Zabalza et al. Journal of Neuroinflammation (2020) 17:161 https://doi.org/10.1186/s12974-020-01840-2 SHORT REPORT Open Access Impact of cytomegalovirus infection on B cell differentiation and cytokine production in multiple sclerosis Ana Zabalza1,2, Andrea Vera1, Elisenda Alari-Pahissa3, Elvira Munteis1, Antía Moreira1,4, Jose Yélamos5,6, Mireia Llop1, Miguel López-Botet3,5,6 and Jose E. Martínez-Rodríguez1* Abstract Background: Human cytomegalovirus (HCMV) infection has been recently associated with a low risk of multiple sclerosis (MS), yet the basis behind this observation remains uncertain. In this study, we aimed to determine in MS patients whether HCMV induces modifications in the peripheral B cell compartment. Methods: HCMV serostatus was determined in 73 MS patients (55 relapsing-remitting MS (RRMS); 18 progressive MS (PMS)) and 30 healthy controls, assessing their B cell immunophenotype and cytokine production (GM-CSF, IL-6, IL-10, and TNFα) by flow cytometry. Results: HCMV seropositivity in untreated MS patients (n = 45) was associated with reduced switched memory B cells, contrasting with an opposite effect in PMS. Expansions of transitional B cells were observed in HCMV(+) IFNβ- treated RRMS patients but not in HCMV(−) cases (p < 0.01), suggesting that HCMV may influence the distribution of B cell subsets modulating the effects of IFNβ. Considering the B cell functional profile, HCMV(−) PMS displayed an increased secretion of proinflammatory cytokines (IL-6, TNFα) as compared to HCMV(+) PMS and RRMS cases (p < 0.001). Conclusions: Our study reveals an influence of HCMV infection on the phenotype and function of B cells, promoting early differentiation stages in RRMS and reducing the proinflammatory cytokine profile in advanced MS forms, which might be related with the putative protective role of this virus in MS. Keywords: Multiple sclerosis, B cells, Human cytomegalovirus, Interferon-beta Introduction reservoir. In this regard, it is conceivable that efficacy of Multiple sclerosis (MS) is a demyelinating disease of the anti-CD20 therapies might depend not only on suppres- central nervous system classically considered to be medi- sion of B cell functions, but also on the control of EBV in- ated by T cells. A relevant role of B cells in MS pathogen- fection [4]. On the other hand, human cytomegalovirus esis is supported by a growing body of evidences, (HCMV) has been associated with a lower MS susceptibil- particularly the successful therapeutic results using anti- ity based on seroepidemiological studies [5, 6], yet the CD20 monoclonal antibodies [1, 2]. Epstein-Barr virus basis for these observations remains poorly understood. (EBV), the main pathogen related to MS [3], establishes a HCMV chronic infection leads to marked changes in persistent infection having memory B cells as the main the immune system [7], mainly characterized by the dif- ferentiation and expansion of specific T cells [8] and adaptive NK cell subsets expressing the CD94/NKG2C * Correspondence: JMartinezR@hospitaldelmar.cat 1 Neurology Department, Hospital del Mar Medical Research Institute (IMIM), activating receptor [9], a phenotypic feature that has Passeig Marítim, 25-29, 08003 Barcelona, Spain been associated with reduced MS progression [10]. In Full list of author information is available at the end of the article © The Author(s). 