Increased DNA and/or RNA content of synovial fluid cells in rheumatoid arthritis: a flow-cytometry study

Pathogenesis of rheumatoid arthritis. The role of T cells and other beasts

The evidence coming from the different experimental approaches reviewed in this article strongly supports the hypothesis that RA is T-cell driven at all stages of the disease. Although the effector phases responsible for the events that lead to joint destruction involve several different cell types, cytokines, and other mediators, T cells still direct operations behind the scenes. Direct experimental proof of this proposition in patients is still lacking, but the development of nondepleting modulating CD4 monoclonal antibodies may provide new tools to test this hypothesis. In this respect, it is encouraging that using one such reagent, we have recently shown that not only did the activity of the disease improve but, more importantly, the inflammatory indices and production of non-T-cell cytokines were reduced. This is not to dissimilar from the results of experiments described in animals, where by blocking synovial T cells, the production of IL-1 beta and TNF alpha could be decreased by more than 90%. From this perspective, it may be predicted that by modulating T cells in the joint, it is possible to achieve our ultimate goal of permanently switching off the disease.

Functional IL-2 receptor beta (CD122) and gamma (CD132) chains are expressed by fibroblast-like synoviocytes: activation by IL-2 stimulates monocyte chemoattractant protein-1 production

The expression of the IL-2R alpha-, beta-, and gamma-chains, CD25, CD122, and CD132, respectively, was investigated on fibroblast-like synoviocytes (FLS) and dermal fibroblasts (DF). Both protein and mRNA for CD122 and CD132 were observed but there was no evidence of CD25 expression. Quantification of the Ag binding sites for CD122 showed that FLS expressed 4 times more receptor molecules than DF. The functional capability of these receptors was confirmed by the production of monocyte chemoattractant protein-1 (MCP-1) in direct response to stimulation by IL-2, which could be inhibited by neutralizing anti-CD122 mAb. Both rheumatoid arthritis (RA) and osteoarthritis (OA) FLS and DF spontaneously produced MCP-1 in culture over a similar range of concentrations. However, RA and OA FLS produced significantly greater levels of MCP-1 following stimulation by IL-2 and IL-1 beta; RA FLS produced significantly more MCP-1 than OA FLS. Addition of exogenous IL-2 caused a slight, but significant, decrease in MCP-1 production by DF. The addition of neutralizing anti-CD122 mAb to FLS cultures partially, but significantly, reduced the IL-2-induced MCP-1 secretion, but did not effect either the spontaneous or IL-1 beta-induced secretion of MCP-1. Increased tyrosine phosphorylation was observed in FLS lysates following 30-min incubation with IL-2. In conclusion, in the inflamed synovium, as activated T cells migrate through the sublining and lining layer, T cell-derived IL-2 may activate FLS to secrete MCP-1, thus recruiting macrophages into the rheumatoid synovium and perpetuating inflammation.

Lack of CD80 expression by fibroblast-like synoviocytes leading to anergy in T lymphocytes

OBJECTIVE

To investigate whether contact with HLA-DR+, but CD80-, fibroblast-like synoviocytes (FLS) in the presence of antigen leads to the induction of anergy in, rather than stimulation of, T cells.

METHODS

Cell surface expression of activation and costimulatory markers on FLS were studied by flow cytometry. Functional changes were investigated by T cell proliferation to tuberculin purified protein derivative or allogeneic responses to FLS, in the presence or absence of DAP3.B7 cells, a human CD80-transfected mouse fibroblast cell line. Induction of anergy was investigated by a 2-stage culture system. T cells were cocultured with allogeneic FLS in the primary culture, rested, and restimulated in the secondary culture by FLS in the presence or absence of DAP3.B7 cells or interleukin-2 (IL-2).

RESULTS

Direct contact between T cells and FLS caused up-regulation of CD69 on T cells and HLA-DR on FLS in both the allogeneic and autologous cultures. The addition of DAP3.B7 cells to FLS-T cell cocultures restored the depressed allogeneic responses of T cells. The allogeneic response by T cells to FLS in the presence of DAP3.B7 cells could be completely inhibited by blocking CD80 with CTLA-4 Ig. Indirect evidence that T cells cocultured with FLS were anergic was the up-regulation of CD25, negligible T cell proliferation, and the restoration of proliferation by the addition of exogenous IL-2. Direct evidence of anergy was obtained when T cells from the primary cultures with FLS remained unresponsive to secondary culture with FLS even in the presence of DAP3.B7 cells. In contrast, primary culture of T cells with FLS plus DAP3.B7 cells initiated a good allogeneic response in all subsequent cultures.

CONCLUSION

It is possible that T cells within the synovium may be anergized by contact with HLA-DR+ CD80- FLS.

