Adult, virgin, female Sprague Dawley rats, weighing 180–220 g, were obtained from the Laboratory Animal Center of the Academy of Military Medical Sciences (Beijing, China). The animals were housed at 20–22°C with a 12-h light/dark cycle and fed standard chow and water ad libitum. The procedures in this study were approved by the Animal Care and Use Committee of Tianjin Medical University (Tianjin, China) and performed in accordance with the guidelines for the use of experimental animals from the National Institutes of Health. Vaginal smears had been obtained daily for >10 days to manifest ≥2 sequential, normal, 5-day vaginal estrus cycles, in order to ensure the availability of experimental animals in this study

The hydrogen-rich saline was prepared as previously described (13). Hydrogen was dissolved in physiological saline for 4 h under 0.4 MPa pressure to a saturated level using a self-designed, hydrogen-rich, water-producing apparatus, which was stored under atmospheric pressure at 4°C in an aluminum bag with no dead volume. To ensure a concentration of >0.6 mmol/l, it was necessary to freshly prepare the hydrogen-rich saline every week. A needle-type hydrogen sensor (Unisense A/S, Aarhus, Denmark) was used to confirm the content of hydrogen in the saline (14).

A total of 240 animals were randomly divided into four groups (n=60 per group): Control (Con), control + hydrogen-rich saline (Con + H₂), cisplatin-induced ovarian injury (OI) and cisplatin-induced ovarian injury + hydrogen-rich saline (OI + H₂). The animal model was established using a previously described method (15) with little change. Cisplatin was diluted in saline immediately before use. In the OI and OI + H₂ groups, the rats received a single dose of cisplatin [5 mg/kg; the lethal dose-50 of cisplatin in rats is 7.4 mg/kg (16)] by intraperitoneal injection on the 1st day. On the 7th day, the rats received another dose of cisplatin in an identical manner, which established the rat models of cisplatin-induced ovarian damage. An identical dose of saline was injected in the Con and Con + H₂ group rats using the same method.

In the Con + H₂ and OI + H₂ groups, the rats were intraperitoneally injected with hydrogen-rich saline (10 ml/kg body weight) once a day between the 1st and the 14th day. Vaginal smears were obtained from each rat at 8:00 a.m. each day for ≥5 days after cisplatin injection to ensure the ovarian injury model was successful.

On the 14th, 28th and 42nd days (T₁, T₂ and T₃) after cisplatin injection, femoral vein blood was collected from the rats (n=10 per group) and centrifuged at 10,000 × g, 4°C for 10 min. At the same time, ovarian tissue was collected, and homogenates were prepared via centrifugation, using the aforementioned conditions. These sample were stored at −80°C until the estrogen (E₂), follicle-stimulating hormone (FSH), superoxide dismutase (SOD), catalase (CAT) and malondialdehyde (MDA) examination. At each time-point, another 10 rats (n=10 per group) were sacrificed for bilateral ovary removal; one was fixed in formalin for follicle-counting analysis, while the other was used for nuclear factor erythroid 2-related factor 2 (Nrf2) detection by western blotting.

Ovarian follicle populations were counted using previously described methods (15). The ovaries were fixed in 10% formalin for 6 h at room temperature, embedded in paraffin and sectioned at a 5-µm thickness. Following deparaffinization and rehydration, the sections were stained with hematoxylin and eosin. The oocyte-containing follicles in the developmental stage were classified, and the antral follicles were counted in every 12th section by two experienced pathologists. Follicle classification and counting was carried out according to the criteria of Oktay et al (17).

The serum E₂ and FSH levels were measured using ELISA kits from R&D Systems (Minneapolis, MN, USA). The assays were performed according to the manufacturer's instructions. The absorbance was read on a microplate reader (Denley Dragon Wellscan MK 3; Thermo Fisher Scientific, Vantaa, Finland), and the concentrations were calculated based on a standard curve. All standards and samples were run in duplicate.

To investigate the mechanism associated with the effects of hydrogen-rich saline, the levels of antioxidant enzymes and oxidation products in the serum and ovarian tissue were measured. The activities of SOD, CAT and MDA were measured using commercial kits purchased from Cayman Chemical Company (Ann Arbor, MI, USA). According to the manufacturer's instructions, total SOD activity was assayed at 450 nm and CAT at 540 nm. Spectrophotometric readings of SOD and CAT were performed using a DU-640B spectrophotometer (Beckman Coulter, Miami, FL, USA), while readings for MDA were obtained using a microplate reader (CA94089; Molecular Devices, Sunnyvale, CA, USA). All standards and samples were run in duplicate.

