DIETARY MODIFICATION AND GENETIC VARIABILITY OF ATHEROSCLEROSIS RISK FACTORS

MAIRERANTALA

Department of Internal Medicine

OULU 2000

DIETARY MODIFICATION AND GENETIC VARIABILITY OF ATHEROSCLEROSIS RISK FACTORS

MAIRE RANTALA

Academic Dissertation to be presented with the assent of the Faculty of Medicine, University of Oulu, for public discussion in Auditorium 10 of the University Hospital of Oulu, on May 31th, 2000, at 10 a.m.

OULUN YLIOPISTO, OULU 2000

Copyright © 2000Oulu University Library, 2000

Manuscript received 4 May 2000Accepted 8 May 2000

Communicated by Professor Markku Laakso Professor Matti Uusitupa

ALSO AVAILABLE IN PRINTED FORMAT

ISBN 951-42-5652-2

ISBN 951-42-5651-4ISSN 0355-3221 (URL: http://herkules.oulu.fi/issn03553221/)

OULU UNIVERSITY LIBRARYOULU 2000

To my late mother,Lahja Laukka

Rantala, Maire, Dietary modification and genetic variability ofatherosclerosis risk factorsDepartment of Internal Medicine and Biocenter Oulu, University of Oulu, FIN-90220Oulu, Finland2000Oulu, Finland(Received 4 May 2000)

Abstract

The risk factors for atherosclerosis and coronary heart disease (CHD) are multiple and mayinteract with each other. Diet has a significant role among the main risk factors foratherosclerosis, as it regulates the levels of plasma lipids and lipoproteins, their oxidativemodification or protection from oxidation, blood pressure, energy balance, andthrombogenesis. Nutrients can transfer their effects directly through plasma concentrationsor modify the cell transduction or gene expression of important regulatory genes. Theresponse to dietary modification varies between individuals. The plasma cholesterolresponse induced by dietary modification is at least partly regulated genetically and someof the variation is explained by other environmental factors.

Apolipoprotein E (apo E) and apolipoprotein B (apo B) are the key regulatory proteinsin cholesterol and lipoprotein metabolism. The genetic variation of apo E is associatedwith the plasma lipid levels and the CHD risk. The polymorphic variation of the apo Bgene is also associated with increased plasma cholesterol and CHD risk. Obesity isassociated with increased morbidity and mortality. Plasma lipid abnormalities, impairedglucose metabolism and increased blood pressure caused by obesity are the main reasonsfor increased CHD mortality among obese subjects.

To study the magnitude of the response to dietary modification, genetically selectedgroups were investigated. Dietary modification had a significant impact on plasma total,LDL, and HDL cholesterol concentrations, and the individual response in plasma LDLcholesterol varied from 3 to 100%. The role of genetic variation in the apo E gene wasnot significant in the lipid response, but the blood pressure response was more distinctamong subjects with the ε4 allele than those with the ε3 allele. The determination of apoB EcoRI and MspI gene polymorphisms revealed subjects with a greater response to diet,a finding which may have clinical importance in the future for the attempt to identifysubjects for effective dietary counselling.

The effect of caloric restriction on gene expression was studied in obese gallstonepatients. Moderate weight reduction during caloric restriction was associated with reducedlipoprotein lipase gene expression, while the cholesteryl ester transfer protein geneexpression remained unchanged. Some of the beneficial changes in plasma lipids andlipoproteins during and after weight reduction may be followed by altered transcription oftheir modifying genes.

Meta-analysis is a modern and generally accepted method. Many clinical uncertaintiescan be solved by combining all the data available to a quantitative and objective analysis.However, the use of meta-analysis do not resolve the problem of the effect of publicationbias.

Keywords: blood pressure, lipoprotein, obesity, polymorphism.

Acknowledgements

This work was carried out at the Department of Internal Medicine and Biocenter Oulu,University of Oulu, Finland.

I wish to express my warmest and most sincere gratitude to Professor and Head of theDepartment, Antero Kesäniemi, M.D., for the opportunity to carry out this work. Hisencouragement and support have been of utmost importance for this study. I am alsograteful for Professor Markku Savolainen, M.D., for his guidance and valuable advice.

I have been privileged to do very close collaboration with Professor Tapio Rantala,PhD, whose experience, optimism and everlasting enthusiasm towards science haveencouraged me throughout these years.

Professor Matti Uusitupa, M.D., and Professor Markku Laakso, M.D., are highlyrespected scientists and experts on nutritional research and atherosclerosis research. Theygave me valuable and constructive criticism in reviewing this thesis.

It is a great pleasure to thank my collaborators Professor Yechiel Friedlander, M.D.,Minna Hannuksela, M.D., Terhi Jokelainen, Leea Järvi, Professor Matti Kairaluoma,M.D., Docent Kari Kervinen, M.D., Jukka Palm, M.D., Marja-Leena Silaste, KaisuSuvanto, and Marjaana Säily, M.D.

Also, I want to express my gratitude to all the skilful technical staff of the researchlaboratory. Especially Saija Kortetjärvi, Marja-Leena Kytökangas, Liisa Mannermaa, SariPyrhönen and Sirpa Rannikko are acknowledged. Their trustworthiness and sense ofhumor have made this work most enjoyable.

Riitta Kaski-Martinviita, Hilkka Karttunen, Jouko Mäkelä and all my friends at thelocal Heart Association of Oulu have given me an opportunity to apply my theoreticalknowledge to practice, and they deserve my most sincere thanks for their support.

Finally, I want to thank my sporty friends and my family for their encouragement.This work was supported financially by the Finnish Foundation for Nutrition

Research, the Sigrid Juselius Foundation, the Farmos Corporation Research Foundation,the Medical Council of the Academy of Finland and the Finnish Heart Association.

Oulu, 23 April 2000Maire Rantala, M.D.

Abbreviations

ADD-1 adipocyte differentiation and determination factor 1ACE angiotensin-converting enzymeapo apolipoproteinBMI body mass indexbp base pairCETP cholesteryl ester transfer proteinC/EBP CCAAT enhancer binding proteinCHD coronary heart diseaseDNA deoxyribonucleic acidE% energy percentFABP2 free fatty acid binding protein 2FFA free fatty acidHDL high-density lipoproteinHL hepatic lipaseHMG-CoA 3-hydroxy-3-methylglutaryl coentzyme-AIDL intermediate-density lipoproteinLCAT lecithin:cholesterol acyltransferaseLDL low-density lipoproteinLPL lipoprotein lipaseLRP LDL receptor related proteinmRNA messenger ribonucleic acidMUFA monounsaturated fatty acidPAI-1 plasminogen activator inhibitor-1PCR polymerase chain reactionPLTP phospholipid transfer proteinPPAR peroxisome proliferator activated receptorPUFA polyunsaturated fatty acidRT–PCR reverse transcription-polymerase chain reactionSAFA saturated fatty acidSD standard deviationSEM standard error of meanSREBP sterol regulatory element binding protein

TNF tumour necrosis factorVLDL very-low-density lipoprotein

List of original papers

This thesis is based on the following articles, which are referred to by their Romannumerals:

I Markku J. Savolainen, Maire Rantala, Kari Kervinen, Leea Järvi, Kaisu Suvanto,Tapio Rantala & Y. Antero Kesäniemi (1991) Magnitude of dietary effects onplasma cholesterol concentration: role of sex and apolipoprotein E phenotype. Atherosclerosis 86: 145–152.

II Maire Rantala, Markku J. Savolainen, Kari Kervinen & Y. Antero Kesäniemi(1997) Apolipoprotein E phenotype and diet-induced alteration in blood pressure.Am J Clin Nutr 65: 543–550.

III Maire Rantala, Tapio T. Rantala, Markku J. Savolainen, Yechiel Friedlander & Y.Antero Kesäniemi (2000) Apolipoprotein B gene polymorphisms and serum lipids:meta-analysis of the role of genetic variation in responsiveness to diet. Am J ClinNutr 71: 713-724.

IV Maire Rantala, Y. Antero Kesäniemi, Matti Kairaluoma, Jukka Palm, MarjaanaSäily, Minna L. Hannuksela & Markku J. Savolainen (2000) Effect ofhypocaloric, low-fat diet on the amount of cholesteryl ester transfer protein andlipoprotein lipase mRNA in obese patients with gallstone disease. Submitted forpublication.

In addition, some unpublished data are presented.

Contents

AbstractAcknowledgementsAbbreviationsList of original papersContents 1. Introduction ............................................................................................ 152. Review of the literature ............................................................................ 17

2.1. Risk factors for atherosclerosis .............................................................. 172.1.1. Plasma lipids and apolipoproteins ..................................................... 17

2.1.1.1. Exogenous and endogenous pathways of lipid transport .................... 172.1.1.2. Plasma lipids and lipoproteins and the risk of atherosclerosis ............ 19

2.1.2. Blood pressure .............................................................................. 202.1.3. Obesity ....................................................................................... 212.1.4. Genetic factors ............................................................................... 212.1.5. Smoking ..................................................................................... 222.1.6. Other risk factors ........................................................................... 24

2.2. Effect of diet on atheroslerosis .............................................................. 242.2.1. Plasma lipids and lipoproteins ......................................................... 252.2.2. Blood pressure .............................................................................. 272.2.3. Obesity ....................................................................................... 282.2.4. Thrombogenesis ............................................................................ 282.2.5. LDL oxidation .............................................................................. 29

2.3. Gene-environment interaction in atherosclerosis ....................................... 292.3.1. Alcohol consumption ..................................................................... 302.3.2. Physical fitness ............................................................................. 312.3.3. Smoking ..................................................................................... 312.3.4. Diet ............................................................................................ 32

2.3.4.1. Mechanisms affecting the individual differences in the response of serum lipids and lipoproteins to changes in dietary cholesterol and fat 322.3.4.2. Genetic variation affecting dietary response in plasma lipids and lipoproteins ............................................................................ 342.3.4.3. Nutritional regulation of gene expression ...................................... 36

2.3.4.4. Obesity and gene expression ....................................................... 373. Aims of the study .................................................................................... 384. Subjects and methods ............................................................................... 39

4.1. Study subjects ................................................................................... 394.2. Diets and study design ......................................................................... 394.3. Clinical measurements ........................................................................ 42

4.3.1. Blood pressure measurements ........................................................... 424.3.2. Body weight measurement ............................................................... 43

4.4. Routine clinical laboratory analyses ....................................................... 434.5. Plasma lipid and lipoprotein analyses ..................................................... 434.6. Determination of apolipoprotein E protein polymorphism........................... 444.7. Determination of plasma CETP activity .................................................. 444.8. Determination of apolipoprotein B DNA polymorphisms ........................... 444.9. Measurement of CETP and LPL mRNA ................................................. 45

4.9.1. Total RNA extraction ..................................................................... 454.9.2. Construction of internal RNA standard .............................................. 45

4.9.3. Reverse transcription ..................................................................... 474.9.2. Polymerase chain reaction ............................................................... 48

4.10. Statistical analyses ............................................................................ 485. Results .................................................................................................. 49

5.1. Diets ................................................................................................ 495.2. Body weight ...................................................................................... 495.3. Plasma lipids and apolipoproteins and the effect of apo E polymorphism ....... 505.4. Blood pressure ................................................................................... 525.5. Plasma CETP activity ......................................................................... 535.6. Liver and adipose tissue CETP mRNA ................................................... 545.7. Adipose tissue LPL mRNA .................................................................. 545.8. Effect of apolipoprotein B DNA polymorphisms on the response to the diet ... 55

6. Discussion ............................................................................................. 576.1. Methodological considerations .............................................................. 57

6.1.1. Study populations ......................................................................... 576.1.2. Dietary interventions, study designs and feasibility of the diets ............... 586.1.3. Quantitative RT-PCR .................................................................... 596.1.4. Meta-analysis ............................................................................... 60

6.2. Dietary effect on lipids and lipoproteins .................................................. 606.3. Role of apolipoprotein E polymorphism in the dietary response .................. 626.4. Role of apolipoprotein B DNA polymorphism on the dietary response ......... 636.5. CETP and the diet .............................................................................. 646.6. LPL gene expression and weight reduction .............................................. 64

7. Conclusions............................................................................................ 668. References .............................................................................................. 68Original articles

1. Introduction

The concept of CHD risk has changed during the past few decades, and the nature of CHDas a multifactorial disease has become clear. It is practical to divide the CHD risk factorsinto three caterogies, namely lifestyle factors, biochemical factors and personalcharacteristics. Personal characteristics, such as age, sex and genes, may play a major rolein the development of CHD, but unfortunately, these factors cannot be changed by thetreatment methods available. The most important biochemical factor is the plasma total(and LDL) cholesterol concentration. After the publication of successful treatment trials ofCHD patients with statins, it is generally recommended to measure the plasma total,HDL and LDL cholesterol and triglycerides in the case of people at high risk to developCHD. Also, measurements of blood pressure and fasting blood glucose, determination ofeventual obesity and even measurements of some thrombogenic factors, at least in youngpatients, help to identify the risk factors that can be modified by modern treatments.

Lifestyle factors have an important role in the CHD risk both at the population leveland at the individual level. The lifestyle factors include a diet high in saturated fat,cholesterol and energy, tobacco smoking, excess alcohol consumption and physicalinactivity. These are factors which, on the one hand, could lower the CHD risksignificantly after modification, but, on the other hand, are the most difficult to modify incommunity-living adult populations. Every smoker who quitts smoking, every physicallyactive individual and every person who limits his/her alcohol consumption benefits inmany ways. The importance of diet is more complicated. The harmful effect of anunhealthy diet probably varies between individuals, as does the effectiveness of a healthydiet, and we lack a good method to identify the responders and non-responders.

In hypercholesterolemia, dietary counselling is the first-line treatment. In slight andmoderate hypertension, smoking cessation and dietary counselling are the first-linetreatments, dietary counselling mainly to achieve weight control and to restrict the intakeof dietary sodium and fat. To rationalize the treatment methods, the significance ofdifferent risk factors should be identified and quantified, the subjects with a particular riskshould be selected, and the responders and non-responders to treatments should beidentified. After identifying a responder, both the patient and the therapist are moremotivated to continue the selected treatment. After identifying a non-responder, othertreatments, such as drug treatment should be started.

16

The complex interaction between diet and genetic factors in lipid metabolism can bestudied in a relatively limited study population, assuming that the genetic factor regulatessome measurable indicator of the metabolic cascade. The present study was conducted tomeasure the magnitude of the effect of a dietary modification on plasma lipid levels and toinvestigate the role of genetic variation in some major proteins of lipid metabolism inresponse to a dietary modification. In addition, the role of the body weight reductionfollowing dietary counselling in altering the expression of lipid-modifying genes wasstudied.

2. Review of the literature

2.1. Risk factors for atherosclerosis

2.1.1. Plasma lipids and apolipoproteins

2.1.1.1. Exogenous and endogenous pathways of lipid transport

Lipids are transported in the circulation by lipoproteins, macromolecular complexes thatconsist of lipids (unesterified and esterified cholesterol, triglycerides, and phospholipids)and proteins called apolipoproteins. Apolipoproteins serve a variety of physiologicalfunctions in lipoprotein metabolism, including cofactors for enzymes, ligands for cell-surface receptors, and structural proteins for lipoproteins. The lipoproteins are separatedon the basis of their hydrated densities and their apolipoprotein composition. The majordensity classes of lipoprotein particles include chylomicrons (density < 0.94 g/ml), very-low-density lipoproteins (VLDL, < 1.006 g/ml), intermediate-density lipoproteins (IDL,1.006-1.019 g/ml), low-density lipoproteins (LDL, 1.019–1.063 g/ml), and high-densitylipoproteins (HDL, 1.063-1.21 g/ml).

Chylomicrons are triglyceride-rich intestinal lipoproteins that contain a form ofapolipoprotein B called apoB-48, which transports dietary lipids to peripheral tissues andthe liver (Fig. 1). The triglycerides in chylomicrons are hydrolysed by the endothelialenzyme, lipoprotein lipase (LPL), which requires apo C-II as a cofactor (1). Chylomicronremnants are removed from the circulation by the liver via a process that involves thebinding of apo E to a putative hepatic remnant (apo E) receptor. Apo E is synthesized inmany different tissues, but the liver is the predominant source of plasma apo E. Apo Econtains a binding site complementary to both remnant and LDL receptors and isresponsible for the uptake of chylomicron and VLDL remnants. It may also play a role inreverse cholesterol transport in plasma and intercellular lipid transport within tissues.

18

Fig. 1. Exogenous pathway of lipid transport. SER denotes smooth endo-plasmic reticulum. RER denotes rough endoplasmic reticulum. Modif iedfrom Schumaker et al. 1992 (2).

VLDL particles are triglyceride-rich lipoproteins secreted by the liver and containing aform of apoB called apoB-100 (Fig. 2). Human plasma apo B-100 is synthesized in theliver, where it is essential for the formation and secretion of VLDL. Firmly attached tothe VLDL, a single apo B-100 remains with each VLDL and LDL during its existence. Itcontains the binding site responsible for the uptake of LDL by the LDL receptor. VLDLtriglycerides are also hydrolyzed by LPL; some VLDL remnants (or IDL) are removedfrom the circulation by the liver via the apoE-mediated process, but others are furtherprocessed by hepatic lipase (3) with conversion to LDL.

LDL transports cholesteryl ester to a variety of peripheral tissues, but a significantamount of plasma LDL is eventually removed from the circulation by the liver via thebinding of apo B-100 to the hepatic LDL receptor (4). The plasma concentration of LDLdepends on the rates of VLDL secretion and the conversion of VLDL to LDL (5) and thefractional clearance rate of LDL, and it is influenced by heritable factors and someenvironmental factors, the most important being the diet. LDL can undergo an oxidativemodification (6), which may have a significant role in the process of atherosclerosis (7,8).

GOLGI

RER(Apo B-48, Apo A)

SER(TG)

LYMPHCHYLOMICRON(Apo B-48, Apo A-I, Apo A-IV)

REMNANT(Apo B-48, Apo E, Apo C,)

LYSOSOME

REMNANTRECEPTOR

IntestinalMucosal Cell

FFAMonoglycerides

Intestinal Lumen

(Apo E, Apo C)

LPLPLASMACHYLOMICRON

FFA

FFA ALBUMIN

Liver Cell

CapillaryEndothelium

HDL(Apo A, Apo E,

Apo C)

(Apo A-IV)

(Apo A-I)

19

Fig. 2. Endogenous pathway of lipid transport. SER and RER are the s a m eas in Fig. 1. Modified from Schumaker et al. 1992 (2).

HDL particles are synthesized and secreted directly as nascent discoidal particles byboth the intestine and the liver (9, 10). Nascent HDL contains a high portion ofunesterified cholesterol and phospholipids, which serve as substrates for the plasmaenzyme lecithin:cholesterol acyltransferase (LCAT), resulting in the formation ofcholesteryl ester (11). HDL cholesteryl esters and phospholipids can be transferred toapoB-containing lipoproteins by the cholesteryl ester transfer protein (CETP) (12) and thephospholipid transfer protein (PLTP) (13). CETP, PLTP and hepatic lipase act in concertto convert larger HDL into smaller particles (10). HDL particles undergo continuousremodelling in the circulation through the action of LCAT, CETP, PLTP, hepatic lipaseand LPL.

2.1.1.2. Plasma lipids and lipoproteins and the risk of atherosclerosis

A high concentration of plasma total cholesterol is a major risk factor for coronary arterydisease (14-16). This risk is mediated through the major cholesterol-carrying lipoprotein,LDL, which is considered the major atherogenic lipoprotein. The evidence supporting thehypothesis that LDL is atherogenic comes from epidemiologic studies (17, 18), clinicaltrials (19-21), studies in laboratory animals (22), heritable hypercholesterolemias (23),

GOLGIRER(Apo B48, Apo A)

SER(TG)

LYSOSOME

FFA

Apo E

FFAALBUMIN

Liver Cell

CapillaryEndothelium

LYSOSOME

B / EReceptor

B / EReceptor

Peripheral Cell

LDL(Apo B)

HDL

REMNANT(Apo B, Apo E, Apo C)

IDL(Apo B, Apo E)

Apo C, Apo ELPL

VLDL(Apo B,Apo E,Apo C)

Apo EApo C

VLDL(Apo B, Apo E,

Apo C) HDL(Apo A,Apo E,Apo C)

20

pathologic investigations (24), and studies in model systems (25). Several theories existconcerning the mechanisms by which LDL produces atherosclerosis. The concentration,the size and the chemical modification of LDL are important for atherogenesis. Clinicaltrials have shown that a reduction of total and LDL cholesterol is followed by a regressionof atherosclerotic manifestations (26). Based on these results, the target plasma totalcholesterol concentration is < 5.0 mmol/l and the target LDL cholesterol concentration <3.0 mmol/l (27).

HDL particles are the smallest lipoproteins and therefore enter and also leave the arterywall easily. The plasma concentrations of HDL cholesterol are inversely associated withthe risk of CHD (28). The plasma concentrations of apo A-I, the structural protein ofHDL, correlate strongly with HDL cholesterol levels and are also inversely associatedwith the CHD risk. HDL cholesterol and apoA-I are anti-atherogenic, as shown by animalstudies and genetic human studies, and the effect has been explained, at least partly, by thereverse cholesterol transport (10, 29, 30). In this mechanism, free cholesterol istransferred from peripheral cells to an acceptor HDL subpopulation. The other potentialanti-atherogenic mechanisms of HDL and apoA-I include the protection of LDL fromoxidation (for a review, see Banka, 1996) (31), the protection of endothelial cells from thecytotoxic effect of LDL, and the stimulation and stabilization of the vasodilatorprostacyclin (for a review, see Barter and Rye, 1996) (29). The goal for plasma HDLcholesterol concentration is > 1.0 mmol/l (27).

Although hypertriglyceridemia has been statistically associated with CHD, theindependent association of plasma triglycerides with the CHD risk is less certain than thatwith LDL, and triglycerides may not be causally related to the development ofatherosclerosis (32). Hypertriglyceridemia may be a secondary phenomenon that occurs inresponse to the metabolism of LDL and HDL.

One additional class of lipoproteins, namely lp(a) concentration, correlate with theCHD risk and may be directly involved in the atherogenic process and, furthermore,interfere with fibrinolysis (33). Lp(a) is largely resistant to modifications, only dietarytrans fatty acids and possibly monoenes and alcohol consumption may alter lp(a)concentrations (34, 35). Nevertheless, lp(a) can be used to identify subjects withincreased risk of CHD.

ApoB is the major protein component of LDL, IDL, VLDL and chylomicrons, and allapoB-containing lipoproteins are atherogenic. Since there is one apoB molecule per eachlipoprotein particle, the apoB concentration is a good indicator of the risk ofatherosclerosis and CHD (36).

