HIF‐1α promotes the proliferation and migration of pulmonary arterial smooth muscle cells via activation of Cx43

Abstract The proliferation of pulmonary artery smooth muscle cells (PASMCs) is an important cause of pulmonary vascular remodelling in hypoxia‐induced pulmonary hypertension (HPH). However, its underlying mechanism has not been well elucidated. Connexin 43 (Cx43) plays crucial roles in vascular smooth muscle cell proliferation in various cardiovascular diseases. Here, the male Sprague‐Dawley (SD) rats were exposed to hypoxia (10% O2) for 21 days to induce rat HPH model. PASMCs were treated with CoCl2 (200 µM) for 24 h to establish the HPH cell model. It was found that hypoxia up‐regulated the expression of Cx43 and phosphorylation of Cx43 at Ser 368 in rat pulmonary arteries and PASMCs, and stimulated the proliferation and migration of PASMCs. HIF‐1α inhibitor echinomycin attenuated the CoCl2‐induced Cx43 expression and phosphorylation of Cx43 at Ser 368 in PASMCs. The interaction between HIF‐1α and Cx43 promotor was also identified using chromatin immunoprecipitation assay. Moreover, Cx43 specific blocker (37,43Gap27) or knockdown of Cx43 efficiently alleviated the proliferation and migration of PASMCs under chemically induced hypoxia. Therefore, the results above suggest that HIF‐1α, as an upstream regulator, promotes the expression of Cx43, and the HIF‐1α/Cx43 axis regulates the proliferation and migration of PASMCs in HPH.

HIF-1α and Cx43 promotor was also identified using chromatin immunoprecipitation assay. Moreover, Cx43 specific blocker ( 37,43 Gap27) or knockdown of Cx43 efficiently alleviated the proliferation and migration of PASMCs under chemically induced hypoxia. Therefore, the results above suggest that HIF-1α, as an upstream regulator, promotes the expression of Cx43, and the HIF-1α/Cx43 axis regulates the proliferation and migration of PASMCs in HPH.

K E Y W O R D S
connexin 43, hypoxia, hypoxia-induced pulmonary hypertension, migration, proliferation, pulmonary artery smooth muscle cells

| INTRODUC TI ON
Chronic hypoxia-induced pulmonary hypertension (HPH), an incurable disease, is often a complication in patients with chronic heart failure, chronic obstructive pulmonary disease and sleep apnoea. 1 Pulmonary vascular remodelling plays an important role in HPH pathology, which is mainly characterized by thickening of the pulmonary artery wall and stenosis of the lumen. 2 Proliferation and migration of pulmonary artery smooth muscle cells (PASMCs) is the main cause of pulmonary vascular remodelling. 3 The inhibition of PASMCs proliferation and migration is expected to be the main pathway for PH treatment.
Connexins (Cxs) are transmembrane proteins that oligomerize to form a pore in the cell membrane known as a hemi-channel and interact with small regulatory molecules such as 3',5'-cyclic adenosine monophosphate (cAMP), adenosine 5' triphosphate (ATP), calcium (Ca 2+ ) and inositol 1,4,5-triphosphate (IP3), which can pass directly through the cell membrane. In the vascular system, the core Cxs are Cx37, Cx40 and Cx43, and deletion of these genes causes severe cardiovascular abnormalities. 4 Cx43 has been shown to engulf vascular smooth muscle cells (VSMCs), resulting in proliferation phenotypes and the development of atherosclerosis. 5 Recent studies have demonstrated that intimal up-regulation of Cx43 induces VSMCs proliferation in atherosclerotic plaques through gap junction generation. 6 In cardiovascular and cerebrovascular diseases, Cx43 up-regulation and smooth muscle coupling has been observed, thereby inducing proliferation and migration through up-regulation of intercellular communication. 7 These findings collectively suggest that alterations in the regulation of Cx43 expression could underlie VSMCs dysfunction.
The role of Cx43 in pulmonary hypertension (PH) has attracted increasing attention. 8 A previous study investigated the role of endothelial Cx43 in pulmonary vascular reactivity in PH, focussing on pulmonary endothelial function and the interaction of pulmonary artery endothelial cells (PAECs) and PASMCs. 9,10 Unfortunately, the biological role of Cx43 in VSMCs function of PH remains largely unknown. In the present study, we established the hypoxia model by culturing PASMCs with cobalt chloride (CoCl 2 ) or by exposing SD rats to 10% O 2 and then observed changes in proliferation and migration under hypoxia. In addition, the effects of Cx43 on CoCl 2 -treatment PASMCs dysfunction, as well as the underlying mechanism, were also explored.

