Pathogenesis and immunopathology of rheumatoid polyarthritis

Etiology of rheumatoid arthritis remains unresolved. Initiating mechanisms include a genetic background of susceptibility, defined by HLA-DR4 and -DR1 antigens and also hormonal factors and infectious events. The synovitis is the hallmark of the disease. Lymphocytes infiltrating the synovial membrane are mainly memory CD4+ T cells associated to dominant clonotypes. The proliferation and cytokine production by synovial tissue T lymphocytes are weak. B cells produce locally auto-antibodies: rheumatoid factors, anti-keratin antibodies.... Type A and B synoviocytes constitute an hyperplastic synovial lining. They synthesize and release large amount of proteolytic enzymes and proinflammatory cytokines, largely participating to inflammation and tissue destruction. Understanding the pathogenesis of rheumatoid arthritis should offer novel approaches to treat rheumatoid arthritis patients.     

The role of CD8+ CD40L+ T cells in the formation of germinal centers in rheumatoid synovitis

In rheumatoid synovitis, lymphocytes can be arranged in follicular structures resembling secondary lymphoid follicles. To understand the organizing principles of this ectopic lymphoid tissue, the cellular components contributing to synovial follicles were examined. In 9 of 24 synovial tissue biopsies, lymphoid aggregates were found consisting of CD4+ T cells and CD20+ B cells. In four of the nine patients, the follicular centers were occupied by CD23+ CD21+ cellular networks representing follicular dendritic cells involved in germinal center reactions. In five patients, CD23+ cells were absent from the centers of the aggregates, suggesting that fully developed germinal centers are generated in only a subset of patients. To identify factors involved in the regulation of the synovial microarchitecture, cell populations contributing to the follicles were quantified by digital image analysis of immunostained tissue and by flow cytometry of tissue-derived lymphocytes. Proportions of CD4+, CD20+, and CD68+ cell subsets were surprisingly invariant, irrespective of the presence or absence of CD23+ follicular dendritic cells. Instead, tissue biopsies with CD23+ germinal center-like regions could be distinguished from those with CD23- T cell-B cell aggregates by a fourfold increase in the frequency of tissue-infiltrating CD8+ T cells, a fraction of which expressed CD40 ligand (CD40L). The data suggest a previously unsuspected role of CD8+ lymphocytes in modulating germinal center formation and raise the possibility that CD8+ CD40L+ T cells are involved in aggravating pathologic immune responses in rheumatoid synovitis.     

Recombinant human growth-regulated oncogene-alpha induces T lymphocyte chemotaxis. A process regulated via IL-8 receptors by IFN-gamma, TNF-alpha, IL-4, IL-10, and IL-13

The human cytokine growth-regulated oncogene (GRO)-alpha is a small glycoprotein secreted by monocytes, endothelial cells, glycoprotein secreted by monocytes, endothelial cells, fibroblasts, synovial cells, and some tumor cells such as melanoma cells. It is structurally related to IL-8 and can activate neutrophils, whereas it induces chemotaxis, exocytosis, and a respiratory burst in neutrophils. To date, its functions on T lymphocytes have not been well established. We report here that recombinant human (rh)GRO-alpha is a potent chemoattractant for freshly isolated T lymphocytes, but not for anti-CD3 mAb-activated T lymphocytes. It attracts CD4+ and CD8+ T lymphocyte subsets to an equal extent. The migrating T lymphocytes toward rhGRO-alpha are predominantly CD45RO+ memory CD4+ and CD8+ subsets. The chemotactic migration of T lymphocytes toward rhGRO-alpha is stimulated via the IL-8 receptors on the cells. This process can be augmented by IFN-gamma and TNF-alpha, and inhibited by IL-4, IL-10, and IL-13. In addition, we also document that on T lymphocytes there exist IL-8 receptors that can be up-regulated by IFN-gamma, TNF-alpha, and IL-2. Our results demonstrate that rhGRO-alpha gene encodes for an inflammatory mediator that stimulates the directional migration of T lymphocytes. It may thus be another important mediator in the diseases in which T lymphocytes form the major constituent of the cellular infiltration.     

