B., Broadhurst, D., Zhang, N., Hayes, A., Walsh, M.C., Berden, J.A.,Brindle, K.M., Kell, D.B., Rowland, J.J., Westerhoff, H.V., van Dam, K., Oliver, S.G., 2001. A functionalgenomics strategy that uses metabolome data to reveal the phenotype of silent mutations. Nat. Biotechnol.19, 45–50.

Radziuk, J., Lee, W.P., 1999. Measurement of gluconeogenesis and mass isotopomer analysis based on[U-13C]glucose. Amer. J. Physiol. 277, E199–E207.

Rapley, R., Walker, J.M., 1998. Molecular Biomethods Handbook. Humana Press, Totowa, NJ.Reeds, P.J., Davis, T.A., 1999. Of flux and flooding: the advantages and problems of different isotopic

methods for quantifying protein turnover in vivo I. Methods based on the dilution of a tracer. Curr.Opin. Clin. Nutr. Metab. Care 2, 23–28.

Reeds, P.J., Hachey, D.L., Patterson, B.W., Motil, K.J., Klein, P.D., 1992. VLDL apolipoprotein B-100,a potential indicator of the isotopic labeling of the hepatic protein synthetic precursor pool in humans:studies with multiple stable isotopically labeled amino acids. J. Nutr. 122, 457–466.

Renaudeau, D., Lebreton, Y., Noblet, J., Dourmad, J.Y., 2002. Measurement of blood flow through themammary gland in lactating sows: methodological aspects. J. Anim. Sci. 80, 196–201.

Rennie, M.J., Smith, K., Watt, P.W., 1994. Measurement of human tissue protein synthesis: an optimalapproach. Amer. J. Physiol. 266, E298–E307.

Reue, K., 1998. mRNA quantitation techniques: considerations for experimental design and application.J. Nutr. 128, 2038–2044.

Roden, M., Perseghin, G., Petersen, K.F., Hwang, J.H., Cline, G.W., Gerow, K., Rothman, D.L.,Shulman, G.I., 1996. The roles of insulin and glucagon in the regulation of hepatic glycogen synthe-sis and turnover in humans. J. Clin. Invest. 97, 642–648.

Roden, M., Shulman, G.I., 1999. Applications of NMR spectroscopy to study muscle glycogen metabo-lism in man. Annu. Rev. Med. 50, 277–290.

Roesler, W.J., 2001. The role of C/EBP in nutrient and hormonal regulation of gene expression. Annu.Rev. Nutr. 21, 141–165.

Rothman, D.L., Magnusson, I., Cline, G., Gerard, D., Kahn, C.R., Shulman, R.G., Shulman, G.I., 1995.Decreased muscle glucose transport/phosphorylation is an early defect in the pathogenesis of non-insulin-dependent diabetes mellitus. Proc. Natl. Acad. Sci. USA 92, 983–987.

Rothman, D.L., Magnusson, I., Katz, L.D., Shulman, R.G., Shulman, G.I., 1991. Quantitation of hepaticglycogenolysis and gluconeogenesis in fasting humans with 13C NMR. Science 254, 573–576.

Ryding, A.D., Sharp, M.G., Mullins, J.J., 2001. Conditional transgenic technologies. J. Endocrinol. 171,1–14.

Sawada, S., Asakura, S., Daimon, H., Furihata, C., 1995. Comparison of autoradiography, liquid scintil-lation counting and immunoenzymatic staining of 5-bromo-2′-deoxyuridine for measurement ofunscheduled DNA synthesis and replicative DNA synthesis in rat liver. Mutat. Res. 344, 109–116.

Shannon, M.F., Rao, S., 2002. Transcription. of chips and ChIPs. Science 296, 666–669.Shen, J., Petersen, K.F., Behar, K.L., Brown, P., Nixon, T.W., Mason, G.F., Petroff, O.A., Shulman, G.I.,

Shulman, R.G., Rothman, D.L., 1999. Determination of the rate of the glutamate/glutamine cycle inthe human brain by in vivo 13C NMR. Proc. Natl. Acad. Sci. USA 96, 8235–8240.

