University of Groningen The significance of serum …1. Choy EH, Panayi GS. Cytokine pathways and joint inflammation in rheumatoid arthritis. N Engl J Med 2001;344(12):907-16. 2. Navarro-Cano

University of Groningen

The significance of serum matrix metalloproteinase 3 in patients with early rheumatoidarthritisPosthumus, Marcel Desiderius

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.

Document VersionPublisher's PDF, also known as Version of record

Publication date:2004

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):Posthumus, M. D. (2004). The significance of serum matrix metalloproteinase 3 in patients with earlyrheumatoid arthritis. s.n.

CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.

Download date: 21-12-2020

https://www.rug.nl/research/portal/en/publications/the-significance-of-serum-matrix-metalloproteinase-3-in-patients-with-early-rheumatoid-arthritis(c0d3fb44-7a77-4454-a58d-dcb93457afdf).html



The significance of

serum matrix metalloproteinase 3 in patients with early rheumatoid arthritis

Marcel D. Posthumus

Studies presented in this thesis were financially supported by the “Dutch Arthritis Association”. The printing of this thesis was financially supported by: the “Dutch Arthritis Association”, Pfizer B.V., Wyeth Pharmaceuticals B.V., Abbott B.V., Schering-Plough B.V., AstraZenica B.V., Merck Sharp & Dohme B.V. & Novartis Pharma B.V. Copyright 2004 Marcel Posthumus ISBN-nummer: 90-367-1994-1 Printed by Drukkerij Van Denderen, Groningen.

Rijksuniversiteit Groningen

The significance of serum matrix metalloproteinase 3

in patients with early rheumatoid arthritis

Proefschrift

ter verkrijging van het doctoraat in de Medische Wetenschappen

aan de Rijksuniversiteit Groningen op gezag van de

Rector Magnificus, dr. F. Zwarts, in het openbaar te verdedigen op

woensdag 14 april 2004 om 14.45 uur

door

Marcel Desiderius Posthumus

geboren op 4 oktober 1961 te Lochem

Promotor: Prof. dr. M.H. van Rijswijk Copromotores: Dr. M.A. van Leeuwen

Dr. P.C. Limburg Beoordelingscommissie: Prof. dr. C.G.M. Kallenberg

Prof. dr. P.L.C.M. van Riel Prof. dr. J.J. Rasker

ISBN-nummer: 90-367-1994-1

Contents

List of abbreviations

7

Chapter 1

Introduction & aim of the thesis

9

Chapter 2

Matrix metalloproteinases in rheumatoid arthritis

15

Chapter 3

Serum levels of matrix metalloproteinase 3 in relation to the development of radiological damage in patients with early rheumatoid arthritis

51 Chapter 4

Serum matrix metalloproteinase 3 in early rheumatoid arthritis is correlated with disease activity and radiological progression

67 Chapter 5

Serum matrix metalloproteinase 3 levels during treatment with sulphasalazine or the combination of methotrexate and sulphasalazine in patients with early rheumatoid arthritis

85 Chapter 6

Serum matrix metalloproteinase 3 levels in comparison to C-reactive protein in periods with and without progression of radiological damage in patients with early rheumatoid arthritis

99 Chapter 7

Matrix metalloproteinase 1 and 3 promoter polymorphism in patients with early rheumatoid arthritis

115 Chapter 8

Summary, general conclusions and future perspectives

137

Chapter 9

Nederlandse samenvatting

147

Dankwoord

153

Curriculum Vitae

155

7

List of abbreviations ADAM a disintegrin and metalloproteinase ADAMTS a disintegrin and metalloproteinase with thrombospondin motifs α2M α2 macroglobulin Anti-CCP anti cyclic citrullinated peptide AP-1 activating protein-1 ATF-2 activating transcription factor 2 CRP C-reactive protein CS chondroitin sulphate DAS disease activity score ECM extra cellular matrix ERK extracellular stimulus regulated kinases ESR erythrocyte sedimentation rate ETS erythroblastosis twenty six FLS fibroblast like synoviocyte GAG glycosaminoglycan HLA human leucocyte antigen IKK inhibitor of κB kinase IκB inhibitor of κB IL-1 interleukin-1 JAK janus kinase JNK c-Jun N terminal kinases KS keratan sulphate MAPK mitogen activated protein kinase MEKK-1 mitogen activated protein kinase/ERK kinase kinase-1 MMP matrix metalloproteinase NF-κB nuclear factor κ B NIK NF-κB inducing kinase OA osteoarthritis PEA-3 polyoma enhancer A-binding protein-3 RA rheumatoid arthritis RF rheumatoid factor SJC swollen joint count TACE TNF-α converting enzyme TIMP tissue inhibitor of metalloproteinase TJC tender joint count TNF-α tumour necrosis factor α STAT signal transducers and activators of transcription

Chapter 1


Marcel D. Posthumus

Chapter 1

10

INTRODUCTION & AIM OF THE THESIS Rheumatoid arthritis (RA) is a complex inflammatory auto-immune disease associated with considerable disability, morbidity, and mortality 1,2. Early identification of patients with aggressive destructive disease is important, not only for prognostic, but also for therapeutic reasons 3. Novel, more aggressive therapies are currently developed 4 and the need for clinically feasible means for assessing the prognosis in the individual patients is becoming increasingly important 5.

Cytokines play a major role in local inflammation and destruction, as well as in the systemic acute phase response 1. The relationship between acute phase proteins such as C-reactive protein (CRP) on the one hand, and disease activity and radiological progression on the other, has been demonstrated in several studies 6,7. Although CRP appears to be a good marker for prognostic purposes and for monitoring treatment effects, it is an indicator of inflammation in general which may be influenced by other stimuli of the acute phase response 8.

The matrix metalloproteinases (MMPs), which are locally produced and activated within the affected joint as a result of cytokine mediated stimulation 9 could be more specific markers for joint inflammation and especially destruction in rheumatic diseases. In RA matrix metalloproteinase 3 (MMP-3=stromelysin 1) is of interest because it is thought to play a prominent role in the pathogenesis of matrix destruction in the inflamed joints 10. MMP-3 is capable of degrading many components of the matrix in the synovial joint 11, and has been localized in fibroblast-like synoviocytes of rheumatoid synovium 12,13 and in RA cartilage 14. The enzyme is secreted as a latent pro-enzyme resulting in elevated levels in local tissue and synovial fluid 15. In RA, serum MMP-3 is thought to originate from the inflamed joint because there are significant correlations between MMP-3 levels in synovial fluid and in serum 16, and between serum MMP-3 levels and the number of clinically active inflamed joints 16,17. Furthermore local therapies, such as intra-articular corticosteroid injections, cause significant reductions in serum MMP-3 levels in synovial fluid and in the systemic circulation of patients with RA 8,18.

The relative balance between activated MMPs and their inhibitors such as the tissue inhibitors of matrix metalloproteinases (TIMPs) and α2-macroglobulin (α2-M) eventually determines the fate of the matrix. In synovial fluid the levels of activated MMP-3 are increased but measurement of activated MMP-3 in the systemic circulation is very difficult, if not impossible, because of the high levels of α2-M in the systemic circulation which encloses the activated enzyme completely 19.


11

Elevated systemic levels of MMP-3 (probably mainly pro-MMP-3) have been found in several other rheumatic diseases such as SLE, systemic sclerosis, ankylosing spondylitis, crystal induced arthritis, and psoriatic arthritis 20,21. Nevertheless, in RA the systemic levels of MMP-3 are considered to be a direct reflection of local synthesis and as such, serum MMP-3 could be a useful marker of disease activity and destruction in early RA.

The aim of the thesis was to investigate the significance of serum MMP-3 in relation to disease activity, and in particular in relation to radiological progression, in patients with early RA. The data were derived from a prospective follow-up study in patients with recent onset RA (disease symptoms < 1 year). In chapter 2 a general review is provided on matrix metalloproteinases, in par-ticular in rheumatoid arthritis. In chapter 3 the significance of serum MMP-3 in relation to the development of radiological damage in early RA is evaluated and in particular its prognostic value at an early time-point. In chapter 4 the clinical significance of serial measurements of serum MMP-3 levels in relation to markers of disease activity and radiological progression in early RA is analyzed, during a follow-up of three years. In chapter 5 the effects of treatment with sulphasalazine, or the combination methotrexate and sulphasalazine, on the serum MMP-3 levels in patients with early RA is investigated. Changes in serum MMP-3 were also compared to changes in CRP levels In chapter 6 serum MMP-3 levels in periods with and without progression of radiological damage in patients with early RA are evaluated and compared to CRP. In chapter 7 functional (at least in in vitro experiments 22) promoter poly-morphisms in MMP genes are evaluated. In this study the significance of MMP-1 and MMP-3 promoter polymorphisms in relation to disease activity and radiological damage in patients with early rheumatoid arthritis is analyzed. In chapter 8 & 9 a summary, the general conclusions and future perspectives are given.

Chapter 1

12

REFERENCES 1. Choy EH, Panayi GS. Cytokine pathways and joint inflammation in rheumatoid arthritis. N Engl J

Med 2001;344(12):907-16. 2. Navarro-Cano G, Del R, I, Pogosian S, Roldan JF, Escalante A. Association of mortality with

disease severity in rheumatoid arthritis, independent of comorbidity. Arthritis Rheum 2003;48(9):2425-33.

3. Emery P, Breedveld FC, Dougados M, Kalden JR, Schiff MH, Smolen JS. Early referral recommendation for newly diagnosed rheumatoid arthritis: evidence based development of a clinical guide. Ann Rheum Dis 2002;61(4):290-7.

4. Smolen JS, Steiner G. Therapeutic strategies for rheumatoid arthritis. Nat Rev Drug Discov 2003; 2(6):473-88.

5. Combe B, Dougados M, Goupille P, Cantagrel A, Eliaou JF, Sibilia J et al. Prognostic factors for radiographic damage in early rheumatoid arthritis: a multiparameter prospective study. Arthritis Rheum 2001;44(8):1736-43.

6. van Leeuwen MA, van der Heijde DM, van Rijswijk MH, Houtman PM, van Riel PL, van de Putte LB et al. Interrelationship of outcome measures and process variables in early rheumatoid arthritis. A comparison of radiologic damage, physical disability, joint counts, and acute phase reactants. J Rheumatol 1994;21(3):425-9.

7. van Leeuwen MA, van Rijswijk MH, van der Heijde DM, Te Meerman G, van Riel PL, Houtman PM et al. The acute-phase response in relation to radiographic progression in early rheumatoid arthritis: a prospective study during the first three years of the disease. Br J Rheumatol 1993;32 Suppl 3:9-13.

8. Taylor DJ, Cheung NT, Dawes PT. Increased serum proMMP-3 in inflammatory arthritides: a potential indicator of synovial inflammatory monokine activity. Ann Rheum Dis 1994;53(11):768-72.

9. Vincenti MP, Brinckerhoff CE. Transcriptional regulation of collagenase (MMP-1, MMP-13) genes in arthritis: integration of complex signaling pathways for the recruitment of gene-specific transcription factors. Arthritis Res 2002;4(3):157-64.

10. Hasty KA, Reife RA, Kang AH, Stuart JM. The role of stromelysin in the cartilage destruction that accompanies inflammatory arthritis. Arthritis Rheum 1990;33(3):388-97.

11. Murphy G, Knauper V, Atkinson S, Butler G, English W, Hutton M et al. Matrix metalloprotein-ases in arthritic disease. Arthritis Res 2002; 4 Suppl 3:S39-S49.

12. Okada Y, Takeuchi N, Tomita K, Nakanishi I, Nagase H. Immunolocalization of matrix metalloproteinase 3 (stromelysin) in rheumatoid synovioblasts (B cells): correlation with rheumatoid arthritis. Ann Rheum Dis 1989;48(8):645-53.

13. Tetlow LC, Lees M, Woolley DE. Comparative studies of collagenase and stromelysin-1 expression by rheumatoid synoviocytes in vitro. Virchows Arch 1995;425(6):569-76.

14. Martel Pelletier J, McCollum R, Fujimoto N, Obata K, Cloutier JM, Pelletier JP. Excess of metalloproteases over tissue inhibitor of metalloprotease may contribute to cartilage degradation in osteoarthritis and rheumatoid arthritis. Lab Invest 1994;70(6):807-15.

15. Yoshihara Y, Nakamura H, Obata K, Yamada H, Hayakawa T, Fujikawa K et al. Matrix metalloproteinases and tissue inhibitors of metalloproteinases in synovial fluids from patients with rheumatoid arthritis or osteoarthritis. Ann Rheum Dis 2000;59(6):455-61.

16. Yoshihara Y, Obata K, Fujimoto N, Yamashita K, Hayakawa T, Shimmei M. Increased levels of stromelysin-1 and tissue inhibitor of metalloproteinases-1 in sera from patients with rheumatoid arthritis. Arthritis Rheum 1995;38(7):969-75.

17. Sasaki S, Iwata H, Ishiguro N, Obata K, Miura T. Detection of stromelysin in synovial fluid and serum from patients with rheumatoid arthritis and osteoarthritis. Clin Rheumatol 1994;13(2):228-33.


13

18. Cheung NT, Mattey DL, Dawes PT, Taylor DJ. Serum pro-matrix metalloproteinase 3 in rheumatoid arthritis: a reflection of local or systemic inflammation? Arthritis Rheum 1996;39(5):884-6.

19. Barrett AJ. Alpha 2-macroglobulin. Methods Enzymol 1981; 80:737-754. 20. Keyszer G, Lambiri I, Nagel R, Keysser C, Keysser M, Gromnica-Ihle E et al. Circulating levels

of matrix metalloproteinases MMP-3 and MMP-1, tissue inhibitor metalloproteinases 1 (TIMP-1), and MMP-1/TIMP-1 complex in rheumatic disease. Correlation with clinical activity of rheumatoid arthritis versus other surrogate markers. J Rheumatol 1999;26:251-8.

21. Zucker S, Lysik RM, Zarrabi MH, Greenwald RA, Gruber B, Tickle SP et al. Elevated plasma stromelysin levels in arthritis. J Rheumatol 1994;21(12):2329-33.

22. Ye S. Polymorphism in matrix metalloproteinase gene promoters: implication in regulation of gene expression and susceptibility of various diseases. Matrix Biol 2000;19(7):623-9.

Chapter 2

Matrix metalloproteinases in rheumatoid arthritis

Marcel D. Posthumus, Pieter C. Limburg, Miek A. van Leeuwen, and Martin H. van Rijswijk.

Submitted.

Chapter 2

16

INTRODUCTION Rheumatoid arthritis (RA) is the most common polyarthritis affecting about 0.5-1% of the general population. Besides a chronic symmetrical polyarthritis of small joints in hands and feet, larger joints are generally involved and extra-articular manifestations are frequently present. It is a complex inflammatory auto-immune disease, associated with considerable disability, morbidity, and mortality 1. I. Pathogenesis of RA Despite intensive research the exact cause of RA remains a mystery. A mixture of environmental and genetic factors seems to be of importance. Environmental pathogens may infect an individual with a predisposing genetic background and by an unknown mechanism the subsequent inflammatory response is mainly targeted to the synovial joints 2.

Based on data from animal models it is likely that involvement of innate immune mechanisms is directive in this process. In the induction phase the innate immunity may prepare the synovium for T cell infiltration and related immune events 3. Subsequently synovial-lining cells are supposed to be engaged, which produce several chemokines that attract monocytes and lymphocytes into the joint. If these cells recognize an antigen in this environment they will be retained and persisting inflammation may ensue. This hypothesis does not require one specific trigger. Any pathogen that gains entry into the joint could be responsible in a given individual. Furthermore there is a possibility of genetically determined variability in response to the same environmental insult.

The external stimulus initiates T cell-mediated responses directed either against the inciting antigen or against a secondary joint-specific target such as type II collagen. As a consequence the typical clinical picture of RA arises, driven by lymphocytes, antigen presenting cells, and macrophages. As the disease progresses, multiple cytokine networks become established and antigen independent processes become more important. The chronic inflammatory process gradually achieves a certain degree of autonomy of non-antigen specific cells, such as Fibroblast Like Synoviocytes (FLS), which will persist even after down regulation of the initial antigen specific T cell response. In addition to this autonomy of the FLS, experimental models of arthritis support arguments for the existence of a “transformed” phenotype of a fraction of RA synoviocytes. For instance cells from the pannus, directly eroding bone or cartilage, have distinctive morphology and features of both FLS and chondrocytes 4,5. These alterations in synoviocyte function might have profound implications for matrix

MMPs in rheumatoid arthritis

17

destruction, as well as for remodeling and repair in individual joints of individual patients.

II. Pathophysiology of joint destruction in RA The Extra-Cellular Matrix (ECM) gives structural support to cells and tissues, and plays a central role in cell adhesion, differentiation, proliferation, and migration of its constituents. The turnover of the ECM usually occurs slowly in mature tissues but is accelerated in conditions such as wound healing, arthritic joint destruction, and malignancy.

In the ECM of joints numerous macromolecules have been identified. The major constituents, especially of cartilage, are collagen and the proteoglycan aggrecan, which forms large aggregated complexes with hyaluronan and link protein (figure 1).

FIGURE 1. Schematic representation of aggrecan with its major proteolytic cleavage sites (1,2). The aggrecan molecule has three globular domains (G1, G2 and G3). The N-terminal G1 domain interacts with hyaluronan and a link protein, forming large aggregates. In the interglobular domain between G2 and G3 keratin- (KS) and chondroitin- (CS) sulphate chains are inserted. The two major cleavage sites of human aggrecan in vivo are the “MMP cleavage site”(1) generating the G1-VDIPEN341 and the “Aggrecanase-cleavage site”(2) generating G1-NITEGE373 fragments. A. Collagen The collagen meshwork in cartilage contains fibrils of type II collagen and to a lesser extent of collagen type IX and XI, linked together by a number of collagen binding glycoproteins on their surface. Type II collagen consists of 3 identical α chains that form triple-helical monomers that associate in a one-quarter stagger to

Chapter 2

18

form fibrils which also contain type IX and XI collagens. This network provides the cartilage with tensile strength and capacity to resist the swelling pressure exerted by the entrapped water molecules in the proteoglycans 6.

There appears to be little capacity for chondrocytes to regenerate collagen architecture once the mature tissue is injured. Collagen degradation is considered to be a critical and probably irreversible step in cartilage destruction in diseases as osteoarthritis and RA 7. B. Aggrecan Aggrecans fill the interstitial spaces of the collagen meshwork by forming large aggregating complexes interacting with hyaluronan and a link protein. Aggrecan is a multidomain glycoprotein composed of a central protein chain, which has 3 globular domains (G1, G2 and G3). In particular the region between G1 and G2 is susceptible to cleavage by members of the Matrix MetalloProteinases (MMPs) and aggrecanases, members of the “A Disintegrin And Metalloproteinase with Thrombospondin Motifs” (ADAMTS) family (figure 1) 8,9. In the interglobular domain between G2 and G3 hydrophylic sulphated glycosaminoglycans (GAGs) such as chondroitin sulphate (CS) and keratan sulphate (KS) chains attract and entrap water molecules, which give cartilage its reversible deforming capacity during joint loading 6.

Proteoglycan degradation is thought to be an early and reversible process, in contrast to collagen degradation, which is believed to be irreversible 9. C. Joint destruction The degradation of ECM components of cartilage, bone, tendons and ligaments is a key process in joint destruction in RA, one of the strongest predictors of long term outcome and disability 10. The balance between synthesis of ECM components and activity of destructive proteases as well as their natural inhibitors determines the fate of the ECM.

Cartilage is destroyed by both enzymatic and mechanical processes. Proteolytic enzymes produced by synoviocytes in the pannus, by granulocytes in the synovial fluid, as well as by chondrocytes themselves degrade proteoglycans, such as aggrecan. As soon as cartilage is depleted from proteoglycans it loses its ability to rebound from a deforming load, leading to increased susceptibility to mechanical fragmentation and eventually to loss of functional integrity 11. Besides proteoglycan loss, collagen degradation is an essential and irreversible process in cartilage disruption 9. This destructive process is mediated by several families of proteases such as aspartic, serine and cysteine proteases. However,


19

the MMPs and aggrecanases, members of the ADAMTS family, are supposed to be key mediators in these processes 8,12,13.

Bone erosion and destruction results mainly from the activation of osteoclasts. RA synovial tissues produce a variety of factors, such as IL-1α/β, TNF-α, IL-11, IL-15, IL-17, and others 14 with the capacity to increase osteoclast differentiation and activation. In addition activated T lymphocytes and FLS express Osteoclast Differentiation Factor (ODF), also known as Receptor Activator of NF-κB Ligand (RANKL), critical in inducing the differentiation of cells of the monocyte/macrophage lineage into osteoclasts. Furthermore ODF has the capacity of directly activating multinucleated osteoclasts which are in contiguity with bone 15 resulting in resorption of the mineral phase of bone followed by degradation of the remaining matrix by cysteine proteases and MMPs 16. MATRIX METALLOPROTEINASE FAMILY Proteolytic enzymes play critical roles in many biological processes such as growth, development, reproduction, and wound healing. Under normal physiological conditions degradation and synthesis of ECM components are precisely regulated. However, if these processes are disrupted, pathologic conditions may arise 17.

The MMPs are thought to be key enzymes involved in remodeling of the ECM in physiological and pathological situations (e.g. arthritis, cancer, atherosclerosis, and periodontal disease). In diseases as rheumatoid arthritis and osteoarthritis MMPs play a major role in matrix degradation in collaboration with other proteinases, in particular members of the ADAM (A Disintegrin And Metalloproteinase) and ADAMTS (A Disintegrin And Metalloproteinase with ThromboSpondin motif) family also designated as aggrecanases 11. I. Matrix metalloproteinases The MMPs are a family of zinc-containing, calcium-dependent endopeptidases that are active at a neutral pH. Today at least 24 different MMPs have been identified of which 23 are found in humans 18. At least 8 of the known human MMP genes are clustered at chromosome 11q21-q23 (MMP-1, -3, -7, -8, -10, -12, -13, and -20). Other known MMPs are scattered among several other chromosomes.

On the basis of their structural and functional properties MMPs are divided into five major groups: I. Collagenases (MMP-1, -8, -13), II. Stromelysins (MMP-3, -10, -11), III. Gelatinases (MMP-2, -9), IV. Membrane-type (MT-) MMPs (MMP-14 to -17, -24, -25), and V. a heterogeneous subgroup (table 1) 19.

Chapter 2

20

Table 1. The matrix metalloproteinases and their substrates I. Collagenases MMP-1 Interstitial collagenase Collagens I, II, III, VII, VIII and X, gelatin, aggrecan, versican, proteoglycan

link protein, casein, α1-proteinase inhibitor, α2-M pregnancy zone protein, ovostatin, nidogen, MBP, proTNF, L-selectin, proMMP-2, proMMP-9

MMP-8 Neutrophil collagenase

Collagens I, II, III, V, VII, VIII and X, gelatin, aggrecan, α1-proteinase inhibitor, α2-antiplasmin, fibronectin

MMP-13 Collagenase-3 Collagens I, II, III and IV, gelatin, plasminogen activator inhibitor 2, aggrecan, perlecan, tenascin

MMP-18 Collagenase-4 (Xenopus)

II. Stromelysins MMP-3 Stromelysin-1 Collagens III, IV, IX and X, gelatin, aggrecan, versican, perlecan, nidogen,

proteoglycan link protein, fibronectin, laminin, elastin, casein, fibrinogen, antithrombin-III, α2M, ovostatin, α1-proteinase inhibitor, MBP, proTNF, proMMP-1, proMMP-7, proMMP-8, proMMP-9, proMMP-13

MMP-10 Stromelysin-2 Collagens III, IV and V, gelatin, casein, aggrecan, elastin, proteinglycan link protein, fibronectin, proMMP-1, proMMP-8

MMP-11 Stromelysin-3 α1-proteinase inhibitor, III. Gelatinases MMP-2 Gelatinase A Collagens I, IV, V, VII, X, XI and XIV, gelatin, elastin, fibronectin, aggrecan,

versican, proteoglycan link protein, MBP, proTNF, α1-proteinase inhibitor, proMMP-9, proMMP-13

MMP-9 Gelatinase B Collagens IV, V, VII, X and XIV, gelatin, elastin, aggrecan, versican, proteoglycan link protein, fibronectin, nidogen, α1-proteinase inhibitor, MBP, proTNF

IV. Membrane-type MMPs MMP-14 MT1-MMP Collagens I, II, II, gelatin, casein, elastin, fibronectin, laminin B chain,

vitronectin, aggrecan, dermatan sulfate proteoglycan, MMP-2, MMP-13, proTNF

MMP-15 MT2-MMP ProMMP-2, gelatin, fibronectin, tenascin, nidogen, laminin MMP-16 MT3-MMP ProMMP-2 MMP-17 MT4-MP MMP-24 MT5-MMP ProMMP-2, proMMP-9, gelatin MMP-25 MT6-MMP Collagen IV, gelatin, fibronectin, fibrin V. Others MMP-7 Matrilysin Collagens IV and X, gelatin, aggrecan, proteoglycan link protein, fibronectin,

laminin, entactin, elastin, casein, transferrin, MBP, α1-proteinase inhibitor, proTNF, proMMP-1, proMMP-2, proMMP9

MMP-12 Metalloelastase Collagen IV, gelatin, elastin, α1-proteinase inhibitor, fibronectin, vitronectin, laminin, proTNF, MBP

MMP-19 Collagen IV, gelatin, laminin, nidogen, tenascin, fibronectin, aggrecan, COMP MMP-20 Enamelysin Amelogenin MMP-21 XMMP (xenopus) MMP-22 (-27)

CMMP (chicken) Gelatin, casein

MMP-23 CA-MMP MMP-26 Matrylisin-2 Collagen IV, fibronectin, fibrinogen, gelatin, α1-proteinase inhibitor, proMMP-

9 MMP-28 Epilysin Casein The matrix metalloproteinases and their substrates. The table was adapted, with permission, from Gillian Murphy,Vera Knäuper, Susan Atkinson, George Butler, William English, Mike Hutton, Jan Stracke and Ian Clark: Matrix metalloproteinases in arthritic disease. Arthritis Res 2002, 4 (suppl 3): S39-S49 (reference 19). http://arthritis-research.com/content/4/S3/S39


21

In general MMPs are composed of a signal peptide at the amino (NH2) terminal followed by a propeptide domain, a zinc-containing catalytic domain and a haemopexin-like domain at the carboxyl (COOH) terminal as shown in figure 2. The gelatinases (MMP-2, -9) have additional features such as three fibronectin type II repeats in the catalytic domain, that provides them gelatin-binding properties. The MT-MMPs are characterized by a COOH-terminal transmembrane region followed by a short intracytoplasmic tail.

FIGURE 2. Schematic representation of the domain structures of the matrix metalloproteinases. The NH2 terminal signal peptide is a hydrophobic sequence of 8-30 residues,

responsible for trafficking of the enzyme through the endoplasmic reticulum and the Golgi apparatus and for its subsequent secretion into the extra cellular space. The signal peptide is cleaved off during the secretion process. The propeptide domain consists of about 80 residues and is responsible for the enzyme latency. It contains one unpaired cysteine in the highly conserved PRCG[V/N]PD motif, the so-called “cysteine switch” (figure 3). In the latent enzyme this sequence of the propeptide is located directly opposite the active site cleft and the SH group of the cysteine interacts with the zinc atom. This prevents the formation of a water-zinc complex that is required for the proteolytic activity. The catalytic domain (about 170 residues) contains the zinc-binding motif HExxHxxGxxH located 50-55 residues from the COOH terminal in which 3 histidines (H) bind to the

Chapter 2

22

catalytic zinc atom. In addition it contains a conserved methionine, which forms a unique “Met-turn” structure. Between the catalytic domain and the COOH terminal haemopexin-like domain lies a variable stretch of 2-72 amino acids, which is named the “hinge region”. The COOH-terminal haemopexin-like domain of about 200 residues has a number of functions including substrate and inhibitor binding or involvement in activation processes 13,19. This haemopexin domain is an absolute requirement for the collagenases to cleave triple helical interstitial collagens. As an illustration the amino acid sequence 20 and the crystal 19 structure of MMP-3 are shown in figure 3 and 4.

FIGURE 3. Amino-acid sequence and specific sites of matrix metalloproteinase 3. The propeptide is represented in italic and ranges from amino acid 18-99. The “bait region” is involved in the activation of the MMPs (see figure 9).

The secreted collagenases (MMP-1, -8 and -13) and membrane bound MT1-MMP are the only enzymes capable of cleaving the intact triple helix of collagen types I, II, and III. The degradation of the triple helical collagen II occurs at a specific site between residues 775 (glycine) and 776 (leucine) within each α chain of the triple helical collagen molecule, resulting in the characteristic TCA (3/4) and TCB (1/4) cleavage products. This opening of the triple helix allows access for other proteinases such as MMP-3, MMP-2, and MMP-9 21. Besides a role in this process MMP-3 is able to degrade numerous other components of the ECM such as aggrecan, fibronectin, laminin, and several other collagens. And, in addition MMP-3 can activate several pro-MMPs such as pro-MMP-1, -7, -8, -9, and -13. One of the central pathophysiological features contributing to cartilage destruction is the catabolism and subsequent loss of aggrecan. One well-characterized MMP-cleavage site is the Asn341-Phe342 bond in the interglobular


23

FIGURE 4. Structure of the catalytic domain of matrix metalloproteinase 3. Strands (sI-sV) and helices (hA-hC) are labelled in black; the catalytic zinc (centre), co-ordinated by three histidine residues (pink), and the structural zinc (Zn) (top) are labelled in black and white, respectively; the NH2 terminus (N) and COOH terminus (C) are labelled in red. The active site cleft and the characteristic methionine (met) turn are indicated by arrows. The figure was adapted, with permission, from Gillian Murphy, Vera Knäuper, Susan Atkinson, George Butler, William English, Mike Hutton, Jan Stracke and Ian Clark: Matrix metalloproteinases in arthritic disease. Arthritis Res 2002, 4 (suppl 3): S39-S49 (reference 19). http://arthritis-research.com/content/4/S3/S39

domain between G1 and G2 (see figure 1). A cleavage by various MMPs in this region separates a major part of aggrecan (MMP-generated G1-VDIPEN341) from its hyaluronan attachment. In 1991 another cleavage site was found at the Glu373-Ala374 bond in the interglobular domain between G1 and G2 resulting in G1-NITEGE373 fragments. The enzyme responsible was referred as "aggrecanase-1", now designated as ADAMTS-4 22,23. II. The ADAM(TS)s The ADAMs are transmembrane proteins. They consist of a NH2-terminal (extra-cellular) signal peptide, a propeptide (about 170 amino acids), a metalloproteinase domain (about 230 amino acids), a disintegrin domain, a cysteine-rich region usually containing an epidermal growth factor-like domain

Chapter 2

24

and a transmembrane domain followed by an intracytoplasmic tail 8. The biological functions of many ADAMs are unknown, but for example ADAM-17 also known as tumour necrosis factor α converting enzyme (TACE), is involved in the release of TNF-α as well as of TNF-α receptors and other surface molecules.

The ADAMTSs are related to the ADAMs but they are not membrane-anchored and contain additional thrombospondin type I motifs (see figure 2) responsible for its anchoring to the ECM 24. The ADAMTS-1, -4, and -5 (the “aggrecanases”) are of major importance in cartilage degradation.

Besides the Glu373-Ala374 bond in the interglobular domain of aggrecan, ADAMTSs cleave several other sites including the MMP-cleavage site Asn341-Phe342 (see figure 1). In arthritic joints MMP-generated G1-VDIPEN341 and ADAMTSs-generated G1-NITEGE373 fragments are found in cartilage 25 and synovial fluids 26. There is discussion which group of enzymes plays the primary role in in-vivo cartilage degradation. In in-vitro cartilage explant systems, the initial enzymes responsible for degrading aggrecan are the ADAMTSs followed by MMPs at a later stage 27. REGULATION OF MMP PRODUCTION To prevent tissue destruction it is important that the activities of MMPs are tightly controlled. The relative balance between activated MMPs and their inhibitors are thought to determine the rate of ECM turnover. Many MMPs are not constitutively produced but inducible and their production can be regulated at several levels such as signal transduction preceding transcription, transcription itself, or posttranscriptional processing. I. Signal transduction pathways Transcription can rise dramatically in response to various extra-cellular stimuli such as inflammatory cytokines, growth factors and matrix proteins. These extra- cellular signals are transduced to the nucleus by specific intra-cellular transcription factors generated by signal transduction pathways such as the Nuclear Factor κ B (NF-κB), the Mitogen-Activated Protein Kinase (MAPK) and the Signal Transducers and Activators of Transcription (STAT) pathways (figure 5-7).

For transcription of MMPs several transcription factor binding sites are of importance such as the NF-κB, Activating Protein-1 (AP-1), STAT and Polyoma Enhancer A-binding Protein-3 (PEA-3) sites (figure 8) 28.


25

A. NF-κB pathway (figure 5) The NF-κB/Rel family includes NF-κB1 (p50/p150), NF-κB2 (p52/p100), p65 (RelA), RelB and c-Rel 29. The most important activated forms are a heterodimer of p65 in combination with a p50 or p52 subunit 30.

FIGURE 5. Nuclear Factor-κB (NF-κB) signal transduction pathway. NIK; NF-κB inducing kinase. MEKK-1; mitogen activated protein kinase/ERK (extra-cellular, stimulus regulated kinase) kinase kinase-1. IKKα/β; inhibitor of κB kinase. IκB; inhibitors of κB.

Upon binding of IL-1 and/or TNF-α to their respective cell-surface receptors several intracellular proteins are recruited to the cytoplasmic domain of these receptors. These complexes then recruit and activate the NF-κB Inducing Kinase (NIK) 31 and mitogen-activated protein kinase/ERK kinase kinase-1 (MEKK-1), members of the MAPK kinase kinase (MAPKKK) family which activate 32 the Inhibitor of κB Kinase (IKK-α, IKK-β and the regulatory unit IKK-γ). These IKKs phosphorylate the inhibitors of κB (IκBα, -β and -ε), which are the cytosolic inhibitors of "pre-made" NF-κB. After phosphorylation IκB becomes

Chapter 2

26

ubiquitinated resulting in proteosome-mediated proteolytic degradation. The p50 and p65 subunits of NF-κB are then translocated to the nucleus where they bind to a NF-κB binding site in promoter regions of MMPs, such as in the MMP-1 promoter region. These binding sites cooperate with other binding sites such as the AP-1 site 33. This synergistic interaction, "cross-talk" between binding sites, seems to be of importance for the transcription of several MMPs such as MMP-3 34 and MMP-9 35. B. MAPK pathway (figure 6) Concomitant with activation of the NF-κB pathway IL-1 and TNF-α can induce the MAPK pathway 28. A MAPK cascade consists of several kinases activating (by phosphorylation) the next kinase, ultimately leading to the activation of transcription factors 36. For MMP transcription three groups of MAPKs are of importance; a. Extra-cellular stimulus-Regulated Kinases (ERK), b. c-Jun N-terminal Kinases (JNK) and c. p38 MAPK (p38). After binding of an extra-cellular stimulus, e.g. a cytokine or growth factor to its receptor, small guanylyl triphosphate (GTP)-binding proteins such as Ras (Rac, Cdc42) are activated which in turn activate MAPK kinase kinase (MAPKKK). MAPKKK phosphorylates and activates MAPKK, which activates the MAPK (ERK, JNK, and p38). Once activated the phosphorylated MAPK translocate to the nucleus where they phosphorylate and activate the actual transcription factors such as Erythroblastosis twenty-six (ETS), jun, and Activating Transcription Factor 2 (ATF-2) (part of respectively the ERK-, JNK- and p38-pathway). These transcription factors bind to specific binding sites in gene promoter regions such as the AP-1 site, which binds fos as well as jun families of transcription factors, and the PEA-3 sites, which bind the ETS family. This binding provides the nuclear signal that initiates the transcription of many genes, not only of target genes coding for MMPs but also of genes for transcription factors themselves such as c-fos, jun, and others.

To date the p38 pathway is not completely elucidated. There seems to be no known target of p38 that directly regulates MMP promoters. However p38 can activate transcription factors such as ATF-2 and Elk-1 (a member of the ETS oncogene superfamily) which drive c-jun and c-fos promoters. Possibly by this route of promoting transcription factors which regulate the expression of AP-1 genes, p38 contributes indirectly to MMP transcription 37.


