Michael Pugia Editor Infl ammatory Pathways in Diabetes
226
Series Editor: Michael J. Parnham Progress in Inflammation Research Michael Pugia Editor Inflammatory Pathways in Diabetes Biomarkers and Clinical Correlates
Transcript of Michael Pugia Editor Infl ammatory Pathways in Diabetes
Series Editor: Michael J. ParnhamProgress in Infl ammation Research
Michael Pugia Editor
Infl ammatory Pathways in DiabetesBiomarkers and Clinical Correlates
Progress in Infl ammation Research
Series Editors Michael J. Parnham Fraunhofer IME & Goethe University Frankfurt am Main , Germany
Achim Schmidtko Witten , Germany
The last few years have seen a revolution in our understanding of how blood and tissue cells interact and of the intracellular mechanisms controlling their activation. This has not only provided multiple targets for potential anti-infl ammatory and immunomodulatory therapy, but has also revealed the underlying infl ammatory pathology of many diseases. This monograph series provides up-to-date information on the latest developments in the pathology, mechanisms and therapy of infl ammatory disease. Areas covered include: vascular responses, skin infl ammation, pain, neuroinfl ammation, arthritis cartilage and bone, airways infl ammation and asthma, allergy, cytokines and infl ammatory mediators, cell signalling, and recent advances in drug therapy. Each volume is edited by acknowledged experts providing succinct overviews on specifi c topics intended to inform and explain. The series is of interest to academic and industrial biomedical researchers, drug development personnel and rheumatologists, allergists, pathologists, dermatologists and other clinicians requiring regular scientifi c updates.
More information about this series at http://www.springer.com/series/4983
Michael Pugia Editor
Infl ammatory Pathways in Diabetes Biomarkers and Clinical Correlates
Additional material to this book can be downloaded from http://extras.springer.com
Progress in Infl ammation Research ISBN 978-3-319-21926-4 ISBN 978-3-319-21927-1 (eBook) DOI 10.1007/978-3-319-21927-1
Library of Congress Control Number: 2015952543
Springer Cham Heidelberg New York Dordrecht London © Springer International Publishing Switzerland 2015 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifi cally the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfi lms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specifi c statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made.
Printed on acid-free paper
Springer International Publishing AG Switzerland is part of Springer Science+Business Media (www.springer.com)
Editor Michael Pugia Strategic Innovation Siemens Healthcare Diagnostics Elkhart , IN USA
v
Abbreviations
ABP-26 Adenosine deaminase binding protein 26 AC Adenylcyclase ACC Acetyl Coenzyme A Carboxylase ACS Acute coronary syndrome Adpn Adiponectin AdipoR Adiponectin Receptor AG Anhydroglucitol AGE Advance glycation end products AGP α-1- acid glycoprotein ALP Alkaline phosphatase Akt Protein kinase B AKI Acute kidney injury AMI Acute myocardial infarction AMBK pre-alpha-1-microglobulin/bikunin AMPK 5′-AMP activated protein kinase ASP Acylation stimulating protein AUC Area under the curve Bik Bikunin BMI Body mass index BNP B-type natriuretic peptide BSA Bovine Serum Albumin CABG Coronary artery bypass grafting CAD Coronary artery disease CBC Complete cell count CCF Cleveland Clinic Foundation CCL Chemokine (C-C motif) ligand CD Clusters of differentiation CHF Congestive heart failure CI Confi dence interval CINC Cytokine-induced neutrophil chemoattractant CKD Chronic kidney disease
vi
CPB Cardiopulmonary bypass CTF C-Terminal Fragment of Adiponectin Receptor CRP C-reactive protein CVD Cardiovascular disease DGAT Acyl CoA diacylglycerol transferase DPP-IV Dipeptidyl peptidase- IV DM Diabetes ECM Extra cellular matrix EGF Epidermal growth factor eGFR Estimated GFR ELISA Enzyme linked immunoabsorbent assay eNOS Endothelial NOS EPO Erythropoietin ESRD End stage renal disease ET Endothelin Erk Extracellular signal-regulated kinase FABP Fatty acid-binding protein Fluo-4 Calcium binding fl uorescence dye FGF Fibroblast growth factor FFA Free fatty acids FXR Farnesoid X receptor a GBM Glomerular basement membrane GDF Growth differentiation factor GFR Glomerular fi ltration rate GH Growth hormone GIP Glucose-dependent insulinotropic polypeptide GLP Glucagon-like peptide-1 GN Glomerular nephritis GSH Glutathione GHSR Growth hormone secretagogue receptor GULT Glucose transporters GCD59 glycated CD59 (aka glyCD59) HAMA Human anti-mouse antibody interference HbA1c Hemogloblin A1c HDL High-density lipoprotein HGF Hepatocyte growth factor HIF Hypoxia-inducible factor HMG 3-Hydroxy-3-methylglutaryl CoA reductase HMW High molecular weight HTN Hypertension ICC Immunocytochemistry IHC Immunohistochemistry I-α-I Inter-α-inhibitor Ig-CTF Immunoglobulin attached C-Terminal Fragment of AdipoR ICC Immuno cytochemistry
Abbreviations
vii
IDE Insulin-degrading enzyme IGF Insulin-like growth factor IGFBP IGF binding proteins IGT Impaired glucose tolerance IHC Immuno histochemistry IL Interleukin cytokines ICAM Intercellular adhesion molecule JAK STAT Janus kinase/ signal transducer and activator of transcription JNK c-Jun N-terminal kinases KIM Kidney injury molecule KLH Keyhole limpet hemocyanin LCN-2 Lipocalin LDL Low-density lipoprotein L-FABP Liver type fatty acid binding protein LMW Low molecular weight LPD Lipidemia LPS Lipopolysaccharide Lp-PLA2 Lipoprotein-associated phospholipase A2 LXR Liver X receptors LYVE Endothelial hyaluronan receptor mAb Monoclonal antibody MAC Membrane attack complex MALDI Matrix-assisted laser desorption/ionization MAPK Mitogen-activated protein kinase MCP Monocyte chemotactic protein MDA Malondialdehyde MetSyn Metabolic syndrome MIC Macrophage inhibitory cytokine MIP Macrophage infammatory protein MMP Matrix metalloproteinase MPO Myeloperoxidase mTOR Mammalian target of rapamycin NADH Nicotinamide adenine dinucleotide NAPLD Non-alcoholic fatty liver disease NGF Nerve growth factor NFκB Nuclear factor kappa-light-chain-enhancer of activated B cells NGAL Neutrophil gelatinase lipocalin NOS Nitric oxide synthase NOX NADPH oxidase NPY Neuropeptide Y ob Obesity OFTT Oral fat tolerance test OGTT Oral glucose tolerance test OHdG Hydroxydeoxyguanosine OR Odds-ratio
Abbreviations
viii
oxLDL Oxidized low density lipoprotein pAb Polyclonal antibody P-α-I pre-interleukin-α-inhibitor PAI-1 Plasminogen activator inhibitor PAR Protease-activated receptors PBS Phosphate buffered saline PDF Placental growth factor PDGF Platelet-derived growth factor PKA Protein kinase A PKC Protein kinase C PI3K Phosphatidylinositol 3-kinase PLA2 Phosphatidylinositol 3-kinase PLC Phospholipase C PNPP Para nitro-phenyl phosphate PPAR Peroxisome proliferator-activated receptor PPG Postprandial glucose PS Phosphatidylserine PSS-380 Phosphatidylserine (PS)-binding fl uorescence dye PR Protienase PYY Neuropeptide like peptide Y RAAS Renin-angiotensin-aldosterone system RAGE Receptors for AGE RFU Relative fl uorescence units ROC Receiver-operator characteristics ROS Reactive oxygen species S100 Calgranulin sCr Serum creatinine SGLT2 Sodium-glucose transporter 2 SREBP Sterol regulatory element binding proteins SMAD Sterile alpha motif domain-containing protein SOD Superoxide dismutase sTNFR Soluble TNF α receptor TACE TNFα alpha cleavage protease TBS Tris buffered saline TFPI Tissue factor pathway inhibitor TFA Trifl uoroacetic acid TG Triglycerides TGF β Transforming growth factor -β TIMP Tissue inhibitor of metalloproteinases TNF α Tumor necrosis factors -α TNFR TNF α receptor THP Tamm-Horsfall protein TLR Toll-like receptors TnI Troponin I tPA Tissue plasminogen activator
Abbreviations
ix
TSP Thrombospondin Tregs Regulatory T cells TZD Thiazolidinediones Uri Uristatin uTi Urinary trypsin inhibitor UTI Urinary tract infection uPA Urokinase-type plasminogen activator uPAR uPA receptor VCAM Vascular cell adhesion molecule VEGF Vascular endothelial growth factor WBC White blood cell WB Western Blot YKL-40 Chitinase-3-like protein 1
Abbreviations
xi
Contents
Part I Introduction
1 Impact of Infl ammation and Innate Immunity Response in Obesity Mediated Diabetes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Michael Pugia
Part II Impact of Complement on Diabetic Disease
2 Complement and Complement Regulatory Proteins in Diabetes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Jose A. Halperin , Pamela Ghosh , Michael Chorev , and Anand Vaidya
Part III Role of C Terminal Fragment of Adiponectin Receptor in Inhibition of Insulin Degradation
3 Development of Adiponectin Receptor C-Terminal Fragment Bioassays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 Michael Pugia and Rui Ma
4 Protease Inhibition and Biological Distribution of the C Terminal Fragment of Adiponectin Receptor. . . . . . . . . . . . 77 Michael Pugia and Rui Ma
5 Cell and Biological Models for the C Terminal Fragment of Adiponectin Receptor . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 Michael Pugia and Rui Ma
6 C-Terminal Fragment of Adiponectin Receptor Clinical Correlations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 Adrian Vella and Michael Pugia
xii
Part IV Uristatin Assay for Prediction of Renal and Other Clinical Events
7 Uristatin Immunoassay Usage in Glomerular Nephritis Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127 Saeed A. Jortani and Michael Pugia
8 Acute Response of Uristatin in Surgery. . . . . . . . . . . . . . . . . . . . . . . . 143 Mitchell H. Rosner and Michael Pugia
9 Cardiovascular and Metabolic Syndrome Impact on Uristatin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157 Saeed A. Jortani and Michael Pugia
10 Uristatin Anti-infl ammatory Cellular Signaling . . . . . . . . . . . . . . . . . 171 Manju Basu , Subhash Basu , and Michael Pugia
Part V Summary
11 Progress in New Markers for Diabetes Infl ammation . . . . . . . . . . . . 193 Michael Pugia
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
Contents
xiii
On-Line Supplements
ELISA method for IgG-CTF assay ELISA method for free CTF assay Western blot methods for studies Tissue and cell staining methods Cell culture methods ELISA method for uristatin assay
xv
Contributors
Manju Basu Department of Chemistry and Biochemistry , University of Notre Dame , Notre Dame , IN , USA
Subhash Basu Department of Chemistry and Biochemistry , University of Notre Dame Notre Dame , IN , USA
Michael Chorev , PhD Laboratory for Translational Research, Division of Hematology, Department of Medicine , Brigham and Women’s Hospital, Harvard Medical School , Boston , MA , USA
Pamela Ghosh , PhD Laboratory for Translational Research, Division of Hematology, Department of Medicine , Brigham and Women’s Hospital, Harvard Medical School , Boston , MA , USA
Jose A. Halperin , MD Laboratory for Translational Research, Division of Hematology, Department of Medicine , Brigham and Women’s Hospital, Harvard Medical School , Boston , MA , USA
Saeed A. Jortani Pathology and Laboratory Medicine, Department of Pathology, School of Medicine , University of Louisville , Louisville , KY , USA
Rui Ma Siemens Healthcare , Pu Dong New Area , People’s Republic of China
Michael Pugia , PhD Strategic Innovation , Siemens Healthcare Diagnostics , Elkhart , IN , USA
Bayer Healthcare Diagnostics , Elkhart , IN , USA
University of Louisville, School of Medicine , Louisville , KY , USA
Department of Chemistry and Biochemistry , University of Notre Dame , Notre Dame , IN , USA
xvi
Mitchell H. Rosner , MD, FACP Division of Nephrology, Department of Medicine , University of Virginia Health System , Charlottesville , VA , USA
Anand Vaidya , MD, MMSc Division of Endocrinology, Diabetes and Hypertension , Brigham and Women’s Hospital, Harvard Medical School , Boston , MA , USA
Adrian Vella , MD Division of Endocrinology and Metabolism , Mayo Clinic , Rochester , MN , USA
Contributors
3© Springer International Publishing Switzerland 2015 M. Pugia (ed.), Infl ammatory Pathways in Diabetes: Biomarkers and Clinical Correlates, Progress in Infl ammation Research, DOI 10.1007/978-3-319-21927-1_1
Chapter 1 Impact of Infl ammation and Innate Immunity Response in Obesity Mediated Diabetes
Michael Pugia
1.1 Healthcare Impact of Diabetes
Diabetes has a signifi cant impact on overall healthcare costs. The population of people in the world who have been diagnosed with diabetes has steadily increased with the prevalence of obesity over the last 20–30 years ( See Fig. 1.1 ). There is a real need to reduce the number of diabetics and to delay end stage complications. This need is diffi cult to address as there is poor diagnostic ability to sort out which pre-diabetics will become diabetics and which diabetics will progress to complications. The frequency and duration of hyperglycemia is the commonly used tool. Hyperglycemia increases as the pre-diabetic patient progresses from impaired glu-cose tolerance (IGT) to fully glucose intolerant. Diabetic hyperglycemia has been correlated with complications such as retinopathy, nephropathy, neuropathy, and increased risk of myocardial infarction, stroke, premature peripheral arterial disease, atherosclerosis, and cardiomyopathy (Nathan et al. 2014 ; Forbes and Copper 2013 ). Control of hyperglycemia has been extensively correlated to reduced complications (Nathan et al. 2014 ). It is now known that diabetic hyperglycemia is clinically cor-related with mild infl ammation (Jenkins et al. 2008 ). This chapter is an overview of the progress made in biochemical relationships of the pathogenesis of diabetes and infl ammation. In Chap. 6 we further explore the tools used in the assessment of hyperglycemia and diabetic infl ammation by reporting clinical testing.
As the pre-diabetic patient progresses to diabetes, signifi cant insulin resistance is observed. Hyperglycemia worsens with increased insulin resistance, as insulin is the major controller of glucose metabolism. It is produced in the pancreas where its secretion is stimulated by a rise in blood glucose. Insulin binds to adipose to inhibit
M. Pugia , PhD Strategic Innovation , Siemens Healthcare Diagnostics , 3400 Middlebury Street , Elkhart , IN 46515 , USA e-mail: [email protected]
4
lipolysis and activates the liver and muscle to store glucose as glycogen, and it pro-motes glucose uptake into muscle and adipose tissues through glucose transporter 4 (GLUT4) (Dugani 2005 ). The insulin resistant state occurs when insulin receptor responsiveness is reduced. Signifi cant progress has been made in the identifi cation of factors that impact insulin sensitivity, such as adipokines, infl ammation signal-ing, complement factors, and immune cell activation. The recently discovered adi-pokine hormones such as leptin, adipionectin, and ghrelin also impact insulin action. Additionally, hormones such as growth hormone (GH), insulin-like growth factor (IGF), insulin-like growth factor binding protein (IGFBP), epinephrine, glucagon, and cortisone are known to infl uence glucose metabolism and secretion and impact insulin action. However, known biomarkers still do not adequately predict insulin resistance (Goldfi ne et al. 2011 ). In Chap. 4 we explore the role of reduced insulin degradation during insulin resistance.
1.1.1 Glycation Stress
Hyperglycemia causes biological stress that produces advance glycation end- products (AGEs) that alter proteins, lipids, glycans, and nucleic acids (Goh et al. 2008 ). AGE are formed either by non-enzymatic glycosylation between reducing sugars and amine residues or by autoxidation of glucose. Non-enzymatic reactions
≥25%15%–19%<10%
20101990
Increase in Obesity
Unmeet need
•Diabetes in US causes 173 billion in costs•Complication cost of disease growing at >25% •HbA1c insensitive to early diabetes•Need to sort emerging diabetics with high specificity •OGTT is to difficult do and clinicians need a spot test
Pre-diabetics
79 mm
Type 1Diabetes
5 mm
Type 2Diabetes
23 mm
Insulin resistance
Hypertrophic adipose
Beta cell loss
IGT Glucose intolerant
Mechanism progress Fat cell signalingInflammation activationInsulin degradationComplement activationInnate immune activation
Leptin ,Adiponectin, etc TNF activationIDE, AdipoR1 CTF
gCD59, C1, C3a,b Adipose involvement, Bik
Fig. 1.1 Diabetes impact on health care cost. The increase in obesity in the U.S. over the last 30 years has increased to more than 25 % of the population in many states. The associated rise in diabetes is now estimated to be 79 million pre-diabetic and 23 million type 2 diabetic. This leaves an unmet need for diagnostics to sort emerging diabetics with highly specifi c and sensitive treat-ments to stall progression and reduce the cost of complications. Mechanistic progress has been made on several fronts and is reviewed in this chapter
M. Pugia
5
of protein amino groups with glucose form a ketoamine adduct (Amadori product) that is dependent on incubation time and glucose concentration. Lysine is the pre-ferred non-enzymatic glycation site over hydroxylysine and arginine (Garlick et al. 1988 ). AGE can form with most molecules in human biological fl uids and tissues at multiple glycation sites (Zhang et al. 2008 ; Ramírez-Boo et al. 2012 ). Basal glyca-tion is typically detectable in relatively high concentration with normal patient sam-ples, and diabetic patient samples are about two to fourfold higher (Calvo et al. 1988 ; Ramírez-Boo et al. 2012 ).
The discovery that glycated hemoglobin (HbA1c) is an AGE component of human hemoglobin turned out to be a useful diagnostic tool for diabetic care (Rahbar 2005 ). Today, HbA1c is well correlated with glycemic control and diabetic compli-cations (Nathan et al. 2014 ). This correlation is based on the intensive glucose con-trol to reduce carotid intima-media thickness and levels of the HbA1c. HbA1c is also highly correlated with the extent and duration of hyperglycemia over 2–3 months. The duration of the prediction is directly due to the natural turnover of red blood cells. Glycation of human serum albumin (HSA) has also been extensively studied and shown to increase in diabetic patients (Wa et al. 2007 ). Non-traditional AGE markers of hyperglycemia (glycated albumin, fructosamine, 1,5- anhydroglucitol (1,5-AG)) are responsive to glycemic control over shorter periods (Poon et al. 2014 ; Parrinello and Selvin 2014 ). While nontraditional markers are often linked to vas-cular complications, they are often found to be inversely related to obesity.
Glycation stress and AGE formation are often proposed as having key mechanis-tic roles in biological damage. AGE formation on endothelial cells and monocytes has been tied to nephropathy, retinopathy, and low grade infl ammation in humans (Goh et al. 2008 ). In the kidneys, AGE formation on glomerular basement mem-brane (GBM) is correlated with membrane thickening, mesangial matrix expansion, and increased collagen, fi bronectin, and laminin (Throckmorton et al. 1995 ; Skolnik et al. 1991 ). Guanidine-insoluble collagen is the site extensively glycated on GBM (Garlick et al. 1988 ). This AGE formation is observed during fi brosis with increased smooth muscle migration and proliferation. Fibrosis requires activation by connec-tive tissue growth factors and Transforming Growth Factor- β (TGF- β). Glycation of lipids is also detectable. For example, high-density lipoprotein (HDL) levels increase by about 400 % in diabetic patients (Calvo et al. 1988 ). AGE of lipid and proteins are theorized to preserve cellular signal patterns in a pathologically stressed state until they are cleared from the cell (Ceriello et al. 2009 ). Inhibition by thera-peutic agents has been shown to reduce AGE formation (Goh et al. 2008 ). However, how AGE stimulate molecular damage is not completely clear.
Receptors for AGE (RAGE) are often implicated as a pathway for molecular damage. RAGE are part of the immunoglobulin superfamily of receptors and lectin. RAGE are found to respond to a wide range of proteins through the glycated lysine residues (Neeper 1992 ). However, the binding of these AGE ligands is not suffi cient to induce infl ammatory signals. The key ligand primarily responsible for RAGE stimulation appears to be calgranulin (S100), a protein that is secreted from innate immune cells (neutrophils and monocytes) independent of glycation (Bierhaus et al. 2005 ; Valencia et al. 2004 ). Activation of RAGE appears to be mainly immune cell
1 Impact of Infl ammation and Innate Immunity Response in Obesity Mediated Diabetes
6
mediated and a pathway for molecular damage. Activation increases vascular endo-thelial cell expression of intercellular adhesion molecule-1 (ICAM-1) and vascular cell adhesion molecule-1 (VCAM-1) and causes innate immune cells to secrete tumor necrosis factors -α (TNF-α) and TGF-β (Valencia et al. 2004 ; Wautier et al. 1996 ). These cytokines induce the infl ammatory pathways involving the nuclear factor kappa-light-chain-enhancer of activated B cells (NFκB) and the mitogen- activated protein kinase phosphatase (MAPK) pathways (Ramasamy et al. 2011 ). Additionally, S100 binding of RAGE on innate immune cells activates phospholi-pase D 2 and phosphokinase C (PKC) which are thought to generate additional reac-tive oxygen species (ROS) by the conversion of nicotinamide adenine dinucleotide phosphate-oxidase (NADPH) to NADPH oxidase (NOX) (Pendyala et al. 2009 ; Curran and Bertics 2011 ).
In summary, while glycation stress and glycemic control can be measured by current AGE markers, there has been only a weak molecular connection between the AGE markers and the causation of complications. Next generation AGE markers seek to make stronger connections with the causation of complications. Chapter 2 explores glycated CD59 as one of these next generation markers. The glycation of CD59 occurs at a specifi c site that blocks CD59 from inhibiting formation of mem-brane attack complex (MAC) (Halperin et al. 2000 ). As a result, glycation increases MAC cell lysis during complement formation. Chapter 2 also reviews the impor-tance of complement in diabetic complications.
1.1.2 Oxidative Stress
Hyperglycemia is shown to increase oxidative stress and the production of ROS. Oxidative stress is a normal byproduct of the aerobic metabolism, where oxy-gen is lost to the formation of superoxide anion free radicals as the predominate ROS (Pitocco 2013 ; Seifert et al. 2010 ). The mitochondria electron transport chain produces ROS during the production of adenosine triphosphate (ATP) from the gly-colytic pathway (citric acid cycle). Hyperglycemia causes the glycolytic pathway to breakdown into more glucose producing NADH (Yan 2014 ). Overproduction of NADH increases mitochondrial oxidative stress byproducts by NADH dehydroge-nase. Additionally, the consumption of glycolytic pathway byproducts by the hexoamine, methlglyloxyal, enediol, and diahydroxyacetone pathways cause addi-tional ROS. Diacylglycerol also activates NOX through PKC and leads to additional ROS (Pendyala et al. 2009 ).
The result of ROS is damage to DNA, lipid, and proteins (Piconi 2003 ; Yan 2014 ; Seifert et al. 2010 ). ROS leads to additional free radicals which cause further damage. For example, production of reactive nitrogen species is formed by free radical exchange with nitric oxide that is produced by nitric oxide synthase (NOS) (Pitocco 2013 ). Oxidative stress is correlated with diabetic infl ammation, insulin sensitivity, and β-cell dysfunction (Bashan et al. 2009 ; Poitout and Robertson 2008 ). Oxidative stress is also associated with aging, cancer, atherosclerosis, neuropathy,
M. Pugia
7
nephropathy, retinopathy, obesity, hypertension, cardiovascular diseases, and heart failure (Lee and Wie 2007 ; Storz 2005 ; Lim et al. 2008 ; Madamanchi et al. 2005 ; Kanauchi et al. 2002 ; Knight 1997 ; Wilkinson-Berka et al. 2013 ; Pitocco 2013 ). ROS are implicated in causing epigenetic changes, endothelial dysfunction, impact-ing vascular tone, apoptosis, cell adhesion, and immune responses (Lim et al. 2008 ; Lee and Wie 2007 ; Zhang and Gutterman 2007 ; Gutterman et al. 2005 ; Chiarugi et al. 2003 ; Grisham 2004 ). Immune cells cause oxidative stress by increasing ROS through immune cell NOX (Hou et al. 2008 ). ROS impact many cellular processes such as proliferation and apoptosis (Droge 2002 ). Infl ammation, growth factors, and cytokines also impact ROS formation. Overall, oxidative stress is one plausible mechanism for biological damage, but it is non-specifi c to hyperglycemia.
1.2 Impact of Obesity
1.2.1 Adipose Signaling
Obesity clearly leads to diabetes, and it increases the infl ammatory response (Brady et al. 2012 ; Horakova et al. 2011 ). Obesity also worsens hyperglycemia and increases ROS and AGE generation. The current epidemic of obesity has increased the population of pre-diabetes with impaired glucose tolerance, insulin resistance, and hypertrophic adipose fat ( See Fig. 1.1 ). The discovery of the hormone, leptin; a product of the obese (ob) gene, opened the door for discoveries of new regulatory factors originating from adipose fat (Zhang et al. 1994 ). Adipose tissue is now accepted as a source of adipokine signals that impact metabolism and are linked to obesity mediated diabetes (Preedy and Hunter 2011 ). Adipose tissue is a complex matrix and contains various cell types such as adipocytes, macrophages, fi broblasts, adipocyte precursors, and others (Ahima 2006 ; Espinoza-Jiménez et al. 2012 ). Of the molecules released primarily from adipocytes during obesity, leptin, and adipo-nectin (Adpn) have the strongest correlated impact on glucose metabolism (Wang and Nakayama 2010 ). Leptin and Adpn have the structural characteristics of a cyto-kine; therefore, they are called adipokines.
Leptin reduces food intake (loss of appetite, satiety) and increases energy expendi-ture which leads to weight loss (Gordon 2003 ). It is produced by the ob gene, and its defi ciency causes obesity. Chapter 5 describes leptin resistant animals which become obese and spontaneously develop diabetes. Leptin signals the level of adipose energy stores to the brain’s hypothalamus and causes decreases in food intake and increases its energy expenditure. This results in more fat oxidization. Leptin activates the Janus kinase-signal transducer-activator of transcription (JAK STAT) pathway to stimulate inhibition of neuropeptides and activation of several anorexigenic peptides (Sweeney 2002 ; Gordon 2003 ). It has an effect on the central nervous system (CNS), it stimu-lates fatty acid oxidation in muscle, and it controls triglyceride synthesis in the liver (Minokoshi et al. 2002 ; Cohen et al. 2002 ). Leptin contributes to insulin resistance and is involved in the release of infl ammatory cytokines (Smith 2012 ).
1 Impact of Infl ammation and Innate Immunity Response in Obesity Mediated Diabetes
8
Adiponectin has structural similarity to the C1q/TNF family of proteins (Preedy and Hunter 2011 ). It activates AMPK and upregulates peroxisome proliferator- activated receptor (PPAR α) to enhance glucose uptake and fatty acid oxidation in muscle (Yamauchi et al. 2002 ) ( See Chap. 5 for further details ). Adpn suppresses glucose output in the liver and improves insulin sensitivity through a mechanism that is not well understood (Wang et al. 2005 ). Adpn is also present in CNS and peripheral tissue. Levels of Adpn correlate inversely with obesity; whereas Leptin levels increase in obesity (Satoh 2004 ). Decreased Adpn is associated with diabetes and cardiovascular disease. Adpn also has anti-infl ammatory properties, where it suppresses cytokines, adhesion compounds, and lipid uptake in monocytes (Yamauchi et al. 2002 ; Ouchi et al. 2000 ). Chapters 5 and 10 show how cellular models can be used to explore cell signal transduction of Adpn and other biochemi-cal substances on the insulin-dependent pathways of phosphatidylinositol 3-kinase (PI3K), protein kinase B (Akt), and insulin-independent pathways of AMP-activated protein kinase (AMPK).
Even with well-established cell and animal models in place, the mechanism by which Adpn delivers improved insulin sensitivity and anti-infl ammatory effects is still disputed. Adpn defi cient mice are reported to show small changes in insulin resistance and glucose intolerance (Kubota et al. 2002 ), while other reports show that Adpn defi cient mice do not show any improvement (Ma et al. 2002 ). Additional stud-ies report higher hepatic glucose production in Adpn defi cient mice (Meada et al. 2002 ). The response appears to be impacted by the combination of a high fat diet and Adpn defi ciency. There is also response dependency on the ob gene and PPARα agonists (Nawrocki et al. 2006 ). Responsiveness correlates with higher tumor necro-sis factors -α (TNF-α), mRNA expression, and reduced insulin receptor activity via an unknown mechanism (Meada et al. 2002 ). Some Adpn oligomers inhibit growth factors (e.g. platelet-derived growth factor (PDGF), fi broblast growth factor (FGF), and epidermal growth factor (EGF)), which explains cytokine production and activa-tion of NFκB (Wang et al. 2005 ). Adpn also stimulates caspase mediated cell apop-tosis during the pro-infl ammatory response (Bråkenhielm et al. 2004 ). The apoptosis and cytokine stimulation aligns with innate immune mediated events that are further described in Sect. 1.4.2 , rather than a direct stimulation by Adpn. The responsiveness to insulin secretion might be indirectly related to infl ammatory signaling.
Chapters 4 and 5 describe how the adiponectin receptor (AdipoR) has a direct impact on insulin by inhibition of insulin-degrading enzyme (IDE) and on TNFα release by inhibition of TACE (TNFα alpha cleavage protease). This new fi nding adds to the previously known functions of the AdipoR to cause glucogenisis and fatty acid oxidation through the AMPK and PPARα pathways (Kadowaki et al. 2005 ). Chapter 3 reviews the discovery of free adiponectin receptor C terminal frag-ment (AdipoR CTF) in human plasma and the development new assays for mea-surement of this signaling receptor. Chapters 4 and 5 make use of these new assays in biochemical studies, animal and cell models, and human clinical trials to explore the impact of AdipoR CTF on the innate immune mediated pathway.
Adipose tissue has additional complex interactions with the brain and other peripheral organs with additional regulatory factors. Ghrelin is a key component of
M. Pugia
9
this interaction and originates from the gastrointestinal tract and indirectly impacts the adipose by functioning as the central “hunger hormone on the brain” (Gumbs et al. 2005 ). As a para gut hormone, ghrelin is an endogenous ligand of the growth hormone secretagogue receptor and stimulates the production of neuropeptide like peptide Y (PPY) through the JAK STAT pathway. Ghrelin levels increase before meals, and it is thought to be involved in meal initiation and shows a rapid postpran-dial decline. Ghrelin administration causes increased caloric intake and initiation of hunger.
Several additional regulatory factors have weaker connection to glucose metabo-lism in obesity. Some are primarily produced and secreted by adipose. These include omentin (interlectin), visfatin (nicotinamide phosphoribosyltransferase), and resis-tin. Other regulatory factors are secondarily secreted by adipose, like retinol- binding protein 4 and vaspin. While they all generally correlate with obesity, the relevance of these regulatory factors is still in question (Preedy and Hunter 2011 ).
1.2.2 Free Fatty Acid Storage
Adipose is the major site of energy storage in the form of triglyceride (TG). Adipocyte can release free fatty acids (FFA) and TG when the body’s energy level demands it. Adipose, FFA, and TG are controlled by several regulatory factors from adipocytes and other tissue. Acylation stimulating protein (ASP), fatty acid-binding protein (FABP), adipose factor (aka angiopoietin-like protein 4), and cytosolic fatty acid synthase (FAS) are examples.
ASP is produced in adipose and other tissue, and it stimulates glucose uptake, clearance of TG, and it inhibits TG release (Cianfl one and Maslowska 1995 ; Cianfl one et al. 2004 ). ASP is produced by an interaction with complement C3 (aka ASP precursor), factor B, and complement factor D (aka adipsin). All can be pro-duced by adipose. ASP increases with increases in weight, diabetes, obesity, meta-bolic syndrome, and dietary fat (chlyomicrons) (Faraj et al. 2008 ; Cianfl one 2008 ). ASP regulates TG synthesis by activating diacylglycerol acyltransferase (DGAT). However, some researchers report no impact of ASP defi ciency in animal models (Phieler et al. 2013 ). Other question the role that ASP plays, whether implicated in complement-mediated injury or in regeneration interaction with complement C3 (Phieler et al. 2013 ).
In healthy individuals, hypoglycemia decreases insulin levels and increases lipase breakdown of TG into FFA (Stralfors and Honnor 1989 ). Albumin and FABP transport fatty acids to muscle and liver for fatty acid oxidation. FABP is widely secreted in the body and causes fatty acid uptake and intra-cellular transport (Chmurzyńska 2006 ). Adipose FABP increases with obesity and correlates with adipose tissue mass and body mass index (BMI) but is poorly predictive of insulin resistance (Horakova et al. 2011 ). Fatty acid oxidation and release of energy in mus-cle and liver occur by activation and transport into the mitochondria, β-oxidation, and the electron transport chain.
1 Impact of Infl ammation and Innate Immunity Response in Obesity Mediated Diabetes
10
Hyperglycemia increases insulin levels and signals the liver and muscle to stop fatty acid oxidation and also signals the adipose to store FFA and TG. Under a con-tinued high energy diet in a low exercise environment, increased fat storage and resulting obesity will occur. Large vacuoles of triglyceride fat accumulate in the liver and adipose. This causes steatosis in the liver and greater lipid loading of adi-pocytes and increased adipose mass in the adipose, which leads to hypertrophic and hypoxic adipose tissue. Adipose factor is produced by adipose and other tissue to inhibit storage of FFA and TG (Preedy and Hunter 2011 ). It is produced in the liver and contributes to the development of diabetic dyslipidemia; it inhibits lipoprotein lipase in adipose tissue lipolysis; and it raises plasma TG and FFA (Mandard et al. 2006 ). However adipose factor has poor diagnostic predictive value for diabetic onset.
Adipose fat synthesis is stimulated by insulin directly by stimulating pyruvate dehydrogenase and the acetyl-CoA carboxylase enzymes and indirectly by glucose uptake, which then stimulates nuclear receptors. Synthesis of most fatty acid moi-eties of the membrane lipids is done by FAS, which synthesizes fatty acids from malonyl CoA. Synthesis and metabolism of both fatty acids and cholesterol lipids is through transcriptional regulation by nuclear receptors such as sterol regulatory ele-ment binding proteins (SREBP), farnesoid X receptor a (FXR), and the liver X receptors (LXR). Activation causes transcription of genes for cholesterol synthesis and the LDL receptor (Ulven et al. 2005 ; Calkin 2013 ). Glucose induces FXR gene expression in a dose- and time-dependent manner. FXR expression decrease triglyc-eride levels by regulating the expression of several lipid-modulating proteins.
1.3 Infl ammation in Obesity and Diabetes
With the development of obesity, there is an increase in adipocyte mass, greater lipid loading (TG, FFA, chylomicrons), tissue hypertrophy, and reduced vascular-ization of the adipose tissue. The reduced capacity of the bloodstream to oxygenate adipose tissue results in hypoxia. Hypoxia then produces a pro-infl ammatory response from adipose through hypoxia-inducible factor (HIF) (Nizet and Johnson 2009 ; Trayhurn et al. 2008 ). Alternative complement pathway reviewed in Chap. 2 is activated in hypoxia stress via C1q complement which causes deposits, tissue injury, and further cellular activation (Collard et al. 1999 ). Alternative complement pathway is not triggered directly by glucose or insulin in high fat diets (Peake et al. 2005 ). Complement activation in adipose tissue, liver, and pancreas occurs under conditions of oxidative stress with upregulated complement receptors (Phieler et al. 2013 ). This activation is highly varied and not completely understood. Plasma C3 levels do appear to be higher in patients with diabetes, and this is mainly from adi-pose activation. However, complement activation is observed in the vicinity of apoptotic cells in the liver, kidney, and pancreas. Islets cells accumulate extracellu-lar amyloid fi brils which can trigger local complement activation via C1q (Phieler et al. 2013 ).
M. Pugia
11
Dysfunctional adipose tissue produces cytokines such as TNFα, TGF-β, and interleukins (IL-1, IL-6, IL-8 IL-18), as well as chemokines like macrophage inhibi-tory cytokine (MIC-1), monocyte chemotactic protein (MCP), and cytokine-induced neutrophil chemoattractant (CINC-1) (Wang and Nakayama 2010 ; Preedy and Hunter 2011 ). Lipocalins such as neutrophil gelatinase-associated lipocalin (NGAL, aka Lipocalin-2 (LCN-2)) are released by neutrophils in obesity (Preedy and Hunter 2011 ). Chitinase-3-like protein 1 (YKL-40, aka CHI3L1) is secreted by macro-phages in obesity. Regulatory factors that impact the endothelial, such as endothelin (ET), thrombospondin (TSP-1), plasminogen activator inhibitor 1 (PAI-1), apelin, and angiotensinogen, are also secreted by adipose in addition to primary excretion by other tissue. The regulatory endothelial factor adrenomedullin is primarily pro-duced by epicardial fat.
The innate immune and complement responses increase as obesity progresses to type 2 diabetes (Espinoza-Jiménez et al. 2012 ; Phieler et al. 2013 ). The innate cells, which include neutrophils, eosinophiles, monocytes, natural killer cells, and macro-phages, increase in diabetes and obesity. In Chap. 7 we review how diabetics are prone to urinary tract infection (UTI) and systemic infections like Upper Respiratory Infection (URI). In Chaps. 7 and 8 we show that non-infective infl ammatory activa-tion is common in diabetes and obesity. Chronic mild infl ammation in the absence of infection is now widely accepted and supported by mechanisms for activation of infl ammation, complement, and innate immunity in diabetes (Tilg 200; Phieler et al. 2013 ; Espinoza-Jiménez et al. 2012 ).
1.3.1 Endothelial Dysfunction
Clinical and experimental data support a link between systemic infl ammation and endothelial dysfunction in diabetes and obesity (Roberts and Porter 2013 ). Endothelial dysfunction occurs in diabetes with impaired endothelial-dependent vasodilation. Disruption of systems controlling vascular permeation can cause endothelial dysfunction. The most signifi cant modulator of endothelial vasodilation is nitric oxide that is produced by endothelial nitric oxide synthase (eNOS) in endo-thelial and muscle cells. Nitric oxide also inhibits platelet aggregation, cell prolif-eration, and immune cell adhesion. Cell adhesion molecules (VCAM-1 & ICAM-1) are reduced. Meanwhile the renin-angiotensin-aldosterone system (RAAS) is the counter balance system, as it increases vasoconstriction, blood pressure, and fl uid volume when angiotensin-converting enzyme forms angiotensin II.
These systems that control vascular permeation are critical to the innate immune response and to tissue repair because they allow migration, adhesion, innate immune cell response, cellular proliferation growth, and remodeling (Pacurari et al. 2014 ). During normal infl ammation that results from injury or infection, muscle cells release serine protease and kallikrein, to form kinins, and they cause vasodilation and hypotension, increased capillary permeability, and infi ltration of innate immune cells to the site of injury (Campbell 2000 , 2001 ). Kinin formation is another mecha-
1 Impact of Infl ammation and Innate Immunity Response in Obesity Mediated Diabetes
12
nism for nitric oxide release and promotion of natriuresis and diuresis. Angiotensin- converting enzyme also reduces kinins. There is a direct correlation between vasodilation and increased blood fl ow to the site of infl ammation with the loss of endothelial barrier being a hallmark of infl ammation.
Endothelial dysfunction starts in diabetes and obesity when hyperglycemia and adipose signals exert signaling control on the endothelial barrier that overrides nor-mal physiological control. Increased vascular permeability occurs in all diabetic vascular complications. Hyperglycemia is shown to diminish normal nitric oxide signaling and reduce permeability by increasing ROS which reduces nitric oxide levels (Roberts and Porter 2013 ). Hyperglycemia also increases vascular permeabil-ity by increasing the kallikrein response from smooth muscle (Feener et al. 2013 ; Riad 2007 ). Hypertrophic adipose diminishes nitric oxide through production of cytokines such as TNF-α, which activates NFκB and impairs eNOS expression (Besler 2011 ). Insulin resistance is another key promotor of vascular dysfunction, as insulin is less able to play a protective role of controlling eNOS expression via Akt- PI3K (Aroor et al. 2013 ).
Fat tissue in an obese individual impacts additional signals that both increase and decrease endothelial permeability. Adipose angiotensin is secreted during obesity and activates the RAAS counter balance system in the absence of injury. Adipose angiotensin increases blood pressure, angiotensin II levels, volume expansion, sodium retention, and it reduces glucose uptake (Preedy and Hunter 2011 ; Aroor et al. 2013 ). Circulating lipids cause activation of PKC which increases expression of vasoconstrictor endothelin (ET-1) (Nishizuka 1995 ). Adrenomedullin is secreted by adipose and increases with cytokine signaling, and it increases vasodilation and hypotension by increasing nitric oxide (Preedy and Hunter 2011 ). Meanwhile, ape-lin is expressed by adipose tissue in endothelial cells to increase vasodilation and hypotension by increasing nitric oxide production through the expression of eNOS during AMPK activation (Preedy and Hunter 2011 ). Adiponectin activation of AMPK also causes eNOS expression (Osuka et al. 2012 ).
Chronic endothelial activation occurs in diabetic and obese individuals even in the absence of infection caused by hyperglycemic and fatty acid stress (Tsuriya et al. 2011 ). This low level chronic activation is thought to be pathological in nature as it increases vascular dysfunction and contributes to both micro-vascular and macro-vascular complications (Feener et al. 2013 ; Kaneko et al. 2011 ). Vasodilation and increased blood fl ow typically allow the innate immune response to preform tissue repair starting with immune cell migration, adhesion, and response and lead-ing to cellular proliferation signaling, tissue growth, and remodeling (Pacurari et al. 2014 ). Constant tissue remodeling leads to stiffness in the extra cellular matrix of the vascular system and loss of the elasticity of endothelium with increased endo-thelial and muscle cell layers. The elastin to collagen ratio is reduced. This reduced elasticity allows damaging innate immune cells to continuously infi ltrate tissue and the vascular bed. Cellular proliferation is mediated through the MAPK/extracellular signal-regulated kinase (MAPK/Erk) pathway. The MAPK/Erk pathway also pro-motes expression of the potent vasoconstrictor endothelin (ET-1) to reverse the vas-cular dilation (Weil et al. 2011 ).
M. Pugia
13
Chapter 9 reviews the innate anti-infl ammatory response which is activated in both diabetes and obesity and increases with vascular dysfunction. Chapters 7 , 8 , 9 , and 10 review the role of Bikunin (Bik) as a key anti-infl ammatory component of this response. Bik is a serpin that inhibits the trypsin family of serine proteases and inhibits kallikrein, vascular dilation, smooth muscle contraction, and coagulation. Additionally, Bik inhibits blood coagulation and serine protease degradation of the endothelial through action of Factors IXa, Xa, XIa and XIIa on plasmin (Kato 2002 ; Pugia et al. 2007 ). In the acute phase of tissue injury, Bik is protective in function, stalling immune mediated tissue damage ( See Chap. 10 ). In the chronic phase, Bik stalls tissue repair and cellular proliferation. A prolonged and chronic anti- infl ammatory response is thought to impair the vascular system. In long standing infl ammatory reactions to chronic diabetic and cardiovascular disease, Bik will be elevated above the normal range for healthy individuals but at levels below the acute response range. Tissue factor pathway inhibitor (TFPI) is another Kunitz-type pro-tease inhibitor that inhibits the initial reactions of blood coagulation and PAR (Kato 2002 ). Adipose tissue has also been implicated as well as several other serpins. PAI-1 is a serpin that increases in obesity due to cytokine expression and causes obesity associated fi brinolysis (Preedy and Hunter 2011 ; Pugia et al. 2007 ). Vaspin is a serpin that is released from adipose and found to inhibit kallkrien (Preedy and Hunter 2011 ). The impact of prolonged exposure of serpins on innate immunity activation is a current active topic in chronic disease pathways.
1.3.2 Immune Mediated Apoptosis
Chronic low-grade infl ammation in adipose tissue induces an innate immune cell response in type 2 diabetes (T2D) and obesity (McNelis and Olefsky 2014 ). Adipose tissue retains macrophages that are recruited by the pro-infl ammatory profi le of cytokines (IL-1β, IL-6, IL-18β, TNFα, TGF-α/β), chemokines (MIC-1, MIP-2, MCP-1, CINC-1), YKL-40, and LCN-2. These pro-infl ammatory signals are released by adipose tissue due to fatty acid and oxidative stress (Espinoza-Jiménez et al. 2012 ) ( See Fig. 1.2 ). Lipids retained in the sub-endothelial space are a leading cause of infi ltration of macrophages. Activation of complement C3 and C5 also participates in macrophage recruitment and polarization (Phieler et al. 2013 ). Adipocyte macrophages are recently suggested to increase the ability of metabolic processing of adipocytes. At the same time macrophages are thought to accelerate the progression of insulin resistance and the development of diabetes (Johnson and Milner 2012 ; Patel et al. 2013 ). Macrophages also release cytokines and chemo-kines which further recruit other innate immune cells (Tilg and Moschen 2006 ).
These innate immune cells not only infi ltrate the adipose but also increasingly infi ltrate across the vascular endothelium into liver, muscle, and pancreas tissues as diabetes progresses. Besides the production of the pro-infl ammatory profi le, stressed adipose tissue also releases resistin (Preedy and Hunter 2011 ). Resistin, also known as adipose tissue-specifi c secretory factor (ADSF), upregulates adhesion molecule
1 Impact of Infl ammation and Innate Immunity Response in Obesity Mediated Diabetes
14
(VCAM-1 & ICAM-1) and chemokine (C-C motif) ligand 2 (CCL2), which cause leukocyte recruitment (Olefsky and Glass 2010 ). Circulating lipids in general cause activation of PKC, which increases and signaling for immune cell adhesion through VCAM-1 & ICAM-1 (Nishizuka 1995 ). Free fatty acids can be recognized by toll- like receptors (TLRs) activation of c-Jun N-terminal kinases (JNK) and NF- κ B sig-naling in various leukocytes and macrophages. Levels of both JNK and NF- κ B are increased in diabetic patients (Masoodi et al. 2015 ; de Jong et al. 2014 ). Activated JNK can inhibit PI3K and promote insulin resistance in tissue.
Macrophages and other innate immune cells such as neutrophils, eosinophiles, monocytes, and natural killer cells have a key function to remove apoptotic and necrotic cells (Espinoza-Jiménez et al. 2012 ). Immune cells mediate the death of damaged cells by causing apoptosis. Innate immune cells and macrophages release serine proteases such as elastase, proteinase 3, cathespin, and granzyme B as a means to cause immune mediated apoptosis (Yang 1996 ; Kondo et al. 2013 ). The granules of neutrophiles in particular, contain these serine proteases in high concen-trations (Mania-Pramanik et al. 2004 ; Johnson et al. 1988 ; Nakatani and Takeshita 1999 ). Degranulation causes release of serine proteases. Cytotoxic T lymphocyte
Cytokine &growth factor
release
Cell shrinking and
membrane blebbing
Extrinsic death factor & NF-KB
activation
Growth factor& cytokine
stress signaling
Innateimmune
cellElastase
Extra cellularTNF-αIL-1βIL-18TGF / βα
IL-1βEGFR TNFR
FasL
Transcriptionof cytokineproduction
Caspase 9
NF-kB
MAPK p38
CytoC
Caspase 3,6,7
Mitochondria Caspase 8 Granzyme B
Apoptosis
DNA fragmentation and no repair
•Nuclear receptors
Transcriptioninitiation complex
•Cell cycle apparatus
Nucleus
Cytoplasm
DNA pol-helicase
MKK3/6
Serineproteaserelease
α
Fig. 1.2 Innate Immune Mediated Apoptosis. Innate immune cells such as neutrophils, eosino-philes, monocytes, natural killer cells, and macrophages release serine proteases such as elastase, proteinase 3, cathespin and granzyme B during infl ammation. These serine proteases mediate apoptosis through the extrinsic death receptor and mediate cytokine stress signaling through release of growth factors and cytokines (e.g. TNF α, IL-1β, IL-18β, and TGF α/β). The extrinsic death receptor signals the caspase apoptosis pathways. Growth factor increases MAPK p38 signal-ing of cytokine transcription. Granzyme B released from immune cells bypass the anti-apoptosis effect of nuclear transcription factor (NF-κB) by directly activating Caspase 3
M. Pugia
15
are natural killer cells that also release granzyme (A, B, H, M), tryptase 2, and mast cell proteases 1 (Trapani 2001 ; Kam et al. 2000 ). Human lymphocytes also contain cell surface elastase. These serine proteases are typically used to remove pathogens by destruction of cells. Activation of cell death occurs through cleavage of proteins in infl amed cells.
Immune mediated apoptosis occurs by the extrinsic death receptor (Yang 1996 ), and serine proteases release growth factors and cytokines to activate this apoptosis ( See Fig. 1.2 ). Cytokines activate the extrinsic death receptor and cause signaling of the caspase pathway. Growth factors increase MAPK p38 signaling of cytokine transcription. The serine protease, granzyme B, is released from immune cells and bypasses the anti-apoptosis effect of NF-κB by directly activating caspase 3. Apoptosis is observed in vivo and in vitro in adipose tissue (Sorisky et al. 2000 ). Bik is a key pathway for inhibition of immune mediated apoptosis (Pugia et al. 2007 ) ( See Chap. 10 for further review ). The complement system is another major compo-nent of cell death ( See Chap. 2 for further review ), and complement activation in hypoxia and ischemia-reperfusion oxidative stress results in cell death and impacts vascular homeostasis (Collard et al. 1999 ).
Cell death initiates an adaptive immune cell response (Kono et al. 2014 ). Macrophages trigger T-cell infi ltration into the adipose (Travers et al. 2015 ). T-lymphocytes, both CD4+ and CD8+, are found in the adipose of obese and dia-betic patients. The role of T-cell activation in the adipose is poorly understood, but it is clear that T-Cell with class II major histocompatibility complex (MHCII) increases in fat stressed adipose (Deng et al. 2013 ). These T-cells have antigen processing and presentation abilities. Macrophages and T cells further interact with each other for activation and differentiation and further trigger the adaptive immune system. Macrophages can express MHC-II molecules (Espinoza-Jiménez et al. 2012 ). Free fatty acids further activate the proliferation of T-cells. There is emerging data supporting different innate immune cell subpopulations and activation in the adipose (Rao et al. 2014 ; Olefsky and Glass 2010 ). Regulatory T cells (Treg cells) that are crucial for mediating immune homeostasis have been observed in the adi-pose (Pucino et al. 2014 ). Fatty acids are reported to enhance CD4(+) T-cell prolif-eration. However, aside from the extremely rare Weber-Christian panniculitis, there are no reports of auto-immune disease of the adipose (Allen-Mersh 1976 ).
1.3.3 Fibrosis
The innate immune system also mediates a repair process under normal conditions. Serine proteases are released to activate protease activate receptors (PAR) ( See Fig. 1.3 ), and these proteases are released by tissues under infl ammatory stress. Trypsin, thrombin, and plasmin are formed by proteolytic cleavage to activate protease- activated receptors (PAR) and lead to the cell signal cascade that affects cell shape, secretion, integrin activation, infl ammatory responses, transcriptional responses, and increased cell motility (Coelho et al. 2003 ). PAR signaling causes
1 Impact of Infl ammation and Innate Immunity Response in Obesity Mediated Diabetes
16
phospholipase C (PLC), adenylcyclase (AC), phosphatidylinositol 3-kinase (PLA2), protein kinase A (PKA), and PKC to activate MAPK/Erk (Cocks and Moffatt 2000 ; Cottrell 2002 ; Miike et al. 2001 ), and activation of MAPK/Erk increases cell prolif-eration. PAR activation also increases cytosolic phospholipase A2 synthesis of pros-taglandins, an intracellular hormone, causing nuclear receptor activation transcription of many proteins. Cell wall bound serine proteases (plasmin) are expressed and cleave urokinase-type plasminogen activator (uPA) and its receptor (uPAR). Activation of these receptors causes the breakdown of basement mem-branes and extracellular matrix during tissue remodeling.
Coagulation is another key component of tissue regeneration during infl amma-tion (Pugia et al. 2007 ). Infl ammation triggers coagulation (Kaplan et al. 1982 ). Vascular endothelium and muscle cells release coagulation serine proteases such as plasmin, thrombin, and Factors VII and X in the activation of the coagulation cas-cade. Coagulation, fi brinolysis, complement activation, and extra-cellular matrix assembly occur. Factors VII & X cause cleavage of pro-thrombin to thrombin dur-ing coagulation. Thrombin forms insoluble fi brin from fi brinogen and forms a mesh of fi bers (a clot) in which blood cells are trapped. Plasmin indirectly causes platelet
H
Increasedproteinexpression
Replication and transcription
• Nuclear receptors
•Transcription•initiation •complex
•Cell cycle apparatus
Increasedhormoneproduction
Cell proliferation and
differentiation
Nucleus
Extra cellularPlasminuPAR
activation
uPA
Membranesynthesis
PAR activation
DNA pol-ahelicase
uPAR
PKC
MAPK/ERK
PLA2PLCAC
PKA
RAS
Trypsinthrombin
Cytoplasm
Inflammation mediated cellproliferation
Fig. 1.3 Infl ammation mediated cell proliferation. Serine proteases trypsin and thrombin are released in endothelial and epithelial autocrine response during infl ammation and activate protease activate receptors ( PAR ). Activation causes signaling by phospholipase C ( PLC ), adenylcyclase ( AC ), and phosphatidylinositol 3-kinase ( PLA2 ); protein kinase C ( PKC ) leads to activation of MAPK/Erk. MAPK/Erk increases cell proliferation during healing repair through the cell cycle apparatus controlled by cyclin-dependent protein kinases. PAR activation also increases cytosolic phospholipase A2 (intracellular hormone) synthesis of prostaglandins, causing nuclear receptor activation transcription of many proteins. Cell wall bound serine proteases (plasmin) are expressed and cleave urokinase-type plasminogen activator ( uPA ) from its receptor ( uPAR ). Activation causes the break-down of basement membranes and extracellular matrix during tissue remodeling
M. Pugia
17
aggregation and endothelial cell proliferation. Thrombospondin 1 from platelets mediates cellular attachment and induces angiogenesis and proliferation in adipose along with activation of TGF-β (Kong et al. 2013 ).
Abnormal tissue repair occurs in diabetic individuals with obesity and leads to fi brosis. Fibrosis is common in diabetic complications and is found in liver, adipose, kidney, heart, nerves, eyes, and vascular system (Ban and Twigg 2008 ). Fibrosis is characterized by an excessive build-up of the extracellular matrix (ECM). Typically, a growth factor excess or defi ciency occurs in diabetic fi brosis ( See Chap. 11 for further detail ). This imbalance can be induced by innate immune cells. For exam-ple, macrophage infi ltration into the adipose causes fi brosis before adipocyte meta-bolic dysfunction and cell death (Vila et al. 2014 ). The presence of functional Toll-like receptor 4 on adipose tissue hematopoietic cells is necessary for the initia-tion of adipose tissue fi brosis. This imbalance can also be the result of TGF-β pro-moting fi brosis ( See Chap. 11 for further detail ).
Several innate immune cell factors that control regeneration and fi brosis are explored in this book. In Chap. 10 , the action of Bik on cell proliferation is explored, and a new pathway to reducing cell growth is described. Bik is a key pathway for protecting the extracellular matrix in arterial walls and connective tissues from abnormal cell growth (Pugia et al. 2007 ). Bik was previously known to inhibit PAR and uPA through its action on trypsin, plasmin, and thrombin. Bik also is known to inhibit coagulation through Factor VIIa (Pugia et al. 2007 ). In Chap. 10 , the action of Bik on cell proliferation and cell growth reduction is presented. Bik suppresses cell growth by activating the Akt-PI3K pathway. In Chap. 6 , the Adiponectin Receptor C terminal fragment (AdipoR CTF) is shown to increase intercellular insulin and reduce insulin sensitivity. These pathways are expected to open new doorways into the understanding of the impact the innate immune systems has on tissue regeneration.
1.4 Progression of Diabetes
It has been proposed that all types of diabetes mellitus are a continuous spectrum of the same disease (Brooks-Worrell and Palmer 2011 ). In theory, diabetes starts with insulin resistance and progresses to Type 2 diabetes as a metabolic disease with glucose intolerance. It then follows with stronger immune system involvement, and it ends with Type 1 diabetes as an autoimmune disease. To stop progression at the start of the spectrum is the goal. The initial relationship between obesity and insulin resistance is well-documented around the world. Lifestyle changes, such as calorie- restricted diets, exercise, and behavior modifi cation that result in weight loss of 5–10 % are the accepted means for signifi cant improvement in insulin resistance and diabetes (Gumbs et al. 2005 ).
At the other end of the spectrum, autoantigen-specifi c infl ammatory CD4+ and CD8+ T cells follow in Type 1 diabetes with destruction of pancreatic insulin- producing β cells. Insulin, glutamic acid decarboxylase, and protein tyrosine phos-
1 Impact of Infl ammation and Innate Immunity Response in Obesity Mediated Diabetes
18
phatase are the most common auto-antigens involved in this process (Winter and Schatz 2011 ). When the majority of β cells are destroyed, the pancreas’s ability to secrete insulin in response to blood glucose levels is lost and results in the disrup-tion of glucose homeostasis (Espinoza-Jiménez et al. 2012 ). As diabetes progresses, complications such as cardiovascular disease, nephropathy, retinopathy, and neu-ropathy increase. All diabetic complications share many common mechanistic behaviors such as oxidative stress, glycation stress, and innate immune involve-ment, and obesity adds the additional adipose and free fatty acid factors.
1.4.1 Cardiovascular Diseases
Diabetes increases the risk of myocardial infarction, stroke, premature peripheral arterial disease, atherosclerosis, and cardiomyopathy (Forbes and Copper 2013 ). Dyslipidemia, impaired glycemic control, and persistent elevations in blood pres-sure are common observation in diabetes. Diabetes is the leading contributor to the 88 million patients in US living with one or more types of cardiovascular disease and to the $444 billion in health care costs. Obesity additionally increases the risk of mortality and morbidity of cardiovascular disease (CVD) (Wang and Nakayama 2010 ).
Infl ammation, innate immunity activation, endothelial dysfunction, and comple-ment responses are common mechanisms in all diabetes complication. The endothe-lium is crucial for maintenance of vascular homeostasis and permeability. Endothelial dysfunction contributes to both micro-vascular and macro-vascular complications in diabetes. Innate immune cells can bind to the vessel wall and cause migration, adhesion and infi ltration of innate immune cell into tissue. The innate immune cell response leads to cellular proliferation growth, tissue remodeling, fi brosis, apoptosis, and T-cell infi ltration. The result is impaired muscle growth and development with reduced muscle mass, reduced myofi bril size, decreased skeletal muscle, capillarization, and angiogenesis (D’Souza et al. 2013 ).
1.4.2 Nephropathy
Kidney disease is a key focus for diabetic disease management as it represents the largest vascular system in the body. Diabetes is the lead cause of chronic kidney disease (CKD) and end stage renal disease (ESRD). It leads to 500,000 cases of ESRD in the US and $33 billion in cost. Infections and nephrotoxic exposure are additional causes of CKD and ESRD. The diabetic kidney disease mechanism is similar to that of diabetic cardiovascular diseases, as it starts with low level infl ammation, progresses through innate immune response, and ends with tissue damage that leads to CKD. Management of kidney disease requires instituting kidney protective strategies before getting to the CKD stage. The pathology
M. Pugia
19
typically starts with infl ammation as glomerulonephritis and ends with wide spread cell loss, fi brosis, vascular dysfunction with GBM thickening and mesan-gial expansion, complement formation with renal lesions, and auto-immunity such as IgA nephropathy (Cochrane et al. 1965 ; Nakatani et al. 2001 ; Huang et al. 2001 ; Hotta et al. 1999 ).
Neutrophilic polymorphonuclear leukocytes and macrophages are accepted as a primary cause of renal capillary wall injury in glomerulonephritis, and they accom-plish this by mediating the release of proteinases (Johnson et al. 1988 ; Nakatani and Takeshita 1999 ). Innate immune cells and ROS lead to TGF formation, increasing collagen IV synthesis, and glomerular mesangial cell growth that leads to glomeru-larsclerosis. Serine proteases are key factors that damage the basement membrane and lead to proteinuria and enlargement of the network structure of the basement membrane and charge barrier. Platelet coagulation and red blood cells also increase permeability of the basement membrane to proteins. This results in impaired kidney cell growth with reduced kidney mass and size, decreased capillarization, and angiogenesis.
Given the morbidity and mortality associated with kidney disease, there is a need to improve patient outcomes. Risk stratifi cation requires better resolution of the causes of kidney injury and better selection of effective therapies. Typically, determination of a loss of renal function is based on the prediction of the glumerular fi ltration rate (GFR). Current biomarkers used to detect and monitor kidney disease include urinary measurement of RBC, albumin, protein, α-1-microglobulin, β-2- microglobulin, transferrin and creatinine, and blood assays for creatinine, phospho-rus, BUN, albumin, protein, cystatin C, and other predictors of GFR. WBC, bacteriuria and nitrite in urine are used as additional predictors of infection. Chapter 7 discusses how once infection is ruled out, there is a need for assays to manage the chronic infl ammatory phase before GFR changes and progression to CKD.
1.4.3 Retinopathy
Diabetic retinopathy is the leading cause of vision loss in adults (Forbes and Copper 2013 ). Proliferative retinopathy is the major vision threatening disorder and is simi-lar in mechanism to microvascular injury in CVD and kidney disease. The inner retina infrastructure supports a high metabolic demand for functions but is limited in size as light must travel through the inner retina to reach the photoreceptors. The retina is vulnerable to oxidative stress, hypoxia, and glycemic stress (Antonetti 2006 ). Degeneration or occlusion of retinal capillaries occurs in diabetes. Diabetes retinopathy is also characterized by reduced growth and remodeling, fi brosis, retinal thickening, and cell death due to apoptosis. Macular edema occurs and is character-ized by a retinal thickening due to fl uid leakage from the breakdown of the blood barrier. This causes blurred vision with reduced visual acuity (Forbes 2013 ). Overall progression of diabetic complications are often studied by retinopathy as the micro-vascular changes are easily measured in the eye.
1 Impact of Infl ammation and Innate Immunity Response in Obesity Mediated Diabetes
20
1.4.4 Neuropathy
More than half of all individuals with diabetes eventually develop neuropathy (Forbes and Copper 2013 ). Diabetics experience sensitivities to vibrations and ther-mal thresholds. Again, the mechanism of glycation and oxidative stress leading to endothelial dysfunction and immune mediated cell damage is used to explain this pathology. Endothelial dysfunction is observed with basement membrane thicken-ing and diminished ability to oxygenate neurons. This dysfunction leads to hypoxia and nerve fi ber deterioration. Neuronal and glial cell death is also associated with infl ammatory signaling and immune mediated cell migration.
1.5 Conclusion
New discoveries in the innate and adaptive immune systems appear key to the rela-tionship of obesity and diabetes. The understanding of the interactions of infl amma-tion in diabetes and obesity is increasing, and several key observations can be made based on today’s understanding. Multiple infl ammatory and immune factors impact insulin sensitivity and glucose tolerance. The initiation of the infl ammatory response by glycemia, adipose, free fatty acid or oxidative stress routes is supported. Endothelial dysfunction and innate immunity activation mediate tissue damage and are consistent factors in the diabetes complications. Taken as a whole, infl ammation will remain an important research topic for new avenues in both the management and treatments of diabetes.
References
Ahima RS (2006) Adipose tissue as an endocrine organ. Obesity 14:242–249 Allen-Mersh TG (1976) Weber-Christian panniculitis and auto-immune disease: a case report. J
Clin Pathol 29(2):144–149 Antonetti DA et al. JDRF Diabetic Retinopathy Center Group. Diabetic retinopathy: seeing beyond
glucose-induced microvascular disease. Diabets. 2006;55(9):2401–2411 Aroor AA et al (2013) The role of tissue renin-angiotensin-aldosterone system in the development
of endothelial dysfunction and arterial stiffness (review). Front Endocrinol (Lausanne) 4:article 161, 7 pgs
Ban CR, Twigg SM (2008) Fibrosis in diabetes complications: pathogenic mechanisms and circu-lating and urinary markers. Vasc Health Risk Manag 4(3):575–596. Review
Bashan N et al (2009) Positive and negative regulation of insulin signaling by reactive oxygen and nitrogen species. Physiol Rev 89:27–71
Besler C (2011) Mechanisms underlying adverse effects of HDL on eNOS-activating pathways in patients with coronary artery disease. J Clin Invest 121(7):2693–2708
Bierhaus A et al (2005) Understanding RAGE, the receptor for advanced glycation end products. J Mol Med (Berl) 83(11):876–886
M. Pugia
21
Brady EM et al (2012) Investigating endothelial activation and oxidative stress in relation to gly-caemic control in a multiethnic population. Exp Diabetes Res 2012:386041
Bråkenhielm E et al (2004) Adiponectin-induced antiangiogenesis and antitumor activity involve caspase-mediated endothelial cell apoptosis. Proc Natl Acad Sci U S A 101(8):2476–2481
Brooks-Worrell B, Palmer JP (2011) Is diabetes mellitus a continuous spectrum? Clin Chem 57(2):2158–2161
Calkin AC, Tontonoz P (2013) Transcriptional integration of metabolism by the nuclear sterol activated receptors LXR and FXR. Nat Rev Mol Cell Biol 13(4):213–224
Calvo C, Ponsin G, Berthezene F (1988) Characterization of the non-enzymatic glycation of high density lipoprotein in diabetic patients. Diabete Metab 14(3):264–269
Campbell DJ (2000) Towards understanding the kallikrein-kinin system: insights from measure-ment of kinin peptides. Br J Med Biol Res 33:665–677
Campbell DJ (2001) The kallikrein-kinin system in humans. Clin Exp Pharmacol Physiol 28:1060–1065
Ceriello A, Ihnat MA, Thorpe JE (2009) Clinical review 2: the “metabolic memory”: is more than just tight glucose control necessary to prevent diabetic complications? J Clin Endocrinol Metab 94(2):410–415
Chiarugi P et al (2003) Reactive oxygen species as essential mediators of cell adhesion: the oxida-tive inhibition of a FAK tyrosine phosphatase is required for cell adhesion. J Cell Biol 161:933–944
Chmurzyńska A (2006) The multigene family of fatty acid-binding proteins (FABPs): function, structure and polymorphism. J Appl Genet 47(1):39–48
Cianfl one K, Maslowska M (1995) Differentiation-induced production of ASP in human adipo-cytes. Eur J Clin Invest 25(11):817–825
Cianfl one K et al (2004) Fasting acylation-stimulating protein is predictive of postprandial triglyc-eride clearance. J Lipid Res 45:124–131
Cianfl one K et al (2008) Acylation stimulating protein but not complement C3 associates with metabolic syndrome components in Chinese children and adolescents. Eur J Endocrinol 159(6):781–790
Cochrane CG, Unanue ER, Dixon FJ (1965) A role of polymorphonuclear leukocyte and comple-ment in nephrotoxic nephritis. J Exp Med 122:99–116
Cocks T, Moffatt J (2000) Protease-activated receptors: sentries for infl ammation. Trends Pharmacol Sci 21:103–108
Coelho AM, Ossovskaya V, Bunnett NW (2003) Proteinase-activated receptor-2: physiological and pathophysiological roles. Curr Med Chem Cardiovasc Hematol Agents 1:61–72
Cohen P et al (2002) Role for stearoyl-CoA desaturase-1 in leptin-mediated weight loss. Science 297:240–243
Collard CD et al (1999) Complement activation following oxidative stress. Mol Immunol 36:941–948
Cottrell G (2002) Protease activated receptors: the role of cell surface proteolysis in signaling. Essays Biochem 38:169–183
Curran CS, Bertics PJ (2011) Human eosinophils express RAGE, produce RAGE ligands, exhibit PKC-delta phosphorylation and enhanced viability in response to the RAGE ligand, S100B. Int Immunol 23(12):713–728
D’Souza D, Al-Sajeea D, Hawke TJ (2013) Diabetic myopathy: impact of diabetes mellitus on skeletal muscle progenitor cells. Front Physiol 4:Article 379, 7 pgs
de Jong AJ et al (2014) Fatty acids, lipid mediators, and T-cell function. Front Immunol 5:483 Deng T et al (2013) Class II major histocompatibility complex plays an essential role in obesity-
induced adipose infl ammation. Cell Metab 17:411–422 Droge W (2002) Free radicals in the physiological control of cell function. Physiol Rev 82:47–95 Dugani CD (2005) Glucose transporter 4: cycling, compartments and controversies (review).
