Vitamin E alleviates energy stress-induced injury in yak granulosa cells by blocking the AMPK/mTOR-mitophagy-ferroptosis axis Authors: Hecheng Zhang, Guorong YE, Linjiao Bai, Peiyan Du, Mengjiao Wang, Furong Feng, Hongliang Li, Xueli Ma, Jiamei Lu, Jiangfeng Fan Published: 2026 ABSTRACT BackgroundEnergy stress-induced dysfunction of granulosa cells (GCs) is a major etiological factor in diminished female reproductive performance. Although vitamin E affords cytoprotection to GCs, its specific mechanisms of action under energy stress conditions in the yak model remain unclear.ResultsThis study aimed to elucidate the pathways and cell fate decisions through which vitamin E alleviates energy stress-induced damage in yak GCs. Our results indicate that energy stress triggers a signaling cascade initiated by the AMPK-mTOR pathway, which functions as an upstream regulator for downstream events. Activation of this pathway promotes PINK1/Parkin-mediated mitophagy, leading to ferroptosis, characterized by the downregulation of SLC7A11 and GPX4 and the upregulation of ACSL4. This cascade ultimately drives the cells toward apoptosis, as evidenced by an increased Bax/Bcl-2 ratio and elevated levels of Cleaved-caspase-3, along with impaired intercellular communication due to downregulation of Cx43 and Cx37. Vitamin E intervention mitigated apoptosis and rescued the expression of gap junction proteins by intercepting this AMPK-mTOR-mitophagy-ferroptosis axis.ConclusionOur study suggests a mechanism by which vitamin E modulates GC fate via this pathway. These findings provide insight into ovarian follicular pathophysiology in yaks and may inform strategies targeted at reproductive disorders associated with energy metabolic dysregulation.Supplementary InformationThe online version contains supplementary material available at 10.1186/s13048-026-02128-4. Background The normal development of ovarian follicles is essential for maintaining female mammalian fertility. This process relies on the functional integrity of granulosa cells, adequate nutrient supply, and energy homeostasis underpinned by healthy mitochondria [1, 2]. Gap junction-mediated communication, particularly via Connexin 37 (Cx37) and Connexin 43 (Cx43), between granulosa cells and the oocyte is crucial for oocyte development and follicular survival [3]. However, follicular utilization efficiency in vivo is remarkably low, as the vast majority of follicles undergo atresia—a degenerative process—prior to ovulation, with only a small fraction progressing to maturity and successful ovulation [4]. Granulosa cell apoptosis is recognized as an event in follicular atresia [5, 6], and perturbations in nutrient and energy availability are key determinants of their fate [7, 8]. Notably, the granulosa cell layer itself remains avascular throughout follicular development until luteinization occurs following ovulation. Consequently, granulosa cells, particularly those in the inner layers of preantral and early antral follicles, rely on diffusion for nutrient and oxygen supply from the vascularized theca layer and follicular fluid, which may result in a relatively hypoxic and metabolically constrained microenvironment. Furthermore, systemic nutritional status, metabolic diseases, or impaired local ovarian perfusion can all contribute to fluctuations or insufficiency in the energy supply within the follicular niche [9, 10]. Therefore, we employed low-glucose culture as an in vitro model of energy stress, which effectively induces granulosa cell apoptosis to mimic these pathophysiological conditions. Although the link between energy stress and apoptosis is established, the precise intracellular signaling cascade, the crosstalk among distinct cell death pathways, and the mechanisms governing the progression from initiation to completion of these processes—ultimately deciding cellular fate—remain to be elucidated. Mitochondria serve as crucial energy hubs and a primary source of reactive oxygen species (ROS) within cells, with their functional integrity and quality control playing a role in cellular fate determination [11, 12]. Energy stress is highly prone to inducing mitochondrial dysfunction, which subsequently activates mitophagy—a quality control mechanism [13, 14]. However, mitophagy acts as a “double-edged sword”: moderate mitophagy serves as an essential quality control mechanism, maintaining cellular homeostasis, whereas excessive or dysregulated mitophagy may exacerbate mitochondrial damage and promote cell death [15–18]. Therefore, whether energy stress can induce mitophagy in granulosa cells and how this process affects cell fate are questions that warrant further investigation. Ferroptosis, a regulated form of cell death characterized by iron dependency and excessive accumulation of lipid peroxides, has been shown to induce substantial accumulation of ROS, disrupt mitochondrial membrane integrity, promote the release of cytochrome c into the cytoplasm, and activate Caspase-3, thereby initiating apoptosis [19]. Notably, studies indicate that oxidative stress and mitophagy are initiating factors for ferroptosis [20–24]. Within the context of energy stress in granulosa cells, it is highly plausible that these two processes are interconnected and