Loss of Ccrk leads to hypoplastic lung development with defective branching morphogenesis
Previously, our group reported that CcrkKO mice display developmental defects in multiple organs that resulted from dysregulation of the primary cilia and Hh signaling in mouse embryos (12). Using whole-mount lacZ staining on heterozygote knockout-first conditional Ccrk embryos (Ccrktm1a/+) at E15.5, we observed higher CCRK expression in embryonic trachea and pulmonary bronchioles (Supplementary Fig. 1). To uncover the role of CCRK in the pulmonary system, we phenotypically analyzed lung development in previously generated Ccrk null mutant embryos (CcrkKO) or control (CcrkWT) littermates (12). Deletion of the Ccrk gene causes hypoplastic lung morphology around E14.5 (Fig. 1A). The lungs of CcrkKO mutant embryos were smaller and had fewer lobes than the control had. To find the cause of the underdeveloped lung in CcrkKO mutants, we characterized the progress of lung development in three distinct stages, including E12.5 (embryonic stage), E14.5 (pseudoglandular stage), and E16.5 (canalicular stage), by gross morphological examination and histological analysis (Fig. 1B). CcrkKO mutant homozygous lungs showed normal morphology on E12.5, whereas they exhibited retarded lung growth associated with a reduction in regularly shaped developing alveoli in the pseudoglandular stages. In addition, alveoli number and size in CcrkKO mutant embryos were markedly reduced and separated with a thickened interstitium at E14.5. These abnormal formations of the terminal bronchioles indicated defects in lung-branching morphogenesis in CcrkKO mutants. We tested whether the loss of Bromi might phenocopy the hypoplastic lung development of CcrkKO mutant embryos. Like the CcrkKO mutant respiratory system, Bromi−/− mutant lungs exhibited fewer lobes, with a significant decrease in lung saccular area and increased mesenchymal thickness (Fig. 1C). This delayed respiratory system development in CcrkKO and Bromi−/− mutants suggests that the CCRK and BROMI complex is required for growth and branching morphogenesis during mouse lung development.
Figure 1. CCRK is a crucial regulator for proper lung development. (A) Whole-mount lungs isolated from E12.5, E14.5, and E16.5 embryos. CcrkKO mutants showed fewer branching lobes and appeared smaller than their wild-type littermates. Scale bars = 1 mm. (B) Histological analysis of embryonic lung sections from indicated embryonic stages. Scale bars = 100 μm and 50 μm. (C) Images from whole-mount lungs isolated from E17.5 wild-type and Bromi−/− mutant embryos. (D) H&E staining of histological sections shows less saccule formation and thickening of the lung parenchyma in Bromi−/− mutants than in controls. Scale bars = 100 μm and 50 μm.
Delayed lung-branching morphogenesis in Ccrk mutant embryos is associated with reduced epithelial-cell proliferation and differentiation
During lung-branching morphogenesis, various developmental signaling pathways involved in crosstalk between the lung endoderm and surrounding mesoderm determine the branching pattern. To evaluate the effect of Ccrk removal in lung-branching morphogenesis, we immunostained airway endodermal epithelial cells with antibodies for endodermal progenitor regional markers, such as SOX2 and SOX9 for the proximal and distal regions, respectively. Although the overall SOX2 and SOX9 expression patterns in whole-mount lung endoderm were not changed at E12.5, bud growth and branching morphogenesis was markedly reduced in the CcrkKO mutant lung epithelium than in the wild-type control (Fig. 2A). We classified the three different sequential developmental stages of terminal bifurcation as early “bud”, “flattened”, and later “cleft” stages based on morphology. CcrkKO mutant lungs exhibited relatively increased early bud stages than in the wild-type controls at E14.5 and E16.5 (Fig. 2B and C); however, morphological analysis of later stages of terminal bifurcation of the buds revealed decreased flattened and bifurcated cleft stages of CcrkKO mutants. Therefore, CcrkKO mutant lungs predominantly remain in the early stages of bud morphogenesis rather than advancing to the more-developed flattened and cleft stages, implying that loss of CCRK leads to delayed bud morphogenesis.
The bronchial tree architecture is formed by extensive proliferation of the distal epithelium and surrounding mesenchyme. We examined coordinated progenitor cell proliferation of the lung epithelium and surrounding mesenchyme at E14.5 and E16.5 by immunohistochemistry using the proliferation markers KI67 or EdU (Fig. 2D and E). We observed that KI67- or EdU-labeled proximal or distal epithelial cells in CcrkKO mutants were significantly diminished (Fig. 2D and E). Moreover, the proliferation of mesenchymal cells in CcrkKO mutant lungs was also decreased, suggesting that the effects of loss of Ccrk on epithelial and mesenchymal progenitor-cell proliferation may be the primary cause of lung hypoplasia and delayed bud morphology. To further examine the effects of delayed terminal bifurcation and impaired epithelial cell proliferation in CcrkKO embryos on SOX2 and SOX9 endoderm progenitor differentiation, we analyzed the expression levels of the endodermal and epithelial lineage cell differentiation markers (Supplementary Fig. 2). All differentiated cell marker expression was significantly diminished in mutant embryos. Therefore, CCRK is important for epithelial progenitor cell proliferation and differentiation in mouse lung development.