2.1.2. Blood pressure

Elevated blood pressure is frequently associated with the other CHD risk factors, such assmoking, diabetes, insulin resistance, dyslipidemias and obesity. According toepidemiological studies and after adjustment for confounding factors high blood pressurealone is an important risk factor for CHD (37). The treatment of hypertension reduces therisk of CHD in clinical drug intervention trials (38), and lifestyle interventions for mildlyelevated blood pressure have also been effective in risk reduction. Blood pressure is

21

controlled by both genetic and environmental factors. The mechanism wherebyhypertension predisposes to cardiovascular disease is probably related to the fact that highblood pressure accelerates the atherosclerotic process as well as to the pressure effectsleading to progressive dilation and rupture of large and small blood vessels. Arecommended level of blood pressure is less than 90 mmHg for diastolic and less than 140mmHg for systolic blood pressure (27).

2.1.3. Obesity

A body mass index (BMI, kg/m2) over 25 is considered overweight, and BMI over 30 isdefined as obesity. Obesity has been associated with excess mortality (39), andprospective studies of cardiovascular morbidity and mortality have shown an associationwith obesity (40, 41). The risk already begins to increase at a moderate level of obesity.Obesity has an adverse influence on blood pressure, plasma lipids and lipoproteins, andglucose tolerance, and further has adverse hemodynamic effects (42). In obese subjects, theexcess lipolysis and release of fatty acids from adipose tissue is followed by increasedVLDL production (5) and secretion from the liver (43) and high plasma triglycerideconcentrations. The activities of plasma and adipose tissue LPL are high in obese subjects(44-46). The HDL cholesterol concentration is usually low in obesity (43). Themechanism for this may be partly regulated by elevated CETP activity (47). Since obesesubjects have supersaturation of cholesterol in bile and increased cholesterol excretion and,furthermore, hypomotility of the gallbladder, cholesterol gallstone formation is enhanced,especially in obese women (48). Weight reduction by either caloric restriction or increasedenergy expenditure is followed by reductions of blood pressure, plasma triglycerideconcentration, increase of plasma HDL cholesterol and normalization of blood glucose(49). During weight reduction, the plasma HDL cholesterol concentration increases onlyafter a stable reduced weight have been reached (50).

2.1.4. Genetic factors

The pathophysiology of atherosclerosis and CHD is characterised by a mixture of chronicprocesses and acute events. The most important pathogenic processes, which aredetermined partly genetically and partly environmentally, are dyslipidemia, hypertension,endothelial dysfunction, diabetes, and cardiac and vascular hypertrophy. Thus, thevariations of many genes may contribute to the disease process, and depending on thecombinations of genetic variations in different subjects, atherosclerosis and CHD may bemanifested as a wide spectrum of phenotypes. Several genes have been investigated inrelation to apolipoproteins, (apo B, apo C-III, apo(a), apo E), lipid transfer proteins(PLTP, CETP), enzymes (LPL, HL, LCAT), receptors (LDL receptor, LDL receptorrelated protein), thrombogenic factors (fibrinogen, PAI-1, glycoprotein IIIa), and others(ACE, angiotensin II-receptor, paraoxonase, methylene tetrahydrofolate reductase,neuropeptide Y, 7-α-hydroxylase), and their respective regulatory genes are also involved.

22

The polymorphism of neuropeptide Y seems to be an important regulator of the plasmaLDL concentration (51). Genetic factors have an influence on the age of onset and theseverity of the disease and also on the response to treatment, including the response ofplasma lipids to dietary treatment.

It has been postulated that common polymorphisms with frequent alleles probablyaccount for most of the genetic component of atherosclerosis and CHD (27), and raremonogenic disorders associated with a high absolute risk for CHD, such as familialhypercholesterolemia, are only observed in a small proportion of the patients with CHD.The rare genetic disorders associated with myocardial infarction or atherosclerosis at youngage, such as familial hypercholesterolemia and familial defective apo B-100, are relativelywell documented (52).

Apo B, which is an essential lipoprotein for lipid metabolism and also for theabsorption of dietary fats and fat-soluble vitamins have many common polymorphicvariations, some of them associating with increased plasma total and LDL cholesterolconcentrations and CHD (53, 54).

Apo E polymorphism is an example of a common polymorphism with three majorisoforms: apo E2, apo E3 and apo E4. Apo E2 differs from E3 by a cysteine for argininesubstitution at amino acid residue 158, whereas apo E4 differs from E3 by an arginine forcysteine substitution at residue 112 (55). A linear correlation has been shown in manypopulations between the presence of E4 allele and the plasma cholesterol level. Despiteits relatively weak effect at the individual level, the apo E polymorphism may explain 5-8% of the attributable risk of CHD in the population (56-59). Other possible genesknown to associate with atherosclerosis and CHD risk through altered lipid metabolismare presented in Table 1.

The simultaneous effects of different genetic variants on lipoprotein metabolism andCHD risk are difficult to estimate at present, because no observational studies ofsufficient scope have been reported. The synergistic effect of different gene variants ontreatment response is also unclear.

2.1.5. Smoking

Smoking is one of the major risk factors for CHD (60, 61). The harmful effect ismediated through altered lipid and lipoprotein metabolism (62) as well as through theinduction of vasoactive, thrombogenic and other atherogenic mechanisms (63-66). Also,smokers exhibit several characteristics of insulin resistance syndrome and the impact ofsmoking on CHD risk is modified by plasma lipid levels (for review, see 27). Stoppingsmoking leads to a considerable risk reduction amon CHD patients.

23

Table 1. Human genes known to associate with atherosclerosis and coronary heart diseasethrough altered lipid metabolism.

Gene Chromosome Majorplace ofsynthesis

ipoproteinparticle

Function

Apo A-I 11 Intestine,liver

HDL Structural protein of HDL,activator of LCAT, role in reversecholesterol transport

Apo A-II 1 Intestine,liver

HDL Structural protein of HDL,activator of LPL, modulator ofLCAT

Apo A-IV 11 Intestine ? Dietary fat absorption, activator ofLPL, modulator of LCAT, reversecholesterol transport

Apo(a) 6 Liver Lp(a) ?

Apo B 2 Liver LDL, VLDL Structural protein of LDL andVLDL, LDL receptor ligand

Apo C-I 19 Liver Inhibitor of LDL receptor andLRP, activator of LCAT

Apo C-II 19 Liver Activator of LPL

Apo C-III 11 Liver,intestine

Chylomicrons,VLDL, HDL

Inhibitor of LPL, apo E mediatedremnant removal

Apo E 19 Liver,intestine

Chylomicrons,VLDL, HDL

LDL receptor, LRP ligand, reversecholesterol transport,immunoregulator, cell growthregulator

CETP 16 Liver,adiposetissue

HDL Exhange of CE betweenlipoproteins, reverse cholesteroltransport

LPL 8 Adiposetissue

Chylomicrons,Triglycerides

Hydrolysis of triglycerides

LDL-receptor

19 Liver Binds apo B-100, apo E,determines plasma LDL level

HL 15 Liver HDL Hydrolysis of HDL2 triglyceridesand phospholipids

24

2.1.6. Other risk factors

The intake of light to moderate amounts of alcohol is associated with reducedmorbidity and mortality from several cardiovascular conditions, particularly CHD (67,68). The beneficial effects of light to moderate alcohol drinking on lipoproteinmetabolism (69, 70), coagulation (71, 72) and antioxidative properties (73, 74) have beenshown in several studies. Since the adverse effects of excess alcohol use are so wellknown, alcohol should not be recommended as a general preventive tool against CHD.

Sedentary lifestyle is associated with an increased risk of death from cardiovasculardisease and an increased CHD risk (75-79), and conversely, a high level of physicalactivity is followed by a reduction of the CHD risk and favourable effects on overweightand plasma lipids and lipoproteins.

2.2. Effect of diet on atherosclerosis

The effect of diet on the development of atherosclerosis and CHD is mediated mainly bythe influence of dietary fat and cholesterol on biological risk factors, such as the plasmaconcentrations of total, LDL and HDL cholesterol, triglycerides, LDL oxidation, bloodpressure, glucose metabolism, some thrombogenic factors, and obesity. The amount andcomposition of the diet regulate the activities of many enzymes of lipid metabolism. Theeffect of diet is partly mediated by the plasma levels of specific dietary components, suchas fuels, hormones or metals. Nutritional regulation mediated intracellularly involves themetabolism of a plasma fuel to an active metabolite or the activation of a hormonallyregulated intracellular signaling pathway or nutrient-induced alteration of gene expression.

It was 90 years ago that experimental atherosclerosis was produced for the first time inrabbits by simply feeding them cholesterol (80). Later, the original observations made inanimals have been confirmed in humans (81), i.e. an increase of dietary cholesterol resultsin higher plasma cholesterol concentrations and an increased risk of atheroscleroticdisease. Several epidemiologic and clinical trials have confirmed the causality between dietand atherosclerosis. Los Angeles Veterans Administration Trial (82), Finnish MentalHospital Study (83), Oslo Trial (84), North Karelia Project (85), and Multiple RiskFactor Intervention Trial (86, 87) are good examples of primary prevention for CHD.These studies were conducted on healthy subjects with dietary intervention, counsellingon smoking cessation and blood pressure control and a long enough follow-up time. Asignificant reduction of CHD morbidity and mortality was achieved in most of thesestudies, and the most marked risk reduction followed a diet-induced alteration in plasmalipids and lipoproteins. Also, a significant risk reduction with dietary interventionhasbeen achieved in several secondary prevention trials for CHD (88-93).

25

2.2.1. Plasma lipids and lipoproteins

Because increased dietary cholesterol leads to increased plasma cholesterol and an increasedrisk for atherosclerosis, the amount of dietary cholesterol should be restricted to minimizethe risk, and the current recommendations restrict the intake of dietary cholesterol to 300mg/d for the general public and to <200 mg/d for subjects still hypercholesterolemic on aless stringent restriction (27). On the other hand, large doses of plant sterols (sitosterol,campesterol) produce a hypocholesterolemic effect when given at or before mealtime.These sterols inhibit the intestinal absorption of cholesterol (94).

The dietary total fat intake and, even more importantly, the fatty acid composition ofdietary fat alter the plasma lipid and lipoprotein concentrations in most individuals. Theeffects of different dietary fatty acids on plasma lipids and lipoproteins are related to theirchemical structure, i.e. the presence, absence or location of the double bonds (Table 2).Dietary saturated long-chain fatty acids (chain lengths of 14-16 carbon atoms) decreaseLDL receptor activity, impair the rate of removal of LDL particles from circulation, andraise the plasma total, LDL and, to some extent, HDL cholesterol concentrations (95,96). Relatively short-chain fatty acids (6-12 carbons) and stearic acid (18 carbons) producelittle or no change in the plasma cholesterol concentration (95-98).

The current recommendations limit the saturated fat intake initially to < 10 %E, and ifthe subjects still have elevated LDL cholesterol concentrations, to < 7 %E (27).

MUFAs were originally considered to have a neutral effect on the plasma cholesterolconcentration (99, 100). More recently, it has been shown that the replacement of SAFAwith MUFA results in a hypocholesterolemic effect while minimizing the decrease in theHDL cholesterol concentration (98, 101). In addition, a diet rich in MUFAs comparedwith PUFAs reduces the susceptibility to oxidative modification of plasma LDL. Thecurrent recommendations suggest that MUFA intake should be <15% of the total energyintake of the diet (27).

Trans-fatty acids are formed when vegetable oils are hydrogenated to convert themfrom a liquid to a semisolid state, and trans-fatty acids are also found in ruminant fats.Trans-fatty acids raise the LDL and lower the HDL cholesterol concentration, and somestudies have also reported an increase of the lp(a) concentration (34, 102).

Dietary PUFAs can be divided into n-3 and n-6 polyunsaturated fatty acids. Theeicosapentaenoic acid and other long-chain n-3 fatty acids found in fish and fish oils mayhave a hypocholesterolemic effect in normal subjects or raise the LDL and HDLcholesterol concentration in hyperlipidemic subjects, and have hypotriglyceridemic effectsin both normal and hyperlipidemic subjects (103, 104). On the other hand, several studieshave shown small increases in lipoprotein oxidation or the plasma levels of lipidperoxides after fish oil supplementation (105). Linoleic acid and other n-6 polyunsaturatedfatty acids are mostly derived from plants. Relative to saturated fatty acids or carbohydrate,these fatty acids have a hypocholesterolemic effect (96-98). Large amounts of dietarypolyunsaturated fatty acids lower the plasma HDL cholesterol concentration and may,furthermore, cause excess oxidability of lipoproteins and potential excess atherogenesis(106). The adequate intake of linoleic acid is probably 5 %E, and 3 %E is probablyadequate to meet the linoleic acid requirements (27).

26

Table 2. Dietary fatty acids, their chemical structure and their effect on plasma lipids andlipoproteins

Effects onplasma lipids

Commonname

Formula ω Sources LDL HDL Tg Lp(a)

SAFAShortchain

Acetic C2:0 - - - -

Butyric C4:0 - - - -

Mediumchain

Caproic C6:0 coconut oil, butter - - - -

Caprylic C8:0 - - - -Capric C10:0 - - - -Lauric C12:0 lamb fat ↑ ↑ - -

Longchain

Myristic C14:0 coconut and palmoil

↑ ↑ - -

Palmitic C16:0 Palm oil, milk,butter, cheese, lard(pig)

↑ ↑ ↑ -

Stearic C18:0 meat, sausages - - ↑? -

Very longchain

Arachidic C20:0 ? ? ? -

Behenic C22:0 ↑ ? ? -Lignoceric C24:0 ? ? ? -

MUFA Palmitoleic C16:1 ↑ - ? ?Oleic C18:1 9 olive oil, lard

(pig),↓ - ↓ -

rapeseed, canola oil,milk

TRANS Elaidic C18:1 9T hydrogenated fats ↑ ↓ ? ↑

PUFA Linoleic C18:2 6 sunflower oil, nut ↓ ↓ ↓ -Arachidonic C20:2 6 ? ? ? -α-linolenic C18:3 3 vegetables,

soybean oil,↓ ↓ ↓ -

γ-linolenic C18:3 6 linseed oil ↓ ↓ ? -Eicosa-pentaenic

C20:5 3 fish ? - ↓ -

Docosa-hexaenic

C22:6 3 fish ? - ↓ -

27

Supplementation of dietary fiber may lower the plasma total and LDL cholesterolconcentrations (107). Especially, water-soluble noncellulose pectin, gums, mucilages andlignin may have specific lipid-lowering effects by increasing the excretion of bile salts.Digestible dietary carbohydrates have only a small direct effect on the plasma lipid andlipoprotein concentrations. Still, the substitution of carbohydrates for saturated fatty acidswill cause a reduction in plasma LDL cholesterol. Quickly absorbed carbohydrates mayincrease the triglyceride concentration by stimulating the production of VLDLtriglycerides, especially in subjects with hypertriglyceridemia, obesity and sedentarylifestyle, and HDL cholesterol concentration may also be decreased. Some types of animalprotein (e.g. casein or meat) may be hypercholesterolemic, and some kinds of vegetableprotein (e.g. soy) may lower plasma cholesterol.

Obesity is associated with a decreased HDL cholesterol level (43). Excess caloriesincrease VLDL production and increase VLDL removal (5), resulting in elevatedtriglyceride levels (43). The restriction of calories in obese subjects will normalize theserum lipids levels. The serum triglycerides are reduced from the beginning of weightreduction, and as soon as the post-weight-reduction stable phase is achieved, the HDLcholesterol concentration increases (50).

The consumption of alcohol increases VLDL production by the liver. Chylomicronsynthesis increases when alcohol is consumed with a fatty meal. These effects of alcoholmay induce hypertriglyceridemia. Alcohol use may increase the plasma HDL cholesterolconcentration through the reduced activity of CETP (108).

2.2.2. Blood pressure

Evidence is accumulating to indicate that diet is an important determinant of bloodpressure. Out of all the known dietary components, sodium intake may be the mostimportant blood pressure regulator. The prevalence of hypertension is very low inpopulations with very low dietary sodium intake (109). A low-salt diet may lower bloodpressure and prevent the increase of blood pressure with age (110, 111).

Blood pressure may be influenced by many other dietary components. Someepidemiological and interventional studies suggest an inverse correlation between bloodpressure and potassium intake (109, 112). Also, dietary calcium and blood pressure areinversely correlated (113, 114). Obesity-related hypertension is common, the incidence ofhypertension in obese subjects being about 2-fold compared to subjects with normalweight (115). A beneficial impact of weight reduction by caloric restriction has clearlybeen documented in short-term trials (116, 117). Heavy consumption of alcohol isassociated with hypertension (118, 119).

The effect of the quantity and quality of dietary fat on blood pressure is somewhatcontroversial. A positive association between the estimated intake of SAFAs and bloodpressure has been reported in some epidemiological studies (120, 121). The dataconcerning the role of PUFAs of the n-6 series or MUFAs in blood pressure regulationhave given contradictory results. Relatively high supplemental doses of n-3 PUFAs ofmarine origin are associated with blood pressure reduction (120, 122). Anyway, thereduction in plasma cholesterol concentration achieved by dietary intervention is

28

associated with lowered blood pressure and even with a reduction in the incidence ofhypertension (123). A diet rich in fruits and vegetables, and low-fat dairy products notonly lowers the plasma LDL cholesterol concentration, but may also favourably influenceblood pressure values. The regulatory role of dietary fiber on blood pressure iscontroversial, though some studies have shown that an increased fibre intake may decreaseblood pressure slightly (124–126).

With appropriate dietary modifications, it may be possible to treat hypertensivepatients with fewer drugs or at least with lower doses (127).

2.2.3. Obesity

The growing prevalence rates of obesity and overweight are attributable to the interactionbetween diet, lifestyle and genetics (128). In western societies, spontaneous and work-related physical activity has decreased, and the possibility to overconsume high-fat andhigh-energy food has simultaneously increased. Obese subjects have a high energy intakecompared to their need for energy. Even though dietary fat does not promote thedevelopment of obesity any more than do other macronutrients under iso-energeticconditions (129), some studies indicate that obesity is more strongly related to high-fatdiets than to the total energy intake (130-132). These results have been explained by thefact that dietary fat is metabolized to body fat at an efficiency rate that is higher than thatfor an equicaloric amount of carbohydrate or protein (133-136). The obesity-promotingeffect of a high-fat diet has been especially pronounced in subjects with a geneticpredisposition to obesity (137, 138) or in subjects with a sedentary lifestyle (139). Thereare specific properties of dietary fat that affect the subject’s decision on food intake andpromote excessive energy consumption and storage. Dietary fat may have a weaker effecton satiety and satiation compared to carbohydrate or protein, which may contribute tooverconsumption of fat (140, 141). High-fat foods are generally considered palatable, andespecially obese individuals have an enhanced sensory preference for high-fat foods (142-144). High-fat food also usually requires less chewing and is therefore eaten more quicklythan foods rich in fibre and carbohydrate. The results of studies conducted withmonozygotic twins have shown that human subjects have different genetic susceptibilitiesto environmental changes, which lead to variability in the degree of positive energybalance and weight gain (145).

2.2.4. Thrombogenesis

There is some evidence to suggest that certain dietary lipids and fatty acids (146-148) anddietary carbohydrates (149) may be antithrombotic. Dietary fat high in trans-fatty acidsincreases the concentration of lp(a), which is associated with both atherogenesis andthrombogenesis. Dietary cholesterol is thrombogenic (150). However, the interplaybetween lipids, atherogenesis, and thrombosis has proven difficult to define in humans.Nevertheless, a diet containing additional nutrients, such as alcohol and folic acid as

29

antithrombogenic factors, natural antioxidants and, furthermore, an appropriate balancebetween fatty acids, may have a significant effect on reducing CHD and acuteatherothrombotic events.

2.2.5. LDL oxidation

Substantial evidence suggests that oxidative modification of LDL critically contributes tothe pathogenesis and progression of human atherosclerosis (27). There is some evidenceto suggest that replacement of dietary SAFAs with PUFAs leads to LDL particlesenriched in PUFA, which are then more susceptible to lipid peroxidation and thus moreatherogenic. On the other hand, diets enriched in MUFA, rather than PUFA, might conferadditional protection by generating LDL particles that are more resistant to oxidationwhile optimizing both LDL and HDL cholesterol levels (151-157). The effect of dietsenriched in long-chain PUFAs, such as eicosapentanoic acid and docosahexaenoic acid,available from fish oil supplementation leads to increased susceptibility to oxidation ofLDL (158-161). Anyhow, the proatherosclerotic effect of PUFA diets through LDLoxidation is cancelled by the lipid-lowering effect of these fatty acids.

Some observational population studies indicate that a high intake of antioxidants(vitamin E, β carotene, vitamin C) may associate with a decrease in CHD risk. However,pharmacological intervention with vitamin E and β carotene failed to decrease CHD, andsome possible adverse effects of β carotene supplementation must be considered,especially among smokers (162, 163). Flavonoids, which are secondary metabolites ofplants, are also powerful antioxidants of LDL cholesterol. Pro-oxidants, such as iron andcopper, may be atherogenic.

2.3. Gene-environment interaction in atherosclerosis

The environmental factors generally connected with atheroslerosis are diet, alcoholconsumption, physical activity and smoking. The genetic basis of lipid and lipoproteinmetabolism and its role in atherosclerosis is relatively well documented. Environmentaland genetic factors work together to shape an individual’s risk for atherosclerosis.Research on the genetic and molecular basis of the gene-environment interaction hastherefore emerged as a major area of investigation. The possible models of gene-environment interactions are presented in Fig. 3 . Diet is the best studied environmentalfactor that affects the lipoprotein phenotype.

30

Model 1 Genotype G ⇒ Environmental stimulus E ⇒ Phenotypic response GEGenotype G' ⇒ Environmental stimulus E ⇒ Phenotypic response G'E

Hypothesis:Apo E4/4 ⇒ Cholesterol-rich diet ⇒ P-LDL ↑↑Apo E3/3 ⇒ Cholesterol-rich diet ⇒ P-LDL ↑

Model 2 Genotype G ⇒ Environmental stimulus E ⇒ Phenotypic response GEGenotype G ⇒ Environmental stimulus E' ⇒ Phenotypic response GE'

Hypothesis:CETP TaqI

B2B2⇒ Non-smoking ⇒ P-HDL B2B2 > B1B1

CETP TaqIB2B2

⇒ Smoking ⇒ P-HDL B2B2 = B1B1

Fig. 3. Two possible models of gene-environmental interaction

2.3.1. Alcohol consumption

The results of observational studies provide evidence that moderate alcohol intake lowersthe CHD risk and that alcohol intake may be causally related to a lower risk of CHDthrough changes in lipids and haemostatic factors (164). There may be some differencesdue to various lifestyle characteristics with relation to the amount of alcohol used, e.g.the percentage of smokers or the percentage of physically inactive subjects increasessimultaneously with the amount of alcohol used (165). There is some evidence ofinteraction between alcohol drinking and a genotype effect on plasma lipids andlipoproteins. The I405V polymorphism of the CETP gene is associated with the plasmaHDL, apo A-I levels and the CHD risk, and this genotype effect is modified by alcoholconsumption (166). The CETP TaqI polymorphism seems to modify the plasma CETPlevel and activity and the plasma HDL cholesterol concentration. This genotype effectinteracts with alcohol consumption, as shown in two separate studies (167, 168). Bloodpressure is not significantly affected by light to moderate alcohol intake (169). Theprevalence of hypertension and the risk of haemorrhagic stroke increase with increasingalcohol consumption (170, 171). In a Finnish study, a significant association was foundbetween alcohol consumption and both systolic and diastolic blood pressure, and the apoE phenotype affected the blood-pressure-increasing effect of alcohol (119).

