Piezo1 channels restrain ILC2s and regulate the development of airway hyperreactivity

Abstract

Mechanosensitive ion channels sense force and pressure in immune cells to drive the inflammatory response in highly mechanical organs. Here, we report that Piezo1 channels repress group 2 innate lymphoid cell (ILC2)-driven type 2 inflammation in the lungs. Piezo1 is induced on lung ILC2s upon activation, as genetic ablation of Piezo1 in ILC2s increases their function and exacerbates the development of airway hyperreactivity (AHR). Conversely, Piezo1 agonist Yoda1 reduces ILC2-driven lung inflammation. Mechanistically, Yoda1 inhibits ILC2 cytokine secretion and proliferation in a KLF2-dependent manner, as we found that Piezo1 engagement reduces ILC2 oxidative metabolism. Consequently, in vivo Yoda1 treatment reduces the development of AHR in experimental models of ILC2-driven allergic asthma. Human-circulating ILC2s express and induce Piezo1 upon activation, as Yoda1 treatment of humanized mice reduces human ILC2-driven AHR. Our studies define Piezo1 as a critical regulator of ILC2s, and we propose the potential of Piezo1 activation as a novel therapeutic approach for the treatment of ILC2-driven allergic asthma.

Graphical abstract

Figure 1.

ILC2 activation induces Piezo1 expression in the lungs. (A) Pure populations of lung ILC2s were FACS-sorted from mice challenged with either PBS or IL-33 and profiled by droplet-based scRNA-seq (Wallrapp et al., 2017). Cells (dots) are colored based on in vivo treatment, either PBS (blue) or IL-33 (red). UMAP projections showing total ILC2s (left panel) and Piezo1⁺ ILC2s (right panel). (B) Corresponding quantitation of Piezo1⁺ ILC2s in PBS vs. IL-33–treated mice. Half violin plots represent overlaid density plots illustrating the distribution of Piezo1 expression levels in each group. (C) Piezo1-positive % fold change (FC) relative to other known MSCs in IL-33–treated lung ILC2s, classified by group/color: red/orange: ENaC/ASIC; yellow: Piezo; green: TREK; blue: TRP; and pink/purple: TMEM16/Ano. (D) Radar plot depicting selected ILC2 signature activation genes induced by IL-33 treatment. Red dots represent log(fold change), and black dots represent −log₁₀ (P value). (E and F) Heat map (E) and dot plot (F) representation of genes isolated in D in comparison to Piezo1^high- and Piezo1^low-expressing cells. (G) Schematic description of ex vivo lung ILC2 Piezo1 analysis. Pure populations of naïve ILC2s were FACS-sorted from the lungs of Piezo1dT mice and cultured ex vivo with rmIL-2 and rmIL-7 (both 10 ng/ml) for the indicated times with or without rmIL-33 (20 ng/ml). The experiment was performed twice. (H) Representative plots of Piezo1 expression at the indicated times and corresponding quantitation presented as Piezo1 mean fluorescence intensity (MFI). n = 3. WT: C57BL/6 ILC2s. (I) Schematic description of in vivo lung immune cell Piezo1 analysis. Cohorts of Piezo1dT mice were challenged i.n. for 3 consecutive days with or without 0.5 µg rmIL-33, and on day 4, lung immune cells were analyzed by flow cytometry. The experiment was performed twice. (J and K) Representative plot of ILC1 (T-bet⁺), ILC2 (GATA3⁺), and ILC3 (RORgt⁺) Piezo1 expression on day 4 (J) and corresponding quantitation presented as Piezo1dT MFI (K). n = 5. (L and M) Representative plot of Lineage⁺ and ILC2 Piezo1 expression on day 4 (L) and corresponding quantitation presented as Piezo1dT MFI (M). n = 5. (N) Piezo1 expression in ILC2s, T cells (CD45⁺CD3⁺), eosinophils (CD45⁺, CD11c⁻, SiglecF⁺), neutrophils (PMN, CD45⁺, SiglecF⁻, Ly6G⁺, CD11b⁺), macrophages (CD45⁺, Ly6G⁻, SiglecF⁻, CD11c⁺), and monocytes (CD45⁺Ly6G⁻, SiglecF⁻Ly6C⁺). n = 5. The experiment was performed twice. (O–Q) Piezo1dT mice were challenged i.n. for 3 consecutive days with rmIL-33 and pure populations of Piezo1dT-low–expressing (Piezo1^low, blue) and Piezo1dT-high–expressing (Piezo1^high, red) ILC2s were FACS-sorted on day 4 from the lungs for the indicated readouts. The experiment was performed twice. (O) Representative dot plots showing gating strategies for Piezo1^high and Piezo1^low populations and pure populations of Piezo1^low and Piezo1^high ILC2s. (P) Representative plots of intranuclear GATA-3 expression and corresponding quantitation were presented as GATA-3 MFI. n = 3. (Q) Pure populations of Piezo1^low and Piezo1^high ILC2s were further cultured ex vivo for 18 h with rmIL-2 and rmIL-7, and levels of IL-5 and IL-13 production in the culture supernatant were measured by ELISA. n = 3. Data are presented as mean ± SEM. A two-tailed Student’s t test for unpaired data was applied for comparisons between two groups (H, K, M, P, and Q), except for multigroup comparisons where Tukey’s multiple comparison one-way ANOVA tests were used (N). *P < 0.05, **P < 0.01, ***P < 0.001, ns: non-significant.

