H₂ (99.999%; Guang Zhou Guang Qi GAS Co., Ltd, Guangdong, China) was stored in the seamless steel gas cylinder, and it was injected into an aseptic soft plastic infusion bag (100 ml; CR Double-Crane Pharmaceuticals Co., Ltd, Anhui, China) under sterile conditions immediately before intraperitoneal injection. ISO (I5627, Sigma–Aldrich, St. Louis, MO, USA) was dissolved in normal saline (5 mg/10 ml) under sterile conditions immediately before subcutaneous injection, and dissolved in double distilled water as 10 mM stock solution 30 min before use. The antibodies against extracellular signal-regulated kinase 1/2 (ERK1/2), p-ERK1/2, p38, p-p38, c-Jun NH2-terminal kinase (JNK), and p-JNK, p67 (phox) were from Cell Signaling Technology (Danvers, MA, USA). The antibody against β-actin was from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Anti-α-actin antibody was from Sigma–Aldrich (St. Louis, MO, USA). The antibody against 3-nitrotyrosine was from Abcam (Cambridge, MA, USA). JC-1 was from Beyotime Biotechnology (C2006, Jiangsu, China).

H₂-rich medium was prepared as previously described (Zhang et al., 2016). The concentration of H₂ was measured by MB-Pt reagent (generously provided by Ming Yan, Shanghai Nanobubble Technology Co., Ltd, Shanghai, China) as previously described (Zhang et al., 2016). The H₂ concentration in our H₂-rich medium was no less than 0.6 ppm (0.6–0.9 ppm).

H9c2 rat cardiac myoblasts (a cardiomyoblast cell line derived from embryonic rat heart tissue; generously provided by Prof. Hongliang Li, Wuhan University, China) were grown in DMEM containing 5.5 mM glucose as described previously (Jeong et al., 2009). To induce hypertrophy, cells were serum starved for 18 h in DMEM containing 1% FBS, and then treated with 10 μM ISO for 48 h (Jeong et al., 2009). In order to investigate the effect of H₂ on the blockage of ISO-induced hypertrophy, H₂-rich medium was added 30 min before ISO administration, the medium, and ISO were re-changed every 24 h, and cardiomyocytes hypertrophic response was examined after 48 h of ISO challenge (Jeong et al., 2009).

The C57BL/6J mice (aged 8–10 weeks, male) were obtained from the Laboratory Animal Center of Sun Yat-sen University. The animals were housed with 12-h light–dark cycles and allowed to obtain food and water ad libitum. All experimental procedures and protocols were approved by Institutional Animal Care and Use Committee (Zhongshan School of Medicine, Sun Yat-sen University), and conformed to the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health (NIH publication NO. 85-23, revised 1996).

Cardiac hypertrophy was induced by subcutaneous injection of ISO (0.5 mg/100g/day) for 7 days as previously revealed (Tshori et al., 2006). Mice were randomly assigned to four groups: Control (Con) group (n = 10), ISO group (n = 12), ISO plus H₂ group (n = 12), and H₂ group (n = 12). H₂ was given at the dose of 1 ml/100g/day by intraperitoneal injection as previously described (Huang et al., 2013; Zhang et al., 2016). Mice in ISO plus H₂ group and H₂ group received H₂ consecutively for 7 days before receiving ISO, and continued for another 7 days. On the 8th day, mice in ISO group, and ISO plus H₂ group received ISO for 7 days until animals were sacrificed on the 15th day. After sacrifice, hearts were excised, rinsed with ice-PBS, and blotted dry. Hearts were weighed; the heart weight/body weight (HW/BW) ratios were calculated and expressed as milligrams HW per gram BW. Then hearts were snap frozen in liquid nitrogen within minutes and stored at -80°C until analyzed.

Transthoracic echocardiography was performed to assess left ventricular function before sacrificed on the 15th day in a blinded manner. Mice were anesthetized with 1.5–2% isoflurane, and hearts were visualized using a RMV707B (30 M Hz) scan-head interfaced with a Vevo-2100 high frequency ultrasound system (VisualSonics Inc., Toronto, Canada) at least three times for each animal indicated (Webb et al., 2010).