2020 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Zabalza et al. Journal of Neuroinflammation (2020) 17:161 Page 2 of 8 this study, we evaluated whether HCMV may have an Statistical analysis influence on the B cell compartment in MS patients ac- Kolmogorov-Smirnov test assessed normal distribution, cording to clinical features of the disease. providing means and standard deviations for parametric variables, and medians and interquartile range for non- Methods parametric variables. Values were compared using Stu- Samples dent’s t test for parametric variables or Mann-Whitney U Blood was collected from healthy controls (HC) and MS test for non-parametric, respectively. Pearson or Spear- patients fulfilling McDonald criteria 2017 at the Neur- man correlation indexes were calculated for pair-wise con- ology Department, Hospital del Mar, Barcelona (Spain), tinuous variables. Results were considered significant at excluding cases with relapses or corticosteroid treatment the two-sided level of 0.05. Data analysis was performed in the previous 30 days, severe concomitant disease, using GraphPad Prism 6.0 software (La Jolla, CA). pregnancy, and disease-modifying therapies known to deplete peripheral lymphocytes or alter their trafficking. Results MS forms were defined as relapsing-remitting MS Demographic description (RRMS) and progressive MS (PMS) [11]. We studied 30 HC and 73 MS patients (55 RRMS, 18 PMS: 8 secondary PMS, 10 primary PMS) (Table 1). Herpesvirus serostatus HCMV serology distribution was comparable in controls Standard serological diagnostic tests were performed to and MS patients. No HCMV-related clinical differences evaluate EBV-(LIASON®) and HCMV-specific antibodies were perceived, other than a lower disability score in (Roche®, Cobas602). HCMV(+) PMS as compared to HCMV(−) PMS patients (MSSS, 5.09 ± 2.23 vs.7.48 ± 1.28, p < 0.05). B cell immunophenotype Peripheral blood mononuclear cells (PBMCs) isolated Distribution of B cell subsets in untreated MS according from blood samples using Ficoll-Hypaque density gradient to clinical forms and HCMV serology centrifugation were analyzed by flow cytometry using a Assessment of the B cell immunophenotype in MS pa- direct staining (Fig. 1) with the following fluorochrome- tients without treatment (n = 45) and in HC did not show conjugated monoclonal antibodies: CD19-PE-Cy7 (Invi- significant differences in the distribution of the main B cell trogen), CD38-APC, CD27-PerCP-Cy5.5 (BD Biosciences), subsets (Additional file 1: Table 1). However, stratifying IgD-FITC (Southern Biotech), CD10-PE (Biolegend), and MS cases according to clinical form revealed that RRMS DAPI (Thermo Fisher). B cell subsets were defined gating displayed lower proportions of TB cells together with in- CD19+ lymphocytes as follows: plasmablasts-plasma cells creased SMB cells as compared to HC and PMS, whereas (PB-PC: CD38hiCD10−), transitional (TB: CD38hiCD10+), PMS had lower proportions of UMB (Fig. 2a). Interest- naive (NB: CD38low/negCD10−IgD + CD27−), unswitched ingly, HCMV seropositivity was associated with greater memory (UMB: CD38low/negCD10-IgD + CD27+), proportions of NB and reduced SMB cells in RRMS, con- switched memory (SMB: CD38low/negCD10−IgD−CD27+), trasting with an opposite effect in PMS (Fig. 2b). No dif- and double negative (DN: CD38low/negCD10−IgD−CD27 ferences in the proportions of the different subsets were −). Samples were analyzed in LSR II Fortessa (BD Biosci- noticed in HC stratified according to HCMV serology. ences) using FlowJo (Tree Star) software. These results suggest that HCMV infection may differ- ently influence the B cell compartment distribution in MS Functional analysis of B cells depending on the clinical form. B cell cytokine production was evaluated in MS patients (RRMS, n = 45; PMS, n = 18) and HC (n = 20) incubat- Impact of HCMV on the B cell compartment in IFNβ- ing PBMCs for 24 h with CpG (10 μg/ml), adding in the treated MS patients last 4 h PMA (50 ng/mL) and ionomycin (1 μg/ml, IFNβ therapy in