Non-cell-autonomous activation of hedgehog signaling contributes to disease progression in a mouse model of renal cystic ciliopathy

Introduction

The primary cilium is a microtubule-based cellular organelle that protrudes from the cell surface into the extracellular matrix or fluid, and functions as an antenna to sense and transduce environmental signals to regulate cellular responses (1–3). Mutations in ciliary genes, encoding proteins required for cilia formation or function, contribute to a wide spectrum of human disorders, including polydactyly, holoprosencephaly, obesity and polycystic kidney disease (PKD), among many others (4–6). Autosomal dominant polycystic kidney disease (ADPKD), the most frequent form of PKD and one of the most common monogenetic diseases in humans, can be attributed to mutations in PKD1 and PKD2, which encode polycystin 1 and 2 (PC1 and PC2), respectively (reviewed in (7)). PC1 and PC2 are trafficked to cilia and their ciliary localization is integral to their function (8–12). A key feature that defines PKD is the formation of epithelial cysts in the kidney. The significance of epithelial cells in PKD pathogenesis is highlighted by the fact that inactivation of ADPKD or cilia biogenesis genes specifically in renal epithelial cells is sufficient to cause PKD in mouse models (13–15). It is postulated that PC1 and PC2 in renal epithelial cells function to inhibit a cilia-dependent and cyst-activating pathway (CDCA) and that unconstrained CDCA leads to cyst formation in Pkd1 or Pkd2 mutants (16). In addition to epithelial cysts, interstitial fibrosis has also been observed in PKD (17–19). Currently, the precise molecular mechanisms of cyst formation and interstitial fibrosis remain poorly understood.

One of the best studied cilia-regulated pathways is the hedgehog (HH) pathway. Three secreted ligands, sonic hedgehog (SHH), indian hedgehog (IHH) and desert hedgehog (DHH), initiate HH signaling by binding to the receptor patched 1/2 (PTCH1/2). In vertebrates, patched is localized to cilia in the absence of its ligands. Binding of ligands activates the HH pathway by displacing ciliary PTCH and allows the G-protein coupled receptor-like protein smoothened (SMO) to enter the cilium. SMO accumulation facilitates the processing of GLI proteins, which are then trafficked to the nucleus and transcribe HH target genes, including Gli1, Ptch1 and Ptch2 (20–22). A functional cilium is therefore required for the activation of the HH pathway. The HH signaling pathway plays a fundamental role in tissue patterning, cell growth and differentiation during development (23–25). In agreement with a central role of cilia in this pathway, abnormal HH signaling underlies polydactyly and craniofacial ciliopathies (26,27). Further, the HH pathway is frequently reactivated during injury and repair in multiple organs, including the lungs and kidneys, and prolonged HH activation contributes to tissue fibrosis (28–32).

Abnormal activation of the HH pathway has been detected in cystic kidney samples in both human and animal models, including Thm1 mouse mutants (33–36). Thm1/Ift139 encodes a component of the intraflagellar transport A complex that negatively regulates HH signaling (37). Defective Thm1 leads to shorter cilia with bulbous tips, upregulation of HH signaling and cystic kidneys (33,37). In concordance with Thm1’s inhibitory role in HH signaling, repressing the HH pathway reduced cyst progression in the Thm1 mutant kidney (33). Moreover, cyst expansion induced by cAMP and Pkd1 inactivation in cultured mouse kidney explants was also sensitive to HH inhibition (33). However, a recent study demonstrated that modulating the HH pathway in renal epithelial cells failed to affect PKD progression in mouse Pkd1 mutants (38), suggesting that the role of HH signaling in different PKD models is complex and not fully understood.

In a previous study, we found that the HH pathway is upregulated in the cystic kidney of Arl13b^f/f;Ksp-Cre mice, where Arl13b is specifically deleted in epithelial cells of the distal nephron (35). Since Arl13b is required for cilia biogenesis, this result seems contradictory to the essential role of cilia in HH signaling. However, HH ligands commonly activate HH signaling non-cell-autonomously during tissue patterning (24,25). Similarly, following acute kidney injury, SHH is produced by epithelial cells, but downstream targets are upregulated in interstitial cells (28,30,39). The role of non-cell-autonomous HH signaling has not been investigated in PKD pathogenesis and is the focus of this study.

We demonstrate that HH signaling is predominantly activated in the interstitium when Arl13b is deleted in renal epithelial cells. In co-culture, Arl13b^−/− epithelial cells produced more SHH and led to activation of the HH pathway in mesenchymal cells. Moreover, treatment of GANT61, a compound that blocks GLI-mediated transcription (40), partially suppressed both renal cyst expansion and fibrosis, and preserved kidney function in Arl13b^f/f;Ksp-Cre mice. Finally, GANT61 treatment reduced the abnormally increased number of proliferating cells in the cystic kidney of Arl13b^f/f;Ksp-Cre mutant mice. Together, our results clarify the role of renal epithelial and interstitial cells in the abnormal activation of HH signaling in cystic kidneys and highlight the significance of epithelial–interstitial cross-talk in this process. Our results also suggest that non-cell-autonomous HH signaling contributes to both renal cyst progression and fibrosis in Arl13b mutant mice and may provide important insight into the molecular etiology of renal cystic diseases and fibrosis in human.

Results

HH signaling is activated non-cell-autonomously in the interstitium of the Arl13b ^f/f;Ksp-Cre kidney

We previously showed that the expressions of HH ligands and target genes are upregulated in the cystic kidney of Arl13b^f/f;Ksp-Cre mice in whole kidney lysates (35). To ascertain tissue specificity of HH activation in the mutant kidney, we took advantage of the Gli1^LacZ/+ reporter line that expresses nuclear localized LacZ (nLacZ) in cells with activated HH signaling (28,41). We generated Arl13b^f/f;Ksp-Cre;Gli1^LacZ/+ mice and performed immunofluorescence analysis for LacZ in sections of postnatal Day 21 (P21) kidneys. In kidneys with functional ARL13B (Arl13b^f/+;Ksp-Cre;Gli1^LacZ/+), we detected LacZ positive nuclei sporadically in both the cortex and medullar region, outside of epithelial tubules, as outlined by anti-laminin signal (Fig. 1A). This expression pattern is consistent with the known distribution pattern of GLI1⁺ cells in the normal kidney (28). In the mutant kidney (Arl13b^f/f;Ksp-Cre;Gli1^LacZ/+), the number of LacZ positive nuclei increased; notably, nLacZ positive cells were still excluded from regions encircled by laminin, suggesting that stromal cells were the HH responding cells (Fig. 1A and Supplementary Material, Fig. S1A, C). Since the morphology of tubules, epithelial cells and the pattern of laminin signal were significantly distorted in the mutant kidney, we validated the identity of nLacZ positive cells, by co-labelling with anti-LacZ and anti-alpha Smooth Muscle Actin (α-SMA), a marker of activated myofibroblasts. In both the cortex and medullar region of the mutant kidney, the signal of α-SMA increased dramatically, as expected (35) (Fig. 1B). Moreover, most nLacZ positive cells were also positive for α-SMA, suggesting that they were myofibroblasts (Fig. 1B and Supplementary Material, Fig. S1B, C). We then investigated whether abnormal increase of nLacZ positive cells occurs at earlier time points. At P14, in comparison to P21, there were more nLacZ positive cells outside of epithelial regions encircled by laminin in control kidneys in both the cortex and medulla; and the increase of these cells in mutant kidneys was more modest and frequently in clusters (Fig. 1C and Supplementary Material, Fig. S1D, F). The increase of α-SMA signal and the percentage of α-SMA positive:nLacZ positive cells in mutant kidneys was more evident (Fig. 1D and Supplementary Material, Fig. S1E, F).