31

2.3.2. Physical fitness

Being physically active helps to reduce weight, increase HDL cholesterol, and lowertriglycerides and the propensity to thrombosis. The role of physical activity in the gene-environment interaction has been only studied in a few trials. LPL is the rate-limitingenzyme in the hydrolysis of the core triglycerides in chylomicrons and VLDL. Thevariation D9N of the LPL gene is associated with increased total cholesterol, apoB andtriglycerides concentration. Physically inactive D9N carriers seem to have higher lipidlevels than physically active ones (172). The G-455A β fibrinogen gene promoterpolymorphism is associated with the plasma fibrinogen levels, and in the determinationof the magnitude of this effect, a significant interaction between the G-455Apolymorphism and physical activity has been demonstrated (173).

2.3.3. Smoking

Smoking is often associated with other atherogenic environmental factors, such assedentary lifestyle, unfavorable nutritional patterns (174) and alcohol drinking, which maysynergistically impair the pathophysiological responses (175,176). The relations betweensome genetic polymorphisms and HDL levels have been shown to be modified bysmoking. Smoking habits modify the association between the CETP TaqI polymorphismand the variation in HDL (177-179) and CETP levels (168). Another polymorphism ofpotential significance is in the apo A-I promoter region, which is associated with HDLlevels in male non-smokers, but not in male smokers or in females (180, 166). Theinteraction could not be confirmed using meta-analysis (181). Also, some interactionbetween smoking, HDL cholesterol and the apoB EcoRI and MspI polymorphisms havebeen shown (182).

The PI(A1/A2) platelet glycoprotein IIIa polymorphism is associated with triglyceridemetabolism and an increased CHD risk, and cigarette smoking has some interaction withthis (183, 184). The serotonin transporter (5-HTT) gene promoter polymorphism isassociated with plasma serotonin levels and vasoconstriction, and there seems to be asignificant interaction between the 5-HTT gene promoter polymorphism and smoking indetermining the risk for CHD (185). Also, a gene-environment interaction has beendetected between the G-455A β fibrinogen gene promoter polymorphism and smoking,the carriers of the A-allele having, on an average, 7-10% higher fibrinogen levels thanthose with the genotype GG, and this effect of the A-allele was connected with thesmoking habits (173). However, no interaction between the D9N mutation in the LPLgene and smoking was found (172).

32

2.3.4. Diet

2.3.4.1. Mechanisms affecting the individual differences in the response ofserum lipids and lipoproteins to changes in dietary cholesterol and fat

The levels of serum lipids and lipoproteins are sensitive to the type and amount of fat andthe amount of cholesterol in the diet. The quantitative effects of dietary changes can beestimated using empirical formulas (186, 187, 104). The individual response variesbetween subjects, and this variation is constant to a certain limit, the standardizedregression coefficients for individual responses in similar experiments being from 0.30 to0.50 (188, 189). A low-fat, low-cholesterol diet results in an average of 10% lowering ofthe plasma cholesterol concentration, while some subjects may respond with a 30%reduction and some may be relatively insensitive to the dietary effect. Studies on inbredstrains of rabbits, mice, rats, and pigeons indicate that the differences in responsivenesshave a genetic basis. The extent of the genetic influence in several animal species isestimated to be about 50-80% (190) but the evidence in humans is less clear.

The metabolic pathways connected with dietary responsiveness are shown in Fig. 4.Cholesterol homeostasis is maintained by balancing the dietary cholesterol absorption andthe endogenous cholesterol synthesis with the excretion of bile acids and biliarycholesterol (191). The differences in individual responses to diet may be followed bydifferences in intestinal absorption efficiency, hepatic cholesterol biosynthesis, excretionof cholesterol, receptor-mediated clearance of LDL, or differences in LDL production(190).

In some animal strains, increased absorption seems to be the cause ofhyperresponsiveness to dietary cholesterol (190, 193). In humans, the average absorptionof dietary cholesterol is about 40%. When the dietary cholesterol or fat intake is increased,the percentage of cholesterol absorption is either reduced (194) or unaffected (195).Anyway, considerable variation can be noted between subjects in the percentage of dietarycholesterol absorption and the respective dietary responsiveness of plasma LDL and HDLcholesterol concentrations (194–196), a finding which suggests the presence of geneticallydetermined differences between individuals in the regulation of dietary cholesterolabsorption. Some studies have shown that intestinal cholesterol absorption is related tothe apolipoprotein E phenotype (197–200), with apo E4 carriers absorbing morecholesterol than non-apo E4 carriers.

The variation in dietary responsiveness may be based on differences in thecompensation of cholesterogenesis (201–203). Cholesterol synthesis is sensitive to manydietary factors, including energy restriction, meal frequency, dietary fat type, andcholesterol and phytosterol content (204). An increased dietary cholesterol and SAFAintakes are associated with reduced cholesterol synthesis (202, 205–207). In theory,subjects with sufficient compensatory suppression of cholesterol synthesis for the influxof exogenous cholesterol showed only a minor response in the plasma lipid level, whereasin the ones whose endogenous cholesterol synthesis is not suppressed, the livercholesterol pool expanded and plasma LDL increased either because of increased excretionof VLDL, IDL and LDL, or because of suppression of LDL receptors, which depressedLDL clearance. At the moment, no studies to confirm this mechanism are available.

33

Fig

. 4.

M

etab

olic

p

ath

way

s co

nn

ecte

d

wit

h

die

tary

re

spon

sive

nes

s.

CH

YLO

MIC

RO

N

RE

MN

AN

TR

EC

EP

TO

R

Inte

stin

alM

uco

sal C

ell

Inte

stin

al L

um

en

LP

L

Liv

er C

ell

Cap

illar

yE

nd

oth

eliu

m

B /

ER

ecep

tor

Per

iph

eral

Cel

l

LDL

HD

L

RE

MN

AN

T

IDL

LP

LV

LDL

HD

L

VLD

L

* C

ET

P ac

tivi

ty ♦

die

tary

cho

lest

erol

and

obe

sity

* L

PL

act

ivit

y ♦

die

tary

fat

and

obe

sity

* H

epat

ic tr

igly

ceri

de ♦

die

tary

PU

FA*

rece

ptor

syn

thes

is ♦

die

tary

cho

lest

erol

* sy

nthe

sis

of b

ile

acid

s ♦

die

tary

fat

and

c

hole

ster

ol ?

* H

epat

ic c

hole

ster

ol

syn

thes

is ♦

die

tary

fat

and

cho

lest

erol

* A

bsor

ptio

n ef

fici

ency

♦ d

ieta

ry f

at a

nd

d

ieta

ry c

hole

ster

ol

* L

DL

mod

ific

atio

n ♦

die

tary

PU

FA

* C

ET

P ac

tivi

ty ♦

die

tary

cho

lest

erol

* re

cept

or s

ynth

esis

♦ d

ieta

ry S

AFA

and

cho

lest

erol

Lym

ph

Cap

illar

yE

nd

oth

eliu

m

* L

PL

act

ivit

y ♦

die

tary

fat

and

obe

sity

Pla

sma

B /

ER

ecep

tor

CH

YLO

MIC

RO

N

CE

TP

CE

TP

CE

TP

HD

L RE

MN

AN

T

34

Differences in the capacity to respond by altered excretion of cholesterol into bilecould explain the differences in dietary responsiveness. This has been confirmed by studiesconducted on laboratory animals (208), but some contradictory findings also exist (209,210). In humans, the consumption of extra cholesterol usually does not lead to enhancedbile acid excretion, and the differences in the responsiveness to dietary cholesterol and fatare most probably not based on differences in cholesterol excretion (190).

The regulation of the activity of the LDL receptor on the cell surface is the majormechanism of cellular cholesterol homeostasis (211). The dietary response of plasmaLDL cholesterol to changes in the dietary saturated fat and cholesterol content is mediatedmainly by the level of hepatic LDL receptors (212). There is some evidence of differencesin the number and activity of LDL receptors between hypo- and hyperresponders,explaining the differences in dietary responses (190). In addition, the binding affinity oflipoproteins to their respective receptors may vary according to the genetic variation ofapoproteins, for example.

The differences in LDL production may explain the differences in dietaryresponsiveness (188) and several mechanisms have been suggested (203). Especially therole of CETP activity seems to be important in the LDL response after the consumptionof diets modified in fat and cholesterol (213–217).

2.3.4.2. Genetic variation affecting dietary response in plasma lipids andlipoproteins

All the genes encoding proteins for the pathway of lipoprotein metabolism (lipoproteins,lipoprotein receptors, lipid transporters and lipid-metabolizing enzymes) and the genesassociated with transcriptional activation and signals modulating the transcription of therespective genes are potential candidates for gene-diet interaction. The dyslipidemic effectsof the rare mutations causing the phenotype of familial hypercholesterolemia and familialdefective apoB are well described (218, 219). In general, their effect on plasma lipid levelsis so great that hyperlipidemia develops even in a low-risk environment. The interactionbetween diet and these rare mutations is not the main interest, because the low-fat, low-cholesterol diet should be recommended to these patients despite the expected cholesterolresponse. The common polymorphisms associated with a relatively small effect on theplasma lipid levels in individuals have a greater influence on hyperlipidemia at thepopulation level than do the rare mutations. Some of the common polymorphisms maypredispose to the development of hyperlipidemia, especially in a high-risk environment,while some of them may protect against developing hyperlipidemia even in a high-riskenvironment. Common polymorphisms are defined as mutations having only a smalleffect on plasma lipid levels (on an average 5-10%) and being present in the generalpopulation at an allele frequency of greater than 1%. The common variations in theknown candidate genes are presented in Table 3. The identification of genetic variationsand the information of the interaction between these genetic variations and diet helps toidentify the subjects who will benefit from dietary counselling.

35

Table 3. Common polymorphisms of human candidate genes regulating the lipid responseto dietary changes.

Gene PolymorphismsEffects on plasma lipids/ CHD risk Function

Apo A-I XmnI, PstI P- HDL, apoA-I Transcription/synthesisrate of apoA-I

Apo A-IV XmnI, SstI, PstI Lipid effect uncertain Binding to lipoprotein;intestinal absorption

Apo B XbaI* P-LDL, CHD risk Linkage disequilibriumwith functionalsequence: transcription/production of apoB

EcoRI, MspI, Bsp I CHD risk Affinity of LDL for itsreceptor

Signal Peptide P-total cholesterol,triglycerides, CHD risk

ApoB secretion

3'VNTR CHD risk ?

Apo C-III SstI*, PvuII P-triglycerides, CHD risk ?

Apo E E2, E3, E4 P-total and LDL cholesterol,apoB, CHD risk

Intestinal cholesterolabsorption; LDL apoBproduction; bile acidand cholesterolsynthesis; postprandialclearance; upregulationof LDL receptor

CETP TaqI, MspI, StuI,EcoRI

P-HDL, apo A-I ?

FABP2 T54 variant P-triglycerides Intestinal fat absorption

LPL HindIII* P-triglycerides Linkage disequilibriumwith functionalsequence; catabolism oftriglyceride richlipoproteins

* No structural or functional effect on gene product.

36

2.3.4.3. Nutritional regulation of gene expression

The most important hormones regulated by dietary factors and involved in lipogenesis andfatty acid synthesis are insulin, glucagon, glucocorticoids and triiodothyronine (220).Dietary carbohydrate and fat are the major components influencing the rate of lipogenesis,and the nutritional states of fasting, feeding, malnutrition and obesity influencelipogenesis and energy metabolism.

The pathway of gene expression comprises the steps presented in Fig. 5.

Protein function

Proteinconcentration

Synthesis Degradation

mRNA amount

Translationefficiency

Transcription

• Initiation of transcription

• Elongation of transcript

• Termination of transcript

• Cleavage and polyadenylation of transcript• Capping of transcript

• Splicing of transcript

• Degradation of transcript

Fig. 5. Pathway of gene expression.

It has been hypothesized that the dietary regulation of gene transcription mightexplain the association between dietary fat and the induction of atherosclerosis (221).Nutrients can influence gene expression directly through gene promoters, through thecontrol of regulatory signals in untranslated regions, and through post-transcriptionalpathways. Dietary PUFAs reduce fatty acid biosynthesis and enzyme activities bydecreasing the respective mRNA levels (223, 223). The alteration in gene transcriptionoccurs so fast (within 3 hours) that the response is not mediated by altered lipid

37

concentrations of cell membranes or by altered signaling mechanisms. The potentialregulatory site for PUFA is the activation/inhibition of the nuclear transcription factor,e.g. the peroxisomal proliferator-activated receptors (PPARs), the hepatic nuclear factor(HNF)-4, the signaling molecules derived from the cholesterol biosynthetic pathway (e.g.sterol regulatory element binding proteins, SREBPs), or the replacement of a positivetrans-factor by a silencer trans-factor. PPARs are members of the steroid/thyroid nuclearreceptor superfamily of ligand-activated transcription factors. To date, three isotypes havebeen identified: α , β and γ , which are encoded by three different genes. The role ofPPARα activation in lipoprotein metabolism and in the inhibition of the development ofatherosclerosis seems obvious (224). All PPARs are activated by fatty acids andderivatives, such as leukotriene B4 and prostaglandin J2.

Dietary cholesterol influences the cellular cholesterol pool and, as a consequence,several genes respond with altered gene expression. The addition of cholesterol to cellsleads to an altered gene expression of the LDL receptor (225, 226), LRP (227), the sterolcarrier protein (228), interleukin-8 (229), the monocyte chemoattractant protein (229),HMGCoA reductase (225), HMGCoA synthase (225), CETP (230, 231) and apo E (232).

2.3.4.4. Obesity and gene expression

Chronic hyperalimentation leads to obesity because the excess dietary energy is convertedto fat. A significant fraction of human obesity is thought to be genetically determined(233, 234). Obesity itself promotes several metabolic disturbances, some of which affectgene expression. The human brain operates through different mechanisms to control thebody’s nutrient storage. Excess availability of nutrients/energy alters the expression of thehypothalamic peptides neuropeptide Y and galanin (235), and excess lipid storage inadipose tissue increases the expression of leptin (236). Adipocyte differentation iscoordinately regulated by several transcription factors (C/EBPβ, C/EBPσ, ADD-1/SREBP-1, PPARγ). Altered activity or expression of these transcription factors isthought to have a role in the pathogenesis of obesity (237). Adipose expression ofPPARγ2 mRNA is increased in human obesity (238). On the other hand, PPARα andPPARγ activators, such as fibrates and thiazolidinediones induce LPL expression, and asequence element in LPL promoter region has been recognized which interferes withPPARs (239). Uncoupling proteins, which permit the oxidation of fuels without thegeneration of adenosine triphosphate, are regulated by dietary factors and may have a rolein the development of human obesity (240). Tumor necrosis factor (TNF) has multipleactions in adipose tissue, including the regulation of LPL activity (241), and TNFαexpression is increased in obesity.

3. Aims of the study

The following specific questions were addressed:

1. What is the magnitude of dietary responsiveness in the plasma lipid and lipoproteinlevels in free-living subjects?

2. Does dietary intervention alter a normotensive blood pressure level, and does theapo E phenotype have an effect on the blood pressure response?

3. Does the apo E phenotype have an effect on the response of blood lipids andlipoproteins to diet?

4. What is the overall effect of genetic variation in the apo B gene on the response todiet?

5. Does weight reduction following caloric restriction alter the gene expression ofCETP and LPL?

4. Subjects and Methods

4.1. Study subjects

The Ethical Committee of the University of Oulu approved the protocols for the studiesreported (I–IV). All study subjects volunteered and gave informed consent beforeparticipation and underwent detailed clinical screening prior to inclusion.

Report I-III: Employees (N=320) of the technical department, administrative office andkitchen of the Oulu University Hospital were invited to participate in screening for thedietary trial. Out of them, two hundred (62.6%) volunteered for the plasma lipid analysesand apolipoprotein E phenotype determination. Out of the population screened for theirapolipoprotein E phenotype, 21 subjects with the phenotype E4/4 or 3/4 volunteered forthe diet intervention trial. These subjects were similar with respect to age, sex and bodymass index with 23 individuals having the phenotype E3/3. The baseline characteristics ofthe study subjects are presented in Table 4.

Report IV: The study subjects for the weight reduction trial were drawn from amongobese female patients scheduled for an elective gallbladder operation in the near future.The patients were invited for laboratory screening and clinical investigation, and out ofthe 40 patients invited, 34 volunteered for the trial. One subject dropped out because shedeveloped atrial fibrillation before her gallbladder operation and her operation wascancelled. The number of study subjects in the weight reduction group was 13. Thesesubjects were matched for age and BMI with a control group of 20 subjects. The baselinecharacteristics of the study subjects in the weight reduction group and control the groupfrom report IV are presented in Table 4.

4.2. Diets and study design

Reports I-III: Seven-day food consumption records were kept during the baseline dietphase. The food records were collected before the trial, and all study subjects were advisedhow to fill in food diaries, and all food diaries were checked by an experienced dietitian.The baseline diet represented a typical Western diet and consisted partly of meals from the

40

hospital kitchen and partly of the subjects' habitual food eaten at home. The baseline dietof the study subjects is presented in Table 5a. The intervention diets were designed on thebasis of the regular hospital meals. The hospital meals were analyzed for seven days.After the trial the nutrient contents of the intervention diets were recalculated according tothe actual daily records. The composition of dietary fat in the intervention diets was asfollows: 1) 26 E% from fat, with 32% from SAFA, 35% from MUFA and 33% fromPUFA, during the low-fat diet; and 2) 38 E% from fat, with 60% from SAFA, 31% fromMUFA and 9% from PUFA, during the high-fat diet. The cholesterol content was 240 mgper day during the low-fat diet and 420 mg per day during the low fat diet.

Table 4. Baseline characteristics of the study subjects in Reports I-III and IV.

Number Percentage Systolic / diastolicof subjects Age

± SDBMI± SD

of smokers blood pressure± SD

(years) (kg/m2) (%) (mmHg)

Report I-III

All 44 38±8 24±3 23 136±14 / 81±8

Men 22 37±7 25±2 32 140±14 / 83±8

Women 22 39±8 23±3 14 131±14 / 80±7

Report IV

All (women) 33 56±8 32±5 18 138±17 / 83±7

Control 20 58±8 32±5 25 137±14 / 82±6

Intervention 13 54±8 31±5 8 140±21 / 83±7

All the food consumed during the intervention periods was supplied from the hospitalkitchen. The two meals comprising about 75% of the daily energy intake were served atthe hospital, and the evening snacks and breakfasts, both to be consumed at home, werealso prepared in the hospital kitchen. The meals for the weekends were prepared anddistributed on Fridays or Saturdays to be consumed at home. The most suitable energyintake to maintain the body weight was chosen according to the individuals' foodconsumption record and daily physical activity, the mean daily energy intakes being 12.6MJ (3000 kcal) for men and 8.4 MJ (2000 kcal) for women. According to the food diariescollected during the baseline period, the amount of alcohol used by the study subjects wasnegligible, and alcohol was therefore not included in the calculations of diets. The studydesign is presented in Fig. 6.

41

Table 5. Composition of baseline diets (/10 MJ) of the study subjects in (a) Reports I-IIIand (b) Report IV.

(a)Reports I-III

All Women Men Apo E3 Apo E4

N 44 22 22 23 21Protein (g) 103 ± 16 103 ± 16 103 ± 15 101 ± 18 105 ± 14Fat (g) 96 ± 16 96 ± 18 95 ± 14 96 ± 18 95 ± 16Carbohydrate (g) 268 ± 35 271 ± 38 265 ± 33 273 ± 36 263 ± 35Alcohol (g) 3.2 ± 5.9 1.1 ± 2.2 5.5 ± 7.6 1.0 ± 2.1 5.3 ± 7.4Fiber (g) 27 ± 7 28 ± 9 26 ± 4 28 ± 8 26 ± 7MONO (g) 32 ± 6 32 ± 7 32 ± 5 31 ± 7 32 ± 6SAFA (g) 42 ± 9 42 ± 10 43 ± 8 43 ± 10 42 ± 9PUFA (g) 13 ± 4 14 ± 4 13 ± 3 13 ± 5 13 ± 3Cholesterol (mg) 387 ± 95 377 ± 75 398 ± 113 376 ± 80 398 ± 108

(b)Report IV

All(women)

Intervention Control

N 33 13 20Protein (g) 102 ± 12 102 ± 13 101 ± 11Fat (g) 92 ± 15 89 ± 14 92 ± 16Carbohydrate (g) 274 ± 36 279 ± 42 276 ± 33Alcohol (g) 4.1 ± 9.2 4.1 ± 12.8 3.5 ± 6.3Fiber (g) 27 ± 6 27 ± 5 27 ± 7MONO (g) 31 ± 6 30 ± 6 31 ± 7SAFA (g) 39 ± 8 39 ± 8 39 ± 8PUFA (g) 14 ± 4 13 ± 4 14 ± 5Cholesterol (mg) 317 ± 79 315 ± 88 310 ± 74

Report IV: Before the weight reduction period, all the study subjects consumed theirhabitual diets. The seven-day food consumption records were collected and checked by anexperienced dietitian, and the baseline diet of the study subjects is presented in Table 5b.The weight reduction program was based on dietary counselling. The intervention subjectswere placed on a hypocaloric AHA step I diet (242) with a recommended daily energycontent of 5.0 MJ (1200 kcal). The control subjects were instructed to continue theirhabitual diet and not to lose weight.

42

Fig. 6. Time line of the study design in Reports I-III and IV. The arrowsindicate timing of the laboratory measurements.

.4.3. Clinical measurements

The same trained physician with specialist qualifications in internal medicine interviewedand examined all the study subjects. Special attention was given to the past medicalhistory, current symptoms, medication, family history, smoking habits, alcoholconsumption and physical activity. Women were classified as postmenopausal if at leastsix months had passed since their latest menstruation and no other relevant reason for thecessation of the menstruation cycle could be identified.

4.3.1. Blood pressure measurements

All blood pressure measurements were made using an automatic, microprosessor-controlled device (Criticon Dinamap 1846 SXiP; Critikon Inc, Tampa, FL), in which anoscillometric technique is used and which determines automatically the systolic bloodpressure, the mean arterial pressure, the diastolic blood pressure, and the pulse rate. Blood

Report I-III:

Report IV:

Baseline period12 weeks

Low-fat period4 weeks

High-fat period4 weeks

Switch-back period8 weeks

Baseline period2–4 weeks

ScreeningpopulationN=40

StudypopulationN=34

DropoutN=1

ScreeningpopulationN=200

StudypopulationN=44

Weight-reduction period8 weeks

Weight-maintaining period8 weeks

ControlgroupN=20

InterventiongroupN=13

DropoutN=1

DropoutN=1

Gallbladder operation

43

pressure was registered by the same experienced nurse on all occasions. Before eachmeasurement, the subjects rested for at least 10 minutes and blood pressure was registeredfrom the right arm with the subject sitting.

