PRMT3-mediated FOXO1 arginine methylation exacerbates oxidative stress-induced decidualization defects in the eutopic endometrium of endometriosis

To investigate the relationship between PRMT3 and decidualization defects associated with endometriosis, we performed IHC and Western blot analysis to examine the expression levels of PRMT3 in patients with endometriosis. The results showed that, compared with secretory endometrium in normal individuals, PRMT3 expression was increased and FOXO1 expression was decreased in EuEM of patients with endometriosis (Fig.  1 A-C). Similarly, compared with proliferative endometrium in normal individuals, PRMT3 expression was increased and FOXO1 expression was decreased in EcEM of patients with endometriosis (Fig. S1 A-C). Interestingly, correlation analysis showed that PRMT3 and FOXO1 were negatively correlated in both EuEM and EcEM (Fig.  1 D and Fig. S1 D, respectively). Fig. 1 PRMT3 expression is increased in EuEM of endometriosis and negatively correlated with FOXO1. ( A ) IHC was used to determine the protein expression levels of PRMT3 and FOXO1 in secretory endometrium of normal individuals ( n  = 30) and in EuEM of patients with endometriosis ( n  = 30) (scale bar = 20 μm) (B) IHC H-score of PRMT3 and FOXO1 in secretory endometrium of normal individuals and in EuEM of patients with endometriosis ( C ) Western blot analysis was used to detect the protein expression levels of PRMT3 and FOXO1 in secretory endometrium of normal individuals ( n  = 5) and in EuEM of patients with endometriosis ( n  = 5) ( D ) Spearman correlation analysis was used to assess the correlation between PRMT3 and FOXO1 expression levels in the EuEM of endometriosis ( n  = 30) (The data in ( B ) were presented as median and quartiles. The normality and variance tests, Mann–Whitney test, and Spearman’s correlation analysis were used for data analysis. * p  < 0.05, ** p  < 0.01, *** p  < 0.001, **** p  < 0.0001) PRMT3 expression is increased in EuEM of endometriosis and negatively correlated with FOXO1. ( A ) IHC was used to determine the protein expression levels of PRMT3 and FOXO1 in secretory endometrium of normal individuals ( n  = 30) and in EuEM of patients with endometriosis ( n  = 30) (scale bar = 20 μm) (B) IHC H-score of PRMT3 and FOXO1 in secretory endometrium of normal individuals and in EuEM of patients with endometriosis ( C ) Western blot analysis was used to detect the protein expression levels of PRMT3 and FOXO1 in secretory endometrium of normal individuals ( n  = 5) and in EuEM of patients with endometriosis ( n  = 5) ( D ) Spearman correlation analysis was used to assess the correlation between PRMT3 and FOXO1 expression levels in the EuEM of endometriosis ( n  = 30) (The data in ( B ) were presented as median and quartiles. The normality and variance tests, Mann–Whitney test, and Spearman’s correlation analysis were used for data analysis. * p  < 0.05, ** p  < 0.01, *** p  < 0.001, **** p  < 0.0001) To further examine the functions of PRMT3 in endometriosis, we conducted relevant experiments in vitro and in vivo. Firstly, we conducted Western blot analysis to validate the overexpression and knockdown effects of PRMT3 in EESCs and 12Z cells. The results showed that after transfection of Flag-PRMT3 into EESCs, the protein expression level of Flag-PRMT3 in cells was significantly increased, while after transfection of siPRMT3, the protein expression level of PRMT3 in cells was significantly reduced (Figure S2 A). Similarly, the same results were observed in 12Z cells (Figure S3 A). The above indicated that we have successfully overexpressed and knocked down PRMT3 in EESCs and 12Z cells. The results of cell proliferation assays showed that overexpression of PRMT3 promoted the proliferation of EESCs and 12Z cells, while knockdown of PRMT3 inhibited cell proliferation (Fig. S2 B and Fig. S3 B, respectively). Similarly, colony formation assays also showed that overexpression of PRMT3 promoted cell proliferation, while knockdown of PRMT3 inhibited cell proliferation (Fig. S2 C and Fig. S3 C, respectively). In addition, migration assays demonstrated that overexpression of PRMT3 promoted the migration of EESCs and 12Z cells, while knockdown of PRMT3 inhibited cell migration (Fig. S2 D and Fig. S3 D, respectively). Wound healing assays demonstrated that overexpression of PRMT3 promoted cell migration, while knockdown of PRMT3 inhibited cell migration (Fig. S2 E and Fig. S3 E, respectively). SGC707 is a potent cell activity allosteric inhibitor of PRMT3. Supplementary Figure S4 A showed the IC50 of SGC707 in EESCs and 12Z cells. We treated cells with SGC707 and conducted proliferation assays, colony formation assays, migration assays, and wound healing assays.The proliferation and colony formation assays showed that SGC707 inhibited the proliferation ability of EESCs and 12Z cells (Fig. S4 B-C). Migration assays and wound healing assays showed that SGC707 inhibited the migration ability of EESCs and 12Z cells (Fig. S4 D-E). Next, we established an animal model of endometriosis (Fig. S5 A). The experimental group was intraperitoneally injected with SGC707 to evaluate the effect of PRMT3 on the progression of endometriosis in vivo. While, there was no difference in weight between the experimental group and the control group mice (Fig. S5 B-C), the number, volume, and weight of endometriosis lesions from mice treated with SGC707 were significantly lower compared to the control group (Fig. S5 D-G). We performed HE staining and IHC for PRMT3 and FOXO1 in ectopic lesions. The results showed that there was no difference in the expression of PRMT3 between the experimental group and the control group, but the expression of FOXO1 in the experimental group was significantly higher than that in the control group (Fig. S5 H). We performed qRT-PCR and found that the expression of decidualization markers PRL and IGFBP1 was elevated in HESCs, demonstrating that we successfully induced decidualization of HESCs in vitro (Fig.  2 A). After in vitro decidualization induction, we found that the mRNA expression of PRMT3 decreased while FOXO1 expression increased in dHESCs (Fig.  2 B). Then, we overexpressed PRMT3 in HESCs and treated them with 8Br-cAMP + MPA for 4 days. The results showed that, compared with the control group, the mRNA expression of PRL, IGFBP1, and FOXO1 was significantly decreased. Conversely, knockdown of PRMT3 resulted in the opposite effect (Fig.  2 C-D). We also performed IF and the results showed that PRL