Stage-dependent DNA damage and mitochondrial dysfunction under simulated microgravity constrain oocyte maturation and are mitigated by melatonin

To determine whether SMG compromises immature oocytes, GV–stage oocytes were cultured in milrinone during a 12-h period of SMG exposure (Fig.  1 A). During SMG exposure, GV morphology remained intact, and milrinone-maintained GV arrest was not significantly altered (Fig.  1 B, C). Consistently, SMG did not increase oocyte degeneration or fragmentation at the GV stage (Fig.  1 D). However, following milrinone removal and subsequent culture under normal gravity (NG), oocyte maturation capacity was markedly impaired (Fig.  1 E, F). Specifically, the frequency of germinal vesicle breakdown (GVBD) was significantly reduced (Fig.  1 E), and the PB1 extrusion rate declined from 84.53 ± 1.24% in NG controls to 40.69 ± 3.05% in SMG-pretreated oocytes (Fig.  1 F). In parallel, the incidence of oocyte degeneration and fragmentation after meiotic maturation was increased (Fig.  1 G). These findings indicate that SMG exposure does not induce premature meiotic resumption at the GV stage but instead compromises the competence of oocytes to complete subsequent meiotic maturation. Fig. 1 Simulated microgravity induces DNA damage and ATM–CHK2-associated DNA damage responses in prophase-arrested oocytes and impairs subsequent meiotic maturation. A Schematic diagram of the experimental design. Immature oocytes from 3-week-old C57 mice were cultured and maintained in prophase arrest with milrinone under NG or SMG for 12 h. B Representative bright-field images of oocytes maintained in milrinone under NG or SMG for 12 h. Scale bar: 100 μm. C , D Proportions of GV-arrested and fragmented or degenerated oocytes after 12 h of culture under NG ( n  = 158) or SMG ( n  = 127). E - G Quantification of GVBD, PB1 extrusion, and fragmentation/degeneration after release from milrinone and subsequent in vitro maturation. H , I Quantification and representative immunofluorescence images of meiotic stage distribution in oocytes released from prophase arrest and cultured for 14 h. Stages include MI, A/T I, and MⅡ, recorded in NG ( n  = 74) and SMG ( n  = 111) oocytes. Chromosomes (magenta), microtubules (green), F-actin (blue). The yellow arrow indicates abnormal chromosome alignment in an SMG-exposed MI oocyte. J , K γ-H2AX staining showing DNA damage foci in NG and SMG oocytes and with quantification. Data were obtained from NG ( n  = 30) and SMG ( n  = 30) oocytes. Chromosomes (magenta), γ-H2AX (cyan). Scale bar: 10 μm. L , M RAD51 immunostaining and fluorescence intensity quantification in NG and SMG oocytes. Data were obtained from NG ( n  = 21) and SMG ( n  = 20) oocytes. Chromosomes (magenta), RAD51 (cyan). Scale bar: 5 μm. N , O Immunostaining and fluorescence intensity quantification of p-ATM and CHK2 in NG and SMG oocytes. Data were obtained from NG ( n  = 35) and SMG ( n  = 48) oocytes. Chromosomes (blue), p-ATM (green), and CHK2 (red). Scale bar: 25 μm. P , Q MitoSOX staining and quantification of mitochondrial superoxide levels in NG and SMG oocytes. Data were obtained from NG ( n  = 53) and SMG ( n  = 52) oocytes. MitoSOX (red). Scale bar: 100 μm. For panels C-I, each data point represents one independent replicate; oocytes from 4–5 mice were pooled in each replicate and randomly assigned to the indicated groups. For panels J - Q , the number of oocytes analyzed is indicated in each panel; in the SMG group, oocytes were pooled from two independent rotating dishes, with each treatment replicate using oocytes collected from 3–4 mice. Data are presented as mean ± SD; ns, not significant; * p  < 0.05; *** p  < 0.001; **** p  < 0.0001 Simulated microgravity induces DNA damage and ATM–CHK2-associated DNA damage responses in prophase-arrested oocytes and impairs subsequent meiotic maturation. A Schematic diagram of the experimental design. Immature oocytes from 3-week-old C57 mice were cultured and maintained in prophase arrest with milrinone under NG or SMG for 12 h. B Representative bright-field images of oocytes maintained in milrinone under NG or SMG for 12 h. Scale bar: 100 μm. C , D Proportions of GV-arrested and fragmented or degenerated oocytes after 12 h of culture under NG ( n  = 158) or SMG ( n  = 127). E - G Quantification of GVBD, PB1 extrusion, and fragmentation/degeneration after release from milrinone and subsequent in vitro maturation. H , I Quantification and representative immunofluorescence images of meiotic stage distribution in oocytes released from prophase arrest and cultured for 14 h. Stages include MI, A/T I, and MⅡ, recorded in NG ( n  = 74) and SMG ( n  = 111) oocytes. Chromosomes (magenta), microtubules (green), F-actin (blue). The yellow arrow indicates abnormal chromosome alignment in an SMG-exposed MI oocyte. J , K γ-H2AX staining showing DNA damage foci in NG and SMG oocytes and with quantification. Data were obtained from NG ( n  = 30) and SMG ( n  = 30) oocytes. Chromosomes (magenta), γ-H2AX (cyan). Scale bar: 10 μm. L , M RAD51 immunostaining and fluorescence intensity quantification in NG and SMG oocytes. Data were obtained from NG ( n  = 21) and SMG ( n  = 20) oocytes. Chromosomes (magenta), RAD51 (cyan). Scale bar: 5 μm. N , O Immunostaining and fluorescence intensity quantification of p-ATM and CHK2 in NG and SMG oocytes. Data were obtained from NG ( n  = 35) and SMG ( n  = 48) oocytes. Chromosomes (blue), p-ATM (green), and CHK2 (red). Scale bar: 25 μm. P , Q MitoSOX staining and quantification of mitochondrial superoxide levels in NG and SMG oocytes. Data were obtained from NG ( n  = 53) and SMG ( n  = 52) oocytes. MitoSOX (red). Scale bar: 100 μm. For panels C-I, each data point represents one independent replicate; oocytes from 4–5 mice were pooled in each replicate and randomly assigned to the indicated groups. For panels J - Q , the number of oocytes analyzed is indicated in each panel; in the SMG group, oocytes were pooled from two independent rotating dishes, with each treatment replicate using oocytes collected from 3–4 mice. Data are presented as mean ± SD; ns, not significant; * p  < 0.05; *** p  < 0.001; **** p  < 0.0001 To define the stage at which meiotic progression was disrupted, spindle–chromosome organization was examined using three-dimensional reconstruction (Fig.  1 H, I). While most NG oocytes progressed to MⅡ, SMG-pretreated oocytes showed a nearly fivefold increase in MI arrest (10.78 ± 2.12% vs. 52.51 ± 3.82%; Fig.  1 H), often with abnormal chromosome alignment (Fig.  1 I, SMG left). The fraction of oocytes at anaphase/telophase I (A/TI) was not significantly altered (Fig.  1 H). To determine whether spindle assembly checkpoint (SAC) activity contributed to this MI arrest, SMG-pretreated oocytes were released from milrinone and cultured to the MI stage, followed by treatment with reversine, an inhibitor of MPS1/TTK kinase activity, for 0.5–2 h (Figure S1A). Consistent with this model, reversine-mediated inhibition of MPS1/TTK increased MI completion (Figure S1B, C), supporting a functional contribution of SAC activity to SMG-associated MI arrest. These data indicate that oocytes exposed to SMG during prophase arrest subsequently exhibit pronounced MI arrest. Given that mammalian oocytes are highly sensitive to DNA damage during prophase arrest, and such damage can activate the SAC to cause MI arrest [ 33 ], we next examined genome integrity. Immunostaining for γH2AX revealed a marked increase in DNA damage foci in GV-stage oocytes following 12 h of SMG exposure compared with NG controls (Fig. 1 J, K). To determine whether the increased DNA damage signal after SMG exposure resulted from enhanced damage induction or insufficient DNA repair, we interfered with two major DSB repair pathways during SMG exposure. GV-stage oocytes were co-cultured under SMG with B02, an HR inhibitor targeting RAD51, or SCR7, an inhibitor of DNA ligase IV–dependent NHEJ (Figure S2A). The oocytes were then released into M16 medium to resume meiosis, allowing us to evaluate subsequent maturation-associated damage (Figure S2B). B02 treatment during SMG exposure aggravated degeneration and fragmentation in prophase-arrested oocytes (Figure S2C). During subsequent meiotic maturation, B02 further reduced PB1 extrusion (Figure S2D) and increased the proportion of degenerated and fragmented oocytes (Figure S2E). In contrast, SCR7 treatment did not produce a significant additional effect under the same conditions (Figure S2C–E). These results suggest that HR participates in the regulation of SMG-associated DNA damage in prophase-arrested oocytes. Previous studies have shown that GV-stage oocytes preferentially rely on HR to repair DNA double-strand breaks [ 17 , 37 , 38 ]. We therefore further examined RAD51, a key HR repair factor, by immunofluorescence staining (Fig. 1 L, M). RAD51 signals were significantly increased in prophase-arrested oocytes after SMG exposure, indicating that SMG-induced DNA damage is accompanied by upregulation of a RAD51-associated HR response. Together, these findings suggest that SMG induces DNA damage in GV-stage oocytes, while the oocytes concurrently enhance RAD51-associated HR repair to partially counteract this damage. To further assess the DNA damage response activated under SMG, we examined p-ATM and CHK2 in GV-stage oocytes. Both p-ATM and CHK2 signals were significantly increased after SMG exposure (Fig. 1 N, O), supporting activation of the ATM–CHK2 during prophase arrest. To assess whether this damage was related to oxidative stress, mitochondrial superoxide was measured using MitoSOX staining. No significant difference in MitoSOX fluorescence intensity was detected between SMG and NG groups (Fig.  1 P, Q). Intracellular ROS levels were also unchanged after SMG exposure (Figure S3A, B). Thus, SMG induces DNA damage in prophase-arrested oocytes, accompanied by RAD51-associated HR engagement and ATM–CHK2 activation, without a detectable increase in mitochondrial superoxide or total cellular ROS. To determine whether MLT mitigates SMG-induced oocyte damage in a stage-dependent