Antigen-dependent interactions between regulatory B cells and T cells at the T:B border inhibit subsequent T cell interactions with DCs

Abstract

IL-10+ regulatory B cells (Bregs) inhibit immune responses in various settings. While Bregs appear to inhibit inflammatory cytokine expression by CD4+ T cells and innate immune cells, their reported impact on CD8+ T cells is contradictory. Moreover, it remains unclear which effects of Bregs are direct versus indirect. Finally, the subanatomical localization of Breg suppressive function and the nature of their intercellular interactions remain unknown. Using novel tamoxifen-inducible B cell-specific IL-10 knockout mice, we found that Bregs inhibit CD8+ T cell proliferation and inhibit inflammatory cytokine expression by both CD4+ and CD8+ T cells. Sort-purified Bregs from IL-10-reporter mice were adoptively transferred into wild-type hosts and examined by live-cell imaging. Bregs localized to the T:B border, specifically entered the T cell zone, and made more frequent and longer contacts with both CD4+ and CD8+ T cells than did non-Bregs. These Breg:T cell interactions were antigen-specific and reduced subsequent T:DC contacts. Thus, Bregs inhibit T cells through direct cognate interactions that subsequently reduce DC:T cell interactions.

Keywords: basic (laboratory) research/science; immunobiology; lymphocyte biology; tolerance: mechanisms.

Figure 1.. T cell produce increased levels of inflammatory cytokines and proliferate more in mice with B cells specific IL-10 deficiency.

hCD20-ERT2.Cre-IL-10fl/fl mice (B-IL-10KO) or Control (IL-10fl/fl) mice were treated with tamoxifen on days −3, −2 and −1 to specifically knockout the IL-10 gene in B cells. All mice were immunized with BALB/c splenocytes on day 0 and spleens harvested 14 days later to examine T cell cytokines. a) Bar graph showing IL-10 expression by B cells from B-IL-10KO vs. Control mice (n = 3). b) IFN-γ and IL-17 production by splenic CD4 T cells (n = 5). c) IFN-γ production by splenic CD8 T cells (n = 5). d) CFSE-stained OT-1 T cells were transferred into tamoxifen treated B-IL-10KO or Control mice, as described above, that were subsequently immunized with NP-OVA-alum i.p. and spleens were harvested on day 5. Histograms show OT-1 CFSE dilution for indicated treatment groups (n = 4). e) IFN-γ production by transferred OT-1 T cells on day 5 (n = 4). Graphs (a, b, c, e) show mean ± SD with Student’s t tests (unpaired, two tailed) to analyze data. f) Kaplan-Meir plots showing allograft survival of Cre+ Control (B6) or B-IL-10KO BALB/c islet recipient with and without anti-TIM-1 treatment. Log-rank test was carried out the analyze survival curve.

Figure 2.. Immunized Tiger-IL-10eGFP reporter mice allow for Breg detection by flow cytometry without ex-vivo stimulation.

a) Splenocytes were isolated from IL-10-reporter mice that were Naïve or immunized (Imm) with allo-antigen (d3). Splenocytes were left unstimulated (Unstim) or stimulated with LPS, PMA, ionomycin and monensin (LPIM) for 5 hours. Subsequently, B cells were examined for IL-10-GFP expression by flow cytometry. Red boxes highlight GFP expression within each plot. b) Cumulative data showing B cell IL-10-GFP expression in LPIM stimulated vs. unstimulated cells. Student’s t tests (unpaired, two tailed) was used to assess statistical differences between groups (n = 3–6/group). c) Percent of B cells expressing IL-10-GFP within each B cell subset from unstimulated vs. LPIM-stimulated splenocytes (MZ, marginal zone; MZP, marginal zone precursor; FOI, follicular I; FOII, follicular II; T1, transitional 1; T2, transitional 2; PC, plasma cells; B1a) (n = 6). d) Percent contribution to total B cell IL-10-GFP expression by each subset. Data are representative of 4–7 mice. Graphs show mean ± SD.

Figure 3.. Localization and interactions of Bregs and Control B cells with OT-1 T cells in the spleen.