Shi, Y., 2002. Mechanisms of caspase activation and inhibition during apoptosis. Mol. Cell 9, 459–470.Short, K.R., Meek, S.E., Moller, N., Ekberg, K., Nair, K.S., 1999. Whole body protein kinetics using Phe

and Tyr tracers: an evaluation of the accuracy of approximated flux values. Amer. J. Physiol. 276,E1194–E1200.

Shulman, G.I., Rothman, D.L., Jue, T., Stein, P., DeFronzo, R.A., Shulman, R.G., 1990. Quantitation ofmuscle glycogen synthesis in normal subjects and subjects with non-insulin-dependent diabetes by13C nuclear magnetic resonance spectroscopy. N. Engl. J. Med. 322, 223–228.


Shulman, R.G., Hyder, F., Rothman, D.L., 2001. Cerebral energetics and the glycogen shunt: neuro-chemical basis of functional imaging. Proc. Natl. Acad. Sci. USA 98, 6417–6422.

Shulman, R.G., Rothman, D.L., 2001. 13C NMR of intermediary metabolism: implications for systemicphysiology. Annu. Rev. Physiol. 63, 15–48.

Sibson, N.R., Dhankhar, A., Mason, G.F., Rothman, D.L., Behar, K.L., Shulman, R.G., 1998.Stoichiometric coupling of brain glucose metabolism and glutamatergic neuronal activity. Proc. Natl.Acad. Sci. USA 95, 316–321.

Sibson, N.R., Mason, G.F., Shen, J., Cline, G.W., Herskovits, A.Z., Wall, J.E., Behar, K.L., Rothman, D.L.,Shulman, R.G., 2001. In vivo 13C NMR measurement of neurotransmitter glutamate cycling, anaplero-sis and TCA cycle flux in rat brain during [2-13C]glucose infusion. J. Neurochem. 76, 975–989.

Sidossis, L.S., Coggan, A.R., Gastaldelli, A., Wolfe, R.R., 1995a. A new correction factor for use in tracerestimations of plasma fatty acid oxidation. Amer. J. Physiol. 269, E649–E656.

Sidossis, L.S., Coggan, A.R., Gastaldelli, A., Wolfe, R.R., 1995b. Pathway of free fatty acid oxidation inhuman subjects: implications for tracer studies. J. Clin. Invest. 95, 278–284.

Sidossis, L.S., Mittendorfer, B., Chinkes, D., Walser, E., Wolfe, R.R., 1999. Effect of hyperglycemia-hyperinsulinemia on whole body and regional fatty acid metabolism. Amer. J. Physiol. 276, E427–E434.

Sidossis, L.S., Mittendorfer, B., Walser, E., Chinkes, D., Wolfe, R.R., 1998. Hyperglycemia-induced inhi-bition of splanchnic fatty acid oxidation increases hepatic triacylglycerol secretion. Amer. J. Physiol.275, E798–E805.

Siler, S.Q., Neese, R.A., Hellerstein, M.K., 1999. De novo lipogenesis, lipid kinetics, and whole-bodylipid balances in humans after acute alcohol consumption. Amer. J. Clin. Nutr. 70, 928–936.

Siler, S.Q., Neese, R.A., Parks, E.J., Hellerstein, M.K., 1998. VLDL-triglyceride production after alcoholingestion, studied using [2-13C1] glycerol. J. Lipid Res. 39, 2319–2328.

Simone, N.L., Paweletz, C.P., Charboneau, L., Petricoin, E.F. 3rd, Liotta, L.A., 2000a. Laser capturemicrodissection: beyond functional genomics to proteomics. Mol. Diagn. 5, 301–307.

Simone, N.L., Remaley, A.T., Charboneau, L., Petricoin, E.F. 3rd, Glickman, J.W., Emmert-Buck, M.R.,Fleisher, T.A., Liotta, L.A., 2000b. Sensitive immunoassay of tissue cell proteins procured by lasercapture microdissection. Amer. J. Pathol. 156, 445–452.

Smolewski, P., Grabarek, J., Halicka, H.D., Darzynkiewicz, Z., 2002. Assay of caspase activation in situcombined with probing plasma membrane integrity to detect three distinct stages of apoptosis. J. Immunol. Methods 265, 111–121.

Sok, J., Wang, X.Z., Batchvarova, N., Kuroda, M., Harding, H., Ron, D., 1999. CHOP-dependent stress-inducible expression of a novel form of carbonic anhydrase VI. Mol. Cell Biol. 19, 495–504.

