HFTMSCs alleviate uterine fibrosis and restore fertility in intrauterine adhesion models.

Primary human fallopian tube mesenchymal stem cells were isolated by enzymatic digestion and expanded in vitro. as summarized in Fig. S2A. Phase-contrast microscopy revealed dynamic morphological changes: initially (days 1–4), mixed epithelial-like and spindle-shaped cells emerged, followed by a homogeneous population of fibroblast-like cells by passage 3 (P3). While proliferation remained stable until P15, noticeable morphological alterations were observed at P20 (Fig.  1 A). Growth curves of P5-P20 HFTMSCs exhibited a typical sigmoidal pattern, with a lag phase (days 1–3), exponential growth (days 3–5), and a plateau (days 5–7) during which contact inhibition and nutrient depletion slowed proliferation. Cell viability remained high (92%−96%). Population doubling time (PDT) was shortest at P5 (25.4 h), increased gradually in later passages (P10: 32.2 h; P15: 34.8 h; P20: 36.69 h), with an average PDT of 32.1 h across P5-P20, as summarized in (Fig. S2E and Supplementary Table 3–4), indicating sustained proliferative capacity within this range. To verify the stem cell characteristics and differentiation capabilities of HFTMSCs, cells underwent specific lineage induction protocols. Under osteogenic conditions, Alizarin Red staining after 21 days revealed prominent calcified nodules with characteristic orange-red deposits confirming their osteogenic differentiation capability. Similarly, adipogenic induction for 21 days resulted in intracellular lipid droplet accumulation, as evidenced by Oil Red O-positive staining. Non-induced control HFTMSCs cultured in standard maintenance medium exhibited no mineralization or lipid vacuole formation (Fig.  1 B). These findings illustrate the differentiation potential of HFTMSCs along mesenchymal lineages, meeting the minimal criteria set by the International Society for Cellular Therapy (ISCT) for defining multipotent mesenchymal stromal cells [ 29 ]. Flow cytometric analysis of P3 HFTMSCs demonstrated robust expression of canonical mesenchymal surface markers, with > 98% of cells positive for CD 90, CD 105, and CD 44(95%) (Fig.  1 C). Critically, these cells exhibited negligible expression of hematopoietic (CD 34, CD 45) and immunogenic (HLA-DR) markers (< 3%), confirming their immunoprivileged phenotype (isotype controls in Fig. S2D). Complementary immunofluorescence imaging further validated these findings, revealing strong membrane-associated staining for CD 90, CD 105, and CD 44. Notably, expression of the pluripotency-associated transcription factor OCT 4 was observed of HFTMSCs, suggesting retention of stemness properties during expansion (Fig.  1 D). Quantitative fluorescence intensity analysis corroborated uniform marker expression across the population (Fig.  1 E). These results indicate that HFTMSCs are a population of multipotent stromal cells with a stable phenotype and meet the criteria established by the International Society for Cell Therapy (ISCT). qRT-PCR analysis confirmed HFTMSCs exhibit molecular hallmarks of bona fide mesenchymal stem cells, with robust expression of CD 44 , CD 90 , and CD 105 , while demonstrating negligible levels of hematopoietic ( CD 34 ) and immunogenic ( HLA-DR ) (Fig. S2B). Strikingly, these cell maintained expression of pluripotency regulators OCT 4 and NANOG , suggesting enhanced stemness preservation compared to conventional MSCs sources. Next cyctogenetic analysis revealed normal diploid karyotypes (46, XX) of P3 HFTMSCs (Fig. S2C), with no detectable chromosomal aberrations. Flow cytometric validation of surface markers demonstrated dual positivity for CD 90/CD 105 and for CD 44 , while HLA-DR remained undetectable (< 3%, Fig. S2D). The findings indicate that HFTMSCs possess high purity, low immunogenicity, self-renewal capabilities, and safe in vitro expansion. These attributes render HFTMSCs valuable cells for applications in regenerative medicine. To assess the safety profile of HFTMSCs, a subcutaneous tumorigenicity assay was conducted in BALB/c-nu, with human induced pluripotent stem cells (HiPSCs) used as a positive control. Strikingly, HFTMSCs-injected mice exhibited tumor-free survival throughout the 8-week observation period (Fig. S2F), while all HiPSCs recipients developed palpable teratomas by Week 6 (Fig. S2G-H). The aforementioned results indicate that HFTMSCs possess a relatively high level of biosafety, which is quite reassuring. Histopathological examination confirmed the absence of proliferative lesions in HFTMSCs-treated animals, whereas HiPSCs-derived tumors displayed classical teratoma features with trilineage differentiation. Endodermal lineage, pseudostratified ciliated columnar epithelium resembling gastrointestinal/respiratory tissues (Fig. S2I). Mesodermal lineage, striated muscle bundles (Fig. S2J). Ectodermal lineage, keratinized epithelial structures with hair follicle differentiation (Fig. S2K). Studies have confirmed the pluripotency and tumorigenic risk of HiPSCs [ 30 ]. The stark contrast with HiPSCs underscores HFTMSCs’ superior safety profile for regenerative applications. To delineate the homing capacity and regenerative dynamics of HFTMSCs, P3 cells were fluorescently tagged with CellTracker™ CM-DiI. The lipophilic orange-red fluorescent dye CellTracker™ CM-DiI, which localizes to cellular membranes and cytoplasmic compartments, effectively labeled HFTMSCs with > 90% efficiency, as confirmed by fluorescence microscopy (Fig.  1 F). Subsequent CCK-8 viability assays across a concentration gradient (0–5 μM) identified 3 μM as the optimal labeling dose. The highest proliferative activity was found at 3uM concentration (Fig.  1 I). Longitudinal immunofluorescence tracking revealed a time-dependent distribution pattern of CM-DiI-labeled HFTMSCs within uterine tissues. At a 14-day therapeutic regimen, CM-DiI⁺ cells were primarily localized in the endometrial region, stromal areas, and myometrium (Fig.  1 G). After a 28-day therapeutic regimen, only sparse fluorescent signals were detected in the endometrial zone, implicating tissue remodeling in endometrial regeneration (Fig.  1 H). These results indicate that CellTracker™ CM-DiI successfully labeled HFTMSCs, and the detected fluorescent signals demonstrated that the transplanted cells could migrate to and be retained within the injured uterine tissues. Whole-body fluorescence imaging was performed following the second (intravenous) administration of CM‑DiI‑labeled HFTMSCs (Fig.  1 J, K). At 24 h post‑injection, prominent fluorescent signals were detected in the liver and pulmonary regions, consistent with the initial filtration and sequestration of intravenously delivered cells. By 7 days post‑injection, fluorescence intensity in these off‑target organs was markedly reduced. Concurrently, distinct fluorescent signals were observed in the uterine region. As noted in the Methods, due to the dual‑route administration of labeled cells, these uterine signals reflect the total engrafted cell population from both intrauterine and intravenous deliveries. Nevertheless, the temporal pattern of an initial systemic distribution shortly after intravenous injection, followed by a sustained uterine‑specific signal, remains compatible with the concept that intravenously delivered HFTMSCs can traffic to sites of uterine injury [ 31 ]. Histopathological evaluation at 14 days post-procedure confirmed successful establishment of the mechanically induced IUA model. Sham-operated controls exhibited intact endometrial architecture with abundant glands (24 ± 6 glands) and minimal collagen deposition (fibrotic area: 10 ± 1.0%) (Fig.  2 A, D). In stark contrast, IUA model mice displayed severe endometrial atrophy (endometrial thickness: 123.7 ± 10.2 μm vs Sham: 333.4 ± 34.2 μm; P  < 0.001), marked glandular depletion (2 ± 1 vs sham 24 ± 6 glands; P  < 0.001), and extensive fibrosis (collagen-positive area: 55.5 ± 5.2% vs sham10 ± 1.0%; P  < 0.001) (Fig.  2 B, C, E). These histopathological findings collectively confirm the successful establishment of the IUA mouse model. Fig. 2 HFTMSCs Therapy Reverses Endometrial Fibrosis and Restores Structural Integrity in an IUA Model. A Representative H&E staining of uterine Sects. 14-day post-intervention, demonstrating severe endometrial thinning and glandular loss in IUA mice compared to Sham controls. B-C Quantitative analyses reveal IUA-induced reductions in endometrial thickness ( P  < 0.001) and gland number ( P  < 0.001) versus Sham. D-E Masson’s trichrome staining and fibrosis quantification confirm pathological collagen accumulation in IUA mice ( P  < 0.001). F-J At 28 days post-treatment, combination therapy with HFTMSCs and E₂ significantly restored endometrial thickness from 194.4 ± 44.3 μm (IUA) to 354.6 ± 24.2 μm ( P  < 0.01 vs. IUA), reaching 89.4% of Sham level (342.1 ± 55.2 μm), and reduced fibrotic area from 20.5 ± 1.1% (IUA) to 10.3 ± 0.8% ( P  < 0.001 vs. IUA), outperforming both HFTMSCs and E₂ monotherapies as well as natural recovery. (Data expressed as mean ± SD; * P  < 0.05, ** P  < 0.01, *** P  < 0.001 vs Sham; # P  < 0.05, ## P  < 0.01, ### P  < 0.001 vs IUA) HFTMSCs Therapy Reverses Endometrial Fibrosis and Restores Structural Integrity in an IUA Model. A Representative H&E staining of uterine Sects. 