Introduction

Autosomal dominant polycystic kidney disease (ADPKD) is the most common inherited kidney disease, with a prevalence of 1/2500–1/1000 worldwide [1]. ADPKD is characterized by excessive proliferation of renal epithelial cells and progressive fluid-filled cysts in bilateral kidneys [2, 3], which is caused by mutations in Pkd1 or Pkd2 gene. Pkd1 and Pkd2 encode polycystin 1 (PC1) and polycystin 2 (PC2), respectively. PC1 and PC2 form a complex co-expressed on the primary cilia of renal tubular epithelial cells [4,5,6]. PC1 converts the mechanical signal alterations in cilia into intracellular calcium signal mediated by PC2 [7]. Defects of polycystin or primary cilia affect intracellular calcium and downstream signal transduction, which result in abnormal proliferation of renal tubular epithelial cells and secretion of cystic fluid, lead to the formation of renal cysts [8].

Most of ADPKD patients develop into end-stage renal disease around the age of 60 requiring kidney dialysis or transplantation [9], which imposes a serious burden to family and society [10, 11]. The clinical treatment of ADPKD is very limited. At present, there is only tolvaptan, a vasopressin type 2 receptor antagonist that is approved as a therapeutic medication for ADPKD [12,13,14]. However, the patients with long-term use of tolvaptan will face a high risk of liver injury [15,16,17,18,19]. Therefore, there is still an urgent need to discover therapeutic drugs that are effective against ADPKD with fewer adverse effects.

Intriguingly, ADPKD has extensive similar pathological features with solid tumors, including excessive cell proliferation, extracellular matrix remodeling, abnormal epithelial cellular polarity and differentiation, dysregulated energetics and alterations in primary cilia [20, 21].

Primary cilia are responsible for sensing fluid flow in eukaryotic cells, transducing extracellular mechanical and chemical signals into the cell and triggering physiological responses. When primary cilia are defective, they will lead to developmental disorders of various tissues and organs in the body, which are called ciliopathies.

The cystic kidney is the common phenotype in numerous ciliopathies. It is suggested that primary cilium defect may be an important trigger of the renal cyst formation [22]. Also, the abnormalities of primary cilia in renal tubular cells have been observed in many models of kidney disorders, indicating the central role of primary cilium in maintaining normal kidney function [23]. Previous study has found abnormalities in the primary cilium structure of renal epithelial cells in ADPKD [24]. Abnormal elongated primary cilia of cyst epithelial cells have been found in multiple PKD animal models such as pcy, jck, Pkd1RC/RC, even in ADPKD patients [25,26,27,28,29]. Especially, abnormally elongated primary cilia were found in the early cystic stage of ADPKD patients. Primary cilium length is correlated with cyst expansion [30]. All these data indicate that primary cilia length is a critical factor of cyst development in ADPKD.

A study suggests that primary cilium functions as a therapeutic target for ADPKD, and targeting primary cilia assembly inhibits renal ciliogenesis [25]. However, there are still many controversies about the morphological and structural abnormalities of renal tubular cilia in different ADPKD models. And the mechanism of primary cilia in regulating renal cysts formation has not been clearly studied. Therefore, in this study we intended to investigate whether the development of ADPKD can be alleviated by directly targeting primary cilia by pharmacological approaches.

Materials and methods

Cell culture and CCK-8 assay

Madin-Darby Canine Kidney (MDCK) cells were cultured with Dulbecco’s Modified Eagle’s Medium (11965175, DMEM, Gibco, Waltham, MA, USA), and mouse inner medullary collecting duct (mIMCD-3) cells were cultured with Dulbecco’s Modified Eagle's Medium/Nutrient Mixture F-12 (DMEM/F12, 12400024, Gibco, Waltham, MA, USA). They were cultured in DMEM/F-12 with 10% fetal bovine serum (FBS, 10099141C, Gibco, Waltham, MA, USA), 100 μg/ml streptomycin, and 100 U/ml penicillin, incubated at 37 °C with 5% CO2. MDCK cells were seeded on a 96-well plate at a density of 5000 cells/well. After treatment with candidate compounds at 25 μM for 24 h, 100 μl 5% CCK-8 solution (CK04, Dojindo Molecular Technologies, Kumamoto, Japan) was added to each well and incubated for 2 h. Absorbance at 450 nm was measured with a microplate reader (MQX200, Biotek, Winooski, VT, USA). When reached to 70% density, mIMCD cells were incubated with 10 μM of forskolin (FSK, 66575-29-9, Sigma, Saint Louis, MO, USA) for 0.5 h or 3 h and treated with different concentrations of 1-Indanone.

MDCK cyst model

MDCK cells were digested and resuspended in type I collagen (C2124, Sigma, Saint Louis, MO, USA) solution containing modified Eagle’s medium and 10 mM HEPES at pH~7.2. The cell suspensions were allocated into 24-well plates and cultured with DMEM-F12 medium with 10% FBS. FSK was supplemented from day 5 to stimulate cysts formation. Medium was changed every 12 h. Cysts were tracked by taking photos (30 cysts/well) every 2 days from day 4 to record their growth curve. Cyst diameter was measured and analyzed with ImageJ software (NIH Image, Bethesda, MD, USA). To assay cyst formation, the numbers of cysts and colonies were counted at day 6 of incubation with the test compounds.