Hearts were excised, washed with ice-PBS, fixed in 10% buffered formalin, and cut transversely close to the apex cordis to visualize the left and right ventricles. Several sections of heart (4–5 μm thickness) were prepared and stained with hematoxylin and eosin (H&E) for histopathology and then visualized by light microscopy.

For immunostaining, anti-sarcomeric α-actin antibody was used to assess the cell surface area of H9c2 cardiomyocytes as described previously (Akimoto et al., 1996). To assess 3-nitrotyrosine levels in heart, which can reflect formation of ONOO–, primary antibody against 3-nitrotyrosine (1:50) was used as previously described (Zhang et al., 2011).

Mitochondrial membrane potential was determined by the dye 5,5′,6,6′-tetrachloro-1,1′,3,3′-tetraethylbenzimidazolcarbo-cyanine iodide (JC-1) as previously described with slight modification (Cossarizza et al., 1993). Briefly, the treated cells were washed with PBS, and then incubated with JC-1 staining dye (culture medium: JC-1 working dye = 1:1) at 37°C in the dark for 20 min and rinsed three times with cold PBS, and analyzed by fluorescence microscope (Axio Observer Z1, Carl Zeiss. Inc.). The JC-1 aggregates, which was accumulated in the inner membrane of mitochondria, emitted red fluorescence and represented the high MMP, while green fluorescence reflected JC-1 monomer which entered in the cytosol following mitochondrial membrane depolarization. When mitochondria is damaged, the red/green ratio decreases. The ratio of JC-1 aggregates to monomer (red/green) intensity for each region was calculated by Image-Pro Plus software (version 6.0).

Total mRNA was extracted from left ventricles and H9c2 cardiomyocytes using TRIZol reagent (15596-026, Invitrogen) according to the manufacturer’s instruction, and cDNA was synthesized using oligo (dT) primers with the Transcriptor First Strand cDNA Synthesis Kit (PrimeScript^TM RT Master Mix, Takara). Selected gene differences were confirmed by qRT-PCR using SYBR green (SYBR^® Premix Ex Taq^TM, Takara). The target gene expression was normalized to GAPDH gene expression. The primers for qRT-PCR are shown in Table 1.

Western blotting was performed as previously described (Zhang et al., 2016). The membranes were incubated with primary (1:2000) and secondary (1:2000) antibodies by standard techniques. Immunodetection was accomplished using enhanced chemiluminescence (ChemiDoc XRS+ System, Bio-Rad, Hercules, CA, USA).

Cardiac total ROS was stained with dihydroethidium (DHE, D-23107; Invitrogen) on fresh frozen sections as previously described (Zhang et al., 2014). Images were immediately acquired using confocal microscopy (Leica Model SPE, Leica Imaging Systems Ltd) using λ_ex 405 nm laser excitation.

Data are expressed as mean ± SD. Differences among groups were tested by one-way ANOVA. Comparisons between two groups were performed by unpaired Student’s t-test. A value of P < 0.05 was considered to be significantly different.

In order to investigate the effects of H₂ on cardiac hypertrophy, ISO was used to induce cardiac hypertrophy in mice. As expected, mice with chronic ISO infusion exhibited cardiac hypertrophy compared to the control group, as indicated by the gross morphology of hearts, heart sections stained with H&E (Figure 1A). The hypertrophic marker gene ANP and BNP mRNA levels (Figures 1B,C, P < 0.05 vs. Con), and HW/BW ratio (Table 2, P < 0.05 vs. Con) were also increased. Pretreatment with H₂ (intraperitoneal injection) at the dose of 1 ml/100g/day reversed these hypertrophic responses (Figure 1; Table 2, P < 0.05 vs. ISO). Moreover, H₂ injection alleviated the impaired left ventricular function, as evidenced by decreasing left ventricular end-systolic diameter (LVESD), left ventricular end-diastolic diameter (LVEDD), and increasing fractional shortening (FS%; Table 2, P < 0.05 vs. ISO). However, there were no significant changes between control group and H₂ group. Collectively, these data suggested that H₂ injection prevented the development of ISO-induced cardiac hypertrophy and preserved cardiac function in vivo.

