Sphingosine-1-phosphate mitigates radiation-induced ovarian injury in rats by suppressing ferroptosis via activating the Nrf2-GCLC-GPX4 axis

Abstract

Background

Ovarian tissue is highly sensitive to ionizing radiation; even low-dose exposure can lead to premature ovarian failure and infertility. However, the molecular mechanisms underlying radiation-induced ovarian damage and potential protective interventions remain incompletely defined.

Methods

Using a rat model of abdominal irradiation, we investigated whether sphingosine-1-phosphate (S1P) protects against radiation-induced ovarian injury by modulating ferroptosis. Rats were pretreated with S1P, with or without the glutamate-cysteine ligase catalytic subunit (GCLC) inhibitor L-buthionine sulfoximine (BSO) or the ferroptosis inhibitor ferrostatin-1 (Fer-1). Ovarian histopathology, follicle counts, serum hormone levels, inflammatory and oxidative stress markers, and ferroptosis-related proteins were analyzed.

Results

S1P treatment significantly alleviated radiation-induced structural and functional ovarian damage by preserving ovarian architecture, increasing follicle number (notably primordial follicles), normalizing serum hormone levels, and reducing inflammatory cytokines. Mechanistically, S1P mitigated mitochondrial injury and oxidative stress by decreasing iron accumulation, lipid peroxidation, and reactive oxygen species (ROS) levels, while restoring glutathione (GSH), superoxide dismutase (SOD), and mitochondrial membrane potential. Mechanistically, S1P enhanced the expression of GCLC and glutathione peroxidase 4 (GPX4), suppressed transferrin receptor 1 (TfR1) expression, and transiently activated nuclear factor erythroid 2-related factor 2 (Nrf2) signaling. Pharmacological inhibition of GCLC by BSO abolished these protective effects, confirming the essential role of the Nrf2–GCLC–GPX4 axis in regulating ferroptosis and redox balance.

Conclusions

S1P protects against radiation-induced ovarian injury by suppressing ferroptosis through upregulation of GCLC and GPX4 downstream of Nrf2 activation. These findings identify S1P as a potential therapeutic agent for preserving ovarian function during radiotherapy and highlight the GCLC-GSH-GPX4 pathway as a critical target for ovarian radioprotection.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s13048-026-02192-w.

Introduction

Radiation therapy remains a crucial cornerstone in cancer treatment, with techniques such as whole-body irradiation or abdominal irradiation being widely used in clinical practice. However, the ovary, an organ that is crucial for women’s reproductive capacity and endocrine balance, is particularly vulnerable to ionizing radiation [1]. This vulnerability is of significant clinical concern, especially for young female cancer survivors. Epidemiological data indicate that cancer remains a leading cause of mortality among women under 40 years of age [2]. A substantial proportion of these patients require gonadotoxic treatments, including chemotherapy and/or radiotherapy, which pose a high risk of inducing premature ovarian failure (POF) [3]. The probability of POF following such treatments varies depending on treatment intensity and patient age but remains clinically significant, thus directly threatening future fertility and long-term hormonal health, thereby significantly impacting quality of life among survivors [4]. Exposure to radiation can severely damage ovarian function, accelerate reproductive aging, leading to adverse outcomes such as decreased fertility, infertility, or increased risks of pregnancy complications [5–7]. Thus, protecting ovarian function is not only a matter of preserving fertility but also a critical component of comprehensive survivorship care for a large and vulnerable patient population. This underscores the significant clinical burden and the urgent need for effective ovarian protection strategies. Although advancements in treatment technologies have minimized the general toxicity [8], managing ovarian damage, particularly resulting from even low-dose radiation, remains a major clinical challenge [9]. Indeed, exposure to 2 Gy of radiation reduces the follicle pool by approximately half, underscoring the sensitivity of follicles (the fundamental functional units of the ovary) to radiation insult [9]. Therefore, there is an urgent need to elucidate the molecular mechanisms underlying radiation-induced ovarian injury and to develop effective protective interventions.

Designing effective ovarian radioprotective agents requires a deep understanding of the molecular mechanisms of radiation-induced damage. In recent years, ferroptosis, as a regulated form of cell death, has emerged as an important mechanism in radiation-induced injury [10–12]. Ferroptosis is an iron-dependent form of cell death characterized by lipid peroxidation and the collapse of the antioxidant defense system, particularly the glutathione peroxidase 4 (GPX4)-glutathione (GSH) axis [13]. The biosynthesis of GSH depends on glutamate-cysteine ligase (GCL), whose catalytic subunit (GCLC) catalyzes the rate-limiting step of forming γ-glutamylcysteine from glutamate and cysteine. Thus, GCLC is essential for maintaining intracellular GSH levels and enabling GPX4 function [14]. When GSH synthesis is inhibited, GPX4 becomes inactive, leading to lipid peroxidation accumulation and an increase in ferroptosis-induced cell death [15]. In irradiated rat ovarian tissues, studies have shown that the expression of GPX4 and GCLC is downregulated, indicating that the inhibition of this pathway is related to radiation-induced ovarian injury [16]. Moreover, inhibiting ferroptosis (e.g., via ferrostatin or other inhibitors) has been shown to ameliorate radiation-induced ovarian damage, providing causal evidence for the involvement of ferroptosis [17]. Nevertheless, the upstream regulatory mechanisms that regulate the GCLC-GSH-GPX4 axis in the ovary under radiation remain underexplored.

