NDUFA4L2 in smooth muscle promotes vascular remodeling in hypoxic pulmonary arterial hypertension

Pulmonary arterial hypertension (PAH) is characterized by a progressive increase in pulmonary vascular resistance and obliterative pulmonary vascular remodelling (PVR). The imbalance between the proliferation and apoptosis of pulmonary artery smooth muscle cells (PASMCs) is an important cause of PVR leading to PAH. Mitochondria play a key role in the production of hypoxia‐induced pulmonary hypertension (HPH). However, there are still many issues worth studying in depth. In this study, we demonstrated that NADH dehydrogenase (ubiquinone) 1 alpha subcomplex 4 like 2 (NDUFA4L2) was a proliferation factor and increased in vivo and in vitro through various molecular biology experiments. HIF‐1α was an upstream target of NDUFA4L2. The plasma levels of 4‐hydroxynonene (4‐HNE) were increased both in PAH patients and hypoxic PAH model rats. Knockdown of NDUFA4L2 decreased the levels of malondialdehyde (MDA) and 4‐HNE in human PASMCs in hypoxia. Elevated MDA and 4‐HNE levels might be associated with excessive ROS generation and increased expression of 5‐lipoxygenase (5‐LO) in hypoxia, but this effect was blocked by siNDUFA4L2. Further research found that p38‐5‐LO was a downstream signalling pathway of PASMCs proliferation induced by NDUFA4L2. Up‐regulated NDUFA4L2 plays a critical role in the development of HPH, which mediates ROS production and proliferation of PASMCs, suggesting NDUFA4L2 as a potential new therapeutic target for PAH.