EMBO Rep 6(12):1137–1142
1 Impact of Infl ammation and Innate Immunity Response in Obesity Mediated Diabetes
22
Espinoza-Jiménez A, Peón AN, Terrazas LI (2012) Alternatively activated macrophages in types 1 and 2 diabetes. Mediators Infl amm 2012:815953
Faraj M et al (2008) Regulation of leptin, adiponectin and acylation-stimulating protein by hyper-insulinaemia and hyperglycaemia in vivo in healthy lean young men. Diabetes Metab 34(4 Pt 1):334–342
Feener EP, Zhou Q, Fickweiler W (2013) Role of plasma kallikrein in diabetes and metabolism (review). Thromb Haemost 110(3):434–441
Forbes JM, Copper MS (2013) Mechanisms of diabetic complications. Physiol Rev 93:137–188 Garlick RL, Bunn HF, Spiro RG (1988) Nonenzymatic glycation of basement membranes from
human glomeruli and bovine sources. Effect of diabetes and age. Diabetes 37(8):1144–1150 Goh SY, Cooper ME et al (2008) Clinical review: the role of advanced glycation end products in
progression and complications of diabetes. J Clin Endocrinol Metab 93(4):1143–1152 Goldfi ne AB et al (2011) Biomarkers in fasting serum to estimate glucose tolerance, insulin sensi-
tivity, and insulin secretion. Clin Chem 57(2):326–337 Gordon P (2003) The clinical uses of leptin. Curr Opin Pharmacol 3:655–659 Grisham MB (2004) Reactive oxygen species in immune responses. Free Radic Biol Med
36:1479–1480 Gumbs AA et al (2005) Changes in insulin resistance following bariatric surgery and the adipoin-
sular axis: role of caloric restriction and weight loss. Obes Surg 15:462–473 Gutterman DD, Miura H, Liu Y (2005) Redox modulation of vascular tone: focus of potassium
channel mechanisms of dilation. Arterioscler Thromb Vasc Biol 25:671–678 Halperin J et al (2000) Molecular basis for a link between complement and the vascular complica-
tions of diabetes. Proc Natl Acad Sci U S A 97(10):5450–5455 Horakova DV et al (2011) Adipocyte fatty acid binding protein and C-reactive protein levels as
indicators of insulin resistance development. Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub 155(4):355–359
Hotta O, Yusa N, Ooyama M, Unno K, Furuta T, Taguma Y (1999) Detection of urinary macro-phages expressing the CD16 molecule: a novel marker of acute infl ammatory glomerular injury. Kidney Int 55:1927–1934
Hou N et al (2008) Reactive oxygen species-mediated pancreatic β-cell death is regulated by inter-actions between stress activated protein kinases, p38 and c-Jun N -terminal kinase and mitogen- activated protein kinase phosphatases. Endocrinology 149:1654–1665
Huang F, Horikoshi S, Kurusu A, Shibata T, Suzuki S, Funabiki K et al (2001) Urinary levels of interleukin-8 (IL-8) and disease activity in patients with IgA nephropathy. J Clin Lab Anal 15:30–34
Jenkins AJ et al. DCCT/EDIC Research Group (2008) Cross-sectional associations of C-reactive protein with vascular risk factors and vascular complications in the DCCT/EDIC cohort. Diabetes Complications 22(3):153–163
Johnson R et al (1988) The human neutrophils serine proteinases, elastase and cathespin G can mediate glomerular injury in vivo. J Exp Med 168:1169–1174
Johnson AR, Milner JJ, Makowski L (2012) The infl ammation highway: metabolism accelerates infl ammatory traffi c in obesity. Immunol Rev 249(1):218–238
Kadowaki T (2005) Adiponectin and adiponectin receptors. Endocr Rev 26(3):439–451 Kam CM, Hudig D, Powers JC (2000) Granzymes (lymphocyte serine proteases): characterization
with natural and synthetic substrates and inhibitors. Biochim Biophys Acta 1477:307–323 Kanauchi M, Nishioka H, Hashimoto T (2002) Oxidative DNA damage and tubulointerstitial
injury in diabetic nephropathy. Nephron 91:327–329 Kaneko H et al (2011) Human C-reactive protein exacerbates metabolic disorders in association
with adipose tissue remodeling. Cardiovasc Res 91:546–555 Kaplan A, Silverberg M, Dunn J, Ghebrehiwet B (1982) Interaction of the clotting, kinin forming,
complement, and fi brinolytic pathways in infl ammation. Ann N Y Acad Sci 389:25–38 Kato H (2002) Regulation of functions of vascular wall cells by tissue factor pathway inhibitor
basic and clinical aspects. Arterioscler Thromb Vasc Biol 22:539–548 Knight JA (1997) Reactive oxygen species and the neurodegenerative disorders. Ann Clin Lab Sci
27:11–25
M. Pugia
23
Kondo T et al (2013) Natural killer T cells in adipose tissue are activated in lean mice. Exp Anim 62(4):319–328
Kong P et al (2013) Thrombospondin-1 regulates adiposity and metabolic dysfunction in diet- induced obesity enhancing adipose infl ammation and stimulating adipocyte proliferation. Am J Physiol Endocrinol Metab 305(3):E439–E450
Kono H, Kimura Y, Latz E (2014) Infl ammasome activation in response to dead cells and their metabolites. Curr Opin Immunol 30:91–98
Kubota N et al (2002) Disruption of adiponectin causes insulin resistance and neointimal forma-tion. J Biol Chem 277(29):25863–25866
Lee HC, Wie YH (2007) Oxidative stress, mitochondrial DNA mutation, and apoptosis in aging. Exp Biol Med 232:592–606
Lim SO et al (2008) Epigenetic changes induced by reactive oxygen species in hepatocellular carcinoma: methylation of the E-cadherin promoter. Gastroenterology 135(6):2128–2140
Ma K et al (2002) Increased β-oxidation but no insulin resistance or glucose intolerance in mice lacking adiponectin. J Biol Chem 277(38):34658–34661
Madamanchi NR, Vendrov A, Runge MS (2005) Oxidative stress and vascular disease. Arterioscler Thromb Vasc Biol 25:29–38
Mandard S et al (2006) The fasting-induced adipose factor/angiopoietin-like protein 4 is physi-cally associated with lipoproteins and governs plasma lipid levels and adiposity. J Biol Chem 281(2):934–944
Mania-Pramanik J et al (2004) Elastase: a predictive marker of infl ammation and/or infection. J Clin Lab Anal 18:153–158
Masoodi M et al (2015) Lipid signaling in adipose tissue: connecting infl ammation & metabolism. Biochim Biophys Acta 1851:503–518
McNelis JC, Olefsky JM (2014) Macrophages, immunity, and metabolic disease. Immunity 41(1):36–48
Meada N et al (2002) Diet-induced insulin resistance in mice lacking adiponectin/ACRP30. Nat Med 8(7):731–737
Miike S, McWilliam A, Kita H (2001) Trypsin induces activation and infl ammatory mediator release from human eosinophils through protease-activated receptor-2. J Immunol 167:6615
Minokoshi Y et al (2002) Leptin stimulates fatty-acid oxidation by activating AMP-activated pro-tein kinase. Nature 415:339–343
Nakatani K, Takeshita S (1999) Vascular endothelial cell injury by activated neutrophil and treat-ment for the injury. Surg Trauma Immunol Response 8:112–114
Nakatani K et al (2001) Inhibitory effect of serine protease inhibitors on neutrophil-mediated endothelial cell injury. J Leukoc Biol 69:241–247
Nathan DM et al (2014) The diabetes control and complications trial/epidemiology of diabetes interventions and complications study at 30 years: overview. Diabetes Care 37(1):9–16
Nawrocki AR et al (2006) Mice Lacking Adiponectin show decreased hepatic insulin sensitivity and reduced pesponsiveness to peroxisome proliferator-activated receptor agonists. J Biol Chem 281(5):2654–2660
Neeper M (1992) Cloning and expression of a cell surface receptor for advanced glycosylation end products of proteins. J Biol Chem 267(21):14998–15004
Nishizuka Y (1995) Protein kinase C and lipid signaling for sustained cellular responses. FASEB J 9(7):484–496
Nizet V, Johnson RS (2009) Inter-dependence of hypoxic and innate immune responses. Nat Rev Immunol 9:609–617
Olefsky JM, Glass CK (2010) Macrophages, infl ammation, and insulin resistance. Annu Rev Physiol 72:219–246
Osuka K et al (2012) Adiponectin activates endothelial nitric oxide synthase through AMPK sig-naling after subarachnoid hemorrhage. Neurosci Lett 514(1):2–5
Ouchi N et al (2000) Adiponectin, an adipocyte-derived plasma protein, inhibits endothelial NF-κB signaling through a cAMP-dependent pathway. Circulation 102:1296–1301
1 Impact of Infl ammation and Innate Immunity Response in Obesity Mediated Diabetes
24
Pacurari M et al (2014) The Renin-angiotensin-aldosterone system in vascular infl ammation and remodeling international (review). J Infl amm 2014:Article ID 689360, 13 pgs
Parrinello CM, Selvin E (2014) Beyond HbA1c and glucose: the role of nontraditional glycemic markers in diabetes diagnosis, prognosis, and management. Curr Diab Rep 14(11):548
Patel PS, Buras ED, Balasubramanyam A (2013) The role of the immune system in obesity and insulin resistance. J Obes 2013:Article ID 616193, 9 pgs
Peake PW et al (2005) Response of the alternative complement pathway to an oral fat load in fi rst- degree relatives of subjects with type II diabetes. Int J Obes (Lond) 29(4):429–435
Pendyala S et al (2009) Regulation of NADPH oxidase in vascular endothelium: the role of phos-pholipases, protein kinases, and cytoskeletal proteins (forum review article). Antioxid Redox Signal 11(4):841–860
Phieler J, Garcia-Martin R, Lambris JD, Chavakis T (2013) The role of the complement system in metabolic organs and metabolic diseases. Semin Immunol 25(1):47–53
Piconi L (2003) Oxidative stress in diabetes. Clin Chem Lab Med 41(9):1144–1149 Pitocco D (2013) Oxidative stress in diabetes: implications for vascular and other complications.
Int J Mol Sci 14:21525–21550 Poitout V, Robertson RP (2008) Glucolipotoxicity: fuel excess and β-cell dysfunction. Endocr Rev
29:351–366 Poon AK et al (2014) Comparative associations of diabetes risk factors with fi ve measures of
hyperglycemia. BMJ Open Diab Res Care 2(1):e000002 Preedy VR, Hunter RJ (eds) (2011) Adipokines, 1st edn. CRC Press LCC, New York, NY.
pp 1–458. ISBN 9781578086894 Pucino V et al (2014) Regulatory T cells, leptin and angiogenesis. Chem Immunol Allergy
99:155–169 Pugia MJ, Valdes R, Jortani SA (2007) Chapter on urinary trypsin inhibitors: structure, biological
relevance and measurement. In: Makowski G (ed) Advances in clinical chemistry, vol 44. Academic Press. Elsevier, New York, NY. ISBN 10: 0-12-373704-4
Rahbar S (2005) The discovery of glycated hemoglobin: a major event in the study of nonenzy-matic chemistry in biological systems. Ann N Y Acad Sci 1043:9–19
Ramasamy R, Yan SF, Schmidt AM (2011) Receptor for AGE (RAGE): signaling mechanisms in the pathogenesis of diabetes and its complications. Ann N Y Acad Sci 1243:88–102
Ramírez-Boo M et al (2012) Characterization of the glycated human cerebrospinal fl uid proteome. J Proteomics 75(15):4766–4782
Rao X, Zhong J, Sun Q (2014) The heterogenic properties of monocytes/macrophages and neutro-phils in infl ammatory response in diabetes. Life Sci 116(2):59–66
Riad A (2007) The role of the renal kallikrein–kinin system in diabetic nephropathy. Curr Opin Nephrol Hypertens 16(1):22–26
Roberts AC, Porter KE (2013) Cellular and molecular mechanisms of endothelial dysfunction in diabetes. Diab Vasc Dis Res 10(6):472–482
Satoh N. Leptin to adiponectin ratio as a potential atherogenic index in obese type 2 diabetic patients. Diabetes Care. 2004;27(10):2488–2490
Seifert EL et al (2010) Electron transport chain-dependent and -independent mechanisms of mito-chondrial H 2 O 2 emission during long-chain fatty acid oxidation. J Biol Chem 285(8):5748–5758
Skolnik EY et al (1991) Human and rat mesangial cell receptors for glucose-modifi ed proteins: potential role in kidney tissue remodelling and diabetic nephropathy. J Exp Med 174:931–939
Smith MM (2012) Minson CT obesity and adipokines: effects on sympathetic overactivity. J Physiol 590(Pt 8):1787–1801
Sorisky A, Magun R, Gagnon AM (2000) Adipose cell apoptosis: death in the energy depot. Int J Obes 24(Suppl 4):S3–S7
Storz P (2005) Reactive oxygen species in tumor progression. Front Biosci 10:1881–1896 Stralfors P, Honnor RC (1989) Insulin-induced dephosphorylation of hormone-sensitive lipase.
Eur J Biochem 182(2):379
M. Pugia
25
Sweeney G (2002) Leptin signaling. Cell Signal 14:655–663 Throckmorton DC et al (1995) PDGF and TGF-β mediate collagen production by mesangial cells
exposed to advanced glycosylation end products. Kidney Int 48:111–117 Tilg H, Moschen AR (2006) Adipocytokines: mediators linking adipose tissue, infl ammation and
immunity. Nat Rev Immunol 6:772–783 Trapani JA (2001) Granzymes: a family of lymphocyte granule serine proteases. Genome Biol
2:30141–30147 Travers RL et al (2015) The impact of adiposity on adipose tissue-resident lymphocyte activation
in humans. Int J Obes (Lond) 39:762–769 Trayhurn P, Bohan Wang B, Wood SI (2008) Hypoxia in adipose tissue: a basis for the dysregula-
tion of tissue function in obesity? Br J Nutr 100:227–235 Tsuriya D et al (2011) Signifi cant correlation between visceral adiposity and high-sensitivity
C-reactive protein (hs-CRP) in Japanese subjects. Intern Med 50:2767–2773 Ulven SM et al (2005) LXR is crucial in lipid metabolism prostaglandins. Leukot Essent Fatty
Acids 73:59–63 Valencia JV et al (2004) Binding of receptor for advanced glycation end products (RAGE) ligands
is not suffi cient to induce infl ammatory signals: lack of activity of endotoxin-free albumin- derived advanced glycation end products. Diabetologia 47(5):844–852
Vila IK et al (2014) Immune cell toll-like receptor 4 mediates the development of obesity- and endotoxemia- associated adipose tissue fi brosis. Cell Rep 7(4):1116–1129
Wa C et al (2007) Characterization of glycation adducts on human serum albumin by matrix- assisted laser desorption/ionization time-of-fl ight mass spectrometry. Clin Chim Acta 385(1–2):48–60
Wang X, Nakayama T (2010) Infl ammation, a link between obesity and cardiovascular disease. Mediators Infl amm 2010:Article ID 535918, 17 pgs
Wang Y et al (2005) Adiponectin inhibits cell proliferation by interacting with several growth fac-tors in an oligomerization-dependent manner. J Biol Chem 280(18):18341–18347
Wautier JL et al (1996) Receptor-mediated endothelial cell dysfunction in diabetic vasculopathy soluble receptor for advanced glycation end products blocks hyperpermeability in diabetic rats. J Clin Invest 97(1):238–243
Weil BR et al (2011) Enhanced endothelin-1 system activity with overweight and obesity. Am J Physiol Heart Circ Physiol 301(3):H689–H695
Wilkinson-Berka JL et al (2013) Reactive oxygen species, Nox and angiotensin II in angiogenesis: implications for retinopathy (review). Clin Sci (Lond) 124(10):597–615
Winter WE, Schatz DA (2011) Autoimmune markers in diabetes. Clin Chem 57(2):168–175 Yamauchi T et al (2002) Adiponectin stimulates glucose utilization and fatty-acid oxidation by
activating AMP-activated protein kinase. Nat Med 8(11):1288–1295 Yan L (2014) Pathogenesis of chronic hyperglycemia: from reductive stress to oxidative stress
(review). J Diab Res 2014:137919 Yang JJ (1996) Apoptosis of endothelial cells induced by the neutrophil serine proteases proteinase
3 and elastase. Am J Pathol 149(5):1617–1626 Zhang DX, Gutterman DD (2007) Mitochondrial reactive oxygen species-mediated signaling in
endothelial cells. Am J Physiol Heart Circ Physiol 292:H2023–H2031 Zhang Y et al (1994) Positional cloning of the mouse obese gene and its human homologue. Nature
372(6505):425–432 Zhang Q et al (2008) Proteomic profi ling of nonenzymatically glycated proteins in human plasma
and erythrocyte membranes. J Proteome Res 7(5):2025–2032
1 Impact of Infl ammation and Innate Immunity Response in Obesity Mediated Diabetes
29© Springer International Publishing Switzerland 2015 M. Pugia (ed.), Infl ammatory Pathways in Diabetes: Biomarkers and Clinical Correlates, Progress in Infl ammation Research, DOI 10.1007/978-3-319-21927-1_2
Chapter 2 Complement and Complement Regulatory Proteins in Diabetes
Jose A. Halperin , Pamela Ghosh , Michael Chorev , and Anand Vaidya
J. A. Halperin , MD (*) • P. Ghosh , PhD • M. Chorev , PhD Laboratory for Translational Research, Division of Hematology, Department of Medicine , Brigham and Women’s Hospital, Harvard Medical School , Boston , MA , USA e-mail: [email protected]
A. Vaidya , MD, MMSc Division of Endocrinology, Diabetes and Hypertension , Brigham and Women’s Hospital, Harvard Medical School , Boston , MA , USA
Abstract In this chapter we have summarized the body of evidence that supports a role for the complement system and complement regulatory proteins in the patho-genesis of diabetic complications, with specifi c emphasis on the novel phenomenon of glycation- inactivation of CD59. Progress in this area has been remarkable con-sidering that (1) ethical and practical issues preclude conducting in-depth mechanis-tic studies in humans, especially for chronic conditions such as the complications of diabetes that take years to develop and (2) non-clinical research is hampered due to the absence of animal models of diabetes that fully recapitulates diabetic vascular disease in the intensity and distribution seen in humans.
Similar to other tightly regulated biological systems, phenotypic endpoints derived from complement activation depend on the delicate balance between com-plement activators and inhibitors. In the specifi c case of complement and its regula-tors, this “delicate balance” paradigm is well exemplifi ed by the human disease Paroxysmal nocturnal hemoglobinuria (PNH), an acquired complement-mediated hemolytic anemia in which blood cells lack glucosylphosphatidylinositol (GPI)-anchored proteins including CD59 (Takeda et al., Cell 73:703–711, 1993; Parker et al., Stem Cells 14:396–411, 1996). Under “normal” conditions, the basal “tick-over” activation of complement is suffi cient to induce only mild degree of hemoly-sis. In contrast, when stressors like infections amplify complement activation, excessive membrane attack complex (MAC) formation on unprotected red blood cells and platelets lacking CD59 results in “paroxysmal” crises of hemolysis, plate-let activation and potentially lethal thrombotic episodes (Hill et al., Blood 121:4985–4996, 2013; Hillmen et al., N Engl J Med 355:1233–1243, 2006; Risitano et al.,
30
Immunobiology 217:1080–1087, 2012). The emerging clinical evidence summa-rized in this article indicates that comparable mechanisms could be operative in the pathogenesis of diabetes complications: on the one hand, biological responses to chronic hyperglycemia like formation of advanced glycated end-products (AGEs) and fructosamine adducts in proteins as well as auto-antibodies against oxidized and/or glycated proteins could activate complement while, on the other hand, glycation- inactivation of CD59 increases the susceptibility of tissues to comple-ment (MAC) mediated damage.
2.1 Introduction
Diabetes is reaching epidemic proportions worldwide and is projected to affect almost 10 % of the world population by year 2035. However, an epidemic of diabe-tes is in fact an epidemic of its complications; diabetes is associated with (1) accel-erated macrovascular disease resulting in atherosclerotic coronary heart disease, stroke and peripheral artery disease, and (2) microvascular diseases that damage the retina leading to blindness, the kidneys leading to end-stage renal failure, and peripheral nerves leading to severe forms of neuropathy, which combined with peripheral artery disease are the leading cause of non-traumatic amputations. The cost of treating complications of diabetes exceeds 10 % of the total healthcare expenditure worldwide.
Large-scale prospective studies for both type 1 and type 2 diabetes including the Diabetes Control and Complications Trial (DCCT) study group (DCCT reports 1995a , b ), the UK Prospective Diabetes Study (UKPDS) (UKPDS reports 1998 ), and the Steno-2 study (Pedersen and Gaede 2003 ) established that the complications of diabetes are caused by prolonged hyperglycemia, and that the extent of tissue damage in individuals with diabetes is infl uenced by genetic determinants of susceptibility and by the presence of accelerating factors such as hypertension and dyslipidemia. A hypothesis summarizing different mechanisms that may underlie the pathogenesis of diabetes complications proposes that hyper-glycemia-induced overproduction of reactive oxygen species (ROS) fuels an increased fl ux of sugars through the polyol pathway, an increased intracellular formation of advanced glycation end-products (AGEs), an increase in reactive carbonyl compounds, an increased expression of the receptor for AGEs and sig-naling from its activating ligands; the activation of protein kinase C (PKC) iso-forms, and an overactivity of the hexosamine pathway (Brownlee 2001 , Giacco and Brownlee 2010 ). However, the actual cellular and molecular mechanisms by which high levels of glucose cause tissue damage in humans are still not fully understood.
A body of clinical and experimental evidence reported in the past few decades supports a link between the complement system, complement regulatory proteins and the pathogenesis of diabetes complications (Acosta et al. 2000 ; Rosoklija et al. 2000 ; Zhang et al. 2002 ; Flyvbjerg 2010 ; Mellbin et al. 2012 ; Hansen et al.
J.A. Halperin et al.
31
2004 ; Hansen 2005 ; Hovind et al. 2005 ; Uesugi et al. 2004 ; Woroniecka et al. 2011 ; Qin et al. 2004 ; Engstrom et al. 2005 ; Wlazlo et al. 2014 ; Gehrs et al. 2010 ; Gerl et al. 2002 ). Emerging evidence also indicates that the complement system is involved in several features of cardiometabolic disease including dysregulation of adipose tissue metabolism, low-grade focal infl ammation, increased expression of adhesion molecules and pro-infl ammatory cytokines in endothelial cells contribut-ing to endothelial dysfunction, and insulin resistance (Hertle et al. 2014 ). Herein we review the biology of complement with particular emphasis on the membrane attack complex (MAC) as a potential effector of pathology seen in target organs of diabetic complications, and of CD59, an extracellular cell membrane-anchored inhibitor of MAC formation that is inactivated by non-enzymatic glycation form-ing glycated CD59 (GCD59), an emerging biomarker for the diagnosis and clini-cal management of diabetes.
2.2 The Complement System
The complement system is an effector of both adaptive and innate immunity. It is composed of more than 30 plasma and cell-membrane proteins that are synthesized by hepatocytes or locally in peripheral tissues and normally circulate as inactive precursors (pro-proteins) (Morgan and Gasque 1997 ; Laufer et al. 2001 ; Andoh et al. 1998 ; Passwell et al. 1990 ). Complement proteins interact with one another in three enzymatic activation cascades known as the classical, the alternative and the mannose-binding lectin (MBL) pathways (See Fig. 2.1 ) (Morgan and Harris 1999 ; Walport 2001a , b ). The three activation pathways eventually converge at the level of C3 and C5; thereafter, the three complement pathways share a common reaction sequence involving the late components C6, C7, C8, and C9, leading to the genera-tion of the membrane attack complex or MAC, the main effector of complement- mediated tissue damage.
2.2.1 Membrane Attack Complex (MAC)
The formation and function of MAC is described in Fig. 2.1 . The MAC is a trans- membrane pore that allows infl ux of salt and water leading to colloidosmotic lysis of MAC-targeted cells. The formation of the MAC pore, however, seems to be the end stage of a process that transits through a reversible phase during which MAC pores can be formed transiently (Halperin et al. 1993 , 1988 ; Acosta et al. 1996 ) generating signifi cant changes in the membrane permeability and internal composition of MAC-targeted cells without compromising their viability (Halperin et al. 1993 , 1989 ; Berger et al. 1993 ). Figure 2.1 illustrates the large increase of intracellular Ca ++ levels caused by transient, non-lytic MAC pores on isolated guinea-pig cardio-myocytes (Berger et al. 1993 ). Thus, while the matured MAC pore can kill non-self
2 Complement and Complement Regulatory Proteins in Diabetes
32
cells by perforating their plasma membrane, the transient MAC pore can induce non-lethal biological responses in “self” cells, and mediate physiological and/or pathological response (Nicholson-Weller and Halperin 1993 ).
2.2.2 MAC Cellular Signaling and Complement Activation
The MAC is a mediator of cellular signaling and an effector of organ pathology. Activation products of complement such as C3b and C4b or the anaphylotoxins C3a and C5a exert their biological activities through binding to specifi c receptors expressed in many cell types. Upon binding to their cognate ligands, these receptors trigger signaling processes mostly linked to infl ammation and immune responses (Carroll 1998 ). In contrast, there are no specifi c receptors for the MAC, which
MASP1
H20MBL
MASP2
Ba
C3bB
Antigen-antibody
Anaphylatoxins
Histamine release
C3
C3
Classical pathway Lectin pathway Alternative pathway
CR1 CR2MCP DAF
D
B
C9 C9
C8C7C6
C5b
MAC
C4
C3bBbC3 convertase
C4b2a
C4b2a3b C3bBbC3b*P
C5b
C5
C5a
C1qC1rC1s
C2b
Factor HFactor I
C3a
C4a
C2
P
C8C7
C6
C9n
C3 convertase
C5 convertase
C5aC3aC4a
Cell lysisSignaling pathwaysPro-inflammatory Pro-thrombotic mitogenic
C5 convertase
CD59Clusterin, S-protein
C3 hydrolysisPathogen surface
J.A. Halperin et al.
33
Fig. 2.1 The Complement Activation Pathways and its regulators. The Complement System: The classical pathway is initiated by immune complexes that activate C1. Activated C1 then recruits C4 and C2 to form a C3-convertase (C4b2a) that fragments C3 into C3a, an anaphylotoxin, and C3b, a molecule at the core of the complement system that binds covalently to hydroxyl groups on car-bohydrates or proteins in bacterial surfaces and/or cell membranes. C3b tags invading microorgan-isms for opsonization and also amplifi es the activation cascade acting as a focal site for further complement activation. The phylogenetically older alternative pathway is activated by low-grade cleavage of C3 in plasma; the resulting C3b binds factor B, a protein homologous to C2, to form the C3bB complex. Bound to C3b, factor B is cleaved by the activator factor D, which circulates in an active form as a serine esterase to form C3bBb: the C3-convertase of the alternative pathway . Because basal “tick over” deposition of C3b occurs in all cells exposed to complement, C3b is always available to prime the alternative pathway . However, continued activation and amplifi ca-tion is normally restricted because cell membranes do not bind factor B effi ciently, and the inhibi-tory factor H prevents the association of factor B with C3b. Amplifi cation of the alternative pathway occurs when factor H binds to foreign carbohydrates leaving factor B free to complex with C3b. Specifi city for activation of the alternative pathway resides in the ability of factor H to discriminate between “self” and foreign carbohydrate determinants (Walport 2001a , b ; Morgan 1995 ). The MBL pathway is initiated when the plasma mannose-binding lectin (MBL) protein, structurally similar to C1q (Matsushita et al. 1998 ), in a complex with the mannose-binding lectin- associated proteases 1 and 2 (MASP1 and MASP2) binds to an array of mannose groups on the surface of microorganisms. This binding activates MASPs, which then cleave C2 and C4 leading to the formation of the C3-convertase C4b2a in a manner comparable to activated C1 (Walport 2001a ). Terminal complement pathway: The three complement activation pathways eventually converge at the level of C3 with formation of C3b and a C5 convertase that cleaves C5 into C5a, an anaphylotoxin, and C5b. This is the last enzymatic step of the activation cascades; thereafter, the complement pathways share a common sequence through the terminal components C6, C7, C8, and C9, leading to the generation of the membrane attack complex or MAC, the main effector of complement-mediated tissue damage. The terminal complement components – C6 through C9 -necessary to form the MAC, are constitutively present in plasma. Formation of the MAC is initiated by C5b followed by the sequential association of C6 into C5b6, and then C7, C8, and C9. C5b6 is a stable protein that binds to C7 to form the C5b-7 complex that inserts into the lipid bilayer of the plasma membrane. Membrane-bound C5b67 exposes a binding site for C8 leading to formation of the C5b-8 complex, which functions as a docking site for C9. A single C9 fi rst binds to C8 in the C5b-8 complex; then C9 polymerizes forming the C5b-9 complex known as the membrane attack complex or MAC. The MAC is a circular polymer of variable number of C9 monomers (Podack 1984 ; Podack and Tschopp 1984 ; Podack et al. 1982 ; Podack and Tschopp 1982 ; Tschopp et al. 1984 ) with the capacity to insert into cell membranes (Laine and Esser 1989 ) and form a transmembrane pore with hydrophobic domains on the outside and hydrophilic domains in the inside, and an effective internal radius of 5–7 nm (Mayer 1984 ; Ramm et al. 1982 , 1985 ). Restriction of complement activity by complement regulators: In the fl uid phase, the C1 inhibitor regulates the classical pathway by targeting C1, factors H and I regulate the alternative pathway, and S-protein, clusterin and serum lipids compete with membrane lipids for reacting with nascent C5b67. In addition to these fl uid phase inhibitors, several membrane proteins, decay- accelerating factor (DAF) (Nicholson-Weller et al. 1985 ), membrane cofactor (MCP) (Brooimans et al. 1992 ; Cervoni et al. 1992 ) and complement receptors 1and 2 (CR1, CR2) regulate the C3-convertases (Beynon et al. 1994 ) a major amplifi cation step in the early complement activation cascade, while clusterin (McDonald and Nelsestuen 1997 ), S-protein (Podack and Muller-Eberhard 1978 ) and CD59 inhibit formation of the MAC (Fletcher et al. 1994 ; Bodian 1997 ; Davies et al. 1989 ; Huang et al. 2006 ; Husler et al. 1994 ; Meri et al. 1990 ; Petranka et al. 1992 ). Massive or unrestricted activation of complement and MAC formation leads to cell lysis, as seen in immune hemolytic anemias or PNH. Basal (tick over) activation with or without reduced restriction in nucleated cells leads to transient MAC pores formation, infl ux and effl ux of ions and macromolecules, and activa-tion of cellular pathways that contribute to proliferation, infl ammation and thrombosis as charac-teristically seen in target organs of diabetic complications
2 Complement and Complement Regulatory Proteins in Diabetes
34
inserts directly into the lipid bilayer of the plasma membrane. Insertion of transient non-lethal MAC pores into cell membranes induces an array of biological responses that promote cell proliferation, infl ammation and thrombosis, as characteristically seen in all diseases in which focal complement activation is a characteristic patho-logical abnormality. These responses are mediated by growth factors and cytokines including bFGF and interleukin-1 (Benzaquen et al. 1994 ; Acosta et al. 1996 ) which are directly released through the MAC pore, and PDGF (Benzaquen et al. 1994 ), monocyte chemotactic protein-1 (MCP-1) (Torzewski et al. 1996 ) and von Willebrand factor (Hattori et al. 1989 ) which are released through canonical secre-tion pathways in response to intracellular perturbations induced by the MAC (See Fig. 2.2 ). Acting auto- or para-crinically on extracellular receptors, these leaked/secreted molecules promote (a) proliferation of fi broblasts, endothelial and smooth muscle cells (Benzaquen et al. 1994 ), (b) infl ammation through the expression of pro-infl ammatory adhesion molecules such as P-selectin, VCAM-I and E-selectin (Marks et al. 1989 ; Libby 2002 ), and the attraction of monocytes and macrophages
Fig. 2.2 Non-lytic transient MAC triggers a large increase of intracellular Ca ++ concentration ([Ca ++ ] i ). The fi gure shows pseudo-color images of isolated adult rat ventricular muscle cell loaded with fura-2, a fl uorescent calcium-indicator dye. Increasing emission intensities were asso-ciated with increasing [Ca ++ ] i , coded in a pseudo-color spectrum from blue (low Ca ++ ) to red (high Ca ++ ). ( a ) Reference, digitalized phase-contrast image. Before adding complement factors, the cell shows low emission intensities ( yellow-green ) ( b ) that does not signifi cantly change after addition of C5b6 plus C7 and C8 ( c ). After addition of C9, the cell showed a marked emission increase ( d – e ), representing an increase in [Ca ++ ] i that lasted ≈1 min. After an additional 3 min, [Ca ++ ] i decreased to near baseline levels ( f – h ) (Figure reproduced from Berger et al. 1993)
J.A. Halperin et al.
35
to the site of focal complement activation (Kostner 2004 ) and (c) thrombosis through the expression of pro-thrombotic tissue factor, and the transient assembly of the prothrombinase complex in both the endothelial cell surface and microvesicles shed from their plasma membrane (Hamilton et al. 1990 ; Christiansen et al. 1997 ; Saadi et al. 1995 ). Our group has also shown that the MAC promotes in vitro accumula-tion of cholesteryl esters into mouse macrophages fostering their conversion into foam cells (Wu et al. 2009a ).
Reversible insertion of transient MAC into cell membranes also activates intra-cellular signaling pathways (Tegla et al. 2011 ) including increased (a) Ca ++ infl ux (Acosta et al. 1996 ; Morgan and Campbell 1985 ) and Ca ++ -activated K + effl ux (Halperin et al. 1989 ) (b) production of reactive oxygen species (ROS) (Adler et al. 1986 ; Bellosillo et al. 2001 ), (c) activation of Ca ++ -sensitive and Ca ++ -insensitive protein kinase C (PKC) (Cybulsky et al. 1990 ; Carney et al. 1990 ), (d) activation of heterotrimeric G proteins of the Gi/Go subfamily (Niculescu et al. 1994 ), (e) mitotic signaling through the small G-protein Ras, that induces Raf-1 translocation trigger-ing activation of the ERK pathway (Niculescu et al. 1997 ), and (f) activation phos-phatidylinositol- 3 kinase (PI3-K) (Niculescu and Rus 2001 ). The MAC also activates the transcription factor NF-κB, which induces the production of IL-6 as well as IL-8 and MCP-1 (Kilgore et al. 1997 ; Viedt et al. 2000 ). In glomerular mesangial cells, insertion of the MAC results in increased production of ROS (Adler et al. 1986 ) and the release and/or up-regulation of (a) bFGF and PDGF (Benzaquen et al. 1994 ) which induce mesangial cell proliferation and transcriptional up-regula-tion of TGF-β (Torbohm et al. 1990 ; Lovett et al. 1987 ); (b) monocyte chemotactic protein- 1 (MCP-1), which attracts monocytes and macrophages to the mesangium (Stahl et al. 1993 ); and (c) transcriptional activation of collagen type IV (Wagner et al. 1994 ).
Important in the context of this chapter, many mediators and cellular signaling pathways activated by the MAC are established players in the tissue damage seen in the target organs of diabetic complications (Kotajima et al. 2000 ; Adler et al. 2000 ) and are common to or intersect with complement-independent mechanisms up- regulated by hyperglycemia such as overproduction of ROS, up-regulation of PKC and NF-κB (Brownlee 2001 , 2005 ; Giacco and Brownlee 2010 ).
2.2.3 Complement Regulation Through CD59
Given the potentially devastating effects of uncontrolled complement activation, it is not surprising that an array of mechanisms have evolved to protect “self” cells from complement attack and MAC formation. On the one hand, the inherent insta-bility of the enzymes in the activation pathways and the extremely short half-life of the activated complement components coupled to the transient nature of the reaction sequences are intrinsic properties of the complement system that keep its activation restricted and limit the effects of complement to the vicinity of the activation site. On the other, a multitude of complement regulatory molecules exist that restrict critical steps of the pathways (See Fig. 2.1 ).
2 Complement and Complement Regulatory Proteins in Diabetes
36
In particular, CD59 is a specifi c inhibitor of MAC formation that restricts C9 polym-erization. Ubiquitously expressed in mammalian cells, CD59 is an 18–20 kDa glyco-sylphosphatidylinositol (GPI)-linked glycoprotein (Kooyman et al. 1995 ; Morgan 1995 ). CD59 inhibits the binding of C9 to C5b-8 by competing for binding to a nascent epitope on C8 thereby affecting the association of C9 to the C5b-8 complex (Ninomiya and Sims 1992 ). CD59 can not only function as an inhibitor of the formation of large MACs but can also allow cells to eliminate newly formed MACs by blocking the early, functional channel formation of MAC complexes (Kimberley et al. 2007 ).
Several lines of evidence indicate that CD59 is the most critical of the comple-ment regulators in protecting human cells from MAC formation and MAC-induced phenomena: erythrocytes from a family carrying the Inab blood group phenotype totally lacked decay-accelerating factor (DAF), an inhibitor of early complement activation, without any associated hemolytic disease (Lin et al. 1988 ; Telen and Green 1989 ; Reid et al. 1991 ; Sun et al. 1999 ); in contrast, individuals carrying CD59 mutations leading to undetectable levels of CD59 on their cell membranes exhibited an early onset hemolytic phenotype associated with vascular disease or chronic relapsing polyneuropathy (Yamashina et al. 1990 ; Nevo et al. 2013 ).
The GPI anchor in GPI-anchored proteins is introduced as a post-translational mod-ifi cation mediated by a protein called Pig-A that is required for the synthesis of the GPI-lipid tail (Luzzatto and Bessler 1996 ; Miyata et al. 1993 , 1994 ; Takeda et al. 1993 ). Defi ciency of GPI-linked proteins including CD59 in erythrocytes and platelets is phenotypically expressed as a complement-mediated hemolytic anemia, known as paroxysmal nocturnal hemoglobinuria (PNH) (Halperin and Nicholson-Weller 1989 ; Parker 1996 ; Risitano 2012 ). This paradigmatic disease, together with the silencing CD59 mutations described above, illustrates how the delicate balance between com-plement activation and restriction can be perturbed by the functional defi ciency of CD59, leading to a potentially harmful pathological response even in the absence of complement- activating stressors.
2.3 Clinical Evidence for a Role of Complement in the Pathogenesis of Diabetes Complications
Evidence for a potential role of complement in the pathogenesis of diabetes compli-cations was initially provided by the identifi cation of activated complement proteins in target organs of diabetes complications biopsied or excised from individuals with diabetes (Rosoklija et al. 2000 ; Zhang et al. 2002 ; Falk et al. 1983a , b , 1987 ).
2.3.1 Diabetic Nephropathy
Falk et al. used antibodies against activated complement proteins including a neo-antigen of the C9 portion the MAC to perform immunoelectron (IE) and immuno-fl uorescent (IF) microscopy on kidney tissue from normal humans and individuals
J.A. Halperin et al.
37
with insulin-dependent diabetes (Falk et al. 1983a , b , 1987 ). Staining of tissues with specifi c antibodies directed against the MAC neoantigen is the most direct method for determining whether complement activation has occurred to completion because the neoantigen is exposed only when the MAC is fully assembled and inserted into a cell membrane. IF microscopy of kidney tissue from individuals with insulin- dependent diabetes and varying degrees of mesangial expansion and glomeruloscle-rosis demonstrated a positive association between the degree of tissue damage and the amount of MAC deposited in the mesangium. IE microscopy of specimens from individuals with diabetic nephropathy revealed that the anti-MAC neoantigen monoclonal antibody reacted with linear and circular membranous structures within the mesangium, tubular basement membranes, blood vessel walls, and the glomeru-lar basement membranes (Falk et al. 1983a , b , 1987 ). Our own studies confi rmed glomerular MAC deposition (co-localized with GCD59, Fig. 2.4b ) in kidney biop-sies from individuals with diabetes (Qin et al. 2004 ). We also reported the presence of extensive MAC deposition in glomeruli and blood vessels in the biopsy of a kidney transplanted 2 years earlier and diagnosed as “recurrent diabetic nephropa-thy with no signs of rejection” (Qin et al. 2004 ).
Also, several studies have found that serum concentrations of mannose-binding lectin (MBL), an indicator of the lectin complement pathway, were signifi cantly elevated in patients with type 1 diabetes (Bouwman et al. 2005 ; Hansen et al. 2003 ) and even more elevated in patients with vascular complications including diabetic nephropathy (Hansen et al. 2004 , 2010 ; Saraheimo et al. 2005 ). Further, the FinnDiane Study showed that the baseline levels of MBL were positively associated with the progression and prognosis of diabetic nephropathy in type-1 diabetes (Hovind et al. 2005 ; Kaunisto et al. 2009 ). Transcriptome analyses of human kid-neys have shown that the canonical complement-signaling pathway is differentially up-regulated in both glomeruli and tubules of individuals with diabetic nephropathy and is associated with increased glomerulosclerosis. Woroniecka et al. showed that increased expression of C3 mRNA and protein in glomeruli was positively correlated with diabetic kidney disease (Woroniecka et al. 2011 ). C3 is a key protein central to all complement activation pathways.
2.3.2 Diabetic Retinopathy
A combination of biochemical, histological and genetic studies suggest that aber-rant function of the complement system plays a key role in the etiology of age- related macular degeneration, a common degenerative ocular disease (Wlazlo et al. 2014 ). Two independent studies have documented that complement activation occurred extensively and to completion in the choriocapillaries of eyes from donors with diabetic retinopathy, while similar deposits were not found in eyes from donors without retinopathy. In the study by Gerl et al., intense positive staining for MAC deposits (neoantigens) and C3d was noted without exception in all 50 eyes with diabetic retinopathy analyzed but only in one of 26 eye donors without diabetic reti-nopathy (Zhang et al. 2002 ; Gerl et al. 2002 ).
2 Complement and Complement Regulatory Proteins in Diabetes
38
2.3.3 Diabetic Neuropathy
Nerve damage through complement activation has been reported in several periph-eral neuropathies (Kaida and Kusunoki 2010 ). Interestingly, a recent study showed that a cell surface defi ciency of the complement regulatory protein CD59 due to a single missense mutation of the CD59 gene was clinically expressed by a form of chronic peripheral neuropathy in children (Nevo et al. 2013 ). The presence of acti-vated complement proteins and MAC neoantigen in sural nerve biopsies from indi-viduals with diabetes was fi rst reported by Hays and collaborators (Rosoklija et al. 2000 ); in the same nerve biopsies analyzed in that study, we observed co- localization of MAC and GCD59 (Qin et al. 2004 ).
2.3.4 Diabetic Cardiovascular Diseases
A relatively large body of evidence supports a role of complement in cardiovascular pathology in general and atherosclerosis in particular (Hertle et al. 2014 ; Kostner 2004 ; Wu et al. 2009a , b ; Haskard et al. 2008 ; Niculescu and Rus 1999 ; Lappegard et al. 2014 ; Niculescu et al. 1987 ; Pang et al. 1979 ; Schmiedt et al. 1998 ; Seifert et al. 1989 ; Seifert and Kazatchkine 1988 ). Diabetes and pre-diabetes are major independent risk factor for cardiovascular disease (Tominaga et al. 1999 ; Jarrett et al. 1982 ). A potential role of C3, its activation products and other components of the early complement cascade in the pathogenesis of type 2 diabetes has been sug-gested from associations found in several cross-sectional and longitudinal human studies, as recently reviewed in (Hertle et al. 2014 ). Also, cleavage of C3 generates the anaphylatoxins, C3a and C5a, and further cleavage of C3a by carboxypeptidases leads to production of acylation stimulating protein (ASP/C3a des Arg), a lipogenic hormone that plays a role in atherosclerosis (Onat et al. 2011 ; Hertle et al. 2012 ; Cianfl one et al. 2003 ). However, the potential role of complement activation in dia-betic cardiovascular disease has not been fully explored in human studies. In a sub- analysis of the DIGAMI-2 trial, which prospectively studied more than 1,200 individuals with type 2 diabetes who had a myocardial infarction, high blood levels of soluble MAC at the time of hospitalization strongly and independently predicted the risk of a cardiovascular event in the subsequent 3.5 years (Mellbin et al. 2012 ). In the same sub-analysis, levels of mannan-binding lectin serine peptidase 2 (MASP-2) appeared to be associated with a poor cardiovascular prognosis, although this did not remain signifi cant after multivariable adjustments.
Together, the fi ndings summarized above suggest an important role of the com-plement system in the development and progression of diabetic complications. However, it is important to note that whether the reported associations of elements of the complement system with diabetes represent causative pathogenic precursors or are just a collateral consequence of diabetes and its complications is still unclear and deserves further investigation. Complimentary to the clinical evidence discussed
J.A. Halperin et al.
39
above, in vitro and animal studies seem to suggest that the lectin MBL pathway could be activated by increased levels of MBL induced by hyperglycemia and by a cooperative binding of MBL to fructosyllysine, the product of the non- enzymatic attachment of glucose to α- or ε-amino groups in lysines (Fortpied et al. 2010 ; Ostergaard et al. 2013 ).
2.4 Glycated CD59
2.4.1 Functional Inactivation of CD59 by Glycation
Glycation-inactivation of CD59 is a molecular link between complement regulation and the complications of diabetes. There are different ways in which the delicate balance between complement activation and restriction can be unfavorably per-turbed in diabetes. On the one hand, autoantibodies to glycated and glycol-oxidized proteins can activate the classical pathway (Uesugi et al. 2004 ; Ostergaard et al. 2013 ; Orchard et al. 1999 ; Witztum et al. 1983 ), fructosamines, reportedly activate the lectin pathway by an increased ligand-receptor binding (Fortpied et al. 2010 ) and increase formation of advanced glycation end products (AGEs) on the arterial walls can serve as neo-epitopes for MBL binding (Flyvbjerg 2010 ). The presence of activated products of C3 (C3d and iC3b), C2, and C4 in target organs of diabetic complications (Gerl et al. 2002 ; Rooijakkers et al. 2009 ) is consistent with activa-tion of complement through either or both of these two pathways. On the other hand, functional inhibition by glycation or reduced expression of complement regu-latory proteins such as CD59 (Acosta et al. 2000 ; Zhang et al. 2002 ) could lead to increased complement-mediated damage and MAC-induced biological processes likely to contribute to the pathogenesis of diabetic complications (Rooijakkers et al. 2009 ; Zhang et al. 2002 ; Falk et al. 1983a , b , 1987 ; Weiss et al. 1990 ).
Of all the above-mentioned putative mechanisms that potentially contribute to complement-mediated tissue damage in diabetes, glycation-inactivation of CD59 is the most extensively studied and documented. Hereafter, we will summarize the experimental and clinical evidence for a role of GCD59 in the pathophysiology of diabetes complications and the potential utility of glycated CD59 as a diagnostic biomarker in diabetes.
Protein glycation on α- or ε-amino groups is a well-established mechanism of tissue damage in diabetes. Despite the many amino groups exposed on the surface of either soluble or membrane proteins, very few ε-amino groups are glycated to a signifi cant extent. For example, despite the very close sequential proximity of lysyl residues K61, K65, and K66 in β-hemoglobin, only the ε-amino group of K66 is preferentially glycated in vivo (Shapiro et al. 1980 ). This is because signifi cant ε-amino glycation requires effective acid-base catalysis, which occurs at “glycation motives” such as histidyl/lysyl residues in which the imidazolyl moiety is located at ≈5A° from the ε-amino group of the lysyl residue or at lysyl-lysyl pairs (Shapiro
2 Complement and Complement Regulatory Proteins in Diabetes
40
et al. 1980 ; Iberg and Fluckiger 1986 ; Shilton and Walton 1991 ; Shilton et al. 1993 ; Calvo et al. 1993 ; Flückiger and Strang 1995 ).
Analyzing the reported NMR structure of human CD59 (Fletcher et al. 1994 ), our group predicted that the protein contains within its active site a glycation-motif formed by amino acid residues K41 and H44 and that glycation could inhibit the activity of CD59 (See Fig. 2.3a ). This idea was further supported by the fact that K41 is adjacent to W40, a conserved residue that is essential for CD59 function (Bodian 1997 ; Yu et al. 1997 ). Exposing purifi ed or recombinant human CD59 to glucose, we demonstrated that the anti-MAC activity of CD59 is indeed inhibited by glycation, as documented with anti-glycated CD59 antibodies (See Fig. 2.3 ) (Acosta et al. 2000 ). Site-directed mutagenesis of human CD59 confi rmed experimentally that (1) human CD59 is inhibited by glycation of its K41 residue and (2) H44 is required for glycation-inactivation of human CD59 (Acosta et al. 2000 ) (See Fig. 2.3b ).
a b
100
GlycatedCD59 Ab
Mutant H44 Q
Mutant K41Q
0
50
0 25 50 10075
Days
rWT CD59Glucose (0.5 M)C
D59
act
ivity
(%
)
Fig. 2.3 Glycation of CD59 ( a ): NMR structure of human CD59 (From Fletcher et al 1994); the fi gure highlights the Lys41 ( red ) and His44 ( green ) amino acid residues that conform the glycation motif and the Trp40 ( blue ) residue, a conserved amino-acid critical for CD59 activity. The His44 of human CD59 is not found in CD59 from any other species sequenced to date. ( b ) Glycation inactivates human CD59: the activity of recombinant human CD59 was measured in a hemolytic assay using guinea-pig erythrocytes ( GPE ) exposed to human MAC formed with purifi ed terminal complement components (as reported in Acosta et al. 2000 ). Incubation of CD59 in the presence of glucose progressively inhibits CD59 activity ( red line ). Inhibition of activity is associated with glycation of CD59, because ( a ) parallel incubation of the same preparation with the non-glycating sugar sorbitol did not affect CD59 activity (see Fig. 2 in Acosta et al. 2000 ). An antibody specifi c for the glycated form of CD59 showed the presence of glycated CD59 in the protein inactivated after incubation with glucose ( inset ). Site directed mutagenesis of either K41 or H44 generated two CD59 mutants (mutant K41Q and mutant H44Q) that were active against human MAC but not inhibited by exposure to glucose (Figure reproduced from Acosta et al. ( 2000 ))
J.A. Halperin et al.
41
2.4.2 Animal Studies
The signifi cance of the glycation motif in human CD59 is highlighted by the fact that the residue H44 is absent in CD59 from any animal species sequenced to date (See Table 1 in Acosta et al. 2000 ). This is remarkable because humans are particu-larly prone to develop diabetic vascular complications, and it is well known that no single animal model of the disease can recapitulate the extensive vascular disease that commonly complicates human diabetes (Vischer 1999 ). This observation has led to the intriguing hypothesis that the presence of a glycation motif K41-H44 in human CD59 (not present in CD59 from other species) could contribute to the unique sensitivity of humans to develop extensive vascular disease in response to hyperglycemia. To test whether the functional inactivation of CD59 confers higher risk for diabetes-induced atherosclerosis, our group has generated molecular engi-neered mice that are completely defi cient in CD59 (mouse CD59 knockout (mCD59KO) mice)), crossed them into the ApoE −/− background (one of the models recommended by the Diabetes Complications Consortium for studying diabetes- induced atherosclerosis (Hsueh et al. 2007 )) and rendered them diabetic by injection of low doses of streptozotocin (an experimental model of type 1 diabetes also rec-ommended by the Diabetes Complications Consortium (Brosius et al. 2009 )). The ablation of CD59 (mCD59 knock out mice in an ApoE−/− background) promoted accelerated atherosclerosis with occlusive coronary disease, vulnerable plaque, and premature death (Wu et al. 2009a ; An et al. 2009 ). In contrast, either transgenic overexpression of human CD59 in the endothelium or the administration of a neu-tralizing anti-C5 monoclonal antibody in these mice signifi cantly attenuated the development of atherosclerosis (Morgan 1995 ). These observations in engineered mice are consistent with a body of experimental evidence supporting a role of com-plement in the development of atherosclerosis (Haskard et al. 2008 ; Niculescu and Rus 1999 ; Pang et al. 1979 ; Schmiedt et al. 1998 ; Buono et al. 2002 ; An et al. 2009 ). Diabetic mice : Initial evidence demonstrated that diabetic mCd59KO/Apoe −/− mice developed a severe and accelerated form of atherosclerosis characterized by signifi -cantly larger atherosclerotic lesion areas in the aorta (>2-fold increased lesion area in en-face aorta preparations) and the aortic roots, as compared to their non-diabetic littermates. This accelerated atherosclerosis observed in diabetic CD59 defi cient mice was associated with signifi cantly increased MAC deposition (detected with anti-mouse C9 antibodies) and occurred at levels of hyperglycemia, lipid profi les and body weights similar to those in diabetic CD59 expressing mice (Liu et al. 2014 ). Interestingly, diabetic mCD59KO mice had signifi cantly increased MAC deposition and higher levels of serum iC3b, a byproduct of C3 cleavage during acti-vation of complement, than non-diabetic mCD59KO mice. These two observations are consistent with the emerging evidence for activation of the complement system in diabetes (Flyvbjerg 2010 ; Uesugi et al. 2004 ; Fortpied et al. 2010 ; Orchard et al. 1999 ), as discussed above.
2 Complement and Complement Regulatory Proteins in Diabetes
42
2.4.3 Functional Studies of Diabetics
The identifi cation of the cellular and molecular mechanisms by which hyperglyce-mia causes tissue damage in humans has been hampered by (1) the lack of animal models that recapitulate the combination and intensity of complications seen in human diabetes (Vischer 1999 ), (2) the chronic nature of the complications that can take decades to become clinically evident, and (3) ethical and practical consider-ations that preclude conduction of the functional studies required to assess mecha-nistic hypotheses in humans. In spite of these limitations, two lines of human experimental evidence seem to support the potential role of glycation-inactivation of CD59 in the pathogenesis of diabetes complications.
Functional inactivation of CD59 was found in individuals with diabetes. If glyca-tion inactivates CD59 in vivo , one would predict an increased sensitivity to MAC- mediated phenomena in cells/tissues carrying glycation-inactivated CD59. This prediction was confi rmed in an ex-vivo study using red blood cells (RBC) from donors with or without diabetes in which we measured both the activity of CD59 in the RBC membranes and the RBC sensitivity to human MAC. The results showed that RBC from individuals with diabetes (selected to carry a HbA1c >8 %) exhib-ited a signifi cantly reduced activity of CD59 and were signifi cantly more sensitive to MAC-mediated lysis than RBC from individuals without diabetes (HbA1c <5.5 %) (See Fig. 2.4 ) (Qin et al. 2004 ). This reduced activity of CD59 in RBC from individuals with diabetes is consistent with inactivation of CD59 by glycation and would explain some reversible hematological abnormalities seen in individuals with poorly controlled diabetes including increased reticulocyte count and shorter RBC lifespan (Peterson et al. 1977 ; Bern et al. 1985 ; Giugliano 1982 ), both suggestive of mild hemolytic anemia. Also, individuals with diabetes experience a high incidence of thrombotic episodes, their platelets are hyperactive and their plasma contains high levels of platelet-derived microparticles (PMPs) (Carr 2001 ). Thrombosis is also a major cause of morbidity and mortality in individuals with PNH (Halperin and Nicholson-Weller 1989 ; Gardner et al. 1958 ; Hill et al. 2013 ) whose CD59-defi cient platelets are exquisitely sensitive to MAC-induced activation and release of PMP (Wiedmer et al. 1993 ); platelets from CD59 knockout mice are also acti-vated spontaneously and release PMPs (Qin et al. 2003 , 2009a , b ). These combined clinical and experimental observations indicate that inactivation of CD59 by glyca-tion could be one factor contributing to the high incidence of thrombosis in indi-viduals with diabetes.
Diabetic patients exhibited glycated CD59 in organs with complications. An antibody that specifi cally recognizes the glycated from of CD59 (GCD59) was able to identify GCD59 colocalized with MAC in target organs of diabetic complica-tions. The presence of GCD59 was confi rmed in renal and sural nerve biopsies from individuals with diabetes. No detectable amounts of GCD59 were found in biopsies from individuals without diabetes (See Fig. 2.4b ) (Qin et al. 2004 ). As expected from the functional inactivation of CD59 by glycation, serial sections of the same
J.A. Halperin et al.
43
renal and nerve biopsies showed co-localization of GCD59 with MAC deposits (Qin et al. 2004 ). Increased levels GCD59 and MAC have also been observed in other tissues extracted from individuals with diabetes, namely skin biopsies and saphenous veins removed for grafting of occluded arteries (unpublished observation (Halperin)).
0
50
100
(n = 10)
(n = 18)
CD59 activity (human RBC)
ND D
% C
D59
Act
ivity
Lysi
s (%
)
HbA1c (%) >8 <5.5
RBC lysis (human MAC)
0
10
20
30
40
(n = 10)
(n = 18)
HbA1c (%)
>8 <5.5
ND D
a
Nodiabetes
Diabetes
GCD59 MACKidneys b
Fig. 2.4 MAC-mediated lysis ( a ) Reduced activity of CD59 and increased sensitivity to MAC- mediated lysis in RBC from individuals with diabetes. Red blood cells ( RBCs ) from individuals with diabetes (selected by HbA1c >8 %) have an 8–10 times lower CD59 activity than RBCs from individuals without diabetes (left panel), and are much more sensitive to cell lysis induced by addi-tion of purifi ed terminal complement components to form the MAC ( right panel ). ( b ) Co-localization of MAC and GCD59 in renal biopsies from individuals with diabetes . GCD59 co-localized with MAC in diabetic human kidneys. Serial sections of the same glomerulus in a kidney from an indi-vidual with and another without diabetes stained with anti-MAC or anti–GCD59 antibodies. Arrows indicate positive staining areas for glycated CD59 and MAC (For details on results depicted in this fi gure see Qin et al. ( 2004 ))
2 Complement and Complement Regulatory Proteins in Diabetes
44
2.5 Complement-Targeted Therapeutics
Progressively recognized as a major mediator of a variety of clinical disorders that include both acute and chronic diseases and affect a wide range of organs, modula-tion of the complement system with either small molecules or biologics is now considered a promising strategy in drug discovery (Ricklin and Lambris 2007 , 2013 ). Fueled by (a) a wealth of high resolution protein structures (Forneris et al. 2010 ; Janssen et al. 2005 , 2006 ; Torreira et al. 2009 ; Wiesmann et al. 2006 ; Wu et al. 2009b ), functional assays (Rooijakkers et al. 2009 ; Garlatti et al. 2010 ; Laursen et al. 2010 ; Shaw et al. 2010 ; Gingras et al. 2011 ; Kajander et al. 2011 ; Morgan et al. 2011 , van den van den Elsen and Isenman 2011 ; Hadders and Gros 2012 ) and genome-wide association studies (Degn et al. 2011 ), which combined offer new insight into the molecular details and ligand interaction sites of comple-ment components; (b) the approval and therapeutic success of the anti-C5 antibody, eculizumab, (Alexion, Cheshire, CT, USA) for the treatment of PNH (Hillmen et al. 2006 , 2007 ; Rother et al. 2007 ) and more recently of atypical hemolytic–uremic syndrome (Legendre et al. 2013 ); and (3) the evidence that even small changes in the activity of individual complement proteins may add up to signifi cantly affect disease progression, the fi eld of complement-targeted drug discovery has seen a surge in approaches to therapeutically intervene at various levels of the complement cascades. Potential therapeutic approaches for interventions include inhibition of C3 activation, inhibition of C3 convertase assembly, acceleration of C3 convertase decay, promotion of C3B proteolysis, inhibition of C3 and/or C5 convertase activi-ties, inhibition of membrane attack complex assembly by CD59, and reestablish-ment of normal alternative complement pathway control with protective Factor H protein. Compounds directed against some of these targets that are currently in early clinical trials for other indications include: compstatin/POT-4, a peptide inhibitor that inhibits the central step of the complement cascade by preventing cleavage of C3and ARC1905, a pegylated, aptamer-based C5 inhibitor that inhibits the cleavage of C5 into C5a and C5b. Other complement pathway-modulating com-pounds currently being considered for and/or are under preclinical development include a humanized anti-factor D antibody (TNX-234), a Fab fragment of an anti- complement factor B (CFB) antibody (TA106), a complement receptor 2 (CR2)-factor H hybrid proteins, a small molecule C5aR peptidomimetic (JPE-1375), and anti-properdin antibody, an inhibitor of C1 esterase inhibitor (C1-INH) and conse-quently the classical pathway, and a soluble form of complement receptor 1 (sCR1).
Based on the extensive clinical and experimental evidence supporting a role of complement in the pathogenesis of diabetes complications, it is conceivable that in the future, complement inhibition could offer an early point of intervention that could potentially retard, prevent, or even reverse progression of those complica-tions. However, given the pleomorphic biological activities of the complement sys-tem, risk-benefi t analyses of therapeutic application targeting complement to prevent or treat complications of diabetes will have to evaluate cautiously the poten-tial systemic effects of chronic complement pathway modulation.
J.A. Halperin et al.
45
2.6 Glycated CD59 as a Biomarker for Human Diabetes
2.6.1 Available Biomarkers to Assess Glucose Physiology
The diagnosis of diabetes, appropriate medical management of diabetes, and risk stratifi cation of diabetic complications all require reliable biomarkers that refl ect integrated glucose handling. The most intuitive potential biomarker is circulating glucose itself; however, circulating glucose concentrations vary signifi cantly based on the temporal relation to meals, physical activity, and stress. Therefore, standard practices include the measurement of a morning blood glucose while fasting, as well as more provocative physiologic maneuvers including measure-ment of glucose following a glucose challenge (such as the 75 g oral glucose tol-erance test). A fasting morning blood glucose measurement provides cross-sectional circulating glucose concentrations independent of a prandial chal-lenge, while the oral glucose tolerance test provides insight into the cross-sec-tional ability to handle a prandial glucose load. While both of these biomarkers of glucose handling are valuable, neither assesses long-term or integrated glucose handling over time.
The hemoglobin A1c (HbA1c) assay overcomes this limitation as it refl ects the approximate average circulating glucose levels over a three month period (Nathan et al. 2008 ). In this regard, HbA1c has become one of the most utilized tests to sur-vey for chronic hyperglycemia, in particular to monitor patients with diabetes on medical therapy, and is increasingly used as a crude screen to diagnose diabetes (American Diabetes Association 2015 ). Similarly, other less utilized biomarkers of chronic glucose handling include fructosamine, glycated albumin, and 1,5- anhydroglucitol. While these assays refl ect long-term circulating glucose con-centrations, none of them are involved in the pathogenesis of diabetes complications related to hyperglycemia; rather these biomarkers represent pathologically unre-lated bystanders that are collaterally glycated.
In this context, glycated CD59 (GCD59) represents a novel biomarker for human glucose handling. The measure of GCD59 refl ects glucose levels that are present in extracellular fl uids and result in glycation and thereby inactivation of CD59. In addition, since glycation-inactivation of CD59 results in a dysinhibition of MAC formation and complement mediated tissue damage, GCD59 levels may also refl ect a measure of hyperglycemia-induced complement-mediated tissue injury (or the risk for developing diabetic complications).
2.6.2 GCD59 Levels in Diabetes
The assessment of GCD59 as a practical marker of human blood glucose and poten-tial predictor of risk for diabetic complications was not possible until the recent development of a highly sensitive and specifi c ELISA to measure serum and plasma
2 Complement and Complement Regulatory Proteins in Diabetes
46
human GCD59 (Ghosh et al. 2013 ). Though CD59 is a cell membrane-bound pro-tein, a soluble form of CD59 that is shed from cell membranes by phospholipases is present in human blood, urine, saliva and other bodily fl uids (Hakulinen and Meri 1995 ; Lehto et al. 1995 ; Meri et al. 1996 ). Ghosh and colleagues reported the opti-mization of a novel sensitive ELISA for measuring circulating blood GCD59 in individuals with and without diabetes: GCD59 was signifi cantly higher in nearly all participants with diabetes when compared to those without diabetes and demon-strated a high sensitivity for the diagnosis of diabetes (Receiver operating character-istics area under the curve = 0.88). The development and availability of this assay enabled a number of subsequent human studies that highlighted the value of GCD59 measurements to diagnose and monitor human diabetes.
2.6.3 GCD59 as a Predictor of Human Glucose Handling
In a series of human studies, Ghosh et al. evaluated how well GCD59 correlated with HbA1c and measures of glucose handling following provocative interventions (Ghosh et al. 2014 ). In their fi rst study, they showed that in a cohort of 400 individu-als with (196) or without (204) type 2 diabetes, the mean circulating GCD59 was nearly fourfold higher among those with than among those without diabetes (See Fig. 2.5 ). Further, a strong and positive association between GCD59 and HbA1c was seen not only among individuals with diabetes, where mean HbA1c levels were 8.4 % (r = 0.58, P < 0.0001), but also among individuals without diabetes where mean HbA1c levels were only 5.6 % (r = 0.45, P < 0.0001). Overall, the authors observed that across the entire spectrum of the population, a positive linear relation-ship between GCD59 and HbA1c existed (r = 0.67, P < 0.0001) and was independent of age, gender, and diabetes diagnosis.
0HbA1c ≤ 6.5 %
n = 204
n = 196
p < 0.0000001
HbA1c ≥ 6.5 %
0.2
0.4GC
D59
(S
PU
)
0.6
0.8
1
1.2
Fig. 2.5 Blood levels of GCD59 classify individuals with and without diabetes. Cross sectional evaluation of GCD59 levels in individuals without ( HbA1c <6.5 %) vs individuals with ( HbA1c >6.5 %) diabetes. Bars represent means ± SEM. Student’s t test p value = 3.7 × 10 −23
J.A. Halperin et al.
47
In a separate study, Ghosh et al. assessed how single fasting plasma GCD59 measurements predicted the glucose response to a 2 h 75 g oral glucose tolerance test (OGTT) (Ghosh et al. 2014 ). They focused on a medication-naive population of participants who carried no prior diagnosis of diabetes (n = 109) but were all deemed to have a high-risk for developing diabetes based on family history of diabetes, personal history of obesity or gestational diabetes or at-risk ethnicity (Hispanic, African-American, or Native-American). They observed a clearly positive and lin-ear relationship between GCD59 and the 2 h glucose response to a 75 g OGTT; more importantly, this relationship was independent of other predictors of glucose handling such as age, gender, race, body-mass index, fasting plasma glucose, blood pressure, HbA1c, and lipoprotein concentrations (β = 19.8, adjusted P = 0.03).
In the context of the aforementioned clinical studies suggesting that higher circulating human GCD59 concentrations strongly predict higher HbA1c and 2 h glucose responses to an OGTT, Ghosh and colleagues hypothesized that interven-tions to lower circulating glucose would result in parallel reductions in GCD59. In an intervention study whereby poorly controlled patients with diabetes were enrolled in a highly monitored program to initiate and titrate subcutaneous insulin therapy to normalize hyperglycemia, the investigators measured serial GCD59 levels. In total, 21 patients with diabetes (8 with type 2 and 13 with type 1) with a mean HbA1c approximating 10 % were initiated on insulin therapy and followed for 8 weeks. Patients were instructed to measure their blood glucose by fi nger stick four times a day for 7 days a week (28 readings per week), and were visited by a trained nurse every week to titrate insulin doses based on the weekly glucose log aiming for pre- prandial blood glucose measurements of 80–120 mg/dL. Blood samples were collected while fasting every week and GCD59 and HbA1c were measured from these weekly blood samples. Within the fi rst two weeks of aggres-sive insulin therapy, the average weekly glucose (average of 28 weekly home measurements) declined as expected from 176.4 mg/dL to 149.4 mg/dL and remained at approximately 150 mg/dL for the remaining 6 weeks of the study. Strikingly, along with the rapid reduction in average weekly glucose, a parallel reduction in GCD59 was observed from 0.94 SPU to 0.73 SPU that also remained stable similarly to glucose for the remaining 6 weeks of the study. Not unexpect-edly, the HbA1c measurements did not signifi cantly change during these fi rst two weeks of insulin therapy, however, did decline to approximately 8 % by the end of the 8 week study period (Ghosh et al. 2014 ).
Together, these initial human studies provide tantalizing preliminary evidence that supports a number of features of plasma GCD59 levels: (1) single plasma GCD59 levels accurately and consistently refl ect mean circulating plasma glucose levels, (2) single plasma GCD59 levels may be a sensitive biomarker for the diag-nosis of diabetes, and (3) plasma GCD59 levels may represent a potential bio-marker for the chronic management of diabetes patients on oral or injectable therapies.
2 Complement and Complement Regulatory Proteins in Diabetes
48
Other complement proteins that have been reportedly used to study the pathways of complement activation and could be potentially explored as diabetes biomarkers include: C1q, C3, C4d (classical and lectin pathways), factors D and Bb (alternative pathway), 3a and C5a (general complement activation), soluble MAC complexes (terminal complement activation), and soluble forms of complement regulatory pro-teins DAF, MCP and CR1 (Hillmen and Richards 2000 ).
2.7 Conclusions
Based on the evidence summarized above, it is tempting to propose a pathogenic model of human diabetic complications in which the delicate balance between com-plement activation and restriction is broken by a combination of glycation- inactivation of CD59 and complement activation by autoantibodies, increased activity of the MBL pathway and possibly other still unidentifi ed mechanisms. This results in an increased MAC deposition as observed in target organs of diabetes complications. Increased MAC deposition activates pathways of intracellular sig-naling, including increased production of ROS, activation of PKC and up-regulation of NF-κB, and induces the release of pro-infl ammatory, prothrombotic cytokines, and growth factors. These complement-dependent mechanisms together, perhaps synergistically, with the multiple complement-independent mechanisms summa-rized by M. Brownlee, promote infl ammation, proliferation and thrombosis as char-acteristically seen in the target organs of diabetes complications (Brownlee 2001 ; Giacco and Brownlee 2010 ; Brownlee 2005 ). Figure 2.6 summarizes these complement- dependent and complement-independent mechanisms highlighting the points of convergence common to both mechanisms such as increased ROS, and activation of PKC and transcription factor NF-κB.
Future research in the area should comprise studies to elucidate further (1) the rela-tive roles of these complement related mechanisms in the pathogenesis of the different diabetes complications, (2) whether differences in complement activity or in expres-sion of complement regulatory proteins such as CD59 could explain the increased risk of diabetes complications among ethnic minorities, (Lanting et al. 2005 ) and (3) how complement-induced biological phenomena and other hyperglycemia- induced cellu-lar responses interplay in the pathogenesis of diabetes complications. Since the com-plement-related mechanisms described in this review are driven by hyperglycemia, it’s conceivable that they will operate similarly in type-1 and type-2 diabetes, a notion to be tested in future human studies. Such research will aid in the development of novel strategies for early stratifi cation of individuals with diabetes who are at higher risk of complications, improve prevention of complications and develop new comple-ment-targeted therapeutic approaches for the complications of diabetes that represent a major cause of morbidity and mortality in adults.
J.A. Halperin et al.
49
Fig. 2.6 A model of the cellular and molecular mechanisms. A model of the cellular and molecu-lar mechanisms leading to cell proliferation, infl ammation and thrombosis characteristically seen in the target organs of diabetic complications is shown. High levels of glucose in diabetes glycates and thereby inactivates CD59 and also increases complement activation. Combined, these effects of hyperglycemia on the complement system trigger more MAC deposition on cell membranes. The fi gure illustrates the different cellular signaling pathways induced by high glucose: the com-plement/MAC-dependent mechanisms summarized in this review are shown on the right side of the fi gure, and the complement-independent mechanisms reviewed in (Brownlee 2001 , 2005 ) are shown on the left side of the fi gure. The center of the fi gure highlights the common signaling path-ways reportedly triggered by either high glucose or the MAC. The inset illustrates how the delicate balance between complement activators and inhibitors can be broken by either increased activation or decreased restriction of the complement pathways. The evidence summarized in this review indicates that both mechanisms are operative in the pathogenesis of diabetes complications
2 Complement and Complement Regulatory Proteins in Diabetes
50
References
Acosta JA, Halperin JA et al (1996) The transient pore formed by homologous terminal comple-ment complexes functions as a bidirectional route for the transport of autocrine and paracrine signals across human cell membranes. Mol Med 2:755–765
Acosta J, Halperin JA et al (2000) Molecular basis for a link between complement and the vascular complications of diabetes. Proc Natl Acad Sci U S A 97:5450–5455
Adler S et al (1986) Complement membrane attack complex stimulates production of reactive oxygen metabolites by cultured rat mesangial cells. J Clin Invest 77:762–767
Adler SG et al (2000) Glomerular type IV collagen in patients with diabetic nephropathy with and without additional glomerular disease. Kidney Int 57:2084–2092
American Diabetes Association (2015) 2. Classifi cation and diagnosis of diabetes. Diabetes Care 38:S8–S16
An G et al (2009) CD59 but not DAF defi ciency accelerates atherosclerosis in female ApoE knock-out mice. Mol Immunol 46:1702–1709
Andoh A et al (1998) Detection of complement C3 and factor B gene expression in normal colorec-tal mucosa, adenomas and carcinomas. Clin Exp Immunol 111:477–483
Bellosillo B et al (2001) Complement-mediated cell death induced by rituximab in B-cell lympho-proliferative disorders is mediated in vitro by a caspase-independent mechanism involving the generation of reactive oxygen species. Blood 98:2771–2777
Benzaquen LR, Nicholson-Weller A, Halperin JA (1994) Terminal complement proteins C5b-9 release basic fi broblast growth factor and platelet-derived growth factor from endothelial cells. J Exp Med 179:985–992
Berger HJ, Halperin JA et al (1993) Activated complement directly modifi es the performance of isolated heart muscle cells from guinea pig and rat. Am J Physiol 265:H267–H272
Bern MM, Busick EJ et al (1985) Disorders of the blood and diabetes. In: Marble A (ed) Joslin’s diabetes mellitus. Lea & Febiger, Philadelphia, p 748
Beynon HL, Davies KA, Walport MJ et al (1994) Erythrocyte complement receptor type 1 and interactions between immune complexes, neutrophils, and endothelium. J Immunol 153:3160–3167
Bodian DL et al (1997) Mutational analysis of the active site and antibodies epitopes of the complement inhibitory glycoprotein CD59. J Exp Med 185:507–516
Bouwman LH et al (2005) Elevated levels of mannose-binding lectin at clinical manifestation of type 1 diabetes in juveniles. Diabetes 54:3002–3006
Brooimans RA, van Wieringen PA, Daha MR et al (1992) Relative roles of decay-accelerating factor, membrane cofactor protein, and CD59 in the protection of human endothelial cells against complement-mediated lysis. Eur J Immunol 22:3135–3140
Brosius FC et al; Animal Models of Diabetic Complications Consortium (2009) Mouse models of diabetic nephropathy. J Am Soc Nephrol 20:2503–2512
Brownlee M (2001) Biochemistry and molecular cell biology of diabetic complications. Nature 414:813–820
Brownlee M (2005) The pathobiology of diabetic complications: a unifying mechanism. Diabetes 54:1615–1625
Buono C et al (2002) Infl uence of C3 defi ciency on atherosclerosis. Circulation 105:3025–3031 Calvo C et al (1993) The preferential site of non-enzymatic glycation of human apolipoprotein A-I
in vivo. Clin Chim Acta 217:193–198 Carney DF, Lang TJ, Shin ML (1990) Multiple signal messengers generated by terminal comple-
ment complexes and their role in terminal complement complex elimination. J Immunol 145:623–629
Carr ME (2001) Diabetes mellitus: a hypercoagulable state. J Diabetes Complications 15:44–54 Carroll MC (1998) The role of complement and complement receptors in induction and regulation
of immunity. Annu Rev Immunol 16:545–568 Cataland SR, Holers VM, Geyer S et al (2014) Biomarkers of terminal complement activation
confi rm the diagnosis of aHUS and differentiate aHUS from TTP. Blood 123:3733–3738
J.A. Halperin et al.