14-day post-intervention, demonstrating severe endometrial thinning and glandular loss in IUA mice compared to Sham controls. B-C Quantitative analyses reveal IUA-induced reductions in endometrial thickness ( P  < 0.001) and gland number ( P  < 0.001) versus Sham. D-E Masson’s trichrome staining and fibrosis quantification confirm pathological collagen accumulation in IUA mice ( P  < 0.001). F-J At 28 days post-treatment, combination therapy with HFTMSCs and E₂ significantly restored endometrial thickness from 194.4 ± 44.3 μm (IUA) to 354.6 ± 24.2 μm ( P  < 0.01 vs. IUA), reaching 89.4% of Sham level (342.1 ± 55.2 μm), and reduced fibrotic area from 20.5 ± 1.1% (IUA) to 10.3 ± 0.8% ( P  < 0.001 vs. IUA), outperforming both HFTMSCs and E₂ monotherapies as well as natural recovery. (Data expressed as mean ± SD; * P  < 0.05, ** P  < 0.01, *** P  < 0.001 vs Sham; # P  < 0.05, ## P  < 0.01, ### P  < 0.001 vs IUA) To evaluate the impact of HFTMSCs on IUA, body weight was monitored for 28 days. A transient, statistically significant weight loss was observed in all surgical groups on the day of model establishment ( P  < 0.05 vs pre-surgery). All groups subsequently regained weight with no significant intergroup differences throughout the follow-up period (Fig. S3D). Estrous cycle monitoring from day 14 to day 28 post-treatment revealed that IUA mice exhibited prolonged diestrus phase and reduced cycle transition frequency compared to Sham controls. While monotherapies (HFTMSCs or E₂ alone) partially restored cycle regularity, the combination therapy restored estrous cycle dynamics to near-normal levels (Fig. S3B-C). Previous studies have suggested that stem cell therapy may improve reproductive function by modulating the hypothalamic-pituitary-ovarian axis [ 32 ]. Thus, we hypothesize that the superior recovery observed in the combination group may involve endocrine modulation, a hypothesis warranting further investigation. Histopathological assessment at 14 days post-intervention revealed distinct therapeutic trajectories. IUA controls showed persistent endometrial atrophy (gland count: 2 ± 1 vs Sham: 24 ± 6; P  < 0.001) and fibrosis (55.3 ± 4.6% vs 9.5 ± 0.5%; P   0.05 vs IUA) and fibrosis reduction (39.8 ± 2.2%, P  > 0.05 vs IUA). In stark contrast, combination therapy with HFTMSCs and E₂ achieved robust structural restoration: endometrial thickness increased from 123.7 ± 10.2 μm (IUA) to 247.4 ± 18.6 μm ( P  < 0.001), a 2.0-fold increase; and fibrotic area was reduced from 55.3 ± 4.6% (IUA) to 32.7 ± 3.1% ( P  < 0.001), corresponding to a 40.9% relative reduction and a 22.6 percentage-point absolute decrease (Fig.  2 B, C, E). Longitudinal evaluation at 28 days post-transplantation revealed progressive but incomplete natural recovery in untreated IUA mice: endometrial thickness reached 194.4 ± 44.3 μm (vs Sham: 342.1 ± 55.2 μm, P  < 0.01), gland count increased to 23 ± 2 (vs Sham: 31 ± 3, P  < 0.05), and fibrotic area decreased to 20.5 ± 1.1% (vs Sham: 8.2 ± 0.7%, P  < 0.01) (Fig.  2 F-J). Strikingly, combination therapy with HFTMSCs and E₂ demonstrated sustained regenerative superiority, achieving near-complete restoration: endometrial thickness (354.6 ± 24.2 μm, P  < 0.01 vs Sham), glandular number (32 ± 4 glands, P  < 0.05 vs IUA), and fibrosis resolution (11.3 ± 0.9%, P  < 0.001 vs IUA) (Fig.  2 G, I, J). Mechanistically, this synergistic effect may be attributed to phased regenerative processes: (i) Estrogen priming phase (days 0–14): E₂ enhances endometrial receptivity via ERα upregulation, facilitating HFTMSCs engraftment [ 33 ]. (ii) Stromal reprogramming phase (days 14–28): HFTMSCs-derived trophic factors promote angiogenesis and matrix remodeling [ 34 ]. Collectively, these findings demonstrate that combined HFTMSCs and E₂ therapy restores both the morphology and function of the injured endometrium while effectively suppressing fibrotic progression. To investigate the regenerative capacity of HFTMSCs in restoring endometrial functionality, we performed immunofluorescence profiling of epithelial proliferation (CK7) and angiogenesis (CD31) across therapeutic phases. At day 14 post-treatment, IUA mice exhibited severe epithelial atrophy (CK7, vs sham, P  < 0.001) and impaired vascularization (CD31, vs sham, P  < 0.001). While monotherapies partially rescued CK7 expression (E2 and HFTMSCs, P  < 0.01 vs IUA), combinatorial therapy achieved more complete epithelial recovery (CK7, vs IUA, P  < 0.001), significantly enhanced angiogenesis (CD31, vs IUA , P  < 0.001; Fig.  3 A-D). On post-treatment day 28, the expression level of CK7 in the IUA group was significantly lower than that in the Sham group ( P  < 0.001). Compared with the IUA group, treatment with E2 alone or HFTMSCs alone could significantly increase the expression of CK7, with effects approaching those of the Sham group, but no statistical difference was observed between the two separate treatment groups. In the HFTMSCs and E2 combined treatment group, the expression level of CK7 showed no significant difference compared to the Sham group (Fig.  3 E; F). Additionally, the expression of CD31 in the IUA group was significantly lower than that in the Sham group ( P  < 0.05), while no statistical difference in CD31 expression was observed among the IUA group treated separately with E2, HFTMSCs, or IUA alone. However, the expression of CD31 in the HFTMSCs and E2 combined treatment group was significantly higher than that in the IUA group ( P  < 0.05) (Fig.  3 G; H), with therapeutic effects approaching those of the Sham group. These findings demonstrate that combination therapy with HFTMSCs and E2 effectively restores the injured endometrium by promoting epithelial proliferation and angiogenesis. Fig. 3 Temporal evaluation of uterine regeneration in IUA mouse models across therapeutic interventions. A Immunofluorescence imaging of CK7 epithelial regeneration at 14-day post-treatment. B Quantitative analysis of CK7 fluorescence intensity. C CD31 vascular endothelial distribution patterns following 14-day interventions. D Angiogenesis quantification by CD31 signal density. E Longitudinal tracking of CK7 epithelial recovery at 28-day. F Quantitative analysis of CK7 fluorescence intensity at 28 days. G CD31 neovascularization patterns at 28-day endpoint. H Late-phase angiogenesis quantification. (Data expressed as mean ± SD; * P  < 0.05, ** P  < 0.01, *** P  < 0.001 vs Sham; # P  < 0.05, ## P  < 0.01, ### P  < 0.001 vs IUA) Temporal evaluation of uterine regeneration in IUA mouse models across therapeutic interventions. A Immunofluorescence imaging of CK7 epithelial regeneration at 14-day post-treatment. B Quantitative analysis of CK7 fluorescence intensity. C CD31 vascular endothelial distribution patterns following 14-day interventions. D Angiogenesis quantification by CD31 signal density. E Longitudinal tracking of CK7 epithelial recovery at 28-day. F Quantitative analysis of CK7 fluorescence intensity at 28 days. G CD31 neovascularization patterns at 28-day endpoint. H Late-phase angiogenesis quantification. (Data expressed as mean ± SD; * P  < 0.05, ** P  < 0.01, *** P  < 0.001 vs Sham; # P  < 0.05, ## P  < 0.01, ### P  < 0.001 vs IUA) We systematically evaluated endometrial receptivity and fibrosis markers at different therapeutic timepoints using immunohistochemical co-localization analysis. At day 14 post-treatment, immunofluorescence quantification revealed significantly diminished expression of endometrial receptivity markers LIF and αvβ3 in the IUA group compared with sham-operated controls(LIF and αvβ3, vs Sham, P  < 0.001). Both monotherapies (E2 alone and HFTMSCs alone) demonstrated marked elevation of these biomarkers relative to IUA ( P  < 0.01) (Fig.  4 A, B). Concurrently, fibrosis markers α-SMA and TGF-β1 exhibited pronounced upregulation in IUA ( P  < 0.001). While single-agent therapies showed moderate reductions ( P  < 0.01, vs Sham) the combination therapy achieved superior suppression of fibrotic markers compared to monotherapies ( P  < 0.001) (Fig.  4 C,D). Day 28, untreated IUA mice demonstrated limited natural recovery (LIF and αvβ3, P  < 0.01, vs Sham). Combinatorial therapy sustained receptivity marker restoration (LIF and αvβ3, P  < 0.05, vs IUA). The HFTMSCs treatment group and the HFTMSCs and E2 combined treatment group had no significant statistical significance between the three groups. Fibrotic resolution progressed temporally, with combinatorial treatment normalizing α-SMA and TGF-β1 ( P  = 0.05, vs IUA), while monotherapies retained elevated fibrosis (α-SMA and TGF-β1, P  < 0.05, vs Sham; Fig.  4 E-H). Western blot quantification corroborated these findings: Day 14: Combinatorial therapy elevated LIF and αvβ3 versus IUA ( P  < 0.001), surpassing monotherapies (Fig.  4 I, K). Day 28: Sustained protein recovery in combinatorial groups (LIF and αvβ3, P  < 0.05 vs IUA) contrasted with monotherapy stagnation (Fig.  4 J, L). Notably, E2 monotherapy exhibited unexpected integrin upregulation at day 28 ( P  < 0.05 vs IUA). We postulate that incomplete receptivity marker restoration may correlate with residual inflammatory microenvironment alterations, particularly sustained elevation of pro-inflammatory cytokines (e.g., IL-6, TNF-α) in IUA models potentially