Mouse neonatal ventricular cardiomyocytes were separated from 1–2-day-old neonatal C57BL/6J mice using enzymatic dissociation and cultured as previously described (26). Briefly, neonatal mice hearts were cut into small pieces and digested in 0.125 mg/ml trypsin (Gibco, #15090046) at 37°C. Then the supernatant was collected with complete Dulbecco's modified Eagle medium: Nutrient Mixture F-12 (DMEM/F12, Gibco, #8120319) and centrifuged at 600 g for 5 min. After repeating this cycle 6–8 times, the cell pellets were resuspended in DMEM/F12 containing 10% fetal bovine serum (FBS, Gibico, #10099133C), passed through a 100 mm cell strainer, and plated onto 10 cm dishes for 1.5 h at 37°C in 5% CO₂. The supernatant was then obtained and plated on dishes for further experiments. For gene overexpression and knockdown studies, cardiomyocytes were infected with lentiviruses according to the manufacturer's protocol. Recombinant lentiviruses were designed and synthesized by Fubio Biological Technology. After lentiviral transduction for 72–96 h, cardiomyocytes were stimulated with 10⁻⁶ M Ang II (Sigma, #A9525) for the indicated time points.

HL-1 cardiac muscle cell line (HL-1 cell) was purchased from Sigma-Aldrich (#SCC065) and maintained in Claycomb medium (Sigma-Aldrich, #51800C) supplemented with 10% FBS and 2 mM L-glutamine. Plasmids were transfected according to the manufacturer's protocol. After starving for 24 h in serum-free Claycomb medium, the cells were treated with 10⁻⁶ M Ang II for 24 h and followed by co-immunoprecipitation (co-IP) experiments.

Pink1-Flag and Samm50-influenza hemagglutinin (HA) complementary DNA (cDNA) were cloned into the pcDNA3.1 vector. Short hairpin RNA (shRNA) against Samm50 and Pink1 were performed using the pLKO vector. The target sequences of shRNA oligonucleotides were listed in Supplementary Table 1. The Vps34 inhibitor was purchased from MCE (#HY-12794). Lipo8000 transfection reagent was purchased from Beyotime (#C0533).

C57BL/6J male mice (10–12 weeks) were subjected to transverse aortic constriction (TAC) to simulate the pressure overload model, and mice were sacrificed two weeks after surgery. After opening the chest cavity and separating the aortic arch, a 27-gauge needle was placed on the aorta between the left common carotid artery and the innominate artery, followed by ligation with 6-0 silk. Then, the needle was removed to generate aortic constriction. The Sham group mice underwent an identical surgery apart from the ligation. The animal study was reviewed and approved by the Animal Care and Use Committee of Zhongshan Hospital, Fudan University.

Total RNA was extracted with the TRIzol reagent (Ambion, #257401). cDNA was generated using PrimeScript^TM RT Reagent Kit with gDNA Eraser (Takara, #RR047A) following the manufacturer's instructions. Quantitative real-time polymerase chain reaction (qRT-PCR) was performed by ChamQ Universal SYBR qPCR Master Mix (Vazyme, #Q711-02) on a Bio-Rad IQ5 multicolor detection system. The program was as follows: 5 min at 95°C followed by 40 cycles of 20 s at 95°C and 30 s at 60°C. The results were analyzed using the 2^−ΔΔCt method (27). The primer sequences were listed in Supplementary Table 2.

Proteins were obtained from tissues and cells lysed in lysis buffer (Beyotime, #P0013B). Then, samples were separated by 10–15% SDS-polyacrylamide electrophoresis gel and transferred to PVDF membranes. After blocking in 5% BSA, blots were incubated with the primary antibodies at 4°C overnight, followed by incubation with peroxidase-conjugated rabbit secondary antibody (Thermo Fisher, #A0545, 1:5000) at room temperature for 1 h. The following antibodies were used in this study: anti-GAPDH (Proteintech, #HRP-60004, 1:10000), anti-LC3 (Cell Signaling Technology, #2775, 1:1000), anti-COX4 (Proteintech, #11242-1-AP, 1:1000), anti-Pink1 (Proteintech, #23274-1-AP, 1:1000), anti-Parkin (Proteintech, #14060-1-AP, 1:1000), anti-Samm50 (Abcam, #ab133709, 1:5000) and anti-TOM20 (Santa Cruz Biotechnology, #sc-11415, 1:1000).

For co-IP analysis, proteins were prepared with non-denaturing lysis buffer and incubated with primary antibodies against HA or isotype control immunoglobulin G (IgG) at 4°C overnight. Then, 50 μl magnetic beads (MCE, #HY-K0205) were added into the mixture and incubated for 4 h with rotation. After removing the surface attachments, samples were obtained from the bead-antibody complexes and subjected to immunoassay as above described. The following antibodies were used in this study: anti-IgG (Proteintech, #30000-0-AP, 1:50) and anti-HA (Proteintech, #51064-2-AP, 1:50).

