The Ankrd1 (ankyrin repeat domain 1) gene is known to be up-regulated in heart failure and acts as a co-activator of p53, modulating its transcriptional activity, but it remains inconclusive whether this gene promotes or inhibits cell apoptosis. In the present study, we attempted to investigate the role of Ankrd1 on AngII (angiotensin II)- or pressure-overload-induced cardiomyocyte apoptosis. In the failing hearts of mice with pressure overload, the protein expression of Ankrd1-encoded CARP (cardiac ankyrin repeat protein) was significantly increased. In NRCs (neonatal rat cardiomyocytes), AngII increased the expression of Ankrd1 and CARP. In the presence of AngII in NRCs, infection with a recombinant adenovirus containing rat Ankrd1 cDNA (Ad-Ankrd1) enhanced the mitochondrial translocation of Bax and phosphorylated p53, increased mitochondrial permeability and cardiomyocyte apoptosis, and reduced cell viability, whereas these effects were antagonized by silencing of Ankrd1. Intra-myocardial injection of Ad-Ankrd1 in mice with TAC (transverse aortic constriction) markedly exacerbated cardiac dysfunction with an increase in the lung weight/body weight ratio and a decrease in left ventricular fractional shortening. Cardiomyocyte apoptosis and the expression of phosphorylated p53 were also significantly increased in Ad-Ankrd1-infected TAC mice, whereas knockdown of Ankrd1 significantly inhibited the apoptotic signal pathway as well as cardiomyocyte apoptosis in pressure-overload mice. These findings indicate that overexpression of Ankrd1 exacerbates pathological cardiac dysfunction through enhancement of cardiomyocyte apoptosis mediated by the up-regulation of p53.
In clinical settings, there is a marked increase in myocardial CARP expression in patients with ischaemic or dilated cardiomyopathy, meanwhile Ankrd1 mutation could again result in dilated cardiomyopathy. However, it remains unknown whether Ankrd1 overexpression is beneficial for, detrimental to or even has no significant influence on HF. Although there is research demonstrating that CARP could activate p53 in skeletal muscle cells, inhibit the phosphorylation of ERK in cardiomyocytes and enhance apoptosis in hepatoma cells, it is unclear whether CARP could also enhance cardiomyocyte apoptosis.
In the present study, we found that overdose infection of Ad-Ankrd1 could directly result in the death of cardiomyocytes, whereas overexpression of Ad-Ankrd1 at a safe dose and under AngII or pressure stimulation could enhance cardiomyocyte apoptosis and thus worsen HF. This leads to further findings that overexpression of Ankrd1 could increase mitochondrial membrane permeability in cardiomyocytes, up-regulate p53 activity in mitochondria, and also increases mitochondrial translocation of Bax.
In chronic HF patients with marked up-regulation of CARP, therapeutic approaches targeting the down-regulation of CARP could possibly improve heart function. In the present study, we found that AngII could up-regulate CARP, indicating that the improvement of HF facilitated by angiotensin-converting enzyme inhibitor and AngII receptor antagonist is achieved at least partially by inhibiting the up-regulation of CARP and consequently reducing cardiomyocyte apoptosis.
Ankrd1 (ankyrin repeat domain 1) was identified as a fetal gene, the expression of which is augmented in both animals and humans with HF (heart failure) [1,2], but the role of Ankrd1/CARP (cardiac ankyrin repeat protein) in HF remains unclear. Although there is clinical evidence that Ankrd1 mutations are involved in the pathogenesis of dilated cardiomyopathy [3–5], there is no consensus on whether accumulation of CARP is beneficial or detrimental to HF.
Apoptosis has been demonstrated to play a critical role in the progression of HF . It is believed that the tumour-suppressor genes p53 and Bax contribute to cardiomyocyte apoptosis [7,8]. Coincidentally, there is recent evidence that CARP acts as a co-activator of p53 and modulates its transcriptional activity , whereas overexpression of CARP inhibits the phosphorylation of ERK (extracellular-signal-regulated kinase)  and impairs the contractility of engineered heart tissues [11,12]. These findings led to the hypothesis that CARP might accelerate the progression of HF by enhancing cardiomyocyte apoptosis. Nowadays, mitochondria are considered to be the central executioners of apoptosis, but the influence of CARP on mitochondrial function is completely unknown.