and IGFBP1 were upregulated in HESCs, and the cell volume increased, demonstrating our successful induction of dHESCs in vitro (Fig.  2 E). The results of IF showed that overexpression of PRMT3 inhibited cellular morphological changes, reduced the expression of PRL and IGFBP1, while knocking down PRMT3 resulted in the contrary effect (Fig.  2 F-G). Then, we treated HESCs with SGC707, and the IC 50 results were presented in Supplementary Figure S6 A. qRT-PCR and IF showed that SGC707 promoted the occurrence of decidualization (Fig. S6 B-C). Fig. 2 PRMT3 inhibits decidualization of HESCs in vitro. ( A - B ) qRT-PCR showed mRNA expression levels of PRL, IGFBP1, PRMT3 and FOXO1 in HESCs treated with or without 8Br-cAMP + MPA ( C - D ) PRMT3 was overexpressed or knocked down in HESCs, then treated with 8Br-cAMP + MPA for 4 days. qRT-PCR showed mRNA expression of PRL, IGFBP1 and FOXO1 ( E ) IF showed the cellular morphological changes and expression of PRL (green) and IGFBP1 (green) in HESCs treated with or without 8Br-cAMP + MPA ( F - G ) PRMT3 was overexpressed or knocked down in HESCs, then treated with 8Br-cAMP + MPA for 4 days. IF showed the cellular morphological changes and expression of PRL (green) and IGFBP1 (green) (Data represent mean ± SEM. The normality and variance tests, Student’s t tests were used for data analysis. * P  < 0.05, ** P  < 0.01, *** P  < 0.001, **** P  < 0.0001) PRMT3 inhibits decidualization of HESCs in vitro. ( A - B ) qRT-PCR showed mRNA expression levels of PRL, IGFBP1, PRMT3 and FOXO1 in HESCs treated with or without 8Br-cAMP + MPA ( C - D ) PRMT3 was overexpressed or knocked down in HESCs, then treated with 8Br-cAMP + MPA for 4 days. qRT-PCR showed mRNA expression of PRL, IGFBP1 and FOXO1 ( E ) IF showed the cellular morphological changes and expression of PRL (green) and IGFBP1 (green) in HESCs treated with or without 8Br-cAMP + MPA ( F - G ) PRMT3 was overexpressed or knocked down in HESCs, then treated with 8Br-cAMP + MPA for 4 days. IF showed the cellular morphological changes and expression of PRL (green) and IGFBP1 (green) (Data represent mean ± SEM. The normality and variance tests, Student’s t tests were used for data analysis. * P  < 0.05, ** P  < 0.01, *** P  < 0.001, **** P  < 0.0001) In order to further investigate the effect of PRMT3 on decidualization in endometriosis in vivo, we artificially induced decidualization in endometriosis mice (Fig.  3 A). We injected sesame oil into the uterine horn of one side of the mouse to form a deciduoma. The results showed that the volume and weight of deciduomas in the experimental group were both greater than those in the control group (Fig.  3 B). Subsequently, we performed HE staining on the unstimulated and stimulated uterus, then we found that the experimental group had larger volume and rounder morphology of decidua cells (Fig.  3 C). And IHC showed that the decidualization markers PRL and IGFBP1 were also upregulated in the experimental group (Fig.  3 D). Then, we conducted embryo formation experiments in mice with endometriosis (Fig.  3 E). Compared with the control group, both the litter size and fetal weight of the experimental group mice were increased, indicating that inhibition of PRMT3 leaded to enhanced embryonic development (Fig.  3 F-G). Fig. 3 PRMT3 inhibits decidualization and affects embryonic development in vivo. ( A ) Artificially induced deciduoma model and medication schedule ( B ) Images of deciduomas and the weight ratio of stimulated and unstimulated uterus ( n  = 7) ( C ) HE staining of unstimulated uterus and stimulated uterus (deciduomas) in the control group and experimental group (scale bar = 20 μm) ( n  = 7) ( D ) Representative IHC micrographs and H-scores of PRL and IGFBP1 in the deciduomas of the control group and experimental group (scale bar = 20 μm) ( n  = 7) ( E ) Embryo formation model and medication schedule ( F ) Images of embryos in the control group and experimental group ( n  = 7) (All animal experiments conducted in this study were approved by the Experimental Animal Ethics Review Committee of Shandong Second Medical University. Six-week-old female ( n  = 42) and eight-week-old male ( n  = 8) C57BL/6 mice were used in the experiment.) Litter size and fetal weight of the control group and experimental group ( n  = 7) (Data represent mean ± SEM. The normality and variance tests, Student’s t tests were used for data analysis. * P  < 0.05, ** P  < 0.01, *** P  < 0.001, **** P  < 0.0001) PRMT3 inhibits decidualization and affects embryonic development in vivo. ( A ) Artificially induced deciduoma model and medication schedule ( B ) Images of deciduomas and the weight ratio of stimulated and unstimulated uterus ( n  = 7) ( C ) HE staining of unstimulated uterus and stimulated uterus (deciduomas) in the control group and experimental group (scale bar = 20 μm) ( n  = 7) ( D ) Representative IHC micrographs and H-scores of PRL and IGFBP1 in the deciduomas of the control group and experimental group (scale bar = 20 μm) ( n  = 7) ( E ) Embryo formation model and medication schedule ( F ) Images of embryos in the control group and experimental group ( n  = 7) (All animal experiments conducted in this study were approved by the Experimental Animal Ethics Review Committee of Shandong Second Medical University. Six-week-old female ( n  = 42) and eight-week-old male ( n  = 8) C57BL/6 mice were used in the experiment.) Litter size and fetal weight of the control group and experimental group ( n  = 7) (Data represent mean ± SEM. The normality and variance tests, Student’s t tests were used for data analysis. * P  < 0.05, ** P  < 0.01, *** P  < 0.001, **** P  < 0.0001) FOXO1 is a key gene for decidualization. Therefore, we used FOXO1 as a bait protein and discovered through IP/MS technology that PRMT3 is a new interacting protein of FOXO1 in HESCs (Fig.  4 A). To further explore the regulatory mechanism, we conducted Co-IP assays in HEK293T cells and HESCs, and found that there was a direct interaction between PRMT3 and FOXO1 (Fig.  4 B-C). In addition, Fig.  4 D showed that PRMT3 was mainly located in the cytoplasm of HESCs and 12Z cells. FOXO1, as a transcription factor, is expressed in both the nucleus and cytoplasm, and our experimental results were consistent with this distribution (Fig.  4 D). The above results indicated that PRMT3 and FOXO1 were co-localized in the cytoplasm of HESCs and 12Z cells. This further confirmed the direct interaction between PRMT3 and FOXO1. Fig. 4 PRMT3 interacts with FOXO1. ( A ) Mass