Experiments were carried out as described in Fig. S2a. (a-c): Immunofluorescence microscopic images of spleen sections of CD45.1 congenic mice 24h after transfer of CD45.2 Bregs and OT-1 cells. a) Section stained with anti-CD3 (T cells), anti-CD45.2 (transferred cells), and anti-IgM/B220 (B cells: same fluorochrome). b) Section stained with anti-CD3 (T cells), anti-IgM/B220 (B cells) and anti-GFP to specifically detect transferred Bregs. c) Inset from (b) zoomed in on a Breg cell showing separated fluorescent signals from each channel. d-e). 2-photon imaging snapshots showing transferred CFSE-stained Control B cells (GFP-, green) (d), or Bregs (GFP+, green) (e), around the periphery or within DS-red OT-1 T cell (red) clusters. f) Graph showing the frequency of Control B cell vs. Control activated B cell (Control Act B cell) vs. Breg cells entering T cell cluster. Each dot represents one individual 20–35 minute recording. Mean ± SD is shown. g) Graph showing the number of contacts between individual Control B cells, Control Act B cells, Bregs pulsed with NP-OVA or Bregs pulsed with NP-CGG with OT-1 T cells. Mean ± SEM is shown. h) Graph showing the duration (minutes) of each contact made by Control B cells, Control Act B cells, Bregs pulsed with NP-OVA or Bregs pulsed with NP-CGG with OT-1 T cells. Red line shows mean contact duration. Imaging results were obtained from 3–4 separate mice for each group with 131 Control B cells, 157 Control Act B cell, 135 Breg-NP-OVA and 90 Breg-NP-CGG cells tracked. Student’s t tests (unpaired, two tailed) was used to assess significant differences between groups.

Figure 4.. Breg, DC, and OT-1 T cell interactions in the spleen.

Experiments were carried out as described in Fig. S2a with minor modification. Mice received 2–3 million BMDC, 2 million Control B cells or Bregs and 1 million DsRed naïve OT-1 T cells. a) 2-photon microscopy imaging snapshot showing transferred Control B cells (green) with OT-1 T cells (blue) and BMDC (pink). Yellow lines represent track path and length for each B cell. b) 2-photon microscopy imaging snapshot showing Bregs (green) with OT-1 T cells (blue) and BMDC (pink). Yellow lines represent track path and length for each B cell. c) Graph showing the number of contacts individual Control B cells or Bregs made with BMDC. d) Graph showing the number of contacts each BMDC made with OT-1 T cells in the presence of Control B cells (98 BMDCs tracked) versus Bregs pulsed with NP-OVA (123 BMDCs tracked) versus Bregs pulsed with NP-CGG (102 BMDCs tracked). Mean (red line) ± SEM is shown. n = 2–4 mice per group. Student’s t tests (unpaired, two tailed) was used to assess significant differences between groups.

Figure 5.. Breg chemokine expression and splenic localization.

Representative histogram overlays showing relative expression of: a) CCR7 expression on Bregs (black) vs. Control B cells (grey). b) Bar graph showing mean MFI ± SD for each cell type in (a). Data is representative of 4 mice. c) CCR7 expression on FO-Bregs (black) vs. MZ-Breg (grey). d) Bar graph showing mean MFI ± SD for each cell type in (c). e) MZ-Bregs were sorted from previously immunized Tiger mice based on established phenotypic markers (Fig. S1b) and transferred into B6 WT mice. To simultaneously detect endogenous FO B cells (B220^hi IgM^lo) and endogenous B cells localizing to the MZ and red pulp (B220^lo IgM^hi), we stained spleen sections with anti-B220 and anti-IgM conjugated to the same fluorophore (blue). The sections were also stained with anti SIGNR-1 (green), which labels MZ macrophages, demarcating the MZ as a narrow band that separates the follicles from the red pulp (RP) and anti-GFP (red) to detect the transferred cells. f) FO-Bregs were sorted from previously immunized Tiger mice based on established phenotypic markers (Fig. S1c) and transferred into B6 WT mice. Sections were stained as in described in (e). Arrows indicate staining for transferred cells. g) Inset from (e) showing zoomed in area of GFP+ cell located in MZ. h) Inset from (f) showing zoomed in area of GFP+ cell located in B cell follicle. i) Summary graph showing quantitation of transferred MZ and FO Breg migration to indicated splenic niche. Mean ± SD is shown for each graph. n = 3–5 mice per group. Student’s t tests (unpaired, two tailed) was used to assess significant differences between groups.