Figure 1

Epithelial-specific inactivation of Arl13b results in non-cell-autonomous activation of HH signaling in stromal cells. (A–D) Immunofluorescence staining of kidney sections of Arl13b^f/f;Ksp-Cre;Gli1^LacZ/+ and Arl13b^f/+;Ksp-Cre;Gli1^LacZ/+ mice at P21 (A, B) and P14 (C, D). Cells with active HH signaling are indicated by nLacZ signal (green, detected by anti-LacZ). Arrows show representative nLacZ positive nuclei. DAPI = nuclear counterstain (blue). Scale bars: 20 μm. In (A) and (C), anti-laminin (Laminin, red) outlines basal border of epithelial tubules; yellow ‘T’ indicates representative tubules. In (B) and (D), anti-α-SMA (red) labels activated myofibroblasts. (E) Cilia in mIMCD3 and 10T1/2 cells detected by anti-acetylated tubulin (green) and anti-ARL13B (red). Nuclei stained with DAPI (blue). Scale bar: 10 μm. (F, G) RT-qPCR analysis of the expression level of HH target genes in 10T1/2 (F) and mIMCD3 (G) cells in response to SAG treatment. Unit 1 is defined by the expression level in vehicle-treated controls. Results represent mean ± SD of indicated number of independent experiments (N). Ctrl: Arl13b^f/+;Ksp-Cre mice; KO: Arl13b^f/f;Ksp-Cre mice; ^*P < 0.05; ^*^*P < 0.01; ^*^*^*P < 0.001; ^*^*^*^*P < 0.0001.

Combined, these results suggest that inactivation of Arl13b in renal epithelial cells activates HH signaling in stromal cells via a non-cell-autonomous mechanism.

Mesenchymal 10T1/2 and epithelial mIMCD3 cells show differential responsiveness to HH stimulation

Both secreted factors and mechanical stimuli could trigger non-cell-autonomous responses in the mutant kidney in vivo. To investigate whether mechanisms independent of cyst expansion could contribute to the observed activation of HH signaling in renal interstitial cells, we cultured mouse inner medullary collecting duct (mIMCD3) cells and the murine mesenchymal cell line, 10T1/2, in vitro. 10T1/2 is a HH responsive cell line that has been used to model GLI1⁺ progenitor cells, a main contributor of activated myofibroblasts after kidney injury (28,42). We first asked whether the two cell lines displayed different sensitivity to HH stimulation. After reaching confluency, both cell lines were switched to a low serum medium (0.5% FBS) to induce the formation of primary cilia. As expected, abundant cilia were detected in mIMCD3 cells (Fig. 1E). In addition, 10T1/2 cells also displayed cilia, shown by co-staining with cilia markers anti-acetylated tubulin and anti-ARL13B (Fig. 1E) or anti-ARL13B alone (Supplementary Material, Fig. S2A). Ciliated cells were subsequently treated with SAG, a small molecule agonist of SMO, for 24 h. The expression level of HH target genes, including Gli1, Ptch1 and Ptch2, was then analyzed by reverse transcription and quantitative PCR (RT-qPCR) to monitor the status of HH signaling. Even at a low SAG concentration (10 nM), the expression level of HH target genes Gli1 (>143-fold), Ptch1 (~7-fold) and Ptch2 (12-fold) was significantly increased in mesenchymal 10T1/2 cells compared to vehicle-treated controls (Fig. 1F). In contrast, and despite that mIMCD3 epithelial cells displayed more prominent cilia (Fig. 1E), mIMCD3 cells showed minimal response to SAG, with a slightly increased expression of Gli1 (1.5-fold), Ptch1 (1.5-fold) and Ptch 2 (1.3-fold) when treated with 10 nM SAG. Increase of SAG concentration (up to 500 nM) failed to elicit more robust responses (Fig. 1G). Combined, these results suggest that although cilia are displayed on a wide range of cell types, the responsiveness to HH activation differs significantly between different cell types.

Arl13b inactivation leads to increased SHH production in cultured cells

Since epithelial cells are the source of HH ligands in injury responses in the kidney (28,30,31,39), we investigated whether inactivation of Arl13b alone is sufficient to induce the expression of HH ligands in epithelial cells. We used an Arl13b^−/− mIMCD3 cell line generated via CRISPR-Cas9 in a previous study (43). To avoid non-specific effect of CRISPR-mediated DNA damage, we used a rescued cell line that re-expresses Arl13b in the Arl13b^−/− mIMCD3 cells as a control. As reported before, cilia biogenesis was disrupted in Arl13b^−/− cells, but longer cilia formed in rescued cells (Supplementary Material, Fig. S2B) (43). It is known that overexpression of Arl13b leads to cilia elongation (44). However, transgenic mice overexpressing Arl13b, even though they display longer cilia at the embryonic stage, show no defect in HH-dependent neural patterning, and are viable and fertile (45).

To dissect the potential role of epithelial–mesenchymal cross-talk free from cyst expansion, we used a co-culture system (Fig. 2A). Briefly, epithelial mIMCD3 cells were seeded onto an inverted transwell and incubated for 0.5 hour to allow for cell attachment to the bottom side of the insert. The insert was then inverted and placed into a fresh well and 10T1/2 cells were seeded into the upper chamber. At confluency, cells were switched to a low serum medium and incubated for an additional 2 days. Both sides of the insert were then scraped to collect mIMCD3 and 10T1/2 cells and RT-qPCR was performed. While the level of Shh mRNA in Arl13b^−/− mIMCD3 cells was significantly increased (19-fold) compared to rescued mIMCD3 cells, the level of Dhh and Ihh mRNA remained comparable between the two mIMCD3 cell lines (Fig. 2B). We also detected increased Shh expression in Arl13b^−/− mIMCD3 cells cultured in the absence of 10T1/2 cells in comparison to rescued cells (Supplementary Material, Fig. S2C). However, the level of Shh expression in mIMCD3 cells cultured alone was very low, with cycle threshold values between 37 and 39, and the increase in mutant cells when cultured alone was modest, which may have contributed to the high variation detected by RT-qPCR (Supplementary Material, Fig. S2C). It is possible that modest differences between Arl13b^−/− and rescued mIMCD3 cells were amplified through epithelial–mesenchymal cross-talk in the co-culture system. However, the precise mechanism for the more robust increase of Shh expression in co-cultured Arl13b^−/− mIMCD3 cells remains undefined.