Analysis of type 1 and type 2 T cells in synovial fluid and peripheral blood of patients with rheumatoid arthritis

OBJECTIVE: It has been reported that CD4+ helper T cells play an important role in the pathogenesis of rheumatoid arthritis (RA). We evaluated the presence of intracellular cytokines interleukin 4 (IL-4) and interferon-gamma (IFN-gamma) produced by CD4+ and CD8+ T cells in the synovial fluid and peripheral blood of patients with RA at the single cell level. METHODS: We used 3 color flow cytometric analysis. Synovial fluid mononuclear cells (SFMC) and peripheral blood mononuclear cells (PBMC) were stimulated with phorbol myristate acetate (PMA) and calcium ionophore. The stimulated SFMC and PBMC were triple stained with conjugated mononuclear antibodies (Mab) against cytokines and surface antigens after fixation and permeabilization with a saponine buffer solution. The cells were analyzed for intracellular cytokines (IFN-gamma, IL-4) and surface antigens (CD3, CD4, CD8) using a flow cytometer. RESULTS: The CD4/CD8 ratio was significantly lower in SFMC than in PBMC. The positive rates of IFN-gamma producing cells among CD4+ T cells were significantly higher than those of IL-4 producing cells in both the SFMC and the PBMC of patients with active RA. In the SF of these patients, we also found CD8+ T cells that produce IL-4 alone, or both IL-4 and IFN-gamma. CONCLUSION: In the SF of patients with RA, CD4+ type 1 T cells, which may infiltrate into the synovium and cause pathogenic immune responses in the tissue, are predominant. We believe this cell type also induces migration and activation of CD8+ type 2 T cells into the active site of inflammation, which appears to downregulate the activity of CD4+ type 1 T cells, modulating the excess immune response.     

Biased T-cell antigen receptor repertoire in Lyme arthritis

A common concern with many autoimmune diseases of unknown etiology is the extent to which tissue T-lymphocyte infiltrates, versus a nonspecific infiltrate, reflect a response to the causative agent. Lyme arthritis can histologically resemble rheumatoid synovitis, particularly the prominent infiltration by T lymphocytes. This has raised speculation about whether Lyme synovitis represents an ongoing response to the causative spirochete, Borrelia burgdorferi, or rather a self-perpetuating autoimmune reaction. In an effort to answer this question, the present study examined the repertoire of infiltrating T cells in synovial fluid from nine Lyme arthritis patients, before and after stimulation with B. burgdorferi. Using a highly sensitive and consistent quantitative PCR technique, a comparison of the T-cell antigen receptor (TCR) beta-chain variable (Vbeta) repertoires of the peripheral blood and synovial fluid showed a statistically significant increase in expression of Vbeta2 and Vbeta6 in the latter. This is remarkably similar to our previous findings in studies of rheumatoid arthritis and to other reports on psoriatic skin lesions. However, stimulation of synovial fluid T cells with B. burgdorferi provoked active proliferation but not a statistically significant increase in expression of any TCR Vbeta, including Vbeta2 and Vbeta6. Collectively, the findings suggest that the skewing of the TCR repertoire of fresh synovial fluid in Lyme arthritis may represent more a synovium-tropic or nonspecific inflammatory response, similar to that occurring in rheumatoid arthritis or psoriasis, rather than a specific Borrelia reaction.     

Acquired heart valve pathology. An update for the millennium

Two inflammatory vascular diseases often show multinucleated macrophages: Takayasu's disease and Horton's disease. Takayasu's disease is a segmentary panarteritis most prominent in the adventitia. Lesions show an inflammatory infiltrate close to the external elastic lamina. Progressive stenosis of the artery, sometimes complicated by calcifying atheroma is the typical course. Horton's disease or temporal arteritis is another segmentary arteritis. Lesions show a mixed inflammatory infiltrate partly localized in the adventitia where there are T CD4+ lymphocytes secreting II-2 and IFN-gamma and also macrophages expressing TGF beta 1, IL-6 and IL-1 beta, and partly situated in the interior part of the wall, around the internal elastic lamina, and mostly made of macrophages and giant cells which produce TNF, collagenase and nitric oxide that are responsible for destruction of the wall. The variety and subtleness of some lesions do not always make a precise diagnosis possible. But any inflammatory vascular lesion, even slight, can reveal a systemic vasculitis.     

