Hippo signaling promotes lung epithelial lineage commitment by curbing Fgf10 and β-catenin signaling

Abstract

Organ growth and tissue homeostasis rely on the proliferation and differentiation of progenitor cell populations. In the developing lung, localized Fgf10 expression maintains distal Sox9-expressing epithelial progenitors and promotes basal cell differentiation in the cartilaginous airways. Mesenchymal Fgf10 expression is induced by Wnt signaling but inhibited by Shh signaling, and epithelial Fgf10 signaling activates β-catenin signaling. The Hippo pathway is a well-conserved signaling cascade that regulates organ size and stem/progenitor cell behavior. Here, we show that Hippo signaling promotes lineage commitment of lung epithelial progenitors by curbing Fgf10 and β-catenin signaling. Our findings show that both inactivation of the Hippo pathway (nuclear Yap) or ablation of Yap result in increased β-catenin and Fgf10 signaling, suggesting a cytoplasmic role for Yap in epithelial lineage commitment. We further demonstrate redundant and non-redundant functions for the two nuclear effectors of the Hippo pathway, Yap and Taz, during lung development.

Keywords: Differentiation; Fgf10; Fgfr2; Hippo; Ilk; Integrin; Lung; Progenitor; Yap; β-Catenin.

Fig. 1.

Epithelial Ilk or Yap inactivation during early lung development activates Wnt7b-Fgf10 crosstalk and expands the distal Sox9^pos epithelial progenitor population. (A) Immunostaining for Sox2 and Sox9, and Yap on E14.5 control Shh-Cre and Shh-Cre;Ilk^f/f lungs. (B) In situ hybridization on E13.5 control Shh-Cre and Shh-Cre;Ilk^f/f lungs. (C) qPCR analysis of E13.5 control Shh-Cre (n=12) and Shh-Cre;Ilk^f/f (ILK KO) lungs (n=14). (D) qPCR of E13.5 control (n=12) and Shh-Cre;Yap^f/f lungs (n=10). (E) Immunostaining for β-catenin on E14.5 control Shh-Cre and Shh-Cre;Yap^f/f lungs. (F) Quantification of Sox9 and Sox2 expression domains on images represented in A and G (n=5 for each genotype). (G) Immunostaining for Sox9 and Sox9 on control Shh-Cre, Shh-Cre;Yap^f/f and Shh-Cre;Yap^f/+ lungs. Data are mean±s.e.m. **P<0.01; *P<0.05, as determined using a two-tailed t-test; n represents the number of biological replicates. Scale bars: 20 µm in E; 100 µm in A,G.

Fig. 2.

Epithelial Ilk activates the Hippo pathway during late lung development. (A-N) Immunostaining and quantification of pixel intensity for Ilk (A-C), merlin (D-F), phospho-Mst1/2 (G-I), Taz (J-L) and pYap (M-O) in E18.5 control (A,D,G,J,M) versus Nkx2.1-Cre;Ilk^f/f (ILK KO) (B,E,H,K,N) airways. (C,F,I,O) Maximum intensity levels. (L) Total intensity per area normalized to control. Data are mean±s.e.m. **P<0.01; *P<0.05, as determined by a two-tailed t-test; n=6 biological replicates for each experimental group. Scale bar: 100 µm.

Fig. 3.

Inactivation of one copy of Yap and Taz simultaneously with both copies of Ilk partially rescues the Nkx2.1-Cre;Ilk^f/f phenotype. (A-T) Immunostaining and quantification for Sftpc and RAGE (A-D), Scgb1a1 and CGRP (E-H), Scgb1a1 and β-tubulin (I-K), K5 and p63 (L-N), and Ilk and E-cadherin (O-T) in E18.5 control, Nkx2.1-Cre;Ilk^f/f and Nkx2.1-Cre;Ilk^f/f;Yap^f/+;Taz^f/+ lungs. n=6 for each experimental group. (U) qPCR analysis of E18.5 control (n=7), Nkx2.1-Cre;Ilk^f/f (ILK KO) (n=6) and Nkx2.1-Cre;Ilk^f/f;Yap^f/+;Taz^f/+ (ILK YAP TAZ rescue) (n=6) lungs. (D) Maximum intensity levels. Data are mean±s.e.m. **P<0.01; *P<0.05, as determined by a two-tailed t-test; n represents the number of biological replicates. Scale bars: 100 µm.

Fig. 4.