Stahelin, B.J., Marti, U., Solioz, M., Zimmermann, H., Reichen, J., 1998. False positive staining in theTUNEL assay to detect apoptosis in liver and intestine is caused by endogenous nucleases and inhib-ited by diethyl pyrocarbonate. Mol. Pathol. 51, 204–208.

Stappenbeck, T.S., Hooper, L.V., Manchester, J.K., Wong, M.H., Gordon, J.I., 2002. Laser capturemicrodissection of mouse intestine: characterizing mRNA and protein expression, and profiling inter-mediary metabolism in specified cell populations. Methods Enzymol. 356, 167–196.

Stillman, B.A., Tonkinson, J.L., 2001. Expression microarray hybridization kinetics depend on length of the immobilized DNA but are independent of immobilization substrate. Anal. Biochem. 295,149–157.

Stoll, B., Burrin, D.G., Henry, J., Jahoor, F., Reeds, P.J., 1997. Phenylalanine utilization by the gut andliver measured with intravenous and intragastric tracers in pigs. Amer. J. Physiol. 273, G1208–G1217.

Stoll, B., Burrin, D.G., Henry, J., Yu, H., Jahoor, F., Reeds, P.J., 1999a. Substrate oxidation by the portaldrained viscera of fed piglets. Amer. J. Physiol. 277, E168–E175.

Stoll, B., Burrin, D.G., Henry, J.F., Jahoor, F., Reeds, P.J., 1999b. Dietary and systemic phenylalanine uti-lization for mucosal and hepatic constitutive protein synthesis in pigs. Amer. J. Physiol. 276, G49–G57.

Stoll, B., Chang, X., Fan, M.Z., Reeds, P.J., Burrin, D.G., 2000. Enteral nutrient intake level determinesintestinal protein synthesis and accretion rates in neonatal pigs. Amer. J. Physiol. Gastrointest. LiverPhysiol. 279, G288–G294.

Stoll, B., Henry, J., Reeds, P.J., Yu, H., Jahoor, F., Burrin, D.G., 1998. Catabolism dominates the first-pass intestinal metabolism of dietary essential amino acids in milk protein-fed piglets. J. Nutr. 128,606–614.

Szczepaniak, L.S., Babcock, E.E., Schick, F., Dobbins, R.L., Garg, A., Burns, D.K., McGarry, J.D., Stein, D.T.,1999. Measurement of intracellular triglyceride stores by H spectroscopy: validation in vivo. Amer.J. Physiol. 276, E977–E989.


Tayek, J.A., Katz, J., 1996. Glucose production, recycling, and gluconeogenesis in normals and diabetics:a mass isotopomer [U-13C]glucose study. Amer. J. Physiol. 270, E709–E717.

Tayek, J.A., Katz, J., 1997. Glucose production, recycling, Cori cycle, and gluconeogenesis in humans:relationship to serum cortisol. Amer. J. Physiol. 272, E476–E484.

Taylor, R., Price, T.B., Rothman, D.L., Shulman, R.G., Shulman, G.I., 1992. Validation of 13C NMRmeasurement of human skeletal muscle glycogen by direct biochemical assay of needle biopsy sam-ples. Magnet. Reson. Med. 27, 13–20.

Tessari, P., Garibotto, G., Inchiostro, S., Robaudo, C., Saffioti, S., Vettore, M., Zanetti, M., Russo, R.,Deferrari, G., 1996. Kidney, splanchnic, and leg protein turnover in humans: insight from leucine andphenylalanine kinetics. J. Clin. Invest. 98, 1481–1492.

Thompson, G.N., Pacy, P.J., Merritt, H., Ford, G.C., Read, M.A., Cheng, K.N., Halliday, D., 1989. Rapidmeasurement of whole body and forearm protein turnover using a [2H5]phenylalanine model. Amer.J. Physiol. 256, E631–E639.

Toth, M.J., MacCoss, M.J., Poehlman, E.T., Matthews, D.E., 2001. Recovery of 13CO2 from infused [1-13C]leucine and [1,2-13C2]leucine in healthy humans. Amer. J. Physiol. Endocrinol. Metab. 281,E233–E241.

