FTO Controls Endometrial Receptivity and Embryo Implantation through Regulating m6A and H3K27me3.

Our study exclusively examined female mice because the disease modeled is only relevant in females. hEEC (Human Endometrial Epithelial Cells), hESC (Human Endometrial Stromal Cells), BeWo and Ishikawa cells were commercially obtained from Procell Biotechnology (hEEC, CP-H058; hESC, CP-H233; BeWo, CL-0500; and Ishikawa, CL-0283). Cells were cultured in complete DMEM (DMEM/FBS/streptomycin = 89/10/1) at 37 °C in 5% CO 2 . At −90% confluency, cells were digested with 0.25% trypsin for 2 min and subcultured. Transient transfections were performed using Lipofectamine 2000 (Invitrogen, 11668019), with medium replaced after 6 h. Cells were harvested 48 h post-transfection. In addition, the vendor website has been provided for reference: https://www.procell.com.cn . Briefly, the Fto knockout mouse used in this study is a global knockout generated by CRISPR/Cas9-mediated editing of exon 3 in zygotes, resulting in a 4-bp deletion in homozygous KO mice. Fto KO and littermate control mice were viable and maintained under specific pathogen-free (SPF) conditions. Throughout the study period, mice were routinely monitored for general health, including gross appearance, activity, body weight, and survival to experimental endpoints, and no overt abnormalities were observed under our experimental conditions. Fto knockout ( Fto KO ) mice were generated on a C57BL/6 J background via intercrossing heterozygotes. Vaginal plugs were monitored to determine gestational day (GD) 0.5, and uteri were collected at GD4.5. Genotyping was performed using tail DNA. Primers are listed in Table  S1 . Primers for FTO and SUZ12 knockdown constructs are listed in Table  S2 . The FTO overexpression plasmid was purchased from MiaoLing Biology. For the construction of the FTO OE plasmids, we followed a procedure 55 . In brief, the FTO fragment was amplified from human cDNA using the FTO-Fwd and FTO-Rev primers. The amplified FTO fragment was subsequently integrated into the mammalian vector pcDNA3.1 (Beijing, Tsingke Biotech). Ishikawa cells were cultured on Poly-HEMA-coated dishes (Sigma, 041M0024U) and transfected in vitro. After 48 h, BeWo cells were suspended, filtered through a 150-mesh sieve (approximately 100 μm pore size) to form spheroids, and seeded onto the Ishikawa monolayer. After 2 h, non-adherent clusters were removed by PBS washing. Adhesion was assessed microscopically after fixation in 4% paraformaldehyde. This in vitro assay is intended as an epithelium-focused readout of early attachment competence, while acknowledging that it falls short of encompassing the full cellular and molecular complexity of a predecidualized endometrium. Total RNA from cells and tissues was extracted using RNAiso Plus (Takara, 9109). DNA was removed with TURBO DNase (Invitrogen, AM2239). RNA concentration was measured using the Qubit RNA HS Assay Kit (Thermo Fisher Scientific, Q32855 ), and integrity was confirmed by gel electrophoresis (28S/18S ratio > 2.0). After 48-h transfection, actinomycin D (5 μg/mL) was added, and cells were collected at 1, 2 and 8 h. Total RNA was isolated with Trizol, and cDNA synthesized using the PrimeScript First Strand cDNA Synthesis Kit (Takara, 6210 A). qRT-PCR was performed using SYBR Green qPCR Mix (BioSharp, BL698A) according to the manufacturer’s protocol. Primers are listed in Table  S1 . Cells and tissues were lysed in RIPA buffer (Beyotime, P0013B) with protease inhibitors (Roche, 11873580001). Lysates were mixed with loading buffer, heated at 100 °C for 10 min, and separated by 12% SDS-PAGE. Proteins were transferred to PVDF membranes (Millipore, IPVH00010) using rapid transfer buffer (New Semet, WB4600). Membranes were blocked in 5% skim milk (BD, 232100) in TBST for 1 h, then incubated with primary and secondary antibodies (Table  S3 ). Signals were visualized via chemiluminescence. Reagent details are in Table  S4 . Serial dilutions of total RNA were spotted on Hybond N + membranes (GE Healthcare, RPN303C), UV-crosslinked twice, and blocked with 5% skim milk for 1 h. Membranes were incubated overnight at 4 °C with anti-m 6 A antibody (CST, 56593), then with secondary antibody (Beyotime, A0208). Methylene blue staining was used for normalization. Signal quantification was performed using ImageJ software. Uterine tissues from GD4.5 mice were fixed in 4% paraformaldehyde for ≥48 h at 4 °C, dehydrated through an ethanol gradient, cleared with xylene, and embedded in paraffin. Sections (5 μm) were cut, deparaffinized, rehydrated, and stained with hematoxylin and eosin. Images were captured using an Olympus microscope. Tissue sections (5 μm) were deparaffinized, and antigen retrieval was performed in citrate buffer (ZSGB-BIO, ZLI-9064). After permeabilization with 0.5% Triton X-100 and blocking with 5% BSA, sections were incubated with primary antibodies overnight at 4 °C and secondary antibodies for 1 h at room temperature. Nuclei were counterstained with DAPI. TUNEL assay was performed using the In Situ Cell Death Detection Kit, TMR Red (Roche, 12156792910). Imaging was performed using an Olympus upright microscope, and data were analyzed with ImageJ. Total RNA was rRNA-depleted using hybridization probes and heat-stable RNase H (Beyotime, R7090M), and DNA was removed with TURBO DNase (Thermo Fisher, AM2239). RNA was precipitated with isopropanol and glycogen (25 μg/mL; Thermo, AM9510), washed, and quantified