Ten tissue samples of patients with hypertrophic cardiomyopathy and 10 from control donors were enrolled in this study. The hypertrophic heart tissues were obtained from patients who were diagnosed with hypertrophic cardiomyopathy (HCM). The control heart tissues were obtained from vehicle accident victims. The samples were obtained and stored in liquid nitrogen after informed consent was signed by the patients or the families of the donors. The study was approved by the institutional review boards for a clinical study at Jiamusi University.

Cardiac hypertrophy in 8-week-old male C57BL/6N mice was induced by chronic infusion of angiotensin II (Sigma, 1.5 mg/kg/day) for a continuous 4 weeks as described previously (Tang et al., 2017). In brief, an Ang II-contained minipump (ALZET 2004) was placed into the mouse subcutaneously. Mice and SD rats were purchased from Charles River. The protocol for the animal study was approved by the ethics review boards at Jiamusi University.

Cardiomyocytes were isolated from the heart ventricles of 1- to 3-day neonatal SD rats as described previously (Tang et al., 2016). Neonatal rat cardiomyocytes (NRCMs) were then cultured in DMEM medium supplied with 10% fetal bovine serum (FBS, Gibco), 1% antibiotic-antimycotic (ThermoFisher) for 48 h. The NRCMs were cultured in DMEM with 1% fetal bovine serum for 24 h and then the hypertrophy of NRCMs was induced by angiotensin II (1 μM) treatment for 48 h as reffed from previous publications (Luo et al., 2017; Tang et al., 2017). Cardiomyocyte size was measured with Image J software. For each group at each experiment, >100 cardiomyocytes were quantified for average cardiomyocyte size, and the results of three independent experiments were shown.

For gene knockdown or overexpression, the adenovirus system was applied. Adenoviral vectors expressing rat Jmjd1c (Ad-Jmjd1c), control construct (Ad-Ctrl), shJmjd1c (Ad-shJmjd1c, shRNA sequence: GCGCTGACCTTCAAACCATTG), or shPrkaa1 (Ad-shPrkaa1, shRNA sequence: GCACGAGTTGACTGGACATAA) and negative control (Ad-shCtrl) were obtained using the AdEasy Vector kit (Jia et al., 2014).

The cultured cells and heart tissues were washed with cold PBS and then subjected to RNA extraction with TRIZOL reagent (ThermoFisher). Next, the SuperScript III first-strand synthesis system kit (Invitrogen) was applied for cDNA synthesis with 1 ug of total RNA. Finally, SYBR green real-time PCR master mixes (Invitrogen) were applied to analyze the expression of target genes with the cDNA. The primers are shown in Table 1. The double delta CT method was applied for the quantification of mRNA expression.

The cultured cells and heart tissues were washed with cold PBS and then subjected to protein extraction with RIPA buffer (Millipore) supplemented with a phosphatase inhibitor cocktail (Roche) and a protease inhibitor cocktail (Roche). Protein concentration was measured with a BCA kit (ThermoFisher). SDS-PAGE was applied to separate the proteins in gel with 30 ug of total protein; then the proteins were transferred to PVDF membranes (Beyotime) followed by blockade with 5% fat-free milk in TBST for 1 h. Then, the membranes were washed with TBST three times and incubated with primary antibodies at 4°C overnight. The next day, the membranes were washed and incubated with HRP-conjected secondary antibodies. Finally, the expression of proteins was analyzed with the chemiluminescent Western blot detection kit (Millipore). The following primary antibodies were used: anti-Tubulin antibody (Cell Signaling Technology, #2144), anti-JMJD1C antibody (ThermoFisher, #PA5-20804), anti-Histone H3 antibody (Abcam, #ab1791), anti-H3K9me1 antibody (Active Motif, #39249), anti-H3K9me2 antibody (Active Motif, #39683), anti-H3K9me3 antibody (Active Motif, #61013), anti-AMPK antibody (Proteintech, #66536-1-Ig), anti-pAMPK antibody (Cell Signaling Technology, #2535), anti-ACC antibody (Proteintech, #67373-1-Ig), anti-pACC antibody (Cell Signaling Technology, #11818), anti-LKB1 antibody (Santa Cruz Biotechnology, #sc-32245), anti-pLKB1 antibody (Santa Cruz Biotechnology, #sc-271924), anti-CAMKK2 antibody (ThermoFisher, #PA5-69921), anti-PGC1a antibody (Cell Signaling Technology, #2178), anti-O-GlcNAcylation antibody (Abcam, #ab2739), anti-p-p70S6K1 antibody (Cell Signaling Technology, #9205), anti-p70S6K1 antibody (Cell Signaling Technology, #2708), anti-p-eEF2 antibody (Cell Signaling Technology, #2331), anti-eEF2 antibody (Cell Signaling Technology, #2332), anti-H3K4me1 antibody (Abcam, #ab8895), anti-H3K4me3 antibody (Abcam, #ab8580), anti-H3K9ac antibody (Active Motif, #61663).