Trimmer, J.K., Schwarz, J.M., Casazza, G.A., Horning, M.A., Rodriguez, N., Brooks, G.A., 2002.Measurement of gluconeogenesis in exercising men by mass isotopomer distribution analysis. J. Appl. Physiol. 93, 233–241.

Trottier, N.L., Shipley, C.F., Easter, R.A., 1997. Plasma amino acid uptake by the mammary gland of thelactating sow. J. Anim. Sci. 75, 1266–1278.

Turner, S.M., Neese, R.A., Murphy, E., Antelo, F., Hellerstein, M., 2002. Measurement of triglyceride(TG) synthesis in vivo by 2H2O incorporation into TG-glycerol and application of mass isotopomerdistribution analysis (MIDA). FASEB J. 16, A400.

Ubeda, M., Wang, X.Z., Zinszner, H., Wu, I., Habener, J.F., Ron, D., 1996. Stress-induced binding of thetranscriptional factor CHOP to a novel DNA control element. Mol. Cell Biol. 16, 1479–1489.

van Goudoever, J.B., Stoll, B., Henry, J.F., Burrin, D.G., Reeds, P.J., 2000. Adaptive regulation of intes-tinal lysine metabolism. Proc. Natl. Acad. Sci. USA 97, 11620–11625.

van Spronsen, F.J., Reijngoud, D.J., Smit, G.P., Nagel, G.T., Stellaard, F., Berger, R., Heymans, H.S.,1998. Phenylketonuria: the in vivo hydroxylation rate of phenylalanine into tyrosine is decreased. J. Clin. Invest. 101, 2875–2880.

Vasquez, K.M., Marburger, K., Intody, Z., Wilson, J.H., 2001. Manipulating the mammalian genome byhomologous recombination. Proc. Natl. Acad. Sci. USA 98, 8403–8410.

Volpi, E., Ferrando, A.A., Yeckel, C.W., Tipton, K.D., Wolfe, R.R., 1998. Exogenous amino acids stimulate net muscle protein synthesis in the elderly. J. Clin. Invest. 101, 2000–2007.

Volpi, E., Jeschke, M.G., Herndon, D.N., Wolfe, R.R., 2000. Measurement of skin protein breakdown ina rat model. Amer. J. Physiol. Endocrinol. Metab. 279, E900–E906.

Walker, J.A., Quirke, P., 2001. Viewing apoptosis through a “TUNEL”. J. Pathol. 195, 275–276.Wang, T., Brown, M.J., 1999. mRNA quantification by real time TaqMan polymerase chain reaction:

validation and comparison with RNase protection. Anal. Biochem. 269, 198–201.Wang, W., Basinger, A., Neese, R.A., Christiansen, M., Hellerstein, M.K., 2000. Effects of nicotinic acid

on fatty acid kinetics, fuel selection, and pathways of glucose production in women. Amer. J. Physiol.279, E50–E59.

Waterlow, J.C., Stephen, J.M., 1967. The measurement of total lysine turnover in the rat by intravenousinfusion of L-[U-14C]lysine. Clin. Sci. 33, 489–506.

Watkins, S.M., Reifsnyder, P.R., Pan, H.J., German, J.B., Leiter, E.H., 2002. Lipid metabolome-wideeffects of the PPARgamma agonist rosiglitazone. J. Lipid Res. 43, 1809–1817.

Watt, P.W., Lindsay, Y., Scrimgeour, C.M., Chien, P.A., Gibson, J.N., Taylor, D.J., Rennie, M.J., 1991.Isolation of aminoacyl-tRNA and its labeling with stable-isotope tracers: use in studies of humantissue protein synthesis. Proc. Natl. Acad. Sci. USA 88, 5892–5896.

Weinmann, A.S., Yan, P.S., Oberley, M.J., Huang, T.H., Farnham, P.J., 2002. Isolating human transcrip-tion factor targets by coupling chromatin immunoprecipitation and CpG island microarray analysis.Gene. Dev. 16, 235–244.

Witzmann, F.A., Li, J., 2002. Cutting-edge technology. II. Proteomics: core technologies and applicationsin physiology. Amer. J. Physiol. 282, G735–G741.

Wolfe, R.R., 1992. Radioactive and Stable Isotope Tracers in Biomedicine: Principles and Practice ofKinetic Analysis. Wiley-Liss, New York.