using the Qubit RNA HS Kit. For m 6 A-RIP, 1 μg rRNA-depleted RNA was fragmented (−200 bp) at 4 °C, then immunoprecipitated using anti-m 6 A antibody (Millipore, ABE572-I) pre-bound to protein A/G beads (Thermo, 10002D/10004D). Beads were washed and RNA eluted using the RNeasy MiniElute Kit (Qiagen, 74104). Libraries were prepared using the SMARTer Stranded Total RNA-Seq Kit v3 (Takara, 634488) with 14 PCR cycles for input and 16 for IP samples. Sequencing was performed on Illumina NovaSeq (150 bp paired-end). Frozen tissue was homogenized in buffer (0.32 M sucrose, 10 mM Tris-HCl pH 7.8, 100 nM EDTA, 5 mM CaCl₂, 3 mM Mg(Ac)₂, 167 μM β-mercaptoethanol), filtered through a 20 μm strainer (pluriSelect, NC1423042), and nuclei were isolated (−50,000/sample). Libraries were prepared using the CUT&Tag Assay Kit (Vazyme, TD903) following the manufacturer’s instructions and sequenced on an Illumina NovaSeq 6000 (150 bp pair-end). All experiments were performed in triplicate. Data are presented as mean ± SD. Statistical significance was determined using unpaired two-tailed Student’s t -test or one-way ANOVA with Tukey-Kramer post hoc tests (GraphPad Prism 9.5). P  < 0.05 was considered significant. See Table  S5 for detailed statistical parameters. The research was approved by the Ethical Committee of the Nanjing Medical University (Approval no. 2014-153). We have complied with all relevant ethical regulations for animal use. Further information on research design is available in the  Nature Portfolio Reporting Summary linked to this article.

To investigate whether FTO knockdown affects endometrial receptivity, we first transfected FTO-targeting plasmids into Ishikawa and human endometrial epithelial cells (hEEC). Western blot analysis confirmed a significant decrease in FTO protein levels following transfection (Fig. 1A, B ). Subsequently, a BeWo spheroid adhesion assay revealed that FTO knockdown markedly reduced the adhesion of BeWo spheroids to Ishikawa cells (Fig. 1C, D ). In parallel, western blot analysis showed that silencing FTO significantly downregulated the expression of estrogen receptor (ER) and progesterone receptor (PGR) in both Ishikawa and hEEC cells (Fig. 1E–H ). Furthermore, assessment of endometrial receptivity markers showed that FTO knockdown led to a significant reduction in leukemia inhibitory factor (LIF) and cytokeratin 18 (CK18) expression, while markedly increasing the expression of Mucin 1 (MUC1) (Fig. 1I–L ). LIF is an IL-6 family cytokine that is indispensable for endometrial receptivity and embryo implantation. CK18 is an epithelial cell marker contributing to endometrial integrity, while MUC1 is a transmembrane glycoprotein that inhibits embryo implantation and is considered a negative marker of endometrial receptivity. Collectively, these results indicate that normal FTO expression is essential for maintaining endometrial receptivity. Fig. 1 FTO knockdown impairs endometrial epithelial cell adhesion and disrupts the expression of receptivity markers. A , B Western blot analysis of FTO expression in Ishikawa and hEEC cells following transfection with FTO-targeting plasmids. n  = 3. C , D BeWo spheroid adhesion assay demonstrating significantly reduced adhesion to Ishikawa cells upon FTO knockdown. n  = 3, Scale bars, 200 μm. E – H Western blot analysis showing decreased expression of estrogen receptor (ER) and progesterone receptor (PGR) in Ishikawa and hEEC cell lines after FTO depletion. n  = 3. I – L FTO knockdown significantly reduced leukemia inhibitory factor (LIF) and cytokeratin 18 (CK18) levels, and increased MUC1 expression, as detected by western blot in Ishikawa or hEEC cells. n  = 3. Data are presented as mean ± SD. SD standard deviation. * P  < 0.05; ** P  < 0.01; *** P  < 0.001. EV empty vector, KD knockdown. A , B Western blot analysis of FTO expression in Ishikawa and hEEC cells following transfection with FTO-targeting plasmids. n  = 3. C , D BeWo spheroid adhesion assay demonstrating significantly reduced adhesion to Ishikawa cells upon FTO knockdown. n  = 3, Scale bars, 200 μm. E – H Western blot analysis showing decreased expression of estrogen receptor (ER) and progesterone receptor (PGR) in Ishikawa and hEEC cell lines after FTO depletion. n  = 3. I – L FTO knockdown significantly reduced leukemia inhibitory factor (LIF) and cytokeratin 18 (CK18) levels, and increased MUC1 expression, as detected by western blot in Ishikawa or hEEC cells. n  = 3. Data are presented as mean ± SD. SD standard deviation. * P  < 0.05; ** P  < 0.01; *** P  < 0.001. EV empty vector, KD knockdown. To further investigate the involvement of FTO-mediated m 6 A demethylation in endometrial receptivity, we first assessed global m 6 A levels using dot blot analysis. FTO knockdown in Ishikawa and hEEC cells resulted in a marked increase in m 6 A signal intensity (Fig.  2A, B ), confirming that FTO functions as an active m 6 A demethylase in endometrial epithelial cells. Fig. 2 FTO regulates m 6 A-dependent SUZ12 mRNA stability and transcriptome-wide methylation patterns. A , B Dot blot analysis showing increased global m 6 A levels in Ishikawa and hEEC cells upon FTO knockdown. C , D Box plot of the m 6 A modification and mRNA levels were identified by m 6 A-RIP-seq in hEEC cells. n  = 3. E Volcano plot of differentially m 6 A-modified genes identified by m 6 A-RIP-seq in hEEC cells. F GO enrichment analysis of differentially methylated transcripts implicates pathways including