To date, there has been a paucity of literature on the topic of Ankrd1 influencing apoptosis, among which three reports from a Korean laboratory have demonstrated that the ectopic expression of Ankrd1 enhances apoptotic cell death in hepatoma cells , whereas it increases anoxia-induced apoptosis in rat embryonic heart-derived H9c2 cells ; however, the down-regulation of Ankrd1 is associated with the increase in cell apoptosis during myocardial ischaemia/reperfusion . These findings suggest that the influence of Ankrd1 on cell apoptosis might be context-dependent. AngII (angiotensin II) and left ventricular overload both can up-regulate Ankrd1 and induce cardiomyocyte apoptosis, but it remains unclear whether the up-regulated Ankrd1 inhibits or enhances cell apoptosis. In the present study, we used AngII-stimulated cultured cardiomyocytes and mice that had undergone TAC (transverse aortic constriction) to examine the following issues: (i) the influence of Ankrd1 overexpression or silencing on apoptosis in cultured cardiomyocytes; (ii) whether Ankrd1/CARP modulates p53 and mitochondrial dysfunction of cardiomyocytes; (iii) the effects of Ankrd1 overexpression or silencing on cardiomyocyte apoptosis in left-ventricular-pressure-overloaded mice.
MATERIALS AND METHODS
C57BL/6 male mice (8–10 weeks old, weighing 22–25 g, provided by the Animal Center of Southern Medical University) were anaesthetized with a mixture of xylazine (5 mg/kg, intraperitoneal) and ketamine (100 mg/kg, intraperitoneal), and the adequacy of anaesthesia was monitored by the disappearance of pedal withdrawal reflex. An endotracheal tube was inserted, and artificial respiration was supported by a respiratory machine. The pressure-overload model was created by TAC as described previously [16,17]. Transthoracic echocardiography was performed at 4 weeks after TAC or sham operation and then the mice were killed by overdose of anaesthesia with pentobarbital sodium (150 mg/kg, intraperitoneal) and cervical dislocation.
All procedures were performed in accordance with our Institutional Guidelines for Animal Research and the investigation conformed to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised in 1996).
NRC (neonatal rat cardiomyocyte) culture
The neonatal rats at 1–3 days after birth were killed by 2% isoflurane inhalation, and the heart of each was removed. Isolation and culture of NRCs was performed as described previously . The cell type was verified using immunochemistry assay (anti-myoactin antibody, Wuhan Boster Biological Technology) as we reported previously . Cells were cultured for 4 days and then treated with 1 μM AngII or recombinant adenovirus containing the rat Ankrd1 cDNA (Ad-Ankrd1) or AAV (adeno-associated virus) carrying shRNA targeting Ankrd1 (sh-Ankrd1). Cell viability was determined using the MTT (Sigma-Aldrich Co) assay according to the corresponding manufacturer's instructions.
Construction and infection by Ad-
Ankrd1 or sh- Ankrd1
Construction of Ad-
The full-length cDNA of rat Ankrd1 was inserted into the vector pUC57, and then subcloned into the shuttle vector pDC316-mCMV-EGFP. The Ankrd1 cDNA clones were sequenced completely to confirm the absence of cloning artefacts and mutations. The Ad-MAX system was used for the generation of recombinant adenovirus carrying Ankrd1 cDNA or empty vector (pEGFP). Briefly, pDC316-mCMV-EGFP-ANKRD1 and virus backbone plasmid pBHGloxdelIE13cre were co-transfected into cultured HEK (human embryonic kidney)-293 cells by using Lipofectamine™ 2000 (Invitrogen), then the recombinant adenovirus was collected and amplified in HEK-293 cells.
The overexpression of Ankrd1 (not an EGFP-fusion protein) was achieved by transfecting cultured cardiomyocytes with Ad-Ankrd1 [MOI (multiplicity of infection) of 10]. For in vivo infection, mice that have received TAC or sham for 2 weeks were anaesthetized as mentioned above, intubated and ventilated with room air. Then a left lateral thoracotomy was made to expose the beating heart followed by myocardial injection of 5×1011 Ad-Ankrd1 or Ad-EGFP into three points of the left ventricular wall.
Construction of sh-
pAAV2/9-CMV-ZsGreen vectors carrying sh-Ankrd1 or negative control were generated by a professional company (Biowit). For in vitro transfection, pAAV2/9-CMV-ZsGreen-sh-Ankrd1 or negative control virus particles (5×105 viral genomes/cell) were added to cultured NRCs. After transfection for 96 h (such a length of time is necessary to increase the efficiency of in vitro AAV infection), infection efficiency and silencing effect were evaluated using fluorescence microscopy and Western blotting respectively. For in vivo infection, pAAV2/9-CMV-ZsGreen-sh-Ankrd1 or control virus particles (1011 viral genomes/ml) were administered by direct injection in the left ventricular free wall (two sites, 10 μl/site) in mice at 4 weeks of age using a syringe with a 30-gauge needle, and 4 weeks later, sham or TAC surgery was performed.
