In Vivo Models of Inflammation

Progress in Inflammation Research Series Editor Prof. Michael J. Parnham PhD Senior Scientific Advisor GSK Research Ce...

Author: Christopher S. Stevenson | Lisa A. Marshall | Douglas W. Morgan

27 downloads 1058 Views 2MB 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

Progress in Inflammation Research

Series Editor Prof. Michael J. Parnham PhD Senior Scientific Advisor GSK Research Centre Zagreb Ltd. Prilaz baruna Filipovic´a 29 HR-10000 Zagreb Croatia Advisory Board G. Z. Feuerstein (Wyeth Research, Collegeville, PA, USA) M. Pairet (Boehringer Ingelheim Pharma KG, Biberach a. d. Riss, Germany) W. van Eden (Universiteit Utrecht, Utrecht, The Netherlands)

Forthcoming titles: Chemokine Biology: Basic Research and Clinical Application, Volume II: Pathophysiology of Chemokines, K. Neote, G. L. Letts, B. Moser (Editors), 2006 The Resolution of Inflammation, A. G. Rossi, D. A. Sawatzky (Editors), 2006 (Already published titles see last page.)

In Vivo Models of Inflammation 2nd Edition, Volume I

Christopher S. Stevenson Lisa A. Marshall Douglas W. Morgan Editors

Birkhäuser Verlag Basel · Boston · Berlin

Editors Christopher S. Stevenson Novartis Institutes for Biomedical Research Respiratory Disease Area Novartis Horsham Research Centre Wimblehurst Road Horsham, West Sussex United Kingdom

Lisa A. Marshall Johnson and Johnson PRD Welsh & McKean Rd Spring House, PA 19477 USA

Douglas W. Morgan Portfolio, Program and Alliance Management BiogenIdec 14 Cambridge Center Cambridge, MA 02142 USA

A CIP catalogue record for this book is available from the Library of Congress, Washington D.C., USA

Bibliographic information published by Die Deutsche Bibliothek Die Deutsche Bibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data is available in the internet at http://dnb.ddb.de

ISBN-10: 3-7643-7519-1 Birkhäuser Verlag, Basel – Boston – Berlin ISBN-13: 978-3-7643-7519-5 Birkhäuser Verlag, Basel – Boston – Berlin The publisher and editor can give no guarantee for the information on drug dosage and administration contained in this publication. The respective user must check its accuracy by consulting other sources of reference in each individual case. The use of registered names, trademarks etc. in this publication, even if not identified as such, does not imply that they are exempt from the relevant protective laws and regulations or free for general use. This work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, re-use of illustrations, recitation, broadcasting, reproduction on microfilms or in other ways, and storage in data banks. For any kind of use, permission of the copyright owner must be obtained. © 2006 Birkhäuser Verlag, P.O. Box 133, CH-4010 Basel, Switzerland Part of Springer Science+Business Media Printed on acid-free paper produced from chlorine-free pulp. TCF ' Cover design: Markus Etterich, Basel Cover illustration: see p. 44; with friendly permission of Leo Joosten Printed in Germany ISBN-10: 3-7643-7519-1 ISBN-13: 978-3-7643-7519-5 987654321

e-ISBN-10: 3-7643-7520-5 e-ISBN-13: 978-3-7643-7520-1 www.birkhauser.ch

Contents Volume I (contents of volume II see last page)

List of contributors

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

vii

Preface to the first edition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

xi

Preface to the second edition. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

xii

Lisa R. Schopf, Karen Anderson and Bruce D. Jaffee Rat models of arthritis: Similarities, differences, advantages, and disadvantages in the identification of novel therapeutics . . . . . . . . . . . . . . . . . . . .

1

Leo A.B. Joosten and Wim B. van den Berg Murine collagen induced arthritis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

35

Stephen A. Stimpson, Virginia B. Kraus and Bajin Han Use of animal models of osteoarthritis in the evaluation of potential new therapeutic agents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

65

Roberto Benelli, Guido Frumento, Adriana Albini and Douglas M. Noonan Models of inflammatory processes in cancer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

83

Antonio Musarò and Nadia Rosenthal Advances in stem cell research: use of stem cells in animal models of muscular dystrophy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103 Karen F. Kozarsky Gene transfer technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125 Matthias Müller and Nicole Avitahl-Curtis Transgenics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139 Sreekant Murthy, Elisabeth Papazoglou, Nandhakumar Kanagarajan and Narasim S. Murthy Nanotechnology: Towards the detection and treatment of inflammatory diseases. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155

Contents

Susan Brain UK legislation of in vivo aspects in inflammation research

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

177

Naoko Kagiyama, Takuya Ikeda and Tatsuji Nomura Japanese guidelines and regulations for scientific and ethical animal experimentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187 Joanne B. Morris, Jeffrey Everitt and Margaret S. Landi United States guidelines and regulations in animal experimentation . . . . . . . . . . . . . 193 Index

vi

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

203


Adriana Albini, Dept of Translational Oncology, Istituto Nazionale per la Ricerca sul Cancro, Genova, Italy; e-mail: [email protected] Karen Anderson, Novartis Institutes for Biomedical Research Inc, 250 Massachusetts Ave, Cambridge, MA 02139, USA; e-mail: [email protected] Nicole Avitahl-Curtis, Novartis Institute for Biomedical Research, 100 Technology Square, Cambridge, MA, USA; e-mail: [email protected] Roberto Benelli, Dept of Translational Oncology, Istituto Nazionale per la Ricerca sul Cancro, Genova, Italy; e-mail: [email protected] Susan Brain, King’s College London, Cardiovascular Division, Guy’s Campus, London SE1 1UL, UK; e-mail: [email protected] Jeffrey Everitt, GlaxoSmithKline Pharmaceuticals, LAS, 709 Swedeland Rd, King of Prussia, PA 19406, USA; e-mail: [email protected] Guido Frumento, Dept of Translational Oncology, Istituto Nazionale per la Ricerca sul Cancro, Genova, Italy; e-mail: [email protected] Bajin Han, GlaxoSmithKline, Research Triangle Park, NC 27709, USA; e-mail: [email protected] Takuya Ikeda, GlaxoSmithKline, Tsukuba Research Laboratories, 43 Wadai, Tsukuba 300-4243, Japan; e-mail: [email protected] Bruce D. Jaffee, Novartis Institutes for Biomedical Research Inc, 250 Massachusetts Ave, Cambridge, MA 02139, USA; e-mail: [email protected]

vii


Leo A.B. Joosten, Rheumatology Research and Advanced Therapeutics, Department of Rheumatology, Radboud University Nijmegen Medical Centre, PO Box 9101, 6500 HB Nijmegen, The Netherlands; e-mail: [email protected] Naoko Kagiyama, Central Institute for Experimental Animals, 1430 Nogawa, Miyamae, Kawasaki 216-0001, Japan; e-mail: [email protected] Nandhakumar Kanagarajan, Division of Gastroenterology and Hepatology, Drexel University College of Medicine, Philadelphia, USA; e-mail: [email protected] Karen F. Kozarsky, Biopharmaceuticals, Center of Excellence for Drug Discovery, GlaxoSmithKline, 709 Swedeland Road, King of Prussia, PA 19406, USA; e-mail: [email protected] Virginia B. Kraus, Duke University Medical Center, Durham, NC 27710, USA; e-mail: [email protected] Margaret S. Landi, GlaxoSmithKline Pharmaceuticals, LAS, 709 Swedeland Rd, King of Prussia, PA 19406, USA; e-mail: [email protected] Joanne B. Morris, GlaxoSmithKline Pharmaceuticals, LAS, 709 Swedeland Rd, King of Prussia, PA 19406, USA; e-mail: [email protected] Matthias Müller, Novartis Pharma AG, WSJ-386.409, 4002 Basel, Switzerland; e-mail: [email protected] Narasim S. Murthy, Associated Radiologists, PA, 322 E. Antietam Street, Suite 106, Hagerstown, MD 21740, USA; e-mail: [email protected] Sreekant Murthy, Division of Gastroenterology and Hepatology and Office of Research, Drexel University College of Medicine, Philadelphia, USA; e-mail: [email protected] Antonio Musarò, Department of Histology and Medical Embryology, CE-BEMM and Interuniversity Institute of Myology, University of Rome “La Sapienza”, Via A. Scarpa 14, 00161 Rome, Italy; e-mail: [email protected] Tatsuji Nomura, Central Institute for Experimental Animals, 1430 Nogawa, Miyamae, Kawasaki 216-0001, Japan; e-mail: [email protected]

viii


Douglas M. Noonan, Dept of Clinical and Biological Sciences, University of Insubria, Varese, Italy; e-mail: [email protected] Elisabeth Papazoglou, School of Biomedical Engineering, Drexel University, Philadelphia, USA Nadia Rosenthal, European Molecular Biology Laboratory, Mouse Biology Unit, Monterotondo, 00016 Rome, Italy; e-mail: [email protected] Lisa R. Schopf, Abbott Bioresearch Center, 100 Research Drive, Worcester, MA 01605, USA; e-mail: [email protected] Stephen A. Stimpson, GlaxoSmithKline, Research Triangle Park, NC 27709, USA; e-mail: [email protected] Wim B. van den Berg, Rheumatology Research and Advanced Therapeutics, Department of Rheumatology, Radboud University Nijmegen Medical Centre, PO Box 9101, 6500 HB Nijmegen, The Netherlands; e-mail: [email protected]

ix

Preface to the first edition

The purpose of this volume in the series Progress in Inflammation Research is to provide the biomedical researcher with a description of the state of the art of the development and use of animal models of diseases with components of inflammation. Particularly highlighted are those models which can serve as in vivo correlates of diseases most commonly targeted for therapeutic intervention. The format is designed with the laboratory in mind; thus it provides detailed descriptions of the methodologies and uses of the most significant models. Also, new approaches to the development of future models in selected therapeutic areas have been highlighted. While emphasis is on the newest models, new information broadening our understanding of several well-known models of proven clinical utility is included. In addition, we have provided coverage of transgenic and gene transfer technologies which will undoubtedly serve as tools for many future approaches. Provocative comments on the cutting edge and future directions are meant to stimulate new thinking. Of course, it is important to recognize that the experimental use of animals for human benefit carries with it a solemn responsibility for the welfare of these animals. The reader is referred to the section on current regulations governing animal use which addresses this concern. To fulfill our purpose, the content is organized according to therapeutic areas with the associated models arranged in subcategories of each therapeutic area. Concepts presented are discussed in the context of their current practice, including intended purpose, methodology, data and limitations. In this way, emphasis is placed on the usefulness of the models and how they work. Data on activities of key reference compounds and/or standards using graphs, tables and figures to illustrate the function of the model are included. The discussions include ideas on a given model’s clinical correlate. For example, we asked our contributors to answer this question: How does the model mimic what is found in human clinical practice? They have answered this question in many interesting ways. We hope the reader will find the information presented here useful for his or her own endeavours investigating processes of inflammation and developing therapeutics to treat inflammatory diseases. October, 1998

Douglas W. Morgan Lisa A. Marshall

xi

Preface to the second edition

Since our first edition of “In Vivo Models of Inflammation” published in 1999, there has been amazing progress, and an abundance of exciting new information in inflammation research: new technologies, new therapeutics, new understanding of inflammatory processes, … and on and on, have emerged in the past 6 years. Supporting all of this are the fundamentals of inflammation research, i.e., the animal models, known mechanisms, and therapeutic standards, that have continued to provide the basis for generating these advances. Given the great progress, we have chosen to provide a second edition to our original text. The second edition of “In Vivo Models of Inflammation” comes to you in two volumes and provides an update of the models included in first edition with expanded coverage and more models. Again, these volumes emphasize the standard models regarded as the most relevant for their disease area. The intent is to provide the scientist with an up-to-date reference manual for selecting the best animal model for their specific question. Updates on previously described models are specifically focused on references to any additional pharmacology that has been conducted using these systems. The sections on arthritis models have been expanded and now include models relating to osteoarthritis. New areas described herein include models of neurogenic, cancer, and vascular inflammation. Additionally, coverage of in vivo technologies includes updates on transgenic and gene transfer technologies, and has also been expanded to include chapters on stem cells and nanotechnologies. The second edition continues to emphasize that conducting in vivo research carries with it a great responsibility for animal respect and welfare. The coverage of this concern has been extended to include chapters describing current regulations in the United States, the United Kingdom, and Japan. The ultimate aim of the second edition is to provide current best practices for obtaining the maximum information from in vivo experimentation, while preserving the dignity and comfort of the animal. We hope the information provided here helps in advancing the reader’s endeavors in investigating processes of inflammation and in developing therapeutics to treat inflammatory diseases. May, 2006

xii

Christopher S. Stevenson Lisa A. Marshall Douglas W. Morgan

Rat models of arthritis: Similarities, differences, advantages, and disadvantages in the identification of novel therapeutics Lisa R. Schopf, Karen Anderson and Bruce D. Jaffee Millennium Pharmaceuticals, Cambridge, MA, USA

Introduction The aim of this chapter is to update and expand the information reviewed by Carlson and Jacobson on the topic of rat models of arthritis [1]. Animal models of rheumatoid arthritis (RA) have been extensively used for many years in the evaluation of anti-arthritic agents [2–4]. The most widely used model, adjuvant-induced arthritis (AA) in rats, is discussed in detail here [4–6]. Another common model, which was not included in the first edition and is perhaps more relevant to human RA in terms of cartilage damage, is collagen-induced arthritis (CIA) in the rat [3, 7, 8], and is also outlined here. These two models are compared with a relatively new model, monoarticular streptococcal cell wall–induced arthritis (SCW) in the rat [1, 3, 9, 10]. All of these models share key features related to human RA that make them critical tools in drug development. They have provided information regarding genetic predisposition, prominent cell types, protein and molecular mediators involved in the immunological and inflammatory processes that leads to arthritic pathology.

Historical background The first reported observation that complete Freund’s adjuvant (CFA) could induce polyarthritis in rats was demonstrated by Stoerk and colleagues in 1954 [11] using spleen extracts emulsified in CFA. Shortly thereafter, Pearson showed that CFA alone could induce arthritis in rats [12]. Over the next decade, the AA rat model was used to test a variety of anti-arthritic therapies such as steroids and nonsteroidal anti-inflammatory drugs (NSAIDs) [1, 4]. More recently, this model has been used to assess immunomodulatory drugs such as methotrexate and cyclosporine A as well as therapies designed to block COX-2, TNF-_ or IL-1 [13–21]. Overall, the AA model has been the most extensively used arthritic rat model by the pharmaceutical industry, and has an excellent track record for predicting both activity and toxicity [19]. In Vivo Models of Inflammation, Vol. I, edited by Christopher S. Stevenson, Lisa A. Marshall and Douglas W. Morgan © 2006 Birkhäuser Verlag Basel/Switzerland

1

Lisa R. Schopf et al.