histone modification and Wnt signaling in hEEC cells. G IGV tracks showing increased m 6 A enrichment at the SUZ12 3′UTR upon FTO depletion in hEEC cells. H , I m 6 A-RIP-qPCR validating enhanced SUZ12 methylation in FTO-deficient hEEC cells. J GSEA plots indicating significant enrichment of transcription and reproduction pathways in hEEC cells. K , L Western blot analysis showing increased SUZ12 protein levels after FTO knockdown in Ishikawa, hESC and hEEC cells. n  = 3. M RNA stability assay demonstrating shortened half-life of SUZ12 mRNA in the absence of FTO in hEEC cells. n  = 3. Data are presented as mean ± SD. SD standard deviation. * P  < 0.05; ** P  < 0.01. EV empty vector, KD knockdown. A , B Dot blot analysis showing increased global m 6 A levels in Ishikawa and hEEC cells upon FTO knockdown. C , D Box plot of the m 6 A modification and mRNA levels were identified by m 6 A-RIP-seq in hEEC cells. n  = 3. E Volcano plot of differentially m 6 A-modified genes identified by m 6 A-RIP-seq in hEEC cells. F GO enrichment analysis of differentially methylated transcripts implicates pathways including histone modification and Wnt signaling in hEEC cells. G IGV tracks showing increased m 6 A enrichment at the SUZ12 3′UTR upon FTO depletion in hEEC cells. H , I m 6 A-RIP-qPCR validating enhanced SUZ12 methylation in FTO-deficient hEEC cells. J GSEA plots indicating significant enrichment of transcription and reproduction pathways in hEEC cells. K , L Western blot analysis showing increased SUZ12 protein levels after FTO knockdown in Ishikawa, hESC and hEEC cells. n  = 3. M RNA stability assay demonstrating shortened half-life of SUZ12 mRNA in the absence of FTO in hEEC cells. n  = 3. Data are presented as mean ± SD. SD standard deviation. * P  < 0.05; ** P  < 0.01. EV empty vector, KD knockdown. To explore the downstream molecular consequences of altered m 6 A methylation, we performed m 6 A RNA immunoprecipitation sequencing (m 6 A-RIP-seq) to identify transcriptome-wide targets of FTO-mediated demethylation. FTO silencing led to a substantial elevation in global m 6 A enrichment, accompanied by a significant increase in the mRNA levels of specific target genes (Fig.  2C, D ). Differential analysis identified 939 genes with increased m 6 A modification and 15 genes with reduced m 6 A levels following FTO knockdown (Fig.  2E ), indicating that the loss of FTO primarily induces hypermethylation at selected transcripts. Gene Ontology (GO) enrichment analysis of these differentially methylated genes revealed significant involvement in biological processes such as histone modification, the Wnt signaling pathway, and transcription factor binding (Fig.  2F ). These pathways are known to be critical for endometrial development and epithelial cell function. Among the m 6 A-enriched transcripts, SUZ12 was identified as a potential direct target of FTO. Integrated Genome Viewer (IGV) analysis showed a notable increase in the m 6 A peak within the 3′-untranslated region (3′-UTR) of SUZ12 mRNA following FTO knockdown (Fig.  2G ). This observation was further supported by m 6 A-RIP-qPCR, which confirmed a significant enrichment of m 6 A on SUZ12 transcripts in the absence of FTO (Fig.  2H, I ). To gain insight into the broader biological implications, we conducted gene set enrichment analysis (GSEA). The results indicated that these m 6 A-differential genes were associated with pathways involved in cis-regulatory sequence-specific DNA binding, transcription factor activity, reproduction, epithelial cell development, the meiotic cell cycle, mitochondrial matrix function and glycosyltransferase activity (Fig.  2J , Supplementary Fig.  1A–D ). Although m 6 A levels were elevated, western blot analysis showed that SUZ12 protein expression also decreased significantly after FTO knockdown (Fig.  2K, L ). Considering the role of m 6 A in post-transcriptional regulation, we next performed RNA stability assays to clarify the mechanism. The results demonstrated that FTO knockdown significantly reduced the half-life of SUZ12 mRNA (Fig.  2M ), suggesting that increased m 6 A levels accelerate SUZ12 transcript degradation. Together, these findings indicate that FTO maintains SUZ12 mRNA stability by removing m 6 A modifications. To investigate whether SUZ12 knockdown affects endometrial receptivity, we first silenced SUZ12 expression in Ishikawa cells. The BeWo spheroid adhesion assay demonstrated that SUZ12 depletion significantly reduced the adhesive capacity of Ishikawa cells (Fig.  3A, B ). In parallel, western blot analysis showed that SUZ12 knockdown led to a marked downregulation of estrogen receptor (ER) expression (Fig.  3C, D ). Consistently, SUZ12 silencing significantly decreased the expression of key endometrial receptivity markers, including LIF and CK18, while markedly increasing the expression of MUC1 (Fig.  3E, F ). Fig. 3 SUZ12 is a downstream effector of FTO in regulating endometrial receptivity. A , B BeWo spheroid adhesion assay showing reduced adhesion following SUZ12 knockdown in Ishikawa cells. n  = 3, Scale bars, 500 μm. C , D Western blot analysis indicating decreased ER expression upon SUZ12 depletion in Ishikawa cells. n  = 3. E , F Loss of SUZ12 downregulated LIF and CK18 while upregulating MUC1 in Ishikawa cells. n  = 3. G , H Rescue of adhesion capacity in FTO-deficient cells by SUZ12 overexpression in Ishikawa cells. n  = 3, Scale bars, 500 μm. I , J Restoration of ER expression upon SUZ12 