MARCH5 restores endothelial cell function against ischaemic/hypoxia injury via Akt/eNOS pathway

Abstract MARCH5 is a critical regulator of mitochondrial dynamics, apoptosis and mitophagy. However, its role in cardiovascular system remains poorly understood. This study aimed to investigate the role of MARCH5 in endothelial cell (ECs) injury and the involvement of the Akt/eNOS signalling pathway in this process. Rat models of myocardial infarction (MI) and human cardiac microvascular endothelial cells (HCMECs) exposed to hypoxia (1% O2) were used in this study. MARCH5 expression was significantly reduced in ECs of MI hearts and ECs exposed to hypoxia. Hypoxia inhibited the proliferation, migration and tube formation of ECs, and these effects were aggravated by knockdown of MARCH5 but antagonized by overexpressed MARCH5. Overexpression of MARCH5 increased nitric oxide (NO) content, p‐eNOS and p‐Akt, while MARCH5 knockdown exerted the opposite effects. The protective effects mediated by MARCH5 overexpression on ECs could be inhibited by eNOS inhibitor L‐NAME and Akt inhibitor LY294002. In conclusion, these results indicated that MARCH5 acts as a protective factor in ischaemia/hypoxia‐induced ECs injury partially through Akt/eNOS pathway.

ischaemia/hypoxia conditions, is crucial to find novel preventive and therapeutic targets for CVDs.
Reactive oxygen species (ROS) are important signalling molecules, mainly generated by mitochondria within most mammalian cells. 5 ECs damage caused by increased levels of ROS is involved in the process of IHD and closely related to ubiquitination modification. 6,7 Ubiquitination is a ubiquitous post-translational modification which attaches a 76-amino-acid peptide called ubiquitin (Ub) to substrate proteins, involved in selective protein degradation, endocytosis as well as signal transduction. The process of ubiquitination is catalysed by three types of enzymes including Ub activating enzymes (E1), Ub conjugating enzymes (E2) and Ub ligases (E3). 8 And the selectivity of Ub system is mainly determined by E3 ligases. 9 Many E3 ligases such as MARCH5 (membrane-associated ring-CH 5, also named MITOL) participated in regulation of mitochondria quality control and intracellular ROS levels.
Thus, the present study aimed to determine (a) the expression change of MARCH5 after myocardial infarction, (b) the potential role of MARCH5 in the pathological process of hypoxia-induced ECs dysfunction and (c) the mechanism in the process of MARCH5regulated ECs function.

| Animal model of myocardial infarction
All animal experiments were approved in accordance with the guidelines of the Institutional Animal Care and the Ethics Committee of Sichuan University. Healthy male Sprague-Dawley (SD) rats (200-220 g) were randomly divided into two groups: (a) myocardial infarction (MI) model group, MI surgery was performed by ligating the left anterior descending (LAD) coronary artery with surgical sutures according to previous reports 20 ; (b) sham group, in which the rats underwent a similar surgery without ligation the artery. After 7 days of ischaemia, TTC staining was used to detect myocardial infarct size as previously described. 21

| Echocardiography
After 7 days of the surgery, the rats were anaesthetized with isoflurane, and cardiac function was assessed by an echocardiographic imaging system (Vinno 6 Lab). The left ventricular ejection fraction (LVEF) and fractional shortening (FS) were measured.

| Immunofluorescence and Immunohistochemistry
The immunofluorescence and immunohistochemistry procedures were performed as previously described. 22 Briefly, hearts were excised, rinsed, fixed, embedded and then sectioned. The sections (4-5 μm) were used. After blocked, the samples were incubated with primary antibodies overnight at 4°C. Images were captured by a light microscope and were determined via Image J (National Institutes of Health, USA), a quantitative digital analysis system. The expression of MARCH5 was evaluated by assigning percentage of positive cells and the staining intensity.

| Small interfering RNA (siRNA) transfection
Cells were seeded in the plates at 30%-50% confluence containing medium without antibiotics before transfection. Specific small interfering RNAs (siRNA) against MARCH5 (against human) were designed and prepared by GenePharma (China), all sequences are shown in Table 1. The siRNAs were transfected into HCMECs using Lipofectamine RNAiMAX (Life Technologies) in serum-free Opti-MEM (Gibco). Cells transfected after 48h were harvested or used for further experiments.

| Adenovirus construction and transfection
Recombinant adenoviruses containing MARCH5 gene (Ad-MARCH5) and corresponding control (Ad-Ctrl) used in this study were designed and produced by Vigene Biology. Transfection of adenoviral particles was performed according to the manufacturer's protocol.

