Epstein-Barr virus reprograms autoreactive B cells as antigen-presenting cells in systemic lupus erythematosus

Abstract

Systemic lupus erythematosus (SLE) is a systemic autoimmune disease characterized by antinuclear antibodies (ANAs). Epstein-Barr virus (EBV) infection has been epidemiologically associated with SLE, yet its role in pathogenesis remains incompletely defined. Here, we developed an EBV-specific single-cell RNA-sequencing platform and used it to demonstrate that EBV infection reprograms autoreactive antinuclear antigen B cells to drive autoimmunity in SLE. We demonstrated that, in SLE, EBV⁺ B cells are predominantly CD27⁺CD21^low memory B cells that are present at increased frequencies and express ZEB2, TBX21 (T-bet), and antigen-presenting cell transcriptional pathways. Integrative analysis of chromatin immunoprecipitation sequencing (ChIP-seq), assay for transposase-accessible chromatin sequencing (ATAC-seq), and RNA polymerase II occupancy data revealed EBV nuclear antigen 2 (EBNA2) binding at the transcriptional start sites and regulatory regions of CD27, ZEB2, and TBX21, as well as the antigen-presenting cell genes demonstrated to be up-regulated in SLE EBV⁺ B cells. We expressed recombinant antibodies from SLE EBV⁺ B cells and demonstrated that they bind prototypical SLE nuclear autoantigens, whereas those from healthy individuals do not. We further found that SLE EBV⁺ B cells can serve as antigen-presenting cells to drive activation of T peripheral helper cells with concomitant activation of related EBV^- antinuclear double-negative 2 B cells and plasmablasts. Our results provide a mechanistic basis for EBV being a driver of SLE through infecting and reprogramming nuclear antigen-reactive B cells to become activated antigen-presenting cells with the potential to promote systemic disease-driving autoimmune responses.

Fig. 1.. EBV-seq enables identification of EBV-infected B cells.

(A) An EBV-seq primer design was incorporated into a 10X bead-based scRNA-seq workflow that included Cellular indexing of transcriptomes and epitopes by sequencing (CITE-seq) performed using oligonucleotide-barcoded antibodies (Abs) specific for the prototypical markers for B cell subsets. B cell receptor (BCR) sequencing was performed by analysis of the expressed variable diversity and joining (VDJ) immunoglobulin genes, followed by analysis of immunoglobulin heavy (H) and light (L) chain gene expression. UMIs, unique molecular identifiers. (B) Shown is the EBV genome annotated with the EBV genes (blue font) and small non-coding RNAs (green font) for which primers were included in EBV-seq. (C) Comparison of standard 10X scRNA-seq with EBV-seq 10X scRNA-seq for detection of EBV⁺ B cells in a PBMC sample from a patient with SLE; EBV UMI counts are represented by intensity of brown dots.

Fig. 2.. EBV-seq analysis of blood from patients with SLE identified EBV⁺ B cells predominantly in latency with some B cells expressing lytic reactivation genes.

(A) Combined EBV-seq, CITE-seq, VDJ-seq, and scRNA-seq were used to sequence B cells isolated from SLE (n=11) and HC (n=10) PBMC samples. (B) Number of EBV⁺ B cells detected in SLE and HC blood, normalized to 10,000 B cells. Each dot represents one individual and bars represent mean ± SEM. ***P<0.001 by Wilcoxon rank sum test. (C) Unsupervised clustering based on differentially expressed genes in the sequenced SLE and HC B cells, with annotation of B cell subsets. (D) EBV gene UMI counts in the sequenced SLE and HC B cells. (E) SLE blood B cell marker expression. The bar graph represents the total EBV UMI counts detected in each B cell subset. (F) Distribution of EBV⁺ B cells across B cell subsets in HC and SLE PBMCs. Data are presented as the normalized number of EBV⁺ B cells per 10,000 B cells of the corresponding B cell subset. Bars represent mean ± SEM. Statistical significance was determined by Kruskal-Wallis test with P values adjusted for multiple testing using the Benjamini–Hochberg method (*P < 0.05, **P < 0.01, ***P < 0.001). Blue asterisks indicate differences in EBV⁺ B cell frequencies between HC and SLE within each subset, and black asterisks indicate pairwise comparisons of CD27⁺CD21^low B cells against other EBV⁺ subsets in SLE. (G) Frequencies of EBV gene UMI counts in B cell subsets from HC and SLE; EBV genes from which UMI gene counts were not detected are not displayed. The heat map represents the number of EBV⁺ B cells per 10,000 B cells per B cell subset. mB, memory B cells.

