KLC3 Regulates Ciliary Trafficking and Cyst Progression in CILK1 Deficiency–Related Polycystic Kidney Disease

Significance Statement

Mutations in ciliogenesis-associated kinase 1 (CILK1) cause ciliopathies. However, the pathogenesis of the ciliary defect in the CILK1-deficient kidney remains unknown. We found that CILK1 deficiency in a mouse model leads to polycystic kidney disease (PKD) with abnormal ciliary trafficking and that kinesin light chain–3 (KLC3), a novel ciliary regulator, interacts with CILK1. Furthermore, KLC3 localizes at cilia bases, where it promotes ciliary trafficking of the IFT-EGFR complex, which contributes to cyst progression. KLC3 knockdown restored abnormal ciliary trafficking and cyst progression caused by CILK1 deficiency. Identifying KLC3 as a ciliary regulator involved in cystogenesis provides insights into the pathogenesis of CILK1 deficiency–related PKD.

Visual Abstract

Abstract

Background

Ciliogenesis-associated kinase 1 (CILK1) is a ciliary gene that localizes in primary cilia and regulates ciliary transport. Mutations in CILK1 cause various ciliopathies. However, the pathogenesis of CILK1-deficient kidney disease is unknown.

Methods

To examine whether CILK1 deficiency causes PKD accompanied by abnormal cilia, we generated mice with deletion of Cilk1 in cells of the renal collecting duct. A yeast two-hybrid system and coimmunoprecipitation (co-IP) were used to identify a novel regulator, kinesin light chain–3 (KLC3), of ciliary trafficking and cyst progression in the Cilk1-deficient model. Immunocytochemistry and co-IP were used to examine the effect of KLC3 on ciliary trafficking of the IFT-B complex and EGFR. We evaluated the effects of these genes on ciliary trafficking and cyst progression by modulating CILK1 and KLC3 expression levels.

Results

CILK1 deficiency leads to PKD accompanied by abnormal ciliary trafficking. KLC3 interacts with CILK1 at cilia bases and is increased in cyst-lining cells of CILK1-deficient mice. KLC3 overexpression promotes ciliary recruitment of IFT-B and EGFR in the CILK1 deficiency condition, which contributes to the ciliary defect in cystogenesis. Reduction in KLC3 rescued the ciliary defects and inhibited cyst progression caused by CILK1 deficiency.

Conclusions

Our findings suggest that CILK1 deficiency in renal collecting ducts leads to PKD and promotes ciliary trafficking via increased KLC3.

Primary cilia are composed of a microtubule-based axoneme that extends from the centrosomal mother centriole.¹ Ciliary transport is involved in ciliogenesis and ciliary maintenance, as appropriate protein delivery is essential for ciliary axoneme and membrane extension.^2–5 There are two main movements of intraflagellar transport (IFT): anterograde intraflagellar transport (IFT-B) and retrograde IFT. IFT-B and retrograde IFT movements are powered by heterotrimeric kinesin-2 and cytoplasmic dynein-2, respectively,^6–9 which transport specific ciliary cargos.^4,10–13 Mutations in the genes that function in ciliary transport cause primary ciliary defects and ciliopathies,¹⁴ such as polycystic kidney disease (PKD).^15,16

Autosomal dominant polycystic kidney disease (ADPKD) is one of the most common genetic diseases, and it is caused by mutations in PKD1 and PKD2.^17,18 PKD1 and PKD2 encode the polycystin-1 and polycystin-2 (PC2) transmembrane proteins, respectively, which form the polycystin complex in renal primary cilia.^19,20 There is ample evidence showing that mutation of cilia-related genes leads to cystic kidney disease, accompanied by defective cilia and the dysregulation of multiple signaling pathways.^15,21–23 Although ciliary protein defects and ciliary malformation have been suggested as the main causative factors of PKD, the ciliary defect mechanisms related to renal cystogenesis are poorly understood.

Ciliogenesis-associated kinase 1 (CILK1), also known as intestinal cell kinase, is localized in the basal bodies and ciliary tips of primary cilia and functions as a regulator of ciliary transport.²⁴ Mutations in CILK1 are associated with ciliopathies, including endocrine cerebro-osteodysplasia syndrome and juvenile myoclonic epilepsy.^24–26 In Cilk1-mutant mice, multiple ciliary defects and phenotypic abnormalities have been observed; however, the mechanisms underlying the ciliary defect in the CILK1-deficient kidney have not been identified.

Kinesin-1 is a microtubule motor that binds and transports multiple cellular cargos and is involved in multiple cell processes.²⁷ The kinesin-1 motor is composed of two kinesin heavy chains (KHCs) and two kinesin light chains (KLCs); it exists in multiple isoforms that carry specific cargos.²⁸ KLCs have two conserved motifs: a heptad repeat domain and tetratricopeptide repeat (TPR) domain.^28,29 The heptad repeat domain is involved in the interaction with the KHC, and the TPR domains are hypothesized to regulate KHC motor activity and bind cargo proteins.^30,31 Despite the involvement of kinesin-1 transport in many cellular processes, the proteins interacting with kinesin-1 remain largely unknown.

Here, we report a novel regulator, KLC3, that promotes ciliary trafficking and cystogenesis in CILK1 deficiency–related PKD. We found that KLC3 overexpression induced ciliary recruitment of IFT-B and EGF receptor (EGFR), which contributed to the ciliary defect involved in cyst progression. Downregulation of KLC3 rescued the ciliary defects and inhibited cyst progression in CILK1-deficient cells, indicating that KLC3 is a ciliary regulator involved in cyst progression in PKD. In brief, we identified a ciliary trafficking mechanism governed by KLC3 and demonstrated that dysregulation of KLC3 contributes to cyst progression in CILK1 deficiency–related PKD.