Wong, M.H., Saam, J.R., Stappenbeck, T.S., Rexer, C.H., Gordon, J.I., 2000. Genetic mosaic analysisbased on Cre recombinase and navigated laser capture microdissection. Proc. Natl. Acad. Sci. USA97, 12601–12606.

Xu, J., Nakamura, M.T., Cho, H.P., Clarke, S.D., 1999. Sterol regulatory element binding protein-1expression is suppressed by dietary polyunsaturated fatty acids: a mechanism for the coordinate sup-pression of lipogenic genes by polyunsaturated fats. J. Biol. Chem. 274, 23577–23583.

Xu, J., Teran-Garcia, M., Park, J.H., Nakamura, M.T., Clarke, S.D., 2001. Polyunsaturated fatty acidssuppress hepatic sterol regulatory element-binding protein-1 expression by accelerating transcriptdecay. J. Biol. Chem. 276, 9800–9807.

Yang, V.W., 1998. Eukaryotic transcription factors: identification, characterization and functions. J. Nutr.128, 2045–2051.

Zaidi, A.U., Enomoto, H., Milbrandt, J., Roth, K.A., 2000. Dual fluorescent in situ hybridization andimmunohistochemical detection with tyramide signal amplification. J. Histochem. Cytochem. 48,1369–1375.

Zhang, X.J., Chinkes, D.L., Sakurai, Y., Wolfe, R.R., 1996. An isotopic method for measurement ofmuscle protein fractional breakdown rate in vivo. Amer. J. Physiol. 270, E759–E767.

Zhang, X.J., Chinkes, D.L., Wolfe, R.R., 2002. Measurement of muscle protein fractional synthesis andbreakdown rates from a pulse tracer injection. Amer. J. Physiol. 283, E753–E764.


Index

3-methyl histidine, 725′deiodinase, 309, 311, 318α-cardiac MHC, 360β-adrenergic agonists, 77, 79, 289–291

A“Absorptive use”, 202Acetyl CoA carboxylase (ACC), 224, 226Acute-phase protein, 84, 85, 90–92Acyl CoA synthetase (ACS), 224, 226Adaptive immunity, 85Adipose tissue, 337Amino acid absorption, 198, 201, 207Amino acid oxidation, 197, 207Amino acid requirements, 128Amino acids, 5, 6, 8–10, 24–27, 29, 35, 51, 53,

55–57, 59, 107–108, 110–113, 116–118,120–122

Ammonia absorption and liver urea synthesis, 208Animal growth, 21, 118Anorexia, 86–88, 90Apoptosis, 303, 313, 315Arachidonic acid

Functional roles, 13–15Infant nutrition, 34–40Metabolism, 8–12, 32–34Milk composition, 26–29Placental transfer, 23–26

Arginine, 161, 166, 168–169, 173, 177–180Arterio-venous tracer balance method, 453ATP, 356, 359, 361–362, 367

BBlood flow, 361Brain development

Docosahexaenoic acid, 18–23N-3 fatty acid deficiency, 15–23Polyunsaturated fatty acid accretion, 29–32Polyunsaturated fatty acid metabolism, 30–34

Branched chain amino acids, 205Bromodeoxyuridine (BrdU) labeling assay, 445Brown adipose tissue (BAT)

β3-adrenergic receptors, 305Adipocytes, 305Brown unknown gene, 305Fatty acid metabolism, 310–311Fetal development, 305Morphology, 304Quantity, 304Thermogenesis, 304Uncoupling protein-1, 304, 308–309,

311–312Brown adipose tissue, 276, 353, 357Brown unknown gene (BUG), 6, 305

CCalcineurin, 13, 44Calpain, 51, 84, 98–99Cardiac output, 353, 356, 361Carnitine palmitoyltransferase (CPT), 224–225,

229, 379, 383Catecholamines, 14,18, 20–21, 368Cathepsin, 84, 99cDNA microarray, 437Cell apoptosis, 446Cell culture, 6, 279Cell number, 277Cell proliferation, 445Cell signalling, 180Cell-mediated immunity, 85Chromatin immunoprecipitation assay, 441Chylomicron, 329–330, 332, 340Citrulline, 166, 169Cold, 354–356, 358, 360, 362, 365, 367–368Colostrum, 2, 4, 15, 17–19, 26, 57–58Compartment modeling, 449Composition, 281Conceptus, 4, 6, 7, 11, 24Confocal laser scanning microscopy, 444Conjugated linoleic acid, 290Constant infusion method, 449Copper, 318Corticotropin-releasing hormone, 87Cortisol, 16, 18–20, 23, 368, 380, 383–384Cost of urea synthesis, 208