jointly determine cellular fate. Recent studies have revealed complex interactions between mitophagy and ferroptosis. However, within the specific biological context of energy stress in granulosa cells, it remains an unresolved question whether mitophagy and ferroptosis function as independent events or constitute an ordered cascade pathway that ultimately determines cell survival or death. Vitamin E (VE), a fat-soluble compound that cannot be synthesized endogenously by animals, is not only an essential nutrient but also a potent antioxidant, is widely used in clinical practice. VE protects organisms through mechanisms such as scavenging free radicals, modulating inflammatory responses, and providing neuroprotection [25]. Moreover, VE and its derivatives are widely used in the clinical treatment or adjunctive therapy of various diseases, particularly in gynecology [26] and cardiovascular medicine [27]. Studies have confirmed that vitamin E can alleviate apoptosis induced by various stress conditions [28, 29]. Further research indicates that it primarily mitigates granulosa cell apoptosis by inhibiting oxidative stress and regulating mitochondrial function, thereby supporting healthy follicular development [30]. In the context of energy stress, where mitochondrial dysfunction readily occurs, investigating whether vitamin E can mitigate the resulting cellular damage and elucidating the specific mechanisms and molecular targets involved are of interest. The AMPK/mTOR signaling axis functions as a hub for cellular energy sensing [31]. Under nutrient stress conditions such as low glucose, the concurrent activation of AMPK and inhibition of mTOR coordinately orchestrate cellular metabolic adaptation [32]. Moreover, this pathway is an upstream regulator of both autophagy and mitophagy [33–35]. Integrating this evidence, we hypothesize that under energy stress, activation of the AMPK/mTOR axis acts as the initiating trigger, leading to mitochondrial dysfunction and excessive mitophagy. These events, in turn, promote the ferroptosis cascade, which culminates in granulosa cell apoptosis and functional impairment. Consequently, the protective effect of vitamin E likely involves intervention in this pathway. Yaks in their natural high-altitude habitat frequently experience seasonal nutritional stress, particularly during the long, cold winters when forage is scarce [36–38]. This nutritional deficiency leads to a state of energy stress at the cellular level, which is a recognized cause of reproductive disorders, including impaired follicular development and delayed puberty [36–38]. This study investigated this signaling axis under energy stress using a primary culture model of yak granulosa cells. Our results indicate that vitamin E significantly alleviated both apoptosis and the downregulation of connexin proteins (Cx37 and Cx43) induced by energy stress. Mechanistically, we elucidated an AMPK/mTOR–mitophagy–ferroptosis signaling axis that drives energy stress-induced granulosa cell death, and this cascade reflects a disruption of mitochondrial quality control. These findings provide insight into energy stress responses in reproductive cells and identify an AMPK/mTOR–mitophagy–ferroptosis regulatory axis. This work thereby provides a mechanistic rationale for mitigating follicular atresia through nutritional interventions like vitamin E or by therapeutically targeting this axis. Materials and methods Isolation and culture of primary yak granulosa cells (YGCs) Ovarian samples were collected from a slaughterhouse in Xining City, Qinghai Province, in strict compliance with national animal welfare regulations. The samples were obtained from adult yaks during the follicular phase. The ovaries were rinsed with 37 °C saline containing penicillin-streptomycin until no visible blood or odor remained. Healthy antral follicles with diameters of 3–8 mm were selected based on visual inspection for clear, vascularized appearance and absence of opacity. Granulosa cells were aspirated from these follicles using a 20-gauge needle, isolated from the follicular fluid via centrifugation, and subsequently cultured in DMEM/F-12 medium under conditions of 5% CO₂ at 37 °C. Cell treatment Prior to the experiment, cells were cultured for 6 h in serum-free DMEM/F12 medium. Of note, no exogenous gonadotropins were added during culture, as the presence of serum provides sufficient survival factors for primary granulosa cells, and we aimed to investigate the direct effects of energy stress without the confounding influence of hormonal stimulation. To confirm that the cultured cells retained their functional identity under these conditions, we examined the expression of follicle-stimulating hormone receptor (FSHR) by immunofluorescence staining, which showed that > 95% of cells remained FSHR-positive (Fig. 1A). Apoptosis was induced by culturing the cells in low-glucose DMEM medium containing 10% FBS and 1% streptomycin/penicillin for 12, 24, and 48 h. For pharmacological studies, cells were incubated for 12 h with or without the following agents: 1 µM Ferrostatin-1 (Fer-1, a ferroptosis inhibitor), 50 µM Chloroquine (CQ, an autophagy flux blocker), 100 nM Rapamycin (Rapa, an mTOR inhibitor), 500 nM Mitoquinone mesylate (MitoQ, a mitochondrial-targeted antioxidant), and 10 µM Compound C (Cc, an AMPK inhibitor). Fig. 1Vitamin E alleviates apoptosis and downregulation of gap junction proteins in granulosa cells (GCs) of yak ovarian follicles under energy stress. A Identification of yak granulosa cells by immunofluorescence staining of follicle-stimulating hormone receptor (FSHR). FSHR expression (red) was detected in the cytoplasm of cultured cells, with nuclei counterstained with DAPI (blue). Scale bar = 100 µm. B Cells were cultured in normal medium for 12, 24, and 48 h, and cell viability was assessed by CCK-8 assay. C-D Western blot analysis of apoptosis-related proteins (Bax, Bcl-2, Cleaved Caspase-3) and gap junction proteins (Cx43, Cx37) in GCs treated with normal medium for 12, 24, or 48 h. E GCs were exposed to low glucose for 12, 24, or 48 h, and cell viability was assessed by CCK-8 assay. F-G Western blot analysis of apoptosis-related proteins (Bax, Bcl-2, Cleaved Caspase-3) and gap junction proteins (Cx43, Cx37) in GCs treated with low glucose for 12, 24, or 48 h. H Cell viability assessed by CCK-8 assay in GCs treated with Vitamin E (VE) at 25, 50, or 100 µM. I-J Western blot analysis of apoptosis-related proteins (Bax, Bcl-2, Cleaved Caspase-3) and gap junction proteins (Cx43, Cx37) in GCs treated with low glucose for 24 h in the presence or absence of VE (25, 50, or 100 µM). K Representative TUNEL staining images of GCs treated with low glucose (24 h) or low glucose plus VE (50 µM) for 24 h. L Estradiol levels in the culture medium were measured by ELISA in granulosa cells treated with low glucose (24 h) or low glucose plus VE (50 µM) for 24 h. Scale bar = 100 µm. Data are presented as mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001; ns, not significant Vitamin E alleviates apoptosis and downregulation of gap junction proteins in granulosa cells (GCs) of yak ovarian follicles under energy stress. A Identification of yak granulosa cells by immunofluorescence staining of follicle-stimulating hormone receptor (FSHR). FSHR expression (red) was detected in the cytoplasm of cultured cells, with nuclei counterstained with DAPI (blue). Scale bar = 100 µm. B Cells were cultured in normal medium for 12, 24, and 48 h, and cell viability was assessed by CCK-8 assay. C-D Western blot analysis of apoptosis-related proteins (Bax, Bcl-2, Cleaved Caspase-3) and gap junction proteins (Cx43, Cx37) in GCs treated with normal medium for 12, 24, or 48 h. E GCs were exposed to low glucose for 12, 24, or 48 h, and cell viability was assessed by CCK-8 assay. F-G Western blot analysis of apoptosis-related proteins (Bax, Bcl-2, Cleaved Caspase-3) and gap junction proteins (Cx43, Cx37) in GCs treated with low glucose for 12, 24, or 48 h. H Cell viability assessed by CCK-8 assay in GCs treated with Vitamin E (VE) at 25, 50, or 100 µM. I-J Western blot analysis of apoptosis-related proteins (Bax, Bcl-2, Cleaved Caspase-3) and gap junction proteins (Cx43, Cx37) in GCs treated with low glucose for 24 h in the presence or absence of VE (25, 50, or 100 µM). K Representative TUNEL staining images of GCs treated with low glucose (24 h) or low glucose plus VE (50 µM) for 24 h. L Estradiol levels in the culture medium were measured by ELISA in granulosa cells treated with low glucose (24 h) or low glucose plus VE (50 µM) for 24 h. Scale bar = 100 µm. Data are presented as mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001; ns, not significant Western blotting Total cellular proteins were extracted according to the instructions of RIPA Tissue Lysate (Transgenic, Beijing, China). The protein was separated by SDS-PAGE, transferred to PVDF and blocked with 5% skim milk, and then specifically bound to primary antibody Bcl-2(1:1000, T40056, Abmart), Bax(1:1000, T40051, Abmart), Cleaved-Caspase3(1:1000, TA7022, Abmart), Connexin 37(1: 500, PA3231, Abmart), Connexin 43(1:500, PA1471, Abmart), LC3 (1:500, 14600-1-AP, Proteintech), p62 (1:1000, TA5384, Abmart), PINK1 (1:1000, PK05715, Abmart), Parkin (1:1000, PA4767, Abmart) Phospho-AMPK(1:1000, TA3423, Abmart), AMPK(1:1000, T55326, Abmart), Phospho-mTOR(1:1000, T56571, Abmart), mTOR(1:1000, T55306, Abmart), SLC7A11(1:1000, 26864-1-AP, Proteintech), GPX4(1:1000, 67763-1-Ig, Proteintech), ACSL4(1:3000, 2241-1-AP, Proteintech) and β-actin (1:3,000 dilution, bs-0061R, Bioss). After washing, the secondary antibody was incubated and a chemiluminescent solution (ECL Luminescent Solution, Abcam, Cambridge, UK) was added dropwise. Imaging and semi-quantitative protein expression were performed using GE AI600 imaging system and Image J 1.51j8 software. Cell viability assay The GC cell viability was determined through the Cell Counting Kit-8 (CCK8, IV08–100T, Invigentech, California, USA) assay, in which the tetrazolium salt (WST-8) is reduced by dehydrogenase activities in viable cells to generate a yellow water-soluble formazan dye. The color intensity is therefore directly proportional to the number of living cells in culture. The experimental procedures were carried out as per the manufacturer’s directions. Briefly, the GC cells were seeded in 96-well plates and grown to 90% confluency for 2 days. After the indicated treatments were completed, the CCK-8 assay reagent (10 