Advances in Clinical Chemistry, Volume 49

ADVANCES IN CLINICAL CHEMISTRY VOLUME 49 This page intentionally left blank Advances in CLINICAL CHEMISTRY Edited b...

Author: Gregory Makowski

53 downloads 1179 Views 1MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

Report copyright / DMCA form

DOWNLOAD PDF

ADVANCES IN CLINICAL CHEMISTRY VOLUME 49

This page intentionally left blank

Advances in CLINICAL CHEMISTRY Edited by GREGORY S. MAKOWSKI Department of Laboratory Medicine University of Connecticut Health Center Farmington, CT, USA

VOLUME 49

AMSTERDAM • BOSTON • HEIDELBERG • LONDON NEW YORK • OXFORD • PARIS • SAN DIEGO SAN FRANCISCO • SINGAPORE • SYDNEY • TOKYO Academic Press is an imprint of Elsevier

Academic Press is an imprint of Elsevier 32 Jamestown Road, London NW1 7BY, UK 30 Corporate Drive, Suite 400, Burlington, MA 01803, USA 525 B Street, Suite 1900, San Diego, CA 92101-4495, USA Radarweg 29, PO Box 211, 1000 AE Amsterdam, The Netherlands This book is printed on acid-free paper. ⬁ Copyright ß 2009, Elsevier Inc. All rights reserved No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means electronic, mechanical, photocopying, recording or otherwise without the prior written permission of the publisher Permissions may be sought directly from Elsevier’s Science & Technology Rights Department in Oxford, UK: phone (+44) (0) 1865 843830; fax (+44) (0) 1865 853333; email: [email protected]. Alternatively you can submit your request online by visiting the Elsevier web site at http://www.elsevier.com/locate/permissions, and selecting Obtaining permission to use Elsevier material Notice No responsibility is assumed by the publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made Library of Congress Cataloging-in-Publication Data A catalog record for this book is available from the Library of Congress British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library ISBN: 978-0-12-374798-3 ISSN: 0065-2423 For information on all Academic Press publications visit our website at www.elsevierdirect.com Printed and bound in USA 09 10 11 12 10 9 8 7 6

5 4 3 2

1

CONTENTS CONTRIBUTORS

................................................................................

ix

PREFACE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

xi

High-Sensitivity Troponin: A New Tool for Pathophysiological Investigation and Clinical Practice ALDO CLERICO, ALBERTO GIANNONI, CONCETTA PRONTERA, AND STEFANIA GIOVANNINI 1. Abstract ... ................................................................................... 2. Background and Aim....................................................................... 3. Introduction: Troponin Framework Within Myocardial Cells and Release Kinetics After Myocardial Damage ............................................. 4. Impact of the New Definition of Myocardial Infarction on Laboratory Practice and Instrumentation: The Need for High-Sensitivity cTnI and cTnT Methods ......................................................................... 5. The Impact of High-Sensitivity cTnI and cTnT Methods on Clinical Practice ...... 6. High-Sensitivity cTnI and cTnT Methods: A Powerful Tool for Monitoring Physiological Renewal and Pathological Remodeling of the Myocardial Tissue? ... 7. Use of High-Sensitivity cTnI and cTnT Methods in a Multimarker Approach for Early Screening: An Increase in Diagnostic and Prognostic Efficiency? ...................................................................... 8. Conclusion ................................................................................... References. ...................................................................................

2 2 4

5 9 19

22 23 24

Biochemical Biomarkers of Oxidative Collagen Damage Y. HENROTIN, M. DEBERG, M. MATHY-HARTERT, AND G. DEBY-DUPONT 1. 2. 3. 4. 5. 6. 7. 8. 9.

Abstract ... ................................................................................... The Collagen Family: Structure and Biochemistry...................................... Chemical Nature and Reactivity of Oxygen ............................................. In Vivo Production of RNOS .............................................................. RNOS Biological Activity.................................................................. In Vivo Markers of Oxidant Stress .. ...................................................... Oxidant-Induced Changes in Collagens .................................................. Biomarkers of Collagen Oxidative Damage ............................................. Critical Comments and Concluding Remarks ... ........................................ References. ................................................................................... v

31 32 33 35 37 39 40 43 48 49

vi

CONTENTS

Biochemical Basis of Fabry Disease with Emphasis on Mitochondrial Function and Protein Trafficking A.M. DAS AND H.Y. NAIM 1. 2. 3. 4. 5. 6. 7.

Abstract....................................................................................... Introduction.................................................................................. Inheritance ................................................................................... Clinical Picture............................................................................... Diagnosis ..................................................................................... Treatment .................................................................................... Biochemical Basis............................................................................ References ....................................................................................

57 58 58 59 59 60 62 68

Urinary Biomarkers for the Detection of Renal Injury MITCHELL H. ROSNER 1. 2. 3. 4. 5. 6.

Abstract....................................................................................... Introduction.................................................................................. Biomarker Development for Kidney Diseases ........................................... Biomarkers for AKI......................................................................... Biomarker Development and Implementation ........................................... Summary ..................................................................................... References ....................................................................................

73 74 75 81 90 91 92

Biomarkers of Bone and Mineral Metabolism Following Bone Marrow Transplantation KI HYUN BAEK AND MOO IL KANG 1. 2. 3. 4. 5. 6. 7. 8. 9.

Abstract....................................................................................... Introduction.................................................................................. Clinical Features of BMT-Related Bone Loss ........................................... Changes in Bone-Turnover Markers After BMT ........................................ Calcium, Parathyroid Hormone (PTH), and Vitamin D.. .............................. Sex Hormones ............................................................................... RANKL and Osteoprotegerin ............................................................. Cytokines and Growth Factors ............................................................ Conclusion ................................................................................... References ....................................................................................

99 100 100 102 109 110 112 114 115 116

Factor V Leiden and Activated Protein C Resistance OLIVIER SEGERS AND ELISABETTA CASTOLDI 1. Abstract....................................................................................... 2. Introduction..................................................................................

121 122

CONTENTS 3. The Protein C System ...................................................................... 4. APC Resistance and FV Leiden........................................................... 5. Conclusions and Perspectives.............................................................. References. ...................................................................................

vii 123 130 143 144

Self-Assembled Tethered Bimolecular Lipid Membranes EVA-KATHRIN SINNER, SANDRA RITZ, RENATE NAUMANN, STEFAN SCHILLER, AND WOLFGANG KNOLL 1. 2. 3. 4. 5. 6.

Abstract ... ................................................................................... Introduction ................................................................................. Assembly of tBLMs from Telechelics and Reconstitution of Proteins................ The Peptide-Tethered Lipid Bilayer Membrane (peptBLM) ........................... Protein-Tethered Bilayer Lipid Membrane (protBLM) ................................ Conclusions .................................................................................. References. ...................................................................................

159 160 162 165 170 174 177

INDEX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . COLOR PLATE SECTION AT THE END OF THE BOOK

181

This page intentionally left blank

CONTRIBUTORS Numbers in parentheses indicate the pages on which the authors’ contributions begin.

KI HYUN BAEK (99), Department of Internal Medicine, The Catholic University of Korea, College of Medicine, 137-701 Seoul, Korea ELISABETTA CASTOLDI (121), Department of Biochemistry, Maastricht University, 6200 MD Maastricht, The Netherlands ALDO CLERICO (1), Scuola Superiore Sant’Anna, 56126 Pisa, Italy; and Laboratory of Cardiovascular Endocrinology and Cell Biology, G. Monasterio Foundation, CNR Regione Toscana, 56126 Pisa, Italy A.M. DAS (57), Department of Pediatrics, Hannover Medical School, Hannover, Germany M. DEBERG (31), Bone and Cartilage Research Unit, Institute of Pathology, University of Lie`ge, CHU-Sart-Tilman, 4000 Lie`ge, Belgium G. DEBY-DUPONT (31), Center for Oxygen Research and Development, Institute of Chemistry, University of Lie`ge, 4000 Lie`ge, Belgium ALBERTO GIANNONI (1), Scuola Superiore Sant’Anna, 56126 Pisa, Italy; and Laboratory of Cardiovascular Endocrinology and Cell Biology, G. Monasterio Foundation, CNR Regione Toscana, 56126 Pisa, Italy STEFANIA GIOVANNINI (1), Laboratory of Cardiovascular Endocrinology and Cell Biology, G. Monasterio Foundation, CNR Regione Toscana, 56126 Pisa, Italy Y. HENROTIN (31), Center for Oxygen Research and Development, Institute of Chemistry, University of Lie`ge, 4000 Lie`ge, Belgium; and Bone and Cartilage Research Unit, Institute of Pathology, University of Lie`ge, CHU-Sart-Tilman, 4000 Lie`ge, Belgium

ix

x

CONTRIBUTORS

MOO IL KANG (99), Department of Internal Medicine, The Catholic University of Korea, College of Medicine, 137-701 Seoul, Korea WOLFGANG KNOLL (159), Austrian Institute of Technology, Vienna, Austria; and Institute of Materials Research and Engineering, Singapore M. MATHY-HARTERT (31), Bone and Cartilage Research Unit, Institute of Pathology, University of Lie`ge, CHU-Sart-Tilman, 4000 Lie`ge, Belgium H.Y. NAIM (57), Department of Physiological Chemistry, University of Veterinary Medicine, Hannover, Germany RENATE NAUMANN (159), Austrian Institute of Technology, Vienna, Austria; and Max Planck Institute for Polymer Research, Mainz, Germany CONCETTA PRONTERA (1), Laboratory of Cardiovascular Endocrinology and Cell Biology, G. Monasterio Foundation, CNR Regione Toscana, 56126 Pisa, Italy SANDRA RITZ (159), Max Planck Institute for Polymer Research, Mainz, Germany MITCHELL H. ROSNER (73), Division of Nephrology, University of Virginia Health System, Box 800133, Charlottesville, Virginia 22908 STEFAN SCHILLER (159), Freiburg Institute for Advanced Studies, Freiburg, Germany OLIVIER SEGERS (121), Department of Biochemistry, Maastricht University, 6200 MD Maastricht, The Netherlands EVA-KATHRIN SINNER (159), Institute of Materials Research and Engineering, Singapore; and Max Planck Institute for Polymer Research, Mainz, Germany

PREFACE I am pleased to present volume forty-nine of Advances in Clinical Chemistry series. In this third volume for 2009, the lead chapter investigates the analytical challenges of developing highly sensitive cardiac troponin immunoassays and their role in advancing clinical diagnosis. Pathophysiologic mechanisms of troponin release are discussed as well as their potential impact on monitoring the heart in a number of chronic disease processes. The next chapter explores eVect of oxidative stress on collagen, a key structural and functional protein within the connective tissue family. The biochemistry of Fabry disease is discussed in the following chapter with particular emphasis on the impact of compromised energy metabolism and abnormal lipid composition. The role of urinary markers in renal disease is presented next. This chapter is noteworthy given the aging population worldwide and the significant cost of chronic kidney disease in healthcare. Bone cell loss and osteoporosis following bone marrow transplantation is a serious endocrine disorder that is elucidated in the next chapter. The state of the art and new developments in factor V Leiden and activated protein C resistance testing are also explored with emphasis on the fundamental biochemistry of this coagulopathic condition. Volume forty-nine is concluded with a revealing manuscript on the biophysical and interfacial studies of lipid bilayers and the importance of lipid–protein interactions within this unique system. I extend my appreciation to each contributor of volume forty-nine and thank colleagues who participated in the peer review process. I would also like to thank my Elsevier editorial liaison, Gayathri Venkatasamy. I hope the third and final volume of 2009 will be enjoyed. I warmly invite comments and suggestions for future review articles for the Advances in Clinical Chemistry readership. In keeping with the tradition of the series, I would like to dedicate volume forty-nine to my grandmother Stephanie and my grandfather Stephen. GREGORY S. MAKOWSKI

xi

This page intentionally left blank

ADVANCES IN CLINICAL CHEMISTRY, VOL. 49

HIGH-SENSITIVITY TROPONIN: A NEW TOOL FOR PATHOPHYSIOLOGICAL INVESTIGATION AND CLINICAL PRACTICE Aldo Clerico,*,†,1 Alberto Giannoni,*,† Concetta Prontera,* and Stefania Giovannini* *Laboratory of Cardiovascular Endocrinology and Cell Biology, G. Monasterio Foundation, CNR Regione Toscana, 56126 Pisa, Italy † Scuola Superiore Sant’Anna, 56126 Pisa, Italy

1. Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Background and Aim. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. Introduction: Troponin Framework Within Myocardial Cells and Release Kinetics After Myocardial Damage. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4. Impact of the New Definition of Myocardial Infarction on Laboratory Practice and Instrumentation: The Need for High-Sensitivity cTnI and cTnT Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Quality Specifications for cTnI and cTnT Immunoassays. . . . . . . . . . . . . . . . . . . 4.2. The Development of High-Sensitivity Immunoassays for cTnI and cTnT Measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3. Definition of Highly (Ultra) Sensitive Immunoassay for cTnI and cTnT . . . . 5. The Impact of High-Sensitivity cTnI and cTnT Methods on Clinical Practice . . . . 5.1. The Problem of Reliable Definition and Accurate Estimation of the 99th Percentile Upper Reference Limit: Can Reference Values Be AVected by any Characteristics of the Reference Population? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2. More Myocardial Infarctions or More False Positive Results?. . . . . . . . . . . . . . 5.3. Clinical Relevance of Serially Measured Troponin Circulating Levels. . . . . . . 5.4. High-Sensitivity Troponin Methods in Patients with HF: A Better Stratification of Cardiovascular Risk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.5. Early Detection of Myocardial Injury in Patients with Extracardiac Diseases or Assuming Potentially Cardiotoxic Drugs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6. High-Sensitivity cTnI and cTnT Methods: A Powerful Tool for Monitoring Physiological Renewal and Pathological Remodeling of the Myocardial Tissue? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1

2 2 4

5 5 7 8 9

10 14 15 16 18

19

Corresponding author: Aldo Clerico, e-mail: [email protected] 1

0065-2423/09 $35.00 DOI: 10.1016/S0065-2423(09)49001-2

Copyright 2009, Elsevier Inc. All rights reserved.

2

CLERICO ET AL.

7. Use of High-Sensitivity cTnI and cTnT Methods in a Multimarker Approach for Early Screening: An Increase in Diagnostic and Prognostic EViciency?. . . . . . . . . . . 8. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

22 23 24

1. Abstract At the dawn of the new century, the advent of more specific myocardial tissue markers, such as cardiac troponin I (cTnI) and T (cTnT), has led to a new definition of acute myocardial infarction (AMI) by international guidelines. If we accept the concept that AMI is the portion of acutely necrotic myocardial tissue (irrespective of size), some patients previously diagnosed with severe angina may be currently considered to present minimal (even microscopic) quantities of myocardial necrosis. Although increased cTnI or cTnT values always indicate myocardial tissue damage, a positive test is not able to identify the mechanism responsible for that cardiac damage (which could be not due to ischemia). New cTnI and cTnT immunoassays with increased analytical sensitivity may increase ‘‘false positive’’ results in patients with cardiovascular disease, especially those with advanced age, heart failure (HF), severe comorbidities (such as chronic renal insuViciency), or assuming potential cardiotoxic drugs. Hence, it may be not clear for most patients and physicians whether high-sensitivity cTnI and cTnT methods will lead to more clarity or confusion. The aim of this review is to update the present knowledge in the field of cTnI and cTnT with particular attention on the impact of immunoassays with increased analytical sensitivity on both laboratory and clinical practice.

2. Background and Aim About 25 years ago, few diagnostic tests were available for clinical practice and considered useful in the assessment of cardiac necrosis, such as those measuring the total enzymatic activity of creatine kinase (CK) and lactate dehydrogenase. Unfortunately, those methods were characterized by low sensitivity and specificity for cardiac damage. Some immunoassay methods for structural proteins and cardiac isoenzymes, such as CK-MB isoenzyme and myoglobin, were then developed (Fig. 1). These markers showed an increased sensitivity, but only a relative specificity for cardiac disease, because these proteins are also present in the skeletal tissue.

HIGH SENSITIVITY TROPONIN ASSAY

3

1950 AST in AMI 1960 CK in AMI 1970

1979 - WHO criterial for AMI

Electrophoresys for CK and LD isoforms CK-MB activity RIA for myoglobin

1980

1990

2000 - New criterial for AMI

2000 2007 - Universal definition for AMI

CK-MB immunoassay Monoclonal antibody MB cTnT in AMI cTnT in unstable angina cTnI in AMI cTnT in cTnI in AMI for risk stratification High sensibility cTnT e cTnl

2010

FIG. 1. Brief history of cardiac marker for myocardial damage.

At the dawn of the new century, the advent of more specific myocardial tissue markers, such as cardiac troponin I (cTnI) and T (cTnT), ledch to the new definition of AMI by international guidelines [1, 2]. If we accept the concept that AMI is the portion of the myocardial tissue (despite size) with acute necrosis due to myocardial ischemia, several patients, previously diagnosed to have a severe angina, should be currently considered to present minimal (even microscopic) quantities of myocardial necrosis [1, 2]. As a result, the new definition of myocardial infarction has had a high impact on both laboratory and clinical practice [3–8]. The clinical application of international guidelines [1] generated main social/economical eVects, leading to a 25–55% increment of diagnosed AMI [3–5]. Although increased cTnI or cTnT values always indicate myocardial tissue damage, a positive test is unable to identify the mechanism responsible for

4

CLERICO ET AL.

that cardiac damage (which could be not due to ischemia). The advent of the new cTnI and cTnT immunoassays with increased analytical sensitivity may increase ‘‘false positive’’ AMI results in patients with cardiovascular disease, especially those with advanced age, HF, severe comorbidities (such as chronic renal insuViciency) or being treated with potential cardiotoxic drugs [3–5]. Hence, it may be not clear for most patients and physicians whether the new high-sensitivity cTnI and cTnT methods will lead to more clarity or confusion. To clarify these important clinical issues, a computerized literature search on National Library of Medicine (i.e., PubMed access to MEDLINE citations, http://www3.ncbi.nlm.nih.gov/PubMed/) was performed in June 2009 using keywords such as ‘‘troponin assays’’ (>7000 articles) and ‘‘high-sensitive troponin assays’’ ( 180 articles). The aim of this review is to update the present knowledge of cTnI and cTnT with particular attention to the impact of these new immunoassays with increased analytical sensitivity (i.e., the so-called high-sensitivity cTnI and cTnT immunoassay methods) on both laboratory and clinical practice.

3. Introduction: Troponin Framework Within Myocardial Cells and Release Kinetics After Myocardial Damage Troponin is a complex of three integrated proteins essential for both muscle contraction and relaxation, regulated by intracellular calcium concentration [9]. The troponin complex plays a fundamental role in the contraction of both cardiac and skeletal muscles, but not of smooth muscles. This complex interacts with two key molecules of the contractile process, the thin actin and the thick myosin filaments. Troponin is linked to the tropomyosin protein and is positioned among actin filaments within the muscle tissue. The three complex subunits, troponin C (TnC), troponin I (TnI), and troponin T (TnT), share diVerent physiologic properties. TnT binds the troponin group to tropomyosin, forming a troponin–tropomyosin complex, which is responsible for contraction. TnI binds to actin, secures the troponin–tropomyosin complex, and leads to muscle relaxation by interrupting the actin–myosin linkage. TnC binds to calcium ions producing a structural change in TnI, in order to interrupt relaxation and to begin the contraction cycle. Skeletal isoforms of TnT and TnI are replaced by cardiac-specific isoforms during fetal development of the human heart. At the end of the last century (Fig. 1), specific immunoassays for identifying cardiac muscle damage were

HIGH SENSITIVITY TROPONIN ASSAY

5

developed using antibodies to cTnI and cTnT. These assays were specific for identifying cardiac muscle damage and were free from interferences due to the presence of skeletal muscle isoforms [10]. First generation cTnT assays were, however, susceptible to false positivity due to cross-reactivity with skeletal TnT antibody [10]. Second generation immunoassay methods, designed using more highly specific antibodies, solved the interference problem with skeletal muscle isoforms and showed comparable results with cTnI assays [11–13]. Substantial data exist today that conclusively demonstrate that methods that rely on cTnI and cTnT detection share absolute specificity for myocardial damage. Cardiac troponins appear in the serum relatively early following onset of AMI (2–10 h), peak at 12–48 h, and remain abnormal for 4–14 days (cTnI 5–10 days and cTnT 5–14 days) [11–13]. These release kinetics can be accounted for by examining the distribution of the proteins within the myocardial cell. The great majority of both cTnI and cTnT is bound to the myofibril (94–97%), and only a relatively small amount ( 3% for cTnI and 6% for cTnT) free in the cytoplasm [11, 14]. Following cardiac cell injury and immediate release of the free cytoplasmic pool, there is a slow, but continuous and prolonged release of troponins presumably from myofibril-bound proteins [11, 14]. It is unclear, however, whether this early releasable troponin pool is actually free in the cytoplasm or loosely bound to myofilaments.

4. Impact of the New Definition of Myocardial Infarction on Laboratory Practice and Instrumentation: The Need for High-Sensitivity cTnI and cTnT Methods 4.1. QUALITY SPECIFICATIONS FOR CTNI AND CTNT IMMUNOASSAYS According to the new definition of AMI [1, 2], cardiac-specific troponins (cTnI and cTnT) are the preferred biomarkers, and if available, they should be measured in all patients with typical chest pain. An increase of cTnI or cTnT levels over the 99th percentile upper reference limit (99th URL) (cut oV value) should be considered clinically relevant. Furthermore, it is recommended that cTnI and cTnT values corresponding to the 99th URL should be measured with an imprecision, or coefficient of variation (CV) 10% [1, 2]. Finally, it has been suggested by international guidelines [1, 2] and quality specifications [6] that each laboratory independently confirm reference intervals, although assay standardization is preferable [4, 6, 7].

6

CLERICO ET AL.

The first important analytical issue is epitope location on the troponin molecule. It is important to note that the amino- and carboxy-terminal ends are more susceptible to proteolysis and this degradation may be related to the degree of tissue ischemia. Interestingly, these modified ‘‘partially degraded’’ products, not intact cTnI, were specifically detected in eVluents from severely ischemic hearts [11]. International guidelines [6] for immunoassay development have recommended that the epitope should be identified and located within a stable region of the cTnI molecule. Furthermore, specific relative responses are required for cTnI forms. These include free cTnI, the I–C binary complex, the T–I–C ternary complex, and oxidized, reduced, and phosphorylated isoforms of the three cTnI forms [6]. cTnI and cTnt can be determined by a number of commercial immunoassays with diVerent epitope-specific antibodies. As such, it can be expected that diVerences in assay response to the various troponin forms probably detect slightly diVerent patient populations depending on the nature and timing of cardiac troponin release [11, 13, 14]. These complications, in addition to diVerences in assay generation, create a substantive problem for clinical and laboratory interpretation of test results. The second important analytical issue is specificity of troponin antibodies. Apart from the cTnT method, which is oVered by one patent-protected vendor, there are more than 20 cTnI immunoassays commercially available [14, 15]. It can be safely assumed that antibodies in these diVerent assays do not bind all to the same epitope and therefore they measure diVerent cTnI forms. In addition, cTnI assays vary with respect to the antigen used for calibration, antibody type itself, and indicator molecule. Detection of antigen–antibody complexes also vary and may involve spectrophotometric, fluorescent, chemiluminescent, or electrochemical methods. Consequently, diVerent TnI assays do not produce equivalent concentration results [4–8] and comparison of absolute troponin concentration should not be made [14]. Indeed, numerous manufacturers have developed their own cTnI assays, leading to a situation in which cTnI measurements, using diVerent methods on identical specimens, have been shown to diVer by more than 20-fold [4–8]. Unfortunately, standardization of cTnI methods, despite continuous solicitation and recommendations [4, 6–8], has been diVicult to achieve and remains in progress [14, 15]. Many, or even most, commercially available cTnI and cTnT methods do not actually report the 99th URL value, nor achieve the precision (10% CV) required for assay reproducibility at the cutoV [4–8]. Increased assay precision and improved standardization is mandatory in order to achieve common reference and decision limits for troponin immunoassays in accordance with international guidelines and quality specifications [1, 2, 6].

7

HIGH SENSITIVITY TROPONIN ASSAY

4.2. THE DEVELOPMENT OF HIGH-SENSITIVITY IMMUNOASSAYS FOR CTNI AND CTNT MEASUREMENT The European Society of Cardiology (ESC), the American College of Cardiology Foundation (ACCF), the American Heart Association (AHA), and the World Heart Federation (WHF) now recommend a single cTnI and cTnT decision cut-point for the diagnosis of myocardial infarction in patients presenting with suspected myocardial necrosis correspond to the 99th percentile upper reference limit (99th URL) [2]. This very low cut oV concentration, however, creates a significant problem because most assays lack the analytical sensitivity to consistently measure troponin in the blood of apparently healthy individuals. This results in a high proportion of reference population values below the limit of detection for most methods. As such, the 99th URL cannot be ascertained with any acceptable degree of analytic certainty or basis [15, 16]. Furthermore, the new definition of AMI [1, 2] that specifically requires assay precision 10% CV for the 99th percentile of the reference population, remains a diVicult challenge for manufacturers of commercial cTnI and cTnT immunoassays. In fact, following establishment of the new AMI definition [1], no commercial immunoassay was able to fulfill this recommendation [7, 8]. The development of more sensitive and better precision assays should permit more reliable estimation of very low cTnI and cTnT concentration. It is likely that significant improvement in troponin assay sensitivity is required to reproducibly measure near or below the ng/L concentration where reference values may be Gaussian-distributed [5, 14]. As a result of this challenge, next generation of cTnI and cTnT assays have been recently developed to improve the analytical performance and standardization [17–29]. It is noteworthy that some of these new methods are characterized by improved low-end analytical sensitivity and precision, which should increase precision at the cutoV (99th percentile of the reference population) to about 10% or even better (Table 1). TABLE 1 DETECTION LIMIT, ANALYTICAL SENSITIVITY, AND 99TH URL OF SOME HIGHLY SENSITIVE IMMUNOASSAY METHODS FOR CTNI AND CTNT

DL (ng/L)

10% CV (ng/L)

99th URL (ng/L)

Ratio

References

cTnI Assay Ultra ADVIA Centaur Singulex Erenna Ultra Accu TnI

6 0.2 6

57 0.91 14

72 9 40

0.8 0.1 0.35

21, 28, 29 24 18, 25, 45

cTnT Assay Elecsys hs TnT

2

12

14

0.85

20, 22, 23

Method

DL, Detection limit.

8

CLERICO ET AL.

4.3. DEFINITION OF HIGHLY (ULTRA) SENSITIVE IMMUNOASSAY FOR CTNI AND CTNT An important issue in the development as well as in the practical use of highly sensitive cTnI and cTnT immunoassays is the appropriate definition of assay sensitivity. This definition directly impacts two aspects of assay performance: limit of detection and assay precision [15]. Accurate discrimination of ‘‘minor’’ myocardial damage versus analytical noise requires assays with excellent limit of detection and a high precision at low troponin concentration. New generation cTnI and cTnT immunoassays have been characterized by a limit of detection at the picogram or subpicogram level (Table 1). A simple calculation may better explain the impact of increased analytical sensitivity in clinical practice. For example, highly sensitive cTnI and cTnT methods have a limit of detection <10 pg/mL (Table 1). Cardiomyocytes contain 70 mg cTnI per gram myocardial tissue [30], 10 pg cTnI is contained in 1 mg myocardial tissue [31]. As such, it is conceivable that necrotic damage to 10–50 mg of myocardial tissue should be detectable by highly sensitive troponin methods. There is a lack of consensus regarding the best method to assess immunoassay sensitivity at very low analyte concentration [15]. The limit of detection has been defined by the Clinical and Laboratory Standards Institute as the lowest amount of analyte in a biological sample that can be reliably detected by a given analytical procedure [32]. The limit of detection for cTnI and cTnT is usually estimated as the protein concentration that corresponds to a signal 2 or 3 standard deviations (SD) above the mean of at least 20 replicates for a sample absent in troponin (zero calibrator). This calculated value should be considered as the limit of blank, which is defined as the highest measurement result likely to be observed (with a stated probability) for a sample that contains no troponin, rather that the true assay limit of detection [15, 32]. Thus, the major diVerence in estimating assay limit of detection and limit of blank is the sample type used for measurement. Zero calibrators, assay diluents, or serum troponin-free (recommended) are only useful for determination of limit of blank, while in order to have an adequate estimation of the immunoassay detection limit, serum samples containing cTnI or cTnI concentrations in the range from the blank value to fourfold the blank value should be used [15, 32]. Therefore, limit of detection values reported in the literature are generally lower than those obtained using the recommended experimental procedure [32]. From a clinical point of view, the most important analytical characteristic should be the limit of quantification [15, 32], also called functional sensitivity [33]. This quantity is defined as the lowest amount of analyte (cTnI or cTnT) that can be quantitatively measured with stated acceptable precision and bias (i.e., measurement uncertainty) [34]. As previously mentioned, international guidelines [1, 2] specifically require an assay precision 10% CV for the 99th

HIGH SENSITIVITY TROPONIN ASSAY

9

percentile of the reference population. It is important to note that this degree of precision (10% CV) is slightly better than an optimal total error goal (12% CV) as suggested by other authors considering the biologic variation of cTnI [14, 35]. According to international guidelines and quality specifications [1, 2, 8, 36], it is conceivable that an immunoassay for the measurement of cTnI or cTnT should be defined to be highly (or ultra) sensitive if it is capable of measuring the 99th percentile of the reference population with an total error <10% CV (10% CV concentration to 99th percentile limit ratio <1). However, there are some analytic and clinical problems concerning this definition. First, there is lack of consensus in the literature about the best method to assess and report precision data [14]. Manufacturer assay package inserts often report only precision based on within-run or between-day evaluation of samples with cTn concentrations much higher than the AMI cutoV. Furthermore, these data do not usually include lot-to-lot and machine-to-machine (interinstrument) variability [14]. As such, cTnI and cTnT assay variability determined within the clinical laboratory is frequently higher than those quoted by the manufacturer. For example, the ADVIA Siemens TnI-UltraTM concentration at 10% CV determined across two hospital sites and five analyzers using three reagent lots was 0.10 g/L [26] compared with 0.03 g/L per package insert and 0.05–0.07 g/L from literature studies [27–29, 37]. Second, some troponin values, measured in the reference population, may be still below the analytical sensitivity of the new generation immunoassay methods [28–39]. This suggests that the precision at low troponin values of these methods should be further improved in order to measure the protein concentration in each plasma sample of all healthy subjects. In consideration of these findings, Apple [39] recently suggested to divide the new cTnI and cTnT methods into four levels, according to the percentage of measurable normal values below the 99th percentile: level 1 (contemporary) <50%; level 2 (first generation with high sensitivity) from 50% to <75%; level 3 (second generation with high sensitivity) from 75% to <95%; level 4 (third generation with high sensitivity) 95%. However, the fraction of measurable normal values may strongly depend on some demographic characteristics (i.e., gender, age, and myocardial ventricular mass) of the reference population studied [16, 21, 28]. These issues will be discussed in the following sections.

5. The Impact of High-Sensitivity cTnI and cTnT Methods on Clinical Practice International guidelines [1, 2] recommend the measurement of cardiac damage markers (usually cTnI or cTnT) in each patient with suspected acute coronary syndrome (ACS). From a clinical point of view, only cardiac

CLERICO ET AL.

10

Acute coronary syndrome 1 step

ECG ! NSTEACS

2 step

@ STEMI

Necrosis biomarker

Diagnostic UA

NSTEMI

Confirmatory STEMI

FIG. 2. Dierential diagnosis of acute coronary syndromes according to consensus document of the Joint European Society of Cardiology/American College of Cardiology Committee for the redefinition of myocardial infarction [1].

troponin measurement is able to distinguish patients with unstable angina (UA) from those with non-ST-segment elevation myocardial infarction (NSTEMI) (Fig. 2) [1, 2]. However, increased cTnI or cTnT values alone are unable to indicate the pathophysiological mechanism underlying the detected myocardial damage, which may be unrelated to ischemia. Therefore, an increased marker value, without clinical indication of myocardial ischemia, should prompt a search for other causes of cardiac damage (Table 2). On the other hand, there is clear evidence that any amount of detectable cardiac troponin is associated with an increased risk of new adverse cardiac events in patients with ACS, with no record of a threshold below which troponin elevation is harmless and meaningless for prognostic stratification [40–43]. These data suggest that there may be some unresolved problems in the definition of decisional (i.e., cut oV) values for diVerential diagnosis in ACS, as well as, risk stratification of patients with cardiovascular diseases. These issues will be discussed and clarified in the following sections.

5.1. THE PROBLEM OF RELIABLE DEFINITION AND ACCURATE ESTIMATION 99TH PERCENTILE UPPER REFERENCE LIMIT: CAN REFERENCE VALUES BE AFFECTED BY ANY CHARACTERISTICS OF THE REFERENCE POPULATION?

OF THE

In their Universal Definition of AMI, the ESC/ACCF/AHA/WHF Task Force for the Redefinition of AMI recommended, as criteria for nonprocedure-related AMI, the evidence of increased or decreased cTnI or

HIGH SENSITIVITY TROPONIN ASSAY

11

TABLE 2 THE MOST FREQUENT CLINICAL CONDITIONS IN WHICH THE CIRCULATING LEVELS OF CARDIAC TROPONINS ARE INCREASED, WITHOUT OVERT CORONARY ARTERY DISEASE

Myocarditis/pericarditis Congestive heart failure Systemic arterial hypertension Systemic arterial hypotension (especially if associated with cardiac arrythmias) Cardiac surgery or catheterization (e.g., ablation) Critically ill patients Hypothyroidism Cardiac trauma Myocardial toxicity from cancer therapy Pulmonary embolism Episode rejection of a cardiac transplant Postoperative noncardiac surgery Chronic renal failure Amyloidosis Sepsis

cTnT with one or more values above the 99th URL, found in a clinical setting suggestive of myocardial ischemia, together with either clinical symptoms, new ischemic ECG changes, or imaging findings of new loss of myocardium [2]. According to this definition, a reliable estimation of the 99th URL assumed a central role in the clinical diagnosis of AMI. In addition to assay sensitivity, the main factor that may influence the 99th URL estimation is the selection of the reference population, including number, type, age, and gender of individuals enrolled in the study [17, 19, 21, 28, 38, 44–46]. Finally, the matrix of the sample employed (serum or plasma) for this specific evaluation may also aVect results [15]. The sample size is an important factor to take into account for the 99th URL estimation. International guidelines recommend a minimum of 120 reference individuals per group for appropriate statistical determination of reference limits [47]. However, a sample size of at least 300 individuals is required to reach the 95% probability, that at least 99% of the population will fall below the highest observed analyte value [48] (Table 3). Under these conditions, the uncertainty in defining true 99th URL is high because the cut oV concentration is approximately equal to the individual having the third highest cTnI or cTnT concentration. Thus, if three more apparently healthy individuals with somewhat increased troponin concentrations were included in the reference group, the calculated 99th URL would be substantially changed [15]. Age- and gender-dependent diVerences may also have significant clinical relevance on the 99th URL estimation. This has been demonstrated in recent

CLERICO ET AL.

12

TABLE 3 SAMPLE SIZES AND ASSOCIATED TOLERANCE LEVELS FOR REPORTING 99TH PERCENTILE BASED ON THE LARGEST OBSERVED VALUE (ACCORDING TO REFERENCE 48) Sample size 100 200 300 400 500

Tolerance level 0.63 0.87 0.95 0.98 0.99

reference population studies that included at least 300 individuals and used highly sensitive methods for cTnI or cTnT [28, 38, 45, 46]. A study from our laboratory [28] included 692 apparently healthy subjects (311 males and 381 females) with a mean (SD) age of 45.3 (17.3) years [range 11–89 years; females 46.5 (17.3) years, males 43.8 (17.1) years]. Our study found significant gender-based diVerence in cTnI values (men: median 0.012 g/L, range from undetectable values to 0.196 g/L; women: median 0.008 g/L, range from undetectable values to 0.130 g/L; p < 0.0001) using the ADVIA TnIUltra method (Siemens Medical Solutions Diagnostics SrL). Undetectable cTnI concentrations were found in 168 individuals (24.3% of the samples tested). All individuals used in this study were screened for preventive medicine programs (laboratory staV, blood donors, or voluntary subjects), with no acute or chronic diseases (excluded by history, accurate clinical examination, ECG, and laboratory tests), nor use of drugs for at least 2 weeks before the sample collection. A gender-dependent cut oV value was also found in another study [45] that used the recently refined Access AccuTnI assay (Beckman-Coulter) to assess the distribution of cTnI in a community population of elderly individuals [PIVUS (Prospective Study of the Vasculature in Uppsala Seniors) study; n ¼ 1005]. Gender-dependent diVerences in 99th URL for the highly sensitive cTnT assay by Roche Diagnostics was also reported by Mingels et al. [46] in a reference population of 479 apparently healthy individuals; the observed 99th percentile was 0.008 g/L in 215 females and 0.018 g/L in 264 males (p < 0.001). In contrast, Collinson et al. [38] reported that the calculated 99th URL of cTnI concentration, measured with the ADVIA TnI-Ultra method, was very similar to that reported by the manufacturer (0.04 g/L) and cTnI values were not age- and gender-dependent [38]. Moreover, 165 (53.4%) of the 309 individuals (127 men, 182 women; median age 53 years, range 45–80 years) enrolled were considered to have no measurable cTnI. The individuals who participated in this study were randomly selected from a population of

13

HIGH SENSITIVITY TROPONIN ASSAY

ostensibly healthy individuals, accurately screened to exclude any history of vascular disease, diabetes mellitus, hypertension, heavy alcohol intake, use of cardiac medication, or pathologic echocardiogram. Some interesting observations may be derived from these studies [28, 38, 45, 46]. First, the selection of reference population may greatly influence the calculation of 99th URL. In particular, the clinical protocol used to exclude the presence of asymptomatic cardiac disease (especially in older subjects) is likely to aVect the statistical analysis with respect to distribution of cTnI measured by highly sensitive methods. Second, the presence of cTnI and cTnT measurable values in apparently healthy subjects requires physiologic explanation. Increased concentrations of cTnI and cTnT have been observed in animal models of ischemia without histologic evidence of irreversible injury [49]. Moreover, apoptotic cells have been described in normal adult hearts, suggesting that myocyte replication is a significant component of normal physiology and cellular processes, even in adults [50]. This finding will be discussed in more detail in the following section. It can be suggested that the release of cTnI from cardiomyocytes of healthy adult subjects may result from a process related to the ‘‘physiological remodeling’’ of human myocardium [28, 31]. Age-dependent increases in cTnI and cTnT in apparently healthy subjects may suggest additional pathophysiological mechanisms. It is well known that the incidence of HF progressively and steeply increases after the age of 55 years (Table 4) and that this disease is the most common cause of death in elderly people [51]. Several histological changes of myocardial tissue characterized by loss of myocytes and subsequent hypertrophy of the remaining cells and calcification of several cardiac structures can be found in most individuals with aging [52, 53]. Moreover, the age-related loss of arterial compliance contributes to isolated systolic hypertension and left ventricular hypertrophy [52, 53]. Despite these changes, for the majority of apparently healthy older

TABLE 4 PREVALENCE OF SYSTOLIC AND DIASTOLIC DYSFUNCTION BY AGE (ACCORDING TO REFERENCE 51) Dysfunction

45–54 years

55–64 years

65–74 years

75 and older

Overall

Diastolic Moderate Severe (restrictive)

1.4% 0%

6.0% 0.4%

9.9% 0.7%

14.6% 3.4%

6.6% 0.7%

Systolic LVEF 50% LVEF 40%

3.0% 0.8%

4.8% 1.3%

7.1% 2.7%

12.9% 4.4%

6.0% 2.0%

LVEF, Left ventricular ejection fraction.

14

CLERICO ET AL.

adults, cardiac output is well preserved by means of the Frank–Starling principle, in the setting of reduced early diastolic filling [53]. In accordance with these findings [52, 53], we suggest that increased levels of cTnI, measured with high-sensitivity immunoassay methods in some apparently healthy older adults [28, 45, 50], are likely to be due to increased remodeling of myocardial tissue in this population. This hypothesis is well in agreement with the results reported by Eggers et al. [54]. This study investigated the prevalence of cTnI elevation in an elderly community population that included 1005 individuals aged 70 years. Using a highly sensitive immunoassay, this study found that increased cTnI was relatively common in elderly subjects and was associated with cardiovascular high-risk features and/or impaired cardiac performance [54]. Cumulatively, these data strongly indicate that calculation of 99th URL is dependent on demographic and clinical characteristics of the reference population used in the study. Hence, clinical cutoVs using highly sensitive cTnI and cTnT assays should be based on analytic definitions (i.e., CV) versus distribution characteristics (i.e., percentiles) such that ‘‘true’’ troponin increase may be identified [38]. Furthermore, the demographic and clinical characteristics of the reference population enrolled for calculation of 99th URL should be clearly delineated by the commercial manufacturers as well as authors of published clinical studies. 5.2. MORE MYOCARDIAL INFARCTIONS OR MORE FALSE POSITIVE RESULTS? It is reasonable that the new generation of high sensitivity cTnI and cTnT methods can detect a greater number of patients with AMI than standard methods, especially those individuals with very small infarct size [5, 55]. Data reported by some recent studies appear to confirm this hypothesis [19, 55–59]. Unlike most commercial cTnI and cTnT methods, the use of highly sensitive methods (i.e., with increased assay sensitivity at very low troponin concentration) will allow earlier diagnosis of AMI (within 1–2 h after thoracic pain onset), as well as recognition of very small (focal) areas of myocardial necrosis as true AMI. Although professional societies have recognized the importance of the enhanced analytic performance of the newer and emerging cardiac troponin assays, the clinical community has not uniformly embraced this trend [55]. Indeed, it is uncertain if the new highly sensitive troponin assays will lead to increased clarity or more confusion for most physicians [59]. Predictably, the application of assays with lower limits of detection has led to an increase in patients evaluated in the emergency setting with detectable cardiac troponin in a variety of acute and chronic medical conditions other than ACS [5, 16, 19, 31, 55, 56].

HIGH SENSITIVITY TROPONIN ASSAY

15

It is also reasonable that increased cardiac troponin in apparently stable populations, such as elderly subjects from the community and patients with previous ACS, may primarily reflect left ventricular hypertrophy and/or myocardial pump failure with a continuous loss of viable cardiac myocytes caused by increased myocardial wall strain, chronic ischemia, or apoptosis. It is well known that these conditions are often associated with ST-T segment abnormalities that may mimic changes related to acute coronary ischemia. Accordingly, several studies have already demonstrated that more sensitive troponin assays increased the number and rapidity of AMI diagnosis, but also increased the number of false positives, that is, non-ACS-related pathology [16, 19, 25, 56, 57]. In fact, Eggers et al. [16] reported a 7% misdiagnosis (i.e., false positivity rate) for AMI in troponin-positive patients with preexisting ST-T segment abnormalities in patients admitted for nonischemic chest pain or other symptoms indicative of myocardial ischemia. In conclusion, these data confirm that it is very diVicult (or even impossible) to reliably diagnose patients suspected with ACS using only one determination of cTnI or cTnT due to the relatively low specificity of existing cardiac troponin assays for ischemic myocardial injury. Indeed, international guidelines [1, 2] recommend at least two samples with a delay of time of 6–12 h for measurement of cTnI and cTnT in these settings. 5.3. CLINICAL RELEVANCE OF SERIALLY MEASURED TROPONIN CIRCULATING LEVELS It is important to note that the detection of a true and significant increase/ decrease in serially measured troponin is of critical importance to correctly establish the diagnosis of AMI in all patients without a diagnostically reliable or recent electrocardiogram [1, 2] (Fig. 2) and to discriminate between ischemic and other causes of troponin increase [1, 2, 16, 35, 60, 61]. Unfortunately, there is no consensus about the required degree of change for serial measurement of cTnI and cTnT in AMI diagnosis in patients with suspected ACS. The National Academy of Clinical Biochemistry has recommended a 20% change as statistically significant [60]. However, these recommendations assume that analytical assays have a precision of 5–7% with three times the SD and produce 99% confidence at limit at the AMI decision point [61]. A statistically more rigorous approach toward assessment of meaningful serial markers in clinical laboratory tests would be to first ascertain biologic variation [14, 35]. Unfortunately, troponin biologic variation cannot be evaluated with certainty using the standard methods due to their inability to detect the protein in the blood of healthy subjects with adequate precision. Using new high-sensitivity assays, Wu et al. [35] were able to demonstrate

16

CLERICO ET AL.

that cTnI biological variation was lower than other cardiovascular biomarkers, that is, cardiac natriuretic peptide hormone, creatine kinase-MB fraction (CK-MB), myoglobin, C-reactive protein (CRP), myeloperoxidase, and serum amyloid A [35]. Because this study was performed in healthy subjects, there is concern regarding the applicability and reliability of these biologic variation parameters in patients with ACS. One would expect that there are diVerences in biologic variation parameters between healthy individuals and patients with ACS. In comparison to cTnI, cTnT has diVerent release kinetics from myocytes and clearance in peripheral tissues and therefore its biologic variation should be evaluated in specific studies [35]. Clearly, more comprehensive studies are required to confirm that measurement relative to biologic variation is useful in evaluating the clinical significance of cardiac troponin in patients with ACS. 5.4. HIGH-SENSITIVITY TROPONIN METHODS IN PATIENTS WITH HF: A BETTER STRATIFICATION OF CARDIOVASCULAR RISK HF is a major public health problem in the North America and Europe [62– 65]. The incidence and prevalence of HF increases significantly with aging in these populations. After the age of 65, the incidence of HF approaches 10 per 1000 of population ( 1:100) [51]. In the United States, HF is the most common hospital discharge diagnosis, and more Medicare dollars are spent for diagnosis and treatment of HF than for any other disease [62–64]. Similar data have been reported from European countries [65]. HF may be considered as the fatal progression of all cardiovascular disorders. For this reason, HF is considered a syndrome rather than a primary diagnosis which results from any structural or functional cardiac disorder that impairs the ability of the heart to function as a pump to supporting physiologic circulation [5, 31]. We can assume that, if heart dysfunction is an inevitable and ultimate fate, the measurement of some highly specific cardiac biomarkers, such as cTnI, cTnT, B-type natriuretic peptide (BNP) and its related forms, should be useful in detection of people at risk of a more rapid progression toward symptomatic HF, thus in need of specific clinical treatment [5, 31]. In 2001, the AHA/ACC task force for the diagnosis and management of chronic HF introduced a new classification that focused on disease evolution [62]. This classification, updated in 2005 [63] and 2009 [64], identified four stages (A, B, C, and D) that account for symptoms, established risk factors for HF development and structural myocardial abnormalities. Stage A includes asymptomatic patients at risk for developing HF with no structural cardiac involvement. Stage B includes asymptomatic patients at risk for developing HF with structural cardiac involvement. Stage C includes

HIGH SENSITIVITY TROPONIN ASSAY

17

patients with past or current symptoms of HF associated with underlying structural heart disease. Stage D includes symptomatic patients with endstage disease who require specialized treatment strategies, such as mechanical circulatory support, continuous inotropic infusions, cardiac transplantation, or hospice care. Unlike the NYHA classification which is based on clinical severity of symptoms, this new classification emphasizes the progressive nature of the pathophysiological processes responsible for development of HF [64]. In fact, the first two stages (A and B) clearly do not include HF but help in the early identification of patients at risk for developing HF [64]. Stages A and B patients are best defined as those individuals with risk factors that clearly predispose to the development of HF. Appropriate risk stratification depends on the availability of specific, accurate, and eVective disease and risk markers [5, 31]. Highly sensitive cTnI and cTnT immunoassay methods share the most important analytic and clinical performance characteristics of an ideal cardiac biomarker (Table 5) [5, 31]. It is well known that a relatively large proportion of HF patients (25–45%), especially those with clinical history of coronary artery disease, has increased cTnI and cTnT, even if measured by standard (i.e., not highly sensitive) methods [64]. Recent studies [16, 54, 58, 59] have suggested that the fraction of patients with HF and troponin values above the 99th URL may further increase when highly sensitive cTnI and cTnT assays are used. In particular, the ValHeFT study (including 4053 randomized patients with symptomatic heart failure) demonstrated cTnT values above the cut-off level in 10.4% of patients studied using a standard assay and this percentage increased to 92% when a more sensitive method was used [66].

TABLE 5 DESIRABLE FEATURES OF AN IDEAL MOLECULAR CARDIAC BIOMARKER

Absolute cardiospecificity Acceptable to patient Stability in vivo and in vitro Adequate analytical sensitivity (functional sensitivity) easy to perform Good degree in reproducibility and accuracy Complete automation of assay Internationally standardized Low cost Low biological variation Reference range and cut off values tested for gender, age, and ethnicity dependence Good diagnostic and prognostic accuracy Favorable cost-benefit ratio

18

CLERICO ET AL.

It is well known that troponin and natriuretic peptide, when used as biomarkers of cardiac disease, furnish complementary clinical information [5, 31]. The progressive increase in both markers with aging [5, 20, 28, 31, 45] suggests that there is progressive decline of cardiac function which can be assessed and monitored. If heart dysfunction is an inevitable and ultimate fate, the measurement of these analytes would be useful to initiate specific clinical care for those individuals at risk of more rapid progression to symptomatic cardiac failure. 5.5. EARLY DETECTION OF MYOCARDIAL INJURY IN PATIENTS WITH EXTRACARDIAC DISEASES OR ASSUMING POTENTIALLY CARDIOTOXIC DRUGS The Dallas Heart Study (3357 subjects) found that the prevalence of increased cTnT in the general population was 1% using electrochemiluminescence immunoassay (ECLIA) (Roche Diagnostics), a standard assay [67]. The prevalence of high-risk cardiovascular features was increased similarly in subjects with cTnT levels in the minimally increased and increased range [67]. These data suggest that cTnT elevation is always indicative of cardiovascular disease or at least a high-risk cardiovascular profile. As such, increased cardiac troponin represents an important finding even in those patients without coronary artery disease. This ‘‘index’’ of cardiac tissue damage may suggest an appropriate diagnosis and, when necessary, a specific treatment. Increased cTnI or cTnT in patients without evidence of ischemic coronary artery disease represent an independent risk of future cardiac events and poor prognosis [5, 31, 60]. It is important to point out that there is no minimal threshold and that cardiovascular risk increases progressively with increased troponin. Increased troponin has been reported in patients treated with potentially cardiotoxic drugs, such as high-dose chemotherapy [68, 69]. Indeed, a specific pattern of troponin release following high-dose chemotherapy identified atrisk patients for cardiac events in the subsequent 3 years [68]. Furthermore, in patients treated with high-dose chemotherapy, increased cTnI suggested early and appropriate treatment and was found to prevent the development of late cardiotoxicity [69]. Another important application of highly sensitive troponin assay (along with cardiac natriuretic peptides) may be the early detection of myocardial damage in patients with systemic acute or chronic inflammatory and rheumatic diseases (i.e., systemic lupus erythematosus, systemic amyloidosis, sarcoidosis, and rheumatoid arthritis) [70–73]. It is noteworthy that mortality risk in these patients is strongly associated with heart complications versus other organ involvement [72, 74–76]. As a result, the early detection of cardiac involvement may have a tremendous clinical impact on the prognosis

HIGH SENSITIVITY TROPONIN ASSAY

19

of patients with chronic inflammatory and rheumatic diseases, especially those with systemic amyloidosis [72]. Early identification of these patients is critical in the initiation of successful treatment strategies. It is well known that cardiovascular events are also the major prognostic determinants in patients with end-stage renal disease, with cardiovascular deaths representing more than 50% of total mortality [77, 78]. The early recognition of conditions such as left ventricular hypertrophy and coronary artery disease may allow the identification of patients with chronic kidney disease at higher risk of developing either HF or other major cardiovascular events with consequent increased mortality [77, 78]. In a considerable number of chronic hemodialysis patients, increased troponin was found despite absence of cardiac ischemia, even if older generation cTnI and cTnT immunoassay methods were used [77, 78]. The elevation of cardiac biomarkers in patients with renal diseases showed a strong prognostic significance with respect to cardiovascular morbidity and mortality [5, 31, 73, 77, 78]. Recent studies have shown that highly sensitive cTnI and cTnT immunoassays were able to detect a greater number of end-stage renal disease patients with increased troponin [58, 78, 79]. As such, high-sensitivity cTnI and cTnT immunoassays may provide a useful rationale for both occult cardiac disease screening and better cardiovascular risk stratification in this unique group of patients. In conclusion, sensitive cTnI and cTnT immunoassays may provide a useful tool for the early screening of occult cardiac disease either in patients with extracardiac diseases (especially renal and chronic inflammatory diseases) or in subjects undergoing treatment with potentially cardiotoxic drugs (such as high-dose chemotherapy). However, further studies are necessary to demonstrate the clinical impact and the cost-eVectiveness of this approach according to the evidence-based laboratory medicine principles (EBLM) [80].

6. High-Sensitivity cTnI and cTnT Methods: A Powerful Tool for Monitoring Physiological Renewal and Pathological Remodeling of the Myocardial Tissue? Increased analytical sensitivity of cTnI and cTnT methods has demonstrated that measurable levels of these proteins are also present in apparently healthy subjects [20, 28, 38] (Fig. 3). These findings suggest some interesting pathophysiological considerations. At present time, the prevailing opinion, based on the aggregate evidence to date, is that any reliably detected elevation of a cardiac troponin is abnormal and might represent cardiac necrosis [81]. However, apoptotic cells have been described in normal adult hearts; thus suggesting that myocyte replication is a significant component of the

CLERICO ET AL.

20 A 80 70 60

cTnI (ng/L)

50 40 30 20 10 0 10

20

30

40

50 Age (years)

60

70

80

90

60

70

80

90

B 100 90 80

cTnI (ng/L)

70 60 50 40 30 20 10 0 10

20

30

40

50 Age (years)

FIG. 3. (Continued)

HIGH SENSITIVITY TROPONIN ASSAY

21

physiological cellular processes even in adults [50]. A very recent experimental study, based on DNA integration of the isotope 14C generated by nuclear bomb tests during the Cold War, was able to establish cardiomyocyte age in humans [82]. The results of this study suggested that cardiomyocytes renew with a gradual decrease of annual turnover from 1% (age 25) to 0.45% (age 75) with fewer than 50% of cardiomyocytes exchanged during a normal life span [82]. At present time, there are no experimental data indicating that troponins are degraded within the cardiomyocytes and released into the interstitial space during apoptosis. There are two potential explanations for the troponin release in absence of lethal sarcolemmal disruption: (1) cellular release of proteolytic troponin degradation products; (2) troponin leaks as an intact nondegraded protein chain from reversibly damaged cardiomyocytes [83, 84]. Mechanical stretch of cardiomyocytes, for example, during pressure or volume overload, may activate some intracellular proteases, such as metalloproteinase, that can degrade cardiac troponin intracellularly [85]. Overload-induced stretch at the cardiomyocyte level is sensed by integrins, which are mechanotransducer molecules that link the extracellular matrix to the intracellular cytoskeleton [86]. Hence, this mechanism may be involved in stretch-induced release of troponin and its degradation products [83]. These findings suggest that stretch stimulation of viable cardiomyocytes may lead to intact cTnI release. Indeed, several studies have demonstrated that mechanically induced transient disruptions (wounding) of the sarcolemma are a constitutive in vivo event [87–90]. This mechanism may account for the release of proteins, like myocyte-derived growth factors that are released despite lack of the classic signal peptide sequence that is normally associated with secretion. These mechanically induced alterations in cardiomyocyte sarcolemmal permeability may similarly be involved in the release of cTnI from cytosolic pools in the absence of necrotic cell death. However, further studies are necessary to accurately describe the cellular mechanisms responsible for release of intact cTnI and cTnT in damaged cardiomyocytes. FIG. 3. (A) Age-dependent distribution of cTnI values measured by the ADVIA method on the Centaur Platform (Siemens Diagnostics) in 269 apparently healthy male subjects (age ranging from 14 to 88 years). There is a very weak, although significant, correlation between age and cTnI values (by Spearman Rank test, Rho ¼ 0.358, p < 0.0001). The trend, assessed by smooth spline analysis, between age and cTnI values is indicated by a continuous line. (B) Age-dependent distribution of cTnI values measured by the ADVIA method on the Centaur Platform (Siemens Diagnostics) in 238 apparently healthy female subjects (age ranging from 14 to 88 years). There is a very weak, although significant, correlation between age and cTnI values (by Spearman Rank test, Rho ¼ 0.258, p < 0.0001). The trend, assessed by smooth spline analysis, between age and cTnI values is indicated by a continuous line. Results obtained in the Authors’ laboratory (see references 21, 28, 29).

22

CLERICO ET AL.

The data, regarding gender- and age-related cTnI and cTnT levels in adult healthy subjects [21, 23, 28, 45], support the hypothesis that small amounts of cTnI can be released from cardiomyocytes even in apparently healthy subjects due to a process related to the ‘‘physiological renewal or remodeling’’ of human myocardium. Moreover, some findings obtained in healthy individuals after endurance exercise appear to confirm this hypothesis. Several studies reported increased circulating cTnI or cTnT after strenuous exercise (such as marathon runs or other endurance races), even in well-trained athletes [91–98]. Middleton et al. [95] suggested that it is unlikely that minor elevations in cTnI or cTnT subsequent to endurance exercise are due to myocardial necrosis. These authors hypothesized that postexercise troponin release represents the reversible cardiomyocyte membrane damage during remodeling processes [95]. According to this hypothesis, in a healthy exercising population, cardiac troponins may be routinely released after periods of increased myocardial demand, such as after endurance exercise. As mentioned earlier, recently developed methods for cTnI and cTnT assays may be able to detect a release of protein from a quantity of myocardial tissue of a few milligrams in size [20–30] (Table 1). Highly sensitive cTnI and cTnT immunoassays should be considered a useful and potentially powerful tool to monitor the continuous and physiological processes related to renewal and remodeling of the myocardial tissue.

7. Use of High-Sensitivity cTnI and cTnT Methods in a Multimarker Approach for Early Screening: An Increase in Diagnostic and Prognostic Efficiency? Many biomarkers exist for the diagnosis and prognosis of individuals with cardiovascular diseases. Methods of assessment using a multimarker approach have been considered the best model for risk prediction in individuals with cardiovascular disease [5, 31, 99]. Despite the many suggested laboratory biomarkers for diagnosis, risk assessment, and follow-up of patients with cardiovascular disease, only cardiac troponins and natriuretic peptides have been shown to have good diagnostic and prognostic accuracy, as well as cost-eVectiveness, according to EBLM principles [5, 31, 99]. Indeed, these two biomarkers are characterized by high analytical and clinical sensitivity as well as by absolute cardiospecificity. Cardiac troponins and natriuretic peptides have diVerent physiological roles. Troponins are structural proteins of actin–myosin complex involved in cardiac contraction and relaxation, whereas the latter are peptide hormones with natriuretic and vasodilative eVects produced by cardiomyocytes. Furthermore, diVerent pathophysiological mechanisms aVect the release of cardiac troponins and

HIGH SENSITIVITY TROPONIN ASSAY

23

natriuretic peptides by cardiomyocytes [5, 31, 100–103]. Owing to their diVerent pathophysiological roles, cardiac troponins and natriuretic peptides may provide independent pathophysiological and clinical information. Due to their unique roles, these markers may be combined in a multimarker model for cardiovascular risk assessment and as predictive variables in practically every acute and chronic cardiac disease, including those due to ischemic, inflammatory, congenital, and traumatic conditions. A substantial number of studies have confirmed this hypothesis [5, 13, 31, 60, 81, 104–109]. Furthermore, both biomarkers can be measured using highly sensitive immunoassays with detection limits below 10 pg/mL [5, 31, 108]. As such, these biomarkers may be useful to assess disease progression and severity in patients with stage B HF (ACC/AHA guidelines) [62–64] and patients with end-stage renal disease [76]. Although some recent studies provide some support [25, 54, 109–111], further and more comprehensive studies are required to demonstrate the eVectiveness of highly sensitive cTnI and cTnT, alone or together with BNP, in early screening of cardiac disease in the general population or in high-risk individuals.

8. Conclusion Improvement of analytical precision and clinical sensitivity of a laboratory test is a goal for all laboratorians. This concept should be applied to cTnI and cTnT immunoassays. Careful evaluation with development of new diagnostic criteria (including 99th URL and decisional cut-oV values) will be necessary to improve patient care. More accurate and earlier diagnosis should lead to more eVective therapies. In particular, highly sensitive cTnI and cTnT assays will better define the true normal levels and the 99th URL. Moreover, the new generation of troponin assays will enable better risk stratification of patients who do not have increased troponin by current assays and may allow risk stratification of patients with chronic stable angina and patients with HF. It is really important to stress that increased cTnI and cTnT represent an index of cardiac tissue damage, even in the case of extracardiac diseases (including chronic inflammatory disease, end-stage renal disease, or treatment with powerful cardiotoxic drugs), suggesting an appropriate diagnosis and, when necessary, a specific treatment. Despite these interesting and preliminary studies, further and more comprehensive studies with highly specific assays are required to firmly establish the clinical usefulness of troponin in a wide range of diseases.