Reports I–III: The blood pressures were measured on at least two occasions during thebaseline period and once a week during both the intervention periods and, furthermore,twice during the switch-back period.

Report IV: The blood pressures were measured on two occasions during the baselineperiod and once every second week during the weight reduction period.

4.3.2. Body weight measurement

All the body weights used in the analyses were recorded using a digital weighing scale(Lindetronic 4000; Lindells, Uppsala, Sweden). The standing heights were also measured,and the body mass index (BMI) was calculated as body weight (kg) divided by the squareof height (m2).

Reports I-III: The study subjects were weighed on two occasions during the baselineperiod and once a week during the intervention diets in our laboratory, in addition towhich they were asked to weigh themselves daily during the trial.

Report IV: The study subjects were weighed on two occasions during the baselineperiod and once every second week during the weight reduction period in our laboratory,and they were asked to weigh themselves daily during the trial and also just before theelective gallbladder operation.

4.4. Routine clinical laboratory analyses

Some clinical laboratory analyses were used to exclude subjects with diseases of majorimportance. The laboratory analyses of blood count, serum alanine amino transferase,serum creatinine and serum thyroxine-stimulating hormone were carried out in the routinelaboratory of the Oulu University Hospital using standard methods.

4.5. Plasma lipid and lipoprotein analyses

Venous blood samples were collected after an overnight fast in EDTA-containing tubes.Plasma was separated by centrifugation at 1200 g for 10 minutes (+4°C). Plasmalipoproteins were analyzed as follows: VLDL fraction (d<1.006 g/ml) was isolated fromthe plasma by ultracentrifugation in a Kontron TFT 45.6 rotor (Kontron Instruments,Zurich, Switzerland) at 114 000 g and 15°C for 18 hours. On ultracentrifugation, theVLDL fraction floated on the surface and was then removed by tube slicing. The plasmaHDL cholesterol concentration was determined from the remaining infranatant afterprecipitation of IDL and LDL with heparin-manganese chloride (243). The plasma LDL

44

cholesterol was calculated according to the Friedewald formula (244). The plasma totalcholesterol and triglyceride concentrations were determined from fresh plasma or itsfractions with enzymatic colorimetric methods (kits by Boehringer Mannheim, GmbH,Germany, cat. no:s 236691 and 701912, respectively) using a Gilford IMPACT 400 EClinical Chemistry Analyzer.

4.6. Determination of apolipoprotein E protein polymorphism

Reports I, II: Apolipoprotein E phenotype was determined from whole plasma afterdelipidation, using the isoelectric focusing and immunoblotting techniques (245, 246)

4.7. Determination of plasma CETP activity

CETP activity was measured as described previously (108). In this method, theapolipoprotein B-containing lipoproteins in the plasma samples are precipitated and theCETP-containing supernatant is incubated with labelled LDL and unlabelled HDL. Afterincubation, LDL is precipitated and the radioactivity of the HDL-containing supernatant iscounted. The LDL and HDL used in the assays were isolated from healthy volunteers.

4.8. Determination of apolipoprotein B DNA polymorphisms

Report III: Analyses of the apolipoprotein B restriction fragment length polymorphismswere performed from DNA samples prepared by the salting-out method as described byMiller et al. (247). The polymorphic sites presented in Table 6 were determined.

Table 6. The apolipoprotein B DNA polymorphisms determined.

Apo B Change in Change inPolymorphism Exon Codon Nucleotide Amino acid

1 XbaI 26 2488 C to T —

2 EcoRI 29 4154 G toA Glu to Lys

3 MspI 26 3611 A to G Arg to Gln

4 Bsp 1261I 4 71 C to T Thr to Ile

5 Insertion/deletion of signal peptide 3-codon (Ala-Leu-Ala) in the region of the human apo B gene.

45

The methods to determine the XbaI and EcoRI DNA polymorphisms have beendescribed before (248). The MspI, Bsp1261I and signal peptide polymorphisms weremeasured using specific oligonucleotide primers, PCR amplifications and digestions ofDNA fragments as previously described (249-251).

4.9. Measurement of CETP and LPL mRNA

4.9.1. Total RNA extraction

Liver tissue samples (20 to 100 mg) were collected by open surgical biopsy during thegallbladder operation. The fresh liver tissue samples were flash-frozen with liquidnitrogen and kept at -70°C until RNA extraction. Adipose tissue samples were collectedduring the gallbladder operation from omental fat or abdominal subcutaneous fat (200 to2000 mg). Excess blood was rinsed with physiological NaCl solution, and the adiposetissue samples were flash-frozen with liquid nitrogen and kept at -70°C until RNAextraction. To extract total cellular RNA, the guanidine thiocyanate method withcentrifugation through a CsCl gradient was used as described previously (252).Theconcentration of total cellular RNA was measured by using spectrophotometry. Theaverage ratios of A260 and A280 nm were 1.8, and the integrity and quality of all RNAsamples were determined by electrophoresis on denaturing agarose gel, which showedintact ribosomal RNA bands. The average yield of total cellular RNA was (±SD) 100 ±25µg/100 mg of liver tissue and 5.0 ±2.0 µg/ 100 mg of adipose tissue. All RNApreparations were found to be free of DNA contamination by amplifying them with thePCR primer pairs used in the studies. The primer pairs were selected to include at leastone intron in the respective DNA sequence, and DNA contamination would thus havebeen distinguished by an extra, larger PCR product than the one obtained from RT-PCRof uncontaminated RNA.

4.9.2. Construction of internal RNA standard

To obtain suitable internal RNA standards for a quantitative reverse transcription-polymerase chain reaction, primer pairs were designed separately for CETP and LPLconstructs. All the primer pairs are presented in Table 7, and the strategy for theproduction of the internal RNA standard is presented on Fig. 7.

The concentration of the constructed internal RNA standard was determined byabsorption spectrophotometry at 260 nm. In designing the primers, the general rules wereused (253). The primers were synthesized at the Department of Biochemistry, Universityof Oulu.

46

Table 7. Primer pairs for the construction of internal RNA standards and for thereverse transcription-polymerase chain reaction (RT-PCR) of the cholesteryl ester transferprotein and lipoprotein lipase.

CETP DNA standard (size of the PCR product 244 bp):CETP a20 (forward)5’-TCCAGATCAGCCACTTGTCC-3’CETP b20c20 (reverse)5’-TCTCGCTCCCCTTGGAGATG*CACTCTACCAGAGTCACAGG-3’

CETP DNA T7–promoter attached (size of the PCR product 267 bp):CETP T7–a20 (forward)5’-TAATACGACTCACTATAGGGAGATCCAGATCAGCCACTTGTCC-3’CETP b20c20 (reverse)

CETP RNA standard transcript (244 bp) and CETP mRNA (292 bp):CETP a20 (forward)CETP c20 (reverse)5’-TCTCGCTCCCCTTGGAGATG-3’

LPL DNA standard (size of the PCR product 234 bp):LPL a21 (forward)5’-GAGATTTCTCTGTATGGCACC-3’LPL b20c21 (reverse)5’-CTGCAAATGAGACACTTTCTC † TACTCTGATCTTCTGAATGG-3’

LPL DNA T7–promoter attached (size of the PCR product 257 bp):LPL T7–a20 (forward)5’-TAATACGACTCACTATAGGGAGAGAGATTTCTCTGTATGGCACC-3’LPL b20c21 (reverse)

LPL RNA standard transcript (212 bp) and LPL mRNA (254 bp):LPL’ a21 (forward)5’-TGGCCGAGAGTAGAACATCCC-3’LPL c21 (reverse)5’-CTGCAAATGAGACACTTTCTC-3’

The primers were selected for CETP to amplify the gene sequence from exon 3to exon 5, and for LPL from exon 7 to exon 9.

47

Fig. 7. Schematic presentation of the construction of an internal RNAstandard for quantitative RT-PCR.

4.9.2. Reverse transcription

The reverse transcription reactions were carried out in 20 µL volumes containing 1.0 to0.5 µg total RNA and two fold serial dilutions of the constructed internal RNA standard(from 0.1 to 1 ng). RNA was reverse-transcribed to first-strand cDNA with Moloneymurine leukemia virus reverse transcriptase (5 U /µl), RNase inhibitor 1.2 U/µl, DTT 4mM and specific oligonucleotide antisense primers (0.2 µM). Negative controls wereincluded into every series of samples.

3’ 5’(1)3’5’

c 20

b 20

mRNA

DNA(2) 3’5’

4dNTPsReverse Transcriptase

a 205’3’

(3)Taq DNA Polymerase

3’5’

5’3’

Wizard™ PCR Preps Purification

(4) T7-

promoter

(5)

Taq DNA Polymerase

Wizard™ PCR Preps Purification4rNTPsT7 RNA Polymerase

3’ 5’ RNA Transcript

(6)

48 bp

3’3’

3’ 3’

3’3’

5’5’

5’

5’5’

5’

48

4.9.3. Polymerase chain reaction

PCRs were performed in a 50 µL volume containing 2.5 mM buffer and 0.4 µM each ofthe gene-specific primers. Accurate pipetting was accomplished by the use of mastermixes. The concentration of dNTP was 0.4 mM of each nucleotide (dATP, dCTP, dGTP,dTTP), that of MgCl2 was 2.5 mM, the Taq DNA polymerase (Finnzymes) concentrationwas 0.025 U/µl, and that of PCR buffer 2.5 mM. The optimal cycling temperatures weretested separately for the CETP gene and the LPL gene, as was the optimal number ofcycles. During the PCR reaction, the duration of the denaturation step was 30 seconds,except for the denaturation of 5 minutes during the first cycle. The annealing step was 30seconds and the extension step one minute, except for the extension of 15 minutes duringthe last cycle. The reproducibility of the quantitative RT-PCR method was testedseparately.

4.10. Statistical analyses

Reports I-III: The power calculations were performed beforehand with the assumption thatthe study subjects are similar to those of North Karelia Project (85) (average change ofplasma total cholesterol is 20 % and sample SD 0.2 %). With the requirement ofstatistical significance level of 5 % the sample size was found to be less than 20 for theestimated effect size. The actual number of study subjects recruited was then planned tobe 20 % more than needed. The minimum number of study subjects in report IV wasdesigned to be 10 in the intervention group and 10 in the control group.

The data are presented as group means ±SD. The differences within groups were testedby paired Student's t-test, and those between groups by analysis of variance or, whennecessary, by analysis of variance for repeated measures. The normality of the variableswas tested by the Shapiro-Wilkinson test. For variables which were not normallydistributed or whose variance was not homogeneous, the Kruskal-Wallis test was used.The analysis of covariance was used for the adjustment for baseline differences. Spearmancorrelation coefficients were calculated for some variables.

To calculate the individual dietary compositions, the Nutrica (Social InsuranceInstitution, Helsinki, Finland) and the AIVO (Aivo, Finland Inc., Turku, Finland)nutrition software programs based on the Finnish nutrient database (254) were used.

Report III: The meta-analysis was based on the method described by Hedges and Olkin(255). The calculations were performed using Microsoft Excel datasheets. Otherwise, allthe calculations were made using the JMP Statistical Software (SAS Institute Inc., Cary,NC).

5. Results

5.1. Diets

(Reports I–III): The habitual diet included 37 E% from total fat, 17 E% from SAFA, 12E% from MUFA and 5 E% from PUFA, and the composition was similar in both apo Egroups during the baseline period. None of the study subjects were on an extreme dietduring the baseline period. The changes in the nutrient compositions of the study dietswere achieved as planned according to the food consumption records and the observedchanges in plasma lipids and lipoproteins. Both the low-fat, low-cholesterol diet and thehigh-fat, high-cholesterol diet provided the same total energy content as the habitual diet.Compared with the habitual diet, during the low-fat, low-cholesterol diet the cholesterolintake decreased 30%, the total fat intake 30%, SAFA intake 56%, MUFA intake 31%,and PUFA intake increased 54%. Compared with the habitual diet, during the high-fat,high-cholesterol diet the cholesterol intake increased 3%, the total fat intake 5%, SAFAintake 30%, MUFA intake decreased 9%, and PUFA intake decreased 31%.

(Report IV): The habitual diet included 34 E% from total fat, 15 E% from SAFA, 12E% from MUFA and 5 E% from PUFA. The hypocaloric, low-fat intervention dietprovided a 1.3 MJ reduction in the total energy content with a reduction of 4% in totalfat, a 5% reduction in SAFA, a 2% reduction in MUFA and a 5% increase in PUFA.

5.2. Body weight

(Report I-III): The body weight remained constant throughout the trial, and only a minorreduction was noticed among the women (Fig. 8a).

(Report IV): The body weight remained constant among the control subjects (Fig. 8b)except in one, who lost 1.5 kg of body weight just before the gallbladder operation, andwas thus excluded from the final analyses. The body weight was reduced significantlyamong the intervention subjects (Fig. 8b), except in one, who regained weight just beforethe gallbladder operation and was thus excluded from the final analyses.

50

Fig. 8. Percentage changes of body weights during dietary interventions i nreport I–III (a) and report IV (b).

5.3. Plasma lipids and apolipoproteins and the effect of apo Epolymorphism

(Report I-II): The plasma total, HDL and LDL cholesterol, apoB, and apo A-Iconcentrations decreased significantly during the low-fat diet and increased during the high-fat diet. In addition, the plasma triglycerides concentration increased significantly duringthe low-fat diet and decreased during the high-fat diet (Figs. 9 and 10). At the end of thestudy, when the subjects returned to their habitual diets, the plasma lipid and lipoproteinconcentrations returned to the baseline values. The baseline concentrations of plasmalipids and lipoproteins were similar in the subjects with different apo E phenotypes. Theabsolute and percentage changes of lipids and lipoproteins were equal in the subjects withthe common apo E phenotype 3/3 and in those homozygous and heterozygous for theepsilon 4 allele (E4/4 and E3/4) (Fig. 9).

(Report IV): The plasma lipid and lipoprotein concentrations were measured everysecond week during the active weight-reducing period among the intervention subjects andevery second week during the stable, weight-maintaining follow-up period in the controlsubjects. In the control subjects, plasma lipids and lipoproteins remained unchanged. Inthe intervention subjects, the plasma triglycerides and serum insulin concentrationsdecreased significantly (–13% and –14%, respectively, p < 0.05). The HDL cholesterolconcentration tended to decrease, but the change was not statistically significant.

-8

-6

-4

-2

0

2

4

All

Female

Male

Low fat High fat Switch back%

cha

nge

of t

he b

ody

wei

ght

-8

-6

-4

-2

0

2

4

Control

Intervention

% c

hang

e of

the

bod

y w

eigh

t

p < 0.001(a) (b)

51

Fig. 9. Percentage changes of plasma lipids and lipoproteins in report I i nsubjects with the apolipoprotein E phenotype 3 (homozygous for t h eepsilon 3 allele) and 4 (homozygous or heterozygous for the epsilon 4a l l e l e ) .

-80

-60

-40

-20

0

20

40

60

80

100

Per

cent

age

chan

ge o

f pl

asm

a lip

ids

and

lipop

rote

ins

Apo E4phenotype

Apo E3phenotype

Apo E4phenotype

Apo E3phenotype

Pla

sma

Tota

l Cho

lest

erol

Pla

sma

HD

L C

hole

ster

ol

Pla

sma

LD

L C

hole

ster

ol

Pla

sma

Apo

A-I

Pla

sma

Tri

glyc

erid

es

Pla

sma

Apo

B

LOW-FAT

HIGH-FAT

52

Fig. 10. Distribution of percentage changes in the plasma LDL c h o l e s t e r o land apoB concentration.

5.4. Blood pressure

(Reports II): The study subjects were normotensive, and none of them were on medicationaffecting blood pressure. Systolic, diastolic and mean arterial pressure decreasedsignificantly when the subjects were switched from their habitual diet to the low-fat dietand increased again on the high-fat diet (Fig. 11). The blood pressure response to dietarychanges differed between the apo E phenotype groups, the apo E 4/4 and E 3/4 subjectshaving a greater reduction in their systolic and mean arterial pressure values (–5.8±1.4%and –6.1±1.2%, p<0.001) when switched from the baseline diet to the low-fat diet thanthe E 3/3 subjects (–1.0±1.3% and –1.5±1.2%, ns)(Fig. 11).

-50

0

50

100

150

Plasma-LDL-CLow-fatdiet

Plasma-LDL-CHigh-fatdiet

Plasma-Apo-BLow-fatdiet

Plasma-Apo-BHigh-fatdiet

Dis

trib

utio

n of

per

cent

age

chan

ge

53

Fig. 11. Percentage changes of blood pressure during the low-fat and h i g h -fat diets. Groups E3 and E4 are the same as in Fig. 9.

(Report IV): The systolic and diastolic blood pressures dropped during the weightreduction, although the changes observed did not differ statistically from those in thecontrol group.

5.5. Plasma CETP activity

(Reports I–III): Plasma CETP activity increased significantly when the subjects switchedfrom the low-fat diet to the high-fat diet (36±20 nmol/ml/h, p < 0.001), the increasebeing similar in men (38±18 nmol/ml/h, p < 0.001) and women (34±22 nmol/ml/h, p <0.001). The increase of plasma CETP activity correlated with the increase in plasma LDLcholesterol (r=0.30, p<0.05).

-8

-6

-4

-2

0

2

4

6

Systolicblood

pressure

Diastolicblood

pressure

Meanarterial

pressure

Low-fat diet period High-fat diet period

Systolicblood

pressure

Diastolicblood

pressure

Meanarterial

pressure

Apo E3N=23

Apo E4N=21

* p<0.05

** p<0.01

Per

cent

age

chan

ge o

f bl

ood

pres

sure

*** p<0.001*** ***

**

**

54

(Report IV): Plasma CETP activity remained unchanged in the control subjects. In theintervention subjects, the plasma CETP activity decreased during the first two weeks ofcaloric restriction (–8±9 nmol/ml/h, p < 0.05), but returned to the basal level at the endof the two-month intervention period.

Fig. 12. Amount of cholesteryl ester transfer protein and l i p o p r o t e i nlipase mRNA in control and intervention subjects. Open circle, mRNAmeasured from visceral adipose tissue, closed circle, mRNA measured fromabdominal subcutaneous adipose tissue.

5.6. Liver and adipose tissue CETP mRNA

(Report IV): The amount of CETP mRNA in the liver tissue of weight-reduced subjectswas not significantly different from that of the controls. The liver CETP mRNAcorrelated with the plasma CETP activity ( r= 0.47, p < 0.05). The adipose tissue CETPmRNA amount tended to be lower in intervention subjects compared to the controls(–20%), especially the amount of mRNA measured from abdominal subcutaneous adiposetissue (Fig. 12). The plasma CETP activity did not correlate with adipose tissue CETPmRNA.

5.7. Adipose tissue LPL mRNA

(Report IV): The expression of LPL mRNA, especially when measured fromsubcutaneous adipose tissue, was significantly lower in the intervention subjects than thecontrol subjects (Fig. 12). Both the weight loss (r = –0.58, p < 0.05) and the reduction intriglycerides (r = 0.70, p < 0.05) correlated with the LPL mRNA amount in theintervention subjects.

0

2

4

6

8

1

Control Intervention N=12 N=8

0

2

4

6

8

10

0

3

6

9

12

15

Control Intervention N=11 N=10

Control Intervention N=18 N=12A

mou

nt o

f ad

ipos

e ti

ssue

CE

TP

mR

NA

x 10

^8

mol

ecul

es/µ

g to

tal R

NA

Am

ount

of

liver

tis

sue

CE

TP

mR

NA

x 10

^8

mol

ecul

es/µ

g to

tal R

NA

Am

ount

of

adip

ose

tiss

ue L

PL

mR

NA

x 10

^8

mol

ecul

es/µ

g to

tal R

NA

55

5.8. Effect of apolipoprotein B DNA polymorphisms on theresponse to the diet

(Report III): Among the 200 subjects screened for the dietary intervention trial, the apo BXbaI, MspI, BspI and apo B signal peptide polymorphisms were associated with theplasma lipid and lipoprotein concentrations (Table 8).

In our dietary intervention trial conducted on 44 healthy subjects, the XbaIpolymorphism significantly affected the diet-induced responses of plasma total and LDLcholesterol concentrations when all the dietary periods were analyzed together (p<0.05,repated measures ANOVA). The subjects with the genotype X-X- had the greatest increasein their plasma total and LDL cholesterol concentrations during the high-fat diet and thegreatest decrease during the low-fat diet. Also, the M+ allele of the apo B MspIpolymorphism and the R- allele of the apo B EcoRI polymorphism were associated with agreater response during the high-fat diet.

For a meta-analysis of the role of the apo B polymorphisms in the response to diet,all published dietary intervention studies were identified by literature searches. Includingour own study, 14 eligible reports were found, including a total of 593 study subjects.The average diet-induced alterations in the plasma total cholesterol, HDL cholesterol, andtriglyceride concentrations did not differ between the genotypes X+X+, X+X- and X-X-,and a comparison of effect sizes between the genotype groups did not reveal anysystematic effect of the XbaI X+ or X– allele. The apo B EcoRI R– allele and the apo BMspI M+ allele were related to a greater response to diet, especially in the plasma apoBconcentration.

56

Table 8. Fasting plasma lipid values of screenees in the Reports I-III. (The values areexpressed as mean ± SEM)

Apo BPoly-morphism

Number ofsubjects(F/M)

Plasmatotalcholesterol(mmol/L) 1

Plasmatriglyceride(mmol/L) 1

PlasmaHDL-cholesterol(mmol/L) 1

PlasmaLDL-cholesterol(mmol/L) 1

All 188 (78/110) 5.39±0.07 1.18±0.05 1.67±0.03 3.23±0.08

XbaI 187 (78/109)X-X- 74 (33/41) 5.25±0.12 1.19±0.09 1.65±0.05 3.12±0.13X+X- 88 (37/51) 5.40±0.11 1.15±0.08 1.65±0.08 3.22±0.12X+X+ 25 (8/17) 5.79±0.20 1.21±0.15 1.74±0.08 3.51±0.21

EcoRI 187 (78/109)R+R+ 111 (46/65) 5.51±0.10 1.22±0.07 1.72±0.04 3.28±0.11R+R- 64 (26/38) 5.26±0.13 1.06±0.09 1.58±0.05 3.24±0.14R-R- 12 (6/6) 5.01±0.29 1.41±0.21 1.66±0.11 2.70±0.32

MspI 185 (77/108)M+M+ 157 (61/96) 5.32±0.08 1.19±0.06 1.66±0.03 3.14±0.09M+M- 28 (17/11) 5.79±0.192 1.10±0.14 1.72±0.08 3.73±0.222

Signal Pept. 185 (77/108)II 104 (40/64) 5.30±0.10 1.25±0.07 1.67±0.04 3.09±0.11ID 70 (32/38) 5.41±0.12 1.05±0.09 1.67±0.05 3.29±0.13DD 11(6/5) 6.22±0.31 1.26±0.22 1.67±0.11 3.98±0.322

Bsp 185 (77/108)B+B+ 100 (38/62) 5.30±0.10 1.20±0.07 1.67±0.04 3.12±0.11B+B- 75 (34/41) 5.41±0.12 1.22±0.09 1.67±0.04 3.25±0.13B-B- 10 (6/4) 6.36±0.323 1.33±0.24 1.67±0.12 4.09±0.332

1Values are mean ±SD.