| Rat model of HPH
After one week under normoxia, SD rats were randomly and equally divided into two groups which named hypoxia group and control group (n = 8 in both groups). Rats were exposed to continuity hypoxia (FIO 2 of 0.10) in an isobaric hypoxic chamber for up to 21 days in hypoxia group while maintained in a normal oxygen condition (FIO 2 of 0.21) for the same time in control group, according to the same methods used in our previous studies. 11 After established HPH rat model, the right ventricle (RV) was separated from left ventricle and septum (LV + S) and weighed. The extent of RVH was calculated as the ratio of RV to (LV + S). In all rats, the freshly isolated pulmonary arterial samples were frozen in liquid nitrogen for mRNA and protein expression analysis. The right lower lung which used in haematoxylin-eosin (HE) staining and in situ hybridization was fixed in 4% paraformaldehyde.

| Histological and Immunohistochemistry analysis
HE staining of right lung was detected in accordance to the same method used in our previous study. 11 In brief, the fixed lungs were sliced mid-sagittally and embedded in paraffin and then cut into approximately 5 μm thick sections by microtome. Then, sections were placed on glass slides and stained by haematoxylin and eosin for morphometric analysis and visualization under an Olympus BX41 microscope (Tokyo, Japan).

| Cell experiments
Normal rat PASMCs were acquired from the American Type Culture

| Cell proliferation assay
Cell proliferation was detected using the Titer 96R AQueous One Solution cell proliferation assay kit (Promega, USA). All protocols were performed according to the manufacturers' instructions as described previously. 11,12 Three repeat holes were implemented in each group, and all results from three independent experiments.

| Wound-healing assay
The wound-healing assay was carried out as described previously. 8,13 PASMCs were plated at a density of 8×10 4 cells/well in 6-well plates.
After the cells reached 85% confluence, wounds were induced by scratching with 200µl pipette tips, followed by washing with the medium to remove cellular debris. Cells were then incubated in normoxia or hypoxia conditions. Wound closure was monitored by comparing digital photographs of the same region of interest taken at 0-and 24-h time points. Pictures were analysed using the Image Pro Plus program. Cell migration was expressed as the percentage of wound area re-covered after 24 h.

| Transwell migration assay
PASMCs ( Table 2. β-actin was used as an endogenous control. The primers used for qRT-PCR are presented in Table 2.

| Cx43 and phosphorylation of Cx43 at Ser 368 (p-Cx43) were up-regulated in HPH rat model
As same as our previous studies, HPH rat model was successfully obtained after 3 weeks exposure to hypoxia presented with hypertrophy of right ventricle ( Figure 1A) and small intrapulmonary arteries remodelling ( Figure 1B,C). In our established HPH models, we found the expression of Cx43 and p-Cx43 in pulmonary artery was significantly up-regulated than in a normoxic rats ( Figure 1D,E).

| Hypoxia up-regulated Cx43 and phosphorylation of Cx43 at Ser 368 expression in PASMCs
Hypoxia has been reported to regulate Cx43 expression and channel activity in different cell models. 14 Figure 2D) and 12 h ( Figure 2E) respectively. The level of phosphorylation of Cx43 at Ser 368 was also up-regulated at 12 h ( Figure 2F). Furthermore, hypoxia also induced PASMCs proliferation ( Figure 2G) and migration ( Figure 2H-K) in a time-dependent manner. These data indicate that Cx43 may play a role in PASMCs proliferation and migration.

| Hypoxia induced PASMC proliferation and migration through Cx43
To determine the role of Cx43 in PASMCs proliferation and migration, we manipulated Cx43 by knocking down its expression with Cx43-targeting shRNA or 37,43 Gap27, which is a specific Cx-mimetic peptide blocker of Cx37 and Cx43. As shown in Figure 3A, exposure to hypoxia for 24 h remarkably induced the proliferation of PASMCs, which was inhibited by 37,43 Gap27. Wound-healing and Transwell assays showed that 37,43 Gap27 prevented PASMCs migration by Then, we used the Cx43 shRNA to specifically block the Cx43 gene. Figure 4A shows that transfection was successful. We transfected different fragments into PASMCs and found that fragment 1 of shRNAs significantly decreased the protein expression of Cx43 ( Figure 4B). Therefore, we chose to use fragment 1 of shRNAs for subsequent experiments. The genetic block of Cx43 by specific shRNA produced a comparable effect to 37,43 Gap27. As shown, exposure to hypoxia significantly induced PASMCs proliferation ( Figure 4C) and migration ( Figure 4D-G), which were reversed by transfection of Cx43 shRNA. Taken together, these results suggest that Cx43 upregulation contributes to PASMCs proliferation and migration.