Sphingosine-1-phosphate (S1P) is a bioactive sphingolipid that plays multifaceted roles in cell survival, proliferation, and stress responses [18, 19]. Notably, previous reports indicate that S1P can protect ovarian function in the context of chemotherapy and radiotherapy. For instance, S1P administration prevents follicular apoptosis in human ovarian xenograft models exposed to chemotherapeutic agents [20]. In irradiated rodents, S1P rescues the number of follicles and preserves ovarian morphology; transcriptomic analysis suggests its protective effect may involve mitochondrial-related genes [16]. Another study demonstrated that S1P can mitigate radiation-induced ovarian cell damage in vitro via inhibiting ferroptosis (an effect associated with upregulation of GPX4) [17]. However, most of these mechanistic studies were conducted in cell lines (e.g. ovarian granulosa KGN cells) or in limited settings; the in vivo effects and upstream regulation remain incompletely characterized. Notably, whether S1P can regulate GCLC (a key upstream regulator of GSH synthesis) in the ovaries after radiation exposure has not been explored.

Given that GCLC is located upstream of the glutathione synthesis pathway and is critical for GPX4 function, and since S1P has been shown to affect GPX4 under in vitro irradiation conditions, we hypothesize that S1P may protect ovaries from radiation-induced ferroptosis by upregulating GCLC, thereby enhancing GSH supply and maintaining GPX4 activity. To verify this hypothesis, we conducted an in vivo study in a rat model, combining abdominal irradiation with S1P treatment. We examined whether S1P could mitigate ovarian damage by inhibiting ferroptosis, and whether its radioprotective effect involves activation of the GCLC-GSH-GPX4 axis. This study fills a critical gap as previous work mainly focused on downstream effectors, such as GPX4 (especially in vitro), but the role of GCLC as an intermediary in the S1P protection mechanism in an in vivo ovarian environment has not yet been established. Ultimately, our goal is to elucidate the mechanistic basis of S1P-based ovarian radioprotection, which may offer novel targets for the protection of fertility in women undergoing radiotherapy.

Materials and methods

Animals and experimental design

Female Sprague-Dawley (SD) rats (6–8 weeks old; 180–220 g) were obtained from the Experimental Animal Center of Soochow University (Suzhou, China). Animals were housed under standard conditions (22 ± 2 °C, 12 h light/dark cycle, 50–60% humidity) with ad libitum access to food and water. All experimental procedures were approved by the Ethics Committee of Lianyungang Maternal and Child Health Hospital (Approval No. xm2023008) and conducted in accordance with institutional and national guidelines for the care and use of laboratory animals.

Following one week of acclimatization, a pilot dose-response study was performed to determine the optimal dose of S1P (Cat. No. 62510, Tocris Bioscience, USA) for protecting against radiation-induced ovarian follicle loss. Rats were randomly assigned to six groups (n = 5 per group): Sham, IR, IR + Vehicle, IR + S1P (0.5 mg/kg), IR + S1P (2.5 mg/kg), and IR + S1P (5 mg/kg). S1P was dissolved in sterile phosphate-buffered saline (PBS) and administered intraperitoneally once daily beginning 72 h before irradiation and continuing until sacrifice. Animals in irradiated groups received a single 2 Gy dose of 6-MV X-rays directed to the lower abdominal region encompassing both ovaries. Twenty-four hours after irradiation, rats were euthanized and ovaries were collected for histological analysis. Primordial follicle numbers were quantified as described in Sect.  2.3. The dose that provided the greatest preservation of primordial follicles without apparent systemic toxicity was selected for subsequent experiments. Based on these results, 2.5 mg/kg S1P was chosen for all further studies (Supplementary Fig. S1A).

In another set of experiments, rats were randomly assigned to six experimental groups (n = 10 per group); (1) control (vehicle only); (2) S1P (Sphingosine-1-Phosphate only); (3) IR (Irradiation only); (4) IR + S1P; (5) IR + Fer-1 (IR + Ferrostatin-1), and (6) IR + S1P + BSO (IR + S1P + L-Buthionine Sulfoximine) for the 24-hour post-irradiation assessment. To evaluate the persistence of S1P-mediated ovarian protection, an additional cohort of rats from the Control, IR, and IR + S1P groups (n = 5 per group) was sacrificed 7 days after irradiation for histological analysis. The estrous cycle was not monitored or synchronized prior to experimentation. However, the relatively large sample size combined with strict randomization was expected to ensure an even distribution of animals at different estrous stages across groups, thereby minimizing potential confounding effects on intergroup comparisons.

For irradiation, animals were anesthetized with 2% isoflurane and exposed to a single dose of 2 Gy 6-MV X-rays (Varian Medical Systems, Palo Alto, USA) directed to the lower abdominal region encompassing the ovaries, as described previously [21, 22].

For SIP treatment, S1P (Cat. No. 62510, Tocris Bioscience, USA) was dissolved in PBS and administered intraperitoneally at 2.5 mg/kg once daily for three consecutive days before irradiation and continued until sacrifice. Animals in the control and IR groups received an equal volume of vehicle (PBS) according to the same schedule. For ferroptosis inhibition, rats in the IR + Fer-1 group were injected intraperitoneally with Ferrostatin-1 (1 mg/kg; Cat. No. HY-100579, MedChemExpress, USA) daily, starting 72 h before irradiation and continuing until sacrifice. For GCLC inhibition, rats in the IR + S1P + BSO group received L-buthionine-(S, R)-sulfoximine (BSO; 10 mg/kg; Cat. No. HY-100031, MedChemExpress, USA) intraperitoneally on the same schedule as S1P.

At 24 h post-irradiation, all rats were euthanized under anesthesia (2% isoflurane). The ovaries were excised, blotted dry, weighed, and immediately processed for subsequent analyses (histology, biochemical assays, and molecular experiments).

Body weight and ovarian weight measurements

Rats were weighed individually using a calibrated digital balance (Sartorius, Germany) immediately before irradiation and again at sacrifice 24 h post-irradiation. After euthanasia, both ovaries were excised, cleared of surrounding fat and oviductal tissue, gently blotted to remove surface fluid, and weighed within 3 min of removal to obtain wet ovarian weight. The ratio of ovarian weight to body weight (mg/g) was calculated to be correct for body size variation. All measurements were performed by an investigator blinded to group allocation.

