Polycystin-2 (PC2) is a key determinant of in vitro myogenesis

Abstract

The development of skeletal muscle (myogenesis) is a well-orchestrated process where myoblasts withdraw from the cell cycle and differentiate into myotubes. Signaling by fluxes in intracellular calcium (Ca²⁺) is known to contribute to myogenesis, and increased mitochondrial biogenesis is required to meet the metabolic demand of mature myotubes. However, gaps remain in the understanding of how intracellular Ca²⁺ signals can govern myogenesis. Polycystin-2 (PC2 or TRPP1) is a nonselective cation channel permeable to Ca²⁺. It can interact with intracellular calcium channels to control Ca²⁺ release and concurrently modulates mitochondrial function and remodeling. Due to these features, we hypothesized that PC2 is a central protein in mediating both the intracellular Ca²⁺ responses and mitochondrial changes seen in myogenesis. To test this hypothesis, we created CRISPR/Cas9 knockout (KO) C2C12 murine myoblast cell lines. PC2 KO cells were unable to differentiate into myotubes, had impaired spontaneous Ca²⁺ oscillations, and did not develop depolarization-evoked Ca²⁺ transients. The autophagic-associated pathway beclin-1 was downregulated in PC2 KO cells, and direct activation of the autophagic pathway resulted in decreased mitochondrial remodeling. Re-expression of full-length PC2, but not a calcium channel dead pathologic mutant, restored the differentiation phenotype and increased the expression of mitochondrial proteins. Our results establish that PC2 is a novel regulator of in vitro myogenesis by integrating PC2-dependent Ca²⁺ signals and metabolic pathways.

INTRODUCTION

The development of skeletal muscle (myogenesis) is a multistep process where pluripotent mesodermal cells (satellite cells) give rise to myoblasts that subsequently withdraw from the cell cycle and differentiate into myotubes. This process of cell cycle withdrawal and initiation of differentiation is largely regulated by myogenic regulatory factors like MyoD and myogenin (1, 2). The end-terminal stage of differentiation and cellular fusion coincides with the expression of myosin heavy chains, muscle-specific proteins, which mark sarcomeric assembly (3, 4). Simultaneously, mitochondrial biogenesis increases, enabling differentiating cells to meet the increased energetic demand. Mitophagy, a mitochondrial-specific autophagic process, has been shown to coordinate with mitochondrial biogenesis and expand mitochondrial mass through the master regulator PGC1α (5–7).

Intracellular calcium (Ca²⁺) signals are mediated by a variety of receptors, including the inositol trisphosphate receptor (InsP₃R; 8), and store-operated Ca²⁺ entry (SOCE) via stromal interaction molecule 1 (STIM1) and Orai (9–14) have been shown to activate downstream pathways involved in the expression of myogenic regulatory factors, autophagy and mitochondrial biogenesis (15). In addition, plasma membrane Ca²⁺ entry via members of the transient receptor potential (TRP) channels also contributes to myogenesis (9, 16–18). TRP channels have well-established functional roles in muscle physiology and pathology, with members of the canonical TRP (TRPC) family implicated in altered Ca²⁺ signaling associated with several muscle diseases (17, 19–21).

However, the function of TRP channels on the endoplasmic reticulum (ER) in regulating Ca²⁺ signaling for myogenesis has not been studied. We and others have found that the TRP channel polycystin-2 (PC2, TRPP1, gene name PKD2) is located on the endoplasmic reticulum (ER) or sarcoplasmic reticulum (SR) in muscle cells. PC2 can interact with other Ca²⁺ release channels in the ER/SR, like InsP₃R and ryanodine receptor (RyR), and modify intracellular Ca²⁺ release (22–24). Most studies of PC2 are focused on its function in the kidney, as mutations cause autosomal dominant polycystic kidney disease (ADPKD). However, expression of PC2 in cardiac, skeletal, and smooth muscle suggests that PC2 may function in maintaining [Ca²⁺] in tissues other than the kidney (25). For example, PC2 is essential during development (26) and for left-right axis determination (27–29). PC2 also contributes to autophagy (22, 30), regulation of muscle atrophy (31), and mitochondrial dynamics (32). Although PC2 is associated with developmental pathways, the requirement for PC2 in the process of myoblast differentiation has not been examined.

The goal of this study was to determine the role of PC2 in C2C12 in vitro myogenesis. Murine C2C12 myoblasts are a well-characterized myogenic cell line able to recapitulate myogenesis in vitro (33). When cultured in a medium containing low concentrations of mitogens, proliferating C2C12 myoblasts withdraw from the cell cycle to express muscle-specific genes and fuse into mature myotubes (34). We hypothesized that PC2 is a key central protein in regulating both Ca²⁺ signaling and mitochondrial changes seen in myogenesis. Knockout (KO) of PC2 in C2C12 myoblast cells inhibited myogenesis by impairment of spontaneous Ca²⁺ oscillations and preventing downstream mitochondrial biogenesis pathways. Re-expression of full-length PC2, but not a calcium impermeant pathogenic mutant, restored the differentiation phenotype and expression of mitochondrial proteins. Our results establish a novel and unique role of PC2 in myogenesis.

MATERIALS AND METHODS

C2C12 Mouse Myoblasts and Differentiation

Murine C2C12 myoblasts were purchased from American Type Culture Collection (ATCC), and cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS; growth medium, GM) and antibiotics at 37°C in a humidified atmosphere with 5% CO₂ and 95% air. Cell cultures were kept to passages no higher than P12. Differentiation was induced when C2C12 myoblast cells were fully confluent by replacing GM with DMEM supplemented with 2% FBS (differentiation medium, DM) for 7 days. The differentiation medium was replaced every 24–48 h.

Generation of PC2 CRISPR KO Cell Lines

We tested three different CRISPR/Cas9 knockout all-in-one ZsGreen pClip lentivirus plasmids each containing a separate guide sequence directed at different mouse Pkd2 loci, along with a control (CTL) template (Transomic Technologies). The lentiviruses were made by cotransfecting pRSV, pMDLg, and pMD2.G along with pClip into HEK293T cells. The supernatant containing virus particles was harvested and used to transduce C2C12 cells. Following 48 h of transduction, ZsGreen fluorescent C2C12 cells were sorted by flow cytometry and single-cell clones expanded. Following expansion, multiple clones of each guide sequence were tested for the PC2 knockout by qPCR and Western blot. Only two guide sequences, starting at base numbers 1088553 and 1155695 were effective in knocking out PC2, and two single clones (“53a” and “95”) were selected for further study.

