Extrafollicular IgD-CD27-CXCR5-CD11c- DN3 B cells infiltrate inflamed tissues in autoimmune fibrosis and in severe COVID-19

Abstract

Although therapeutic B cell depletion dramatically resolves inflammation in many diseases in which antibodies appear not to play a central role, distinct extrafollicular pathogenic B cell subsets that accumulate in disease lesions have hitherto not been identified. The circulating immunoglobulin D (IgD)^-CD27^-CXCR5^-CD11c⁺ DN2 B cell subset has been previously studied in some autoimmune diseases. A distinct IgD^-CD27^-CXCR5^-CD11c^- DN3 B cell subset accumulates in the blood both in IgG4-related disease, an autoimmune disease in which inflammation and fibrosis can be reversed by B cell depletion, and in severe COVID-19. These DN3 B cells prominently accumulate in the end organs of IgG4-related disease and in lung lesions in COVID-19, and double-negative B cells prominently cluster with CD4⁺ T cells in these lesions. Extrafollicular DN3 B cells may participate in tissue inflammation and fibrosis in autoimmune fibrotic diseases, as well as in COVID-19.

Keywords: B cell depletion in disease; COVID-19 pathogenesis; CP: Immunology; DN3 B cells; IgG4-related disease; T-B collaboration in disease; double-negative B cells; extrafollicular B cells; inflammatory fibrosis.

Graphical abstract

Figure 1

Several populations of circulating activated B cells distinguish patients with IgG4-RD from healthy controls (A) Multiparametric analysis of circulating B cells using tSNE with overlay of the relative expression of key markers. (B) Gating strategy to examine B cell subsets, segregating plasmablasts and then categorizing IgD⁻CD27⁻ B cells into DN1, DN2, DN3, and DN4 subsets. (C) Patients with IgG4-RD (n = 5) exhibited four distinctive populations of DN (IgD⁻CD27⁻) B cells distinct from those in patients living with HIV, which are FcRL4⁺ (n = 5). (D) DN B cells from IgG4-RD (n = 38) are different from the DN B cells described in lupus (n = 5). (E) Frequencies of different B cell populations and of SWM B cells, ABC-like B cells, and activated naive B cells in IgG4-RD and controls (n = 38). Multiple comparisons are controlled for by Kruskal-Wallis test. Error bars represent mean ± SEM. ^∗p < 0.05; ^∗∗p < 0.01; ^∗∗∗p < 0.001; ^∗∗∗∗p < 0.0001.

Figure 2

Circulating DN B cell populations are enriched in IgG4-RD and COVID-19 (A) Frequencies of circulating B cell subpopulations in patients with IgG4-RD compared with healthy volunteers. (B) DN B cell subsets correlated with plasmablast frequencies. The strongest correlation being with DN3 and DN2 B cells (r = 0.66, r = 0.45) suggesting co-occurrence of these B cell in IgG4-RD. (C) Frequency of DN B cell subsets in controls and moderate, severe, and convalescent (conval) COVID-19 patients from an MGH cohort. The data were extracted from the analyses in Kaneko et al. Two-tailed Mann-Whitney U test was used for pairwise comparisons, and multiple comparisons were controlled for by Kruskal-Wallis test. Error bars represent mean ± SEM. ^∗p < 0.05; ^∗∗p < 0.01; ^∗∗∗p < 0.001; ^∗∗∗∗p < 0.0001. (D) Using a logistic regression model and the maximum likelihood method, we confirmed that plasmablasts discriminate between patients with IgG4-RD and healthy controls. We additionally showed that total DN B cells as well as DN2, DN3, and DN4 B cells can independently distinguish IgG4-RD. Significance was tested using the Wald test.