Transduction efficiency of in vivo gene transfer by adenovirus or AAV was assessed by EGFP fluorescence (510 nm) in cryosectioned heart slices using fluorescence microscopy.
Total RNA of homogenized murine whole heart or cultured cardiomyocytes was isolated using Total RNA Kit II (Omega) according to the protocol provided by the manufacturer. Quantitative real-time PCR using a Quantitect SYBR Green RT–PCR kit (Qiagen) and an Applied Biosystems 7500 system targeting the Ankrd1, Anp (atrial natriuretic peptide) and Gapdh (glyceraldehyde-3-phosphate dehydrogenase) genes was performed. Primers are shown in Table 1.
|Transcripts .||Forward primer (5′→3′) .||Reverse primer (5′→3′) .||Size (bp) .|
|Transcripts .||Forward primer (5′→3′) .||Reverse primer (5′→3′) .||Size (bp) .|
Proteins were extracted from the cultured cardiomyocytes or their mitochondria, or the murine heart. Immunoblotting was performed using antibodies against CARP (Santa Cruz Biotechnology), p53, phosphorylated p53 (p-p53) and Bax (Cell Signaling Technology). Blotting of β-actin (Santa Cruz Biotechnology) or COX IV (cytochrome c oxidase subunit IV) (Cell Signaling Technology) or GAPDH (Santa Cruz Biotechnology) was used as a loading control. Immunoreactive bands were visualized with Odyssey® Infrared Imaging System (LI-COR® Bioscience) and quantified by densitometry with ImageJ software (NIH).
Measurements of mitochondrial membrane potential and cell apoptosis
To analyse the mitochondrial depolarization, the NRCs were labelled with TMRE (tetramethylrhodamine ethyl ester) (Molecular Probes) and DAPI. Myocyte suspensions were loaded with 100 nM TMRE by incubation for 15 min at room temperature. The fluorescence intensity of TMRE was monitored at 582 nm.
Apoptosis in NRCs was detected by Hoechst staining as we described previously ; cells were fixed with 4% (w/v) formaldehyde in PBS for 10 min, and then stained with Hoechst 33258 (10 mg/l) (Beyotime) for 15 min, and the results were observed under fluorescence microscopy. Apoptotic cells were indicated by bright white nuclei. Apoptosis in the myocardium was determined using TUNEL (terminal deoxynucleotidyltransferase-mediated dUTP nick-end labelling) assay . Briefly, apoptotic cells were detected with an In Situ Cell Death Detection Kit, TMR red (Roche). Tissue samples of heart embedded in paraffin were used. After deparaffinization, the sections were treated with proteinase K for 20 min, incubated with TUNEL reaction mixture or negative control solution for 60 min at 37°C and then stained with DAPI solution for 10 min. Slides were rinsed twice with PBS between each step. The positive rate of TUNEL-labelled nuclei was calculated from four different randomly selected areas under confocal microscopy.
Data are expressed as the means±S.E.M. Statistical significance was analysed using Student's unpaired t test, or one- or two-way ANOVA, followed by Bonferroni's correction for post-hoc multiple comparisons. In addition, the least-squares method was used to analyse the linear correlation of the selected variables. In all analyses, P<0.05 was considered to indicate statistical significance.
Ankrd1/CARP was up-regulated in response to HF and AngII stimulation
Because ANP is a recognized marker of HF, we analysed the correlation between Ankrd1 and Anp mRNA levels in mice with TAC or sham-operated mice over 4–8 weeks. Real-time PCR revealed that the level of Ankrd1 expression was positively correlated with that of Anp, as shown in Figure 1(A) (r=0.914, P<0.01, n=14). Myocardial CARP was also markedly up-regulated in mice with HF induced by TAC according to Western blot analysis (Figure 1B). CARP levels were also positively correlated with the LW/BW (lung weight/body weight ratio) (Figure 1C), a critical parameter of congestive HF in the pressure-overload model. In cultured cardiomyocytes, AngII stimulation for 24 h markedly increased the expression of CARP (Figure 1D), in agreement with our recent study . These findings indicate that the expression pattern of myocardial Ankrd1/CARP is similar to that of natriuretic peptides in response to the pathological stimulation of HF. To clarify the role of Ankrd1 in AngII-stimulated cardiomyocytes, we infected NRCs with Ad-Ankrd1 or sh-Ankrd1. As a result, the infective efficiency was satisfactory at an MOI of 10 in Ad- or 5×105 viral genomes/cell in AAV-infected cells (Figure 1E), and the level of CARP significantly increased ~4-fold in Ad-Ankrd1-infected cells and decreased by ~70% in sh-Ankrd1-infected cells (Figure 1F).
