mTORC1 as a metabolic rheostat coordinates hormonal signals and cellular proliferation to sustain endometrial competence.

To evaluate the involvement of mTORC1 signaling in thin endometrium pathogenesis, we initially analyzed the GSE160633 dataset sourced from the GEO database. This dataset comprises gene expression profiles from eight samples of thin endometrium and their adjacent normal endometrial tissues, all collected during the mid-luteal phase [ 18 ]. Gene set enrichment analysis (GSEA) of these sequencing data demonstrated significant downregulation of mTORC1 signaling targets in thin endometrium compared with normal tissue (Fig.  1 A, B). Subsequently, we assessed mTORC1 signaling pathway activity in both thin endometrium and normal controls using immunohistochemistry. The immunohistochemical quantification revealed that the phosphorylation level of the mTORC1 signaling molecule S6 (Ser235/236) was significantly decreased in thin endometrium, particularly in epithelial cells, compared with normal controls (Fig.  1 C, D). These consistent findings across transcriptomic and protein levels establish that mTORC1 hypoactivation is a hallmark feature of thin endometrium, hinting at a potential association between mTORC1 signaling deficiency and the pathogenesis of thin endometrium. Fig. 1 The mTORC1 signaling activity decreases in patients with thin endometrium A. Gene set enrichment analysis (GSEA) enrichment plot of the mTORC1 signaling in the endometrium transcriptome of patients with thin endometrium (Thin) to controls (NC) ( GSE160633 ). The normalized enrichment score (NES), p -value, and q -value are obtained from the GSEA algorithm. B. A corresponding heatmap of the gene expression levels associated with the leading edge of mTORC1 signaling in thin endometrium compared with controls. C. Immunohistochemical staining for p-S6Ser235/236 in both normal (left) and thin endometrium (right), with nuclei counterstained with hematoxylin. Scale bars: 200 μm. D. Quantification the integrated optical density (IDO) values of the p-S6Ser235/236 in both normal (NC, n  = 5) and thin endometrium (Thin, n  = 5) using ImageJ software. IOD was calculated as the sum of the optical densities of all pixels within the area of interest (IOD = Mean Density × Area). Data are presented as mean ± SD. Statistical significance was determined by a two-tailed Student’s t -test. ***p < 0.0001, relative to control The mTORC1 signaling activity decreases in patients with thin endometrium A. Gene set enrichment analysis (GSEA) enrichment plot of the mTORC1 signaling in the endometrium transcriptome of patients with thin endometrium (Thin) to controls (NC) ( GSE160633 ). The normalized enrichment score (NES), p -value, and q -value are obtained from the GSEA algorithm. B. A corresponding heatmap of the gene expression levels associated with the leading edge of mTORC1 signaling in thin endometrium compared with controls. C. Immunohistochemical staining for p-S6Ser235/236 in both normal (left) and thin endometrium (right), with nuclei counterstained with hematoxylin. Scale bars: 200 μm. D. Quantification the integrated optical density (IDO) values of the p-S6Ser235/236 in both normal (NC, n  = 5) and thin endometrium (Thin, n  = 5) using ImageJ software. IOD was calculated as the sum of the optical densities of all pixels within the area of interest (IOD = Mean Density × Area). Data are presented as mean ± SD. Statistical significance was determined by a two-tailed Student’s t -test. ***p < 0.0001, relative to control To establish the causal relationship between mTORC1 dysfunction and thin endometrium development, we generated uterine-specific Raptor knockout mice ( Rptor fl/fl Pgr cre/ + ) using the Cre-loxP system (Supplementary Fig. S1A). Immunofluorescence analysis demonstrated efficient and specific deletion of Raptor in uterine epithelial and stromal cells, accompanied by significant downregulation of both Raptor protein and its downstream phosphorylation target p-S6Ser235/236 (Supplementary Fig. S1B, C), thereby validating successful mTORC1 pathway disruption. The absence of Rptor in the uterus did not impair the somatic growth of the female mice, as evidenced by the comparable weights observed between Rptor -deficient and control mice (Supplementary Fig. S1D). It is known that throughout the estrus cycle and pregnancy of healthy mice, the uterus undergoes significant structural and functional changes driven by hormones and other regulatory factors [ 19 ]. Control mice exhibited characteristic cyclic uterine remodeling, including dynamic epithelial proliferation and stromal expansion during proestrus to estrus, followed by tissue regression in diestrus. In contrast, Rptor -deficient mice displayed a persistent diestrus-like atrophic morphology (Fig.  2 A, B), indicating a fundamental loss of endometrial cycling capacity. As previously reported, Rptor -deficient mice exhibited normal follicle development and hormone levels comparable to those of the control mice [ 20 ]. Consequently, the disruption in endometrial cycling observed in Rptor fl/fl Pgr cre/ + mice cannot be attributed to impairments in ovarian function. Advanced imaging quantification through T2-weighted MRI with 3D volumetric reconstruction objectively demonstrated significant uterine hypoplasia in Rptor fl/fl Pgr cre/ + mice (Fig.  2 C, D). Notably, despite possessing definitive uterine compartments-including the epithelium, stroma, and myometrium, Rptor fl/fl Pgr cre/ + mice exhibited significant decreases in luminal epithelial height and endometrial diameter, as shown by quantitative immunofluorescence of endometrial compartment markers (Fig.  2 E, F). Moreover, immunofluorescence analysis demonstrated high FOXA2 expression (a master regulator of uterine gland differentiation) in control mice, whereas Rptor fl/fl Pgr cre/ + mice exhibited significant reduction in FOXA2-positive cells accompanied by a marked decrease in glandular density (Fig.  2 G, H), suggesting critical defects in glandular development following Raptor ablation. This morphological aberration was further corroborated at the transcriptional level, with qPCR analysis showing significant downregulation of key adenogenesis markers in Rptor fl/fl Pgr cre/ + mice compared with controls (Fig.  2 I). These findings indicate that Raptor-mediated mTORC1 signaling is indispensable for proper glandular morphogenesis through coordinated regulation of FOXA2 expression and gland-specific transcriptional programs. In addition, the expression of CD31, a marker for endothelial cells lining blood vessels, was decreased in Rptor fl/fl Pgr cre/ + mice compared with controls, implying that Rptor ablation affected the formation of uterine blood vessels (Fig.  2 J, K). These multi-dimensional defects collectively establish Rptor deficiency as a novel molecular mechanism disrupting the coordinated epithelial–stromal–vascular interactions essential for endometrial homeostasis, thereby accurately modeling the pathology of human thin endometrium syndrome. Fig. 2 The uterine phenotype of adult female mice with Rptor defects. Representative images of the reproductive tract of adult Rptor fl/fl and Rptor fl/fl Pgr cre/ + mice across various estrus stages. Scale bars: 0.5 cm. B. Histological examination of uterine structure throughout the mouse estrus cycle, using HE staining. Scale bars: 200 μm. C. Three-dimensional (3D) renderings of the female reproductive tract in adult Rptor f/f and Rptor fl/fl Pgr cre/ + mice, based on MRI T2-weighted images. Scale bars: 5 mm. D. Precise measurements of uterine horn diameters are derived from the 3D renderings in (C) ( n  = 6). The results represent the mean ± SD. *** p  = 0.0004 by a two-tailed Student’s t -test. E. Immunostaining analysis of cytokeratin 8 (CK8, red) and α-smooth muscle actin (α-SMA, green) in the uteri of Rptor fl/fl and Rptor fl/fl Pgr cre/ + mice at diestrus stage. Nuclei are stained with DAPI. Scale bars: 250  μm. F. Quantitative assessment of endometrial area, luminal epithelial height, and the proportions of stromal and myometrial areas within the endometrial area in Rptor fl/fl ( n  = 8) and Rptor fl/fl Pgr cre/ + mice ( n  = 5). The results represent the mean ± SD. ** p  = 0.0034 (endometrial area), * p  = 0.0129 (luminal epithelial height), by a two-tailed Student’s t -test. G. Immunostaining analysis of FOXA2 in the uteri of Rptor fl/fl and Rptor fl/fl Pgr cre/ + mice at diestrus stage. Nuclei are stained with DAPI. Scale bars: 100  μm. H. Quantitative analysis of FOXA2-positive glands per section in the uteri of adult female Rptor fl/fl (24 sections from six mice) and Rptor fl/fl Pgr cre/ + mice (10 sections from five mice). The results represent the mean ± SD. *** p  < 0.0001 by a two-tailed Student’s t -test. I. qPCR analysis of gene expression related to gland development in the uteri of adult female Rptor fl/fl ( n  = 6) and Rptor fl/fl Pgr cre/ + ( n  = 5) mice at the diestrus stage. The results represent the mean ± SD. Statistical significance was determined by a two-tailed Student’s t -test. Wnt5a (* p  = 0.0128); Wnt11 (** p  = 0.0025) ; Fzd6 (* p  = 0.0236) ; Ctnnb1 (* p  = 0.0393) ; Lef1 (*** p  = 0.0004) ; Hoxa11 (* p  = 0.0164) ; Vangl2 (* p  = 0.0115) ; Cdh1 (** p  = 0.0069) ; Wfdc3 (* p  = 0.0397) . J. Immunofluorescence staining for CD31 in the uteri of adult female Rptor fl/fl and Rptor fl/fl Pgr cre/ + mice at the diestrus stage. Nuclei are stained with DAPI. Scale bars: 100  μm. K. The percentage of CD31 + blood vessel area in the uterine stromal of adult female Rptor