henhaobio
Sign in

Abstract

Multiple sclerosis (MS) occurs when the central nervous system (CNS) is damaged by misguided adaptive immune responses, likely caused by a combination of environmental factors in genetically susceptible individuals. A known prerequisite for disease is Epstein-Barr virus (EBV) infection, and previous studies have demonstrated elevated Epstein-Barr virus nuclear antigen 1 (EBNA1) antibodies which cross-react with the calcium-activated chloride channel anoctamin-2 (ANO2) in persons with MS (pwMS). ANO2-reactive antibodies have been associated with greater neuroaxonal damage in MS, although their exact effector function is still uncertain. Here, we demonstrate that ANO2 is also the target of IFN{gamma}-producing CD4+ T cells, which are more frequent in untreated and natalizumab-treated pwMS compared to control individuals. Immunisation of SJL/J mice with either ANO2 or EBNA1 elicited cross-reactive CD4+ T cell and antibody responses in vivo. Pre-immunisation of young mice with ANO2 worsened proteolipid protein (PLP)-induced experimental autoimmune encephalomyelitis (EAE), which in older mice included atypical clinical phenotype, immune infiltration into the brain and reduced survival. EAE exacerbation was recapitulated with the adoptive co-transfer of ANO2 and myelin antigen-specific CD4+ T cells, and ANO2-specific T cells alone could induce the cell death of ANO2-expressing glial cells in vitro. T cell clones with cross-reactivity to both EBNA1 and ANO2 antigens could be isolated from natalizumab-treated pwMS. Single cell sequencing of EBNA1 and ANO2-specific T cell receptors (TCR) from four pwMS revealed a significant overlap between their antigen-specific expanded TCR repertoires within donors and transcriptomic analysis showed cross-reactive T cells to have predominantly activated and cytotoxic phenotypes. In summary, we report the first mechanistic evidence that EBNA1 CD4+ T cells can target the MS-associated autoantigen ANO2, thereby establishing a link between EBV infection and development of autoimmune neuroinflammatory disease.

In-depth summary

Claude Opus 5.5 · xhigh effort · 5 Nov 2026 · from the full text of version 1 (5 Dec 2025)

CD4+ T cells cross-reactive between Epstein-Barr virus EBNA1 and the central nervous system channel anoctamin-2 (ANO2) are elevated in multiple sclerosis and exacerbate experimental autoimmune encephalomyelitis in mice.

Epstein-Barr virus (EBV) infection is a prerequisite for multiple sclerosis (MS) and cross-reactive antibodies between EBNA1 and anoctamin-2 (ANO2) associate with neuroaxonal injury, yet antibodies alone cannot initiate demyelinating disease. This study shows that EBNA1-primed CD4+ T cells cross-react with ANO2 in MS patients and drive CNS vascular leakage, glial killing, and disease exacerbation in mice.

Key results

  1. ShownANO2-reactive pro-inflammatory T cells are significantly elevated in untreated and natalizumab-treated persons with multiple sclerosis and strongly discriminate cases from controls.

    In FluoroSpot assays with ANO2(79-168) beads, 56.7% of untreated pwMS (n = 30) had >10 IFN-gamma SFU per 2.5x10^5 PBMCs vs healthy controls (n = 20) and OND (n = 19), yielding ROC AUCs of 0.77 for IFN-gamma (P = 0.0002), 0.75 for IL-17A (P = 0.0005), and 0.85 for IL-22 (P < 0.0001), with spectral flow cytometry confirming higher CD4+ IFN-gamma+ ANO2 responses in natalizumab-treated pwMS (n = 12 vs 9 HC). · T cell responses targeting Anoctamin-2 are…; T cell responses to other regions of ANO2 are…; Figure legends

  2. ShownImmunisation of SJL/J mice with ANO2 or EBNA1 reciprocally cross-primes CD4+ T cells in vivo and exacerbates proteolipid protein-induced experimental autoimmune encephalomyelitis independently of autoantibodies.

    Immunising SJL/J mice with ANO2(1-275) or EBNA1(380-641) proteins, peptide pools, or homologous minimal peptides ANO2(137-151) and EBNA1(428-444) cross-primed CD4+ IFN-gamma+ T cells to the reciprocal antigen; pre-immunisation with either ANO2(1-275) protein or overlapping 15-mer peptides worsened subsequent PLP(139-151)-induced EAE scores and AUC at day 31 despite lacking anti-ANO2 or anti-EBNA1 IgG in peptide mice. · ANO21-275 and EBNA1380-641 immunisation…; ANO21-275 pre-immunisation cross-primes…; Protein versus peptide ANO21-275…; Figure legends

  3. ShownANO2 localizes to CD31+ vasculature in the mouse cerebellum, septum, and inferior olivary nucleus, where ANO2-specific CD4+ T cells exacerbate atypical EAE, blood-brain barrier breakdown, and oligodendrocyte apoptosis.