Staphylococcal enterotoxin B-specific adhesion of murine splenic T cells to a human endothelial cell line

The presence of a putative autoantigen of autoimmune disorder in a target organ may cause accumulation of specific T cells in the inflammatory region. One of the mechanisms of such accumulation involves the migration of specific-circulating T cells through the endothelial cells into the target lesion. The presence of only a few specific T cells responsive to a putative autoantigen has hampered the investigation of specific migration of circulating T cells to the target organ. We used a superantigen to investigate specific T-cell adhesion to endothelial cells, because it stimulates a large proportion of T cells with particular V beta elements and adhesion of T cells to the endothelium is a vital step in the migration process. Adhesion of murine T cells to the human endothelial cell line, EA.hy926, was specifically increased in the presence of staphylococcal enterotoxin B (SEB). The increase was interferon-gamma (IFN-gamma)-dependent, and consisted mainly of CD4+ T cells. V beta 8.1,2+ T cells preferentially adhered to endothelial cells in the presence of SEB compared with V beta 6+ T cells. Pretreatment of endothelial cells with SEB increased the adherence of V beta 8.1,2+ T cells, while anti-human leucocyte antigen (HLA)-DR and -DQ antibodies inhibited the increased adherence of V beta 8.1,2+ T cells. Our results demonstrate that increased T-cell adhesion to endothelial cells is SEB specific, and that the specificity is dependent on major histocompatibility complex (MHC) class II molecules expressed on endothelial cells and on the recognition of the SEB-MHC class II complex by V beta 8.1,2+ T cells.

Evidence for the continuous recruitment and activation of T cells into the joints of patients with rheumatoid arthritis

Rheumatoid arthritis (RA) synovial fluid (SF) T cells express the activation markers CD69, HLA-DR and very late antigen (VLA)-1, but surprisingly few bear interleukin-2 receptors (CD25). This unusual activation state is commonly assumed to be due to stimulation by local antigen, yet T cells activated in vitro express activation antigens in the clearly defined sequence: CD69, CD25, HLA-DR and finally VLA-1. Two possible explanations for the activation state of SF cells are: first, they comprise several subpopulations each expressing different activation antigens or, second, activation markers are up-regulated by mechanisms other than antigen stimulation. To examine these hypotheses, double- and triple-color immunofluorescence techniques were applied to four T cell populations: normal peripheral blood T cells activated in vitro, RA SF T cells, T cells from an in vivo model of migration [tuberculin purified protein derivative (PPD)-induced skin blisters] and T cells co-cultured with endothelial cells (EC). The results confirmed that in vitro activated T cells expressed activation markers in the sequence described above, with significant CD25 expression and few cells co-expressing CD69 with HLA-DR or VLA-1. In contrast, almost half the SF T cells were CD69+HLA-DR+ but CD25-; a significant minority were CD69+VLA-1+. T cells from PPD-induced skin blisters were already HLA-DR+ and VLA-1+ at 24 h, although, in vitro, PPD-activated T cells up-regulated HLA-DR and VLA-1 only after 1 week, suggesting that pre-activated T cells were preferentially recruited into the blisters. Finally, T cells were found to up-regulate CD69 and, to a lesser extent, HLA-DR after adhering to EC in vitro. In summary, the paradoxical activation state of SF T cells cannot be explained solely by single or multiple rounds of activation in situ. At least two other mechanisms, the preferential recruitment of pre-activated T cells and the induction of HLA-DR and especially CD69 by endothelial contact during migration, may also play a role.

Increased expression of VLA-5 adhesion molecules on synovial fluid T lymphocytes in chronic polyarthritis: a consequence of T-cell activation

This study focuses on the consequences of T-lymphocyte activation in chronic polyarthritis in terms of expression of cell surface receptors interacting with extracellular matrix (ECM). The expression of the VLA group of integrins was studied on in vitro-stimulated peripheral-blood T cells, and on peripheral-blood and synovial-fluid mononuclear cells (MNC) of patients with polyarthritis. The VLA expression was measured by flow cytometry using monoclonal antibodies (MoAbs) against alpha-subunits of the VLA family. VLA-alpha 4 and VLA-alpha 5, but not VLA-alpha 1, were expressed on a major fraction of unstimulated peripheral-blood T cells both in the patients with polyarthritis and in healthy individuals. Two distinct populations, VLA-alpha 5-high and VLA-alpha 5-low, were found in resting peripheral-blood T lymphocytes. Two days after stimulation by phorbol 12-myristate 13-acetate (PMA) and concanavalin A, most T cells became VLA-alpha 5-high. In patients with chronic polyarthritis, the expression of VLA-alpha 1 and VLA-alpha 5 was always higher on synovial-fluid T cells than on peripheral-blood T cells. These results give further support to the hypothesis that upon activation the induction of the VLA adhesion-molecule expression may be a factor contributing to the accumulation of T cells in the inflamed synovium.