Histological analysis and ovarian follicle quantification

Following weighing, the ovaries from each rat were fixed in 4% paraformaldehyde (Sigma-Aldrich, USA) for 24 h at 4 °C, dehydrated through graded ethanol, cleared in xylene, and embedded in paraffin using a Leica tissue processor. Serial Sect. (5 μm thick) were prepared using a rotary microtome (Leica RM2235) and mounted onto glass slides for subsequent histological evaluation. To ensure systematic and unbiased follicle quantification while minimizing the risk of duplicate counting, every fifth section throughout the entire ovary was selected for analysis. Depending on ovarian size, approximately 8–12 sections per ovary were evaluated. This systematic random sampling approach has been previously validated to provide reliable estimates of follicle populations [23]. Sections were stained with hematoxylin and eosin (H&E) using a commercial kit (Cat. No. C0105S, Beyotime, China) according to the manufacturer’s instructions. Ovarian morphology was examined under a light microscope (Nikon Eclipse E200, Japan), and representative images were captured digitally. Follicles were counted only when a clearly visible oocyte nucleus was present to prevent duplication across adjacent sections. Based on established morphological criteria, follicles were classified as primordial (single layer of flattened granulosa cells), primary (single layer of cuboidal granulosa cells), secondary (two or more granulosa layers without antrum), or antral (presence of one or more antral cavities). The total number of each follicle type per ovary was estimated by multiplying the counted number by the inverse of the section sampling interval (×5). All quantifications were performed independently by two blind observers.

Immunohistochemical (IHC) staining

Immunohistochemical staining was performed to evaluate follicular structures and ferroptosis-related protein expression in ovarian tissues. Ovaries were fixed in 4% paraformaldehyde (Sigma-Aldrich, USA) for 24 h, dehydrated through graded ethanol (70–100%), cleared in xylene, and paraffin-embedded. Sections (4–5 μm) were deparaffinized, rehydrated, and subjected to antigen retrieval by heating in 10 mM sodium citrate buffer (pH 6.0) for 15 min at 100 °C. Endogenous peroxidase activity was quenched with 3% hydrogen peroxide for 15 min, followed by blocking with 5% normal goat serum (Beyotime Biotechnology, China) for 30 min at room temperature. Sections were incubated overnight at 4 °C with the following primary antibodies: anti-MSY-2 (1:200, Abcam, Cat. No. ab183879), anti-GCLC (1:200, Proteintech, Cat. No. 12601-1-AP), anti-GPX4 (1:200, Abcam, Cat. No. ab125066), and anti-TfR1 (1:200, Cell Signaling Technology, Cat. No. 13208). After washing three times with TBST (Tris-buffered saline with 0.1% Tween-20), the slides were incubated for 1 h at room temperature with HRP-conjugated secondary antibody (goat anti-rabbit or anti-mouse IgG, 1:500, Beyotime, China). Color development was achieved using 3,3′-diaminobenzidine (DAB) substrate (Beyotime, China) and counterstained with hematoxylin. Images were captured using a Nikon Eclipse E100 microscope (Tokyo, Japan). Ovarian tissues from at least five animals per group were analyzed, and representative fields were selected for imaging. The number of MSY-2-positive follicles was manually counted from three randomly selected sections per ovary, as previously described [16]. Quantitative assessment of GCLC, GPX4, and TfR1 expression was performed by measuring the integrated optical density (IOD) of positive staining using ImageJ software (NIH, USA).

Serum hormone analysis

Following anesthesia with 2% isoflurane, blood samples were collected from the left ventricle of each rat and immediately transferred to collection tubes. The blood was allowed to clot, and the serum was subsequently harvested by centrifugation at 10,000×g for 15 min at 4 ℃. Serum samples were stored at -80 ℃ until the assay was performed. The serum concentrations of reproductive hormones, including Follicle-Stimulating Hormone (FSH) (Cat. #D731057, Sangon Biotech, Shanghai, China), Estradiol (E2) (Cat. #D731284, Sangon Biotech, Shanghai, China), and Anti-Müllerian Hormone (AMH) (Cat. #D731257, Sangon Biotech, Shanghai, China), were measured using commercial ELISA kits according to the manufacturer’s instructions.

Ovarian inflammatory cytokine analysis

Ovarian tissue samples were homogenized in RIPA lysis buffer and centrifuged to obtain the supernatant. The levels of inflammatory cytokines, specifically tumor necrosis factor-alpha (TNF-α) and interleukin-1 beta (IL-1β), within the rat ovarian tissue homogenates were measured using respective commercial ELISA kits (Beyotime Biotechnology, Shanghai, China), following the manufacturer’s protocols. Absorbance was measured at 450 nm using a microplate reader (BioTek Epoch, USA).

Transmission electron microscopy (TEM)

Transmission electron microscopy was used to examine mitochondrial morphology in oocytes. Ovarian tissues from all groups were dissected and immediately cut into small fragments (~ 1 mm³). Samples were fixed in 2.5% glutaraldehyde (Servicebio, Wuhan, China) in 0.1 M phosphate buffer (pH 7.4) at 4 °C overnight. After three rinses in the same buffer, tissues were post-fixed with 1% osmium tetroxide (Ted Pella, USA) for 2 h, rinsed again, and dehydrated through a graded ethanol series (50%, 70%, 90%, 95%, and 100%) followed by acetone (90% and 100%). The specimens were then embedded in epoxy resin (Epon 812, SPI Supplies, USA) and polymerized at 60 °C for 48 h. Ultrathin sections (~ 70 nm) were obtained using an ultramicrotome (Leica EM UC7, Germany), mounted on copper grids, and double-stained with 3% uranyl acetate and lead citrate. Mitochondrial morphology in oocytes was observed using a transmission electron microscope (Hitachi HT7800, Japan) operated at 80 kV, and representative images were captured from at least five animals per group.

