SEARCH

SEARCH BY CITATION

Current views of rheumatoid arthritis (RA) pathogenesis emphasize an autoimmune etiology marshalling components of the innate and adaptive immune responses (1). As a consequence of rheumatoid inflammation, joint synovium develops into the erosive granulation tissue known as pannus, which is a pathologic hallmark of RA (2). Within pannus tissue, there is an enormous increase in the number of fibroblast-like synoviocytes, persistent inflammatory foci, lymphoid germinal centers, and angiogenesis. In vitro studies have established that fibroblasts isolated from the synovia of patients with RA exhibit an activated phenotype with enhanced proliferation, invasive properties, and increased secretion of growth-promoting cytokines and matrix-degrading enzymes.

Rheumatoid synovial fibroblasts proliferate in an anchorage-independent manner, escape contact inhibition, and invade cartilage implants (2–5). Remarkably, this aggressive phenotype endures after multiple cell passages and prolonged culture, suggesting that synovial fibroblasts are persistently, if not irreversibly, altered in the rheumatoid joint. The precise manner in which synovial fibroblasts maintain this activated state after removal from the inflammatory articular milieu is unknown, but such maintenance may result from a self-perpetuating or reentrant stimulatory pathway involving autocrine mediators and sustained intracellular signaling. There is evidence in rheumatoid synovial fibroblasts for protooncogene activation and for the functional (and perhaps mutational) inactivation of the tumor suppressor gene p53, which regulates cell cycle progression (6).

Although the notion of a mutational origin for pannus tissue has been largely discounted (and does not account for multicentric rheumatoid synovitis), long-term epigenetic changes may explain the activated phenotype of rheumatoid synovial fibroblasts. Beyond exhibiting an invasive phenotype, the hyperplastic synovium also supports leukocyte recruitment and may be directly activating to the immune system (7, 8). Why pannus develops in RA but not in other arthritides also is unknown, but pannus development is central to the progression of joint destruction. Another important clinical question is whether the currently available therapies are effective for treating rheumatoid pannus or whether, as in systemic sclerosis, activated connective tissue is intrinsically resistant to immunosuppression and should be targeted pharmacologically in order to achieve effective disease control.

In this issue of Arthritis & Rheumatism, Galligan and Fish present experimental evidence for the participation of circulating fibrocytes in the pathogenesis of inflammatory synovitis (9). The fibrocyte is a monocyte-derived cell that expresses features of both a macrophage and a connective tissue cell. First identified in the 1990s by their coexpression of hematopoietic progenitor markers together with extracellular matrix proteins, fibrocytes are among the earliest responding cells in the innate response to injury or tissue invasion (10). Fibrocytes normally comprise a minor population of circulating leukocytes, but they are mobilized from the bone marrow and circulate in increased numbers during the systemic inflammatory response, in the setting of ongoing fibrogenesis, or as a consequence of aging (11).

Galligan and Fish approached their investigation under the hypothesis that an alteration in proliferation or apoptosis rates alone cannot account for the kinetics of synovial hyperplasia. Thymidine uptake is sparse, and there are few mitotic figures in pannus. This dilemma, which was thoughtfully reviewed by Nathan Zvaifler (12), prompted consideration of alternative mechanisms for pannus development. Unremitting inflammation, initiated by immune complex deposition and an innate immune response leading to endothelial activation, may create an environment conducive to the entry and residence of mesenchymal cell precursors that become established within the synovium. The circulating fibrocyte is one such candidate mesenchymal cell that already has been shown to contribute to tissue remodeling in the lung (13, 14), the kidney (15), and the liver (16). Galligan and colleagues previously proposed fibrocytes to be precursors of fibroblast-like synoviocytes (17), which is an attractive idea given the ability of fibrocytes to produce abundant amounts of inflammatory and proliferative cytokines and to present antigen to naive T cells (18, 19). Indeed, those investigators observed an increase in the signal activation profile of circulating fibrocytes in patients with RA and an increase in a similar population of STAT-5–positive fibrocytes in the circulation and synovia of mice with collagen-induced arthritis (17).