Detection of Borrelia burgdorferi-specific CD8+ cytotoxic T cells in patients with Lyme arthritis

T cells are believed to play an important role in the pathogenesis of Lyme arthritis (LA), an inflammatory joint disease caused by the spirochete Borrelia burgdorferi (Bb). The presence or absence of certain Bb-specific CD4+ T helper cells has been associated with prognosis. Since recent observations suggested the activation of CD8+ T cells during infection with Bb, we searched for CD8+ cytotoxic T lymphocytes in patients with LA. CD8+ T cell lines were generated from peripheral blood and synovial fluid of five patients with LA. In addition, CD8+ T cells were expanded by Ag-specific stimulation in bulk cultures. A cytotoxicity assay was established using target cells infected with recombinant vaccinia viruses expressing the borrelial proteins outer surface protein (Osp) A, OspB, or flagellin. We found Bb-specific CTL lines derived from the peripheral blood of three patients with LA with specificity for flagellin, OspA, and OspB. All Bb-specific CTL lines were CD3+, CD8+, and TCRalphabeta, and cytotoxic activity was HLA class I restricted. Moreover, CD8+ T cells expanded by Ag-specific stimulation in vitro demonstrated Bb-specific and HLA class I-restricted lysis toward individual borrelial proteins. Interestingly, Bb-specific lytic activity was only detected in patient samples obtained after the disappearance of arthritis. We report the detection of Bb-specific cytotoxic CD8+ T cells in patients with LA. The induction of specific CD8+ T cells may play an important role in disease control and may have important bearings for the development of effective vaccines against Lyme borreliosis.     

IL-6, IFN-gamma and TNF-alpha production by liver-associated T cells and acute liver injury in rats administered concanavalin A

The relationship between the development of acute hepatitis and the production of TNF-alpha IFN-gamma and IL-6 by liver-associated T lymphocytes following intravenous injection of concanavalin A (Con A) was studied in rats. Following a single injection of Con A, there was a dose and time-dependent correlation in the serum levels of serum alanine aminotransferase (ALT), IL-6, IFN-gamma and TNF-alpha. These increases correlated with an increase in the numbers of CD4+, CD8+ and CD25+ T cells in blood and CD4+ and CD25+ T cells in the liver perfusate, but not with CD8+ T cells in liver perfusate. Increased levels of IL-6, IFN-gamma and TNF-alpha were constitutively produced by liver-associated CD4+ T cells when cultured. In Con A-stimulated cultures, liver-associated CD4+ T cells secreted increasing levels of TNF-alpha in a time-dependent manner following Con A injection, but TNF-alpha production by peripheral blood lymphocytes was transient with peak levels detected at 1 h which then declined over 24 h. Histological examination of the liver revealed fatty change, hepatocyte degeneration and necrosis, with an associated cell infiltrate of neutrophils and CD4+ T cells both in the portal areas and around the central veins. These results support the hypothesis that Con A-induced liver damage is mediated by CD4+ T cells acting within the liver, at least in part through the secretion of TNF-alpha, IFN-gamma and IL-6.     

Interactions of lymphocytes from patients with psoriatic arthritis or healthy controls and cultured endothelial cells

Psoriatic arthritis (PA) is an inflammatory rheumatic disease that can concomitantly occur in patients with psoriasis vulgaris. Psoriatic synovitis shows alterations of the synovial microvasculature. Inflammatory cells adhere to endothelial cells (EC) and migrate through the vascular wall of postcapillary venules located in the subintimal layer of the synovial membrane. The aim of our study was to investigate, first, the phenotype of lymphocytes (LC) of PA patients using flow cytometry (FC) with regard to activation antigens and adhesion molecules; second, the adhesion of LC of PA patients on cultivated resting or activated (with thrombin, LPS, IFN-gamma, or TNF-alpha) human umbilical vein endothelial cells (HUVEC) by counting the Feulgen-stained nuclei of both adherent LC and HUVEC using image analysis; and third, the synthesis of IL-6 and IL-8 in both LC and HUVEC 24 hr after cell contact. These cytokines were determined qualitatively by immunofluorescence and quantitatively at the single-cell level by FC as well as in the supernatants of the cultures using commercial cytokine ELISAs. Fourth, we investigated whether or not the LC adhesion on HUVEC as well as the cytokine production could be inhibited by monoclonal antibodies against LC- or EC-specific adhesion molecules. In contrast to controls PA patients showed an increased surface expression of CD11a, b, and c as well as of CD44 but a reduced surface expression of CD49d/CD29, and CD49e/CD29, and cell-bound fibronectin on CD3+ LC. The activation markers CD25 and HLA-DR were found to be slightly enhanced in PA. The cell adhesion was generally enhanced in PA patients vs controls. It could be reduced with monoclonal antibodies (MoAbs) against CD11a and CD18 on IFN-gamma- or TNF-alpha-activated HUVEC but was generally enhanced after treatment of HUVEC with MoAbs against CD54, CD62E, or CD106. Due to LC adhesion on HUVEC IL-6 and IL-8 were produced in significantly higher amounts in PA patients compared to controls. This effect occurred already in resting but was enhanced in activated HUVEC. While IL-6 is mainly produced by HUVEC but also in smaller quantities by LC, IL-8 is synthesized only by HUVEC and could be modified by preincubation with MoAbs against LC- or EC-specific adhesion molecules in parallel to the cell adhesion. The experiments show that the main adhesion pathway in LC homing of PA patients is the interaction of the LC adhesion molecule CD11a/CD18 with CD54 on EC followed by an enhanced synthesis of proinflammatory and chemotactic cytokines. These results favor the hypothesis that the pathological alterations of the microvasculature in PA patients are generated by altered homing processes.     