A ChIP assay was performed with the chromatin immunoprecipitation (ChIP) assay kit (17-295) with ChIP-grade primary antibodies. Immunoprecipitated chromatin fragments were quantified by SYBR-based qRT-PCR, normalized using the input.

The protein synthesis assay was performed by analyzing the incorporation of [¹⁴C]-phenylalanine into proteins of cultured NRVMs as described previously (Sundaresan et al., 2009).

The statistical analysis was performed with SPSS software, and all the values are expressed as mean ± SD. Student's t-test was applied for the analysis of the difference between the two groups. One-way ANOVA followed by Tukey's post hoc test was used when more than two groups exist.

To understand the functions of JMJD1C in cardiac hypertrophy, JMJD1C expression was examined in hypertrophic hearts of humans and mice. We collected heart tissues from 10 cases of hypertrophic cardiomyopathy (HCM) and from 10 control donors. We confirmed the upregulation of hypertrophy-related fetal genes (e.g., ANP, BNP, and MYH7) in HCM tissues compared with the control donors (Figure 1A). Compared with control tissues, the expression levels of JMJD1C mRNA and protein were significantly increased in HCM tissues (Figures 1B,C). Next, we observed that the levels of H3K9me1, H3Kme2, and H3K9me3 were downregulated in HCM tissues, which was associated with the upregulation of JMJD1C (Figure 1C). In addition, cardiac hypertrophy was induced in mice by chronic infusion of Ang II for 4 weeks. The overexpression of hypertrophy-related fetal genes was confirmed in hypertrophic hearts (Figure 1D). In hypertrophic hearts, the expression of JMJD1C was remarkably increased (Figures 1E,F), and the levels of H3K9me1, H3K9me2, and H3K9me3 were decreased (Figure 1F). In addition, we observed the increase in methylation of H3K4 (H3K4me1 and H3K4me3) as well as acetylation of H3K9 in hypertrophic mouse hearts (Supplementary Figure 1). Collectively, these findings demonstrated that JMJD1C was overexpressed in cardiac hypertrophy, which was associated with the downregulation of methylation of H3K9.

Next, we investigated the roles of JMJD1C in the regulation of cardiac hypertrophy using neonatal rat cardiomyocyte (NRCMs). We knocked down the expression of Jmjd1c in NRCMs with adenovirus-mediated shRNA (Figure 2A) and induced cardiomyocyte hypertrophy with Ang II. Treatment with Ang II increased the size of cardiomyocytes and led to overexpression of the hypertrophy-related fetal genes Anp, Bnp, and Myh7. Knockdown of Jmjd1c repressed hypertrophy of cardiomyocytes induced by Ang II treatment (Figures 2B,C). We also overexpressed Jmjd1c in NRCMs with adenovirus-mediated overexpression (Figure 2D). Jmjd1c overexpression promoted the Ang II-mediated increase in the size of cardiomyocytes and the overexpression of hypertrophy-related fetal markers (Figures 2E,F). Protein synthesis is a hallmark of cardiac hypertrophy (Tang et al., 2020a); thus, we tested the effects of JMJD1C on protein synthesis by performing [¹⁴C]-phenylalanine incorporation assay. We observed that JMJD1C knockdown reduced Ang II-induced protein synthesis (Figure 3A), whereas JMJD1C overexpression promoted Ang II-induced protein synthesis (Figure 3B). In addition, we tested the effects of JMJD1C on proteins that critically participate in protein synthesis. We analyzed the phosphorylations of p70S6K1 and eEF2. The results showed that JMJD1C knockdown inhibited the phosphorylations of p70S6K1 and Eef2, whereas the opposite effects were observed in JMJD1C-overexpressed cardiomyocytes (Figures 3C,D). Taken together, these findings revealed that JMJD1C promotes cardiomyocyte hypertrophy partially via regulating protein synthesis.