Figure S1.

Gating strategies and viability of Piezo1^high and Piezo1^low ILC2s. (A and B) ILC2 gating strategy (A) and myeloid cell/T cell gating strategy (B). (C) Gating strategy used for the intranuclear staining of ILC1, ILC2, and ILC3. (D) Representative plots of live, early apoptotic (E.A.), and late apoptotic/necrotic (L.A.) ILC2s and corresponding quantitation presented as the frequencies (%) of live, E.A., and L.A. ILC2s. n = 3. The experiment was performed twice. (E) Pure populations of lung ILC2s were FACS-sorted from mice challenged with either PBS or IL-33 and profiled by droplet-based scRNA-seq (Wallrapp et al., 2017). Gene set enrichment analysis performed by IPA depicting top actin remodeling pathways induced by IL-33 treatment in murine lung ILC2s. All pathways shown are enriched in IL-33–treated ILC2s compared with PBS-treated ILC2s. The size of the dots represents the number of gene overlaps in each pathway, colors indicated P value. Data are presented as mean ± SEM. A two-tailed Student’s t test for unpaired data was applied for comparisons between two groups (D). **P < 0.01.

Figure S2.

Effects of IL-33, IL-25, TSLP, NMU, and VIP on ILC2 Piezo1 expression. (A) Cohorts of Piezo1dT mice were challenged i.n. for 3 consecutive days (0.5 µg rmIL-33) or 4 consecutive days (100 µg A. Alternata). Cohorts of mice challenged with PBS were used as controls. The expression of Piezo1 in lung ILC2s was measured 24 h after the last intranasal challenge by flow cytometry. (B) Representative flow cytometry plots of Piezo1 expression and corresponding quantitation, presented as Piezo1dT MFI. n = 5. The experiment was performed three times. (C) Cohorts of Piezo1dT mice were challenged i.n. for 3 consecutive days with either rmIL-33, rmIL-25, or rmTSLP (all 0.5 µg/mouse). Cohorts of mice challenged with PBS were used as controls. The expression of Piezo1 in lung ILC2s was measured 24 h after the last intranasal challenge by flow cytometry. (D) Representative flow cytometry plots of Piezo1 expression and corresponding quantitation, presented as Piezo1dT MFI. n = 5. The experiment was performed three times. (E) Schematic description of ex vivo lung ILC2 Piezo1 analysis. Pure populations of naïve ILC2s were FACS-sorted from the lungs of Piezo1dT mice and cultured ex vivo with rmIL-2 and rmIL-7 (both 10 ng/ml) for the indicated times with or without rmIL-33 (20 ng/ml), VIP (10 µM), or NMU (100 ng/ml) for 72 h. (F) Representative flow cytometry plots of Piezo1 expression and corresponding quantitation, presented as Piezo1dT MFI. n = 3. The experiment was performed twice. (G) Pure populations of lung ILC2s were FACS-sorted from mice challenged with either PBS, IL-33, IL-25, NMU, or NMU+IL-25 and profiled by droplet-based scRNA-seq (Wallrapp et al., 2017). Cells (dots) are colored based on in vivo treatment, either PBS (blue) or IL-33 (red), NMU (purple), IL-25 (green), and NMU+IL-25 (orange). tSNE1 projections showing total ILC2s (left panel) and Piezo1⁺ ILC2s (middle panel) and an overlay (right panel). (H) Frequencies of Piezo1⁺ cells in each experimental condition. Data are presented as mean ± SEM. Tukey’s multiple comparison one-way ANOVA tests were used (B, D, and F). ***P < 0.001, ns: non-significant.