RRMS patients (n = 23) was associated Sigma-Aldrich), monensin (GolgiStop), and brefeldin with increased proportions of TB and NB cells, and a re- (GolgiPlug, BD Biosciences). Samples were stained using ciprocal reduction of UMB, SMB, and DN cells (Fig. 3a), CD19-PE-Cy7 (Invitrogen) and CD3-perCP (BD Biosci- in line with previous studies describing an effect of this ences), permeabilized, fixed, and stained with the follow- cytokine on the redistribution of B cell subsets in MS pa- ing cytokine-specific monoclonal antibodies: anti-GM- tients [12]. Glatiramer acetate treatment (n = 5) was not CSF-APC, anti-IL-6-FITC (Biolegend), anti-IL-10-PE associated with modifications in B cell subsets (Fig. 3a). (BD Biosciences), and anti-TNFα conjugated with the Remarkably, the effect of IFNβ therapy on the propor- CFP fluorochrome. Samples were acquired in LSR II tions of TB cells differed according to HCMV serostatus, Fortessa and data analyzed by FlowJo software, defining observing an increase in treated HCMV(+) MS patients the proportions of B cells stained for specific cytokines. but not in seronegative cases (Fig. 3b). Moreover, IFNβ Zabalza et al. Journal of Neuroinflammation (2020) 17:161 Page 3 of 8 Fig. 1 Gating strategy for B cell subset analysis. Peripheral blood mononuclear cells were gated based on forward (FSC) and side scatter (SSC) (a). After exclusion of DAPI(+) dead cells (b) and selection of CD45(+) lymphocytes (c), CD19+ B cell subsets (d) were defined as plasmablasts-plasma cells (PB-PC: CD38hiCD10−) and transitional (TB: CD38hiCD10+) (e), gating in the CD38lowCD10− B cells (f) to define naive (NB: CD38low/negCD10 −IgD+CD27−), unswitched memory (UMB: CD38low/negCD10−IgD+CD27+), switched memory (SMB: CD38low/negCD10−IgD−CD27+), and double negative (DN: CD38low/negCD10−IgD−CD27−). A representative case is illustrated in the figure, showing proportions of B cell subsets therapy in HCMV(−) RRMS was associated with in- increased GM-CSF by B cells from PMS as compared to creased NB and reduced SMB cells, resembling the effect RRMS patients (Fig. 4b). Considering B cell subsets, IL- of HCMV infection on these B cell subsets described 10 production in MS was directly related to proportions above, which dampened the differences with HCMV(+) of TB cells (R 0.35, p < 0.05) and inversely to SMB cells treated cases (Fig. 3b). These results revealed that (R − 0.39, p < 0.05), whereas TNFα appeared inversely HCMV infection may influence the distribution of B cell related to TB (R − 0.43, p < 0.01) and directly to UMB subsets in MS patients modulating the effects of IFNβ. (R 0.49, p < 0.01). IFNβ therapy, age, and disability scores appeared unrelated with the B cell cytokine pro- Functional implications of HCMV on the cytokine profile duction (data not shown). On the other hand, HCMV(−) of B cells in MS PMS displayed greater IL-6 and TNFα production as com- B cell cytokine production was evaluated as described in pared to HCMV(+) PMS and HCMV(−) RRMS patients, the “Methods” section (Fig. 4a). No differences were contrasting with comparable values of cytokine produc- found between HC and the whole group of MS patients, tion in HCMV(+) cases (Fig. 4c). These results suggest though we observed a greater IL-10 secretion by B cells that HCMV infection in PMS patients may decrease the from RRMS and PMS patients as compared to HC, and production of proinflammatory cytokines by B cells. Zabalza et al. Journal of Neuroinflammation (2020) 17:161 Page 4 of 8 Table 1 Demographic and clinical characteristics of MS patients and controls HC (n = 30) RRMS (n = 55) PMS (n = 18) p value, HC vs MS p value, RRMS vs PMS p value, RRMS vs HC Age, years (mean ± SD) 41.1 ± 