Detection of mitochondrial membrane potential

Mitochondrial membrane potential (ΔΨm) was evaluated using the JC-1 assay kit (Cat. No. C2006, Beyotime Biotechnology, Shanghai, China) following the manufacturer’s instructions. Fresh ovarian tissues were minced and gently homogenized in ice-cold PBS, then filtered through a 200-mesh nylon filter to obtain a single-cell suspension. Approximately 1 × 10⁵ to 6 × 10⁵ cells were resuspended in 0.5 mL of culture medium and incubated with JC-1 working solution at 37 °C for 20 min in the dark. After incubation, cells were washed twice with JC-1 buffer and observed under a microscope. In healthy mitochondria, JC-1 forms aggregates emitting red fluorescence (Ex/Em = 525/590 nm), whereas in depolarized mitochondria, it remains as monomers emitting green fluorescence (Ex/Em = 490/530 nm). The ratio of red-to-green fluorescence intensity was calculated to assess mitochondrial membrane potential, with a decrease indicating mitochondrial depolarization.

Iron detection assay

Total iron content in ovarian tissues was quantified using a Colorimetric Iron Assay Kit (Cat. No. ab83366, Abcam, UK) according to the manufacturer’s instructions. Briefly, approximately 20 mg of ovarian tissue was homogenized in the provided iron assay buffer on ice and centrifuged at 16,000 × g for 10 min at 4 °C. The supernatant was collected, and 100 µL of each sample was mixed with iron reducer solution to convert ferric (Fe³⁺) to ferrous (Fe²⁺) ions. Following incubation with the iron probe at room temperature for 30 min, absorbance was measured at 562 nm using a microplate reader (Infinite M200 PRO, Tecan, Switzerland). Iron concentration was calculated from a standard curve generated with known Fe²⁺ standards and normalized to total protein concentration, expressed as µmol/mg protein.

Measurement of oxidative stress indices

Oxidative stress in ovarian tissues was evaluated by determining the levels of malondialdehyde (MDA), superoxide dismutase (SOD), glutathione (GSH), and reactive oxygen species (ROS) using commercial assay kits (Beyotime Biotechnology, Shanghai, China). Briefly, ovarian samples were washed with cold PBS, homogenized on ice in the appropriate extraction buffer, and centrifuged at 12,000 × g for 10 min at 4 °C to obtain supernatants.

The MDA level was measured using a thiobarbituric acid reactive substances (TBARS) assay kit (Cat. No. S0131S) to evaluate lipid peroxidation, with absorbance recorded at 532 nm. SOD activity (Cat. No. S0101S) was determined based on inhibition of nitroblue tetrazolium (NBT) reduction and expressed as units per milligram of protein. GSH content (Cat. No. S0053) was quantified using a 5,5’-dithiobis-(2-nitrobenzoic acid) (DTNB) colorimetric method, and ROS levels (Cat. No. S0033S) were detected by fluorescence intensity of DCFH-DA (10 µM) using a microplate reader (Tecan Infinite M200, Switzerland) with excitation/emission wavelengths of 488/525 nm. All measurements were normalized to total protein concentrations determined by the BCA Protein Assay Kit (Beyotime Biotechnology, Shanghai, China), and each assay was performed in triplicate.

Western blot analysis

As described previously [24], total protein was extracted from ovarian tissues using RIPA lysis buffer (Cat. No. 89900, Thermo Fisher Scientific, USA) supplemented with protease and phosphatase inhibitor cocktails (Cat. No. P1045, Beyotime, China). Protein concentration was determined using the BCA Protein Assay Kit (Cat. No. P0012, Beyotime, China) following the manufacturer’s protocol. Equal amounts of protein (30 µg per lane) were separated on 10% SDS-PAGE gels and transferred onto nitrocellulose membranes (PALL Corporation, USA). Membranes were blocked with 5% non-fat milk in TBST for 2 h at room temperature and then incubated overnight at 4 °C with primary antibodies against TfR1 (1:1000, Cat. No. ab84036, Abcam, UK), Nrf2 (1:1000, Cat. No. ab62352, Abcam, UK), phospho-Nrf2 (Ser40) (1:1000, Cat. No. PA5-105674, Thermo Fisher, USA), GPX4 (1:2000, Cat. No. ab125066, Abcam, UK), GCLC (1:1000, Cat. No. 12601-1-AP, Proteintech, USA), and β-actin (1:5000, Cat. No. 20536-1-AP, Proteintech, USA) as loading control. After three washes with TBST, membranes were incubated with HRP-conjugated goat anti-rabbit IgG secondary antibody (1:5000, Cat. No. A0208, Beyotime, China) for 1 h at room temperature. Protein bands were visualized using an enhanced chemiluminescence reagent (Cat. No. P0018S, Beyotime, China) and imaged with the Bio-Rad ChemiDoc™ XRS+ Imaging System (Bio-Rad, Mississauga, Ontario, Canada). Band intensities were quantified using ImageJ software (NIH, USA) and normalized to β-actin expression. β-actin was verified to be unaffected by the treatments in our experimental conditions. A control group sample was included on each gel as an internal reference, and a lane loaded with protein loading buffer alone was used as a negative control to confirm signal specificity.

Statistical analysis

Statistical analyses were performed using SPSS version 25.0 (IBM Corp., Armonk, NY, USA) and GraphPad Prism version 10.0 (GraphPad Software, San Diego, CA, USA). Data distribution was assessed using the Shapiro-Wilk test (α = 0.1). Normally distributed data were expressed as mean ± standard deviation (SD). Multiple-group comparisons were analyzed by two-way analysis of variance (ANOVA) followed by Tukey’s post hoc test. The effect size was calculated using Cohen’s d, with |d| > 0.5 considered biologically meaningful. Sample size was estimated by power analysis using G*Power software (version 3.1; effect size = 1.2, power = 0.8, α = 0.05). All tests were two-sided, and statistical significance was defined as P < 0.05.

