A Ciliary Phosphoinositide Pathway Regulates the Dosage of Polycystins in Primary Cilia

Visual Abstract

Abstract

Key Points

• Inositol polyphosphate-5-phosphatase E (INPP5E) and type Ig phosphatidylinositol-4-phosphate 5-kinase (PIPKIγ) coordinated the homeostasis of phosphatidylinositol-4-phosphate and phosphatidylinositol-4,5-bisphosphate in primary cilia.

• Modulating INPP5E or PIPKIγ activity changed the level of polycystin-1 and polycystin-2 in primary cilia.

• INPP5E inhibition increased the hypomorphic PKD1 variants (PC1-R3277C) in cilia and reduced cystogenesis in the kidney in vitro.

Background

Autosomal dominant polycystic kidney disease is mainly caused by mutations in PKD1 and PKD2, which encode polycystin-1 (PC1) and polycystin-2 (PC2), respectively. PC1 and PC2 assemble a cation channel complex enriched in primary cilium, a sensory organelle associated with various developmental diseases, including polycystic kidney disease (PKD). Accumulating evidence supports the necessity of functional polycystin (PC) complex in cilia to prevent cystogenesis in the kidney, indicating that improving their ciliary levels may ameliorate defects underlying PKD pathogenesis. Yet, molecular mechanisms underlying the ciliary targeting and homeostasis of the PC complex are not fully understood.

Methods

Indirect immunofluorescence microscopy was used to monitor ciliary levels of PC1 and PC2 in renal epithelial cells. Electrophysiology analysis in oocytes was used to determine the channel activity of the PC complex. Cystogenesis in the kidney was measured using in vitro 3D-Matrigel cell models and ex vivo mouse embryonic kidney models.

Results

Suppressing inositol polyphosphate-5-phosphatase E (INPP5E) or activating type Ig phosphatidylinositol-4-phosphate 5-kinase raised ciliary levels of the PC complex in both normal renal epithelial cells and cells carrying autosomal dominant polycystic kidney disease mutations that interrupt the trafficking of PCs into cilia, including GANAB inactivation and the trafficking PKD1 mutation p.Arg3277Cys (RC). PC1^RC formed a complex with PC2 and exhibited normal channel activity in vitro. An INPP5E that increases PC1 and PC2 in cilia, suppressed in vitro forskolin-induced cystogenesis of inner medullary collecting duct epithelial cell line 3 cells in 3D Matrigel and ex vivo cyst formation in embryonic Pkd1^RC/RC mouse kidneys.

Conclusions

Our results demonstrated that increasing the ciliary level of PCs, by manipulating a ciliary phosphoinositide signaling axis, enhanced the functionality of PCs and suppressed cystogenesis of renal epithelial cells in vitro.

Introduction

Affecting approximately one in 500–1000 live births, autosomal dominant polycystic kidney disease (ADPKD) is the leading monogenic (third most common overall) cause of kidney failure in adults, with approximately 50% of patients needing dialysis or kidney transplant by the age of 60.^1–3 Most of the ADPKD cases are caused by monoallelic mutations in PKD1 or PKD2,⁴ encoding the multi-transmembrane proteins polycystin-1 (PC1)⁵ and polycystin-2 (PC2),^6,7 respectively. Strong evidence indicates that PC1 and PC2 form a nonselective cation channel complex^6–8 that regulates ion transport and calcium homeostasis in primary cilia.^9–12 Primary cilia are microtubule-based sensory organelles essential for the development and homeostasis of tissues and organs and are thus associated with various genetic diseases collectively termed ciliopathies, including polycystic kidney disease (PKD).^13–22

Compelling evidence from patients with ADPKD and rodent models demonstrate that the likelihood of cyst formation in the kidneys substantially increases when the level of functional PC1 or PC2 drops below a critical threshold.^23–28 Apart from loss-of-function (truncating) mutations in PKD1 and PKD2 identified in patients with ADPKD,²⁹ many PKD1/2 disease mutations are nontruncating; with abnormal folding/maturation (through glycosylation defects) and membrane/ciliary translocation a likely pathogenic mechanism.^23,30–33 Mice expressing the ciliary targeting defective PC2 mutant (PC2^E442G), which retains intact channel activity, develop cysts in the kidney indistinguishable from Pkd2-null mice.³⁴ Recent studies have revealed the importance of RAB11-RAB8 centered, microtubule-dependent trafficking machinery in transporting PCs to the ciliary base and suggested the interdependence of PC1 and PC2 for their appropriate trafficking to the cell surface, including cilia.³⁵ However, the molecular mechanism underlying the ciliary entry and homeostasis of PC1 and/or PC2 in cilia are not fully understood.

The ciliary membrane has a unique molecular composition that is distinct from the plasma membrane,³⁶ including specific compartmentalization of phosphatidylinositol-4-phosphate (PI[4]P) and phosphatidylinositol-4,5-bisphosphate (PI[4,5]P₂).^37–39 Type Ig phosphatidylinositol-4-phosphate 5-kinase (PIPKIγ) and INPP5E, the counteracting phosphoinositide kinase and phosphatase mediating the conversion between PI(4,5)P₂ and PI(4)P, are separately localized to the base and the stalk of cilia, respectively,^40–42 contributing to the enrichment of PI(4)P in the ciliary membrane^{37,38,40–47} and PI(4,5)P₂ around the ciliary base.⁴⁰ Consistent with the critical roles of phosphoinositides in fundamental cellular events,^44,48,49 phosphoinositide-metabolizing enzymes are frequently associated with human diseases.^46,50 A homozygous mutation in PIP5K1C that abrogates the kinase activity of PIPKIγ leads to lethal congenital contracture syndrome type 3 featured with severe defects in joint and muscle development and death from respiratory insufficiency.⁵¹ Similarly, PIPKIγ knockout causes embryonic or postnatal lethality in mice.^52,53 On the other hand, whereas INPP5E knockout mice exhibit typical ciliopathy symptoms including cystogenesis in developing kidneys and die after birth,⁵⁴ biallelic INPP5E mutations are also associated with a subtype of Joubert syndrome, a ciliopathy often including a nephronophthisis phenotype.^55,56

Loss of INPP5E leads to an increase of ciliary PI(4,5)P₂, the PI(4,5)P₂-bound adaptor protein, TULP3, and subsequently, TULP3-mediated trafficking of various ciliary receptors.^57–59 Moreover, small interfering RNA-mediated knockdown of INPP5E results in enhanced ciliary targeting of PC2 through PI(4,5)P₂-dependent accumulation of TULP3.⁶⁰ In this study, we examined whether the ciliary levels of polycystin variants can be modified by changing the PI(4,5)P₂ content within primary cilia and how it may affect the development of kidney cysts.