manner, GV-stage oocytes were assigned to three experimental groups (Fig.  2 A): SMG (12 h SMG exposure under GV arrest in milrinone followed by 14 h maturation under normal gravity), SMM (MLT supplementation restricted to the 14 h maturation phase), and MLT (10⁻⁵ M MLT supplementation during the 12 h SMG exposure followed by maturation without MLT). Following in vitro maturation, both GVBD and PB1 extrusion rates were significantly higher in the MLT group compared with SMG group (Fig.  2 B-D). By contrast, no improvement was observed in the SMM group (Fig.  2 B-D). Oocytes receiving MLT only during maturation failed to recover maturation competence, exhibiting GVBD and PB1 extrusion rates comparable to those of SMG controls. The incidence of degeneration and fragmentation did not differ significantly among groups (Fig.  2 E). These results indicate that MLT restores oocyte developmental competence only when present during SMG exposure, whereas delayed supplementation after SMG is ineffective. Fig. 2 Melatonin (MLT) alleviates SMG-induced defects in meiotic maturation, DNA damage, and early apoptosis of prophase-arrested oocytes. A Schematic diagram of the experimental design. Oocytes were assigned to normal gravity (NG), simulated microgravity (SMG), SMM (MLT supplementation during meiotic maturation only), or MLT (MLT treatment during SMG exposure) groups. B Representative bright-field images of oocytes after in vitro maturation in NG, SMG, MLT, and SMM groups. The enlarged images show representative abnormal oocytes: in the SMG group, the upper inset shows an oocyte that failed to undergo GVBD, and the lower inset shows a degenerated oocyte; in the SMM group, the upper inset shows an oocyte arrested without PB1 extrusion, and the lower inset shows a degenerated oocyte. Scale bar: 100 μm. C - E Quantification of GVBD, PB1 extrusion, and oocytes fragmentation/ degeneration after in vitro culture, recorded in NG ( n  = 75), SMG ( n  = 63), SMM ( n  = 60) and MLT ( n  = 60) oocytes. F , G Representative γ-H2AX staining (cyan) with chromosomes (magenta) and quantification of γ-H2AX foci per oocyte, recorded in NG ( n  = 21), SMG ( n  = 22) and MLT ( n  = 20) oocytes. Scale bar: 10 μm. H , I MitoTracker Green staining showing mitochondrial distribution in oocytes, with the relative proportions of perinuclear versus cortical localization quantified for NG ( n  = 43), SMG ( n  = 38), and MLT ( n  = 39) oocytes. Scale bar: 100 μm. J , K Annexin V staining for detection of early apoptosis, with representative images and quantification across groups, recorded in NG ( n  = 42), SMG ( n  = 38) and MLT ( n  = 44) oocytes. Scale bar: 25 μm. For panel F , the number of oocytes analyzed is indicated in the figure; in the SMG group, oocytes were pooled from two independent rotating dishes. For the remaining quantitative panels, each data point represents one independent replicate; in each replicate, oocytes were pooled from 4–6 mice and randomly assigned to the indicated groups. Data are presented as mean ± SD; ns, not significant; * p  < 0.05; ** p  < 0.01; *** p  < 0.001; **** p  < 0.0001 Melatonin (MLT) alleviates SMG-induced defects in meiotic maturation, DNA damage, and early apoptosis of prophase-arrested oocytes. A Schematic diagram of the experimental design. Oocytes were assigned to normal gravity (NG), simulated microgravity (SMG), SMM (MLT supplementation during meiotic maturation only), or MLT (MLT treatment during SMG exposure) groups. B Representative bright-field images of oocytes after in vitro maturation in NG, SMG, MLT, and SMM groups. The enlarged images show representative abnormal oocytes: in the SMG group, the upper inset shows an oocyte that failed to undergo GVBD, and the lower inset shows a degenerated oocyte; in the SMM group, the upper inset shows an oocyte arrested without PB1 extrusion, and the lower inset shows a degenerated oocyte. Scale bar: 100 μm. C - E Quantification of GVBD, PB1 extrusion, and oocytes fragmentation/ degeneration after in vitro culture, recorded in NG ( n  = 75), SMG ( n  = 63), SMM ( n  = 60) and MLT ( n  = 60) oocytes. F , G Representative γ-H2AX staining (cyan) with chromosomes (magenta) and quantification of γ-H2AX foci per oocyte, recorded in NG ( n  = 21), SMG ( n  = 22) and MLT ( n  = 20) oocytes. Scale bar: 10 μm. H , I MitoTracker Green staining showing mitochondrial distribution in oocytes, with the relative proportions of perinuclear versus cortical localization quantified for NG ( n  = 43), SMG ( n  = 38), and MLT ( n  = 39) oocytes. Scale bar: 100 μm. J , K Annexin V staining for detection of early apoptosis, with representative images and quantification across groups, recorded in NG ( n  = 42), SMG ( n  = 38) and MLT ( n  = 44) oocytes. Scale bar: 25 μm. For panel F , the number of oocytes analyzed is indicated in the figure; in the SMG group, oocytes were pooled from two independent rotating dishes. For the remaining quantitative panels, each data point represents one independent replicate; in each replicate, oocytes were pooled from 4–6 mice and randomly assigned to the indicated groups. Data are presented as mean ± SD; ns, not significant; * p  < 0.05; ** p  < 0.01; *** p  < 0.001; **** p  < 0.0001 Consistent with the improved maturation outcomes, MLT supplementation during SMG exposure markedly reduced DNA damage in GV-stage oocytes, as indicated by decreased γH2AX foci compared with SMG controls (Fig. 2 F, G). Given our previous finding that SMG disrupts mitochondrial organization [ 39 ], and the close functional linkage between mitochondrial dynamics and genome stability [ 40 , 41 ], we next analyzed mitochondrial distribution patterns (Fig. 2 H). In NG oocytes, mitochondria were predominantly perinuclear, whereas SMG exposure induced a pronounced redistribution toward the oocyte cortex, increasing cortical localization nearly tenfold (2.56 ± 4.44% vs. 25.79 ± 9.55%). MLT supplementation during SMG largely reversed this redistribution, enhancing perinuclear distribution and reducing aberrant cortical accumulation (Fig. 2 I). In addition, MLT significantly lowered the incidence of oocytes undergoing early apoptosis, as assessed by Annexin V staining (Fig. 2 J, K). Together, these findings demonstrate that MLT counteracts SMG-induced oocyte defects when administered during the prophase-arrested stage, concomitantly reducing DNA damage, normalizing mitochondrial distribution, and improving meiotic competence. MLT improved subsequent maturation competence only when administered during prophase-arrested SMG exposure, whereas the same dose failed to enhance maturation when added after the SMG exposure period. This timing-dependent effect suggested that MLT protection might be linked to the repair of DNA damage incurred during prophase arrest. Previous studies have reported that MLT facilitates DNA damage repair through enhancement of the NHEJ pathway [ 25 ]. To determine whether a similar mechanism operates in SMG-exposed oocytes, we next examined the involvement of NHEJ in MLT-mediated repair. To this end, the NHEJ inhibitor SCR7 was applied during SMG treatment. If MLT alleviates SMG-induced DNA damage through NHEJ-dependent repair, inhibition of this pathway would be expected to abolish its protective effects. Consistent with this prediction, γH2AX immunostaining showed that SMG exposure markedly increased DNA damage foci, while MLT supplementation significantly reduced γH2AX signals. This reduction was abolished by SCR7 co-treatment, with SMG + MLT + SCR7 oocytes showing γH2AX foci at levels comparable to those in SMG-exposed oocytes (Fig. 3 A, B). We further examined DNA-PKcs (Fig. 3 C, D), a core factor involved in NHEJ-mediated DNA double-strand break repair [ 42 ]. DNA-PKcs fluorescence intensity was reduced after SMG exposure but was markedly increased by MLT treatment under SMG conditions. These data suggest that MLT promotes a DNA-PKcs-associated NHEJ response in prophase-arrested oocytes exposed to SMG. Fig. 3 MLT alleviates SMG-induced DNA damage and meiotic arrest through a DNA-PKcs-associated NHEJ pathway. A , B Representative γ-H2AX immunofluorescence staining (cyan) with chromosomes (magenta) and quantification of γ-H2AX foci number per oocyte in oocytes exposed to SMG with or without MLT and NHEJ inhibitor SCR7 treatment, recorded in NG ( n  = 21), SMG ( n  = 22), SMG + MLT ( n  = 22) and SMG + MLT + SCR7 ( n  = 22) oocytes. Scale bar: 10 μm. C , D Quantification and representative immunofluorescence images of DNA-PKcs in NG, SMG, and SMG + MLT oocytes. Data were obtained from NG ( n  = 24), SMG ( n  = 29), and SMG + MLT ( n  = 37) oocytes. Chromosomes (magenta), DNA-PKcs (cyan). Scale bar: 25 μm. E Representative bright-field images of MⅡ oocytes matured in vitro with MLT or MLT + SCR7. Scale bar: 100 μm. F , G Quantification of GVBD and PB1 extrusion across groups, recorded in NG ( n  = 76), SMG ( n  = 74), SMG + MLT ( n  = 74) and SMG + MLT + SCR7 ( n  = 129) oocytes. H , I Immunofluorescence staining of spindle organization (microtubules, green), chromosome alignment (magenta), and F-actin (blue) in oocytes released from prophase arrest and cultured for 14 h. The distribution of oocytes at MI and MⅡ stages was quantified for NG ( n  = 64), SMG ( n  = 56), MLT ( n  = 71) and SCR7 ( n  = 68) oocytes. Red arrows indicate abnormal chromosome alignment in SMG-treated oocytes, and yellow arrows indicate abnormal chromosome alignment after SCR7 co-treatment. For panels A-D, the number of oocytes analyzed is indicated in the figure; in the SMG groups, oocytes were pooled from two independent rotating dishes, and each treatment replicate used oocytes collected from 3–4 mice. For panels E - I , each data point represents one independent replicate, and oocytes in each replicate were pooled from 3–4 mice. Data are presented as mean ± SD; ns, not significant; * p  < 0.05; ** p  < 0.01; *** p  < 0.001; **** p  < 0.0001 MLT alleviates SMG-induced DNA damage and meiotic arrest through a DNA-PKcs-associated NHEJ pathway. A , B Representative γ-H2AX immunofluorescence staining (cyan) with chromosomes (magenta) and