Figure 2

Co-culturing with Arl13b^−/− epithelial cells (KO) leads to activation of HH signaling in mesenchymal cells. (A) Schematic diagram of the co-culture system. (B) Expression levels of HH ligands, Dhh, Ihh and Shh, in KO and rescued mIMCD3 cells co-cultured with 10T1/2 cells, assayed by RT-qPCR. Unit 1 is defined by the expression level in rescued cells. (C) Concentration of the N-terminal fragment of SHH (SHH) in supernatants of co-cultured mIMCD3 and 10T1/2 cells on Days 2, 5 and 10 of incubation in low serum medium. (D) Expression levels of HH targets Gli1, Ptch1 and Ptch2 in 10T1/2 cells co-cultured with KO and rescued mIMCD3 cells, assayed by RT-qPCR. Unit 1 is defined by the expression level in 10T1/2 cells co-cultured with rescued mIMCD3 cells. (E) Western blot showing protein levels of VIM and α-SMA in 10T1/2 cells after 2, 5 and 10 days of co-culturing with KO and rescued mIMCD3 cells. GAPDH is used as a loading control. Adjusted density (Adj. density) is defined as relative density of the band of interest normalized by that of GAPDH. In all RT-qPCR, Gapdh was used for normalization. All quantitative results represent mean ± SD of the indicated number of independent experiments (N). ns, not significant (P > 0.05); ^*P < 0.05; ^*^*P < 0.01; ^*^*^*P < 0.001; ^*^*^*^*P < 0.0001.

We further determined the level of the active N-terminal fragment of SHH in supernatants collected from both the top and bottom culture chambers using a SHH N-terminus Quantikine enzyme-linked immuno-sorbent assay (ELISA) assay. Although the source of N-SHH in the bottom or top chamber could not be exclusively attributed to cells cultured in corresponding chambers, since the level was elevated in both chambers in Arl13b^−/− co-cultures, the total amount of secreted N-SHH increased significantly in Arl13b^−/− co-cultures compared to the Arl13b rescued co-cultures from Day 2 and remained significantly elevated at Days 5 and 10 under low serum condition (Fig. 2C).

The HH pathway is activated in mesenchymal cells co-cultured with Arl13b^−/− mIMCD3 cells

We used the co-culture system to investigate the impact of Arl13b^−/− mutant epithelial cells on co-cultured mesenchymal cells. 10T1/2 cells co-cultured with Arl13b^−/− mIMCD3 cells for 5 days showed significantly higher expression levels of HH target genes Gli1, Ptch1 and Ptch2 compared to those cultured with rescued mIMCD3 cells, revealing an elevated level of HH signaling in these cells (Fig. 2D).

We then tested whether co-culturing with mutant epithelial cells was sufficient to trigger a profibrotic response in mesenchymal cells by western blot analysis of the early myofibroblast activation marker vimentin (VIM) and the myofibroblast marker α-SMA (46,47). Co-culturing with Arl13b^−/− mIMCD3 cells led to an increased level of VIM in 10T1/2 cells from Day 2 after switching to low serum medium compared to co-culturing with rescued mIMCD3 cells, while the level of α-SMA was less sensitive (Fig. 2E).

Together, these results suggest that abnormal communications between co-cultured Arl13b^−/− mIMCD3 and 10T1/2 cells lead to increased HH signaling in 10T1/2 cells. However, this activation of HH signaling is insufficient to trigger a full profibrotic response in these cells.

The GLI inhibitor GANT61 partially suppresses renal cyst progression in Arl13b^f/f;Ksp-Cre mice

Since HH signaling is activated non-cell-autonomously in the Arl13b^f/f;Ksp-Cre kidney, to investigate the functional significance of this abnormal activation in PKD progression, we used GANT61, a small molecule that interferes with the transcriptional activities of both GLI1 and GLI2 (40), to achieve whole body HH inhibition. Our previous study established that the Arl13b^f/f;Ksp-Cre kidney is non-cystic at P7 and cystic at P14 (35). Because of technical difficulties of intraperitoneal (i.p.) injection into P7 mice, we analyzed key characteristics of the Arl13b^f/f;Ksp-Cre kidney at P10. Results showed that the Arl13b^f/f;Ksp-Cre kidney was slightly cystic at this stage (Supplementary Material, Fig. S3A). The kidney weight to body weight (KWB) ratio and cystic index were increased in both female and male mutant mice (Supplementary Material, Fig. S3B–E), although at levels much milder than P14 mutants (35). In addition, blood urea nitrogen (BUN) level was increased when male and female mice were analyzed together or in male mice, but did not reach statistical significance in female mice alone (Supplementary Material, Fig. S3F, G). We further investigated the status of HH activation at this stage. As expected, nLacZ positive cells were observed in both wild-type and mutant kidneys (Supplementary Material, Fig. S3H). Interestingly, the early myofibroblast activation marker VIM was already detected in clusters of cells in the cortex region of the Arl13b^f/f;Ksp-Cre kidney, while it was largely absent in the cortex of control kidneys (Supplementary Material, Fig. S3I). No obvious difference of VIM in the medullar region, which already contains VIM positive cells in the control animal, was detected in the mutant kidney (not shown). Moreover, in the cortex region, in addition to nLacZ positive:VIM negative cells, almost all VIM positive cells in the mutant kidney were GLI1⁺ as indicated by the GLI1-nLacZ signal, consistent with GLI1⁺ cells as progenitors for activated myofibroblasts as shown after injury (42) (Supplementary Material, Fig. S3I). In contrast, no obvious difference of α-SMA signal was detected, suggesting an early stage of myofibroblast activation in the mutant kidney at P10 (Supplementary Material, Fig. S3J). We therefore started GANT61 treatment from P10.

Arl13b^f/f;Ksp-Cre and control mice (Arl13b^f/+;Ksp-Cre) were subjected to daily i.p. injection of GANT61 or vehicle alone (Fig. 3A). After GANT61 treatment, RT-qPCR using total kidney lysates verified that this treatment led to reduced expression levels of the HH target genes Gli1, Ptch1 and Ptch2 in mutant kidneys, and to a lesser degree in control kidneys (Fig. 3B). Both female and male mice treated with GANT61 showed no significant changes in body weight in either the control or mutant group (Fig. 3C). However, the treated mutants show significantly reduced KBW ratio, while no difference in KBW was seen in treated control animals (Fig. 3D). There was also no significant difference between male and female mice in response to the GANT61 treatment (Fig. 3E). Taking advantage of variations in phenotypic severity between individual mice, we asked whether the level of Gli1 expression correlates with KBW ratio by Pearson correlation analysis. Interestingly, while there was no significant correlation between Gli1 level and KBW ratio in vehicle-treated mutants, the correlation coefficient was statistically significant in GANT61-treated mutants, suggesting a significant correlation between the levels of HH signaling and KBW in this group (Fig. 3F, G). In concordance with reduced KBW, the BUN level was significantly reduced in GANT61-treated mutants, indicating preserved kidney function (Fig. 3H). No difference in BUN level was detected between male and female mice (Fig. 3I).