Methods

Cell Lines

Inner medullary collecting duct epithelial cell line 3 (IMCD3) cells were purchased from American Type Culture Collection. Parental and GANAB^−/− renal cortical tubule epithelial (RCTE) cells were reported previously.³⁰ Immortalized Inpp5e^+/− and Inpp5e^−/− mouse embryonic fibroblasts (MEFs) were from Dr. Jeremy Reiter (University of California, San Francisco).³⁸ Primary Pkd1^RC/RC MEFs were generated following standard methods as previously reported.⁶¹ IMCD3^Flip-In cells and the immortalized Pkd2Halo Sstr3GFP murine renal epithelial cell lines with or without Inpp5e inactivated were obtained through the Polycystic Kidney Disease Research Resources Consortium. TransIT-X2 (Mirus Bio) or Lipofectamine RNAiMAX (Invitrogen) was used to transfect mammalian cells. IMCD3^Flp-In cells were cotransfected with pCAG-Flpo (Addgene 60662) and pcDNA5-mPC1 or pcDNA5-mPC2 and then treated with hygromycin to select positive single clones. The immortalized Pkd1^RC/flox murine collecting duct cell line was reported previously.⁶²

Fluorescence Microscopy

Standard immunofluorescence labeling protocol⁶³ was used, except for PC1 labeling described in Supplemental Methods. Immunofluorescence staining of PI(4,5)P₂ was performed following published methods.^37,38,64 PC2Halo was labeled using HaloTag TMRDirect ligand (Promega) following manufacture's instruction. Fluorescent images were acquired using Nikon Eclipse Ti2-E with NIS-Elements. Cilia length and fluorescence intensity were quantified using NIS-Elements.

Electrophysiology Analysis

Standard protocol described previously⁶⁵ was followed. Technical details are presented in Supplemental Methods.

In Vitro 3D Culture and Ex Vivo Cystogenesis

3D cystogenesis was conducted using IMCD3 cells as previously reported.⁶⁶ Cells were plated in 8-well chambered slides (10,000/well) and cultured for 5 days with daily medium change.

Metanephric kidneys from wild-type (WT) and Pkd1^RC/RC mice were dissected at E13.5 and cultured in 0.4 µm Transwell 6-well plates (353090, Corning) using DMEM/F-12 medium containing 10% FBS, 1% penicillin/streptomycin, 5 μg/ml transferrin, 2.8 nM selenium, 25 ng/ml PG E1, 3 ng/ml T3, and 5 μg/ml insulin. Kidneys were cultured for 5 days with daily medium change and imaged daily using Nikon Eclipse-Ci. Based on published studies,^67,68 each experiment used 2–3 female mice (3–5 embryonic kidneys/experimental group) and was repeated three times to ensure reproducibility. Studies using mice adhered to the National Institutes of Health Guide for the Care and Use of Laboratory Animals or the equivalent.

PC1 Glycosylation Analyses

Crude membrane proteins were isolated from cells following published methods.^23,66 The resulting membrane pellet was used for immunoprecipitation of PC1 with PC1 antibody (EB08670, Everest Biotech). Technical details are described in Supplemental Methods. The resulted agarose beads were eluted with 100 ml of 1× glycoprotein denaturing buffer New England Biolabs [NEB] at 80°C for 20 minutes. The beads were divided into three equal parts, and two of them were treated using either Endo H (P0702L, NEB) or PNGase F (P0704L, NEB) kits following the manufacturer's instructions. Then, all samples were subjected to immunoblotting using 3%–8% Tris-acetate gradient gels (EA0378BOX, Invitrogen) and PC1 antibody (sc-130554, Santa Cruz).

Statistical Analyses

Results are presented as mean plus SD. Statistical significance was assessed using GraphPad Prism 9.3.1. A two-tailed Student's t test was used for comparisons between two groups, and one-way ANOVA was performed for comparisons involving more than two groups.

Other experimental methods are provided in Supplemental Methods.

Results

INPP5E and PIPKIγ Played Opposite Roles in Regulating the Ciliary Level of PC2

In MEFs isolated from Inpp5e^−/− mice,³⁸ both the percentage of PC2-positive cilia and the amount of PC2 in cilia were >3-fold greater than in Inpp5e^+/− MEFs (Figure 1A). This increase of PC2 in Inpp5e^−/− cilia was suppressed by depleting PIPKIγ (Supplemental Figure 1B) that localized at the base of primary cilia^40,69 (Figure 1A). Consistent with previous reports,^37,38 Inpp5e^−/− cilia harbored much higher level of PI(4,5)P₂ along the shaft compared with Inpp5e^+/− cilia, which was suppressed by depletion of PIPKIγ (Supplemental Figure 1A). Interestingly, incubating Inpp5e^+/− MEFs with diC8-PI(4,5)P₂, the short chain, water soluble synthetic analogue of PI(4,5)P₂, also strongly boosted PC2 levels in cilia (Supplemental Figure 1C). Moreover, the ciliary level of PC2 proportionally increased in mouse embryonic fibroblast (MEF) cells carrying two, one, or zero Inpp5e alleles, exhibiting a negative correlation with the level of INPP5E (Figure 1B). Together, these results suggested that INPP5E and PIPKIγ might regulate ciliary levels of PC2 by coordinating PI(4,5)P₂ levels in the ciliary membrane.

Figure 1.