spectrometry analysis was performed on FOXO1 in HESCs ( B ) HEK293T cells were transfected with Flag-PRMT3 and HA-FOXO1. The interaction between exogenous PRMT3 and FOXO1 was verified through Co-IP and Western blot analysis using specific antibodies ( C ) The interaction between endogenous PRMT3 and FOXO1 in HESCs was verified by performing Co-IP and Western blot analysis using specific antibodies ( D ) IF was performed to determine the colocalization of PRMT3 (green) and FOXO1 (red) in HESCs and 12Z cells PRMT3 interacts with FOXO1. ( A ) Mass spectrometry analysis was performed on FOXO1 in HESCs ( B ) HEK293T cells were transfected with Flag-PRMT3 and HA-FOXO1. The interaction between exogenous PRMT3 and FOXO1 was verified through Co-IP and Western blot analysis using specific antibodies ( C ) The interaction between endogenous PRMT3 and FOXO1 in HESCs was verified by performing Co-IP and Western blot analysis using specific antibodies ( D ) IF was performed to determine the colocalization of PRMT3 (green) and FOXO1 (red) in HESCs and 12Z cells Given that PRMT3 is an arginine methyltransferase, we explored whether PRMT3 can directly methylate FOXO1. We overexpressed PRMT3 and FOXO1 in HEK293T cells, and Co-IP assay demonstrated that PRMT3 directly methylated FOXO1 (Fig. 5 A). On the contrary, knockdown PRMT3 in HESCs reduced methylation modifications (Fig. 5 B). According to reports, PRMT3 is specific for the substrate-arginine RXR motif [ 38 ]. We analyzed the amino acid sequence of FOXO1 and found that it contains several RXR motifs. Then, combined with GPS-MSP web site analysis ( http://msp.biocuckoo.org/ ), we predicted the arginine sites (R19, R21, R251, R252, R253, R267, R269, R314, R316) (Fig. 5 C). We constructed FOXO1 arginine site mutants and transfected them into HEK293T cells and then used Co-IP to detect ADMA modification signals. The results showed that in cells transfected with the R253K mutant plasmid, the ADMA signal of FOXO1 protein was significantly reduced, but there was no significant change in the ADMA signals of other cells (Fig. 5 D-E). Fig. 5 PRMT3 methylates FOXO1 at the Arg253 site. ( A ) HEK293T cells were transfected with Flag-FOXO1 and HA-PRMT3 + Flag-FOXO1, respectively. Then, the effect of PRMT3 on FOXO1 methylation was validated using specific antibodies for Co-IP and Western blot analysis ( B ) HESCs were transfected with siPRMT3. Then, the effect of endogenous PRMT3 on FOXO1 methylation was validated using specific antibodies for Co-IP and Western blot analysis ( C ) Identify the potential methylation sites of PRMT3 in the FOXO1 protein sequence ( D ) HEK293T cells were transfected with HA-PRMT3 + Flag-FOXO1 (WT) and HA-PRMT3 + Flag-FOXO1 (mutant), respectively. Then, the effect of PRMT3 on FOXO1 (WT or mutant) methylation was validated using specific antibodies for Co-IP and Western blot analysis ( E ) HEK293T cells were transfected with HA-PRMT3 + Flag-FOXO1 (WT) and HA-PRMT3 + Flag-FOXO1 (R253K), respectively. Then, the effect of PRMT3 on FOXO1 (WT or R253K) methylation was validated using specific antibodies for Co-IP and Western blot analysis PRMT3 methylates FOXO1 at the Arg253 site. ( A ) HEK293T cells were transfected with Flag-FOXO1 and HA-PRMT3 + Flag-FOXO1, respectively. Then, the effect of PRMT3 on FOXO1 methylation was validated using specific antibodies for Co-IP and Western blot analysis ( B ) HESCs were transfected with siPRMT3. Then, the effect of endogenous PRMT3 on FOXO1 methylation was validated using specific antibodies for Co-IP and Western blot analysis ( C ) Identify the potential methylation sites of PRMT3 in the FOXO1 protein sequence ( D ) HEK293T cells were transfected with HA-PRMT3 + Flag-FOXO1 (WT) and HA-PRMT3 + Flag-FOXO1 (mutant), respectively. Then, the effect of PRMT3 on FOXO1 (WT or mutant) methylation was validated using specific antibodies for Co-IP and Western blot analysis ( E ) HEK293T cells were transfected with HA-PRMT3 + Flag-FOXO1 (WT) and HA-PRMT3 + Flag-FOXO1 (R253K), respectively. Then, the effect of PRMT3 on FOXO1 (WT or R253K) methylation was validated using specific antibodies for Co-IP and Western blot analysis To further elucidate the regulatory mechanism of PRMT3 on FOXO1, we investigated whether PRMT3 affects the protein stability of FOXO1. Firstly, we transfected Flag-PRMT3 and HA-FOXO1 into HEK293T cells, and we found transfection of Flag-PRMT3 reduced the protein stability of HA-FOXO1 (Fig.  6 A). Interestingly, as the transfection amount of Flag-PRMT3 increases, the protein level of HA-FOXO1 decreases more and more (Fig.  6 B). Afterwards, we assessed the protein stability of endogenous FOXO1 in HESCs and 12Z cells. We found that overexpression of PRMT3 in HESCs and 12Z cells reduced endogenous FOXO1 protein expression, whereas knockdown of PRMT3 in HESCs and 12Z cells increased FOXO1 protein expression (Fig.  6 C-F). Similarly, when Flag-PRMT3 and HA-FOXO1 (R253K) were transfected into HEK293T cells, we found that overexpression of PRMT3 did not reduce FOXO1 protein in cells transfected with HA-FOXO1 (R253K) plasmid (Fig.  6 G-H). Finally, through CHX experiments, we found that PRMT3 reduced the stability of FOXO1 by shortening its half-life (Fig.  6 I). To determine whether PRMT3 affects the stability of FOXO1 through the proteasome pathway, we overexpressed PRMT3 in HESCs and treated them with MG132. The results showed that MG132 reversed the PRMT3-induced downregulation of FOXO1 (Fig.  6 J). Next, we verified whether PRMT3 regulated the ubiquitination level of FOXO1. We found that PRMT3 overexpression in HEK293T cells promoted FOXO1 ubiquitination (Fig.  6 K). Subsequently, we transfected HA-PRMT3 and Flag-FOXO1 (WT, R253K) into HEK293T cells and performed ubiquitination assays. Interestingly, cells transfected with FOXO1 (R253K) exhibited lower ubiquitination levels compared to those transfected with FOXO1 (WT) (Fig.  6 L). Collectively, these findings indicated that PRMT3 promoted the degradation of FOXO1 at the Arg253 site through the ubiquitin proteasome pathway. In addition, IF results indicated that knocking down PRMT3 in HESCs and 12Z cells leaded to an increase in FOXO1 expression levels and FOXO1 exhibited significant nuclear translocation (Fig.  6 M-N). Fig. 6 PRMT3 downregulates the stability of FOXO1 protein. ( A ) HEK293T cells were transfected with HA-FOXO1 and Flag-PRMT3 + HA-FOXO1, respectively, followed by Western blot analysis to measure the expression of HA-FOXO1 ( B ) HEK293T cells were transfected with