51
Cervoni F et al (1992) Identifi cation and characterization of membrane cofactor protein of human spermatozoa. J Immunol 148:1431–1437
Christiansen VJ, Sims PJ, Hamilton KK (1997) Complement C5b-9 increases plasminogen bind-ing and activation on human endothelial cells. Arterioscler Thromb Vasc Biol 17:164–171
Cianfl one K, Xia Z, Chen LY (2003) Critical review of acylation-stimulating protein physiology in humans and rodents. Biochim Biophys Acta 1609:127–143
Cybulsky AV et al (1990) Cytosolic calcium and protein kinase C reduce complement-mediated glomerular epithelial injury. Kidney Int 38:803–811
Davies A, Lachmann PJ, Waldmann H et al (1989) CD59, an LY-6-like protein expressed in human lymphoid cells, regulates the action of the complement membrane attack complex on homolo-gous cells. J Exp Med 170:637–654
Degn SE, Jensenius JC, Thiel S (2011) Disease-causing mutations in genes of the complement system. Am J Hum Genet 88:689–705
Diabetes Control and Complications Trial (DCCT) Group (1995a) Effect of intensive diabetes management on macrovascular events and risk factors in the Diabetes Control and Complications Trial. Am J Cardiol 75:894–903
Diabetes Control and Complications Trial Group (1995b) Effect of intensive therapy on the develop-ment and progression of diabetic nephropathy in the Diabetes Control and Complications Trial. The Diabetes Control and Complications (DCCT) Research Group. Kidney Int 47:1703–1720
Engstrom G et al (2005) Complement C3 is a risk factor for the development of diabetes: a population- based cohort study. Diabetes 54:570–575
Falk RJ, Dalmasso AP et al (1983a) Neoantigen of the polymerized ninth component of comple-ment. Characterization of a monoclonal antibody and immunohistochemical localization in renal disease. J Clin Invest 72:560–573
Falk RJ et al (1983b) Polyantigenic expansion of basement membrane constituents in diabetic nephropathy. Diabetes 32(Suppl 2):34–39
Falk RJ et al (1987) Ultrastructural localization of the membrane attack complex of complement in human renal tissues. Am J Kidney Dis 9:121–128
Fletcher CM et al (1994) Structure of a soluble, glycosylated form of the human complement regu-latory protein CD59. Structure 2:185–199
Flückiger R, Strang CJ (1995) Structural requirements for protein glycation. Protein Sci 4:186A Flyvbjerg A (2010) Diabetic angiopathy, the complement system and the tumor necrosis factor
superfamily. Nat Rev Endocrinol 6:94–101 Forneris F, Ricklin D, Gros P et al (2010) Structures of C3b in complex with factors B and D give
insight into complement convertase formation. Science 330:1816–1820 Fortpied J, Vertommen D, Van Schaftingen E (2010) Binding of mannose-binding lectin to fruc-
tosamines: a potential link between hyperglycaemia and complement activation in diabetes. Diabetes Metab Res Rev 26:254–260
Gardner FH et al (1958) Some pathophysiologic aspects of paroxysmal nocturnal hemoglobinuria. J Clin Invest 37:895
Garlatti V et al (2010) Cutting edge: C1q binds deoxyribose and heparan sulfate through neighbor-ing sites of its recognition domain. J Immunol 185:808–812
Gehrs KM et al (2010) Complement, age-related macular degeneration and a vision of the future. Arch Ophthalmol 128:349–358
Gerl VB, Bohl J, Pitz S, Stoffelns B, Pfeiffer N, Bhakdi S (2002) Extensive deposits of comple-ment C3d and C5b-9 in the choriocapillaris of eyes of patients with diabetic retinopathy. Invest Ophthalmol Vis Sci 43:1104–1108
Ghosh P, Chorev M, Halperin JA et al (2013) A specifi c and sensitive assay for blood levels of glycated CD59: a novel biomarker for diabetes. Am J Hematol 88:670–676
Ghosh P, Chorev M, Halperin JA et al (2014) Glycation of the complement regulatory protein CD59 is a novel biomarker for glucose handling in humans. J Clin Endocrinol Metab 99(6):E999–E1006
Giacco F, Brownlee M (2010) Oxidative stress and diabetic complications. Circ Res 107:1058–1070 Gingras AR et al (2011) Structural basis of mannan-binding lectin recognition by its associated
serine protease MASP-1: implications for complement activation. Structure 19:1635–1643
2 Complement and Complement Regulatory Proteins in Diabetes
52
Giugliano D (1982) Glycosylated haemoglobin and reticulocyte count in diabetes. Diabetologia 22:223
Hadders MA, Gros P (2012) Assembly and regulation of the membrane attack complex based on structures of C5b6 and sC5b9. Cell rep 1:200–207
Hakulinen J, Meri S (1995) Shedding and enrichment of the glycolipid-anchored complement lysis inhibitor protectin (CD59) into milk fat globules. Immunology 85:495–501
Halperin JA, Nicholson-Weller A (1989) Paroxysmal nocturnal hemoglobinuria. A complement- mediated disease. Complement Infl amm 6:65–72
Halperin JA et al (1988) Complement induces a transient increase in membrane permeability in unlysed erythrocytes. J Clin Invest 82:594–600
Halperin JA, Brugnara C, Nicholson-Weller A (1989) Ca2 + −activated K+ effl ux limits complement- mediated lysis of human erythrocytes. J Clin Invest 83:1466–1471
Halperin JA et al (1993) Transient changes in erythrocyte membrane permeability are induced by sublytic amounts of the complement membrane attack complex (C5b-9). Blood 81:200–205
Hamilton KK et al (1990) Complement proteins C5b-9 induce vesiculation of the endothelial plasma membrane and expose catalytic surface for assembly of the prothrombinase enzyme complex. J Biol Chem 265:3809–3814
Hansen TK (2005) Mannose-binding lectin (MBL) and vascular complications in diabetes. Horm Metab Res 37(Suppl 1):95–98
Hansen TK et al (2003) Elevated levels of mannan-binding lectin in patients with type 1 diabetes. J Clin Endocrinol Metab 88:4857–4861
Hansen TK et al (2004) Association between mannose-binding lectin and vascular complications in type 1 diabetes. Diabetes 53:1570–1576
Hansen TK et al (2010) Association between mannose-binding lectin, high-sensitivity C-reactive protein and the progression of diabetic nephropathy in type 1 diabetes. Diabetologia 53:1517–1524
Haskard DO, Boyle JJ, Mason JC (2008) The role of complement in atherosclerosis. Curr Opin Lipidol 19:478–482
Hattori R et al (1989) Complement proteins C5b-9 induce secretion of high molecular weight multimers of endothelial von Willebrand factor and translocation of granule membrane protein GMP-140 to the cell surface. J Biol Chem 264:9053–9060
Hertle E, van Greevenbroek MM, Stehouwer CD (2012) Complement C3: an emerging risk factor in cardiometabolic disease. Diabetologia 55:881–884
Hertle E, Stehouwer CD, van Greevenbroek MM (2014) The complement system in human cardio-metabolic disease. Mol Immunol 61:135–148
Hill A, Kelly RJ, Hillmen P (2013) Thrombosis in paroxysmal nocturnal hemoglobinuria. Blood 121:4985–4996; quiz 5105
Hillmen P, Richards SJ (2000) Implications of recent insights into the pathophysiology of parox-ysmal nocturnal haemoglobinuria. Br J Haematol 108:470–479
Hillmen P et al (2006) The complement inhibitor eculizumab in paroxysmal nocturnal hemoglo-binuria. N Engl J Med 355:1233–1243
Hillmen P et al (2007) Effect of the complement inhibitor eculizumab on thromboembolism in patients with paroxysmal nocturnal hemoglobinuria. Blood 110:4123–4128
Hovind P, Hansen TK et al (2005) Mannose-binding lectin as a predictor of microalbuminuria in type 1 diabetes: an inception cohort study. Diabetes 54:1523–1527
Hsueh W et al (2007) Recipes for creating animal models of diabetic cardiovascular disease. Circ Res 100:1415–1427
Huang Y, Qiao F, Abagyan R et al (2006) Defi ning the CD59-C9 binding interaction. J Biol Chem 281:27398–27404
Husler T, Sodetz JM, Sims PJ et al (1994) Chimeras of human complement C9 reveal the site recognized by complement regulatory protein CD59. J Biol Chem 270:3483–3486
Iberg N, Fluckiger R (1986) Nonenzymatic glycosylation of albumin in vivo. Identifi cation of multiple glycosylated sites. J Biol Chem 261:13542–13545
J.A. Halperin et al.
53
Janssen BJ, Gros P et al (2005) Structures of complement component C3 provide insights into the function and evolution of immunity. Nature 437:505–511
Janssen BJ, Gros P et al (2006) Structure of C3b reveals conformational changes that underlie complement activity. Nature 444:213–216
Jarrett RJ, McCartney P, Keen H (1982) The Bedford survey: ten year mortality rates in newly diagnosed diabetics, borderline diabetics and normoglycaemic controls and risk indices for coronary heart disease in borderline diabetics. Diabetologia 22:79–84
Kaida K, Kusunoki S (2010) Antibodies to gangliosides and ganglioside complexes in Guillain- Barre syndrome and Fisher syndrome: mini-review. J Neuroimmunol 223:5–12
Kajander T et al (2011) Dual interaction of factor H with C3d and glycosaminoglycans in host- nonhost discrimination by complement. Proc Natl Acad Sci U S A 108:2897–2902
Kaunisto MA et al (2009) Elevated MBL concentrations are not an indication of association between the MBL2 gene and type 1 diabetes or diabetic nephropathy. Diabetes 58:1710–1714
Kilgore KS et al (1997) Sublytic concentrations of the membrane attack complex of complement induce endothelial interleukin-8 and monocyte chemoattractant protein-1 through nuclear fac-tor- kB activation. Am J Pathol 150:2019–2031
Kimberley FC, Sivasankar B, Morgan PB (2007) Alternative roles for CD59. Mol Immunol 44:73–81
Kooyman DL et al (1995) In vivo transfer of GPI-linked complement restriction factors from erythrocytes to the endothelium. Science 269:89–92
Kostner KM (2004) Activation of the complement system: a crucial link between infl ammation and atherosclerosis? Eur J Clin Invest 34:800–802
Kotajima N et al (2000) Type IV collagen as an early marker for diabetic nephropathy in non- insulin-dependent diabetes mellitus. J Diabetes Complications 14:13–17
Laine RO, Esser AF (1989) Detection of refolding conformers of complement protein C9 during insertion into membranes. Nature 341:63–65
Lanting LC, Joung IM, Mackenbach JP et al (2005) Ethnic differences in mortality, end-stage complications, and quality of care among diabetic patients: a review. Diabetes Care 28:2280–2288
Lappegard KT et al (2014) A vital role for complement in heart disease. Mol Immunol 61:126–134 Laufer J, Katz Y, Passwell JH (2001) Extrahepatic synthesis of complement proteins in infl amma-
tion. Mol Immunol 38:221–229 Laursen NS et al (2010) Structural basis for inhibition of complement C5 by the SSL7 protein from
Staphylococcus aureus. Proc Natl Acad Sci U S A 107:3681–3686 Legendre CM et al (2013) Terminal complement inhibitor eculizumab in atypical hemolytic-
uremic syndrome. N Engl J Med 368:2169–2181 Lehto T, Honkanen E, Teppo AM et al (1995) Urinary excretion of protectin (CD59), complement
SC5b-9 and cytokines in membranous glomerulonephritis. Kidney Int 47:1403–1411 Libby P (2002) Infl ammation in atherosclerosis. Nature 420:868–874 Lin RC et al (1988) A family showing inheritance of the Inab phenotype. Transfusion 28:427–429 Liu F, Halperin J et al (2014) The defi ciency of CD59 accelerates the development of diabetic
atherosclerosis in mice. Mol Immunol 61:223 Lovett D et al (1987) Activation of glomerular mesangial cells by the terminal membrane attack
complex of complement. J Immunol 138:2473–2480 Luzzatto L, Bessler M (1996) The dual pathogenesis of paroxysmal nocturnal hemoglobinuria.
Curr Opin Hematol 3:101–110 Marks RM, Todd RF III, Ward P (1989) Rapid induction of neutrophil-endothelial adhesion by
endothelial complement fi xation. Nature 339:314–317 Matsushita M, Endo Y, Fujita T et al (1998) Complement-related serine proteases in tunicates and
vertebrates. Curr Opin Immunol 10:29–35 Mayer MM (1984) Complement: historical perspective and some current issues. Complement
1:2–26
2 Complement and Complement Regulatory Proteins in Diabetes
54
McDonald JF, Nelsestuen GL (1997) Potent inhibition of terminal complement assembly by clus-terin: characterization of its impact on C9 polymerization. Biochemistry 36:7464–7473
Mellbin LG, Bjerre M, Thiel S et al (2012) Complement activation and prognosis in patients with type 2 diabetes and myocardial infarction: a report from the DIGAMI 2 trial. Diabetes Care 35:911–917
Meri S, Podack E, Lachmann PJ et al (1990) Human protectin (CD59), an 18-20-kD homologous complement restriction factor, does not restrict perforin-mediated lysis. J Exp Med 172:367–370
Meri S, Lehto T, Sutton CW et al (1996) Structural composition and functional characterization of soluble CD59: heterogeneity of the oligosaccharide and glycophosphoinositol (GPI) anchor revealed by laser-desorption mass spectrometric analysis. Biochem J 316(Pt 3):923–935
Miyata T et al (1993) The cloning of PIG-A, a component in the early step of GPI-anchor biosyn-thesis. Science 259:1318–1320
Miyata T et al (1994) Abnormalities of PIG-A transcripts in granulocytes from patients with par-oxysmal nocturnal hemoglobinuria. N Engl J Med 330:249–255
Morgan BP (1995) Complement regulatory molecules: application to therapy and transplantation. Immunol Today 16:257–259
Morgan BP, Campbell AK (1985) The recovery of human polymorphonuclear leucocytes from sublytic complement attack is mediated by changes in intracellular free calcium. Biochem J 231:205–208
Morgan BP, Gasque P (1997) Extrahepatic complement biosynthesis: where, when and why? Clin Exp Immunol 107:1–7
Morgan BP, Harris CL (1999) Complement regulatory proteins. Academic, London Morgan HP et al (2011) Structural basis for engagement by complement factor H of C3b on a self
surface. Nat Struct Mol Biol 18:463–470 Nathan DM et al (2008) Translating the A1C assay into estimated average glucose values. Diabetes
Care 31:1473–1478 Nevo Y et al (2013) CD59 defi ciency is associated with chronic hemolysis and childhood relapsing
immune-mediated polyneuropathy. Blood 121:129–135 Nicholson-Weller A, Halperin JA (1993) Membrane signaling by complement C5b-9, the mem-
brane attack complex. Immunol Res 12:244–257 Nicholson-Weller A et al (1985) Surface membrane expression by human blood leukocytes and
platelets of decay-accelerating factor, a regulatory protein of the complement system. Blood 65:1237–1244
Niculescu F, Rus H (1999) Complement activation and atherosclerosis. Mol Immunol 36:949–955 Niculescu F, Rus H (2001) Mechanisms of signal transduction activated by sublytic assembly of
terminal complement complexes on nucleated cells. Immunol Res 24:191–199 Niculescu F, Rus HG, Vlaicu R (1987) Activation of the human terminal complement pathway in
atherosclerosis. Clin Immunol Immunopathol 45:147–155 Niculescu F, Rus H, Shin ML (1994) Receptor-independent activation of guanine nucleotide-
binding regulatory proteins by terminal complement complexes. J Biol Chem 269:4417–4423 Niculescu F, Shin ML et al (1997) Activation of Ras and mitogen-activated protein kinase pathway
by terminal complement complexes is G protein dependent. J Immunol 158:4405–4412 Ninomiya H, Sims PJ (1992) The human complement regulatory protein CD59 binds to the alpha-
chain of C8 and to the “b”domain of C9. J Biol Chem 267:13675–13680 Onat A, Cianfl one K et al (2011) Complement C3 and cleavage products in cardiometabolic risk.
Clin Chim Acta 412:1171–1179 Orchard TJ et al (1999) Antibodies to oxidized LDL predict coronary artery disease in type 1 dia-
betes: a nested case–control study from the Pittsburgh Epidemiology of Diabetes Complications Study. Diabetes 48:1454–1458
Ostergaard JA et al (2013) Diabetes-induced changes in mannan-binding lectin levels and comple-ment activation in a mouse model of type 1 diabetes. Scand J Immunol 77(3):187–194
Pang AS, Katz A, Minta JO (1979) C3 deposition in cholesterol-induced atherosclerosis in rabbits: a possible etiologic role for complement in atherogenesis. J Immunol 123:1117–1122
J.A. Halperin et al.
55
Parker CJ (1996) Molecular basis of paroxysmal nocturnal hemoglobinuria. Stem Cells 14:396–411
Passwell JH et al (1990) Complement gene expression in hepatic and extrahepatic tissues of NZB and NZB x W (F1) mouse strains. Immunology 71:290–294
Pedersen O, Gaede P (2003) Intensifi ed multifactorial intervention and cardiovascular outcome in type 2 diabetes: the Steno-2 study. Metabolism 52:19–23
Peterson CM et al (1977) Reversible hematologic sequelae of diabetes mellitus. Ann Intern Med 86:425–429
Petranka JG, Kaufman RE, Rosse WF et al (1992) Structure of the CD59-encoding gene: further evidence of a relationship to murine lymphocyte antigen Ly-6 protein. Proc Natl Acad Sci U S A 89:7876–7879
Podack ER (1984) Molecular composition of the tubular structure of the membrane attack complex of complement. J Biol Chem 259:8641–8647
Podack ER, Muller-Eberhard HJ (1978) Binding of desoxycholate, phosphatidylcholine vesicles, lipoprotein and of the S-protein to complexes of terminal complement components. J Immunol 121:1025–1030
Podack ER, Tschopp J (1982) Polymerization of the ninth component of complement (C9): forma-tion of poly(C9) with a tubular ultrastructure resembling the membrane attack complex of complement. Proc Natl Acad Sci U S A 79:574–578
Podack ER, Tschopp J (1984) Membrane attack by complement. Mol Immunol 21:589–603 Podack ER, Horst H, Hoppe W et al (1982) Membrane attach complex of complement (MAC): three-
dimensional analysis of MAC-phospholipid vesicle recombinants. J Immunol 128:2353–2357 Qin X, Halperin JA et al (2003) Defi ciency of the mouse complement regulatory protein mCd59b
results in spontaneous hemolytic anemia with platelet activation and progressive male infertil-ity. Immunity 18:217–227
Qin X, Halperin JA et al (2004) Glycation inactivation of the complement regulatory protein CD59: a possible role in the pathogenesis of the vascular complications of human diabetes. Diabetes 53:2653–2661
Qin X et al (2009a) Balancing role of nitric oxide in complement-mediated activation of platelets from mCd59a and mCd59b double-knockout mice. Am J Hematol 84:221–227
Qin X, Halperin JA et al (2009b) Generation and phenotyping of mCd59a and mCd59b double- knockout mice. Am J Hematol 84:65–70
Ramm LE, Whitlow MB, Mayer MM (1982) Transmembrane channel formation by complement: functional analysis of the number of C5b6, C7, C8, and C9 molecules required for a single channel. Proc Natl Acad Sci U S A 79:4751–4755
Ramm LE, Whitlow MB, Mayer MM (1985) The relationship between channel size and the num-ber of C9 molecules in the C5b-9 complex. J Immunol 134:2594–2599
Reid ME et al (1991) Biochemical studies on red blood cells from a patient with the Inab pheno-type (decay-accelerating factor defi ciency). Blood 78:3291–3297
Ricklin D, Lambris JD (2007) Complement-targeted therapeutics. Nat Biotechnol 25:1265–1275 Ricklin D, Lambris JD (2013) Progress and trends in complement therapeutics. Adv Exp Med Biol
735:1–22 Risitano AM (2012) Paroxysmal nocturnal hemoglobinuria and other complement-mediated
hematological disorders. Immunobiology 217:1080–1087 Rooijakkers SH, Gros P et al (2009) Structural and functional implications of the alternative com-
plement pathway C3 convertase stabilized by a staphylococcal inhibitor. Nat Immunol 10:721–727
Rosoklija GB et al (2000) Local activation of the complement system in endoneurial microvessels of diabetic neuropathy. Acta Neuropathol 99:55–62
Rother RP et al (2007) Discovery and development of the complement inhibitor eculizumab for the treatment of paroxysmal nocturnal hemoglobinuria. Nat Biotechnol 25:1256–1264
Saadi S et al (1995) Complement-mediated regulation of tissue factor activity in endothelium. J Exp Med 182:1807–1814
2 Complement and Complement Regulatory Proteins in Diabetes
56
Saraheimo M et al (2005) Increased levels of mannan-binding lectin in type 1 diabetic patients with incipient and overt nephropathy. Diabetologia 48:198–202
Schmiedt W et al (1998) Complement C6 defi ciency protects against diet-induced atherosclerosis in rabbits. Arterioscler Thromb Vasc Biol 18:1790–1795
Seifert PS, Kazatchkine MD (1988) The complement system in atherosclerosis. Atherosclerosis 73:91–104
Seifert PS et al (1989) Prelesional complement activation in experimental atherosclerosis. Terminal C5b-9 complement deposition coincides with cholesterol accumulation in the aortic intima of hypercholesterolemic rabbits. Lab Invest 60:747–754
Shapiro R et al (1980) Sites of nonenzymatic glycosylation of human hemoglobin A. J Biol Chem 255:3120–3127
Shaw CD et al (2010) Delineation of the complement receptor type 2-C3d complex by site-directed mutagenesis and molecular docking. J Mol Biol 404:697–710
Shilton BH, Walton DJ (1991) Sites of glycation of human and horse liver alcohol dehydrogenase in vivo. J Biol Chem 266:5587–5592
Shilton BH, Campbell RL, Walton DJ (1993) Site specifi city of glycation of horse liver alcohol dehydrogenase in vitro. Eur J Biochem 215:567–572
Stahl RA et al (1993) Increased expression of monocyte chemoattractant protein-1 in anti- thymocyte antibody-induced glomerulonephritis. Kidney Int 44:1036–1047
Sun X et al (1999) Role of decay-accelerating factor in regulating complement activation on the erythrocyte surface as revealed by gene targeting. Proc Natl Acad Sci U S A 96:628–633
Takeda J et al (1993) Defi ciency of the GPI anchor caused by a somatic mutation of the PIG-A gene in paroxysmal nocturnal hemoglobinuria. Cell 73:703–711
Tegla CA, Cudrici C, Patel S, Trippe R 3rd, Rus V, Niculescu F, Rus H (2011) Membrane attack by complement: the assembly and biology of terminal complement complexes. Immunol Res 51:45–60
Telen MJ, Green AM (1989) The Inab phenotype: characterization of the membrane protein and complement regulatory defect. Blood 74:437–441
Tominaga M et al (1999) Impaired glucose tolerance is a risk factor for cardiovascular disease, but not impaired fasting glucose. The Funagata diabetes Study. Diabetes Care 22:920–924
Torbohm I et al (1990) C5b-8 and C5b-9 modulate the collagen release of human glomerular epi-thelial cells. Kidney Int 37:1098–1104
Torreira E et al (2009) 3D structure of the C3bB complex provides insights into the activation and regulation of the complement alternative pathway convertase. Proc Natl Acad Sci U S A 106:882–887
Torzewski J et al (1996) Complement-induced release of monocyte chemotactic protein-1 from human smooth muscle cells. A possible initiating event in the atherosclerotic lesion formation. Arterioscler Thromb Vasc Biol 16:673–677
Tschopp J, Engel A, Podack ER (1984) Molecular weight of poly(C9). 12 to 18 C9 molecules form the transmembrane channel of complement. J Biol Chem 259:1922–1928
Uesugi N et al (2004) Possible mechanism for medial smooth muscle cell injury in diabetic nephropathy: glycoxidation-mediated local complement activation. Am J Kidney Dis 44:224–238
UK Prospective Diabetes Study (UKPDS) Group (1998) Intensive blood-glucose control with sul-phonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). Lancet 352:837–853
van den Elsen JM, Isenman DE (2011) A crystal structure of the complex between human comple-ment receptor 2 and its ligand C3d. Science 332:608–611
Viedt C et al (2000) The terminal complement complex C5b-9 stimulates interleukin-6 production in human smooth muscle cells through activation of transcription factors NF-kappaB and AP-1 [In Process Citation]. Faseb J 14:2370–2372
Vischer UM (1999) Hyperglycemia and the pathogenesis of atherosclerosis: lessons from murine models. Eur J Endocrinol 140:1–3
J.A. Halperin et al.
57
Wagner C et al (1994) Induction of matrix protein synthesis in human glomerular mesangial cells by the terminal complement complex. Exp Nephrol 2:51–56
Walport MJ (2001a) Complement. First of two parts. N Engl J Med 344:1058–1066 Walport MJ (2001b) Complement. Second of two parts. N Engl J Med 344:1140–1144 Weiss JS, Sang DN, Albert DM (1990) Immunofl uorescent characteristics of the diabetic cornea.
Cornea 9:131–138 Wiedmer T et al (1993) Complement-induced vesiculation and exposure of membrane prothrom-
binase sites in platelets of paroxysmal nocturnal hemoglobinuria. Blood 82:1192–1196 Wiesmann C et al (2006) Structure of C3b in complex with CRIg gives insights into regulation of
complement activation. Nature 444:217–220 Witztum JL, Fisher M, Kesaniemi A et al (1983) Nonenzymatic glucosylation of homologous low
density lipoprotein and albumin renders them immunogenic in the guinea pig. Proc Natl Acad Sci U S A 80:2757–2761
Wlazlo N et al (2014) Complement factor 3 is associated with insulin resistance and with incident type 2 diabetes over a 7-year follow-up period: the CODAM Study. Diabetes Care 37:1900–1909
Woroniecka KI et al (2011) Transcriptome analysis of human diabetic kidney disease. Diabetes 60:2354–2369
Wu G, Halperin JA, Qin X et al (2009a) Complement regulator CD59 protects against atheroscle-rosis by restricting the formation of complement membrane attack complex. Circ Res 104:550–558
Wu J, Gros P et al (2009b) Structure of complement fragment C3b-factor H and implications for host protection by complement regulators. Nat Immunol 10:728–733
Yamashina M et al (1990) Inherited complete defi ciency of 20-kilodalton homologous restriction factor (CD59) as a cause of paroxysmal nocturnal hemoglobinuria. N Engl J Med 323:1184–1189
Yu J et al (1997) Mapping the active site of CD59. J Exp Med 185:745–753 Zhang J, Gerhardinger C, Lorenzi M (2002) Early complement activation and decreased levels
of glycosylphosphatidylinositol-anchored complement inhibitors in human and experimental diabetic retinopathy. Diabetes 51:3499–3504
2 Complement and Complement Regulatory Proteins in Diabetes
Part III Role of C Terminal Fragment
of Adiponectin Receptor in Inhibition of Insulin Degradation
61© Springer International Publishing Switzerland 2015 M. Pugia (ed.), Infl ammatory Pathways in Diabetes: Biomarkers and Clinical Correlates, Progress in Infl ammation Research, DOI 10.1007/978-3-319-21927-1_3
Chapter 3 Development of Adiponectin Receptor C-Terminal Fragment Bioassays
Michael Pugia and Rui Ma
Abstract The presence of adiponectin receptor C terminal fragment (AdipoR CTF) in human blood was discovered using immune-affi nity mass spectroscopy techniques (Pugia et al., Clin Proteomics 5:156–162, 2009). Defi ciency in diabetic patients was observed and followed up by enzyme linked immunoabsorbent assays (ELISA) with generation of a set of antibodies for AdipoR1 and R2 CTF. Additional western blot analysis identifi ed the majority of AdipoR CTF in human, rat and rab-bit blood was covalently bound to immunoglobulins (Ig-CTF) as the primary native form. Bound CTF in patient plasma was found at 95–4900 ng/mL and to be <0.2 % of all plasma IgG. Free CTF was found at 0.05–5.0 ng/mL. CTF was shown to be covalently bound to immunoglobulin and only released upon basic conditions and not simply by breaking the disulfi de bonds. Mass spectroscopy peptide enrichment assays utilizing trypsin digests were unable to digest bound CTF and basic treat-ment was required to liberate free form. Antibodies for AdipoR1 and R2 CTF were paired in sandwich ELISA suitable for measurement Ig-CTF bound forms for clini-cal patient testing and free CTF forms for animal model testing. Analytical valida-tion demonstrated <10 % between sample and <10 % within patient reproducibility. Calibration, precision, and stability were acceptable. No cross reactivity or interfer-ence was observed. Simple sample preparation was demonstrated as a benefi t over mass spectroscopic peptide enrichment assays analysis.
M. Pugia , PhD (*) Strategic Innovation , Siemens Healthcare Diagnostics , 3400 Middlebury Street , Elkhart IN 46515 , USA e-mail: [email protected]
R. Ma Siemens Healthcare , 278 Zhouzhu Road , Pu Dong New Area 201318 , People’s Republic of China
62
3.1 Introduction
3.1.1 Adiponectin Response
Adiponectin (Adpn) is secreted by adipose and regulates energy metabolism and infl ammation (Scherer et al. 1995 ; Maeda et al. 1996 ; Spiegelman et al. 1996 ; Nakano et al. 1996 ; Gil-Campos et al. 2004 ) with its insulin-sensitizing (Kadowaki et al. 2006 ; Ouchi and Walsh 2007 ), anti-infl ammatory (Szmitko et al. 2007 ) and cardioprotective (Kubota et al. 2002 ) properties. Reduction of Adpn in obese mice and humans correlates with diabetes, cardiovascular disease and insulin resistance (Kadowaki et al. 2003 ). Adpn is very abundant in the bloodstream with a sequence homology to complement C1q without the Fc binding site. It is found in three major oligomeric isoforms, with the high molecular weight (HMW) form most often cited as clinically relevant (Shapiro and Scherer 1998 ; Menzaghi et al. 2007 ). Surprisingly, X-ray crystallography analysis of adiponectin’s globular head showed similarity to tumor necrosis factor-α (TNF-α) structure without signifi cant amino acid sequence homology (Maeda et al. 2002 ). These similarities suggest the possible regulation mechanism of Adpn in the process of infl ammation and complement.
In a healthy patient, plasma Adpn increases with weight as adipose tissue increases and in response to insulin and fatty acids ( See Fig. 3.1 ). Adpn is released into the plasma and signals metabolism in the muscle and liver. Adpn increases glucose uptake in muscle, decreases gluconeogenesis in liver, and increases fatty- acid oxidation in both ( See Fig. 3.1 ). This action occurs through the activation of acetyl coenzyme A carboxylase (ACC) and 5′-AMP activated protein kinase (AMPK) (Ma et al. 2002 ; Kadowaki et al. 2002 ; Mao et al. 2006 ; Sattar et al. 2006 ; Daimon et al. 2003 ). Adpn is also an insulin sensitizing agent. Fatty tissues become hypertrophic in obese patients due to lipid loading. Hypertrophic adipose tissue secretes less Adpn. Lower plasma Adpn correlates with increased risks of hyper-glycemia, diabetes, insulin resistance and cardiovascular disease (Shapiro and Scherer 1998 ; Hara et al. 2006 ; Phillips et al. 2003 ; Mao et al. 2006 ). Plasma Adpn can be increased with thiazolidinediones (TZD) treatment (Tsao et al. 2002 ). This anti- diabetic activates the peroxisome proliferator-activated receptor (PPAR γ) increasing Apdn secretion and sensitizing insulin. How Adpn sensitizes insulin is unclear.
The clinical sensitivity of Adpn to diagnose and monitor diabetes is poor with many false negatives and positives. A greater understanding of the Adpn pathway is needed to improve clinical accuracy. It is unknown how Adpn impacts insulin sensi-tivity. Adpn alone has no effect on protein kinase B (Akt) phosphorylation (Tsao et al. 2002 ). There is also no explanation for Apdn infl ammatory behaviors. Adpn inhibits tumor necrosis factors -α (TNF-α) induced adhesion molecule expression, inhibits nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) acti-vation in endothelial cells and reduces macrophage production OD TNF-α (Ouchi and Walsh 2007 ; Yamauchi et al. 2003 ). Adpn also activates mitogen-activated pro-tein kinase (MAPK p38) and induces endothelial apoptosis (Bråkenhielm et al. 2004 ).
M. Pugia and R. Ma
63
3.1.2 Role of Adiponectin Receptor
The adiponectin receptor is a membrane bound G protein-coupled receptor with seven transmembrane domains; the N-terminal in the cytosol and the C-terminal in an unexpected extra-cellular position ( See Fig. 3.2 ). Two adiponectin receptors have been identifi ed. AdipoR1 is ubiquitously expressed and abundant in skeletal muscle. AdipoR2 is expressed mostly in the liver (Kadowaki et al. 2003 , Yamauchi Kadowaki et al. 2003 ; Bråkenhielm et al. 2004 ). T-cadherin is a novel receptor found for high-molecular- weight (HMW) adiponectin multimers (Hug et al. 2004 ). The internal N-terminus of AdipoR1 was found to interact with the adaptor protein containing a pleckstrin homology domain, phosphotyrosine binding domain, and leucine zipper motif to initiate the downstream cascade AMPK (Ouchi and Walsh 2007 ). The external C-termini is of unknown function. It is unclear if Adpn interacts with the C-terminus or the loops between the transmembrane domains.
Our goal was to investigate the function of the C -terminal fragment (CTF) as a potential new avenue in this mechanism. We previously reported that the C-terminal
Adiponectin(multimeric)
MyocyteMuscle
AdipocyteFat
PPAR
Therapies(TZD)
Glucoseproduction
AdipoR2
Fatty acids
HepatocyteLiver
AdipoR1
Glycolysisand fatty acidoxidation
Full-lengthAdiponectin
AMPK ACC
Globularadiponectin
Insulinreceptor
Insulinsensitivity
γ
Fig. 3.1 Adiponectin Receptor System. In normal patients, the adiponectin receptor system allows adiponectin to reduce glucose production in the liver (AdipoR2) and causes glycolysis and fatty acid oxidation in muscle (AdipoR1) through signaling cell metabolism via 5′-AMP activated pro-tein kinase (AMPK). Adiponectin also activates cell growth (MAPK p38) and anti-atherogenic responses (PPAR α). AdipoR2 reacts to full length adiponectin in low, medium and high molecular weight forms but not globular adiponectin (gAdpn). AdipoR1 is more responsive to gAdpn. In metabolic syndrome patients, adipocytes become hypertrophic due to lipid accumulation. The result is less adiponectin is produced and a reduced the AMPK response in muscle and liver. In the liver heptocytes, glucose production is not shut down and in the muscle myocytes, less glycolysis and fatty acid oxidation occurs
3 Development of Adiponectin Receptor C-Terminal Fragment Bioassays
64
fragments of adiponectin receptor (named as AdipoR CTF) existed in human plasma, and the free CTF was low or missing in most diabetic patients tested using immuno-affi nity mass spectroscopy (Pugia et al. 2009 ). CTF was signifi cantly higher (P > 0.001) in non-diabetics (n = 10, ave 3.1 ng/ml, sd 0.6) than diabetics (n = 10, ave 0.4 ng/ml, sd 0.1). This chapter demonstrates methods used to translate a proteomic discovery into a clinical assay. We report methods for measurement of Adiponectin receptor (AdipoR) in blood which offer tools to for researches to fur-ther explore this pathway (Pugia 2012 ; Pugia and Ma 2013 ).
3.2 Methods
3.2.1 AdipoR1 and R2 CTF Peptide Preparation
Peptides were made to match the clinical form, to cover the cleavage site, to serve as soluble standards, to produce antibodies and to study the inhibition of proteases. ( See Table 3.1 and supplemental table on CTF peptide sequences). AdipoR1 and R2 peptides were either synthesized by Siemens Healthcare Diagnostics (Walpole MA) or under contract to Celtek Biosciences LLC (Nashville TN). The 17 mers were obtained from Phoenix Pharamaceuticals Inc (Belmont, CA). All peptides were characterized by an HPLC C-18 column and matrix-assisted laser desorption/ion-ization (MALDI) mass spectroscopy and typically >90 % pure.
277267
328
298
Cytoplasm
Extracellular matrix
179
245
344
307
342
C-terminal fragment CTF
376
kfd
wi
fqsh
hfi
q
q
r
ga
vflg
slgl
vfa
aa
vvlv
ip
acg
vcv
lgvt
at
kvf g
at
p
hk
rqt
isl
iy
lr p
qlyy
f sycsp
lpa
fm
ymnp
flg
li
tmlr
g t e
h atw i
lvfg
n
l
g mwffl p
s y di ga
l l i m gs
l
vke
f vgmfflga q
CO2Hll t dd t c
ggr
fe
y
q
g l e
fl n s v g
h
f
l
p
g
r
erf
a ay
lv
t
iymva
m
298
e aitfhm
tp v v g
277
frd
qw avii
gisa
ksf
rt
fvpw
svk
se
hc y
vlsf
w slfht
lfl
y l
i arfi h
l
t
Transmembrane
267f srac p
r phgh pkl d n dwm
kf
s
pv i
r grwr
e
l pvdy d
v kw y f ev e k em m h ha a l pq l
a n p k ap e e qe t p vc p e eq e e ve r
t
vl
ivr k g k e e l l p g l e a l e v t d a e r n s a
m s s h k g s v v a p g n g
pa
Area of interest
y
Fig. 3.2 AdipoR1 inverted G protein-coupled receptor. The adiponectin receptor is an inverted membrane bound G protein-coupled receptor with seven transmembrane domains and the N-terminal in the cytosol and the C-terminal in an unexpected extra-cellular position. The C-Terminal for G protein-coupled receptors is typically on cytoplasm side
M. Pugia and R. Ma
65
The peptides matching the masses in human clinical blood were previous identi-fi ed by immune-affi nity mass spectroscopy techniques as the 32 mers of AdipoR1 CTF 343-375 (Pugia et al. 2009 ). These 32 mer peptides had poor solubility due to Val-Val- Ala-Ala (Gly)-Ala-Phe-Val sequence and therefore were of lower purity and diffi cult to work with. The 25 mer peptides AdipoR1 CTF 351-375 had good solubility and were suitable materials for standards. This material was dissolved at 2 mg/mL in 60 % DMSO-water. All peptides containing the cysteine at position 269 readily dimerized unless dissolved in 60 % acetonitrile-water with 0.1 % TFA. Mixing pep-tides in the absence of trifl uoroacetic acid (TFA) allowed the R1-R2, R1-R1 and R2-R2 25 mer dimers to be synthesized in high purity.
3.2.2 CTF Antibodies
Mouse monoclonal antibody (mAb) clones 444, 515, 510, and 461 were raised against R1 25-mer (soluble section), R1 9-mer (tail), R2 9-mer (tail) and R1 15-mer (cleavage site) peptides using standard methods ( See Table 3.2 ). Biacore testing against the R1 and R2 peptides for 9-mer, 12-mer, 25-mer and 32-mers showed all mAb to be medium affi nity except 515 which was low affi nity. A high affi nity rabbit monoclonal was made to R1 12-mer (active site) conjugated to keyhole limpet hemocyanin (KLH) carriers for immunization (Epitomics Inc. Burlingame, CA). Binding kinetics were measured using surface plasmon resonance (Biacore et al. 3000 instrument).
Table 3.1 Adiponectin receptor C- terminal fragment peptides
Peptide name Number of amino acids (mer) Description of peptide
AdipoR1 CTF 343-375 33 R1 Longer form AdipoR1 CTF 344-375 32 R1 Clinical form AdipoR2 CTF 344-375 32 R2 Clinical form AdipoR1 CTF 343-375 33 Used for polyclonal antibodies AdipoR1 CTF 341-370 30 Covers the cleavage site AdipoR1 CTF 341-355 15 Covers the cleavage site AdipoR1 CTF 351-375 25 R1 Soluble portion AdipoR2 CTF 351-375 25 R2 Soluble portion AdipoR2 CTF 351-375 13 R2 used for standard AdipoR1 CTF 358-375 17 Non inhibitory domain AdipoR2 CTF 358-375 17 Non inhibitory domain AdipoR1 CTF 351-362 12 Inhibitory domain (active site) AdipoR1 CTF 344-353 10 Non inhibitory domain AdipoR1 CTF 367-375 9 Tail (end of C terminal) AdipoR2 CTF 367-375 9 Tail (end of C terminal)
3 Development of Adiponectin Receptor C-Terminal Fragment Bioassays
66
Rabbit polyclonal antibodies (pAb) were raised against CTF R1 33-mer and 25-mer using standard methods ( See Table 3.2 ). Rabbits did not produce antibodies to 33 mers conjugated with bovine serum albumin (BSA) but did with conjugation to KLH. Both BSA and KLH conjugates to 25 mer produced antibodies. Cross reactiv-ity and paring studies were conducted on all antibodies by ELISA ( See Table 3.2 ). The sequence for rat was only one amino acid different and all clones cross-reacted.
3.2.3 Conjugated Antibodies and Standards
Conjugates were made for ELISA, Western blot (WB), immunohistochemistry (IHC) and immunocytochemistry (ICC) analysis. ALP and FITC conjugated antibodies were tested for uses and cross reactivity ( See Table 3.3 ). Immunoglobulin CTF standard was made by conjugating AdipoR1 CTF 375-351 and R2 CTF 375-363 to human IgG. AdipoR1 CTF 375-351 was used as standard for free
Table 3.2 Adiponectin receptor CTF antibodies & cross reactivity and pairing
Antibody Binding site Raised against Clone identifi cation
Mouse mAb 444 R1CTF
Active site AdipoR1 CTF 351-375 ATCC 444-1D12.1H7 PTA 12564
Mouse mAb 515 R1CTF
Tail AdipoR1 CTF 367-375 ATCC 515-4D6.1B10 PTA 12563
Mouse mAb 510 R2CTF
Tail AdipoR2 CTF 367-375 ATCC 510-6B6-1G1 PTA 12565
Mouse mAb 461 R1CTF
Cleavage site AdipoR1 CTF 341-354 ATCC 461-4H11.2A4 PTA 12562
Rabbit mAb R1CTF Active site AdipoR1 CTF 351-362 SAT-56-1 IgG Rabbit pAb R1CTF Soluble portion AdipoR1 CTF 351-375 NA Rabbit pAb R1CTF Active site AdipoR1 CTF 351-362 NA
Antibody Reacted with No cross reaction with Pairing
ATCC 444-1D12.1H7
R1 12-, 25-, 32-mer peptides
R2 peptides or the R1 9-mer peptide
pAb against CTF 25- mer mAb 461 or 515
ATCC 515-4D6.1B10
R1 9-, 25-, 32- mer peptides
R2 peptides or the R1 12-mer peptide
–
ATCC 510-6B6-1G1
R2 9-, 25-, 32- mer peptides
R1 peptides or the R2 12- mer peptide
–
ATCC 461-4H11.2A4
R1 32- mer peptide R2 peptides or the R1 9-, 12- or 25- mer peptides
–
Rabbit monoclonal R1 12-, 25-, 32- mer peptides
R2 peptides or the R1 9- mer peptides
pAb against CTF 25- mer mAb 461 or 515
ATCC deposited antibody cells lines are available from ATCC upon request at [email protected] Legend: Antibodies and their antigenic sites are shown along with cross reactivates and pairing. Note pairing with mouse mAb 515 was dependent on conjugation, N-hydroxysuccinimide conju-gation method did not allow pairing but carbodiimide conjugation did
M. Pugia and R. Ma
67
CTF ELISA. All materials were stored in 0.1 M Tris, 0.25 M NaCl, pH 7.3, 0.5 % BSA and 0.1 % NaN 3 .
3.2.4 Western Blot Methods
Western methods for measuring CTF R1 or R2 and the related protein were done per standard denatured SDS PAGE gels or native gel techniques (See on-line supple-ment for full procedure at www.extras.springer.com/2015/978-3-319-21926-4 ). Low molecular weights peptides were analyzed with 20 % Tris -Tricine SDS PAGE gels. Transfer membranes were incubated with antibodies for CTF R1 or R2 conju-gated to alkaline phosphate and developed with substrate.
3.2.5 ELISA Methods
ELISA methods for measuring the immunoglobulin bound form of CTF R1 or R2 were developed by are coating ELISA plates with anti CTF R1 or R2 antibody, washing the plate with tris buffered saline (TBS) before blocking the plate (See on- line supplement for full procedure at www.extras.springer.com/2015/978-3- 319- 21926-4 ). Samples are collected in sterile BD Vacutainer® tubes containing 10.8 mg K 2 EDTA. Samples were diluted 1:48 in phosphate buffered saline contain-ing 1 % BSA, 0.1 M citric acid, 0.27 g/dL EDTA and pH adjusted to 6.4. A second monoclonal antibody for Anti-Human IgG produced in mouse (binds human Fab) and conjugated to alkaline phosphatase was used for detection with para-nitro
Table 3.3 Antibodies conjugates and uses
Conjugates and standards Binding site Std Pair with Use Assay for
Anti- R1CTF pAb 25 mer -ALP Multiple R1 25 mer
444 ELISA, WB
R1CTF
Anti- R2CTF mAb 510 -FITC Tail – – IHC, ICC R2 CTF Ig- R2CTF
Anti- R1CTF mAb 461 -FITC Cleavage site
– – IHC, ICC R1CTF
Anti- R1CTF mAb 444 -FITC Active site – – IHC, ICC R1 CTF Ig- R1CTF
Anti human IgG mAb to ALP
Fab Ig-R1 CTF
444 ELISA Ig- R1CTF
Ig-R2 CTF
510 Ig- R2CTF
Anti- R1CTF mAb 444 -ALP Active site – – WB R1CTF Ig- R1CTF
3 Development of Adiponectin Receptor C-Terminal Fragment Bioassays
68
phenyl phosphate (PNPP) substrate. Calibrators and standards were made by conju-gating IgG (Fitzgerald 30-AI17) to AdipoR1 CTF 375-351 (25 mer).
ELISA methods for measuring the free forms of CTF R1 utilized a polyclonal rabbit antibody raised to CTF 25 mer and conjugated to ALP (See on-line supple-ment for full procedure at www.extras.springer.com/2015/978-3-319-21926-4 ). The R1 CTF 25 mer is used for calibration and control. Samples were diluted 1–12 in TBS containing protease inhibitor. For human plasmas, Protein A is used to remove Ig-CTF from human plasma prior to assay for free form.
3.2.6 Mass Spectroscopy Peptide Enrichment Assays
Immuno-affi nity enrichment after trypsin digestion using magnetic particle was conducted to measure AdipoR CTF. The SISCAPA method (SISCAPA Assay Technologies Inc.) and the immuno-affi nity MALDI (iMALDI) methods (MRM Proteomics Inc.) were followed according to literature (Whiteaker et al. 2007 ; Camenzind et al. 2013 ). In brief, both methods used magnetic particles with anti-bodies to enrichment for peptides of interested. The protocols starts fi rst with serum or plasma samples being denatured, reduced, alkylated and digested with trypsin to yield sample digests; stable-isotope labeled synthetic versions of target peptides are added in known amounts as internal standards; target-specifi c anti- AdipoR CTF peptide antibodies are captured on protein G magnetic beads; the magnetic bead bound antibodies are added directly to the sample digests where they capture target peptides; and the beads are then washed to remove unbound peptides.
Both methods used arginine isotope labels ( 13 C6; U- 15 N4) in a heavy form of the target VHFYGVSNLQEFR peptide as an internal standard with a 10 D mass differ-ence from native endogenous peptide. The heavy peptide was >90 % purity and stable in 30 % acetonitrile and 0.1 % formic acid at −80 °C. Limits of detection were generated by titrating unlabeled light peptides from 10 pmol to 56 amol, while the heavy version of the peptide remained constant at 500 fmol.
In the SISCAPA approach, bound heavy and light target peptides are eluted from the antibody by low pH and the target peptides and their respective internal standards are quantitated by mass using a LC-MS/MS. ZORBAX 300 SB-C8 column held at 40 °C on an Agilent 6490 triple quadrupole mass spectrometer. The target peptides were separated using a 3-min gradient with 0.1 % formic acid in water as solvent A and 90 % acetonitrile in 0.1 % formic acid in water as solvent B. Source conditions included drying gas at 200 °C, sheath gas at 250 °C and 11 L/min fl ow for both drying and sheath gases. Eight transitions were monitored using optimized collision energies: four equivalent transitions for each of the light (endogenous) and heavy (IS) peptides.
In the iMALDI approach, beads with bound heavy and light target peptides are placed directly on the MALDI target plate, without prior elution. The beads were washed three times with 15 % ACN, 25 mM ammonium bicarbonate prior to addi-tion. After washing, a fi nal 3 μL of 25 mM ammonium bicarbonate were added to the beads, the solution centrifuged briefl y and spotted onto a MALDI target plate
M. Pugia and R. Ma
69
using an AB/Sciex 4800 MALDI-TOF/TOF system (Applied Biosystems, Foster City, CA, USA). Next, 1 μL of CHCA matrix was added on top of the dried spot. The samples analyzed in refl ector positive mode with delayed extraction over a mass range of 800–4000 m/z, collecting 1250 laser shots per spot. The AB 4000-Series Explorer software was used for instrument control and the AB 4000 Series Data Explorer (v4.9) was used for data analysis.
3.3 Results
3.3.1 Mass Spectroscopy Discovery Technique
The AdipoR1 C-terminal fragment (AdipoR CTF) was discovered in plasma using immunoaffi nity mass spectroscopy and anti-CTF antibodies without any protease treatment of sample (Pugia et al. 2009 ). The original SELDI-MALDI method (Ciphergen) detected the native free peptide mass in circulation without protease digestion ( See Fig. 3.3 ). Two CTF masses observed were at 3378–3381 Da and 3450–3498 Da corresponding to 32 amino acid peptide (AdipoR
3381
3498
3100 3200 3300 3400 3500 3600
3100 3200 3300 3400 3500 3600
5 X mag
Normal, n = 5
Diabetic, n = 10
Fig. 3.3 Mass spectra of peptide after antibody separation from plasma. SELDI-MALDI mass spectrograph is shown after using antibody for surface capture for AdipoR CTF 344-374 as native free peptide without protease digestion mass in plasma of fi ve diabetes is compared to fi ve non- diabetics. Two CTF masses observed were at 3378–3381 Da and 3450–3498 Da corresponding to 32 amino acid peptide (AdipoR CTF 344-374 of 3475 Da) and 31 amino acid peptide (AdipoR CTF 343- 374 of 3376 Da)
3 Development of Adiponectin Receptor C-Terminal Fragment Bioassays
70
CTF 344-374 of 3475 Da) and 31 amino acid peptide (AdipoR CTF 343-374 of 3376 Da). The masses were confi rmed with the new two monoclonal antibodies spanning the length of the 32 amino acid peptide (ATCC 444-1D12-1H7 & ATCC 461-4H11.2A4). Mass spectroscopy showed CTF dimers in plasma. The R1-R1, R2-R2 and R1-R2 dimers were confi rmed in plasma by western blot and by ELISA. Finally, this method showed CTF 32-mer did not bind to immobilized adiponectin supporting the fragments function was not part of receptor binding of adiponectin.
3.3.2 Discovery of Immunoglobulin CTF
The mass spectra discovery was followed up with western blots using the monoclo-nal antibodies to CTF. Native gel and denatured gels of plasmas from patients with normal, impaired glucose tolerance (IGT), and diabetes by oral glucose tolerance testing (OGTT) showed cross reactivity to Anti R1 CTF 25 mer pAb –ALP (See on-line supplement data for western blot method at www.extras.springer.com/2015/978-3-319-21926-4 ). Antibodies bound to unknown high molecular weight plasma components at 27, 42, 50 and higher kDa.
These forms were identifi ed as CTF bound to immunoglobulin (Ig CTF). Native and denatured gels showed the CTF bands to cross react to antibodies for immuno-globulin chains. These high molecular weight forms of CTF were also found in rat, mouse and rabbit blood. While free CTF was observed in plasma it was primarily as the dimer. The full receptor was only found in tissue samples and not plasma. While free CTF is often undetected in diabetics, the Ig CTF forms was found in the plasma of normal, pre- diabetic, and diabetic patients. Similar results were also observed for the Adiponectin Receptor 2 fragment.
Commercial grade immunoglobulins (IgG) showed cross reactivity to Anti- R1CTF mAb 444 –ALP. The western blot of these materials exhibited binding to the same high molecular weight fragments (27, 50 and higher kDa) seen in human plasmas. Filtration to remove proteins >10 kDa, also removed any detectable cross reactivity. Denaturing gels under SDS-PAGE reducing gel 18 % with 2-mercapto-ethanol did not break bonds to liberate CTF from immunoglobulins.
3.3.3 Immunoglublin Binding Mechanism
CTF was thought to be covalently bound to immunoglobulin (Ig CTF) as it was not released by simply breaking the disulfi de bonds. Alkaline conditions (250 mM Tris buffer, pH 11.5, and heated at 37 °C at 2 h) did liberate CTF from immunoglobulin (See on-line appendix for western blots at www.extras.springer.com/2015/978-3- 319-21926-4 ). Re-blotting same gel showed IgG remained intact on gel and still reacted to anti-IgG but not to anti-CTF. Alkaline conditions are also required to
M. Pugia and R. Ma
71
break the ester between C3b and immunoglobulin chains (Sim and Twose 1981 ; Lutz and Stammler 1993 ; Suhn and Pangburn 1994 ). Therefore it was reasoned that CTF could behave in a similar mechanism. Native complement component C3 binds C3b to IgG by means of A reactive thioester as a means for immune complex solubilization, clearance of soluble immune complexes, phagocytosis of opsonized target cells and in stimulation of the alternative complement pathway.
The thioester of C3b is between Cysteine and Glutamine in the peptide sequence Gly-Cys-Gly-Glu-Glx- Asn-Met. These reactive thioesters are capable of forming esters or amides with peptides when immunoglobulin is brought into proximity. A mechanism for CTF attachment to IgG that is the same as C3b would require an internal thioester to be formed between cysteine and glutamic acids in the peptide R1 CTF sequence Glu-Gly-Gly-Cys-Thr-Asp-Asp-Thr-Leu-Leu and R2 CTF sequence Gly-Gly-Gly-Cys-Ser-Glu-Glu-Asp-Ala-Leu.
3.3.4 Immunoglublin Chain Identifi cation
Analysis of purifi ed human IgG showed CTF attaches to the 50 kDa gamma heavy chain and 26 kDa kappa/lambda light chains (see on-line appendix for western blot at www.extras.springer.com/2015/978-3-319-21926-4 ). CTF also attached to the light chains of IgA and IgM. Smaller light chain fragments of 15, 13 and 12 kDa also contained CTF. The Fc effectors function domain of the gamma heavy chain did not contain CTF. This indicates that CTF is attached to gamma and kappa/ lambda chains in the Fab region of IgG.
ELISA confi rmed CTF is integrated into IgG, IgA, IgM, IgD and IgE isotypes using an anti- kappa and anti CTF monoclonal sandwich assay to measure immuno-globulin bound CTF (Ig-CTF). Anti-gamma chain and anti CTF monoclonal sand-wich assay measured bound CTF only in the Ig isotype. Similarly, other specifi c monoclonal antibodies allowed sandwich assays that measured CTF bound to IgG, IgA, IgM, IgD or IgE (See on-line appendix for ELISA method for immunoglobulin bound CTF at www.extras.springer.com/2015/978-3-319-21926-4 ). It appears Ig-CTF is polyclonal as both IgG and IgM bands in native and denatured western blots were diffuse bands.
3.3.5 Mass Spectroscopy Peptide Enrichment Assays
Immuno-affi nity enrichment after trypsin digest using magnetic particles has become a common approach to develop clinical assays by mass spectrometers. Two such methods, the SISCAPA and iMALDI techniques were tested as potential CTF assays (Anderson et al. 2004 ; Whiteaker et al. 2007 , 2010 ; Eyford et al. 2009 ; Camenzind et al. 2013 ; Anderson et al. 2012 ). These methods are optimized to detect fM levels of peptides in the 400–3000 Da range. We found both methods
3 Development of Adiponectin Receptor C-Terminal Fragment Bioassays
72
required high-affi nity anti-peptide antibodies and typical affi nity mouse mAb were not effective. Either a rabbit mAb or rabbit pAb proved effective. The rabbit mAb clones were selected based upon nanomolar affi nity and slow off-rates. The best binding kinetics were: ka of 1.38E6 M-1s-1, kd of 1.24E-4s-1, KD of 0.09 nM and half off-time of 93.0 min.
Both SISCAPA and iMALDI methods, used the antibodies against the 12-mer fragment (AdipoR1 CTF 351-362 ) as this peptide was ionized very effi ciently and detection at lower concentrations . The 32-mer, 25-mer, or 19-mer fragments did not ionize well by electrospray or by MALDI using a variety of matrices. As a result the clinical 32-mer was not detectable in plasma at clinical concentrations and required over 18–35 ng in the sample.
Instead both methods had to form the 12-mer fragment by trypsin treatment of the 32-mer. The SISCAPA method showed excellent quantifi cation for 12-mer added to human plasma with a LOD of 11.1 fmol and LOQ of 24.6 fmol. The iMALDI method had an LOD of 1.28 fmol for 12-mer in buffer but only an LOD of 500 fmol in plasma. Non-specifi cally bound plasma peptides may be causing ion-ization suppression of the 12-mer signal or the complex nature of the plasma digest inhibited the polyclonal antibodies ability to bind and retain the 12-mer through the iMALDI procedure.
However, endogenous CTF (bound or unbound) in the native sample could not be found. It was confi rmed that the immunoglobulin linkage of CTF interfered with tryptic cleavage. The 12-mer CTF peptide was formed after dissociating the endog-enous CTF from the immunoglobulin by bringing the sample to pH ~11.5 using Tris- Base and incubating at 37 °C for 2 h prior to digesting the sample. However, this increased the complexity of the assay and proved no more sensitive than ELISA.
3.3.6 Analytical Validation of ELISA
3.3.6.1 Cross Reactivity
Low molecular weight peptide analysis gels (10–20 % Tris-Tricine) showed the free CTF peptide standards readily dimerized and were detected at ~6–7 kDa and contained no masses at 27, 42, 50 kDa or higher. The R1 antibodies showed no R2 cross reactivity nor did the R2 antibodies show any R1 cross reactivity.
3.3.6.2 Calibration Reproducibility and Stability
Calibrator reproducibility and stability were achieved by use of citrate-phosphate- BSA buffer at pH 6.4. The calibrator stock used for daily dilution was stable stored 5–8 °C for 6 months and at −20 °C for 4 months. Calibrators had to be fresh daily and used within 24 h at 5–8 °C. Freezing dilute calibrators depressed values.
M. Pugia and R. Ma
73
3.3.6.3 Analytical Validation
Precision and Accuracy testing was conducted on the final Ig-CTF ELISA method over 3 days for ten sample levels on four separate plates for each day. Each plate used with separate calibration curves. The overall repeatability of the assay was a 2.6 % and the overall accuracy was 11.2 % in the reportable range of 95–4900 ng/mL. Accuracy was reduced at lower concentrations. The sensitivity limit was 13.5 ng/mL. No prozone effect was observed up to 21,200 ng/mL.
3.3.6.4 Within Patient Reproducibility and Stability
Specimens were collected in EDTA, heparin and sodium fl uoride tubes on day 1 as non-fasting samples, days 2 and 45 after overnight fasts. Patient weight, height and fasting blood glucose meter readings were taken (10 % were diabetic and 40 % were obese). Samples were kept at 5 °C for <8 h until whole blood and plasma were separated and sample dilutions made. Dilutions were stable for 1 day at 5–25 °C and for 6 months at −20 °C. Samples were stable for −20 °C for 12 months. Three freeze/thaw cycles were tested. Variance was tested by three separate dilution and duplicate determinations were made for each specimen condition.
3.3.6.5 Recovery
Recovery was tested on specimens collected from seven patients across the report-able range. The samples were diluted 1:12: 1:24, 1:48, 1:96 and 1:187. The target dilution was 1:48. Results at twice the target dilution and half the target dilution recovered as expected. Whole blood and plasma from the same patient showed the same recovery. The recovery fell off at dilution were 1:12 and 1:187. The recovery at 1:12 was 46 % of expected values, while recovery at 1:187 was 71 % of expected values.
3.3.6.6 Interference Testing
Interference testing was conducted by contriving six plasma samples with the potential interference. Hemolysis, icterus, lipemia, anti-coagulates, proteinemic and human anti-mouse antibody interference (HAMA) impact was tested by addition of 500 mg/mL hemoglobin, 60 mg/dL bilirubin, 500 mg/dL cholesterol, 10,000 mg/dL triglycerides, 2 IU/mL heparin, 12.0 g/dL of albumin, 60 mg/mL goat and mouse IgG. Human immunoglobulin (IgG and IgM) may contain natural CTF-Ig (show by western blot) and is not suitable for cross reactivity testing.
3 Development of Adiponectin Receptor C-Terminal Fragment Bioassays
74
3.3.6.7 Blood Levels of CTF and Ig-CTF by ELISA
Analysis of patient plasma by ELISA was possible for immunoglobulin bound CTF (Ig-CTF) and for free CTF (See on-line appendix for ELISA method for free CTF at www.extras.springer.com/2015/978-3-319-21926-4 ). Bound CTF in patient plasma was found to be <0.2 % of all plasma IgG. This percentage was similar whether measuring Ig CTF R1 or Ig CTF R2. The percentage of IgM with CTF was typically lower, <0.03 % of all plasma IgM. The number of CTF attached to IgG was typically 1–2, in most patients with a range of 1–5.
Assays for bound Ig-CTF and free CTF by ELISA were possible. To measure free CTF, the bound form was removed by Protein A column as the antibodies do no distin-guish between the free and the bound forms. The amount of Ig-CTF in plasma, 95–4900 ng/mL was signifi cantly greater than amount of free CTF, 0.05–5.0 ng/mL with less than 0.1 % in free from. While the free CTF ELISA assay works for rat blood, the interference in human blood was too large to develop a clinical assay for free CTF.
3.4 Conclusion
Discoveries based on peptides detected by immune affi nity mass spectroscopy method such as SELDI can be useful discovery tools when trypsin is not used. However these methods can be misleading as to true complexity of protein forms in native samples. Mass spectroscopy based on trypsin digestion of native samples such as SISCAPA and iMALDI were misleading in this case as native forms were resistant to treatment trypin hydrolysis. The mass spectroscopy peptide enrichment was unable to detect the free peptide until the bound CTF was released by base treatment to liberate free form which could be trypsinized.
Traditional bioassay such as Western blot and ELISA are recommended. These methods were suitable for testing of patient and biological samples (See on-line supplement for full procedures at www.extras.springer.com/2015/978-3- 319-21926-4 ) and can be used for studies on Ig CTF formation, absorption into tissue (liver, pancreas, brain), and CTF mechanism action. These assays could be useful for screening of compounds that block or lead to Ig CTF formation, and for investigation of mechanism action.
Acknowledgements We would like to acknowledge the work done by Siemens Healthcare Diagnostic people including Mary Foltz, Jill Perry, David Brock, Qingping Jiang, Yu Hui Lin, Jim Freeman and others who helped prepare antibody, conjugates, peptides, and assay formats. We would like to acknowledge the mass spectroscopy work done by the SELDI group of Dr. Saeed A. Jortani, Department of Pathology and Laboratory Medicine, University of Louisville, School of Medicine, by the SISCAPA group of Dr. Terry W. Pearson and Dr. Matthew Pope of SISCAPA Technologies at the University of Victoria BC Canada Victoria., and by the i-MALDI group of Dr. Christoph Borchers & Dr Andrew Munk of MRM Proteomics at the University of Victoria BC Canada Victoria.
M. Pugia and R. Ma
75
3.5 Supplements
• CTF Peptide Sequences • ELISA Method for IgG-CTF assay • ELISA Method for free CTF assay • Western Blot Methods
References
Anderson NL et al (2004) Mass spectrometric quantitation of peptides and proteins using Stable Isotope Standards and Capture by Anti-Peptide Antibodies (SISCAPA). J Proteome Res 3:235–244
Anderson NL et al (2012) Precision of heavy-light peptide ratios measured by MALDI-TOF mass spectrometry. J Proteome Res 11:1868–1878
Bråkenhielm E et al (2004) Adiponectin-induced antiangiogenesis and antitumor activity involve caspase-mediated endothelial cell apoptosis. Proc Natl Acad Sci U S A 101(8):2476–2481
Camenzind AG et al (2013) Development and evaluation of an immuno-MALDI (iMALDI) assay for angiotensin I and the diagnosis of secondary hypertension. Clin Proteomics 10:20
Daimon M, Kato T et al (2003) Funagata study. Decreased serum levels of adiponectin are a risk factor for the progression to type 2 diabetes in the Japanese Population: the Funagata study. Diabetes Care 26(7):2015–2020
Eyford BA, Anderson NL, Pearson TW (2009) Anti-peptide antibody screening: selection of high affi nity monoclonal reagents by a refi ned surface plasmon resonance technique. J Immunol Methods 341:86–96
Gil-Campos M, Canete RR, Gil A (2004) Adiponectin, the missing link in insulin resistance and obesity. Clin Nutr 23:963–974
Hara K et al (2006) Measurement of the high-molecular weight form of adiponectin in plasma is useful for the prediction of insulin resistance and metabolic syndrome. Diabetes Care 29(6):1357–1362
Hug C et al (2004) T-cadherin is a receptor for hexameric and high-molecular-weight forms of Acrp30/adiponectin. Proc Natl Acad Sci U S A 101(28):10308–10313
Kadowaki T et al (2002) Adiponectin stimulates glucose utilization and fatty-acid oxidation by activating AMP-activated protein kinase. Nat Med 8:1288–1295
Kadowaki T et al (2003) Impaired multimerization of human adiponectin mutants associated with diabetes. J Biol Chem 278:40352–40363
Kadowaki T, Yamauchi T, Tobe K et al (2006) Adiponectin and adiponectin receptors in insulin resistance, diabetes, and the metabolic syndrome. J Clin Invest 116:1784–1792
Kubota N, Yamauchi T, Kadowaki T et al (2002) Disruption of adiponectin causes insulin resis-tance and neointimal formation. J Biol Chem 277:25863–25866
Lutz HU, Stammler P (1993) Preferential formation of C3b-IgG complexes in vitro and in vivo from nascent C3b and naturally occurring anti-band 3 antibodies. J Biol Chem 268:17418–17426
Ma K, Chan L et al (2002) Increased beta -oxidation but no insulin resistance or glucose intoler-ance in mice lacking adiponectin. J Biol Chem 277:34658–34661
Maeda K et al (1996) cDNA cloning and expression of a novel adipose specifi c collagen-like fac-tor, apM1 (adipose most abundant gene transcript 1). Biochem Biophys Res Commun 221:286–289
Maeda N, Matsuzawa Y et al (2002) Diet-induced insulin resistance in mice lacking adiponectin/ACRP30. Nat Med 8:731–737
3 Development of Adiponectin Receptor C-Terminal Fragment Bioassays
76
Mao X et al (2006) APPL1 binds to adiponectin receptors and mediates adiponectin signalling and function. Nat Cell Biol 8:516–523
Menzaghi C, Trischitta V, Doria A (2007) Genetic infl uences of adiponectin on insulin resistance, type 2 diabetes, and cardiovascular disease. Diabetes 56:1198–1209
Nakano Y et al (1996) Isolation and characterization of CBP28, a novel gelatin-binding protein purifi ed from human plasma. J Biochem 120:803–812
Ouchi N, Walsh K (2007) Adiponectin as an anti-infl ammatory factor. Clin Chim Acta 380(1–2):24–30
Phillips SA et al (2003) Modulation of circulating and adipose tissue adiponectin levels by antidia-betic therapy. Diabetes 52(3):667–674
Pugia M (2012) Detection of soluble adiponectin receptor peptides and use in diagnostics and therapeutics. US Patent 8,093,017, 10 Jan 2012
Pugia M, Ma R (2013) Adiponectin receptor fragments and methods of use. US Patent 8,017,573, 13 Sept 2011 and US Patent 8,377,889, 19 Feb 2013
Pugia MJ et al (2009) Adiponectin receptor-1 C-terminal fragment (CTF) in plasma: putative bio-marker for diabetes. Clin Proteomics 5:156–162
Sattar N et al (2006) Adiponectin and coronary heart disease: a prospective study and meta- analysis. Circulation 114(7):623–629
Scherer PE et al (1995) A novel serum protein similar to C1q, produced exclusively in adipocytes. J Biol Chem 270:26746–26749
Shapiro L, Scherer PE (1998) The crystal structure of a complement-1q family protein suggests an evolutionary link to tumor necrosis factor. Curr Biol 8:335–338
Sim RB, Twose TM (1981) The covalent-binding reaction of complement component C3. Biochem J 193:115–127
Spiegelman BM et al (1996) AdipoQ is a novel adipose-specifi c gene dysregulated in obesity. J Biol Chem 271:10697–10703
Suhn A, Pangburn MK (1994) Covalent attachment of human complement C3 to IgG. J Biol Chem 269:28997–29002
Szmitko PE, Teoh H, Stewart DJ et al (2007) Adiponectin and cardiovascular disease: state of the art? Am J Physiol Heart Circ Physiol 292:H1655–H1663
Tsao TS et al (2002) Oligomerization state-dependent activation of NF-kappa B signaling pathway by adipocyte complement-related protein of 30 kDa (Acrp30). J Biol Chem 277(33):29359–29362
Whiteaker R et al (2007) Antibody-based enrichment of peptides on magnetic beads for massspectrometry- based quantifi cation of serum biomarkers. Anal Biochem 362:44–54
Whiteaker R et al (2010) An automated and multiplexed method for high throughput peptide immunoaffi nity enrichment and multiple reaction monitoring mass spectrometry based quanti-fi cation of protein biomarkers. Mol Cell Proteomics 9:184–196
Yamauchi T, Kadowaki T et al (2003) Cloning of adiponectin receptors that mediate antidiabetic metabolic effects. Nature 423(6941):762–769
M. Pugia and R. Ma
77© Springer International Publishing Switzerland 2015 M. Pugia (ed.), Infl ammatory Pathways in Diabetes: Biomarkers and Clinical Correlates, Progress in Infl ammation Research, DOI 10.1007/978-3-319-21927-1_4
Chapter 4 Protease Inhibition and Biological Distribution of the C Terminal Fragment of Adiponectin Receptor
Michael Pugia and Rui Ma
Abstract Protease inhibition studies showed Adiponectin Receptor C- Terminal Fragment (AipoR CTF) inhibited insulin-degrading enzyme (IDE) and TNF alpha cleavage protease (TACE). Cleavage of Adiponectin Receptor by TACE to form CTF was demonstrated by western blot assays analysis of recombinate AdipoR1 treated with TACE and by inhibition studies. Tissue and blood analysis by standard immuno histochemistry (IHC) and immuno ctyochemistry (ICC) methods demon-strated CTF is primarily absorbed into the liver, pancreas and brain tissue where it is co-located with IDE. CTF formation by TACE appears to occur in fat and muscle tissues and to be carried in circulation by an immunoglobulin bound form (Ig-CTF) that freely circulates and is immune cell bound in both blood and lymph nodes. Tissue from animal models showed levels of CTF increase with disease progression from non-diabetic to diabetic states. All of these data unveiled a novel regulation mechanism of AdiopR CTF in its free and Ig-associated forms which could impact insulin sensitivity and prevent insulin degradation by IDE in tissue and cells.
4.1 Introduction
This chapter demonstrates methods to determine the function and biological loca-tions of peptides such as 32 amino acid peptide (AdipoR CTF 344-374 ) discovered in the plasma. Identifi cation of peptidases likely to interact with the peptide is a useful starting point. The peptide sequence can be compared to the peptide cleavage
M. Pugia , PhD (*) Strategic Innovation , Siemens Healthcare Diagnostics , 3400 Middlebury Street , Elkhart , IN 46515 , USA e-mail: [email protected]
R. Ma Siemens Healthcare , 278 Zhouzhu Road , Pu Dong New Area 201318 , People’s Republic of China
78
sequences for known peptidase. Once identifi ed, peptidases of interest can be used to explore the mechanism of peptide formation and function. Integrated data bases source such as MEROPS that have a hierarchical classifi cation system that allows identifi cation of homologous sets of peptidase and protein inhibitor sequences (Rawlings et al. 2006 ). A MEROPS search identifi ed insulin-degrading enzyme (IDE) and TNF α alpha cleavage protease (TACE) as potential peptidases for CTF ( See Sect. 4.3.1 ). As Adiponectin has insulin sensitizing properties and ability to inhibit TNF α signaling that cannot be fully explained, both IDE and TACE are candidates of signifi cant interest to offer new understandings of the pathway (Gil-Camposa et al. 2004 ; Ouchi et al. 2000 ).
4.1.1 Insulin-Degrading Enzyme (IDE)
Insulin-degrading enzyme (IDE) is also known as β -secretase (BACE) or insulysin and fragments the α , β chains of insulin at nine cleave sites (Duckworth et al. 1998 ). Insulin degradation is a regulated process controlling the removal and inactivation of the hormone. Degradation occurs through an endosomal degradation pathway which controls the levels of intracellular insulin. Intracellular insulin impacts the recycling of the insulin receptor and receptor sensitivity (Kahn et al. 1973 ). While circulating insulin is primarily cleared in the liver and kidneys, IDE is present in all insulin sensitive tissues and is alternative feedback mechanism for insulin clearance and receptor sensitivity (Kuo et al. 1993 ).
Pancreatic secretion increases blood concentrations of insulin and c-peptide in obesity (Tura et al. 2001 ). The patient presents with hyperinsulinemia without gly-caemia and is categorized as insulin resistant. The degree of insulin resistance worsen through stages of worsening glucose intolerance and is measured using fre-quent sampled insulin and glucose values during an oral glucose tolerance test (Sathananthan 2012 ). As obesity progresses to diabetic glucose intolerance, glycae-mia starts to be routinely observed and insulin secretion starts to decrease below normal levels (Bonora et al. 1983 ).
Insulin clearance rates decrease in obesity and diabetes and correlate with reduced IDE effectiveness to degrade insulin (Duckworth et al. 1998 ; Abdul-Hay et al. 2011 ). Deletion of IDE in animal models does cause insulinemia without immediate glycaemia. Insulin resistance with reduced insulin receptor and reduced glucose tolerance does occur as the animals ages in response to chronic hyperinsu-linemia (Abdul-Hay et al. 2011 ). Inhibition of IDE may constitute a valid approach to the treatment of diabetes and Alzheimer’s disease with synthesized inhibitors (Leissring et al. 2010 ; Maianti et al. 2014 )
IDE also cleaves on insulin-like growth factor I and II (IGF), transforming growth factor α (TGF), Aβ protein precursor (APP), β - amyloid peptide (Aβ), and glucagon (Duckworth et al. 1998 ; Kuo et al. 1993 ). Alzheimer’s disease (AD) is caused by aggregates of the amyloid peptide (Aβ), which is generated by cleavage of the Aβ
M. Pugia and R. Ma
79
protein precursor (APP) by IDE followed by γ-secretase (Farris et al. 2003 ). Therefore IDE contributes to Aβ 42 oligomers noted in amyloid plaques and tau pathology in cognitive defects (Tamboli et al. 2010 ). Breakdown of extracellular Aβ by insulin-degrading enzyme in microglia is thought to be central to Alzheimer diseases (Tamboli et al. 2010 ). However while advances in synthetic IDE inhibitor design are reported natural inhibitors of IDE remain unknown (Maianti et al. 2014 ; Olson et al. 2007 )
4.1.2 TNF-α–Converting Enzyme (TACE)
TNF-α–converting enzyme (TACE) processes pro- tumor necrosis factor -α (TNF- α) to active TNF-α in the bloodstream and is central to the infl ammatory response (Black 2002 ). Insulin sensitivity, glucose, and lipid metabolism are associated with TNF-α. Deletion of TACE or TNF-α in animal models protect from insulin resistance and obesity (Serino et al. 2007 ; Black 2002 ). TNF-α mediates cytokine signaling and cytokine activation of c-Jun N-terminal kinase (c-JNK) which can phosphorylate the insulin receptor substrate and inhibit insulin signaling (Lorenzo et al. 2008 ). TNF- α neutralization decreases circulating free fatty acids in obese animals and lowers plasma glucose concentrations and weight increases (Weickert and Pfeiffer 2006 ). Adipose tissue of obese humans with insulin resistance also expresses elevated levels of TNF- α mRNA relative to lean individuals (Hotamisligil 1994 ) .