Embryonic kidney cyst model

Embryonic kidney cyst model was set up as described [31]. Briefly, embryos at day 13.5 were harvested from pregnant ICR mice. Embryonic kidneys were placed on transwell filters (CLS3401-48EA, Corning, NY, USA) filled with conditioned medium as described. DMEM-F12 supplemented with 2 mM L-glutamine, 250 U/ml penicillin, 250 μg/ml streptomycin, 10 mM HEPES, 5 μg/ml insulin, 5 μg/ml transferrin, 2.8 nM selenium, 25 ng/ml prostaglandin E1, 32 pg/ml T3, and 100 μM 8-bromoadenosine-3’,5’-cyclic monophosphate (8-Br-cAMP, 23583-48-4, Sigma, MO, USA). Test compounds were supplemented from day 0 and medium was changed every 12 h. Through 6 days culture, embryonic kidney cysts grew massively under stimulation of 100 μM 8-Br-cAMP. The kidney micrographs were tracked on day 0, 2, 4, and 6. Cystic area and total kidney area were analyzed using ImageJ software (NIH Image, Bethesda, MD, USA).

Mouse model

Pkd1flox/flox: Ksp-Cre mice (C57BL/6 genetic background) were generated by intercrossing Pkd1flox/+ mice (from the laboratory of Dr. Stefan Somlo) with Ksp-Cre transgenic mice (from the laboratory of Dr. Peter Igarashi). Genotyping was conducted on postnatal day 1. From postnatal day 1 to day 5, mice were injected with candidate compound at 100 mg· kg−1 ·d−1 or vehicle solution subcutaneously. Mice were sacrificed on postnatal day 5. All experiments conducted on animals were according to animal protocols approved by the Peking University Health Science Center Committee on Animal Research (26 Febrary 2021; protocol no. LA2021080).

Molecular docking studies

1-Indanone was drawn in “ChemBioDraw Ultra” 14.0, transformed to 3D conformations and minimized using the “LigPrep” module in “Schrödinger 2018”. Microtubule (PDB code: 6EW0) was prepared for docking using the “Protein Preparation” module, and a grid box was created using “Receptor Grid Generation”. Prepared 1-Indanone was docked with microtuble using “Ligand Docking”. Bound complexes were visualized in “Schrödinger 2018”.

Measurements of cilium length

Imaging and quantification of cilium length was performed using the Galvano scanner of a Nikon A1+ confocal Ti microscope system with a ×100 immersion objective lens and ×4 optical zoom. Primary cilia were acquired from a minimum of 50 cells for each group and cilium length was measured using the “z projection” function of ImageJ software (NIH Image, Bethesda, MD, USA). The distance from the base to the tip of the cilium was measured by tracing the length of the cilium, using ImageJ’s segmented line tool. The cilium length of renal tubules or cysts was measured manually using ImageJ software (NIH Image, Bethesda, MD, USA) and converted in micrometers (0.031 μm/pixel).

Tubulin isolation and purification

Microtubule protein was purified from porcine brain by three cycles of temperature-dependent assembly and disassembly in buffer A (0.1 M Pipes, 2 mM EGTA, 1 mM MgSO4, pH 6.4) as described [32]. The first polymerization step was carried out in the presence of 4 M glycerol and the two further polymerization steps were performed without glycerol. Microtubule protein was stored at −70 °C in buffer A at a protein concentration of ~15–20 mg/ml.

Tubulin polymerization assay

Tubulin was reconstituted to 3 mg/ml using G-PEM buffer. The tubulin (100 μl) was added to each well of a pre-warmed 96-well plate and exposed to test compounds at 25 µM. The absorbance at 340 nm was recorded every 1 min for 65 min using a TECAN GENios multifunction microplate reader (IC 12833, TECAN U.S. Inc., Research Triangle Park, NC, USA) at 37 °C. The dose-response curves were plotted using Graphpad Prism (Graphpad Software Inc., San Diego, CA, USA).

Western blot

Proteins were separated on SDS-polyacrylamide gel though electrophoresis and then transferred to polyvinylidene difluoride membranes (IPVH00010, Millipore, Saint Louis, MO, USA). After blocking with 5% bovine serum albumin for 2 h at room temperature, membranes were incubated with primary antibodies (Table 1) overnight at 4 °C. Membranes were washed and incubated with goat anti-mouse IgG or goat anti-rabbit IgG secondary antibodies for 1 h at room temperature. Blots were then developed with the ECL kit (BF06053S-100, Biodragon, Suzhou, China) and detected with a chemiluminescence detection system (GeneGnomeXRQ, Syngene, Cambridge, United Kingdom). The expression levels of proteins were quantified relative to β-actin expression.

Table 1 Antibody information.
Full size table

Histological staining and immunofluorescence staining

For hematoxylin and eosin (H&E) and Masson’s trichome staining, 5 μm of kidney paraffin sections were used. Cyst diameter and cystic index were calculated using ImageJ software (NIH Image, Bethesda, MD, USA). Cystic area was measured based on the definition that a cyst has spherical structure with diameter ≥ 50 μm (cyst index = total cystic area/total kidney area) × 100%. For immunofluorescence staining, sections were blocked with 5% (w/v) goat serum at room temperature for 1 h and then incubated with a primary antibody for Ki-67 (Abcam, Cat# ab15580, RRID: AB_443209, 1:500 dilution), anti-ARL13B (Thermo Fisher Scientific, Cat# PA5-92354, RRID: AB_2806450, 1:200 dilution) at 4 °C overnight. After washing three times, sections were incubated with secondary antibody or Lotus tetragonolobus lectin (LTL, Vector Laboratories, Cat# FL-1321, RRID: AB_2336559, 1:400 dilution) and Dolichos biflorus agglutinin (DBA, Vector Laboratories, Cat# RL-1032, RRID: AB_2336396, 1:400 dilution) for 1 h. Hoechst dye (Sigma Aldrich, Cat# 94403, 1:1000 dilution) was used to stain nuclei. All images were captured with a Nikon fluorescence microscope.