Figure S3.

Generation and validation of Il7r^cre Piezo1^fl/fl mice. (A) Representative bands from DNA gels for genotyping and generation of Il7r^crePiezo1^fl/fl mice. (B) Il7r^cre and Il7r^cre Piezo1^fl/fl mice were i.n. challenged on days 1–3 with 0.5 µg IL-33. On day 4, lungs were processed to single-cell suspensions and stained for ILC2s and Piezo1. Piezo1 MFI. n = 3. The experiment was performed twice. (C) Piezo1 expression in indicated ILC, T cell, and myeloid populations isolated from the lungs of Il7r^cre and Il7r^crePiezo1^fl/fl mice. n = 3. The experiment was performed twice. (D) Representative flow cytometry plots of ILC2 intranuclear KLF2 expression and quantitation presented as KLF2 MFI. n = 5. The experiment was performed twice. Data are presented as mean ± SEM. A two-tailed Student’s t test for unpaired data was applied for comparisons between two groups (B–D). ***P < 0.001. ns: non-significant. Source data are available for this figure: SourceData FS3.

Figure 2.

ILC2s lacking Piezo1 are metabolically and functionally more active. Il7r^cre and Il7r^crePiezo1^fl/fl mice were challenged i.n. for 3 consecutive days with rmIL-33, and pure populations of lung ILC2s were isolated on day 4 and cultured with rmIL-2 and rmIL-7 for 18 h. Cells were then collected for the indicated readouts. (A) Levels of IL-5, IL-6, and IL-13 production in the culture supernatant. n = 3. The experiment was performed three times. (B) Representative plots of intranuclear GATA-3 expression and corresponding quantitation are presented as GATA-3 MFI. n = 3. The experiment was performed twice. (C–H) Cells were collected, and RNA was extracted to perform a bulk transcriptomic analysis. (C) Volcano plot comparison between Il7r^cre vs. Il7r^crePiezo1^fl/fl. (D) PCA plot. (E) Venn diagram depicting the number of transcripts differentially modulated between Il7r^cre and Il7r^crePiezo1^fl/fl ILC2s differential analysis, with a statistical cutoff of P < 0.05. (F) Chord plot representing the highest differentially expressed genes (DEG) from the top five upregulated pathways. Specific pathways are color-coded and represented in the right inner bands, where chords gather. Outer bands on the right depict the IPA −log₁₀ P value. The left inner bands represent the gene −log₁₀ P value. Outer bands on the left represent the gene log₂(fold change). (G) Radar plot depicting selected genes from the ILC2 inflammation and the Piezo1 pathway. Red dots represent log(fold change), and black dots represent −log₁₀ (P value). (H) Differential expression of TCA cycle-related genes between Il7r^cre and Il7r^crePiezo1^fl/fl ILC2s. Gray histograms represent −log₁₀ (P value), and the dotted line represents P < 0.05. Inner bands represent gene log(fold change). (I) Differential expression of genes between Il7r^cre and Il7r^crePiezo1^fl/fl ILC2s involved in the mitochondrial respirasome and OXPHOS. Gray histograms represent −log₁₀ (P value), and the dotted line represents P < 0.05. Inner bands represent gene log(fold change). (J–M) Mitochondrial respiratory profile showing OCRs in response to oligomycin (ATP synthase inhibitor), BAM15 (mitochondrial uncoupler), and rotenone + antimycin A (complex I and II inhibitors) sequential injections (J) and corresponding key parameters of mitochondrial function (K) basal respiration, (L) spare respiratory capacity, and (M) ATP production rates. n = 3. The experiment was performed twice. Data are presented as mean ± SEM. A two-tailed Student’s t test for unpaired data was applied for comparisons between two groups (A, B, and K–M). *P < 0.05, **P < 0.01, ***P < 0.001, and ns: non-significant.