Figure 3

GANT61 suppresses renal phenotypes in Arl13b^f/f;Ksp-Cre (KO) mice. (A) Schematic diagram of the treatment schedule. P10 and P11 mice were subjected to daily i.p. injection of 35 mg/kg/day GANT61, followed by daily i.p. injection of 50 mg/kg/day GANT61 from P12 to P20. (B) GANT61 treatment reduced the expression level of HH target genes Gli1, Ptch1 and Ptch2 in whole kidney lysates of KO and Arl13b^f/+;Ksp-Cre (Ctrl) mice, assayed by RT-qPCR. Gapdh was used for normalization. Unit 1 is defined by the expression level in vehicle-treated animals. Results represent mean ± SD of indicated independent experiments. (C) Body weight in vehicle (Veh) and GANT61-treated redundant with KO and Arl13b^f/+;Ksp-Cre (Ctrl) mice. (D, E) KWB ratio in vehicle (Veh) and GANT61-treated redundant with KO and Arl13b^f/+;Ksp-Cre (Ctrl) mice. (F, G) Pearson correlation between the level of whole kidney Gli1 expression and KWB ratio in vehicle (F) and GANT61 (G) treated Arl13b^f/f;Ksp-Cre mice. (H, I) BUN level in vehicle (Veh) and GANT61-treated redundant with KO and Arl13b^f/+;Ksp-Cre (Ctrl) mice. All quantitative results represent mean ± SD of the indicated number of independent experiments (N). ns, not significant (P > 0.05); ^*^*^*P < 0.001; ^*^*^*^*P < 0.0001.

We then investigated the impact of GANT61 treatment on cyst progression by quantifying cystic index on histological sections (Fig. 4A). Results showed significant reduction of cystic index by GANT61 treatment in mutant mice while no difference was detected between male and female mice (Fig. 4B, C).

Figure 4

GANT61 suppresses both cyst progression and renal fibrosis in Arl13b^f/f;Ksp-Cre (KO) mice. (A) Hematoxylin- and eosin-stained kidney sections from GANT61 and vehicle Arl13b^f/+;Ksp-Cre (Ctrl) and KO mice at P21. Scale bar: 2 mm. (B, C) Cystic index of GANT61 and vehicle (Veh) treated Arl13b^f/+;Ksp-Cre (Ctrl) and KO kidneys at P21. (D) Immunofluorescence staining of kidney sections of vehicle- and GANT61-treated KO mice at P21. Anti-VIM in green, anti-α-SMA in red. DAPI, blue. Scale bar: 20 μm. (E) Trichrome-stained kidney sections of vehicle- and GANT61-treated KO mice at P21. Blue color indicates collagen deposition. Scale bar: 20 μm. (F) Left panel: western blot showing the protein level of collagen I (Col I), α-SMA, PCNA, cyclin D1 and pHH3 in whole kidney lysates of KO mice treated with GANT61 (+) or vehicle (−) at P21. GAPDH is used as a loading control. M: male; F: female. Right panel: quantification of signals on western blot. Adjusted density is defined as relative density of the band of interest normalized by that of GAPDH. Results represent the mean ± SD of indicated number of independent experiments (N). ns, P > 0.05; ^*^*^*^*P < 0.0001.

Taken together, these results demonstrate that GANT61 treatment ameliorates PKD progression in the mutant kidney and this effect correlates with the inhibition of the HH pathway.

GANT61 partially suppresses interstitial fibrosis in the Arl13b^f/f;Ksp-Cre kidney

We then investigated the impact of the GANT61 treatment on interstitial fibrosis in the kidney of Arl13b^f/f;Ksp-Cre mice. Immunofluorescence analysis of kidney sections revealed that VIM and α-SMA signals in the interstitium of the renal cortex and medullar region were decreased in GANT61-treated mutants (Fig. 4D). In addition, trichrome staining showed that the blue staining, i.e. collagen deposition, was reduced in treated mutants, suggesting that GANT61 treatment reduced interstitial fibrosis in mutant kidneys (Fig. 4E). Furthermore, although anti-VIM signal was too noisy on western blot of whole kidney lysates, western blot showed that the protein level of α-SMA and collagen I in whole kidney lysates of mutant mice was reduced by GANT61 treatment (Fig. 4F). Combined, these results suggest that GANT61 treatment reduces myofibroblasts and inhibits fibrosis in the mutant kidney.

GANT61 treatment reduces the number of proliferating epithelial and interstitial cells in the Arl13b^f/f;Ksp-Cre kidney

Since over-proliferation of cyst-lining epithelial cells is thought to contribute to cyst progression (48–50), we used proliferating cell nuclear antigen (PCNA), a nuclear marker for cell turnover, to monitor the impact of GANT61 treatment in tubular cells. Immunofluorescence staining of kidney sections revealed a vast increase of PCNA positive cells in the mutant kidney (Supplementary Material, Fig. S4). To quantify PCNA positive cells in tubules specifically, we performed immunohistochemical analysis, in which histological features facilitated the identification of tubular versus interstitial cells (Fig. 5A). In vehicle-treated samples, the number of nuclear PCNA positive tubular cells were increased in the Arl13b^f/f;Ksp-Cre kidney in both the cortex and medullar region in comparison to the control kidney (Fig. 5A, B). GANT61 treatment reduced the number of PCNA positive tubular cells, although only in the cortex region (Fig. 5B).

Figure 5

GANT61 reduces PCNA positive cells in Arl13b^f/f;Ksp-Cre (KO) kidneys. (A) PCNA staining of kidney sections from GANT61 and vehicle Arl13b^f/+;Ksp-Cre (Ctrl) and KO mice at P21. Brown staining in the nucleus indicates positive cells. Arrows show representative positive tubular cells and arrow heads show representative positive interstitial cells. Scale bar: 20 μm. (B, C) Quantification of PCNA positive cells in tubular (B) and interstitial (C) cells. (D) A model for the functional consequence of non-cell-autonomous activation of the HH pathway. Increased production of HH ligands leads to over-proliferation of GLI1⁺ cells, which are further activated to myofibroblasts and contribute to fibrosis. Results represent the mean ± SD of indicated number of independent experiments (N). ns, P > 0.05; ^*P < 0.05; ^*^*P < 0.01; ^*^*^*P < 0.001; ^*^*^*^*P < 0.0001.

We additionally investigate the number of proliferating cells in the interstitium. Interestingly, PCNA positive interstitial cells increased significantly as well, in both the cortex and medullar region of the mutant kidney and this increase was suppressed by GANT61 treatment (Fig. 5A, C).

In agreement with the immunostaining result, the level of PCNA protein in whole kidney lysates of mutant mice was increased in comparison to control mice; and GANT61 treatment inhibited this abnormal increase, as shown by western blot (Fig. 4F). We further analyzed the level of cyclin D1, which is required for cell proliferation and prevents cells enter quiescence, and phosphorylated histone H3 (pHH3), a marker of mitosis, in whole kidney lysates (51). Results showed that while both cyclin D1 and pHH3 were increased in the mutant kidney, GANT61 treatment reduced their levels (Fig. 4F).

In wild-type control kidneys, while GANT61 treatment had no impact on the number of PCNA positive tubular cells, it modestly reduced the number of proliferating interstitial cells, although to a much lesser degree than in mutant kidneys (Fig. 5B, C).