The CIA model was first described in 1977 by Trentham and colleagues, and has since gained favor by providing clues into the pathogenesis of arthritis and related disorders as well as its predictive value for testing anti-rheumatic therapies [7, 8]. This model has been used to evaluate NSAIDs, methotrexate and cyclosporine A as well as newer therapies, which block TNF-_ and/or IL-1 [19, 22–27]. Although there are more data using the AA model, the rat CIA model has also proven to have predictive value for many current therapies and tends to be favored when examining protection against cartilage destruction because the lesion is more comparable to human RA than in the AA model [19]. Although we have known since the 1950s that injections of streptococcal cell wall components or more specifically covalent complexes of peptidoglycan and polysaccharide (PG-PS) from group A streptococci can induce rheumatic-like lesions, the monoarticular SCW model which is described in this chapter, was not developed until the mid-1980s, and has not been routinely used for pharmacological screening [1, 28–31]. However, in this review we provide details on methods and disease parameters such as joint swelling, histopathology, gene expression, serum acute-phase proteins and cartilage and bone markers. We also report efficacy determinations using this model to evaluate some of today’s current therapeutics.

Drug therapies There has been and continues to be extensive research in the area of drug development to treat human RA, but it is not the intention of this review to fully recount or explore all of these efforts [1–3, 32–36]. However, we have provided information on some of the more commonly used RA therapies and their efficacy in the three rat arthritic models (Tab. 1). This table focuses solely on therapeutic dosing regimes, although there is a vast amount of literature that explores prophylactic treatments as well [1, 2, 13, 14, 22, 23, 30, 37–44]. When patients first present with symptoms, the primary care physician will typically suggest the use of NSAIDs to provide some relief from pain and stiffness. Two examples of commonly used NSAIDs are ibuprofen and naproxen in Table 1, where we provide efficacy data in the rat models of arthritis. The main disadvantage to these treatments is that they only provide partial relief from pain and stiffness, but do not radically change the course of disease progression, as also predicted using the animal models (AA and CIA) [40]. They are typically only tolerated for short periods of time, after which patients can experience any number of gastrointestinal toxicity problems. Sometimes, corticosteroids are prescribed at this early stage, but, despite their potent anti-inflammatory action, they too have many dose-dependent side effects. Again, both the AA and CIA rat models predicted this outcome observed in patients [2]. Alternatively, corticosteroids such as prednisone are used

2

Rat models of arthritis: Similarities, differences, advantages, and disadvantages…

Table 1 - Efficacy of standard RA drugs using therapeutic dosing regimes in rat models of arthritis: AA, CIA and monoarticular SCW Drug class

AA ED50 (mg/kg)

Ref.

CIA ED50 (mg/kg)

75a; 65a

[123, 124]

Naproxen Corticosteroids Prednisolone Dexamethasone DMARDS Methotrexate

7a

[125, 126]

49% @ [42] 25 mg/kge ND

ND

0.3a 0.005a; 0.01b

[124] [1]; MPI

ND 0.01f

ND 0.01i

Inactivea,b

[1]; MPI

0.1g

Cyclosporine A Leflunomide

2.4c 53% @ 32 mg/kgd

[127] [128]

Enhancede,f MPI; [42, 43] ND

56% @ 0.1–0.5i 18i ND

11h

6000× 6× 7×

2–3× 2–5×

2–3× 2–3×

ND ND

15–50× 15–40×

15–40× 3–20×

40× 15×

50× 5–10× 20×

20× 2–5× 10×

6× 5× ND

*Higher baseline associated with mineral oil injected control group (AA baseline) compared to saline or IFA injected controls (SCW, CIA baseline)

COX-1, and B29 (Ig beta) failed to demonstrate significant change in expression or differential expression between the experimental groups in all models.

Cytokine levels Plasma/serum and tissue cytokine levels have been studied by several investigators, using a variety of bioassay and ELISA methods. In AA, Philippe et al. [5] noted a spike in serum levels of TNF-_ and IL-6 (not IL-1`, which remained unchanged) between 6 and 12 h after CFA injection. These levels returned to baseline and then gradually increased up to day 20, with a greater magnitude of increase observed for IL-6 than TNF-_. Szekanecz et al. [89] reported concomitant increases in serum and joint cytokine levels between days 11 and 25 for TNF-_ and IL-6, and increased

12


joint IL-1` as well. Smith-Oliver et al. [107] were able to detect TNF-_ in AA joints at day 20. In CIA joints (days 14–28), Magari et al. [108] found elevated levels of IL-1` and IL-6 protein, but not TNF-_. Rioja et al. [30] have examined joint cytokine levels in monoarticular SCW and found a good correlation of mRNA and protein increases for IL-6 and IL-1`, both immediately following the i.a. injection and 3 days after the i.v. challenge. Increased TNF-_ mRNA, but not protein, was also detected. The consistent detection of elevated joint TNF-_ mRNA (Tab. 4), as well as the anti-arthritic efficacy of anti-TNF-_ therapies, in all of the models supports the concept that perhaps the current methods are not sufficiently sensitive to detect small but significant increases of TNF-_ protein in these joints.

Cartilage pathology The articular cartilage can be assessed using histological methods, although routine hematoxylin and eosin (H&E) staining is generally not sufficient for visualizing subtle changes. Special stains for proteoglycan such as toluidine blue or SafraninO can be used to demonstrate a compromise of the cartilage layer. However, histology-based methods are not quantitative. Methods of measuring cartilage synthesis using uptake of radiolabeled sulfate or glycosaminoglycan quantitation has been used to some extent and show reduced cartilage formation in arthritis [5, 109]. However, these methods require the sacrifice of the animal. Recently, assays have been developed that utilize non-invasive, blood-based parameters to assess cartilage status. Useful analytes include, cartilage-associated oligomeric matrix protein (COMP), which can be measured in the blood as an indicator of cartilage turnover, and an assay for C terminal collagen type II fragments (CartilapsTM), which can be measured in serum or urine. The latter is generated by the proteolytic cleavage (destruction) of cartilage. Increased levels of COMP have been detected in CIA by us (Tab. 3) and by others [110, 111]; these levels correlated reasonably well with the clinical score at the end of the study. Regardless of method of assessment, cartilage destruction is a prominent feature of both AA and CIA. In AA, however, cartilage destruction often pales in comparison to the striking destruction of bone. Therefore, it is possible that loss of cartilage may in fact be due at least in part to the loss of the underlying bone. In CIA, there is a strong humoral immune response to type II collagen, with high titers of anti-collagen IgG antibodies; thus the articular cartilage is a primary target. In monoarticular SCW, at the peak of ankle swelling, i.e., 3 days following i.v. challenge with the bacterial component, there are minimal changes in the cartilage or bone of the joints. This is likely due to the short course (3 days) of the response. It has been observed that cartilage damage does occur if a greater post-inoculation interval (10–20 days) and/or additional reactivation of the disease is used [29, 31]. Figure 1 illustrates the H&Estained histopathology of all three models.

13


Figure 1 The hind paws were decalcified in Immunocal (American MasterTech Scientific Inc., Lodi, CA) for 4 days. After decalcification paws were embedded in paraffin, sectioned, and stained with H&E. Histological evaluation of the ankle joints (peak disease) was based on five parameters, each scored on a scale of 0–4: composite inflammation (average of cellular infiltrate, edema and joint/tendon effusion), composite bony change (average of periosteal new bone formation and osteolysis), synovial alteration, pannus (fibrinocellular debris/granulation tissue within joint space), and cartilage degeneration. Higher scores indicated more severe disease as defined by each histopathological parameter. H&E-stained sagittal sections through the tibiotalar joint (upper panels) and midfoot region (lower panels) of a normal rat and diseased rats from three different models of arthritis. In the normal rat, the bone (B) and cartilage (C) are intact, and the synovium (S) is of low cellularity, consisting of mainly adipose cells with a synovial cell lining that is only a few cells thick. In the monoarticular SCW 100P joint, taken 3 days after i.v. reactivation, the synovium is hypercellular due to infiltrating inflammatory cells and synoviocyte proliferation. Bone and cartilage are intact, however. In the CIA joint, harvested on day 21 after collagen injection, synovial inflammation is evident, and there are foci in the distal tibia, talus, and midfoot where pannus tissue (black arrows) is invading the bone and cartilage. In AA, at day 19 after CFA injection, synovial inflammation is marked, and the invasion by pannus and resulting bone destruction is dramatic. Total magnification of images is 50×; black bar represents 150 µm.

Bone pathology Changes in bone are prominent features of AA and CIA. By histopathology, there is an extensive effacement of bone by the invasive inflammatory pannus tissue, and there is marked periosteal new bone formation and osteoclastic bone resorption.

14


Figure 2 Comparing bone destruction by micro-computed tomography in rat models of arthritis. µCT imaging was performed on a Scanco µCT-40 (Scanco Medical AG, Zurich, Switzerland). The excised hind paws were secured in 36-mm imaging tubes and bathed in 10% buffered formalin. Approximately 200–300 slices, with 37.6-µm slice thickness, were acquired on a 1024 × 1024 image matrix with digital resolution of 36 µm × 36 µm × 37.6 µm. Other imaging parameters included a 150-ms exposure time, 55 peak kilovolts (kVp) 145 milliamp (mA) and 1000 projections over 360°. A normal rat ankle is shown for reference. In AA (day 19), severe bone loss is evident in the distal tibia and midfoot such that normal structures are difficult to recognize. This loss of bone volume can be quantitated and is significant. In contrast, bone volume loss is less in CIA (day 21). An alternative parameter, bone roughness, has been developed to quantitate the more subtle bony changes that occur in CIA (see Table 3). In monoarticular SCW (day 56), there is minimal bone damage and it is too little for quantitation. On day 24, after the first reactivation flare, in SCW (not shown) there is essentially no bone damage.

These features are seen in both models but are more dramatic and extensive in AA. Chronologically, investigators have noted the presence of large numbers of osteoclasts and their precursors, and bone destruction as early as day 5 in AA, and day 10 onward in CIA [49]. Bone pathology can also be assessed by newer, imagingbased methods. Micro-computed tomography (µCT) has been used by ourselves and others [112–114]. The µCT appearance of a representative arthritic joint from each model is shown in Figure 2, illustrating the differing degrees of bony involvement. The information captured by µCT can be used to quantitate the bone volume loss in AA (% decrease of 40 ± 20), and prove that it is indeed greater than CIA (% decrease, 16 ± 6) (Tab. 3) [114]. Typically, the bone volume loss in CIA is too small

15


to afford a sufficient window for evaluating therapeutics. To address this issue, an alternative µCT-generated parameter, bone roughness, has been developed [114] to quantitate the more subtle bone changes seen in CIA. It is now possible to assess these bony changes in live, anesthetized rats as well as in post-mortem specimens. Magnetic resonance imaging (MRI) has also been used by some investigators, although effective quantitative methods are still under development [1, 115]. Other direct, quantitative methods that have been used include measurement of bone mineral density, and bone mineral content [113, 116, 117]. Indirect methods include blood-based assays for bone metabolites that offer a means of assessing bone destruction in the living rat. Useful analytes include collagen type I fragments (RatLapsTM) [112] (Tab. 3), osteocalcin [115, 118–120], and bone sialoprotein (BSP) [111].

Methods Adjuvant-induced arthritis As previously stated, there are multiple rat strains that can be used in the AA model and there appears to be no differences in susceptibility due to gender, however, in our laboratory we have exclusively used female Lewis rats with a starting weight of 150–170 g. Rats (n = 12) are randomly divided into experimental groups and weighed to determine the average body weight of each group. We also measure each rat’s ankle and paw (maximal lateral) individually with a plethysmometer (Ugo Basile Biological research Apparatus, Italy) to determine the baseline ankle measurement. Calipers can be used to measure ankle thickness as well but using a volume displacement method tends to be less biased by the researcher taking the measurement and encompasses the entire ankle instead a of single location. Rats receive an 0.1-mL injection using a 30 gauge needle of CFA (containing Mycobacterium tuberculosis H37 Ra ATCC 25177 heat killed from Difco Laboratories, Lee Labs, Grayson, GA) at a concentration of 3 mg/mL in mineral oil intradermally (i.d.) into the right hind footpad. Control rats are injected with an equal volume of mineral oil alone. The rats are lightly anesthetized during this injection process. Disease can also be induced if the injection is given i.d. at the base of the tail; however, we have found the robustness of disease between rats is much more consistent when the footpad injection site is used. Additionally, we have reduced the typical dose of 1.0–0.5 mg CFA to 0.3 mg. This does not compromise the therapeutic window, but has reduced the primary lesion on the injected footpad, thereby reducing pain and distress. The onset of arthritis, as indicated by contralateral paw swelling, appears on approximately day 10 post injection (Fig. 3). Both paw volume and body weights are measured throughout the study every 2–3 days. We have found for typical compound screening experiments that 19 days is sufficient to evaluate novel compounds

16


Figure 3 Rat adjuvant-induced arthritis model (AA)

by providing a large enough therapeutic window to determine accurate half maximal effective dose ED50. The longer the disease progresses, the more bone damage occurs and the lesion on the injected paw worsens and even begins to develop on the contralateral paw. There can actually be a reduction in paw swelling over time due to reduced edema and inflammation. In our hands, we have not found the need to cull animals due to lack of disease development or too severe lesion development. However, if this is an issue, refer to an excellent review of this practice in the previous edition of this chapter [1]. Compound testing can be either prophylactic (beginning just prior to CFA injection and continuing throughout the experiment, typically 19 days) or therapeutic (beginning on day 10 and continuing to day 19). To get a sense of drug coverage over the course of one day, we serially bleed animals for pharmacokinetic (PK) analysis. Usually, for PK analyses, we add four additional animals per group and encompass the time points of 0, 1, 3, 6, 8 and 24 h. Therefore, these animals are bled serially five times by restraining the animal using a rat holder and warming the tail with a heat lamp or immersing it in warm water to dilate the vessels. The last time point (24 h) should be done after euthanasia via cardiac puncture. If the day chosen for the full PK determinations is late in disease (day 18 or later) it can be dif-

17


ficult to obtain blood from the tail vein because arthritis in evident in the tail due to the cartilage destruction within the tail. We continue measuring each rat's ankle individually with a plethysmometer (paw volume meter) every 2–3 days to determine the response. We also weigh each rat in each group and determine the average body weight of each group every 2–3 days. A local nonspecific inflammatory reaction occurs at the site of the CFA injection. The initial acute inflammation is observed around day 3 and can progress over time into a lesion. The contralateral hind paw volume increase 65–70% over time (plateau occurs around day 19). Arthritic rats also experience about a 20% reduction in body weight when compared to nonarthritic controls by days 16–60. Body weight loss should be monitored carefully, because if rats experience more than a 20% reduction in their initial body weight, they should be humanely euthanized. As disease progresses, both hind limbs become immobile, and animals drag themselves with their front paws, but they are perfectly capable of reaching their food and water sources. However, to allow for easier access and to encourage them to eat, food pellets or gel packs can be placed within their cages. Swelling can still be observed up to 60 days. No response should be observed in the negative (mineral oil) control group of rats. The typical duration of each novel compound study is 19 days, but on occasion it may be necessary to extend the time period, e.g., to provide a longer treatment period to test for improvement in efficacy, or for removal of a compound after successful efficacy is achieved, to see if disease reduction is maintained or if the disease returns. Other physical changes observed as the disease progresses are decreased food intake, roughening of the hair, and lethargy [4]. Another common sign of distress is chromodacryorrhea [121]. This condition can be mistaken for a bloody discharge from the eyes and nose, but is actually caused by an ocular secretion of a porphyrin-containing pigment deep within the orbits and has been correlated with a stress response [1].