INPP5E and PIPKIγ play opposite roles in regulating the ciliary trafficking of PC2. (A) INPP5E and PIPKIγ coordinate the PC2 level in primary cilia. Inpp5e^−/− MEFs were transfected with control (siNC) or PIPKIγ-specific siRNA for 48 hours. (B) INPP5E dosage negatively correlates with PC2 levels in cilia. (C) INPP5E activity in cilia regulates the ciliary level of PC2. Inpp5e^−/− MEFs were transiently transfected with the empty pcDNA3 vector (Mock) or pcDNA3 expressing the Flag-tagged INPP5E variants, including the WT (INPP5E^WT), nonciliary C-terminus truncated INPP5E (INPP5E^ΔCT, aa 1-604), and the catalytic dead INPP5E (INPP5E^CD) for 48 hours. (D–G) Interrupting the ciliary entry of INPP5E increases the ciliary level of PC2. (D) RCTE cells transfected with control (siNC) or PDE6δ-specific (siPDE6δ) siRNAs for 48 hours followed by 24-hour serum starvation. (E) RCTE cells were treated with empty vehicle (0.1% DMSO) or Deltarasin (5 μM) in serum-free media for 24 hours. (F–G) Activation of PIPKIγ at the ciliary base increases the ciliary level of PC2. (F) RCTE cells were transiently transfected with empty vector (Mock) or PACT-fused HA-PIPKIγ variants, including the WT (PACT-HA-Iγ) or the kinase dead (PACT-HA-Iγ-KD) for 48 hours followed by 24-hour serum starvation. (G) RCTE cells were transiently transfected with the empty vector (Mock) or Flag-HYLS1 for 48 hours followed by 24-hour serum starvation. (A–G) Cells were serum-starved for 24 hours and subjected to indirect IF microscopy to visualize PC2 in primary cilia. Antibody against acetylated Ac-tub, the marker for the ciliary axoneme, was used to visualize the primary cilium. Scale bar, 2  μm. The percentage of PC2-positive primary cilia under each condition was calculated by evaluating 100–110 cilia in each experimental group. Data from three independent experiments were statistically analyzed using one-way ANOVA (A and F) or a two-tailed Student's t test (B–E and G), and results were plotted as mean±SD. Ac-tub, α-tubulin; CD, catalytically dead; ΔCT, nonciliary targeting; HA-PIPKIγ, HA-tagged PIPKIγ; HYLS1, hydrolethalus syndrome protein 1; IF, immunofluorescence; MEF, mouse embryonic fibroblast; PACT, pericentrin centrosomal targeting domain; PACT-HA-Iγ, PACT-fused, HA-tagged PIPKIγ; PACT-HA-Iγ-KD, PACT-fused, HA-tagged PIPKIγ kinase dead; PC2, polycystin-2; PIPKIγ, type Iγ phosphatidylinositol-4-phosphate 5-kinase; RCTE, renal cortical tubule epithelial; siNC, siRNA nonspecific control; siRNA, small interfering RNA; WT, wild-type.

To test this possibility, we re-expressed Flag-tagged INPP5E in Inpp5e^−/− MEFs, including the WT, nonciliary targeting (ΔCT, aa 1-604),⁷⁰ and catalytically dead (CD, D512/515W)⁴¹ variants (Supplemental Figure 1D). As shown in Figure 1C, INPP5E^WT expressed in Inpp5e^−/− cells significantly suppressed the ciliary level of PC2, which was not achieved by expressing comparable levels of INPP5E^ΔCT or INPP5E^CD. Moreover, the overall cellular PC2 protein level was not affected by the INPP5E variants (Supplemental Figure 1D).

To confirm that INPP5E activity in primary cilia suppressed PC2 accumulation in cilia, we switched to the human renal epithelial cell line RCTE.⁷¹ First, RNAi-mediated decrease of INPP5E led to an approximately two-fold increase in the ciliary PC2 level in RCTE cells (Supplemental Figure 1E), recapitulating the results observed in Inpp5e^−/− MEFs. Next, depletion of PDE6δ, which disrupted the ciliary targeting of INPP5E⁷² (Supplemental Figure 1F), significantly increased PC2 levels in RCTE cilia (Figure 1D) without affecting the overall PC2 levels (Supplemental Figure 1F). Similarly, when RCTE cells were treated with Deltarasin, which prevents INPP5E's ciliary entry by blocking INPP5E-PDE6δ binding⁷³ (Supplemental Figure 1G, left panel), ciliary PC2 was also notably elevated (Figure 1E), with the cellular INPP5E and PC2 levels not affected (Supplemental Figure 1G, right panel). Furthermore, using mouse renal epithelial cell lines expressing PC2Halo,⁷⁴ we observed the same phenomena that inactivating Inpp5e (Supplemental Figure 1I), inhibiting INPP5E, or adding diC8-PI(4,5)P₂ increases the number of PC2Halo-positive cilia (Supplemental Figure 1H).

Collectively, our results suggested that INPP5E suppressed the ciliary level of PC2 by maintaining a PI(4,5)P₂-low environment in cilia. This was consistent with increasing ciliary PI(4,5)P₂ facilitating the ciliary entry of TULP3 and the intraflagella transport complex A complex,⁵⁹ which are necessary for the ciliary trafficking of PC2.^57,58 To further test this possibility, we overexpressed a pericentrin centrosomal targeting domain (PACT)-fused, HA-tagged PIPKIγ, which primarily targets to the basal body of primary cilia with minimal diffusing cytoplasmic expression.⁴⁰ As shown in Figure 1F, expression of WT PIPKIγ, but not the kinase-dead mutant, at the basal body in RCTE cells increased PC2 levels in cilia without changing the overall PC2 levels (Supplemental Figure 1H). In addition, overexpression of hydrolethalus syndrome protein 1 (HYLS1), the PIPKIγ activator at the basal body,⁶³ also increased PC2 levels in cilia (Figure 1G). Together, these results consistently showed that suppression of INPP5E and activation of PIPKIγ, both increasing PI(4,5)P₂ in cilia, enhanced the ciliary trafficking of PC2 in normal renal epithelial cells.

Suppression of INPP5E or Activation of PIPKIγ Increased Ciliary PC2 in Cells Carrying ADPKD Mutations

Studies using in vitro and in vivo models have concluded that altered ciliary trafficking of PCs underlies PKD pathogenicity in some cases.^34,75 The minor ADPKD gene, GANAB, which encodes the α subunit of glucosidase 2, has been implicated in PC1 and PC2 glycosylation, maturation, and trafficking.³⁰ Indeed, compared with parental RCTE cells where PC2 was visible in approximately 35% of cilia, only 8% of cilia in GANAB^−/− RCTE cells showed PC2 signal (Figure 2A). Depletion of INPP5E (Figure 2B) or PDE6δ (Figure 2C) in GANAB^−/− RCTE cells significantly increased the PC2 trafficking into cilia without affecting total PC2 levels (Supplemental Figure 2A). Overexpression of PACT-PIPKIγ (Figure 2D) or the PIPKIγ activator HYLS1 (Figure 2E), which both localized to the ciliary base and augmented PIPKIγ activity in situ, also led to the accumulation of PC2 in primary cilia in GANAB^−/− cells. We also found that PC1 levels in primary cilia were significantly diminished in GANAB^−/− RCTE cells (Supplemental Figure 2B), which was considerably improved when INPP5E was depleted (Figure 2F).

Figure 2.