Together, these results suggest that GANT61 treatment reduces the number of proliferating epithelial and interstitial cells in the mutant kidney.

Discussion

The primary cilium plays a key role in both PKD pathogenesis and HH signaling. However, the relationship between the two is less clear and results from different studies are contradictory. For example, it was shown that inhibiting HH signaling in mouse mutants of Thm1, a negative regulator of HH signaling, partially suppresses cyst progression (33). However, in mouse Pkd1 mutant models, cell-autonomous inhibition of HH signaling, through genetic inactivation of Smo in epithelial cells, does not alter PKD progression (38). These interesting studies suggest that different PKD or ciliary genes might have distinct relationships with HH signaling, hence the different response to HH inhibition. Alternatively, HH signaling could be activated non-cell-autonomously in cystic kidneys, resulting in the lack of response by cell-autonomous inhibition of HH. We therefore tested the mode of HH activity in a model of cilia dysfunction and PKD, using a HH reporter in Arl13b^f/f;Ksp-Cre mutant mice. We show that when Arl13b is deleted in epithelial cells, HH signaling is activated non-cell-autonomously in interstitial cells, consistent with the pattern of HH activation during development and tissue repair (24,25,28,30,39). Although abnormal activation of HH signaling is consistent with multiple renal ciliopathy models, and though Arl13b is a cilia biogenesis gene, the possibility that defective extra-ciliary function of ARL13B is responsible for the abnormal activation of HH signaling in the Arl13b^f/f;Ksp-Cre kidney cannot be ruled out at the current stage. Testing additional cilia biogenesis mutants will further clarify the role of defective cilia in non-cell-autonomous activation of HH signaling.

Non-cell-autonomous HH signaling could be activated through secreted ligands or even through mechanical stress exerted by expanding cysts. To investigate whether mechanisms independent of cyst expansion contribute to the activation of HH signaling in interstitial cells, we used cultured cells and showed that 10T1/2 mesenchymal cells are ciliated and respond robustly to HH stimulation. In contrast, differentiated renal epithelial cells respond poorly to HH stimulation, potentially providing a mechanism for tissue specific response of HH activation. Moreover, when co-cultured with mesenchymal cells, Arl13b^−/− epithelial cells produce more SHH, accompanied by increased expression of HH target genes in the co-cultured mesenchymal cells. These results suggest that the production of HH ligands in epithelial cells can be induced in the absence of cyst expansion; the mechanism, however, remains undefined and we cannot rule out the possibility that cilia elongation in rescued cells inhibits HH ligand production. Cross-talk in vivo would be vastly more complex. In addition to a cyst independent and cell-autonomous mechanism for increased Shh expression in mutant epithelial cells, it is plausible that injury signals could additionally induce Shh expression indirectly. It is also plausible that disturbed luminal fluid flow in tubules and mechanical stress exerted by cyst expansion could further enhance the HH pathway directly or indirectly; and more cell types are most likely involved.

The severe phenotypic consequences of defective epithelial cilia in the Arl13b^f/f;Ksp-Cre kidney, coupled with the lack of strong HH response in epithelial cells in this model, suggest the existence of a separate cilia-mediated pathway in epithelial cells, disruption of which triggers a cascade of events leading to the formation of epithelial cysts and non-cell-autonomous activation of HH signaling in interstitial cells. The relationship between this undefined ciliary pathway and the CDCA pathway, which also functions in epithelial cilia, is unknown. It will be interesting to investigate whether the CDCA pathway plays a role in non-cell-autonomous activation of the HH pathway in cystic kidney models.

Given the non-cell-autonomous nature of HH activation in the Arl13b^f/f;Ksp-Cre kidney, we sought to investigate the functional significance of abnormal activation of HH signaling by using the small molecule GLI inhibitor GANT61 to reduce HH signaling globally. GANT61 treatment attenuated both renal fibrosis and cyst progression in Arl13b^f/f;Ksp-Cre mutants. The impact on fibrosis is consistent with the role of HH in renal fibrosis triggered by kidney injury. Previous studies showed that HH signaling is activated after kidney injury and this activation is integral to repair (28,30,39,47,52). Functionally, both global pharmacological inhibition of HH signaling and targeted genetic deletion of Shh in renal epithelial cells attenuated myofibroblast activation after kidney injury (30,39,47,52). GANT61 could similarly attenuate fibrosis in the Arl13b^f/f;Ksp-Cre kidney through inhibiting myofibroblast activation. The reduced level of α-SMA and VIM in treated mutant kidneys is consistent with this model. However, co-culturing with Arl13b^−/− mIMCD3 was insufficient to induce 10T1/2 cells into α-SMA positive cells, despite the activation of HH signaling in these cells. In contrast, 10T1/2 cells are capable of differentiating into α-SMA positive cells when treated with TGFβ (53). Notably, exogenous SHH stimulates the proliferation of 10T1/2 cells and inhibition of HH signaling causes cell cycle arrest in 10T1/2 cells (28,52). Moreover, it was shown that in a mouse medulloblastoma cell line, transient cilia in cycling cells mediate HH-dependent proliferation (54). Furthermore, we showed that the number of proliferating interstitial cells is increased in the Arl13b^f/f;Ksp-Cre kidney and that this increase is attenuated by GANT61 treatment. Based on these results, we propose that Arl13b^−/− epithelial cells trigger activation of HH signaling in GLI1⁺ interstitial cells, leading to increased proliferation of interstitial cells, which are further activated into myofibroblasts by additional signals such as TGFβ (Fig. 5D). By inhibiting the proliferation of GLI1⁺ interstitial cells, GANT61 treatment reduces the number of activated myofibroblasts and interstitial fibrosis in the Arl13b^f/f;Ksp-Cre kidney.

Interestingly, GANT61 treatment also reduced the number of proliferating epithelial cells and ameliorated cyst progression. Whether this effect is through reduced myofibroblasts or a more indirect mechanism such as macrophage recruitment, which has an established role in cyst progression, remains to be investigated (55,56). In addition, while ligand-mediated activation of the HH pathway is abrogated in cilia biogenesis mutants, the processing of GLI3 to the repressor GLI3R, is also reduced in these mutants (20). This suggests that cilia-less mutant cells could have a higher basal level of GLI-mediated transcription, the target of GANT61 (40). In contrast, Smo deletion would be ineffective as the cilium is downstream from SMO in HH signaling. Given the modest reduction of cyst progression by GANT61 treatment, abnormal activation of HH is likely a contributing factor of cystogenesis. In addition, since HH signaling plays a critical role in tissue homeostasis and repair, toxicity of GANT61 may prevent its clinical use against PKD. It will be vital for our understanding of PKD to further dissect the pathways mediated by cilia in epithelial cells and investigate whether interstitial cells are also HH responsive cells in additional PKD models, such as polycystin mutants and mutants affecting different aspects of cilia biogenesis and signaling.

