HRSS and media (HRM) were produced as previously described (20). Briefly, hydrogen gas was dissolved in saline and in Dulbecco’s modified Eagle’s serum (DMEM; Gibco, USA) supplemented with 10% fetal bovine serum (FBS). We obtained saturation using a pressure of 0.4 MPa for 6 h. HRSS and HRM were maintained at 4°C and fresh solutions were prepared each week. Gas chromatography was used to confirm the hydrogen levels by the method described by Ohsawa et al (6). Cells were cultured in closed culture flasks.

All experimental procedures involving animals were approved by the Institutional Animal Care and Use Committee of the Hebei Medical University (Shijiazhuang, China). A total of 30 male Sprague-Dawley rats (weight, 280–320 g) were purchased from the animal center of the University and were divided into five groups: the sham group (no angioplasty) and the balloon-injured groups including the control group and 3 groups administered with HRSS at 2.5, 5, 10 ml/kg, intraperitoneally, respectively.

Rats were anaesthetized with 3.6% (w/v) chlorohydrate (1 ml/100 g, intraperitoneally). Left common carotid artery (CCA) was balloon-denuded as previously described (21,22). In brief, a median incision was performed on the anterior neck and left carotid arteries were isolated. A ligature in external carotid artery (ECA) was distally placed and the proximal ends of the CCA and internal carotid arteries (ICA) were clamped. An incision was performed in the ECA. Following blood removal, a 0.13 mm balloon catheter was delicately introduced into the CCA using the incision in the ECA. The balloon was inflated using saline to distend the CCA and pulled back to the ECA. The catheter was removed and the ECA’s proximal end was ligated. Clamps were removed to re-establish blood flow.

Rats were then housed (12 h light cycles) and fed for 2 weeks (free access to food and water), and were then sacrificed using a pentobarbital overdose. Sections from carotid arteries from both sides were excised and fixed using 4% paraformaldehyde. Histologically stained sections were used to assess the extent of neointimal formation using computed planimetry. The intima-to-media (I/M) area ratio was calculated as the mean of these determinations (23).

VSMCs were isolated from the thoracic aorta of 10–12-week-old male Sprague-Dawley rats (23–25) and were cultured (37°C in a humidified 5% CO₂ incubator) in DMEM with 10% FBS, penicillin 100 U/ml and streptomycin 100 μg/ml. Purity was confirmed by immunocytochemical localization of α-smooth muscle actin. Upon confluence, cells were sub-cultured using 0.5% (w/v) trypsin. Media were changed every 3 days and experiments were performed between 3 to 6 passages.

VSMCs (4×10³ cells/well) were placed in a 96-well plate. Following incubation, sterile MTT (20 μl of 5 mg/ml) was added, and incubated for 4 h at 37°C; 150 μl of dimethyl sulfoxide (DMSO) was then added. Absorbance (490 nm) was measured using a microplate reader (26).

VSMCs were synchronized by serum deprivation and incubated with 10% FBS for 24 h with or without HRM. BrdU labeling mixture was then added and cultures were incubated for 12 h at 37°C. Cell numbers were measured using a cell proliferation ELISA (Roche Molecular Biochemicals, Germany) (9,27).

The migration of VSMCs was evaluated by performing a cell-wounding assay according to a previously described method (28). Cells grown to 100% confluence on glass slides were scraped off the slides using a cell scraper. These cells were used to create a 3-mm-wide wound and were then incubated at 37°C for 24 h in DMEM containing 10% FBS. Cells were then fixed with methanol and stained with hexamethyl pararosaniline. Cell migration activity was expressed as the number of cells that migrated into the wound area in each field.

VSMCs were synchronized at the G0-phase by serum depletion for 48 h. After replenishment with fresh DMEM, the cells were preincubated with hydrogen-rich medium, and 10% FBS was added to allow progression of cell cycle. After 48 h, cells were trypsinized, centrifuged at 1,250 x g for 5 min. DNA was stained with propidium iodide (PI) (50 μg/ml) for 30 min at 37°C, and 5×10³ cells were then analyzed by flow cytometry.

