Rapid group-2 innate lymphoid cell mobilization from the intestine aids in early lung defense and repair

Abstract

Circulating group 2 innate lymphoid cells (ILC2s) comprise a third of all lung ILC2s shortly after helminth infection, providing an early source of type-2 cytokines. However, the origin of circulating ILC2s and their relationship with ILC progenitors and tissue-resident ILC2s remain unresolved. Single-cell RNA-sequencing and trajectory analysis of ILC2 and ILC2 progenitors in the lung, bone marrow (BM), and intestine draining lymph nodes (LNs) revealed that circulating ILC2s in the lung mirror a population of ILC2s in the mesenteric LNs. Conversely, lung-resident ILC2s closely resembled BM ILC progenitors. These BM progenitors gave rise to lung-resident ILC2s but not circulating ILC2s after adoptive transfer. Definitive proof of an intestinal origin for circulating ILC2s in the lung was achieved through in vivo photoconversion of the intestine. These findings emphasize how rapid deployment of intestinal ILC2s to distal sites of type-2 inflammation bolsters barrier immunity to allow time for tissue-resident ILC2 expansion.

Keywords: CP: Immunology; IL-13: iILC2; IL-4; ILC2; Type-2 inflammation; gut-lung axis; helminth; innate lymphoid cells; intestine; lung.

Figure 1.. Single-cell transcriptomic and trajectory analysis support an intestinal origin for pulmonary iILC2

(A) Schematic of the scRNA-seq workflow. See also Figure S1. (B) UMAP plot showing unsupervised clustering (C1–C7) of cells from BM, Mes LN, and lung of three mice combined. See also Figure S2. (C) Heatmap of the top 10 differentially expressed genes of each cluster (C1–C6) relative to all clusters. Cluster C4 is shown twice to represent manually separated tissue subsets: the top C4 row shows genes in the Mes LN ILC2-B cluster, and the bottom C4 row represents mostly expressed genes in Lung iILC2-A cluster. (D) Violin plots showing the distribution and level of expression of canonical ILC-related genes across clusters (C1–C6). See also Figure S3. (E) UMAP plot showing trajectory analysis predicting the progression of CHILPs (red star) into mature ILC2 populations based on data from 5 dpi. The red line indicates predicted branching points in the trajectory suggestive of key lineage bifurcation points as determined by Monocle3. Black arrows summarize likely trajectory paths (P1–P3).

Figure 2.. Despite evidence of historical IL-13 expression among all nILC2s, migrating iILC2s serve as the predominant source of IL-13 in the lung early after infection

(A) Dot plot showing cytokine expression profile of mesenteric (Mes) LN and lung clusters from the scRNA-seq data presented in Figure 1 5 days after N. brasilensis infection. Dot size represents the frequency of cells expressing a given gene, while color indicates the level of expression. (B) Contour plots of Lin-CD45⁺CD127⁺KLRG1⁺ILC2 from the lungs pre-infection (0 dpi; top) and 5 dpi (bottom). Gates reflect historical (tdTomato+YFP−) or active (tdTomato+YFP+) IL-13 production. (C) Contour plots gated from historical and active IL-13-producing populations in (B) at 0 dpi (top) and 5 dpi (middle and bottom). Gates represent the percentage of CD90⁺KLRG1^low nILC2 (blue) and CD90^low/−KLRG1^high iILC2 (red) among total ILC2s. (D) Bar graphs show percentage of nILC2s and iILC2s at stated days post infection. Mean ± SEM; n = 7–8 mice, from two independent experiments. Two-tailed paired t test (*p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001).

Figure 3.. Temporal analysis of ILC2 populations across tissues after Nippostrongylus brasiliensis infection

(A) Contour plots of Lin-CD45⁺CD127⁺KLRG1⁺ ILC2 in the lung, mediastinal (Med) LN, mesenteric (Mes) LN, bone marrow (BM), and blood. Blue gates represent Arg1⁺ α4β7⁻nILC2 and red gates represent Arg1⁻ α4β7⁺ iILC2. See also Figure S4. (B) Graphs represent the number of nILC2s and iILC2s in each tissue over time. Mean ± SEM. (C) Bar graphs of iILC2 (red) and nILC2 (blue) numbers at 0, 3, 4, and 5 dpi in indicated tissues. Mean ± SEM; n = 6–8 mice, from two independent experiments. One-way ANOVA followed by Tukey’s post hoc test (*p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001).

Figure 4.. FTY720 treatment promotes iILC2 accumulation in the Mes LN with a corresponding loss of iILC2 in the lung and BM after N. brasiliensis infection

(A) Contour plots of Lin-CD45⁺CD127⁺KLRG1⁺ILC2 in the lung, Mes LN, and BM at 5 dpi without (left) or with (right) FTY720 treatment. Gates depict Arg1⁺ α4β7⁻nILC2 and Arg1⁻ α4β7⁺ iILC2. (B) Bar graphs of nILC2 and iILC2 numbers in each tissue of untreated (black) or treated (red) mice. Means ± SEM; n = 6 mice, from two independent experiments. Two-tailed unpaired t test (*p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001).

Figure 5.. Adoptively transferred CHILPs and ILC2ps from bone marrow give rise to nILC2 but not iILC2

(A) Schematic representation of the adoptive transfer experiment. See also Figure S5. (B) Representative contour plot of total Lin⁻CD45⁺CD127⁺KLRG1⁺ ILC2s 5 dpi with gates depicting Arg1⁺ KLRG1^low nILC2 and Arg1⁻ KLRG1^high ILC2 subsets. (C and D) Contour plots (C) and bar graphs (D) representing the percentage and number of transferred cells found in nILC2 (top) and iILC2 (bottom) gates normalized to input. Mean ± SEM; n = 5 mice, from four independent experiments. Two-tailed unpaired t test (*p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001).

Figure 6.. Bone marrow cells do not contribute to pulmonary ILC2s 5 days after helminth infection

(A) Contour plots of Lin⁻CD45⁺CD127⁺ bone marrow and Lin⁻CD45⁺CD127⁺KLRG1⁺ lung cells at 5 dpi. (B and C) Contour plots (B) and bar graphs (C) representing the number of photoactivated KikRed⁺Lin⁻CD45⁺CD127⁺ BM cells (top) and KikRed⁺CD90⁺KLRG1^low lung nILC2 and KikRed⁺CD90^low/−KLRG1high lung iILC2s (bottom). Mean ± SEM; n = 9 mice, from two independent experiments. Two-tailed unpaired t test (*p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001).

Figure 7.. Photoactivated intestinal cells arrive in the lungs as iILC2s during helminth infection

(A) Contour plots depict Lin⁻CD45⁺CD127⁺KLRG1⁺ Mes LN (top) and lung (bottom) cells at 5 dpi. See also Figure S6. (B and C) Representative contour plots (B) and bar graphs (C) showing the number of KikRed⁺Lin⁻CD45⁺CD127⁺KLRG1⁺ Mes LN cells (top; n = 4 mice, from one experiment) and KikRed⁺CD90⁺KLRG1^low nILC2 and KikRed⁺CD90^low/−KLRG1^high iILC2s from the lung (middle and bottom); n = 6–8 mice, from two independent experiments). Mean ± SEM. Two-tailed unpaired t test (*p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001).