| INTRODUC TI ON
Pulmonary artery hypertension (PAH) is a type of cardiovascular disease with a high degree of malignancy. The diagnostic criteria is based on the 2018 6th World Pulmonary Hypertension Conference (WSPH) redefining the definition of pulmonary hypertension (PH) as mPAP >20 mmHg measured by the right heart catheter at rest.
In addition, the definition of all precapillary PH also includes pulmonary vascular resistance ≥3 Wood units. 1 The worsening symptoms and poor prognosis caused by the development of PAH are associated with elevated pulmonary arterial pressure and right ventricular hypertrophy (RVH) eventually leading to right heart failure. 2 Longterm oxygen therapy can effectively prolong survival, but can only slightly reduce pulmonary arterial pressure. So far, no targeted drugs including pulmonary vasodilators have shown effective therapeutic effects in the treatment of PAH. 3 Although the natural survival time of patients after diagnosis has been increased from the original 2.8-6 years, it still causes a huge clinical and economic burden.
Although the number of PAH-related hospital admissions decreased between 2001 and 2012, the average cost and length of hospital admissions per PAH-related admissions increased, and hospital mortality did not decrease significantly. 3 Therefore, it is important to understand the pathogenesis of PAH and find early diagnosis, prognostic judgment and new molecular markers.
The aetiology of PAH involves environmental and genetic factors, and its pathogenesis is complex, so far there is no complete explanation. Hypoxia is an important cause. Long-term exposure to hypoxic conditions can cause inflammation, vasoconstriction, smooth muscle cell proliferation, muscleization of the anterior capillaries and loss of distal pulmonary vessels, which are key pathophysiological processes of PAH.
After the pulmonary artery is subjected to various injuries or hypoxia, the vascular wall tissue structure and its function undergo pathological changes, mainly including the two pathological processes of pulmonary angiogenesis and pulmonary artery middle smooth muscle thickening. Although thickening of all layers of the pulmonary vessel wall intima, media and adventitia can cause PAH, vascular remodelling caused by medial thickening plays a major role in its development. In the pulmonary blood vessels, pulmonary artery smooth muscle cells (PASMCs) are the main cells constituting the pulmonary artery wall. Their hypertrophy is an important pathological feature of hypoxic pulmonary vascular structural changes.
When PAH occurs, PASMCs proliferation increases, apoptosis decreases, and DNA synthesis increases. These abnormal changes promote thickening of the pulmonary artery wall and narrowing of the lumen, which in turn leads to increased pulmonary vascular resistance, RVH and continuous increase in pulmonary artery pressure. 4 Therefore, further searching for new targets for anti-PASMCs proliferation and hypertrophy, and exploring related intracellular molecular mechanisms have important scientific value and clinical application prospects.
Numerous studies have showed that oxidative stress plays a crucial role in the development of PAH. 5,6 Whether in the animal model of PAH induced by SU5416-hypoxia 7 or in the rat PAH model induced by hypoxia 8 and PAH patient specimens, 9 reactive oxygen species (ROS) and other oxidative stress indicators have increased significantly. In particular, the release of ROS from mitochondria caused by hypoxia attacks and promotes the oxidation of polyunsaturated fatty acids, a process called lipid peroxidation. 10,11 Lipid peroxidation produces a variety of oxidation products, with aldehydes being the major end products. Toxic aldehydes are highly reactive and interact with cellular macromolecules (including nucleic acids and proteins) to produce a variety of adducts, resulting in DNA damage and protein inactivation. Among the products of lipid peroxidation, 4-hydroxynonene (4-HNE) is considered the most toxic aldehyde and malondialdehyde (MDA) appears to be the most mutagenic aldehyde. 12 Lipid oxidation has been reported to participate in the proliferation of PASMCs and exacerbate the development of PAH. 12 However, it remains uncertain whether the production of toxic aldehydes contributes to hypoxic pulmonary vascular remodelling (PVR).
NADH dehydrogenase (ubiquinone) 1 alpha subcomplex 4 like 2 (NDUFA4L2), as part of the electron transport chain (ETC) complex I (complex I) subunit, belongs to the NDUFA4 subunit family of the complex I subunit, can fine-tune the activity of complex I, thereby mediating mitochondria to activate oxidative phosphorylation and produce ROS. 13 NDUFA4L2, which is one of the important components of the ETC complex I subunit in mitochondria, is an important site for ROS generation and is widely involved in the regulation of biological processes. For example, NDUFA4L2 is abnormally expressed in many types of cancer, including malignant hepatocellular carcinoma, 14 clear cell renal cell carcinoma 15,16 and colorectal cancer. 17 NDUFA4L2 inactivation increases mitochondrial activity and oxygen consumption, leading to accumulation of ROS and apoptosis in hepatocellular carcinoma. 14 NDUFA4L2 has been reported to promote hypoxic cell proliferation by increasing mitochondrial ROS and nucleic acid production. 18 However, the molecular mechanism of hypoxic-induced PASMCs proliferation is unknown, and the role of NDUFA4L2 in vascular remodelling of PAH has not been reported.
In the present study, we found that the production of 4-HNE was increased in the plasma of PAH patients and chronic hypoxia-induced PAH rat animal models. Using multiple approaches, we have demonstrated that inhibition of NDUFA4L2 reduced 4-HNE levels and attenuated the severity of hypoxic PAH. NDUFA4L promoted the proliferation of PASMCs by regulating the upstream HIF1α pathway and the downstream p38-5-lipoxygenase (5-LO) signal, thereby promoting PVR and inducing PAH.

| Animal models
Adult Wistar male rats with an average weight of 200 g were obtained from the Experimental Animal Center of Xuzhou Medical University (Grade II). Rats were either placed in normal air or in a normobaric hypoxic chamber (FiO 2 10%) for 3 weeks. All animal procedures were in accordance with the National Institutes of Health Guidelines and were approved by the Animal Use and Care Committee of the First People's Hospital of Lianyungang. To understand the role of NDUFA4L2 in hypoxia-induced PAH. Adult Wistar male rats were randomly divided into four groups (n = 6). Chemically modified anti-small interfering RNA (siRNA) oligonucleotides targeting NDUFA4L2 (siNDUFA4L2) or nontargeting control (siNC) were administered intraperitoneally for two times a week for 3 weeks in the chronic hypoxia-induced PAH rats. The primer sequences were designed and synthesized by Ribobio co., LTD based on the mRNA sequences obtained from the NCBI database as follows: si-r-NDUFA4L2:GTTTCCACCGACTACAAGA. Wistar male rats treated with siNC were fed in normal air or in a normobaric hypoxic chamber (FiO 2 10%) for 3 weeks. Wistar male rats treated with siNDUFA4L2 were fed in normal air or normal pressure hypoxic chamber (FiO 2 10%) for 3 weeks.