Ca²⁺ Transient Measurements

To measure intracellular calcium [Ca²⁺]_i, we used the calcium indicator Fluo-4 AM (Molecular Probes/Invitrogen, Carlsbad, CA). Cells grown on a laminin-coated coverslip were incubated at 37°C with 2.2 μM Fluo-4 AM for 30 min in Tyrode’s solution (in mM: 140 NaCl, 4 KCl, 1.8 CaCl₂, 1 MgCl₂, 10 glucose, 10 HEPES, pH 7.4), followed by a 5-min wash in Tyrode’s solution. For ATP experiments, cells were stimulated with 10 µM ATP. For depolarization evoked experiments, electrical field stimulation was achieved by using a pair of platinum electrodes 1 cm distance apart, which were connected to a Grass stimulator (Astro-Med, Inc.) set at 40 V and a frequency of 0.5 Hz. For calcium entry experiments, zero Ca²⁺ Tyrode’s solution was used (in mM: 140 NaCl, 4 KCl, 2 MgCl₂, 5 EGTA, 10 glucose, 10 HEPES, pH 7.4). Cells were stimulated with 1 µM thapsigargin, and then 1.8 mM calcium was added back. Fluo-4 was excited with a 488 nm light emiting diode (LED) (Lumencor Spectra X lamp) and emitted fluorescence filtered with a band pass filter (515–530 nm, Chroma). Resultant fluorescent images were acquired with a sCMOS camera (Orca Flash, Hamamatsu) on a wide-field fluorescence Zeiss microscope, equipped. Images and videos were recorded using Zen Blue software (Zeiss, Germany). Fluorescence intensity was measured as the ratio of the fluorescence (F) to the basal fluorescence (F0).

Western Blot Analysis

Total protein extracts were prepared by lysing C2C12 cells with radioimmunoprecipitation assay (RIPA) buffer (in mM: 10 Tris-Cl, 1 EDTA, 0.5 EGTA, 1% Triton-X, 0.1% sodium deoxycholate, 0.1% SDS, 140 NaCl) containing protease inhibitor cocktail (Sigma-Aldrich), phosphatase inhibitors NaF, and sodium orthovanadate (Alfa Aesar). Protein concentrations of the resulting supernatants were measured with a Pierce bicinchoninic acid (BCA) protein assay kit (Thermo Fisher Scientific). Equal amounts of protein (15–30 µg) were separated by SDS-PAGE (Bio-Rad, 4%–20% gradient gels) and transferred to PVDF membranes via wet transfer. Validation of the PC2 D-3 sc-28331 Santa Cruz antibody was established by comparing the expected molecular weight (∼110 kDa) and loss of reactivity in the PC2-KO cell lines. Commercially purchased antibodies were chosen based on available published literature and compared with the expected molecular weight. Primary antibodies for α-tubulin (1:1,000, No. 2125S; Cell Signaling Technology), GAPDH (1:1,000, No. 6004-Ig; Proteintech), PC2 (D-3, 1:500, sc-28331; Santa Cruz Biotechnology), beclin-1 (D40C5, 1:500, No. 3495S; Cell signaling Technology), cytochrome C (Cyt. C; 136F3, 1:250, No. 4280S; Cell Signaling Technology), PINK1 (D8G3, 1:750, No. 6946S, Cell Signaling Technology), LC3A/B (D3U4C, 1:1,000, No. 12741S; Cell Signaling Technology), MFN2 (mitofusin 2; D2D10, 1:1,000, No. 9482S, Cell Signaling Technology), parkin (Prk8, 1:500, No. 2132S, Cell Signaling Technology), RyR1/RyR2 (34C, 1:500, Developmental Studies Hybridoma Bank), InsP₃R1 [T1C, 1:500, affinity-purified peptide-derived antibody, gift of Dr. G. Mignery, directed to the COOH terminus of the receptor (35)] were applied to the membranes. Horseradish peroxidase (HRP)-conjugated secondary antibodies were applied (Immun-Star goat anti-mouse, 1:20,000, 1705047 and Immun-Star goat anti-rabbit, 1:20,000, 1705046, Biorad), and then activated with Clarity Max Western ECL (Bio-Rad). Chemiluminescence was imaged with a ChemiDoc MP imager (Bio-Rad); signal intensity of each protein was measured with ImageLab software (Bio-Rad, v. 6.0) and normalized to levels of GAPDH or α-tubulin. At least five biological replicates were analyzed.

Autophagy-Bafilomycin Experiment

To assess for the levels of autophagy, cells grown in 10% FBS GM were switched to a serum-free DMEM media for 1 h. Cells were incubated with 200 nM bafilomycin at 37°C for 3 h before harvesting for protein or fixing cells for immunohistochemistry analysis.

Quantitative Real-Time Polymerase Chain Reaction

Total RNA was extracted from the C2C12 cells using Direct-zol RNA MiniPrep (Zymo), including a DNAse digestion step. Eluted RNA concentration was measured with a Nanodrop One (Thermo Fisher Scientific) and reverse-transcribed to cDNA with a high-capacity cDNA reverse transcription kit (Applied Biosystems). Specific primers for PC2, myogenin, myosin heavy chain subunit 7 (MYH7), CaV1.1α, ATG5, PGC1α, NRF2, SP1, with Accuris Green master mix (Dot Scientific) were used to determine levels of cDNA with a QuantStudio 3 real-time PCR system (Applied Biosystems). cDNA levels were normalized to GAPDH and analyzed as mean fold changes in expression using the 2^−ΔΔCT method (36).

Next-Generation RNA Sequencing

Extracted cells were harvested for 24 h in GM and changed to DM for 4 days. Analysis was performed comparing the control cell line to the PC2-KO mutant in DM. Snap frozen cell pellets were shipped to GeneWiz, where RNA was extracted for next-generation sequencing, Illumina platform, HiSeq (GeneWiz). FASTQ sequences were analyzed using the Galaxy platform (37). Quality control of the FASTQ sequences was determined by FastQC, and the sequence reads were mapped against the GRCm38 genome reference. STAR and DESeq2 were utilized for the identification of differentially expressed genes (DEGs). Metascape (38; http://metascape.org) was used to detect enriched pathways among the DEGs with significant change (P value < 0.05).