24

CLERICO ET AL.

REFERENCES [1] J.S. Alpert, K. Thygesen, E. Antman, J.P. Bassand, Myocardial infarction redefined: a consensus document of the Joint European Society of Cardiology/American College of Cardiology Committee for the redefinition of myocardial infarction, J. Am. Coll. Cardiol. 36 (2000) 959–969. [2] K. Thygesen, J.S. Alpert, H.D. White, Joint ESC/ACCF/AHA/WHF Task Force for the Redefinition of Myocardial Infarction, Universal definition of myocardial infarction, J. Am. Coll. Cardiol. 50 (2007) 2173–2195. [3] J. Trevelyan, E.W.A. Needham, S.C.H. Smith, R.K. Mattu, Impact of the recommendations for the redefinition of miocardial infarction on diagnosis and prognosis in an unselected United Kingdom cohort with suspected cardiac chest pain, Am. J. Cardiol. 93 (2004) 817–821. [4] M. Panteghini, The new definition of myocardial infarction and the impact of troponin determination on clinical practice, Int. J. Cardiol. 106 (2006) 298–306. [5] S. Vittorini, A. Clerico, Cardiovascular biomarkers: increasing impact of laboratory medicine in cardiology practice, Clin. Chem. Lab. Med. 46 (2008) 748–763. [6] F.S. Apple, R.L. Jesse, L.K. Newby, et al., National Academy of Clinical Biochemistry and IFCC Committee for Standardization of Markers of Cardiac Damage Laboratory Medicine Practice Guidelines: analytical issues for biochemical markers of acute coronary syndromes, Clin. Chem. 53 (2007) 547–551. [7] R.H. Christenson, S.H. Duh, F.S. Apple, et al., Standardization of cardiac troponin I assays: round robin of ten candidate reference materials, Clin. Chem. 47 (2001) 431–437. [8] M. Panteghini, F. Pagani, K.T.J. Yeo, et al., Evaluation of imprecision for cardiac troponin assays at low-range concentrations, Clin. Chem. 50 (2004) 327–332. [9] S. Takeda, A. Yamashita, K. Maeda, Y. Mae´da, Structure of the core domain of human cardiac troponin in the Ca(2þ)-saturated form, Nature 424 (6944) (2003) 35–41. [10] F.S. Apple, Cardiac troponin testing in renal failure and skeletal muscle disease patients, in: A.H.B. Wu (Ed.), Cardiac Markers, second ed., Homan Press, Totowa, NJ, 2003, pp. 139–147. [11] M. Panteghini, Acute coronary syndrome. Biochemical strategies in the troponin era, Chest 122 (2002) 1428–1435. [12] F.S. Apple, M.M. Murakami, The diagnostic utility of cardiac biomarkers in detecting myocardial infarction, Clin. Cornerstone 7 (Suppl. 1) (2005) S25–S30. [13] A.S. JaVe, L. Babuin, F.S. Apple, Biomarkers in acute cardiac disease. The present and the future, J. Am. Coll. Cardiol. 48 (2006) 1–11. [14] M. Panteghini, Assay-related issues in the measurement of cardiac troponins, Clin. Chim. Acta 402 (2009) 88–93. [15] M. Panteghini, D.M. Bunk, R.H. Christenson, et al., Standardization of troponin I measurements: an update, Clin. Chem. Lab. Med. 46 (2008) 1501–1506. [16] K.M. Eggers, L. Lind, P. Venge, B. Lindahl, Will the universal definition of myocardial infarction criteria result in an overdiagnosis of myocardial infarction? Am. J. Cardiol. 103 (2009) 588–591. [17] F.S. Apple, M.A.M. Muratami, Serum and plasma cardiac troponin I 99th percentile reference values for 3 2nd generation assays, Clin. Chem. 53 (2007) 1558–1560. [18] K.M. Eggers, B. Lagerquist, P. Venge, L. Wallentin, B. Lindahl, Persistent troponin I elevation in stabilized patients after an episode of acute coronary syndrome predicts longterm mortality, Circulation 107 (2007) 1907–1914. [19] S.E. Melanson, D.A. Morrow, P. Jarolim, Earlier detection of myocardial injury in a preliminary evaluation using a new troponin I assay with improved sensitivity, Am. J. Clin. Pathol. 128 (2007) 282–286.

HIGH SENSITIVITY TROPONIN ASSAY

25

[20] D. Hermsen, F. Apple, L. Garcia-Beltra`n, et al., Results from a multicenter evaluation of the 4th generation Elecsys Troponin T assay, Clin. Lab. 53 (2007) 1–9. [21] C. Prontera, A. Fortunato, S. Storti, et al., Evaluation of analytical performance of the siemens ADVIA TnI ultra immunoassay, Clin. Chem. 53 (2007) 1722–1723. [22] J. Jarausch, S. Braun, A. Dolci, et al., Evaluation of a development version of the Elecsys high sensitive troponin T assay, Clin. Chem. 54 (Suppl. 6) (2008) A91. [23] R. Beyrau, S. Braun, R. Cooray, et al., Multicentre evaluation of a high sensitive Elecsys troponin T assay, Clin. Chem. Lab. Med. 47 (2009) S128. [24] M.S. Sabatine, D.A. Morrow, J.A. de Lemos, P. Jarolim, E. Braunwald, Detection of acute changes in circulating troponin in the setting of transient stress test-induced myocardial ischaemia using an ultrasensitive assay: results from TIMI 35, Eur. Heart J. 30 (2009) 162–169. [25] P. Venge, S. James, L. Jansson, B. Lindahl, Clinical performance of two highly sensitive cardiac troponin I assays, Clin. Chem. 55 (2009) 109–116. [26] M. Redpath, G. Chalmers, C. Dibden, B. Morris, A. Blumsohn, Evaluation of imprecision of ADVIA CentaurW TNI-UltraTM automated immunoassay, Ann. Clin. Biochem. 45 (Suppl. 1) (2008) 57. [27] D. Van de Kerkhof, B. Peters, V. Scharnhorst, Performance of the Advia Centaur secondgeneration troponin assay TnI-Ultra compared with the first-generation cTnI assay, Ann. Clin. Biochem. 45 (2008) 316–317. [28] A. Clerico, A. Fortunato, C. Prontera, A. Ripoli, G.C. Zucchelli, M. Emdin, Distribution of plasma cardiac troponin-I values in healthy subjects: pathophysiological considerations, Clin. Chem. Lab. Med. 46 (2008) 804–808. [29] C. Prontera, A. Fortunato, S. Storti, et al., Evaluation of analytical performance of Advia TnI ultra immunoassay and comparison with Access AccuTnI method, Immuno-Analyse Biol. Spec. 23 (2008) 311–318. [30] J.C.J.M. Swaanenburg, P.J. Visser-VanBrummen, M.J.L. DeJongste, Tiebosch ATHM. The content and distribution of troponin I, troponin T, myoglobin, and alpha-hydrobytyric acid dehydrogenase in the human heart, Am. J. Clin. Pathol. 115 (2001) 770–777. [31] M. Emdin, S. Vittorini, C. Passino, A. Clerico, Old and new biomarkers of heart failure, Eur. J. Heart Fail. 11 (2009) 331–335. [32] Clinical and Laboratory Standards Institute, Approved guideline;Protocols for determination of limits of detection and limits of quantitation, (2004) CLSI document EP17-A. [33] C.A. Spencer, M. Takeuchi, M. Kazarosyan, F. MacKenzie, G.J. Beckett, E. Wilkinson, Interlaboratory/intermethod diVerences in functional sensitivity of immunometric assays of thyrotropin (TSH) and impact on reliability of measurement of subnormal concentrations of TSH, Clin. Chem. 41 (1995) 367–374. [34] International vocabulary of metrology, Basic and general concepts and associated terms (VIM). Document produced by the Working Group 2 of the Joint Committee for Guides in Metrology (JCGM/WG 2), JCGM, (2008) 200. [35] A.H.B. Wu, Q.A. Lu, J. Todd, J. Moecks, F. Wians, Short- and long-term biological variation in cardiac troponin I measured with a high sensitivity assay: implications for clinical practice, Clin. Chem. 55 (2009) 52–58. [36] M. Zaninotto, M. Mion, S. Altinier, M. Forni, M. Plebani, Quality specifications for biochemical markers of myocardial injury, Clin. Chim. Acta 346 (2004) 65–72. [37] J.R. Tate, W. Ferguson, R. Bais, K. Kostner, T. Marwick, A. Carter, The determination of the 99th centile level for troponin assays in an Australian reference population, Ann. Clin. Biochem. 45 (2008) 275–288. [38] P.O. Collinson, O. CliVord-Mobley, D. Gaze, F. Boa, R. Senior, Assay imprecision and 99th-percentile reference value of a high-sensitivity cardiac troponin I assay, Clin. Chem. 55 (2009) 1433–1434.

26

CLERICO ET AL.

[39] F.S. Apple, A new season for cardiac troponin assays: it’s time to keep a scorecard, Clin. Chem. 55 (2009) 1303–1306. [40] S. James, P. Armstrong, R. CaliV, et al., Troponin T levels and risk of 30-day outcomes in patients with the acute coronary syndrome: prospective verification in the GUSTO-IV trial, Am. J. Med. 115 (2003) 178–184. [41] B. Lindahl, E. Diderholm, B. Lagerqvist, P. Venge, L. Wallentin, Mechanisms behind the prognostic value of troponin T in unstable coronary artery disease: a FRISC II substudy, J. Am. Coll. Cardiol. 38 (2001) 979–986. [42] D.A. Morrow, C.P. Cannon, N. Rifai, et al., Ability of minor elevations of troponins I and T to predict benefit from an early invasive strategy in patients with unstable angina and non-ST elevation myocardial infarction: results from a randomized trial, JAMA 286 (2001) 2405–2412. [43] P. Venge, B. Lagerqvist, E. Diderholm, B. Lindahl, L. Wallentin, Clinical performance of three cardiac troponin assays in patients with unstable coronary artery disease, Am. J. Cardiol. 89 (2002) 1035–1041. [44] J.R. Tale, Troponin revisited 2008: assay performance, Clin. Chem. Lab. Med. 46 (2008) 1489–1500. [45] K.M. Eggers, A.S. JaVe, L. Lind, P. Venge, B. Lindahl, Value of cardiac troponin I cutoV concentrations below the 99th percentile for clinical decision-making, Clin. Chem. 55 (2009) 85–92. [46] A. Mingels, L. Jacobs, E. Michielsen, J. Swaanenburg, W. Wodzig, M. van Dieijen-Visser, Reference population and marathon runner sera assessed by highly sensitive cardiac troponin T and commercial cardiac troponin T and I assays, Clin. Chem. 55 (2009) 101–108. [47] Clinical and Laboratory Standards Institute, Defining, establishing, and verifying reference intervals in the clinical laboratory; Proposed guideline, third ed., Clinical and Laboratory Standards Institute, Wayne, PA, 2008 CLSI document C28-P3. [48] P.E. Hickman, T. Badrick, S.R. Wilson, D. McGill, Reporting of cardiac troponin— problems with the 99th population centile, Clin. Chim. Acta 381 (2007) 182–183. [49] Y. Chen, R.C. Serfass, S.M. Mackey-Bojack, K.L. Kelly, J.L. Titus, F.S. Apple, Cardiac troponin T alterations in myocardium and serum of rats after stressful, prolonged intense exercise, J. Appl. Physiol. 88 (2000) 1749–1755. [50] P. Anversa, A. Leri, J. Kajstura, B. Nadal-Ginard, Myocyte growth and cardiac repair, J. Mol. Cell. Cardiol. 34 (2002) 91–105. [51] R.K. Illango, Utilizing NT-proBNP in the selection of risks for life insurance, J. Insur. Med. 39 (2007) 182–191. [52] J.M. Ribera-Casado, Ageing and the cardiovascular system, Z. Gerontol. Geriatr. 32 (1999) 412–419. [53] K.G. Pugh, J.Y. Wei, Clinical implications of physiological changes in the aging heart, Drugs Aging 18 (2001) 263–276. [54] K.M. Eggers, L. Lind, H. Ahlstrom, et al., Prevalence and pathophysiological mechanisms of elevated cardiac troponin I levels in a population-based sample of elderly subjects, Eur. Heart J. 29 (2008) 2252–2258. [55] D.A. Morrow, E.M. Antman, Evaluation of high-sensitivity assays for cardiac troponin, Clin. Chem. 55 (2009) 5–9. [56] S.E. Melanson, M.J. Conrad, N. Mosammaparast, P. Jarolim, Implementation of a highly sensitive cardiac troponin I assay: test volumes, positivity rates and interpretation of results, Clin. Chim. Acta 395 (2008) 57–61. [57] P.A. Kavsak, A.R. MacRae, M.J. Yerna, A.S. JaVe, Analytical and clinical utility of a next-generation, highly sensitive cardiac troponin I assay for early detection of myocardial injury, Clin. Chem. 55 (2009) 573–577.

HIGH SENSITIVITY TROPONIN ASSAY

27

[58] A.H. Wu, Interpretation of high sensitivity cardiac troponin I results: reference to biological variability in patients who present to the emergency room with chest pain: case report series, Clin. Chim. Acta 401 (2009) 170–174. [59] H.D. White, Will new higher-precision troponins lead to clarify or confusion? Curr. Opin. Cradiol. 23 (2008) 292–295. [60] A.H. Wu, A.S. JaVe, F.S. Apple, et al., NACB Writing Group; Cannon CP, Storrow AB; NACB Committee, National Academy of Clinical Biochemistry Laboratory Medicine practice guidelines: use of cardiac troponin and B-type natriuretic peptide or N-terminal proB-type natriuretic peptide for etiologies other than acute coronary syndromes and heart failure, Clin. Chem. 53 (2007) 2086–2096. [61] A.H. Wu, A.S. JaVe, The clinical need for high-sensitivity cardiac troponin assays for acute coronary syndromes and the role for serial testing, Am. Heart J. 155 (2008) 208–214. [62] S.A. Hunt, D.W. Baker, M.H. Chin, et al., ACC/AHA guidelines for the evaluation and management of chronic heart failure in the adult: executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Revise the 1995 Guidelines for the Evaluation and Management of Heart Failure), J. Am. Coll. Cardiol. 38 (2001) 2101–2113. [63] S.A. Hunt, American College of Cardiology; American Heart Association Task Force on Practice Guidelines (Writing Committee to Update the 2001 Guidelines for the Evaluation and Management of Heart Failure), ACC/AHA 2005 guideline update for the diagnosis and management of chronic heart failure in the adult: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Update the 2001 Guidelines for the Evaluation and Management of Heart Failure), J. Am. Coll. Cardiol. 46 (2005) e1–e82. [64] S.A. Hunt, W.T. Abraham, M.H. Chin, American College of Cardiology Foundation; American Heart Association, 2009 Focused update incorporated into the ACC/AHA 2005 Guidelines for the Diagnosis and Management of Heart Failure in Adults A Report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines Developed in Collaboration With the International Society for Heart and Lung Transplantation, J. Am. Coll. Cardiol. 53 (2009) e1–e90. [65] K. Swedberg, J. Cleland, H. Dargie, et al., Task force for the diagnosis and treatment of chronic heart failure of the European Society of Cardiology. Guidelines for the diagnosis and treatment of chronic heart failure: executive summary (update 2005): the Task Force for the Diagnosis and Treatment of Chronic Heart Failure of the European Society of Cardiology, Eur. Heart J. 26 (2005) 1115–1140. [66] R. Latini, S. Masson, I.S. Anand, et al., Prognostic value of very low plasma concentrations of troponin T in patients with stable chronic heart failure, Circulation 116 (2007) 1242–1249. [67] T.W. Wallace, S.M. Abdullah, M.H. Drazner, et al., Prevalence and determinants of troponin T elevation in the general population, Circulation 113 (2006) 1958–1965. [68] D. Cardinale, M.T. Sandri, A. Colombo, et al., Prognostic value of troponin I in cardiac risk stratification of cancer patients undergoing high-dose chemotherapy, Circulation 109 (2004) 2749–2754. [69] D. Cardinale, A. Colombo, M.T. Sandri, et al., Prevention of high-dose chemotherapyinduced cardiotoxicity in high-risk patients by angiotensin-converting enzyme inhibition, Circulation 114 (2006) 2474–2481. [70] M. Maeder, T. Fehr, H. Rickli, P. Ammann, Sepsis-associated myocardial dysfunction: diagnostic and prognostic impact of cardiac troponins and natriuretic peptides, Chest 129 (2006) 1349–1366.

28

CLERICO ET AL.

[71] A. Dispenzieri, M.A. Gertz, R.A. Kyle, et al., Serum cardiac troponins and N-terminal pro-brain natriuretic peptide: a staging system for primary systemic amyloidosis, J. Clin. Oncol. 22 (2004) 3751–3757. [72] G. Merlini, G. Palladini, Amyloidosis is a cure possible? Ann. Oncol. 19 (Suppl. 4) (2008) iv63–iv66. [73] C. Passino, R. Poletti, M. Fontana, et al., Clinical relevance of non-cardiac determinants of natriuretic peptide levels, Clin. Chem. Lab. Med. 46 (2008) 1515–1523. [74] T. Pham, L. Gossec, A. Constantin, et al., Cardiovascular risk and rheumatoid arthritis: clinical practice guidelines based on published evidence and expert opinion, Joint Bone Spine 73 (2006) 379–387. [75] A. Kahan, Y. Allanore, Primary myocardial involvement in systemic sclerosis, Rheumatology 45 (Suppl. 4) (2006) 14–17. [76] N. Assous, E. Touze´, C. Meune, A. Kahan, Y. Allanore, Cardiovascular disease in rheumatoid arthritis: single-center hospital-based cohort study in France, Joint Bone Spine 74 (2007) 66–72. [77] A.Y. Wang, K.N. Lai, Use of cardiac biomarkers in end-stage renal disease, J. Am. Soc. Nephrol. 19 (2008) 1643–1652. [78] C. Sommerer, E. Giannitsis, V. Schwenger, M. Zeier, Cardiac biomarkers in haemodialysis patients: the prognostic value of amino-terminal pro-B-type natriuretic peptide and cardiac troponin T, Nephron Clin. Pract. 107 (2007) c77–c81. [79] L.H. Jacobs, J. van de Kerkhof, A.M. Mingels, et al., Haemodialysis patients longitudinally assessed by highly sensitive cardiac troponin T and commercial cardiac troponin T and cardiac troponin I assays, Ann. Clin. Biochem. 46 (2009) 283–290. [80] D.A. Marshall, B.J. O’Brien, Economic evaluation of diagnostic tests, in: P.C. Price, R.H. Christensen (Eds.), Evidence-based laboratory medicine—from principles to outcomes, AACC Press, Washington, DC, 2003, pp. 159–186. [81] D.A. Morrow, C.P. Cannon, R.L. Jesse, et al., National Academy of Clinical Biochemistry Laboratory Medicine Practice Guidelines: clinical characteristics and utilization of biochemical markers in acute coronary syndromes, Clin. Chem. 53 (2007) 552–574. [82] O. Bergmann, R.D. Bhardwaj, S. Bernard, et al., Evidence for cardiomyocyte renewal in humans, Science 324 (5923) (2009) 98–102. [83] M.H. Hessel, D.E. Atsma, E.J. van der Valk, W.H. Bax, M.J. Schalij, A. van der Laarse, Release of cardiac troponin I from viable cardiomyocytes is mediated by integrin stimulation, Pflugers Arch. 455 (2008) 979–986. [84] M.H. Hessel, E.C. Michielsen, D.E. Atsma, et al., Release kinetics of intact and degraded troponin I and T after irreversible cell damage, Exp. Mol. Pathol. 85 (2008) 90–92. [85] W. Wang, C.J. Schulze, W.L. Suarez-Pinzon, J.R. Dyck, G. Sawicki, R. Schulz, Intracellular action of matrix metalloproteinase-2 accounts for acute myocardial ischemia and reperfusion injury, Circulation 106 (2002) 1543–1549. [86] J.Y. Shyy, S. Chien, Role of integrins in cellular responses to mechanical stress and adhesion, Curr. Opin. Cell Biol. 9 (1997) 707–713. [87] M.S. Clarke, R.W. Caldwell, H. Chiao, K. Miyake, P.L. McNeil, Contraction-induced cell wounding and release of fibroblast growth factor in heart, Circ. Res. 76 (1995) 927–934. [88] D. Kaye, D. Pimental, S. Prasad, et al., Role of transiently altered sarcolemmal membrane permeability and basic fibroblast growth factor release in the hypertrophic response of adult rat ventricular myocytes to increased mechanical activity in vitro, J. Clin. Invest. 97 (1996) 281–291. [89] P.L. McNeil, R. Khakee, Disruptions of muscle fiber plasma membranes. Role in exerciseinduced damage, Am. J. Pathol. 140 (1992) 1097–1109.

HIGH SENSITIVITY TROPONIN ASSAY

29

[90] E. Page, J. Upshaw-Earley, G. Goings, Permeability of rat atrial endocardium, epicardium, and myocardium to large molecules. Stretch-dependent eVects, Circ. Res. 71 (1992) 159–173. [91] R.E. Shave, E. Dawson, G. Whyte, et al., Evidence of exercise-induced cardiac dysfunction and elevated cTnT in separate cohorts competing in an ultra-endurance mountain marathon race, Int. J. Sports Med. 23 (2002) 489–494. [92] M. Herrmann, J. Scharhag, M. Miclea, A. Urhausen, W. Herrmann, W. Kindermann, Post-race kinetics of cardiac troponin T and I and N-terminal pro-brain natriuretic peptide in marathon runners, Clin. Chem. 49 (2003) 831–834. [93] H.J. Roth, R.M. Leitha¨user, H. Doppelmayr, et al., Cardiospecificity of the 3rd generation cardiac troponin T assay during and after a 216 km ultra-endurance marathon run in Death Valley, Clin. Res. Cardiol. 96 (2007) 359–364. [94] E.B. Fortescue, A.Y. Shin, D.S. Greenes, et al., Cardiac troponin increases among runners in the Boston Marathon, Ann. Emerg. Med. 49 (2007) 137–143. [95] N. Middleton, K. George, G. Whyte, D. Gaze, P. Collinson, R. Shave, Cardiac troponin release is stimulated by endurance exercise in healthy humans, J. Am. Coll. Cardiol. 52 (2008) 1813–1814. [96] K.M. Hubble, D.M. Fatovich, J.M. Grasko, S.D. Vasikaran, Cardiac troponin increases among marathon runners in the Perth Marathon: the Troponin in Marathons (TRIM) study, Med. J. Aust. 190 (2009) 91–93. [97] N. Mousavi, A. Czarnecki, K. Kumar, et al., Relation of biomarkers and cardiac magnetic resonance imaging after marathon running, Am. J. Cardiol. 103 (2009) 1467–1472. [98] D.S. Jassal, D. MoVat, J. Krahn, et al., Cardiac injury markers in non-elite marathon runners, Int. J. Sports Med. 30 (2009) 75–79. [99] S. Vasan, Biomarkers of cardiovascular disease: molecular basis and practical considerations, Circulation 113 (2006) 2335–2362. [100] A. Clerico, S. Vittorini, C. Passino, M. Emdin, New biomarkers of heart failure, Crit. Rev. Clin. Lab. Sci. 46 (2009) 107–128. [101] A. Clerico, F.A. Recchia, C. Passino, M. Emdin, Cardiac endocrine function is an essential component of the homeostatic regulation network: physiological and clinical implications, Am. J. Physiol. Heart Circ. Physiol. 290 (2006) H17–H29. [102] A. Clerico, M. Fontana, S. Vittorini, M. Emdin, The search for a pathophysiological link between gender, cardiac endocrine function, body mass regulation and cardiac mortality: proposal for a working hypothesis, Clin. Chim. Acta 405 (2009) 1–7. [103] A. De Bold, Determinants of brain natriuretic peptide gene expression and secretion in acute cardiac allograft rejection, Curr. Opin. Cardiol. 22 (2007) 146–150. [104] S. Potluri, H.O. Ventura, M. Mulumudi, M.R. Mehra, Cardiac troponin levels in heart failure, Cardiol. Rev. 12 (2004) 21–25. [105] S.D. Wiviott, J.A. de Lemos, D.A. Morrow, Pathophysiology, prognostic significance and clinical utility of B-type natriuretic peptide in acute coronary syndromes, Clin. Chim. Acta 346 (2004) 119–128. [106] P. Venge, J. Arnlo¨v, B. Zethelius, Seemingly healthy 71-year-old men with minor elevations of cardiac troponin I and at risk of premature death in CVD have elevated levels of NT-proBNP: report from the ULSAM study, Scand. Clin. Lab. Invest. 69 (2009) 418–424. [107] J.C. Lega, Y. Lacasse, L. Lakhal, S. Provencher, Natriuretic peptides and troponins in pulmonary embolism: a meta-analysis, Thorax 64 (2009) 868–875. [108] A. Clerico, M. Emdin, Diagnostic accuracy and prognostic relevance of the measurement of the cardiac natriuretic peptides: a review, Clin. Chem. 50 (2004) 33–50.

30

CLERICO ET AL.

[109] M. Emdin, C. Passino, C. Prontera, et al., Comparison of brain natriuretic peptide (BNP) and amino-terminal proBNP for early diagnosis of heart failure, Clin. Chem. 53 (2007) 1264–1272. [110] J. Sundstro¨m, E. Ingelsson, L. Berglund, et al., Cardiac troponin-I and risk of heart failure: a community-based cohort study, Eur. Heart J. 30 (2009) 773–781. [111] G.C. Fonarow, T.B. Horwich, Combining natriuretic peptides and necrosis markers in determining prognosis in heart failure, Rev. Cardiovasc. Med. 4 (Suppl. 4) (2003) S20–S28.

ADVANCES IN CLINICAL CHEMISTRY, VOL. 49

BIOCHEMICAL BIOMARKERS OF OXIDATIVE COLLAGEN DAMAGE Y. Henrotin,*,†,1 M. Deberg,* M Mathy-Hartert,* and G. Deby-Dupont† *Bone and Cartilage Research Unit, Institute of ` ge, CHU-Sart-Tilman, Pathology, University of Lie ` 4000 Liege, Belgium † Center for Oxygen Research and Development, Institute of ` ge, 4000 Lie ` ge, Belgium Chemistry, University of Lie

1. Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. The Collagen Family: Structure and Biochemistry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. Chemical Nature and Reactivity of Oxygen. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. The Surprising Inertia of Oxygen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. The RNOS Family . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4. In Vivo Production of RNOS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5. RNOS Biological Activity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6. In Vivo Markers of Oxidant Stress. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7. Oxidant-Induced Changes in Collagens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.1. Oxidative Cleavage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2. Glycoxidation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.3. Amino Acid Modifications. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8. Biomarkers of Collagen Oxidative Damage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.1. Glycoxidation Products. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.2. Nitrated Peptides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9. Critical Comments and Concluding Remarks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

31 32 33 33 34 35 37 39 40 40 41 43 43 43 45 48 49

1. Abstract Collagens are major constituents of connective tissues in the animal kingdom. During aging and inflammatory-related diseases, the collagen network undergoes oxidation that leads to structural and biochemical alterations within 1

Corresponding author: Y. Henrotin, e-mail: [email protected] 31

0065-2423/09 $35.00 DOI: 10.1016/S0065-2423(09)49002-4

Copyright 2009, Elsevier Inc. All rights reserved.

32

HENROTIN ET AL.

the collagen molecule. Collagen oxidation appears to be a key determinant of aging and a critical physiopathologic mechanism of numerous diseases. Further, the detection of oxidized-collagen peptides seems to be a promising approach for the diagnosis and the prognosis of inflammatory diseases. This chapter reviews the structural and biochemical changes to collagen induced by reactive oxygen and nitrogen species and discusses recent data on the use of collagen-derived biomarkers for measuring oxidative damage.

2. The Collagen Family: Structure and Biochemistry Collagens are the most abundant proteins found in connective tissues. There are at least 12 types of collagen. Collagen types I, II, and III, composed of identical or different a chains, are the most abundant and form fibrils of similar structure. Each chain consists of 1050 amino acids wound around each other in a characteristic right-handed triple helix. There are three amino acids per turn of the helix and every third amino acid is a glycine residue. Collagens are also rich in proline and hydroxyproline. Lateral interactions of the triple helix result in formation of fibrils of roughly 50 nm in diameter. Collagens are synthesized as longer precursor proteins called procollagens. Type I procollagen contains additional 150 amino acids at its N terminus and 250 at its C terminus. Following synthesis, collagen fibers begin assembly in the endoplasmic reticulum (ER) and Golgi complexes. During this time, the signal sequence is removed and numerous but specific modifications take place. For example, specific proline residues are hydroxylated by the enzymes prolyl 4-hydroxylase and prolyl 3-hydroxylase and specific lysine residues also are hydroxylated by lysyl hydroxylase. Both prolyl and lysyl hydroxylases are absolutely dependent upon vitamin C as a cofactor. Glycosylations of the O-linked type also occurs during Golgi transit. Following processing, the procollagens are secreted into the extracellular space in which extracellular enzymes remove the prodomains. Groups of mature collagen fibrils then associate in regular staggered arrays and undergo cross-linking to form large fibrils and fibers [1]. Accompanying fibril formation is the oxidation of certain lysine residues by the extracellular enzyme lysyl oxidase thus forming reactive aldehydes. These reactive aldehydes form specific cross-links between chains and thereby stabilize the staggered array of the collagen fibril. Fibrillar collagens are cleaved by collagenases into two fragments: a 3 /4-length fragment (also referred TCA) and a ¼-length fragment (TCB). In addition to enzymatic cleavage, collagens undergo oxidative damage that, not only contribute to destabilize collagen network, but also increase the susceptibility of collagen fibrils to proteases in general.

33

BIOMARKERS OXIDATIVE COLLAGEN DAMAGE

After a short review of the main characteristics of oxygen (O2) metabolism in vivo, this chapter will focus on the structural and functional changes induced by reactive nitrogen and oxygen species (RNOS) in collagens. We also discuss the opportunity to use biomarkers of oxidative collagen damage for diagnosis and prognosis of disease states.

3. Chemical Nature and Reactivity of Oxygen Despite its discovery in the eighteenth century by Priestly [2], oxygen (O2) was simply considered the electron acceptor at the end of the mitochondrial electron transport until the 1950s. The existence of O2 metabolism, not only restricted to its use by mitochondria, as well as the possibility of an in vivo excessive production of reactive oxygen species (ROS) with a subsequent alteration of the redox equilibrium [3–6] started mainly with the pioneer research of Fridovich on superoxide anion (O2 ) [7], and were complicated by the discovery of nitric oxide ( NO) and the reactive nitrogen species (RNS), a subgroup of ROS including diverse nitrogen derivatives (N2O3, NO2–, ONOO–, and others). In this chapter, we shall generally use the term RNOS to refer to the entire family of reactive species derived from oxygen and nitrogen (for more details see Refs. [8–10].

3.1. THE SURPRISING INERTIA OF OXYGEN Being a triplet (two unpaired electrons in the ground state), molecular oxygen is unreactive toward organic molecules at low temperature. The reaction of dioxygen with the single state of organic substrates is spin forbidden. Consequently, the oxygenation of organic molecules, at physiological temperatures, must involve the modification of the electronic structure of one of the patterns. Therefore, according to the rules of quantic chemistry, O2 which is a biradical may not react with molecules which are not radicals. To overcome the quantic inhibition and to allow the reaction of O2 with organic molecules, O2 must gain one electron and become a monoradical, or it is the organic molecule which must become a monoradical by the loss of one electron. This transformation needs either the furniture of a considerable energy, what is unacceptable for living matter, or the use of catalysts which lower the energy barrier and allow the reaction to occur by multiple low-energy steps. In vivo, the catalysts are enzymes, the oxidases and oxygenases, which transform either O2 or the organic molecule in a monoradical. The oxidases

HENROTIN ET AL.

34

of the NOX family such as the NADPH-oxidase of neutrophils (NOX2) transfer one electron from a donor molecule to O2, which becomes the superoxide anion (O2 ). The oxygenases activate the organic molecule which becomes a free radical and binds either an atom of oxygen (monooxygenase) or O2 (dioxygenase). Cyclooxygenases and lipoxygenases are well-known examples: they activate the substrate (an unsaturated fatty acid) into a radicular state by removing H (a hydrogen atom with its electron), allowing the newly formed radical to react with O2.

3.2. THE RNOS FAMILY

The superoxide anion (O2 ) is the first element of ROS or RNOS (Fig. 1) [11]. By dismutation catalyzed by superoxide dismutase, O2 produces hydrogen peroxide (H2O2). O2 also reacts with NO generated by NO synthase (NOS), to produce peroxynitrite (ONOO), a powerful unspecific oxidant for most of the organic molecules. ONOO is unstable and decomposes into new active species, some of them being free radicals such as hydroxyl radical ( OH) and nitryl radical ( NO2 ), responsible for hydroxylation and nitration [12–14].

Other reactive species (HOCI, ferryls)

Oxidation chlorination

Hydroxylation

Peroxidases Complexed iron Mitochondria NADPH-oxidase Xanthine-oxidase

SOD O2–•

H2O2

•OH

ONOO−

Oxidation of lipids, proteins and DNA

NO•2 No-synthases

NO• Nitration

FIG. 1. Reactive oxygen species produced from the metabolism of molecular oxygen (by macrophages, neutrophils, alveolar cells, endothelial cells, etc.). O2 , superoxide anion; H2O2, hydrogen peroxide; OH, hydroxyl radical; ONOO–, peroxynitrite; NO , nitric oxide; HOCl, hypochlorous acid; NO2 , nitryl radical.

35

BIOMARKERS OXIDATIVE COLLAGEN DAMAGE

NO also reacts with hemic structures, –SH functions and FeS-bearing enzymes, inducing alterations of the mitochondrial respiratory chain enzymes and the inactivation of antioxidant enzymes such as catalase, GSH-peroxidase, and SOD [15]. Nitrite (NO2) derived from the reaction of NO with oxygen is a substrate for myeloperoxidase (MPO) activity [16]. H2O2 is more stable than the previously described RNOS and is a substrate for peroxidase. Catalase destroys it without producing reactive intermediate species but MPO uses H2O2 to produce hypochlorous acid (HOCl), by a complex way of reaction including the generation of unstable radical species [17]. HOCl is not a radical but a powerful oxidant, with a sufficient lifetime to diffuse inside and in the immediate vicinity of the cells. It may also reacts with H2O2 to produce singlet oxygen (1O2), a nonradical form of oxygen, but highly reactive, especially to induce lipoperoxidation [18, 19]. In vitro in the presence of complexed ferrous iron (Fe2þ), H2O2 generates OH by the Fenton reaction, but the in vivo production of OH by the Fenton reaction is questionable [20], and it seems more probable that in vivo OH derives from the unstable ferryl and oxoferryl species produced by the reaction of O2 with complexed iron (i.e., iron of hemoglobin). Finally, O2 contributes to prolong the RNOS lifetime, by reacting with organic radicals (R ) to forms peroxyl radicals (ROO ) which initiate lipid peroxidation. Almost all the RNOS are oxidant species, except O2 . Its important role in the RNOS cascade is not linked to its oxidant properties, but to its position of intermediate between fundamental oxygen and the organic molecules of the living matter.

4. In Vivo Production of RNOS The ‘‘redox’’ status of a biological medium can be defined as a regulated equilibrium between oxidants and reducing compounds, and in normal conditions, the intracellular medium seems to be reductor with an excess of small molecular weight reductor compounds (thiols such as glutathione, NADH, NADPH, ascorbate, urate, glucose, etc.), antioxidant enzymes and proteins (thioredoxins, glutaredoxins, etc.). An excessive production of RNOS disturbs the normal intracellular equilibrium, and can lead to an ‘‘oxidant stress’’ [21]. However, it must be kept in mind that ‘‘oxidant stress’’ is an imprecise expression which does not take into account the level, the duration, and the cellular or tissular extension of the redox imbalance. In vivo, the production of RNOS results from the activity of specialized enzymes in numerous cells, but with variable levels of RNOS production. Fibroblasts, endothelial cells, or chondrocytes, for example, produce low

HENROTIN ET AL.

36

levels of RNOS if compared to the phagocytes (monocytes/macrophages and PMNs) which need high levels of RNOS for their phagocyte function. The enzymes involved in the intracellular production of RNOS are mostly hemic enzymes, including: (A) the NOX family, of which the more famous is the phagocyte membrane enzymatic complex NADPH-oxidase (NOX 2). It converts O2 into O2 by a monoelectronic reduction of O2 [22]. The enzyme is normally quiescent, but is quickly activated (assembly of the protein subunits) by soluble and receptor binding mediators (e.g., IL-1b, TNFa, endotoxins, etc.). NOX enzymes are present in many other cell types, such as lymphocytes, fibroblasts, endothelial cells, myocytes, and chondrocytes. In these cells, NOX enzymes normally produce moderate levels of ROS, which act as regulators of cell responses such as cell growth [22–24]. (B) the NOS which produce NO from O2 and L-arginine, in presence of NADPH and biopterin as cofactor. In particular conditions linked to substrate and cofactor concentrations, NOS can produce O2 simultaneously with NO, increasing the risk of in situ generation of ONOO [25]. At least three isoforms of NOS have been identified. Endothelial cells and neuronal cells contain a constitutive enzyme (eNOs and nNOS, respectively), considered as a ‘‘housekeeping’’ enzyme, producing physiological levels of NO needed for vascular tone regulation or neurotransmission. In inflammation, an inducible NOS (iNOS) is triggered in many cells (macrophages, neutrophils, endothelial cells, smooth muscle cells, hepatocytes, chondrocytes, etc.) considerably increasing the production of NO [26–28]. An NOS synthase is also present in mitochondria [29]. (C) the MPO, a specific enzyme of neutrophil azurophilic granules, which uses H2O2 and Cl (or other halide ions) to form HOCl [18, 19], but also reacts with nitrite ions and ONOO and, by this way, is responsible for tyrosyl residue nitration in peptides and proteins [16, 17, 30, 31]. The presence of MPO in blood is considered as a marker of PMNs activation [32, 33], but the enzyme is present at lower levels in monocytes [34] and seems inducible by an oxidative stress in other cells such as endothelial cells. (D) the mitochondria enzymes. In the mitochondrion, there is a weak but regular production of O2 , which is can be formed by an electron ‘‘leak’’ from the electron transport chain, at the level of the NADPH deshydrogenase, the ubiquinone/cytochrome c intersection, and the cytochrome oxidase [35, 36]. It appears that 2% of the O2 used by the mitochondria is transformed into O2 and its derived product, H2O2 [14]. Mitochondria also contain NOS, suggesting a possible intramitochondrial production of ONOO [37, 38]. (E) the xanthine oxidase, a nonhemic cytosolic enzyme converting xanthine or hypoxanthine into uric acid. Xanthine oxidase derives from xanthine dehydrogenase by proteolytic activation (by a Ca2þ-dependent calpain) during hypoxia. It uses O2 as an electron acceptor, forming O2 and H2O2.

BIOMARKERS OXIDATIVE COLLAGEN DAMAGE

37

(F) the ‘‘radical’’ enzymes, which generate RNOS during their enzymatic activity. In this group are included intracellular enzymes (lipoxygenases, PGHsynthases, cyclooxygenases), peroxysomic enzymes (glycolate oxidase, D-amino acid oxidase, urate oxidase, ‘‘fatty acyl Co-A’’ oxidase, etc.), and even some DNA methylating enzymes and enzymes involved in the synthesis of hormones and neurotransmitters [39].

5. RNOS Biological Activity When they are produced in controlled conditions, RNOS play a normal physiological role, mainly as temporarily and local intermediates necessary for many enzymatic activities, and as active agents in the fight of phagocytes against microorganisms. The role of RNOS becomes pathological when, escaping to physiological controls, RNOS-generating enzymes produced abnormal amounts of RNOS, either acutely at high levels or chronically at lower levels. RNOS play a role in intracellular signal transduction, in the maintenance of vascular tone, in neuronal transmission, in host defence and inflammation, and in the pathological conditions of anoxia or hyperoxia. In signal transduction, modest modifications of the redox equilibrium by a local increase of RNOS triggers a cascade of responses. By this way, RNOS participate to the control of basal cellular functions including proliferation, migration, and extracellular matrix remodeling [40–44]. The mechanism of action of RNOS on cell signaling pathways would implicate a direct action on cellular messengers (–SH modifications, protein oxidation, and nitration) or a reduction of the intracellular level of antioxidant molecules (GSH, NADH, NADPH, etc.) [45]. A fine control of RNOS in signal transduction seems to be exerted by intracellular antioxidants: catalase, superoxide dismutase, peroxyredoxins, peptidyl methionine sulfoxyde reductase, glutathione regulating enzymes, glutaredoxin, thioredoxins, or iron transport regulating proteins (transferrine, ferritine, etc.) [46]. In normal physiological conditions, NO is produced at low levels (in the range of picomoles) in endothelial cells [26] and is important for the regulation of vascular tone. NO also modulates neuronal function. It is probably implicated in the regulation of excitability and firing, in depression and memory processes, by modulating neurotransmitter release, such as acetylcholine, catecholamines, excitatory and inhibitory amino acids, serotonin, histamine, and adenosine [47], by acting on the cell receptors (nitrosylation and even a toxic effect of nitration and oxidation). RNOS produced at high level by activated PMNs intervene in the host defence against pathogens [48], but an acute or a prolonged chronic inflammatory response may result in an excessive release of RNOS by phagocytes, an subsequently in host tissue damages. The deleterious role of RNOS has

HENROTIN ET AL.

38

been reported, for example, in asthma, cirrhosis, atherosclerosis, cancers, AIDS, and rheumatoid arthritis (RA) [3, 49, 50]. During a local inflammatory reaction, RNOS are mainly produced by macrophages and neutrophils, in response to endotoxin, cytokines, or products of tissue degradation. RNOS can clearly degrade hyaluronic acid, modify collagen and proteoglycan structure and/or synthesis, alter and interact with immunoglobulins, activate enzymes and inactivate their inhibitors, and possibly participate in chemotaxis [8, 9]. Ischemia occurs in many diseases (heart diseases, cerebrovascular attacks, artery occlusion disease, etc.) with variable structural and functional impacts according the sensitivity of the organ to O2. Brain, kidney, and heart are highly sensitive to O2 depletion; cartilage displays a metabolism adapted to hypoxic conditions, however, ischemic conditions seems to increase RNOS production in the joint. It has been demonstrated in cell cultures that the RNOS production starts in the anoxia period, and increases at reoxygenation: free radicals (superoxide anion and lipidic radicals) were monitored by electron spin resonance (ESR) [51]. RNOS production participates to tissue and cell alterations with edema leading to ‘‘no-reflow’’ phenomenon [52] (Fig. 2). The most oxygen-sensitive cells are those with an active metabolism highly dependent from aerobic respiration for the continuous production of ATP

RNOS Ischemia/reperfusion (I/R) Anaerobic metabolites

RNOS Cell damage (mitochondria)

Cytokines Increase of adhesion Receptors

Cell death by necrosis

Cell death by apoptosis

Neutrophils

Inflammation Amplification of alterations FIG. 2. Cell metabolism changes associated with ischemia/reperfusion. RNOS, reactive nitrogen and oxygen species.

39

BIOMARKERS OXIDATIVE COLLAGEN DAMAGE

indispensable to their metabolic functions. During the ischemic phase, the mitochondria are rapidly altered [53], the enzymatic complexes of the electron transport chain are damaged and use O2 to produce O2 , mainly at the level of the ubiquinone/cytochrome bc1 (complex III). The proton gradient falls in with an arrest of the ATP synthetase and ATP production, followed by a rapid fall of adenylated nucleotides. In consequence, the function of the plasmic membrane and ER ionic pumps are disturbed, leading to an increase of cytosolic Ca2þ and activation of Ca2þ-dependent enzymes (phospholipases, serine proteases). The proteases convert xanthine dehydrogenase to xanthine oxidase, which forms O2 and H2O2. The manganese SOD of mitochondria is able to neutralize O2 , but the mitochondrial NOS, of which the activity is enhanced by anoxia-reoxygenation, releases NO able to inhibit the mitochondrial respiration and to react with O2 to form ONOO– [54]. Anoxia also stimulates extramitochondrial enzymes to produce RNOS: NOS, PGHsynthases, lipoxygenases, and NOX enzymes, especially in endothelial cells [44]. However, a moderate period of anoxia may trigger a defense response to a second hit of anoxia, a phenomenon which is described as a preconditioning reaction 55], a field of intense research. Hyperoxia is a clinical situation encountered in the treatment of intensive care patients with acute respiratory failure. Like hypoxia, hyperoxia has noxious effects correlated with an RNOS production with effects similar to those described for anoxia: mitochondrial dysfunction, DNA damages, peroxidation of lipidic membranes, protein oxidation, stimulation of the nuclear transcription factors, inflammation, and cell death by apoptosis or necrosis [56]. The role of mitochondria is essential, by releasing cytochrome c and proapoptotic factors, two factors which stimulate the caspase cascade. But, as for hypoxia, low doses of RNOS act as a preconditioning treatment, and adapt the cells to a further hit of hyperoxia by induction of protective enzymes and proteins [57, 58].

6. In Vivo Markers of Oxidant Stress Free radicals can be directly measured by ESR, most often coupled to spin trapping to increase the sensibility of the method. However, ESR is difficult to use in vivo in human [59, 60]. Many indirect methods of RNOS measurements have been proposed, based on the use of antioxidants and enzyme inhibitors or on the measurement of stable compounds derived from RNOS activity and considered as ‘‘markers of oxidant stress.’’ Isoprostanes [61], hydroxynonenal and lipid peroxides [62], nitrated and oxidized proteins [63, 64], chlorinated compounds, carbonylated proteins [65], oxidized glutathione, and malondialdehyde (MDA) (detected as thiobarbituric acid reactants;

HENROTIN ET AL.

40

TBARs) [66] are measured as markers of oxidant stress, but the measurements are not all specific for oxidative stress, and their measurements are at risk of artifacts. For example, isoprostanes can be produced by platelets independently of oxidant stress [67, 68], and the chemical reaction of MDA detection is influenced by the presence of iron in the sample [68, 69]. Moreover, these markers are not specific of one particular RNOS, and are not specific for the tissue in which oxidative stress occurs. All RNOS do not generate one particular by-product. Therefore, the measurement of more than one marker is needed to assess ‘‘oxidative stress,’’ and should be measured ideally in the tissue or fluids of the organ where the oxidative stress has occurred: in bronchoalveolar lavage (BAL) fluid for lung [70] or in synovial fluid for the joint.

7. Oxidant-Induced Changes in Collagens 7.1. OXIDATIVE CLEAVAGE

Collagen is the only protein susceptible to fragmentation by O2 [71]. In comparison, proteins such as serum albumin or various enzymes are not degraded by O2 [72]. This O2 -induced collagen degradation was characterized by the release of small 4-hydroproline-containing peptides, suggesting scissions in the triple-helical part of the collagen molecule [73, 74]. This collagen oxidative degradation was inhibited by SOD but not by catalase or chelating agents such deferroxamine or diethylene triamine penta-acetate (DTPA), confirming the key role played by O2 in the process. Nevertheless, the action of OH on collagen remains questionable since it was demonstrated that its action is quite different in the absence or in the presence of O2 [75]. In the presence of O2, OH generated by gamma radiolysis released pattern of peptides different from that generated by O2 . OH-generated peptides are characterized by an increase of aspartic and glutamic acid residues and a decrease in the amount or 4-hydroproline and proline residues. In contrast, when irradiations of collagen are performed in the absence of O2, no collagen cleavage is observed, but a polymerization of collagen. Hypochlorite (HOCL/OCl–), within the predicted range generated by PMNs or monocytes at sites of inflammation (10–50 M), does not cause fragmentation of collagen I or II [76, 77]. Only the supraphysiological concentrations of 1–5 mM cause extensive fragmentation of collagen [78]. N-chloramine (5–50 M) does not cause fragmentation but, greatly increases the degradation of collagen by collagenase and elastase. The mechanism by which N-chloramines, and probably other oxidants, increase the proteolytic susceptibility of collagen is not clearly determined, although it is assumed

BIOMARKERS OXIDATIVE COLLAGEN DAMAGE

41

that N-chloramines react with amino groups and disrupt the secondary and tertiary structure of collagen molecules [77]. Disruption of the tertiary structure of collagen by oxidation exposes hydrophobic regions and promotes the degradation of fibrillar collagens by proteases. Another explanation would be that oxidation and disruption of pyridinoline cross-links could results in the loss of functional interactions of collagen fibrils and consequently an increase in the susceptibility of collagen to proteolytic degradation. Pyridinoline cross-links are derived from hydroxylated lysine residues located within the collagen telopeptides. They are markers of collagen degradation. Finally, exposure of proline peptides to a Fenton system (Cu(II)/peroxide) results in conversion of some proline residues into hydroxyproline, along with formation of g-aminobutyric acid [79] and exposure of purified type II collagen to FeSO4-EDTA ( OH source) or xanthine oxidase-hypoxanthine system (O2 source) induced a cleavage of the proline producing more terminal glutamate residues, which are Ca2þ affinity ligand. This oxidativeinduced change promotes crystal formation which is an important feature in some rheumatic diseases including OA, gut, or Kaschi-Beck’s disease [80].

7.2. GLYCOXIDATION The covalent binding of the aldehyde group of glucose to a free amino group of a protein (mainly lysine amino group) is called glycation. The initial reaction is the formation of Schiff base followed by a spontaneous Amadori rearrangement. Both the Schiff base and the Amadori products may subsequently undergo oxidation, particularly in the presence of metal ions (i.e., iron and copper ions) and RNOS (mainly OH) [81, 82], and generate carbonyls and RNOS that lead to an extensive modification of the protein [83–85]. This type of reaction has been referred to as glycoxidation. The stable end products of these reactions, mainly carboxymethyl lysine (CML) and pentosidine, are known collectively as advanced glycation end products (AGEs) [86]. CML can be formed in two pathways: oxidation of fructoselysine [84], and reaction of protein with glycoxal, which is an autoxidation product of glucose or Schiff base adduct [87, 88]. Pentosidine is a cross-link between lysine and arginine residues resulting from the glycoxidation of Amadori products or the reaction of arabinose, which is an autooxidative product of glucose [87]. Collagen is a protein with slow turnover, and so is highly susceptible to glycoxidation in vivo. This is one reason for which collagen glycoxidation was commonly used as biological markers of disease progression, complication occurrence, or severity. Glycoxidation endproducts accumulation in cartilage is particularly well documented. Pentosidine levels are three- to fourfold higher in cartilage collagen than in skin collagen [89, 90], probably as a consequence of the low turnover of type

42

HENROTIN ET AL.

II collagen. The levels of pentosidine and CML have been shown to progressively accumulate with age in skin and cartilage collagens [91]. In contrast, in the age range 50–95 years, pentosidine concentration in bone trabecular collagens did not depend on age [92]. After maturity (>20 years), pentosidine and CML accumulate linearly with age in human articular cartilage collagens [90]. After maturity, their concentrations in cartilage collagen increase 33-fold and 27-fold with age, respectively. The rate of pentosidine accumulation in collagen is 0.32 nmol g 1 protein year 1 or 0.64 mol pentosidine mol 1 Arg year 1, or 1.40 mol pentosidine mol 1 Lys year 1. In healthy adult cartilage, more than 80% of the pentosidine is present in collagen (84.4% at age 70) and less than 10% in aggrecan (9.0 % at age 70) [93]. The glycation of collagen and accumulation of AGE contribute to the pathogenic mechanisms of many age-related diseases including atherosclerosis, diabetes mellitus, and osteoarthritis (OA). Glycation induced major changes in the physicochemical properties of collagen including stiffness, mechanical strength, temperature instability and insolubility, and high resistance to collagenase digestion. Further, glycated products are potentially active in the pathogenesis of some diseases. In bone, increased pentosidine concentration is associated with a decline in mechanical properties of bone [92, 94, 95], indicating that glycoxidation end-products accumulation contributes to increase the propensity of individuals to microfracture. AGE elicit a wide range of cell-mediated responses leading to cartilage degradation, vascular dysfunction, and atherosclerosis. Among these effects, glycoxidized collagen has been reported to increase the oxidation of unsaturated lipids and to lead to the formation of reactive products including MDA or 4-hydroxynonenal. The adhesion of neutrophils to in vitro-glycoxidized collagen I and IV is significantly increased when compared with native collagens. Glycoxidized type I collagen increases the chemotactic properties of neutrophils without significant stimulatory effect on respiratory burst, whereas preincubation of neutrophils with glycoxidized type I collagen induced a priming on subsequent stimulation by N-formyl-methionyl-leucylphenylalanine. In contrast, glycoxidation of type IV collagen suppresses its inhibitory effect on further neutrophils stimulation or migration. These findings suggest that glycosidized type IV and I collagens act synergically to increase extravascular migration and inappropriate activation of neutrophils [96]. The deleterious effects of glycoxidized products are mediated through receptors. At least four different AGE receptors have been characterized, of which two are scavenger receptors [97]. One AGE receptor, RAGE, which has a wide tissue distribution, when interacting with AGE, results in the generation of cellular oxidant stress manifested by the appearance of malonedialdehyde and activation of NF-B [98]. AGE may also induce activation of DAG protein kinase C (PKC) pathway [99].

BIOMARKERS OXIDATIVE COLLAGEN DAMAGE

43

7.3. AMINO ACID MODIFICATIONS In some inflammatory diseases, particularly in RA, HOCl/OCl– mainly produced by activated neutrophils plays an important role in tissue destruction. HOCl/OCl mediates some major type II collagen modifications including desamination with consecutive carbonyl group formation and transformation of tyrosines residues to dichlorotyrosine, fragmentation in small peptides (<10 kDa) and decrease of radius aggregates [78]. These modifications were observed at low concentrations (less than 1 mM) suggesting that they can occur in pathological synovial fluids. Interestingly, pyridinoline cross-links, which determine the stability of the fibrillar collagen network, are also potential target of HOCl/OCl. HOCl/OCl reacts with the pyridinium ring structure and disrupts the ring structure, probably secondary to the formation of aldehydes [76]. Although both collagen type I and II are oxidized by HOCl/OCl, the reactions differ. Oxidized type II collagen shows greater amount of carbonyl formation as compared with oxidized type I collagen. An additional novel mechanism particularly relevant to inflammationrelated diseases involves the nonenzymatic nitrite (NEN) modification of connective tissue proteins. Protein nitration is mediated by peroxynitrite, mostly by the derived nitryl radical (NO2 ), at the level of aromatic amino acids (e.g., tyrosine). Recently, it was reported that nitration of fibrillar type I collagen inhibits the ability of primary adult cardiac fibroblasts to remodel type I collagen gels and reduces the deformability of type I collagen gels subjected to mechanical testing [100]. This latter finding correlates with the degree of cross-linking. Beside glycation, collagen nitration probably contributes to the alterations in the biomechanical properties of collagen-containing tissues consistent with age-related functional decline observed in human disease.

8. Biomarkers of Collagen Oxidative Damage 8.1. GLYCOXIDATION PRODUCTS In 1997, Horie et al. [101] demonstrated by immunohistochemistry methods that CML and pentosidine accumulated in the expanded mesangial matrix and thickened glomerular capillary walls of early diabetic nephropathy and in nodular lesions and arterial walls of advanced diabetic nephropathy, but were absent in control kidneys. No association was observed for pentosidine or CML and type III, IV, V, or VI collagens, suggesting that glycoxidation process was not associated with a specific collagen type. In contrast, an immunohistochemical colocalization of glycoxidation products and MDA was observed, assuming that local oxidative stress contributes to

44

HENROTIN ET AL.

CML and pentosidine in situ generation and that they could be markers of both protein glycation and oxidation. Subsequently, it was also shown that glycoxidation rate determined by pentosidine in skin collagen was inversely related to maximum life span in several mammalian species [102]. Further, in mice of the C57BL/6NNIA strain, CML and pentosidine formation rates (dCML or pentosidine/dt) are predictors of the individual longevity, but to a lesser extent than furosine, a glycation indicator [103]. In the diabetes control and complication trial (DCCT) including 216 type 1 diabetes participants, skin collagen levels of pentosidine and CML were found to correlate with duration of diabetes and with the severity of the microvascular complication (retinopathy, nephropathy, and neuropathy) [104–106]. Further, long-term (5 years) intensive control of glycemia was able to significantly decrease skin collagen glycation/glycoxidation products (including pentosidine and CML levels) [107]. Pentosidine and CML levels were negatively correlated with acid- and pepsin-collagen solubility, indicating that glycoxidation is involved in physicochemical changes of collagens in diabetes. They are also correlated with mean HbA1c levels up to biopsy, mean HbA1c over the past year, and HbA1c at time of skin biopsy, but the correlations were weaker than those of collagen glycation (furosine). These latter findings suggest that glycoxidation products in skin provide a biological index of prior glycemic status over a considerable period of time. These differences between glycation and glycoxidation products may be explained by the fact that glycoxidation is modulated by tissues levels of metal ions, ROS, and antioxidants. Analysis of age-adjusted levels of CML and pentosidine in diabetic skin collagen suggests that the increase in the AGE in diabetes can be explained by the increase of glycemia (substrate stress) without invoking an increase in oxidative stress [108]. The predictive value of glycation and AGE skin levels of the diabetes complication has been determined in a cohort of 211 diabetes volunteers included in the DCCT in 1992 who continued to be followed in the Epidemiology of Diabetes and Complication study for 10 years [109]. Using the 1999 DCCT skin biopsy data, CML, pentosidine, and furosine were predictors of the risk of progression of retinopathy at 4 years and nephropathy at 7–8 years. More precisely, for each 1 standard deviation (SD) increase of furosine and CML, the odd ratio (OR) for progression of nephropathy were 2.03 (95% CI 1.38–2.38, P ¼ 0.0003) and 1.81 (1.24–2.64, P ¼ 0.002), respectively, and the OR for progression of retinopathy were 3.52 (95% CI 2.32–5.34, P < 0.0001) and 2.45 (1.68–3.57, P < 0.0001), respectively. Further, CML and furosine, are stronger predictors of the future risk of complication than HbA1c. Their predictive effects are independent of the preceding HbA1c levels and those later present during progression of these complications.

BIOMARKERS OXIDATIVE COLLAGEN DAMAGE

45

Articular cartilage derives its mechanical properties from its extracellular matrix. This matrix is composed of type II collagen, which forms a threedimensional network that provides the cartilage with resistance to tensile forces. In cartilage, type II collagen turnover is very slow (half-life of type II collagen 117 years), allowing AGE accumulation. Some different authors have observed that pentosidine levels increase 50-fold from age 20–80 years [90, 110, 111]. Recently, it was reported that AGE accumulation level in cartilage predisposes to the development of OA in dog with anterior cruciate ligament transection [112]. Considering that one of the first characteristics of OA is damage to the collagen network, we can suggest the possibility that glycoxidized type II collagen peptides could be relevant biomarkers of early cartilage degradation. 8.2. NITRATED PEPTIDES Usually, OA progression is monitored by measurement of changes in joint space width on plain X-ray with a graduated magnifying lens or with a computer after digitization of the radiograph. This must be considered a rather indirect measure of cartilage integrity, as articular cartilage itself is invisible on the radiographs and thus has to be assessed indirectly from the spacing between the subchondral bone ends of the joint. Furthermore, joint space width does not allow detection of early structural damage and remains difficult to use in daily practice. Magnetic resonance imaging (MRI) is a promising noninvasive tool for evaluation of cartilage, but access to this technique is confined and very expansive. Further, MRI, such as other standard X-ray, gives a static picture of the cartilage lesion, but fails to explore the metabolic changes occurring in OA cartilage. Alternatively, biochemical factors of bone synovium or cartilage turnover have been proposed as potential tools for diagnostic, prognostic, and monitoring treatment efficacy (Table 1) [113–116]. Type II collagen is possibly the ideal marker for studying cartilage remodeling. First, this collagen is relatively specific to articular cartilage, although it is also present in the vitreous humor of the eye, the nucleus pulposus of vertebral discs, and the meniscus [117]. Second, it is the most abundant protein in cartilage, representing 25% of the wet weight, 50% of the dry weight, and 90–95% of the total collagen content. Recently, we have published a method for the assessment of oxidative damage of type II collagen in cartilage and biological fluids (serum, urine, and synovial fluid) [63, 118, 119]. This original approach is based on the detection in biological fluids or in tissue of a nitrated peptide release from type II collagen during proteolytic and/or oxidative cartilage degradation. Our strategy was based on the following points: (1) type II collagen is specific for cartilage and is the most abundant collagen in the extracellular matrix, (2)

HENROTIN ET AL.

