The PCP genes Celsr1 and Vangl2 are required for normal lung branching morphogenesis

Abstract

The lungs are generated by branching morphogenesis as a result of reciprocal signalling interactions between the epithelium and mesenchyme during development. Mutations that disrupt formation of either the correct number or shape of epithelial branches affect lung function. This, in turn, can lead to congenital abnormalities such as cystadenomatoid malformations, pulmonary hypertension or lung hypoplasia. Defects in lung architecture are also associated with adult lung disease, particularly in cases of idiopathic lung fibrosis. Identifying the signalling pathways which drive epithelial tube formation will likely shed light on both congenital and adult lung disease. Here we show that mutations in the planar cell polarity (PCP) genes Celsr1 and Vangl2 lead to disrupted lung development and defects in lung architecture. Lungs from Celsr1(Crsh) and Vangl2(Lp) mouse mutants are small and misshapen with fewer branches, and by late gestation exhibit thickened interstitial mesenchyme and defective saccular formation. We observe a recapitulation of these branching defects following inhibition of Rho kinase, an important downstream effector of the PCP signalling pathway. Moreover, epithelial integrity is disrupted, cytoskeletal remodelling perturbed and mutant endoderm does not branch normally in response to the chemoattractant FGF10. We further show that Celsr1 and Vangl2 proteins are present in restricted spatial domains within lung epithelium. Our data show that the PCP genes Celsr1 and Vangl2 are required for foetal lung development thereby revealing a novel signalling pathway critical for this process that will enhance our understanding of congenital and adult lung diseases and may in future lead to novel therapeutic strategies.

Figure 1.

Disruption of Celsr1 or Vangl2 causes lung morphogenesis defects. Analysis of E18.5 separated mouse embryonic lung lobes (A–C) reveals visibly misshapen lung lobes of Celsr1^Crsh (B) and Vangl2^Lp (C) lungs compared with wild-type (A). H&E staining of sections of E14.5 Celsr1^Crsh (E) and Vangl2^Lp (F) lungs compared with wild-type (D). (P) There was a significant reduction in the number of epithelial branches in both mutants at E14.5: wild-type 44.42, ±2.43, n = 12; Celsr1^Crsh 20.00, ±1.3, n = 12; Vangl2^Lp 18.58, ±1.48, n=12. H&E sections from a minimum of three mutants for each genotype were used for counts. Airway lumina were narrow or absent in many Crsh/Crsh and Lp/Lp airways and epithelium had a multilayered morphology, in contrast to the wider lumina surrounded by single layered epithelium that predominates in wild-type. H&E staining of sections of E18.5 wild-type (G), Celsr1^Crsh (H), Vangl2^Lp (I) lungs. Control lung sections display typical saccular structure and evidence of septation (arrows in G). In contrast, mutant lung sections (H, I) show no evidence of septation. Number (Q) and width (R) of airways is dramatically reduced at E18.5. Number or width of airways was determined by counting airways visible in a complete section of E18.5 lungs from a minimum of two separate embryos (Q) wild-type 90.89, ±3.97; Celsr1^Crsh 52.22, ±3.37; Vangl2^Lp 69.11, ±3.43, n = 9 for each genotype (R) wild-type 12.13, ±0.05, Celsr1^Crsh 4.21, ±0.03, Vangl2^Lp 4.63, ±0.03, n = 10 for each genotype. E14.5 cryosections immunostained for expression of pan-cytokeratin (J–L, inserts show airways at higher magnification), corresponding sections to (J–L) counterstained with DAPI (M–O), dashed lines outline cytokeratin positive cells (as seen in J–L) difficult to distinguish from surrounding cells in Celsr1^Crsh (N) and Vangl2^Lp (O), airways are easily visualized in wild-type (M). Scale bars: (A–C) 62.5 µM; (D–F) 50 µM; (G–O) 12.5 µM, inserts in (D–F); (J–L) 5 µM; *P < 0.05.

Figure 2.