Collagen-induced arthritis The basic principles and procedures, as well as the clinical signs of disease for the CIA model are very similar to those just described for AA in the rat [8]. The biggest difference, as the name implies, is the material used to induce the disease and how it is introduced. On days 0 and 7, shaved female Lewis rats are administered three i.d. injections at the base of the tail using a 25–27 gauge needle (100 µg/0.1 mL/site) of incomplete Freund’s adjuvant (IFA; Difco, Franklin Lakes, NJ) only or IFA plus type II collagen (CII). We use nasal bovine type II collagen (CN276) obtained from Elastin Products Company, Inc. (Owensville, MO) at an initial concentration of 2 mg/mL in 0.01 M acetic acid. This solution is stirred overnight, and then emulsified with equal volume of IFA. A total volume of 0.3 mL is injected at multiple locations (0.1 mL/site); other sites such as intrascapular and flank regions

18


Figure 4 Rat collagen-induced arthritis model (CIA)

can be used too. The rats are anesthetized with isoflurane for the injections. The rats get a booster injection again on day 7. As in the AA model, both paw volume and body weights are measured throughout the study approximately every 2–3 days. Paw volumes are measured using a water displacement plethysmometer and the onset of arthritis, as indicated by increased paw volume, appears on approximately day 14 following the initial injection of collagen. Disease onset and plateau are slightly delayed compared to the AA model and less robust (Fig. 4). No response should be observed in the negative (IFA only) control group of rats. The typical duration of a study for each novel compound is 21 days, but once again it may be necessary to extend the time period. A local nonspecific inflammatory reaction occurs at the site of the collagen injections, which also progresses into a lesion over time. The hind paw volumes increase 45% over time and plateaus around day 21. Arthritic rats may experience up to a 20% reduction in body weight compared to non-arthritic control animals at peak disease (day 21). Arthritic rats then start to gain weight again more rapidly and return to a body weight near that of their non-arthritic controls by day 45. One advantage of this model is that both hind paws can be assessed as part of the disease measurements, whereas in the AA model only the contralateral is appropriate. Although clearly showing the signs and symp-

19


toms of distress as described for the AA model, in all cases it appears to be milder disease. The preparation of the collagen emulsion is one of the most critical steps in this model. The collagen mixture should be added dropwise to the IFA while gently sonicating the resulting mixture. After all the collagen is added, sonicate the solution more robustly for about 30 s. Care needs to be taken not to overheat the collagen, and to maintain a uniform, cool temperature, this entire procedure should be done on ice. The collagen emulsion thickens and should not spill out of beaker if turned upside-down. If the emulsion does not thicken, it should not be used. The collagen model requires more time and skill in the preparation when compared to AA, but there are also advantages. As previously mentioned, both hind limbs can be used for analysis and the milder disease causes less distress to the animals. Other advantages, as discussed in the previous sections, include the antibody component and the directed nature of the cartilage destruction.

Streptococcal cell-wall-induced arthritis Once again many of the comments regarding procedure and general principles of rat arthritis models have already been discussed and apply to SCW as well. There are two different versions of SCW, a polyarthritic or systemic 10S model and the monoarticular 100P model. In the polyarthritic model, a single injection of the 10S fraction of a SCW preparation is given intraperitoneally (i.p.) [31, 70, 122]. An initial acute response occurs and consists of a sharp increase in ankle measurements, which are typically 25% above baseline measurements. This rise reaches a peak in 3–5 days and is followed by a decline in the measurements on subsequent days. A chronic response should manifest around 12–14 days post injection and be seen as a slow increase in the ankle measurements. This is typically seen as 35% over baseline measurements (Fig. 5). The chronic response is remittent and erosive, and should persist for the remainder of the experimental period. Not all animals (~25%) exhibit this initial swelling and are considered non-responders and must be removed, this is a distinct disadvantage of this model. Additionally, the SCW material is much more costly than either adjuvant or collagen preparation. Lastly, we hypothesize that because the antigen is given i.p., animals have organ damage particularly the liver, spleen and kidney. The model we prefer, although it has not been used extensively in the pharmaceutical industry to date, is the monoarticular 100P model [1, 9, 29, 31]. Technically, this model is a bit more difficult, but it provides some advantages over the traditional AA and CIA models. First, in regard to the preparation of the SCW material, PG-PS 100P (Lee Laboratories) is sonicated and diluted in sterile, pyrogen-free saline to a concentration of 500 µg/mL rhamnose units. To induce disease, 10 µL (5 µg rhamnose units per joint) is injected i.a. into the tibiotalar joint space using a 27-

20


Figure 5 Rat streptococcal cell wall-induced systemic arthritis model (SCW)

gauge needle on day 0 and slowly i.v. (400 µL, 250 µg/mL rhamnose) on day 21 via tail vein. Rats are lightly anesthesized during the i.a. injection to relieve the pain associated with the injection and to prevent movement, aiding in accuracy. The first administration of PG-PS 100P causes an acute response with minor paw swelling that peaks after 2–3 days (typically 20% above baseline) and then declines returning to baseline. The i.v. boost causes paw swelling to increase typically 35% above baseline (Fig. 6). [9, 10]. The monoarticular SCW model most closely resembles an arthritic flare seen in patients and has the additional advantage of fewer systemic complications. Overall, this model is less severe than the previously discussed models; as a comparison the AA model has an increase of 65–70% in paw swelling and CIA has a 45% increase over a 10-day period. In the SCW model, the rats continue to gain weight between days 0 and 21 and then experience a 5–10% drop in weight after i.v. reactivation of disease. No response should be observed in the normal (saline) control group of rats. Additionally, to ensure that the model represents a reactivation of disease, it is important to establish that the i.v. injection of PG-PS 1000P alone is nonarthropathic [31].

21


Figure 6 Rat streptococcal cell wall-induced monoarticular arthritis model (SCW)

Conclusions This chapter has provided information on three distinct rat arthritic models, and has demonstrated that they are perhaps not that dissimilar. A summary of disease-related parameters for each of these models is given in Table 5. These models have few qualitative differences; the major departures are quantitative in nature. The AA model clearly exhibits the greatest magnitude of disease as measured by edema, cellular influx, cytokine levels, and bone destruction within the joint. Systemic markers of disease, such as APPs and markers of cartilage and bone destruction, are also the most elevated in this model of arthritis. The CIA model in rats is very similar to the AA model in terms of disease outcome as characterized by the parameters given in Table 5; however, the magnitude of such responses is approximately one third to one half less than what is observed in AA. One important difference that should be highlighted is that cartilage is the primary target in the CIA model, whereas cartilage destruction is a secondary consequence of bone damage in the AA model. Also, the monoarticular SCW model exhibits very little cartilage damage after its first reactive flare (day 24). Another distinction of the CIA model is that autoantibodies to type II collagen have been identified and are known to participate in the disease process, and to date no such autoantibodies have been clearly identified as disease

22

Table 5 - Summary of rat models of arthritis, AA, CIA and monoarticular SCW CIA

SCW (monoarticular SCW; not SCW polyarthritis)

Initiation

Complete Freund’s adjuvant (contains Mycobacteria); ID; in footpad Single injection, day 0

PGPS 100 intra-articularly in ankle on day 0 followed by 3 week sensitization period; Inflammation reactivated by PGPS 100 IV on day 21

Paws affected

Polyarthritis; only contralateral ankle is relevant

Type II collagen in incomplete Freund’s Adjuvant; ID; in base of tail day 0 plus boost required at day 7 Polyarthritis; Both ankles and knees can be taken

Paw volume

GREATEST in magnitude 4.0 mL volume 2.5mL volume Earliest onset day 10; day 14; max/plateaus around day 19 max/plateaus around day 21 Equivalent in magnitude of Onset score to CIA (day 13) Earlier onset in AA (day 9) Equivalent CD4 & CD14 mRNA Equivalent CD4 & CD14 mRNA CD11b 2X greater in AA than CIA Cartilage destruction is secondary Cartilage is primary target; to bone & soft tissue damage Greater loss of proteoglycans Higher serum COMP levels

Pathology score

Cellular influx Cartilage destruction/ turnover (cartilage oligomeric matrix protein =COMP) Bone destruction/ turnover (RatLaps measures type I collagen)

23

Greatest (RATLAPS scores higher) (µCT shows significant bone volume loss)

Inflammation can be induced in both ankles or one side can be used as internal control Small window in paw swelling; 2.0 mL volume Maximum swelling 2–3 days after IV Lower pathology scores (No/minimal bone or cartilage destruction; soft tissue changes primarily) Lower CD4, CD14, CD11b mRNA than both AA, CIA Essentially none

Bone damage; µCT detects Essentially none; significant bone roughness µCT supports histopathology but usually not significant bone volume loss


AA

AA

CIA

Other pro-inflammatory molecules: IL-1` TNF-_ IL-6 COX-2 INOS MMP13

2× higher level of joint mRNA in AA than CIA (Joint cytokine levels also higher in AA) Serum cytokine levels in all three models are generally not detectably increased Acute phase proteins Markedly increased levels of haptoglobin, _1-acid glycoprotein (8–10× over normal) Other systemic inflammation Neutrophils increased to 5–7× indicators: total neutrophils, normal numbers; Fibrinogen 4× fibrinogen levels increased

SCW (monoarticular SCW; not SCW polyarthritis)

Lower mRNA levels than AA or CIA of all of these except COX-2, IL-6 which are comparable to AA

Levels 1/2 to 1/3 of AA levels (3× over normal)

Levels 1/2 to 1/3 of AA levels (3× over normal)

Neutrophils increased to 3–4× normal numbers; Fibrinogen 2× increased

Neutrophils not significantly increased; Fibrinogen 2× increased


24

Table 5 (continued)


contributors in the AA or the SCW models. The monoarticular SCW model most closely resembles an arthritic flare seen in human patients, and although this model shares many of the described features, it is much milder in terms of joint damage and systemic markers when compared to AA and in most cases CIA as well. As related to human disease, these models progress much more rapidly and exhibit marked bone resorption and formation, which is perhaps more similar to ankylosing spondylitis than RA. However, as highlighted in this chapter, these animal models have proven effective in predicting the outcome of many of the current drugs used in the treatment of RA and will continue to be useful tools for discovery of new therapies.

Acknowledgements The authors would like to sincerely thank Anneli Savinainen, Julie Kujawa, Michelle DuPont, Kristina Perry, Matt Silva, Elizabeth Siebert, and Sudeep Chandra from Millennium Pharmaceuticals for their technical support and dedication without which the quality data presented here would not have been possible.

References 1

2 3 4

5

6

7 8

Carlson RP, Jacobson PB (1999) Comparison of adjuvant and streptococcal cell wallinduced arthritis. In: DW Morgan, LA Marshall (eds): In vivo Models of Inflammation. Birkhäuser Verlag, Basel/Boston/Berlin, 1–50 Bendele A (2001) Animal models of rheumatoid arthritis. J Musculoskelet Neuronal Interact 1(4): 377–385 Kannan K, Ortmann RA, Kimpel D (2005) Animal models of rheumatoid arthritis and their relevance to human disease. Pathophysiology 12(3): 167–181 Weichman B (1989) Rat Adjuvant Arthritis: A Model of Chronic Inflammtion. In: JY Chang, AJ Lewis (eds): Pharmacological Methods in the Control of Inflammation. Alan R. Liss, New York, 363–380 Philippe L, Gegout-Pottie P, Guingamp C, Bordji K, Terlain B, Netter P, Gillet P (1997) Relations between functional, inflammatory, and degenerative parameters during adjuvant arthritis in rats. Am J Physiol 273(4 Pt 2): R1550–1556 Van Eden W, Waksman BH (2003) Immune regulation in adjuvant-induced arthritis: possible implications for innovative therapeutic strategies in arthritis. Arthritis Rheum 48(7): 1788–1796 Trentham DE, Townes AS, Kang AH (1977) Autoimmunity to type II collagen an experimental model of arthritis. J Exp Med 146(3): 857-868 Trentham DE, Dynesius-Trentham RA (1989) Type II ccollagen-induced arthritis in the rat. In: JY Chang, AJ Lewis (eds): Pharmacological Methods in the Control of Inflammation. Alan R. Liss, New York, 395–413

25


9

10

11 12 13

14

15

16

17

18

19

20

21

26

Schimmer RC, Schrier DJ, Flory CM, Laemont KD, Tung D, Metz AL, Friedl HP, Conroy MC, Warren JS, Beck B et al (1998) Streptococcal cell wall-induced arthritis: requirements for IL-4, IL-10, IFN-gamma, and monocyte chemoattractant protein-1. J Immunol 160(3): 1466–1471 Schwab JH, Anderle SK, Brown RR, Dalldorf FG, Thompson RC (1991) Pro- and antiinflammatory roles of interleukin-1 in recurrence of bacterial cell wall-induced arthritis in rats. Infect Immun 59(12): 4436–4442 Stoerk H (1954) Chronic polyarthritis in rats injected with spleen in adjuvants. Am J Pathol 30: 616–621 Pearson CM (1956) Development of arthritis, periarthritis and periostitis in rats given adjuvants. Proc Soc Exp Biol Med 91(1): 95–101 Jaffee BD, Kerr JS, Jones EA, Giannaras JV, McGowan M, Ackerman NR (1989) The effect of immunomodulating drugs on adjuvant-induced arthritis in Lewis rats. Agents Actions 27(3–4): 344–346 Rovensky J, Svik K, Stancikova M, Istok R (2003) Effect of immunostimulatory ribomunyl on the preventive treatment of rat adjuvant arthritis with cyclosporine and methotrexate. J Rheumatol 30(9): 2027–2032 Rovensky J, Svik K, Matha V, Istok R, Kamarad V, Ebringer L, Ferencik M, Stancikova M (2005) Combination treatment of rat adjuvant-induced arthritis with methotrexate, probiotic bacteria Enterococcus faecium, and selenium. Ann NY Acad Sci 1051: 570–581 Silva MA, Ishii-Iwamoto EL, Bracht A, Caparroz-Assef SM, Kimura E, Cuman RK, Bersani-Amado CA (2005) Efficiency of combined methotrexate/chloroquine therapy in adjuvant-induced arthritis. Fundam Clin Pharmacol 19(4): 479–489 Noguchi M, Kimoto A, Kobayashi S, Yoshino T, Miyata K, Sasamata M (2005) Effect of celecoxib, a cyclooxygenase-2 inhibitor, on the pathophysiology of adjuvant arthritis in rat. Eur J Pharmacol 513(3): 229–235 Bendele A, McAbee T, Sennello G, Frazier J, Chlipala E, McCabe D (1999) Efficacy of sustained blood levels of interleukin-1 receptor antagonist in animal models of arthritis: comparison of efficacy in animal models with human clinical data. Arthritis Rheum 42(3): 498–506 Bendele AM, Chlipala ES, Scherrer J, Frazier J, Sennello G, Rich WJ, Edwards CK, 3rd (2000) Combination benefit of treatment with the cytokine inhibitors interleukin-1 receptor antagonist and PEGylated soluble tumor necrosis factor receptor type I in animal models of rheumatoid arthritis. Arthritis Rheum 43(12): 2648–2659 Miwatashi S, Arikawa Y, Kotani E, Miyamoto M, Naruo K, Kimura H, Tanaka T, Asahi S, Ohkawa S (2005) Novel inhibitor of p38 MAP kinase as an anti-TNF-alpha drug: discovery of N-{4-[2-ethyl-4-(3-methylphenyl)-1,3-thiazol-5-yl]-2-pyridyl}benzamide (TAK-715) as a potent and orally active anti-rheumatoid arthritis agent. J Med Chem 48(19): 5966–5979 Cole P, Rabassed, X (2004) The soluble tumor necrosis factor receptor etanercept: a new