Ankrd1/CARP up-regulation in heart failure
Ankrd1 overexpression increased cardiomyocyte apoptosis
High-dose infection (MOI of 30) of Ad-Ankrd1 resulted in the direct death of most cardiomyocytes, whereas there was no significant cell death when the cardiomyocytes were infected with the same dose of Ad-EGFP (Figure 2A), and it was safe when the MOI of Ad-Ankrd1 was no larger than 20. Forced Ankrd1 overexpression (MOI of 10) of Ad-Ankrd1 significantly reduced the viability of NRCs (Figure 2B) and increased apoptosis in the presence/absence of AngII stimulation (Figures 2C and 2D), whereas the decrease in cell viability and the increase in apoptosis induced by AngII were antagonized partially by addition of sh-Ankrd1 (Figures 2B–2D). These findings suggest that AngII exerted Ankrd1-dependent and -independent effects on cardiomyocyte apoptosis.
Ankrd1 overexpression or knockdown on apoptosis in NRCs
Ankrd1 exerted effects on p53 activity and Bax expression
In order to clarify the mechanism of the pro-apoptotic effect of CARP on cardiomyocytes, we examined the changes of apoptosis-related molecules by Western blotting, and found that Ankrd1 overexpression significantly increased the levels of p53, p-p53 (Figure 3A) and mitochondrial Bax (Figure 3B). Using immunocytochemistry, we found that there was a significant increase in the phosphorylation of p53 in the mitochondria of Ad-Ankrd1-infected cardiomyocytes (Figure 3C). Western blot analysis confirmed further that Ad-Ankrd1 reduced nuclear p-p53 and increased mitochondrial p-p53, suggesting a mitochondrial translocation of p-p53 (Figure 3D).
Effects of Ad-
Ankrd1 infection on apoptotic signal in NRCs
Effects of sh-
Ankrd1 infection on apoptotic signal in NRCs
Ankrd1 increased mitochondrial permeability
In addition, mitochondrial depolarization was investigated using TMRE staining. We found that TMRE fluorescence was markedly reduced in the NRCs exposed to AngII, whereas the decrease in fluorescence was augmented when cells were infected with Ad-Ankrd1 (Figure 5). AngII-induced TMRE fluorescence reduction was antagonized by sh-Ankrd1 (Figure 5). These findings suggest that Ankrd1 would enhance AngII-induced increase in mitochondrial permeability.
Effect of Ad-
Ankrd1 or sh- Ankrd1 on mitochondrial permeability in NRCs
Cardiac overexpression of
Ankrd1 exaggerated HF
To investigate the role of CARP in vivo, cardiac overexpression of Ankrd1/CARP was achieved by intramyocardial injection of Ad-Ankrd1 at 2 weeks after TAC. Another 2 weeks later, total myocardial CARP expression was significantly higher in TAC+Ad-Ankrd1 mice than in TAC+Ad-EGFP mice (Figures 6A and 6B). Echocardiography showed that the LVEDd (left ventricular end-diastolic dimension) and LVESd (left ventricular end-systolic dimension) were larger in TAC+Ad-Ankrd1 mice than in TAC+Ad-EGFP mice (Figures 6C and 6D). To investigate the influence of Ankrd1 overexpression on cardiac function, the LW/BW was determined as an index of pulmonary congestion and LVFS (left ventricular fractional shortening) was measured by echocardiography to assess systolic function in Ad-Ankrd1-infected mice. LVFS showed a marked decrease in TAC+Ad-EGFP mice (Figure 6E). The LW/BW was significantly larger in TAC+Ad-Ankrd1 mice than in TAC+Ad-EGFP mice (11.0±1.25 mg/g compared with 6.97±0.38 mg/g, P<0.05; Figure 6F).
