Cilia-enriched oxysterol 7β,27-DHC is required for polycystin ion channel activation

Abstract

Polycystin-1 (PC-1) and PC-2 form a heteromeric ion channel complex that is abundantly expressed in primary cilia of renal epithelial cells. This complex functions as a non-selective cation channel, and mutations within the polycystin complex cause autosomal dominant polycystic kidney disease (ADPKD). The spatial and temporal regulation of the polycystin complex within the ciliary membrane remains poorly understood. Using both whole-cell and ciliary patch-clamp recordings, we identify a cilia-enriched oxysterol, 7β,27-dihydroxycholesterol (DHC), that serves as a necessary activator of the polycystin complex. We further identify an oxysterol-binding pocket within PC-2 and showed that mutations within this binding pocket disrupt 7β,27-DHC-dependent polycystin activation. Pharmacologic and genetic inhibition of oxysterol synthesis reduces channel activity in primary cilia. In summary, our findings reveal a regulator of the polycystin complex. This oxysterol-binding pocket in PC-2 may provide a specific target for potential ADPKD therapeutics.

Fig. 1. 7β,27-DHC activates the polycystin complex on the plasma membrane.

A Strategy to express the polycystin complex in the plasma membrane. The black dotted arrow indicates ciliary trafficking of PC-1 with endogenous signal peptide and PC-2. Red arrow indicates redirected plasma membrane trafficking using a IgG kappa-derived secretion peptide on PC-1. B Immunofluorescent staining of HEK293 and IMCD-3 cells overexpressing the polycystin complex. Surface HA staining indicates that polycystin complex accumulates in plasma membrane of HEK cells (Top) or cilia (IMCD-3 cells, bottom). Green, surface HA; Red, total HA or cilia marker Arl13b; Magenta, PC-2; Images are representative for at least three independent experiments. scale bar = 10 µM. C Structure of 7β,27-dihydrocholesterol (DHC). D Whole-cell patch-clamp traces recorded from parental HEK293 cells or overexpressing sPC-1/2, PC-1 or PC-2. Voltage step pulse from −100 mV to +180 mV in +20 mV increments with 0 mV holding potential and 5 μM 7β,27-DHC in the pipette solution. E, F Current (I)–voltage (V) relationship and current density for HEK293 cells overexpressing sPC-1/2, PC-1 or PC-2. Each value was obtained at the maximum current (pA) from −100 mV to +180 mV and divided by capacitance (pF); sPC-1/2 (red, n = 16), PC-1 (orange, n = 12), PC-2 (green, n = 10), and HEK293 (n = 15). F Quantification of current density I (pA/pF) at +180 mV. Two-tailed unpaired Student’s t-test *P < 0.05. G Effect of extracellular application of 7β,27-DHC on polycystin channel activation. 5 μM 7β,27-DHC was applied in bath solution. Ramp pulse applied from −100 mV to +100 mV with 500 ms duration. Each value (pA) at −100 mV (gray square) and +100 mV (gray blank square) is divided by capacitance (pF) and plotted over time. Orange line indicates holding potential (0 mV). H, I Extracellular NMDG inhibits 7β,27-DHC evoked PC−1/2 currents. Blank and black squares indicate the current density (pA/pF) obtained at +100 mV and −100 mV, respectively. i Left. I–V relationship of polycystin currents in presence of Na⁺ (red) and NMDG (gray). Blue insert shows reversal potential shift as the extracellular solution changes from Na⁺ to NMDG⁺. Right. (E) I (pA/pF) at +100 mV and −100 mV before (red) and after (gray) NMDG perfusion. n = 5. Two-tailed paired Student’s t-test, *P < 0.05. Error bars: mean ± SEM.

Fig. 2. Higher concentration of 7β,27-DHC treatment increase open probability of sPC-1/PC-2 inward currents.

A Representative inside-out single channel recordings of cells overexpressing sPC-1/2 with intracellular application of 1 μM to 50 μM 7β,27-DHC. Red dotted lines indicate closed (C) and open (O) states of a single channel. B Single-channel conductance after intracellular application of 1 μM to 50 μM 7β,27-DHC. green: 1 μM (n = 3), blue: 2.5 μM (n = 3), black: 5 μM (n = 4) and orange: 50 μM (n = 4) C Normalized absolute open probability of sPC-1/2 obtained from −100 mV to +100 mV during 10 s of recording. The absolute open probability was normalized to the maximum open probability. 1 μM (n = 3), 2.5 μM (n = 3), 5 μM (n = 4) 50 μM (n = 4). Two-tailed unpaired Student’s t-test, *P < 0.05. Error bars: mean ± SEM.

Fig. 3. 7β,27-DHC binds to PC-2 subunit.

A Other oxysterols closely related 7β,27-DHC. Molecular structures of three oxysterols, 7β−HC, 7α,27-DHC, and 7α− HC closely related to 7β,27-DHC. Hydroxyl groups colored in red indicate unique hallmarks in each structure. B Current density from whole-cell recordings expressing sPC-1/2. 5 μM 7β, 7α,27-DHC, and 7α were included in the intracellular recording solution. Current density is plotted at +180 mV. 7β,27 (n = 16), 7α,27 (n = 13) 7β (n = 7) 7α (n = 10), Two-tailed unpaired Student’s t-test, **P < 0.01. Data are presented as mean values ± SEM. C Molecular dynamics simulations of oxysterol binding within the PC-2 structure (PDB: 5T4D). The top image shows the system set-up with eight 7β,27-DHC molecules placed in solution (cyan spheres), above the upper leaflet and below the lower leaflet. 7β,27-DHC molecules are free to bind to either extracellular or intracellular leaflet of PC-2 during simulation. Bottom image represents a binding event of one 7β,27-DHC molecule (orange spheres) to the intracellular leaflet of PC-2. Protein is represented as cartoon, and phosphate heads of lipids as yellow spheres to show membrane boundaries. D Snapshots of 7β,27-DHC binding mode during molecular dynamics simulations. Once one 7β,27-DHC molecule binds to the intracellular region between pre-S1 and the S4-S5 linker, it stably interacts with E208 and R581 throughout 1.4 μs of simulation time. The protein is shown as white cartoon, 7β,27-DHC as orange/red sticks and residues within 4 Å as cyan/red/blue sticks.