11.4 44.7 ± 12.1 60.1 ± 8.0 < 0.01 < 0.01 0.190 Female sex, n (%) 18 (60.0%) 31 (56.40%) 11 (61.10%) 0.498 0.914 0.173 HCMV(+) serology, n (%) 23 (76.7%) 35 (63.6%) 11 (61.1%) 0.133 0.400 0.093 EBV(+) serology, n (%) 24 (85.7%) 54 (98.2%) 17 (94.4%) < 0.05 0.075 < 0.05 MS duration (years) – 10.9 (5.5–16.0) 22.6 (13.9–33.3) – < 0.01 – EDSS – 2.0 (1.0–3.0) 6.0 (4.4–7.0) – < 0.01 – MSSS – 2.08 (1.01–4.17) 6.74 (3.74–7.46) – < 0.01 – 2y-RR – 0.0 (0.0–1.0) 0.0 (0.0–0.0) – < 0.01 – ARR – 0.31 (0.17–0.57) 0.0 (0–0.13) – < 0.01 – DMT None – 27 18 < 0.01 – IFNβ – 23 0 GA – 5 0 Values are expressed as mean ± SD for parametric variables and as median (interquartile range) for nonparametric variables HC healthy controls, RRMS relapsing-remitting MS, PMS progressive MS, HCMV human cytomegalovirus, EBV Epstein-Barr virus, EDSS expanded disability status scale, MSSS Multiple Sclerosis Severity Score, 2y-RR relapse rate in the previous 2 years, ARR annualized relapse rate, DMT disease-modifying therapy, IFNβ interferon-β, GA glatiramer acetate Discussion influence of this viral infection in MS, as suggested by In the present study, we provide data supporting that previous studies describing a low risk for MS in HCMV HCMV infection alters the phenotypic and function of B seropositive cases [5, 6]. Although an epiphenomenon cells in MS, modulating the influence of IFNβ and redu- linked to the exposure to other environmental factors cing the proinflammatory B cell profile. These observa- could not be excluded in the light of these observations, tions may contribute to understand the putative putative mechanisms of heterologous immunity between Fig. 2 B cell subsets in controls and untreated MS patients according to clinical form and HCMV. a Percentages of B cell subsets within total B cells (CD19+) in healthy controls (HC) and MS patients classified based on clinical form. b HC and MS patients according to HCMV serostatus. RRMS: relapsing-remitting MS; PMS: progressive MS; PB-PC: plasmablasts-plasma cells; TB: transitional B cells; NB: naive B cells; UMB: unswitched memory B cells; SMB: memory switched B cells, DN: double negative B cells. p values: *< 0.05, ***< 0.001 Zabalza et al. Journal of Neuroinflammation (2020) 17:161 Page 5 of 8 Fig. 3 Impact of HCMV on the B cell compartment in IFNβ-treated MS patients. a B cell subsets in untreated (n = 27), IFNβ-treated (n = 23), and glatiramer-acetate-treated (n = 5) RRMS patients. b Untreated and IFNβ-treated MS patients classified based on HCMV serostatus. PB-PC: plasmablasts-plasma cells; TB: transitional B cells; NB: naive B cells; UMB: unswitched memory B cells; SMB: memory switched B cells; and DN: double negative B cells. p values: *< 0.05, **< 0.01, ***< 0.001, ****< 0.0001 viruses [13–15] and HCMV-induced modifications on subset distribution, namely the proportions of NB and the immune system may be implicated in the effect of SMB, was perceived when MS form was considered, with HCMV on MS [7, 8, 16, 17]. In this regard, HCMV has an opposite effect in RRMS as compared to PMS pa- been described as an environmental factor inducing tients. The reduced proportions of memory B cells ob- striking effects on the immune system [7]. Nevertheless, served in HCMV(+) RRMS appear consistent with a in contrast to the well-known role of HCMV as a driving protective effect of the viral infection in MS described in force behind the expansion of late-differentiated T lym- previous studies [5, 6]. Intriguingly, no influence of phocytes [8, 18] and adaptive NK cells [9], the influence HCMV was detected in the B cell subset distribution in of chronic HCMV infection on B cells is minor and controls, in line with previous reports