4.1.3 Goal
Our goal was to investigate the function of the C -terminal fragment (CTF) as a potential new avenue for the mechanism of action of adiponectin. We explored TACE enzyme as the likely upstream protease to release CTF and IDE as the likely downstream protease to be inhibited by CTF. We also wanted to determine the potential tissues for formation and accumulation of CTF.
4.2 Methods
4.2.1 Protease Identifi cation
The 32 amino acid peptide (AdipoR CTF 344-374 ) found in plasma was submitted to the MEROPS peptidase database ( http://merops.sanger.ac.uk/ ) for potential prote-ase search. TNF-α–converting enzyme (TACE, also known as A disintegrin and metalloproteinases 17, ADAM17) was the only recorded protease to cleave peptide
4 Protease Inhibition and Biological Distribution of the C Terminal Fragment
80
sequence X-X-X-X + Val-Val-Ala-Ala matching AdipoR CTF 344-374 cleavage site. Another protease able to cleave AdipoR CTF 344-374 and sequences of X-X-X- Leu + Val- X-X-X is IDE (insulin-degrading enzyme or Insulysin). Other protease able to cleave this simple sequence include, pepsin (A&B), renin, cathepsin (D, E, L, G & H), leucolysin, enprilysin, endothelin converting enzyme, aminopeptidase Ap1, deutrolysin, camelysin and peptidase C1 & K.
4.2.2 Inhibition Studies
Inhibition studies were performed using InnoZyme™ Insulysin-IDE Immunocapture Activity Assay Kit and InnoZyme™ TACE Activity Kit purchased from EMD Chemicals, Inc. (Gibbstown, NJ). Recombinant human TACE standard was purchased from R&D Systems Inc. (Minneapolis, MN). IDE standard was from the kit. CTF peptides at concentrations of 0, 12.5, 25, and 50 mg/L were incubated with IDE or TACE standard coated on the assay plate along with the fl uorescent substrates and controls. The assay was read over 0–300 min to monitor the change of relative fl uorescence units (RFU) recorded at Ex325/Em405 using an FLx800 Fluorescence Microplate Reader (BioTek Instruments, Inc., Winooski, VT).
4.2.3 Tissue Sample Preparation
At termination, animals were perfused with saline. Brain, pancreas, liver, lymph nodes, fat and quadriceps muscle were harvested from rats graded as normal (n = 3), short term diabetic (n = 3), on long term diabetic (n = 3). Tissues were also harvested from rabbits which did (n = 3) and did not (n = 3) produce polyclonal antibodies to CTF. After PBS perfusion tissues was embedded for frozen tissue slices with OCT compound, placed in disposable base in ice bath and then frozen at −70 °C (Ramos-Vara 2005 ). Cyro slices at 8 μm were made on to permafrost Fisher slides (AML laboratories). Separate slices (n = 20) were obtained for each tissue sample with additional slices for left, right cerebral hemisphere and cerebellum. Slides with tis-sue slices were stored at −70 °C.
4.2.4 Blood Sample Preparation
Blood samples from rats and human were collected fresh in EDTA tubes and stored at 4° for less than 8 h.
M. Pugia and R. Ma
81
4.2.5 Tissue and Blood Analysis
Tissues and blood sample were stained using standard immuno histochemistry (IHC) and immuno ctyochemistry (ICC) methods (See on-line supplement for full procedures at www.extras.springer.com/2015/978-3-319-21926-4 ). The slide area containing tissue or a 10 μL whole blood smear was circled with a hydrophobic pen to create an area for holding liquid. Then 100 μL of 5 % formaldehyde in phosphate buffered saline (PBS) was added to the holding area followed by incubating for 15 min at RT. The slide was permeabilized in PBS with 0.2 % Triton-X 100 for 5–15 min. After air drying, the Fc receptors were blocked with 50 μL IgG in PBS (5 mg/ml G4368 Sigma Aldrich) for 15 min RT, washed in PBS with 0.1 % Tween 20 at 5 min RT. Slides were stained with 50 μL of 4,6 diamidino-2- phenylindole dihydrochloride (DAPI) (0.01 mg/mL) & 100 μL of one or more antibody with fl uorophore (FITC, TR or CY5). Each antibody was at 0.1 mg/mL in PBS. The slide was incubated for 15 min at 37 °C and washed in cell wash buffer (PBS/Tween 20 at 0.1 %) in a wash bath for 5 min. Cover slip was applied with 5 μL DAPCO.
Phase contrast and fl uorescence microscopy was conducted with Leica DM5000 and using the fi lter sets for reading four fl uorophores (DAPI, FITC, TR and CY5) signals. The slides are scanned at 40 X using 50–300 ms exposures. A grading sys-tem was established for consistency across all slides in an experiment.
4.2.6 Cleavage Study
Analysis of recombinant AdipoR1 (62 kDa from Novus) before and after treatment with TACE (R&D Systems) was conducted by Western Blot techniques (See on-line supplement for full procedure at www.extras.springer.com/2015/978-3- 319-21926-4 ) using with 100 ng AdipoR1 and 1 ug TACE incubated 37 °C and 24 h followed by western blot denatured gel with Anti R1CTF25 pAb-ALP (5 μg/ml).
4.3 Results
4.3.1 Protease Inhibition Studies
Both ADAM17/TACE and IDE activity showed inhibitory effects due to AdipoR1 CTF 344–375 ( See Figs . 4.1 and 4.2 ). A structure reactivity study was conducted for a series of AdipoR1 and R2 CTF peptides as inhibitors for ADAM17/TACE and IDE ( See Table 4.1 ). As a negative control, the lack of elastase or trypsin inhibition by AdipoR1 CTF 344-375 was confi rmed.
4 Protease Inhibition and Biological Distribution of the C Terminal Fragment
82
-20.0%
0.0%
20.0%
40.0%
60.0%
80.0%
100.0%
120.0%
0 30 60 90 120 150 180 210 240 270 300
% T
AC
E C
on
tro
l Act
ivit
y .
Time (Minutes)
CTF 32 mer 0µg/ml
CTF 32 mer 12.5 µg/mlCTF 32 mer 25 µg/ml
Fig. 4.1 Effects of AdipoR1 CTF 344-375 on the activities of ADAM17/TACE. Synthetic peptide Adiponectin Receptor C terminal fragment 32 mer peptide (AdipoR1 CTF 344-375 ) at concentrations of 0–50 μg/mL was incubated with ADAMI7/TACE from 10 to 120 min. The percent activities relative to that of the control sample (0 mg/L CTF1) were plotted
0%
20%
40%
60%
80%
100%
120%
0 50 100 150 200 250 300
100%
Co
ntr
ol A
ctiv
ity
Time (Minute)s
CTF 32 mer 0µg/ml
CTF 32 mer 12.5µg/ml
CTF 32 mer 25µg/ml
CTF 32 mer 50µg/ml
Fig. 4.2 Effects of AdipoR1 CTF 344-375 on the activities of IDE. Synthetic peptide Adiponectin Receptor C terminal fragment 32 mer peptide (AdipoR1 CTF 344-375 ) At concentrations of 0–50 μg/mL was incubated with IDE from 60 to 300 min. The percent activities (RFU) relative to that of the control sample (0 mg/L CTF1) were plotted
M. Pugia and R. Ma
83
A sharp dose- and time-dependent inhibition of ADAM17/TACE activity was observed with only the full length 32 mer sequences (See AdipoR1 and R2 CTF 344-375 in Table 4.1 and Fig . 4.1 ). Activity resumed and remained at about 50 % after 60 min of incubation. The acute decrease of activity and recovery supported TACE is the possible cleavage protease. The sequence of AdipoR1 and R2 CTF 344-375 of Val-Val-Ala-Ala at the cleavage points matches the known cleavage sequence for ADAM17/TACE.
Additional experiments supported that ADAM-17/TACE was the cleavage prote-ase. Western blots of recombinant AdipoR1 (62 kDa from Novus) before and after treatment with TACE (R&D Systems) showed AdipoR1 cleaved into a 27 kDa frag-ment and a 7 kDa fragment matching a CTF dimer (See on-line supplemental west-ern blot data at www.extras.springer.com/2015/978-3-319-21926-4 ).
Table 4.1 Adiponectin Receptor CTF peptide inhibition data
Peptide name
Number of amino acids (mer) Form Description
ADAM-17 inhibition (observed)
IDE inhibition (concentration at 90 % inhibition)
AdipoR1 CTF 344-375
32 Monomer R1 clinical form Positive 150.0 μg/ml
AdipoR1 CTF 351-375
25 Monomer R1 soluble portion Negative 25.4 μg/ml
AdipoR1 CTF 358-375
17 Monomer Non inhibitory domain
Negative Negative
AdipoR1 CTF 367-375
9 Monomer Tail (end of C terminal)
Negative Negative
AdipoR1 CTF 344-353
10 Monomer Non inhibitory domain
Negative Negative
AdipoR1 CTF 351-362
12 Monomer Inhibitory domain (active site)
Negative 20.2 μg/ml
AdipoR2 CTF 344-375
32 Monomer R1 clinical form Positive 140.0 μg/ml
AdipoR2 CTF 351-375
25 Monomer R1 soluble portion Negative 9.0 μg/ml
AdipoR2 CTF 367-375
9 Monomer Tail (end of C terminal)
Negative Negative
AdipoR1 CTF 351-375 AdipoR1 CTF 351-375
25 Homodimer R1 CTF25 R1 CTF25
Negative 12.5 μg/ml
AdipoR2 CTF 351-375 AdipoR2 CTF 351-375
25 Homodimer R2 CTF25 R2 CTF25
Negative 9.0 μg/ml
AdipoR1 CTF 351-375 AdipoR2 CTF 351-375
25 Heterodimer R1 CTF25 R2 CTF25
Negative 5.2 μg/ml
4 Protease Inhibition and Biological Distribution of the C Terminal Fragment
84
A time-independent inhibition of IDE activity was observed with all sequences of CTF containing AdipoR1 CTF 351-362 , considered the active inhibitor site (See Table 4.1 and Fig. 4.2 ). The impact was not reversed with time. The amount of CTF needed for inhibition of 90 % (IC90) of the IDE activity was determined. The solu-ble 25- and 12-mer peptides were more active inhibitors, with the 12-mer identifi ed as the key inhibitory site.
The homology between the R1 CTF and R2 CTF is very high with key differ-ences only in the last nine peptides of the C terminal. No large inhibitory differences were observed between R1 and R2. R2 was slightly more inhibitory. The dimers for CTF were not stronger inhibitors than monomers. The hetero-dimer of R1-R2 is slightly active than the homo-dimers.
Structural comparison of insulin beta chain and CTF 32 mer was made as both bind IDE. There was no sequence homology between them. Molecular modeling showed structural similarity in the size and orientation of the α helixes (see Fig. 4.3 ). The inhibitory domain AdipoR1 CTF 351-362 is located the in bend between ß sheet located at AdipoR1 CTF 344-354 between the α helix located at AdipoR1 CTF 356-364 . The interaction point between the sheet and helix is Phe 353 and Arg 361. IDE has multiple insulin cleavage sites at multiple sites for insulin (Duckworth et al. 1998 ). The inhibitory domain of the CTF inhibitor could association with IDE and block insulin binding.
Adiponectin ReceptorC terminal fragment (CTF)
Insulin
Fig. 4.3 Molecular structure comparison of AdipoR CTF to insulin. I-TASSER structure for CTF 32-mer heterodimer of CTF R1 and CTF R2 is shown. The two cysteines were placed in docked orientation (ZDOCK). The α helixes of AdipoR1 and R2 located at CTF 356-364. are shown along with the between ß sheets AdipoR1 and R located at CTF 344-354. Insulin ß chain dimer complex of 4-hydroxybenzamine was taken from the protein database (PDB-ID: 2bn3, Resolution 1.40 Å). The α helix of insulin ß chain is from Val-Asn-Gln-His-Leu-Cys-Gly-Ser-His-Leu-Val-Glu-Ala-Leu- Tyr-Leu-Val-Cys-Gly (Images are shown as captured in RasMol viewer)
M. Pugia and R. Ma
85
4.3.2 Tissue Forms and Distribution
Multiple forms of bound CTF found were found in tissue and cell cultures by west-ern blots with monoclonal antibodies to CTF (See Table 4.2 and on-line supplement data for western blots at www.extras.springer.com/2015/978-3- 319- 21926-4 ). CTF bound to Ig, Fab2, IgG or IgM immunoglobulin (Ig CTF) was only observed in liver, muscle, and myocyte cell culture. The full receptor was found in fat, muscle and liver tissue samples and not plasma. Lower molecular weight forms of CTF bound to gamma chain or kappa light chain were observed in pancreas and mesen-teric fat tissues. TACE and IDE were found in liver, pancreas, muscle and mesen-teric fat of normal and diabetic rats. Normal and diabetic rat treated in one exposure with AdipoR1 30 and 32 mer exhibited no changes in tissue forms.
4.3.3 Plasma Forms
All Ig CTF and free CTF forms were observed in plasma ( See Table 4.2 ). Follow up with immuno ctyochemistry (ICC) of human whole blood showed Ig-CTF was freely circulating in plasma As well As on washed white blood cells (WBC) but not red blood cells (RBC) (See on-line supplement for cell and tissue staining proce-dure at www.extras.springer.com/2015/978-3-319-21926-4 ). WBC binding to Ig-CTF occurred on some neutrophils with polymorphonuclear pattern, basophils, T cell (CD3 positive by counter staining) and Beta cell lymphocytes (CD 19 posi-tive by counter staining). As the Fc receptor interaction was blocked, the WBC binding was assumed to be through CTF binding or absorption into cells. It was not determined whether CTF was in free form or IgCTF bound form when associated with WBC. Levels of CTF detected were above any expression level of AdipoR1 on WBC surface (Golz et al. 2007 ). Additional free circulating level TACE and IDE were also found in human blood by westerns blot with some differences noted in molecular weight between normal and diabetic samoles.
Table 4.2 CTF Immunoglobulin in human tissue and blood
Ig-CTF band (KD)
Observed kDa in Western blots
Human plasma
Fat tissue
Liver tissue
Muscle tissue
Pancreas tissue
Myocyte cell culture
IgM-CTF 250 – – 200 – 200 IgG-CTF 160 – – – – Fab2-CTF 100 – 100 110 – 100–110 Ig-CTF – 70 80 – 60 Gamma heavy chain-CTF 55 – 55 50 – AdipoR1 37–45 43 37–45 – 47 Light chain CTF 25 25 25 25 25 – Light chain CTF 22 22 22 22 22 – CTF dimer 7 – – – – –
4 Protease Inhibition and Biological Distribution of the C Terminal Fragment
86
4.3.4 Primary Tissues of Action
Immuno histochemistry (IHC) of rabbit and rat tissue showed that Ig-CTF is most strongly absorbed into the liver followed by the brain and pancreas but not the adipose (See Table 4.3 and on-line supplement for cell and tissue staining procedure at www.extras.springer.com/2015/978-3-319-21926-4 ). The liver is therefore a primary organ for CTF absorption. Counter staining for IgG showed also immunoglobulin built up. The liver, brain and pancreas are also where IDE primarily resides in agreement with IDE as the site of CTF action. The muscle and fat were the primary location for TACE in agree-ment with these tissue being primary sites for releasing CTF.
4.3.5 Fat Morphologic Observations
CTF was not detected intracellular and was located on outer membrane of white adi-pose. Cells were IgG negative on counter staining with anti-IgG antibody and CTF observed appears to be due to AdipoR. CTF was also clearly seen in outer membrane of brown fat and connective tissue. TACE counter staining was positive on the cell surface and co-located with CTF. Fat appears to be a source of CTF from but not a target of free CTF or CTF-Ig.
4.3.6 Muscle Morphologic Observations
CTF was not detected intracellular and was located on outer membrane muscle. Cells were IgG negative. Staining for CTF with mAb 461 was negative as this anti-body only detects membrane released CTF. Staining for CTF with mAb 444 was positive as this antibody detects CTF still bound to full AdipoR receptor. TACE Counteres staining was positive on the cell surface and co-located with CTF. Muscle appears to be a source of CTF Ig but not a target of free CTF or CTF-Ig.
Table 4.3 Intracellular CTF observed in IHC fl uorescent staining
Observed immuno histological stain results
Liver Cerebellum Cerebral right & left Pancreas Fat (adipose) Muscle
CTF Pos. 5+ Pos 1+ Pos 1+ Pos1+ Neg. Pos 1+ TACE Neg. Neg. Neg. Neg. Pos Pos IDE Pos Pos Pos Pos Neg. Neg. IgG counter staining
Pos where CTF found
Pos where CTF found
Pos where CTF found
Neg Neg Neg
CTF Staining grade on scale of 1–5 across all tissue on same day with replicates of three animal per determination. Constant microscopic exposure times were used to grade to same scale. For fat tissues like white adipose too fragile for making slice, a tissue homogenization method was used to make the slide
M. Pugia and R. Ma
87
4.3.7 Liver Morphologic Observations
CTF was detected in intra cellular regions of liver tissue with both CTF antibodies and was detected even after Fc blocking. TACE staining was negative. IDE staining was positive. IgG was detected inside and outside the cells. Liver could not be as source of CTF IgG but could a target of CTF or CTF-Ig to act on IDE.
4.3.8 Pancreas Morphologic Observations
CTF was not generally detected in intra cellular areas with either CTF antibodies even after Fc blocking. CTF was detected in the interstitial space, especially in areas which had some infl ammation with observed lesions and white blood cells. These WBC cells were IgG and CTF positive. There was IgG detected migrating into the tissue with CTF positivity near lesions. This behavior was observed with both CTF antibodies. CTF detected as receptor free epitope using mAb 444 was more often extra cellular as IgG-CTF and CTF detected as the membrane epitope using mAb 461 was more often inter cellular. This tissue was TACE negative and therefore not a source of CTF-Ig but a target of CTF Ig.
4.3.9 Brain Morphologic Observations
CTF was detected in both intra and extra cellular areas with either antibodies and all types of Fc blocking. No differences in right or left cerebral or cerebellum were observed. The brain tissues were TACE negative and IDE positive therefore not a source of CTF-Ig but a target tissue. CTF R1 free fragments (mAb 461) were nega-tive for free CTF. CTF-Ig (mAb 444) was detected in interstitial areas surrounding the tissue especially around endothelial. WBC migration into the interstitial CSF space was detect, especially in areas which had some infl ammation lesions. These interstitial WBC cells and areas were positive for IgG and CTF co-location. The brain tissues were negative for IgG inside the cells.
4.3.10 Tissue Absorption in Disease
Staining of tissue from a diabetic rat model showed an increase in CTF absorption as rats became diabetic ( See Chap. 5 for description of animal moedels ). The increase occurred in the liver, brain and lymph nodes ( See Table 4.4 ; Fig. 4.4 ). Absorption into lymph nodes occurred earlier while liver absorption continued to increase with length of diabetes. Cell binding of CTF was clearly visible in the lymph nodes. Diseased liver tissues contained immune cells with CTF such as mac-rophages and monocytes.
4 Protease Inhibition and Biological Distribution of the C Terminal Fragment
88
The impact of an immune response to CTF was shown using rabbits which did and did not produce antibodies to CTF. The immune response greatly increased CTF absorption into the liver ( See Table 4.5 ). Circulating IgG immune complexes are primarily eliminated by the liver (Datta-Mannan et al. 2007 ). The serum half- life of IgG is ~330 h and absorption into endothelial cells is regulated by the Fc receptor (Johansson et al. 1996 ). Antibodies binding CTF would bring CTF into the liver. The presence of CTF in liver of diabetic animals is therefore more likely due to natural degradation of immunoglobulin. No antibodies to CTF were observed in rat models or human patients.
4.4 Conclusion
This data that supports that CTF could contribute to inhibition of insulin degrada-tion through IDE. The 32-mer CTF peptide was less soluble due to the N- terminal hydrophilic peptides. The iso-electric point of 4.1, net charge at pH 7.0 of −3 and average hydrophilicity of −0.4 is favored by the acidic endosome environment. Therefore it was proposed AdipoR1 CTF 344-375 is formed inside the endosome by TACE cleavage allowing CTF regulation of insulin concentration by inhibiting insulin degradation by IDE ( See Fig . 4.5 ).
This could prevent IDE from lowering intracellular insulin. Higher insulin would be expected to decrease insulin receptor sensitivity to bind plasma insulin ( See Fig . 4.5 ). At the same time TACE activity is needed for CTF formation. Competitive binding of CTF to TACE could reduce free TNF-α formation and decrease its infl ammatory effects.
In abnormal patients such as long term diabetics and chronic infl ammatory dis-eases, AdipoR1 CTF formed in tissue and blood is expected to inhibit insulin deg-radation. The result could be CTF build up in liver, pancreas and other tissue causing a reduced insulin response. The CTF absorption would be expected to increase as diabetes progresses. Additionally AdipoR1 CTF is carried on antibodies and carried into cells to tissue. It is also possible this mechanism is used to alter the immune cell response in the lymphnodes. Any resultant increase in cellular insulin would cause reduced insulin receptor response (Kahn et al. 1973 ).
Table 4.4 CTF in tissue by immunuo fl uorescent staining
CTF Normal rats N = 3
Short term diabetics rats N = 3
Long term diabetic rats N = 3
Liver Neg Pos 2+, Neg. Neg.
Pos Pos 5+, Pos 4+, Pos 3+
Pos Pos 6+, Pos 7+, Pos 5+
Lymph Neg Pos 5+, Neg, Neg.
Pos Pos 10+, Pos 11+, Pos 8+
Pos Pos 5+, Pos 11+, Pos 5+
Staining of tissue from a diabetic rat model showed CTF was absorped into tissues as rats became more diabetic over 16 weeks. Staining was graded on scale to 1–40
M. Pugia and R. Ma
89
Table 4.5 CTF in tissue by immunuo fl uorescent staining
CTF in liver CTF in brain CTF in pancreas
Non producers N = 3
Pos Pos 5+, Pos 6+, Pos 5+
Neg Pos 1+, Neg, Neg
Pos Pos 1+, Pos 1+, Pos 1+
Producers N = 3
Strongly pos Pos 35+, Pos 40+, Pos 30+
Pos Pos 10+, Pos 11+, Pos 8+
Pos Pos 8+, Pos 12+, Pos 8+
Staining of tissue from a rabbits with and without producing CTF polyclonal antibodies showed CTF was absorption into tissues. All rabbits were exposed to CTF antigen for 3 months and some liver absorption occurred with our immune response. Staining was graded on scale to 40. Example images shown above, Blue is DAPI stain showing nuclei and green is CTF
Lymph nodesLiver
Normal
Early Diabetic
LateDiabetics
Fig. 4.4 Tissue staining for Adiponectin receptor C terminal fragment (AdipoR CTF) . Absorption of AdipoR CTF into tissue as with diabetic disease progression is shown. Images are of IHC stain-ing of tissue from diabetic rats which were normal or diabetic for short or long periods of time. Blue color is DAPI staining showing nuclei and green color is AdipoR CTF from antibody to AdipoR CTF labeled with green fl uorescent dye. Images shown are at 400× magnifi cation. Frames show CTF was absorption into tissues as rats became more diabetic over 16 weeks
4 Protease Inhibition and Biological Distribution of the C Terminal Fragment
90
The CTF compound can be useful for increasing insulin (proliferation). The assays can be useful for diabetic drug screening, for screening of compounds to block CTF absorption into tissue (liver, pancreas, brain) and formation (adipose, muscle and lymph nodes). The IHC and ICC methods are useful for study of CTF on immune cells (from blood and lymph nodes) and in tissues for study of immune activation and tissue absorption in disease. A recent study showed a contradictive result of effect of IDE inhibition by synthetic inhibitors on the oral glucose toler-ance and glucose injection (Maianti et al. 2014 ), which may suggest the inhibition of IDE activity can result in an complicated insulin desensitizing effect worth fur-ther story.
Acknowledgments We would like to acknowledge the work done by Siemens Healthcare Diagnostic development people including Stoney Jackson, Karen Marfurt and others who helped with animal tissues harvesting. We are grateful to Dr Carl Marfurt of Indiana University School of Medicine-South Bend for his help on methods for tissue preparation. We thank Boston University’s Zlab for the generation of the TASSER structures for AdipoR1 CTF 344-375 and AdipoR2 CTF 344-375
Insulin Receptorresponse
Endosome
IDE
CTF
AdipoR1
Adiponectin
Pro-TNF-α
TNF-α
Insulin DegradingEnzyme (IDE)
IDE
Insulin
TACE
Fig. 4.5 Proposed role of Adiponectin receptor C terminal fragment (AdipoR CTF) in insulin receptor response. The proposed mechanism of AdipoR CTF formation and function in regulating insulin resistance is shown. The cleavage of CTF by ADAM17/TACE in the endosome is proposed. Intercellular CTF would inhibit insulin degradation through insulin degradation enzyme (IDE) and increase cellular insulin levels. Increased intercellular CTF and insulin are observed and proposed to reduce insulin receptor response. The inhibition of TACE by interaction with AdipoR CTF could also inhibit TNF-α release. Adiponectin (Adpn) is normally present due to fat cell signaling. Adpn binds to AdipoR potentially modulating CTF release by TACE with other unknown factors. Adiponectin secretion is suppressed in diabetes and obesity, as is CTF whereas tissue levels of CTF increase. Tissues with CTF inhibiting (IDE) would be less responsive to insulin
M. Pugia and R. Ma
91
Supplements
• IHC and ICC Cell Staining Method • Western Blot Methods
References
Abdul-Hay SO, Kang D, McBride M, Li L, Zhao J, Leissring MA (2011) Deletion of insulin- degrading enzyme elicits antipodal, age-dependent effects on glucose and insulin tolerance. PLoS One 6(6):e20818
Black RA (2002) Tumor necrosis factor-alpha converting enzyme. Review. Int J Biochem Cell Biol 34(1):1–5
Bonora E, Zavaroni I, Coscelli C, Butturini U (1983) Decreased hepatic insulin extraction in sub-jects with mild glucose intolerance. Metabolism 32:438–446
Datta-Mannan A et al (2007) Monoclonal antibody clearance impact of modulating the interaction of IgG with the neonatal Fc receptor. J Biol Chem 282(3):1709–1717
Duckworth WC, Bennett RG, Hamel FG (1998) Insulin degradation: progress and potential. Endocr Rev 19(5):608–624
Farris W et al (2003) Insulin-degrading enzyme regulates the levels of insulin, amyloid beta -pro-tein, and the beta-amyloid precursor protein intracellular domain in vivo. Proc Natl Acad Sci U S A 100(7):4162–4167
Gil-Camposa M, Canetea R, Gilb A (2004) Adiponectin, the missing link in insulin resistance and obesity. Clin Nutr 23:963–974
Golz S, Bruggemeier U, Geerts A (2007) Diagnostic and therapeutic for diseases associated with GPCR AdipoR2 and R1. US2007/0053913, US2007/0037226
Hotamisligil GS, Budavari A, Murray D, Spiegelman BM (1994) Reduced tyrosine kinase activity of the insulin receptor in obesity-diabetes. Central role of tumor necrosis factor-alpha. J Clin Invest 94(4):1543–1549
Johansson AG et al (1996) Liver cell uptake and degradation of soluble immunoglobulin G immune complexes in vivo and in vitro in rats. Hepatology 24:169–176
Kahn R, Neville DM, Roth J (1973) Insulin-receptor interaction in the obese-hyperglycemic mouse. A model of insulin resistance. J Biol Chem 248:244–250
Kuo WL, Montag AG, Rosner MA (1993) Insulin degrading enzyme is differentially expressed and developmentally regulated in various rat tissues. Endocrinology 132(2):604–611
Leissring MA et al (2010) Designed inhibitors of insulin-degrading enzyme regulate the catabo-lism and activity of insulin. PLoS One 5(5):e10504
Lorenzo M et al (2008) Insulin resistance induced by tumor necrosis factor-alpha in myocytes and brown adipocytes. J Anim Sci 86(14):E94–E104
Maianti JP et al (2014) Anti-diabetic activity of insulin-degrading enzyme inhibitors mediated by multiple hormones. Nature 511:94–98
Olson RE, Marcin LR (2007) Chapter 3. Secretase inhibitors and modulators for the treatment of Alzheimer’s disease. In: Macor JE (ed) Annual reports in medicinal chemistry, vol 42. Elsevier Inc, Wallingford CT, USA, pp 27–48. ISSN 978-0-12-373912-4
Ouchi N et al (2000) Adiponectin, an adipocyte-derived plasma protein, inhibits endothelial NF-kappaB signaling through a cAMP-dependent pathway. Circulation 102(11):1296–1301
Ramos-Vara JA (2005) Technical aspects of immunohistochemistry. Vet Pathol 42(4):405–426 Rawlings ND, Morton FR, Barrett AJ (2006) MEROPS: the peptidase database. Nucleic Acids Res
34:D270–D272 Sathananthan A (2012) A concerted decline in insulin secretion and action occurs across the spec-
trum of fasting and post challenge glucose concentrations. Clin Endocrinol (Oxf) 76(2):212–219
4 Protease Inhibition and Biological Distribution of the C Terminal Fragment
92
Serino M et al (2007) Mice heterozygous for tumor necrosis factor-alpha converting enzyme are protected from obesity-induced insulin resistance and diabetes. Diabetes 56(10):2541–2546
Tamboli IY et al (2010) Statins promote the degradation of extracellular amyloid beta peptide by microglia via stimulation of exosome-associated insulin-degrading enzyme (IDE) secretion. J Biol Chem 285(48):37405–37414
Tura A, Ludvik B, Nolan JJ, Pacini G, Thomaseth K (2001) Insulin and C-peptide secretion and kinetics in humans: direct and model based measurements during OGTT. Am J Physiol Endocrinol Metab 281:E966–E974
Weickert MO, Pfeiffer AF (2006) Signaling mechanisms linking hepatic glucose and lipid metabo-lism. Diabetologia 49:1732–1741
M. Pugia and R. Ma
93© Springer International Publishing Switzerland 2015 M. Pugia (ed.), Infl ammatory Pathways in Diabetes: Biomarkers and Clinical Correlates, Progress in Infl ammation Research, DOI 10.1007/978-3-319-21927-1_5
Chapter 5 Cell and Biological Models for the C Terminal Fragment of Adiponectin Receptor
Michael Pugia and Rui Ma
Abstract The impact of the peptides such as 32 amino acid peptide AdipoR1 C -terminal fragment (CTF) (AdipoR CTF 344-374 ) can be studied in cell and animal models provided the conditions are defi ned for the biological system tested. In this work, calcium depleted cardiomyocytes (C2C12) were found useful as cell model for study of Ig-CTF formation and release. This cell model could be a useful tool for screening of compouns to block formation of CTF. A breast cancer cell model (SKBR-3) with labeled insulin was found useful for measurement of cellular load-ing and inhibition of insulin degradation. This cell model could be a useful tool for screening of compounds to block CTF inhibition of IDE. Leptin resistant rat models were shown to become hyper-insulinemic when treated with CTF allowing a model for inducing insulin resistance. The progression of leptin resistant rats to diabetes was measure-able by their oral glucose tolerance and plasma CTF levels. This ani-mal model supported that AdipoR CTF raises intracellular plasma insulin but not glucose and is an important factor in diabetic progression. Together these models allow exploring formation and signaling of new inhibitory plasma peptides and demonstrate how to explore bioactivity of proteomic discoveries with cell and ani-mal models.
5.1 Introduction
Metabolic syndrome is a collection of conditions including glucose intolerance, obesity, hypertension, hypertriglyceridemia, and hypercholesterolemia that precede diabetes mellitus and cardiovascular disease (Shin et al. 2013 ). The ability of insulin to signal glucose transport by the glucose transporter (GULT1/4) is reduced.
M. Pugia , PhD (*) Strategic Innovation , Siemens Healthcare Diagnostics , 3400 Middlebury Street , Elkhart , IN 46515 , USA e-mail: [email protected]
R. Ma Siemens Healthcare , 278 Zhouzhu Road , Pu Dong New Area 201318 , People’s Republic of China
94
Reduced insulin response impairs the mitochondrial metabolism of glucose and cell survival though the phosphatidylinositol 3-kinase (PI3K), protein kinase B (AKT) and extracellular-signal-regulated kinase (ERK) pathways (Chang et al. 2004 ; Dugani and Klip 2006 ; Copps and White 2012 ; Siddle 2011 ).
Exercise increases mitochondrial metabolism of glucose generating fuel-depleted cells with increased intracellular calcium levels (Ojuka 2004 ; Hardie 2004 ). Glucose transport by GULT1/4 is activated during exercise via the AMP-activated protein kinase (AMPK) (Jessen and Goodyear 2005 ). Activation of AMPK also increases fatty acid oxidation by inactivating acetyl-CoA carboxylase (ACC) and lowering malonyl-CoA content (Hayashi et al. 2000 ). Adipokines such as leptin and adipo-nectin are key insulin-independent activators of AMPK (Carling 2005 ; Yamauchi et al. 2002 ). Lipid loading during obesity causes hyperplastic-hypertrophic fat pro-ducing less adipokines and lessening AMPK activation.
Cardiomyocytes have become important cellular models for the study of AMPK activation (An 2006 ). Cellular growth rates and kinase activation are carefully mon-itored while the cultures are exposed to various stimulates. Adiponectin activates cardiomyocyte AMPK, but inhibits cell proliferation by suppressing extracellular-signal- regulated kinase (ERK) activation and inducing apoptosis (Wang et al. 2005 ; Shibata et al. 2004 ; Bråkenhielm et al. 2004 ).
Metabolic syndrome can also be studied by animal models that develop disease prematurally in life (de Artiñano and Miguel Castro 2009 ). This allows rapid assess-ments of new markers, pathways and treatments. The degree of metabolic syndrome worsens through the life of animal with increased hyperglycemia, hyperinsulinemia, weight gain and glucose intolerance (Schmidt 2003 ). Rats resistant to leptin are one of the most widely used metabolic syndrome model. Breeds such as ZDF, Zucker, BB/ZDR, JCR, SHHF, SHROB and ZSF were created by classical breeding meth-ods and all shown to carry a defective leptin receptor with a gene mutation in fa allele (Clark et al. 1983 ; Kava et al. 1990 ; Tirabassi et al. 2004 ).
Leptin resistance animals become obese and spontaneous develop diabetes, met-abolic syndrome, nephropathy, neuropathy, hepatic steatosis and congestive heart failure at an age of 3 months (An 2006 ; Schmidt 2003 ; Rodrigues 2006 ). Leptin is synthesized in adipose cells inducing satiety and weight loss by acting on leptin receptors in the hypothalamus and central nervous system (Gorden and Gavrilovay 2003 ). Leptin replacement improves hyperglycemia, hyper-insulinemia, and tri-glyceride levels reducing hyperplastic-hypertrophic fatty in adipocytes.
Leptin resistance rat models have been used to show the benefi ts of many new treatments pathways. The value of incretines, a major pharmacological agent for diabetic care, was fi rst shown in the ZDF rat model (Doyle and Egan 2007 ). Incretines improved glucose tolerance through GLP-1 receptor signaling mediated by increased β cell proliferation and decreased apoptosis allowing islet regeneration and include gastric inhibitory peptide (GIP), glucagon-like peptide-I (GLP –I) and pituitary adenyl-cyclase-activating peptide. Dipeptidyl peptidase IV (DPPIV) inhibitors preventing incretine degradation have similarly been studied. Nesfatin for appetite control (Shimizu et al. 2009 ), antagonist of the cannabinoid CB1 receptor (Pavón et al. 2008 ), and the impact of free acid amide hydrolase FAAH activity (Cable et al. 2014 ) have been demonstration in the ZDF rat model.
M. Pugia and R. Ma
95
The goal of any new metabolic syndrome treatment is the improvement of energy metabolism resulting in increased glucose uptake, glycolysis, pyruvate oxidation and fatty acid consumption. Our goal was to investigate the impact of the AdipoR1 C -terminal fragment (CTF) to determine potential benefi ts. Therefore we explored the impact of CTF in cardiomyocytes and cancer cell lines by measuring formation of Ig-CTF, intracellular calcium and insulin, and cellular proliferation. We also mea-sured the impact of CTF on plasma insulin and glucose tolerance in a leptin resistant rat model and measured the plasma CTF with progression to a diabetic state.
5.2 Methods
5.2.1 Cell Culture Methods and Bioassays
The materials and bioassay used for AdipoR CTF were as described in previous chapter. Globular Adpn (16.8 kDa) was obtained from Alpco Diagnostics (Salem NH) and confi rmed to be 14–18 kDa by western blot. Full length Adpn (25.4 kDa) was purchases a human recombinant protein from BioVendor, LLC (Candler, NC) and R&D Systems (Minneapolis MN) and confi rmed to be 26–36 kDa by western blot. As a result of glycosylation, the recombinant protein can migrate as high as approximately 36 kDa band in SDS-PAGE under reducing conditions. The low, middle and high molecular weight oligomeric forms of Adpn activate different signaling pathways through the full length and globular peptide domain. Both Adpn forms were tested. Bikunin was obtained from Scipac Ltd (Sittingbourne, Kent UK).
5.2.2 Cell Lines
Cell culture mouse myocytes (C2C12 CRL1772) and human cancer cells (SKBR-3) (American Type Cell Culture, ATCC, Manassa VA) were maintained in culture. IMDM + 10 % FBS growth media was used for SKBR-3, while C2C12 were main-tained with Dulbeccor Modifi ed Eagle’s medium (MEM), glutamine and 10 % fetal bovine serum. Serum contains native insulin and adiponectin. Cells were grown at 37 °C in an atmosphere of 95 % air and 5 % CO 2 and propagated by splitting with one to four dilutions using PBS/EDTA at cell growth.
5.2.3 Cell Culture Treatment
Myocytes are then grown in Modifi ed Eagle Medium (MEM) supplemented with 2 mM glutamine in addition to penicillin (50 units/ml), streptomycin (50 μg/ml) and 10 % fetal bovine serum in an atmosphere of 95 % air and 5 % CO 2 . The cells were
5 Cell and Biological Models for the C Terminal Fragment of Adiponectin Receptor
96
propagated by sub-culturing at their confl uence (4–5 days) in one to four dilutions using either trypsin or PBS/EDTA. Cell confl uence was around fi ve million cells per T-75 cm 2 fl ask. Semi-confl uent cell cultures were synchronized twice using 0.5 mM hydroxyurea (sterile) in MEM, containing 2 mM glutamine without serum, for two 24 h period. After synchronization the hydroxyurea was removed, cells were har-vested asceptically using trypin- EDTA, counted, and replenished with serum free MEM at 0.3 × 10 6 cells/ml into either multiple T-25 cm 2 fl asks or by adding 100 μL to separate micro-titre-plate wells.
The cells are grown for another 24 h serum free MEM followed by induction with various concentrations of Bik (0, 20, 40 μg/mL), Adpn (0, 20, 40 μg/L), CTF 25 mer (0, 20, 40 μg/L) and combination in serum free MEM under sterile condition. The extracellular Ca 2+ concentration is raised to 0.5 mM with calcium chloride and the cells were allowed to grow additional 24 h before harvesting using PBS/EDTA. The cells were harvested with trypsin, and centrifuged for 5 min @ 2000 rpm followed by decanting supernanant and adding small amount (~200 μL) media. The cells are sus-pended in 1:3 ratio with PBS after saving an aliquot for cell counts and later analysis.
5.2.4 ELISA and Western Blot Analysis
ELISA for Adpn was performed using monoclonal antibody (mAb) directed against the globular domain in kits from B-Bridge International Inc (Mountain View, CA). CTF and CTF-Ig were measured as previously described. Analysis for p38 MAPK was performed using rabbit monoclonal antibodies from Cell Signaling Technology Inc (Danvers MA) by western blot method. Measurements of the relative increase in ERK phosphoralyation were preformed according to the ELISA manufacturer’s instruction (Assay Designs, Ann Arbor MI).
5.2.5 Intra-cellular Calcium Analysis
Measurements of cytosolic calcium were done by infusing cells with the membrane permeable calcium sensitive Fluo-4 AM dye (Molecular Probes, Eugene, OR). Packed viable cells (10 6 cell/μL) in 100 μL PBS were treated with 5 μL of fl uo-4, AM dye solution. The dye solution was made by dissolving 50 μg of Fluo-4 AM into 100 μL of a 20 % solution Pluronic® F-127 in DMSO. Cells were incubated for 30 min at 37 °C to allow permeabilization of the dye. Cells were centrifuged for 10 min at 7000 rpm (12,000 × g) and pellets resuspended in 100 μL PBS to wash out the extra-cellular fl uorescent dye. Intracellular proteases hydrolyzed Fluo-4 AM which caused fl uorescence in the presence of calcium. A 5 μL sample was analyzed on a glass slide with an attached covers using fl uorescent microscopics.
Fluorescent images were obtained at room temperature using a Leica DM5000 confocal microscope with a 63 1.2 NA objective in a darken room. Phase contrast
M. Pugia and R. Ma
97
images were used to locate regions of intracellular space and measure fl uorescent signals. The Fluo-4, excitation and emission wavelengths were 488 nm and 510–590 nm, respectively. Fluorescent images were collected with a frequency of 0.6–1.0 frame/s for 5 min. Calibration was done by measuring the fl uorescent after mixing 5 μL fl uo-4 dye solution, 5 μL of 1 N NaOH with 100 μL calcium calibration solution at 0 or 100 nM calcium chloride.
5.2.6 Intracellular Insulin Measurements
Human cancer cells (SKBR-3) were grown to confl uency. A CTF 25 mer 1 mg/mL stock solution in DMSO was used to make 0.5 mg/mL in 1 % DMSO/media while 0.1 mg/mL insulin-FITC (Sigma Aldrich) in media was made by diluting 1 mg/mL stock with ten-fold with cell growth media. Media was fi ltered through 0.22 μ syringe fi lter unit after addition of CTF or insulin-FITC. Cells were treated with the fresh media. SKBR-3 cells were treated with different concentrations of insulin- FITC (0, 1, 10 μg/mL), CTF 25 mer (0, 50 μg/L) and with the combination added under sterile condition. The cells were incubated for ~2 days @37C, 5 % CO 2 mix cells in microtiter plate shaker on low speed. Cultures were split into triplicates by placing 0.4 mL into separate wells of a Lab-Tek II 4-well chamber slide (Sigma Aldrich). All slides were measured at 1, 2, 3 days for uptake of insulin-FITC after washing with 2 × 1 mL PBS. Fluorescent images were obtained at room temperature using a Leica DM5000 and the L5 fi lter set for FITC (excitation 494 nm/emission 518 nm). The slides are scanned at 40 × using 50–300 ms exposures.
5.2.7 Cell Proliferation Counts
A 10 μl cell suspension was added to 80 ul PBS and 10 ul Trypan blue (1 %) and 4 squares counted on the slide in hemacytometer (area: 0.1 cm × 0.1 cm × 0.01 cm). Count all nine chambers and repeat the count with second dilution in trypan blue calculation: total # cells in 9 chambers × 4 × 10,000/9 = # cells per mL, Typically cells were grown to 5.5 × 10 5 cells/mL.
5.2.8 Animal Models
Animal model studies were performed at PreClinOmics, Inc. (Indianapolis, IN). Normal and diabetic rats were injected with CTF as part of a glucose tolerance chal-lenge. The concentrations of glucose, insulin, Adpn, and free CTF were measured pre- dose and at 30, 60, 90, and 120 min after injection. Normal Hsd:SD rats were divided into two groups (n = 7 as controls and n = 8 for CTF vehicle). Diabetic ZDF rats (Charles
5 Cell and Biological Models for the C Terminal Fragment of Adiponectin Receptor
98
River Laboratories) (n = 12) were also divided into two groups (n = 6 for each control or CTF vehicle). The disease rats were at 438–593 g and on diabetic diet initiated at 25–26 weeks age. The animals were fasted for 16 h overnight before injection with CTF and glucose. Injection vehicle was prepared by weighing 2–10 mg of AdipoR1 CTF341-370 (30 mer) or AdipoR1 CTF344-375 (32 mer) as compound into 5 ml ster-ile saline with 0.4 g/ml (2 g/kg) glucose with 1 % albumin (Sigma, Lot #: 067K0136).
5.2.9 Disease Progression Study
Groups of non-diabetic (n = 6), short term diabetic (n = 6), and long term diabetic (n = 6) were determined by glucose tolerance. Rats (ZDSD) used were at 438–593 g and on diabetic diet imitated at 25–26 weeks age. At time of study animals were dosed with 2 g/kg glucose (Sigma, Lot #: 067K0136), p.o. following a 16 h overnight fast. Whole blood was collected from the tail vein at pre-dose, 60 and 120 min following glucose chal-lenge. The glucose tolerance curve in the OGTT varied depending on the age, strain, degree of diabetes and other test conditions. The glucose typically peaks at 30 min at 175–200 mg/dl in a “normal” rat. Insulin will typically peak somewhat later at several times the baseline level. Observation of the test subject occurred once daily during the accumulation and study period. Body weight was measured on Day 0 for sorting and on Day 1 to determine dose volumes. Food consumption was not measured.
5.2.10 Biomarker Measurements
Whole blood was collected from the tail vein at 0, 30, 60, 90, and 120 min (150 μL for each time point into Li-Hep) following glucose challenge. Whole blood glucose was measured by Beckman CX4. HbA1c was analyzed on an AU480 clinical chem-istry analyzer. Insulin was analyzed on serum samples by MSD (K112BZC) or ELISA (Alpco, Salem, NH). A sample of 50 μL of whole blood was collected and immediately plated into a 15 ml conical tube and diluted with 3.6 mL TBS and fro-zen at −20 °C until measurement of free CTF by ELISA (See previous section). An additional 100 μL of whole blood was collected for the measurements of adiponec-tin (Adpn) levels were measured with a rat ELISA kit from Alpco (Salem, NH).
5.3 Results
5.3.1 Cardiomyocyte Models
Cardiomyocytes (C2C12) produced Ig-CTF when grown in culture with fetal bovine serum containing immunoglobulins. Western blot analysis detected IgM-CTF, Fab2-CTF, Ig-CTF and AdipoR1 in cell membranes lysates ( See Table 4.3 of Chap. 4
M. Pugia and R. Ma
99
for western blots data ). Ig-CTF remained membrane bound. Cardiomyocytes did not release any detectable free CTF or Ig-CTF into culture media measured by ELISA.
The causes or inhibition of Ig CTF formation and release was examined by inducing cardiomyocytes culture media with additional biochemical factors. Adpn was present in fetal bovine serum culture media but was not observed in washed cell membranes. Treating cell cultures with additional Adpn did drive receptor binding and Adpn was observed in washed cells membranes. This additional Adpn did not increase or inhibit Ig-CTF formation or cause CTF release from the membrane. Binding studies showed neither full-length nor globular Adpn bound directly to the CTF of the Adiponectin receptor.
Intracellular calcium (Ca 2+ ) was also measured on cardiomyocytes using a Ca 2+ sensitive dye (Fluo-4 AM) after exposure of viable cells to media containing Ca 2+ and washing to remove extra-cellular and unabsorbed Ca 2+ (See Table 5.1 ). Green fl uorescence from Fluo-4 AM was observed inside the cell indicating the presence of calcium and membrane permeation. Microscopic analysis of the phase contrast images and fl uorescence images was used to determine the amount and intensity of fl uorescence inside the intracellular area of the cell in the range of 0.5–100 mM. Calibration solutions were used to quantitate the intracellular Ca 2+ con-centrations. Cells were treated with CTF and Apdn to determine impact.
Cardiomyocytes (C2C12) cultures typically maintain intra-cellular Ca 2+ in an 80–90 nM constant range. Intra-cellular Ca 2+ decreased to >5 nM with when cells were exposed to 20 μg/mL bikunin (Bik). The addition of bikunin (Bik) served as a means to strip cells of intra-cellular calcium and allowing means to diminished Ca 2+ to zero base line for measuring increases in viable cells (See Chap. 12 for further information). Cells lysates showed Bik was maintained on cell surfaces during this treatment. Cells were further treated with Adpn or CTF 25 mer in presence and absence of Bik.
The addition of Adpn to Ca 2+ depleted cell to restored intra-cellular Ca 2+ to 91 nM (Table 5.1 ). Adpn binding of AdipoR signals activation of AMPK which regen-erates the ATP major pathway and promotes the cytosolic localization of liver kinase B and a minor pathway (the phospholipase Ca 2+ /calmodulin-dependent pro-tein kinase kinase-dependent pathway) that stimulates Ca 2+ release from intracel-lular stores restoring cytosolic calcium (Jessen and Goodyear 2005 ; Zhou et al. 2009 ). Thus Ca 2+ restoration is a good measure of Adpn activation of AMPK. The addition of AdipoR1 357-375 CTF had no impact on intra-cellular Ca 2+ whether cells were depleted by Bik or loaded with calcium (See Table 5.1 ). The Adpn restoration is useful tool to determine that receptor function is not impaired during CTF for-mation and release.
Over all, the mechanism of Ig-CTF formation and release remains unclear. However calcium depleted cardiomyocytes cell model do appear useful for study of enzymes and proteins required for Ig-CTF formation and release. As a pathway tool, cardiomyocytes could be grown with enzyme inhibitors and blockers in serum free media to examine factors required for Ig-CTF formation and release. Cardiomyocytes also allow monitoring Adpn activation of AMBK via normal adiponectin receptor function.
5 Cell and Biological Models for the C Terminal Fragment of Adiponectin Receptor
100
5.3.2 Cellular Insulin Uptake
Breast cancer cells (SKBR-3) did not contain any detectable CTF or AdipoR1 but did contain IDE and TACE (See on-line supplement for cell and tissue staining procedure at www.extras.springer.com/2015/978-3-319-21926-4 ). As CTF is absent in the cell but the target downstream protease IDE are present, SKBR-3 makes a good model for measuring the impact of CTF on intra-cellular insulin. Treatment of SKBR-3 with CTF increases the amount of intercellular insulin ( See Fig . 5.1 ). The number of insulin loaded cells double when exposed to 10 ug/mL of CTF. The insu-lin loaded cells clearly showed intercellular fl uorescence from the labeled insulin. The insulin loading cell model allows screening of compound to block CTF inhibi-tion of IDE.
Cells treated with CTF maintain viability (proliferation) with no observable cell death (apoptosis) ( See Fig . 5.1 ). Cardiomyocytes exposed to CTF also did not modu-late phosphorylation of ERK and MAPK-p38 or slow proliferation ( See Table 5.1 ). No observable cell death was noted. As insulin signaling of the PI3K/AKT/mTOR path-way increases proliferation and decreases apoptosis, insulin loading of cells would be expected to have reduce insulin receptor response and proliferation. However no
Table 5.1 Intra-cellular signals after exposure to CTF and Adpn
Protein and Exposure level (μg/mL) a Intra-cellular signals
Bik Adpn AdipoR1 357-375 Calcium (nM) b
MAPK p38 c (% change)
MAPK ERK c (% change)
0 0 0 93 8 % 0 % 0 20 0 82 −14 % 7 % 0 40 0 Not measured 10 −33 %
20 0 0 4 −17 % −5 % 0 0 20 84 0 % 5 % 0 0 40 Not measured 13 % 6 % 0 20 20 Not measured 19 % 7 % 0 40 40 Not measured −14 % −38 %
20 20 20 72 nm nm 20 0 20 9 −13 % 6 % 20 20 0 91 −10 % −14 %
a Synchronized cultures of myocyte (C2C12) were incubated 5 h with Bikunin (Bik), Adiponectin (ADPN) and/or AdipoR1 357-375 at 0, 20, or 40 μg/mL. Cells were harvested for western blots and whole cell used for intra-cellular calcium measurement b Intra-cellular calcium was measured by incubating with cell the membrane permeable calcium sensitive Fluo-4 AM for 30 min at 36 °C to allow intracellular hydrolysis to cause fl uorescence in the presence of calcium. Cells were washed and centrifuged to remove extra-cellular dye and fl uo-rescent images captured c Measurements of the relative change in phosphorylation of MAPK ERK by ELISA and MAPK p38 by Western Blot were measured at 0, 6, 12 and 24 h after cells were revived in media contain-
ing serum
M. Pugia and R. Ma
101
impact to proliferation observed. It appears that CTF does not inhibit PI3K/AKT/mTOR intercellular signaling. This could be due to the excess of insulin used in cell culture media. An insulin straved model was not done. This proliferation effect con-trasts the apoptotic impact of Adpn, which did suppress phosphorylation of ERK in cardiomyocytes ( See Table 5.1 ) and restored intra-cellular Ca 2+ . Restored intra-cellular Ca 2 would inhibit calcium dependent PI3K pathway and desensitize insulin response.
5.3.3 Impact on In-Vivo Insulin
Normal and diabetic rats were injected with CTF as part of a glucose tolerance chal-lenge ( See Tables 5.2 – 5.4 and Fig . 5.2 ). Normal and diabetic rats showed different behaviors. In normal rat showed substantial plasma insulin increases compared to control ( See top two graphs of Fig . 5.2 ). Glucose was not reduced and insulin levels in plasma doubled in normal rats. This behavior matched an insulin resistance behavior. Diabetic rats which already had with impair insulin production (<10 % plasma insulin levels of normals) did not exhibit an increase in plasma insulin upon
Control (no CTF)
+0.5mg/ml CTF
Effect of CTF on Insulin Uptake
0
20
40
60
80
100
120
1 ug/mL
insulin, day 1
10 ug/mL
insulin,day 1
10 ug/mL
insulin, day 2
insu
lin u
pta
ke (
# ce
lls)
Effect of CTF on Cell Viability
0
20
40
60
80
100
1ug/mL insulin 10ug/mL insulince
ll vi
ablil
ity
(%)
(no CTF) (CTF)
Fig. 5.1 Effect of CTF-25 mer on insulin uptake and cell viability in SKBR3 cell model. Breast cancer cells (SKBR) did not contain any detectable CTF or AdipoR1 but did contain IDE and TACE ( by ICC method ). This makes SKBR a good model for measuring the impact of CTF on cellular insulin degradation. Treatment of SKBR with CTF doubled the cellular labeled insulin. Cells treated with CTF maintain viability (proliferation) with no observable cell death (apoptosis). Insulin loading was demonstrated by fl uorescence inside cells due to label on insulin
5 Cell and Biological Models for the C Terminal Fragment of Adiponectin Receptor
102
Tabl
e 5.
2 G
luco
se, i
nsul
in, H
bA1c
and
CT
F re
spon
se in
nor
mal
and
dia
betic
rat
s: n
on d
iabe
tic r
at g
roup
Ani
mal
#
wee
k
Ani
mal
ID
B
ody
wei
ght (
g)
HbA
1c %
Glu
cose
(m
g/dl
) In
sulin
(pg
/mL
) C
TF
(pg/
mL
)
Min
utes
aft
er d
ose
– pr
e 60
12
0 pr
e 60
12
0 pr
e 60
12
0
ZD
SD r
at 1
0
8150
0400
1 54
4 5.
25
105.
0 26
3.0
164.
0 59
5.6
1086
98
4.4
1.3
113.
8 28
5.7
ZD
SD r
at 2
0
8150
0400
2 42
8 4.
43
101.
0 24
6.0
172.
0 57
5.4
874.
5 96
3.4
41.8
26
1.7
110.
0 Z
DSD
rat
3
0 81
5004
003
544
4.74
11
0.0
236.
0 18
0.0
868.
4 13
07
1167
67
.0
319.
7 18
1.3
ZD
SD r
at 4
0
8150
0400
4 53
1 5.
33
105.
0 25
7.0
234.
0 49
0.6
770.
1 10
93
78.0
35
0.5
139.
8 Z
DSD
rat
5
0 81
5004
005
483
5.17
99
.0
304.
0 25
0.0
595.
1 80
8.4
876.
4 11
1.8
184.
0 74
.8
ZD
SD r
at 6
0
8150
0400
6 55
2 5.
50
108.
0 26
1.0
176.
0 55
9.9
791.
9 12
22.
28.5
18
2.0
135.
5 –
– M
ean
513.
67
5.07
10
4.7
261.
2 19
6.0
614.
2 93
9.5
1051
. 54
.7
235.
3 15
4.5
– –
St. D
ev .
48.7
6 0.
40
4.1
23.3
36
.4
130.
5 21
3.6
132.
1 39
.1
90.9
73
.3
– –
SEM
19
.91
0.16
1.
7 9.
5 14
.9
53.3
87
.2
53.9
16
.0
37.1
29
.9
M. Pugia and R. Ma
103
CTF exposure ( See bottom two graphs of Fig . 5.2 ). It was further shown that dia-betic rats post treatment had CTF loading in liver, pancreas, brain and lymph nodes where as normal rats did not ( See Chap. 4 ).
The levels of CTF in the animals were monitored during this study. Plasma free CTF was signifi cantly (P < 0.003) lower in diabetic rats (63.7 ± 3.3 ng/mL) than in normal rats (98.8 ± 8.1 ng/mL) prior to the glucose tolerance challenge. In contrast, no difference in plasma adiponectin was detected in normal (24.1 ± 3.7 μg/mL) or dia-betic rats (20.9 ± 3.9 μg/mL) prior to the glucose tolerance challenge. Once the animal was injected with CTF, the plasma CTF levels rose 20–35 ng/mL over 120 min. This level of CTF was limited by the insolubility of the CTF peptides used. Unfortunately, neither CTF 30 mer or 32 mer compounds tested in this study were soluble forms. Comparison to a more soluble CTF 25 mer form is expected to allow higher plasma CTF levels cause additional insulin loading in tissue and plasma. High activity of CTF 25 mer could also signifi cantly increase IDE inhibition response.
AdipoR1 CTF344-375 32 mer(2mg/kg)
AdipoR1 CTF341-370 30 mer(2mg/kg)
0%
100%
200%
0 20 40 60 80 100 120Minutes
GlucoseInsulin
0%
100%
200%
0 20 40 60 80 100 120Minutes
100%
of
Co
ntr
ol
GlucoseInsulin
0%
100%
0 20 40 60 80 100 120
Minutes
100%
of
Co
ntr
ol
GlucoseInsulin
0%
100%
0 20 40 60 80 100 120Minutes
GlucoseInsulin
No
rmal
rat
s
100%
of
Co
ntr
ol
Dia
bet
ic r
ats
100%
of
Co
ntr
ol
Fig. 5.2 Glucose and insulin response in disease. A study was conducted by dosing animal ZDF diabetic rats and HD normal rats with Adiponectin Receptor C- Terminal fragment (CTF 32 mer and CTF 30 mer). Normal and diabetic rats were injected +/− CTF just prior to a glucose tolerance challenge. The concentrations of glucose, insulin, Adpn, and free CTF were measured pre-dose and at 30, 60, 90, and 120 min after injection. Adpn did not signifi cant change with time or diabe-tes. CTF was signifi cantly lower in diabetic rats (63.7 ± 3.3 ng/mL) than in normal rats (98.8 ± 8.1 ng/mL) and rose 20–35 ng/mL over 120 min with CTF injection. Substantial insulin increases occurred for normal rats ( top two graphs) but glucose was not reduced and insulin resistance behavior occurred. Diabetic rats only produced <10 % of insulin of normal rat and did not show substantial insulin increases when dosed with CTF
5 Cell and Biological Models for the C Terminal Fragment of Adiponectin Receptor
104
5.3.4 Disease Progression Study
A study was conducted with ZDF rats on a high fat diet separated into groups of non-diabetic and rats that progressed to diabetes ( See Tables 5.2 – 5.4 ). Groups of rats that were diabetic for a short term where further separated from those who were diabetic for a longer term. Blood glucose, insulin, HbA1c and free CTF were mea-sure before and during a glucose tolerance test. Blood glucose and HbA1c values increased as rats became diabetic ( See Tables 5.3 and 5.4 ). Insulin secretion decreased in longer term diabetes which also showed an impaired glucose tolerance.
An increase in free CTF in blood was observed in early and late stage diabetic rats. Normal rats had near zero values of free CTF until glucose challenge. Free CTF in plasma increased during the glucose challenge, peaking at 60 min and still elevated 120 min. Analytical validation showed the free CTF ELISA was suit-able for animal models but not human clinical study. The Ig-CTF ELISA assay was usable for human clinical but not the rat studies. In human studies, Ig-CTF was unchanged during OGTT of 120 min ( See Chap. 6 ). This suggested Ig-CTF is more of a chronic marker while free CTF is more responsive in short term (acute marker). In agreement tissue staining of long term rats from this study showed signifi cant increases in Ig-CTF in the liver, brains, pancreas and lymph nodes when compared to normal and short term diabetics.
5.3.5 Suggested Mechanism of Action
Based on data on hand, it appears that CTF-IgG forms in the muscle by IgG com-ing into proximity of CTF during release from AdipoR1 by TACE ( See Fig . 5.3 ). While all the contributors to this mechanism are not known, it is likely that a reac-tive thioester of CTF is protected until the release is triggered. Here, adiponectin could provide blocking role preventing insulin receptor desensitization by CTF, however inhibition of formation of CTF-IgG has not been demonstrated to date.
It is clear that CTF is carried on antibodies and cells and transported into tissue more so in disease ( See Chap. 4 ). The attraction of Ig-CTF to certain tissue is unknown. One possibility is this mechanism is used to stimulate immune cell response to certain antigens. Another possibility is that absorption is part of immunoglobulin clearance and infl ammatory response. However in any case, CTF Accumulation in tissue correlates with the progress of diabetes conditions. As tissues absorb more Ig-CTF they would be expected to inhibit insulin degrada-tion by IDE. The insulin inside the cell increases and which decreases the insulin receptor response. The result is lower insulin receptor response and higher levels of plasma insulin must be secreted to stimulate the insulin receptor in response to glucose.
M. Pugia and R. Ma
105
Tabl
e 5.
3 G
luco
se, i
nsul
in, H
bA1c
and
CT
F re
spon
se in
nor
mal
and
dia
betic
rat
s: s
hort
term
dia
betic
rat
gro
up
Ani
mal
#
Wee
ks
Ani
mal
ID
B
ody
wei
ght (
g)
HbA
1c
%
Glu
cose
(m
g/dl
) In
sulin
(pg
/mL
) C
TF
(pg/
mL
)
Min
utes
aft
er
dose
–
pre
60
120
pre
60
120
pre
60
120
ZD
SD r
at 1
2
8150
0401
3 53
4 9.
64
202.
0 51
7.0
522.
0 34
5.3
370.
2 25
4.3
173.
5 18
6.8
437.
0 Z
DSD
rat
2
1 81
5004
014
536
7.77
11
9.0
461.
0 34
0.0
879.
7 10
18
835.
7 24
8.5
266.
5 43
1.5
ZD
SD r
at 3
1
8150
0401
5 56
9 6.
98
122.
0 40
0.0
224.
0 46
6.1
786.
2 39
8.3
302.
5 31
2.3
340.
0 Z
DSD
rat
4
1 81
5004
016
550
7.81
10
0.0
409.
0 28
0.0
545.
3 69
3.1
663.
6 83
.0
203.
3 11
1.0
ZD
SD r
at 5
2
8150
0401
7 51
7 8.
29
98.0
43
3.0
322.
0 85
2.2
865.
0 68
2.3
70.0
15
4.0
363.
0 Z
DSD
rat
6
2 81
5004
018
497
9.26
16
5.0
531.
0 39
8.0
558.
8 42
8.8
427.
9 12
4.5
249.
3 19
5.0
– –
Mea
n 53
3.83
7.
46
134.
3 45
8.5
347.
7 60
7.9
693.
6 54
3.7
167.
0 22
8.7
312.
9 –
– St
. Dev
. 25
.07
2.79
41
.0
55.1
10
3.5
214.
0 25
2.2
217.
8 93
.1
58.0
13
2.2
– –
SEM
10
.24
1.14
16
.7
22.5
42
.2
87.4
10
3.0
88.9
38
.0
23.7
54
.0
5 Cell and Biological Models for the C Terminal Fragment of Adiponectin Receptor
106
Tabl
e 5.
4 G
luco
se, i
nsul
in, H
bA1c
and
CT
F re
spon
se in
nor
mal
and
dia
betic
rat
s: lo
ng te
rm d
iabe
tic r
at g
roup
Ani
mal
#
Wee
ks
Ani
mal
ID
B
ody
wei
ght (
g)
HbA
1c
%
Glu
cose
(m
g/dl
) In
sulin
(pg
/mL
) C
TF
(pg/
mL
)
Min
utes
aft
er
dose
–
pre
60
120
pre
60
120
pre
60
120
ZD
SD r
at 1
8
8150
0400
7 46
8 10
.3
306.
0 61
0.0
507.
0 13
8.6
170.
6 19
6.0
143.
3 28
0.5
81.8
Z
DSD
rat
2
9 81
5004
008
394
12.3
51
2.0
678.
0 53
4.0
230.
6 19
2.7
140.
8 96
.0
201.
3 22
7.0
ZD
SD r
at 3
9
8150
0400
9 41
3 11
.9
471.
0 67
2.0
583.
0 25
6.1
212.
9 13
7.7
60.5
18
0.3
79.5
Z
DSD
rat
4
9 81
5004
010
394
11.9
43
1.0
668.
0 58
1.0
137.
5 11
2.9
105.
0 20
8.8
377.
0 51
9.3
ZD
SD r
at 5
9
8150
0401
1 44
0 11
.3
249.
0 60
7.0
499.
0 23
4.7
211.
1 22
6.5
69.5
37
3.0
290.
0 Z
DSD
rat
6
7 81
5004
012
472
10.3
23
4.0
566.
0 52
5.0
314.
1 30
4.5
299.
5 15
4.0
343.
8 15
5.5
– –
Mea
n 43
0.17
11
.4
367.
2 63
3.5
538.
2 21
8.6
200.
8 18
4.2
122.
0 29
2.6
225.
5 –
– St
. Dev
. 35
.18
0.86
11
9.4
45.7
36
.2
69.2
62
.8
71.5
56
.9
86.4
16
5.7
– –
SEM
14
.36
0.35
48
.7
18.7
14
.8
28.2
25
.6
29.2
23
.2
35.3
67
.7
M. Pugia and R. Ma
107
5.4 Conclusion
The results of both cellular and animal studies provided new insights into the role of AdipoR1 C -terminal fragment (CTF). Plasma levels of free CTF decrease with disease progression from non-diabetic to diabetic states. As animal became more diabetic, CTF is cleared from the plasma and absorbed into tissue inhibiting insulin degradation. The result is CTF build up in liver, pancreas, brain, lymph nodes and other tissues lead-ing to a reduced insulin response. Treatment with CTF can be used to cause insulin levels to increase in animal models and human cells. Free CTF (acute marker) increases with glucose over 120 min in a OGTT while Ig-CTF was unchanged (chronic marker)
The leptin resistant rats and cell models described allow determining new aspects of this pathway which could stall the negative impacts of CTF during diabetes. Screening of compound to block CTF inhibition of IDE or loading of with in tissues with CTF appears possible through the animal and cell models described.
Acknowledgements We would like to acknowledge the animal work done by PreClinOmics Inc (Indianapolis, IN) including Dr. Richard Peterson, Dr. WK Yeh, Troy Gobbett and others who conducted animals trials, samples and performed on-site testing. We would like to acknowledge the work done inside Bayer Healthcare Diagnostic on cell models by Dr. Manju Basu, Karen Marfurt and Ronald Sommer.
AdipoR
Antibody binding
CTF-IgG released from musclefrom AdipoR via TACE
CTF transportinto
cell/tissue
CTFAntigenIDEInsulin
Key:
IDE
Insulin
WBC
IgG(blood) Impact:
Glucoseconsumption
Protein Synthesis
Cell Growth
Proteolysis
Insulinreceptor
WBC binding
Ig-CTF absorbed into, liver, brain
pancreas
Cell boundIg-CTF
CirculatingIg-CTF
IDE
Fig. 5.3 Formation of Ig-CTF and impact on insulin response in disease. There are antibodies in the human body with CTF attached. This attachment appears to occur in myoctyes with TACE activity. As tissues absorb more Ig-CTF, they inhibit insulin degradation by IDE. The insulin inside the cell increases and which decreases the insulin receptor response. The result is higher levels of plasma insulin must be secreted to stimulate the insulin receptor. Insulin secretion is typically determined by the level of blood sugar. This mechanism could cause a localized insulin resistance
5 Cell and Biological Models for the C Terminal Fragment of Adiponectin Receptor
108
Supplements
• IHC and ICC Cell Staining Methods • Western Blot Methods
References
An D, Rodrigues B (2006) Role of changes in cardiac metabolism in development of diabetic cardiomyopathy. Am J Physiol Heart Circ Physiol 291(4):H1489–H1506
Bråkenhielm E et al (2004) Adiponectin-induced antiangiogenesis and antitumor activity involve caspase-mediated endothelial cell apoptosis. Proc Natl Acad Sci U S A 101(8):2476–2481
Cable JC et al (2014) The effects of obesity, diabetes and metabolic syndrome on the hydrolytic enzymes of the endocannabinoid system in animal and human adipocytes. Lipids Health Dis 13:43
Carling D (2005) AMP-activated protein kinase: balancing the scales. Biochimie 95:87–91 Chang L, Chiang SH, Saltiel AR (2004) Insulin signaling and the regulation of glucose transport.
Mol Med 10:65–71 Clark J, Palmer CJ, Shaw WN (1983) The diabetic Zucker fatty rat. Proc Soc Exp Biol Med
173:68–75 Copps KD, White MF (2012) Regulation of insulin sensitivity by serine/threonine phosphorylation
of insulin receptor substrate proteins IRS1 and IRS2. Diabetologia 55(10):2565–2582 de Artiñano A, Miguel Castro M (2009) Experimental rat models to study the metabolic syndrome.
Br J Nutr 102(9):1246–1253 Doyle ME, Egan JM (2007) Mechanisms of action of glucagon-like peptide 1 in the pancreas.
Pharmacol Ther 113:546–593 Dugani CB, Klip A (2006) Glucose transporter 4: cycling, compartments and controversies.
EMBO Rep 6(12):1137–1142 Gorden P, Gavrilovay O (2003) The clinical uses of leptin. Curr Opin Pharmacol 3:655–659 Hardie DG (2004) The AMP-activated protein kinase pathway – new players upstream and down-
stream. J Cell Sci 117:5479–5487 Hayashi T et al (2000) metabolic stress and altered glucose transport activation of AMP-activated
protein kinase as a unifying coupling mechanism. Diabetes 49:1–5 Jessen N, Goodyear LJ (2005) Contraction signaling to glucose transport in skeletal muscle. J Appl
Physiol 99:330–337 Kava R, Greenwood MRC, Johnson PR (1990) Zucker (fa/fa) rat. ILAR News 32(3):4–8 Ojuka EO (2004) Role of calcium and AMP kinase in the regulation of mitochondrial biogenesis
and GLUT4 levels in muscle. Proc Nutr Soc 63:275–278 Pavón FJ, Serrano A, Pérez-Valero V, Jagerovic N, Hernández-Folgado L, Bermúdez-Silva FJ,
Macías M, Goya P, de Fonseca FR (2008) Central versus peripheral antagonism of cannabinoid CB1 receptor in obesity: effects of LH-21, a peripherally acting neutral cannabinoid receptor antagonist, in Zucker rats. J Neuroendocrinol 20 Suppl 1:116–123
Schmidt RE (2003) Analysis of the zucker diabetic fatty (ZDF) type 2 diabetic rat model suggests a neurotrophic role for insulin/IGF-1 in diabetic autonomic neuropathy. Am J Pathol 163:21–28
Shibata R et al (2004) Adiponectin-mediated modulation of hypertrophic signals in the heart. Nat Med 10(12):1384–1389
Shimizu H, Oh-I S, Okada S, Mori M (2009) Nesfatin-1: an overview and future clinical applica-tion. Endocr J 56(4):537–543
M. Pugia and R. Ma
109
Shin JA et al (2013) Metabolic syndrome as a predictor of type 2 diabetes, and its clinical interpre-tations and usefulness. J Diabetes Investig 4(4):334–343
Siddle K (2011) Signaling by insulin and IGF receptors: supporting acts and new players. J Mole Endocrinol 47:R1–R10
Tirabassi RS et al (2004) The BBZDR/Wor rat model for investigating the complications of type 2 diabetes mellitus. ILAR J 45(3):292–302
Wang Y et al (2005) Adiponectin Inhibits cell proliferation by interacting with several growth fac-tors in an oligomerization-dependent manner. J Biol Chem 280(18):18341–18347
Yamauchi T et al (2002) Adiponectin stimulates glucose utilization and fatty-acid oxidation by activating AMP-activated protein kinase. Nat Med 8(11):1288–1295
Zhou L et al (2009) Adiponectin activates AMP-activated protein kinase in muscle cells via APPL1/LKB1-dependent and phospholipase C/Ca2+/Ca2+/calmodulin-dependent protein kinase kinase-dependent pathways. J Biol Chem 284(33):22426–22435
5 Cell and Biological Models for the C Terminal Fragment of Adiponectin Receptor
111© Springer International Publishing Switzerland 2015 M. Pugia (ed.), Infl ammatory Pathways in Diabetes: Biomarkers and Clinical Correlates, Progress in Infl ammation Research, DOI 10.1007/978-3-319-21927-1_6
Chapter 6 C-Terminal Fragment of Adiponectin Receptor Clinical Correlations
Adrian Vella and Michael Pugia
Abstract The Ig-CTF immunoassay was tested in serum, plasma and whole blood samples for healthy volunteers with normal glucose (n = 126), gestational pre-diabetics (n = 48), long term diabetics (n = 63) and in a population undergoing pre- diabetic assessment (n = 240). Patients with infl ammation (n = 13), using immunosuppressants (n = 20), anti TNF alpha therapies (n = 8) or with breast can-cer (n = 5), liver disease (n = 23), or pancreatic cancer (n = 10) were also assessed. The values for Ig-CTF were compared to Impaired fasting glucose (IFG) and impaired glucose tolerance (IGT) detected by OGTT results and HbA1c test result. The circulating Ig-CTF values are suppressed in diabetics and infl ammation. Patients using anti- TNF α therapies (Remicade® Infl iximab) exhibited suppressed Ig-CTF. The Ig-CTF values did not change following glucose administration or over 45 days without signifi cant weight changes and the within patient variability was <10 %. The predictive value of Ig-CTF for an advanced diabetic disease to healthy control diabetes was a 92 % sensitivity and 93 % specifi city. Ig-CTF was unable to resolve early diabetes from normals Ig-CTF did not meet any predictive value criteria. Only 10 % of earth diabetics and pre-diabetics had abnormally low Ig-CTF values.