At each time-point, ovarian tissue was collected to lyse on ice in 100 µl radioimmunoprecipitation assay buffer [50 mmol/l Tris-HCl (pH 7.4), 150 mmol/l NaCl, 1 mmol/l phenylmethanesulfonyl fluoride, 1 mmol/l ethylenediaminetetraacetic acid, 1% Triton X-100, 0.5% sodium deoxycholate and 0.1% sodium dodecyl sulfate]. The lysates were cleared by centrifugation at 15,000 × g for 10 min, and the supernatant protein samples were denatured at 100°C for 5 min, separated on 10% acrylamide gels and then electrotransferred to polyvinylidene fluoride membranes. Following the blocking of the membranes with 5% non-fat dried milk in Tris buffer saline tween 20 buffer at room temperature for 2 h, primary rabbit polyclonal antibodies against Nrf2 (1:200; ab92946; Abcam, Cambridge, MA,USA) and β-actin (1:2,000; ab129348; Abcam) were added for incubation overnight at 4°C. The immunoblots were subsequently incubated with horseradish peroxidase-conjugated goat anti-rabbit IgG (ab175773; Abcam, Cambridge, MA,USA and detected using enhanced chemiluminescence reagent (Merck Millipore, Molsheim, France). Images of the immunoblots were visualized and captured using the Quantity One® quantitative gel analysis system (Bio-Rad, Tokyo, Japan). All western blot analyses were carried out ≥3 times.

Differences between the groups were analyzed using one-way analysis of variance followed by the Tukey comparison. Results are expressed as the mean ± standard deviation of ≥3 independent experiments, and P<0.05 was considered to indicate a statistically significant difference between groups.

Following the injection of two single doses of 5 mg/kg cisplatin in the rats, abnormal levels of the sex hormones appeared in the serum. Compared with group Con, no significant difference in the release of FSH was found in group Con + H₂ (P>0.05, Fig. 1); however, the release of FSH was increased and E₂ release was reduced in group OI (P<0.05, Fig. 1). The administration of hydrogen-rich saline (10 ml/kg body weight) for 2 weeks during cisplatin injection attenuated the FSH release and elevated the level of E₂ at T₁, T₂ and T₃ compared with group OI (P<0.05, Fig. 1).

All developmental stages of the follicles, and the structure of the ovarian cortex and medulla, could be clearly seen in group Con. Compared with group Con, few follicles were observed in group OI; furthermore, the group OI rats exhibited significant damage to the cortex, which could not be distinguished from the medulla. Compared with group OI, the development of the follicles in group OI + H₂ was much more regular; all developmental stages of the follicles could be seen, and the damage to the cortex was less severe (P<0.05, Fig. 2).

To determine the underlying mechanism of the effects observed, antioxidant enzymes and oxidation products were detected in the serum and ovarian tissue of the rats at the T₁, T₂ and T₃ time-points. As shown in Fig. 3, when compared with group Con, cisplatin induced an excessive release of MDA and inhibited the activity of SOD and CAT in the serum and ovarian tissue of group OI (P<0.05). Hydrogen-rich saline significantly reduced the levels of MDA and elevated the activity of SOD and CAT in the serum and ovarian tissue of the cisplatin-challenged rats (P<0.05). The results revealed that cisplatin led to oxidative stress by increasing the levels of oxidation products and attenuating the activity of antioxidant enzymes, which could be reversed by hydrogen-rich saline treatment.

Nrf2 is the key molecule of the Nrf2/antioxidant response element (ARE) signaling pathway, which is a relatively conservative and important endogenous antioxidative system (18). Cisplatin induced Nrf2 expression in the ovarian tissue of the rats between T₁ and T₃ (P<0.05, Fig. 4). Compared with group OI, hydrogen-rich saline treatment further increased the Nrf2 expression between T₁ and T₃. Based on the results, it was suggested that hydrogen-rich saline may regulate the Nrf2/ARE signaling pathway to protect cells from immoderate oxidative stress.