Immunofluorescence Microscopy

C2C12 myoblasts grown on laminin-coated coverslips were cultured in growth media (GM), for 24 h and then subjected to the differentiation protocol. Cells were fixed in 2% paraformaldehyde (PFA) for 20 min at room temperature, washed three times in phosphate-buffered saline (PBS), and blocked for 45 min in 2% BSA blocking solution with 0.2% triton-X. The cells were incubated with antibodies against PC2 YCE2 (1:100, sc-47734, Santa Cruz Biotechnology), MFN2 (1:100, D2D10, Cell Signaling Technology), cytochrome C (1:100, 136F3, Cell Signaling Technology), Ki-67 (1:200, D3B5, Cell Signaling Technology), LC3A/B (D3U4C, 1:1,000, No. 12741S; Cell Signaling Technology), and beclin-1 (D40C5, 1:500, No. 3495S; Cell signaling Technology) overnight at 4°C, followed by the appropriate secondary antibody [Alexa Fluor 488 donkey anti-mouse IgG (1:800, A21202, Invitrogen), Alexa fluor 488 AffiniPure Fab fragment donkey anti-rabbit IgG 488 (1:800, 711 547 003, Jackson laboratory), Alexa Fluor 546 donkey anti-rabbit IgG (1:800, A10040, Invitrogen), Rhodamine Red-X AffiniPure Fab fragment donkey anti-mouse IgG (1:800, 715 297 003, Jackson laboratory)] for 1 h, then washed three times in PBS. For samples with RyR1/RyR2 (34C, 1:100, Developmental Studies Hybridoma Bank), cells were permeabilized with 2% triton-X for 10 min before the blocking step. Some slides were coincubated with 647-phalloidin (1:300, 20555, Cayman Chemicals) during the primary incubation step. Coverslips were mounted with Prolong-Diamond mounting media with DAPI (Invitrogen). After curing, cells were imaged with a ×43 oil (numerical aperture (N.A.) 1.2) or ×63 oil (N.A. 1.4) objective on an 880 Zeiss laser-scanning microscope with Airyscan (Zeiss, Germany) with up to ×2 zoom. Images were postprocessed with Zen Black software (Zeiss, Germany) and FIJI (National Institute of Health, NIH, 39).

Live-Cell Imaging

Cells were plated on glass bottom plates (CellVis) and transiently transfected with gCaMP6f (cytosolic calcium indicator). For overnight experiments, cells were imaged in DMEM without phenol red, at 37°C with 5% CO₂ (InCell, Cytiva). Images were acquired once every minute. Calcium imaging movies were analyzed using Fiji (39) and the Time Series Analyzer V3 plugin.

Generation of D511V Mutant

The D511V mutation was introduced into PC2-mCherry by Q5 site-directed mutagenesis kit (New England Bioscience, NEB), with mutagenesis primers designed according to the manufacturer’s guidelines. The point mutation was verified by sequencing (GeneWiz).

Plasmid Transfection of PC2 Constructs

PC2-mCherry [gift of Dr. A. Hofer, (32)], mCherryER, PC2-703-myc-his [gift of Dr. Y. Cai (40)], D511V PC2-mCherry, and MFN2-YFP plasmids [Addgene, 28010, (41)] were transfected into control or 53a PC2 KO C2C12 myoblasts cells using continuum reagent (Gemini Bio), according to the manufacturer’s instructions. C2C12 myoblasts were incubated with the plasmid-continuum complex for 12–16 h; afterward, media was changed to DM to induce differentiation. The cells were cultured for 7 days in the differentiation media.

Statistical Analysis

Data were plotted using GraphPad Prism 9. For comparisons between more than two groups, statistical significance was determined using two-way ANOVA followed by Tukey’s multiple comparison test. Differences were measured relative to growth media controls. For comparisons between two groups, statistical significance was evaluated using two-tailed t test. For comparisons in which Gaussian distribution was not assumed, nonparametric tests were used followed by Mann–Whitney test. Groups were considered significantly different if P values < 0.05. Error bars indicate standard deviation (SD).

RESULTS

PC2 KO Myoblasts Do Not Differentiate into Myotubes

We used Western blot to determine if CRISPR-Cas9 guide sequences against Pkd2 effectively knocked out protein expression levels of PC2 (Fig. 1A, clone 53a). Using immunofluorescence microscopy, we further established that PC2 was not expressed in the PC2 KO cells (Fig. 1B, clone 53a and Supplemental Fig. S1A; all Supplemental material is available at https://doi.org/10.6084/m9.figshare.19726198, clone 95). We examined relative Pkd2 mRNA expression and found that Pkd2 was downregulated by greater than 75% in the clone 53a PC2 KO cells (Supplemental Fig. S1B).

Figure 1.

Gene disruption of Pkd2 by CRISPR/Cas9 prevented C2C12 myoblast differentiation into myotubes. A: polycystin-2 (PC2) protein expression in C2C12 cells was greatly diminished following gene disruption of Pkd2 using a CRISPR/Cas9 system to generate PC2 knockout (PC2 KO) cells. Representative Western blot of PC2 in CTL (C2C12 transduced with a nontemplate control) and PC2 KO (clone 53a) C2C12 cells cultured in DMEM supplemented with 10% FBS media (growth media, GM). Each lane represents an independent experiment (left). Quantification of PC2 expression in C2C12 CTL and PC2 KO cells (right). n = 9 independent experiments. B: representative immunofluorescent images of PC2 (green), actin (phalloidin stain, white), and nuclei (DAPI, blue) in C2C12 myoblasts cultured for 24 h with GM. CTL (left) and PC2 KO (right). Scale bar, 10 µm. C: representative fluorescent images of C2C12 cells loaded with 2.2 µM of Fluo 4-AM. Top left: CTL cells 24 h cultured in GM. Top right: CTL cells after 7 days in DM. Bottom left: PC2 KO cells cultured for 24 h in GM. Bottom right: PC2 KO cells after 7 days in DM. Scale bar, 10 µm. Relative myogenin (D) and myosin heavy chain subunit 7 (MYH7; E) mRNA expression in CTL and PC2 KO after 24 h in GM (day 0), and various days under DM as determined by RT-qPCR. Levels were normalized to GAPDH and expressed as fold change of that determined in CTL cells under GM. Data plotted represent mean and SD (n = 3–4 independent experiments) are indicated with individual symbols. F: depiction of two gene ontology pathways that are significantly impacted following PC2 KO, as assessed by RNAseq. *P < 0.05, **P < 0.01, ***P < 0.001. Statistical significance was evaluated by two-way ANOVA followed by Tukey’s multiple comparison test. CTL, control; DM, differentiation medium; DMEM, Dulbecco’s modified Eagle’s medium; FBS, fetal bovine serum; SD, standard deviation.

We then analyzed the morphological changes upon myogenic differentiation in control (CTL) cells and PC2 KO cells. To induce differentiation of myoblasts to myotubes, we switched the cells plated at the same density in growth media (GM-DMEM supplemented with 10% FBS) to differentiation media (DM-DMEM supplemented with 2% FBS) for 7 days. As expected, in CTL cells, we observed elongation, fusion, and alignment of the differentiated cells, typical characteristics of myotube formation. In contrast, we observed that PC2 KO (both 53a and 95 clones) cells did not align or fuse together (Fig. 1C). KO of PC2 was sustained in differentiated cells (DM; Supplemental Fig. S1C).