Methods

Mice

Mice carrying the Cilk1^tm1a allele,²⁴ obtained from Hyuk Wan Ko (Yonsei University, Seoul, Korea), were crossed with FLPe mice to generate progeny harboring the conditional allele Cilk1^flox, in which exon 6 is excised by Cre recombinase in a conditional manner. Mice with specific deletion of Cilk1 in the renal collecting ducts were generated by crossing the Cilk1^flox/flox mice with HoxB7-Cre transgenic mice,³² which were obtained from Minho Shong (Chungnam National University, Daejeon, Korea). Mice with the Cilk1^flox/+:HoxB7-Cre genotype were used as wild-type controls, and Cilk1^flox/flox: HoxB7-Cre mice were used as experimental mice. All animal experiments were approved by the Institutional Animal Care and Use Committee of Sookmyung Women’s University and Dongguk University. Pkd1 ^flox and Pkd2 ^flox mice were obtained from Stefan Somlo (Yale University, New Haven, CT), and specific deletions of Pkd1 and Pkd2 were generated by crossing with HoxB7-Cre transgenic mice.³³

Cell Culture

Mouse inner medullary collecting duct (mIMCD-3) cells were cultured in DMEM/F-12 (Welgene) supplemented with penicillin-streptomycin and 10% (vol/vol) FBS. Human embryonic kidney 293T (HEK293T) cells were cultured in DMEM (Welgene) supplemented with penicillin-streptomycin and 10% (vol/vol) FBS. For three-dimensional culture, suspensions of mIMCD-3 cells (2.0–3.0 × 10⁵ cells/ml) were mixed with Matrigel (354230; Corning) at a 1:1 ratio and plated in an eight-well chamber slide system (C7182; Sigma). Phase-contrast micrographs of individual spheroids were obtained using a model IX70 light microscope (Olympus). Cyst areas were measured using ISP capture software (Olympus). All cells were cultured at 37°C in a humidified atmosphere of 5% CO₂.

Transfection

For siRNA transfection, Cilk1 and Klc3 siRNAs were purchased from Santa Cruz Biotechnology and Bioneer, respectively. Cells were transfected with these siRNAs using Lipofectamine RNAiMAX (Invitrogen). For plasmid transfection, the mouse Cilk1 coding sequence was cloned into the pCMV-HA-N vector, and mouse KLC3 and KLC3ΔTPR coding sequences were cloned into both pCMV-Tag2B and pEGFP-C2. Cells were transfected with these plasmids using FuGene (Promega). The mutant mouse Cilk1 TDY plasmid was obtained from Hyuk Wan Ko.

Yeast Two-Hybrid Screening

CILK1 285–629 cDNA (1038 bp) was PCR amplified using the following primers: CILK1 285–629F, 5′-GCGAATTCCAGATCGGACACCCACTGG-3′; CILK1 285–629R, 5′-CGGCTCGAGTCACCGCCGGGATGGGTA-3′. The PCR product was cloned into the pGBKT7 vector, which contains the DNA binding domain of GAL4. Saccharomyces cerevisiae AH109 was cotransformed with GAL4 DNA binding domain–fused CILK1 285–629 and a human kidney cDNA activation domain library. Two reporter genes, HIS3 and ADE2, were used as selection markers. Yeast transformants were spread on selection medium lacking leucine, tryptophan, adenine, and histidine. Of the 3.0 × 10⁶ colonies that were screened, >400 grew on selection medium lacking leucine, tryptophan, adenine, and histidine. To confirm the interaction, selected candidate prey genes were PCR amplified or transformed into Escherichia coli; then, they were retransformed into S. cerevisiae AH109 with the CILK1 285–629 bait plasmid. Screening was conducted at PanBionet Corp.

Coimmunoprecipitation

Cells were lysed with 50 mM Tris-HCl, pH 8.0, 50 mM NaCl, 1 mM EDTA, 1% Triton X-100, and 0.5% NP-40 supplemented with a protease and phosphatase inhibitor mixture (Roche) at 4°C for 5 minutes. Lysates were clarified by centrifugation at 4°C at 20,000×g for 15 minutes. For immunoprecipitation, lysates were incubated with anti-FLAG (A1804; Sigma), followed by incubation with Dynabeads Protein G (10004D; Invitrogen). Precipitates were washed three times with the lysis buffer described above. Precipitates were separated by SDS-PAGE, and immunoblotting was performed.

Quantitative RT-PCR

RNA was prepared from cell lysates using the NucleoSpin RNA/Protein kit (MACHEREY-NAGEL). RNA was reverse transcribed into cDNA using M-MLV reverse transcription (Promega) per the supplier’s protocol. PCR was performed with SYBR Green qPCR Master Mix (PCR Biosystem) using a LightCycler 96 System (Roche). Relative target gene expression levels were calculated using the 2^−ΔΔCq method and were normalized to the expression levels of GAPDH or ACTB.

Histologic Analysis

Formalin-fixed kidneys were embedded in paraffin and sectioned at 5-μm thickness. The sections were rehydrated in a graded ethanol series and stained with hematoxylin and eosin.

Human Tissue Collection

The study was conducted in accordance with the Declaration of Helsinki and was approved by the Institutional Review Board of Seoul National University Hospital (H-0701–033–195) and Samsung Medical Center (SMC 2019–08–074). Paraffin-embedded ADPKD and non-ADPKD tissue blocks were obtained from Seoul National University Hospital and Samsung Medical Center, respectively. As a non-ADPKD tissue control, noncancerous tissue regions were obtained from patients with renal cell carcinoma who underwent radical nephrectomy.

