Group 2 innate lymphoid cells are a non-redundant source of interleukin-5 required for development and function of murine B1 cells

Abstract

Tissue-resident immune cells, such as innate lymphoid cells, mediate protective or detrimental immune responses at barrier surfaces. Upon activation by stromal or epithelial cell-derived alarmins, group 2 innate lymphoid cells (ILC2s) are a rapid source of type 2 cytokines, such as IL-5. However, due to the overlap in effector functions, it remains unresolved whether ILC2s are an essential component of the type 2 response or whether their function can be compensated by other cells, such as T cells. Here we show a non-redundant role of ILC2s in supporting the development and function of B1 cells. We demonstrate that B1 cells fail to develop properly in the absence of ILC2s and identify the IL-33 receptor on ILC2s as an essential cell-intrinsic regulator of IL-5 production. Further, conditional deletion of Il5 in ILC2s results in defective B1 cell development and immunoglobulin production. Consequently, B1 cells with phosphatidylcholine specific B cell receptor rearrangements are diminished in ILC2-deficient mice. Thus, our data establish an essential function of ILC2s in supporting B1 cells and antibody production at barrier surfaces.

Fig. 1. ILC2s shape the B1 cell pool.

a Flow cytometric plots of B1 cells in Nmur1^iCre-eGFP Id2^fl/fl (blue) and littermate Id2^fl/fl mice (white) in the peritoneal cavity in steady state. B1 cells were pre-gated on live CD45⁺ TCRβ^-. First gate shows B1 cells gated as IgM⁺ CD23⁻. Second gate shows B1 cells gated as CD19⁺ CD23⁻, further gating of B1 cells included CD11b. B1a and B1b cells were subdivided using the marker CD5. b Quantification of B1 cells as in (a) (Id2^fl/fl n = 4, Nmur1^iCre-eGFP Id2^fl/fl n = 3). c–f Flow cytometric plots and quantification of B1 cells in the (c), Thoracal lavage (Id2^fl/fl n = 6, Nmur1^iCre-eGFP Id2^fl/fl n = 4), d Omentum (Id2^fl/fl n = 4, Nmur1^iCre-eGFP Id2^fl/fl n = 3) e Lung (Id2^fl/fl n = 5, Nmur1^iCre-eGFP Id2^fl/fl n = 4) and (f) spleen (Id2^fl/fl n = 8, Nmur1^iCre-eGFP Id2^fl/fl n = 7). g Relative and absolute cell numbers of B1 cells across different ages in Nmur1^iCre-eGFP Id2^fl/fl (blue, 3 weeks n = 3, 13 weeks n = 4, 18 weeks n = 5, 30 weeks n = 4) and littermate Id2^fl/fl mice (black, 3 weeks n = 4, 13 weeks n = 4, 18 weeks n = 4, 30 weeks n = 5). h, i Flow cytometric plots of B1 cells in Nmur1^iCre-eGFP Gata3^fl/fl (green) and littermate Gata3^fl/fl (white) mice in the (h), peritoneal cavity (Gata3^fl/fl n = 4, Nmur1^iCre-eGFP Id2^fl/fl n = 5) and (i), thoracal lavage (Gata3^fl/fl n = 4, Nmur1^iCre-eGFP Id2^fl/fl n = 3) at steady state. j Histology section of omental Fat-Associated-Lymphoid-Clusters (FALCs) from the ILC2 reporter mouse Nmur1^iCre-eGFP Rosa26^flSTOP-RFP/+ (violet signal) stained with anti-KLRG1 (light-blue), anti-CD5 (dark blue), anti-B220 (red), anti-CD43 (green), and DAPI. Yellow arrows identify ILC2s (RFP⁺ KLRG1⁺ CD5⁻), blue arrows point towards B1 cells (B220^low, CD43⁺). Scale bar overview 100 µm, zoom 20 µm. Each symbol represents data from one mouse, mean +/− SD, all data are representative of at least two independent experiments. Statistical significance was determined by two-tailed unpaired Student’s t-test (b–i), *p < 0.05 **p < 0.01, ***p < 0.001, ****p < 0.0001. Source data, including exact p-values, are provided as a Source data file Fig. 1.