Scanning electronic microscope (SEM)

The kidneys of mice were perfused and fixed as described [33]. One cubic millimeter tissues were collected and stored in fixative (2% glutaraldehyde in 0.1 M cacodylate buffer, pH range 6.8–7.4) for 24 h at 4 °C. Then, tissues were subsequently washed with PBS three times. The samples were kept in 1% osmium tetroxide fixative for 3 h. The samples were rinsed with distilled water. The processed tissues were dehydrated in different concentration of ethanol: 50%, 70%, 95%, and 100% with each step for 10–20 min. And the tissue samples were mounted onto metal stubs and coated with gold. Samples were viewed using a Hitachi tm4000 plusII (Hitachi, Tokyo, Japan) scanning electron microscope.

Transmission electronic microscope (TEM)

For TEM, tissue samples were fixed and dehydrated same as SEM. The samples were then transferred to embedding resin in acetone. Small pieces of kidney were embedded with TEM of epoxy resin embedded tissue kit (Genmed Gene Pharmaceutical Co., Ltd, Shanghai, China). Semithin sections were cut with a LKB III ultratome, and stained with the electron dense agent uranyl acetate and lead citrate to provide contrast. Staining was achieved by floating grids specimen side down on a small drop of stain on a piece of Parafilm, with extensive washing with distilled water after each step. Then ultrathin sections were ready for viewing and capturing.

Statistical analysis

All results were analyzed with GraphPad Prism software and expressed as mean values ± SEM. Statistics were analyzed by Student’s t test or one-way ANOVA followed by the Tukey’s multiple comparison tests was performed in. P value < 0.05 was considered statistically significant in all tests.

Results

1-Indanone was found to inhibit cyst development

Twenty-four compounds were virtually screened from the natural compound library (from Targetmol Biochemical) based on their inhibitory activity on cell proliferation signaling, such as Ras/MAPK, mTOR and Wnt signaling pathways. Compounds C, N, O, Q, T were excluded due to their cytotoxicity on MDCK cells by CCK-8 assay (Fig. 1a). The inhibitory activity on cysts of 19 remainders was determined in MDCK cyst model. The experimental results showed that compounds A, E, F, G, J, X remarkably inhibited the enlargement of MDCK cysts (Fig. 1b, c). The active compounds E, F, G, and J with the remarkable inhibitory effect on MDCK cyst were selected for further investigation in the embryonic kidney cyst model. Compounds E, F and G markedly suppressed the cyst expansion of embryonic kidney and reduced the cyst index without effect on kidney development (Fig. 1d). However, in the kidney-specific Pkd1 knockout (Pkd1flox/flox: Ksp-Cre) mouse model, only compound G showed notable inhibition on the development of renal cysts (Fig. 1e). These data suggested that compound G, which chemical name is 1-Indanone (Fig. 1f), had an inhibitory effect on cysts and was conducted in further pharmacological and mechanistic studies.

Fig. 1: 1-Indanone is identified as an inhibitor of renal cysts.
figure 1

a Cell viability examined by CCK-8. n = 6. b Cyst inhibition activity of tested compounds in MDCK cyst model. n ≥ 30. c Representative micrographs of MDCK cysts treated with compounds A, E, F, G, J, and X. Bar = 200 μm. d Representative micrographs of embryonic kidneys treated with compounds E, F, G, and J. Bar = 1 mm. e Representative photographs of wild-type (WT) and Pkd1 knockout (PKD) kidneys treated with compounds E, F, and G. Bar = 1 cm. f The chemical structure of 1-Indanone (compound G). Data are presented as means ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001 vs. Ctrl group.

Full size image

1-Indanone was pharmacologically confirmed to suppress cyst development

Further, we adopted Pkd1flox/flox; Ksp-Cre mice to test the inhibitory effect of 1-Indanone on renal cysts in vivo. From postnatal day 1 to day 5, Pkd1 knockout (PKD) mice were subcutaneously injected with 1-Indanone at 25, 50 or 100 mg· kg−1 ·d−1 or vehicle solution. There was no notable difference in appearance (Fig. 2a), body weight (Fig. 2b) or the liver index (Fig. 2c) between mice with or without 1-Indanone treatment. The organ indexes of brain, heart and gut in the mice of each group indicated 1-indanone had no obvious toxicity in vivo (data not shown). The kidney volume of PKD mice was dose-dependently decreased under the treatment with 1-Indanone at 25, 50, or 100 mg· kg−1 ·d−1 (Fig. 2d) and the kidney index of PKD mice was pronouncedly reduced by 1-Indanone at 100 mg· kg−1 ·d−1 (Fig. 2e). Kidney sections stained with H&E showed that cystic area in PKD kidneys was conspicuously reduced by 1-Indanone at 100 mg· kg−1 ·d−1 (Fig. 2f, g). The immunostaining results of LTL and DBA revealed that 1-Indanone mainly inhibited the expansion of cysts derived from the collecting duct in PKD mice (Fig. 2h).

