Down-regulation of the T cell receptor CD3 zeta chain in rheumatoid arthritis (RA) and its influence on T cell responsiveness

Abstract

T cells implicated in chronic inflammatory diseases such as RA respond weakly when stimulated in vitro with mitogen or antigen. The mechanism behind this hyporesponsiveness is unclear, but a depressed expression of the T cell receptor (TCR)-associated CD3zeta chain has been suggested. In the present work we describe a low expression of CD3zeta in synovial fluid (SF) T cells from RA patients compared with peripheral blood (PB) T cells, but no difference in CD3zeta expression between RA and healthy control PB T cells. In vitro studies demonstrated that granulocytes but not SF macrophages are able to down-regulate the expression of CD3zeta. Through stimulation with anti-CD3 antibodies we demonstrated that the TCR-dependent proliferative response was decreased in SF T cells compared with PB T cells. Stimulation with phorbol ester and ionomycin also resulted in a low proliferative response of SF T cells, indicating that both signal transduction through the TCR (stimulation with anti-CD3) and events further downstream in the signalling pathways (stimulation with phorbol ester and ionomycin) are affected. A similar depression of T cell activity was observed when induction of IL-2 and IL-4 was measured. However, SF T cells were not defective in the induction of interferon-gamma (IFN-gamma) when stimulated with phorbol myristate acetate (PMA)/ionomycin, in contrast to the diminished IFN-gamma response observed after stimulation with anti-CD3. This indicates that the hyporesponsiveness of SF T cells can not be generalized to all T cell functions. The differential response to external stimuli is likely to be of importance for the capacity of SF T cells to influence inflammatory reactions.

Fig. 1

Down-regulated CD3ζ expression in RA synovial fluid mononuclear cells (SFMC) but not in RA peripheral blood mononuclear cells (PBMC). Expression of CD3ζ was analysed by intracellular cytometric staining on density-separated PBMC and SFMC from 19 RA patients and PBMC from 16 healthy controls (HC). Median fluorescence intensity of CD3ζ staining was analysed in cells expressing CD3ε (defining all T cells), CD4 or CD8.

Fig. 2

Down‐regulation of CD3ζ in T cells by granulocytes but not by macrophages. Purified RA peripheral blood (PB) T cells were coincubated with autologous synovial fluid (SF) macrophages or autologous SF granulocytes overnight and the T cell expression of CD3ζ was measured. One representative experiment of three is illustrated.

Fig. 3

Defective proliferative response of RA synovial fluid (SF) T cells but not of RA peripheral blood (PB) T cells. Proliferation in response to (a) anti-CD3 stimulation and (b) phorbol 12-myristate 13-acetate (PMA)/ionomycin stimulation was measured by flow cytometric analysis of incorporated 5-bromo-2′-deoxyuridine (BrdU) in combination with cell surface staining of CD3ε, CD4 or CD8. The percentage of proliferating cells was compared between RA PB (n = 12) and healthy control (HC) PB (n = 10) and between RA PB and RA SF (n = 11), respectively.

Fig. 4

Cytokine concentration in supernatants after stimulation of peripheral blood mononuclear cells (PBMC) or synovial fluid mononuclear cells (SFMC) with (a) anti-CD3 or (b) phorbol 12-myristate 13-acetate (PMA)/ionomycin. Cells were incubated for 24 h (IL-4) or 3 days (IFN-γ, IL-2 and IL-10) before collection of supernatants. Analyses were performed by cytokine-specific ELISAs. Comparisons were made between RA PBMC (n = 12) and healthy control (HC) PBMC (n = 10) or between RA PBMC and SFMC (n = 11), respectively. IL-2 secretion after CD3 stimulation could not be detected.

Fig. 5

Flow cytometric analysis of IFN-γ-producing cells in RA peripheral blood mononuclear cells (PBMC) and synovial fluid mononuclear cells (SFMC). Cells were stimulated with or without phorbol 12-myristate 13-acetate (PMA)/ionomycin overnight in the presence of brefeldin A. Flow cytometry triple staining was performed by combining intracellular staining of IFN-γ with cell surface staining of CD3 and CD94. Thin and thick lines represent unstimulated and PMA/ionomycin-stimulated cells, respectively. Dot plots of gated IFN-γ⁺ (black) and IFN-γ⁻ (red) cells after stimulation with PMA/ionomycin are indicated below. One representative experiment of five is illustrated.

Fig. 6

Identification of IFN-γ-producing cells after phorbol 12-myristate 13-acetate (PMA)/ionomycin stimulation. RA peripheral blood mononuclear cells (PBMC; n = 5), synovial fluid mononuclear cells (SFMC; n = 6) and healthy control (HC) PBMC (n = 5) were stimulated with PMA/ionomycin overnight in the presence of brefeldin A. Intracellular staining of IFN-γ combined with surface staining of either CD3 and CD94 or CD4 and CD8 was performed. (a) Proportion of IFN-γ-expressing cells in different subsets of T cells and natural killer (NK) cells. Gates were set for T cells (CD3⁺CD94⁻, CD3⁺CD94⁺, CD4⁺ or CD8⁺) and NK cells (CD3⁻CD94⁺), and the percentage of IFN-γ⁺ cells within each cell population was calculated. (b) Gates were set for IFN-γ⁺ cells and the distribution of the cytokine-expressing cells among T cells (CD3⁺CD94⁻, CD3⁺CD94⁺, CD4⁺ or CD8⁺) and NK cells (CD3⁻CD94⁺) was calculated (median values).