| Real-time polymerase chain reaction (RT-PCR) analysis
The total RNA was extracted from tissues or cell lysates using TRIzol (Invitrogen). One microgram of extracted RNA was reverse transcribed into cDNA using the PrimeScript RT Reagent Kit (Takara). The RT-PCR assay was performed on the CFX96TM Realtime PCR Detection System (Bio-Rad) using EvaGreen Supermix Kit (Bio-Rad). The specific primers were synthesized by Sangon and details were shown in Table 2. All samples were run in triplicate, and data were averaged. Relative mRNA expression was normalized to β-actin.

| Western blotting analysis
Cells were lysed by RIPA buffer (Beyotime) with protease and phosphatase inhibitor cocktail. Equal amounts of protein were loaded on 10% SDS-PAGE gels and then transferred to PVDF membranes (Millipore). The membranes were incubated with corresponding primary antibodies and HRP-conjugated secondary antibodies. Bands were detected by an ECL system and blotting density was determined using Bio-Rad analysis software Image Lab.

| Cell proliferation assay
Cell proliferation was detected using the enhanced cell counting kit-8 (CCK8) assay (Beyotime) following the manufacturer's instructions. Cells were seeded in 96-well plates at a density of 2 × 10 4 per well. Transfection and treatment of cells were performed as described above. Then, CCK8 reagent was added to each well for a further 2 hours incubation. Subsequently, cell viability was evaluated by detecting the absorbance at 450 nm and quantified by calculating the absorbance percentage of the treated group vs control group.

| Cell migration assay
The migration of ECs was evaluated using a wound healing assay.
For migration, cells (5 × 10 5 ) were seeded in 6-well plates and grown to confluency. HCMECs monolayers were scratched with a sterile pipette tip. The remaining area of the scratch was measured using Image J software.

| Measurement of NO production
Intracellular NO production was detected using a total nitric oxide assay kit (Beyotime) according to the manufacturer's recommendations.

| Statistical analysis
All data were presented as mean ± SD. Statistical analysis was performed using GraphPad Prism 7.0. Comparisons between two groups were analysed by Student's t test, and multigroup comparisons were F I G U R E 1 MARCH5 expression was decreased in rat myocardial infarction (MI) model. A, Myocardial infarct size was determined by TTC staining. B-C, Cardiac function was determined by echocardiogram after 7 d left anterior descending coronary artery ligation in each group. D, Quantitative analysis of MARCH5 expression by immunohistochemistry (MARCH5-brown, Nucleus-blue). E, Co-staining of MARCH5 and CD31, a classic endothelial cell marker (CD31-red, MARCH5-green, DAPI-blue). Data were shown as mean ± SD (n ≥ 3), *P < .05, **P < .01 vs sham group analysed by one-way ANOVA followed by the Dunnett's test. A value of P < .05 was considered statistically significant.

| MARCH5 expression is decreased in heart tissues of MI rats and hypoxic endothelial cells
Firstly, TTC staining showed that rats with LAD coronary artery ligation exhibited increased infarct size ( Figure 1A). In addition, the MI rats showed a lower LVEF and decreased FS compared with the sham group ( Figure 1B,C). These results suggested the MI model was established successfully. Then, RT-PCR and Western blot were used to quantify MARCH5 alterations in the context of ischaemia injury. Compared with the sham group, MARCH5 was significantly decreased in the infarct cardiac tissues both at transcription and protein expression levels ( Figure S1A,B). Immunohistochemistry staining reconfirmed that MARCH5 expression was down-regulated in response to the ischaemia injury ( Figure 1D). To observe MARCH5 expression changes in cardiac ECs, CD31 and MARCH5 co-staining was performed. As shown in Figure 1E, ischaemia injury significantly reduced MARCH5 expression in ECs.
To further investigate the role of MARCH5 in ischaemia injury, HCMECs were incubated with 1% O 2 for 0, 6, 16 and 24 hours. From CCK8 assay, we observed that hypoxia reduced ECs proliferation  Figure 2E,F, the expression of MARCH5 decreased gradually with the prolongation of hypoxia. These results provided us with a potential link between MARCH5 reduction and hypoxia injury.