re-expression in FTO-silenced Ishikawa cells. n  = 3. K , L Rescue of LIF and MUC1 expression levels by SUZ12 overexpression in Ishikawa cells. n  = 3. Data are presented as mean ± SD. SD, standard deviation. */# P  < 0.05; ** P  < 0.01. EV empty vector, KD knockdown, OE overexpression. * indicates comparison between the EV group and the SUZ12 KD group, or between the EV + pcDNA3.1 group and the FTO KD  + pcDNA3.1 group; # indicates comparison between the FTO KD  + pcDNA3.1 group and the FTO KD  + SUZ12 OE group. A , B BeWo spheroid adhesion assay showing reduced adhesion following SUZ12 knockdown in Ishikawa cells. n  = 3, Scale bars, 500 μm. C , D Western blot analysis indicating decreased ER expression upon SUZ12 depletion in Ishikawa cells. n  = 3. E , F Loss of SUZ12 downregulated LIF and CK18 while upregulating MUC1 in Ishikawa cells. n  = 3. G , H Rescue of adhesion capacity in FTO-deficient cells by SUZ12 overexpression in Ishikawa cells. n  = 3, Scale bars, 500 μm. I , J Restoration of ER expression upon SUZ12 re-expression in FTO-silenced Ishikawa cells. n  = 3. K , L Rescue of LIF and MUC1 expression levels by SUZ12 overexpression in Ishikawa cells. n  = 3. Data are presented as mean ± SD. SD, standard deviation. */# P  < 0.05; ** P  < 0.01. EV empty vector, KD knockdown, OE overexpression. * indicates comparison between the EV group and the SUZ12 KD group, or between the EV + pcDNA3.1 group and the FTO KD  + pcDNA3.1 group; # indicates comparison between the FTO KD  + pcDNA3.1 group and the FTO KD  + SUZ12 OE group. To further determine whether SUZ12 mediates the functional effects of FTO, we performed a complementation assay. Overexpression of SUZ12 in FTO-deficient Ishikawa cells significantly restored BeWo spheroid adhesion capacity (Fig.  3G, H ). In addition, SUZ12 overexpression reversed the downregulation of ER expression caused by FTO knockdown (Fig.  3I, J ). It also rescued the reduced expression of LIF and attenuated MUC1 upregulation observed in FTO-silenced cells (Fig.  3K, L ). Together, these results demonstrate that SUZ12 functions as a key downstream effector of FTO, mediating its regulatory role in endometrial receptivity through modulation of cell adhesion and receptivity-related gene expression. To investigate the role of Fto in endometrial receptivity, we generated an Fto knockout ( Fto KO ) mouse model using CRISPR/Cas9-mediated genome editing in our previous study 31 . A 4 bp deletion was introduced into exon 3 of the Fto gene (Fig.  4A, B ). Western blot analysis confirmed a significant reduction in Fto protein levels in the uterus of Fto KO mice (Fig.  4C, D ). Consistently, immunofluorescence staining of uterine tissue at gestational day 4.5 (GD4.5) revealed a marked loss of Fto signal (Fig.  4E, F ). Fig. 4 Fto knockout leads to uterine epithelial cell apoptosis and glandular loss in vivo . A , B Schematic and genotyping results showing a 4-bp deletion in exon 3 of Fto gene in Fto KO mice. C , D Western blot confirms reduction of Fto protein in uterus of Fto KO mice. E , F Immunofluorescence staining at GD4.5 confirms Fto deletion in uterine epithelium. Scale bars, 50 μm. G – I Gross morphology and weight of uteri showing significant reduction in uterine weight in Fto KO mice. J , K H&E staining reveals reduced gland numbers and disrupted epithelial architecture in Fto KO uterus. Scale bars, 50 μm. L , M Immunofluorescence staining showing decreased Ki67-positive and increased TUNEL-positive epithelial cells, with elevated Bcl2 expression in Fto -deficient uteri. Scale bars, 50 μm. Data are presented as mean ± SD. SD standard deviation. * P  < 0.05; *** P  < 0.001; **** P  < 0.0001; ns presents no significance. For all IF and HE, n   =  3. WT wildtype, KO knockout. A , B Schematic and genotyping results showing a 4-bp deletion in exon 3 of Fto gene in Fto KO mice. C , D Western blot confirms reduction of Fto protein in uterus of Fto KO mice. E , F Immunofluorescence staining at GD4.5 confirms Fto deletion in uterine epithelium. Scale bars, 50 μm. G – I Gross morphology and weight of uteri showing significant reduction in uterine weight in Fto KO mice. J , K H&E staining reveals reduced gland numbers and disrupted epithelial architecture in Fto KO uterus. Scale bars, 50 μm. L , M Immunofluorescence staining showing decreased Ki67-positive and increased TUNEL-positive epithelial cells, with elevated Bcl2 expression in Fto -deficient uteri. Scale bars, 50 μm. Data are presented as mean ± SD. SD standard deviation. * P  < 0.05; *** P  < 0.001; **** P  < 0.0001; ns presents no significance. For all IF and HE, n   =  3. WT wildtype, KO knockout. To assess whether Fto deficiency affects uterine function, we examined gross uterine morphology at GD4.5. Although no significant differences in uterine size were observed, uterine weight was significantly reduced in Fto KO mice compared with controls (Fig.  4G–I ). Histological analysis using hematoxylin and eosin (H&E) staining further revealed a notable reduction in the number of glandular epithelial cells, indicating compromised uterine epithelial architecture in the absence of Fto (Fig.  4J, K ). Moreover, immunofluorescence analysis at GD4.5 demonstrated that Fto deficiency led to reduced epithelial cell proliferation, as evidenced by a marked decrease in Ki67-positive cells. This was accompanied by increased apoptotic activity, evidenced by enhanced TUNEL staining, along with decreased expression of the anti-apoptotic