fl/fl (17 sections from six mice) and Rptor fl/fl Pgr cre/ + mice (seven sections from five mice). Data are presented as mean ± SD. ** p  = 0.0017 by a two-tailed Student’s t -test The uterine phenotype of adult female mice with Rptor defects. Representative images of the reproductive tract of adult Rptor fl/fl and Rptor fl/fl Pgr cre/ + mice across various estrus stages. Scale bars: 0.5 cm. B. Histological examination of uterine structure throughout the mouse estrus cycle, using HE staining. Scale bars: 200 μm. C. Three-dimensional (3D) renderings of the female reproductive tract in adult Rptor f/f and Rptor fl/fl Pgr cre/ + mice, based on MRI T2-weighted images. Scale bars: 5 mm. D. Precise measurements of uterine horn diameters are derived from the 3D renderings in (C) ( n  = 6). The results represent the mean ± SD. *** p  = 0.0004 by a two-tailed Student’s t -test. E. Immunostaining analysis of cytokeratin 8 (CK8, red) and α-smooth muscle actin (α-SMA, green) in the uteri of Rptor fl/fl and Rptor fl/fl Pgr cre/ + mice at diestrus stage. Nuclei are stained with DAPI. Scale bars: 250  μm. F. Quantitative assessment of endometrial area, luminal epithelial height, and the proportions of stromal and myometrial areas within the endometrial area in Rptor fl/fl ( n  = 8) and Rptor fl/fl Pgr cre/ + mice ( n  = 5). The results represent the mean ± SD. ** p  = 0.0034 (endometrial area), * p  = 0.0129 (luminal epithelial height), by a two-tailed Student’s t -test. G. Immunostaining analysis of FOXA2 in the uteri of Rptor fl/fl and Rptor fl/fl Pgr cre/ + mice at diestrus stage. Nuclei are stained with DAPI. Scale bars: 100  μm. H. Quantitative analysis of FOXA2-positive glands per section in the uteri of adult female Rptor fl/fl (24 sections from six mice) and Rptor fl/fl Pgr cre/ + mice (10 sections from five mice). The results represent the mean ± SD. *** p  < 0.0001 by a two-tailed Student’s t -test. I. qPCR analysis of gene expression related to gland development in the uteri of adult female Rptor fl/fl ( n  = 6) and Rptor fl/fl Pgr cre/ + ( n  = 5) mice at the diestrus stage. The results represent the mean ± SD. Statistical significance was determined by a two-tailed Student’s t -test. Wnt5a (* p  = 0.0128); Wnt11 (** p  = 0.0025) ; Fzd6 (* p  = 0.0236) ; Ctnnb1 (* p  = 0.0393) ; Lef1 (*** p  = 0.0004) ; Hoxa11 (* p  = 0.0164) ; Vangl2 (* p  = 0.0115) ; Cdh1 (** p  = 0.0069) ; Wfdc3 (* p  = 0.0397) . J. Immunofluorescence staining for CD31 in the uteri of adult female Rptor fl/fl and Rptor fl/fl Pgr cre/ + mice at the diestrus stage. Nuclei are stained with DAPI. Scale bars: 100  μm. K. The percentage of CD31 + blood vessel area in the uterine stromal of adult female Rptor fl/fl (17 sections from six mice) and Rptor fl/fl Pgr cre/ + mice (seven sections from five mice). Data are presented as mean ± SD. ** p  = 0.0017 by a two-tailed Student’s t -test The implantation window in mice generally occurs around gestation day (GD) 4–5, marked by a transition from an estrogen (E2)-driven proliferative phase to a P4-responsive state [ 21 ]. To examine the regulatory role of mTORC1 signaling in endometrial receptivity, we first evaluated the expression patterns of Raptor and phosphorylated S6Ser235/236 residues during this period. Immunofluorescence analysis indicated a significant upregulation of both Raptor and its downstream effector, p-S6Ser235/236, in wild-type mice at GD4 compared with GD1 (Fig.  3 A, B), suggesting active mTORC1 signaling during this critical period. As detailed in our previous publication [ 20 ], these Rptor fl/fl Pgr cre/ + mice persist in a diestrus-like state, characterized by an atrophied epithelium and a significantly reduced external vaginal orifice. Consequently, natural mating and the formation of vaginal plugs do not occur, resulting in infertility. Therefore, we utilized an ovariectomized model with precisely timed hormone replacement to mimic GD4 endometrial conditions [ 15 ] (Fig.  3 C). Remarkably, while control mice exhibited the expected proliferative quiescence in the luminal epithelium (Ki67-negative) with active stromal proliferation (Ki67-positive), Rptor -deficient mice showed significantly impaired stromal cell proliferation (Fig.  3 D, E), indicating the role of mTORC1 in activating the stromal compartment. Previous studies have suggested that the loss of luminal epithelial cell polarity facilitates embryo implantation, whereas preserved polarity may impede it [ 22 ]. However, both Rptor fl/fl Pgr cre/ + mice and controls maintained similar levels of the cell polarity marker E-cadherin (Fig.  3 F, G) and Muc1(Supplementary Fig. S2A, B) expression, suggesting that mTORC1 signaling does not control uterine luminal epithelial polarity. Since P4 receptor (PR) downregulation is associated with endometrial receptivity [ 23 ], we observed a slight increase in PR density in the luminal epithelium of Rptor fl/fl Pgr cre/ + mice compared with controls (Fig.  3 H, I). Recognizing that Indian Hedgehog (IHH) is a major mediator of progesterone signaling for epithelial-stromal crosstalk, we investigated this specific axis. Immunohistochemistry analyses revealed a significant downregulation of IHH in Rptor -deficient uteri (Supplementary Fig. S2C, D). Consistent with the marked reduction in glandular density (Fig.  2 E), the transcript levels of Lif , a gland-derived cytokine essential for implantation, were also significantly diminished in Rptor -deficient uteri (Supplementary Fig. S2E). In contrast, other receptivity markers such as Ptch1 and Hand2 showed no significant alterations (Supplementary Fig. S2E). Notably, uterine estrogen receptor (ER) levels remained unchanged between the groups (Fig.  3 J, K), indicating that mTORC1 signaling does not regulate ER expression. These findings demonstrate that uterine mTORC1 signaling specifically coordinates stromal proliferation and PR downregulation, but not epithelial polarity or ER expression, during the establishment of endometrial receptivity. Fig. 3 The specific deletion of Rptor in the uterus impairs endometrial receptivity . A, B. Immunofluorescence staining for Raptor (green) ( A ) and p-S6Ser235/236 (red) ( B ) in the uteri of WT adult mice on GD1 and GD4. Nuclei were counterstained with DAPI. Scale bars: 75  μm. C. Schematic representation of the artificial pregnancy induction experiment (Pollard experiment). D, F, H, J. Representative immunofluorescence images of Ki67 ( D ), E-cadherin ( F ), PR ( H ), and ER ( J ) in the uteri of Rptor fl/fl and Rptor fl/fl Pgr cre/ + mice following the Pollard experiment. Nuclei were counterstained with DAPI. Scale bars: 75 μm. E, G, I, K. Quantification of the percentage of Ki67 + cells ( E ), fluorescence intensity of E-cadherin ( G ), PR ( I ), and ER ( K ) in the uteri of Rptor fl/fl ( n  = 4) and Rptor fl/fl Pgr cre/ + mice ( n  = 5) using ImageJ software. The results represent the mean ± SD. *** p  < 0.0001 (Ki67), ** p  = 0.0021 (Epithelial PR) by a two-tailed Student’s t -test The specific deletion of Rptor in the uterus impairs endometrial receptivity . A, B. Immunofluorescence staining for Raptor (green) ( A ) and p-S6Ser235/236 (red) ( B ) in the uteri of WT adult mice on GD1 and GD4. Nuclei were counterstained with DAPI. Scale bars: 75  μm. C. Schematic representation of the artificial pregnancy induction experiment (Pollard experiment). D, F, H, J. Representative immunofluorescence images of Ki67 ( D ), E-cadherin ( F ), PR ( H ), and ER ( J ) in the uteri of Rptor fl/fl and Rptor fl/fl Pgr cre/ + mice following the Pollard experiment. Nuclei were counterstained with DAPI. Scale bars: 75 μm. E, G, I, K. Quantification of the percentage of Ki67 + cells ( E ), fluorescence intensity of E-cadherin ( G ), PR ( I ), and ER ( K ) in the uteri of Rptor fl/fl ( n  = 4) and Rptor fl/fl Pgr cre/ + mice ( n  = 5) using ImageJ software. The results represent the mean ± SD. *** p  < 0.0001 (Ki67), ** p  = 0.0021 (Epithelial PR) by a two-tailed Student’s t -test Decidualization is a critical postimplantation process characterized by extensive stromal cell proliferation and differentiation into specialized decidual cells that provide essential embryonic support [ 24 ]. Immunofluorescence revealed significant upregulation of both Raptor and its downstream target p-S6Ser235/236 in WT uteri during natural decidualization at GD5 compared with GD1 (Fig.  4 A, B). To overcome the reproductive impairments in Rptor fl/fl Pgr cre/ + mice, we established an artificial decidualization model that recapitulated the robust induction of mTORC1 signaling (Raptor and p-S6Ser235/236) in WT uteri following oil stimulation (Fig.  4 C–E). Notably, while control uteri exhibited characteristic expansion and histological transformation of stromal compartments, Rptor -deficient uteri remained completely unresponsive (Fig.  4 F–H). Molecular profiling confirmed this functional dichotomy: oil-stimulated control horns showed marked upregulation of key decidual markers ( Wnt4, Bmp8a, Prl8a2, and Bmp2 ), whereas Rptor -deficient uteri failed to activate this genetic program (Fig.  4 I). To rigorously distinguish between a delayed decidual response and a complete physiological failure, we further analyzed the temporal dynamics at decidual day 2 (early phase) and day 5 (peak phase). Time-course analysis of gross morphology and uterine weight revealed that while control uteri underwent progressive expansion from day 2 to day 5, Rptor -deficient uteri remained completely unresponsive at both time points (Supplementary Fig. S3A, B). Ki67 immunohistochemistry showed