Cyst growth in ADPKD is prevented by pharmacological and genetic inhibition of TMEM16A in vivo

Abstract

In autosomal dominant polycystic kidney disease (ADPKD) multiple bilateral renal cysts gradually enlarge, leading to a decline in renal function. Transepithelial chloride secretion through cystic fibrosis transmembrane conductance regulator (CFTR) and TMEM16A (anoctamin 1) are known to drive cyst enlargement. Here we demonstrate that loss of Pkd1 increased expression of TMEM16A and CFTR and Cl^- secretion in murine kidneys, with TMEM16A essentially contributing to cyst growth. Upregulated TMEM16A enhanced intracellular Ca²⁺ signaling and proliferation of Pkd1-deficient renal epithelial cells. In contrast, increase in Ca²⁺ signaling, cell proliferation and CFTR expression was not observed in Pkd1/Tmem16a double knockout mice. Knockout of Tmem16a or inhibition of TMEM16A in vivo by the FDA-approved drugs niclosamide and benzbromarone, as well as the TMEM16A-specific inhibitor Ani9 largely reduced cyst enlargement and abnormal cyst cell proliferation. The present data establish a therapeutic concept for the treatment of ADPKD.

Fig. 1. Cyst formation by Polycystin-1-deficient cells depends on cAMP- and Ca²⁺-activated Cl⁻ secretion.

a, b Wild type (Pkd1^+/+) and polycystin-1-deficient (Pkd1^−/−) principal-like MDCK cells (clones #1) were cultured in a collagen matrix in the absence (−FSK) or presence (−FSK) of 10 µM forskolin. Forskolin induced cyst formation by PKD1^+/+ cells within 5 days (^§P < 0.0001). Pkd1^−/− cells demonstrated cyst formation even in the absence of FSK (^#P < 0.0001) and formed larger cysts in the presence of FSK (^#P < 0.0001) (n = 105 cysts examined in n = 3 independent experiments). c Deletion of Polycystin-1 in Pkd1^−/− cells induced an increase in basal intracellular cAMP concentrations (−FSK, ^#P = 0.041). FSK-stimulation further enhanced cAMP levels in both Pkd1^+/+ and Pkd1^−/− cells (^§P = 0.0008) (each n = 3 independent experiments). cAMP levels in Pkd1^+/+ cells were set to 100%). d, e Cyst growth was strongly inhibited in both Pkd1^+/+ and Pkd1^−/− cells by CFTRinh172 (CFTRinh; 10 µM, ^§P < 0.0001), niclosamide (Niclo; 1 µM, ^§P = 0.0001), and suramin (Sur; 100 µM, ^§P = 0.002), but was further augmented by ATP (10 µM, ^§P = 0.002). Bars 200 μm (n = 87 cysts examined in n = 3 independent experiments). Mean and error bars indicating ±SEM. ^#Unpaired two-sided t test comparing Pkd1^+/+ with Pkd1^−/−, ^§unpaired two-sided t test comparing effects by forskolin, ATP or inhibitors. Source data are provided as a Source Data file.

Fig. 2. Tubule-specific deletion of TMEM16a reduces cyst progression in an ADPKD mouse model.

a–d Western blotting from whole kidney lysates detecting tubule-specific deletion of Pkd1 (450 kDa) (^#P = 0.03) and increase in TMEM16A expression (glycosylated 130 kDa and non-glycosylated forms) (^#P = 0.003) (10 weeks after induction at postnatal days 20–22 using tamoxifen. Mice co-deleted for Pkd1 and Tmem16a (Pkd1^−/−/T16a^−/−) showed reduced Pkd1 expression (^#P = 0.022) and TMEM16A expression (^§P = 0.0004) (n = 3 animals each). e, f Tubule-specific knockout of Pkd1 (Pkd1^−/−) induced polycystic kidney disease (^#P < 0.0001, n = 11 animals), which was largely reduced in mice with an additional knockout of Tmem16a (Pkd1^−/−/T16a^−/−) (^§P = 0.0001, n = 11 animals). Non-induced Pkd1^fl;fl mice served as control (Pkd1^+/+) (n = 8 animals). Corresponding cystic indices (defined as cortical cyst area normalized to whole cortex area). Mean and error bars indicating ±SEM. ^#Unpaired two-sided t test comparing Pkd1^+/+ with Pkd1^−/−, ^§unpaired two-sided t test comparing Pkd1^−/− with Pkd1^−/−/T16a^−/−. Source data are provided as a Source Data file.