protein Bcl2 (Fig.  4L, M ). To further verify the role of Fto in regulating endometrial receptivity in vivo, immunofluorescence staining was performed. The results showed that Fto knockout significantly reduced the expression of key functional markers, including cytokeratin 18 (Ck18), vimentin, and forkhead box A2 (Foxa2) (Fig.  5A–D ). Consistently, Fto deficiency also caused a significant decrease in estrogen receptor (Er) expression (Fig.  5E, F ). In line with these findings, the expression levels of canonical receptivity markers, such as Hoxa10 and Lif, were markedly downregulated, while the receptivity-inhibitory marker Muc1 was upregulated (Fig.  5G, H ). Together, these findings indicate that Fto is critical for maintaining uterine structural integrity, supporting epithelial proliferation and preserving endometrial receptivity during implantation. Fig. 5 Fto deficiency impairs endometrial receptivity marker expression during implantation window. A – D Immunofluorescence analysis of Ck18, Vimentin, and Foxa2 in uterine tissue at GD 4.5 shows reduced expression in Fto KO mice. Scale bars, 50 μm. E , F Decreased Er expression in the endometrium of Fto-deficient mice. Scale bars, 50 μm. G , H Immunofluorescence analysis showing downregulation of Hoxa10 and Lif, and upregulation of Muc1 in Fto KO uteri. Scale bars, 50 μm. Data are presented as mean ± SD. SD standard deviation. * P  < 0.05; ** P  < 0.01; **** P  < 0.0001. For all IF, n   =  3. WT wildtype, KO knockout. A – D Immunofluorescence analysis of Ck18, Vimentin, and Foxa2 in uterine tissue at GD 4.5 shows reduced expression in Fto KO mice. Scale bars, 50 μm. E , F Decreased Er expression in the endometrium of Fto-deficient mice. Scale bars, 50 μm. G , H Immunofluorescence analysis showing downregulation of Hoxa10 and Lif, and upregulation of Muc1 in Fto KO uteri. Scale bars, 50 μm. Data are presented as mean ± SD. SD standard deviation. * P  < 0.05; ** P  < 0.01; **** P  < 0.0001. For all IF, n   =  3. WT wildtype, KO knockout. Given that Fto deficiency alters gene transcription, we next examined whether it also affects histone modifications, with a particular focus on H3K27me3, a well-known epigenetic repressive mark involved in transcriptional regulation. To further explore whether Fto influences endometrial receptivity through Suz12-mediated histone H3K27me3 modification, we analyzed both H3K27me3 levels and transcriptional changes in uterine tissues from Fto KO mice. Immunofluorescence analysis of uterine tissue at gestational day 4.5 (GD4.5) further demonstrated that Fto deficiency led to a marked reduction in Suz12 and H3K27me3-positive cells (Fig.  6A, B ). CUT&Tag sequencing revealed that the Fto KO group displayed markedly reduced H3K27me3 enrichment around transcription start sites (±3 kb) (Fig.  6C ), indicating a global alteration in transcriptional regulation. Further analysis identified 457 sites showed decreased and 744 sites showed increased H3K27me3 in the Fto KO group relative to the WT group (Fig.  6D ). GO analysis showed that genes with differential H3K27me3 modification were enriched in cell-substrate adhesion, reproductive structure development, Wnt signaling pathway, and transcription regulator complex (Fig.  6E ). Two overlapping Wnt pathway-related genes, WNT5B and CELSR2, were identified between upregulated m 6 A-methylated transcripts and genes with decreased H3K27me3 modification (Fig.  6F ). Integrated Genome Viewer (IGV) analysis showed a notable increase in the m 6 A peak within the untranslated regions of WNT5B or CELSR2 mRNA following FTO knockdown (Supplementary Fig.  2A ). This observation was further supported by m 6 A-RIP-qPCR, which confirmed an increased enrichment of m 6 A on WNT5B and CELSR2 transcripts in the absence of FTO (Supplementary Fig.  2B, C ). Fig. 6 The Fto/Suz12/H3K27me3/Wnt5b axis modulates endometrial receptivity via the Wnt signaling pathway. A , B Immunofluorescence staining showing reduced Suz12 and H3K27me3 in uterine epithelium of Fto KO mice at GD 4.5. Scale bars, 50 μm. C , D CUT&Tag-seq identifies genes with altered H3K27me3 modification following Fto deletion. n  = 3. E Functional enrichment analysis of genes with differential H3K27me3 signals. F Venn diagram identifying WNT5B and CELSR2 as overlapping targets of m 6 A and H3K27me3 regulation. G qRT-PCR validation of WNT5B upregulation in FTO-deficient hEEC cells. H IGV tracks showing loss of H3K27me3 at Wnt5b locus in Fto-deficient uterus. I CUT&Tag-qPCR confirms H3K27me3 reduction at Wnt5b promoter in Fto-deficient uterus. J , K Immunofluorescence reveals increased Wnt5b and decreased c-Myc and β-catenin in Fto-deficient uterus. Scale bars, 50 μm. Data are presented as mean ± SD. SD standard deviation. * P  < 0.05; ** P  < 0.01. For all IF, n   =  3. WT wildtype, KO knockout. A , B Immunofluorescence staining showing reduced Suz12 and H3K27me3 in uterine epithelium of Fto KO mice at GD 4.5. Scale bars, 50 μm. C , D CUT&Tag-seq identifies genes with altered H3K27me3 modification following Fto deletion. n  = 3. E Functional enrichment analysis of genes with differential H3K27me3 signals. F Venn diagram identifying WNT5B and CELSR2 as overlapping targets of m 6 A and H3K27me3 regulation. G qRT-PCR validation of WNT5B upregulation in FTO-deficient hEEC cells. H IGV tracks showing loss of H3K27me3 at Wnt5b locus in Fto-deficient uterus. I CUT&Tag-qPCR confirms H3K27me3 reduction at Wnt5b promoter in Fto-deficient uterus. J , K Immunofluorescence reveals increased Wnt5b and decreased c-Myc and β-catenin in Fto-deficient uterus. Scale bars, 50 μm. Data are presented as mean ± SD. SD standard deviation. * P  < 0.05; ** P  < 0.01. For all IF, n   =  3. WT wildtype, KO knockout. To investigate whether FTO regulates the expression of WNT5B and CELSR2 through mRNA stability, RNA decay assays were performed. However, the results showed that FTO knockdown had no significant effect on the half-life of either transcript (Supplementary Fig.  2D ), indicating that FTO does not modulate their expression via RNA stability. Meanwhile, qRT-PCR analysis showed that WNT5B expression was upregulated at the mRNA level, whereas CELSR2 expression remained unchanged (Fig.  6G , Supplementary Fig.  2E ). These findings suggest that WNT5B expression may be regulated by H3K27me3-mediated epigenetic mechanisms. IGV analysis showed a notable decrease in the H3K27me3 peak within the gene body of Wnt5b following Fto knockout (Fig.  6H ). This observation was further supported by CUT&Tag-qPCR, which confirmed a significant enrichment of H3K27me3 on the promoter of Wnt5b in the absence of Fto (Fig.  6I ). Immunofluorescence analysis of uterine tissue at gestational day 4.5 (GD4.5) further revealed a significant accumulation of Wnt5b-positive cells in the Fto-deficient uterus. However, c-Myc and β-catenin positive cells in the downstream of the classical Wnt signaling pathway were significantly reduced (Fig.  6J, K ). Together, these findings indicate that Fto deficiency disrupts epigenetic homeostasis by reducing H3K27me3 modification at the Wnt5b locus, thereby upregulating Wnt5b expression but attenuating canonical Wnt signaling activity in the uterus.

Increasing evidence has demonstrated the importance of epitranscriptomic regulation in establishing endometrial receptivity, particularly through N 6 -methyladenosine (m 6 A) methylation 32 . However, whether the m 6 A demethylase FTO plays a distinct and critical role in the endometrial epithelium remains unclear. In this study, we show that Fto knockout ( Fto KO ) mice exhibit impaired endometrial receptivity, characterized by disrupted epithelial integrity, reduced proliferation, and aberrant expression of receptivity markers. Transcriptomic and epigenetic analyses further reveal that the loss of FTO induces widespread alterations in gene expression and m 6 A methylation, especially affecting pathways involved in Wnt signaling and histone modification. These findings demonstrate that FTO exerts essential and specific regulatory functions during the implantation window, acting through coordinated epitranscriptomic and epigenetic mechanisms to support the acquisition of a receptive endometrial phenotype. FTO modulates m 6 A demethylation on mRNA transcripts, thereby influencing transcript stability, splicing, and translation, all of which are essential for endometrial remodeling during the window of implantation 32 , 33 . Previous studies have implicated aberrant FTO activity in uterine pathologies associated with implantation failure, supporting its essential role in uterine function 34 , 35 . Notably, recent studies have also implicated FTO in regulating 6-methyladenine (6 mA) modifications in genomic DNA, adding an additional layer of complexity to its epigenetic functions 36 . Although the presence of 6 mA in mammalian genomes remains relatively low and its biological functions remain under investigation, emerging evidence suggests that it may participate in gene transcription activation 37 , DNA repair 38 , and chromatin organization 39 . The interplay between m 6 A and 6 mA modifications mediated by FTO may therefore fine-tune gene expression programs necessary for establishing a receptive endometrial phenotype. In our study, SUZ12 was identified as a novel m 6 A-dependent target that is downregulated upon FTO knockdown. Notably, overexpression of SUZ12 fully rescued the expression of key receptivity-related genes impaired by FTO deficiency, highlighting a potential epigenetic axis through which FTO operates dependently of m 6 A demethylation. These findings imply that m 6 A-mediated mechanisms may have a more prominent impact than 6 mA modifications in regulating endometrial receptivity. Histone H3 lysine 27 trimethylation (H3K27me3), catalyzed by the Polycomb Repressive Complex 2 (PRC2), is a well-characterized repressive epigenetic mark in with SUZ12 serves as an essential structural and functional component 40 . This modification plays a critical role in transcriptional repression by promoting chromatin condensation at developmentally regulated gene loci. In the endometrium, dynamic remodeling of H3K27me3 across the menstrual cycle and the implantation window ensures proper timing of gene activation and repression required for uterine receptivity 41 . Disruption of H3K27me3 patterns has been associated with impaired decidualization, defective implantation, and reproductive failure 42 , 43 . Our findings indicate that the RNA demethylase FTO directly regulates H3K27me3 by modulating the post-transcriptional stability of SUZ12. Specifically, FTO knockdown increases m 6 A methylation on SUZ12 mRNA, thereby reducing transcript stability and lowering SUZ12 protein levels. Consequently, the impairment of PRC2 activity leads to diminished H3K27me3 deposition at key regulatory gene loci, including those involved in WNT signaling. We observed that H3K27me3 loss at the promoter region of Wnt5b