that in the antimesometrial (AM) zone of control mice, stromal cells differentiated into decidual cells characterized by enlarged mono or bi-nucleated cells morphology (polyploidization) at decidual day 2 and day 5. Notably, cell proliferation in controls decreased from day 2 to day 5, indicating physiological cell cycle exit. In contrast, Rptor -deficient cells retained a fibroblast-like morphology and maintained a consistent proliferative status (Supplementary Fig. S3C). Together, these data demonstrate a complete blockade of the decidual program rather than a delay, thereby establishing mTORC1 signaling as a master regulator of stromal cell decidualization that coordinates both the morphological transformation and genetic reprogramming essential for successful embryo implantation and maintenance. Fig. 4 Uterine-specific deletion of Rptor impairs decidualization . A, B. Immunofluorescence staining of Raptor (green) ( A ) and p-S6Ser235/236 (red) ( B ) in the uteri of WT adult female mice GD1 and GD5, with nuclei counterstained with DAPI. Scale bars: 75 μm. C. Schematic outline of the artificial decidualization experimental procedure. D, E. Immunofluorescence staining of Raptor (red) ( D ) and p-S6Ser235/236 (red) ( E ) in both unstimulated and stimulated uterine horns of WT adult female mice, with nuclei counterstained with DAPI. Scale bars: 75 μm. F. Representative images of the gross morphology from oil-treated uterine horns (right horn) and untreated uterine horns (left horn) from Rptor fl/fl and Rptor fl/fl Pgr cre/ + mice collected 5 days after oil injection. Scale bars: 0.5 cm. G. Comparative analysis of weight per unit length between oil-stimulated and unstimulated uterine horns from Rptor fl/fl ( n  = 8) and Rptor fl/fl Pgr cre/ + mice ( n  = 6). The results represent the mean ± SD. *** p  < 0.0001 by Two-way analysis of variance (ANOVA) with Tukey’s multiple comparisons test. H. Histological analysis of oil-stimulated and unstimulated uterine horns from Rptor fl/fl and Rptor fl/fl Pgr cre/ + mice, visualized through HE staining. Scale bars: 200  μm. I. Relative mRNA expression levels of decidualization marker genes ( Wnt4 , Bmp8a , Bmp2 , and Prl8a2 ) in oil-stimulated versus unstimulated uterine horns from Rptor fl/fl ( n  = 5) and Rptor fl/fl Pgr cre/ + mice ( n  = 5). Data in are presented as mean ± SD. Wnt4 (*** p  < 0.0001), Bmp8a (*** p  < 0.0001), Bmp2 (*** p  < 0.0001), Prl8a2 (** p  = 0.0013) by a Two-way ANOVA with Tukey’s multiple comparisons test Uterine-specific deletion of Rptor impairs decidualization . A, B. Immunofluorescence staining of Raptor (green) ( A ) and p-S6Ser235/236 (red) ( B ) in the uteri of WT adult female mice GD1 and GD5, with nuclei counterstained with DAPI. Scale bars: 75 μm. C. Schematic outline of the artificial decidualization experimental procedure. D, E. Immunofluorescence staining of Raptor (red) ( D ) and p-S6Ser235/236 (red) ( E ) in both unstimulated and stimulated uterine horns of WT adult female mice, with nuclei counterstained with DAPI. Scale bars: 75 μm. F. Representative images of the gross morphology from oil-treated uterine horns (right horn) and untreated uterine horns (left horn) from Rptor fl/fl and Rptor fl/fl Pgr cre/ + mice collected 5 days after oil injection. Scale bars: 0.5 cm. G. Comparative analysis of weight per unit length between oil-stimulated and unstimulated uterine horns from Rptor fl/fl ( n  = 8) and Rptor fl/fl Pgr cre/ + mice ( n  = 6). The results represent the mean ± SD. *** p  < 0.0001 by Two-way analysis of variance (ANOVA) with Tukey’s multiple comparisons test. H. Histological analysis of oil-stimulated and unstimulated uterine horns from Rptor fl/fl and Rptor fl/fl Pgr cre/ + mice, visualized through HE staining. Scale bars: 200  μm. I. Relative mRNA expression levels of decidualization marker genes ( Wnt4 , Bmp8a , Bmp2 , and Prl8a2 ) in oil-stimulated versus unstimulated uterine horns from Rptor fl/fl ( n  = 5) and Rptor fl/fl Pgr cre/ + mice ( n  = 5). Data in are presented as mean ± SD. Wnt4 (*** p  < 0.0001), Bmp8a (*** p  < 0.0001), Bmp2 (*** p  < 0.0001), Prl8a2 (** p  = 0.0013) by a Two-way ANOVA with Tukey’s multiple comparisons test To systematically elucidate the molecular pathogenesis of thin endometrium resulting from mTORC1 signaling inactivation, while accounting for hormonal variations during the estrous cycle, we conducted a comprehensive uterine transcriptomic analysis in OVX Rptor fl/fl Pgr cre/ + ( n  = 3) and Rptor fl/fl ( n  = 3) mice, 24 h postestrogen stimulation. The transcriptome data analysis revealed 2178 significantly differentially expressed genes (DEGs) (|fold change|> 2, q  < 0.05) in the Rptor -deficient group compared with the control group, consisting of 1350 upregulated and 828 downregulated transcripts (Fig.  5 A, B). Functional annotation indicated that these DEGs were predominantly associated with steroid and cholesterol metabolism, with Gene Ontology (GO) analysis showing significant enrichment in biological processes such as “steroid metabolic process,” “cholesterol metabolic process,” and “cholesterol biosynthetic process.” At the molecular function level, we observed significant enrichment for “cholesterol transfer activity,” “lipid transfer activity,” and “cholesterol binding” (Fig.  5 C, Supplementary Fig. S4). Consistent with these observations, pathway enrichment analysis of the downregulated genes using the MSigDB gene sets revealed that these suppressed transcripts were significantly enriched in “cholesterol biosynthesis,” “metabolism of steroids,” “metabolism of lipids,” and “steroid biosynthesis” (Fig.  5 D). These convergent results from multiple analytical approaches strongly implicate mTORC1 signaling as a central regulator of endometrial metabolic homeostasis, primarily through its governance of the sterol–lipid axis that orchestrates cholesterol biosynthesis and metabolic flux. Fig. 5 Deletion of Rptor specific to the uterus alters gene expression profiles . A. Volcano plot of genes upregulated (red) or downregulated (blue) by at least twofold in the uteri of OVX Rptor fl/fl Pgr cre/ + mice ( n  = 3) compared with Rptor fl/fl mice ( n  = 3) treated with 1 day of E2 treatment. B. Heatmap of differentially expressed genes ( n  = 3 mice per group). C. Gene Ontology (GO) analysis of the differentially expressed genes in the uterus of OVX Rptor fl/fl Pgr cre/ + mice compared with Rptor fl/fl mice with  day of E2 treatment. D. Pathway enrichment of downregulated genes in the uterus of OVX Rptor fl/fl Pgr cre/ + mice compared with Rptor fl/fl mice following 1 day of E2 treatment via MSigDB gene sets Deletion of Rptor specific to the uterus alters gene expression profiles . A. Volcano plot of genes upregulated (red) or downregulated (blue) by at least twofold in the uteri of OVX Rptor fl/fl Pgr cre/ + mice ( n  = 3) compared with Rptor fl/fl mice ( n  = 3) treated with 1 day of E2 treatment. B. Heatmap of differentially expressed genes ( n  = 3 mice per group). C. Gene Ontology (GO) analysis of the differentially expressed genes in the uterus of OVX Rptor fl/fl Pgr cre/ + mice compared with Rptor fl/fl mice with  day of E2 treatment. D. Pathway enrichment of downregulated genes in the uterus of OVX Rptor fl/fl Pgr cre/ + mice compared with Rptor fl/fl mice following 1 day of E2 treatment via MSigDB gene sets Transcriptomic profiling revealed significant dysregulation of estrogen metabolic pathways in Rptor -deficient endometria (Fig.  5 C, S4). To assess estrogen-regulated transcriptional changes, we integrated differentially expressed genes from the GSE23072 dataset (24 h E2 treatment) with our RNA-seq data [ 25 ]. Cross-dataset analysis demonstrated attenuated activation of E2-induced genes and enhanced expression of E2-repressed genes in Rptor -deficient uterine relative to controls (Supplementary Fig. S5A, B). This was corroborated by GSEA showing negative enrichment of “REACTOME_Estrogen_ Dependent_Gene_Expression” in Rptor fl/fl Pgr cre/ + mice (Fig.  6 A). Consistent with this, transcriptional profiles derived from human thin endometrium datasets ( GSE160633 ) exhibited a suppression of estrogen-responsive signatures (Fig.  6 B). Quantitative real-time PCR analysis revealed that the expression levels of key estrogen-responsive genes ( MKI67, PCNA, CCND1, CCNE1, CCNA2, GREB1, TFF3 ) in thin endometrial tissue were significantly lower than those in normal endometrial tissue (Supplementary Fig. S6). To explore potential regulatory interactions between mTOR activity and estrogen signaling, we first analyzed the expression of Raptor and p-S6Ser235/236 across different estrous cycle phases. The longitudinal studies revealed that both Raptor protein and its downstream effector p-S6Ser235/236 exhibited cycle-phase-dependent expression dynamics, peaking during the proestrus and estrus stages, which coincided with serum estradiol fluctuations (Supplementary Fig. S7A, B). Next, we compared the uterine morphology of mice at postnatal day 28 (PND28) before puberty. In this case, ovarian hormones had the least effect on mice. Notably, prepubertal Rptor fl/fl Pgr cre/ + mice exhibited comparable uterine morphology, structure, and glandular density to controls (Supplementary Fig. S8A–D). Consistently, adenogenesis-related transcripts were comparable between prepubertal Rptor fl/fl Pgr cre/ + mice and controls (Supplementary Fig. S8E). Ovariectomy experiments further demonstrated that estradiol supplementation induced robust mTORC1 activation in WT mice but failed to activate this pathway in Rptor fl/fl Pgr cre/ + mice (Supplementary Fig. S9A–D), suggesting the role of mTORC1 manifests primarily during hormone-responsive phases. To further investigate the role of mTORC1 signaling in estrogen-mediated endometrial remodeling, we evaluated tissue homeostasis in OVX mice before and after a 3-day estrogen regimen. Compared with the control group, Rptor -deficient uteri failed to undergo estrogen-induced expansion, with attenuated morphological differentiation, weight gain, and epithelial thickening (Fig.  6 C, D, Supplementary Fig. S10A). In the absence of E2, PR is expressed in epithelial cells, and E2 downregulates PR expression. While estrogen downregulated epithelial PR expression in controls, this regulation was abolished in Rptor fl/fl Pgr cre/ + mice (Fig.  6 E). Equivalent ERα expression between genotypes excluded receptor-level differences as a contributing factor (Supplementary Fig. S10B). These findings position mTORC1 as an essential point integrating metabolic and hormonal signals to regulate endometrial remodeling, where its ablation disrupts estrogen responsiveness independent of receptor availability. Fig. 6 Depletion of Rptor inhibits estrogen response and cell proliferation . A. GSEA of the “Estrogen Dependent Gene Expression” gene set in uterine tissue from OVX Rptor fl/fl Pgr cre/ + mice ( n  = 3) relative to Rptor fl/fl mice ( n  = 3) 1 day of E2 treatment. B. GSEA enrichment plots of “Estrogen Dependent Gene Expression” gene set in endometrial samples from control and thin endometrium patients ( GSE160633 ). C. Representative images of the HE staining of the uteri from Rptor fl/fl Pgr cre/ + and Rptor fl/fl mice administrated daily with E2 for 3 consecutive days. Scale bars: 200 μm. D. Quantitative analysis of the endometrial diameter and the uteri weight in the Rptor fl/fl Pgr cre/ + ( n  = 8) and Rptor fl/fl ( n  = 8) mice. Data are presented as mean ± SD. *** p  < 0.0001 by Two-way ANOVA with Tukey’s multiple comparisons test. E. Representative immunohistochemical images of PR protein in the uteri of OVX Rptor fl/fl Pgr cre/ + and Rptor fl/fl mice before and after E2 stimulation. Scale bars: 200 μm. F. Proliferation-related pathways were enriched in the uteri of OVX Rptor fl/fl Pgr cre/ + mice, compared with Rptor fl/fl mice subjected to a 1-day E2 treatment, as indicated by GSEA analysis, with the Normalized Enrichment Scores (NES) displayed in a radar chart. G, H. Representative immunohistochemistry images ( G ) and the percentage of Ki67 + epithelial cells ( H ) in the uteri of OVX Rptor fl/fl ( n  = 5) and Rptor fl/fl Pgr cre/ + ( n  = 5) mice with 1 day of E2 administration. Scale bars: 200 μm. The results represent the mean ± SD. *** p  < 0.0001 by a two-tailed Student’s t -test. I, J. Representative immunofluorescence images ( I ) and the percentage of Brdu + epithelial cells ( J ) in the uteri of OVX Rptor fl/fl ( n  = 5) and Rptor fl/fl Pgr cre/ + ( n  = 6) with 1 day of E2 administration. Scale bars: 100 μm. The results represent the mean ± SD. *** p  < 0.0001 by a two-tailed Student’s t -test. K, L. Representative of flow cytometric plots of CFSE-dilution analysis ( K ) and quantification ( L ) of primary uterus stromal cells isolated from Rptor fl/fl Pgr cre/ + ( n  = 4) and Rptor fl/fl ( n  = 4) mice treated with or without E2 for 48 h. The results represent the mean ± SD. *** p  < 0.0001, ** p  = 0.0076, by two-way ANOVA with Tukey’s multiple comparisons test. M, N. Representative of flow cytometric plots ( M ) and quantification ( N ) of Ki67 expression in primary uterus stromal cells isolated from Rptor fl/fl Pgr cre/ + ( n  = 3) and Rptor fl/fl ( n  = 3) treated with or without E2 for 48 h. The results represent the mean ± SD. *** p  = 0.0009 ( Rptor fl/fl Vehicle versus E2), * p  = 0.0465 ( Rptor fl/fl Pgr cre/ + vs. Rptor fl/fl under Vehicle), ** p  = 0.0034 ( Rptor fl/fl Pgr cre/ + vs. Rptor fl/fl under E2), ** p  = 0.0092 ( Rptor fl/fl Pgr cre/ + Vehicle versus E2), by two-way ANOVA with Tukey’s multiple comparisons test Depletion of Rptor inhibits estrogen response and cell proliferation . A. GSEA of the “Estrogen Dependent Gene Expression” gene set in uterine tissue from OVX Rptor fl/fl Pgr cre/ + mice ( n  = 3) relative to Rptor fl/fl mice ( n  = 3) 1 day of E2 treatment. B. GSEA enrichment plots of “Estrogen Dependent Gene Expression” gene set in endometrial samples from control and thin endometrium patients ( GSE160633 ). C. Representative images of the HE staining of the uteri from Rptor fl/fl Pgr cre/ + and Rptor fl/fl mice administrated daily with E2 for 3 consecutive days. Scale bars: 200 μm. D. Quantitative analysis of the endometrial diameter and the uteri weight in the Rptor fl/fl Pgr cre/ + ( n  = 8) and Rptor fl/fl ( n  = 8) mice. Data are presented as mean ± SD. *** p  < 0.0001 by Two-way ANOVA with Tukey’s multiple comparisons test. E. Representative immunohistochemical images of PR protein in the uteri of OVX Rptor fl/fl Pgr cre/ + and Rptor fl/fl mice before and after E2 stimulation. Scale bars: 200 μm. F. Proliferation-related pathways were enriched in the uteri of OVX Rptor fl/fl Pgr cre/ + mice, compared with Rptor fl/fl mice subjected to a 1-day E2 treatment, as indicated by GSEA analysis, with the Normalized Enrichment Scores (NES) displayed in a radar chart. G, H. Representative immunohistochemistry images ( G ) and the percentage of Ki67 + epithelial cells ( H ) in the uteri of OVX Rptor fl/fl ( n  = 5) and Rptor fl/fl Pgr cre/ + ( n  = 5) mice with 1 day of E2 administration. Scale bars: 200 μm. The results represent the mean ± SD. *** p  < 0.0001 by a two-tailed Student’s t -test. I, J. Representative immunofluorescence images ( I ) and the percentage of Brdu + epithelial cells ( J ) in the uteri of OVX Rptor fl/fl ( n  = 5) and Rptor fl/fl Pgr cre/ + ( n  = 6) with 1 day of E2 administration. Scale bars: 100 μm. The results represent the mean ± SD. *** p  < 0.0001 by a two-tailed Student’s t -test. K, L. Representative of flow cytometric plots of CFSE-dilution analysis ( K ) and quantification ( L ) of primary uterus stromal cells isolated from Rptor fl/fl Pgr cre/ + ( n  = 4) and Rptor fl/fl ( n  = 4) mice treated with or without E2 for 48 h. The results represent the mean ± SD. *** p  < 0.0001, ** p  = 0.0076, by two-way ANOVA with Tukey’s multiple comparisons test. M, N. Representative of flow cytometric plots ( M ) and quantification ( N ) of Ki67 expression in primary uterus stromal cells isolated from Rptor fl/fl Pgr cre/ + ( n  = 3) and Rptor fl/fl ( n  = 3) treated with or without E2 for 48 h. The results represent the mean ± SD. *** p  = 0.0009 ( Rptor fl/fl Vehicle versus E2), * p  = 0.0465 ( Rptor fl/fl Pgr cre/ + vs. Rptor fl/fl under Vehicle), ** p  = 0.0034 ( Rptor fl/fl Pgr cre/ + vs. Rptor fl/fl under E2), ** p  = 0.0092 ( Rptor fl/fl Pgr cre/ + Vehicle versus E2), by two-way ANOVA with Tukey’s multiple comparisons test Studies have confirmed that estrogen stimulation promotes the proliferation and differentiation of endometrial cells. Consistently, the majority of transcripts linked to “KEGG_CELL_CYCLE” and “REACTOME_DNA_REPLICATION” were downregulated, evident from the overlaid volcano plot derived from our RNA-seq data (Supplementary Fig. S11A, B). Further analysis uncovered a notable decrease in several terms related to cellular replication and proliferation in the uteri of Rptor fl/fl Pgr cre/ + mice compared with the control group, including “DNA_REPLICATION,” “PROLIFERATION,” “S_PHASE,” “CELL_CYCLE,” “CELL_CYCLE_G2_M,” “MITOTIC_SPINDLE,” “MITOTIC_PROMETAPHASE,” and “CELL_CYCLE_M_G1” (Fig.  6 F). As anticipated, enrichment analysis of downregulated genes in the E2-induced group also showed enrichment in "CELL_CYCLE" (Supplementary Fig. S12A). These observations suggest that deletion of Raptor leads to suppression of the cell cycle and cell proliferation in the uterus. We also assessed the expression levels of COUP-TFII (NR2F2) and IHH, both recognized as key regulators of hormone-dependent proliferation in the uterus. Our findings revealed that the expression level of the Ihh gene was reduced in Rptor -deficient endometrial cells compared with the control (Supplementary Fig. S13A). In addition, immunohistochemical analysis demonstrated a considerable decline in COUP-TFII expression of Rptor -deficient mice relative to the control mice (Supplementary Fig. S13B, C). Furthermore, the proliferation of uterine epithelial cells was corroborated by assessing specific biomarkers, such as Ki67-labeled G1-S phase and BrdU-labeled S phase. Consistent with our expectations, the proportions of uterine epithelia exhibiting positive signals for Ki67 and BrdU decreased in OVX Rptor fl/fl Pgr cre/ + mice compared with control mice after E2 treatment, accounting for the impaired growth (Fig.  6 G–J). The validity of these findings was further reinforced through proliferation analysis conducted on thin endometrium samples. Notably, when compared with the normal control group, the patients with thin endometrium exhibited a markedly diminished proliferation activity in their endometrium cells (Supplementary Fig. S14A, B). Further in vitro studies have shown that the proliferation and division ability of endometrial cells lacking Raptor is significantly lower than that of the control group. Moreover, estrogen strongly stimulated the proliferation of endometrial cells in control mice, but Raptor-deficient cells showed underlying proliferation impairment and complete estrogen nonresponse (Fig.  6 K–N). These findings collectively demonstrate that the ablation of Raptor in the uterus