All procedures involved in human subjects were approved by the Ethics Committee of the First Affiliated Hospital of Xi’an Jiaotong University according to those of the 1964 Helsinki Declaration and its later amendments for ethical research involving human subjects. Forty-one health people with systolic blood pressure 120 mmHg and diastolic blood pressure 80 mmHg were recruited at the First Affiliated Hospital of Xi’an Jiaotong University. Sixty-eight hypertensive patients with systolic blood pressure $\ge$140 mmHg and/or diastolic blood pressure $\ge$90 mmHg were recruited at the First Affiliated Hospital of Xi’an Jiaotong University. Written informed consents were obtained from all the human subjects. Patients who had cognitive impairment or incomplete tests or data were excluded in this study. Three mL peripheral blood from all the human subjects was collected in tubes containing EDTA. The samples were then centrifuged to separate the plasma at 1000 g for 20 minutes. RNAs were extracted from the plasma via mirVana™ PARIS™ kit (Ambion, Austin, TX, USA) for the subsequently analysis. The clinical characteristics of patients with hypertension were shown in Table 1.

HASMCs (human aortic smooth muscle cells) or human vascular smooth muscle cells (HVSMCs) were acquired from Lonza (Rockland, ME, USA) and cultured in Smooth Muscle Cell Medium (SciencCell, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (Lonza) at 37 ${}^{\circ }$C humidified incubator with 5% CO${}_2$. For the establishment of Ang II-induced hypertension cell model, HASMCs or HVSMCs were cultured in Smooth Muscle Cell Medium containing 2 $\mu$M Ang II for 12 hours before the functional assays. For the treatment of Ang II, HASMCs were cultured in Smooth Muscle Cell Medium containing 2 $\mu$M Ang II and 200 or 500 ng/mL human Wnt4 recombinant protein (R&D Systems, Minneapolis, MN, USA) for 12 hours before the functional assays.

Mimic (5${}^{\prime }$-AACCAGCACCCCAACUUUGGACGGTATTCGCACTGGATACGACGAACTTT-3${}^{\prime }$) and inhibitor (5${}^{\prime }$-CACUACUUUUGUGUCCCACUU-3${}^{\prime }$) of miR-634, as well as the corresponding negative controls (NC mimic, NC inhibitor), were acquired from Ribobio (Guangzhou, China). Full length of Wnt4 was constructed into pcDNA3.1 vector (Invitrogen, Carlsbad, CA, USA). HASMCs were transfected with miR-634 mimic, inhibitor, NC mimic, NC inhibitor, pcDNA-Wnt4 or pcDNA vector by Lipofectamine 3000 (Invitrogen). Twenty-four hours later, cells were treated with or without 2 $\mu$M Ang II, 200 or 500 ng/mL human Wnt4 recombinant protein for 12 hours before the functional assays.

HASMCs (4000 cells/well) were seeded and then treated with or without different treatment or transfection. After the different treatment, cultured medium of HASMCs were changed to fresh medium containing 50 $\mu$M EdU (Sigma-Aldrich, St. Lou-is, MO, USA) for 4 hours. Cells were then fixed in 4% paraformaldehyde, and incubated with EdU antibody for 2 hours. After washing with phosphate-buffered saline, cells were incubated with 200 $\mu$L Apollo staining reaction buffer. The immunostaining was in 5 randomly selected fields measured under fluorescence microscopy (Nikon, Tokyo, Japan) with nuclei staining with DAPI.

HASMCs (30000 cells/well) in serum-free medium were seeded to the top of the chamber with 8 $\mu$m pore size (Coatar, Cambridge, MA, USA). Cultured medium containing 20% fetal bovine serum was added to the low chamber. Twenty-four hours after incubation, migrating cells to the low chamber were washed, fixed and stained with crystal violet. Cells in 5 randomly selected fields were measured under microscope (Nikon) using Image J (NIH, Bethesda, MD, USA).