HA-FOXO1, Flag-PRMT3 (0.5 µg) + HA-FOXO1 and Flag-PRMT3 (1 µg) + HA-FOXO1, respectively, followed by Western blot analysis to measure the expression of HA-FOXO1 ( C - D ) HESCs were transfected with Flag-PRMT3 or siPRMT3, followed by Western blot analysis to measure the expression of endogenous PRMT3 and FOXO1 ( E - F ) 12Z cells were transfected with Flag-PRMT3 or siPRMT3, followed by Western blot analysis to measure the expression of endogenous PRMT3 and FOXO1 ( G ) HEK293T cells were transfected with HA-FOXO1 (WT), Flag-PRMT3 + HA-FOXO1 (WT), HA-FOXO1 (R253K) and Flag-PRMT3 + HA-FOXO1 (R253K), respectively, followed by Western blot analysis to measure the expression of HA-FOXO1 ( H ) HEK293T cells were transfected with HA-FOXO1 (WT), Flag-PRMT3 + HA-FOXO1 (WT) and Flag-PRMT3 + HA-FOXO1 (R253K), respectively, followed by Western blot analysis to measure the expression of HA-FOXO1 ( I ) HEK293T cells were transfected with HA-FOXO1 and Flag-PRMT3 + HA-FOXO1, respectively, then treated with or without CHX (100 µg/mL) for the indicated times, followed by Western blot analysis to measure the expression of HA-FOXO1 ( J ) HESCs were transfected with Flag-PRMT3, then treated with or without MG132 (100 µmol/L) for 8 h, followed by Western blot analysis to measure the expression of endogenous FOXO1 ( K ) HEK293T cells were transfected with Flag-FOXO1 + HA-UB and HA-PRMT3 + Flag-FOXO1 + HA-UB, respectively. After 48 h, the cells were treated with MG132 (100 µmol/L) for 8 h, followed by IP and Western blot analysis to detect the ubiquitination levels ( L ) HEK293T cells were transfected with HA-PRMT3 + Flag-FOXO1 (WT)+HA-UB and HA-PRMT3 + Flag-FOXO1 ༈R253K༉+HA-UB, respectively. After 48 h, the cells were treated with MG132 (100 µmol/L) for 8 h, followed by IP and Western blot analysis to detect the ubiquitination levels ( M ) IF showed the effect of knocking down PRMT3 (green) on the expression and localization of FOXO1 (red) in HESCs and 12Z cells ( N ) Quantitative analysis of the fluorescence intensity of PRMT3 and FOXO1 (Data represent mean ± SEM. The normality and variance tests, Student’s t tests were used for data analysis. * P  < 0.05, ** P  < 0.01, *** P  < 0.001, **** P  < 0.0001) PRMT3 downregulates the stability of FOXO1 protein. ( A ) HEK293T cells were transfected with HA-FOXO1 and Flag-PRMT3 + HA-FOXO1, respectively, followed by Western blot analysis to measure the expression of HA-FOXO1 ( B ) HEK293T cells were transfected with HA-FOXO1, Flag-PRMT3 (0.5 µg) + HA-FOXO1 and Flag-PRMT3 (1 µg) + HA-FOXO1, respectively, followed by Western blot analysis to measure the expression of HA-FOXO1 ( C - D ) HESCs were transfected with Flag-PRMT3 or siPRMT3, followed by Western blot analysis to measure the expression of endogenous PRMT3 and FOXO1 ( E - F ) 12Z cells were transfected with Flag-PRMT3 or siPRMT3, followed by Western blot analysis to measure the expression of endogenous PRMT3 and FOXO1 ( G ) HEK293T cells were transfected with HA-FOXO1 (WT), Flag-PRMT3 + HA-FOXO1 (WT), HA-FOXO1 (R253K) and Flag-PRMT3 + HA-FOXO1 (R253K), respectively, followed by Western blot analysis to measure the expression of HA-FOXO1 ( H ) HEK293T cells were transfected with HA-FOXO1 (WT), Flag-PRMT3 + HA-FOXO1 (WT) and Flag-PRMT3 + HA-FOXO1 (R253K), respectively, followed by Western blot analysis to measure the expression of HA-FOXO1 ( I ) HEK293T cells were transfected with HA-FOXO1 and Flag-PRMT3 + HA-FOXO1, respectively, then treated with or without CHX (100 µg/mL) for the indicated times, followed by Western blot analysis to measure the expression of HA-FOXO1 ( J ) HESCs were transfected with Flag-PRMT3, then treated with or without MG132 (100 µmol/L) for 8 h, followed by Western blot analysis to measure the expression of endogenous FOXO1 ( K ) HEK293T cells were transfected with Flag-FOXO1 + HA-UB and HA-PRMT3 + Flag-FOXO1 + HA-UB, respectively. After 48 h, the cells were treated with MG132 (100 µmol/L) for 8 h, followed by IP and Western blot analysis to detect the ubiquitination levels ( L ) HEK293T cells were transfected with HA-PRMT3 + Flag-FOXO1 (WT)+HA-UB and HA-PRMT3 + Flag-FOXO1 ༈R253K༉+HA-UB, respectively. After 48 h, the cells were treated with MG132 (100 µmol/L) for 8 h, followed by IP and Western blot analysis to detect the ubiquitination levels ( M ) IF showed the effect of knocking down PRMT3 (green) on the expression and localization of FOXO1 (red) in HESCs and 12Z cells ( N ) Quantitative analysis of the fluorescence intensity of PRMT3 and FOXO1 (Data represent mean ± SEM. The normality and variance tests, Student’s t tests were used for data analysis. * P  < 0.05, ** P  < 0.01, *** P  < 0.001, **** P  < 0.0001) As is well known, OS damages decidualization, and our data showed that PRMT3 also damaged decidualization. Therefore, we speculated whether PRMT3 may be related to OS. We transfected Flag-PRMT3 into HESCs and used corresponding assay kits to detect the content of the oxidative stress indicator MDA and the activity of the antioxidant stress indicators SOD and CAT. The results showed that overexpression of PRMT3 increased MDA content and decreased SOD and CAT activity (Fig.  7 A). On the contrary, knockdown of PRMT3 or treatment with the PRMT3 inhibitor SGC707 in HESCs led to decreased MDA content and increased activities of SOD and CAT (Fig.  7 B-C). This indicated that PRMT3 promoted OS. Considering that FOXO1 is an important antioxidant factor and our data suggested that PRMT3 reduced the protein stability of FOXO1, we speculated whether the degradation of FOXO1 may be involved in the impact of PRMT3 on OS. Therefore, we transfected HA-FOXO1 into HESCs on the basis of overexpression of PRMT3. We found that FOXO1 reduced MDA content while restoring SOD and CAT activity (Fig.  7 D). Based on these data, we further investigated whether PRMT3 affects cell decidualization through the OS. We treated dHESCs with the antioxidant NAC (10 mM). The results of qRT-PCR and IF showed that cells treated with NAC exhibited larger volumes, rounder shapes, and higher expression levels of PRL and IGFBP1 compared to the group without NAC, indicating that NAC restored the inhibitory effect of PRMT3 on decidualization (Fig.  7 E-F). Fig. 7 PRMT3 enhances oxidative stress-induced decidualization defects by regulating FOXO1. ( A ) HESCs were transfected with empty vector or Flag-PRMT3, then detected the oxidative stress indicator MDA and the antioxidant stress indicators SOD and CAT ( B ) HESCs were transfected with siNC or siPRMT3, then detected