A. Vella , MD (*) Division of Endocrinology and Metabolism , Mayo Clinic , 200 First ST SW , Rochester , MN 55905 , USA e-mail: [email protected]
M. Pugia Strategic Innovation , Siemens Healthcare Diagnostics , 3400 Middlebury Street , Elkhart , IN 46515 , USA e-mail: [email protected]
112
6.1 Introduction
Diabetes is considered a disease that progresses through overlapping groups of dis-ease phenotypes and pathologies (Brooks-Worrell and Palmer 2011 ). The measure-ment of the progression is of great epidemiological importance especially as there are over 470 million pre-diabetics worldwide (Tabák et al. 2012 ). Diagnosis of the on-set of diabetes in pre-diabetics is typically preformed by determination of glu-cose intolerance. The common clinical practice to detect on-set is by a combination of testing including an oral glucose tolerance test (OGTT), fasting blood glucose (FBG) and/or hemoglobin A1c (HbA1c) (Sacks et al. 2011 ). However predicting which pre- diabetic will become diabetic is particularly diffi cult as on-set might not occur for many years. Only small changes in beta-cell function are detectable in pre-diabetes and arise from an interaction between insulin resistance and a decrease in beta-cell responsivity to glucose (Sathananthan et al. 2012 ). The CDC estimates application of common practices to the 72 million U.S pre-diabetics would add 4.52 mill newly diagnosed diabetics if the testing was done by OGTT, 2.25 mill if tested by fasting blood glucose (FBG) or 1.58 million if tested by HbA1c (CDC.gov). Any method used for prediction and early intervention of which patients would develop diabetes would have to be highly specifi c. Otherwise, even a small % false positive would result signifi cant follow-up of the pre-diabetics population. At the same time the diagnostic method must be sensitive, reproduce-able within an individual and consistent with common practice. Any changes in diagnostic methodology could cause signifi cant numbers of patients to be impacted.
The OGTT is most sensitive and specifi c measure on onset, but least practical to applay to every suspected case as timed sample collection is needed. Impaired fasting glucose (IFG) and impaired glucose tolerance (IGT) are detected by OGTT and allow placing individuals at higher risk of overt type 2 diabetes. The OGTT best correlates with insulin resistance/sensitivity measures. The FBG measurement itself is a very simple test, but very non-specifi c with false positives and has high inter-subject vari-ance requiring repeat testing if discrete diagnostic cut-offs are applied. Fasting is required for both and tests can suffer from glucose instability requiring specifi c sam-ple tubes and storage consideration.
The HbA1c measurement is a highly specifi c measure of persistent diabetes with a >95 % specifi city at the >6.5 % HbA1c threshold. The HbA1c measurement has the advantages of allowing non-fasting measurement with low inter individual variability and has also been tied to retinopathy complications (Lopes-Virella et al. 2008 ). However, the HbA1c measurement is insensitive to detection of on-set in pre-diabetes and gestational diabetes and only detects persistent hyperglycemia. The HbA1c is not very predictive of a positive OGTT, insulin resistance or β-cell dysfunction (Dods and Bolmey 1979 ).
Surrogate indexes of insulin sensitivity/resistance derived multiple time points in the OGTT are available (e.g. QUICKI, HOMA, Log HOMA, 1/HOMA, Matusda index and 1/Fasting insulin). While reported as useful methods for diagnosing insulin resistance, and follow-up during the treatment of patients with diabetes (Katsuki 2001 ), the correlations of most surrogates indexes to glucose clamp
A. Vella and M. Pugia
113
testing are modest at best and do not offer much to clinical practice over direct measurement (Lee et al. 2011 ).
Alternative approaches to sorting out pre-diabetic progression measure the underlying insulin sensitivity/resistance (Muniyappa et al. 2008 ). These include direct measurements based on hyperinsulinemic euglycemic glucose clamp or insu-lin suppression test. The glucose clamp is widely accepted as the reference stan-dards for measuring insulin sensitivity/resistance (DeFronzo et al. 1979 ). Hepatic catheterization allowing measurement of hepatic insulin extraction and systemic insulin/c-peptide clearance does confi rm the kinetics of altered insulin secretion and absorption in obese and diabetic subjects (Tura et al. 2001 ). However these methods can only done in highly controlled research studies where insulin is infused in a complex time-consuming, labor-intensive, and invasive procedure.
Indirect measurements of insulin sensitivity/resistance are more commonly uti-lized as they require fasting followed by a bolus of glucose and plasma glucose, c-peptide and insulin measurements. Indirect measurements by the frequently sam-pled intravenous glucose tolerance test are most practically used and methods for estimating insulin sensitivity/resistance. Here model-derived estimates allow calcu-lation of indices for insulin action and β-cell responsiveness.
New diagnostic methodologies are often tested and compared to results obtained in the screening process for glucose intolerance during gestation ( Tieu et al. 2010 ). This cohort allows a short duration for prediction of on-set based on patients who progress to gestational diabetes. More pregnant women received glucose testing than any other risk factor group, making this a good means to obtain samples for the study of on-set of diabetes (Catalano 2010 ). This risk group also as an increasing number of individuals classifi ed as having either impaired fasting glucose (33.8 %) or impaired glucose tolerance (15.4 %) which allows reasonable sampling numbers (CDC.gov).
Inhibition of insulin degradation by CTF suggests a potential to impact this cor-relation with insulin resistance and the possibility to further characterize diabetes phenotypes. However, any new surrogate marker intended to replace or supplement OGTT will need to achieve better clinical performance for detection of onset and for monitoring of progression. Ability to fl ag at risk populations with glucose intoler-ance is also an attractive feature. The new marker must surpass the performance of HbA1c or FBG. Our goal was to test the Iq-CTF clinical assay against an improved clinical performance specifi cation of a specifi city of >95 % and a sensitivity of >95 % for diabetes as determined OGTT. Additional clinical specifi city of >95 % and a sensitivity of >65 % for IGT would also be a signifi cant improvement (3X). As glucose is the primary treatment indicator for any surrogate marker would need a correlation of at least an r >0.5 to OGTT for improved performance.
The Ig-CTF immunoassay was tested for patients with gestational diabetes, pre- diabetes, and diagnosed with diabetes. Reference range testing was also conducted using healthy volunteers. The impact of sampling, sample type, fasting and days between sample were determined. Within individual variability was measured. Additionally the impact of patients having infl ammation, cancer and liver disease was determined. The impact of anti-infl ammatory and immunosuppressant treatment was assessed by measuring patients under treatment. Results were compared to a combination of testing including hemoglobin A1c (HbA1c), fasting blood glucose
6 C-Terminal Fragment of Adiponectin Receptor Clinical Correlations
114
(FBG) and/or an oral glucose tolerance test (OGTT). As a new surrogate marker intended to replace or supplement OGTT, higher better clinical performance criteria were put forth and compared to HbA1c or FBG performance. Additional compari-sons were made to indirect model-derived estimates of insulin sensitivity/resistance.
6.2 Methods
6.2.1 Reference Range Testing
Clinical specimens were collected and refrigerated (5–8 °C) from time of collection until frozen (−20 °C) shortly after they were poured off. Before additional testing, specimens were thawed to 5–8 °C and used to make dilutions for analysis. Matched serum, plasma and whole blood samples were collected from 25 health volunteer monthly for 45 days. All healthy controls maintained normal fasting blood glucose (<99 mg/dL whole blood and <115 mg/dL plasma for venous samples). Whole blood samples and plasma samples were obtained for patients outside of normal range with either impaired fasting glucose (IFG), impaired glucose tolerance (IGT) or diabetes. IFG was indicated with glucose of 115–125 mg/dL at 0 min and IGT was indicated with glucose of 140–200 mg/dL at 120 min. Diabetes was indicated with glucose of ≥126 mg/dL at 0 min or ≥200 mg/dL at 120 min or >6.5 % hbA1c.
The impact of cancer and liver disease was assessed by comparison of serum samples from normal patients no history of cancer, liver disease, diabetes or meta-bolic syndrome, or infl ammation to those with liver disease or cancer (Promeddx, Norton MA). Equal numbers of male and female patients were used. Patients with liver disease had either alcoholic hepatitis, nonalcoholic steatohepatitis (NASH), autoimmune hepatitis or hepatocellular carcinoma, and cancer patients had either pancreatic or breast cancer.
The impact of infl ammation was also assessed by comparison of whole blood samples from normal patient (n = 101) to patients (n = 13) with active pro- infl ammatory response (CPR >5 mg/dl or CBC of >10 white blood cells/mL). Also patients using anti-infl ammatory therapies such as Remicade® (Infl iximab) were compared. Infl iximab is a monoclonal antibody against tumor necrosis factor alpha (TNF-α) used to treat autoimmune diseases. Transplant patients on immunosup-pressant were also tested. These additional samples were obtained from Bioreclamation LLC of Westbury NY and Promeddx of Norton MA.
6.2.2 Gestational Diabetic Collection
Whole blood fasting venous specimens were collected from gestational patients at Methodist Medical Center of Illinois, Chicago IL from Nov 2011 to Feb 2012. Patients were fasting for 8 hours prior to sample collection. Capillary fasting POC glucose test (fi nger stick/glucose meter) was checked fi rst, if >135 mg/dL then a
A. Vella and M. Pugia
115
grey top tube collected and the OGTT given. The gestational diabetes OGTT proce-dure calls for 100 g glucose instead of 75 g glucose. Samples were collection at 0, 60, 120 or 180 min. Most physicians chose to collect a fasting, 60 min and 120 min specimen. Whole blood fasting venous samples were sent immediately to chemistry for laboratory glucose determination. Plasma glucose values for 0 and 2 h were obtained.
Specimens for 48 female patients were fl agged as pre-diabetic by glucose meter were obtained for 0 and 120 min time points. Specimens for 101 female patients with normal fasting plasma glucose were also obtained from a comprehensive meta-bolic panel performed with an EDTA tube. Fasting plasma blood glucose were obtained and evaluated against the glucose limits for fasting venous samples <115 mg/dL. Of the 48 diabetics patients fl agged pre-diabetic, 28 were IGT, 2 were diabetic and 18 were in the normal range
6.2.3 Pre-diabetic Collection
Whole blood and plasma fasting venous specimens were collected from normal, pre-diabetic and diabetic at Mayo Clinic (Sathananthan et al. 2012 ). A total of 240 subjects (143 women, 97 men) gave informed written consent to participate in the study. All subjects were in good health and were at a stable weight and did not engage in regular vigorous exercise. All subjects were instructed to follow a weight- maintenance diet containing 55 % carbohydrate, 30 % fat and 15 % protein for at least 3 days prior to the study. Samples were collected form patients at 0, 10, 20, 30, 60, 90 and 120 time points after an overnight fast and using a 75-g oral glucose tolerance test (OGTT). Patients were characterized as normal glucose tolerance (NGT), normal fasting glucose (NFG), impaired glucose tolerance (IGT), impaired fasting glucose (IFG), and diabetes (DM) using analytical values for insulin and glucose measurements modeled at 7 time points out to 2 h. Additional values for c-peptide, pancreatic glucagon, human growth hormone, and cortisol were exam-ined as well. Insulin action (Si) was estimated in this study using the oral glucose minimal model, and β-cell responsively indices (Phi total) were estimated using the oral C-peptide minimal model. The disposition index (DI) for each individual was calculated
Subjects with a prior diagnosis of diabetes or a fasting glucose >126 mg/dl (>7 mm) at the time of screening were excluded, as were those taking medications that could affect glucose metabolism (such as insulin, GLP, DPP-4 inhibitors, sulfo-nylureas, meglitinides, alpha-glucosidase inhibitors and thiazolidinediones (TZD)) or those taking medications which induce diabetic symptoms (glucosamine, rifam-picin, isoniazid, olanzapine, risperidone, progestogens, corticosteroids, glucocorti-coids, methadone many anti-retrovirals, D5W/glucagon and TPN-induced multiple enteral feedings). Patients were excluded with infectious diseases (HIV, HCV, HBV and immunodefi ciency), certain cancers (leukemia, multiple myeloma, and lym-phoma), certain liver disease (hepatitis & cirrhosis) those taking TNF alpha inhibi-tors (Infl iximab and others).
6 C-Terminal Fragment of Adiponectin Receptor Clinical Correlations
116
The hbA1c values were measured on the thawed whole blood samples using the DCA. Iq-CTF-Ig values were measured blinded on the thawed whole blood and plasma samples using the ELISA method in duplicate for all patients in four sepa-rate run. Siemens results were sent to Mayo for examination prior to receiving Mayo analytical data. Samples were placed in ice, centrifuged at 4 °C, separated and stored at −20 °C until assayed. Samples were thawed to 4 °C not more than three times.
6.2.4 Diabetic Collection
Whole blood venous specimens were also collected from 68 long term diabetic undergoing regular HbA1c monitoring at Diabetes Diagnostic Laboratory, University of Missouri School of Medicine, Columbia MS from Jan 2011 to Mar 2012. The patients were not presumed fasted and collected in either grey top or EDTA tubes. The hbA1c values were measured on the thawed whole blood samples using the DCA Advantage instrument in duplicate for all patients. Microscopic examination of blood sample confi rmed freezing caused complete blood lysis.
6.2.5 Ig-CTF and CTF Measurements
The Ig-CTF and CTF bioassay descriptions and analytical validation data for repro-ducibility & stability are shown in Chap. 3 . The method calibration, precision, accu-racy, interference and recovery were demonstrated. Variability of the Ig-CTF and CTF determinations required use of preservatives in the blood tube and the dilution buffer to allow an acceptable allowable total error. The addition of EDTA, citrate, and BSA at pH 6.4 to the dilution buffer was required to allow to −20 °C storage freezing of diluted plasma. Deactivation of platelets in the sample was critical for decreasing variance and the impact of freezing. Collections with EDTA tubes were less variable than heparin and sodium fl uoride tubes. Addition of heparin instead of EDTA was required for freezing diluted whole blood. With the sample stabilized, the average bias was 4.5 % CV for repeated plasma samples. Whole blood samples were more variable 7 % CV. The data showed a spot measurement was possible. Plasma, serum and whole blood use was possible.
6.3 Results
6.3.1 Normal Range and Individual Variability Testing
The values for Ig-CTF apparently healthy volunteers were measured for the three sample types (See Table 6.1 ). Matching serum, plasma and whole blood samples were collected from the same donor. Whole blood values were highest followed by
A. Vella and M. Pugia
117
plasma and then by serum values. The correlation between whole blood and plasma for the same patients was R^2 of 0.63 and a slope of 1.17 for higher whole blood values compared to plasma. Higher whole blood values were expected by white blood cell binding of IgG (Dickler 1973 ). Platelet binding of Ig could explain higher values in plasma than serum (Worth et al. 2006 ).
The values for Ig-CTF were next compared with patients undergoing glucose and HbA1c testing. The Ig-CTF plasma values for patient with pre diabetes and diabetes were signifi cantly lower than patients with normal glucose tolerance (P > 0.01). Overnight fasting did not impact Ig-CTF values in health donors or diabetics with all values within 9–14 % of non-fasting values. The Ig-CTF val-ues did not change during the 120 min following glucose administration. Values did not change over 45 days. This is consistent with the 14 day plasma half-life of immunoglobulin (Hinton et al. 2006 ). Values did changed in a patients who
Table 6.1 Impact of infl ammation on Ig-CTF R1 values
Specimen type Patient group
– Ig CTF ng/mL
n Aver sd min max
Serum Normal 25 322.5 173.4 55.5 778.8 Serum Liver disease 23 1141.2 1514.3 293.8 6854.7 Serum Breast cancer 5 794.6 260.7 340.7 1140.0 Serum Pancreatic cancer 10 581.2 430.5 293.2 1330.7 – – – – – – – Plasma Normal 25 727.5 343.0 290.2 1723.4 Plasma Anti TNF α therapy 8 404.1 130.1 194.1 601.0 Plasma Pre-diabetic (IGT) 119 349.5 268.1 95.1 2346.3 Plasma Pre-diabetic (IFG) 100 335.3 256.4 99.0 2346.3 Plasma Diabetic 24 268.0 141.1 91.3 645.7 – – – – – – – Whole blood Normal 101 1234.1 651.9 178.6 4758.9 Whole blood Infl ammation 13 402.1 173.6 145.2 713.2 Whole blood Immunosuppressant 20 566.0 361.1 160.7 1427.1 Whole blood Pre-diabetic 48 527.0 195.6 250.9 1055.2 Whole blood Diabetic 63 222.3 97.3 51.6 444.1
The values for Ig-CTF were measured in serum, plasma and whole blood for patients and health donors. Patients were characterized by oral glucose tolerance test if sample was plasma. Donors lacked infl ammatory conditions (CBC & CRP normal) unless indicated. The HbA1c and FBG was measured if the sample was whole blood. Matching samples for all three sample type were com-pared across 25 health donor and 48 pre-diabetics and produce values similar to those shown
Sample collection had to be optimized to reduce within patient variability to <10 % to suppress coagulation. We found plasma with activated platelets caused 4-fold increase in Ig-CTF values compared to the same plasma with deactivated platelets. Activation caused a visible cloudiness to occured in diluted sample. The impact on the assay was explained by IgG binding to Fc receptors on platelets and causing platelet aggregation artifi cially elevating values (Worth et al. 2006 ). Complement components also bind to activated normal platelets and promote binding of immuno-globulin through C1 q (Hamad et al. 2010 ). As a control it was verifi ed neither C1q nor Adpn bound to Ig-CTF as tested by ELISA and platelet binding Ig-CTF is a likely through the immuno-globulin Fc receptor
6 C-Terminal Fragment of Adiponectin Receptor Clinical Correlations
118
underwent signifi cant a weight change (>20 % drop) after 45 days. This con-trasts the rapid changes of free CTF measured in animals during the 120 min following glucose administration (See Chap. 5 ). Ig-CTF appears to be more a chronic phase marker.
6.3.2 Correlation to Infl ammation
The values for Ig-CTF measured shown for normal and diabetic donors were in cases lacking infl ammation. Infl ammatory conditions were ruled out by a CPR <5 mg/dl or CBC of <10 white blood cells/mL. These cut-off do not rule out mild elevation of infl ammation. The patients with signifi cant infl ammation, as indicated a CRP and CBC positive, did exhibited a signifi cantly lower Ig-CTF (P > 0.01) than normal patients (Table 6.1 ). This decrease was of a similar magnitude to a decrease in the Ig-CTF seen in diabetes.
Patients using anti-infl ammatory therapies such as Remicade® (Infl iximab) exhibited suppressed Ig-CTF values. Transplant patients on immunosuppressants also had suppressed Ig-CTF values. The impact of anti-TNF alpha inhibitor drugs, would be expected to prevent CTF formation by inhibition of TACE. Others drug expected to impact the assay would include adalimumab (Humira), certolizumab pegol (Cimzia), golimumab (Simponi), or etanercept (Enbrel). Immunosuppressants would also be expected to lower immunoglobulin levels and suppress Ig-CTF formation.
Patients who had diseases that would be expected to increase blood immuno-globulin levels, also had changes in the values for Ig-CTF. Liver diseases represent a signifi cant infl ammatory disease group. Liver diseases (Hepatitis & cirrhosis) elevate serum immunoglobulin. Values for Ig-CTF were signifi cantly increased with auto-immuno hepatitis patients having the highest values (Table 6.2 ). Patients with leukemia, multiple myeloma, and lymphoma were excluded as these cancers would clearly elevate immunoglobulin levels. However, cancer is generally associated with greater blood immunoglobulin levels and many patients with breast and pan-creatic carcinomas did exhibited increased Ig-CTF values.
6.3.3 Correlation in Gestational Diabetes
Whole blood fasting venous specimens were collected from gestational patients. Patients were fl agged by fasting POC glucose as normal (n = 101) or pre-diabetic (n = 48). Additional whole blood fasting venous specimens were collected from confi rmed diabetics by HbA1c >6.5 % (n = 68). The diabetic and normal groups are well separated using Ig-CTF values and cutoff of 500 ng/mL (Fig. 6.1 ). The pre- diabetic group had Ig-CTF values in the middle of the normal and diabetic groups
A. Vella and M. Pugia
119
Correlations of Ig-CTF were made between non-diabetic, pre-diabetic, gesta-tional diabetic and diabetic populations. Overnight fasting did not impact the Ig-CTF results. Results for Ig-CTF were compared to diagnosis based on fast blood glucose (FBG), oral glucose tolerance test (OGTT), and hemoglobin A1c (hbA1c). The clinical correlation of Ig CTF was highest for a glucose tolerance test (OGTT). The average variance between samples was 9 % with a minimum variance of 2 % and maximum variance of 29 % in one sample. This reinforced that one CTF-Ig measurement could be taken and a timed collection was not needed.
The correlation between values for Ig-CTF and glucose at 120 min is shown in Fig. 6.2 . In general, the higher Ig-CTF values were more likely it is to have lower glu-cose values which were in the normal range. However, the correlation to glucose was poor, and Ig-CTF did not change with glucose administration. Ig-CTF values were unchanged between the samples collected at 0 min and 120 min OGTT time points. In contrast free CTF in animals had a higher correlation with glucose and does increase sharply during the 120 min OGTT challenge after glucose administration (See Chap. 5 ).
Table 6.2 The predictive values of CTF-Ig for in gestational study
Study 1 Study 2
TN 95 110 FN 0 10 FP 7 10 TP 64 84 Sens 100.0 % 89.4 % Spec 93 % 92 % PPV 90 % 89 % NPV 100 % 92 %
Study 1 is a comparison of normal vs. diabetics with pre-dia-betic considered normal and diagnosis based solely on FBG and HbA1c values. Study 2 is a comparison of normal vs. dia-betics with IGT pre-diabetics considered positive with dia-betic and diagnosis based solely on OGTT and HbA1c values. The effi ciency of ruling in and out impaired glucose tolerance (IGT) was also tested. Of the 48 patients fl agged by FBG as impaired, only 2 patients were confi rmed by OGTT as a dia-betic and 30 were classifi eds as impaired glucose tolerance (IGT). The Ig-CTF value eliminated 10 of the 30 follow ups fl agged by FBG and agreed with 13 of the 16 patients cor-rectly rule out by FBG
Two patients were clearly Ig-CTF false negatives. Patient A with fasting plasma glucose of 428 mg/dL, abnormal hbA1c of 9.1 % and a normal Ig-CTF of 858 ng/mL and patient B with abnormal hbA1c of 10.2 % and a very high normal Ig-CTF of 2625 ng/mL. In both cases, elevated Ig-CTF values could be due to generally elevated immunoglobulin. There were seven patients with normal fasting glucose but “false positive” by Ig-CTF value or <500 ng/mL. These specimens were reanalyzed with new dilutions and plates. As the results were repeated they were considered as clinical “false positive”
6 C-Terminal Fragment of Adiponectin Receptor Clinical Correlations
120
The predictive value of Ig-CTF for diabetes in a gestational population is shown in Table 6.2 . Both sensitivity and specifi city were high for assessing diabetes (100 % sens/93 % spec) and pre-diabetes or diabetes (89 % sens/92 % spec) (Study 1 and 2). The Ig-CTF measurements based on the Adiponectin receptor 2 produced a similar correlation. This meets the target criteria of >95 % spec for diabetes in combination with HbA1c with a >70 % sensitivity for diabetic detection. In the second study, the fasting blood glucose (FBG) was not as predictive as CTF-Ig hav-ing poor specifi city of 30 % but good sensitivity of 95 % for diagnosis of diabetes based on OGTT. The hbA1c was also not as predictive as CTF-Ig having poor sen-sitivity. HbA1c values of ≥6.5 % only confi rmed 68 of the 129 diabetes patients (roughly half). HbA1c had a good specifi city of 95 %. However none of the pre-diabetic population (0 of 48) had positive hbA1c and one of the normal population (n = 102) had an hbA1c of 7.3 %.
0 %
10 %
20 %
30 %
40 %
50 %
60 %
Ig-CTF ng/mL
% o
f p
atie
nt
in g
rou
p
Diabetic
Prediabetic
Normals
Liver cirrhosis& cancer
ImmunosuppressantsInflammationAnti-TNF therapies
Fig. 6.1 Distribution of Ig-CTF in Diabetes and Infl ammatory Diseases. Whole blood values of Ig-CTF were measured for non-diabetic pregnancies, gestational diabetic and long term diabetic populations characterized by fast blood glucose (FBG), oral glucose tolerance test (OGTT), and hemoglobin A1c (hbA1c). Values for long term diabetics were lower than non-diabetics with pre- diabetics falling in-between. The patients with infl ammation, using anti-infl ammatory or immuno-suppressant therapies exhibited a lower Ig-CTF. Liver diseases (Hepatitis & cirrhosis) and cancers which elevate immunoglobulin levels exhibited a higher Ig-CTF
A. Vella and M. Pugia
121
6.3.4 Correlation in Pre-diabetics
The predictive value of Ig-CTF for diabetes in a population with many pre-diabetic and not well separated from normal is shown in Table 6.3 . Whole blood and plasma specimens were collected for a group of patients characterized by FBG, OGTT and HbA1c. No patient was >6.5 % HbA1c. Patients were characterized into fi ve groups. Normal patients were normal glucose tolerance (NGT) & normal fasting glucose (NFG) (n = 66, 27 % of population). Pre-diabetics (n = 150, 63 % of population) were either impaired glucose tolerance (IGT) and NFG (n = 63) or impaired fasting glucose (IFG) and NGT (n = 26) or IFG and IGT (n = 61). Diabetic patients (DM) were positive for the OGTT (n = 24, 10 % of population).
The Ig CTF values did not separate pre- diabetics or diabetics in this population (See Table 6.3 ). Ig-CTF did not meet any predictive value criteria separating pre- diabetic from normal and achieved only a ~50 % specifi city. Only 10 % of diabetics and pre-diabetics had abnormally low Ig-CTF values. The predictive value for Ig-CTF also did not meet the ideal criteria achieving only a >50 % specifi city for pre-diabetes and clinical sensitivity of >56 % for IGT and of >75 % for DM (See Table 6.3 ). While Ig-CTF was not predictive in this population, HbA1c was also
y = – 3.705x + 1090.7R2= 0.218
0.0
200.0
400.0
600.0
800.0
1000.0
1200.0
100 120 140 160 180 200 220 240 260
Glucose at 120 min
CT
F Ig
ng
/mL
Fig. 6.2 Correlation of Ig-CTF to the 120 min OGTT glucose value. Whole blood values of Ig-CTF are plotted against the glucose value at 120 min in the oral glucose tolerance test (OGTT). The liner correlation coeffi cient was not an r > .5 indicating a weak correlation
6 C-Terminal Fragment of Adiponectin Receptor Clinical Correlations
122
completely insensitive for pre-diabetes and diabetes. Meanwhile this data demon-strated the false positive discordance of FBG for predicting IGT or OGTT.
Patients were also fully characterized by insulin action measures (DI, Si, Phi) using timed measurement of insulin, glucagon, glucose, and c-peptide (Sathananthan et al. 2012 ). Ig-CTF did not meet any predictive value criteria and was only ~50 % specifi c and sensitive for predicting any difference in indices of insulin secretion and
Table 6.3 The predictive values of Ig- CTF for pre-diabetic and diabetics
– IgG CTF – – – Patient group High Low Cat 1 Cat2 Normal (NFG & NGT) 34 32 TN 104 34 Pre-diabetics (NFG & IGT) 29 34 FN 6 76 Pre-diabetics (IFG & NGT) 13 13 FP 112 32 Pre-diabetics (IFG & IGT) 28 33 TP 18 98 Diabetics (DM) 6 18 sens 75.0 % 56.3 % – – – spec 48 % 52 % Total 110 130 PPV 14 % 75 % – – – NPV 95 % 31 % – – – – – – – IgG CTF – – – Phenotype group High Low – Cat 2 Normal 46 43 TN – 47 Flat curve (Cortisol high) 1 0 FN – 59 Pre-diabetic with normal insulin curve 26 27 FP – 43 Pre-diabetic with insulin resistance curve 27 39 TP – 86 Type 2 Diabetic -normal insulin curve 3 7 sens – 59.3 % Type 2 Diabetic -insulin resistance curve 3 13 spec – 52 % – – – PPV – 67 % Total 106 129 NPV – 44 %
Plasma correlations shown were stronger than those observed in whole blood. However the impacts observed in plasma were mirrored in whole blood. The clinical thresholds for an abnormal Ig-CTF used were <300 ng/mL for plasma and <650 ng/mL for whole blood. A hemoglobin ratio helped plasma and whole blood correlation, and is not shown. Categorization 1 (Cat 1) is a comparison of normals vs. diabetics with pre-diabetics is considered normal. Categorization 2 (Cat 2) is a com-parison of normals vs. diabetics and pre-diabetics
Despite the lack of diagnostic correlation, there were several notable trends which agreed with biochemical model. Patient with lower plasma Ig-CTF were more likely to exhibit increased glu-cose of >125 mg/dL over time. Patient with lower plasma Ig-CTF were more likely to exhibit smaller changes in insulin of <20 or C-peptide of <1. Greater changes in Phi total (>50), Si(>20) or DI(>1000) were more likely with lower CTF. No correlation in pancreatic glucagon was observed
Several un-expected correlations were also found. Females had lower Ig-CTF (Avg (sd) of 269(150) ng/mL) compared to males (Avg(sd) of 373(266) ng/mL). This gender trend was in all patient categories. Patients with high plasma Ig-CTF were more likely to exhibit with no changes in growth hormone or high cortisone values. However, only a small percentage of normal and pre- diabetic patients exhibited high Ig-CTF values. Five patients had Ig-CTF of >2500 ng/mL. One patient had Ig-CTF of ~6000 ng/mL. At the same time about 2.9 % of all patients exhibited extremely low Ig-CTF values (<100 ng/mL)
A. Vella and M. Pugia
123
action (data not shown). Insulin resistance phenotype was also defi ned by a hyperin-sulinemia (fasting 60 min insulin of >48 IU/mL) and impaired glucose tolerance (fasting 120 min glucose value from 140 to 200 mg/dL in OGTT). The plasma and whole blood Ig CTF values could not separate this phenotype in either the pre-dia-betic or diabetic groups (See Table 6.3 ).
6.4 Conclusions
The circulating Ig-CTF values are suppressed in diabetes in and infl ammation. As a diagnostic this marker was un-able to resolve early diabetes. Advanced diabetic disease was resolved from healthy controls. The drop in circulating Ig-CTF is con-sistent with greater Ig-CTF absorption in liver, pancreas and other tissue during diabetes. The circulating Ig-CTF values are also elevated during cancer and liver disease with elevated immunoglobulins. The suppression by anti-TNF and anti- immune therapies suggests immune system modulation can impact the amount of circulating Ig-CTF. As Ig-CTF is carried on activated white blood cells to tissue, it is possible Ig-CTF could stimulate immune cell response to certain antigens. The Ig-CTF assay could potential identify patients with immune mediated diabetic dis-eases (kidney, eye, liver, other) and impaired tissues loading of CTF (liver, pan-creas, brain), and those with liver disease, cancer or other chronic infl ammatory diseases. Additional clinical investigations are required.
Acknowledgements We would like to acknowledge the samples collected by Sandra D. Pottorf, CLS, ASCP (Methodist Medical Center of Illinois, Chicago IL); the samples removed from stor-age by Jessica Cray and Linda Sanders (Mayo Clinic Rochester, MN); the samples provided by Randie R. Little, PhD (Diabetes Diagnostic Laboratory, University of Missouri School of Medicine, Columbia MS); and the samples procurement help from Denise Cervelli of (Siemens Healthcare Diagnostic Newark DE).
Supplements
• ELISA Method for Ig-CTF assay • ELISA Method for free CTF assay
References
Brooks-Worrell B, Palmer JP (2011) Diabetes mellitus a continuous spectrum? Clin Chem 57(21):58–161
Catalano PM (2010) Obesity, insulin resistance and pregnancy outcome. Reproduction 140(3):365–371
6 C-Terminal Fragment of Adiponectin Receptor Clinical Correlations
124
DeFronzo RA et al (1979) Glucose clamp technique: a method for quantifying insulin secretion and resistance. Am J Physiol 237(3):E214–E223
Dickler HB (1973) Lymphocyte binding of aggregated IgG and surface Ig staining in chronic lym-phocytic leukaemia. Clin Exp Immunol 14:97–106
Dods RF, Bolmey C (1979) Glycosylated hemoglobin assay and oral glucose tolerance test com-pared for detection of diabetes mellitus. Clin Chem 25(5):764–768
Hamad OA et al (2010) Complement component C3 binds to activated normal platelets without preceding proteolytic activation and promotes binding to complement receptor 1. J Immunol 184(5):2686–2692
Hinton PR et al (2006) An engineered human IgG1 antibody with longer serum half life. J Immunol 176:346–356
Katsuki A (2001) Homeostasis model assessment is a reliable indicator of insulin resistance during follow-up of patients with type 2 diabetes. Diabetes Care 24:362–365
Lee HW et al (2011) Comparison between surrogate indexes of insulin sensitivity/resistance glu-cose clamps in rhesus monkeys. Endocrinology 152(2):414–423
Lopes-Virella MF et al (2008) Diabetes control and complications trial/epidemiology of diabetes intervention and complications cohort study group. Diabetes Care 31(10):2006–2012
Muniyappa R et al (2008) Current approaches for assessing insulin sensitivity and resistance in vivo: advantages, limitations, and appropriate usage. Am J Physiol Endocrinol Metab 294:E15–E26
Sacks DB et al; National Academy of Clinical Biochemistry; Evidence-Based Laboratory Medicine Committee of the American Association for Clinical Chemistry (2011) Guidelines and recommendations for laboratory analysis in the diagnosis and management of diabetes melli-tus. Diabetes Care 34(6):e61–e99
Sathananthan A et al (2012) A concerted decline in insulin secretion and action occurs across the spectrum of fasting and postchallenge glucose concentrations. Clin Endocrinol (Oxf) 76(2):212–219
Tabák AG et al (2012) Prediabetes: a high-risk state for diabetes development. Lancet 379(9833):2279–2290
Tieu J et al (2010) Screening and subsequent management for gestational diabetes for improving maternal and infant health. Cochrane Database Syst Rev 7:CD007222
Tura A et al (2001) Insulin and C-peptide secretion and kinetics in OGTT humans: direct and model-based measurements during. Am J Physiol Endocrinol Metab 281:E966–E974
Worth RG et al (2006) Platelet Fc gamma RIIA binds and internalizes IgG-containing complexes. Exp Hematol 34(11):1490–1495
A. Vella and M. Pugia
127© Springer International Publishing Switzerland 2015 M. Pugia (ed.), Infl ammatory Pathways in Diabetes: Biomarkers and Clinical Correlates, Progress in Infl ammation Research, DOI 10.1007/978-3-319-21927-1_7
Chapter 7 Uristatin Immunoassay Usage in Glomerular Nephritis Assessment
Saeed A. Jortani and Michael Pugia
Abstract A novel immunoassays for Uristatin (Uri) was used for testing in patients glomerular nephritis (GN) by using of an extensive clinical data and urine sample bank from previous studies (n = 6292 patients). Proteinuira or hematuria was found in 1407 of 6292 patients and used an indicator for potential cases needing glomeru-lar nephritis follow up. Of these, 351 patients had asymptomatic GN without infec-tion or Uri. The remaining 1194 patients had potential infections based on symptoms of urinary tract infection, systemic infections, and/or a positive Uri value >12 mg/L. Urine culture and CBC were used to characterize 419 patients as actually having infections. Those remaining were sorted according to Uri value as 198 cases of symptomatic GN with anti-infl ammatory stress (Uri >12 mg/L) and 533 cases without this stress (Uri <12 mg/L). The Uri method allowed sorting atypical anti- infl ammatory stress patient in glomerular nephritis.
7.1 Introduction
Diabetes leads to infl ammation, higher chances of infection and impaired wound healing (Larkin et al. 1985 ). During the progression of this disease, an innate immune system response is necessary to prevent damage to normal tissue (Donath et al. 2013 ). Urinary trypsin inhibitors (uTi) are a key part of the anti-infl ammatory
S. A. Jortani Department of Pathology , University of Louisville, School of Medicine , 511 S. Floyd Street (Room 227) , Louisville , KY 40202 , USA e-mail: [email protected]
M. Pugia (*) Department of Pathology , University of Louisville, School of Medicine , Louisville , KY 40202 , USA
Strategic Innovation , Siemens Healthcare Diagnostics , 3400 Middlebury Street , Elkhart , IN 46530 , USA e-mail: [email protected]
128
response functioning intrinsically as an inhibitor of the serine proteases causing infl ammation (Pugia et al. 2007b ; Pugia and Lott 2005 ). The uTi suppress infl am-mation and then readily pass into urine (Fig. 7.1 ). Once uTi suppresses infl amma-tion, they are readily fi ltered by the glomeruli (>99 %) through the glomerular basement membrane and accumulating in the lysosomes of proximal tubular epithe-lial cells. Thus the kidney is exposed to the majority of uTi and their anti- infl ammatory properties, and urine is the primary sample for testing.
Urinary tract infections are the primary cause of glomerular nephritis as immune cells responding to infection damage tissue and cause a pro-infl ammatory response (Hooton and Stamm 1997 ; Barnett and Stephens 1997 ; Saint et al. 1991 ). The
..from tissues forcellular proliferation
..from immune cell for bacteria phagocytosis
Bikunin
Inflammation
1. Release of trypsinproteases
4. Inhibition ofcellular proliferation
>99%cleared
into urine
Inter-a-Inhibitor
pro-Inhibitor always in blood
ElastaseCathepsinTryptase TrypsinKallikreinThrombin Plasmin Factors VII & X
Neutrophils,Monocytes,Eosinophils,Natural killer cells,Macrophages,Mast cells
Epithelial cell,Endothelial cell,Smooth muscle,Fibroblast,Platelets,Neoplastic
2. Releaseof inhibitor
3. Inhibition ofImmune mediatedapoptosis
BikuninH2
H1
Additionalfragmentation
Uristatin
Urinarytrypsininhibitors
Fig. 7.1 Urinary trypsin inhibitors are release from non-inhibitory pro-inhibitors. (Step 1 ) Serine proteases of the trypsin family (Elastase, Cathepsin, Tryptase, Trypsin, Kallikrein, Thrombin, Plasmin and Factors VII & X) are increased during infl ammation by innate immune cells (e.g. White Blood Cells; Neutrophils, Monocytes, Eosinophils, Natural killer cells, Macrophages, Mast cells) and affected tissues (Epithelial, Endothelial, Smooth muscle, Fibroblast, Platelets and Neoplastic). Step 2) These proteases produce infl ammatory response of vascular dilation, coagula-tion, and leukocyte infi ltration with destruction of pathogens. Step 3) Serine proteases (mainly elastase) liberate plasma bikunin (Bik) the primary uTi from pro-inhibitor, inter-α-inhibitor (I-α-I), activating the inhibitory response. Bik is rapidly excreted into urine. Step 4) Bik inhibits serine proteases as an anti-infl ammatory response controlling remodeling, smooth muscle contraction, fl uid/electrolyte balance and protecting tissue from leukocyte damage. Step 5) Bik signals cells to reduce proliferation, infl ammation cytokine mediator release, growth factor activation, uPA activa-tion and intra-cellular Ca+ slowing cell growth. Step 6) Bik is rapidly metabolized and excreted into urine avoiding a prolonged suppression of infl ammatory system. Uristatin (Uri) is a primary form in urine and lacks the O-linked glycoside
S.A. Jortani and M. Pugia
129
kidney model shown in Figs. 7.2 , 7.3 and 7.4 can be used to explain the impact of infl ammation and uTi on the kidney. The kidney components and blood and urine fl ow are shown in Fig. 7.2 . Pro- and anti-infl ammatory processes involved in kidney damage and regeneration can be explained by this model. In normal kidneys, only low molecular weight (LMW) proteins (<30 kDa) pass the basement membrane of the glomerular capillary into the distal collecting tube where proximal tubular cells re-absorb this protein back into the blood leaving no proteinuria (Fig. 7.3 ). In abnor-mal kidneys, where the basement membrane has been damaged, large proteins (> = 30 kDa) pass the basement membrane of the glomerular capillary into the distal collecting tube and proximal tubular cells can fail to re-absorb high amounts of protein into the circulation leaving gross proteinuria and loss of kidney function (decline of glomerular fi ltration rate (GFR)) (Fig. 7.4 ).
As kidney tissue becomes infl amed, white blood cells (WBC) migrate into the interstitial space to fi ght infection. Neutrophilic elastase is released from these WBC. Elastase proteolysis of pro-inhibitors give rise to uTi. Interleukin-α-inhibitor (I-α-I) and pre-interleukin-α-inhibitor (P-α-I) are the primary pro-inhibitor forms (Fig. 7.1 ). More than 98 % of the uTi in blood is bound to these pro-inhibitor forms and is inactive from inhibition of serine proteases. Bikunin (Bik) and Uristatin (Uri) are two key forms of uTi. Uristatins (Uri) are small derivatives of Bikunin lacking the O-linked glycoconjugates.
The uTi released during infections has a potent anti-infl ammatory response (Fig. 7.1 ). The presence of uTi correlates with febrile proteinuria and elevated white
Blood flow
Arteriole
Urine
Mesangial cellEndothial cellBasement membraneEpithelial cells (podoctye)
Glomerular capillary
Epithelial cell
Distal collecting tubeBrush border
Basement membrane & blood capillary
Bowman’s capsule
Proximal tubular
Fig. 7.2 Kidney model. The kidney model is to explain the impact uristatin on kidney during urinary tract infection and infl ammation. Key kidney components for glomerular capillary and proximal tubular are shown
7 Uristatin Immunoassay Usage in Glomerular Nephritis Assessment
130
blood cells (WBC) during urinary tract and respiratory infections (Pugia et al. 2002 , 2004 ). Low molecular weight (LMW) urinary proteins also increase during febrile proteinuria. These LMW proteins are not reabsorbed by the proximal tubular. Microalbuminuria (30–300 mg/L of albumin) is also detected during fever (Pugia et al. 2002 ). In cell models uTi has been shown to shut down the proteases need for tubular re-absorption and potentially facilitating this excretion of LMW proteinuria of during fever (See Chap. 10 ). Febrile proteinuria is gone once WBC returns to normal and fever is gone. This temporary condition does not cause loss of renal function or change glomerular fi ltration rates.
This potent anti-infl ammatory agent has been used as an anti-shock treatment and to protect tissues by suppressing vascular dilation, coagulation, and smooth muscle contraction during wound healing (Pugia and Lott 2005 ). Studies have shown uTi to protect kidney tissues from immune mediated apoptosis in animal models during glomerulonephritis (Koizumi et al. 2000 ; Takahashi et al. 2001 ; Nakakuki et al. 1996 ). There are also uTi released in response to chronically infl amed tissue. The benefi ts of these protective effects of uTi have been shown during disseminated intra-vascular coagulation, ischemia-reperfusion injury, endothelial injury and acute renal
Large proteins >30 kDa
Small proteins <30 kDa
Normal Inflammation
Immune cells >5 uM
Urinary trypsin inhibitors <30 kDa
Blood flow
Blood
Basementmembrane
Glomerular capillary
Re absorption inepithelial cell
Proximaltubular
Urineflow
Small proteins <30 kDa
Fig. 7.3 Early kidney infl ammation. In normal kidney, only the small proteins pass the basement membrane of the glomerular capillary into the distal collecting tube and proximal tubular cells re- absorb this this protein into the blood leaving no proteinuria. In kidney where the basement mem-brane has been damaged, large proteins pass the basement membrane of the glomerular capillary into the distal collecting and tube proximal tubular cells fail to re-absorb this protein into the blood leaving proteinuria. Prior to iamass immune cells such as white blood cells response to infection and migrate to infl amed tissue. The WBCs cause uTi to be released into the urine. The uTi correlates low molecular weight proteinuria presumed to be due to shut down tubular re-absorption. Infl amed tissue can also release serine proteases that are able to generate uTi from its pro-inhibitor
S.A. Jortani and M. Pugia
131
failure (Eguchi 2003 ; Murakami 1988 ; Yamaguchi et al. 2000 ; Nakatani et al. 2001 ; Onbe et al. 1991 ). uTi has also been shown to reduce release of growth factors and cytokines, inhibit growth factor activation, and stop protease-activated receptor (PAR) signaling in cell models (Fig. 7.1 ) (Pugia et al. 2007b ).
Persistent white blood cells (WBC) will mediate cell death in tissues and lead to glomerular nephritis causing lesions in basement membrane (Cochrane et al. 1965 ). Glomerular nephritis is characterized by immune complex deposition in kidney and is observed with hematuria and proteinuria (McCluskey 1970 ; Hatano 1984 ; Couser 1998 ). Damage can occur through an immune mediated apoptosis process (Bonventre and Weinberg 2003 ). Untreated glomerular nephritis progresses to chronic kidney disease (CKD). The infl ammatory response will transition from the innate immune response to complement and auto immune responses (McCluskey 1970 ; Pickering et al. 2013 ). Chronically infl amed tissue will develop glomerular sclerosis with glomerular C3 fragment deposition (Pickering et al. 2013 ; Cameron 2003 ). As glomerular sclerosis develops, the kidney tissue fails to regenerate to normal state. These chronically infl amed tissues promote proliferation by release of serine proteases to activate protease-activated receptors (PAR) (Vesey et al. 2013 ;
Tissue damage
Blood flow
Urine
Blood
Urine
Blood
Glomerular leakage
Tissue specific markersMicroalbuminuria & proteinuria
Tissuerecovery
Diseaseprogression
Fig. 7.4 Tissue recovery or disease progression. White blood cells such as Neutrophils and Macrophages lead to lesions in the basement membrane. Damage continues without a strong anti-infl ammatory response by urinary trypsin inhibitors (uTi) to shut down the immune mediated apoptosis. Microalbuminuria is fi rst detectable before the loss of renal function or of change in glomerular fi ltration by blood markers. As tissue damages progresses and the glomerular leakage of high molecular weight protein occurs and increases the amount of protein spilled into urine. Regeneration of renal cells such as podocytes is possible at early stages and recovery of kidney possible. In absence of regeneration, complement form and the disease will progress
7 Uristatin Immunoassay Usage in Glomerular Nephritis Assessment
132
Coelho et al. 2003 ; Tull et al. 2012 ). The key proteases causing this apoptosis and proliferation are known to be inhibited by uTi (Pugia et al. 2007b ). In the absence of uTi protecting normal tissue, the damage increases and proliferation occurs lead-ing to proteinuria (>300 mg/L) and decreased glomerular fi ltration rate (GFR) (Fig. 7.4 ).
The kidneys can recover if protected from immune cells and allowed to regener-ate normally (Yoshida and Honma 2014 ). The anti-infl ammatory response appears to be central to the ability of the kidney to regenerate and naturally reduce these complications. Here, uTi is an ideal candidate for assessment of immune cell medi-ated kidney damage in chronically infl amed kidney tissue. However the role of uTi in progression from a healthy kidney to damaged tissue is not completely known. uTi offers value for identifying glomerular nephritis patients under anti-infl amma-tory stress.
Urinary trypsin inhibitors (uTi) in a urine specimen can be measured by inhibi-tion of trypsin in a dipstick method based on a newly found enzyme substrate (Pugia et al. 2002 , 2004 ; Jortani et al. 2004 ). The dipstick is a urine test alternative to a blood C-reactive protein (CRP), a complete blood count (CBC) or sedimentation rate in patients with upper respiratory infections and other infections owing to bac-teria and surpassed these measures in patients with urinary tract infections (UTI). Clinical use of inhibition assays for urinary trypsin inhibitors were also demon-strated useful as acute phase reactant and for renal disease by other groups (Kuwajima et al. 1989 , 1992 ; Takemura et al. 1994 ; Mizon et al. 2002 ).
Immunoassays for uTi are typically unable to distinguish among the various forms of urinary trypsin inhibitors from pro-inhibitors (I-α-I and P-α-I) (Pugia et al. 2002 ; Trefz et al. 1991 ). The cross-reactivity negates blood specimens where I-α-I and P-α-I are present in healthy patients. Western blot analysis of urine identifi ed uristatin and bikunin were key forms of uTi (Pugia et al. 2002 ). A monoclonal anti-body (IATI5) produced against I-α-I also recognized P-α-I, AMBK and bikunin (Trefz et al. 1991 ). Others showed monoclonal antibodies raised against AMBK were reactive to uristatin, AMBK, and bikunin (Bik) (Kobayashi et al. 2000 ).
New monoclonal immunoassays were developed that do not show an interfering cross reactivity to pro-inhibitors in blood and urine (Pugia et al. 2007a ). The anti-bodies uncovered an impact of glycoconjugation, measured certain glycoforms could detect uristatin Uri and bikunin. The primary uTi remaining in blood is Bik with two Kunitz inhibitor domains and O-linked and N-linked polysaccharides. In urine, Uri are derivatives of Bik, lacking the O-linked glycoconjugates chondroitin sulfate chains are often found along with Bik. The urine immunoassay has been used to measure both Bik and Uri in urine.
The immunoassay detects forms that are more common in kidney disease, thus we explored the application of this immunoassays for kidney diseases by making use of an extensive clinical data and sample bank from previous studies (n = 6294 patients). It was possible that these new uTi assays could predict glomerular nephri-tis progressing to chronic kidney disease as an indicator of damaging infl ammation and infection. The goal was to test the role of Uri and Bik and to better understand the involvement in the stages of nephritis.
S.A. Jortani and M. Pugia
133
7.2 Methods
7.2.1 Immunoassay for Clinical Analysis
Patients’ urine specimens were tested using a competitive ELISA method previ-ously described with new antibodies (See Table 7.1 ) (Pugia et al. 2007a ). The immunoassay measures both Uri and Bik but does not cross-react with IαI or pαI. Calibration of the immunoassay used Uri isolated from human patients with chronic renal failure (SciPac, Sittingbourne, Kent, UK, product code P392-1). All lots were stable at −20 °C to −70 °C, either in solution or as lyophilized solids and previously described (Pugia et al. 2007a ).
The ELISA Method for Uri is initiated by coating each well with 0.1 ug/well Uri in TBS (Tris buffered saline) overnight at 4 °C followed by washing fi ve times with 300 μL of TBS per well (El x 405 Auto Plate Washer, Biotek Instruments, Winooski, VT). The plate is blocked with 300 μL of SuperBlocker® per well, incubated for 60 min at 37.5 °C (JitterBug Microplate Shaker, Beokel Scientifi c, Feasterville, PA), and fi nally washed the plates fi ve times with 300 μL of TBS-with 05 % Tween- 20 per well. Calibrators made with Uri are diluted with uristatin antibody conjugated to alkaline phoshatase (ALP-Mab) conjugates in TBS at 0.084 μg/L (mg/L). A sample is made by diluting urine with the required a 100-fold dilution of ALP- Mab. Diluted calibrator or specimen in 100 μl volumes are loaded into the coated wells, incubated at 35 °C for 1 h, and washed fi ve times with 300 μLTBS-Tween per well. The antibody binding is measured at 37 °C over 30 min after addi-tion of PNPP 100 μL/well (100 ng/mL) (1-Step™ solution, Pierce), by reading the change in 410 nm absorbance on a FLx 800 microplate Reader (Biotek Instruments). (See on- line supplement for ELISA Method for Uristatin for additional details at www.extras.springer.com/2015/978-3-319-21926-4 ). A uristatin positive is consi-derd as ≥12 mg/L.
Table 7.1 Antibodies for Uristatin and Bikunin
Antibody Raised against Clone/lot
Uristatin/Bikunin N-link glycan binding Human Uristatin ATCC 420-5D11.5G8.1E4 PTA-7744
Bikunin/Uristatin tyrpsin inhibitory peptide domain binding
Human Uristatin ATCC 421-3G5.4C5.3B6 PTA-7746 ATCC 421-5G8.1AB.5C1 PTA-7745
mAb for Uristatin/Bikunin conjugated to ALP
Human Uristatin ATCC 421-3G5.4C5.3B6
pAb for Uristatin/Bikunin conjugated to ALP
Urinary trypsin inhibitors
Rabbit polyconal
Legend Antibodies used for clinical analysis are shown. The urinary trypsin containing used to produce the polyclonal contained both Uri and Bik. Monoclonal antibodies were produced with only purifi ed Uri lacking Bik. ATCC deposited cell lines are available from ATCC upon request a [email protected]
7 Uristatin Immunoassay Usage in Glomerular Nephritis Assessment
134
Some patients’ urine specimens were also tested using a western blot method with rabbit polyclonal antibody conjugated to ALP A described (Table 7.1 ) (Pugia et al. 2004 ). The antibody cross-reacted with Uri, Bik, P-a-I, I-a-I, and AMBP. Urine specimens from diabetics were characterized using a commercial pre-cast SDS- PAGE gel system (BioRad). In brief, urine from diabetic patients was concentrated to about 20 μg total protein and loaded onto the gel along with molecular-weight markers and Bik standards. The SDS-PAGE gel was ran in 25 mM Tris, pH 8.3, 192 mM glycine, and 0.1 % SDS buffer at 110 V in for 1–1.5 h until complete. The gel was transferred to the blotting cassette using fi lter paper. The gel soaked in transfer buffer and proteins transfers to nitrocellulose paper at 30 V in a 4 °C cham-ber overnight. The membrane was washed in TBS buffer plus 0.1 % Tween-20, blocked with 5 % BSA (Sigma A3059) and reacted rabbit anti-uristatin antiserum at a dilution of 1:250,000 at 4 °C overnight with gentle agitation. The membrane was incubated with the alkaline phosphate substrate and color developed and recorded. (See on- line supplement for Western Blot Methods for additional details at www.extras.springer.com/2015/978-3-319-21926-4 )
7.2.2 Sample and Data Bank
All clinical and diagnostic data was collated into one database from four previous studies. The combined population (n = 6294) was intended to mimic the natural prevalence of renal disease (3 %), urinary tract infection (3 %), nephritis (15 %) and systemic infection (2 %) expected in a general population without exclusion or selection criteria. Clinical samples stored at −70 °C were thawed and used once for additional testing for uTi by the Uri ELISA immunoassay methods. Uri values by ELISA of <12 mg/L was considered normal. Test strips for urinary tract infection had been tested in original studies. Here a Uri by dipstick of < = 7.5 mg uristatin/g creatinine or <12.5 mg/L Uri was considered normal.
The retested samples from the original cohorts are summarized in Table 7.2 along with the diagnosis of sub-groups within each and criteria used. Urinalysis strip testing included pads for detection of albumin, protein, creatinine, occult blood, leukocytes, glucose, and ketone (MULTISTIX PRO®, Siemens Healthcare Diagnostics). Potential nephritis was indicated with as a urinalysis strip with a posi-tive for albumin, protein or occult blood.
7.2.3 Health Screen Collection
Random urine specimens were collected from a Japanese health screening study as part of an annual physical examination by a General Practice physician (Pugia et al. 2002 ). Of the original study samples, 4022 were available. Of these, 3665 were
S.A. Jortani and M. Pugia
135
urine samples were from apparently healthy patients, 292 were urine samples from patients with potential renal disease with previous abnormal blood urea nitrogen (BUN) or serum creatinine and were follow-up visits, and 205 urine samples patients with diagnoses of potential infects and/or fever (> = 38 o C) were followed up erythrocyte sediment rate (ESR) or abnormal serum C-reactive protein (CRP).
Table 7.2 Clinical sample and data bank used for uTi immunoassay testing
Study Diagnosis a Urinalysis Criteria b Additional criteria used c
Health screen (n = 4022)
Normal (n = 3665) Urinalysis strip negative
none
Asymptomatic GN(n = 292) Urinalysis strip positive
Neg. BUN, CRE
Bacteriuria (n = 3) Urinalysis strip positive
Pos. urine culture
Systemic infection (n = 62) Not used Pos. CBC, ESR, CRP or fever
Urinary tract infection (n = 1743)
Normal (n = 788) Urinalysis strip negative
Neg. urine culture
Symptomatic GN (n = 731) Urinalysis strip positive
Neg. urine culture
Bacteriuria (n = 224) Not used Pos. urine culture Hospital infections (n = 341)
Normal (n = 137) Urinalysis strip negative
Neg. CBC, CRP & bloodculture
Asymptomatic GN (n = 67) Urinalysis strip positive
Neg. CBC, CRP, blood culture
Bacteriuria (n = 77) Not used Pos. urine culture Systemic infection (n = 58) Not used Pos CBC, CRP or blood
culture Hospital patients (n = 188)
Normal (n = 137) Urinalysis strip negative
Neg CBC
Asymptomatic GN (n = 46) Urinalysis strip positive
Neg CBC
Bacteriuria (n = 3) Not used Pos. urine culture Systemic infection (n = 2) Not used Pos CBC
Legend a Diagnosis follows typical standard of practice with appropriate follow up testing and physician diagnostic judgment based on these tests b Urinalysis strips were run on all samples, and used as fi rst test result and additional follow up testing used for diagnosis . For indication of nephritis, only the protein and occult blood result were used. For normal, all results were considered c In cases of pre-existing renal disease, the strip result was not used for diagnosis and blood results are used. In the case of infection the follow up urine culture and blood testing were used and strip
result was not used for diagnosis
7 Uristatin Immunoassay Usage in Glomerular Nephritis Assessment
136
7.2.4 Urinary Tract Infection Collection
Uri Immunoassays were run on 1743 of the clean catch urine samples were col-lected patients with suspected urinary tract infection (Pugia et al. 2004 ). These samples were consecutive samples sent for urine culture. Of the 1743 specimens, 244 had bacteruria with positive cultures for gram-positive bacteria and/or gram- negative bacteria at ≥10 5 organisms/ml. Urine sediments were centrifuged and examined by microscopic to estimate of the counts of erythrocytes, leukocytes, number and types of casts, epithelial cells, crystals, clarity, color, and mucus (Freeman and Beeler 1983 ; Schumann 1996 ). A fi nding of yeast was considered as a negative result for bacterial infection. Of the 1743 specimens, 19 had renal disease based on urine microscopic (renal cells cast) and gross proteinuria.
7.2.5 Hospital Infection Collection
Immunoassays for Uri were run on 341 collected patients with presenting with upper respiratory infection and several other systemic types of infection or pre-sumed healthy (Jortani et al. 2004 ). Matched blood samples were collected and all samples test by CBC, CRP, and erythrocyte sedimentation rate (ESR). Patient with infections were diagnosed as bacterial, probable bacteria, viral and probable viral by two reviewers (board-certifi ed microbiologist/pathologist and a clinical chemist) in a double blind procedure. Blood culture and urine culture were run for all patients with suspected infection. Patients having been diagnosed with HIV, leukemia, para-sitic or yeast infections were excluded from this collection.
7.2.6 Hospital Patient Collection
Immunoassays for Uri were run on 188 patients presenting with cardiovascular dis-ease or presumed healthy (See Chap. 9 ). Cardiology work up was provided for all patients and match blood and urines samples were collected. All samples were test by CBC and CRP and systemic infections identifi ed (n = 2). Chart review only indi-cated 1 with pre-existing renal disease patients.
7.3 Results
Uri and Bik were commonly found by western blot in urines collected from diabet-ics with glomerular nephritis (See on-line supplement for example data at www.extras.springer.com/2015/978-3-319-21926-4 ). No detectable urinary P-a-I or I-a-I
S.A. Jortani and M. Pugia
137
was found in any urine by western blots. Smaller Uri fragments were found most diabetics with chronic kidney disease (CKD) or urinary tract infection. The Uri immunoassay agreed with western blot and allowed measurements of both Uri and Bik to be made on large sample banks.
Uri immunoassay results for 6292 urine samples were used to test for correlation to glomerular nephritis (Table 7.3 ). Urine samples from health patients (n = 4747) lacked Uri, with 95.7 % of patients sample having less than 12 mg/L of Uri and 98.7 % less than <25 mg/L. These patients all were negative for glomerular nephri-tis by the criteria of lacking proteinuria or hematuria or CKD. Hospital patients had other conditions such as cardiovascular diseases, cancers, and metabolic diseases (e.g. diabetes, etc.) but lacked known infections. Cases with suspected urinary tract infections were considered normal if negative by urine bacteria culture (<10 4 cell/mL). Cases with suspected systemic infections were considered normal by CBC and CRP. The high % of true negatives by Uri immunoassay makes it suitable for follow up of proteinuria or hematuria for general population and gives a very high negative predictive value.
Suspected infections are followed-up regardless of the Uri immunoassay result. In the presence of systemic bacterial infection or UTI, the Uri can be positive due to the innate anti-infl ammatory response to the infection. Uri and Bik abnormally elevate when white blood cells in blood and urine are clearly abnormally elevated. Elevated WBCs are characteristic of acute systemic bacterial and urinary tract infections and an infection work up is required (Table 7.4 ). Uri immunoassay results were ≥12 mg/L in 35 % of UTI patients and ≥25 mg/L in 12 %. Of these UTI patients, 84 % were
Table 7.3 Uri immunoassay results for glomerular nephritis follow up
Diagnosis Uristatin <12 mg/L
Uristatin > = 12 mg/L
Uristatin <25 mg/L
Uristatin > = 25 mg/L
Normal (n = 4747) 4544 203 4683 64 Asymptomatic Glomerular nephritis (n = 385)
351 34 378 7
Symptomatic Glomerular nephritis (n = 731)
533 198 648 83
Urinary Tract Infection (n = 307) 200 107 257 50 Systemic bacterial infection (n = 122) 53 69 86 36 total 5681 611 6052 240
Table 7.4 Expected Uri during infection follow up
Condition Uristatin Urine culture Blood test CBC Blood test CRP
Normal neg neg neg neg Viral Infection neg neg pos pos Bacterial Infection pos neg pos pos Urinary tract infection pos pos neg neg Glomerular nephritis pos neg neg neg
7 Uristatin Immunoassay Usage in Glomerular Nephritis Assessment
138
either uTi, WBC, or NIT strip positive. In systemic bacterial infection patients, Uri values were ≥12 mg/L in 57 % of patients and ≥25 mg/L in 41 % of patients. CRP and CBC did detect systemic infection but did not detect UTI. Follow up for infection with any positive Uri value is recommended (Fig. 7.5 ). Uristatin’s ability to detect infections helps kidney disease management, as these infections are known to damage kidneys. Management of infections is part of typical standards of care for diabetics.
Uri immunoassay results, unlike CRP, were not elevated by viral infections. Common viral infections (CBC negative for granulocytosis) did not elevate uTi, such as viral tonsillitis (Adenovirus), pneumonia, enterocolitis, gastroentritis, viral menin-gitis & bronchiolitis (viral respiratory), mumps (Rubulavirus) varicella (chickenpox), rotavirus, infl uenza (A/B), measles (rubeola) mononucleosis (EBV), hepatitis, and herpes zoster. Uri was also not elevated in conditions such as asthma, allergies or auto-immune diseases like systemic lupus erythematosus or amyloidosis. However several conditions do interfere with Uri usage. AIDS/HIV patients can be falsely negative for Uri. Uri is elevated in cancers which increase white blood cells like multiple myeloma.
No
Yes
Yes
No
No Yes
Infection oruristatinIndicated?
All patients (n=6292)
SymptomaticGlomerular
Nephritis with Anti-inflammatory
response (n=198)
UristatinStill positive ?
GomerularNephritis (n=1407)
Potentialinfection (n=1194)
Resolved infection (n=419)
Proteinuria or Hematuria ?
AsympotmaticGomerular
Nephritis (n=351)
Normal (n=4887)
Follow up forinfection
SymptomaticGlomerular
Nephritis(n=533)
Fig. 7.5 Flow chart. Proteinuira or hematuria was an indicator of glomerular nephritis in 1407 of 6292 patients tested. Of these, only 351 patients had asymptomatic GN. The remaining 1194 patients had potential infections based on symptoms of urinary tract infection, systemic infections, and/or a positive Uri value >12 mg/L. These were all followed by urine culture and CBC. This resulted in 419 patients as needing to be treated for infections. Those remaining were sorted according to Uri value as 198 cases of symptomatic GN with anti-infl ammatory stress and 533 cases without this stress. Proteinuria or hematuria is often resolved with anti-infl ammatories, how-ever persistent proteinuria or hematuria in absence of infection continues in progressing glomeru-lar nephritis. Patients with positive Uristatin indicate with atypical anti-infl ammatory stress
S.A. Jortani and M. Pugia
139
Predicting progression of patients with glomerular nephritis (GN) to CKD is an important unmet need. An initial fi nding of GN based on proteinuria or hematuria in urinalysis requires signifi cant follow up (Pugia et al. 2000 ). In diabetes, the prev-alence of diabetic hematuria or proteinuria is as high as 40 %. However, progression to CKD occurs in only <13 % of cases and end stage renal disease (ESRD) occurs in <6 % of population (Park 2004 ). As infl ammation becomes more chronic, the anti-infl ammatory response increases and Uri values become abnormally elevated. Patients with asymptomatic GN who have hematuria or proteinuria start to report the classical symptoms of chronic infl ammation with pain and swelling. Uri immu-noassay results were ≥12 mg/L in 9 % of asymptomatic GN patients and ≥25 mg/L in 2 %. Patients with symptomatic GN were Uri positives with ≥12 mg/L in 27 % of cases and ≥25 mg/L in 11 %.
The anti-infl ammatory status in these 27 % of patients with elevated Uri is expected to impact the prognosis. While Uri is renal protective, a population with chronic exposure to infl ammation is expected to have reduced kidney regeneration due to anti-proliferative cell signaling of Uri (Chap. 10 ). Uri does protect kidney damage by the white blood cells until tissue could clear the WBCs after infection is addressed. The immune response often continues in the absence of detectable infec-tion. Uri also inhibits proteolysis in the renal tubular cell, which correlates with reduced re-absorption and adds additional proteinuria. Controlling the action of this cascade would seem necessary to prevent further damage to the normal tissues. Patients with symptoms of chronic urinary infl ammation without overt infection are diffi cult to resolve. Here, Uri allows identifi cation of patients under anti- infl ammatory stress (Fig. 7.5 ).
7.4 Conclusion
In this report, we have shown the value of urinary trypsin inhibitors (uTi) as a mea-sure of anti-infl ammatory stress during glomerular nephritis (GN). In management of diabetic GN today, yearly urinalysis screening is recommended with identifi ed positives follow up regularly with repeat testing. Once infection is ruled out, patients in the chronic infl ammatory phase without active infection are not further assessed. Use of Uri as a measure of chronic anti-infl ammatory stress is a possible tool to assess the risk but requires follow up studies to determine the impact on progression to CKD. The Uri immunoassay could aid in the development of drugs for glomeru-lar nephritis by allowing screening of patients for renal regeneration therapies, immune cell suppression therapies and to allow identifi cation of non-responders.
Acknowledgements We would like to thank Ronald Sommer, Dr Hai-Hang Kuo, and Dr. Manju Basu for their many hours of effort in performing uristatin ELISA testing. We would also like to thank, Dr. Roland Valdes, Mr. Glen Renfrew, Dr. Darryl L. Gopual, Dr. John Lott and Dr. Chris Price for their helpful guidance to build this patient sample bank, with helpful advice over many years and the various studies.
7 Uristatin Immunoassay Usage in Glomerular Nephritis Assessment
140
Supplements
• ELISA Method for uristatin • Western Blot methods
References
Barnett BJ, Stephens DS (1997) Urinary tract Infection: an overview. Am J Med Sci 314:245–249
Bonventre JV, Weinberg JM (2003) Recent advances in the pathophysiology of ischemic acute renal failure. J Am Soc Nephrol 14(8):2199–2210
Cameron JS (2003) Focal segmental glomerulosclerosis in adults. Nephrol Dial Transplant 18(Suppl 6):45–51
Cochrane CG, Unanue ER, Dixon FJ (1965) A role of polymorphonuclear leukocyte and comple-ment in nephrotoxic nephritis. J Exp Med 122:99–116
Coelho AM, Ossovskaya V, Bunnett NW (2003) Proteinase-activated receptor-2: physiological and pathophysiological roles. Curr Med Chem Cardiovasc Hematol Agents 1:61–72
Couser WG (1998) Pathogenesis of glomerular damage in glomerulonephritis. Nephrol Dial Transplant 13(Suppl 1):10–15
Donath MY et al (2013) Infl ammation in obesity and diabetes: islet dysfunction and therapeutic opportunity. Cell Metab 17(6):860–872
Eguchi Y (2003) DIC-induced organ failure. Igaku Ayumi 206:19–22 Freeman JA, Beeler MF (1983) Laboratory medicine/urinalysis medical microscopy, 2nd edn. Lea
& Febiger, Philadelphia, pp 553–555, 81–110 Hatano M (1984) Clinical studies on primary glomerular disease. Jpn J Med 23(3):264–267 Hooton TN, Stamm WE (1997) Diagnosis and treatment of uncomplicated urinary tract infection.