| Hypoxia increased Cx43 promoter activity through the HIF-1α transcription factor
To pinpoint the molecular mechanism by which hypoxia up-regulates the expression of Cx43, we next focussed on HIF-1α, which is a hallmark of HPH and promotes the progress of vascular remodelling through regulating PASMCs proliferation and migration. 16 We first tested whether hypoxia regulates HIF-1α expression in PASMCs.
As shown, hypoxia significantly increased HIF-1α mRNA ( Figure 5A) and protein expression ( Figure 5B), and these changes were correlated with the induction of Cx43. Since HIF-1α can bind to the Cx43 promoter region, we performed ChIP analysis to show that hypoxic conditions induced more binding of HIF-1α protein to the Cx43 promoter compared with control conditions (Figure 5C,D). Therefore, our data indicate that HIF-1α is involved in the induction of Cx43 by hypoxia. 37,43 Gap27 prevented CoCl 2 -treatment PASMCs proliferation and migration. Proliferation of PASMCs was measured by MTS assay (A). Migration of PASMCs was measured by Wound-healing assay (B and C) and Transwell migration assay (D and E). All results from three independent experiments. Scale bar, 100 μm. Data are means ± SEM, ** p < 0.01 vs. Control, ## p < 0.01 vs. CoCl 2

| Hypoxia increased the expression of Cx43 in PASMCs via HIF-1α-dependent transcriptional activation
Previous studies have shown that elevated pressure contributed to the increase of Cx43 protein expression in cultured cells from aorta and Cx43 expression in the lung was increased in Sprague-Dawley rats exposed to chronic hypoxia (CH) or treated with monocrotaline In the current study, we found that hypoxia up-regulated Cx43 gene, the level of Cx43 protein expression and phosphorylation of Cx43 at Ser368, followed by proliferation and migration in PASMCs. In contrast, in cardiomyocytes, hypoxia decreased Cx43 and then induced cell death. 29 In many cancer cells, hypoxia can downregulate the originally highly expressed Cx43, thereby inhibiting proliferation and migration and promoting apoptosis and autophagy. 28,30 In addition, studies have been reported that Cx43 to sustain cell survival under hypoxic stress. 33 A previous study verified many HIF-1α target genes, such as GLUT-1, GLUT-3, Hx-1, Hx-2 and Cx43. 33 Hypoxic stress reportedly increases the expression of Cx43 by transcriptional activation through HIF-1α in melanoma cells in cancer. 30 Here, we identified Cx43 as one of the targets of HIF-1α in PH. HIF-1α induces Cx43 expression and its phosphorylation at Ser 368 activity in PASMCs through binding to the promoter of the Cx43 gene (GJA1). We provide a novel physiological role for HIF-1α in proliferation and migration of PASMCs through its induction of Cx43 expression and phosphorylation of Cx43 at Ser 368 activity.
In line with this result, we showed that CoCl 2 -treatment induced the expression of Cx43 in PASMCs in a time-dependent manner.
Research shows that HIF-1α is involved in hypoxia-induced PASMCs proliferation and migration. 34 and astrocytes. 37 Our results suggest that reduced HIF-1α expression is accompanied by a decrease in the levels of Cx43 mRNA, Cx43 protein and phosphorylation of Cx43 at Ser368, showing the existence of a negative feedback regulation between HIF-1α and Cx43 in PASMCs.
The present study demonstrates that hypoxia promotes the proliferation and migration of PASMCs via the HIF-1α/Cx43 axis.
Expression of Cx43 and its phosphorylation activity are necessary for PASMCs proliferation and migration under hypoxia. These findings suggest that Cx43 is a novel therapeutic target for PH.

ACK N OWLED G EM ENT
We acknowledge Miss Ming-Fang Liao for the technical assistance in CHIP experiments and appreciate Dr. Yang Wang for his valuable comments on this paper.

CO N FLI C T O F I NTE R E S T S
The authors have declared that no competing interest exists.

DATA AVA I L A B I L I T Y S TAT E M E N T
The authors confirm that the data supporting the findings of this study are available within the article.