We next explored the mechanism of JMJD1C function in hypertrophy of cardiomyocytes. Metabolism is critically involved in the physiological and pathological progress of the heart (Tang et al., 2020b). The metabolic sensor AMPK serves as one of the key regulators of glucose and fatty acid metabolism of cardiomyocytes, and the deregulation of AMPK signaling participates in hypertrophic cardiomyopathy and other cardiac diseases. In this study, we observed that the phosphorylation levels of the kinase AMPK and its downstream factor ACC were significantly downregulated in hypertrophic hearts in humans and mice (Figures 4A,B). Additionally, we observed that Jmjd1c knockdown promoted the activation of AMPK signaling, whereas Jmjd1c overexpression repressed AMPK signaling activation in Ang II-treated NRCMs (Figures 4C,D). AMPK was reported to regulate cardiac hypertrophy by inducing PGC1a (Peroxisome proliferator–activated receptor gamma coactivator 1-alpha) expression and reducing protein O-GlcNAcylation (Watanabe et al., 2014; Gelinas et al., 2018). Indeed, we observed that Jmjd1c knockdown increased the PGC1a protein level and reduced total protein O-GlcNAcylation, which was blocked by AMPK knockdown (Supplementary Figure 2). Therefore, these findings support the notion that JMJD1C represses AMPK signaling during cardiac hypertrophy.

We attempted to investigate the molecular mechanism underlying JMJD1C-mediated AMPK inactivation. AMPK is generally activated by two kinases, LKB1 and CAMKK2. We observed that LKB1 expression and activation and its interaction partners STRAD and MO25 were not affected by Jmjd1c knockdown or overexpression (Figures 5A,B). Interestingly, we observed that Jmjd1c knockdown increased the protein levels of CAMKK2, whereas Jmjc1c overexpression downregulated the protein levels of CAMKK2 in NRCMs (Figure 5C). Importantly, we showed that JMJD1C controlled the expression of CAMKK2 at the transcriptional level because JMJD1C overexpression reduced and knockdown increased the levels of CAMKK2 mRNA (Figure 5E). Next, we analyzed the mechanism by which JMJD1C regulated CAMKK2 expression. JMJD2A and JMJD3 were reported to regulate cardiac hypertrophy by epigenetically targeting the FHL1 (four-and-a-half LIM domains 1) and the β-MHC promoter, respectively (Zhang et al., 2011; Guo et al., 2018). Our chromatin immunoprecipitation assay did not reveal the enrichment of JMJD2A and JMJD3 at the CAMKK2 promoter nor of JMJD1C at the FHL1 or MHC promoters. By contrast, JMJD1C selectively bound the CAMKK2 promoter (Figure 5F). In addition, overexpression of JMJD1C reduced H3K9me1/2/3 levels at the CAMKK2 promoter (Figure 5G, Supplementary Figure 3). Finally, we treated NRCMs with the CAMKK2 inhibitor STO-609 and observed that STO-609 treatment blocked the roles of JMJD1C on the activation of the AMPK pathway (Figure 5H). These results demonstrated that JMJD1C controls AMPK signaling via a CAMKK2-dependent mechanism.

Finally, we investigated whether JMJD1C-mediated AMPK inactivation was critically involved in the function of JMJD1C. We knocked down the expression of AMPK (Prkaa1) in NRCMs (Figure 6A) and induced cardiomyocyte hypertrophy with Ang II. We observed that AMPK knockdown promoted the hypertrophic response induced by Ang II and blocked the prevention of cardiomyocyte hypertrophy mediated by Jmjd1c knockdown (Figures 6B,C). In addition, we observed that metformin, an AMPK activator, rescued AMPK activation in NRCMs with Jmjd1c overexpression (Figure 6D). Significantly, metformin treatment repressed cardiomyocyte hypertrophy and blocked the function of JMJD1C overexpression (Figures 6E,F). Finally, we also tested whether JMJD1C promoted cardiomyocyte hypertrophy via CAMKK2. We inhibited CAMKK2 with STO-609 or knocked down Camkk2 expression and observed that either inhibition of CAMKK2 with STO-609 or Camkk2 knockdown blocked the inhibitory effects of Jmjd1c knockdown on cardiomyocyte hypertrophy (Figures 6G,H). Therefore, AMPK signaling critically participated in the roles of JMJD1C during the hypertrophic response of cardiomyocytes.