Fig. 4. E208 and R581 within PC-2 are key amino acids required for 7β,27-DHC-dependent activation.

A, B Mutation of E208A or R581A does not affect plasma membrane or ciliary trafficking of the polycystin complex. Live cell immunostaining of HEK cells (top: DOX-induced, bottom: DOX-uninduced) stably expressing sPC-1/PC-2, sPC-1/PC-2_E208A or sPC-1/PC-2_R581A using an anti-HA antibody (green), indicative of sPC-1 trafficking. Data represent one intensity value per image. sPC-1/PC-2 (+Dox: n = 3, −Dox: n = 1), sPC-1/PC-2_E208A (+Dox: n = 5, −Dox: n = 1), sPC-1/PC-2_R581A (+Dox: n = 5, −Dox: n = 1). Two-tailed unpaired student’s T-test. C sPC-1/PC-2 E208A or R581A are insensitive to 7β,27-DHC-dependent activation. Whole-cell patch-clamp recordings of HEK cells stably expressing sPC-1/PC-2, sPC-1/PC-2_E208A or sPC-1/PC-2_R581A and respective double mutant with additional GOF F604P. The same voltage step pulse protocol from Fig. 1D was applied to test binding mutants. sPC-1/2_E208A and sPC-1/2_R581A were tested with 5 μM 7β,27-DHC added to the intracellular solution, while the binding mutants with F604P (sPC-1/_2E208A-F604P and sPC-1/2_R581A-F604P) were tested without 5 μM 7β,27-DHC. D Comparison of current density at + 180 mV from whole-cell patch-clamp recordings of the two oxysterol-binding mutants. The current amplitudes at +180 mV with E208A (green, n = 10) and R581A (yellow, n = 11) were compared to those of the double mutants, E208A-F604P (green, n = 6) and R581A-F604P (yellow, n = 5). Two-tailed unpaired Student’s t-test *P < 0.05, ***P < 0.0001. E, F Immunofluorescent imaging analysis of ciliary expression of sPC-1 with PC-2_E208A or PC-2_R581A. Ciliary localization of sPC-1 co-expressed with PC-2, PC-2_E208A and PC-2_R581A was confirmed using anti-PC-2 (red) and anti-acetylated tubulin antibodies (cilia marker, green). Each n indicates the number of cilia for each group. PC-2 (n = 328), PC-2_E208A (n = 289) and PC-2_R581A (n = 374). **P < 0.005 between the indicated groups tested by One-way ANOVA. G Representative excised ciliary single channel recordings. Single channel recordings of parent PC-2 knockout IMCD-3 cells or overexpressing sPC-1 with PC-2, PC-2_E208A or PC-2_R581A without intracellular 7β,27-DHC application are shown at +100 mV and −100 mV. Error bar, mean ± SEM.

Fig. 5. 7β,27-DHC potentiates ciliary polycystin-mediated currents.

A Representative excised inside-out single channel recordings of primary cilia in IMCD-3 cell without (left) and with 5 μM 7β,27-DHC (right). Single channel recordings were obtained between +100 mV and −100 mV at 20 mV increments. Red dotted line indicates the close state of the channel. B Representative image of ARL13B-GFP expressing IMCD-3 cells. Primary cilia (green) form a giga-ohm seal with glass electrode (right). C Normalized absolute open probability of ciliary channels. P_open of ciliary channels from IMCD-3 cells without (n = 4) and with 5 μM 7β,27-DHC (n = 4) D Single-channel conductance of ciliary channel recordings shown in 3 Å. Black and red squares indicate the averaged current amplitudes of ciliary channels from IMCD-3 cells without (n = 4) or with 5 μM 7β,27-DHC (n = 4) application, respectively. Dotted line indicates the fitting to the linear equation. Bar graph insert shows comparison of ciliary channel conductance without or with 5 μM 7β,27-DHC. Error bars: mean ± SEM.

Fig. 6. 7β,27-DHC synthesis is critical for basal activity of ciliary polycystin channels.

A Schematic pathway of 7β,27-DHC synthesis. Schematic diagram illustrating key role of β1/2-hydroxysteroid dehydrogenase type 11 (β1/2-HSD11) enzymes for conversion of 7-keto and 27-HC to 7β,27-DHC. Carbenoloxone (CNX) inhibits enzymatic activity of β2-HSD11. B Excised ciliary single-channel recording of IMCD-3 cells after 72-h CNX incubation. Holding potentials were given from +100 mV to −100 mV. Red dotted line indicates closed state. C Model of CNX-dependent polycystin inhibition. D Ciliary single-channel recordings after β1/2-HSD11 knockdown. IMCD-3 cells were transfected with scrambled siRNA control (top), β2-HSD11-siRNA (middle), and β1-HSD11-siRNA (bottom). Knockdown efficiency was 95 ± 7% for β2-HSD11 and 90 ± 5% for β1-HSD11. Single channels with recorded in ciliary excised patches at +100 mV and −100 mV. E Quantification of open probability of ciliary recordings shown in D. Open events of single-channel recordings were counted at +100 mV and −100 mV. Scr (scrambled) n = 5, B1 (β1-HSD11) n = 21, B2 (β2-HSD11) n = 23. Error bars, mean ± SEM.