RANTES expression and contribution to monocyte chemotaxis in arthritis

Rheumatoid arthritis (RA) is characterized by recruitment of leukocytes from the vasculature into inflamed synovial tissue (ST) and synovial fluid (SF), which depends, in part, upon the continued maintenance of chemotactic stimuli. RANTES is a potent chemoattractant for leukocytes including monocytes and CD45RO+ memory T lymphocytes. The aim of this study was to determine the production, the source, and the function of antigenic RANTES in arthritis. We detected antigenic RANTES in SFs from RA and OA patients (100 +/- 22.7 and 72 +/- 30.7 pg/ml, respectively). CM from RA ST fibroblasts stimulated with interleukin-1beta or tumor necrosis factor-alpha contained significantly more antigenic RANTES than unstimulated CM (452 +/- 181.6 and 581 +/- 200.2 pg/ml, respectively, versus 12 +/- 4.4 pg/ml, P < 0.05). PHA-stimulated RA SF mononuclear cells secreted 5- to 15-fold more antigenic RANTES than did nonstimulated mononuclear cells, while LPS induced secretion up to 4-fold. We immunolocalized antigenic RANTES to sublining macrophages (28 +/- 3.7 and 8 +/- 2.0% immunopositive cells), perivascular macrophages (56 +/- 6.9 and 19 +/- 3.4%), and synovial lining cells (37 +/- 5.8 and 60 +/- 10.4%) in RA and OA tissue, respectively. Anti-RANTES neutralized 20.2 +/- 1.3% of the RA SF chemotactic activity for normal peripheral blood monocytes (P < 0.05). These results demonstrate antigenic RANTES in RA and OA ST and SF and identify RANTES as a chemoattractant for monocytes in the RA joint.     

Imbalance between interstitial collagenase and tissue inhibitor of metalloproteinases 1 in synoviocytes and fibroblasts upon direct contact with stimulated T lymphocytes: involvement of membrane-associated cytokines

