Interleukin-6 Signaling Blockade Induces Regulatory Plasmablasts in Neuromyelitis Optica Spectrum Disorder

Abstract

Background and objectives: Interleukin-6 receptor antibodies (IL-6R Abs), including satralizumab, are increasingly used to prevent relapse for neuromyelitis optica spectrum disorder (NMOSD). However, the detailed mechanism of action of this treatment on the lymphocyte phenotype remains unclear. This study focused on B cells in patients with NMOSD, hypothesizing that IL-6R Ab enables B cells to acquire regulatory functions by producing the anti-inflammatory cytokine IL-10.

Methods: Peripheral blood mononuclear cells were stimulated in vitro to induce the expansion of B-cell subsets, double-negative B cells (DNs; CD19⁺ IgD^-, CD27^-) and plasmablasts (PBs; CD19⁺, CD27^hi, CD38^hi). Whole B cells, DNs, or PBs were isolated after culture with IL-6R Ab, and IL-10 expression was quantified using quantitative PCR and a cytometric bead array. RNA sequencing was performed to identify the marker of regulatory PBs induced by IL-6R Ab.

Results: DNs and PBs were observed to expand in patients with NMSOD during the acute attacks. In the in vitro model, IL-6R Ab increased IL-10 expression in B cells. Notably, IL-10 expression increased in PBs but not in DNs. Using RNA sequencing, CD200 was identified as a marker of regulatory PBs among the differentially expressed upregulated genes. CD200⁺ PBs produced more IL-10 than CD200^- PBs. Furthermore, patients with NMOSD who received satralizumab had a higher proportion of CD200⁺ PBs than patients during the acute attacks.

Discussion: Treatment with IL-6 signaling blockade elicited a regulatory phenotype in B cells and PBs. CD200⁺ PBs may be a marker of treatment responsiveness in the context of NMOSD pathophysiology.

Figure 1. Increased Proportion of DNs and PBs in the Acute Phase of NMOSD Attacks and Their Changes Over Time

(A) Left: Representative flow cytometry analysis of DNs (CD3⁻, CD19⁺, IgD⁻, CD27⁻) from healthy controls (HC) and patients with NMOSD. The numbers in the areas represent percentages of the population. Right: Proportion of DNs. The open and filled symbols represent PBMC data obtained from 23 HC and 19 patients with acute NMOSD attacks, respectively (****p < 0.0001; Mann-Whitney test, the bar charts represent mean ± SEM). (B) Left: Representative flow cytometry analysis of PBs (CD3⁻, CD19⁺, CD27^hi, and CD38^hi) from HC and patients with NMOSD. Right: Proportion of PBs from 23 HC and 19 patients with NMOSD (*p < 0.05; Mann-Whitney test). (C, D) Longitudinal analysis of DNs and PBs. Dashed lines indicate that the samples are from the same patient. The mean interval between attack and remission #1 was 11.3 months (range 5–33). Sampling intervals for remissions #1, #2, and #3 were generally set at 6 months (**p < 0.01, *p < 0.05; fixed-effect p value, mixed-effect model; attack, n = 19; remission #1, n = 21; remission #2, n = 15; remission #3, n = 10). For cases experiencing relapse during the observation period, data acquired before the relapse were excluded because these showed an increase in PBs during the relapse compared with before, aligning the data with the attack as the first time point. DNs = double-negative B cells; HC = healthy controls; NMOSD = neuromyelitis optica spectrum disorder; PBs = plasmablasts. *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001.