| Haemodynamic experiments
Right ventricular systolic pressure (RVSP) was analysed as previously described. 19 Rats were anaesthetized with 3% pentobarbital sodium 30 mg/kg by intraperitoneal injection. After anaesthesia, the right jugular vein was peeled off, and a heparin anticoagulated PV-I polyethylene catheter (curved tip) was inserted into the right ventricle (RV) through the right jugular vein, followed by blood into the pulmonary arteries (PAs). Pulmonary arterial pressure was analysed by a 1.4 F pressure transducer catheter (Millar Instruments) and AcqKnowledge software (Biopac Systems Inc). 20 After analysing the hemodynamic data, the rat thorax was opened. The lungs and heart were moved together into a petri dish filled with cold PBS. The ratio of RV/left ventricle + septum (RV/LV + S) was analysed as an indicator of RVH. The lungs were then immersed in 4% paraformaldehyde overnight and wrapped with wax blocks for HE staining and immunohistochemistry. The remaining lungs were frozen in a refrigerator at −80°C for subsequent experiments.

| Histology and immunohistochemistry
Lung tissue was fixed with 4% formaldehyde overnight, then dehydrated, cleared and embedded in paraffin. The wax blocks were cut into 4 μm thick sections for H&E and immunohistochemistry staining. For immunohistochemistry, 4-µm paraffin-embedded tissue sections were dewaxed and hydrated. The immunohistochemistry method was based on the technique described previously. 21 Antibody was incubated with NDUFA4L2 (Catalogue number: 16480-1-AP; ProteinTech). Brown indicated positive staining.

| Cell culture
Human PASMCs (HPASMCs) and pulmonary artery endothelial cells (HPAECs) were used in vitro experimental studies. Primary HPASMCs were derived from PAs of organ donor lungs. The PAs were cut into 1-2 mm with ophthalmic scissors, and then digested with a mixture containing 2 mg/mL collagenase, 0.5 mg/mL elastase and 1.5 mg/ mL BSA for 1-2 hours until the tissue mass became a cell mass. Then, HPASMCs were cultured in SmGM-2 BulletKit medium (Lonza) containing 10% FBS and placed in a 5% CO 2 incubator at 37°C for 1 week. 22 The purity of HPASMC was identified by a specific monoclonal antibody against smooth muscle α-actin.

| Microarray analysis of mRNA expression
RNAs were extracted and quantified from HPASMCs according to our previously published protocol. 19 The microarray analysis work was carried out by OE Biotech. Co., Ltd.

| Transfection
The sequence of siRNA against NDUFA4L2 and negative control (NC) were synthesized from GenePharma (Shanghai GenePharma Co). The target sequences for using in transfection were follows: si-h-

| Cell cycle analysis
We used PI single staining method to detect cell cycle. Before staining, cells were washed the fixative solution with PBS (if necessary, filter the cell suspension once with a 200 mesh sieve). Hundred microliter RNase A was added to the cells at 37°C for 30 minutes. Finally, cells were added with 400 μL PI dark room to protect from light for 10 minutes. DNA fluorescence measurements were analysed using BD flow cytometry.

| Cell Counting Kit-8 (CCK-8) assay
The HPASMCs were cultured in a 96-well plate, and then cells were starved for 24 hours after growing to about 60%. The cells were transfected with Nor-siNC, Hyp-siNC, Hyp-si-NDUFA4L2 or Hyp-si-5-LO. After incubation for 24 hours at 37°C, cell proliferation was detected with CCK8 kit (Dojin Laboratories, catalog no: CK04) according to the manufacturer's instruction. CCK-8 solution was added to each well of the plate for 10 μL. The plate was incubated for 1-4 hours in the incubator. The absorbance was measured at 450 nm using a microplate reader.