Fig. 2: 1-Indanone suppresses cyst expansion in PKD mice.
figure 2

a Representative photographs of wild-type (WT) and Pkd1 knockout (PKD) mice treated without (Ctrl) or with 1-Indanone (Ind) at 100mg· kg−1 ·d−1. Bar = 2 cm. b Body weight. n ≥ 6. c Liver index (percentage of liver weight to body weight). n ≥ 6. d Representative photographs of WT and PKD kidneys treated without or with 1-Indanone at 25, 50, 100 mg· kg−1· day−1 from postnatal day 1 to day 5. Bar = 1 cm. e Kidney index (percentage of kidney weight to body weight) of WT and PKD mice treated with 1-Indanone at 100 mg· kg−1 ·d−1. n ≥ 6. ***P < 0.001 vs. Ctrl group in WT mice. ###P < 0.001 vs. Ctrl group in PKD mice. f Representative kidney sections with H&E staining from PKD mice treated with vehicle (left) or with 1-Indanone (right) (100 mg ·kg−1 ·d−1). Bar = 500 μm. g Cystic index (percentage of cystic area to whole kidney area). n = 9. **P < 0.01 vs. Ctrl group in PKD mice. h Immunofluorescence staining of LTL and DBA on kidney sections from WT or PKD mice treated without or with Ind. Bar = 1 mm. All data are presented as mean values ± SEM.

Full size image

To determine whether 1-Indanone could affect cyst formation, we examined the effect of 1-Indanone on MDCK cyst model. MDCK cells were cultured in 3D collagen gel with FSK (10 μM) for 6 days without or with 1-Indanone. The statistic results showed that 25 μM of 1-Indanone notably reduced the number of MDCK cysts without affecting the number of total cell colonies (Fig. 3a), which indicated 25 μM of 1-Indanone inhibited MDCK cyst formation. For assaying the effect of 1-Indanone on the enlargement of MDCK cyst, MDCK cells were cultured in 3D collagen gel with 10 μM FSK for 4 days. Then the MDCK cysts were treated with 1-Indanone at 1, 5 or 25 μM from day 5 to day 12. The MDCK cysts treated with FSK alone were enlarged excessively with diameters around 350 μm on day 12 (Supplementary Fig. S1). The enlargement of MDCK cysts was suppressed with 1-Indanone in a dose-dependent manner (Fig. 3b). We further investigated whether the inhibitory effect of 1-Indanone is reversible. After 1-Indanone was washed out on day 8, the cysts were re-enlarged with FSK stimulation, suggesting the inhibitory effect of 1-Indanone is reversible (Fig. 3c).

Fig. 3: 1-Indanone retards cyst progression dose-dependently.
figure 3

a Numbers of total colonies and MDCK cysts treated without (Ctrl) or with 25 μM of 1-Indanone (Ind) for 6 days. n = 10. b Growth curves of MDCK cysts treated with Ind at 1, 5, 25 μM. n = 50 in each group. c Growth curves of MDCK cysts treated with 25 μM 1-Indanone from day 4 to day 12 (solid line) or from day 4 to day 8 (dashed line). n = 50. d Representative micrographs of embryonic kidneys treated with 1-Indanone. Bar = 1 mm. Black lines means the time of 1-Indanone treatment. e Cyst index of embryonic kidneys treated with 1-Indanone at 1, 5, 25 μM for 6 days or 25 μM for 4 days. n ≥ 8. Data are presented as mean values ± SEM. Data are analyzed with one-way ANOVA multiple comparisons. *P < 0.05, **P < 0.01, ***P < 0.001 vs. Ctrl group. ###P < 0.001 vs. the group treat with 25 μM 1-Indanone for 6 days.

Full size image

The reversibly inhibitory effect of 1-Indanone was confirmed in embryonic kidney cyst model. E13.5 embryonic kidneys were stimulated by 100 μM 8-Br-cAMP for 6 days without or with 1-Indanone treatment for 4 days. The experimental results showed that 8-Br-cAMP-induced cyst expansion in embryonic kidneys was markedly retarded by 25 μM 1-Indanone (Fig. 3d, e). However, after 1-Indanone was washed out on day 4, renal cysts regrew progressively, which indicated that 1-Indanone reversibly inhibited cyst development in embryonic kidneys.

1-Indanone inhibited hyperproliferation of cystic epithelial cells

Hyperproliferation of cystic epithelial cells is the main pathological feature of ADPKD. Immunofluorescence showed that the number of Ki67-positive cells in the kidneys of PKD mice was significantly reduced after 1-Indanone treatment (Fig. 4a). Consistently, the up-regulated PCNA, a proliferation marker protein, was remarkably suppressed by 1-Indanone in PKD kidneys (Fig. 4b). In order to further verify the effect of 1-Indanone on cell proliferation, we detected the activation of cAMP response element binding protein (CREB) and Ras/MAPK and mTOR signaling pathways, which play critical roles in regulating cellular over-proliferation in ADPKD [34]. It was found that 1-Indanone inhibited the excessive activation of the CREB and ERK in Ras/MAPK signaling pathway. The expression of H-ras and B-raf tended to be down-regulated by 1-Indanone, but there was no statistical difference (Fig. 4c and Supplementary Fig. S2a). Moreover, the expression and activation of mTOR, S6K, S6 in mTOR signaling pathway in PKD mice were also retarded after 1-Indanone treatment (Fig. 4d and Supplementary Fig. S2b). These results indicated that 1-Indanone inhibited the cellular hyperproliferation in PKD by mTOR and MAPK pathways.