Fig. 2. ILC2s expand IL5ra⁺ B1 cells.

a Clustering of flow-sorted and sequenced B cells at a resolution 0.3. b Proportions of the clusters of Nmur1^iCre-eGFP Id2^fl/fl vs. littermate Id2^fl/fl mice. c B1- and B2-specific genes in each Cluster were plotted according to their average expression and percentage of expression. d Annotated and color-coded B1 vs. B2 cells. e Heatmap of top 30 differentially-expressed genes in B1 cells of Nmur1^iCre-eGFP Id2^fl/fl vs. littermate Id2^fl/fl mice. f UMAPs of Ccnd2, Ass1, Anxa2, Zcwpw1. g Validation by qPCR of downregulated B1 cell genes. B1 cells were sort-purified from the peritoneal cavity (Id2^fl/fl n = 5, Nmur1^iCre-eGFP Id2^fl/fl n = 3). h Top 10 differentially-regulated pathways in Cluster 0 of the UMAP in a. i UMAP showing Mki67 in B cells, j flow cytometric plots and quantification of Ki67⁺ B1 cells (Pre-gated on Live CD45⁺ CD19⁺ CD23^-) in Nmur1^iCre-eGFP Id2^fl/fl (blue, PL, peritoneal lavage n = 7, TL, thoracal lavage n = 5, Spleen n = 4, Omentum n = 7) and littermate Id2^flox/flox mice (white, PL n = 6, TL n = 6, Spleen n = 5, Omentum n = 6). k Expression of cytokine-receptors on B1 and B2 cells in ILC2-sufficient mice. l UMAP of IL5ra⁺ B cells. m differentially- regulated gene expression of reported IL-5-induced B1 cell genes of Nmur1^iCre-eGFP Id2^fl/fl vs. (blue) and littermate Id2^flox/flox mice. n B1 cells were sort-purified from Il5^Cre/Cre (yellow, n = 5) and littermate Il5^Cre/+ mice (white, n = 7) and qPCR was performed for the indicated genes. Each symbol represents data from one mouse, mean +/− SD, data in (g, j, n) are representative of two independent experiments. Statistical significance was determined by two-tailed unpaired Student’s t-test (g) or Mann Whitney test (j, n), *p < 0.05 **p < 0.01, ***p < 0.001, ****p < 0.0001. Source data, including exact p-values, are provided as a Source data file Fig. 2.

Fig. 3. ILC2-derived IL-5 is required for proper B1 cell development.

a Flow cytometric plots and quantification of B1 cells in Il5^Cre/Cre (yellow, n = 4) and littermate Il5^Cre/+ mice (white, n = 7) in the peritoneal cavity and thoracal lavage at steady state. b Histology section of omental FALCs from Il5^Cre/+ Rosa26^flSTOP-YFP/+ mice. Staining with anti-B220 (red), anti-CD43 (green), anti-KLRG1 (light-blue), anti-CD5 (blue), endogenous IL-5 (Violet) and DAPI. Il-5 producing ILC2s (yellow arrows, Il5⁺KLRG1⁺CD5^-) reside in close proximity to B1 cells (blue arrows, B220^lowCD43⁺). Scale bars 50 µm. c Validation of the mouse line Nmur1^iCre-eGFP Il5^fl/fl by qPCR of sort-purified ILC2s from the lung (Il5^fl/fl n = 4, Nmur1^iCre-eGFP Il5^fl/fl n = 3). d IL-5 and Il-13 secretion was determined by flow cytometry of cultured immune cells isolated from the small intestine of Nmur1^iCre-eGFP Il5^fl/fl mice and littermate controls, stimulated with PMA/ionomycin (PMA/Iono) for 4 h (Il5^fl/fl n = 3, Nmur1^iCre-eGFP Il5^fl/+ n = 3, Nmur1^iCre-eGFP Il5^fl/fl n = 3). e, f Relative ILC2 numbers in the (e), small intestine and (f), mesenteric lymph nodes (MLN) (Il5^fl/fl n = 3, Nmur1^iCre-eGFP Il5^fl/fl n = 3). g–l Flow cytometric plots and quantification (Il5^fl/fl n = 3, Nmur1^iCre-eGFP Il5^fl/fl n = 3) of (g, i, k), B1 cells (cells were pre-gated on live CD45⁺ TCRβ⁻) and (h, j, l), CD11b⁺ B1 cells in the (g, h) peritoneal lavage, (i, j), the thoracal lavage and (k, l), the omentum. m Flow cytometric quantification of B1 cells in 2 weeks old Nmur1^iCre-eGFP Il5^fl/fl mice and littermate controls (Il5^fl/fl n = 9, Nmur1^iCre-eGFP Il5^fl/fl n = 8). n, o Flow cytometric plots and quantification of B1 cells in the (n), peritoneal cavity and the (o), thoracal lavage in littermate, co-housed Cd4^Cre Il5^fl/fl mice and controls (Il5^fl/fl n = 7, Cd4^Cre Il5^fl/fl n = 5). Each symbol represents data from one mouse, mean +/− SD, all data are representative of at least two independent experiments. Statistical significance was determined by two-tailed unpaired Student’s t-test (a, e–o) or one-way ANOVA (c, d), n.s. non-significant, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Source data, including exact p-values, are provided as a Source data file Fig. 3.

