ndufa7 plays a critical role in cardiac hypertrophy

Abstract Cardiac hypertrophy is a common pathological change in patients with progressive cardiac function failure, which can be caused by hypertrophic cardiomyopathy (HCM), dilated cardiomyopathy (DCM) or arterial hypertension. Despite years of study, there is still limited knowledge about the underlying molecular mechanisms for cardiac hypertrophy. NDUFA7, a subunit of NADH:ubiquinone oxidoreductase (complex I), has been reported to be a novel HCM associated gene. However, the biological role of NDUFA7 in heart remains unknown. In this study, we found that NDUFA7 exhibited high expression in the heart, and its level was significantly decreased in mice model of cardiac hypertrophy. Moreover, we demonstrated that ndufa7 knockdown in developing zebrafish embryos resulted in cardiac development and functional defects, associated with increased expression of pathological hypertrophy biomarkers nppa (ANP) and nppb (BNP). Mechanistic study demonstrated that ndufa7 depletion promoted ROS production and calcineurin signalling activation. Moreover, NDUFA7 depletion contributed to cardiac cell hypertrophy. Together, these results report for the first time that ndufa7 is implicated in pathological cardiac hypertrophy.


| INTRODUC TI ON
Cardiac hypertrophy is an adaptive response of the heart to haemodynamic and neurohormonal stress to preserve cardiac function. 1 Prolonged cardiac hypertrophy can contribute to functional decompensation, cardiac fibrosis, even progress to heart failure or sudden death, leading to a high mortality worldwide. 2 Cardiac hypertrophy is accompanied by reactivation of a set of cardiac foetal genes, including atrial natriuretic peptide (ANP, nppa), brain natriuretic peptide (BNP, nppb) and β-myosin heavy chain (β-MHC), as well as an increase in cell size and protein synthesis, suggesting that molecular events controlling heart development are redeployed to regulate hypertrophic growth. 3 Among many intracellular signalling pathways involved in hypertrophic development, calcineurin/ nuclear factor of activated T cell (NFAT) pathway plays a key role in pathological cardiac hypertrophy. 4 NFAT, resides in the cytoplasm in a hyper-phosphorylated state, is dephosphorylated by the calcium/ calmodulin-activated phosphatase calcineurin and then translocate to the nucleus where it regulates transcription of hypertrophic genes. 5 The heart is the greatest oxygen-consuming organ in the body, which requires ATP production from mitochondrial respiration to maintain normal cardiac mechanical function. 6 Therefore, mitochondrial dysfunction plays a key role in the pathogenesis and development of cardiac hypertrophy. 7 Reactive oxygen species (ROS) are a natural by-product of mitochondrial energy production. 8 ROS are generated during excessive oxidative stress, which has been implicated in the pathogenesis of cardiac hypertrophy and heart failure. 9 Interestingly, it has been reported that calcineurin regulates the pathogenesis of cardiac hypertrophy with accompanying by the intracellular ROS production. 10 It has been demonstrated that ROS stimulates the signal transduction in cardiomyocytes during pathological conditions by activating NFAT. 11 NDUFA7, also known as B14. 5, encodes a subunit of NADH:ubiquinone oxidoreductase (complex I) in the mitochondrial respiratory chain. It has been reported that an association is observed between NDUFA7 and rheumatoid arthritis (RA) with severe erosive arthritis. 12 Our previous work of whole exome sequencing analysis on hypertrophic cardiomyopathy patients found that NDUFA7 might be a novel candidate gene associated with cardiomyopathy. 13 However, the biological role of NDUFA7 in cardiac diseases has not been explored. Here we present the first evidence to our knowledge that ndufa7 depletion contributes to pathological cardiac hypertrophy.

| Tissue expression profiles and animal model data
The Genotype-Tissue Expression database (GTEx, http://www. gtexp ortal.org/) was used to investigate the expression of candidate genes in multiple human tissues. The zebrafish information network (ZFIN, http://zfin.org/) and the mouse genome database (MGD, http://www.infor matics.jax.org/) databases, as well as PubMed, were used to investigate the phenotype of candidate genes in zebrafish and mice. Clustal Omega software (https://www.ebi.ac.uk/ Tools/ msa/clust alo/) was used for multiple sequence alignment.
As a negative control, a standard control oligonucleotide (Control MO: 5'-CCTCTTACCTCAGTTACAATTTATA-3') was injected at the same dose.

