Background Follicular helper T cells (Tfh), a CD4 T helper subset localized in lymphoid organs, help B cell differentiation and function. Circulating CD4 T cells expressing CXCR5 together with ICOS and/or PD-1 are considered as counterparts of Tfh, can function as B cell helpers, and can be subdivided into three subpopulations based on the expression of CCR6 and/or CXCR3: CXCR5+CXCR3+CCR6- (Tfh-Th1), CXCR5+CXCR3-CCR6+ (Tfh-Th17) and CXCR5+CXCR3-CCR6- (Tfh-Th2). Only Tfh-Th17 and Tfh-Th2, as opposed to Tfh-Th1, seem to display functional properties of Tfh cells. An altered proportion of these subpopulations has been associated with autoimmune diseases.
Objectives To study the frequency of circulating Tfh and Tfh cell subsets together with the frequency of circulating plasmablasts (CD19+CD20-CD27+CD38high B cells) in patients with Ankylosing Spondilytis (AS).
Methods Peripheral blood was drawn from healthy controls (n=50), AS patients receiving NSAIDs (n=25) and AS patients treated with TNF blocking agents (n=25). After isolation by Ficoll-Hypaque gradient, PBMCs were stained with antibodies to CD3, CD4, CXCR5, ICOS, PD-1, CCR6, CXCR3, CD19, CD20, CD27, and CD38, and examined by flow cytometry. The percentages of CXCR5+CXCR3+ CCR6- (Tfh-Th1), CXCR5+CXCR3- CCR6+ (Tfh-Th17) and CXCR5+CXCR3-CCR6- (Tfh/Th2) cells were calculated after gating for CD3, CD4 and CXCR5+. The percentage of CD20-CD38high cells was calculated after gating for CD19+ and CD27+.
Results The frequency of circulating CXCR5+ cells gated for CD4+ T cells was not different among the studied groups. In contrast, AS patients receiving NSAIDs but not those on TNF blockers, demonstrated a decreased frequency of CD4+CXCR5+ICOS+ PD-1+ cells. Furthermore, in AS patients receiving NSAIDs, the frequency of Tfh-Th1 cells was significantly increased and the frequency of Tfh-Th17 and Tfh-Th2 cells was significantly decreased as compared with controls. Subsequently, the ratio (Tfh-Th17+Tfh-Th2)/Tfh-Th1 was decreased in these patients. At the same time, the frequency of circulating plasmablasts was decreased in AS patients on NSAIDs. Interestingly, there was a significantly positive correlation between the percentage of circulating plasmablasts and the frequency of CXCR5+ICOS+PD-1+CD4+ T cells. In addition, the frequency of circulating plasmablasts showed a positive correlation with the ratio (Tfh-Th17+Tfh-Th2)/Tfh-Th1.
Conclusions AS patients treated with NSAIDS, but not those receiving TNF blockers, demonstrate a decreased frequency of circulating Tfh counterparts (CXCR5+ICOS+PD-1+CD4+ T cells) and altered proportions of circulating Tfh subpopulations, with underrepresentation of subsets bearing a phenotype associated with B cell helping capacity. At the same time, a decreased proportion of circulating plasmablasts is apparent in AS patients treated with NSAIDs.
Disclosure of Interest None declared