VIEWS: 36 PAGES: 12 POSTED ON: 12/5/2009
Progress in Cell Cycle Research, Vol. 5, 59-70, (2003) (Meijer, L., Jézéquel, A., and Roberge, M., eds.) chapter 7 Rheumatoid arthritis: targeting the proliferative fibroblasts 1 2 3 4 IJsbrand Kramer *, Auragun Wibulswas , Daniel Croft , and Elisabeth Genot 1 Section of Cellular and Molecular Biology, European Institute of Chemistry and Biology, Bat B2, University of Bordeaux 1, 33405 Talence, France. 2 Thammasat University (Rangsit campus), Patum Thani 12121 Thailand. 3 Abramson Family Cancer Research Institute, University of Pennsylvania, Philadelphia, PA 19104-6160, USA. 4 Section of Cellular and Molecular Biology, European Institute of Chemistry and Biology, U441 INSERM , Avenue du Haut Lévêque, 33600 Pessac, France. * To whom correspondence should be addressed Our flexible joints are synovial joints composed of bone, hyaline cartilage, synovial membrane, ligaments and tendons. Rheumatoid arthritis is a disease that affects multiple synovial joints and involves inflammation of the synovial membrane, often resulting in loss of function due to erosion of bone and cartilage. Inflammation is accompanied by an influx of immune-competent cells and by aberrant proliferation of resident fibroblast- like synoviocytes. Accretion of fibroblasts directly contributes to joint destruction, through enhanced production of matrix-degrading enzymes, and indirectly, through excessive release of cytokines that boost the immune system. Targeting the proliferative fibroblast could facilitate regeneration of synovial joints. ARCHITECTURE OF THE SYNOVIAL JOINT renders it rather permeable to solutes. For instance, Our flexible joints are synovial joints. They own large molecules like hyaluronan leak out at a constant, their name to the presence of a visco-elastic solution, slow rate. the synovial fluid, which resembles eggwhite, (syn RHEUMATOID ARTHRITIS ovum). The synovial fluid is composed mainly of the Clinical symptoms and diagnosis glycosaminoglycan hyaluronan, a polymer composed Rheumatoid arthritis is described as a pattern of of repeating β-1,4-glucuronate-β-1,3-N-acetylglucosa- illness, of unknown aetiology, involving chronic mine dis a c c ha r ide s . Hya l ur o na n is the ma j o r synovitis (inflammation of the synovial membrane) of component of the synovial space and it provides multiple joints often resulting in loss of joint function hydraulic resistance (shock absorbing) due to its fluid due to severe erosion of bone, cartilage and ligaments. retaining capacity (1). A precise definition is based on a number of criteria Synovial joints comprise a number of separate tis- laid out by the American College of Rheumatologists sues: bone, hyaline cartilage, synovial membrane, in 1987 (4). Aspects that are considered are: at least ligaments and tendons, together creating a structure three joints with active disease causing morning that moves with a remarkably low level of wear and stiffness, joint swelling, tenderness, radiographic tear (Figure 1). The synovial membrane, the focus of changes (Figure 2), prolonged symptoms, involvement this chapter, plays an important role in the function of of metatarsophalangeal joints and the presence of the synovial joint. It is responsible for the production auto-antibodies such as rheumatoid factor (RF, of hyaluronan and it forms an interface between the representing IgM directed against the Fc-region of IgG blood circulation and the synovial fluid, and it is antibodies) or IgG antifilaggrin auto-antibodies (a implicated in the regulation of immune surveillance of citrullinated antigen) (5, 6). Rheumatoid arthritis may synovial joints. The synovial membrane is made up of be distinguished from other forms of chronic arthritis, a cellular inner layer, the synovial lining, mainly such as osteoarthritis, by the characteristic pattern of composed of macrophages (type-A synoviocytes) and juxta-articular cartilage and bone erosion and the fibroblast-like cells (inner lining cells or type-B occurrence of fever, weight loss, anorexia, severe synoviocytes) (2). The macrophages are identified by muscle atrophy and symmetry of joint involvement their expression of CD68 (macrosialin) and IgG-Fc (reviewed in 7). Tenderness, or at times straight- receptors, and by their capacity to phagocytose. forward pain, is a prominent feature of arthritic joints Fibroblast-like synoviocytes are recognized through and arises from sensory wind up in the spinal cord their constitutive expression of the adhesion molecule mediated b y i n f l a mma t o ry me d i a t o rs s u ch a s VCAM-1 (3) and the fibroblast marker hydroxyproline. prostanoids, bradykinin, protons and cytokines such These cells are responsible for the production of the as IL-1β, IL-15 and TNF-α that sensitize peripheral extracellular matrix components, in particular nociceptors, causing greater activation of spinal hyaluronan. The outer layer of the membrane, the sensory neurones for a given stimulus (8). Pain sub-lining, is mainly composed of a collagenous perception occurs from innervated tissues such as extracellular matrix containing few dendritic cells and subchondral bone, synovial membrane, joint capsule, a vascular bed. The cellular components of the ligaments, and muscle. synovial membrane do not form extensive cellular contacts, as in epithelial or endothelial tissues, and this 59 I. KRAMER, A. WIBULSWAS, D. CROFT AND E. GENOT Figure 1. Synovial joints in health and disease. Synovial joints comprise a number of tissues, bone, car- tilage, capsule and synovial membrane. The joint space is filled with a solution of hyaluronan. Characteristics of synovial joints obtained from patients with rheuma- toid arthritis are hyperplasia of the synovial membrane, as a consequence of prolife- ration of resident fibroblast- like synoviocytes and an influx of immunecompetent cells. Bone and cartilage are covered by pannus. Cartilage and bone show signs of erosion. Figure 2. Progressive destruction of synovial joints. A thinning of cartilage is manifested by a reduction of the distance between the opposing bones. At a later stage erosion of bone and cartilage is visible and the joint has lost its characteristic architecture. normal thinning erosion joint of cartilage of bone and cartilage Figure 3. A model depicting a dissociation between clinical symp- toms, pain, swelling, and erosion of bone (as observed by radiogra- phical analysis). Prognosis; the relationship between clinical symp- sive disease at 2 years are unlikely to show subsequent toms and joint erosion erosive damage but not all patients with erosion at 2 years The disease has erosive and non-erosive variants, subsequently have progressive joint damage. Damage each causing pain and dysfunction, but the latter not and disability are unrelated in early disease (<5 years) and causing visible erosions of bone and cartilage and this relationship only becomes significant after 5 years being self-limiting. After two years of clinical symp- with a strong correlation after 12 years. From this it fol- toms, there is a turning point in the development of lows that early clinical symptoms are primarily influen- erosions visible by radiographic analysis (9). However, ced by disease activity (inflammation), pain and psycho- subsequent