quantification of γ-H2AX foci number per oocyte in oocytes exposed to SMG with or without MLT and NHEJ inhibitor SCR7 treatment, recorded in NG ( n  = 21), SMG ( n  = 22), SMG + MLT ( n  = 22) and SMG + MLT + SCR7 ( n  = 22) oocytes. Scale bar: 10 μm. C , D Quantification and representative immunofluorescence images of DNA-PKcs in NG, SMG, and SMG + MLT oocytes. Data were obtained from NG ( n  = 24), SMG ( n  = 29), and SMG + MLT ( n  = 37) oocytes. Chromosomes (magenta), DNA-PKcs (cyan). Scale bar: 25 μm. E Representative bright-field images of MⅡ oocytes matured in vitro with MLT or MLT + SCR7. Scale bar: 100 μm. F , G Quantification of GVBD and PB1 extrusion across groups, recorded in NG ( n  = 76), SMG ( n  = 74), SMG + MLT ( n  = 74) and SMG + MLT + SCR7 ( n  = 129) oocytes. H , I Immunofluorescence staining of spindle organization (microtubules, green), chromosome alignment (magenta), and F-actin (blue) in oocytes released from prophase arrest and cultured for 14 h. The distribution of oocytes at MI and MⅡ stages was quantified for NG ( n  = 64), SMG ( n  = 56), MLT ( n  = 71) and SCR7 ( n  = 68) oocytes. Red arrows indicate abnormal chromosome alignment in SMG-treated oocytes, and yellow arrows indicate abnormal chromosome alignment after SCR7 co-treatment. For panels A-D, the number of oocytes analyzed is indicated in the figure; in the SMG groups, oocytes were pooled from two independent rotating dishes, and each treatment replicate used oocytes collected from 3–4 mice. For panels E - I , each data point represents one independent replicate, and oocytes in each replicate were pooled from 3–4 mice. Data are presented as mean ± SD; ns, not significant; * p  < 0.05; ** p  < 0.01; *** p  < 0.001; **** p  < 0.0001 Following milrinone removal and meiotic resumption, MLT significantly improved both GVBD and PB1 extrusion rates, whereas SCR7 co-treatment eliminated these improvements, yielding maturation outcomes comparable to those of SMG alone (Fig.  3 E-G). Consistently, three-dimensional spindle imaging further confirmed that MLT reduced the proportion of oocytes arrested at MI and promoted progression to MⅡ. In contrast, SCR7 co-treatment negated these effects, resulting in MI arrest frequencies comparable to SMG alone (Fig.  3 H, I). Together, these findings indicate that MLT alleviates SMG-induced oocyte damage through an NHEJ-dependent DNA repair mechanism. Blocking NHEJ prevents MLT-mediated reduction of DNA damage and compromises the subsequent recovery of meiotic maturation, thereby maintaining a high frequency of MI arrest after SMG exposure. Previous studies have shown that, as oocytes progress from the GV stage to the MⅡ stage, DNA double-strand break repair shifts from predominant reliance on HR toward greater use of NHEJ [ 38 , 43 ]. Consistent with this idea, inhibition of NHEJ during oocyte maturation has been reported to impair meiotic maturation [ 17 ]. Given that MLT supplementation during the maturation phase at the original concentration did not improve PB1 extrusion (Fig. 2 C, D), whereas our preceding results linked MLT-mediated protection to NHEJ activity, we next tested whether the maturation-phase response to MLT depends on both MLT concentration and the degree of NHEJ inhibition. Oocytes were first exposed to SMG for 12 h under prophase arrest and then matured for 14 h in the presence of different concentrations of SCR7, with or without a higher concentration of MLT (Figure S4A). Low-dose SCR7 (5 µM) reduced PB1 extrusion, and a higher SCR7 concentration (20 µM) further impaired PB1 extrusion during maturation (Figure S4B, C). Increasing MLT to 10 − 3 M improved PB1 extrusion after SMG exposure. This improvement was maintained in the presence of 5 µM SCR7 but was lost when SCR7 was increased to 20 µM (Figure S4B, C). These data indicate that high-dose MLT can partially improve maturation outcomes after SMG exposure, whereas stronger NHEJ inhibition abolishes this effect. MLT exerts its biological effects through two potential mechanisms: by activating its membrane receptors MT1 and MT2, or through receptor-independent pathways that directly influence DNA damage. To determine whether MT1/MT2 signaling is required for MLT-mediated protection under SMG, we used a receptor-specific agonist ramelteon (50 nM) and antagonist luzindole (100 nM). γH2AX immunostaining revealed that ramelteon administration during SMG exposure did not reduce DNA damage foci, with γH2AX levels remaining comparable to those observed in SMG controls (Fig.  4 A, B). Thus, activation of MT1/MT2 receptors alone was insufficient to reproduce the DNA-protective effect of MLT. In contrast, co-treatment with MLT and luzindole did not abolish the MLT-induced reduction in γH2AX foci, as DNA damage levels remained significantly lower than those in SMG-only oocytes (Fig.  4 C, D). These data indicate that MLT alleviates SMG-induced DNA damage through a mechanism that does not require MT1/MT2 receptor signaling. Fig. 4 MLT-induced DNA damage repair under SMG is independent of canonical melatonin receptor signaling. A , B Representative γ-H2AX immunofluorescence staining (cyan) with chromosomes (magenta) and quantification of γ-H2AX foci in oocytes cultured under SMG with MLT or the melatonin receptor agonist ramelteon, recorded in NG ( n  = 14), SMG ( n  = 15), SMG + MLT ( n  = 15) and SMG + ramelteon ( n  = 20) oocytes. Scale bar: 10 μm. C , D γ-H2AX immunofluorescence staining and corresponding quantification in oocytes cultured under SMG with MLT in the presence or absence of the melatonin receptor antagonist luzindole, recorded in NG ( n  = 19), SMG ( n  = 20), SMG + MLT ( n  = 20) and SMG + MLT + luzindole ( n  = 23) oocytes. Scale bar: 10 μm. E , F Transcriptomic and translatomic expression profiles of MT1 ( Mtnr1a / MTNR1A ) and MT2 ( Mtnr1b / MTNR1B ) in GV-stage mouse and human oocytes. Zp3 / ZP3 was used as a highly expressed oocyte reference gene; its mean expression values were 4985 and 6348 in mouse transcriptome and translatome datasets, and 6417 and 4354 in human transcriptome and translatome datasets, respectively. G Representative bright-field images of MⅡ oocytes after in vitro maturation under SMG, SMG + MLT, SMG + ramelteon, or SMG + MLT + luzindole conditions. Scale bar: 100 μm. H , I Quantification of GVBD and PB1 extrusion across groups, recorded in NG ( n  = 73), SMG ( n  = 69), SMG + MLT ( n  = 65), SMG + ramelteon ( n  = 85), and SMG + MLT + luzindole ( n  = 79) oocytes. For panels A - D , the number of oocytes analyzed is indicated in the figure; in the SMG groups, oocytes were pooled from two independent rotating dishes, and each treatment replicate used oocytes collected from 3–4 mice. For panels G - I , each data point represents one independent replicate, and oocytes in each replicate were pooled from 3–4 mice. Data are presented as mean ± SD; ns, not significant; *** p  < 0.001; **** p  < 0.0001 MLT-induced DNA damage repair under SMG is independent of canonical melatonin receptor signaling. A , B Representative γ-H2AX immunofluorescence staining (cyan) with chromosomes (magenta) and quantification of γ-H2AX foci in oocytes cultured under SMG with MLT or the melatonin receptor agonist ramelteon, recorded in NG ( n  = 14), SMG ( n  = 15), SMG + MLT ( n  = 15) and SMG + ramelteon ( n  = 20) oocytes. Scale bar: 10 μm. C , D γ-H2AX immunofluorescence staining and corresponding quantification in oocytes cultured under SMG with MLT in the presence or absence of the melatonin receptor antagonist luzindole, recorded in NG ( n  = 19), SMG ( n  = 20), SMG + MLT ( n  = 20) and SMG + MLT + luzindole ( n  = 23) oocytes. Scale bar: 10 μm. E , F Transcriptomic and translatomic expression profiles of MT1 ( Mtnr1a / MTNR1A ) and MT2 ( Mtnr1b / MTNR1B ) in GV-stage mouse and human oocytes. Zp3 / ZP3 was used as a highly expressed oocyte reference gene; its mean expression values were 4985 and 6348 in mouse transcriptome and translatome datasets, and 6417 and 4354 in human transcriptome and translatome datasets, respectively. G Representative bright-field images of MⅡ oocytes after in vitro maturation under SMG, SMG + MLT, SMG + ramelteon, or SMG + MLT + luzindole conditions. Scale bar: 100 μm. H , I Quantification of GVBD and PB1 extrusion across groups, recorded in NG ( n  = 73), SMG ( n  = 69), SMG + MLT ( n  = 65), SMG + ramelteon ( n  = 85), and SMG + MLT + luzindole ( n  = 79) oocytes. For panels A - D , the number of oocytes analyzed is indicated in the figure; in the SMG groups, oocytes were pooled from two independent rotating dishes, and each treatment replicate used oocytes collected from 3–4 mice. For panels G - I , each data point represents one independent replicate, and oocytes in each replicate were pooled from 3–4 mice. Data are presented as mean ± SD; ns, not significant; *** p  < 0.001; **** p  < 0.0001 To further evaluate the potential contribution of canonical melatonin receptors, we analyzed recently published Smart-seq and translatome datasets from mouse and human oocytes [ 44 ]. In mouse GV-stage oocytes, Mtnr1a , which encodes MT1, was undetectable, whereas Mtnr1b , which encodes MT2, showed absent or only minimal expression at the transcriptomic and translatomic levels (Fig. 4 E). A similar pattern was observed in human GV-stage oocytes, in which MTNR1A and MTNR1B were absent or expressed at very low levels in most datasets (Fig. 4 F). These expression profiles further support the possibility that GV-stage oocytes respond to MLT primarily through receptor-independent mechanisms. We next asked whether this receptor independence also extended to the recovery of meiotic maturation after SMG exposure. For this purpose, oocytes treated under prophase arrest were released from milrinone and allowed to resume meiosis, and maturation outcomes were used to assess whether the protective effect of MLT on SMG-induced oocyte damage depended on MT1/MT2 signaling (Fig.  4 G). Consistent with the DNA damage data, Ramelteon treatment failed to improve either GVBD or PB1 extrusion rates. By contrast, MLT significantly restored maturation competence, and