    In 6-month-old SJL/J mice (n = 23 ANO2/PLP vs 22 adjuvant/PLP), ANO2(1-275) pre-immunisation reduced survival and doubled atypical EAE; RNAscope and iDISCO+ (n = 9 brains) showed ANO2 co-expression with CD31+ endothelium in cerebellum (76%), septum (85%), and inferior olivary nucleus (72%); co-transferring ANO2- and MOG-specific CD4+ T cells increased BBB leakage, and ANO2 T cells killed oligodendrocytes in vitro. · ANO21-275 pre-immunisation of adult mice gives…; ANO2 co-localises with CD31 staining in the…; Adoptive transfer of ANO21-275-specific T cells…; ANO21-275-specific T cells can directly target…

  4. ShownSingle-cell TCR and transcriptome sequencing of CD4+ T cells from pwMS reveals substantial repertoire overlap between EBNA1 and ANO2 with activated and cytotoxic phenotypes.

    In HLA-DRB1*15:01+ pwMS, isolated T cell clones and HLA-DR2a/DR2b-restricted lines recognized both EBNA1(380-641) and ANO2; across 27,524 single-cell TCRs from 4 natalizumab-treated pwMS, ~25-26% of expanded EBNA1 and ANO2 TCRs were shared (55 clonotypes shared exclusively between EBNA1 and ANO2 and 43 tri-cross-reactive with CRYAB), mapping to activated (S1PR1, CCR6, RORC) and cytotoxic (GZMB, PRF1, GNLY) clusters. · Dual-specific T cell clones with reactivity to…; Single cell TCR sequencing of EBNA1- and…; EBNA1-, ANO2-specific and cross-reactive T cells…; Figure legends

Methods, data and reproducibility

Measured ex vivo T cell and IgG reactivity to recombinant ANO2 and EBNA1 proteins and peptides in untreated and natalizumab-treated pwMS and controls via FluoroSpot, Cytek Aurora spectral flow cytometry, and Luminex bead arrays. Assayed in vivo cross-priming, EAE severity, iDISCO+ clearing, Atto647-plasma BBB leakage, and Incucyte glial killing in mice, plus 10x 5' scRNA/TCR-seq of 4 MS-Nat donors.

  • Human PBMCs from untreated MS (MS-Un), natalizumab-treated MS (MS-Nat), healthy controls (HC), and other neurological diseases (OND) for FluoroSpot and spectral flow cytometry · human · n = Cohort 1: MS-Un n=30, HC n=20, OND n=19; Cohort 2: MS-Nat n=61, HC n=28
  • Plasma samples from the Swedish Epidemiological Investigation of Multiple Sclerosis (EIMS) cohort for suspension bead array autoantibody profiling · human · n = 713 pwMS and 722 matched population controls
  • 10x Chromium 5' single-cell RNA and TCR sequencing of CFSE-dim CD45RA-CD4+ T cells stimulated with anti-CD3, ABD, EBNA1(380-641), ANO2(1-275), and CRYAB · human · n = 4 natalizumab-treated pwMS; 27,524 TCRs and 27,076 single-cell transcriptomes
  • Mouse antigen recall, EAE pre-immunisation, adoptive transfer, iDISCO+ whole-mount brain imaging, and primary glial/neuronal co-cultures · mouse (C57BL/6Ntac, NOD, SJL/J, and SJL/J x C57BL/6-GFP F1) · n = 9 brains for iDISCO+; 6-mo SJL/J EAE: n=22 adjuvant/PLP, n=23 ANO2/PLP · https://mouse.brain-map.org/
  • Public MS blood and CSF bulk TCRbeta CDR3 repertoire datasets (Gottlieb et al. and Amoriello et al.) · human · GSE250037
Methods & tools
CellRanger, Seurat, DropletUtils, Scirpy, Scanpy, Muon, Harmonypy, PyDESeq2, fgsea, FlowJo, ImageJ (Fiji), Imaris
Software & environment
R 4.2.3, Python 3.10, Scirpy 0.19.0, Scanpy 1.10.3, Muon 0.1.7, DropletUtils 1.18.1, Seurat 4.3.0, AnndataR 0.99.0, vegan 2.6-4, msigdbr 7.5.1, ImageJ (Fiji) 1.54j, Imaris 10.2.0, Amira 3D 2024.1, GraphPad Prism v10
Reproducibility gaps
Full analysis code is stated to be supplied upon request; no public accession identifier is printed for the newly generated 10x single-cell RNA and V(D)J sequencing dataset.
Limitations (3)
  • Single-cell TCR and transcriptome sequencing was restricted to in vitro antigen-expanded CD4+ T cells from 4 natalizumab-treated pwMS and did not include healthy controls or patients with other EBV-associated autoimmune diseases. · stated by the authors
  • Natalizumab blocks lymphocyte trafficking into the CNS and gut and promotes peripheral lymphocyte activation, which could enrich autoreactive T cells and influence ex vivo and expansion assays. · stated by the authors
  • Overlap analysis with external blood and CSF TCR repertoires relied on bulk TCRVbeta CDR3b chains without paired TCRValpha sequences, preventing definitive confirmation of shared antigen specificity. · stated by the authors

Read this if you investigate Epstein-Barr virus molecular mimicry, autoreactive T cell repertoires, or neuroinflammation in multiple sclerosis and experimental autoimmune encephalomyelitis.

Citations
40
FWCI
182.97
Field percentile
top 1%

Reproducibility ledger

Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.

Sign in to file a reproducibility report.

Similarity Map
SelectedSame fieldCross-field
THISYounis '26Kristensen '26Allanach '22Kristensen '25Valkiers '24Reijm '26
83% semantic matchImmunology
Shady Younis, Sajede Rasouli, Jacob W. Loeffler+15 more·bioRxiv·Feb 12, 2026· 5 citations