Assignment: American Psychological Assoc
Assignment: American Psychological Assoc
Permalink:
Goronzy, J. J., & Weyand, C. M. (2005). Rheumatoid arthritis. Immunological Reviews, 204(1), 55-73. doi:10.1111/j.0105-2896.2005.00245.x
tment of Medicine, Kathleen B. and Mason I. Lowance Center for Human Immunology, Emory School of Medicine, Atlanta, GA, USA. Correspondence to: Jo¨rg J. Goronzy, MD, PhD Lowance Center for Human Immunology Emory University School of Medicine Room 1003 Woodruff Memorial Research Building 101 Woodruff Circle Atlanta, GA 30322 USA Tel.: þ1 404 727 7310 Fax: þ1 404 727 7371 E-mail: jgoronz@emory.edu Acknowledgements We thank Dr S. Pryshchep for help in the preparation of the figures, L. Arneson for secretarial support, and T. Yeargin for editorial support. This work was funded in part by grants from the National Institutes of Health (RO1 AR 42527, RO1 AI 44142, RO1 AR 41974, and RO1 AG 15043). Summary: Therapeutic efficacy of depleting B cells or blocking T-cell costimulation in rheumatoid arthritis (RA) has confirmed the critical pathogenic role of adaptive immune responses. Yet, RA preferentially affects elderly individuals, in whom adaptive immunity to exogenous antigens begins to fail. Here, we propose that senescence of the immune system is a risk factor for RA, with chronic inflammation resulting from the accumulation of degenerate T cells that have a low threshold for activation and utilize a spectrum of novel receptors to respond to microenvironmental cues. The process of immunosenescence is accelerated in RA and precedes the onset of disease, the acceleration, in part, being conferred by the HLA-DR4 haplotype. Naive CD4þ T cells in RA are contracted in diversity and restricted in clonal burst. Senescence of effector CD4þ T cells is associated with the loss of CD28 and the de novo expression of KIR2DS2, NKG2D, and CX3CR1, all of which function as costimulatory molecules and reduce the threshold for T-cell activation. The synovial microenvironment promotes chronic persistent immune responses by facilitating ectopic lymphoid neogenesis, such as the formation of aberrant germinal centers. With the propensity to develop complex lymphoid architectures and to provide optimal activation conditions for senescent CD4þ T cells, the synovium becomes a natural target for pathogenic immune responses in prematurely aged individuals. Introduction In the broad spectrum of inflammatory joint diseases, rheumatoid arthritis (RA) has a prominent position. By causing symmetrical and destructive inflammation in the small and large joints, RA leads to pain and joint failure, eventually resulting in disfiguration and disability. RA is a quintessential example of an immune-mediated disease with chronic smoldering immune responses causing tissue damage. The syndrome is unequivocally systemic in nature, but synovitis of the diathrodial joints is the most visible manifestation. The most significant risk factors for RA are female gender, advanced age, and a positive family history. In recent years, the therapeutic management of patients with RA has seen a major evolution (1). A number of biological reagents have been developed that allow for marked suppression of inflammation, at least in a subset of patients. Spearheaded by Immunological Reviews 2005 Vol. 204: 55–73 Printed in Singapore. All rights reserved Copyright Blackwell Munksgaard 2005 Immunological Reviews 0105-2896 55 the efficacy of tumor necrosis factor-a (TNF-a) blocking agents (2, 3), a number of other anti-cytokine therapies are making their way into clinical application. Preliminary data suggest that anti-interleukin-6 (anti-IL-6) therapy has profound anti-inflammatory activity (4), and encouraging results for blockade of IL-15 are emerging (5). Other therapeutic strategies have targeted cellular pathways in the rheumatoid disease process. Remarkably, impressive therapeutic responses are being reported for B-cell-depleting regimens using anti-CD20 antibodies (6). Similarly, blocking CD28-mediated costimulation with cytotoxic T-lymphocyte antigen 4-immunoglobulin fusion protein (CTLA4-Ig) has shown convincing therapeutic benefits, and follow-up studies of patients treated with CTLA4-Ig have suggested therapeutic responses qualifying as remission (7). The clinical efficacy of many novel immunosuppressive therapies in RA has emphasized the complexity of the disease process and has challenged a disease model that proposes a single and straightforward pathomechanism. A multitude of pathways appear to contribute to persistent and tissue-injurious inflammation. While initial success of TNF-a blockade suggested a gatekeeper function of this cytokine in rheumatoid synovitis (3), similar claims can now be made for B cells, T-cell costimulation, and an array of other cytokines. A number of important lessons have been learned from examining the therapeutic effects of immunomodulatory interventions. While diverse in the molecular targets, multiple strategies have been successful. As a rule, treatment responses are partial instead of complete. Continuous immunosuppression is required to maintain clinical benefits, documenting that the key abnormalities have not been removed. Accumulating clinical experience supports heterogeneity in the disease process (8), emphasizes the coexistence of systemic and joint-directed disease components (9), and stresses the need for identifying the ultimate principle causing autoimmunity in RA. Current animal models recapitulate just a single facet of the disease and fail to provide a unifying concept. Considering that all novel immunosuppressants