Figure 3.

Lack of Piezo1 in ILC2s increases AHR and lung inflammation. (A) Schematic description of in vivo effects of Piezo1-deficient ILC2s on the development of AHR. Il7r^cre and Il7r^crePiezo1^fl/fl mice received 0.5 µg rmIL-33 or PBS i.n. on days 1–3. On day 4, lung function, lung ILC2s, BAL cellularity, and histology were analyzed. The experiment was performed twice. (B and C) Lung resistance (B) and dynamic compliance (C) in response to increasing doses of methacholine. n = 4. (D) Total number of ILC2s per lung. n = 6. (E) Representative plots of live, early apoptotic (E.A.), and late apoptotic/necrotic (L.A.) ILC2s and corresponding quantitation presented as the frequencies (%) of live, E.A., and L.A. ILC2s. n = 6. (F and G) Representative plots of intranuclear (F) Ki67 and (G) GATA-3 expressions and corresponding quantitations are presented as MFI. n = 6. (H) Total number of CD45⁺ lymphoid cells in the BAL. n = 6. (I) Numbers of BAL eosinophils (CD45⁺, CD11c⁻, SiglecF⁺), neutrophils (PMN, CD45⁺, SiglecF⁻, Ly6G⁺, and CD11b⁺), macrophages (CD45⁺, Ly6G⁻, SiglecF⁻, and CD11c⁺), and other. n = 6. (J) Lung histology. Scale bars, 50 µm. (K) Average alveolar epithelial thickness. n = 5. (L) Schematic representation of the induction of airway inflammation by adoptively transferred Il7r^cre or Il7r^crePiezo1^fl/fl ILC2s in alymphoid recipients. Il7r^cre and Il7r^crePiezo1^fl/fl mice were challenged i.n. on 3 consecutive days with rmIL-33, and pure populations of lung ILC2s were isolated on day 4. 5 × 10⁴ Il7r^cre or Il7r^crePiezo1^fl/fl ILC2s were then subsequently transferred intravenously to Rag^−/−GC^−/− recipient mice, who then received 0.5 µg rmIL-33 i.n. on days 1–3. On day 4, lung function, lung ILC2s, and BAL cellularity were analyzed. The experiment was performed three times. (M) Lung resistance in response to increasing doses of methacholine. n = 6. (N) Total number of ILC2s per lung. n = 6. (O) Total number of BAL eosinophils (CD45⁺, CD11c⁻, SiglecF⁺). n = 6. Data are presented as mean ± SEM. A two-tailed Student’s t test for unpaired data was applied for comparisons between two groups (D–I, N, and O), except for multigroup comparisons where Tukey’s multiple comparison one-way ANOVA tests were used (B, C, K, and M). *P < 0.05, **P < 0.01, ***P < 0.001, and ns: non-significant.

Figure 4.