of minor HCMV- barely evaluated in previous studies. induced modifications in the B cell compartment as B cells have been implicated in MS pathogenesis, with compared to T cells [18]. In this regard, HCMV-related a special relevance of the memory subset [2, 19]. Of differences in B cell subsets and the functional profile note, no HCMV-related differences in B cell subsets be- observed in our MS patients appeared uncoupled to tween controls and MS patients were found in our study. markers of terminal differentiation of T cells and expan- However, an influence of the viral infection on the B cell sion of adaptive NKG2C+ NK cells (data not shown), Zabalza et al. Journal of Neuroinflammation (2020) 17:161 Page 6 of 8 Fig. 4 B cell cytokine production in controls and MS patients. a Representative cases illustrating the gating strategy and staining of pro- inflammatory (GM-CSF, IL-6, TNFα) and anti-inflammatory cytokines (IL-10) in B cells in resting conditions and stimulated using CpG and PMA/ ionomycin. b B cell cytokine production in controls and MS patients according to MS clinical form. c Impact of HCMV on B cell cytokine production in controls and MS patients classified based on clinical form. p values: *< 0.05, **< 0.01, ***< 0.001 suggesting specific changes on the B cell compartment previously described increase of TB cells associated with induced by HCMV in the setting of MS. IFNβ [12] was only observed in HCMV(+) RRMS patients. Of note, the shift in the B cell subset distribution associ- In view of the implications of B cells in the mechanism of ated with HCMV in our untreated MS patients was rem- action of this cytokine [21], it is reasonable to hypothesize iniscent of the impact of IFNβ therapy in HCMV that HCMV infection in MS patients may modulate the seronegative RRMS patients, an effect that might be hypo- influence of IFNβ on B cell differentiation. thetically related with interferon-signaling pathways in- B cells in MS have been shown to display a pro- duced by persistent HCMV infection [20]. In addition, the inflammatory cytokine profile [22, 23], which was also Zabalza et al. Journal of Neuroinflammation (2020) 17:161 Page 7 of 8 related with MS clinical form when B cells from CSF written informed consent to enter the study and provide peripheral blood were evaluated in PMS cases [24]. In our study, B cells samples. from HCMV(−) PMS patients displayed an enhanced Consent for publication pro-inflammatory profile as compared to HCMV(+) Not applicable. PMS cases, independently of other clinical variables. Al- though limited by the reduced sample size of PMS pa- Competing interests tients evaluated, these results suggest that persistent A. Zabalza has received travel funding from Biogen Idec, Novartis, and HCMV infection may reduce the inflammatory profile of Genzyme. A. Moreira has received travel funding from Teva, Biogen Idec, Novartis, Almirall, Bayer, and Genzyme, and personal fees for lectures from B cells in MS, in line with previous reports describing Genzyme. E. Munteis has received personal fees for consulting services and differences in IL-10 production by B cells from monozy- lectures from Merck-Serono, Biogen Idec, Teva, Genzyme, Novartis, Bayer, and gotic healthy twins discordant for HCMV [7]. Almirall. JE. Martínez-Rodríguez has participated as principal investigator in pharmaceutical company-sponsored clinical trials including Novartis, Roche, Merck-Serono, Actelion, and Celgene, and personal fees for consulting services Conclusions and lectures from Novartis, Biogen Idec, and Merck-Serono. A Vera, E. Alari- Altogether, our study supports that HCMV infection Pahissa, J. Yélamos, M. Llop, and M. López-Botet declare that they have no competing interests. modulates the distribution of B cell subsets and the IFNβ response in MS patients, and associated with a re- Author details 1 duced pro-inflammatory cytokine profile in PMS, thus Neurology Department, Hospital del Mar Medical Research Institute (IMIM), Passeig Marítim, 25-29, 08003 Barcelona, Spain. 