suppressing endometrial receptivity [ 35 ]. Fig. 4 Temporal restoration of endometrial receptivity and fibrotic regression in IUA therapeutics. A Immunofluorescence of endometrial receptivity markers (LIF, red; αvβ3, green) at 14-day post-treatment. B Quantification of LIF/αvβ3 fluorescence intensity. C Fibrotic marker expression (α-SMA, red; TGF-β1, green) following 14-day interventions. D α-SMA/TGF-β1 signal density analysis. E Longitudinal tracking of LIF/αvβ3 expression at 28 days. F Quantitative statistical chart of LIF and αvβ3 at 28-day post-treatment. G Quantitative statistical chart of α-SMA and TGF-β1 at 28-day post-treatment. H Fibrotic marker expression (α-SMA, red; TGF-β1, green) following 28-day interventions. I Western blot analysis of LIF/αvβ3 protein levels (14-day). J Protein expression profiles at 28-day. K Quantitative analysis of LIF and αvβ3 protein expression levels in each group at 14-day post-treatment. L Quantitative analysis of LIF and αvβ3 protein expression levels in each group at 28-day post-treatment. (Data expressed as mean ± SD; * P  < 0.05, ** P  < 0.01, *** P  < 0.001 vs Sham; # P  < 0.05, ## P  < 0.01, ### P  < 0.001 vs IUA) Temporal restoration of endometrial receptivity and fibrotic regression in IUA therapeutics. A Immunofluorescence of endometrial receptivity markers (LIF, red; αvβ3, green) at 14-day post-treatment. B Quantification of LIF/αvβ3 fluorescence intensity. C Fibrotic marker expression (α-SMA, red; TGF-β1, green) following 14-day interventions. D α-SMA/TGF-β1 signal density analysis. E Longitudinal tracking of LIF/αvβ3 expression at 28 days. F Quantitative statistical chart of LIF and αvβ3 at 28-day post-treatment. G Quantitative statistical chart of α-SMA and TGF-β1 at 28-day post-treatment. H Fibrotic marker expression (α-SMA, red; TGF-β1, green) following 28-day interventions. I Western blot analysis of LIF/αvβ3 protein levels (14-day). J Protein expression profiles at 28-day. K Quantitative analysis of LIF and αvβ3 protein expression levels in each group at 14-day post-treatment. L Quantitative analysis of LIF and αvβ3 protein expression levels in each group at 28-day post-treatment. (Data expressed as mean ± SD; * P  < 0.05, ** P  < 0.01, *** P  < 0.001 vs Sham; # P  < 0.05, ## P  < 0.01, ### P  < 0.001 vs IUA) Uterine anatomical morphology was systematically evaluated at postoperative day 14. Compared with sham-operated controls, IUA model mice exhibited significant uterine distension with marked intrauterine fluid accumulation (Fig.  5 A). Remarkably, both HFTMSCs monotherapy and estrogen (E2) treatment groups demonstrated resolution of uterine cavity effusion, though residual hyperemia persisted. The combined HFTMSCs and E2 cohort achieved complete resolution of pathological distension, restoring near-normal uterine architecture (Fig.  5 A; comparative histology in Fig.  2 A, D). No intergroup morphological differences were observed at postoperative day 28. Fig. 5 Functional fertility recovery in IUA mouse models across therapeutic regimens. A Representative gross morphology of uteri at 14 days post-intervention. B Embryo implantation sites at gestational day 8.0. C Pregnancy rate of mouse. D Neonatal viability assessment through live pup counts. E Morphometric analysis of embryonic implantation efficiency. F Quantitative evaluation of reproductive outcomes. ( * P  < 0.05, ** P  < 0.01, *** P  < 0.001 vs Sham; # P  < 0.05, ## P  < 0.01, ### P  < 0.001 vs IUA) Functional fertility recovery in IUA mouse models across therapeutic regimens. A Representative gross morphology of uteri at 14 days post-intervention. B Embryo implantation sites at gestational day 8.0. C Pregnancy rate of mouse. D Neonatal viability assessment through live pup counts. E Morphometric analysis of embryonic implantation efficiency. F Quantitative evaluation of reproductive outcomes. ( * P  < 0.05, ** P  < 0.01, *** P  < 0.001 vs Sham; # P  < 0.05, ## P  < 0.01, ### P  < 0.001 vs IUA) To assess functional endometrial recovery, IUA mice were mated with proven-fertile males 28 days post-intervention. Uterine horns were examined at gestational day 8.0 (post-vaginal plug detection). Sham controls exhibited symmetrical uterine horns with uniformly sized embryos, whereas IUA mice showed near-complete fertility loss (Fig.  5 B). The combined treatment group demonstrated significantly higher embryo counts (9.33 ± 2.08) compared to HFTMSCs (1.66 ± 2.08) or E2 monotherapy (1.66 ± 2.08), approaching sham-operated controls levels (7.33 ± 0.577) without statistical significance ( P  > 0.05) (Fig.  5 E). Pregnancy outcomes across groups revealed marked contrasts (Fig.  5 C and Supplementary Table 5). Sham-operated controls exhibited a 100% pregnancy rate, which plummeted to 33.33% in IUA models. Both HFTMSCs and E2 monotherapy partially restored fertility, achieving a pregnancy rate of 66.67%. Notably, the combined treatment with HFTMSCs and E2 significantly restored reproductive capacity, reaching a pregnancy rate of 100%, which was comparable to that of the Sham-operated controls. This synergistically enhanced efficacy may be attributed to the following mechanisms: HFTMSCs + E2 co-treatment promotes endometrial gland regeneration, substantially improves the endometrial microenvironment to facilitate embryo implantation [ 36 ], and exerts dual anti‑fibrotic and pro‑proliferative actions. Differences in neonatal outcomes were observed among the groups. Longitudinal observation revealed that healthy offspring were delivered in each treatment group, with successful births of viable pups, normal postpartum vitality, and no significant abnormalities in growth and development (Fig.  5 D, F). Together, these findings demonstrate that treatment with HFTMSCs combined with estradiol comprehensively improves reproductive parameters in mice with IUA, including the restoration of fertility, increased pregnancy rate, and higher litter size. Notably, the untreated IUA model group also delivered viable pups after natural mating at 2 months post-surgery, suggesting spontaneous recovery capacity in IUA induced by mechanical injury. To elucidate the therapeutic mechanisms of HFTMSCs in IUA, uterine tissue samples exhibiting significant therapeutic effects after 14-day of treatment were selected from three experimental groups (HFTMSCs-treated, IUA model, and sham-operated groups; n  = 3/group, totaling 9 samples) for RNA sequencing (RNA-seq) analysis. The RNA-seq data analysis revealed that principal component analysis (PCA) showed clear separation among the different treatment groups (Fig.  6 A), indicating excellent reproducibility and reliability of the data. Differential expression gene (DEGs) profiling identified 4,583 injury-responsive genes (923 upregulated, 3,660 downregulated) in IUA versus Sham controls. HFTMSCs administration conversely modulated 3,471 genes (2,855 upregulated, 616 downregulated) compared to untreated IUA models (Fig.  6 B, C). Venn analysis pinpointed 445 fibrogenesis-related genes overexpressed post-injury yet suppressed by HFTMSCs therapy, alongside 2,188 regeneration-associated genes downregulated in IUA but reactivated through treatment (Fig.  6 D, E).KEGG and GO enrichment analyses indicated that the major pathways involved in fibrosis induced by IUA injury in the endometrium included the Wnt signaling pathway, extracellular matrix-related pathways, TGF-β signaling pathway, MAPK signaling pathway, and PI3K-Akt signaling pathway (Fig.  6 F-G). After HFTMSC treatment, the upregulated pathways were associated with cell adhesion molecules, cytokine receptor interactions, cell differentiation, immune regulation (Fig.  6 H-I). Further analysis revealed that the TGF-β and PI3K/Akt signaling pathways may play pivotal roles in the pathogenesis and progression of IUA. Leukemia inhibitory factor (LIF), an IL-6 family cytokine, exhibited multifaceted interactions with Wnt signaling through three principal modalities: 1) direct molecular cooperation exemplified by STAT3-β-catenin complex formation; 2) indirect regulatory mechanisms via the PI3K-Akt/GSK3β signaling axis. These synergistic interactions collectively orchestrate stem cell fate specification, tissue regeneration dynamics, and pathological progression [ 37 ]. HFTMSCs were identified as the cellular therapeutic agent in this study. Through immunofluorescence and Western blot analyses at post-treatment days 14 and 28, we observed incomplete restoration of LIF expression in untreated IUA specimens, whereas HFTMSCs administration induced dose-dependent recovery of LIF levels. This expression pattern substantiates the mechanistic role of LIF upregulation in endometrial regeneration and receptivity restoration within the injury microenvironment. Fig. 6 Mechanistic elucidation of HFTMSCs-mediated uterine regeneration via fibrotic pathway modulation. A Principal component analysis (PCA) demonstrating distinct transcriptional clustering among Sham controls, IUA models, and HFTMSCs-treated groups ( n  = 3). B Volcano plot of differentially expressed genes (DEGs) between IUA and Sham cohorts (4,583 DEGs: 923 upregulated, 3,660 downregulated). C Volcano plot of DEGs modulated by HFTMSCs therapy versus untreated IUA models (3,471 DEGs: 2,855 upregulated, 616 downregulated). D Venn diagram identifying 