| RNA isolation, reverse transcription (RT-PCR) and quantitative real-time PCR (qPCR)
Total RNA was isolated from whole embryos, larval hearts, cells or tissues with Trizol reagent (Invitrogen) and reverse-transcribed into cDNA with an Oligo dT primer and SuperScript II reverse transcriptase (Invitrogen). The efficiency of the ndufa7 splice MO was tested by carrying out RT-PCR from RNA extracted from zebrafish whole embryos at 2dpf with ndufa7 primers. For the analysis of specific gene expression in zebrafish larval hearts or cells, realtime PCR was performed on all samples in triplicate with a Power SYBR Green PCR Master Mix (Applied Biosystems) as described previously. 14,15 The 2 -△△Ct method was used to normalize the gene of interest to the endogenous housekeeping gene rpl13a (ribosomal protein L13a) or GAPDH and determine the fold change relative to control. Primers used for experiments were listed in Table 1.

| Whole-mount in situ hybridization (ISH)
For all manipulations, embryos at the appropriate developmental stage, 1 cell, 24 hpf (hour post-fertilization), 48 hpf, 72 hpf and 96 hpf were fixed in 4% paraformaldehyde/PBS. In situ hybridization and immunohistochemistry were performed using standard protocols. Digoxigenin-labelled riboprobes were prepared from PCR templates with T3 and T7 polymerases as recommended by the supplier (Roche). Staining was visualized using an anti-DIG-alkaline phosphatase-conjugated antibody and NBT/BCIP (Roche). After probe detection, embryos were cleared and photographed in glycerol.

| Luciferase assay
The nppb::F-Luc embryos injected with morpholinos or treated with chemicals were placed into a 96-well microtiter plate with 100 μL buffered embryo water and incubated at 28°C for embryonic TA B L E 1 Primers for qPCR or ISH analysis

| Statistical analysis
Data are presented as mean ± SEM using Image GraphPad Prism 5.0 software (GraphPad Software). Analysis of statistical significance was performed by the Student's t test for comparison between two groups. A P value of less than .05 was deemed statistically significant.

| The expression level of NDUFA7 is markedly decreased in cardiac hypertrophy
We first checked the expression level of NDUFA7 in tissues using NIH's Genotype-Tissue Expression (GTEx) database (https://commo nfund.nih.gov/gtex) and found that NDUFA7 displayed relatively high expression in human heart and muscle ( Figure 1A). We extracted RNA from different tissues of mice, performed RT-qPCR analysis F I G U R E 1 NDUFA7 is involved in cardiac hypertrophy induced by ISO infusion. (A) Tissue expression profiles of NDUFA7 are generated from GTEx database (https://commo nfund.nih.gov/gtex). Median RPKM level was shown. (B) RNA was extracted from different tissues of mice, and RT-qPCR analysis of NDUFA7 and GAPDH expression was then performed. (C) NDUFA7 expression in hearts of mice subjected to cardiac pressure overload by transverse aortic constriction (TAC) or in control group (GSE2459). *P < .05 compared with healthy control. (D) Similarity of Ndufa7 protein sequence from different species including human, chimpanzee, dog, cattle, mouse, rat, chicken and zebrafish and found high expression level of NDUFA7 mRNA in mice heart, which is consistent with GTEx database finding ( Figure 1B). We then searched the GEO (gene expression omnibus) database to explore whether NDUFA7 is involved in cardiac diseases. Compared with control group, NDUFA7 expression was significantly decreased in the mice heart with pressure overload left ventricle hypertrophy caused by transverse aortic constriction (TAC; Figure 1C). 17 We further analysed the homology of NDUFA7 by aligning the protein sequences from human, chimpanzee, dog, cattle, mouse, rat, chicken as well as zebrafish and found that NDUFA7 is highly conserved among species ( Figure 1D). These data indicate that NDUFA7 displays relatively high expression in heart, and its level decreases in cardiac hypertrophy. The fluorescent image demonstrated heart phenotype from the enlarged box region. Scale bar, 300 μm in bright field (BF) and zoom in BF, 150 μm in GFP field. V represents ventricle, and a represents atrium. The experiments were performed in triplicate, processing 40 embryos per condition. (H) WT embryos were injected with MOs at 1-cell stage, and heart rate was counted via a recorded video captured with the aid of a microscope. Statistical test: Student's t test. **P < .01 compared with controls. n = 12 measurements per condition. (I) Fractional shortening (FS) of the ventricular chamber in control and ndufa7 morphants was measured at the indicated developmental stages. Statistical test: Student's t test. *P < .05 compared with controls. n = 5 measurements per condition