As the heart primarily consists of cardiomyocyte and fibroblast, therefore, we investigated whether H₂ could target cardiomyocyte for hypertrophic inhibition. H₂-rich medium and H9c2 cardiomyocytes were used for in vitro studies. First, we used CCK8 to investigate the possible cytotoxity of H₂-rich medium on H9c2 cardiomyocyte. H₂ was shown to be non-cytotoxic for cardiomyocyte treating with H₂-rich medium for 48 h (data not shown). After 48 h of ISO stimulation, cardiomyocyte surface areas, and the hypertrophic marker gene ANP and BNP mRNA levels were significantly increased in H9c2 cardiomyocyte (Figures 2A–D, P < 0.05 vs. Con). H₂-rich medium attenuated these hypertrophic responses of H9c2 cardiomyocyte (Figures 2A–D, P < 0.05 vs. ISO). These data indicated that H₂ could also inhibit ISO-induced cardiomyocyte hypertrophy in vitro.

ROS play a critical role in the development of cardiac hypertrophy and heart failure (Burgoyne et al., 2012). ROS levels were increased in the left ventricular of ISO-infused mice compared with control mice, and this increase was inhibited by pretreatment with H₂ at the dose of 1ml/100g/day (Figure 3A). Moreover, another oxidative stress marker, 3-nitrotyrosine (3-NT), which reflects the formation of ONOO–, was also upregulated by ISO stimuli, and suppressed by H₂ (Figure 3B). To confirm these in vivo findings, we evaluated the effects of H₂-rich medium on the levels of 3-NT stimulated by ISO in vitro. The accumulation of 3-NT was increased after ISO stimulation, while H₂-rich medium attenuated this effects (Figure 3C, P < 0.05).

To further understand the mechanism of H₂ in blocking ROS accumulation, we tested the NADPH oxidase subunit p67 (phox) expression. Immunoblotting revealed the expression of p67 (phox) was increased in left ventricular of ISO-infused mice, and this increase was alleviated by H₂ (Figure 3D, P < 0.05). As we have mentioned above, NADPH oxidase-derived ROS can stimulate and amplify mitochondrial ROS production and induce mitochondrial dysfunction (Zorov et al., 2000; Dai et al., 2011a), and these can be reflected by the change of MMP. ISO induced the depression of MMP, as indicated by high levels of green fluorescence and low levels of red fluorescence. Interestingly, H₂-rich medium blocked the depression of MMP induced by ISO (Figures 4A,B, P < 0.05). Therefore, these data indicated that H₂ inhibited the excess ROS accumulation following ISO stimuli through attenuating NADPH oxidase expression and mitochondrial damage.

Based on the inhibitory effect of H₂ on the ISO-induced excess accumulation of ROS in vitro and in vivo, we further investigated its effect on the downstream hypertrophic targets, such as mitogen-activated protein kinases (MAPKs) signaling pathways. Following ISO stimuli, the phosphorylation of ERK1/2, p38 MAPK (p38), and c-Jun NH2-terminal kinase (JNK) were increased to the high level at 5 min, and came to the base line at 30 min (Figure 5, P < 0.05 vs. 0 min). These enhanced activation of MAPKs could be blocked by H₂-rich medium in vitro (Figure 6; P < 0.05 vs. ISO). Similarly, the activation of MAPKs were enhanced in the hearts of ISO-infused mice compared with control group (Figure 7, P < 0.05 vs. Con). Such changes were inhibited by pretreatment with H₂ in vivo (Figure 7, P < 0.05 vs. ISO). Thus, H₂ suppressed the enhanced phosphorylation of ERK1/2, p38, and JNK to alleviate ISO-mediated cardiac hypertrophy in vivo and cardiomyocyte hypertrophy in vitro.