Cells were fixed with 4% paraformaldehyde for 30 min, permeabilized with 0.1% Triton X-100 for 5 min, and blocked in 5% BSA for 1 h. Then cells were stained with the following primary antibodies overnight at 4°C: Cardiac troponin T (Abcam, #ab8295, 1:200), Samm50 (Proteintech, #28679-1-AP, 1:50), Pink1 (Proteintech, #23274-1-AP, 1:100) and VDAC1 (Santa Cruz Biotechnology, #sc-390996, 1:50). After washing with PBS 3 times, cells were incubated with specific secondary antibodies conjugated to Alexa Fluor for 1 h at 37°C. Five minutes after co-staining with 4',6-diamidino-2-phenylindole (DAPI, Invitrogen, #D1306), the cells were observed by a fluorescence microscope, and cardiomyocyte surface area was calculated by Image-Pro Plus software.

Heart sections were deparaffinized and rehydrated and then antigenically retrieved in sodium citrate buffer for 30 min. The following procedures were conducted according to the protocol described for cardiomyocytes. Co-localization was analyzed by Image-Pro Plus software.

Adenovirus harboring tandem fluorescent mRFP-GFP-LC3 system (adenovirus-tf-LC3) was used to evaluate autophagy flux as previously described (28). Cardiomyocytes plated on coverslips were transfected with adenovirus-tf-LC3 at 10 MOI for 24 h and treated with Ang II for 6 h. Then cells were fixed with 4% paraformaldehyde, stained with DAPI, and observed under a fluorescence microscope. The number of GFP and mRFP dots were recorded by manual counting of fluorescent puncta from at least 50 cells. The number of DAPI-stained nuclei were recorded to represent the nuclear number.

All data were presented as mean ± s.e.m. All statistical results were analyzed using GraphPad Prism Software. Statistical analysis of two sets was performed by Student's t-test. Multiple comparisons were conducted by one-way analysis of variance with the Newman–Keuls test. A value of P < 0.05 was considered statistically significant.

To determine whether Samm50 expression is associated with cardiac hypertrophy, we performed the TAC mice model (Supplementary Figures 1A–D) and found the mRNA, and protein levels of Samm50 were remarkably decreased in response to pressure overload (Figures 1A,B). Parallelly, in the isolated mice neonatal cardiomyocytes, qRT-PCR and western bolt (WB) data showed that Samm50 expression levels were significantly reduced by Ang II stimulation (Figures 1D,E). Interestingly, Samm50 expression was unchanged in cardiac fibroblasts under Ang II treatment (Supplementary Figures 1E,F). The immunofluorescence staining also showed a consistent result of decreased Samm50 levels in TAC-induced hypertrophic hearts and Ang II-treated cardiomyocytes (Figures 1C,F). These results revealed that Samm50 was dramatically downregulated both in vivo and in vitro in response to hypertrophic stimuli, indicating a potential role of Samm50 in the regulation of cardiac hypertrophy.

To investigate the potential effect of Samm50 on cardiomyocyte hypertrophy, we transfected lentivirus overexpressed Samm50 into isolated and cultured mice neonatal cardiomyocytes and verified Samm50 expression by WB and qRT-PCR analysis (Figures 2A,B). After Ang II administration, the expressions of hypertrophic markers such as nppb (natriuretic peptide B), c-jun (jun proto-oncogene), c-fos (Fos proto-oncogene), and rcan1.4 (regulator of calcineurin 1, transcript 4) were obviously elevated in the Samm50-overexpressed group compared with the control group (Figure 2C). Furthermore, cardiomyocytes were incubated with cardiac troponin T to evaluate the cardiomyocyte surface area. As shown in Figure 2D, Samm50 overexpression markedly enlarged the cardiomyocyte size in response to Ang II treatment. These data indicated that Samm50 exacerbates cardiac hypertrophy.