22

23

24

25

26

27

28

29

30

31

32 33 34

strategy for the treatment of autoimmune rheumatic disease. Drugs Today (Barc) 40(4): 281–324 Kamada H, Goto M, Matsuura S, Takaoka Y, Nagai H (1997) Immunopharmacological studies on collagen-induced arthritis in dark Agouti (DA) rats. Jpn J Pharmacol 74(4): 313–322 Smith RJ, Sly LM (1996) Type II collagen-induced arthritis in the diabetic-resistant biobreeding rat: inflammatory and histopathological features of joint pathology and effects of antiinflammatory and antirheumatic drugs on this chronic arthritic process. J Pharmacol Exp Ther 277(3): 1801–1813 Cuzzocrea S, Mazzon E, di Paola R, Genovese T, Muia C, Caputi AP, Salvemini D (2005) Synergistic interaction between methotrexate and a superoxide dismutase mimetic: pharmacologic and potential clinical significance. Arthritis Rheum 52(12): 3755–3760 Brahn E, Peacock DJ, Banquerigo ML (1991) Suppression of collagen-induced arthritis by combination cyclosporin A and methotrexate therapy. Arthritis Rheum 34(10): 1282–1288 Lange F, Bajtner E, Rintisch C, Nandakumar KS, Sack U, Holmdahl R (2005) Methotrexate ameliorates T cell dependent autoimmune arthritis and encephalomyelitis but not antibody induced or fibroblast induced arthritis. Ann Rheum Dis 64(4): 599–605 Nishikawa M, Myoui A, Tomita T, Takahi K, Nampei A, Yoshikawa H (2003) Prevention of the onset and progression of collagen-induced arthritis in rats by the potent p38 mitogen-activated protein kinase inhibitor FR167653. Arthritis Rheum 48(9): 2670–2681 Esser RE, Anderle SK, Chetty C, Stimpson SA, Cromartie WJ, Schwab JH (1986) Comparison of inflammatory reactions induced by intraarticular injection of bacterial cell wall polymers. Am J Pathol 122(2): 323–334 Esser RE, Stimpson SA, Cromartie WJ, Schwab JH (1985) Reactivation of streptococcal cell wall-induced arthritis by homologous and heterologous cell wall polymers. Arthritis Rheum 28(12): 1402–1411 Rioja I, Bush KA, Buckton JB, Dickson MC, Life, PF (2004) Joint cytokine quantification in two rodent arthritis models: kinetics of expression, correlation of mRNA and protein levels and response to prednisolone treatment. Clin Exp Immunol 137(1): 65–73 Stimpson SA, Schwab, JH (1989) Chronic remittent erosive arthritis induced by bacterial peptidoglycan-polysaccharide structures. In: JY Chang, AJ Lewis (eds): Pharmacological Methods in the Control of Inflammation. Alan R. Liss, New York, 381–394 Voll RE, Kalden JR (2005) Do we need new treatment that goes beyond tumor necrosis factor blockers for rheumatoid arthritis? Ann NY Acad Sci 1051: 799–810 Smolen JS, Steiner G (2003) Therapeutic strategies for rheumatoid arthritis. Nat Rev Drug Discov 2(6): 473–488 O’Dell JR (2004) Therapeutic strategies for rheumatoid arthritis. N Engl J Med 350(25): 2591–2602

27


35 36 37

38

39

40 41

42

43 44

45 46

47

48 49

50

28

Olsen NJ, Stein CM (2004) New drugs for rheumatoid arthritis. N Engl J Med 350(21): 2167–2179 Lee DM, Weinblatt ME (2001) Rheumatoid arthritis. Lancet 358(9285): 903–911 Inoue K, Fujisawa H, Motonaga A, Inoue Y, Kyoi T, Ueda F, Kimura K (1994) Antiinflammatory effects of etodolac: comparison with other non-steroidal anti-inflammatory drugs. Biol Pharm Bull 17(12): 1577–1583 Bartlett RR, Schleyerbach R (1985) Immunopharmacological profile of a novel isoxazol derivative, HWA 486, with potential antirheumatic activity – I. Disease modifying action on adjuvant arthritis of the rat. Int J Immunopharmacol 7(1): 7–18 Sakuma S, Nishigaki F, Magari K, Ogawa T, Miyata S, Ohkubo Y, Goto T (2001) FK506 is superior to methotrexate in therapeutic effects on advanced stage of rat adjuvant-induced arthritis. Inflamm Res 50(10): 509–514 Bendele A, McComb J, Gould T, McAbee T, Sennello G, Chlipala E, Guy M (1999) Animal models of arthritis: relevance to human disease. Toxicol Pathol 27(1): 134–142 Segawa Y, Tsuzuike N, Itokazu Y, Omata T, Inoue N, Aota S, Okazaki Y, Nakamura T (1999) Age-depending effects of methotrexate treatment on systemic bone turnover in experimental adjuvant arthritis. Arzneimittelforschung 49(1): 38–43 Komoriya K, Nagata I, Kunisawa K, Takeshita T, Naruchi T (1987) Anti-arthritic and immunoregulatory effects of TI-31 on collagen-induced arthritis. Jpn J Pharmacol 45(3): 389–396 Kaibara N, Hotokebuchi T, Takagishi K, Katsuki I (1983) Paradoxical effects of cyclosporin A on collagen arthritis in rats. J Exp Med 158(6): 2007–2015 Kliwinski C, Kukral D, Postelnek J, Krishnan B, Killar L, Lewin A, Nadler S, Townsend R (2005) Prophylactic administration of abatacept prevents disease and bone destruction in a rat model of collagen-induced arthritis. J Autoimmun 25(3): 165–171 Kremer JM, Cannon GW (2004) Benefit/risk of leflunomide in rheumatoid arthritis. Clin Exp Rheumatol 22(5 Suppl 35): S95–100 Ofosu-Appiah WA, Morgan K, Holt PJ (1984) Native type II collagen-induced arthritis in the rat. Studies of the humoral response to collagen at the cellular level. J Rheumatol 11(4): 432–437 Bendele AM, McComb J, Gould T, Frazier J, Chlipala E, Seely J, Kieft G, Edwards CK, 3rd (1999) Effects of PEGylated soluble tumor necrosis factor receptor type I (PEG sTNF-RI) alone and in combination with methotrexate in adjuvant arthritic rats. Clin Exp Rheumatol 17(5): 553–560 Jones RS, Ward JR (1963) Studies on adjuvant-induced polyarthritis in rats. II. Histogenesis of joint and visceral lesions. Arthritis Rheum 6: 23–35 Schett G, Middleton S, Bolon B, Stolina M, Brown H, Zhu L, Pretorius J, Zack DJ, Kostenuik P, Feige U (2005) Additive bone-protective effects of anabolic treatment when used in conjunction with RANKL and tumor necrosis factor inhibition in two rat arthritis models. Arthritis Rheum 52(5): 1604–1611 Kohashi O, Kuwata J, Umehara K, Uemura F, Takahashi T, Ozawa A (1979) Suscepti-


51

52

53

54

55 56

57

58 59

60 61

62 63 64

bility to adjuvant-induced arthritis among germfree, specific-pathogen-free, and conventional rats. Infect Immun 26(3): 791–794 van den Broek MF, van Bruggen MC, Koopman JP, Hazenberg MP, van den Berg WB (1992) Gut flora induces and maintains resistance against streptococcal cell wallinduced arthritis in F344 rats. Clin Exp Immunol 88(2): 313–317 Fathalla B, Hamada K, Vannier E, Smith D, Edwards C, 3rd, and Roubenoff R (2004) Effects of aging and cytokine blockade on inflammatory cachexia. Clin Exp Rheumatol 22(1): 85–90 Sternberg EM, Hill JM, Chrousos GP, Kamilaris T, Listwak SJ, Gold PW, Wilder RL (1989) Inflammatory mediator-induced hypothalamic-pituitary-adrenal axis activation is defective in streptococcal cell wall arthritis-susceptible Lewis rats. Proc Natl Acad Sci USA 86(7): 2374–2378 Waksman BH, Pearson CM, Sharp JT (1960) Studies of arthritis and other lesions induced in rats by injection of mycobacterial adjuvant. II. Evidence that the disease is a disseminated immunologic response to exogenous antigen. J Immunol 85: 403–417 Waksman BH (2002) Immune regulation in adjuvant disease and other arthritis models: relevance to pathogenesis of chronic arthritis. Scand J Immunol 56(1): 12–34 Whitehouse DJ, Whitehouse MW, Pearson CM (1969) Passive transfer of adjuvantinduced arthritis and allergic encephalomyelitis in rats using thoracic duct lymphocytes. Nature 224(226): 1322 Allen JB, Malone DG, Wahl SM, Calandra GB, Wilder RL (1985) Role of the thymus in streptococcal cell wall-induced arthritis and hepatic granuloma formation. Comparative studies of pathology and cell wall distribution in athymic and euthymic rats. J Clin Invest 76(3): 1042–1056 Ridge SC, Zabriske JB, Oronsky AL, Kerwar SS (1985) Streptococcal cell wall arthritis: studies with nude (athymic) inbred Lewis rats. Cell Immunol 96(1): 231–234 Santos LL, Morand EF, Hutchinson P, Boyce NW, Holdsworth SR (1997) Anti-neutrophil monoclonal antibody therapy inhibits the development of adjuvant arthritis. Clin Exp Immunol 107(2): 248–253 Brahn E, Trentham DE (1984) Effect of antithymocyte serum on collagen arthritis in rats: evidence that T cells are involved in its pathogenesis. Cell Immunol 86(2): 421–428 DeJoy SQ, Ferguson KM, Sapp TM, Zabriskie JB, Oronsky AL, Kerwar SS (1989) Streptococcal cell wall arthritis. Passive transfer of disease with a T cell line and crossreactivity of streptococcal cell wall antigens with Mycobacterium tuberculosis. J Exp Med 170(2): 369–382 Firestein GS (2005) Immunologic mechanisms in the pathogenesis of rheumatoid arthritis. J Clin Rheumatol 11(3 Suppl): S39–44 Firestein GS, Zvaifler NJ (1990) How important are T cells in chronic rheumatoid synovitis? Arthritis Rheum 33(6): 768–773 Kremer JM, Dougados M, Emery P, Durez P, Sibilia J, Shergy W, Steinfeld S, Tindall E, Becker JC, Li T et al (2005) Treatment of rheumatoid arthritis with the selective cos-

29


65

66

67

68

69

70 71

72

73

74

75

76 77

78

30

timulation modulator abatacept: twelve-month results of a phase iib, double-blind, randomized, placebo-controlled trial. Arthritis Rheum 52(8): 2263–2271 van Eden W, Thole JE, van der Zee R, Noordzij A, van Embden JD, Hensen EJ, Cohen IR (1988) Cloning of the mycobacterial epitope recognized by T lymphocytes in adjuvant arthritis. Nature 331(6152): 171–173 Prakken BJ, Roord S, Ronaghy A, Wauben M, Albani S, van Eden W (2003) Heat shock protein 60 and adjuvant arthritis: a model for T cell regulation in human arthritis. Springer Semin Immunopathol 25(1): 47–63 van Eden W, Holoshitz J, Nevo Z, Frenkel A, Klajman A, Cohen IR (1985) Arthritis induced by a T-lymphocyte clone that responds to Mycobacterium tuberculosis and to cartilage proteoglycans. Proc Natl Acad Sci USA 82(15): 5117–5120 Billingham ME, Carney S, Butler R, Colston MJ (1990) A mycobacterial 65-kD heat shock protein induces antigen-specific suppression of adjuvant arthritis, but is not itself arthritogenic. J Exp Med 171(1): 339–344 Prakken BJ, Roord S, van Kooten PJ, Wagenaar JP, van Eden W, Albani S, Wauben MH (2002) Inhibition of adjuvant-induced arthritis by interleukin-10-driven regulatory cells induced via nasal administration of a peptide analog of an arthritis-related heat-shock protein 60 T cell epitope. Arthritis Rheum 46(7): 1937–1946 van den Broek MF (1989) Streptococcal cell wall-induced polyarthritis in the rat. Mechanisms for chronicity and regulation of susceptibility. APMIS 97(10): 861–878 Beech JT, Siew LK, Ghoraishian M, Stasiuk LM, Elson CJ, Thompson SJ (1997) CD4+ Th2 cells specific for mycobacterial 65-kilodalton heat shock protein protect against pristane-induced arthritis. J Immunol 159(8): 3692–3697 Paul AG, van Kooten PJ, van Eden W, van der Zee R (2000) Highly autoproliferative T cells specific for 60-kDa heat shock protein produce IL-4/IL-10 and IFN-gamma and are protective in adjuvant arthritis. J Immunol 165(12): 7270–7277 Wilder RL, Allen JB, Hansen C (1987) Thymus-dependent and -independent regulation of Ia antigen expression in situ by cells in the synovium of rats with streptococcal cell wall-induced arthritis. Differences in site and intensity of expression in euthymic, athymic, and cyclosporin A-treated LEW and F344 rats. J Clin Invest 79(4): 1160–1171 Stimpson SA, Esser RE, Carter PB, Sartor RB, Cromartie WJ, Schwab JH (1987) Lipopolysaccharide induces recurrence of arthritis in rat joints previously injured by peptidoglycan-polysaccharide. J Exp Med 165(6): 1688–1702 Lichtman SN, Wang J, Sartor RB, Zhang C, Bender D, Dalldorf FG, Schwab JH (1995) Reactivation of arthritis induced by small bowel bacterial overgrowth in rats: role of cytokines, bacteria, and bacterial polymers. Infect Immun 63(6): 2295–2301 Stuart JM, Townes AS, Kang AH (1982) The role of collagen autoimmunity in animal models and human diseases. J Invest Dermatol 79 Suppl 1: 121s–127s Stuart JM, Tomoda K, Yoo TJ, Townes AS, Kang AH (1983) Serum transfer of collageninduced arthritis. II. Identification and localization of autoantibody to type II collagen in donor and recipient rats. Arthritis Rheum 26(10): 1237–1244 Cremer MA, Rosloniec EF, Kang AH (1998) The cartilage collagens: a review of their