46

TABLE 1 THE PRINCIPAL COLLAGEN-DERIVED BIOMARKERS Synthesis

Breakdown

Bone

N-and-C-terminal propeptides of collagen I (PICP and PINP)

N- and C-terminal telopeptides of collagen type I (NTX-I and CTX-I )

Cartilage

N- and C-propeptides of collagen type II (PIICP, PIIAN and, PIIBNP)

Deoxypyridinoline Pyridinoline C-terminal telopeptide of collagen type II (CTX-II, Coll2-2) Collagenase-generated cleavage epitopes of collagen type II (C2C,C1,2C ) Denaturation epitopes localized to the triple-helical domain (Coll 2-1, helix-II) Nitrated peptides (Coll2-1, NO2, Coll 2-2 NO2) Synovium Liver

N-terminal propeptide of collagen type III (PIIINP)

CTX-I and NTX-I Type IV collagen 7S domain

peptide nitration results from the reaction of aromatic acid with –ONOO–, (3) in pathological circumstances, chondrocytes, but also synovium cells (mainly macrophages) produced high levels of NO and O2 , (4) type II collagen contains two tyrosine residues, but not other aromatic amino acid, one located in the triple helix (Coll2-1) and the other in the telopeptide of the C-terminal end (Coll2-2). We have then developed specific immunoassays, one for the peptide 108HRGYPGLDG116 (Coll2-1) and the other for the nitrated form of this peptide 108HRGY(NO2)PGLDG116 (Coll2-1NO2) (Fig. 3). This strategy allows the calculation of the ratio Coll2-1NO2/ Coll2-1 which reflects the oxidative-related damage of the triple-helical area of the molecule. These immunoassays have been validated in animal models and in human clinical surveys, with the following conclusions:

– In mice, Coll2-1 levels are significantly higher in the serum of biglycan/ fibromodulin double deficient mice (DKO), which develop premature and severe knee OA compared to wild-type mice (WT). In contrast, Coll2-1NO2 displayed cyclic variations with a DKO/WT ratio 1.60 at day 49 and 95 but ¼ 0.86 at intermediate time points (day 81 and 141). In both genotypes, immunostainings with Coll2-1 and Coll2-1NO2 labeled some fibroblasts in tendons and menisci as well as the chondrocytes above the tidemark in articular cartilage whereas chondrocytes in the growth plate remained unstained. For the two biomarkers, extracellular staining was limited to

BIOMARKERS OXIDATIVE COLLAGEN DAMAGE

Telopeptide OH

47

Telopeptide OH OH

OH OH

OH OH

OH OH

HRGYGLRG (Coll 2–1) NO2

OH OH

OH OH

OH OH

OH OH

LQYMRA (Coll 2–2)

NO2

FIG. 3. Localization of the epitopes Coll2-1 and Coll2-2 in type II collagen.

fibrocartilage areas in tendons and menisci from both genotypes and to some of the focal lesions of the biglycan/fibromodulin-deficient cartilage. No extracellular staining was observed in WT cartilage. The different DKO/ WT ratios observed with Coll2-1 and Coll2-1NO2 suggest that these serum biomarkers give complementary rather than redundant information on type II collagen catabolism [120]. – In horses, plasmatic levels of Coll2-1 were significantly lower in animals suffering from osteochondrosis, a disease characterized by a trouble of the endochondral ossification with further damage to the cartilage, compared to healthy horses. In contrast, Coll2-1NO2 levels were significantly higher, indicating an uncoupling variation of these two markers. This study demonstrated that the markers were able to differentiate the group of horses suffering from osteochondrosis from the group of healthy ones [121]. – In healthy humans, the concentration of Coll2-1 in serum remains stable over life. In contrast, Coll2-1NO2 concentration is more elevated in young subjects (26–30 years old). These two markers were not affected by circadian rhythm [119]. Further, Coll2-1 and Coll2-1NO2 concentration were elevated in serum of patients with primary knee OA and early RA compared to healthy subjects. Interestingly, Coll2-1NO2/Coll2-1 ratio was higher in RA (0.25) than in OA (0.15) suggesting that Coll2-1 nitration is related to synovium inflammation. Further, Coll2-1NO2, but not Coll2-1 was significantly correlated with C-reactive protein in serum of arthritic patients. These findings suggest that Coll2-1 NO2 could be a promising marker of arthritic disease activity and could advantageously replace other systemic inflammatory biomarkers to monitor specifically joint inflammation and more particularly oxidative-related cartilage degradation. – These markers have also been measured in urine of a cohort of 75 patients with knee OA corresponding to the placebo arm of a large 3-year, randomized, double-blind study comparing the efficacy of glucosamine

48

HENROTIN ET AL.

sulfate with placebo [122]. In this study, we have observed that elevated urinary levels of Coll2-1 and Coll2-1NO2 at baseline were associated with a higher disability evaluated with the Western Ontario and McMaster University Osteoarthritis Index (WOMAC). This finding indicates that Coll2-1 and Coll2-1NO2 in urinary levels reflect the clinical severity of OA disease. Another objective of our study was to investigate whether our biochemical markers could predict the radiological severity progression of OA. Interestingly, at baseline, significant correlation were found between the first year change of these markers in urine and the 3-year change of the medial joint space width, suggesting that the increase of these peptides over 1 year could be predictive of the radiological progression of knee OA [118].

9. Critical Comments and Concluding Remarks Glycoxidation markers are potential markers of glycemic-and/or oxidantrelated collagen network damage in diabetes and some aged-related diseases including OA. The two main glycoxidation products investigated are pentosidine and CML. However, these markers are associated with the etiology of a variety of different diseases (tumor, diabetes, etc.) and their measurement in serum fails to reflect changes in one particular organ. For this reason, they are relevant as prognostic markers of disease progression or complication occurrence and severity and as markers of efficacy of intervention, but their use for the diagnosis or burden of disease is questionable. To investigate trouble in one particular tissue, these markers need to be measured in biopsies and require invasive taking out methods. Another limitation on the use of pentosidine or CML as disease markers is that protein glycoxidation results of a two subsequent mechanisms, the nonenzymatic glycation and the oxidation of glycated products by ROS. These mechanisms are not necessarily coupled and may occur individually. This finding suggests that an increase of the glycoxidation products may result of either an increase of glycation or an increase of ROS production. The increase of glycoxidation in one particular tissue may reflect a local disorders and/or systemic trouble affecting the antioxidant status or the glucose level. Increased glycation may result of an increased glucose uptake or a high rate of glycolysis and seems to be dependent of the protein turnover rate. In parallel, accumulation of oxidated glycation product may results of a decrease of the antioxidant defences or an increase of ROS. Serum proteinbound CML is likely to represent a mixture of oxidative and carbonyl stress stemming in part from oxidized Amadori compounds. Cellular sources of these metabolites include peroxynitrites and hydroxyl radicals originating

BIOMARKERS OXIDATIVE COLLAGEN DAMAGE

49

from Haber–Weiss cycle as well as glyoxal and glycoaldehyde originating from food and digestion. Therefore, these markers are not tissue specific, not disease specific, and not the result of different pathogenic mechanisms. There is a need to develop cell-based or specific matrix-bound AGE to improve the clinical relevance of these potential biomarkers. The methods used to measure these markers are also questionable. Actually, pentosidine must be assayed by high-performance liquid chromatography (HPLC) because pentosidine enzyme-linked immunosorbent assays (ELISA) suffer from low sensitivity. These techniques need a previous acid extraction to release pentosidine which can generate artifact. CML is most precisely assayed by gas chromatography/mass spectrometry (GC/MS) [123] or liquid chromatography/mass spectrometry (LC/MS) [124]. However, either is somewhat inconvenient for large numbers of samples. Various investigators have developed CML poly- and monoclonal ELISA using N-epsilon-(carboxymethyl)lysine as immunogen [125]. Many of these assays are likely to suffer from a lack of linearity, especially at lower CML concentrations, and perform poorly with intact proteins. Novel technologies are needed to determine glycoxidation products in tissues. One alternative would be to produce antibodies which recognize tissue-specific molecule modified by the oxidative process. This approach has been developed by our laboratory which set up new immunoassays for nitrated type II collagen. In conclusion, to develop biochemical for measuring oxidative-related collagen damage is an important challenge for the next decade. They will allow a better understanding of the pathology affecting connective tissues and will be helpful tools for monitoring the efficiency of antioxidants on the natural evolution of these diseases. REFERENCES [1] K. Gelse, E. Poschl, T. Aigner, Collagens—structure, function, and biosynthesis, Adv. Drug Deliv. Rev. 55 (12) (2003) 1531–1546. [2] R.J. Foltz, C.A. Piantadosi, J.D. Crapo, Oxygen toxicity, in: R.G. Crystal, J.B. West, P.J. Barnes, E.R. Weibel (Eds.), The Lung: Scientific Foundations, Lippincott-Raven, Philadelphia, 1997, pp. 2713–2722. [3] J.A. Knight, Free radicals: their history and current status in aging and disease, Ann. Clin. Lab. Sci. 28 (6) (1998) 331–346. [4] G. Deby-Dupont, C. Deby, M. Lamy, Oxygen therapy in intensive care patients: a vital poison? in: J.L. Vincent (Ed.), Yearbook of Intensive Care and Emergency Medicine, Springer Verlag, Berlin, 1999, pp. 417–432. [5] S. Cuzzocrea, et al., Antioxidant therapy: a new pharmacological approach in shock, inflammation, and ischemia/reperfusion injury, Pharmacol. Rev. 53 (1) (2001) 135–159. [6] W. Droge, Free radicals in the physiological control of cell function, Physiol. Rev. 82 (1) (2002) 47–95.

50

HENROTIN ET AL.

[7] J.M. McCord, I. Fridovich, Superoxide dismutase. An enzymic function for erythrocuprein (hemocuprein), J. Biol. Chem. 244 (22) (1969) 6049–6055. [8] Y. Henrotin, B. Kurz, T. Aigner, Oxygen and reactive oxygen species in cartilage degradation: friends or foes? Osteoarthr. Cartil. 13 (8) (2005) 643–654. [9] Y.E. Henrotin, P. Bruckner, J.P. Pujol, The role of reactive oxygen species in homeostasis and degradation of cartilage, Osteoarthr. Cartil. 11 (10) (2003) 747–755. [10] D.B Zorov, et al., Reactive oxygen and nitrogen species: friends or foes? Biochemistry (Mosc) 70 (2) (2005) 215–221. [11] C. Deby, R. Goutier, New perspectives on the biochemistry of superoxide anion and the efficiency of superoxide dismutases, Biochem. Pharmacol. 39 (3) (1990) 399–405. [12] J.S. Beckman, et al., Apparent hydroxyl radical production by peroxynitrite: implications for endothelial injury from nitric oxide and superoxide, Proc. Natl. Acad. Sci. USA 87 (4) (1990) 1620–1624. [13] J.S. Beckman, Oxidative damage and tyrosine nitration from peroxynitrite, Chem. Res. Toxicol. 9 (5) (1996) 836–844. [14] A. Mouithys-Mickalad, et al., Peroxynitrite reacts with biological nitrogen-containing cyclic molecules by a radical pathway, as demonstrated by ultraweak luminescence coupled with ESR technique, Biochem. Biophys. Res. Commun. 259 (2) (1999) 460–464. [15] J.R. Lancaster Jr., K. Xie, Tumors face NO problems? Cancer Res. 66 (13) (2006) 6459–6462. [16] J.B. Sampson, et al., Myeloperoxidase and horseradish peroxidase catalyze tyrosine nitration in proteins from nitrite and hydrogen peroxide, Arch. Biochem. Biophys. 356 (2) (1998) 207–213. [17] J.P. Eiserich, R.P. Patel, V.B. O’Donnell, Pathophysiology of nitric oxide and related species: free radical reactions and modification of biomolecules, Mol. Aspects Med. 19 (4–5) (1998) 221–357. [18] G. Deby-Dupont, C. Deby, M. Lamy, Neutrophil myeloperoxidase: effector of defense and host damage, in: J.L. Vincent (Ed.), Yearbook of Intensive Care and Emergency Medicine, Springer Verlag, Berlin, 1998, pp. 75–89. [19] G. Deby-Dupont, C. Deby, M. Lamy, Neutrophil myeloperoxidase revisited: its role in health and disease, Intensivmed (1999) 500–5013. [20] W.H. Koppenol, The Haber–Weiss cycle—70 years later, Redox Rep. 6 (4) (2001) 229–234. [21] D.P. Jones, Redefining oxidative stress, Antioxid. Redox Signal. 8 (9–10) (2006) 1865–1879. [22] B.M. Babior, NADPH oxidase: an update, Blood 93 (5) (1999) 1464–1476. [23] V.J. Thannickal, B.L. Fanburg, Reactive oxygen species in cell signaling, Am. J. Physiol. Lung Cell. Mol. Physiol. 279 (6) (2000) L1005–L1028. [24] K.K. Griendling, D. Sorescu, M. Ushio-Fukai, NAD(P)H oxidase: role in cardiovascular biology and disease, Circ. Res. 86 (5) (2000) 494–501. [25] S. Milstien, Z. Katusic, Oxidation of tetrahydrobiopterin by peroxynitrite: implications for vascular endothelial function, Biochem. Biophys. Res. Commun. 263 (3) (1999) 681–684. [26] D.J. Stuehr, Mammalian nitric oxide synthases, Biochim. Biophys. Acta 1411 (2–3) (1999) 217–230. [27] S. Baldus, et al., Heparins increase endothelial nitric oxide bioavailability by liberating vessel-immobilized myeloperoxidase, Circulation 113 (15) (2006) 1871–1878. [28] Y.E. Henrotin, et al., Nitric oxide downregulates interleukin 1beta (IL-1beta) stimulated IL-6, IL-8, and prostaglandin E2 production by human chondrocytes, J. Rheumatol. 25 (8) (1998) 1595–1601. [29] P. Ghafourifar, E. Cadenas, Mitochondrial nitric oxide synthase, Trends Pharmacol. Sci. 26 (4) (2005) 190–195.

BIOMARKERS OXIDATIVE COLLAGEN DAMAGE

51

[30] S.L. Hazen, et al., Formation of nitric oxide-derived oxidants by myeloperoxidase in monocytes: pathways for monocyte-mediated protein nitration and lipid peroxidation in vivo, Circ. Res. 85 (10) (1999) 950–958. [31] C.J. van Dalen, et al., Nitrite as a substrate and inhibitor of myeloperoxidase. Implications for nitration and hypochlorous acid production at sites of inflammation, J. Biol. Chem. 275 (16) (2000) 11638–11644. [32] D. Lau, S. Baldus, Myeloperoxidase and its contributory role in inflammatory vascular disease, Pharmacol. Ther. 111 (1) (2006) 16–26. [33] M.L. Brennan, S.L. Hazen, Emerging role of myeloperoxidase and oxidant stress markers in cardiovascular risk assessment, Curr. Opin. Lipidol. 14 (4) (2003) 353–359. [34] M. Mathy-Hartert, et al., Bactericidal activity against Pseudomonas aeruginosa is acquired by cultured human monocyte-derived macrophages after uptake of myeloperoxidase, Experientia 52 (2) (1996) 167–174. [35] J.F. Turrens, Superoxide production by the mitochondrial respiratory chain, Biosci. Rep. 17 (1) (1997) 3–8. [36] M.L. Genova, et al., The site of production of superoxide radical in mitochondrial Complex I is not a bound ubisemiquinone but presumably iron-sulfur cluster N2, FEBS Lett. 505 (3) (2001) 364–368. [37] J.P. Bolanos, et al., Nitric oxide-mediated mitochondrial damage in the brain: mechanisms and implications for neurodegenerative diseases, J. Neurochem. 68 (6) (1997) 2227–2240. [38] T. Koeck, et al., Rapid and selective oxygen-regulated protein tyrosine denitration and nitration in mitochondria, J. Biol. Chem. 279 (26) (2004) 27257–27262. [39] B.F. McAdam, et al., Effect of regulated expression of human cyclooxygenase isoforms on eicosanoid and isoeicosanoid production in inflammation, J. Clin. Invest. 105 (10) (2000) 1473–1482. [40] D.M. Hockenbery, et al., Bcl-2 functions in an antioxidant pathway to prevent apoptosis, Cell 75 (2) (1993) 241–251. [41] Y.J. Suzuki, H.J. Forman, A. Sevanian, Oxidants as stimulators of signal transduction, Free Radic. Biol. Med. 22 (1–2) (1997) 269–285. [42] J. Chandra, A. Samali, S. Orrenius, Triggering and modulation of apoptosis by oxidative stress, Free Radic. Biol. Med. 29 (3–4) (2000) 323–333. [43] J.F. Curtin, M. Donovan, T.G. Cotter, Regulation and measurement of oxidative stress in apoptosis, J. Immunol. Methods 265 (1–2) (2002) 49–72. [44] M. Ushio-Fukai, Redox signaling in angiogenesis: role of NADPH-oxidase, Cardiovasc. Res. 71 (2006) 226–235. [45] J. Cai, D.P. Jones, Superoxide in apoptosis. Mitochondrial generation triggered by cytochrome c loss, J. Biol. Chem. 273 (19) (1998) 11401–11404. [46] E.S. Arner, A. Holmgren, Physiological functions of thioredoxin and thioredoxin reductase, Eur. J. Biochem. 267 (20) (2000) 6102–6109. [47] A. Smolenski, et al., Functional analysis of cGMP-dependent protein kinases I and II as mediators of NO/cGMP effects, Naunyn Schmiedebergs Arch. Pharmacol. 358 (1) (1998) 134–139. [48] E. Linares, et al., Role of peroxynitrite in macrophage microbicidal mechanisms in vivo revealed by protein nitration and hydroxylation, Free Radic. Biol. Med. 30 (11) (2001) 1234–1242. [49] S. Magder, Reactive oxygen species: toxic molecules or spark of life? Crit. Care 10 (1) (2006) 208. [50] E. Crimi, et al., The role of oxidative stress in adult critical care, Free Radic. Biol. Med. 40 (3) (2006) 398–406.

52

HENROTIN ET AL.

[51] A. Mouithys-Mickalad, et al., Oxygen consumption and electron spin resonance studies of free radical production by alveolar cells exposed to anoxia: inhibiting effects of the antibiotic ceftazidime, Redox Rep. 7 (2) (2002) 85–94. [52] S.H. Rezkalla, R.A. Kloner, No-reflow phenomenon, Circulation 105 (5) (2002) 656–662. [53] S.S. Sheu, D. Nauduri, M.W. Anders, Targeting antioxidants to mitochondria: a new therapeutic direction, Biochim. Biophys. Acta 1762 (2) (2006) 256–265. [54] G.A. Walford, et al., Hypoxia potentiates nitric oxide-mediated apoptosis in endothelial cells via peroxynitrite-induced activation of mitochondria-dependent and-independent pathways, J. Biol. Chem. 279 (6) (2004) 4425–4432. [55] D.K. Das, R.M. Engelman, N. Maulik, Oxygen free radical signaling in ischemic preconditioning, Ann. N. Y. Acad. Sci. 874 (1999) 49–65. [56] A.P. Autor, A.C. Bonham, R.L. Thies, Toxicity of oxygen radicals in cultured pulmonary endothelial cells, J. Toxicol. Environ. Health 13 (2–3) (1984) 387–395. [57] A. Ahmad, et al., Endothelial Akt activation by hyperoxia: role in cell survival, Free Radic. Biol. Med. 40 (7) (2006) 1108–1118. [58] J.D. Crapo, et al., Structural and biochemical changes in rat lungs occurring during exposures to lethal and adaptive doses of oxygen, Am. Rev. Respir. Dis. 122 (1) (1980) 123–143. [59] L. Valgimigli, G.F. Pedulli, M. Paolini, Measurement of oxidative stress by EPR radicalprobe technique, Free Radic. Biol. Med. 31 (6) (2001) 708–716. [60] D.A. Svistunenko, et al., Mitochondrial dysfunction in patients with severe sepsis: an EPR interrogation of individual respiratory chain components, Biochim. Biophys. Acta 1757 (4) (2006) 262–272. [61] D. Pratico, et al., Formation of 8-iso-prostaglandin F2 alpha by human platelets, Agents Actions Suppl. 45 (1995) 27–31. [62] A Ben Baouali, et al., Plasma lipid peroxidation in critically ill patients: importance of mechanical ventilation, Free Radic. Biol. Med. 16 (2) (1994) 223–227. [63] Y. Henrotin, et al., Type II collagen peptides for measuring cartilage degradation, Biorheology 41 (3–4) (2004) 543–547. [64] M. Mathy-Hartert, et al., Nitrated proteins in bronchoalveolar lavage fluid of patients at risk of ventilator-associated bronchopneumonia, Eur. Respir. J. 16 (2) (2000) 296–301. [65] H. Buss, et al., Protein carbonyl measurement by a sensitive ELISA method, Free Radic. Biol. Med. 23 (3) (1997) 361–366. [66] I.S. Young, E.R. Trimble, Measurement of malondialdehyde in plasma by high performance liquid chromatography with fluorimetric detection, Ann. Clin. Biochem. 28 (Pt. 5) (1991) 504–508. [67] J.D. Morrow, L.J. Roberts II, The isoprostanes. Current knowledge and directions for future research, Biochem. Pharmacol. 51 (1) (1996) 1–9. [68] H.H. Draper, M. Hadley, Malondialdehyde determination as index of lipid peroxidation, Methods Enzymol. 186 (1990) 421–431. [69] H. Esterbauer, K.H. Cheeseman, Determination of aldehydic lipid peroxidation products: malonaldehyde and 4-hydroxynonenal, Methods Enzymol. 186 (1990) 407–421. [70] I. Rahman, F. Kelly, Biomarkers in breath condensate: a promising new non-invasive technique in free radical research, Free Radic. Res. 37 (12) (2003) 1253–1266. [71] J.C. Monboisse, J.P. Borel, Oxidative damage to collagen, Exs 62 (1992) 323–327. [72] H. Schuessler, K. Schilling, Oxygen effect in the radiolysis of proteins. Part 2. Bovine serum albumin, Int. J. Radiat. Biol. Relat. Stud. Phys. Chem. Med. 45 (3) (1984) 267–281. [73] J.C. Monboisse, et al., Effect of oxy radicals on several types of collagen, Int. J. Tissue React. 6 (5) (1984) 385–390.

BIOMARKERS OXIDATIVE COLLAGEN DAMAGE

53

[74] J.C. Monboisse, et al., Non-enzymatic degradation of acid-soluble calf skin collagen by superoxide ion: protective effect of flavonoids, Biochem. Pharmacol. 32 (1) (1983) 53–58. [75] V. Monboisse, et al., Nonisotopic evaluation of collagen in fibroblasts cultures, Anal. Biochem. 176 (2) (1989) 395–399. [76] K.M. Daumer, A.U. Khan, M.J. Steinbeck, Chlorination of pyridinium compounds. Possible role of hypochlorite, N-chloramines, and chlorine in the oxidation of pyridinoline cross-links of articular cartilage collagen type II during acute inflammation, J. Biol. Chem. 275 (44) (2000) 34681–34692. [77] J.M. Davies, D.A. Horwitz, K.J. Davies, Potential roles of hypochlorous acid and N-chloroamines in collagen breakdown by phagocytic cells in synovitis, Free Radic. Biol. Med. 15 (6) (1993) 637–643. [78] S. Olszowski, et al., Collagen type II modification by hypochlorite, Acta Biochim. Pol. 50 (2) (2003) 471–479. [79] K. Uchida, Y. Kato, S. Kawakishi, A novel mechanism for oxidative cleavage of prolyl peptides induced by the hydroxyl radical, Biochem. Biophys. Res. Commun. 169 (1) (1990) 265–271. [80] K. Wang, et al., Free radicals-induced abnormal chondrocytes, matrix and mineralization. A new concept of Kaschin-Beck’s disease, Chin. Med. J. (Engl) 104 (4) (1991) 307–312. [81] J.R. Requena, et al., Lipoxidation products as biomarkers of oxidative damage to proteins during lipid peroxidation reactions, Nephrol. Dial. Transplant. 11 (Suppl. 5) (1996) 48–53. [82] K.N. Sulochana, et al., Glycation and glycoxidation studies in vitro on isolated human vitreous collagen, Med. Sci. Monit. 9 (6) (2003) BR220–BR224. [83] S.P. Wolff, R.T. Dean, Glucose autoxidation and protein modification. The potential role of ‘autoxidative glycosylation’ in diabetes, Biochem. J. 245 (1) (1987) 243–250. [84] M.U. Ahmed, S.R. Thorpe, J.W. Baynes, Identification of N epsilon-carboxymethyllysine as a degradation product of fructoselysine in glycated protein, J. Biol. Chem. 261 (11) (1986) 4889–4894. [85] K.V. Chace, R. Carubelli, R.E. Nordquist, The role of nonenzymatic glycosylation, transition metals, and free radicals in the formation of collagen aggregates, Arch. Biochem. Biophys. 288 (2) (1991) 473–480. [86] G.B. Sajithlal, P. Chithra, G. Chandrakasan, An in vitro study on the role of metal catalyzed oxidation in glycation and crosslinking of collagen, Mol. Cell. Biochem. 194 (1–2) (1999) 257–263. [87] M.C. Wells-Knecht, S.R. Thorpe, J.W. Baynes, Pathways of formation of glycoxidation products during glycation of collagen, Biochemistry 34 (46) (1995) 15134–15141. [88] M.A. Glomb, V.M. Monnier, Mechanism of protein modification by glyoxal and glycolaldehyde, reactive intermediates of the Maillard reaction, J. Biol. Chem. 270 (17) (1995) 10017–10026. [89] D.R. Sell, V.M. Monnier, Isolation, purification and partial characterization of novel fluorophores from aging human insoluble collagen-rich tissue, Connect. Tissue Res. 19 (1) (1989) 77–92. [90] N. Verzijl, et al., Age-related accumulation of Maillard reaction products in human articular cartilage collagen, Biochem. J. 350 (Pt. 2) (2000) 381–387. [91] D.R. Sell, et al., Pentosidine: a molecular marker for the cumulative damage to proteins in diabetes, aging, and uremia, Diabetes Metab. Rev. 7 (4) (1991) 239–251. [92] C.J. Hernandez, et al., Trabecular microfracture and the influence of pyridinium and nonenzymatic glycation-mediated collagen cross-links, Bone 37 (6) (2005) 825–832. [93] N. Verzijl, et al., Age-related accumulation of the advanced glycation endproduct pentosidine in human articular cartilage aggrecan: the use of pentosidine levels as a quantitative measure of protein turnover, Matrix Biol. 20 (7) (2001) 409–417.

54

HENROTIN ET AL.

[94] X. Wang, et al., Age-related changes in the collagen network and toughness of bone, Bone 31 (1) (2002) 1–7. [95] S. Viguet-Carrin, et al., Contribution of the advanced glycation end product pentosidine and of maturation of type I collagen to compressive biomechanical properties of human lumbar vertebrae, Bone (2006). [96] J.C. Monboisse, et al., In vitro glycoxidation alters the interactions between collagens and human polymorphonuclear leucocytes, Biochem. J. 350 (Pt. 3) (2000) 777–783. [97] S. Horiuchi, et al., Extra- and intracellular localization of advanced glycation endproducts in human atherosclerotic lesions, Nephrol. Dial. Transplant. 11 (Suppl. 5) (1996) 81–86. [98] H.M. Lander, et al., Activation of the receptor for advanced glycation end products triggers a p21(ras)-dependent mitogen-activated protein kinase pathway regulated by oxidant stress, J. Biol. Chem. 272 (28) (1997) 17810–17814. [99] Y. Kaji, T. Oshika, [Role of advanced glycation end products and activation of PKC in diabetic retinopathy], Nippon Rinsho 63 (Suppl. 6) (2005) 188–193. [100] D.C. Paik, et al., Nitrite-induced cross-linking alters remodeling and mechanical properties of collagenous engineered tissues, Connect. Tissue Res. 47 (3) (2006) 163–176. [101] K. Horie, et al., Immunohistochemical colocalization of glycoxidation products and lipid peroxidation products in diabetic renal glomerular lesions. Implication for glycoxidative stress in the pathogenesis of diabetic nephropathy, J. Clin. Invest. 100 (12) (1997) 2995–3004. [102] D.R. Sell, et al., Longevity and the genetic determination of collagen glycoxidation kinetics in mammalian senescence, Proc. Natl. Acad. Sci. USA 93 (1) (1996) 485–490. [103] D.R. Sell, N.R. Kleinman, V.M. Monnier, Longitudinal determination of skin collagen glycation and glycoxidation rates predicts early death in C57BL/6NNIA mice, FASEB J. 14 (1) (2000) 145–156. [104] V.M. Monnier, et al., Relation between complications of type I diabetes mellitus and collagen-linked fluorescence, N. Engl. J. Med. 314 (7) (1986) 403–408. [105] D.R. Sell, et al., Pentosidine formation in skin correlates with severity of complications in individuals with long-standing IDDM, Diabetes 41 (10) (1992) 1286–1292. [106] D.R. McCance, et al., Long-term glycemic control and neurological function in IDDM patients, Diabetes Care 16 (9) (1993) 1291–1293. [107] V.M. Monnier, et al., Skin collagen glycation, glycoxidation, and crosslinking are lower in subjects with long-term intensive versus conventional therapy of type 1 diabetes: relevance of glycated collagen products versus HbA1c as markers of diabetic complications, Diabetes 48 (4) (1999) 870–880. DCCT Skin Collagen Ancillary Study Group. Diabetes Control and Complications Trial. [108] D.G. Dyer, et al., Accumulation of Maillard reaction products in skin collagen in diabetes and aging, J. Clin. Invest. 91 (6) (1993) 2463–2469. [109] S. Genuth, et al., Glycation and carboxymethyllysine levels in skin collagen predict the risk of future 10-year progression of diabetic retinopathy and nephropathy in the diabetes control and complications trial and epidemiology of diabetes interventions and complications participants with type 1 diabetes, Diabetes 54 (11) (2005) 3103–3111. [110] A. Maroudas, G. Palla, E. Gilav, Racemization of aspartic acid in human articular cartilage, Connect. Tissue Res. 28 (3) (1992) 161–169. [111] R.A. Bank, et al., Ageing and zonal variation in post-translational modification of collagen in normal human articular cartilage. The age-related increase in non-enzymatic glycation affects biomechanical properties of cartilage, Biochem. J. 330 (Pt. 1) (1998) 345–351. [112] J. DeGroot, et al., Accumulation of advanced glycation end products as a molecular mechanism for aging as a risk factor in osteoarthritis, Arthritis Rheum. 50 (4) (2004) 1207–1215.

BIOMARKERS OXIDATIVE COLLAGEN DAMAGE

55

[113] D.C. Bauer, et al., Classification of osteoarthritis biomarkers: a proposed approach, Osteoarthr. Cartil. 14 (8) (2006) 723–727. [114] P. Garnero, Biochemical markers in osteoarthritis: will they measure up? Nat. Clin. Pract. Rheumatol. 2 (3) (2006) 116–117. [115] P. Garnero, Use of biochemical markers to study and follow patients with osteoarthritis, Curr. Rheumatol. Rep. 8 (1) (2006) 37–44. [116] V.B. Kraus, Do biochemical markers have a role in osteoarthritis diagnosis and treatment? Best Pract. Res. Clin. Rheumatol. 20 (1) (2006) 69–80. [117] K.A. Elsaid, C.O. Chichester, Review: collagen markers in early arthritic diseases, Clin. Chim. Acta 365 (1–2) (2006) 68–77. [118] M.A. Deberg, et al., One-year increase of Coll 2–1, a new marker of type II collagen degradation, in urine is highly predictive of radiological OA progression, Osteoarthr. Cartil. 13 (12) (2005) 1059–1065. [119] M. Deberg, et al., New serum biochemical markers (Coll 2–1 and Coll 2–1 NO2) for studying oxidative-related type II collagen network degradation in patients with osteoarthritis and rheumatoid arthritis, Osteoarthr. Cartil. 13 (3) (2005) 258–265. [120] Y. Henrotin, et al., Dietary supplementation with chondroitin sulfate and glucosamineHCL does not slow down articular cartilage erosion in biglycan/fibromodulin deficient mice, Osteoarthr. Cartil. 12 (Suppl. B) (2004) S33. [121] M. Gangl, et al., A type II-collagen derived peptide and its nitrated form as new markers of inflammation and cartilage degradation in equine osteochondral lesions, Res. Vet. Sci. (2006). [122] J.Y. Reginster, et al., Long-term effects of glucosamine sulphate on osteoarthritis progression: a randomised, placebo-controlled clinical trial, Lancet 357 (9252) (2001) 251–256. [123] J.N. Shaw, J.W. Baynes, S.R. Thorpe, N epsilon-(carboxymethyl)lysine (CML) as a biomarker of oxidative stress in long-lived tissue proteins, Methods Mol. Biol. 186 (2002) 129–137. [124] T. Teerlink, et al., Measurement of N epsilon-(carboxymethyl)lysine and N epsilon-(carboxyethyl)lysine in human plasma protein by stable-isotope-dilution tandem mass spectrometry, Clin. Chem. 50 (7) (2004) 1222–1228. [125] S. Reddy, et al., N epsilon-(carboxymethyl)lysine is a dominant advanced glycation end product (AGE) antigen in tissue proteins, Biochemistry 34 (34) (1995) 10872–10878.

This page intentionally left blank

ADVANCES IN CLINICAL CHEMISTRY, VOL. 49

BIOCHEMICAL BASIS OF FABRY DISEASE WITH EMPHASIS ON MITOCHONDRIAL FUNCTION AND PROTEIN TRAFFICKING A.M. Das*,1 and H.Y. Naim† *Department of Pediatrics, Hannover Medical School, Hannover, Germany † Department of Physiological Chemistry, University of Veterinary Medicine, Hannover, Germany

1. 2. 3. 4. 5. 6.

Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Inheritance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Clinical Picture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Diagnosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.1. Symptomatic Treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.2. Enzyme Replacement Therapy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.3. Substrate Reduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.4. Chaperones. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7. Biochemical Basis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.1. Enzyme Deficiency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2. Mutations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.3. Pathology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.4. Secondary Biochemical EVects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

57 58 58 59 59 60 60 60 60 61 62 62 63 63 63 68

1. Abstract Fabry disease, also known as Anderson–Fabry disease, is an X-linked lysosomal storage disorder. The clinical picture is highly variable and usually milder in females. It is a multisystemic disease involving many organs. Fabry disease is due to a deficiency of alpha-galactosidase A caused by 1

Corresponding author: A.M. Das, e-mail: [email protected] 57

0065-2423/09 $35.00 DOI: 10.1016/S0065-2423(09)49003-6

Copyright 2009, Elsevier Inc. All rights reserved.

58

DAS AND NAIM

diVerent usually ‘‘private’’ mutations. Enzyme replacement therapy (ERT) has been established, other therapeutic options are at an experimental stage. Classically, mechanical deposition of storage material in blood vessels was believed to lead to decreased blood supply with consecutive organ dysfunction. Recently, however, many secondary biochemical processes have been discussed to be involved in the pathogenesis of Fabry disease. For example, compromised energy metabolism has been found both in vitro and in vivo, altered lipid composition of membranes can lead to abnormalities in traYcking and sorting of rafts-associated proteins. We discuss the role of these secondary phenomena in the pathogenesis of Fabry disease. 2. Introduction The clinical picture of Fabry disease (OMIM 301500)—also known as Anderson disease in the Angloamerican literature—has been independently described by the German dermatologist Johannes Fabry and his British colleague William Anderson back in 1898 [1, 2]. Pathogenesis of the disease was then completely unknown. The discovery of lysosomes by Christian de Duve and coworkers about 50 years ago [3] paved the way for defining Fabry disease as a lysosomal storage disease. In 1963, Ge´ry Hers demonstrated deficiency of acid alpha-glucosidase in Pompe disease [4], another lysosomal storage disease. Subsequently, other diseases were characterized as lysosomal storage diseases, in Fabry disease deficiency of alpha-galactosidase A (AGAL) was demonstrated [5, 6]. It was Kornreich who deciphered the genetic code of AGAL in 1989 [7]. Although the deficient enzyme is known, the precise pathogenetic mechanism leading to clinical symptoms in Fabry disease is not completely understood. Classically, it was assumed that mechanical deposition of storage compounds leads to compromised vessel lumina restricting blood flow (ischemia). However, this hypothesis has been questioned. We believe that biophysical and biochemical phenomena play a role, too. In this manuscript, we will discuss putative mechanisms leading to clinical symptoms in Fabry disease. Two diVerent enzyme preparations, agalsidase alpha and agalsidase beta, have been developed for enzyme replacement therapy (ERT), substrate depletion and chaperones are more experimental therapeutic options which will be discussed. 3. Inheritance Fabry disease is inherited as an X-linked trait. Hemizygous males, therefore, frequently show the full-blown clinical picture while the disease in females is usually milder. Disease severity in females can vary enormously

BIOCHEMICAL BASIS OF FABRY DISEASE

59

due to lyonization [8]. The defective gene encoding AGAL is located on the long arm of the X-chromosome (Xq22.1). More than 200 mutations have been described so far, many of them are so-called ‘‘private’’ mutations specific for one family [9]. Fabry disease is a panethnic disease, its prevalence has been estimated to be 1:40,000 to 1:100,000, however, due to the unspecific clinical features of Fabry disease the figures probably underestimate the true incidence [9].

4. Clinical Picture Fabry disease may aVect many organs, it is a multisystemic disease. In the male hemizygous patient clinical onset usually occurs in childhood to adolescence. Often cutaneous vascular lesions (so-called angiokeratoma) are the first presenting sign, accompanied by severe pain in the extremities (acroparesthesias) and hypohidrosis. Pain is triggered by fever/infection or high ambient temperatures. Opacities in the eye can also be the presenting sign (‘‘cornea verticillata’’). Later on kidney dysfunction (often leading to renal failure requiring dialysis/transplantation), hearing loss, gastrointestinal symptoms, cardiomyopathy, cerebral symptoms like stroke or psychiatric disease, gastrointestinal symptoms, hypothyroidism, obstructive pulmonary disease, musculoskeletal disease, and endocrinological dysfunction may occur. Some of these symptoms may lead to premature death, in average 20 years earlier compared to the general population [10]. In female heterozygous patients, the clinical course is highly variable due to lyonization and ranges from asymptomatic to severe disease as observed in male patients. On average, in females clinical symptoms set in 10 years later compared to males and may lead to premature death, in average 15 years earlier than in the general population [11]. Therapeutic intervention probably leads to a modification of the clinical course; long-term observations are needed to assess the impact of causal therapy on the clinical course.

5. Diagnosis If Fabry disease is clinically suspected, the diagnosis has to be confirmed by biological tests. A selective screening test is available measuring elevated Gb3-levels in urine or blood, however, this method is not reliable. Another test relies on the determination of alpha-galactosidase activity in dried blood spots [12]; however, enzyme activity is often normal in females or males with atypically mild forms of disease. Genetic testing is more reliable, most of the reported mutations are ‘‘private,’’ that is, confined to one specific

60

DAS AND NAIM

pedigree. In many patients, the diagnosis of Fabry disease is through selective testing based on an index case in the family. Prenatal diagnosis is possible from chorionic villus sampling or amniocentesis either using enzymatic testing and/or mutation analysis, the latter is more reliable. Newborn mass screening in the first days of life has been advocated [13–15] using diVerent analytic methods (Gb3, enzyme activity, mutation analysis), however, it is still not clear whether presymptomatic treatment of children with Fabry disease is able to improve the outcome and thus would justify presymptomatic screening (and therapy) in the general population (ethical and financial concerns).

6. Treatment 6.1. SYMPTOMATIC TREATMENT Pain relief may be achieved by drugs like diphenylhydantoin and carbamazepine. If renal failure occurs, dialysis and possibly renal transplantation are life-saving interventions. In cardiac arrhythmias, a pacemaker may be necessary, but in some cases is not able to prevent severe complications. If gastrointestinal symptoms exist, replacement of pancreatic enzymes may be helpful. 6.2. ENZYME REPLACEMENT THERAPY Two enzyme preparations of the defective AGAL are commercially available, Agalsidase alpha (ReplagalW, Shire) and beta (FabrazymeW, Genzyme). Agalsidase alpha is biotechnologically produced from human fibroblasts while Agalsidase beta is produced in Chinese hamster ovary cells. Both preparations are eYcient and lead to a stabilization or improvement of the disease [16]. The enzyme has to be infused every other week. Antibody formation may occur. This therapeutic modality in most cases allows reduction of symptomatic therapy, clinical improvement is often reported by the patient. The enzymes are not able to cross the blood–brain barrier, therefore neurological symptoms caused by brain dysfunction cannot be influenced. 6.3. SUBSTRATE REDUCTION A less well-evaluated therapeutic option is substrate reduction which blocks sphingolipid synthesis and will presumably attenuate storage phenomena [17–19].

BIOCHEMICAL BASIS OF FABRY DISEASE

61

The balance between synthesis and catabolism of lipid storage compounds in Fabry disease may be redressed by inhibiting the synthesis of storage material. N-butyldeoxynojirimycin (OGT 918, miglustat, ZavescaW, Actelion) inhibits glucosylceramide-synthetase (Fig. 1). This substance is commercially available for the treatment of two other lysosomal storage diseases (M. Gaucher, M. Niemann-Pick type C). This small molecule is able to cross the blood–brain barrier and thus can influence intracerebral manifestations. Further controlled studies are necessary to assess clinical benefit of this new therapeutic option in Fabry disease. 6.4. CHAPERONES Miglustat may have—in addition to substrate reduction—a stabilizing eVect on the dysfunction of the mutated enzyme (chaperone function). Other substances have similar eVects increasing residual enzyme activity [20].

Gal

Gal

Ceramide

Glc

Globotriaosylceramide a-galactosidase A

Gal

Gal

Galactose

Glc

Ceramide

Lactosylceramide

Substrate reduction therapy Dihydrosphingosine

N-acetyldihydrosphingosine

Ceramide

Globotriaosylceramide

Miglustat Glucosylceramide

Lactosylceramide

FIG. 1. Enzyme defect in Fabry disease and principle of substrate reduction therapy. Deficiency of alpha-galactosidase A leads to the accumulation of globotriaosylceramide while substrate reduction therapy results in decreased production of globotriaosylceramide.

62

DAS AND NAIM

7. Biochemical Basis 7.1. ENZYME DEFICIENCY Fabry disease is caused by deficiency of the lysosomal enzyme AGAL. This enzyme is responsible for the degradation of globotriaosylceramide (Gb3) (Fig. 1). Several classes of enzyme deficiency have been described. There are classical hemizygotes with no detectable AGAL-activity and absence of enzyme protein, classic hemizygotes with no enzyme activity but detectable enzyme protein and atypical hemizygotes with residual enzyme activity [9]. Absence of enzyme protein was immunologically shown using rabbit-anti human AGAL antibodies [21–24] and the term cross-reacting immunologic material (CRIM)-negative was coined for these cases. If enzyme protein is present but enzyme activity is markedly compromised traYcking of the enzyme may be disturbed. By confocal laser microscopy, we have recently shown reduced amounts of labeled AGAL in the lysosome assessed by colocalization with LAMP II as a lysosomal marker (Fig. 2). Km-variants of AGAL have been described in atypical hemizygotes [25]. In a recent study, globotriaosylsphingosine which accumulates both in plasma and tissue of aVected patients has been shown to inhibit AGAL and to promote proliferation of vascular smooth muscle cells [26]. This mechanism may play a vital role in the pathogenesis of clinical symptoms though the molecular mechanism has not been elucidated yet.

WT (>95% colocal.)

Patient (85% colocal.)

FIG. 2. Confocal microscopy of AGAL and LAMP II. AGAL is marked in green, LAMP II as a lysosomal marker in red, yellow indicates colocalization.

BIOCHEMICAL BASIS OF FABRY DISEASE

63

7.2. MUTATIONS As many as 200 diVerent mutations have been described so far, most are so-called ‘‘private’’ mutations (i.e., confined to a single pedigree). Common mutations are N251S, R227Q, R227X, and R342Q. Genotype–phenotype correlation is poor. 7.3. PATHOLOGY Deficiency of alpha-galactosidase leads to the accumulation of glycosphingolipids, predominantly globotriaosylceramide (Gb3, also referred to as ceramidetrihexoside), to a lesser extent galabiosylceramide and blood group B and P glycosphingolipids [9]. It is classically believed that deposition of glycosphingolipids in vascular endothelial and smooth muscle cells leads to a compromised blood supply. Subsequent ischemia/anoxia is then responsible for organ dysfunction. However, glycosphingolipid deposits can also be found in organ cells like cardiomyocytes, corneal cells, glomeruli and tubules of the kidney, ganglia of the autonomous nervous system, epithelial and goblet cells of bronchi, etc. [27–29]. In the central nervous system, Gb3-storage has been reported to be limited to scattered neurons in the spinal cord, brainstem, amygdala, hypothalamus, and entorhinal cortex [30]. 7.4. SECONDARY BIOCHEMICAL EFFECTS The primary goal of lysosomal processing is the degradation and recycling of breakdown products for biosynthetic purposes, often referred to as ‘‘metabolic salvage’’ [31]. However, lysosomes have many other functions. Apart from regulating the composition of cellular membranes they modulate signaling processes, for example, by controlling the availability of neurotransmitter receptors and growth factors in the brain [32]. Hypothesizing that membrane lipid composition is altered in Fabry disease, we recently examined fibroblasts from patients with Fabry disease for defects in mitochondrial respiratory chain enzymes as integrity of the mitochondrial inner membrane is essential for function of the respiratory chain. Decreased activities of several mitochondrial respiratory chain complexes were found, which was reflected in decreased cellular levels of ADP and creatinephosphate [33]. These findings are of relevance in human Fabrypatients in vivo as reduced levels of creatinephosphate and ATP were found in heart. ERT led to partial recovery of creatinephosphate- and ATP-contents (M. Beer, personal communication). The molecular mechanism for these abnormalities in oxidative phosphorylation is unclear. How does a lysosomal enzyme deficiency translate into

64

DAS AND NAIM

mitochondrial dysfunction? Cellular lipid traYcking is disturbed in lysosomal storage diseases [34]. In healthy cells, membrane lipid components are hydrolyzed inside lysosomes. In M. Fabry, this process is disturbed leading to lysosomal trapping of Gb3 and other lipid compounds and liberation of malprocessed lipids. Trapping of sphingolipids and their metabolites in the lysosome may lead to a reduced sphingolipid content of the recycled lipids which has a direct eVect on the lipid composition of membranes in diVerent organelles, for example, the inner mitochondrial membrane. This has serious consequences for the biophysical properties of the membrane bilayer. Membranes can generally change from an ordered to a disordered state depending on their lipid composition [34]. The inner mitochondrial membrane constitutes a permeability barrier for protons which allows maintenance of the electrochemical proton gradient created via electron transport over the mitochondrial respiratory chain [35]. Sphingolipids favor the ordered state of membranes [34]. A lack of sphingolipids in the inner mitochondrial membrane of Fabry cells would, therefore, result in a disordered state of the membrane. This could lead to an increased permeability of the membrane as polar substances like protons can more easily intercalate into the disordered lipids (uncoupling or ‘‘loose coupling’’ of mitochondria). On the other hand, the membrane geometry is changed. In the disordered state, the membrane thickness decreases while the membrane area increases (Fig. 3). If we consider the inner mitochondrial membrane as a dielectric of the proton gradient the membrane potential c is defined as follows: Dc ¼

Qt e0 er A

where Q is the electrical charge of the proton gradient, e0 is the dielectrical constant, er the dielectrical constant of the membrane, A is the surface area of the inner mitochondrial membrane, and t its thickness. If A goes up and t is reduced due to a transition from ordered to disordered state, the membrane potential will be decreased. These changes in the membrane geometry lead to a reduction of the electrochemical potential across the inner mitochondrial membrane. A reduced membrane potential leads to increased binding of the naturally occurring inhibitor protein IF1 of the mitochondrial ATPsynthase [36–41] which would explain the reduced activity of the ATPsynthase observed in Fabry fibroblasts. ‘‘Loose coupling’’ of mitochondria would also lead to a reduced membrane potential with increased binding of IF1 and reduced activity of the mitochondrial ATPsynthase. The respiratory chain complexes I, II, III, and IV are embedded in the inner mitochondrial membrane and therefore could be directly aVected by changes in membrane lipid composition.

BIOCHEMICAL BASIS OF FABRY DISEASE

65

A Lipid rafts

Cholesterol/sphingolipid-rich domains B Lipid organization and membrane geometry

Ordered state

Disordered state: d & A

FIG. 3. Structural characteristics of lipid rafts. Lipid rafts are specialized membrane domains enriched in cholesterol and glycosphingolipids (A). The extended fatty-acid chains of lipids within these membrane structures generate a more tightly packed domain with higher order; lipids with long, straight acyl chains are preferentially incorporated into the rafts and also alter membrane geometry (B).

The same mechanism of altered lipid composition may lead to inhibition of fatty-acid oxidation enzymes located at or in the inner mitochondrial membrane as very-long-chain acyl-CoA dehydrogenase (VLCAD), carnitine-palmitoyl transferase II (CPT II), the carnitine-acylcarnitine carrier (CAC), and the trifunctional protein which is a multienzyme complex of fatty-acid oxidation in the inner mitochondrial membrane bearing longchain 2-enoyl-CoA hydratase, long-chain 3-hydroxyacyl-CoA dehydrogenase, and long-chain 3-ketoacyl-CoA thiolase activities. Other enzymes of b-oxidation like medium-chain-acyl-CoA dehydrogenase (MCAD), shortchain-acyl-CoA dehydrogenase (SCAD) and short-chain 3-hydroxyacylCoA dehydrogenase (SCHAD) are located in the mitochondrial matrix and may serve as reference enzymes, while carnitine-palmitoyl transferase I (CPT I) is located in the outer mitochondrial membrane. No data on the activities of fatty-acid oxidation enzymes in Fabry disease are available, so far.

66

DAS AND NAIM

If the abnormalities of energy metabolism were expressed in vascular endothelium and/or smooth muscle cells they might be responsible for vascular dysfunction especially in the brain leading to features like cerebrovascular events and white matter lesions which are probably a consequence of microangiopathy [42, 43]. Relaxation of vascular smooth muscle requires energy for the reduction of myoplasmic calcium concentrations, in case of reduced energy generation the vascular lumen may be compromised resulting in ischemia [44]. On the other hand, nitric oxide has been shown to be involved in vascular relaxation using a cGMP-dependent pathway [44]. Under energy depletion, mitochondrial NO production may be reduced leading to vasoconstriction [44, 45]. Energy depletion may favor the energetically eYcient ‘‘latch state’’ of vascular muscle, hence increase vascular tone [44, 45]. In an attempt to reverse the abnormalities of energy metabolism observed, we incubated cultured fibroblasts from Fabry patients with Agalsidase alpha for up to 10 days and saw partial recovery of respiratory chain complexes (unpublished data). It has previously been shown that Gb3 is cleared from fibroblasts after a few hours to days of enzyme incubation in vitro [46, 47]. The changes in energy metabolism persist longer than Gb3 storage which would fit into our concept of changes in lipid composition of membranes. The slow response to ERT in vivo in patients with M. Fabry may also reflect persisting abnormalities of membrane lipid composition. Particularly important are those eVects that are associated with alteration of membrane subdomains known as lipid rafts or detergent-resistant membranes (DRMs). Lipid rafts are specialized membrane domains enriched in cholesterol and glycosphingolipids (Fig. 3A) [48]. The extended fatty-acid chains of lipids within these membrane structures generate a more tightly packed domain with higher order; lipids with long, straight acyl chains are preferentially incorporated into the rafts (Fig. 3B). These structural characteristics propose that lipid rafts exit as a separate ordered phase that floats in a sea of poorly ordered lipids. Lipid rafts are implicated in a variety of cellular functions. For example, it has been shown that proteins involved in signaling events can be brought together at the cell surface via clustering of these microdomains [49]. Lipid rafts can act as docking sites for pathogens and toxins on host cells [50] and their role in cholesterol homeostasis has been demonstrated [51]. One of the most important functions for lipid rafts is its relation to protein sorting along the secretory pathway. While the role of lipid rafts as a driving vehicle for cell surface targeting of a subset of proteins has been demonstrated in the early secretory pathway, that is, between the ER and the Golgi apparatus [52], the most established function of lipid rafts is their association with polarized protein sorting in epithelial cells. In fact, a subset of apical proteins interact with lipid rafts and their sorting occurs based on signals

BIOCHEMICAL BASIS OF FABRY DISEASE

67

such as N- or O-linked glycans [53], lipid anchors (e.g., glycosylphosphatidylinositol (GPI), acylation) [49], or signals residing in transmembrane domains [54]. The sorting function of membrane rafts within the secretory pathway occurs from the trans-Golgi network (TGN) onward. Selective delivery of certain proteins in TGN-derived vesicles enriched in certain lipids results in a distinct distribution of these molecules among membrane domains in polarized cells and maintains membrane identity and specialized functionality. Previously, it has been shown that many lipids processed in lysosomes reside close to raft-related compounds like cholesterol and sphingomyelin [55–57]. Furthermore, some of these raft-related compounds are processed in lysosomes. Therefore, the malfunctioning of a-galactosidase in Fabry disease strongly suggests that the membrane lipid composition is altered. In particular glycolipids, the target of a-galactosidase, are not metabolized leading to disturbances in the membrane homeostasis with subsequent implications on the intracellular vesicular traYcking. In fact, comparative analysis of the membrane lipid composition of normal fibroblasts and fibroblasts from patients carrying the mutations c.658C > T and c.959T > G in Fabry disease revealed significant fluctuations in the levels of membrane lipids, whereby cholesterol and sphingomyelin were substantially reduced. These findings prompted us to examine the traYcking of lipid raft-associated and nonassociated proteins to the plasma membrane. For this, we utilized dipeptidylpeptidase IV, a type II ubiquitous membrane glycoprotein that is expressed in a variety of polarized and non‐polarized cells and is sorted to the apical membrane via association with lipid rafts in the TGN [58]. Aminopeptidase N, on the other hand, was investigated as a representative of proteins that are not associated with lipid rafts. These proteins were conjugated to the green fluorescent protein (GFP) and expressed in wild-type fibroblasts and fibroblasts of a patient with Fabry disease. Fig. 4 shows that aminopeptidase N-GFP was normally localized at the cell surface in both cell types (A), while a strong intracellular staining of dipeptidylpeptidase IV was observed in the Fabry fibroblasts indicating an intracellular block of this protein and impaired traYcking to the cell surface (B). It is obvious, therefore, that the changes in the lipid composition of cell membranes in Fabry disease have implications on the traYcking and sorting of rafts-associated proteins and on the overall protein–lipid interaction at the cell membrane. As a consequence signaling at the cell surface, gene regulation, and subsequent protein expression may be altered. In line with these considerations increased expression of CD 77 (Gb3) was recently observed in a cellular model of Fabry disease. In response to ERT CD3 expression decreased [59]. As already mentioned in Section 7.1, globotriaosylsphingosine has been shown to promote proliferation of vascular smooth muscle cells which may contribute to vascular dysfunction [26].

68

DAS AND NAIM

Aminopeptidase N-GFP A

Patient

Wild type

Dipeptidylpeptidase IV-GFP B

Wild type

Patient

FIG. 4. Cellular localization of the membrane proteins aminopeptidase N and dipeptidylpeptidase IV. Aminopeptidase N is transported to the cell surface independent of lipid rafts, while dipeptidylpeptidase IV associates with lipid rafts in the trans-Golgi network prior to tracking to the cell surface. These proteins were conjugated to the green fluorescent protein (GFP) and expressed in wild-type fibroblasts and fibroblasts of a patient with Fabry disease. Aminopeptidase N-GFP was normally localized at the cell surface, while a strong intracellular staining of dipeptidylpeptidase IV was observed in the Fabry fibroblasts indicating an intracellular block of this protein.

REFERENCES [1] J. Fabry, Ein Eintrag zur Kenntnis der Purpura haemorrhagica nodularis (Purpura papulosa haemorrhagica Hebrae), Arch. Derm. Syphil. (Berlin) 43 (1898) 187–200. [2] W. Anderson, A case of angio-keratoma, Brit. J. Derm. 10 (1898) 113–117. [3] C. de Duve, The lysosome turns fifty, Med. Sci. (Paris) 21 (2005) 12–15. [4] H.G. Hers, Alpha-glucosidase deficiency in generalized glycogenstorage disease (Pompe’s disease), Biochem. J. 86 (1963) 11–16. [5] R.O. Brady, A.E. Gal, R.M. Bradley, E. Martensson, A.L. Warshaw, L. Laster, Enzymatic defect in Fabry’s disease. Ceramidetrihexosidase deficiency, N. Engl. J. Med. 276 (1967) 1163–1167.

BIOCHEMICAL BASIS OF FABRY DISEASE

69

[6] J.A. Kint, Fabry’s disease: alpha-galactosidase deficiency, Science 167 (1970) 1268–1269. [7] R. Kornreich, R.J. Desnick, D.F. Bishop, Nucleotide sequence of the human alphagalactosidase A gene, Nucleic Acids Res. 17 (1989) 3301–3302. [8] M.F. Lyon, X-chromosome inactivation as a system of gene dosage compensation to regulate gene expression, Prog. Nucleic Acid Res. Mol. Biol. 36 (1989) 119–130. [9] R.J. Desnick, Y.A. Ioannu, C.M. Eng, Alpha galactosidase deficiency: Fabry disease, in: C.R. Scriver, A.L. Beaudet, W.S. Sly, D. Valle (Eds.), The Metabolic and Molecular Bases of Inherited Disease. McGraw-Hill, New York, 2001, pp. 3733–3774. [10] K.D. MacDermot, A. Holmes, A.H. Miners, Anderson–Fabry disease: clinical manifestations and impact of disease in a cohort of 98 hemizygous males, J. Med. Genet. 38 (2001) 750–760. [11] K.D. MacDermot, A. Holmes, A.H. Miners, Anderson–Fabry disease: clinical manifestations and impact of disease in a cohort of 60 obligate carrier females, J. Med. Genet. 38 (2001) 769–775. [12] N.A. Chamoles, M. Blanco, D. Gaggioli, Fabry disease: enzymatic diagnosis in dried blood spots on filter paper, Clin. Chim. Acta 308 (2001) 195–196. [13] O.A. Bodamer, Newborn screening in Fabry disease: what can be achieved with early diagnosis? Clin. Ther. 30 (Suppl. B) (2008) S41. [14] X.K. Zhang, C.S. Elbin, W.L. Chuang, et al., Multiplex enzyme assay screening of dried blood spots for lysosomal storage disorders by using tandem mass spectrometry, Clin. Chem. 54 (2008) 1725–1728. [15] D. Matern, Newborn screening for lysosomal storage disorders, Acta Paediatr. Suppl. 97 (2008) 33–37. [16] A.C. Vedder, G.E. Linthorst, G. Houge, et al., Treatment of Fabry disease: outcome of a comparative trial with agalsidase alfa or beta at a dose of 0.2 mg/kg, PLoS ONE 2 (2007) e598. [17] J.M. Aerts, C. Hollak, R. Boot, A. Groener, Biochemistry of glycosphingolipid storage disorders: implications for therapeutic intervention, Philos. Trans. R. Soc. Lond. B Biol. Sci. 358 (2003) 905–914. [18] A. Abe, S. Gregory, L. Lee, et al., Reduction of globotriaosylceramide in Fabry disease mice by substrate deprivation, J. Clin. Invest. 105 (2000) 1563–1571. [19] A. Abe, S.R. Wild, W.L. Lee, J.A. Shayman, Agents for the treatment of glycosphingolipid storage disorders, Curr. Drug Metab. 2 (2001) 331–338. [20] T. Okumiya, S. Ishii, T. Takenaka, et al., Galactose stabilizes various missense mutants of alpha-galactosidase in Fabry disease, Biochem. Biophys. Res. Commun. 214 (1995) 1219–1224. [21] E. Beutler, W. Kuhl, Relationship between human alpha-galactosidase isozymes, Nat. New Biol. 239 (1972) 207–208. [22] E. Beutler, W. Kuhl, Absence of cross-reactive antigen in Fabry’s disease, N. Engl. J. Med. 289 (1973) 694–695. [23] P.J. Rietra, J.L. Molenaar, M.N. Hamers, J.M. Tager, P. Borst, Investigation of the alphagalactosidase deficiency in Fabry’s disease using antibodies against the purified enzyme, Eur. J. Biochem. 46 (1974) 89–98. [24] M.N. Hamers, D. Wise, A. Ejiofor, A. Strijland, D. Robinson, J.M. Tager, Relationship between biochemical and clinical features in an English Anderson–Fabry family, Acta Med. Scand. 206 (1979) 5–10. [25] G. Bach, E. Rosenmann, A. Karni, T. Cohen, Pseudodeficiency of alpha-galactosidase A, Clin. Genet. 21 (1982) 59–64. [26] J.M. Aerts, J.E. Groener, S. Kuiper, et al., Elevated globotriaosylsphingosine is a hallmark of Fabry disease, Proc. Natl. Acad. Sci. USA 105 (2008) 2812–2817.

