B cells in the pneumococcus-infected lung are heterogeneous and require CD4+ T cell help including CD40L to become resident memory B cells

Abstract

Recovery from respiratory pneumococcal infections generates lung-localized protection against heterotypic bacteria, mediated by resident memory lymphocytes. Optimal protection in mice requires re-exposure to pneumococcus within days of initial infection. Serial surface marker phenotyping of B cell populations in a model of pneumococcal heterotypic immunity revealed that bacterial re-exposure stimulates the immediate accumulation of dynamic and heterogeneous populations of B cells in the lung, and is essential for the establishment of lung resident memory B (B_RM) cells. The B cells in the early wave were activated, proliferating locally, and associated with both CD4⁺ T cells and CXCL13. Antagonist- and antibody-mediated interventions were implemented during this early timeframe to demonstrate that lymphocyte recirculation, CD4⁺ cells, and CD40 ligand (CD40L) signaling were all needed for lung B_RM cell establishment, whereas CXCL13 signaling was not. While most prominent as aggregates in the loose connective tissue of bronchovascular bundles, morphometry and live lung imaging analyses showed that lung B_RM cells were equally numerous as single cells dispersed throughout the alveolar septae. We propose that CD40L signaling from antigen-stimulated CD4⁺ T cells in the infected lung is critical to establishment of local B_RM cells, which subsequently protect the airways and parenchyma against future potential infections.

Keywords: B cells; adaptive immunity; lung immunology; mucosal immunity; pneumonia.

Figure 1

Heterogeneous lung EV B cell populations generated after Sp19F infections. (A) Timeline to study lymphocyte kinetics after 1 or 2 Sp19F infections. Adult B6 mice received intratracheal (i.t.) instillations of Sp19F or saline into the left lung on day 0 and day 7 and were rested for 4 weeks to generate immunity. (B) Average numbers of lung EV B cells (i.v.CD45^-CD45⁺CD3^-CD4^-CD19⁺ B cells) at each collection timepoint, n=6 per timepoint, 2 biological replicates. (C) Individual op-tSNE plots with overlaid PhenoGraph clusters depicting live EV B cells after 1 or 2 infections with Sp19F on day 10, day 14, and day 35; n=6 per group. (D) Merged opt-SNE plot from each of the projections in panel (C) with overlaid and labeled PhenoGraph clusters. (E) Heatmap of average MFI per marker normalized to minimal MFI and maximal MFI generated with Morpheus (28). Hierarchical clustering and dendrogram construction were performed using Pearson’s correlation distances.

Figure 2

Kinetics of EV B cells after 1 or 2 self-limiting Sp infections. (A) Numbers of live EV CSW B2 (i.v.CD45.2^-CD19⁺CD43^loB220^hiIgD^-IgM^-), (B) CSW B1 (i.v.CD45.2^-CD19⁺CD43^hiB220^loIgD^-IgM^-), (C) GC/pre-GC (i.v.CD45.2^-CD19⁺GL7⁺), (D) naïve (i.v.CD45.2^-CD19⁺IgD⁺), (E) IgM⁺ B2 (i.v.CD45.2^-CD19⁺CD43^loB220^hiIgD^-IgM⁺), (F) IgM⁺ B1 (i.v.CD45.2^-CD19⁺CD43^hiB220^loIgD^-IgM⁺) B cells from lungs extracted at indicated timepoints following 1 (blue circles) or 2 infections (red squares) with Sp19F. (G) Proportion of live EV B cells that exhibit the memory B cell profile IgD^-PD-L2⁺CD73⁺ at the indicated timepoints. (H) Representative flow plots of these populations from day 35 with overlays of Sp19F x1 (blue) and Sp19F x2 (red) groups, as well as numbers of these populations from each group. Mann Whitney test, **p<0.01. Analogous plots are depicted for EV B cells with the non-naive activation/residency markers IgD^-CD69⁺ over the time course (I) and as flow plots with numbers at day 35 (J). n=6 per group, 2 biological replicates.