| Knockdown of MARCH5 deteriorates endothelial function
To examine the effects of MARCH5 on ECs function in response to hypoxia, ECs were transfected with specific siRNAs of MARCH5 (siMA-RCH5) for 24 hours before hypoxia treatment. After transfection, MARCH5 expression was down-regulated in siMARCH5 group compared with control (siCtrl) group, indicating that the transfection was successful ( Figure 3A,B). The siMARCH5-2 with the best silencing efficacy was selected for further studies. The ECs proliferation was inhibited in siMARCH5 group under both normoxia and hypoxia conditions compared with siCtrl group ( Figure 3C). Then, ECs migration analysis using wound healing assay was performed, results revealed that the cell migration was reduced in siMARCH5 group ( Figure 3D). Besides, we employed the EC tube formation assay on Matrigel matrix and found that MARCH5-deficient ECs exhibited markedly blunted angiogenesis ( Figure 3E). These results indicated that knockdown of MARCH5 impaired endothelial proliferation, migration and angiogenesis.

| Overexpression of MARCH5 alleviates endothelial dysfunction
Because loss of MARCH5 mediated endothelial dysfunction, we

| MARCH5 affects ECs function through eNOSdependent mechanism
Endothelial nitric oxide synthase (eNOS, also named nitric oxide synthase-3 [NOS3]), one of the critical enzymes regulating nitric oxide (NO) production, plays a key role in ECs function. 23,24 Therefore, eNOS expression levels during hypoxia were measured next step. RT-PCR showed eNOS mRNA was decreased during hypoxia ( Figure 5A). The decrease in the expression of total eNOS and p-eNOS at Ser1177 was also confirmed by western blot analysis ( Figure 5B).
To investigate whether MARCH5 affects ECs function through eNOS signalling pathway, firstly, we explored the relationship between NO content and MARCH5 expression. Results showed that NO content in ECs was significantly decreased after silencing MARCH5 and significantly increased after transfection with Ad-MARCH5 ( Figure 5C).
Secondly, eNOS expression levels after specific silencing or overexpression of MARCH5 were detected. Knockdown of MARCH5 attenuated the levels of eNOS, while overexpression MARCH5 increased the expression of eNOS in both normoxia and hypoxia conditions ( Figure 5D,E). These data suggested that MARCH5 may affect ECs function through NO synthesis regulated by eNOS.
Furthermore, ECs were pre-treated with NG-nitrol-arginine methyl ester (L-NAME), an inhibitor of nitric oxide synthase (NOS), followed by hypoxia stimulation. Incubation with L-NAME abolished

MARCH5-induced enhancement of migration and tube formation of
ECs ( Figure 5F-I). These data indicated that MARCH5 regulates ECs function through the eNOS-dependent mechanism. To further test our hypothesis, we treated ECs with an Akt phosphorylation inhibitor (LY294002) and an Akt phosphorylation agonist (SC79). As shown in Figure 6C-F, administration with LY294002  Figure 6G,H).

| D ISCUSS I ON
In the present study, the role of MARCH5 on ECs under ischaemia/ hypoxia injury was investigated. We found the followings: (a) the stress. [10][11][12]25 Studies showed that MARCH5 exerted potential regulator effects in hypoxia stimulation 15  Akt at both Ser473 and Thr308 residues ( Figure 6A,B, Figure S1D,E).

| CON CLUS IONS
This study firstly proved that MARCH5 played a protective role in maintaining ECs function under hypoxic stress by regulating Akt/ eNOS signalling, which might be a potential therapeutic target for the clinical management of ischemic heart diseases.

ACK N OWLED G EM ENTS
The work was supported by the National Natural Science Foundation

CO N FLI C T O F I NTE R E S T
The authors declare that there are no conflicts of interest.

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 from the correspondence author on reasonable request.
F I G U R E 6 MARCH5 protects ECs against hypoxia injury via Akt/eNOS pathway. A, Knockdown of MARCH5 decreased eNOS, p-eNOS and p-Akt. B, Overexpression of MARCH5 up-regulated eNOS, p-eNOS and p-Akt. Ad-Ctrl and Ad-MARCH5 ECs under hypoxia were incubated with DMSO or the Akt inhibitor LY294002 (10 μg/mL). C, Wound healing assay was performed in each group. D, Tube formation. E, F, Quantitative analysis of C and D. Akt phosphorylation inhibition with LY294002 aborted cellular proliferation and tube formation of Ad-MARCH5 ECs vs Ad-Ctrl group under hypoxia stress. G, Protein expression of eNOS and p-eNOS in Ad-Ctrl and Ad-MARCH5 ECs incubated with DMSO or LY294002 under hypoxia. E, Protein expression of eNOS and p-eNOS in siCtrl and siMARCH5 ECs incubated with DMSO or Akt activator SC79 (8 μmol/L) under hypoxia. Data were shown as mean ± SD (n ≥ 3), *P < .05, **P < .01