Fig. 3. Expression of CFTR and cell proliferation in Pkd1−/− kidneys depends on TMEM16A.

a Enhanced tubular epithelial expression of CFTR and TMEM16A in kidneys from Pkd1^−/− mice that was abolished in kidneys from Pkd1^−/−/T16a^−/− double-knockout mice. Both, CFTR and TMEM16A were colocalized in the apical membrane of the cyst epithelium. (blue; DAPI). b–d Expression of TMEM16A (^#P < 0.0001) and CFTR (^#P < 0.0001) were enhanced (n = 25 independent images over 5 animals) in Pkd1^−/− mice. Significant colocalization of TMEM16A and CFTR was detected (^#P = 0.0006) using colocalization finder algorithm (ImageJ, V.1.49 by Laummonerie and Mutterer). Expression of CFTR (^§P = 0.0006) and TMEM16A (^§P = 0.0006), as well as colocalization (^§P = 0.0001) were lower in Pkd1^−/−/T16a^−/− mice. e, f ki-67 staining was enhanced in tubular epithelial cells from Pkd1^+/+ (^#P < 0.0001) animals but was lower in cells from Pkd1^−/−/T16a^−/− kidneys (^§P < 0.0001) (n = 25 independent images from 5 animals). Mean and error bars indicating ±SEM. ^#§one-way ANOVA and Tukey’s post-hoc test comparing Pkd1^+/+ with Pkd1^−/− and Pkd1^−/− with Pkd1^−/−/T16a^−/−, respectively. Source data are provided as a Source Data file.

Fig. 4. Cl⁻ secretion and enhanced Ca²⁺ signaling by knockdown of Polycystin-1 in primary renal epithelial cells.

a Original Ussing chamber recordings from polarized grown primary medullary epithelial cells from Pkd1^+/+, Pkd1^−/−, and Pkd1^−/−/T16a^−/− kidneys. b, c Summaries for calculated basal and UTP (100 µM) activated short-circuit currents (I_sc) demonstrating enhanced basal (^#P = 0.0007) and UTP-activated (^#P = 0.0028) Cl⁻ secretion in epithelia lacking expression of Polycystin-1 (n = 45 independent monolayers from 5 animals each). Enhanced basal and UTP-activated Cl⁻ secretion was abolished in Pkd1^−/−/T16a^−/− cells (^§P < 0.024 and ^§P < 0.001, respectively). d–f Basal and ATP (100 µM) induced Ca²⁺ increase were larger (^#P = 0.038 and ^#P = 0.019, respectively) in primary epithelial cells from Pkd1^−/− but from Pkd1^−/−/T16a^−/− kidneys (n = 44 cells from 4 independent cultures of 4 animals each). g, h Original recording and summary demonstrating the inhibitory effect of benzbromarone (BBR, 10 µM) on ATP (100 µM) induced Ca²⁺ peak (^#P < 0.0001) and plateau (^#P = 0.005) (n = 129 cells from 5 independent cultures of 5 animals each). i, j Original recordings and summaries of cyclopiazonic acid (CPA, 10 µM) induced ER Ca²⁺ store release (Store) and activation of store-operated Ca²⁺ influx (SOCE) in medullary primary epithelial cells. Store release (^#P = 0.005) and SOCE (^#P < 0.0001) were larger in Pkd1^−/− cells, but were reduced in cells from Pkd1^−/−/T16^−/− kidneys (^§P < 0.0001 and ^§P < 0.0001, respectively) (n = 117 cells from 4 independent cultures of 5 animals each). Mean and error bars indicating ±SEM. ^#§One-way ANOVA and Tukey’s post-hoc test comparing Pkd1^+/+ with Pkd1^−/− and Pkd1^−/− with Pkd1^−/−/T16a^−/−, respectively. Source data are provided as a Source Data file.

Fig. 5. Inhibition of enhanced store-operated Ca²⁺ entry (SOCE) by TRP and Orai channel inhibitors.

a–d Original recordings and summaries showing inhibition of upregulated ER Ca²⁺ store release (Store) and store-operated Ca²⁺ entry (SOCE) in primary renal epithelial cells from Pkd1^−/− mice in the presence of 5 µM Orai channel inhibitor YM58483 (^#P < 0.0001 and ^#P < 0.0001, respectively) and 5 µM TRP-channel inhibitor SK&F96365 (^#P < 0.0001 and ^#P < 0.0001, respectively) (n = 73 cells from 5 independent cultures of 4 animals each). e, f Semiquantitative RT-PCR analysis of Orai1 and TRP-channel (Trpc1, Trpc6, Trpm6, Trpv4, Trpv5) expression in Pkd1^+/+ and Pkd1^−/− renal epithelial cells. In Pkd1^−/− cells mRNA for Pkd1 (^#P = 0.03) and Trpc6 (^#P = 0.03) were reduced, while mRNA for Trpm1 was enhanced (^#P = 0.03) (n = 5 reactions from 3 animals each). Mean and error bars indicating ±SEM. ^#§One-way ANOVA and Tukey’s post-hoc test comparing Pkd1^+/+ with Pkd1^−/− and Pkd1^−/− with Pkd1^−/−/T16a^−/−, respectively. Source data are provided as a Source Data file.