Results

S1P treatment attenuates radiation-induced ovarian injury in rats

To establish an appropriate dosing regimen for ovarian radioprotection, we first performed a pilot dose-escalation study evaluating three intraperitoneal doses of S1P (0.5, 2.5, and 5 mg/kg) in the rat irradiation model. The number of primordial follicles per ovary was used as the primary endpoint because it reflects the size of the ovarian reserve. As shown in Supplementary Fig. 1A, irradiation alone resulted in a marked reduction in primordial follicle numbers compared with the sham group. Administration of S1P at 0.5 mg/kg partially preserved the primordial follicle pool (118 ± 10 follicles/ovary). Increasing the dose to 2.5 mg/kg further improved follicle preservation (152 ± 11 follicles/ovary), yielding significantly higher counts than those observed with 0.5 mg/kg treatment and restoring follicle numbers to levels comparable with those of the sham group. No additional benefit was observed at 5 mg/kg (149 ± 13 follicles/ovary; n.s. versus 2.5 mg/kg). No mortality or obvious behavioral abnormalities were detected in any treatment group (data not shown). Based on these findings, 2.5 mg/kg was selected for all subsequent experiments.

Due to the extreme sensitivity of ovarian tissue to ionizing radiation and the rapid atrophy of follicles after radiation exposure [25], we first examined whether S1P could mitigate early ovarian injury. To this end, rats were pretreated with S1P for 3 days before receiving 2 Gy abdominal X-ray irradiation and then evaluated the ovaries 24 h post-irradiation. Results showed that S1P treatment alone did not significantly alter the body weight or wet ovarian weight compared to the control group, confirming it’s in vivo safety. In contrast, rats in the IR group exhibited a significant reduction in both parameters, an effect that was effectively rescued by S1P pre-treatment (Figs. 1A-B). The histological analysis using H&E staining revealed that the ovarian structures of the control group and the S1P group were normal. However, radiation exposure led to obvious pathological changes, including vascular congestion, edema, and hemorrhage, and S1P treatment effectively alleviated these symptoms (Fig. 1C). To further confirm the protective effect on oocytes, ovarian sections were stained for MSY-2, an oocyte-specific protein. MSY-2 expression was unchanged between the control and S1P groups but was significantly reduced following radiation exposure; notably, S1P treatment restored MSY-2 levels to near-normal intensity (Fig. 1D). Monitoring the dynamic changes in the number of follicles showed that the average number of ovarian follicles in the IR group was significantly lower (Fig. 1E). Specifically, the number of primordial follicles decreased sharply (approximately 55% in the IR group), but the IR + S1P group showed a significantly attenuated reduction, demonstrating that S1P has a protective effect on oocyte loss. Furthermore, ELISA showed that the expression of inflammatory cytokines, TNF-α and IL-1β in the IR group significantly increased, while S1P treatment significantly reduced this inflammatory response (Figs. 1F-G). Finally, regarding ovarian function, radiation exposure significantly reduced the serum levels of FSH, estradiol (E2), and AMH, while S1P pre-treatment successfully restored these endocrine markers (Figs. 1H-J). To determine whether the protective effects of S1P persisted beyond the acute 24 h phase, ovarian histology was additionally evaluated at 7 days post-irradiation in a separate cohort of rats. As shown in Figures S1B-C, the IR group continued to exhibit severe follicular depletion at day 7, with primordial follicle numbers reduced by approximately 52% compared with the control group. In contrast, the IR + S1P group retained significantly higher numbers of primordial and primary follicles, indicating sustained preservation of ovarian reserves. Together, these findings indicate that S1P protects against radiation-induced ovarian damage by preserving follicular reserve and oocyte integrity, suppressing inflammation, and maintaining ovarian endocrine function, with protective effects persisting beyond the acute phase of radiation exposure.

Fig. 1.

S1P treatment attenuates radiation-induced ovarian injury in rats. A Mean body weight of rats in each group. B Wet ovarian weight of rats in each group. C Representative HE staining images showing histopathological changes in ovarian tissues from each group. D IHC analysis was conducted to detect the expression of MYS2 in rat ovarian tissues. E Quantification of primordial, primary, secondary, and antral follicles per group. F ELISA-based measurement of proinflammatory cytokines TNF-α and IL-1β in ovarian tissues. Serum levels of FSH, estradiol, and AMH were measured by ELISA. Data are shown as the mean ± SD of n = 10 rats per group, and statistical analyses were performed using two-way analysis of variance (ANOVA) with Tukey’s post-hoc test. ***P < 0.001, ^#P < 0.05, ^###P < 0.001

S1P protects ovaries from radiation-induced ferroptosis

Given that ferroptosis plays a crucial role in radiation-induced ovarian injury [16, 17], we next explored whether S1P exerts radioprotective effects by inhibiting ferroptosis. TEM revealed significant mitochondrial changes among the groups. The oocytes in the control group had intact mitochondria with clearly distinguishable cristae, while the oocytes exposed to radiation showed mitochondrial swelling, cristae destruction, and outer membrane rupture. In the IR + S1P group, these ultrastructural changes were significantly reduced, indicating that S1P can maintain mitochondrial integrity under oxidative stress conditions (Fig. 2A). Consistent with this, the iron content in the ovarian tissues after radiation significantly increased, while S1P treatment reduced it (Fig. 2B). Oxidative stress assays further demonstrated elevated MDA levels and decreased SOD and GSH contents in the IR group, whereas S1P treatment reversed these alterations (Figs. 2C-E). Additionally, the ROS level in the radiation group significantly increased, while S1P treatment significantly reduced it (Fig. 2F). Consistent with the TEM findings, JC-1 staining revealed that radiation led to depolarization of mitochondria (an increase in the ratio of green/red fluorescence), while S1P restored ΔΨm toward normal levels (Fig. 2G). Collectively, these results show that S1P ameliorates radiation-induced ferroptosis by reducing iron overload, lipid peroxidation, and ROS accumulation, while maintaining mitochondrial structure and function.