progression of joint damage is variable and social factors, and little by structural abnormalities (10). non-predictable. It is true that patients with a non-ero- The discrepancy between disease activity and erosion is 60 CHAPTER 7 / CELL CYCLE AND RHEUMATOID ARTHRITIS also highlighted by the observation that a large number of 19). The presence of elevated levels of inflammatory persons over the age of 50 have some radiographic detec- cytokines enables a number of aspects of inflamma- table erosion without clinical symptoms. It thus tion: the influx of immune competent cells, the phago- appears that at least two pathological processes are cytic function of neutrophils and macrophages, anti- simultaneously at work in the joint, one causing clini- gen presentation by dendritic cells, activation of T- cal symptoms and one being responsible for erosions cells, differentiation of B-cells into plasma cells, pro- (see Figure 3, adapted from 11). Novel criteria to defi- duction of immunoglobulins and, last but not least, ne prognosis in early arthritis are clearly needed to induction of expression of matrix metalloproteinases. enable differentiation between patients who will deve- Abundant evidence has been provided that destruc- lop a progressive erosive disease (and require an tion of cartilage is a consequence of the action of metal- "aggressive" therapy) and those with a self-limiting or loproteinases. These zinc-dependent endopeptidases a non-erosive variant. cleave a large number of extracellular components Changes in the synovial membrane such as collagens and proteoglycans (such as aggregan In rheumatoid arthritis, the synovial membrane in cartilage) (20, 21). Elevated levels of matrix metallo- undergoes a number of changes. Macroscopically, there is proteinase-1 (MMP-1 or interstitial collagenase), a thickening of the membrane and the development of MMP-3 (stromelysin-1) and MMP-9 are detected in the numerous projections at the inner side (villi) (Figure 1). At synovial membrane, the synovial fluid and at the sites the microscopical level a massive increase in the number of cartilage erosion (22). MMPs are produced by fibro- of immune-competent cells and fibroblast-like cells is obs- blast-like synoviocytes, scattered cells in the sublining erved, a phenomenon commonly referred to as hyperpla- region and by synovial endothelial cells (23, 24). Other sia (12). A redistribution of fibroblast-like synoviocytes is proteinases, such as cathepsin and elastase, are also also evident. These cells gradually invade deeper layers of detected. These cysteine or serine proteinases are pro- the synovial membrane and also the bone and cartilage, duced by fibroblast-like synoviocytes, macrophages where they form a cell mass, together with macrophages, and neutrophilic granulocytes and are involved in the that covers these tissues and is therefore named pannus degradation of a wide range of proteins, not exclusive- (from Latin meaning cover or coat). The omnipresence of ly components of the extracellular matrix (25, 26). fibroblast-like synoviocytes in the synovial joint is striking Various knock-out or knock-in mice confirm the and gives the impression of a neoplastic disorder, as if the participation of the above described mediators in cells were indeed transformed. These cells have an increa- murine models of arthritis and three examples are pre- sed proliferative capacity and are highly invasive (revie- sented here. Firstly, transgenic mice expressing human wed in 13). We shall return to this subject in detail in later TNF-α spontaneously develop erosive arthritis (27). sections of this chapter. Secondly, mice lacking MMP-3 (stromelysin-1) fail to The presence of immune-competent cells, like B- and cleave collagen and have limited destruction of pro- T-lymphocytes as well as monocytes, clearly illustrates the teoglycans after the induction of joint inflammation inflammatory component of this disease. B-cells are main- with an emulsion of albumin/Freund's adjuvant and ly present as highly differentiated plasma cells secreting heat-inactivated Bordetella pertussis (antigen-induced auto-antibodies like rheumatoid factor (anti-Fc-IgG IgM) arthritis) (28). Lastly, mice lacking IL-6 resist the deve- or anti-citrullinated flaggrine antibodies. They are found lopment of a chronic response after antigen-induced in the sub-lining of the synovial membrane in the vicinity arthritis (29, 30). The above-mentioned observations of dendritic cells (antigen-presenting cells). Natural killer indicate that rheumatoid arthritis is certainly a conse- cells and mast cells have been observed. Chondrocyte- quence of an elaborated network of cytokines and pro- derived cells are also known to appear in the proliferating teinases. Regardless of the origin of the pathology, the pannus. The appearance of pathological synovial tissue is disease could thus be described as "an inflammatory conspicuous in early rheumatoid arthritis, for instance ele- disorder that fails to resolve itself" (Figure 4) but as we vated levels of VCAM-expressing cells, fibroblast-like will see later this is certainly an oversimplification. It is synoviocytes, and large numbers of macrophages are pre- true that removing or inactivating any one of the com- sent in synovial tissue within one year after diagnosis ponents of this network ameliorates the general condi- (14). Lastly, infiltrating neutrophilic granulocytes are tion, both in animal models and in humans, but the also observed but these reside mainly in the synovial question remains whether or not this cures the disease. fluid. Auto-antigens and auto-immunity Inflammatory cytokines and metalloproteases Immune tolerance to self is not absolute. Given The clinical symptoms are accompanied by eleva- the breadth of protein epitopes in higher eukaryotes, ted levels of pro-inflammatory cytokines such as IL1β, an immune system from which all potential self-reacti- IL-6, IL-15 and TNF-α, and an insufficient presence of vity had been deleted would probably not respond to anti-inflammatory cytokines such as IL-4, IL-10, IL-13, anything. The self-reacting T cells essentially ignore IL-1ra and TGFβ1 (15, 16, 17). All cells of the synovial the self-antigens because they are seen with low affini- membrane contribute to their production, with macro- ty, they occur at low abundance or because they are phages and lymphocytes being dominant in the pro- hidden behind tissue or processing barriers. The self- duction of IL-1β, IL-15 and TNF-β and fibroblast-like tolerance may be lost because of aberrant processing of synoviocytes being the predominant source of IL-6 (18, proteins or because of the homing of infectious orga- 61 I. KRAMER, A. WIBULSWAS, D. CROFT AND E. GENOT Figure 4. Inflammatory mediators present in the diseased joint. In patients with rheumatoid arthritis the balance between pro- and anti-inflammatory mediators is disturbed. An excess of pro- inflammatory mediators creates a chronic inflammatory environ- ment causing clinical symptoms like joint swelling and pain but also a thinning of cartilage. Functional blockade of the pro- inflammatory mediators improves the general condition of the synovial joint. Figure 5. The three phases of an auto-immune response causing rheumatoid arthritis. Stage 