disrupts estrogen signaling transduction and compromises cell proliferation, ultimately resulting in endometrial hypoplasia. The mTORC1 signaling pathway functions as a master metabolic regulator that integrates nutritional cues and growth factor signals to coordinate cellular proliferation and differentiation processes. Our comprehensive investigation into mTORC1-mediated uterine regeneration mechanisms uncovered profound alterations in cholesterol homeostasis-related gene expression following Rptor deletion (Fig.  5 C, Supplementary Fig. S4). Initial GSEA demonstrated substantial downregulation of cholesterol biosynthesis and metabolic pathway genes in Rptor fl/fl Pgr cre/ + mice (Fig.  7 A, B), a finding that was further corroborated by transcriptomic analysis of thin endometrium datasets ( GSE160633 ), which revealed concomitant reductions in both inflammatory response genes and cholesterol regulatory elements (Fig.  7 C). The observed suppression of cholesterol biosynthesis is particularly noteworthy given the pathway’s complexity, involving at least 21 enzymatic steps with HMG-CoA reductase (HMGCR) and squalene epoxidase (SQLE) serving as critical rate-limiting enzymes that govern systemic cholesterol synthesis and circulating levels (Fig.  7 D). Supporting these findings, our RNA-seq data quantitatively confirmed the significant suppression of cholesterol biosynthesis and metabolic genes in Rptor fl/fl Pgr cre/ + mice relative to controls (Fig.  7 E). Furthermore, quantitative real-time PCR analysis revealed that the expression levels of key cholesterol synthesis genes ( HMGCR, HMGCS1, SQLE, SREBF2, DHCR24 ) were significantly lower in thin endometrium tissues compared with normal uterine tissues (Supplementary Fig. S15A). Furthermore, we found that under the combined treatment of E2 and statins, WT cells exhibited a strong accumulation of active n-SREBP2 (a key transcription factor regulating cholesterol biosynthesis), confirming their responsiveness to estrogen and metabolic cues. In contrast, Rptor cKO cells showed impaired n-SREBP2 production even under E2 stimulation (Supplementary Fig. S16A–D). This provides direct mechanistic evidence that mTORC1 signaling is essential for the generation of active nuclear SREBP2, thereby explaining the downregulation of cholesterol biosynthesis genes ( Hmgcr, Sqle ) observed in our RNA-seq data. Genes that were downregulated in Rptor -deficient endometria despite being normally estrogen-induced showed significant enrichment in “CHOLESTEROL_HOMEOSTASIS” and “CHOLESTEROL_BIOSYNTHESIS” pathways (Supplementary Fig. S12A), collectively establishing mTORC1 as a crucial nexus integrating metabolic and hormonal signals. Fig. 7 The mTORC1 signaling regulates endometrial proliferation through cholesterol biosynthesis. A, B. GSEA plots of significant downregulation of cholesterol biosynthesis ( A ) and homeostasis ( B ) pathways in uterine tissues from Rptor fl/fl Pgr cre/ + mice versus Rptor fl/fl mice. C. GSEA plots of “Inflammatory Response and Cholesterol Up” gene signatures in thin endometrium patient samples compared to controls ( GSE160633 ). D. Schematic representation of the cholesterol biosynthesis pathway from acetyl-CoA, with core pathway genes highlighted in blue and rate-limiting enzymes HMG-CoA reductase (HMGCR) and squalene epoxidase (SQLE) marked in red. E. Heatmap of RNA-seq data showing differential expression of cholesterol biosynthesis and homeostasis genes in Rptor fl/fl Pgr cre/ + mice versus Rptor fl/fl mice. F. Quantitative analysis of the number of uterine stromal cells isolated from Rptor fl/fl Pgr cre/ + ( n  = 4) and Rptor fl/fl ( n  = 4) mice treated with cholesterol or vehicle control for 96 h. Data represents three independent experiments. The results represent the mean ± SD. ** p  = 0.0027, *** p  < 0.0001, by two-way ANOVA with Tukey’s multiple comparisons test. G, H. Representative flow cytometric plots ( G ) and quantification ( H ) for CFSE dilution of the uterus stromal cells isolated from Rptor fl/fl Pgr cre/ + ( n  = 4) and Rptor fl/fl ( n  = 4) mice treated with cholesterol or vehicle control for 96 h. The results represent the mean ± SD. ** p  = 0.0028, *** p  < 0.0001, by two-way ANOVA with Tukey’s multiple comparisons test. I. Quantification of Ki67 + cells in the uterus stromal cells isolated from Rptor fl/fl Pgr cre/ + and Rptor fl/fl mice treated with cholesterol or vehicle control for 96 h. The results represent the mean ± SD. * p  = 0.019, *** p  < 0.0001, by two-way ANOVA with Tukey’s multiple comparisons test. J. Quantitative analysis of WT uterine stromal cell number treated with atorvastatin ( n  = 5) or DMSO ( n  = 5) for 96 h. The results represent the mean ± SD. *** p  = 0.001 by a two-tailed Student’s t -test. K, L. Representative flow cytometric plots ( K ) and quantification ( L ) of CFSE dilution of WT primary uterine stromal cells treated with atorvastatin ( n  = 5) or dimethyl sulfoxide (DMSO) ( n  = 5) for 96 h. The results represent the mean ± SD. *** p  < 0.0001 by a two-tailed Student’s t -test. M. Quantification of Ki67 + cells in primary uterine stromal cells treated with atorvastatin ( n  = 5) or DMSO ( n  = 5) for 96 h. The results represent the mean ± SD. *** p  < 0.0001 by a two-tailed Student’s t -test The mTORC1 signaling regulates endometrial proliferation through cholesterol biosynthesis. A, B. GSEA plots of significant downregulation of cholesterol biosynthesis ( A ) and homeostasis ( B ) pathways in uterine tissues from Rptor fl/fl Pgr cre/ + mice versus Rptor fl/fl mice. C. GSEA plots of “Inflammatory Response and Cholesterol Up” gene signatures in thin endometrium patient samples compared to controls ( GSE160633 ). D. Schematic representation of the cholesterol biosynthesis pathway from acetyl-CoA, with core pathway genes highlighted in blue and rate-limiting enzymes HMG-CoA reductase (HMGCR) and squalene epoxidase (SQLE) marked in red. E. Heatmap of RNA-seq data showing differential expression of cholesterol biosynthesis and homeostasis genes in Rptor fl/fl Pgr cre/ + mice versus Rptor fl/fl mice. F. Quantitative analysis of the number of uterine stromal cells isolated from Rptor fl/fl Pgr cre/ + ( n  = 4) and Rptor fl/fl ( n  = 4) mice treated with cholesterol or vehicle control for 96 h. Data represents three independent experiments. The results represent the mean ± SD. ** p  = 0.0027, *** p  < 0.0001, by two-way ANOVA with Tukey’s multiple comparisons test. G, H. Representative flow cytometric plots ( G ) and quantification ( H ) for CFSE dilution of the uterus stromal cells isolated from Rptor fl/fl Pgr cre/ + ( n  = 4) and Rptor fl/fl ( n  = 4) mice treated with cholesterol or vehicle control for 96 h. The results represent the mean ± SD. ** p  = 0.0028, *** p  < 0.0001, by two-way ANOVA with Tukey’s multiple comparisons test. I. Quantification of Ki67 + cells in the uterus stromal cells isolated from Rptor fl/fl Pgr cre/ + and Rptor fl/fl mice treated with cholesterol or vehicle control for 96 h. The results represent the mean ± SD. * p  = 0.019, *** p  < 0.0001, by two-way ANOVA with Tukey’s multiple comparisons test. J. Quantitative analysis of WT uterine stromal cell number treated with atorvastatin ( n  = 5) or DMSO ( n  = 5) for 96 h. The results represent the mean ± SD. *** p  = 0.001 by a two-tailed Student’s t -test. K, L. Representative flow cytometric plots ( K ) and quantification ( L ) of CFSE dilution of WT primary uterine stromal cells treated with atorvastatin ( n  = 5) or dimethyl sulfoxide (DMSO) ( n  = 5) for 96 h. The results represent the mean ± SD. *** p  < 0.0001 by a two-tailed Student’s t -test. M. Quantification of Ki67 + cells in primary uterine stromal cells treated with atorvastatin ( n  = 5) or DMSO ( n  = 5) for 96 h. The results represent the mean ± SD. *** p  < 0.0001 by a two-tailed Student’s t -test To mechanistically dissect cholesterol biosynthesis in uterine regeneration, we established highly purified primary uterine stromal cell cultures from both Rptor fl/fl Pgr cre/ + and control mice (Supplementary Fig. S17A). Functional assays demonstrated that while exogenous cholesterol supplementation robustly enhanced proliferative capacity in control stromal cells—as quantitatively evidenced by increased cell counts, accelerated CFSE dilution kinetics, and elevated Ki67 expression—it completely failed to rescue the proliferation defect in Rptor -deficient cells, which maintained baseline proliferation rates indistinguishable from untreated controls (Fig.  7 F–I). The effect of cholesterol on promoting the proliferation of stromal cells can also be observed to be similar in human endometrial stromal cells (Supplementary Fig. S18A–C). This striking differential response underscores the essentiality of mTORC1-mediated cholesterol biosynthesis rather than extracellular cholesterol availability for endometrial proliferation. This conclusion was further reinforced by reciprocal pharmacological experiments where Atorvastatin-mediated inhibition of HMG-CoA reductase in WT cells faithfully recapitulated the Rptor -deficient phenotype: manifesting as reduced cell numbers, impaired CFSE dilution profiles, and significantly diminished Ki67 + cell populations relative to untreated controls (Fig.  7 J–M). Taken together, these multifaceted experimental approaches provide compelling evidence that mTORC1 signaling critically regulates the hormonally-responsive proliferative capacity of endometrial cells through its precise control of intracellular cholesterol biosynthesis and metabolic homeostasis.

