Follicular helper and follicular cytotoxic T cells in primary Sjögren’s Syndrome: clues for an abnormal antiviral response as a pathogenic mechanism

Abstract

Introduction: In chronic viral infections, T follicular helper (Tfh) cells are crucial to sustain CD8 cytotoxic T cells in face of the persistence of viral antigens. Moreover, the usual Th1 profile that is associated with antiviral responses in acute infections is replaced by a Tfh profile [1]. Primary Sjögren’s Syndrome (pSS) is a systemic autoimmune disease (SAD) with progressive destruction of exocrine glands. Viral infections have been proposed as etiological agents for pSS, but the connection between an abnormal Tfh-mediated antiviral response and pSS pathophysiology has not yet been established. Furthermore, as far as we know, the recently described T follicular cytotoxic cells (Tcf), also implicated in antiviral responses, have never been assessed in the context of pSS.

Materials and methods: Five groups of patients were enrolled for this study: pSS patients classified according to the AECG criteria (n = 57), sicca syndrome patients with (n = 38) and without (n = 30) criteria for undifferentiated connective tissue disease (UCTD), patients with rheumatoid arthritis (RA; n = 20) and healthy controls (HC; n = 24). Peripheral blood samples from all patients and controls were analysed by 4-color flow cytometry, in a BD FACS Calibur. CD3, CD4, CXCR5, and CCR7 were used for the characterization of Tfh and Tcf cells. Tfh and Tfc were identified as the CXCR5 + CCR7- subset within CD4+ and CD4-(CD8+) T cells, respectively. The expression of IL-21 was also assessed in CD4 and CD8 T cells after cell stimulation with PMA + ionomycin. In the pSS group, EULAR Sjögren’s Syndrome Disease Activity Index (ESSDAI) scale was used to classify patients according to the disease activity. Patients were further divided in two groups: pSS with ESSDAI <5 (low activity) and pSS with ESSDAI ≥5 (moderate to high activity). Normality of cell subset data was assessed using D’Agostino & Pearson omnibus normality test. Unpaired Student’s t test with Welch’s correction was applied to compare normal variables while Mann-Whitney test was used for non-normal variables (p < 0.05 considered for both tests).

Results: No differences were observed in pSS patients when compared to all other groups regarding either Tfh and Tcf cells. Nevertheless, IL-21 secreting CD4 T cells were increased in pSS patients compared to: HC (p = 0.0267); sicca patients without UCTD (p = 0.0158); sicca patients with UCTD (p = 0.0456) and RA patients (p = 0.0455). Similarly, IL-21 secreting CD8 T cells were also increased in pSS patients (p = 0.029) compared to HC. No differences were found in Tfh cells and Tcf cells, nor in IL-21 expression by T cells comparing either RA or sicca patients to HC. Interestingly, in pSS patients, Tcf cells were positively correlated with ESSDAI scores, particularly in those patients with longer disease duration (r = 0.430; p = 0.029). Considering disease activity, pSS patients with ESSDAI ≥5 (n = 8) presented a significantly increased percentage of Tcf cells (p = 0.0103). No other correlations between ESSDAI scores and IL-21-secreting T cells were observed.

Discussion and conclusions: The increase in cells secreting IL-21, the signature cytokine for follicular T cells, seems to be a feature of pSS, not observed in RA nor in sicca syndrome. On the other hand, our data suggest Tcf cells are related to the disease activity in pSS, eventually highlighting their implication in the disease. Future studies approaching follicular T cell subsets, relevant in both autoimmune and antiviral responses could bring new insights for the comprehension of autoimmune diseases such as pSS.