1. Auto-antigens, spilled into cellular environment, are presented by antigen-presenting cells in the context of MHC classII. This complex is recognized by the T-cell receptor (TCR) causing activation of naïve T-cells and resulting in clonal expansion of T- helper cells. These play a role in the subsequent activation of macrophages but more importantly of B-cells, which become anti-self antibody- producing plasma cells. Stage 3. Antibody/antigen complexes accumulate and activate complement with the generation of the proteolytic fragments, C3bi and C5a. The immune complexes bind to macrophages, together with C3bi, and initiate phagocytosis causing the production of toxic oxygen metabolites and the secretion of metalloproteinases and inflammatory cytokines. The complement factors C3a and C5a cause activation of mast cells, with release of histamine, and recruitment of phagocytes (through their chemokinetic action). Why synovial joints are favourable sites for accumulation of immune complexes and why macrophages react so vigorously remains to be resolved. nisms, viral, bacterial or mycoplasmale, into certain tis- T-cells. Activated T-cells, in turn, communicate with sues. The local production of inflammatory cytokines macrophages (TH1) to boost production of inflammato- and the ensuing cellular damage, cause release of cyto- ry cytokines and, importantly, with B-lymphocytes solic components and will wake up self-reactive T- (TH2) to boost the production of antigen-specific lymphocytes and confront them with their antigen but immunoglobulins (vital for the rapid elimination of this time in an effective way (through presentation in "foreign"). The antibody/antigen complexes that are an MHC classII context by antigen-presenting cells formed, often coupled to components of the comple- together with lots of co-stimuli) (31) (Figure 5). In the ment system (C3bi), are deleterious because they inten- case of arthritis, not only infection by a pathogen, but sify the inflammatory response by stimulating macro- also repeated mechanical insults may create a favoura- phages. In the case of the synovial joint, they do this by ble environment for the activation of normally tolerant 62 CHAPTER 7 / CELL CYCLE AND RHEUMATOID ARTHRITIS binding to the IgG-Fc type IIIa receptor (FcRIIIA or but not from those with osteoarthritis, spondylarthro- CD16a) and the C3bi receptor (CD11b or mac-1). pathy or other diagnoses (38, 39). Anti-citrullinated Macrophages only carry FcRIIIA in the tissues normal- filaggrin antibody-producing plasma cells are also ly affected by rheumatoid arthritis, including joint detected in rheumatoid synovial tissue and these fin- lining, the lungs and the pericardial lining of the heart, dings have raised the hypothesis that citrullinated pro- so inflammation only occurs in these tissues. teins could trigger an autoimmune response in rheu- Activation leads to phagocytosis of the complexes with matoid arthritis. Post-translational modifications of the subsequent release of proteinases (members of the proteins have been reported to be the cause of other elastase and cathepsin family), chemokines (32) and autoimmune diseases, mainly in systemic lupus ery- inflammatory cytokines, such as IL-1α and TNF-α thrematosus and scleroderma. (33). It also causes activation of the respiratory burst The autoimmune model has gained renewed with release of toxic oxygen metabolites (34) (for a attention with the fortuitous generation in mice of an review on this subject, 35). The activation of comple- illness resembling rheumatoid arthritis, obtained by ment not only causes the production of C3bi, it also crossing a T cell-receptor transgenic line (KRN) with results in the cleavage of C5 into C5a and C5b. C5a, an the auto-immune disease sensitive-NOD strain. After anaphylatoxin and chemotaxin, localizes elements of two weeks, all offspring (K/BxN) developed joint dis- the inflammatory response (through chemotaxis), ease (40). At that time point T- and B-cells infiltrated degranulates neutrophils (thereby causing release of into the synovial joints and the mice started to present free oxygen radicals and prostaglandins), and also sti- signs of inflammation. The reason why these mice mulates Il-1 and Il-6 production. In contrast, C5b teams developed arthritis was recognition of self. The auto- up with C6, C7, C8 and C9 to form a pore-forming antigen is glucose-6-phosphate isomerase (GPI), and it structure that causes leakage through the plasma is recognized, in the context of MHC classII, by the membrane followed by cell death (the lytic pathway, (monoclonal) T-cell receptor. GPI is a glycolytic enzy- an important step in protection against micro-orga- me and not at all a joint-specific antigen (41). Despite nisms). GPI being a ubiquitously expressed antigen, anti-GPI The generation of C5a and the release of inflam- immune complexes, combined with complement pro- matory cytokines will cause further recruitment of ducts, appeared to accumulate at the cartilage surface immune-competent cells thereby generating an inflam- of synovial joints of the mice (and caused the delete- matory response that lasts until the antibody-antigen rious inflammatory response in the manner illustrated complexes are eliminated. In the case of auto-immuni- in the previous paragraphs) (42). ty, this point will never be reached and a chronic Arthritis could also be provoked by transfer of inflammatory condition will ensue. anti-GPI immunoglobulins, obtained from diseased A number of auto-antibodies have been found in mice, into healthy lymphocyte-deficient recipients. patients with rheumatoid arhritis, none of which are This finding provided ample evidence for the role of directed against joint-specific epitopes. As already auto-antibodies and innate immune players in the mentioned, one of them is rheumatoid factor. This development of the disease. Finally, and importantly, represents mainly IgM antibodies directed against the in the immunoglobulin transfer experiments, not all Fc portion of IgG. Rheumatoid factor is not highly spe- recipient mice developed arthritis, and a search for cific for rheumatoid arthritis nor is it a very useful pro- genetic influences on susceptiblity highlighted one gnostic factor for the development of the erosive genomic region that contained the complement factor variant. Moreover, it can be found in other autoimmu- C5 gene (43). Interestingly, this gene has also been ne diseases, malignancies, chronic infections and even identified as a susceptibility locus for murine experi- healthy persons (36). Other auto-antigens that have mental allergic asthma (44a) and myasthenia gravis been detected are directed against perinuclear factor (destruction of motor endplate AchR) (44b). From this (APF), keratin (AKA), collagen, Epstein-Barr nuclear information it follows that rheumatoid arthritis could antigen and proteins such as RA33, Sa and p68. be interpreted as an auto-immune disease, causing the Antibodies against APF and AKA are of great interest. formation of immune complexes, which in turn activa- These are in fact directed against a polypeptide te complement with the production of complement sequence containing a modified arginine residue, cal- factors C3bi, C5a etc, resulting in a runaway activation led