the oxidative stress indicator MDA and the antioxidant stress indicators SOD and CAT ( C ) HESCs were treated with or without SGC707, then detected the oxidative stress indicator MDA and the antioxidant stress indicators SOD and CAT ( D ) HESCs were transfected with Flag-PRMT3 and Flag-PRMT3 + HA-FOXO1, respectively, then detected the oxidative stress indicator MDA and the antioxidant stress indicators SOD and CAT ( E ) HESCs were transfected with empty vector or Flag-PRMT3 and treated with 8Br-cAMP + MPA for 4 days, and were treated with or without NAC (10mM) concurrently. qRT-PCR showed the mRNA expression of PRL and IGFBP1 ( F ) HESCs were transfected with empty vector or Flag-PRMT3 and treated with 8Br-cAMP + MPA for 4 days, and were treated with or without NAC (10mM) concurrently. IF showed the cellular morphological changes and expression of PRL (green) and IGFBP1 (green) (Data represent mean ± SEM. The normality and variance tests, Student’s t tests, one-way ANOVA were used for data analysis. * P  < 0.05, ** P  < 0.01, *** P  < 0.001, **** P  < 0.0001) PRMT3 enhances oxidative stress-induced decidualization defects by regulating FOXO1. ( A ) HESCs were transfected with empty vector or Flag-PRMT3, then detected the oxidative stress indicator MDA and the antioxidant stress indicators SOD and CAT ( B ) HESCs were transfected with siNC or siPRMT3, then detected the oxidative stress indicator MDA and the antioxidant stress indicators SOD and CAT ( C ) HESCs were treated with or without SGC707, then detected the oxidative stress indicator MDA and the antioxidant stress indicators SOD and CAT ( D ) HESCs were transfected with Flag-PRMT3 and Flag-PRMT3 + HA-FOXO1, respectively, then detected the oxidative stress indicator MDA and the antioxidant stress indicators SOD and CAT ( E ) HESCs were transfected with empty vector or Flag-PRMT3 and treated with 8Br-cAMP + MPA for 4 days, and were treated with or without NAC (10mM) concurrently. qRT-PCR showed the mRNA expression of PRL and IGFBP1 ( F ) HESCs were transfected with empty vector or Flag-PRMT3 and treated with 8Br-cAMP + MPA for 4 days, and were treated with or without NAC (10mM) concurrently. IF showed the cellular morphological changes and expression of PRL (green) and IGFBP1 (green) (Data represent mean ± SEM. The normality and variance tests, Student’s t tests, one-way ANOVA were used for data analysis. * P  < 0.05, ** P  < 0.01, *** P  < 0.001, **** P  < 0.0001) To investigate whether FOXO1 plays a role at its Arg253 site. We transfected HA-FOXO1 (WT) and HA-FOXO1 (R253K) separately into EESCs for wound healing assay, migration assay, colony formation assay, and cell proliferation assay. The results showed that the proliferation and migration abilities of cells transfected with HA-FOXO1 (R253K) were similar to those of the control group, while the proliferation and migration abilities of cells transfected with HA-FOXO1 (WT) were inhibited (Fig. S7 A-D). The same conclusion was also confirmed in 12Z cells (Fig. S7 A-D). Afterwards, through qRT-PCR and IF experiments in HESCs, we found that cells transfected with HA-FOXO1 (R253K) expressed less PRL and IGFBP1 than those transfected with HA-FOXO1 (WT) in the control group, and the morphological changes of decidual cells were also smaller (Fig.  8 A-B). Fig. 8 PRMT3 inhibits decidualization through the Arg253 site of FOXO1 ( A ) FOXO1 was knocked down with siFOXO1, and then HA-FOXO1 (empty vector or WT or R253K) was transfected in HESCs, and treated with 8Br-cAMP + MPA for 4 days. qRT-PCR showed mRNA expression of PRL and IGFBP1 ( B ) FOXO1 was knocked down with siFOXO1, and then HA-FOXO1 (empty vector or WT or R253K) was transfected in HESCs, and treated with 8Br-cAMP + MPA for 4 days. IF showed the cellular morphological changes and expression of PRL (green) and IGFBP1 (green) ( C ) HESCs were transfected with empty vector, Flag-PRMT3 and Flag-PRMT3 + HA-FOXO1, respectively, and treated with 8Br-cAMP + MPA for 4 days. Then qRT-PCR showed mRNA expression of PRL and IGFBP1 ( D ) HESCs were transfected with empty vector, Flag-PRMT3 and Flag-PRMT3 + HA-FOXO1, respectively, and treated with 8Br-cAMP + MPA for 4 days. Then IF showed the cellular morphological changes and expression of PRL (green) and IGFBP1 (green) ( E ) The mechanism by which PRMT3 promotes endometriosis and inhibits decidualization (Data represent mean ± SEM. The normality and variance tests, one-way ANOVA were used for data analysis. * P  < 0.05, ** P  < 0.01, *** P  < 0.001, **** P  < 0.0001) PRMT3 inhibits decidualization through the Arg253 site of FOXO1 ( A ) FOXO1 was knocked down with siFOXO1, and then HA-FOXO1 (empty vector or WT or R253K) was transfected in HESCs, and treated with 8Br-cAMP + MPA for 4 days. qRT-PCR showed mRNA expression of PRL and IGFBP1 ( B ) FOXO1 was knocked down with siFOXO1, and then HA-FOXO1 (empty vector or WT or R253K) was transfected in HESCs, and treated with 8Br-cAMP + MPA for 4 days. IF showed the cellular morphological changes and expression of PRL (green) and IGFBP1 (green) ( C ) HESCs were transfected with empty vector, Flag-PRMT3 and Flag-PRMT3 + HA-FOXO1, respectively, and treated with 8Br-cAMP + MPA for 4 days. Then qRT-PCR showed mRNA expression of PRL and IGFBP1 ( D ) HESCs were transfected with empty vector, Flag-PRMT3 and Flag-PRMT3 + HA-FOXO1, respectively, and treated with 8Br-cAMP + MPA for 4 days. Then IF showed the cellular morphological changes and expression of PRL (green) and IGFBP1 (green) ( E ) The mechanism by which PRMT3 promotes endometriosis and inhibits decidualization (Data represent mean ± SEM. The normality and variance tests, one-way ANOVA were used for data analysis. * P  < 0.05, ** P  < 0.01, *** P  < 0.001, **** P  < 0.0001) Then, we verified whether the promoting effect of PRMT3 on the biological behavior of endometriosis and its inhibitory effect on decidualization depend on FOXO1. Firstly, we transfected Flag-PRMT3 and HA-FOXO1 into EESCs and 12Z cells. The results of migration assay, wound healing assay, colony formation assay, and cell proliferation assay showed that PRMT3 promoted the proliferation and migration of endometriosis cells, but overexpression of FOXO1 weakened the promoting effect of PRMT3 (Fig. S8 A-D). Similarly, we transfected Flag-PRMT3 and HA-FOXO1 into HESCs and treated them with 8Br-cAMP + MPA for 4 days. The results of qRT-PCR and IF showed that the inhibitory effect of PRMT3 on decidualization of HESCs was also weakened by overexpression of FOXO1 (Fig.  8 C-D).