2p<0.05 and

3p<0.001, Tukey’s test.

6. Discussion

6.1. Methodological considerations

6.1.1. Study populations

The Finnish population is genetically more homogeneous than many other populations(256). This can be explained by the small original population (about 1000 subjects), bythe fact that Finland (about 2000 years ago), and especially Northern Finland (about 400-500 years ago), was populated relatively recently and that the Finnish population hasgrown without immigration. The genetic homogeneity is one of the reasons why theFinnish population has been as ideal target for genetic clinical studies.

Even though the incidence of and mortality from CHD in Finland has decreased duringthe last thirty years, we are still one of the leading countries in the incidence of CHDmortality in Western Europe (257, 258). Genetic factors explain about 40% of themorbidity of CHD among middle-aged patients (259), and environmental factors such assmoking, diet and sedentary lifestyle, interacting with genetic factors explain the rest ofit. In Finland, the prevalence of familial hypercholesterolemia is not different from thatseen in other countries but the apo B 3500 mutation has not been found in the Finnishpopulation. Thus, according to the data available, no single genetic factor explaining thehigh CHD morbidity in Finland has been found.

In Finland, the average plasma total cholesterol concentration is high compared withthe other countries in Western Europe, and the prevalence of hypertension is high (260).The traditional Finnish diet contains large amounts of fatty dairy products and animal fat.Of the environmental factors, the diet has changed most dramatically during the last threedecades, which explains a notable part of the decrease in CHD mortality and morbidity.The successful education programs, such as the North Karelia Project, were followed by adecrease of the average plasma total cholesterol concentration. But we still have to payattention to the dietary counselling and education.

Because the individual responsiveness to dietary changes varies a lot and the cause forindividual variation in response to diet can be regulated by some genetic factors, the

58

search for such causes is suitable to be conducted especially among Finns, who aregenetically homogeneous.

The risk factors for gallstones are obesity, female sex, age and genetic predisposition.The apo E4 allele is associated with supersaturation of bile and an increased risk ofgallstones (261). We selected obese female patients with gallstone disease for the weightreduction trial, because the elective gallbladder operation would provide tissue samples formeasuring the gene expression after weight reduction. Most of the gallstones arecholesterol stones. Gallstone patients may have genetic factors that predispose to bothobesity and lipid abnormalities as well as to gallstones, but these factors are so farunknown.

Because our weight reduction trial was conducted on human subjects, it was notpossible to obtain tissue samples (liver and adipose tissue) on two occasions (before andafter weight reduction). We thus needed a control group of obese gallstone patients similarto the intervention group. Using several criteria (age, sex, lipid levels, general health), acareful selection of all study subjects was done before randomization into the control orweight reduction groups. The group sizes were relatively small because of the nature ofthe study.

6.1.2. Dietary interventions, study designs and feasibility ofthe diets

Epidemiological studies may not be so practical as intervention trials when the causalitybetween specific genes and environmental factors is investigated. In epidemiologicalstudies, the accuracy of dietary data may not be so good as during intervention trials.Dietary habits in a single population may vary so little that it is hard to detect smalldifferences in dietary intakes and, as a consequence, relationship between dietary intake offat and plasma lipids remains unconfirmed (262). In such a case, any significantinteraction between diet and genetic variation regulating the plasma lipids and lipoproteinsmay be difficult to be found. The number of study subjects needed would be high, and thetime needed for follow-up would be long.

In human dietary intervention studies, some problems may arise. The way in whichthe dietary intervention is performed dictates the accuracy and reliability of the study.Experimental studies conducted on metabolic wards are relatively free of problemsconcerning the feasibility of the diet. Anyway, during studies conducted oninstitutionalized subjects, their daily activities are modified, too, and it may thus bedifficult to generalize the results to community-living subjects. Placebo-controlled designis almost impossible in dietary trials.

Controlled prospective dietary studies in which the research unit prepares and servesall the meals can be considered a reliable and feasible method to investigate the role ofgenetic variation in response to diet (263). This was the design in our dietary interventionstudy (Reports I-II). In report III, the apo B DNA polymorphisms were determined afterthe dietary intervention. Thus, the study design cannot be considered prospective and theresults must be interpreted carefully. The data analyzed in meta-analysis (Report III) werepartly from prospective and partly from retrospective studies; the dietary interventions

59

were performed either on metabolic wards, where the meals were served under strictcontrol, or in research units, where the meals were partly served and partly delivered, orthe interventions were conducted on community-living subjects, who were given detailedwritten and/or oral instructions of the study diets. The weight reduction trial (Report IV)was based on dietary counselling. Better feasibility and, eventually, more effective weightreduction and greater changes of plasma lipids and lipoproteins could have been achievedwith controlled diets. At any rate, the target of one kilogram weight reduction per twoweeks was achieved.

Both interventions in our controlled dietary study (Reports I-III) were conducted forone month. Blood lipids and lipoproteins were measured each week and, depending on thechanges in the plasma total cholesterol concentration, the maximal change was achievedafter two weeks (Report II). The average duration of dietary interventions among thestudies relevant for meta-analysis was six weeks (from three to twelve weeks). In theweight reduction trial (Report IV), the average two months duration of the study wasconsidered suitable for several reasons. Firstly, the study patients had been scheduled for agallbladder operation and, two months is the average waiting time in our hospital.Secondly, the presumed alteration of gene expression was estimated to start as soon as asignificant weight reduction (more than 5 % of body weight) was achieved.

6.1.3. Quantitative RT-PCR

The quantifications of adipose tissue CETP and LPL mRNA and hepatic CETP mRNAwere performed in Report IV with competitive RT-PCR. The sensitivity of the RT-PCRmethod depends on the strategy selected. PCR can be performed without a standard, andthe PCR product is quantified during the exponential phase of amplification. This strategyis not so sensitive as competitive PCR, in which an internal standard is used. Theinternal standard has the same primer-binding sites as the target (264).

In our study, a specific RNA standard for competitive RT-PCR was developed, and theuse of a RNA standard is generally regarded as an ideal way to increase the sensitivity ofthe competitive RT-PCR method. Our RNA standard contained a small deletion (48 bpdeletion for CETP RNA standard and 42 bp deletion for LPL RNA standard), making itpossible to separate the standard band from the target band on electrophoresis. Thespecificity of the method was verified by sequencing PCR products. Thus, the resultsobtained from RT-PCR are reliable. Because the sequence lengths of the standard and thetarget were almost equal, they were amplified in PCR with the same efficiency. Accordingto gel electrophoresis and the sequence of the PCR products, no signs of heterodimerswere found. The reproducibility of the RT-PCR method was good because thecoefficients of variation in the analysis ranged from 2 to 14%, which values are betterthan those published earlier (264).

60

6.1.4. Meta-analysis

Meta-analysis is defined as a quantitative analysis of primary data which yields an overallsummary statistics and is, to some degree, more objective and more quantitative than aconventional review (265). By using meta-analysis in analyzing the results of separatedietary studies, a large amount of information can be combined, and even though themethod is time-consuming, it is efficient and both quicker and less costly than a newdietary trial. Also, the possibilities to generalize the findings, to increase the statisticalpower and to reduce the risk of random and systematic errors are better in meta-analysiscompared to conventional review.

During the processing of the meta-analysis, the explanations for the inconsistenciesand contradictions in the data must be clarified. The background information from everystudy included is needed to minimize the clinical and statistical heterogeneity of the data.The sources of clinical heterogeneity in dietary studies are variations in patient selection(healthy / diseased), randomization, losses on follow-up (all data should be included withthe principle of an intention to treat), baseline differences in relevant variables(normocholesterolemia / hypercholesterolemia) and the duration of follow-up (couple ofweeks / several months). Statistical heterogeneity may be caused by clinical heterogeneityor methodological differences between studies. Also, some unknown or unrecordedvariables may cause statistical heterogeneity (266).

By using meta-analysis, new reliable findings can be achieved and some of them mayeven lead to new treatment practices, as was the case with the use of streptokinase inacute myocardial infarction (267). Anyway, it is critical to identify all relevant studies.With a computerized literature search, only about 50% of relevant studies are identified,partly because some of the journals are not indexed (268) or the relevant study is notindexed correctly. Also, the publication bias, i.e. the fact that some of the studies arenever published, mainly because of statistically insignificant result, may lead to biaseddata for meta-analyses. In a meta-analysis, it is assumed that when effect sizes areestimated, the regressions are linear. In the dietary intervention studies analyzed in ReportIII, the variables of interest were the plasma lipids and lipoproteins, which have a linearrelation to dietary changes. The genetic effect was also assumed to be linear.

6.2. Dietary effect on lipids and lipoproteins

The following lipid abnormalities are associated with obesity: increased concentrations ofplasma triglycerides and decreased plasma HDL cholesterol concentration (45). The plasmatriglycerides were slightly increased and the HDL cholesterol concentration was slightlyreduced among the obese gallstone patients (Report IV) compared with the normal-weightfemale subjects (Report I). The mechanism by which triglyceride levels are elevated is anincrease of the adipose triglyceride pool and overproduction of VLDL triglycerides.Several mechanisms may explain the reduced HDL cholesterol concentrations in obesity.Changes may be related to altered triglyceride metabolism, or excess adipose tissue mayremove HDL from circulation and thereby lower serum HDL levels (96). In addition,

61

hyperinsulinemia and insulin resistance, which are often associated with obesity, maycontribute to abnormalities in plasma lipids and lipoproteins (269).

In Reports I-III, the dietary intervention was designed to lower the plasma totalcholesterol concentration with a low-fat diet and, on the other hand, to increase it with ahigh-fat diet. Dietary changes were achieved by changing the amount of ordinaryvegetables, whole-grain products and some vegetable oils and by changing the amount offatty dairy products and cold meats in the diets. This was a controlled study, and thefeasibility of the diets can be considered good. The observed changes in plasma lipids andlipoproteins can be considered to be caused only by the dietary changes because nosignificant changes of body weights were noticed.

Excessive dietary cholesterol and saturated fat leads to an increased plasma cholesterolconcentration, and decreased dietary fat and cholesterol leads to a decreased plasmacholesterol concentration. In addition to the reduction of dietary saturated fat andcholesterol, the low-fat diet included other changes leading to reduced plasma total andLDL cholesterol: the amount of MUFA, PUFA and dietary fiber increased. On the high-fat diet, opposite changes took place.

According to the empirical formulas (104,186,187) designed to estimate thequantitative effects of dietary changes, the observed changes in plasma lipids andlipoproteins (Reports I-III) were even greater than expected in our study population. Infact, during the high-fat diet, the increase of the plasma LDL cholesterol concentration inour study group was greater (39%) than in the dietary studies relevant for the meta-analysis (3-35%). A probable explanation could be the ideal study diets and also somegenetic properties of the study subjects.

The reduction of dietary total fat and cholesterol and the replacement of SAFA withPUFA were accompanied by a reduced plasma HDL cholesterol concentration (98, 101).This also happened during the low-fat diet (Report I). The diet-induced reduction of theplasma HDL cholesterol concentration may be considered harmless, especially if theLDL/HDL ratio remains constant (101).

There are several mechanisms by which the diet alters lipid metabolism and theplasma lipid levels. Raising the dietary cholesterol intake suppresses the activity of LDLreceptors and simultaneously increases the hepatic synthesis of apoB-containinglipoproteins (96, 270). The mechanism by which dietary SAFAs manifest theircholesterol-raising effect is not completely understood, but SAFAs seem to interfere withLDL-receptor mediated clearance of LDL (96). The cholesterol-lowering effect of MUFAsis unknown, but it can be the release of suppression of SAFA on LDL receptor activity.The reducing effect of PUFAs on plasma triglycerides is probably mediated by inhibitionof the hepatic synthesis of apoB-containing lipoproteins and the cholesterol-reducingeffect by increasing LDL receptor activity. Dietary carbohydrates increase triglycerideconcentrations (271, 272), and diets high in carbohydrates and low in fat have theparadoxical action of raising plasma triglycerides (271-274). This was observed in ourlow-fat diet, too (Report I). Some of the mechanisms described above are mediated byaltered gene expression, e.g. dietary cholesterol inhibits the transcription of HMG-CoAreductase gene (275).

62

6.3. Role of apolipoprotein E polymorphism in the dietaryresponse

The interindividual variation of responsiveness to diet in plasma lipids can be partlyexplained by the basal plasma lipid levels and the age of the subjects. Also, some resultssupport the theory of differences in responsiveness between male and female subjects(189). A certain part of the serum lipoprotein response to dietary manipulation have agenetic component. The polymorphisms of the candidate genes can be used to study theinteractions between genes and environmental factors (276).

The genetic variation of apo E is associated with the atherogenic lipoproteinphenotype (56, 277). 20% of the variance in the apo E level and 14% of that in theplasma cholesterol level is determined by the apo E protein polymorphism (56), subjectswith the apo E4 phenotype having the highest initial cholesterol level. In theory, themechanism which could explain the eventual differences in dietary responsiveness is thedifference in the fractional efficiency of cholesterol absorption from the gut between apoE3 and apo E4 subjects (197-200). In addition, the apo E4 affinity to LDL receptors maybe higher, leading to larger up-regulation of the LDL receptor than for apo E3 or apo E2.Also, differences in apoB production, bile acid synthesis, cholesterol synthesis andpostprandial lipoprotein clearance could explain the differences in the response to dietchanges, but final confirmation is still missing.

The apo E phenotype was not associated with dietary responsiveness of plasma lipidsand lipoproteins in our trial (Report I). This finding has been confirmed by other trials(278-288), although some reports showed greater responsiveness in subjects with the ε4allele (289-298, 199). The differences of basal lipids may cause some confounding effectwhen the role of apo E polymorphism and the response to diet are investigated becausethe basal cholesterol level contributes to the sensitivity to at least dietary cholesterol. Thebasal lipid values were equal among the subjects with the apo E3 and apo E4 phenotypesin our trial. In addition, the study design may have some impact on the outcome; thedesign of our study, and most of the other studies confirming our result later, has beenprospective, whereas the design in most of the intervention studies showing differencesin responsiveness between the apo E phenotypes, has been retrospective. On thecontrary, a recent study by Sarkkinen et al. (298) was prospective, and the effect of dietaryfat and cholesterol modification were studied separately. The finding of this study was thatthe apo E polymorphism did have an effect on lipid response to both modification ofdietary fat and cholesterol, E4/4 subjects having the greatest response in plasma totalcholesterol.

It has been hypothesized that because the genetic variation is not independent ofenvironmental and ethnic factors, the association between apo E4 and the elevated plasmatotal cholesterol concentration is expressed especially in populations consuming diets richin SAFA and cholesterol (299). According to this hypothesis, the ε4 allele effect wouldbe expressed especially during dietary interventions in which the amounts of SAFA anddietary cholesterol are increased simultaneously. Our finding did not confirm thishypothesis. Also, the modification of dietary total fat irrespective of dietary cholesterol ordietary fat saturation has been suggested to be an important way to reveal apo E4-relatedhyperresponsiveness (299).

63

Postprandial lipoprotein responses to fat ingestion may play a significant role inatherogenesis, and there are some reports showing differences in the postprandialresponses between subjects with ε3 and ε4 alleles (300, 301), suggesting that the ε4allele may be associated with accumulation of chylomicron remnants postprandially. Onthe other hand, the ε4 allele is associated with hyporesponsiveness to HMG-CoAreductase inhibitor therapy (302, 303).

Our result show (Report II) that a significant diet-induced alteration in blood pressurecan be achieved, the response being more marked among apo E4 subjects than apo E3subjects. There is some evidence of an interaction between blood pressure and the plasmacholesterol concentration (123, 304). Because the correlation between blood pressure andthe plasma cholesterol concentration was stonger in subjects with the apo E4 phenotypethan in subjects with the apo E3 phenotype, it can be hypothesized that apo E may act asa link between blood pressure and the plasma cholesterol concentration. Anyway, theexact mechanism is currently unknown.

6.4. Role of apolipoprotein B DNA polymorphism on thedietary response

Another logical candidate gene affecting the dietary responsiveness is the apo B gene. Itscommon variations have been determined in several populations, and their associationswith differences in plasma lipid levels are well established. The apo B gene has severalpolymorphisms, some of them causing amino acid changes, which can then alter thetranscription of the gene or affinity of LDL to its receptor. The secretion of apo B canalso be altered, and its structural stability or the interactions of apo B-containinglipoproteins with other lipoproteins, cells, or enzymes may be altered. The XbaIpolymorphism explains 3-8% of the variance in plasma cholesterol, but the functionalsequence variation in the apo B gene is not the XbaI polymorphism, because it does notchange the amino acid sequence. Thus, the functional sequence variation may be inlinkage disequilibrium with the XbaI polymorphism.

Several studies elucidated the role of the apo B polymorphism in response to diet(278, 286, 287, 294, 305-313), and showed an association in some (278, 287, 294, 305-309, 312) but not all cases (286, 310, 311, 313). A meta-analysis conducted on therelevant data did not confirm the role of the apo B XbaI polymorphism (Report III),although meta-analysis supported the finding of the significant role of the EcoRI andMspI polymorphisms in response to diet. The major problems when combining data fromdifferent dietary trials are the variations in the study groups (ages, genetic backgrounds),the differences in cultural backgrounds, the dietary protocols, and the difficulties to ensurecompliance in each trial. Taken together, the data suggest that although the effectsassociated with the apo B genotype are likely to make a small contribution, it is clear thatthe genes involved in determining the strong genetic effects that have been observed in thedietary response have yet to be identified. Further studies are needed to provide a betterunderstanding of the genetic component of the dietary response and to help to identifyindividuals who would benefit from dietary intervention.

64

6.5. CETP and the diet

LDL formation requires lipolysis of most VLDL triglyceride, transfer of freecholesterol and phospholipid to HDL, loss of all VLDL proteins except apo B, and asignificant increase in the cholesteryl ester content in VLDL. Cholesteryl ester istransferred from HDL in exchange for LDL triglyceride. The exchange reaction is catalysedby plasma CETP. A contribution of CETP activity to differences in plasma LDLcholesterol between individuals was suggested (314), and several dietary studies providedata supportive of the contribution of CETP to diet-induced changes in LDL cholesterollevels (215, 216, 213). According to these studies, CETP could be directly responsible forabout one third of the increase observed in LDL cholesterol in response to dietarysaturated fat. Trans-fatty acids also increase CETP activity (315). In our dietaryintervention the plasma CETP activity was increased upon a shift from a low-fat diet to ahigh-fat diet by 36±20 nmol/ml/h, and it correlated with the increase in plasma LDLcholesterol (r=0.30, p<0.05). Dietary cholesterol and fatty acids regulate the hepatic geneexpression of the CETP, which is the main regulator of the plasma CETP activity andconcentration.

Plasma CETP activities in obese subjects are increased and may explain, at leastpartly, the lipid abnormalities observed in obesity. During weight reduction somebeneficial effects on plasma lipids and lipoproteins are usually observed, and a meta-analysis has yielded quantitative estimate (50). In the weight reduction trial, the plasmaCETP activity decreased during the first two weeks of caloric restriction, but returned tothe basal level at the end of the two-month intervention period. The CETP mRNAamount was also almost the same among the control subjects and the weight-reducedsubjects when measured at the end of the study, and even plasma LDL cholesterol levelsremained unchanged during a moderate weight reduction. Only plasma triglycerides werereduced significantly. Adipose tissue CETP mRNA did not correlate with plasma CETPactivity, but hepatic CETP mRNA correlated with plasma CETP activity. Thus, adiposetissue CETP expression may have a lesser role in lipid abnormalities in obesity.

6.6. LPL gene expression and weight reduction

LPL regulates the size, density and composition of HDL and triglyceride-containinglipoproteins. The fasting plasma and adipose tissue LPL activities are increased inobesity. LPL is responsible for the extrahepatic hydrolysis of circulating triglycerides.LPL-mediated lipolysis of triglycerides provides a large amount of free fatty acids toextrahepatic tissues. The main sites for LPL production are adipose tissue and muscle. Inaddition to its hydrolytic activity, LPL might play an important role as a ligand orbridging factor for lipoprotein receptors, thereby facilitating the cellular uptake oflipoprotein particles by cells (316). LPL has been thought to play a role in the initiationand development of adiposity, and variation of LPL activity in various adipose depotswas shown to correlate with regional differences in the development of obesity (317).

65

Thus, the tissue-specific changes in LPL activity may have major effects on lipoproteinand energy metabolism.

The effect of weight reduction on plasma and adipose tissue LPL activities has beenvariable and seems to depend on the degree of obesity, the amount of weight lost and thephase of weight reduction. Also, the dietary change to achieve weight reduction may playa role. In a Finnish study, plasma and adipose tissue LPL activities were reduced by 50%during moderate weight reduction when the stable weight maintaining phase was notreached (318). During the caloric restriction in Report IV, plasma triglycerides werereduced significantly, and the LPL mRNA mainly from the subcutaneous adipose tissuedecreased among the intervention subjects compared to the controls. The weight reductionachieved correlated significantly with the LPL mRNA level. Extensive weight reductionwas achieved during a relatively short period with caloric restriction in the studies by Ongand Kern (46, 319). In these studies, at the post-weight-reduction stable phase, increasedexpression and activity of LPL was observed, and the finding was explained by themechanism in obesity-prone subjects to regain body weight.

7. Conclusions

The present study was designed to evaluate the magnitude of the response in plasma lipidsand lipoproteins in genetically selected subjects. The roles of specific genetic variantswere studied to find out possible indicators of responsive and non-responsive groups.Also, the effect of dietary modifications on the blood pressure level in subjects withnormal blood pressure was studied and the role of genetic variation was evaluated. Inaddition, to study the role of altered gene expression in lipoprotein metabolism duringdietary manipulation, a weight reduction trial was conducted.

The study subjects were healthy, middle-aged men and women in northern Finland(Reports I-III) and obese women with gallstone disease who were otherwise healthy(Report IV), and the meta-analysis data were obtained from studies conducted on healthymen and women who volunteered for dietary intervention trials in southern Finland,eastern Finland, Canada, New Zealand, Israel, USA and Great Britain.

The diets (Reports I-III) were designed based on ordinary food items: in the low-fat dietthe current recommendations concerning the amount of total fat, the fatty acidcomposition and the amount of dietary cholesterol were taken into consideration. In thehigh-fat diet, the fat content, fatty acid composition and amount of dietary cholesterolwere nearly the same as in ordinary Finnish food. The caloric restriction (Report IV) wasbased on dietary counselling, the energy content was advised to be 5.0 MJ daily, and thegoal to reduce weight was 0.5 kg of body weight each week, which was quite moderate.

In conclusion, the plasma total and LDL cholesterol concentrations can be changed bydietary modifications significantly in community-living, healthy subjects. The individualvariation in the genetically relatively homogeneous group was high, as it has been shownto be in other populations, too. The role of environmental factors in determining theplasma total and LDL cholesterol variation in a population is important; 60-70% of thevariation is regulated by environmental factors, the most important of them being thediet. The genetic variation of apolipoprotein E and apolipoprotein B are likely to makeonly a small contribution in the response to diet, and the genes involved in determiningthe major genetic effects have yet to be identified. The body weight reduction isaccompanied by a reduction of plasma triglycerides and serum insulin. Simultaneously,the gene expression of lipoprotein lipase is reduced. These changes are beneficial andencourage physicians to recommend moderate weight reduction programs for obesesubjects.