CPT-1, 365–366Cysteine, 169–170, 179, 181Cytochrome c oxidase, 311–312Cytokine, 83–100Cytokines, 178, 182

DDevelopment, 108, 119, 277, 282Differentiation, 277–280Distribution, 276DNase I protection assay, 441Docosahexaenoic acid

Behaviour, 18–23Brain accretion, 29–32Brain function, 18–23, 37–40Dopamine, 19–20Functional roles, 241–243Infant nutrition, 254Metabolism, 237–241, 253–254Milk composition, 248–249Neurotransmitters, 19–21Placental transfer, 23–26Serotonin, 21Visual function, 15–18, 36–40

EElectrophoretic mobility shift assay, 441Endocrine functions, 282Endocrine regulation–anabolism, 289–290tEndocrine regulation–catabolism, 18–22tEndogenous amino acid secretions, 198, 206Energy metabolism, 353, 355, 359, 368–369Energy stores, 356, 363, 369Enteral, 161, 169–173, 176Enterocytes, 108–118, 120–121Essential fatty acids

Requirements, 40–43Deficiency, 7,12,13Gene expression, 15Infant nutrition, 34–40Metabolism, 8–12, 32–34Milk composition, 26–29Placental transfer, 23–26

Essential fatty acids, 324, 334–335Esterification, 329, 339, 342Eukaryotic initiation factor, 94–96Eukaryotic initiation factors, 29, 50–53, 55–57Excess protein digestion, 208Expression of enzyme data, 18, 287

FFast–twitch muscle fibres, 22, 38, 44–45,

53, 55Fast-twitch muscle, 91Fatty acid binding protein, 328, 336Fatty acid oxidation, 342–343, 353

Fatty acid synthesis, 282–290Fatty acids

Metabolism, 310–311Polyunsaturated (PUFA), 316–317Saturated, 316–317

Fatty acids, 5, 8–11, 23Feed intake, 1–9, 83–88, 90, 94Feeding, 51–57Fetal programming, 4, 23Fetus, 5–8, 11, 14–16, 19, 21First pass metabolism, 141First-pass utilization, 458Flooding dose method, 451Functions, 276

GGastrocnemius muscle, 90–91, 95–97Gene expression

Polyunsaturated fatty acids, 15Gene expression, 400Glucagon, 15, 158, 164, 176–177, 379–380,

383–385Glucocorticoids, 176, 379, 383–385Gluconeogenesis, 6, 10, 12, 15, 213, 353, 367,

462–463Glucose transport, 395Glucose, 5–8, 10, 12, 14, 17–18, 21, 23, 406Glucose-6-phosphatase, 377–378Glutamate, 161, 166–171, 173, 179, 182Glutamine, 92, 160–161, 166, 168–169, 173,

176–182Glutathione, 24, 39, 164, 169–171, 179, 181Glycogen synthesis, 466Glycogen, 16, 17, 378–381Glycogenolysis, 466Glycolytic, 16, 22, 31, 38, 44–45, 48, 53Growth hormone releasing hormone, 84, 93Growth hormone, 15–17, 19–20, 22, 24, 29, 37, 41,

45, 56–58, 77, 84, 93, 383–384

HHigh-density lipoprotein, 333, 341Humoral immunity, 85Hydroxy-methylglutarylCoA synthase

(HMGCS), 224Hyperplasia, 277–278Hypertrophy, 281–282

IIGF-1, 25, 30–31, 77, 175, 369Immune system, 83–86, 90, 93, 99–100Immunonutrients, 39, 157, 179, 183in situ hybridization, 439Indicator amino acid oxidation, 128Innate immunity, 4–5Insulin receptor substrate, 16, 93

Index480

Insulin, 12, 14–15, 18–19, 22–23, 28, 30, 46, 50,53–59, 158, 164, 174, 176, 180–181,288–289, 292, 379–383, 385