27

FIGURE 6. Mitogen Activated Protein Kinase (MAPK) signal transduction pathway. MEKK; mitogen activated protein kinase/ERK kinase kinase-1. ASK-1; apoptosis signal-regulating kinase 1. ERK; extra-cellular stimulus-regulated kinase. JNK; c-Jun N-terminal kinase. ETS; erythroblastosis twenty six. ATF-2; activating transcription factor-2. PEA-3; polyoma enhancer A-binding protein-3 binding site. AP-1; activating protein-1 binding site. C. STAT pathway (figure 7) Upon binding of an extra cellular stimulus to its receptor Janus Kinases (JAKs) are activated followed by activation of STATs. Once phosphorylated STATs dimerise and translate to the nucleus where they bind to specific transcription response elements in MMP genes 28. Recently Li and colleagues showed that Oncostatin, a member of the IL-6 superfamily of cytokines induces MMP-1, -3 and -13 and TIMP-3 expression in chondrocytes by activating JAK/STAT and MAPK signaling cascades 38.

Chapter 2

28

FIGURE 7. JAK; Janus kinases. Signal Transducers and Activators of Transcription (STAT) signal transduction pathway. D. Transcription repression Transcription can also be inhibited. Glucocorticoids bind to an intracellular glucocorticoid receptor followed by nuclear translocation and binding to glucocorticoid response elements in gene promoters. By this mechanism glucocorticoids can inhibit the expression of multiple inflammatory genes. In addition the glucorticoid receptor can interfere with the activity of key transcription factors such as NF-κB by activating the IκB-α gene and AP-1 by physical interference with the fos and jun subunits of AP-1 39. II. Transcription itself: promoter polymorphisms Next to these signal transduction pathways Single Nucleotide Polymorphisms (SNPs) in the promoter regions of MMPs seem to be of importance. Promoter regions of many MMP genes contain binding sites for several transcription factors (figure 8).


29

FIGURE 8. Promoter region of MMP-3 with transcription factor binding sites. ZBP-89; 89 kDa zinc binding protein. SPRE; stromelysin 1 PDGF responsive element. NF- κB; nuclear factor κB. TCF; T-cell factor. CIZ; Cas-interacting zinc-finger protein. PEA-3; polyomavirus enhancer A binding protein 3. AP-1; activator protein. TBS; TEL (translocation-ETS-leukaemia) Binding site. STAT?: possible binding site of signal transducers and activators of transcription.TATA; TATA box. Reprinted with permission, from Nat Rev Cancer 2002;2(9):657-72 (reference 75), copyright 2002 Macmillan Magazines Ltd. These elements play important roles in the regulation of MMP transcription both at the basal level and in response to various stimuli including cytokines and growth factors. Naturally occurring sequence variations in the promoters of these genes may result in different levels of expression of MMPs 40. Approximately 90% of DNA polymorphisms are SNPs due to single base substitutions or insertions 41. Although the majority is probably functionally neutral, a number of them can exert allele specific effects on the regulation of gene expression of the coded protein. Promoter polymorphisms have been identified in a number of MMP genes such as MMP-1 and MMP-3. In the MMP-1 gene (locus 11q22-q23) two alleles have been detected, one having a single guanine (1G) and the other having two guanines (2G) at positions -1607 (-1607:1G/2G). The two MMP-3 alleles (at locus 11q23) are defined by the presence of a run of either 5 or 6 adenines at position -1171 (-1171:5A/6A). These polymorphisms have been shown to influence MMP gene expression and are associated with susceptibility and/or severity of diseases such as cancer and atherosclerosis.

In the MMP-1 promoter, at -1607 (and -1608) 2 guanines together with an adjacent adenine can create a core binding site (5’-GGA-3’) for the ETS family of transcription factors. This 2G allele binds substantially more ETS-1 and has substantially more transcriptional activity compared to the 1G allele in normal fibroblasts and melanoma cells 42 as well as in ovarian tumour tissue 43. In the study of Kanamori et al, concerning ovarian tumours, the proportion of 2G carriers was also significantly higher in patients than in healthy controls which might imply an influence on susceptibility 43. A relation with disease severity was found in a study on malignant melanoma. In this study the 2G allele was associated with deep invasiveness and thereby a worse prognosis 44.

Chapter 2

30

Studies concerning the -1607:1G/2G MMP-1 polymorphism in RA patients are scarce. In a recent study Constantin et al could not find an association between this SNP in the MMP-1 promoter and the susceptibility to or severity of RA 45.

The MMP-3 promoter polymorphism -1171:5A/6A appears to be functionally important as well. Cultured fibroblasts and vascular smooth muscle cells transfected with constructs containing the run of 6 adenines expressed a roughly 2-fold lower amount of reporter gene product as compared with the transfectants of the constructs containing a run of 5 adenines 46. Additional studies revealed binding of putative transcription factors to this region. The difference in promoter activity is probably due to preferential binding of a transcription repressor to the 6A allele 46,47. Influences of this polymorphism have been evaluated in atherosclerosis and cancer. The 5A allele was found to be associated with susceptibility of acute myocardial infarction 48,49 and abdominal aortic aneurysm 50, diseases associated with weakening of the matrix possibly as a result of increased MMP-3 expression. On the other hand the 6A allele seems to be associated with accelerated growth of coronary atherosclerosis 51 possibly due to accumulation of ECM as a result of an insufficient MMP-3 expression, production, and activation 40. In breast cancer the presence of the 5A allele seems to be associated with susceptibility and severity 52. On the other hand in colorectal cancer the 6A/6A genotype was found more often in patients compared to controls. In RA patients Constantin et al found no association between the -1171:5A/6A polymorphism and disease susceptibility. However, severity and progression, represented by radiologically detectable joint damage, appeared to be associated with the 6A/6A genotype 54, which is associated with a lower transcriptional activity, in in vitro experiments compared to the 5A genotype. This is somewhat unexpected but possibly linkage disequilibrium is of importance. In the MMP gene cluster the MMP-1 locus is adjacent to the MMP-3 locus and MMP-1 and MMP-3 promoter linkage disequilibrium has been reported. In colorectal cancer the 2G MMP-1/6A MMP-3 haplotype was significantly increased 53. And in RA patients Keyszer et al showed in a preliminary report that almost all haplotypes were found to be either 1G MMP-1/5A MMP-3 or 2G MMP-1/6A MMP-3 55. III. Posttranscriptional regulation Posttranscriptional regulation (the control of mRNA stability and translation) of pro-inflammatory gene expression is thought to be mediated by the adenosine/ uridine rich elements (AREs) within the 3' untranslated regions of the mRNA flanking the protein coding sequence 56. AREs have been shown to destabilize


31

heterologous transcripts into which they are inserted 57. Deletion of ARE elements from the TNF-α locus in the mouse caused an increase in TNF-α mRNA stability and a loss of the inhibitory effects of specific p38 MAPK inhibitors upon TNF-α biosynthesis 58. A mouse knockout study suggested that the effects of p38 may be mediated by Mitogen Activated Protein Kinase-Activated Protein Kinase 2 (MAPKAPK2) a kinase activated by p38. In absence of the MAPKAPK2 activity, reduced quantities of TNF-α protein were synthesized whereas the induction of TNF-α mRNA was unaffected 59. Taken together these findings suggest that p38 may regulate TNF-α mRNA translation by inducing MAPKAPK2 activity and by a mechanism involving the ARE region in the TNF-α gene. In contrast Brook et al reported that TNF-α mRNA was clearly destabilized by the p38 MAPK inhibitor SB203580, strongly indicating a role of p38 in the regulation of TNF-α mRNA stability 60. In addition p38 and MAPKAPK2 regulated the stability rather than the translation of IL-6, IL-8, GM-CSF and Cox-2 mRNAs in the human HeLa cell line 61. Besides control over mRNA turnover in cytokine and prostaglandin genes, AREs are also required for the regulated decay of MMP-transcripts such as those of MMP-1 and possibly MMP-13 28. REGULATION OF MMP ACTIVITY I. MMP activation Most MMPs are synthesized as inactive prepro-enzymes and are secreted as inactive pro-enzymes, which have to be activated by proteolytic cleavage. In vivo activating proteases (other MMPs, plasmin, trypsin) attack the so called “bait region” (figure 3 & 9) in the propeptide, resulting in an intermediate peptide with a destabilized cysteine-zinc interaction rendering the final activation site susceptible to a second proteolytic attack usually catalyzed by MMPs but not by the trigger protease 62. Some MMPs are activated intracellular such as MMP-11, whereas other MMPs are activated at the cell surface such as MMP-2 18.

Studies concerning the regulation of the ADAMTSs activity are inconsistent and results are probably influenced by species and age of the used tissue and culture conditions. The ADAMTSs possess a furin cleavage site just before the zinc-containing catalytic domain and are therefore probably activated intracellularly and secreted as active enzymes. Further regulation may be transcriptional and/or posttranscriptional at the level of mRNA stability and/or translation 8.

Chapter 2

32

FIGURE 9. Stepwise activation of proMMPs. Zn; zinc. C; cystein. Cleavage sites are indicated by an arrow (↑ ↓) II. MMP inhibition Activated MMPs can be inactivated by inhibitors such as the Tissue Inhibitors of MetalloProteinases (TIMPs) or α-2 macroglobulin (α-2M). Whereas α-2M acts primarily as a regulator in the systemic circulation, the TIMPs are considered to be the key inhibitors locally in the tissues. TIMPs are 2 domain molecules, consisting of an N-terminal domain of about 125 amino acids and a smaller C-terminal domain of about 65 residues, each domain being stabilized by three disulfide bonds. The shape of the TIMP molecule is like a wedge, which slots into the active-site cleft of an MMP in a manner similar to that of the substrate 18. Although the four different TIMPs bind to most activated MMPs by forming a 1:1 enzyme-inhibitor complex, significant differences have been reported between different TIMPs. For example TIMP-3 is a good inhibitor of TACE 63 and seems to be the most important inhibitor of ADAMTS-4 and -5 64,65. Besides inhibition of activated MMPs, TIMPs are important for pro-MMP activation (pro-MMP-2), cell growth promotion, inhibition of angiogenesis, and induction of apoptosis (reviewed by Brew et al 63). α-2M is a plasma glycoprotein synthesized by the liver and by macrophages

and fibroblasts. It is a large (Mr 725,000) plasma protein and responsible for inhibition of activated proteases in the systemic circulation. α-2M is active against many endopeptidases regardless of their specificity. The activated protease attacks α-2M in the “bait-region”. This triggers a conformational change in the α-2M molecule, which in turn entraps and inactivates the enzyme. Subsequently the α-2M-proteinase complexes are rapidly cleared by macrophages and fibroblasts by receptor-mediated endocytosis 66.


33

III. Balance Under physiological conditions the levels of protease inhibitors exceed the levels of their targeted proteases. Controlled activation of MMPs and inhibition by local inhibitors such as TIMPs will allow a low levels of turnover to occur. Many proteases, including MMPs are not constitutively produced but inducible and in pathological conditions transcription, translation and activation can increase substantially. In contrast, protease inhibitors are constitutive proteins of which the production can also be increased by cytokines such as IL-1, IL-6 and TNF-α. Given the important role of MMPs in tissue destruction, the relative balance between activated MMPs and their inhibitors such as TIMPs and α-2M will ultimately determine the fate of the ECM. ROLES OF MMPs I. Physiologic Because MMPs can degrade almost all components of the ECM their main function is presumed to be ECM remodeling e.g. during embryonic development, tissue growth, and morphogenesis. In fact the discovery of the MMPs in 1962 was based on the observation that during amphibian metamorphosis a collagenolytic activity had to be present to digest the collagens in tadpole tails 67. Ongoing research has shown that MMPs not only remodel the ECM but may also be important in a number of other physiologic processes: a. MMPs may affect cell migration by changing cells from an adhesive to a non-adhesive phenotype and/or by ECM degradation, b. MMPs may alter the micro-environment of the ECM leading to alterations in cell behaviour (proliferation, differentiation, organization, survival), c. MMPs may modulate the activity of growth factors or their receptors by releasing them from the ECM, and d. MMPs themselves may alter the balance of other protease activities by cleaving these enzymes and/or their receptors 68. It has to be emphasized that most data concerning the roles of MMPs are derived from in-vitro experiments. In vivo experiments may well yield different results implicating compensatory mechanisms.

II. Pathologic The roles of MMPs have been described in many diseases associated with an unbalanced degradation of ECM such as in rheumatic diseases but also in non-rheumatic diseases such as cardiovascular disease, cancer, lung-, inflammatory bowel-, and renal disease.

Chapter 2

34

A. Non-rheumatic diseases Cardiac and arterial remodeling by MMPs occurs in various cardiovascular diseases such as abdominal aortic aneurysms 69, atherosclerosis 70, and heart failure 71. Several preclinical studies with MMP-inhibitors in animal models with heart failure have shown less damage of matrix collagen, favourable matrix remodeling, and improved cardiac structure and function 71.

In tumour invasion at least three steps are of importance. First the affinity of cells either for each other or for the ECM should decrease in order to allow cell release from the primary tumour. Second, the surrounding ECM should be remodeled by the production of proteolytic enzymes in order to allow the third step, cell migration. There is considerable evidence to suggest that MMPs are not only involved in this local invasion but also in tumour growth (by altering the local environment, e.g. by allowing the access or release of growth factors into or from the ECM) and in the metastatic process 72. The increased MMP-expression in host stromal cells rather than in tumour cells themselves underlines the importance of tumour-stroma interactions 73.

In view of the role of MMPs in cancer the inhibition of their activity is a novel target for the treatment of malignancies. Several synthetic MMP-inhibitors have been developed and tested in phase I, II, and III clinical trials and have been reviewed elsewhere 74,75.

In various other pathological conditions the MMPs seem to play important roles such as in lung disease 76, neurological diseases 77,78, and in gastrointestinal disease 79. B. Rheumatic diseases, non-RA Due to the important role of MMPs in matrix remodeling and degradation, many studies “on MMPs” have been reported in rheumatoid arthritis and to a lesser extent in other rheumatic conditions such as osteoarthritis, SLE, scleroderma, psoriatic arthritis, ankylosing spondylitis, crystal arthropathies, etc. In many of these studies systemic levels of MMPs were evaluated, in particular MMP-3.

Osteoarthritis (OA) is a slowly progressive disorder characterized by destruction of joint cartilage and subchondral bone, with “mild” inflammatory alterations. The MMPs are thought to play an important role in the degradation of the ECM in OA 17. Several studies have shown an increase in MMP gene expression 80-82 and MMP protein production 83,84 as well as in collagenase-mediated type II collagen degradation products 85 in tissues of OA patients. In synovial fluid of the knee of patients with OA several MMPs can be detected with a predominance of MMP-1, MMP-2, and in particular MMP-3 86. In the


35

systemic circulation of OA patients MMP levels are generally low or undetectable. Some studies report elevated systemic MMP-3 levels 87-89 whereas other investigators found no differences compared to healthy controls 90,91. In a subgroup of patients with rapidly destructive osteoarthritis of the hip elevated systemic levels of MMP-3 and MMP-9 have been reported 92. Although there is consensus that these enzymes do contribute to the degenerative changes, their relative role is still uncertain 85.

In SLE systemic MMP-3 levels have been reported to be elevated but serial measurements did not reveal a relationship to disease activity 93. On the other hand in a study of Kotajima et al, patients with clinical and serological features of lupus nephritis had significantly increased serum MMP-3 levels 94. Since arthritis is less frequent in SLE MMP-3 might be produced extra-articularly, possibly in the kidney.

In scleroderma both normal and elevated systemic MMP-1 and MMP-3 levels have been reported. In most studies TIMP-1 levels were increased 95,96. This suggests that tissue matrix accumulation in scleroderma may be due to increased levels of inhibitors rather than to decreased MMPs 73.

In several other rheumatic diseases such as psoriatic arthritis 91,97, ankylosing spondylitis97, and crystal arthritis 89,91 elevated systemic MMP-3 levels have been reported mainly in view of the comparison with levels primarily determined in RA patients. MMPs IN RHEUMATOID ARTHRITIS I. Introduction RA is characterized by chronic inflammation of synovial tissue and in most cases progressive destruction of cartilage and bone 98. Progressive joint destruction is one of the strongest predictors of long-term outcome and disability in RA 10. The MMPs are considered to be key mediators in this destructive process. The principle targets of destruction are interstitial collagens, and secreted collagenases (MMP-1, -8 and -13) have a major role in this process 37. These three collagenases are upregulated and found locally in inflamed synovial tissue 99,100, chondrocytes 101, or synovial fluid 86 of inflamed joints. Besides these collagenases, MMP-3 has been mentioned as an important mediator in matrix degradation 102. II. Role of MMP-3 MMP-3 is thought to play a prominent role in the pathogenesis of matrix degradation 12,17,102 even though it is not the only key enzyme 103. MMP-3 is

Chapter 2

36

capable of degrading many components of the matrix in the synovial joint including proteoglycans, gelatins, laminin, fibronectin, and collagen III, IV, and IX 17,102,104,105. Moreover, MMP-3 is able to activate other matrix metalloprotein-ases such as MMP-1, MMP-7, MMP-8, MMP-9, and MMP-13 62.

MMP-3 is produced abundantly in the inflamed joints of RA patients. The enzyme has been localized in the fibroblast-like synoviocytes of rheumatoid synovium 106-108, in RA cartilage 83,109, at sites of cartilage erosion 101, in synovial fluid 85,87,109-114, and in the systemic circulation 88-90,97,116,117.

MMP-3 is locally produced and activated within the inflamed joint and released into the synovial fluid and systemic circulation. Systemic MMP-3 levels are considered to be a reflection of local synthesis. As such, serum MMP-3 can be used as a systemic marker of local joint inflammation 88,97,112,117,118 and/or destruction 90,119,120. Therefore serum MMP-3 may reflect joint inflammation and destruction more directly compared to C-reactive protein (CRP) which is produced indirectly by the liver after cytokine stimulation 121. This difference in marker characteristics can be of importance for new therapies. For example specific MMP-inhibitors may uncouple the relation between surrogate markers of joint inflammation (such as CRP) and joint damage. Radiological progression might be stopped by such agents without inhibition of inflammation and the acute phase response. In that case new markers of joint destruction, such as MMPs may become essential. III. Future perspectives, MMP-inhibition Despite optimal care and recent new therapeutic advances in RA, such as TNF-α blocking agents and IL-1 receptor antagonist, it is often not possible to completely stop joint destruction. In RA and OA progressive destruction of cartilage and bone is considered to be a consequence of a dysbalance between an excess of activated MMPs and inadequate levels of their inhibitors. Therapeutic interventions might be focused on a. inhibition of the synthesis and release of proMMPs, b. inhibition of the activation of proMMPs, c. inhibition of activated MMPs, d. stimulation of TIMP synthesis or e. inhibition of TIMP inactivation. One of the key requirements in developing specific MMP-inhibitors is to determine in vivo the importance of individual MMPs in the pathogenesis of arthritis. Two experimental approaches have been used: a. knockout and transgenic mouse technologies and b. effects of MMP-inhibitors in cultured cells and animal models. The MMP-3 knockout mice show a normal development and demonstrate in the collagen-induced arthritis model, similar severity of arthritis and cartilage


37

degradation as wild type animals 103. MMP-2 deficient mice also develop normally but display reduced angiogenesis and tumour progression 122. In an antibody-induced arthritis model the MMP-2 knockout mice exhibited both a clinically and histologically severe arthritis when compared to wild type mice or MMP-9 deficient animals 123. MMP-9 deficient mice exhibit an abnormal vascularization and ossification of the skeletal growth plate but will eventually develop normally 124. The mild phenotypes of these knockout mice suggest that these individual MMPs (MMP-2, -3, -7, -9, and -12) are not critical during normal development and can be compensated for by other MMP family members 125. The MT1-MMP knockout mice display a severe phenotype with dwarfism, osteopenia and arthritis 125, which suggests severe side effects to be expected in case of specific inhibition. In a transgenic mouse model overexpression of a constitutively active form of the MMP-13 gene induced cartilage erosions resembling human OA due to collagen and proteoglycan degradation 126. In addition to these knockout and transgenic studies both broad-spectrum and more specific MMP-inhibitors have provided insight into the role of individual MMPs in joint destruction. The importance of collagenases in RA and OA has been demonstrated with Ro-32-3555 (TrocadeTM) a compound, which ”selectively” inhibits MMP-1, -8, and -13. In a Propionibacterium acnes rat model Ro-32-3555 did block cartilage degradation despite continuing synovitis 127. In the STR/ORT mouse model of OA it effectively inhibited cartilage degradation, joint space narrowing and osteophyte formation 128. These data provide evidence for an important role of MMPs in joint destruction and the potential effects of MMP-inhibitors. However, it also highlights the fact that these compounds may not affect the inflammatory process itself, which would have to be controlled by additional treatments. Over the last 10-15 years a number of MMP-inhibitors have been developed. In addition to the rheumatic diseases, the main therapeutic focus has been directed to the treatment of cancer. The different compounds have been extensively reviewed elsewhere 74. Treatment strategies are now focused on inhibition of MMP activity and/or MMP gene expression 129. A. Inhibition of MMP activity Originally MMP-inhibitors were designed to mimic peptide sequences surrounding the point in the collagen molecule first cleaved by collagenases. These “peptomimics” fit into the active sites of MMPs and chelate the zinc atom

Chapter 2

38

through a zinc-binding group. A major drawback of the approach is that these agents inhibit a broad spectrum of MMPs and are therefore likely to cause unwanted side effects. For example Marimastat®, a broad spectrum MMP-inhibitor has been shown to improve the survival of patients with gastric cancer 130 but the drug caused musculoskeletal pains, mainly in shoulders and hands as a side effect, possibly caused by inhibiting physiological levels of other MMPs. Despite great efforts in this therapeutic approach the applicability of MMP inhibition as treatment of RA or OA is still undetermined. Ro-32-3555 (TrocadeTM), a hydroxamic acid that “selectively” inhibits the collagenases and has little activity against other MMPs is the only MMP-inhibitor having completely entered clinical trials to assess efficacy. However, despite good preclinical data 131 the radiographic scores of patients with RA did not differ from placebo and the study was terminated 132. Currently the only MMP inhibitor on the market is the tetracycline periostat, which is licensed for periodontal disease. Tetracyclines are able to suppress MMP activity in e.g. adjuvant arthritis 133. In patients with joint disease a significant reduction in MMP activity within the joints is observed after oral minocyclin 134 or doxycycline 135. Clinical trials with tetracyclines in RA show promising results with respect to disease activity 136 but failed to show clear effects on radiological progression 137. Insufficient dosing and residual proteolytic activity, not inhibited by tetracyclines, are reported as possible explanations 138. In addition several other issues are of importance. In the first place almost all “specific” MMP inhibitors do not influence the inflammatory components of a disease such as RA 127,132. Consequently patients still require their background drug treatment. This might have consequences, not only for compliance (the patient feels no short-term benefit), but also for the evaluation of additional effects of MMP-inhibitors. Differences between groups will become smaller, especially when more effective therapies such as anti-TNF-α are used. Probably large long-term trials will be necessary with patients on a ”stable” DMARD regime, with prescribing guidelines, and stratification. Secondly it is of importance to consider the outcome and the assessment of the outcome. Although prevention of radiologically detectable damage is an established goal, there is still the question how to translate this into clinical benefit for the individual patient in terms of disability 132. Furthermore new DMARDs, such as anti-TNF-α are very successful in inhibiting radiological progression in many cases. In that situation an additional effect of a specific MMP-inhibitor is difficult to detect.


39

In the third place safety issues have already been mentioned. Several broad-spectrum MMP inhibitors were complicated by musculoskeletal pains, probably due to inhibition of other MMPs. Moreover, because of the role of MMPs in physiological processes such as embryonic development, tissue growth, and morphogenesis MMP inhibitors are likely to be teratogenic. B. TACE (TNF-α converting enzyme) Another type of inhibition of MMP activity is the interference with TACE. This enzyme, an ADAM, releases the soluble TNF-α from its membrane and thereby enables the cytokine to act in a paracrine and endocrine manner. TACE is over-expressed in synovial tissue of patients with RA 139 and its inhibition has been shown to be efficacious in experimental arthritis 140. Phase I studies, of TACE inhibitors have been performed and results from Phase II studies are awaited 141. C. Inhibition of MMP gene expression Due to the difficulties encountered with MMP active site inhibitors, other approaches are currently evaluated. A more selective inhibition of inducible pathological gene expression, without affecting constitutive physiological gene expression is such a line. In particular genes coding for cytokines, prostaglandins and MMPs are of interest. Selective inhibition of inducible MMP genes in RA without affecting constitutive, physiological MMP expression, could have major advantages with regard to potential side effects.

Recent studies have focused at the signal transduction pathways. Blockade of the MAPK and/or NFκB pathways leading to inhibition of MMP expression are effective in tissue culture experiments and in animal models of arthritis 129,142,143. Clinical studies targeted at signal transduction pathways are currently evolving 141.

By the way, since multiple steps and pathways contribute to MMP synthesis it is important to emphasize that therapies targeted at one step may be only partially effective. Moreover, many intracellular signaling enzymes are not completely specific for one signaling cascade, a phenomenon which bears the risk of a broader spectrum of side effects.

Non-specific inhibitors of MMP gene expression are already used in the treatment of arthritis. Corticosteroids block gene transcription by direct binding to glucocorticoid responsive elements in gene promoters or by interference with transcription factors such as NFκB and AP-1. However, corticosteroids are not specific. They inhibit transcription of many other genes and the frequently occurring side effects associated with these drugs suggest the need for more

Chapter 2

40

specific MMP gene expression inhibitors. D. Biologicals Another way to reduce MMP levels is by direct blockade of inflammatory cytokines such as TNF-α 144-146. Anti-TNF-α therapies reduce inflammation and presumably also reduce MMP synthesis, resulting in reduced joint destruction 146-

148. Radiological progression is even absent in infliximab treated patients without a clinical response 147, an argument for a possible dissociation of the two pathophysiologic mechanisms of joint inflammation and articular destruction 149.

In the future, the concept of combination therapy may become more “evidence based” focused on specific signal transduction pathways or genes. In that context it is essential to realize that further research with respect to markers of disease activity and destruction is of great importance. Not only because specific MMP inhibition might result in an uncoupling of inflammation and destruction but also because currently used parameters of inflammation such as the acute phase proteins such as CRP are mediated by similar signal transduction pathways 150. ACKNOWLEDGEMENTS This work was supported by a grant from the “Dutch Arthritis Association”, The Netherlands.


41

REFERENCES 1. Harris ED, Jr. Clinical Features of Rheumatoid Arthritis. In: Ruddy S, Harris-ED J, Sledge CB,

Budd RC, Sergent JS, editors. Kelley's Textbook of rheumatology. Philadelphia: Sanders,W.B., 2001:967-1000.

2. Firestein GS. Etiology and Pathogenesis of Rheumatoid Arthritis. In: S., Harris-E, Jr, Sledge, B., Budd et al., editors. Kelley's Textbook of rheumatology. Philadelphia: Sanders,W.B., 2001:921-66.

3. Firestein GS, Zvaifler NJ. How important are T cells in chronic rheumatoid synovitis?: II. T cell-independent mechanisms from beginning to end. Arthritis Rheum 2002;46:298-308.

4. Zvaifler NJ, Tsai V, Alsalameh S, von Kempis J, Firestein GS, Lotz M. Pannocytes: distinctive cells found in rheumatoid arthritis articular cartilage erosions. Am J Pathol 1997;150:1125-38.

5. Xue C, Takahashi M, Hasunuma T, Aono H, Yamamoto K, Yoshino S et al. Characterisation of fibroblast-like cells in pannus lesions of patients with rheumatoid arthritis sharing properties of fibroblasts and chondrocytes. Ann Rheum Dis 1997;56:262-7.

6. Heinegård D, Lorenzo P, Saxne T. Matrix Glycoproteins, Proteoglycans, and Cartilage. In: Ruddy S, Harris-ED J, Sledge CB, Budd RC, Sergent JS, editors. Kelley's Textbook of rheumatology. Philadelphia: Sanders,W.B., 2001:41-53.

7. Eyre D. Collagen of articular cartilage. Arthritis Res 2002;4:30-5. 8. Nagase H, Kashiwagi M. Aggrecanases and cartilage matrix degradation. Arthritis Res 2003;5:94-

103. 9. Mort JS, Billington CJ. Articular cartilage and changes in arthritis: matrix degradation. Arthritis

Res 2001;3:337-41. 10. Kirwan JR. Links between radiological change, disability, and pathology in rheumatoid arthritis. J

Rheumatol 2001;28:881-6. 11. Caterson B, Flannery CR, Hughes CE, Little CB. Mechanisms involved in cartilage proteoglycan

catabolism. Matrix Biol 2000;19:333-44. 12. Okada Y. Proteinases and matrix degradation. In: Ruddy S, Harris ED, Sledge CB, Budd RC,

Sergent JS, editors. Kelley's Textbook of rheumatology. Philadelphia: Sanders,W.B., 2001:55-72. 13. Nagase H, Woessner-JF J. Matrix metalloproteinases. J Biol Chem 1999;274:21491-4. 14. Goldring SR. Pathogenesis of bone and cartilage destruction in rheumatoid arthritis.

Rheumatology (Oxford) 2003;42 Suppl 2:ii11-6. 15. Gravallese EM, Manning C, Tsay A, Naito A, Pan C, Amento E et al. Synovial tissue in

rheumatoid arthritis is a source of osteoclast differentiation factor. Arthritis Rheum 2000;43:250-8. 16. Teitelbaum SL. Bone resorption by osteoclasts. Science 2000;289:1504-8. 17. Nagase H, Okada Y. Proteinases and matrix degradation. In: Kelley WN, Harris ED, Ruddy S,

Sledge CB, editors. Textbook of rheumatology. Philadelphia: Saunders,W.B., 1997:323-41. 18. Visse R, Nagase H. Matrix metalloproteinases and tissue inhibitors of metalloproteinases:

structure, function, and biochemistry. Circ Res 2003;92:827-39. 19. Murphy G, Knauper V, Atkinson S, Butler G, English W, Hutton M et al. Matrix

metalloproteinases in arthritic disease. Arthritis Res 2002;4 Suppl 3:S39-49. 20. Saus J, Quinones S, Otani Y, Nagase H, Harris ED, Jr., Kurkinen M. The complete primary

structure of human matrix metalloproteinase- 3. Identity with stromelysin. J Biol Chem 1988;263:6742-5.

21. Billinghurst RC, Dahlberg L, Ionescu M, Reiner A, Bourne R, Rorabeck C et al. Enhanced cleavage of type II collagen by collagenases in osteoarthritic articular cartilage. J Clin Invest 1997;99:1534-45.

22. Sandy JD, Neame PJ, Boynton RE, Flannery CR. Catabolism of aggrecan in cartilage explants. Identification of a major cleavage site within the interglobular domain. J Biol Chem 1991;266:8683-5.

Chapter 2

42

23. Tortorella MD, Burn TC, Pratta MA, Abbaszade I, Hollis JM, Liu R et al. Purification and cloning of aggrecanase-1: a member of the ADAMTS family of proteins. Science 1999;284:1664-6.

24. Kuno K, Matsushima K. ADAMTS-1 protein anchors at the extracellular matrix through the thrombospondin type I motifs and its spacing region. J Biol Chem 1998;273:13912-7.

25. Lark MW, Bayne EK, Flanagan J, Harper CF, Hoerrner LA, Hutchinson NI et al. Aggrecan degradation in human cartilage. Evidence for both matrix metalloproteinase and aggrecanase activity in normal, osteoarthritic, and rheumatoid joints. J Clin Invest 1997;100:93-106.

26. Lohmander LS, Neame PJ, Sandy JD. The structure of aggrecan fragments in human synovial fluid. Evidence that aggrecanase mediates cartilage degradation in inflammatory joint disease, joint injury, and osteoarthritis. Arthritis Rheum 1993;36):1214-22.

27. Little CB, Flannery CR, Hughes CE, Mort JS, Roughley PJ, Dent C et al. Aggrecanase versus matrix metalloproteinases in the catabolism of the interglobular domain of aggrecan in vitro. Biochem J 1999;344:61-8.

28. Vincenti MP. The matrix metalloproteinase (MMP) and tissue inhibitor of metalloproteinase (TIMP) genes. Transcriptional and posttranscriptional regulation, signal transduction and cell-type-specific expression. Methods Mol Biol 2001;151:121-48.

29. Chen F, Castranova V, Shi X, Demers LM. New insights into the role of nuclear factor-kappaB, a ubiquitous transcription factor in the initiation of diseases. Clin Chem 1999;45:7-17.

30. Tak PP, Firestein GS. NF-kappaB: a key role in inflammatory diseases. J Clin Invest 2001;107:7-11.

31. Malinin NL, Boldin MP, Kovalenko AV, Wallach D. MAP3K-related kinase involved in NF-kappaB induction by TNF, CD95 and IL-1. Nature 1997;385:540-4.

32. Lee FS, Hagler J, Chen ZJ, Maniatis T. Activation of the IkappaB alpha kinase complex by MEKK1, a kinase of the JNK pathway. Cell 1997;88:213-22.

33. Vincenti MP, Coon CI, Brinckerhoff CE. Nuclear factor kappaB/p50 activates an element in the distal matrix metalloproteinase 1 promoter in interleukin-1beta-stimulated synovial fibroblasts. Arthritis Rheum 1998;41:1987-94.

34. Kirstein M, Sanz L, Quinones S, Moscat J, Diaz-Meco MT, Saus J. Cross-talk between different enhancer elements during mitogenic induction of the human stromelysin-1 gene. J Biol Chem 1996;271:18231-6.

35. Yokoo T, Kitamura M. Dual regulation of IL-1 beta-mediated matrix metalloproteinase-9 expression in mesangial cells by NF-kappa B and AP-1. Am J Physiol 1996;270:F123-30.

36. Garrington TP, Johnson GL. Organization and regulation of mitogen-activated protein kinase signaling pathways. Curr Opin Cell Biol 1999;11:211-8.

37. Vincenti MP, Brinckerhoff CE. Transcriptional regulation of collagenase (MMP-1, MMP-13) genes in arthritis: integration of complex signaling pathways for the recruitment of gene-specific transcription factors. Arthritis Res 2002;4:157-64.

38. Li WQ, Dehnade F, Zafarullah M. Oncostatin M-induced matrix metalloproteinase and tissue inhibitor of metalloproteinase-3 genes expression in chondrocytes requires Janus kinase/STAT signaling pathway. J Immunol 2001;166:3491-8.

39. Firestein GS, Manning AM. Signal transduction and transcription factors in rheumatic disease. Arthritis Rheum 1999;42:609-21.

40. Ye S. Polymorphism in matrix metalloproteinase gene promoters: implication in regulation of gene expression and susceptibility of various diseases. Matrix Biol 2000;19:623-29.

41. Collins FS, Brooks LD, Chakravarti A. A DNA Polymorphism Discovery Resource for Research on Human Genetic Variation. Genome Res 1998;8:1229.

42. Rutter JL, Mitchell TI, Buttice G, Meyers J, Gusella JF, Ozelius LJ et al. A single nucleotide polymorphism in the matrix metalloproteinase-1 promoter creates an Ets binding site and augments transcription. Cancer Res 1998;58:5321-5.


43

43. Kanamori Y, Matsushima M, Minaguchi T, Kobayashi K, Sagae S, Kudo R et al. Correlation between expression of the matrix metalloproteinase-1 gene in ovarian cancers and an insertion/deletion polymorphism in its promoter region. Cancer Res 1999; 59:4225-7.

44. Ye S, Dhillon S, Turner SJ, Bateman AC, Theaker JM, Pickering RM et al. Invasiveness of cutaneous malignant melanoma is influenced by matrix metalloproteinase 1 gene polymorphism. Cancer Res 2001;61:1296-8.

45. Constantin A, Lauwers-Cances V, Navaux F, Abbal M, van MJ, Mazieres B et al. Collagenase-1 (MMP-1) and HLA-DRB1 gene polymorphisms in rheumatoid arthritis: a prospective longitudinal study. J Rheumatol 2002;29:15-20.

46. Ye S, Eriksson P, Hamsten A, Kurkinen M, Humphries SE, Henney AM. Progression of coronary atherosclerosis is associated with a common genetic variant of the human stromelysin-1 promoter which results in reduced gene expression. J Biol Chem 1996;271:13055-60.

47. Borghaei RC, Sullivan C, Mochan E. Identification of a cytokine-induced repressor of interleukin-1 stimulated expression of stromelysin 1 (MMP-3). J Biol Chem 1999;274:2126-31.

48. Terashima M, Akita H, Kanazawa K, Inoue N, Yamada S, Ito K et al. Stromelysin promoter 5A/6A polymorphism is associated with acute myocardial infarction. Circulation 1999;99:2717-9.