OBJECTIVE: To determine whether direct cell-cell contact with stimulated T lymphocytes (a) differentially modulates the production of interstitial collagenase (matrix metalloproteinase 1 [MMP-1]) and tissue inhibitor of metalloproteinases 1 (TIMP-1) on human synoviocytes and dermal fibroblasts, and (b) induces the production of prostaglandin E2 (PGE2); and to identify the membrane-associated factors on T cell surfaces involved in these mechanisms. METHODS: Dermal fibroblasts and fibroblast-like synovial cells (synoviocytes) were cultured with fixed T cells, isolated plasma membranes from T cells, interleukin-1beta (IL-1beta; 250 pg/ml), or transforming growth factor beta (TGFbeta; 5 ng/ml). Culture supernatants were assayed for the production of MMP-1, TIMP-1, and PGE2. The expression of MMP-1 and TIMP-1 messenger RNA was analyzed by Northern blot of total fibroblast RNA. RESULTS: Membranes of stimulated T cells, i.e., human peripheral blood T lymphocytes (PBTL) and the human T cell line HUT-78, induced the production of PGE2 and MMP-1 on both synoviocytes and dermal fibroblasts. TIMP-1 production was enhanced upon contact with PBTL stimulated for short periods of time (2-4 hours) but not for longer periods. Similar results were obtained with CD4+ and CD8+ synovial tissue T cell clones (TCCs), which induced the production of TIMP-1 by fibroblasts when stimulated for short (2-4 hours), but not long, periods of time. This time dependency was not observed with HUT-78 cells. The production of MMP-1 by fibroblasts and synoviocytes upon cellular contact with stimulated T cells was higher than that induced by an optimum concentration of IL-1beta, whereas the production of PGE2 was equivalent or slightly lower. Cell membrane-associated IL-1alpha and tumor necrosis factor a, but not CD69, CD40 ligand, or CD11b, were involved in the induction of MMP-1 and PGE2 production, as shown by blockade experiments using monoclonal antibodies and cytokine antagonists. CONCLUSION: Synovial tissue TCCs and PBTL stimulated for long periods of time trigger the production of PGE2 and MMP-1, but not TIMP-1, in synoviocytes and dermal fibroblasts, thus inducing an imbalance between the metalloenzyme and its inhibitor. These results demonstrate that T cells may affect fibroblast and synoviocyte functions directly (i.e., by contact activation) and indirectly (i.e., by activation of cytokine production in monocyte/macrophages, which in turn, trigger stromal cell functions). Since the production of MMPs in monocyte/macrophages is also induced upon contact with stimulated T cells, our results strongly suggest that contact of synovial cells with chronically stimulated T lymphocytes favors matrix catabolism. By analogy, this mechanism may trigger tissue destruction in vivo and, thus, may potentiate tissue destruction in chronic inflammatory diseases such as RA.     

Evidence for selective in vivo expansion of synovial tissue-infiltrating CD4+ CD45RO+ T lymphocytes on the basis of CDR3 diversity

In this study we have analyzed the TCR V alpha and V beta regions at the DNA level in the CD4+CD45RO+ memory T cell population of synovial tissue infiltrating T lymphocytes of three rheumatoid arthritis (RA) patients and one patient with chronic arthritis. Cell lines of CD4+CD45RO+, CD4+CD45RO-, CD8+CD45RO+ and CD8+CD45RO- T lymphocyte populations were generated following FACS cell sorting of freshly isolated synovial tissue mononuclear cell infiltrates (STMC) and of freshly isolated peripheral blood mononuclear cells (PBMC) of these patients. The phenotypic and molecular analyses have revealed the following. (i) The TCR repertoires of tissue infiltrating T lymphocytes in the various subsets were extensive on the basis of TCR V gene family usage. (ii) Furthermore, each patient displayed individual specific TCR V gene expression patterns in the various STMC and PBMC derived T cell subsets. However, the majority of these arthritis patients manifested increased expression of multiple TCR V gene families in the synovial tissue derived CD4+CD45RO+ T cell population when compared with the peripheral blood derived CD4+CD45RO+ subset. Of these gene families, we found enhanced expression of the TCR V alpha 7 and V beta 11 gene segments in synovial tissue to be shared by all four patients analyzed. (iii) Nucleotide sequence analysis of the CDR3 regions of a number of TCR V regions in the CD4+CD45RO+ T cell subsets has revealed that the CDR3 regions comprised within synovial tissue derived TCR V regions differed from those found in peripheral blood derived TCR V regions. These differences in CDR3 diversity might be the consequence of a specific interaction with particular MHC-peptide complexes expressed at the site of inflammation. (iv) The CDR3 region analysis also showed individual specific amino acid motifs within the N-D-N regions of all analyzed TCR V beta genes derived from PBMC as well as STMC.     

IL-5 is required for eosinophil recruitment, crystal deposition, and mononuclear cell recruitment during a pulmonary Cryptococcus neoformans infection in genetically susceptible mice (C57BL/6)

