Hydrogen-rich saline solution was prepared as previously described (Cai et al., 2009). Briefly, hydrogen was dissolved in saline solution for 6 h under high pressure (0.4 MPa) using an apparatus manufactured in our department. The saturated hydrogen saline solution was stored under atmospheric pressure at 4°C in an aluminum bag with no dead volume. HRSS was sterilized by gamma radiation. To ensure that a concentration of 0.6 mmol/L was maintained, HRSS was freshly prepared every week. The hydrogen content of HRSS was confirmed using the gas chromatography method described by Ohsawa et al. (2007).

Sprague Dawley (SD) rats (250–300 g) were kept in standard cages under a 12-h light/dark cycle. All experimental procedures were approved by the Institutional Animal Care and Use Committee of Jinling Hospital (Nanjing, China).

Animals were divided into four groups: (1) sham-operated rats receiving vehicle physiological saline solution treatment (control group, n = 30); (2) I/R rats receiving vehicle physiological saline solution treatment (I/R group, n = 30); (3) I/R rats receiving HRSS treatment at 4 h intervals for a total of 108 h (long-term treatment group, n = 30); (4) I/R rats receiving HRSS treatment for 24 h and then vehicle physiological saline solution treatment for 84 h (short-term treatment group, n = 30). In the control group, rats received a sham operation, followed by a continuous intraperitoneal administration of 0.9% NaCl solution (1 ml/kg). In the I/R group, the right kidney was surgically removed, and the left renal artery was occluded. This operation was followed by intraperitoneal administration of 0.9% NaCl solution (1 ml/kg) at 4 h intervals. Left renal artery occlusion was performed for 45 min under pentobarbital anesthesia. In the HRSS-I/R groups, the right kidney was surgically removed, and the left renal artery was occluded (Shingu et al., 2010; Wang et al., 2011). This was followed by intraperitoneal administration of HRSS (1 ml/kg) immediately after reperfusion at 4 h intervals. Occlusion of the left renal artery was performed as described above. Rats received saline solution or HRSS immediately after reperfusion at 4 h intervals for a total of 108 h. The animals were anesthetized by intraperitoneal injection of pentobarbital (Biomics Laibo, Beijing; 0.7 mg/kg) before any surgical treatments.

Blood urea nitrogen and Cr were measured to evaluate renal function. Samples were analyzed using commercial kits (Sigma, St. Louis, MO, USA) and a COBAS Mira chemical analyzer (Roche, Basel, Switzerland).

Renal samples were embedded in paraffin, cut into 4 μm sections and stained with hematoxylin and eosin (H&E). Samples evaluation was performed under light microscopy by experienced technicians blinded to the groups. These expert observers made all assessments. To assess tubular injury, the percentages of tubules in the outer medulla and corticomedullary junction with tubular cell necrosis, cytoplasmic vacuole formation, hemorrhage and tubular dilatation were measured. Semi-quantitative injury scores ranged from 0 to 5 (0: normal kidney; 1: 0–10% injury; 2: 11–25% injury; 3: 26–45% injury; 4: 46–75% injury; 5: 76–100% injury).

Kidneys were perfused with PBS (50 ml) using the transcardiac approach, followed by 4% phosphate-buffered formalin. Perfusion-fixed kidney tissues were further fixed overnight in a solution of 4% paraformaldehyde in PBS, embedded in paraffin and cut into 4 μm serial sections. Terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) staining was performed on paraffin embedded sections by using an in situ cell death detection kit (Roche, Indianapolis, USA). According to standard protocols, the sections were dewaxed and rehydrated by heating the slides at 60°C. Then, these sections were incubated in a proteinase K working solution (20 mg/ml) for 15 min at room temperature. The slides were rinsed three times with PBS before the incubation in the TUNEL reaction mixture for 1 h at 37°C. The area around the sample was dried by filter paper, and Converter-AP was added to the samples for 1 h at 37°C. After rinsing with PBS (5 min, three times), sections were stained with nitroblue tetrazolium (NBT) and 5-bromo-4-chloro-3-indolylphosphate (BCIP) until they became dark. Four slide fields were randomly examined using a defined rectangular field area at 100× magnification. One hundred cells were counted in each field. Data are shown as percentage of TUNEL-positive cells of the total cell nuclei per field.

Bax and Bcl-2 imunohistochemical evaluation was carried out as previously described (Nakamura et al., 2007). Five μm tissue sections were deparaffinized, rehydrated in gradient alcohols and processed using the streptavidin immunoperoxidase method. In brief, sections were submitted to antigen retrieval by microwave oven treatment for 10 min in 0.01 mol/L citrate buffer (pH 6.0). Slides were then incubated in 10% normal serum for 30 min, followed by an overnight incubation at 4°C with the appropriately diluted primary antibody. Bax and Bcl-2 antibodies (Beyotime Inc., China) were used at 1:100 dilution. Next, samples were incubated with biotinylated anti-mice or anti-rabbit immunoglobulins for 15 min at 37°C. Then, they were incubated with streptavidin peroxidase complexes for 15 min at 37°C. Four slide fields were randomly examined using a defined rectangular field area at 400× magnification. Bax and Bcl-2 positive cells were counted in each field. Data are shown as the number of Bax and Bcl-2 positive cells per field.