this effect persisted in the presence of luzindole (Fig.  4 H, I). Collectively, these results demonstrate that MLT reduces SMG-induced DNA damage and restores oocyte maturation capacity through receptor-independent mechanisms, distinguishing its actions from those mediated by canonical melatonin receptor signaling. Our previous work demonstrated that SMG impairs oocyte meiotic maturation [ 7 ]. We therefore next examined whether MLT could alleviate SMG-induced defects during the maturation phase itself. GV-stage oocytes were cultured under SMG conditions in the presence of increasing concentrations of MLT (10 − 9 M, 10 − 7 M, and 10 − 5 M), and maturation outcomes were evaluated following 14 h of in vitro culture (Fig. 5 A, B). Across all experimental groups, GVBD rates remained comparable, indicating that SMG and MLT treatment did not affect meiotic resumption per se (Fig. 5 C). In contrast, PB1 extrusion exhibited a clear concentration-dependent response to MLT, increasing significantly at 10 − 9 M, reaching a maximum at 10 − 7 M, and showing no further enhancement at 10 − 5 M (Fig. 5 D). In parallel, the incidence of oocyte fragmentation and degeneration was consistently reduced at all three concentrations relative to SMG controls (Fig. 5 E). Based on these results, 10⁻⁷ M MLT, which conferred maximal improvement in maturation efficiency, was selected for subsequent experiments. Fig. 5 MLT improves SMG-induced meiotic defects during oocyte maturation in a receptor-independent manner. A Experimental design for assessing the effects of MLT on oocyte maturation under SMG. B Representative bright-field images of oocytes matured for 14 h under SMG in the presence or absence of MLT. Scale bar: 100 μm. C - E Quantification of GVBD, PB1 extrusion, and oocyte fragmentation/degeneration after 14 h maturation, recorded in NG ( n  = 193), SMG ( n  = 180), MLT (10 -9 M) ( n  = 191), MLT (10 -7 M) ( n  = 239) and MLT (10 -5 M) ( n  = 211) oocytes. F Transcriptomic and translatomic expression profiles of MT1 ( Mtnr1a / MTNR1A ) and MT2 ( Mtnr1b / MTNR1B ) in MⅡ-stage mouse oocytes and human MⅡ-stage oocytes or morulae. G Transcriptomic expression profiles of Mtnr1a and Mtnr1b in mouse oocytes matured under NG or SMG, and in SMG-arrested oocytes (SMF). H , I Representative images of oocytes matured for 14 h under NG + ramelteon, SMG, or SMG + ramelteon conditions, and quantification of PB1 extrusion. Scale bar: 100 μm. J , K Representative images of oocytes matured for 14 h under NG + luzindole, SMG + MLT, or SMG + MLT + luzindole conditions, and quantification of PB1 extrusion. Scale bar: 100 μm. For panels I and K , a total of seven experimental groups were analyzed: NG ( n  = 86), NG + ramelteon ( n  = 81), NG + luzindole ( n  = 91), SMG ( n  = 103), SMG + MLT ( n  = 106), SMG + ramelteon ( n  = 103), and SMG + MLT + luzindole ( n  = 116) oocytes. For panels C - E and H - K , each data point represents one independent replicate; oocytes in each replicate were pooled from 4–6 mice and randomly assigned to the indicated groups. The total number of oocytes analyzed is indicated in each panel. Data are presented as mean ± SD; ns, not significant; * p  < 0.05; ** p  < 0.01; *** p  < 0.001; **** p  < 0.0001 MLT improves SMG-induced meiotic defects during oocyte maturation in a receptor-independent manner. A Experimental design for assessing the effects of MLT on oocyte maturation under SMG. B Representative bright-field images of oocytes matured for 14 h under SMG in the presence or absence of MLT. Scale bar: 100 μm. C - E Quantification of GVBD, PB1 extrusion, and oocyte fragmentation/degeneration after 14 h maturation, recorded in NG ( n  = 193), SMG ( n  = 180), MLT (10 -9 M) ( n  = 191), MLT (10 -7 M) ( n  = 239) and MLT (10 -5 M) ( n  = 211) oocytes. F Transcriptomic and translatomic expression profiles of MT1 ( Mtnr1a / MTNR1A ) and MT2 ( Mtnr1b / MTNR1B ) in MⅡ-stage mouse oocytes and human MⅡ-stage oocytes or morulae. G Transcriptomic expression profiles of Mtnr1a and Mtnr1b in mouse oocytes matured under NG or SMG, and in SMG-arrested oocytes (SMF). H , I Representative images of oocytes matured for 14 h under NG + ramelteon, SMG, or SMG + ramelteon conditions, and quantification of PB1 extrusion. Scale bar: 100 μm. J , K Representative images of oocytes matured for 14 h under NG + luzindole, SMG + MLT, or SMG + MLT + luzindole conditions, and quantification of PB1 extrusion. Scale bar: 100 μm. For panels I and K , a total of seven experimental groups were analyzed: NG ( n  = 86), NG + ramelteon ( n  = 81), NG + luzindole ( n  = 91), SMG ( n  = 103), SMG + MLT ( n  = 106), SMG + ramelteon ( n  = 103), and SMG + MLT + luzindole ( n  = 116) oocytes. For panels C - E and H - K , each data point represents one independent replicate; oocytes in each replicate were pooled from 4–6 mice and randomly assigned to the indicated groups. The total number of oocytes analyzed is indicated in each panel. Data are presented as mean ± SD; ns, not significant; * p  < 0.05; ** p  < 0.01; *** p  < 0.001; **** p  < 0.0001 To assess whether canonical MLT receptor signaling might contribute to this maturation-stage effect, we first examined MT1/MT2 expression using published transcriptomic and translatomic datasets from mouse and human MⅡ-stage oocytes [ 44 ]. MT1- and MT2-encoding genes showed absent or very low expression at both transcriptomic and translatomic levels (Fig. 5 F). We further analyzed our previously generated Smart-seq2 dataset from oocytes matured under NG or SMG conditions [ 7 ]. Mtnr1a and Mtnr1b were absent or expressed at only minimal levels, with no apparent difference among NG oocytes, SMG-exposed oocytes that matured successfully, and SMG-exposed oocytes that failed to mature (Fig. 5 G). These expression profiles suggested that the beneficial effect of MLT on SMG-impaired oocyte maturation is unlikely to be primarily mediated by MT1 or MT2. To determine whether this effect depended on MT1/MT2 receptors, we tested the agonist ramelteon and the antagonist luzindole during oocyte maturation. Ramelteon treatment under NG conditions did not alter PB1 extrusion, confirming that the drug itself does not affect meiotic progression (Fig.  5 H, left; Fig.  5 I). Under SMG conditions, ramelteon likewise failed to improve PB1 extrusion compared with untreated controls (Fig.  5 H, right; Fig.  5 I). Similarly, luzindole alone did not impair oocyte maturation under NG conditions (Fig.  5 J, left; Fig.  5 K). Importantly, co-treatment with MLT and luzindole under SMG conditions significantly increased PB1 extrusion, with no difference from the MLT-only group (Fig.  5 J, middle and right; Fig.  5 K). Together, these results indicate that MLT rescues SMG-induced oocyte maturation arrest independently of MT1/MT2 receptor signaling, supporting the involvement of non-receptor–mediated mechanisms. Proper spindle assembly is essential for accurate chromosome segregation during meiotic division. Consistent with our previous findings, SMG markedly disrupted spindle assembly in MI oocytes, producing a higher frequency of aberrant spindles, including multipolar spindles and spindles with wide poles. To assess whether MLT mitigates these defects, spindle and chromosome organization were examined by α-tubulin immunofluorescence with chromosomal counterstaining in MI oocytes (Fig.  6 A). Quantitative analysis revealed that SMG exposure significantly increased the proportion of oocytes exhibiting aberrant spindles, whereas MLT co-treatment partially reversed this effect, reducing the incidence of abnormal spindle structures from 69.91 ± 6.06% to 45.39 ± 8.34% (Fig.  6 B). Further subclassification confirmed that both multipolar and wide-pole spindles were significantly reduced in the presence of MLT compared with SMG alone (Fig.  6 C, D). To further quantify these architectural defects, we measured MI-stage spindle morphology (Figure S5A). SMG increased spindle length, spindle width, and MI plate width, whereas MLT reduced these abnormalities toward the NG range (Figure S5B, C, E). The spindle length-to-width ratio was unchanged among groups (Figure S5D). These data support the categorical analysis and indicate that MLT partially alleviates SMG-induced spindle defects in MI oocytes. Fig. 6 MLT promotes spindle assembly and MTOC coalescence in MI oocytes under SMG. A Representative immunofluorescence images of spindle morphology and chromosome alignment in NG, SMG, and MLT oocytes. Microtubules (tubulin) are shown in green and chromosomes in magenta. Red and yellow arrows indicate representative spindle pole abnormalities and misaligned chromosomes, respectively. Scale bar: 25 μm. B - D Quantification of spindle abnormalities in MI oocytes, including overall aberrant spindles, multipolar spindles as shown in the left SMG representative image in panel A , and wide-pole spindles as shown in the middle SMG representative image in panel A . Data were recorded in NG ( n  = 51), SMG ( n  = 64) and MLT ( n  = 52) oocytes. E Quantification of the frequency of aberrant MTOC distribution in MI oocytes, recorded in NG ( n  = 66), SMG ( n  = 72), and MLT ( n  = 75) oocytes. F Representative immunofluorescence images of pericentrin-labeled MTOCs (red) with microtubules (green) and chromosomes (blue), together with quantitative analysis of MTOC distribution along the spindle axis (0–20, normalized distance). Red curves indicate MTOCs and blue curves indicate chromosome distribution. Scale bar: 25 μm. Each data point represents one independent replicate; oocytes in each replicate were pooled from 3–4 mice. Data are presented as mean ± SD; * p  < 0.05; ** p  < 0.01; *** p  < 0.001 MLT promotes spindle assembly and MTOC coalescence in MI oocytes under SMG. A Representative immunofluorescence images of spindle morphology and chromosome alignment in NG, SMG, and MLT oocytes. Microtubules (tubulin) are shown in green and chromosomes in magenta. Red and yellow arrows indicate representative spindle pole abnormalities and misaligned chromosomes, respectively. Scale bar: 25 μm. B - D Quantification of spindle abnormalities