Fig. 4. ILC2s support the expansion of phosphatidylcholine specific B1 cells.

a Clustering of flow-sorted and sequenced B cells from the peritoneal lavage at a resolution 0.3. b Proportions of the clusters of Nmur1^iCre-eGFP Id2^fl/fl vs. littermate Id2^fl/fl mice. c Annotated and color-coded B1 and B2 cells in the peritoneal lavage. d UMAP of IL5ra⁺ B cells. e UMAPs of Ighm and Ighd. f Heatmap of differentially-regulated genes including BCR genes comparing Nmur1^iCre-eGFP Id2^fl/fl and littermate Id2^fl/fl mice. g UMAPs and (h), quantification of the most important phosphatidyl-specific heavy- and corresponding light chains, Ighv11-2, Igkv14-126 and Ighv12−3, Igkv4-91 in Nmur1^iCre-eGFP Id2^fl/fl and littermate Id2^fl/fl mice. i Ighv usage in single B1 cells from 5 mice from Nmur1^iCre-eGFP Id2^fl/fl (blue) and littermate Id2^fl/fl mice (white). Ighv gene segments are ordered by their relative proximity to the D segments. Number of cells is depicted in the Figure.

Fig. 5. Serum IgM and IgE levels are dependent on the presence of ILC2s.

a–c Immunglobulin concentrations in the serum of Nmur1^iCre-eGFP Id2^fl/fl and littermate Id2^fl/fl mice were determined by using the LEGENDplex multiplex beads-based assay. a Total immunoglobulin concentration in steady state (Id2^fl/fl n = 8, Nmur1^iCre-eGFP Id2^fl/fl n = 8). b Mice were infected with N. brasiliensis and serum was harvested at d11 post-infection (Id2^fl/fl n = 8, Nmur1^iCre-eGFP Id2^fl/fl n = 9). c Inflammation was induced by intranasal administration of Alternaria alternata (A. alternata) extract. Sera were harvested at d7 after induction of inflammation (Untreated Id2^fl/fl n = 4, A. alternata Id2^fl/fl n = 12, A. alternata Nmur1^iCre-eGFP Id2^fl/fl n = 12). Each symbol represents data from one mouse, mean +/− SD, data are pooled from two (a, b) or three (c) independent experiments. Statistical significance was determined by two-tailed unpaired Mann Whitney test (a–c), *p < 0.05, **p < 0.01. Source data, including exact p-values, are provided as a Source data file Fig. 5.

Fig. 6. IL-33 triggers IL-5 production in ILC2 to stimulate B1 cells.

a–d Flow cytometric plots and quantification of B1 cells in (a), the peritoneal lavage (Il33^+/+ n = 7, Il33^−/− n = 7), b the thoracal lavage (Il33^+/+ n = 6, Il33^-/- n = 4), c omentum (Il33^+/+ n = 6, Il33^−/− n = 3) and (d) lung (Il33^+/+ n = 5, Il33^-/- n = 3) of Il33^-/- and control mice. e, f Flow-cytometric analysis of Ki-67 of B1 cells (Pre-gated on Live CD45⁺ CD19⁺ CD23⁻) in Il33^−/− versus control mice in the (e), peritoneal lavage (Il33^+/+ n = 5, Il33^-/- n = 4) and the (f) thoracal lavage (Il33^+/+ n = 4, Il33^−/− n = 3). g–j Flow-cytometric plots and quantification of B1 cells of Nmur1^iCre-eGFP Il1r1^fl/fl (ST2) mice and littermate controls in (g), the peritoneal lavage (Il1rl1^fl/fl n = 4, Nmur1^iCre-eGFP Il1rl1^fl/fl n = 5) h the thoracal lavage (Il1rl1^fl/fl n = 4, Nmur1^iCre-eGFP Il1rl1^fl/fl n = 4), i omentum (Il1rl1^fl/fl n = 4, Nmur1^iCre-eGFP Il1rl1^fl/fl n = 4), and (j) lung (Il1rl1^fl/fl n = 4, Nmur1^iCre-eGFP Il1rl1^fl/fl n = 4). k Sort-purified small intestinal ILC2s gated as Live CD45⁺ Lin⁻ (CD3, CD5, CD19, Fcεrl, Ly6G) CD127⁺ KLRG1⁺ of Tslpr^−/− (Tslpr^+/+ n = 5, Tslpr^−/− n = 5), Il17rb^−/− (Il17rb^+/+ n = 5, Il17rb^−/− n = 4)⁻, Il33^−/− (Il33^+/+ n = 5, Il33^−/− n = 5) and Nmur1^iCre-eGFP Il1rl1^fl/fl (Il1rl1^fl/fl n = 6, Nmur1^iCre-eGFP Il1rl1^fl/fl n = 7) and qPCR for Il5. Each symbol represents data from one mouse, mean +/− SD, data are representative of at least two independent experiments. Statistical significance was determined by two-tailed unpaired Student’s t-test (a–j) or Mann Whitney test (k), n.s. non-significant, *p < 0.05, **p < 0.01, ***p < 0.001. Source data, including exact p-values, are provided as a Source data file Fig. 6.