Mitochondrial dynamics are critical to maintain cardiomyocyte function, especially under pathological stress. Given that Samm50 is associated with mitochondrial homeostasis and mitophagy (25), we explored whether Samm50 aggravates cardiac hypertrophy by affecting mitochondria function. We first evaluated mitochondrial biogenesis by detecting the expression of mitochondrial NADH dehydrogenase 1 (mtND1). The results showed that mtND1 was not altered in response to Samm50 overexpression (Supplementary Figure 1G). Then, autophagy markers and mitochondria protein levels were assessed by immunoblotting. Compared with the control group, overexpression of Samm50 displayed decreased LC3-II/LC3-I ratio and increased levels of TOM20 and COX4, indicating inhibition of mitophagy in cardiomyocytes (Figure 2E). Besides, the adenovirus-tf-LC3 was generated to evaluate autophagic flux. We monitored the number of red and green puncta by fluorescence microscope. The red puncta overlapped with the green ones are indicators of autophagosomes, while the free red puncta represent autolysosomes (29). As shown in Figure 2F, the yellow and red puncta were both increased in Ang II-treated cardiomyocytes, whereas Samm50 overexpression evidently decreased the number of red puncta, and most of them overlaid with the green ones in merged images, indicating Samm50 inhibits autophagic flux. Overall, these data demonstrated that Samm50 exacerbates Ang II-induced cardiomyocyte hypertrophy and inhibits mitophagy.

To further verify the effects of Samm50 on cardiac hypertrophy and heart failure, we performed lentivirus-delivered shRNA to ablate samm50 expression. The efficacy of samm50 depletion was confirmed by WB and qRT-PCR analysis, respectively (Figures 3A,B). In accordance with the results observed in Samm50 overexpression, Samm50 deficiency remarkably reduced the expression of hypertrophic markers and ameliorated cardiomyocyte size compared with the control group, as evidenced by qRT-PCR and immunofluorescence (Figures 3C,D). Knockdown of Samm50 also exhibited increased mitophagy, manifested by an increase in LC3-II/LC3-I ratio and a decrease in TOM20 and COX4. Furthermore, in response to Ang II treatment, Samm50 ablation caused a significant accumulation of red puncta, which was greater than that in yellow puncta, indicating Samm50 increases autophagosomes more than autophagosomes and thus stimulates autophagic flux (Figure 3F). These results suggested that Samm50 deficiency ameliorates cardiomyocyte hypertrophy and enhanced mitophagy.

To evaluate the role of mitophagy in Samm50-mediated cardiac hypertrophy, we introduced an autophagy inhibitor into the Samm50-depleted cardiomyocyte prior to Ang II stimulation. Notably, the protective effect of Samm50-deficiency on cardiomyocyte hypertrophy was largely diminished in the presence of autophagy inhibitors, as demonstrated by the increases of hypertrophic markers and cardiomyocyte size (Figures 3G,H), further supporting the idea that mitophagy is required for Samm50-mediated cardiomyocyte hypertrophy.

The previous study has suggested that Samm50-mediated mitophagy is dependent on the Pink1-Parkin pathway under cancer conditions (25). To further elucidate the underlying mechanism of Samm50 in the regulation of myocardial hypertrophy, we first detected the expression of Pink1 and Parkin in response to Ang II stimulation. As seen in Figure 4A, knockdown of Samm50 caused accumulation of Pink1 and Parkin, indicating the activation of mitophagy in cardiomyocytes. However, Pink1 and Parkin expression had no obvious difference between Samm50 overexpression and control group (Figure 4B). To evaluate whether Pink1 is required for Samm50-mediated cardiac hypertrophy, we interfered with the expression of Pink1 in the absence of Samm50. Our data revealed that Samm50 depletion-trigged activation of mitophagy was blunted in cardiomyocytes with simultaneous Pink1 knockout, as indicated by decreased LC3-II/LC3-I ratio and increased protein level of TOM20 and COX4 (Figure 4C). Consistent with immunoblotting results, the protective effect of Samm50 deficiency in response to Ang II induction was blocked by Pink1 knockdown, which was manifested by enhanced hypertrophic markers and cardiomyocyte enlargement (Figures 4D,E).

Based on Samm50 interaction with Pink1 and regulation of its stability (25), we transfected HL-1 cells with HA-tagged Samm50 and Flag-tagged Pink1, followed by Ang II treatment and co-IP analysis to explore whether their physical interaction changes under Ang II stimulation. The co-IP of Samm50 by HA antibody exhibited an interaction between Samm50 and Pink1 in HL-1 cells, while there was no significant difference in the level of Pink1 detected in the absence or presence of Ang II, indicating the interaction between Samm50 and Pink1 was not influenced by Ang II stimulation (Figure 5A).

Pink1 and Parkin were recruited during mitophagy and accumulated in mitochondria (16). To further verify the activation of the Pink1/Parkin pathway in Samm50-mediated hypertrophy, we utilized the immunofluorescent staining to visualize the subcellular localization of Parkin and mitochondria in Samm50-deficient cardiomyocytes. The results showed that Parkin was predominantly stained with mitochondria marker VDAC1 (Figure 5B), indicating the accumulation of Parkin in mitochondria. Overall, these results demonstrate that Samm50 regulates cardiomyocyte hypertrophy in a Pink1/Parkin-dependent manner.