2Departament de Medicina, providing mechanistic insights on the putative beneficial Universitat Autònoma de Barcelona, Barcelona, Spain. 3University Pompeu influence of HCMV in MS. Further studies are required Fabra, Barcelona, Spain. 4Neurology Department, Althaia, Xarxa Assistencial i to assess whether HCMV may also modulate the im- Universitària de Manresa, Manresa, Spain. 5Hospital del Mar Medical Research Institute (IMIM), Barcelona, Spain. 6Immunology laboratory, Hospital del Mar, munological response to other MS therapies targeting B Barcelona, Spain. cells [19, 25]. Received: 19 March 2020 Accepted: 11 May 2020 Supplementary information Supplementary information accompanies this paper at https://doi.org/10. 1186/s12974-020-01840-2. References 1. Hauser SL, Bar-Or A, Comi G, Giovannoni G, Hartung H-P, Hemmer B, et al. Ocrelizumab versus interferon beta-1a in relapsing multiple sclerosis. N Engl Additional file 1: Table 1. Proportions of B cell subsets in controls and J Med. 2017;376:221–34. MS patients according to HCMV serostatus. 2. Li R, Patterson KR, Bar-Or A. Reassessing B cell contributions in multiple sclerosis. Nat Immunol. 2018;19:696–707. Abbreviations 3. Ascherio A, Munger KL, Lennette ET, Spiegelman D, Hernán MA, Olek MJ, EBV: Epstein-Barr virus; HCMV: Human cytomegalovirus; IFNβ: Interferon-beta; et al. Epstein-Barr virus antibodies and risk of multiple sclerosis: a MS: Multiple sclerosis; PBMCs: Peripheral blood mononuclear cells; prospective study. JAMA. 2001;286:3083–8. PMS: Progressive multiple sclerosis; RRMS: Relapsing-remitting multiple 4. ’t Hart BA, Jagessar SA, Haanstra K, Verschoor E, Laman JD, Kap YS. The sclerosis primate EAE model points at EBV-infected B cells as a preferential therapy target in multiple sclerosis. Front Immunol 2013;4:145. 5. Waubant E, Mowry EM, Krupp L, Chitnis T, Yeh EA, Kuntz N, et al. Common Acknowledgements viruses associated with lower pediatric multiple sclerosis risk. Neurology. Not applicable. 2011;76:1989–95. 6. Sundqvist E, Bergström T, Daialhosein H, Nyström M, Sundström P, Hillert J, Authors’ contributions et al. Cytomegalovirus seropositivity is negatively associated with multiple AZ, AV, EA-P, AM, and JM-R designed the study, executed experiments, and sclerosis. Mult Scler J. 2014;20:165–73. performed the statistical analysis. EM and ML contributed to the acquisition 7. Brodin P, Jojic V, Gao T, Bhattacharya S, Angel CJL, Furman D, et al. Variation and analysis of data. AZ, EA-P, JY, ML-B, and JM-R contributed to interpretation in the human immune system is largely driven by non-heritable influences. of the results and wrote the final draft that was revised for all authors. ML-B and Cell. 2015;160:37–47. JM-R contributed to data interpretation and critically reviewed the manuscript. 8. Klenerman P, Oxenius A. T cell responses to cytomegalovirus. Nat Rev The authors read and approved the final manuscript. Immunol. 2016;16:367–77. 9. Gumá M, Angulo A, Vilches C, Gómez-Lozano N, Malats N, López-Botet M. Funding Imprint of human cytomegalovirus infection on the NK cell receptor This work was supported by grants FIS/PI17/00254, SAF 2016-80363-C2-1-R repertoire. Blood. 