in MI oocytes, including overall aberrant spindles, multipolar spindles as shown in the left SMG representative image in panel A , and wide-pole spindles as shown in the middle SMG representative image in panel A . Data were recorded in NG ( n  = 51), SMG ( n  = 64) and MLT ( n  = 52) oocytes. E Quantification of the frequency of aberrant MTOC distribution in MI oocytes, recorded in NG ( n  = 66), SMG ( n  = 72), and MLT ( n  = 75) oocytes. F Representative immunofluorescence images of pericentrin-labeled MTOCs (red) with microtubules (green) and chromosomes (blue), together with quantitative analysis of MTOC distribution along the spindle axis (0–20, normalized distance). Red curves indicate MTOCs and blue curves indicate chromosome distribution. Scale bar: 25 μm. Each data point represents one independent replicate; oocytes in each replicate were pooled from 3–4 mice. Data are presented as mean ± SD; * p  < 0.05; ** p  < 0.01; *** p  < 0.001 Given that multipolar and wide-pole spindles often arise from defective MTOC fusion, we next examined MTOC distribution using pericentrin immunostaining. Under SMG conditions, pericentrin-positive foci failed to consolidate at discrete spindle poles and instead displayed a dispersed distribution, frequently overlapping with chromosome regions (Fig.  6 F, black arrows), a pattern consistent with impaired MTOC fusion and broadened metaphase plates (Fig.  6 E, F middle). By contrast, MLT treatment promoted the re-aggregation of pericentrin foci toward bipolar spindle poles, significantly reducing the proportion of oocytes with aberrant MTOC distribution (Fig.  6 E, F down). In summary, MLT significantly alleviates SMG-induced spindle assembly defects in MI oocytes and partially restores spindle polarity by improving MTOC coalescence, thereby supporting more orderly spindle organization under SMG conditions. Our previous work showed that exposure to SMG during oocyte maturation induces hyperactivation of oxidative phosphorylation (OXPHOS) and excessive stimulation of the UPR mt , contributing to meiotic arrest. To determine whether MLT modulates these processes, we first assessed UPR mt activation and mitochondrial distribution under SMG conditions. MI-stage oocytes were immunolabeled with α-tubulin to visualize spindles and co-stained for HSPE1 to assess UPR mt (Fig. 7 A) [ 45 , 46 ]. HSPE1 fluorescence intensity was significantly higher in SMG-treated oocytes than in NG controls (Fig. 7 B). Importantly, HSPE1 levels were comparable between oocytes exhibiting morphologically normal and abnormal spindles under SMG, indicating that UPR mt activation occurs broadly in response to SMG exposure rather than being restricted to oocytes with overt spindle defects (Fig. 7 B). MLT treatment markedly reduced HSPE1 fluorescence intensity relative to SMG alone (Fig. 7 B), indicating that MLT attenuates SMG-induced UPR mt overactivation. Fig. 7 MLT restores mitochondrial homeostasis, improves spindle organization, reduces early apoptosis, and enhances developmental competence under SMG. A , B HSPE1 immunostaining and fluorescence intensity quantification in MI oocytes. Data were recorded in NG ( n  = 24), SMG ( n  = 31), and MLT ( n  = 21) oocytes. Under SMG, HSPE1 levels were comparable between oocytes with abnormal ( n  = 17) and normal ( n  = 14) spindles. Chromosomes (blue), tubulin (green), HSPE1 (magenta). Scale bar: 25 μm. C , D JC-1 staining and quantification of mitochondrial membrane potential, shown as the aggregate-to-monomer ratio, in MI oocytes. Data were recorded in NG ( n  = 30), SMG ( n  = 35), and MLT ( n  = 33) oocytes. JC-1 aggregates (red), JC-1 monomers (green). Scale bar: 100 μm. E , F Representative immunofluorescence images and quantification of mitochondrial distribution patterns in MⅡ oocytes, recorded in NG ( n  = 44), SMG ( n  = 39) and MLT ( n  = 56) oocytes. Scale bar: 25 μm. G - I Representative immunofluorescence images of spindle organization and chromosome alignment in MⅡ oocytes stained for microtubules (tubulin, green), chromosomes (magenta), and F-actin (blue), together with quantification of aberrant spindle checkpoint complexes (SCCs) and telophase I arrest, recorded in NG ( n  = 55), SMG ( n  = 61) and MLT ( n  = 62) oocytes. Scale bar: 25 μm. J , K Annexin V staining and quantification of early apoptotic MⅡ oocytes across groups, recorded in NG ( n  = 62), SMG ( n  = 85) and MLT ( n  = 69) oocytes. Scale bar: 25 μm. L , M Representative images and quantification of parthenogenetically activated embryos at the 2-cell and blastocyst stages. Activated oocytes were assessed in SMG ( n  = 99) and MLT ( n  = 95) groups; blastocyst formation was assessed in SMG ( n  = 53) and MLT ( n  = 54) groups. Scale bar: 100 μm. For panels A - D , the number of oocytes analyzed is indicated in the figure; in the SMG groups, oocytes were pooled from two independent rotating dishes, and each treatment replicate used oocytes collected from 4–5 mice. For panels E - M , each data point represents one independent replicate, and oocytes or embryos in each replicate were pooled from 3–4 mice. Data are presented as mean ± SD; ns, not significant; * p  < 0.05; ** p  < 0.01; *** p  < 0.001; **** p  < 0.0001 MLT restores mitochondrial homeostasis, improves spindle organization, reduces early apoptosis, and enhances developmental competence under SMG. A , B HSPE1 immunostaining and fluorescence intensity quantification in MI oocytes. Data were recorded in NG ( n  = 24), SMG ( n  = 31), and MLT ( n  = 21) oocytes. Under SMG, HSPE1 levels were comparable between oocytes with abnormal ( n  = 17) and normal ( n  = 14) spindles. Chromosomes (blue), tubulin (green), HSPE1 (magenta). Scale bar: 25 μm. C , D JC-1 staining and quantification of mitochondrial membrane potential, shown as the aggregate-to-monomer ratio, in MI oocytes. Data were recorded in NG ( n  = 30), SMG ( n  = 35), and MLT ( n  = 33) oocytes. JC-1 aggregates (red), JC-1 monomers (green). Scale bar: 100 μm. E , F Representative immunofluorescence images and quantification of mitochondrial distribution patterns in MⅡ oocytes, recorded in NG ( n  = 44), SMG ( n  = 39) and MLT ( n  = 56) oocytes. Scale bar: 25 μm. G - I Representative immunofluorescence images of spindle organization and chromosome alignment in MⅡ oocytes stained for microtubules (tubulin, green), chromosomes (magenta), and F-actin (blue), together with quantification of aberrant spindle checkpoint complexes (SCCs) and telophase I arrest, recorded in NG ( n  = 55), SMG ( n  = 61) and MLT ( n  = 62) oocytes. Scale bar: 25 μm. J , K Annexin V staining and quantification of early apoptotic MⅡ oocytes across groups, recorded in NG ( n  = 62), SMG ( n  = 85) and MLT ( n  = 69) oocytes. Scale bar: 25 μm. L , M Representative images and quantification of parthenogenetically activated embryos at the 2-cell and blastocyst stages. Activated oocytes were assessed in SMG ( n  = 99) and MLT ( n  = 95) groups; blastocyst formation was assessed in SMG ( n  = 53) and MLT ( n  = 54) groups. Scale bar: 100 μm. For panels A - D , the number of oocytes analyzed is indicated in the figure; in the SMG groups, oocytes were pooled from two independent rotating dishes, and each treatment replicate used oocytes collected from 4–5 mice. For panels E - M , each data point represents one independent replicate, and oocytes or embryos in each replicate were pooled from 3–4 mice. Data are presented as mean ± SD; ns, not significant; * p  < 0.05; ** p  < 0.01; *** p  < 0.001; **** p  < 0.0001 Mitochondrial membrane potential is a key indicator of mitochondrial functional status and is closely linked to ATP production, redox balance, and mitochondrial stress responses [ 47 ]. We therefore assessed mitochondrial membrane potential using JC-1 staining (Fig. 7 D). SMG exposure increased the JC-1 aggregate/monomer ratio, indicating mitochondrial hyperpolarization in MI oocytes (Fig. 7 C). MLT treatment significantly reduced this ratio compared with SMG alone, suggesting that MLT alleviates SMG-induced mitochondrial hyperpolarization (Fig. 7 C, D). We next analyzed mitochondrial distribution using MitoTracker staining (Fig.  7 E). SMG exposure significantly increased the proportion of oocytes displaying abnormal mitochondrial distribution. MLT treatment reduced this proportion and restored a more homogeneous intracellular distribution (Fig.  7 F). Spindle organization was then evaluated in MⅡ-stage oocytes by α-tubulin and chromosomal staining (Fig.  7 G). SMG markedly increased the frequency of abnormal spindle structures, whereas MLT treatment significantly reduced this proportion (Fig.  7 H). In parallel, SMG led to a higher incidence of telophase I (TI) arrest, which was partially reversed by MLT supplementation (Fig.  7 I). Finally, early apoptosis was assessed by Annexin-V staining in MⅡ-stage oocytes (Fig.  7 J). SMG significantly elevated the proportion of oocytes undergoing early apoptosis, whereas MLT treatment during maturation substantially reduced apoptotic incidence (Fig.  7 K). We then assessed whether the improvement in oocyte quality was accompanied by enhanced developmental competence using parthenogenetic activation. Before testing SMG-exposed oocytes, the activation system was validated using in vivo- and in vitro-matured oocytes (Figure S6). Oocytes matured under SMG conditions with or without MLT were then subjected to parthenogenetic activation (Fig.  7 L). MLT-treated oocytes showed a trend toward a higher activation rate compared with SMG controls, although this difference did not reach statistical significance ( p  = 0.0808; Fig.  7 M). Notably, MLT significantly increased the blastocyst formation rate after activation (Fig.  7 M), indicating improved intrinsic developmental competence of SMG-exposed oocytes. Collectively, these results demonstrate that MLT supplementation during oocyte maturation under SMG conditions alleviates mitochondrial stress, improves mitochondrial distribution, reduces spindle assembly abnormalities, and limits early apoptotic progression, and improves subsequent developmental competence.