| 5-ethynyl-20-deoxyuridine (EDU) staining
EdU staining was analysed using a kFluor555 Click-It EDU Kit according to the manufacturer's protocol (KeyGen Biotech). HPASMCs were seeded into 96-well plates at a density of 5 × 10 3 cells/well followed by incubation for 24 hours in serum-free DMEM. After the cells were transfected with Nor-siNC, Hyp-siNC, Hyp-si-NDUFA4L2 or Hyp-si-5-LO, HPASMCs were treated with 10 μM EdU for 4 hours at 37°C.
Then cells were fixed with 4% paraformaldehyde for 30 minutes at room temperature and treated with 0.5% Triton X-100 in PBS for 20 minutes to permeabilize cells according to the manufacturer's instruction.

| Real-time PCR
Total RNA was extracted from HPASMCs and HPAECs using TRIzol according to the manufacturer's specifications. The yield of RNA was determined using Synergy H1 Microplate reader (BioTek), and the integrity was evaluated using agarose gel electrophoresis stained with ethidium bromide. Quantification was performed with a two-step reaction process: reverse transcription (RT) and real-time PCR according to the

| Western blotting
The PAs homogenate tissue and collected HPASMCs and HPAECs

| Determination of mitochondrial ROS production
Cells were pre-treated with different reagents for 24 hours. The culture medium was then removed, and ROS detection reagents DCFH-DA (10 μM) or Rosup (5 mg/mL) (Beyotime Biotechnology) or MitoSOX Red (5 mM) (Yeasen Biotech Co., Ltd) were applied according the manufacturer's instructions. An Olympus IX73 fluorescence microscopy was used for acquisition of fluorescent images.

| Oxygen consumption
The HPASMCs were seeded in a 96-well culture plate and treated with hypoxia and hypoxia + siNDUFA4L2. Oxygen consumption was analysed by extracellular oxygen consumption assay kit according to the manufacturer's instructions (Abcam, ab197243).

| Complex I activity
The HPASMCs were seeded in a 96-well plate at a density 20 000 cells per well before the initiation of the experiment. The activity of Complex I was analysed using the Complex I Enzyme Activity Assay Kit according to the manufacturer's instructions (Abcam, ab109721).

| Statistical analysis
The data are expressed as mean ± SEM of at least three independent experiments. One-way ANOVA or Student's t test was used to determine the significance of differences between the means of different groups, followed by a Bonferroni test using the Prism software package (version GraphPad Prism 5.0). A P value <.05 was considered statistically significant.

| mRNA expression profile in hypoxic human pulmonary artery smooth muscle cells
In order to find related factors that promote cell proliferation and trigger vascular remodelling in PASMCs, we compared mRNA expression profiles in normal and hypoxic HPASMCs with mRNA chips in a recognized model of hypoxic proliferation of PASMCs.
We found that 17 mRNAs were significantly up-regulated and Results found that NDUFA4L2 was a candidate gene for us because of its large multiples (≥8 times) (Table 1), and its results verified in subsequent in vitro and in vivo models and the chip results are reproducible ( Figure 1A,B).

| Expression of NDUFA4L2 in the PAs of PAH patients and hypoxic PAH model rats
Our Immunohistochemical experiments confirmed that NDUFA4L2 was expressed in both the medial and intimal membranes, and the median membrane was more pronounced (Figure 2C,D). Because NDUFA4L2 was a direct target of HIF-1α, we also detected the expression of HIF-1α. We found that HIF-1α expression also increased in the PAs of PAH patients and hypoxic PAH model rats ( Figure 2E,F).

| Expression of NDUFA4L2 in HPASMCs and HPAECs
Pulmonary vascular remodelling is a significant feature of PAH for- The hypoxia-induced NDUFA4L2 expression was markedly abolished when HIF1a was knocked down ( Figure 3E,F).