Fig. 4: 1-Indanone inhibits abnormal cell proliferation in PKD mice.
figure 4

a Representative immunofluorescence images of the kidney sections stained with Ki-67 in PKD mice without (Ctrl) or with 1-Indanone (Ind) treatment. Magnification bar = 200 μm. Bar = 800 μm. b Representative Western blots (upper) and relative protein expression (lower) of PCNA in kidneys of PKD and wild-type (WT) mice treated without or with 1-Indanone at 100 mg· kg−1 ·d−1 for 4 days. n = 7. c Representative Western blots (left) and relative protein expression (right) of proteins in Ras/MAPK signaling pathway in the kidneys of WT mice and PKD mice without or with 1-Indanone treatment. n ≥ 6. d Representative Western blots (left) and relative protein expression (right) of proteins in mTOR signaling pathway in the kidneys of WT mice and PKD mice with vehicle or 1-Indanone treatment. n ≥ 6. Data are presented by mean values ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001 vs. Ctrl group in WT mice, #P < 0.05, ##P < 0.01, ###P < 0.001 vs. Ctrl group in PKD mice. Data are analyzed with one-way ANOVA.

Full size image

1-Indanone inhibited the abnormally elongation of primary cilia in cystic epithelium

It is well established that the defects in the structure of primary cilia are related to renal cyst development [35]. There is evidence proved that primary cilia length in cystic epithelial cells correlated with ADPKD progression [30]. Therefore, we investigated the structural change of the primary cilia of renal tubular epithelial cells in PKD mice and the effect of 1-Indanone on primary cilia. Immunofluorescence staining with ARL13B showed that the lengthes of primary cilia in cystic and tubular epithelial cells of PKD mice were abnormally increased (Fig. 5a). 1-Indanone notably inhibited the abnormal elongation of primary cilia in PKD kidney cystic epithelial cells, and had no notable effect on that in wild-type kidneys. We further observed the structure of primary cilia using transmission electronic microscope (TEM) and scanning electronic microscope (SEM). Consistent with immunofluorescence results, TEM and SEM images also showed that primary cilia were abnormally lengthened and thinner in PKD kidneys, which were inhibited by 1-Indanone (Fig. 5b, c).

Fig. 5: 1-Indanone reduces the primary cilium length of renal tubular epithelium in PKD mice.
figure 5

a Representative immunofluorescence images (left) and length of primary cilia in tubular and cystic epithelial cells in kidneys of PKD mice treated without (Ctrl) or with 1-Indanone (Ind) (right). Bar = 20 μm. n > 60, in three mice of each group. Data are presented by mean values ± SEM. Data are analyzed with one-way ANOVA. ***P < 0.001 vs. Ctrl group in WT mice, ###P < 0.001 vs. Ctrl group in PKD mice. b Representative TEM images of primary cilia in PKD mice treated without or with Ind. Bar = 1 μm. c Representative SEM images of primary cilia in each group. Bar = 1 μm.

Full size image

To confirm the effect of 1-Indanone on primary cilia length, we stimulated mIMCD cells with cAMP agonist FSK and detected primary cilia by immunofluorescence staining with ARL13B. Consistent with the results of PKD mice, after stimulation with 100 μM FSK for 3 h, the primary cilia of mIMCD cells were prolonged excessively, even reaching 2-fold of those in the ctrl group (Fig. 6a). Studies have confirmed that activation of mTORC1 or PKA will promote the elongation of primary cilia [36, 37], therefore mTORC1 inhibitor rapamycin and PKA inhibitor H-89 were used as positive controls in this study. It was found that 1-Indanone remarkably inhibited the excessive elongation of cilia by cAMP activation in mIMCD cells, similar to the effect of rapamycin (Fig. 6a) or H-89 (Fig. 6b). There was no significant difference in the length of primary cilia in mIMCD cells without and with 1-Indanone treatment (Fig. 6a, b). In addition, we detected the effect of 1-Indanone on the ciliary basal body by gamma-tubulin immunofluorescence staining. There was no difference in the structure of the ciliary basal body between 1-Indanone-treated group and the other groups in mIMCD cells (Supplementary Fig. S3), which indicated ciliary structural integrity was not disrupted by 1-Indanone.

Fig. 6: 1-Indanone retards primary cilia prolongation with FSK stimulation in mIMCD cells.
figure 6

a Representative image of primary cilium and quantification results of cilium length in mIMCD cells (Ctrl) or forskolin stimulated (FSK, 100 μM) mIMCD cells incubated with 1-Indanone (Ind, 25 μM) or rapamycin (Rapa, 20 nM), respectively. n = 39–98 in each group. Bar = 10 μm. ***P < 0.001. b Representative image of primary cilium and quantification results of cilium length in mIMCD ctrl cells or FSK-stimulated mIMCD cells incubated with 1-Indanone (Ind, 25 μM) or H-89 (10 μM), respectively. n > 60 in each group. Bar = 10 μm. Data are presented by mean values ± SEM. ***P < 0.001. Data are analyzed with one-way ANOVA.