citrulline. This epitope is present in filaggrin, abun- of the innate immune response. This then results in dant in keratinocytes, but is also present in other pro- joint destruction. According to this model, therapy teins of unknown identity. The conversion of arginine should focus on prevention of the C5a product being to citrulline in proteins is an enzymatic process that is formed. Quite independently from these findings, carried out by peptidylarginine deiminase (PAD), an beneficial effects of anti-C5a monoclonal antibodies on enzyme that appears to be hormonally controlled (37). collagen-induced arthritis in mice has already been Through altered post-translational processing, a modi- demonstrated (45). fied putative auto-antigen emerges which bypasses CURRENT TREATMENT OF ARTHRITIS immune tolerance. The presence of citrullinated pro- teins has recently been detected in a small number of Based on the knowledge presented in the pre- synovial cells from patients with rheumatoid arthritis vious sections, therapy has focussed on the attenuation 63 I. KRAMER, A. WIBULSWAS, D. CROFT AND E. GENOT T-cell/B-cell expansion - nucleotide metabolism inhibitors (methotrexate) - disease-modifying anti-rheumatic drugs (DMARDS) (gold salts, D-penicillamine, sulfasalazine) - immunosuppressors (cyclosporin) Inflammation - disease-modifying anti-rheumatic drugs - non-steroidal anti-inflammatory drugs (NSAIDS) (aspirin, indomethacin or aceclofenac) - cytokine inhibitors (TNF-α antibody (Infliximab), TNF-α soluble receptor (Etanercept) and IL-1β receptor antagonist (Anakinra) - glucocorticoids (prednisolone) Table 1. Therapeutic strategies of the immune response and more recently on the inac- the above-mentioned treatments, surgery can have a tivation of particular inflammatory cytokines. profoundly positive impact on their ability to carry out Conventional therapy includes the so-called disease- their daily activities . The usual indications for surgery modifying anti-rheumatic drugs (DMARDS) like gold include pain and/or disability that are unacceptable salts, D-penicillamine or sulfasalazine, the non-steroi- and functionally limiting. Surgery of hand and wrist, dal anti-inflammatory drugs (NSAIDS) like aspirin, shoulder, elbow, knee or total hip replacement are all indomethacin or aceclofenac, and the nucleotide meta- common treatments. In some case surgical removal of bolite inhibitors such as methotrexate (Table 1). The the synovial membrane, synovectomy, suffices. treatment rapidly suppresses inflammation and causes THE TRANSFORMED FIBROBLAST-LIKE a reduction in cytokine expression and mononuclear SYNOVIOCYTE cell infiltration. This effect could be explained by the observed reduction in the production of TNF-α and The previous sections were intended not only to lower levels of expression of the endothelial adhesion give the reader information about rheumatoid arthritis molecules, E-selectin and ICAM-1, instrumental in the but also to provide the necessary context in which to recruitment of immune-competent cells (46). place the final section dealing with cell proliferation Combination therapies are also applied, comprising and disease. In this section we add another level of sulfasalazine, methotrexate and the immune-suppres- complexity, the contribution of the fibroblast-like sing glucocorticoids such as prednisolone (over a per- synoviocytes to the destruction of synovial joints, inde- iod of 28 to 40 weeks). Patients treated with such com- pendent of T- or B-cells, auto-antigens or complement. bination therapy show less radiological detectable As mentioned in the first section of this chapter, the damage at 4 years compared with those treated with onset of rheumatoid arthritis is characterized by an "monotherapy". Novel therapies that have been tested increasing number of fibroblast-like synoviocytes and extensively over the last three years, are administra- a redistribution of these cells throughout the synovial tion of antibodies, soluble receptors or receptor anta- joint. The origin of these cells is, however, not clear. At gonists that block the action of TNF-α (Infliximab or best they can be described as a heterogeneous popula- Etanercept) or IL-1β (Anakinra) combined with metho- tion comprising proliferating specialized joint-lining trexate (47). Using these particular therapies, arrest of cells, common interstitial fibroblasts and recruited pri- joint damage was observed, even in patients who did mitive mesenchymal cells (49). not respond clinically (and retained symptoms of swel- FIBROBLAST-LIKE SYNOVIOCYTES AS ling, tenderness etc). In general, infiltration of T-cells AGGRESSORS (CD3+), B-cells (CD22+) and macrophages (CD68+) was greatly diminished, as was the expression of adhe- These fibroblast-like synoviocytes, when derived sion molecules on fibroblast-like synoviocytes and vas- from the joints of patients with rheumatoid arthritis, cular endothelial cells. (48). At present this treatment is are often characterized as aggressive or transformed reserved for patients who do not respond well to cells. This term should not be interpreted as these cells conventional therapy. Thus far, limited adverse effects being malignant because they are not. For instance, have been recorded with antibody treatments, when kept in culture they reach senescence and exp- amongst which are opportunistic infections, tuberculo- ress cell cycle inhibitors that are related to fibroblast sis (awakening of a latent infection), multiple sclerosis senescence (in particular p16INK4a) (50). Moreover, or intestinal perforation. Immediate re-appearance of there is little evidence for rheumatoid arthritis-specific clinical symptoms follows the arrest of treatment. gene mutations (with the exception of p53, see below) or gene rearrangements, nor is there evidence for Finally, in those patients who develop joint dama- microsatellite instability, commonly detected in true ge with deformity and functional limitation, despite malignant cells. It is also not clear whether this "trans- 64 CHAPTER 7 / CELL CYCLE AND RHEUMATOID ARTHRITIS Figure 6. "Transformation" of fibroblast-like synoviocytes. The inflammatory environment brings about proliferation of resident fibro- blast-like synoviocytes (formation of granula- tion tissue) through (over)expression of genes implicated in cellular proliferation and de-dif- ferentiation. Prolonged exposure may cause permanent changes, like mutations in the p53 gene, expression of splice variants of CD44. The cells also acquire the unique capacity to express p16INKa but this remains suppressed in the synovial environment. The result is the formation of a proliferative and invasive fibroblast-like synoviocytes that contributes to joint destruction, either directly, through release of their matrix metalloproteases, or indirectly, through release of inflammatory cytokines that boost the immune system. formed" phenotype is an exclusive synovial joint-phe- med" fibroblast-like synoviocytes are an elevated level nomenon or a characteristic of any granulation tissue of expression of the oncogenes c-myc , c-fos and c-jun that forms during injury or chronic inflammation (and (59, 60) a capacity to form colonies in soft agar (ancho- needed for the reconstruction of extracellular matrix). rage-independent