Figure 3

Lymphocyte recirculation during repeat Sp19F exposure is necessary for early antigen-activated B cell accumulation and lung EV B cell maintenance. (A) Before, during, and after a second Sp19F infection, mice were treated with i.p. FTY720 or saline vehicle at the timepoints indicated by the # symbol. Numbers of live (B) IV (i.v.CD45.2⁺CD19⁺), (C) EV (i.v.CD45.2^-CD19⁺), (D) EV non-I (i.v.CD45.2^-CD19⁺IgD^-), (E) EV naïve (i.v.CD45.2^-CD19⁺IgD⁺), and (F) EV CD69⁺ (i.v.CD45.2^-CD19⁺CD69⁺) B cells were assessed at 2 dpsi, 7 dpsi, and 28 dpsi between vehicle (empty circles) and FTY720 (filled circles) treated groups. (G) GC B cells (i.v.CD45.2^-CD19⁺GL7⁺CD38^lo) and (H) pre-GC B cells (i.v.CD45.2^-CD19⁺GL7⁺CD38^hi) were enumerated at 2dpsi and 7dpsi only. Statistical significance was determined using multiple Mann-Whitney tests, *p<0.05, **p<0.01, ns, nonsignificant. n=5-6 per group, 2 biological replicates. y-axes are formatted in a log scale.

Figure 4

Lung EV B cells exposed to 2 infections undergo self-renewal. (A) Representative flow plots demonstrating manual gating of EdU⁺ EV B cells, gated based on fluorescence minus one controls. Numbers of EdU⁺ (B) IV (i.v.CD45.2⁺CD19⁺), (C) EV (i.v.CD45.2^-CD19⁺), (D) non-naive B2 (i.v.CD45.2^-CD19⁺IgD^-CD43^loB220^hi), (E) naïve (i.v.CD45.2^-CD19⁺IgD⁺), (F) B1 (i.v.CD45.2^-CD19⁺CD43^hiB220^lo), (G) and GC/pre-GC (i.v.CD45.2^-CD19⁺GL7⁺) B cells at day 9, day 14, and day 35 after 1 (empty circles) or 2 (filled circles) Sp19F infections. n=6-9 per group, 3 biological replicates. Analyzed with multiple Mann-Whitney tests, *p<0.05, **p<0.01, ***p<0.001. ns, not significant.

Figure 5

CD4⁺ cell depletion during antigen presentation in the lung affects early EV B cell accumulation. (A) Representative IF image of naïve mouse and Sp-experienced mouse lung BV bundles and alveolar regions stained for CD19 (magenta), CD3 (green), and DAPI (blue) at day 0 and at day 9 (2 dpsi). Scale bars signify 100 μm. (B) GK1.5 vs. IgG2b isotype control was administered i.p. and i.n. to B6 mice before, during, and after the second infection with Sp19F at the timepoints indicated by the # symbol. Mice were euthanized 2 dpsi for flow cytometric analysis. (C) Enumeration of live EV CD4+ T (i.v.CD45.2^-CD4⁺CD3⁺), (D) IV B (i.v.CD45.2⁺CD19⁺), (E) EV B (i.v.CD45.2^-CD19⁺), (F) non-naive B2 B (i.v.CD45.2^-CD19⁺IgD^-CD43^loB220^hi), and (G) CD69⁺ B2 B (i.v.CD45.2^-CD19⁺IgD^-CD43^loB220^hiCD69⁺) cells after treatment with GK1.5 (white bars) or IgG2b (grey bars). (H) Representative overlaid EV B2 B cell CD69 MFI plots normalized to mode from mice treated with GK1.5 (red) vs. IgG2b (blue), with accompanying individual sample MFIs. (I) Enumeration of EV B1 B (i.v.CD45.2^-CD19⁺CD43^hiB220^lo) and (J) EV CD69⁺ B1 B (i.v.CD45.2^-CD19⁺CD43^hiB220^loCD69⁺) cells. (K) Representative overlaid EV B1 B cell CD69 MFI plots normalized to mode from mice treated with GK1.5 (red) vs. IgG2b (blue), with accompanying individual sample MFIs. n=9 per group, 3 biological replicates. Mann-Whitney tests; *p<0.05, **p<0.01, ****p<0.0001; ns, not significant.

Figure 6

CD4⁺ cell depletion during second Sp19F infection abrogates lung B_RM cells. (A) GK1.5 vs. IgG2b isotype control was administered i.p. and i.n. to B6 mice before, during, and after the second infection with Sp19F at the timepoints indicated by the # symbol. Mice were euthanized at 28 dpsi for flow cytometric analysis. (B) Enumeration of live EV B (i.v.CD45.2^-CD19⁺), (C) non-naive B2 B (i.v.CD45.2^-CD19⁺IgD^-CD43^loB220^hi), (D) CSW B (i.v.CD45.2^-CD19⁺IgD^-IgM^-), and (E) B1 B (i.v.CD45.2^-CD19⁺CD43^hiB220^lo) cells after treatment with GK1.5 (white bars with grey stripes) or IgG2b (grey bars with black stripes). Representative flow plots depicting memory (IgD^-PD-L2⁺CD73⁺), (F) and resident (IgD^-CD69⁺), (G) EV B cell populations with overlaid IgG2b (green) and GK1.5-treatment (purple) conditions. Associated graphs depicting each population as a percentage of EV B cells for individual samples are provided. n=5-6 per group, 2 biological replicates. Mann-Whitney tests; *p<0.05, **p<0.01; ns, not significant.