For all subsequent results, we proceeded to further analyze the 53a clone. To start to understand why PC2 KO cells did not differentiate and remained morphologically as myoblasts, we tested the mRNA expression of transcription factors and muscle-specific genes involved in myogenesis. Upon serum starvation, in CTL cells, the transcription levels of myogenin progressively increased over 7 days by 500-fold (Fig. 1D, black bars). PC2 KO cells progressed to similar levels as the CTL cells during the first 3 days in DM to a 200-fold change. However, expression of myogenin remained constant in the PC2 KO cells over days 3–7 (Fig. 1D, gray bars). We then tested the expression of a muscle gene marker, myosin heavy chain subunit 7 (MYH7), which is upregulated upon myotube formation (42–44). MYH7 expression was upregulated by 25-fold in CTL cells after 7 days in DM but was not significantly upregulated in PC2 KO cells after 7 days in DM. In comparison with CTL cells in the same conditions, there was a 20-fold decrease in MYH7 expression in the PC2 KO cells (Fig. 1E).

Myoblasts are proliferating cells and when serum is withdrawn from myoblast cultures, these proliferating cells exit from the cell cycle and initiate differentiation to become myotubes (43). We used the proliferation marker, Ki-67, to assess the percentage of proliferating cells. As expected, we found that both CTL myoblasts and PC2 KO myoblasts in GM had a high percentage of proliferating cells, 75% and 50%, respectively (Supplemental Fig. S1D). After differentiation, proliferation was completely halted in CTL cells. However, ∼20% of PC2 KO cells in DM remained Ki-67 positive (Supplemental Fig. S1D). These results suggest that PC2 KO impairs myotube formation by preventing the required mRNA upregulation of muscle transcription factors and markers. In addition, as PC2 KO cells did not undergo complete withdrawal from the cell cycle and remained as myoblasts, the cells could not complete myogenesis and muscle formation.

As further verification that the myogenesis pathway is impacted by PC2 KO, we conducted an unbiased next-generation RNAseq analysis comparing the RNA gene profiles of CTL and PC2 KO cells after 4 days in DM. We chose to analyze the 4-day time point as the myogenin data (Fig. 1D) indicated that there was a bifurcation point in differentiation gene expression at this time. We conducted a gene ontology Metascape analysis of the genes significantly different between the two groups. The top two Metascape pathways affected at the 4-day time point in differentiation media were muscle structure development (98 genes differentially regulated, representing ∼8.7% of the gene in the gene ontology biological process) and muscle system processes (67 genes affected, representing 6% of the genes associated in muscle system processes, Fig. 1F). These data confirmed that the muscle differentiation process was indeed impaired in the PC2 KO cells.

PC2 KO Cells Do Not Functionally Differentiate

To functionally assess the impairment of muscle differentiation, we analyzed the ability of the CTL and PC2 KO cells after 7 days in DM to respond to voltage stimulation. Excitation-contraction (EC) coupling through depolarization-evoked Ca²⁺ transients is a hallmark of mature myotubes formation (45). In order for EC coupling to occur, the RyR in the SR and the voltage sensor, CaV1.1α (dihydropyridine receptor, DHPR) in the plasma membrane need to be in close contact (46). Ca²⁺ responses were assessed with the fluorescent calcium indicator Fluo-4AM in response to 40 V stimulation at 0.5 Hz (Supplemental Video S1). Representative tracings of Ca²⁺ transients in CTL myotubes and PC2 KO are shown (Fig. 2A). CTL myotubes showed robust synchronized calcium responses to voltage stimulation (around 65% of total cells analyzed). In contrast, PC2 KO cells showed a diminished amplitude response to voltage stimulation (less than 15% of total cells analyzed responded; Fig. 2B).

Figure 2.

C2C12 cells lacking PC2 have altered Ca²⁺ machinery. A: Ca²⁺ transients evoked by 0.5 Hz, 40 V stimulation in CTL (black) and PC2 KO (gray) cells under 7 days in DM and recorded in 1.5 mM Ca²⁺ containing Tyrode’s solution. B: quantification of peak Ca²⁺ amplitude. C: DHPR (CaV 1.1α) mRNA expression in CTL and PC2 KO after 24 h in GM and after 7 days in DM. RT-qPCR levels were normalized to GAPDH and expressed as fold change of those determined in CTL cells under GM. D: representative immunostaining of RyR1/RyR2 (green) and nuclei (DAPI, blue) of C2C12 cells after 7 days in DM, CTL (left), and PC2 KO (right). E: example images of CTL (top) and PC2 KO cells transfected with the calcium indicator gCaMP6F. Numbers refer to minutes. F: example traces of CTL (black) and PC2 KO (gray) cells demonstrating spontaneous Ca²⁺ oscillations. G: quantification of Ca²⁺ peaks over the 6 h imaging period. Scale bar 10 µm. Mean and SD (n = 3–4 independent experiments for RT-qPCR) are indicated with individual symbols. For imaging studies, cells are represented by individual symbols (n = 3–4 independent experiments). ****P < 0.0001. Statistical significance was evaluated using two-way ANOVA followed by Tukey’s multiple comparison test. Ca²⁺, calcium; CTL, control; DM, differentiation medium; GM, growth media; KO, knockout; PC2, polycystin-2.

To understand the molecular basis of the previous result, we measured mRNA levels of CaV1.1α. In CTL cells, there was a 550-fold increase in mRNA levels of CaV1.1α when differentiation was induced in the cells (Fig. 2C). Surprisingly, mRNA levels of CaV1.1α in PC2 KO cells were upregulated to a similar magnitude after differentiation (Fig. 2C). We then tested RyR1/RyR2 protein expression levels with immunofluorescence microscopy (Fig. 2D). Only CTL myotubes showed expression of RyR protein, which was almost completely absent in PC2 KO cells (Fig. 2D).

PC2 KO Cells Have Similar Evoked Calcium Responses, but Not Spontaneous Calcium Signals

We then turned our focus to addressing why PC2 KO prevented myogenic differentiation. PC2 as a TRP channel can allow for Ca²⁺ flux, and various Ca²⁺ processes have been implicated in myogenesis, including oscillations driven by InsP₃R and store-operated calcium entry. As PC2 can interact with both InsP₃R and SOCE and act as a calcium flux channel, we examined all three processes in turn. InsP₃R1-mediated Ca²⁺ signals have been suggested to be associated with the early steps of myoblasts differentiation (8, 47). To evoke Ca²⁺ release mediated by the InsP₃R, we applied ATP (10 µM), which activates the PIP₂-PLC-IP₃ pathway. In GM, the area under the curve (AUC) was higher in the PC2 KO compared with the CTL cells, but the amplitude did not change (Supplemental Fig. S2, A–C). In DM, the amplitude of PC2 KO cells was larger compared with CTL cells, but the AUC was smaller in PC2 KO compared with CTL cells (Supplemental Fig. S2, D–F). The expression of InsP₃R1, analyzed by Western blot, was unchanged in GM between CTL and PC2 KO cells, but in DM, there was a significant reduction in the PC2 KO cells compared with the CTL (Supplemental Fig. S2, G and H). As the agonist-induced Ca²⁺ response to ATP was larger in the GM in the PC2 KO cells, it seemed unlikely that the increased InsP₃R dependent pathway response seen in PC2 KO cells would be responsible for a loss in differentiation signals.

