To investigate CXCL12 therapeutic potential in endometrial repair post intrauterine adhesion (IUA) surgery.

A murine model of IUA was established using mechanical curettage. A total of 48 healthy female C57BL/6 mice (aged 7 to 8 weeks, weighing 18 to 20 g) were randomly allocated to 4 groups (n=12 per group): the sham-operated group (Sham), the model group (Model), the umbilical cord mesenchymal stem cell treatment group (UC-MSCs treatment group), and the CXCL12 treatment group. In the Sham group, only the abdominal wall was incised without intrauterine manipulation. In the Model group, after mechanical curettage, 100 μL of normal saline was infused into the uterine cavity. In the UC-MSCs treatment group, after curettage, 100 μL of UC-MSCs suspension (1×10⁶ cells) was infused into the uterine cavity. In the CXCL12 treatment group, after curettage, 100 μL of CXCL12 solution (containing 200 ng CXCL12) was infused into the uterine cavity. At 7 d post-modeling, histological examination was performed to confirm model establishment, and immunofluorescence staining was used to assess the chemotactic effect of CXCL12 on MSCs. At 21 d post-modeling, hematoxylin and eosin (HE) staining was used to evaluate endometrial thickness and gland number; Masson’s trichrome staining assessed the extent of endometrial fibrosis. Immunohistochemical staining was employed to determine the microvessel density (MVD) and the expression levels of VEGF, Ki-67, and Sirt1. Spearman correlation coefficient was used to assess the correlation between Sirt1 expression levels and endometrial fibrosis.

At 7 d post-modeling, histological examination confirmed the successful establishment of the IUA mouse model. Immunofluorescence staining demonstrated significantly increased MSC homing to the endometrium in both the CXCL12 and UC-MSCs treatment groups versus the Sham and Model groups (P<0.001), with no significant difference between the 2 treatment groups. At 21 d post-modeling, both the CXCL12 and UC-MSCs treatment groups exhibited significantly greater endometrial thickness (P<0.001) and gland number (P<0.05) relative to the Model group. Endometrial fibrosis was significantly reduced in the treatment groups compared with the Model group (P<0.001). Microvessel density (MVD) was significantly higher in the CXCL12 and UC-MSCs groups than in the Model group (P<0.05). Immunohistochemical analysis revealed significantly elevated expression of VEGF (P<0.001), Ki-67 (P<0.001), and Sirt1 (P<0.001) in both treatment groups versus the Model group. After controlling for group allocation, Sirt1 expression showed a significant negative correlation with endometrial fibrosis area (r_s=-0.73, P<0.05, t=-3.86).

Through enhancing MSC homing to the endometrium in IUA mice, CXCL12 exerts therapeutic effects comparable to UC-MSCs therapy, effectively promoting endometrial regeneration and reducing stromal fibrosis, with upregulated Sirt1 expression potentially mediating these effects.