70

DAS AND NAIM

[27] M.M. Kelly, R. Leigh, R. McKenzie, D. Kamada, E.H. Ramsdale, F.E. Hargreave, Induced sputum examination: diagnosis of pulmonary involvement in Fabry’s disease, Thorax 55 (2000) 720–721. [28] M. Elleder, V. Bradova, F. Smid, et al., Cardiocyte storage and hypertrophy as a sole manifestation of Fabry’s disease. Report on a case simulating hypertrophic non-obstructive cardiomyopathy, Virchows Arch. A Pathol. Anat. Histopathol. 417 (1990) 449–455. [29] L.K. Brown, A. Miller, A. Bhuptani, et al., Pulmonary involvement in Fabry disease, Am. J. Respir. Crit. Care Med. 155 (1997) 1004–1010. [30] G.A. deVeber, G.A. Schwarting, E.H. Kolodny, N.W. Kowall, Fabry disease: immunocytochemical characterization of neuronal involvement, Ann. Neurol. 31 (1992) 409–415. [31] G. Tettamanti, R. Bassi, P. Viani, L. Riboni, Salvage pathways in glycosphingolipid metabolism, Biochimie 85 (2003) 423–437. [32] S.U. Walkley, Pathogenic cascades in lysosomal disease-Why so complex? J. Inherit. Metab. Dis. 32 (2009) 181–189. [33] T. Lucke, W. Hoppner, E. Schmidt, S. Illsinger, A.M. Das, Fabry disease: reduced activities of respiratory chain enzymes with decreased levels of energy-rich phosphates in fibroblasts, Mol. Genet. Metab. 82 (2004) 93–97. [34] F.R. Maxfield, I. Tabas, Role of cholesterol and lipid organization in disease, Nature 438 (2005) 612–621. [35] P. Mitchell, Chemiosmotic coupling in energy transduction: a logical development of biochemical knowledge, J. Bioenerg. 3 (1972) 5–24. [36] A.M. Das, Regulation of the mitochondrial ATP-synthase in health and disease, Mol. Genet. Metab. 79 (2003) 71–82. [37] A.M. Das, D.A. Harris, Intracellular calcium as a regulator of the mitochondrial ATP synthase in cultured cardiomyocytes, Biochem. Soc. Trans. 18 (1990) 554–555. [38] A.M. Das, D.A. Harris, Control of mitochondrial ATP synthase in heart cells: inactive to active transitions caused by beating or positive inotropic agents, Cardiovasc. Res. 24 (1990) 411–417. [39] A.M. Das, D.A. Harris, Regulation of the mitochondrial ATP synthase in intact rat cardiomyocytes, Biochem. J. 266 (1990) 355–361. [40] G. Lippe, M.C. Sorgato, D.A. Harris, Kinetics of the release of the mitochondrial inhibitor protein. Correlation with synthesis and hydrolysis of ATP, Biochim. Biophys. Acta 933 (1988) 1–11. [41] G. Lippe, M.C. Sorgato, D.A. Harris, The binding and release of the inhibitor protein are governed independently by ATP and membrane potential in ox-heart submitochondrial vesicles, Biochim. Biophys. Acta 933 (1988) 12–21. [42] M. Nill, M.J. Muller, M. Beck, P. Stoeter, A. Fellgiebel, Pathophysiological aspects of brain structural disturbances in patients with Fabry disease: literature review, Fortschr. Neurol. Psychiatr. 74 (2006) 687–695. [43] L.B. Jardim, F. Aesse, L.M. Vedolin, et al., White matter lesions in Fabry disease before and after enzyme replacement therapy: a 2-year follow-up, Arq. Neuropsiquiatr. 64 (2006) 711–717. [44] R.A. Murphy, J.S. Walker, Inhibitory mechanisms for cross-bridge cycling: the nitric oxidecGMP signal transduction pathway in smooth muscle relaxation, Acta Physiol. Scand. 164 (1998) 373–380. [45] H. Tanaka, K. Homma, H.D. White, T. Yanagida, M. Ikebe, Smooth muscle myosin phosphorylated at single head shows sustained mechanical activity, J. Biol. Chem. 283 (2008) 15611–15618.

BIOCHEMICAL BASIS OF FABRY DISEASE

71

[46] D. Blom, D. Speijer, G.E. Linthorst, W.G. Donker-Koopman, A. Strijland, J.M. Aerts, Recombinant enzyme therapy for Fabry disease: absence of editing of human alphagalactosidase A mRNA, Am. J. Hum. Genet. 72 (2003) 23–31. [47] R.N. Sifers, J.S. Mayes, R.E. Nordquist, Loss of electron-dense lamellar material from Fabry’s disease fibroblasts after enzyme replacement, Hum. Genet. 65 (1983) 85–87. [48] K. Simons, D. Toomre, Lipid rafts and signal transduction, Nat. Rev. Mol. Cell. Biol. 1 (2000) 31–39. [49] D.A. Brown, J.K. Rose, Sorting of GPI-anchored proteins to glycolipid-enriched membrane subdomains during transport to the apical cell surface, Cell 68 (1992) 533–544. [50] F.G. van der Goot, T. Harder, Raft membrane domains: from a liquid-ordered membrane phase to a site of pathogen attack, Semin. Immunol. 13 (2001) 89–97. [51] S. Heino, S. Lusa, P. Somerharju, C. Ehnholm, V.M. Olkkonen, E. Ikonen, Dissecting the role of the golgi complex and lipid rafts in biosynthetic transport of cholesterol to the cell surface, Proc. Natl. Acad. Sci. USA 97 (2000) 8375–8380. [52] M. Alfalah, G. Wetzel, I. Fischer, et al., A novel type of detergent-resistant membranes may contribute to an early protein sorting event in epithelial cells, J. Biol. Chem. 280 (2005) 42636–42643. [53] M. Alfalah, R. Jacob, U. Preuss, K.P. Zimmer, H. Naim, H.Y. Naim, O-linked glycans mediate apical sorting of human intestinal sucrase-isomaltase through association with lipid rafts, Curr. Biol. 9 (1999) 593–596. [54] P. ScheiVele, M.G. Roth, K. Simons, Interaction of influenza virus haemagglutinin with sphingolipid-cholesterol membrane domains via its transmembrane domain, EMBO J. 16 (1997) 5501–5508. [55] T. Kolter, K. SandhoV, Principles of lysosomal membrane digestion: stimulation of sphingolipid degradation by sphingolipid activator proteins and anionic lysosomal lipids, Annu. Rev. Cell Dev. Biol. 21 (2005) 81–103. [56] T. Kolter, F. Winau, U.E. Schaible, M. Leippe, K. SandhoV, Lipid-binding proteins in membrane digestion, antigen presentation, and antimicrobial defense, J. Biol. Chem. 280 (2005) 41125–41128. [57] G. Schwarzmann, K. SandhoV, Metabolism and intracellular transport of glycosphingolipids, Biochemistry 29 (1990) 10865–10871. [58] M. Alfalah, R. Jacob, H.Y. Naim, Intestinal dipeptidyl peptidase IV is eYciently sorted to the apical membrane through the concerted action of N- and O-glycans as well as association with lipid microdomains, J. Biol. Chem. 277 (2002) 10683–10690. [59] T. Thomaidis, M. Relle, M. Golbas, et al., Downregulation of alpha-galactosidase A upregulates CD77: functional impact for Fabry nephropathy, Kidney Int. 75 (2009) 399–407.

This page intentionally left blank

ADVANCES IN CLINICAL CHEMISTRY, VOL. 49

URINARY BIOMARKERS FOR THE DETECTION OF RENAL INJURY Mitchell H. Rosner1 Division of Nephrology, University of Virginia Health System, Box 800133, Charlottesville, Virginia 22908

1. Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. Biomarker Development for Kidney Diseases. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Definition of a Biomarker . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Defining AKI. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3. Types and Characteristics of Biomarkers for the Diagnosis of AKI. . . . . . . . . 3.4. The Ideal Biomarker. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5. Sources of Biomarker Discovery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4. Biomarkers for AKI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2. N-Acetyl-b-Glucosaminidase. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3. Kidney Injury Molecule-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4. Neutrophil Gelatinase-Associated Lipocalin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.5. Interleukin-18 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.6. Fatty-Acid Binding Protein . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5. Biomarker Development and Implementation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6. Summary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

73 74 75 75 76 78 80 80 81 81 83 84 86 88 89 90 91 92

1. Abstract Despite the well-known limitations, currently the most widely used biomarkers for the early detection of chronic kidney disease or acute kidney injury are proteinuria, serum creatinine, and blood urea nitrogen. All of these are less than optimal and tend to focus attention on later stages of injury when therapies may be less effective. Recently, there has been a great surge of interest in identifying novel biomarkers that can be easily detected in the urine 1

Corresponding author: Mitchell H. Rosner, E-mail: [email protected] 73

0065-2423/09 $35.00 DOI: 10.1016/S0065-2423(09)49004-8

Copyright 2009, Elsevier Inc. All rights reserved.

74

MITCHELL H. ROSNER

that can diagnose renal injury at the earliest stages. A variety of methods have been employed to identify these biomarkers including transcriptomics, proteomics, metabolomics, lipidomics, and gene arrays. Currently, several candidate biomarkers have been identified and studied in different renal injury states. These include kidney injury molecule-1 (KIM-1), neutrophil gelatinase-associated lipocalin (NGAL), interleukin (IL)-18, and fatty-acid binding proteins (FABPs). This review will highlight the current state of knowledge of these biomarkers as well as the limitation of these biomarkers in the early diagnosis of renal injury. 2. Introduction Kidney disease continues to remain a significant clinical problem. Patients with chronic kidney disease (CKD) suffer from a number of complications leading to early death, dialysis, or transplantation. The latest United States Renal Data Systems analysis indicates that there are 20 million Americans with CKD [1]. Although diabetes mellitus and hypertension are the most common cause of ESRD, acute kidney injury (AKI) is now recognized as an important and growing contributor of CKD. Overall mortality associated with AKI has been reported to be 40–60% in critically ill patients [2–4]. Critically important is that in patients suffering from AKI, 13.4% of patients (or 30% of patients with AKI superimposed on CKD) will progress to ESRD in 3 years [2–4]. Advances in long-term improvement and outcomes of patients with kidney disease will require the use of novel biomarkers to identify patients at high risk for kidney disease and to diagnose kidney disease at its earliest stage to allow for effective treatment. This is the potential promise of biomarkers. A biomarker is a substance found in the blood, body fluids, or tissues that provides a measure of normal biological or pathological processes or response to pharmacological compounds or drugs. There are a wide variety of biomarkers including but not limited to mRNA, proteins and peptides, and lipid molecules. In AKI, important pathophysiological processes such as inflammation, apoptotic and necrotic cell death, and tubule regeneration may be reflected in blood or urine and indicated by the assay of a biomarker. In diseases associated with progressive CKD, the detection of an informative biomarker may allow determination of whether specific pathophysiological processes are active and allow for tailored therapy. An array of candidate markers along with clinical information in long-term clinical studies with appropriate analytical methodologies will likely provide prognostic information that will be critical in furthering the understanding of kidney disease. Despite well-known limitations, currently the most widely used biomarkers for the early diagnosis of CKD and AKI are proteinuria, serum creatinine,

URINARY BIOMARKERS FOR THE DETECTION OF RENAL INJURY

75

and blood urea nitrogen. Most clinicians are aware that serum creatinine and blood urea nitrogen are poor biomarkers due to inherent characteristics of these molecules and handling by the kidney. Creatinine is secreted and urea nitrogen is reabsorbed by the renal tubules. Many endogenous substances interfere in the assay for creatinine. Most importantly, serum creatinine and urea appear late after AKI and the serum levels in part depend on the generation (large or small body mass). AKI is a nonsteady-state condition thus serum creatinine and urea nitrogen will lag behind kidney injury [5]. Furthermore, while serum creatinine and urea nitrogen levels may continue to rise, the pathological processes that led to the injury may no longer be operative. Finally, serum creatinine only measures glomerular filtration and thus may not give a broader indication of renal injury. For these reasons, new biomarkers are imperative that allow for rapid, early diagnosis of injury, allow determination of pathophysiological processes, and allow tailored therapy. With knowledge of these limitations of current biomarkers and the lack of progress in reducing the mortality and morbidity from kidney disease, there has been a great surge of interest in identifying novel biomarkers with a particular emphasis on the early diagnosis of kidney disease. A variety of methods have been employed including transcriptomics, proteomics, gene arrays, metabolomics, and lipidomics. Currently, candidate biomarkers have been found in different disorders and have been tested in humans and many candidate biomarkers have yet to be identified. Most studies to date are preliminary and require validation in large multicenter studies followed by commercial assay development validation and testing. This review outlines the rapid advances made in the field of biomarker development for kidney disease in which a variety of novel molecules have been identified and studied in humans. The review focuses on biomarker development for AKI where the need for rapid and early diagnosis is particularly vital. There is much hope that in the near future advances will be made in reducing the morbidity and mortality of kidney disease. An important and necessary step to achieving this goal will be to identify novel biomarkers of kidney disease.

3. Biomarker Development for Kidney Diseases 3.1. DEFINITION OF A BIOMARKER A biomarker is any biological parameter that can be objectively quantified or evaluated to determine the status of a biological process, whether it be a normal occurrence, a pathogenic process, or a response to therapeutic

76

MITCHELL H. ROSNER

intervention. Examples include blood pressure, enzyme activity, protein concentration, mRNA expression profiles, bone density, gram staining, or electrical potential. The most valuable biomarkers track the severity of disease with high fidelity, including rate of onset, response to therapy, and remission. In the kidney, biomarkers are desired that either directly measure renal structural damage, especially at its earliest stages (similar to troponin for cardiac disease), or be integral to the disease pathway (such as the activity of a transporter in the luminal membrane of the renal tubule) so that the biomarker result cannot be dissociated from the hard clinical outcome it is intended to track. Thus, the surrogate biomarker can predict a well-defined clinical endpoint, and greatly speed the drug discovery process by shortening the length and size of a clinical trial. However, validation of a surrogate endpoint is extremely difficult and costly unless it is integrated into existing clinical research projects or clinical trials. This is especially problematic for those biomarkers developed to detect low-frequency events. 3.2. DEFINING AKI Critical to the development of biomarkers is a clear definition of the outcome of interest to which the biomarker will be correlated. For acute renal failure (ARF), up until recently, this has been very problematic as there was a lack of consensus regarding the criteria used to define this outcome. It is estimated that more than 30 different definitions of ARF exist in the published literature [6], ranging from severe (ARF requiring dialysis) to mild (modest observable increases in serum creatinine) [7]. As a result of the disparate clinical and physiologic endpoints used to guide investigation, epidemiologic studies as well as trials of prevention and intervention are often not comparable. This morass would make biomarker development nearly impossible. As part of the Acute Dialysis Quality Initiative (ADQI) 2nd International Consensus Conference, the RIFLE (Risk of kidney dysfunction, Injury to the kidney, Failure of kidney function, Loss of kidney function, and End stage kidney disease) classification scheme was derived to provide standardized criteria for defining ARF with the goal of facilitating comparison of outcomes across studies, development of prognostic scoring systems, interpretation of therapeutic intervention strategies, and design of multicenter studies [3]. This work ultimately led to the introduction of the term AKI as the preferred nomenclature for the clinical disorder formally called ARF (from this point on AKI will be used to describe the clinical condition formally referred to as ARF). This transition served to emphasize the notion that the spectrum of disease is much broader than that subset of patients who experience failure requiring dialysis support [8]. This new nomenclature

URINARY BIOMARKERS FOR THE DETECTION OF RENAL INJURY

77

underscores the fact that kidney injury exists along a continuum: the more severe the injury, the more likely the overall outcome will be unfavorable. More recently, the acute kidney injury network (AKIN) was formed in an effort to facilitate improved care of patients with or at risk for AKI. In recognition of increasing data suggesting that even small changes in serum creatinine are associated with poorer outcome as measured by mortality [9, 10], the AKIN committee proposed uniform standards for defining, diagnosing, and classifying AKI. This group defined AKI as ‘‘functional or structural abnormalities or markers of kidney damage including abnormalities in blood, urine, or tissues tests or imaging studies present for less than three months [11].’’ The committee set diagnostic criteria for AKI as ‘‘an abrupt (within 48 h) reduction in kidney function currently defined as an absolute increase in serum creatinine of either 0.3 mg/dl or a percentage increase of 50% or a documented reduction in urine output (documented oliguria of <0.5 ml/kg/h for >6 h [11].’’ Appropriate modifications to the RIFLE criteria were made to modify the risk criteria to include an absolute increase in serum creatinine 0.3 mg/dl (Table 1). All future studies of biomarker development for the diagnosis of AKI should utilize this consensus definition. Although clear progress has been made in formulating a consensus definition, diagnostic criteria, and classification scheme for AKI, it is notable that all are heavily dependent on changes in serum creatinine as a marker of injury. Unfortunately, creatinine is a suboptimal marker following injury; levels are often not reflective of glomerular filtration rate (GFR) due to a number of renal and nonrenal influences [12]. In the setting of AKI, the

TABLE 1 PROPOSED CLASSIFICATION SCHEME FOR ACUTE KIDNEY INJURY (AKI) (MODIFIED RIFLE CRITERIA)a Stage 1 (risk) 2 (injury) 3 (failure)

Creatinine criteria

Urine output criteria

Increased SCr of 0.3 mg/dl or increase to 150–200% from baseline Increased SCr to >200–300% from baseline Increased SCr to >300% from baseline or SCr 4 mg/dl (acute rise 0.5 mg/dl)

UO < 0.5 ml/kg/h for >6 h UO < 0.5 ml/kg/h for >12 h UO < 0.3 ml/kg/h 24 h or anuria 12 h

Stages eliminated from original RIFLE criteria Loss Persistent ARF ¼ complete loss of kidney function >4 weeks ESKD End stage kidney disease (>3 months) ARF, acute renal failure; ESKD, end stage kidney disease; RIFLE, risk of renal dysfunction, injury to the kidney, failure of kidney function, loss of kidney function, end stage kidney disease; SCr, serum creatinine; UO, urine output. a Adapted from Ref. [11].

78

MITCHELL H. ROSNER

dynamic relationship between serum creatinine and GFR inhibits the ability to accurately estimate timing of injury and severity of dysfunction following injury [13]. A sudden fall in GFR to a constant low level causes a gradual increase in serum creatinine until a new steady state between generation and excretion is achieved. The rate of rise following AKI is dependent on many factors including the new GFR, rate of tubular secretion, rate of generation, and volume of distribution [12, 13]. As a result, large changes in GFR may be associated with relatively small changes in serum creatinine in the first 24–48 h following AKI, resulting not only in delayed diagnosis and intervention, but also in underestimation of the degree of injury [12]. In addition, there is considerable variability among patients in the correlation between serum creatinine and baseline GFR, in the magnitude of functional renal reserve, and in creatinine synthesis rates. As a result, a renal injury of comparable magnitude may result in disparate alterations in creatinine concentration in different individuals [12]. The insensitivity and nonspecificity of serum creatinine as well other traditionally used markers of renal injury, including blood urea nitrogen, urine sediment, and urinary indices (fractional excretion of sodium, urine osmolality, etc.), have been major obstacles in developing strategies to ameliorate AKI. Results from interventional trials suggesting inefficacy of putative therapies of AKI are by definition confounded by delayed diagnosis and treatment. This paradigm is analogous to the initiation of therapy in patients with myocardial infarction or stroke 48 h after the onset of ischemia [14]. Figure 1 depicts the rise in serum creatinine in relationship to the fall in GFR after an injurious event. The utility of a biomarker would be that it would detect renal injury in a sensitive and specific manner at an early stage before the rise in serum creatinine and thus at a time potentially amenable to intervention. 3.3. TYPES AND CHARACTERISTICS OF BIOMARKERS FOR THE DIAGNOSIS OF AKI Intensive investigative efforts have led to the identification and evaluation of many urinary and serum proteins as potential biomarkers of AKI [13, 14]. In general, serum markers of nephron damage may be of relatively limited utility if they are highly sensitive to modification by any factor that may alter renal perfusion with changes in filtration that may or may not be associated with injury (e.g., volume depletion, hemorrhage, or decreased effective intravascular volume in congestive heart failure or cirrhosis). In addition, when elevated serum levels are observed in the setting of a primary renal insult, serum biomarkers have limited utility in determining the location or mechanism of injury unless they are known to only derive from the kidney [15]. As a result, much of the focus in new biomarker development has focused on the examination of urine proteins and metabolites. Studies have yielded many

URINARY BIOMARKERS FOR THE DETECTION OF RENAL INJURY

79

Injury

GFR (ml/min)

100

10

Novel biomarkers are needed 0

2

Creatinine (mg/dl)

10

3

6

8

10

By the time scr rises, the injury has already occurred and interventions may be too late

8 6 4 2 0

2

3 6 Time (days)

8

10

FIG. 1. Time course of fall in glomerular filtration rate as compared to rise in serum creatinine and rise in a hypothetical biomarker (bar) in relation to an injurious insult.

promising urinary candidates for the early detection of AKI and further characterization is anticipated to aid in earlier diagnosis, identification of mechanism of injury, assessment of site, and severity of injury. Hopefully one or more of these biomarkers, either alone or in combination, will prove to be useful in guiding targeted intervention, and monitoring of disease progression and resolution. The large majority of existing studies have focused on animal models and adult patients at risk for or with established AKI. It should be emphasized, however, that pediatric patients represent an important subpopulation for study as they generally lack comorbidities such as hypertension, atherosclerosis, and diabetes that affect kidney structure and function and as a result may prove to have very different biomarker profiles than adults [16]. Existing biomarkers of AKI may be broadly classified into two categories: (1) proteins with enzymatic activity which are leaked into the urine following injury, and (2) urinary proteins without enzymatic activity which are either upregulated or specifically released into the urine in the setting of cellular injury. Of particular interest are applying these biomarkers to aid in (1) early detection of AKI to improve testing of early disease preemption strategies in situations where the overall incidence of AKI is low (i.e., postcardiac surgery) or the disease is complicated (sepsis), (2) diagnosing AKI accurately and

80

MITCHELL H. ROSNER

differentiating AKI from other forms of renal injury (such as prerenal azotemia), (3) ascertaining the etiology and site of renal injury to select appropriate therapy, (4) predicting the severity of AKI to aid in stratifying patients in clinical trials, predicting prognosis, and outcome to assess who needs drug treatment or dialysis, and addressing ‘‘when to start renal replacement therapy,’’ and (5) monitoring the effects of an intervention (intermediate or surrogate outcome biomarkers) for initial dosing and Phase 2 proof of concept clinical trials, and even larger Phase 3 efficacy trials. 3.4. THE IDEAL BIOMARKER Clinically, applicable biomarkers should be accurate, relatively noninvasive tests that can be easily performed at the bedside or in a clinical laboratory. The most desirable tests involve a blood or spot urine specimen, can be measured efficiently, have a rapid turn-around time, and are stable over time. A useful AKI biomarker should also have a robust, standardized, and scalable assay that is cost-effective and is validated by prospective studies to have a high predictive ability (both sensitive and specific for AKI). The most effective biomarkers augment rather than replace conventional clinical observations. The specificity (i.e., lack of cross-reactivity with other non-AKI renal diseases) must be high to avoid expensive or hazardous additional diagnostic or therapeutic intervention. Sensitivity is essential, especially for early detection biomarkers. A wide dynamic range is desirable since it has the potential to determine the timing or severity of injury, rather than interpretation as a qualitative yes/no answer. Typically a biomarker will not clearly distinguish between normal and ill patients; therefore, a cutoff point must be chosen, which represents a tradeoff between specificity and sensitivity. Because a single cutoff point is qualitative, a receiver operating characteristic (ROC) curve demonstrates in detail the relationship between specificity and sensitivity along a continuum of cutoff values. The overall predictive ability of a biomarker is typically measured by the area under the ROC curve (AUC), in which 1.0 reflects perfect diagnostic value and 0.5 reflects no diagnostic value. Individual biomarkers rarely have an AUC of greater than 0.80. An AUC of 0.80 for an individual biomarker is too low for clinical utility; however, when combined with clinical information and/or the results of other biomarkers, the AUC of the diagnostic panel should approach 1.0 [17]. 3.5. SOURCES OF BIOMARKER DISCOVERY Blood, urine, and kidney tissue from biopsy are traditional sources of biomarker development for renal disease. Of these, urine has built-in advantages for biomarker discovery because it is noninvasive and its proximity to

URINARY BIOMARKERS FOR THE DETECTION OF RENAL INJURY

81

the kidney reduces signals from other organs. Clinical urinalysis has not changed over several decades; it is useful for general renal disease diagnosis, but with little specificity. Protein biomarkers should exist that are more specific and sensitive for each type of specific renal cell damage, but they have not yet been identified, primarily because of their low abundance in urine, as well as interference by the most abundant proteins albumin and uromodulin/Tamm Horsfall protein. To overcome this problem, subfractionation of urine has been the most common strategy. Typically, a chemical property, such as hydrophobicity or ionic charge [18] is used to divide the urine proteome into manageable subpopulations of proteins. Another approach that is commonly used in serum and plasma is to subtract the abundant proteins either by chemical [19] or affinity [20–22] methods. A complementary approach is to enrich for a class of proteins by affinity chromatography [23–25] or to select for a restricted proteome, such as that found in exosomes. Exosomes are a distinct subcellular structure found in normal human urine [26]. They are small (<100 nm diameter) vesicles that contain both membrane and cytosolic proteins and are normally shed into the urine from all segments of the nephron. [26]. Recently, urinary Fetuin-A, a novel biomarker which can detect AKI earlier than serum creatinine, was discovered in urine exosomes by proteomic methods [27].

4. Biomarkers for AKI 4.1. OVERVIEW Numerous urine biomarkers have been identified for the potential early diagnosis of AKI. These include urinary proteins with enzymatic activity such as the proximal tubule brush border enzyme alanine aminopeptidase (AA) [28], the proximal tubule cytosolic enzymes a- or -glutathione-Stransferase [29, 30], and the proximal tubule lysosomal enzyme N-acetyl-b-D-glucosaminidase [31]. In the setting of acute or chronic damage, enzymes physiologically present in the tubular epithelium may be released into the urine secondary to leakage from damaged cells or secondary to intensified enzyme induction during the repair and regeneration process [32, 33]. Preclinical and clinical studies have lead to extensive knowledge regarding the segmental nephron distribution and cellular ultrastructural (brush border, lysosome, or cytoplasm) location of tubular enzymes. As a result, enzyme detection in the urine potentially provides valuable information not only pertinent to the site of tubular injury (proximal vs. distal tubule) but also the severity of injury. As a general rule, brush border enzymuria is indicative of a less severe injury than lysosomal or cytosolic enzymuria

82

MITCHELL H. ROSNER

[32–34]. It should be noted, however, that despite the theoretical advantage of identifying the primary site of nephron injury, thus far enzymatic markers have not proved effective in differentiating between predominate proximal or distal tubular involvement [32]. Measurement of some urinary enzymes is readily performed with existing colorimetric assays and commercially available ELISA kits. Although these quantification methods have been extensively validated and are highly reproducible, they do not lend themselves to the high-throughput analysis preferable when analyzing large numbers of samples over extended time courses [33]. A number of important disadvantages have been noted with respect to the use of enzymuria in the setting AKI. Although highly sensitive for renal injury, the utility of urinary enzymes has been clouded by the low threshold for release in response to physiologic conditions that may not precede clinically significant AKI [35]. Perhaps more important from a practical viewpoint is the instability of many urinary enzymes and the specific processing necessary to ensure integrity in samples [34]. For example, the brush border enzymes alanine aminopeptidase (AA), alkaline phosphatase (AP), and g-glutamyl transpeptidase (g-GT) are stable for only 4 h after urine collection, and samples require gel filtration to eliminate interfering substances [29, 30]. Alpha-glutathioneS-transferase (a-GST) and -glutathione-S-transferase (-GST), cytosolic enzymes that have been found to be rapidly released in the urine following proximal and distal tubular injury, respectively, require a specific stabilization buffer to ensure appropriate quantification [35]. In general, the utility of urinary enzyme excretion as diagnostic or predictive biomarkers for AKI remains an area that warrants additional investigation. Further characterization is needed to determine threshold levels indicative of clinically significant injury and if the theoretic advantage of identifying the primary site of tubule injury will translate to a clinically important advantage by guiding intervention. In addition, it will be important to develop user friendly approaches to sample preservation and quantification to allow for high-throughput processing of samples necessary for large-scale studies and clinical assessment. A number of other urinary antigens have been evaluated as potential biomarkers of AKI. Filtered low molecular weight proteins, including retinol binding protein (RBP), a1-microglobulin, b1-microglobulin, and cystatin-C have been studied extensively over the last 2 decades [36–40]. More recently, proteomic and genomic screening modalities have identified numerous tubular proteins that are markedly upregulated and/or excreted in the setting of renal injury. These novel proteins have generated significant excitement in the research community not only as candidate biomarkers of AKI, but also as molecules that may play critical roles in the regulation of cell dedifferentiation, migration, and proliferation in response to injury. It is

URINARY BIOMARKERS FOR THE DETECTION OF RENAL INJURY

83

anticipated that further investigative efforts will not only define the utility of these proteins as biomarkers, but also enhance our understanding of AKI pathophysiology and aid in the development of targeted interventions to ameliorate injury with the ultimate goal of improving outcomes. Given the large number of these potential urinary biomarkers, this review will focus on those that have undergone the most extensive preclinical and clinical validation. These include N-acetyl-b-glucosaminidase (NAG), kidney injury molecule-1 (KIM-1), neutrophil gelatinase-associated lipocalin (NGAL), interleukin (IL)-18, and fatty-acid binding protein (FABP).

4.2. N-ACETYL-b-GLUCOSAMINIDASE NAG, a proximal tubule lysosomal enzyme, has been extensively studied in both the adult and pediatric population and has proved to be a sensitive, persistent, and robust indicator of AKI. Increased NAG levels have been reported with nephrotoxicant exposure [32, 33], delayed renal allograft function [41], chronic glomerular disease [42], diabetic nephropathy[43], as well as following cardiopulmonary bypass procedures [44]. Westhuyzen et al. [34] reported that urinary NAG levels (in addition to other tubular enzymes) were highly sensitive in detecting AKI in a population of critically ill adult patients, preceding increases in serum creatinine by 12 h to 4 days. Chew et al. [45] reported a poorer outcome (death in hospital, requirement for longterm renal replacement therapy) in patients with higher urinary NAG levels on admission to a renal care unit, indicating a dose response by injury. The greater the urinary NAG in patients already diagnosed with AKI clinical criteria, the greater the incidence of the combined end point of dialysis or death [46]. In a recent study, urinary NAG levels were evaluated in a heterogenous population with AKI [47]. In this study, urine samples were collected from 44 patients with various acute and CKDs, and from 30 normal subjects in a cross-sectional study. The receiver-operator characteristic area under the curve (AUC) for the diagnosis of AKI using urinary NAG was an impressive 0.97. In a case-control study in a subset of patients developing AKI after cardiac bypass surgery, the rise in urinary NAG levels occurred 6 h after surgery and approximately 18 h before the rise in serum creatinine [47]. A possible cofounding factor is that urinary NAG activity has been found to be inhibited by endogenous urea [48] as well as a number of nephrotoxicants and heavy metals [13]. In addition, given the various conditions that have been associated with increased NAG excretion, nonspecificity for AKI may limit its use as a biomarker and more studies are required across a spectrum of causes of AKI to determine its utility.

84

MITCHELL H. ROSNER

4.3. KIDNEY INJURY MOLECULE-1 KIM-1 is a type I cell membrane glycoprotein containing a unique sixcysteine immunoglobulin-like domain and a mucin domain in its extracellular portion [49]. Rat and human cDNAs encoding KIM-1 (Kim-1 in the rat) were initially identified using representational difference analysis, a polymerase-chain reaction-based cDNA subtraction analysis designed to identify genes with differential expression between normal and regenerating kidneys following ischemia/reperfusion (I/R) injury [49]. Kim-1 was subsequently found to be expressed at low to undetectable levels in the normal adult rat kidney but dramatically upregulated in the postischemic kidney. In situ hybridization and immunohistochemistry studies demonstrated that Kim-1 mRNA and protein are expressed in proximal tubule epithelial cells in damaged regions. Kim-1 immunoreactivity was colocalized with the dedifferentiation protein vimentin and the proliferation marker BrdU [49]. Independently, a large pharmaceutical consortium, using an unbiased genomic approach, determined that Kim-1 was upregulated more than any other of the 30,000 genes tested in response to the nephrotoxin cisplatin [50]. Thus, structure and expression data suggest that KIM-1 is an epithelial adhesion molecule upregulated in dedifferentiating and regenerating tubule epithelial cells following injury and may play a role in the restoration of morphological integrity of the tubule [49]. It was later found that the KIM-1 ectodomain was shed from cells in vitro [51] and into the urine in vivo in rodents and humans after proximal tubular injury of varying etiology [52–55]. Urinary KIM-1 quantification was initially performed via an enzyme-linked immunosorbent assay (ELISA) and subsequently by a high-throughput microbead-based assay that offers increased sensitivity and a greater dynamic range [56, 57]. In animal and human studies, KIM-1 has been found to be an early indicator of AKI that compares favorably to a number of conventional biomarkers and tubular enzymes [57, 58]. In the setting of cisplatin exposure as well as renal I/R in rats, Kim-1 was found to have superior sensitivity for detecting AKI than serum creatinine and BUN or urinary NAG, glycosuria, and proteinuria [57]. Adjusted for age, gender, and length of time delay between insult and sampling, a one-unit increase in normalized (to urine creatinine) KIM-1 was associated with a greater than 12-fold increase in the presence of acute tubular necrosis (ATN) [58]. The diagnostic utility of urinary KIM-1 was evaluated for the early detection of postoperative AKI in a prospective study of 90 adults undergoing cardiac surgery. Urinary KIM-1 was measured at five time points for the first 24 h after operation and normalized to the urinary creatinine concentration after cardiac surgery. ROC curves were generated and the areas under the curve (AUCs) compared for performance

URINARY BIOMARKERS FOR THE DETECTION OF RENAL INJURY

85

of biomarkers in detection of postoperative AKI. In this study, 36 patients developed AKI, defined as an increase in SCr of > or ¼0.3 mg/dl within 72 h after surgery. The AUCs for KIM-1 to predict AKI immediately and 3 h after operation were 0.68 and 0.65 [59]. In another study, the diagnostic performance of nine urinary biomarkers of AKI including KIM-1 was evaluated in a cross-sectional comparison of 204 patients with or without clinically documented AKI. In the case of each biomarker, the median urinary concentrations were significantly higher in patients with AKI than in those without AKI. The area under the ROC curve (AUC–ROC) for KIM-1 was 0.95 when the AKI patients were compared to healthy control patients [60]. In studies of kidney transplantation, human KIM-1 protein expression was quantified in renal transplant biopsies by immunohistochemistry and correlated these findings with renal functional indices [61]. Although protocol biopsies showed no detectable tubular injury on histological examination, there was focal positive KIM-1 expression in 28% of the cases. It is thought that this does not reflect a false-positive test but rather the result of superior sensitivity of KIM-1 expression in detecting proximal tubule injury when compared to morphology alone. In this study of renal allografts, KIM-1 expression was detected in 100% of biopsies from patients with deterioration in kidney function and histological changes indicative of tubular damage [61]. KIM-1 expression was significantly correlated with levels of serum creatinine and BUN concentrations and inversely correlated with estimated GFR on the biopsy day. KIM-1 was expressed focally in affected tubules in 92% of kidney biopsies from patients with acute cellular rejection reflecting the epithelial cell injury that results from a severe cellular rejection [61]. Van Timmeren et al. also evaluated the utility of urinary KIM-1 in renal transplant recipients. They looked at a cohort with a median of 6 years posttransplant, and we measured baseline KIM-1 excretion in stored urine samples. Graft loss was monitored over time [62]. The occurrence of graft loss increased with increasing tertiles of KIM-1 excretion. High KIM-1 levels were associated with low creatinine clearance, proteinuria, and high donor age. KIM-1 levels predicted graft loss independent of creatinine clearance, proteinuria, and donor age. These two studies show that monitoring of urinary KIM-1 levels may offer promise as a noninvasive method for monitoring renal allograft function. Most recently, Kim-1 has been qualified by the Food and Drug Administration and European Medicines Agency as a highly sensitive and specific urinary biomarker to monitor drug-induced kidney injury in preclinical studies and on a case-by-case basis in clinical trials [63]. To facilitate the bedside use of Kim-1 as a diagnostic test, a rapid direct immunochromatographic lateral flow 15-min assay for detection of urinary Kim-1 (rat) or

86

MITCHELL H. ROSNER

KIM-1 (human) had been developed as a urine dipstick test. In initial studies, the urinary Kim-1 band intensity using the rat Kim-1 dipstick significantly correlated with levels of Kim-1 as measured by a microbead-based assay, histopathological damage, and immunohistochemical assessment of renal Kim-1 in a dose- and time-dependent manner [63]. Kim-1 was detected following kidney injury induced in rats by cadmium, gentamicin, or bilateral renal ischemia/reperfusion [63]. In humans, the urinary KIM-1 band intensity was significantly greater in urine from patients with AKI than in urine from healthy volunteers [63]. The KIM-1 dipstick also enabled temporal evaluation of kidney injury and recovery in two patients who developed postoperative AKI following cytoreductive surgery for malignant mesothelioma with intraoperative local cisplatin administration [63]. More extensive studies will be needed to confirm the utility of these results and hopefully demonstrate that the Kim-1/KIM-1 dipsticks can provide a sensitive and accurate detection of Kim-1/KIM-1, thereby providing a rapid diagnostic assay for kidney damage and facilitating the rapid and early detection of kidney injury in preclinical and clinical studies. 4.4. NEUTROPHIL GELATINASE-ASSOCIATED LIPOCALIN Human NGAL is a 25 kDa protein initially identified bound to gelatinase in specific granules of the neutrophil. NGAL is synthesized during a narrow window of granulocyte maturation in the bone marrow [64], but also may be induced in epithelial cells in the setting of inflammation or malignancy [65]. Cowland and Borregaard demonstrated varying degrees of NGAL gene expression in a number of other human tissues including the uterus, prostate, salivary gland, lung, trachea, stomach, colon, and kidney [66]. Using cDNA microarray screening techniques, Devarajan and colleagues identified NGAL as one of seven genes whose expression was upregulated >10-fold within the first few hours after ischemic renal injury in a mouse model [67]. Immunohistochemistry studies demonstrated minimal NGAL expression in control mouse kidneys, but marked upregulation in proximal tubules within 3 h of ischemia [67]. Examination of serial kidney sections revealed significant colocalization of NGAL and the proliferative marker PCNA [67]. In addition, it has been reported that NGAL induces the conversion of rat kidney progenitors into tubules and epithelia [68]. As a result, it has been hypothesized that NGAL may play a role in the induction of tubular reepithelialization in the setting of AKI. Increased NGAL levels were readily detected in the urine of rodents following I/R injury and after cisplatin exposure, far preceding changes in serum creatinine as well as the appearance of urinary b2M and NAG [67, 69]. As a result, NGAL has generated much interest as a sensitive early

URINARY BIOMARKERS FOR THE DETECTION OF RENAL INJURY

87

biomarker for diagnosing AKI. A prospective study of pediatric patients undergoing cardiopulmonary bypass for cardiac corrective surgery found urinary NGAL to be a powerful early marker of AKI, preceding any increase in serum creatinine by 1–3 days [70]. A similar study of adult patients showed urinary NGAL levels at 1, 3, and 18 h after cardiac surgery to be significantly higher in patients who went on to develop clinically significant AKI [71]. In the previously described study by Han et al., the AUC for NGAL predicting postcardiac bypass AKI both immediately and at 3 h postsurgery was poor (0.59 and 0.65, respectively) [59]. Similarly, in a study by Wagener et al., the diagnostic utility of urine NGAL to predict AKI defined by change in serum creatinine after cardiac surgery was poor [72]. Areas under the ROC curve for urinary NGAL immediately after and 3, 18, and 24 h later as a predictor for AKI were 0.573, 0.603, 0.611, and 0.584, respectively [72]. Thus, at least in adult cardiac surgery, the predictive value for AKI of NGAL appears to have wide variability. The choice of definition of AKI might, at least in part, account for such variability and was investigated by HaaseFielitz and colleagues [73]. In a prospective study of 100 adult cardiac surgery patients, they assessed the value of postoperative plasma NGAL in predicting AKI according to the degree of severity used for its definition. The predictive value of plasma NGAL varied according to the AKI definition used and was higher for more severe AKI (increase in creatinine >50%: mean AUC–ROC 0.79 0.01) compared to less severe AKI (>25%: mean AUC–ROC 0.65 0.02). The discriminatory ability of NGAL for AKI also increased with increasing RIFLE classes (AUC–ROC R: 0.72, I: 0.79, F: 0.80) or AKIN stages (AUC–ROC 1: 0.75, 2: 0.78, 3: 0.81); P ¼ 0.015. It was highest for the prediction of renal replacement therapy (AUC–ROC: 0.83). Thus, in adult cardiac surgery patients, the predictive value of NGAL increases with grade of AKI. This observation needs to be taken into account when interpreting any future studies of this biomarker. In a different clinical cause of AKI, a retrospective analysis of urine samples from patients with diarrhea-associated hemolytic uremic syndrome revealed that normal urinary NGAL excretion during the early stages of hospitalization had a high negative predictive value of the need for dialysis; however, high urinary NGAL levels were not a reliable predictor of need for dialysis [74]. It should be noted that serum NGAL levels are known to rise in the setting of a number of inflammatory and infective conditions [75, 76]. Further studies are required to determine specificity of urinary NGAL for AKI in the setting of sepsis, a condition frequently associated with clinically significant renal injury. Most recently, two studies have assessed the utility of serum and urine NGAL to assess severity and predict progression of diabetic nephropathy [77, 78]. In both studies, serum and urine NGAL increased in parallel with

88

MITCHELL H. ROSNER

the severity of renal disease, reaching higher levels in patients who manifest diabetic nephropathy. These findings need to be expanded in larger, prospective studies but indicate that NGAL measurement might become a useful and noninvasive tool for the evaluation of renal involvement in diabetic patients and for the diagnosis of incipient diabetic nephropathy. 4.5. INTERLEUKIN-18 IL-18, formerly known as interferon-g-inducing factor, is produced as a 24 kDa inactive precursor that is cleaved by caspase-1 to generate its mature, biologically active form [79–81]. IL-18 has been found to have broad immunomodulatory properties and appears to play a critical role in host defense against a number of infections [80]. In addition, IL-18 activity has been described in a number of inflammatory diseases across a broad range of tissues, including inflammatory arthritis, multiple sclerosis, inflammatory bowel disease, chronic hepatitis, systemic lupus erythematosis, and psoriasis [80]. There are reports indicating that IL-18 is an important mediator in tissue I/R injury. In an I/R model of human myocardium, diminished IL-18 activity through selective caspase-1 inhibition was found to protect against injury and resulted in significant preservation of myocardial contractile force [82]. In preclinical studies, caspase-1 inhibition or the administration of IL-18-neutralizing serum was found to protect against ischemic AKI in mice [83, 84]. Parikh et al. reported increased levels of IL-18 in patients with AKI of varying etiology, especially those with delayed renal allograft function and ischemic ATN [85]. Following transplantation, a rapid decline in urinary IL-18 levels was predictive of a steeper decline in serum creatinine concentrations postoperative days 0–4 [85]. Immunohistochemical staining of renal transplant protocol biopsies revealed constitutive IL-18 expression in the distal tubular epithelium. There was strong positive immunoreactivity in the proximal tubules of patients with acute rejection. There was also strong immunoreactivity in infiltrating leukocytes, and endothelium, suggesting upregulation in the setting of immunopathological reactions [86]. In a study of critically ill adult patients with acute respiratory distress syndrome (ARDS), increased urinary IL-18 was found to be an early marker of AKI, preceding changes in serum creatinine by 1–2 days, and was an independent predictor of death [87]. More recent studies have questioned the diagnostic utility of IL-18 [88, 89]. In the first study, patients developing AKI after percutaneous coronary interventions (presumably due to iodinated contrast-induced renal injury) were no more likely to have rises in urine IL-18 levels than controls not

URINARY BIOMARKERS FOR THE DETECTION OF RENAL INJURY

89

developing AKI [88]. The second study was a single-center prospective observational cohort study in which 100 adult cardiac surgical patients undergoing cardiopulmonary bypass at a tertiary hospital were enrolled [89]. The investigators measured the urinary concentration of IL-18 and creatinine preoperatively, on arrival in the intensive care unit, and 24 h postoperatively, as well as assessing urinary IL-18 concentration and urinary IL18/urinary creatinine ratio in relation to the postoperative development of AKI defined as an increase in serum creatinine of greater than 50% from preoperative to postoperative peak value within 48 h after surgery. Urinary IL-18 was not better than chance in predicting AKI either immediately postoperative (AUC–ROC 0.55) or 24 h after surgery (AUC–ROC 0.53). In this model system, IL-18 correlated with the duration of cardiopulmonary bypass and the changes in IL-18 levels may represent a nonspecific marker of cardiopulmonary bypass-associated systemic inflammation. Further studies will be required to elucidate the role of IL-18 as a biomarker for AKI across a spectrum of etiologies.

4.6. FATTY-ACID BINDING PROTEIN The FAPB are small cytoplasmic proteins abundantly expressed in tissues with an active fatty-acid binding metabolism. Nine different types have been identified with each named for the initial site of identification [90]. The primary function of FABP is the facilitation of long-chain free fatty-acid transport from the plasma membrane to sites for oxidation (mitochondria, peroxisomes) [90, 91]. Increased levels of cystosolic free fatty acids with attendant increased FABP expression may be seen in response to a variety of pathophysiologic tubular stresses [90]. There is evidence that FABP may serve as an endogenous antioxidant, not only binding polyunsaturated fatty acids and protecting them from oxidation but also binding fatty-acid oxidation products, thereby limiting the toxic effects of oxidative intermediates on cellular membranes [92]. Two types of FABP have been identified in the human kidney, liver-type FABP (L-FABP) in the proximal tubule and heart-type FABP (H-FABP) in the distal tubule [92, 93]. H-FABP levels have been found to be a sensitive marker for nephrotoxicantinduced kidney injury in rats [90]. In addition, H-FABP has been studied as a marker of tissue damage resulting from pretransplantation machine perfusion in the preparation of nonheart beating donor (NHBD) organs for kidney transplantation. Higher levels of H-FABP, as well as other biomarkers, in kidney perfusates were determined to be useful adjuncts to routine indicators (donor age, donor medical history, macroscopic appearance, warm ischemic time, and ex vivo perfusion) of suitability of NHBD kidneys for transplantation

90

MITCHELL H. ROSNER

[90, 94–96]. Clinical studies in the utility of H-FABP as a urinary biomarker in more conventional models of AKI are lacking. Urinary L-FABP has been studied extensively in preclinical and clinical models and has been found to be a potential biomarker in a number of pathologic conditions, including CKD, diabetic nephropathy, IgA nephropathy, and contrast nephropathy. Using human L-FABP (hL-FABP) chromosomal transgenic mice, it has been demonstrated that protein-overload nephropathy and unilateral ureteral obstruction, two models of renal interstitial injury, are associated with increased expression and urinary excretion of hL-FABP [96, 97]. In both injury models, attenuation of tubulointerstitial damage was observed in the transgenic mice when compared to wild-type mice, supporting the notion that L-FABP plays a protective role in the setting of increased renal tubular stress [96, 97]. In clinical studies, L-FABP has been advocated as a potential biomarker for monitoring progression of CKD. Kamijo et al. reported increasing L-FABP levels with deterioration of renal function in those with nondiabetic CKD [96, 97]. Further studies in type II diabetics have shown an association between the stage of diabetic nephropathy and urinary L-FAPB levels [98, 99] as well as the potential benefit of renin–angiotensin system blockade in this population, reflected in decreased L-FABP excretion [99]. In addition, Nakamura et al. have reported that urinary L-FABP may serve as a noninvasive biomarker to discriminate between IgA nephropathy and thin basement membrane disease [100] as well as a potential predictive marker for contrast-induced nephropathy [101, 102]. While L-FAPB appears to be an attractive candidate biomarker for a number of renal diseases, additional studies are needed to determine the utility of L-FABP in AKI, especially in the setting of ischemia/reperfusion injury, nephrotoxin exposure, and sepsis. Lastly, urinary L-FABP may be nonspecific for AKI in the setting of acute liver injury. Although Kamijo et al. reported urinary L-FABP levels in patients with liver disease to be similar to levels in healthy volunteers, it is unclear if these patients suffered from acute or chronic liver injury [98]. This will be important to investigate as AKI and acute liver injury commonly cooccur in the critically ill population.

5. Biomarker Development and Implementation As described previously, a host of biomarkers have been identified that show promise in aiding the rapid and sensitive diagnosis of AKI. However, with the exception of NGAL (and less true for KIM-1), most of these biomarkers have not been studied in large-scale prospective confirmatory studies. Furthermore, most studies have focused on AKI in very specific

URINARY BIOMARKERS FOR THE DETECTION OF RENAL INJURY

91

clinical settings (such as postbypass surgery) and these results may not be generalizable. These specific clinical settings are often those were the insults to the kidney are single and well-timed and more complex clinical situations such as sepsis in the intensive care unit need to be studied. Ideally, several biomarkers should be compared using the same samples and candidate biomarkers can then be ranked on their utility based on the AUC–ROC values. Easily reproducible and performed clinical assays need to be developed (as has been done with KIM-1) for the most promising biomarkers and validated at several independent laboratories. Finally, specific performance criteria, such as the exact timing of biomarker appearance relative to rises in serum creatinine and to the injury, cut off values for maximal sensitivity and specificity, and false-positive and false-negative rates will need to be determined. Ideally, studies that gauge the impact of the diagnostic biomarkers on clinical decision making and ultimately, clinical outcomes should be performed. Given the complexity of AKI and the various etiological agents and pathophysiological causes (toxins, hypoperfusion, inflammation, etc.), it is reasonable to expect that each individual biomarker may have varying sensitivities and specificities depending upon the clinical scenario. For example, IL-18 may be particularly good if the cause of AKI was dependent upon an inflammatory stimulus. Thus, a ‘‘panel’’ approach, whereby several biomarkers are concurrently assayed and which cover different pathophysiological aspects of AKI may be critically important. Development of such a panel will require large, well-designed prospective studies comparing multiple biomarkers in the same set of urine/plasma samples. Such studies will allow temporal patterns of biomarker elevation to be established, patterns that may be specific to the mechanism of injury, population of interest, and/or concurrent disease states (diabetes, heart failure, CKD, sepsis, etc.). The cost-effectiveness of these biomarkers in how they will be used to change the care of patients at risk for AKI must be evaluated. These biomarkers will also aid in the development of clinical trials for new pharmacological agents for the treatment and prevention of AKI by allowing researchers to rapidly identify patients at the earliest stage of renal injury.

6. Summary There is a clear need for the rapid and precise diagnosis of AKI, a condition associated with substantial morbidity and mortality. A step-wise approach of redefining the clinical condition has occurred so that all studies can be standardized to the same definition. This has been followed by an intensive search for candidate biomarkers that has yielded some promising

92

MITCHELL H. ROSNER

results. It is anticipated that continued advances in exosome isolation and urinary proteomics will lead to continued discovery of new biomarkers. Ultimately, a panel of informative biomarkers will be developed that can be utilized to screen at-risk patients for the early detection of AKI. This would then allow early therapy with the hope that early intervention may lead to improved outcomes. REFERENCES [1] United States Renal Data System. U.S.R.D.S, Annual Data Report., The National Institutes of Diabetes and Digestive and Kidney Diseases, Bethesda, 2006. [2] S. Uchino, J.A. Kellum, R. Bellomo, et al., Acute renal failure in critically ill patients: a multinational, multicenter study, JAMA 294 (2005) 813–818. [3] R. Bellomo, C. Ronco, J.A. Kellum, R.L. Mehta, P. Palevsky, Acute renal failure— definition, outcome measures, animal models, fluid therapy and information technology needs: the Second International Consensus Conference of the Acute Dialysis Quality Initiative (ADQI) Group, Crit. Care 8 (2004) R204–R212. [4] F. Liano, J. Pascual, Outcomes in acute renal failure, Semin. Nephrol. 18 (1998) 541–550. [5] R.A. Star, Treatment of acute renal failure, Kidney Int. 54 (1998) 1817–1831. [6] J.A. Kellum, N. Levin, C. Bouman, N. Lameire, Developing a consensus classification system for acute renal failure, Curr. Opin. Crit. Care 8 (2002) 509–514. [7] R.L. Mehta, G.M. Chertow, Acute renal failure definitions and classification: time for change? J. Am. Soc. Nephrol. 14 (2003) 2178–2187. [8] D.G. Warnock, Towards a definition and classification of acute kidney injury, J. Am. Soc. Nephrol. 16 (2005) 3149–3150. [9] G.M. Chertow, E. Burdick, M. Honour, J.V. Bonventre, D.W. Bates, Acute kidney injury, mortality, length of stay, and costs in hospitalized patients, J. Am. Soc. Nephrol. 16 (2005) 3365–3370. [10] A. Lassnigg, D. Schmidlin, M. Mouhieddine, et al., Minimal changes of serum creatinine predict prognosis in patients after cardiothoracic surgery: a prospective cohort study, J. Am. Soc. Nephrol. 15 (2004) 1597–1605. [11] R.L. Mehta, J.A. Kellum, S. Shah, et al., AKIN: Acute Kidney Injury Network: report of an initiative, in: AKIN Summit, Amsterdam, 2006. [12] S.M. Moran, B.D. Myers, Course of acute renal failure studied by a model of creatinine kinetics, Kidney Int. 27 (1985) 928–937. [13] V.S. Vaidya, J.V. Bonventre, Mechanistic biomarkers for cytotoxic acute kidney injury, Expert Opin. Drug Metab. Toxicol. 2 (2006) 697–713. [14] R.J. Trof, F. Di Maggio, J. Leemreis, A.B. Groeneveld, Biomarkers of acute renal injury and renal failure, Shock 26 (2006) 245–253. [15] S.G. Emeigh Hart, Assessment of renal injury in vivo, J. Pharmacol. Toxicol. Methods 52 (2005) 30–45. [16] S.L. Goldstein, Pediatric acute kidney injury: it’s time for real progress, Pediatr. Nephrol. 21 (2006) 891–895. [17] S.M. Hewitt, J. Dear, R.A. Star, Discovery of protein biomarkers for renal diseases, J. Am. Soc. Nephrol. 15 (2004) 1677–1689. [18] V. Thongboonkerd, T. Semangoen, S. Chutipongtanate, Enrichment of the basic/cationic urinary proteome using ion exchange chromatography and batch adsorption, J. Proteome Res. 6 (2007) 1209–1214.

URINARY BIOMARKERS FOR THE DETECTION OF RENAL INJURY

93

[19] Q. Fu, C.P. Garnham, S.T. Elliott, D.E. Bovenkamp, J.E. Van Eyk, A robust, streamlined, and reproducible method for proteomic analysis of serum by delipidation, albumin and IgG depletion, and two-dimensional gel electrophoresis, Proteomics 5 (2005) 2656–2664. [20] K. Bjorhall, T. Miliotis, P. Davidsson, Comparison of different depletion strategies for improved resolution in proteomic analysis of human serum samples, Proteomics 5 (2005) 307–317. [21] A.K. Yocum, K. Yu, T. Oe, I.A. Blair, Effect of immunoaffinity depletion of human serum during proteomic investigations, J. Proteome Res. 4 (2005) 1722–1731. [22] T. Liu, W.J. Qian, H.M. Mottaz, et al., Evaluation of multiprotein immunoaffinity subtraction for plasma proteomics and candidate biomarker discovery using mass spectrometry, Mol. Cell. Proteomics 5 (2006) 2167–2174. [23] H. Zhang, X.J. Li, D.B. Martin, R. Aebersold, Identification and quantification of N-linked glycoproteins using hydrazide chemistry, stable isotope labeling and mass spectrometry, Nat. Biotechnol. 21 (2003) 660–666. [24] W.C. Lee, K.H. Lee, Applications of affinity chromatography in proteomics, Anal. Biochem. 324 (2004) 1–10. [25] R. Kange, U. Selditz, M. Granberg, et al., Comparison of different IMAC techniques used for enrichment of phosphorylated peptides, J. Biomol. Tech. 16 (2005) 91–103. [26] T. Pisitkun, R.F. Shen, M.A. Knepper, Identification and proteomic profiling of exosomes in human urine, Proc. Natl. Acad. Sci. USA 101 (2004) 13368–13373. [27] H. Zhou, T. Pisitkun, A. Aponte, et al., Exosomal Fetuin-A identified by proteomics: a novel urinary biomarker for detecting acute kidney injury, Kidney Int. 70 (2006) 1847–1857. [28] B. Holdt-Lehmann, A. Lehmann, G. Korten, H. Nagel, H. Nizze, P. Schuff-Werner, Diagnostic value of urinary alanine aminopeptidase and N-acetyl-beta-D-glucosaminidase in comparison to alpha 1-microglobulin as a marker in evaluating tubular dysfunction in glomerulonephritis patients, Clin. Chim. Acta 297 (2000) 93–102. [29] G. Cressey, D.R. Roberts, C.P. Snowden, Renal tubular injury after infrarenal aortic aneurysm repair, J. Cardiothorac. Vasc. Anesth. 16 (2002) 290–293. [30] B.K. Van Kreel, M.A. Janssen, G. Kootstra, Functional relationship of alpha-glutathioneS-transferase and glutathione-S-transferase activity in machine-preserved non-heart-beating donor kidneys, Transpl. Int. 15 (2002) 546–549. [31] M.T. Bondiou, R. Bourbouze, M. Bernard, F. Percheron, N. Perez-Gonzalez, J.A. Cabezas, Inhibition of A and B N-acetyl-beta-D-glucosaminidase urinary isoenzymes by urea, Clin. Chim. Acta 149 (1985) 67–73. [32] S.G. Emeigh Hart, Assessment of renal injury in vivo, J. Pharmacol. Toxicol. Methods 52 (2005) 30–45. [33] G. D’Amico, C. Bazzi, Urinary protein and enzyme excretion as markers of tubular damage, Curr. Opin. Nephrol. Hypertens. 12 (2003) 639–643. [34] J. Westhuyzen, Z.H. Endre, G. Reece, D.M. Reith, D. Saltissi, T.J. Morgan, Measurement of tubular enzymuria facilitates early detection of acute renal impairment in the intensive care unit, Nephrol. Dial. Transplant. 18 (2003) 543–551. [35] K. Usuda, K. Kono, T. Dote, et al., Urinary biomarkers monitoring for experimental fluoride nephrotoxicity, Arch. Toxicol. 72 (1998) 104–109. [36] A.M. Bernard, A.A. Vyskocil, P. Mahieu, R.R. Lauwerys, Assessment of urinary retinolbinding protein as an index of proximal tubular injury, Clin. Chem. 33 (1987) 775–779. [37] M.W. Wolf, J. Boldt, Kidney specific proteins: markers for detection of renal dysfunction after cardiac surgery? Clin. Res. Cardiol. Suppl. 2 (2007) S103–S107. [38] M.G. Dehne, J. Boldt, D. Heise, A. Sablotzki, G. Hempelmann, Tamm-Horsfall protein, alpha-1- and beta-2-microglobulin as kidney function markers in heart surgery, Anaesthesist 44 (1995) 545–551.