growth) (61), exorbitant release of However, numerous studies indicate that the fibro- IL-6 in the absence of detectable levels of IL-1β or TNF- blast-like synoviocytes derived from rheumatoid α (62) and an increased production of matrix metallo- arthritic joints have undergone a series of permanent proteases like MMP-1 and MMP-3 (63). Furthermore changes and are clearly distinguishable from fibro- these cells express elevated levels of Frizzled (Fz5) and blasts derived from non-diseased joints, from osteo- its ligand Wnt-5A (64) as well as Notch-1 (65), HOXD9 arthritic joints or from other tissues (skin or lung fibro- (66) and the EGF receptor ErbB-2 (67). The presence of blasts) (Figure 6). Different experimental approaches Notch-1, Wnt, Fz and HOXD9 is interpreted as a sign have demonstrated an increased percentage of fibro- of cellular de-differentiation because these are genes blast-like cells in S-phase in the rheumatoid joint (51, expressed throughout embryological development 52 , 53, 54). The proliferative advantage persists when and that play an important role in tissue development. whole tissue specimens or dissociated synovial cells The "transformed" cells are also characterized by are brought into culture (55, 56). Our studies have expression of CD44-splice variants, in different spli- shown that VCAM-1-expressing-cells derived from cing combinations (68). Fibroblast-like synoviocytes rheumatoid arthritic joints can be either fast or slow (identified as the VCAM-1-positive fibroblast popula- dividing. Fast dividers have a doubling time of 55 tion) expressing variants containing exons CD44v7/8 hours whereas slow dividers double every 69 hours were shown to have a proliferative advantage over (57). These estimations have been made throughout those that do not express them (57). These findings the second week in culture in the presence of 10% fetal confirm the previous mentioned observation that cell bovine serum. Other studies have shown that, under mass in pannus can arise from a monoclonal origin and similar conditions, cells from highly inflamed joints that clonal expansion is an important cause of increa- double every 42 hours whereas those from osteo- sed cellularity of the synovial membrane. Aberrant arthritic joints take 72 hours (55). Proliferation in the expression of CD44-splice variants is often associated synovial joint may be a consequence of the presence of with true cell transformation (69). Lastly, in fibroblast- high levels of inflammatory cytokines but this does not like synoviocytes of some patients, overexpression or explain the difference in proliferation seen under cul- mutation of the tumour-suppressor p53 have been obs- ture conditions. After all, under culture conditions the erved (70, 71). Good evidence has been provided that cells all experience a highly inflammatory environ- dominant negative mutations of p53 increased their ment because of the presence of serum (the soluble invasive capacity, their resistance to apoptosis and component of clotted blood). Evidence has also been their production and release of cytokines and metallo- provided that synoviocytes obtained from pannus proteinase (72, 73, 74). What perhaps is lacking is a clear lesions exhibit a monoclonal pattern of restriction frag- demonstration that fibroblast-like synoviocytes contai- ment length polymorphism (RFLP) of the phosphogly- ning the p53 mutations colonize the synovial joint and cerate kinase 1 (PGK-1) gene, suggesting that this cell concentrate at the interface between pannus and cartilage mass originates from locally proliferating cells. The (because of their survival advantage plus invasive quali- pannus cells had high levels of expression of TGF-β ties). In addition, no reports have shown that these cells and PDGF (58). Other characteristics of the "transfor- dominate cell culture if kept for some weeks. It should be 65 I. KRAMER, A. WIBULSWAS, D. CROFT AND E. GENOT noted that numerous studies have not observed muta- viocytes in different forms of joint athropathies (85-87). tions in patients with rheumatoid arthritis (75, 76). Inhibition of proliferation reduces joint destruction Cellular changes persist ex vivo If transformation of the fibroblast-like synoviocy- When derived from rheumatoid arthritic joints tes contributes to joint destruction, inhibition of their and brought into culture, the fibroblast-like synoviocy- proliferative or invasive capacities should provide pro- tes are highly invasive, whereas those derived from tection against joint destruction. In studying expres- non-diseased joints are devoid of such activity. Cells sion of cyclin-dependent kinase inhibitors (CDI's) in from osteo-arthritic tissue have rather variable levels fibroblast-like synoviocytes, it was found that during of invasion but always well below those observed in serum starvation, cells derived from rheumatoid patients with rheumatoid arthritis (77). Invasion did arthritic joints had the unique quality of expressing not require high concentrations of growth factors, indi- p16INK4a, in addition to the expression of p21CIP1 and cating an intrinsic quality of the cells rather than sti- p27KIP1 (88). Expression of p16INK4a could not be detec- mulation by environmental factors, and was compara- ted in fibroblast-like synoviocytes in the synovial ble with v-fos-transfected rat fibroblast cells (a cell line membrane. Because this inhibitor is implicated in the used to study the mechanisms of cellular invasion of induction of a differentiation programme resulting in extracellular matrix) (78). In these studies it was shown senescence of human fibroblasts, one could conclude that entry of cells into Matrigel could be inhibited up that the synovial joint environment represses fibro- to 60% by prior treatment of the fibroblast-like syno- blast differentiation (89). Under similar experimental viocytes with antibodies recognizing the variant exons conditions, cells obtained from other tissues or from CD44v3 or CD44v6. The aberrant expression of CD44- other types of athropathies only expressed p21Cip1 and splice variants thus has important functional conse- p27KIP1. After serum starvation, fibroblast-like synovio- quences and could play a role in the altered phenotype cytes derived from rheumatoid arthritic joints, again of the fibroblast-like synoviocytes. Importantly, inva- unlike other fibroblasts, were unable to reactivate the sion and destruction of cartilage can also occur in a cell cycle machinery when given medium with a high more physiological model comprising synovial tissue serum content. Ectopic expression of p16INK4a in fibro- or isolated fibroblast-like synoviocytes co-implanted blast-like synoviocytes, using an adenovirus infection with human cartilage under the skin of severe combi- technique, abrogated the proliferative response nor- ned immunodeficient (SCID) mice (lacking T- and B- mally induced by the inflammatory cytokines TNF-α lymphocytes). Cells obtained from non-diseased tissue and IL-1β. To test whether this irreversible inhibition do not cause erosion (79, 80). Various elements play a could be of therapeutic advantage, similar experiments role in the damage of cartilage: elevated levels of were performed in rats, after intra-articular application cathepsins L