We further assessed EBI expression in
We further assessed EBI2 expression in RRMS patients before and during NTZ treatment. A significant increase of EBI2 expression was observed in memory CD4+ T cell subpopulations, but not in CD8+ T cell or in CD19+ B cell subsets of patients undergoing NTZ treatment, with a parallel gain in the migration rate toward 7α,25-OHC compared to before treatment onset. Those changes are likely specific to NTZ, because other DMTs do not elicit such a response and EBI2 expression was not correlated with disease severity or duration. The effects of NTZ treatment on EBI2 could be either direct or indirect. We favor the second postulate, because the sole addition of NTZ on PBMCs was not sufficient to induce EBI2 at the protein level. This is reminiscent of the effect of NTZ on lymphocyte proliferation described previously, in which the increased number of lymphocytes observed in the blood of patients under NTZ treatment did not correlate with an increase in lymphocyte proliferation obtained from PBMCs when NTZ was added in vitro (Kivisäkk et al., 2009). As another option, EBI2 upregulation in memory CD4+ T cells could be driven in a different compartment than blood (such as in the bone marrow or in secondary lymphoid organs), potentially requiring interactions in vivo that are still to be explored. Besides, the increase in EBI2 expression is mirrored by a decrease in CD11a (LFA-1) expression. On this point, the upregulation of one receptor can be potentially related to the downregulation of another through imbalances in recruitment of common adaptor molecules. In addition, EBI2 is mainly co-expressed on memory CD4+ T cells with LFA-1 in untreated RRMS patients, while under NTZ treatment, a significant subset of CD4+ T cells were EBI2+ but LFA-1−. These changes in receptor co-expression might affect the migration of CD4+ T cells and the disease evolution in particular under NTZ treatment. Further investigations are needed to better understand the interplay among integrins, chemokine receptors, and EBI2 during MS and in particular under DMTs such as NTZ. NTZ treatment is associated with an increased risk of progressive multifocal leukoencephalopathy (PML), due to John Cunningham (JC) virus reactivation or infection (McGuigan et al., 2016). Because T cells are important for adrenergic receptors immune surveillance (Dubois et al., 2015), an increase in EBI2 detection and function in the context of NTZ administration could be a compensatory reaction to favor CD4+ T cell homing to the brain and/or cerebrospinal fluid (CSF). It would be interesting to test whether a lack of EBI2 upregulation may affect PML development. However, it is not clear whether this increase in EBI2 expression under NTZ plays a role in the “rebound” effect of disease burden, which often follows NTZ arrest. Brain inflammatory infiltrates in patients with rebound effect are mainly constituted of CD8+ T cells with CD4+ T cells concomitantly enriched in CSF, while during NTZ treatment, CD4+ T cells are dramatically diminished in CSF, suggesting an important role for the latter cells in this compartment (Larochelle et al., 2016, Stüve et al., 2006). EBI2 might be involved in the localization of these cells; therefore, its expression changes under NTZ could have potential effects on disease activity. Previous works have highlighted significant participation for EBI2 in B and Tfh cells to regulate B cell maturation and T-dependent antibody response, with simultaneous modulation of the receptor expression to changes in 7α,25-OHC production by surrounding cells (Hannedouche et al., 2011, Li et al., 2016, Suan et al., 2015, Yi et al., 2012). Therefore, we can hypothesize that there may be EBI2-oxysterol cross-talk between memory CD4+ T cells and the blood brain barrier (BBB). This interaction could be affected by NTZ treatment; thus, T cell migration may be affected. It remains to be unraveled whether an oxysterol gradient can be generated by the BBB, potentially involving 7α,25-OHC production by the endothelium, pericytes, or perivascular space macrophages. The 7α,25-OHC synthesizing enzymes are expressed in blood endothelial cells of lymph nodes (Yi et al., 2012), in macrophages (Preuss et al., 2014), and in astrocytes (Rutkowska et al., 2016), reinforcing a potential role for the BBB in oxysterol synthesis. In the setting of the rebound effect, further investigations are required to better understand the participation of EBI2 in the migration of pathogenic CD4+ T cells beyond the BBB. Because an inhibitor is available with NIBR189, which efficiently blocks EBI2 in human lymphocytes and in NTZ-treated patients, this could open alternative therapeutic opportunities.