T lymphocyte adhesion to fibronectin (FN): a possible mechanism for T cell accumulation in the rheumatoid joint

The accumulation of T cells within the joint is responsible for the perpetuation of synovitis. This process is partly regulated by selective binding to endothelium. However, adhesion to extra-cellular matrix proteins, like FN, may also be important. FN binding is mediated by certain members of the VLA (beta 1 integrin) family of proteins. To investigate the role of Tc-FN interactions in synovitis the binding of synovial fluid (SF) and peripheral blood (PB) T cells to FN-coated wells, and the expression of cell surface VLA molecules on these cells by double label immunofluorescence, were studied. SF T cells bound better to FN than PB T cells. VLA alpha 4 and VLA beta 1 but not VLA alpha 5 were up-regulated on SF compared with PB T cells. Anti-VLA alpha 4, VLA beta 1 and VLA alpha 5 MoAbs inhibited the binding of SF T cells to FN. The increased binding of SF T cells to FN could have been related to activation and/or to their predominantly memory phenotype. Purified resting memory or naive T cells bound poorly to FN. In contrast, compared with SF T cells, concanavalin A-activated T cells showed a very similar level of binding to FN, comparable expression of VLA molecules and the same pattern of inhibition of binding to FN by MoAbs. Thus, VLA molecules may play an important role in the retention of T cells in the joint and since T cells can be activated via VLA-FN interactions, this mechanism may perpetuate chronic inflammation.

The importance of the T cell in initiating and maintaining the chronic synovitis of rheumatoid arthritis

We hope that this short review has produced convincing, although indirect, evidence that in the case of rheumatoid synovitis, the T cell is the conductor of an orchestra playing a tune written by an antigen-presenting cell. The tune develops and becomes modified with time, so that it is difficult to discern with clarity much of the original melody. We believe that it may be possible to substitute a new score that will bring this symphony to a harmonious end. We also strongly believe that this is preferable to augmenting the string section or suppressing the percussion section of the orchestra in the vain hope that in the end harmony can be achieved. The final musical analogy must be the hope that the first page of the score can be found, so that we may discover the main theme, the leitmotif, of the rheumatoid-specific antigenic peptide.

Selective migration of the human helper-inducer memory T cell subset: confirmation by in vivo cellular kinetic studies

The vast majority of T cells present in chronic inflammatory lesions are of the helper-inducer/memory (CD45RO+CD29+) phenotype; suppressor-inducer/naive cells (CD45RA+) are virtually absent. Furthermore, CD4+ T cells are found more frequently than CD8+ cells. Previous in vitro studies have suggested that this may be, in part, due to the increased capacity of CD45RO+CD29+ T cells to bind to endothelium and, thus, enter inflammatory foci but no in vivo evidence for preferential migration exists. To investigate this, suction blisters were generated over a purified protein derivative of tuberculin-induced delayed-type hypersensitivity lesions in humans and the phenotype of the blister cells was studied. At all time points, a preponderance of CD45RO+CD29+ cells over CD45RA+ cells and of CD4 over CD8 cells was demonstrated. Because of the rapid kinetics, this appears to represent preferential migration of these cell types rather than in situ proliferation or phenotype conversion. In addition the expression of the CD45RO but not the CD45RA antigen was up-regulated on blister T cells compared to blood T cells. Analysis of blister fluid showed high concentrations of interleukin 6 but not tumor necrosis factor-alpha or lymphotoxin. This study shows for the first time directly in vivo that CD45RO+ T cells migrate preferentially into inflammatory lesions. Furthermore, it illustrates the potential usefulness of this system in the analysis of the early phases of the immune inflammatory response.

Cellular fibronectin in rheumatoid synovium and synovial fluid: a possible factor contributing to lymphocytic infiltration

Mouse monoclonal antibodies against ED sequence-containing cellular fibronectin (cFn) were used to show that Fn in the inflamed synovium is distinct from the major form of plasma Fn (pFn). An accumulation of cFn was seen at sites of hyperplasia of the rheumatoid synovial membrane and in the walls of small vessels in the synovium by immunofluorescence microscopy. cFn was also found in rheumatoid synovial fluid by immunoblotting. Approximately one-fifth of the T lymphocytes from rheumatoid synovial fluid bound to Fn. The binding of synovial fluid T cells was always higher than that from peripheral blood. These results have two implications. On the one hand, the cellular type of Fn may be an indicator of synovial inflammation. On the other hand, the deposition of Fn may be a factor contributing to the infiltration of mononuclear cells into the synovium.