and B as well as MMP-1 and MMP-13 (81) of the virus containing p16INK4a. The animals showed and activation of plasminogen (82). The interplay bet- reduced symptoms of arthritis upon treatment with ween fibroblast-like synoviocytes and chondrocytes of mycobacterium butyricum-containing adjuvant and his- the cartilage is also of importance because inhibition of tological analysis showed reduced mononuclear cell chondrocyte protein synthesis reduced the invasive infiltration, less pannus formation and less thinning of capacity of the synoviocytes (83). These studies show cartilage. Interestingly, significant differences between unmistakably that the transformed phenotype persists p16INK4a-expressing and non-expressing animals emer- even under conditions where inflammatory cytokines ged only after two weeks of the adjuvant treatment. or auto-antigens are absent, and that fibroblast-like This finding supports the notion that a synergy may synoviocytes cause cartilage erosion independent of an exist between the adjuvant-induced deleterious immu- intact immune system. It therefore has been proposed ne response (early symptoms) and the ensuing trans- that, in parallel to a cytokine network, an oncogene formation of the fibroblast-like synoviocytes (amplifi- network acts as a pathogenic mechanism in synovial cation of damage in the long run). Similar experiments joint-destruction (84). More studies are required to pre- were performed, this time with the ectopic expression cisely define these molecular alterations and to deter- of p21CIP1 or p16INK4a, in a mouse model where arthritis mine how they relate to disease development. For is induced by immunisation against collagen. Both instance, among patients showing true symptoms of cyclin-dependent inhibitors proved to be beneficial. rheumatoid arthritis for two years, some never deve- They not only prevented synovial overgrowth but also lop full-blown joint destruction. This indicates that if improved the inflammatory environment by keeping irreversible changes in the fibroblast-like synoviocytes low the levels of IL-1ss, IL-6 and TNF-β (90). do occur, they must occur, or not, at that particular Induction of fibroblast-like synoviocyte differen- stage. If markers for such permanent changes were tiation has also been obtained by ectopic expression of found, they could prove useful in the prognosis of the Jun-D, a transcription factor known to counteract the development of an erosive or non-erosive variant of effects of Ras-mediated cell transformation or c-Jun- the disease. A number of laboratories have taken up and jun B-mediated fibroblast proliferation (91). As the genomic, transcriptomic or proteomic approach to with ectopic expression of the cyclin-dependent kinase characterize further alterations in fibroblast-like syno- inhibitors, induction of differentiation by Jun-D reduced 66 CHAPTER 7 / CELL CYCLE AND RHEUMATOID ARTHRITIS inflammatory cytokine expression but also diminished hand-in-hand is the targeting of cyclo-oxygenase COX- the content of MMPs, through a down-modulation of 2. This enzyme is the inducible variant of cyclo-oxyge- AP-1 activity. Beneficial effects of Jun-D expression in nase, is highly expressed in inflamed tissues, and murine models of arthritis have not yet been reported. converts arachidonic acid into prostaglandins of which The aberrant expression of CD44-splice variants PGD2, PGE1 and PGE2 are relatively stable end pro- provides proliferative and invasive advantages to the ducts. Of these, PGE2 has been linked to inflammation. fibroblast-like synoviocytes. Aberrant expression of Mice lacking COX-2, but not the mice lacking COX-1, adhesion molecules may be one way in which trans- have a reduced incidence and severity of collagen- formed cells obtain additional signals supporting cell induced arthritis, with a reduction in both prolifera- proliferation (92). In our studies, we have shown that tion of synovial lining and infiltration of immune-com- addition of anti-CD44v7/8 antibodies to fibroblast-like petent cells (99). Transfer of antibodies obtained from synoviocytes in culture, causes expression of the the diseased mice ("arthritogenic" auto-antibodies) cyclin-dependent inhibitor p21CIP1 and of the growth could not induce arthritis in the naive COX-2 -/- mice. arrest and DNA-damage-induced proteins GADD45 These findings could be interpreted as COX-2 being an and GADD153 concomitant with an inhibition of cell important regulatory enzyme for the execution of both proliferation. No signs of apoptosis were detected in an acquired and an innate immune response. the first 5 days of treatment (57). From these studies we However, it may be more than that. A recent study has have learned that adhesion molecules also can be tar- shown that PGE2 is also a potent transactivator of the gets to restrain excessive cell proliferation. The results EGF receptor (100) and this mechanism plays a vital role presented above, provide ideas or suggestions, of how in the promotion of gastro-intestinal hypertrophy (cau- we could possibly target proliferation of fibroblast-like sing polyps). An additional interpretation could thus be synoviocytes. The hard work, translating them into a that the absence of COX-2 prevents proliferation of fibro- therapy for rheumatoid arthritis, still lies ahead. blast-like synoviocytes, thereby providing extra protec- tion against these potentially aggressive cells. Induction of apoptosis would be another way to reduce or eliminate the excess of fibroblast-like syno- A PHARMACOLOGICAL INTERVENTION viocytes. This could be particularly advantageous THAT TARGETS CELL PROLIFERATION when the synovial joints are already populated by AND REDUCES INFLAMMATION large numbers of cells. It is known that transformation Rheumatoid arthritis can be described as a disea- renders cells more susceptible to apoptosis, and trans- se with at least three pathogenic processes; 1) forma- formed cells have often put into place additional res- tion of auto-antibodies/complement complexes, 2) cue pathways (93). These rescue pathways provide excessive production of inflammatory cytokines and 3) potential pharmacological targets as their blockade "transformation" of fibroblast-like synoviocytes. could impose apoptosis. There is evidence that the pro- Attempts to interfere with any of these processes have tein kinase PKB (Akt) provides such a rescue signal, all proven to be beneficial in murine models of rheu- and that its activation is elevated in cells derived from matoid arthritis. In the case of the inhibitors of TNF-α rheumatoid arthritic joints (compared with those or IL-1β, benefits have also been demonstrated in obtained from patients with osteoarthritis) (94). humans suffering from the real disease. Drawbacks of Fibroblast-like synoviocytes express Fas, a cell death the current therapies are either limited protection receptor, and apoptotic cells are apparent in the syno- against joint destruction (DMARDs and glucocorti- vial membrane of patients with rheumatoid arthritis coids) or excessive costs (antibody treatment). The stu- (95). When Fas binds to its ligand FasL, receptor clus- dies presented in this chapter clearly provide a ratio- ters are formed and these induce a death signal through nale for the development of a