Urinary Tract Infect 1:551–581 Jortani SA et al (2004) A sensitive non-invasive marker for diagnosis of infection and infl amma-
tion. J Clin Lab Anal 18:289–295 Kobayashi H et al (2000) Identity of urinary trypsin inhibitor-binding protein to link protein. J Biol
Chem 275:21185–21191 Koizumi R, Kanai H, Maezawa A, Kanda T, Nojima Y, Naruse T (2000) Therapeutic effects of
ulinastatin on experimental crescentic glomerulonephritis in rats. Nephron 84:347–353 Kuwajima S et al (1989) Urinary trypsin inhibitor and its clinical usefulness for diagnosis of acute
phase reactant and renal disease. Ensho 9:175–182 Kuwajima S et al (1992) Automated measurement of urinary trypsin inhibitor, an acute phase
reactant in urine. International Congress Series. Prog Clin Biochem 991:317–320 Larkin JG, Frier BM, Ireland JT (1985) Diabetes mellitus and infection. Postgrad Med J
61(713):233–237 McCluskey RT (1970) Evidence for immunologic mechanisms in several forms of human glo-
merular diseases. Bull N Y Acad Med 46(10):769–788 Mizon C et al (2002) Urinary bikunin determination provides insight into proteinase/proteinase
inhibitor imbalance in patients with infl ammatory diseases. Clin Chem Lab Med 40:579–586 Murakami M (1988) Effect of protease inhibitors on endotoxin-induced disseminated intravascular
coagulation in rats. Kyoto Ika Daigaku Zasshi 97:1155–1165 Nakakuki M, Yamasaki F, Shinkawa T et al (1996) Protective effect of human ulinastatin against
gentamicin-induced acute renal failure in rats. Can J Physiol Pharmacol 74:104–111 Nakatani K, Takeshita S, Tsujimoto H, Kawamura Y, Sekine I (2001) Inhibitory effect of serine
protease inhibitors on neutrophil-mediated endothelial cell injury. J Leukoc Biol 69:241–247
S.A. Jortani and M. Pugia
141
Onbe T, Makino H, Haramoto T et al (1991) Effect of protease inhibitor on primary glomerulone-phritis and the mechanism of the effect. Nihon Jinzo Gakkai Shi 33:753–759
Park CW (2004) Diabetic kidney disease: from epidemiology to clinical perspectives. Diabetes Metab J 38(4):252–260
Pickering MC et al (2013) C3 glomerulopathy: consensus report. Kidney Int 84(6):1079–1089 Pugia MJ, Lott JA (2005) Pathophysiology and diagnostic value of urinary trypsin inhibitors
(review article). Clin Chem Lab Med 43:1–16 Pugia MJ et al (1999) Screening school children for albuminuria, proteinuria and occult blood with
dipsticks. Clin Chem Lab Med 37:49–57 Pugia MJ et al (2000) Screening for kidney disorders in school aged children with urine reagent
strips. Clin Chem Lab Med 38:975–982 Pugia MJ et al (2002) Clinical utility of a rapid test for uristatin. Clin Biochem 35:105–110 Pugia MJ et al (2004) The uristatin dipstick is useful in distinguishing upper respiratory from uri-
nary tract infections. Clin Chim Acta 341:73–81 Pugia MJ et al (2007a) Immunological evaluation of urinary trypsin inhibitors in blood and urine:
Role of N- & O-linked glycoproteins. Glycoconjugate J 24(1):5–15 Pugia MJ, Valdes R, Jortani SA (2007b) Chapter on urinary trypsin inhibitors: structure, biological
relevance and measurement. In: G Makowski (ed) Advances in clinical chemistry, vol 44. Elsevier, Morristown, NJ, USA, ISBN- 10: 0-12-373704-4
Saint S, Scholes D, Fihn SD, Farrell RG, Stamm WE (1991) The effectiveness of a clinical practice guideline for the management of presumed uncomplicated urinary tract infection in women. Am J Med 106:635–641
Schumann GB (1996) Comparing slide systems for microscopic analysis. Lab Med 27:1–5 Takahashi M, Sawaguchi T, Sawaguchi A, Suzuki T (2001) The cytoprotective effect of protease
inhibitor on programmed cell death of endothelial cell. Tokyo Joshi Ika Daigaku Zasshi 71:669–678
Takemura T et al (1994) A clinical study of urinary trypsin inhibitor, an acute phase reactant in urine, in acute pediatric infectious diseases. J Infl amm 14:53–57
Trefz G et al (1991) Establishment of an enzyme-linked immunosorbent assay for urinary trypsin inhibitor by using a monoclonal antibody. J Immunoassay 12:347–369
Tull SP et al (2012) PR3 and elastase alter PAR1 signaling and trigger vWF release via a calcium- independent mechanism from glomerular endothelial cells. PLoS One 7(8):e43916
Vesey DA et al (2013) PAR2-induced infl ammatory responses in human kidney tubular epithelial cells. Am J Physiol Renal Physiol 304(6):F737–F750
Yamaguchi Y, Ohshiro H, Nagao Y, Odawara K, Okabe K, Hidaka H et al (2000) Urinary trypsin inhibitor reduces C-X-C chemokine production in rat liver ischemia/reperfusion. J Surg Res 94:107–115
Yoshida M, Honma S (2014) Regeneration of injured renal tubules. J Pharmacol Sci 124(2):117–122
7 Uristatin Immunoassay Usage in Glomerular Nephritis Assessment
143© Springer International Publishing Switzerland 2015 M. Pugia (ed.), Infl ammatory Pathways in Diabetes: Biomarkers and Clinical Correlates, Progress in Infl ammation Research, DOI 10.1007/978-3-319-21927-1_8
Chapter 8 Acute Response of Uristatin in Surgery
Mitchell H. Rosner and Michael Pugia
Abstract Prediction of those patients at highest risk for the development of acute kidney injury (AKI) allows for institution of specifi c kidney protective strategies. Novel urinary biomarkers could serve as important pre-exposure risk stratifi cation tools. In this study, we identify a novel urinary biomarker, uristatin, that has the abil-ity to aid in the determination of the risk of AKI in patients undergoing cardiac bypass (CPB) surgery in a single center. Thirty-six patients were enrolled and underwent CPB surgery. Fifteen patients developed AKI and pre-operative predic-tors of AKI on univariate analysis included: male gender (OR 6.88 (CI 1.56, 30.36); ejection fraction <35 % (OR 3.18 (CI 1.37, 7.38)) and CPB time >85 min (OR 2.54 (CI 1.183, 5, 47). On multivariate analysis, those patients with a urinary uristatin level <40 mg/L had an OR for developing AKI of 3.41 (CI 1.23, 9.44). There was an inverse correlation between pre-operative uristatin levels and the risk of AKI with those patients having a baseline level >80 mg/L having at least a fi vefold lower risk of AKI than those patients with baseline levels <40 mg/L. Urine uristatin as a pre- operative predictive marker for development of AKI post-operatively had an area under the curve (AUC) of 0.855. In this cohort, pre-operative levels of either urine or serum IL-6 were not predictive for the development of AKI. Urine uristatin levels may allow for better discrimination of pre-operative AKI risk and allow for targeted preventative strategies
M. H. Rosner , MD, FACP (*) Division of Nephrology, Department of Medicine , University of Virginia HSC , Charlottesville , VA 22908 , USA e-mail: [email protected]
M. Pugia Strategic Innovation , Siemens Healthcare Diagnostics , 3400 Middlebury Street , Elkhart , IN 46530 , USA e-mail: [email protected]
144
8.1 Introduction
Acute kidney injury (AKI) can occur in critically ill patients (Uchino 2005 ), after surgery (Rosner and Okusa 2006 ) and due to drug or contrast agent nephrotoxicity (Park et al. 2010 ; Alexopoulos et al. 2010 ; Nadim 2008 ). Defi ning patients at high risk for AKI has typically relied on risk stratifi cation tools that utilize a combination of clinical and laboratory parameters to defi ne a risk score. However, these scoring systems typically have good negative predictive power but suffer from poor positive predictive power. Biomarker-based risk profi les have sought to improve predictive power (Rosner 2009 ).
An ideal setting for testing such risk profi ling may be in those patients undergo-ing cardiopulmonary bypass surgery (CPB). Various risk assessment tools have been developed and validated for specifi c clinical scenarios: for instance, pre- cardiac catheterization or pre-CPB surgery (Thakar et al. 2005 ; Chertow et al. 1997 ). AKI occurs in about 18 % of CPB patients and 2–6 % require hemodialysis (Khilji and Khan 2004 ). In all settings, AKI is associated with signifi cant morbidity and mortality and efforts to fi nd therapeutic agents to prevent or hasten resolution of AKI have been largely unsuccessful (Bagshaw et al. 2005 ; Hoste et al. 2003 ). Reliance on serum creatinine for the diagnosis of AKI has hampered the use of therapeutics (Jo et al. 2007 ; Star 1998 ; Liu and Glidden 2009 ). Serum creatinine assessments in the fi rst 24 h do not predict the glomerular fi ltration rate (GFR) (Soni et al. 2009 ; Moore 2010 ). Serum creatinine tends to rise 24–48 h after surgery and too late for therapeutic intervention.
Recently, several pro-infl ammatory biomarkers have shown promise for the early detection of AKI (Kinsey et al. 2008 ). Pro-infl ammatory cytokines such as interleu-kin (IL) play a role in infl ammatory processes involved in the development of AKI when measured post-surgically but not pre surgically (Liu 2009 ; Parikh et al. 2004 ). Neutrophil Gelatinase-Associated Lipocalin (NGAL) released from immune cells during infl ammation precedes the rise in serum creatinine hours after surgery in patients progressing to AKI (Haase et al. 2009 ; Bennett et al. 2008 ; Mishra et al. 2005 ). Acute tubular necrosis markers such as Adenosine Deaminase Binding Protein (ABP-26) and liver type Fatty Acid Binding Protein (L-FABP) are useful measures of renal injury from nephrotoxin exposure (Nakamura et al. 2006 ; Doi et al. 2010 ; Mueller et al. 1990 ; Thompson et al. 1985 ; Moore 2010 ). Cystatin C and Kidney Injury Molecule 1 (KIM) have shown benefi ts for detection of pre-existing GFR dysfunction over creatinine (Koyner et al. 2008 ; Bonventre 2008 ).
Anti-infl ammatory proteins such as Adiponectin (Adpn) and have also been cor-related to renal status (Shen et al. 2008 ). Urinary trypsin inhibitors (uTi) such as uristatin (Uri) and bikunin (Bik) play a key role in AKI by halting immune medi-ated infl ammatory processes (Pugia et al. 2007a ). These serine protease inhibitors reduce cytokine and growth factor release during stresses such as renal ischemia by shutting down the trypsin family of serine proteases such as trypsin, elastase, kal-likrein, chymotrypsin, plasmin and cathepsin (Pugia et al. 2007b ; Fries and Blom 2000 ). uTi are readily fi ltered by the glomeruli and are increased in the urine of
M.H. Rosner and M. Pugia
145
patients with most infl ammatory disorders (Piette et al. 1992 ),bacterial infections (Pugia et al. 2002 ), neoplasia (Takubo et al. 2001 ), and kidney stones (Moriyama et al. 2001 ).
The release of elastase and cathepsin G from immune cells is implicated in base-ment membrane damage (Johnson et al. 1998 ; Takahashi et al. 2001 ). uTi have been shown to protect endothelial cells during pre-eclampsia (Maradny et al. 1994 ), sup-presses neutrophile elastase and protects respiratory function in CPB patients (Sugita et al. 2002 ), have a protective effect against disseminated intravascular coagulation (DIC) (Murakami 1988 ) and hepatic microvascular injury (Itabashi et al. 2003 ). As such, uTi represent a potential target for biomarker development based upon a putative pathophysiological role. As an acute-phase indicator, uTi are generated rapidly owing to the pro-inhibitor inter-α-inhibitor being available in blood prior to the time of injury and before surgery due to innate response to organ damage. Patients with impaired innate response have lower uTi values (Masaki et al. 2003 ). Surgery induces an increase in uTi which parallels tissue damage (Endo 2003 ; Masaki et al. 2003 ). uTi usually continue to rise during the course of trauma. The amount of urine uTi excreted typically increases 24 after the surgery to a peak on about the third day. By the seventh day, uTi usually decreases to its basal value.
Our goal was to test the Uri immunoassay which detected Uri and Bik in urine samples using a CPB patient cohort for risk prediction for the development of AKI with pre and post-operative samples. An improved prediction model for the proba-bility of developing AKI after cardiopulmonary bypass was tested. Comparisons were made between several biomarkers.
8.2 Methods
8.2.1 Patient Population
The patient population, was enrolled prospectively, and is described in Table 8.1 . The protocol was reviewed and approved by the Institutional Review Board of the University of Virginia. We screened 78 patients and enrolled 36 patients undergoing CPB at a single institution. Reasons for screening failure included: failure to obtain consent (16 patients), baseline serum creatinine >2.0 mg/dL (12 patients), WBC >10,000/mm 3 (4 patients), recent IV contrast exposure (4 patients), and post- operative anuria (1 patient). AKI as defi ned as a persistent (>48 h) rise in serum creatinine >0.3 mg/dL occurred in 15 patients (42 %) within the 48 h post- CPB. Based upon the RIFLE criteria (using serum creatinine criteria), 6 patients were in the Risk, 7 patients in the Injury and 3 patients in the Failure categories. The characteristics of the patients who developed AKI and those who did not are dis-played in Table 8.1 . While the baseline serum creatinine and eGFR were higher in the group that developed AKI, there was no statistically signifi cant difference between baseline renal function between the two groups.
8 Acute Response of Uristatin in Surgery
146
A preoperative risk assessment for the occurrence of AKI was performed using both the renal risk stratifi cation system of Chertow as well as the Cleveland Clinic Foundation (CCF) scoring system (excludes systemic infection) (Thakar et al. 2005 ; Chertow et al. 1997 ). Based upon these scoring system which utilize clinical vari-ables and baseline renal function, there was no statistically signifi cant difference in baseline risk between those patients who developed AKI and those who did not. However, there was a trend toward higher CCF scores in the group that developed AKI (see Table 8.1 ). Most patients who developed AKI had white blood cells in urine (n = 13) indicating an innate immune mediated infl ammatory reaction. Patients who did not develop AKI did not have WBC in their urine.
Exclusion criteria included: age <18 years, failure to obtain consent, recent (within 3 month) development of AKI (defi ned as a rise in serum creatinine greater than 0.3 mg/dL from a stable baseline creatinine if this data was available), recent infectious disease (defi ned as a documented fever >38 °C within 1 week of surgery, use of antibiotics within 7 days of surgery with the exclusion of pre- operative or intra-operative prophylactic antibiotics), or a white blood cell count (WBC) >10,000/mm 3 , end-stage renal disease, baseline serum creatinine >2.0 mg/dL, use
Table 8.1 Characteristics of the patient population
Variable Patients with AKI (n = 15)
Patients without AKI (n = 21) P value
Age in years 63.8 ± 7.88 67.4 ± 7.18 0.87 Gender (male (%)) 11 (73) 6 (29) 0.20 Diabetes (%) 11 (73) 15 (7) 0.76 Baseline sCr (mg/dL) 1.69 ± 0.44 1.25 ± 0.35 0.10 sCr at 36 h post-op (mg/dL) 2.26 ± 0.54 1.43 ± 0.36 0.04 Baseline eGFR (ml/min/1.73 m 2 ) 53 ± 15 74 ± 18 0.08 Left ventricular ejection fraction <40 % (%)
5 (33) 8 (38) 0.91
Peripheral vascular disease (%) 3 (20) 4 (19) 0.89 Baseline WBC (cells/mm 3 ) 8.7 ± 3.1 7.66 ± 2.6 0.54 Surgery type: CABG (%) Valve only (%) Combined (%)
7 (47) 5 (33) 3 (20)
12 (57) 7 (33) 2 (10)
0.56
CPB time (min) 118 ± 33 103 ± 29 0.64 Cross clamp time (min) 80 ± 12 70 ± 9 0.10 Use of pressors for >24 h post- operatively (%)
5 (33) 5 (24) 0.21
CCF risk score 6.9 ± 2.1 4.87 ± 1.9 0.06 Need for post-operative dialysis (%) 3 (20) 0 0.07 In-hospital death (%) 2 (13) 1 (5) 0.4
Legend: sCr serum creatinine, eGFR estimated glomerular fi ltration rate, CABG coronary artery bypass grafting, CPB cardiopulmonary bypass, AKI acute kidney injury, CCF Cleveland Clinic Foundation
M.H. Rosner and M. Pugia
147
of intra-aortic balloon pump, pre-operative use of inotropic agents or vasopressors, pre-operative cardiogenic shock, post-operative anuria and pre-operative use of intravenous contrast within 7 days of the surgery. At total of 78 patients were screened and 36 enrolled. Post-operatively all patient physiological parameters (blood pressure, pulse, temperature, urine output, use of inotropes/vasopressors) were recorded.
8.2.2 Sample Collection
Time matched blood and urine specimens were collected immediately pre- operatively, and then at 1–3 h, 5–7 h and 9–11 h on the day of surgery and every day thereafter out to 6 days post-operatively. Blood was analyzed for creatinine, electrolytes, white blood cell count, hemoglobin and biomarkers as described below. Duplicate urine and blood specimens were stored at −70 °C until analyzed for biomarkers. A protease inhibitor cocktail (contains 0.1 mg/mL Pefabloc SC® (Centerchem Inc, Norwalk, CT), 0.1 mg/mL leupeptin (SigmaAldrich, St. Louis MO), and 1 mM sodium azide (SigmaAldrich, St. Louis MO)) was added to each urine sample. Ejection fraction was estimated visually by surgeon. Surgical time for the bypass (min) and urinary output (ml/h) were recorded. Granular casts were measured on urine sample at time of collection by spinning the urine and viewing under high power light microscopy. Serum creatinine values and a complete white blood cell count (WBC) was obtained using specimens sent to the hospital laboratory. Duplicate specimens were stored at −70 °C until analyzer for remaining biomarkers.
8.2.3 Biomarker Analysis
Quantitative urinalysis for albumin and creatinine were measured using the DCA Vantage instrument (Siemens Healthcare Diagnostics, Tarrytown, New York). Urinalysis strip results for occult blood, creatinine, protein, albumin, nitrite, ketone, glucose, pH and luekocytes were measured using the CLINITEK Status instrument (Siemens Healthcare Diagnostics, Tarrytown, New York) with the MULTISTIX PRO® 10LS and CLINITEK Microalbumin®. Casts are determined by spinning the urine and viewing under high power light microscopy
Enzyme-linked immunosorbent assay (ELISA) measurements were made with commercially available kits. Urine and serum IL-6 used kits from E-Bioscience, San Diego, CA (Cat No. 88–706). Control and standards were preformed according to manufacturer’s instructions. Serum measurement of Cystatin C used kits from Biovendor Laboratory Medicine Inc, Czech Republic (Cat No RD191009100). Serum measurement of Adpn used kits from Panomics Inc, Fremont, CA (Cat No K1001-1). Urine measurement of NGAL used kits from Affi nity BioReagents Inc, Colden, CO (Cat No 037). All kits used monoclonal antibodies directed against the
8 Acute Response of Uristatin in Surgery
148
human peptide in a sandwich assay format with enzyme labels. Urine assays for Uristatin were performed by Brendan Bioscience (Hopewell MA) following litera-ture competitive antibody ELISA method (Pugia et al. 2007b ) (see Chap. 8 and Supplement Materials). Control and standards were preformed according to manu-factures instructions.
Urine assays for ABP-26 was performed by Brendan Bioscience (Hopewell, MA) following the literature ELISA Nephroscreen method (Thompson et al. 1985 ). Briefl y, ELISA plates (Corning®, 9018) were coated overnight at 4–8 °C with anti- ABP-26 (Uro 4 antibody, 8040–99, Covance) in bicarbonate buffer, pH 9.6 at 8 ug/ml. The plates were washed 2X with coating buffer then blocked with bicarbonate buffer containing 0.2 % bovine serum albumin (Sigma, A8022) for 2 h at 4–8 °C. The block was decanted and the plates were stabilized with 10 % sucrose for 1 h at room temperature (RT). The stabilizer was decanted and the plates were dried in vaco then sealed in a mylar pouch with desiccant. The detection conjugate was made by conjugating a second anti-ABP26 antibody (Uro-4a, 8045–99, Covance Inc) to HRP (Uro4a-HRP). The standard curve was constructed by diluting ABP26 (R&D Systems, 1180-SE) from 500 to 7.81 ng/ml in wash/sample diluent (W/SD) prior to adding it to the plate along with the samples which were diluted 1:100 in W/SD.
Urine samples and standards were diluted and 100 μL was added to each well and incubated for 4 h at RT. The fl uid was decanted and the plate was washed 3X. The Uro4a-HRP was added to each well and incubated for 2 h at RT. The con-jugate was decanted and the plate was washed 3X. The OPD substrate (Sigma, P-8412, 4 mg/ml in 100 mM citrate, pH 5.0) was added and incubated at RT for 20–30 min. The reaction was stopped using 3 N H 2 SO 4 and the absorbance was read at A492. A linear regression was performed using the data from the standard curve. The concentration of ABP-26 was calculated for each sample using the regression values derived from the standard curve.
8.2.4 Statistical Analysis
AKI was defi ned as a rise in serum creatinine >0.3 mg/dL occurring within 48 h after CPB surgery and sustained for >48 h. Patients were parsed into two groups (AKI v. no AKI) based upon meeting this criteria of AKI. SAS version 8.2 was used for analyses. Values for groups are shown in Tables 8.2 and 8.3 . To compare con-tinuous variables, we used a two-sample t test or Mann–Whitney rank sum test, and to compare categorical variables we used the χ 2 or Fisher’s exact test, as indicated. To measure the sensitivity and specifi city for urine uristatin and IL-6 at different cutoff values, a conventional receiver-operating characteristic (ROC) curve was generated for preoperative uristatin and IL-6. We calculated the area under the curve to ascertain the quality of uristatin as a biomarker. An area of 0.5 is no better than expected by chance, whereas a value of 1.0 signifi es a perfect biomarker. Univariate and multivariate stepwise multiple logistic regression analyses were undertaken to assess predictors of acute kidney injury. Potential independent predictor variables
M.H. Rosner and M. Pugia
149
Tabl
e 8.
2 Su
mm
ary
stat
istic
s fo
r th
e de
mog
raph
ic a
nd c
linic
al v
aria
bles
Clin
ical
va
riab
les
AR
F n
Mea
n SD
SE
M
edia
n G
M
Min
M
ax
Q25
Q
75
Age
0
22
63.8
2 7.
88
1.68
64
.00
63.3
6 51
.00
79.0
0 58
.25
67.0
0 1
15
67.4
0 7.
18
1.85
68
.00
67.0
4 56
.00
78.0
0 63
.50
72.5
0 E
ject
ion
frac
tion
0 22
40
.45
8.99
1.
92
37.5
0 39
.50
25.0
0 55
.00
35.0
0 45
.00
1 15
28
.67
11.4
1 2.
95
25.0
0 26
.89
15.0
0 55
.00
20.0
0 30
.00
Uri
ne b
ypas
s tim
e 0
22
70.0
5 8.
38
1.79
68
.50
69.6
0 60
.00
94.0
0 65
.00
73.5
0 1
15
80.1
3 12
.08
3.12
78
.00
79.2
7 60
.00
98.0
0 72
.00
89.5
0 U
rine
out
put
0 22
96
.36
23.9
6 5.
11
95.0
0 93
.62
60.0
0 15
0.00
80
.00
118.
75
1 15
87
.33
32.9
4 8.
51
80.0
0 76
.94
10.0
0 12
5.00
75
.00
120.
00
WB
C
0 22
7.
66
2.61
0.
56
7.35
7.
25
4.30
13
.40
5.48
8.
98
1 15
8.
69
3.07
0.
79
8.00
8.
27
5.60
16
.80
6.55
9.
65
8 Acute Response of Uristatin in Surgery
150
Tabl
e 8.
3 Su
mm
ary
stat
istic
s fo
r th
e m
ean
of th
e pa
tient
’s o
bser
vatio
ns u
p to
and
incl
udin
g th
e tim
e po
int w
hen
the
patie
nt w
as d
eter
min
e to
be
AR
F
Res
pons
e A
RF
n M
ean
SD
SE
Med
ian
GM
M
in
Max
Q
25
Q75
Seru
m c
reat
inin
e 0
22
1.43
0.
36
0.08
1.
45
1.39
0.
80
2.50
1.
23
1.50
1
15
2.26
0.
54
0.14
2.
10
2.20
1.
50
3.30
1.
95
2.40
Se
rum
IL
6 0
22
749.
68
949.
41
202.
41
313.
00
418.
21
68.9
0 37
51.0
0 19
0.50
70
9.75
1
15
836.
35
1021
.92
263.
86
327.
00
425.
07
83.2
0 29
09.0
0 16
7.00
86
5.00
U
rine
IL
6 0
22
567.
32
2321
.93
495.
04
58.1
0 57
.31
10.8
0 10
954.
00
18.2
3 82
.18
1 15
15
0.26
23
0.66
59
.56
43.2
0 53
.51
3.20
68
5.00
18
.85
150.
00
Uri
ne A
BP-
26
0 22
68
9.02
48
1.11
10
2.57
57
4.45
52
0.87
61
.10
1721
.80
337.
85
933.
30
1 15
50
3.88
56
5.47
14
6.00
37
9.00
29
7.86
23
.40
2352
.00
212.
50
605.
20
Uri
ne a
lbum
in
0 22
13
5.49
10
7.77
22
.98
79.4
0 99
.66
21.1
0 30
0.00
53
.28
267.
25
1 15
14
0.09
12
4.96
32
.26
67.8
0 82
.63
13.6
0 30
0.00
29
.65
300.
00
Uri
ne c
reat
inin
e 0
22
140.
64
40.8
7 8.
71
121.
50
135.
56
89.0
0 23
0.00
10
9.25
15
8.75
1
15
96.8
3 54
.43
14.0
5 85
.30
80.3
1 20
.90
204.
00
73.2
0 12
8.50
A
lbum
in/c
reat
ine
0 22
18
2.18
18
3.69
39
.16
127.
50
129.
21
30.0
0 87
3.00
73
.25
211.
50
1 15
34
0.47
43
0.72
11
1.21
16
9.00
16
3.26
23
.00
1350
.00
66.0
0 44
0.50
M.H. Rosner and M. Pugia
151
included age, sex, ethnic origin, cardiopulmonary bypass time, ejection fraction, urine albumin/creatinine, urine and serum IL-6, and urine uristatin. We judged p < 0 · 05 signifi cant
8.3 Results
8.3.1 Pre-operative Predictors of AKI
Thirty-six patients were enrolled and underwent CPB surgery. Fifteen patients developed AKI and pre-operative predictors of AKI on univariate analysis included: male gender (odds ratio (OR) 6.88 (CI 1.56, 30.36)); ejection fraction <35 % (OR 3.18 (CI 1.37, 7.38)) and CPB time >85 min (OR 2.54 (CI 1.183, 5, 47). The unad-justed odds of developing AKI based upon pre-operative clinical and laboratory features are shown in Table 8.4 . The pre-operative risk factors identifi ed are similar to other studies in this population (Thiele et al. 2014 ). On multivariate analysis, those patients with a urinary Uri level <40 mg/L had an odds ratio (OR) for develop-ing AKI of 3.41 (95 % confi dence interval 1.23–9.44).
Of the 36 patients 15 had abnormally high Uri prior to surgery (>12 mg/L) indi-cating pre-existing acute anti-infl ammatory response. Analysis of pre-surgical val-ues for other biomarkers tested including IL-6 (serum and urine), neutrophil gelatinase lipocalin (NGAL), dipeptidyl peptidase-4 (adenosine deaminase binding protein ADP-26), cystatin C, protienuria, microabluminuria, hematuria and urinary adiponectin, had no predictive value for the development of AKI.
Table 8.4 Odds of developing AKI according to baseline clinical and laboratory characteristics
Variable Unadjusted odds ratio Age (per decade) 1.812 (0.73, 4.52) Gender (male) 6.875 (1.56, 30.36) Race (African American) 0.458 (0.09, 2.45) Ejection Fraction (less than <35 %) 3.179 (1.37, 7.38) Cardiac Bypass Time (>85 min) 2.543 (1.183, 5.47) Urine Albumin/Creatinine >200 1.193 (0.94, 1.52) Serum IL-6 >500 pg/ml 1.021 (0.94, 1.11) Urine IL-6 >90 pg/ml 0.99 (0.99, 1.02) Urine Uristatin <40 mg/L 3.41 (1.23, 9.44) – – Variable Odds ratio adjusted for age, gender, ejection fraction,
bypass time Urine Albumin/Creatinine >200 1.159 (0.85, 1.58) Serum IL-6 >500 pg/ml 1.039 (0.78, 1.38) Urine IL-6 >90 pg/ml 0.99 (0.93, 1.042) Urine Uristatin <40 mg/L 1.262 (0.96, 1.624)
8 Acute Response of Uristatin in Surgery
152
Pre-operative levels of Uri did not correlate with the level of RIFLE criteria, however the numbers of individuals in higher levels of RIFLE AKI were small and thus this analysis is limited. This small number of patients limited the utility of any analysis. In an attempt to account for potential confounders in the relationship between baseline laboratory parameters and the development of AKI, a step-wise logistic regression analysis was performed utilizing the signifi cant clinical features found in the unadjusted analysis (see Table 8.4 ). This regression analysis does reduce the potential impact of pre-operative Uri and without additional sample sized the true predictive value cannot be uncovered.
8.3.2 Post-operative Prediction of AKI with Urinary Biomarkers
Urinary biomarkers were measured serially after CPB surgery. Neither urine uristatin nor serum and urine IL-6 measured at time periods immediately post- operative were predictive of rises in serum creatinine of greater than 0.3 mg/dL (data not shown). Uri continued to rise in patients after surgery to two to six times higher than pre-surgical values. A rise in Uri occurred in days 1–6 for all patients except 2 that failed to pro-duce Uri. At day 6, Uri values still did not return to normal but appeared to peak for most patients. This agrees with previous fi ndings that Uri is part of the acute infl am-matory response in wound healing increase during the fi rst week after major surgery.
Of the other biomarkers tested, neutrophil gelatinase lipocalin (NGAL) a mea-sure of immune cell activation and Cystatin C a measure of glomerular fi ltration response had the highest areas under the curve (AUC) of 0.6–0.7. Urinary Ig com-plement, a measure of auto-immune response, was not predictive. Adiponectin was more likely to be lower in the failure group but was not sensitive. There were only two failure patients with low adiponectin. Dipeptidyl peptidase-4 (adenosine deaminase binding protein ADP-26) a well-studied marker of tubular necrosis marker (cell disruption) was not sensitive for failure due to infl ammation. There were only fi ve failure patients with acute tubular necrosis. Protienuria, microablu-minuria, and hematuria were too sensitive and not specifi c.
8.3.3 Uristatin and Pre-operative Prediction of AKI
The performance characteristics of Uri for the pre-operative prediction of AKI are shown in Figs. 8.1 and 8.2 . Urine Uri as a pre-operative predictive marker for development of AKI post-operatively had an area under the curve (AUC) of 0.855 (see Fig. 8.1 ). In patients undergoing CPB-surgery there was an inverse correlation between pre-operative uristatin levels. The probability of AKI is shown in Fig. 8.2 along with the 95 % confi dence intervals. There is an inverse relationship between pre-operative urine uristatin levels and the risk for post-operative AKI and the
M.H. Rosner and M. Pugia
153
subsequent development of AKI. Those patients with urine uristatin levels >80 mg/L had at least a fi vefold lower risk of AKI as compared to those patients with baseline uristatin levels <40 mg/L. At urinary uristatin levels <40 mg/dL there is nearly a 50 % or greater risk of developing AKI, independent of inclusion of any
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
1.00.90.80.70.60.50.40.30.20.10.0
1 - Specificity
Sen
sitiv
ity
ROC Curve
Model ROC CurveRandom discrimination
A
Fig. 8.1 Receiver operator curve analysis of pre- operative urine uristatin levels for prediction of AKI
1.0
0.8
0.6
0.4
0.2
0.0
0 20 40 60 80 100
Maximum Uristatin (mg/L)Model ROC = 0.855
Pro
babi
lity
of A
PF
Fig. 8.2 Relationship between the probability of AKI and baseline uristatin level
8 Acute Response of Uristatin in Surgery
154
clinical variables. The predictive ability of this biomarker compares well to risk stratifi cation based upon clinical variables alone (for instance the AUC for the prediction of AKI for the CCF scoring system is 0.81 and 0.68 for the Chertow scoring system).
Based upon the known effects of infl ammation to cause AKI, it is not surprising that a urinary anti-infl ammatory protein may have either have protective effects or be a marker for renal protection (Bennett et al. 2008 ). It is interesting that another urinary marker that has been found to predict pre-operative risk (β-defensin-1) is also an anti-infl ammatory protein. Importantly, uristatin levels post-CPB surgery were not predictive of later AKI nor were serum or urine IL-6 levels. Thus, pre- operative urine uristatin levels may be helpful in identifying those patients at high risk of AKI and allow for targeted strategies of renal protection.
This study is limited by the size of the cohort and that the study occurred at a single-center. It should also be cautioned that this cohort of patients was selected carefully based upon stringent inclusion and exclusion criteria. Further studies in a larger multi-center cohort representative of the group of patients undergoing CPB surgery are required to validate these fi ndings as well as determine how a uristatin- based predictive model would operate in clinical practice and whether interventions based upon a pre-operative uristatin level would lead to renal protection. Furthermore, it is intriguing to speculate that recombinant uristatin or other protease-inhibitor therapies may offer therapeutic benefi t in the prevention of AKI. Future animal studies and trials will need to address these possibilities.
8.4 Conclusion
In this study, we show uristatin provides pre-operative risk assessment for the devel-opment of AKI post-CPB. The presence of uristatin in the urine and its potent anti-infl ammatory effects made it an ideal candidate biomarker for the assessment of pre-operative risk for AKI.
This study supports the premise that uTi protects patients from renal failure and decreases the likelihood of immune-mediated kidney injury. As uTi is induced to clinically signifi cant levels during acute events, it is an important fi nding for assess-ment of patient. Monitoring in an acute setting presurgically and after event for return to baseline could be helpful for predicting patient recovery. If uTi continues to be induced in chronically infl amed tissue, the prolonged protective state could slow renal regeneration (see Chap. 10 ).
Acknowledgements We thank Brent Dorval of Brendan BioScience (Hopedale, MA) for his contributions of Uristatin testing that made this study possible. We thank Dr Manju Basu who participated in the analysis of the urine and serum samples. We thank Glenn Fung of Siemens Healthcare Image and Knowledge Management, Malvern, Pennsylvania for additional statistical analysis.
M.H. Rosner and M. Pugia
155
References
Alexopoulos E et al (2010) Contrast-induced acute kidney injury in patients with renal dysfunction undergoing a coronary procedure and receiving non-ionic low-osmolar versus iso-osmolar con-trast media. Am J Med Sci 339(1):25–30
Bagshaw SM et al (2005) Prognosis for long-term survival and renal recovery in critically ill patients with severe acute renal failure: a population based study. Crit Care 9:R700–R709
Bennett M, Dent CL et al (2008) NGAL predicts severity of acute kidney injury after cardiac sur-gery: a prospective study. Clin J Am Soc Nephrol 3(3):665–673
Bonventre JV (2008) Kidney injury molecule-1; a specifi c and sensitive biomarker of kidney injury. Scand J Clin Lab Invest Suppl 241:79–83
Chertow GM, Lazarus JM, Christiansen CL, Cook F, Hammermeister KE, Grover F, Daley J (1997) Preoperative renal risk stratifi cation. Circulation 95:878–885
Doi K, Noiri E, Sugaya T (2010) Urinary L-type fatty acid-binding protein as a new renal bio-marker in critical care. Curr Opin Crit Care 16:545–549
Endo Y (2003) Antishock action of ulinastatin. Surg Trauma Immunol Respon 12:29–38 Fries E, Blom AM (2000) Bikunin: not just a plasma proteinase inhibitor. Int J Biochem Cell Biol
32:125–137 Haase M et al (2009) NGAL Meta-analysis Investigator Group. Accuracy of neutrophil gelatinase-
associated lipocalin (NGAL) in diagnosis and prognosis in acute kidney injury: a systematic review and meta-analysis. Am J Kidney Dis 54(6):1012–1024
Hoste EA et al (2003) Acute renal failure in patients with sepsis in a surgical ICU: predictive fac-tors, incidence, comorbidity, and outcome. J Am Soc Nephrol 14:1022–1030
Itabashi K et al (2003) Protective effects of urinary trypsin inhibitor on hepatic microvascular injury in hypotensive brain-dead rats. Transplant Proc 35:114–116
Jo SK, Rosner MH, Okusa MD (2007) Pharmacological treatment of acute kidney injury: why drugs haven’t worked and what is on the horizon. Clin J Am Soc Nephrol 2:356–365
Johnson RJ, Couser WG, Alpers CE, Vissers M, Schulze M, Klebanoff SJ (1998) The human neutrophil serine proteinases, elastase and cathepsin G can mediate glomerular injury. J Exp Med 168:1169–1174
Khilji SA, Khan AH (2004) Acute renal failure after cardiopulmonary bypass surgery. J Ayub Med Coll Abbottabad 16(3):25–28
Kinsey GR, Li L, Okusa MD (2008) Infl ammation in acute kidney injury. Nephron Exp Nephrol 109:e102–e107
Koyner JL, Bennett MR, Worcester EM, Ma Q, Raman J, Jeevanandam V, Kasza KE, O’Connor MF, Konczal DJ, Trevino S, Devarajan P, Murray PT (2008) Urinary cystatin C as an early biomarker of acute kidney injury following adult cardiothoracic surgery. Kidney Int 74(8):1059–1069
Liu KD, Glidden DV (2009) Clinical trials for acute kidney injury: design challenges and possible solutions. Curr Drug Targets 10:1190–1195
Liu KD et al (2009) Serum interleukin-6 and interleukin-8 are early biomarkers of acute kidney injury and predict prolonged mechanical ventilation in children undergoing cardiac surgery: a case–control study. Crit Care 13(4):R104
Liu KD (2014) Therapeutic strategies for clinical trials targeting clinical recovery. Nephron Clin Pract 127(1–4):113–116
Maradny E et al (1994) Urinary trypsin inhibitor may have a protective effect on endothelial cells in pre-eclampsia. Acta Obstet Gynecol Scand 73:755–757
Masaki M et al (2003) Changes in urinary trypsin inhibitor in blood and urine as well as serum cytokines in living related liver transplantation MASUI. Anesthesia 52:251–256
Mishra J et al (2005) Neutrophil gelatinase-associated lipocalin (NGAL) as a biomarker for acute renal injury after cardiac surgery. Lancet 365:1231–1238
8 Acute Response of Uristatin in Surgery
156
Moore E (2010) Biomarkers of acute kidney injury in anesthesia, intensive care and major surgery: from the bench to clinical research to clinical practice. Minerva Anestesiol 76(60):425–437
Moriyama MT, Glenton PA, Khan SA (2001) Expression of inter-α-inhibitor related proteins in kidneys and urine of hyperoxaluric rats. J Urol 165:1687–1692
Mueller PW et al (1990) Indicators of acute renal-transplant rejection in patients treated with cyclosporine. Clin Chem 36(5):759–764
Murakami M (1988) Effect of protease inhibitors on endotoxin-induced disseminated intravascular coagulation in rats. Kyoto Ika Daigaku Zasshi 97:1155–1165
Nadim MK (2008) An overview of drug-induced acute kidney injury. Crit Care Med 36(4 Suppl):S216–S223
Nakamura T et al (2006) Urinary excretion of L-type fatty acid binding protein in contrast medium- induced nephropathy. Am J Kidney Dis 47:439–444
Park WY et al (2010) The risk factors and outcome of acute kidney injury in the intensive care units. Korean J Intern Med 25:181–187
Parikh CR et al (2004) Urinary interleukin-18 is a marker of human acute tubular necrosis. Am J Kidney Dis 4:405–414
Piette A et al (1992) Urinary trypsin inhibitory activity for the diagnosis of bacterial infection: a prospective study of 690 patients. Eur J Med 1:273–276
Pugia MJ, Takemura T, Kuwajima S, Suzuki M, Cast TK, Profi tt JA, Schulman LS, Otta Y, Lott JA (2002) Clinical utility of a rapid test for uristatin. Clin Biochem 35:105–110
Pugia MJ, Valdes R Jr, Jortani SA (2007a) Bikunin (urinary trypsin inhibitor): structure, biological relevance, and measurement. Adv Clin Chem 44:223–245
Pugia MJ et al (2007b) Immunological evaluation of urinary trypsin inhibitors in blood and urine: role of N- & O-linked glycoproteins. Glycoconj J 24(1):5–15
Rosner MH, Okusa MD (2006) Acute kidney injury associated with cardiac surgery. Clin J Am Soc Nephrol 1(1):19–32
Rosner MH (2009) Urinary biomarkers for the detection of renal injury. Adv Clin Chem 49:73–97
Shen YY, Peake PW, Charlesworth JA (2008) Review article: Adiponectin: its role in kidney dis-ease. Nephrology 13(6):528–534
Star RA (1998) Treatment of acute renal failure. Kidney Int 54:1817–1831 Soni SS et al (2009) Early diagnosis of acute kidney injury: the promise of novel biomarkers.
Blood Purif 28(3):165–174 Sugita T et al (2002) Effect of a human urinary protease inhibitor (ulinstatin) on respiratory func-
tion in pediatric patients undergoing cardiopulmonary bypass. J Cardiovasc Surg 43:437–440 Takahashi M, Sawaguchi T, Sawaguchi A, Suzuki T (2001) The cytoprotective effect of protease
inhibitor on programmed cell death of endothelial cell. Tokyo Joshi Ika Daigaku Zasshi 71:669–678
Takubo T et al (2001) Urinary trypsin inhibitor levels in the urine of patients with haematological malignancies. Haematologia 31:267–272
Thakar CV, Arrigain S, Worley S, Yared J-P, Paganini EP (2005) A clinical score to predict acute renal failure after cardiac surgery. J Am Soc Nephrol 16:162–168
Thiele RH, Isbell JM, Rosner MH (2014) AKI associated with cardiac surgery. Clin J Am Soc Nephrol. pii: CJN.07830814. [Epub ahead of print]
Thompson RE et al (1985) Adenosine deaminase binding protein, a new diagnostic marker for kidney disease. Clin Chem 31(5):679–683
Uchino S et al (2005) Acute renal failure in critically ill patients: A multinational multicenter study. JAMA 294(7):813–818
M.H. Rosner and M. Pugia
157© Springer International Publishing Switzerland 2015 M. Pugia (ed.), Infl ammatory Pathways in Diabetes: Biomarkers and Clinical Correlates, Progress in Infl ammation Research, DOI 10.1007/978-3-319-21927-1_9
Chapter 9 Cardiovascular and Metabolic Syndrome Impact on Uristatin
Saeed A. Jortani and Michael Pugia
Abstract Urinary trypsin inhibitors (uTi) were modestly elevated in cardiovascular disease patients (CVD), coronary artery disease (CAD) or metabolic syndrome patients (MetSyn). Urinary and plasma immunoassay values for Uristatin (Uri) and Bikunin (Bik) were within their normal ranges for 94 % of CVD patients and 97 % of MetSyn patients. Urinary and plasma Bik and Uri values were also only slightly elevate with metabolic syndrome risk factors such as diabetes (DM), lipidemia (LIP), hypertension (HTN), and obesity (OBS). Bik and Uri were typically elevated in CVAD patients with clinically signifi cant events such as acute myocardial infarc-tion (AMI) or congestive heart failure (CHF). Here urinary trypsin inhibitors (uTi) are indicative of a white blood cell (WBC) response, as acute events also elevated granulocytes in blood. Granulocytes values signifi cantly correlated (P < .005) with the degree of coronary artery stenosis and coronary angiography. Generally patients with cardiovascular and metabolic conditions and without renal disease or infection had uTi values below the acute phase level (<12 mg/L).
9.1 Introduction
Cardiovascular disease (CVD) is associated with elevated lipids which are deposited in the arteries thereby causing atherosclerosis. Cardiovascular disease is also charac-terized by a number of metabolic syndrome (MetSyn) risk factors (Libby et al. 2002 ).
S. A. Jortani Department of Pathology and Laboratory Medicine , University of Louisville, School of Medicine , 511 S. Floyd Street (Room 227) , Louisville , KY 40202 , USA e-mail: [email protected]
M. Pugia (*) Department of Pathology and Laboratory Medicine , University of Louisville, School of Medicine , Louisville , KY USA
Strategic Innovation , Siemens Healthcare Diagnostics , 3400 Middlebury Street , Elkhart , IN 46530 , USA
e-mail: [email protected]
158
These indicators include; smoking, diabetes, sedentary lifestyle, hyperlipidemia, homocysteine levels and total cholesterol to HDL ratio. Patients with metabolic syn-drome have an a higher likelihood of fi nding cardiovascular disease (CVD), coro-nary artery disease (CAD), acute coronary syndrome (ACS). MetSyn also increases the chance of vascular damage, atherosclerosis, cardiomyopath, congestive heart failure (CHF) and acute myocardial infarction (AMI).
Accumulating epidemiological evidence currently indicates that infl ammation also plays a signifi cant role in the initiation and progression of CVD and CAD (Libby et al. 2002 ; Ridker et al. 2002 ; Danesh et al. 1998 ). Infl ammation plays a signifi cant role in the initiation and progression of all these conditions. As CVD and CAD progresses from MetSyn risk factors to full disease, the infl ammatory response further impacts the vascular system. The likelihood of fi nding CVD, CAD, and ACS is associated with increased basal concentrations of acute phase reactants, cytokines and adhesion molecules (Blake and Ridker 2003 ; Avanzas et al. 2004 ; Szmitko et al. 2003 ).
Atherosclerosis occurs after lipid accumulation causes the endothelial cells in the artery wall to express molecules that attract various classes of leukocytes lead-ing to plaque destabilization and rupture (Libby et al. 2002 ). The immune cell response mediates platelet activation, coagulation, and endothelial damage (Szmitko et al. 2003 ). Lipoprotein-associated phospholipase A2 (Lp-PLA2) has been linked to the oxidized low-density lipoprotein (oxLDL) activation of macrophages (Avanzas et al. 2004 ). Acute phase reactants, such as myeloperoxidase, placental growth factor, gelatinases and tissue factor pathway inhibitor have been linked to oxLDL uptake by macrophages and monocyte release of cytokines and chemokines (Heeschen et al. 2004 ; Brennan et al. 2003 ; Nguyen et al. 2001 ; Kato 2002 ). These markers have been used to estimate the immune cell recruitment, adhesion and acti-vation. The end result is formation of macrophage foam cells and ruptured plaques triggering thrombus potentially followed by ischemia, endothelial degradation or acute myocardial infarction (AMI).
Infl ammation increases the risk of AMI and stroke. Acute phase reactants such as high sensitivity C-reactive protein (CRP) have been used for cardiovascular risk assessment (Ridker et al. 1998a , b , 2000 ). The highest quartile of CRP has twice the risk of future stroke and three times the risk of future myocardial infarction (Ridker et al. 1997 ; Pasceri et al. 2000 ). Infl ammation markers are present in 10–15 % of events but provide increased hazard risk ratings of 2–4. Other complications such as arrhythmias and heart failure can follow these processes as well. While the amount and nature of infl ammation in a plaque is quite variable, persistently high or increas-ing concentrations of many acute phase biomarkers are indicators of a poor progno-sis and increase risk of development of an event.
Anti-infl ammatory responses by urinary trypsin inhibitors (uTi) such as Bikunin (Bik) and Uristatin (Uri) are typically present during infl ammation (Pugia and Lott 2005 ; Pugia et al. 2007a ). Serine protease inhibition by uTi impacts cell signaling and many of the processes known to arise in CVD and CAD. Inhibition serine pro-teases prevent signaling of Protease Activated Receptors (PAR). This subgroup of the G-protein-coupled receptors uses proteases to initiate cell signaling via specifi c
S.A. Jortani and M. Pugia
159
cleavage sites for thrombin, trypsin and possibly other serine proteases (Olear and Nouza 1999 ). PAR activation has been proposed as a key part of the vascular infl am-mation mechanism (Szmitko et al. 2003 ; Miike et al. 2001 ). PAR contribute to cel-lular changes as part of the chronic infl ammatory defense and precedes abnormal function (Cottrell 2002 ; Coelho et al. 2003 ). PAR also regulate vascular tone and participate in the response to vascular injury (Bono et al. 1997 ). uTi is shown to inhibit PAR activation during coronary artery bypass grafting surgery (CABG) (Shibutani and Kunihiro 1986 ). uTi also inhibits blood coagulation through action on plasmin, Factors IXa, Xa, XIa and XIIa (Nii et al. 1994 , 1995 ; Egeblad and Astrup 1966 ). Inhibition of plasmin by uTi prevents activation of Urokinase-type Plasminogen Activator and slows tissue remodeling. Finally, inhibition of serine proteases from white blood cells (WBC) prevents release of cytokines and growth factors, and signaling of apoptosis.
Urinary trypsin inhibitors have not been studied in progression of CVD or CAD. The impact of MetSyn risk factors is also unknown. Our goal was to measure the impact of vascular infl ammation and associated risk factors on the uTi measure-ments in blood and urine. We wanted to compare immunoassays for Bik, Uri and inter-α-inhibitor (I-α-I) to CRP, WBC and uTi inhibition assays in patients with and without vascular infl ammation and associated risk. Secondarily, we wanted to ret-rospectively compare the infl ammation markers with the degree of arterial block-age and by presence of AMI or CHF. Correlations between infl ammatory markers were tested to determine synergic and independent relationships. As a fi nal objec-tive, we also wanted to establish the normal ranges for the Bik, Uri and I-α-I immu-noassay in apparently healthy adults.
9.2 Methods
9.2.1 Patient Assessment
We collected matched urine, plasma and serum from 188 patients from several hos-pital settings including the cardiac unit. Patients were risk-stratifi ed. Those without angina pectoris (chest pain within the last 48 h), with no more than one cardiovas-cular disease risk factor, and no existing cardiovascular disease were considered normal and without coronary artery disease (n = 101). The remaining patients (n = 87) underwent coronary artery angiograms. Angiographic measurements were made of the degree of stenosis in the left anterior descending, left coronary artery, right coronary artery, ramus intermedius, graft-immediate distal vessel, and the cir-cumfl ex artery. Two board-certifi ed cardiologists reviewed the angiographs and grouped the patients according to their risk fi ndings. Angiogram patients were con-sidered normal (n = 32) when they uniformly had <30 % stenosis in all the coronary arteries and as affected with acute coronary syndrome when having at least >50 % stenosis in any artery (n = 55).
9 Cardiovascular and Metabolic Syndrome Impact on Uristatin
160
We followed the IRB practices of our institutions. The analysts were blinded to all test results. Specimens were collected during the initial assessment and prior to the angiographic evaluations. Most of the patients undergoing angiography had multiple cardiovascular risk factors. Data on age, gender, BMI, blood pressure, and plasma lipids were collected along with medical histories over the last 5 years. A systolic pressure of >140 mmHg, a diastolic pressure of >90 mmHg, or oral use of anti-hypertensive agents were used as an indication of hypertension (n = 57). We used a BMI cutoff of >30 ((weight in lb/(height in inches^2))*703) to defi ne over-weight (n = 60). Patients with hyperlipidemia (n = 100) were those with a ratio of cholesterol/ HDL >5 or if they were getting lipid-lowering agents such as a statin drug. Patients with diabetes mellitus (n = 28), and (or) smoker status (n = 69) were identifi ed and counted. While patients were not using anti-infl ammatory agents dur-ing the angiography and sample collection period, their history defi ned regular use, anti-infl ammatory medicines, anti-thrombotic or nitroglycerine.
Patients with infection(s), HIV, AIDS, or leukemia were excluded. One patient had an elevated CRP >10 mg/L and a WBC of >12.4 cell/nL indicating possible unknown infection. No patient had abnormal kidney function as judged by a creati-nine clearance of >100 mL/min. Clinical evaluation, and diagnosis of CHF (n = 7) was made by echocardiographic and B-type natriuretic peptide (BNP) of >100 ng/L assays (Remme 2002 ). A diagnosis of AMI (n = 10) was based on abnormal EKG fi ndings, troponin I >1.5 μg/L and clinical evaluation by our cardiologists. A diag-nosis of cardiovascular disease (AMI nor CHF) was not considered abnormal for CVD unless the angiogram was positive.
9.2.2 Sample Collection and Marker Analysis
We obtained clean-catch midstream urines specimens from all patients and con-trols. We stored the specimens at 4 °C for testing within 24 h. Storage was at – 20 °C for samples used in assays performed after 24 h. We collected plasma and serum from all patients and controls in Vacutainer® tubes (BD, Franklin Lakes, NJ). We separated the serum or plasma (EDTA) immediately after collection. We performed urine dipstick tests for urinary trypsin inhibitors by enzyme-inhibition assays on a Clinitek® 50 strip reader to obtain concentrations of 0 (undetectable), <6.25, 12.5, 25, and 50 mg/L by a method described elsewhere (Siemens Healthcare Diagnostics) (Pugia 2007b ). We measured for uristatin (Uri) and bikunin in blood and urine and I-α-I in blood by ELISA methods (Siemens Healthcare Diagnostics) (Pugia et al. 2007b ). Enzyme linked immunosorbent assays (ELISA) for adiponec-tin were performed with plasma according to manufacturers instructions (B-Bridge International, Inc Mountain View, CA). The MULTISTIX PRO® strip was used to measure blood, nitrite, leukocyte, protein, and creatinine ratios (Siemens Healthcare Diagnostics). The percent granulocytes and lymphocytes were determined on a Cell-Dyne 4000 cell counter, (Abbott Diagnostics, Santa Clara, CA). All CBCs were performed within 1 h of collection. We determined the total cholesterol, HDL, LDL, and triglycerides on a Vitros 950 analyzer (Ortho Clinical Diagnostics,
S.A. Jortani and M. Pugia
161
Raritan, NJ) along with a high sensitivity CRP test (BN II nephelometer Dade Behring, Deerfi eld IL). Analysis of plasma for B-type natriuretic peptide (BNP) and troponin I (TnI) were performed by immunoassay (ADVIA® Centaur™, Siemens Healthcare Diagnostics).
9.2.3 Patient Characterization
Comparisons of the cardiovascular risk factors, demographics, and medications are shown for the control (unaffected) population and the affected group with risk of CVD or CAD in Table 9.1 . Comparisons of the standard diagnostic tests are shown in Table 9.2 . The control population was characterized as either having no risk of CVD or CAD, if they had more than one risk factor, they had no evidence of steno-sis by angiogram. The affected group included at risk patients who underwent angi-ography with abnormal outcome. The affected group consisted of patients characterized at least >50 % stenosis in one artery and the expected differences in risk factors. We found that any chest pain did not signifi cantly predict the degree of vascular stenosis. All confi rmed patients with AMI had stenosis while most patients with CHF had stenosis. Medication usage was signifi cantly higher for the disease group than the control group regardless of gender.
9.2.4 Statistics Methods
Comparisons (as P -values) between these two groups, again for each sex (and both together) are included. For the qualitative factors (e.g., smoking status) the number, percent and the Clopper-Pearson 95 % confi dence interval for the percent are tabu-lated. The P -value is that for the difference in proportions between the control and affected groups calculated using the normal approximation to the binomial distribu-tion with continuity correction. For the quantitative factors (e.g. age and diagnostic test), the mean and standard deviations (SDs) are tabulated. The P -values are for the difference between each group’s mean assuming a t distribution. Before calculating the t value, the two groups’ SDs are compared using the F distribution. If they were not statistically different, (two-sided, 95 % confi dence), the SDs were pooled and the SD of the difference is estimated with the pooled SD. Otherwise, the SD of the difference is estimated from the two separate SDs and its number of degrees of freedom estimated using Satterthwaite’s approximation (rounded down). All P -values are two-sided. Differences between the control and affected groups are deemed statistically signifi cant if the corresponding P- values are less than 0.05. For Bik and the cell counts, exact non-parametric P- values are calculated using the Wilcox on scores (equivalent to the Mann-Whitney U test). For the cell counts, the exact P- values are estimated using a Monte Carlo approximation with 10,000 sam-ples and a random (system clock generated) seed. Again, all P- values are two-sided.
9 Cardiovascular and Metabolic Syndrome Impact on Uristatin
162
Tabl
e 9.
1 C
ohor
t dem
ogra
phic
cha
ract
eris
tics
Para
met
er
Con
trol
gro
up
Aff
ect g
roup
P-
valu
e, a
ffec
ted
vs. c
ontr
ol a , f
All
Mal
e Fe
mal
e A
ll M
ale
Fem
ale
All
Mal
e Fe
mal
e
– n
= 1
33
n =
54
n =
79
n =
55
n =
25
n =
30
– –
– –
(n, %
) (n
, %)
(n, %
) (n
, %)
(n, %
) –
– –
– Sm
oker
s,
34 (
25.6
%)
22 (
40.7
%)
12 (
15.2
%)
35 (
63.6
%)
17 (
68.0
%)
18 (
60.0
%)
<0.
005
0.02
<
0.00
5 H
yper
tens
ive b
27 (
20.3
%)
13 (
24.1
%)
14 (
17.7
%)
50 (
90.9
%)
22 (
88.0
%)
28 (
93.3
%)
<0.
005
<0.
005
<0.
005
Hyp
erlip
idem
ic c
55 (
41.4
%)
21 (
38.9
%)
34 (
43.0
%)
45 (
81.8
%)
19 (
76.0
%)
26 (
86.7
%)
<0.
005
<0.
005
<0.
005
Dia
betic
9
(6.8
%)
3 (5
.6 %
) 6
(7.6
%)
17 (
30.9
%)
7 (2
8.0
%)
10 (
33.3
%)
<0.
005
0.01
<
0.00
5 O
besi
ty d
30 (
22.6
%)
11 (
20.4
%)
19 (
24.1
%)
30 (
54.5
%)
11 (
44.0
%)
19 (
63.3
%)
<0.
005
0.03
<
0.00
5 C
HF
patie
nts
3 (2
.3 %
) 2
(3.7
%)
1 (1
.3 %
) 3
(2.3
%)
2 (3
.7 %
) 1
(1.3
%)
<0.
005
0.01
0.
48
Hav
e ha
d A
MI
1 (0
.8 %
) 0
(0.0
%)
1 (1
.3 %
) 6
(10.
9 %
) 4
(16.
0 %
) 2
(6.7
%)
<0.
005
<0.
005
0.13
A
nti-
imfl a
mm
ator
y m
edic
atio
ns e
26 (
19.5
%)
14 (
25.9
%)
12 (
15.2
%)
34 (
61.8
%)
16 (
64.0
%)
18 (
60.0
%)
<0.
005
<0.
005
<0.
005
Ant
i-hy
pert
ensi
ve
med
icat
ions
b, e
20 (
15.0
%)
10 (
18.5
%)
10 (
12.7
%)
37 (
67.3
%)
13 (
52.0
%)
24 (
80.0
%)
<0.
005
<0.
005
<0.
005
Lip
id-r
educ
ing
med
icat
ions
c, e
10 (
7.5
%)
5 (9
.3 %
) 5
(6.3
%)
26 (
47.3
%)
9 (3
6.0
%)
17 (
56.7
%)
<0.
005
<0.
005
<0.
005
Ant
i-th
rom
botic
m
edic
atio
ns
2 (1
.5 %
) 2
(3.7
%)
0 (0
.0 %
) 12
(21
.8 %
) 6
(24.
0 %
) 6
(20.
0 %
) <
0.00
5 0.
01
<0.
005
Use
s ni
trog
lyce
rine
10
(7.
5 %
) 6
(11.
1 %
) 4
(5.1
%)
23 (
41.8
%)
10 (
40.0
%)
13 (
43.3
%)
<0.
005
<0.
005
<0.
005
a Com
pari
sons
wer
e m
ade
for
cont
rol (
unaf
fect
ed)
popu
latio
n an
d th
e af
fect
ed g
roup
with
ris
k of
CV
D o
r C
AD
The
aff
ecte
d gr
oup
had
at le
ast 5
0 %
ste
nosi
s in
at
leas
t one
art
ery.
The
num
ber
of p
atie
nts
with
the
para
met
er a
nd th
e %
of
popu
latio
n w
ith th
e pa
ram
eter
are
sho
wn
b A s
ysto
lic B
P >
140
mm
Hg,
a d
iast
olic
BP
> 9
0 m
mH
g, a
nd/o
r th
e us
e of
ant
i-hy
pert
ensi
ves
defi n
es a
hyp
erte
nsiv
e pa
tient
c P
atie
nts
with
a v
alue
>5
and/
or u
sing
lipi
d re
duce
rs a
re d
efi n
ed a
s hy
perl
ipid
emic
d O
besi
ty d
efi n
ed a
s a
body
mas
s in
dex
(BM
I) >
30
e Ant
i-in
fl am
mat
orie
s in
clud
e as
piri
n (A
SA)
and
non-
ster
oida
l an
ti-in
fl am
mat
orie
s (N
SAID
s). L
ipid
-red
ucin
g m
edic
atio
ns a
re s
tatin
s. A
nti-
hype
rten
sive
s ar
e A
CE
inhi
bito
rs a
nd/o
r be
ta b
lock
ers
f P-v
alue
s ar
e fo
r di
ffer
ence
bet
wee
n th
e pr
opor
tion
in th
e af
fect
ed a
nd in
the
cont
rol g
roup
s
S.A. Jortani and M. Pugia
163
Tabl
e 9.
2 C
ohor
t dia
gnos
tic r
isk
fact
or c
hara
cter
istic
s
Para
met
er
Con
trol
gro
up
Aff
ect g
roup
P-
valu
e, a
ffec
ted
vs. c
ontr
ol a
All
Mal
e Fe
mal
e A
ll M
ale
Fem
ale
All
Mal
e Fe
mal
e
– n
= 1
33
n =
54
n =
79
n =
55
n =
25
n =
30
– –
– –
X (
sd)
X (
sd)
X (
sd)
X (
sd)
X (
sd)
X (
sd)
– –
– A
ge y
ears
50
.2 (
8.9)
51
.4 (
9.4)
49
.4 (
8.4)
56
.1 (
9.2)
53
.9 (
9.1)
57
.9 (
9.0)
<
0.00
5 0.
02
<0.
005
Wei
ght l
bs
171.
4 (5
1.4)
18
7.9
(45.
8)
160.
1 (5
2.3)
19
3.2
(42.
0)
211.
0 (4
7.6)
17
8.4
(30.
3)
0.01
0.
04
0.03
H
eigh
t in
66.7
(4.
1)
69.6
(2.
9)
64.7
(3.
5)
65.7
(4.
5)
69.3
(3.
9)
62.7
(2.
1)
0.15
0.
70
<0.
005
BM
I b 27
.0 (
7.2)
27
.2 (
5.8)
26
.8 (
8.1)
31
.4 (
5.7)
30
.7 (
5.9)
31
.9 (
5.6)
<
0.00
5 0.
01
<0.
005
Syst
olic
BP
mm
Hg
131.
1 (1
0.1)
13
1.0
(12.
2)
131.
2 (8
.4)
151.
1 (2
7.5)
15
2.1
(30.
0)
150.
2 (2
5.7)
<
0.00
5 <
0.00
5 <
0.00
5 D
iast
olic
BP
mm
Hg
72.9
(9.
4)
74.2
(10
.7)
72.0
(8.
3)
84.4
(14
.9)
87.9
(14
.1)
81.5
(15
.1)
0.00
5 <
0.00
5 <
0.00
5 To
tal c
hole
ster
ol
(mm
ol/L
) 20
2.0
(45.
2)
198.
9 (4
5.5)
20
4.2
(45.
2)
202.
0 (4
5.2)
19
4.1
(44.
1)
204.
5 (5
4.8)
0.
76
0.67
0.
98
Tri
glyc
erid
es
(mm
ol/L
) 12
9.2
(58.
8)
133.
1 (7
5.7)
12
6.6
(44.
1)
133.
6 (5
6.1)
12
7.8
(62.
9)
138.
4 (5
0.4)
0.
64
0.76
0.
23
HD
L c
hole
ster
ol
(mm
ol/L
) 46
.5 (
16.6
) 44
.6 (
14.5
) 47
.8 (
17.8
) 42
.5 (
13.3
) 43
.5 (
15.9
) 41
.6 (
10.9
) 0.
11
0.76
0.
03
LD
L c
hole
ster
ol
(mm
ol/L
) 12
6.7
(45.
5)
124.
1 (5
6.2)
12
8.5
(36.
7)
126.
3 (4
1.8)
12
3.2
(39.
9)
129.
0 (4
3.9)
0.
96
0.94
0.
96
Lip
id r
atio
c 4.
7 (1
.6)
4.8
(1.7
) 4.
7 (1
.5)
5.1
(1.8
) 5.
0 (2
.0)
5.1
(1.6
) 0.
19
0.60
0.
20
(n, %
) (n
, %)
n, (
%)
(n, %
) (n
, %)
(n, %
) Pr
otei
unur
ia d
13 (
9.8
%)
8 (1
4.8
%)
5 (6
.3 %
) 14
(25
.5 %
) 10
(40
.0 %
) 4
(13.
3 %
) <
0.00
5 <
0.00
5 <
0.00
5
a Com
pari
sons
wer
e m
ade
for
cont
rol (
unaf
fect
ed)
popu
latio
n an
d th
e af
fect
ed g
roup
with
ris
k of
CV
D o
r C
AD
The
aff
ecte
d gr
oup
had
at le
ast 5
0 %
ste
nosi
s in
at
leas
t one
art
ery
b Bod
y M
ass
Inde
x (B
MI)
is c
alcu
late
d 70
3 ×
wei
ght (
lb)/
heig
ht (
in) 2 .
A B
MI
>30
indi
cate
s ov
erw
eigh
t (ob
ese
patie
nt)
c The
lipi
d ra
tio is
that
of
tota
l to
HD
L c
hole
ster
ol
d Uri
ne p
rote
in to
cre
atin
ine
ratio
is e
stim
ated
with
a s
trip
indi
catin
g “N
orm
al”
or “
Dilu
te N
orm
al”
or n
umer
ical
val
ues.
An
elev
ated
rat
io is
indi
cate
d by
any
nu
mer
ical
val
ue
9 Cardiovascular and Metabolic Syndrome Impact on Uristatin
164
9.3 Results
9.3.1 Correlation to Risk Factors
In Tables 9.1 and 9.2 , we compared the demographic and diagnostic risk factor for the controls to patients with stenosis. Patients with stenosis were more likely to be smok-ers, to be obese by weight or BMI, and to have high blood pressure, be older or diabet-ics. Of the diagnostic tests evaluated, proteinuria was the most signifi cantly different between the control and affected groups. Proteins in urine are independent risk factor of CVD in essential hypertension (Pontremoli et al. 1997 ). Lipids were not signifi -cantly different as most of the affected population was on lipid-reducing medications. The usage of nitroglycerines, anti-hypertension, anti-infl ammatory and anti-throm-botic medications were also signifi cantly more used in the affected population.
9.3.2 Correlation of Tested Markers
In Table 9.3 , we compared the values of the infl ammation tests between controls and affected patients with stenosis. As expected, the mean values of most infl ammatory tests were slightly higher in the stenosis patient groups. The exception was I-α-I where mean values decreased. Of the tests evaluated, granuloctye counts were the most signifi cantly difference ( P < .005) between the control and affected groups. While Uri and uTi values were elevated, 98 % were below 12 mg/L and below.
Predictive value of infl ammatory response in vascular stenosis of these markers was measured. The thresholds for a positive result are shown above in Table 9.4 . For CRP, WBC, granulocytes, and lymphocytes these thresholds were obtained from the literature. We used the high sensitivity threshold for CRP. We found that a threshold of ≥36 mg/L for I-α-I was best for reporting the control group (total n = 133) as normal. For the Uri immunoassay and the uTi strip we used a normal urine concentration limit of <7.8 mg/L based on previous studies with health individual. In practice, only results ≥12 mg/L are considered abnormal. In this case, 98 % of
Table 9.3 Impact of vascular stenosis on infl ammation markers
Assay Control group Affected group P value for control vs affected
Uri (mg/dL) 3.6 (5.9) 6.6 (15.0) .16 uTi Strip (mg/dL) 1.9 (6.4) 5.9 (12.8) .02 Bik (mg/L) 0.9 (1.0) 1.1 (1.5) .34 I-α-I (mg/L) 87.5 (26.4) 82.6 (30.7) .27 CRP (mg/L) 0.4 (0.5) 1.4 (3.6) .04 Total WBC (cell/nL) 6.2 (2.0) 8.6 (2.5) <.005 Granulocytes (cell/nL) 3.6 (1.6) 5.5 (2.4) <.005 Lymphocytes (cell/nL) 1.9 (0.6) 2.2 (0.9) .19
S.A. Jortani and M. Pugia
165
results were normal. The strip method for Bik gave more false positives with pro-teinuria and was less specifi c. Here, only 90 % of the controls were reported as normal. For blood values, we used a threshold of ≥2.5 mg/L for Bik based on reporting a comparable number of control group as normal.
Table 9.4 Predictive value of infl ammatory response in vascular stenosis
Assay(s) a Positive cutoff
True neg
False pos
True pos
False neg
%Specifi city d, e (95 % Conf)
%Sensitivity e, f (95 % Conf)
Uri (mg/dL) >7.8 125 8 11 44 94.0 (88.5–97.4)
20.0 (10.4–33.0)
Bik strip (mg/dL)
>7.8 120 13 13 42 90.2 (83.9–94.7)
23.6 (13.2–37.0)
Bik (mg/L) >2.5 128 5 9 46 96.2 (91.4–98.8)
16.4 (7.8–28.8)
I-α-I (mg/L) <36 129 4 5 50 97.0 (92.5–99.2)
9.1 (3.0–20.0)
CRP (mg/L) >2 129 4 6 49 97.0 (92.5–99.2)
10.9 (4.1–22.2)
Total WBC (cell/nL) b
>10 124 9 15 40 93.2 (87.5–96.9)
27.3 (16.1–41.0)
Granulocytes (cell/nL) b
>7 127 6 14 41 95.5 (90.4–98.3)
25.5 (14.7–39.0)
Lymphocytes (cell/nL) b
>3 125 8 10 45 94.0 (88.5–97.4)
18.2 (9.1–30.9)
Bik or Uri c – 121 12 19 36 91.0 (84.8–95.3)
34.5 (22.2–48.6)
Bik or Uri or CRP c
– 120 13 21 34 90.2 (83.9–94.7)
38.2 (25.4–52.3)
Bik or Uri or I-α-I c
– 117 16 23 32 88.0 (81.2–93.0)
41.8 (28.7–55.9)
Bik or Uri or granulocytes c
– 116 17 29 26 87.2 (80.3–92.4)
52.7 (38.8–66.3)
Bik or Uri or granulocytes or CRP c
– 15 18 30 25 86.5 (79.5–91.8)
54.5 (40.6–68.0)
Legend : a Uri, Bik, I-α-I and CRP were measured by immunoassay. In this table, Uri is a urine value, and Bik a plasma value. Note that both measure all inhibitory fragments in specimens (e.g., both Uri and Bik). I-α-I is a plasma immunoassay for the family of I-α-I pro-inhibitors. Bik is a strip assay performed on a CLINITEK analyzer and based on trypsin enzyme inhibition. Values above the cutoff (except for I-α-I) are taken as positive b 1 Cell/nL = 10 3 cell/mL c The combination tests are positive if any one of the assays being combined is positive. Compared non-parametrically with the Mann-Whitney U test and exact, two-sided P value calculations using the SAS NPAR1WAY procedure. The P values are two sided. A value less than .05 is considered as statistically signifi cant d Specifi city = 100 × True Negatives/(True Negatives + False Positives) e 95 % confi dence intervals are Clopper-Pearson intervals f Sensitivity = 100 × True Positives/(True Positives + False Negatives)
9 Cardiovascular and Metabolic Syndrome Impact on Uristatin
166
The specifi cities and sensitivities for the various infl ammation markers for detecting vascular stenosis (>50 %) were calculated as shown in Table 9.4 . For all the markers, at least 93–97 % of the controls were considered normal and only 9–27 % of the stenosis group were considered positive. The WBC test was most sensitive with 27 % of the affected groups with vascular stenosis reported. The com-binations of pro-infl ammatory (CRP, granulocytes) and anti-infl ammatory (Bik, Uri) biomarkers as a panel indication of infl ammation increased the sensitivity to stenosis to 55 %. But not all patients with vascular stenosis had a positive infl amma-tion test. Moreover, increasing the number of tests in a panel increased the number of false positives and reduced the specifi city from 93 to 87 %. The immunoassay Uri and Bik values were typically within the normal ranges (see thresholds in Table 9.4 ) except when patient had elevated granuloctye counts. Also half of the CHF and MI patients (n = 13) had elevated Uri or Bik and this contributes signifi cantly to the positives observed.
9.3.3 Correlation to Metabolic Syndrome
In Table 9.5 , we compared the values of the infl ammation tests between controls and patients meeting metabolic syndrome (MetSyn) criteria or only a single risk factor such as lipidemia (LPD), hypertension (HTN), diabetes (DM) or obesity (OB) (Deen 2004 ). The average patient ages in all groups was 51–52 years and not signifi cantly different between groups. Males and females were evenly distributed. As expected, the mean values of most infl ammatory tests were slightly higher in the stenosis patient groups.
In addition, we tested all samples for plasma adiponectin (Adpn). Adpn is sig-nifi cantly lowered in patients with metabolic syndrome or diabetes (Daimon et al. 2003 ; Kadowaki et al. 2006 ). This adipocytokine is released by adipose tissue to slow liver production of glucose and to initiate muscle to oxidize fatty acids and glucose (Maeda et al. 1996 ). MetSyn patients tend to have infl ammation and have diffi cultly controlling insulin and obesity (Hotamisligil et al. 1995 ; Kahn et al. 2006 ). Lipid accumulation and infl ammation in the adipose is thought to lower Adpn by inhibiting expression by the Peroxisome Proliferator-Activated Receptor (PPARγ). Hypertrophic adipose requires more insulin to activate PPARγ, thus increasing insulin resistance. Activating PPARγ with thiazolidinediones (TZD) is therapeutic intervention for hyperglycemia and hypoadiponectinemia (Tsuchida et al. 2006 ; Kubota et al. 2005 ).
9.3.4 Pro-infl ammatory Response
The pro-infl ammatory response measured by White Blood Cell counts (WBC) and CRP was increased in metabolic syndrome and risk groups compared to the control group in a sub group of patients with all metabolic factors measured (see
S.A. Jortani and M. Pugia
167
Tabl
e 9.
5 C
ompa
riso
n of
met
abol
ic s
yndr
ome
fact
ors
Para
met
er
Nor
mal
Met
abol
ic
synd
rom
e (M
etSy
n)
Hyp
erte
nsio
n (H
TN
) O
besi
ty (
OB
) L
ipid
emia
(L
PD)
Dia
bete
s (D
M)
Cou
nt a,
b
84
40
36
34
25
15
avg
sd
avg
sd
avg
sd
avg
sd
avg
sd
avg
sd
Age
(ye
ar)
51.9
9.
1 50
.3
6.4
51.7
7.
1 50
.9
6.9
51.0
7.
6 50
.9
6.4
Plas
ma
Adi
pone
ctin
(m
g/L
) 10
.9
6.9
5.8
2.4
6.2
2.7
6.9
4.3
8.1
4.5
6.0
2.6
Uri
(m
g/L
) 2.
6 2.
0 4.
6 6.
0 4.
7 6.
1 4.
9 6.
3 2.
9 2.
7 5.
3 7.
1 To
tal W
BC
(ce
ll/nL
) c 5.
6 1.
2 7.
4 2.
4 7.
2 2.
4 7.
5 2.
2 6.
7 2.
4 7.
0 2.
4 G
ranu
locy
tes
(cel
l/nL
) c 3.
1 1.
0 4.
3 1.
8 4.
1 1.
8 4.
4 1.
8 3.
9 1.
8 4.
1 2.
0 Ly
mph
ocyt
es (
cell/
nL) c
1.9
0.5
2.3
0.8
2.3
0.7
2.3
0.7
2.2
0.7
2.0
0.8
CR
P (m
g/L
) 0.
3 0.
3 0.
6 0.
6 0.
6 0.
7 0.
6 0.
6 0.
5 0.
5 0.
4 0.