Galligan and Fish now report the outcome of adoptive transfer studies in which fibrocytes isolated from the bloodstream of mice with collagen-induced arthritis, a chronic and T cell–dependent form of disease, are injected into the circulation of mice with anti–collagen antibody–induced arthritis, which is more acute than collagen-induced arthritis and is T cell independent. By this manipulation, the authors observed that the transferred fibrocytes migrated into synovia and exacerbated joint pathology, with an attendant increase in neutrophil content and the expression of tissue-destructive matrix metalloproteinases. Enhanced STAT-1 and JNK phosphorylation as well as increased class II major histocompatibility complex expression also were noted in the circulating leukocytes of fibrocyte-recipient mice. These data provide first support for the notion that circulating fibrocytes contribute to the activated phenotype of synovial stroma and may participate directly in tissue-destructive inflammatory responses.

These new findings provide opportunities for new lines of research. It would be of great interest to identify the specific signals necessary for fibrocyte trafficking and accumulation in synovial tissue and for fibrocyte mobilization from bone marrow progenitors (Figure 1). Such pathways have been probed in models of pulmonary inflammation leading to lung remodeling, and they offer attractive access points for new therapeutic intervention. The chemokines CXCL12 (stromal cell–derived factor 1), CCL2 (secondary lymphoid chemokine), and semaphorin 7A may be such candidates (14, 20, 21). CD4+ T cells are critical to RA pathogenesis and regulate fibrocyte differentiation (22). Notably, it appears that the precise context of T cell activation, whether stimulation proceeds in the presence of calcineurin or mammalian target of rapamycin inhibition for instance, can determine whether fibrocyte differentiation is supported or blocked. A link between fibrocytes and autoimmunity also has been provided by recent studies in Graves' ophthalmopathy (23).

thumbnail image

Figure 1. Scheme for the proposed role of fibrocytes in rheumatoid synovitis, showing multiple steps in mobilization from bone marrow, differentiation from monocytic precursors, inflammatory activation, and recruitment to synovium. Col I = type I collagen.

Download figure to PowerPoint

Galligan and Fish demonstrate a link between circulating fibrocytes and synovial inflammation and propose both a new participant in pathogenesis as well as a new target for therapy. Therapeutic prospects must be tempered by clinical experience, which indicates that RA is a heterogeneous disease, with common phenotypic features expressed over a diverse genetic background of susceptibility. Immunopathogenesis thus may proceed by more than one avenue. The authors' observations nevertheless arrive at a propitious time, because pharmacologic fibrocyte blockade is advancing into phase II clinical testing (24, 25). Equally exciting is ongoing work showing that enumeration of circulating fibrocytes offers information about clinical progression in different fibrosing disorders (26–28). This advance may be applied to RA and conceivably guide the definition of RA subtypes or the application of synoviocyte-directed therapies. Whether rheumatoid pannus develops by the intermediation of circulating fibrocytes ultimately can be tested only by pharmacologic intervention. Nevertheless, this study can be welcomed for providing insight into a novel and potentially important feature of pannus biology.

AUTHOR CONTRIBUTIONS

  1. Top of page
  2. AUTHOR CONTRIBUTIONS
  3. Acknowledgements
  4. REFERENCES

Dr. Bucala drafted the article, revised it critically for important intellectual content, and approved the final version to be published.

Acknowledgements

  1. Top of page
  2. AUTHOR CONTRIBUTIONS
  3. Acknowledgements
  4. REFERENCES

Gratitude is expressed to Prof. Erica Herzog and Adriana Blakaj for their review of the manuscript.

REFERENCES

  1. Top of page
  2. AUTHOR CONTRIBUTIONS
  3. Acknowledgements
  4. REFERENCES