µl) was added to each well containing 100 µl medium and incubated in the dark for 2 h at 37 ℃. The absorbance at 450 nm was measured to assess formazan formation. Apoptosis analysis Apoptosis in tissues and cells was analyzed using the terminal deoxynucleotidyl transferase (TdT) dUTP nickend labeling (TUNEL) assay. In short, after tissue and cell treatment, the TUNEL mixture (KeyGEN, Nanjing, China) was added to the cells, and the cells were incubated for 60 min. DAPI was added for reverse staining. Visualization and image acquisition was performed through fluorescence microscopy (Apexbio, Revolve Omega, Suzhou, China). Measurement of mitochondrial membrane potential (MMP) The MMP was determined using the JC-1 Mitochon-drial Membrane Potential Assay Kit (Beyotime, Shanghai, China). After collection, the cells were resuspended in 0.5 mL of JC-1 working solution at a density of 2 × 105 cells per sample and incubated at 37 °C for 20 min. Images were obtained with a confocal microscope (Olympus, Tokyo, Japan). ATP levels GCs were plated at a density of 1.0 × 105 cells per well in 6-well plates and incubated overnight. Following treatment, the cells were harvested and lysed for ATP measurement. ATP levels were assessed using a commercial ATP Assay Kit (Beyotime, Shanghai, China) and detected with a multifunction microplate reader (PerkinElmer, Wal-tham, MA, USA). Mitochondrial staining GCs were plated in confocal dishes at a density of 1.0 × 105 cells per well and cultured overnight. Following treatment, Mito-Tracker Red 633 (Beyotime, Shanghai, China) was added, and the cells were incubated with the dye at 37 °C for 30 min in darkness. Detection of Ad-mCherry-GFP-LC3B GCs were labeled with Ad-mCherry-GFP-LC3B for 24 h. After treatment of the cells, the cells were fixed with paraformaldehyde, and the nucleus was stained with DAPI and blocked with an anti-fluorescence quencher. Autophagy flux was assessed by imaging using laser confocal fluorescence microscopy. Detection of Lipid Peroxidation 1 ml of BODIPY 581/591 C11 working solution was added to each well. Cells were incubated at 37 °C for 10 min in a cell culture incubator. After incubation, the supernatant was removed, and the cells were washed twice with PBS. Subsequently, 2 ml of PBS was added, and the samples were observed under a fluorescence microscope. Statistical analysis Statistical analysis was performed using GraphPad Prism 9.5. Data are presented as mean ± standard deviation (SD) from at least three independent experiments (n = 3). Comparisons between two groups were analyzed by unpaired Student’s t-test, while multiple comparisons among three or more groups were analyzed by one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test. Statistical significance was set at *p < 0.05, **p < 0.01, and ***p < 0.001; ns, not significant. Results Vitamin E alleviates apoptosis and downregulation of gap junction proteins in granulosa cells of yak follicles under energy stress To confirm the identity of isolated yak granulosa cells, immunofluorescence staining for follicle-stimulating hormone receptor (FSHR) was performed. FSHR expression was detected in the cytoplasm of cultured cells, confirming their granulosa cell identity(Fig. 1A). To exclude the potential confounding effect of culture duration, yak granulosa cells were cultured under normal conditions for 12, 24, and 48 h, and cell viability was assessed by CCK-8 assay (Fig. 1B). Using the 0 h time point as the reference, cell viability exhibited a slight but non-significant decrease at 12 h. In contrast, a slight but significant increase was detected at both 24 h and 48 h. Western blotting analysis showed that the expression levels of Bcl-2, Bax, Bax/Bcl-2 ratio, Cx43, and Cx37 remained unchanged across all time points, with no significant differences detected (Fig. 1C-D). CCK-8 assay showed that exposure to low-glucose medium for 12, 24, and 48 h significantly reduced the viability of granulosa cells (GCs) in a time-dependent manner (Fig. 1E). Western blot analysis revealed that under low-glucose conditions across different time points, the ratio of pro-apoptotic protein Bax to anti-apoptotic protein Bcl-2 increased. This was accompanied by upregulation of Cleaved Caspase-3 and significant downregulation of the gap junction proteins Cx37 and Cx43 (Fig. 1F-G). Interestingly, as shown by CCK-8 results, treatment with VE at different concentrations effectively rescued the low-glucose-induced reduction in cell viability (Fig. 1H). Consistently, Western blot analysis indicates that, compared to the low-glucose group, VE treatment decreased the Bax/Bcl-2 ratio, downregulated Cleaved Caspase-3 expression, and significantly increased the levels of Cx37 and Cx43 (Fig. 1I-J). These protein-level findings were in agreement with the reduction in apoptotic cells observed by TUNEL staining (Fig. 1K). ELISA was performed to measure estradiol (E2) levels in yak granulosa cells under different culture conditions. Low glucose treatment significantly decreased estradiol production compared with the control group, whereas co-treatment with vitamin E (VE) significantly restored estradiol levels, reversing the inhibitory effect induced by low glucose (Fig. 1L). Taken together, these results indicate