Figure 7

CD40L is dispensable for initial lung B cell accumulation but is required for lung B cell maintenance and the establishment of B_RM cells. (A) MR1 vs. IgG isotype control was administered i.p. and i.n. to B6 mice before, during, and after the second infection with Sp19F at the timepoints indicated by the # symbol. Mice were euthanized at 2 dpsi and 28 dpsi for flow cytometric analysis. Enumeration of live (B) EV B and (C) non-naive B2 B cells (i.v.CD45.2^-CD19⁺IgD^-CD43^loB220^hi) at 2 dpsi after treatment with MR1 (white bars) or IgG (grey bars). (D) Representative overlaid EV B2 B cell CD69 MFI plots normalized to mode from mice treated with MR1 (red) vs. IgG (blue), with accompanying individual sample MFIs. (E) Enumeration of live EV B (i.v.CD45.2^-CD19⁺), (F) non-naive B2 B (i.v.CD45.2^-CD19⁺IgD^-CD43^loB220^hi), (G) CSW B (i.v.CD45.2^-CD19⁺IgD^-IgM^-), and (H) B1 B (i.v.CD45.2^-CD19⁺CD43^hiB220^lo) cells at 28 dpsi after treatment with MR1 (white bars with grey stripes) or IgG (grey bars with black stripes). Representative flow plots depicting memory (IgD^-PD-L2⁺CD73⁺), (I) and resident (IgD^-CD69⁺), (J) EV B cell populations with overlaid IgG (green) and MR1-treatment (purple) conditions. Associated graphs depicting each population as a percentage of EV B cells for individual samples are provided. n=5-6 per group, 2 biological replicates. Mann-Whitney tests; **p<0.01; ns, not significant.

Figure 8

CD4⁺ cells are not required for the recall function of Sp-reactive BRM cells. (A) Acapsular Sp3-reactive IgG, IgA, and IgM titers in BALF from Sp19F-experienced mice in 24 hour increments after respiratory challenge infection with Sp3. n = 3-6 per timepoint across 2 independent experiments. 2way ANOVA; *p<0.05; **p<0.01; ***p<0.001. (B) Experimental timeline to generate anti-pneumococcal immunity in B6 mice followed by treatment with CD4 cell-depleting GK1.5 or IgG2b isotype control antibody 72 and 24 hours before and 24 hours after Sp3 challenge as indicated by the # symbol. (C) Acapsular Sp3-reactive IgG, IgA, and IgM titers in BALF from Sp19F-experienced mice treated with GK1.5 vs. IgG2b during Sp3 challenge were measured via whole-cell pneumococcal ELISAs 48 hpi. n = 5-6 per group across 2 independent experiments. 2way ANOVA.

Figure 9

Lung B and B_RM cells localize to BV bundles and across alveoli in the Sp-experienced lung. (A) Representative immunohistochemical staining of CD19⁺ cells extracted from B6 mouse lungs infected twice with Sp19F i.t. at day 0 and (B) day 35. Scale bars represent 50 μm. (C) Example image tile before and after markup applied in QuPath. BV bundles and associated infiltrates are outlined in blue. Airway and vessel lumens are outlined in grey for exclusion. Positively and negatively stained cells are traced in red and blue, respectively. (D) Quantification of the numbers of positively stained CD19⁺ cells per area (mm²) of alveolar or bronchovascular tissue at each timepoint (day 0 - open circles, day 35 - filled circles) after infection. Slides for blinded analysis were selected at random from 3 independent experiments, totaling 4-5 mice per timepoint across replicates. Multiple Mann-Whitney tests, *p<0.05. (E) Experimental timeline to generate Sp-experience in CD19-Cre PZTD mice followed by challenge with Sp3 in recovered mice. (F) Lung and heart blocks were extracted from Sp-experienced CD19-Cre PZTD mice for live ex vivo imaging; screen captures before and (G) after 24 hpi with Sp3 are presented here. Pulmonary vasculature is marked with fluorescent Cascade blue dextran (blue). The TdTomato endogenous reporter marks CD19⁺PD-L2⁺ cells (magenta), examples denoted by white arrows.