Materials and Methods

Animals

Mouse experiments in this study were carried out at Yale University School of Medicine in accordance with the Animal Use Protocols as approved by the Institutional Animal Care and Use Committee. Arl13b^f/f;Ksp-Cre mice have been previously described (35). Gli1^lacZ mice (Jackson Laboratory, Bar Harbor, ME, USA) were provided by the Liem lab. To generate Arl13b^f/f;Ksp-Cre;Gl1i^lacZ/+ mice and control littermates (Arl13b^f/+;Ksp-Cre;Gli1^lacZ/+), Arl13b^f/f;Gli1^lacZ/+ mice were crossed to Arl13b^f/+;Ksp-Cre mice.

GANT61 treatment

GANT61 (ApexBio Technology, A1615, 25 mg) was dissolved in ethanol (1.75 mL) and stored at −80°C. For mouse injection, the ethanol solution was further diluted in PBS (3:7) immediately before injection. GANT61 was intraperitoneally administered to male and female mice at a dose of 30 mg/kg body weight (BW)/day at both P10 and P11, and 50 mg/kg BW/day from P12 to P20. Vehicle alone was injected following the same schedule. Mice were sacrificed at P21.

Monolayer cell culture

Arl13b^−/− and rescued mIMCD3 cell lines (43) were cultured in Dulbecco’s modified Eagle’s medium/Nutrient Mixture F-12 (Invitrogen, 11330-032) supplemented with 10% heat-inactivated fetal bovine serum (FBS, Invitrogen, 16410) and 1% Antibiotic-Antimycotic solution (Gibco, 15240062). C3H/10T1/2, clone 8 (10T1/2) cells were obtained from the American Type Culture Collection (ATCC, CCL-226) and cultured in Basal Medium Eagle (BEM, Gibco, 21010046) supplemented with 10% FBS, 2 mm l-glutamine (Gibco, 25030081) and 1% Antibiotic-Antimycotic solution. To induce primary cilium formation, cells were switched to low-serum (0.5% FBS) medium after reaching confluency and maintained for the duration indicated in each experiment. For SAG treatment, SAG (Cayman Chemical, 11914) was dissolved in DMSO to make a stock solution (5 mm). After 1 day of culturing in low serum medium, cells were treated with varying concentrations of SAG in medium containing 0.5% FBS for 24 h. The same concentration of DMSO was used as vehicle control.

Co-culture of epithelial and mesenchymal cells

mIMCD3 cells (2 × 10⁵/well) were seeded on the underside of 6-well PET transwell inserts (0.4 μm pore membrane, Fisher Scientific, 08-771) for 5 h. Inserts were then inverted, and 10T1/2 cells (1.22 × 10⁵/well) were added on the top side of each insert. Co-cultured cells were maintained in BEM supplemented with 10% FBS, 2 mm l-glutamine and 1% Antibiotic-Antimycotic solution until confluent. After 2 days (when cells reached confluency), medium was replaced with BEM supplemented with 0.5% FBS, 2 mm l-glutamine and 1% Antibiotic-Antimycotic solution. After incubation, cells were collected by scraping, transferred to microfuge tubes and centrifuged at 500g for 10 min at 4°C to obtain cell pellets for further analyses.

Determining the level of the active N-terminal fragment of SHH in culture medium

The level of the activate N-terminal fragment of SHH in culture supernatants was determined with ELISA using mouse SHH N-Terminus Quantitkine ELISA kit (R&D Systems, MSHH00) following the protocol provided by the manufacturer.

Western blot using cultured cells and mouse kidney tissues

Cell pellets and kidney tissue samples were homogenized in whole cell extract (WCE) buffer containing 20 mm HEPES, pH 7.4, 0.2 M NaCl, 0.5% Triton X-100, 5% glycerol, 1 mm EDTA, 1 mm EGTA, 1 mm DTT, Phosphatase Inhibitor Cocktail (Thermo Fisher Scientific, 78440) and Protease Inhibitor Cocktail (Roche, 11697498001). Samples were lysed by trituration through a 25-gauge needle 10 times on ice, rotated for 30 min at 4°C, and centrifuged (21 000g, 30 min, 4°C) to obtain WCEs. Protein concentration was determined by Bradford assay (Bio-Rad). Lysates were boiled in 1× loading buffer (5× stock contains 250 mm Tris–HCl, pH 6.8, 10% SDS, 0.05% bromophenol blue, 50% glycerol, 25% β-mercaptoethanol), subjected to electrophoretic separation by 4–15% SDS–PAGE (Bio-Rad, 5000006) and transferred to Immobilon-P/PVDF membrane (Millipore, IPVH00010). Blots were probed overnight at 4°C with the following primary antibodies: rabbit anti-vimentin at 1:2000 (Proteintech Group Inc., 10366-1-AP), rabbit anti-collagen I at 1:2000 (Proteintech Group Inc., 14695-1-AP), mouse anti-α-SMA at 1:2000 (Abcam, ab7817), rabbit anti-PCNA (Proteintech, 10205-2-AP) at 1:2000, rabbit anti-cylcin D1 (Cell Signaling, 55506) at 1:2000, mouse anti-pHH3 (Cell Signaling, 9706) at 1:2000 and rabbit anti-GAPDH at 1:5000 (GeneTex, GTX100118). Blots were washed and then probed with respective HRP-conjugated secondary antibodies at 1:5000 (Cell Signaling Technology, 7076 and 7074). Signals were subsequently detected by enhanced chemiluminescence reaction (Thermo Fisher Scientific, 34095). For quantification, film was scanned and analyzed using Fiji. The relative density value of the band of interest was calculated as the density value of the band divided by the density value of the control band. Adjusted density was defined as relative density normalized with the corresponding relative density of GAPDH.

RT-qPCR

Total RNA from cell pellets and kidney tissues was isolated using TRIzol reagent (Invitrogen, 1559602) according to manufacturer’s instructions. cDNA was synthesized by Superscript II reverse transcriptase (Invitrogen, 18 064 071) with random hexamers. Primers for real-time PCR assays are listed in Supplementary Material, Table S1. cDNA levels were determined quantitatively by real-time PCR using the Bio-Rad iTaq Universal SYBR Green Supermix system (1725125) and normalized with Gapdh. All results represent the mean ± SD of indicated number of independent experiments.

Histology

Kidneys were fixed in 4% paraformaldehyde (PFA) overnight at 4°C, embedded in paraffin wax, sectioned at 5 μm and stained with hematoxylin and eosin (H&E) or trichrome staining.

Immunofluorescent staining

Kidney samples were fixed in 4% PFA overnight at 4°C, embedded in OCT (Sakura Finetek, 4583) and cryosectioned into 5-μm sections. Cells cultured on slides were fixed in 4% PFA for 5 min. Slides with embedded tissue sections or cultured cells were washed with PBS twice and permeabilized with 0.5% Triton X-100/0.1% Tween-20/PBS at room temperature for 20 min. Slides were then blocked in blocking buffer (R.T.U. Animal Free Block and Diluent, Vector Laboratories, SP-5035-100) and incubated overnight with the following primary antibodies: rabbit anti-laminin (1:200, NB300-144, Novus Biologicals), chicken anti-LacZ (1:3000, BGL-1010, Aves Labs), rabbit anti-vimentin (1:200), mouse anti-α-SMA (1:200), rabbit anti-ARL13B (1:200, Proteintech Group, 17 711-1-AP) and mouse anti-acetylated tubulin (1:5000, Sigma-Aldrich, clone 6–11B-1). Slides were then washed three times in PBS and incubated for 1 h at room temperature in blocking buffer containing secondary antibodies conjugated to Alexa Fluor 568, Alexa Fluor 488 (Invitrogen, A-10680 and A-11011) or Fluorescein (Aves Labs, F-1005). Slides were washed three times in PBS and then mounted with Vectashield Vibrance mounting medium with DAPI (Vector Laboratories, H-1800).