Yoda1 limits ILC2 function ex vivo and in vivo in a Piezo1-dependent manner. (A) Representative diagram detailing the activation of the Piezo1 channel by shear stress or agonistic activation by Yoda1 leading to an influx of cations, resulting in regulation of transcription. (B) Schematic description of ex vivo ILC2 Piezo1 activation. BALB/c mice were challenged i.n. for 3 consecutive days with rmIL-33, and a pure population of lung ILC2s was isolated on day 4 and cultured with rmIL-2, rmIL-7 with or without Yoda1 (10 µM) for 18 h. (C) Levels of IL-5, IL-6, and IL-13 production in the culture supernatant. n = 6. The experiment was performed three times. (D) Representative plots of live, early apoptotic (E.A.), and late apoptotic/necrotic (L.A.) ILC2s and corresponding quantitation presented as the frequencies (%) of live, E.A., and L.A. ILC2s. n = 3. The experiment was performed twice. (E and F) Representative plots of intranuclear (E) GATA-3 (n = 5) and (F) Ki67 (n = 4) expressions and corresponding quantitation are presented as MFI. The experiments were performed three times. (G) Schematic description of ex vivo ILC2 Piezo1 activation. Il7r^crePiezo1^fl/fl mice were challenged i.n. for 3 consecutive days with rmIL-33, and a pure population of lung ILC2s was isolated on day 4 and cultured with rmIL-2, rmIL-7 with or without Yoda1 (10 µM) for 18 h. (H) Levels of IL-5, IL-6, and IL-13 production in the culture supernatant. n = 3. The experiment was performed twice. (I) Representative plots of intranuclear Ki67 expressions and corresponding quantitation are presented as MFI. n = 3. The experiment was performed twice. (J) Schematic description of in vivo effects of Piezo1 activation. BALB/c mice were challenged i.n. on 3 consecutive days with rmIL-33 and Yoda1 (213 µg/kg/day) or vehicle intraperitoneally. On day 4, lungs were collected and processed to single-cell suspensions for the selected readouts. The experiment was performed twice. (K) Numbers of ILC2s per lung. n = 4. (L) Representative plots of live, early apoptotic (E.A.), and late apoptotic/necrotic (L.A.) ILC2s and corresponding quantitation presented as the frequencies (%) of live, E.A., and L.A. ILC2s. n = 4. (M and N) Representative plots of intranuclear (M) GATA-3 and (N) Ki67 expressions and corresponding quantitation are presented as MFI. n = 4. (O and P) Representative plots of levels of intracellular IL-13 (O) and (P) corresponding quantitation of IL-13–positive ILC2s from the vehicle or Yoda1-treated mice. n = 4. (Q and R) Representative plots of levels of intracellular IL-5 (O) and (R) corresponding quantitation of IL-5–positive ILC2s from the vehicle or Yoda1-treated mice. n = 4. Data are presented as mean ± SEM. A two-tailed Student’s t test for unpaired data was applied for comparisons between two groups (C–F, H, I, K–N, P, and R). *P < 0.05, **P < 0.01, ***P < 0.001, and ns: non-significant.

Figure 5.