Fig. 2.

S1P protects rat ovaries from radiation-induced ferroptosis. A Representative TEM images of mitochondrial morphology in ovarian tissues in each group. B Quantification of iron accumulation in ovarian tissues using iron assay kits. C-E Levels of MDA, SOD, and GSH were measured in ovarian tissues using biochemical kits. F Intracellular ROS levels in ovarian tissues were detected using flow cytometry. G JC-1 staining was used to evaluate mitochondrial membrane potential. Data are shown as the mean ± SD of n = 10 rats per group, and statistical analyses were performed using two-way analysis of variance (ANOVA) with Tukey’s post-hoc test. **P < 0.01, ***P < 0.001, ^#P < 0.05, ^###P < 0.001

S1P upregulates the Nrf2-GCLC-GPX4 antioxidant axis and limits TfR1 expression in radiation-exposed ovaries

Given that S1P alleviated radiation-induced ferroptosis, we next investigated the underlying molecular mechanism. The nuclear factor-erythroid 2-related factor 2 (Nrf2) pathway is a key regulator of oxidative stress responses and is involved in the protection against radiation damage [21, 22]. After activation, Nrf2 induces the expression of antioxidant and iron metabolism-related genes such as GCLC and GPX4, which maintain redox balance and suppress lipid peroxidation [26], while TfR1 promotes iron uptake and ferroptosis damage [27]. Since S1P has been reported to enhance the Nrf2 signaling [28], we examined whether the Nrf2-GCLC-GPX4-TfR1 axis mediates its radioprotective effects in the ovary. Western blot analyses revealed that radiation exposure markedly decreased the protein levels of GPX4 and GCLC compared to the control group, whereas S1P pre-treatment significantly restored their expression. In contrast, TfR1 expression was notably upregulated after irradiation, but it was decreased by S1P administration. Notably, the expression of the nuclear phosphorylated form of Nrf2 (p-Nrf2) significantly increased after radiation treatment alone. Importantly, compared to the control group, the p-Nrf2 level in the experimental mice treated with S1P remained at a higher level, and there was no significant difference compared to the IR group, indicating that S1P promotes the activation of the Nrf2 signaling pathway (Fig. 3A). IHC staining further confirmed that radiation exposure markedly reduced the expression of GCLC and GPX4 in the ovary and induced a strong upregulation of TfR1 compared to the control group. Specifically, S1P treatment partially restored GCLC and GPX4 levels and mitigated TfR1 overexpression (Fig. 3B). Collectively, these findings demonstrate that S1P enhances the antioxidant defense system of the ovary by activating the Nrf2-GCLC-GPX4 axis, while inhibiting TfR1-mediated iron overload, thereby contributing to its protective effect against radiation-induced oxidative and ferroptotic damage.

Fig. 3.

S1P activates the NRF2/GCLC pathway following radiation exposure. A Western blot analysis of GCLC, GPX4, TfR1, nuclear p-NRF2, and total NRF2 levels in ovarian tissues at 24 h post-irradiation in each group. B IHC analysis was conducted to detect the expression of GCLC, GPX4, and TfR1 in ovarian tissues in each group. Data are shown as the mean ± SD of n = 10 rats per group, and statistical analyses were performed using two-way analysis of variance (ANOVA) with Tukey’s post-hoc test. ***P < 0.001, ^###P < 0.001

GCLC mediates the radioprotective and anti-ferroptotic effects of S1P in ovaries

To confirm the involvement of the GCLC pathway in S1P-mediated radioprotection, we employed two complementary pharmacological approaches: administering ferroptosis inhibitor ferrostatin-1 (Fer-1), and the GCLC inhibitor butyrolactone sulfonate (BSO). Western blot analysis revealed that Fer-1 or S1P increased the levels of GCLC and GPX4, while inhibiting the expression of TfR1, whereas co-treatment with SIP and BSO reversed these changes and restored TfR1 upregulation. Importantly, the total and phosphorylated Nrf2 levels remained largely unchanged, indicating that S1P mainly functions through the GCLC-dependent mechanism downstream of Nrf2 activation (Fig. 4A). Immunohistochemistry also confirmed these protein changes, showing that the levels of GCLC and GPX4 increased after Fer-1 or S1P treatment and BSO treatment decreased them (Fig. 4B). Together, these findings demonstrate that S1P requires GCLC activity to maintain the expression of GPX4 and inhibit ferroptosis mediated by TfR1 in irradiated ovaries.

Fig. 4.

GCLC mediates the radioprotective and anti-ferroptotic effects of S1P in rat ovaries. A Western blot was conducted to detect the protein levels of GCLC, GPX4, TfR1, and nuclear p-NRF2 and total NRF2 levels in ovarian tissues in the indicated groups. B IHC staining of GCLC, GPX4, and TfR1 in ovarian tissues of each group. Data are shown as the mean ± SD of n = 10 rats per group, and statistical analyses were performed using two-way analysis of variance (ANOVA) with Tukey’s post-hoc test. ***P < 0.001, ^###P < 0.001

GCLC inhibition reverses the protective and anti-inflammatory effects of S1P in irradiated ovaries