Fig. 3.. SLE EBV⁺ CD27⁺CD21^low memory B cells exhibit activation as APCs.

(A) Differential gene expression analysis of SLE blood EBV⁺ versus EBV⁻ CD27⁺CD21^low memory B cells. (B) Volcano plot of differentially expressed genes in SLE EBV⁺ versus EBV⁻ CD27⁺CD21^low memory B cells. (C) Activated pathways of the differentially expressed genes in SLE EBV⁺ versus EBV⁻ CD27⁺CD21^low memory B cells. (D) Differential gene expression analysis EBV⁺ CD27⁺CD21^low memory B cells in SLE versus HC blood. (E) Volcano plot of differentially expressed genes in EBV⁺ CD27⁺CD21^low memory B cells in SLE versus HC blood. (F) Activated pathways of the differentially expressed genes in EBV⁺ CD27⁺CD21^low memory B cells in SLE versus HC blood. mB, memory B cells. FDR, false discovery rate.

Fig. 4.. EBV⁺ B cells in SLE harbor distinct transcriptional programs and developmental trajectories toward CD27⁺CD21^low memory and plasmablast phenotypes.

(A) Network analysis of differentially expressed genes differentially expressed genes (DEGs) in EBV⁺ cells as compared to EBV⁻ B cell subsets. Endoplasmic-reticulum-associated protein degradation (ERAD); mB, memory B cell. (B) B cell activation score, APC score, and interferon score across all B cells. Scores were defined as described in the Methods. (C to E) B cell activation score (C), APC score (D), and interferon score (E) of the un-switched memory B cell, switched memory B cell, CD27⁺CD21^low memory B cell, DN2 B cell, DN1 B cell, non-proliferating Ki67⁻ plasmablast (PB), and proliferating Ki67⁺ PB subsets in EBV⁺ and EBV⁻ B cells in peripheral blood from individuals with SLE and HCs. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001 by Mann-Whitney U test with Benjamini–Hochberg correction for multiple testing. (F) Trajectory analysis of EBV⁺ B cells. Solid arrows indicate inferred directional flows or transitions among subsets as computed by RNA velocity.

Fig. 5.. SLE EBV⁺ B cells express and EBNA2 binds regions implicated in transcriptional and epigenetic regulation of CD27, ZEB2, TBX21 (Tbet) and APC genes in EBV⁺ B cells from individuals with SLE.

(A) Expression of B cell markers and transcription factors in SLE blood EBV⁺ B cells across B cell subsets. (B) Expression of ZEB2, TBX21, other select markers, and the APC gene module in HC EBV⁺ B cells in latency 0-1 and in SLE EBV⁺ B cells in latency 0-1, latency 2-3, or expressing lytic genes. (C) EBNA2 ChIP-seq peak density at transcriptional start site (TSS) regions and gene bodies of all up-regulated genes (left) and up-regulated genes related to antigen presentation (right) in SLE EBV⁺ versus EBV⁻ CD27⁺CD21^low memory B cells. Histograms show EBNA2 peak frequency (FDR <0.01) across genomic regions. (D) Integrative genomic analysis of ChIP-seq, ATAC-seq, and RNA polymerase II occupancy data from EBV⁺ B cell lines (Mutu III and GM12878 LCL) for EBNA2, focusing on the genes encoding CD27 and the transcription factors Zeb2 and Tbet (TBX21), which are up-regulated in EBV⁺ B cells (Fig. 5, A and B) and implicated in SLE pathogenesis. Highlighted red regions indicate EBNA2 binding (FDR < 0.01).

Fig. 6.. SLE EBV⁺ B cells, their clonal family EBV⁻ B cells, and their related family EBV⁻ B cells, encode anti-nuclear BCRs.