67

Combination of data from separate studies is one way to save the limited resourcesand could be ideal in finding out the genetic factors regulating the response to diet. Tomake this possible, the collaboration between research centers should be easy and open,and the study plans should take into consideration all possible confounding factors. Thenutritional regulation of gene expression is one of the important topics at the moment. Inthe future, we hope to have better information on the transcriptional regulation ofdifferent genes by specific nutrients.

8. References

1. Fojo SS (1992) Hypertriglyceridaemia due to genetic defect in lipoprotein lipase andapolipoprotein C-II. J Intern Med 231:669-677.

2. Schumaker V & Lambertas A (1992) Lipoprotein metabolism: chylomicrons, very-low-density lipoproteins and low-density lipoproteins. In: Lusis AJ, Rotter JI,Sparkes RS (ed) Molecular genetics of coronary artery disease. Monographs inhuman genetics 14:98-139. Karger, Basel.

3. Murdoch SJ (1995) Influence of lipoprotein lipase on the transformation of VLDLand HDL during lipolysis of VLDL. Atherosclerosis 118:193-212.

4. Brown MS & Goldstein JL (1986) A receptor-mediated pathway for cholesterolhomeostasis. Science 232:34-47.

5. Kesäniemi YA, Beltz WF & Grundy SM (1985) Comparisons of metabolism ofapoprotein B in normal subjects, obese patients, and patients with coronary heartdisease. J Clin Invest 76:586-595.

6. Steinberg D, Parthasarathy S, Carew TE, Khoo JC & Witztum JL (1989) Beyondcholesterol: Modifications of low-density lipoprotein that increase its atherogenicity.N Engl J Med 320:915-924.

7. Witztum JL & Steinberg D (1991) Role of oxidized low density lipoprotein inatherogenesis. J Clin Invest 88:1785-1792.

8. Steinberg D (1997) Low density lipoprotein oxidation and the pathobiologicalsignificance. J Biol Chem 272:20963-20966.

9. Brewer HB Jr & Rader DJ (1991) HDL: structure, function and metabolism. ProgLipid Res 30:139-144.

10. Tall AR. Plasma high density lipoproteins. Metabolism and relationship toatherogenesis. J Clin Invest 1990;86:379-384

11. Jonas A (1991) Lecithin-cholesterol acyltransferase in the metabolism of high-density lipoproteins. Biochim Biophys Acta 1084:205-220.

12. Swenson TL (1991) The role of the cholesterol ester transfer protein protein inlipoprotein metabolism. Diabetes Metab Rev 7:139-153.

13. Pussinen P (1997) Plasma phospholipid transfer protein (PLTP) in high densitylipoprotein (HDL) metabolism. National Public Health Institute A7. Helsinki,Finland.

69

14. Neaton JD, Blackburn H, Jacobs D, Kuller L, Lee DJ, Sherwin R, Shih J, Stamler J& Wentworth D (1992) Serum cholesterol level and mortality findings for menscreened in the Multiple Risk Factor Intervention Trial. Arch Intern Med 152:1490-1500.

15. Smith GD, Shipley MJ, Marmot MG & Rose G (1992) Plasma cholesterol andmortality: the Whitehall Study. JAMA 267:70-76.

16. Chen Z, Peto R, Collins R, MacMahon S, Lu J & Li W (1991) Serum cholesterolconcentration and coronary heart disease in a population with low cholesterolconcentrations. Br Med J 303:276-282.

17. Keys A (1970) Coronary heart disease in seven countries. Circulation 41(Suppl 1):1-211.

18. Kannel WB, Castelli WP & Gordon T (1979) Cholesterol in the prediction ofatherosclerotic disease: new perspectives based on the Framingham study. AnnIntern Med 90:85-91.

19. West of Scotland Coronary Prevention Study Group (1998) Influence of pravastatinand plasma lipids on clinical events in the West of Scotland Coronary PreventionStudy (WOSCOPS). Circulation 97:1440-1445.

20. Scandinavian Simvastatin Survival Study Group. Pedersen TR, Olsson AG,Faergeman O, Kjekshus J, Wedel H, Berg K, Wilhelmsen L, Haghfelt T,Thorgeirsson G, Pyörälä K, Miettinen T, Christophersen B, Tobert JA, MuslinerTA & Cook TJ (1998) Lipoprotein changes and reduction in the incidence of majorcoronary heart disease events in the Scandinavian Simvastatin Survival Study (4S).Circulation 97:1453-1460.

21. Law MR, Wald NJ & Thompson SG (1994) By how much and how quickly doesreduction in serum cholesterol concentration lower risk of ischaemic heart disease?Br Med J 308:367-372.

22. McGill HC Jr, McMahan CA, Kruski AW & Mott GE (1981) Relationship oflipoprotein cholesterol concentrations to experimental atherosclerosis in baboons.Arteriosclerosis 1:3-12.

23. Scientific Steering commitee on behalf of the Simon Broome Register Group (1991)Risk of fatal coronary heart disease in familial hypercholesterolaemia. Br Med J303:893-896.

24. Pathobiological determinants of atherosclerosis in youth (PDAY) Research Group(1990) Relationship of atherosclerosis in young men to serum lipoproteincholesterol concentrations and smoking. JAMA 264:3018-3024.

25. Brown MS & Goldstein JL (1983) Lipoprotein metabolism in the macrophage:implications for cholesterol deposition in atherosclerosis. Annu Rev Biochem52:223-261.

26. Nguyen TT (1998) Cholesterol-lowering trials and coronary heart disease in theUnited States. In: Roberts WO (ed) Managing high cholesterol. A postgraduatemedicine special report. Nov: 15 – 26. The McGraw–Hill Companies, Inc.Minneapolis.

27. Wood D, De Backer G, Faergeman O, Graham I, Mancia G & Pyörälä K (1998)Prevention of coronary heart disease in clinical practice: Recommendations of theSecond Joint Task Force of European and other Societes of Coronary Prevention.Atherosclerosis 140:199-270.

70

28. Abbott RD, Wilson PW, Kannel WB & Castelli WP (1988) High densitylipoprotein, total cholesterol screening, and myocardial infarction. The FraminghamStudy. Arteriosclerosis 8:207-211.

29. Barter PJ & Rye KA (1996) High density lipoproteins and coronary heart disease.Atherosclerosis 121:1-12.

30. Schmitz G & Lackner KJ (1993) High-density lipoproteins and atherosclerosis. CurrOpin Lipidol 4:392-400.

31. Banka CL (1996) High density lipoprotein and lipoprotein oxidation. Curr OpinLipidol 7:139-142.

32. Kesäniemi YA (1998) Relevance of the reduction of triglycerides in the preventionof coronary heart disease. Curr Opin Lipidol 9:571-574.

33. Dieplinger H & Kronenberg F (1999) Genetics and metabolism of lipoprotein(a) andtheir clinical implications (Part 1) Wien Klin Wochenschr 111:5-20.

34. Zock PL & Mensink RP (1996) Dietary trans-fatty acids and serum lipoproteins inhumans. Curr Opin Lipidol 7:34-37.

35. Paassilta M (1998) Lipoprotein(a) levels: modifications by alcohol, hormones andthrombolytic treatment. Acta Univ Oul D 465. Oulu, Finland.

36. Sniderman AD, Shapiro S, Marpole DG, Skinner BF, Teng B & Kwiterovich PO Jr(1980) Association of coronary atherosclerosis with hyperbetalipoproteinemia(increased protein but normal cholesterol levels in human plasma low-density (B)lipoproteins. Proc Natl Acad Sci 77:604-608.

37. Kannel WB (1996) Blood pressure as a cardiovascular risk factor: prevention andtreatment. JAMA 275:1571-1576.

38. Collins R, Peto R, MacMahon S, Hebert P, Fiebach NH, Eberlein KA, Godwin J,Qizilbash N, Taylor JO & Hennekens CH (1990) Blood pressure, stroke andcoronary heart disease. Part 2, Short term reductions in blood pressure: overview ofrandomized drug trials in their epidemiological context. Lancet 335:827-838.

39. Jung RT (1997) Obesity as a disease. Br Med Bull 53:307-321. 40. Jousilahti P, Tuomilehto J, Vartiainen E, Pekkanen J & Puska P (1996) Body

weight, cardiovascular risk factors, and coronary mortality. 15-year follow-up ofmiddle-aged men and women in Eastern Finland. Circulation 93:1372-1379.

41. Shaper AG, Wannamethee SG & Walker M (1997) Body weight: implication for theprevention of coronary heart disease, stroke, and diabetes mellitus in a cohort studyof middle-aged men. Br Med J 314:1311-1317.

42. Kannel WB & Gordon,T (1979) Physiological and medical concomitans of obesity:The Framingham Study. In: Bray G A (ed) Obesity in America 359:125-163. NIHPublication, Washington, DC.

43. Despres J-P (1994) Dyslipidaemia and obesity. Bailliere’s Clinical Endocrinologyand Metabolism 8: 629-660.

44. Taskinen MR & Nikkilä EA (1977) Lipoprotein lipase activity in adipose tissue andin postheparin plasma in human obesity. Acta Med Scand 202:399-408.

45. Bosello O, Cigolino M, Battaggia A, Ferrari F, Micciolo R, Olivetti R & CorsatoM (1984) Adipose tissue lipoprotein-lipase activity in obesity. Int J Obes 8:213-220.

71

46. Ong JM & Kern PA (1989) Effect of feeding and obesity on lipoprotein lipaseactivity, immunoreactive protein, and messenger RNA levels in human adiposetissue. J Clin Invest 84:305-311.

47. Arai T, Yamashita S, Hirano K, Sakai N, Kotani K, Fujioka S, Nozaki S, Keno Y,Yamane M & Shinohara E (1994) Increased plasma cholesteryl ester transfer proteinin obese subjects. A possible mechanism for the reduction of serum HDLcholesterol levels in obesity. Arterioscler Thromb 14:1129-1136.

48. Hayes KC, Livingstone A & Trautwein EA (1992) Dietary impact on biliary lipidsand gallstones. Annu Rev Nutr 12:229-326.

49. Leive A (1998) What are the benefits of moderate weight loss? Exp Clin EndocrinolDiab 106:10-13.

50. Dattilo AM & Kris-Etherton PM (1992) Effects of weight reduction on blood lipidsand lipoproteins: a meta-analysis. Am J Clin Nutr 56:320-328.

51. Karvonen MK, Pesonen U, Koulu M, Niskanen L, Laakso M, Rissanen A, DekkerJM, Hart LM, Valve R & Uusitupa MI (1998) Association of a leucine (7)-to-proline (7) polymorphism in the signal peptide of neuropeptide Y with high serumcholesterol and LDL cholesterol levels. Nature Med 4:1434-1437.

52. Funke H & Assman G (1999) Strategies for the assessment of genetic coronaryartery disease risk. Curr Opin Lipidol 10:285-291.

53. Berg K (1986) DNA polymorphism at the apolipoprotein B locus is associated withlipoprotein level. Clin Genet 30:515-520.

54. Ilmonen M (1999) Genetic variation of apolipoprotein B in the Finns: Effects onserum lipid levels. Univ. Helsinki, Dept of Medicine, Helsinki, Finland.

55. de Knijff P, Van den Maagdenberg AM, Frants RR & Havekes LM (1994) Geneticheterogeneity of apolipoprotein E and its influence on plasma lipid and lipoproteinlevels. Hum Mutation 4:178-194.

56. Davignon J, Gregg RE & Sing CF (1988) Apolipoprotein E polymorphism andatherosclerosis. Atherosclerosis 8:1-21.

57. Lenzen HJ, Assman G, Buchwalsky R & Schulte H (1986) Association ofapolipoprotein E polymorphism, low-density lipoprotein cholesterol and coronaryartery disease. Clin Chem 32:778-781.

58. De Knijff P & Havekes LM (1996) Apolipoprotein E as a risk factor for coronaryheart disease: a genetic and molecular biology approach. Curr Opin Lipidol 7:59-63.

59. Wilson PWF, Schaefer EJ, Larson MG & Ordovas JM (1996) Apo E allele and riskof coronary disease. A meta-analysis. Arterioscler Thromb Vasc Biol 16:1250-1255.

60. Kannel WB (1981) Update on the role of cigarette smoking in coronary arterydisease. Am Heart J 101:319-328.

61. Lakier JB (1995) Smoking and cardiovascular disease. Am J Med 93(Suppl):8-12. 62. Freeman DJ & Packard CJ (1995) Smoking and plasma lipoprotein metabolism.

Clin Sci 89:333-342. 63. Lassila R, Seyberth HW, Haapanen A, Schweer H, Koskenvuo M, Laustiola KE

(1988) Vasoactive and atherogenic effects of cigarette smoking: a study ofmonozygotic twins discordant for smoking. Br Med J 297:955-957.

64. Heitzer T, Ylä-Herttuala S, Luoma J, Kurz S, Manzel T, Just H, Olschewski M &Drexler H (1996) Cigarette smoking potentiates endothelial dysfunction of forearm

72

resistance vessels in patients with hypercholesterolemia. Role of oxidized LDL.Circulation 93:1346-1353.

65. Reilly M, Delanty N, Lawsin JA & FitzGerald GA (1996) Modulation of oxidantstress in vivo in chronic cigarette smokers. Circulation 94:19-25.

66. Dullaart RPF, Hoogenberg K, Dikkeschei BD & Van Tol A (1994) Higher plasmalipid transfer protein activities and unfavorable lipoprotein changes in cigarettesmoking men. Arterioscler Thromb Vasc Biol 14:1581-1585.

67. Thun MJ, Peto R, Lopez AD, Monaco JH, Henley SJ, Heath CW & Doll R (1997)Alcohol consumption and mortality among middle-aged and elderly U.S. adults. NEngl J Med 337:1705-1714.

68. Klatsky AL, Armstrong MA & Friedman GD (1990) Risk of cardiovascularmortality in alcohol drinkers, ex-drinkers and non-drinkers. Am J Cardiol 66:1237-1242.

69. Savolainen MJ & Kesäniemi YA (1995) Effects of alcohol on lipoproteins inrelation to coronary heart disease. Curr Opin Lipidol 6:243-250.

70. Porkka KVK & Ehnholm C (1996) Smoking, alcohol and lipoprotein metabolism.Curr Opin Lipidol 7:162-166.

71. McKenzie CR & Eisenberg PR. (1996) Alcohol, coagulation and arterialthrombosis. In: Zakhari S, Wassef M. (eds.) Alcohol and cardiovascular system.Research monograph. 31(No. 96-4133):413-439. NIH Publication, Washington.DC.

72. Elwood PC, Hughes C & O’Brien JR (1998) Platelets, aspirin, and cardiovasculardisease. Postgrad Med J 74:587-591.

73. Fuhrman B, Lavy A & Aviram M (1995) Consumption of red wine with mealsreduces the susceptibility of human plasma and low-density lipoprotein to lipidperoxidation. Am J Clin Nutr 61:549-554.

74. Reinke LA & McCay PB (1996) Interaction between alcohol and antioxidants. In:Zakhari S, Wassef M (eds) Alcohol and cardiovascular system. Research monograph.31(No. 96-4133):441-452. NIH Publication, Washington. DC.

75. Blair SN (1995) Changes in physical fitness and all-cause mortality: A prospectivestudy of healthy and unhealthy men. JAMA 273:1093-1098.

76. Rosengren A & Wilhelmsen L (1997) Physical activity protects against coronarydeath and deaths from all causes in middle-aged men. Evidence from a 20-yearfollow-up of the primary prevention study in Göteborg. Ann Epidemiol 7:69-75.

77. Wannamethee SG, Sharper AG & Walker M (1998) Changes in physical activity,mortality, and incidence of coronary heart disease in older man. Lancet 351:1603-1608.

78. Goldbourt U (1997) Physical activity, long-term CHD mortality and longevity: areview of studies over the last 30 years. World Rev Nutr Diet 82:229-239.

79. Leon AS, Myers MT & Connett J (1997) Leisure time physical activity and the 16-year risks of mortality from coronary heart disease and all-causes in the MultipleRisk Factor Intervention Trial (MRFIT). Int J Sports Med 18(Suppl 3):208-215.

80. Anitschkow N (1913) Uber die veränderungen der kaninchenaorta bei experimentellercholesterinsteatose. Beitr Path Anat Allg Path 56:379-404.

81. McNamara DJ (1990) Relationship between blood and dietary cholesterol. In:Pearson AM, Dutson TR (eds) Meat and health. Advances in meat research. 6:63-87.Elsevier, London, England.

73

82. Dayton S, Pearce ML, Hashimoto S, Dixon WJ & Tomiyasy U (1969) AControlled clinical trial of a diet high in unsaturated fat in preventing complicationsof atherosclerosis. Circulation 39/40(Suppl II):1-63.

83. Turpeinen O, Karvonen MJ, Pekkarinen M, Miettinen M, Elosuo R & PaavilainenE (1979) Dietary prevention of coronary health disease: the Finnish Mental HospitalStudy. Int J Epidemiol 8:99-118.

84. Hjermann I, Velve BK, Holme I & Leren P (1981) Effect of diet and smokingintervention on the incidence of coronary heart disease. Lancet 8259:1303-1310.

85. Puska P, Salonen JT, Nissinen A, Tuomilehto J, Vartiainen E, Korhonen H,Tanskanen A, Rönnqvist P, Koskela KAJ & Huttunen J (1983) Change in riskfactors for coronary heart disease during 10 years of a community interventionprogramme (North Karelia Project). Br Med J 287:1840-1844.

86. Multiple Risk Factor Intervention Trial Research Group (1982) Multiple RiskFactor Intervention Trial. Risk factor changes and mortality results. JAMA248:1465-1477.

87. Multiple Risk Factor Intervention Trial Research Group (1990) Mortality after 101/2 years for hypertensive participants in the multiple risk factor intervention trial.Circulation 82:1616-1628.

88. Research Committee to the Medical Research Council (1965) Low-fat diet inmyocardial infarction - a controlled trial. Lancet 2:501-504.

89. Leren P (1970) The Oslo Diet Heart Study. Eleven-year report. Circulation 42:935-942.

90. Arntzenius AC, Kromhout D, Barth JD, Reiber JHC, Bruchke AVG, Buis B, vanGent CM, Kempen-Voogd N, Strikwerda S & van der Velde EA (1985) Diet,lipoproteins, and the progression of coronary atherosclerosis. The LeidenIntervention Trial. N Engl J Med 312:805-811.

91. Ornish D, Brown SE, Scherwitz LW, Billings JH, Armstrong WT, Ports TA,McLanahan SM, Kirkeeide RL, Brand RJ & Gould KL (1990) Can lifestyle changesreverse coronary heart disease? The Lifestyle Heart Trial. Lancet 336:129-133.

92. Gould KL, Ornish D, Scherwitz L, Brown S, Edens RP Hess MJ, Mullani N,Boloney L, Dobbs F & Armstrong WT (1995) Changes in myocardial perfusionabnormalities by position emission tomography after long-term, intense risk factormodification. JAMA 274:894-901.

93. Watts GF, Lewis B, Brunt JN, Lewis ES, Coltart DJ, Smith LD & Swan AV(1992) Effects on coronary artery disease of lipid-lowering diet, or diet pluscholestyramine, in the St Thomas’ Atherosclerosis Regression Study (STARS).Lancet 339:563-569.

94. Miettinen TA, Gylling H (1999) Regulation of cholesterol metabolism by dietaryplant sterols. Curr Opin Lipidol 10:9-14.

95. Clarke R, Frost C, Collins R, Appleby P & Peto R (1997) Dietary lipids and bloodcholesterol: quantitative meta-analysis of metabolic ward studies. Br Med J 314:112-117.

96. Grundy SM & Denke MA (1990) Dietary influences on serum lipids andlipoproteins. J Lipid Res 31:1149-1172.

97. Schaefer EJ, Lichtenstein AH, Lamon-Fava S, McNamara J & Ordovas JM (1995)Lipoproteins, nutrition, aging, and atherosclerosis. Am J Clin Nutr 61(Suppl):726-740.

74

98. Schaefer EJ & Brousseau ME (1998) Diet, lipoproteins, and coronary heart disease.Endocrinology and metabolism . Clinics of North America 27:711-733.

99. Keys A, Anderson JT & Grande F (1957) Prediction of serum cholesterol responsesof man to changes in fats in the diet. Lancet 2:959-966.

100. Hegsted DM, McGandy RB, Myers ML & Stare FJ (1965) Quantitative effects ofdietary fat on serum cholesterol in man. Am J Clin Nutr 17:281-295.

101. Mattson FH & Grundy SM (1985) Comparison of effects of dietary saturated,monounsaturated, and polyunsaturated fatty acids on plasma lipids and lipoproteinsin man. J Lipid Res 26:194-202.

102. Mann GV (1994) Metabolic consequences of dietary trans fatty acids. Lancet343:1268-1271.

103. Harris WS (1996) Dietary fish oil and blood lipids. Curr Opin Lipidol 7:3-7.104. Mensink RP & Katan MB (1992) Effect of dietary fatty acids on serum lipids and

lipoproteins: A meta-analysis of 27 trials. Atheroscler Thromb 12:911-919.105. Lussier-Cacan S,Dubreuil-Quidoz S, Roederer G, Leboneuf N, Boulet L,

deLangavant GC, Davignon J & Naruszewicz M (1993) Influence of probucol onebhanced LDL oxidation after fish oil treatment on hypertriglyceridemic patients.Arterioscler. Thromb 13:1790-1797.

106. Reaven P, Parthasarathy S, Grasse BJ, Miller E, Steinberg D & Witztum JL (1993)Effects of oleate-rich and linoleate-rich diets on the susceptibility of low densitylipoprotein to oxidative modification in mildly hypercholesterolemic subjects. JClin Invest 91:668-676.

107. Brown L, Rosner B, Willett W & Sacks FM (1999) Cholesterol-lowering effects ofdietary fiber: a meta-analysis. Am J Clin Nutr 69:30-42.

108. Hannuksela M, Marcel YL, Kesäniemi YA & Savolainen MJ (1992) Reduction inthe concentration and activity of plasma cholesteol ester transfer protein by alcohol.J Lipid Res 33:737-744

109. Intersalt Cooperative Research Group. (1988) Intersalt: an international study ofelectrolyte excretion and blood pressure. Result for 24 hour urinary sodium andpotassium excretion. Br Med J 297:319-328.

110. Elliott P, Stamler J, Nichols R, Dyer AR, Stamler R, Kesteloot H & Marmot M(1997) Intersalt revisited: further analyses of 24 hour sodium excretion and bloodpressure within and across populations. Intersalt Cooperative Research Group. BrMed J 312:1249-1253.

111. Law MR, Frost CD, Wald NJ (1991) By how much does dietary salt reduction lowerblood pressure? III-Analysis of observational data among populations. Br Med J302:811-815.