To determine whether GCLC inhibition affects the physiological protection conferred by S1P, we next examined ovarian morphology, inflammation, and hormone secretion across treatment groups. Both S1P and Fer-1 significantly increased the body weight and ovarian weight of irradiated rats, while BSO co-treatment abolished these benefits (Figs. 5A-B). Histological assessment revealed that S1P and Fer-1 alleviated radiation-induced structural damage, whereas S1P + BSO treatment led to further deterioration of the structure (Fig. 5C). Follicle counts confirmed that S1P and Fer-1 restored the numbers of primordial, primary, and secondary follicles to normal, while BSO reversed these gains (Fig. 5D). Similarly, the elevated levels of TNF-α and IL-1β caused by radiation were inhibited in the S1P or Fer-1 treatment groups, but not in the S1P + BSO group (Figs. 5E-F). Hormonal analysis further revealed that the declines in FSH, E2, and AMH levels, caused by radiation were restored in the S1P and Fer-1 treatment groups, but were significantly reduced again when combined with BSO (Figs. 5G-I). These results collectively indicate that the radioprotective, anti-ferroptosis, and anti-inflammatory effects of S1P are achieved through the antioxidant regulatory mechanism mediated by GCLC and via the Nrf2-GCLC-GPX4 axis.

Fig. 5.

GCLC inhibition reverses the protective and anti-inflammatory effects of S1P in irradiated ovaries. A Rat body weight and B wet ovarian weight in each treatment group. C HE staining was used to analyze the pathological changes in ovarian tissues in each group. D Quantification of follicle subtypes (primordial, primary, secondary, antral) in each group. (E-F) ELISA quantification of TNF-α and IL-1β in ovarian tissues

(G-I) Serum levels of FSH, estradiol, and AMH were determined by ELISA. Data are shown as the mean ± SD of n = 10 rats per group, and statistical analyses were performed using two-way analysis of variance (ANOVA) with Tukey’s post-hoc test. ***P < 0.001, ^###P < 0.001

S1P mitigates radiation-induced ferroptosis by upregulating GCLC

To further clarify the mechanistic role of GCLC in S1P-mediated anti-ferroptosis protection, we examined the ferroptosis markers and mitochondrial integrity in irradiated rats treated with S1P, Fer-1, or S1P + BSO. Both Fer-1 and S1P significantly reduced iron accumulation and MDA levels in irradiated ovaries, whereas co-administration of BSO abrogated these effects (Figs. 6A-B). Similarly, S1P and Fer-1 restored SOD and GSH levels, but these levels dropped again after BSO treatment (Figs. 6C-D). Also, Fer-1 or S1P inhibited ROS production, while GCLC inhibition increased it (Fig. 6E). The ultrastructural observations using TEM further confirmed that Fer-1 and S1P mitigated mitochondrial swelling and cristae disruption caused by radiation, while BSO reversed these morphological improvements (Fig. 6F). Moreover, JC-1 staining showed that S1P and Fer-1 restored mitochondrial membrane potential but significantly reduced with BSO co-treatment (Fig. 6G). These findings collectively demonstrate that S1P upregulates GCLC to prevent radiation-induced ferroptosis, thereby maintaining redox balance, mitochondrial integrity, and ovarian function under radiation stress conditions.

Fig. 6.

S1P alleviates radiation-induced ferroptosis by upregulating GCLC. A Iron accumulation in ovarian tissues was measured using an iron assay kit. B-D Levels of MDA, SOD, and GSH in ovarian tissues were assessed using commercial kits. E Flow cytometry was conducted to detect the levels of ROS in rat ovarian tissues. F The TEM images were used to explore the morphological changes of mitochondria in ovarian tissues in each group. G JC-1 staining was performed to detect the mitochondrial membrane potential in ovarian tissues. Data are shown as the mean ± SD of n = 10 rats per group, and statistical analyses were performed using two-way analysis of variance (ANOVA) with Tukey’s post-hoc test. ***P < 0.001, ^###P < 0.001

Discussion

Radiation therapy remains a primary treatment modality for cancer, yet it inherently poses a risk of damaging healthy tissues, notably the highly sensitive ovary [1, 29–31]. Therefore, developing radioprotective strategies is crucial for reducing complications in reproductive and endocrine aspects for female cancer patients. In this study, we demonstrated that S1P confers significant protection against radiation-induced ovarian injury in rats. S1P treatment, after abdominal exposure to 2 Gy radiation, preserved follicular structure, alleviated inflammation, and restored endocrine function. Mechanistically, S1P suppressed ferroptosis by reducing iron accumulation, lipid peroxidation, and ROS generation, while maintaining mitochondrial integrity and membrane potential. At the molecular level, S1P enhanced the expression of antioxidant regulators GCLC and GPX4, attenuated radiation-induced upregulation of TfR1, and was associated with the transient activation of the Nrf2 signaling pathway, indicating that the enhancement of the Nrf2-GCLC-GPX4 axis underlies its anti-ferroptosis and radioprotective effects.

The female ovary is highly sensitive to radiation, which can prevent the regeneration of primary follicles and lead to premature ovarian failure and infertility [32]. Our findings align with previous reports showing that abdominal radiation can cause severe pathological changes, including vascular damage and hemorrhage, as well as a reduction in the number of follicles and a decline in ovarian reserve function. This is reflected in the decreased levels of FSH, E2, and AMH in the serum [1, 29]. S1P is a bioactive sphingolipid implicated in cell survival and proliferation [33–35]. Previously, it has been shown that it can protect germ cells from apoptosis and protect ovarian cells from radiation/cytotoxic damage [20, 36, 37]. Consistently, our data show that S1P treatment significantly alleviated radiation-induced pathological changes and restored the numbers of follicles numbers. Using the oocyte-specific marker MSY-2 to confirm and quantify the number of follicles further strengthens our conclusion that S1P could preserve ovarian reserve capacity. Additionally, S1P inhibited the increase in inflammatory cytokine levels (TNF-α and IL-1β) caused by radiation, highlighting its anti-inflammatory properties, which complement its direct cell-protective effect.