Immunohistochemical analysis of lung epithelial cell differentiation in Celsr1^Crsh and Vangl2^Lp compared with wild-type. Comparison of E18.5 control lung sections (A, D) with Celsr1^Crsh (B, E), and Vangl2^Lp (C, F) showed no apparent alteration in the expression of proximal airway marker, alpha smooth muscle actin (A–C) or distal airway marker, pro-surfactant C (D–F) Graphical comparison of the percentage of Pro-SP-C positive cells in wild-type and mutant lungs (G) confirmed no significant difference between control and Celsr1^Crsh or Vangl2^Lp lungs, n = 8 for each genotype. Counts were performed on sections of left lung lobe obtained from three mutants for each genotype, n = 8 for each genotype. Wild-type 45.12%, ±3.00; Celsr1^Crsh 42.60%, ±3.72; Vangl2^Lp 44.93%, ±1.43. Scale bars: (A–F) 12.5 µM.

Figure 3.

Ex vivo lung cultures reveal that mutations in Celsr1 and Vangl2 phenocopy treatment with Rho kinase inhibitor and Rho kinase activation partially rescues the branching defect in Celsr1^Crsh. E11.5 left lung lobes from wild-type (A, D); Vangl2^Lp (B, E) and Celsr1^Crsh (C, F) were cultured for 48 h. t = 0 mutant lungs (B, C) were indistinguishable from wild-type lungs (A). t = 48h wild-type lungs exhibit increased number of terminal buds (D, G). A smaller increase in terminal buds is observed in Vangl2^Lp (E, G) and Celsr1^Crsh (F, G) explants: wild-type 9.67, ±0.33, n = 14; Crsh 6.43, ±0.48, n = 8; Lp 4.75, ±0.2, n = 6. Mutant lung buds are also expanded (E, F, H) relative to wild-type (D, H): wild-type 67.47, ±4.54, n = 10; Crsh 81.85, ±4.34, n = 16, Lp 96.5, ±10.6, n = 8. Addition of Y27632 severely inhibits branching morphogenesis of a wild-type lung in a dose dependent manner (I, L): lung control 13.14, ±0.51, n = 7; lung 10 µM Y27632 5.60, ±0.51, n = 5; lung 30 µM Y27632 4.33, ±0.42, n = 6. Disruption of the cytoskeleton with 100 ng/ml Cytochalasin D also inhibits lung branching morphogenesis (J, M): wild-type 12.33, ±0.97, n = 27; CytoD 5.28, ±0.32, n = 18. Explants in (I) and (J) have broad branches and appear similar to those from Vangl2^Lp (E) and Celsr1^Crsh (F). Wild-type lung explants treated with 20 ng/ml CNF-1 show a small increase in number of epithelial buds formed (N): wild-type (untreated) 7, ±0.94, n = 6; wild-type with CNF-1 8.33, ±1.13, n = 9, two-tailed P-value = 0.4819. This result was consistent though not statistically significant. Whereas in a separate set of experiments, treatment of Crsh homozygous lung explants with 20 ng/ml CNF-1 results in a greater, statistically significant increase in bud numbers (K, N): Crsh (untreated) 6.25, ±0.52, n = 8; Crsh with CNF-1 8.18, ±0.63, n = 11. Explants were stage matched at t = 0 to allow direct comparison of the number of end buds between control and treated or wild-type and mutant lungs within each experiment. Scale bars: (A–F, I–K) 63 µM. *P < 0.05.

Figure 4.

Celsr1 and Vangl2 are expressed in all modes of branching. Confocal images of wild-type E14.5 (A, B, G, H) and Celsr1^Crsh (C, D) or E12.5 wild-type (E, F) whole lungs immunostained with anti-cytokeratin (A, C, D, G), anti-Celsr1 (B, E, F) or anti-Vangl2 (H) antibodies. Pictures are slices showing a single layer from a z-stack. In wild-type lungs Celsr1 and Vangl2 are expressed in all three branching modes as described by Metzger et al. (34); examples of planar branching (yellow lines in A, C, D, G); domain branching (red lines in A, D, G) and orthogonal branching (white lines in A, C, G) are highlighted. All three modes are present in Celsr1^Crsh and Vangl2^Lp lungs (C, D and data not shown) indicating the defects in Celsr1 and Vangl2 are not restricted to a specific mode of branching. Scale bars: (A, B) 125 µM plus ×1.2 zoom, (C) 125 µM plus ×1.2 zoom, (D) 125 µM plus ×1.7 zoom. (E, F) 125 µM plus ×6.3 zoom, (G, H) 125 µM plus ×1.7 zoom.

Figure 5.