that vitamin E protects yak granulosa cells from low glucose-induced apoptosis, gap junction dysfunction, and steroidogenic impairment. Energy stress-induced mitochondrial dysfunction and mitophagy activation in ovarian granulosa cells To assess the effects of energy stress on mitochondrial function in granulosa cells, we measured relevant indicators. Low-glucose treatment significantly reduced the mitochondrial membrane potential (Fig. 2A) and concurrently decreased intracellular ATP levels (Fig. 2B), indicating impaired mitochondrial function. These effects were significantly reversed by co-treatment with vitamin E (VE), which restored both mitochondrial membrane potential and ATP levels. Western blot analysis showed that low-glucose treatment significantly increased the level of LC3-II but decreased that of p62. The expression of mitophagy proteins, PINK1 and Parkin, was also markedly upregulated (Fig. 2C-D). Using Ad-mCherry-GFP-LC3B and MitoTracker for colocalization analysis, we observed that LC3 protein localized to damaged mitochondria (Fig. 2E), which provided direct evidence for mitophagy induction. Fig. 2Energy stress induces mitochondrial dysfunction and activates mitophagy in ovarian granulosa cells. A Mitochondrial membrane potential was assessed by JC-1 staining in granulosa cells treated with low glucose or low glucose plus VE (50 µM) for 24 h. CCCP, a mitochondrial uncoupler, was used as a positive control to disrupt mitochondrial membrane potential. Scale bar = 100 μm. B ATP content was measured in granulosa cells treated with low glucose or low glucose plus VE (50 µM) for 24 h. C-D Western blot analysis showing the expression levels of autophagy- and mitophagy-related proteins (LC3-II, p62, PINK1, Parkin) in granulosa cells treated with low glucose for 24 h. E Representative images showing colocalization of mitochondria (MitoTracker Deep Red 633, red) with autophagosomes (Ad-mCherry-GFP-LC3B, green) in granulosa cells treated with low glucose for 24 h. Scale bar = 50 μm. Data are presented as mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001; ns, not significant Energy stress induces mitochondrial dysfunction and activates mitophagy in ovarian granulosa cells. A Mitochondrial membrane potential was assessed by JC-1 staining in granulosa cells treated with low glucose or low glucose plus VE (50 µM) for 24 h. CCCP, a mitochondrial uncoupler, was used as a positive control to disrupt mitochondrial membrane potential. Scale bar = 100 μm. B ATP content was measured in granulosa cells treated with low glucose or low glucose plus VE (50 µM) for 24 h. C-D Western blot analysis showing the expression levels of autophagy- and mitophagy-related proteins (LC3-II, p62, PINK1, Parkin) in granulosa cells treated with low glucose for 24 h. E Representative images showing colocalization of mitochondria (MitoTracker Deep Red 633, red) with autophagosomes (Ad-mCherry-GFP-LC3B, green) in granulosa cells treated with low glucose for 24 h. Scale bar = 50 μm. Data are presented as mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001; ns, not significant Vitamin E alleviates energy stress-induced mitophagy and apoptosis in ovarian granulosa cells by mitigating mitochondrial dysfunction The Western blot results showed that VE treatment, following low-glucose exposure, significantly downregulated the expression of LC3-II and upregulated p62. Concurrently, the expression of mitophagy proteins, PINK1 and Parkin, was also decreased (Fig. 3A, B). The Ad-mCherry-GFP-LC3 reporter system confirmed that low-glucose treatment induced complete mitophagic flux (Fig. 3C). Notably, the mitochondrial-targeted antioxidant MitoQ produced protective effects similar to those of VE (Fig. 3D-E). To investigate the functional role of this autophagic process, we inhibited autophagosome degradation using Chloroquine (CQ). CQ treatment alleviated cellular injury, whereas treatment with Rapamycin (Rapa) exacerbated it (Fig. 3F-H). This finding indicates that under energy stress, the successful completion of the autophagic process, rather than its mere initiation, is involved in determining cell fate. It further suggests that the mitophagy induced by low glucose in this context is an excessive and detrimental process. Fig. 3Vitamin E alleviates energy stress-induced mitophagy and apoptosis in ovarian granulosa cells by mitigating mitochondrial dysfunction. A-B Western blot analysis of LC3-II, p62, PINK1, and Parkin expression in GCs treated with low glucose (24 h) alone or in combination with VE (25, 50, or 100 µM). C Representative images of autophagosome formation assessed using the Ad-mCherry-GFP-LC3B reporter in GCs treated with low glucose (24 h) with or without VE (50 µM). Scale bar = 10 μm. D-E Western blot analysis of LC3-II, p62, PINK1, and Parkin expression in GCs treated with low glucose (24 h) alone, low glucose plus VE (50 µM), or low glucose plus the mitochondrial antioxidant MitoQ (500 nM). F-H Western blot analysis of proteins involved in mitophagy (LC3-II, p62, PINK1, Parkin), apoptosis (Bax, Bcl-2, Cleaved Caspase-3), and gap junction communication (Cx43,Cx37) in GCs treated under the following conditions for 24 h: low glucose (LG), LG + VE (50 µM), LG + Chloroquine (CQ; 50 µM), and