4
Lege
nd :
a Uri
ne, p
lasm
a an
d se
rum
wer
e co
llect
ed f
rom
pat
ient
s di
agno
sed
with
(n
= 4
0) a
nd w
ithou
t (n
= 8
4) m
etab
olic
syn
drom
e ac
cord
ing
to A
FP d
efi n
ition
(D
een
2004
) as
eith
er a
dia
gnos
is o
f in
sulin
res
ista
nce
or h
avin
g m
ore
than
tw
o m
etab
olic
ris
k fa
ctor
s. P
atie
nts
(n =
124
) w
ith a
t le
ast
one
risk
fac
tor:
hyp
erte
nsio
n (H
TN
), o
besi
ty (
OB
), li
pide
mia
(L
PD)
or d
iabe
tes
(DM
) ar
e al
so s
how
n b I
n th
is c
ohor
t, al
l pat
ient
s la
cked
kid
ney
dise
ase
and
infe
ctio
ns in
clud
ing
urin
ary
trac
t inf
ectio
n c 1
Cel
l/nL
= 1
0 3 cel
l/mL
9 Cardiovascular and Metabolic Syndrome Impact on Uristatin
168
Table 9.3 ). Values for WBC increased by 32 % and values for CRP increased by 149 % for patients meeting metabolic syndrome criteria (see Table 9.3 ). The WBC differences were the more signifi cant: cardiovascular (P = 0.017), HTN (P = 0.030), OB (P = 0.014), LPD (P = 0.086) and DM (P = 0.075). Differential counts showed granulocytes correlated strongly while lymphocytes changes were less signifi cant. Serum CRP differences were signifi cant for cardiovascular (P = 0.033), HTN (P = 0.045), OB (P = 0.036) but not for LPD or DM (P > 0.1). Most values (<99 %) were well within normal reference ranges of <12.4 cell/nL for WBC and <2.0 mg/L for high sensitivity threshold for a negative CRP. A few patients were above the WBC (n = 1) or CRP (n = 4) clinical thresholds.
9.3.5 Adiponectin Response
Plasma Adpn was signifi cantly reduced in patients with cardiovascular (P = 0.019), HTN (P = 0.034), OB (P = 0.064) and DM (P = 0.054) but not those with LPD (P = 0.195). The mean Adpn value for the cardiovascular group was 55 % lower than for the group without cardiovascular. A clinical threshold of <5 mg/L for a positive result gave a sensitivity of 50 % and specifi city of 87 % for the detection of MetSyn patients. Urinary levels of Adpn were not correlated to the presence cardiovascular or risk factors.
9.3.6 Uristatin Response
Uri values were higher in 79 % metabolic syndrome (MetSyn) patients and those with lipidemia (LPD), hypertension (HTN), diabetes (DM) or obesity (OB). Normal reference ranges for Uri values were <7.8 mg/L for urine. Most (>99 %) patients and controls had values below this threshold. Therefore the clinical sensitivity for metabolic factors was only 5–15 %. A few patients had values above these clinical thresholds in either urine (n = 7) or plasma (n = 3) and this was likely due to elevated WBC.
9.4 Conclusion
Urisatin can be used for assessment of kidney disease and infection in patients with cardiovascular disease patients (CVD), coronary artery disease (CAD) or metabolic syndrome patients (MetSyn) without signifi cant interference from underlying con-ditions. This allows potential applications for diabetic nephropathy and other nephropathy.
S.A. Jortani and M. Pugia
169
Acknowledgements We would like to acknowledge the work done by Department of Pathology and Laboratory Medicine, University of Louisville, School of Medicine. We thank Dr. Manju Basu and Ron Sommer for running ELISA as needed. We also thank Dr. Robert Dwyer, Paul W. Dillon PhD, Prakash Tewari, Laura Tinker, and Peter Connolly for the ADVIA® Centaur™ results, and assistance of Dr. Sumanth Prabhu (University of Louisville Cardiology) in recruiting some of the patients into this study and Mr. Edward P. Womack from Diagnostic Reference Laboratory at University of Louisville is acknowledged for technical assistance.
References
Avanzas P, Arroyo-Espliguero R, Cosin-Sales J, Aldama G, Pizzi C, Quiles J, Kaski JC (2004) Markers of infl ammation and multiple complex stenoses (pancoronary plaque vulnerability) in patients with non-ST segment elevation acute coronary syndromes. Heart 90:847–852
Blake J, Ridker PM (2003) C-reactive protein and other infl ammatory risk markers in acute coro-nary syndromes. J Am Coll Cardiol 41:37S–42S
Bono F, Lamarche I, Herbert JM (1997) Induction of vascular smooth muscle cell growth by selec-tive activation of the proteinase activated receptor-2 (PAR-2). Biochem Biophys Res Commun 241:762–764
Brennan ML, Penn MS, Van Lente F, Nambi V, Shishehbor MH, Aviles RJ, Goormastic M, Pepoy ML, McErlean ES, Topol EJ, Nissen SE, Hazen SL (2003) Prognostic value of myeloperoxi-dase in patients with chest pain. N Engl J Med 349:1595–1604
Coelho AM, Ossovskaya V, Bunnett NW (2003) Proteinase-activated receptor-2: physiological and pathophysiological roles. Curr Med Chem Cardiovasc Hematol Agents 1:61–72
Cottrell G (2002) Protease activated receptors: the role of cell surface proteolysis in signaling. Essays Biochem 38:169–183
Daimon M et al (2003) Decreased serum levels of adiponectin are a risk factor for the progression to type 2 diabetes in the Japanese Population Funagata study. Diabetes Care 26:2015–2020
Danesh J et al (1998) Association of fi brinogen, C-reactive protein, albumin, or leucocyte count with coronary heart disease: meta-analysis of perspective studies. JAMA 279:1477–1482
Deen D (2004) Metabolic syndrome: time for action. Am Fam Physician 69:2875–2882 Egeblad K, Astrup T (1966) Fibrinolysis and the trypsin inhibitor in human urine. Scand J Clin
Lab Invest 18:181–190 Heeschen C, Dimmeler S, Fichtlscherer S, Hamm CW, Berger J, Simoons ML, Zeihe AM (2004)
Prognostic value of placental growth factor in patients with acute chest pain. JAMA 291:435–441 Hotamisligil GS, Arner P, Caro JF, Atkinson RL, Spiegelman BM (1995) Increased adipose tissue
expression of tumor necrosis factor in human obesity and insulin resistance. J Clin Invest 95:2409–2415
Kadowaki T, Yamauchi T, Kubota N, Hara K, Kohjiro Ueki K, Tobe K (2006) Adiponectin and adiponectin receptors in insulin resistance, diabetes, and the metabolic syndrome. J Clin Invest 116:1784–1792
Kato H et al (2002) Regulation of functions of vascular wall cells by tissue factor pathway inhibitor basic and clinical aspects arterioscler. Throub Vasc Biol 22:539–548
Kahn SE, Hull RL, Utzschneider KM (2006) Mechanisms linking obesity to insulin resistance and type 2 diabetes. Nature 444:840–846
Kubota N et al (2005) Pioglitazone ameliorates insulin resistance and diabetes by both adiponectin- dependent and -independent pathways. Diabetes 54(12):3358–3370
Libby P, Ridker PM, Maseri A (2002) Infl ammation and atherosclerosis. Circulation 105:1135–1143 Maeda K, Okubo K, Shimomura I, Funahashi T, Matsuzawa Y, Matsubara K (1996) cDNA cloning
and expression of a novel adipose specifi c collagen-like factor, apM1 (AdiPose Most abundant Gene transcript 1). Biochem Biophys Res Commun 221:286–289
9 Cardiovascular and Metabolic Syndrome Impact on Uristatin
170
Miike S, McWilliam A, Kita H (2001) Trypsin induces activation and infl ammatory mediator release from human eosinophils through protease-activated receptor-2. J Immunol 167:6615–6622
Nii A, Morishita H, Hirose J, Yamakawa T, Kanamori T (1995) Novel blood coagulation factor inhibitory activities of the second domain of urinary trypsin inhibitor and its variants. Nippon Kessen Shiketsu Gakkaishi 6:203–207
Nii A, Morishita H, Yamakawa T, Matsusue T, Hirose J, Miura T et al (1994) Design of variants of the second domain of urinary trypsin inhibitor (R-020) with increased factor Xa inhibitory activity. J Biochem 115:1107–1112
Nguyen M, Arkell J, Jackson CJ (2001) Human endothelial gelatinases and angiogenesis. Int J Biochem Cell Biol 33:960–970
Olear T, Nouza K (1999) Thrombin and trypsin receptors: the same mechanism of signaling on cellular surfaces. Bratisl Lek Listy 100:75–79
Pasceri V, Willerson JT, Yeh ETH (2000) Direct proinfl ammatory effect of C-reactive protein on human endothelial cells. Circulation 102:2165–2168
Pontremoli R et al (1997) Microalbuminuria: a marker of cardiovascular risk and organ damage in essential hypertension. Kidney Int Suppl 52:S163–S165
Pugia MJ, Lott JA (2005) Pathophysiology and diagnostic value of urinary trypsin inhibitors (review article). Clin Chem Lab Med 43:1–16
Pugia MJ, Valdes R, Jortani SA (2007a) Chapter on urinary trypsin inhibitors: structure, biological relevance and measurement. In: Dr. Greg Makowski (ed) Adv Clin Chem, vol 44, Elsevier, Morristown, NJ, USA, pp 223–245. ISBN- 10: 0-12-373704-4
Pugia MJ et al (2007b) Immunological evaluation of urinary trypsin inhibitors in blood and urine: role of N- & O-linked glycoproteins. Glycoconj J 24(1):5–15
Remme WJ, Swedberg K (2002) European Society of Cardiology Comprehensive guidelines for the diagnosis and treatment of chronic heart failure. Task force for the diagnosis and treatment of chronic heart failure of the European Society of Cardiology. Eur Heart J 4:11–12
Ridker P, Rifai N, Rose L, Buring J, Cook N (2002) Comparison of C-reactive protein and low- density lipoprotein cholesterol levels in the prediction of fi rst cardiovascular events. N Engl J Med 347:1557–1565
Ridker PM, Cushman M, Stampfer MJ, Tracy RP, Hennekens CH (1997) Infl ammation, aspirin, and the risk of cardiovascular disease in apparently healthy men. N Engl J Med 336:973–979
Ridker PM, Cushman M, Stampfer MJ, Tracy RP, Hennekens CH (1998a) Plasma concentration of C-reactive protein and risk of developing peripheral vascular disease. Circulation 97:425–428
Ridker PM et al (1998b) Prospective study of C-reactive protein and the risk of future cardiovas-cular events among apparently healthy women. Circulation 98:731–733
Ridker PM et al (2000) C-reactive protein and other markers of infl ammation in the prediction of cardiovascular disease in women. N Engl J Med 342:836–843
Shibutani Y, Kunihiro Y (1986) Preventive effects of urinastatin on tissue degradation. Yakuri Chiryo 14:6057–6072
Szmitko PE, Wang CH, Weisel RD, Jeffries GA, Anderson TJ, Verma S (2003) Biomarkers of vas-cular disease linking infl ammation to endothelial activation part II. Circulation 108:2041–2048
Tsuchida A et al (2006) Peroxisome proliferator-activated receptor (PPAR) alpha activation increases adiponectin receptors and reduces obesity-related infl ammation in adipose tissue: comparison of activation of PPAR alpha, PPAR gamma, and their combination. J Biol Chem 281(13):8748–8755
S.A. Jortani and M. Pugia
171© Springer International Publishing Switzerland 2015 M. Pugia (ed.), Infl ammatory Pathways in Diabetes: Biomarkers and Clinical Correlates, Progress in Infl ammation Research, DOI 10.1007/978-3-319-21927-1_10
Chapter 10 Uristatin Anti-infl ammatory Cellular Signaling
Manju Basu , Subhash Basu , and Michael Pugia
Abstract Bikunin (Bik) and uristatin (Uri) impacted cellular signaling of kidney proximal tubular, kidney gluomerlar, muscle, and carcinoma cells. Cell cultures were induced with Bik and Uri at 12–40 mg/L and compared to unexposed cell cultures. The balance in serine protease trypsin activity was shifted to complete trypsin inhibition. Cell proliferation was slowed to 50 %. Cell death by apoptosis signaling though caspases, was not activated by the presence of Bik or Uri. Proliferation signaling though the mitogen-activated protein kinases (MAPK) was active. Phosphatidylinositol 3-kinase (PIK3) and protein kinase B (Akt) signaling were activated. Cell death occurs through a non-programmed cell death. Exposed cells showed increased release of a membrane-bound protein, dipeptidyl peptidase 4 (DPP-4). Membrane disruption was also evidenced by phosphatidylserine (PS) externalization measured by binding a fl uorescence dye (PSS-380). Exposed cells were depleted of intercellular calcium as shown by using calcium- binding fl uores-cence dye (Fluo-4). Blockage of the Ca 2+ sensitive potassium (K Ca ) channels (or calcium-activated potassium (BK) channels) is thought to lower the intracellular Ca 2+ . All of the information was supported by a kidney cell model, of Bik/Uri being protective of immune mediated apoptosis and proliferation during the infl ammatory process, but causing cell loss by decreased renal regeneration by Akt-PIK3 signaling.
M. Basu (*) • S. Basu Department of Chemistry and Biochemistry , University of Notre Dame , Notre Dame , IN 46556 , USA e-mail: [email protected]
M. Pugia Department of Chemistry and Biochemistry , University of Notre Dame , Notre Dame , IN 46530 , USA
Strategic Innovation , Siemens Healthcare Diagnostics , 3400 Middlebury Street , Elkhart , IN 46530 , USA e-mail: [email protected]
172
10.1 Introduction
Infl ammatory processes promote the release of trypsin serine proteases. Proper con-trol of infl ammation is necessary to prevent further damage to normal tissue. Urinary trypsin inhibitors (uTi) are small proteins serving as Kunitz type inhibitors of serine proteases that can be detected in human urine during infl ammation (Pugia and Lott 2005 ; Pugia et al. 2007b ). Bikunin (Bik) is one of the key forms of uTi with N- and O- linked complex glycans. Uristatins (Uri) are another key form lacking the O-linked glycan. Bik binds to renal cell surfaces not only through the peptide Kunitz-type domains but also through the large complex glycans. The O-glycoside chain binding increases the affi nity to cells, calcium and TSG-6 (Hirashima et al. 2001 ; Kanayama et al. 1997 ; Milner and Day 2003 ). The sequences of the carbohy-drate chains are known to change during disease progression (Hochstrasser et al. 1967 ; Zhuo et al. 2002 ; Lindstroem et al. 1997 ).
Many researchers have shown uTi plays a key role in the human anti- infl ammatory system to protect tissues and suppress vascular dilation, coagulation, and smooth muscle contraction (Pugia et al. 2007b ). Bik inhibits trypsin, chymotrypsin, kalli-krein and plasmin, elastase, cathepsin Factors IXa, Xa, XIa and XIIa through bind-ing with either of two Kunitz binding domains. uTi reduces release of growth factors and cytokines inhibiting growth factor activation (Huang et al. 2001 ). uTi has been measured in multiple patient groups and proved to be a sensitive diagnostic marker for infections (Pugia et al. 2004 ; Jortani et al. 2004 ). uTi is almost entirely excreted into urine to prevent its extended negative impact on the anti-infl ammatory pro-cesses. Rapid renal clearance is due to a glomerular fi ltration rate 80-fold higher than that of albumin (Lindstroem et al. 1997 ; Joberg et al. 1995 ; Ohlson et al. 2001 ). Thus the kidney is exposed to the majority of uTi. This interaction likely impacts the infl ammatory repair and healing processes of the kidney. However, the cellular impact of Bik and Uri exposure is not fully understood.
Kunitz-type protease inhibitors such as aprotinin have been shown to slow cell proliferation, protect renal cells, and signifi cantly decreased the rise in serum cre-atinine and apoptosis caused by kidney ischemia-reperfusion injury (Bono et al. 1997 ; Yamaski et al. 1996 ; Kher et al. 2005 ). Similarly, Uri and Bik have been shown to protect against renal failure induced in animal models and programmed cell death (Nakakuki et al. 1996 ; Takahashi et al. 2001 ) Aprotinin is shown to prevent protease- activated receptor (PAR) signaling of proliferation through the p38 mitogen- activated protein kinase activation (MAPK) and protein kinase A and B (PKA, PKB) (Bono et al. 1997 ). Aprotinin also blocks calcium-sensitive potassium channels (Moczydlowski et al. 1992 ). uTi have also been shown to inhibition of Ca 2+ infl ux in vascular smooth muscle (Kanayama et al. 1998 ). The Kunitz-type peptide domains Bik and Ur are homologous to aprotinin (Pugia et al. 2007b ). A key difference is aprotinin lacks the added N- and O- linked complex glycans of Bik.
In this study, we explored the responses in cultured cells to pathological levels of Bik and Uri to evaluate which signal transduction pathway is involved in the inner- cellular signaling. Additional we want to compare Bik to Uri to uncover any
M. Basu et al.
173
potential impact of glyco-conjugation. Therefore active site on Bik was also charac-terized to understand its the distinct functions of Bik protein domains.
10.2 Methods
10.2.1 Bikunin and Uristatin
Urinary trypsin inhibitors were isolated from chronic renal failure patients (SciPac, Sittingbourne, Kent, UK) (Sumi et al. 1981 ). Two isolations of >90 % purity were obtained: Bik (~30 kDa) and Uristatin (Uri) (17 kDa).
10.2.2 Cell Culturing Procedure
Normal cells used were African green monkey proximal tubular kidney cells (Vero), mouse myocytes (C2C12) (American Type Cell Culture, ATCC, Manassa VA) and human mesangial gluomerlar cells (NHMC) (Cambrex Inc, Cambridge MA). Carcinoma cells used were human breast cancer lines SKBR-3 and MDA-468 (ATCC). Carcinoma cells were maintained in culture using RPMI-1640 media, while normal cells were maintained with modifi ed Eagle’s medium (MEM). Both media contained 2 mM glutamine, penicillin (50 units/ml), streptomycin (50 μg/ml), and 10 % fetal bovine serum (Invitrogen, Carlsbad, CA).
Cells were grown at 37 °C in an atmosphere of 95 % air and 5 % CO 2 , and p propagated by sub-culturing at confl uence by asceptically treating with trypsin- EDTA, and replenishing with media. Cell confl uence was about 5 million cells per T-75 cm 2 fl ask. Cells were grown to 70–80 % and synchronized by adding hydroxy-urea (0.5 mM, sterile, Sigma-Aldrich, St. Louis, MO) 2 times for 24 h each. Synchronized cells were induced with various concentrations of Bik (5, 50, 500 μg/mL) in full media under sterile conditions. After induction, the cell media was dis-carded and aliquots taken for cell counts.
Cells were harvested at 4, 24, and 48 h from start of induction. The cells were harvested in PBS/EDTA and suspended in 2 ml PBS. Cell counts were taken imme-diately using Trypan Blue staining. Harvested packed cells were suspended in two volumes of HEPES buffer, 0.05 M and pH 7.2. The cells were sonicated at 4 °C for 15 s with 30 s rest intervals three times. The homogenates were centrifuged at 12,000 rpm (50,000 g) for 15 min at 4 °C. The resulting supernates were saved in small aliquots at −20 °C. The lysate residues (pellets) were suspended again in twice the volume of HEPES with 0.1 % Triton X-100 stored at −20 °C until tested. The protein content in cell lysates was measured in triplicate by the Bradford dye- binding procedure (BioRad Laboratories Hercules, CA). Exposures and analysis were repeated on three separate trials. Cells for kinase phosphorylation assays were added to the mircotiter plate wells and fi xed with 100 uL of 4 % formaldehyde in
10 Uristatin Anti-infl ammatory Cellular Signaling
174
PBS. The plates were kept for an additional 20 min at room temperature and then covered with parafi lm, and stored at 4 °C until assayed.
10.2.3 Serine Protease Activity and Inhibition Assays
Serine protease inhibitor concentrations were determined by the inhibition of tryp-sin hydrolysis of a chromogenic arginine peptide substrate (3-(N-tosyl-N-nitro arginyloxy)-5-phenylpyrrole) placed in a dipstick (Siemens Healthcare Diagnostics, IN). The rate of substrate disappearance was measured by the rate of color forma-tion with 2-methoxy-4-morpholino-benzenediazonium on a Clinitek® 50 strip reader (Siemens Healthcare Diagnostics, IN), as previously published. The samples were read in duplicate and the value determined from a standard curve obtained with Bik standard solutions containing 0, 12.5, 25, and 50 mg/l (SciPac, Sittingbourne, Kent, UK, product code P205-1). Serine protease activity was also measured using another trypsin specifi c chromogenic substrate BAPNA (Nα-Benzoyl-DL-arginine 4-nitroanilide, Sigma-Aldrich, St. Louis, MO) and mon-itoring trypsin hydrolysis at 405 nm. BAPNA at 18.3 mM was prepared in 20 % dimethyl formamide: 80 % water. Triethanolamine reaction buffer was prepared at 250 mM and pH 7.8. Samples of cell extracts (6 μL) were diluted into 75 μL water. Trypsin calibration samples included TPCK treated from bovine pancreas: 7300 unit/mg was from Sigma-Aldrich, St. Louis, MO. At time zero, 15 μL of water and 25 μL of BAPNA were added to 100 μL of reaction buffer and 75 μL sample under light-proof conditions. Samples were read in triplicate every 10 min for 1 h at 37 °C. The results were plotted to determine the serine protease amount in mg/L in cell extracts.
10.2.4 DPP-4 Peptidase Activity Assay
Serine peptidase activity was measured using the chromogenic substrate GPN (H-Gly-Pro-p-nitroanilide, Sigma-Aldrich, St. Louis, MO) specifi c for dipeptidyl peptidase 4 (DPPIV porcine, Sigma- Aldrich, St. Louis, MO). Cell extracts (100 uL) or DPP-4 at 0, 3.6, 7.2, and 14.4 mg/L were added to 100 uL of 100 mM Tris buffer (pH 8.0). At time zero, 25 uL of GPN (0.76 mM in Tris) was added and triplicate samples read every 1 min for 20 min at 37 °C. The hydrolysis rates were monitored at 405 nm and plotted to determine peptidase amount in mg/L. One miligram equals 0.91 units of DPP-4 activity. Bik did not inhibit DPP-4 and was not an interferent.
10.2.4.1 Enzyme Linked Immunosorbent Assays (ELISA)
ELISA for DPP-4 was performed with cell extracts according to manufacturers instructions (Nephroscreen kits, Brendan Bioscience, LLC of Hopedale, MA).
M. Basu et al.
175
10.2.5 Western-Blot Analysis
Western blots were performed against anti-caspase antibodies (3/8/9) and MAPK anti-body (phosporylated p38 and ERK) (Cell Signaling Technology, Inc. Danvers, MA) and the monoclonal antibodies against Bik/Uri (421-3G5) and DPP-4 (Anti-URO- 4 mAb S27, Abcam, Cambridge, MA). Proteins separated on SDS-PAGE electrophore-sis were transferred to a nitrocellulose membrane in 1x Tris-Glycine buffer (Bio-rad, Hercules, CA) and 20 % methanol. The transferred proteins reacted with the specifi c fi rst and ALP-conjugated second antibodies at a desired concentration after 1 h block-ing with 5 % BSA blocker in TBS-0.1 % Tween-20. The target binds were visualized using an ALP substrate BCIP/NBT tablet dissolved in ddH2O (Sigma, St. Louis, MO).
10.2.6 Akt and PIK3 Activation Measurements
Measurements of the relative increase in phosphoralyation of Akt and PIK3 were performed according to the ELISA manufacturer’s instruction (Superarray, Frederick, MD). Results were repeated in 3 independent experiments with new cul-tures. Harvesting of packed cells was accomplished by suspending cells in 2X vol-ume of HEPES buffer, 0.05 M and pH 7.2. The cells were sonicated at 4 °C for 15 s with 30 s rest intervals three times. The homogenates were centrifuged at 12,000 rpm (50,000 g) for 15 min at 4 °C. The resulting supernates were saved in small aliquots at −20 °C. The membrane pellets were suspended again in twice the volume of HEPES with 0.1 % Triton X-100 stored at −20 °C until tested. Cells for kinase phosphorylation assays were added to the mircotiter plate wells and fi xed with 100 uL of 4 % formaldehyde in PBS. The plates were kept for an additional 20 min at room temp and then covered with parafi lm and stored at 4 °C until assayed
Kinase phosphorylation assays can generally be conducted by adding cells to the mircotiter plate wells and fi xing with 100 uL of 4 % formaldehyde in PBS. The plates were kept for an additional 20 min at room temperature and then covered with parafi lm and stored at 4 °C until assayed. ELISA measurements of the relative increase in phosphoralyation of Akt and PIK3 were performed according to the ELISA manufacturer’s instruction (Superarray, Frederick, MD). Measurements of the relative increase in MAPK ERK phosphoralyation were preformed according to the ELISA manufacturer’s instruction (Assay Designs, Ann Arbor MI). Results were repeated in 3 independent experiments with new cultures. Cells were fi xed to the micro-titer plate followed by removal of the fi xing buffer and cell quenching performed with H 2 O 2 and NaN 3 . The quenching buffer was removed and the plate was heated, washed, and blocked. This was followed by 1 h incubation with primary antibodies for Akt, Phos-Akt, PI3K, and Phos-PI3K (1:100 dilution). Each antibody and induction condition was tested in separate triplicate wells. The unbound pri-mary antibodies were removed by washing and the plate was then incubated for 1 h with a secondary antibody conjugated to HRP (1:16 dilution). The plate was washed twice and developing solution was added. After incubation, the resultant dark blue color formed after the addition of stop solution was read at 450 nm.
10 Uristatin Anti-infl ammatory Cellular Signaling
176
10.2.7 Apoptosis Assays
Apoptosis assays by caspase activation are generally conducted by using various cell extracts and supernatants. Western blots were performed against anti-caspase anti-bodies (3/8/9) (Cell Signaling Technology, Inc. Danvers, MA). The electrophoresis was run under standard conditions using Tris-Glycine-SDS buffer at 18–20 mAmp with 3 μg BioRad prestained standards. After electrophoresis, the proteins were transferred from the gel onto a nitrocellulose membrane at 80 V for 1 h or 40 V over-night at 4 °C. The transferred proteins were reacted with the specifi c antibodies (1 mg/mL) mentioned above after 1 h blocking with casein blocker. After washing off the unbound antibody and blocking with TBS containing Tween20, the blotted membranes were incubated with ALP-conjugated corresponding second antibody for another hour. Finally, the cross-reactivity of the samples with the antibodies was monitored by the addition of 66 ul of nitro blue tetrazolium (50 mg/ml 70 % DMF) and 35 ul of bromo chloro indolyl phosphate (50 mg/ml DMF, 100 %) in 10 mL TBS.
10.2.8 Intra-cellular Analysis
Measurements of cytosolic calcium were done by infusing cells with the membrane permeable calcium sensitive Fluo-4 AM dye as previously described in Chap. 5 . Apoptosis assays using phospatidyl serine analysis were conducted by using viable cells. Cells induced with CTF or Adpn were treated with PSS-380 dye to recognize phosphatidyl serine on the outer leafl et of the apoptotic cells (Molecular Targeting Technologies Inc). Cells were plated onto a Falcon Microslide System (Fisher) and washed two times with TES buffer (5 mM N-tris [Hydroxymethyl]-2-aminoethane- sulfonic acid: TES, 150 mM NaCl, pH7.4) followed by incubation with 200 μL new TES buffer containing 25 μM PSS-380 at 37 °C for 10 min. The buffer was removed after staining, and the cells were washed with TES buffer once before observation for fl uorescence. Phase contrast images were used to locate regions of fl uorescent signals and to calculate the percentage of affected cells. Vero cells induced with Bik at 10, 25, and 50 mg/L were treated with PSS-380 dye to recognize phosphatidyl serine externalization as previously described (Smith et al. 2003 ).
10.2.9 Active Site Studies
Trypsin inhibition by different Bik peptides (A–E), Bikunin, Uristatin, and apro-tinin on trypsin inhibitory was measured in the presence of antibody to Bik (mAb 421.3G5) using measurement by trypsin-specifi c chromogenic substrate (N-α- benzoyl-DL-arginine 4-nitroanilide, BAPNA, Sigma-Aldrich) at multiple trypsin and inhibitor levels (10–500 mg/L). Hydrolysis rates are plotted to determine the average inhibition expressed as mg trypsin to mg inhibitor. Five peptides (A–E)
M. Basu et al.
177
represent different regions in Bikunin sequence were chemically synthesized without the attached glycoconjugates. Trypsin inhibition was observed by Bik (0.76 mg/mg), Uri (0.89 mg/mg) and aprotinin (0.58 mg/mg) but not synthetic pep-tides A–E even if these overlapped the predicted active sites. Competitive inhibition experiments were preformed by adding peptide (A–E) or protein (Bik, Uri or apro-tinin) to the sample of Bikunin and mAb.
The hydrolysis rates were plotted to determine the average percentage of inhibi-tion w/w and the amount of restored inhibition calculated from the Bikunin inhibi-tion rate without mAb. Inhibition of trypsin by Bik or Uri (0.4 mg/L) was reduced by 60–70 % when either were incubated for 1 h at 37 °C with Uristatin monoclonal antibody (mAb 3G5) (1.6 mg/L). Only peptide B restored any trypsin inhibition when added in 50-fold excess as a competitive binder. This supported an active site inhibition, and mAb 3G5 binding was in peptide B region. Trypsin inhibition by BAPNA assay was plotted to determine the average percentage of inhibition for equal weight of trypsin to inhibitor (mg/mg). Binding of the Bikunin proteins and peptide analogs by mAb (421.3G5) was measured by ELISA with the full protein or peptide sequence (1–1000 ng/well).
10.3 Results
10.3.1 Protease and Inhibitor Balance
The levels of serine proteases (Trypsin type) and inhibitor, and DPP-4 were measured in normal proximal tubular kidney cells (Vero): primary human kidney mesangial gluomerlar cells (NHMC); mouse C1C12 myoctyes cells; and human breast carci-noma cells SKBR-3 and MDA-468. All regular growing cells exhibited signifi cant levels of trypsin enzymatic activities as shown in Table 10.1 . Proximal tubular kidney and carcinoma cells had the most activity. Glomerular mesangial cells and myocytes had the least trypsin activity. Trypsin levels were higher in membrane pellets than in soluble cellular supernate. Trypsin inhibitors were detected by inhibition assays. However, the balance of trypsin/inhibitor in these cells favored greater trypsin than inhibitor by 8–70 times (ug/ug). All cells were shown to lack the Bik inhibitor in both ELISA and western blot immunoassays. Exposure of cells to 50 mg/L Bik shifted the protease to inhibitor balance in favor of greater inhibitor by a 1.5 to 10-fold (ug/ug).
10.3.2 Cellular Signal Response Due to Bik
Vero cells were induced with sterile Bik in serum-growth media at the G0 phase. Cell numbers declined by approximately half of initial cell counts at 24 h after Bik exposure, as shown in Fig. 10.1 (closed circles). Control cell cultures were revived with serum growth media and continued to double during the same time period.
10 Uristatin Anti-infl ammatory Cellular Signaling
178
Table 10.1 Protease and inhibitor balance
Cell type a – Trypsin b (μg/mg protein)
Trypsin Inhibitors c, d (μg/mg protein)
DPP-4 e (ng/mg protein)
Proximal tubular Supernatant 132 8.9 2.2 Pellet 243 15.2 19.3
Glomerular mesangial
Pellet 17 2.4 32.4
Myoctyes Pellet 20 1.9 nil Carcinoma Supernatant 213 3 467
Pellet 445 6.3 –
Legend: a Normal proximal tubular kidney cells were African green monkey cells (Vero). A primary human kidney cell line was used for mesangial gluomerlar cells (NHMC). Myoctyes were mouse C1C12 cells. Human breast carcinoma cells were SKBR-3 and MDA-468. Cells were lyzed and homog-enates separated in membrane-bound pellet residues and supernate b Trypsin protease activity was measured using trypsin specifi c chromogenic substrate calibrated to trypsin standards c Trypsin inhibitors were measured by the inhibition of trypsin substrate hydrolysis d Bikunin (Bik) levels were ≤0.1 mg/L for all cell extracts measured by ELISA and western blot immunoassays e Dipeptidyl Peptidase 4 (DPP-4), was measured by ELISA
0
2.5
5
7.5
10
12.5
15
17.5
20
22.5
0 10 20 30 40 50 60
Bik (mg/L)
Kid
ney
cel
l x10
6 /mL
0
50
100
150
200
250
300
350
400
450
DP
P-4
(n
g/m
L)
Fig. 10.1 Normal cell Loss and DPP-4 release upon exposure to Bik. Live cell count was per-formed with Trypan Blue staining. Cell cultures exposed to 25 mg/L Bik lost 50 % of cells within a 24 h period. Kidney cell (Vero) counts ( closed circles ) decreased with increasing Bik. In addi-tion, Vero released DPP-4 ( open circles ) upon increased Bik exposure (units on right hand axis). Exposures and analysis were repeated in three separate trials
M. Basu et al.
179
Phosphorylation of MAPK-ERK and MAPK-p38 was unchanged between control and exposed cells (data not shown here).
10.3.3 Membrane Disruption Due to Bik
Membrane disruption after Bik exposure was detected by membrane phosphatidyl-serine (PS)-binding dyes PSS-380 (see Table 10.2 ). PSS-380 binds PS on outer leafl et of cell membrane which presents only when the cell undergoes the dying process (Ma et al. 2004 ) Phosphatidylserine externalization was detected after 4 h of Bik treatment and increased with Bik amounts from 10 to 50 mg/L, as shown by the increased blue PSS-380 staining. Incorporation of the propidium iodide into the cell nucleus DNA was observed at a lesser extent after 24 h (data not shown here).
Apoptosis, or programmed cell death, was measured by western blot using anti-bodies for the intra-cellular caspase proteases 3, 8, and 9 (see Table 10.2 ). The absence of apoptosis signaling was confi rmed in all cell models before and after exposure to Bik. Western blots showed no cleavage of caspase 3 into active forms (see Table 10.2 ). Inactive caspases 9 confi rmed no intrinsic apoptosis signaling occurred. Caspases 8 remained inactivate which supported no extrinsic apoptosis signaling. There was no activation of apoptosis at extreme Bik exposure levels of 500 mg/L.
10.3.4 Calcium In-fl ux Blocking by Bik
The cellular calcium (Ca 2+ ) was measured by fl uorescent microscopy using a Ca 2+ sensitive dye (Fluo-4 AM) (as previously described in Chap. 7 ). Harvested myo-cytes were treated with Bik or Uri. Muscle, and kidney cell cultures typically main-tained intra-cellular Ca 2 in the range of 0.08–0.12 μM. Cells exposed to Bik exhibited signifi cantly reduced intra-cellular Ca 2 concentration. Intra-cellular Ca 2 decreased to 0.01 μM with increasing Bik concentrations. Cell lysates showed Bik was maintained on cell surfaces after washing.
10.3.5 Release of DPP-4 Due to Bik
We found kidney cell cultures did not release DPP-4 into the cell supernate (<10 ng/mL) unless cell cultures were exposed to Bik (see Table 10.2 and Fig. 10.1 ). This release increased with concentrations of Bik. Western blots showed released DPP-4 from membrane as the expected 110 kDa monomeric form along with bands 77, 50, and 45 kDa for fragments. Release of DPP-4 occurred within 4 h of Bik exposures and was reproducible in separate experiments.
Dipeptidyl Peptidase 4 (DPP-4) was detected more in cell membranes com-pared to cell supernate. In agreement with literature, DPP-4 was detected in kidney
10 Uristatin Anti-infl ammatory Cellular Signaling
180
Tabl
e 10
.2
Stre
ss r
espo
nse
in k
idne
y an
d m
yocy
tes
cells
exp
osur
e to
Bik
Bik
mg/
L a
Cas
pase
3 b
Cas
pase
8 b
Cas
pase
9 b
Intr
a-
cellu
lar
calc
ium
(μ
M) c
% C
ells
with
out
er
mem
bran
e di
srup
tion d
% C
ells
with
in
ner
mem
bran
e di
srup
tion d
Inac
tive
(32
kDa)
A
ctiv
e (1
7 kD
a)
Inac
tive
(57
kDa)
A
ctiv
e (4
3/18
kD
a)
Inac
tive
(47
kDa)
A
ctiv
e (3
7 kD
a)
At 4
h
At 4
h
At 2
4 h
At 4
h
At 2
4 h
0 +
–
+
– +
–
0.93
0
%
2 %
0
%
0 %
12
.5
+
– +
–
+
– 0.
47
17 %
42
%
0 %
5
%
25
+
– +
–
+
– 0.
09
NM
N
M
NM
N
M
50
+
– +
–
+
– 0.
01
26 %
57
%
0 %
14
%
500
+
– +
–
+
– 0.
01
NM
N
M
NM
N
M
a Syn
chro
nize
d cu
lture
s fo
r no
rmal
kid
ney
cells
(V
ero)
and
myo
cyte
s (C
2C12
) w
ere
indu
ced
with
Bik
at
0, 1
2.5,
25,
50,
and
500
mg/
L. C
ells
wer
e ha
rves
ted
afte
r 4
and
24 h
. Exp
osur
es w
ere
repe
ated
in th
ree
sepa
rate
tria
ls
b Apo
ptot
ic c
aspa
se c
leav
ages
wer
e m
easu
red
by w
este
rn b
lots
usi
ng c
ell
extr
acts
with
ant
ibod
ies
for
casp
ase
prot
ease
s (3
/8/9
). I
nact
ive
and
activ
e ba
nds
are
iden
tifi e
d by
mol
ecul
ar w
eigh
ts a
s sh
own
abov
e an
d ar
e in
dica
ted
(+)
for
pres
ent a
nd (
−)
for
abse
nt
c Int
ra-c
ellu
lar c
alci
um w
as m
easu
red
by in
cuba
ting
the
Ver
o ce
lls w
ith m
embr
ane
perm
eabl
e ca
lciu
m s
ensi
tive
Fluo
-4 A
M fo
r 30
min
at 3
6 °C
to a
llow
intr
acel
-lu
lar h
ydro
lysi
s to
cau
se fl
uore
scen
ce in
the
pres
ence
of
calc
ium
. Cel
ls w
ere
was
hed
and
cent
rifu
ged
to r
emov
e ex
tra-
cellu
lar
dye,
and
fl uo
resc
ent i
mag
es w
ere
capt
ured
. Mic
rosc
opic
imag
es o
f B
ik tr
eate
d ki
dney
cel
ls w
ere
colle
cted
by
phas
e co
ntra
st a
nd fl
uore
scen
ce m
icro
scop
e. T
he %
of
cells
with
fl uo
resc
ence
was
de
term
ined
by
area
com
pari
sons
d M
easu
rem
ents
of
oute
r m
embr
ane
disr
uptio
n w
ere
done
with
cel
l m
embr
ane-
bind
ing
dye
PPS-
380
whi
le i
nner
mem
bran
e di
srup
tion
was
mon
itore
d w
ith
DN
A-b
indi
ng d
ye p
ropi
dium
iodi
de. (
NM
= n
ot m
easu
red)
M. Basu et al.
181
and carcinoma cells but not myocytes (see Table 10.1 ) (Mentlein 2004 ). The high concentration of DPP-4 in proximal tubular cells was expected as the peptidase is part of the brush border membrane and participates in protein absorption by hydro-lysis of proline containing peptides (Mueller et al. 1990 ; Thompson et al. 1985 ). However the presence of DPP-4 was not previously reported for glomerular mesangial cells.
10.3.6 Proliferation
The high level of Trypsin and Trypsin inhibitor expression in the proximal tubular cells refl ect their dynamic living status. Proximal tubular epithelial cells play a piv-otal role in kidney disease. Most renal cell carcinoma and kidney cancer arises from this region. During infl ammation, endothelial and epithelial cellular proliferation is signaled by trypsin-type serine proteases which generate wound healing. Trypsin and thrombin cleave Protease Activated Receptors (PAR) which increases phospho-lipase C (PLC), adenylcyclase (AC), and phosphatidylinositol 3-kinase (PIK3) acti-vation of ERK 1/2 mitogen-activated protein kinase (MAPK/ERK). MAPK/ERK activation plays a key role in cell growth and differentiation through transcriptional activation. Active PAR also increases cytosolic phospholipase A 2 (PLA2) synthesis of prostaglandins causing nuclear receptor signaling of transcriptions.
10.3.7 Activation of PI3K-Akt Pathway by Bik
Phosphorylated and non-phosphorylated forms of PI3K were measured by ELISA with cell cultures before and after exposure to Bik and Uri (see Fig. 10.2 ). Activated PI3K was increased in cells when exposed to Bik and Uri. Increased activity would reduce immobilization of calcium from storage organelles by inositol 1,4,5 triphos-phate. Phosphorylation of PI3K increased more with the exposure to Uri. Protein kinase B (Akt) is an important target of increased PI3K phosphorylation. Phosphorylated and non-phosphorylated forms of Akt were measured at the same time points using ELISA. Cell exposed to Bik did not increase in Akt phosphoryla-tion. Cells exposed to Uri showed Akt activation in a dose-dependent manner (see Fig. 10.2 ). The highest level of Akt phosphorylation occurred at 500 mg/L of Uri. A drop in PI3K activation was noted at 500 mg/L.
10.3.8 Bik Active Site Study
The structure indicates the two Kunitz domains and two oligosaccharide chains. Colored peptides were labeled with letters in parentheses “(A) to (E)” (see Fig. 10.3 ).
10 Uristatin Anti-infl ammatory Cellular Signaling
182
The mAb 3G5 responded well to Bikunin and Uristatin at ~1 ng/well and showed no cross-reactivity to aprotinin even at an excess of 10,000 ng/well. The mAb 3G5 also responded to peptide B at 362 ng/well, that supporting the epitope in this region. The affi nity for purifi ed Bikunin standard was compared to that for native Bikunin in diabetic urine specimens. The antibody affi nity of ~2.9 × 10 11 was reduced for many diabetics when the urine was not diluted. Diluting urine (5 μL) in PBS (45 μL) eliminated the urine sample matrix affect. Undiluted urine inhibiting mAb binding in some patients supports additional associating molecules.
Trypsin inhibition was observed with Bik, Uri, and aprotinin (see Table 10.3 ). The loss of the O-linked glycan in Uri did not impact inhibitory strength. Synthetic peptides A–E were not inhibitory to trypsin. The results suggested that a secondary structure for the bikunin molecule is needed for trypsin inhibition. The mAb 421.3G5 was able to bind only fragment B, which is the region for N-linked polysaccharide. The affi nity is much lower than for the purifi ed Bik standard (362 ng/well vs. 1 ng/well), indicating that this mAb recognizes both the amino acids and N-linked sugars of fragment B. Inhibition of trypsin was prevented when Bikunin was bound to mAb 3G5 in peptide region B (see Table 10.3 an Fig. 10.3 ). This supports the explanation that mAb was binding the region B of Bik where is active sitde of inhibition.
0.80
0.90
1.00
1.10
1.20
1.30
1.40
1.50
1.60R
elat
ive
incr
ease
from
bas
al a
ctiv
ity
Control Bik(5 mg/L)
Uri(5 mg/L)
Uri(50 mg/L)
Uri (500 mg/L)
Bik (50 mg/L)
Bik(500 mg/L)
Fig. 10.2 Activation of PI3K/Akt with Bikunin and Uristatin. The relative increases in phospho-ralyation of phosphatidylinosinol 3 kinase (PI3K) ( solid bars ) and Akt ( open bars ) are shown for epitheilal (Vero) cell cultures simulated with 0, 5, 50, and 500 mg/L of Bik or Uri. Measurements were performed using immunoassays for both phosphoralyation and non- phosphoralyation forms of PI3K and Akt. Exposures and analysis were repeated in three separate trials
M. Basu et al.
183
Bikunin active sites are often studied by comparison to aprotinin’s single Kunitz binding domain. For Bikunin Kunitz domain 1 (see Fig. 10.3 ), however, the location of the predicted active site lacks an arginine (R) or lysine (K) residue which can be locked into the trypsin pocket. The likely binding site is the arginine peptide in the middle of peptide region B. The region B arginine is connected to N-linked glycan (RYFYN), an atypical structure of a glycoconjugated protease inhibitor (Otlewski et al. 2001 ). For Bikunin Kunitz domain 2, the predicted trypsin binding site is an arginine (R)-containing peptide. In our study, we characterized the mAb 3G5 bind-ing site as the active site for serine protease inhibition, located in the N-linked oli-gosaccharide region.
GDGD EELLRFSN
DSCQLGYSAG PCMGMTSRYF YNGTSMACET FQYGGCMGNG NNFVTEKECL QTCRTV
AACNLPIVRG PCRAFIQLWA FDAVKGKCVL FPYGGCQGNG NKFYSEKECR EYCGVP
Kunitz domain 1
Kunitz domain 2
C-terminal tail
AVLPQEEEGS GGGQLVTEVT KK
N-terminal tail
A
Xyl-Gal-Gal-GlcUA-(NeuAc-GlcUA)5-(GalNAc-GlcUA)20-GalNAc-GlcUA
GlcNAc-GlcNAc-Man-(Man-GlcNAc-Gal-NeuAc)2
ED
CBTrypsin
inhibition site
N-linked glycan
O-linked glycan with chondroitin sulphate
Fig. 10.3 Peptide analogs and aprotinin comparison to Bik predicted structure. The predicted full length Bikunin (Bik) peptide and carbohydrate sequence is compared to fi ve synthetic peptides and aprotinin. Peptides in bold are a match to the aprotinin sequence. Underlined peptides are the pre-dicted contact points between trypsin- aproptinin based on X-ray structure. This single Kunitz binding domain has matching peptides ( darkened circles ) to both inhibitor domains of Bik. The location of the predicted active site ( Underlined circles ), an arginine (R) or lysine (K) peptide which can be locked into the trypsin pocket is shown for Bik domain 1 and 2. The likely antibody binding site is the arginine peptide in the middle of peptide region B. The region B arginine is con-nected to N-linked glycan (RYFYN) and an example of a glycoconjugated protease inhibitor. For Bik domain II, the predicted trypsin binding site is an arginine (R) peptide. The other peptides of the domain II active site also match the preferred and strongest trypsin binding sequences of GPCR and are examples of a peptide-only protease inhibitor This active site was not mimicked by peptide D as the native sequence must be in a distorted conformation to cause trypsin inhibition
10 Uristatin Anti-infl ammatory Cellular Signaling
184
10.4 Discussion
Bik readily passes through the glomerular basement membrane and accumulates in the lysosomes of proximal tubular epithelial cells, and is not reabsorbed in the con-densing vesicles. As Bik inhibits the proteolytic process in proximal tubular epithe-lial cells and shifts the protease activity to inhibitor balance, it is reasonable that Bik would inhibit the proximal tubular from proteolysis and protein reabsorption in the kidney. Inhibition of tubular reabsorption would result in LMW proteinuria during periods of infl ammation. This is in agreement with clinical observation that Bik cor-relates well to proteinuria during fever and infection (Pugia et al. 2002 ; Pugia et al. 2004 ; Jortani et al. 2004 ).
Glomerulonephritis is the lead cause of renal injury leading to failure and is typi-cally associated with infections (Hotta et al. 1999 ; Huang et al. 2001 ). Tissue dam-age by White Blood Cell (Neutrophilic polymorphonuclear leukocytes and macrophages) does cause capillary wall injury in glomerulonephritis mediated by the release of proteinases (Nakakuki et al. 2001 ). Elastase and cathespin are known to damage the basement membrane leading to proteinuria and enlargement of the network structure the basement membrane and charge barrier (Cochrane et al. 1965 ). Serine proteases such as elastase, proteinase 3, and cathespin cause the release of apoptotic cytokines such as tumor necrosis factor-α (TNFα), interleukin-1β (IL-1β), interleukin- 18 (IL-18β), and transforming growth factor -α (TGF α) (Mania-Pramanik et al. 2004 ; Johnson et al. 1988 ; Nakatani and Takeshita 1999 )
The intrinsic and extrinsic caspase pathways were not activated after Bik expo-sure in this study (Basu et al. 2014 ). Bik inhibitors, like aprotinin, have been shown to prevent apoptosis in animal models of ischemia-mediated kidney damage
Table 10.3 Bikunin active site study for trypsin inhibition and antibody binding
Inhibitor a
% Trypsin inhibition b Antibody binding in ng/well c
Restored of trypsin inhibition d
Average (standard deviation)
Average (standard deviation)
Average (standard deviation)
Peptide A 0 % NB 0 % Peptide B 0 % 362 (0.66) 35 % (16 %) Peptide C 0 % NB 0 % Peptide D 0 % NB 0 % Peptide E 0 % NB 0 % Bik 76 % (3 %) 1.25 (0.43) 89 % (7 %) Uri 89 % (5 %) 1.03 (0.12) 72 % (12 %) Aprotinin 58 % (2 %) NB 0 %
a Synthetic bikunin peptide fragments A to E, native Bik, Uri and aprotinin standards were used to study the trypsin inhibitory effect b (column 2), antibody binding ability to mAb 421.3G5 c (column 3), and restoration of trypsin inhibition activity d (column 4). NB = “No Binding”
M. Basu et al.
185
(Kher et al. 2005 ). Elastase, proteinase 3, and cathespin, are all inhibited preventing subsequent infl ammatory cytokine and growth factors release. Our results in normal cell models, without infl ammatory cytokine stimulation, did show that Bik does not activate apoptosis supporting the anti-apoptotic action of Bik on leukocytes prote-ases. Additionally, Bik activation of PI3K-Akt could activate NF-κB independently of TNF-α which prevents Caspase 3 activation (Vivanco and Sawyers 2002 ).
Necrosis is a distinguished from of cell death where membrane disruption occurs without the programmed process that can be observed in apoptosis (Skulachev 2006 ). Osmotic changes within the cell cause outer membranes to expand and rup-ture during necrosis. Necrotic cell death, not apoptosis, occurred in normal cells exposed to Bik. As Fig. 10.4 shows, during this process, Bik continued to protect against immune mediated apoptosis and infl ammatory signaling of cell prolifera-tion. However, Bik exposure induced stress on the cell. Membrane disruption was
Ca+
Out-fluxNF-κBActivation
No Apoptosis(Caspases 3, 8, 9)
InhibitsPAR
activation
MAPK/ERK MAPK p38
Reducedcytokine &
growth factor release
Inhibitedgrowth factor
signaling
Bik-RCD44 Bik
ReduceduPAR
activation
• uPARLP
BikuPA
BikCa+
KCa ChannelK+
K+
K+
K+
Ca+
Ca+Ca+
Ca Channel
Bik
Bik
ReducedK + influx
Inhibited extrinsic death factor
& NF-KB signaling
Akt&PI3KActivation
Slows cell Inflammatory proliferation
(~50% cell/24h)
Granzyme Binhibition
Bik
Membrane disruption via osmotic changes
Fig. 10.4 Impact of Bik on cell signaling. Kunitz inhibitor are known to stop PAR activation reducing cell proliferation through MAPK/ERK. Additionally, activation of MAPK is inhibited through CD44 binding to growth factors (EGF, TGF-α, and TGF-β) upon Bikunin (Bik) has been shown vis interaction with its receptor (Bik-R). Bik is also known to inhibits release of growth factors and cytokines, subsequently reducing extrinsic death factor (FasL) and NF-KB signaling (TNFR). Bik reduces expression of uPA and inhibits membrane-bound plasmin, impacting mem-brane stabilization through binding to link protein (LP). Bik is also known to inhibit Granzyme B from activating apoptosis. We here report, no activation of apoptosis via Caspase 3, 8, and 9 was observed after Bik exposure. There was no release of DNA in fl uorescence microscopic cell images. No activation of MAPK p38 was observed after Bik exposure. PI3K-Akt phosphorylation was observed after Bik exposure. Bik reduced intracellular Ca 2+ in the fl uorescence microscope cell images which could explain PIK3-Akt activation. Membrane disruption and cell death was observed by phosphatidyl serine membrane-changes in fl uorescence microscope cell images and release of membrane bound ADBP-26 increased. Exposure causes cell proliferation slows to half the rate seen with non-exposed cells (~50 % cell/24 h)
10 Uristatin Anti-infl ammatory Cellular Signaling
186
clearly evident by phosphatidylserine externalization along with decreases in micro-scopic cells counts. This change could be partially due to the blocking the K Ca chan-nels with the changes of membrane permeability. Additionally, DPP-4 was released from the membrane. This is consistent with DPP-4 used as a diagnostic marker of acute renal necrosis (Mueller et al. 1990 ; Thompson et al. 1985 ).
Proximal tubular epithelial cells play a pivotal role in kidney disease. The high level of Trypsin and Trypsin-inhibitor expression in the proximal tubular cells refl ect their dynamic living status. Most renal cell carcinoma and kidney cancer arise from this region (Tomita 2006 ). Normal cellular proliferation is central to kid-ney regeneration and repair. During infl ammation, endothelial and epithelial cellu-lar proliferation is signaled by trypsin-type serine proteases which generate wound healing. Trypsin and thrombin cleave Protease Activated Receptors (PAR) which increases phospholipase C (PLC), adenylcyclase (AC), and phosphatidylinositol 3-kinase (PIK3) activation of ERK 1/2 mitogen-activated protein kinase (MAPK/ERK) (Bono et al. 1997 ). MAPK/ERK activation plays a key role in cell growth and differentiation through transcriptional activation. Active PAR also increases cytosolic phospholipase A 2 (PLA2) synthesis of prostaglandins causing nuclear receptor signaling of transcriptions.
Previous groups have shown Bik inhibits cell proliferation in a carcinoma cell model (Kobayashi 2001 ). In carcinoma, infl ammatory signaling is an important component of proliferation. Bik inhibition of trypsin and thrombin prevents activa-tion of PAR. Bik inhibition of plasmin prevents activation of urokinase-type plas-minogen activator which breaks down the basement membrane during tissue remodeling. Infl ammatory MAPK activation is also inhibited through Bik interac-tion with CD44 binding (Suzuki et al. 2002 ). Our studies with normal cells, showed Bik does not completely stop cell proliferation and that non-infl ammatory activa-tion of MAPK is not impacted. Activation of MAPK ERK through the tyrosine- specifi c protein receptors or the G-protein-coupled receptors is not impacted with Bik. This is consistent with Bik only inhibiting proliferation due to infl ammation signaling through PAR and uPAR.
Bik exposure caused signifi cant loss of intra-cellular Ca 2+ favoring osmotic stress and cell lysis. Reduction of intracellular Ca 2+ is consistent with suppression of vas-cular smooth muscle contraction by Bik (Moczydlowski et al. 1992 ). Kunitz-type protease inhibitors are known to bind Ca 2+ activated K + channels causing modula-tion (Moczydlowski et al. 1992 ). Bik is similar in peptide sequence to aprotinin and might be acting through same mechanism. The binding causes a blockage of the channel whether by partial occlusion, repulsion or allosteric interaction. Blockage of the Ca 2+ sensitive potassium (K Ca ) channels (or calcium-activated potassium (BK) channels) is thought to lower the intracellular Ca 2+ . The ability of K + to exit the cell is responsible for the resting membrane potential and controls the Ca 2+ in- fl ux through the voltage sensitive Ca 2+ channels (Jensen et al. 2001 ). A reduced potas-sium in-fl ux is counterbalanced by the calcium channels (Kanayama et al. 1997 ).
This change in intracellular Ca 2+ would be expected to impact the calcium dependent pathways The calcium channel potentially explains the observed activa-tion of PI3K and Akt signaling. Other groups have recently confi rmed uTi that
M. Basu et al.
187
activate PI3K-Akt and ERK signal transduction and inhibit of p38 MAPK and JNK (Kim et al. 2009 ). The uTi affect is of great interest as the PI3K-Akt -mTOR path-way is an important topic in cancer where cells have reduced apoptosis and increased proliferation (Vivanco and Sawyers 2002 ). Protein kinase C and phospholipase C could also be impacted by decreased intra-cellular calcium. The difference in PI3K and Akt signaling observed between the Uri and Bik could be due to Uri lacking the O-linked chondrotin sulfate chain which promotes calcium binding. Previous stud-ies have shown the chondroitin sulfate chains of Bik impact muscle contraction (Kanayama et al. 1997 ).
We also found adiponectin (Adpn) signaling of G-protein coupled receptor (AipoR1) restored cytosolic calcium in the presence of Bik. Therefore Adpn and Bik activate the anti- infl ammatory responses through separate but dependent cell signal pathways of metabolic importance. Adpn initiates glycolysis through AMPK (Mao et al. 2006 ; Deepa and Dong 2009 ). Bik activates PIK3-Akt and could interfere with insulin receptor signaling (Deepa and Dong 2009 ; Kobayashi et al. 2005 ; Bertrand et al. 2008 ). Bik inhibits proliferation without apoptosis and decreases intra-cellular Ca 2 . Adpn counteracts the impact of Bik on the intra-cellular Ca 2+ and could be modulating intra-cellular Ca 2+ through PI3K (Marcantoni et al. 2006 ; Lu et al. 2005 ).
10.5 Conclusion
Our results suggest that kidney exposure to Bik during abnormal stress might con-tribute to reduced renal regeneration with the signal transduction shown below. Bik/Uri slowed kidney cell proliferation by more than half with exposure at 25 mg/L.
• Bik/Uri did not activate caspases or mitogen-activated protein kinases (MAPK). • Natural cell death increased with over exposure of cells to Bik/Uri. • Bik/Uri caused a release of membrane-bound proteins, Dipeptidyl Peptidase 4
(DPP-4) and phosphatidylserine externalization. • Bik/Uri activated Phosphatidylinositol 3-kinase (PIK3) and protein kinase B
(Akt) • Bik/Uri reduced intra-cellular calcium in treated cells • The N-Glycan structure of the Uri and Bik is needed for trypsin inhibition. • The mAb 421.3G5 was able to bind the inhibitor region of Uri and Bik.
Under infl ammatory stress condition the amounts Bik inhibitors overwhelm the proteases involved in proliferation and proteolysis. While these biochemical cell models support that Bik does not cause apoptosis and could protect tissues damag-ing immune cells, endothelial and epithelial proliferation is also slowed cell growth which would reduce regeneration and repair (during wound healing) and cell prolif-eration (PKI3-Akt).
All of the information discussed here may provide new drug target design clues for treatment of kidney infl ammatory diseases. For example, the Bik site could be blocked. Previously, a report showed a synthetic trypsin inhibitor could reduce the
10 Uristatin Anti-infl ammatory Cellular Signaling
188
urinary albumin excretion in diabetic rats and human patients during the diabetic renal hypertrophy (Ikeda and Hoshino 1996 ; Ikeda et al. 1999 ).
Acknowledgements We would like to acknowledge the laboratory work of the Notre Dame undergraduate researchers in enzyme studies and cell models done by: Vesta Anilus, Yoonie Cho, Kunal Saxena, Kimberly St. Jean, and Gabriel J. Crawford. We would like thank Dr. Sipra Banerjee of the Cleveland Clinic Foundation for the gift of cancer cell lines. We would like thank Dr. Rui Ma for his helpful efforts with advise to the students. Part of this research was supported by the Jacobs Javits Research Grant Award from NIH-NINDS (NS-18005) and NCI R01 Grant (CA- 14764) awarded to SB and a grant-in-aid from Colman Cancer Foundation to MB.
Supplements
• Western Blot Methods • ELISA Method for Uristatin
References
Basu S, Ma R, Moskal JR, Basu M (2014) Regulations of glycolipids:glycosyltransferases (GSL:GLTs) involved in SALex biosynthesis in carcinoma cells. XII. Biosimilar apoptotic agents- Potential Anticancer Drugs. Adv Expt Med Biol ed Chakrabarti, C, Surolia A Springer 842:329–345
Bertrand L et al (2008) Insulin signaling in the heart. Cardiovasc Res 79(2):238–248 Bono F, Lamarche I, Herbert JM (1997) Induction of vascular smooth muscle cell growth by selec-
tive activation of the proteinase activated receptor-2 (PAR-2). Biochem Biophys Res Commun 241:762–764
Cochrane CG, Unanue ER, Dixon FJ (1965) A role of polymorphonuclear leukocyte and comple-ment in nephrotoxic nephritis. J Exp Med 122:99–116
Deepa SS, Dong LQ (2009) APPL1: role in adiponectin signaling and beyond. Am J Physiol Endocrinol Metab 296(1):E22–E36
Hirashima Y, Kobayashi H, Suzuki M et al (2001) Characterization of binding properties of urinary trypsin inhibitor to cell-associated binding sites on human chondrosarcoma cell line HCS-2/8. J Biol Chem 276:13650–13656
Hochstrasser K, Wachter E, Bretzel G (1967) Liberation of Kunitz-type inhibitors from the inter-alpha- trypsin-inhibitor by limited proteolysis. Proc FEBS 47:225–234, Proteolytic enzymes and their inhibitors
Hotta O, Yusa N, Ooyama M, Unno K, Furuta T, Taguma Y (1999) Detection of urinary macro-phages expressing the CD16 molecule: a novel marker of acute infl ammatory glomerular injury. Kidney Int 55:1927–1934
Huang F, Horikoshi S, Kurusu A, Shibata T, Suzuki S, Funabiki K et al (2001) Urinary levels of interleukin-8 (IL-8) and disease activity in patients with IgA nephropathy. J Clin Lab Anal 15:30–34
Ikeda T, Hoshino T (1996) An inhibition of urinary albumin excretion by protease inhibitor in streptozotocin-diabetic rats. Nephron 74(4):709–712
Ikeda Y, Ito H, Hashimoto K (1999) Effect of camostat mesilate on urinary protein excretion in three patients with advanced diabetic nephropathy. J Diabetes Complications 13(1):56–58
M. Basu et al.
189
Jensen BS et al (2001) The Ca2+- activated K+ channel of intermediate conductance: a molecular target for novel treatments? Curr Drug Targets 2(4):401–422
Joberg ME, Blom A, Larsson BS, Alston-Smith J, Mats S, Fries E (1995) Plasma clearance of rat bikunin: evidence for receptor-mediated uptake. Biochem J 308:881–887
Johnson R, Couser W, Alpers C, Vissers M, Schulze M (1988) The human neutrophils serine pro-teinases, elastase and cathespin G can mediate glomerular injury in vivo. J Exp Med 168:1169–1174
Jortani SA et al (2004) Sensitive noninvasive marker for the diagnosis of probable bacterial or viral infection. J Clin Lab Anal 18(6):289–295
Kanayama N, Maehara K, Suzuki M et al (1997) The role of chondroitin sulfate chains of urinary trypsin inhibitor in inhibition of LPS-induced increase of cytosolic free Ca2+ in HL60 cells and HUVEC cells. Biochem Biophys Res Commun 238:560–564
Kanayama N et al (1998) Urinary trypsin inhibitor suppresses vascular smooth muscle contraction by inhibition of Ca2+ infl ux. Biochim Biophys Acta 1381:139–146
Kher A et al (2005) Aprotinin improves kidney function and decreases tubular cell apoptosis and proapoptotic signaling after renal ischemia-reperfusion. J Thorac Cardiovasc Surg 130(3):662–669
Kim SJ et al (2009) Urinary trypsin inhibitors afford cardioprotective effects through activation of PI3K-Akt and ERK signal transduction and inhibition of p38 MAPK and JNK. Cardiology 114(4):264–270
Kobayashi H (2001) Suppression of urokinase expression and tumor metastasis by bikunin over-expression [mini-review]. Hum Cell 14(3):233–236
Kobayashi T, Matsumoto T, Kamata K (2005) The PI3-K/Akt pathway: roles related to alterations in vasomotor responses in diabetic models. J Smooth Muscle Res 41(6):283–302
Lindstroem KE, Blom A, Johnsson E, Haraldsson B, Fries E (1997) High glomerular permeability of bikunin despite similarity in charge and hydrodynamic size to serum albumin. Kidney Int 51:1053–1058
Lu Z et al (2005) Galpha q inhibits cardiac L-type Ca2+ channels through phosphatidylinositol 3-kinase. J Biol Chem 280(48):40347–40354
Ma R et al (2004) Apoptosis of human breast carcinoma cells in the presence of disialosyl ganglio-sides: II. Treatment of SKBR3 cells with GD3 and GD1b gangliosides. Glycoconj J 20(5):319–330
Mania-Pramanik J et al (2004) Elastase: a predictive marker of infl ammation and/or infection. J Clin Lab Anal 18:153–158
Mao X, Hong JY, Dong LQ (2006) The adiponectin signaling pathway as a novel pharmacological target. Mini Rev Med Chem 6(12):1331–1340
Marcantoni A et al (2006) Phosphoinositide 3-kinasegamma (PI3Kgamma) controls L-type cal-cium current (ICa, L) through its positive modulation of type-3 phosphodiesterase (PDE3). J Cell Physiol 206(2):329–336
Mentlein R (2004) Cell-surface peptidases. Int Rev Cytol 235:165–213 Milner CM, Day AJ (2003) TSG-6: a multifunctional protein associated with infl ammation. J Cell
Sci 116:1863–1873 Moczydlowski E, Moss GW, Lucchesi KJ (1992) Bovine pancreatic trypsin inhibitor as a probe of
large conductance Ca(2+)-activated K+ channels at an internal site of interaction. Biochem Pharmacol 43(1):21–28
Mueller PW et al (1990) Indicators of acute renal-transplant rejection in patients treated with cyclosporine. Clin Chem 36(5):759–764
Nakakuki M et al (1996) Protective effect of human ulinastatin against gentamicin-induced acute renal failure in rats. Can J Physiol Pharmacol 74:104–111
Nakakuki M et al (2001) Inhibitory effect of serine protease inhibitors on neutrophil-mediated endothelial cell injury. J Leukoc Biol 69:241–247
Nakatani K, Takeshita S (1999) Vascular endothelial cell injury by activated neutrophil and treat-ment for the injury. Surg Trauma Immunol Respon 8:112–114
10 Uristatin Anti-infl ammatory Cellular Signaling
190
Ohlson M, Sorensson J, Lindstrom K, Blom AM, Fries E, Haraldsson B (2001) Effects of fi ltration rate on the glomerular barrier and clearance of four differently shaped molecules. Am J Physiol 281:F103–F113
Otlewski J et al (2001) Structure-function relationship of serine protease-protein inhibitor interac-tion. Acta Biochim Pol 48(2):419–428
Pugia MJ, Lott JA (2005) Pathophysiology and diagnostic value of urinary trypsin inhibitors. Clin Chem Lab Med 43(1):1–16
Pugia MJ et al (2002) Clinical utility of a rapid test for uristatin. Clin Biochem 35(2):105–110 Pugia MJ et al (2004) The uristatin dipstick is useful in distinguishing upper respiratory from uri-
nary tract infections. Clin Chim Acta 341(1–2):73–81 Pugia MJ et al (2007a) Immunological evaluation of urinary trypsin inhibitors in blood and urine:
role of N- & O-linked glycoproteins. Glycoconj J 24(1):5–15 Pugia MJ, Valdes R Jr, Jortani SA (2007b) Bikunin (urinary trypsin inhibitor): structure, biological
relevance, and measurement. Adv Clin Chem 44:223–245 Skulachev VP (2006) Bioenergetic aspects of apoptosis, necrosis and mitoptosis. Apoptosis
11(4):473–485 Smith BD et al (2003) Detection of apoptotic cells using a synthetic fl uorescent sensor for mem-
brane surfaces that contain phosphatidylserine. Cell Death Differ 10:1357–1359 Sumi H et al (1981) Human urinary trypsin inhibitors purifi ed by affi nity chromatography on
trypsin- sepharose. Nihon Ketsueki Gakkai Zasshi 44(1):146–153 Suzuki M et al (2002) Kunitz-type protease inhibitor bikunin disrupts phorbol ester-induced oligo-
merization of CD44 variant isoforms containing epitope v9 and subsequently suppresses expression of urokinase-type plasminogen activator in human chondrosarcoma cells. J Biol Chem 277(10):8022–8032
Takahashi M, Sawaguchi T, Sawaguchi A, Suzuki T (2001) The cytoprotective effect of protease inhibitor on programmed cell death of endothelial cell. Tokyo Joshi Ika Daigaku Zasshi 71:669–678
Thompson RE et al (1985) Adenosine deaminase binding protein, a new diagnostic marker for kidney disease. Clin Chem 31(5):679–683
Tomita Y (2006) Early renal cell cancer. Int J Clin Oncol 11(1):22–27 Vivanco I, Sawyers CL (2002) The phosphatidylinositol 3-Kinase AKT pathway in human cancer.
Nat Rev Cancer 2(7):489–501 Yamaski F et al (1996) Uptake of human urinary trypsin inhibitor by kidney epithelial cell line,
LLC-PK 1 . Pfl ügers Arch Eur J Physiol 433:9–15 Zhuo L, Salustri A, Kimata K (2002) A physiological function of serum proteoglycan bikunin: the
chondroitin sulfate moiety plays a central role. Glycoconj J 19:241–247
M. Basu et al.
193© Springer International Publishing Switzerland 2015 M. Pugia (ed.), Infl ammatory Pathways in Diabetes: Biomarkers and Clinical Correlates, Progress in Infl ammation Research, DOI 10.1007/978-3-319-21927-1_11
Chapter 11 Progress in New Markers for Diabetes Infl ammation
Michael Pugia
11.1 Introduction
The impact of hyperglycemia, obesity, and infl ammation add signifi cant tissue stress during diabetes. These stress factors cause damaging pathologies in the key affected tissues, namely the liver, brain, pancreas, muscle, adipose, and vascular system (Table 11.1 ).
11.2 Hyperglycemia Stress Markers
It has been clearly shown that hyperglycemia correlates to diabetic infl ammation, complications, and worsened progression to end stage disease. Under normal condi-tions, hyperglycemia induces insulin secretion by the pancreas (Table 11.1 ). Insulin reduces production of glucose and fatty acid oxidation by the liver as well as glu-cose stored by muscle (Akt/PIK3 pathway), and it decreases adipose energy storage and lipolysis. Glucose consumption by the brain and red blood cells is stimulated by glucose transport (GLUT1/4), and appetite suppression is signaled by the brain.
Insulin resistance is an early indicator of diabetic conditions, and this impair-ment leads to hyperglycemia. As such, glucose measurement remains the primary tool relied upon for diabetes diagnosis and monitoring. During fasting, the liver is the sole source of glucose production with the major route being gluconeogenesis (production of new glucose from noncarbohydrate precursors) and the minor route being glycogenolysis (break down of glycogen) Adipose is the major energy store
M. Pugia , PhD Strategic Innovation , Siemens Healthcare Diagnostics , 3400 Middlebury Street , Elkhart , IN 46515 , USA e-mail: [email protected]
194
of noncarbohydrate precursors (e.g. triglycerides lipolysis to free fatty acids and glycerol) for gluconeogenesis, and the brain and red blood cells are the major con-sumers of glucose. These organs normally act in concert, and impaired insulin metabolism leads to hyperglycemia stress across all organs along with reduced fatty
Table 11.1 Impact of stress factor on key tissue
Hyperglycemia Obesity Infl ammation
Liver ↑ Glucose production ↑ Insulin ↑ ROS ↑ AGE ↓ Fatty acid oxidation
↑Hyperglycemia ↑Insulin resistance ↑Cytokines/chemokines ↑Fatty acids (fatty liver) ↑ Innate immune cells
↓ Glucose production ↓ Insulin degradation/sens. ↑ Immune cell adhesion ↑ Apoptosis (↑ damage) ↑ Fibrosis (↓ response) ↑ IgG & complement
Brain ↑ Glucose consumption ↑Appetite suppression ↑ROS
↓Appetite suppression ↓Glucose consumption ↓ Insulin degradation/sens. ↑ IgG deposits ↑ Amyloid deposits
Pancreas ↑Insulin production ↑Insulin production ↓Insulin production ↓Glucose sens. ↑ Immune cell adhesion ↑ Apoptosis (↑ damage) ↑ Fibrosis (↓ response) ↑IgG & complement ↑ Auto-immune response
Adipose ↑ Adipokine response ↑ Energy storage ↓ Lipolysis ↑ FFA/TG
↓Adipokine response ↑Cytokines/chemokines ↑Hypoxia ↑Lipid loading ↓Lipolysis
↓ Insulin sens. ↑ Immune cell adhesion ↑ Apoptosis (↑ damage) ↑ Fibrosis (↓ response) ↑ IgG & complement
Muscle ↑ Glucose uptake ↑ Mitochondria ↓Fatty acid oxidation ↑ROS ↑AGE ↑Kallikrein (Kinins) ↑RAGE
↓ Glucose uptake ↑ IgG-CTF generation
↓Glucose consumption ↓Insulin degradation/sens. ↑ Immune cell adhesion ↑ Apoptosis (↑ damage) ↑ Fibrosis (↓ response)
Vascular system
↑ Vascular dysfunction ↓ NO (↑ROS) ↓ NO (↓eNOS by insulin) ↑Kinins ↑AGE (HbA1c) ↑glyCD59 (complement) ↑RAGE
↑ Vascular dysfunction ↓ NO (↑cytokines) ↑ Apidose AgnII ↑ Immune cell adhesion ↑ eNOS (AMBK)
↑ Vascular dysfunction ↑ Immune cell adhesion ↑ Fibrosis (↓elasticity) ↑ Membrane thickening ↑ Apoptosis ↑ Coagulation ↑ Complement
M. Pugia
195
acid metabolism. Hyperglycemia stress increases reactive oxidative species (ROS), advanced glycation end products (AGE), vascular dysfunction, and complement activation in the key tissues.
