Pulmonary Epithelial Cell-Derived Cytokine TGF-β1 Is a Critical Cofactor for Enhanced Innate Lymphoid Cell Function

Abstract

Epithelial cells orchestrate pulmonary homeostasis and pathogen defense and play a crucial role in the initiation of allergic immune responses. Maintaining the balance between homeostasis and inappropriate immune activation and associated pathology is particularly complex at mucosal sites that are exposed to billions of potentially antigenic particles daily. We demonstrated that epithelial cell-derived cytokine TGF-β had a central role in the generation of the pulmonary immune response. Mice that specifically lacked epithelial cell-derived TGF-β1 displayed a reduction in type 2 innate lymphoid cells (ILCs), resulting in suppression of interleukin-13 and hallmark features of the allergic response including airway hyperreactivity. ILCs in the airway lumen were primed to respond to TGF-β by expressing the receptor TGF-βRII and ILC chemoactivity was enhanced by TGF-β. These data demonstrate that resident epithelial cells instruct immune cells, highlighting the central role of the local environmental niche in defining the nature and magnitude of immune reactions.

Graphical abstract

Figure 1

Mice Lacking Epithelial-Derived TGF-β Have Reduced Airway Hyperreactivity, Airway Inflammation, and Eosinophilia (A) TGF-β1 expression in the epithelium of Ccsp-creTgfb1^−/− (DOX)- and Tgfb1^+/+ (mock)-treated mice after intranasal house dust mite (HDM) or PBS administration for 3 weeks. Original magnification 20×. (B) Airway hyperreactivity measured by airway resistance to ascending methacholine concentration (baseline; BL). (C and D) Cell counts in the (C) airways and (D) lung tissue. (E) Hematoxylin and eosin-stained lung tissue after HDM administration. (F) Numbers of macrophages (MAC), eosinophils (EOS), and neutrophils (NEU) in the airways. Mann-Whitney ^∗p < 0.05, ^∗∗p < 0.01, and ^∗∗∗p < 0.001. Data shown are from one experiment representative of two independent experiments with a total of n = 10–12 mice per group. Box and whisker plots depict the median and IQR and minimum and maximum values. Line graphs and bar charts are expressed as mean ± SEM. See also Figures S1 and S2.

Figure 2

Epithelial-Derived TGF-β Is Necessary for the Generation of Hallmark Features of Allergic Airways Disease (A and B) Levels of eotaxin-2 in the (A) BAL and (B) lung of Ccsp-creTgfb1^−/− (DOX) and Tgfb1^+/+ (mock) mice after intranasal house dust mite (HDM) or PBS administration. (C and D) IL-5 levels in the (C) BAL and (D) lung. (E) Concentration of IL-4 in lung tissue. (F and G) Serum levels of (F) IgE and (G) HDM-specific IgE. (H) IgA in lung tissue. (I and J) Levels of IL-13 in the (I) lung tissue and (J) BAL fluid. (K) Il13 mRNA levels in lung tissue. Mann-Whitney ^∗p < 0.05 and ^∗∗p < 0.01. Data shown are from one experiment representative of two independent experiments with a total of n = 10–12 mice per group. Box and whisker plots depict the median and IQR and minimum and maximum values. See also Figure S3.

Figure 3

IL-13⁺ ILCs rather than T Effector Cell Subsets Are Preferentially Reduced in Ccsp-creTgfb1^−/− Mice (A and B) Frequencies of Th2 cells in the (A) lung tissue and (B) airways of Ccsp-creTgfb1^−/− (DOX)- and Tgfb1^+/+ (mock)-treated mice after intranasal house dust mite (HDM) or PBS administration. (C–F) Numbers of (C) Th17 cells, (D) Th1 cells, (E) IL-10⁺ Treg cells, and (F) FOXP3⁺ Treg cells in the lung. (G and H) Frequencies of IL-13⁺lineage⁻CD45⁺ICOS⁺ ILCs in the (G) lung and (H) airways. Mann-Whitney ^∗∗p < 0.01. Data shown are from one experiment representative of two independent experiments with a total of n = 10–12 mice per group. Box and whisker plots depict the median and IQR and minimum and maximum values. See also Figures S3–S5.