To examine if SOCE was affected in the PC2 KO cells, we measured the amount of calcium influx following the emptying of stores and SERCA inhibition with thapsigargin (Supplemental Fig. S3A). Although the thapsigargin response was lower in the PC2 KO, the amplitude and area under the curve of the SOCE response were not changed between the CTL and PC2 KO cells (Supplemental Fig. S3B). These data suggested that it was unlikely that SOCE was responsible for the changes.

Finally, we tested spontaneous Ca²⁺ signals in the PC2 KO cells, as these would give rise to Ca²⁺ oscillations that could lead to cell differentiation. Over a 6 h period, there were significantly more spontaneous Ca²⁺ transients that lasted for longer periods of time in the CTL cells compared with the PC2 KO cells (Fig. 2, E–G). Collectively, these results showed that PC2 KO cells had impaired spontaneous Ca²⁺ signaling.

Autophagy is Downregulated in PC2 KO Cells following Induction of Myogenic Cues

To better understand how PC2 KO contributed to the lack of differentiation, we returned to the RNAseq data to examine pathways known to regulate myogenic differentiation. We noted that several genes in the autophagy pathway were significantly downregulated (Supplemental Table S1). This caught our attention, as the C2C12 differentiation is partially dependent on autophagy (5, 48). PC2 has been implicated in mediating autophagy via beclin-1 (22, 30), and beclin-1 has previously been shown to contribute to autophagy induction and formation of the autophagosomes (49). We examined the protein expression of beclin-1 and found that it was significantly downregulated under DM conditions (Fig. 3, Aand B). Beclin-1 expression was not significantly different in GM between CTL and PC2 KO cells, suggesting it was only impacted when differentiation was evoked (Fig. 3, Aand B).

Figure 3.

Autophagy initiation is downregulated in PC2 KO cells following induction of myogenic cues. A: representative Western blot of beclin-1 in CTL and PC2 KO C2C12 cells after 7 days in DM. GAPDH was used as a loading control. B: quantification of beclin-1 in CTL and PC2 KO C2C12 cells in GM and after 7 days in DM. Protein levels were normalized to GAPDH. C: representative Western blot of PINK1 in CTL and PC2 KO C2C12 cells after 7 days in DM. GAPDH was used as a loading control. D: representative Western blot of parkin in CTL and PC2 KO C2C12 cells after 7 days in DM. GAPDH was used as a loading control. E: quantification of PINK1 in CTL and PC2 KO C2C12 cells in GM and after 7 days in DM. Protein levels were normalized by GAPDH. F: quantification of parkin in CTL and PC2 KO C2C12 cells after 7 days in DM. Protein levels were normalized by GAPDH. n = 5 independent experiments. CTL, control; DM, differentiation medium; GM, growth media; KO, knockout; PC2, polycystin-2. **P < 0.01.

We also examined the mitophagy pathway (the autophagy of mitochondria) by examining the protein expression of the kinase PINK1, which is activated and recruited to mitochondria under conditions of mitophagy. Under GM conditions, there was no difference in PINK1 levels (Supplemental Fig. S4), however, the PC2 KO cells in DM expressed significantly higher levels of the kinase PINK1 compared with CTL (Fig. 3, Cand E). However, the immediate downstream ligase parkin was unchanged, indicating a possible impairment of the mitophagic and autophagic pathways (Fig. 3, Dand F).

To further confirm that the autophagic pathway was affected by PC2 KO, we challenged myoblast cells in GM with bafilomycin (200 nM for 3 h). Bafilomycin inhibits autophagy by preventing the fusion of autophagosomes and lysosomes. In CTL cells, addition of bafilomycin caused an increase of LC3-II, which was not observed in PC2 KO cells, as assessed by Western blot (Fig. 4, A–C). We also verified by immunohistochemistry where CTL cells had significantly more LC3 puncta than the PC2 KO cells (Fig. 4, Dand E). These results suggest a reduction in autophagic flux in PC2 KO cells.

Figure 4.

Induced autophagic and mitochondrial pathways are impaired in PC2 KO cells. A: example Western blot of LC3 expression following the addition of bafilomycin for 3 h in GM. B: quantification of LC3-II expression after the addition of bafilomycin in CTL and PC2 KO cells. Protein levels were normalized to GAPDH. C: quantification of LC3-II/I ratio expression after the addition of bafilomycin in CTL and PC2 KO cells. D: representative immunofluorescence images of LC3 puncta (magenta) in CTL and PC2 KO cells following the addition of bafilomycin for 3 h. E: quantification of LC3 puncta CTL and PC2 KO C2C12 cells in GM following the addition of bafilomycin for 3 h. *P < 0.05, **P < 0.01, ****P < 0.0001. Individual symbols in imaging experiments represent single cells. Data are representative of 3–4 independent experiments. Individual symbols in Western blot quantification represent independent experiments (n = 6). Statistical significance was evaluated using two-way ANOVA followed by Tukey’s multiple comparison test. CTL, control; DM, differentiation medium; GM, growth media; KO, knockout; PC2, polycystin-2.

Mitochondrial Biogenesis and Remodeling Are Impaired in PC2 KO Cells under Serum Starvation

We reasoned that impairment of the mitophagic and autophagic pathways would also inhibit mitochondrial biogenesis, a required pathway for cells to adapt to the new metabolic demands during myotube differentiation (5–7, 50). We assessed mitochondrial biogenesis in the early stages of myogenesis (day 3) by measuring mRNA expression of transcription factors involved in mitochondrial biogenesis (51): PGC1α (Fig. 5A), NRF2 (Fig. 5B), and SP1 (Fig. 5C). We found that 3 days into differentiation, PGC1α mRNA expression was upregulated threefold in the CTL cells compared with the undifferentiated cells. On the other hand, PGC1α was not significantly upregulated in PC2 KO cells after 3 days in differentiation (Fig. 5A). Similar results were found with NRF2, in which mRNA expression was significantly upregulated in CTL cells but not in PC2 KO (Fig. 5B). SP1, in CTL cells, showed an increasing trend in the transcription levels after 3 days in differentiation, but this trend did not occur in the PC2 KO cells (Fig. 5C). To evaluate whether these transcriptional factors resulted in increased mitochondrial mass, we examined the mitochondrial protein cytochrome C (Cyt. C) at the final stage of differentiation (Fig. 5, Dand E). Protein expression of Cyt. C was significantly upregulated in differentiated CTL cells. Consistent with the PGC1α and NRF2 mRNA results, there was no protein upregulation of Cyt. C in PC2 KO cells. We further confirmed the increase in mitochondrial mass by examining immunofluorescence staining of Cyt. C. Compared with robust Cyt. C immunostaining in CTL cells, staining in differentiated PC2 KO cells was almost completely absent (Fig. 5F).