Piezo1 activation and a KLF2 inducer inhibit mitochondrial respiration in ILC2s. BALB/c mice were challenged i.n. on 3 consecutive days with rmIL-33, and a pure population of lung ILC2s was isolated on day 4 and cultured with rmIL-2, rmIL-7 without or with Yoda1 (10 µM) for 18 h. Cells were collected, and RNA was extracted to perform a bulk transcriptomic analysis. (A) Volcano plot comparison between DMSO-treated and Yoda1-treated ILC2s. (B) PCA plot. (C) Venn diagram depicting the number of transcripts differentially modulated between DMSO-treated and Yoda1-treated ILC2s differential analysis, with a statistical cutoff of P < 0.05. (D) Chord plot representing the highest DEG from the top five downregulated pathways. Specific pathways are color-coded and represented in the right inner bands, where chords gather. Outer bands on the right depict the IPA −log₁₀ P value. The left inner bands represent the gene −log₁₀ P value. Outer bands on the left represent the gene log₂(fold change). (E) Differential expression of genes between DMSO-treated and Yoda1-treated ILC2s involved in the mitochondrial respirasome, TCA cycle, and OXPHOS pathways. Gray histograms represent −log₁₀ (P value); dotted line represents P < 0.05. Inner bands represent gene log(fold change). (F–I) Mitochondrial respiratory profile showing OCRs in response to oligomycin (ATP synthase inhibitor), BAM15 (mitochondrial uncoupler), and rotenone + antimycin A (complex I and II inhibitors) sequential injections (F) and corresponding key parameters of mitochondrial function (G) basal respiration, (H) spare respiratory capacity, and (I) ATP production rates. n = 3. The experiment was performed twice. (J) Representative diagram detailing the signaling of Piezo1 leading to downstream activation of KLF2/4 resulting in inhibition of the NF-κB pathway and downstream effects on TCA cycle and ILC2 function. (K) Radar plot depicting selected genes from the ILC2 inflammation and the KLF2/4 pathway. Red dots represent log(fold change), and black dots represent −log₁₀ (P value). (L) Protein levels of KLF2 and HSP90 monitored by western blot and quantified with ImageJ. The experiment was performed twice. (M) Representative plots of intranuclear p65 expression and corresponding quantitation presented as MFI. n = 3. The experiment was performed twice. (N) Schematic description of ex vivo ILC2 incubation with SAHA. BALB/c mice were challenged i.n. for 3 consecutive days with rmIL-33, and pure populations of lung ILC2s were isolated and cultured on day 4 with rmIL-2, rmIL-7 with or without SAHA (1 µM) for 18 h. (O) Protein levels of KLF2 and HSP90 monitored by western blot and quantified with ImageJ. The experiment was performed twice. (P–R) Representative plots of intranuclear (P) p65, (Q) GATA-3, and (R) Ki67 expressions and corresponding quantitation presented as MFI. n = 3. These experiments were performed twice. (S–U) Levels of (S) IL-5, (T) IL-6, and (U) IL-13 production in the culture supernatant. n = 6. The experiment was performed twice. (V–Y) Mitochondrial respiratory profile showing OCRs in response to oligomycin (ATP synthase inhibitor), BAM15 (mitochondrial uncoupler), and rotenone + antimycin A (complex I and II inhibitors) sequential injections (V) and corresponding key parameters of mitochondrial function (W) basal respiration, (X) spare respiratory capacity, and (Y) ATP production rates. n = 3. The experiment was performed twice. Data are presented as mean ± SEM. A two-tailed Student’s t test for unpaired data was applied for comparisons between two groups (G–I, M, P–U, and W–Y). *P < 0.05, **P < 0.01, ***P < 0.001. Source data are available for this figure: SourceData F5.

Figure S4.

Effects of Yoda1 and SAHA on ILC2s. (A–C) BALB/c mice were challenged i.n. on 3 consecutive days with rmIL-33, and a pure population of lung ILC2s was isolated on day 4 and cultured with rmIL-2, rmIL-7 without or with (A and B) Yoda1 (10 µM) or (C) increasing doses of SAHA for 18 h. (A) Cells were collected, and RNA was extracted to perform a bulk transcriptomic analysis. Dot plot representation depicting fatty acid degradation and glycolysis gene signatures in Yoda1 or control-treated ILC2s. (B) Representative dot plots of ILC2 intranuclear KLF2 expression and corresponding quantitation presented as KLF2 MFI. n = 3. The experiment was performed twice. (C) Frequencies of live cells in response to increasing doses of SAHA analyzed using the live/dead stain after 18 h of culture. n = 4. The experiment was performed twice. (D) Schematic description of in vivo SAHA administration. Rag2^−/− mice received intraperitoneal injections of 100 mg/kg SAHA or vehicle control and 0.5 µg rmIL-33 or PBS i.n. on days 1–3. On day 4, lung ILC2s were analyzed. The experiment was performed twice. (E and F) Representative plots of intranuclear (E) GATA-3 and (F) Ki67 expressions and corresponding quantitation are presented as MFI. n = 4. Data are presented as mean ± SEM. A two-tailed Student’s t test for unpaired data was applied for comparisons between two groups (B, E, and F). *P < 0.05.

Figure 6.