pharmacological inter- activation of intracellular proteinases called caspases. In vention that targets the cell cycle of fibroblast-like an attempt to test if Fas expression could be exploited to synoviocytes (Table 2). Possible means to restrain the kill fibroblast-like synoviocytes, irradiated Fas-ligand- fibroblast-like synoviocytes are induction of expres- expressing cells (introduced through transfection) were sion of p16INK4a or p21CIP1, abrogation of cell-matrix combined with fibroblast-like synoviocytes and co- interactions or interference with the transactivation- implanted with cartilage under the skin of SCID mice. effect of PGE2 (COX-2-dependent). These approaches This treatment effectively eliminated fibroblast-like syno- might not only reduce the direct damaging effect of viocytes through cell-cell interaction (96). these cells on bone and cartilage, but also dampen the One example where inhibition of cell proliferation inflammatory response. We are of the opinion that one and attenuation of the immune response might go cannot ignore the accretion of fibroblast-like synovio- Inhibiting of the cell cycle - Induction of expression of senescence genes (p16INK4a) - induction of expression of cell cycle inhibitors (p21CIP1, GADDs) - blockade of proliferation promoting signals through specific adhesion molecules (CD44-splice variants, integrins) Sensitising the cells to apoptosis - inhibition of apoptosis rescue pathhways (PKB) Table 2. Targeting the fibroblast-like synoviocytes 67 I. KRAMER, A. WIBULSWAS, D. CROFT AND E. GENOT cytes in the diseased joint for the simple reason that 20. Shapiro, S.D. (1998) Curr Opin Cell Biol 10, 602-608. their exorbitant production of IL-6 makes the rheuma- 21. van Meurs, J., van Lent, P., Holthuysen, A., toid arthritic joint an impossible environment in which Lambrou, D., Bayne, E., Singer, I., and van den to control the immune system. Under these conditions, Berg, W. (1999) J Immunol 163, 5633-5639. any attempt to reverse inflammation or abrogate the 22. Cunnane, G., FirtzGerald, O., Beeton, C., Cawston, T.E., formation of auto-immune complexes resembles and Breshinan, B. (2002) Arthritis Rheum 44, 2263-2274. "mopping the floor with the water tap wide open" (Dutch proverb) (18, 97, 98). 23. McCachren, S.S. (1991) Arthritis Rheum 34, 1085-1093. 24. Gravallese, E.M., Darling, J.M., Ladd, A.L., Katz, J.N., ACKNOWLEDGEMENT and Glimcher, L.H. (1991) Arthritis Rheum 34, 1076-1084. The authors would like to thank Paul-Peter Tak, 25. Adkison, A.M., Raptis, S.Z., Kelley, D.G., and Academic Medical Center Amsterdam, The Netherlands, Pham, C.T. (2002) J Clin Invest 109, 363-371. for his comments on this manuscript. 26. Hou, W.S., Li, W., Keyszer, G., Weber, E., Levy, R., REFERENCES Klein, M.J., Gravallese, E.M., Goldring, S.R., and 1. Scott, D., Coleman, P.J., Abiona, A., Ashhurst, Bromme, D. (2002) Arthritis Rheum 46, 663-674. D.E., Mason, R.M., and Levick, J.R. (1998) J Physiol 27. Keffer, J., Probert, L., Cazlaris, H., Georgopoulos, 511 (Pt 2), 629-643. S., Kaslaris, E., Kioussis, D., and Kollias, G. (1991) 2. Henderson and Ewards (1987). The synovial lining EMBO J 10, 4025-4031. in health and disease. Chapman and Hall, London. 28. Meurs, J., van Lent, P., Stoop, R., Holthuysen, A., 3. Morales-Ducret, J., Wayner, E., Elices, M.J., Singer, I., Bayne, E., Mudgett, J., Poole, R., Alvaro-Gracia, J.M., Zvaifler, N.J., and Firestein, Billinghurst, C., van der Kraan, P., Buma, P., and van G.S. (1992) J Immunol 149, 1424-1431 den Berg, W. (1999) Arthritis Rheum 42, 2074-2084. 4. Arnett, F.C., Edworthy, S.M., Bloch, D.A., 29. Alonzi, T., Fattori, E., Lazzaro, D., Costa, P., McShane, D.J., Fries, J.F., and Cooper, N.S. et al. Probert, L., Kollias, G., De Benedetti, F., Poli, V., (1988) Arthritis Rheum 31, 315-324. and Ciliberto, G.I. (1998) J Exp Med 187, 461-468. 5. Scott, D.L. (2002) Arthritis Rheum 46, 286-290. 30. de Hooge, A.S., van De Loo, F.A., Arntz, O.J., and van Den Berg, W.B. (2000) Am J Pathol 157, 2081-2091. 6. van Boekel, M.A., Vossenaar, E.R., van den Hoogen, F.H., and van Venrooij, W.J. (2002) Arthritis Res 4, 87-93. 31. Horwitz, M.S., Bradley, L.M., Harbertson, J., Krahl, T., Lee J., and Sarvetnick, N. (1998) Nat Med 4, 781-785. 7. Tak, P.P., and Bresnihan, B. (2000) Arthritis Rheum. 43, 2619-2633. 32. Chantry, D., DeMaggio, A.J., Brammer, H., Raport, C.J., Wood, C.L., Schweickart, V.L., Epp, A., Smith, 8. Dickenson, A.H. (1990) Trends Pharmacol Sci. 11, 307-309. A., Stine, J.T., Walton, K., Tjoelker, L., Godiska, R., 9. Wofle, F.A. (2000) Arthritis Rheum 43, 2751-2761. and Gray, P.W. (1998) J Leukoc Biol 64, 49-54. 10. van der Heijde, D.M. (1995) Br J Rheumatol 34 33. Brown, E.J. (1995) Bioessays 17, 109-117. Suppl 2, 74-78. 34. Segal, A.W., and Shatwell, K.P. (1997) Ann N Y 11. Kirwan (1999) J. Rheum 26, 720-725. Acad Sci 832, 215-222. 12. Tak, P.P., Smeets, T.J., Daha, M.R., Kluin, P.M., 35. Arend, W.P. (2001) Arthritis Rheum 44, 2224-2234. Meijers, K.A., Brand, R., Meinders, A.E., and 36. Williams, D.G. (1994) Autoimmunity in rheumatoid Breedveld, F.C. (1997)Arthritis Rheum. 40, 217-225. arthritis In: Klippel, J.H., and Dieppe, P.A. eds. 13. Zvaifler, N.J., and Firestein, G.S. (1994) Arthritis Rheumatology, London: Mosby, 9.1-9.14. Rheum 37, 783-789. 37. van Venrooij, W.J., and Pruijn, G.J. (2000) Arthritis 14. Tak, P.P., Thurkow, E.W., Daha, M.R., Kluin, P.M., Res 2, 249-225. Smeets, T.J., Meinders, A.E., and Breedveld, F.C. 38. Baeten, D., Peene, I., Union, A., Meheus, L., (1995) Clin Immunol Immunopathol 77, 236-242. Sebbag, M., Serre, G., Veys, E.M., and de Keyser, 15. Feldmann, M., Brennan, F.M., Chantry, D., F. (2001) Arthritis and Rheum 44, 2255-2262. Haworth, C., Turner, M., Abney, E., Buchan, G., 39. Smolen, J.S., and Steiner, G., (2001) Arthritis and Barrett, K., Barkley, D., Chu, A., Field, M., and Rheum 44, 2218-2220. Maini, R.N. (1990). Ann Rheum Dis 49, 480-486. 40. Kouskoff, V., Korganow, A.S., Duchatelle, V., 16. Arend, W.P., and Dayer, J.M., (1990). Arthritis Degott, C., and Benoist, C., (1996) Cell 87, 811-822. Rheum 33, 305-315. 41. Matsumoto, I., Staub, A., Benoist, C., and Mathis, 17. Arend, W.P., and Dayer, J.M., (1995). Arthritis D. (1999) Science 286, 1732-1735. Rheum 38, 151-160. 42. Matsumoto, I., Maccioni, M., Lee, D.M., Maurice, 18. Chen, V., Croft, D., Purkis, P., and Kramer, I.M. M., Simmons, B, Brenner, M., Mathis, D., and (1998) Br J Rheumatol 37, 148-156. Benoist, C. (2002) Nat Immunol. 3, 360-365. 19. Thurkow, E.W., van der Heijden, I.M., Breedveld, 43. Ji, H., Gauguier, D., Ohmura, K., Gonzalez, A., F.C., Smeets, T.J., Daha, M.R., Kluin, P.M., Meinders, Duchatelle, V., Danoy, P., Garchon, H.J., Degott, A.E., and Tak, P.P. (1997) J Pathol 181, 444-450. C., Lathrop, M., Benoist, C., and Mathis, D. (2001) 68 CHAPTER 7 / CELL CYCLE AND RHEUMATOID ARTHRITIS J Exp Med. 194, 321-330. Pierce, R.A., Welgus, H.G., and Dayer, J.M. (1998) 44a. Karp, C.L., Grupe, A., Schadt, E., Ewart, S.L., Arthritis Rheum 41, 1748-1759. Keane-Moore, M., Cuomo, P.J., Kohl, J., Wahl, L., 64. Sen, M., Lauterbach, K., El-Gabalawy, H., Kuperman, D., Germer, S., Aud, D., Peltz, G., and Firestein, G.S., Corr, M., and Carson, D.A. (2000) Wills-Karp, M. (2000) Nat Immunol 1, 221-226. Proc Natl Acad Sci USA 97, 2791-2796. 