49. Yamada Y, Izawa H, Ichihara S, Takatsu F, Ishihara H, Hirayama H et al. Prediction of the risk of myocardial infarction from polymorphisms in candidate genes. N Engl J Med 2002;347:1916-23.

50. Yoon S, Tromp G, Vongpunsawad S, Ronkainen A, Juvonen T, Kuivaniemi H. Genetic analysis of MMP3, MMP9, and PAI-1 in Finnish patients with abdominal aortic or intracranial aneurysms. Biochem Biophys Res Commun 1999;265:563-8.

51. Ye S, Watts GF, Mandalia S, Humphries SE, Henney AM. Preliminary report: genetic variation in the human stromelysin promoter is associated with progression of coronary atherosclerosis. Br Heart J 1995;73:209-15.

52. Ghilardi G, Biondi ML, Caputo M, Leviti S, DeMonti M, Guagnellini E et al. A single nucleotide polymorphism in the matrix metalloproteinase-3 promoter enhances breast cancer susceptibility. Clin Cancer Res 2002;8:3820-3.

53. Hinoda Y, Okayama N, Takano N, Fujimura K, Suehiro Y, Hamanaka Y et al. Association of functional polymorphisms of matrix metalloproteinase (MMP)-1 and MMP-3 genes with colorectal cancer. Int J Cancer 2002;102:526-9.

54. Constantin A, Lauwers-Cances V, Navaux F, Abbal M, van Meerwijk J, Mazieres B et al. Stromelysin 1 (matrix metalloproteinase 3) and HLA-DRB1 gene polymorphisms: Association with severity and progression of rheumatoid arthritis in a prospective study. Arthritis Rheum 2002;46:1754-62.

55. Keyszer G, Doerr S, Lechtenboehmer N, Rau R, Herborn G, Keysser M et al. Haplotype analysis of matrix metalloproteinase 1 (MMP-1) and MMP-3 alleles in rheumatoid arthritis: association of the 1G/5A haplotype with radiological progression. Arthritis Rheum 2002;46:S97.

56. Clark A. Post-transcriptional regulation of pro-inflammatory gene expression. Arthritis Res 2000;2:172-4.

57. Chen CY, Shyu AB. AU-rich elements: characterization and importance in mRNA degradation. Trends Biochem Sci 1995;20:465-70.

58. Kontoyiannis D, Pasparakis M, Pizarro TT, Cominelli F, Kollias G. Impaired on/off regulation of TNF biosynthesis in mice lacking TNF AU-rich elements: implications for joint and gut-associated immunopathologies. Immunity 1999;10:387-98.

59. Kotlyarov A, Neininger A, Schubert C, Eckert R, Birchmeijer C, Volk HD et al. MAPKAP kinase 2 is essential for LPS-induced TNF alpha biosynthesis. Nat Cell Biol 1999;1:94-97.

60. Brook M, Sully G, Clark AR, Saklatvala J. Regulation of tumour necrosis factor alpha mRNA stability by the mitogen-activated protein kinase p38 signalling cascade. FEBS Lett 2000;483:57-61.

Chapter 2

44

61. Winzen R, Kracht M, Ritter B, Wilhelm A, Chen CY, Shyu AB et al. The p38 MAP kinase pathway signals for cytokine-induced mRNA stabilization via MAP kinase-activated protein kinase 2 and an AU-rich region-targeted mechanism. EMBO J 1999;18:4969-80.

62. Nagase H. Activation mechanisms of matrix metalloproteinases. Biol Chem 1997;378:151-60. 63. Brew K, Dinakarpandian D, Nagase H. Tissue inhibitors of metalloproteinases: evolution,

structure and function. Biochim Biophys Acta 2000;1477:267-83. 64. Hashimoto G, Aoki T, Nakamura H, Tanzawa K, Okada Y. Inhibition of ADAMTS4

(aggrecanase-1) by tissue inhibitors of metalloproteinases (TIMP-1, 2, 3 and 4). FEBS Lett 2001;494:192-5.

65. Kashiwagi M, Tortorella M, Nagase H, Brew K. TIMP-3 is a potent inhibitor of aggrecanase 1 (ADAM-TS4) and aggrecanase 2 (ADAM-TS5). J Biol Chem 2001;276:12501-4.

66. Barrett AJ. Alpha 2-macroglobulin. Methods Enzymol 1981;80:737-54. 67. Gross J, Lapiere CM. Collagenolytic activity in amphibian tissues: a tissue culture assay. Proc Natl

Acad Sci U S A 1962;48:1014-22. 68. Vu TH, Werb Z. Matrix metalloproteinases: effectors of development and normal physiology.

Genes Dev 2000;14:2123-33. 69. Carrell TW, Burnand KG, Wells GM, Clements JM, Smith A. Stromelysin-1 (matrix

metalloproteinase-3) and tissue inhibitor of metalloproteinase-3 are overexpressed in the wall of abdominal aortic aneurysms. Circulation 2002;105:477-82.

70. Galis ZS, Khatri JJ. Matrix metalloproteinases in vascular remodeling and atherogenesis: the good, the bad, and the ugly. Circ Res 2002;90:251-62.

71. Li YY, Feldman AM. Matrix metalloproteinases in the progression of heart failure: potential therapeutic implications. Drugs 2001;61:1239-52.

72. Cox G, O'Byrne KJ. Matrix metalloproteinases and cancer. Anticancer Res 2001;21:4207-19. 73. Chung AS, Yoon SO, Park SJ, Yun CH. Roles of matrix metalloproteinases in tumor metastasis

and angiogenesis. J Biochem Mol Biol 2003;36:128-37. 74. Brown PD. Ongoing trials with matrix metalloproteinase inhibitors. Expert Opin Investig Drugs

2000;9:2167-77. 75. Overall CM, Lopez-Otin C. Strategies for MMP inhibition in cancer: innovations for the post-trial

era. Nat Rev Cancer 2002;2:657-72. 76. Ohbayashi H. Matrix metalloproteinases in lung diseases. Curr Protein Pept Sci 2002;3:409-21. 77. Yong VW, Krekoski CA, Forsyth PA, Bell R, Edwards DR. Matrix metalloproteinases and

diseases of the CNS. Trends Neurosci 1998;21:75-80. 78. Rosenberg GA. Matrix metalloproteinases in neuroinflammation. Glia 2002;39:279-91. 79. von Lampe B, Barthel B, Coupland SE, Riecken EO, Rosewicz S. Differential expression of

matrix metalloproteinases and their tissue inhibitors in colon mucosa of patients with inflammatory bowel disease. Gut 2000;47:63-73.

80. Wolfe GC, MacNaul KL, Buechel FF, McDonnell J, Hoerrner LA, Lark MW et al. Differential in vivo expression of collagenase messenger RNA in synovium and cartilage. Quantitative comparison with stromelysin messenger RNA levels in human rheumatoid arthritis and osteoarthritis patients and in two animal models of acute inflammatory arthritis. Arthritis Rheum 1993;36:1540-7.

81. Mitchell PG, Magna HA, Reeves LM, Lopresti-Morrow LL, Yocum SA, Rosner PJ et al. Cloning, expression, and type II collagenolytic activity of matrix metalloproteinase-13 from human osteoarthritic cartilage. J Clin Invest 1996;97:761-8.

82. Freemont AJ, Hampson V, Tilman R, Goupille P, Taiwo Y, Hoyland JA. Gene expression of matrix metalloproteinases 1, 3, and 9 by chondrocytes in osteoarthritic human knee articular cartilage is zone and grade specific. Ann Rheum Dis 1997;56:542-9.


45

83. Martel Pelletier J, McCollum R, Fujimoto N, Obata K, Cloutier JM, Pelletier JP. Excess of metalloproteases over tissue inhibitor of metalloprotease may contribute to cartilage degradation in osteoarthritis and rheumatoid arthritis. Lab Invest 1994;70:807-15.

84. Tetlow LC, Adlam DJ, Woolley DE. Matrix metalloproteinase and proinflammatory cytokine production by chondrocytes of human osteoarthritic cartilage: associations with degenerative changes. Arthritis Rheum 2001;44:585-94.

85. Freemont AJ, Byers RJ, Taiwo YO, Hoyland JA. In situ zymographic localisation of type II collagen degrading activity in osteoarthritic human articular cartilage. Ann Rheum Dis 1999;58:357-65.

86. Yoshihara Y, Nakamura H, Obata K, Yamada H, Hayakawa T, Fujikawa K et al. Matrix metalloproteinases and tissue inhibitors of metalloproteinases in synovial fluids from patients with rheumatoid arthritis or osteoarthritis. Ann Rheum Dis 2000;59:455-61.

87. Manicourt DH, Fujimoto N, Obata K, Thonar EJ. Serum levels of collagenase, stromelysin-1, and TIMP-1. Age- and sex-related differences in normal subjects and relationship to the extent of joint involvement and serum levels of antigenic keratan sulfate in patients with osteoarthritis. Arthritis Rheum 1994;37:1774-83.

88. Yoshihara Y, Obata K, Fujimoto N, Yamashita K, Hayakawa T, Shimmei M. Increased levels of stromelysin-1 and tissue inhibitor of metalloproteinases-1 in sera from patients with rheumatoid arthritis. Arthritis Rheum 1995;38:969-75.

89. Zucker S, Lysik RM, Zarrabi MH, Greenwald RA, Gruber B, Tickle SP et al. Elevated plasma stromelysin levels in arthritis. J Rheumatol 1994;21:2329-33.

90. Posthumus MD, Limburg PC, Westra J, van Leeuwen MA, van Rijswijk MH. Serum levels of matrix metalloproteinase-3 in relation to the development of radiological damage in patients with early rheumatoid arthritis. Rheumatology 1999;38:1081-7.

91. Ribbens C, Porras M, Franchimont N, Kaiser MJ, Jaspar JM, Damas P et al. Increased matrix metalloproteinase-3 serum levels in rheumatic diseases: relationship with synovitis and steroid treatment. Ann Rheum Dis 2002;61:161-6.

92. Masuhara K, Nakai T, Yamaguchi K, Yamasaki S, Sasaguri Y. Significant increases in serum and plasma concentrations of matrix metalloproteinases 3 and 9 in patients with rapidly destructive osteoarthritis of the hip. Arthritis Rheum 2002;46:2625-31.

93. Zucker S, Mian N, Drews M, Conner C, Davidson A, Miller F et al. Increased serum stromelysin-1 levels in systemic lupus erythematosus: lack of correlation with disease activity. J Rheumatol 1999;26:78-80.

94. Kotajima L, Aotsuka S, Fujimani M, Okawa-Takatsuji M, Kinoshita M, Sumiya M et al. Increased levels of matrix metalloproteinase-3 in sera from patients with active lupus nephritis. Clin Exp Immunol 1998;16:409-15.

95. Toubi E, Kessel A, Grushko G, Sabo E, Rozenbaum M, Rosner I. The association of serum matrix metalloproteinases and their tissue inhibitor levels with scleroderma disease severity. Clin Exp Rheumatol 2002;20:221-4.

96. Young-Min SA, Beeton C, Laughton R, Plumpton T, Bartram S, Murphy G et al. Serum TIMP-1, TIMP-2, and MMP-1 in patients with systemic sclerosis, primary Raynaud's phenomenon, and in normal controls. Ann Rheum Dis 2001;60:846-51.

97. Keyszer G, Lambiri I, Nagel R, Keysser C, Keysser M, Gromnica-Ihle E et al. Circulating levels of matrix metalloproteinases MMP-3 and MMP-1, tissue inhibitor metalloproteinases 1 (TIMP-1), and MMP-1/TIMP-1 complex in rheumatic disease. Correlation with clinical activity of rheumatoid arthritis versus other surrogate markers. J Rheumatol 1999;26:251-8.

98. Choy EH, Panayi GS. Cytokine pathways and joint inflammation in rheumatoid arthritis. N Engl J Med 2001;344:907-16.

Chapter 2

46

99. Konttinen YT, Ainola M, Valleala H, Ma J, Ida H, Mandelin J et al. Analysis of 16 different matrix metalloproteinases (MMP-1 to MMP-20) in the synovial membrane: different profiles in trauma and rheumatoid arthritis. Ann Rheum Dis 1999;58:691-7.

100. Jain A, Brennan F, Troeberg L, Nanchahal J. The role of matrix metalloproteinases in rheumatoid tendon disease. J Hand Surg [Am] 2002;27:1059-64.

101. Shlopov BV, Lie WR, Mainardi CL, Cole AA, Chubinskaya S, Hasty KA. Osteoarthritic lesions: involvement of three different collagenases. Arthritis Rheum 1997;40:2065-74.

102. Hasty KA, Reife RA, Kang AH, Stuart JM. The role of stromelysin in the cartilage destruction that accompanies inflammatory arthritis. Arthritis Rheum 1990;33:388-97.

103. Mudgett JS, Hutchinson NI, Chartrain NA, Forsyth AJ, McDonnell J, Singer II et al. Susceptibility of stromelysin 1-deficient mice to collagen-induced arthritis and cartilage destruction. Arthritis Rheum 1998;41:110-21.

104. Singer II, Kawka DW, Bayne EK, Donatelli SA, Weidner JR, Williams HR et al. VDIPEN, a metalloproteinase-generated neoepitope, is induced and immunolocalized in articular cartilage during inflammatory arthritis. J Clin Invest 1995;95:2178-86.

105. Wu JJ, Lark MW, Chun LE, Eyre DR. Sites of stromelysin cleavage in collagen types II, IX, X, and XI of cartilage. J Biol Chem 1991;266:5625-8.

106. McCachren SS. Expression of metalloproteinases and metalloproteinase inhibitor in human arthritic synovium. Arthritis Rheum 1991;34:1085-93.

107. Okada Y, Takeuchi N, Tomita K, Nakanishi I, Nagase H. Immunolocalization of matrix metalloproteinase 3 (stromelysin) in rheumatoid synovioblasts (B cells): correlation with rheumatoid arthritis. Ann Rheum Dis 1989;48:645-53.

108. Tetlow LC, Lees M, Woolley DE. Comparative studies of collagenase and stromelysin-1 expression by rheumatoid synoviocytes in vitro. Virchows Arch 1995;425:569-76.

109. Martel Pelletier J, Cloutier JM, Howell DS, Pelletier JP. Human rheumatoid arthritic cartilage and its neutral proteoglycan-degrading proteases. The effects of antirheumatic drugs. Arthritis Rheum 1985;28:405-12.

110. Ishiguro N, Ito T, Miyazaki K, Iwata H. Matrix metalloproteinases, tissue inhibitors of metalloproteinases, and glycosaminoglycans in synovial fluid from patients with rheumatoid arthritis. J Rheumatol 1999;26:34-40.

111. Ishiguro N, Ito T, Oguchi T, Kojima T, Iwata H, Ionescu M et al. Relationships of matrix metalloproteinases and their inhibitors to cartilage proteoglycan and collagen turnover and inflammation as revealed by analyses of synovial fluids from patients with rheumatoid arthritis. Arthritis Rheum 2001;44:2503-11.

112. Sasaki S, Iwata H, Ishiguro N, Obata K, Miura T. Detection of stromelysin in synovial fluid and serum from patients with rheumatoid arthritis and osteoarthritis. Clin Rheumatol 1994;13:228-33.

113. Ishiguro N, Ito T, Obata K, Fujimoto N, Iwata H. Determination of stromelysin-1, 72 and 92 kDa type IV collagenase, tissue inhibitor of metalloproteinase-1 (TIMP-1), and TIMP-2 in synovial fluid and serum from patients with rheumatoid arthritis. J Rheumatol 1996;23:1599-604.

114. Cheung NT, Mattey DL, Dawes PT, Taylor DJ. Serum pro-matrix metalloproteinase 3 in rheumatoid arthritis: a reflection of local or systemic inflammation? Arthritis Rheum 1996;39:884-6.

115. Walakovits LA, Moore VL, Bhardwaj N, Gallick GS, Lark MW. Detection of stromelysin and collagenase in synovial fluid from patients with rheumatoid arthritis and posttraumatic knee injury. Arthritis Rheum 1992;35:35-42.

116. Manicourt DH, Fujimoto N, Obata K, Thonar EJ. Levels of circulating collagenase, stromelysin-1, and tissue inhibitor of matrix metalloproteinases 1 in patients with rheumatoid arthritis. Relationship to serum levels of antigenic keratan sulfate and systemic parameters of inflammation. Arthritis Rheum 1995;38:1031-9.


47

117. Taylor DJ, Cheung NT, Dawes PT. Increased serum proMMP-3 in inflammatory arthritides: a potential indicator of synovial inflammatory monokine activity. Ann Rheum Dis 1994;53:768-72.

118. Ribbens C, Andre B, Jaspar JM, Kaye O, Kaiser MJ, De Groote D et al. Matrix metalloproteinase-3 serum levels are correlated with disease activity and predict clinical response in rheumatoid arthritis. J Rheumatol 2000;27:888-93.

119. Posthumus MD, Limburg PC, Westra J, van Leeuwen MA, van Rijswijk MH. Serum matrix metalloproteinase 3 in early rheumatoid arthritis is correlated with disease activity and radiological progression. J Rheumatol 2000;27:2761-8.

120. Yamanaka H, Matsuda Y, Tanaka M, Sendo W, Nakajima H, Taniguchi A et al. Serum matrix metalloproteinase 3 as a predictor of the degree of joint destruction during the six months after measurement, in patients with early rheumatoid arthritis. Arthritis Rheum 2000;43:852-8.

121. Gabay C, Kushner I. Acute-phase proteins and other systemic responses to inflammation. N Engl J Med 1999;340:448-54.

122. Itoh T, Tanioka M, Yoshida H, Yoshioka T, Nishimoto H, Itohara S. Reduced angiogenesis and tumor progression in gelatinase A-deficient mice. Cancer Res 1998;58:1048-51.

123. Itoh T, Matsuda H, Tanioka M, Kuwabara K, Itohara S, Suzuki R. The role of matrix metalloproteinase-2 and matrix metalloproteinase-9 in antibody-induced arthritis. J Immunol 2002;169:2643-7.

124. Vu TH, Shipley JM, Bergers G, Berger JE, Helms JA, Hanahan D et al. MMP-9/gelatinase B is a key regulator of growth plate angiogenesis and apoptosis of hypertrophic chondrocytes. Cell 1998;93:411-22.

125. Holmbeck K, Bianco P, Caterina J, Yamada S, Kromer M, Kuznetsov SA et al. MT1-MMP-deficient mice develop dwarfism, osteopenia, arthritis, and connective tissue disease due to inadequate collagen turnover. Cell 1999;99:81-92.

126. Neuhold LA, Killar L, Zhao W, Sung ML, Warner L, Kulik J et al. Postnatal expression in hyaline cartilage of constitutively active human collagenase-3 (MMP-13) induces osteoarthritis in mice. J Clin Invest 2001;107:35-44.

127. Lewis EJ, Bishop J, Bottomley KM, Bradshaw D, Brewster M, Broadhurst MJ et al. Ro 32-3555, an orally active collagenase inhibitor, prevents cartilage breakdown in vitro and in vivo. Br J Pharmacol 1997;121:540-6.

128. Brewster M, Lewis EJ, Wilson KL, Greenham AK, Bottomley KM. Ro 32-3555, an orally active collagenase selective inhibitor, prevents structural damage in the STR/ORT mouse model of osteoarthritis. Arthritis Rheum 1998;41:1639-44.

129. Mengshol JA, Mix KS, Brinckerhoff CE. Matrix metalloproteinases as therapeutic targets in arthritic diseases: bull's-eye or missing the mark? Arthritis Rheum 2002;46:13-20.

130. Bramhall SR, Hallissey MT, Whiting J, Scholefield J, Tierney G, Stuart RC et al. Marimastat as maintenance therapy for patients with advanced gastric cancer: a randomised trial. Br J Cancer 2002;86:1864-70.

131. Hemmings FJ, Farhan M, Rowland J, Banken L, Jain R. Tolerability and pharmacokinetics of the collagenase-selective inhibitor Trocade in patients with rheumatoid arthritis. Rheumatology (Oxford) 2001;40:537-43.

132. Close DR. Matrix metalloproteinase inhibitors in rheumatic diseases. Ann Rheum Dis 2001;60 Suppl 3:iii62-7.

133. Greenwald RA, Moak SA, Ramamurthy NS, Golub LM. Tetracyclines suppress matrix metalloproteinase activity in adjuvant arthritis and in combination with flurbiprofen, ameliorate bone damage. J Rheumatol 1992;19:927-38.

134. Greenwald RA, Golub LM, Lavietes B, Ramamurthy NS, Gruber B, Laskin RS et al. Tetracyclines inhibit human synovial collagenase in vivo and in vitro. J Rheumatol 1987;14:28-32.

Chapter 2

48

135. Smith GN, Jr., Yu LP, Jr., Brandt KD, Capello WN. Oral administration of doxycycline reduces collagenase and gelatinase activities in extracts of human osteoarthritic cartilage. J Rheumatol 1998;25:532-5.

136. O'Dell JR, Blakely KW, Mallek JA, Eckhoff PJ, Leff RD, Wees SJ et al. Treatment of early seropositive rheumatoid arthritis: a two-year, double-blind comparison of minocycline and hydroxychloroquine. Arthritis Rheum 2001;44:2235-41.

137. Bluhm GB, Sharp JT, Tilley BC, Alarcon GS, Cooper SM, Pillemer SR et al. Radiographic results from the Minocycline in Rheumatoid Arthritis (MIRA) Trial. J Rheumatol 1997;24:1295-1302.

138. van der Laan W, Molenaar E, Ronday K, Verheijen J, Breedveld F, Greenwald R et al. Lack of effect of doxycycline on disease activity and joint damage in patients with rheumatoid arthritis. A double blind, placebo controlled trial. J Rheumatol 2001;28:1967-74.

139. Ohta S, Harigai M, Tanaka M, Kawaguchi Y, Sugiura T, Takagi K et al. Tumor necrosis factor-alpha (TNF-alpha) converting enzyme contributes to production of TNF-alpha in synovial tissues from patients with rheumatoid arthritis. J Rheumatol 2001;28:1756-63.

140. Beck G, Bottomley G, Bradshaw D, Brewster M, Broadhurst M, Devos R et al. (E)-2(R)-[1(S)-(Hydroxycarbamoyl)-4-phenyl-3-butenyl]-2'-isobutyl-2'-(methanesulfonyl)-4-methylvalerohydra-zide (Ro 32-7315), a selective and orally active inhibitor of tumor necrosis factor-alpha convertase. J Pharmacol Exp Ther 2002;302:390-6.

141. Smolen JS, Steiner G. Therapeutic strategies for rheumatoid arthritis. Nat Rev Drug Discov 2003;2:473-88.

142. Mengshol JA, Vincenti MP, Coon CI, Barchowsky A, Brinckerhoff CE. Interleukin-1 induction of collagenase 3 (matrix metalloproteinase 13) gene expression in chondrocytes requires p38, c-Jun N-terminal kinase, and nuclear factor kappaB: differential regulation of collagenase 1 and collagenase 3. Arthritis Rheum 2000;43:801-11.

143. Liacini A, Sylvester J, Li WQ, Zafarullah M. Inhibition of interleukin-1-stimulated MAP kinases, activating protein-1 (AP-1) and nuclear factor kappa B (NF-kappa B) transcription factors down-regulates matrix metalloproteinase gene expression in articular chondrocytes. Matrix Biol 2002;21:251-62.

144. Brennan FM, Browne KA, Green PA, Jaspar JM, Maini RN, Feldmann M. Reduction of serum matrix metalloproteinase 1 and matrix metalloproteinase 3 in rheumatoid arthritis patients following anti-tumour necrosis factor-alpha (cA2) therapy. Br J Rheumatol 1997;36:643-50.

145. Catrina AI, Lampa J, Ernestam S, Af KE, Bratt J, Klareskog L et al. Anti-tumour necrosis factor (TNF)-alpha therapy (etanercept) down- regulates serum matrix metalloproteinase (MMP)-3 and MMP-1 in rheumatoid arthritis. Rheumatology (Oxford) 2002;41:484-9.

146. den Broeder AA, Joosten LA, Saxne T, Heinegard D, Fenner H, Miltenburg AM et al. Long term anti-tumour necrosis factor alpha monotherapy in rheumatoid arthritis: effect on radiological course and prognostic value of markers of cartilage turnover and endothelial activation. Ann Rheum Dis 2002;61:311-8.

147. Lipsky PE, van der Heijde DM, St Clair EW, Furst DE, Breedveld FC, Kalden JR et al. 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 2000;343:1594-1602.

148. Genovese MC, Bathon JM, Martin RW, Fleischmann RM, Tesser JR, Schiff MH et al. Etanercept versus methotrexate in patients with early rheumatoid arthritis: two-year radiographic and clinical outcomes. Arthritis Rheum 2002;46:1443-50.

149. Cunnane G, Fitzgerald O, Beeton C, Cawston TE, Bresnihan B. Early joint erosions and serum levels of matrix metalloproteinase 1, matrix metalloproteinase 3, and tissue inhibitor of metalloproteinases 1 in rheumatoid arthritis. Arthritis Rheum 2001;44:2263-74.


49

150. Agrawal A, Cha-Molstad H, Samols D, Kushner I. Overexpressed nuclear factor-kappaB can participate in endogenous C-reactive protein induction, and enhances the effects of C/EBPbeta and signal transducer and activator of transcription-3. Immunology 2003;108:539-47.

Chapter 3

Serum levels of matrix metalloproteinase 3 in relation to the development of radiological

damage in patients with early rheumatoid arthritis

Marcel D. Posthumus, Pieter C. Limburg, Johanna Westra, Hans A. Cats, Roy E. Stewart, Miek A. van Leeuwen, and Martin H. van Rijswijk.

Rheumatology 1999;38:1081-1087.

Chapter 3

52

SUMMARY Objective. To evaluate the significance of serum matrix metalloproteinase 3 (MMP-3) levels in relation to the development of radiological damage (X-ray damage) in early rheumatoid arthritis (RA). Methods. Serum MMP-3 levels were measured in 46 healthy controls (CTRL), 19 osteoarthritis (OA) and in 78 RA patients with joint symptoms < 1 year at presentation (T0): 48 patients without and 30 with X-ray damage at T0. Serum MMP-3, measured by ELISA, and X-ray damage, scored according to Sharp’s-method were assessed at 0, 6, 12 and 24 months. Results. MMP-3 levels in CTRL and OA were low or undetectable with no differences between the groups (p = 0.19). Levels in RA were higher than in CTRL (p < 0.01). Initial MMP-3 levels in patients with X-ray damage at T0 (n=30) were higher than the levels in patients without any X-ray damage during follow-up (n=19) (p < 0.01), but were not different from patients who developed X-ray damage during the study (n=29, p = 0.11). In the patients without X-ray damage at T0 there was a significant correlation between MMP-3 at T0 and the total X-ray damage after 6 months (r = 0.34, p = 0.02) and 12 months (r = 0.32, p = 0.03). This correlation was almost exclusively determined by joint space narrowing in the Sharp score. Conclusion. Serum MMP-3 level seems to be an indicator for the development of radiological damage in patients with early RA and appears to be particularly indicative of cartilage degradation. Key words: serum matrix metalloproteinase 3, early rheumatoid arthritis, radiolo-gical damage.

Serum MMP-3 and radiological damage in early RA

53

INTRODUCTION Matrix metalloproteinases (MMPs) are proteolytic enzymes that can degrade extra-cellular matrix components 1. These enzymes are synthesized as latent pro-enzymes, contain a zinc binding active site and require Ca2+ and proteolytic cleavage for activation. The MMP-family includes four different groups of enzymes: (I) collagenases (interstitial collagenase 1 or MMP-1, neutrophil-type collagenase 2 or MMP-8 and collagenase 3 or MMP-13), (II) gelatinases (gelatinase A or MMP-2 and gelatinase B or MMP-9), (III) stromelysins (stromelysin-1 or MMP-3, stromelysin-2 or MMP-10) and (IV) others (matrilysin, stromelysin-3, metalloelastase and membrane-type MMPs which are not secreted but anchored into the membrane). The activity of the MMPs can be regulated either by modulation of pro-enzyme production and/or activation or by changes in the levels of inhibitors of MMPs such as α2-macroglobulin and tissue inhibitors of matrix metalloproteinases (TIMPs) 2,3. An imbalance between these inhibitors and the proteinases conceivably leads to matrix degradation 4,5. The MMPs play an important role both in normal physiological processes, and in pathological conditions such as arthritis. In patients with rheumatoid arthritis (RA), the MMPs play a major role in the destruction of cartilage and other components of connective tissue in the joints. MMPs are released by synovial fibroblasts, chondrocytes, macrophages, neutrophils and endothelial cells in response to pro-inflammatory cytokines like interleukin-1 (IL-1), tumour necrosis factor α (TNF-α), and growth factors such as epidermal growth factor and platelet derived growth factor 6,7. Of the MMP family MMP-3 is thought to play a prominent role in the pathogenesis of RA even though it is not the only key enzyme in matrix degradation in this disease 8. MMP-3 is capable of degrading many components of matrix proteins in the synovial joint including proteoglycans, gelatins, laminin, fibronectin and collagens III, IV, IX, X 1,9-11. The enzyme has been localized in the fibroblast-like synovial lining cell of rheumatoid synovium 12,13 and in RA cartilage 4,5. It is secreted as a latent pro-enzyme resulting in highly elevated MMP-3 levels in synovial fluid of RA patients 14-16. Although serum levels are (300-500 times) lower there is a highly significant correlation with matched synovial fluid levels indicating that serum-MMP-3 is derived mainly from the inflamed synovium 14,16-20. Levels of activated MMP-3 are also highly increased in synovial fluid 21 but measurement of activated MMP-3 in serum will be very difficult, if not impossible, because of the high serum levels of α2-macroglobulin which encloses the activated proteinase completely 3.

Chapter 3

54

In RA serum MMP-3 could be a specific marker of joint inflammation and destruction because it is almost exclusively produced locally, in the inflamed synovium. In that case it could be used in the early identification of patients with aggressive destructive disease which is important for therapeutic reasons 22. Several other features such as age, HLA-DR4, rheumatoid factor and C-reactive protein (CRP)-production also appear to be prognostic for the final outcome 23. Although CRP is a good parameter for prognostic purposes 24 and for monitoring treatment effects, it is an indicator of inflammation in general which may be influen-ced by other stimuli of the acute phase response, this in contrast with MMP-3 17. Previous studies of serum MMP-3 levels in RA concerned patients with a long disease duration 14,16,18,20. Investigation in an early phase of the disease and in particular in a group of patients without initial radiological damage (X-ray damage) has, to our knowledge, not been reported. In the present study we measured the levels of MMP-3 in sera from patients with early rheumatoid arthritis without X-ray damage at presentation and compared these with serum levels in RA patients with already X-ray damage at presentation, osteoarthritis (OA) patients and healthy controls (CTRL). In the group of RA patients without initial X-ray damage we analyzed whether serum MMP-3 levels have a predictive value for the development of X-ray damage. Furthermore we investigated in this selected group of RA patients the relationship between serum MMP-3 levels and CRP. PATIENTS AND METHODS Patients Seventy-eight patients with rheumatoid arthritis (RA), 19 patients with OA and 46 CTRL were studied. The RA patients were selected from a cohort of patients with RA, according to the 1987 ACR criteria 25, with joint symptoms existing less than one year at presentation and who had not previously received disease modifying antirheumatic drugs (DMARDs). These patients participated in a prospective follow-up study at the department of rheumatology at the Groningen University Hospital. During follow-up patients were treated with non-steroidal anti-inflammatory drugs and DMARDs as indicated clinically. Guidelines for the sequence of the different second-line drugs were as follows: hydroxychloroquine or sulphasalazine as first choice therapy, followed in order by intramuscular gold, D-penicillamine, azathioprine or methotrexate. Low dose corticosteroids could be administered as adjuvant therapy. Out of this cohort 48 consecutive RA patients without initial X-ray damage and a randomly chosen group of 30 RA patients with X-ray damage at presentation, due to


55

RA, were selected. In the group of 48 RA patients without initial X-ray damage 19 patients did not develop X-ray damage during the 2 years of follow-up (RA-00 group; initially no X-ray damage (=0) and after 2 years no X-ray damage (=0)). The remaining 29 patients developed X-ray damage (1 or more Sharp points) during the 2 years of follow-up (RA-0X group; initially no X-ray damage (=0), after 2 years X-ray damage (=X)). Clinical (number of painful joints, number of swollen joints, Ritchie articular index) and laboratory (CRP) investigations were performed at monthly visits during 2 years. Radiographs of hands and feet were obtained at study entry and every 6 months during 2 years. Joint damage in the hands and feet was assessed by Sharp’s method with some modifications, in particular inclusion of foot joints, as described by Van der Heijde et al 26,27. By this method joint space narrowing and erosions are scored separately and combined to a total X-ray damage score. The radiographs were scored without knowledge of clinical and laboratory data in chronological order per patient by two observers (M.v.L. and H.A.C.). The inter-observer agreement was 0.90 and the intra-observer agreements were 0.96 and 0.99 for the two observers respectively. The radiographic score in the group of RA-patients with initial X-ray damage (RA-XX) was 5 or more at study entry. For analysis we used the variables at presentation (T0) and after 6, 12 and 24 months. Of the group of 30 RA-patients with X-ray damage at presentation (RA-XX) only the variables at study entry were used. Osteoarthritis was diagnosed from clinical and radiological investigations using the ACR criteria for osteoarthritis 28-30. The CTRL were bloodbank donors. Measurement of MMP-3 Serum samples were obtained at 0, 6, 12 and 24 months. MMP-3 levels were determined with an MMP-3 ELISA developed in our laboratory. In short, 96-well plates were precoated with F(ab)2 fragment of goat-anti-mouse IgG, 1µg/ml (Jackson Immunoresearch Labs, West Grove, PN, USA). Next a mouse monoclonal antibody against human MMP-3, clone 55-2-A4 (clone B) 31, (Oncogene, Cambridge, MA, USA) was coated, 0.2 µg/ml. Serum samples were analyzed in two-fold serial dilutions in High Performance ELISA buffer (CLB, Amsterdam, NL) and incubated during 1 hr. After washing bound MMP-3 was detected with a biotinylated polyclonal sheep-anti-MMP-3 (The Binding Site, Birmingham, UK) in combination with streptavidin-E+ (CLB, Amsterdam, NL). Peroxidase activity was determined using tetramethyl-benzidin as substrate. MMP-3 levels were calculated at the linear range of the assay from a standard curve (10-1500 ng/ml) using a RA synovial fluid, which was standardized against the BIOTRAK® standard (see below). The intra-

Chapter 3

56

assay coefficient of variation (CV) was 8.9%, the inter-assay CV 10.1%. With an immunoblot we have demonstrated that both the monoclonal and the polyclonal antibody reacted with active MMP-3 as well as with pro-MMP-3 (data not shown). Treatment of the sera with p-aminophenyl mercury acetate (APMA) 32,33 to activate pro-MMP-3 did not influence results in the ELISA. Furthermore it was demonstrated that rheumatoid factors do not react in this assay and do not interfere with measurement of MMP-3. In the group of 48 patients without initial radiological damage (RA-00/0X) serum MMP-3 levels were also measured in the BIOTRAK® MMP-3 ELISA kit (Amersham,‘s Hertogenbosch, NL) according to the manufacturers recommen-dations at 0, 6, 12 and 24 months. Other laboratory measurements In patients with RA C-reactive protein (CRP) and rheumatoid factor (RF) were measured by rate nephelometry (Behring; normal values: CRP < 3 mg/l, RF < 12 IU/ml) at 0, 6, 12 and 24 months. Statistical analysis Correlations were analyzed using Pearson correlations and Spearman’s rank correlation coefficient. Differences between groups were analyzed by Chi square and the Mann-Whitney U test. Logistic regression analysis was performed with radiological damage as dependent variable and serum MMP-3 as independent variable at different time points. P values less than 0.05 were considered significant. RESULTS Characteristics of control subjects, OA- and RA patients are summarized in table 1. All OA-patients had osteoarthritis of the hands according to the ACR criteria. Nine out of 19 patients had also osteoarthritis of the knees or hips. There was one male OA patient with a high CRP of 170 mg/l due to an epididymitis and one female OA patient with a CRP level of 44 mg/l diagnosed as having a polymyalgia rheumatica as well. The CRP-levels of the other 17 OA patients were low (median: 3 mg/l, range 2-21 mg/l). In the group of 48 RA 00/0X patients 44 patients (92%) were treated with DMARDs and low dose corticosteroids were given as adjuvant treatment in 6 patients (13%).