Immunostaining

For immunocytochemistry, mIMCD-3 cells were fixed with cold methanol at –20°C for 10 minutes. Fixed cells were washed with PBS and incubated with primary antibodies in permeabilizing solution at 4°C overnight. After washing off the primary antibodies, fluorescence-conjugated secondary antibodies were added, and cells were further incubated at room temperature for 2 hours. Cells were mounted using fluorescent mounting medium (S3023; Dako).

For immunofluorescence, paraffin-embedded sections were deparaffinized and rehydrated. The sections were heated in a Borg Decloaker RTU (BD1000G1; Biocare Medical) for antigen retrieval and blocked with blocking solution PK-6200 (Vector Laboratories) for 1 hour followed by overnight incubation with the appropriate primary antibody at 4°C. After washing, fluorescence-conjugated secondary antibodies were added, and sections were incubated at room temperature for 2 hours. Cells were mounted using mounting medium with DAPI (H-1500; Vector Laboratories).

Primary antibodies against the following proteins were used: pERK (4370), Ac-α-tubulin (5335), EGFR (4267), pEGFR Y845 (6963), and pEGFR Y1068 (3777) from Cell Signaling Technology; rhodamine Dolichos biflorus agglutinin (RL-1032), fluorescein D. biflorus agglutinin (RL-1031), and Lotus tetragonolobus lectin (FL-1321) from Vector Laboratories); acetylated tubulin (T6793) and γ-tubulin (T6557) from Sigma; CILK1 (H00022858-A01) from Abnova; α-Sma (ab5694), Ki67 (ab16667), KLC3 (ab180523), IFT46 (ab122422), KIF3A (ab11259), KIF17 (ab11261), and EGFR (ab52894) from Abcam; PC2 (sc-25749 or sc-10376), GFP (sc-9996), and KLC3 (sc-398492) from Santa Cruz Biotechnology; and ARL13B (17711–1-AP), IFT20 (13615–1-AP), IFT25 (15732–1-AP), IFT52 (17534–1-AP), IFT88 (13967–1-AP), and IFT140 (17460–1-AP) from Proteintech. Immunofluorescence images were captured using an LSM-700 confocal laser scanning microscope (Carl Zeiss).

Western Blotting

Proteins were extracted using a NucleoSpin RNA/Protein kit (MACHEREY-NAGEL), separated by SDS-PAGE, and transferred to a PVDF membrane. The following primary antibodies were used: pERK (4370), ERK (9102) from CST; FLAG (F7425) from Sigma; HA (PRB-101P) from BioLegend; IFT46 (ab122422) from Abcam; IFT20 (13615–1-AP), IFT25 (15732–1-AP), IFT52 (17534–1-AP), IFT88 (13967–1-AP), and IFT140 (17460–1-AP) from Proteintech; β-actin (A300–491A) from Bethyl; and PC2 (sc-25749) from Santa Cruz Biotechnology. Immunoreactive proteins were detected using EzWestLumi plus (ATTO) and analyzed using an Amersham Imager 600 (GE Healthcare).

Zebrafish

All zebrafish husbandry and experimental protocols complied with the institutional guidelines and were approved by local ethics boards (Sookmyung Women’s University Animal Care and Use Committee, SMWU-IACUC-1712–036). Zebrafish were maintained at 28.5°C under standard conditions and a 14-hour light/10-hour dark cycle. AB* wild-type embryos were obtained through natural crosses. Embryos were treated with 1-phenyl-2-thiourea (Sigma) to suppress melanin synthesis.

Morpholino and RNA Injections

Splice-blocking morpholinos (MOs) for cilk1 and klc3 were designed using Gene Tools and microinjected into one-cell AB* zebrafish embryos (male germ cell–associated kinase [mak] MO: 5′-CATCTGTCTGACACACCTCTTGATG-3′ and klc3 MO: 5′-ACACTCAGACCCTTAATTCACAGTA-3′).

For mouse Cilk1 overexpression experiments, full-length Cilk1 cDNA was amplified by RT-PCR from total RNA isolated from mIMCD-3 cells using sense (5′-CCATCGATTCGAATTCATGAATAGATACACAACGATCAAGC-3′) and antisense (5′-GAGAGGCCTTGAATTCTCACCGCCGGGATGGGTACTTGG-3′) primers and then cloned between EcoRI restriction sites of the pCS2+ vector. For in vitro transcription, Cilk1 RNA was synthesized using an mMESSAGE mMACHINE kit (Ambion) and injected into one-cell embryos. RNAs were microinjected at 100–250 ng/μl in nuclease-free water. Injected embryos were incubated until 30 hours postfertilization or 2 days postfertilization and imaged on a bright-field dissection microscope (Nikon SMZ1500).

Statistical Analyses

GraphPad Prism 5 (GraphPad Software) was used for statistical analysis. Data are presented as the mean±SD and were analyzed using the unpaired one-tailed t test. P<0.05 was considered significant.

Results

Deletion of Cilk1 in Renal Collecting Ducts Causes PKD Accompanied by Ciliary Defect

To identify the postnatal kidney and renal cilia phenotypes induced by CILK1 deficiency, Cilk1-floxed mice harboring two LoxP regions in Cilk1 exon 6 (Supplemental Figure 1A) were mated with HoxB7-Cre transgenic mice to delete Cilk1 only in renal collecting duct cells.

Renal cysts derived from collecting ducts of Cilk1^f/f:HoxB7-Cre mice at 2 weeks of age were small but grew in size and number with aging (Figure 1A, Supplemental Figure 1B). Cilk1^f/f:HoxB7-Cre mice died of PKD at 17–30 weeks (Figure 1B), and the two kidney weight–body weight ratio was elevated in Cilk1^f/f:HoxB7-Cre mice (Figure 1C). Cilk1^f/f:HoxB7-Cre kidneys showed renal fibrosis and increased cell proliferation (Figure 1, D and E, Supplemental Figure 1C). These findings suggested that Cilk1 deletion in renal collecting duct cells causes PKD.