facilitates its transcriptional upregulation. In addition to modulating the expression of chromatin regulators at the post-transcriptional level, m 6 A has also been shown to indirectly influence gene expression by recruiting histone-modifying enzymes to specific genomic loci. Recent studies have demonstrated that certain m 6 A reader proteins, such as YTHDC1, can interact with components of chromatin remodeling complexes and facilitate the localization of histone modifiers to transcriptionally relevant regions 44 . For example, m 6 A-modified RNAs have been reported to act as scaffolds for PRC2 localization, thereby promoting H3K27me3 deposition and transcriptional silencing 32 . This RNA-mediated recruitment mechanism adds a co-transcriptional layer of regulation, where m 6 A not only influences transcript fate but also serves as a molecular bridge connecting RNA to chromatin regulator. These findings highlight the dual role of m 6 A in epigenetic control, regulating both the histone modifier expression and the spatial guidance of these complexes to chromatin. A limitation of the current study is that, although we demonstrate that FTO regulates H3K27me3 through m 6 A-dependent control of SUZ12 mRNA stability, whether m 6 A-modified RNAs or m 6 A reader proteins contribute to PRC2 recruitment to chromatin in endometrial cells remains to be determined. WNT5B, a member of the non-canonical WNT family, plays a pivotal role in modulating WNT signaling dynamics by antagonizing the classical WNT/β-catenin pathway 45 . Canonical WNT signaling is initiated by ligands such as WNT1 or WNT3A binding to Frizzled (FZD) receptors and LRP5/6 co-receptors, stabilizing β-catenin and promoting its nuclear translocation, WNT5B signals independently of β-catenin 46 . Instead, WNT5B activates alternative pathways, including the planar cell polarity (PCP) and WNT/Ca²⁺ pathways 28 , which are associated with cytoskeletal remodeling, cell motility, and calcium flux. Accumulating evidence indicates that WNT5B inhibits β-catenin signaling by preventing LRP6 phosphorylation and maintaining the integrity of the β-catenin destruction complex. This preservation of the complex, which includes APC, AXIN, and GSK3β, facilitates continuous phosphorylation, ubiquitination, and proteasomal degradation of β-catenin 47 . Meanwhile, WNT5B also induces the expression of ubiquitin ligases or phosphatases that directly promote β-catenin turnover. These mechanisms converge to repress β-catenin/TCF-mediated transcription of genes essential for proliferation, differentiation, and stemness 48 . Notably, we observed a significant decrease in H3K27me3 occupancy at the promoter region of WNT5B, leading to its transcriptional upregulation. Elevated WNT5B expression has been reported to antagonize canonical WNT/β-catenin signaling by promoting β-catenin degradation or blocking upstream activation, thereby suppressing transcriptional programs required for endometrial receptivity 49 . Consistent with this, our findings suggest that the FTO-SUZ12-H3K27me3 axis impairs endometrial receptivity through epigenetic repression of WNT5B, thereby suppressing canonical WNT signaling. This regulatory cascade highlights a novel epigenetic mechanism through which RNA methylation modulates histone modifications and downstream signaling pathways essential for uterine function. Another key limitation of this study is that the Fto knockout model is global (whole-body) rather than uterine epithelium-specific; therefore, the uterine phenotypes observed may reflect not only epithelial-intrinsic loss of Fto but also contributions from stromal or myometrial compartments and/or systemic effects. Recent high-resolution single-cell and spatiotemporal atlases further highlight that endometrial function and receptivity rely on coordinated stroma-epithelium crosstalk and compartment-specific programs 50 , 51 . In addition, genetic evidence indicates that commonly used uterine drivers such as Pgr -Cre can affect multiple uterine compartments, whereas epithelial-restricted strategies provide clearer compartment attribution 52 , 53 . Future studies using epithelium-restricted Fto deletion and compartment-resolved analyses will be required to conclusively establish epithelial causality in vivo 54 . In summary, our findings uncover a previously unrecognized role for FTO in promoting endometrial receptivity through coordinated epitranscriptomic and epigenetic mechanisms. Loss of FTO increased m 6 A methylation and reduced stability of SUZ12 mRNA, resulting in diminished SUZ12 expression and a global decrease in H3K27me3 deposition. This epigenetic dysregulation disrupted transcriptional repression at key loci such as WNT5B , resulting in aberrant activation of signaling pathways that impair epithelial integrity and receptivity. The identification of the FTO/SUZ12/H3K27me3 regulatory axis highlights a critical interface between RNA modification and chromatin regulation, providing novel mechanistic insight into the molecular basis of embryo implantation and potential therapeutic targets for reproductive disorders (Fig.  7 ). Fig. 7 FTO regulates endometrial receptivity through the SUZ12/H3K27me3/WNT5B axis. The schematic diagram illustrates that the FTO/SUZ12/H3K27me3/Wnt5B axis modulates endometrial receptivity through the Wnt signaling pathway. The schematic diagram illustrates that the FTO/SUZ12/H3K27me3/Wnt5B axis modulates endometrial receptivity through the Wnt signaling pathway.