To quantify nLacZ positive cells, cryosections of kidneys were subjected to immunostaining with anti-LacZ in combination with anti-laminin or anti-α-SMA. Five randomly selected high-power fields (400×) and 100–300 cells in each field were evaluated for each mouse. Interstitial cells were defined as cells outside of regions encircled by anti-laminin signal. Two mice were analyzed for each condition. Pie-chart was drawn using GraphPad Prism 8.

Cyst index analysis

Cystic index was quantified as previously described (35). Briefly, H&E-stained sagittal sections of kidney were analyzed using Metamorph v.7.1 acquisition software (Universal Imaging). Cystic index was calculated by dividing cyst-containing area with total kidney area.

BUN measurement

Total blood was collected from mouse in a heparin-coated tube (BD Microtainer, BD Inc.), and centrifuged (6000 rpm, 5 min) to obtain plasma. Plasma BUN was measured using a Urea Nitrogen colorimetric detection kit (Invitrogen, EIABUN) following the manufacturer’s instructions.

Quantification of PCNA positive cells

Paraffin-embedded kidney sections were used for immunohistochemical staining of PCNA (NeoMarkers, ms106) by the Research Histology Facility at Yale School of Medicine. PCNA-positive cells were defined as those with clear, distinct brown nuclear staining. PCNA positive cells were counted in six randomly selected high-power fields (400×). More than 200 cells were evaluated in each chosen field and summarized to obtain a single value for each individual mouse. PCNA-positive tubular or interstitial cells were assessed as the percentage of all analyzed tubular or interstitial cells, respectively.

Statistical analysis

Data are presented as mean with error bars indicating the standard deviation (SD). Statistical significance was calculated by unpaired two-tailed t-test and Pearson correlation analysis using GraphPad Prism 8.

Acknowledgements

We thank the Liem lab for Gli1^lacZ mice; the members of Somlo laboratory and Sun laboratory for helpful discussions; A. Cox for critical reading of the manuscript; S. Mentone in the Microscopy and Imaging Core of Cellular and Molecular Physiology Department and the Research Histology Facility of Yale School of Medicine for histology assistance.

Conflict of Interest statement. None declared.

Funding

National Institute of Diabetes and Digestive and Kidney Diseases (R01DK113135 to Z.S.) and The Eunice Kennedy Shriver National Institute of Child Health and Human Development (R01HD093608 to Z.S.).

References

1.

Goetz

,

S.C.

and

Anderson

,

K.V.

(

2010

)

The primary cilium: a signalling centre during vertebrate development

.

Nat. Rev. Genet.

,

11

,

331

–

344

.

2.

Wheway

,

G.

,

Nazlamova

,

L.

and

Hancock

,

J.T.

(

2018

)

Signaling through the primary cilium

.

Front. Cell Dev. Biol.

,

6

,

8

.

3.

Anvarian

,

Z.

,

Mykytyn

,

K.

,

Mukhopadhyay

,

S.

,

Pedersen

,

L.B.

and

Christensen

,

S.T.

(

2019

)

Cellular signalling by primary cilia in development, organ function and disease

.

Nat. Rev. Nephrol.

,

15

,

199

–

219

.

4.

Waters

,

A.M.

and

Beales

,

P.L.

(

2011

)

Ciliopathies: an expanding disease spectrum

.

Pediatr. Nephrol.

,

26

,

1039

–

1056

.

5.

Novarino

,

G.

,

Akizu

,

N.

and

Gleeson

,

J.G.

(

2011

)

Modeling human disease in humans: the ciliopathies

.

Cell

,

147

,

70

–

79

.

6.

Hildebrandt

,

F.

,

Benzing

,

T.

and

Katsanis

,

N.

(

2011

)

Ciliopathies

.

N. Engl. J. Med.

,

364

,

1533

–

1543

.

7.

Bergmann

,

C.

,

Guay-Woodford

,

L.M.

,

Harris

,

P.C.

,

Horie

,

S.

,

Peters

,

D.J.M.

and

Torres

,

V.E.

(

2018

)

Polycystic kidney disease

.

Nat. Rev. Dis. Primers.

,

4

,

50

.

8.

Barr

,

M.M.

and

Sternberg

,

P.W.

(

1999

)

A polycystic kidney-disease gene homologue required for male mating behaviour in C. elegans

.

Nature

,

401

,

386

–

389

.

PubMed

9.

Cai

,

Y.

,

Fedeles

,

S.V.

,

Dong

,

K.

,

Anyatonwu

,

G.

,

Onoe

,

T.

,

Mitobe

,

M.

,

Gao

,

J.D.

,

Okuhara

,

D.

,

Tian

,

X.

,

Gallagher

,

A.R.

et al. (

2014

)

Altered trafficking and stability of polycystins underlie polycystic kidney disease

.

J. Clin. Invest.

,

124

,

5129

–

5144

.

10.

Pazour

,

G.J.

,

San Agustin

,

J.T.

,

Follit

,

J.A.

,

Rosenbaum

,

J.L.

and

Witman

,

G.B.

(

2002

)

Polycystin-2 localizes to kidney cilia and the ciliary level is elevated in orpk mice with polycystic kidney disease

.

Curr. Biol.

,

12

,

R378

–

R380

.

11.

Yoder

,

B.K.

,

Hou

,

X.

and

Guay-Woodford

,

L.M.

(

2002

)

The polycystic kidney disease proteins, polycystin-1, polycystin-2, polaris, and cystin, are co-localized in renal cilia

.

J. Am. Soc. Nephrol.

,

13

,

2508

–

2516

.

12.

Yoshiba

,

S.

,

Shiratori

,

H.

,

Kuo

,

I.Y.

,

Kawasumi

,

A.

,

Shinohara

,

K.

,

Nonaka

,

S.

,

Asai

,

Y.

,

Sasaki

,

G.

,

Belo

,

J.A.

,

Sasaki

,

H.

et al. (

2012

)

Cilia at the node of mouse embryos sense fluid flow for left-right determination via Pkd2

.

Science

,

338

,

226

–

231

.

13.

Kim

,

I.

,

Ding

,

T.

,

Fu

,

Y.

,

Li

,

C.

,

Cui

,

L.

,

Li

,

A.

,

Lian

,

P.

,

Liang

,

D.

,

Wang

,

D.W.

,

Guo

,

C.

et al. (

2009

)

Conditional mutation of Pkd2 causes cystogenesis and upregulates beta-catenin

.

J. Am. Soc. Nephrol.

,

20

,

2556

–

2569

.

14.