94

MITCHELL H. ROSNER

[39] F. Fernandez, M.D. de Miguel, V. Barrio, J. Mallol, Beta-2-microglobulin as an index of renal function after cardiopulmonary bypass surgery in children, Child Nephrol. Urol. 9 (1998) 326–330. [40] M. Conti, S. Moutereau, M. Zater, et al., Urinary cystatin C as a specific marker of tubular dysfunction, Clin. Chem. Lab. Med. 44 (2006) 288–291. [41] B. Mukhopadhyay, S. Chinchole, V. Lobo, S. Gang, M. Rajapurkar, Enzymuria pattern in early port renal tranplant period: diagnostic usefulness in graft dysfunction, Indian J. Clin. Biochem. 19 (2004) 14–19. [42] C. Bazzi, C. Petrini, V. Rizza, et al., Urinary N-acetyl-beta-glucosaminidase excretion is a marker of tubular cell dysfunction and a predictor of outcome in primary glomerulonephritis, Nephrol. Dial. Transplant. 17 (2002) 1890–1896. [43] H. Ikenaga, H. Suzuki, N. Ishii, H. Itoh, T. Saruta, Enzymuria in non-insulin-dependent diabetic patients: signs of tubular cell dysfunction, Clin. Sci. (Lond.) 84 (1993) 469–475. [44] R. Ascione, C.T. Lloyd, M.J. Underwood, W.J. Gomes, G.D. Angelini, On-pump versus off-pump coronary revascularization: evaluation of renal function, Ann. Thorac. Surg. 68 (1999) 493–498. [45] S.L. Chew, R.L. Lins, R. Daelemans, G.D. Nuyts, M.E. De Broe, Urinary enzymes in acute renal failure, Nephrol. Dial. Transplant. 8 (1993) 507–511. [46] O. Liangos, M.C. Perianayagam, V.S. Vaidya, et al., Urinary N-acetyl-beta-(D)-glucosaminidase activity and kidney injury molecule-1 level are associated with adverse outcomes in acute renal failure, J. Am. Soc. Nephrol. 18 (2007) 904–912. [47] W.K. Han, S.S. Waikar, A. Johnson, et al., Urinary biomarkers in the early diagnosis of acute kidney injury, Kidney Int. 73 (2008) 863–869. [48] M.T. Bondiou, R. Bourbouze, M. Bernard, F. Percheron, N. Perez-Gonzalez, J.A. Cabezas, Inhibition of A and B N-acetyl-beta-D-glucosaminidase urinary isoenzymes by urea, Clin. Chim. Acta 149 (1985) 67–73. [49] T. Ichimura, J.V. Bonventre, V. Bailly, et al., Kidney injury molecule-1 (KIM-1), a putative epithelial cell adhesion molecule containing a novel immunoglobulin domain, is up-regulated in renal cells after injury, J. Biol. Chem. 273 (1998) 4135–4142. [50] R.P. Amin, A.E. Vickers, F. Sistare, et al., Identification of putative gene based markers of renal toxicity, Environ. Health Perspect. 112 (2004) 465–479. [51] V. Bailly, Z. Zhang, W. Meier, R. Cate, M. Sanicola, J.V. Bonventre, Shedding of kidney injury molecule-1, a putative adhesion protein involved in renal regeneration, J. Biol. Chem. 277 (2002) 39739–39748. [52] P.L. Goering, V.S. Vaidya, R.P. Brown, et al., Kidney injury molecule-1 (kim-1) expression in kidney and urine following acute exposure to gentamicin and mercury, Toxicologist 90 (2006) 341. [53] B.D. Humphreys, V.S. Vaidya, R. Samarakoon, D.M. Hentschel, K.M. Park, J.V. Bonventre, Early and sustained expression of kidney injury molecule-1 (Kim-1) after unilateral ureteral obstruction, J. Am. Soc. Nephrol. 16 (2005) 425A. [54] T. Ichimura, C.C. Hung, S.A. Yang, J.L. Stevens, J.V. Bonventre, Kidney injury molecule1: a tissue and urinary biomarker for nephrotoxicant-induced renal injury, Am. J. Physiol. Renal Physiol. 286 (2004) F552–F563. [55] W.C. Prozialeck, V.S. Vaidya, A.M. Johnson, et al., Kidney injury molecule-1 (kim-1) as an early biomarker of cadmium (cd) nephrotoxicity, Toxicologist 90 (2006) 340. [56] V.S. Vaidya, V. Ramirez, N.A. Bobadilla, J.V. Bonventre, A microfluidics based assay to measure Kidney Injury Molecule-1 (Kim-1) in the urine as a biomarker for early diagnosis of acute kidney injury, J. Am. Soc. Nephrol. 16 (2005) 192A.

URINARY BIOMARKERS FOR THE DETECTION OF RENAL INJURY

95

[57] V.S. Vaidya, V. Ramirez, T. Ichimura, N.A. Bobadilla, J.V. Bonventre, Urinary kidney injury molecule-1: a sensitive quantitative biomarker for early detection of kidney tubular injury, Am. J. Physiol. Renal Physiol. 290 (2006) F517–F529. [58] W.K. Han, V. Bailly, R. Abichandani, R. Thadhani, J.V. Bonventre, Kidney Injury Molecule-1 (KIM-1): a novel biomarker for human renal proximal tubule injury, Kidney Int. 62 (2002) 237–244. [59] W.K. Han, G. Wagener, Y. Zhu, S. Wang, H.T. Lee, Urinary biomarkers in the early detection of acute kidney injury after cardiac surgery, Clin. J. Am. Soc. Nephrol. 4 (2009) 873–882. [60] V.S. Vaidya, S.S. Waikar, M.A. Ferguson, et al., Urinary biomarkers for sensitive and specific detection of acute kidney injury in humans, Clin. Transl. Sci. 1 (2008) 200–208. [61] P.L. Zhang, L.I. Rothblum, W.K. Han, et al., Kidney injury molecule-1 expression in transplant biopsies is a sensitive measure of cell injury, Kidney Int. 73 (2008) 608–614. [62] M.M. van Timmeren, V.S. Vaidya, R.M. van Ree, et al., High urinary excretion of kidney injury molecule-1 is an independent predictor of graft loss in renal transplant recipients, Transplantation 84 (2007) 1625–1630. [63] V.S. Vaidya, G.M. Ford, S.S. Waikar, et al., A rapid urine test for early detection of kidney injury, Kidney Int. (2009) epub 22 April 2009; doi:10.1038/ki.2009.96. [64] N. Borregaard, M. Sehested, B.S. Nielsen, H. Sengelov, L. Kjeldsen, Biosynthesis of granule proteins in normal human bone marrow cells. Gelatinase is a marker of terminal neutrophil differentiation, Blood 85 (1995) 812–817. [65] B.S. Nielsen, N. Borregaard, J.R. Bundgaard, S. Timshel, M. Sehested, L. Kjeldsen, Induction of NGAL synthesis in epithelial cells of human colorectal neoplasia and inflammatory bowel diseases, Gut 38 (1996) 414–420. [66] J.B. Cowland, N. Borregaard, Molecular characterization and pattern of tissue expression of the gene for neutrophil gelatinase-associated lipocalin from humans, Genomics 45 (1997) 17–23. [67] J. Mishra, Q. Ma, A. Prada, et al., Identification of neutrophil gelatinase-associated lipocalin as a novel early urinary biomarker for ischemic renal injury, J. Am. Soc. Nephrol. 14 (2003) 2534–2543. [68] J. Yang, D. Goetz, J.Y. Li, et al., An iron delivery pathway mediated by a lipocalin, Mol. Cell 10 (2002) 1045–1056. [69] J. Mishra, K. Mori, Q. Ma, C. Kelly, J. Barasch, P. Devarajan, Neutrophil gelatinaseassociated lipocalin: a novel early urinary biomarker for cisplatin nephrotoxicity, Am. J. Nephrol. 24 (2004) 307–315. [70] J. Mishra, C. Dent, R. Tarabishi, et al., Neutrophil gelatinase-associated lipocalin (NGAL) as a biomarker for acute renal injury after cardiac surgery, Lancet 365 (2005) 1231–1238. [71] G. Wagener, M. Jan, M. Kim, et al., Association between increases in urinary neutrophil gelatinase-associated lipocalin and acute renal dysfunction after adult cardiac surgery, Anesthesiology 105 (2006) 485–491. [72] G. Wagener, G. Gubitosa, S. Wang, N. Borregaard, M. Kim, H.T. Lee, Urinary neutrophil gelatinase-associated lipocalin and acute kidney injury after cardiac surgery, Am. J. Kidney Dis. 52 (2008) 425–433. [73] A. Haase-Fielitz, R. Bellomo, P. Devarajan, et al., The predictive performance of plasma neutrophil gelatinase-associated lipocalin increases with grade of acute kidney injury, Nephrol. Dial. Transplant. (2009) Epub ahead of print. [74] H. Trachtman, E. Christen, A. Cnaan, et al., Urinary neutrophil gelatinase-associated lipocalcin in DþHUS: a novel marker of renal injury, Pediatr. Nephrol. 21 (2006) 989–994.

96

MITCHELL H. ROSNER

[75] S. Ohlsson, J. Wieslander, M. Segelmark, Increased circulating levels of proteinase 3 in patients with anti-neutrophilic cytoplasmic autoantibodies-associated systemic vasculitis in remission, Clin. Exp. Immunol. 131 (2003) 528–535. [76] S.Y. Xu, K. Pauksen, P. Venge, Serum measurements of human neutrophil lipocalin (HNL) discriminate between acute bacterial and viral infections, Scand. J. Clin. Lab. Invest. 55 (1995) 125–131. [77] D. Bolignano, A. Lacquaniti, G. Coppolino, et al., Neutrophil gelatinase-associated lipocalin as an early biomarker of nephropathy in diabetic patients, Kidney Blood Press. Res. 32 (2009) 91–98. [78] Y.H. Yang, X.J. He, S.R. Chen, L. Wang, E.M. Li, L.Y. Xu, Changes of serum and urine neutrophil gelatinase-associated lipocalin in type 2-diabetic patients with nephropathy: one year observational follow-up study, Endocrine (2009) Epub ahead of print. [79] T. Ghayur, S. Banerjee, M. Hugunin, et al., Caspase-1 processes IFN-gamma-inducing factor and regulates LPS-induced IFN-gamma production, Nature 386 (1997) 619–623. [80] J.A. Gracie, S.E. Robertson, I.B. McInnes, Interleukin-18, J. Leukoc. Biol. 73 (2003) 213–224. [81] Y. Gu, K. Kuida, H. Tsutsui, et al., Activation of interferon-gamma inducing factor mediated by interleukin-1beta converting enzyme, Science 275 (1997) 206–209. [82] B.J. Pomerantz, L.L. Reznikov, A.H. Harken, C.A. Dinarello, Inhibition of caspase 1 reduces human myocardial ischemic dysfunction via inhibition of IL-18 and IL-1beta, Proc. Natl. Acad. Sci. USA 98 (2001) 2871–2876. [83] V.Y. Melnikov, T. Ecder, G. Fantuzzi, et al., Impaired IL-18 processing protects caspase1-deficient mice from ischemic acute renal failure, J. Clin. Invest. 107 (2001) 1145–1152. [84] V.Y. Melnikov, S. Faubel, B. Siegmund, M.S. Lucia, D. Ljubanovic, C.L. Edelstein, Neutrophil-independent mechanisms of caspase-1- and IL-18-mediated ischemic acute tubular necrosis in mice, J. Clin. Invest. 110 (2002) 1083–1091. [85] C.R. Parikh, A. Jani, V.Y. Melnikov, S. Faubel, C.L. Edelstein, Urinary interleukin-18 is a marker of human acute tubular necrosis, Am. J. Kidney Dis. 43 (2004) 405–414. [86] I. Striz, E. Krasna, E. Honsova, et al., Interleukin 18 (IL-18) upregulation in acute rejection of kidney allograft, Immunol. Lett. 99 (2005) 30–35. [87] C.R. Parikh, E. Abraham, M. Ancukiewicz, C.L. Edelstein, Urine IL-18 is an early diagnostic marker for acute kidney injury and predicts mortality in the intensive care unit, J. Am. Soc. Nephrol. 16 (2005) 3046–3052. [88] C.B. Bulent Gul, M. Gullulu, B. Oral, et al., Urinary IL-18: a marker of contrast-induced nephropathy following percutaneous coronary intervention? Clin. Biochem. 41 (2008) 544–547. [89] M. Haase, R. Bellomo, D. Story, P. Davenport, A. Haase-Fielitz, Urinary interleukin-18 does not predict acute kidney injury after adult cardiac surgery: a prospective observational cohort study, Crit. Care 12 (2008) R96. [90] M.M. Pelsers, W.T. Hermens, J.F. Glatz, Fatty acid-binding proteins as plasma markers of tissue injury, Clin. Chim. Acta 352 (2005) 15–35. [91] Y. Oyama, T. Takeda, H. Hama, et al., Evidence for megalin-mediated proximal tubular uptake of L-FABP, a carrier of potentially nephrotoxic molecules, Lab. Invest. 85 (2005) 522–531. [92] B.A. Ek-Von Mentzer, F. Zhang, J.A Hamilton, Binding of 13-HODE and 15-HETE to phospholipid bilayers, albumin, and intracellular fatty acid binding proteins. Implications for transmembrane and intracellular transport and for protection from lipid peroxidation, J. Biol. Chem. 276 (2001) 15575–15580.

URINARY BIOMARKERS FOR THE DETECTION OF RENAL INJURY

97

[93] R.G. Maatman, E.M. van de Westerlo, T.H. van Kuppevelt, J.H. Veerkamp, Molecular identification of the liver- and the heart-type fatty acid-binding proteins in human and rat kidney. Use of the reverse transcriptase polymerase chain reaction, Biochem. J. 288 (Pt. 1) (1992) 285–290. [94] R.G. Maatman, T.H. Van Kuppevelt, J.H. Veerkamp, Two types of fatty acid-binding protein in human kidney. Isolation, characterization and localization, Biochem. J. 273 (Pt. 3) (1991) 759–766. [95] M.A. Gok, M. Pelsers, J.F. Glatz, et al., Use of two biomarkers of renal ischemia to assess machine-perfused non-heart-beating donor kidneys, Clin. Chem. 49 (2003) 172–175. [96] M.A. Gok, M. Pelzers, J.F. Glatz, et al., Do tissue damage biomarkers used to assess machine-perfused NHBD kidneys predict long-term renal function post-transplant? Clin. Chim. Acta 338 (2003) 33–43. [97] A. Kamijo, T. Sugaya, A. Hikawa, et al., Urinary excretion of fatty acid-binding protein reflects stress overload on the proximal tubules, Am. J. Pathol. 165 (2004) 1243–1255. [98] A. Kamijo, T. Sugaya, A. Hikawa, et al., Urinary liver-type fatty acid binding protein as a useful biomarker in chronic kidney disease, Mol. Cell. Biochem. 284 (2006) 175–182. [99] K. Suzuki, T. Babazono, H. Murata, Y. Iwamoto, Clinical significance of urinary livertype fatty acid-binding protein in patients with diabetic nephropathy, Diabetes Care 28 (2005) 2038–2039. [100] T. Nakamura, T. Sugaya, H. Koide, Angiotensin II receptor antagonist reduces urinary liver-type fatty acid-binding protein levels in patients with diabetic nephropathy and chronic renal failure, Diabetologia 50 (2007) 490–492. [101] T. Nakamura, T. Sugaya, I. Ebihara, H. Koide, Urinary liver-type fatty acid-binding protein: discrimination between IgA nephropathy and thin basement membrane nephropathy, Am. J. Nephrol. 25 (2005) 447–450. [102] T. Nakamura, T. Sugaya, K. Node, Y. Ueda, H. Koide, Urinary excretion of liver-type fatty acid-binding protein in contrast medium-induced nephropathy, Am. J. Kidney Dis. 47 (2006) 439–444.

This page intentionally left blank

ADVANCES IN CLINICAL CHEMISTRY, VOL. 49

BIOMARKERS OF BONE AND MINERAL METABOLISM FOLLOWING BONE MARROW TRANSPLANTATION Ki Hyun Baek and Moo Il Kang1 Department of Internal Medicine, The Catholic University of Korea, College of Medicine, 137–701 Seoul, Korea

1. 2. 3. 4.

5. 6. 7. 8. 9.

Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Clinical Features of BMT-Related Bone Loss. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Changes in Bone-Turnover Markers After BMT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Normal Bone Modeling and Remodeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2. Biochemical Markers of Bone Turnover . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3. Biochemical Markers of Bone Turnover After BMT . . . . . . . . . . . . . . . . . . . . . . . Calcium, Parathyroid Hormone (PTH), and Vitamin D . . . . . . . . . . . . . . . . . . . . . . . . . Sex Hormones. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RANKL and Osteoprotegerin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cytokines and Growth Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

99 100 100 102 102 104 106 109 110 112 114 115 116

1. Abstract The loss of bone mass often occurs after patients undergo bone marrow transplantation (BMT). The rapid impairment of bone formation and the increase in bone resorption, as mirrored by the biochemical markers of bone turnover, might play a role in this bone loss, and especially during the immediate post-BMT period. The possible direct causes for this paradoxical uncoupling are exposure to immunosuppressants, hypogonadism, the changes of cytokines, the changes of the bone growth factors, and the damage to the osteoprogenitor cells because of myeloablative therapy. In this chapter, we discuss the general aspects of post-BMT bone loss with a 1

Corresponding author: Moo Il Kang, e-mail: [email protected] 99

0065-2423/09 $35.00 DOI: 10.1016/S0065-2423(09)49005-X

Copyright 2009, Elsevier Inc. All rights reserved.

100

BAEK AND KANG

peculiar focus on the remodeling imbalance of bone and its relation to the use of immunosuppressants and the changes of sex hormones, growth factors, and cytokines.

2. Introduction Bone marrow transplantation (BMT) is the treatment of choice for many hematological diseases and it is also performed for treating other advanced or relapsed cancers and nonmalignant disorders. BMT is currently the most common form of transplantation and the number of long-term survivors has remarkably increased over the recent decades. BMT is frequently complicated by endocrine abnormalities (diabetes mellitus, thyroid dysfunction, hypogonadism, hypopituitarism, growth retardation, adrenal insufficiency, etc.), and the loss of bone mass has also been well documented as a sequela of BMT [1–5]. Multiple mechanisms are responsible for the altered bone metabolism caused by BMT and these mechanisms are not completely understood [6]. The major risk factors for transplant-related osteoporosis include the underlying disease itself; the myeloablative conditioning regimens; the drastic cytokine changes at the time of transplantation; exposure to longlasting posttransplant immunosuppressants; hypogonadism; immobilization; and decreased kidney, liver, and bowel function; these result in the reduced intake and altered homeostasis of calcium and vitamin D [7, 8]. In addition, the osteogenic potential of the bone marrow stromal cells from the recipient is lower after BMT [9, 10]. For adults undergoing allogeneic BMT, the mean lumbar spine and hip BMDs show the rapidest decline during the first 6 months after transplantation [1–3, 5]. This loss of bone mass might be caused by an imbalance between the increased bone resorption and the reduced bone formation. This chapter describes the general aspects of post-BMT bone loss and the characteristic changes of bone remodeling which is an important biomarker of bone and mineral metabolism. This review will also discuss the factors that contribute to the unique bone remodeling that occurs after BMT.

3. Clinical Features of BMT-Related Bone Loss Patients awaiting BMT are exposed to many factors that may influence their bone and mineral metabolism. Induction/consolidation chemotherapy, conditioning regimens (which may involve total body irradiation, TBI), and immobility may all contribute to altered bone and mineral metabolism. For a group of patients who were studied prior to transplantation, the majority of the

BIOMARKERS FOLLOWING BONE MARROW TRANSPLANTATION

101

patients (72%) showed normal bone mineral density (BMD), whereas a lower BMD (osteopenia in 24% and osteoporosis in 4%) was observed in the patients who received high-dose chemotherapy as compared with those who received no chemotherapy or only hydroxyurea [1]. However, the extent of attenuation in the BMD is generally small in the BMT recipients compared to that in the recipients of solid organ transplantation. The subjects undergoing solid organ transplantation are generally older and they have a longer duration of disease or a longer duration of treatment before transplantation [11–13]. Many previous studies have described the features of bone loss consequent to BMT. The results from cross-sectional studies have demonstrated that the prevalence of low bone mass (T-scores <1.0) ranged 28–52% at the lumbar spine or femoral neck at various time points after allogeneic BMT [14–16]. After a median period of 3 years, a significant decrease in BMD was detected at all the skeletal sites, including the lumbar spine and femoral neck, when compared with the age- and gender-matched controls [10]. The prospective studies on hematopoietic stem cell recipients have documented that the mean rate of bone loss during the first year was 2–7% at the lumbar spine and 6–12% at the femoral neck [1–4, 9, 17, 18]. The temporal pattern of bone loss showed that a dramatic bone loss preferentially occurred during the first 6 months after the graft [1–3, 5]. The majority of patients lose more bone from the femoral neck, which is rich in cortical bone, than from the spine, which is rich in trabecular bone, during the first posttransplant year [1–4]. The losses from the lumbar spine may start to recover 6–12 months after BMT [2, 19, 20] and the lumbar spine BMD returned to the baseline level at month 48 [20, 21]. However, the losses from the proximal femur are not regained. Although the bone loss at the femoral neck is attenuated by the end of the first year, the bone losses continue with the lowest BMD levels being observed at 24 months after transplantation [20, 21] and the BMD did not return to baseline until month 48 [21]. At a median of 75 months after BMT, the decrease in the T‐score less than 1 SD was documented in 26% and 41% of the patients at the lumbar spine and the femoral neck, respectively [22]. Autologous BMT differs from an allotransplant in that autologous transplant patients may be older, and they do not require immunosuppressive therapy, including prednisone and cyclosporine, after transplantation. Gandhi et al. [23] found a similar trend in the bone loss following autologous BMT, although the magnitude of the decrease was small compared to allogeneic BMT. The BMD decreased by 1% at the lumbar spine and 4% at the femoral neck at 1 year after autologous BMT. Additional cross-sectional analysis in the long-term survivors following autologous BMT (the BMD was measured at a median of 4.2 years) also found reduced BMD levels compared with age-matched healthy control subjects at the femoral neck, but not at the lumbar spine [24].

102

BAEK AND KANG

The incidence of bone fractures after allo- or auto-BMT has not been well characterized. In one study, nontraumatic fractures occurred in 10.6% of the patients within 3 years of allogeneic transplantation [19]. The same study also showed the decrease in height between pretransplantation and 12 months posttransplantation. In a study on autologous transplantation, the first short-term data did not indicate a significantly increased risk of fracture [23]. However, no data are available on nonvertebral fractures and the long-term risk of vertebral fractures in patients who undergo both types of BMT. Considering that the BMD can predict the future risk of fracture [25, 26], and approximately 10% BMD loss at the proximal femur persisting 4–6 year after BMT [21], this might suggest a two- to threefold increase in the risk for hip fracture [27]. Because increased bone resorption contributes to a considerable degree to the bone loss after BMT, it is reasonable to use antiresorptive agents such as bisphosphonates for preventing post-BMT bone loss. There have been several recent clinical trials that have examined the efficacy of antiresorptive therapy to prevent post-BMT bone loss. The patients received bisphosphonate pamidronate for 1 year beginning before they started conditioning and their BMD was compared with that of the control groups who received estrogen or testosterone replacement therapy only. Pamidronate markedly reduced bone loss, but it did not completely prevent bone loss especially at the femoral sites [28, 29]. Other investigators have reported the beneficial effect of more potent bisphosphonates, zoledronic acid [30, 31], or risedronate [32], even at the hip area. However, it is not clear from these studies whether potent bisphosphonates would also prevent the rapid early bone loss from the femoral neck because these studies started therapy in the patients who had undergone BMT at least several months to years before they took bisphosphonates. Additional studies are needed that will compare the efficacy of more potent bisphosphonates, and particularly when this treatment is commenced immediately before BMT. Sex hormone replacement therapy, vitamin D, or calcitonin has also been shown to be ineffective to prevent post-BMT bone loss [2, 28].

4. Changes in Bone-Turnover Markers After BMT 4.1. NORMAL BONE MODELING AND REMODELING Bone constantly undergoes modeling during life to help it adapt to changing biomechanical forces. Bone also undergoes remodeling to remove old, microdamaged bone and replace it with new, mechanically stronger bone to preserve the strength of bone.

BIOMARKERS FOLLOWING BONE MARROW TRANSPLANTATION

103

Modeling is the process by which bones change their overall shape in response to physiologic influences or mechanical forces, and this leads to gradual adjustment of the skeleton to the forces that it encounters. During bone modeling, bone formation and resorption are not tightly coupled. New bone can be deposited without previous bone resorption, and bone resorption without subsequent bone formation is also possible. By modeling, it is possible to excavate a marrow cavity during growth and establish the bone’s cortical and trabecular architecture [33]. Bone remodeling is the process by which bone is renewed to maintain bone strength and mineral homeostasis. In bone remodeling, the resorption of a volume of bone occurs at focally discrete points on the bone’s inner surface (the endocortical, intracortical, and trabecular surfaces), and this is followed by the deposition of a comparable volume of new bone. The bone remodeling unit is comprised of tightly coupled groups of osteoclasts and osteoblasts that sequentially carry out resorption of old bone and formation of new bone. The resorption phase of the remodeling cycle removes damaged bones (e.g., microcracks), and the formation phase restores the structure to prevent accumulation of bone microdamage. Provided that the cellular machinery is intact and it functions normally, equal volumes of bone are both removed and formed in this process. The resorption phase lasts 2–3 weeks and the formation phase lasts 2–3 months. Bone formation may be coupled with bone resorption by the products produced from the resorbed matrix and from the osteoclast themselves [34]. The newly deposited osteoid undergoes rapid primary mineralization as it is laid down, and then secondary mineralization (enlargement of the deposited calcium hydroxyapatite-like crystals) occurs during the next 12 months. It is largely unknown what signal initiates osteoclastic bone resorption in the bone remodeling process. However, the osteocyte is regarded as another participant in bone remodeling. The signals produced by the deformation or death of osteocytes instruct the osteoclasts to be formed and then they resorb bone [35]. The apoptotic death of osteocytes induced by an estrogen deficiency, corticosteroid therapy, advancing age, or damage to bone probably heralds the presence of damage and its location, and the initiation of targeted remodeling [36]. During growth, the balance between resorption and formation in the bone remodeling unit is positive, so that each remodeling event adds a small moiety of bone, and this leads to bone growth. In adults, bone formation by osteoblasts declines in the bone remodeling unit. When bone formation is less than the prior bone resorption, each bone remodeling cycle removes a small moiety of bone from the skeleton and this results in net bone loss. The magnitude of the positive or negative balance in each bone remodeling unit during growth or aging is small. The rate of remodeling also plays an

104

BAEK AND KANG

important role in determining the net gain or loss of bone. Estrogen deficiency induces a rapid rate of remodeling and it increases the volume of bone resorption while it decreases the volume of bone formation in each bone remodeling unit. The combination of a rapid rate of remodeling and an increased imbalance in the bone remodeling unit accelerates bone loss. This explains the cause of many cases of pathologic bone loss.

4.2. BIOCHEMICAL MARKERS OF BONE TURNOVER Biochemical markers of bone turnover are broadly divided into two categories: (i) the markers of bone resorption and (ii) the markers of bone formation. In the usual clinical circumstances, the two arms of bone remodeling are coupled and they change in parallel. The bone resorption markers reflect osteoclast activity and they are, for the most part, degradation products of type I collagen. Osteoclast activity can also be monitored by the serum levels of tartrate-resistant acid phosphatase and cathepsin K [37], although the use of these markers is more limited. The bone formation markers reflect osteoblast activity and they are the byproducts of collagen synthesis, matrix proteins, or osteoblastic enzymes (Table 1). In several previous studies, bone loss has been shown to correlate with the markers of bone turnover, although this finding is not universal. Most of the studies have indicated that there is a relationship between high levels of boneturnover markers and increased bone loss. However, these bone-turnover markers are not sufficiently sensitive to be used to predict bone loss in an individual patient [38]. While the biochemical markers of bone turnover may be able to predict bone loss and hence the risk of fracture, they may also predict the risk of fracture independently of the BMD. Women with increased levels of bone resorption markers (but not bone formation markers) had a twofold relative risk of hip fractures and this is independent of their bone density and physical performance [39, 40]. Combining a bone-turnover marker with BMD showed an additive effect on the risk of fracture [41]. High bone turnover is associated with an increased risk of fracture because with high bone turnover, more densely mineralized bone is replaced with less densely mineralized bone and the resorption sites remain temporarily unfilled and this predisposes bone to microdamage. In addition, increased remodeling impairs the isomerization and maturation of collagen, which causes the bone to become fragile [36]. Bone markers can also be used to monitor the efficacy of antiresorptive therapy, such as hormone-replacement therapy and bisphosphonate therapy, in individual patients [42, 43].

TABLE 1 BIOCHEMICAL MARKERS OF BONE TURNOVER Bone formation Marker 1. Byproduct of collagen synthesis Procollagen type I N-terminal propeptide (PINP) Procollagen type I C-terminal propeptide (PICP) 2. Matrix protein Osteocalcin (OC) 3. Osteoblast enzyme Bone-specific alkaline phosphatase (BALP) Total alkaline phosphatase (ALP)

Bone resorption Clinical source

Serum Serum

Serum Serum Serum

Marker 1. Type I collagen breakdown product Amino-terminal crosslinking telopeptides of type I collagen (generated by cathepsin K, NTX) Carboxy-terminal crosslinking telopeptides of type I collagen (generated by cathepsin K, CTX) Large carboxy-terminal crosslinking telopeptides of type I collagen (generated by matrix metalloproteinase, ICTP) 2. Collagen degradation product Pyridinoline (PYD) Deoxypyridinoline (DPD) Hydroxyproline 3. Osteoclast enzyme Tartrate-resistant acid phosphatase (TRACP) Cathepsin K

Clinical source

Serum or urine Serum or urine Serum

Urine Urine Urine Serum Serum

106

BAEK AND KANG

4.3. BIOCHEMICAL MARKERS OF BONE TURNOVER AFTER BMT About 10 studies have reported observing the changes of bone-turnover markers before and after BMT; more than half of these studies are prospective studies (Table 2). The bone resorption markers increased immediately after BMT and they continued to increase for up to 6 months. The markers of bone formation decreased until 1 month after BMT. This level then increased transiently at 3 and 6 months, and then it returned to the basal level after 1 year (Fig. 1) [2, 17, 18, 44]. For adults who undergo allogeneic BMT, the mean lumbar spine and proximal femur BMDs show the rapidest decline during the first 6 months after the graft [1–3, 5]. This loss of bone mass might result from an imbalance between the increased bone resorption and reduced bone formation, namely, biochemical uncoupling. Since glucocorticoids have been known to suppress bone formation and increase bone resorption, they appear to be primarily responsible for the observed uncoupling. Glucocorticoid-induced osteoporosis is characterized by direct and profound reductions in bone formation [45]. Glucocorticoids lower the number of osteoblasts by decreasing their replication and differentiation, and glucocorticoids shorten osteoblasts’ lifespan by causing apoptosis of osteoblasts and osteocytes. The direct and indirect effects of glucocorticoids on bone resorption are less profound than their negative effects on bone formation. Early during the course of their administration, glucocorticoids may increase bone resorption. However, with chronic longterm use, glucocorticoid administration is associated with decreased bone resorption, and the patients exposed to glucocorticoids eventually develop a state of decreased bone remodeling [46, 47]. Although we and other research groups have repeatedly found that both the progressive increase in bone resorption and the decreased bone formation during the post-BMT period are correlated with the dose of steroid [4, 44, 48], glucocorticoids appear to be only one of the contributing factors, especially from the view point of the bone resorption. In addition, glucocorticoids predominately reduce the BMD at trabecular sites, while most of the BMT recipients show more bone loss at the femoral neck, which is a site rich in cortical bone. This may be more indirect evidence that glucocorticoid is not a sole player, and other factors also contribute to the paradoxical uncoupling of bone turnover and bone loss. Other immunosuppressive therapies, including cyclosporine and tacrolimus, may also influence bone remodeling during the post-BMT period. In rodent models, cyclosporine administration resulted in significant losses of trabecular bone and marked increases in bone resorption and formation [49–51]. The results of animal studies suggest that cyclosporine may be responsible for the high-turnover aspects of post-BMT bone disease. For

TABLE 2 BIOCHEMICAL MARKERS OF BONE TURNOVER IN STUDIES ON POST-BMT BONE AND MINERAL METABOLISM

Study

Type of transplantation

Type of study

No. of patients

Carlson et al. [86]

10 allo- and 14 auto-

Prospective

24

Schulte et al. [1]

Allo-

Prospective

81

Va¨lima¨ki et al. [2, 22]

Allo-

Prospective

44

Massenkeil et al. [5]

Allo-

Prospective

67

Gandhi et al. [23]

6 allo- and 38 auto-

Prospective

44

Baek et al. [18, 44]

Allo-

Prospective

36

Kerschan-Schindl et al. [15] Castan˜eda et al. [16]

Allo-

Cross-sectional

22

14 allo- and 13 auto-

Cross-sectional

27

Kauppila et al. [58]

Allo-

Cross-sectional

25

Follow-up after HSCT (BMD, months) Before, 0, W1, W2, W3, and W12 0, 6, 12, and 24 0, 6, 12, and median 75 (range 54–96) 0, 6, and 12 Before, 3, 6, 12, and 24 Before and 12

Median 73 (range 65–207) Mean 34 (range 7–158) Mean 36 (range 12–120)

Bone-turnover markers OC, BALP, ICTP BALP, NTX, and DPD BALP, PICP, PINP, and ICTP BALP, DPD, and PYD BALP and OC OC and ICTP

OC and CTX PICP, ICTP, and TRAP BALP, PINP, and ICTP

Measurement of BTM Before, 0, W1, W2, W3, and W12 Before, D14, and D28 Before, W3, W6, M3, M6, M12, and M75 Before, weekly to 12 weeks, M6, and M12 Before, M1, and M3 Before, W1, W2, W3, W4, M3, M6, and M12

108

BAEK AND KANG 18

12 Osteocalcin ICTP Creatinine

16

a

12

8 b b

10

6 8 6

Creatinine (mg/dl)

Osteocalcin (ng/ml), ICTP (ng/ml)

10

b

14

4

4

a

b 2

2 0

0 Pre-BMT

1 week

2 weeks

3 weeks

4 weeks 3 months 6 months 12 months

FIG. 1. The changes in the serum bone-turnover markers before and after allogenic BMT. The data are reported as mean values 95% CI. a, p < 0.01; b, p < 0.05 versus the basal value. Adapted from Ref. [44].

BMT patients, the duration of cyclosporine therapy has shown correlation with the loss of BMD [4, 19]. Hypogonadism by any cause results in a rapid increase in the levels of bone-turnover markers. Ovarian insufficiency develops in up to 92–100% of women soon after TBI and high-dose chemotherapy [7]. Once the estradiol levels decrease, the markers of bone resorption and formation increase [52]. Our observations showed that biochemical uncoupling persists until about 1–2 months after BMT, and this is followed by a high remodeling state until 6 months (Fig. 1) [18, 44]. This high bone-turnover state appears as a result of the lack of enough sex hormone and the use of cyclosporine. There are several possible causes for the decreased bone formation that continues about 1–2 months during the immediate post-BMT period. The use of glucocorticoids, as described earlier, might be the first explanation for the decreased osteoblast activity. During the immediate period after BMT, the doses of glucocorticoids are generally high enough to profoundly suppress bone formation. According to our observations, the mean daily dose of prednisolone was 30.4 mg during the first month, while it decreased to 14 mg by 3 month after BMT [44]. A second explanation is that a substantial number of

BIOMARKERS FOLLOWING BONE MARROW TRANSPLANTATION

109

osteoprogenitor cells are damaged by the myeloablative therapy. The pretreatment, including the TBI and high-dose chemotherapy, causes direct damage to the osteoprogenitor cells [53, 54], and the differentiation of the bone marrow stromal cells into osteoblasts is impaired after BMT [9, 10], A third possibility is that a marked cytokine release at the time of transplantation might inhibit the function of osteoblasts. IL-1 and TNF-a not only stimulate bone resorption, but they also inhibit bone formation [55, 56]. Changes in the levels of these proinflammatory cytokines have been well documented and these levels have been reported to increase immediately after BMT [57]. The antiresorptive or bone-forming agents usually suppress or increase bone remodeling, respectively. This has been demonstrated in many studies that examined the efficacy of antiosteoporotic drugs on the prevention or treatment of osteoporosis. In fact, the bone resorption and bone formation markers were also significantly reduced by pamidronate therapy in allogeneic BMT recipients [28].

5. Calcium, Parathyroid Hormone (PTH), and Vitamin D PTH and 1,25(OH)2VitD3 are the principal regulators of calcium homeostasis for most vertebrates, including humans. Massenkeil et al. [5] prospectively studied 67 patients after allogeneic transplantation, and they reported a prolonged vitamin D deficiency for more than 6 months in all patients. The pretransplant values were reached after 1 year. In their study, the PTH values increased slightly and the calcium concentration remained grossly unchanged after BMT. Schulte et al. [1] also pointed out that the serum 25(OH)D levels were low even before BMT and they were further decreased 4 weeks after BMT when the PTH levels rose, and this happened despite vitamin D supplementation. The PTH levels were not significantly altered in the autologous HSCT recipients, although they were in the upper normal range [23]. The vitamin D status and the PTH levels were grossly normal in the long-term follow-up, cross-sectional studies [15, 16, 58], although one group demonstrated a long-lasting vitamin D deficiency beyond a median of 75 months after BMT [22]. Immobilization of hospitalized patients is known to lead to rapid changes in calcium metabolism along with decreased vitamin D levels [59]. Due to isolation, multiple infusions, general weakness, and infections, the physical activity of patients is reduced during high-dose chemotherapy and bone marrow aplasia in BMT recipients. Nutritional restrictions, the generally recommended UV protection and gastrointestinal disturbances may lead to prolonged vitamin D deficiency even when patients have left the hospital. Normal bone metabolism depends on adequate serum levels of vitamin D.

110

BAEK AND KANG

The compensatory response to vitamin D deficiency is the stimulation of PTH secretion, and this secondary hyperparathyroidism could lead to an increase in bone turnover and bone loss [60]. However, when compared to the solid organ transplantation recipients [61], BMT recipients have higher levels of 25(OH)D [15, 16]. The reason for this may be that BMT recipients are generally younger, more mobile and less deconditioned than the solid organ transplant recipients. In addition, glucocorticoids inhibit calcium absorption from the gut by opposing the actions of vitamin D and by decreasing the expression of specific calcium channels in the duodenum [62]. Renal tubular calcium reabsorption also is inhibited by glucocorticoids [63]. As a consequence of these effects, secondary hyperparathyroidism could also exist in the context of glucocorticoid use. In summary, hypovitaminosis D and/or the use of glucocorticoids theoretically could provoke a state of PTH excess after BMT. However, many of the BMT patients do not exhibit serum levels of PTH that are frankly elevated. These observations indicate that hyperparathyroidism does not play a central role in the development of the skeletal manifestations of BMT patients. Also, there is only sparse and inconsistent data on the temporal sequence of the PTH changes in BMT patients. Further studies are needed to illuminate in detail the regulation of PTH release in the setting of BMT.

6. Sex Hormones Estrogen plays a pivotal role in maintaining bone mass, and women undergo significant bone loss during menopause [64]. It is also recognized that while estrogen plays a dominant role in regulating bone metabolism, testosterone also has some antiresorptive effects, and testosterone clearly helps maintain bone formation [65]. We previously studied the acute effect of BMT on changes of the pituitary– gonad axis during the early post-BMT period (Fig. 2) [44]. In this study, the levels of luteinizing hormone (LH) and follicle-stimulating hormone (FSH) showed a rapid decrease immediately after BMT in the female BMT recipients, and thereafter their levels increased to the postmenopausal levels by 3 months. It is likely that transient gonadotropin insufficiency occurs during the early post-BMT period due to the combination of radiation treatment and the cytotoxic chemotherapy. Glucocorticoids also have a suppressive effect on the hypothalamic–gonad axis [66]. The estradiol levels also decrease immediately after BMT, and these levels then transiently increase at 3 weeks; thereafter, they decrease to the menopausal level. We think that in the female BMT recipients, a combined primary and secondary hypogonadism may occur during the early post-BMT period (e.g., before 1 month); after then,

BIOMARKERS FOLLOWING BONE MARROW TRANSPLANTATION

A

111

B 400

100 a

350

a

80

300 a

60

c

250

a

b

200 40

150 100

20 a 0

b

a

LH (IU/l)

b a

a

50 0

FSH (IU/l)

C

E2 (pmol/l)

D 40

50

40

Pre-BMT 1 week 3 weeks 3 months 6 months

30 b

30

20 20

a b

10

10

0

LH (IU/l)

FSH (IU/l)

0

Testosterone (nmol/l)

FIG. 2. The changes in the serum LH, FSH (A), and E2 (B) levels in female recipients (n ¼ 20). The changes in the serum LH, FSH (C), and testosterone (D) levels in male recipients (n ¼ 19). Data are reported as mean values 95% CI. a, p < 0.01; b, p < 0.05 versus the basal value. c, p < 0.05 between 1 and 3 weeks in female recipients. Adapted from Ref. [44].

the primary gonadal dysfunction supervenes. However, these changes of gonadotropin were not demonstrated in the male recipients. In addition, the testosterone levels were only temporarily suppressed during the early post-BMT period. Besides, the male patients also showed increased LH and FSH levels at 6 months post-BMT, which suggests that a relatively subclinical gonadal dysfunction exists after BMT. It seems likely that the pituitary–gonad axis of women is more vulnerable to the combined TBI and

112

BAEK AND KANG

high-dose chemotherapy than is the pituitary–gonad axis of men. These gender differences in the changes of sex steroids after BMT may explain, at least in part, the greater susceptibility of women to bone loss after BMT than is observed for men [17]. The long-term prospects of gonadal failure after BMT have been well established. The major cause of gonadal damage leading to hypergonadotropic hypogonadism is irradiation [67]. In females, the ovaries are more susceptible to irradiation and chemotherapy than are the testes in males, and hypergonadotropic hypogonadism is almost the rule. Busulfan is one of the most gonadotoxic agents, whereas cyclophosphamide is usually associated with only minor effects on the gonadal function [68]. In males, the testicular germinal epithelium (the Sertoli cells), which is where spermatogenesis occurs, is more vulnerable to radiation and chemotherapy than the testicular Leydig cell component which is involved in testosterone secretion. Therefore, the testosterone levels are usually normal even when spermatogenesis is reduced or absent in the men who undergo BMT. The serum FSH level is typically elevated, whereas the LH levels may remain in the normal range [68]. This happens because male FSH secretion is predominantly regulated by inhibin B, which is a product of Sertoli cells, and LH secretion is regulated by testosterone. The decreased estrogen levels during the post-BMT period most likely cause abnormal bone remodeling and bone loss. It has been widely accepted that declining estrogen levels are associated with an increase in bone resorption and a relative deficit in bone formation, and this all results in accelerated bone loss [65]. Estrogen also has an important role in men. Estrogen accounts for 70% or more of the total effect of sex steroids on the bone resorption in men, whereas testosterone could account for no more than 30% of the effect [69]. Measuring the estrogen levels in male BMT patients could give us additional information, but there are currently no such data. In men, 85% or more of the circulating estrogen levels are derived not from direct testicular secretion, but rather from peripheral aromatization of testosterone [70]. So, it could be speculated that the decreased testosterone levels in men during the early post-BMT period and the subsequent subclinical hypogonadism both provoke a negative balance of bone. Testosterone is also important for the maintenance of bone formation and it has been shown to enhance periosteal apposition [69].

7. RANKL and Osteoprotegerin The receptor activator of nuclear factor B ligand (RANKL) and its receptor RANK are the key regulators of bone remodeling, and both are essential for the development and activation of osteoclasts [71]. The effects of

BIOMARKERS FOLLOWING BONE MARROW TRANSPLANTATION

113

RANKL are counterbalanced by osteoprotegerin (OPG), which acts as an endogenous soluble receptor antagonist [72]. Changes in the status of RANKL and OPG may be the important cause of the remodeling imbalance during the post-BMT period. In a study by Kananen et al. [73], the serum OPG level increased by 27% from baseline at 6 months post-BMT while the serum soluble RANKL (sRANKL) level did not change significantly. They also reported that the OPG levels, but not the sRANKL levels, were elevated in comparison to that of the healthy controls at a median of 122 days after BMT. Ricci et al. [74] investigated OPG and sRANKL in both the blood serum and the marrow plasma of long-term survivors. The serum OPG level was significantly higher in the patients than that in the controls, whereas the sRANKL levels were similar in the two groups. Interestingly, the OPG level in the bone marrow plasma was lower in patients than that in controls, whereas the sRANKL levels in the bone marrow plasma were similar in both groups. They suggested that the low OPG levels in the marrow plasma are likely a result of a decreased number of bone-forming cells, which are responsible for OPG production, and a nonskeletal source of OPG is likely responsible for the posttransplant serum increase of OPG. In our study [44], the serum sRANKL and OPG levels and the sRANKL/ OPG ratio showed their peak at post-BMT 3 weeks and these declined within 3 months after transplant. Also, the degrees of the changes of sRANKL and sRANKL/OPG changes were correlated with those of the bone resorption markers. When the findings about the changes of sRANKL, OPG, and the sRANKL/OPG ratio are taken together, they all suggest that during the early post-BMT period enhanced bone resorption is mediated, at least in part, by an enhancement of RANKL. However, the OPG also increases, and this suggests that the decoy receptor tries to act in opposition to the effects of RANKL. This mechanism seems to be ineffective for fully counterbalancing the increased RANKL activity; thus, the result is an increased RANKL/OPG ratio and increased bone resorption. However, the interpretation of sRANKL and OPG assays need to be done with a bit of caution. OPG is produced in various tissues, including bone, skin, stomach, intestine, lung, heart, and placenta [72]. Therefore, the serum concentrations of OPG may not accurately reflect its levels in the bones’ microenvironment. In fact, no correlation was found between the OPG or sRANKL levels measured in the serum and those levels measured in marrow plasma [74]. Also, laboratory measurements of sRANKL are done by an assay that is limited by the relative instability of serum RANKL [75]. Physiological factors such as cyclic variation, the menstrual status, age, and gender must also be considered when interpreting the results of RANKL assays [76].

114

BAEK AND KANG

8. Cytokines and Growth Factors The procedural requirements of BMT predispose these patients to the dramatic induction of local and systemic inflammatory states. Inflammatory responses can be induced in relation to conditioning therapy, the activation and expansion of alloreactive T cells, and graft-versus-host disease (GVHD) [77]. Alloreactive T cells can lead to dramatic and often fulminant inflammatory states, which are responsible for many transplantation-related complications, including bone loss. A marked increase of the plasma IL-6 level was demonstrated in patients after BMT, and this was an independent predictor of the bone resorption marker urinary pyridinoline [78]. In recent studies, the amounts of IL-6 and TNF-a were increased during the first month after BMT, and these were decreased by 3 months. Enhanced bone resorption following BMT is related to increased levels of bone marrow IL-6, which is a potent stimulator of bone resorption in vivo [48]. Also, an increase in IL-6 in the early post-BMT period predicted the amount of bone loss during the first year after BMT [79]. In a study by Ricci et al. [74], an increase in the levels of serum interferon-g and TNF-a, but not IL-6, was observed in patients who were evaluated 12–72 months after allo-BMT. In their study, no relationship was found between the IL-6 levels and the BMD values. These findings suggest that cytokines may play an important role in bone loss immediately after grafting, whereas their production and influence on BMD decreases thereafter. The IL-7 levels have also been noted to be increased during the early postBMT period [44]. An increased IL-7 level has been associated with suppressed bone formation and enhanced bone resorption, and it also predicted the amount of bone loss during the first 12 months after transplant [44]. IL-7 has been known to enhance the expression of RANKL by the T cells and the stromal cells and so stimulate osteoclastogenesis [80]. In addition, in the ovariectomized mouse model, IL-7 is upregulated as a consequence of estrogen deficiency and then it promotes bone resorption. Moreover, IL-7 has been reported to inhibit bone formation [81]. The growth factors produced by osteoblasts, including insulin-like growth factor (IGF), fibroblast growth factor (FGF), bone morphogenetic protein and TGF-b, are important for bone growth and osteogenesis [82]. They are capable of stimulating both osteoblast cell proliferation and differentiation. According to our observations [18], the serum IGF-I levels decreased progressively until 3 weeks after BMT and then they increased to the basal value at 3 months. Decreased IGF-I levels were correlated with the serum osteocalcin levels and the patients who had lower IGF-I levels during the early post-BMT period lost more bone at the proximal femur during the first year after BMT. TBI-induced hypothalamic pituitary injury has been shown to

BIOMARKERS FOLLOWING BONE MARROW TRANSPLANTATION

115

actually decrease growth hormone (GH) secretion [68], and it appears to contribute to low IGF-I levels. Inadvertent irradiation of the liver and other tissues including bone during the TBI contributes to defective production of IGF-I [67]. Glucocorticoids are possibly another cause for the decline of IGF-I levels during the early post-BMT period. Glucocorticoids directly decrease IGF-I synthesis in osteoblasts [39], and GH secretion is blunted by glucocorticoids due to an increase in the hypothalamic somatostatin tone, which may alter the systemic GH/IGF-I axis [83]. The serum FGF-2 level showed an abrupt decline immediately after the transplant and it returned to its basal level 3 months after BMT [18]. Okunieff et al. [84] also reported a significant decrease in the serum FGF-2 level during and after TBI. It is conceivable that myeloablation-induced tissue injury, including that to the bone, caused the suppressed FGF-2 levels. However, the exact mechanism must be determined in future studies. Although FGF-2 is known to play many important physiological roles in bone growth, remodeling, and repair, one study found no correlations between the FGF-2 levels and either the bone-turnover markers or the BMD changes [18]. Macrophage colony-stimulating factor (M-CSF) is one of a family of growth factors for the cells of the mononuclear phagocyte system and M-CSF plays a crucial role in osteoclast formation and bone resorption [85]. We confirmed that the M-CSF levels increased during the early post-BMT period. In addition, the increase in bone resorption was significantly associated with the M-CSF levels 3 weeks after BMT. Moreover, the bone loss at the proximal femur appeared to be affected by the increased M-CSF level [18].

9. Conclusion Post-BMT bone and mineral metabolism is represented by acute bone loss and unique changes of bone remodeling. The rapid impairment of bone formation and the increase in bone resorption, as shown by the biochemical markers of bone turnover, might play a role for the post-BMT bone loss. The immunosuppressants, including glucocorticoids and cyclosporine, are strong determinants of the biochemical uncoupling of the bone maintenance system. High-dose chemo/radiotherapy that is administered before a graft usually induces clinical or subclinical hypogonadism and this could also contribute to the imbalance of bone remodeling and the subsequent bone loss. Low vitamin D and a negative calcium balance theoretically could provoke secondary hyperthyroidism, but do not appear to play a decisive role in the development of osteoporosis in the BMT recipients. Alterations in the RANKL/OPG system seem to be crucial in the pathogenesis of posttransplant bone resorption, but more direct studies on this are surely needed. Changes of growth

116

BAEK AND KANG

factors also have an influence on the bone metabolism after transplantation. Finally, drastic increases of cytokines are common for a BMT procedure, and these increased cytokine levels can also induce osteoclastic activation. REFERENCES [1] C. Schulte, D.W. Beelen, U.W. Schaefer, K. Mann, Bone loss in long-term survivors after transplantation of hematopoietic stem cells: a prospective study, Osteoporos. Int. 11 (2000) 344–353. [2] M.J. Va¨lima¨ki, K. Kinnunen, L. Volin, et al., A prospective study of bone loss and turnover after allogeneic bone marrow transplantation: effect of calcium supplementation with or without calcitonin, Bone Marrow Transplant. 23 (1999) 355–361. [3] A. Kashyap, F. Kandeel, D. Yamauchi, et al., Effects of allogeneic bone marrow transplantation on recipient bone mineral density: a prospective study, Biol. Blood Marrow Transplant. 6 (2000) 344–351. [4] P.R. Ebeling, D.M. Thomas, B. Erbas, J.L. Hopper, J. Szer, A.P. Grigg, Mechanisms of bone loss following allogeneic and autologous hemopoietic stem cell transplantation, J. Bone Miner. Res. 14 (1999) 342–350. [5] G. Massenkeil, C. Fiene, O. Rosen, R. Michael, W. Reisinger, R. Arnold, Loss of bone mass and vitamin D deficiency after hematopoietic stem cell transplantation: standard prophylactic measures fail to prevent osteoporosis, Leukemia 15 (2001) 1701–1705. [6] L. Tauchmanova`, A. Colao, G. Lombardi, B. Rotoli, C. Selleri, Bone loss and its management in long-term survivors from allogeneic stem cell transplantation, J. Clin. Endocrinol. Metab. 92 (2007) 4536–4545. [7] K.N. Weilbaecher, Mechanisms of osteoporosis after hematopoietic cell transplantation, Biol. Blood Marrow Transplant. 6 (2000) 165–174. [8] A.D. Schimmer, M.D. Minden, A. Keating, Osteoporosis after blood and marrow transplantation: clinical aspects, Biol. Blood Marrow Transplant. 6 (2000) 175–181. [9] W.Y. Lee, S.W. Cho, E.S. Oh, et al., The effect of bone marrow transplantation on the osteoblastic differentiation of human bone marrow stromal cells, J. Clin. Endocrinol. Metab. 87 (2002) 329–335. [10] L. Tauchmanova`, B. Serio, A. Del Puente, et al., Long-lasting bone damage detected by dual-energy X-ray absorptiometry, phalangeal osteosonogrammetry, and in vitro growth of marrow stromal cells after allogeneic stem cell transplantation, J. Clin. Endocrinol. Metab. 87 (2002) 5058–5065. [11] E. Shane, M. Rivas, D.J. McMahon, et al., Bone loss and turnover after cardiac transplantation, J. Clin. Endocrinol. Metab. 82 (1997) 1497–1506. [12] S.L. Ferrari, L.P. Nicod, J. Hamacher, et al., Osteoporosis in patients undergoing lung transplantation, Eur. Respir. J. 9 (1996) 2378–2382. [13] A. Trombetti, M.W. Gerbase, A. Spiliopoulos, D.O. Slosman, L.P. Nicod, R. Rizzoli, Bone mineral density in lung-transplant recipients before and after graft: prevention of lumbar spine post-transplantation-accelerated bone loss by pamidronate, J. Heart Lung Transplant. 19 (2000) 736–743. [14] N. Buchs, C. Helg, C. Collao, et al., Allogeneic bone marrow transplantation is associated with a preferential femoral neck bone loss, Osteoporos. Int. 12 (2001) 880–886. [15] K. Kerschan-Schindl, M. Mitterbauer, W. Fu¨reder, et al., Bone metabolism in patients more than five years after bone marrow transplantation, Bone Marrow Transplant. 34 (2004) 491–496.

BIOMARKERS FOLLOWING BONE MARROW TRANSPLANTATION

117

[16] S. Castan˜eda, L. Carmona, I. Carvajal, R. Arranz, A. Dı´az, A. Garcı´a-Vadillo, Reduction of bone mass in women after bone marrow transplantation, Calcif. Tissue Int. 60 (1997) 343–347. [17] M.I. Kang, W.Y. Lee, K.W. Oh, et al., The short-term changes of bone mineral metabolism following bone marrow transplantation, Bone 26 (2000) 275–279. [18] K.H. Baek, W.Y. Lee, K.W. Oh, et al., Changes in the serum growth factors and osteoprotegerin after bone marrow transplantation: impact on bone and mineral metabolism, J. Clin. Endocrinol. Metab. 89 (2004) 1246–1254. [19] J.M. Stern, K.M. Sullivan, S.M. Ott, et al., Bone density loss after allogeneic hematopoietic stem cell transplantation: a prospective study, Biol. Blood Marrow Transplant. 7 (2001) 257–264. [20] W.Y. Lee, M.I. Kang, K.H. Baek, et al., The skeletal site-differential changes in bone mineral density following bone marrow transplantation: 3-year prospective study, J. Korean Med. Sci. 17 (2002) 749–754. [21] C.M. Schulte, D.W. Beelen, Bone loss following hematopoietic stem cell transplantation: a long-term follow-up, Blood 103 (2004) 3635–3643. [22] K. Kananen, L. Volin, R. Ta¨htela¨, K. Laitinen, T. Ruutu, M.J. Va¨lima¨ki, Recovery of bone mass and normalization of bone turnover in long-term survivors of allogeneic bone marrow transplantation, Bone Marrow Transplant. 29 (2002) 33–39. [23] M.K. Gandhi, S. Lekamwasam, I. Inman, et al., Significant and persistent loss of bone mineral density in the femoral neck after haematopoietic stem cell transplantation: longterm follow-up of a prospective study, Br. J. Haematol. 121 (2003) 462–468. [24] A.D. Schimmer, K. Mah, L. Bordeleau, et al., Decreased bone mineral density is common after autologous blood or marrow transplantation, Bone Marrow Transplant. 28 (2001) 387–391. [25] D. Marshall, O. Johnell, H. Wedel, Meta-analysis of how well measures of bone mineral density predict occurrence of osteoporotic fractures, Br. Med. J. 312 (1996) 1254–1259. [26] M. Lunt, D. Felsenberg, J. Reeve, et al., Bone density variation and its effects on risk of vertebral deformity in men and women studied in thirteen European centers: the EVOS Study, J. Bone Miner. Res. 12 (1997) 1883–1894. [27] S.R. Cummings, D.M. Black, M.C. Nevitt, et al., Bone density at various sites for prediction of hip fractures. The Study of Osteoporotic Fractures Research Group, Lancet 341 (1993) 72–75. [28] K. Kananen, L. Volin, K. Laitinen, H. Alfthan, T. Ruutu, M.J. Va¨lima¨ki, Prevention of bone loss after allogeneic stem cell transplantation by calcium, vitamin D, and sex hormone replacement with or without pamidronate, J. Clin. Endocrinol. Metab. 90 (2005) 3877–3885. [29] A.P. Grigg, P. Shuttleworth, J. Reynolds, et al., Pamidronate reduces bone loss after allogeneic stem cell transplantation, J. Clin. Endocrinol. Metab. 91 (2006) 3835–3843. [30] A.B. D’Souza, A.P. Grigg, J. Szer, P.R. Ebeling, Zoledronic acid prevents bone loss after allogeneic haemopoietic stem cell transplantation, Intern. Med. J. 36 (2006) 600–603. [31] L. Tauchmanova`, P. Ricci, B. Serio, et al., Short-term zoledronic acid treatment increases bone mineral density and marrow clonogenic fibroblast progenitors after allogeneic stem cell transplantation, J. Clin. Endocrinol. Metab. 90 (2005) 627–634. [32] L. Tauchmanova`, C. Selleri, M. Esposito, et al., Beneficial treatment with risedronate in long-term survivors after allogeneic stem cell transplantation for hematological malignancies, Osteoporos. Int. 14 (2003) 1013–1019. [33] B. Clarke, Normal bone anatomy and physiology, Clin. J. Am. Soc. Nephrol. 3 (Suppl. 3) (2008) 131–139. [34] T.J. Martin, N.A. Sims, Osteoclast-derived activity in the coupling of bone formation to resorption, Trends Mol. Med. 11 (2005) 76–81.