Full size image

These results demonstrated that 1-Indanone inhibited the elongation of the primary cilia caused by activation of PKA and cAMP in tubular epithelial cells.

1-Indanone inhibited excessive elongation of primary cilia in PKD by stabilizing tubulin

As the principal component of ciliated axoneme, microtubule dynamics directly modulate the cilia assembly [38]. Therefore, we explored whether modulating cilia length by 1-Indanone is associated with changes in tubulin dynamics. Virtual docking model showed 1-Indanone forms a hydrogen bond with arginine 359 in the paclitaxel binding site of tubulin (PDB ID: 6EW0) (Fig. 7a). Immunofluorescence staining results showed 25 μM of 1-Indanone promoted tubulin accumulation around the nucleus (Fig. 7b). We further detected the effect of 1-Indanone on tubulin polymerization. The absorbance at 340 nm of tubulin showed 25 μM of 1-Indanone promoted tubulin polymerization (Fig. 7c). These results indicated that 1-Indanone inhibited the abnormal elongation of primary cilia through tubulin stabilization. As the key component of the spindle apparatus, microtubule participates in mitosis regulating the cell cycle [39]. Therefore, we tested the effect of 1-Indanone on the cell cycle of mIMCD cells with or without FSK stimulation. It was found that FSK stimulation increased the number of mIMCD cells in G0/G1 phases, and the cells in S and G2 phases were decreased compared with ctrl group. 1-Indanone inhibited the increased FSK-treated mIMCD cells in G0/G1 phases, but had no obvious effect on the cell cycle of those in the ctrl group (at least at the concentration we tested) (Supplementary Fig. S4).

Fig. 7: 1-Indanone modulates the structure of primary cilia by stabilizing tubulin.
figure 7

a Virtual docking results showing 1-Indanone formed a hydrogen bond with arginine at 359 in the paclitaxel binding site of tubulin (PDB ID: 6EW0). b Representative immunofluorescence images of α-tubulin in mIMCD cells (Ctrl) and FSK-treated mIMCD cells (FSK) incubated with 10 μM of paclitaxel (FSK+Taxol) or 25 μM of 1-Indanone (FSK+Ind). Bar = 20 μm. c The absorbance curves of tubulin treated with 1-Indanone (25 μM) incubated under 37 °C for 65 min at 340 nm.

Full size image

1-Indanone down-regulated the anterograde transport of primary cilia in PKD mice

Cilia assembly and disassembly were determined by anterograde or retrograde transport process. The elongation of cilia relies on the anterograde intraflagellar process regulated by IFT-B and kinesin II [40]. Therefore, we detected the expressions of critical components of IFT-B and kinesin II, IFT88, KIF3A and KIF3B. The results showed that the expression of IFT88 was up-regulated impressively in PKD kidneys, which was retarded by 1-Indanone (Fig. 8a). And 1-Indanone notably reversed the abnormal up-regulation of both KIF3A and KIF3B in PKD kidneys (Fig. 8b, c), demonstrating 1-Indanone affected cilium assembly through down-regulating ciliary anterograde IFT progress.

Fig. 8: 1-Indanone down-regulates anterograde transport of primary cilia.
figure 8

a Representative Western blots and relative protein expression of IFT88 in the kidneys of wild-type (WT) and PKD mice without (Ctrl) or with 1-Indanone (Ind) treatment. n = 11. b Representative Western blots and relative protein expression of KIF3A. c Representative Western blots and relative and protein expression of KIF3B. n ≥ 6. d Representative Western blots and relative protein expression of IFT20 and KIF17. n = 5–6. All data are presented by mean values ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001 vs. Ctrl group in WT mice, #P < 0.05, ##P < 0.01, ###P < 0.001 vs. Ctrl group in PKD mice.

Full size image

We further detected the expression of IFT20, another subunit of IFT-B, which regulates primary cilia growth as well [41]. The results showed consistent with IFT88, the expression of IFT20 was markedly increased in PKD kidneys and was inhibited by 1-Indanone. Another motor protein in kinesin II, KIF17, has been reported to alter the dynamics of microtubules by promoting its polymerization. We found that the expression of KIF17 was reduced in PKD, and 1-Indanone had no notable effect on its expression (Fig. 8d).

The ciliary membrane protein Arl13b was reported to regulate cilia length by inducing cilia protrusion [42, 43]. And acetylated tubulin modulates cilia length by regulating cilia axoneme extension [44]. However, there was no notable difference in the expression of ARL13B and acetylated tubulin among each group, suggested that 1-Indanone does not affect the extension of ciliary membranes and axoneme by affecting the modification of microtubule (Supplementary Fig. S5).