To assess apoptosis, we suspended HRM-cultured cells in buffer (400 μl; 10 mM HEPES/NaOH; 140 mM NaCl; 2.5 mM CaCl₂; pH 7.4) after a careful wash. FITC-conjugated Annexin V (5 μl; BD Biosciences, San Diego, CA, USA) was added and cells were incubated for 15 min at 4–8°C in the dark. PI (10 μl) was added. After 5 min, cells were washed again and 400 μl of buffer was added. Cells were counted using a FACSCalibur (BD Biosciences).

Total RNA was extracted using the TRIzol reagent method and cDNA was obtained using a cDNA synthesis kit (TransScript first-strand) according to the manufacturer’s instructions. Bcl-2 primers were: ATGGGGTGAAC TGGGGGAGGATTG (forward) and TTTCATATTTGTTT GGGGCAGGTC (reverse). Bax primers were: GAGAGGAT GGCTGGGGAGAC (forward) and GGTGAGCGAGGCGG TGAGGACT (reverse). DNA fragments were analyzed on a 1.5% agarose gel. Quantitative RNA data were expressed by normalizing band density to GAPDH (internal control).

The cells and arterial tissue were solubilized or homogenized in lysate buffer on ice. Homogenates were centrifuged at 12,000 x g for 30 min at 4°C, yielding a cell protein supernatant. Proteins were subjected to SDS-PAGE and transferred to polyvinylidene difluoride membranes (PVDF; Millipore, Billerica, MA, USA). Samples were then probed with antibodies against: Ras, phosphorylated and nonphosphorylated ERK1/2, MEK1/2, proliferative cell nuclear antigen (PCNA), Akt and β-actin. Non-specific binding was blocked with 1.5% (w/v) evaporated skimmed milk (Difco, Franklin Lakes, NJ, USA) in TBS (154 mM NaCl, 10 mM Tris base). Anti-rabbit or anti-mouse secondary antibodies conjugated to IRDYe700DX and IRDYe800 (1:5,000; Rockland Immunochemicals, Gilbertsville, PA, USA) were used to probe primary antibodies. Protein bands were detected and quantified on an Odyssey 2-color infrared imaging system (LI-COR Biosciences, Lincoln, NE, USA). The integrated signal densities were normalized first to β-actin (the loading control) and subsequently expressed in terms of the fraction abundance relative to control cells or arterial tissues. These experiments were performed in triplicate.

Immunohistochemistry was performed as previously described (29). The control and sham groups were treated with anti-PCNA (1:100) antibody on Day 14 following balloon injury. Sections were counterstained with hematoxylin. Staining intensities were determined by measurement of the integrated optical density (IOD) with light microscopy using a computer-based Image-Pro Morphometric system. Measurements were conducted by 2 independent observers in a double-blind manner.

VSMCs (1×10⁵ cells) were incubated with 10% FBS in the presence or absence of HRM for 12 h. Cells were stained with 10 μM of DCFH-DA for 30 min at 37°C, detached with trypsin/EDTA, washed, re-suspended in PBS, and immediately analyzed by flow cytometry using a FACScan by fluorescence intensity (FL-1,525 nm) of 10,000 cells (23).

Following incubation with HRM for 24 h, cells were collected and washed with cold PBS, and hyperplastic neointimal tissues were homogenized. Tissue samples (0.1 ml), 8.1% SDS (0.2 ml), 20% acetic acid buffer solution (1.5 ml, pH 3.5), 1% total bile acid (TBA) thiobarbituric solution (1.5 ml), and distilled water (1 ml) were added. The solution was heated in a 95°C water bath for 40 min and then allowed to cool; it was then centrifuged at 3,500 rpm/min for 15 min and measured at 532 nm using a spectrophotometer (14,30).

The 8-OHdG amounts were determined in VSMCs and hyperplastic neointimal tissues using a Bioxytech 8-OHdG-EIA kit. DNA was isolated using the DNAzol reagent and quantified. DNA (400 μg) was resuspended in 50 μl reaction mixture containing 100 mmol/l sodium acetate (pH 5.0) and 5 mmol/l MgCl₂ and digested using DNaseI. Assay was performed according to the manufacturer’s instructions (Shanghai Lanji Biological Institute, China).

Data are provided as the means ± SEM. One-way analysis of variance (ANOVA) followed by the LSD-t-test were performed using SPSS 19.0 software. P<0.05 was considered to indicate a statistically significant difference.