57

TABLE 1. Characteristics of the rheumatoid arthritis (RA), osteoarthritis (OA) patients and healthy controls at study entry. Patients RA without

radiological damage at

presentation

RA with radiological damage at

presentation

Osteoarthritis Controls

(RA-00 & RA-0X) (RA-XX) (OA) (CTRL) (N=48) (N=30) (N=19) (N=46) Age 47.5 (16-76) 45.0 (18-71) 59.0 (52-73) 41.5 (23-65) Gender, female/ male (% female)

31/17 (65%) 18/12 (60%) 17/2 (89%) 32/14 (70%)

Disease duration (months)

4.6 (1.5-12) † 8.0 (1.5-12) - -

Rheumatoid factor, +/- (%+)

34/14 (71%) 24/6 (80%) - -

Joint pain 16.0 (2-44) 14.5 (1-30) - - Joint swelling 12.5 (1-36) 10.5 (1-28) - - Ritchie 13.5 (2-50) 12.0 (1-30) - - CRP 11.0 (1-232) ¶ 32.0 (1-250) - - Sharp score 0 12.0 (5-46) - - Values are the median and range. RA-00: at presentation no radiological damage (=0), after 2 years no radiological damage (=0). RA-0X: at presentation no radiological damage (=0), after 2 years radiological damage; ≥ 1 Sharp point (=X). RA-XX: at presentation already radiological damage; ≥ 5 Sharp points (=X), after 2 years radiological damage (=X). † RA-00/0X vs. RA-XX p < 0.01; ¶ RA-00/0X vs. RA-XX p = 0.05.

There were no differences between the RA groups in age, gender, rheumatoid factor positivity, joint pain score, joint swelling score and Ritchie articular index. Disease duration was shorter in the RA-00/0X group (p < 0.01) There was a borderline significant difference in initial CRP levels between the RA-00/0X and RA-XX groups (p = 0.05). Measurement of MMP-3 In the group of 48 patients without initial radiological damage (RA-00/0X) the serum MMP-3 levels were determined with the MMP-3 ELISA developed in our own laboratory and with the BIOTRAK® MMP-3 ELISA. In 192 serum samples there was a good correlation between the two tests (r = 0.96, p < 0.01). Further serum MMP-3 measurements and calculations were performed with our own MMP-3 ELISA.

Chapter 3

58

Initial serum MMP-3 level in CTRL, OA- and RA patients Serum levels of MMP-3 at study entry are summarized in table 2 and figure 1.

TABLE 2. Serum levels of matrix metalloproteinase 3 (MMP-3), ng/ml in rheumatoid arthritis (RA), osteoarthritis (OA) patients and healthy controls (CTRL) at study entry. Patients RA without

radiological damage at presentation

RA with radiological damage at

presentation

Osteoarthritis Controls

(RA-00 & RA-0X) (RA-XX) (OA) (CTRL) (N=48) (N=30) (N=19) (N=46) Female (n=32) Male (n=14) MMP-3 33.5 (5-1460) †§ 95.5 (5-1800) ‡ 20.0 (5-66) 5.0 (5-22) 19.6 (5-61) ¶ Values are the median and range. RA-00: at presentation no radiological damage (=0), after 2 years no radiological damage (=0). RA-0X: at presentation no radiological damage (=0), after 2 years radiological damage; ≥ 1 Sharp point (=X). RA-XX: at presentation already radiological damage; ≥ 5 Sharp points (=X), after 2 years radiological damage (=X). ¶ CTRL-male vs. CTRL-female p < 0.01; † RA-00/0X vs. CTRL p < 0.01; ‡ RA-XX vs. CTRL p < 0.01; § RA-00/0X vs. RA-XX p < 0.01.

In CTRL the serum MMP-3 levels were undetectable (< 10.0 ng/ml) or low. In these control subjects the serum MMP-3 levels in men were higher than those in women (p < 0.01). Sex differences were not found in the groups of OA and RA patients. No differences were found in serum MMP-3 levels between CTRL and OA patients (p = 0.19). The serum MMP-3 levels in RA patients were higher in comparison with the CTRL (RA-00/0X vs. CTRL p < 0.01 and RA-XX vs. CTRL p < 0.01). Initial serum MMP-3 levels in RA patients in relation to X-ray damage at presentation Serum MMP-3 levels were lower in the RA-00/0X group in comparison with the RA-XX group (p < 0.01).


59

FIGURE 1. MMP-3 serum levels in the female controls (CTRL-F), male controls (CTRL-M), osteoarthritis (OA) patients and the 3 groups of rheumatoid arthritis patients. The horizontal line is the median. RA-00: at presentation no radiological damage (=0), after 2 years no radiological damage (=0) RA-0X: at presentation no radiological damage (=0), after 2 years radiological damage; > 1 Sharp point (=X) RA-XX: at presentation already radiological damage; > 5 Sharp points (=X), after 2 years radiological

damage (=X). Initial serum MMP-3 levels in the RA-00, RA-0X and RA-XX subgroups During 2 years of follow-up 29 of the 48 patients developed X-ray damage (≥ 1 Sharp-point). Characteristics at study entry of the 3 subgroups are summarized in table 3. There were no significant differences between the 3 groups in age, gender, joint pain score, joint swelling score and Ritchie articular index. CRP levels were lower in the RA-00 group (p < 0.01). In the RA-00 group (n=19) 53% had a positive rheumatoid factor test in contrast to 83% in the RA-0X group (n=29) and 80% in the RA-XX group (n=30) (p = 0.05).

Chapter 3

60

TABLE 3. Characteristics at entry of the 19 RA patients without radiological damage after 2 years (RA-00), the 29 RA patients with ≥ 1 Sharp point after 2 years (RA-0X) and the 30 RA patients with at start of the study already ≥ 5 Sharp points. Patients RA-00 patients

(N=19) RA-0X patients

(N=29) RA-XX patients

(N=30) Age 50.0 (16-76) 44.0 (17-74) 45.0 (18-71) Gender, female/male (% female) 13/6 (68%) 18/11 (62%) 18/12 (60%) Rheumatoid factor, +/- (%+) 10/9 (53%) 24/5 (83%) 24/6 (80%) Joint pain 16.0 (2-44) 15.0 (2-36) 14.5 (1-30) Joint swelling 9.0 (1-33) 13.0 (1-36) 10.5 (1-28) Ritchie 15.0 (2-48) 12.0 (2-50) 12.0 (1-30) CRP 3.0 (1-38)¶ 29.0 (2-232) 32.0 (1-250) MMP-3 18.0 (5-85)†‡ 38.0 (5-1460) 95.5 (5-1800) Values are the median and range. RA-00: at presentation no radiological damage (=0), after 2 years no radiological damage (=0). RA-0X: at presentation no radiological damage (=0), after 2 years radiological damage; ≥ 1 Sharp point (=X). RA-XX: at presentation already radiological damage; ≥ 5 Sharp points (=X), after 2 years radiological damage (=X). ¶ RA-00 vs. RA-0X or RA-XX p < 0.01; † RA-00 vs. RA-0X p = 0.05; ‡ RA-00 vs RA-XX p < 0.01.

In the RA-00 group the serum MMP-3 levels at study entry were lower than in the RA-XX group (p < 0.01). The difference in serum MMP-3 levels at study entry between RA-00 and RA-0X patients reached borderline significance (p = 0.05). There were no significant differences in serum MMP-3 levels at study entry between the 29 patients who developed X-ray damage (RA-0X group) compared to the 30 RA-XX patients with X-ray damage at presentation (p = 0.11). Initial serum MMP-3 levels in relation to development of X-ray damage In the complete group of 48 RA-00/0X patients serum MMP-3 level at study entry correlated significantly with the total Sharp score at 6 months (r = 0.34, p = 0.02) and 12 months (r = 0.32, p = 0.03). The correlation between serum MMP-3 level at study entry and total Sharp score at 24 months reached borderline significance (r = 0.28, p = 0.05). Further separate analysis of joint space narrowing and erosions demonstrated a significant correlation between serum MMP-3 levels at study entry and joint space narrowing at 6 months (r = 0.41, p < 0.01), 12 months (r = 0.37, p < 0.01) and 24 months (r = 0.32, p = 0.03). In the subgroup of 29 RA-0X patients there was also a significant correlation between serum MMP-3 at presentation and joint space narrowing at 6 months (r = 0.42, p = 0.02) and 12 months (r = 0.37, p = 0.04). There were no correlations between initial serum MMP-3 levels and erosions at any time point.


61

At a cut-off point for serum MMP-3 of 80 ng/ml (the upper limit of CTRL and OA patients) we found a positive predictive value for the development of X-ray damage within 2 years of 91.7% (sensitivity 37.9%, specificity 94.7%). Serum MMP-3 levels in relation to clinical parameters and CRP in the 48 patients without initial X-ray damage Serum MMP-3 correlated with joint swelling at study entry and after 24 months. The other clinical parameters did not correlate with MMP-3 at any time point. Serum MMP-3 levels correlated with CRP at study entry (r = 0.42, p < 0.01), 6 months (r = 0.56, p < 0.01), 12 months (r = 0.52, p < 0.01) and 24 months (r = 0.38, p < 0.01). A logistic regression model, with X-ray damage at 6, 12, 24 months as dependent variable and MMP-3 level and/or CRP level at study entry as independent variables reached significance for CRP levels at study entry and X-ray damage at 6 and 24 months. In the combined model of MMP-3 and CRP, MMP-3 had no additive predictive value for the explained variance of X-ray damage at these time points. Separate analysis of joint space narrowing and erosions did not influence these results. DISCUSSION Research on matrix metalloproteinases in an early phase of RA is of interest because early identification of patients with aggressive destructive disease is important for prognostic and therapeutic reasons. Novel more aggressive therapies, including matrix metalloproteinase inhibitors, are currently developed 34-38 and thus the need for clinically feasible means for assessing the prognosis in the individual patient is becoming more and more important. Even though an elevated serum MMP-3 level is not unique for RA (elevated levels have also been found in patients with for example SLE, MCTD, and psoriatic arthritis 39,40) it could be a specific marker of joint inflammation and destruction in RA because it is almost exclusively produced in the inflamed synovium. Recent studies showed close correlations between serum MMP-3 and disease activity in RA.39,41 As far as we know there are no studies reported that link serum MMP-3 to the development of radiological damage in RA. In the present study we evaluated the significance of serum MMP-3 levels in relation to the development of radiological damage in early rheumatoid arthritis. Furthermore we investigated the relationship between serum MMP-3 and CRP. As in other studies 14,20,31 we found that serum MMP-3 levels were undetectable or low in normal CTRL and OA patients. In healthy male CTRL serum MMP-3 levels

Chapter 3

62

appeared to be higher than in healthy female CTRL, a phenomenon which has been reported before 14,40. Sex differences were not found in OA patients (possibly due to the skewed distribution) nor in RA patients (possibly due to the fact that small sex differences were overruled by the massive production of MMP-3 in these patients). Further analysis of serum MMP-3 levels in normal controls with respect to gender, age and variation in time is of interest because these variables could be important in determining the cut-off point between normal and abnormal values. Serum MMP-3 levels in RA patients were higher than in CTRL. Initial MMP-3 levels in patients with X-ray damage at presentation (RA-XX) were higher than the levels in patients without any X-ray damage during follow-up (RA-00), but were not different from patients who developed X-ray damage during the study (RA-0X). In other words a high initial serum MMP-3 level was associated with the development of radiological damage (positive predictive value for the development of X-ray damage within 2 years of 91.7%, at a cut-off point for serum MMP-3 of 80 ng/ml). In the group of 48 patients without initial X-ray damage (RA-00/0X) we found a significant correlation between initial serum MMP-3 levels and the development of X-ray damage which was mainly determined by joint space narrowing in the Sharp score. In the subgroup of 29 patients who developed X-ray damage during follow-up (RA-0X) there was also a significant correlation between initial serum MMP-3 levels and joint space narrowing in the Sharp score. The finding that initial serum MMP-3 levels correlate with joint space narrowing fits with the fact that the main targets of MMP-3 are localized in the matrix of cartilage, like proteoglycans 1,9-11 and the fact that in animal models MMP-3 was inducible and detectable in the chondrocytes in a very early phase of the arthritis 11. Serum MMP-3 and CRP showed significant correlations at all time points. Since the production of acute phase proteins like CRP as well as the production of MMP-3 is stimulated by pro-inflammatory cytokines such as IL-1 and TNF-α these correlations could be expected. Logistic regression with X-ray damage (as total Sharp score or separated into joint space narrowing or erosions) as dependent variable did not yield statistical significance when serum MMP-3 at study entry was used as independent variable, in contrast to CRP. An explanation for this finding could be the hypothesis that it takes some time to produce enough MMP-3 to be detectable in serum. Taylor et al found a higher median serum MMP-3 in patients with established RA (mean disease duration 121.5 months (SD 73.1)) than in those with recently diagnosed RA (mean disease duration 12.8 months (SD 9.0)). They suggest that the serum MMP-3 level is affected by the amount of inflammatory tissue within the joints which increases over the years in this condition 19. The disease duration in our patients groups was even


63

shorter (see table 1) compared to their “early RA group”. In this context further analysis concerning serum MMP-3 in the early phases of RA seems to be of interest. In the combined logistic regression model of MMP-3 and CRP, MMP-3 had no additive predictive value for the explained variance of X-ray damage at these time points. The fact that both CRP and MMP-3 production are stimulated by the same cytokines could be one explanation for this result. A direct comparison between serum MMP-3 and CRP to analyze which one is the best in predicting radiological damage was not the major goal of our study. As we look at our results serum MMP-3 is not superior to CRP. CRP is still an excellent parameter of disease activity and for monitoring the progression of radiological damage. On the other hand the good correlation between initial serum MMP-3 and the development of radiological damage is of importance for analyzing mechanisms of drug therapy in relation to disease activity and the development of radiological damage, especially when specific MMP-inhibitors will be introduced in the treatment of RA. In conclusion, serum MMP-3 level seems to be an indicator for the development of radiological damage in a significant number of patients with early RA and appears to be particularly indicative of cartilage destruction. Further longitudinal analysis with multiple measurements in early RA patients seems to be warranted to analyze whether MMP-3 is suitable for the monitoring of disease activity and/or progression of radiological damage. ACKNOWLEDGMENTS This study was supported by a grant from “The Dutch Arthritis Association”, The Netherlands.

Chapter 3

64

REFERENCES 1. Nagase H, Okada Y. Proteinases and matrix degradation. In: Kelley WN, Harris ED, Ruddy S, Sledge

CB, eds. Textbook of rheumatology, 5th ed. Philadelphia: WB Saunders, 1997: 323-41. 2. Nagase H. Activation mechanisms of matrix metalloproteinases. Biol Chem 1997;378:151-60. 3. Barrett AJ. Alpha 2-macroglobulin. Methods Enzymol 1981;80 Pt C:737-54. 4. Martel Pelletier J, McCollum R, Fujimoto N, Obata K, Cloutier JM, Pelletier JP. Excess of

metalloproteases over tissue inhibitor of metalloprotease may contribute to cartilage degradation in osteoarthritis and rheumatoid arthritis. Lab Invest 1994;70:807-15.


6. MacNaul KL, Chartrain N, Lark M, Tocci MJ, Hutchinson NI. Discoordinate expression of stromelysin, collagenase, and tissue inhibitor of metalloproteinases-1 in rheumatoid human synovial fibroblasts. Synergistic effects of interleukin-1 and tumor necrosis factor-alpha on stromelysin expression. J Biol Chem 1990;265:17238-45.

7. Mauviel A. Cytokine regulation of metalloproteinase gene expression. J Cell Biochem 1993;53:288-95.

8. Mudgett JS, Hutchinson NI, Chartrain NA, et al. Susceptibility of stromelysin 1-deficient mice to collagen-induced arthritis and cartilage destruction. Arthritis Rheum 1998;41:110-21.

9. Singer II, Kawka DW, Bayne EK, Donatelli SA, Weidner JR, Williams HR, Ayala JM, Mumford RA, Lark MW, Glant TT, et al. VDIPEN, a metalloproteinase-generated neoepitope, is induced and immunolocalized in articular cartilage during inflammatory arthritis. J Clin Invest 1995;95:2178-86.








17. Cheung NT, Mattey DL, Dawes PT, Taylor DJ. Serum pro-matrix metalloproteinase 3 in rheumatoid arthritis: a reflection of local or systemic inflammation?. Arthritis Rheum 1996;39:884-6.




65


21. Beekman B, van El B, Drijfhout JW, Ronday HK, TeKoppele JM. Highly increased levels of active stromelysin in rheumatoid synovial fluid determined by a selective fluorogenic assay. FEBS Lett. 1997;418:305-9.

22. Emery P, Salmon M. Early rheumatoid arthritis: time to aim for remission? Ann Rheum Dis 1995;54:944-7.

23. van der Heijde DM, van Riel PL, van Leeuwen MA, van 't Hof MA, van Rijswijk MH, van de Putte LB. Prognostic factors for radiographic damage and physical disability in early rheumatoid arthritis. A prospective follow-up study of 147 patients. Br J Rheumatol 1992;31:519-25.

24. van Leeuwen MA, van der Heijde DM, van Rijswijk MH, et al. Interrelationship of outcome measures and process variables in early rheumatoid arthritis. A comparison of radiologic damage, physical disability, joint counts, and acute phase reactants. J Rheumatol 1994;21:425-9.

25. Arnett FC, Edworthy SM, Bloch DA, et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 1988;31:315-24.

26. van der Heijde DM. Plain X-rays in rheumatoid arthritis: overview of scoring methods, their reliability and applicability. Baillieres Clin Rheumatol 1996;10:435-53.

27. van der Heijde DM, van Leeuwen MA, van Riel PL, van de Putte LB. Radiographic progression on radiographs of hands and feet during the first 3 years of rheumatoid arthritis measured according to Sharp's method (van der Heijde modification). J Rheumatol 1995;22:1792-6.

28. Altman R, Alarcón G, Appelrouth D, et al. The American College of Rheumatology criteria for the classification and reporting of osteoarthritis of the hand. Arthritis Rheum 1990;33:1601-10.

29. Altman R, Alarcón G, Appelrouth D, et al. The American College of Rheumatology criteria for the classification and reporting of osteoarthritis of the hip. Arthritis Rheum 1991;34:504-14 .

30. Altman R, Asch E, Bloch G, et al. Development of criteria for the classification and reporting of osteoarthritis: classification of osteoarthritis of the knee. Arthritis Rheum 1986;29:1039-49.

31. Obata K, Iwata K, Okada Y, et al. A one-step sandwich enzyme immunoassay for human matrix metalloproteinase 3 (stromelysin-1) using monoclonal antibodies. Clin Chim Acta 1992;211:59-72.

32. Smith-GN J, Brandt KD, Hasty KA. Activation of recombinant human neutrophil procollagenase in the presence of doxycycline results in fragmentation of the enzyme and loss of enzyme activity. Arthritis Rheum 1996;39:235-44.

33. Okada Y, Harris-ED J, Nagase H. The precursor of a metalloendopeptidase from human rheumatoid synovial fibroblasts. Purification and mechanisms of activation by endopeptidases and 4-aminophenylmercuric acetate. Biochem J 1988;254:731-41.


35. Cawston TE. Metalloproteinase inhibitors and the prevention of connective tissue breakdown. Pharmacol Ther 1996;70:163-82.

36. Cawston TE, Rowan A. Prevention of cartilage breakdown by matrix metalloproteinase inhibition - a realistic therapeutic target? Br J Rheumatol 1998;37:353-6.

37. Zask A, Levin JI, Killar LM, Skotnicki JS. Inhibition of matrix metalloproteinases: structure based design. Current Pharmaceutical Design 1996;2:624-61.

38. Vincenti MP, Clark IM, Brinckerhoff CE. Using inhibitors of metalloproteinases to treat arthritis. Easier said than done?. Arthritis Rheum 1994;37:1115-26.

Chapter 3

66

39. Keyszer G, Lambiri I, Nagel R, et al. Circulating levels of matrix metalloproteinases MMP-3 and MMP-1, tissue inhibitor of metalloproteinase 1 (TIMP-1), and MMP-1/TIMP-1 complex in rheumatic disease. Correlation with clinical activity of rheumatoid arthritis versus other surrogate markers. J Rheumatol 1999;26:251-8.

40. Zucker S, Lysik RM, Zarrabi MH, et al. Elevated plasma stromelysin levels in arthritis. J Rheumatol 1994;21:2329-33.

41. Ichikawa Y, Yamada C, Horiki T, Hoshina Y, Uchiyama M. Uchiyama. Serum matrix metalloproteinase-3 and fibrin degradation product levels correlate with clinical disease activity in rheumatoid arthritis. Clin Exp Rheum 1998;16:533-40.

Chapter 4

Serum matrix metalloproteinase 3 in early rheumatoid arthritis is correlated with

disease activity and radiological progression

Marcel D. Posthumus, Pieter C. Limburg, Johanna Westra, Miek A. van Leeuwen, and Martin H. van Rijswijk.

Journal of Rheumatology 2000;27:2761-2768.

Chapter 4

68

SUMMARY Objective. To analyze the clinical significance of serial measurements of serum

matrix metalloproteinase 3 (MMP-3) levels in relation to markers of disease activity and radiological progression in early rheumatoid arthritis (RA). Methods. In a 3 year prospective study of 33 patients with early RA (symptoms < 1 year at entry) monthly measurements of serum MMP-3 were transformed into time-integrated values for 6 months periods for comparison with other markers of disease activity like swollen joint count (SJC), tender joint count (TJC), Ritchie articular index (RAI), the disease activity score (DAS), erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), and radiological progression, scored according to Sharp’s method, in which erosions and joint space narrowing are scored separately and combined to a total Sharp score.

Results. Significant correlations were found between serum MMP-3 and SJC, ESR, and CRP during all periods and between 6 and 30 months with the DAS. There were no correlations between serum MMP-3 and TJC or the RAI. During the first 12 months serum MMP-3 was only correlated with the item joint space narrowing of the Sharp score. After 12 months of follow-up it was also correlated with the total Sharp score and after 18 months it was correlated with all 3 items of the Sharp score. There was a wide inter-individual variation in the relation between serum MMP-3 and radiological progression but intra-individually this relation seemed to be rather constant.

Conclusion. Time-integrated values of serum MMP-3 are correlated with time-integrated values of other markers of disease activity such as joint swelling, ESR, CRP, and the DAS. Of the radiological scores, as outcome measures, especially joint space narrowing is closely correlated with cumulative serum MMP-3. Key words: serum matrix metalloproteinase 3, stromelysin 1, early rheumatoid arthritis, disease activity, radiological damage.

Serum MMP-3 in relation to disease activity and radiological progression

69

INTRODUCTION In rheumatoid arthritis (RA) cytokines play a major role in local joint inflammation and destruction as well as in the systemic acute phase response. The relationship between acute phase proteins, like C-reactive protein (CRP) and disease activity and radiological progression has been described 1,2. Although CRP appears to be a good parameter for prognostic purposes 3 and for monitoring treatment effects 4, it is an indicator of inflammation in general which may be influenced by other stimuli of the acute phase response 5. Matrix metalloproteinases (MMPs), which are locally produced and activated within the affected joint as a result of cytokine mediated stimulation, could be more specific markers for joint inflammation 6 and especially destruction 7,8 in rheumatic diseases. In RA matrix metalloproteinase 3 (MMP-3 = stromelysin 1) is of interest because it is thought to play a prominent role in the pathogenesis of matrix degradation in RA 9 even though it is not the only key enzyme 10,11. MMP-3 is capable of degrading many components of the matrix in the synovial joint including proteoglycans, gelatins, laminin, fibronectin and collagen III, IV, IX, X 9,10,12,13. Moreover, MMP-3 is able to activate other matrix metalloproteinases like MMP-1, MMP-7, MMP-8, MMP-9 and MMP-1314. The enzyme has been localized in the fibroblast-like synovial lining cells of rheumatoid synovium 15-17, and in RA cartilage 18,19. It is secreted as a latent pro-enzyme resulting in highly elevated MMP-3 levels in synovial fluid (SF) 5,7,20-22. In RA serum MMP-3 is thought to originate from the inflamed joint because there are significant correlations between MMP-3 levels in serum and in SF 5-7,20, and between serum MMP-3 levels and the number of swollen joints and/or Lansbury articular index 7,20. Furthermore local therapies, like intra-articular corticosteroid injections cause significant reductions in serum MMP-3 levels 5,6. In SF, levels of activated MMP-3 are also highly increased 23, but measurement of activated MMP-3 in serum is very difficult, if not impossible, because of the high serum levels of α2-macroglobulin that encloses the activated enzyme completely 24. Elevated levels of serum MMP-3 have also been found in patients with SLE 25-27, MCTD 28, systemic sclerosis 28, gout 7,25, ankylosing spondylitis 6,28, calcium pyrophosphate arthritis 29, and psoriatic arthritis 28. Nevertheless, in RA serum MMP-3 could be a useful parameter to evaluate disease activity and outcome. Especially in early disease the use of markers of joint inflammation (disease activity) and joint destruction (radiological damage) are of importance for prognostic and therapeutical reasons 30. Therefore we analyzed in a prospective study the clinical significance of serial measurements of serum MMP-3 levels in relation to

Chapter 4

70

other markers of disease activity and radiological progression in early rheumatoid arthritis. PATIENTS AND METHODS Patients Thirty-three RA patients with radiological progression during a follow-up of 3 years were selected from a cohort of 149 patients with RA according to the 1987 ACR criteria 31 with joint symptoms existing less than one year at presentation and who had not previously received disease modifying antirheumatic drugs (DMARDs). These patients participated in a prospective follow-up study at the department of rheumatology at the Groningen University Hospital. During follow-up patients were treated with non-steroidal anti-inflammatory drugs and DMARDs as indicated clinically. Guidelines for the sequence of the different DMARDs were as follows: hydroxychloroquine or sulphasalazine as first choice therapy, followed in order by intramuscular gold, D-penicillamine, azathioprine, or methotrexate. Low dose corticosteroids could be administered as adjuvant therapy. In this cohort of 149 patients 43% had normal X-rays at presentation. Out of this cohort of 149 RA patients 33 consecutive patients were selected who showed radiological progression of at least 10 or more points according to Sharp’s method during a follow-up of 3 years. Clinical and laboratory investigations were performed at monthly intervals during a follow-up of 3 year. Radiographs of hands and feet were obtained at study entry and every 6 months during follow-up. Clinical markers of disease activity Fifty-two peripheral joints were examined for tenderness and soft tissue swelling. The following articular indices were determined: Ritchie articular index (RAI) 32, tender joint count (TJC), swollen joint count (SJC), and the disease activity score (DAS) according to Van der Heijde with 3 variables: Ritchie articular index, number of swollen joints, and erythrocyte sedimentation rate (ESR) 33. Radiological analysis Radiological damage in hands and feet was assessed by Sharp’s method with some modifications as described by Van der Heijde et al 34,35. By this method joint space narrowing (JSN) and erosions (ER) are scored separately and combined to a total Sharp score (TSS) with a maximum TSS of 448 points. The radiographs were scored without knowledge of clinical and laboratory data in chronological order per patient


71

by two observers. The inter-observer agreement was 0.90 and the intra-observer agreements were 0.96 and 0.99 for the two observers respectively. Laboratory analysis Serum MMP-3 levels were determined with a MMP-3 ELISA developed in our own laboratory 8. In short, 96 well plates were precoated with F(ab)2 fragment of goat-anti-mouse IgG, 1µg/ml (Jackson Immunoresearch Labs, West Grove, PN, USA). Next a mouse monoclonal antibody against human MMP-3, clone 55-2-A4 (clone B) 36, (Oncogene, Cambridge, MA, USA) was coated, 0.2 µg/ml. Serum samples were analyzed in two-fold serial dilutions in high performance ELISA buffer (CLB, Amsterdam, NL) and incubated during 1 hour. After washing bound MMP-3 was detected with a biotinylated polyclonal sheep-anti-MMP-3 (The Binding Site, Birmingham, UK) in combination with streptavidin-polyHRP (CLB, Amsterdam, NL). Peroxidase activity was determined using tetramethylbenzidin as substrate. MMP-3 levels were calculated at the linear range of the assay from a standard curve (10-1500 ng/ml) using a RA synovial fluid, which was standardized against the BIOTRAK® MMP-3 ELISA kit (Amersham,’s Hertogenbosch, NL). The intra-assay coefficient of variation (CV) was 8.9%, the inter-assay CV 10.1%. With an immunoblot we demonstrated that both the monoclonal and the polyclonal antibody reacted with active MMP-3, pro-MMP-3 as well as with MMP-3 bound to tissue inhibitor of matrix metalloproteinases (TIMP). Treatment of the sera with p-aminophenyl mercury acetate (APMA) 37,38 to activate pro-MMP-3 did not influence results in the ELISA. Furthermore it was demonstrated that rheumatoid factors do not react in this assay and do not interfere with measurement of MMP-3 (data not shown). For normal values of serum MMP-3 the 95 percentile in healthy bloodbank donors was used (female < 20 ng/ml, male < 60 ng/ml). CRP was measured by ELISA 39, ESR according to Westgren. IgM rheumatoid factor (RF) was measured by ELISA 40 (normal value: < 10 IU/ml)). Statistical analysis As progression of radiological damage is a cumulative process, the production of MMP-3 was also expressed as cumulative values. Monthly levels of serum MMP-3 were plotted against time (weeks) and time-integrated values were obtained by calculation of the area under the curve (AUC). Cumulative values were calculated for each 6 month period concomitant with the X-ray intervals. Time-integrated values for the 6 month periods were indicated as δ-MMP-3, δ-TSS etc. In order to obtain comparable variables the serial data of CRP and the clinical variables were also transformed into time-integrated values.

Chapter 4

72

Spearman’s rank correlation coefficient was used for the assessment of correlations between the different variables. RESULTS Baseline characteristics of the 33 patients with early RA are summarized in table 1.

TABLE 1. Characteristics of the 33 patients with early rheumatoid arthritis at study entry Age (years) 50 (17-72) Gender, female/male (% female) 22/11 (67%) Disease duration (months) 6.5 (1.5-12) IgM-RF positive (%) 29 (88%) Tender joint count 19 (1-45) Swollen joint count 13 (1-28) Ritchie articular index 16 (1-50) Disease activity score 4.5 (2.0-7.0) CRP (mg/l.) 25 (0.6-172) ESR (mm/h.) 43 (5-102) MMP-3 (ng/ml.) 77 (5-1807) Sharp score 3 (0-37) Number of patients with normal X-rays (%) 9 (27%) Values are the median and range. Maximal scores: Tender and swollen joint count; 52, Ritchie articular index; 78, Disease activity score; 10.0, Sharp score; 448.

During follow-up 31 out of the 33 patients were treated with DMARDs. Mean values of serum MMP-3, CRP, joint swelling (of serum MMP-3, CRP, joint swelling etc) and radiological progression over 6 months periods are shown in figure 1.


73

FIGURE 1. Mean value over six months of monthly measured variables in an individual patient (•). The variables are MMP-3 (A), CRP (B), and swollen joint count (C). For radiological progression (D) the increment in total Sharp score over a period of six months in an individual patient is shown. The horizontal line represents the median.

Chapter 4

74

Serum MMP-3 vs markers of disease activity Correlation coefficients between time-integrated values for 6 months periods of serum MMP-3 (δ-MMP-3) and markers of disease activity, joint scores, DAS, ESR, and CRP are shown in table 2.

TABLE 2. Spearman correlations between time-integrated values (δ) for 6 months periods of serum MMP-3 with CRP, ESR, disease activity score (DAS), swollen joint count (SJC), tender joint count (TJC), and Ritchie articular index (RAI) (n=33). Time interval months

δ-CRP δ-ESR δ-DAS (δ-SJC) (δ-TJC) δ-RAI

0-6 0.81 *** 0.62 *** 0.23 0.54 *** 0.15 0.17 6-12 0.90 *** 0.57 *** 0.45 ** 0.56 *** 0.21 0.23 12-18 0.88 *** 0.74 *** 0.43 * 0.57 *** 0.17 0.18 18-24 0.86 *** 0.69 *** 0.62 *** 0.78 *** 0.28 0.25 24-30 0.81 *** 0.76 *** 0.61 *** 0.71 *** 0.34 0.30 30-36 0.72 *** 0.68 *** 0.33 0.49 ** 0.12 0.07 0-36 0.90 *** 0.74 *** 0.49 ** 0.70 *** 0.20 0.19 Values are Spearman r. * = P value < 0.05, ** = P value < 0.01, *** = P value < 0.001.

δ-MMP-3 was closely correlated with δ-SJC, δ-ESR and δ-CRP during all periods. Between 6 and 30 months δ-MMP-3 was also correlated with the DAS. There were no correlations between δ-MMP-3 and δ-TJC or δ-RAI. Time-integrated values of serum MMP-3 for the total follow-up period of 36 months showed the same pattern. In particular time-integrated values of serum MMP-3 and CRP were closely related. Figure 2 and 3 (see next page) show the relation between time-integrated values of serum MMP-3 and CRP in the first period of 6 months and over the total follow-up period of 36 months of the total group (n=33).


75

FIGURE 2. Time-integrated serum MMP-3 levels in relation to time-integrated CRP levels of 33 patients in the first period of 6 months. Spearman (r = 0.81, p < 0.0001).

FIGURE 3. Time-integrated serum MMP-3 levels in relation to time-integrated CRP levels over a 3 year follow-up period. Spearman (r = 0.90 p < 0.0001). Also intra-individually there were significant correlations between serum MMP-3 and CRP in 25 of the 33 patients. In 8 patients there were no correlations between serum MMP-3 and CRP. In 2 of these 8 patients serum MMP-3 could not be detected.

Chapter 4

76

Figure 4 shows serial serum MMP-3 and CRP levels of a patient with a close correlation between serum MMP-3 and CRP.

FIGURE 4. Patient data with significant correlations between serial serum MMP-3 (•) and CRP (o) (r = 0.58, p < 0.001) and serial MMP-3 and swollen joint count (shaded area) (r = 0.73, p < 0.001). Inset showing radiological progression. Serum MMP-3 vs radiological damage Table 3 shows the correlation coefficients between δ-MMP-3 for 6 month periods and the progression in radiological scores: δ-JSN, δ-ER and δ-TSS.

TABLE 3. Spearman correlations between time-integrated values for 6 months periods of serum MMP-3 with progression in joint space narrowing (JSN), erosions (ER) and total Sharp score (TSS) (n=33). Time interval (months)

δ-Joint space narrowing (δ-JSN)

δ-Erosions (δ-ER)

δ-Total Sharp score (δ-TSS)

0-6 0.46 ** 0.19 0.33 6-12 0.45 ** 0.12 0.34 12-18 0.73 *** 0.11 0.53 ** 18-24 0.51 ** 0.38 * 0.53 ** 24-30 0.58 *** 0.52 ** 0.64 *** 30-36 0.48 ** 0.49 ** 0.52 ** 0-36 0.56 *** 0.31 0.49 ** Values are Spearman r. * = P value < 0.05, ** = P value < 0.01 and *** = P value < 0.001.


77

During the first 2 periods of 6 months δ-MMP-3 was only correlated with δ-JSN. After 12 months of follow-up it was also correlated with δ-TSS and after 18 months it was correlated with all 3 items of the Sharp score. The intra-individual relations between δ-MMP-3 and δ-TSS during follow-up are shown in figure 5. Despite great differences between individuals, intra-individually the relation between δ-MMP-3 and δ-TSS appeared to be rather constant during follow-up.