79 80

81 82

83

84

85

86

87 88

89

90

91 92

93

structure, organization, and role in the pathogenesis of experimental arthritis in animals and in human rheumatic disease. J Mol Med 76(3–4): 275–288 Chaiamnuay S, Bridges SL Jr (2005) The role of B cells and autoantibodies in rheumatoid arthritis. Pathophysiology 12(3): 203–216 Edwards JC, Szczepanski L, Szechinski J, Filipowicz-Sosnowska A, Emery P, Close DR, Stevens RM, Shaw T (2004) Efficacy of B-cell-targeted therapy with rituximab in patients with rheumatoid arthritis. N Engl J Med 350(25): 2572–2581 Cohen SB (2005) B-cell depletion for rheumatic diseases: where are we? MedGenMed 7(2): 72 Vossenaar ER, van Boekel MA, van Venrooij WJ, Lopez-Hoyoz M, Merino J, Merino R, Joosten LA (2004) Absence of citrulline-specific autoantibodies in animal models of autoimmunity. Arthritis Rheum 50(7): 2370–2372 Caulfield JP, Hein A, Dynesius-Trentham R, Trentham DE (1982) Morphologic demonstration of two stages in the development of type II collagen-induced arthritis. Lab Invest 46(3): 321–343 Harris HE, Liljestrom M, Lindroos E, Kleinau S (1997) Characterization of adhesion molecule expression in the pathogenesis of homologous collagen-induced arthritis in the DA rat. Clin Immunol Immunopathol 85(1): 40–46 Pearson CM, Wood FD (1963) Studies of arthritis and other lesions induced in rats by the injection of mycobacterial adjuvant. VII. Pathologic details of the arthritis and spondylitis. Am J Pathol 42: 73–95 Issekutz AC, Issekutz TB (1991) Quantitation and kinetics of polymorphonuclear leukocyte and lymphocyte accumulation in joints during adjuvant arthritis in the rat. Lab Invest 64(5): 656–663 Ward PA (1997) Neutrophils and adjuvant arthritis. Clin Exp Immunol 107(2): 225–226 Bush KA, Walker JS, Lee CS, Kirkham BW (2001) Cytokine expression and synovial pathology in the initiation and spontaneous resolution phases of adjuvant arthritis: interleukin-17 expression is upregulated in early disease. Clin Exp Immunol 123(3): 487–495 Szekanecz Z, Halloran MM, Volin MV, Woods JM, Strieter RM, Kenneth Haines G 3rd, Kunkel SL, Burdick MD, Koch AE (2000) Temporal expression of inflammatory cytokines and chemokines in rat adjuvant-induced arthritis. Arthritis Rheum 43(6): 1266–1277 Suda T, Takahashi N, Udagawa N, Jimi E, Gillespie MT, Martin TJ (1999) Modulation of osteoclast differentiation and function by the new members of the tumor necrosis factor receptor and ligand families. Endocr Rev 20(3): 345–357 Teitelbaum SL (2000) Bone resorption by osteoclasts. Science 289(5484): 1504–1508 Tyler JA, Benton HP (1988) Synthesis of type II collagen is decreased in cartilage cultured with interleukin 1 while the rate of intracellular degradation remains unchanged. Coll Relat Res 8(5): 393–405 Reginato AM, Sanz-Rodriguez C, Diaz A, Dharmavaram RM, Jimenez SA (1993) Tran-

31


94 95 96 97 98

99

100

101

102

103

104

105 106

32

scriptional modulation of cartilage-specific collagen gene expression by interferon gamma and tumour necrosis factor alpha in cultured human chondrocytes. Biochem J 294 (Pt 3): 761–769 Saklatvala J (1986) Tumour necrosis factor alpha stimulates resorption and inhibits synthesis of proteoglycan in cartilage. Nature 322(6079): 547–549 Goldring SR (2003) Inflammatory mediators as essential elements in bone remodeling. Calcif Tissue Int 73(2): 97–100 Goldring SR (2003) Pathogenesis of bone and cartilage destruction in rheumatoid arthritis. Rheumatology (Oxford) 42 (Suppl 2): ii11–16 Kirwan JR (2001) Links between radiological change, disability, and pathology in rheumatoid arthritis. J Rheumatol 28(4): 881–886 Kirwan J, Byron M, Watt I (2001) The relationship between soft tissue swelling, joint space narrowing and erosive damage in hand X-rays of patients with rheumatoid arthritis. Rheumatology (Oxford) 40(3): 297–301 Molenaar ET, Voskuyl AE, Dinant HJ, Bezemer PD, Boers M, Dijkmans BA (2004) Progression of radiologic damage in patients with rheumatoid arthritis in clinical remission. Arthritis Rheum 50(1): 36–42 Lipsky PE, van der Heijde DM, St Clair EW, Furst DE, Breedveld FC, Kalden JR, Smolen JS, Weisman M, Emery P, Feldmann M et al (2000) Infliximab and methotrexate in the treatment of rheumatoid arthritis. Anti-Tumor Necrosis Factor Trial in Rheumatoid Arthritis with Concomitant Therapy Study Group. N Engl J Med 343(22): 1594–1602 Smolen JS, Han C, Bala M, Maini RN, Kalden JR, van der Heijde D, Breedveld FC, Furst DE, Lipsky PE (2005) Evidence of radiographic benefit of treatment with infliximab plus methotrexate in rheumatoid arthritis patients who had no clinical improvement: a detailed subanalysis of data from the anti-tumor necrosis factor trial in rheumatoid arthritis with concomitant therapy study. Arthritis Rheum 52(4): 1020–1030 Bresnihan B, Alvaro-Gracia JM, Cobby M, Doherty M, Domljan Z, Emery P, Nuki G, Pavelka K, Rau R, Rozman B et al (1998) Treatment of rheumatoid arthritis with recombinant human interleukin-1 receptor antagonist. Arthritis Rheum 41(12): 2196–2204 Fujikawa Y, Quinn JM, Sabokbar A, McGee JO, Athanasou NA (1996) The human osteoclast precursor circulates in the monocyte fraction. Endocrinology 137(9): 4058–4060 Takayanagi H, Iizuka H, Juji T, Nakagawa T, Yamamoto A, Miyazaki T, Koshihara Y, Oda H, Nakamura K, Tanaka S (2000) Involvement of receptor activator of nuclear factor kappaB ligand/osteoclast differentiation factor in osteoclastogenesis from synoviocytes in rheumatoid arthritis. Arthritis Rheum 43(2): 259–269 Billingham ME (1983) Models of arthritis and the search for anti-arthritic drugs. Pharmacol Ther 21(3): 389–428 Schmidt-Weber CB, Pohlers D, Siegling A, Schadlich H, Buchner E, Volk HD, PalomboKinne E, Emmrich F, Kinne RW (1999) Cytokine gene activation in synovial membrane, regional lymph nodes, and spleen during the course of rat adjuvant arthritis. Cell Immunol 195(1): 53–65


107 Smith-Oliver T, Noel LS, Stimpson SS, Yarnall DP, Connolly KM (1993) Elevated levels of TNF in the joints of adjuvant arthritic rats. Cytokine 5(4): 298–304 108 Magari K, Miyata S, Ohkubo Y, Mutoh S (2004) Inflammatory cytokine levels in paw tissues during development of rat collagen-induced arthritis: effect of FK506, an inhibitor of T cell activation. Inflamm Res 53(9): 469–474 109 Seed MP, Gardner CR (2005) The modulation of intra-articular inflammation, cartilage matrix and bone loss in mono-articular arthritis induced by heat-killed Myobacterium tuberculosis. Inflammopharmacology 12(5): 551–567 110 Larsson E, Erlandsson Harris H, Larsson A, Mansson B, Saxne T, Klareskog L (2004) Corticosteroid treatment of experimental arthritis retards cartilage destruction as determined by histology and serum COMP. Rheumatology (Oxford) 43(4): 428–434 111 Larsson E, Mussener A, Heinegard D, Klareskog L, Saxne T (1997) Increased serum levels of cartilage oligomeric matrix protein and bone sialoprotein in rats with collagen arthritis. Br J Rheumatol 36(12): 1258–1261 112 Sims NA, Green JR, Glatt M, Schlict S, Martin TJ, Gillespie MT, Romas E (2004) Targeting osteoclasts with zoledronic acid prevents bone destruction in collagen-induced arthritis. Arthritis Rheum 50(7): 2338–2346 113 Badger AM, Griswold DE, Kapadia R, Blake S, Swift BA, Hoffman SJ, Stroup GB, Webb E, Rieman DJ, Gowen M et al (2000) Disease-modifying activity of SB 242235, a selective inhibitor of p38 mitogen-activated protein kinase, in rat adjuvant-induced arthritis. Arthritis Rheum 43(1): 175–183 114 Silva MD, Savinainen A, Kapadia R, Ruan J, Siebert E, Avitahl N, Mosher R, Anderson K, Jaffee B, Schopf L et al (2004) Quantitative analysis of micro-CT imaging and histopathological signatures of experimental arthritis in rats. Mol Imaging 3(4): 312–318 115 Jacobson PB, Morgan SJ, Wilcox DM, Nguyen P, Ratajczak CA, Carlson RP, Harris RR, Nuss M (1999) A new spin on an old model: in vivo evaluation of disease progression by magnetic resonance imaging with respect to standard inflammatory parameters and histopathology in the adjuvant arthritic rat. Arthritis Rheum 42(10): 2060–2073 116 Yamane I, Hagino H, Okano T, Enokida M, Yamasaki D, Teshima R (2003) Effect of minodronic acid (ONO-5920) on bone mineral density and arthritis in adult rats with collagen-induced arthritis. Arthritis Rheum 48(6): 1732–1741 117 Okazaki Y, Tsurukami H, Nishida S, Okimoto N, Aota S, Takeda S, Nakamura T (1998) Prednisolone prevents decreases in trabecular bone mass and strength by reducing bone resorption and bone formation defect in adjuvant-induced arthritic rats. Bone 23(4): 353–360 118 Okamoto A, Yamamura M, Iwahashi M, Aita T, Ueno A, Kawashima M, Yamana J, Kagawa H, Makino H (2003) Pathophysiological functions of CD30+ CD4+ T cells in rheumatoid arthritis. Acta Med Okayama 57(6): 267–277 119 Osterman T, Virtamo T, Lauren L, Kippo K, Pasanen I, Hannuniemi R, Vaananen K, Sellman R (1997) Slow-release clodronate in prevention of inflammation and bone loss associated with adjuvant arthritis. J Pharmacol Exp Ther 280(2): 1001–1007

33


120 Osterman T, Kippo K, Lauren L, Hannuniemi R, Sellman R (1995) Effect of clodronate on established collagen-induced arthritis in rats. Inflamm Res 44(6): 258–263 121 Harkness JE, Ridgway MD (1980) Chromodacryorrhea in laboratory rats (Rattus norvegicus): etiologic considerations. Lab Anim Sci 30(5): 841–844 122 Wilder RL (1988) Streptococcal cell-wall-induced arthritis in rats: an overview. Int J Tissue React 10(1): 1–5 123 Neuman RG, Wilson BD, Barkley M, Kimball ES, Weichman BM, Wood DD (1987) Inhibition of prostaglandin biosynthesis by etodolac. I. Selective activities in arthritis. Agents Actions 21(1–2): 160–166 124 Tanaka K, Shimotori T, Makino S, Aikawa Y, Inaba T, Yoshida C, Takano S (1992) Pharmacological studies of the new antiinflammatory agent 3-formylamino-7-methylsulfonylamino-6-phenoxy-4H-1-benzopyran-4-o ne. 1st communication: antiinflammatory, analgesic and other related properties. Arzneimittelforschung 42(7): 935–944 125 Calhoun W, Gilman SC, Datko LJ, Copenhaver TW, Carlson RP (1992) Interaction studies of tilomisole, aspirin, and naproxen in acute and chronic inflammation with assessment of gastrointestinal irritancy in the rat. Agents Actions 36(1–2): 99–106 126 Ackerman NR, Rooks WH, 2nd, Shott L, Genant H, Maloney P, West E (1979) Effects of naproxen on connective tissue changes in the adjuvant arthritic rat. Arthritis Rheum 22(12): 1365–1374 127 Carlson RP, Hartman DA, Tomchek LA, Walter TL, Lugay JR, Calhoun W, Sehgal SN, Chang JY (1993) Rapamycin, a potential disease-modifying antiarthritic drug. J Pharmacol Exp Ther 266(2): 1125–1138 128 Magari K, Miyata S, Nishigaki F, Ohkubo Y, Mutoh S (2004) Comparison of antiarthritic properties of leflunomide with methotrexate and FK506: effect on T cell activation-induced inflammatory cytokine production in vitro and rat adjuvant-induced arthritis. Inflamm Res 53(10): 544–550

34

Murine collagen induced arthritis Leo A. B. Joosten and Wim B. van den Berg Rheumatology Research and Advanced Therapeutics, Department of Rheumatology, Radboud University Nijmegen Medical Center, PO Box 9101, 6500 HB Nijmegen, The Netherlands

Introduction A major research goal in the field of arthritis is to unravel the pathogenesis of chronic arthritis and the concomitant joint destruction. A second, more practical goal is to define targeted therapies, selectively inhibiting the progression of destructive arthritis, yet leaving host defense mechanisms virtually intact. Although animal models are not ideal in terms of precise mimicry of human arthritic disease, they do reflect key aspects of their human counterparts and offer a useful approach to understand arthritic processes and to improve therapeutic treatment. Rheumatoid arthritis (RA) is characterized by chronic inflammation in the joints and progressive destruction of cartilage and bone. Histopathological features include immune complexes in the articular cartilage layers and variable amounts of macrophages and T cells in the synovium, accompanied by fibrosis and synovial hyperplasia. The disease is often considered as an autoimmune process, the articular cartilage being an intriguing component, since it is the victim but also a likely trigger of the disease. Arguments for this are based on the observation that destructive forms of RA tend to decline when the cartilage is fully destroyed. Moreover, total joint replacement often results in a complete remission of arthritis in that particular joint, without the need of concomitant synovectomy. This is compatible with cartilage components being joint specific autoantigens or cartilage tissue functioning as an avascular reservoir, retaining yet unidentified arthritogenic triggers. Models have been developed that have proved the arthritogenic potential of cartilage autoantigens, such as collagen type II (CII) and proteoglycan [1–4]. More recently [5–7], arthritogenic potential has been demonstrated for novel cartilage components such as collagen types IX and XI, cartilagederived oligomeric protein (COMP) and hyaline cartilage glycoprotein 39 (HC gp-39). These models all elude to the same principle: arthritis due to the loss of tolerance against a cartilage-specific autoantigen. Based on the hypothesis that RA is initiated by cross-reactivity of T cells to bacterial fragments and cartilage components, models of experimental arthritis were generated using bacteria as antiIn Vivo Models of Inflammation, Vol. I, edited by Christopher S. Stevenson, Lisa A. Marshall and Douglas W. Morgan © 2006 Birkhäuser Verlag Basel/Switzerland

35

Leo A. B. Joosten and Wim B. van den Berg

gens. Adjuvant arthritis and streptococcal cell wall-induced arthritis are the most common models of arthritis, using Mycobacterium tuberculosis and group A streptococci, respectively [8–11]. The present chapter is confined to detailed discussion of key events in arthritis induced with CII. Collagen-induced arthritis (CIA) is a widely accepted arthritis model, based on T cell and antibody-mediated autoimmune reactivity against cartilage CII. The model is characterized by severe cartilage and bone erosions. Induction has been demonstrated in various strains of rats and mice, susceptibility showing tight genetic restriction. More recently, CIA has been induced in non-human primates as well. The model of CIA is highly suited to analyze principles of autoimmune disease expression and antigen-specific immunosuppression. Moreover, it can be used to study mechanism and mediators involved in autoimmune cartilage and bone destruction. The following sections mainly deal with features of CIA in the mouse.