Figure 2. In Vitro IL-6R Blockade Induces IL10 Expression in Whole B Cells and PBs But Not in DNs

(A) Representative flow cytometry analysis before and after in vitro stimulation with PBMCs from HC. Both DNs and PBs expand on stimulation. (B) Left: Relative expression of IL10 in B cells from 13 HC and 13 patients with NMOSD, with IL-6R Ab or isotype controls quantified by real-time qPCR (IL-6R Ab vs isotype; **p < 0.01, *p < 0.05; Wilcoxon matched-pairs signed-rank test; HC vs NMOSD; *p < 0.05, Mann-Whitney test). Right: Levels of IL-10 in the culture supernatant measured by the cytometric bead array (**p < 0.01; Wilcoxon matched-pairs signed-rank test). (C) Relative expression of IL10 in DNs and PBs from 11 HC with IL-6R Ab or isotype controls. (D) Proportion of IL-10⁺ cells in induced PBs from PBMCs of 6 HC with IL-6R Ab or isotype controls, which is obtained using the cytokine secretion assay (*p < 0.05; Wilcoxon matched-pairs signed-rank test). Data are from 3 or more independent experiments. ACTB = actin β; DNs = double-negative B cells; HC = healthy controls; IL-10 = interleukin-10; IL-6R Ab = IL-6 receptor antibody; NMOSD = neuromyelitis optica spectrum disorder; PBs = plasmablasts.

Figure 3. Transcriptional Profiles Showing CD200 as a Candidate for Regulatory PBs

(A) Differential expression analysis with RNA-seq data from PBs induced by in vitro stimulation with IL-6R Ab or isotype control. The heatmap shows expression levels of sorted PBs derived from stimulated PBMCs of healthy controls (HC) (n = 4). The color intensity represents the column Z score, with red indicating high expression and blue indicating low expression. The regions outlined in green squares are transcripts upregulated by IL-6R Ab, and their contents are listed on the right. (B) Left: Representative flow cytometry analysis of CD200 expression in circulating B cells and non-B cells of HC. Most B cells express CD200 (75.2%), whereas most non-B cells do not (4.15%). Right: CD200 expression on canonical B-cell subsets. The gray lines indicate negative expression based on samples of fluorescence minus one. (C) Representative flow cytometry analysis and proportions of IL-10–producing cells within CD200⁺ PBs compared with CD200^- PBs based on the cytokine secretion assay (samples from HC, n = 6, *p < 0.05; Wilcoxon matched-pairs signed-rank test). (D) IL-10 and immunoglobulin quantification of culture supernatants from CD200⁺ NAVs and CD200⁺ MEMs from 3 HC. Samples below the detection sensitivity described on the y-axis are marked as 0. DNs = double-negative B cells; HC = healthy controls; IL-10 = interleukin-10; IL-6R Ab = IL-6 receptor antibody; MEMs = memory B cells; NAVs = naïve B cells; PBs = plasmablasts; SWMs = switched memory B cells; USMs = unswitched memory B cells.

Figure 4. CD200+ PBs as Treatment Markers With Satralizumab for NMOSD

(A) Proportion of IL-10⁺ cells in induced PBs from PBMCs of 6 patients with NMOSD with IL-6R Ab or isotype controls, which is obtained using the cytokine secretion assay (*p < 0.05; Wilcoxon matched-pairs signed-rank test). (B) Representative flow cytometry analysis of circulating CD200⁺ cells in PBs. (C) Cumulative data show that PBs from patients with NMOSD receiving satralizumab in the remission phase have a higher percentage of CD200⁺ cells compared with patients during an attack. PBMCs were obtained ≥4 months after initiating satralizumab treatment. Orange dots indicate samples taken <12 months after the most recent attack (HC; n = 28, attack; n = 7, remission on PSL; n = 9, remission on satralizumab; n = 10, **p < 0.01; Kruskal-Wallis test with Dunn multiple comparisons test). (D) Correlation between frequencies of CD200⁺ PBs and age (NMOSD on satralizumab; n = 10, r = −0.7455, *p = 0.0174, HC; n = 28, r = −0.0312, p = 0.8747) (*p < 0.05; Spearman correlation, where r represents the correlation coefficient). HC = healthy controls; NMOSD = neuromyelitis optica spectrum disorder; PBs = plasmablasts; PSL = prednisolone.