A growing body of evidence indicates that radiation-induced damage is closely linked to redox imbalance, leading to the accumulation of ROS and potentially triggering ferroptosis [38]. Our current research results strongly support this paradigm. We observed the key features of ferroptosis: mitochondrial damage (swelling, cristae destruction, loss of ΔΨm), iron accumulation, elevated lipid peroxidation levels, and the reduction of key antioxidant GSH and its regulating enzyme SOD. The ability of S1P to reverse these indicators, reducing ROS and MDA while increasing GSH and SOD, aligns with prior in vitro studies showing that S1P protects ovarian granulosa cells from ferroptosis [17, 39]. Our TEM analysis, which showed that S1P preserved mitochondrial integrity and ΔΨm, is consistent with the view in the literature that the radioprotective effects of S1P is closely related to maintaining mitochondrial function [16, 40, 41]. This suggests that S1P exerts its powerful antioxidant effect by possibly scavenging ROS and, more fundamentally, by promoting the glutathione biosynthesis (which is necessary for GPX4 activity) [42, 43].

To further elucidate the underlying molecular mechanism, we focused on the Nrf2 signaling pathway, a master regulator of cellular antioxidant defense that transcriptionally activates genes such as GCLC and GPX4 [44, 45]. The radiation-induced downregulation of GPX4 and GCLC observed in our study implies the collapse of the GSH-dependent anti-ferroptosis system, which is consistent with the initiation of ferroptosis [46–48]. The concurrent upregulation of TfR1 also shows the radiation-induced iron overload mechanism, which catalyzes lipid peroxidation via the Fenton reaction, which is consistent with the previous findings that TfR1 overexpression promotes ferroptosis [44]. S1P stimulation effectively restored the expression of both GPX4 and GCLC and suppressed TfR1. Interestingly, at 24 h post-radiation, the levels of Nrf2 and p-Nrf2 were upregulated, suggesting an initial compensatory defense mechanism [49], but this activation diminished by 72 h (data not shown), suggesting that sustained oxidative stress overwhelmed endogenous defense mechanisms. S1P maintained elevated Nrf2 levels at 24 h; however, the absence of a statistically significant difference compared with the radiation-alone group indicates that its sustained protection may rely less on persistent Nrf2 activation per se and more on reinforcement of key downstream effectors, particularly GCLC as suggested elsewhere [17].

Indeed, the central role of GCLC was confirmed by pharmacological inhibition experiments. Treatment with BSO, a specific GCLC inhibitor, completely abrogated the protective effects of S1P, negating improvements in ferroptosis-related markers (MDA, GSH, ROS, mitochondrial morphology) as well as physiological outcomes (follicle counts, organ weight, hormone levels). This functional evidence establishes GCLC as a non-redundant effector of S1P-mediated radioprotection. The sustained enhancement of the GCLC-GSH-GPX4 axis, in contrast to the more transient modulation of Nrf2 and TfR1, suggests that maintenance of glutathione synthesis and GPX4 activity constitutes the principal long-term mechanism by which S1P suppresses ferroptosis and preserves ovarian function.

Despite these promising findings, several limitations should be acknowledged. First, although systemic S1P administration conferred ovarian protection, the specific receptor subtypes (S1PR1-5) and downstream signaling pathways responsible for mediating these effects were not delineated. Given the pleiotropic roles of S1P in vascular tone regulation and cardiac electrophysiology, potential systemic side effects must be carefully considered. Second, the possibility that S1P might inadvertently protect tumor cells from radiation-induced ferroptosis warrants rigorous evaluation before clinical translation. Finally, as the current study was conducted in rats, species-specific metabolic and reproductive differences may limit direct extrapolation to humans. Future investigations should therefore incorporate long-term reproductive assessments and utilize genetic or pharmacological approaches to dissect receptor-specific mechanisms. Localized delivery strategies, such as intraovarian administration, or the development of receptor subtype-selective modulators may enhance ovarian specificity while minimizing systemic and potential onco-protective effects. Moreover, defining whether tumor cells and ovarian follicles depend on distinct ferroptosis regulatory nodes could facilitate the development of selectively protective strategies. Collectively, such efforts will be essential to establish S1P or its derivatives as clinically viable anti-ferroptotic radioprotectors in female cancer patients.

Conclusions

Our results suggest that S1P protects against radiation-induced ovarian injury by inhibiting ferroptosis and restoring redox homeostasis. The sustained protective effects observed at both 24 h and 7 days post-irradiation indicate that S1P not only suppresses the early initiation of ferroptotic damage but also preserves long-term follicular integrity. Mechanistically, S1P-mediated protection is associated with activation of the Nrf2-GCLC-GPX4 axis, enhanced GSH synthesis, and reduced iron-dependent lipid peroxidation. Moreover, the similar protective effects observed with ferrostatin-1 further support ferroptosis inhibition as a key mechanism underlying the radioprotective action of S1P. These findings provide new mechanistic insight into how sphingolipid signaling interacts with the iron death pathway during the maintenance of ovarian function after radiation exposure.

Authors' contributions

JZ: Conceptualization, Methodology, Software, Resources, Writing; FC: Conceptualization, Methodology, Software, Resources; DL: Conceptualization, Software, Validation; MW: Data Curation, Visualization; LT: Conceptualization, Validation; HZ: Conceptualization, Validation, Supervision, Writing- Reviewing and Editing, Funding acquisition. All authors have read the final manuscript and approved it.

Funding

This study was supported by the Project of State Key Laboratory of Radiation Medicine and Protection, Soochow University (No. GZK1202236).

Data availability

The datasets generated and/or analyzed during the current study are presented in the manuscript and can be requested from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

All experimental procedures involving animals were conducted in accordance with ARRIVE guidelines and approved by the Ethics Committee of Lianyungang Maternal and Child Health Hospital (Approval No. xm2023008). The study involved no human subjects.

Consent for publication

Not Applicable.

Competing interests

The authors declare no competing interests.