118

BAEK AND KANG

[35] Y. Han, S.C. Cowin, M.B. Schaffler, S. Weinbaum, Mechanotransduction and strain amplification in osteocyte cell processes, Proc. Natl. Acad. Sci. USA 101 (2004) 16689–16694. [36] E. Seeman, P.D. Delmas, Bone quality—the material and structural basis of bone strength and fragility, N. Engl. J. Med. 354 (2006) 2250–2261. [37] G. Holzer, H. Noske, T. Lang, L. Holzer, U. Willinger, Soluble cathepsin K: a novel marker for the prediction of nontraumatic fractures? J. Lab. Clin. Med. 146 (2005) 13–17. [38] D.J. Leeming, P. Alexandersen, M.A. Karsdal, P. Qvist, S. Schaller, L.B. Tanko´, An update on biomarkers of bone turnover and their utility in biomedical research and clinical practice, Eur. J. Clin. Pharmacol. 62 (2006) 781–792. [39] P. Garnero, E. Sornay-Rendu, B. Claustrat, P.D. Delmas, Biochemical markers of bone turnover, endogenous hormones and the risk of fractures in postmenopausal women: the OFELY study, J. Bone Miner. Res. 15 (2000) 1526–1536. [40] P. Garnero, E. Hausherr, M.C. Chapuy, et al., Markers of bone resorption predict hip fracture in elderly women: the EPIDOS Prospective Study, J. Bone Miner. Res. 11 (1996) 1531–1538. [41] S. Sarkar, J.Y. Reginster, G.G. Crans, A. Diez-Perez, K.V. Pinette, P.D. Delmas, Relationship between changes in biochemical markers of bone turnover and BMD to predict vertebral fracture risk, J. Bone Miner. Res. 19 (2004) 394–401. [42] S.L. Greenspan, H.N. Rosen, R.A. Parker, Early changes in serum N-telopeptide and C-telopeptide cross-linked collagen type 1 predict long-term response to alendronate therapy in elderly women, J. Clin. Endocrinol. Metab. 85 (2000) 3537–3540. [43] S.L. Greenspan, N.M. Resnick, R.A. Parker, Early changes in biochemical markers of bone turnover are associated with long-term changes in bone mineral density in elderly women on alendronate, hormone replacement therapy, or combination therapy: a three-year, doubleblind, placebo-controlled, randomized clinical trial, J. Clin. Endocrinol. Metab. 90 (2005) 2762–2767. [44] K.H. Baek, K.W. Oh, W.Y. Lee, et al., Changes in the serum sex steroids, IL-7 and RANKL-OPG system after bone marrow transplantation: influences on bone and mineral metabolism, Bone 39 (2006) 1352–1360. [45] R.S. Weinstein, R.L. Jilka, A.M. Parfitt, S.C. Manolagas, Inhibition of osteoblastogenesis and promotion of apoptosis of osteoblasts and osteocytes by glucocorticoids. Potential mechanisms of their deleterious effects on bone, J. Clin. Invest. 102 (1998) 274–282. [46] E. Canalis, A.M. Delany, Mechanisms of glucocorticoid action in bone, Ann. N. Y. Acad. Sci. 966 (2002) 73–81. [47] S. Minisola, R. Del Fiacco, S. Piemonte, et al., Biochemical markers in glucocorticoidinduced osteoporosis, J. Endocrinol. Invest. 31 (2008) 28–32. [48] W.Y. Lee, M.I. Kang, E.S. Oh, et al., The role of cytokines in the changes in bone turnover following bone marrow transplantation, Osteoporos. Int. 13 (2002) 62–68. [49] S. Epstein, Post-transplantation bone disease: the role of immunosuppressive agents and the skeleton, J. Bone Miner. Res. 11 (1996) 1–7. [50] C. Movsowitz, S. Epstein, M. Fallon, F. Ismail, S. Thomas, Cyclosporin-A in vivo produces severe osteopenia in the rat: effect of dose and duration of administration, Endocrinology 123 (1988) 2571–2572. [51] C. Movsowitz, S. Epstein, F. Ismail, M. Fallon, S. Thomas, Cyclosporin A in the oophorectomized rat: unexpected severe bone resorption, J. Bone Miner. Res. 4 (1989) 393–398. [52] A. Griesmacher, P. Peichl, P. Pointinger, et al., Biochemical markers in menopausal women, Scand. J. Clin. Lab. Invest. Suppl. 227 (1997) 64–72. [53] A. Banfi, M. Podesta`, L. Fazzuoli, et al., High-dose chemotherapy shows a dose-dependent toxicity to bone marrow osteoprogenitors: a mechanism for post-bone marrow transplantation osteopenia, Cancer 92 (2001) 2419–2428.

BIOMARKERS FOLLOWING BONE MARROW TRANSPLANTATION

119

[54] M. Galotto, G. Berisso, L. Delfino, et al., Stromal damage as consequence of high-dose chemo/radiotherapy in bone marrow transplant recipients, Exp. Hematol. 27 (1999) 1460–1466. [55] L. Nguyen, F.E. Dewhirst, P.V. Hauschka, P. Stashenko, Interleukin-1 beta stimulates bone resorption and inhibits bone formation in vivo, Lymphokine Cytokine Res. 10 (1991) 15–21. [56] D.R. Bertolini, G.E. Nedwin, T.S. Bringman, D.D. Smith, G.R. Mundy, Stimulation of bone resorption and inhibition of bone formation in vitro by human tumour necrosis factors, Nature 319 (1986) 516–518. [57] R. Schots, L. Kaufman, I. Van Riet, et al., Proinflammatory cytokines and their role in the development of major transplant-related complications in the early phase after allogeneic bone marrow transplantation, Leukemia 17 (2003) 1150–1156. [58] M. Kauppila, K. Irjala, P. Koskinen, et al., Bone mineral density after allogeneic bone marrow transplantation, Bone Marrow Transplant. 24 (1999) 885–889. [59] Y. Sato, H. Kuno, M. Kaji, K. Etoh, K. Oizumi, Influence of immobilization upon calcium metabolism in the week following hemiplegic stroke, J. Neurol. Sci. 175 (2000) 135–139. [60] D.T. Villareal, R. Civitelli, A. Chines, L.V. Avioli, Subclinical vitamin D deficiency in postmenopausal women with low vertebral bone mass, J. Clin. Endocrinol. Metab. 72 (1991) 628–634. [61] K. Kerschan-Schindl, J. Strametz-Juranek, G. Heinze, et al., Pathogenesis of bone loss in heart transplant candidates and recipients, J. Heart Lung Transplant. 22 (2003) 843–850. [62] S. Huybers, T.H. Naber, R.J. Bindels, J.G. Hoenderop, Prednisolone-induced Ca2þ malabsorption is caused by diminished expression of the epithelial Ca2þ channel TRPV6, Am. J. Physiol. Gastrointest. Liver Physiol. 292 (2007) G92–G97. [63] G. Mazziotti, A. Angeli, J.P. Bilezikian, E. Canalis, A. Giustina, Glucocorticoid-induced osteoporosis: an update, Trends Endocrinol. Metab. 17 (2006) 144–149. [64] R. Eastell, Pathogenesis of postmenopausal osteoporosis, in: M. Favus (Ed.), Primer on the Metabolic Bone Diseases and Disorders of Mineral Metabolism. fourth ed., Vol. 1, (1999), Lippincott Williams and Wilkins, Philadelphia. [65] F. Syed, S. Khosla, Mechanisms of sex steroid effects on bone, Biochem. Biophys. Res. Commun. 328 (2005) 688–696. [66] M. Sakakura, K. Takebe, S. Nakagawa, Inhibition of luteinizing hormone secretion induced by synthetic LRH by long-term treatment with glucocorticoids in human subjects, J. Clin. Endocrinol. Metab. 40 (1975) 774–779. [67] M. Kauppila, P. Koskinen, K. Irjala, K. Remes, J. Viikari, Long-term effects of allogeneic bone marrow transplantation (BMT) on pituitary, gonad, thyroid and adrenal function in adults, Bone Marrow Transplant. 22 (1998) 331–337. [68] B.M. Brennan, S.M. Shalet, Endocrine late effects after bone marrow transplant, Br. J. Haematol. 118 (2002) 58–66. [69] S. Khosla, S. Amin, E. Orwoll, Osteoporosis in men, Endocr. Rev. 29 (2008) 441–464. [70] L. Gennari, R. Nuti, J.P. Bilezikian, Aromatase activity and bone homeostasis in men, J. Clin. Endocrinol. Metab. 89 (2004) 5898–5907. [71] D.L. Lacey, E. Timms, H.L. Tan, et al., Osteoprotegerin ligand is a cytokine that regulates osteoclast differentiation and activation, Cell 93 (1998) 165–176. [72] W.S. Simonet, D.L. Lacey, C.R. Dunstan, et al., Osteoprotegerin: a novel secreted protein involved in the regulation of bone density, Cell 89 (1997) 309–319. [73] K. Kananen, L. Volin, K. Laitinen, T. Ruutu, M.J. Va¨lima¨ki, Serum osteoprotegerin and receptor activator of nuclear factor-kappaB ligand (RANKL) concentrations in allogeneic stem cell transplant-recipients: a role in bone loss? Osteoporos. Int. 17 (2006) 724–730.

120

BAEK AND KANG

[74] P. Ricci, L. Tauchmanova, A.M. Risitano, et al., Imbalance of the osteoprotegerin/ RANKL ratio in bone marrow microenvironment after allogeneic hemopoietic stem cell transplantation, Transplantation 82 (2006) 1449–1456. [75] G. Hawa, N. Brinskelle-Schmal, K. Glatz, S. Maitzen, W. Woloszczuk, Immunoassay for soluble RANKL (receptor activator of NF-kappaB ligand) in serum, Clin. Lab. 49 (2003) 461–463. [76] D. Vega, N.M. Maalouf, K. Sakhaee, The role of receptor activator of nuclear factorkappaB (RANK)/RANK ligand/osteoprotegerin: clinical implications, J. Clin. Endocrinol. Metab. 92 (2007) 4514–4521. [77] G.R. Hill, Inflammation and bone marrow transplantation, Biol. Blood Marrow Transplant. 15 (Suppl.) (2008) 139–141. [78] W. Withold, H.H. Wolf, S. Kollbach, et al., Relationship between bone metabolism and plasma cytokine levels in patients at risk of post-transplantation bone disease after bone marrow transplantation, Eur. J. Clin. Chem. Clin. Biochem. 34 (1996) 295–299. [79] W.Y. Lee, K.H. Baek, E.J. Rhee, et al., Impact of circulating bone-resorbing cytokines on the subsequent bone loss following bone marrow transplantation, Bone Marrow Transplant. 34 (2004) 89–94. [80] G. Toraldo, C. Roggia, W.P. Qian, R. Pacifici, M.N. Weitzmann, IL-7 induces bone loss in vivo by induction of receptor activator of nuclear factor kappa B ligand and tumor necrosis factor alpha from T cells, Proc. Natl. Acad. Sci. USA 100 (2003) 125–130. [81] M.N. Weitzmann, C. Roggia, G. Toraldo, L. Weitzmann, R. Pacifici, Increased production of IL-7 uncouples bone formation from bone resorption during estrogen deficiency, J. Clin. Invest. 110 (2002) 1643–1650. [82] E. Canalis, Skeletal growth factors, in: R. Marcus, D. Feldman, J. Kelsey (Eds.), Osteoporosis, second ed., Academic Press, San Diego, 2001, pp. 405–431. [83] A. Giustina, J.D. Veldhuis, Pathophysiology of the neuroregulation of growth hormone secretion in experimental animals and the human, Endocr. Rev. 19 (1998) 717–797. [84] P. Okunieff, A.J. Barrett, S.E. Phang, et al., Circulating basic fibroblast growth factor declines during Cy/TBI bone marrow transplantation, Bone Marrow Transplant. 23 (1999) 1117–1121. [85] Y. Fujikawa, A. Sabokbar, S.D. Neale, I. Itonaga, T. Torisu, Athanasou NA The effect of macrophage-colony stimulating factor and other humoral factors on human osteoclast formation from circulating cells, Bone 28 (2001) 261–267. [86] K. Carlson, B. Simonsson, S. Ljunghall, Acute effects of high-dose chemotherapy followed by bone marrow transplantation on serum markers of bone metabolism, Calcif. Tissue Int. 55 (1994) 408–411.

ADVANCES IN CLINICAL CHEMISTRY, VOL. 49

FACTOR V LEIDEN AND ACTIVATED PROTEIN C RESISTANCE Olivier Segers and Elisabetta Castoldi1 Department of Biochemistry, Maastricht University, 6200 MD Maastricht, The Netherlands

1. Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. The Protein C System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. The Key Players: Protein C and Protein S. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Protein C Activation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3. Anticoagulant Activity of APC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4. APC Resistance and FV Leiden . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. The APC Resistance Phenotype. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2. The FV Leiden Mutation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3. Other Causes of APC Resistance. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5. Conclusions and Perspectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

121 122 123 123 124 125 130 130 132 137 143 144

1. Abstract Activated protein C (APC) proteolytically inactivates factors Va (FVa) and VIIIa (FVIIIa), which in turn control two key steps of the coagulation cascade. The pathophysiological importance of this anticoagulant mechanism is illustrated by the severe prothrombotic diathesis associated with the congenital deficiencies of protein C and its cofactor protein S. A poor anticoagulant response of plasma to APC (APC resistance) was first described in a thrombotic patient in 1993 and soon recognized as the most common risk factor for venous thrombosis. The underlying genetic defect was identified one year later as the FV Arg506Gln mutation (FV Leiden), which abolishes one of the APC-cleavage sites on FVa. These ground-breaking discoveries 1

Corresponding author: Elisabetta Castoldi, e-mail: [email protected] 121

0065-2423/09 $35.00 DOI: 10.1016/S0065-2423(09)49006-1

Copyright 2009, Elsevier Inc. All rights reserved.

122

SEGERS AND CASTOLDI

have stimulated numerous researches into the workings of the protein C pathway, the molecular mechanisms of APC resistance in carriers and noncarriers of FV Leiden, and the clinical significance of APC resistance. This chapter reviews the most important findings, summarizes the state of the art, and discusses new developments in this rapidly evolving research area.

2. Introduction Venous thrombosis, the obstruction of a vein by an intravascular clot, is a serious condition affecting 1/1000 people per year. The most common manifestation is deep-vein thrombosis of the lower limbs. In 1–2% of cases, deepvein thrombosis evolves in life-threatening pulmonary embolism. Additional complications are the high risk of recurrence ( 25% within 5 years of discontinuation of the anticoagulation treatment) and an invalidating postthrombotic syndrome in 20% of the patients [1]. The pathogenesis of venous thrombosis is typically multifactorial, involving complex interactions of several genetic and acquired factors. Most known risk factors act by inducing a hypercoagulable state [2]. The protein C pathway (reviewed in Ref. [3]) is a major anticoagulant system that leads to the inactivation of coagulation factors Va (FVa) and VIIIa (FVIIIa), the essential cofactors of the prothrombinase and tenase complexes, respectively. The pathway is initiated by the activation of protein C via the thrombin/thrombomodulin (TM) complex on the surface of endothelial cells. Activated protein C (APC) subsequently inactivates membrane-bound FVa and FVIIIa by limited proteolysis at specific sites. Both inactivation reactions are stimulated by protein S, the nonenzymatic cofactor of APC. Moreover, factor V (FV) acts as an additional APCcofactor in the inactivation of FVIIIa. The pivotal role of the protein C pathway in downregulating the coagulation cascade is underscored by the severe prothrombotic diathesis associated with congenital deficiencies of protein C [4] and protein S [5, 6]. However, these deficiencies are rare in the general population. In contrast, functional abnormalities of the protein C pathway, which manifest themselves as a poor sensitivity of plasma to added APC (APC resistance [7]), are common. APC resistance is most often caused by the FV Arg506Gln (FV Leiden) mutation [8], which abolishes one of the cleavage sites recognized by APC on FV(a). The identification of APC resistance and the FV Leiden mutation as the most common risk factors for venous thrombosis represented a major breakthrough, and has fostered a large number of mechanistic and epidemiological studies. This chapter reviews the most important findings and summarizes

FACTOR V LEIDEN AND ACTIVATED PROTEIN C RESISTANCE

123

our current understanding of the pathophysiology of the protein C pathway. After a general introduction to the protein C system, we describe the discovery of APC resistance and FV Leiden and discuss their molecular mechanisms and clinical implications. We conclude with an overview of other causes of APC resistance and with some future perspectives of this rapidly developing research area.

3. The Protein C System 3.1. THE KEY PLAYERS: PROTEIN C AND PROTEIN S Protein C [9] is a 62-kDa vitamin K-dependent glycoprotein circulating in plasma at a concentration of 65 nM. Following biosynthesis in the liver, protein C is secreted in the circulation as a single-chain molecule, which is subsequently cleaved at Lys156 to yield a light chain and a heavy chain held together by a disulfide bond [10]. Most protein C present in plasma (85–95%) is in this processed two-chain form. Protein C is a zymogen and is converted to the active serine protease (APC) by limited proteolysis of a single peptide bond [11]. APC is a multifunctional enzyme with anticoagulant, cytoprotective, antiapoptotic, and anti-inflammatory properties (reviewed in Refs. [12, 13]). This chapter focuses on the anticoagulant function of APC, which consists in the proteolytic inactivation of FVa and FVIIIa. The gene for human protein C (PROC) is located on chromosome 2 (q13–q14) [14]. PROC comprises nine exons which code for a 419-amino acid mature protein highly homologous to prothrombin and coagulation factors VII (FVII), IX (FIX), and X (FX) [15]. Protein C is composed of an N-terminal g-carboxyglutamic acid (Gla) domain, two epidermal growth factor-like (EGF-like) domains, a short activation peptide, and a C-terminal serine protease domain. The Gla domain, containing Gla residues, is important for proper protein folding and calcium-dependent binding to anionic phospholipid surfaces. Differently, the EGF-like domains are likely involved in the interaction between APC and its cofactors and substrates [16]. Protein S [17] is a 75-kDa vitamin K-dependent glycoprotein which is mainly synthesized in hepatocytes and endothelial cells. It circulates in plasma at a concentration of 350 nM, of which 60% is noncovalently bound to the complement regulatory factor C4b-binding protein (C4BP) [18]. Within the protein C pathway, protein S functions as a nonenzymatic cofactor of APC in the inactivation of FVa and FVIIIa [19]. However, protein S also has an APCindependent anticoagulant activity by acting as a cofactor of tissue factor pathway inhibitor (TFPI) in the inhibition of FXa [20, 21].

124

SEGERS AND CASTOLDI

The gene for protein S (PROS1) spans 80 kb on chromosome 3 (q11.2) and contains 15 exons [22]. Mature protein S is a 635-amino acid protein comprising an N-terminal Gla domain, a thrombin-sensitive region (TSR), four EGF-like domains, and a large sex hormone-binding globulin (SHBG)like domain [23]. The Gla domain is essential for the calcium-dependent binding of protein S to anionic phospholipid membranes as well as for the expression of APC-cofactor activity. Moreover, several studies indicate that an intact TSR and the first two EGF-like domains are necessary for the expression of APC-cofactor activity, whereas the SHBG-like domain contains interaction sites for C4BP and the other APC-cofactor FV (reviewed in Ref. [24]).

3.2. PROTEIN C ACTIVATION Protein C is activated by thrombin bound to TM on the surface of endothelial cells (Fig. 1). Thrombin cleaves a single peptide bond (Arg169) in the heavy chain of protein C, thereby releasing an activation peptide of 12 amino acids and converting the serine protease domain to its active conformation [11]. Protein C activation is stimulated 1000-fold by TM [25, 26] and an additional 20-fold by the endothelial protein C receptor (EPCR)

APC Cleavage at Arg169

TM

TM

PC IIa

EPCR

12 a.a.

IIa

EPCR

FIG. 1. Protein C activation. Protein C (PC; red) is activated by thrombin (IIa; yellow) on the surface of endothelial cells. During this process, the transmembrane receptors thrombomodulin (TM; green) and endothelial protein C receptor (EPCR; blue) bind thrombin and protein C, respectively, and closely align them for optimal cleavage. Cleavage of a single peptide bond (Arg169) converts protein C to its active form, activated protein C (APC; red), which is released into the circulation with the capacity to inactivate procoagulant cofactors FVa and FVIIIa. Modified from Ref. [235].

FACTOR V LEIDEN AND ACTIVATED PROTEIN C RESISTANCE

125

[27, 28]. These integral membrane proteins act as receptors for thrombin and protein C, respectively, and act in concert to position enzyme and substrate in an optimal reciprocal orientation for cleavage (Fig. 1). Once released into the circulation, APC has a relatively long half-life (15–20 min) and is eventually inhibited by protein C inhibitor (PCI), a1-antitrypsin, or a2-macroglobulin [29]. TM is a transmembrane glycoprotein with a complex multidomain structure consisting of an N-terminal lectin domain, six EGF-like domains, a Ser/Thr-rich region, a transmembrane section, and a short C-terminal cytoplasmatic tail. TM is expressed at high concentrations on the endothelial cells of virtually all vascular beds [30], but especially in the microcirculation. Following high-affinity interaction with TM, thrombin undergoes a conformational change induced by the occupation of thrombin exosite I by EGFlike domains 5 and 6 of TM [31]. As a consequence of this conformational modification, TM-bound thrombin loses its procoagulant activities [32–34] and acquires new substrate specificities. Protein C becomes the preferred substrate and is cleaved in a highly Ca2þ-dependent manner [35]. The thrombin/TM complex is then rapidly inactivated, as it is very susceptible to inhibition by antithrombin and PCI [36]. The EPCR belongs to the major histocompatibility complex (MHC) class I/CD1 family of membrane proteins [37] and is predominantly expressed on endothelial cells of large vessels [38]. Binding of circulating protein C to the EPCR, which is mediated by the Gla domain [39, 40], decreases the Km for protein C activation by closely aligning protein C with the activating thrombin/TM complex [27, 28]. Platelet factor 4 (PF4), an abundant platelet a-granule protein, has also been shown to stimulate protein C activation by the thrombin/TM complex [41]. PF4 is believed to interact with the Gla domain of protein C and to induce a conformational change that increases the affinity of protein C for the thrombin/TM complex. In contrast, fibrinogen has been reported to inhibit protein C activation by competing with TM for thrombin exosite I [42]. 3.3. ANTICOAGULANT ACTIVITY OF APC APC downregulates coagulation by proteolytically inactivating FVa and FVIIIa, the essential nonenzymatic cofactors of the prothrombinase and intrinsic tenase complexes, respectively. FV and FVIII are highly homologous proteins that share the same domain structure (A1–A2–B–A3–C1–C2) [43]. Both of them circulate in plasma as inactive single-chain procofactors and are activated via limited proteolysis by thrombin or FXa. Activation of FV by cleavage at Arg709, Arg1018, and Arg1545 removes the B domain and yields a heterodimer comprising a heavy chain (A1–A2) and a light chain (A3–C1–C2) noncovalently associated via a Ca2þ ion. This

126

SEGERS AND CASTOLDI

activated form (FVa) binds to FXa on negatively charged phospholipid surfaces in the presence of Ca2þ ions (prothrombinase complex) and enhances FXa-catalyzed prothrombin activation by several orders of magnitude (reviewed in Ref. [44]). Similarly, FVIII is cleaved by thrombin or FXa at Arg740 and Arg1689, which leads to the elimination of the B domain and the formation of a heavy and a light chain. The heavy chain is subsequently cleaved at Arg372. The resulting FVIIIa is, therefore, a heterotrimer composed of noncovalently associated A1, A2, and A3–C1–C2 fragments. FVIIIa binds to FIXa on negatively charged phospholipid surfaces in the presence of Ca2þ ions (tenase complex) and enhances FIXa-catalyzed FX activation by several orders of magnitude (reviewed in Ref. [45]). Since FXa and FIXa in the absence of their respective cofactors are very inefficient activators of prothrombin and FX, respectively, APC-mediated inactivation of FVa and FVIIIa shuts down all prothrombinase and tenase activity. In this way, the protein C pathway importantly contributes to the downregulation of coagulation. 3.3.1. APC-Mediated Inactivation of FVa APC inactivates FVa via limited proteolysis of the heavy chain. This reaction is greatly stimulated by negatively charged phospholipids and, to a lesser extent, by the APC-cofactor protein S [46, 47] (Fig. 2A). The composition of the lipid surface affects the rate of APC-mediated FVa inactivation [47–49]. In particular, endothelial cells offer a more favorable surface for FVa inactivation than platelets [50]. In the absence of phospholipids, APCmediated inactivation of FVa is extremely slow and not influenced by protein S [46, 47]. The detailed mechanism by which APC inactivates FVa is presently well understood. APC cleaves FVa at residues Arg306, Arg506, and Arg679 in the heavy chain [51, 52] (Fig. 2A). Complete loss of cofactor activity correlates with cleavage at Arg306, but Arg506 is cleaved at a 20-fold higher rate. Cleavage at Arg679 is only relevant in the absence of lipids. Due to the kinetic advantage of the Arg506 site, most FVa molecules undergo initial rapid cleavage at Arg506, which yields an intermediate that has reduced affinity for FXa, but still retains 40% of the cofactor activity. This intermediate is then fully inactivated by slow cleavage at Arg306. However, FVa can also be inactivated by direct slow cleavage at Arg306, which leads to complete loss of the cofactor activity [52]. In particular, the latter is the only available mechanism for the inactivation of FVaLeiden, which lacks the Arg506 cleavage site (see Section 4.2.2). Following proteolysis by APC, FVa loses its binding affinity for FXa and hence its FXa-cofactor activity in prothrombin activation [53, 54].

FACTOR V LEIDEN AND ACTIVATED PROTEIN C RESISTANCE

127

A

Arg Arg Arg 306 506 679

A2

A1

A1 A2 A3

PS APC

C1 C2

A3

C1 C2

A3

C1 C2

Ca2+

+ Protein S FXa Prothrombin −

+ − −

Arg 562

B Arg 336

A2 A1

FV

A2 A1 A3

PS APC

C1 C2

Ca2+ Protein S FV FIXa FX

+ + −

+ + −

FIG. 2. Anticoagulant activity of APC. (A) APC-mediated inactivation of FVa. APC (red) inactivates FVa (green) via limited proteolysis at residues Arg306, Arg506, and Arg679 in the heavy chain. This reaction occurs on a phospholipid surface and is greatly stimulated by the APC cofactor protein S (PS; purple). The eects of protein S, FXa, and prothrombin on the individual cleavage sites (þ, stimulation; , inhibition) are indicated. (B) APC-mediated inactivation of FVIIIa. APC (red) inactivates FVIIIa (orange) via limited proteolysis at residues Arg336 (A1 domain) and Arg562 (A2 domain). This reaction occurs on a phospholipid surface and is greatly stimulated by the APC cofactors protein S (PS; purple) and FV (green). The eects of protein S, FV, FIXa, and FX on the individual cleavage sites (þ, stimulation; , inhibition) are indicated.

128

SEGERS AND CASTOLDI

The exact role of protein S in FVa inactivation has also been elucidated. Free protein S enhances APC-catalyzed cleavage of FVa at Arg306 20-fold [49, 55], whereas it only has a minor effect (three- to fivefold stimulation, depending on the lipid composition) on cleavage at Arg506 [49] (Fig. 2A). Therefore, protein S is especially important for the second (slow) phase of FVa inactivation, and for direct single-cleavage inactivation at Arg306 (such in FVaLeiden). These findings explain why earlier studies reported only a modest stimulation of FVa inactivation by protein S, as these studies concentrated on the initial rate of inactivation, which is determined by the largely protein S-independent Arg506 cleavage. In addition, it has been recently shown that C4BP-bound protein S, which was long considered to be devoid of APC-cofactor activity [56, 57], stimulates cleavage at Arg306 more than 10-fold, while inhibiting cleavage at Arg506 three- to fourfold [58]. Two mechanisms have been proposed to account for the APC-cofactor activity of protein S: (1) protein S would enhance the binding of APC to phospholipids [46] and (2) protein S would relocate the active site of APC closer to the membrane for optimal FVa cleavage [59, 60]. Several studies have indicated that, when FVa is incorporated in the prothrombinase complex, it is protected from APC-mediated inactivation by interactions with FXa [55, 61–64] and prothrombin [65–68]. However, protein S has been reported to abrogate the FXa-mediated protection, making it possible for APC to inactivate FXa-bound FVa [69]. The mechanisms underlying these phenomena have also been investigated in detail. Binding of FXa to FVa specifically decreases the rate of cleavage at Arg506 20-fold [55, 63] (Fig. 2A), but this effect is lost in the presence of physiological concentrations of protein S. This is not due to direct competition between FXa and protein S for binding to FVa, but rather to the fact that protein S stimulates the inactivation of free (non-FXa-bound) FVa molecules [49]. This causes more FVa to dissociate from FXa and be inactivated, until no FXa-bound FVa is left anymore [63]. This mechanism accounts for the apparent abrogation of the FXa-mediated protection of FVa by protein S [69]. Prothrombin delays the inactivation of FVa by inhibiting APC-catalyzed cleavage at both Arg306 and Arg506 [67, 68] (Fig. 2A). However, this effect can be counteracted by high concentrations of protein S, suggesting that prothrombin and protein S share a binding site on FVa [67]. The mechanism by which prothrombin protects FVa remains unclear, but it probably involves the disruption of some interaction between APC and FVa [67, 68]. Finally, the platelet-derived chemokine PF4 (which promotes protein C activation by the thrombin/TM complex, see Section 3.2) has been shown to inhibit APC-mediated FVa inactivation and particularly cleavage at Arg306 [70].

FACTOR V LEIDEN AND ACTIVATED PROTEIN C RESISTANCE

129

3.3.2. APC-Mediated Inactivation of FVIIIa The FVIIIa heterotrimer, composed of loosely associated A1, A2, and A3– C1–C2 subunits, is rather unstable and loses cofactor activity by the spontaneous dissociation of the A2 subunit, which causes a decrease of binding affinity for FIXa [71]. However, spontaneous inactivation is counteracted by the incorporation of FVIIIa in the tenase complex [72, 73]. Alternatively, FVIIIa can be inactivated through limited proteolysis by APC. The APCmediated inactivation of FVIIIa (Fig. 2B) shows several analogies with FVa inactivation. In particular, cleavage only occurs in the heavy chain [74] and is stimulated by negatively charged phospholipids [75] and protein S [76]. Moreover, since the discovery of APC resistance it is known that efficient APC-catalyzed inactivation of FVIIIa requires FV as an additional APCcofactor [7, 77]. To function as an APC cofactor, FV must retain (part of) the B domain (FVa is therefore devoid of APC-cofactor activity) [78–81] and to be cleaved by APC at Arg506 [79, 82]. Experiments in model systems indicate that FV is a more potent APC cofactor than protein S, but its effect is largely dependent on the concomitant presence of protein S [80, 83, 84]. Therefore, protein S and FV act as synergistic cofactors in the APC-mediated inactivation of FVIIIa. C4BP-bound protein S has also been reported to enhance the APC-mediated inactivation of factor VIIIa but, in contrast to free protein S, it does not synergize with FV [85]. The molecular mechanism of APC-catalyzed FVIIIa inactivation has been elucidated to a great extent. APC cleaves FVIIIa at Arg336 and Arg562 in the A1 and A2 subunits, respectively [74] (Fig. 2B). Cleavage at Arg336 alters the interaction between the A1 and A2 subunits, favoring A2 dissociation and resulting in a decreased kcat and an increased Km for FX activation [86, 87]. Differently, cleavage at Arg562 disrupts an important FIXa interaction site, severely decreasing the binding affinity of FVIIIa for FIXa [88]. While both cleavages contribute to FVIIIa inactivation, only cleavage at Arg562 results in complete loss of cofactor activity [84]. When FVIIIa is incorporated in the tenase complex, it is efficiently protected from inactivation by APC. In particular, FIXa protects the Arg562 cleavage site, whereas FX protects the Arg336 cleavage site (Fig. 2B). However, FX-mediated blockage of the Arg336 cleavage site can be abrogated by protein S [89]. Cleavage at Arg336 is kinetically favored over cleavage at Arg562 and usually occurs first. However, analysis of APC-catalyzed inactivation of an artificial FVIII(a) mutant lacking the Arg336 cleavage site shows that prior cleavage at Arg336 is not strictly required for cleavage at Arg562 [90]. Protein S and FV greatly enhance APC-mediated FVIIIa inactivation by stimulating proteolysis at both cleavage sites (Fig. 2B), their effect being more potent on the Arg562 cleavage [84, 89]. As a result, in the presence of both protein S and FV cleavage at Arg336 is only twofold faster than cleavage at Arg562 [84].

130

SEGERS AND CASTOLDI

4. APC Resistance and FV Leiden 4.1. THE APC RESISTANCE PHENOTYPE 4.1.1. Discovery of APC Resistance Since APC is a potent inhibitor of FVa and FVIIIa, the addition of exogenous APC to a plasma sample results in a dose-dependent prolongation of the clotting time. However, in 1993 Dahlba¨ck et al. [7] described a thrombotic patient whose plasma showed a poor anticoagulant response to APC, as added APC failed to prolong the plasma clotting time in an activated partial thromboplastin time (APTT) assay. This plasma phenotype was, therefore, referred to as ‘‘APC resistance’’ (Fig. 3). Interestingly, the patient had a strong personal and family history of thrombosis, and several family members showed the same plasma abnormality, suggesting an inherited defect. After excluding the presence of a circulating APC inhibitor or protein S deficiency as possible causes of APC resistance, it was hypothesized that the patient (and his affected family members) might carry a mutation that made their FVa or FVIIIa resistant to APC-mediated inactivation. To explore this possibility, the plasma’s response to APC was tested in FXa and

A

B

120 −APC +APC

APTT reagent CaCl2

Plasma

APTT reagent APC/CaCl2

Clotting time (s)

100 80 60 40 20 Clotting time−APC

Clotting time+APC

0 Normal plasma

APC-resistant plasma

FIG. 3. (A) APTT-based APC resistance assay. The APTT-based APC resistance assay is based on the measurement of the plasma clotting time after activation of the intrinsic coagulation pathway in the presence and absence of added APC. (B) APC resistance. In normal plasma, the clotting time obtained in the presence of APC is approximately three times longer than the clotting time in the absence of APC. In APC-resistant plasma APC fails to significantly prolong the clotting time. Clotting times were derived from the original publication by Dahlba¨ck et al. [7].

FACTOR V LEIDEN AND ACTIVATED PROTEIN C RESISTANCE

131

FIXa-initiated clotting assays. Since APC resistance was much more pronounced in the FIXa-initiated assay than in the FXa-initiated assay, it was assumed that impaired FVIIIa inactivation rather than impaired FVa inactivation was responsible for APC resistance. However, polymorphism analysis in the family excluded linkage of the APC resistance phenotype to both the FVIII and von Willebrand factor (vWF) genes. Based on the observation that the addition of normal plasma dose-dependently corrected the APC resistance phenotype, it was eventually concluded that the poor anticoagulant response to APC was due to an inherited deficiency of a new (unidentified) cofactor of APC [7]. Shortly afterwards, the novel APC cofactor was purified from normal plasma and found to be indistinguishable from well-known coagulation FV [77]. Accordingly, normal FV proved able to stimulate APC-catalyzed FVIIIa inactivation in model systems containing purified proteins [78]. The involvement of FV in APC resistance was further supported by the observation that APC resistance could be corrected by all coagulation factor-depleted plasmas, except FV-depleted plasma [8]. Moreover, plasma reconstitution experiments indicated that FV purified from APC-resistant plasma was sufficient to confer the abnormal phenotype to FV-deficient plasma, whereas APC-resistant plasma depleted of endogenous FV and reconstituted with normal FV showed a normal APC response [91]. Finally, genetic analysis in two large thrombophilic families showed linkage of the APC resistance phenotype to the FV gene [8, 92] and led to the identification of the FV G1691A mutation in exon 10, predicting the substitution of Arg506 by a Gln (FV Leiden), as the cause of APC resistance [8, 92–94] (Fig. 4). The causative role of this mutation was later confirmed by the ability of plasma-purified as well as recombinant FVLeiden to confer APC resistance to FV-deficient plasma [95–98]. A passionate account of the exciting quest for the molecular defect underlying APC resistance can be found in Ref. [99]. 4.1.2. Prevalence and Thrombosis Risk Soon after the discovery of APC resistance, an association between this plasma abnormality and the risk of venous thrombosis was established by several epidemiological studies [100–102]. Moreover, it became obvious that this condition was very frequent, being found in approximately 5% of healthy controls from the general population, 20% of consecutive patients with venous thrombosis, and up to 60% of selected patients with a family history of thrombosis. In the Leiden Thrombophilia study (LETS), a populationbased case-control study, APC resistance was estimated to confer a 6.6-fold increased risk of thrombosis [100]. These figures make APC resistance the most common risk factor for venous thrombosis known to date.

132

SEGERS AND CASTOLDI

Coagulation cascade

Protein C pathway

TF FVIIa

APC PS

FIXa FVIIIa

FVIIIi

FVAPC FV

FV Leiden

FXa FVa

APC PS

FVi

FV Leiden APC APC APC

Thrombin A1

Fibrin clot

A2

B

A3

C1 C2

Arg306 Gln506 Arg679

FIG. 4. The FV Leiden mutation. The FV Leiden mutation (inset) abolishes the APC-cleavage site at Arg506 in the heavy chain of FV(a). As a consequence, FVaLeiden is inactivated by APC at a lower rate and single-chain FVLeiden cannot be converted into a functional APC cofactor for FVIIIa inactivation. The result is delayed inactivation of both FVa and FVIIIa.

Although the FV Leiden mutation is the most important cause of APC resistance (at least in the Caucasian population), we now know that other genetic and acquired conditions are associated with APC resistance (see Section 4.3). Interestingly, APC resistance increases the risk of venous thrombosis also in the absence of FV Leiden [103, 104]. 4.2. THE FV LEIDEN MUTATION 4.2.1. Population Genetics The FV Arg506Gln mutation (FV Leiden) is almost exclusively present in the Caucasian population [105–107], where it accounts for >90% of individuals with inherited APC resistance. Moreover, the mutation has also been detected in the Middle East and India. Differently, it is (virtually) absent in natives of Africa, Asia, America, and Australia, suggesting that other genetic mutations are responsible for APC resistance in these ethnic groups [108]. Within Europe, the prevalence (carrier frequency) of FV Leiden ranges

FACTOR V LEIDEN AND ACTIVATED PROTEIN C RESISTANCE

133

between 2% and 15% in different populations [109]. In particular, the mutation is very common in the south of Sweden (where APC resistance was discovered!) and its frequency decreases along a characteristic north–south gradient (Fig. 5). However, a high prevalence has also been reported in Greek Cypriots and some Middle Eastern populations [106]. Several studies have shown that all FV Leiden carriers share one and the same FV gene haplotype [8, 64, 110–112], which suggests that they have

3.1

5.9

1.5

4.1

7.8

4.0 1.3 1.7

2.8

0.6

3.9 1.5

4.0 2.0

2.1

3.0 0.3

1.8

1.3

1.7 1.7

1.0

4.8

2.0 2.8 2.9 2.8 1.5 5.0

4.5 4.2 4.9

7.0

4.7 6.5

FIG. 5. Geographical distribution of the FV Leiden mutation in Europe. The numbers represent allele frequencies. The data have been extracted from the following sources: Albania [202], Austria [203], Azerbaijan [204], Belarus [205], Bulgaria [206], Croatia [207], Cyprus [64,208], Denmark [209], Finland [210], France [211–213], Germany [214,215], Greece [216], Holland [217], Hungary [218], Iceland [219], Ireland [220], Italy [221,222], Macedonia [202], Norway [223], Poland [224], Portugal [225], Serbia [226], Slovakia [227], Spain [228,229], Sweden [230,231], Turkey [232,233], and United Kingdom [234].

134

SEGERS AND CASTOLDI

descended from a common ancestor. On the basis of small variations in the common FV Leiden haplotype, the FV Leiden mutation has been estimated to be between 21,000 and 34,000 years old [112]. This places the origin of the mutation long after the out-of-Africa migration of modern humans, but also after the divergence of Caucasoid and Mongoloid populations, which explains why the FV Leiden mutation is confined to Caucasians. The reason for the relatively high frequency of FV Leiden in Caucasians has been a matter of debate [106]. The single origin of the mutation and the regional variations in its prevalence suggest that founder effects and genetic drift have played a major role in spreading the mutation throughout Europe. However, it has also been proposed that FV Leiden might have been maintained (and enriched) in the population due to a selective advantage, such as reduced bleeding after trauma during warfare or hunting and/or at delivery. Interestingly, it has been shown that female carriers of FV Leiden indeed have a lower risk of severe intrapartum bleeding [113] and a better hemoglobin status [114] than female noncarriers. Positive effects of FV Leiden on male (but not female) fecundity [115] and on embryo implantation [116, 117] have also been reported, but these reproductive advantages might be offset by later pregnancy complications associated with thrombophilia. Finally, a recent study suggests that FV Leiden heterozygotes (but not homozygotes) might have a substantial survival advantage in severe sepsis [118]. Although in modern times of antibiotics and oral contraceptives, FV Leiden carriership might represent a disadvantage rather than an advantage, the fact that FV Leiden is not underrepresented in people aged 90 years and older suggests that it is well compatible with successful aging [119, 120]. 4.2.2. Molecular Mechanisms By replacing Arg506 with a Gln, the FV Leiden mutation abolishes the preferential cleavage site recognized by APC on FV(a) (Fig. 4). Since APCcatalyzed cleavage at Arg506 is the first step in the mainstream pathway of FVa inactivation (see Section 3.3.1) and is also required for the conversion of single-chain FV in a functional APC-cofactor (see Section 3.3.2), both processes are affected by the mutation (Fig. 4). The effect of the FV Leiden mutation on FVa inactivation has been extensively studied in model systems using both plasma-purified [52, 55, 95, 96, 121] and recombinant FVLeiden [97, 122]. These experiments have shown that the APC-catalyzed inactivation of FVaLeiden is 10–20 times slower than the inactivation of normal FVa, but complete loss of FXa-cofactor activity is eventually achieved. In the absence of the Arg506 cleavage site, loss of cofactor activity in FVaLeiden reflects slow cleavage at Arg306, and is therefore greatly stimulated by protein S [55]. As a matter of fact, it has been shown that the presence of protein S (which stimulates cleavage at Arg306) and FXa

FACTOR V LEIDEN AND ACTIVATED PROTEIN C RESISTANCE

135

(which protects the Arg506 cleavage site in normal FVa) almost completely eliminates the difference between FVaLeiden and normal FVa in rate of APCmediated inactivation [55]. In plasma (and hence in vivo), where optimal concentrations of protein S and FXa are present, FVaLeiden and normal FVa are therefore inactivated at similar rates. This finding explains why APC resistance was poorly detected in a FXa-initiated clotting assay (see Section 4.1.1) and suggests that delayed APC-mediated inactivation of FVaLeiden cannot per se account for the pronounced APC resistance observed in plasma from FV Leiden carriers. The original observations by Dahlba¨ck et al. indicated that APC resistance was caused by an inherited deficiency of a new APC cofactor (especially needed for FVIIIa inactivation), which later turned out to be FV (see Section 4.1.1). Following the discovery of FV Leiden as the main cause of APC resistance, the ability of plasma-purified or recombinant FVLeiden to stimulate APC-mediated FVIIIa inactivation was investigated in model systems [80, 82]. Due to the lack of the Arg506 cleavage site, which must be cleaved to convert FV in a functional APC cofactor [82], FVLeiden expresses 10-fold reduced APC-cofactor activity compared to normal FV [80, 82]. Since FV is required for efficient APC-catalyzed FVIIIa inactivation, the impaired APC-cofactor activity of FVLeiden severely hampers FVIII inactivation, leading to sustained FXa generation. In summary, the hypercoagulable state associated with the FV Leiden mutation is attributable to a combination of delayed FVa and FVIIIa inactivation, resulting in increased thrombin formation. While both components contribute approximately equally to plasma APC resistance as determined with the APTT-based assay [123], failure of FVLeiden to stimulate APC-catalyzed FVIIIa inactivation is likely to be the predominant mechanism underlying thrombosis risk in FV Leiden carriers in vivo [124]. This conclusion is supported by thrombin generation experiments in FV-deficient plasma reconstituted with normal FV and/or FVLeiden to simulate plasma from heterozygous, homozygous, and pseudo-homozygous carriers of FV Leiden. FV Leiden pseudo-homzygotes are FV Leiden heterozygotes that carry a null mutation on the non-Leiden FV allele [125]. As a consequence, their plasma contains as much FVLeiden as plasma from FV Leiden heterozygotes, but no normal FV. In the presence of APC, thrombin generation in pseudo-homozygous plasma (50% FVLeiden) was the same as in homozygous plasma (100% FVLeiden) and twice as high as in heterozygous plasma (50% FVLeiden and 50% normal FV) [123], in line with the respective thrombosis risks [126]. Therefore, the thrombosis risk is not determined by the absolute amount of FVLeiden present in plasma, but rather by the presence or absence of normal FV which, besides its procoagulant activity, expresses anticoagulant activity as a cofactor of APC in FVIIIa inactivation [123].

136

SEGERS AND CASTOLDI

In addition to increasing thrombin formation and fibrin deposition, FVLeiden also has an antifibrinolytic effect, possibly due to increased activation of the thrombin-activatable fibrinolysis inhibitor (TAFI) [127, 128]. 4.2.3. Thrombosis Risk Carriership of the FV Leiden mutation confers a life-long increased risk of venous thrombosis. In the LETS case-control study, thrombosis risk was estimated to be increased 7-fold in FV Leiden heterozygotes and 80-fold in homozygotes [129]. However, these values might be overestimates, as they were not corrected for concomitant genetic and acquired risk factors. There is suggestive evidence that FV Leiden might be a stronger risk factor for deep-vein thrombosis (OR 6.0) than for pulmonary embolism (OR 2.5), a discrepancy that is referred to as the ‘‘FV Leiden paradox’’ [130, 131]. Moreover, the risk of recurrent thromboembolism in FV Leiden carriers is much lower ( 1.4) than the risk of a first thrombotic event [132, 133]. The reasons for these inconsistencies are presently unknown, although the antifibrinolytic effect of FV Leiden (which leads to more stable thrombi) has been proposed as a possible explanation for the low risk of pulmonary embolism in FV Leiden carriers [130]. Due to its high prevalence in the (Caucasian) population, FV Leiden is found in 20% of consecutive patients with venous thrombosis and in 50% of all thrombophilic families [134]. As a consequence, FV Leiden accounts for 22% of all thrombosis cases in the general population [135]. For comparison, the attributable risk of the deficiencies of the natural anticoagulants (antithrombin, protein C, and protein S) is <5% altogether. Despite their hypercoagulable state, FV Leiden carriers have only a 10% lifetime risk of thrombosis, and the majority of them never develop clinical symptoms [136]. Accordingly, it has been shown that FV Leiden carriership is compatible with extreme longevity and successful aging [119]. Therefore, the FV Leiden mutation can be considered a mild risk factor for venous thrombosis, which only becomes symptomatic in combination with additional genetic and/or acquired risk factors. Several thrombophilic defects, including protein C deficiency, protein S deficiency, and the prothrombin G20210A mutation, have been shown to cosegregate with FV Leiden in thrombophilic families, leading to earlier and more severe thrombotic manifestations in carriers of double defects [137]. Among acquired factors, the use of oral contraceptives strongly interacts with FV Leiden, increasing the risk of young female carriers from 8-fold in nonusers to >30-fold in users [138, 139]. The identification of modulators of thrombosis risk in FV Leiden carriers beyond the well-known risk factors for venous thrombosis is an active area of research, as is the development of functional tests able to distinguish FV Leiden carriers at high and low risk of thrombosis.

FACTOR V LEIDEN AND ACTIVATED PROTEIN C RESISTANCE

137

While FV Leiden is an established risk factor for venous thrombosis, its association with the risk of arterial thrombosis is far less clear, if any at all. According to recent meta-analyses on thousands of patients and controls, carriership of FV Leiden might be a mild risk factor (RR 1.2) for ischemic arterial events [140] and coronary artery disease [141], particularly in highly selected subgroups of patients. However, more studies are warranted to understand the role of FV Leiden in arterial disease. A murine model of FV Leiden, in which the Leiden mutation has been introduced at the homologous position in the mouse FV gene, is currently available [142]. This model shows potential for the study of the molecular mechanisms underlying the risk of venous [142, 143] and possibly arterial thrombosis associated with FV Leiden [144].

4.3. OTHER CAUSES OF APC RESISTANCE 4.3.1. Laboratory Assessment of APC Resistance The original APC resistance assay developed by Dahlba¨ck and coworkers [7] is based on the comparison of the plasma clotting times obtained in the absence and presence of added APC after intrinsic initiation of coagulation (Fig. 3). The assay result is expressed as the ratio of the clotting times (CTþ APC/CT APC), the so-called APC-sensitivity ratio (APCsr). Usually, the APCsr is normalized against the APCsr of a reference plasma to yield the nAPCsr. When the (n)APCsr falls below a certain cutoff value, the plasma is considered to be APC-resistant. However, the (n)APCsr ranges of FV Leiden carriers and noncarriers are rather broad and partially overlap, suggesting the existence of factors modulating the APC resistance phenotype in both carriers and noncarriers of the mutation. After the discovery of APC resistance and FV Leiden as the most common risk factors for venous thrombosis, the APTT-based APC resistance assay rapidly spread to all major coagulation laboratories throughout the world. As more and more individuals were tested, it became apparent that the APTT-based APC resistance assay did not completely correlate with FV Leiden carriership as determined with a genetic test. In fact, many individuals that did not carry the mutation tested positive in the functional assay. In this way, it was soon realized that several conditions that were frequent among thrombotic patients, such as elevated FVIII levels [145, 146], lupus anticoagulants [147, 148], and the use of oral contraceptives [149], could interfere with the assay outcome. To improve on the ability of the functional assay to predict the FV Leiden genotype, a modification of the assay was introduced consisting in a 1:5 predilution of the test plasma in FV-deficient plasma [150, 151]. This would eliminate any unwanted interferences, as all

138

SEGERS AND CASTOLDI

coagulation factors and inhibitors except FV are contributed by the FV-deficient plasma in this assay setup. Although the modified test was indeed a better predictor of FV Leiden genotype [152], occasional discrepancies between the genetic test and the functional assay were still observed, suggesting that other variants in the FV gene could modulate the APC-resistance phenotype. The first to be identified was the FV R2 haplotype, which was highly enriched among individuals who showed APC resistance in the modified test without carrying the FV Leiden mutation [153, 154]. Since then, several other FV gene mutations associated with plasma APC resistance have been described and characterized (Table 1). These are discussed in detail in Section 4.3.2. In addition, we now know that the APC-resistance phenotype of FV Leiden heterozygotes is strongly influenced by the degree of expression of the normal (non-Leiden) FV allele [155]. This is dramatically illustrated by the condition known as pseudo-homozygous APC resistance [125], where genetically heterozygous FV Leiden carriers show an APC resistance level (and thrombosis risk) typical of homozygotes due to complete nonexpression of the normal allele (see Section 4.2.2). Apart from the APTT-based assay, several other functional assays have been developed to probe the plasma anticoagulant response to APC (a selection is shown in Table 2). These assays differ in the assay material (full or diluted plasma), in the trigger used to activate coagulation (intrinsic or extrinsic), and in the assay endpoint (clotting, thrombin generation, or FVa/FVIIIa activity). Accordingly, they show differential sensitivities for FVa and FVIIIa inactivation. In particular, while the APTT-based and endogenous thrombin potential (ETP)-based assays [156] probe the overall anticoagulant response of plasma to APC, the prothrombinase-based assay [157] is only sensitive to FVa inactivation, and the ImmunochromW assay [158] is only sensitive to FVIIIa inactivation. Although all assays detect FV Leiden-associated APC resistance, each of them is also sensitive to quantitative and qualitative defects of other plasma factors (Table 2), the so-called ‘‘assay determinants’’ [159]. This automatically defines (genetic or acquired) alterations in these coagulation factors or inhibitors as new (FV-independent) causes of APC resistance. These additional determinants of APC resistance are discussed in Section 4.3.3.

4.3.2. FV Gene Mutations Associated with APC Resistance In addition to FV Leiden, several other mutations and polymorphisms in the FV gene have been shown to cause some degree of plasma APC resistance. The most important are listed in Table 1 and briefly discussed in the following section.

TABLE 1 FV GENE VARIANTS ASSOCIATED WITH APC RESISTANCE

Common name

Nucleotide change

Amino acid change

Ethnic distribution

FV level

APC-mediated inactivation

APC-cofactor activity

References

FV Leiden FV Cambridge FV Hong Kong FV Liverpool FV R2a

G1691A G1091C A1090G T1250C A6755G

Arg506Gln Arg306Thr Arg306Gly Ile359Thr Asp2194Gly

Caucasians (5%) Caucasians (sporadic) Chinese (4.5%) Caucasians (single family) All populations ( 10%)

#

## # # #

## # # ## #

[8, 52, 80] [160, 166] [164, 166] [168, 169] [153, 154, 174]

Percentages represent prevalences (carrier frequencies); , no change. FV R2 denotes a whole haplotype including several genetically linked polymorphisms in the FV gene. Asp2194Gly is most probably the functional variant. The prevalence of FV R2 varies widely among populations. a

TABLE 2 APC RESISTANCE ASSAYS Assay APTT-based ETP-based Prothrombinase-based ImmunochromW

Assay material

Trigger

Endpoint

Major determinants

References

Full plasma Full plasma Diluted plasma (1:1000) Diluted plasma (1:126)

Kaolin/PL/Ca2þ APC TF/PL/Ca2þ APC Thrombin/PL/Ca2þ APC Thrombin/PL/Ca2þ APC

Clotting time Thrombin generation FVa activity FVIIIa activity

FVIII, PT, FV TFPI, PS, PT, FV, FX FV FV, FVIII, PS

[7] [156] [157] [158]

ETP, endogenous thrombin potential; PL, phospholipids; TF, tissue factor; PT, prothrombin; PS, protein S.

FACTOR V LEIDEN AND ACTIVATED PROTEIN C RESISTANCE

141

4.3.2.1. Mutations at the Arg306 Cleavage Site (FV Cambridge and FV Hong Kong). Two distinct mutations affecting the Arg306 cleavage site in FV have been reported. FV Cambridge, changing Arg306 into a Thr, was identified in a patient with APC resistance and thrombosis [160] and later sporadically found in the Caucasian population [161–163]. FV Hong Kong, replacing Arg306 with a Gly [164], is relatively prevalent (4.5%) among Hong Kong Chinese [165]. Both mutants have been expressed as recombinant molecules and shown to confer mild APC resistance when reconstituted into FV-deficient plasma [98, 166]. When APC-catalyzed FVa inactivation was studied in a model system, both Arg306-mutants showed a normal initial inactivation phase (reflecting cleavage at Arg506), but no further degradation of the Arg506-cleaved intermediate (consistent with the absence of the Arg306 cleavage site). However, in the presence of protein S, they showed a slow loss of cofactor activity, which might be due to stimulation of cleavage at Arg679 [166] or possibly at the protein S-dependent alternative APC-cleavage site located near Arg306 [167]. For both mutants, the APC-cofactor activity in FVIIIa inactivation was intermediate between that of normal FV and FVLeiden [166]. 4.3.2.2. FV Ile359Thr (FV Liverpool). FV Liverpool was identified in two siblings who had experienced severe thrombotic manifestations before the age of 20 years [168]. Both siblings carried the mutation in the ‘‘pseudohomozygous’’ condition, as a null mutation precluded the expression of the counterpart allele. The amino acid change in FV Liverpool introduces a consensus sequence for N-glycosylation at Asn357 in the heavy chain of FV. Glycosylation at Asn357 was confirmed by expressing the mutant as a recombinant molecule [169]. Moreover, FV-deficient plasma reconstituted with the mutant showed mild APC resistance. Characterization of the mutant in model systems showed that FVaLiverpool was inactivated less efficiently than normal FV (due to poor cleavage at Arg306) and had a higher protein S requirement than normal FVa for stimulation of cleavage at Arg306. Moreover, FVLiverpool had severely impaired APC-cofactor activity in FVIIIa inactivation. These effects are attributable to the steric hindrance caused by the bulky carbohydrate moiety at Asn357 [169]. 4.3.2.3. FV R2. The FV R2 haplotype, marked by the A4070G (His1299Arg) substitution, was initially reported for its lowering effect on FV levels [170]. Later, this effect was shown to be mediated by the genetically linked Asp2194Gly variant [171, 172]. The R2 haplotype was also found at a high frequency among individuals with unexplained APC resistance and among FV Leiden heterozygotes with particularly marked APC resistance [153]. The association between FV R2 and APC resistance was later confirmed and shown to be assay-dependent [154]. FVaR2 is inactivated by APC at a normal rate, but it expresses reduced APC-cofactor activity in FVIIIa

142

SEGERS AND CASTOLDI

inactivation [154]. The underlying mechanism is not fully clear [173], but it probably involves a combination of (1) amino acid substitutions in the B domain [153]; (2) a shift in the relative proportions of the two FV isoforms in favor of the more thrombogenic FV1 isoform [174]; and (3) a possible associated increase in FVIII levels [175]. FV R2 has a worldwide distribution and might therefore contribute to APC resistance in ethnic groups that lack FV Leiden. Moreover, due to its high prevalence in Caucasians (10–15%), it can be coinherited with FV Leiden leading to enhanced APC resistance in double heterozygotes [153]. Whether FV R2 carriership also increases thrombosis risk remains a matter of debate [176]. 4.3.3. Altered Levels of Coagulation Factors/Inhibitors Causing APC Resistance As part of a large population survey conducted in Italy, 2580 healthy individuals randomly extracted from the general population were tested for APC resistance (using the original APTT-based assay) on two different occasions 5 years apart. Among the subjects who did not carry the FV Leiden mutation (n¼2506), 14.3% and 11.5% turned out to be APC-resistant in the first and second visit, respectively, while 4.9% were APC-resistant in both visits [177]. These data indicate that genetic and acquired causes of APC resistance different from the FV Leiden mutation are prevalent in the general population. Some of these have been identified as increased levels of coagulation factors (particularly FVIII and prothrombin) and decreased levels of anticoagulant factors (particularly protein S and TFPI) [159]. 4.3.3.1. FVIII. Elevated FVIII levels are highly prevalent in the general population and have been recognized as a potent effector of APC resistance from the early days of APC resistance [145, 146]. In the population survey mentioned above, elevated levels of FVIII accounted for as much as onethird of FV Leiden-negative individuals that were APC-resistant in both visits [177]. Apart from the ABO blood group [178], the genetic determinants of high FVIII levels are poorly understood, although several loci showing genetic linkage to FVIII levels have been identified [179–181]. Remarkably, despite active search, FVIII gene mutations at the APC-cleavage sites in FVIIIa have never been found in conjunction with APC resistance [182–185]. 4.3.3.2. Prothrombin. High prothrombin levels, due to genetic or acquired causes, are associated with APC resistance both in the APTT-based [186–188] and ETP-based APC-resistance assay [188, 189]. The underlying reason is unclear, but might be related to the ability of prothrombin to inhibit APC-catalyzed FVa inactivation (see Section 3.3.1). In Caucasians, hyperprothrombinemia is often (though not always) caused by the G20210A mutation in the prothrombin gene [190]. Interestingly, carriers of this mutation have a more pronounced APC resistance than noncarriers with similar

FACTOR V LEIDEN AND ACTIVATED PROTEIN C RESISTANCE

143

prothrombin levels because of other causes [189]. This is due to the higher prothrombin/protein S ratio in carriers of the prothrombin G20210A mutation, as protein S counteracts APC resistance due to high prothrombin levels [188]. 4.3.3.3. Protein S. Since protein S is a cofactor of APC, low protein S levels are associated with APC resistance. However, while the ETP-based assay is also sensitive to variations in protein S levels within the normal range, the APTT-based assay only recognizes protein S levels <50% [188]. Low levels of protein S are caused by rare loss-of-function mutations in the protein S gene or by common acquired conditions such as oral contraceptive use and pregnancy [191]. 4.3.3.4. TFPI. Low levels of TFPI are only detected in the ETP-based assay [159], where tissue factor is used as a trigger of thrombin generation. Although TFPI has no direct role in the protein C pathway, it is a specific inhibitor of FXa. Therefore, failure to downregulate FXa at low TFPI levels could lead to excessive protection of FVa from APC-catalyzed inactivation. Low TFPI levels are usually the result of acquired conditions such as oral contraceptive use [192] and pregnancy [193]. The ability of altered levels of coagulation factors and inhibitors to produce an APC resistance phenotype provides an explanation for acquired forms of APC resistance, such as those related to oral contraceptive use [194], pregnancy [195, 196], and cancer [197]. In particular, decreased levels of protein S and TFPI as well as increased prothrombin levels underlie acquired APC resistance in oral contraceptive users [198], while low levels of protein S have been implicated in APC resistance associated with hematological malignancies [199, 200]. Due to its high sensitivity to all relevant determinants (Table 2), the ETP-based assay appears the most suitable for studies on acquired forms of APC resistance. In conclusion, although FV Leiden remains the most important cause of plasma APC resistance (at least in the Caucasian population), APC resistance is actually a complex phenotype influenced by several genetic and acquired factors [201]. In this respect, it should be emphasized that not only FV Leiden-related APC resistance, but also APC resistance due to other cause, dose-dependently increases the risk of venous thrombosis [103, 104].

5. Conclusions and Perspectives The discovery of APC resistance and the FV Leiden mutation has opened a new era in thrombosis research by providing an explanation for a large fraction of previously unexplained thrombosis cases. Not only FV Leiden,

144

SEGERS AND CASTOLDI

but also several other genetic and acquired risk factors for venous thrombosis (including elevated levels of prothrombin and FVIII, decreased levels of protein S and TFPI, as well as pregnancy, oral contraceptive use, and cancer) act by inducing an APC resistance state, which underscores the centrality of the protein C pathway in the downregulation of coagulation. Despite the impressive amount of knowledge accumulated in the past 15 years, a number of questions still remain unanswered. In particular, little information is available on the genetic factors involved in the fine modulation of the APC resistance phenotype in FV Leiden carriers. Moreover, the molecular mechanisms of APC resistance associated with altered levels of coagulation factors and inhibitors are not understood in detail. Finally, the genetic bases of APC resistance in populations that lack FV Leiden are largely unexplored. These open issues represent new challenges for future research. ACKNOWLEDGMENT E. Castoldi is supported by a VIDI grant (nr. 917-76-312) from the Dutch Organisation for Scientific Research (NWO).