Figure 1.

Deletion of Cilk1 in renal collecting ducts causes PKD. (A) Hematoxylin and eosin–stained sections of Cilk1^f/+:HoxB7-Cre (Cilk1 conditional wild type [cWT]) and Cilk1^f/f:HoxB7-Cre (Cilk1 conditional knockout [cKO]) kidneys at 2, 6, and 8 weeks. (B) Survival rates of Cilk1 cWT and cKO mice. (C) Ratio of two kidney weight (2KW) to total body weight (BW). Data are presented as the mean±SD. (D) Masson trichrome staining of Cilk1 cWT and cKO kidney at 8 weeks. (E) Cilk1 cWT and cKO kidneys stained for nuclei, α-Sma, and D. biflorus agglutinin (DBA) at 6 weeks. (F) Cilk1 cWT and cKO kidneys stained for nuclei, acetylated α-tubulin, and DBA at postnatal day 1 (P1) and 2 and 4 weeks. (G) Cilk1 cWT and cKO kidneys stained for nuclei, collecting ducts, Ac-α-tub, and PC2 at 8 weeks. DBA was used as a collecting duct marker, and acetylated α-tubulin was used as a cilia marker. Cy, Cyst.

Cilk1 deletion induces ciliary defects in mice.^24,34 However, the phenotype of primary cilia in CILK1 deficiency–related PKD has not been elucidated. Cilia were retained in cyst-lining cells of Cilk1^f/f:HoxB7-Cre kidneys on postnatal day 1 and at 2 and 4 weeks (Figure 1F). To assess whether the cilia were defective, we examined the PC2 expression pattern in the cilia of CILK1-deficient cyst-lining cells. PC2 had accumulated in the ciliary tip of cyst-lining cells in Cilk1^f/f:HoxB7-Cre mice (Figure 1G). Collectively, CILK1 deficiency induces PKD accompanied by abnormal cilia.

KLC3 Interacts with CILK1 at the Bases of the Primary Cilia

To identify interactor proteins involved in ciliogenesis and cystogenesis in CILK1 deficiency–related PKD, a yeast two-hybrid screen was performed using the C terminus of CILK1 as the bait. We identified KLC3 as an interactor of CILK1. We then generated HA-tagged CILK1 and FLAG-tagged KLC3 for coimmunoprecipitation, which confirmed that CILK1 and KLC3 interact (Figure 2A).

Figure 2.

CILK1 and KLC3 interact at the bases of primary cilia. (A) Coimmunoprecipitation of HA-CILK1 and FLAG-KLC3 from HEK293T cell lysates. Immunoprecipitation (IP) was performed using anti-FLAG. IB, immunoblot. EV, empty vector. (B) Staining of mIMCD-3 cells with nuclei, CILK1, KLC3, and Ac-α-tubulin. (C) Schematic representations of the KLC3 and KLC3ΔTPR constructs. HR, heptad repeat. (D and E) mIMCD-3 cells transfected with EGFP-KLC3 or EGFP-KLC3ΔTPR were stained for Ac-α-tubulin, γ-tubulin, and Cilk1. γ-tubulin was used as a basal body marker. DAPI was used for nuclear staining.

Next, we investigated the localization of KLC3 and CILK1 in mIMCD-3 cells. Consistent with previous findings,²⁴ CILK1 was located in the basal bodies of the primary cilia (Figure 2B). KLC3 localized to the proximity of the cilia bases and partially colocalized with CILK1 in the basal bodies. These findings were consistent with those made in mIMCD-3 cells transfected with EGFP-tagged KLC3 (Figure 2D). To confirm whether the TPR domains of KLC3 contribute to its localization, we generated a TPR domains deletion construct (KLC3ΔTPR) (Figure 2C). EGFP-tagged KLC3ΔTPR also localized to basal bodies but accumulated around the bases of cilia to a lesser extent than full-length KLC3 (Figure 2E), suggesting that the TPR domains of KLC3 function in determining its localization to the base of primary cilia. Collectively, KLC3 is localized to the bases of primary cilia and interacts with CILK1 in the basal bodies of ciliated renal epithelial cells.

KLC3 Promotes Ciliary Trafficking of the IFT-B Core-2 Complex

We transfected mIMCD-3 cells with EGFP-tagged KLC3 or KLC3ΔTPR and observed the ciliary phenotypes. The proportion of cilia-positive cells was three-fold higher in KLC3-overexpressing cells, and the cilia of KLC3-overexpressing cells were nearly 1.5 times longer than those of control (Figure 3A). In contrast, KLC3ΔTPR-transfected cells had a smaller proportion of cilia-positive cells, with shorter cilia than KLC3-overexpressing cells. In addition, the distances between mother and daughter centrioles in KLC3ΔTPR-transfected cells were longer than those in control and KLC3-overexpressing cells, which provides evidence that KLC3 might interact with proteins involved in regulating basal bodies via its TPR domain. Collectively, these results show that KLC3 functions in cilia regulation in renal epithelial cells.

Figure 3.

KLC3 interacts with and is involved in the ciliary trafficking of IFT46, IFT52, and IFT88. (A and C–E) For immunostaining, EGFP-tagged empty vector, EGFP-KLC3, or EGFP-KLC3ΔTPR was transfected into mIMCD-3 cells. (A) Transfected cells were stained with antibodies targeting Ac-α-tubulin and γ-tubulin. Cilia length and the distance between centrosomes were measured using ImageJ. (B) For coimmunoprecipitation, FLAG-tagged empty vector, FLAG-KLC3, or FLAG-KLC3ΔTPR was transfected into HEK293T cells and pulled down with anti-FLAG. (C–E) Transfected cells were stained with antibodies targeting Ac-α-tubulin, (C) IFT46, (D) IFT52, and (E) IFT88. The graphs show representative images indicating the mean distribution of fluorescence intensities of IFT proteins quantified using Zen software. Data are presented as the mean±SD. EV, empty vector; IB, immunoblot; IP, immunoprecipitation; ns, not significant.