In vivo Yoda1 treatment reduces AHR and lung inflammation. (A) Schematic description of in vivo effects of Piezo1 activation on the development of AHR. BALB/c mice received intraperitoneal injections of 213 µg/kg/day Yoda1 or vehicle control and 0.5 µg rmIL-33 or PBS i.n. on days 1–3. On day 4, lung function, lung ILC2s, BAL cellularity, and histology were analyzed. (B and C) Lung resistance (B) and dynamic compliance (C) in response to increasing doses of methacholine. n = 5. The experiment was performed twice. (D) Total number of ILC2s per lung. n = 6. The experiment was performed twice. (E) Levels of IL-4, IL-5, IL-6, and IL-13 in the BAL. n = 8. The experiment was performed twice. (F) Total number of CD45⁺ lymphoid cells in the BAL. n = 5. The experiment was performed twice. (G) Numbers of BAL eosinophils (CD45⁺, CD11c⁻, SiglecF⁺), neutrophils (PMN, CD45⁺, SiglecF^-, Ly6G⁺, CD11b⁺), macrophages (CD45⁺, Ly6G⁻, SiglecF⁻, CD11c⁺), T cells (CD45⁺, CD3⁺) and others. n = 5. The experiment was performed twice. (H) Lung histology. Scale bars, 50 µm. The experiment was performed twice. (I) Average alveolar epithelial thickness. n = 5. (J) Schematic description of in vivo effects of Piezo1 activation on the development of AHR. Rag2^−/− mice received intraperitoneal injections of 213 µg/kg/day Yoda1 or vehicle control and 0.5 µg rmIL-33 or PBS i.n. on days 1–3. On day 4, lung function, lung ILC2s, BAL cellularity, and histology were analyzed. (K and L) Lung resistance (K) and dynamic compliance (L) in response to increasing doses of methacholine. n = 5. The experiment was performed three times. (M) Total number of ILC2s per lung. n = 4. The experiment was performed three times. (N) Levels of IL-4, IL-5, IL-6, and IL-13 in the BAL. n = 6. The experiment was performed three times. (O) Total number of CD45⁺ lymphoid cells in the BAL. n = 5. The experiment was performed three times. (P) Numbers of BAL eosinophils (CD45⁺, CD11c⁻, SiglecF⁺), neutrophils (PMN, CD45⁺, SiglecF⁻, Ly6G⁺, CD11b⁺), macrophages (CD45⁺, Ly6G⁻, SiglecF⁻, CD11c⁺), T cells (CD45⁺, CD3⁺), and others. n = 5. The experiment was performed three times. (Q) Lung histology. Scale bars, 50 µm. The experiment was performed three times. (R) Average alveolar epithelial thickness. n = 5. Data are presented as mean ± SEM. A two-tailed Student’s t test for unpaired data was applied for comparisons between two groups (D–G and M–P), except for multi-group comparisons where Tukey’s multiple comparison one-way ANOVA tests were used (B, C, I, K, L, and R). *P < 0.05, **P < 0.01, ***P < 0.001.

Figure S5.

In vivo Yoda1 treatment reduces AHR and lung inflammation after A. alternata challenge. (A) Schematic description of in vivo effects of Piezo1 activation on the development of AHR. Rag2^−/− mice received intraperitoneal injections of 213 µg/kg/day Yoda1 or vehicle control and 100 µg A. Alternata or PBS i.n. on days 1–4. On day 5, lung function, lung ILC2s, BAL cellularity, and histology were analyzed. The experiment was performed twice. (B and C) Lung resistance (B) and dynamic compliance (C) in response to increasing doses of methacholine. n = 5. (D) Total number of ILC2s per lung. n = 5. (E) Levels of IL-4, IL-5, IL-6, and IL-13 in the BAL. n = 8. (F) Total number of CD45⁺ lymphoid cells in the BAL. n = 5. (G) Numbers of BAL eosinophils (CD45⁺, CD11c⁻, SiglecF⁺), neutrophils (PMN, CD45⁺, SiglecF⁻, Ly6G⁺, CD11b⁺), macrophages (CD45⁺, Ly6G⁻, SiglecF⁻, CD11c⁺), T cells (CD45⁺, CD3⁺), and others. n = 5. (H) Lung histology. Scale bars, 50 µm. (I) Average alveolar epithelial thickness. n = 5. (J) Schematic representation of the induction of airway inflammation by adoptively transferred human donor ILC2s in alymphoid recipients. PBMCs were isolated from 500 ml of blood from healthy subjects. After isolation of CD45⁺ Lineage⁻ CRTH2⁺ CD127⁺ hILC2s, cells were cultured (5 × 10⁵/ml) in rhIL-2, rhIL-7, (both 20 ng/ml) and decreasing doses of rhIL-33 (100, 50, 25, and 10 ng/ml) every 72 h or until the required number of cells was achieved. 1 × 10⁵ human donor ILC2s or vehicle were then subsequently transferred intravenously to Rag2^−/−GC^−/− recipient mice, who then received 0.5 µg rmIL-33 i.n. on days 1–3. On day 4, lung function was analyzed. The experiment was performed twice. (K) Lung resistance in response to increasing doses of methacholine. n = 3. (L) Cohorts of Rag2^−/−GC^−/− mice received intraperitoneal injections of 213 µg/kg/day Yoda1 or vehicle control and 0.5 µg hIL-33 i.n. on days 1–3. On day 4, lung function was analyzed. The experiment was performed twice. (M) Lung resistance in response to increasing doses of methacholine. n = 3. Data are presented as mean ± SEM. A two-tailed Student’s t test for unpaired data was applied for comparisons between two groups (D–G), except for multigroup comparisons where Tukey’s multiple comparison one-way ANOVA tests were used (B, C, and I). *P < 0.05, **P < 0.01, ***P < 0.001. ns: non-significant.