REFERENCES [1] F.R. Rosendaal, Venous thrombosis: a multicausal disease, Lancet 353 (9159) (1999) 1167–1173. [2] B. Dahlba¨ck, Advances in understanding pathogenic mechanisms of thrombophilic disorders, Blood 112 (1) (2008) 19–27. [3] B. Dahlba¨ck, B.O. Villoutreix, The anticoagulant protein C pathway, FEBS Lett. 579 (15) (2005) 3310–3316. [4] J.H. Griffin, B. Evatt, T.S. Zimmerman, A.J. Kleiss, C. Wideman, Deficiency of protein C in congenital thrombotic disease, J. Clin. Invest. 68 (5) (1981) 1370–1373. [5] P.C. Comp, C.T. Esmon, Recurrent venous thromboembolism in patients with a partial deficiency of protein S, N. Engl. J. Med. 311 (24) (1984) 1525–1528. [6] H.P. Schwarz, M. Fischer, P. Hopmeier, M.A. Batard, J.H. Griffin, Plasma protein S deficiency in familial thrombotic disease, Blood 64 (6) (1984) 1297–1300. [7] B. Dahlba¨ck, M. Carlsson, P.J. Svensson, Familial thrombophilia due to a previously unrecognized mechanism characterized by poor anticoagulant response to activated protein C: prediction of a cofactor to activated protein C, Proc. Natl. Acad. Sci. USA 90 (3) (1993) 1004–1008. [8] R.M. Bertina, B.P. Koeleman, T. Koster, F.R. Rosendaal, R.J. Dirven, H. de Ronde, et al., Mutation in blood coagulation factor V associated with resistance to activated protein C, Nature 369 (6475) (1994) 64–67. [9] J. Stenflo, A new vitamin K-dependent protein. Purification from bovine plasma and preliminary characterization, J. Biol. Chem. 251 (2) (1976) 355–363. [10] J.P. Miletich, G.J. Broze, Jr., Beta protein C is not glycosylated at asparagine 329. The rate of translation may influence the frequency of usage at asparagine-X-cysteine sites, J. Biol. Chem. 265 (19) (1990) 11397–11404.

FACTOR V LEIDEN AND ACTIVATED PROTEIN C RESISTANCE

145

[11] W. Kisiel, Human plasma protein C: isolation, characterization, and mechanism of activation by alpha-thrombin, J. Clin. Invest. 64 (3) (1979) 761–769. [12] L.O. Mosnier, B.V. Zlokovic, J.H. Griffin, The cytoprotective protein C pathway, Blood 109 (8) (2007) 3161–3172. [13] C.J. Jackson, M. Xue, Activated protein C—an anticoagulant that does more than stop clots, Int. J. Biochem. Cell Biol. 40 (12) (2008) 2692–2697. [14] P. Patracchini, V. Aiello, P. Palazzi, E. Calzolari, F. Bernardi, Sublocalization of the human protein C gene on chromosome 2q13–q14, Hum. Genet. 81 (2) (1989) 191–192. [15] D.C. Foster, S. Yoshitake, E.W. Davie, The nucleotide sequence of the gene for human protein C, Proc. Natl. Acad. Sci. USA 82 (14) (1985) 4673–4677. [16] J. Stenflo, Y. Stenberg, A. Muranyi, Calcium-binding EGF-like modules in coagulation proteinases: function of the calcium ion in module interactions, Biochim. Biophys. Acta 1477 (1–2) (2000) 51–63. [17] R.G. Di Scipio, M.A. Hermodson, S.G. Yates, E.W. Davie, A comparison of human prothrombin, factor IX (Christmas factor), factor X (Stuart factor), and protein S, Biochemistry 16 (4) (1977) 698–706. [18] B. Dahlba¨ck, J. Stenflo, High molecular weight complex in human plasma between vitamin K-dependent protein S and complement component C4b-binding protein, Proc. Natl. Acad. Sci. USA 78 (4) (1981) 2512–2516. [19] F.J. Walker, Regulation of activated protein C by a new protein. A possible function for bovine protein S, J. Biol. Chem. 255 (12) (1980) 5521–5524. [20] T.M. Hackeng, K.M. Sere, G. Tans, J. Rosing, Protein S stimulates inhibition of the tissue factor pathway by tissue factor pathway inhibitor, Proc. Natl. Acad. Sci. USA 103 (9) (2006) 3106–3111. [21] M. Ndonwi, G. Broze, Jr., Protein S enhances the tissue factor pathway inhibitor inhibition of factor Xa but not its inhibition of factor VIIa-tissue factor, J. Thromb. Haemost. 6 (6) (2008) 1044–1046. [22] H.K. Ploos van Amstel, P.H. Reitsma, C.P. van der Logt, R.M. Bertina, Intron–exon organization of the active human protein S gene PS alpha and its pseudogene PS beta: duplication and silencing during primate evolution, Biochemistry 29 (34) (1990) 7853–7861. [23] B. Dahlba¨ck, A. Lundwall, J. Stenflo, Primary structure of bovine vitamin K-dependent protein S, Proc. Natl. Acad. Sci. USA 83 (12) (1986) 4199–4203. [24] S.M. Rezende, R.E. Simmonds, D.A. Lane, Coagulation, inflammation, and apoptosis: different roles for protein S and the protein S-C4b binding protein complex, Blood 103 (4) (2004) 1192–1201. [25] C.T. Esmon, W.G. Owen, Identification of an endothelial cell cofactor for thrombincatalyzed activation of protein C, Proc. Natl. Acad. Sci. USA 78 (4) (1981) 2249–2252. [26] N.L. Esmon, W.G. Owen, C.T. Esmon, Isolation of a membrane-bound cofactor for thrombin-catalyzed activation of protein C, J. Biol. Chem. 257 (2) (1982) 859–864. [27] D.J. Stearns-Kurosawa, S. Kurosawa, J.S. Mollica, G.L. Ferrell, C.T. Esmon, The endothelial cell protein C receptor augments protein C activation by the thrombin– thrombomodulin complex, Proc. Natl. Acad. Sci. USA 93 (19) (1996) 10212–10216. [28] K. Fukudome, X. Ye, N. Tsuneyoshi, O. Tokunaga, K. Sugawara, H. Mizokami, et al., Activation mechanism of anticoagulant protein C in large blood vessels involving the endothelial cell protein C receptor, J. Exp. Med. 187 (7) (1998) 1029–1035. [29] M.F. Scully, C.H. Toh, H. Hoogendoorn, R.P. Manuel, M.E. Nesheim, S. Solymoss, et al., Activation of protein C and its distribution between its inhibitors, protein C inhibitor, alpha 1-antitrypsin and alpha 2-macroglobulin, in patients with disseminated intravascular coagulation, Thromb. Haemost. 69 (5) (1993) 448–453.

146

SEGERS AND CASTOLDI

[30] H. Ishii, H.H. Salem, C.E. Bell, E.A. Laposata, P.W. Majerus, Thrombomodulin, an endothelial anticoagulant protein, is absent from the human brain, Blood 67 (2) (1986) 362–365. [31] J. Ye, L.W. Liu, C.T. Esmon, A.E. Johnson, The fifth and sixth growth factor-like domains of thrombomodulin bind to the anion-binding exosite of thrombin and alter its specificity, J. Biol. Chem. 267 (16) (1992) 11023–11028. [32] C.T. Esmon, N.L. Esmon, K.W. Harris, Complex formation between thrombin and thrombomodulin inhibits both thrombin-catalyzed fibrin formation and factor V activation, J. Biol. Chem. 257 (14) (1982) 7944–7947. [33] N.L. Esmon, R.C. Carroll, C.T. Esmon, Thrombomodulin blocks the ability of thrombin to activate platelets, J. Biol. Chem. 258 (20) (1983) 12238–12242. [34] J. Polgar, I. Lerant, L. Muszbek, R. Machovich, Thrombomodulin inhibits the activation of factor XIII by thrombin, Thromb. Res. 43 (5) (1986) 685–690. [35] A.E. Johnson, N.L. Esmon, T.M. Laue, C.T. Esmon, Structural changes required for activation of protein C are induced by Ca2þ binding to a high affinity site that does not contain gamma-carboxyglutamic acid, J. Biol. Chem. 258 (9) (1983) 5554–5560. [36] A.R. Rezaie, S.T. Cooper, F.C. Church, C.T. Esmon, Protein C inhibitor is a potent inhibitor of the thrombin–thrombomodulin complex, J. Biol. Chem. 270 (43) (1995) 25336–25339. [37] K. Fukudome, C.T. Esmon, Identification, cloning, and regulation of a novel endothelial cell protein C/activated protein C receptor, J. Biol. Chem. 269 (42) (1994) 26486–26491. [38] Z. Laszik, A. Mitro, F.B. Taylor, Jr., G. Ferrell, C.T. Esmon, Human protein C receptor is present primarily on endothelium of large blood vessels: implications for the control of the protein C pathway, Circulation 96 (10) (1997) 3633–3640. [39] N.L. Esmon, L.E. DeBault, C.T. Esmon, Proteolytic formation and properties of gammacarboxyglutamic acid-domainless protein C, J. Biol. Chem. 258 (9) (1983) 5548–5553. [40] L.M. Regan, J.S. Mollica, A.R. Rezaie, C.T. Esmon, The interaction between the endothelial cell protein C receptor and protein C is dictated by the gamma-carboxyglutamic acid domain of protein C, J. Biol. Chem. 272 (42) (1997) 26279–26284. [41] A. Slungaard, J.A. Fernandez, J.H. Griffin, N.S. Key, J.R. Long, D.J. Piegors, et al., Platelet factor 4 enhances generation of activated protein C in vitro and in vivo, Blood 102 (1) (2003) 146–151. [42] N. Diez, R. Montes, A. Alonso, P. Medina, S. Navarro, F. Espana, et al., Association of increased fibrinogen concentration with impaired activation of anticoagulant protein C, J. Thromb. Haemost. 4 (2) (2006) 398–402. [43] W.H. Kane, E.W. Davie, Blood coagulation factors V and VIII: structural and functional similarities and their relationship to hemorrhagic and thrombotic disorders, Blood 71 (3) (1988) 539–555. [44] K. Segers, B. Dahlba¨ck, G.A. Nicolaes, Coagulation factor V and thrombophilia: background and mechanisms, Thromb. Haemost. 98 (3) (2007) 530–542. [45] P.J. Fay, Factor VIII structure and function, Int. J. Hematol. 83 (2) (2006) 103–108. [46] F.J. Walker, Regulation of activated protein C by protein S. The role of phospholipid in factor Va inactivation, J. Biol. Chem. 256 (21) (1981) 11128–11131. [47] H.M. Bakker, G. Tans, T. Janssen-Claessen, M.C. Thomassen, H.C. Hemker, J.H. Griffin, et al., The effect of phospholipids, calcium ions and protein S on rate constants of human factor Va inactivation by activated human protein C, Eur. J. Biochem. 208 (1) (1992) 171–178. [48] M.D. Smirnov, C.T. Esmon, Phosphatidylethanolamine incorporation into vesicles selectively enhances factor Va inactivation by activated protein C, J. Biol. Chem. 269 (2) (1994) 816–819.

FACTOR V LEIDEN AND ACTIVATED PROTEIN C RESISTANCE

147

[49] E.A. Norstrøm, M. Steen, S. Tran, B. Dahlba¨ck, Importance of protein S and phospholipid for activated protein C-mediated cleavages in factor Va, J. Biol. Chem. 278 (27) (2003) 24904–24911. [50] J.A. Oliver, D.M. Monroe, F.C. Church, H.R. Roberts, M. Hoffman, Activated protein C cleaves factor Va more efficiently on endothelium than on platelet surfaces, Blood 100 (2) (2002) 539–546. [51] M. Kalafatis, M.D. Rand, K.G. Mann, The mechanism of inactivation of human factor V and human factor Va by activated protein C, J. Biol. Chem. 269 (50) (1994) 31869–31880. [52] G.A. Nicolaes, G. Tans, M.C. Thomassen, H.C. Hemker, I. Pabinger, K. Va´radi, et al., Peptide bond cleavages and loss of functional activity during inactivation of factor Va and factor VaR506Q by activated protein C, J. Biol. Chem. 270 (36) (1995) 21158–21166. [53] E.R. Guinto, C.T. Esmon, Loss of prothrombin and of factor Xa-factor Va interactions upon inactivation of factor Va by activated protein C, J. Biol. Chem. 259 (22) (1984) 13986–13992. [54] A.J. Gale, X. Xu, J.L. Pellequer, E.D. Getzoff, J.H. Griffin, Interdomain engineered disulfide bond permitting elucidation of mechanisms of inactivation of coagulation factor Va by activated protein C, Protein Sci. 11 (9) (2002) 2091–2101. [55] J. Rosing, L. Hoekema, G.A. Nicolaes, M.C. Thomassen, H.C. Hemker, K. Va´radi, et al., Effects of protein S and factor Xa on peptide bond cleavages during inactivation of factor Va and factor VaR506Q by activated protein C, J. Biol. Chem. 270 (46) (1995) 27852–27858. [56] B. Dahlba¨ck, Inhibition of protein Ca cofactor function of human and bovine protein S by C4b-binding protein, J. Biol. Chem. 261 (26) (1986) 12022–12027. [57] R.H. van de Poel, J.C. Meijers, J. Rosing, G. Tans, B.N. Bouma, C4b-binding protein protects coagulation factor Va from inactivation by activated protein C, Biochemistry 39 (47) (2000) 14543–14548. [58] L.F. Maurissen, M.C. Thomassen, G.A. Nicolaes, B. Dahlba¨ck, G. Tans, J. Rosing, et al., Re-evaluation of the role of the protein S-C4b binding protein complex in activated protein C-catalyzed factor Va-inactivation, Blood 111 (6) (2008) 3034–3041. [59] S. Yegneswaran, G.M. Wood, C.T. Esmon, A.E. Johnson, Protein S alters the active site location of activated protein C above the membrane surface. A fluorescence resonance energy transfer study of topography, J. Biol. Chem. 272 (40) (1997) 25013–25021. [60] S. Yegneswaran, M.D. Smirnov, O. Safa, N.L. Esmon, C.T. Esmon, A.E. Johnson, Relocating the active site of activated protein C eliminates the need for its protein S cofactor. A fluorescence resonance energy transfer study, J. Biol. Chem. 274 (9) (1999) 5462–5468. [61] M.E. Nesheim, W.M. Canfield, W. Kisiel, K.G. Mann, Studies of the capacity of factor Xa to protect factor Va from inactivation by activated protein C, J. Biol. Chem. 257 (3) (1982) 1443–1447. [62] K. Suzuki, J. Stenflo, B. Dahlba¨ck, B. Teodorsson, Inactivation of human coagulation factor V by activated protein C, J. Biol. Chem. 258 (3) (1983) 1914–1920. [63] E.A. Norstrøm, S. Tran, M. Steen, B. Dahlba¨ck, Effects of factor Xa and protein S on the individual activated protein C-mediated cleavages of coagulation factor Va, J. Biol. Chem. 281 (42) (2006) 31486–31494. [64] E. Castoldi, B. Lunghi, F. Mingozzi, P. Ioannou, G. Marchetti, F. Bernardi, New coagulation factor V gene polymorphisms define a single and infrequent haplotype underlying the factor V Leiden mutation in Mediterranean populations and Indians, Thromb. Haemost. 78 (3) (1997) 1037–1041.

148

SEGERS AND CASTOLDI

[65] C.A. Mitchell, S.M. Jane, H.H. Salem, Inhibition of the anticoagulant activity of protein S by prothrombin, J. Clin. Invest. 82 (6) (1988) 2142–2147. [66] M.D. Smirnov, O. Safa, N.L. Esmon, C.T. Esmon, Inhibition of activated protein C anticoagulant activity by prothrombin, Blood 94 (11) (1999) 3839–3846. [67] S. Tran, E. Norstrøm, B. Dahlba¨ck, Effects of prothrombin on the individual activated protein C-mediated cleavages of coagulation factor Va, J. Biol. Chem. 283 (11) (2008) 6648–6655. [68] S. Yegneswaran, P.M. Nguyen, A.J. Gale, J.H. Griffin, Prothrombin amino terminal region helps protect coagulation factor Va from proteolytic inactivation by activated protein C, Thromb. Haemost. 101 (1) (2009) 55–61. [69] S. Solymoss, M.M. Tucker, P.B. Tracy, Kinetics of inactivation of membrane-bound factor Va by activated protein C. Protein S modulates factor Xa protection, J. Biol. Chem. 263 (29) (1988) 14884–14890. [70] R.J. Preston, S. Tran, J.A. Johnson, F.N. Ainle, S. Harmon, B. White, et al., Platelet factor 4 impairs the anticoagulant activity of activated protein C, J. Biol. Chem. 284 (9) (2009) 5869–5875. [71] P.J. Fay, Factor VIII structure and function, Thromb. Haemost. 70 (1) (1993) 63–67. [72] L.M. Regan, B.J. Lamphear, C.F. Huggins, F.J. Walker, P.J. Fay, Factor IXa protects factor VIIIa from activated protein C. Factor IXa inhibits activated protein C-catalyzed cleavage of factor VIIIa at Arg562, J. Biol. Chem. 269 (13) (1994) 9445–9452. [73] P.J. Fay, T.L. Beattie, L.M. Regan, L.M. O’Brien, R.J. Kaufman, Model for the factor VIIIa-dependent decay of the intrinsic factor Xase. Role of subunit dissociation and factor IXa-catalyzed proteolysis, J. Biol. Chem. 271 (11) (1996) 6027–6032. [74] P.J. Fay, T.M. Smudzin, F.J. Walker, Activated protein C-catalyzed inactivation of human factor VIII and factor VIIIa. Identification of cleavage sites and correlation of proteolysis with cofactor activity, J. Biol. Chem. 266 (30) (1991) 20139–20145. [75] P.J. Fay, F.J. Walker, Inactivation of human factor VIII by activated protein C: evidence that the factor VIII light chain contains the activated protein C binding site, Biochim. Biophys. Acta 994 (2) (1989) 142–148. [76] J.A. Koedam, J.C. Meijers, J.J. Sixma, B.N. Bouma, Inactivation of human factor VIII by activated protein C. Cofactor activity of protein S and protective effect of von Willebrand factor, J. Clin. Invest. 82 (4) (1988) 1236–1243. [77] B. Dahlba¨ck, B. Hildebrand, Inherited resistance to activated protein C is corrected by anticoagulant cofactor activity found to be a property of factor V, Proc. Natl. Acad. Sci. USA 91 (4) (1994) 1396–1400. [78] L. Shen, B. Dahlba¨ck, Factor V and protein S as synergistic cofactors to activated protein C in degradation of factor VIIIa, J. Biol. Chem. 269 (29) (1994) 18735–18738. [79] D. Lu, M. Kalafatis, K.G. Mann, G.L. Long, Comparison of activated protein C/protein S-mediated inactivation of human factor VIII and factor V, Blood 87 (11) (1996) 4708–4717. [80] K. Va´radi, J. Rosing, G. Tans, I. Pabinger, B. Keil, H.P. Schwarz, Factor V enhances the cofactor function of protein S in the APC-mediated inactivation of factor VIII: influence of the factor VR506Q mutation, Thromb. Haemost. 76 (2) (1996) 208–214. [81] E. Thorelli, R.J. Kaufman, B. Dahlba¨ck, The C-terminal region of the factor V B-domain is crucial for the anticoagulant activity of factor V, J. Biol. Chem. 273 (26) (1998) 16140–16145. [82] E. Thorelli, R.J. Kaufman, B. Dahlba¨ck, Cleavage of factor V at Arg 506 by activated protein C and the expression of anticoagulant activity of factor V, Blood 93 (8) (1999) 2552–2558.

FACTOR V LEIDEN AND ACTIVATED PROTEIN C RESISTANCE

149

[83] L. Shen, X. He, B. Dahlba¨ck, Synergistic cofactor function of factor V and protein S to activated protein C in the inactivation of the factor VIIIa–factor IXa complex—species specific interactions of components of the protein C anticoagulant system, Thromb. Haemost. 78 (3) (1997) 1030–1036. [84] A.J. Gale, T.J. Cramer, D. Rozenshteyn, J.R. Cruz, Detailed mechanisms of the inactivation of factor VIIIa by activated protein C in the presence of its cofactors, protein S and factor V, J. Biol. Chem. 283 (24) (2008) 16355–16362. [85] R.H. van de Poel, J.C. Meijers, B.N. Bouma, C4b-binding protein inhibits the factor V-dependent but not the factor V-independent cofactor activity of protein S in the activated protein C-mediated inactivation of factor VIIIa, Thromb. Haemost. 85 (5) (2001) 761–765. [86] K. Nogami, H. Wakabayashi, P.J. Fay, Mechanisms of factor Xa-catalyzed cleavage of the factor VIIIa A1 subunit resulting in cofactor inactivation, J. Biol. Chem. 278 (19) (2003) 16502–16509. [87] K. Nogami, H. Wakabayashi, K. Schmidt, P.J. Fay, Altered interactions between the A1 and A2 subunits of factor VIIIa following cleavage of A1 subunit by factor Xa, J. Biol. Chem. 278 (3) (2003) 1634–1641. [88] P.J. Fay, T. Beattie, C.F. Huggins, L.M. Regan, Factor VIIIa A2 subunit residues 558–565 represent a factor IXa interactive site, J. Biol. Chem. 269 (32) (1994) 20522–20527. [89] L.M. O’Brien, M. Mastri, P.J. Fay, Regulation of factor VIIIa by human activated protein C and protein S: inactivation of cofactor in the intrinsic factor Xase, Blood 95 (5) (2000) 1714–1720. [90] F. Varfaj, J. Neuberg, P.V. Jenkins, H. Wakabayashi, P.J. Fay, Role of P1 residues Arg336 and Arg562 in the activated-Protein-C-catalysed inactivation of Factor VIIIa, Biochem. J. 396 (2) (2006) 355–362. [91] X. Sun, B. Evatt, J.H. Griffin, Blood coagulation factor Va abnormality associated with resistance to activated protein C in venous thrombophilia, Blood 83 (11) (1994) 3120–3125. [92] B. Zo¨ller, B. Dahlba¨ck, Linkage between inherited resistance to activated protein C and factor V gene mutation in venous thrombosis, Lancet 343 (8912) (1994) 1536–1538. [93] J. Voorberg, J. Roelse, R. Koopman, H. Buller, F. Berends, J.W. ten Cate, et al., Association of idiopathic venous thromboembolism with single point-mutation at Arg506 of factor V, Lancet 343 (8912) (1994) 1535–1536. [94] J.S. Greengard, X. Sun, X. Xu, J.A. Fernandez, J.H. Griffin, B. Evatt, Activated protein C resistance caused by Arg506Gln mutation in factor Va, Lancet 343 (8909) (1994) 1361–1362. [95] M.J. Heeb, Y. Kojima, J.S. Greengard, J.H. Griffin, Activated protein C resistance: molecular mechanisms based on studies using purified Gln506-factor V, Blood 85 (12) (1995) 3405–3411. [96] C. Aparicio, B. Dahlba¨ck, Molecular mechanisms of activated protein C resistance. Properties of factor V isolated from an individual with homozygosity for the Arg506 to Gln mutation in the factor V gene, Biochem. J. 313 (Pt. 2) (1996) 467–472. [97] M.J. Heeb, A. Rehemtulla, M. Moussalli, Y. Kojima, R.J. Kaufman, Importance of individual activated protein C cleavage site regions in coagulation factor V for factor Va inactivation and for factor Xa activation, Eur. J. Biochem. 260 (1) (1999) 64–75. [98] M. van der Neut Kolfschoten, R.J. Dirven, G. Tans, J. Rosing, H.L. Vos, R.M. Bertina, The activated protein C (APC)-resistant phenotype of APC cleavage site mutants of recombinant factor V in a reconstituted plasma model, Blood Coagul. Fibrinolysis 13 (3) (2002) 207–215. [99] B. Dahlba¨ck, The discovery of activated protein C resistance, J. Thromb. Haemost. 1 (1) (2003) 3–9.

150

SEGERS AND CASTOLDI

[100] T. Koster, F.R. Rosendaal, H. de Ronde, E. Briet, J.P. Vandenbroucke, R.M. Bertina, Venous thrombosis due to poor anticoagulant response to activated protein C: Leiden Thrombophilia Study, Lancet 342 (8886–8887) (1993) 1503–1506. [101] J.H. Griffin, B. Evatt, C. Wideman, J.A. Fernandez, Anticoagulant protein C pathway defective in majority of thrombophilic patients, Blood 82 (7) (1993) 1989–1993. [102] P.J. Svensson, B. Dahlba¨ck, Resistance to activated protein C as a basis for venous thrombosis, N. Engl. J. Med. 330 (8) (1994) 517–522. [103] M.C. de Visser, F.R. Rosendaal, R.M. Bertina, A reduced sensitivity for activated protein C in the absence of factor V Leiden increases the risk of venous thrombosis, Blood 93 (4) (1999) 1271–1276. [104] F. Rodeghiero, A. Tosetto, Activated protein C resistance and factor V Leiden mutation are independent risk factors for venous thromboembolism, Ann. Intern. Med. 130 (8) (1999) 643–650. [105] D.C. Rees, M. Cox, J.B. Clegg, World distribution of factor V Leiden, Lancet 346 (8983) (1995) 1133–1134. [106] D.C. Rees, The population genetics of factor V Leiden (Arg506Gln), Br. J. Haematol. 95 (4) (1996) 579–586. [107] F.H. Herrmann, M. Koesling, W. Schroder, R. Altman, R. Jimenez Bonilla, S. Lopaciuk, et al., Prevalence of factor V Leiden mutation in various populations, Genet. Epidemiol. 14 (4) (1997) 403–411. [108] H. Cai, B. Hua, L. Fan, Q. Wang, S. Wang, Y. Zhao, A novel mutation (G2172C) of factor V gene in a chinese family with activated protein C resistance associated with venous thromboembolism, J. Thromb. Haemost. 7 (Suppl. 2) (2009) 1044 (Abstract). [109] G. Lucotte, G. Mercier, Population genetics of factor V Leiden in Europe, Blood Cells Mol. Dis. 27 (2) (2001) 362–367. [110] M.J. Cox, D.C. Rees, J.J. Martinson, J.B. Clegg, Evidence for a single origin of factor V Leiden, Br. J. Haematol. 92 (4) (1996) 1022–1025. [111] B. Zo¨ller, L. Norlund, H. Leksell, J.E. Nilsson, H. von Schenck, U. Rosen, et al., High prevalence of the FVR506Q mutation causing APC resistance in a region of southern Sweden with a high incidence of venous thrombosis, Thromb. Res. 83 (6) (1996) 475–477. [112] A. Zivelin, J.H. Griffin, X. Xu, I. Pabinger, M. Samama, J. Conard, et al., A single genetic origin for a common Caucasian risk factor for venous thrombosis, Blood 89 (2) (1997) 397–402. [113] P.G. Lindqvist, P.J. Svensson, B. Dahlba¨ck, K. Marsal, Factor V Q506 mutation (activated protein C resistance) associated with reduced intrapartum blood loss—a possible evolutionary selection mechanism, Thromb. Haemost. 79 (1) (1998) 69–73. [114] P.G. Lindqvist, B. Zo¨ller, B. Dahlba¨ck, Improved hemoglobin status and reduced menstrual blood loss among female carriers of factor V Leiden—an evolutionary advantage? Thromb. Haemost. 86 (4) (2001) 1122–1123. [115] F.M. van Dunne, A.J. de Craen, B.T. Heijmans, F.M. Helmerhorst, R.G. Westendorp, Gender-specific association of the factor V Leiden mutation with fertility and fecundity in a historic cohort. The Leiden 85-Plus Study, Hum. Reprod. 21 (4) (2006) 967–971. [116] W. Gopel, M. Ludwig, A.K. Junge, T. Kohlmann, K. Diedrich, J. Moller, Selection pressure for the factor-V-Leiden mutation and embryo implantation, Lancet 358 (9289) (2001) 1238–1239. [117] F.M. van Dunne, C.J. Doggen, M. Heemskerk, F.R. Rosendaal, F.M. Helmerhorst, Factor V Leiden mutation in relation to fecundity and miscarriage in women with venous thrombosis, Hum. Reprod. 20 (3) (2005) 802–806.

FACTOR V LEIDEN AND ACTIVATED PROTEIN C RESISTANCE

151

[118] B.A. Kerlin, S.B. Yan, B.H. Isermann, J.T. Brandt, R. Sood, B.R. Basson, et al., Survival advantage associated with heterozygous factor V Leiden mutation in patients with severe sepsis and in mouse endotoxemia, Blood 102 (9) (2003) 3085–3092. [119] D. Mari, P.M. Mannucci, F. Duca, S. Bertolini, C. Franceschi, Mutant factor V (Arg506Gln) in healthy centenarians, Lancet 347 (9007) (1996) 1044. [120] D.C. Rees, Y.T. Liu, M.J. Cox, P. Elliott, J.S. Wainscoat, Factor V Leiden and thermolabile methylenetetrahydrofolate reductase in extreme old age, Thromb. Haemost. 78 (5) (1997) 1357–1359. [121] M. Kalafatis, R.M. Bertina, M.D. Rand, K.G. Mann, Characterization of the molecular defect in factor VR506Q, J. Biol. Chem. 270 (8) (1995) 4053–4057. [122] J.O. Egan, M. Kalafatis, K.G. Mann, The effect of Arg306–>Ala and Arg506–>Gln substitutions in the inactivation of recombinant human factor Va by activated protein C and protein S, Protein Sci. 6 (9) (1997) 2016–2027. [123] E. Castoldi, J.M. Brugge, G.A. Nicolaes, D. Girelli, G. Tans, J. Rosing, Impaired APC cofactor activity of factor V plays a major role in the APC resistance associated with the factor V Leiden (R506Q) and R2 (H1299R) mutations, Blood 103 (11) (2004) 4173–4179. [124] E. Castoldi, J. Rosing, Factor V Leiden: a disorder of factor V anticoagulant function, Curr. Opin. Hematol. 11 (3) (2004) 176–181. [125] P. Simioni, A. Scudeller, P. Radossi, S. Gavasso, B. Girolami, D. Tormene, et al., "Pseudo homozygous" activated protein C resistance due to double heterozygous factor V defects (factor V Leiden mutation and type I quantitative factor V defect) associated with thrombosis: report of two cases belonging to two unrelated kindreds, Thromb. Haemost. 75 (3) (1996) 422–426. [126] P. Simioni, E. Castoldi, B. Lunghi, D. Tormene, J. Rosing, F. Bernardi, An underestimated combination of opposites resulting in enhanced thrombotic tendency, Blood 106 (7) (2005) 2363–2365. [127] L. Bajzar, M. Kalafatis, P. Simioni, P.B. Tracy, An antifibrinolytic mechanism describing the prothrombotic effect associated with factor VLeiden, J. Biol. Chem. 271 (38) (1996) 22949–22952. [128] A.C. Parker, L.V. Mundada, A.H. Schmaier, W.P. Fay, Factor VLeiden inhibits fibrinolysis in vivo, Circulation 110 (23) (2004) 3594–3598. [129] F.R. Rosendaal, T. Koster, J.P. Vandenbroucke, P.H. Reitsma, High risk of thrombosis in patients homozygous for factor V Leiden (activated protein C resistance), Blood 85 (6) (1995) 1504–1508. [130] H. Bounameaux, Factor V Leiden paradox: risk of deep-vein thrombosis but not of pulmonary embolism, Lancet 356 (9225) (2000) 182–183. [131] K.J. van Stralen, C.J. Doggen, I.D. Bezemer, E.R. Pomp, T. Lisman, F.R. Rosendaal, Mechanisms of the factor V Leiden paradox, Arterioscler. Thromb. Vasc. Biol. 28 (10) (2008) 1872–1877. [132] W.K. Ho, G.J. Hankey, D.J. Quinlan, J.W. Eikelboom, Risk of recurrent venous thromboembolism in patients with common thrombophilia: a systematic review, Arch. Intern. Med. 166 (7) (2006) 729–736. [133] A. Marchiori, L. Mosena, M.H. Prins, P. Prandoni, The risk of recurrent venous thromboembolism among heterozygous carriers of factor V Leiden or prothrombin G20210A mutation. A systematic review of prospective studies, Haematologica 92 (8) (2007) 1107–1114. [134] R.P. Lensen, R.M. Bertina, H. de Ronde, J.P. Vandenbroucke, F.R. Rosendaal, Venous thrombotic risk in family members of unselected individuals with factor V Leiden, Thromb. Haemost. 83 (6) (2000) 817–821.

152

SEGERS AND CASTOLDI

[135] F.R. Rosendaal, Risk factors for venous thrombotic disease, Thromb. Haemost. 82 (2) (1999) 610–619. [136] J.A. Heit, J.L. Sobell, H. Li, S.S. Sommer, The incidence of venous thromboembolism among Factor V Leiden carriers: a community-based cohort study, J. Thromb. Haemost. 3 (2) (2005) 305–311. [137] R.F. Franco, P.H. Reitsma, Genetic risk factors of venous thrombosis, Hum. Genet. 109 (4) (2001) 369–384. [138] J.P. Vandenbroucke, T. Koster, E. Briet, P.H. Reitsma, R.M. Bertina, F.R. Rosendaal, Increased risk of venous thrombosis in oral-contraceptive users who are carriers of factor V Leiden mutation, Lancet 344 (8935) (1994) 1453–1457. [139] J. Rosing, G. Tans, G.A. Nicolaes, M.C. Thomassen, R. van Oerle, P.M. van der Ploeg, et al., Oral contraceptives and venous thrombosis: different sensitivities to activated protein C in women using second- and third-generation oral contraceptives, Br. J. Haematol. 97 (1) (1997) 233–238. [140] R.J. Kim, R.C. Becker, Association between factor V Leiden, prothrombin G20210A, and methylenetetrahydrofolate reductase C677T mutations and events of the arterial circulatory system: a meta-analysis of published studies, Am. Heart J. 146 (6) (2003) 948–957. [141] Z. Ye, E.H. Liu, J.P. Higgins, B.D. Keavney, G.D. Lowe, R. Collins, et al., Seven haemostatic gene polymorphisms in coronary disease: meta-analysis of 66, 155 cases and 91, 307 controls, Lancet 367 (9511) (2006) 651–658. [142] J. Cui, D.T. Eitzman, R.J. Westrick, P.D. Christie, Z.J. Xu, A.Y. Yang, et al., Spontaneous thrombosis in mice carrying the factor V Leiden mutation, Blood 96 (13) (2000) 4222–4226. [143] D.T. Eitzman, R.J. Westrick, X. Bi, S.L. Manning, J.E. Wilkinson, G.J. Broze, et al., Lethal perinatal thrombosis in mice resulting from the interaction of tissue factor pathway inhibitor deficiency and factor V Leiden, Circulation 105 (18) (2002) 2139–2142. [144] D.T. Eitzman, R.J. Westrick, Y. Shen, P.F. Bodary, S. Gu, S.L. Manning, et al., Homozygosity for factor V Leiden leads to enhanced thrombosis and atherosclerosis in mice, Circulation 111 (14) (2005) 1822–1825. [145] C.M. Henkens, V.J. Bom, J. van der Meer, Lowered APC-sensitivity ratio related to increased factor VIII-clotting activity, Thromb. Haemost. 74 (4) (1995) 1198–1199. [146] M.A. Laffan, R. Manning, The influence of factor VIII on measurement of activated protein C resistance, Blood Coagul. Fibrinolysis 7 (8) (1996) 761–765. [147] J. Matsuda, K. Gohchi, M. Tsukamoto, M. Gotoh, N. Saitoh, K. Kawasugi, Resistance to activated protein C in systemic lupus erythematosus patients with antiphospholipid antibodies, Eur J Haematol 53 (3) (1994) 188–189. [148] S. Ehrenforth, K.P. Radtke, I. Scharrer, Acquired activated protein C-resistance in patients with lupus anticoagulants, Thromb. Haemost. 74 (2) (1995) 797–798. [149] O. Olivieri, S. Friso, F. Manzato, A. Guella, F. Bernardi, B. Lunghi, et al., Resistance to activated protein C in healthy women taking oral contraceptives, Br. J. Haematol. 91 (2) (1995) 465–470. [150] J.I. Jorquera, J.M. Montoro, M.A. Fernandez, J.A. Aznar, J. Aznar, Modified test for activated protein C resistance, Lancet 344 (8930) (1994) 1162–1163. [151] M. Trossaert, J. Conard, M.H. Horellou, M.M. Samama, H. Ireland, T.A. Bayston, et al., Modified APC resistance assay for patients on oral anticoagulants, Lancet 344 (8938) (1994) 1709. [152] H. de Ronde, R.M. Bertina, Careful selection of sample dilution and factor-V-deficient plasma makes the modified activated protein C resistance test highly specific for the factor V Leiden mutation, Blood Coagul. Fibrinolysis 10 (1) (1999) 7–17.

FACTOR V LEIDEN AND ACTIVATED PROTEIN C RESISTANCE

153

[153] F. Bernardi, E.M. Faioni, E. Castoldi, B. Lunghi, G. Castaman, E. Sacchi, et al., A factor V genetic component differing from factor V R506Q contributes to the activated protein C resistance phenotype, Blood 90 (4) (1997) 1552–1557. [154] L. Hoekema, E. Castoldi, G. Tans, D. Girelli, D. Gemmati, F. Bernardi, et al., Functional properties of factor V and factor Va encoded by the R2-gene, Thromb. Haemost. 85 (1) (2001) 75–81. [155] J.M. Brugge, P. Simioni, F. Bernardi, D. Tormene, B. Lunghi, G. Tans, et al., Expression of the normal factor V allele modulates the APC resistance phenotype in heterozygous carriers of the factor V Leiden mutation, J. Thromb. Haemost. 3 (12) (2005) 2695–2702. [156] G.A. Nicolaes, M.C. Thomassen, G. Tans, J. Rosing, H.C. Hemker, Effect of activated protein C on thrombin generation and on the thrombin potential in plasma of normal and APC-resistant individuals, Blood Coagul. Fibrinolysis 8 (1) (1997) 28–38. [157] G.A. Nicolaes, M.C. Thomassen, R. van Oerle, K. Hamulya´k, H.C. Hemker, G. Tans, et al., A prothrombinase-based assay for detection of resistance to activated protein C, Thromb. Haemost. 76 (3) (1996) 404–410. [158] K. Va´radi, J. Rosing, G. Tans, H.P. Schwarz, Influence of factor V and factor Va on APCinduced cleavage of human factor VIII, Thromb. Haemost. 73 (4) (1995) 730–731. [159] M.C. de Visser, A. van Hylckama Vlieg, G. Tans, J. Rosing, A.E. Dahm, P.M. Sandset, et al., Determinants of the APTT- and ETP-based APC sensitivity tests, J. Thromb. Haemost. 3 (7) (2005) 1488–1494. [160] D. Williamson, K. Brown, R. Luddington, C. Baglin, T. Baglin, Factor V Cambridge: a new mutation (Arg306–>Thr) associated with resistance to activated protein C, Blood 91 (4) (1998) 1140–1144. [161] A.J. ten Cate, Y.T. van de Hoek, P.H. Reitsma, H. ten Cate, P. Smits, Mutation screening for thrombophilia: two cases with factor V Cambridge without activated protein C resistance, Thromb. Haemost. 87 (5) (2002) 919–920. [162] A. Santamaria, J.M. Soria, I. Tirado, J. Mateo, I. Coll, J.C. Souto, et al., Double heterozygosity for Factor V Leiden and Factor V Cambridge mutations associated with low levels of activated protein C resistance in a Spanish thrombophilic family, Thromb. Haemost. 93 (6) (2005) 1193–1195. [163] V. Le Cam-Duchez, M.H. Chretien, P. Saugier-Veber, J.Y. Borg, Factor V Cambridge mutation and activated protein C resistance assays, Thromb. Haemost. 95 (3) (2006) 581–583. [164] W.P. Chan, C.K. Lee, Y.L. Kwong, C.K. Lam, R. Liang, A novel mutation of Arg306 of factor V gene in Hong Kong Chinese, Blood 91 (4) (1998) 1135–1139. [165] R. Liang, C.K. Lee, M.S. Wat, Y.L. Kwong, C.K. Lam, H.W. Liu, Clinical significance of Arg306 mutations of factor V gene, Blood 92 (7) (1998) 2599–2600. [166] E. Norstrøm, E. Thorelli, B. Dahlba¨ck, Functional characterization of recombinant FV Hong Kong and FV Cambridge, Blood 100 (2) (2002) 524–530. [167] M. van der Neut Kolfschoten, R.J. Dirven, H.L. Vos, G. Tans, J. Rosing, R.M. Bertina, Factor Va is inactivated by activated protein C in the absence of cleavage sites at Arg-306, Arg-506, and Arg-679, J. Biol. Chem. 279 (8) (2004) 6567–6575. [168] A.D. Mumford, J.H. McVey, C.V. Morse, K. Gomez, M. Steen, E.A. Norstrøm, et al., Factor V I359T: a novel mutation associated with thrombosis and resistance to activated protein C, Br. J. Haematol. 123 (3) (2003) 496–501. [169] M. Steen, E.A. Norstrøm, A.L. Tholander, P.H. Bolton-Maggs, A. Mumford, J.H. McVey, et al., Functional characterization of factor V-Ile359Thr: a novel mutation associated with thrombosis, Blood 103 (9) (2004) 3381–3387. [170] B. Lunghi, L. Iacoviello, D. Gemmati, M.G. Dilasio, E. Castoldi, M. Pinotti, et al., Detection of new polymorphic markers in the factor V gene: association with factor V levels in plasma, Thromb. Haemost. 75 (1) (1996) 45–48.

154

SEGERS AND CASTOLDI

[171] T. Yamazaki, G.A. Nicolaes, K.W. Sorensen, B. Dahlba¨ck, Molecular basis of quantitative factor V deficiency associated with factor V R2 haplotype, Blood 100 (7) (2002) 2515–2521. [172] M. van der Neut Kolfschoten, R.J. Dirven, H.L. Vos, R.M. Bertina, The R2-haplotype associated Asp2194Gly mutation in the light chain of human factor V results in lower expression levels of FV, but has no influence on the glycosylation of Asn2181, Thromb. Haemost. 89 (3) (2003) 429–437. [173] J.W. Govers-Riemslag, E. Castoldi, G.A. Nicolaes, G. Tans, J. Rosing, Reduced factor V concentration and altered FV1/FV2 ratio do not fully explain R2-associated APCresistance, Thromb. Haemost. 88 (3) (2002) 444–449. [174] E. Castoldi, J. Rosing, D. Girelli, L. Hoekema, B. Lunghi, F. Mingozzi, et al., Mutations in the R2 FV gene affect the ratio between the two FV isoforms in plasma, Thromb. Haemost. 83 (3) (2000) 362–365. [175] N. Martinelli, D. Girelli, P. Ferraresi, O. Olivieri, B. Lunghi, F. Manzato, et al., Increased factor VIII coagulant activity levels in male carriers of the factor V R2 polymorphism, Blood Coagul. Fibrinolysis 18 (2) (2007) 125–129. [176] G. Castaman, E.M. Faioni, A. Tosetto, F. Bernardi, The factor V HR2 haplotype and the risk of venous thrombosis: a meta-analysis, Haematologica 88 (10) (2003) 1182–1189. [177] A. Tosetto, M. Simioni, D. Madeo, F. Rodeghiero, Intraindividual consistency of the activated protein C resistance phenotype, Br. J. Haematol. 126 (3) (2004) 405–409. [178] K.H. Orstavik, P. Magnus, H. Reisner, K. Berg, J.B. Graham, W. Nance, Factor VIII and factor IX in a twin population. Evidence for a major effect of ABO locus on factor VIII level, Am. J. Hum. Genet. 37 (1) (1985) 89–101. [179] J.M. Soria, L. Almasy, J.C. Souto, A. Buil, E. Martinez-Sanchez, J. Mateo, et al., A new locus on chromosome 18 that influences normal variation in activated protein C resistance phenotype and factor VIII activity and its relation to thrombosis susceptibility, Blood 101 (1) (2003) 163–167. [180] M. Berger, M. Mattheisen, B. Kulle, H. Schmidt, J. Oldenburg, H. Bickeboller, et al., High factor VIII levels in venous thromboembolism show linkage to imprinted loci on chromosomes 5 and 11, Blood 105 (2) (2005) 638–644. [181] M. Berger, H. Moscatelli, B. Kulle, B. Luxembourg, K. Blouin, M. Spannagl, et al., Association of ADAMDEC1 haplotype with high factor VIII levels in venous thromboembolism, Thromb. Haemost. 99 (5) (2008) 905–908. [182] J.C. Roelse, M.M. Koopman, H.R. Buller, J.W. ten Cate, B. Montaruli, J.A. van Mourik, et al., Absence of mutations at the activated protein C cleavage sites of factor VIII in 125 patients with venous thrombosis, Br. J. Haematol. 92 (3) (1996) 740–743. [183] M.I. Bokarewa, G. Falk, K. Bremme, M. Blomback, B. Wiman, Search for mutations in the genes for coagulation factors V and VIII with a possible predisposition to activated protein C resistance, Eur. J. Clin. Invest. 27 (4) (1997) 340–345. [184] W.C. Hooper, A. Dilley, H. Austin, N.K. Wenger, J. Benson, B.L. Evatt, et al., Absence of mutations at APC cleavage sites Arg306 in factor V and Arg336, Arg562 in factor VIII in African-Americans, Thromb. Haemost. 79 (1) (1998) 236. [185] J. Brummer, J. Groth, R. Flayeh, C. Wagener, R. Jung, Absence of mutations at the APC interacting sites of factor VIII in Caucasians, Thromb. Haemost. 87 (1) (2002) 170. [186] A. Tripodi, V. Chantarangkul, P.M. Mannucci, Hyperprothrombinemia may result in scquired activated protein C reistance, Blood 96 (9) (2000) 3295–3296. [187] G. Castaman, A. Tosetto, M. Simioni, M. Ruggeri, D. Madeo, F. Rodeghiero, Phenotypic APC resistance in carriers of the A20210 prothrombin mutation is associated with an increased risk of venous thrombosis, Thromb. Haemost. 86 (3) (2001) 804–808.

FACTOR V LEIDEN AND ACTIVATED PROTEIN C RESISTANCE

155

[188] J.M. Brugge, G. Tans, J. Rosing, E. Castoldi, Protein S levels modulate the activated protein C resistance phenotype induced by elevated prothrombin levels, Thromb. Haemost. 95 (2) (2006) 236–242. [189] E. Castoldi, P. Simioni, D. Tormene, M.C. Thomassen, L. Spiezia, S. Gavasso, et al., Differential effects of high prothrombin levels on thrombin generation depending on the cause of the hyperprothrombinemia, J. Thromb. Haemost. 5 (5) (2007) 971–979. [190] S.R. Poort, F.R. Rosendaal, P.H. Reitsma, R.M. Bertina, A common genetic variation in the 3’-untranslated region of the prothrombin gene is associated with elevated plasma prothrombin levels and an increase in venous thrombosis, Blood 88 (10) (1996) 3698–3703. [191] E. Castoldi, T.M. Hackeng, Regulation of coagulation by protein S, Curr. Opin. Hematol. 15 (5) (2008) 529–536. [192] A. Dahm, A. Van Hylckama Vlieg, B. Bendz, F. Rosendaal, R.M. Bertina, P.M. Sandset, Low levels of tissue factor pathway inhibitor (TFPI) increase the risk of venous thrombosis, Blood 101 (11) (2003) 4387–4392. [193] S.N. Tchaikovski, B. van Vlijmen, M. Thomassen, E. Castoldi, L.L. Peeters, G. Tans, et al., Changes in the coagulation system in humans and mice during pregnancy, J. Thromb. Haemost. 7 (2009) Abstracts from XXII ISTH Congress. [194] J. Rosing, Mechanisms of OC related thrombosis, Thromb. Res. 115 (Suppl. 1) (2005) 81–83. [195] B. Brenner, Haemostatic changes in pregnancy, Thromb. Res. 114 (5–6) (2004) 409–414. [196] A.M. Cumming, R.C. Tait, S. Fildes, A. Yoong, S. Keeney, C.R. Hay, Development of resistance to activated protein C during pregnancy, Br. J. Haematol. 90 (3) (1995) 725–727. [197] G. Sarig, Y. Michaeli, N. Lanir, B. Brenner, N. Haim, Mechanisms for acquired activated protein C resistance in cancer patients, J. Thromb. Haemost. 3 (3) (2005) 589–590. [198] H.A. van Vliet, R.M. Bertina, A.E. Dahm, F.R. Rosendaal, J. Rosing, P.M. Sandset, et al., Different effects of oral contraceptives containing different progestogens on protein S and tissue factor pathway inhibitor, J. Thromb. Haemost. 6 (2) (2008) 346–351. [199] H.F. Negaard, P.O. Iversen, B. Ostenstad, M.C. Mowinckel, P.M. Sandset, Increased acquired activated protein C resistance in unselected patients with hematological malignancies, J. Thromb. Haemost. 6 (9) (2008) 1482–1487. [200] M. Marchetti, E. Castoldi, H.M. Spronk, R. van Oerle, D. Balducci, T. Barbui, et al., Thrombin generation and activated protein C resistance in patients with essential thrombocythemia and polycythemia vera, Blood 112 (10) (2008) 4061–4068. [201] C. Taralunga, R. Gueguen, S. Visvikis, V. Regnault, C. Sass, G. Siest, et al., Phenotypic sensitivity to activated protein C in healthy families: importance of genetic components and environmental factors, Br. J. Haematol. 126 (3) (2004) 392–397. [202] T. Arsov, D. Miladinova, M. Spiroski, Factor V Leiden is associated with higher risk of deep venous thrombosis of large blood vessels, Croat. Med. J. 47 (3) (2006) 433–439. [203] T. Mueller, R. Marschon, B. Dieplinger, D. Haidinger, A. Gegenhuber, W. Poelz, et al., Factor V Leiden, prothrombin G20210A, and methylenetetrahydrofolate reductase C677T mutations are not associated with chronic limb ischemia: the Linz Peripheral Arterial Disease (LIPAD) study, J. Vasc. Surg. 41 (5) (2005) 808–815. [204] A. Gurgey, R. Rustemov, H. Parlak, G. Balta, Prevalence of factor V Leiden and methylenetetrahydrofolate reductase C677T mutations in Azerbaijan, Thromb. Haemost. 80 (3) (1998) 520–521. [205] N.V. Lipay, V.V. Dmitriev, M.B. Borisenok, Thrombotic complications during cancer treatment in children, Exp. Oncol. 29 (3) (2007) 231–235. [206] B. Boyanovsky, M. Russeva, V. Ganev, M. Penev, M. Baleva, Prevalence of factor V Leiden and prothrombin 20210 A variant in Bulgarian patients with pulmonary thromboembolism and deep venous thrombosis, Blood Coagul. Fibrinolysis 12 (8) (2001) 639–642.

156

SEGERS AND CASTOLDI

[207] V. Cikes, I. Abaza, V. Krzelj, I.M. Terzic, R. Tafra, A. Trlaja, et al., Prevalence of factor V Leiden and G6PD 1311 silent mutations in Dalmatian population, Arch. Med. Res. 35 (6) (2004) 546–548. [208] S.L. Xenophontos, M. Hadjivassiliou, N. Ayrton, A. Karagrigoriou, M. Pantzaris, A.N. Nicolaides, et al., Spectrum and prevalence of prothrombotic single nucleotide polymorphism profiles in the Greek Cypriot population, Int. Angiol. 21 (4) (2002) 322–329. [209] T.L. Benfield, M. Dahl, B.G. Nordestgaard, A. Tybjaerg-Hansen, Influence of the factor V Leiden mutation on infectious disease susceptibility and outcome: a population-based study, J. Infect. Dis. 192 (10) (2005) 1851–1857. [210] K. Kontula, A. Ylikorkala, H. Miettinen, A. Vuorio, R. Kauppinen-Makelin, L. Hamalainen, et al., Arg506Gln factor V mutation (factor V Leiden) in patients with ischaemic cerebrovascular disease and survivors of myocardial infarction, Thromb. Haemost. 73 (4) (1995) 558–560. [211] F. Bauduer, A. Zivelin, L. Ducout, E. Shpringer, U. Seligsohn, The prevalence of factor V G1691A but not of prothrombin G20210A and methylenetetrahydrofolate reductase C677T is remarkably low in French Basques, J. Thromb. Haemost. 2 (2) (2004) 361–362. [212] E. Cochery-Nouvellon, F. Vitry, P. Cornillet-Lefebvre, N. Hezard, L. Gillot, P. Nguyen, Interleukin-10 promoter polymorphism and venous thrombosis: a case-control study, Thromb. Haemost. 96 (1) (2006) 24–28. [213] E. Oger, K. Lacut, G. Le Gal, F. Couturaud, J.H. Abalain, B. Mercier, et al., Interrelation of hyperhomocysteinemia and inherited risk factors for venous thromboembolism. Results from the E.D.I.TH. study: a hospital-based case-control study, Thromb. Res. 120 (2) (2007) 207–214. [214] M. Aleksic, P. Jahn, J. Heckenkamp, K. Wielckens, J. Brunkwall, Comparison of the prevalence of APC-resistance in vascular patients and in a normal population cohort in Western Germany, Eur. J. Vasc. Endovasc. Surg. 30 (2) (2005) 160–163. [215] B. Hoppe, F. Tolou, T. Dorner, H. Kiesewetter, A. Salama, Gene polymorphisms implicated in influencing susceptibility to venous and arterial thromboembolism: frequency distribution in a healthy German population, Thromb. Haemost. 96 (4) (2006) 465–470. [216] A. Gialeraki, M. Politou, L. Rallidis, E. Merkouri, C. Markatos, D. Kremastinos, et al., Prevalence of prothrombotic polymorphisms in Greece, Genet. Test 12 (4) (2008) 541–547. [217] J.W. Blom, C.J. Doggen, S. Osanto, F.R. Rosendaal, Malignancies, prothrombotic mutations, and the risk of venous thrombosis, JAMA 293 (6) (2005) 715–722. [218] G. Mozsik, G. Rumi, A. Domotor, M. Figler, B. Gasztonyi, E. Papp, et al., Involvement of serum retinoids and Leiden mutation in patients with esophageal, gastric, liver, pancreatic, and colorectal cancers in Hungary, World J Gastroenterol. 11 (48) (2005) 7646–7650. [219] I. Olafsson, S. Hjaltadottir, P.T. Onundarson, R. Porarinsdottir, V. Haraldsdottir, Prevalence of factor V(Q506) and prothrombin 20210 A mutations in an apparently healthy Icelandic population and patients suffering from venous thrombosis, Thromb. Haemost. 79 (3) (1998) 685–686. [220] A.D. Pherwani, P.C. Winter, P.T. McNamee, C.C. Patterson, C.M. Hill, J.K. Connolly, et al., Is screening for factor V Leiden and prothrombin G20210A mutations in renal transplantation worthwhile? Results of a large single-center U.K. study, Transplantation 76 (3) (2003) 603–605. [221] C.H. Martini, C.J. Doggen, C. Cavallini, F.R. Rosendaal, P.M. Mannucci, No effect of polymorphisms in prothrombotic genes on the risk of myocardial infarction in young adults without cardiovascular risk factors, J. Thromb. Haemost. 3 (1) (2005) 177–179.

FACTOR V LEIDEN AND ACTIVATED PROTEIN C RESISTANCE

157

[222] G. Sottilotta, C. Mammi, G. Furlo, V. Oriana, C. Latella, V. Trapani Lombardo, High Incidence of Factor V Leiden and Prothrombin G20210A in Healthy Southern Italians, Clin. Appl. Thromb. Hemost. 15 (3) (2009) 356–359. [223] E. Berge, K.B. Haug, E.C. Sandset, K.K. Haugbro, M. Turkovic, P.M. Sandset, The factor V Leiden, prothrombin gene 20210GA, methylenetetrahydrofolate reductase 677CT and platelet glycoprotein IIIa 1565TC mutations in patients with acute ischemic stroke and atrial fibrillation, Stroke 38 (3) (2007) 1069–1071. [224] E. Nizankowska-Mogilnicka, L. Adamek, P. Grzanka, T.B. Domagala, M. Sanak, M. Krzanowski, et al., Genetic polymorphisms associated with acute pulmonary embolism and deep venous thrombosis, Eur. Respir. J. 21 (1) (2003) 25–30. [225] S. Barreirinho, A. Ferro, M. Santos, E. Costa, J. Pinto-Basto, A. Sousa, et al., Inherited and acquired risk factors and their combined effects in pediatric stroke, Pediatr. Neurol. 28 (2) (2003) 134–138. [226] V. Djordjevic, L.J. Rakicevic, D. Mikovic, M. Kovac, P. Miljic, D. Radojkovic, et al., Prevalence of factor V leiden, factor V cambridge, factor II G20210A and methylenetetrahydrofolate reductase C677T mutations in healthy and thrombophilic Serbian populations, Acta Haematol. 112 (4) (2004) 227–229. [227] J. Hudecek, M. Dobrotova, J. Hybenova, J. Ivankova, V. Melus, R. Pullmann, et al., Factor V Leiden and the Slovak population, Vnitr. Lek. 49 (11) (2003) 845–850. [228] J.M. Garcia-Gala, V. Alvarez, C.R. Pinto, I. Soto, M.F. Urgelles, M.J. Menendez, et al., Factor V Leiden (R506Q) and risk of venous thromboembolism: a case-control study based on the Spanish population, Clin. Genet. 52 (4) (1997) 206–210. [229] F. Frances, O. Portoles, F. Gabriel, D. Corella, J.V. Sorli, A. Sabater, et al., Factor V Leiden (G1691A) and prothrombin-G20210A alleles among patients with deep venous thrombosis and in the general population from Spain, Rev. Med. Chil. 134 (1) (2006) 13–20. [230] J. Holm, B. Zo¨ller, E. Berntorp, L. Erhardt, B. Dahlba¨ck, Prevalence of factor V gene mutation amongst myocardial infarction patients and healthy controls is higher in Sweden than in other countries, J. Intern. Med. 239 (3) (1996) 221–226. [231] J. Holm, B. Zo¨ller, P.J. Svensson, E. Berntorp, L. Erhardt, B. Dahlba¨ck, Myocardial infarction associated with homozygous resistance to activated protein C, Lancet 344 (8927) (1994) 952–953. [232] N. Akar, E. Akar, G. Dalgin, A. Sozuoz, K. Omurlu, S. Cin, Frequency of Factor V (1691 G –> A) mutation in Turkish population, Thromb. Haemost. 78 (6) (1997) 1527–1528. [233] S. Kabukcu, N. Keskin, A. Keskin, E. Atalay, The frequency of factor V Leiden and concomitance of factor V Leiden with prothrombin G20210A mutation and methylene tetrahydrofolate reductase C677T gene mutation in healthy population of Denizli, Aegean region of Turkey, Clin. Appl. Thromb. Hemost. 13 (2) (2007) 166–171. [234] N.J. Beauchamp, M.E. Daly, K.K. Hampton, P.C. Cooper, F.E. Preston, I.R. Peake, High prevalence of a mutation in the factor V gene within the U.K. population: relationship to activated protein C resistance and familial thrombosis, Br. J. Haematol. 88 (1) (1994) 219–222. [235] J.H. Griffin, J.A. Ferna´ndez, A.J. Gale, L.O. Mosnier, J. Thromb. Haemost. 5 (Suppl. 1) (2007) 73–80.

This page intentionally left blank

ADVANCES IN CLINICAL CHEMISTRY, VOL. 49

SELF-ASSEMBLED TETHERED BIMOLECULAR LIPID MEMBRANES Eva-Kathrin Sinner,*,†,1 Sandra Ritz,* Renate Naumann,*,‡ Stefan Schiller,§ and Wolfgang Knoll†,‡ *Max Planck Institute for Polymer Research, Mainz, Germany † Institute of Materials Research and Engineering, Singapore ‡ Austrian Institute of Technology, Vienna, Austria § Freiburg Institute for Advanced Studies, Freiburg, Germany

1. 2. 3. 4. 5. 6.

Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Assembly of tBLMs from Telechelics and Reconstitution of Proteins . . . . . . . . . . . . The Peptide-Tethered Lipid Bilayer Membrane (peptBLM) . . . . . . . . . . . . . . . . . . . . . . Protein-Tethered Bilayer Lipid Membrane (protBLM). . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

159 160 162 165 170 174 177

1. Abstract This chapter describes some of the strategies developed in our group for designing, constructing and structurally and functionally characterizing tethered bimolecular lipid membranes (tBLM). We introduce this platform as a novel model membrane system that complements the existing ones, for example, Langmuir monolayers, vesicular liposomal dispersions and bimolecular (‘‘black’’) lipid membranes. Moreover, it offers the additional advantage of allowing for studies of the influence of membrane structure and order on the function of integral proteins, for example, on how the composition and organization of lipids in a mixed membrane influence the ion translocation activity of integral channel proteins.