Insulin-like growth factor 1, 16, 20Insulin-like growth factor 2, 15–16, 18, 20–22Insulin-like growth factor binding protein, 19–21,

94–96Insulin-like growth factor-I, 84, 92–96Insulin-like growth factors, 40–41, 48, 56–58Interferon, 85, 94Interleukin-1, 83–91, 93, 97Interleukin-6, 84–91, 93, 97–99Interleukin-1 receptor antagonist, 88, 90, 95Intestine, lipid metabolism, 328Intrauterine growth retardation, 3, 11, 19, 21, 23Involution, 310–311Isotopic labelling, 201–202, 206

KKetogenesis, 222–223, 225, 227–229, 394,

396–398, 400

LLactate, 375, 377–383, 386, 410Laser capture microdissection, 447Leptin, 19, 22, 89, 337, 341Leucine, 161, 164, 167–168, 172–173, 176, 180–181Linoleic acid

Placental transfer, 23–26Deficiency, 7, 12–13Dietary requirements, 40–43Functional roles, 13–15Infant nutrition, 34–40Metabolism, 8–12, 32–34Milk composition, 26–29

Linolenic acid (alpha)Deficiency, 7,12–14Dietary requirements, 40–43Functional roles, 13–23Infant nutrition, 34–40Metabolism, 8–12, 32–34Milk, 26–29

Lipid degradation, 290f, 292Lipid digestion, preruminants, 324, 327Lipid digestion, ruminants, 326–327Lipid metabolism, 337, 341Lipid synthesis, 13f, 282Lipogenesis, 337, 339, 353, 366, 461, 463Lipolysis, 13, 24–26, 282f, 290–291, 340Lipopolysaccharide, 6–8, 10–13, 16, 19, 23, 86Lipoprotein lipase, 13, 20, 282, 288, 330Lipoprotein metabolism, 329–330, 332, 341Liver amino acid metabolism, 211Liver, lipid metabolism, 325, 341Long chain polyunsaturated fatty acid See also

docosahexaenoic acid, arachidonic acidBrain, 243–246, 251–252Dietary requirements, 253–258

Infant nutrition, 254Metabolism, 237–241, 253–254Milk, 248–249Placental transfer, 246–248

Low-density lipoprotein, 341Lysine, 165, 167, 173, 176

MMammalian target of rapamycin, 50, 52–53Mass isotopomer distribution analysis, 462MCFA, 364–365Melanocyte-stimulating hormone, 7Membrane function

By contractile activity, 75By feeding and diet, 12–15, 76By inflammation and injury, 75tBy stress, 76–77Endocrine and autocrine controls, 15–16, 75Genetic makeup, 74

Metabolism, 408Metabolomics, 467Methionine, 23, 41–43, 169–170, 176–177, 179,

181–182Methionyl-tRNA, 50Microflora, 43, 163, 177–178, 182Milk

Polyunsaturated fatty acids, 241–246Mitochondria, 306, 312, 353, 357–359, 362,

364–365Molecular aspects–differentiation, 278–279mRNA quantitative technique, 438Mucin, 164, 169–170Multi-catheterization, 202Muscle, 1, 5, 7–12, 14, 16–18, 25Muscle hyperplasia, 37, 39–41, 59Muscle hypertrophy, 40, 47Muscle, lipid metabolism, 335Myoblasts, 41–42MyoD, 39, 41Myofibres, 37, 40–42, 58Myofibril, 356, 359–360Myofibrillar protein, 12, 21, 23–24, 26Myofibrillar proteins, 38, 42, 46, 53, 57, 59Myogenesis, 39–40Myogenic regulatory factors, 24, 27–30Myogenin, 39, 41Myosin, 36, 38, 43–44Myostatin, 40–41, 58, 60 Myotubes, 41–42

NNeuropeptide Y, 87Newborn mammals, 275–276Nonesterified fatty acid uptake and oxidation,

460–461Norepinephrine (NE), 311, 316Northern blotting analysis, 438Nuclear magnet resonance spectroscopy, 465

Index 481

Nucleotides, 3, 169, 179–180Nutritional efficiency, 121

OOntogeny, 399Ornithine, 166, 168, 177, 180Oxidation, 157, 160, 165–168, 171, 173, 177Oxidative fuels, 161, 165–166, 168, 182Oxygen, 5–7, 18