Induction of CIA The model of CIA was first described in 1977 by Trentham and colleagues [1], as a coincidental finding in protocols to induce autoantibodies to purified collagen preparations. The initial observations indicated that arthritis was confined to sensitization with native CII; denatured CII or native CI not showing arthritogenicity. The crucial element in this arthritis is the induction of immunity to foreign CII, subsequently cross-reacting with homologous CII. Plain immunization with homologous instead of heterologous CII can also be used, but then much stronger immunization regimens are needed to override natural tolerance. In Lewis rats, a single immunization with CII in complete Freund's adjuvant (CFA) at the base of the tail is sufficient to get full-blown expression of a polyarthritis within 14 days. In mice, the disease expression is more gradual, starting after 3–4 weeks in some animals, whereas a 10% incidence commonly takes 8–10 weeks. Both chicken and bovine CII preparations are proper heterologous antigens for inducing CIA in mice. Susceptible mouse strains include DBA/1j and B10RIII mice, which have the H-2q and H-2r haplotype, respectively. The dominant epitopes of the CII molecule differ between the DBA/1j and B10RIII mice, consistent with the different haplotype [7, 8]. Male mice show higher susceptibility as compared to female mice. In general, we use DBA/1j male mice and bovine CII to induce this model. Bulk quantities of bovine CII can be isolated from articular cartilage slices, taken from a knee joint of 1–2-year-old cows, according to Miller and Rhodes [9]. In brief, proteoglycans are extracted from the cartilage with 4 M guanidinium chloride in a neutral 0.05 M Tris buffer, for 24 h at room temperature. After washing, the cartilage is digested with pepsin (1 mg/ml in 0.1 M acetic acid) for 48 h at room tem-

36

Murine collagen induced arthritis

perature. The suspension is then centrifuged at 1000 g and the supernatant is adjusted to pH 7.4 with 2 M NaOH. Bulk protein is precipitated with addition of solid NaCl, reaching a final concentration of 20% NaCl, and equilibration for 2 days at 4°C. After centrifugation at 27 000 g for 30 min, the pellet is resolved in 0.5 M acetic acid and dialyzed against this solution overnight at room temperature. The collagen is then redissolved, and remaining material is removed by centrifugation (10 min at 2000 g). The collagen is then selectively precipitated at a final concentration of 5% NaCl, overnight at 4°C. The precipitate is spun down, redissolved and reprecipitated with 5% NaCl twice more to obtain a purer collagen preparation. Final dialyzation is done against 0.05 M acetic acid and the preparation can be stored as such at –20°C, or lyophilized. Pure, native CII preparations are poorly soluble in water, which provides a first, simple check on quality. Purity can be analyzed by gel electrophoresis. In addition, newly prepared CII batches are first screened in an ELISA, and compared with former arthritogenic CII batches and the use of a standard set of anti-CII antibodies, obtained from a pool of arthritic mice. To obtain a defined solution for immunization, CII is slowly dissolved in 0.05 M acetic acid overnight at 4°C (concentration 2 mg/ml) and then emulsified with an equal volume of CFA, containing 2 mg/ml Mycobacterium tuberculosis, strain H37Ra. Mice are then immunized intradermally at the base of the tail with 100 µl emulsion (100 µg CII). On day 21 a booster injection is given with 100 µg CII in 100 µl PBS, administered intraperitoneally. Onset of arthritis starts around days 25–28, often first affecting some digits of hind and fore paws, then spreading to multiple sites in the paw, including the ankle compartments (Fig. 1). When not heavily boosted the onset may be rather gradual, not even reaching a 100% incidence at 8 weeks and with limited numbers of joints affected (Fig. 2A). The model is a mixture of an immune complex disease and a delayed-type hypersensitivity reaction in the joint. Although anti-CII antibodies alone are able to induce arthritis after passive transfer, high concentrations are needed, in particular of complement-activating subclasses, recognizing multiple epitopes, and even then at best a transient arthritis occurs [10]. Passive transfer with bulk T cells or T cell clones has also been shown to yield poor disease expression [11]. Antibodies are probably needed to bind to the cartilage surface, promoting there the further release of collagen epitopes upon complement fixation and the attraction of leukocytes. Attachment of granulocytes to the cartilage surface is a crucial element of CIA. Subsequent influx of anti-CII-specific Th1 cells further drives the arthritic process. Of interest, the cytokine pattern in the lymphoid organs shows a dominant Th1 pattern after immunization with CII in CFA [12]. Susceptibility to CIA is enhanced when the amount of M. tuberculosis in the CFA preparation is enhanced. Moreover, severe arthritis can also be induced upon immunization with CII in incomplete Freund's adjuvant (IFA), provided that the mice are then treated with recombinant IL-12 dur-

37


Figure 1 Macroscopic appearance of murine CIA, ranging from normal (A) to one affected toe (B) and full expression in the whole paw (C).

ing the immunization period, which strongly promotes a Th1 response. Bacterial preparations are potent inducers of IL-12. Remarkably, high doses of IL-12 given during standard immunization with CII in CFA were shown to suppress the CIA, associated with a marked reduction in CII-specific antibodies [13]. Although there is no doubt that Th1 reactivity is needed, the critical importance of high levels of anti-CII antibodies is further underlined in studies in susceptible and resistant mice strains. Additional IL-12 treatment during immunization with CII in CFA was shown to enhance CII-specific Th1 responses in C57BL/6 and B10.Q mice, but this protocol still failed to induce arthritis in these mice. Analysis of anti-CII antibody titers revealed that the levels remained markedly lower in these strains as compared to those found in susceptible DBA mice [14].

38


Figure 2 (A) Incidence of CIA, as scored in one selected paw or in all paws. Note that the incidence in the right hind paw is still less than 50% after 50 days. Reflects a group of 20 mice. (B) Accelerated expression of CIA and more severe arthritis after i.p. injection of Zymosan at day 26. Reflects groups of 10 mice.

Expression of arthritis Apart from the generation of adequate levels of complement fixing anti-CII antibodies and the presence of anti-CII-specific Th1 cells, it is clear that expression of

39


autoimmune arthritis depends on local conditions in joint tissues. CIA shows a higher incidence in male as compared to female mice, whereas other models such as antigen- or Zymosan-induced arthritis show the opposite sex preponderance. Apart from the autoimmune character, the elicitation of the joint inflammation by direct injection into the knee joint is a major difference between these models and CIA. In CIA no local insults are given and arthritis develops “spontaneously”. The male preponderance of CIA might be linked to the impact of hormones on this arthritic process, but it is tempting to suggest that the consistent fighting of male mice also makes a major contribution, causing microtrauma in joint tissues, which are crucial in triggering of onset of arthritis. In line with this, it is also our experience that disease incidence is generally higher when the mice are housed in large groups instead of small groups. Threatening autoimmune reactivity is generated by the immunization protocol, but precipitation of the autoimmune process in the joint is facilitated by nonspecific inflammation at such sites or systemic generation of proinflammatory mediators. It has long been recognized that systemic administration of IL-1, shortly before onset of the disease, markedly accelerates CIA expression [15]. This seems related to activation of endothelium, facilitating influx of inflammatory cells. Moreover, IL1 is a potent cartilage destruction mediator, causing loss of cartilage proteoglycans and thereby denuding the autoimmune target in CIA, the CII in the articular cartilage. In addition, it has been shown that local injection of TNF-_ or TGF-` potentiates CIA expression in the injected joint [16, 17]. TNF-_ is a pivotal proinflammatory cytokine in arthritis and an inducer of IL-1. TGF-`, although having immunosuppressive potential, is a potent chemoattractant. All of this fits with unmasking of dormant autoimmune reactivity by nonspecific attraction of inflammatory cells to the joints, including CII-reactive T cells, and amplification of the process by inflammation-mediated exposure of autoimmune epitopes. A single injection of LPS provides an elegant alternative for the acceleration of CIA by systemic administration of recombinant IL-1 [18]. In our standard protocol we give a booster immunization with 100 µg CII at day 21 and a single i.p. injection of 10–40 µg LPS around day 28. This not only greatly enhances the severity of the arthritis, but also synchronizes the expression in a group of mice and enlarges the number of affected joints in one animal. This is useful since spontaneous expression of CIA after plain immunization often affects only a limited number of joints, such as one or two toes, complicating grading and histological analysis. A critical prerequisite for the accelerated expression still remains the proper immunization with CII. In animals showing poor immunity to CII, the CIA can not be accelerated, either with a single or with repeated LPS challenge. The mechanism behind CIA acceleration with LPS can be linked to the generation of TNF-_ and IL-1, as well as marked IL-12 production. LPS-induced acceleration can be blocked with antibodies against IL-12, and acceleration can be induced with systemic administration of recombinant IL-12 [19].

40


Table 1 - Factors influencing expression of murine CIA -

Isotype of the anti-CII antibodies Degree of Th1 anti-CII reactivity Acceleration with the cytokines IL-1, TNF-_ and TGF-` Control by the suppressive cytokines IL-4 and IL-10 (TGF-`) Nonspecific trauma in joint tissues promotes expression

Apart from LPS-induced cytokine generation, it is conceivable that the LPSinduced elicitation of IL-12 promotes Th1 generation. When applied shortly after the booster injection with CII, it will influence the critical process of generation of cross-reactive T cell activity from heterologous CII to the homologous CII of the mouse, including the process of epitope spreading. Although CIA can be induced by immunization with a small CII fragment, containing the dominant epitope, the arthritis is less severe as compared to the one induced with the whole CII molecule, suggesting multiple epitope involvement in classic CIA [9]. In line with the notion that any inflammatory stimulus, generating proinflammatory mediators such as TNF-_ or IL-1, will accelerate CIA expression (Tab. 1), we have demonstrated that a systemic injection with Zymosan (yeast particles) highly accelerates CIA in DBA/1j mice [20]. Consistent enhancement of CIA incidence and severity was seen with a dose of 3 mg Zymosan, injected i.p. This injection induces a marked peritonitis and onset of accelerated CIA expression was just noted after a few days. With increasing dosages of Zymosan, the arthritis expression could not be further enhanced (Fig. 2B). We even observed a delay in day of onset, apparently linked to a more prolonged, distracting inflammation in the peritoneal cavity.

Unilateral CIA We have developed a variant of the polyarthritic Zymosan-induced CIA, by local injection of Zymosan in one knee joint [20]. This injection is given around day 25 after the first immunization with CII at day 0 and boosting at day 21. Upon local injection in nonimmunized DBA mice, Zymosan induces a transient arthritis, with reversible cartilage proteoglycan depletion. When injected in CII-immunized DBA mice, a dose of 60 µg Zymosan is sufficient to accelerate expression of CIA in that knee joint, as reflected by the characteristic, aggressive cartilage destruction and prolonged joint inflammation. When a higher dose is injected, 180 µg Zymosan, the expression of CIA is not restricted to the injected joint, but also extends to the ipsilateral ankle joint, whereas the expression in the other paws was not enhanced. With the 180 µg dose of Zymosan the accelerated and synchronized expression in

41


the ankle reached an incidence of 90%, whereas the incidence dropped to 10–50% with lower Zymosan dosages. Anti-TNF-_ treatment markedly reduced this spreading to the ipsilateral paw, whereas anti-IL-1 treatment fully prevented the expression in that ankle joint, making it likely that the Zymosan-induced expression is related to TNF-_ and IL-1, produced locally in the knee joint and diffusing to the ankle. To obtain optimal advantage of a fully synchronized and localized expression model of CIA, it is essential that the initial immunization and boosting with CII is not too optimal, creating already a considerable number of mice with affected ankles around day 25. In daily practice, when aiming for this unilateral model, we apply normal amounts of M. tuberculosis (1 mg) in the initial immunization, give no boosting with LPS and perform the CII boosting at day 21 in saline and not in Freund's adjuvant. Finally, we do a prescreening of the mice at day 25, discarding all animals having any sign of arthritis from the experiment and performing the local Zymosan acceleration in this negative-selection group. This approach highly reduces the variation normally encountered in the spontaneous polyarthritic CIA.

CIA in C57BL/6 mice It was shown previously that susceptibility of CIA was linked to the H-2q haplotype (DBA-1) and that H-2b (C57BL/6) mice were less sensitive for developing CIA. Recently, several studies have demonstrated that it is possible to induce CIA in C57BL/6 mice, although the disease incidence is lower than in the DBA-1 mice [21, 22]. C57BL/6 mice are immunized with 100 µg chick CII in enriched CFA (5 mg/ml M. tuberculosis) at the base of the tail at days 0 and 21. The first signs of CIA can be seen at day 30, and about 60–70% of the animals will get CIA at 40–50 days after the first immunization. The clinical and histological appearance of CIA in the C57BL/6 mice resembles the CIA in the DBA-1 mice. One of the great benefit of CIA in C57BL/6 mice is the applicability of this arthritis model in gene-deficient mice, since the most gene-deficient mice are on the C57BL/6 background.

Passive transfer model of CIA There is a growing interest in the use of passive immune complex models of RA, along with the availability of transgenic animals to unravel crucial pathways of inflammation and tissue destruction. The advantage of passive immune complex models is the applicability in several mouse strains. The first study was performed by transfer of whole serum or immunoglobulins concentrates isolated from arthritic DBA-1 mice to naïve DBA-1 mice [23]. Histopathologically, this passively transferred arthritis resembled the early disease of immunized DBA-1 mice. Several years later Terato et al. [24] described a passive transfer model using monoclonal anti-

42


bodies directed against CII. The monoclonal antibodies against CII were developed from DBA/1 mice immunized with chick CII. Several antibodies were reactive with CB11, previously identified as containing a major immunogenic and arthritogenic epitope, which cross-reacted strongly with mouse CII. Individual antibodies were able to induce mild lesions consisting of minimal synovial proliferation but not overt arthritis. However, a combination of four antibodies induced severe arthritis with marked destruction of articular cartilage. Arthritis developed within 48–72 h after injection of the antibodies and persisted for the duration of the observation period of 3 weeks. Sets of antibody pairs are commercially available at the moment. In combination with LPS, the amount of antibodies needed for the passive transfer model can be reduced. Although this arthritis model can be induced in all mouse strains, DBA-1 is the most sensitive strain.