| NDUFA4L2 silencing ameliorates hypoxiainduced PAH in rat model
To understand the role of NDUFA4L2 in hypoxia-induced PAH, Wistar rats were randomly divided into four groups (n = 6): NC control , NC hypoxia, siNDUFA4L2 control, siNDUFA4L2 hypoxia . Real-time PCR experiment was used to examine the interference efficiency of siNDUFA4L2 on rat lung ( Figure 4A). To examine that NDUFA4L2 was involved in the remodelling of pulmonary arterial wall, we first compared the thickness of PA between NC control and NC hypoxia group.
We found significant increase of pulmonary arterial wall thickness in NC hypoxia group compared with NC control group, but NDUFA4L2 silencing effectively inhibited the thickening of PAs under hypoxia ( Figure 4B,C). In addition, hypoxia significantly increased RVSP in NC treatment, knockdown of NDUFA4L2 led to significant decreases in RVSP in siNDUFA4L2 hypoxia group compared with NC hypoxia group ( Figure 4D). Then, we observed obvious hypertrophy of RV in NC hypoxia group, but NDUFA4L2 silencing partially reversed RV hypertrophy. Quantitative analysis showed that RV/(LV + S) in NC hypoxia significantly increased compared with NC control , but NDUFA4L2 silencing significantly decreased RV/(LV + S) in hypoxia condition ( Figure 4E).

| The activation of NDUFA4L2 promoted proliferation of hypoxic HPASMCs
To examine the effect of NDUFA4L2 on the proliferation of hypoxic HPASMCs, we knocked out the gene NDUFA4L2 and treated the HPASMCs with hypoxia. We designed siRNA to inhibit NDUFA4L2 expression. The efficiency of siRNA transfection was tested and confirmed by Western blotting ( Figure 5A). CCK-8 assay indicated that the proliferation of PASMCs was inhibited in the hypoxia + si- silencing ( Figure 5G,H). These data indicated that NDUFA4L2 knockdown decreased hypoxia-induced proliferation in HPASMCs.

| Hypoxia increases the production of 4-HNE, MDA and NDUFA4L2 is involved in the hypoxiainduced increase in ROS production and decrease in complex I activity and oxygen consumption during hypoxia
Hypoxia causes mitochondria to produce excessive ROS to attack and promote oxidation of lipid, especially polyunsaturated fatty

| p38-5-LO may be the downstream pathway of NDUFA4L2 inducing PAH
Next, we want to know how NDUFA4L2 works? Our results indi-   Figure 8F). Then, we directly detected the F I G U R E 6 Lipid peroxidation was increased in pulmonary arterial hypertension and NDUFA4L2 silencing-mediated inhibition of hypoxiainduced oxidative stress and mitochondrial dysfunction. A, Plasma 4-HNE levels were determined by ELISA (n = 6); (B) After knockdown of NDUFA4L2 in HPASMCs in vitro, ELISA was detected the expression of lipid peroxidation metabolites 4-HNE and MDA in different groups of Nor-siNC, Hyp-siNC and Hyp-siNDUFA4L2 (n = 6); (C) Immunofluorescence was analysed the ROS release after interference with siNDUFA4L2 in HPASMCs (n = 6); (D) Representative images of six independent experiments showing MitoSOX intensity as a measure of mitochondrial superoxide levels (n = 6); ROS production was increased in hypoxia-treated NC, which was reversed by silencing of NDUFA4L2; (E) Complex I activity was measured in HPASMCs transfected with siNC or NDUFA4L2 siRNA, then exposed to normoxic or hypoxic (3% O 2 ) conditions for 24 h (n = 6); (F) Oxygen consumption ratio in siNC or NDUFA4L2 siRNA, then exposed to normoxic or hypoxic (3% O 2 ) conditions for 24 h (n = 3). NC, negative control; Nor, normoxia; Hyp, hypoxia. *P < .05 vs donor or normoxia group; * P < .05,    and in PAEC exposed to ET-1 33 or ADMA. 34  MAPK. 26 This report suggests that p38 MAPK may be an important downstream molecule for NDUFA4L2 to mediate cell proliferation. p38 MAPK activation could increase 5-LO activity by promoting phosphorylation of 5-LO in cells. 27 The increase activity of 5-LO tends to aggravate PAH. 28  In conclusion, the present study demonstrated that NDUFA4L2 played a key role in the development of HPH by activating HIF-1α, inducing phosphorylation of p38, promoting lipid oxidation, which in turn activates 5-LO, and subsequently enhancing PASMC proliferation ( Figure 9). Furthermore, HIF-1α-NDUFA4L2-p38/5-LO could be a potential pharmacological target for the therapy of HPH and other lipid peroxidation related diseases.

CO N FLI C T O F I NTE R E S T
The authors declare no conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data used to support the findings of this study are available from the corresponding author upon request.