Figure 2. Ccrk mutant lungs have delayed branching morphogenesis because of decreased epithelial proliferation. (A) Representative whole-mount image of immunofluorescence staining for SOX2 and SOX9 in wild-type and CcrkKO mutant lungs, which were tissue clearing using clarity. Scale bars = 200 μm. (B) Sectioned lung tissue were immunostained with SOX2 and SOX9 antibodies. Nuclei were counterstained with DAPI. Scale bars = 100 μm and 50 μm. (C) Statistical analysis of bud morphology during branch bifurcation. (D) We counted KI67 or EdU immuno-positive proliferating SOX2-, SOX9- or α-SMA-positive cells, and obtained the percentage of proliferating cells. Scale bars = 50 μm. All data are presented as P values from Student’s t-tests: ***P < 0.001.
CCRK is essential for Hh signaling in lung mesenchymal progenitor cells via regulation of primary cilia formation
The SHH signaling pathway positively regulates epithelial-cell proliferation by increasing FGF10 expression in mesenchymal cells during the process of lung branching morphogenesis. SHH expression is restricted to the distal epithelium in developing lungs. Ptch1 and Gli1 are highly expressed in the mesenchyme that surrounds epithelial cells around the branching tips. In addition, bone morphogenic protein 4 (BMP4) is also expressed in the distal epithelium during the pseudoglandular stage and is required for promoting epithelial cell proliferation for branch morphogenesis. To investigate the defect in signaling pathways that regulate epithelial-cell proliferation in CcrkKO mutant lungs, we examined the expression of Shh and Bmp4 by RNA in situ hybridization analysis at E14.5. We found that CcrkKO mutants showed significantly downregulated expression of Shh and Bmp4 in the epithelium. Also, the expression of SHH target genes was also downregulated in CcrkKO mutant lungs. These results revealed a significant impairment of Hh signaling in mesenchymal cells in CcrkKO mutant lungs (Fig. 3A). Moreover, BMP4, which is normally modulated by SHH-signal transduction in the distal epithelium, was downregulated in the CcrkKO mutants, suggesting that SHH signaling is affected by loss of CCRK function. To confirm the compromised SHH signaling in CcrkKO mutants, we examined the gene expression of signaling pathways that regulate branching morphogenesis using quantitative real-time PCR. Consistently, the SHH target genes were downregulated in CcrkKO mutant lungs. Bmp4 was also significantly downregulated in CcrkKO mutant lungs. In contrast, expression of genes, such as Wnt and Fgf10, involved in epithelial and mesenchymal proliferation during branch morphogenesis, were not affected (Fig. 3B). These data suggest that CCRK functions as a positive regulator of Hh signaling during lung development and that the defects of epithelial and mesenchymal cell proliferation observed in CcrkKO mutants may be due, in part, to altered Hh and BMP4 signaling. Although GLI3 exists in a full-length activator form or a processed repressor form, functioning as both positive and negative regulators of the SHH signaling pathway, the repressor function has a dominant role in embryonic lung development. We evaluated whether GLI3 processing is also affected in CcrkKO mutant lungs. We found that GLI3 processing was decreased in CcrkKO mutant lungs, resulting in an increased ratio of full-length GLI3 to GLI3 repressor, thus indicating compromised Hh signaling (Fig. 3C and D).
Figure 3. Loss of Ccrk results in downregulation of the SHH signaling pathway and ciliary defects in the pseudoglandular stage. (A) In situ hybridization lung-tissue sections displayed expression of Shh, Gli1, Ptch1, and Bmp4. Scale bars = 50 μm. (B) Target gene expression for SHH, FGF, and Wnt pathways in the developing lungs of wild-type and CcrkKO mutant lungs. P values are from Student’s t-tests: ***P < 0.001, *P < 0.05; n.s., not significant. (C) Immunoblots of GLI3 in E14.5 whole-lung tissue extracts. β-actin served as a loading control. (D) Quantification of GLI3-FL/Rep ratios. Two samples for each genotype were analyzed. P values are from Student’s t-tests: ***P < 0.001. (E) We analyzed primary cilia in the E14.5 CcrkKO mutant and wild-type lung epithelium and mesenchyme using confocal microscopy by staining with ARL13B and γ-TUB antibodies. Scale bars = 5 μm. (F) We counted the number of primary cilia in each genotype and then divided by the number of total γ-TUB-positive cells. P values are from Student’s t-tests: ***P < 0.005. (G) Distribution of primary cilia in CcrkKO mutants and wild-type lungs was measured. Average lengths and standard deviations for distributions are shown. P values are from Student’s t-tests: ***P < 0.005; n.s., not significant.