Fig. 6. Niclosamide inhibits polycystic kidney disease in a mouse model for ADPKD.

a, b Representative images of kidneys from non-induced (Pkd1^+/+) (n = 4 animals) and induced (Pkd1^−/−) mice, untreated (n = 5 animals) or treated with niclosamide (n = 8 animals). Scale bar 5000 µm. Tubule-specific deletion of Pkd1 (Pkd1^−/−) was induced at postnatal days 20–22. Thereafter, mice were either treated with standard diet or diet supplemented with 0.2% niclosamide (Niclo) for 10 weeks. Pkd1^+/+ mice fed with standard diet served as control. Cyst formation in Pkd1^−/− (^#P = 0.003) was strongly inhibited by niclosamide treatment (^§P = 0.005). c, d Proliferating cells identified in Pkd1^+/+ and Pkd1^−/− kidneys by the proliferation marker ki-67 (red). Proliferation was enhanced in Pkd1^−/− kidneys (^#P < 0.0001), but was significantly inhibited by niclosamide (^§P < 0.0001). Number of ki-67 positive cells was normalized to the tissue area (n = 25 independent images from 5 animals each). Mean and error bars indicating ±SEM. ^#§One-way ANOVA and Tukey’s post-hoc test. Source data are provided as a Source Data file.

Fig. 7. Benzbromarone inhibits polycystic kidney disease in a mouse model for ADPKD.

a Representative images of kidneys from non-induced (Pkd1^+/+) and induced (Pkd1^−/−) mice (n = 5 animals each), untreated (daily i.p. injection of corn oil) or treated with benzbromarone in corn oil (BBR, 1 mg/kg/day, i.p.) for 30 days. Scale bars: 1000 µm. b Cystic index (%) 8 weeks after induction of Pkd1^+/+ and Pkd1^−/− mice, untreated or treated with benzbromarone. Cyst growth was induced in Pkd1^−/− mice (^#P < 0.0001), but was inhibited by benzbromarone (^§P < 0.0001) (n = 10 kidneys analyzed from n = 5 animals each). c Time course of body weight for untreated and benzbromarone-treated Pkd1^+/+ and Pkd1^−/− mice. Body weight gain was enhanced in Pkd1^−/− mice (^#P < 0.05), but was inhibited by benzbromarone (^§P < 0.05) (n = 8 kidneys analyzed from n = 5 animals each). d, e Proliferating cells identified in Pkd1^+/+ and Pkd1^−/− kidneys by the proliferation marker ki-67 (red). Proliferation was upregulated in Pkd1^−/− kidneys (^#P < 0.0001), but was inhibited by benzbromarone (^§P < 0.0001) (25 independent images from n = 5 animals each). Scale bar 50 µm. Number of ki-67 positive cells was normalized to the tissue area. Mean and error bars indicating ±SEM. ^#§One-way ANOVA and Tukey’s post-hoc test comparing Pkd1^+/+ with Pkd1^−/− and Pkd1^−/− with Pkd1^−/−/T16a^−/−, respectively. Source data are provided as a Source Data file.

Fig. 8. Ani9 inhibits polycystic kidney disease in a mouse model for ADPKD.

a Representative images of kidneys from induced (Pkd1^−/−) mice, untreated (daily i.p. injection of corn oil) (n = 8 kidneys analyzed from n = 4 animals) or treated with Ani9 in corn oil (Ani9, 0.5 mg/kg/day, i.p.) for 30 days (n = 8 kidneys analyzed from n = 4 animals). Scale bars 1000 µm. b Cystic index (%) 8 weeks after induction of Pkd1^+/+ and Pkd1^−/− mice, untreated or treated with Ani9. Ani9 reduced cyst growth in Pkd1^−/− mice (^§P = 0.001). ^§Unpaired two-sided t test. c, d Proliferating cells identified in Pkd1^+/+ and Pkd1^−/− kidneys by the proliferation marker ki-67 (red). Scale bars 50 µm. Proliferation was upregulated in Pkd1^−/− mice (^#P < 0.0001), but was largely inhibited by Ani9 (^§P = 0.021). ^#§One-way ANOVA and Tukey’s post-hoc test. Number of ki-67 positive cells was normalized to the tissue area. Mean and error bars indicating ±SEM (25 independent images from n = 8 kidneys of n = 4 animals each). Mean and error bars indicating ±SEM. Source data are provided as a Source Data file.

Fig. 9. Expression of TMEM16A correlates with cyst size in ADPKD.

Reanalysis of the public data set (GSE7869) from a global gene profiling study on renal cysts reported by Song et al.. The data set comprises microarray expression data from cysts of different sizes from n = 5 PKD1-patients and from n = 3 kidneys of healthy patients. Reanalysis was performed using the Transcriptome Analysis Console (Thermo Fisher). Inspecting the probe covering TMEM16A (ANO1), a significant increase in TMEM16A-mRNA expression was found in the cystic tissue when compared to normal samples. Increase in TMEM16A-expression correlated with cyst size. Mean and error bars indicating ±SEM. Source data are provided as a Source Data file.