Rheumatoid arthritis (RA) is a chronic inflammatory disease of still-unknown etiology that leads to progressive destruction of affected joints by extracellular matrix degradation. The rheumatoid synovium is characterized by marked hyperplasia of synovial lining cells and prominent T lymphocyte infiltration (1). Fibroblast-like synoviocytes (FLS) and synovial macrophages have previously been shown to promote cartilage destruction and bone resorption by the production of matrix metalloproteinases (MMPs) (2), which are a family of zinc-dependent endopeptidases, and proinflammatory cytokines (3), such as interleukin-1β (IL-1β) and tumor necrosis factor α (TNFα). The contribution of T cells to the chronic inflammatory process in RA, however, remains to be defined. Association of the disease with HLA–DR4 antigens (4) and prominent accumulation of T cells in RA synovium (5) support a central role for these cells in the pathogenesis of RA. Until now, the mechanisms promoting T cell recruitment to sites of chronic inflammation have not been fully elucidated. T cell trafficking has previously been characterized as a multistep process involving selectins, adhesion molecules, and chemokines (6, 7).
Chemokines are secreted, low molecular weight proteins known to play a crucial role in transendothelial leukocyte migration and activation along chemoattractant gradients during inflammation (8). These molecules are structurally related and have previously been classified according to the organization of the N-terminal conserved cysteine motif into 4 groups, designated as chemokines CC, CXC, CX3C, and C (9). Whereas the CXC and CC groups both include several members, the C and CX3C chemokine subfamilies are represented so far by only 1 member each: lymphotactin (Lptn; XCL1) and fractalkine (or, neurotactin; CX3CL1), respectively. The 4 groups can also be distinguished according to their chromosomal localization and their biologic activities: CXC chemokines mainly target neutrophils and T cells, while CC chemokines generally attract monocytes and T cells. In contrast, both C and CX3C chemokines have a more restricted specificity for T cells.
The Lptn molecule was independently detected by Kelner et al (10), who called it lymphotactin, Yoshida et al (11), who called it single C motif 1 (SCM-1), and Müller et al (12), who called it activation-induced, T cell–derived, and chemokine-related molecule (ATAC). This novel chemokine is structurally related to the CC chemokine subfamily that lacks the first and third cysteine residues, and it is thus considered to represent the C chemokine subfamily. Two highly homologous genes encoding for XCL1 and XCL2, respectively, have thus far been detected in humans. ATAC/Lptn was found to be selectively expressed in activated CD8+ T cells and in a small proportion of activated CD4+ T cells (12), α/β-type thymocytes (10), intraepithelial γ/δ-type T cells (13), mast cells (14), and natural killer (NK) cells (15). ATAC/Lptn acts via a unique G protein–coupled receptor (XCR1) (16, 17).
The full spectrum of biologic functions of ATAC/Lptn is still unknown, but the functional properties that have been detected include chemotactic activity for both CD4+ and CD8+ T cells (18, 19) and induction of migratory responses in NK cells after activation with IL-2 (20). The ability of ATAC/Lptn to specifically recruit T cells and NK cells has been used successfully in gene therapy. Genetic modification of tumor RNA–pulsed dendritic cells to secrete ATAC/Lptn was shown to be associated with an enhanced therapeutic efficacy of dendritic cell–based tumor vaccines (21). Moreover, results of recent studies in a murine model of listeriosis suggest that ATAC/Lptn, macrophage inflammatory protein 1α (MIP-1α; CCL3), MIP-1β (CCL4), and RANTES (CCL5) are cosecreted with interferon-γ (IFNγ) by activated NK cells, CD8+ T cells, and CD4+ Th1 cells, and function as type 1 cytokines by up-regulating CD40, IL-12, and TNFα in macrophages (22).
Expression of ATAC/Lptn has been studied in several clinical and experimental models of inflammatory disease, such as acute allograft rejection (23, 24), autoimmune diabetes (25), encephalomyelitis (26), experimental crescentic glomerulonephritis (27), and chronic inflammatory bowel disease (28). Results of these studies support the concept of a potential role of ATAC/Lptn in Th1-type inflammatory processes. To analyze whether ATAC/Lptn may also participate in the pathogenesis of RA, we investigated the expression of ATAC/Lptn and its receptor in synovial tissues from RA patients and compared it with the expression in synovial tissues from osteoarthritis (OA) control patients. Our findings indicate that there is enhanced expression of ATAC/Lptn in rheumatoid synovium, on both CD4+ and CD8+ T cells. In addition, this study is the first to demonstrate the expression of the ATAC/Lptn receptor (XCR1) on FLS. Functional analysis revealed a marked down-regulation of MMP-2 production in cultured FLS during in vitro stimulation with ATAC/Lptn. These findings suggest a novel immunoregulatory function for ATAC/Lptn in RA.