2004;104:3664–71. (Spanish Ministry of Economy and Competitiveness and FEDER), the EU FP7- 10. Martínez-Rodríguez JE, Cobo-Calvo A, Villar LM, Munteis E, Blanco Y, Rasal R, MINECO Infect-ERA Program (PCIN-2015-191-C02-01), and Red Española de et al. Adaptive natural killer cell response to cytomegalovirus and disability Esclerosis Múltiple (REEM) from the Instituto de Salud Carlos III, the European progression in multiple sclerosis. Mult Scler. 2016;22:741–52. Regional Development Fund (Grant RD16/0015/0011), and the Spanish Ministry 11. Lublin FD, Reingold SC, Cohen JA, Cutter GR, Sorensen PS, Thompson AJ, of Economy and Competitiveness. et al. Defining the clinical course of multiple sclerosis: the 2013 revisions. Neurology. 2014;83:278–86. Availability of data and materials 12. Haas J, Bekeredjian-Ding I, Milkova M, Balint B, Schwarz A, Korporal M, et al. The datasets used and/or analysed during the current study are available B cells undergo unique compartmentalized redistribution in multiple from the corresponding author on reasonable request. sclerosis. J Autoimmun. 2011;37:289–99. 13. Welsh RM, Che JW, Brehm MA, Selin LK. Heterologous immunity between Ethics approval and consent to participate viruses. Immunol Rev. 2010;235:244–66. The study was approved by the local ethics committee (Hospital del Mar 14. White DW, Suzanne Beard R, Barton ES. Immune modulation during latent Research Institute, IMIM, ethic approval number: 2016/6607/I). Patients gave herpesvirus infection. Immunol Rev. 2012;245:189–208. Zabalza et al. Journal of Neuroinflammation (2020) 17:161 Page 8 of 8 15. López-Montañés M, Alari-Pahissa E, Sintes J, Martínez-Rodríguez JE, Muntasell A, López-Botet M. Antibody-dependent NK cell activation differentially targets EBV-infected cells in lytic cycle and bystander B lymphocytes bound to viral antigen–containing particles. J Immunol. 2017; 199:656–65. 16. Alari-Pahissa E, Moreira A, Zabalza A, Alvarez-Lafuente R, Munteis E, Vera A, et al. Low cytomegalovirus seroprevalence in early multiple sclerosis: a case for the ‘hygiene hypothesis’? Eur J Neurol. 2018;25:925–33. 17. Moreira A, Alari-Pahissa E, Munteis E, Vera A, Zabalza A, Llop M, et al. Adaptive features of natural killer cells in multiple sclerosis. Front Immunol. 2019;10:2403. 18. Chidrawar S, Khan N, Wei W, McLarnon A, Smith N, Nayak L, et al. Cytomegalovirus-seropositivity has a profound influence on the magnitude of major lymphoid subsets within healthy individuals. Clin Exp Immunol. 2009;155:423–32. 19. Baker D, Marta M, Pryce G, Giovannoni G, Schmierer K. Memory B cells are major targets for effective immunotherapy in relapsing multiple sclerosis. EBioMedicine. 2017;16:41–50. 20. Lee GC, Song BH, Lee CH. Increase in the expression of human leukocyte antigen class I in human fibroblasts by soluble factors secreted from human cytomegalovirus-infected cells. Mol Cells. 2001;11:392–8. 21. Schubert RD, Hu Y, Kumar G, Szeto S, Abraham P, Winderl J, et al. IFN-β treatment requires B cells for efficacy in neuroautoimmunity. J Immunol. 2015;194:2110–6. 22. Duddy M, Niino M, Adatia F, Hebert S, Freedman M, Atkins H, et al. Distinct effector cytokine profiles of memory and naive human B cell subsets and implication in multiple sclerosis. J Immunol. 2007;178:6092–9. 23. Bar-Or A, Fawaz L, Fan B, Darlington PJ, Rieger A, Ghorayeb C, et al. Abnormal B-cell cytokine responses a trigger of T-cell-mediated disease in MS? Ann Neurol. 2010;67:452–61. 24. Stein J, Xu Q, Jackson KC, Romm E, Wuest SC, Kosa P, et al. Intrathecal B cells in MS have significantly greater lymphangiogenic potential compared to B cells derived from non-MS subjects. Front Neurol. 2018;9:554. 25. Longbrake EE, Cross AH. Effect of multiple sclerosis disease-modifying therapies on B cells and humoral immunity. JAMA Neurol. 2016;73:219–25. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.