1

Corresponding author: Eva-Kathrin Sinner 159

0065-2423/09 $35.00 DOI: 10.1016/S0065-2423(09)49007-3

Copyright 2009, Elsevier Inc. All rights reserved.

160

SINNER ET AL.

The first strategy that we introduce concerns the preparation of tethered monolayers by the self-assembly of telechelics. Their molecular architecture with a headgroup, a spacer unit (the ‘‘tether’’) and the amphiphile that mimics the lipid molecule allows them to bind specifically to the solid support thus forming the proximal layer of the final architecture. After fusion of vesicles that could contain reconstituted proteins from a liposomal dispersion in contact to this monolayer the tethered bimolecular lipid membrane is obtained. This can then be characterized by a broad range of surface analytical techniques, including surface plasmon spectroscopies, the quartz crystal microbalance, fluorescence and IR spectroscopies, and electrochemical techniques, to mention a few. It is shown that this concept allows for the construction of tethered lipid bilayers with outstanding electrical properties including resistivities in excess of 10 M cm2. A modified strategy uses the assembly of peptides as spacers that couple covalently via their engineered sulfhydryl or lipoic acid groups at the N-terminus to the employed gold substrate, while their C-terminus is being activated afterward for the coupling of, for example, dimyristoylphosphatidylethanol amine (DMPE) lipid molecules via the NH2 moiety of their headgroups. It is demonstrated that these membranes are well suited for the in situ synthesis of membrane protein by a cell-free expression approach. The vectorial integration of an in vitro synthesized odorant receptor, OR5 from the rat, is demonstrated by means of antibodies that specifically bind to a tag at the N‐terminus of the receptor and is read out by surface plasmon fluorescence spectroscopy. A completely different strategy employs his-tagged membrane proteins in their solubilized form binding to a surface-attached Ni+–NTA monolayer generating a well-oriented protein layer the density of which can be easily controlled by online monitoring the binding (assembly) step by surface plasmon spectroscopy. Moreover, the attachment of the his-tag to either the C- or the N‐terminus allows for the complete control of the protein orientation. After the exchange of the detergent micelle by a lipid bilayer via a surface dialysis procedure an electrically very well isolating protein-tethered membrane is formed. We show that this ‘‘wiring’’ of the functional units allows for the (external) manipulation of the oxidation state of the redox–protein cytochrome c Oxidase by the control of the potential applied to the gold substrate which is used as the working electrode in an electrochemical attachment.

2. Introduction The classical portfolio of model membrane systems used for biophysical and interfacial studies of lipid (bi)layers and lipid/protein composites includes Langmuir monolayers assembled at the water/air interface, (uni- and

SELF-ASSEMBLED TETHERED BIMOLECULAR LIPID MEMBRANES

161

multilamellar) vesicles in a bulk (liposomal) dispersion, the bimolecular lipid membrane (BLMs), and various types of solid supported membranes [1]. All these have their advantages but also serious drawbacks. For example, in Langmuir monolayers spread and compressed at the water surface of a Langmuir trough, the packing density and the corresponding phase behavior of lipids in these ‘‘half-membranes’’ can be manipulated and very well characterized by a variety of experimental techniques [2]. What is problematic is the incorporation of transmembrane proteins, because the other half of the membrane is missing. Vesicles in either the unilamellar (i.e., one lipid bilayer thick) or multilamellar form are the classical model membrane system, perfectly suited and often used for structural studies [1]. What is nearly impossible to do is the simultaneous investigation of functional aspects of the bare membranes and those found after incorporation of functional units like proteins. Hence, also no correlation between structure and function can be deduced in either systems, neither the Langmuir monolayers nor the vesicular system. On the other hand, certain functions of lipid bilayer/protein composites, for example, the translocation of ions across the hydrophobic barrier of a membrane can be studied nicely with the BLMs, the ‘‘black’’ lipid membranes [3]. However, owing to their fragile nature nearly no structural studies on incorporated protein moieties are possible and were reported. The various ‘‘solid supported membranes’’ promise to bridge this gap by offering the possibility of very detailed biophysical studies of membrane structure, order, and dynamics and allowing for experimental strategies to elucidate the much-needed correlation of these parameters with the function of incorporated (or surface-associated) proteins or protein aggregates [4]. Introduced in the 1980s by the McConnell group [5], supported membranes rapidly demonstrated their enormous potential for the application of a broad spectrum of experimental techniques. By virtue of the fact that the (fluid) lipid bilayer in this platform is weakly coupled to a robust solid support a largely enhanced stability allows for the use of a variety of surface-analytical tools for structural characterizations, ranging from X-ray and neutron reflectometry, optical techniques, including ellipsometry, surface plasmon and waveguide spectroscopies, vibrational spectroscopies, to fluorescence-based techniques, scanning probe methods, and many more. Additionally, functional aspects, for example, the behavior of membrane-integral units diffusing in the (liquid-crystalline) two-dimensional matrix of the lipid bilayer, the binding of ligands to membrane-integral receptors, or the translocation of ions across the hydrophobic barrier of the bilayer could be studied in parallel and interpreted on the basis of the simultaneously monitored structural data. However, the fact that in these systems the bilayers are only physisorbed to the substrate, that is, are only floating on top of a very thin water layer eventually can lead to delamination and the destruction of the membrane architecture. Hence, attempts were made to further stabilize the lipid matrix

162

SINNER ET AL.

and the incorporated proteins by chemically tethering either one of them to the solid support [6]. In some cases, this was done via flexible spacers that coupled the membranes by ‘‘anchor lipids’’ in a stable and robust way to the substrates while, at the same time, decoupled the membranes sufficiently from the solid surface thus preventing the incorporated proteins from being denatured by their strong interaction with the polar groups of the hydrophilic support. An alternative approach was recently reported by which the membranes were coupled (wired) to the support via tethered proteins [7]. The process starts with the self-assembly of the tether binding with the substrate-specific functional groups to the solid support. The other end can then be used to couple (solubilized membrane-) proteins, for example, via the his-tag strategy, followed by the build-up of the surrounding membrane by a surface-dialysis protocol [8]. Each of these strategies for the build-up of stable tethered bimolecular lipid membranes (tBLMs) are based on the mere self-assembly of specifically custom-tailored (tether) molecules that bind and, thus, couple—structurally and functionally—the membranes and/or incorporated proteins to a solid substrate. In the following, we will briefly summarize some key features of these concepts and give a few examples of the obtained membrane architectures and their functional properties.

3. Assembly of tBLMs from Telechelics and Reconstitution of Proteins The first concept of tethering a lipid membrane to a solid support is based on the use of anchor lipids. These telechelics are composed of three distinct molecular parts: (i) an amphiphile that becomes part of the proximal monolayer of the final bilayer architecture, (ii) a spacer unit that decouples the bilayer from the substrate thus guaranteeing sufficient space for integral proteins, and (iii) a substrate-specific head group, for example, based on thiol, disulfides, lipoic acids, or alike for Au supports [9]. The whole process of the functional membrane fabrication is schematically summarized in the carton given in Fig. 1. It starts with the assembly of telechelic lipid derivatives that are designed to covalently bind to and self-organize at the substrate thus constituting the proximal monolayer of the final membrane bilayer. The fusion of vesicles results in the formation of the distal monolayer completing the lipid bilayer followed by the eventual incorporation of a variety of functional structures such as peptides or proteins. The inset shows a kinetic surface plasmon optical recording taken during the assembly of the lipid molecule 2,3-di-O-phytanyl-sn-glycerin-1-tetraethylenglycol-lipoic acid ester (DPTL). The change in reflected intensity monitored at a fixed angle of

Thickness d (nm)

SELF-ASSEMBLED TETHERED BIMOLECULAR LIPID MEMBRANES

163

5.0 Protein incorporation

2.5

0 0

300 Time t (min)

600

Selfassembly

Protein− functionalized membrane Bilayer Monolayer

Vesicle fusion

Au substrate FIG. 1. The assembly process of a protein-functionalized tethered membrane, stating with a flat Au substrate, the assembly of telechelics from solution, the formation of the bilayer by fusion of vesicles to the tethered monolayer and the incorporation of proteins by reconstitution. The inset shows the SPR kinetic data of this adsorption process of the tethered lipid molecules from solution to the Au substrate.

observation as a function of time is converted to a thickness increase of the interfacial layer with a final value of ca. 3.5 nm for the tethered monolayer. The whole assembly process takes several hours to be completed, well in line with observations with other self-assembling thiols on gold. Other techniques applied to the sample for the further characterization of the structural properties of the resulting monolayer are then also used to characterize the final double layer architecture after the distal lipid monolayer was added by fusing lipid vesicles from a liposomal dispersion. A very sensitive way to analyze the functional characteristics of the bare lipid membrane is given by electrochemical impedance spectroscopy (EIS). One of the first examples that we found to satisfy the needs of a membrane that imposes a real barrier against the mere passive permeation of ions across its hydrophobic barrier is given in Fig. 2. Shown are the impedance data in the form of a Bode plot (A) and as frequency normalized admittance plot (B). The analysis of the data based on the equivalent circuit shown in the inset of the left panel results in the following values: the specific membrane capacitance is found to be Cm ¼ 0.52 F cm 2, in excellent agreement with values that were reported for black lipid membranes [10]. The real breakthrough

SINNER ET AL.

164 A

108

100

107

80

106

40

104 Rm

Rex

103 102

Cin

20

Cm

101 10–3 10–2

10–1

100

q (des)

Z/W

60 105

0 101

102

103

104

105

106

Frequency (Hz) Rbilayer = 14.9 MΩ cm2 B

0

Y ⬙/w

2.5⫻10–7

5.0⫻10–7

7.5⫻10–7 0

2.5⫻10–7

5.0⫻10–7

7.5⫻10–7

Y⬘/w Cbilayer = 0.52 mF cm–2 FIG. 2. Electrical properties of tBLM quantified by electrochemical impedance spectroscopy (EIS).

toward using tethered lipid bilayer membranes for ion translocation studies was the specific membrane resistance that was found for this system to be Rm ¼ 14.9 M cm2. To appreciate what this value means let’s assume a membrane area of, for example, 100 100 m2. With the given resistance and upon the application of a potential of U ¼ 100 mV, the current across

SELF-ASSEMBLED TETHERED BIMOLECULAR LIPID MEMBRANES

165

the membrane corresponds to I ¼ 1 pA, which has to be seen in comparison to the currents that typically flow cross a single channel in the open state, that is, currents in the range of 10 pA. It has been shown, indeed, that the electrical barrier of the unmodified tBLMs is good enough to monitor the current fluctuations induced by the opening and closing of single individual channels [11]. We should also mention that these excellent barrier properties are the result of an optimized molecular architecture of the self-assembling lipid molecule with typically a thiol or lipoic acid anchor group, a short ethylenoxid unit as the spacer, and a (ether bond coupled) lipid analogue with two phytanoyl chains that improve the fluidic (hence, sealing) character of the hydrophobic core, and the use of an ultra-flat Au substrate prepared by the so-called template stripping [12]. For an electrically tight bilayer it seems to be important to use a substrate with a residual roughness of not more than 10% of the typical thickness of the multilayer architecture assembled on top in order to allow for a rather undisturbed self-organizing of the lipid molecules into the fluid tethered lipid bilayer membrane.

4. The Peptide-Tethered Lipid Bilayer Membrane (peptBLM) peptBLMs are supported membranes tethered to the support by a peptide spacer. The peptides are attached to typically a gold substrate by S–Au interaction. Self‐assembled monolayers (SAMs) are prepared from synthetic or native thiopeptides or thiolipopeptides. The conformation of the secondary structure is determined by the amino acid sequence (a-helical/b-sheet/ random coil). Lipid bilayers having peptide spacers were first prepared on polymer beads. A spacer of a certain length (e.g., penta-alanin) was a prerequisite for the successful insertion of the proton pump bacteriorhodopsin in a functionally active form [13]. The peptide-tethered lipid concept was then extended to planar gold surfaces thus providing the possibility to apply electrochemical and optical surface-analytical methods [6]. In these studies, slightly more hydrophilic peptides were used as compared to the penta-alanine, functionalized with terminal sulfur groups such as cysteine or lipoic acid designed for self-assembly on gold (Fig. 3). The terminal carboxyl group of the peptide was used to covalently attach the amino group of the phospholipid dimyristoylphosphatidylethanol amine (DMPE), using the active ester approach, thus forming a peptide-tethered lipid monolayer. Subsequently, a lipid bilayer was formed by fusion of liposomes with and without the reconstituted protein of interest. In this way, the incorporation of the Hþ-ATPases from chloroplasts and

SINNER ET AL.

166 Thiopeptide P5:

Lip-Ala-Ala-Ala-Ala-Ala-COOH

Thiopeptide P7:

HS-(CH2)2-Ala-Ser-Ser-Ala-Ala-Ser-Ala-COOH

Thiopeptide P19:

HS-Cys-Arg-Ala-Arg-Lys-Gln-Ala-Ala-Ser-Ile-Lys-Val-Ala-Val-Ser-Ala-Asp-Arg-COOH

Thiolipopeptide LP12: HS-Cys-Ala-Ser-Ala-Ala-Ser-Ser-Ala-Pro-Ser-Ser-Lys(Myr)-Myr Lip−lipoic acid Myr−myristoic acid, C14

FIG. 3. Amino acid sequences of thio- and thiolipopeptides used for coupling membranes to solid substrates.

Escherichia coli was verified by EIS, cyclic voltammetry (CV), and surface plasmon resonance (SPR) spectroscopy [6, 14, 15]. Other proteins such as the dimer species of the nicotinic acetylcholine receptor (nAChR), isolated from the electric organ of Torpedo californica and cytochrome c oxidase (CcO) from bovine heart were also incorporated. In the case of the nAChR, the insertion was followed by combined SPR and SPFS measurements and the receptor functionality was analyzed by the binding of the snake toxin a-bungarotoxin [16]. Electrochemical measurements on the CcO, inserted into a preformed peptide-tethered bilayer (P7) by detergent dilution below the critical micelle concentration showed active proton transport using impedance spectroscopy [17]. PeptBLMs were thus shown to be well suited for the incorporation of membrane proteins in a functionally active form with the advantage of providing a very rigid and biocompatible spacer moiety. However, resistance and capacitance of the pure peptBLM were only in the range of 0.8 M cm2 and 15 Fcm 2 indicating an incomplete coverage ( 70%) of the substrate by the bilayer. Hence, these systems are not particularly suited for electrochemical measurements. They are, however, very well designed to preserve the integrity and functionality of membrane proteins. Hence they were used as a model system for the in situ cell-free expression of proteins. For this purpose and the ease of preparation, the membrane assembly process was transferred from organic solvents into water using a water-soluble natural peptide P19. P19 is part of the a-subunit of laminin, which contains various charged amino acids that increase the hydrophilicity of the tethering layer [18]. Laminin is a complex glycoprotein of the extracellular matrix, consisting of three different polypeptide chains, a, b, and g. Through the interaction with cellular receptors (e.g., integrins), they critically contribute to cell attachment, differentiation, cell shape, and movement. Whereas the SAM formation of the shorter peptides (P7/P11) could only be achieved in organic

SELF-ASSEMBLED TETHERED BIMOLECULAR LIPID MEMBRANES

167

solvents (trifluoroacetic acid and dimethylformamide) and needed 96 h to be completed, the water-soluble laminin peptide P19 assembles more quickly (1–3 h) as confirmed by SPR, EIS, and ATR-IR measurements [18–20]. Two principle strategies were used for the stepwise assembly of the asymmetric lipid bilayer to the water-soluble P19 peptide tether. Firstly, the amino moieties of DMPE (1 mg ml1 solubilized in PBS with 0.003 % (w/v) Triton X-100) were covalently linked in a flow through process to the NHS/EDC ester-activated carboxyl groups of the peptide layer and fused with unilamellar vesicles containing reconstituted proteins (Fig. 4A and B) [18, 21]. Secondly, DMPE (0.5 mg ml1 in chloroform) was spread at the air/water interface and bound by Langmuir–Blottget transfer ( ¼ 30 mN m1) to the activated P19 layer. The distal lipid monolayer was obtained by Langmuir–Schaefer transfer [20]. Owing to the transfer of the membrane assembly from an organic solvent into water, the functional incorporation of the cell adhesion receptor integrin avb3 (cancer metastasis/angiogenesis) and a1b1 (wound healing) into the P19tethered bilayer could be monitored online by SPR spectroscopy and surface plasmon fluorescence spectroscopy (SPFS) measurements in a flow cell. The functionality of the integrin receptors were proven by the binding of the ECM proteins vitronectin and collagen type IV. Surface regeneration with EDTA, chelating all divalent cations which are mandatory for the integrin– ligand binding, allowed for repeated integrin–ligand binding over a time period of 3 days [18]. Since the stepwise assembly allows for an easy exchange of the distal layer, the effect of several lipid mixtures (e.g., POPC/cholesterol/sphingomyelin) were analyzed with respect to the lipid adsorption of the amyloid b-peptide (Ab40). Amyloid fibrils of Ab are a major component of the extracellular plaques in Alzheimer’s disease and pathological interaction between cell lipid raft domains containing a mixture of cholesterol/sphingomyelin and Ab were described [22–24]. The binding study of Ab to peptide-tethered bilayers with different lipid compositions supported this hypothesis, because Ab bound specifically to membranes containing sphingomyelin. Furthermore, the adsorption was amplified by the addition of cholesterol [20]. Combining cell-free expression of proteins with the peptide-tethered lipid bilayer provides the basis for membrane-based screening platforms. The basic idea is to bypass the difficult expression, purification, and reconstitution procedures inherent when dealing with complex membrane proteins like, for example, G protein-coupled receptors (GPCRs). GPCRs are key targets of pharmaceutical drug development, because of their involvement in main cellular signaling pathways [25]. A common problem is the detergent-based purification and refolding of aggregated membrane proteins into their functional native conformation.

SINNER ET AL.

168 A 14

1666, amide I

Absorption (10−3 a.u.)

12 1562, amide II

10 8 6

7.5 min 15 min 22.5 min 30 min 37.5 min 45 min 1430, 1400, C=O C=O

4 2 0 −2 −4 1700

B

1600 1500 Wavenumber (cm−1)

1400

1300

0.60 Rinse PBS

Reflectifity (% at 30⬚)

0.55 NHS/EDC

0.50

Rinse PBS

0.45 0.40

PC

DMPE

Water rinse

0.35 0.30

P19

0.25 0.20 0

60

120

180 240 Time (min)

300

360

FIG. 4. Kinetic assembly of the P19 peptide-tethered membrane. (A) Adsorption kinetic of P19 on rough gold measured by surface-enhanced infrared adsorption reflection spectroscopy (SEIRAS). (B) Membrane assembly followed by surface plasmon resonance spectroscopy (SPR).

Cell-free expression systems are based on gene expression in the presence of cell extracts of various species like bacteria (E. coli), insects (Spodoptera frugiperda), plants (wheat germ) or eukaryotes (rabbit reticulocytes). Thus, the relevant gene, provided as cDNA, is mixed with the cell lysate and inserted in vitro by a co- or posttranslational process into the peptide-tethered membrane mimicking the biological membrane (Fig. 5A). The vectorial

A

mRNA Ribosome

In vitro expression cDNA

RNA polymerase

mRNA

Plasmid

tBLM

C 5⫻106

Cy5 Antimouse-Cy5 (goat)

Fluorescence (cps)

B

Anti-VSV (mouse)

I

II

III

IV

V

VI

4⫻106 3⫻106

N-terminal VSV-tag

2⫻106 1⫻106

Reference: no plasmid

VII

C-terminal VSV-tag

0 COO−

0

10

20

30

40

50

Time (min) FIG. 5. In vitro expression of membrane proteins. (A) The protein of interest is cloned into an appropriate plasmid-vector. A mixture of cDNA and cell lysate leads to the expression and incorporation of the membrane protein into the tBLM. (B) Schematic of the immunofluorescent analysis for the vectorial insertion of the odorant receptor OR5. The monoclonal antibody binds to the amino-terminal VSV tag and the secondary Cy5‐labeled antibody provides the specific fluorescence signal. (C) SPFS analysis for the vectorial insertion of OR5.

170

SINNER ET AL.

insertion of the odorant receptor OR5, a member of the seven-transmembrane GPCR family, into a peptide-tethered membrane composed of P19/DMPE and soybean phosphatidylcholine (PC) has been shown by Robelek et al. [21]. Immunolabeling with antibodies against amino and carboxy terminal tags, validated that the amino terminus of OR5 was outside of the membrane (Fig. 5B and C). All together, peptide-tethered membranes provide a biomimetic platform which is appropriate for the functional insertion of membrane proteins of different species and complexity. The peptide spacer forms a sufficiently rigid structure to allow for an aqueous layer of 1–3 nm in thickness compared to 0.12 nm in the case of poly-(oxyethylene) spacer [26] and up to 10 nm in the case of dextrane [27].

5. Protein-Tethered Bilayer Lipid Membrane (protBLM) A limited integration capacity for membrane proteins, particularly those composed of large subunits was demonstrated in the case of tBLMs from telechelics. Even small peptides such as melittin and gramicidin could only be incorporated in tBLMs made from mixed monolayers of DPTL with complementary dilution molecules [28]. In an attempt to create protein-rich biomimetic membrane systems, we began using the proteins themselves as coupling units tethered to the electrode surfaces through a histidine (his)-tag engineered onto the enzyme (Fig. 6). A lipid bilayer was then reconstituted in situ around the bound proteins by surface-dialysis, forming a protein-tethered bilayer lipid membrane (protBLM) [7]. The electrical properties of these systems showed resistances high enough (>1 M cm2) to compare favourably with BLMs and tBLMs mentioned above, with capacitances ( 7 F cm 2) determined by the proteins rather than the lipids and, hence, not comparable with pure lipid bilayers. They are, however, well suited for electrochemical investigations. A significant advantage of these systems is the possibility to immobilize the proteins as a function of packing density [8] as well as in a strict orientation. In the case of cytochrome c oxidase (CcO), for example, the enzyme could be immobilized with the cyt c binding site directed toward the electrode or pointing away from it (Fig. 7). Moreover, this strategy made it possible to apply a combination of spectroscopy and electrochemistry techniques in order to investigate electron and proton transfer processes through multiredox site proteins such as the CcO. Surface-enhanced IR absorption spectroscopy (SEIRAS) [29] and surfaceenhanced resonance Raman spectroscopy (SERRS) [30] could thus be applied to the protBLM under a defined electric field. Electron transfer (ET) to the CcO could be initiated, for example, by the interaction of reduced

SELF-ASSEMBLED TETHERED BIMOLECULAR LIPID MEMBRANES

171

O O

O

O

NH

NH

N N

O

N

Cu2+ O

O

N

O

O

O

N

OH

OH

N

N O

S

O S

FIG. 6. The concept of a protein-tBLM on a solid support.

A

B Cytochrome c H+

H+-pumping direction

e−

H+

e− H+

Au His-tag at subunit II

Au His-tag at subunit I

Electrode-activated (direct electron transfer)

(reduced) Cyt c activated

H+

FIG. 7. Oriented immobilization of CcO with the cyt c binding domain facing to the substrate (A) or being exposed to the free aqueous phase (B).

172

SINNER ET AL.

cytochrome c, if the cyt c binding site was pointing to the outside of the membrane. In this case, the enzyme was bound to the surface via the his-tag attached to subunit I (SU I, cf. Fig. 7B) and proton pumping activity was detected by EIS and CV. Catalytic currents were measured compatible with turnover rates determined under physiological conditions. In the reverse orientation, the enzyme was immobilized via the his-tag attached to SU II. In this case, the cyt c binding site is directed toward the electrode surface and ET to the enzyme could be initiated directly from the electrode by electronic wiring (cf. Fig. 8A) [31]. Direct ET was demonstrated by CV. Under anaerobic conditions, a single reduction peak was shown at 274 7 mV and a corresponding oxidation peak at 209 6 mV (Fig. 8A) at low scan rates (<1 V s 1). At scan rates above 1 V s 1 the peaks become more asymmetric; a second maximum gradually appears as the scan rate exceeds 20 V s 1. These complex voltammograms were deconvoluted into four Gaussian components with the area of each peak corresponding to one electron transferred in each step. ET rates could thus be calculated to every one of the four redox centers of the CcO, that is, CuA, heme a, heme a3, and CuB. The heterogeneous rate constants to the first electron acceptor CuA was shown to be relatively high (k0 > 4000 s 1). Evidence for the integrity of the enzyme undergoing electrochemically induced ET was derived from Soret band excited SERRS spectra taken as a function of the applied potential, also under strictly anaerobic conditions (Fig. 8B). Utilizing an excitation wavelength of ¼ 413nm of the Krypton ion laser, the selective resonant enhancement of the vibrational modes of the heme sites was employed, the absorption maxima of which are located at ¼ 410 nm. Redox changes of the heme centers could be observed in the SERR spectra at exactly the same potential at which the peaks were seen in the CV scans. In order to investigate the enzyme undergoing catalytic turnover, aerobic conditions were used. The peak at 200 mV was shown to be significantly amplified [31] compared to the peak under anaerobic conditions (Fig. 8A). This is a clear indication of the catalytic turnover of the enzyme. Electrons transferred from the electrode to the redox centers of CcO are irreversibly transferred to oxygen, leading to a continuous ET. As a consequence of catalytic turnover of the CcO in the orientation with the his-tag attached to SU II, protons are pumped from the bulk solution into the interstitial space between the electrode and the protBLM. Due to its small volume, this interstitial space is thereby highly acidified. These protons are electrochemically reduced to H2 giving rise to a second peak at 400 mV. These investigations clearly demonstrate the ‘‘wired’’ protBLM to be a valuable tool to gain information about the kinetics and mechanisms of proton and ET processes in multiredox site proteins.

SELF-ASSEMBLED TETHERED BIMOLECULAR LIPID MEMBRANES

A

173

0.9 0.6

j (mA cm−2)

0.3 0.0 −0.3 −0.6 −0.9 −1.2 −0.6

−0.4

−0.2

B a

0.0 0.2 E/V vs. NHE

0.6

1585 1646 1671

1370

1226 1358 1247

0.4

1500

−150 mV −175 mV −200 mV

Intensity (a.u.)

−225 mV −250 mV

−275 mV 1471

−300 mV −325 mV

1517 −350 mV 1610 1663 1200

1300

1400 1500 Δν (cm−1)

1600

1700

FIG. 8. Direct electron transfer (ET) as seen in the CV scans (A), and in the SERRS spectra taken as a function of the applied potential (B).

174

SINNER ET AL.

6. Conclusions We described some of our efforts in designing, synthesizing, assembling, and structurally and functionally characterizing tethered lipid bilayer membranes (tBLMs) as a novel platform for biophysical studies of and with artificial membranes. Chemical coupling schemes, for example, based on thiol groups for gold (Au) substrates, allowed for the covalent and, hence, chemically and mechanically robust attachment of telechelics as anchor lipids to the solid support, stabilizing the proximal layer of a tethered membrane on the transducer surface. By assembling the distal lipid layer, for example, by fusion of vesicles from a liposomal dispersion the final tethered lipid bilayer was obtained. A whole battery of surface-analytical methods, for example, surface plasmon optics, the quartz crystal microbalance, fluorescence- and IR spectroscopies, and electrochemical techniques were used to characterize these complex supramolecular interfacial architectures with respect to their assembly, structure, and function. By this we demonstrated, in particular, that these bilayers show the fluid character of a liquid-crystalline membrane with the capacity typical for such ultrathin layers, that is, in the range of 0.5 F cm 2, and a high specific electrical resistance of better than 10 M cm2. Although this approach resulted in very stable and electrically insulating membranes the incorporation of proteins, however, followed a rather stochastic mechanism: Unless any specific asymmetry, for example, pronounced shape anisotropy with a large external part of the protein sticking out of the bilayer induced a preferred insertion, the orientation was not controlled at all. Moreover, the packing density of the reconstituted proteins in the lipid bilayer could barely be manipulated either. Hence, any correlation between their functional performance and their orientation and number density were not possible. This was an obvious disadvantage for the development of biosensors, employing, for example, membrane-integral receptor proteins. Additionally, in some other cases it might be desirable to have the lipids completely decoupled from the solid substrate and stabilize the membrane via the incorporated proteins that are attached to the support. Examples are redox proteins that one might want to connect electronically to the base electrode because they need to be ‘‘wired’’ to the support in order to allow for efficient heterogeneous ET between the external circuit and the redox center of the protein. From the membrane proteins that we reconstituted in this way we described results obtained with the redox–protein cytochrome c oxidase. Here, we also used a genetically modified mutant with the his-tag at either the C- or the N-terminus for the oriented attachment of the protein via the NTA/ Ni2þ approach. With this strategy, we not only could control the density of the immobilized functional units, but we also introduced a completely new

SELF-ASSEMBLED TETHERED BIMOLECULAR LIPID MEMBRANES

175

and alternative concept for the stabilization of lipid bilayers, that is, the protein-tethered membrane. The last part of this report referred to an alternate approach for the assembling of tBLMs, this time based on a sequential adsorption and chemical reaction of an appropriate peptide sequence used as the final tether structure that coupled the anchor lipids to the substrate. With its terminal cystein group at one end it could covalently bind to the Au substrates used for surface plasmon optical and simultaneous electrochemical characterization, whereas the activation of the C-terminal carboxy group allowed for the covalent binding of the ethanolamine head group of a corresponding lipid derivative. Typically, also here a fusion process was used to assemble the distal lipid layer. For this peptide-tethered membrane system we introduced a totally novel approach for the functional incorporation of membrane proteins, that is, by their cell-free in vitro expression and in situ reconstitution in the presence of tBLMs. We focused on the rabbit reticulocyte expression system for the synthesis of the olfactory receptor species OR5 from Rattus norvegicus. By the combination of the corresponding coding DNA with the protein synthesis machinery of a cell extract (in vitro transcription and translation) we observed spontaneous and vectorial insertion of the olfactory receptor protein into the tethered lipid membrane. The various concepts presented here for the build-up of tethered lipid bilayers by relatively simple self-assembly strategies result in membranes that differ significantly in their property profiles and, hence, offer different advantages for the study of basic properties of model membranes in general and for the quantitative characterization of functional units like ionophores or proteins incorporated into these tethered architectures. Common to all the presented membranes is their stability and robustness which is guaranteed by the covalent attachment of the lipid bilayer to the solid support via the various spacer concepts. This way, all the systems discussed are very well suited for general biophysical studies; however, additionally, they are far more promising for practical applications as biosensors, for example, for the development of membrane chips. This is particularly interesting because more than 50% of current targets in drug discovery efforts address membrane-integral receptors which alternatively can be characterized only by very time-intensive patch-clamp techniques. Hence, tethered lipid bilayers bear an enormous potential for the next generation of advanced biosensor devices. By far the best electrical barrier properties are found for membranes prepared by the assembly of telechelics onto flat substrates. The latter can be prepared in large areas (several cm2) by template stripping and, hence, could be produced also in large quantities, a fact which is important for technical applications. The obtained conductivities result in background currents in the range of a few picoamperes only and, hence, can even exceed

176

SINNER ET AL.

the values of the best model system known for its isolation properties, that is, the classical BLMs. As mentioned before it is thus possible to monitor the minute current fluctuations through the membrane that occur upon the statistical opening and closing of single channels structures (synthetic pores or natural proteins) incorporated into the bilayer [11]. Somewhat surprising was the finding that despite the fact that all lipids of the proximal monolayer in these experiments were grafted to the solid support the translocation of ions mediated by the carrier valinomycin was easily possible and led to a conductivity increase of several orders of magnitude [32] with the discrimination between Kþ and Naþ that showed the expected ration of the conductance of 104. This was remarkable as it is well known that the molecular mechanism of the ion transport mediated by valinomycin involves the complexation of the ion by the carrier on one side of the membrane, the translocation of the charged complex across the hydrophobic barrier to the other side, and the dissociation and release of the ion into the opposite aqueous phase. This requires a considerable fluidity—at least locally—to allow for the diffusion of the complex across the bilayer. However, this fluidity also explains why these membranes are so well isolating: a thin liquid layer by definition has no holes (that could act as defects for the unspecific transfer of ions). If one is interested in the incorporation of (large) proteins or even protein complexes, the tBLM architecture is not ideal because the lipids in the proximal layer attached to the solid support are rather densely packed and, hence, do not allow for the easy incorporation of proteins. This limitation could be overcome by the reverse mode of action in the assembly procedure: by immobilizing the (detergent-solubilized) membrane proteins first, followed by the fusion of the surrounding bilayer in a surface-dialysis step that then replaces all the detergent molecules surrounding the protein any desired protein density could be achieved. This control of the final membrane architecture is a clear advantage over the mere statistical incorporation of proteins in the tethered bilayer architecture with no control of their number density nor their orientation. As described above, this full control of the orientation of the tethered proteins allows for a much better design of meaningful experiments, for example, for the quantitative determination of electron and proton exchange reaction between the electrode and redox–protein complexes incorporated into the membrane. Somewhat surprising was the finding that for these preparations the surface-dialysis step, indeed, leads to membranes that are almost as isolating as the tBLMs made from telechelics. This, however, helps the design and the interpretation of electrochemical experiments as it limits the contributions from charge translocation across defects that otherwise would interfere with the specific current contributions from the redox reactions.

SELF-ASSEMBLED TETHERED BIMOLECULAR LIPID MEMBRANES

177

On the other hand, this approach is viable only for studies of membrane properties or protein functions in which the lateral mobility of the protein is irrelevant (since all the proteins are immobilized—at least in the studies presented here). One could imagine, however, tethering the membranes to the substrate via one type of proteins that act as anchors just like the ankyrins couple the cytoskeleton to the plasma membrane. The concept for the build-up of tethered membranes by peptide spacers described in the last section of this report clearly led to only partial surface coverages of the (electrode) substrate resulting in rather high leakage currents. However, these defects obviously provided the lateral space needed for the efficient incorporation of functional units by either classical reconstitution or—as a complete paradigm chance in membrane research—by the cell-free expression of membrane-integral proteins. We employed the in vitro synthesis of, for example, GPCRs using the protein synthesis machinery of a cell extract in order to accomplish the direct incorporation of the nascent protein into a membrane mimicking structure. So far, this concept was demonstrated not only for the olfactory receptor from rat presented above but was also successfully applied to the synthesis and incorporation of the nicotinic aetylcholine receptor from mouse, claudine proteins from humans, the ubichinol cytochrome c oxidase from bacteria, bovine rhodopsin, and recently, the light harvesting complex LHCIIb from spinach. These results again demonstrate the usefulness and the enormous potential of the tethered membrane platform for basic studies in membrane biophysics but also for practical applications in the development of biosensors based on membrane-related process. Examples are the search for novel antimicrobial peptides, the screening of drugs interacting with membrane receptors, or the development of a smell sensor. ACKNOWLEDGMENTS We are grateful to a number of colleagues for stimulating discussions, in particular, to Marcel G. Friedrich, Frank Giess, Vincent Kirste, Ingo Ko¨per, R. Robelek, and B. Wiltschi. Partial support for this work came from the EU through the FuSyMem research project under the Sixth Research Framework program (FP6-2005-NEST-PATH).

REFERENCES [1] Y.H.M. Chan, S.G. Boxer, Model membrane systems and their applications, Curr. Opin. Chem. Biol. 11 (6) (2007) 581–587. [2] V. Erokhin, Langmuir–Blodgett films of biological molecules 1 (2002) 523–557. [3] H.T. Tien, A.L. Ottova, The lipid bilayer concept and its experimental realization: from soap bubbles, kitchen sink, to bilayer lipid membranes, J. Memb. Sci. 189 (1) (2001) 83–117. [4] In Focus: Structure and Properties of Soft Organic-Aqueous Interfaces Biointerphases 3 (3) (2001) FC1–93.

178

SINNER ET AL.

[5] L.K. Tamm, H.M. McConnell, Supported phospholipid-bilayers, Biophys. J. 47 (1) (1985) 105–113. [6] R. Naumann, A. Jonczyk, R. Kopp, J. Vanesch, H. Ringsdorf, W. Knoll, et al., Incorporation of membrane-proteins in solid-supported lipid layers, Angew. Chem. Int. Ed. Engl. 34 (18) (1995) 2056–2058. [7] F. Giess, M.G. Friedrich, J. Heberle, R.L. Naumann, W. Knoll, The protein-tethered lipid bilayer: a novel mimic of the biological membrane, Biophys. J. 87 (5) (2004) 3213–3220. [8] M.G. Friedrich, V.U. Kirste, J.P. Zhu, R.B. Gennis, W. Knoll, R.L.C. Naumann, Activity of membrane proteins immobilized on surfaces as a function of packing density, J. Phys. Chem. B 112 (10) (2008) 3193–3201. [9] S.M. Schiller, R. Naumann, K. Lovejoy, H. Kunz, W. Knoll, Archaea analogue thiolipids for tethered bilayer lipid membranes on ultrasmooth gold surfaces, Angew. Chem. Int. Ed. 42 (2) (2003) 208–211. [10] P. Mueller, D.O. Rudin, H.T. Tien, W.C. Wescott, Reconstitution of cell membrane structure in vitro and its transformation into an excitable system, Nature 194 (4832) (1962) 979–980. [11] H.M. Keizer, B.R. Dorvel, M. Andersson, D. Fine, R.B. Price, J.R. Long, et al., Functional ion channels in tethered bilayer membranes—implications for biosensors, Chembiochem 8 (11) (2007) 1246–1250. [12] R. Naumann, S.M. Schiller, F. Giess, B. Grohe, K.B. Hartman, I. Karcher, et al., Tethered lipid bilayers on ultraflat gold surfaces, Langmuir 19 (13) (2003) 5435–5443. [13] U. Rothe, H. Aurich, Lipid-coated particles—a new approach to fix membrane-bound enzymes onto carrier surfaces, Biotechnol. Appl. Biochem. 11 (1) (1989) 18–30. [14] R. Naumann, T. Baumgart, P. Graber, A Jonczyk, A. Offenhausser, W. Knoll, Proton transport through a peptide-tethered bilayer lipid membrane by the H(þ)-ATP synthase from chloroplasts measured by impedance spectroscopy, Biosens. Bioelectron. 17 (1–2) (2002) 25–34. [15] N. Bunjes, E.K. Schmidt, A. Jonczyk, F. Rippmann, D. Beyer, H. Ringsdorf, et al., Thiopeptide-supported lipid layers on solid substrates, Langmuir 13 (23) (1997) 6188–6194. [16] E.K. Schmidt, T. Liebermann, M. Kreiter, A. Jonczyk, R. Naumann, A. Offenhausser, et al., Incorporation of the acetylcholine receptor dimer from Torpedo californica in a peptide supported lipid membrane investigated by surface plasmon and fluorescence spectroscopy, Biosens. Bioelectron. 13 (6) (1998) 585–591. [17] R. Naumann, E.K. Schmidt, A. Jonczyk, K. Fendler, B. Kadenbach, T. Liebermann, et al., The peptide-tethered lipid membrane as a biomimetic system to incorporate cytochrome c oxidase in a functionally active form, Biosens. Bioelectron. 14 (7) (1999) 651–662. [18] E.K. Sinner, U. Reuning, F.N. Kok, B. Sacca, L. Moroder, W. Knoll, et al., Incorporation of integrins into artificial planar lipid membranes: characterization by plasmon-enhanced fluorescence spectroscopy, Anal. Biochem. 333 (2) (2004) 216–224. [19] B. Wiltschi, W. Knoll, E.K. Sinner, Binding assays with artificial tethered membranes using surface plasmon resonance, Methods 39 (2) (2006) 134–146. [20] H. Song, E.K. Sinner, W. Knoll, Peptid-tethered bilayer lipid membranes and their interaction with amyloid beta-peptide, Biointerphases 2 (4) (2007) 151–158. [21] R. Robelek, E.S. Lemker, B. Wiltschi, V. Kirste, R. Naumann, D. Oesterhelt, et al., Incorporation of in vitro synthesized GPCR into a tethered artificial lipid membrane system, Angew. Chem. Int. Ed. Engl. 46 (4) (2007) 605–608. [22] R. Mahfoud, N. Garmy, M. Maresca, N. Yahi, A. Puigserver, J. Fantini, Identification of a common sphingolipid-binding domain in Alzheimer, prion, and HIV-1 proteins, J. Biol. Chem. 277 (13) (2002) 11292–11296.

SELF-ASSEMBLED TETHERED BIMOLECULAR LIPID MEMBRANES

179

[23] A. Kakio, Y. Yano, D. Takai, Y. Kuroda, O. Matsumoto, Y. Kozutsumi, et al., Interaction between amyloid beta-protein aggregates and membranes, J. Pept. Sci. 10 (10) (2004) 612–621. [24] A. Kakio, S.I. Nishimoto, K. Yanagisawa, Y. Kozutsumi, K. Matsuzaki, Cholesteroldependent formation of GM1 ganglioside-bound amyloid beta-protein, an endogenous seed for Alzheimer amyloid, J. Biol. Chem. 276 (27) (2001) 24985–24990. [25] W.R. Leiefert, A.L. Aloia, O. Bucco, R.V. Glatz, E.J. McMurchie, G-protein-coupled receptors in drug discovery: nanosizing using cell-free technologies and molecular biology approaches, J. Biomol. Screening 10 (8) (2005) 765–779. [26] C. Duschl, M. Liley, H. Lang, A. Ghandi, S.M. Zakeeruddin, H. Stahlberg, et al., Sulphurbearing lipids for the covalent attachment of supported lipid bilayers to gold surfaces: a detailed characterisation and analysis, Mater. Sci. Eng. C Biomim. Mater. Sens. Syst. 4 (1) (1996) 7–18. [27] G. Elender, M. Kuhner, E. Sackmann, Functionalisation of Si/SiO2 and glass surfaces with ultrathin dextran films and deposition of lipid bilayers, Biosens. Bioelectron. 11 (6–7) (1996) 565–577. [28] L. He, J.W. Robertson, J. Li, I. Karcher, S.M. Schiller, W. Knoll, et al., Tethered bilayer lipid membranes based on monolayers of thiolipids mixed with a complementary dilution molecule. 1. Incorporation of channel peptides, Langmuir 21 (25) (2005) 11666–11672. [29] K. Ataka, F. Giess, W. Knoll, R. Naumann, S. Haber-Pohlmeier, B. Richter, et al., Oriented attachment and membrane reconstitution of His-tagged cytochrome c oxidase to a gold electrode: in situ monitoring by surface-enhanced infrared absorption spectroscopy, J. Am. Chem. Soc. 126 (49) (2004) 16199–16206. [30] M.G. Friedrich, F. Giebeta, R. Naumann, W. Knoll, K. Ataka, J. Heberle, et al., Active site structure and redox processes of cytochrome c oxidase immobilised in a novel biomimetic lipid membrane on an electrode, Chem. Commun. (Camb) (21) (2004) 2376–2377. [31] M.G. Friedrich, J.W. Robertson, D. Walz, W. Knoll, R.L Naumann, Electronic wiring of a multi-redox site membrane protein in a biomimetic surface architecture, Biophys. J. 94 (9) (2008) 3698–3705. [32] R. Naumann, D. Walz, S.M. Schiller, W. Knoll, Kinetics of valinomycin-mediated Kþ ion transport through tethered bilayer lipid membranes, J. Electroanal. Chem. 550 (2003) 241–252.

This page intentionally left blank

INDEX A Activated protein C (APC) system. See also Factor V Leiden mutation FVa inactivation FXa interaction, 128 mechanism, 126 platelet‐derived chemokine PF4, 128 protein S role, 128 FVIIIa inactivation, 129 FV Leiden mutation molecular mechanisms, 134–136 population genetics, 132–134 thrombosis risk, 136–137 resistance altered coagulation factors, 142 Arg360 cleavage site mutation, 141 discovery, 130–131, 132 FV gene mutations, 138–142 FVIII level, 142 FV liverpool, 141 FV R2 haplotype, 141–142 laboratory assessment, 137–138 prevalence, 131 protein S, 143 prothrombin, 142–143 TFPI level, 143 thrombosis risk, 131–132 Acute kidney injury (AKI) Acute Kidney Injury Network (AKIN), 77 acute renal failure (ARF), 76 biomarkers, 90–91 development and implementation, 90–91 discovery, sources, 80–81 enzymuria, 81–82 fatty‐acid binding protein (FAPB), 89–90 interleukin‐18, 88–89 kidney injury molecule‐1 (KIM‐1), 84–86 limitations, 75 N‐acetyl‐b‐glucosaminidase, 83 neutrophil gelatinase‐associated lipocalin (NGAL), 86–88

specificity and sensitivity, 80 types and characteristics, 78–80 uninary proteins and antigens, 82–83 classification scheme, 77 pathophysiological processes, 74 serum creatinine and GFR, 77–79 Acute myocardial infarction (AMI) cardiac troponin, 3–4 (see also High‐sensitivity troponins) CK‐MB isoenzyme and myoglobin, 3–4 Alpha‐galactosidase A (AGAL) deficiency, 58 enzyme replacement therapy, 60 inheritance, 59 inhibition, globotriaosylsphingosine, 62 Anderson–Fabry disease. See Fabry disease Anoxia, 39 Apoptosis, 19, 21 B Biomarkers acute kidney injury (AKI), 90–91 development and implementation, 90–91 discovery, sources, 80–81 enzymuria, 81–82 fatty‐acid binding protein (FAPB), 89–90 interleukin‐18, 88–89 kidney injury molecule‐1 (KIM‐1), 84–86 limitations, 75 N‐acetyl‐b‐glucosaminidase, 83 neutrophil gelatinase‐associated lipocalin (NGAL), 86–88 specificity and sensitivity, 80 types and characteristics, 78–80 uninary proteins and antigens, 82–83 bone‐turnover markers, 107 antiosteoporotic drugs, 109 decreased estrogen level, 112 osteoprogenitor cells, 108–109 pituitary–gonad axis, 110–112 serum changes, 108 collagen oxidative damage

181

182

INDEX

Biomarkers (cont. ) amino acid modifications, 43 collagen family, 32–33 glycoxidation, 41–42 glycoxidation products, 43–45 nitrated peptides, 45–47, 45–48 oxidative cleavage, 40–41 Bisphosphonates, 102 Bone marrow transplantation (BMT). See also Post‐BMT bone and mineral metabolism bone loss, clinical features autologous vs. allotransplant patients, 101 bisphosphonates, 102 fractures, 102 hematopoietic stem cell recipients, 101 bone‐turnover biochemical markers, 104–109 bone modeling and remodeling, 102–104 sex hormones, 110–112 calcium, 109 cytokines, 114 growth factors fibroblast growth factor (FGF), 115 insulin‐like growth factor (IGF), 114–115 macrophage‐colony stimulating factor (M‐CSF), 115 osteoporosis antiosteoporotic drugs, 109 glucocorticoids, 106 parathyroid hormone (PTH), 109–110 vitamin D level, 109–110 C Calcium, 109 Cardiac troponins. See High‐sensitivity troponins Chaperones, 61 Chronic kidney disease (CKD), 74–75 Collagen oxidative damage biomarkers glycoxidation products, 43–45 nitrated peptides, 45–48 collagen family biochemistry, 32–33 structure, 32 nitrated peptides, 45–47 oxidant‐induced changes amino acid modifications, 43 glycoxidation, 41–42

oxidative cleavage, 40–41 Cytochrome c oxidase (CcO) peptide‐tethered lipid bilayer membrane (peptBLM), 166 protein‐tethered bilayer lipid membrane (protBLM) catalytic turnover, 172–173 oriented immobilization, 170–172 reverse orientation, 172, 173 E Enzyme replacement therapy, 60 Enzymuria, 81–82 F Fabry disease. See also Alpha‐galactosidase A (AGAL) deficiency biochemical basis enzyme deficiency, 62 mutations, 63 pathology, 63 secondary biochemical effects, 63–68 clinical picture, 59 diagnosis, 59–60 inheritance, 58–59 treatment chaperones, 61 enzyme replacement therapy, 60 substrate reduction, 60–61 symptomatic, 60 Factor V Leiden mutation, 132. See also Activated protein C (APC) system APC resistance Arg360 cleavage site mutation, 141 FV gene variants, 139–140 FV liverpool, 141 FV R2 haplotype, 141–142 molecular mechanisms APC‐cofactor activity, 135 Arg506 cleavage, 134 FVa inactivation, 134–135 thrombin formation, 135–136 population genetics, 132–134 thrombosis risk, 136–137 Fatty‐acid binding protein (FAPB) expression, 89 heart‐type FABP (H‐FABP), 89–90 liver‐type FABP (L‐FABP), 90

INDEX Fibrillar collagens, 32 Fibroblast growth factor (FGF), 115 G Glomerular filtration rate (GFR), 77–79 Glucocorticoid‐induced osteoporosis, 106 H Heart‐type fatty‐acid binding protein (H‐FABP), 89–90 High‐sensitivity troponins acute coronary syndromes, 9–10 cardiac muscle damage, 4 functions, 4 heart failure (HF) coronary artery disease, 17–18 diagnosis and management, 16–17 incidence and prevalence, 16 molecular cardiac biomarker, 17 progression, 16 risk stratification, 17 multimarker approach natriuretic peptides, 23 troponins, 22–23 myocardial cells, 4–5 myocardial infarction analytic and clinical problems, 9 assay precision, 8–9 detection limit, 8 development, 7 functional sensitivity, 8–9 immunoassays, 5–6 myocardial injury detection cardiovascular events, 19 cTnT elevation, 18 high‐dose chemotherapy, 18 inflammatory and rheumatic diseases, 18–19 myocardial tissue, renewal and remodeling age‐dependent distribution, 20–21 apoptosis, 19, 21 cardiomyocytes, 21–22 gender‐ and age‐related cTnI and cTnT levels, 22 sarcolemmal disruption, 21 release, myocardial damage, 5 serial measurement biologic variation, ACS patients, 15–16 changes, AMI diagnosis, 15

183

subunits, 4 upper reference limit (URL) estimation age‐ and gender‐dependent differences, 11–14 sample size, 11, 12 Hyperoxia, 39 I Insulin‐like growth factor (IGF), 114–115 Interleukin‐18 caspase‐1 inhibition, 88 diagnostic utility, 88–89 inflammatory diseases, 88 transplantation, 88 Ischemia, 38–39 K Kidney diseases. See Acute kidney injury (AKI); Chronic kidney disease (CKD) Kidney injury molecule‐1 (KIM‐1) immunohistochemical assessment, 85–86 quantification areas under the curve (AUC), 84–85 transplant biopsies, 85 structure and expression data, 84 L Langmuir monolayers, 160–161 Liver‐type fatty‐acid binding protein (H‐FABP), 90 M Macrophage‐colony stimulating factor (M‐CSF), 115 Myeloperoxidase (MPO), 36 Myocardial infarction, cTNI and cTNT high‐sensitivity immunoassays analytic and clinical problems, 9 assay precision, 8–9 detection limit, 8 development, 7 functional sensitivity, 8–9 immunoassays 99th URL, 5 antibody specificity, 6 epitope location, 6

184

INDEX N

N‐Acetyl‐b‐glucosaminidase, 83 Natriuretic peptides, 23 Neutrophil gelatinase‐associated lipocalin (NGAL) cardiac surgery, 86–87 diabetic nephropathy, 87–88 diarrhea‐associated hemolytic uremic syndrome, 87 structure and expression, 86 Nitrated peptides, 45–47 Non ST‐segment elevation myocardial infarction (NSTEMI), 10 O Osteoblasts fibroblast growth factor (FGF), 115 insulin‐like growth factor (IGF), 114–115 macrophage‐colony stimulating factor (M‐CSF), 115 Osteoporosis antiosteoporotic drugs, 109 glucocorticoids, 106 Osteoprotegerin, 112–113 Oxidative collagen damage. See also Reactive nitrogen and oxygen species (RNOS) biomarkers glycoxidation products, 43–45 nitrated peptides, 45–48 collagen family biochemistry, 32–33 structure, 32 oxidant stress, 39–40 oxygen, chemical nature and reactivity inertia, 33–34 RNOS family, 34–35 RNOS biological activity, 37–39 in vivo production, 35–37 P P19 peptide‐tethered membrane, 166–168 Parathyroid hormone (PTH), 109–110 Peptide‐tethered lipid bilayer membrane (peptBLM) cell‐free expression systems, 167–170 cytochrome c oxidase (CcO) insertion, 166

G protein‐coupled receptors (GPCRs), 167, 177 Hþ‐ATPases incorporation, 165–166 nicotinic acetylcholine receptor (nAChR) insertion, 166 P19 peptide, 166–168 peptide spacer, 165, 170 self‐assembly, gold, 165, 166 Post‐BMT bone and mineral metabolism. See also Bone marrow transplantation (BMT) bone‐turnover markers, 107 antiosteoporotic drugs, 109 decreased estrogen level, 112 osteoprogenitor cells, 108–109 pituitary–gonad axis, 110–112 serum changes, 108 cyclosporine, 106, 108 glucocorticoids, 106 Private mutations, 59, 63 Protein C system. See also Activated protein C (APC) system activated protein C (APC) system altered coagulation factors, 142 Arg360 cleavage site mutation, 141 discovery, 130–131, 132 FVa inactivation, 126, 128 FV gene mutations, 138–142 FVIIIa inactivation, 129 FVIII level, 142 FV Leiden mutation, 132–137 FV liverpool, 141 FV R2 haplotype, 141–142 laboratory assessment, 137–138 prevalence, 131 protein S, 143 prothrombin, 142–143 TFPI level, 143 thrombosis risk, 131–132 coagulation factors, 122 protein C activation, 124–125 structure, 123 protein S structure, 124 synthesis, 123 Protein‐tethered bilayer lipid membrane (protBLM) advantage, 170 cytochrome c oxidase (CcO)

INDEX catalytic turnover, 172–173 electron transfer, 170, 172 oriented immobilization, 170–172 reverse orientation, 172, 173 electrical properties, 170 formation, 170 R Reactive nitrogen and oxygen species (RNOS) biological activity anoxia, 39 host defence, 37–38 hyperoxia, 39 inflammatory response, 37–38 ischemia, 38–39 neuronal transmisson, 37 signal transduction, 37 vascular tone maintenance, 37 hydrogen peroxidase, 35 superoxide anion, 34–35 in vivo production, enzymes cells, 35–36 mitochondrial enzymes, 36 myeloperoxidase (MPO), 36 NADPH‐oxidase (NOX) family, 36 nitric oxide synthase (NOS), 36 radical enzymes, 37 xanthine oxidase, 36 Receptor activator of nuclear factor B ligand (RANKL), 112–113

185

electrical barrier properties, 175–176 ion translocation, 176 Langmuir monolayers, 160–161 limitation, 176 NTA/Ni2þ approach, 174–175 peptide‐tethered lipid bilayer membrane (peptBLM) cell‐free expression systems, 167–170 cytochrome c oxidase (CcO) insertion, 166 Hþ‐ATPases incorporation, 165–166 nicotinic acetylcholine receptor (nAChR) insertion, 166 P19 peptide, 166–168 peptide spacer, 165, 170 self‐assembly, gold, 165, 166 protein‐tethered bilayer lipid membrane (protBLM) advantage, 170 cytochrome c oxidase (CcO), 171–173 electrical properties, 170 formation, 170 stochastic mechanism, 174 surface‐dialysis, 176–177 telechelics self‐assembly assembly process, 162–163 barrier properties, 165 components, 162 electrochemical impedance spectroscopy (EIS), 163–165 Thrombosis. See Venous thrombosis Troponins. See High‐sensitivity troponins

S U Solid supported membranes, 161 Surface‐enhanced IR absorption spectroscopy (SEIRAS), 168, 170 T Telechelics self‐assembly assembly process, 162–163 barrier properties, 165 components, 162 electrochemical impedance spectroscopy (EIS), 163–165 Tethered bimolecular lipid membranes (tBLMs) biosensors, 175 chemical coupling scheme, 174

Upper reference limit (URL) estimation, troponins age‐ and gender‐dependent differences demographic and clinical characteristics, 14 ostensibly healthy individuals, 12–13 reference population studies, 11–12 systolic and diastolic dysfunction, 13–14 sample size, 11, 12 Urinary biomarkers enzymuria, 81–82 fatty‐acid binding protein (FAPB), 89–90 interleukin‐18, 88–89 kidney injury molecule‐1 (KIM‐1), 84–86 N‐acetyl‐b‐glucosaminidase, 83

186

INDEX

Urinary biomarkers (cont. ) neutrophil gelatinase‐associated lipocalin (NGAL), 86–88 serum creatinine and GFR, 77–79 uninary proteins and antigens, 82–83 V Venous thrombosis. See also Protein C system APC resistance altered coagulation factors, 142 Arg360 cleavage site mutation, 141 discovery, 130–131, 132 FV gene mutations, 138–142 FVIII level, 142 FV liverpool, 141 FV R2 haplotype, 141–142 laboratory assessment, 137–138 prevalence, 131

protein S, 143 prothrombin, 142–143 TFPI level, 143 thrombosis risk, 131–132 Factor V Leiden mutation APC‐cofactor activity, 135 Arg506 cleavage, 134 FVa inactivation, 134–135 population genetics, 132–134 thrombin formation, 135–136 thrombosis risk, 136–137 pathogenesis, 122 Vitamin D, 109–110 X Xanthine oxidase, 36 X‐linked lysosomal storage disorder. See Fabry disease

A Lipid rafts

Cholesterol/sphingolipid-rich domains B Lipid organization and membrane geometry

Ordered state

Disordered state: d & A

FIG. 3, DAS AND NAIM, Structural characteristics of lipid rafts. Lipid rafts are specialized membrane domains enriched in cholesterol and glycosphingolipids (A). The extended fatty-acid chains of lipids within these membrane structures generate a more tightly packed domain with higher order; lipids with long, straight acyl chains are preferentially incorporated into the rafts and also alter membrane geometry (B).

APC Cleavage at Arg169

TM

TM

PC IIa

EPCR

12 a.a.

IIa

EPCR

FIG. 1, SEGERS AND CASTOLDI, Protein C activation. Protein C (PC; red) is activated by thrombin (IIa; yellow) on the surface of endothelial cells. During this process, the transmembrane receptors thrombomodulin (TM; green) and endothelial protein C receptor (EPCR; blue) bind thrombin and protein C, respectively, and closely align them for optimal cleavage. Cleavage of a single peptide bond (Arg169) converts protein C to its active form, activated protein C (APC; red), which is released into the circulation with the capacity to inactivate procoagulant cofactors FVa and FVIIIa. Modified from Ref. [235].

A

Arg Arg Arg 306 506 679

A2

A1

A1 A2 A3

PS APC

C1 C2

A3

C1 C2

A3

C1 C2

Ca2+

+ Protein S FXa Prothrombin −

+ − −

Arg 562

B Arg 336

A2 A1

FV

A2 A1 A3

PS APC

C1 C2

Ca2+ Protein S FV FIXa FX

+ + −

+ + −

FIG. 2, SEGERS AND CASTOLDI, Anticoagulant activity of APC. (A) APC-mediated inactivation of FVa. APC (red) inactivates FVa (green) via limited proteolysis at residues Arg306, Arg506, and Arg679 in the heavy chain. This reaction occurs on a phospholipid surface and is greatly stimulated by the APC cofactor protein S (PS; purple). The effects of protein S, FXa, and prothrombin on the individual cleavage sites (þ, stimulation; , inhibition) are indicated. (B) APC-mediated inactivation of FVIIIa. APC (red) inactivates FVIIIa (orange) via limited proteolysis at residues Arg336 (A1 domain) and Arg562 (A2 domain). This reaction occurs on a phospholipid surface and is greatly stimulated by the APC cofactors protein S (PS; purple) and FV (green). The effects of protein S, FV, FIXa, and FX on the individual cleavage sites (þ, stimulation; , inhibition) are indicated.

A

mRNA Ribosome

In vitro expression cDNA

RNA polymerase

mRNA

Plasmid

tBLM

C 5⫻106

Cy5 Antimouse-Cy5 (goat)

Fluorescence (cps)

B

Anti-VSV (mouse)

I

II

III

IV

V

VI

4⫻106 3⫻106

N-terminal VSV-tag

2⫻106 1⫻106

Reference: no plasmid

VII

C-terminal VSV-tag

0 COO−

0

10

20

30

40

50

Time (min) FIG. 5, SINNER ET AL., In vitro expression of membrane proteins. (A) The protein of interest is cloned into an appropriate plasmid-vector. A mixture of cDNA and cell lysate leads to the expression and incorporation of the membrane protein into the tBLM. (B) Schematic of the immunofluorescent analysis for the vectorial insertion of the odorant receptor OR5. The monoclonal antibody binds to the amino-terminal VSV tag and the secondary Cy5‐labeled antibody provides the specific fluorescence signal. (C) SPFS analysis for the vectorial insertion of OR5.