(A) Representative phylogenetic tree of IGHV genes expressed by SLE blood EBV⁺ and EBV⁻ B cells. Examples of EBV⁺ B cells, and their clonal family EBV⁻ B cells that utilize the same V(D)J rearrangement indicative of a shared progenitor B cell (dark blue), or their related family EBV⁻ B cells that utilize different V(D)J rearrangements indicative of different progenitor B cells (teal). (B) Distribution of SLE EBV⁻ and EBV⁺ B cell clonal family sizes. For the EBV⁻ B cells, only the EBV⁻ B cells in EBV⁺ B cell clonal families were included. ****P<0.0001, by Chi-Squared test. (C) SLE EBV⁺ and EBV⁻ B cell immunoglobulin isotype usage. (D) SLE and HC EBV⁺ B cell-derived mAb immunostaining of HEp-2 cells. (E) BCR profiles, cell type and SHM, and HEp-2 staining of SLE EBV⁺ clonal family B cells encoding anti-Sm antibodies. The left schematic shows a clonal family with the same V(D)J rearrangement; node color indicates EBV status of the cells (red = EBV⁺, blue = EBV⁻) (F) Binding affinity analysis of EBV⁺ and EBV⁻ B cell mAbs from the SLE EBV⁺ clonal family for Sm. (G) BCR profiles, properties, and HEp-2 staining of SLE EBV⁺ related family B cells encoding anti-Sm antibodies. (H) Binding affinity analysis of EBV⁺ and EBV⁻ B cell mAbs from the SLE EBV⁺ related family for Sm. For (F) and (H), P<0.01 by Student’s t-test.

Fig. 7.. SLE EBV⁺ B cell mAb HEp-2 cell immunostaining patterns and their reactivity to prototypical SLE autoantigens.

(A) Representative examples of immunostaining patterns of SLE EBV⁺ B cell-derived mAb on HEp-2 cells. Nuclear homogeneous includes both homogeneous and fine speckled staining patterns. (B) Representative examples of immunostaining of HC and MS EBV⁺ B cell-derived mAbs on HEp-2 cells. (C) ELISA analysis of the reactivity of SLE, HC, and MS EBV⁺ B cell-derived mAbs to prototypical SLE autoantigens, including the ANA screen ELISA (Nuclear Ag cocktail) and individual nuclear antigens, and for EBV EBNA1. As an assessment for potential polyreactive antibodies, LPS ELISAs were performed. The Nuclear Antigen (Ag) cocktail, comprising Sm (Smith), RNP/Sm, Scl-70, SS-A (Ro) (52 kDa and 60 kDa), SS-B (La), Jo-1, U1-SmRNP, CENP-B, dsDNA, Histones, and C1q was obtained from a commercial source different from that used for the individual antigens in the heatmap. N = number of mAbs tested per group.

Fig. 8.. SLE EBV⁺ LCLs serve as APCs that promote expansion of Tph and EBV⁻ ANA-encoding DN2 B cells and plasmablasts.

(A) Schematic overview of LCL co-cultures with T and B cells. LCLs were generated from patients with SLE possessing anti-DNA autoantibodies. SLE LCLs were loaded with chromatin using a combination of anti-IgG/M-avidin and anti-DNA-biotin antibodies (free chromatin is present in culture supernatants due to low amounts of dying cells). Chromatin-loaded LCLs were washed and then co-cultured with T and B from the same patient (SLE MHC-matched B and T cells), or from a different patient with anti-DNA autoantibodies (SLE MHC-mismatched B and T cells). Healthy control LCLs, as well as T and B cells, were used as controls. (B) Frequency of T cell subsets, including PD1^hiCXCR5⁻CD4⁺ Tph cells, at day 10 among non-expanding and expanding CD4⁺ T cell clones as measured by scRNA-seq. ****P<0.0001 by Chi-Squared test. (C) Frequencies of the B cell subsets among the expanded EBV⁻ SLE B cells as measured by scRNA-seq. ****P<0.0001 by Chi-Squared test. (D) Immunostaining of HEp-2 cells with expressed mAbs from expanded DN2 B cell and plasmablast (PB) clones in co-culture from (C). (E) Immunostaining of HEp-2 cells using culture supernatants collected at day 10. The top bar graph presents the mean fluorescence of HEp-2 staining, with the values representing the average pixel intensity within a selected region of interest (ROI) measured by ImageJ. Data are shown as mean ± SEM. ****P<0.0001 by Student’s t test.