Myocardial injection of Ad-
Ankrd1 in mice promoted cardiac remodelling
Cardiac overexpression of
Ankrd1 enhanced apoptosis
Expression of p53 and p-p53 protein and apoptosis were evaluated. We noted that phosphorylation of p53 was increased in TAC and TAC+Ad-EGFP groups, and further increased in TAC+Ad-Ankrd1 mice detected by Western blotting (Figures 7A and 7B). TUNEL assay showed that myocardial apoptosis was significantly enhanced in TAC and TAC+Ad-EGFP groups, and the percentage of cell apoptosis in TAC+Ad-Ankrd1 mice was markedly higher than in the TAC+Ad-EGFP group (Figures 7C and 7D). Immunostaining of Bax determined by immunohistochemistry and Western blotting was also enhanced in TAC+Ad-Ankrd1 mice (Figures 7E and 7F).
Ankrd1 overexpression enhanced TAC-induced cardiac apoptosis
Ankrd1 inhibited apoptosis in pressure-overload mice
Intramyocardial injection of Sh-Ankrd1 4 weeks before TAC or sham produced satisfactory infection efficiency (Figures 8A and 8B). TUNEL assay showed that myocardial apoptosis was significantly increased in TAC and TAC+AAV-zsGreen groups, and the percentage of cell apoptosis in TAC+sh-Ankrd1 mice was significantly lower than in the TAC+AAV-zsGreen group (Figures 8C and 8D). Immunostaining of Bax determined by immunohistochemistry and Western blotting was also reduced in TAC+sh-Ankrd1 mice (Figures 8E and 8F), so did myocardial p53 and p-p53 (Figure 8G).
Ankrd1 attenuated TAC-induced cardiac apoptosis
It is generally believed that Ankrd1/CARP is a versatile factor which exerts pleiotropic effects on cardiomyocytes at multiple levels . Although Ankrd1/CARP expression is known to be up-regulated in response to HF with different aetiology [2,21,22], it remains unclear whether myocardial overexpression of Ankrd1 in HF is causally related to the development of a malignant cardiac phenotype or whether it is merely an adaptive response that delays the progression of HF. In the present study, we provided both in vitro and in vivo evidence indicating that overexpression of myocardial Ankrd1/CARP worsens heart failure, and these deleterious effects were associated with the enhancement of apoptosis mediated at least partly by the accumulation of p53 and mitochondrial dysfunction (Figure 9).
Model of CARP involvement in signal pathway in cardiomyocyte apoptosis
We demonstrated that overdose (MOI of 30) infection of Ad-Ankrd1 in cultured cardiomyocytes directly induced cell death, whereas the cells survived well with the same dose of Ad-EGFP infection, suggesting cytotoxicity of Ankrd1 overexpression. We also noticed that with the infective dose (MOI of 10) that would not influence cell viability when there was no pathological stimulation, but the viability of cardiomyocytes would decrease while the apoptosis would increase when co-treated with AngII. In pressure-overloaded hearts infected with Ad-Ankrd1 in vivo, cardiomyocyte apoptosis also significantly increased. These results indicate that overexpression of Ankrd1 enhances apoptosis in cardiomyocytes under AngII or pressure-overload stimulation.
It is well known that p53 protein has a pro-apoptotic effect. A close relationship between p53 and muscle ankyrin repeat proteins has been reported [9,23]. Ankrd1/CARP was demonstrated to act as a co-activator of p53  and also inhibit ERK phosphorylation , implying that CARP is a pro-apoptotic factor. In the present study, we demonstrated that CARP promotes cardiomyocyte apoptosis via a p53-dependent mitochondrial pathway. Mitochondrial dysfunction also induces the activation of calcineurin, leading to apoptosis . It has been demonstrated that activation of Bax by p53 is the decisive signal leading to cell death , which is in support of our findings that overexpression of Ankrd1/CARP promoted apoptosis by inducing p53 activation, increasing mitochondrial translocation of Bax, and mitochondrial membrane permeability. In addition, AngII can induce both mitochondria-dependent and DNA-damage-dependent apoptosis [26,27]. Our finding that Ad-Ankrd1 led to a translocation of p-p53 from nuclei to mitochondria suggests that the Ankrd1-induced apoptosis is independent of DNA damage.
There are multiple levels of regulation of mitochondria-directed apoptosis by p53. When the cells are exposed to oxidative stress, p53 accumulates in the mitochondrial matrix and triggers opening of the MPTP (mitochondrial permeability transition pore) , where MPTP opening is well known to induce apoptosis [18,29]. The present study implies that the mitochondrial Ankrd1–p53 axis is a contributor to AngII-induced cardiomyocyte apoptosis.