Endometrial receptivity is a temporally restricted, tightly regulated biological process that enables successful embryo implantation 1 . This state is established during the “window of implantation” a brief period characterized by extensive molecular and structural remodeling of the endometrial epithelium that facilitates blastocyst attachment and invasion 2 . At the cellular level, the luminal epithelium undergoes polarization, junctional reorganization, and secretory activation to enable embryo adhesion, while stromal cells initiate decidualization to accommodate trophoblast invasion 3 . Disruption of these tightly coordinated processes can result in implantation failure, recurrent pregnancy loss, or infertility 4 . While the roles of endocrine and immune factors have been partially elucidated 5 , the integrated molecular mechanisms involving both epigenetic and post-transcriptional regulation during this transition remain incompletely understood. Among emerging epitranscriptomic regulators, N 6 -methyladenosine (m 6 A) is the most abundant internal RNA modification in eukaryotes and plays a pivotal role in regulating RNA stability, splicing, export, and translation 6 . The fat mass and obesity-associated protein (FTO) functions as an m 6 A demethylase that removes methylation marks from target transcripts, thereby modulating their post-transcriptional fate 7 . Although the functions of FTO have been extensively studied in metabolism 8 , oncogenesis 9 , and neural development 10 , its regulatory role in reproductive tissues, particularly in the endometrial epithelium, remains poorly defined. Recent studies have demonstrated that the expression of m 6 A-related enzymes is dynamically regulated across the menstrual cycle 11 . m 6 A methylation has been implicated in essential reproductive events such as endometrial stromal decidualization 12 and ovarian folliculogenesis 13 . Moreover, altered FTO expression has been reported in gynecologic disorders such as polycystic ovary syndrome and endometriosis, which are frequently associated with impaired endometrial receptivity 14 . These findings suggest that FTO may be an important post-transcriptional regulator during the acquisition of a receptive endometrial phenotype. Epigenetic mechanisms, particularly histone modifications, are essential for establishing the gene expression landscape that underlies endometrial receptivity. The Polycomb Repressive Complex 2 (PRC2) catalyzes trimethylation of histone H3 at lysine 27 (H3K27me3), a repressive mark that governs transcriptional silencing during tissue development and homeostasis 15 . Suppressor of Zeste 12 (SUZ12) is a core structural component of PRC2 and is indispensable for maintaining the integrity and enzymatic activity of the complex 16 . Although PRC2-mediated repression has been implicated in uterine development and pathophysiology 17 , the specific contribution of SUZ12 to the establishment of endometrial receptivity remains largely unexplored. Chromatin immunoprecipitation and transcriptome analyses have revealed that H3K27me3 marks are redistributed in a cycle-dependent manner, correlating with the expression of genes involved in cell adhesion 18 , cytokine signaling 19 , and epithelial remodeling 20 . Notably, emerging evidence indicates that SUZ12 mRNA may be subject to regulation by m 6 A methylation 21 . This suggests that FTO may stabilize SUZ12 transcripts, thereby linking m 6 A-mediated post-transcriptional control with chromatin-based transcriptional repression in the receptive endometrium. Wnt signaling plays a central role in the regulation of cellular polarity 22 , adhesion 23 , and differentiation 24 , all of which are critical for establishing endometrial receptivity. During the receptive phase, canonical Wnt signaling must be precisely downregulated in the luminal epithelium, as excessive activation of β-catenin has been shown to impair embryo attachment 25 . Conversely, Wnt activity in the stromal compartment supports decidualization 26 and trophoblast invasion 27 . Among the Wnt ligands, Wnt5b has emerged as a non-canonical modulator that antagonizes β-catenin signaling by activating planar cell polarity and calcium-dependent pathways 28 . Although Wnt5b is expressed in reproductive tissues, its specific regulatory role in the endometrium remains poorly characterized. Dysregulated expression or epigenetic repression of Wnt5b may result in aberrant Wnt pathway activity, impairing epithelial quiescence and disrupting the embryo-endometrium interaction. In addition, Wnt5b has been implicated in the regulation of inflammatory signaling 29 and cellular senescence 30 , both of which are tightly associated with the establishment and maintenance of a receptive endometrial microenvironment. These observations suggest that appropriate regulation of Wnt5b is essential for preserving epithelial integrity and immunological balance during the implantation window. In this study, we identify a previously unrecognized regulatory axis involving FTO, SUZ12, H3K27me3, and WNT5B that coordinates epitranscriptomic and epigenetic mechanisms during the acquisition of endometrial receptivity. We demonstrate that FTO-mediated m 6 A demethylation stabilizes SUZ12 mRNA, thereby enabling the PRC2 complex to repress WNT5B transcription through H3K27me3 deposition at its promoter region. Repression of WNT5B attenuates epithelial Wnt signaling, restrains premature senescence, and promotes cell adhesion, ultimately supporting the establishment of a functionally receptive endometrial state. These findings provide mechanistic insight into the molecular regulation of endometrial receptivity and suggest potential therapeutic targets for implantation failure and infertility.

Supplementary Information Supplementary Data 1 Description of additional supplementary file Reporting Summary Supplementary Information Supplementary Data 1 Description of additional supplementary file Reporting Summary