Reduction of elevated Gli3 does not alter the progression of autosomal recessive polycystic kidney disease

Abstract

Polycystic kidney diseases (PKD) are genetic disorders which disrupt kidney architecture and function. Autosomal recessive PKD (ARPKD) is a rare form of PKD, caused by mutations in PKHD1, and clinically more severe than the more common autosomal dominant PKD (ADPKD). Prior studies have implicated Hedgehog (Hh) signaling in ADPKD, with increased levels of Hh components in experimental ADPKD and reduced cystogenesis following pharmacological Hh inhibition. In contrast, the role of the Hh pathway in ARPKD is poorly understood. We hypothesized that Hh pathway activity would be elevated during ARPKD pathogenesis, and its modulation may slow disease progression. We utilized Cpk mice which phenocopy ARPKD and generated a PKHD1-mutant spheroid model in human collecting ducts. Significantly elevated levels of the Hh transcriptional effector Gli3 were found in Cpk mice, a finding replicated in PKHD1-mutant spheroids. In Cpk mice, total GLI3 and GLI3 repressor protein levels were also increased. Reduction of increased Gli3 levels via heterozygous genetic deletion in Cpk mice did not affect cyst formation. Additionally, lowering GLI3 transcripts to wildtype levels did not influence PKHD1-mutant spheroid size. Collectively, these data suggest attenuation of elevated Gli3 does not modulate murine and human models of ARPKD.

Keywords: autosomal recessive polycystic kidney disease; cystogenesis; hedgehog signaling; human model; mouse model.

FIGURE 1

Gli3 is elevated at both the transcript and protein level in the cystic kidneys of Cpk ^−/− mice. (a) Representative images of periodic acid‐Schiff staining on 5 μm sections of Cpk ^+/+ and Cpk ^−/− kidneys at P10, P14, and P21. Images taken at 10x magnification. Scale bars: 100 μm. Quantitative RT‐PCR analysis of Ihh, Shh, Ptch1, Smo, Sufu, Gli1, Gli2, and Gli3 transcript levels relative to Gapdh from whole kidney RNA from Cpk ^+/+ and Cpk ^−/− mice at (b) P10, (c) P14, and (d) P21. Ct values of each gene of interest were normalized to littermate controls at each timepoint (Unpaired t‐test, n = 7, 5, 3 for Cpk ^+/+ , n = 4, 6, 5 for Cpk ^−/− mice at P10, P14, and P21, respectively). (e) Western blot analysis of the levels of full‐length GLI3 activator (GLI3A) and truncated GLI3 repressor (GLI3R) in P14 kidney lysates from Cpk ^+/+ (n = 6) and Cpk ^−/− (n = 6). (f) The relative intensity of total GLI3, a sum of GLI3A and GLI3R, (g) GLI3A and (h) GLI3R were quantified by densitometry in arbitrary units relative to endogenous alpha‐tubulin. (i) The ratio of GLI3A: GLI3R was quantified (Mann–Whitney t‐test, n = 6 mice per group). Data represents mean ± SD.

FIGURE 2

Reduced levels of Gli3 have no effect on cyst progression in Cpk mice. (a) Representative images of periodic acid‐Schiff (PAS) staining on 5 μm sections of wildtype, Gli3 ^+/XtJ , Cpk ^−/− , and Gli3 ^+/XtJ ; Cpk ^−/− kidneys at P14. 10× magnification, scale bars: 1 mm. (b) Quantitative RT‐PCR analysis of Gli3 relative to Hprt from whole kidney RNA at P14. Relative Ct values of each gene of interest were normalized to Cpk ^−/− littermates (Unpaired t‐test). (c) Kidney: Body weight ratio and (d) blood urea nitrogen (BUN) levels for each group. One‐way ANOVA with Tukey's multiple comparisons test. Quantification of PAS staining was performed on Cpk ^−/− and Gli3 ^+/XtJ ; Cpk ^−/− mice to determine (e) cystic index, the percentage of total cyst area/total kidney area, (f) average cyst size (μm²) and (g) total number of cysts per kidney section. Cysts were defined as ≥10,000 μm². Unpaired t‐test. Data represents mean ± SD. n = 6 for wildtype, n = 6 for Gli3 ^+/XtJ , n = 5 for Cpk ^−/− , and n = 7 for Gli3 ^+/XtJ ; Cpk ^−/− mice.

FIGURE 3

Generation of a human in vitro model of PKHD1‐mutant spheroids. (a) CRISPR‐Cas9 was utilized to generate mutations within exon 5 of PKHD1 in a human collecting duct cell line. Alignment of Sanger sequencing data of PKHD1 exon 5 from clonal cell lines generated was analyzed using Synthego ICE analysis. PKHD1 wildtype sequence is underlined and sgRNA‐binding site is highlighted in pink. Nucleotide deletion sites are indicated with a dash. (b) Representative images of isogenic wildtype control and PKHD1‐mutant 3D spheroids formed in Matrigel following 6 days in culture. (c) The cross‐sectional area of each spheroid was quantified, and the average spheroid area was calculated for each independent repeat. (d) Area of each individual spheroid quantified across all repeats. n = 5 independent repeats. (e) Quantitative RT‐PCR analysis of GLI3 relative to GAPDH from RNA extracted from isogenic wildtype control and PKHD1‐mutant HCD cells cultured for 6 days. Relative Ct values of each gene of interest were normalized to isogenic wildtype control HCD cells. Data represent mean ± SD. Unpaired t‐test. n = 4 independent repeats.

FIGURE 4

Cyclopamine treatment inhibits PKHD1‐mutant spheroid size in vitro. Isogenic wildtype control and PKHD1‐mutant HCD cells were embedded in Matrigel and cultured for 6 days to generate 3D spheroids. Cells were left untreated or treated with 10 μM cyclopamine, a SMO inhibitor, every 2 days. (a) Representative images of isogenic wildtype control and PKHD1‐mutant spheroids at day 6, following 10 μM cyclopamine treatment. (b) The cross‐sectional area of each spheroid was quantified, and the average spheroid area was calculated for each independent repeat. (c) Area of each individual spheroid quantified across all repeats. n = 3 independent repeats. Data represent mean ± SD. Two‐way ANOVA with Tukey's multiple comparisons test. Statistical comparisons are shown relative to the untreated isogenic wildtype control and between conditions of the same cell type.

FIGURE 5

GLI3 knockdown does not alter size of PKHD1‐mutant spheroids. Isogenic wildtype control and PKHD1‐mutant cells were transfected with non‐targeting siRNA or GLI3 siRNA and embedded in Matrigel to generate spheroid structures. (a) Quantitative RT‐PCR analysis of GLI3 relative to GAPDH from RNA extracted from control and PKHD1‐mutant HCD cells for all conditions 6 days after transfection. Relative Ct values of each gene of interest were normalized to untreated isogenic control HCD cells. (b) Representative images of isogenic control and PKHD1‐mutant 3D spheroids that were not transfected or transfected with non‐targeting siRNA or GLI3 siRNA. Images of the spheroids were taken at Day 6 at 20× magnification. (c) The cross‐sectional area of each spheroid was quantified, and the average spheroid area was calculated for each independent repeat. Data represent mean ± SD. Two‐way ANOVA with Tukey's multiple comparisons test. Statistical comparisons are shown relative to the untransfected isogenic wildtype control and between conditions of the same cell type.