44b. Christadoss, P. (1988) J Immunol 140, 2589-2592. 65. Nakazawa, M., Ishii H., Aono, H., Takai, M., 45. Wang, Y., Rollins, S.A., Madri, J.A., and Matis, Honda, T., Aratani, S., Fukamizu, A., Nakamura, L.A. (1995) Proc Natl Acad Sci USA 92, 8955-8959. H., Yoshino S., Kobata, T., Nishioka, K., Nakajima, T. (2001) Arthritis Rheum 44, 1545-1554. 46. Walters, M.T., Smith, J.L., Moore, K., Evans, P.R., and Cawley, M.I.D. (1987) Ann Rheum Dis 44, 7-16. 66. Khoa, N.D., Nakazawa, M., Hasunuma, T., Nakajima, T., Nakamura, H., Kobata, T., 47. Maini, R., St Clair, E.W., Breedveld, F., Furst, D., Nishioka, K. (2001) Arthritis Rheum 44, 1013-1021. Kalden, J., Weisman, M., Smolen, J., Emery, P., Harriman, G., Feldmann, M., and Lipsky P. (1999) 67. Satoh K., Kikuchi S., Sekimata M., Kabuyama Y., Lancet 354, 1932-1939. Homma M.K., Homma Y. (2001) Arthritis Rheum 44, 260-265. 48. Tak, P.P., Taylor, P.C., Breedveld, F.C., Smeets, T.J., Daha, M.R., Kluin, P.M., Meinders, A.E., and 68. Croft, D.R., Dall, P., Davies, D., Jackson, D.G., Maini, R.N. (1996) Arthritis Rheum. 39, 1077-1081. McIntyre, P., and Kramer, I.M. (1997) Eur J Immunol 27, 1680-1684. 49. Marinova-Mutafchieva, L., Williams, R.O., Funa, K., Maini, R.N., and Zvaifler, N.J. (2002) Arthritis Rheum 46, 69. Herrlich, P., Morrison, H., Sleeman, J., Orian- 507-513. Rousseau, V., Konig, H., Weg-Remers, S., and Ponta, H. (2000) Ann N Y Acad Sci 910, 106-118; 50. Taniguchi, K., Kohsaka, H., Inoue, N., Terada, Y., Ito, H., discussion 118-120. Hirokawa, K., and Miyasaka, N. (1999) Nat Med 5, 760-767. 70. Tak, P.P., Smeets, T.J., Boyle, D.L., Kraan, M.C., Shi Y, 51. Lalor, P.A., Mapp, P.I., Hall, P.A., and Revell, P.A. Zhuang, S., Zvaifler, N.J., Breedveld, F.C., and (1987) Rheumatol Int. 7, 183-186. Firestein, G.S. (1999) Arthritis Rheum 42, 948-953. 52. Konttinen, Y.T., Nykanen, P., Nordstrom, D., 71. Firestein, GS, Echeverri, F, Yeo, M, Zvaifler, NJ, Saari, H., Sandelin, J., Santavirta, S., and Kouri, T. and Green, DR. (1997) Proc Natl Acad Sci USA 94, (1989) J Rheumatol 16, 339-345. 10895-10900. 53. Qu, Z., Garcia, C.H., O'Rourke, L.M., Planck, S.R., 72. Sun, Y., Cheung, J.M., Martel-Pelletier, J., Kohli, M., and Rosenbaum, J.T. (1994) Arthritis Rheum Pelletier, J.P., Wenger, L., Altman, R.D., Howell, 37, 212-220. D.S., and Cheung, H.S. (2000) J Biol Chem 275, 54. Ceponis, A., Konttinen, Y.T., Imai, S., 11327-11332. Tamulaitiene ,M., Li, T.F., Xu, J.W., Hietanen, J., 73. Yao, Q., Wang, S., Glorioso, J.C., Evans, C.H., Santavirta, S., and Fassbender, H.G. (1998) Br J Robbins, P.D., Ghivizzani, S.C., and Oligino, T.J. Rheumatol 37, 170-178. (2001) Mol Ther 3, 901-910. 55. Anastassiades, T.P., Ley, J., Wood, A., and Irwin, 74. Han, Z., Boyle, D.L., Shi, Y., Green, D.R., and D. (1978) Arthritis Rheum 21, 461-466 Firestein, G.S. (1999) Arthritis Rheum 42, 1088-1092. 56. Mohr, W., Beneke, G, Mohing, W. (1975 ) Ann 75. Kitasato, H, Okamoto, R, and Kawai, S. (2000) Rheum Dis 34, 219-224. Arthritis Rheum 43, 469-470. 57. Wibulswas, A., Croft D., Bacarese-Hamilton, I., 76. Kullmann, F., Judex M., Neudecker, I., Lechner, McIntyre, P., Genot, E., and Kramer, I.M. (2000) S., Justen, H.P., Green, D.R., Wessinghage, D., Am J Pathol 157, 2037-2044. Firestein, G.S., Gay, S., Scholmerich, J., and 58. Imamura, F., Aono, H., Hasunuma, T., Sumida, T., Muller-Ladner, U. (1999) Arthritis Rheum 42, 1594-1600. Tateishi, H., Maruo, S., and Nishioka, K. (1998) 77. Wibulswas, A., Croft, D., Pitsillides, A.A., Arthritis Rheum 41, 1979-1986. Bacarese-Hamilton, I., McIntyre, P., Genot, E., and 59. Qu, Z., Garcia, C.H., O'Rourke, L.M., Planck, S.R., Kramer, I.M. (2002) Arthritis Rheum 46, in press. Kohli, M., and Rosenbaum, J.T. (1994) Arthritis 78. Lamb, R.F., Hennigan, R.F., Turnbull, K., Rheum 37, 212-220. Katsanakis, K.D., MacKenzie, E.D., Birnie, G.D., 60. Xue, C., Takahashi, M., Hasunuma, T., Aono, H., Ozanne, B.W. (1997) Mol Cell Biol 17, 963-976. Yamamoto, K., Yoshino, S., Sumida, T., and 79. Geiler, T., Kriegsmann, J., Keyszer, G.M., Gay, R.E., Nishioka, K. (1997) Ann Rheum Dis 56, 262-267. and Gay, S. (1994) Arthritis Rheum 37, 1664-1671. 61. Lafyatis, R., Remmers, E.F., Roberts, A.B., Yocum, 80. Muller-Ladner, U., Kriegsmann, J., Franklin, B.N., D.E., Sporn, M.B., and Wilder, R.L.(1989) J Clin Invest Matsumoto, S., Geiler, T., Gay, R.E., and Gay, S. 83, 1267-1276. (1996) Am J Pathol 149, 1607-1615. 62. Chen, V., Croft, D., Purkis, P., and Kramer I.M. 81. Wernicke, D., Schulze-Westhoff, C., Brauer, R., (1998) Br J Rheumatol 37, 148-156. Petrow, P., Zacher, J., Gay, S., and Gromnica-Ihle, 63. Burger, D., Rezzonico, R., Li, J.M., Modoux, C., E. (2002) Arthritis Rheum 46, 64-74. 69 I. KRAMER, A. WIBULSWAS, D. CROFT AND E. GENOT 82. van der Laan, W.H., Pap, T., Ronday, H.K., 93. Evan, G. (1997) Int J Cancer 71, 709-711. Grimbergen, J.M., Huisman, L.G., TeKoppele, 94. Zhang, H.G., Wang, Y., Xie, J.F., Liang, X., Liu, D., J.M., Breedveld, F.C., Gay, R.E., Gay, S., Huizinga, Yang, P., Hsu, H.C., Ray, R.B., and Mountz, J.D. T.W., Verheijen, J.H., and Quax, P.H. (2000) (2001) Arthritis Rheum 44, 1555-1567. Arthritis Rheum 43, 1710-1718. 95. Asahara, H., Hasunuma, T., Kobata, T., Inoue, H., 83. Pap, T., van der Laan, W.H., Aupperle, K.R., Gay, Muller-Ladner, U., Gay, S., Sumida, T., and R.E., Verheijen, J.H., Firestein, G.S., Gay, S., and Nishioka, K. (1997) J Rheumatol 24(3), 430-435. Neidhart, M. (2000) Arthritis Rheum 43, 2531-2536. 96. Okamoto, K., Asahara, H., Kobayashi, T., 84. Muller-Ladner, U., Kriegsmann, J., Gay, R.E., and Matsuno, H., Hasunuma, T., Kobata, T., Sumida, Gay S. (1995) Clin North Am 21, 675-690. T., and Nishioka, K. (1998) Gene Ther 5, 331-338. 85. Neumann, E., Kullmann, F., Judex, M., Justen, H.P., 97. Ayroldi, E, Zollo, O, Cannarile, L, D' Adamio, F, Wessinghage, D., Gay, S., Scholmerich, J., and Grohmann U, Delfino, DV, and Riccardi, C. (1998) Muller-Ladner, U. (2002) Arthritis Rheum 46, 52-63. Blood 92, 4212-4219. 86. Ritchlin, C., Dwyer, E., Bucala, R., and 98. Atreya, R., Mudter, J., Finotto, S., Mullberg, J., Jostock, Winchester, R. (1994) Scand J Immunol 40, 292-298. T., Wirtz, S., Schutz M., Bartsch, B., Holtmann, M., 87. Utz, P.J. (2001) 21st European Workshop for Becker, C., Strand, D., Czaja, J., Schlaak, J.F., Lehr, Rheumatology Research, Vienna, Austria, H.A., Autschbach, F., Schurmann, G., Nishimoto, N., Arthritis Res 3, 237-240. Yoshizaki, K., Ito, H., Kishimoto, T., Galle, P.R., Rose- 88. Taniguchi, K., Kohsaka, H., Inoue, N., Terada, Y., Ito, John, S., and Neurath, M.F. (2000) Nat Med 6, 583-538. H., Hirokawa, K., and Miyasaka, N. (1999) Nat Med 99. Myers, L.K., Kang, A.H., Postlethwaite, A.E., 5, 760-767. Rosloniec, E.F., Morham, S.G., Shlopov, B.V., 89. Stein, G.H., Drullinger, L.F., Soulard, A., and Goorha, S., and Ballou, L.R. (2000) Arthritis Rheum Dulic, V. (1999) Mol Cell Biol 19, 2109-2117. 43, 2687-2693. 90. Nasu, K., Kohsaka, H., Nonomura, Y., Terada, Y., 100. Pai, R., Soreghan, B., Szabo, I.L., Pavelka, M., Baatar, Ito, H., Hirokawa, K., and Miyasaka, N. (2000) J D., and Tarnawski, A.S. (2002) Nat Med 8, 289-293. Immunol 165, 7246-7252. 91. Wakisaka, S., Suzuki, N., Saito, N., Ochi, T., and Sakane, T. (1998) Arthritis Rheum 41, 470-481. 92. Assoian, R.K., and Zhu, X. (1997) Curr Opin Cell Bio 9, 93-98. 70
"Rheumatoid arthritis _ targeting"