Arthritis score and histopathology The course of CIA is routinely scored by macroscopic analysis of arthritic signs in peripheral joints. In general, mice are examined every other day, from day 25, to get an impression of the course of the disease. Clinical severity of arthritis is graded on a scale of 0–2 for each paw, according to changes in redness and swelling. At late stages of the disease ankylosis of the ankles can be included. The macroscopic score is expressed as a selective value in one paw or as a cumulative value for all paws, with a maximum of 8. Histologically, the inflammation is characterized by a florid exudate in the joint space, containing numerous amounts of granulocytes, and a progressive destruction of the articular cartilage. Erosion of bone is pronounced, but periosteal new bone formation is also seen. Bone marrow is affected, but markedly less as compared to the polyarthritic adjuvant disease. Apart from the high number of granulocytes in the joint cavity, the synovial tissue contains large numbers of macrophages and lymphocytes. However, in this compartment granulocytes are also prominent in the first 2 weeks after onset. The most characteristic feature of CIA is the aggressive attack of the inflammatory process at the articular cartilage (Fig. 3A–F). In this model, heavy adherence of granulocytes at the cartilage surface is a common finding. Moreover, cartilage damage is not limited to loss of proteoglycans from the matrix, but in a short period of time roughening and deep erosions of the surface are consistently observed, further facilitating attachment of granulocytes at these sites. In that sense, granulocytes probably play an active role in the cartilage destruction, linked to sticking to anti-CII immune complexes in the surface layers. It has long been recognized that granulocytes contain high amounts of elastase and cathepsin G, which are potent mediators of cartilage proteoglycan depletion in co-cultures of cartilage and activated granulocytes in vitro. When such cultures are done in the presence of full serum or synovial fluid, the destruction is limited, due to large

43


Figure 3 Characteristic histopathology of murine CIA in the area of the patella and opposite femur of the knee. Homogeneous Saffranin O staining of proteoglycans in cartilage surface layers of a naïve mouse (A). Depletion of proteoglycan in the superficial cartilage but still intact surface in arthritic knee joint (B) and marked depletion as well as severe erosion in late stage CIA (C). Focal bone destruction in the femur between the growth plate and the cartilage layer. Note the invasive synovium that erodes the bone from the outside (D, E). Severe cartilage erosion but also bone erosion underneath by ingrowth of granulation tissue in CIA (F).

44


amounts of high molecular weight enzyme inhibitors in these fluids. However, when the granulocytes are pelleted on the cartilage surface, heavy destruction is again noted, implying direct extrusion of enzymes in the matrix and escape from natural inhibitors once the cells are in full contact with the cartilage surface [25]. This process of pronounced sticking is seen in vivo in the model of CIA, provided that the anti-CII antibody levels are high. In addition, we also only noted such sticking in murine mBSA-induced arthritis when the joints are immobilized [26]. Apparently, the immune complex formation between antibodies and cationic mBSA, planted in the cartilage surface, is on its own insufficient to generate heavy sticking under normal movement conditions, but when the cells are allowed to settle under immobilized conditions, the interaction remains intact. Like the situation in CIA, this condition results in erosion of the surface and further settlement and digging in into the roughened surface by the attached granulocytes. With respect to granulocyte involvement in cartilage destruction in human RA, it is clear that immune complexes can be found in the articular cartilage surface of a large number of patients. Whether the concentration is high enough to allow for consistent granulocyte attachment is yet unclear, and considerable variation between different RA patients seems obvious. Apart from this process, cartilage erosion at the cartilage margins, linked to pannus overgrowth, is considered to make a significant contribution in RA patients. At later stages of CIA, cartilage erosion at the margins and pannus formation is a prominent feature as well (Fig. 3E, F). After a few weeks, the model often progresses to complete loss of the whole cartilage, ending up in bone to bone contact and variable degree of ankylosis. This dramatic destruction of the cartilage reflects the directed autoimmune attack at the cartilage, and the arthritis in the synovial tissue burns out in a particular joint, when the cartilage is fully destroyed. The synovium then displays a mixture of an immune infiltrate, macrophages and a pronounced fibrotic reaction. The highly destructive character of CIA is also reflected in the massive occurrence of the proteoglycan breakdown neoepitope VDIPEN throughout the cartilage. This epitope is indicative for the involvement of metalloproteases, in particular stromelysin. In contrast to the lack of such epitopes in reversible cartilage proteoglycan depletion in Zymosan arthritis, and the variable degree of these neoepitopes in murine antigen induced arthritis, only showing expression at particular sites of the cartilage displaying irreversible lesions, the expression is fast and much more pronounced in CIA (Fig. 4). Detailed studies in antigen-induced arthritis made it clear that VDIPEN expression is linked to IL-1-driven processes [27] and colocalizes with collagen breakdown neoepitopes. All of this is compatible with a role of stromelysin in activation of collagenase, and a dominant role of this process in cartilage erosion in CIA. Stromelysin is produced in a latent form after activation of synovial cells or cartilage with IL-1, suggesting that further activation by granulocyte enzymes may contribute as well, elastase being a likely candidate in this process. Recent studies with elastase inhibitors revealed efficacy in murine CIA [28].

45


Figure 4 Expression of cartilage proteoglycan breakdown neoepitope VDIPEN in the tibia-femur region. Note the absence in normal cartilage (A); local expression in murine CIA (B); fully affected cartilage in CIA (C).

Given the rapid development of the arthritic changes in this model, it is clear that consistent histological scoring of the severity of the arthritis is seriously complicated by variable days of onset of arthritis in individual mice, variation in onset between different paws or even between digits in one paw. The latter variability furthermore asks for highly standardized semi-serial sectioning of complicated joint structures of the whole paw. All of this flaws the design of proper drug studies. Attempts, discussed above, to synchronize expression in groups of mice, or perhaps even better, to precipitate the arthritis in a given joint by a local inflammatory insult, provide valuable improvements of applicability of the CIA model.

46


Table 2 - Histology of CIA: comparison of affected knee and ankle joints

Infiltrate Cartilage damage Proteoglycan depletion

Day 37* Knee joint

Ankle joint

Day 42 Knee joint

Ankle joint

1.5 ± 0.6 1.0 ± 0.6 1.9 ± 0.8

1.5 ± 0.3 0.9 ± 0.4 1.6 ± 0.4

1.3 ± 0.4 1.2 ± 0.3 2.2 ± 0.5

1.7 ± 0.4 1.1 ± 0.2 1.9 ± 0.5

*Days after the first immunization with CII. The values represent the mean ± SD of at least 20 knee or ankle joints. Histology was scored on a scale ranging from 0 to 3. Infiltrate is scored as the amount inflammatory cells in synovial tissue and joint cavity. Cartilage damage reflects surface erosions, proteoglycan depletion indicates loss of Saffranin O staining.

Although the classic macroscopic scoring of CIA is always done in paws, with additional analysis of histology of the ankle joints, it is our experience that there is a high correlation between occurrence of arthritis in the knee and the ankle. Since the standardized joint sectioning is much easier in the knee compared to the ankle, histological analysis should preferably be done in the knee. We have carefully compared the characteristic histopathology in ankles and knees and the patterns of synovitis and cartilage destruction are very similar (Tab. 2). It should be noted that this is unlike the situation in adjuvant arthritis in the rat. The latter model of arthritis shows predominant expression in the ankles, whereas knee joints are rarely affected.

Cytokine involvement In line with a major role of the cytokine TNF-_ in human RA, the onset of CIA is TNF-_ dependent. Studies with neutralizing anti-TNF-_ antibodies or soluble TNF_ receptors revealed a major suppressive effect, when treatment was started shortly before onset of CIA [29, 30]. When the arthritis is fully expressed, subsequent blocking of TNF-_ appeared only marginally effective, implying that TNF-_ is crucial in the onset but less so in propagation of arthritis. In clear contrast, blocking of IL-1 with neutralizing antibodies or IL-1 receptor antagonist (IL-1Ra) markedly reduced severity of the arthritis [31], also when the arthritis was fully established (Fig. 5A, B). Moreover, anti-IL-1 treatment markedly reduced cartilage damage. Elegant studies in IL-1`-deficient mice showed full resistance to CIA induction and the critical importance of IL-1` was also emphasized by greatly reduced CIA in IL-1converting enzyme (ICE)-deficient mice and the efficacy of ICE inhibitors in CIA in

47


Figure 5 CIA is treated with systemic administration of neutralizing antibodies against TNF-_ (A) or IL-1_,` (B). Treatment was started in the various groups at either day 28 or 32. Anti-TNF-_ is still effective shortly after onset, but not in established disease. In contrast, anti-IL-1 is highly effective, even in late arthritis.

48


normal mice. Studies in TNF-_ receptor-knockout mice revealed a lower incidence and a milder form of CIA in the absence of proper TNF-_-receptor interaction. However, once a joint was afflicted, the progression of arthritis in that joint was indistinguishable from that in wild-type mice [32], again underlining the limited role of TNF-_ in propagation and cartilage destruction. It emphasizes that TNF-_ is helpful in acceleration of arthritis expression, but that TNF-_-independent onset can occur as well. Although it is claimed in human RA that TNF-_ is driving most of the IL-1 production and that TNF-_ blocking would be sufficient to block the whole arthritic process, this is not found in CIA. Recent studies in murine streptococcal cell wall-induced arthritis also revealed major TNF-_ dependence of initial joint swelling, but IL-1` dependence of cartilage destruction [33]. This was found using neutralizing antibodies and confirmed in TNF-_- and IL-1`-knockout mice. Again, TNF-_ blocking did not sufficiently prevent IL-1 production. These findings demonstrated that anti-TNF-_ treatment in RA patients would be beneficial when the disease is in fact a chronic process, due to repeated flares, with each acute exacerbation showing strong TNF-_ dependency. Of interest, when expression of CIA is not highly stimulated by additional boosting or synchronizing injections with LPS or additional cytokines, onset of arthritis starts only in a small number of joints, with gradual involvement of additional joints with time. This creates a seemingly extended period of TNF-_ dependency of the model, which is lost upon synchronization and speedy propagation to established arthritis in most joints. An intriguing element in control of CIA expression is formed by the synovial lining cells. This layer consists of synovial fibroblasts and macrophages. When macrophages are selectively depleted from this layer by local injection of toxic liposomes and the subsequent process of engulfment of liposomes by these phagocytes and subsequent apoptotic cell death, such a joint becomes refractory to the onset of CIA [34]. Further analysis revealed that the lining macrophages are a major source of chemotactic factors, needed to direct the initiating leukocyte influx into the joint. TNF-_ and IL-1` are potent inducers of chemokine production in lining cells, and when these recombinant cytokines are injected in a lining-depleted, naive joint, they do not induce leukocyte influx. In contrast, C5a is still fully capable of attracting leukocytes in such a joint. This further establishes the TNF-_/IL-1` dependence, with an intermediate role of the lining cells, in CIA. Subsequent studies in other models revealed that immune complex arthritis was totally abolished in lining depleted joints, whereas a strong T cell driven arthritis was hardly affected. Moreover, immune complex arthritis showed strong IL-1 dependence, sharing this feature with CIA, and further emphasizing that onset of CIA is more an immune complex phenomenon than a T cell process. Apart from a pivotal role of TNF-_ and IL-1` in onset and propagation of CIA, regulation of the arthritis is exerted by the cytokines IL-4 and IL-10. These socalled modulatory cytokines have a critical impact on the arthritic process at vari-

49


ous levels, including the control of the Th1/Th2 balance, inhibition of macrophage TNF-_/IL-1` production and stimulation of chondrocytes in the articular cartilage. Expression of CIA is under the control of endogenous IL-10 [35]. High levels are found in the synovial tissue and anti-IL-10 antibodies, given shortly before expected onset, enhance incidence and severity. Anti-IL-4 antibodies were without effect, in line with the difficulty to detect significant levels of IL-4 in the synovium, but combined anti-IL-4/IL-10 treatment promoted the strongest expression of CIA (Fig. 6A). The opposite approach, i.e., treatment of CIA with systemically injected recombinant IL-4 or IL-10 revealed that IL-10 reduced the joint swelling in CIA, but more marked suppression, including reduced cartilage destruction was noted with the combination treatment with IL-4 and IL-10 (Fig. 6B, [36]). Of interest, IL-10 is a potent reducer of macrophage TNF-_ production, but IL-1 production was only suppressed with the combination of IL-10 and IL-4. Moreover, this combination also up-regulated the IL-1Ra/IL-1 balance, both in the synovium and in the cartilage. Additional studies, using gene transfer technology to overexpress IL4 locally in the knee joints of collagen-immunized DBA-1 mice showed clearly that IL-4 protects against severe cartilage and bone destruction. Moreover, overexpression of IL-4 generation reduces the local IL-1`, IL-6, IL-17 and RANKL levels [37]. The members of the IL-12 family represent another group of cytokines deserving major attention. These cytokines, including IL-12 and IL-23 originate from macrophages and are produced after activation with bacterial components. IL-12 is a potent inducer of IFN-a and promotes Th1 generation and propagation. As stated above, IL-12 when given at the expected onset of CIA, greatly enhances incidence and severity and systemic anti-IL-12 treatment prevented LPS-accelerated CIA expression. This in line with the reduced incidence and severity of CIA in IL12-deficient mice [38]. DBA/1 mice primed with CII in IFA treated with IL-12 developed significantly higher incidence and more severe disease compared with controls. These were elevated further by combination treatment with IL-12 and IL18. The IL-12/IL-18 treatment led to markedly enhanced synovial hyperplasia, cellular infiltration, and cartilage erosion compared with controls [39]. In contrast, when anti-IL-12 was applied in the established phase of CIA, it appeared poorly suppressive and upon interruption of anti-IL-12 treatment we noted a marked exacerbation. Moreover, late treatment with recombinant IL-12 suppressed instead of enhanced the arthritis, prolonged IL-12 treatment markedly enhanced IL-10 levels and the suppressive effect of IL-12 could be abrogated with anti-IL-10 [19]. This suggests a dual role of IL-12 in early and late disease. The potent induction of IL10 reflects an intriguing feedback pathway to control for excessive and prolonged Th1 responses, but seriously hampers therapeutic targeting of IL-12 in autoimmune arthritis. It has been suggested that the inflammatory cytokine IL-15 plays an important role in the development of several autoimmune diseases, including RA. IL-15 is

50


Figure 6 Treatment of CIA with systemic neutralizing antibodies against IL-4 and/or IL-10 (A) or recombinant proteins IL-4 and/or IL-10 (B). Antibodies given at days 28, 32 and 36. Recombinant cytokines given daily, from day 34. For experiment in (A), mice were selected at day 28, having no signs of arthritis. In (B), groups of mice are depicted which were challenged with LPS at day 28, to obtain high arthritis expression in the control mice. Spontaneous CIA expression is under the control of endogenous IL-10. Suppression of established arthritis is still possible with additional IL-4/IL-10.