The study enrolled women aged 20–40 years with thin endometrium ( n  = 5), on the basis of predefined criteria: ultrasonographically confirmed endometrial thickness  2 consecutive failed embryo transfer cycles with persistent endometrial thinness (< 7 mm) across three cycles. Exclusion criteria comprised: intrauterine pathologies (adhesions, tuberculosis), endocrine disorders (thyroid dysfunction, hyperprolactinemia), pelvic irradiation history, or chronic systemic diseases. Age-matched controls ( n  = 5) exhibited normal endometrial parameters: mid-luteal thickness 8–14 mm, regular menstrual cycles (21–35 days), and preserved ovarian reserve. All participants underwent comprehensive screening including: infectious disease panel (human immunodeficiency virus (HIV), hepatitis B virus (HBV), hepatitis C virus (HCV), syphilis), chromosomal analysis, assessment of recent (< 3 months) hormonal therapy use, and evaluation for systemic comorbidities (diabetes, thyroid dysfunction). Rptor flox/flox mice (Jackson Laboratory, stock #013191) were crossed with progesterone receptor-Cre ( PR-Cre ) knock-in mice (Cyagen Biosciences, Guangzhou, China) to generate uterine-specific knockout mice ( Rptor fl/fl Pgr cre/ + ). Rptor fl/fl littermates served as controls. For pregnancy studies, control females were paired with proven-fertile WT males overnight; successful mating was confirmed by vaginal plug observation the following morning (designated GD 1). Uterine tissues were fixed in 4% paraformaldehyde (PFA; Servicebio, G1101-500ML) for 24 h at 4  °C, followed by paraffin embedding and sectioning at 5 μm thickness. Sections were dried at 37  °C overnight before processing. Standard hematoxylin and eosin (HE) staining was performed, with microscopic evaluation conducted using an Olympus BX53 bright-field microscope. Representative images were selected for presentation. The 8-week-old female mice (4–5 per genotype) were acclimated in specific pathogen free (SPF) conditions for 21 days prior to analysis. Daily vaginal cytology was performed between 09:00–10:00 using 200 μl phosphate buffered saline (PBS) lavage. Cell suspensions were spread on slides, air-dried at 37 °C (45 min), stained with 0.1% crystal violet (Sangon Biotech, E607309) for 2 min, and PBS-washed (3 × 2 min). Slides were mounted with neutral balsam (Solarbio, G8590-100ML) under 24 × 50 mm coverslips. Estrous cycle staging was determined at 40 × magnification using an Olympus BX53 microscope. Uterine imaging was conducted using a 9.4 T Bruker Biospec preclinical magnetic resonance imaging (MRI) system (Bruker Corp., USA) with a dedicated 35 mm mouse body radiofrequency coil. High-resolution T2-weighted anatomical images of the reproductive tract were acquired using optimized Spin Echo sequence parameters: TR = 3650 ms, TE = 22 ms, Echo Train Length = 8, Flip Angle = 180°, NEX = 4, and Slice Thickness = 0.5 mm. Image postprocessing included three-dimensional reconstruction and segmentation using 3D Slicer software (v4.11.20210226), followed by triplicate measurements of uterine cross-sectional diameters using the integrated digital caliper tool. All quantitative analyses were performed independently by two trained operators to ensure measurement reproducibility and minimize observer bias. For estradiol response evaluation, ovariectomized (OVX) mice received daily subcutaneous injections of 17β-estradiol (E2; Sigma E8875, 100 ng in 100 μl sesame oil) for three consecutive days, with tissue collection 24 h postfinal injection. In separate experiments: for RNA-seq analysis, uterine tissues were harvested from OVX Rptor fl/fl PR Cre/ + mice and littermate controls 24 h after a single 100 ng E2 injection; For mTORC1 activation assessment, OVX mice received a single 100 ng E2 dose and were euthanized 24 h later for immunofluorescence analysis of Raptor and phospho-S6 ribosomal protein (Ser235/236) expression patterns. The peri-implantation hormonal milieu was replicated in 6–8-week-old OVX mice using an established steroid replacement protocol [ 14 – 17 ]. Following bilateral ovariectomy under tribromoethanol anesthesia (200 μl/20 g IP), mice underwent a 14-day hormone washout period. Hormonal stimulation consisted of: daily 100 ng E2 injections for 2 days; 48 h rest; daily 1 mg P4 for 3 days and combined 100 ng E2 + 1 mg P4 on day 4. Mice were euthanized 15 h postfinal injection, with uteri collected for downstream analyses. Uterine decidualization capacity was assessed in OVX mice primed with daily 100 ng E2 injections for 3 days, followed by 48 h rest. Mice then received daily 1 mg P4 + 6.7 ng E2 injections for 3 days, On the third day of P4 + E2 priming, one uterine horn was injected with 50 μl of sesame oil to induce decidualization, while the contralateral horn served as the unstimulated control. To evaluate the temporal kinetics of the decidual response, mice were sacrificed at two distinct time points: decidual day 2 (DD2, 48 h poststimulation) representing the early proliferative phase, and decidual day 5 (DD5, 120 h poststimulation) representing the peak differentiation phase. Uterine wet weight was measured, and the ratio of the stimulated horn to the unstimulated horn was calculated. Uterine tissues were subsequently collected for histological (HE, IHC) and molecular (qPCR) analyses. For immunofluorescence, uterine paraffin sections (5 μm) were subjected to antigen retrieval by microwave irradiation in 10 mM sodium citrate buffer (pH 6.0, 0.05% Tween 20) for 15 min. After permeabilization with 0.2% Triton X-100/PBS (45 min), nonspecific binding was blocked with 1% BSA/10% goat serum in PBS (1 h, RT). Primary antibody incubations were performed overnight at 4  °C using: Raptor (Santa Cruz, sc-81537, 1:200), p-S6 (Cell Signaling Technology, E2R10, 1:400), α-SMA (Cell Signaling Technology, D4K9N, 1:800), Vimentin (Cell Signaling Technology, D21H3, 1:800), CK8 (DSHB, 1:800), CD31 (Cell Signaling Technology, 77699 T, 1:400), IHH (Immunoway, PT1268R,1:200), Muc1(PTM BIO, PTM-5373, 1:200), and E-cadherin (Cell Signaling Technology, 24E10, 1:600). Species-matched secondary antibodies (Jackson ImmunoResearch: Alexa Fluor 488 anti-rabbit 111–545-144, 1:500; Alexa Fluor 594 anti-mouse 115–585-146, 1:500) were applied for 1 h at RT. Nuclei were counterstained with DAPI (5 μg/ml, 5 min) before imaging on a Leica SP8 confocal microscope. For IHC, sections were incubated with: PR (Cell Signaling Technology, D8Q2J, 1:400), ERα (Abcam, ab32063, 1:400), Ki67 (Abcam, ab15580, 1:500), COUP-TFII (Selleck, F1422, 1:200), or FOXA2 (Abcam, ab108422, 1:600) overnight at 4 °C. After PBS washes, biotinylated secondaries (Vector Labs, BA-1000–1.5, 1 h RT) and Vectastain ABC reagent (Vector Labs, PK-6100, 30 min) were applied. DAB development (Vector Labs, SK-4103) was followed by hematoxylin counterstaining (Servicebio, G1140) and ethanol dehydration. Total RNA was isolated from uterine tissues using TRIzol (Invitrogen) following manufacturer’s protocols. cDNA synthesis (500 ng RNA input) employed PrimeScript RT Master Mix (Takara, RR036A). qPCR was performed on a CFX Connect system (Bio-Rad) using SYBR Green (Takara, RR820A). Relative quantification used the 2-ΔΔCt method with Actb normalization. Primer sequences are provided in Table  1 . Table 1 List of primer sequences for RT-qPCR Primer name Sequences(5′-3′) Wnt5a Forward CAACTGGCAGGACTTTCTCAA Wnt5a Reversed CATCTCCGATGCCGGAACT Wnt11 Forward ATGCGTCTACACAACAGTGAAG Wnt11 Reversed GTAGCGGGTCTTGAGGTCAG Fzd6 Forward ATGGAAAGGTCCCCGTTICTG Fzd6 Reversed GGGAAGAACGTCATGTTGTAAGT Ctnnb1 Forward ATGGAGCCGGACAGAAAAGC Ctnnb1 Reversed CTTGCCACTCAGGGAAGGA Lef1 Forward GCCACCGATGAGATGATCCC Lef1 Reversed TTGATGTCGGCTAAGTCGCC Hoxa11 Forward TTTGATGAGCGTGGTCCCTG Hoxa11 Reversed AGGAGTAGGAGTATGTCATTGGG Vangl2 Forward ACTCGGGCTATTCCTACAAGT Vangl2 Reversed TGATTTATCTCCACGACTCCCAT Cdh1 Forward CAGGTCTCCTCATGGCTTTGC Cdh1 Reversed CTTCCGAAAAGAAGGCTGTCC Wfdc3 Forward GAGAGCACGCATTGAGAGGTG Wfdc3 Reversed ACAGGATTCGTCTCCGGTACA Wnt4 Forward AGACGTGCGAGAAACTCAAAG Wnt4 Reversed GGAACTGGTATTGGCACTCCT Bmp2 Forward TGCTTCTTAGACGGACTGCG Bmp2 Reversed CTGGGGAAGCAGCAACACTA Bmp8a Forward CCTGGTCATGAGCTTCGTCA Bmp8a Reversed AGCAGGGATCTGGGTTAGGT Prl8a2 Forward TTATGGGTGCATGGATCACTCC Prl8a2 Reversed CCCACGTAAGGTCATCATGGAT DHCR24 Forward CTCCTGCCGCTCTCGCTTATC Prl8a2 Reversed GCTACCCTGCTCCTTCCATTCC HMGCR Forward GCAGGACCCCTTTGCTTAGA HMGCR Reversed GCACCTCCACCAAGACCTAT SQLE Forward GTTCGCCCTCTTCTCGGATATT SQLE Reversed GGTTCCTTTTCTGCGCCTCCT HMGCS1 Forward TGGCAGGGAGTCTTGGTACT HMGCS1 Reversed TCCCACTCCAAATGATGACA SREBF2 Forward AACGGTCATTCACCCAGGTC SREBF2 Reversed GGCTGAAGAATAGGAGTTGCC Lif Forward ATTGTGCCCTTACTGCTGCTG Lif Reversed