445 fibrogenesis-associated genes upregulated post-IUA injury but suppressed by HFTMSCs intervention. E Venn diagram revealing 2,188 regeneration-related genes downregulated in IUA pathology but reactivated through HFTMSCs treatment (blue: downregulation, yellow: upregulation). F KEGG enrichment of pro-fibrotic pathways upregulated in IUA endometria, including Wnt/β-catenin-mediated ECM remodeling and TGF-β/Smad-driven EMT (IUA group). G GO enrichment network of fibrotic DEGs upregulated post-IUA modeling and suppressed by HFTMSCs (IUA group). H KEGG analysis of regenerative pathways activated by HFTMSCs transplantation, featuring cell adhesion molecule signaling and MAPK-regulated proliferation (T group). I GO enrichment of regenerative DEGs upregulated post-HFTMSCs transplantation and downregulated in IUA pathology (T: group). abbreviation: T Group: Treatment group Mechanistic elucidation of HFTMSCs-mediated uterine regeneration via fibrotic pathway modulation. A Principal component analysis (PCA) demonstrating distinct transcriptional clustering among Sham controls, IUA models, and HFTMSCs-treated groups ( n  = 3). B Volcano plot of differentially expressed genes (DEGs) between IUA and Sham cohorts (4,583 DEGs: 923 upregulated, 3,660 downregulated). C Volcano plot of DEGs modulated by HFTMSCs therapy versus untreated IUA models (3,471 DEGs: 2,855 upregulated, 616 downregulated). D Venn diagram identifying 445 fibrogenesis-associated genes upregulated post-IUA injury but suppressed by HFTMSCs intervention. E Venn diagram revealing 2,188 regeneration-related genes downregulated in IUA pathology but reactivated through HFTMSCs treatment (blue: downregulation, yellow: upregulation). F KEGG enrichment of pro-fibrotic pathways upregulated in IUA endometria, including Wnt/β-catenin-mediated ECM remodeling and TGF-β/Smad-driven EMT (IUA group). G GO enrichment network of fibrotic DEGs upregulated post-IUA modeling and suppressed by HFTMSCs (IUA group). H KEGG analysis of regenerative pathways activated by HFTMSCs transplantation, featuring cell adhesion molecule signaling and MAPK-regulated proliferation (T group). I GO enrichment of regenerative DEGs upregulated post-HFTMSCs transplantation and downregulated in IUA pathology (T: group). abbreviation: T Group: Treatment group

All animal studies complied with the Guidelines for the Care and Use of Laboratory Animals, and the use of all experimental animals was approved by the Ethics Committee of Ningxia Medical University (Certification No. IACUC-NYLAC-2024–149). All laboratory animals were obtained from the Laboratory Animal Center of Ningxia Medical University. Female C57BL/6 J mice (aged 6–8 weeks, weighing 18 ± 0.5 g) were maintained in a sterile Specific Pathogen Free (SPF) environment under controlled conditions (room temperature: 22 ± 0.5℃, humidity: 55 ± 2%, a 12-h light/dark cycle, with ad libitum access to drinking water and standard rodent chow). The IUA model was established in female C57BL/6 J mice via bilateral mechanical injury to the uterine horns, as previously described [ 22 – 25 ]. The detailed experimental workflow is illustrated in Fig. S1A. Briefly, the estrus phase was confirmed in all mice by vaginal cytology prior to surgery. Anesthesia was induced by intraperitoneal injection of sodium pentobarbital (1%, 30 mg/kg). A midline abdominal incision (approx. 10 mm) was made to expose the uterine horns. Endometrial injury was inflicted using a custom-made curette, applying fifty longitudinal strokes per horn. Hemorrhagic hyperemia was visually confirmed before returning the uterus to the abdominal cavity and closing the incision in layers. The surgical procedure is detailed in Fig. S3A. Seventy-five mice were randomly divided into five groups ( n  = 15/group): Group 1 (Sham): Underwent sham surgery (exposure without curettage). Group 2 (IUA Model): Underwent curettage-induced injury. Served as the disease control. Group 3 (HFTMSCs Monotherapy): Underwent curettage-induced injury, followed by a spatiotemporal transplantation of HFTMSCs (see Sect. "  Therapeutic intervention protocols "). Group 4 (E₂ Monotherapy): Underwent curettage-induced injury, followed by estradiol treatment. Group 5 (HFTMSCs + E₂ Combination Therapy): Underwent curettage-induced injury, followed by the same spatiotemporal HFTMSCs transplantation combined with estradiol therapy. Group 1 (Sham): Underwent sham surgery (exposure without curettage). Group 2 (IUA Model): Underwent curettage-induced injury. Served as the disease control. Group 3 (HFTMSCs Monotherapy): Underwent curettage-induced injury, followed by a spatiotemporal transplantation of HFTMSCs (see Sect. "  Therapeutic intervention protocols "). Group 4 (E₂ Monotherapy): Underwent curettage-induced injury, followed by estradiol treatment. Group 5 (HFTMSCs + E₂ Combination Therapy): Underwent curettage-induced injury, followed by the same spatiotemporal HFTMSCs transplantation combined with estradiol therapy. HFTMSCs at passages 3 to 5 (P3-P5), which met the International Society for Cellular Therapy criteria (viability > 95%; positive for CD90, CD105, CD44 > 98%; negative for CD34, CD45, HLA-DR < 2%; Fig.  1 C), were used for transplantation. For administration, P3 HFTMSCs were resuspended in Dulbecco’s phosphate-buffered saline (DPBS, PWL052, Meilun Bio, Dalian, China) at a concentration of 1 × 10⁶ cells/mL. Fig. 1 Isolation, Characterization, and In Vitro / In Vivo Tracking of HFTMSCs. A Representative morphological characteristics of HFTMSCs at different culture passages (P0, P1, P3, P20). B Multilineage differentiation potential: a , b Undifferentiated cells (negative control) refer to cells grown in complete α-MEM medium; c Osteogenic differentiation evidenced by Alizarin Red-stained calcium deposits; d Adipogenic differentiation revealed through Oil Red O-labeled lipid droplets. C Flow cytometric profiling of P3 HFTMSCs confirming CD 90/CD 105/CD 44 positivity (> 95%) and CD 34/CD 45/HLA-DR negativity (< 3%) (Isotype controls: Fig. S1D). D-E Immunofluorescence validation of mesenchymal (CD 90/CD 105/CD 44)/pluripotency (OCT 4) markers. F CM-Dil-labeled P3 HFTMSCs displayed orange-red fluorescence (upright microscope). G-H CM-DiI-labeled HFTMSCs localized to endometrial, myometrial, and stromal regions at 14 days, with sparse endometrial signals observed at 28 days. (E = endometrial epithelium, S = endometrial stroma, M = myometrium). I CCK-8 proliferation assay confirming preserved viability across CM-DiI concentrations (0–5 μM; ns: no significance). J-K Biodistribution of CM-DiI-labeled HFTMSCs after tail‑vein injection. Systemic retention in major organs was assessed by fluorescence imaging at 1 and 7 days post‑injection (dpi) Isolation, Characterization, and In Vitro / In Vivo Tracking of HFTMSCs. A Representative morphological characteristics of HFTMSCs at different culture passages (P0, P1, P3, P20). B Multilineage differentiation potential: a , b Undifferentiated cells (negative control) refer to cells grown in complete α-MEM medium; c Osteogenic differentiation evidenced by Alizarin Red-stained calcium deposits; d Adipogenic differentiation revealed through Oil Red O-labeled lipid droplets. C Flow cytometric profiling of P3 HFTMSCs confirming CD 90/CD 105/CD 44 positivity (> 95%) and CD 34/CD 45/HLA-DR negativity (< 3%) (Isotype controls: Fig. S1D). D-E Immunofluorescence validation of mesenchymal (CD 90/CD 105/CD 44)/pluripotency (OCT 4) markers. F CM-Dil-labeled P3 HFTMSCs displayed orange-red fluorescence (upright microscope). G-H CM-DiI-labeled HFTMSCs localized to endometrial, myometrial, and stromal regions at 14 days, with sparse endometrial signals observed at 28 days. (E = endometrial epithelium, S = endometrial stroma, M = myometrium). I CCK-8 proliferation assay confirming preserved viability across CM-DiI concentrations (0–5 μM; ns: no significance). J-K Biodistribution of CM-DiI-labeled HFTMSCs after tail‑vein injection. Systemic retention in major organs was assessed by fluorescence imaging at 1 and 7 days post‑injection (dpi) HFTMSCs Transplantation (Spatiotemporal Strategy): A dual-route transplantation strategy was employed, adapted from previous work [ 26 ]. Immediately after curettage, mice in Groups 3 (HFTMSCs Monotherapy) and 5 (Combination Therapy) received a local intrauterine injection of 50 µL cell suspension (1 × 10⁶ cells/mL) into each horn [ 27 ]. Mice in Groups 1 (Sham), 2 (IUA Model), and 4 (E₂ Monotherapy) received an equal volume (50 µL per horn) of sterile DPBS vehicle via the same route. On postoperative day 7, mice in Groups 3 and 5 received a systemic intravenous injection via the tail vein of 100 µL of the same cell suspension (total 1 × 10 6 cells). Estradiol (E₂) Treatment: Mice in Groups 4 (E₂ Monotherapy) and 5 (Combination Therapy) received intramuscular injections of 17β-estradiol (E₂, Sigma E8515) at a dose of 0.1 mg/kg every four days for a total of 21 days, starting immediately after surgery. Mice in Groups 1, 2, and 3 received vehicle injections on the same schedule. To monitor the homing and retention of transplanted cells, a subset of HFTMSCs was labeled with the fluorescent dye CM-DiI (see Sect. "  CellTracker™ CM-DiI labeled HFTMSCs and proliferation assays "). Only mice in Groups 3 and 5 designated for this