51


derived from several cell types, including macrophages and fibroblasts. IL-15 induces T cell chemotaxis and activation, together with B cell maturation. Furthermore, it enhances NK cell cytotoxicity and cytokine production, activates neutrophils and modifies monokine secretion [40, 41]. IL-15 mediates several diverse effects at multiple stages of the immune response. Interestingly, administration of soluble IL-15 receptor a-chain prevents the onset of CIA, indicating a role of IL-15 in antigen-induced immunopathology [42]. In addition, it was elegantly shown that targeting of IL-15 receptor-bearing cells with an antagonist mutant IL-15/Fc protein prevents disease development and progression in murine CIA [43]. These data indicate that IL-15 may be an interesting target in autoimmune disease, like RA. The first clinical studies in RA that neutralized IL-15 revealed promising results. IL-17A is a pro-inflammatory cytokine, produced by activated CD4+ memory T cells, and it induces production of other pro-inflammatory cytokines by stromal cells [44]. It has been described in RA synovium and shows similar cellular responses to IL-1 [45]. IL-17A belongs to a family of proteins, including IL-17B, IL-17C, IL-17E and IL-17F. The trigger for IL-17 has not been fully identified; however, IL23 promotes the production of IL-17 and a strong correlation between IL-15 and IL-17 levels in synovial fluid has been observed. IL-17 induces NF-gB activation and IL-1, IL-6, IL-8, G-CSF and TNF-_ production in both fibroblasts and macrophages [46]. Interestingly, synergistic effects of IL-17 and IL-1 as well as TNF-_ were reported. In addition, IL-17 was associated with cartilage destruction and inhibition of chondrocyte proteoglycan synthesis due to increased catabolic enzymes and NO production [47]. Furthermore, IL-17 was shown to be a potent stimulator of osteoclastic bone resorption by enhanced prostaglandin E2 synthesis. Recently, it was nicely demonstrated that blocking of endogenous IL-17 in experimental arthritis models, such as CIA, using neutralizing antibodies against IL-17 prevented bone destruction [48]. In line with these observations, overexpression of IL-17 during onset of CIA clearly aggravated the joint destruction through loss of the receptor activator of NF-gB ligand/osteoprotegerin (OPG) balance [49]. It would be of high importance to perform blocking studies with anti-IL-17 in experimental arthritis models, such as CIA to unravel distinct IL-17 activity or its synergy with IL-1, IL15 or TNF-_. A novel factor for osteoclast differentiation, receptor activator of nuclear factor B ligand (RANKL), has been identified [50]. This TNF-related cytokine (also called OPG ligand, osteoclast differentiation factor, and TRANCE) is an essential factor for osteoclast differentiation and activation. RANKL-deficient mice have a complete absence of osteoclasts and exhibit osteopetrosis. In addition, RANKL is involved in the interaction of T cells and dendritic cells, and plays a role in immune cell differentiation [51]. OPG is a naturally occurring decoy receptor for RANKL. When bound to RANKL, OPG prevents the binding of RANKL to RANK, and thus inhibits the biological activity of RANKL. The relative local expression levels of RANKL and OPG (often represented as the RANKL/OPG ratio), is instrumental in

52


determining the degree of osteoclast-mediated bone resorption. RANKL binds to a cell-associated TNF receptor-related protein called receptor activator of nuclear factor B (RANK). This type I transmembrane protein is expressed on many cell types, including osteoclasts and osteoclast precursor cells, certain B and T cells, and dendritic cells. The interaction between RANKL and its receptor, RANK, has been shown to be critical in osteoclastogenesis and bone resorption. RANKL and OPG are important positive and negative regulators of osteoclastogenesis and bone resorption. Treatment of murine CIA or rat adjuvant arthritis with OPG reduces the number of osteoclasts and prevents bone erosion [52, 53]. IL-18 is a cytokine originally identified as IFN-a-inducing factor (IGIF), it is a member of the IL-1 family of proteins [54]. As IL-1`, IL-18 is produced as an inactive precursor and it is cleaved by ICE to the biologically active form. IL-18 acts synergistically with IL-12, IL-2 and antigens to induce the production of IFN-a. The crucial role of IL-18 in IFN-a synthesis was demonstrated in IL-18-deficient mice, where the IFN-a production was markedly reduced after injection of endotoxin, despite of normal IL-12 production in these animals [55]. IL-18 is produced by human articular chondrocytes, and it induces pro-inflammatory cytokines and catabolic factors like NO, cyclooxygenase and stromelysin. Studies in IL-18-deficient mice or blocking studies with anti-IL-18 demonstrated the proinflammatory role of IL-18 in arthritis [56, 57]. Recently, IL-18 binding protein (IL-18BP) was isolated and cloned. This protein blocks the endotoxin-induced IFN-a production in mice and belongs to a member of novel soluble receptors [58]. Since IL-18 is an early promotor of Th1 cells, IL-18BP probably plays a crucial role in the regulation of immune response. Studies in models of arthritis showed the potential therapeutic value of IL-18BP for treatment of RA [59]. IL-23 is a new member of the IL-12 family of regulatory cytokines produced by activated macrophages and dendritic cells. IL-23 is a heterodimeric cytokine composed of a p19 subunit and the p40 subunit of IL-12 [60]. IL-23 affects memory T cell and inflammatory macrophage function through engagement of a novel receptor (IL-23R) on these cells. Using gene-targeted mice lacking only IL-12 (p35–/–) or IL-23 (p19–/–), it was demonstrated that the specific absence of IL-23 is protective, whereas loss of IL-12 exacerbates CIA. IL-23 gene-targeted mice do not develop clinical signs of disease and are completely resistant to the development of joint and bone pathology. Resistance correlates with an absence of IL-17-producing CD4+ T cells despite normal induction of collagen-specific, IFN-a-producing T helper 1 cells. In contrast, IL-12-deficient p35–/– mice developed more IL-17-producing CD4+ T cells, as well as elevated mRNA expression of proinflammatory TNF-_, IL-1`, IL6, and IL-17 in affected tissues of diseased mice. These data indicate that IL-23 is an essential promoter of end-stage joint autoimmune inflammation, whereas IL-12 paradoxically mediates protection from autoimmune inflammation [61]. Figure 7 shows a schematic presentation of the cytokines involved in pathways of synovitis and concomitant cartilage and bone destruction.

53


Figure 7 Schematic presentation of pathways of synovitis and concomitant cartilage and bone destruction. Note the amplifying elements through T cell activation and generation of autoantibodies. The latter will trigger macrophages after immune complex (IC) formation, through Fca receptors. T cell-derived IL-17 contributes to bone destruction via induction of RANKL in co-operation with IL-1 and TNF-_. Ag, antigen; APC, antigen-presenting cell; Ch, chondrocyte; Fibro, fibroblast; IFN, IFN-a.

Applicability of the model The model of CIA establishes that an autoimmune reaction to a cartilage component can lead to a chronic, destructive polyarthritis. Although it is far from accepted that CII is a crucial antigen in human RA, the findings in the model may exemplify common principles in arthritis directed against cartilage autoantigens. The model is highly suitable and widely used to try to understand the immunoregulation in autoimmune arthritis and to identify ways to induce tolerance using peptide fragments, or to selectively target the T cell receptors involved in collagen epitope recog-

54


nition. Detailed discussion of these topics goes beyond the scope of the present chapter and a recent review of Myers et al. [9] as well as the chapter of Griffiths et al in this volume is advised for further reading. Apart from the immunoregulation, the model is highly suitable to try to understand the complex cytokine interplay in onset and propagation of arthritis, and to identify therapies aimed at prevention of cartilage destruction. Examples of cytokine involvement are already discussed above. As mentioned before, the onset of CIA is an immune complex phenomenon. This stage shows high sensitivity to NSAIDs and in fact, all therapies which will interfere with the initiating leukocyte influx will show efficacy. The more interesting part of the model is the established phase of the arthritis and the ongoing destruction of the articular cartilage. When the efficacy of drugs at onset of arthritis is investigated, it should be realized that this stage is rather stress sensitive. Daily treatment by i.p. or oral injection may have a large impact at that stage, and handling of mice should be done with great subtlety. We have often noted a significant suppression of arthritis onset with prophylactic daily vehicle treatment, whereas this effect was absent at later stages, when the arthritis is fully established. Treatment that is started after onset of arthritis has the further advantage that grouping of the mice can be done by weighted randomization, creating a similar mean index of arthritis at the start of the various control and treatment groups. When randomization has to be done before onset, higher variation between groups is unavoidable and in general, the use of at least 10 mice per experimental group in such protocols is warranted. To illustrate the efficacy of some drugs in murine CIA a few examples are discussed in more detail. The onset of CIA is sensitive to treatment with indomethacin. A dose of 1 mg/kg significantly suppressed the macroscopic signs of arthritis. Intriguingly, when low dosages of indomethacin are used, sensitivity is lost. However, when treatment with indomethacin is combined with a leukotriene synthesis inhibitor, marked synergistic suppression was observed (Fig. 8A). This clearly illustrates that both prostaglandins and leukotrienes are of importance at arthritis onset. The role of leukotrienes was also nicely illustrated in the poor induction of CIA in lipoxygenase-deficient mice [62]. Of interest, cartilage destruction was also markedly reduced with the combination treatment (Tab. 3). It suggests that NSAIDs with a combined profile would be the better anti-arthritic drug. More recent interest focused on the dominant COX-1 and/or COX-2 inhibitory pattern of NSAIDs, further pinpointing the profiling of suitable NSAIDs. A second example of therapeutic approaches in this model of arthritis is provided by the demonstration of synergy between steroids and IL-10. Steroids are potent suppressors of arthritis, but their clinical application is seriously hampered by side effects such as osteoporosis. It would be desirable to find ways to combine drugs at lower, nontoxic concentrations, yet retaining the beneficial effects. We found that prednisolone treatment suppressed CIA at a dose of 1–5 mg/kg per day, but dosages of 0.05 or 0.1 mg/kg were without effect. Interestingly, daily IL-10 treatment at a

55


Figure 8 (A) CIA mice received daily oral treatment with indomethacin and/or a leukotriene synthesis inhibitor (Bay W 5676), from day 28 for 14 consecutive days. The data represent groups of 10 mice. Note the marked synergy. (B) CIA mice received daily i.p. treatment with prednisolone and/or murine IL-10. Note the synergy between low-dose steroid and IL-10.

56


Table 3 - Histology of CIA after treatment with either COI, LSI or combination of COI/LSI Treatment

Dose

Infiltrate

Vehicle COI COI LSI COI/LSI

– 1 mg/kg 0.1 mg/kg 200 mg/kg 0.1/200

1.3 0.7 1.4 1.5 0.6

± ± ± ± ±

Cartilage damage

0.7 0.6 0.9 1.0 0.4

1.5 0.8 1.6 1.4 1.0

± ± ± ± ±

Proteoglycan depletion

0.7 0.5 0.6 0.8 0.3

2.0 1.5 2.2 2.1 1.2

± ± ± ± ±

0.7 0.8 0.9 0.5 0.8

Treatment of arthritic mice was started at day 28 after immunization with CII. Mice were injected i.p. twice a day with cyclooxygenase inhibitor (COI) indomethacin or leukotriene synthase inhibitor (LSI) Bay W 5676, or the combination for 14 consecutive days. The data represent the mean ± SD of at least ten mice per group. Histology was scored on a scale ranging from 0 to 3.

Table 4 - Histology of CIA after treatment with IL-10, prednisolone or IL-10/prednisolone Treatment

Dose

Infiltrate

Cartilage damage

Vehicle IL-10 Prednisolone IL-10/pred.

– 5 µg/day 0.05 mg/kg 5/0.05

1.7 1.3 1.9 1.0

1.5 1.2 1.7 0.8

± ± ± ±

0.9 0.7 1.0 0.8

± ± ± ±

1.0 0.6 0.9 0.7*

Proteoglycan depletion 2.3 2.0 2.1 1.5

± ± ± ±

1.3 1.0 1.0 1.0*

COMP (µg/ml) 8.2 8.0 10.2 4.8

± ± ± ±

0.8 0.7 1.8 0.6*

Treatment of arthritic mice was started at day 28 after immunization with CII. Mice were injected i.p. twice a day with murine IL-10, prednisolone or the combination for 14 consecutive days. The data represent the mean ± SD of at least ten mice per group. Histology was scored on a scale ranging from 0 to 3. Serum COMP levels were determined by ELISA, levels in normal sera amount 4.2 ± 0.6 µg/ml. *P

In Vivo Models of Inflammation

In Vivo Models of Inflammation, 2e, Vol. II

Hepatotoxicity: From Genomics to In Vitro and In Vivo Models

Microarrays in Inflammation (Progress in Inflammation Research)

Hepatotoxicity: From Genomics to In Vitro and In Vivo Models

The Resolution of Inflammation (Progress in Inflammation Research) (Progress in Inflammation Research)

In Vivo Neuromethods

Fundamentals of Inflammation

Regulatory T Cells in Inflammation (Progress in Inflammation Research)

Proteases and Their Receptors in Inflammation (Progress in Inflammation Research)

Textbook of in vivo Imaging in Vertebrates

Histamine in Inflammation

Endothelial Dysfunction and Inflammation (Progress in Inflammation Research)

In Vivo Imaging of Cancer Therapy

Cell Signaling in Vascular Inflammation

In Vivo Optical Imaging of Brain Function,

In Vivo Optical Imaging of Brain Function

The Activities of Bacterial Pathogens in Vivo

Neurokinetics: The Dynamics of Neurobiology in Vivo

Inflammation Protocols

Meglio Morto Che Vivo

Vivo o morto

The NPY Family of Peptides in Immune Disorders, Inflammation, Angiogenesis, and Cancer (Progress in Inflammation Research)

Molecular and Cellular Basis of Inflammation (Current Inflammation Research)

Practical Manual of Intraocular Inflammation

In vivo Models of HIV Disease and Control (Infectious Agents and Pathogenesis)

The NPY Family of Peptides in Immune Disorders, Inflammation, Angiogenesis, and Cancer (Progress in Inflammation Research)

In Vivo Optical Imaging of Brain Function (Frontiers in Neuroscience)

In Vivo Optical Imaging of Brain Function (Frontiers in Neuroscience)

Chemosensitivity: In Vivo Models, Imaging, and Molecular Regulators: Volume 2 (Methods in Molecular Medicine)

Turning up the Heat on Pain: TRPV1 Receptors in Pain and Inflammation (Progress in Inflammation Research)

In Vivo Models of Inflammation