Female C57BL/6J mice (3 weeks old) were used (Approval Number: SIAT-IACUC-210427-YYS-LXH-A1514-01). MLT (M5250, Sigma Aldrich) was applied at final concentrations of 10 − 9 , 10 − 7 , and 10 − 5 M. The NHEJ inhibitor SCR7 (HY-107845, MCE) was used at 20 µM. The HR inhibitor B02 (HY-101462, MCE) was used at 50 µM, luzindole (0877, Tocris Bioscience) at 100 nM, and ramelteon (6777, Tocris Bioscience) at 50 nM. All compounds were dissolved in DMSO or ethanol and diluted to final concentrations not exceeding 0.1%. Mice received 5 IU pregnant mare serum gonadotropins (PMSG) by subcutaneous injection, and cumulus–oocyte complexes (COCs) were collected 44–46 h later. Fully grown oocytes arrested at prophase I were isolated in M2 medium (Sigma-Aldrich, USA) supplemented with 2.5 µM milrinone (MCE, USA). For experiments with prophase-arrested oocytes, the cells were cultured for 12 h in M16 medium supplemented with 2.5 µM milrinone under SMG using a RCCS. The rotation rate for inducing SMG was set at 15 rotations per min [ 7 , 8 ]. Normal gravity (NG) oocytes were used as controls, while the melatonin-treated group received 10 − 5 M MLT under SMG for the same duration. This concentration was selected based on prior work showing that 10 − 3 M melatonin alleviated DNA damage in prophase-arrested oocytes [ 25 ]; we used a lower dose (10 − 5 M) to reduce potential off-target effects while maintaining efficacy. Following treatment, cumulus cells were mechanically removed. Germinal vesicle (GV)-stage oocytes were either imaged immediately by differential interference contrast (DIC) microscopy, fixed immediately for subsequent immunostaining, or GV oocytes from both groups were transferred to M16 medium covered with mineral oil (M5310, Sigma Aldrich) for in vitro maturation. For maturation-stage experiments, oocytes were cultured for 14 h in 10 mL RCCS vessels containing M16 medium under SMG, with MLT supplementation at 10 − 9 , 10 − 7 , and 10 − 5 M. Unless otherwise specified, 10 − 7 M MLT was used in subsequent analyses. 10⁻³ M MLT was dissolved in M16 medium and prepared directly at the working concentration. The solution was filter-sterilized and used on the same day. Reversine treatment was performed as previously described, with minor modifications [ 33 – 35 ]. Briefly, after 12 h of SMG exposure under prophase arrest, oocytes were released into M16 medium and allowed to resume meiosis for 12 h. Oocytes that had not extruded the PB1 were then transferred to M16 medium with or without 100 nM reversine and cultured for an additional 2 h. PB1 extrusion was recorded every 0.5 h during this period. Oocytes were fixed in 3.7% paraformaldehyde for 30 min, followed by permeabilization in 0.5% Triton X-100 for 1 h. After blocking in PBS with 1% bovine serum albumin (BSA), oocytes were stained with primary antibodies: γH2AX (1:500, Abcam, ab22551), mouse anti-α-tubulin (1:300, Sigma-Aldrich, F2168), rabbit anti-HSPE1 (1:300, Sigma-Aldrich, SAB4501465), rabbit anti-RAD51 (1:200, Abcam, ab133534), mouse anti-ATM (phospho S1981) (1:100, Santa Cruz Biotechnology, sc-47739), rabbit anti-CHK2 (phospho T68) (1:200, Cell Signaling Technology, 2661T), and rabbit anti-DNA-PKcs (phospho S2056) (1:200, Abcam, ab18192). For RAD51 staining, antigen retrieval was performed before blocking, as previously described [ 36 ]. Briefly, oocytes were incubated with Quick Antigen Retrieval Solution for Frozen Sections (Beyotime, P0090) for 10 min at room temperature, followed by the standard blocking and staining procedures described above. Actin was labeled with Actin-Tracker Red-555 (1:200, Beyotime, C2203S). Secondary antibodies were Alexa-488 goat anti-mouse (1:200, Invitrogen, A11001), Alexa-488 goat anti-rabbit (1:200, Invitrogen, A11008), Alexa-568 donkey anti-mouse (1:300, Invitrogen, A11037) and Alexa-568 donkey anti-rabbit (1:300, Invitrogen, A11042). DNA was counterstained with Hoechst 33,342 (10 µg/mL, Sigma-Aldrich). Images were acquired using Leica SP8 and NEXCOPE NCF1000 confocal microscopes. Images of p-ATM, CHK2, RAD51, and DNA-PKcs staining in GV-stage oocytes were acquired using the NEXCOPE NCF1000 system; all other fluorescence images were acquired using the Leica SP8 confocal microscope. For γH2AX foci quantification, images of the nuclear equatorial plane were acquired using identical imaging parameters across groups. The γH2AX channel was converted to 8-bit images in ImageJ, and the nuclear region was manually defined as the regions of interest (ROI). Background and noise thresholds were determined using the non-nuclear cytoplasmic region, and foci within the nuclear ROI were counted using the Find Maxima function with the same noise tolerance across all groups within the same experiment. The resulting foci counts were exported for statistical analysis. For RAD51, p-ATM, CHK2, and DNA-PKcs quantification, images were acquired at the nuclear equatorial plane using identical imaging parameters across groups. Nuclear ROI were manually defined in ImageJ, and fluorescence intensity within each nuclear ROI was measured. Values were normalized to the mean fluorescence intensity of the NG group before statistical comparison. For HSPE1 quantification, equatorial optical sections of oocytes were acquired using identical imaging parameters across groups. Fluorescence intensity within the oocyte area was measured in ImageJ and normalized to the mean intensity of the NG group. These values are also presented as normalized arbitrary units. GV (2.5 µM milrinone) and metaphase Ⅱ (MⅡ) oocytes were incubated in M16 medium containing 500 nM MitoTracker Green (Thermo Fisher) for 30 min, mounted on glass slides, and analyzed using confocal microscopy (Leica SP8).

SMG exerts heterogeneous, stage-dependent effects on the female reproductive system, reflecting differences in developmental timing and cellular context. In this study, we demonstrate that oocytes arrested at the first meiotic prophase sustain DNA damage upon SMG exposure in the absence of detectable global cellular ROS or mitochondrial superoxide accumulation, and that this early damage compromises their subsequent meiotic maturation capacity. A central finding of this work is that the reparative action of MLT is strongly shaped by the developmental stage at which it is administered. MLT supplementation during SMG exposure reduced DSBs by engaging a DNA-PKcs-associated NHEJ repair response, thereby alleviating maturation arrest and partially restoring developmental competence. By contrast, MLT supplementation after SMG exposure, at the dose effective during prophase arrest, was insufficient to rescue PB1 extrusion, although a higher MLT concentration partially improved maturation outcomes in an NHEJ-sensitive manner. These findings underscore the critical importance of the prophase-arrested stage as a window for effective damage resolution. In parallel, when administered during oocyte maturation under SMG conditions, MLT alleviated mitochondrial dysfunction, reduced spindle assembly defects, facilitated MTOC coalescence, and lowered early apoptosis, and improved oocyte-intrinsic developmental competence after parthenogenetic activation. These effects occurred largely independently of canonical melatonin receptor signaling, highlighting MLT’s capacity to improve oocyte quality under SMG through distinct, stage-specific mechanisms (Fig.  8 ). Fig. 8 Stage-specific effects of SMG on oocytes and protective actions of MLT. Top panel: SMG exposure induces DNA DSBs in GV-stage oocytes without detectable ROS accumulation, leading to subsequent meiotic arrest. During meiotic maturation, SMG disrupts spindle assembly, delays MTOC coalescence, triggers UPR mt overactivation, causes abnormal mitochondrial distribution, and reduced PB1 extrusion. Bottom panel: MLT attenuates SMG-induced DSBs in GV-stage oocytes via an NHEJ-associated repair pathway, thereby alleviating meiotic arrest, largely independent of canonical melatonin receptor (MT1/MT2) signaling. During maturation under SMG conditions, MLT improves spindle assembly, promotes MTOC fusion, suppresses UPR mt overactivation, restores mitochondrial distribution, and reduces early apoptosis, collectively leading to enhanced PB1 extrusion, and improves subsequent developmental competence. Abbreviations: GV, germinal vesicle; MI/MⅡ, metaphase I/Ⅱ,; PB1, first polar body; DSB, double-strand break; MTOC, microtubule-organizing center; UPR mt , mitochondrial unfolded protein response Stage-specific effects of SMG on oocytes and protective actions of MLT. Top panel: SMG exposure induces DNA DSBs in GV-stage oocytes without detectable ROS accumulation, leading to subsequent meiotic arrest. During meiotic maturation, SMG disrupts spindle assembly, delays MTOC coalescence, triggers UPR mt overactivation, causes abnormal mitochondrial distribution, and reduced PB1 extrusion. Bottom panel: MLT attenuates SMG-induced DSBs in GV-stage oocytes via an NHEJ-associated repair pathway, thereby alleviating meiotic arrest, largely independent of canonical melatonin receptor (MT1/MT2) signaling. During maturation under SMG conditions, MLT improves spindle assembly, promotes MTOC fusion, suppresses UPR mt overactivation, restores mitochondrial distribution, and reduces early apoptosis, collectively leading to enhanced PB1 extrusion, and improves subsequent developmental competence. Abbreviations: GV, germinal vesicle; MI/MⅡ, metaphase I/Ⅱ,; PB1, first polar body; DSB, double-strand break; MTOC, microtubule-organizing center; UPR mt , mitochondrial unfolded protein response Our data demonstrate that oocytes arrested in meiotic prophase exhibit pronounced DNA damage upon SMG exposure. Despite lacking robust DNA damage checkpoint responses in oocytes [ 12 ], SMG-induced DSBs markedly impaired subsequent meiotic progression, reflected by reduced GVBD and PB1 extrusion. At the same time, SMG-induced DNA damage was accompanied by activation of an endogenous HR-associated repair response, as evidenced by increased RAD51 levels and the aggravation of oocyte damage after RAD51 inhibition. This is consistent with previous studies showing that GV-stage oocytes preferentially use HR to repair DNA damage [ 17 , 37 ]. Interestingly, inhibition of NHEJ by SCR7 during SMG exposure did not further worsen subsequent maturation outcomes in the absence of MLT, whereas DNA-PKcs levels were reduced under SMG. These findings suggest that, during the initial response to SMG-induced DNA damage, GV-stage oocytes may rely more heavily on HR, while NHEJ activity is relatively limited or insufficiently engaged. This interpretation is also supported by previous evidence that NHEJ deficiency does not necessarily increase DNA damage in GV-stage oocytes under normal developmental conditions, but compromises DNA repair under stress, and that HR cannot fully compensate for the loss of NHEJ [ 48 ]. Similarly, aged GV-stage oocytes with high DNA damage show increased RAD51 expression together with reduced expression of NHEJ-related factors such as XRCC4 and Ligase4 [ 49 ], suggesting that altered HR–NHEJ balance may be a broader feature of stressed or damaged oocytes. In this context, the ability of MLT to facilitate NHEJ engagement under SMG conditions therefore suggests a mechanism by which oocyte DNA repair pathway usage can be modulated in response to environmental stress. These findings are consistent with prior studies showing that MLT enhances repair of etoposide-induced DNA damage in oocytes [ 25 ], reinforcing its function in preserving genomic stability. More broadly, our results suggest that strengthening NHEJ-mediated repair may be both critical and feasible for resolving DNA damage in prophase-arrested oocytes under stress conditions. This may be particularly relevant because primate oocytes and early embryos have relatively limited capacity for DSB repair [ 50 ], making the GV stage and peri-ovulatory window important opportunities for intervention. A key finding of this study is that SMG-induced DSBs in prophase-arrested oocytes occur without detectable accumulation of total cellular ROS or mitochondrial superoxide, indicating that oxidative stress is unlikely to be the primary driver of genotoxic injury at this stage. This stage-specific response may be related to the relatively quiescent