make the patient an immuno-compromised host, susceptible to life-threatening infections and possibly tumors (10), efforts should go towards developing therapies that restore instead of impair immunocompetence. The rationale for current therapeutic strategies has derived from a focus on common pathways in tissue inflammation, many of them not at all unique for RA. It is unlikely that disrupting or suppressing generalized pathways of inflammation can eradicate RA. In the search for characteristics that are typical for RA, two features of this autoimmune disorder come to mind. While systemic in nature, the chronic persistent immune response seeks out the synovial membrane of selected joints. The classical paradigm that joint specificity can be explained as an immune response to joint-specific antigens has lost attraction. Other elements of tissue susceptibility have to be considered that define the preference for the synovial space (11). Also, RA is a disease of the elderly, not of the young. Incidence rates of RA are highest in postmenopausal women and increase steadily with advancing age (12). How can a host in whom the aging process gradually but profoundly compromises immunity be most susceptible to self-aggressive immune responses? The traditional concepts of a defect in central and peripheral tolerance were developed with the young and evolving immune system in mind; new theories are needed to understand loss of tolerance during the second half of life. This review will summarize recent data on the synovial microenvironment as a disease susceptibility factor and the contribution of immunosenescence in the derailing of immune tolerance. Tissue specificity and autoantigen expression In the classical model of autoimmunity, tissue specificity is determined by the tissue-specific recognition of autoantigens that elicit adaptive immune responses. Consequently, studies into the pathogenesis of RA have implicated antigens that are locally expressed in the joint. Prominent candidate antigens have included type II collagen, the cartilage protein gp39, and proteoglycans. These antigens have been shown to induce autoimmunity and arthritis in rodent models. However, their relevance for the pathogenesis of RA is at best uncertain. None of these autoantigen-directed responses have been found to associate uniformly with RA. Autoantibodies that are being produced in RA are generally specific for ubiquitous antigens. One of the best examples is the rheumatoid factor, an autoantibody specific for the Fc portion of IgG. While rheumatoid factor is occasionally found in some other autoimmune diseases, it remains one of the most sensitive laboratory tests for RA. Even higher specificity and equal sensitivity have been reported for antibodies that are directed to citrullinated proteins (13). These antibodies that detect cyclic citrullinated peptides in many different proteins, such as fillagrin, are detected in about 80% of patients with RA at a specificity of 98%. Several lines of evidence suggest that citrullinated antigens have direct involvement in the rheumatoid disease process. Citrullination is mediated by peptidylarginine deiminases. One of the deiminases, PAD4, has been identified as an RA susceptibility gene (14). Moreover, citrullinated peptides Goronzy & Weyand Rheumatoid arthritis 56 Immunological Reviews 204/2005 appear to bind with high affinity to the RA-associated human leukocyte antigen (HLA)-DRB1*04 alleles (15). Citrullinated proteins may be present in the RA synovium. However, this association is controversial, and possible candidate autoantigens, such as citrullinated fibrins, citrullinated vimentin, and citrullinated histones, are certainly not synovium specific (13). Also, anti-cyclic citrullin peptide antibodies precede the clinical development of synovitis by many years (16, 17). Together, these data support the concept that RA-associated autoantibodies are indicators of general immune defects and are not the mediators of classical autoimmune responses to autoantigens uniquely available in the synovium. This argument holds for antibody and T-cell responses to a variety of other autoantigens that have been described in patients with RA, including responses to several heat-shock proteins, such as BiP (18). The rheumatoid synovium: facilitator of aberrant immune responses The tissue preference of the immune response in RA appears not to be conferred by tissue-specific expression of autoantigens. In an alternative model, the rheumatoid synovium provides an exceptional microenvironment for abnormal immune responses that settle in while immune abnormalities occur systemically. Examples for the importance of the tissue microarchitecture in directing and facilitating immune responses are widely known from developmental biology (19, 20). In fact, the formation of secondary lymphoid structures in lymph nodes and the gastrointestinal tract is a prime example of how the structural organization of the tissue site determines whether and how immune responses occur. This concept has been explored less in diseased non-lymphoid tissues, where it is likely to play a major role in explaining the pathogenesis of chronic inflammatory diseases (21). Numerous studies in the last two decades have stressed the point that activation events in the rheumatoid synovium and, in particular, the interactions between T cells and fibroblastlike synoviocytes, are cell–cell contact dependent and therefore assign particular importance to tissue architecture and spatial relationships (22–26). A major reason why this issue has not been more aggressively