Figure 4

Epithelial-Derived TGF-β Enhances IL-33-Mediated Inflammation (A) Levels of TGF-β in the BAL of control (Tgfb1^+/+) and Ccsp-creTgfb1^−/− mice treated with a single dose of either rIL-33 or HDM as measured by TGF-β bioassay. (B) TGF-β in the BAL of Ccsp-creTgfb1^−/− (DOX) and Tgfb1^+/+ (mock) mice after 1 week of rIL-33 (or PBS) administration. (C) TGF-β1 expression in the epithelium in DOX- and mock-treated mice. Original magnification 20×. (D and E) Cell counts in the (D) airways and (E) lung tissue. (F) Numbers of macrophages (MAC), eosinophils (EOS), and neutrophils (NEU) in the airways. (G–I) Levels of (G) Eotaxin-2 and (H) IL-5 in the BAL and (I) Il5 mRNA levels in the lung tissue. Mann-Whitney ^∗p < 0.05 and ^∗∗p < 0.01. Data shown are from one experiment representative of two independent experiments with a total of n = 10–12 mice per group. Box and whisker plots depict the median and IQR and minimum and maximum values. Bar charts are expressed as mean ± SEM. See also Figures S5 and S6.

Figure 5

Epithelial-Derived TGF-β Enhances IL-13⁺ ILC Responses during IL-33-Driven Inflammation (A and B) Frequencies of IL-13⁺lineage⁻CD45⁺ICOS⁺ ILCs in the (A) airways and (B) lung tissue in Ccsp-creTgfb1^−/− (DOX) and Tgfb1^+/+ (mock) mice after rIL-33 (or PBS) administration. (C and D) Th2 cells in the (C) lung tissue and (D) airways. (E and F) Levels of IL-13 in the (E) BAL fluid and (F) lung tissue. (G) Il13 mRNA levels in lung tissue. Mann-Whitney ^∗p < 0.05, ^∗∗p < 0.01. Data shown are from one experiment representative of two independent experiments with a total of n = 10–12 mice per group. Box and whisker plots depict the median and IQR and minimum and maximum values.

Figure 6

TGF-β Acts to Enhance ILC2 Activation (A) TGF-βRII expression on ILCs defined as lineage⁻CD45⁺ICOS⁺ in different organs (lymph nodes [LN]), with resident ILC populations in naive control (Tgfb1^+/+ ) mice. (B) FACs plot showing TGF-βRII expression on ILCs. (C) TGF-βRII expression on ILCs from HDM and rIL-33-treated control mice. (D and E) Proliferation defined as Ki67 expression on (D) ILC and (E) ILC2 (IL-13⁺) populations in the lung of Ccsp-creTgfb1^−/− (DOX) and Tgfb1^+/+ (MOCK) mice after rIL-33 (or PBS) administration. (F and G) GATA-3 expression on (F) ILC and (G) ILC2 (IL-13⁺) populations in the lung. (H) mRNA levels of Gata3 in the lung. (I–L) mRNA levels of GATA-3 regulatory proteins (I) Zfpm1, (J) Lef1, (K) Zbtb32, and (L) Sox4 in lung tissue. (M) Sox4 mRNA levels in FACs-sorted CD4⁺ T cells and ILCs. Mann-Whitney ^∗p < 0.05; n = 7–12 mice per group. Box and whisker plots depict the median and IQR and minimum and maximum values. Bar charts are expressed as mean ± SEM.

Figure 7

TGF-β Induced Chemoactivity (A) Plots showing cell tracking of individual migrating cells exposed to chemoattractant gradients over 60 min. Cells were realigned to show the same point of origin. (B–D) Accumulated distance (B), track velocity (C), and directionality (D) of individual cells. ILCs (lineage⁻CD45⁺ICOS⁺) were recovered from the airway lumen of control (Tgfb1^+/+) mice treated with rIL-33 for 1 week. Cells were then exposed to gradients of rIL-33 (20 μg/ml), rTGF-β1 (5 μg/ml), or PBS and cell movement assessed by TAXIScan methodology. Data were pooled from three individual experiments with a minimum total of n ≥ 50 cells tracked per group. Dot plots depict median values. See also Figure S6.