1-Indanone down-regulated the ciliary coordinated signaling pathways

Above results have confirmed that 1-Indanone modulates the primary cilium structure, however it is unclear whether 1-Indanone affects the function of cilia signal transduction. Various signaling pathways such as Hedgehog, Wnt, PDGF, and mTOR have been reported to be related to primary cilia [45, 46]. First, we detected the expressions of GLI family members in the Hedgehog signaling pathway. The expressions of GLI1, 2, 3 increased remarkably in the PKD kidneys, and 1-Indanone down-regulated the over-activated molecules in Hedgehog signaling (Fig. 9a–c). Wnt/β-catenin is another signaling pathway closely related to primary cilia [47, 48]. Our results showed that the expressions of active (non-phosphorylated) β-catenin, total β-catenin and its downstream molecules c-myc and cyclin D1 were up-regulated in PKD kidneys. 1-Indanone notably reversed the increased expression of active β-catenin, total β-catenin and cyclin D1. The phosphorylation level of gsk-3β increased markedly in PKD, and was also retarded by 1-Indanone. It was confirmed that Wnt signal was activated in PKD and returned to normal level under the treatment of 1-Indanone. These data indicated that 1-Indanone suppressed the up-regulation of ciliary coordinated signaling pathways in PKD (Fig. 9d).

Fig. 9: 1-Indanone down-regulates signaling pathways coordinated with primary cilia.
figure 9

a Representative Western blots and relative protein expression of GLI1 in wild-type (WT) and PKD kidneys treated without (Ctrl) or with 1-Indanone (Ind). n ≥ 6. b Representative Western blots and relative protein expression of GLI2. n ≥ 6. c Representative Western blots and relative protein expression of GLI3. n ≥ 6. d Representative Western blots and relative protein expression of signature proteins in canonical Wnt signaling. n ≥ 6. Data are presented by mean values ± SEM. *P < 0.05, ***P < 0.001 vs. Ctrl group in WT mice, #P < 0.05, ##P < 0.01, ###P < 0.001 vs. Ctrl group in PKD mice.

Full size image

Discussion

The goals of this study were to find active compounds with efficacy to retard cyst development in ADPKD from the natural compounds with anti-tumor activity and to determine the pharmacological mechanisms. We used distinct in vitro and in vivo experimental models of PKD to obtain proof of natural products as PKD therapeutics. By screening candidate compounds that strongly inhibited cyst formation and growth in vivo and in vitro, without cytotoxicity, compound 1-Indanone was found to have the best efficacy in inhibiting cyst development.

Previous studies have found that some natural products with anti-tumor activity, such as Curcumin, Ganoderma triterpenes and Cardamonin [49,50,51], slowed down the development of renal cyst in ADPKD. These data suggest that the natural compounds with anti-tumor activity might slow down the development of renal cyst in ADPKD. In this study, based on the inhibitory effect on the signaling pathways modulating cellular abnormal proliferation both in tumor cells and renal cystic cells, we found an active compound 1-Indanone from the natural product library. 1-Indanone is isolated from marine cyanobacteria Lyngbya majuscule [52, 53]. Indanone analogs possess a variety of biological activities, which include anti-tumor, anti-Alzheimer’s disease, anti-inflammation, anti-bacteria and anti-malaria. Previous studies reported that 1-Indanone inhibited the growth of various tumor cell lines by suppressing cell proliferation, arresting cell cycle in G2/M phases [54]. In this study, it was found that 1-Indanone had a considerable inhibitory activity on the renal cyst development. In both in vitro MDCK cyst model and ex vitro embryonic kidney model, 1-Indanone reversibly inhibited the formation and growth of cysts in a dose-dependent manner. The experiments with Pkd1 knockout mice confirmed that 1-Indanone markedly retarded the growth of renal cysts in vivo. Notably, our study found that 1-Indanone normalized the abnormal elongated primary cilia of cystic epithelial cells.

It is well established that primary cilium has a critical role in maintaining the normal function of the kidney. The primary cilium functions as a mechano-sensor in renal tubular epithelial cells to convert mechanical sensation into intracellular chemical signals [55]. Abnormalities of primary cilia of renal epithelial cells have been observed in various kidney diseases, such as acute tubular necrosis and unilateral nephrectomy [56,57,58]. Ciliopathies, such as Bardet-Biedl (BBS), Joubert (JBTS), Meckel-Gruber (MKS), are often accompanied by cystic kidney and defects of cilia biogenesis and ciliary assembly genes will lead to ADPKD, suggesting that defect in primary cilium is an important trigger of renal cyst development [22, 59,60,61].

However, the relationship between primary cilia length and renal cysts formation in PKD remains elusive. In PKD, cyst formation is driven by cAMP-mediated hyperproliferation of cystic epithelial cells [34]. It is reported that the activation of cAMP or PKA will conspicuously prolong primary cilia [62]. And studies have shown that down-regulating TSC1/TSC2, negative regulators of mTORC1 promoted the elongation of cilia [36] and administration with mTORC1 inhibitor rapamycin reversed the increase of primary cilia length [63]. Therefore, it indicated in our study that the over-activated mTOR signaling in ADPKD renal tubular epithelial cells may contribute to the abnormalities of primary cilia length. Conversely, defects of primary cilia structure will promote cyst expansion in ADPKD, suggesting there is a regulatory loop between primary cilia and cellular proliferation. In this study, 1-Indanone inhibited the excessive elongation of primary cilia in cystic epithelial cells. Simultaneously, 1-Indanone suppressed the activation of the cAMP/PKA, and down-regulated mTOR signaling pathways and retarded the resulting cellular hyperproliferation. Thus, 1-Indanone broke the regulatory loop between the hyperproliferation and primary cilia elongation in renal cystic epithelial cells.