112. Whelton PK, He J, Cutler JA, Brancati FL, Appel LJ, Follmann D & Klag MJ(1997) Effects of oral potassium on blood pressure. Meta-analysis of randomizedcontrolled clinical trials. JAMA 277:1624-1632.

113. McCarron DA, Morris CD, Henry HJ & Stanton JL (1984) Blood pressure andnutrient intake in the United States. Science 224:1392-1398.

114. Gruchow HW, Sobocinski KA & Barboriak JJ (1985) Alcohol, nutrient intake andhypertension in US adults. JAMA 253:1567-1570.

115. Havlik RJ, Hubert HB, Fabsitz RR & Feinleib M (1983) Weight and hypertension.Ann Intern Med 98:855-859.

75

116. Goldstein DJ (1992) Beneficial effects of moderate weight loss. Int J Obesit 16:397-415.

117. deLeiva A (1998) What are the benefits of moderate weight loss? Exp Clin EndocrinDiab106(Suppl) 2:10-13.

118. Marmot MG, Elliott P, Shipley MJ, Dyer AR, Ueshira H, Beevers DG, Stamler R,Kesteloot H, Rose G & Stamler J (1994) Alcohol and blood pressure: theINTERSALT study. Br Med J 308:1263-1267.

119. Kauma H, Savolainen MJ, Rantala AO, Lilja M, Kervinen K, Reunanen A &Kesäniemi YA (1998) Apolipoprotein E phenotype determines the effect of alcoholon blood pressure in middle-aged men. Am J Hypertens 11:1334-1343.

120. Salonen JT (1991) Dietary fats, antioxidants and blood pressure. Ann Med 23:295-298.

121. Beegom R & Singh RB (1997) Association of higher saturated fat intake withhigher risk of hypertension in an urban population of Trivandrum in south India. IntJ Cardiol 58:63-70.

122. Pauletto P, Puato M, Caroli MG, Gasiglia E, Muuhambo AE, Cazzolato G, BittoloBon G, Angeli MT, Galli C & Pessina AC (1996) Blood pressure and atherogeniclipoprotein profiles of fish-diet and vegetarian villagers in Tanzania: the Lugalawastudy. Lancet 348:784-788.

123. Goode GK, Miller JP & Heagerty AM (1995) Hyperlipidaemia, hypertension, andcoronary heart disease. Lancet 345:362-364.

124. Nami R, Gallo V, Pavese G, Panza F & Gennari C (1995) Antihypertensive activityof a vegetable fibre preparation: a preliminary, double-blind, placebo controlledstudy. Eur J Clin Nutr 49(Suppl1):S201-206.

125. Schlamowitz P, Halberg T, Warnoe O, Wilstrup F & Ryttig K (1987) Treatment ofmild to moderate hypertension with dietary fibre. Lancet 2:622-623.

126. Eliasson K, Ryttig KR, Hylander B & Rossner S (1992) A dietary fibre supplementin the treatment of mild hypertension. A randomized, double-blind, placebocontrolled trial. J Hypertens 10:195-199.

127. Preuss HG, Gondal JA & Lieberman S (1996) Association of macronutrients andenergy intake with hypertension. Am J Coll Nutr 15:21-35.

128. Prentice AM & Jobb SA (1995) Obesity in Britain: gluttony or sloth? Br Med J311:437-439.

129. Seidell JC (1998) Dietary fat and obesity: an epidemiologic perspective. Am J ClinNutr 67(Suppl3):546-550.

130. Romieu I, Willett WC, Stampfer MJ, Colditz GA, Sampson L, Rosner B,Hennekens CH & Speizer FE (1988) Energy intake and other determinants ofrelative weight Am J Clin Nutr 47:406-412.

131. Tremblay A, Plourde G, Despres JP & Bouchard C (1989) Impact of dietary fatcontent and fat axidation on energy intake in humans. Am J Clin Nutr 49:799-805.

132. Golay A & Bobbioni E (1997) The role of dietary fat in obesity. Int J Obes21(Suppl):2-11.

133. Donato K (1987) Efficiency and utilization of various energy sources for growth.Am J Clin Nutr 45:164-167.

76

134. Schutz Y, Flatt JP & Jequier E (1989) Failure of dietary fat intake to promote fatoxidation: a factor favoring the development of obesity. Am J Clin Nutr 50:307-314.

135. Horton TJ, Drousas H, Brachey A, Reed GW, Peters JC & Hill JO (1995) Fat andcarbohydrate overfeeding in humans: Different effects on energy storage. Am J ClinNutr 62:19-29.

136. Lean MJ& James WP (1988) Metabolic effects of isoenergetic nutrient exchangeover 24 hours in relation to obesity in women. Int J Obesit 12:15-27.

137. Astrup A (1993) Dietary composition substrates balances and body fat in subjectswith predisposition to obesity. Int J Obes Relat Metab Disord17:(Suppl 3)32-36.

138. Astrup A & Raben A (1996) Glucostatic control of intake and obesity. Proc NutrSoc 55:485-495.

139. Pate RR & Blair SN (1995) Physical activity and public health. A recommendationfrom the Centers for Disease Control and Prevention and the American College ofSports Medicine. JAMA 273:402-407.

140. Rolls B & Shide D (1992) The influence of dietary fat on food intake and bodyweight. Nutr Rev 50:283-290.

141. Blundell JE, Green S & Burley V (1994) Carbohydrates and human appetite. Am JClin Nutr 59(Suppl):728-734.

142. Mela D (1991) Sensory preference for fats: what, who, why? Food Quality Pref1:71-73.

143. Drewnovski A, Kurth C, Holden-Wiltse J & Saari J (1992) Food preferences andobesity: carbohydrates versus fats. Appetite 18:207-221.

144. Prusse L & Bouchard C (1994) Genetics of energy intake and food preferences. In:Bouchard C (ed). The Genetics of Obesity. Boca Raton: CRC press.

145. Bouchard C, Tremblay A, Despres JP, Nadeau F, Lupien PJ, Theriault G, DussaultJ, Moorjami S, Pinault S & Fournier G (1990) The response to long-termoverfeeding in identical twins. N Engl J Med 322:1477-1482.

146. Kim DN, Eastman A, Baker JE, Mastrangels A, Sethi S, Ross JS, Schnee J &Thomas WA (1995) Fish oil, atherogenesis, and thrombogenesis. Ann N Y AcadSci.748:474-480.

147. Larsen LF, Jespersen J & Marckmann P (1999) Are olive oil diets antithrombotic?Diets enriched with olive oil, rapeseed, or sunflower oil affect postprandial factor VIIdifferently. Am J Clin Nutr 70:976-982.

148. Perez-Jimenez F, Lopez-Miranda J, Paz-Rojas E, Lopez-Segura F, Velasco F, MarinC, Fuentes F & Ordovas JM (1999) Circulating levels of endothelial function aremodulated by dietary monounsaturated fat. Atherosclerosis, 145:351-358.

149. Rankinen T, Rauranmaa R, Väisänen S, Penttilä IM, Uusitupa M, Relation ofhabitual diet and cardiorespiratory fitness to blood coaculation and fibrinolyticfactors. Thromb Haemost 1994 71:180-183.

150. Vogel RA, Corretti MC & Gellman J (1998) Cholesterol, cholesterol lowering, andendothelial function. Prog Cardiovasc Dis 41:117-136.

151. Jenkinson A, Franklin MF, Wahle K & Duthie GG (1999) Dietary intakes ofpolyunsaturated fatty acids and indices of oxidative stress in human volunteers. Eur JClin Nutr 53:523-528.

77

152. Reaven P, Parthasarathy S, Grasse BJ, Miller E, Steinberg D & Witztum JL (1993)Effects of oleate-rich and linoleate-rich on the susceptibility of low densitylipoprotein to oxidative modification in mildly hypercholesterolemic subjects. JClin Invest. 91:668-676.

153. Reaven PD, Parthasarthy S, Grasse BJ, Miller E, Almazan F, Mattson FH, KhooJC, Steinberg D& Witztum JL (1991) Feasibility of using an oleate-rich diet toreduce the susceptibility of low-density lipoprotein oxidative modification inhumans. Am J Clin Nutr 54:701-706.

154. Bonamone A, Pagnan A, Biffanti S, Opportuno A, Sortago F, Dorella M, MaiorinoM & Ursini F (1992) Effects of dietary monunsaturated and polyunsaturated fattyacids on susceptibility of plasma low density lipoproteins to oxidative modification.Arterioscler Thromb 12:529-533.

155. Abbey M, Belling GB, Noakes M, Hirata F & Nestel PJ (1991) Oxidation of lowdensity lipoproteins: intraindividual variabilitu and the effect of dietary linoleatesupplementation. Am J Clin Nutr 57:391-398.

156. Berry EM, Eisenberg S, Friedlander Y, Harats D, Kaufmann NA, Norman Y &Stein Y (1992) Effects of diets rich in monounsaturated fatty acids on plasmalipoproteins—the Jerusalem Nutrition Study.II. Monounsaturated fatty acids vscarbohydrates. Am J Clin Nutr 56:394-403.

157. Aviram M & Elias K (1993) Dietary olive oil reduces low-density lipoproteinuptake by macrophages and decreases the susceptibility of the lipoprotein to undergolipid peroxidation. Ann Nutr Metab 37:75-84.

158. Harats D, Dabach Y, Hollander G, Ben-Naim M, Schwarz R, Berry EM, Stein O &Stein Y (1991) Fish oil ingestion in smokers and non-smokers enhancesperoxidation of plasma lipoproteins. Atherosclerosis 90:127-139.

159. Harris WS, Windsor SL & Caspermayer JJ (1993) Modification of lipid-related. JNutr Biochem 4:706-712.

160. Lussier-Cacan S, Dubreuil-Quidoz S, Roederer G, Leboeuf N, Boulet L,deLangavant GC, Davignon J & Naruszewicz M (1993) Influence of probucol onenhanced LDL oxidation after fish oil treatment of hypertriglyceridemic patients.Arterioscler Thromb 13:1790-1797.

161. Meydani M, Natiello F, Goldin B & Free N (1991) Effect of long-term fish oilsupplementation on vitamin E status and lipid peroxidation in women. J Nutr.121:484-491.

162. The Alpha-Tocopherol, Beta Carotene Cancer Prevention Study Group (1994) Theeffect of vitamin E and beta carotene on the incidence of lung cancer and othercancers in male smokers. N Engl J Med 330:1029-1035.

163. Rapola JM, Virtamo J, Ripatti S, Huttunen JK, Albanes D, Taylor PR & HeinonenOP (1997) Randomized trial of alpha-tocopherol and beta-carotene supplements onincidence of major coronary events in men with previous myocardial infarction.Lancet 349:1715-1720.

164. Rimm E, Williams P, Fosher K, Criqui M & Stampfer MJ (1999) Moderate alcoholintake and lower risk of coronary heart disease: meta-analysisi of effects on lipidsand haemostatic factors. Br Med J 319:1523-1528.

165. Fagrell B, De Faire U, Bondy S, Criqui M, Gaziano M, Gronback M, Jackson R,Klatsky A, Salonen J & Shaper AG (1999) The effects of light to moderate drinkingon cardiovascular diseases. J Intern Med 246:331-340.

78

166. Gudnason V, Thormar K & Humphries SE (1997) Interaction of the cholesteryl estertransfer protein I405V polymorphism with alcohol consumption in smoking andnon-smoking healthy men, and the effect on plasma HDL cholesterol and apo AIconcentration. Clin Genet 51:15-21.

167. Fumeron F, Betoulle D, Luc G, Behagere I, Ricard S, Pairier O, Jemaa R, Evans A,Arveller D & Marques-Vidal P (1995) Alcohol intake modulates the effect of apolymorphism of the cholesteryl ester transfer protein gene on plasma high densitylipoprotein and the risk of myocardial infarction. J Clin Invest 96:1664-1671.

168. Hannuksela ML, Liinamaa MJ, Kesäniemi YA & Savolainen MJ (1994) Relation ofpolymorphisms in the cholesteryl ester transfer protein gene to transfer proteinactivity and plasma lipoprotein levels in alcohol drinkers. Atherosclerosis;30:35-44.

169. Klatsky AL (1999) Blood pressure and alcohol intake. In: Izzo JL, Black HL (eds.)Hypertension Primer. Dallas: American Heart Association:164-169.

170. Klatsky AL (1996) Alcohol, coronary heart disease, and hypertension. Ann Rev Med47:149-160.

171. Hillbom M, Numminen H & Juvela S (1999) Recent heavy drinking of alcohol andembolic stroke. Stroke 30:2307-2312.

172. Boer JM, Kuivenhoven JA, Feskens EJ, Schauten EG, Havekes LM, Seidell JC &Kastelein JJ (1999) Physical activity modulates the effect of lipoprotein lipasemutation (D9N) on plasma lipids and lipoproteins. Clin Genet 56:158-163.

173. Humphries SE, Henry JA & Montgomery HE (1999) Gene-environment interactionin the determination of levels of haemostatic variables involved in thrombogenesisand fibrinolysis. Blood Coagul Fibrinolysis 10(Suppl):17-21.

174. Stamler J, Rains-Clearman D, Lenz-Litzow K, Tillotson JL & Grandits GA (1997)Relation of smoking at baseline and during trial years 1-6 to food and nutrientintakes in the special intervention and usual care groups in the Multiple Risk FactorIntervention Trial. Am J Clin Nutr 65(Suppl):374-402.

175. Heitzer T, Ylä-Herttuala S, Luoma J, Kurz S, Munzel T, Just H, Olschewski M &Drexler H (1996) Cigarette smoking potentiates endothelial dysfunction of forearmresistance vessels in patients with hypercholesterolemia. Role of oxidized LDL.Circulation 93:1346-1353.

176. Reilly M, Delanty N, Lawsin JA & FitzGerald GA (1996) Modulation of oxidantstress in vivo in chronic cigarette smokers. Circulation 94:19-25.

177. Kondo I, Berg K, Drayna D & Lawn R (1989) DNA polymorphism at the locus forhuman cholesteryl ester transfer protein (CETP) is associated with high densitylipoprotein cholesterol and apolopoprotein levels. Clin Genet 35:49-56.

178. Freeman DJ, Packard CJ, Shepherd J & Gaffney D (1990) Polymorphisms in thegene coding for cholesteryl ester transfer protein are related to plasma high-densitylipoprotein cholesterol and transfer protein activity. Clin Sci 79:575-581.

179. Freeman DJ, Griffin BA, Holmes AP, Lidsay GM, Gaffney D, Packard CJ &Shepherd J (1994) Regulation of plasma HDL cholesterol and subfractiondistribution by genetic and environmental factors. Associations between TaqIBRFLP in the CETP gene and smoking and obesity. Arterioscler Thromb 14:336-344.

180. Sigurdsson G Jr, Gudnason V, Sigurdsson G & Humphries SE (1992) Interactionbetween polymorphism of the apo A-I promoter region and smoking determinesplasma levels of HDL and apo A-I. Arterioscler Thromb 12:1017-1022.

79

181. Juo SH, Wyszynski DF, Beaty TH, Huang HY, Bailey M & Wilson JE (1999) Mildassociation between the A/G polymorphism in the promoter of the apolipoproteinA-I gene and apolipoprotein A-I levels: a meta-analysis. Am Med Genet 82:235-241.

182. Glisic S, Savic I & Alavantic D (1995) Apolipoprotein B gene DNApolymorphisms (EcoRI and MspI) and serum lipid levels in the Serbian healthypopulation: interaction of rare alleles and smoking and cholesterol levels. GenetEpidemiol 12:499-508.

183. Senti M, Aubo C & Bosch M (1998) The relationship between smoking andtriglyceride-rich lipoproteins is modulated by genetic variation in the glycoproteinIIIa gene. Metabolism 47:1040-1041.

184. Ardissino D, Mannucci PM, Merlini PA, Duca F, Fetiveau R, Tagliabue L, TubaroM, Ottani F, Ferrario M, Corral J & Margaglione M (1999) Prothrombotic geneticrisk factors in young survivors of myocardial infarction. Blood 94:46-51.

185. Arinami T, Ohtsuki T, Yamakawa-Kobayashi K, Amamiya H, Fujiwara H, KawataK, Ishiguro H & Hanaguchi H (1999) A synergistic effect of serotonin transportergene polymorphism and smoking in association with coronary heart disease.Thromb Haemost 81:853-856.

186. Keys A, Anderson JT & Grande F (1965) Serum cholesterol response to changes inthe diet. IV. Particular saturated fatty acids in the diet. Metabolism 14:747-787.

187. Hegsted DM, Ausman LM, Johnson JA & Dallal GE (1993) Dietary fat and serumlipids: An evaluation of the experimental data. Am J Clin Nutr 57:873-875.

188. Katan MB, Beynen AC, de Vries JHM & Nobels A (1986) Existence of consistenthypo- and hyperresponders to dietary cholesterol in man. Am J Epidemiol 123:221-234.

189. Glatz JFC, Turner PR, Katan MB, Stalenhoef AFH & Lewis B (1993) Hypo- andhyperresponse of serum cholesterol level and low density lipoprotein production anddegradation to dietary cholesteol in man. In: Lee KT, Oike Y, Kanazawa T (eds.)Annals of the New York Academy of Sciences. 676:163-178.

190. Beynen AC, Katan MB & van Zutphen LF (1987) Hypo- and hyperresponders:Individual differences in the response of serum cholesterol concentration to changesin diet. Adv Lipid Res. 22:115-171.

191. Hofman AF (1994) Intestinal absorption of bile acids and biliary constituents: Theintestinal component of the enterohepatic circulation and the integrated system. In:Johnson LR (ed) Physiology of the gastrointestinal tract.1845-1865. Raven Press.New York.

192. Beynen AC, Meijer GW, Lemmens AG, Glatz JF, Versluis A, Katan MB & VanZutphen LF (1989) Sterol balance and cholesterol absorption in inbredd strains ofrabbits hypo- and hyperresponsive to dietary cholesterol. Atherosclerosis 77:151-157.

193. Jolley CD, Dietschy JM & Turley SD (1999) Genetic differences in choleterolabsorption in 129/Sv and C57BL/6 mice: effect on cholesterol responsiveness. AmJ Physiol276:1117-1124.

194. Ostlund RE Jr, Bosner MS & Stenson WF (1999) Cholesterol absorption efficiencydeclines at moderate dietary doses in normal human subjects. J Lipid Res 40:1453-1458.

80

195. Sehayek E, Nath C, Heinemann T, McGee M, Seidman CE, Samuel P & BreslowJL (1998) U-shape relationship between change in dietary cholesterol absorption andplasma lipoprotein responsiveness and evidence for extreme interindividual variationin dietary cholesterol absorption in humans. J Lipid Res 39:2415-2411.

196. Kesäniemi YA & Miettinen TA (1987) Cholesterol absorption efficiency regulatesplasma cholesterol level in the Finnish population. Eur J Clin Invest 17:391-395.

197. Kesäniemi YA, Ehnholm C & Miettinen TA (1987) Intestinal cholesterolabsorption efficiency in man is related to apolipoprotein E phenotype. J Clin Invest80:578581

198. Miettinen TA, Gylling H, Vanhanen H& Ollus A (1992) Cholesterol absorption,elimination, and synthesis related to LDL kinetics during varying fat intake in menwith different apoprotein E phenotypes. Arterioscler Thromb 12:1044-1052.

199. Miettinen TA & Vanhanen H (1994) Dietary sitostanol related to absorption,synthesis and serum level of cholesterol in different apolipoprotein E phenotypes.Atherosclerosis 105:217-226.

200. Gylling H, Kontula K & Miettinen TA. Cholesterol absorption and metabolism andLDL kinetics in healthy men with different apoprotein E phenotypes and apoproteinB XbaI and LDL receptor PvuII genotypes. Arterioscler Thromb VascBiol1995;15:208-213.

201. Quintao E, Grundy SM, Ahrens EH (1971) Effects of dietary cholesterol on theregulation of total body cholesterol in man. J Lipid Res 12:233-247.

202. Nestel PJ & Poyser A (1976) Changes in cholesterol synthesis and excretion whencholesterol intake is increased. Metabolism 24:1591-1599.

203. Fielding CJ (1997) Response of low density lipoprotein cholesterol levels to dietarychange: contributions of different mechanisms. Curr Opin Lipidol 8:39-42.

204. Jones PJ (1997) Regulation of cholesterol biosynthesis by diet in humans. Am JClin Nutr 66:438-446.

205. Lin DS & Connor WE (1980) The long term effects of dietary cholesterol uponplasma lipids, lipoproteins, cholesterol absorption and the sterol balance in man: thedemonstration of feedback inhibition of cholesterol biosynthesis and increased bileacid secretion. J Lipid Res.;21:1042-1052.

206. McMurry MP, Connor WE, Lin DS, Cerquiera MT & Connors SL (1985) Theabsorption of cholesterol and the sterol balance in the Tarahumara Indians of Mexicofed cholesterol-free and high cholesterol diets. Am J Clin Nutr 41:1289-1298.

207. Jones PJH, Pappu AS, Hatcher L, Li Z-C, Illingworth DR & Connor WE (1996)Dietary cholesterol feeding supresses human cholesterol synthesis measured bydeuterium incorporation and urinary mevalonic acid levels. Arterioscler ThrombVasc Biol 16:1222-1228.

208. Lofland HB Jr, Clarkson TB, St Clair RW & Lehner ND (1972) Studies on theregulation of plasma cholesterol levels in squirrel monkeys of two phenotypes. JLipid Res 13:39-47.

209. Eggen DA (1976) Cholesterol metabolism in groups of rhesus monkeys with highor low response of serum cholesterol to an atherogenic diet. J Lipid Res 17:663-673.

210. Parks JS, Lehner ND, St Clair RW & Lofland HB (1977) Whole body cholesterolmetabolism in cholesterol-fed African green monkeys with a variablehypercholesterolemic response. J Lab Clin Med 90:1021-1034.

81

211. Goldstein JL & Brown MS (1977) The low-density lipoprotein pathway and itsrelation to atherosclerosis. Annu Rev Biochem 46:897-930.

212. Spady DK, Woollett LA & Dietchy JM (1993) Regulation of plasma LDL-cholesterol levels by dietary cholesterol and fatty acids. Ann Rev Nutr 13:355-382.

213. Groener JEM, van Ramshorst EM, Katan MB, Mensink RP & van Tol A (1991)Diet-induced alteration in the activity of plasma lipid transfer protein innormolipidemic human subjects. Atherosclerosis 87:221-226.

214. Cox C, Mann J, Sutherland W, Chisholm A & Skeaff M (1995) Effects of coconutoil, butter and safflower oil in persons with moderately elevated cholesterol levels. JLipid Res 36:1787-1795.

215. Schwab US, Maliranta HM, Sarkkinen ES, Savolainen MJ, Kesäniemi YA &Uusitupa MIJ (1996) Different effects of palmitic and stearic acid enriched diets onserum lipids and lipoproteins and cholesteryl ester transfer protein activity in healthyyoung women. Metabolism 45:143-149.

216. Abbey M & Nestel PJ (1994) Plasma cholesteryl ester transfer protein in increasedwhen trans-elaidic acid is substituted for cis-oleic acid in the diet. Atherosclerosis106:99-107.