PPapillae development, 399Parenteral, 27, 170, 173Peroxisomal β-oxidation, 342Peroxisome, 223–224, 227, 230Phosphoenolpyruvate carboxykinase, 377–378,

382–385Pig, 353Placenta

Polyunsaturated fatty acid metabolism, 23Polyunsaturated fatty acid transfer, 23–26

Placenta, 5–11, 14, 19Placental lactogen, 16, 18Polyamines, 168, 170Portal absorption, 160Portal-drained viscera, 158–159, 161, 164, 167,

169–170, 173, 176Portal-drained visceral amino acid sequestration,

198–199Post mortem, 78Preadipocytes, 279–280Proliferation, 44, 157, 163, 169, 173,

178–182Proline, 10, 20, 161, 163, 166, 168–169Protease

ATP-ubiquitin-dependent, 6, 17, 71, 74Calpain, 71, 74Gene expression, 73Lysosomal, 71, 74Matrix metalloprotease, 72

Proteasome, 22, 96–97Protein accretion 83, 83–87, 90–92, 98–100Protein breakdown, 448, 454Protein degradation, 3, 23, 25, 48, 51, 56, 59,

70, 83, 90–92, 96–100, 163, 172–173, 180

Determination of, 73-methylhistidine, 72Difference methods, 72In vitro approaches, 74Isotopic tracer approaches, 73

Protein kinase B, 53–54, 57Protein microarray, 444Protein synthesis, 4, 7, 10, 15, 37, 40, 46–57, 59,

69, 72, 83, 86, 90–95, 100, 157–158, 162,170, 172–182, 448, 463

Pyruvate carboxylase, 375, 378

RRates, 69, 70Real-time RT-PCR, 440Recycling of nitrogen, 198, 207, 214Redox status, 157, 181–182Regulation

Polyunsaturated fatty acids, 13–23Regulation-fatty acid synthesis, 282, 284–286Regulation-lipolysis, 282tRegulation-triacylglycerol, 288–289t, 290Ribosomal protein S6 kinase, 26, 51–52,

54–55, 57Ribosomes, 48–49, 57RNA interference, 436RNase protection assay, 439Rumen acidosis, 395Ruminant, 405

SSatellite cells, 39, 42, 47–48, 58–60Sepsis, 91, 95–96, 98, 99Shivering, 356–357, 359–360, 364Short-chain fatty acids, 414Skeletal muscle, 14, 19, 21, 70, 73, 83–86,

90–100Slow-twitch (SO) muscle fibres, 38, 44,

49, 55Slow-twitch muscle, 91Small intestine, 107–121Somatotrophs, 93Somatotropic axis, 83–84, 92Somatotropin, 289

β3-Adrenergic receptors, 7, 305, 309

TThermogenesis, 2, 17, 311–312, 316, 318,

356–357, 368Threonine, 169, 171, 177, 181Thyroid hormone, 14, 16, 44–45Thyroid hormones, 353, 367Thyroxine, 18Total parenteral nutrition, 168, 172–173, 177Tracee release method, 452Transgenic technique, 435Translation initiation, 84, 95Translation, 26, 28, 49–50, 52, 54–57, 59Transsulfuration, 170, 177, 179, 182Triacylglycerol synthesis, 21, 288tTriads, 360Triglyceride synthesis, 461Triiodothyronine, 18Tumor necrosis factor binding protein, 90, 95Tumor necrosis factor-α, 83–86, 88–91, 93, 95,

97–99TUNEL method, 446Two-dimensional gel electrophoresis, 443Tyramide signal amplification, 439

Index482

UUbiquitin, 51, 84, 96–97, 99UCP, 356–357Uncoupling protein (UCP), 309, 311, 314, 318,

312t, 312fUrea synthesis, 207

VVagus nerve, 89Very low-density lipoprotein, 329, 330, 332, 334,

341–342

VFA metabolism, 392–394Visual function

Docosahexaenoic acid, 15–18, 36–40

WWhite adipose tissue (WAT), 304, 311, 313Whole-body lipolysis, 460Whole-body protein kinetics, 448

Index 483

Biology of Metabolism in Growing Animals

Documents

Transcript of Biology of Metabolism in Growing Animals