Suppression of INPP5E or activation of PIPKIγ increased the ciliary levels of PCs in GANAB-null RCTE cells. (A) Loss of GANAB inhibits PC2 levels in cilia. Parental RCTE cells (GANAB^+/+) and GANAB^−/− RCTE cells were serum starved for 24 hours and then subjected to IF microscopy to visualize PC2 in cilia (labeled by Ac-tub). (B–E) Suppressing INPP5E (B and C) or activating PIPKIγ (D and E) recovers PC2 in cilia of GANAB^−/− cells. GANAB^−/− RCTE cells were transfected with siRNAs (B, siNC or siINPP5E; C, siNC or siPDE6δ) or protein-expressing plasmids (D, empty vector or PACT-HA-PIPKIγ; E, empty vector or Flag-HYLS1) for 48 hours, serum starved for 24 hours, and then subjected to IF microscopy to visualize PC2. (F) Loss of INPP5E increases PC1 in GANAB^−/− cilia. GANAB^−/− RCTE cells were transfected with siNC or siINPP5E for 48 hours, serum starved for 24 hours, and then subjected to IF microscopy to visualize PC1. Primary cilia were labeled by Ac-tub. Scale bar, 2  μm. The percentage of PC2-positive or PC1-positive cilia and the mean fluorescence intensity of PC2 or PC1 were quantified in 100–110 cilia per experimental group. Results from three independent experiments were statistically analyzed using a two-tailed Student's t test and plotted as mean±SD. PACT-HA-PIPKIγ, PACT-fused, HA-tagged PIPKIγ; PC1, polycystin-1.

Next, we examined PKD1 p.R3277C that encodes a temperature-sensitive folding mutant (PC1^RC), which reduces the level of mature PC1, yielding a functionally hypomorphic variant of PC1.²³ To determine whether the ciliary levels of PCs can be affected by PIPKIγ or INPP5E, we used MEFs isolated from Pkd1^RC/RC mice, which develop adult-onset progressive PKD, similar to human patients.²³ Compared with MEFs isolated from WT mice (Pkd1^+/+), Pkd1^RC/RC MEFs showed significantly lower PC2 localization and intensity in primary cilia (Figure 3A). When INPP5E was depleted (Figure 3B) or its ciliary targeting was interrupted by an PDE6δ inhibitor (Deltarasin, Figure 3C), PC2 targeting to and levels in cilia were increased significantly in Pkd1^RC/RC MEFs. When we overexpressed PACT-PIPKIγ (Figure 3D) or HYLS1 (Figure 3E) to enhance PIPKIγ activity at the ciliary base in Pkd1^RC/RC MEFs, both conditions elevated the ciliary level of PC2. Moreover, the mouse PC1^RC level in cilia of Pkd1^RC/RC MEFs was increased when INPP5E was depleted (Figure 3F). Depletion of INPP5E or overexpression of PACT-PIPKIγ or HYLS1 had no effect on the overall protein level of PC2 in Pkd1^RC/RC MEFs, however (Supplemental Figure 2, C–E). Together, our results showed that suppressing INPP5E or enhancing PIPKIγ activity in cilia might increase the ciliary accumulation of PC1 and PC2, not only in their WT forms but also for PC1^RC, a hypomorphic, folding mutant form of PC1.

Figure 3.

Suppression of INPP5E or activation of PIPKIγ recovered the ciliary levels of the PCs in MEFs isolated from Pkd1^RC/RC mice. (A) PC2 levels in Pkd1^RC cilia are decreased compared with WT cilia (Pkd1⁺). MEF cells isolated from WT mice (Pkd1^+/+) or Pkd1^RC/RC mice were serum starved for 24 hours and then subjected to IF microscopy to visualize PC2 in cilia (labeled by Ac-tub). (B–F) Suppressing INPP5E (B and C) or activating PIPKIγ (D and E) recovers PC2 in Pkd1^RC cilia. Pkd1^RC/RC MEFs were transfected with siRNAs for 48 hours (B, siNC or siINPP5E), treated with DMSO or Deltarasin for 24 hours (C), or transfected with indicated plasmids for 48 hours (D, empty vector or PACT-HA-PIPKIγ; E, empty vector or Flag-HYLS1). Cells were then serum starved for 24 hours and subjected to IF microscopy to visualize PC2. (F) Loss of INPP5E increases the level of PC1^RC in cilia. Pkd1^RC/RC MEFs were transfected with siNC or siINPP5E for 48 hours, serum starved for 24 hours, and then subjected to IF microscopy to visualize PC1. Primary cilia were labeled by Ac-tub. Scale bar, 2  μm. The percentage of PC2-positive or PC1-positive cilia and the mean fluorescence intensity of PC2 or PC1 were quantified in 100–110 cilia per experimental group. Results from three independent experiments were statistically analyzed using a two-tailed Student's t test and plotted as mean±SD.

INPP5E Inhibition Increased the Ciliary Levels of PC1 and PC2 in ADPKD Cells

To further examine our genetic findings, we tested whether small molecule compounds inhibiting INPP5E activity could increase polycystin (PC) levels in primary cilia in the same cellular systems. Pirruccello et al. reported several small molecules that could inhibit the activity of various inositol 5-phosphatases to different extents, although none of them are INPP5E-specific.⁷⁶ Among the small molecules they characterized, YU144369 inhibits INPP5E and other PI 5-phosphatases (including OCRL1, INPP5B, SHIP1, and SYNJ1) with similar IC50; whereas YU142670 is a potent inhibitor for OCRL1 and INPP5B, but not INPP5E.⁷⁶ SHIP1 and SYNJ1 are predominantly expressed in hematopoietic cells and neurons, respectively,⁴⁶ while OCRL1 and INPP5B share similar substrate/product specificity as INPP5E and also localize to cilia.⁷⁷ Thus, we treated the GANAB^−/− RCTE cells with YU144369 and YU142670 to distinguish the effects caused by inhibiting INPP5E from inhibiting OCRL1/INPP5B, respectively. Interestingly, YU144369 (inhibiting INPP5E, OCRL1, and INPP5B) increased PC2 levels in GANAB^−/− cilia in a dose-dependent manner, whereas YU142670 (inhibiting OCRL1 and INPP5B) had no such effect (Figure 4A), confirming that the suppression of INPP5E activity specifically enhances ciliary trafficking of PC2. In alignment with this, YU144369 increased PI(4,5)P₂ in cilia of RCTE cells, so did Deltarasin that inhibits INPP5E entering cilia (Supplemental Figure 3A). Consistent with the siRNA-mediated INPP5E depletion (Figure 2E), YU144369 treatment also increased PC1 levels in cilia of GANAB^−/− RCTE cells (Figure 4B). However, we did not observe a meaningful global improvement of PC1 maturation in GANAB^−/− RCTE cells by inhibiting INPP5E activity (Figure 4C and Supplemental Figure 3B).

Figure 4.