The study was approved by the Ethics Committee of Shandong Second Medical University, with all participants providing written consent. We collected endometrial tissues from 60 patients with ovarian endometriosis (as experimental group), including 30 cases of eutopic endometrial tissues (in the secretory phase) (EuEM) and 30 cases of ectopic endometrial tissues (EcEM). In addition, we collected tissues from 60 patients with non-ovarian endometriosis (as contral group), including 30 cases of secretory endometrial tissues and 30 cases of proliferative endometrial tissues. The control group of secretory endometrium and EuEM for decidualization related studies. The control group of proliferative endometrium and EcEM were used for endometriosis related research. Patients with ovarian endometriosis were defined as those with a histological diagnosis of endometriosis but no adenomyosis. In contrast, patients without ovarian endometriosis were defined as those with tubal infertility and those who had surgery for non-endometriotic ovarian cysts. All patients had not received hormone therapy for at least 6 months. Clinical details are in Supplementary Table S1 . Perform a series of operations such as paraffin embedding, sectioning, staining, photography, and analysis on the above-mentioned tissues. These methods were carried out as described earlier [ 30 ]. We scored the stained tissue according to the formula H-score=∑ Pi (i). i is the staining intensity (1 point indicates weak, 2 points indicates moderate, and 3 points indicates strong), and Pi is the percentage of stained cells at each intensity (ranging from 0 to 100%). The total score is 0–300. We used the above method [ 31 ] to separate primary eutopic endometrial stromal cells (HESCs) and ectopic endometrial stromal cells (EESCs) from patients with endometriosis. In short, cut the tissues into 1 mm 3 fragments, digested them with type IV collagen (Sigma-Aldrich, St. Louis, MO, USA) for one and a half hours at 37 °C. Finally, used nylon mesh with apertures of 37 mm and 76 mm for separation. Immortalized human endometriotic cell line (12Z) and HEK293T cells were purchased from ATCC (Manassas, VA, USA). Cultivated HESCs, EESCs, and 12Z cells in DMEM/F12 (HyClone) and cultivated HEK293T cells in DMEM (HyClone). Penicillin, streptomycin 100 µg/mL each, and 10% fetal bovine serum (HyClone) were added to both culture media. All cells were in an environment at 37 °C and containing 5% CO2. We constructed recombinant plasmids based on previous articles [ 32 ]. The constructed plasmid includes: Flag-PRMT3, HA-PRMT3, Flag-FOXO1, HA-FOXO1, Flag-FOXO1 mutant (R19K, R21K, R267K, R269K, R251K, R252K, R253K, R314K, R316K)、HA-FOXO1 mutant (R19K, R21K, R267K, R269K, R251K, R252K, R253K, R314K, R316K). Sequence of siPRMT3: 5’-CACTGTCTGCTGAAGCCGCATT-3’, siFOXO1: 5’-CCCAGUCUGUCUGAGAUAATT-3′. Cells inoculated in a six-well plate were transfected using Lipofectamine 2000 (Invitrogen, Cat#11668019). Lysis buffer (Beyotime, P0013) was added to the cells, followed by centrifugation at 12,000 rpm for 10 min at 4 °C. The supernatant obtained was mixed with indicator beads and incubated overnight at 4 °C. Subsequently, 2 × loading buffer (Solarbio, P1040) was added, and the samples were boiled for 10 min at 100 °C. Western blot analysis was then performed according to the method described earlier [ 33 ]. Proteins were visualized using the Odyssey ® Fc imaging system. The experiments were repeated three times ( n = 3). All antibodies used in this experiment are listed in Supplementary Table S2 . For wound healing assay, the confluence of the transfected cells in the six-well plate was 100%, scratched with a medium pipette tip, washed with PBS, and then photographed. photographed again after 24 h of cultivation. The experiments were repeated three times ( n  = 3). For colony formation assay, reseeded 200–800 cells/well in a six-well plate and incubated in 37 °C incubator for 10–14 days. Fixed the cells with 4% paraformaldehyde for 15 min, then stained with crystal violet for 15 min, and finally photographed with gel imaging system (G: BOX F3 gel Document system). The experiments were repeated three times ( n  = 3). For cell proliferation analysis, inoculated the transfected cells into a 24-well plate (20000–25000 cells/well) and counted continuously for 4 days starting from the next day. The experiments were repeated three times ( n  = 3). For migration assay, added 600 µL of culture medium containing 10% FBS to a 24-well plate and put the chamber in it. Then, added 200 µL of serum-free DMEM/F12 culture medium containing transfected cells (80000 cells/well) to the upper chamber. After 12–24 h of cultivation, fixed the cells with 4% paraformaldehyde for 15 min and stained with crystal violet for 10 min. Photographed and counted the number of cells. The experiments were repeated three times ( n  = 3). 0.5 mM 8Br-cAMP (Sigma-Aldrich, St. Louis, MO, USA) and 1 mM medroxyprogesterone acetate (MPA) (Sigma-Aldrich, St. Louis, MO, USA) were added to a six-well plate inoculated with HESCs [ 6 ]. After 48 h of cultivation, 2 ml of new culture medium (containing 8Br-cAMP and MPA) was replaced. The cultivation lasted for a total of 96 h. Total RNA was separated and RNA was reverse transcribed into cDNA using the Trizol assay kit (Takara) and the cDNA synthesis assay kit (Takara), respectively. The TB Green ® Fast qPCR Mix (Takara) and the CFX96 Real-time PCR Detection System (Bio-Rad) are used to detect mRNA expression levels. The experiments were repeated three times ( n  = 3). The summary of quantitative PCR primer sequences is presented in Supplementary Table S3 . 1.2 × 10 5 cells were inoculated on a cell slide. After 24 h, the cells were fixed with 4% paraformaldehyde for 10 min, treated with 0.05% Triton X-100 for 10 min, and blocked with 1% BSA for 1 h. The cells were incubated overnight with the primary antibody at 4 °C. Afterwards, the cells were incubated with the secondary antibody for 1 h at room temperature and finally stained with DAPI. After drying, the slides were examined using a ZEISS fluorescence microscope to capture images. The experiments were repeated three times ( n  = 3). All animal experiments conducted in this study were approved by the Experimental Animal Ethics Review Committee of Shandong Second Medical University. Six-week-old female ( n  = 42) and eight-week-old male ( n  = 8) C57BL/6 mice were used in the experiment. A mouse model of endometriosis was established as mentioned earlier [ 34 ]. The endometrial development in donor mice ( n = 14) was promoted with estradiol benzoate (100 µg/kg) (injected into thigh muscles every two days, three times in total). The uterus of the donor mouse was cut into 1 mm 3 fragments and injected intraperitoneally into the recipient mice ( n = 28) (one donor mouse corresponds to two recipient mice). The recipient mice were also injected with estradiol benzoate using the above method. One month after establishing the endometriosis model, the mice with endometriosis were divided into an experimental group ( n = 7) and a control group ( n = 7), and a decidualization model was established as described in the previous article [ 35 ]. In short, after removing the bilateral ovaries of the mice for two weeks, 100 ng of estradiol was injected into the muscles of the mice (three consecutive days). After resting for two days, progesterone (1 mg) and estradiol (10 ng) were injected into the muscles of mice (three consecutive days). After the third injection for 6 h, 20 µL sesame oil was injected into one side of the mouse’s uterine horn, and the other side was not injected. Progesterone and estradiol were injected continuously for 5 days. During this process, the experimental group mice were intraperitoneally injected with SGC707 five times (30 mg/kg) (MCE, HY-19715). SGC707 is a potent and cell-active allosteric inhibitor of PRMT3 [ 36 , 37 ]. On the second day after the last injection of estrogen into endometriotic mice, the mice were divided into an experimental group ( n  = 7) and a control group ( n  = 7). Then, the experimental group was injected with inhibitors for two consecutive days. Next day, in the experiment, the female mice with endometriosis and control male mice were caged together (from 8pm to 8am the next day). After observing the vaginal plug, the inhibitor was injected once into the experimental group. Afterwards, the inhibitor was injected into the experimental group every two days for a total of six injections. Mice were euthanized at 8.5 days of pregnancy (GD8.5). Cycloheximide (CHX, 100 µg/mL) (MCE, HY-12320) was added to the transfected cells. Cells were collected at 0, 2, 4 and 8 h after adding CHX. The experiments were repeated three times ( n  = 3). We transfected HA-PRMT3, Flag-FOXO1 (WT, R253K), and HA-UB into cells, and added MG132 (100 µmol/L) (MCE, HY-13259) to the cells 8 h before collection. Then, we performed Co-IP and Western blot experiments. The experiments were repeated three times ( n  = 3). After extracting the protein, the content of MDA (Beyotime, S0131S), as well as the activities of SOD (Beyotime, S0101S) and CAT (Beyotime, S0051), were determined using the reagent kit. According to experimental requirements, added antioxidant NAC (10mM) (Beyotime, S0077) 96 h before collecting cells. The experiments were repeated three times ( n  = 3). Data were analyzed with GraphPad Prism 9.0, presented as mean ± SEM (median and quartiles for Mann-Whitney tests). Methods included normality and variance tests, one-way ANOVA, Student’s t tests, Mann-Whitney, and Welch’s tests. Spearman correlation analysis is used to measure the correlation between two variables. P  < 0.05 was significant; P  ≥ 0.05 (n.s.) was not significant.