Figure 5.

Mitochondrial biogenesis and remodeling are impaired in PC2 KO cells under serum starvation. PGC1α (A), NRF2 (B), and SP1 (C) mRNA expression was determined by RT-qPCR in CTL and PC2 KO after 24 h in GM (day 0), and 1 and 3 days in DM. mRNA levels were normalized to GAPDH and expressed as fold change of those determined in CTL cells under GM. D: representative Western blot of cytochrome C (Cyt. C) in CTL and PC2 KO C2C12 cells in GM (left) and after 7 days in DM (right). GAPDH was used as a loading control. E: quantification of Cyt. C protein in CTL and PC2 KO C2C12 cells in GM and after 7 days in DM. Protein levels were normalized by GAPDH. F: representative immunostaining of Cyt. C (magenta) and nuclei (DAPI, blue) in C2C12 cells after 7 days in DM, CTL (left) and PC2 KO (right). Scale bar, 10 µm. G: representative Western blot of MFN2 protein in CTL and PC2 KO C2C12 cells under GM (left) and after 7 days in DM (right). H: quantification of MFN2 protein levels in CTL and PC2 KO C2C12 cells under GM and after 7 days in DM. Protein levels were normalized by GAPDH and expressed as fold change of those determined in CTL cells in GM. I: representative immunostaining of MFN2 (green) and nuclei (DAPI, blue) in C2C12 cells after 7 days in DM, CTL (left) and PC2 KO (right). Scale bar, 20 µm. Mean and SD (n = 3–5 independent experiments for RT-qPCR and Western blot) are indicated with individual values. *P < 0.05, **P < 0.01, ***P < 0.001. Statistical significance was evaluated using two-way ANOVA followed by Tukey’s multiple comparison test. CTL, control; DM, differentiation medium; GM, growth media; KO, knockout; MFN2, mitofusin 2; PC2, polycystin-2; SD, standard deviation.

Recently, PC2 has been shown to interact with MFN2, a protein located in the ER-mitochondria-associated membranes, which is known to mediate mitochondria-ER tethering and fusion (32). Moreover, it is known that mitochondrial fusion is upregulated during myogenic differentiation of C2C12 myoblasts, resulting in the formation of mitochondrial networks (6). Therefore, we hypothesized that PC2 KO cells would experience disruption in mitochondrial remodeling. We looked at MFN2 protein expression levels by Western blot (Fig. 5G) and by immunofluorescence microscopy (Fig. 5I). We found that in differentiated cells, there was upregulation of MFN2 protein in CTL cells compared with undifferentiated cells. However, there was no upregulation of MFN2 protein expression in differentiated PC2 KO cells (Fig. 5H). The earlier results indicate that upon differentiation stimulus, PC2 KO disrupts mitochondrial biogenesis causing a decrease in mitochondrial mass. In addition, PC2 KO causes impairment to mitochondrial-ER fusion and remodeling.

Acute Knockdown of PC2 in Differentiated Cells Results in Atrophy

To assess if PC2 was required for the maintenance of mature myotubes, we differentiated CTL cells for 7 days in DM, and then transduced the cultures with lentivirus containing the CRISPR KO 95 and 53 sequences, or no-template control. After 48–72 h, the majority of the transfected or transduced myotubes (as evidenced by green fluorescence) dedifferentiated or atrophied (Supplemental Fig. S5A). In calcium imaging experiments, the amplitude of the ATP calcium response was unchanged (Supplemental Fig. S5B), although the response time was slower. Cells that were transduced with PC2 KO no longer responded to a voltage stimulus (Supplemental Fig. S5, C and D). As PC2 KO caused rapid dedifferentiation and atrophy, it is unlikely that the same signaling pathway is being affected with acute PC2 KO in mature myotubes compared with the differentiation from myoblasts. These data suggest that PC2 is not only required for differentiation from myoblasts to myotubes, but also for the maintenance of in vitro myotubes, and this is consistent with previous findings (31).

Re-Expression of Full-Length PC2 Is Required to Restore the Myotubes Phenotype, and the Re-Expression of Mitochondrial Proteins in PC2 KO Cells

To further examine the requirement of PC2 for in vitro myogenesis, we re-expressed full-length human PC2 in the PC2 KO cells. We transiently transfected CTL cells with mCherryER (Fig. 6A), or PC2 KO cells with mCherryER, PC2-mCherry (overexpressed, PC2-OE), or MFN2-YFP (Fig. 6A), and cultured the cells for 7 days in DM. PC2 KO cells transfected with mCherryER or MFN2-YFP did not differentiate into myotubes (Fig. 6A). However, we saw the restoration of differentiation in PC2 KO cells that were transfected with PC2. These data indicate that full-length PC2 expression was required to restore the morphological linkage between the ER and mitochondria, which is essential for myogenesis.

Figure 6.

Re-expression of full-length PC2 restores the myotube phenotype in PC2 KO cells. A: example images of C2C12 CTL cells (top left) and PC2 KO cells transfected with mCherryER (top middle), PC2-mCherry (top right), MFN2-YFP (bottom left), 703X-His (bottom middle), or D511V (bottom right), and cultured for 7 days in DM. B: quantification of myogenin mRNA levels in mCherryER, PC2-mCherry over expressing, or D511V PC2 KO cells after 7 days in DM. C: quantification of MYH7 mRNA levels in mCherryER, PC2-mCherry over expressing, or D511V PC2 KO cells after 7 days in DM. D: representative immunofluorescence images of Cyt. C expression (green) in mCherryER, PC2-mCherry over expressing, or D511V PC2 KO cells after 7 days in DM. Quantification of staining in transfected cells (right). E: representative immunofluorescence images of ryanodine receptor expression (green) in mCherryER, PC2-mCherry over expressing, or D511V PC2 KO cells after 7 days in DM. Quantification of staining in transfected cells (right). CTL, control; DM, differentiation medium; GM, growth media; KO, knockout; MFN2, mitofusin 2; OE, overexpression; PC2, polycystin-2.