Inhibition of INPP5E activity increased the ciliary levels of PCs in ADPKD cells. (A–B) Inhibiting INPP5E activity increased the level of PCs in GANAB^−/− RCTE cells. Cells were treated with DMSO, YU142670 (84 μM), or YU144369 (23 μM and 69 μM) for 24 hours in serum-free medium and then subjected to indirect IF microscopy to analyze the ciliary PC2 (A) or PC1 (B). (C) INPP5E inhibition had no effect on the glycosylation of PC1 in GANAB^−/− RCTE cells. The membrane fraction was prepared from parental RCTE cells treated with DMSO or GANAB^−/− RCTE cells treated with DMSO or YU144369 (69 μM). With or without (Un) the deglycosylation treatment using endoglycosidase H (+E) or PNGase F (+P), membrane fractions were subjected to IP using the C-terminal PC1 antibody (1:50). The precipitates were analyzed by IB using the N-terminal PC1 antibody (1:100). GANAB^−/− cells exhibited a complete loss of mature PC1 (NTR, Endo H-resistant), with more abundant FL and NTS PC1, which was not affected by YU144369 treatment. (D–E) Inhibiting INPP5E activity increased the level of PC1^RC and PC2 in Pkd1^RC/RC MEFs. Pkd1^RC/RC MEFs were treated with DMSO, YU144369 (23 μM or 69 μM), or YU142670 (84 μM) for 24 hours in serum-free medium. Cells were then subjected to indirect IF microscopy with indicated antibodies to visualize ciliary PC2 (D) or PC1 (E). (A, B, D, and E) Ac-tub, acetylated Ac-tub. Scale bar, 2  μm. The percentage of PC2-positive or PC1-positive cilia was quantified in 100–110 cilia per experimental condition. Results from three independent experiments were statistically analyzed using one-way ANOVA and plotted as mean±SD. ADPKD, autosomal dominant polycystic kidney disease; FL, full-length; IB, immunoblotting; IP, immunoprecipitation; NTR, N-terminal glycoform; NTS, N-terminal Endo H-sensitive.

In addition to increasing the ciliary levels of WT PC1 and PC2 with reduced glycosylation in GANAB^−/− RCTE cells, YU144369 also increased the ciliary levels of PC1^RC and PC2 in Pkd1^RC/RC MEFs (Figure 4, D and E), further supporting the concept that small molecule inhibition of INPP5E promotes ciliary trafficking of PCs in both WT and ADPKD variant (trafficking perturbed) contexts. These experiments were repeated using Pkd1^RC/flox renal epithelial cells⁶² and the derived Pkd1^RC/− cells by exogenously expressing Cre. PC2 levels in Pkd1^RC/− renal epithelial cilia were lower compared with the control Pkd1^RC/flox cilia (Supplemental Figure 3C) and similar to our observations in MEFs. YU144369 and Deltarasin both significantly increased the ciliary level of PC2 in Pkd1^RC/flox and Pkd1^RC/− renal epithelial cells, with a greater increase of PC2 in Pkd1^RC/− cilia (Supplemental Figure 3C).

To determine whether other ciliary trafficking mutations in PKD1 or PKD2 could be ameliorated with INPP5E inhibition, we established mouse renal epithelial IMCD3^Flp-In cell lines to test the exogenous expression and ciliary targeting of PC variants. Epitope tag antibodies detected strong Flag-PC1^WT and HA-PC2^WT expression in primary cilia, which were further increased by YU144369 treatment (Supplemental Figure 4, A and B), verifying the reliability of this system. The pathogenic PC mutants we examined included mouse PC1^RC (Pkd1^RC/RC, PKD1 p.Arg3277Cys),²³ PC1^TV (Pkd1^V/V, mPC1^T3041V),^78,79 PC2^WG (PKD2 p.Trp414Gly),⁸⁰ PC2^EG (Pkd2^lrm4/lrm4, mPC1^E442G),³⁴ and a newly identified VUS PC2^LP (PKD2 p.Leu884Pro, mPC2^L882P). Consistently, YU144369 increased the number of PC1^RC-positive cilia from approximately 30% PC1^WT levels (approximately 18%) to almost 60% (approximately 32%; Supplemental Figure 4A). However, PC1^TV remained absent from primary cilia after YU144369 treatment, likewise for PC2^WG and PC2^EG (Supplemental Figure 4, A and B), although these variants were expressed at a similar level (Supplemental Figure 4, C and D). Interestingly, YU144369 increased the PC2^LP-positive cilia from none to almost 20% (Supplemental Figure 4B). Thus, INPP5E inhibition might improve the ciliary targeting of some hypomorphic variants, but not all pathogenic missense changes.

PC1^RC maintained ion channel function within the PC1/PC2 complex expressed in Xenopus oocytes.

The correlation of PC dosage with disease severity and progression of ADPKD³ raised a question of whether increasing PC levels in cilia may compensate for the pathogenic defects leading to cystogenesis in the kidneys. Because the hypomorphic PC1^RC is a well-characterized pathogenic variant with a mouse model available, we conducted in vitro, ex vivo, and in vivo studies to investigate the functionality of the PC1^RC/PC2 complex.

We recently generated a gain-of-function (GOF) mutant of the PC1/PC2 ion channel complex by introducing the L677A/N681A double mutation into PC2 (PC2^AA).⁶⁵ We have also showed that the C-terminal eleven-transmembrane fragment of PC1, resulting from G protein–coupled receptor proteolysis site cleavage, is adequate to form a complex with PC2 and conducts cation channel function.⁶⁵ Thus, we coexpressed PC1-CTF^RC with PC2^AA in Xenopus Laevis oocytes and compared its channel activity with the PC1-CTF/PC2^AA heteromeric channel and the PC2^AA homomeric channel. Immunoblotting confirmed robust expression of both PC1-CTF/PC2^AA and PC1-CTF^RC/PC2^AA in Xenopus oocytes (Supplemental Figure 5). In a bath solution containing 100 mM Na⁺ without (Figure 5A) or with (Figure 5B) 2 mM Ca²⁺, oocytes expressing PC1-CTF^RC/PC2^AA produced currents nearly identical to those expressing the PC1-CTF/PC2^AA complex. It was reported that extracellular Ca²⁺ blocks the inward currents of PC2 homomeric GOF channels but not PC1/PC2 heteromeric GOF channels.^65,81 We observed the same effect of Ca²⁺ on the inward current of our PC2^AA homomeric channel and not the PC1-CTF/PC2^AA or PC1-CTF^RC/PC2^AA heteromeric channels (Figure 5B), highlighting that currents recorded in these oocytes were mediated by the PC1/PC2 heteromeric complex.

Figure 5.