Lantinga-van Leeuwen

,

I.S.

,

Leonhard

,

W.N.

,

van der

Wal

,

A.

,

Breuning

,

M.H.

,

de

Heer

,

E.

and

Peters

,

D.J.

(

2007

)

Kidney-specific inactivation of the Pkd1 gene induces rapid cyst formation in developing kidneys and a slow onset of disease in adult mice

.

Hum. Mol. Genet.

,

16

,

3188

–

3196

.

15.

Lin

,

F.

,

Hiesberger

,

T.

,

Cordes

,

K.

,

Sinclair

,

A.M.

,

Goldstein

,

L.S.

,

Somlo

,

S.

and

Igarashi

,

P.

(

2003

)

Kidney-specific inactivation of the KIF3A subunit of kinesin-II inhibits renal ciliogenesis and produces polycystic kidney disease

.

Proc. Natl. Acad. Sci. U. S. A.

,

100

,

5286

–

5291

.

16.

Ma

,

M.

,

Tian

,

X.

,

Igarashi

,

P.

,

Pazour

,

G.J.

and

Somlo

,

S.

(

2013

)

Loss of cilia suppresses cyst growth in genetic models of autosomal dominant polycystic kidney disease

.

Nat. Genet.

,

45

,

1004

–

1012

.

17.

Dwivedi

,

N.

,

Tao

,

S.

,

Jamadar

,

A.

,

Sinha

,

S.

,

Howard

,

C.

,

Wallace

,

D.P.

,

Fields

,

T.A.

,

Leask

,

A.

,

Calvet

,

J.P.

and

Rao

,

R.

(

2020

)

Epithelial vasopressin Type-2 receptors regulate myofibroblasts by a YAP-CCN2-dependent mechanism in polycystic kidney disease

.

J. Am. Soc. Nephrol.

,

31

,

1697

–

1710

.

18.

Hassane

,

S.

,

Leonhard

,

W.N.

,

van der

Wal

,

A.

,

Hawinkels

,

L.J.

,

Lantinga-van Leeuwen

,

I.S.

,

ten

Dijke

,

P.

,

Breuning

,

M.H.

,

de

Heer

,

E.

and

Peters

,

D.J.

(

2010

)

Elevated TGFbeta-Smad signalling in experimental Pkd1 models and human patients with polycystic kidney disease

.

J. Pathol.

,

222

,

21

–

31

.

19.

Song

,

C.J.

,

Zimmerman

,

K.A.

,

Henke

,

S.J.

and

Yoder

,

B.K.

(

2017

)

Inflammation and fibrosis in polycystic kidney disease

.

Results Probl. Cell Differ.

,

60

,

323

–

344

.

20.

Haycraft

,

C.J.

,

Banizs

,

B.

,

Aydin-Son

,

Y.

,

Zhang

,

Q.

,

Michaud

,

E.J.

and

Yoder

,

B.K.

(

2005

)

Gli2 and Gli3 localize to cilia and require the intraflagellar transport protein polaris for processing and function

.

PLoS Genet.

,

1

, e53.

21.

Huangfu

,

D.

,

Liu

,

A.

,

Rakeman

,

A.S.

,

Murcia

,

N.S.

,

Niswander

,

L.

and

Anderson

,

K.V.

(

2003

)

Hedgehog signalling in the mouse requires intraflagellar transport proteins

.

Nature

,

426

,

83

–

87

.

22.

Rohatgi

,

R.

,

Milenkovic

,

L.

and

Scott

,

M.P.

(

2007

)

Patched1 regulates hedgehog signaling at the primary cilium

.

Science

,

317

,

372

–

376

.

23.

Yu

,

J.

,

Carroll

,

T.J.

and

McMahon

,

A.P.

(

2002

)

Sonic hedgehog regulates proliferation and differentiation of mesenchymal cells in the mouse metanephric kidney

.

Development

,

129

,

5301

–

5312

.

24.

McMahon

,

A.P.

,

Ingham

,

P.W.

and

Tabin

,

C.J.

(

2003

)

Developmental roles and clinical significance of hedgehog signaling

.

Curr. Top. Dev. Biol.

,

53

,

1

–

114

.

25.

Lum

,

L.

and

Beachy

,

P.A.

(

2004

)

The Hedgehog response network: sensors, switches, and routers

.

Science

,

304

,

1755

–

1759

.

26.

Chang

,

C.F.

,

Chang

,

Y.T.

,

Millington

,

G.

and

Brugmann

,

S.A.

(

2016

)

Craniofacial ciliopathies reveal specific requirements for GLI proteins during development of the facial midline

.

PLoS Genet.

,

12

, e1006351.

27.

Liu

,

A.

,

Wang

,

B.

and

Niswander

,

L.A.

(

2005

)

Mouse intraflagellar transport proteins regulate both the activator and repressor functions of Gli transcription factors

.

Development

,

132

,

3103

–

3111

.

28.

Fabian

,

S.L.

,

Penchev

,

R.R.

,

St-Jacques

,

B.

,

Rao

,

A.N.

,

Sipila

,

P.

,

West

,

K.A.

,

McMahon

,

A.P.

and

Humphreys

,

B.D.

(

2012

)

Hedgehog-Gli pathway activation during kidney fibrosis

.

Am. J. Pathol.

,

180

,

1441

–

1453

.

29.

Horn

,

A.

,

Palumbo

,

K.

,

Cordazzo

,

C.

,

Dees

,

C.

,

Akhmetshina

,

A.

,

Tomcik

,

M.

,

Zerr

,

P.

,

Avouac

,

J.

,

Gusinde

,

J.

,

Zwerina

,

J.

et al. (

2012

)

Hedgehog signaling controls fibroblast activation and tissue fibrosis in systemic sclerosis

.

Arthritis Rheum.

,

64

,

2724

–

2733

.

30.

Zhou

,

D.

,

Li

,

Y.

,

Zhou

,

L.

,

Tan

,

R.J.

,

Xiao

,

L.

,

Liang

,

M.

,

Hou

,

F.F.

and

Liu

,

Y.

(

2014

)

Sonic hedgehog is a novel tubule-derived growth factor for interstitial fibroblasts after kidney injury

.

J. Am. Soc. Nephrol.

,

25

,

2187

–

2200

.

31.

Hu

,

B.

,

Liu

,

J.

,

Wu

,

Z.

,

Liu

,

T.

,

Ullenbruch

,

M.R.

,

Ding

,

L.

,

Henke

,

C.A.

,

Bitterman

,

P.B.

and

Phan

,

S.H.

(

2015

)

Reemergence of hedgehog mediates epithelial-mesenchymal crosstalk in pulmonary fibrosis

.

Am. J. Respir. Cell Mol. Biol.

,

52

,

418

–

428

.

32.

Jung

,

I.H.

,

Jung

,

D.E.

,

Park

,

Y.N.

,

Song

,

S.Y.

and

Park

,

S.W.

(

2011

)

Aberrant Hedgehog ligands induce progressive pancreatic fibrosis by paracrine activation of myofibroblasts and ductular cells in transgenic zebrafish

.

PLoS One

,

6

, e27941.