Primary cilia are temporally present in progenitor cells and differentiate to multi-cilia, but other cell types maintain the non-motile primary cilia after cell differentiation. To characterize the morphology and frequency of primary cilia in CcrkKO mutant lungs at the pseudoglandular stage, we examined the primary cilia by immunostaining for the ciliary marker ARL13B and the basal body marker γ-tubulin (γ-TUB) (Fig. 3D). Both epithelial and mesenchymal cells in CcrkKO mutant embryonic lungs displayed swelling at their distal tips and shortened ciliary length compared with those in wild-type embryonic lungs. We measured the ciliary frequency of wild-type and CcrkKO mutant embryonic lung epithelial and mesenchymal cells (Fig. 3E). CcrkKO showed reduced ciliary frequency in both the epithelium and the mesenchyme in embryonic lungs and the ciliary lengths in CcrkKO mutant more broadly distributed than control (Fig. 3F). Therefore, CCRK regulates the proper length of primary cilia in epithelium and mesenchymal progenitor cells to regulate Hh signaling; these events may be important for development of branching morphogenesis.
Next, to evaluate the motile multi-ciliogenesis of ciliated epithelial cells under conditions of disrupted primary ciliogenesis in CcrkKO mutants, we examined the generation of multi-cilia with antibodies to cilia markers ARL13B and acetylated α-tubulin (Ac-TUB) as well as the motile cilia marker β-TUB IV with ARL13B to confirm the ciliary type (Supplementary Fig. 3). Similar to primary ciliogenesis defects in progenitor cells, the removal of the Ccrk gene in differentiated ciliated cells impaired motile multi-cilia formation at the canalicular stage. Moreover, we analyzed the expression profiles of key ciliogenic-transcription factors and common ciliary genes during primary cilia (E14.5) and multi-cilia stages (E17.5) in CcrkKO mutant and control embryos (Supplementary Fig. 4). Interestingly, the Ccrk mutant displays a significantly reduced expression of ciliary genes at E17.5 but not in E14.5. These results indicate that CCRK might involve in two ciliogenesis processes with different mechanisms. Collectively, CCRK is required for proper assembly of ciliogenesis in both primary cilia and multiple motile cilia.
Figure 4. SAG treatment of CcrkKO mutant explant lungs rescues Hh signaling and lung branching morphogenesis. (A) Embryonic lungs were dissected at E12.5 and treated for 48 h with or without SAG-enriched media with serum. Scale bars = 100 μm. (B) The q-PCR analysis of Gli1 and Ptch1 in explant lungs treated with SAG or DMSO. (C) Sectioned lung tissue was stained for SOX2 and SOX9 antibodies to show the development of lungs. Scale bars = 50 μm. (D) Statistical analysis compared bud morphology during branch bifurcation. All data are presented as P values from Student’s t-tests: ***P < 0.001; n.s., not significant.
Activation of Hh signaling partially rescues the branching morphogenesis defect in Ccrk mutant explant lungs
To examine whether the decreased Hh signaling pathway in CcrkKO mutants is the primary cause of dysregulation in branching morphogenesis in vitro, we manipulated the Hh signaling pathway in an explant culture system by using small chemical Smoothened agonist SAG. After about 48 h of SAG treatment, wild-type explants did not exhibit significant alterations in bud-branching morphogenesis (Fig. 4A). The CcrkKO mutant lungs cultured in vitro exhibited branching defects similar to those we observed in vivo, displaying a significantly decreased number of branches after 48 h of culture. However, treatment with 500 nM SAG efficiently increased the growth of bud branching and the number of branches in the mutant lungs, yielding results comparable with the control lungs. Consistently, SAG treatment of wild-type lung explants for 48 h increased the expression of the Hh target genes Gli1 and Ptch1 in a dose-dependent manner. Although the responsiveness to SAG in CcrkKO mutants was significantly lessened compared to that of the wild-type controls, it modestly increased Gli1 and Ptch1 expression compared to that of the DMSO control (Fig. 4B). The increased expression levels of Gli1 and Ptch1 in CcrkKO explant lungs following 500 nM SAG treatment was comparable to that in DMSO-treated wild-type explant lungs. Moreover, E12.5 explant lungs from wild-type and mutant embryos were further treated with DMSO or SAG for 48 h and sectioned for analysis of terminal bifurcation via detection of SOX2 and SOX9 (Fig. 4C and D). The 500-nM SAG treatment significantly increased the SOX9-positive nascent bud size and number in the distal epithelium of the CcrkKO mutant explant lungs. Although the relative percentage of the cleft form did not change in 200-nM SAG-treated CcrkKO mutant explant lungs, 500-nM SAG treatment increased the cleft and flattened forms in mutants at the expense of the bud form. These findings demonstrate that the effective increase of Hh response in developing lungs of CcrkKO mutants partially recovers the branching morphogenesis, implying that defects in progenitor-cell proliferation and lung-branching morphogenesis may result from the primary decrease in Hh signaling in CcrkKO mutants.