Phenotypic analysis of synovial tissue and peripheral blood lymphocytes isolated from patients with rheumatoid arthritis

Cytofluorometric analysis was performed to characterize the surface phenotype and activation status of freshly isolated synovial tissue lymphocytes (STL) and peripheral blood lymphocytes (PBL) from 7 patients with rheumatoid arthritis (RA). Proliferative synovium was enzymatically digested to obtain tissue-derived lymphocytes. Indirect immunofluorescent staining of patient PBL and STL with a variety of monoclonal antibodies failed to reveal a consistent alteration in the number of CD4+ (helper/inducer) PBL or STL. However, there was a significant decrease in the number of CD8+ (suppressor/cytotoxic) cells in rheumatoid STL (P less than 0.05). A significant reduction in the density of the T cell differentiation antigens CD3 and CD4 was observed in RA PBL and STL, compared with control PBL. These differences in antigen density were not seen when normal PBL were subjected to the same enzymatic digestion. Both RA PBL and STL manifested increased expression of HLA-DR antigens, without augmentation of interleukin-2 receptor expression. Alterations in the expression of the T cell differentiation antigens and activation antigens by patient PBL closely paralleled the abnormalities observed in STL. In contrast, STL of patients with RA exhibited an increase in the expression of the adhesion-related glycoproteins (leukocyte function-associated 1 [LFA-1] and very late activation 1 [VLA-1] antigens), not observed with autologous PBL. These studies demonstrate that lymphocytes isolated from the synovial tissues of RA patients bear an activated phenotype, exemplified by the modulation of CD3 and CD4 and the expression of HLA-DR.(ABSTRACT TRUNCATED AT 250 WORDS)

The preferential accumulation of helper-inducer T lymphocytes in inflammatory lesions: evidence for regulation by selective endothelial and homotypic adhesion

The mechanisms which lead to the accumulation of T lymphocytes into inflammatory lesions are not clearly understood. We have previously shown that synovial CD4 T lymphocytes are mostly CDw29+UCHL1+ (helper-inducer cells) and very few carry the CD45R antigen which identifies the suppressor-inducer subset. Synovial CD8+ cells are also CDw29+UCHL1+CD45R-. In the present study, lymphocytes from pleural and peritoneal inflammatory infiltrates were shown to have a similar phenotypic pattern. Furthermore, it was demonstrated that the CDw29+UCHL1+ subset had a greater ability than CD45R+ cells to adhere to endothelial cells and to form homotypic clusters. Differential surface expression of LFA-1 on the two subsets was also shown, but this could not account for the demonstrated adhesion differences. Differences in adhesion between CDw29+/UCHL1+ and CD45R+ cells may explain the preferential accumulation of CDw29+/UCHL1+ cells in inflammatory infiltrates and underlie some of the functional differences between cells taken from sites of chronic inflammation and those from peripheral blood.

Abnormal helper-inducer/suppressor-inducer T-cell subset distribution and T-cell activation status are common to all types of chronic synovitis

We have previously shown that rheumatoid synovial T cells are virtually all helper-inducer (CD4+4B4+UCHL1+) rather than suppressor-inducer (CD4+2H4+) cells. CD8 cells were also largely 4B4+. In addition, the majority of T cells were HLA-DR+. To investigate whether these findings were specific for rheumatoid disease, we studied the prevalence of these markers in a variety of chronic inflammatory arthropathies such as ankylosing spondylitis, Reiter's syndrome, and psoriatic arthritis. Again, almost 90% of the T cells were 4B4+UCHL1+ and only 11% were 2H4+; 50% expressed the HLA DR antigen. Thus this phenotypic distribution represents a final common pathway of chronic synovitis and may help to explain the immunopathology of the lesion.

Characterization of a T-cell subset prevalent in immunoregulatory disorders in humans

T cells from individuals with certain autoimmune diseases (rheumatoid arthritis, graft-versus-host disease, acquired immunodeficiency syndrome) express high levels of a cell surface sialoglycoprotein with a molecular weight of 140 kDa (gp140). Although a low frequency of gp 140+ T cells was detected in the blood of normal individuals, upon stimulation with autologous EBV-transformed B cells (AMLR), the frequency of expression of gp140 was increased threefold. To further characterize gp 140+ T cells, rosetting techniques with ox erythrocytes coated with monoclonal anti-gp 140 antibody were used to isolate T-cell subsets for phenotypic, cell cycle, and functional analysis. The majority of gp140+ T cells expressed cytotoxic/suppressor (CD8+) phenotype in both normal and AMLR-activated states. Unstimulated gp140+ T cells had significantly greater nucleic acid content, as measured by acridine orange and flow cytometry, than gp140- T cells. Surprisingly, the gp140+ T-cell subset had a less proliferative response in vitro to pokeweed or phytohemaglutinin mitogens. These results suggest that gp140+ T cells in normal individuals and in patients with autoimmune diseases may have been activated previously in vivo and that they are relatively resistant to reactivation in vitro.