tracking experiment received CM-DiI-labeled cells via the spatiotemporal strategy described above. The distribution of fluorescent signals was assessed using an in vivo imaging system at 24 h (Day 1) and 168 h (Day 7) after the systemic injection (i.e., on post-operative days 8 and 14). Uterine tissues from these time points were also harvested for fluorescence microscopy analysis. To systematically evaluate therapeutic efficacy, the 75 mice (15 per group) were allocated to different endpoints as detailed in Supplementary Table 1. Uterine tissues were harvested at postoperative days 14 and 28 for macroscopic examination, histopathological analysis (H&E and Masson's trichrome staining), and molecular profiling (immunofluorescence and Western blot). Additionally, separate independent cohorts (not included in the aforementioned 75 mice) were used for the in vivo cell tracking experiment (Sect. "  CellTracker™ CM-DiI labeled HFTMSCs and proliferation assays ") and fertility testing on day 28 (Sect. "  Fertility testing "). The overall experimental workflow is illustrated in Fig. S1B. Human fallopian tube tissues were obtained from three ( n  = 3) patients undergoing hysterectomy with bilateral salpingectomy for uterine leiomyoma at the General Hospital of Ningxia Medical University, following written informed consent and ethical approval (Approval No. KYLL-2024–0529). Following surgical resection, fallopian tube tissues were transported to the laboratory on ice within 2 h. The tissues were rinsed with pre-cooling 1 × PBS to remove residual blood, after which the serosal layer was carefully dissected and discarded under sterile conditions. The remaining mucosal and muscular layers were minced into a paste-like consistency using sterile ophthalmic scissors and transferred into 15 mL conical tubes. Sequential enzymatic digestion was then performed, tissues were first incubated with 1 mg/mL collagenase type IV (17,104,019, Thermo Fisher, USA) at 37℃ for 70 min under continuous agitation, the digested material was subjected to centrifugation at 1,000 revolutions per minute for 10 min at ambient temperature to collect the fragmented components. The supernatant was carefully removed, and the remaining tissue was then treated with a 0.25% trypsin–EDTA solution (G4011, Servicebio, Wuhan, China) for 8 min under gentle agitation at 37℃. Following the addition of complete culture medium to neutralize the solution, the mixture was again centrifuged at 1,000 rpm for 5 min at room temperature. The isolated cell pellet was resuspended in α-MEM (12,571,048, Thermo Fisher, USA) containing 10% fetal bovine serum (PWL002-1, MeiluniBo, China), 5 ng/mL recombinant human basic fibroblast growth factor (bFGF, 100-18B, PeproTech, USA), and 1% penicillin–streptomycin (10,378,016, Thermo Fisher, USA). The cells were then seeded into 60-mm culture plates and cultured in a 5% CO₂ incubator at 37℃. After an initial 3-day adaptation period, the medium was refreshed every 48 h to eliminate non-adherent cells and debris. By the 12th day of cultivation, the adherent cells had reached approximately 90% confluence, exhibiting a uniform spindle-shaped morphology typical of mesenchymal stem cells. For subsequent passage, the cells were detached using a trypsin–EDTA solution and diluted at a 1:5 ratio. By the third passage, a homogeneous population of spindle-shaped mesenchymal stem cells was achieved. Prior to experimental use, the cells were characterized for their surface marker expression and multipotency, including their ability to differentiate into adipocytes and osteoblasts, to confirm their mesenchymal stem cell properties. Surface antigen profiling of P3 HFTMSCs was conducted in accordance with established flow cytometry protocols to characterize cellular phenotypes. Cells cultured in 60-mm dishes were harvested at 70–80% confluence via trypsinization (0.25% trypsin–EDTA, 37℃ for 2 min) and neutralized with 2 mL complete medium. The cell suspension was centrifuged at 1,000 rpm for 5 min, washed twice with DPBS, and resuspended in 200 μL DPBS at a density of 1 × 10⁶ cells/mL. Per tube were incubated with fluorochrome-conjugated monoclonal antibodies (BD Biosciences, USA) for 30 min at room temperature in the dark, using the following antibody panel, Positive markers, CD 105-PerCP (Cat#560,819), CD 90-PerCP (Cat#561,557), CD 44-APC (Cat#560,532). Negative markers, HLA-DR-PE (Cat#562,304), CD 34-PE (Cat#555,822), CD 45-PE (Cat#666,483). Isotype controls: IgG1-PE (Cat#562,306), IgG1-PerCP (Cat#560,819), IgG1-APC (Cat#560,183), IgG1-PE (Cat#557,872). Following antibody incubation, cells were washed twice with DPBS (1,000 rpm, 5 min) and resuspended in 250 μL pre-cooling DPBS. Samples were immediately analyzed on a BD FACSVerse flow cytometer (BD Biosciences). Data acquisition was performed using BD FACSuite software (v1.6.1). To determine the background fluorescence thresholds, isotype control antibodies matched in species, subtype, and labeling were processed in parallel. To verify the expression of mesenchymal stem cell markers, third-passage HFTMSCs were plated on glass coverslips in 24-well plates at 2 × 10 3 cells/well (500 μL medium) and cultured until 70% confluency. Cells were fixed with 4% paraformaldehyde for 20 min at room temperature, rinsed three times with PBS (5 min each), and then incubated with 3% bovine serum albumin (BSA) for 1 h to block non-specific binding. After three additional PBS washes, cells were incubated overnight at 4℃ with primary antibodies diluted in PBS. CD 90 (1:50, Cat#222,519, Zen BioScience, China). CD 105 (1:50, Cat#222,133, Zen BioScience). CD 44 (1:50, Cat#200,840, Zen BioScience). HLA-DR (1:50, Cat#R381263, Zen BioScience). OCT 4 (1:50, Cat#sc-5279, Santa Cruz Biotechnology, USA). After incubation with primary antibodies, cells were washed three times with PBS and incubated for 1 h at room temperature with Alexa Fluor 488-conjugated goat anti-mouse IgG secondary antibody (1:200, Zen BioScience) in the dark. The nuclei were counterstained with 5 μg/mL DAPI for 10 min. Negative controls were processed identically but without primary antibody incubation. Fluorescent images were acquired using a Nikon confocal laser scanning microscope (Nikon DS-Ri1, Japan). To evaluate gene expression profiles, total RNA was extracted from P3–P5 HFTMSCs using the Total RNA Kit I (Cat# R6834, Omega Bio-tek, USA) following the manufacturer's instructions. RNA purity and concentration were measured using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, USA), and only samples with A260/A280 ratios between 1.8 and 2.0 were used for further analysis. Reverse transcription was carried out using 1 μg of total RNA and the PrimeScript RT Reagent Kit (Takara Bio, Japan) under the conditions: 37℃ for 15 min and 85℃ for 5 s. After RNA extraction and cDNA synthesis. Each 25 μL reaction mixture included 2 μL of cDNA template, 0.5 μM forward and reverse primers (Sangon Biotech, Shanghai), 12.5 μL of 2 × Taq PCR Master Mix (Tiangen Biotech, China), and nuclease-free water. Amplification was performed at 95℃ for 3 min, followed by 35 cycles of 95℃ for 30 s, annealing at a primer-specific temperature (55–62℃) for 30 s, and extension at 72℃ for 45 s, with a final extension at 72℃ for 5 min. Amplified products were separated on 1.5% agarose gels in 1 × TAE buffer (Servicebio, G3001, pH 8.0–8.3). Electrophoresis was conducted at 120 V for 40 min using a Bio-Rad PowerPac Basic system. DNA bands were visualized and analyzed using Bio-Rad Gel Doc XR + and Image Lab 6.1 software. Gene expression levels were normalized to the housekeeping gene GAPDH . And DL 500 DNA Marker (Takara, 3590Q). Primer sequences (synthesized by Sangon Biotech, Shanghai, China) are detailed in Supplementary Table 2. To validate the multipotent differentiation capacity of HFTMSCs, P3 cells were cultured in 6-well plates and induced toward adipogenic and osteogenic lineages using commercially available differentiation media. For adipogenic differentiation, cells were maintained in HyCyte™ Adipogenic Induction Medium (Cat# UCHX-D102R, Haixing Biotechnology, Suzhou, China) for 14–21 days, with medium changes every 3 days. Similarly, osteogenic differentiation was initiated using HyCyte™ Osteogenic Induction Medium (Cat# UCHX-D101R, Haixing Biotechnology) under identical culture conditions. Non-induced counterparts cultured in standard growth medium served as negative controls. Following the induction period, adipogenic differentiation was confirmed by Oil Red O staining: cells were fixed with 4% paraformaldehyde, and stained with Oil Red O solution to visualize lipid droplet formation. For osteogenic assessment, calcium deposition was detected via Alizarin Red S staining. Cells were fixed and incubated with Alizarin Red S for 10 min, followed by extensive washing to remove nonspecific staining. All stained samples were imaged under phase-contrast microscopy (Nikon, Japan). To assess the chromosomal stability of HFTMSCs, P3 HFTMSCs were utilized. By incubating the cells with Colcemid solution (0.1 μg/mL, Gibco, USA) at 37℃ for 2 h, the process of cellular division was blocked, causing HFTMSCs to arrest at metaphase. Subsequently, the cell pellet was resuspended in a hypotonic solution and incubated at 37℃ for 15 min. The cells were