The 3${}^{\prime }$ UTR of Wnt4 containing miR-634 binding site or mutant control without the binding site were subcloned into pGL3 vector (GenePharma, Suzhou, China). HASMCs (20000 cells/well) were seeded and then co-transfected with pRL-TK (10 ng), miR-634 mimic or NC mimic (20 nM) and the pGL3 vectors (100 ng). Luciferase activities were performed via Dual luciferase reporter assay kit (Promega, Madison, WI, USA) 48 hours after transfection.

Plasma RNAs, miRNAs or RNAs extracted from HASMCs were isolated and reverse-transcribed into cDNAs. qRT-PCR analysis was conducted with QuantiTect SYBR Green PCR Master Mix (Qiagen, Valencia, CA, USA). GAPDH or U6 were used as endogenous controls. The primer sequences were showed as below: miR-634 (F: 5${}^{\prime }$-ATTATGTTAGTTAGGATGGTTTCGA-3${}^{\prime }$; R: 5${}^{\prime }$-ATATCCACAAACAAATAACTTCGTT-3${}^{\prime }$); Wnt4 (F: 5${}^{\prime }$-AGGAAGCCAGGTAAAGTGCTC-3${}^{\prime }$; R: 5${}^{\prime }$-CAGCCAGACCACTCAAAGTTG-3${}^{\prime }$); GAPDH (F: 5${}^{\prime }$-ACCACAGTCCATGCCATCAC-3${}^{\prime }$; R: 5${}^{\prime }$-TCCACCACCCTGTTGCTGTA-3${}^{\prime }$); U6 (F: 5${}^{\prime }$-CTCGCTTCGGCAGCACATA-3${}^{\prime }$; R: 5${}^{\prime }$-AACGATTCACGAATTTGCGT-3${}^{\prime }$).

For isolation of cytoplasm and nuclear fractionation, HASMCs were harvested and suspended in isolation buffer A containing protease inhibitors (HiScript II First Strand cDNA Synthesis Kit; Vazyme Biotech, Nanjing, Jiangsu, China). After rotating for 1 minute and centrifuging at 12000 g for 5 minutes, supernatant with cytoplasm fraction was collected. The debris was suspended in isolation buffer B containing protease inhibitors, and then rotated for the collection of nuclear fractionation. For western blot analysis, proteins (30 $\mu$g) extracted from HASMCs were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis, and then electro-transferred onto nitrocellulose membrane. Membranes were blocked in 5% skim milk, and incubated with primary antibodies: anti-Wnt4 (1:1500; Abcam, Cambridge, MA, USA), anti-$\beta$-catenin (1:2500; Abcam, Cambridge, MA, USA), anti-Histone, anti-GAPDH and anti-$\beta$-actin (1:3000; Abcam, Cambridge, MA, USA). After incubation with horseradish peroxidase labeled secondary antibody (1:5000; Abcam, Cambridge, MA, USA), the signals were determined by Amersham enhanced chemiluminescence detection system (GE Healthcare Life Sciences, Little Chalfont, UK).

All the experiments were performed at least in triplicates. Data were expressed as mean $\pm$ SD, and performed by SPSS 19.0 software (SPSS, Chicago, IL, USA). Statistical analysis was determined by Student’s t-test or one-way analysis of variance. p value 0.05 was considered to be statistically significant.

To establish cell model of hypertension, HASMCs were incubated with culture medium containing 2 $\mu$M Ang II for 12 hours. Ang II treatment promoted HASMCs proliferation, as demonstrated by increased EdU incorporation (Fig. 1A). Ang II treatment also promoted migration of HASMCs by (Fig. 1B), as shown by increased cell migration in HASMCs treated with Ang II compared with the control (Fig. 1C), suggesting successful cell model of hypertension. miR-634 was down-regulated in Ang II-induced HASMCs in a time dependent manner (Fig. 1D), suggesting that miR-634 might be involved in regulation of HASMCs proliferation and migration.