LG + VE + Rapamycin (Rapa; 100 nM). Data are presented as mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001; ns, not significant Vitamin E alleviates energy stress-induced mitophagy and apoptosis in ovarian granulosa cells by mitigating mitochondrial dysfunction. A-B Western blot analysis of LC3-II, p62, PINK1, and Parkin expression in GCs treated with low glucose (24 h) alone or in combination with VE (25, 50, or 100 µM). C Representative images of autophagosome formation assessed using the Ad-mCherry-GFP-LC3B reporter in GCs treated with low glucose (24 h) with or without VE (50 µM). Scale bar = 10 μm. D-E Western blot analysis of LC3-II, p62, PINK1, and Parkin expression in GCs treated with low glucose (24 h) alone, low glucose plus VE (50 µM), or low glucose plus the mitochondrial antioxidant MitoQ (500 nM). F-H Western blot analysis of proteins involved in mitophagy (LC3-II, p62, PINK1, Parkin), apoptosis (Bax, Bcl-2, Cleaved Caspase-3), and gap junction communication (Cx43,Cx37) in GCs treated under the following conditions for 24 h: low glucose (LG), LG + VE (50 µM), LG + Chloroquine (CQ; 50 µM), and LG + VE + Rapamycin (Rapa; 100 nM). Data are presented as mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001; ns, not significant Vitamin E protects granulosa cells from energy stress-induced apoptosis and downregulation of gap junction proteins by suppressing AMPK/mTOR signaling-mediated mitophagy The AMPK/mTOR pathway is a regulator of cellular energy stress. We next investigated whether it plays a decisive role in determining cell fate in our model. Low-glucose treatment significantly activated AMPK (increased p-AMPK) and inhibited mTOR (decreased p-mTOR) (Fig. 4A-B). Both VE and MitoQ treatment effectively normalized these changes by reducing p-AMPK and restoring p-mTOR levels (Fig. 4C-D), indicating that these effects are mediated, at least in part, through this pathway. To define the necessity and hierarchical position of AMPK/mTOR signaling in VE’s action, we used the AMPK-specific inhibitor Compound C. The results indicate that AMPK inhibition with Compound C mimicked the protective effects of VE. Specifically, it reduced the levels of mitophagy markers (LC3-II, PINK1, Parkin), increased p62 accumulation, suppressed apoptosis (as indicated by a decreased Bax/Bcl-2 ratio, reduced Cleaved Caspase-3 expression, and fewer TUNEL-positive cells), and restored the expression of gap junction proteins (Cx43, Cx37) (Fig. 4E-G). These results suggest that AMPK activation acts as an upstream driver of the downstream pathological cascade induced by energy stress, and that VE inhibits this AMPK activation and restores mTOR activity. Fig. 4Vitamin E protects against energy stress-induced granulosa cell apoptosis and downregulation of gap junction proteins by inhibiting AMPK/mTOR-mediated mitophagy. A-B Western blot analysis of AMPK, phospho-AMPK (p-AMPK), mTOR, and phospho-mTOR (p-mTOR) in GCs treated with low glucose (LG) for 24 h. C-D Western blot analysis of AMPK, p-AMPK, mTOR, and p-mTOR in GCs treated under the following conditions for 24 h: control, LG, LG + Vitamin E (VE; 50 µM), and LG + MitoQ (500 nM). E, G Western blot analysis of proteins involved in mitophagy (LC3-II, p62, PINK1, Parkin), apoptosis (Bax, Bcl-2, Cleaved Caspase-3), and gap junctions (Cx43, Cx37) in GCs treated under the following conditions for 24 h: control, LG, LG + VE (50 µM), and LG + Compound C (10 µM). (F) Representative TUNEL staining images of GCs treated for 24 h under the indicated conditions: control, LG, LG + VE (50 µM), and LG + Compound C (10 µM). Scale bar = 100 μm. Data are presented as mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001; ns, not significant Vitamin E protects against energy stress-induced granulosa cell apoptosis and downregulation of gap junction proteins by inhibiting AMPK/mTOR-mediated mitophagy. A-B Western blot analysis of AMPK, phospho-AMPK (p-AMPK), mTOR, and phospho-mTOR (p-mTOR) in GCs treated with low glucose (LG) for 24 h. C-D Western blot analysis of AMPK, p-AMPK, mTOR, and p-mTOR in GCs treated under the following conditions for 24 h: control, LG, LG + Vitamin E (VE; 50 µM), and LG + MitoQ (500 nM). E, G Western blot analysis of proteins involved in mitophagy (LC3-II, p62, PINK1, Parkin), apoptosis (Bax, Bcl-2, Cleaved Caspase-3), and gap junctions (Cx43, Cx37) in GCs treated under the following conditions for 24 h: control, LG, LG + VE (50 µM), and LG + Compound C (10 µM). (F) Representative TUNEL staining images of GCs treated for 24 h under the indicated conditions: control, LG, LG + VE (50 µM), and LG + Compound C (10 µM). Scale bar = 100 μm. Data are presented as mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001; ns, not significant Vitamin E prevents ferroptosis by disrupting the AMPK/mTOR-mediated mitophagy axis Having confirmed that low glucose induces mitophagy, we next investigated whether it concurrently triggers ferroptosis, another form of programmed cell death. Western blot analysis showed that low-glucose treatment significantly downregulated the ferroptosis inhibitors SLC7A11 and GPX4, while upregulating the pro-ferroptotic protein ACSL4 (Fig. 5A-B). Both VE and MitoQ treatment effectively reversed these changes (Fig. 