CBA/J (highly resistant), BALB/c (moderately resistant), and C57BL/6 (susceptible) mice displayed three resistance patterns following intratracheal inoculation of Cryptococcus neoformans 52. The inability to clear the infection correlated with the duration of the eosinophil infiltrate in the lungs. The role of IL-5 in promoting the pulmonary eosinophilia and subsequent inflammatory damage in susceptible C57BL/6 mice was investigated. C57BL/6 mice developed a chronic alveolar, peribronchiolar, and perivascular eosinophilia following C. neoformans infection. This resulted in the accumulation of intracellular Charcot-Leyden-like crystals in alveolar macrophages by wk 4 and the extracellular deposition of these crystals in the bronchioles with associated destruction of airway epithelium by wk 6. IL-5 mRNA was expressed in the lungs, and injections of anti-IL-5 mAb prevented eosinophil recruitment and crystal deposition but did not alter cryptococcal clearance. Depletion of CD4+ T cells (but not CD8+) ablated IL-5 production by lung leukocytes in vitro and eosinophil recruitment in vivo. Neutralization of IL-5 also inhibited the recruitment of macrophages, CD8+ T lymphocytes, and B lymphocytes by 47 to 57%. Anti-IL-5 mAb inhibited CD4+ T lymphocyte recruitment by 30% but did not affect neutrophil recruitment. Thus, the development of a chronic eosinophil infiltrate in the lungs of C. neoformans-infected C57BL/6 mice is a nonprotective immune response that causes significant lung pathology. Furthermore, IL-5 promotes the recruitment and activation of eosinophils, resulting in the recruitment of additional macrophages and lymphocytes into the lungs.     

Primary surgical repair of traumatic lacerations of the lacrimal canaliculi

Normal human peritoneal cells collected during elective laparatomy from patients with gallbladder stones without clinically detectable inflammatory changes were characterized phenotypically with immunocytochemical method and flow cytometry, with special attention paid to the presence of memory cells. The responsiveness of normal PCs to mitogen and, specifically, the role of peritoneal macrophages in this process was studied. The peritoneal cells consisted of 45% of monocytes/ macrophages (CD68+), as many as CD2+ T lymphocytes, 8% CD57+ NK and K 2% CD22+ B, cells. The CD4/CD8 ratio was 0.4. The peritoneal cells did not express interleukin-2 (CD25+) and transferrin receptors (CD71+) on their surface. Approximately 49% of the peritoneal cells were class II MHC antigen positive cells. Two per cent of S100+ dendritic cells were found. Flow cytometric two-colour analysis revealed that the majority of peritoneal CD4+ (92.4%) and CD8+ (73.1%) lymphocytes, while only 50.2% of CD4+ and 30.1% CD8+ peripheral blood cells expressed simultaneously the CD45R0 (UCHL1) molecule, which is characteristic to the memory/effector T-cell subpopulation. Peritoneal T lymphocytes responded to the mitogens less than peripheral blood lymphocytes of the same individual. Supplementation of cell culture with anti-macrophage (anti-CD68) and anti-HLA-DR MoAb brought about a dose-dependent decrease of proliferative peritoneal cell response to ConA. The authors conclude that human peritoneal cell population comprises a high proportion of T lymphocytes and macrophages capable of presenting antigens to peritoneal lymphocytes. High prevalence of memory lymphocytes points to the preparedness of these cells to react with invading antigens most likely of gut bacterial origin.     

IL-17 is produced by some proinflammatory Th1/Th0 cells but not by Th2 cells

IL-17 is defined as a proinflammatory cytokine and produced by activated CD4+ T cells. In rheumatoid arthritis synovial tissue, high levels of IL-17 contribute to IL-6 production by synoviocytes. The present study was performed to see whether Th cells that produce IL-17 are associated with the Th1, Th2, or Th0 subset. Thirty-three CD4+, alphabeta+ T cell clones were developed from synovial membranes and synovial fluid of rheumatoid arthritis patients. Thirteen clones were defined as Th1 since they produced IFN-gamma but not IL-4, and four clones were defined as Th0 type that produced both IL-4 and IFN-gamma. Sixteen clones were defined as Th2 since they produced high levels of IL-4 and/or IL-10 but not IFN-gamma. IL-17 was measured in a bioassay, where IL-6 production from synoviocytes was a measurement for IL-17 activity in the presence and absence of blocking anti-IL-17 mAb. Three Th1 clones and two Th0 clones produced IL-17. In contrast, none of the sixteen Th2 clones analyzed produced IL-17. In addition, six Th2 clones were further cultured in conditions that induced a switch to Th1 type. Induction of this Th1 phenotype also led to production of IL-17 in two of these clones. The results demonstrate that some cells of the Th1/Th0 phenotype produce IL-17 but not cells of the Th2 phenotype. Thus, IL-17 may define a new subset of T cells, and IL-17 production appears to be a mechanism for Th1/Th0 cells, the most frequent Th subtype present in the rheumatoid synovium, to contribute to the local inflammatory reactions.