then fixed with methanol at 4℃ for 30 min, followed by centrifugation and supernatant removal. Pre-chilled fixative solution was added to resuspend the cells into a single-cell suspension. After spreading onto chilled glass slides, the samples were treated at 75℃ for 3 h and subjected to Giemsa staining, followed by drying at 37℃. Metaphase images were captured using a Leica fully automated scanner (Leica, Germany), and karyotype analysis was performed with CytoVision software (Leica, Germany). To monitor the fate and retention of transplanted HFTMSCs, cells were fluorescently labeled using CellTracker™ CM-DiI (Cat# C7001, Thermo Fisher Scientific, USA). Briefly, HFTMSCs at passage 3 were resuspended in DPBS and incubated with 3 μM CM-DiI for 10 min at 37℃ under 5% CO₂, followed by a 15-min incubation at 4℃ to stabilize the dye. After two DPBS washes, labeled cells was centrifuged at 1,000 rpm for 5 min, resuspended in fresh DPBS, and fluorescence intensity was verified using an Olympus BX51 fluorescence microscope (Olympus, Japan). For in vivo tracking, CM-DiI-labeled HFTMSCs were transplanted into the uterine horns of IUA mice (n = 4) via intrauterine injection. On postoperative day 7, an additional 1 × 10⁶ labeled cells were administered via tail vein injection. Mice were euthanized at 1-and 7-days post-injection (dpi) for comprehensive biodistribution analysis. To assess the in vivo homing and retention of the systemically administered (tail‑vein) HFTMSCs, a subset of CM‑DiI‑labeled cells was used. Mice were euthanized at 1 and 7 days after the second (tail‑vein) injection (i.e., 8 and 14 days post‑surgery) for comprehensive biodistribution analysis. At each endpoint, uteri and major organs (heart, liver, spleen, lungs, kidneys) were harvested and imaged ex vivo using a Carestream In-Vivo FX Pro imaging system (Carestream Health, USA) with excitation/emission settings at 553/570 nm. For cellular localization studies, uterine tissues collected at 14 and 28 days were embedded in OCT compound, sectioned at 10 μm thickness, and nuclear staining with DAPI (5 μg/mL, 10 min). Fluorescent signals from CM-DiI-labeled cells were visualized using a Nikon DS-Ri1 confocal microscope (Nikon, Japan). Interpretation Note for the Tracking Experiment: It is important to note that the mice used for in vivo fluorescence tracking received CM-DiI-labeled HFTMSCs via both intrauterine (at surgery) and intravenous (on day 7) routes, following the therapeutic spatiotemporal strategy. Therefore, the fluorescent signals detected in the uterus and other organs at subsequent time points represent the composite presence of cells delivered through both pathways. This design allows for the assessment of overall cell retention and distribution patterns but does not permit the definitive distinction of signals originating specifically from the intravenously administered cell fraction. To assess the potential cytotoxic effects of CM-DiI labeling, HFTMSCs were seeded in 96-well plates at a density of 4 × 10 3 cells/well and divided into three groups, unlabeled control, CM-DiI-labeled experimental, and medium-only blank. Cells in the experimental group were incubated with CM-DiI at concentrations ranging from 0 to 5 μM (1 μM increments) for 24 h. Cellular proliferation was evaluated at days 1, 3, 5, and 7 post-labeling using the CCK-8 assay (Beyotime Biotechnology, China). At each timepoint, 10 μL of CCK-8 reagent was added to each well, followed by a 2-h incubation at 37℃. Absorbance at 450 nm was measured using a SpectraMax M5 microplate reader (Molecular Devices, USA), with background subtraction based on blank controls. Cell viability was calculated as: Viability (%) = (Experimental OD—Blank OD)/(Control OD450—Blank OD450) × 100%. All experiments were performed in triplicate, with data normalized to day 1 viability (100%) and presented as mean ± SD. To systematically evaluate the proliferation capacity and senescence characteristics of HFTMSCs during in vitro expansion, cells at passages 5, 10, 15, and 20 (P5, P10, P15, P20) in the logarithmic growth phase were selected for detailed proliferation curve plotting and population doubling time (PDT) determination. HFTMSCs at P5, P10, P15, and P20 in the logarithmic growth phase were cultured in complete α-MEM medium supplemented with 10% fetal bovine serum and 1% penicillin‑streptomycin. After digestion with trypsin, the cells were resuspended and adjusted to a density of 1 × 10 4 cells/mL. Subsequently, 2 mL of cell suspension (approximately 2 × 10 4 cells) was seeded into each well of a 6 ‑ well plate, with three biological replicates per group. The cells were maintained at 37 ℃ under 5% CO₂, and the medium was replaced every 2—3 days. On days 1, 3, 5, and 7 after seeding, cells were harvested by trypsinization, washed with PBS, and stained with trypan blue for viable cell counting, with a viability rate > 90% ensured. A growth curve was plotted with culture time as the horizontal axis and the number of cells as the vertical axis. Using data from the exponential growth phase (days 3—5, corresponding to a 48 ‑ hour interval), the population doubling time was calculated according to the formula: DT (h) = (T × lg2)/lg(Nₜ/N₀). Uterine tissue samples were collected from 6 and 3 mice per group at 14-day and 28-day post-treatment, respectively, for each group of mice, uterine tissue samples were collected for the detection of the expression levels of endometrial receptivity markers LIF and αvβ3. Initially, proteins were extracted from these samples using RIPA lysis buffer containing PMSF (P0013B, Beyotime, Shanghai, China). Protein quantification was performed using the BCA method (G2026, Servicebio, Wuhan, China), ensuring that 30 μg of total protein was loaded onto each lane for subsequent Western blot analysis. Proteins were separated by SDS-PAGE and subsequently transferred to a PVDF membrane (Millipore, Bedford, MA, USA). The membrane was blocked with 5% skim milk at room temperature for 1.5 h and then incubated overnight at 4℃ with primary antibodies against LIF (1:1000, Proteintech), αvβ3 (1:500, Affinity), and alpha-tubulin (1:500, abcam, USA). Subsequently, the membrane was treated with a secondary antibody (Goat Anti-Rabbit IgG, Servicebio, Wuhan, China) at room temperature for one hour. The target protein bands were then detected using an enhanced chemiluminescence (ECL) detection kit (G2020, Servicebio, Wuhan, China). Uterine tissues were collected at 14 days and 28 days post-treatment, with 6 and 3 mice per group at the respective time points. Specimens were fixed in 4% paraformaldehyde (24 h), paraffin-embedded, and sectioned at 5 μm thickness slices. The sections were baked at 65℃ for 1 h to enhance adhesion, then deparaffinized with xylene (for 5 min), hydrated using graded ethanol, and finally rinsed with PBS. The sections were placed in citrate buffer (pH 6.0) in a pressure cooker and boiled for 3 min to perform antigen retrieval, and then washed three times with PBS (5 min each time). For immunostaining, tissue sections were incubated overnight at 4℃ with primary antibodies diluted in PBS. Primary antibodies used included Leukemia Inhibitory Factor (LIF) (1:200, Cat# GB12147, Serviceo, Wuhan, China). Integrin αvβ3 (1:200, Cat# AF6086, Affinity Biosciences, China). α-Smooth Muscle Actin (α-SMA) (1:200, Cat# 67,735–1-Ig, Proteintech). Transforming Growth Factor-β1 (TGF-β1) (1:200, Cat# AF1027, Affinity Biosciences). Cytokeratin 7 (CK7) (1:200, Cat# GB11063, Servicebio, China). CD31 (1:200, Cat# GB112225 , Servicebio). On the second day, secondary antibodies, namely Goat Anti-Mouse IgG conjugated with Alexa Fluor 594 (1:500, Invitrogen, USA) and Alexa Fluor 488 Goat Anti-Rabbit IgG (1:200, Servicebio), were applied and incubated at room temperature for 1 h. Nuclear staining was performed using DAPI. The tissue sections were imaged under a fluorescence confocal microscope (Nikon, Japan). Immunofluorescence was quantified by measuring mean fluorescence intensity (MFI) using ImageJ 8.0. For each biological replicate, three non‑consecutive sections were analyzed. Within the endometrial region of each section, three random high‑power fields (400 ×) were captured. MFI was measured in manually defined ROIs following background subtraction from adjacent unstained areas. The average MFI per animal served as a single data point for statistical analysis. Uterine tissues collected from 6 and 3 mice per group at 14-day and 28-day post-treatment were fixed in 4% paraformaldehyde (24 h), dehydrated through a graded ethanol series, and paraffin-embedded. Transverse Sects. (5 μm) were deparaffinized, rehydrated, and subjected to antigen retrieval in citrate buffer (pH 6.0) using a pressure cooker. For morphological evaluation, sections were stained with hematoxylin and eosin (H&E). Endometrial thickness (vertical distance from basal layer to luminal surface) and glandular numbers were quantified in four randomly selected areas per section using ImageJ 8.0. To characterize fibrotic remodeling, parallel sections were subjected to Masson's trichrome staining (G1340; Solarbio, China) following manufacturer's protocols. Collagen deposition was quantified by calculating the percentage of blue-stained fibrotic areas relative to total tissue area through threshold-based