| Knockdown of ndufa7 results in cardiac defect in developing zebrafish embryos
Given the high homology of ndufa7 among species, we studied the function of ndufa7 using zebrafish. To characterize the spatiotemporal expression pattern of ndufa7, we first performed wholemount in situ hybridization with different stages of WT zebrafish embryos. ndufa7 was detected in 1-cell embryo, suggesting maternal expression of this gene (Figure 2A). We also observed specific somite expression of ndufa7 in 24 hpf embryos, and heart expression in 48 hpf embryos ( Figure 2B, C). RT-PCR analysis of RNA isolated from 2 dpf cmlc2::GFP zebrafish whole embryo, tail and heart further verified the expression of ndufa7 in the somite and heart ( Figure 2D).  Figure 2F). We then injected control or ndufa7 MO into 1-cell cmlc2::GFP transgenic zebrafish embryos and found that ndufa7 morphants displayed a slightly curved, short and roughed tail, as well as a small head at both 2 dpf and 3 dpf stages ( Figure 2G). Furthermore, a slight deformed ventricle was observed in 2 dpf embryos with an even more severe heart phenotype presented in 3 dpf embryos. To further investigate the role that ndufa7 plays in myocardial function, the heart-beating videos in live embryos were taken. We found that the heart rate of ndufa7 morphant was markedly reduced compared to that of control group ( Figure 2H). Moreover, fractional shortening (FS) measurements showed that ndufa7 depletion leads to a significant reduction in ventricular function, decreasing from around 34% to 20% at 48 hpf and from approximately 33% to 14% at 72 hpf ( Figure 2I). These data demonstrate that ndufa7 depletion contributes to myocardial dysfunction in zebrafish embryos.

| ndufa7 inhibition contributes to cardiac structural defects
To investigate whether the ndufa7 MO-induced cardiac dysfunction is due to the structure defect, we evaluated the cardiac phenotype with heart-chamber marker vmhc (ventricular myosin heavy chain).
Whole-mount ISH performed on 3 dpf zebrafish embryos showed that vmhc mRNA is specifically expressed in the skeletal muscle of the trunk as well as in cardiac ventricular muscle ( Figure 3A). Compared with the control group, ndufa7 morphants exhibited a small head and disorganized somite structure missing the paralleled V shape ( Figure 3A-a,b,a',b'). The dorsal view of the embryo hearts showed that ndufa7 morphants exhibit an enlarged ventricle and wide outflow track ( Figure 3A-a",b"). The size of the ventricle in ndufa7 morphant group was around 1.9 times that of the control group ( Figure 3B). We then injected cmlc2::GFP embryos with ndufa7 MO or control MO and extracted RNA from collected embryo hearts at 3 dpf. The qPCR analysis showed that depletion of ndufa7 significantly increases the expression level of vmhc in the heart ( Figure 3C).
To verify the above result, we further performed whole-mount ISH with another heart-chamber marker cmlc2 (cardiac myosin light chain 2). cmlc2 was shown to be predominantly expressed in the  Figure 3D). Similarly, compared with the control group, ndufa7 morphants displayed a swelling ventricle with wide outcurvature, as well as elongated outflow track. Furthermore, the expression level of cmlc2 in the heart was significantly upregulated upon ndufa7 knockdown as measure by qPCR ( Figure 3E). These data show that inhibition of ndufa7 leads to altered cardiac structure.