Fig. 1.

miR-634 was reduced in Ang II-induced HASMCs.
(A) The effect of 2 $\mu$M Ang II on cell proliferation of HASMCs was detected by EdU staining. (B) The effect of 2 $\mu$M Ang II on cell migration of HASMCs was detected by transwell. (C) Invasion cells number affected by Ang II treatment. (D) The effect of 2 $\mu$M Ang II on miR-634 expression of HASMCs in different time. ** p 0.01. Scale bars: 50 $\mu$m.

To evaluate the role of miR-634 in HASMCs proliferation and migration, HASMCs or HVSMCs were transfected with miR-634 mimic and then treated with Ang II. qRT-PCR confirmed the up-regulation of miR-634 in HASMCs (Fig. 2A) and HVSMCs (Supplementary Fig. 1A) transfected with miR-634 mimic. Over-expression of miR-634 suppressed cell proliferation (Fig. 2B) and migration (Fig. 2C,D) of HASMCs, as evidenced by the decreased EdU incorporation (Fig. 2B) and invasion cells number (Fig. 2D) in HASMCs transfected with miR-634 mimic compared with cells transfected with negative control. Moreover, transfection with miR-634 mimic suppressed the cell proliferation (Fig. 2B) and migration (Fig. 2C,D) of Ang II-induced HASMCs, suggesting that miR-634 might contribute to amelioration of hypertension. The cell proliferation (Supplementary Fig. 1B) and migration (Supplementary Fig. 1C,D) of HVSMCs were also promoted by Ang II treatment, while reduced by over-expression of miR-634.

Fig. 2.

miR-634 suppressed Ang II-induced HASMCs proliferation and migration.
(A) Transfection efficiency of miR-634 mimic in HASMCs was detected by qRT-PCR. (B) Effect of miR-634 on cell proliferation of HASMCs with or without Ang II treatment. (C) Effect of miR-634 on cell migration of HASMCs with or without Ang II treatment. (D) Effect of miR-634 on invasion cells number of HASMCs with or without Ang II treatment. *, **, p 0.05, p 0.01. NC mimic: negative control of miR-634 mimic. Scale bars: 50 $\mu$m.

To investigate the mechanism of miR-634 in HASMCs, the target gene of miR-634 was predicted as Wnt4 via Targetscan (http://www.targetscan.org/vert_72/) (Fig. 3A). Luciferase activity of pGL3-WT-Wnt4 was decreased in HASMCs transfected with miR-634 mimic compared with the negative control (Fig. 3B), while activity of pGL3-MUT-Wnt4 was not affected by miR-634 mimic compared with the negative control (Fig. 3B), suggesting that miR-634 could target Wnt4 in HASMCs. To validate effect of miR-634 on Wnt4 expression, HASMCs were transfected with miR-634 mimic or inhibitor (Fig. 3C). Results revealed that mRNA (Fig. 3D) and protein (Fig. 3E) of Wnt4 were reduced by miR-634 mimic, while enhanced by the inhibitor. Moreover, Wnt4 was up-regulated in HASMCs after 12 hours of Ang II treatment (Supplementary Fig. 2A,B). These results indicated that miR-634 could target Wnt4 and repress its expression.

Fig. 3.

Target gene of miR-634.
(A) Potential binding target of miR-634 was predicted as Wnt4. (B) The effect of miR-634 on luciferase activities of pGL3-WT-Wnt4 and pGL3-MUT-Wnt4. (C) Transfection efficiency of miR-634 mimic or inhibitor in HASMCs was detected by qRT-PCR. (D) Effect of miR-634 on mRNA expression of Wnt4 in HASMCs. (E) Effect of miR-634 on protein expression of Wnt4 in HASMCs. ** p 0.01. NC inhibitor: negative control of miR-634 inhibitor.