Measurements of glucose, insulin, and glycated hemoglobin (HbA1c) are essen-tial tools to manage hyperglycemia (Table 11.2 ) where the goal of hyperglycemic management is to use diet, exercise, and medications to obtain glucose values in a normal range. Insulin resistance is often the underlying cause of an inability to man-age hyperglycemia. As reviewed in Chap. 1 , many factors impact and correlate with insulin sensitivity. Insulin degradation is providing new understandings of insulin sensitivity. The direct or indirect measurement of insulin resistance is diffi cult to obtain and is best estimated using the traditional methods. However, the oral glu-cose tolerance test (OGTT) is not a spot test, and HbA1c is a late marker for predic-tion of diabetes in a group of prediabetes patients, and diabetes cannot be predicted from a single new marker. This reduces the ability to identify patients who are likely to progress to diabetes and respond to therapy. Additionally, hormones such as growth hormone growth hormone (GH), insulin-like growth factors (IGF), IGF binding proteins (IGFBP), epinephrine, glucagon, and cortisone impact the glucose metabolic pathways.
Advanced glycation end products (AGE) such as HbA1c have shown value for measurement of glycemic control over periods of 3–4 months. Whereas, other non-traditional AGE assays, such as glycated albumin, fructosamine, and 1,5- anhydroglucitol (1,5-AG), which measure glycemic control over a shorter period, have not proved more valuable than HbA1c (Beck et al. 2011 ). Correlations of complications with nontraditional AGE assays are typically less signifi cant.
All current AGE assays lack a direct mechanistic link to complications, and the receptors for advanced glycation end products (RAGE) have not provided that mechanistic link. While insulin increases the receptors for AGE (RAGE), activation of RAGE requires immune cell activation, and sRAGE levels do not correlate with glycemic control or complications (Lam et al. 2013 ; Motawi et al. 2013 ). The for-mation of AGE on glomerular basement membrane (GBM) directly correlates with HbA1c but is not linked to tissue damage or cellular proliferation without immune cell involvement. Next generation AGE markers such as glycated CD59 offer hope of a true mechanistic link (Chap. 2 ). Glycated CD59 could offer improvement over HbA1c by its correlation with glucose intolerance in short term glycemic control and by offering a molecular connection that is tied to a direct cell death pathway, as
Table 11.2 Markers of glycation stress
Measurement Markers Key cellular pathways
Insulin resistance/sensitivity Glucose AKT-PIK3 Insulin GLUT1/4 C peptide IDE
Glycemic control HbA1c Membrane glycation Glycemic control & complement stress
Glycated CD59 MAC (complement activation)
11 Progress in New Markers for Diabetes Infl ammation
196
exemplifi ed in the literature by an increased interest in the formation of membrane attack complex (MAC) in the compliment cascade (Qin et al. 2004 ).
Many therapeutic approaches have been allied to lowering hyperglycemia (Mazzola 2012 ), and insulin is by far the most effective means used. Other approaches include sulfonylureas and glinides, which lower glucose by enhancing insulin secretion; metformin which suppresses glucose production in the liver; α-glucosidase inhibitors, which reduce the rate of digestion of polysaccharides in the small intestine; thiazolidinediones (TZD), which activate PPARs (peroxisome proliferator-activated receptors) to increase insulin sensitivity of muscle, fat, and liver; incretins of the secretin family of hormones like Glucose-dependent insulino-tropic polypeptide (GIP) and glucagon-like peptide-1 (GLP-1), which improve hyperglycemia by indirectly stimulating insulin secretion and suppressing glucagon secretion; dipeptidyl peptidase-4 (DPP-4) inhibitors, which prolong incretin half- life; amylin agonists, which act as a neuroendocrine hormone to enhance satiety and reduce food intake; and sodium-glucose transporter 2 (SGLT2) inhibitors, which reduce glucose reabsorption in the proximal tubule of the kidney.
11.3 Markers of Oxidative Stress
While it appears clear that hyperglycemia leads to oxidative stress damage in tis-sues, the actual clinical value of measurements of ROS is less clear (Table 11.3 ). ROS reacts with DNA as the primary target to produce various byproducts such as 8-hydroxydeoxyguanosine (8-OHdG) (Hwang and Kim 2007 ). Oxidative stress can also be measured by oxidative modifi cation of proteins, peptides and amino acids, lipids, cell membranes, and small molecules. For example, markers such as nitro- tyrosine, modifi ed albumin, oxidized low density lipoprotein (ox-LDL), malondial-dehyde (MDA), and arachidonic acid derived oxidation products (plasma F(2)-isoprostanes) have been used to measure ROS.
Many enzymes, such as superoxide dismutase (SOD), nitric oxide synthase (NOS), myeloperoxidase (MPO), and glutathione peroxidase, process nitric oxide, a key ROS. In addition, small molecule REDOX acceptors also process ROS. These include vitamin C, vitamin E, and uric acid (Pitocco 2013 ; Sung et al. 2013 ). Glutathione (GSH) levels are a direct measurement of ROS generation (Motawi et al. 2013 ). For example, GSH levels decrease in diabetic patients due to increased mitochondrial oxidative activity and the formation of byproducts of NADH dehy-drogenase. Oxidative damage of DNA, amino acids, lipids, and small molecules also has unique repair systems such as DNA ligase and methionine sulfoxide reduc-tase (Hwang and Kim 2007 ). Lipoprotein-associated phospholipase A2 (Lp-PLA2) hydrolyzes oxidized phospholipids to generate lysophosphatidylcholine and oxi-dized fatty acids, and Lp-PLA2 activates the innate immune system.
Oxidative repair and formation is in fl ux, and clinical correlations often use several biomarker measurements to account for overall oxidative stress (Sung et al. 2013 ). Oxidative stress is further complicated by the amplifi cation caused by pro-
M. Pugia
197
infl ammatory cytokines released by immune cells and stimulated by bacterial lipo-polysaccharides (LPS) during infections. Therefore, many biomarkers, including those for immune cell activation, are accounted for in clinical assessments. Urinary biomarkers of oxidative damage has been an attractive topic of research due to the noninvasive sampling process involved (Il’yasova et al. 2012 ). Thus, the range and depth of biomarkers tested is broad and wide, and correlations vary greatly between the actual species measured. Of the biomarkers tested to date, 8-OHdG, ox LDL, and MDA are among the most clinically verifi ed ( See Table 11.3 ).
In obesity and metabolic syndrome, rapid weight loss correlates with reduced oxidative stress as measured by 8-OHdG, oxLDL, and MDA. Oxidative stress mea-sured by these three biomarkers signifi cantly increases (P < 0.05) with fast blood glucose and HbA1c in type 1 diabetics (Goodarzi et al. 2010 ; Holvoet et al. 1998 ). Statistically signifi cant higher values of ROS markers were observed in diabetic patients along with slight reduction in the synthesis of nitric oxide and a decrease in the antioxidant levels (Ramakrishna and Jailkhani 2007 ; Al-Rawi 2011 ). Of the oxidative stress marker measured, markers of DNA damage, 8-OHdG and MDA, appear to be the most clinically sensitive to diabetes. Dieting and improved antioxi-dant intakes signifi cantly improves the levels of these markers in diabetic subjects, and the improvements correlate with improved fasting plasma glucose value and HbA1c (Armstrong et al. 1996 ).
Alternative approaches to measure oxidative stress by NO, SOD, GSH, or REDOX acceptors poorly differentiate normal patients from diabetics with hyper-glycemia (Motawi et al. 2013 ). Small decreases in MPO (15 %) and phospholipase activities (Lp-PLA2) (4 %), after adjustment for age and sex, were observed in dia-betic patients (Tumova et al. 2013 ). The levels of plasma F(2)-isoprostanes, ox- LDL, and Lp-PLA2 as markers of oxidative damage were unchanged across the risk categories of metabolic syndrome (Seet et al. 2010 ). A small decrease of 12 % in oxidized low density lipoprotein (ox-LDL) levels was reported in diabetics with hyperglycemia. Additionally, ox-LDL levels increase in diabetic patients with coro-nary artery disease (CAD) but do not increase in diabetic patients without CAD, and these levels are impacted by sulfonylureas therapy (Motawi et al. 2013 ).
Hypoxia and uremic toxins increase oxidative stress during kidney and heart disease. Antioxidant therapies show signifi cant reduction in event risk but have not
Table 11.3 Markers of oxidative stress
Measurement Markers Key cellular pathways
DNA damage 8-OHdG NOS DNA ligase SOD NOX NADH/GSH Anti-oxidants (diet) Innate immune cells
Lipid oxidation
ox LDL, MDA Lipase Innate immune cells
11 Progress in New Markers for Diabetes Infl ammation
198
been shown to improved survival (Sung et al. 2013 ). Innate immune cell involve-ment shows a signifi cant increase in oxidative stress and event risk. However, the predictive value of ROS markers is weakly correlated (R ~ 0.4) with diabetes given the complexity of oxidative stress status of the body. Many factors, such as age, diet, and other diseases, contribute to oxidative stress levels. Therefore, biomarker use in a clinical setting is impractical for general diabetic patient care, as all factors must be accounted for (Motawi et al. 2013 ). Although recent trials on anti-oxidants treatments (like alpha-lipoic acid) in type 2 diabetes have not shown decreases in ROS or AGE levels (Porasuphatana et al. 2012 ), ROS markers remain a valuable research tool.
11.4 Impact of Obesity
Adipose tissue is now recognized as having an endocrine function that serves as a source of adipokines and as having an innate immune function that serves as a source of cytokine and chemokine signals (Preedy and Hunter 2011 ). Release of adipokines allows insulin independent glucose uptake and fatty acid oxidation in muscle and liver. Obesity impairs the secretion of adipokines and causes insu-linemia, hyperglycemina, lipidemia, and reduced appetite suppression ( See Table 11.1 ). Impaired adipokine secretion helps explain the impact of excess fat storage during glycemia where an impaired AMP-activated protein kinase (AMBK) pathway results in decreased glycolysis and reduced fatty acid oxidation. Obesity and insulin resistance increase the release of free fatty acids from the liver. Lipid loading causes the adipose tissue to become infl amed and hypoxic, and it induces cytokine and chemokine release and stimulation of innate immune cells. However, these new understandings of how adipokines relate to insulin resistance and the resultant hyperglycemia has had limited benefi ts for today’s diagnosis and treatment.
Lifestyle changes such as exercise and diet for weight loss and glycemic control have not stemmed the tide of the epidemic of obese adults with diabetes. Bariatric surgery has become a common treatment for weight loss and glycemic control in obese adults with diabetes (Maggard-Gibbons 2013 ), and studies have shown that at the 2-year follow up after bariatric surgery, adiponectin, leptin, insulin resistance, and body mass indexes (BMI) continue to show improvement (Trakhtenbroit et al. 2009 ; Ballantyne 2005 ). Glucose metabolism and insulin resistance are improved short term by caloric restriction, and decreased fat mass leads to long-term improve-ments in insulin resistance and returns adipokines secretion closer to normal levels (Cumb 2005; Sasaki et al. 2014 ). It is common for bariatric surgery patients to also have non- alcoholic fatty liver disease (NAPLD), and NAPLD progresses to non-alcoholic steatohepatitis (NASH) in morbidly obese subjects. Hepatic insulin resis-tance and steatosis precede the development of type 2 diabetes. Thus, bariatric surgery improves steatosis, fi brosis, lipidemia, and infl ammation markers in NASH and NAPLD patients (Sasaki et al. 2014 ).
M. Pugia
199
Adipokines have led to several new therapeutic approaches for treatment of obe-sity and diabetes. Adiponectin administration in rodents is insulin-sensitizing and decreases body weight, but this result is not applied to humans (Haluzik 2005 ). Thiazolidinedione (TZD) activation of peroxisome proliferator-activated receptors (PPARγ) increases adiponectin and transcription of important regulators in trans-port, synthesis, storage, and oxidation of fatty acids. Leptin therapy normalizes gonadotrophin secretion and menstrual function but is only used in rare human disorders that cause congenital leptin defi ciency and lipodystrophy. Leptin treat-ment causes reductions in hyperglycemia, hypertriglyceridemia, and hepatic steato-sis (Gorden 2003 )
Gherlin and peptide tyrosine tyrosine (aka peptide YY, PYY, pancreatic peptide, and YY3-36) are also important factors that impact obesity through appetite. Gherlin and PYY are released by cells in the ileum and colon in response to feeding (Soares et al. 2008 ; Knudsen et al. 2014 ). Inhibition of ghrelin impacts growth hormone secretagogue receptor (GHSR) and neuropeptide Y (NPY) pathways and reduce appetite, food intake, body weight, and adiposity. Thus, the gherlin pathway is an attractive therapeutic target. When PYY is injected and enters the central nervous system, it acts as an orexigenic and increases appetite. It also has functions related to digestion and absorption and is associated with both increased and decreased appetite. Obese people secrete less PYY than non-obese people, and studies have shown that it inhibits appetite in rodents and primates. Leptin treatments have also been shown to reduce appetite, but obese people develop a resistance to leptin, so it is not an effective treatment for obesity.
11.5 Adipokine Markers
While adipokine biomarkers are closely associated with obesity and diabetes, the predictive value of these markers has not made the grade for routine diagnostic use (See Table 11.1 ). Adiponectin, leptin, and gherlin have the strongest correlations and are useful for pathway research (See Chap. 1 and Table 11.4 ), and levels of these three markers improve after bariatric surgery (Ballantyne 2005 ; Gumbs et al. 2005 ).
Table 11.4 Markers of adipose stress
Measurement Markers Key cellular pathway
Impaired glycolysis and fatty acid oxidation
Adiponectin AMPK PPAR α
Appetite suppression Gherlin GHSR Neuropeptide Y (NPY) JAK STAT
Fat mass and energy balance (intake/exercise)
Leptin GHSR Neuropeptide Y (NPY) JAK STAT
11 Progress in New Markers for Diabetes Infl ammation
200
Leptin is directly correlated to patient fat mass and diet/exercise balance (Meiera and Gressner 2004 ). In addition, leptin and adiponectin correlate with hypertrophic adipose tissue in obesity and reduced glycolysis, reduced fatty acid oxidation, insu-lin resistance, and type 2 diabetes. An increase in the ratio of adiponectin to leptin improves the correlations (Satoh 2004 ). Ghrelin levels also generally correlate with obesity. Leptin and ghrelin exert important roles on body weight regulation, eating behavior, and reproduction by acting on the central nervous system and target repro-ductive organs. The relevance of other adipokines has not been resolved or strongly correlated and are still of question (See Chap. 1 ).
11.6 Fatty Acid Markers
Obesity increases adipose release of free fatty acids (FFA) and triglycerides (TG), and adipose tissues can produce FFA and reduce TG by the acylation stimulating protein (ASP). Interaction of the ASP precursor (compliment C3), factor B, and adipsin produce ASP. Elevated FFA concentrations are predictive of pre-diabetic progression to diabetes but are not diagnostic of diabetes, and a decrease in fasting FFA adversely impacts glucose-induced insulin release by pancreatic ß-cells. FFA acid uptake and metabolism are required for normal islet function (Lam et al. 2002 ). Insulin increases the production of ASP. Fatty acid oxidation, PPAR gamma activa-tion, and mitochondrial activity in the muscle and liver are also upregulated by insulin (Turner et al. 2007 ). Fasting plasma ASP concentration correlates with fast-ing serum triglyceride during the oral fat tolerance test (OFTT) and negatively cor-relates with the glucose disposal rate and high-density lipoprotein (HDL) concentrations (Koistinen 2001 ). However, there is no correlation between fasting plasma ASP concentration and TG in lean healthy men, and fasting plasma ASP concentrations are only mildly elevated in obese and diabetic subjects (Koistinen 2001 ; Peake et al. 2005 ). Given these correlations, fatty acid stress is currently best measured by typical and proven diagnostic markers of lipidemia (See Table 11.5 ). ASP is highly correlated to ketosis-prone diabetes (Liu et al. 2014 ). Other adipose markers such as lipase, FABP, and adipose factor are involved in fatty acid stress during obesity but have poor diagnostic predictive value for diabetic onset (See Chap. 1 ).
Nuclear receptors such as sterol regulatory binding protein (SREBP), farnesiod X receptor (FXR), and liver X receptor (LXR) are pharmacological targets for improving lipidemia, bile acid production, and glucose metabolism in type 2 diabe-
Table 11.5 Markers of fatty acid stress
Category Markers Key cellular pathway
Impair fatty acid tolerance Fatty acids (FFA) ASP Triglycerides (TG) DCAT Lipids (HDL/LDL) SREBP, FXR and LXR
M. Pugia
201
tes (Ding 2014 ). Synthetic derivatives of triiodothyronine, retinoic acids, fatty acids, eicosanoids, oxysterols, bile acids, steroids, xenobiotics, mineralocorticoids, glucocorticoids, androgens, estrogens, progesterone, retinoic acids, synthetic estro-gens, riccardin, and podocarpic acid amides have all been identifi ed as potent ago-nists of SREBP, FXR, and LXR (Lui 2005 ; Makishima 2005 ). Acyl CoA diacylglycerol transferase (DGAT) is an important enzyme in TG synthesis and another important pharmacological target (Subauste and Burant 2003 ). Treatment with n-3 fatty acids as inhibitors of DGAT decreases the availability of fatty acids for triglyceride synthesis (Subauste and Burant 2003 ; DeVita and Pinto 2013 ). Statins are inhibitors of 3-hydroxy-3-methylglutaryl CoA reductase (HMG) and lower cholesterol. Reduced cholesterol directly reduces DGAT expression by increasing the transcription factor SREBP and upregulates fatty acid oxidizing enzymes and inhibitors (Chen and Farese 2005 ).
11.7 Infl ammation Markers in Diabetes
Adipocytes produce tumor necrosis factor alpha (TNFα), IL-1β, and IL-6 during obesity and are shown to cause insulin resistance (Kern et al. 2001 ; Lorenzo et al. 2008 ; Cawthorn and Sethi 2008 ; Coppack 2001 ). Adipose tissue also contains a signifi cant amount of stromovascular fraction that is made up of preadipocytes, endothelial cells, smooth muscle cells, fi broblasts, leukocytes, and macrophages which can produce substantially more TNFα than adipocytes (Cawthorn and Sethi 2008 ). Production of IL-1 and IL-6 is regulated differently from TNFα, and interac-tions between macrophages and adipocytes control adipocyte differentiation and cytokine expression (Xie et al. 2010 ; Cawthorn and Sethi 2008 ). TNFα induces cytokine secretion by infl amed cells and is cytotoxic. IL-6 promotes differentiation of B cells and stimulates antibody secretion by plasma cells. IL-1 stimulates T cells, attracts macrophages and neutrophils, and activates NK cells (Gao et al. 2014 ). Cell signaling by TNFα, IL-1β, and IL-6 decreases GLUT4 in adipocytes. Both IL-6 and TNF- α inhibit insulin action through the Akt-PIK3 pathway. They also both initiate a host of other cytokines (e.g. IL-8, IL-18β) and chemokines (e.g. MIC-1, MIP-2, MCP-1, CINC-1) to recruit macrophages and other innate immune cells to infl amed tissue. Chemokines direct chemotaxis to distressed cells and cause neutrophils, macrophages, T-lymphocytes, eosinophils, natural killer cells, and other innate immune cells to build-up in the adipose. These pathways can be characterized by several markers ( See Table 11.6 ).
During insulin resistance, cytokines and chemokine secretion is elevated in adi-pose tissue and muscle (Beavers and Nicklas 2011 ; Cawthorn and Sethi 2008 ). Adipose tissue produces a fi ve-fold increase in TNFα or IL-1β and a 40-fold increase in IL-6 secretion which correlates with obesity measured by BMI (Kern et al. 2001 ; Zhao et al. 2014 ). Plasma values of IL-8, IL-18, and Monocyte chemotactic protein (MCP-1) are elevate to a lesser extent during insulin resistance, and plasma levels of IL-1β, TNFα, and IL-6 also correlate with insulin resistance and adipose dys-
11 Progress in New Markers for Diabetes Infl ammation
202
function (Beavers and Nicklas 2011 ; Kern et al. 2001 ). However, clinical utility is limited, as values of IL-1β, TNFα, and IL-6 show high intra and inter-individual variability with overlap between healthy and diabetic individuals (Kern et al. 2001 ; Breslin et al. 2012 ; Beavers and Nicklas 2011 ). Nonspecifi city is also an issue, as cytokines are also markers of endothelial dysfunction and vascular infl ammation (Beavers and Nicklas 2011 ).
Tumor necrosis factor receptor (TNFR) is released as a soluble form (sTNFR) into plasma during diabetic infl ammation (Cawthorn and Sethi 2008 ), and in animal studies, neutralization of TNFα by sTNFR leads to a signifi cant improvement in insulin resistance (Hotamisligil et al. 1993 ). However, in both obese and nonobese adults with damaging lipidemia, plasma sTNFR levels increase and are not more correlated to obese adipose than TNFα (Cawthorn and Sethi 2008 ; Good et al. 2006 ). Plasma sTNFR has been correlated to progression of chronic kidney disease (CKD) (Krolewski et al. 2012 ). This chronic response likely refl ects prolong TNFα activation and not acute innate immune activity ( See Table 11.6 ). In contrast, lipo-calin (LCN-2), aka neutrophil gelatinase-associated lipocalin (NGAL), is secreted by neutrophils during renal ischemia reperfusion injury and is a marker of acute kidney injury (AKI) (Soni et al. 2009 ). However NGAL is only weakly associated with obesity and diabetes through levels of fasting triglycerides and LPS-binding protein and FFA intake (Moreno-Navarrete et al. 2010 )
C-Reactive protein (CRP) synthesis is upregulated by cytokines, predominantly IL-6, and is a precise measure of systemic mild infl ammation that returns to base-line 12–24 months after weight loss from bariatric surgery (Gebhart et al. 2014 ). Mild infl ammation as measured by CRP is consistently shown to be an independent predictor of metabolic syndrome, obesity, and diabetes (Seet et al. 2010 ) ( See Table 11.6 ). CRP activates the complement pathway, binds phosphocholine, and promotes clearance of apoptotic cells (Szalai 2004 ). CRP levels are signifi cantly increased in obese patients during hyperglycemic and diabetic ketoacidosis crises as compared to obese controls (Stentz et al. 2004 ), and these CRP values return to normal after insulin treatment and resolution (Stentz et al. 2004 ; Peuchant 1997 ). Increased fat mass correlates with CRP (Tsuriya et al. 2011 ). However, these CRP changes are small in magnitude and cannot predict type 2 diabetes or the prediabetic state but are clearly indicative of prolonged chronic infl ammation.
Mild elevations of peripheral white blood cell (WBC), neutrophil, granulocyte, lymphocyte, and monocyte counts are associated with the risk of diabetes (Gkrania-
Table 11.6 Infl ammatory markers adipose stress
Category Markers Key cellular pathway
Cytotoxic adipose response in obesity TNF-α, IL-6, and IL-1
NFκB activation Akt-PIK3 inhibition
Prolonged infl ammation exposure due to obesity and diabetes
CRP TNFRs
TNF-α IL-1 IL-6
Macrophage response YKL-40, IL-6 , MCP-1
M2 pro-infl ammatory polarization
M. Pugia
203
Klotsas et al. 2010 ; Vuong et al. 2014 ). However, these innate immune cell levels are not able to differentiate progression to diabetes. In diseased adipose tissue, mac-rophages are polarized into pro-infl ammatory M1 forms by IL-10, CD8 positive T cells, interferon (INF-γ), and immunoglobulins from B cells. Alternatively, macro-phages are polarized in healthy adipose tissue into anti-infl ammatory M2 forms by regulatory T cells (Tregs), eosinophils, and IL-4. As polarization occurs in tissue, few blood based markers for macrophage responsivity exist today. However, gen-eral macrophage markers such as chitinase-3-like protein 1 (YKL-40), which is secreted by macrophages in infl amed tissues and causes activation of the Akt pro- survival signaling (anti-apoptotic pathway), can be used. YKL-40 levels are ele-vated in patients with vascular disease, hypertension, diabetes, and obesity (Ridker 2014 ). However, YKL-40 is also elevated in neutrophils. MCP-1 is another example of a marker secreted primarily by macrophages, endothelial cells, and adipocytes (Patel et al. 2013 ).
11.8 Endothelial System Markers
Vascular dysfunction occurs in diabetic and obesity as the endothelium loses it selectivity as a permeable barrier (Versari et al. 2009 ). The eNOS, RAAS and kallikrein- kinin response systems all fail to maintain normal vascular homeostasis with hyperglycemia, reactive oxidative species (ROS), free fatty acid (FFA) stress and pro- infl ammatory signaling (Roberts and Porter 2013 ; Brady et al. 2012 ). Blood pressure increases in diabetic obesity. Circulating pro-infl ammatory cytokine signaling (TNF-α, IL-1β) activates MAPK p38 and causes releases of vascular adhesion molecules (ICAM, VCAM, E-selectin) by endothelial cell (Szmitko et al. 2003b ; Collins et al. 1995 ) ( See Chap. 1 ). Vascular adhesion molecules attract innate immune cells to endothelium. Tissue remodeling with hypertrophy increases in the heart, kidney, liver, peripheral vascular and artery systems (Versari et al. 2009 ).
When innate immune cells pass the endothelium barrier and infi ltrate the sur-rounding tissue, they also initiate repair and wound healing mechanisms (Szmitko et al. 2003a ). Serine proteases that are released from innate immune cells activate protease-activated receptors (PARs). This group of receptors activates platelet aggregation, vascular smooth muscle proliferation, promotes thrombosis, prolongs infl ammation, stimulates endothelial cell proliferation, increases collagen synthesis, stimulates connective tissue synthesis, and activates platelet-derived growth factor production (Barry et al. 2006 ; Szmitko et al. 2003b ). The activation of PAR increases endothelial cell release of plasmin, thrombin, and trypsin and allows urokinase (uPA) activation of tissue remodeling ( See Fig. 1.2 of Chap. 1 ) . Plasminogen activator inhibitor 1 (PAI-1) is a modulator of fi brinolysis that slows cell prolifera-tion and restores normal cell differentiation (Ghosh and Vaughan 2012 ). PAI-1 cor-relates to fi brosis (decreasing elasticity) and membrane thickening throughout the body.
11 Progress in New Markers for Diabetes Infl ammation
204
There are several markers of endothelial dysfunction in diabetes and obesity that have value for patient assessments ( See Table 11.7 ). Plasminogen activator inhibi-tor- 1 (PAI-1) levels signifi cantly increase during hyperglycemic and diabetic keto-acidosis in obese patients as compared to obese controls (Stentz et al. 2004 ; Lopes-Virella et al. 2008 ). PAI-1 activity assays correlate with metabolic syndrome better than PAI-1 antigen assays do (Al-Hamodi et al. 2011 ), and tissue plasmino-gen activator (tPA) activity is not elevated in these patients. Uristatin (Uri) and Bikunin (Bik) respond to tissue regeneration by inhibiting PAR and uPA (Pugia et al. 2007 ), and Bik and Uri are urinary markers that correlate with glomerular nephritis and suppress cellular proliferation while activating Akt-PIK3.
There are several additional markers of endothelial dysfunction that are involved in diabetic conditions but have not yet proven clinical value. Endogenous tissue inhibitor of metalloproteinases (TIMP), angiotensin I converting enzyme, and angio-tensinogen correlate with atherosclerosis but do not correlate directly with diabetes (Szmitko et al. 2003a , b ; Goldfi ne et al. 2011 ). Endothelin-1 (ET-1) is a potent vaso-constrictor that is released by the endothelium in obese subjects. However, circulat-ing levels of ET-1 refl ect overall clearance and not vascular production (Weil et al. 2011 ). Vascular adhesion molecules (ICAM, VCAM, E-selectin) are increased in poorly controlled diabetes (Motawi et al. 2013 ; Ridker 2014 ; Lopes-Virella et al. 2008 ). However, vascular adhesion molecules and fi brinogen are less predictive of hypertension, diabetes, and obesity than infl ammation markers (Ridker 2014 ; Motawi et al. 2013 ). Matrix metalloproteinase (MMP) levels decrease in patients with diabe-tes and hyperglycaemia and correlates with reduced C-peptide (Lewandowski et al. 2011 ; Shiau et al. 2006 ). However, MMP levels are increased in patients with athero-sclerosis (Szmitko et al. 2003a ; Wang 2003). Both sCD40 and adrenomedullin levels are increased in patients with atherosclerosis and diabetes but play an unspecifi c role in diabetes (Szmitko et al. 2003a ; Wong et al. 2014 ; Gottsäter et al. 2013 ).
11.9 Markers of Dysfunctional Cell Growth
Obesity mediated diabetes is characterized by infl amed, hypertorphic, and dysfunc-tional tissues. This dysfunctional tissue is observed in the adipose, liver, pancreas, vascular system, and kidney (Halberg et al. 2009 ; Suzuki et al. 2014 ; Li et al. 2014 ; Rosenberger et al. 2008 ; Haligur et al. 2012 ), and these tissues all share the charac-teristic of being under oxygenated. Hypoxia typically induces an angiogenic
Table 11.7 Markers of endothelial stress in diabetes
Category Markers Key cellular pathway
Vascular fi brinolysis PAI-1 PAR 1-4 VCAM-1/ICAM-1 uPA/tPA
Anti-proliferative factor
Bik/Uri PAR inhibition uPA inhibition Akt-PIK3 activation
M. Pugia
205
response with endothelial cell migration and organization into capillary-like struc-tures. However, dysfunctional tissue growth occurs in obesity and diabetes as char-acterized by fi brosis and innate immune cell infi ltration. For example, the control of pre-adipocytes, specialized fi broblast-like progenitor cells that normally differenti-ate into mature adipocytes under control of growth factors, is over-ridden in diabetic and obese patients (Sorisky et al. 2000 ).
Hypoxic tissues release TGF and IL-6, which chemotactically attract macro-phages and promote fi broblast, endothelial, and progenitor cell growth (Bourlier et al. 2012 ; Halberg et al. 2009 ). Matrix metalloproteinase (MMP) is responsible for the proteolytic activation of TGF, and TGF causes endothelial cell expression of fi broblast growth factor (FGF) and vascular endothelial growth factor (VEGF) which block apoptosis via MAPK p38 and allow cell growth during tissue remodel-ing (Ferrari et al. 2006 ). Additionally, endothelial hyaluronan receptor 1 (LYVE-1) and hypoxia-inducible factor (HIF) are upregulated in hypoxic tissue and contribute to dysfunctional tissue growth and fi brosis (Sun et al. 2013 ). Leptin defi ciency con-tributes to transcription of hypoxia related genes (Yingzhong et al. 2006 ). Other growth factors such as growth hormone (GH), insulin-like growth factors (IGF), epidermal growth factor (EGF), placental growth factor (PDF), hepatocyte growth factor (HGF), nerve growth factor (NGF), platelet-derived growth factor (PDGF), and growth differentiation factors (GDF) are also important players in diabetes, and they work together to control differentiation of various cell types and connective tissue (Chiarella et al. 2004 ; Doronzo et al. 2006 ; Lu et al. 1999 ; Cook et al. 2002 ; Seedorf 1995 , Cirri et al. 2005 ). Cell signaling to generate normal tissue growth versus dysfunctional tissue growth under stress conditions is a complex process (Seedorf 1995 ).
New markers are needed to identify and monitor the repair processes during diabetic hypoxia ( See Table 11.8 ). Hypoxic conditions increase HIF and TGF values which induce changes in the expression of >1000 genes (Jiang et al. 2011 ; Ranganathan et al. 2007 ). Serum TGF-β1 levels increase in diabetic nephropathy (El Mesallamy 2012 ). Plasma TGF-β2 decreases, and urinary TGF-β1 increases in type-1 Diabetes Mellitus (Azar et al. 1999 ; Flores et al. 2004 ). TGF-β1 directly cor-relates to insulin administration and is an effective measure of therapeutic response to pro-fi brotic stimulus (Flores et al. 2004 ; Pscherer et al. 2013 ). Reduction of hypoxia signaling also improves insulin sensitivity and decreases adiposity (Jiang et al. 2011 ). Plasma HIF-1α levels are closely correlated and have predictive value to coronary calcifi cation in diabetes (Li et al. 2014 ). Furthermore, VEGF levels signifi cantly increase in diabetes and atherosclerosis, whereas VEGF receptor (sol-uble Flt-1) does not signifi cantly increase in diabetic patients but has value as a marker of angiogenesis (Blann et al. 2002 ). Many other growth factors such as FGF-19, GDF-15, HGF, NGF, and IGFBP are insulin dependent and undergo changes in diabetes (Vila et al. 2011 ; Goldfi ne et al. 2011 ; Cirri et al. 2005 ; Matsumoto et al. 2007 ; Gerhard et al. 2013 ; Azar et al. 1999 ). Since these growth factors correlate with insulin, measuring insulin sensitivity is more diagnostically addressable. Urinary EGF levels are insulin independent and are lower in diabetic patients, but these changes are thought to be related to declining kidney function and age (Lev-Ran et al. 1990 ; Kawaguchi et al. 1993 ).
11 Progress in New Markers for Diabetes Infl ammation
206
Insulin and growth hormone have a strong impact on cell growth and differentia-tion (Chiarella et al. 2004 ). Tissue and cell growth and survival are activated through the Akt-PI3K pathway (Siddle 2011 ; Chiarella et al. 2004 ; Sorisky et al. 2000 ). Insulin activation of Akt-PI3K increases expression of HIP-1, VEGF, GDF, PPARα, IGF, and IGFBP (Doronzo et al. 2006 ; Lu et al. 1999 ; Cook et al. 2002 ). Insulin inhibits PDGF (Cirri et al. 2005 ). FGF and EGF are insulin independent growth factors that activate MAPK ERK (Kir et al. 2011 ). EGF also activates HGF (Scheving et al. 2002 ). However, MAPK ERK is also stimulated by insulin recep-tors and IGFR and causes cross talk with EGF (Seedorf 1995 ; Kir et al. 2011 ; Cross et al. 1997 ). The Akt-PI3K pathway is therefore of great regulatory interest for tis-sue growth in diabetes.
As shown in this volume, inhibition of insulin degradation through adiponectin receptor-1 C-terminal fragments (AdipoR1 CTF) is providing new understanding of tissue insulin resistance and intracellular insulin control ( See Table 11.8 ). Oxygen and glucose deprivation increase expression of TNFα converting enzyme (TACE) (Hurtado et al. 2001 ), and TACE is key in AdipoR CTF-Ig formation. TACE is also the mechanism of release for soluble forms of TNF, TGF, and IL-6 from their membrane- bound precursors in hypoxia and is essential in all steps of immune cell recruitment and infl ammatory response (Black 2002 ; Garton et al. 2006 ; Ichikawa et al. 2004 ). Overactive mitochondria activity in the muscle could increase AdipoR CTF as a means for desensitizing insulin action that is directed by immunoglobulin attachment. Patients undergoing anti-TNF therapies have lower plasma Ig-CTF levels that are similar to levels observed in long term diabetes patients ( See Chaps. 5 and 6 ).
11.10 Innate Immunity Markers
Infl ammation, vascular dysfunction, and hypoxia attract innate immune cells such as neutrophils, macrophages, natural killer cells, and eosinophils into the diabetic tissue (Lee and Lee 2014 ; Olefsky and Glass 2010 ; Johnson et al. 2012 ).
Table 11.8 Markers of dysfunctional cell growth in diabetes
Category Markers Key cellular pathway
Pro-fi brotic stimulus
TGF- β SMAD MAPK p38 Akt-PI3K
Hypoxia HIF NOS VEGF EPO
Insulin desensitizing Ig-CTF TACE IDE
Angiogenesis VEGF PKC MAPK ERK Paxillin Akt-PI3K
M. Pugia
207
Immunoglobulin and complement proteins are also attracted. Dysfunctional tissue is cleaned-up through immune mediated apoptosis (Fink and Cookson 2005 ; Olefsky and Glass 2010 ) (See Chap. 1 , Fig. 1.3 ). Regulated cell death (apoptosis) causes cell loss in the adipose, liver, pancreas, and kidney during diabetes (Sorisky et al. 2000 ; Younossi et al. 2008 ; Chandra et al. 2001 ). Apoptosis is initiated after the innate immune cells infi ltrate the endothelium and release serine proteases, granzyme B, cathepsin G and proteinase 3 (PR 3) (Wiedow and Meyer-Hoffert 2005 ; Johnson et al. 2003 ). Pro-infl ammatory signaling (e.g. TNF-α, IL-1β) and activation of NFκB remove protection from granzyme B apoptosis ( See Chap. 1 ) and trigger apoptosis through caspase 8 activation (Cawthorn and Sethi 2008 ). Apoptosis prevents the maturation of pre-adipocytes into normal healthy tissue (Sorisky et al. 2000 ). Neutrophils, macrophages, T-lymphocytes, eosinophils, natu-ral killer cells and other innate immune cells also mediated damage by secreting TNF-α (Olefsky and Glass 2010 ). Immune cell activation of alternative complement pathway is through C3b deposition followed by factor B activation and ending with opsonized and lysed tissue cells (Carroll 2004a , b ; Mamane et al. 2009 ; Phieler et al. 2013 ; Schulze et al. 1993 )
T- Cell and B- cell lymphocytes and other adaptive immune cells are also clearly observed in diabetic tissues (Travers et al. 2015 ; Johnson et al. 2012 ; Lalor 2002 ; Descamps-Latscha 1993 ). Classical complement pathway deposits immune com-plexes (IgG & IgM) followed by factor B activation and ending with opsonized and lysis of tissue cells (Carroll 2004a , b , 4; Alsaad and Herzenberg 2007 ; Callard 1975 ). Apoptotic cell antigens are detected by IgM on B cell which are activated by T cells in the lymphoid compartment to produce IgG. This is followed by plasma and memory cell production by the spleen and bone marrow. Apoptotic antigens are then detected in dysfunction tissue by free IgG and activate the complement com-plex. The complex attracts neutrophils and macrophages to induce cell death and clean up, and the diabetic autoimmune response destroys the ability of the pancreas to produce insulin (Yoon and Jun 2005 ). Work in this volume showed CTF adsorp-tion causes insulin resistance. Antibodies attached to CTF are polyclonal and are directed at multiple antigens but could play a role in the adaptive response. Insulin sensitivity is required for B-cell and T-cell activation. Whether Ig-CTF protects cells from adaptive immune response or further activates the response, remains to be seen.
11.11 Conclusion
New markers of adipose, glycation, endothelial, and oxidative stress can be com-bined with markers of dysfunctional cell growth, infl ammation, and fatty acid stress to give a systematic view of the diabetic infl ammatory process. However, there is still a need to identify and monitor injury due to immune cell activation during dia-betes, and markers are needed to sort out the complex interaction of innate and adaptive immune cells within tissues. Anti- infl ammatory drugs to date have had limited effectiveness to reduce diabetic complications (Agrawal and Kant 2014 ).
11 Progress in New Markers for Diabetes Infl ammation
208
The goal of new markers should be to improve therapeutic outcomes and reduce the cost of diabetic complications. Therapies directed toward reducing the infl amma-tory response and protecting specifi c tissues such as the kidney, vascular systems, liver, pancreas, and eyes are needed.
References
Agrawal NR, Kant S (2014) Targeting infl ammation in diabetes: newer therapeutic options. World J Diab 5(5):697–710
Al-Hamodi Z et al (2011) Association of plasminogen activator inhibitor-1 and tissue plasminogen activator with type 2 diabetes and metabolic syndrome in Malaysian subjects. Cardiovasc Diabetol 10:23
Al-Rawi NH (2011) Oxidative stress, antioxidant status and lipid profi le in the saliva of type 2 diabetics. Diab Vasc Dis Res 8(1):22–28
Alsaad KO, Herzenberg AM (2007) Distinguishing diabetic nephropathy from other causes of glomerulosclerosis: an update. Clin Pathol 60(1):18–26
Armstrong AM, Chestnutt JE, Gormley MJ, Young IS (1996) The effect of dietary treatment on lipid peroxidation and antioxidant status in newly diagnosed noninsulin dependent diabetes. Free Radic Biol Med 21(5):719–726
Azar ST, Major SC, Safi eh-Garabedian B (1999) Altered plasma levels of nerve growth factor and transformation growth factor-β 2 in type-1 diabetes mellitus. Brain Behav Immun 13(4):361–366
Ballantyne GH, Gumbs A, Modlin IM (2005) Changes in insulin resistance following bariatric surgery and the adipoinsular axis: role of the adipocytokines, leptin, adiponectin and resistin. Obes Surg. 15(5):692–699
Barry GD, Le GT, Fairlie DP (2006) Agonists and antagonists of protease activated receptors (PARs). Curr Med Chem 13(3):243–262
Beavers KM, Nicklas BJ (2011) Effects of lifestyle interventions on infl ammatory markers in the metabolic syndrome. Front Biosci 3:168–177
Beck R et al (2011) The interrelationships of glycemic control measures: HbA1c, glycated albu-min, fructosamine, 1,5-anhydroglucitrol, and continuous glucose monitoring. Pediatr Diab 12(8):690–695
Black RA (2002) Tumor necrosis factor-alpha converting enzyme. Int J Biochem Cell Biol 34(1):1–5
Blann AD et al (2002) Vascular endothelial growth factor and its receptor, Flt-1, in the plasma of patients with coronary or peripheral atherosclerosis, or type II diabetes. Clin Sci (Lond) 102(2):187–194
Bourlier V et al (2012) TGFbeta family members are key mediators in the induction of myofi bro-blast phenotype of human adipose tissue progenitor cells by macrophages. PLoS One 7(2):e31274
Brady EM et al (2012) Investigating endothelial activation and oxidative stress in relation to gly-caemic control in a multiethnic population. Exp Diabetes Res 2012:e386041
Breslin WL et al (2012) Obese Mexican American children have elevated MCP-1, TNF-α, mono-cyte concentration, and dyslipidemia. Pediatrics 129(5):e1180–e1186
Callard P (1975) Immune complex type glomerulonephritis in cirrhosis of the liver. Am J Pathol 80(2):329–340
Carroll MC (2004a) A protective role for innate immunity in systemic lupus erythematosus. Nat Rev Immunol 4:825–831
Carroll MC (2004b) The complement system in regulation of adaptive immunity (review). Nat Immunol 3(10):981–986
M. Pugia
209
Cawthorn WP, Sethi JK (2008) TNF-alpha and adipocyte biology. FEBS Lett 582(1):117–131 Chandra J et al (2001) Role of apoptosis in pancreatic β cell death in diabetes. Diabetes 50(Suppl
1):S44–S47 Chen HC, Farese RV (2005) Inhibition of triglyceride synthesis as a treatment strategy for obesity
lessons from DGAT1-defi cient mice. Arterioscler Thromb Vasc Biol 25:482–486 Chiarella F, Giannini C, Mohn A (2004) Growth, growth factors and diabetes. Eur J Endocrinol
151:U109–U117 Cirri P et al (2005) Insulin inhibits platelet-derived growth factor-induced cell proliferation. Mol
Biol Cell 16(1):73–83 Collins T et al (1995) Transcriptional regulation of endothelial cell adhesion molecules: NF-kappa
B and cytokine-inducible enhancers. FASEB J 9(10):899–909 Cook SA et al (2002) Transcriptional effects of chronic Akt activation in the heart. J Biol Chem
227:22528–22533 Coppack SW (2001) Pro-infl ammatory cytokines and adipose tissue. Proc Nutr Soc
60(3):349–356 Cross DAE et al (1997) Insulin activates protein kinase B, inhibits glycogen synthase kinase-3 and
activates glycogen synthase by rapamycin-insensitive pathways in skeletal muscle and adipose tissue. FEBS Lett 406(1–2):211–215
Descamps-Latscha B (1993) The immune system in end-stage renal disease. Curr Opin Nephrol Hypertens 2(6):883–891
DeVita RJ, Pinto S (2013) Current status of the research and development of diacylglycerol O-acyltransferase 1 (DGAT1) inhibitors. J Med Chem 56(24):9820–9825
Ding L (2014) Coordinated actions of FXR and LXR in metabolism: from pathogenesis to phar-macological targets for type 2 diabetes (review). Int J Endocrinol 2014;Article ID 751859, 13 pgs
Doronzo G et al (2006) Insulin activates hypoxia-inducible factor-1 in human and rat vascular smooth muscle cells via phosphatidylinositol-3 kinase and mitogen-activated protein kinase pathways: impairment in insulin resistance owing to defects in insulin signaling. Diabetologia 49:1049–1063
El Mesallamy HO (2012) Clinical signifi cance of infl ammatory and fi brogenic cytokines in dia-betic nephropathy. Clin Biochem 45(9):646–650
Ferrari G et al (2006) VEGF, a pro-survival factor, acts in concert with TGF-β1 to induce endothe-lial cell apoptosis. Proc Natl Acad Sci U S A 103(46):17260–17265
Fink SL, Cookson BT (2005) Apoptosis, pyroptosis, and necrosis: mechanistic description of dead and dying eukaryotic cells. Infect Immun 73(4):1907–1916
Flores L et al (2004) Transforming growth factor β1 at clinical onset of type 1 diabetes mellitus. A pilot study. Diabet Med 21(8):818–822
Gao D et al (2014) Interleukin-1β mediates macrophage-induced impairment of insulin signaling in human primary adipocytes. Am J Physiol Endocrinol Metab 307:E289–E304
Garton KJ, Gough PJ, Raines EW (2006) Emerging roles for ectodomain shedding in the regula-tion of infl ammatory responses. J Leukoc Biol 79:1105–1116
Gebhart A et al (2014) Changes in high-sensitivity C-reactive protein levels after laparoscopic gastric stapling procedures versus laparoscopic gastric banding. Am Surg 80(10):1044–1048
Gerhard GS et al (2013) A role for fi broblast growth factor 19 and bile acids in diabetes remission after Roux-en-Y gastric bypass. Diabetes Care 36(7):1859–1864
Ghosh AK, Vaughan DE (2012) PAI-1 in tissue fi brosis (review). J Cell Physiol 227(2):493–507 Gkrania-Klotsas E et al (2010) Differential white blood cell count and type 2 diabetes: system-
atic review and m-analysis of cross-sectional and prospective studies. PLoS One 5(10):e13405
Goldfi ne AB et al (2011) Biomarkers in fasting serum to estimate glucose tolerance, insulin sensi-tivity, and insulin secretion. Clin Chem 57(2):326–337
Good M et al (2006) TNF and TNF receptor expression and insulin sensitivity in human omental and subcutaneous adipose tissue – infl uence of BMI and adipose distribution. Diab Vasc Dis Res 3(1):26–33
11 Progress in New Markers for Diabetes Infl ammation
210
Goodarzi MT et al (2010) Oxidative damage to DNA and lipids: correlation with protein glycation in patients with type 1 diabetes. J Clin Lab Anal 24:72–76
Gottsäter M et al (2013) Adrenomedullin is a marker of carotid plaques and intima-media thick-ness as well as brachial pulse pressure. J Hypertens 31(10):1959–1965
Gorden P, Gavrilovay O (2003) The clinical uses of leptin. Curr Opin Pharma. 3:655–659 Gumbs AA, Modlin IM, Ballantyne GH (2005) Changes in insulin resistance following bariatric
surgery: role of caloric restriction and weight loss. Obes Surg. 15(4):462–473 Halberg N et al (2009) Hypoxia-inducible factor 1alpha induces fi brosis and insulin resistance in
white adipose tissue. Mol Cell Biol 29(16):4467–4483 Haligur M, Topsakal S, Ozmen O (2012) Early degenerative effects of diabetes mellitus on pan-
creas, liver, and kidney in rats: an immunohistochemical study. Exp Diab Res 2012:1–10 Haluzik M (2005) Adiponectin and its potential in the treatment of obesity, diabetes and insulin
resistance. Curr Opin Investig Drugs 6(10):988–993 Holvoet P et al (1998) Oxidized LDL and malondialdehyde modifi ed LDL in patients with acute
coronary syndromes and stable coronary artery disease. Circulation 98:1487–1494 Hotamisligil GS, Shargill NS, Spiegelman BM (1993) Adipose expression of tumor necrosis
factor- alpha: direct role in obesity-linked insulin resistance. Science 259:87–91 Hurtado O et al (2001) Up-regulation of TNF-alpha convertase (TACE/ADAM-17) after oxygen-
glucose deprivation in the rat forebrain slices. Neuropharmacology 40(8):1094–1102 Hwang ES, Kim GH (2007) Biomarkers for oxidative stress status of DNA, lipids, and proteins
in vitro and in vivo cancer research. Toxicology 229(1–2):1–10 Ichikawa Y et al (2004) The role of ADAM protease in the tyrosine kinase-mediated trigger mech-
anism of ischemic preconditioning. Cardiovasc Res 62:167–175 Il’yasova D, Scarbrough P, Spasojevic I (2012) Urinary biomarkers of oxidative status. Clin Chim
Acta 413(19–20):1446–1453 Jiang C et al (2011) Disruption of hypoxia-inducible factor 1 in adipocytes improves insulin sen-
sitivity and decreases adiposity in high-fat diet-fed mice. Diabetes 60(10):2484–2495 Johnson H et al (2003) Cell death induced by granzyme C. Blood 101(8):3093–3101 Johnson AR, Milner JJ, Makowski L (2012) The infl ammation highway: metabolism accelerates
infl ammatory traffi c in obesity. Immunol Rev 249(1):218–238 Kawaguchi M et al (1993) Excretion of urinary epidermal growth factor in non-insulin dependent
diabetes mellitus. Life Sci 52(14):1181–1186 Kern PA et al (2001) Adipose tissue tumor necrosis factor and interleukin-6 expression in human
obesity and insulin resistance. Am J Physiol Endocrinol Metab 280(5):E745–E751 Kir S et al (2011) FGF19 as a postprandial, insulin-independent activator of hepatic protein and
glycogen synthesis. Science 331(6024):1621–1624 Knudsen SH, Karstoft K, Solomon TP (2014) Hyperglycemia abolishes meal-induced satiety by a
dysregulation of ghrelin and peptide YY3-36 in healthy overweight/obese humans. Am J Physiol Endocrinol Metab 306(2):E225–E231
Koistinen HA (2001) Plasma acylation stimulating protein concentration and subcutaneous adi-pose tissue C3 mRNA expression in nondiabetic and type 2 diabetic men. Arterioscler Thromb Vasc Biol 21:1034–1039
Krolewski AJ et al (2012) Circulating TNF receptors 1 and 2 predict stage 3 CKD in type 1 diabe-tes. J Am Soc Nephrol 23:516–524
Lalor PF (2002) Recruitment of lymphocytes to the human liver. Immunol Cell Biol 80:52–64 Lam TK et al (2002) Free fatty acid-induced hepatic insulin resistance: a potential role for protein
kinase C-delta. Am J Physiol Endocrinol Metab 283(4):E682–E691 Lam JK et al (2013) Effect of insulin on the soluble receptor for advanced glycation end products
(RAGE). Diabet Med 30(6):702–709 Lee BC, Lee J (2014) Cellular and molecular players in adipose tissue infl ammation in the devel-
opment of obesity-induced insulin resistance. Biochim Biophys Acta 1842:446–462 Lev-Ran A et al (1990) Excretion of epidermal growth factor (EGF) in diabetes. Clin Chim Acta
192(3):201–206 Lewandowski KC et al (2011) Matrix metalloproteinases in type 2 diabetes and non-diabetic con-
trols: effects of short-term and chronic hyperglycaemia. Arch Med Sci 7(2):294–303
M. Pugia
211
Li G et al (2014) The relationship between serum hypoxia-inducible factor 1α and coronary artery calcifi cation in asymptomatic type 2 diabetic patients. Cardiovasc Diab 13:52
Liu Y et al (2014) Acylation stimulating protein, complement C3 and lipid metabolism in ketosis- prone diabetic subjects. PLoS One 9(10):e109237
Lopes-Virella MF et al (2008) Risk factors related to infl ammation and endothelial dysfunction in the DCCT/EDIC cohort and their relationship with nephropathy and macrovascular complica-tions. Diabetes Care 31:2006–2012
Lorenzo M et al (2008) Insulin resistance induced by tumor necrosis factor-a in myocytes and brown adipose. J Anim Sci 86(14 Suppl):E94–E104
Lu M et al (1999) Insulin-induced vascular endothelial growth factor expression in retina. Invest Ophthalmol Vis Sci 40(13):3281–3286
Lui W (2005) Design, synthesis, and structure–activity relationship of podocarpic acid amides as liver X receptor agonists for potential treatment of atherosclerosis. Bioorg Med Chem Lett 15:4574–4578
Maggard-Gibbons MA et al (2013) Bariatric surgery for weight loss and glycemic control in non-morbidly obese adults with diabetes: a systematic review. JAMA. 309(21):2250–2261
Makishima M (2005) Nuclear receptors as targets for drug development: regulation of cholesterol and bile acid metabolism by nuclear receptors (review). J Pharmacol Sci 97:177–183
Mamane Y et al (2009) The C3a anaphylatoxin receptor is a key mediator of insulin resistance and functions by modulating adipose tissue macrophage infi ltration and activation. Diabetes 58:2006–2017
Matsumoto K et al (2007) Inverse correlation between activated protein C generation and carotid atherosclerosis in type 2 diabetic patients. Diabet Med 24(12):1322–1328
Mazzola N (2012) Review of current and emerging therapies in type 2 diabetes mellitus. Am J Manag Care 1:S17–S26
Meiera U, Gressner AM (2004) Endocrine regulation of energy metabolism: review of pathobio-chemical and clinical chemical aspects of leptin, ghrelin, adiponectin, and resistin. Clin Chem 50(9):1511–1525
Moreno-Navarrete JM et al (2010) Metabolic endotoxemia and saturated fat contribute to circulat-ing NGAL concentrations in subjects with insulin resistance. Int J Obes (Lond) 34(2):240–249
Motawi TMK et al (2013) Effect of glycemic control on soluble RAGE and oxidative stress in type 2 diabetic patients. BMC Endocr Disord 13:32
Olefsky JM, Glass CK (2010) Macrophages, infl ammation, and insulin resistance. Annu Rev Physiol 72(21):1–28
Patel PS, Buras ED, Balasubramanyam A (2013) The role of the immune system in obesity and insulin resistance (review articles). J Obes 2013:Article ID 616193, 9 pgs
Peake PW et al (2005) Response of the alternative complement pathway to an oral fat load in fi rst- degree relatives of subjects with type II diabetes. Int J Obes (Lond) 29(4):429–435
Peuchant E (1997) Short-term insulin therapy and normoglycemia. Effects on erythrocyte lipid peroxidation in NIDDM patients. Diabetes Care 20:202–207
Phieler J, Garcia-Martin R, Lambris JD, Chavakis T (2013) The role of the complement system in metabolic organs and metabolic diseases. Semin Immunol 25(1):47–53
Pitocco D (2013) Oxidative stress in diabetes: implications for vascular and other complications. Int J Mol Sci 14:21525–21550
Porasuphatana S et al (2012) Glycemic and oxidative status of patients with type 2 diabetes mel-litus following oral administration of alpha-lipoic acid: a randomized double-blinded placebo- controlled study. Asia Pac J Clin Nutr 21(1):12–21
Preedy VR, Hunter RJ (eds) (2011) Adipokines, 1st edn. CRC Press LCC, New York, NY. pp 1–458. ISBN 9781578086894
Pscherer S et al (2013) Anti-diabetic treatment regulates pro-fi brotic TGF-ß serum levels in type 2 diabetics. Diabetol Metab Syndr 5(1):1–6
Pugia MJ, Valdes R, Jortani SA (2007) Chapter on urinary trypsin inhibitors: structure, biological relevance and measurement. In: Makowski G (ed) Advances in clinical chemistry, vol 44. Academic Press. Elsevier, New York, NY. ISBN 10: 0-12-373704-4
11 Progress in New Markers for Diabetes Infl ammation
212
Qin X et al (2004) Glycation inactivation of the complement regulatory protein CD59: a possible role in the pathogenesis of the vascular complications of human diabetes. Diabetes 53(10):2653–2661
Ramakrishna V, Jailkhani R (2007) Evaluation of oxidative stress in insulin dependent diabetes mellitus (IDDM) patients. Diagn Pathol 2:22
Ranganathan P et al (2007) Expression profi ling of genes regulated by TGF-beta: differential regu-lation in normal and tumour cells. BMC Genomics 8:98
Ridker PM (2014) Plasma levels of the pro-infl ammatory chitin-binding glycoprotein YKL-40, variation in the chitinase 3-like 1 gene (CHI3L1), and incident cardiovascular events. J Am Heart Assoc 3(3):e000897
Roberts AC, Porter KE (2013) Cellular and molecular mechanisms of endothelial dysfunction in diabetes. Diab Vasc Dis Res 10(6):472–482
Rosenberger C et al (2008) Adaptation to hypoxia in the diabetic rat kidney. Kidney Int 73(1):34–42
Sasaki A et al (2014) Bariatric surgery and non-alcoholic Fatty liver disease: current and potential future treatments. Front Endocrinol (Lausanne). 5;164
Satoh N (2004) Leptin to adiponectin ratio as a potential atherogenic index in obese type 2 diabetic patients. Diabetes Care 27(10):2488–2490
Scheving LA et al (2002) Integral role of the EGF receptor in HGF-mediated hepatocyte prolifera-tion. Biochem Biophys Res Commun 290(1):197–203
Schulze M et al (1993) Glomerular C3c localization indicates ongoing immune deposit formation and complement activation in experimental glomerulonephritis. Am J Pathol 142(1):179–187
Seedorf K (1995) Intracellular signaling by growth factors. Metabolism 44:24–32 Seet RC et al (2010) Markers of oxidative damage are not elevated in otherwise healthy individuals
with the metabolic syndrome. Diabetes Care 33(5):1140–1142 Shiau MY et al (2006) Increased circulatory MMP-2 and MMP-9 levels and activities in patients
with type 1 diabetes mellitus. Mt Sinai J Med 73(7):1024–1028 Siddle K (2011) Signaling by insulin and IGF receptors: supporting acts and new players. J Mol
Endocrinol 47:R1–R10 Soares JB, Roncon-Albuquerque R, Leite-Moreira A (2008) Ghrelin and ghrelin receptor inhibi-
tors: agents in the treatment of obesity. Expert Opin Ther Targets 12(9):1177–1189 Soni SS et al (2009) Early diagnosis of acute kidney injury: the promise of novel biomarkers.
Blood Purif 28:165–174 Sorisky A, Magun R, Gagnon AM (2000) Adipose cell apoptosis: death in the energy depot. Int J
Obes 24(4):S3–S7 Stentz FB, Umpierrez GE, Cuervo R, Kitabchi AE (2004) Proinfl ammatory cytokines, markers of
cardiovascular risks, oxidative stress, and lipid peroxidation in patients with hyperglycemic crises. Diabetes 53(8):2079–2086
Subauste A, Burant CF (2003) DGAT: novel therapeutic target for obesity and type 2 diabetes mel-litus. Curr Drug Targets 3:263–270
Sun K et al (2013) Fibrosis and adipose tissue dysfunction. Cell Metab 18(4):470–477 Sung CC, Hsu YC, Chen CC, Lin YF, Wu CC (2013) Oxidative stress and nucleic acid oxidation
in patients with chronic kidney disease. Oxid Med Cell Longev 2013:Article ID 301982, 15 pgs Suzuki T et al (2014) Hypoxia and fatty liver. World J Gastroenterol 20(41):15087–15097 Szalai AJ (2004) C-reactive protein (CRP) and autoimmune disease: facts and conjectures. Clin
Dev Immunol 11(3–4):221–226 Szmitko PE et al (2003a) Biomarkers of vascular disease linking infl ammation to endothelial acti-
vation: part 2. Circulation 108:2041–2048 Szmitko PE et al (2003b) Biomarkers of vascular disease linking infl ammation to endothelial acti-
vation: part 1. Circulation 108:1917–1923 Trakhtenbroit MA et al (2009) Body weight, insulin resistance, and serum adipokine levels 2 years
after 2 types of baraitric surgery. Am J Med. 122(5):435–442 Travers RL et al (2015) The impact of adiposity on adipose tissue-resident lymphocyte activation
in humans. Int J Obes (Lond) 39(8):762–769
M. Pugia
213
Tsuriya D et al (2011) Signifi cant correlation between visceral adiposity and high-sensitivity C-reactive protein (hs-CRP) in Japanese subjects. Intern Med 50:2767–2773
Tumova E et al (2013) The impact of rapid weight loss on oxidative stress markers and the expres-sion of the metabolic syndrome in obese individuals. J Obes 2013:729515
Turner N et al (2007) Excess lipid availability increases mitochondrial fatty acid oxidative capacity in muscle: evidence against a role for reduced fatty acid oxidation in lipid-induced insulin resistance in rodents. Diabetes 56(8):2085–2092
Versari D et al (2009) Endothelial dysfunction as a target for prevention of cardiovascular disease. Diabetes Care 32(Suppl 2):S312–S321
Vila G et al (2011) The relationship between insulin resistance and the cardiovascular biomarker growth differentiation factor-15 in obese patients. Clin Chem 57(2):309–316
Vuong J et al (2014) Reference intervals of complete blood count constituents are highly correlated to waist circumference: should obese patients have their own “normal values?”. Am J Hematol 89(7):671–677
Weil BR et al (2011) Enhanced endothelin-1 system activity with overweight and obesity. Am J Physiol Heart Circ Physiol 301:H689–H695
Wiedow C, Meyer-Hoffert UM (2005) Neutrophil serine protease: potential key regulators of cell signaling during infl ammation. J Intern Med 257:319–328
Wong HK et al (2014) Adrenomedullin and diabetes. Rev World J Diabetes 5(3):364–371 Xie L et al (2010) Interactive changes between macrophages and adipocytes. Clin Vaccine
Immunol 17(4):651–659 Yingzhong Y, Droma Y, Rili G, Kubo K (2006) Regulation of body weight by leptin, with special
reference to hypoxia-induced regulation. Intern Med 45(16):941–946 Yoon JW, Jun HS (2005) Autoimmune destruction of pancreatic β cells. Am J Ther 12:580–591 Younossi ZM et al (2008) A novel diagnostic biomarker panel for obesity-related nonalcoholic
steatohepatitis (NASH). Obes Surg 18:1430–1437 Zhao G et al (2014) Possible role of interleukin-1b in type 2 diabetes onset and implications for
anti-infl ammatory therapy. PLoS Comput Biol 10(8):e1003798
11 Progress in New Markers for Diabetes Infl ammation
215© Springer International Publishing Switzerland 2015 M. Pugia (ed.), Infl ammatory Pathways in Diabetes: Biomarkers and Clinical Correlates, Progress in Infl ammation Research, DOI 10.1007/978-3-319-21927-1
A Acute kidney injury (AKI)
biomarker analysis , 147–148 in cardiopulmonary bypass surgery
patients , 144 patient population , 145–147 post-operative prediction , 152 pre-operative predictors , 151–154 pro-infl ammatory biomarkers , 144 risk assessment tools , 144 risk score , 144 sample collection , 147 statistical analysis , 148–151 uTi , 144, 145
Adipokine markers , 194, 199–200 Adiponectin receptor C-terminal fragment
(AdipoR/AipoR CTF) animal models , 97–98 antibodies , 65–66 bioassays , 95 biomarker measurements , 98 blood sample preparation , 80 brain morphologic observations , 87 cell and tissue staining procedure , 86 cell culture methods , 95, 99 cell culture treatment , 95–96, 99 cell lines , 95 cell proliferation counts , 97 cellular insulin , 97, 100–102 conjugates and standards , 66–67 correlation to infl ammation , 118, 119 diabetic collection , 116 disease progression , 98, 102,
104–106 ELISA , 96
blood levels , 74 calibrator reproducibility and stability , 72 cross reactivity , 72 immunoglobulin bound form
measurement , 67–68 immunoglublin chain
identifi cation , 71 interference testing , 73 patient recovery , 73
fat morphologic observations , 86 fatty-acid oxidation , 62, 63 FBG measurement , 112 gestational diabetes
collection , 114–115 Ig-CTF correlations , 119, 121 Ig-CTF distribution , 118, 120 predictive values, Ig-CTF , 120, 122
gluconeogenesis , 62, 63 glucose clamp , 113 G protein-coupled receptor , 63–64 HbA1c measurement , 112 IDE , 78–79 Ig-CTF and CTF measurements , 116 Ig-CTF R1 values, sample measurement ,
116–118 iMALDI approach , 68–69 immunoglobulin binding mechanism , 70–71 immunuo fl uorescent staining , 87–89 inhibition studies
ADAM17/TACE and IDE activity , 81–84 InnoZyme™ Insulysin-IDE
Immunocapture Activity Assay Kit , 80
InnoZyme™ TACE Activity Kit , 80 insulin
Index
216
Adiponectin receptor C-terminal fragment (AdipoR/AipoR CTF) (cont.)
receptor response , 88, 90 sensitivity/resistance , 112, 113 sensitizing agent , 62
intracellular calcium , 96–97, 99–100 in-vivo insulin , 101–103, 105–106 leptin resistance , 94 liver morphologic observations , 87 mass spectroscopy peptide enrichment
assays , 71–72 mechanism of action , 104, 107 MEROPS search , 78 muscle
glucose uptake , 62, 63 morphologic observations , 86
OGTT , 112 oligomeric isoforms , 62 pancreas morphologic observations , 87 patient reproducibility and stability , 73 peptide preparation , 64–65 plasma forms , 85 precision and accuracy testing , 73 pre-diabetic
collection , 115, 116 correlation , 121–123
protease identifi cation , 79–80 reference range testing , 114 SELDI-MALDI method , 69 SISCAPA method , 68 TACE , 79 T-cadherin , 63 tissue
and blood analysis , 81 forms and distribution , 85 sample preparation , 80
TNF-α , 62 TZD treatment , 62 western blot method , 67, 70, 96, 98–99 western blot techniques , 81 X-ray crystallography analysis , 62
Adipose stress markers , 201–203 Advanced glycation end products (AGEs) ,
195–196 vs. CD59 markers , 6 glycated hemoglobin , 5 human serum albumin , 5 molecular damages , 5 nontraditional markers , 5 receptors activation , 5–6
AGEs See Advanced glycation end products (AGEs)
Aprotinin , 172, 183–185
B β –secretase (BACE) , 78–79 Biacore testing , 65 Bikunin (Bik)
anti-infl ammatory cellular signaling aprotinin , 183–185 Ca 2+ in-fl ux , 179, 186 cellular signal response , 177–179 DPP-4 , 178–181 membrane disruption , 179, 180,
185–186 necrosis , 185 peptide analogs , 181–183 PI3K-Akt pathway , 181, 182,
186–187 proliferation , 186 proteinuria , 184 trypsin inhibition , 182, 184
urinary trypsin inhibitor , 172 Bioassays , 95
C Cardiomyocyte (C2C12) models
cell culture treatment , 99 intracellular calcium , 99–100 western blot analysis , 98–99
Cardiovascular disease (CVD) , 18, 38–39 adiponectin response , 168 atherosclerosis , 157 demographic characteristics , 161, 162, 164 diagnostic risk factor , 161, 163, 164 epidemiological evidence , 158 indicators , 157–158 PAR activation , 158–159 statistics methods , 161
CD59 glycoprotein animal experimental model , 41 glycated ( see Glycated CD59 (GCD59)) glycation-inactivation , 39–40, 42–43 MAC inhibition mechanism , 35–36
Chronic kidney disease (CKD) , 131, 137, 139 Complement system
diabetic complications cardiovascular disease , 38–39 nephropathy , 36–37 neuropathy , 38 retinopathy , 37
MAC alternative pathway , 31, 33 classical pathway , 31, 33 inhibitory protein ( see CD59
glycoprotein)
Index
217
MBL/lectin pathway , 31, 33 non-lytic effects , 34–35
therapeutic application , 44 C-reactive protein (CRP)
acute phase reactants , 158 pro-infl ammatory response , 166, 168 vascular stenosis, infl ammation markers ,
164–166 C-terminal fragment (CTF) See Adiponectin
receptor C-terminal fragment (AdipoR/AipoR CTF)
CVD See Cardiovascular disease (CVD)
D Dipeptidyl peptidase 4 (DPP-4) , 174, 178–181 Dysfunctional cell growth markers
AdipoR1 CTF , 206 Akt-PI3K pathway , 206 hypoxia , 204–206
E ELISA See Enzyme linked immunosorbent
assays (ELISA) Endothelial dysfunction
adipose angiotensin , 12 Bikunin , 13 chronic activation , 12 kinin formation , 11–12 nitric oxide , 11 TFPI action , 13
Endothelial stress markers , 203–204 Enzyme linked immunosorbent assays
(ELISA) , 96, 174 blood levels , 74 calibrator reproducibility and stability , 72 cross reactivity , 72 immunoglobulin bound form measurement ,
67–68 immunoglublin chain identifi cation , 71 interference testing , 73 patient recovery , 73
F Fasting blood glucose (FBG) , 112 Fatty acid markers , 200–201 Fibrosis
Bikunin action , 17 coagulation , 16 mechanism , 16, 17 PAR activation , 15–16
Free fatty acids (FFAs) , 9–10
G GCD59 See Glycated CD59 (GCD59) Gestational diabetes
collection , 114–115 Ig-CTF correlations , 119, 121 Ig-CTF distribution , 118, 120 predictive values, Ig-CTF , 120, 122
Glomerular fi ltration rate (GFR) , 129, 131, 132
Glycated CD59 (GCD59) ELISA , 46 HbA1c assay , 45 OGTT test , 47 plasma level , 46–47
H Hyperglycemia
complications with , 3 glycation stress ( see Advanced glycation
end products (AGEs)) insulin resistance , 3–4 oxidative stress , 6–7 stress markers
AGE , 195–196 glycation stress markers , 195 impaired insulin metabolism , 194–195 insulin resistance , 193–195 stress factor , 193, 194 therapeutic approaches , 196
I Infl ammation
endothelial dysfunction adipose angiotensin , 12 Bikunin , 13 chronic activation , 12 kinin formation , 11–12 nitric oxide , 11 TFPI action , 13
fi brosis Bikunin action , 17 coagulation , 16 mechanism , 16, 17 PAR activation , 15–16
innate immune apoptosis complement system , 15 cytokines , 14–15 lymphocytes , 14–15 mechanism , 13–14 serine proteases , 15 T-cells , 15
Index
218
Infl iximab , 114 Innate immune mediated apoptosis
complement system , 15 cytokines , 14–15 lymphocytes , 14–15 mechanism , 13–14 serine proteases , 15 T-cells , 15
Innate immunity markers , 206–207 Insulin-degrading enzyme (IDE) , 78–79 Insulin resistance , 3–4
M Membrane attack complex (MAC)
alternative pathway , 31, 33 classical pathway , 31, 33 inhibitory protein ( see CD59 glycoprotein) MBL/lectin pathway , 31, 33 non-lytic effects , 34–35
MEROPS search , 78 Metabolic syndrome (MetSyn) , 166, 167 Modifi ed Eagle Medium (MEM) , 95–96
N Nephropathy , 18–19, 36–37 Neuropathy , 20, 38
O Obesity , 194, 198–199
adipose signaling adiponectin , 8 ghrelin level , 8–9 hypoxia , 10 leptin signals , 7
free fatty acids , 9–10 infl ammatory response ( see Infl ammation) prevalence, U.S. , 3–4
Oral glucose tolerance test (OGTT) , 112 Oxidative stress markers , 197
biomarker measurements , 196–197 ROS , 196, 198
P Protease activate receptors (PAR) signalling ,
15–16
R Reactive oxidative species , 196–198 Reactive oxygen species (ROS) , 6–7
Retinopathy , 19, 37 ROS See Reactive oxidative species; Reactive
oxygen species (ROS)
T Thiazolidinediones (TZD) treatment , 62 TNF-α–converting enzyme (TACE) , 79
U Urinary trypsin inhibitor (uTi) , 144, 145
anti-infl ammatory cellular signaling Akt and PIK3 measurements , 175 apoptosis assays , 176 cell culturing procedure , 173–174 DPP-4 peptidase activity assay , 174 intra-cellular analysis , 176 proliferation , 181, 186 protease and inhibitor balance , 177,
178 serine protease activity , 174 trypsin-specifi c chromogenic substrate ,
176–177 western-blot analysis , 175
CRP acute phase reactants , 158 pro-infl ammatory response , 166, 168 vascular stenosis, infl ammation
markers , 164–166 CVD
adiponectin response , 168 atherosclerosis , 157 demographic characteristics , 161, 162,
164 diagnostic risk factor , 161, 163, 164 epidemiological evidence , 158 indicators , 157–158 PAR activation , 158–159 statistics methods , 161
Kunitz type inhibitor , 172 marker analysis , 160–161 metabolic syndrome (MetSyn) , 166, 167 patient assessment , 159–160 sample collection , 160–161 uristatin immunoassay
anti-infl ammatory response , 127–131, 139
clinical analysis , 133–134 cross-reactivity , 132 dipstick method , 132 follow-up care , 137–138 health screen collection , 134–135 hospital infection collection , 136
Index
219
hospital patient collection , 136 kidney model , 128–131 proteinuria/hematuria , 138–139 sample and data bank , 134–135 urinary tract infection collection , 136 urine samples , 137 viral infections , 138 WBC , 129–132 Western blot , 136–137
uristatin response , 168 Uristatin (Uri) immunoassay
anti-infl ammatory response , 127–131, 139
clinical analysis , 133–134 cross-reactivity , 132 dipstick method , 132 follow-up , 137–138 health screen collection , 134–135
hospital infection collection , 136 hospital patient collection , 136 kidney model , 128–131 proteinuria/hematuria , 138–139 sample and data bank , 134–135 urinary tract infection collection , 136 urine samples , 137 viral infections , 138 WBC ( see White blood cells (WBCs)) western blot , 136–137
uTi See Urinary trypsin inhibitor (uTi)
W White blood cells (WBCs)
anti-infl ammatory response , 129–130 cell death , 131–132 neutrophilic elastase , 129
Index