5C-D), indicating their ability to inhibit ferroptosis. To delineate the relationship between autophagy and ferroptosis, we employed specific modulators. Notably, both Chloroquine (CQ, an inhibitor of autophagosome-lysosome fusion) and Compound C (an AMPK inhibitor) alleviated ferroptosis, with CQ exhibiting a stronger effect than Compound C. Conversely, treatment with Rapamycin (Rapa) exacerbated ferroptosis (Fig. 5E-F). This finding, corroborated by a lipid peroxidation assay (Fig. 5G), indicates that the completion of autophagic flux is required to drive ferroptosis. To investigate the role of ferroptosis in the damage cascade, we used the specific ferroptosis inhibitor Ferrostatin-1 (Fer-1). Fer-1 not only normalized ferroptosis markers but also reduced apoptosis and restored gap junction protein expression, exhibiting protection comparable to VE (Fig. 5H-K). Collectively, these results suggest a pathological axis wherein energy stress activates AMPK/mTOR signaling, leading to excessive mitophagy, which in turn drives ferroptosis and culminates in granulosa cell apoptosis and dysfunction. Fig. 5Vitamin E prevents ferroptosis by disrupting the AMPK/mTOR-mediated mitophagy axis. A-B Western blot analysis of ferroptosis regulators (SLC7A11, GPX4, ACSL4) in granulosa cells (GCs) treated with low glucose (LG) for 24 h. C-D Western blot analysis of SLC7A11, GPX4, and ACSL4 in GCs treated under the following conditions for 24 h: control, LG, LG + Vitamin E (VE; 50 µM), and LG + MitoQ (500 nM). E-F Western blot analysis of SLC7A11, GPX4, and ACSL4 in GCs treated under the following conditions for 24 h: control, LG, LG + Vitamin E (VE; 50 µM), LG + Chloroquine (CQ; 50 µM), LG + VE + Rapamycin (Rapa; 100 nM), and LG + Compound C (10 µM). G Representative fluorescence images showing lipid peroxidation (C11-BODIPY 581/591) in GCs treated as indicated in panel. Scale bar = 100 μm. H-J Western blot analysis of ferroptosis markers (SLC7A11, GPX4, ACSL4), apoptosis markers (Bax, Bcl-2, Cleaved Caspase-3), and gap junction proteins (Cx43, Cx37) in GCs treated under the following conditions for 24 h: control, LG, LG + VE (50 µM), and LG + Ferrostatin-1 (Fer-1; 1 µM). (K) Representative TUNEL staining images of GCs treated for 24 h under the indicated conditions: control, LG, LG + VE (50 µM), and LG + Fer-1 (1 µM). Scale bar = 100 μm. Data are presented as mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001; ns, not significant Vitamin E prevents ferroptosis by disrupting the AMPK/mTOR-mediated mitophagy axis. A-B Western blot analysis of ferroptosis regulators (SLC7A11, GPX4, ACSL4) in granulosa cells (GCs) treated with low glucose (LG) for 24 h. C-D Western blot analysis of SLC7A11, GPX4, and ACSL4 in GCs treated under the following conditions for 24 h: control, LG, LG + Vitamin E (VE; 50 µM), and LG + MitoQ (500 nM). E-F Western blot analysis of SLC7A11, GPX4, and ACSL4 in GCs treated under the following conditions for 24 h: control, LG, LG + Vitamin E (VE; 50 µM), LG + Chloroquine (CQ; 50 µM), LG + VE + Rapamycin (Rapa; 100 nM), and LG + Compound C (10 µM). G Representative fluorescence images showing lipid peroxidation (C11-BODIPY 581/591) in GCs treated as indicated in panel. Scale bar = 100 μm. H-J Western blot analysis of ferroptosis markers (SLC7A11, GPX4, ACSL4), apoptosis markers (Bax, Bcl-2, Cleaved Caspase-3), and gap junction proteins (Cx43, Cx37) in GCs treated under the following conditions for 24 h: control, LG, LG + VE (50 µM), and LG + Ferrostatin-1 (Fer-1; 1 µM). (K) Representative TUNEL staining images of GCs treated for 24 h under the indicated conditions: control, LG, LG + VE (50 µM), and LG + Fer-1 (1 µM). Scale bar = 100 μm. Data are presented as mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001; ns, not significant Discussion In this study, we used low-glucose-cultured yak granulosa cells to investigate energy stress and to assess cellular responses. We show that in yak granulosa cells, vitamin E alleviates energy stress-induced apoptosis and gap junction impairment by disrupting a coordinated AMPK/mTOR–mitophagy–ferroptosis axis. This finding suggests a cell death mode in which prolonged energy stress subverts the homeostatic process of mitophagy, transforming it into a driver of ferroptosis. In this pathogenic shift, the cellular quality-control machinery switches from a protective to a destructive program, ultimately committing the cell to death via ferroptosis. Our observations confirm that mitochondria are primary targets of energy stress in granulosa cells. Low-glucose stress induced characteristic signs of mitochondrial dysfunction—dissipation of membrane potential and reduced ATP production—which triggered a mitophagic response, as evidenced by upregulation of PINK1 and Parkin and recruitment of LC3 to mitochondria. Notably, both VE and the mitochondrial-targeted antioxidant MitoQ similarly attenuated these events, indicating a role for mitochondrial oxidative stress in this cascade. Importantly, inhibition of autophagosome-lysosome fusion with Chloroquine (CQ) mitigated apoptosis and restored gap junction protein levels, demonstrating that the completion of mitophagy, rather than its initiation, is a step committing cells to death. Thus, VE appears