FIGURE 5. Intra-individual relations between cumulative serum MMP-3 levels and radiological progression (total Sharp score) in 33 patients during a follow-up of 3 year. DISCUSSION In the present study we found that time-integrated values of serum MMP-3 are correlated with time-integrated values of other markers of disease activity as joint swelling, ESR, CRP and the DAS. Of the radiological scores especially joint space narrowing is closely correlated with cumulative serum MMP-3. In rheumatoid arthritis a distinction must be made between process variables like joint tenderness, joint swelling, ESR, CRP etc. and outcome measures like radiological progression. The course of the disease is generally monitored by serial measurements of one or more process variables. As radiological outcome is essentially the result of what has happened during the course of the disease, theoretically the area under the curve of serially measured process variables meets the requirements of a proper outcome measure. Such a transformation into time-

Chapter 4

78

integrated values enables the comparison of process variables with outcome measures 1,41. As shown in table 2 time-integrated values for 6 months periods of serum MMP-3 were closely correlated with time-integrated values of joint swelling, ESR, and CRP and to a lesser extent with the DAS (as markers of disease activity). There were no correlations with the tender joint count and RAI, scores in which the variable pain plays a dominant role. Correlations between circulating MMP-3 and joint swelling, ESR, or CRP have also been reported by others 6,7,20,27,28,42-44. Our study with serial measurements and transformation of the data into time-integrated values confirms these findings. Theoretically serum MMP-3 could have some advantages as a marker of disease activity in comparison to CRP. In RA MMP-3 is mainly locally produced and activated in the affected joints and in that way it is a more direct reflection of joint inflammation. The acute phase protein CRP is produced in an indirect manner by the liver after cytokine stimulation. This production can also be influenced by other stimuli of the acute phase response, as for example bacterial infections. Nevertheless, there are no stronger associations between serum MMP-3 and markers of disease activity in comparison with CRP, neither in our study nor in other studies28. This confirms the value of CRP level as a good marker for disease activity of RA in clinical practice. In contrast to parameters such as joint swelling, ESR, and CRP, no significant correlations were found between time-integrated values of serum MMP-3 and the RAI and tender joint count. This could also explain the less strong correlation with the DAS, which includes the item Ritchie articular index. In a recent study by Keyszer et al only a weak correlation was found between a visual analog scale of joint pain and serum MMP-3 28. Van Leeuwen et al found close correlations between CRP and joint swelling but not with the number of tender joints or the RAI 1. These results suggest different pathogenetic mechanisms of pain and joint inflammation. In addition nearly all patients used NSAIDs which change symptoms of pain and stiffness without changing the course of the synovitis, joint swelling or the acute phase response. There was a wide variation in the absolute values of serum MMP-3 and CRP between patients conceivably due to inter-individual differences in extent of disease, involved joints, used medication and differences in locally produced cytokines and cytokine inhibitors. Despite these variations we found significant intra-individual correlations between serum MMP-3 and CRP in 25 of 33 patients (data not shown). In 8 patients there were no correlations between serum MMP-3 and CRP. In 2 of these 8 patients serum MMP-3 could not be detected possibly due to a limited


79

sensitivity of the test or to the fact that MMP-3 was not produced at all. In the remaining 6 patients documented infections resulted in clear discrepancies between serum MMP-3 and CRP but there were also several CRP peaks in all 33 patients that could not be explained with the available data. Long-term follow-up studies that link serum MMP-3 to radiological progression are rare. In a recent study by So et al 44 in patients with long standing RA no correlations were found with erosive disease. Recently our group evaluated the prognostic significance of serum MMP-3 in relation to the development of radiological damage in patients with early RA without radiological damage at presentation. At a cut-off point of 80 ng/ml (the upper limit of controls and osteoarthritis patients) a positive predictive value of 91% was found for the development of radiological damage within 2 years. In addition serum MMP-3 at presentation was correlated with the total Sharp score after 6 and 12 months of follow-up. This correlation was almost exclusively determined by the item joint space narrowing in the Sharp score 8. As shown in table 3, time-integrated values of serum MMP-3 in patients with early RA are closely correlated with the total Sharp score, but in particular with joint space narrowing, representing destruction of cartilage. Although cause and effect remains to be elucidated, these results fit with the fact that the main targets of MMP-3 are localized in the matrix of cartilage, like proteoglycans 9,10,12,13 and the fact that in animal models MMP-3 was inducible and detectable in the chondrocytes in a very early phase of the arthritis 9. These findings are of importance for the unraveling of the pathogenesis of RA but also for analyzing mechanisms of drug therapy. For example specific MMP-inhibitors may uncouple the relation between CRP and radiological damage. Radiological progression could be stopped by these agents although inflammation and acute phase response may continue. In that case new markers, not only stromelysins (MMP-3), but also upregulated collagenases or gelatinases could become essential. As shown in figure 5 there were great inter-individual differences in the relation between serum MMP-3 level and radiological progression. Intra-individually this relation seemed to be rather constant over time. The inter-individual differences might be explained by differences in involved joints. For example, large joints, like the knee, producing much MMP-3, could have a great influence on the serum MMP-3 level while in Sharp’s method only small hand and feet joints are scored. Another explanation could be that different RA patients have different profiles of cytokines and cytokine inhibitors resulting in different responses of MMP production, activation and joint destruction. Furthermore, in RA MMP-3 is produced in the affected joint as pro-MMP-3, which is locally activated by other proteinases. After activation it can degrade many components of the matrix or it is inactivated by tissue

Chapter 4

80

inhibitors of matrix metalloproteinases (TIMPs) and/or by α2 macroglobulin. Inter-individual differences in production site and/or activation level of pro MMP-3 and levels of inhibitors could explain the wide inter-individual variations. Finally, MMP-3 is not the only important enzyme. Inter-individual differences in the upregulation of other MMPs, like collagenases could also have an influence on the individual relation between serum MMP-3 and radiological damage. The rather constant intra-individual relation between serum MMP-3 and radiological progression resembles the relation between CRP and radiological progression 2,45, a result which could be expected because of the close relation between CRP and MMP-3. In conclusion time-integrated values of serial serum MMP-3 measurements showed close correlations with other markers of disease activity as joint swelling, ESR, CRP, and the DAS in patients with early RA. Of the radiological scores, as outcome measures, especially the item joint space narrowing was closely correlated with cumulative serum MMP-3. We therefore propose that serum MMP-3 may be an additional serological marker indicative of joint inflammation and cartilage degrading activity in rheumatoid arthritis. ACKNOWLEDGEMENTS This study was supported by a grant from “The Dutch Arthritis Association”, The Netherlands.


81

REFERENCES 1. van Leeuwen MA, van der Heijde DM, van Rijswijk MH, Houtman PM, van Riel PL, van de Putte

LB, et al. Interrelationship of outcome measures and process variables in early rheumatoid arthritis. A comparison of radiologic damage, physical disability, joint counts, and acute phase reactants. J Rheumatol 1994;21:425-9.

2. van Leeuwen MA, van Rijswijk MH, van der Heijde DM, Te Meerman G, van Riel PL, Houtman PM, et al. The acute-phase response in relation to radiographic progression in early rheumatoid arthritis: a prospective study during the first three years of the disease. Br J Rheumatol 1993;32 Suppl 3:9-13.

3. van der Heijde DM, van Riel PL, van Leeuwen MA, van 't Hof MA, van Rijswijk MH, van de Putte LB. Prognostic factors for radiographic damage and physical disability in early rheumatoid arthritis. A prospective follow-up study of 147 patients. Br J Rheumatol 1992;31:519-25.

4. Stenger AA, van LM, Houtman PM, Bruyn GA, Speerstra F, Barendsen BC, et al. Early effective suppression of inflammation in rheumatoid arthritis reduces radiographic progression. Br J Rheumatol 1998;37:1157-63.






10. Nagase H, Okada Y. Proteinases and matrix degradation. In: Kelley WN, Harris ED, Ruddy S, Sledge CB, editors. Textbook of rheumatology. 5th ed. Philadelphia: Saunders,W.B., 1997:323-41.

11. Mudgett JS, Hutchinson NI, Chartrain NA, Forsyth AJ, McDonnell J, Singer II, et al. Susceptibility of stromelysin 1-deficient mice to collagen-induced arthritis and cartilage destruction. Arthritis Rheum 1998;41:110-21.

12. Singer II, Kawka DW, Bayne EK, Donatelli SA, Weidner JR, Williams HR, et al. VDIPEN, a metalloproteinase-generated neoepitope, is induced and immunolocalized in articular cartilage during inflammatory arthritis. J Clin Invest 1995;95:2178-86.


14. Nagase H. Activation mechanisms of matrix metalloproteinases. Biol Chem 1997;378:151-60. 15. McCachren SS. Expression of metalloproteinases and metalloproteinase inhibitor in human arthritic

synovium. Arthritis Rheum 1991;34:1085-93. 16. Okada Y, Takeuchi N, Tomita K, Nakanishi I, Nagase H. Immunolocalization of matrix

metalloproteinase 3 (stromelysin) in rheumatoid synovioblasts (B cells): correlation with rheumatoid arthritis. Ann Rheum Dis 1989;48:645-53.



Chapter 4

82





23. Beekman B, van EB, Drijfhout JW, Ronday HK, TeKoppele JM. Highly increased levels of active stromelysin in rheumatoid synovial fluid determined by a selective fluorogenic assay. FEBS Lett 1997;418:305-9.

24. Barrett AJ. Alpha 2-macroglobulin. Methods Enzymol 1981;80 Pt C:737-54. 25. Zucker S, Lysik RM, Zarrabi MH, Greenwald RA, Gruber B, Tickle SP, et al. Elevated plasma

stromelysin levels in arthritis. J Rheumatol 1994;21:2329-33. 26. Kotajima L, Aotsuka S, Fujimani M, Okawa-Takatsuji M, Kinoshita M, Sumiya M, et al. Increased

levels of matrix metalloproteinase-3 in sera from patients with active lupus nephritis. Clin Exp Immunol 1998;16:409-15.

27. Ichikawa Y, Yamada C, Horiki T, Hoshina Y, Uchiyama M. Serum matrix metalloproteinase-3 and fibrin dagradation product levels correlate with clinical disease activity in rheumatoid arthritis. Clin Exp Rheumatol 1999;16:533-40.

28. Keyszer G, Lambiri I, Nagel R, Keysser C, Keysser M, Gromnica-Ihle E, et al.. Circulating levels of matrix metalloproteinases MMP-3 and MMP-1, tissue inhibitor metalloproteinases 1 (TIMP-1), and MMP-1/TIMP-1 complex in rheumatic disease. Correlation with clinical activity of rheumatoid arthritis versus other surrogate markers. J Rheumatol 1999;26:251-8.

29. Lohmander LS, Hoerrner LA, Lark MW. Metalloproteinases, tissue inhibitor, and proteoglycan fragments in knee synovial fluid in human osteoarthritis. Arthritis Rheum 1993;36:181-9.


31. Arnett FC, Edworthy SM, Bloch DA, McShane DJ, Fries JF, Cooper NS, et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 1988;31:315-24.

32. Ritchie DM, Boyle JA, McInnes JM, Jasani MK, Dalakos TG, Grieveson P, et al. Clinical studies with an articular index for the assessment of joint tenderness in patients with rheumatoid arthritis. Q J Med 1968;37:393-406.

33. van der Heijde DM, van 't Hof MA, van Riel PL, Theunisse LA, Lubberts EW, van Leeuwen MA, et al. Judging disease activity in clinical practice in rheumatoid arthritis: first step in the development of a disease activity score. Ann Rheum Dis 1990;49:916-20.



36. Obata K, Iwata K, Okada Y, Kohrin Y, Ohuchi E, Yoshida S, et al. A one-step sandwich enzyme immunoassay for human matrix metalloproteinase 3 (stromelysin-1) using monoclonal antibodies. Clin Chim Acta 1992;211:59-72.


83

37. Smith-GN J, Brandt KD, Hasty KA. Activation of recombinant human neutrophil procollagenase in the presence of doxycycline results in fragmentation of the enzyme and loss of enzyme activity. Arthritis Rheum 1996;39:235-44.

38. Okada Y, Harris-ED J, Nagase H. The precursor of a metalloendopeptidase from human rheumatoid synovial fibroblasts. Purification and mechanisms of activation by endopeptidases and 4-aminophenylmercuric acetate. Biochem J 1988;254:731-41.

39. van Leeuwen MA, van Rijswijk MH, Marrink J, Westra J, de Jong HJ. CRP measurements in rheumatic disorders. Protides of the biological fluids 1986;34:315-8.

40. van Leeuwen MA, Westra J, van Riel PL, Limburg PC, van Rijswijk MH. IgM, IgA, and IgG rheumatoid factors in early rheumatoid arthritis predictive of radiological progression? Scand J Rheumatol 1995;24:146-53.

41. Matthews JN, Altman DG, Campbell MJ, Royston P. Analysis of serial measurements in medical research. BMJ 1990;300:230-5.



44. So A, Chamot AM, Peclat V, Gerster JC. Serum MMP-3 in rheumatoid arthritis: correlation with systemic inflammation but not with erosive status. Br J Rheumatol 1999;38:407-10.

45. van Leeuwen MA, van Rijswijk MH, Sluiter WJ, van Riel PL, Kuper IH, van-de Putte LB, et al. Individual relationship between progression of radiological damage and the acute phase response in early rheumatoid arthritis. Towards development of a decision support system. J Rheumatol 1997;24:20-7.

Chapter 5

Serum matrix metalloproteinase 3 levels during treatment with sulphasalazine or the combination

of methotrexate and sulphasalazine in patients with early rheumatoid arthritis

Marcel D. Posthumus, Pieter C. Limburg, Johanna Westra, Miek A. van Leeuwen, and Martin H. van Rijswijk.

Journal of Rheumatology 2002;29:883-889.

Chapter 5

86

SUMMARY Objective. To determine the effects of treatment with sulphasalazine (SSZ) or the

combination of methotrexate (MTX) and SSZ on serum matrix metalloproteinase 3 (MMP-3) levels in patients with early rheumatoid arthritis (RA).

Methods. Eighty-two patients with early RA (symptoms < 1 year and DMARD-naive at presentation) were selected who had been treated with SSZ (2000 mg/day) or with the combination of MTX (7.5-15 mg/week) and SSZ. Serum MMP-3 levels, C-reactive protein (CRP), erythrocyte sedimentation rate (ESR), swollen joint count (SJC), tender joint count (TJC), Ritchie articular index (RAI), and the disease activity score (DAS) were determined at 4 week intervals during a follow-up of 28 weeks for each treatment group. Response was based on clinical grounds and CRP at 12, 20, and 28 weeks.

Results. SSZ responders (n=52) had lower baseline values of serum MMP-3, CRP, and ESR, compared to partial/non-responders (n=30), but did not differ in joint scores and DAS. In the SSZ responder group all variables decreased. In the SSZ partial/non-responders CRP, ESR, and SJC decreased in contrast to serum MMP-3, TJC, RAI, and DAS-3. After addition of MTX all variables decreased in 24 out of the 30 patients who had shown a partial or no response on SSZ. In the SSZ responders there was a delayed decrease in serum MMP-3 compared to CRP.

Conclusions. Serum MMP-3 levels decrease in early RA patients who respond to SSZ or to the combination of MTX and SSZ. In patients who respond to SSZ the changes in serum MMP-3 levels indicate a delayed response compared to CRP. Key words: serum matrix metalloproteinase 3, stromelysin 1, early rheumatoid arthritis, disease modifying antirheumatic drugs, sulphasalazine, methotrexate.

Serum MMP-3 during DMARD treatment

87

INTRODUCTION In patients with rheumatoid arthritis (RA) matrix metalloproteinase 3 (MMP-3 = stromelysin 1) is of interest because this proteolytic enzyme is thought to play a prominent role in the pathogenesis of matrix degradation 1-3 even though it is not the only key enzyme 4. In RA MMP-3 is locally produced and activated within the inflamed joint and released into the peripheral blood. Systemic MMP-3 levels are a reflection of local synthesis. As such, serum MMP-3 can be used as a systemic marker of local joint inflammation 5-9 and/or destruction 10,11. Studies concerning MMPs are not only of importance for the unraveling of the pathogenesis of RA but also for analyzing mechanisms of drug therapy. For example specific MMP-inhibitors may uncouple the relation between surrogate markers of joint inflammation (for example C-reactive protein (CRP)) and joint damage. Radiological progression might be stopped by these new agents without inhibiting inflammation and acute phase response. In that case, especially in early disease, new markers like systemic MMP, including its level of activation, are essential 12. Sulphasalazine (SSZ) and methotrexate (MTX) are disease modifying antirheumatic drugs (DMARDs), widely used in early RA to suppress disease activity and thereby to prevent or delay joint destruction 13,14. Studies investigating the effects of SSZ or MTX on MMPs, especially on MMP-3, are rare. Therefore we analyzed in a prospective follow-up study of early RA patients (symptoms < 1 year) whether serum matrix metalloproteinase 3 levels are influenced by treatment with SSZ or the combination of MTX and SSZ. In responders to these treatments we also analyzed the influence on serum MMP-3 levels in relation to the influence of treatment on conventional clinical and other biochemical disease activity measures. PATIENTS AND METHODS Patients Eighty-two patients with early rheumatoid arthritis were selected from a cohort of patients with RA, according to the 1987 ACR criteria 15, with joint symptoms existing less than one year at presentation and who had not previously received disease modifying anti rheumatic drugs (DMARDs). These patients participated in a prospective follow-up study at the department of Rheumatology at the Groningen University Hospital. Clinical and laboratory investigations were performed at monthly intervals during a follow-up of 28 weeks. Clinical markers of disease activity Fifty-two peripheral joints were examined for tenderness and soft tissue swelling. The following articular indices were determined: Ritchie articular index (RAI) 16,

Chapter 5

88

tender joint count (TJC), swollen joint count (SJC) and the disease activity score (DAS) according to Van der Heijde with 3 variables 17. Laboratory analysis Serum MMP-3 levels were determined by a MMP-3 ELISA developed at our laboratory. In short, 96 well plates were precoated with F(ab)2 fragment of goat-anti-mouse IgG, 1µg/ml (Jackson Immunoresearch Labs, West Grove, PN, USA). Next a mouse monoclonal antibody against human MMP-3, clone 10D6 (R&D Systems, Abingdon, UK) was coated at 0.1 µg/ml. Serum samples were analyzed in two-fold serial dilutions in high performance ELISA buffer (CLB, Amsterdam, NL) and incubated during 1 hour. After washing bound MMP-3 was detected with a polyclonal rabbit anti-human MMP-3 (AB 810, Chemicon, Temecula, CA, USA), followed by horseradish-peroxidase-labelled F(ab)2 fragment of goat-anti-rabbit IgG (Zymed, San Francisco, CA, USA). Peroxidase activity was determined using tetramethylbenzidin as substrate. MMP-3 levels were calculated at the linear range of the assay from a standard curve (3-400 ng/ml) using pro-MMP-3, which was purified from serum free supernatant of IL-1β stimulated rheumatoid arthritis synovial fibroblasts 18. The intra-assay coefficient of variation (CV) was 6.8%, the inter-assay CV 8.8%. With an immunoblot we demonstrated that both the monoclonal and the polyclonal antibody reacted with active MMP-3, pro-MMP-3 as well as with MMP-3 bound to tissue inhibitor of matrix metalloproteinases (TIMPs). Furthermore it was demonstrated that rheumatoid factors do not react in this assay and do not interfere with measurement of MMP-3 (data not shown). For normal values of serum MMP-3 the 95 percentile in healthy bloodbank donors (n=80) was used (female < 20 ng/ml, male < 60 ng/ml). CRP was measured by ELISA 19, ESR according to Westgren. IgM rheumatoid factor (RF) was measured by Dade/Behring BN-2 nephelometer (normal value: < 15 IU/ml). DMARD treatment and definition of response During follow-up patients were treated with non-steroidal anti-inflammatory drugs (NSAIDs) as indicated clinically. It is our policy to use an intensive treatment strategy in early RA patients consisting of an immediate start of DMARDs and rapid adjustment of dosage and/or drugs (step up method) in case of an insufficient response 20. DMARD treatment was instituted according to the following guidelines: at study entry all patients with active disease started with SSZ 500 mg/day, increased to 2000 mg/day in weekly increments of 500 mg. In case of an insufficient response at week 12, MTX 7,5 mg/week could be added to SSZ 2000 mg/day. At week 20 the


89

MTX dose could be increased to 15 mg/week. Corticosteroids were allowed as adjuvant therapy. Decisions about intensifying the treatment with DMARDs, at 12 and 20 weeks were discussed by an independent observer and the patients rheumatologist. These decisions were based on clinical markers of disease activity in combination with the CRP level as effect measures. Patients with a sufficient response, ≥ 50% reduction in joint scores or CRP (or normalization of CRP (< 5 mg/l)) were assumed to be SSZ responders. A SSZ partial responder showed some improvement but less than 50% in joint scores or CRP. SSZ non-responders did not show any response or had deteriorated. Statistical analysis Chi-square and Mann Whitney U were used to detect differences between groups. The Friedman test with the Dunnett’s post test were used to analyze differences within groups. RESULTS All patients started with SSZ. After 12 weeks of treatment there were 2 groups: SSZ responders (n=52), and SSZ partial/non-responders (n=30). Of these 30 SSZ partial/non responders 24 patients had a sufficient response on subsequent combination therapy with MTX/SSZ (MTX/SSZ responders (n=24)). The 6 MTX/SSZ non responders were subsequently treated with the combination of MTX/cyclosporine. SSZ responders After 12 weeks of SSZ treatment the response was determined as described in the method section. Sixteen patients (31%) had ≥ 50% reduction in CRP, 13 patients (25%) ≥ 50% reduction in joint scores and 23 patients (44%) ≥ 50% in both variables. In table 1 the baseline characteristics of the SSZ responders (n=52) and the SSZ partial/non responders (n=30) are shown. The SSZ responders had lower baseline values of serum MMP-3, CRP, and ESR levels but did not differ in joint scores or DAS.

Chapter 5

90

TABLE 1. Baseline characteristics of 82 patients with early rheumatoid arthritis; sulfasalazine (SSZ) responders (n=52) and SSZ partial/non-responders (n=30).

SSZ responders (N=52)

SSZ partial/non responders (N=30)

Age (years) 49.8 (20.3-78.6) 51.4 (21.5-66.2) Gender, female/male (% female) 37/15 (71) 16/14 (53) IgM-RF positive (%) 45 (87) 26 (87) Tender joint count 12 (0-35) 12 (0-38) Swollen joint count 11 (3-29) 12 (4-28) Ritchie articular index 6 (0-34) 6 (0-24) Disease activity score 3.52 (1.73-5.69) 3.76 (1.88-5.55) Serum MMP-3 (ng/ml.) 64 (14-495) 112 (17-1290) ¶ CRP (mg/l.) 14.5 (2-110) 33 (2-139) ¶ ESR (mm/h.) 27 (5-114) 40 (9-114) ¶ Sharp > 0 (%) 32 (63) 16 (53) Values are the median and range. Chi-square for gender, IgM-RF positivity and Sharp > 0. Mann Whitney U for the other variables. ¶: SSZ-responders vs. SSZ-partial/non-responders p < 0.01.

After 12 weeks of SSZ all variables were significantly decreased in the SSZ

responders (table 2). In the SSZ partial/non responders CRP, ESR, and SJC decreased in contrast to serum MMP-3, TJC, RAI and DAS-3 (table 3).

TABLE 2. Median values of clinical and biochemical variables in SSZ responders (n=52) during 28 weeks of follow-up. Week MMP-3

(ng/ml) CRP (mg/l)

ESR (mm/h)

SJC TJC RAI DAS-3

0 64 14 27 11 12 6 3.52 4 68 13 22 9 7 5 2.97 8 56 9 15 7 6 4 2.54 12 ¶ 55 8 13 5 4 2 2.32 16 44 6 10 3 2 2 1.89 20 38 4 8 3 2 2 1.96 24 39 4 8 1 1 1 1.72 28 ¶ 38 4 9 2 1 1 1.56 ¶: after 12 weeks of sulphasalazine the response was determined. In the SSZ responders all variables were significantly decreased at week 12 and at the end of the follow-up, at 28 weeks (P < 0.05, Friedman test).


91

TABLE 3. Median values of clinical and biochemical variables in SSZ partial/non responders (n=30) at study entry and at 4, 8 and 12 weeks.

Week MMP-3 (ng/ml)

CRP (mg/l)

ESR (mm/h)

SJC TJC RAI DAS-3

0 112 33 40 12 12 6 3.76 4 136 35 36 13 14 7 3.67 8 121 30 35 12 9 7 3.40 12 ¶ 132 17 30 9 10 6 3.27 P ns < 0.01 < 0.01 0.03 ns ns ns ¶: after 12 weeks of sulphasalazine the response was determined. In the SSZ partial/non responders CRP, ESR and swollen joint count were significantly decreased (P < 0.05, Friedman test) in contrast to serum MMP-3, TJC, RAI and DAS-3.

Effects of MTX in SSZ partial/non responders Study variables at the moment of the addition of MTX were taken as a new baseline (week 0 MTX). In most patients MTX was added at week 12 or 16. Due to this variation the values at week 12 in SSZ partial/non responders (table 3) were not exactly the same as the values at week 0 MTX (table 4). Ten patients (42%) had a ≥ 50% reduction in CRP, 4 patients (16%) a ≥ 50% reduction in joint scores and 10 patients (42%) a ≥ 50% in both variables. Because of the small size of the non responder group (n=6) only the responders on MTX/SSZ (n=24) were evaluated.

The effects of adding MTX to SSZ partial/non-responders during a follow-up of 28 weeks is shown in table 4. All variables decreased significantly.

TABLE 4. Median values of clinical and biochemical variables in MTX/SSZ responders (n=24) during 28 weeks after the addition of MTX.

Weeks MTX

MMP-3 (ng/ml)

CRP (mg/l)

ESR (mm/h)

SJC TJC RAI DAS-3

0 145 29 26 11 11 6 3.31 4 102 15 18 7 6 5 2.72 8 79 13 13 7 3 3 2.61 12 73 12 8 5 7 4 2.48 16 64 10 10 5 4 3 2.16 20 50 7 10 2 2 2 1.89 24 45 6 7 3 2 2 1.93 28 46 6 10 2 1 1 1.68 P ¶ < 0.01 < 0.01 < 0.01 < 0.01 < 0.01 < 0.01 < 0.01 ¶: During follow-up all variables decreased significantly (P < 0.05, Friedman test)

Chapter 5

92

Serum MMP-3 levels compared to other disease activity variables SSZ responders (n = 52) and MTX/SSZ responders (n = 24) were eventually followed for 28 weeks. In both groups serum MMP-3 and CRP decreased during follow-up (tables 2 and 4). To evaluate differences between serum MMP-3 and CRP, changes of each variable were expressed as a percentage of the initial level. Serum MMP-3 was significantly decreased after 16 weeks in both groups (figure 1). A significant reduction in CRP was reached after 8 weeks of treatment in both groups. In both responder groups there were some patients with serum MMP-3 levels in the normal range (female < 20 ng/ml, male < 60 ng/ml). To analyze the influence of these ”normal levels” on the overall results we separately evaluated the patients with an elevated serum MMP-3 level at study entry. The results were the same in the SSZ responders (n=42 out of 52). In the MTX/SSZ responders (n=22 out of 24) serum MMP-3 and CRP were both significantly decreased after 12 weeks (data not shown).

We evaluated differences between serum MMP-3 and CRP by changes expressed as a percentage of the initial level. Small variations in or close by the normal range could result in exceptional relative (in terms of percentage) elevations. In some instances serum MMP-3 and CRP were, for example > 400% from baseline (see figure 1). Further evaluation of these data showed that these exceptional, relative elevations were indeed mainly caused by variations within or close by the normal range.


93

FIGURE 1. Serum MMP-3 and CRP as a % of the initial level in SSZ responders (A & B (n=52)) and MTX/SSZ responders (C & D (n=24)) All variables decreased during a follow-up of 28 weeks. P < 0.05 was reached after 16 weeks for serum MMP-3 and after 8 weeks for CRP (Friedman with Dunnett's post test on the original crude data).

Chapter 5

94

Inter-individual differences in the course of serum MMP-3 and CRP levels Data from 4 randomly selected patients are shown in figure 2. These data are illustrative for the wide variations in absolute values of serum MMP-3 and CRP, for the close relation between MMP-3 and CRP, and for the individual response on treatment.

FIGURE 2. Individual serum MMP-3 (•) and CRP (o) levels in SSZ-responders (patient A and B) and MTX/SSZ responders (patient C and D). Broken line represents the normal value for serum MMP-3: for women < 20 (patients A, B, C) and for men < 60 ng/ml (patient D)/ DISCUSSION Our data show that serum MMP-3 levels decrease in early RA patients who respond to SSZ or to the combination of MTX and SSZ. Actually all variables, including CRP, ESR, and clinical variables decreased, confirming the close relation between serum MMP-3 and markers of disease activity. The MMPs are thought to play an important role in the pathogenesis of RA based on their capacity to degrade many matrix components, their local expression in synovial tissue, and their increased levels in synovial fluid and serum 2,3. Inhibition of the production and/or activation of these MMPs could be an explanation for the


95

restraining effects of DMARDs on radiological progression. Therefore it is of interest to investigate the influence of DMARDs like SSZ and MTX on MMPs. In RA MMP-3 is locally produced and activated in the inflamed joints and systemic levels are a direct reflection of this local synthesis. This is in contrast to CRP which is an indicator of inflammation in general that may be influenced by other stimuli of the acute phase response, like infections 21,22. Especially in early disease the use of markers of joint inflammation and destruction are of importance for prognostic and therapeutical reasons 23. Therefore serum MMP-3 is an interesting marker to investigate the influence of SSZ or the combination of MTX/SSZ on MMP production. Studies concerning effects of DMARDS, such as sulfasalazine and methotrexate on MMPs are scarce. There is growing evidence that Nuclear Factor κB (NF-κB) is involved in MMP induction. NF-κB is an important transcription factor for inflammatory cytokine genes such as TNF-α, IL-1 and IL-6 24 as well as MMPs such as MMP-3 25,26. SSZ is a potent and specific inhibitor of NF-κB in in-vitro cell cultures by interfering with IκBα phosphorylation 27. By this path the effects of SSZ on MMP-3 and thereby on serum MMP-3 levels could be explained. MTX in combination with steroids was reported to be effective in reducing neutral protease activity in RA synovium and cartilage 28. In a more recent study MTX therapy decreased collagenase (MMP-1) but not stromelysin-1 (MMP-3) gene expression in synovium of RA patients 29. More detailed studies concerning the effects of MTX on for example transcription factors have, to our knowledge, not been published 30. Nevertheless, MTX interferes with pro-inflammatory cytokines like IL-1 which is a potent inducer of MMPs by activation of transcription factors such as activator protein 1 (AP-1) and NF-κB 31. With regard to serum MMP-3 levels, treatment with anti-TNF-α (chimeric monoclonal antibody) resulted in a significant and rapid fall of serum MMP-3 32.

Studies concerning corticosteroids showed conflicting results: intra-articular steroids resulted in a decrease 5 but systemic corticosteroids resulted in an increase in serum MMP-3 33. A prospective open label trial concerning MTX and Tenidap showed close correlations between the DAS, ESR, and CRP, and serum MMP-3 but effects on the absolute values were not given 6. In our study all variables decreased in the SSZ and MTX/SSZ responders. This was to be expected considering the clinical definition of response and the close relation between serum MMP-3 and markers of disease activity like swollen joint count, CRP, and ESR 5,7-9,11,32,34-36. In the SSZ partial/non responders ESR, CRP, and SJC decreased, but obviously not sufficient to the opinion of the patients

Chapter 5

96

rheumatologist (table 3). After the addition of MTX 24 out of 30 patients showed a sufficient response and in these patients all variables decreased significantly (table 4). The serum MMP-3 levels decreased in both responder groups; however, there were differences in comparison with CRP. Especially in the SSZ responders serum MMP-3 showed a delayed response in comparison with CRP (figure 1). This difference did not change when only patients with an initially elevated serum MMP-3 (n=42) were evaluated. In the MTX/SSZ responders serum MMP-3 showed only a delayed response when the complete group was evaluated. Analysis of only the patients with an initially elevated serum MMP-3 (n=22) showed a significant response of serum MMP-3 comparable to the CRP response. This delayed response of serum MMP-3 in comparison with CRP could be explained by differences in metabolism of MMP-3 but, as far as we know there are no data concerning the half-life and clearance of MMP-3 or MMP-3/TIMP complexes. Furthermore there is the possibility of an uncoupled relationship between joint inflammation and joint damage. The differences between CRP and serum MMP-3 may indicate that inhibition of the inflammatory response, resulting in decreased production of IL-6 and subsequent decline in serum CRP does not immediately result in decline in MMP production. This may implicate a temporary cytokine independent production (”autonomous production”) of MMP by synovial cells reflecting continuing matrix degradation. Data from individual patients are shown in figure 2. Although there was a wide inter-individual variation in absolute values, intra-individually there was a close relation between serum MMP-3 and CRP, consistent with our previous study 11. In conclusion serum MMP-3 levels decrease in early RA patients who respond to sulphasalazine or to the combination of methotrexate and sulphasalazine. In patients who respond to sulphasalazine the changes in serum MMP-3 levels indicate a delayed response compared to CRP. ACKNOWLEDGEMENTS This study was supported by a grant from “The Dutch Arthritis Association”, The Netherlands.


97

REFERENCES 1. Hasty KA, Reife RA, Kang AH, Stuart JM. The role of stromelysin in the cartilage destruction that

accompanies inflammatory arthritis. Arthritis Rheum 1990;33:388-97. 2. Nagase H, Okada Y. Proteinases and matrix degradation. In: Kelley WN, Harris ED, Ruddy S, Sledge

CB, editors. Textbook of rheumatology. Philadelphia: Saunders,W.B., 1997:323-41. 3. Okada Y. Proteinases and matrix degradation. In: Ruddy S, Harris-ED J, Sledge CB, Budd RC,

Sergent JS, editors. Kelley's Textbook of rheumatology. Philadelphia: Sanders,W.B., 2001:55-72. 4. Mudgett JS, Hutchinson NI, Chartrain NA, Forsyth AJ, McDonnell J, Singer II et al. Susceptibility of

stromelysin 1-deficient mice to collagen-induced arthritis and cartilage destruction. Arthritis Rheum 1998;41:110-21.








12. Greenwald RA. Thirty-six years in the clinic without an MMP inhibitor. What hath collagenase wrought? Ann N Y Acad Sci 1999;878:413-9.

13. Fries JF. Current treatment paradigms in rheumatoid arthritis. Rheumatology (Oxford) 2000; 39 Suppl 1:30-5.

14. Pincus T, O'Dell JR, Kremer JM. Combination therapy with multiple disease-modifying antirheumatic drugs in rheumatoid arthritis: a preventive strategy. Ann Intern Med 1999;131:768-74.

15. Arnett FC, Edworthy SM, Bloch DA, McShane DJ, Fries JF, Cooper NS et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 1988; 31:315-24.

16. Ritchie DM, Boyle JA, McInnes JM, Jasani MK, Dalakos TG, Grieveson P et al. Clinical studies with an articular index for the assessment of joint tenderness in patients with rheumatoid arthritis. Q J Med 1968;37:393-406.

17. van der Heijde DM, van 't Hof M, van Riel PL, Theunisse LA, Lubberts EW, van Leeuwen MA et al. Judging disease activity in clinical practice in rheumatoid arthritis: first step in the development of a disease activity score. Ann Rheum Dis 1990;49:916-20.

18. Lark MW, Walakovits LA, Shah TK, Vanmiddlesworth J, Cameron PM, Lin TY. Production and purification of prostromelysin and procollagenase from IL-1 beta-stimulated human gingival fibroblasts. Connect Tissue Res 1990;25:49-65.

Chapter 5

98


20. Stenger AA, van LM, Houtman PM, Bruyn GA, Speerstra F, Barendsen BC et al. Early effective suppression of inflammation in rheumatoid arthritis reduces radiographic progression. Br J Rheumatol 1998; 37(11):1157-63.




24. Epstein FH. Nuclear factor-kB. A pivotal transcription factor in chronic inflammatory diseases. N Engl J Med 2001;336:1066-71.

25. Han Z, Boyle DL, Manning AM, Firestein GS. AP-1 and NF-kappaB regulation in rheumatoid arthritis and murine collagen-induced arthritis. Autoimmunity 1998;28:197-208.

26. Bondeson J, Brennan F, Foxwell B, Feldmann M. Effective adenoviral transfer of IkappaBalpha into human fibroblasts and chondrosarcoma cells reveals that the induction of matrix metalloproteinases and proinflammatory cytokines is nuclear factor- kappaB dependent. J Rheumatol 2000;27:2078-89.

27. Wahl C, Liptay S, Adler G, Schmid RM. Sulfasalazine: a potent and specific inhibitor of nuclear factor kappa B. J Clin Invest 1998;101:1163-74.

28. Martel Pelletier J, Cloutier JM, Pelletier JP. In vivo effects of antirheumatic drugs on neutral collagenolytic proteases in human rheumatoid arthritis cartilage and synovium [see comments]. J Rheumatol 1988;15:1198-1204.

29. Firestein GS, Paine MM, Boyle DL. Mechanisms of methotrexate action in rheumatoid arthritis. Selective decrease in synovial collagenase gene expression. Arthritis Rheum 1994;37:193-200.

30. Genestier L, Paillot R, Quemeneur L, Izeradjene K, Revillard JP. Mechanisms of action of methotrexate. Immunopharmacology 2000;47:247-57.