image analysis in ImageJ. To assess the impact of HFTMSCs on murine estrous cycles, daily vaginal cytology was performed at 10:00 a.m. beginning on post-treatment day 14 for 14 consecutive days. Exfoliated epithelial cells were collected by gentle introduction of saline-moistened sterile swabs into the vaginal lumen. The obtained cellular material was uniformly spread onto glass slides and allowed to air-dry for a quarter of an hour prior to processing with an UltraFast Modified Papanicolaou Stain Kit (Nanjing Jiancheng Bioengineering Institute, Cat# D022-1–1, China) according to the manufacturer’s protocol. The estrous cycle stages were categorized according to the cellular morphology observed under light microscopy [ 28 ]. Proestrus (the prevalence of nucleated epithelial cells). Estrus (Most cornified squamous epithelial cells). Metestrus (a blend of keratinized cells and white blood cells). Diestrus (Significant presence of leukocytes). A typical estrous cycle is characterized by a sequential progression through proestrus, estrus, metestrus, and diestrus phases within a 4—5 day period. To evaluate uterine receptivity and embryo developmental capacity, the remaining six mice per group ( n  = 6) underwent fertility assessment starting at 28 days post-treatment. Females from each group were co-housed with proven fertile 8-week-old male mice at a 2:1 ratio. Vaginal plugs were inspected daily at 08:00, and the day of detection was designated as gestational day 0.5 (GD 0.5). At GD 8.0, three mice per group were euthanized for embryo counting, and their uteri were collected for systematic quantification of implantation sites via morphological analysis (Fig. S1C). The remaining three pregnant mice per group were allowed to deliver at term for assessment of litter size. In order to assess the tumorigenic potential of HFTMSCs, six 4-week-old female BALB/c-nu (Beijing HuaFuKang Biotechnology Co., Ltd., License No. SCSK (Jing) 2024–0003) were randomized into two groups ( n  = 3) based on body weight: test group receiving HFTMSCs (1 × 10⁷ cells/mouse) and positive sham-operated controls administered human induced pluripotent stem cells (HiPSCs, 1 × 10⁷ cells/mouse). In the experiment, a single 0.2 mL subcutaneous injection was administered to the left scapular region of the mice. During the 8-week observation period, animals were monitored daily for clinical signs and weekly for palpable tumor formation (using caliper measurements). At the end of the study, mice were euthanized via cervical dislocation. Autopsy included a gross examination of major organs (heart, liver, spleen, lungs, kidneys) and histopathological analysis of tissues at the injection sites. Suspected tumors and lesions were fixed in 4% paraformaldehyde, embedded in paraffin, and sectioned for H&E staining. To elucidate the mechanisms of HFTMSCs-mediated uterine repair, uterine tissues were collected at the 14-day time point post-treatment from three groups ( n  = 3): the sham control, the IUA model, and the HFTMSCs monotherapy group, as detailed in Supplementary Table 1.Total RNA was isolated with TRIzol Reagent (Thermo Fisher Scientific), followed by verification of purity (NanoDrop 2000; A260/A280 = 1.8–2.0) and integrity (Agilent 2100 Bioanalyzer; RIN ≥ 8.0). Ribosomal RNA-depleted samples were used to construct strand-specific mRNA libraries (Hieff NGS® Ultima Dual-mode mRNA Library Prep Kit, Yeasen), with quality assessed via Qubit fluorometry and Agilent TapeStation. Paired-end sequencing (150 bp, ≥ 40 million reads/sample) was performed on an Illumina NovaSeq 6000 platform (Guangzhou Genedenovo Biotechnology). Raw data underwent stringent quality control: Trimmomatic (v0.39) trimmed adapters, and FastQC (v0.12.1) filtered low-quality reads (Phred score < 30). HISAT2 (v2.2.1) aligned high-quality reads to the Mus musculus genome (GRCm38/mm10), followed by transcript quantification with featureCounts (v2.0.3). Differential gene expression analysis using DESeq2 (v1.34.0) identified DEGs under thresholds of adjusted p-value  1. Functional annotation of DEGs included Gene Ontology (GO: biological process, cellular component, molecular function) enrichment and KEGG pathway mapping via clusterProfiler. For statistical comparisons, one-way ANOVA was utilized to evaluate intergroup differences, with data presented as mean ± SD. Significance thresholds were defined as * P  < 0.05, ** P  < 0.01, *** P  < 0.001 vs Sham group, while # P  < 0.05, ## P  < 0.01, ### P  < 0.001 vs IUA group. All analyses were executed using GraphPad Prism 8.0 (GraphPad Software, USA).

Intrauterine adhesions (IUA), also known as Asherman syndrome, a pathological condition resulting from damage to the basal layer of endometrium, lead to the replacement of tissue with avascular fibrous tissue and myofibroblasts [ 1 , 2 ], causing partial or complete blockage of the uterine cavity [ 3 – 5 ]. IUA commonly occurs following procedures such as recurrent abortions, dilation and curettage (D&C), infections, or postpartum hemorrhage. The primary pathological consequence is compromised female fertility. Clinical symptoms include reduced menstruation, amenorrhea, secondary infertility, endometriosis, and other related manifestations [ 6 ]. Hysteroscopic transcervical adhesiectomy (TCRA) is the standard treatment for IUA; however, the postoperative recurrence rate remains notably high, ranging from 40 to 62.5% [ 7 – 9 ]. Despite this, accumulating clinical evidence suggests that conventional methods have limited efficacy in restoring the functional morphology of endometrium [ 8 , 10 ]. Consequently, there is an urgent clinical imperative to develop effective treatments aimed at mitigating endometrial fibrosis and enhancing endometrial resilience, thereby preventing the occurrence of IUA and improving clinical outcomes [ 11 ]. Recently, spurred by progress in regenerative medicine, stem cell therapy has gained recognition as a potential treatment for IUA [ 12 – 14 ]. Mesenchymal stem cells (MSCs), derived from a variety of tissue sources, exhibit significant therapeutic potential in preclinical models across a spectrum of pathologies. Their therapeutic modalities involve five core pathways: immune regulation, inflammation suppression, neovascularization stimulation, tissue-specific migration capacity, and secretory signaling functions [ 15 , 16 ]. Current studies demonstrate that MSCs from various sources, including umbilical cord, menstrual blood, and bone marrow, exhibit differing levels of efficacy in repairing IUA and promoting the recovery of damaged endometrium [ 17 , 18 ]. Notably, the therapeutic application of HFTMSCs, a novel MSCs subpopulation with unique topographic advantages in reproductive tissue repair, remains unexplored in IUA therapeutics. HFTMSCs, an emerging entity in regenerative medicine, share an ontogenic lineage with endometrial basal stem cells (EBSCs) due to their common mesodermal origin. These progenitor cells possess essential characteristics, including multilineage differentiation potential, feasibility of in vitro isolation and expansion, and minimal immunogenicity [ 11 , 19 ]. Their shared developmental origins and biological characteristics provide a robust theoretical foundation for advancing research in endometrial regeneration. Although stem cell therapy has shown great potential in regenerative medicine, its clinical translation remains fraught with challenges. The optimal dosage and delivery route for stem cells are yet to be established. Existing studies have explored various administration methods, such as intrauterine injection, intraperitoneal injection, and tail vein injection, each exhibiting distinct differences in cell engraftment efficiency and therapeutic outcomes [ 20 ]. Moreover, the insufficient retention of transplanted cells in vivo, coupled with concerns regarding the long‑term safety and durability of therapeutic effects [ 21 ], continues to impede clinical translation. Notably, most investigations into stem cell therapy for IUA have been confined to animal models, with a dearth of comprehensive assessments regarding the feasibility of translating these strategies into clinical practice. The findings are intended to provide a scientific basis for optimizing stem cell‑based therapies for clinical application. Therefore, this study aimed to evaluate the therapeutic potential of HFTMSCs in repairing damaged endometrial tissues. To this end, HFTMSCs were isolated from human fallopian tubes, characterized, and then xenotransplanted as monotherapy or in combination with estradiol (E2) into a C57BL/6 J mouse model of experimentally induced IUA.

We successfully isolated and characterized mesenchymal stem cells derived from human fallopian tubes. In vitro experiments demonstrated the multi-lineage differentiation potential of these cells, and they were efficiently tracked using CellTracker™ CM-DiI. In vivo experiments, involving subcutaneous tumor formation in nude mice, confirmed the non-tumorigenic nature of HFTMSCs, thereby demonstrating their safety. Additionally, in a mouse model of IUA, the combination of HFTMSCs with E2 significantly repaired the damaged endometrium, providing a theoretical foundation and suggesting potential translational implications for future clinical applications.

