Adenosine metabolized from extracellular ATP promotes type 2 immunity through triggering A2BAR signaling in intestinal epithelial cells

Abstract

Intestinal nematode parasites can cross the epithelial barrier, causing tissue damage and release of danger-associated molecular patterns (DAMPs) that may promote host protective type 2 immunity. We investigate whether adenosine binding to the A_2B adenosine receptor (A_2BAR) on intestinal epithelial cells (IECs) plays an important role. Specific blockade of IEC A_2BAR inhibits the host protective memory response to the enteric helminth, Heligmosomoides polygyrus bakeri (Hpb), including disruption of granuloma development at the host-parasite interface. Memory T cell development is blocked during the primary response, and transcriptional analyses reveal profound impairment of IEC activation. Extracellular ATP is visualized 24 h after inoculation and is shown in CD39-deficient mice to be critical for the adenosine production mediating the initiation of type 2 immunity. Our studies indicate a potent adenosine-mediated IEC pathway that, along with the tuft cell circuit, is critical for the activation of type 2 immunity.

Keywords: CP: Immunology; CP: Molecular biology.

Figure 1.. Intestinal epithelial cell (IEC) A_2BAR signaling promotes worm expulsion.

Villin^Cre-A_2BAR^fl/fl and corresponding Villin^Cre control mice were orally inoculated with 200 Hpb L3. Fourteen days later, mice were treated with the anti-helminthic drug pyrantel pamoate. Six weeks post-clearance, mice were given a secondary (2′) Hpb inoculation; controls included mice given primary (1′) Hpb inoculation (A–C). On day 14, after 2′ inoculation, luminal worm burden was assessed (B), while on day 11, parasite metabolic activity (C) was determined. LysM^cre-A_2BAR^fl/fl mice and LysMCre control mice were orally inoculated as with Villin^Cre-A_2BAR^fl/fl (D and E). Luminal worm burden on day 14 (D) and metabolic activity on day 11 (E) were assessed. Data shown are the means and SEMs from 5–7 individual mice per group (1-way ANOVA, multiple comparisons; *p < 0.05, ***p < 0.001). Experiments were repeated at least 2 times with similar results.

Figure 2.. IEC A_2BAR signaling promotes recruitment of CD4⁺ T cells and alternatively activated (M2) macrophages required to mediate resistance at the host-parasite interface

Control Villin^Cre (A–D) and Villin^Cre-A_2BAR^fl/fl (F–H) mice were orally inoculated with 200 L3 Hpb, and 14 days later, mice were treated with an anti-helminthic drug, pyrantel pamoate. At 6 weeks post-clearance, mice were challenged with a 2′ Hpb inoculation. On day 4 post-2′ inoculation, small intestines were collected. Frozen Swiss-roll sections, 4 μm, stained with Alexa Fluor (AF) 488-F4/80 for macrophages (green), AF647-CD206 for M2 (red) (A and E), AF647-CD4 for T cells (red) (B and F), AF488-MBP for eosinophils (green) (C and G), AF647-CD206 (red), AF488-Ly6G (green) for neutrophils (D and H), and Hoechst 33342 (blue). Arrows, Hpb. Scale bar, 100 μm. Figures are representative of 5 individual mice per group. Experiments were repeated at least 2 times with similar results.

Figure 3.. A_2BAR deficiency in IECs impairs memory CD4⁺ T cell development and initiation of type 2 response after primary Hpb inoculation.

Villin^Cre-A_2BAR^fl/fl and Villin^Cre control mice were orally inoculated with 200 L3 Hpb, and 14 days later, mice were treated with pyrantel pamoate to expulse parasites. Six weeks post-clearance, CD4⁺ T cells were magnetically sorted from mesenteric lymph nodes (MLNs) and spleens. Then, 5 × 10⁶ CD4⁺ T cells from both donor treatment groups were transferred to naive Villin^Cre-A_2BAR^fl/fl recipient mice. Controls included CD4⁺ T cells from naive Villin^Cre-A_2BAR^fl/fl and Villin^Cre mice injected into naive Villin^Cre-A_2BAR^fl/fl recipient mice. Two days post-transfer, mice were inoculated with 200 L3 Hpb (A); 14 days post-infection, worm burden was assessed (B), and small intestine tissues (C) were analyzed by qPCR. (D–F) Naive Villin^Cre-A_2BAR^fl/fl and control Villin^Cre mice were given a primary inoculation with 200 L3 Hpb for 8 days, and controls included naive mice orally gavaged with PBS. Representative flow cytometric analyses of CD4⁺ T cells expressing intracellular phosphorylated STAT6 (D) and the mean percentage of CD4⁺ T cells representing pSTAT6 (E) are shown from MLNs. (F) MLN CD4⁺ T cells from pooled mice were sorted, and the number of IL-4-secreting CD4⁺ T cells determined by ELISpot is shown with technical replicates. (G–J) Villin^Cre-A_2BAR^fl/fl mice and Villin^Cre controls were inoculated with 200 L3 Hpb for 24 h. Small intestinal tissue was analyzed by qPCR. Data from both experiments show the means and SEMs from 4–6 individual mice per group and are representative of at least 2 independent experiments (1-way ANOVA, multiple comparisons; *p < 0.05, **p < 0.01, ***p < 0.001).