217. van Tol A, Zock PL, van Gent T, Scheek LM & Katan MB (1995) Dietary transfatty acids increase serum cholesteryl ester transfer protein activity in man.Atherosclerosis 115:129-134.

218. Dammerman M & Breslow JL (1995) Genetic basis of lipoprotein disorders.Circulation 91:505-512.

219. Humphries SE, Peacock RE & Talmud PJ (1995) The genetic determinants ofplasma cholesterol and response to diet. In: Betteridge DJ (ed). Baillières’s clinicalendocrinology and metabolism. International practice and research. 9:797-823.Saunders Company.

220. Sul HS & Wang D (1998) Nutritional and hormonal regulation of enzymes in fatsynthesis: studies of fatty acid synthase and mitochondrial glycerol-3-phosphateacyltransferase gene transcription. Annu Rev Nutr 18:331-351.

221. Baillie RA, Jump DB & Clarke SD (1996) Specific effects of polyunsaturated fattyacids on gene expression. Curr Opin Lipidol 7:53-55.

222. Clarke SD & Jump DB (1994) Dietary polyunsaturated fatty acid regulation of genetranscription. Annu Rev Nutr 14:83-98.

223. Clarke SD, Baillie R, Jump DB & Nakamura MT (1997) Fatty acid regulation ofgene expression. Its role in fuel partitioning and insulin resistance. Ann NY AcadSci 827:178-187.

224. Fruchart J-C, Duriez P & Staels B (1999) Peroxisome proliferator-activated receptor-alpha activators regulate genes governing lipoprotein metabolism, vascularinflammation and atherosclerosis. Curr Opin Lipidol 10:245-257.

225. Brown MS & Goldstein JL (1997) The SREBP pathway: regulation of cholesterolmetabolism by proteolysis of a membrane-bound transcription factor. Cell 89:331-340.

226. Edwards PA & Ericsson J (1998) Signaling molecules derived from the cholesterolbiosynthetic pathway: mechanisms of action and possible roles in human disease.Curr Opin Lipidol 9:433-440.

227. Boucher P, de Lorgeril M, Salen P, Crozier P, Delaye J, Vallon JJ, Geyssant A &Dante R (1998) Effect of dietary cholesterol on low density lipoprotein-receptor, 3-

82

hydroxy-3-methylglutaryl-CoA reductase, and low density lipoprotein receptor-related protein mRNA expression in healthy humans. Lipids 33:1177-1186.

228. Kraemer R, Pomeranz KB, Kesav S, Scallen TJ & Hajjar DP (1995) Cholesterolenrichment enhances expression of sterol carrier protein-2: implication inintracellular cholesterol trafficking. J Lipid Res 36:2630-2638.

229. Wang N, Tabas I, Wunchester R, Ravalli S, Rabbani LE & Tall A (1996)Interleukin 8 is induced by cholesterol loading of macrophages and expressed bymacrophage foam cells in human atheroma. J Biol Chem 271:8837-8842.

230. Quinet EM, Agellon LB, Kroon PA, Marcel YL, Lee YC, Whitlock ME & Tall AR(1990) Atherogenic diet increases cholesteryl ester transfer protein messenger RNAlevels in rabbit liver. J Clin Invest 85:357-363.

231. Oliveira HCF, Chouinard RA, Agellon LB, Bruce C, Ma L & Walsh A (1996)Human cholesteryl ester transfer protein gene proximal promoter contains dietarycholesterol positive responsive elements and mediates expression in small intestineand periphery while predominant liver and spleen expression is controlled by 5’-distal sequences mapped in transgenic mice. J Biol Chem 271:31831-31838.

232. Srivastava RA (1996) Regulation of the apolipoprotein E by dietary lipids occurs bytranscriptional and post-transcriptional mechanisms. Mol Cell Biochem 155:153-162.

233. Chagnon YC, Perusse L, Weisnagel SJ, Rankinen T & Bouchard C (2000) Thehuman obesity gene map: the 1999 update. Obes Res 8:89-117.

234. Widen E, Lehto M, Kanninen T, Walston J, Schuldiner HR & Groop LC (1995)Association of a polymorphism in the beta-3-adrenergic-receptor gene with featuresof insulin resistence syndrome in Finns. N Engl J Med 333:348-351.

235. Leibowitz SF (1995) Brain peptides and obesity: pharmacologic treatment. Obes Res3(Suppl):573-589.

236. Wang J, Liu R, Hawkins M, Barzilai N & Rossetti L (1998) A nutrient-sensingpathway regulates leptin gene expression in muscle and fat. Nature 393:684-688.

237. Fajas L, Fruchart J-C & Auwerx J (1998) Transcriptional control of adipogenesis.Curr Opin Cell Biol 10:165-173.

238. Vidal-Puig AJ, Considine RV, Jimenez-Linan M, Werman A, Pories WJ, Caro JF& Flier JS (1997) Peroxisome proliferator-activated receptor gene expression inhuman tissues. Effects of obesity, weight loss, and regulation by insulin andglucocorticoids. J Clin Invest 99:2416-2422.

239. Schoonjans K, Peinado-Onsurbe J, Lefebvre AM, Heyman RA, Brigs M, Deeb S,Staels B & Auwerx J (1996) PPARalpha and PPARgamma activators direct adistinct tissue-specific transcriptional response via a PPRE in lipoprotein lipasegene. EMBO J 15:5336-5348.

240. Freake HC (1998) Uncoupling proteins: beyond brown adipose tissue. Nutr Rev56:185-188.

241. Fried SK & Zehner R (1989) Cachectin/tumor necrosis factor decreases humanadipose tissue lipoprotein lipase mRNA level, synthesis, and activity. J Lipid Res30:1917-1923.

242. Nicklas BJ, Katzel LI, Bunyard LB, Dennis KE & Goldberg AP (1997) Effects of anAmerican Heart Association diet and weight loss on lipoprotein lipids in obese,postmenopausal women. Am J Clin Nutr 66:853-859.

83

243. Bachorik PS & Albers JJ (1986) Precipitation methods for quantification onlipoproteins. Methods Enzymol 129:78-100.

244. Friedewald WT, Levy RI & Fredrickson DS (1972) Estimation of the concentrationof low-density lipoprotein cholesterol in plasma, without use of preparativeultracentrifuge. Clin Chem 18:499-502.

245. Ehnholm C, Lukka M, Kuusi T, Nikkilä E & Utermann G (1986) Apolipoprotein Epolymorphism in the Finnish population: gene frequences and relation tolipoprotein concentrations. J Lipid Res 27:227-235.

246. Menzel HJ & Utermann G (1986) Apolipoprotein E phenotyping from serum bywestern blotting. Electrophoresis 7:492.

247. Miller SA, Dykes DD & Polesky HF (1988) A simple salting out procedure forextracting DNA from human nucleated cells. Nucleic Acids Res 16:1215.

248. Ukkola O, Savolainen MJ, Salmela PI, von Dickhoff K & Kesäniemi YA (1993)Apolipoprotein B gene DNA polymorphisms are associated with macro- andmicroangiopathy in non-insulin-dependent diabetes mellitus. Clin Genet 44:177-84.

249. Huang LS, de Graaf J Breslow JL (1988) ApoB gene MspI RFLP in exon 26changes amino acid 3611 from Arg to Gln. J Lipid Res 29:63-67.

250. Hallman DM, Visvikis S, Steinmetz J & Boerwinkle E (1994) The effect ofvariation in the apolipoprotein B gene on plasma lipid and apolipoprotein Blevels.I.A likelihood-based approach to cladistic analysis. Ann Hum Genet 58:35-64.

251. Visvikis S, Chan L, Siest G, Drouin P & Boerwinkle E (1990) An insertiondeletion polymorphism in the signal peptide of the human apolipoprotein B gene.Human Genet 84:373-75.

252. Chirgwin JM, Przybyla AE, MacDonald RJ & Rutter WJ (1979) Isolation ofbiologically active ribonucleic acid from sources enriched in ribonuclease.Biochemistry 18:5294-5299.

253. PCR Protocols. A Quide to methods and applications. Innis MA, Gelfand DH,Sninsky JJ, White TJ (eds) .1-75. Academic Press Inc. San Diego, CA.

254. Rastas M, Seppänen R, Knuts L-R, Karvetti R-L & Varo P (1989) Nutrientcomposition of foods. Helsinki: Publication of the Social Insurance Institution.

255. Hedges LV & Olkin I (1985) Statistical methods for meta-analysis. Academic Press.Montreal.

256. Nevanlinna HR (1972) The Finnish population structure. A genetic and genealogicalstudy. Hereditas 71:195-236.

257. On behalf of the Task Froce, Sans S, Kesteloot H & Kromhout D (1997) Theburden of the European Society of Cardiology on Cardiovascular Mortality andMorbidity Statistics in Europe. Eur Heart J 18:1231-124.

258. World Healt Statistics Annual 1996, Geneve 1998.259. Koskenvuo M, Kaprio J & Romanov K (1992) Twin studies in metabolic disease.

Ann Med 24:379-381.260. Nissinen A, Piha T, Tuomilehto J, Romo M & Puska P (1987) Comparison of the

levels of cardiovascular risk factors between eastern and south-western Finland 1982.Acta Med Scand 222:389-400.

84

261. Juvonen T, Kervinen K, Kairaluoma MI, Lajunen LHJ & Kesäniemi YA (1993)Gallstone cholesterol content is related to apolipoprotein E polymorphism.Gastroenterology 104:1806-1813.

262. Wilson PW, Castelli WP & Kannel WB (1987) Coronary risk prediction in adults(the Framingham Heart Study). Am J Cardiol 59:91-94.

263. Sarkkinen E (1995) Long-term feasibility and effects of three fat-modified diets infree-living hypercholesterolemic subjects. Kuopio University Publications D.Medical Sciences 72. Kuopio, Finland.

264. Hiltunen TP (1997) Expression of lipoprotein receptors, 15-lipoxygenase and β-amyloid precursor protein in atherosclerotic lesions. University of Tampere,Tampere, Finland.

265. Mulrow C (1995) Rationale for systematic reviews In: Chalmers I, Altman DG(eds). Systematic reviews:1-8. BMJ Publishing Group, London

266. Thompson SG (1995) Why sources of heterogeneity in meta-analysis should beinvestigated. In: Chalmers I, Altman DG (eds). Systematic reviews:48-63. BMJPublishing Group, London.

267. Lau J, Antman EM, Jimenez-Silva J, Kubelnic B, Mosteller F & Chalmers TC(1992) Cumulative meta-analysis of therapeutic trials for myocardial infarction. NEngl J Med 327:248-254.

268. Dickersin K, Scherer R & Lefebvre C (1995) Identifying relevant studies forsystematic reviews. In: Chalmers I, Altman DG (eds). Systematic reviews:17-36.BMJ Publishing Group, London.

269. Laakso M, Sarlund H & Mykkänen L (1990) Insulin resistance is associated withlipid and lipoprotein abnormalities in subjects with varying degrees of glucosetolerance. Arteriosclerosis 10:223-231.

270. Packard CJ, McKinney L, Carr K & Shepherd J (1983) Cholesterol feeding increaseslow density lipoprotein synthesis. J Clin Invest 72:45-51.

271. Knittle JL & Ahrens EH (1964) Carbohydrate metabolism in two form ofhyperglyceridemia. J Clin Invest 43:485-495.

272. Ahrens EH, Hirsch J, Oette K & Farquhar JW (1961) Carbohydrate-induced and fat-induced lipemia. Trans Assoc Am Physicians 74:134-146.

273. Mancini M, Mattock M, Rabays E, Chait A & Lewis B (1973) Studies of themechanisms of carbohydrate-induced lipaemia in normal man. Atherosclerosis17:445-454.

274. Ginsberg H, Olefsky JM, Kimmerling G, Crapo P & Reaven GM (1976) Inductionof hypertriglyceridemia by a low-fat diet. J Clin Endocrin Metab 42:729-735.

275. Osborn TF, Goldstein JL & Brown MS (1985) 5’-end of HMG CoA reductase genecontains sequences responsible for cholesterol mediated inhibition of transcription.Cell 42:203-212.

276. Berg K (1989) Predictive genetic testing to control CHD and hyperlipidemia.Arteriosclerosis 9(Suppl 1):150-158.

277. Stuyt PM, Brenninkmeijer BJ, Demacker PN, Hendriks JC, vanElteren P, StaenhoefAF & van’tLaar A (1991) Apolipoprotein E phenotypes, serum lipoproteins andapolipoproteins in angiographically assesed coronary heart disease. Scand J Clin LabInvest 51:425-435.

85

278. Xu CF, Boerwinkle E, Tikkanen MJ, Huttunen JK, Humphries SE & Talmud PJ(1990) Genetic variation at the apolipoprotein gene loci contribute to response ofplasma lipids to dietary change. Genet Epidemiol 7:261-275.

279. Cobb MM, Teitlebaum H, Risch N, Jekel J & Ostfeld A (1992) Influence of dietaryfat, apolipoprotein E, and sex on plasma lipoprotein levels. Circulation 86:849-857.

280. Sarkkinen ES, Uusitupa MI, Pietinen P, Aro A, Ahola I, Penttilä I, Kervinen K &Kesäniemi YA (1994) Long-term effects of three fat-modified diets inhypercholesterolemic subjects. Atherosclerosis 105:9-23.

281. Zambon D, Ros E, Casals E, Saullehy C, Bartomey A & Campero I (1995) Effectof apolipoprotein E polymorphismon the serum lipid response to a hypolipidemicdiet rish in monounsaturated fatty acids in patients with hypercholesterolemia andcombined hyperlipidemia. Am J Clin Nutr 61:141-148.

282. Martin LJ, Connelly PW, Naneoo D, Wood N, Zhang ZJ, Maquire G, Quinet E,Tall AR, Marcel YL & McPherson R (1993) Cholesterol ester transfer protein andhigh density lipoprotein responses to cholesterol feeding in men: relationship toapolipoprotein E phenotype. J Lipid Res 34:437-446.

283. Jones PJ, Main BF & Frohlich JJ (1993) Response of cholesterol synthesis tocholesterol feeding in men with different apolipoprotein E phenotypes. Metabolism42:1065-1071.

284. Glatz JF, Demacker PNM, Turner PR & Katan MB (1991) Response of serumcholesterol in relation to apolipoprotein E phenotype. Nutr Metab Cardiovasc Dis1:13-17.

285. Marshall JA, Kamboh MI, Bessesen DH, Hoag S, Hamman RF & Ferrell RE(1996) Association between dietary factors and serum lipids by apolipoprotein Epolymorphism. Am J Clin Nutr 63:87-95.

286. Boerwinkle E, Brown SA, Rohrbach K, Gotto AM Jr & Patsch W (1991) Role ofapolipoprotein E and B variation in determining response of lipid, lipoprotein, andapolipoprotein levels to increased dietary cholesterol. Am J Hum Genet 49:1145-1154.

287. Friedlander Y, Berry EM, Eisenberg S, Stein Y & Leitersdorf E (1995) Plasmalipids and lipoproteins in response to a dietary challenge: analysis of four candidategenes. Clin Genet 47:1-12.

288. Ginsberg H, Karmally W, Siddiqui M, Holleran S, Tall AR, Rumsey SC,Deckelbaum RJ, Blaner WS & Ranakrishnan R (1994) A dose-response study of theeffects of dietary cholesterol on fasting and postprandial lipid and lipoproteinmetabolism in healthy young men. Arterioscler Thromb 14:576-586.

289. Lehtimäki T, Moilanen T, Solakivi T, Laippala T & Ehnholm C (1992)Cholesterol-rich diet induced changes in plasma lipids in relation to apolipoproteinE phenotype in healthy students. Ann Med 24:61-66.

290. Gylling H & Miettinen TA (1992) Cholesterol absorption and synthesis related tolow density lipoprotein metabolism during varying cholesterol intake in men withdifferent apo E phenotypes. J Lipid Res 33:1361-1371.

291. Mänttäri M, Koskinen P, Ehnholm C, Huttunen JK & Manninen V (1991)Apolipoprotein E polymorphism influences the serum cholesterol response todietary intervention. Metabolism 40:217-221.

86

292. Tikkanen MJ, Huttunen JK, Ehnholm C & Pietinen P (1990) Apolipoprotein E4homozygosity predisposes to serum cholesterol elevation during high fat diet.Arteriosclerosis 10:285-288.

293. Lopez-Miranda J, Ordovas JM, Mata P, Lichtenstein AH, Clevidence S, Judd JT &Schaefer EJ (1994) Effects of apolipoprotein E phenotype on diet induced loweringof plasma low density lipoprotein cholesterol. J Lipid Res 35:1965-1975.

294. Gylling H, Kontula K, Koivisto UM, Miettinen HE & Miettinen TA (1997)Polymorphism of the genes encoding apoproteins A-I, C-III, and E and LDLreceptor, and cholesterol and LDL metabolism during increased cholesterol intake.Common alleles of the apoprotein E show the greatest regulatory impact.Arterioscler Thromb Vasc Biol 17:38-44.

295. Dreon DM, Fernstrom HA, Miller B & Krauss RM (1995) Apolipoprotein Eisoform phenotype and LDL subclass response to a reduced-fat diet. ArteriosclerThromb Vasc Biol 15:105-111.

296. Fisher EA, Blum CB, Zannis VI & Breslow JL (1983) Independent effects of dietarysaturated fat and cholesterol on plasma lipids, lipoproteins, and apolipoprotein E. JLipid Res 24:1039-1048.

297. Miettinen TA, Gylling H & Vanhanen H (1988) Serum cholesterol response todietary cholesterol and apolipoprotein E phenotype. Lancet 2:1261.

298. Sarkkinen E, Korhonen M, Erkkilä A, Ebeling T & Uusitupa M (1998) Effect ofapolipoprotein E polymorphism on serum lipid response to the separatemodification of dietary fat and dietary cholesterol. Am J Clin Nutr 68:1215-1222.

299. Ordovas JM, Lopez-Miranda J, Mata P, Perez-Jimenez F, Lichtenstein AH &Schaefer EJ (1995) Gene-diet interaction in determining plasma lipid response todietary intervention. Atherosclerosis 118(Suppl):11-27.

300. Bergeron N & Havel RJ (1996) Prolonged postprandial responses of lipids andapolipoproteins in triglyceride-rich lipoproteins of individuals expressing anapolipoprotein epsilon 4 allele. J Clin Invest 97:65-72.

301. Reznik Y, Pousse P, Herrou M, Morello R, Mahoudeau J, Drosdovksy MA &Fradin S (1996) Postprandial lipoprotein metabolism in normotriglyceridemic non-insulin dependent diabetic patients: influence of apolipoprotein E polymorphism.Metabolism 45:63-71

302. Ordovas JM, Lopez-Miranda J, Perez-Jimenez F, Rodriquez C, Park JS, Cole T &Schaefer EJ (1995) Effect of apolipoprotein E and A-IV phenotypes on the lowdensity lipoprotein response to HMG CoA redustase inhibitor therapy.Atherosclerosis 113:157-166.

303. Carmena R, Roederer G, Mailloux H, Lussier-Cacan S & Davignon J (1993) Theresponse to lovastatin treatment in patients with heterozygous familialhypercholesterolemia is modulated by apolipoprotein E polymorphism. Metabolism42:895-898.

304. Hjermann I, Helgeland A, Holme I, Lund-Larsen PG & Leren P (1978) Theassociation between blood pressure and serum cholesterol in healthy men: the OsloStudy. J Epidemiol Community Health 32:117-123.

305. Tikkanen MJ, Xu CF, Hämäläinen T, Talmud P, Sarna S, Huttunen JK, Pietinen P& Humphries S (1990) XbaI polymorphism of the apolipoprotein B gene influencesplasma lipid response to diet intervention. Clin Genet 37:327-334.

87

306. Xu CF, Tikkanen MJ, Huttunen JK, Pietinen P, Butler R, Humphries S & TalmudP (1990) Apolipoprotein B signal peptide insertion/deletion polymorphism isassociated with Ag epitopes and involved in the determination of serum triglyceridelevels. J Lipid Res 31:1255-1261.

307. Talmud PJ, Boerwinkle E, Xu CF, Tikkanen MJ, Pietinen P, Huttunen JK &Humphries S (1992) Dietary intake and gene variation influence the response ofplasma lipids to dietary intervention. Genet Epidemiol 31:1255-1261.

308. Tikkanen MJ & Heliö T (1992) Genetic variants of apolipoprotein B: relation toserum lipid levels and coronary heart disease among the Finns. Ann Med 24:357-361.

309. Abbey M, Belling B, Clifton P & Nestel PJ (1991) Apolipoprotein B genepolymorphism associates with plasma cholesterol changes induced by dietary fat andcholesterol. Nutr Metab Cardiovasc Dis 1:10-12.

310. Friedlander Y, Kaufman NA, Cedar H & Kark JD (1993) XbaI polymorphism of theapolipoprotein B gene and plasma lipid and lipoprotein response. Clin Genet43:223-231.

311. Abbey M, Hirata F, Chen GZ, Ross R, Noakes M, Belling B, Clifton P & NestelPJ (1995) Restriction fragment length polymorphism of the apolipoprotein B geneand response to dietary fat and cholesterol. Can J Cardiol 11(Suppl):79-85.

312. Pajukanta PE, Valsta LM, Aro A, Pietinen P, Heliö T & Tikkanen MJ (1996) Theeffects of the apolipoprotein B signal peptide (ins/del) and XbaI polymorphism ofplasma lipid responses to dietary change. Atherosclerosis 122:1-10.

313. Humphries SE, Talmud PJ, Cox C, Sutherland W & Mann J (1996) Genetic factorsaffecting the consistency and magnitude of change in plasma cholesterol in responseto dietary challenge. Q J Med 89:671-680.

314. Tato F, Vega GL, Tall AR & Grundy SM (1995) Relation between cholesteryl estertransfer protein activities and lipoprotein cholesterol in patients withhypercholesterolemia and combined hyperlipidemia. Arterioscler Thromb Vasc Biol15:112-120.

315. Katan MB, Zock PL & Mensink RP (1995) Trans fatty acids and their effects onlipoproteins in humans. Ann Rev Nutr 15:473-493.

316. Zehner R (1997) The tissue-specific expression of lipoprotein lipase: implicationsfor energy and lipoprotein metabolism. Curr Opin Lipidol 8:77-88.

317. Mauriege P, Marette A, Atgie C, Bouchard C, Theriault G, Bukowiecki LK,Marceau P, Biron S, Nadeau A & Despres JP (1995) Regional variation in adiposetissue metabolism of severly obese premenopausal women. J Lipid Res 36:672-684.

318. Taskinen MR & Nikkilä EA (1987) Basal and postprandial lipoprotein lipaseactivity in adipose tissue during caloric restriction and refeeding. Metabolism36:625-630.

319. Kern PA, Saghizadeh M, Ong JM, Bosch RJ, Deem R & Simsolo RB (1995) Theexpression of tumor necrosis factor in human adipose tissue. Regulation by obesity,weight loss, and relationship to lipoprotein lipase. J Clin Invest 95:2111-2119.