There is a relative paucity of literature addressing the role of Ankrd1 on apoptosis. Park et al.  reported that the ectopic expression of Ankrd1 leads to a decrease in colony formation and an increase in apoptotic cell death in hepatoma cells, which was consistent with our findings in the present study, suggesting that Ankrd1 up-regulation is positively correlated with apoptotic cell death. In addition, mitogen-activated protein kinase p38, Rac1 and α-adrenergic stimulation were demonstrated to stimulate Ankrd1 promoter activity [30,31], whereas p38, Rac1 and α-adrenergic signalling have been implicated in cardiomyocyte apoptosis. Moreover, Ankrd1 was first identified as a cytokine-inducible gene. TNFα (tumour necrosis factor α), IL-1β (interleukin 1β) or TGFβ (transforming growth factor β) can up-regulate Ankrd1 [32,33], whereas it is well known that AngII is able to induce the production of cytokines such as IL-1β and TNFα as well as up-regulation of TGFβ. Reports from our laboratory  and others  showed further that pressure-overload or AngII can increase the expression of the chemokine fractalkine; moreover, we confirmed that fractalkine significantly increases the expression of both IL-1β and TNFα . These lines of evidence suggest that pro-apoptotic factors are able to up-regulate Ankrd1 (Figure 9). Unexpectedly, the down-regulation of Ankrd1 secondary to overexpression of GADD153 (growth-arrest and DNA-damage-inducible protein 153), an apoptosis-related gene, was noted to contribute to the apoptosis induced by mimicked ischaemia/reperfusion injury in cultured cardiomyocytes, whereas forced expression of Ankrd1 was reported to enhance the resistance to hypoxia-induced apoptosis in H9c2 cardiomyocytes [14,15]. This discrepancy to our findings suggests that the role of Ankrd1 in cardiomyocyte apoptosis depends on the type of pathological stimulation.
In HF induced by ischaemia, dilated cardiomyopathy or arrhythmogenic right ventricular dysplasia, there was marked up-regulation of Ankrd1 expression, and a previous study has shown that CARP up-regulation is correlated with pro-ANP expression levels . However in adriamycin-induced HF, there was down-regulation of Ankrd1. It was therefore hard to find out the cause-and-effect relationship between Ankrd1 and HF. Ankrd1 gene mutation as a gain of function could result in hypertrophic cardiomyopathy , whereas Ankrd1 gene mutation as a loss of function could result in dilated cardiomyopathy [3,5], which indicates that the normally expressed Ankrd1 gene plays an important role in maintaining myocardial function. Among the up-regulated genes during HF, there are not only genes such as those encoding ANP and BNP (brain natriuretic peptide) [29,30] that compensatively inhibit HF, but also ones such as those encoding p53 [35,36] that accelerate the progression of HF. It is therefore necessary to know how exactly Ankrd1 would influence HF. Our research has found that, during chronic HF, the expression of Ankrd1 was positively correlated with the expression of ANP; however, overexpression of Ankrd1 would result in cardiomyocyte apoptosis and thus worsen HF, which indicates a therapeutic potential in regulating the expression of Ankrd1 in cardiomyocytes.
recombinant adenovirus containing Ankrd1
recombinant adenovirus containing EGFP
ankyrin repeat domain 1
atrial natriuretic peptide
cardiac ankyrin repeat protein
IV, cytochrome c oxidase subunit IV
human embryonic kidney
left ventricular fractional shortening
lung weight/body weight ratio
multiplicity of infection
mitochondrial permeability transition pore
neonatal rat cardiomyocyte
shRNA targeting Ankrd1
transverse aortic constriction
transforming growth factor β
tetramethylrhodamine ethyl ester
tumour necrosis factor α
terminal deoxynucleotidyltransferase-mediated dUTP nick-end labelling
Liang Shen and Ci Chen designed the experiments, performed most of the in vitro and in vivo experiments and interpreted the data. Xuan Wei, Xixian Li, Guangjin Luo and Jingwen Zhang carried out the experiments of cell culture and pressure overload. Ci Chen and Jingwen Zhang wrote the paper. Jianping Bin, Xiaobo Huang, Shiping Cao and Guofeng Li analysed and interpreted the data. Yulin Liao designed the study, analysed and interpreted the data, wrote and revised the paper.
This work was supported by grants from the National Natural Science Foundation of China [grant numbers 81170146 and 31271513 (to Y.L.)], the National Program on Key Basic Research Project [grant number 2012CB945100 (to Y.L.)], the Team Program of the Natural Science Foundation of Guangdong Province, China [grant number S2011030003134 (to Y.L. and B.J.)].
These authors contributed equally to this work.