Figure 4.. A_2BAR signaling in IECs is required for their activation at the initiation of the primary response to Hpb.

(A–E) Villin^Cre-A_2BAR^fl/fl and corresponding control Villin^Cre mice were orally inoculated with 200 Hpb L3; controls included naive mice orally gavaged with PBS. At 24 h after Hpb inoculation, IECs were sort purified, RNA was extracted, and RNA-seq was performed. Analyses included a heatmap representing the Euclidean distance matrix of the total transcriptome of each sample, along with hierarchical clustering (A); principal-component analysis (PCA) of the total transcriptome of each sample (B); and volcano plots (C); Venn diagram depicting the number of upregulated and downregulated genes with a shared and unique expression from Villin^Cre-A_2BAR^fl/fl and corresponding controls Villin^Cre mice relative to naive (D); regulated gene pathways and processes as determined by enrichment Ingenuity Pathway Analysis (IPA) (E). (F and G) Further analyses revealed heatmaps representing G protein-coupled receptor signaling (F) and type 2 immune response genes (G). (H) Villin^Cre-A_2BAR^fl/fl mice and Villin^Cre controls were inoculated with 200 L3 Hpb for 24 h. RNA from sorted IECs (EpCAM⁺CD45⁻DAPI⁻) was extracted and analyzed for gene expression of Il33 by qPCR. Data shown are the means and SEMs from 4 individual mice per group. qPCR experiment shows the means and SEMs from 4–6 individual mice per group and are representative of at least 2 independent experiments (1-way ANOVA, multiple comparisons; *p < 0.05).

Figure 5.. A_2BAR signaling on IECs activates and induces the production of IL-33 and type 2 markers

(A–D) Villin^Cre-A_2BAR^fl/fl and corresponding control Villin^Cre mice were orally inoculated with 200 Hpb L3 for 24 h; controls included naive mice orally gavaged with PBS. Western blot analysis of IL-33 was performed on whole duodenal intestinal tissue lysates from naive and Hpb-inoculated mice (A). Immunofluorescence staining for IL-33 was detected with anti-IL-33 (green), epithelial cells with anti-EpCAM (red), and nuclear stain Hoechst 3342 (blue) in the proximal small intestine at 20× and 60× (B and C). Quantitation of IL-33⁺ cells (D). (E–I) Villin^Cre-A_2BAR^fl/fl and corresponding control Villin^Cre mice were orally inoculated with 200 Hpb L3 for 48 h; controls included naive mice orally gavaged with PBS. RNA from sorted lamina propria (LP) cells (EpCAM^™CD45⁺DAPI^™) was extracted and analyzed by qPCR for type 2 cytokines and M2 macrophage markers. Blot was cropped to center on an area of interest. Scale bar, 100 μm for 20× and 25 μm for 60× magnification. Data shown are the means and SEMs from 3 individual mice per group and are representative of at least 2 independent experiments (1-way ANOVA, multiple comparisons; *p < 0.05, **p < 0.01, ***p < 0.001).

Figure 6.. IEC A_2BAR signaling contributes to tuft cell hyperplasia

Villin^Cre-A_2BAR^fl/fl and control Villin^Cre mice were orally inoculated with 200 Hpb L3 for 24 h; controls included naive mice orally gavaged with PBS. (A–D) Tuft cell frequency in the proximal small intestine was assessed by staining with α-DCLK1 (green), marking tuft cells and nuclear stain Hoechst 3342 (blue). (E) Quantitation of the tuft cells. Scale bars, 100 μm. Data show the means and SEMs from 4–5 individual mice per group and are representative of at least 2 independent experiments (1-way ANOVA, multiple comparisons; **p < 0.01).

Figure 7.. Adenosine metabolized from extracellular ATP promotes type 2 immunity through triggering A_2BAR signaling on IECs.

(A and B) pmeLuc mice were orally inoculated with 200 L3 Hpb or PBS vehicle. Mice were imaged 24 h post-gavage using the Xenogen IVIS-200 System for the detection of ATP (A) and measurement of total flux (B). (C) CD39^™/™ mice and WT controls were orally inoculated with 200 L3 Hpb. Mice were treated with pyrantel pamoate 14 days later. At 6 weeks post-clearance, mice were challenged with a 2′ Hpb inoculation; controls included naive mice given 1’ Hpb inoculation. Resistance was assessed on day 14 by determining luminal worm burden. CD39^™/™ mice and WT controls were infected with 200 L3 Hpb for 8 days, and controls included naive mice orally gavaged with PBS. (D–F) MLN tissue was analyzed by qPCR for type 2 and type 1 cytokines. (G and H) Cell suspensions from MLNs were assessed for intracellular pSTAT6 cell surface expression by CD4⁺ T cells, with a representative plot (G) and mean percentage of CD4⁺ T cells expressing pSTAT6 (H). (I–N) Small intestinal tissue was analyzed by qPCR for type 2 cytokines at 8 days (I–K) and 24 h after inoculation (I–N). Data from both experiments show the means and SEMs from 4–5 individual mice per group and are representative of at least 2 independent experiments (1-way ANOVA, multiple comparisons; *p < 0.05, **p < 0.01, ***p < 0.001).