Figure 3

T cell-B cell conjugates in IgG4-RD end organs (A) Immunofluorescence staining of CD19 (red), CD27 (cyan), IgD (yellow), CD20 (purple), CD4 (green), CD8 (orange), and DAPI (blue) in a submandibular gland (SMG) from a patient with IgG4-RD. (B) Absolute numbers and percentages of IgD⁻CD27⁻ DN B cells and IgD⁻CD27+ B cells in IgG4-RD (n = 7). This combines data for which a representative image is shown in Figure S1 and from (A). (C) Absolute numbers of DN B cells (blue), IgD⁻CD27⁺ B cells (gold), and plasmablasts (gray) in IgG4-RD (n = 4). (D) Immunofluorescence staining of CD19 (red), CD27 (cyan), IgD (yellow), CD20 (purple), CD4 (green), CD8 (orange), and DAPI (blue) and visualization of physically interacting T and B cells in an SMG from a patient with IgG4-RD using the StrataQuest cell-to-cell contact application. Nuclei circled in red depict CD19⁺ B cell in physical contact with CD4⁺ T cells. T cells and B cells formed close and extensive intercellular plasma membrane contacts. (E) Absolute numbers of DN B cells (blue), SWM B cells (yellow), and plasmablasts (purple) in physical contact with CD4⁺ T cells in patients with IgG4-RD (n = 4). (F) Absolute numbers of DN B cells, SWM B cells, and plasmablasts conjugated with CD4⁺ T cells (blue) and CD8⁺ T cells (yellow) in patients with IgG4-RD (n = 4).

Figure 4

Comparison of transcriptomes of distinct circulating DN B cell subsets (A) The 4 subsets of DN B cells, isolated from the peripheral blood, of IgG4-RD are transcriptomically distinct and harbor unique signatures that suggest different, non-overlapping biological properties of each population. (B) A list of the representative Gene Ontology annotations that were examined. (C and D) The DN3 subset of DN B cells exhibits signatures of proliferating cells and the unfolded protein response seen in antibody-secreting cells.

Figure 5

DN B cells in draining lymph nodes and in the lungs of COVID-19 patients (A) Representative multicolor immunofluorescence images of CD19 (red), IgD/CD27 (yellow), CXCR5 (green), CD11c (purple), and DAPI (blue) staining in a lymph node from a late COVID-19 patient. Arrows indicate DN B cells. (B) Relative proportions of DN B cells in the pool of CD19⁺ B cells in lymph nodes from late COVID-19 patients (n = 6). (C) Absolute numbers of each subset of DN B cells in late COVID-19 patients (n = 6) (DN1, blue; DN2, gold; DN3, gray; DN4, cyan). (D) Representative multicolor immunofluorescence images of CD19 (red), IgD/CD27 (yellow), CXCR5 (green), CD11c (purple), and DAPI (blue) staining in a lung from a late COVID-19 patient. Arrows indicate DN B cells. (E) Relative proportions (left) and absolute numbers (right) of DN B cells in the pool of CD19⁺ B cells in lungs from late COVID-19 (blue) (n = 6) and non-COVID-19 patients (gold) (n = 6). (F) Absolute numbers of each subset of DN B cells in late COVID-19 patients (blue) (n = 6) and non-COVID-19 patients (gold) (n = 6). (G) Relative proportions of each DN B cells in the pool of DN B cells in lungs from late COVID-19 patients (n = 6). Mann-Whitney U test used to calculate p value. Error bars represent mean ± SEM. ^∗p < 0.05.

Figure 6

DN B cells infiltrate the SMGs in IgG4-related disease (A) Representative multicolor immunofluorescence images of CD19 (red), IgD (orange), CD27 (green), and DAPI (blue) staining in an SMG from a patient with IgG4-RD. Arrows indicate DN B cells. (B) Relative proportions (left) and absolute numbers (right) of DN B cells in the pool of CD19⁺ B cells in SMGs from IgG4-RD (blue) (n = 10) and chronic sialadenitis (CS) (gold) (n = 7). (C) Representative multicolor immunofluorescence images of CD19 (red), IgD/CD27 (yellow), CD11c (purple), CXCR5 (green), and DAPI (blue) staining in an SMG from a patient with IgG4-RD (DN1, top left; DN2, top right; DN3, bottom left; DN4, bottom right). (D) Absolute numbers of each subset of DN B cells in IgG4-RD (n = 10) (blue) and CS (gold) (n = 7). (E) Relative proportions of each DN B cell in the pool of DN B cells in SMGs from IgG4-RD (blue) (n = 10) and CS (gold) (n = 7). Mann-Whitney U test used to calculate p value. Error bars represent mean ± SEM. ^∗p < 0.05; ^∗∗p < 0.01.