To determine if the calcium channel function of PC2 was required for myogenesis, we tested two known pathological variants of PC2, both of which localize, like the full-length PC2, to the ER. The first was a truncation, in which the C-terminus of the protein was cut off (PC2-703X). This pathological mutation is known to abolish Ca²⁺-mediated PC2 responses (52). When PC2-703X was expressed in PC2 KO cells, there was no myotube differentiation (Fig. 6A). To determine if calcium from PC2 was required for the differentiation, we expressed a channel dead mutant, D511V. PC2 KO cells transfected with the D511V mutation did not differentiate. To further confirm that the ATP-evoked response in C2C12 cells is not altered by PC2 expression, we examined evoked ATP responses in PC2 KO cells either transfected with PC2 or with the PC2 D511V mutation. We found no difference in ATP responses (Supplemental Fig. S6A).

To assess which differentiation pathways were affected by the calcium from PC2, we re-examined the differentiation markers that were significantly downregulated in PC2 KO cells. Whereas PC2 overexpressed (OE) resulted in an increase in differentiation markers myogenin and MYH7, this was not seen in the PC2 KO cells expressing the D511V mutant (Fig. 6, Band C). The mitochondrial pathway was partially restored in PC2 OE cells, with an increasing trend in PGC1α expression (Supplemental Fig. S6B). As the number of cells that retained the transfection following the 7-day period was low, we used immunofluorescence analysis instead of Western blot to assess Cyt. C and RyR expression. We then selected cells expressing the mCherry tag to denote cells that were transfected. Cyt. C was upregulated in the PC2 OE transfected cells compared with mCherry only cells, and trended upwards, but was not significantly upregulated in PC2 D511V cells (Fig. 6D). RyR expression was significantly upregulated in both PC2 OE cells and PC2 D511V OE cells compared with mCherry control (Fig. 6E).

As our results suggested that PC2 may directly modulate the autophagic pathway to alter myogenesis, we examined beclin-1 expression after 24 h in GM (Fig. 7, A–C). Measurements of total beclin-1 intensity showed higher beclin-1 expression in PC2 OE and PC2 D511V transfected cells compared with the mCherry-ER PC2 KO cells (Fig. 7B). PC2 D511V transfected cells showed an increase in the size of beclin-1 particles (Fig. 7C). Beclin-1 has been shown to translocate to the mitochondria to regulate mitophagy (53) and its well-described role in general autophagy. Because our data in Fig. 3 pointed to a function for PC2 in mitophagy, we, therefore, examined the location of the beclin-1 in PC2 OE and PC2 D511V after 4 days in DM. Colocalization of beclin-1 to the mitochondria, as examined by Pearson’s correlation with the mitochondrial marker, VDAC, was significantly higher in the PC2 OE and PC2 D511V (Fig. 7, Dand E). We also observed the formation of aggregates in PC2 D511V expressing cells (Fig. 7D, white arrow). We quantified these aggregates by counting the number of beclin-1 puncta, and found, consistent with the aggregates, fewer individual beclin-1 puncta in the PC2 OE and PC2 D11V-transfected cells compared with the mCherry-ER-transfected cells (Fig. 7F). As we did not see the restoration of myotube formation in the PC2 D511V-transfected cells, we wondered if the aggregates of beclin-1 and VDAC were indicative of dysregulated mitophagy and increased autophagy. We measured LC3-II and found that there was a significantly higher level of LC3-II in the D511V expressing cells but not in the PC2 OE cells (Fig. 7, Gand H). These data suggested that overexpression of a pathogenic PC2 mutation, which does not conduct calcium, caused increased autophagy. Our results also suggest that restoration of the beclin-1 pathway for functional mitochondrial remodeling requires a calcium-dependent PC2 signal.

Figure 7.

Re-expression of full-length cation conducting PC2 enables beclin-1 VDAC interaction. A: representative immunofluorescence images of beclin-1 (red) and PC2 expression (green) in PC2 KO+mCherry-ER (top left), PC2 KO+PC2-mCherry over expressing (OE; top middle), or PC2 KO+D511V cells (top right) after 24 h in GM. Bottom: beclin-1 expression of the cell lines of the top. B: quantification of beclin-1 intensity in the cell lines described in A. C: quantification of beclin-1 size in mCherry-ER expressing, PC2 over-expressing, or D511V expression in PC2 KO cells. D: representative images of beclin-1 (red), VDAC (green), and actin (phalloidin, magenta) in PC2 KO+mCherry-ER (top left), PC2 KO+PC2-mCherry over expressing (top middle), or PC2 KO+D511V cells (top right) after 4 days in DM. Bottom left are zoomed in images of mitochondrial morphology highlighted in white squares. E: quantification of Pearson’s correlation between beclin-1 and VDAC staining in cell lines described on D. F: quantification of number of beclin-1 particles expressed in the cell lines described on D. G: example Western blot of LC3 expression in PC2 KO+mCherry-ER, PC2 KO+PC2-mCherry over expressing, or PC2 KO+D511V cells after 24 h in GM. H: quantification of LC3-II expression in the three transfection conditions described in G. Protein levels were normalized to GAPDH. Mean and SD are indicated with individual values. n = 3 independent experiments for immunofluorescence studies. Statistical significance was evaluated using two-way ANOVA followed by Tukey’s multiple comparison test. CTL, control; DM, differentiation medium; GM, growth media; KO, knockout; PC2, polycystin-2; SD, standard deviation.

DISCUSSION

PC2 is known to cause renal cysts, but the expression of PC2 outside of the kidneys suggests that it may have an importance in other tissue organs. In this study, we demonstrate that the knockout of an ER-localized TRP channel, PC2, prevents C2C12 myoblast differentiation into myotubes. Our rescue and pathogenic mutation studies suggest that the physical presence of PC2 is sufficient for myogenic pathways such as mitochondrial upregulation to take place, but that a calcium signal mediated by PC2 is necessary for a properly timed myogenic pathway that includes the upregulation of myogenin and the fusion of myoblast cells.

We found that PC2 KO cells have impaired spontaneous Ca²⁺ oscillations without agonist stimulation, suggesting that PC2 in myoblast cells may be acting as a spontaneous Ca²⁺ leak channel. Ca²⁺ oscillations have previously been implicated in the initial steps of myogenesis (8, 47). Rather than being an InsP₃R1-mediated Ca²⁺ signaling pathway, our findings that the ATP-evoked response was unchanged between PC2 expressing and the D511V PC2 channel mutation indicate that these signaling pathways are InsP₃R independent but PC2-Ca²⁺ dependent (54, 55). Several studies have demonstrated the importance of calcium entry via STIM1, plasma membrane TRPs, and Orai in initiating differentiation pathways (9, 13, 16). Our SOCE experiments suggest that overall SOCE may not be affected, and PC2 acts upstream of SOCE and independent of InsP₃R to modulate myogenesis.