In this study, we found that PRMT3 expression was elevated in both the EuEM and EcEM of endometriosis, and PRMT3 was negatively correlated with FOXO1. PRMT3 reduced the protein stability of FOXO1 and inhibited its nuclear translocation by methylating the Arg253 site of FOXO1. PRMT3 enhances oxidative stress-induced decidualization defects by degrading FOXO1. Interestingly, the PRMT3 inhibitor SGC707 reversed the aforementioned effects (Fig.  8 E). Moreover, in animal experiments, we found that SGC707 inhibited the number and size of endometriosis lesions, while it promoted deciduomas formation and embryonic development. Endometriosis is a common benign gynecological disease [ 39 ]. It is the main cause of chronic pelvic pain in women and seriously affects fertility [ 40 ]. Decidualization is a prerequisite for a normal pregnancy, accompanied by the formation and remodeling of blood vessels, providing nutrients for embryonic development, and altering the receptivity of the endometrium to synchronize with embryonic development [ 41 ]. Recent studies have used single-cell sequencing technology to detect endometrial shedding cells during menstruation, and find that the IGFBP1 + decidualized endometrial stromal cell subpopulation in patients with endometriosis is significantly lower than that in the normal individuals, indicating impaired decidualization of endometrial stromal cells in patients with endometriosis [ 42 ]. Damage to endometrial decidualization in eutopic endometrium is an important factor contributing to infertility in patients with endometriosis [ 43 ]. However, the mechanism of decidualization damage in endometriosis remains to be explored. PRMT3 is a member of PRMTs that catalyze ADMA [ 44 ]. PRMT3 affects the occurrence of various diseases by methylating downstream target genes. Such as, PRMT3 mediates IGF2BP1 arginine methylation in liver cancer, promoting cell resistance to oxaliplatin [ 45 ]. In endometrial cancer, PRMT3 regulates iron apoptosis through methylation modification of METTL14, promoting tumor progression and resistance to platinum based chemotherapy [ 46 ]. However, there have been no reports on the effects and mechanisms of PRMT3 on endometriosis and decidualization. Here, we found through IHC and Western blot analysis that PRMT3 was upregulated in both eutopic and ectopic endometrium of endometriosis, indicating that PRMT3 was a differentially expressed gene in endometriosis. We altered the expression of PRMT3 in vitro and conducted a series of related studies on cell biology behaviors. The results indicated that overexpression of PRMT3 promoted the proliferation and migration of endometriosis cells, while knockdown of PRMT3 inhibited the proliferation and migration of endometriosis cells. In addition, we used SGC707 to inhibit the enzymatic activity of PRMT3 methylation, which inhibited the promoting effect of PRMT3 on endometriosis cells. Similarly, we found that SGC707 reduced the volume and weight of endometriosis lesions in mice. As can be seen from the results, PRMT3 promoted the progression of endometriosis both in vivo and in vitro. As is well known, human decidualization occurs in the secretory phase of the endometrium and is driven by an increase in progesterone levels and local cyclic AMP production after ovulation [ 47 ]. We found that the expression of PRMT3 was higher in the secretory phase endometrium of endometriosis than in normal individuals, while PRMT3 mRNA expression was reduced in dHESCs. This indicated that decidualization in endometriosis was influenced by PRMT3. Our data showed that overexpression of PRMT3 led to a decrease in the expression levels of PRL and IGFBP1 in dHESCs, while knockdown of PRMT3 or treatment with SGC707 resulted in an increase in the expression levels of PRL and IGFBP1. This indicated that PRMT3 inhibited the decidualization process in vitro. Meanwhile, we investigated the effect of PRMT3 on decidualization in vivo using the inhibitor SGC707. As hypothesized, SGC707 increased the volume and weight of deciduomas, promoted hypertrophy of decidual cells, and increased the expression of key decidual markers such as PRL and IGFBP1, compared to the control group. In addition, it also increased litter size and fetal weight in embryo formation. These findings indicated that treatment with SGC707 not only inhibits the progression of endometriosis but also promotes the decidualization of the endometrium. SGC707 is a potent inhibitor of PRMT3, with an IC 50 of 31 nM and a K D of 53 nM. It exhibits exceptional selectivity, showing no significant inhibitory activity against 31 other methyltransferases and over 250 non-epigenetic targets. In terms of in vivo performance, SGC707 demonstrates good bioavailability in mice: following intraperitoneal injection at a dose of 30 mg/kg, the plasma concentration remains above 208 nM at 6 h post-administration, which is significantly higher than its cell-based IC 50 value. Importantly, this inhibitor displays favorable tolerability in short-term administration regimens in animal models, with no mortality or significant weight loss observed during short-term experiments, making it an ideal tool for preclinical research [ 36 ]. However, there are currently no public reports on the off-target effects of SGC707 following systemic administration, and this research gap urgently requires systematic investigation to be addressed. Based on the results of this study, our future studies will assess the long-term efficacy of SGC707 on reducing endometriosis, enhancing endometrial decidualization, and improving fertility outcomes in patients with endometriosis. FOXO1 plays a crucial role in the differentiation of endometrial stromal cells into decidual cells [ 48 ]. We used FOXO1 as a bait protein for IP/MS and found that PRMT3 is a new interacting protein of FOXO1 in endometriosis cells. Later, we further verified this interaction using Co-IP. We found that PRMT3 and FOXO1 directly interact with each other, whether endogenous or exogenous. For the mechanism of action between PRMT3 and FOXO1, we found that PRMT3, as a methyltransferase, methylated the Arg253 site of FOXO1, reduced the protein stability of FOXO1, and inhibited its nuclear translocation. And PRMT3 promoted the degradation of FOXO1 at the Arg253 site through the ubiquitin proteasome pathway. Previous studies have confirmed that oxidative stress significantly inhibits endometrial decidualization. As core transcription factors of the cellular antioxidant system, FOXO1 and Nrf2 form a FOXO1-Nrf2 complex through crosstalk, cooperatively bind and activate antioxidant response elements (ARE), thereby upregulating antioxidant genes such as SOD and CAT, enhancing the clearance of oxidative damage, and maintaining decidualization homeostasis [ 49 , 50 ]. Interestingly, our data showed that PRMT3 inhibited the decidualization process and reduced the expression of FOXO1. Therefore, we conducted a series of studies to investigate whether FOXO1 degradation mediated by PRMT3 leads to oxidative stress and subsequently results in decidualization defects. We found that PRMT3 promoted the production of the oxidative stress marker MDA and inhibited the enzymatic activity of antioxidant enzymes SOD and CAT, indicating that PRMT3 promotes OS. Interestingly, overexpression of FOXO1 reversed this effect. Subsequently, we found that the OS inhibitor NAC reversed the inhibitory effect of PRMT3 on decidualization. In summary, PRMT3 enhances oxidative stress-induced decidualization defects by degrading FOXO1. We previously found that PRMT3 methylates the Arg253 site of FOXO1, but we were still unclear whether the Arg253 site of FOXO1 is involved in the effects of PRMT3 on endometriosis and decidualization. Therefore, we validated the role of the Arg253 site of FOXO1 in endometriosis and decidualization. The results showed that cells transfected with the HA-FOXO1 (R253K) plasmid exhibited increased proliferation and migration compared to those transfected with the HA-FOXO1 (WT) plasmid, but displayed impaired decidualization capacity. The results indicated that the Arg253 site of FOXO1 was involved in inhibiting endometriosis and promoting decidualization. Meanwhile, our dependency experiments showed that FOXO1 reversed the effects of PRMT3 on endometriosis and decidualization. In a word, PRMT3 promoted endometriosis and inhibited decidualization, which depends on the methylation of the Arg253 site of FOXO1. In summary, we discovered that PRMT3 is a novel molecular marker of abnormal expression in the endometrium of endometriosis. FOXO1 is a new substrate for PRMT3. PRMT3 mediates methylation of FOXO1 at the Arg253 site, which enhances oxidative stress-induced decidualization defects by degrading FOXO1. These findings suggest the potential for targeted PRMT3 therapy to address decidualization defects and may provide new therapeutic avenues for treating infertility associated with endometriosis.