GCCAGTTGATTCTTGATCTGGT Hand2 Forward GCAGGACTCAGAGCATCAACA Hand2 Reversed AGGTAGGCGATGTATCTGGTG Pthc1 Forward AAAGAACTGCGGCAAGTTTTTG Pthc1 Reversed CTTCTCCTATCTTCTGACGGGT Muc1 Forward GGCATTCGGGCTCCTTTCTT Muc1 Reversed TGGAGTGGTAGTCGATGCTAAG MKI67 Forward ACGCCTGGTTACTATCAAAAGG MKI67 Reversed CAGACCCATTTACTTGTGTTGGA PCNA Forward CCTGCTGGGATATTAGCTCCA PCNA Reversed CAGCGGTAGGTGTCGAAGC CCND1 Forward GCTGCGAAGTGGAAACCATC CCND1 Reversed CCTCCTTCTGCACACATTTGAA CCNE1 Forward AAGGAGCGGGACACCATGA CCNE1 Reversed ACGGTCACGTTTGCCTTCC CCNA2 Forward CGCTGGCGGTACTGAAGTC CCNA2 Reversed GAGGAACGGTGACATGCTCAT GREB1 Forward ATGGGAAATTCTTACGCTGGAC GREB1 Reversed CACTCGGCTACCACCTTCT TFF3 Forward CCAAGCAAACAATCCAGAGCA TFF3 Reversed GCTCAGGACTCGCTTCATGG List of primer sequences for RT-qPCR To investigate the transcriptional profile of uterine tissues, OVX Rptor fl/fl PR Cre/ + ( n  = 3) and Rptor fl/fl control ( n  = 3) mice received a single subcutaneous injection of 100 ng E2 in 100 μl sesame oil and were euthanized 24 h post-treatment. Uterine tissues were immediately snap-frozen in liquid nitrogen. Total RNA was isolated using TRIzol reagent (Invitrogen, 15596026) following the manufacturer’s protocol. RNA quality was assessed by a NanoDrop 2000 spectrophotometer (Thermo Scientific) for purity (A260/280 ratio > 1.8) and concentration, and RNA integrity was verified using an Agilent 2100 Bioanalyzer (RIN > 8.0 for all samples). Stranded mRNA libraries were prepared using the TruSeq Stranded mRNA LT Sample Prep Kit (Illumina, RS-122–2101) and sequenced on a Novaseq6000 platform. Clean reads were aligned to the mouse reference genome (GRCm38.p6) using HISAT2. Gene expression levels were quantified as fragments per kilobase of transcript per million mapped reads (FPKM) using Cufflinks and read counts for each gene were obtained via HTSeq-count (v0.13.5). Differential gene expression analysis was performed using DESeq2 (v1.30.1) with thresholds of q -value  2. The transcriptome sequencing and analysis were conducted by OE Biotech Co., Ltd. (Shanghai, China). Primary uterine stromal cells were isolated from 8-week-old female mice as previously described [ 17 ]. Briefly, uterine horns were excised and longitudinally dissected in ice-cold Hank’s Balanced Salt Solution (HBSS; Servicebio, G4204-500ML) to remove blood. Tissue fragments were subjected to sequential enzymatic digestion; initially treated with 1% (w/v) trypsin (Sigma-Aldrich, P7545) and 6 mg/mL dispase II (Sigma, D4693) in HBSS at 4 °C for 1 h, followed by an additional hour at RT and 10 min at 37 °C with orbital shaking at 100 rpm. Subsequent to three washes with HBSS, the epithelial-rich cell suspensions were filtered through a 100 μm nylon mesh. Residual tissues underwent a secondary digestion with 0.15 mg/mL collagenase I (Thermo Fisher, 17100017) at 37 °C for 30 min, with intermittent vortexing. Following three additional washes with HBSS, stromal cell fractions were collected through 100 μm filtration. Cell purity (> 90%) was confirmed by flow cytometry analysis of stromal marker vimentin. Cells were cultured in Dulbecco’s modified eagle medium (DMEM)/F12 (Wisent, 319–080-CL) supplemented with 10% charcoal–dextran-treated FBS (Biological Industries, 04–201-1A) at 37 °C in 5% CO 2 . For proliferation assays, cells (1 × 10 5 /well) were treated with 20 μM atorvastatin (MCE, HY-B0589) or 10 μg/mL cholesterol (Selleck, S4154) for 96 h, and proliferation was monitored by Ki67 or CFSE dilution using a Cytek flow cytometer. All quantitative data are presented as the mean ± SD from at least three independent biological replicates. Statistical analyses were performed using GraphPad Prism 9.0. For comparisons between two groups, a two-tailed Student’s t -test was used. For comparisons involving three or more independent groups defined by a single factor, data were analyzed by one-way analysis of variance (ANOVA) followed by Tukey’s multiple comparisons test. For comparisons involving multiple factors, data were analyzed by two-way analysis of variance (ANOVA) followed by Tukey’s multiple comparisons test. A p -value of < 0.05 was considered statistically significant. Detailed statistical parameters, including the exact sample size ( n ), specific statistical tests, and exact p -values, are provided in the respective figure legends.

The endometrium, a dynamically remodeled tissue, provides the essential microenvironment for successful embryo implantation, with its functional quality being principally determined by optimal thickness and receptivity status during the implantation window [ 1 ]. The cyclical regeneration and function of the human endometrium are dynamically and precisely regulated by sex steroid hormones, including estrogen (E2) and progesterone (P4) [ 2 , 3 ]. Clinical evidence indicates that while one-third of implantation failures stem from embryonic abnormalities, the majority result from endometrial dysfunction, particularly impaired receptivity and disrupted embryo-endometrial crosstalk [ 4 ]. Among various endometrial pathologies, a thin endometrium (< 7 mm in mid-cycle) represents one of the most challenging causes of infertility, with its treatment presenting numerous obstacles [ 5 , 6 ]. Impaired angiogenesis and adenogenesis (gland formation) are key features of thin endometrium, accompanied by compromised steroid hormone responsiveness and regenerative defects. Such inadequate endometrial thickness compromises its receptivity, thereby reducing embryonic implantation rates and ultimately leading to significant declines in clinical pregnancy rates and live birth rates [ 4 , 7 ]. Currently, the precise pathogenesis of thin endometrium remains unclear, potentially involving multifactorial contributions, including systemic factors (e.g., age and endocrine status), local factors (e.g., history of uterine cavity surgery or uterine diseases), external drug effects, and undefined latent factors [ 8 ]. However, there is still a lack of a systematic theoretical framework to comprehensively elucidate the onset and progression of this disease, as well as effective treatment measures. This knowledge gap highlights the urgent need to clarify the molecular determinants of endometrial homeostasis, which may reveal new therapeutic targets for the etiology of this intractable infertility. Cellular proliferation and differentiation are highly dependent on the activation status of intracellular metabolic signaling pathways, among which the mammalian target of rapamycin (mTOR) signaling pathway serves as a central metabolic integrator. mTOR signaling molecules orchestrate cellular proliferation and differentiation through precise regulation of anabolic processes, including gene transcription, protein translation, and ribosome biogenesis in response to growth factors (e.g., IGF-1, EGF), hormones (e.g., estrogen), and nutrient availability [ 9 , 10 ]. Emerging evidence highlights its crucial involvement in reproductive physiology. Mouse experiments have shown that the activity of mTOR increases during the early stage of pregnancy (peaking at gestational day 5), temporally coinciding with stromal cell proliferation and coinciding with the window of implantation [ 11 ]. Pharmacological inhibition using intrauterine rapamycin administration on gestation day 4 significantly impairs embryo implantation, while L-arginine supplementation enhances implantation rates through PI3K/AKT/mTOR-mediated nitric oxide synthesis [ 12 ]. A recent report showed that the activation level of the mTOR signaling pathway in thin endometrium is significantly downregulated compared with normal endometrial tissue [ 13 ]. Despite these observations, the underlying mechanistic connections among mTOR dysregulation, metabolic reprogramming, and the failure of endometrial regeneration continue to be inadequately comprehended, posing a significant knowledge deficit in the field of reproductive medicine. Our research systematically clarified the relationship between mTORC1 signaling reduction and the pathogenesis of thin endometrium. Phenotypic characterization revealed that uterine-specific Rptor ablation recapitulated key features of human thin endometrium syndrome, including reduction in endometrial thickness, impaired vascular density, and defective glandular development. Mechanistically, Raptor ablation disrupts estrogen-induced proliferation by suppressing cholesterol synthesis. This metabolic defect resulted in functional consequences, including impaired endometrial receptivity and defective decidual response. Our findings establish mTORC1 as a master regulator of endometrial homeostasis that couples hormonal signals to metabolic reprogramming for tissue regeneration.