CILK1 regulates anterograde transport in primary cilia.^34,35 As KLC3 belongs to the kinesin family, we assessed whether it transported any cargos, such as IFT proteins. We found that KLC3 interacts with IFT46, IFT52, IFT88, and IFT140 but not with IFT20 and IFT25 (Figure 3B). Surprisingly, KLC3ΔTPR barely interacted with any of these proteins. Remarkably, IFT46, -52, and -88 are involved in the IFT-B core-2 complex,³⁶ implying that KLC3 regulates ciliary trafficking of the IFT-B core-2 complex. KLC3-overexpressing cells simultaneously displayed elongated cilia and the accumulation of IFT46, IFT52, and IFT88 proteins in the ciliary tips, whereas these IFT proteins did not accumulate in the ciliary tips in KLC3ΔTPR-overexpressing cells (Figure 3, C–E). We measured the fluorescence intensities of IFT46, -52, and -88 along the proximal to distal end of cilia in control and KLC3-overexpressing cells. These IFT proteins were significantly increased in at least one point of three along the distal end; IFT46 was strongly accumulated in the ciliary tip, and IFT52 and IFT88 were partially accumulated at the ciliary tip of KLC3-overexpresseing cells (Figure 3, C–E).

Kinesin-2 primarily regulates IFT-B trafficking.³⁶ Thus, we further examined the relation between kinesin-2 family proteins, KIF3A and KIF17, and KLC3. Both KIF3A and KIF17 interacted with KLC3 via its TPR domains (Supplemental Figure 2A), but only KIF3A tended to accumulate along the ciliary axoneme in KLC3-overexpressing cells (Supplemental Figure 2, B and C), implying that KLC3 is selectively involved in the ciliary trafficking of its binding proteins. Because PC2 accumulated in the ciliary tips of Cilk1^f/f:HoxB7-Cre cyst-lining cells, we confirmed that KLC3 could regulate ciliary PC2. PC2 did not interact with KLC3 (Supplemental Figure 2D), whereas the ratio of PC2 to cilia length was increased in KLC3-overexpressing cells (Supplemental Figure 2E). This result implies that KLC3 is indirectly involved in ciliary PC2 regulation. Collectively, KLC3 contributes to not only ciliogenesis but also, ciliary recruitment of the IFT-B core-2 complex via its TPR domains.

Increased KLC3 Expression Causes Ciliary IFT Trafficking Defects in CILK1-Deficient Cells

We examined KLC3 expression in the cyst-lining cells of CILK1-deficient kidneys. KLC3 expression was elevated along Cilk1^f/f:HoxB7-Cre cyst-lining cells (Figure 4A). Furthermore, IFT46, IFT52, IFT88, and IFT140 proteins tended to accumulate in the ciliary tips of Cilk1^f/f:HoxB7-Cre cyst-lining cells, whereas IFT20 did not (Figure 4B, Supplemental Figure 3A). We confirmed that KLC3 accumulated in the proximity of primary cilia bases in Cilk1 siRNA-transfected mIMCD-3 cells, accompanied by ciliary IFT88 accumulation (Figure 4C). Although KLC3 proteins accumulated in Cilk1 siRNA-transfected cells, its transcript level remained unchanged, indicating that CILK1 negatively regulates KLC3 expression at the protein level.

Figure 4.

KLC3 overexpression increases ciliary IFT trafficking in CILK1-deficient cells. (A) Immunostaining of nuclei, D. biflorus agglutinin (DBA), and KLC3 in Cilk1 conditional wild-type (cWT) and conditional knockout (cKO) kidneys at 8 weeks. (B) Immunostaining of nuclei, DBA, Ac-α-tubulin, IFT46, IFT52, and IFT88 in Cilk1 cWT and Cilk1 cKO kidneys at 8 weeks. (C) Immunostaining of nuclei, IFT88, and KLC3 in mIMCD-3 cells transfected with control or Cilk1 siRNA (left panel). The bar graph shows the relative mRNA levels of Cilk1 and KLC3 in siRNA-transfected mIMCD-3 cells (right panel). (D–F) Immunostaining of nuclei, Ac-α-tubulin, (D) IFT46, (E) IFT52, and (F) IFT88 in mIMCD-3 cells transfected with control, Cilk1, or Cilk1 and Klc3 siRNAs. Signal intensities of IFT proteins at basal bodies and the distal end of each cilium were quantified using ImageJ; data are presented as the mean±SD. The bar graph shows the relative mRNA levels of Cilk1 and KLC3 in siRNA-transfected mIMCD-3 cells. Cy, Cyst.

The abnormal accumulation of IFT-B core-2 proteins in the cilia in cyst-lining cells was in agreement with the above finding that KLC3 interacts with these proteins and is involved in their trafficking (Figure 3, B–E). In addition, KIF3A accumulated in the ciliary tips of CILK1-deficient cells, whereas KIF17 did not (Supplemental Figure 3, B and C). These data indicate that IFT-B core-2 proteins accumulate in the ciliary tips upon overexpression of KLC3 and thus, that KLC3 is involved in the ciliary recruitment of these proteins in CILK1-deficient cyst-lining cells.