| The cardiac hypertrophy biomarkers nppb and nppa are upregulated by ndufa7 depletion
To examine whether natriuretic peptide signalling is implicated in ndufa7-MO-induced cardiac hypertrophy, we first performed whole-mount ISH to examine the alteration of nppb upon ndufa7 depletion. As shown in Figure 4A, nppb was mainly expressed in ventricle, with its expression level decreased during the development in the control group. In contrast, the nppb expression level in ndufa7 morphants remained high and elevated during the development. In particular, the ventricle chamber of ndufa7 morphant was enlarged compared to the control group at 3 dpf. Moreover, the qPCR analysis showed that ndufa7 morphants exhibited a significantly higher nppb mRNA expression, with an average 1-fold increase at 3 dpf and 2.8-fold increase at 4 dpf ( Figure 4B). We then performed a luciferase assay with nppb::F-Luc transgenic zebrafish line, a genetic model to study HCM signalling. In comparison with the control group, depletion of ndufa7 significantly increased the nppb promoter activity, increasing luciferase activity by around 1.5-fold ( Figure 4C).
To further confirm the role of ndufa7 in cardiac hypertrophy, we performed whole-mount ISH with another hypertrophic marker nppa.
It has been shown that nppa is expressed both in ventricle and atrium. Similar to nppb, its expression level decreased during development in the control group and the expression level of nppa in ndufa7 morphants was increased compared with control group ( Figure 4D). Interestingly, the staining pattern of nppa at 3 dpf larval zebrafish also indicated that the ventricle chamber of ndufa7 morphant is enlarged as compared with control. We also showed that knockdown of ndufa7 enhances expression level of nppa in the heart by qPCR analysis, with an average 1.4-fold increase at 3 dpf and 5.6-fold increase at 4 dpf ( Figure 4E).
These results show that ndufa7 depletion contributes to increased expression of pathological hypertrophy markers.

| Calcineurin signalling is involved in ndufa7 inhibition induced cardiac hypertrophy
To gain mechanistic insight into the role of ndufa7 in cardiac hypertrophy, we first examined the impact of ndufa7 deficiency on ROS level by staining embryos with 2',7'-dichlorofluorescin diacetate. A F I G U R E 6 Depletion of NDUFA7 leads to cardiac hypertrophy in H9c2 cells. (A) H9c2 cells were transfected with control or NDUFA7 siRNAs for 72 hours, and expression level of NDUFA7 and GAPDH were examined. (B) H9c2 cells transfected with control or NDUFA7 siRNAs were stained for ROS using 2',7'-dichlorofluorescin diacetate, and then visualized under the fluorescence microscope. Scale bar, 50 µm. (C) Experiments were performed as in (B), and the ROS intensity in each group were examined by ImageJ. *P < .05 compared with control group. (D) H9c2 cells transfected with control or NDUFA7 siRNAs for 72 h, RNA was then extracted for RT-qPCR analysis. **P < .01 compared with control group. (E) Schematic representation of the proposed mechanism. Depletion of ndufa7 triggered ROS production and subsequent calcineurin signalling activation, which further led to the expression of cardiac hypertrophy genes significant increase in ROS level especially in zebrafish heart was observed in ndufa7 morphant compared with control group (Figure 5A,   B). It has been reported that calcineurin signalling regulates cardiac hypertrophy with accompanying by the intracellular ROS production. 18 We further examined several important genes in calcineurin signalling and found that ndufa7 MO significantly reduces serca2 level and increase calcineurin level in embryonic hearts ( Figure 5C).
To further verify our hypothesis, we treated ndufa7-MO-injected embryos with the calcineurin inhibitor, FK506. In the vehicle (DMSO) treated group, ndufa7 morphant presented a curved and roughed tail and evidently defected ventricle ( Figure 5D). However, ndufa7 morphant treated with FK506 exhibited a straight tail and less defected ventricle. Besides, whole-mount ISH analysis with cmlc2 showed that FK506 treatment restored ndufa7 depletion induced swelling ventricle and wide outcurvature to an extent ( Figure 5E). Moreover, FK506 treatment partially rescued defects in heart rate and ventricle function in ndufa7 morphant ( Figure 5F, G). To further investigate whether calcineurin signalling is involved in ndufa7 inhibition induced cardiac hypertrophy, we performed a luciferase assay with nppb:Luc zebrafish embryos treated with FK506. As shown in Figure 5H, FK506 treatment restored the ndufa7 MO-induced upregulation of nppb. Moreover, we found that inhibition of calcineurin restored the expression level of both nppa and nppb in embryo hearts by qPCR analysis (Figure 5I, J). These results show that calcineurin signalling is involved in ndufa7 induced cardiac hypertrophy.

| Depletion of NDUFA7 leads to cardiac cell hypertrophy
To verify the above findings in zebrafish, we further examined the role of NDUFA7 in cardiac cells. H9c2 cells, derived from embryonic rat heart tissue, have been used as a cardiac cell model in many cardiac hypertrophy studies. 19,20 Immunoblotting revealed that NDUFA7 siRNAs decreased the protein level of NDUFA7 effectively ( Figure 6A). Moreover, NDUFA7 depletion promotes ROS generation, which is consistent with the phenotype in zebrafish heart ( Figure 6B, C). We further examined expression of cardiac hypertrophy markers by qPCR analysis and found that NDUFA7 inhibition markedly increase the mRNA level of ANP and BNP, indicating that depletion of NDUFA7 leads to cardiac hypertrophy ( Figure 6D). We proposed that depletion of ndufa7 promoted ROS production and calcineurin signalling activation, leading to the expression of cardiac hypertrophy genes, thus contributing to cardiac hypertrophy ( Figure 6E).