PC1^RC formed a complex with PC2 and exhibited comparable in vitro channel activity with the WT PC1. (A–B) Representative I–V curves (left) and scatter plots and bar graphs (right) show the currents from oocytes expressing PC2^AA alone (green), PC1-CTF with PC2^AA (red), PC1-CTF^RC with PC2^AA (blue), and from un-injected oocytes (black). Currents were recorded in bath solutions containing 100 mM Na⁺ (A) or 100 mM Na⁺ and 2 mM Ca²⁺ (B). Scatter plots and bar graphs illustrate currents between +100 mV and −100 mV. (C) Representative I–V curves of the PC2_AA (green), PC1-CTF/PC2^AA (red), and PC1-CTF^RC/PC2^AA (blue) channels in divalent ion-free bath solutions containing 100 mM of Na⁺ (atomic mass 23), DMA⁺ (MW 45.1), or NMDG⁺ (MW 195.2). (D) Scatter plots and bar graphs show the reversal potentials of currents recorded in (C). The currents of the PC2 and PC1/PC2 channels exhibit significantly different RP in DMA⁺ solution. (A–D) Oocyte numbers are indicated in parentheses. Data are presented as mean±SD, and currents were compared with the Student's t test. DMA⁺, dimethylamine⁺; NMDG⁺, N-Methyl-D-glutamine⁺; PC1-CTF, C-terminal eleven-transmembrane fragment of PC1; RP, reverse potential.

In addition to current amplitude, ion channel function is also determined by its ion permeability, which is governed by the conformation of the ion-conducting pore, especially the ion selectivity filter within it. To evaluate the permeability of our channel complexes, we used three different cations including a small cation, Na⁺, an intermediate-sized cation, dimethylamine⁺, and a large cation, N-Methyl-D-glutamine⁺. Although all three complexes exhibited similar permeability to the smallest and largest cations (Figure 5, C and D), the permeability of the intermediate-sized dimethylamine⁺ differed significantly between the homomeric PC2 and the heteromeric PC1/PC2 channels, supporting a unique contribution of PC1 in the permeability of the PC complex channel. Comparing with the PC2^AA homomeric channel showed near impermeability to this ion, as indicated by the highly negative reversal potential, −70 mV, both PC1-CTF/PC2^AA and PC1-CTF^RC/PC2^AA channels demonstrated substantial, comparable permeability (reversal potential >+5 mV; Figure 5, C and D). Together, our data indicated that the PC1^RC mutation had little effect on the activity and ion permeability of the PC1/PC2 channel complex and the functionality of PC1^RC, once appropriately localized, most likely remained comparable with its WT counterpart. Thus, it is plausible that increasing the levels of PC1 or PC1^RC in primary cilia can recover sufficient function of the PC complex to reduce cystogenesis in the kidneys.

INPP5E Inhibition Suppressed Cystogenesis in the Kidneys

To test this possibility, we first used an in vitro cystogenesis model using IMCD3 cells. A recent study showed that activating a cAMP signalosome specifically in primary cilia, but not in the cytoplasm, efficiently transformed the morphology of IMCD3 cells grown in 3D Matrigel from tubules into cysts,⁸² strengthening the physiological relevance of this in vitro cystogenesis model. After confirming that the treatment of YU144369 (INPP5E inhibitor) led to a substantial increase of PC1 in primary cilia in IMCD3^Flp-In cells (Supplemental Figure 6A), we performed the 3D Matrigel cystogenesis assay with or without various concentrations of YU144369. As shown in Figure 6A, YU144369 significantly reduced the expansion of IMCD3 cystic spheroids induced by forskolin in a dose-dependent manner, whereas the total number of IMCD3 spheroids was not affected by YU144369 (Supplemental Figure 6B). Similarly, the PDE6δ inhibitor Deltarasin reduced the size of IMCD3 cysts (Supplemental Figure 6C). In addition to PI(4,5)P₂, INPP5E can also dephosphorylate PI(3,4,5)P₃ in vitro⁷⁶ and thus may suppress the protein kinase B (AKT)-dependent cell proliferation/survival.⁵⁴ However, YU144369 treatment did not change AKT activity (Supplemental Figure 6D) or the overall growth of IMCD3 cells cultured on 2D surface (Supplemental Figure 6E). Inpp5e-null IMCD3 cells (Supplemental Figure 6F) also showed similar levels of AKT activity and cell proliferation compared with parental cells (Supplemental Figure 6, G and H), although exhibiting significantly increased PC2 in cilia (Supplemental Figure 6I). Indeed, Inpp5e-null IMCD3 cysts were less responsive to forskolin stimulation (Supplemental Figure 6J), consistent with the effect of YU144369 on inhibiting INPP5E activity. Moreover, the level of AKT activity in MEF cells did not correlate with Inpp5e expression (Supplemental Figure 6K) or cell proliferation (Supplemental Figure 6L). It was reported that the activation of hedgehog signaling by smoothened agonist (SAG) increased the ciliary PI(4,5)P₂ and PC2 in Inpp5e-null MEFs, likely by disrupting the transition zone structure.⁸³ However, we did not observe such effects of SAG on the transition zone protein TCTN1 and PC2 levels in Inpp5e-null cilia (Supplemental Figure 7, A–C). This inconsistency could be attributed to cell type differences because our results were obtained in IMCD3 cells and MEF cells from a different mouse line.³⁸ Consequently, SAG did not suppress the forskolin-stimulated cystogenesis of IMCD3 cells cultured in 3D Matrigel (Supplemental Figure 7D). Owing to the inconsistent performance of PI(4,5)P₂ antibodies from different lots and the overwhelming plasma membrane signal detected by the antibody in mouse cells, we were unable to draw a definitive conclusion regarding the increase of ciliary PI(4,5)P₂ levels by SAG⁸³ in our Inpp5e-null mouse cells. Nevertheless, the overall PI(4,5)P₂ levels appeared unchanged before and after SAG treatment.

Figure 6.

INPP5E inhibition suppressed cystogenesis in the kidneys. (A) YU144369 inhibited the cyst formation of IMCD3 cells cultured in 3D Matrigel in a dose-dependent manner. Twenty-four hours after seeding cells in Matrigel, the cells are treated with empty vehicle (0.1% DMSO) or YU144369 (23, 69, or 116 μM) combined with 10 μM Forskolin for 5 days. Samples were examined under microscope (4× lens) daily, and representative images are shown. Scale bar, 50 μm. The average diameter of cysts in each group was calculated (n=100–110, from five to seven randomly selected fields) and normalized against the average cyst diameter of the DMSO group on day 1. Results from three independent experiments were statistically analyzed and plotted as mean±SD. (B–C) YU144369 suppressed cystogenesis in the kidneys induced by forskolin in WT mouse embryonic kidneys (B), without causing negative effects on normal kidney growth (C). (D–E) YU144369 inhibited the ex vivo cystogenesis in Pkd1^RC/RC embryonic kidneys. (B–E) Embryonic kidneys were isolated from WT or Pkd1^RC/RC C57BL/6 embryos (E13.5) and cultured in transparent Transwells (0.4 μm pore size). Kidneys were treated with DMSO or YU144369 (23 μM and 69 μM) combined with (B) and (D) or without (C) and (E) 10 μM Forskolin for 1–6 days. Kidneys were imaged every day during culture. Scale bar, 400 μm. The total and cystic areas of kidneys were analyzed using Adobe Photoshop and Image-Pro, and the cyst index was calculated as the ratio of cystic area and total area. Each experiment used 3–5 embryonic kidneys per experimental group from two to three pregnant female mice, and three independent experiments were conducted for each study. Results were statistically analyzed using the Student's t test (A–C, E) or one-way ANOVA (D) and plotted as mean±SD. (F) A working model that summarizes our observation: Increasing PI(4,5)P₂ in primary cilia, for example, by inhibiting INPP5E in a dosage-dependent manner, enhances the ciliary trafficking of WT PCs and trafficking-deficient variant like PC1-R3277C, leading to the recovery of PC functions that weaken the PKD-pathogenic signals. IMCD3, inner medullary collecting duct epithelial cell line 3; PKD, polycystic kidney disease.