How might the Ca²⁺ mediated by PC2 be contributing to the myogenic pathway? The use of the D511V pathogenic PC2 mutation enabled a finer dissection of the structural restoration of PC2 protein compared with the actual ability of the channel to conduct cations. Intriguingly, the expression of the D511V phenocopied the lack of differentiation, but some molecular pathways of myogenesis, including partial upregulation of myogenin and Cyt. C were restored. What was noticeably absent was the lack of fusion in the D511V-transfected cells and the lack of restoration in late differentiation proteins, such as RyR. These data suggest that the Ca²⁺ function of PC2 is a requirement for these processes. Not all myogenic pathways were inhibited by the loss of PC2. For example, PC2 KO and CTL cells had similar upregulation of CaV1.1α gene expression after 7 days in DM. These data suggest that PC2 and its downstream targets do not regulate the transcriptional activation of CaV1.1α in C2C12 cells, which may be under the control of other regulatory pathways. Our findings suggest that CaV1.1α expression was not affected by PC2 KO is consistent with what we have previously reported with cardiomyocyte-expressed CaV1.2 (23). However, this is in contrast to the other polycystin protein, PC1, which has been demonstrated to regulate the expression of the β-subunit of CaV1.2 in cardiomyocytes (56).

We saw the largest divergence between the CTL and PC2 KO cells after 3 days in DM, a time point that coincides with a progressively increased metabolic demand of the myocytes, as well as the rate of mitochondrial biogenesis. We observed a striking halt in the upregulation of myogenin in the PC2 KO cells after 3 days of differentiation (Fig. 1D). Therefore, the halt of myogenin after 3 days in the PC2 KO cells was likely to be a lack of metabolic demand in the PC2 KO cells rather than an insufficiency of the KO cells to keep up with the expected metabolic demand (5, 43). Controlled autophagy is linked with cell metabolism and regulates mitochondrial biogenesis and remodeling to adapt to the increased metabolic demands of the newly formed myotubes (5, 7, 50, 57). A key ramification following PC2 KO is an impairment in mitochondrial biogenesis, as we show that mitochondrial biogenesis regulatory genes like PGC1α, NRF2, and SP1 (51) increased in CTL cells after 3 days in DM but not in PC2 KO cells (Fig. 5, A–C). Inhibition of mitochondrial biogenesis in PC2 KO cells is likely due to initial downregulation of PC2-mediated autophagy, which would result in impairment of mitochondrial mass in the later stages of myogenesis. In addition, it has been shown that PGC1α regulates MFN2 expression, resulting in mitochondrial remodeling (58). We saw an increase in MFN2 protein during differentiation, and this was halted in PC2 KO cells. In our studies, overexpression of MFN2 in PC2 KO cells did not recover the myotubes phenotype. We suggest that mitochondrial remodeling alone is not sufficient to promote myogenesis, and this supports our finding that PC2-calcium signaling is necessary during C2C12 myoblast differentiation.

Mechanistically, autophagy is required for the maintenance of skeletal muscle mass, as well as metabolism and regeneration (59). In addition, several skeletal muscle studies have shown that an increase in autophagy is crucial for myogenesis to occur (6, 48, 59, 60). PC2 has been shown to induce autophagy through an interaction with beclin-1 (22, 30). Our data suggest that PC2 KO results in decreased beclin-1 protein expression and loss of autophagic responses (Figs. 3 and 4). These results are consistent with cardiomyocytes and human embyronic kidney (HEK) cells autophagy-induced by PC2 Ca²⁺-mediated signals in complex with beclin-1 (22, 30). Moreover, silencing of beclin-1 in myoblasts has been shown to reduce fusion ability (48). In addition, we observed a decoupling of the mitophagy pathway, where PC2 KO leads to a striking increase in the expression of PINK1 but unaltered parkin expression levels. These findings are consistent with previous studies on the PC2 partner protein, polycystin 1 (PC1), where knockdown of PC1 leads to an upregulation of PINK1, but no change in parkin (61). Moreover, it has been demonstrated that beclin-1 interaction with parkin aids the translocation of parkin to the mitochondria to interact with PINK1 for mitophagy (53). The lack of restoration of the myotube formation in the non-Ca²⁺ conducting PC2 D511V mutant and the formation of VDAC-beclin-1 aggregates (Fig. 7) suggested that the Ca²⁺ signal mediated via PC2 may be required for functional mitophagy.

Taken together, our results suggest that one functional role for the Ca²⁺-mediated signals by PC2 is in activating beclin-1-mediated autophagy. The absence of PC2 leads to an impaired mitophagy as evidenced by elevated PINK1 levels, and unchanged parkin levels due to downregulation of beclin-1, which is needed to facilitate parkin translocation to the mitochondria. In the context of muscle differentiation, this impairment would be expected to decrease mitochondrial biogenesis, and indirectly, may be linked to the lack of transcriptional upregulation of myogenic genes and myoblast fusion (48). Although skeletal muscle dysfunction has not been reported in the ADPKD clinical literature (where patients carry mutations in an autosomal dominant fashion), skeletal and muscle deformation has been reported in whole body homozygous animal studies (62). It remains to be determined if PC2 is essential in mediating in vivo muscle repair and injuries.

Conclusions

In this study, we demonstrate that a TRP channel localized on the ER, PC2, is a novel regulator of C2C12 myoblast differentiation. Our data demonstrate that PC2 integrates early intracellular Ca²⁺ signals and metabolic pathways that are necessary for in vitro myogenesis.

SUPPLEMENTAL DATA

Supplemental Table S1, Supplemental Figs. S1–S6, and Supplemental Video S1: https://doi.org/10.6084/m9.figshare.19726198.

GRANTS

This work was supported by National Institutes of Health Grant R00DK101585 and a pilot award 5UL1TR002389-02, which funds the Institute for Translational Medicine (ITM; to I.Y.K). Research reported was also supported by the National Institute of Diabetes and Digestive and Kidney Diseases of the National Institute of Health under award Numbers U2CDK129917 and TL1DK132769.

DISCLAIMERS

The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health (to K.M.M.N).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

K.M.M.N., V.V., and I.Y.K. conceived and designed research; K.M.M.N., V.V., E.D., S.O.V., and I.Y.K. performed experiments; K.M.M.N., V.V., and I.Y.K. analyzed data; K.M.M.N., V.V., and I.Y.K. interpreted results of experiments; K.M.M.N., V.V., and I.Y.K. prepared figures; K.M.M.N. and V.V. drafted manuscript; K.M.M.N., V.V., E.D., and I.Y.K. edited and revised manuscript; K.M.M.N., V.V., E.D., S.O.V., and I.Y.K. approved final version of manuscript.