To investigate the role of miR-634/Wnt4 axis in HASMCs, Ang II-induced HASMCs were transfected with pcDNA-Wnt4 or cotransfected with miR-634 mimic and pcDNA-Wnt4. The transfection efficiency of pcDNA-Wnt4 was shown in Supplementary Fig. 3. Results showed that over-expression of Wnt4 aggravated the promotive effects of Ang II on cell proliferation (Fig. 4A) and migration (Fig. 4B,C). However, cotransfection with miR-634 mimic and pcDNA-Wnt4 attenuated the promotive effects of Wnt4 on Ang II-induced HASMCs proliferation (Fig. 4A) and migration (Fig. 4B,C). These results showed that miR-634/Wnt4 axis participated in regulation of Ang II-induced HASMCs proliferation and migration. As a downstream target of Wnt4, nuclear distribution of $\beta$-catenin (Fig. 4D) was decreased in Ang II-induced HASMCs transfected with miR-634 mimic, while the cytoplasmic distribution was increased (Fig. 4D). Moreover, over-expression of Wnt4 enhanced the nuclear distribution of $\beta$-catenin (Fig. 4D) while decreased the cytoplasmic distribution (Fig. 4D). Furthermore, over-expression of Wnt4 reversed the inhibitory effect of miR-634 on nuclear distribution of $\beta$-catenin (Fig. 4D) and the promotive effect on the cytoplasmic distribution (Fig. 4D). The subcellular distribution of $\beta$-catenin indicated that function of miR-634/Wnt axis was dependent on $\beta$-catenin pathway. In vitro treatment with Wnt4 protein significantly promoted cell proliferation (Fig. 4E) and migration (Fig. 4F) in a dose-dependent manner, as demonstrated that 500 ng/mL Wnt4 treatment showed more EdU incorporation (Fig. 4E) and migration cells (Fig. 4G) of Ang II-induced HASMCs than 200 ng/mL Wnt4 treatment. Moreover, nuclear distribution of $\beta$-catenin (Fig. 4H) was promoted by Wnt4 treatment, while cytoplasmic distribution (Fig. 4G) was suppressed. Therefore, miR-634 could decrease Wnt4 to reduce nuclear distribution of $\beta$-catenin, thus suppressing Ang II-induced HASMCs proliferation and migration.

Fig. 4.

Over-expression of Wnt4 counteracted the suppressive effects of miR-634 on Ang II-induced HASMCs proliferation and migration.
(A) Effect of miR-634 mimic and pcDNA-Wnt4 on cell proliferation of HASMCs with Ang II treatment. (B) Effect of miR-634 mimic and pcDNA-Wnt4 on cell migration of HASMCs with Ang II treatment. (C) Effect of miR-634 mimic and pcDNA-Wnt4 on cell migration of HASMCs with Ang II treatment. (D) Effect of miR-634 mimic and pcDNA-Wnt4 on nucleus and cytoplasmic distribution of $\beta$-catenin in HASMCs with Ang II treatment. (E) The effect of in vitro treatment with 200 or 500 ng/mL Wnt4 protein on cell proliferation of miR-634 mimic-transfected HASMCs under Ang II treatment. (F) The effect of in vitro treatment with 200 or 500 ng/mL Wnt4 protein on cell migration of miR-634 mimic-transfected HASMCs under Ang II treatment. (G) The effect of in vitro treatment with 200 or 500 ng/mL Wnt4 protein on invasion cells number of miR-634 mimic-transfected HASMCs under Ang II treatment. (H) The effect of in vitro treatment with 200 or 500 ng/mL Wnt4 protein on nucleus distribution and decreased cytoplasmic distribution of $\beta$-catenin in miR-634 mimic-transfected HASMCs under Ang II treatment. ** p 0.01. Vector: pcDNA-3.1 vector. Scale bars: 50 $\mu$m.

Expression level of miR-634 in hypertensive patients was evaluated by qRT-PCR. Plasma samples from healthy individuals or hypertensive patients were analyzed and the result showed a significant decrease in miR-634 in hypertensive patients compared with health individuals (p 0.01, Fig. 5). Analysis between miR-634 expression and clinical characteristics of patients with hypertension was shown in Table 1, suggesting that miR-634 might regulate hypertension.

Fig. 5.

miR-634 was reduced in hypertensive patients.
Expression level of miR-634 in hypertensive patients and healthy individuals. ** p 0.01.