Balloon-induced intimal hyperplasia was evident in treated rats compared with rats of the control group (Fig. 1A). Results showed that 3 doses of HRSS were effective for prevention of neointimal formation in a dose-dependent manner (Fig. 1A). These findings showed neointimal area ratio reductions of 36.9, 52.2 and 71.8% (P<0.01) in HRSS-treated groups using 2.5, 5 ml and 10 ml/kg doses, respectively, compared with the balloon-injured control group (Fig. 1B).

Western blot analysis revealed that the activation of ERK1/2, MEK1/2 and Ras was strongly suppressed in a dose-dependent manner in 14-day HRSS-treated samples compared with control samples (P<0.01) (Fig. 1E).

PCNA-positive cells were abundant in the balloon injury group. However, PCNA immunostaining was much less apparent in cells treated with HRSS. Staining intensities were lower in HRSS-treated rats (P<0.01) (Fig. 1C and D).

Treatment with HRSS significantly inhibited the levels of carotid artery MDA and 8-OHdG in a dose-dependent manner (Fig. 2).

MTT assay showed that VSMC viability in the HRSS group was significantly lower than in the control group (32.4±5.2 vs. 73.3±5.1%; P<0.01). Two VSMC cultures were treated with HRM for 48 h; hydrogen was then ceased for one culture, while it was continued in the other. Following the removal of hydrogen, cells resumed viability (Fig. 3A). Cells without hydrogen treatment exhibited a BrdU index of 64.1±11.1%, compared to 26.7±4.9% in the hydrogen group (Fig. 3B).

VSMC cell count was smaller in HRM-treated cells after 48 and 72 h (40 and 51% reductions, respectively; P<0.01) (Fig. 4A and B). VSMCs treated with FBS for 24 h showed greater mobility (FBS 58.3±13.9 vs. hydrogen 18.6±5.8 cells; P<0.01) (Fig. 4C and D).

Treatment with HRM inhibited FBS-induced G1-S progression, as demonstrated by the increase in G0/G1 cells (68.8±2.2%) accompanied by concurrent decrease in S-phase cells (16.7±2.0%) (Fig. 5A). These results suggest that hydrogen may prevent FBS-induced S-phase entry in VSMCs via a G0-G1 blocking mechanism.

Proportion of early apoptotic cells (Fig. 6A and B; lower right) increased from 1.36% (FBS-treated) to 2.1% (FBS + hydrogen-treated) (P<0.05). These results also showed that the proportion of late apoptotic cells (Fig. 6A and B; upper right) increased from 6.96 to 13.47% (P<0.01).

Following HRM treatment, a significant increase in Bax/Bcl-2 ratio (P<0.01) indicated that VSMCs were, in fact, progressing towards apoptosis (Fig. 6C and D).

FBS treatment significantly induced intracellular peroxide production (Fig. 7A and B). ROS generation in cells treated with HRM was 395.6±16.1, compared with 483.0±15.1 for controls (P<0.01).

The 8-OHdG levels in cells treated with HRM were 1.3±0.0 ng/ml, compared to 1.6±0.1 ng/ml in controls (P<0.01) (Fig. 7C). Fig. 7D shows that treatment with HRM significantly inhibited MDA generation (FBS 3.7±0.2 vs. hydrogen 2.9±0.1; P<0.01).

Stimulation of cells with 10% FBS induced Ras activation. Cells treated with HRM exhibited a slight inhibition of FBS-induced Ras activation (Fig. 8A). Treatment of FBS-stimulated VSMCs with HRM markedly decreased phosphorylated MEK1/2 levels (Fig. 8C). Data showed that FBS also induced a profound increase in ERK1/2 activation. Treatment with HRM significantly inhibited FBS-stimulated phosphorylation of ERK1/2 (Fig. 8B). By contrast, total ERK1/2 protein levels were not altered by treatment with hydrogen. FBS induced a profound increase in Akt activation. The level of Akt phosphorylation following FBS stimulation was also significantly inhibited by treatment with HRM (Fig. 5E). As shown in Fig. 8D, FBS-induced PCNA expression was significantly inhibited by HRM by 50% (P<0.01).