To examine whether the abnormal accumulation of IFT-B core-2 proteins in CILK1-depleted cells could be rescued by suppression of KLC3, we simultaneously stained mIMCD-3 cells to detect cilia and IFT46, IFT52, and IFT88, and then we measured individual IFT protein signal intensities from basal bodies to the distal end of the cilia (Figure 4, D–F). In mIMCD-3 cells transfected with a Cilk1-targeted siRNA, the staining intensities of all IFT-B core-2 proteins along the primary cilia were significantly increased compared with controls. Interestingly, IFT protein accumulation in the cilia was rescued by cotransfection of Cilk1 and Klc3 siRNAs. To sum up, CILK1 depletion results in the accumulation of IFT-B core-2 proteins in primary cilia via an increase of KLC3.

Suppression of KLC3 Relieves Cyst Progression Induced by CILK1 Depletion

As KLC3 was increased in CILK1-deficient cyst-lining cells, we hypothesized that it also plays a role in cyst progression. We conducted an in vitro cyst formation assay and found that KLC3 overexpression resulted in an increase in cyst lumen size (Figure 5A). Next, we examined whether knockdown of KLC3 in CILK1-deficient cysts would inhibit cyst progression. Cyst progression was not affected in cells transfected with a CILK1 kinase-inactive form (CILK1 TDY mutant) (Supplemental Figure 4), whereas decreased expression of CILK1 resulted in enlarged cysts (Figure 5B). Interestingly, suppression of KLC3 relieved in vitro cyst enlargement induced by CILK1 deficiency (Figure 5B).

Figure 5.

Cyst enlargement caused by Cilk1 knockdown is rescued by Klc3 knockdown. (A and B) Cyst formation was observed using Matrigel-based three-dimensional culture. (A) Images of cysts from mIMCD-3 cells transfected with EGFP-tagged empty vector or EGFP-KLC3 and (B) transfected with control, Cilk1, or Cilk1 and Klc3 siRNAs. The graph showing cyst lumen areas was plotted using data obtained from the images. Data are presented as the mean±SD. (C) Representative images of zebrafish embryos injected with the indicated MOs at 30 hours postfertilization (hpf). The inset shows a magnification of the dilated pronephric duct in the black-boxed region. (D) Graph showing the percentages of pronephric dilation occurrence of zebrafish embryos of the indicated groups at 30 hpf. (E) Histogram showing the percentages of zebrafish embryos with normal, mild, and severe curly tail phenotype at 30 hpf. (F) Expression of pERK and ERK in zebrafish embryos of the indicated groups. β-actin was used as the loading control (left panel). The bar graph shows the band intensity quantification presented as the mean±SD (right panel).

Cilk1 is highly conserved³⁷ and is hypothesized to be the counterpart of the mak gene in zebrafish. The viability of zebrafish embryos injected with an mak MO was reduced to ≤10% at 2 days postfertilization but was maintained at approximately 80% upon Cilk1 RNA coinjection (Supplemental Figure 5A), indicating that the mak gene in zebrafish has a function similar to that of Cilk1; thus, we used zebrafish embryos injected with mak MO as a Cilk1-deficient model. MO-induced knockdown of mak resulted in pronephric dilation and curly tail (Figure 5C). In addition, the klc3 MO injection significantly decreased the occurrence of pronephric dilation and suppressed the severe curly tail of the mak MO group (Figure 5, D and E). The phenotype of the klc3 MO group was generally normal except for MO injection damage (Figure 5, C–E), implying that a decrease of klc3 in the mak MO group influences pathogenesis, but a decrease in klc3 in the normal group does not directly contribute to phenotypic change.

To confirm the off-target effect of MO, we used p53 MO to suppress cell death.³⁸ We found little phenotypic difference in the groups where the p53 MO was coinjected and in the groups where it was not (Supplemental Figure 5B), indicating that the zebrafish phenotype induced by mak and klc3 MO could be considered as unrelated to the off-target effect of MO. Next, we examined alterations in pERK related to cystogenesis in zebrafish. The increase in pERK expression in mak MO was suppressed by klc3 MO coinjection (Figure 5F). Taken together, KLC3 plays a role in renal cyst progression caused by CILK1 deficiency.

Increased KLC3 Expression Induces Ciliary EGFR Accumulation in CILK1-Deficient Cells

We evaluated whether the abnormal cilia in CILK1-deficient cells are involved in cyst progression. The cilia frequencies of CILK1-deficient cells were decreased upon genetic knockdown of IFT46 or KIF3A. Next, we compared cyst size in three-dimensional culture. Silencing of CILK1 alone led to increased cyst size, whereas cysts derived from cells transfected with siRNAs targeting CILK1 and IFT46 or KIF3A were decreased compared with CILK1-deficient cysts (Supplemental Figure 6, A–C). These data indicated that the defective cilia of CILK1-deficient cells may contribute to cyst progression.

Because PC2 was accumulated in the ciliary tips in the CILK1-deficient mouse, we examined whether CILK1 deficiency alters PKD1 and PKD2 expression levels. PKD1 mRNA expression was only increased in vivo, whereas PKD2 mRNA expression was increased in both in vivo and in vitro Cilk1-deficient models (Supplemental Figure 6, F and G). Thus, we determined that the reduced PKD2 could rescue the enlargement of cysts induced by CILK1 deficiency (Supplemental Figure 6, D–F). Interestingly, the cells treated with both CILK1 and PKD2 siRNA did not display reduced cysts compared with the cells treated with CILK1 siRNA only. These results suggested that the accumulation of PC2 caused by excessive ciliary transport was not involved in CILK1-deficient cystogenesis.