Endometriosis is an estrogen dependent disease characterized by the presence of endometrial like tissue (glands and stroma) outside the uterine cavity [ 1 ]. Endometriosis is closely related to infertility. It is reported that the incidence rate of infertility in patients with endometriosis reaches 50% [ 2 ], and the incidence of endometriosis among women with infertility reaches 20–30% [ 3 ]. Although the incidence rate of endometriosis is rising, its pathogenesis and the mechanism leading to fertility reduction are still unclear. Decidualization refers to the transformation of endometrial stromal fibroblasts into large, epithelioid decidual cells in the presence of estrogen and progesterone [ 4 ]. After decidualization, mesenchymal cells have secretory functions and can secrete insulin-like growth factor binding protein 1 (IGFBP1) and prolactin (PRL), which are considered markers of decidualization [ 5 ]. Decidualization is an important step in successful pregnancy [ 6 ]. Recent studies have shown that abnormal gene expression in endometriosis leads to abnormal decidualization of the eutopic endometrium. For example, hypoxia leads to increased expression of EZH2 and H3K27Me3 in endometriosis, promoting decidualization damage by targeting IGFBP1 [ 7 ]. HDAC3 targets COL1A1 and COL1A2 in patients with endometriosis, inducing decidual defects, suggesting that HDAC3 plays an important role in maintaining endometrial receptivity [ 8 ]. However, the genes abnormally expressed in endometriosis and their molecular mechanisms leading to the decidualization damage of eutopic endometrium in patients with endometriosis remain to be elucidated. The process of decidualization in the endometrium is regulated by multiple key genes, including the FOXO1 transcription factor [ 9 ]. FOXO1 belongs to the FOX family, which includes various members involved in cell differentiation and proliferation, cell cycle, apoptosis, oxidative stress, and DNA damage repair [ 10 – 12 ]. Research has found that the expression level of FOXO1 is lower in the ectopic and eutopic endometrium of patients with endometriosis and is regulated by multiple genes. Such as, the m6A modification regulated by METTL3 promotes the degradation of FOXO1 mRNA mediated by YTHDF2, thereby impairing the decidual process of endometrial stromal cells [ 13 ]. In endometriosis, the PI3K/AKT signaling pathway phosphorylates FOXO1, leading to its degradation after enucleation and ultimately damaging decidualization [ 14 ]. However, the upstream genes that can target FOXO1 to affect the decidualization of eutopic endometrium in patients with endometriosis remain largely unclear. The Protein arginine methyltransferase (PRMTs) family is the most critical regulator of protein methylation modification. The PRMTs family could be divided into three types according to the property of the methylarginine. Type I PRMTs (PRMT1, PRMT2, PRMT3, PRMT4, PRMT6, PRMT8) catalyze asymmetric dimethylarginine modification (ADMA), type II PRMTs (PRMT5, PRMT9) catalyze symmetric dimethylarginine modification (SDMA), and type III PRMTs (PRMT7) catalyze monoarginine modification (MMA) [ 15 ]. Unlike other members, PRMT3 has a unique localization (mainly in the cytoplasm) and structure (containing zinc finger domains) [ 16 – 18 ]. Due to these characteristics, the amount of its substrates is relatively small, and its role in cells is not fully understood. This is especially true in the endometrial stromal cells of patients with endometriosis, where research is very limited. Therefore, strengthening research on PRMT3 in endometriosis may provide new ideas for the treatment of endometriosis and related decidualization damage in endometriosis. Oxidative stress (OS) is a state caused by an imbalance between pro-oxidants and antioxidants [ 19 ]. It has multiple biomarkers, such as the oxidative stress indicator malondialdehyde (MDA) and the antioxidant stress indicators superoxide dismutase 2 (SOD2) and catalase (CAT) [ 20 – 22 ]. OS damages the decidualization of the endometrium and is associated with various reproductive disorders such as endometriosis [ 23 – 25 ]. The OS inhibitor N-acetylcysteine (NAC) can effectively inhibit the proliferation of endometriosis cells and alleviate decidualization damage [ 26 , 27 ]. However, the relationship between OS and decidualization damage in endometriosis still needs to be explored. FOXO1 is an important regulatory factor for cellular antioxidant stress [ 28 ]. Studies have shown that, transcriptional coactivator with PDZ-binding motif (TAZ) improves oxidative damage in stromal cell differentiation by enhancing the antioxidant capacity dependent on the Nrf2/ARE/Foxo1 pathway [ 29 ]. However, the mechanism by which FOXO1 regulates antioxidant stress remains elusive in endometriosis. Herein, we identify PRMT3 as a novel upstream regulatory factor of FOXO1 and establish the relationship between PRMT3 and FOXO1. We elucidate a novel molecular mechanism by studying the methylation modification of FOXO1 by PRMT3 and its impact on the biological functions of FOXO1. Specifically, PRMT3-mediated methylation of FOXO1 promotes FOXO1 degradation, thereby exacerbating oxidative stress-induced decidualization defects in the eutopic endometrium of endometriosis. This may lay a theoretical foundation for developing new strategies targeting PRMT3 for endometriosis-related infertility.

Below is the link to the electronic supplementary material. Supplementary Material 1 (DOCX 8.38 MB) Supplementary Material 1 (DOCX 8.38 MB)