| D ISCUSS I ON
Due to the embryo transparency and the ease of direct embryonic manipulation, the zebrafish has emerged as a promising research tool to model cardiovascular diseases including cardiac hypertrophy, congenital heart defects, arrhythmia, as well as cardiomyopathy. For example, a zebrafish model of human cardiac troponin T (TNNT2) mutation known to cause HCM has been generated. 21 The morphant zebrafish embryos with tnnt2 knockdown displayed sarcomere disarray and a robust induction of myocardial hypertrophic pathways, which is similar to humans with HCM. 22 Moreover, zebrafish with pickwick m171 mutation demonstrated reduced cardiac contractile function and pericardial oedema. 23 The mutation was attributed to the gene ttn, encoding the protein Titin, which is an important cause of human idiopathic dilated cardiomyopathy. 24 Besides, the establishment of zebrafish models to dissect ANP/BNP signalling pathway further prompts the functional studies of cardiac hypertrophy pathogenesis, disease mechanisms and potentially drug screens. [25][26][27] NDUFA7 has been reported to be associated with rheumatoid arthritis (RA) with severe erosive arthritis. 12 Till now, the biological functions of NDUFA7 remain unknown. In this study, we employed zebrafish models extensively to explore ndufa7 function in pathological cardiac hypertrophy. Consistent with tissue expression database, zebrafish ndufa7 is expressed in the heart and muscle during embryonic development. We further provide several lines of evidence showing the important role of ndufa7 in cardiac hypertrophy: Knockdown of ndufa7 leads to cardiac defect in developing zebrafish embryos; ndufa7 depletion contributes to the elevated expression of hypertrophic biomarkers nppb and nppa; calcineurin signalling is involved in ndufa7 inhibition induced cardiac hypertrophy. These findings were further verified in the model of cardiac cells. It is possible that ndufa7 depletion activates calcineurin signalling, which allows the dephosphorylated NFAT to be imported into the nucleus, thus leading to the expression of cardiac hypertrophy genes. It will be interesting to explore in further studies, which will deepen our understanding of the biological function of ndufa7.
Given the high density of mitochondria in cardiomyocytes, it is not surprising that ndufa7 plays an important role in cardiac function. 28 Mitochondrial function is essential for proper heart function by producing a constant supply of energy to accomplish complex cellular processes including continuous repetitive contraction and maintenance of Ca 2+ homeostasis. 29 Oxidative stress, characterized by the accumulation of ROS, is known to exert detrimental influence on the myocardium such as the induction of apoptotic cell death, hypertrophy and dysfunction. Increasing evidence suggests that cardiac hypertrophy induced by mechanical left ventricular wall stress is partially triggered by ROS generation. 30,31 It has been reported that calcineurin regulates the pathogenesis of cardiac hypertrophy with accompanying by the intracellular ROS production. 10 Mitochondrial respiratory complex I has been considered to be the major ROS generation site because large changes in the potential energy of the electrons occur in the sites. 32 In a recent study, it is indicated that ROS might involve in the NDUFA7 induced rheumatoid arthritis. 12 In this study, we demonstrate that depletion of ndufa7 promotes ROS production and calcineurin signalling activation, leading to cardiac hypertrophy. Furthermore, alterations in mitochondrial metabolism have been reported in pathological cardiac hypertrophy, including dysfunction of the electron transport chain and reduced capacity of ATP synthesis. 33 It is possible that ndufa7 might be associated with cardiac hypertrophy by regulating the activity of mitochondria complex I, as well as the ATP generation of mitochondrial oxidative phosphorylation. Elucidation of this will help to fully understand the role of ndufa7 in the pathogenesis of cardiac hypertrophy.

ACK N OWLED G EM ENTS
We would like to thank Koroboshka Brand-Arzamendi (St. Michael's Hospital), Dr Kim A Connelly (St. Michael's Hospital) for technical help and Dr Peter P. Liu (University of Ottawa Heart Institute) for critical comments on the manuscript.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no competing interests.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.