cellular state of GV-stage oocytes. Fully grown GV oocytes are transcriptionally inactive or exhibit markedly reduced transcriptional activity [ 51 , 52 ], and their mitochondrial network is comparatively stable, with limited fission–fusion remodeling [ 53 ]. These features may partly explain why SMG exposure during prophase arrest did not elicit measurable ROS accumulation. By contrast, during meiotic maturation, mitochondria undergo marked redistribution toward the spindle region and support the ATP demand required for chromosome segregation and cytoplasmic remodeling, conditions that may increase susceptibility to mitochondrial stress and oxidative imbalance [ 39 , 54 ]. ROS accumulation may also represent a more integrated outcome of prolonged or multifactorial oocyte damage, as observed in aged mouse or human oocytes [ 55 , 56 ], or may arise secondarily when severe DNA damage and repair place additional energetic demands on the cell [ 25 ]. Conversely, oocyte maturation defects can also occur through ROS-independent mechanisms, and reducing ROS does not necessarily further improve PB1 extrusion under some stress conditions [ 57 ]. Accordingly, the beneficial effect of MLT at this stage appears to derive primarily from direct modulation of DNA repair capacity rather than its classical free radical scavenging. Moreover, this effect was not mediated by the canonical melatonin receptors MT1 or MT2. Taken together, these findings support a model in which MLT acts independently of membrane receptor signaling, not only by accessing intracellular compartments such as mitochondria [ 58 , 59 ], but also by directly influencing components of the DNA end-joining machinery, thereby facilitating efficient DSB repair in prophase-arrested oocytes. Once DNA damage had been established by SMG exposure, MLT supplementation during subsequent meiotic maturation required a higher concentration to improve PB1 extrusion, underscoring the spatiotemporal restriction of effective repair-associated intervention during meiotic progression. Although NHEJ can theoretically operate throughout the cell cycle [ 15 , 17 ], recent studies indicate that oocytes undergoing maturation and MⅡ-stage oocytes preferentially rely on NHEJ to repair DNA damage [ 17 , 38 ]. This shift in repair pathway usage may partly explain why post-exposure MLT supplementation during maturation required a higher concentration to produce a measurable improvement. At this stage, endogenous NHEJ activity may already be engaged by meiotic progression, leaving a narrower or more demanding window for exogenous modulation. This observation emphasizes the developmental stage–specific heterogeneity of SMG-induced damage and underscores the need for precisely timed interventions with both optimized timing and dose. This stage dependence also carries implications for assisted reproductive technologies (ART). Specifically, when the goal is to reduce oocyte DNA damage, administration of MLT or other DNA repair–modulating agents may be most effective prior to meiotic resumption, whereas delayed intervention after this window may fail to confer DNA repair–dependent benefits unless the dose and treatment context are carefully optimized. In addition, the efficiency of ROS reduction itself may also be temporally constrained. Consistent with this notion, our previous work in ovarian endometriosis (OEM) demonstrated that oral MLT administration during controlled ovarian stimulation more effectively reduced oocyte ROS levels, whereas MLT exposure during oocyte maturation exerted a comparatively weaker antioxidative effect [ 26 ]. Our prior work demonstrated that SMG accelerates meiotic progression through increased OXPHOS and elevated ΔΨm, thereby compressing the time window in which spindle and MTOC structures can assemble properly and ultimately leading to maturation arrest [ 7 ]. Here, we show that MLT supplementation during meiotic maturation effectively improves spindle assembly and MTOC fusion at the MI stage under SMG conditions, while simultaneously attenuating overactivation of the UPR mt . As a result, PB1 extrusion rates, mitochondrial distribution and SMG-induced mitochondrial hyperpolarization were improved, and the incidence of early apoptosis was reduced. Consistent with prior clinical and animal studies, as well as our own work in OEM, MLT alleviates DNA damage and maturation defects both in vivo and in vitro [ 26 , 60 ]. The current findings extend the potential application of MLT to space biology, where reproductive health is challenged by microgravity. Notably, the protective effects of MLT under SMG were not mediated by the canonical membrane receptors MT1 or MT2. This conclusion was supported by pharmacological activation and inhibition of MT1/MT2 signaling using ramelteon and luzindole, together with analysis of published human and mouse oocyte datasets showing absent or very low expression of MT1- and MT2-encoding genes in oocytes and early developmental stages [ 44 ]. These observations suggest that non-receptor-mediated actions of MLT deserve greater attention in oocytes and early embryos. However, the direct molecular targets responsible for these receptor-independent effects remain unresolved. Identifying stage- and pathway-specific intracellular targets or MLT-interacting proteins may help explain why the optimal effective concentration differs between DNA repair during prophase arrest and mitochondrial–meiotic regulation during maturation. This remains an important direction for future investigation. Notably, UPR mt activation was observed under SMG regardless of whether spindle morphology was overtly abnormal, indicating that mitochondrial stress and spindle defects represent partially independent pathological processes. Accordingly, MLT appears to act on mitochondrial function and spindle assembly through parallel, rather than sequential, mechanisms. A similar dissociation was also suggested by the relationship between spindle morphology and MTOC organization. For example, in some SMG-exposed oocytes, spindle structure appeared relatively preserved, whereas pericentrin-positive MTOCs remained incompletely coalesced at the two spindle poles. Moreover, the extent to which MLT improved spindle abnormalities and aberrant MTOC distribution was not identical, suggesting that spindle architecture and MTOC coalescence may be regulated through related but distinct mechanisms. Further work will be needed to define how MTOC organization, microtubule assembly, and chromosome alignment influence one another under SMG conditions and how MLT intervenes in these processes. This complexity is consistent with the broader view that oocyte quality is shaped by crosstalk among multiple subcellular systems rather than by a single damaged organelle or pathway. For instance, spindle assembly has also been implicated in DNA damage repair, suggesting that cytoskeletal organization and genome maintenance may be functionally connected in oocytes [ 61 ]. In the present study, MLT reduced SMG-induced mitochondrial hyperpolarization. Interestingly, previous studies in other oocyte damage models have reported that MLT can restore reduced mitochondrial membrane potential [ 39 , 62 – 65 ]. Thus, the effect of MLT on mitochondrial membrane potential is not simply unidirectional. Instead, MLT may help stabilize upstream mitochondrial physiology, allowing ΔΨm to return toward a more appropriate functional range depending on the initial direction of mitochondrial disturbance. A limitation of the present study is that the RCCS-based SMG model cannot fully separate simulated gravitational unloading from rotation-associated physical cues. In our previous work, an NG-RCV control showed that the RCV dish itself did not impair GVBD, PB1 extrusion, or oocyte degeneration/fragmentation under normal gravity conditions [ 39 ]. In addition, mouse embryos have been reported to develop from the 2-cell to blastocyst stage in the RCCS system without obvious culture-system-associated arrest [ 8 ]. These observations suggest that vessel-related conditions are unlikely to fully account for the oocyte DNA damage observed here. Nevertheless, low-shear fluid movement or mild mechanical stimulation may still contribute to the RCCS-SMG response. Another unresolved issue is how SMG induces DSBs in the absence of detectable ROS accumulation. Non-oxidative mechanisms, such as altered nuclear architecture, chromatin organization, or DNA damage sensing and repair dynamics, may be involved, but were not directly examined in this study [ 66 ]. In addition, whether SMG alters oocyte biomechanical properties, including membrane/cortical tension, stiffness, or zona pellucida mechanics, remains unknown. Future studies focusing on nuclear structure, chromatin–repair dynamics, and oocyte biomechanical properties will be important for defining the upstream trigger of DNA damage in prophase-arrested oocytes under SMG. We also previously showed that extending M phase with APCin alleviates SMG-induced maturation arrest, but systemic application disrupts cell cycle control, limiting translational potential [ 7 ]. In contrast, MLT is safe for oral administration and has been validated in improving oocyte quality both in vitro and in vivo, with encouraging clinical outcomes [ 67 , 68 ]. Although its efficacy in enhancing PB1 extrusion under SMG was less than that of APCin, its favorable safety profile and clinical feasibility render MLT a more practical candidate for intervention. Furthermore, follicular development and oocyte maturation form a continuous biological process [ 69 ]. Prior studies have shown that culturing primary follicles under SMG impairs oocyte–granulosa cell communication, thereby reducing oocyte quality [ 6 ]. Similarly, our work in OEM demonstrated that disrupted oocyte–cumulus interactions reduce transzonal projection (TZP) formation and impair oocyte quality, effects that were mitigated by MLT [ 26 ]. These parallels suggest that MLT may be broadly effective against microgravity-induced disruptions of follicular development and maturation. Nevertheless, the 12 h and 14 h exposure windows used here should be viewed as acute, stage-defined SMG models rather than direct simulations of long-duration spaceflight. Whether oocyte DNA damage accumulates linearly during chronic microgravity exposure remains unknown, and long-term reproductive outcomes may also be shaped by follicular-stage exposure, endocrine regulation, and systemic physiological adaptation. Moreover, the real space environment involves combined stressors, particularly space radiation, which may act together with microgravity to produce more severe or qualitatively distinct reproductive damage [ 8 , 66 , 70 , 71 ]. Thus, the optimal use of MLT in spaceflight, whether continuous, intermittent, or stage-targeted, will require chronic in vivo and spaceflight-based validation. Beyond reproduction, SMG affects multiple physiological systems, including skeletal muscle, bone, and the central nervous system [ 72 – 75 ]. MLT has been shown to attenuate bone loss and cognitive decline in these contexts [ 31 , 76 – 78 ]. This raises the possibility that MLT could provide systemic protection during spaceflight, complementing its reproductive benefits. Together, these findings outline a stage-specific model of oocyte injury under simulated microgravity. During prolonged prophase arrest, SMG primarily compromises oocyte genome integrity through the induction of DNA damage without detectable global cellular ROS or mitochondrial superoxide accumulation, which can be alleviated by MLT via a DNA-PKcs-associated NHEJ repair response. By contrast, once oocytes enter meiotic maturation, SMG-induced defects manifest predominantly as mitochondrial stress and spindle assembly abnormalities, which are mitigated by MLT through receptor-independent regulation of mitochondrial homeostasis and cytoskeletal organization. By delineating distinct temporal windows of vulnerability and repair competence, this work positions MLT as a stage-dependent modulator of oocyte quality, with translational relevance not only for preserving female reproductive health during spaceflight but also for informing the timing and dosing of MLT-based interventions in clinical reproductive settings.