Together, our results indicated that without a cystogenesis signal, such as forskolin-induced increase of cAMP or a mutation to the PC-complex, INPP5E inhibition did not affect global AKT activity, and likely had no adverse effect on normal cell growth. Based on our previous data, YU144369-mediated INPP5E inhibition plausibly counteracted the cystogenesis effect by increasing the level of functional PCs in cilia. To test this possibility, we examined the effect of INPP5E inhibition on cystogenesis in the kidneys using an ex vivo organ culture model. When cultured in forskolin-containing medium, embryonic kidneys from WT mice developed mild cysts in 5 days, which was inhibited by YU144369 (Figure 6B), consistent with what we observed with the in vitro 3D cystogenesis model using IMCD3 cells (Figure 6A). WT mouse kidneys did not form cysts in forskolin-free medium and grew at a comparable rate with or without YU144369 (Figure 6C), indicating that YU144369 did not disturb normal kidney development. In embryonic kidneys from Pkd1^RC/RC mice, forskolin induced much more aggressive cystogenesis, and YU144369 treatment yielded a dose-dependent, significant suppression of cyst progression (Figure 6D). Importantly, we constantly observed forskolin-independent cyst formation in Pkd1^RC/RC kidneys by day 5, and this spontaneous cystogenesis caused by PC1^RC was also suppressed by YU144369 (Figure 6E). Considering our previous data that inhibition of INPP5E increased the levels of PC1^RC and PC2 in primary cilia and the PC1^RC/PC2 complex retained channel function comparable with the WT PC complex, these results suggested that restoring the level of WT and some hypomorphic PC variants in primary cilia might represent a therapeutic approach to ameliorate or slow PKD progression.

Discussion

Recent studies showed that re-expression of PC1 or PC2 in patient cells or mice carrying Pkd1 or Pkd2 mutations suppressed cystogenesis in the kidneys or even reversed the disease.^84–86 Similar therapeutic strategies have been successfully applied to cystic fibrosis caused by mutations in the CFTR gene to potentiate channel function, correct membrane targeting, or enhance translation, which are dampened by corresponding pathogenic CFTR variants.^87–89 Similarly, genotype-specific targeted therapies aiming to correct/improve the functionality of specific PC1 or PC2 variants could represent promising precision medicine. About 30%–35% of PKD1 (7%–10% for PKD2) pathogenic alleles are predicted to be nontruncating, in many cases resulting in partial loss-of-function, hypomorphic proteins,³³ which often show defects in ciliary trafficking.⁷⁵ Regardless of the mutation, most ADPKD kidney cells likely have one normal PKD1 (or PKD2) allele, at least at the early stage of the disease. In this context, raising the level of PCs in primary cilia, either the WT and/or hypomorphic, trafficking defective variants, could make up for the primary pathogenic functional loss and thus delay the initiation and slow or even reverse the progression of ADPKD.

Aligning with this concept, we provided an example of increasing PCs' levels in cilia by elevating ciliary PI(4,5)P₂ through manipulating the activity of INPP5E or PIPKIγ. WT PCs could be increased in cilia by suppressing INPP5E or activating PIPKIγ in normal or GANAB-null RCTE cells. We did not observe PC1 glycosylation improved by suppressing INPP5E, likely because the assay we used is not sensitive enough to detect a small increase of mature PC1. Another possibility is that INPP5E has no effect on PC maturation/folding that occurs in the endoplasmic reticulum/Golgi compartments; instead, it maintains a PI(4,5)P₂-low environment in cilia that is necessary for a tight gating mechanism to repel the immature PCs. Notably, we demonstrated that PC1^RC retained normal abilities to form a complex with PC2 and mediated cation transport in vitro, suggesting that the functional reduction of PC1^RC is rooted from its defective ciliary trafficking. Correspondingly, YU144369 that brought the ciliary PC1^RC level back to approximately 60% of WT levels, which should be above the proposed pathogenic threshold,²³ indeed delayed both forskolin-induced and spontaneous cystogenesis in Pkd1^RC embryonic kidneys ex vivo.

Yet, the efficacy of INPP5E inhibitor to boost the ciliary level of PC variants varied significantly. Among the tested nonresponsive variants, the PKD1 p.Thr3041Val completely disrupts the G protein–coupled receptor proteolysis site cleavage and maturation and ciliary targeting of PC1⁹⁰; whereas PKD2 p.Trp414Gly (encoding human PC2^W414G) and Pkd2^lrm4 (encoding mouse PC2^E442G) carry mutations at the top surface of the tetragonal opening for polycystins domain that are important for stabilizing the structure and supporting the formation of PC2 homotetramers or PC1/PC2 heterotetramers.⁹¹ Thus, these variants are likely not hypomorphic and possibly result in severely misassembled PC complexes, whose defective ciliary trafficking may be insensitive to correction by increasing PI(4,5)P₂ in cilia. Future studies are needed to more broadly understand which PC1 and PC2 variants can be rescued regarding ciliary trafficking.