Kidney tissues were collected and washed with normal saline solution. Then, they were immediately homogenized in 1 ml of cold normal saline solution (4°C) on ice. The homogenates were centrifuged at 3,000 × g at 4°C for 15 min. Caspase-3, caspase-9, and caspase-8 activities were evaluated using a Fluorometric Assay Kit (BIOVISION Research Products, Mountain View, CA, USA), according to the manufacturer’s instructions. IL-6 and TNF-α were measured with a commercial ELISA kit, according to the manufacturer’s instructions (Dakawe, Shenzhen, China).

Data are expressed as mean ± standard deviation (SD). Each experiment was repeated three times, and data represent the means from at least three parallel samples. SPSS version 11.0 (SPSS Inc., Chicago, USA) was used for all statistical analyses. Analysis of variance (ANOVA) and Student’s t-test were used to determine the differences among and between groups, respectively. P-values of less than 0.05 were considered statistically significant.

In a preliminary study of hydrogen’s protective effect in kidneys, we analyzed kidney morphology and function at 24 h after I/R injury, and we found similar results to previous studies (Shingu et al., 2010; Wang et al., 2011, 2014). As shown in Figure 1, HRSS significantly reduced renal I/R injury. After HRSS treatment, renal tubule denaturation and edema were significantly attenuated (Figure 1A). Histology results in the normal kidney show that the renal glomerulus and tubules structures were clear without hemorrhage in the renal tubule, without loss of epithelial cells. In contrast, renal tubules in the I/R injured kidney showed evident edema, an almost complete loss of the epithelium and infiltration of red blood cells. The HRSS-treated kidney showed a partial loss of the epithelium in the renal tubules, and the number of red blood cells was significantly decreased, compared to the I/R injured kidney (Figure 1B). Additionally, serum Cr and BUN levels of the rats in the HRSS treatment group were significantly reduced when compared to the control group (P < 0.01; Figure 1C).

Additionally, we detected renal cell apoptosis by TUNEL immunostaining. As shown in Figures 4A,B, TUNEL-positive renal cells were significantly reduced at 24 h after I/R injury in the short-term and long-term treatment groups when compared to the I/R group. However, in the short-term treatment group, renal cell apoptosis still occurred after discontinuation of HRSS treatment, while it remained significantly reduced in the long-term treatment group.

In order to further verify HRSS therapeutic effect on renal I/R injury, we analyzed kidney morphology by H&E staining after I/R injury. At 24 h after I/R injury, renal tubular necrosis was significantly reduced in the short-term and long-term treatment groups. Renal hemorrhage and edema were also significantly attenuated, compared to the I/R group. Continuous HRSS treatment accelerated the recovery of the renal tubules function, when compared with short-term treatment. Histology scores for renal injury showed similar results (Figure 3).

Previous studies on hydrogen and I/R injury mainly focus on the ability of hydrogen on inhibiting ROS to reduce renal I/R injury. It has been well established, however, that the oxidation reaction after I/R injury is a lengthy process and can cause sustained renal injury. Therefore, we speculated that continuous HRSS treatment could promote renal function recovery. To test this hypothesis, we monitored renal function at different time points after I/R injury. When compared with the I/R group, the long-term and short-term treatment groups showed significantly reduced renal function impairment, and the renal function was rapidly recovered. However, compared with the short-term treatment group, renal function recovery in the long-term treatment group was faster. In addition, we also unexpectedly found that serum Cr and BUN increased in the short-term treatment group after discontinuation of HRSS treatment (Figure 2). These results indicated that after renal I/R injury, continuous HRSS treatment exerted a significantly enhanced therapeutic effect and promoted more rapid renal function recovery than short-term HRSS treatment.

In order to determine the effect of HRSS treatment on inflammatory cytokine levels in kidney tissue, we measured TNF-α and IL-6 expression by ELISA. As shown in Figures 4C,D, inflammatory cytokines levels in the short-term and long-term treatment groups were significantly reduced at 24 h after I/R injury, when compared to the I/R group. However, after discontinuation of HRSS treatment, the inflammatory cytokine levels in the kidney tissues of the short-term treatment group were increased, while in the long-term treatment group these levels continuously decreased.

To further study the mechanism by which HRSS reduces renal cell apoptosis, Bcl-2 and Bax renal expressions were analyzed at 24 h after I/R injury by immunohistochemistry (Figure 5A). Bcl-2 expression in the HRSS treatment group was increased compared to the I/R group, while Bax expression was slightly decreased. However, these differences between the two groups were not statistically significant (P > 0.05; Figures 5B,C). We also compared the Bcl-2/Bax ratios in the I/R and hydrogen treatment groups, and we found that this ratio was significantly increased in the HRSS treatment group (P < 0.05; Figure 5D).

Because activation of caspase-3, caspase-8, and caspase-9 plays an important role in apoptosis induction, we analyzed their expression in kidney tissue. As shown in Figure 4E, caspase-3, caspase-8, and caspase-9 levels in the short-term and long-term treatment groups were significantly reduced compared to the I/R group (P < 0.01; Figure 4E).