In this study, we employed HFTMSCs derived from reproductive tissues to treat murine IUA. Through combined intrauterine injection and tail vein systemic administration, dual transplantation routes of HFTMSCs demonstrated that neither HFTMSCs nor E2 monotherapy showed significant differences in restoring glandular count, endometrial thickness, angiogenesis, or receptivity marker expression in IUA mice. However, when HFTMSCs were co-administered with E2, the therapeutic effects were significantly enhanced. These findings suggest that HFTMSCs represent a novel cellular source with potential value in IUA treatment, and indicate a possible synergistic interaction between stem cell transplantation and hormonal modulation that may optimize therapeutic outcomes. MSCs isolated from reproductive tissues have become a research hotspot in endometrial regeneration due to their unique tissue-repair capabilities. The therapeutic effects of HFTMSCs observed in this study align with preclinical evidence from other reproductive tissue-derived MSCs, such as umbilical cord- and amniotic membrane-derived MSCs. Fallopian tube-derived MSCs have demonstrated efficacy in bone regeneration and autoimmune encephalomyelitis, confirming their osteogenic, xenotransplantable, and immunomodulatory properties [ 38 , 39 ]. Their capacity to shift cytokine balance toward anti-inflammatory profiles is directly relevant to the fibrotic microenvironment of intrauterine adhesions, supporting the translational potential of HFTMSCs for endometrial repair. Notably, clinical-grade HUC-MSCs loaded onto collagen scaffolds have demonstrated both safety and pro-regenerative efficacy in patients with recurrent IUA, highlighting the translational promise of such strategies [ 40 ]. Animal studies further reveal that HUC-MSCs not only promote glandular regeneration by upregulating estrogen receptor (ER) and Ki-67 expression but also remodel the microenvironment through modulation of fibrotic pathways such as TGF-β [ 41 ]. Importantly, pretreatment with melatonin or optimization of delivery methods (e.g. intraperitoneal injection of 1 × 10⁶ cells) significantly enhances therapeutic efficacy [ 42 , 43 ]. underscoring the value of functional modulation strategies for MSCs in IUA treatment. Collectively, these studies affirm the therapeutic versatility of reproductive tissue-derived MSCs in mitigating IUA pathology. Recent studies have shown that the co-transplantation of estrogen and human amnion mesenchymal stem cells (HAMSCs) can effectively promote the differentiation of HAMSs into endometrial epithelial cells by activating the Notch signaling pathway [ 44 ]. In addition, exogenous estrogen can promote the differentiation of endometrial basal stem cells [ 45 ]. Previous studies have reported that the combined use of bone marrow mesenchymal stem cells (BMSCs) and estrogen can significantly promote endometrial repair [ 46 ].Clinical evidence indicates that estrogen mitigates glandular atrophy by driving the differentiation of basal endometrial stem cells [ 47 ]. however, its long-term administration may increase the risk of endometrial carcinogenesis [ 48 ], necessitating safer alternatives. The innovation of this work lies in identifying the high homology between HFTMSCs and endometrial stem cells, a tissue-specific advantage that may circumvent the side effects of exogenous estrogen. Randomized controlled trials (RCTs) confirm comparable efficacy among different estrogen formulations in improving endometrial receptivity [ 49 ], yet combining stem cell therapy could overcome the dose limitations of current regimens. In our model, mechanical injury disrupted endometrial blood supply, leading to fibrotic scar formation and intrauterine fluid accumulation due to impaired drainage of secretions [ 2 ]. Subsequent recruitment of neutrophils and macrophages exacerbated fibrosis via TGF-β1 upregulation, the master regulator of extracellular matrix (ECM) deposition [ 9 , 50 ]. Concomitantly, this mechanical injury directly targets the endometrial epithelium, resulting in substantial loss of CK7-positive luminal and glandular cells. The marked downregulation of CK7, a key cytoskeletal protein essential for epithelial integrity, aligns with established IUA pathology and validates the model’s ability to recapitulate genuine epithelial depletion [ 51 , 52 ]. This epithelial injury provides a clear pathological basis for assessing regenerative therapies. Histopathological analyses at 14 days post-injury confirmed severe structural compromise in IUA mice, marked by reduced glandular counts, diminished endometrial thickness, and elevated fibrosis. Strikingly, HFTMSCs and E2 therapy reversed these pathologies, restoring near-normal endometrial morphology. By day 28, untreated IUA mice exhibited partial spontaneous recovery, likely attributable to attenuated inflammation and microenvironmental improvements. Moreover, Immunofluorescence analysis showed that on day 14 of treatment, HFTMSCs and E2 administration suppressed fibrotic markers and restored receptivity factors, effects that persisted at day 28. Western blot analysis corroborated these findings. Notably, our in vitro characterization revealed that prolonged culture to P20 induced morphological alterations in HFTMSCs, including elongation and a pronounced fibroblast-like phenotype. This change, observed alongside reduced proliferative kinetics at P20, may reflect metabolic adaptation or early signs of replicative senescence, potentially involving shifts in secretory profiles, or cell cycle regulation. Although P20 cells retained considerable proliferative capacity, this morphological transition suggests that lower-passage cells (e.g., P5-P10) should be prioritized in clinical translation to ensure optimal cellular potency and safety. These observations lead us hypothesize that the incomplete restoration of endometrial receptivity markers (αvβ3 and LIF) at 28 days post-treatment may stem from limitations in HFTMSCs differentiation efficiency. Specifically, the restricted proportion of HFTMSCs differentiating into functional endometrial epithelial cells likely contributed to the partial recovery of αvβ3 expression. Previous studies have suggested that combining MSCs with scaffold materials could enhance differentiation efficiency, a strategy that warrants further exploration in this context [ 53 ]. The mechanisms underlying MSCs-mediated endometrial repair remain debated. Our data suggest that HFTMSCs home to injury sites, aligning with prior reports of MSC tropism [ 54 ]. However, the ongoing scientific discourse continues to investigate whether their regenerative potential primarily stems from their differentiation capacity or their paracrine secretory functions [ 55 ]. RNA-seq analysis reveals that HFTMSCs promote endometrial regeneration through a coordinated modulation of fibrotic and regenerative pathways in IUA. RNA-seq profiling of uterine tissues at day 14 post-treatment demonstrated that HFTMSCs significantly altered the expression of 3,471 genes compared to the IUA model, with 2,855 upregulated and 616 downregulated. Notably, Venn analysis identified 445 fibrosis-associated genes suppressed by HFTMSCs treatment, concomitant with the reactivation of 2,188 tissue repair genes, demonstrating a dual mechanism that concurrently inhibits pathological fibrosis and restores regenerative capacity. Pathway enrichment analysis underscored the pivotal roles of ECM remodeling and the Wnt, TGF-β, MAPK, and PI3K-Akt signaling pathways during IUA fibrosis. Abnormal activation of these pathways may underlie IUA development. Conversely, HFTMSCs treatment influenced pathways such as cell adhesion and cytokine-receptor interactions, thereby facilitating tissue repair and suppressing fibrosis. The TGF-β pathway promotes fibrosis by enhancing ECM accumulation and fibroblast activation, whereas the PI3K-Akt pathway supports cell survival and functional recovery. Collectively, we speculate that the therapeutic mechanism of HFTMSCs in IUA involves activating regenerative pathways and inhibiting fibrotic signaling. Leukemia inhibitory factor (LIF), a pleiotropic cytokine critical for endometrial receptivity, was dose-dependently restored by HFTMSCs, further supporting their role in niche remodeling [ 56 ]. Concurrently, TGF-β1, a key driver of ECM deposition, was robustly suppressed, underscoring the dual anti-fibrotic and regenerative actions of HFTMSCs. TGF-β1 plays a crucial role in the pathogenesis of endometrial fibrosis. Following tissue injury, its aberrant upregulation stimulates α-smooth muscle actin (α-SMA) expression in fibroblast-like cells and promotes excessive ECM deposition. Upregulated α-SMA expression is acknowledged as a hallmark characteristic of fibroblast-to-myofibroblast differentiation, a pivotal process in fibrosis development [ 57 ]. Beyond its regulatory roles in cellular growth, development, and tissue remodeling, TGF-β1 directly participates in fibrotic pathogenesis by persistently disrupting ECM homeostasis and sustaining the activation of pro-fibrotic signaling pathways [ 50 ]. Despite the promising preclinical efficacy of HFTMSCs in treating IUA, several limitations must be addressed before clinical translation. First, the current isolation method relies on salpingectomy specimens, which inherently conflicts with the clinical goal of fertility preservation. Although minimally invasive alternatives, including laparoscopic salpingoscopy-guided mucosal biopsy, laparoscopic-guided tubal mucosal biopsy, and combined hysteroscopic-salpingoscopic sampling, have been proposed, their feasibility and stem cell yield require systematic validation. Existing literature indicates that mucosal biopsies performed under laparoscopic salpingoscopy can provide sufficient tissue for cellular and immunohistochemical analyses while maintaining tubal patency [ 58 , 59 ], and sub-millimeter endoscopic methods have been shown to enable in vivo collection of tubal cells with minimal tissue damage [ 60 ]. However, whether these approaches can yield adequate HFTMSCs for autologous expansion and therapy without compromising reproductive anatomy remains to be determined. Second, the use of a single fluorescent label (CM-DiI) for both intrauterine and intravenous delivery precludes precise quantification of route-specific homing efficiency; future studies employing distinct, pathway-specific labels (e.g., different dyes or reporter genes) are needed to delineate the contribution of each administration route to endometrial engraftment. Third, while our murine model recapitulates key features of human IUA, it does not fully replicate the complexity of the human endometrial microenvironment. Therefore, validation in large animal models that more closely approximate human reproductive physiology is imperative, followed by comparative morphological analyses in human IUA patients to confirm therapeutic efficacy. Additionally, comprehensive assessments of long-term safety, immunogenicity, and engraftment dynamics in humanized models are essential to support clinical application. Building on these findings, future research will pursue three complementary directions. First, we will optimize fertility-preserving collection protocols and systematically compare the stem cell characteristics and therapeutic potency of biopsy-derived HFTMSCs with those obtained from resected specimens. Second, to circumvent the inherent risks of direct cell transplantation, we will focus on engineering HFTMSC-derived exosome or secretome-based acellular therapies, capitalizing on their paracrine reparative functions. Third, leveraging advances in 3D organoid technology, we plan to establish human endometrial organoid co-culture systems to investigate the direct cellular crosstalk and paracrine signaling mechanisms underlying HFTMSC-mediated regeneration [ 61 , 62 ]. This physiologically relevant in vitro model will enable precise dissection of anti-fibrotic, pro-angiogenic, and epithelial regenerative pathways, thereby providing a robust platform for mechanistic validation and therapeutic optimization. Collectively, these efforts will be pivotal in translating HFTMSC-based therapies into safe and effective clinical interventions for IUA.

Supplementary Material 1. Supplementary Material 2. Supplementary Material 1. Supplementary Material 2.