The endometrium plays a pivotal role in establishing the optimal microenvironment for successful embryo implantation, with its functional competence primarily determined by achieving appropriate thickness and receptivity during the implantation window [ 1 ]. Our study provides compelling evidence that mTORC1 signaling serves as a central regulator of endometrial homeostasis and reproductive function by integrating hormonal signaling with metabolic control and cellular proliferation pathways. Through the generation and analysis of a conditional knockout mouse model with Raptor ablation specifically in progesterone receptor-positive cells, we demonstrated that mTORC1 inactivation induces a pronounced endometrial thinning phenotype, manifesting as structural impairments in both angiogenesis and adenogenesis; significantly diminished endometrial receptivity, and profound decidualization defects. At the molecular mechanistic level, our findings demonstrate that uterine Raptor deficiency disrupts fundamental metabolic processes, particularly intracellular cholesterol biosynthesis and metabolic homeostasis. These perturbations directly impair the estrogen-induced proliferative response in endometrial epithelial cells. Importantly, our data establishes that the mTORC1 signaling pathway coordinates hormone-dependent endometrial proliferation through metabolic reprogramming, thereby revealing its dual regulatory role in both establishing endometrial receptivity and facilitating proper stromal decidualization during early pregnancy. The establishment of optimal endometrial receptivity constitutes a fundamental prerequisite for successful embryo implantation, with pregnancy outcomes being critically dependent on achieving both adequate endometrial thickness and proper developmental competence. Clinical evidence indicates that endometrial factors contribute to approximately two-thirds of implantation failure cases [ 7 ]. Among these, thin endometrium represents one of the most clinically challenging etiologies of infertility, with current therapeutic strategies showing limited efficacy [ 26 ]. The precise molecular mechanisms governing the pathogenesis and progression of this condition remain incompletely understood. Our earlier report revealed a notable downregulation in the expression of the epigenetic regulator METTL3 within the uteri of infertile women suffering from endometriosis or recurrent implantation failure [ 17 ]. Notably, METTL3-dependent m6A methylation is crucial for maintaining female fertility by harmonizing estrogen and progesterone signaling pathways. Emerging single-cell transcriptomic analyses of endometrial tissues have revealed a potential downregulation of mTOR signaling activity in thin endometrium compared with normal controls [ 13 ]. Cellular proliferation and differentiation processes are tightly regulated by intracellular metabolic signaling networks, with the mTOR pathway serving as a central regulator of cellular metabolism [ 27 ]. We previously found that mTOR signaling plays an important role in estrogen-dependent vaginal epithelial cell proliferation and differentiation [ 20 ]. In the present study, we present direct evidence indicating a marked decrease in mTORC1 activation within thin endometrial specimens. Of particular significance, genetic ablation of Rptor (an essential mTORC1 component) in murine endometrial cells recapitulated the hallmark phenotypic features of thin endometrium, including reduced uterine and endometrial dimensions, impaired vascular network formation, and defective glandular development. These collective findings strongly suggest that dysregulation of mTORC1 signaling activity may represent a fundamental pathophysiological mechanism underlying thin endometrium development. Moreover, our genetically engineered mice provide a physiologically relevant preclinical model for developing therapies targeting thin endometrium. The endometrium undergoes precisely coordinated cyclic remodeling through an integrated network of autocrine, paracrine, and endocrine signaling pathways, with estrogen and progesterone serving as the principal regulators of endometrial cellular dynamics through their opposing actions on proliferation and differentiation. Importantly, we observed synchronous cyclical variations in endometrial mTORC1 activity corresponding to estrous cycle phases, with exogenous E2 administration in OVX mice significantly enhancing mTORC1 activation. The Raptor knockout model demonstrated substantially blunted hormonal responsiveness, accompanied by significant impairment of hormone-induced endometrial cell proliferation, thereby establishing mTORC1 as both a crucial downstream mediator of estrogen signaling and a fundamental regulator of endometrial proliferative potential. These findings suggest that attenuated mTORC1 signaling may induce endometrial hormone resistance, leading to suppressed cellular proliferation and contributing to the development of pathological endometrial thinning. Clinical correlations indicate that mTORC1 hypoactivity contributes to hormonal resistance in thin endometrium, suboptimal therapeutic responses to hormonal regimens, and impaired decidual transformation, highlighting its potential as a therapeutic target for endometrial disorders. mTOR signaling pathway has been established as a critical regulator of endometrial receptivity and embryo implantation [ 11 ], with our current findings providing genetic validation of its pivotal role in uterine physiology. Previous investigations demonstrated that cortisone acetate administration induces endometrial receptivity defects through coordinated suppression of ERK1/2 and mTOR phosphorylation [ 28 ]. Similarly, pharmacological mTOR inhibition with PP242 reduced leukemia inhibitory factor (LIF) expression—a critical implantation mediator. Chen et al. further elucidated the temporal dynamics of mTOR regulation during early pregnancy, documenting progressive mTOR activation from GD3 to peak levels at GD5 (the implantation window) compared with nonpregnant controls [ 11 ]. The functional significance of this temporal regulation was confirmed by rapamycin-mediated mTOR inhibition on GD4, which significantly impaired embryo implantation efficiency. Our genetic approach employing uterine-specific Raptor knockout mice provides mechanistic insights by demonstrating that mTORC1 deficiency disrupts two fundamental processes: endothelial cell proliferation and decidual transformation, thereby establishing its nonredundant role in hormone-dependent uterine remodeling. Collectively, these studies reveal that mTORC1 orchestrates endometrial receptivity through a tripartite mechanism: transcriptional regulation of implantation-critical genes, control of stromal cell proliferation dynamics, and maintenance of decidualization competence. The pathway’s unique capacity to simultaneously regulate both proliferative and differentiative processes positions it as a promising therapeutic target for implantation failure disorders. As an energy metabolism regulatory molecule, the mTORC1 signaling pathway serves as a major metabolic integrator. It dynamically regulates cellular energy homeostasis by promoting anabolic processes and energy storage when nutrition is adequate and simultaneously activates catabolic pathways to save energy under starvation conditions [ 10 , 29 ]. Within lipid metabolism specifically, mTORC1 exerts multifaceted control through two principal mechanisms: activation of sterol regulatory element-binding proteins (SREBPs) and phosphorylation-mediated regulation of HMG-CoA reductase (HMGCR), the rate-limiting enzyme in the USP20-dependent cholesterol biosynthesis pathway [ 30 ]. The emerging paradigm of lipid metabolism in reproductive physiology reveals that lipids serve dual essential functions during endometrial decidualization, as both bioenergetic substrates and critical structural components. This role is exemplified by carnitine palmitoyltransferase-1 (CPT1), the gatekeeper enzyme of fatty acid β-oxidation, whose genetic knockdown or pharmacological inhibition significantly impairs human endometrial decidualization [ 31 ]. Equally crucial is the maintenance of lipid peroxidation homeostasis by glutathione peroxidase 4 (GPX4), as evidenced by the complete infertility observed in mice with uterine epithelium-specific GPX4 deletion [ 32 ]. Our study uncovers a previously unrecognized metabolic constraint in endometrial cells: Raptor ablation disrupts de novo cholesterol biosynthesis, and intriguingly, this defect cannot be rescued by exogenous cholesterol supplementation. This cell-intrinsic dependence on mTORC1-regulated cholesterol production likely reflects the endometrium’s unique requirement for localized steroidogenesis during the critical windows of implantation and decidualization. The observed metabolic inflexibility provides a mechanistic explanation for the limited clinical efficacy of systemic cholesterol supplementation in thin endometrium treatment, while simultaneously highlighting the therapeutic potential of precisely modulating endometrial mTORC1 activity to restore proper metabolic function. In conclusion, our study establishes mTORC1 as a central metabolic orchestrator that integrates hormonal signaling, proliferative capacity, and lipid metabolic pathways to maintain endometrial functional competence. The failure of exogenous cholesterol to restore proliferative activity in Rptor -deficient endometrial cells underscores the complexity of endometrial metabolic regulation and challenges existing paradigms for treating thin endometrium. Collectively, this work not only advances our understanding of endometrial physiology but also opens new avenues for treating endometrial insufficiency through metabolism-focused therapeutic strategies. While our research offers significant insights into how mTORC1 signaling influences endometrial function, it is important to recognize certain limitations. Firstly, our study utilized the Pgr-Cre driver, which efficiently deletes Rptor in the uterus but also targets the vaginal epithelium. As we previously documented, Rptor fl/fl Pgr cre/ + mice exhibit vaginal opening defects that prevent successful mating. While our study utilized the Pollard hormonal replacement model to mimic the window of implantation due to this infertility, we acknowledge the limitations of this approach. Our data revealed a selective impairment of the Lif and IHH–COUP–TFII signaling axes, while other markers such as Muc1 and Ptch1 remained unaffected, pointing to a specific regulatory role of mTORC1 in stromal–epithelial communication. However, we recognize that the Pollard protocol artificially simulates the hormonal milieu and may not perfectly recapitulate the complex, dynamic physiological fluctuations of the natural gestation day 4–5 window, particularly the precise temporal synchronization of estrogen and progesterone peaks. Therefore, future studies utilizing inducible deletion models (e.g., Ltf-iCre ) that preserve natural fertility would be valuable to further validate these mechanistic findings in a natural pregnancy context. Secondly, another limitation of our study is that the transcriptomic validation of human thin endometrium relies on a single public dataset ( GSE160633 ) due to the scarcity of high-throughput sequencing data in this specific pathology. While our RT-qPCR and IHC analyses on collected clinical samples support the key findings, future studies with larger, independent transcriptomic cohorts are warranted to further bolster the generalizability of these pathway-level alterations. Finally, while we observed that mTORC1 activity is notably lower in thin endometrium compared with controls, and the upstream mechanisms driving this suppression remain to be thoroughly examined in future investigations.

Our study establishes that mTORC1 signaling orchestrates estrogen-induced endometrial proliferation through cholesterol biosynthesis and metabolic homeostasis. Crucially, mTORC1 coordinates hormonal signals via metabolic reprogramming, demonstrating dual regulatory roles in establishing endometrial receptivity and facilitating stromal decidualization. By linking mTORC1 dysfunction to impaired cholesterol synthesis, estrogen unresponsiveness, and regenerative failure, we resolve a key mechanistic gap in thin endometrium pathogenesis. These findings provide the first evidence for uterine-specific metabolic reprogramming in endometrial incompetence, offering actionable targets for steroid-resistant infertility treatment.

Supplementary Material 1. Supplementary Material 1.