to act upstream by restraining mitophagy. The AMPK/mTOR pathway functions as a cellular energy sensor [39–41], and its sustained activation reflects a cellular response to energy crisis [42, 43]. This aligns with our findings, wherein sustained energy stress induced by low-glucose treatment activated AMPK (as indicated by increased p-AMPK levels) and inhibited mTOR activity (as indicated by decreased p-mTOR levels). Importantly, treatment with the AMPK inhibitor Compound C mimicked the protective effects of VE—reducing mitophagy, suppressing apoptosis, and restoring gap junction protein expression. These findings suggest AMPK activation as an upstream driver of the downstream pathological cascade and indicate that VE acts by inhibiting this excessive AMPK activation and restoring mTOR activity. The sustained activation of this axis under prolonged stress may represent a juncture, where the cell’s primary energy-sensing mechanism transitions from a survival response to a death signal. Our data suggest a sequential relationship wherein mitophagy-triggered ferroptosis acts as a downstream effector of cell death. Low-glucose treatment upregulated the pro-ferroptotic protein ACSL4 while downregulating the ferroptosis inhibitors SLC7A11 and GPX4. Notably, inhibition of autophagic flux with CQ recapitulated the protective effects of VE, MitoQ, and Compound C against ferroptosis, while the ferroptosis inhibitor Ferrostatin-1 rescued both apoptosis and gap junction impairment, confirming ferroptosis as a mediator of the damage. We propose two complementary mechanisms through which mitophagy drives ferroptosis: First, mitophagy supplies ferroptosis substrates. The degradation of damaged mitochondria releases mitochondrial membrane-derived PUFAs into the cytosol [44–46], providing abundant substrate for lipid peroxidation. Consistent with this, inhibiting autophagic flux with CQ alleviated ferroptosis. Second, mitophagy compromises cellular defense. PINK1/PARKIN-mediated mitophagy may also promote the degradation of GPX4, an anti-ferroptotic enzyme [44–46]. Supporting this, CQ treatment preserved the highest GPX4 levels, whereas the autophagy inducer Rapamycin (Rapa) caused the greatest GPX4 depletion—an effect not rescued by VE. However, Rapamycin is a specific inhibitor of mTOR, and its effects are not limited to the induction of autophagy, as mTOR signaling regulates a wide range of cellular processes including metabolism, transcription, and translation [47]. Therefore, the exacerbation of ferroptosis observed with Rapamycin treatment could potentially involve mTOR-dependent but autophagy-independent mechanisms. Together, these data highlight the execution of mitophagy—the clearance process itself—as a commitment step in cell death under energy stress. Thus, VE confers protection by suppressing this integrated AMPK/mTOR–mitophagy–ferroptosis axis, thereby preventing the pathological execution of the cell death program. Both vitamin E and the mitochondria-targeted antioxidant MitoQ exerted protective effects in our model. This observation suggests two points. First, it establishes oxidative damage as a unifying thread linking all levels of the AMPK/mTOR–mitophagy–ferroptosis axis. Second, it demonstrates that effective blockade of this pathological cascade requires an upstream intervention that scavenges ROS and preserves mitochondrial integrity, thereby preventing the initiation of the pathological cascade. Together, these findings reinforce the hierarchical architecture of this axis and underscore the therapeutic rationale for targeting its earliest trigger, as exemplified by VE. Conclusions Our data position the energy stress-activated AMPK/mTOR–mitophagy–ferroptosis axis as a mechanistic driver of granulosa cell apoptosis and follicular atresia. Notably, dysregulation of AMPK/mTOR signaling, mitochondrial dysfunction, and oxidative stress—components of this axis—are recognized features across a spectrum of reproductive pathologies, including polycystic ovary syndrome [48], diabetic reproductive dysfunction [49], age-related ovarian decline [50], and follicular stagnation [51]. Consequently, this signaling axis may represent a common pathway linking dysregulated energy metabolism and aberrant cell death across multiple reproductive disorders. Accumulating evidence suggests that the core components of this signaling axis are evolutionarily conserved. The AMPK-mTOR module is conserved across the three domains of life [52], while ferroptosis represents a fundamental and distinct cell death modality [53]. Furthermore, the regulatory interplay between AMPK and ferroptosis has been implicated in various pathological contexts, including cancer [47]. Together with our findings in yak granulosa cells, these studies suggest that the mechanisms governing energy stress responses and cell fate may be conserved across species. Nevertheless, the specific integration of these pathways into an AMPK/mTOR–mitophagy–ferroptosis axis, and its generalizability across mammalian species, warrants further comparative investigation. Supplementary Information Supplementary Material 1. Supplementary Material 1. Supplementary Material 2. Supplementary Material 2. Supplementary Material 3. Supplementary Material 3.