Human space exploration has entered a new era focused on long-duration missions. The space environment exposes organisms to microgravity and ionizing radiation—two stressors that jeopardize health during long missions and affect multiple physiological systems [ 1 – 3 ]. Whether mammals can successfully reproduce under these conditions remains unresolved. Several non-mammalian species (sea urchins, fish, amphibians) can complete reproduction in microgravity, whereas mammalian oocytes appear considerably more vulnerable [ 4 , 5 ]. Simulated microgravity (SMG) generated by a rotating cell culture system (RCCS) disrupts oocyte–cumulus cell communication and impairs follicular development [ 6 ]. Our previous work further showed that SMG accelerates meiotic progression in mouse oocytes, elevates mitochondrial membrane potential, and triggers the mitochondrial unfolded protein response (UPR mt ), leading to maturation arrest and reduced developmental potential in a subset of oocytes [ 7 ]. Embryos from the 2-cell to blastocyst stages tolerate microgravity relatively well yet remain highly sensitive to low-dose radiation [ 8 ]. These observations suggest that oocytes represent a particularly vulnerable reproductive window under SMG and motivate a focused examination of genome integrity at distinct meiotic stages. Previous studies have assessed the impact of SMG on follicular development using first polar body (PB1) extrusion as an indicator [ 6 ]. Oocyte arrest rates were significantly higher when SMG was applied from earlier stages compared with exposure restricted to the maturation phase, suggesting cumulative injury to immature oocytes, although the underlying mechanism is unclear. SMG has also been reported to induce sperm DNA damage and function as a weak genotoxic stressor that exacerbates genomic instability in cells with compromised DNA-damage responses [ 9 , 10 ]. Mammalian oocytes remain arrested for prolonged periods at the first meiotic prophase and depend on precise coordination with surrounding granulosa cells to sustain development until meiotic resumption. During this stage, they must maintain a dynamic balance between the formation and repair of DNA double-strand breaks (DSBs) [ 11 , 12 ], and excessive or unrepaired DSBs compromise meiotic maturation, fertilization, and subsequent embryogenesis [ 13 ]. Whether SMG perturbs this balance during prophase arrest, thereby predisposing oocytes to later maturation failure, remains unknown. DSBs in oocytes are repaired primarily through homologous recombination (HR) and non-homologous end-joining (NHEJ) [ 14 ]. HR uses a sister chromatid template to restore sequence with high fidelity [ 15 ], whereas NHEJ rejoins DNA ends more rapidly with lower accuracy. Both pathways operate in mammalian oocytes and contribute to genome maintenance across meiotic stages [ 16 , 17 ]. However, it remains unclear whether SMG directly induces DSBs in oocytes, whether specific repair pathways are preferentially engaged, and whether what treatments can effectively mitigate such damage in a stage-dependent manner. Melatonin (MLT), a pineal hormone best known for circadian control, also supports reproduction through receptor-dependent and receptor-independent actions. Through its receptor MT1, MLT supports ovarian activity, delays reproductive decline in female animals [ 18 ] and improves meiotic progression in mouse and goat oocytes [ 19 , 20 ]. Through MT2, MLT promotes proliferation of porcine granulosa cells, inhibits apoptosis, regulates steroidogenesis [ 21 ], and enhances cumulus expansion to support embryonic development [ 22 ]. Independently of its membrane receptors, MLT can enter oocytes, reduce oxidative stress, and limit DNA damage [ 23 , 24 ], including through modulation of NHEJ-mediated repair [ 25 ]. Our prior work showed that MLT mitigates oxidative stress and restores cumulus–oocyte communication in ovarian endometriosis while lowering reactive oxygen species (ROS) levels [ 26 ]. MLT also alleviates UPR–related stress in porcine oocytes and testicular germ cells [ 27 , 28 ]. In addition, accumulating evidence indicates that MLT confers protective effects against SMG-induced cellular injury in diverse systems, including hepatocytes, osteogenic differentiation, and plant growth [ 29 – 32 ], suggesting its potential as a broadly acting countermeasure under microgravity conditions. Together, these findings nominate MLT treatment as a plausible intervention for microgravity-related reproductive injury. In this study, we employed an SMG model to dissect stage-specific effects of microgravity-associated stress on oocyte genome integrity and to test the protective role of MLT, with particular emphasis on the prophase-arrested stage and the subsequent meiotic maturation phase. We demonstrate that SMG induces DNA damage in prophase-arrested oocytes and that MLT mitigates this damage through a DNA-dependent protein kinase catalytic subunit (DNA-PKcs)-associated NHEJ repair response. During meiotic maturation, MLT alleviates mitochondrial stress, improves spindle organization, and promotes microtubule organizing center (MTOC) coalescence, thereby improving maturation competence. Collectively, these findings reveal a stage-dependent pattern of oocyte vulnerability under SMG and position MLT as a promising stage-sensitive intervention to safeguard reproductive health in space environments.

Mitochondrial Superoxide (MitoSOX) levels in GV oocytes were measured using MitoSOX™ Red (500nM, Thermo, M36007). Global intracellular ROS levels in GV oocytes were measured using DCFH-DA (10 µM, Beyotime, S0033S). Oocytes were incubated in M16 medium supplemented with 2.5 µM milrinone and MitoSOX™ Red or DCFH-DA for 30 min, and fluorescence intensities were quantified using confocal microscopy (Leica SP8). Fluorescence intensity within the oocyte boundary was measured in ImageJ by defining the intracellular region as the ROI, and background-corrected fluorescence values were used for statistical analysis. Mitochondrial membrane potential (ΔΨm) in metaphase (MI)-stage oocytes was assessed using JC-1 dye (1:200; Beyotime Biotech). Briefly, oocytes in the NG, SMG, and MLT groups were cultured for 8 h to reach the MI stage and then incubated with JC-1 staining solution for 20 min. After staining, oocytes were washed and imaged using a NEXCOPE NCF1000 confocal microscope under identical acquisition settings across groups. Mitochondrial membrane potential was quantified as the fluorescence intensity ratio of JC-1 aggregates to JC-1 monomers. Apoptosis in GV (2.5 µM milrinone) and MⅡ oocytes was assessed using an Annexin V Staining Kit (C1062M, Beyotime Biotech). Oocytes were incubated in 95 µL binding buffer containing 5 µL Annexin-V-FITC for 20 min, washed, and imaged using a Leica SP8 confocal microscope. For parthenogenetic activation, oocytes that had extruded the first polar body after 14 h of in vitro maturation were selected from the NG, SMG, and MLT groups. In vivo-matured oocytes were used as an additional control to validate the activation system. Briefly, 3-week-old C57BL/6J mice were subcutaneously injected with 5 IU PMSG, followed by intraperitoneal injection of 5 IU human chorionic gonadotropin (hCG) 44–48 h later. After 14 h, COCs were collected from the oviductal ampullae, and cumulus cells were removed using hyaluronidase. Parthenogenetic activation was performed using a commercial activation kit (Nanjing Aibei, M2755). Briefly, oocytes were activated for 4 h in calcium-free CZB medium containing 10 mM SrCl₂·6 H₂O and 10 µM cytochalasin B (CB), and then transferred to KSOM medium (Sigma, MR-101-D) for further culture. The number of 2-cell embryos was recorded 24 h after activation, and the activation rate was calculated using the number of MⅡ oocytes subjected to activation as the denominator. Blastocyst formation was recorded at 96 h, and the blastocyst rate was calculated using the number of 2-cell embryos as the denominator. Data are presented as mean ± standard deviation (SD). Statistical analyses were performed using Prism 9 (GraphPad, USA). For comparisons between two groups, an unpaired two-tailed Student’s t-test was used. For comparisons among three or more groups, data normality was assessed when applicable using the Shapiro–Wilk test. When the normality assumption was met, one-way ANOVA followed by Tukey’s multiple-comparisons test was applied. The number of oocytes or embryos analyzed and the number of independent replicates for each experiment are specified in the corresponding figure legends. Statistical significance was indicated as ns, not significant; * p  < 0.05, ** p  < 0.01, *** p  < 0.001, **** p  < 0.0001.

Supplementary Material 1. Supplementary Material 1.