Although the regulatory circuit of INPP5E suppressing the trafficking of TULP3-dependent ciliary cargos (including PCs) through constraining ciliary PI(4,5)P₂ has been extensively studied,^37,38,57,59 the current literature suggests that this INPP5E–TULP3 axis plays a complex role in ciliary trafficking and kidney development and maintenance. Loss of INPP5E that triggers TULP3 increase in cilia causes cystic kidney,^41,42,54,92 but inactivating Tulp3 in embryonic kidneys also leads to cystogenesis^57,58; and interestingly, inactivating Tulp3 in adult Pkd1^RC/RC kidneys slows cyst formation.⁵⁸ These results reflect different roles of these proteins during development compared with later, suggesting a crosstalk of them with the cilia-dependent cyst activation mechanism.^93–95 In this context, the positive effect of decreasing INPP5E on PC2 trafficking and suppression of cystogenesis that we observed in vitro could be due to the extent and timing of suppressing INPP5E in cilia. Other paradoxical data are that although loss of inpp5e causes kidney cysts in zebrafish,⁹⁶ activating the fish PIPKIγ paralogue suppresses kidney cysts in inpp5e-depleted fish by maintaining the level of apical PI(4,5)P₂ and promoting ciliogenesis.⁹⁷ Under physiological conditions, PIPKIγ and INPP5E activities likely need to be fine-tuned in a temporal manner to coordinate the delicate PI(4,5)P₂ dynamics for boosting the ciliary dosage of PCs, while maintaining normality of epithelial polarity, ciliogenesis, and other signaling pathways harbored by the primary cilium.

Our study provided an example supporting the therapeutic potential of increasing the ciliary level of PCs for PKD treatment (Figure 6F). To treat lifelong genetic diseases, therapies directly targeting the underlying defective proteins often yield better efficacy with less side effects, than managing the consequences downstream of mutated proteins. Future research efforts are necessary to develop genotype-specific targeted therapies to improve the functionality of PCs, including increasing their ciliary trafficking.

Supplementary Material

jasn-37-0944-s001.pdf ^{(1.4MB, pdf)}

jasn-37-0944-s002.pdf ^{(1.3MB, pdf)}

jasn-37-0944-s003.pdf ^{(2MB, pdf)}

Disclosures

Disclosure forms, as provided by each author, are available with the online version of the article at http://links.lww.com/JSN/F548.

Author Contributions

Conceptualization: Peter C. Harris, Jinghua Hu, Kun Ling, Yong Yu.

Data curation: Ariana Aghevli, Chuan Chen, Yue Gao, Biyun Ji, Zhifei Wang, Qingwen Xu.

Formal analysis: Ariana Aghevli, Chuan Chen, Chunhua Chen, Madilyn R. Ellis, Yue Gao, Zhifei Wang, Qingwen Xu.

Funding acquisition: Peter C. Harris, Jinghua Hu, Kun Ling, Yong Yu.

Investigation: Chuan Chen, Yue Gao, Zhifei Wang, Qingwen Xu.

Methodology: Ariana Aghevli, Chuan Chen, Chunhua Chen, Mandy Cryole, Madilyn R. Ellis, Courtney J. Haycraft, Kai He, Yan Huang, Jielu H. Robinchaud, Cynthia J. Sieben, Zhifei Wang, Qingwen Xu.

Project administration: Peter C. Harris, Jinghua Hu, Kun Ling, Yong Yu.

Resources: Chunhua Chen, Mandy Cryole, Madilyn R. Ellis, Courtney J. Haycraft, Kai He, Jinghua Hu, Yan Huang, Biyun Ji, Kun Ling, Cynthia J. Sieben, Bradley K. Yoder, Yong Yu.

Supervision: Peter C. Harris, Kun Ling, Yong Yu.

Validation: Chuan Chen, Madilyn R. Ellis, Zhifei Wang, Qingwen Xu.

Visualization: Chuan Chen, Chunhua Chen, Madilyn R. Ellis, Qingwen Xu.

Writing – original draft: Chuan Chen, Yue Gao, Kun Ling, Zhifei Wang, Yong Yu.

Writing – review & editing: Chuan Chen, Peter C. Harris, Jinghua Hu, Kun Ling, Cynthia J. Sieben, Zhifei Wang, Yong Yu.

Funding

K. Ling: U.S. Department of Defense (W81XWH2010214, HT94252410154), National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK, R01DK139607), and Pilot and Feasibility sub-awards from Mayo Clinic Translational PKD Center (P30DK90728). P.C. Harris: U.S. Department of Defense (HT94252410155, W81XWH2010215) and NIDDK (R01DK58816, R01DK139607). J. Hu: NIDDK (R01DK090038, R01DK099160) and Pilot and Feasibility sub-awards from Mayo Clinic Translational PKD Center (P30DK90728). Y. Yu: NIDDK (R01DK125404) and U54 Center Grant to Maryland PKD Research and Translation Core Center (U54DK126114). C. Chen: U.S. Department of Defense (W81XWH2010214) and Pilot and Feasibility sub-awards from Mayo Clinic Translational PKD Center (P30DK90728). C. Chen: U.S. Department of Defense (W81XWH2010214) and Pilot and Feasibility sub-awards from Mayo Clinic Translational PKD Center (P30DK90728). M.R. Ellis: U.S. Department of Defense (W81XWH2010214). Y. Huang: U.S. Department of Defense (W81XWH2010214), NIDDK (R01DK090038, R01DK099160), and Pilot and Feasibility sub-awards from Mayo Clinic Translational PKD Center (P30DK90728). C.J. Sieben: NIDDK (R01DK58816). Z. Wang: NIDDK (R01DK125404). K. He: Pilot and Feasibility sub-awards from Mayo Clinic Translational PKD Center (P30DK90728) and NIDDK (R01DK090038, R01DK099160). Y. Gao and B. Ji: U.S. Department of Defense (HT94252410154) and NIDDK (R01DK139607). This study was supported by Baltimore PKD Center (2P30DK090868).

Declarative Statements

All animal experiments were conducted in accordance with the NIH Guide for the Care and Use of Laboratory Animals or an equivalent standard that meets or exceeds the ethical and welfare requirements outlined in the NIH Guide. All protocols were approved by the appropriate institutional animal care and use committee.

Data Availability Statements

Original data generated for the study will be made available upon reasonable request to the corresponding author. Data Type: Published Material; Image Data; Raw Data/Source Data; Observational Data; Research Protocols. Reason for Restricted Access: Data generated from this study are low-throughput, thus are saved in institutional-based and lab-based servers. All published data, materials, and reagents can be accessed and shared upon reasonable request aligning with National Institutes of Health and institutional policies. All data generated and analyzed during this study are available from the corresponding author upon reasonable request.

Supplemental Material

This article contains the following supplemental material online at http://links.lww.com/JSN/F549.

Supplemental Methods

Supplemental Figure 1. Supporting data for Figure 1.

Supplemental Figure 2. Supporting data for Figure 2 and Figure 3.

Supplemental Figure 3. Supporting data for Figure 4.

Supplemental Figure 4. INPP5E inhibitor exhibited different affects on PC1 and PC2 variants.

Supplemental Figure 5. Supporting data for Figure 5.

Supplemental Figure 6. Supporting data for Figure 6.

Supplemental Figure 7. Smoothened agonist-induced activation of the Hedgehog pathway showed little effect on primary cilia with or without INPP5E.

Supplemental Figure 8. Uncropped immunoblotting images.