Figure 7.

Piezo1 controls human ILC2 function and development of AHR in humanized mice. (A) Schematic description of human blood ILC2 isolation and experimental design. PBMCs were isolated from 500 ml of blood of six healthy subjects. After isolation of CD45⁺ Lineage⁻ CRTH2⁺ CD127⁺ hILC2s, cells were cultured (2 × 10⁴/ml) in rhIL-2, rhIL-7 (both 20 ng/ml) with or without rhIL-33 (100 ng/ml) for 72 h. (B) Representative plots of hILC2 isolation purity. The experiment was performed six times. (C and D) Representative plots of Piezo1 expression from the six subjects (C) and corresponding quantitation presented as MFI (D). Gray histograms represent FMO: full-minus-one staining control. n = 6. (E) Levels of IL-5, IL-6, and IL-13 in the culture supernatants following treatment with 10 µM Yoda1. n = 4. (F) Representative plots of intranuclear GATA-3 expression and corresponding quantitation presented as MFI. n = 4. The experiment was performed three times. (G–J) Mitochondrial respiratory profile showing OCRs in response to oligomycin (ATP synthase inhibitor), BAM15 (mitochondrial uncoupler), and rotenone + antimycin A (complex I and II inhibitors) sequential injections (G) and corresponding key parameters of mitochondrial function (H) basal respiration, (I) spare respiratory, and (J) respiratory-coupled ATP. n = 3. The experiment was performed twice. (K) Schematic representation of the induction of airway inflammation by adoptively transferred human donor ILC2s in alymphoid recipients. PBMCs were isolated from 500 ml of blood from healthy subjects. After isolation of CD45⁺ Lineage⁻ CRTH2⁺ CD127⁺ hILC2s, cells were cultured (5 × 10⁵/ml) in rhIL-2, rhIL-7 (both 20 ng/ml), and decreasing doses of rhIL-33 (100 ng/ml, 50 ng/ml, 25 ng/ml and 10 ng/ml) every 72 h or until the required number of cells was achieved. 1 × 10⁵ human donor ILC2s were then subsequently transferred intravenously to Rag2^−/−GC^−/− recipient mice, who then received 0.5 µg rmIL-33 i.n. on days 1–3. On day 4, lung function, lung ILC2s, and BAL cellularity were analyzed. The experiment was performed twice. (L) Lung resistance in response to increasing doses of methacholine. n = 3. (M) Total number of human ILC2s per lung. n = 5. (N) Total number of BAL eosinophils (CD45⁺, CD11c⁻, SiglecF⁺). n = 5. Data are presented as mean ± SEM and were performed with a total of eight donors. A two-tailed Student’s t test for paired data was applied for comparisons between two groups (D and E). A two-tailed Student’s t test for unpaired data was applied for comparisons between two groups (F and H–J), except for multigroup comparisons where Tukey’s multiple comparison one-way ANOVA tests were used (L–N). *P < 0.05 and ns: non-significant.