To provide the disease-promoting effect of the defective cilia of CILK1-deficient cells, we focused on EGFR signaling. EGFR promotes renal cyst formation and is one of the RTKs localized to the primary cilia.³⁹ The expression levels of pEGFR and its downstream target pERK were increased in CILK1-deficient kidneys (Figure 6, A and B). Treatment with the EGFR inhibitor EKI-785 reduced cyst enlargement in KLC3-overexpressing or siCilk1-transfected cells (Supplemental Figure 7), indicating that the activation of EGFR signaling can be considered as a cyst development–accelerating pathway in this model. In addition, EGFR protein accumulated in the cilia of CILK1-deficient cyst-lining cells (Figure 6A), implying that defective cilia that accumulate EGFR drive cyst progression in CILK1-deficient cells.

Figure 6.

Ciliary EGFR accumulation is mediated by the KLC3-IFT-B pathway in the CILK1-deficient kidney. (A) Immunostaining of nuclei, D. biflorus agglutinin (DBA), pERK, Ac-α-tubulin, EGFR, and pEGFR (Y1068) in Cilk1 conditional wild-type (cWT) and conditional knockout (cKO) kidneys. (B) Expression of EGFR signaling pathway components in Cilk1 cWT and cKO kidneys at 4 weeks (left panel). Bar graph showing the band intensity quantification presented as the mean±SD (right panel). (C) For coimmunoprecipitation, FLAG-tagged empty vector, FLAG-KLC3, or FLAG-KLC3ΔTPR was transfected into HEK293T cells and pulled down with anti-FLAG. (D and E) Immunostaining of nuclei, Ac-α-tub, and EGFR or pEGFR Y1068 in mIMCD-3 cells transfected with EGFP-tagged empty vector, EGFP-KLC3, or EGFP-KLC3 ΔTPR. (F) For coimmunoprecipitation, FLAG-tagged empty vector or FLAG-IFT46 was transfected into HEK293T cells and pulled down with anti-FLAG. (G) Immunostaining of nuclei, Ac-α-tub, and EGFR in mIMCD-3 cells transfected with EGFP-IFT46 vector. ; Cy, Cyst. IB, immunoblot; IP, immunoprecipitation.

Our data indicated that KLC3 modulates cyst progression via controlling ciliary trafficking. Given the role of KLC3 in cilia and cyst progression, we speculated that KLC3 is involved in ciliary EGFR trafficking. KLC3 interacted with EGFR via its TPR domains (Figure 6C), with accumulated ciliary EGFR and pEGFR in KLC3-overexpressing cells (Figure 6, D and E). Moreover, EGFR was found to interact with IFT46, and IFT46 partially colocalized with EGFR on the ciliary axoneme (Figure 6, F and G). Together, KLC3 overexpression leads to ciliary accumulation of the IFT-B-EGFR in CILK1-deficient cells, which contributes to cyst progression.

The Inverse Correlation between CILK1 and KLC3 Expression Is Related to Human ADPKD Pathogenesis Accompanied by Ciliary Defects

Next, we examined CILK1 and KLC3 expression in human ADPKD tissues to assess whether changes in their expression are clinically relevant. CILK1 expression was decreased in ADPKD tissues as compared with non-ADPKD tissues (Figure 7A). KLC3 was expressed at higher levels in cyst-lining cells of patients with ADPKD than in those of control subjects (Figure 7A). Interestingly, KLC3 was increased in cyst-lining cells of other PKD mouse models, Pkd1^f/f:HoxB7-Cre and Pkd2^f/f:HoxB7-Cre, compared with their respective controls (Supplemental Figure 8, A and B). We found that both of the in vitro cyst enlargements induced by decreased PKD1 and PKD2 were rescued by depleting KLC3 (Supplemental Figure 8C). These results suggested that there are strong correlations between increased KLC3 and PKD development in humans and mouse PKD models.

Figure 7.

Abnormal ciliary trafficking accompanied by increased KLC3 expression is observed in cyst-lining cells of patients with ADPKD. (A) Immunostaining of nuclei, CILK1, and KLC3 in non-ADPKD and ADPKD kidneys (left panel). Fluorescence intensities of CILK1 and KLC3 were quantified using ImageJ (right panel). Data are presented as the mean±SD. (B, C and D) Immunostaining of nuclei, Ac-α-tub, (B) IFT46, IFT52, IFT88, (C) EGFR, and (D) PC2 in non-ADPKD and ADPKD kidneys. Cy, Cyst. (E) Overview of KLC3 function in ciliary trafficking involved in cyst progression in CILK1 deficiency–related PKD. KLC3 expression increases in cyst-lining cells in CILK1 deficiency–related PKD and promotes ciliary trafficking of IFT-B. KLC3 interacts with IFT-B and EGFR, leading to the accumulation of the IFT-EGFR complex in the cilia, which results in abnormal cilia that activate cyst progression. Downregulation of KLC3 relieves both cyst progression and abnormal cilia trafficking, suggesting that KLC3 functions as a pathologic factor in ciliary defects in CILK1 deficiency–related PKD. Cy, Cyst.

We found that primary cilia were still present in those of patients with ADPKD, but IFT46, IFT52, IFT88, and EGFR accumulated in the ciliary tips (Figure 7, B and C). We also confirmed that PC2 was accumulated in the cilia of patients with ADPKD (Figure 7D). Collectively, increased KLC3 is strongly related to cyst progression accompanied by cilia defects not only in CILK1-deficient PKD but also, in universal PKD models.

Discussion

Mutations in Cilk1 have been associated with ciliary defects accompanied by multiple ciliopathies; however, the mechanisms underlying ciliary defects and cyst progression in CILK1-deficient kidneys remained largely undescribed. In this study, we generated Cilk1^f/f:HoxB7-Cre mice exhibiting a PKD phenotype and identified KLC3 as a novel ciliary regulator involved in cyst progression in CILK1 deficiency–related PKD (Figure 7E).