Mutant lungs display aberrant cell architecture, yet apical–basal polarity remains intact. Cryosections of E14.5 wild-type (A, D, G, J, M, P) Celsr1^Crsh (B, E, H, K, N, Q) and Vangl2^Lp (C, F, I, L, O, R). Rhodamine phalloidin staining of F-actin reveals cytoskeletal defects in mutants at E14.5 (B, C) and E18.5 (E, F) compared with wild-type (A, D), as does staining with antibodies to non-muscle myosin IIA in mutants (H, I), compared with wild-type (G). Anti-β-catenin staining appears normal in Celsr1^Crsh (K) and Vangl2^Lp (L), compared with wild-type (J). Normal apical–basal polarity is revealed in lung by double-labelling of wild-type (M), Celsr1^Crsh (N) and Vangl2^Lp (O) airways with ZO-2 (red) and laminin (green in M–O and red in P–R) and by double-labelling of wild-type (P) Celsr1^Crsh (Q) and Vangl2^Lp (R) with aPKCζ (green). Scale bars: (A–F) 5 µM, (G–L) 5 µM, (G–L) 25 µM, (M–R) 12.5 µM plus 3× zoom.

Figure 6.

Mutant lung endoderm responds to FGF10 but is unable to undergo branching. (A, B) Wild-type lung endoderm branches in response to 400 ng/ml FGF10 (mean bud number = 11, n = 6). (C, D) mutant lungs do not branch in response to FGF10 (mean bud number Crsh = 0.9, n = 6). P-ERK1/2 staining of E14.5 lung, shows no significant difference in percentage of P-ERK1/2 positive cells in the epithelium of mutants compared with wild-type controls (E): wild-type 55.17%, ±3, n = 10; Crsh 51.97%, ±6.3, n = 7, P = 0.6074; Lp 56.45%, ±3.9, n = 9, P = 0.7939. Data from three embryos for each genotype. Western blotting reveals no change in the relative levels of P-ERK1/2 in wild-type and Lp/Lp or Crsh/Crsh littermates (F). Scale bars: (A–D) 5 µM.

Figure 7.

Celsr1 and Vangl2 proteins are spatially restricted and differentially expressed in lung epithelia. (A, B) E11.5 transverse cryosections of lungs immunostained with antibodies against Celsr1 (A) and Vangl2 (B). (C–F) Immunostaining of transverse cryosections of E14.5 wild-type (C, E) with anti-Celsr1 (C) and anti-Vangl2 (E) antibodies. Corresponding protein levels are dramatically decreased in lung tissue of mutants; anti-Celsr1 in Celsr1^Crsh (D) and anti-Vangl2 in Vangl2^Lp (F) compared with wild-type (C, E). Immunostaining of Celsr1 protein in Vangl2^Lp (G) reveals no change in spatial localization. Enrichment of Vangl2 protein adjacent to the lumen is lost in Celsr1^Crsh (H) mutant lung sections. Scale bars: (A, B) 125 µM plus ×1.9 zoom, (C) 125 µM plus ×2.5 zoom, (D) 125 µM plus ×1 zoom, (E, H) 125 µM plus ×2 zoom, (F, G) 125 µM plus ×3 zoom.

Figure 8.

Differential expression of Celsr1 and Vangl2 is observed in branching lung endoderm explants and morpholino knockdown highlights a role for Celsr1 in bifurcation. Double-labelling of wild-type E14.5 cryosections with Celsr1 (A, C) and laminin (B, C) antibodies. E11.5 lung endoderm explants cultured for 48 h in 400 ng/ml FGF10 and double labelled with phalloidin (H–K) and Celsr1 (D, E) or Vangl2 (F, G) antibodies. High levels of Celsr1 expression are present in regions of restricted tissue growth such as points of bifurcation (D, E, H, I). Vangl2 is most highly expressed at the luminal surface of outgrowing buds (F, G, J, K). E11.5 lung explants from β-actin promoter driven GFP embryos were cultured for 48 h in the presence of control (L–N) or Celsr1 (O–Q) morpholinos and subsequently imaged over a 24 h period. Images show three timepoints from this series. Scale bars: (A–C) 125 µM ×2 zoom (D, H, F, J) 125 µM ×2.7 zoom, (E, I) 125 µM ×10 zoom, (G, K) 125 µM ×8 zoom, (L–Q) 50 µM.