31. Firestein GS, Manning AM. Signal transduction and transcription factors in rheumatic disease. Arthritis Rheum 1999;42:609-21.


33. Sharif M, Salisbury C, Taylor DJ, Kirwan JR. Changes in biochemical markers of joint tissue metabolism in a randomized controlled trial of glucocorticoid in early rheumatoid arthritis. Arthritis Rheum 1998; 41:1203-9.


35. Manicourt DH, Fujimoto N, Obata K, Thonar EJ. Levels of circulating collagenase, stromelysin-1, and tissue inhibitor of matrix metalloproteinases 1 in patients with rheumatoid arthritis. Relationship to serum levels of antigenic keratan sulfate and systemic parameters of inflammation. Arthritis Rheum 1995; 38:1031-9.


Chapter 6

Serum matrix metalloproteinase 3 levels in comparison to C-reactive protein in periods with and without progression of radiological damage

in patients with early rheumatoid arthritis

Marcel D. Posthumus, Pieter C. Limburg, Johanna Westra Miek A. van Leeuwen, and Martin H. van Rijswijk.

Clinical and Experimental Rheumatology 2003;21:465-472.

Chapter 6

100

SUMMARY Objective. To evaluate serum matrix metalloproteinase 3 (MMP-3) levels in

comparison to C-reactive protein (CRP) in periods with and without progression of radiological damage in patients with early rheumatoid arthritis (RA).

Methods. Thirty-two patients with RA and radiological progression (> 5 points according to the Sharp/van der Heijde method) during 6 months followed by a 6 months period without radiological progression (< 1 point) were selected from a prospective follow-up study of early RA patients. Serum MMP-3 levels, CRP, erythrocyte sedimentation rate (ESR), disease activity index (DAS), swollen joint count (SJC), tender joint count (TJC), Ritchie articular index (RAI) were measured monthly and results were transformed into mean values for the 6 months periods.

Results. During the period with radiological progression the mean serum MMP-3 correlated significantly with the mean CRP (r = 0.68, p < 0.001)), ESR (r = 0.54, p = 0.001) and swollen joint count (r = 0.48, p = 0.006). In the period without radiological progression the mean serum MMP-3 only correlated with the mean CRP (r = 0.44, p = 0.012). Individual changes - in terms of percentage (%) - between the 2 periods showed a decrease in both mean serum MMP-3 and CRP in 19 and an increase in 3 patients, in parallel with other markers of disease activity in these patients (69% of cases). The individual change (%) in mean serum MMP-3 or CRP did not correlate with the difference in radiological progression between the 2 periods.

Conclusions. Serum MMP-3 and CRP are closely related and there seems to be no difference between serum MMP-3 and CRP with regard to the monitoring of the progression of radiological damage. Key words: serum matrix metalloproteinase 3, stromelysin 1, early rheumatoid arthritis, radiological damage.

Serum MMP-3 and radiological damage

101

INTRODUCTION Rheumatoid arthritis (RA) is characterized by chronic inflammation of synovial tissue and in most cases progressive destruction of cartilage and bone 1. The matrix metalloproteinases (MMPs) are thought to play a critical role in the degra-dation of many components of the extracellular matrix in the synovial joint 2,3.

Matrix metalloproteinase 3 (MMP-3, stromelysin 1) is of interest because this proteolytic enzyme is produced abundantly in the inflamed joints and plays a prominent role in the pathogenesis of matrix degradation in RA 2-4, even though it is not the only key-enzyme 5. MMP-3 is capable of degrading many components of the matrix in the synovial joint including proteoglycans, gelatins, laminin, fibronectin, and collagen III, IV, IX 2,4,6,7. Moreover MMP-3 is able to activate other matrix metalloproteinases such as MMP-1, MMP-7, MMP-8, MMP9 and MMP-13 8. The enzyme has been localized in the fibroblast-like synoviocytes of rheumatoid synovium 9-11, in RA cartilage 12,13, at sites of cartilage erosion 4, in synovial fluid 14-21, and in serum 20,22-26. Systemic MMP-3 levels are supposed to be a reflection of local synthesis induced by pro-inflammatory cytokines. As such, serum MMP-3 can be used as a systemic marker of local joint inflammation 16,20,24,25,27,28 and/or destruction 26,29,30. In this respect serum MMP-3 may reflect joint inflammation and destruction more directly compared to C-reactive protein (CRP) which is produced indirectly by the liver after cytokine stimulation 31. In addition it has been suggested that the pathophysiologic mechanisms of joint inflammation, as reflected by CRP may be partially independent of destruction, which is also determined by the prominent local cytokine, protease, and inhibitor environment 32-34. This possible uncoupling of inflammation and destruction could be one of the explanations for the wide inter-individual variation in radiological damage despite comparable inflammation as defined by, for example CRP 35.

We hypothesize that serum MMP-3 is closer related to progression of radiological damage in comparison with CRP. Therefore we analyzed mean serum MMP-3 levels in a 6 months period with and a consecutive 6 months period without progression of radiological damage and evaluated differences between serum MMP-3 and CRP. PATIENTS AND METHODS Thirty-two patients with RA and progression of radiological damage (> 5 points according to the Sharp/van der Heijde method) during a 6 months period followed by a 6 months period with no radiological progression (< 1 point) were selected from a cohort of patients with RA according to the 1987 American

Chapter 6

102

College of Rheumatology criteria 36 with joint symptoms existing less than one year at presentation and who had not previously received DMARDs. These patients participated in a prospective follow-up study at the Department of Rheumatology at the Groningen University Hospital. According to the protocol clinical and laboratory investigations were performed at monthly intervals and radiographs of hands and feet were obtained every 6 months of follow-up.

During follow-up patients were treated with nonsteroidal anti-inflammatory drugs and disease modifying antirheumatic drugs (DMARDs) as indicated clinically. Guidelines for the sequence of the different second-line drugs were as follows: hydroxychloroquine or sulphasalazine as first choice therapy, followed in order by intramuscular gold, D-penicillamine, azathioprine, or methotrexate. Low dose corticosteroids could be administered as adjuvant therapy. Clinical markers of disease activity Fifty-two peripheral joints were examined for tenderness and soft tissue swelling. The following articular indices were determined: Ritchie articular index (RAI) 37, tender joint count (TJC), swollen joint count (SJC) and the disease activity score (DAS) according to Van der Heijde with 3 variables: RAI, number of swollen joints, and erythrocyte sedimentation rate (ESR) 38. The maximal scores are: RAI 78, TJC 52, SJC 52 and DAS 10.0. Laboratory analysis Serum MMP-3 levels were determined by a MMP-3 ELISA developed at our laboratory 39. In short, 96 well plates were precoated with F(ab)2 fragment of goat-antimouse IgG, 1µg/ml (Jackson Immunoresearch Labs, West Grove, PN, USA). Next a mouse monoclonal antibody against human MMP-3, clone 10D6 (R&D Systems, Abingdon, UK) was coated at 0.1 µg/ml. Serum samples were analyzed in two-fold serial dilutions in high performance ELISA buffer (CLB, Amsterdam, NL) and incubated during 1 hour. After washing bound MMP-3 was detected with a polyclonal rabbit-anti-human MMP-3 (AB 810, Chemicon, Temecula, CA, USA), followed by horseradish-peroxidase-labelled F(ab)2 fragment of goat-anti-rabbit IgG (Zymed, San Francisco, CA, USA). Peroxidase activity was determined using tetramethylbenzidin as substrate. MMP-3 levels were calculated at the linear range of the assay from a standard curve (3-400 ng/ml) using pro-MMP-3, purified from serum free supernatant of IL-1β stimulated rheumatoid arthritis synovial fibroblasts 40. The intra-assay coefficient of variation (CV) was 6.8%, the inter-assay CV 8.8%. With an immunoblot we demonstrated that both the monoclonal and the polyclonal antibody reacted with


103

active MMP-3, pro-MMP-3, as well as with MMP-3 bound to tissue inhibitor of matrix metalloproteinases (TIMPs). Furthermore it was demonstrated that rheumatoid factors do not react in this assay and do not interfere with measurement of MMP-3 (data not shown). For normal values of serum MMP-3 the 95 percentile in healthy bloodbank donors (n=80) was used (female < 20 ng/ml, male < 60 ng/ml) 39.

CRP was measured by ELISA 41 (lowest limit of detection: 2 mg/l), ESR according to Westgren. IgM rheumatoid factor (RF) was measured by Dade/Behring BN-2 nephelometer (normal value: < 15 IU/ml). Radiological analysis Radiological damage in hands and feet was assessed by Sharp’s method with some modifications as described by Van der Heijde et al 42,43. By this method joint space narrowing (JSN) and erosions (ER) are scored separately and combined to a total Sharp score (TSS) with a maximum TSS of 448 points. The radiographs were scored without knowledge of clinical and laboratory data in chronological order per patient by two observers. The inter-observer agreement was 0.90 and the intra observer agreements were 0.96 and 0.99 for the two observers respectively. Radiological progression was defined as > 5 Sharp points and no radiological progression as < 1 Sharp point in a 6 months period. Statistical analysis Monthly determined values of clinical and laboratory variables were transformed in a mean value over a 6 months period. Individual changes between the 2 periods were expressed in terms of percentage. For example: ∆ serum MMP-3 =

Individual mean serum MMP-3 in the non-progressive period

Individual mean serum MMP-3 in the progressive period Spearman’s rank correlation coefficients were used for assessment of correlation, for differences between periods the paired T-test. P values of < 0.05 were considered significant.

X 100 %

Chapter 6

104

RESULTS Thirty-two patients with RA and radiological progression (> 5 points according to Sharp’s method) during a 6 months period followed by a 6 months period without radiological progression (< 1 point) were selected from a cohort of early RA patients. The median time between the moment of diagnosis RA and such a period was 10 months with a range of 0-31 months. The characteristics of these 32 patients with RA at the beginning of the period with radiological progression are summarized in table 1.

TABLE 1. Characteristics of the 32 patients with RA at the beginning of the 6 months period with radiological progression. Age (years) 54 (20-78) Gender, female/male (% female) 23/9 (72%) Time until begin progressive period ¶ months 10 (0-31) IgM RF positive (%) 28 (88%) Tender joint count 7 (1-44) Ritchie articular index 6 (1-40) Swollen joint count 6 (1-31) Disease activity score 2.9 (1.3-5.7) CRP (mg/l) 11 (2-117) ESR (mm/h) 25 (4-97) MMP-3 (ng/ml) 54 (5-2370) Sharp score † T0

T6 T12

11 18.5 19

(0-65) (7-71) (8-71)

Values are expressed as the median and range. ¶: Time between the moment of diagnosis of rheumatoid arthritis and the beginning of the period with radiological progression. †: T0 is the Sharp score at the beginning and T6 the Sharp score at the end of the 6-month period with radiological progression. T12 is the Sharp score after the subsequent 6 months without radiological progression.


105

The individual mean values of serum MMP-3, CRP, ESR and disease activity score in the period with and without radiological progression are shown in figure 1A-D.

FIGURE 1. Individual mean values for serum MMP-3 (A), CRP (B), ESR (C), DAS (disease activity score (D) in the 6 months period with vs. the consecutive 6 months period without progression of radiological damage. Each dot represents 1 patient (n=32). All mean values decreased significantly (p < 0.05). DAS < 1.6 = remission, DAS 1.6-2.4 = low disease activity, DAS > 2.4 = moderate to high disease activity.

Chapter 6

106

Individual changes - in terms of percentage – of the mean values between the two periods (figure 2) showed a decrease in both serum MMP-3 and CRP in 19 and an increase in 3 patients, in parallel with other markers of disease activity in the patients (69% of cases).

FIGURE 2. Individual changes - in terms of percentage (∆) - of the mean values of serum MMP-3 in relation to CRP (A), ESR (B), and DAS (disease activity score (C)) between the 2 periods. Each dot represents 1 patient (n=32). Details concerning MMP-3:

Individual mean serum MMP-3 in the non-progressive period Individual mean serum MMP-3 in the progressive period

A decrease in mean serum MMP-3 is represented by a ∆ serum MMP-3 of < 100 %, an increase by a ∆ serum MMP-3 of > 100 %. Spearman correlation showed that ∆ serum MMP-3 correlated significantly with ∆ ESR (r = 0.47, p < 0.01) and ∆ DAS (r = 0.43, p < 0.01). The correlation between ∆ serum MMP-3 and ∆ CRP was borderline significant (r = 0.33, p = 0.06). A separate analysis excluding the data of the “CRP non-responders” showed a correlation coefficient of 0.40 with a p value of 0.03.

∆ serum MMP-3 = X 100 %


107

There was a discrepancy between change in mean serum MMP-3 and CRP in 6 patients (19% of cases). In 5 of these patients serum MMP-3 levels ranged from 22-39 ng/ml and in 2 patients CRP levels ranged from 2-8 mg/l. Small changes in measured serum MMP-3 levels resulted in great changes, in terms of percentage. In the remaining 4 patients (12% of cases) serum MMP-3 decreased but no CRP response could be registered because CRP levels were within the normal range (< 2 mg/l) in both periods. Spearman correlation showed that ∆ serum MMP-3 correlated significantly with ∆ ESR (r = 0.47, p < 0.01) and ∆ DAS (r = 0.43, p < 0.01). The correlation between ∆ serum MMP-3 and ∆ CRP was borderline significant (r = 0.33, p = 0.06). A separate analysis excluding the data of the “CRP non-responders” (n=4) showed a correlation coefficient of 0.40 and p = 0.03.

The median change in serum MMP-3 was 61.8% (range: 4-204%), the median change in CRP 54.7% (range: 13-228%).

The individual change - in terms of percentage - in mean serum MMP-3 or CRP did not correlated with the difference in radiological progression (figure 3) between the 2 periods (∆ serum MMP-3 vs. ∆ Sharp: r = -0.14, p = 0.45 and ∆ CRP vs. ∆ Sharp: r = -0.12, p = 0.50).

FIGURE 3. Individual changes - in terms of percentage (∆) – of the mean values of serum MMP-3 (A) and CRP (B) in relation to the difference in the Sharp score between the 2 periods. Each dot represents 1 patient (n=32). The individual change (mainly decrease, defined as < 100%) in mean MMP-3 or CRP did not correlated with the difference in radiological progression between the two periods (∆ serum MMP-3 vs. ∆ Sharp: r = -0.14, p = 0.45 and ∆ CRP vs. ∆ Sharp: r = -0.12, p = 0.50).

Chapter 6

108

In the period without radiological progression disease activity was moderate-high (DAS > 2.4) in 9 patients, low (DAS 1.6 – 2.4) in 16 patients and 7 patients were in remission (DAS < 1.6), (figure 1D). Patients with an increase in serum MMP-3, CRP or ESR in the period without radiological progression (n=3) had DAS scores of > 1.6. DISCUSSION In RA markers of disease activity and destruction are important, not only for prognostic but also for therapeutical reasons. Acute phase proteins, such as CRP are indirectly produced by the liver after stimulation by cytokines produced at the inflammatory site. Serum MMP-3 is locally produced and activated in the inflamed joint and systemic levels are supposed to be a direct reflection of local synthesis. As such serum MMP-3 could be used as a systemic marker of local inflammation and destruction.

In RA serum MMP-3 is closely correlated to other markers of disease activity like swollen joint count, CRP, and ESR 16,20,22,24,25,27-29,34,44,45. Cross-sectional and long-term follow-up studies also show correlation between serum MMP-3 and radiological detectable damage 16,29. Recent studies suggest that the pathophysiologic mechanisms of joint inflammation, as reflected by CRP and destruction may be partially independent 32-34. Because of the key role of MMP-3 in matrix degradation and the relation between serum MMP-3 levels and radiological damage, we evaluated the serum MMP-3 levels in periods with and without progression of radiological damage in patients with early RA. We hypothesized that serum MMP-3 is closer related to progression of radiological damage in comparison with CRP.

In rheumatoid arthritis a distinction must be made between process variables such as CRP and outcome measures like radiological progression. The course of the disease is generally monitored by serial measurements of one or more process variables. As radiological outcome is essentially the result of what has happened during the course of the disease, theoretically the area under the curve of serially measured process variables meets the requirements of a proper outcome measure 46. For equally spaced observations, as is the case in this study, the area under the curve is essentially the same as the mean of all measurements 47. Such a transformation of data enables the comparison of process variables with outcome measures.

In this select group of RA patients we found, in the first place a close correlation between mean serum MMP-3 and CRP in both periods. Secondly, evaluation of individual changes - in terms of percentage - between the 2 periods


109

showed a discrepancy between change in mean serum MMP-3 and CRP in only a minority of patients, especially in patients with low or close to normal values. In the third place, individual changes - in terms of percentage - in mean serum MMP-3 or CRP did not correlate to the differences in radiological progression between the 2 periods. In this study we did not observe a difference between serum MMP-3 and CRP in relation to the progression of radiological damage. Or in other words, in this select group of RA patients we could not find arguments for an uncoupling of inflammation as reflected by CRP and destruction as reflected by serum MMP-3. An influence of differences in therapy was considered but could, unfortunately not be evaluated due to diversity in used DMARDs in this relative small number of RA patients.

The lack of a difference between serum MMP-3 and CRP in relation to progression of radiological damage could have several reasons. In the first place patient selection could be of influence. In our attempt to look for differences between serum MMP-3 and CRP in relation to progression of radiological damage we have chosen a radiological criterium. Patients with radiological progression (> 5 points according to Sharp’s method) during 6 months followed by a 6 months period with no radiological progression (< 1 point) were selected from a large follow-up cohort of RA patients. The cut-off point of > 5 Sharp points was inspired by the publications concerning minimal clinical important difference and the smallest detectable difference in the Sharp score which scores damage in hands and feet 48 With these criteria, even in the non-progressive period, 9 patients had a high-moderate disease activity score (see figure 1D) probably due to disease activity in other joints. This persistent disease activity makes it difficult to evaluate the uncoupling of disease activity assessed by CRP, joint counts, DAS and radiological damage scored by X-rays of hands and feet. In addition it is of importance to notice that there were some patients with radiological progression despite a relatively low disease activity, according to the DAS score. This emphasizes the inter-individual difference in the relation between parameters of disease activity and radiological damage.

Secondly it is conceivable that a lag time could play a role 35,49. Radiological detectable damage could lag behind disease activity as measured by CRP or serum MMP-3. It remains debatable if all the 32 patients were actually in “radiological remission”.

In the third place sample size could be of importance. With the criteria mentioned we could select 32 patients out of a cohort of 405 RA patients. This number of patients could have been too small because serum MMP-3 and CRP

Chapter 6

110

are closely related, which implies a relative great sample size. Finally, in RA MMP-3 is produced in the inflamed joints as an inactive pro-

enzyme. Upon activation it is able to degrade many components of the matrix. Inactivation occurs by inhibitors such as TIMPs (tissue inhibitors of matrix metalloproteinases). It is conceivable that an imbalance between production, activation and inhibition results in joint destruction (radiological damage). In our hypothesis we assume that serum MMP-3, consisting mainly pro-MMP-3, is a direct reflection of local pro-MMP-3 production as well as of its local activation and subsequent joint destruction. In this respect serum MMP-3 could reflect joint destruction more directly compared to CRP which is produced indirectly by the liver after cytokine stimulation. The fact that we did not find a difference between serum MMP-3 and CRP with regard to radiological progression implies that not only production but rather the balance between activation and inhibition is of major importance. Unfortunately these mechanisms are difficult to evaluate.

The optimal biochemical marker for monitoring destruction has still to be defined 50. Therefore, although serum MMP-3 remains an interesting marker in patients with rheumatoid arthritis, it is not superior to CRP. CRP is still an excellent parameter not only for monitoring disease activity but also for monitoring progression of radiological detectable damage.

In conclusion, serum MMP-3 and CRP are closely related and there seems to be no difference between serum MMP-3 and CRP with regard to the monitoring of the progression of radiological damage. ACKNOWLEDGEMENTS This study was supported by a grant from “The Dutch Arthritis Association”, The Netherlands.


111

REFERENCES 1. Choy EH, Panayi GS.Cytokine pathways and joint inflammation in rheumatoid arthritis. N Engl J

Med 2001;344:907-16. 2. Nagase H, Okada Y. Proteinases and matrix degradation. In Kelley WN, Harris ED, Ruddy S,

Sledge CB, (Eds.). Textbook of rheumatology. Philadelphia, Saunders,W.B., 1997:323-41. 3. Okada Y. Proteinases and matrix degradation. In Ruddy S, Harris-ED J, Sledge CB, Budd RC,

Sergent JS, (Eds.). Kelley's Textbook of rheumatology. Philadelphia, Sanders,W.B., 2001:55-72. 4. Hasty KA, Reife RA, Kang AH, Stuart JM. The role of stromelysin in the cartilage destruction that

accompanies inflammatory arthritis. Arthritis Rheum 1990;33:388-97. 5. Mudgett JS, Hutchinson NI, Chartrain NA et al. Susceptibility of stromelysin 1-deficient mice to

collagen-induced arthritis and cartilage destruction. Arthritis Rheum 1998;41:110-21. 6. Singer II, Kawka DW, Bayne EK et al. VDIPEN, a metalloproteinase-generated neoepitope, is

induced and immunolocalized in articular cartilage during inflammatory arthritis. J Clin Invest 1995;95:2178-86.


8. Nagase H. Activation mechanisms of matrix metalloproteinases. Biol Chem 1997;378:151-60. 9. McCachren SS. Expression of metalloproteinases and metalloproteinase inhibitor in human

arthritic synovium. Arthritis Rheum 1991;34:1085-93. 10. Okada Y, Takeuchi N, Tomita K, Nakanishi I, Nagase H. Immunolocalization of matrix metallo-

proteinase 3 (stromelysin) in rheumatoid synovioblasts (B cells): correlation with rheumatoid arthritis. Ann Rheum Dis 1989;48:645-53.





15. Ishiguro N, Ito T, Oguchi T et al. Relationships of matrix metalloproteinases and their inhibitors to cartilage proteoglycan and collagen turnover and inflammation as revealed by analyses of synovial fluids from patients with rheumatoid arthritis. Arthritis Rheum 2001;44:2503-11.






Chapter 6

112

21. Yoshihara Y, Nakamura H, Obata K et al. Matrix metalloproteinases and tissue inhibitors of metalloproteinases in synovial fluids from patients with rheumatoid arthritis or osteoarthritis. Ann Rheum Dis 2000;59:455-61.


23. Zucker S, Lysik RM, Zarrabi MH et al. Elevated plasma stromelysin levels in arthritis. J Rheumatol 1994;21:2329-33.

24. Taylor DJ, Cheung NT, Dawes PT: Increased serum proMMP-3 in inflammatory arthritides: a potential indicator of synovial inflammatory monokine activity. Ann Rheum Dis 1994;53:768-72.

25. Keyszer G, Lambiri I, Nagel R et al. Circulating levels of matrix metalloproteinases MMP-3 and MMP-1, tissue inhibitor metalloproteinases 1 (TIMP-1), and MMP-1/TIMP-1 complex in rheumatic disease. Correlation with clinical activity of rheumatoid arthritis versus other surrogate markers. J Rheumatol 1999;26:251-8.



28. Ribbens C, Andre B, Jaspar JM et al. Matrix metalloproteinase-3 serum levels are correlated with disease activity and predict clinical response in rheumatoid arthritis. J Rheumatol 2000;27:888-93.


30. Yamanaka H, Matsuda Y, Tanaka M et al.. Serum matrix metalloproteinase 3 as a predictor of the degree of joint destruction during the six months after measurement, in patients with early rheumatoid arthritis. Arthritis Rheum 2000;43:852-8.


32. Mulherin D, Fitzgerald O, Bresnihan B. Clinical improvement and radiological deterioration in rheumatoid arthritis: evidence that the pathogenesis of synovial inflammation and articular erosion may differ. Br J Rheumatol 1996;35:1263-8.

33. Kirwan JR. The relationship between synovitis and erosions in rheumatoid arthritis. Br J Rheumatol 1997; 36:225-8.

34. Cunnane G, Fitzgerald O, Beeton C, Cawston TE, Bresnihan B. Early joint erosions and serum levels of matrix metalloproteinase 1, matrix metalloproteinase 3, and tissue inhibitor of metalloproteinases 1 in rheumatoid arthritis. Arthritis Rheum 2001;44:2263-74.

35. van Leeuwen MA, van Rijswijk MH, van der Heijde DM et al. The acute-phase response in relation to radiographic progression in early rheumatoid arthritis: a prospective study during the first three years of the disease. Br J Rheumatol 1993;32 Suppl 3:9-13.

36. Arnett FC, Edworthy SM, Bloch DA et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 1988;31:315-24.

37. Ritchie DM, Boyle JA, McInnes JM et al. Clinical studies with an articular index for the assessment of joint tenderness in patients with rheumatoid arthritis. Q J Med 1968;37:393-406.

38. van der Heijde DM, van 't Hof M, van Riel PL et al. Judging disease activity in clinical practice in rheumatoid arthritis: first step in the development of a disease activity score. Ann Rheum Dis 1990;49:916-20.


113

39. Posthumus MD, Limburg PC, Westra J, van Leeuwen MA, van Rijswijk MH. Serum matrix metalloproteinase 3 levels during treatment with sulfasalazine or combination of methotrexate and sulfasalazine in patients with early rheumatoid arthritis. J Rheumatol 2002;29:883-9.







46. Matthews JN, Altman DG, Campbell MJ, Royston P. Analysis of serial measurements in medical research. BMJ 1990;300:230-5.

47. Altman DG. Some common problems in medical research. In Altman DG, (Ed.). Practical statistics for medical research. Chapman & Hall/ CRC press, 1999:396-439.

48. Bruynesteyn K, van der HD, Boers M et al. Minimal clinically important difference in radiological progression of joint damage over 1 year in rheumatoid arthritis: preliminary results of a validation study with clinical experts. J Rheumatol 2001;28:904-10.

49. Matsuda Y, Yamanaka H, Higami K, Kashiwazaki S. Time lag between active joint inflammation and radiological progression in patients with early rheumatoid arthritis. J Rheumatol 1998;25:427-32.

50. Garnero P, Rousseau JC, Delmas PD. Molecular basis and clinical use of biochemical markers of bone, cartilage, and synovium in joint diseases. Arthritis Rheum 2000;43:953-68.

Chapter 7

Matrix metalloproteinase 1 and 3 promoter polymorphism in patients with

early rheumatoid arthritis

Marcel D. Posthumus, Miek A. van Leeuwen, Johan Bijzet, Bouke G. Hepkema, Ilja M. Nolte, Martin H. van Rijswijk, and

Pieter C. Limburg.

Submitted.

Chapter 7

116

SUMMARY Objective. To evaluate the significance of matrix metalloproteinase-1 (MMP-1)

and MMP-3 promoter polymorphism in relation to disease activity and radiological damage in patients with early rheumatoid arthritis (RA). Methods. MMP-1 and MMP-3 promoter polymorphisms were determined in a cohort of 448 early RA patients. Clinical and laboratory markers of disease activity and severity at presentation were related to these polymorphisms. In addition total radiological damage and radiological progression (Sharp’s method, modified by van der Heijde) after 2 years were evaluated. In a subgroup initial serum MMP-3 (sMMP-3), cumulative sMMP-3 (serum MMP-3 area under the curve; sMMP-3AUC) over 0.5 and 1 year and sMMP-3AUC relative to CRPAUC in relation to the MMP-3 promoter polymorphism were studied.

Results. An association between MMP-1 and MMP-3 promoter polymorphism and disease activity, severity or radiological damage at presentation was not found. MMP-1 and MMP-3 promoter polymorphisms showed strong linkage disequilibrium, but associations with “risk factors” such as RF, anti-CCP and HLA-DR4 status were not found. The MMP-1 and MMP-3 promoter polymorphisms were not associated with radiological damage or radiological progression after a follow-up of 2 years. Serum MMP-3AUC correlated significantly with radiological progression after 6 (r = 0.26, p < 0.01) and 12 months (r = 0.36, p < 0.01) The sMMP-3AUC after 6 and 12 months did not differ across the three MMP-3 promoter polymorphism groups. sMMP-3AUC/CRPAUC ratios differed after 12 months across the three promoter polymorphism groups with an increase from the 5A/5A to the 6A/6A carrier genotype (p < 0.01). Conclusion. In our study in early RA patients we could not find an association between MMP-1 and MMP-3 promoter polymorphism and disease activity or severity at presentation. In addition the MMP-1 and MMP-3 promoter polymorphisms were not associated with radiological damage or radiological progression after a follow-up of 2 years. Key words: matrix metalloproteinase 1, matrix metalloproteinase 3, promoter polymorphism, early rheumatoid arthritis, disease activity, radiological progression.

MMP-1 and MMP-3 promoter polymorphism in early RA

117

INTRODUCTION Rheumatoid arthritis (RA) is a systemic inflammatory disorder characterized by chronic inflammation of synovial tissue and in most cases progressive destruction of cartilage and bone 1. It affects about 0.5 to 1% of the population and has substantial impact in terms of disability, morbidity, mortality, and costs 2,3. Early identification of patients with aggressive destructive disease is important not only for prognostic but also for therapeutic reasons 4. Novel, more aggressive therapies are currently developed 5 and the need for clinically feasible means for assessing the prognosis in the individual patient is becoming increasingly important 6. Disease activity, rheumatoid factor positivity, and early radiological abnormalities are considered to be important in relation to outcome such as radiological damage. Genetic studies show associations between radiological damage and HLA-DRB1 shared epitope 7 and recently with matrix metalloproteinase (MMP) promoter polymorphisms 8,9.

MMPs are proteolytic enzymes involved in many physiological processes and pathological conditions. These enzymes are capable of degrading many components of the extra-cellular matrix and/or activate other MMPs 10-12. It is conceivable that an imbalance between production, activation, and inhibition of MMPs results in matrix degradation.

In RA, the MMPs are locally produced and activated within the affected joint as a result of cytokine mediated stimulation of synovial cells. In particular matrix metalloproteinase 3 (MMP-3) is of interest because this proteolytic enzyme is produced abundantly in the inflamed joints and plays a prominent role in the pathogenesis of matrix degradation in RA 10,11,13, even though it is not the only key-enzyme 14. MMP-3 is capable of degrading many components of the matrix in the synovial joint including proteoglycans, gelatins, laminin, fibronectin, and collagen III, IV, IX 10,13,15,16. Moreover, MMP-3 is able to activate other matrix metalloproteinases such as MMP-1, MMP-7, MMP-8, MMP-9 and MMP-13 12.

The enzyme has been localized in the fibroblast-like synoviocytes of rheumatoid synovium 17-19, in RA cartilage 20,21, at sites of cartilage erosion 13, in synovial fluid 22-29, and in serum 28,30-34. Systemic MMP-3 levels are supposed to reflect local synthesis induced by pro-inflammatory cytokines. As such, serum MMP-3 can be used as a systemic marker of local joint inflammation 24,28,32,33,35,36 and/or destruction 34,37-39.

Several regulatory mechanisms influence the ultimate effect of a MMP on extra-cellular matrix, such as regulation of transcription, activation of latent proMMPs, and inhibition of MMP activity by inhibitors such as TIMPs (tissue inhibitors of matrix metalloproteinases) 40,41. At the level of transcription,

Chapter 7

118

functional promoter polymorphisms in the genes of important MMPs might have implications for susceptibility and/or severity of diseases 42.

With respect to RA, functional MMP promoter polymorphisms have been reported, which seem to be of importance in the regulation of MMP gene expression 42. For example in the promoter region of the MMP-1 gene, at position -1607, two alleles have been detected; one is having a single guanine (1G) and the other having two guanines (2G). In in vitro experiments the 2G allele had substantially more transcriptional activity compared to the 1G allele 43,44.

In the promoter of MMP-3 a polymorphism at position -1171 appears to be functionally important, at least in in vitro experiments. Cultured cells transfected with constructs containing a run of 6 adenines (6A) expressed a roughly 2-fold lower amount of reporter gene product as compared with the transfectants of the constructs containing a run of 5 adenines (5A) 45.

In previous studies in RA patients an association between the 1G-1607/2G promoter polymorphism in the MMP-1 gene and radiological damage could not be confirmed 46. However in another study by the same group of investigators an association was found between the 5A-1171/6A promoter polymorphism in the MMP-3 gene and radiological damage and progression 8. In particular patients homozygous for the 6A allele had the highest radiological progression. This seems to contrast with in vitro experiments in which this 6A/6A polymorphism expressed a lower amount of gene product 42-45.

In the present prospective follow-up study of early RA patients we evaluated the significance of MMP-1 and MMP-3 promoter polymorphisms in relation to disease activity and radiological damage. In a subgroup of patients the association between the MMP-3 promoter polymorphism and serum levels of MMP-3 (sMMP-3) was evaluated. PATIENTS AND METHODS Patients Patients with early RA who participated in a prospective follow-up study between January 1985 and December 2000 at the Department of Rheumatology of the Groningen University Hospital were included in this study. All patients met the following criteria: a. the 1987 American College of Rheumatology criteria 47 b. joint symptoms existing less than one year at presentation c. no previous disease modifying anti rheumatic drugs (DMARDs) and d. two years of clinical and radiological follow-up.

Clinical and laboratory investigations were performed at monthly intervals and


119

radiographs of hands and feet were obtained every 6 months during follow-up. Treatment During follow-up, patients were treated with non-steroidal anti-inflammatory drugs as indicated clinically. Low dose corticosteroids could be administered as adjuvant therapy. From January 1985 until January 1991 the guidelines for the sequence of the different DMARDs were as follows: Hydroxychloroquine or Sulphasalazine (SASP) as first choice therapy, followed in order by intramuscular Gold, D-penicillamine, Azathioprine, or Methotrexate (MTX). These patients are denoted as traditional treatment cohort. From January 1991 until December 1993, patients participated in a study in which this traditional regime was compared with an early, intensive drug treatment 48. Since January 1994 it is our policy to use this intensive treatment strategy in early RA patients consisting of an immediate start of DMARDs and rapid adjustment of dosage and/or drugs (step up method) in case of an insufficient response. In this intensive treatment cohort all patients started with SASP 2000-3000 mg/day. In case of an insufficient response MTX, in increasing dosages up to 25 mg/week, could be added. If there was still an insufficient response, SASP was replaced by either cyclosporine or more recently by a TNF-α blocking agent.

Decisions about intensifying the treatment with DMARDs were discussed at monthly visits by an independent observer and the patient’s rheumatologist. These decisions were based on clinical markers of disease activity in combination with the CRP level as effect measures. A sufficient response was defined as a > 50% reduction in joint scores or CRP level (or normalization of CRP (< 2 mg/l)). Clinical markers of disease activity Fifty-two peripheral joints were examined for tenderness and soft tissue swelling. The following articular indices were determined: Ritchie articular index (RAI) 49, tender joint count (TJC), swollen joint count (SJC), and the disease activity score (DAS) according to Van der Heijde with 3 variables: RAI, number of swollen joints, and erythrocyte sedimentation rate (ESR) 50. The maximal scores are: RAI 78, TJC 52, SJC 52 and DAS 10.0. HLA-DR4 typing HLA typing was performed by serologically using two coloured fluorescence technique 51, and since 1994 by DNA using sequence specific primers 52. In case only a single HLA-DR4 allele was found, the alleles were assumed to be present

Chapter 7

120

homozygously. MMP-1 and MMP-3 polymorphism genotyping Genomic DNA from each individual was isolated from a whole blood sample by using the Puregene® DNA isolation kit (Gentra Systems, Minneapolis, USA). DNA from smaller amounts of frozen lymphocyte suspensions was isolated using the QIAmp DNA Blood mini kit (QIAGEN, Hilden, Germany).

The final DNA samples were measured by using a Bio Photometer (Eppendorf, Hamburg, Germany) calculating the concentration and purity of the samples. For MMP polymorphism analyses the Taqman assay (Applied Biosystems, Foster City, USA) was used. Briefly this assay enables the discrimination of two alleles based on the hybridization of labeled specific probes and the 5’-3’ exonuclease activity of Taq DNA polymerase. The exonuclease will digest hybridized probes, separating the reporter and quencher, leading to one or two specific fluorescence signals 53. Primer and labeled probes for MMP-1 polymorphism (at position -1607) and MMP-3 polymorphism (at position -1171) were developed by Applied Biosystems (Assay by Design). The sequences for MMP-1 primers were 5’-GTT ATG CCA CTT AGA TGA GGA AAT TG (forward) and 5’-CAT AAA CAA TAC TTC AGT ATA TCT TGG ATT GA (reverse), and for MMP-3 the primers were 5’-TGG TTC TCC ATT CCT TTG ATG GG (forward) and 5’-TCA ATG TGG CCA AAT ATT TTC CCT GTA (reverse). The dye-labeled probes for MMP-1 were 6-FAM-AGT CAT ATC TTT CTA ATT AT-NFQ-MGB (1G) and VIC-AGA TAA GTC ATA TCC TTT CT-NFQ-MGB (2G) in which NFQ is the Non-Fluorescent Quencher, FAM and VIC the reporter dye, and MGB the Minor Groove Binder for increase of annealing temperature to enable use of shorter probes. The FAM probe detects the 1G and the VIC probe the 2G polymorphism of MMP-1. For MMP-3 the 5A allele was detected by the 6-FAM-AAG ACA TGG TTT TTC-NFQ-MGB probe and the 6A allele by the VIC-AAG-ACA-TGG-TTT-TTT-C-NFQ-MGB probe.