The abnormal ciliary phenotype in CILK1-deficient models appears to differ depending on the tissue.^24,34 We confirmed the ciliary phenotype in the CILK1-deficient kidney. Primary cilia were not observed in PKD models targeted by other cilia regulators,^22,40 whereas cilia were still present in CILK1-deficient kidneys. These results suggest that CILK1 deficiency contributes to cyst progression without complete loss of cilia in the kidneys.

KLC3, an interactor of CILK1, belongs to the kinesin-1 family. In support of the involvement of the kinesin-1 family in cilia regulation,⁴¹ we found that KLC3 localized to proximity of cilia bases, where CILK1 is localized, and that KLC3 overexpression resulted in elongated cilia. On the basis of reports that most ciliary proteins are localized in or around cilia,^42–44 it can be inferred that KLC3 is involved in cilia regulation.

As mislocalized IFT proteins implicated in defects of ciliary signaling pathways cause ciliopathies,^34,45,46 identification of regulators in ciliary IFT machinery is important to understand pathogenesis. CILK1 regulates the ciliary localization of IFT proteins.³⁴ Consistent with those findings, we observed KLC3 overexpression and ciliary accumulation of IFT-B proteins in CILK1-deficient kidneys. KLC3 interacts with KIF5C,⁴⁷ but cargos recruited by KLC3 remained unidentified. Here, we demonstrated that KLC3 interacts with the IFT-B core-2 complex³⁶ and promotes ciliary trafficking of this complex. Further, reduction in KLC3 diminished the ciliary accumulation of IFT-B caused by CILK1 depletion. On the basis of these observations, we propose a model in which KLC3 overexpression contributes to ciliary accumulation of IFT-B in the CILK1-deficient kidney.

KLC3 expression was increased in the cyst-lining cells in CILK1-deficient kidneys, which prompted us to validate the function of KLC3 overexpression in not only ciliary defects but also, cystogenesis. Overexpressed KLC3 resulted in cyst enlargement, and KLC3 suppression relieved cystogenesis induced by CILK1 depletion. This rescue effect of KLC3 downregulation on cyst progression was verified in an mak-targeted zebrafish model. These results indicate that increased expression of KLC3 is involved in renal cyst progression accompanied by ciliary defects.

EGFR localizes to the primary cilia of renal epithelial cells.⁴⁸ However, its ciliary trafficking is poorly understood. We found that KLC3 interacts with EGFR and promotes ciliary trafficking of the IFT-EGFR. EGFR signaling can be considered to be involved in cyst progression as EGFR inhibition attenuated cyst progression in PKD.^49,50 Therefore, we suggest that ciliary defects caused by KLC3 overexpression contribute to cyst progression via the activation of ciliary EGFR signaling in CILK1 deficiency–related PKD.

In conclusion, our findings provide new insights into the pathogenesis of CILK1 deficiency–related PKD through KLC3. KLC3 overexpression promotes ciliary trafficking of IFT-B and EGFR, which contributes to PKD progression. Interestingly, KLC3 overexpression was detected in cyst-lining cells in other rodent PKD tissues, suggesting that KLC3 is universally involved in the pathogenesis of PKD. Further studies of the mediators playing roles in the regulation of KLC3 will advance our understanding of the defective cilia–induced cystogenesis in PKD.

Disclosures

Y.K. Oh reports research funding from Bayer Korea, Otsuka Korea, and Sanofi. All remaining authors have nothing to disclose.

Funding

This work was supported by grants from the National Research Foundation of Korea (NRF) funded by the Korean government (MSIT) (2022R1A2C3002899, 2020R1I1A1A01072944, and 2016R1A5A1011974).

Author Contributions

J.H. Park, J.Y. Ko, and G. Rah conceptualized the study; J.H. Park, J.Y. Ko, and G. Rah were responsible for data curation; J.H. Park, J.Y. Ko, G. Rah, H. Cha, Joohee Kim, and M.J. Kim were responsible for investigation; J.H. Park, J.Y. Ko, G. Rah, Joohee Kim, and M.J. Kim were responsible for formal analysis; G. Rah, J.Y. Ko, H. Cha, J. Song, Jongmin Kim, K.H. Yoo, and H.W. Ko were responsible for methodology; J.H. Park was responsible for project administration; H. Kim, Y.K. Oh, C. Ahn, M. Kang, and H.W. Ko were responsible for resources; G. Rah, J.Y. Ko, and Joohee Kim were responsible for validation; J.Y. Ko and G. Rah were responsible for visualization; J.Y. Ko and J.H. Park were responsible for funding acquisition; J.H. Park provided supervision; G. Rah, J.Y. Ko, M.J. Kim and J.H. Park wrote the original draft; and G. Rah, J.Y. Ko, and J.H. Park reviewed and edited the manuscript.

Data Sharing Statement

All data used in this study are available in this article.

Supplemental Material

This article contains the following supplemental material online at http://jasn.asnjournals.org/lookup/suppl/doi:10.1681/ASN.2021111455/-/DCSupplemental.

Supplemental Figure 1. Deletion of Cilk1 in mouse renal collecting ducts.

Supplemental Figure 2. KLC3 promotes ciliary trafficking.

Supplemental Figure 3. Localization of Ift20, Ift140, Kif3α, and Kif17 in Cilk1-deficient cyst-lining cells.

Supplemental Figure 4. Effect of Cilk1 TDY on cyst formation in vitro.

Supplemental Figure 5. Phenotype of zebrafish embryos in various groups.

Supplemental Figure 6. The effect of ciliary proteins in vitro cyst formation in Cilk1 siRNA-treated cells.

Supplemental Figure 7. Inhibition of EGFR signaling suppresses cyst formation in vitro.

Supplemental Figure 8. Cyst progression induced by increased KLC3 expression in cyst-lining cells of Pkd1- and Pkd2-targeted models.