Fgf10 overexpression during late lung development phenocopies the Nkx2.1;Ilk^f/f late lung phenotype, which itself can be rescued by blocking Fgf10 signaling. (A,B) Co-immunostaining for Fgf10^LacZ (β-gal) and ADRP (adipose differentiation-related protein), a lipofibroblast marker, on E18.5 control Fgf10^LacZ lungs (A) and Nkx2.1-Cre;Ilk^f/f;Fgf10^LacZ (B) lungs. (C) Quantification of the number of Fgf10⁺ cells per image in A,B. Ctrl, n=6; Ilk KO, n=5. **P<0.01. (D-X) Immunostaining for Sox9; Sftpc (AT2 cells and distal tip epithelial progenitor marker) and α-SMA (smooth muscle); Rage (AT1 cells); K5 and p63 (basal cells); or Scgb1a1 (club cells) and CGRP (neuroendocrine cells) on E18.5 control (D,H,L,P,T), Nkx2.1-Cre;Ilk^f/f (E,I,M,Q,U), Rosa26-rtTa;Tet-Fgf10 lungs induced from E15.5 (F,J,N,R,V) and rescued Nkx2.1-Cre;Ilk^f/f;Rosa26-rtTa;Tet-sFgfr2b (G,K,O,S,W) lungs induced from E15.5. In A,B,D-O, the boxed areas are shown at higher magnification in the insets. (X) qPCR analysis of E18.5 control, Nkx2.1-Cre;Ilk^f/f (ILK KO), Nkx2.1-Cre;Ilk^f/f;Rosa26-rtTa;Tet-sFgfr2b (ILK rescue) and Rosa26-rtTa;Tet-Fgf10 (Tet-Fgf10) lungs. n=6 for each experimental group. (Y) Quantification of number of K5^posP63^pos basal cells in P-S. n=6 for each experimental group. (Z) Quantification of normalized maximum Sox9 and Rage intensity in D-G and L-O, respectively, and total Sftpc intensity per area in H-K. n=6 for each experimental group. **P<0.01; *P<0.05, as determined by a two-tailed t-test; n represents the number of biological replicates. Data are mean±s.e.m. Scale bars: 100 µm.

Fig. 5.

Inactivating β-catenin rescues the cystic Nkx2.1-Cre;Yap^f/f phenotype. (A-P) Immunostaining on E18.5 control, Nkx2.1-Cre;Yap^f/f and Nkx2.1-Cre;Yap^f/f;βcat^f/f lungs for Sftpc and RAGE (A-F), Taz (G-J), Yap and β-catenin (K-M), and K5 and p63 (N-P). (Q) qPCR analysis of E18.5 control, Nkx2.1-Cre;Yap^f/f (YAP KO) and Nkx2.1-Cre;Yap^f/f;βcat^f/f (YAP rescue) lungs. n=6 biological replicates for each experimental group. (J) Maximum intensity levels for Taz (n=6 biological replicates for each experimental group). Data are mean±s.e.m. **P<0.01; *P<0.05, as determined by a two-tailed t-test. Scale bars: 500 µm in A-C; 100 µm in D-I; 50 µm in K-P.

Fig. 6.

Taz promotes AT1 cell differentiation. (A-O) Immunostaining on P21 control, Nkx2.1-Cre;Taz^f/f and Nkx2.1-Cre;Taz^f/f;Yap^f/+ lungs for Sftpc and RAGE (A-C), Sox9 (D-F), Acta2 (G-L), or Taz, Spc and Hopx (M-O). In A-F,M-O, the boxed areas are shown at higher magnification in the insets. J-K show the boxed areas in G-I, respectively. (P) qPCR analysis of P21 control, Nkx2.1-Cre;Taz^f/f (Taz KO) and Nkx2.1-Cre;Taz^f/f;Yap^f/+ (Taz KO Yap het) lungs. Data are mean±s.e.m. **P<0.01; *P<0.05, as determined by a two-tailed t-test; n=6 biological replicates for each experimental group. Scale bars: 100 µm in A-F,J-O; 500 µm in G-I.

Fig. 7.

Hippo promotes lung epithelial lineage commitment by inhibiting Fgf10 and nuclear β-catenin. During the branching phase of lung development, which lasts until ∼E16, distal Sox9-positive epithelial progenitors give rise to Sox2-positive bronchial epithelial cells. Subsequently, lungs undergo an alveolar differentiation program during which Sox9-positive distal epithelial progenitors give rise to alveolar epithelial cells. Nuclear Yap (n-Yap) in the distal tip epithelium allows the nuclear localization of β-catenin, which induces Sox9. In addition, nuclear Yap, through β-catenin, drives Wnt7b expression, which acts on the distal mesenchyme to promote the release of Fgf10. Fgf10 in turn drives nuclear β-catenin in the distal tip epithelium, both of which induce Sox9. Cytoplasmic Yap (c-Yap) in the proximal epithelium promotes the degradation of β-catenin and blocks Fgf10 expression in the adjacent mesenchyme, thereby promoting differentiation of the distal tip epithelial progenitors.