Genomic DNA (10 ng) was added to a 384-well microtitre plate and dried. The polymerase chain reaction (PCR) and measuring of the fluorescence was performed under standard conditions after adding 5 µl reaction mix (containing Taqman Universal Master Mix, primers and probes) by using the ABI Prism 7900HT Sequence Detection System (Applied Biosystems, Foster City, USA). The intensities of the allele specific fluorophores were plotted against each other and 3 groups were identified as shown in figure 1A & B.


121

FIGURE 1. Representation of the results of the genotyping for (A) MMP-1 and (B) MMP-3. In both plots three groups were identified. H20 controls are water controls. Serum MMP-3 Serum MMP-3 (sMMP-3) levels were determined by a MMP-3 ELISA developed at our laboratory 54. In short, 96 well plates were precoated with F(ab)2 fragment of goat-antimouse IgG, 1µg/ml (Jackson Immunoresearch Labs, West Grove, PN, USA). Next a mouse monoclonal antibody against human MMP-3, clone 10D6 (R&D Systems, Abingdon, UK) was coated at 0.1 µg/ml. Serum samples were analyzed in two-fold serial dilutions in high performance ELISA buffer (CLB, Amsterdam, NL) and incubated during 1 hour. After washing bound MMP-3 was detected with a polyclonal rabbit-anti-human MMP-3 (AB 810, Chemicon, Temecula, CA, USA), followed by horseradish-peroxidase-labelled F(ab)2 fragment of goat-anti-rabbit IgG (Zymed, San Francisco, CA, USA). Peroxidase activity was determined using tetramethylbenzidin as substrate. MMP-3 levels were calculated at the linear range of the assay from a standard curve (3-400 ng/ml) using pro-MMP-3, purified from serum free supernatant of IL-1β stimulated rheumatoid arthritis synovial fibroblasts 55. The intra-assay coefficient of variation (CV) was 6.8%, the inter-assay CV 8.8%. With an immunoblot we demonstrated that both the monoclonal and the polyclonal antibody reacted with active MMP-3, pro-MMP-3 as well as with MMP-3 bound to tissue inhibitor of matrix metalloproteinases (TIMPs) (data not shown). Furthermore it was demonstrated that rheumatoid factors do not react in this assay and do not interfere with measurement of MMP-3 (data not shown).

Chapter 7

122

For normal values of sMMP-3 the 95 percentile in healthy bloodbank donors (n=80) was used (female < 20 ng/ml, male < 60 ng/ml) 54.

Anti-CCP Serum anti cyclic citrullinated peptides (anti-CCP) levels were determined using an ELISA kit as described by the manufacturer (Euro-Diagnostica, Arnhem, NL). Serum or plasma was added to microtitre plate wells coated with citrullinated synthetic peptides. The bound antibodies are detected by adding a HRP labelled antibody to human IgG and subsequently visualized with a color reaction. The measured absorbency of the unknown samples is calculated against a standard curve, ranging from 25-1600 Units/ml., using the software package Softmax Pro (Molecular Devices, Sunnyvale, USA). Normal or negative values are defined as < 25 Units/ ml. CRP, ESR and Rheumatoid factor C-reactive protein (CRP) was measured by ELISA 56 (normal value: < 2 mg/l), ESR according to Westgren. IgM rheumatoid factor (RF) was measured by Dade/Behring BN-2 nephelometer (normal value: < 15 IU/ml). Radiological analysis Radiological damage in hands and feet was assessed by Sharp’s method with some modifications as described by Van der Heijde et al 57,58. By this method joint space narrowing (JSN) and erosions (ER) are scored separately and combined to a total Sharp score (TSS) with a maximum TSS of 448 points. The radiographs were scored without knowledge of clinical and laboratory data in chronological order per patient by two observers. In a random sample of 20 patients the inter-observer agreement was 0.90 and the intra-observer agreements were 0.96 and 0.99 for the two observers respectively. Statistical analysis To investigate the association between MMP promoter polymorphisms and quantitative markers of disease activity, such as the DAS and CRP, and radiological damage (TSS) at presentation the Kruskal-Wallis test was used.

For evaluation of an association between MMP promoter polymorphisms and other “riskfactors” at presentation such as HLA-DR4, RF and anti-CCP the Chi-square test was applied. Linkage disequilibrium between MMP-1 and MMP-3 promoter polymorphism was evaluated using the software package Arlequin.

To determine the additional value of MMP promoter polymorphisms to the


123

prediction of progression of radiological damage (∆ TSS) an univariate generalized linear model was used with incorporation of all other relevant prognostic factors. To investigate the significance of MMP-3 promoter polymorphism with respect to sMMP-3 levels monthly sMMP-3 (and CRP) levels were plotted against time and time-integrated values were obtained by calculation of the area under the curve. Cumulative values were determined over 6 months periods (sMMP-3AUC and CRPAUC). The association between MMP-3 promoter polymorphism and sMMP-3AUC and sMMP-3AUC/CRPAUC ratios during respectively the first 6 and 12 months of follow-up was evaluated by using the Kruskal-Wallis test.

To evaluate the relation between sMMP-3AUC over 6 months periods and radiological progression during the first year of follow-up the Spearman’s rank correlation coefficient was determined. Statistical analysis was carried out using the SPSS statistical package, 10.0. RESULTS Table 1 shows the characteristics of the 448 early RA patients at presentation.

TABLE 1. Characteristics of the study population at presentation (n = 448). Age yrs. 53.3 16.3-82.5 Gender % female 67.6 % IgM-RF % + 82.6 % Anti-CCP (n=446) % + 75.4 % HLA-DR4 (n=432) % + 56.5 %

1G/ 1G 28.9 % 1G/ 2G 51.0 %

MMP-1 polymorphism (n=394)

2G/ 2G 20.1 % 5A/ 5A 25.0 % 5A/ 6A 52.8 %

MMP-3 polymorphism (n=396)

6A/ 6A 22.2 % SJC 11 0-36 TJC 12 0-48 RAI 9 0-54 DAS 6.4 3.2 -10.0 Serum MMP-3 (n=99) ng/ml 77 < 5-1807 CRP mg/l 22 < 2-398 ESR mm/hr 36 2-130 Total Sharp score 1 0-97 Total Sharp score % 0 42 % Demographic and clinical data of the patients at study entry. The absolute values are the median and range. SJC; swollen joint count. TJC; tender joint count. RAI; Ritchie articular index. DAS; disease activity score with 3 variables. % +; percentage positive. % 0; percentage with Sharp score of 0.

Chapter 7

124

Suitable materials to determine MMP-1 and MMP-3 promoter polymorphisms were available in 394 and 396 patients respectively. The described assay to determine MMP-1 and MMP-3 promoter polymorphism identified 3 groups (figure 1) with frequencies (see table 1) comparable to the literature 8,46.

In table 2 the MMP-1 and MMP-3 promoter polymorphisms are described in detail. Haplotypes were not determined and thereby unknown in 120 patients. However by using the Arlequin software package strong linkage disequilibrium between the two promoter polymorphisms of MMP-1 and MMP-3 was found (p < 0.001). The 1G and 2G alleles were more frequently (than expected) linked to the 5A and 6A alleles respectively.

There were no associations between MMP-1 and/or MMP-3 promoter polymorphism and RF, anti-CCP, and HLA-DR4 status at presentation (Chi square). TABLE 2. MMP-1 and MMP-3 promoter polymorphisms in the study population (n=448). MMP-1

1G/1G 1G/2G 2G/2G nd # Total 5A/5A 60 27 9 3 99 5A/6A 47 120 ¶ 34 8 209 6A/6A 6 43 35 4 88

nd # 1 11 1 39 52 Total 114 201 79 54 448 # nd; not done = inadequate material to determine promoter polymorphisms. ¶ Patients with an unknown haplotype; 1G-5A/2G-6A or 2G-5A/1G-6A.

MMP-1 and MMP-3 promoter polymorphism in relation to activity and severity of RA at presentation The DAS at presentation did not differ among the three MMP-1 and the three MMP-3 polymorphism groups. Analysis of the DAS at presentation in subgroups stratified by RF, anti-CCP status, HLA-DR4 positivity, or a total Sharp score of < or > 10 points at presentation did not show any significant differences either.

The same analysis was performed with the CRP level, total Sharp score and serum MMP-3 level at presentation. The distribution of all these variables did not significantly differ across the different MMP-1 or MMP-3 promoter polymorphism groups, neither in the whole group nor in subgroups.

A subanalysis of the two items of the total Sharp score, erosions and joint space narrowing did not reveal significant differences across the MMP-1 or MMP-3 genotypes (table 3).


125

TABLE 3. MMP1 and MMP-3 promoter polymorphisms and radiological severity at presentation. MMP-1 MMP-3 1G/1G 1G/2G 2G/2G 5A/5A 5A/6A 6A/6A Baseline (n=114) (n=201) (n=79) (n = 99) (n = 209) (n = 88) TSS 1 (0-3) 1 (0-4) 1 (0-6) 1 (0-5) 1 (0-4) 1 (0-5) ER 0 (0-2) 1 (0-3) 1 (0-3.5) 0 (0-3) 0 (0-2) 0 (0-2) JSN 0 (0-1) 0 (0-1) 0 (0-1) 0 (0-1) 0 (0-1) 0 (0-2) Median and interquartile range of the Total Sharp Score (TSS), Erosions (ER) and Joint Space Narrowing (JSN) at presentation in the MMP-1 and MMP-3 promoter polymorphism groups. These variables were not different across the MMP-1 and MMP-3 polymorphism groups. (Kruskal-Wallis test).

MMP-1 and MMP-3 promoter polymorphism in relation to radiological progression In an univariate generalized linear model no association was found between the MMP-1 or MMP-3 promoter polymorphisms and radiological progression (∆ TSS) or the TSS at 24 months. In contrast, variables such as treatment strategy, anti-CCP status, ESR, CRP and TSS at presentation were significantly associated with ∆TSS and the absolute TSS at 24 months (data not shown).

Because of clear differences in TSS and ∆TSS (table 4) a subgroup analysis was done among the different treatment strategy groups. However, this did not reveal significant associations between MMP-1 or MMP-3 promoter polymorphisms and ∆TSS or the absolute TSS at 24 months. TABLE 4. Total Sharp score and radiological progression in different treatment groups. TSS ∆ TSS 0 12 24 0-24 Traditional treatment (n=132) 1 0-4.5 11 2-27.5 21.5 4-51 15 3-38 Intermediate group (n=143) 2 0-6 15 6.5-27 23 9.5-42 20 8-33.5 Intensive treatment (n=173) 1 0-3 5 0.5-11 8 1-17 ¶ 6 1-15 ¶ All patients (n=448) 1 0-4 9 2-21 15 4-31 12 2-26 Total sharp score (TSS) at presentation (0 months), after 12 and 24 months of follow-up and radiological progression after 24 months (∆ TSS) in the 3 different treatment groups. ¶ P < 0.001 compared to traditional and intermediate treatment group (Kruskal-Wallis test).

MMP-3 promoter polymorphism in relation to sMMP-3AUC Monthly sMMP-3 levels in combination with MMP-3 promoter polymorphism were available in 99 patients during the first 6 months and in 66 patients during 12 months. To investigate the significance of an MMP-3 promoter polymorphism with respect to serum levels monthly sMMP-3 levels were plotted against time and time-integrated values were obtained by calculation of the area under the curve (sMMP-3AUC). Cumulative values were determined over 6 months periods.

This sMMP-3AUC after 6 and 12 months correlated significantly with ∆TSS

Chapter 7

126

after 6 months (Spearman's r = 0.26, p < 0.01) and after 12 months (r = 0.36, p < 0.01).

The sMMP-3AUC after 6 and 12 months did not differ across the MMP-3 promoter polymorphism groups (Kruskal-Wallis test). There was a trend to a higher sMMP-3AUC in the 6A/6A group but this did not reach statistical significance (figure 2).

FIGURE 2. Serum MMP-3 and CRP area under the curve (AUC) after 6 and 12 months of follow-up across the MMP-3 promoter polymorphisms. Statistical significant differences were not found (Kruskal-Wallis) although the sMMP-3AUC had a tendency to higher levels in the 6A/6A group.


127

MMP-3 promoter polymorphism in relation to sMMP-3AUC/CRPAUC ratios Serum MMP-3 and CRP are closely correlated in patients with RA 28,33,37,59. To evaluate an additive effect of MMP-3 promoter polymorphism “correction for inflammation” was made by evaluation of sMMP-3AUC/CRPAUC ratios. After 6 months of follow-up the ratios were not different across the MMP-3 promoter polymorphism groups. However after 12 months a significant difference in ratios was found (p < 0.01) between the 5A/5A & 5A/6A and 5A/5A & 6A/6A groups (Kruskal-Wallis with Dunn’s multiple comparison test) (figure 3).

FIGURE 3. Serum MMP-3 area under the curve/CRP area under the curve (AUC) ratio’s across the MMP-3 promoter polymorphism groups after 6 and 12 months of follow-up. After 6 months of follow-up the ratios were not different across the MMP-3 promoter polymorphism groups. After 12 months a significant difference in ratio’s was found (p < 0.01) determined by a difference between the 5A/5A & 5A/6A (p < 0.01) and 5A/5A & 6A/6A groups (p < 0.01) (Kruskal-Wallis with Dunn’s multiple comparison test). DISCUSSION In the present study concerning 448 early RA patients with a follow-up of 2 years we could not find any association between MMP-1 and MMP-3 promoter polymorphism and disease activity or severity at presentation. MMP-1 and MMP-3 promoter polymorphisms showed strong linkage disequilibrium, but associations with “risk factors” such as RF, anti-CCP, and HLA-DR4 status were not found. In addition the MMP-1 and MMP-3 promoter polymorphisms were not associated with radiological damage or radiological progression after a follow-up of 2 years. Because aggressive treatment seems to influence the

Chapter 7

128

association between genetic markers and radiological damage 60 a subgroup analysis was done in the 3 different treatment groups with clearly different radiological progression (table 4). This analysis did not reveal different results. In accordance with previous studies we confirmed that the cumulative serum MMP-3 designated by the serum MMP-3 area under the curve (sMMP-3AUC) is correlated with radiological progression 37. The sMMP-3AUC after 6 and 12 months of follow-up was however not different across the three MMP-3 promoter polymorphism groups. sMMP-3AUC/CRPAUC ratios differed after 12 months across the three promoter polymorphism groups with an increase from the 5A/5A to the 6A/6A carrier genotype possibly due to the higher (although not statistically significant) sMMP-3AUC in this group. In RA the MMPs are important mediators in the process of joint destruction.61 In particular MMP-3 is of interest because of its key role in the destructive process 10,11,13 and its associations with disease activity 32,33,36,37 and joint damage 34,38,39. Most MMPs are inducible and their production can be regulated at different levels such as signal transduction preceding transcription, transcription itself, or posttranscriptional processing. At the level of transcription functional promoter polymorphisms in the genes of important MMPs might have implications for disease susceptibility and/or severity of various diseases 42.

From in vitro experiments it appears that a polymorphism in the promoter region of the MMP-1 may be of functional importance. In the MMP-1 gene (locus 11q22-q23) two alleles have been detected; one having a single guanine (1G) and the other having two guanines (2G) at position -1607 (1G-1607/2G). The 2 guanines together with an adjacent adenine can create a core binding site (5’-GGA-3’) for the Ets family of transcription factors. This 2G allele binds substantially more Ets-1 and has substantially more transcriptional activity compared to the 1G allele in normal fibroblasts and melanoma cells4 3 as well as in ovarian tumor tissue 44. Influences of this promoter polymorphism have been evaluated in cancer 42,62, pulmonary disease 63, and scleroderma 64. In RA patients Constantin et al could not find an association between this polymorphism in the MMP-1 promoter and the susceptibility to or severity of RA 46.

With respect to our data in early RA patients the results concerning the MMP-1 promoter polymorphism are consistent with previous reported studies in pulmonary disease 63, scleroderma 64, and RA 46. These studies and our results do not support the hypothesis of an association between this MMP-1 promoter polymorphism and disease activity and severity of RA. Another, in vitro functional, polymorphism has been described in the promoter region of the MMP-3 gene (locus 11q23), at position -1171 (5A-1171/6A).


129

Cultured fibroblasts and vascular smooth muscle cells transfected with constructs containing the run of 6 adenines (6A) expressed a roughly 2-fold lower amount of reporter gene product as compared with the transfectants of the constructs containing a run of 5 adenines (5A) 45. Additional studies revealed binding of putative transcription factors to this region. The difference in promoter activity is probably due to preferential binding of a transcription repressor to the 6A allele 45,65. Influences of this polymorphism have been evaluated in atherosclerosis66-69 and cancer 70,71. In RA patients, Constantin et al found no association between the 5A-1171/6A polymorphism and disease susceptibility. However severity and progression of RA, represented by radiologically detectable joint damage, appeared to be associated with the 6A/6A genotype 8, which in in vitro experiments appears to be associated with a lower transcriptional activity, compared to the 5A carrier genotype.

Our data concerning the MMP-3 promoter polymorphism differ from the study by Constantin et al 8. In contrast to our results they found a significant association between MMP-3 promoter polymorphism and radiological damage after 4 years of follow-up of 96 RA patients. This discrepancy could be due to a number of reasons.

Firstly, there is a difference in duration of follow-up between the two studies, with a follow-up of 2 years in our study and a follow-up of 4 years in the study of Constantin. Theoretically a relative small effect on radiological damage due to e.g. an MMP-3 promoter polymorphism can eventually become statistically significant if the follow-up is long enough. On the other hand such a statistical significance after a long follow-up would imply a decreased clinical relevance.

Secondly, differences in disease activity could be of importance. For example the mean disease activity score in the study of Constantin et al at presentation was 2.8 (SD ± 1.1) and in our study 6.4 (SD ± 1.4). Many of their patients (41%) already used DMARDs in contrast with our patients who were DMARD-naive. High disease activity with active cytokine networks and signal transduction pathways causing extensive mRNA expression of multiple MMP genes could theoretically overrule a relatively small difference in production due to promoter polymorphisms. This last mechanism might also explain why we could not find a significant difference in sMMP-3AUC across the different MMP-3 promoter polymorphism groups.

Thirdly, it is striking that, in the study of Constantin et al, there seems to be an association between the in vitro “low producer” 6A/6A promoter polymorphism and radiological damage. In our study the sMMP-3AUC was not statistically different across the polymorphism groups although the levels tended to increase


131

ACKNOWLEDGEMENTS This study was financially supported by the "Dutch Arthritis Association", The Netherlands.

Chapter 7

130

towards the 6A/6A genotype (figure 2). The sMMP-3AUC/CRPAUC ratios, used as a “correction for inflammation” were significantly higher in the 5A/6A and 6A/6A groups after 12 months. So in vivo, in early RA the 5A/6A and 6A/6A genotypes appear to produce more MMP-3 relative to CRP. A straightforward explanation for these conflicting data is not available but apparently important differences exist between in vitro experiments and in vivo disease with regard to regulatory mechanisms of transcription as well as pre- and posttranscriptional processes.

Finally, linkage with other MMPs could be of influence. At chromosome 11q21-23 at least 8 known human MMP genes (MMP-1, -3, -7, -8, -10, -12, -13 and -20) are clustered and, for example the MMP-1 locus is adjacent to the MMP-3 locus. MMP-1 and MMP-3 promoter linkage disequilibrium has been reported. In colorectal cancer the 2G MMP-1/6A MMP-3 haplotype was significantly increased 71 and in RA patients Keyszer et al reported that almost all haplotypes were found to be either 1G MMP-1/5A MMP-3 or 2G MMP-1/6A MMP-3 9. These last data are in line with our own results with a more frequent prevalence of the 1G/5A and 2G/6A combinations. So linkage in this region is demonstrated and possibly of importance. Theoretically the 5A-1171/6A MMP-3 promoter polymorphism could be linked to other much more important MMP genes with respect to the destructive processes in RA. In that case the described association between the 5A-1171/6A MMP-3 promoter polymorphism and radiological damage might be just an epiphenomenon. Despite the fact that were not able to demonstrate any association between MMP-1 or MMP-3 promoter polymorphism and radiological progression, sMMP-3AUC itself appeared to be correlated with radiological progression after 6 and 12 months. This confirms previous results from our 37 and other groups 38,39

that serum MMP-3 is associated with radiological progression in early RA. These results underline serum MMP-3 as an intriguing marker of disease severity. In conclusion in our study we could not find an association between MMP-1 and MMP-3 polymorphism and disease activity or severity at presentation. In addition the MMP-1 and MMP-3 promoter polymorphisms were not associated with radiological damage or radiological progression after a follow-up of 2 years. However MMP-3 promoter polymorphisms 5A/6A and 6A/6A were associated with increased serum MMP-3 levels relative to CRP, as analyzed over a 12 months period. Although cumulative MMP-3 levels are associated with radiological progression in RA, the influence of the MMP-3 promoter polymorphism is apparently not of clinical relevance.

Chapter 7

132

REFERENCES 1. Choy EH, Panayi GS. Cytokine pathways and joint inflammation in rheumatoid arthritis. N Engl J

Med 2001;344:907-16. 2. Navarro-Cano G, Del R, I, Pogosian S, Roldan JF, Escalante A. Association of mortality with

disease severity in rheumatoid arthritis, independent of comorbidity. Arthritis Rheum 2003;48:2425-33.

3. Michaud K, Messer J, Choi HK, Wolfe F. Direct medical costs and their predictors in patients with rheumatoid arthritis. Arthritis Rheum 2003;48:2750-62.

4. Emery P, Breedveld FC, Dougados M, Kalden JR, Schiff MH, Smolen JS. Early referral recommendation for newly diagnosed rheumatoid arthritis: evidence based development of a clinical guide. Ann Rheum Dis 2002;61:290-7.

5. Smolen JS, Steiner G. Therapeutic strategies for rheumatoid arthritis. Nat Rev Drug Discov 2003;2:473-88.

6. Combe B, Dougados M, Goupille P, Cantagrel A, Eliaou JF, Sibilia J et al. Prognostic factors for radiographic damage in early rheumatoid arthritis: a multiparameter prospective study. Arthritis Rheum 2001;44:1736-43.

7. Reveille JD. The genetic contribution to the pathogenesis of rheumatoid arthritis. Curr Opin Rheumatol 1998;10:187-200.

8. Constantin A, Lauwers-Cances V, Navaux F, Abbal M, van Meerwijk J, Mazieres B et al. Stromelysin 1 (matrix metalloproteinase 3) and HLA-DRB1 gene polymorphisms: Association with severity and progression of rheumatoid arthritis in a prospective study. Arthritis Rheum 2002;46:1754-62.

9. Keyszer G, Doerr S, Lechtenboehmer N, Rau R, Herborn G, Keysser M et al. Haplotype analysis of matrix metalloproteinase 1 (MMP-1) and MMP-3 alleles in rheumatoid arthritis: association of the 1G/5A haplotype with radiographic progression. Arthritis Rheum 2002;46:S97 (abstract 158).

10. Nagase H, Okada Y. Proteinases and matrix degradation. In: Kelley WN, Harris ED, Ruddy S, Sledge CB, editors. Textbook of rheumatology. Philadelphia: Saunders,W.B., 1997: 323-341.

11. Okada Y. Proteinases and matrix degradation. In: Ruddy S, Harris ED, Sledge CB, Budd RC, Sergent JS, editors. Kelley's Textbook of rheumatology. Philadelphia: Sanders,W.B., 2001: 55-72.

12. Nagase H. Activation mechanisms of matrix metalloproteinases. Biol Chem 1997;378:151-60. 13. Hasty KA, Reife RA, Kang AH, Stuart JM. The role of stromelysin in the cartilage destruction that

accompanies inflammatory arthritis. Arthritis Rheum 1990;33:388-97. 14. Mudgett JS, Hutchinson NI, Chartrain NA, Forsyth AJ, McDonnell J, Singer II et al. Susceptibility

of stromelysin 1-deficient mice to collagen-induced arthritis and cartilage destruction. Arthritis Rheum 1998;41:110-21.

15. Singer II, Kawka DW, Bayne EK, Donatelli SA, Weidner JR, Williams HR et al. VDIPEN, a metalloproteinase-generated neoepitope, is induced and immunolocalized in articular cartilage during inflammatory arthritis. J Clin Invest 1995;95:2178-86.






133




23. Ishiguro N, Ito T, Oguchi T, Kojima T, Iwata H, Ionescu M et al. Relationships of matrix metalloproteinases and their inhibitors to cartilage proteoglycan and collagen turnover and inflammation as revealed by analyses of synovial fluids from patients with rheumatoid arthritis. Arthritis Rheum 2001;44:2503-11.






29. Yoshihara Y, Nakamura H, Obata K, Yamada H, Hayakawa T, Fujikawa K et al. Matrix metalloproteinases and tissue inhibitors of metalloproteinases in synovial fluids from patients with rheumatoid arthritis or osteoarthritis. Ann Rheum Dis 2000;59:455-61.


31. Zucker S, Lysik RM, Zarrabi MH, Greenwald RA, Gruber B, Tickle SP et al. Elevated plasma stromelysin levels in arthritis. J Rheumatol 1994;21:2329-33.





Chapter 7

134



38. Yamanaka H, Matsuda Y, Tanaka M, Sendo W, Nakajima H, Taniguchi A et al. Serum matrix metalloproteinase 3 as a predictor of the degree of joint destruction during the six months after measurement, in patients with early rheumatoid arthritis. Arthritis Rheum 2000;43:852-8.

39. Tchetverikov I, Lard LR, DeGroot J, Verzijl N, TeKoppele JM, Breedveld FC et al. Matrix metalloproteinases-3, -8, -9 as markers of disease activity and joint damage progression in early rheumatoid arthritis. Ann Rheum Dis 2003;62:1094-9.

40. Nagase H, Woessner-JF J. Matrix metalloproteinases. J Biol Chem 1999;274:21491-4. 41. Visse R, Nagase H. Matrix metalloproteinases and tissue inhibitors of metalloproteinases:

structure, function, and biochemistry. Circ Res 2003;92:827-39. 42. Ye S. Polymorphism in matrix metalloproteinase gene promoters: implication in regulation of

gene expression and susceptibility of various diseases. Matrix Biol 2000;19:623-9. 43. Rutter JL, Mitchell TI, Buttice G, Meyers J, Gusella JF, Ozelius LJ et al. A single nucleotide

polymorphism in the matrix metalloproteinase-1 promoter creates an Ets binding site and augments transcription. Cancer Res 1998;58:5321-5.

44. Kanamori Y, Matsushima M, Minaguchi T, Kobayashi K, Sagae S, Kudo R et al. Correlation between expression of the matrix metalloproteinase-1 gene in ovarian cancers and an insertion/deletion polymorphism in its promoter region. Cancer Res 1999;59:4225-7.

45. Ye S, Eriksson P, Hamsten A, Kurkinen M, Humphries SE, Henney AM. Progression of coronary atherosclerosis is associated with a common genetic variant of the human stromelysin-1 promoter which results in reduced gene expression. J Biol Chem 1996;271:13055-60.

46. Constantin A, Lauwers-Cances V, Navaux F, Abbal M, van MJ, Mazieres B et al. Collagenase-1 (MMP-1) and HLA-DRB1 gene polymorphisms in rheumatoid arthritis: a prospective longitudinal study. J Rheumatol 2002;29:15-20.

47. Arnett FC, Edworthy SM, Bloch DA, McShane DJ, Fries JF, Cooper NS et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 1988;31:315-24.

48. Stenger AA, van Leeuwen MA, Houtman PM, Bruyn GA, Speerstra F, Barendsen BC et al. Early effective suppression of inflammation in rheumatoid arthritis reduces radiographic progression. Br J Rheumatol 1998;37:1157-63.

49. Ritchie DM, Boyle JA, McInnes JM, Jasani MK, Dalakos TG, Grieveson P et al. Clinical studies with an articular index for the assessment of joint tenderness in patients with rheumatoid arthritis. Q J Med 1968;37:393-406.

50. van der Heijde DM, van 't Hof M, van Riel PL, Theunisse LA, Lubberts EW, van Leeuwen MA et al. Judging disease activity in clinical practice in rheumatoid arthritis: first step in the development of a disease activity score. Ann Rheum Dis 1990;49:916-20.

51. van Rood JJ, van Leeuwen A, Ploem JS. Simultaneous detection of two cell populations by two-colour fluorescence and application to the recognition of B-cell determinants. Nature 1976;262:795-7.

52. Olerup O, Zetterquist H. HLA-DR typing by PCR amplification with sequence-specific primers (PCR-SSP) in 2 hours: an alternative to serological DR typing in clinical practice including donor-recipient matching in cadaveric transplantation. Tissue Antigens 1992;39:225-35.

53. Livak KJ. SNP genotyping by the 5'-nuclease reaction. Methods Mol Biol 2003;212:129-47.


135

54. Posthumus MD, Limburg PC, Westra J, van Leeuwen MA, van Rijswijk MH. Serum matrix metalloproteinase 3 levels during treatment with sulfasalazine or combination of methotrexate and sulfasalazine in patients with early rheumatoid arthritis. J Rheumatol 2002;29:883-9.






60. Lard LR, Boers M, Verhoeven A, Vos K, Visser H, Hazes JM et al. Early and aggressive treatment of rheumatoid arthritis patients affects the association of HLA class II antigens with progression of joint damage. Arthritis Rheum 2002;46:899-905.

61. Murphy G, Knauper V, Atkinson S, Butler G, English W, Hutton M et al. Matrix metalloproteinases in arthritic disease. Arthritis Res 2002;4 Suppl 3:S39-S49.

62. Ye S, Dhillon S, Turner SJ, Bateman AC, Theaker JM, Pickering RM et al. Invasiveness of cutaneous malignant melanoma is influenced by matrix metalloproteinase 1 gene polymorphism. Cancer Res 2001;61:1296-8.

63. Joos L, He JQ, Shepherdson MB, Connett JE, Anthonisen NR, Pare PD et al. The role of matrix metalloproteinase polymorphisms in the rate of decline in lung function. Hum Mol Genet 2002;11:569-76.

64. Johnson RW, Reveille JD, McNearney T, Fischbach M, Friedman AW, Ahn C et al. Lack of association of a functionally relevant single nucleotide polymorphism of matrix metalloproteinase-1 promoter with systemic sclerosis (scleroderma). Genes Immun 2001;2:273-5.

65. Borghaei RC, Sullivan C, Mochan E. Identification of a cytokine-induced repressor of interleukin-1 stimulated expression of stromelysin 1 (MMP-3). J Biol Chem 1999;274:2126-31.

66. Ye S, Watts GF, Mandalia S, Humphries SE, Henney AM. Preliminary report: genetic variation in the human stromelysin promoter is associated with progression of coronary atherosclerosis. Br Heart J 1995;73:209-15.

67. Terashima M, Akita H, Kanazawa K, Inoue N, Yamada S, Ito K et al. Stromelysin promoter 5A/6A polymorphism is associated with acute myocardial infarction. Circulation 1999;99:2717-9.

68. Yoon S, Tromp G, Vongpunsawad S, Ronkainen A, Juvonen T, Kuivaniemi H. Genetic analysis of MMP3, MMP9, and PAI-1 in Finnish patients with abdominal aortic or intracranial aneurysms. Biochem Biophys Res Commun 1999;265:563-8.

69. Yamada Y, Izawa H, Ichihara S, Takatsu F, Ishihara H, Hirayama H et al. Prediction of the risk of myocardial infarction from polymorphisms in candidate genes. N Engl J Med 2002;347:1916-23.

70. Ghilardi G, Biondi ML, Caputo M, Leviti S, DeMonti M, Guagnellini E et al. A single nucleotide polymorphism in the matrix metalloproteinase-3 promoter enhances breast cancer susceptibility. Clin Cancer Res 2002;8:3820-3.

71. Hinoda Y, Okayama N, Takano N, Fujimura K, Suehiro Y, Hamanaka Y et al. Association of functional polymorphisms of matrix metalloproteinase (MMP)-1 and MMP-3 genes with colorectal cancer. Int J Cancer 2002;102:526-9.

Chapter 8

Summary, general conclusions and future perspectives

Marcel D. Posthumus

Chapter 8

138

SUMMARY Rheumatoid arthritis (RA) is a systemic inflammatory disease, characterized by a chronic inflammation of synovial tissue and in most cases progressive destruction of cartilage and bone 1. In particular joint destruction is one of the strongest predictors of long term outcome and disability in RA 2.

The matrix metalloproteinases (MMPs) are considered to be key mediators in the destruction of the matrix components of cartilage and bone in inflamed RA joints 3. In synovial tissue, chondrocytes and synovial fluid of inflamed RA joints many MMPs are upregulated 4,5. In particular MMP-3 is of interest because it is supposed to play a prominent role in the pathogenesis of matrix degradation in the inflamed joint 6, although it is not the only key enzyme 7. It is locally produced as a latent pro-enzyme and activated within the affected joint, resulting in elevated levels in local tissue 8-10 and synovial fluid 4,11.

Serum MMP-3 in RA is thought to originate from these inflamed joints because there are significant correlations between MMP-3 levels in synovial fluid and serum 12, as well as between serum MMP-3 levels and the number of clinically active inflamed joints 12,13. Furthermore local therapies, such as intra-articular corticosteroid injections, cause significant reductions in serum MMP-3 levels in synovial fluid and in the systemic circulation of patients with RA 14,15.

Elevated serum MMP-3 levels are not specific for RA, they are also found in several other rheumatic and non-rheumatic diseases (as described in chapter 1). Nevertheless, in RA the systemic MMP-3 levels are considered to be a direct reflection of local synthesis induced by pro-inflammatory cytokines. As such, serum MMP-3 could be a useful, systemic marker of disease activity 16 and joint destruction 17,18 in RA. In this respect serum MMP-3 may reflect joint inflammation and/or destruction more directly than C-reactive protein (CRP), which is produced by the liver indirecly after cytokine stimulation 19. In addition it has been suggested that the pathophysiologic mechanisms of joint inflamma-tion, as reflected by CRP, may be partially independent of the mechanisms of destruction 20-22, which are also determined by the prominent local cytokine-, protease-, and inhibitor-environment. This possible uncoupling of inflammation and destruction could be one of the explanations for the wide inter-individual variation in radiological damage despite comparable inflammation as reflected by, for example, CRP 23. Moreover it might imply re-evaluation of traditional indicators and a search for new, more specific markers of disease activity and joint destruction.

Summary, conclusions and future perspectives

139

The aim of the thesis was to investigate the significance of serum MMP-3 in relation to disease activity, and in particular in relation to radiological progression, in patients with early RA. The results of the clinical studies are based on data derived from a prospective observational study in patients with early RA (disease symptoms < 1 year and disease modifying anti-rheumatic drugs (DMARDs) naive at study entry). The follow-up protocol included monthly assessments of disease activity and X-rays of hands and feet every 6 months.

In chapter 2 a general review is provided on matrix metalloproteinases. MMPs are a family of zinc-containing proteases and are thought to be key mediators involved in the remodeling of the extra-cellular matrix. The regulation of MMP production, activation and inhibition is discussed, as well as the role of MMPs in physiological and pathological conditions, in particular in RA. Finally, therapeutic options of MMP-inhibition are set in perspective. In chapter 3 the significance of serum MMP-3 levels in relation to the development of radiological damage in early RA patients was evaluated.

Serum MMP-3 levels were measured in 46 healthy controls, 19 osteoarthritis patients and in 78 early RA patients (joint symptoms < 1 year at presentation (T0)): 48 patients without and 30 with radiological damage at T0. Serum MMP-3, measured by ELISA, and radiological damage, scored according to Sharp/van der Heijde method were assessed at 0, 6, 12 and 24 months. MMP-3 levels in controls and osteoarthritis patients were low or undetectable with no differences between the groups. Levels in RA were higher than in controls. Initial MMP-3 levels in the patients with radiological damage at T0 (n=30) were higher than the levels in patients without any radiological damage during follow-up (n=19), but were not different from patients who developed radiological damage during the study (n=29). In the patients without radiological damage at T0 there was a significant correlation between MMP-3 at T0 and the total radiological damage after 6 months and 12 months. This correlation was almost exclusively determined by the item joint space narrowing in the Sharp score. In conclusion the initial serum MMP-3 level seems to be an indicator for the development of radiological damage in patients with early RA and appears to be particularly indicative of cartilage degradation.

