The CXCL13/CXCR5-chemokine axis in neuroinflammation: evidence of CXCR5+CD4 T cell recruitment to CSF

Abstract

Abstract Background C-X-C chemokine ligand 13 (CXCL13) is frequently elevated in cerebrospinal fluid (CSF) in a variety of inflammatory central nervous system (CNS) diseases, has been detected in meningeal B cell aggregates in brain tissues of multiple sclerosis patients, and proposedly recruits B cells into the inflamed CNS. Besides B cells also follicular helper T (Tfh) cells express the cognate receptor C-X-C chemokine receptor type 5 (CXCR5) and follow CXCL13 gradients in lymphoid tissues. These highly specialized B cell helper T cells are indispensable for B cell responses to infection and vaccination and involved in autoimmune diseases. Phenotypically and functionally related circulating CXCR5+CD4 T cells occur in blood. Their co-recruitment to the inflamed CSF is feasible but unresolved. Methods We approached this question with a retrospective study including data of all patients between 2017 and 2019 of whom immune phenotyping data of CXCR5 expression and CSF CXCL13 concentrations were available. Discharge diagnoses and CSF laboratory parameters were retrieved from records. Patients were categorized as pyogenic/aseptic meningoencephalitis (ME, n = 29), neuroimmunological diseases (NIMM, n = 22), and non-inflammatory neurological diseases (NIND, n = 6). ANOVA models and Spearman’s Rank-Order correlation were used for group comparisons and associations of CXCL13 levels with immune phenotyping data. Results In fact, intrathecal CXCL13 elevations strongly correlated with CXCR5+CD4 T cell frequencies in the total cohort (p < 0.0001, r = 0.59), and ME (p = 0.003, r = 0.54) and NIMM (p = 0.043, r = 0.44) patients. Moreover, the ratio of CSF-to-peripheral blood (CSF/PB) frequencies of CXCR5+CD4 T cells strongly correlated with CXCL13 levels both in the total cohort (p = 0.001, r = 0.45) and ME subgroup (p = 0.005, r = 0.50), indicating selective accumulation. ME, NIMM and NIND groups differed with regard to CSF cell counts, albumin quotient, intrathecal IgG, CXCL13 elevations and CXCR5+CD4 T cells, which were higher in inflammatory subgroups. Conclusion The observed link between intrathecal CXCL13 elevations and CXCR5+CD4 T cell frequencies does not prove but suggests recruitment of possible professional B cell helpers to the inflamed CSF. This highlights CSF CXCR5+CD4 T cells a key target and potential missing link to the poorly understood phenomenon of intrathecal B cell and antibody responses with relevance for infection control, chronic inflammation and CNS autoimmunity.