- Top of page
- PATIENTS AND METHODS
To our knowledge, this is the first study to demonstrate a prominent expression of ATAC/Lptn and its receptor XCR1 in the synovial tissue of RA patients. We established a highly sensitive ELISA to examine the levels of ATAC/Lptn in the peripheral blood and in synovial fluid from affected joints. ATAC/Lptn levels in serum and synovial fluid from RA patients were not significantly elevated levels compared with those in healthy controls and in OA control patients, respectively. This finding is in contrast to the observed overexpression of ATAC/Lptn mRNA and protein in RA compared with OA synovium. A possible explanation for this discrepancy is a strictly local induction of ATAC/Lptn in RA synovium. ATAC/Lptn may be secreted locally and may exert its effects directly at the site of chronic inflammation in RA. A tight regulation of ATAC/Lptn production, with induction only in response to proinflammatory stimuli, is also suggested by our finding that intracellular ATAC/Lptn was detectable in PBMC cultures only after stimulation with PMA and ionomycin.
Consistent with previous results (12), ATAC/Lptn expression was predominantly detected in CD8+ T cells and in a small proportion of CD4+ T cells. Analysis of the immunophenotype of ATAC/Lptn-positive T cells revealed that the majority of CD4+ and the CD8+ T cells containing ATAC/Lptn lacked expression of the costimulatory molecule CD28. Interestingly, expansion of CD4+,CD28− and CD8+,CD28− T cell clones in RA has previously been identified (39–41).
CD4+,CD28− T cells are infrequent in normal subjects, but account for up to 50% of CD4+ T cells in RA patients (42). In line with a role of these cells in the pathogenesis of RA, a correlation between the expansion of CD4+,CD28− T cells and the manifestation of clinical symptoms has previously been observed: Patients with extraarticular disease manifestations were found to carry the highest frequency of this unusual T cell subset. CD4+,CD28− T cells were shown to infiltrate the synovial lesions, to produce high amounts of IFNγ (43), and to lyse target cells by producing perforin and granzyme B (44). These properties of CD4+,CD28− T cells further support a direct contribution of this T cell subset to the chronic inflammatory process in RA.
Quantitative RT-PCR for the detection of ATAC/Lptn mRNA expression revealed that ATAC/Lptn transcript levels were significantly elevated in RA synovium compared with OA synovium (P < 0.001). In situ hybridization and immunohistochemistry studies further demonstrated that on both the transcript (mRNA) level and the protein level, ATAC/Lptn-positive cells were mainly localized in lymphocytic infiltrates of the sublining layer. Double immunostaining for cell surface antigens revealed that most of these ATAC/Lptn-positive cells were CD3+ T cells, whereas cells of the monocyte/macrophage lineage did not contain ATAC/Lptn.
Further characterization of the phenotype of ATAC/Lptn-positive T lymphocytes revealed the expression of ATAC/Lptn in both the CD8+ and the CD4+ T cell subsets. These findings were consistent with those of previous studies demonstrating that activated CD8+ T cells and a small proportion of CD4+ T cells represent the most important source of this chemokine in the peripheral blood (19). In RA synovium, ATAC/Lptn expression could also be demonstrated in mast cells, NK cells, and dendritic cells. Expression of ATAC/Lptn in these 3 different cell types has previously been described. Rumsaeng et al (14) demonstrated that ATAC/Lptn synthesis and release could be induced in mast cells in response to Fcε receptor I aggregation. In addition, ATAC/Lptn expression has been demonstrated in NK cells (15) and was shown to be up-regulated in these cells by activating Ly-49 NK receptors (45). In Crohn's disease, ATAC/Lptn expression has been demonstrated in dendritic cells (27). Since there is evidence that expression of ATAC/Lptn is up-regulated in dendritic cells during maturation (46), our finding of ATAC/Lptn-positive dendritic cells in RA synovium may indicate the presence of mature dendritic cells in the inflamed synovium. Our present data and our previous data (12) indicate the absence of ATAC/Lptn in cells of the monocyte/macrophage lineage, including cytokine-activated monocytes (U937), fibroblasts, and HeLa cells.