Chapter 8

140

In chapter 4 the clinical significance of serial measurements of serum MMP-3 levels in relation to markers of disease activity and radiological progression in early RA was analyzed. In a 3 year prospective follow-up study of 33 patients with early RA monthly measurements of serum MMP-3 were transformed into time-integrated values for 6 months periods for comparison with other markers of disease activity such as swollen joint count (SJC), tender joint count (TJC), Ritchie articular index (RAI), the disease activity score (DAS), erythrocyte sedimentation rate (ESR), CRP, and radiological progression, assessed according to the Sharp/van der Heijde method.

Significant correlations were found between serum MMP-3 and SJC, ESR, and CRP during all periods and for most 6-months period with the DAS. There were no correlations between serum MMP-3 and TJC or the RAI. During the first 12 months serum MMP-3 only correlated with the item joint space narrowing of the Sharp score. After 12 months of follow-up it also correlated with the total Sharp score and after 18 months it was correlated with all 3 items of the Sharp score. There was a wide inter-individual variation in the relation between serum MMP-3 and radiological progression but intra-individually this relation was rather constant. In summary, in this study the time-integrated values of serum MMP-3 correlated with time-integrated values of other markers of disease activity. Of the radiological scores, as outcome measures, especially joint space narrowing was closely correlated with cumulative serum MMP-3. In chapter 5 the effects of treatment with sulphasalazine (SSZ) or the combination of methotrexate (MTX) and SSZ on serum MMP-3 levels in patients with early RA were investigated

Eighty-two patients with early RA and DMARD-naive at presentation were selected who were treated with SSZ (2000 mg/day) or with the combination of MTX (7.5-15 mg/week) and SSZ. Serum MMP-3 levels, CRP, ESR, SJC, TJC, RAI, and the DAS were determined at 4-week intervals during a follow-up of 28 weeks for each treatment group. Response or non-response was based on clinical grounds in combination with the influence on CRP levels (> 50% reduction in joint scores or CRP (or normalization of CRP)) at 12, 20, and 28 weeks.

SSZ responders (n=52) had lower baseline values of serum MMP-3, CRP, and ESR, compared to partial/non-responders (n=30), but did not differ in joint scores and DAS at entrance. In the SSZ responder group all variables decreased. In the SSZ partial/non-responders CRP, ESR, and SJC decreased in contrast to serum MMP-3, TJC, RAI, and DAS-3. After addition of MTX all variables decreased in


141

24 out of the 30 patients who had shown a partial or no response on SSZ previously. In the SSZ responders there was a delayed decrease in serum MMP-3 compared to CRP.

In conclusion the serum MMP-3 levels decrease in early RA patients who respond to SSZ or to the combination of MTX and SSZ. In patients who respond to SSZ the changes in serum MMP-3 levels show a delayed response compared to CRP. In chapter 6 serum MMP-3 levels in comparison to CRP in periods with and without progression of radiological damage in patients with early RA were evaluated.

Thirty-two patients with RA and radiological progression (> 5 points according to the Sharp/van der Heijde method) during 6 months followed by a 6 months period without radiological progression (< 1 point) were selected from a prospective follow-up study of early RA patients. Serum MMP-3 levels, CRP, ESR, DAS, SJC, TJC and, RAI were measured monthly and results were transformed into mean values for the 6 months periods.

During the period with radiological progression the mean serum MMP-3 correlated significantly with the mean CRP, ESR and swollen joint count. In the period without radiological progression the mean serum MMP-3 only correlated with the mean CRP. Individual changes - in terms of percentage (%) - between the 2 periods showed a decrease in both mean serum MMP-3 and CRP in most patients, in parallel with other markers of disease activity. However this individual change (%) in the mean serum MMP-3 or CRP was not correlated with the difference in radiological progression between the two consecutive periods.

So serum MMP-3 and CRP were closely related and there seemed to be no difference between serum MMP-3 and CRP with regard to the monitoring of the progression of radiological damage. In chapter 7 the significance of MMP-1 and MMP-3 promoter polymorphisms in relation to disease activity and radiological damage in patients with early RA was analyzed. Functional MMP promoter polymorphisms have been reported for MMP-1 and MMP-3 24 resulting in different amounts of gene product 25-27. Consequently this might lead to different levels of MMPs and more or less joint destruction after the same stimulus. MMP-1 (1G-1607/2G) and MMP-3 (5A-1171/6A) promoter polymorphisms were determined in a cohort of 448 early RA patients. Clinical and laboratory

Chapter 8

142

markers of disease activity and severity at presentation (T0) were related to these polymorphisms. In addition the relations with total radiological damage and radiological progression (Sharp/van der Heijde method) after 2 years were evaluated. In a subgroup initial serum MMP-3, cumulative serum MMP-3 (serum MMP-3 area under the curve; sMMP-3AUC) over 0.5 and 1 year and serum MMP-3AUC/CRPAUC ratio in relation to the MMP-3 promoter polymorphism were studied.

No association was found between MMP-1 and MMP-3 promoter polymorphisms and disease activity, severity or radiological damage at T0. MMP-1 and MMP-3 promoter polymorphisms showed strong linkage disequilibrium, but no associations were found with “risk factors” such as rheumatoid factor (RF), anti-cyclic citrullinated peptide (anti-CCP) and HLA-DR4 status. The MMP-1 and MMP-3 promoter polymorphisms were not associated with radiological damage or radiological progression after a follow-up of 2 years. Serum MMP-3AUC correlated significantly with radiological progression after 6 and 12 months. However the serum MMP-3AUC after 6 and 12 months did not differ across the three MMP-3 promoter polymorphism groups. On the other hand the MMP-3 promoter polymorphisms 5A/6A and 6A/6A were associated with higher serum MMP-3 levels relative to CRP (serum MMP-3AUC/CRPAUC ratios), as analyzed over a 12 months period.

These last results are in contrast with in vitro experiments in which the 6A/6A genotype seems to be a “low producer” of MMP-3 transcripts 27. On the other hand, our results might be in line with the data of Constantin et al who found more radiological damage in the 5A/6A and 6A/6A genotypes.

So, although cumulative MMP-3 levels are associated with radiological progression in RA, the influence of the MMP-3 promoter polymorphism is apparently not of great clinical relevance.


143

CONCLUSIONS The presented studies reveal several interesting conclusions with respect to the significance of serum MMP-3 in early RA: 1. The initial serum MMP-3 level is an prognostic factor for the development of

radiological damage (chapter 3) 2. The initial serum MMP-3 and cumulative serum MMP-3 (time integrated

values) levels are correlated with radiological progression, in particular with joint space narrowing (chapter 3 and 4)

3. Serum MMP-3 is correlated with other markers of disease activity, such as joint swelling, ESR, CRP and the DAS (chapter 4)

4. Serum MMP-3 levels decrease in patients who respond to treatment with the DMARDs sulphasalazine or to the combination methotrexate/sulphasalazine. In the patients who respond to sulphasalazine the changes in serum MMP-3 show a delayed response compared to CRP (chapter 5)

5. Serum MMP-3 and CRP are closely correlated (chapter 4 and 6) and there seems to be no difference between serum MMP-3 and CRP with regard to the monitoring of the progression of radiological damage (chapter 6)

6. MMP-1 and MMP-3 promoter polymorphisms were not associated with disease activity or severity at presentation. In addition the MMP-1 and MMP-3 promoter polymorphisms were not associated with radiological damage or radiological progression after a follow-up of 2 years. However the MMP-3 promoter polymorphisms 5A/6A and 6A/6A were associated with increased serum MMP-3 levels relative to CRP, as analyzed over a 12 months period. (chapter 7).

In summary: I. Serum MMP-3 is an indicator (marker) of disease activity and radiological

damage/progression II. Serum MMP-3 is not superior to CRP with respect to the monitoring of

disease activity and radiological progression, probably due to the close correlation between serum MMP-3 and CRP

III. MMP-3 promoter polymorphisms are apparently not of clinical significance in patients with early RA.

Chapter 8

144

FUTURE PERSPECTIVES Despite optimal care and recent new therapeutic advances in RA, such as tumor necrosis factor α (TNF-α) blocking agents and Interleukin 1 (IL-1) receptor antagonist, it is often not possible to completely stop joint destruction 28. This progressive joint destruction is considered to be a consequence of a dysbalance between an excess of activated MMPs and inadequate levels of their inhibitors.

Over the last 10-15 years a number of “specific” MMP-inhibitors, mainly focused on inhibition of MMP activity, have been developed, with disappointing results (see chapter 1) 29.

Recent studies are focused at the signal transduction pathways. Blockade of the MAPK, NFκB and JAK/STAT pathways lead to inhibition of MMP expression in tissue culture experiments and in animal models of arthritis 30-32. This implicates that this type of intervention leads to both a direct inhibition of MMP production as well as an indirect inhibition of MMP production by inhibition of the pro-inflammatory cytokines. Clinical studies targeted at these signal transduction pathways are currently evolving 33.

In this context it is essential to realize that further research with respect to markers of disease activity and destruction is of great importance. Not only because more specific MMP inhibition might result in an uncoupling of inflammation and destruction but also because currently used parameters of inflammation like the acute phase proteins such as CRP are mediated by similar signal transduction pathways 34. This means that the use of CRP and possibly other acute phase proteins to monitor inflammatory activity during treatment with signal transduction inhibitors, such as p38 MAPK inhibitors, may not be valid. The production may be influenced both at the level of cytokine production (inhibition of TNF-α, IL-1, IL-6) and at the protein synthesis level of the liver-response itself.

Future studies will be focused on these signal transduction pathways. Not only with respect to effects on disease activity and joint destruction but also regarding the effects on markers of inflammation and destruction.


145

REFERENCES 1. Choy EH, Panayi GS. Cytokine pathways and joint inflammation in rheumatoid arthritis. N Eng J

Med 2001;344(12):907-16. 2. Kirwan JR. Links between radiological change, disability, and pathology in rheumatoid arthritis. J

Rheumatol 2001;28(4):881-6. 3. Murphy G, Knauper V, Atkinson S, Butler G, English W, Hutton M et al. Matrix metalloprotein-ases

in arthritic disease. Arthritis Res 2002;4 Suppl 3:S39-S49. 4. Yoshihara Y, Nakamura H, Obata K, Yamada H, Hayakawa T, Fujikawa K et al. Matrix

metalloproteinases and tissue inhibitors of metalloproteinases in synovial fluids from patients with rheumatoid arthritis or osteoarthritis. Ann Rheum Dis 2000;59(6):455-61.

5. Konttinen YT, Ainola M, Valleala H, Ma J, Ida H, Mandelin J et al. Analysis of 16 different matrix metalloproteinases (MMP-1 to MMP-20) in the synovial membrane: different profiles in trauma and rheumatoid arthritis. Ann Rheum Dis 1999;58(11):691-7.

6. Hasty KA, Reife RA, Kang AH, Stuart JM. The role of stromelysin in the cartilage destruction that accompanies inflammatory arthritis. Arthritis Rheum 1990;33(3):388-97.

7. Mudgett JS, Hutchinson NI, Chartrain NA, Forsyth AJ, McDonnell J, Singer II et al. Susceptibility of stromelysin 1-deficient mice to collagen-induced arthritis and cartilage destruction. Arthritis Rheum 1998;41(1):110-21.

8. Okada Y, Takeuchi N, Tomita K, Nakanishi I, Nagase H. Immunolocalization of matrix metalloproteinase 3 (stromelysin) in rheumatoid synovioblasts (B cells): correlation with rheumatoid arthritis. Ann Rheum Dis 1989;48(8):645-53.

9. Tetlow LC, Lees M, Woolley DE. Comparative studies of collagenase and stromelysin-1 expression by rheumatoid synoviocytes in vitro. Virchows Arch 1995;425(6):569-76.

10. Martel Pelletier J, McCollum R, Fujimoto N, Obata K, Cloutier JM, Pelletier JP. Excess of metalloproteases over tissue inhibitor of metalloprotease may contribute to cartilage degradation in osteoarthritis and rheumatoid arthritis. Lab Invest 1994;70(6):807-15.

11. Ishiguro N, Ito T, Miyazaki K, Iwata H. Matrix metalloproteinases, tissue inhibitors of metalloproteinases, and glycosaminoglycans in synovial fluid from patients with rheumatoid arthritis. J Rheumatol 1999;26(1):34-40.

12. Yoshihara Y, Obata K, Fujimoto N, Yamashita K, Hayakawa T, Shimmei M. Increased levels of stromelysin-1 and tissue inhibitor of metalloproteinases-1 in sera from patients with rheumatoid arthritis. Arthritis Rheum 1995;38(7):969-75.

13. Sasaki S, Iwata H, Ishiguro N, Obata K, Miura T. Detection of stromelysin in synovial fluid and serum from patients with rheumatoid arthritis and osteoarthritis. Clin Rheumatol 1994;13(2):228-33.

14. Cheung NT, Mattey DL, Dawes PT, Taylor DJ. Serum pro-matrix metalloproteinase 3 in rheumatoid arthritis: a reflection of local or systemic inflammation? Arthritis Rheum 1996;39(5):884-6.

15. Taylor DJ, Cheung NT, Dawes PT. Increased serum proMMP-3 in inflammatory arthritides: a potential indicator of synovial inflammatory monokine activity. Ann Rheum Dis 1994;53(11):768-72.

16. Ribbens C, Andre B, Jaspar JM, Kaye O, Kaiser MJ, De Groote D et al. Matrix metalloproteinase-3 serum levels are correlated with disease activity and predict clinical response in rheumatoid arthritis. J Rheumatol 2000;27(4):888-93.

17. Yamanaka H, Matsuda Y, Tanaka M, Sendo W, Nakajima H, Taniguchi A et al. Serum matrix metalloproteinase 3 as a predictor of the degree of joint destruction during the six months after measurement, in patients with early rheumatoid arthritis. Arthritis Rheum 2000;43(4):852-8.

18. Tchetverikov I, Lard LR, DeGroot J, Verzijl N, TeKoppele JM, Breedveld FC et al. Matrix metalloproteinases-3, -8, -9 as markers of disease activity and joint damage progression in early rheumatoid arthritis. Ann Rheum Dis 2003;62(11):1094-9.

19. Gabay C, Kushner I. Acute-phase proteins and other systemic responses to inflammation. N Engl J Med 1999;340(6):448-54.

20. Mulherin D, Fitzgerald O, Bresnihan B. Clinical improvement and radiological deterioration in rheumatoid arthritis: evidence that the pathogenesis of synovial inflammation and articular erosion may differ. Br J Rheumatol 1996;35(12):1263-8.

Chapter 8

146

21. Kirwan JR. The relationship between synovitis and erosions in rheumatoid arthritis. Br J Rheumatol 1997;36(2):225-8.

22. Cunnane G, Fitzgerald O, Beeton C, Cawston TE, Bresnihan B. Early joint erosions and serum levels of matrix metalloproteinase 1, matrix metalloproteinase 3, and tissue inhibitor of metalloproteinases 1 in rheumatoid arthritis. Arthritis Rheum 2001;44(10):2263-74.

23. van Leeuwen MA, van Rijswijk MH, van der Heijde DM, Te Meerman G, van Riel PL, Houtman PM et al. The acute-phase response in relation to radiographic progression in early rheumatoid arthritis: a prospective study during the first three years of the disease. Br J Rheumatol 1993;32 Suppl 3:9-13.

24. Ye S. Polymorphism in matrix metalloproteinase gene promoters: implication in regulation of gene expression and susceptibility of various diseases. Matrix Biol 2000;19(7):623-9.

25. Rutter JL, Mitchell TI, Buttice G, Meyers J, Gusella JF, Ozelius LJ et al. A single nucleotide polymorphism in the matrix metalloproteinase-1 promoter creates an Ets binding site and augments transcription. Cancer Res 1998;58(23):5321-5.

26. Kanamori Y, Matsushima M, Minaguchi T, Kobayashi K, Sagae S, Kudo R et al. Correlation between expression of the matrix metalloproteinase-1 gene in ovarian cancers and an insertion/deletion polymorphism in its promoter region. Cancer Res 1999;59(17):4225-7.

27. Ye S, Eriksson P, Hamsten A, Kurkinen M, Humphries SE, Henney AM. Progression of coronary atherosclerosis is associated with a common genetic variant of the human stromelysin-1 promoter which results in reduced gene expression. J Biol Chem 1996;271(22):13055-60.

28. Genovese MC, Bathon JM, Martin RW, Fleischmann RM, Tesser JR, Schiff MH et al. Etanercept versus methotrexate in patients with early rheumatoid arthritis: two-year radiographic and clinical outcomes. Arthritis Rheum 2002;46(6):1443-50.

29. Close DR. Matrix metalloproteinase inhibitors in rheumatic diseases. Ann Rheum Dis 2001;60 Suppl 3:iii62-iii67.

30. Mengshol JA, Vincenti MP, Coon CI, Barchowsky A, Brinckerhoff CE. Interleukin-1 induction of collagenase 3 (matrix metalloproteinase 13) gene expression in chondrocytes requires p38, c-Jun N-terminal kinase, and nuclear factor kappaB: differential regulation of collagenase 1 and collagenase 3. Arthritis Rheum 2000;43(4):801-11.

31. Mengshol JA, Mix KS, Brinckerhoff CE. Matrix metalloproteinases as therapeutic targets in arthritic diseases: bull's-eye or missing the mark? Arthritis Rheum 2002;46(1):13-20.

32. Liacini A, Sylvester J, Li WQ, Zafarullah M. Inhibition of interleukin-1-stimulated MAP kinases, activating protein-1 (AP-1) and nuclear factor kappa B (NF-kappa B) transcription factors down-regulates matrix metalloproteinase gene expression in articular chondrocytes. Matrix Biol 2002;21(3):251-62.

33. Smolen JS, Steiner G. Therapeutic strategies for rheumatoid arthritis. Nat Rev Drug Discov 2003;2(6):473-88.

34. Agrawal A, Cha-Molstad H, Samols D, Kushner I. Overexpressed nuclear factor-kappaB can participate in endogenous C-reactive protein induction, and enhances the effects of C/EBPbeta and signal transducer and activator of transcription-3. Immunology 2003;108(4):539-47.

Chapter 9

Samenvatting

Marcel D. Posthumus

Chapter 9

148

SAMENVATTING Reumatoïde artritis (RA) is een aandoening die voorkomt bij 0,5-1% van de bevolking en die gekenmerkt wordt door een chronische ontsteking van meerdere gewrichten (polyartritis). Behalve een symmetrische artritis van kleine hand- en voetgewrichten zijn ook grote gewrichten vaak aangedaan. Daarnaast komen extra-articulaire verschijnselen, zoals reuma-noduli frequent voor.

Naast de tekenen van ontsteking ontstaat er bij de meeste patiënten helaas ook schade aan het bot en het kraakbeen. Vooral dit laatste gaat gepaard met misvormingen en handicaps.

De zogenaamde matrix metalloproteinasen (MMPs) blijken een belangrijke rol te spelen in dit proces van de gewrichtsbeschadiging. Het zijn enzymen (zink bevattende proteasen), die belangrijke onderdelen van de matrix van kraakbeen en bot, zoals proteoglycanen en collagenen, afbreken. Op dit moment zijn er in de mens 23 MMPs beschreven.

Aan het MMP-3 wordt een belangrijke rol toebedacht omdat het vele onderdelen van de matrix op belangrijke plaatsen kan splitsen en het enzym duidelijk aanwezig is in het ontstoken gewricht. Bovendien is het in het bloed meetbaar en op die manier zou het wellicht als indicator gebruikt kunnen worden voor de mate van ontstekingsactiviteit en gewrichtsbeschadiging.

De laatste jaren is de behandeling van patiënten met RA enorm veranderd. Meer uitleg/informatie, intensievere begeleiding, maar met name een snellere behandeling met veel betere geneesmiddelen heeft de prognose sterk verbeterd. Voor een dergelijke vroege behandeling is het belangrijk om vooral patiënten met een ernstig, agressief beloop snel op te sporen en te behandelen. Maten (“markers”) van ziekteactiviteit, zoals het aantal ontstoken gewrichten of de “bloedbezinking” (BSE), en maten van gewrichtsbeschadiging, zoals de op röntgenfoto’s waarneembare schade, spelen daarbij een belangrijke rol. Ook het effect van een behandeling wordt mede beoordeeld op grond van deze maten. Markers die gemakkelijk te meten zijn (bijv. in het bloed) en die een directe en goede afspiegeling geven van wat er precies in het ontstoken gewricht gebeurt zijn dan essentieel.

Het MMP-3 zou een dergelijk hulpmiddel kunnen zijn omdat het ter plekke, in het ontstoken gewricht, wordt geproduceerd. Het zou op die manier een directe afspiegeling kunnen zijn van het locale proces van ontsteking en schade. Dit in tegenstelling tot de op dit moment veel gebruikte “bloedbezinking” (BSE) en het C-reactive protein (CRP) -gehalte. Deze laatste twee markers worden via een indirecte weg door de lever geproduceerd o.i.v. zogenaamde cytokinen, die afkomstig zijn vanuit het ontstoken gewricht. Dergelijke cytokinen worden ook

Samenvatting

149

verhoogd aangemaakt bij andere vormen van ontsteking, zodat bijvoorbeeld ook infecties e.d. de BSE of het CRP kunnen doen stijgen.

Doel van het onderzoek (hoofdstuk 1) Ons onderzoek heeft zich vooral toegespitst op het belang van het serum MMP-3 bij patiënten met een recent gediagnosticeerde RA. De gegevens werden verkregen tijdens een prospectief follow-up onderzoek. Bij insluiting in het onderzoek hadden alle patiënten < 1 jaar gewrichtsklachten.

In hoofdstuk 2 wordt een algemeen overzicht gegeven van de MMPs en de wijze waarop zij geproduceerd, geactiveerd en geremd worden. Naast de fysiologische betekenis wordt ook hun belang bij diverse aandoeningen, maar vooral bij RA besproken.

In hoofdstuk 3 wordt de relatie beschreven tussen het serum MMP-3 en de ontwikkeling van radiologische waarneembare schade bij patiënten met een recent gediagnosticeerde RA.

Het serum MMP-3 gehalte, gemeten bij aanvang bleek geassocieerd te zijn met de ontwikkeling van radiologisch waarneembare schade en was vooral een indicator voor kraakbeenverlies.

In hoofdstuk 4 is een studie beschreven waarin de patiënten 3 jaar lang zijn gevolgd en waarin gekeken is naar de relatie tussen het serum MMP-3 en proces variabelen (zoals het aantal ontstoken gewrichten, de BSE, het CRP, de disease activity score (DAS)) en uitkomstvariabelen (zoals de radiologische scores, waarmee de mate van gewrichtsbeschadiging in getal wordt uitgedrukt). Daarbij is, voor een goede vergelijking, de oppervlakte onder de curve (“area under the curve”=AUC) berekend van de diverse procesvariabelen (markers). Op die manier wordt een maat verkregen die de totale productie gedurende een bepaalde periode (bijv. 6 maanden) weergeeft. Een dergelijke berekening van een proces variabele is vervolgens goed te vergelijken met een uitkomst variabele zoals de mate van afwijkingen op een röntgenfoto.

Serum MMP-3AUC bleek geassocieerd te zijn met markers voor ziekteactiviteit zoals de CRPAUC, BSEAUC, aantal gezwollen gewrichtenAUC en de DASAUC. Van de radiologische scores bleek vooral de gewrichtsspleetversmalling (een maat voor kraakbeenverlies) geassocieerd te zijn met de serum MMP-3AUC.

In hoofdstuk 5 wordt een studie beschreven waarin gekeken is naar het effect van sulphasalazine (SSZ) of de combinatie SSZ/methotrexaat (MTX) op de serum MMP-3 spiegels. SSZ en MTX zijn belangrijke disease modifying anti-rheumatic drugs (DMARDs), medicamenten die de potentie hebben de activiteit

Chapter 9

150

van de ontsteking in de gewrichten en de progressie van gewrichtsschade te remmen. Serum MMP-3 spiegels bleken te dalen bij patiënten die goed reageerden op SSZ of de combinatie SSZ/MTX. In de groep die goed reageerde op SSZ daalde het CRP sneller dan het serum MMP-3.

In hoofdstuk 6 is gekeken naar het serum MMP-3 in vergelijking met het CRP gedurende perioden meten perioden zonder radiologische progressie.

Ook in deze studie bleken het serum MMP-3 en het CRP nauw gecorreleerd te zijn. Bovendien bleken er geen verschillen te zijn tussen het serum MMP-3 en het CRP wat betreft hun betekenis voor het vervolgen van de progressie van radiologische schade.

In hoofdstuk 7 is een studie beschreven betreffende MMP-1 and MMP-3 promoter polymorfismen. In de promoter van het gen voor MMP-1 en MMP-3 zijn polymorfismen (kleine verschillen) beschreven welke, in ieder geval in het laboratorium (in vitro), van belang lijken te zijn. Bijvoorbeeld op positie -1171 in de MMP-3 promoter kunnen of 5 of 6 adeninen voorkomen. De 5A variant zou, tenminste in vitro, méér gen product (mRNA) tot expressie brengen. Meer gen product zou kunnen leiden tot meer MMP-3, meer destructie en op die manier zou het MMP-3 promoter polymorfisme een prognostische factor voor destructie kunnen zijn. In deze studie met 448 RA patiënten werd geen associatie tussen deze promoter polymorfismen en ziekteactiviteit en/of ernst bij presentatie gevonden. Ook radiologische schade en/of progressie waren niet geassocieerd met deze promoter polymorfismen. Samenvattend kan het volgende geconcludeerd worden: • De initiële serum MMP-3 spiegel is een prognostische factor voor het

ontstaan van radiologische schade (hoofdstuk 3) • De initiële en cumulatieve serum MMP-3 spiegels zijn gecorreleerd met

radiologische progressie en vooral met gewrichtsspleetversmalling (hoofdstuk 3 en 4)

• Het serum MMP-3 is gecorreleerd met andere markers van ziekteactiviteit zoals de BSE, het CRP, het aantal ontstoken gewrichten en de disease activity score (hoofdstuk 4)

• Het serum MMP-3 daalt bij patiënten die goed reageren op een behandeling met SSZ of de combinatie SSZ/MTX. In de SSZ monotherapie responders daalt het serum MMP-3 trager dan het CRP (hoofdstuk 5)

Samenvatting

151

• Het serum MMP-3 en het CRP zijn nauw gecorreleerd (hoofdstuk 4 en 6) en er zijn geen verschillen tussen het serum MMP-3 en het CRP wat betreft de monitoring van radiologische progressie (hoofdstuk 6)

• MMP-1 en MMP-3 promoter polymorfismen zijn niet geassocieerd met ziekte activiteit en/of ernst bij presentatie. Ook radiologische schade en/of progressie daarvan zijn niet geassocieerd met deze promoter polymorfismen. De MMP-3 promoter polymorfismen 5A/6A en 6A/6A zijn geassocieerd met een hoger serum MMP-3 relatief t.o.v. CRP (hoofdstuk 7).

Kortom: 1. Serum MMP-3 is een marker voor ziekteactiviteit en radiologische

schade/progressie 2. Serum MMP-3 is niet beter als instrument voor de monitoring van

ziekteactiviteit en radiologische progressie dan het CRP, waarschijnlijk omdat deze twee markers onderling nauw gecorreleerd zijn

3. MMP-3 (en MMP-1) promoter polymorfismen zijn klinisch niet van groot belang voor RA patiënten.

TOEKOMSTIG ONDERZOEK In de afgelopen jaren is, wereldwijd veel onderzoek gedaan in de richting van de rol van de MMPs in het algemeen en naar het nut van het meten van systemische MMP spiegels als markers van ziekteactiviteit en gewrichtsschade bij RA. Tegelijkertijd is ook MMP-remming als medicamenteuze interventie onderzocht, niet alleen bij gewrichtsaandoeningen maar vooral ook in de oncologie. Helaas zijn de resultaten tot nu toe teleurstellend gebleken (zie hoofdstuk 2). Het huidige onderzoek lijkt zich steeds meer toe te spitsen op remming van de zogenaamde signaal transductie routes. Deze routes regelen de signaal overdracht vanaf de celmembraan naar het genetisch materiaal in de kern van de cel (zie hoofdstuk 2 en 7). Via die weg kan de productie van vele pro-inflammatoire cytokinen, maar ook van vele MMPs worden beïnvloed. Specifieke remming van deze routes kan echter, naast beïnvloeding van de ontsteking en destructie, ook belangrijke andere effecten hebben. Bijvoorbeeld ook de productie van traditionele indicatoren van ziekteactiviteit en gewrichtsschade, zoals de BSE en het CRP, verlopen via vergelijkbare routes. Ons toekomstig onderzoek spitst zich toe op de “signal transduction pathways” in vroege RA, niet alleen m.b.t. de effecten op ziekteactiviteit en gewrichts-destructie, maar ook wat betreft de effecten op markers van deze processen.

153

Dankwoord Op het moment dat ik dit dankwoord schreef bedacht ik me dat ik niet goed meer wist wanneer “het project promotie” eigenlijk begonnen is. Ik weet nog wel dat Hans Rasker mij destijds, op zijn bekende positieve manier enthousiast heeft gemaakt voor de reumatologie. Het beviel zelfs zo goed dat er plannen ontstonden te promoveren in Enschede. Maar soms lopen de dingen anders, krijg je een kans die je niet kan laten lopen en werk je voor je het weet in Groningen.

Of is het misschien al begonnen tijdens het EULAR congres in Amsterdam toen ik met Piet Limburg en Martin Gerding (internist en paranimf) zat te eten op een klein terrasje aan de kant van de weg? Werden toen de ideeën door Piet in de week gelegd?

Het eerste werk kreeg al snel vorm en ging over de voorspellende waarde van het serum MMP-3 met betrekking tot de ontwikkeling van radiologische schade bij patiënten met een recent ontstane reumatoïde artritis. Het resulteerde in een “oral presentation” (jawel, in colbert) tijdens de “American College of Rheumatology” in 1998. Vervolgens kwam artikel 2, gevolgd door nummer 3 en zo waren alle hoofdstukken al snel ingevuld en werd het “traject” uitgestippeld.

Natuurlijk kan een dergelijk promotietraject nooit zonder hulp van vele anderen die je helpen en steunen tijdens deze proeve van bekwaamheid en vooral doorzettingsvermogen. Allereerst wil ik Piet Limburg bedanken. Voor je originele ideeën, je scherpe oog voor detail, het immense geduld en de altijd positieve benadering. Minstens evenveel dank ben ik verschuldigd aan Miek van Leeuwen. Al die versies van artikelen die je hebt gelezen, al die vragen die je hebt beantwoord evenals je steunende woorden tijdens sombere momenten! Niet alleen als promotor maar ook als motivator wil ik Martin van Rijswijk bedanken voor alle ondersteunende faciliteiten en inspirerende ideeën. Hannie Westra wil ik vooral bedanken voor het opzetten van de MMP-3 ELISA, de metingen en het helpen verzamelen van alle data uit de database en de bekende mapjes. Johan Bijzet verrichte veel werk aan de promoter polymorphismen en was van grote waarde bij het maken van de “plaatjes” en tijdens vele andere computertechnische calamiteiten.

Vanzelfsprekend had al dit werk niet verricht kunnen worden zonder hulp van mijn collega reumatologen en reumatologen in opleiding. In de tijd dat ik zat te ploeteren met de data en de computer deden zij de patiëntenzorg, het onderwijs etc. Vooral tijdens de finale (2003) werd e.e.a. duidelijk projectmatig aangepakt met uiteindelijk dit proefschrift als resultaat. Bouke Hazenberg bedankt voor het waarnemen van taken als het onderwijs, het kritisch doorlezen van het manuscript, het aanhoren van de klaagzangen en je hulp als paranimf. Hendrika

154

Bootsma, als kamergenoot heb je van dichtbij meegeleefd en meegeleden. Martha Leysma, Liesbeth Brouwer, Thea Jagt en Ingrid van Gameren wil ik bedanken voor het overnemen van vooral patiëntenzorg taken, evenals Wietie Lolkema voor het verzamelen van de klinische gegevens van de patiënten en het opslaan van de data. Ilja Nolte, Roy Stewart en Wim Sluiter bedankt voor de statistische steun en Bouke Hepkema voor de hulp bij het uitzoeken van de “primers & probes” voor het MMP promoter polymorphisme stuk. Vooral tijdens de eindsprint was Janny Havinga voor het maken van de lay-out van het proefschrift en alle andere administratieve handelingen rondom het verzenden van het proefschrift belangrijk.

En tenslotte wil ik Herma en de kinderen bedanken voor de relativerende noot en de kritische, taaltechnische opmerkingen over de 3 MMPs of waren het de MMP-3s?

Marcel Posthumus.

155

Curriculum vitae Marcel Posthumus werd op 4 oktober 1961 geboren in Lochem. Na de lagere school en het atheneum werd het doctoraal examen Geneeskunde in 1985 afgelegd aan de Katholieke Universiteit te Nijmegen. Vervolgens werd de overstap gemaakt naar de Rijksuniversiteit in Groningen en volgden de co-schappen in ziekenhuis Ziekenzorg (tegenwoordig opgegaan in het Medisch Spectrum Twente) te Enschede. Na het basisartsexamen in 1987 werd in 1988 gestart met de opleiding tot algemeen internist, de eerste 2 jaar in Leeuwarden (opleider: Marten Leemhuis), gevolgd door 3 jaar in Enschede (opleider: Hein Jordans). In de laatste periode ontstonden de contacten met het vakgebied reumatologie en Hans Rasker als opleider reumatologie te Enschede. In januari 1993 vond de registratie tot internist plaats en in mei 1995 werd deze vervangen door de registratie tot reumatoloog. Per mei 1995 is hij werkzaam als staflid op de afdeling reumatologie in het Academisch ziekenhuis in Groningen.

Stellingen

Behorende bij het proefschrift “The significance of serum matrix metalloproteinase 3in patients with early rheumatoid arthritis”, door Marcel Posthumus.

• De initiële serum MMP-3 spiegel bij patiënten met recent ontstane reumatoïdeartritis is een prognostische factor voor de ontwikkeling van radiologischwaarneembare gewrichtsschade. Dit is vooral belangrijk voor patiënten met bijpresentatie een normale röntgenfoto (dit proefschrift)

• Het systemisch meetbare serum MMP-3 bij patiënten met een recent ontstanereumatoïde artritis is geassocieerd met maten van ziekteactiviteit en maten vangewrichtsbeschadiging en op die manier een directe afspiegeling van het localeproces van ontsteking en schade (dit proefschrift)

• Het MMP-3 promoter polymorfisme heeft voor patiënten met een recent ontstanereumatoïde artritis geen klinische relevantie (dit proefschrift)

• Hoewel het serum MMP-3 een directe afspiegeling lijkt te zijn van het localeontstekingsproces is het tot dusver niet beter gebleken dan het CRP als maat voorziekteactiviteit en radiologisch waarneembare gewrichtsschade bij patiënten metreumatoïde artritis (dit proefschrift)

• Het is de vraag of de beschreven ontkoppeling van het inflammatoire endestructieve proces bij patiënten met reumatoïde artritis niet berust op een onjuisteweergave van het proces door een inadequate marker

• Het tot in detail ontrafelen van de signaaltransductie routes biedt mogelijkhedenvoor verdere differentiatie in medicamenteuze behandelmogelijkheden vanreumatoïde artritis

• Ondanks alle nieuwe kennis, hulpmiddelen en medicamenten blijft de behandelingvan de patiënt met een chronische gewrichtsaandoening individueel maatwerk

• Nieuwe diagnostische hulpmiddelen resulteren helaas vaak in overdiagnostiek

• Een expert is iemand die steeds meer weet over steeds minder

• Het lopen van de “New York Marathon”, het fietsen van “La Marmotte” of hetrijden op een “Moto Guzzi Le Mans 1000” is niets in vergelijking met het afrondenvan een promotie onderzoek.

• Je kunt niet alles hebben…waar zou je het moeten laten

Documents

University of Groningen The significance of serum …1. Choy EH, Panayi GS. Cytokine pathways and joint inflammation in rheumatoid arthritis. N Engl J Med 2001;344(12):907-16. 2. Navarro-Cano