The assembly and elongation of cilia depend on the anterograde or retrograde transport of component proteins. Previous studies showed that inhibiting efficiently the expression of primary cilia-related proteins KIF3A, IFT88, IFT20 in Pkd1 or Pkd2 knockout mice suppressed renal cyst progression [24, 64], suggesting modulating the transport process of the primary cilia component proteins may be a therapeutic approach for ADPKD. Our study found that the expressions of IFT88, IFT20, KIF3A and KIF3B were notably up-regulated in PKD mouse kidney, which were down-regulated by 1-Indanone. The experimental results indicate that 1-Indanone reduced the abnormal cilium length by affecting primary cilium assembly. The cell cycle is another factor that closely affects the biogenesis and assembly of primary cilia. The primary cilia are assembled in the G0/G1 phase and reabsorbed into the cell entering S phase [65]. Recently, several studies have illustrated that the cell cycle-dependent kinases that modulate the G1-S cell cycle transition promote the cyst formation in ADPKD [66,67,68]. In this study, we verified that with FSK stimulation, the cells in G0/G1 phases were increased remarkably compared with ctrl group, which indicated except for excessively increase in cilia length, ciliogenesis in cystic epithelial cells was also elevated. 1-Indanone inhibited the increase in G0/G1 phase cell number induced by FSK treatment, illustrating that 1-Indanone may inhibit ciliogenesis in cystic epithelial cells by cell cycle arresting.

The primary cilium is a microtubule-based organelle. Cilium assembly depends on importing tubulin to cilium axoneme and tubulin at cilium tip. It has been reported the microtubule stabilizer strikingly reduced the length of primary cilia [69], and retarded renal cysts expansion in PKD [70]. Our previous target prediction for 1-Indanone from “Similarity Ensemble Approach server database” indicated that 1-Indanone might bind to the heavy chain of α-tubulin. Our docking results showed 1-Indanone formed a hydrogen bond with arginine at 359 in the paclitaxel binding site of tubulin, which suggests that 1-Indanone might promote tubulin polymerization. In addition, we found that the expression of KIF17, another component of kinesin, is reduced in ADPKD. KIF17 has been reported to stabilize tubulin. Knocking out KIF17 in vitro leads to the formation of cysts [71]. All these data suggest that 1-Indanone inhibited the extension of primary cilia by down-regulating the anterograde transport of soluble tubulin along the cilia axoneme.

The primary cilium is considered as a hub of signaling transduction. A variety of oncogenic signaling coordinated with primary cilia, such as hedgehog and Wnt/β-catenin signaling pathways [72], both of which facilitate cyst development [73, 74]. It has been reported that the hedgehog signaling transduction relies on primary cilia [75, 76]. In this study, we found three GLI family members were up-regulated impressively in the kidneys of PKD mice, which were rescued by 1-Indanone. Besides, it is well elaborated that primary cilium played an important role in Wnt signaling transduction, and defects in cilia lead to over-activation of Wnt signaling [77, 78]. In our study, it was determined that 1-Indanone suppressed the activation of β-catenin and the increase of its downstream effector cyclin-D1 in PKD mice, which suggested 1-Indanone might influence cyst initiation regulated by β-catenin.

However, this study still has many limitations. First, whether there is a direct regulatory relationship between cytoskeletal dynamics and signaling pathways co-transduced by primary cilia remains to be investigated. In addition, the velocity of primary ciliary transport is also an important indicator for regulating ciliary biogenesis and assembly. However, due to limited experimental conditions, we did not detect it. ADPKD is a slowly progressive disease. The Pkd1 knockout mice used in this study is a rapidly progressive model and can not accurately simulate the process of human ADPKD. We intend to examine the effect of 1-Indanone on primary cilia length as the cyst progresses using the slowly progressive ADPKD model in the future. Besides, it is also planned to examine the specific expression of hedgehog signaling and Wnt signaling effector molecules in cilia, but precise measurements may be difficult due to the limitations of current technology. Moreover, we may examine the pharmacokinetic characteristics of 1-Indanone by oral administration and optimize its chemical structure if necessary.

In conclusion, in this study we found the potential of 1-Indanone suppressing renal cyst development through in vitro and in vivo ADPKD models. 1-Indanone inhibited the abnormal elongation of primary cilia by stabilizing tubulin and decreasing anterograde transporting soluble tubulin to cilia axoneme, and suppressed the over-activation of ciliary coordinated hedgehog and Wnt signaling pathways. Therefore, 1-Indanone inhibited the ciliary defects in signal transduction and fluid sensation caused by alterations in the primary cilia structure. Thereby, 1-Indanone down-regulated the over-activation of cAMP/PKA and downstream MAPK and mTOR signaling pathways and regulated excessive proliferation in cystic epithelial cells. As a result, 1-Indanone broke the regulation loop between excessive cell proliferation and alteration in primary cilia structure. Eventually, 1-Indanone retarded renal cysts growth and delayed ADPKD progression (Fig. 10). Our data indicate that normalizing the length of primary cilia in cystic epithelial cells is a promising therapeutic strategy for ADPKD and 1-Indanone is a potential therapeutic candidate for ADPKD treatment.

Fig. 10: The diagram of mechanism in which 1-Indanone delays cyst development in ADPKD.
figure 10

1-Indanone inhibits the abnormal elongation of cystic epithelial cilia via blocking soluble tubulin transport to cilium axoneme, which down-regulates Wnt/β-catenin, hedgehog, Ras/MAPK and mTOR signaling pathways, eventually ameliorates the epithelial cell proliferation of renal cysts in ADPKD.

Full size image