Analysis of ATAC/Lptn expression in other inflammatory arthritides, including OA, reactive arthritis, and psoriatic arthritis, revealed a prominent staining of inflammatory cells only in psoriatic arthritis, which suggests a role for ATAC/Lptn in Th1 cytokine–associated disease states. Comparison of early-stage and late-stage RA demonstrated a more pronounced expression of ATAC/Lptn in lymphocytic infiltrates of the sublining layer in early RA, which also indicates that ATAC/Lptn might exert an important function in the early phase of RA pathogenesis.
To further elucidate a potential functional role in RA, we studied the expression of the ATAC/Lptn receptor XCR1 in different cell types known to be involved in the pathogenic process of RA. As detected by RT-PCR, T cells and, surprisingly, FLS isolated from the synovial tissue of RA patients were shown to express the ATAC/Lptn receptor mRNA. XCR1 has been identified as a specific functional, G protein–coupled receptor for ATAC/Lptn and has been detected in selected tissues, including placenta, spleen, and thymus, but was rarely found in PBMCs (16, 17). In addition, B cells and neutrophils have also been shown to express the ATAC/Lptn receptor (47) and to respond to ATAC/Lptn by chemotaxis. XCR1 expression was recently detected in tissue macrophages in a murine model of listeriosis (22). The RT-PCR results in our study indicate that XCR1 is also highly expressed in the rheumatoid synovium.
Since T cells represent the main source of ATAC/Lptn expression in RA synovium, expression of XCR1 on these cells may indicate autocrine and/or paracrine regulation by ATAC/Lptn. The finding that FLS were also positive for XCR1 expression is rather striking and suggests that these cells also represent targets of ATAC/Lptn activity in RA synovium. The presence of functional ATAC/Lptn receptors on FLS is supported by our finding that ATAC/Lptn dose-dependently regulated the production of MMP-2 in cultured FLS at the mRNA and protein levels.
Previous studies have largely focused on the chemoattractant properties of ATAC/Lptn. It has been shown to induce chemotactic responses in T cells (10, 18, 48), NK cells (15), and B cells and neutrophils (47), but this functional activity has not been confirmed by other investigators (16, 19). More recent data indicate that Lptn may particularly enhance the migration of antigen-activated CD62Llow T cells (49) and memory T cells (50). In addition, ATAC/Lptn gene–modified dendritic cells were capable of attracting T cells and NK cells in chemotaxis assays (21) and were successfully used to enhance antitumor immunity. In light of these in vitro functions of ATAC/Lptn, the expression of ATAC/Lptn in RA synovium may contribute to the selective recruitment of T cells to the synovium of affected joints, where accumulated T cells are known to be harbored.
Recent studies further suggest that apart from its chemotactic properties, ATAC/Lptn might exert multiple biologic functions. In a murine model of listeriosis, ATAC/Lptn as well as MIP-1α, MIP-1β, and RANTES were shown to be cosecreted with IFNγ and to act as type 1 cytokines by up-regulating CD40, IL-12, and TNFα in macrophages (22). As part of a proinflammatory cytokine milieu in RA, ATAC/Lptn may enhance the expression of these mediators in synovial macrophages as well, and thus may perpetuate the inflammatory destruction of the articular tissue. However, these hypothesized effects of ATAC/Lptn stimulation on synovial macrophage function have not yet been analyzed in detail.
Our data on the regulation of MMP-2 in FLS suggest a more complex role of ATAC/Lptn in RA that may include additional immunomodulatory effects or even antiinflammatory effects. Down-regulation of MMP-2 production was demonstrated on both the RNA and protein levels in cultured FLS after stimulation with ATAC/Lptn. This effect of ATAC/Lptn could be abrogated by the addition of neutralizing anti-ATAC/Lptn antibodies. Thus, apart from T cell recruitment, ATAC/Lptn might have an additional function in regulating the mechanisms of disease progression in RA. However, at this point, it cannot be excluded that the in vitro effects observed in cultured FLS may not correlate with the effects in vivo, especially in the environment of a proinflammatory cytokine milieu characteristic of RA. Further studies, including studies in animal models, are therefore needed to more precisely characterize the role of ATAC/Lptn in RA.
In summary, our study demonstrates an increased expression of ATAC/Lptn in CD4+,CD28− T cells in the peripheral blood as well as in CD4+ and CD8+ T cells in the synovial infiltrate in patients with RA. Our study also shows the expression of the ATAC/Lptn receptor XCR1 in lymphocytes, synovial macrophages, and, interestingly, FLS. Functional analysis of cultured FLS stimulated with ATAC/Lptn in vitro suggests that this chemokine may exert an immunomodulatory effect in RA in addition to its known chemotactic activity for T cells. Our findings suggest a crucial role for this chemokine in the pathogenesis of the chronic inflammatory process in RA.