Exploring the mechanism of action of aspirin in improving endometrial receptivity in PCOS rats based on uterine lavage fluid metabolomics

Abstract

Backgrounds: Aspirin has been shown to enhance endometrial receptivity (ER) during the window of implantation in patients with polycystic ovary syndrome (PCOS). However, the underlying mechanisms remain unclear. This study aimed to elucidate the mechanisms by which aspirin improves ER through metabolic analysis of uterine lavage fluid.

Methods: A PCOS rat model was established using letrozole. Body weight and estrous cycles were monitored, and the number of implanted embryos was assessed across groups. We evaluated endometrial ultrastructure, ovarian and endometrial histomorphometry. Serum levels of estradiol(E2) and progesterone(P)were measured. Moreover, through ultra-performance liquid chromatography-mass spectrometry, the study of uterine lavage fluid metabolites revealed the potential mechanism of action of aspirin.

Results: Compared with the model group, aspirin treatment significantly increased embryo implantation rates, improved endometrial morphology and hormone levels. Metabolomic analysis identified 48 differential metabolites, among which five-2, 6-dihydroxypurine, gluconolactone, Oxaceprol, PC (18:1/18:1), and PC (20:3e/17:1)-were identified as potential biomarkers for aspirin-mediated improvement of ER in PCOS rats. Pathway analysis revealed that aspirin primarily modulates the pentose phosphate pathway, arginine and proline metabolism, and glycerophospholipid metabolism.

Conclusions: Aspirin may enhance glucose metabolism, alleviate insulin resistance, promote angiogenesis, and improve vascular permeability and endometrial receptivity. These effects are likely mediated through the regulation of biomarkers involved in the pentose phosphate pathway, arginine and proline metabolism, and glycerophospholipid pathways in uterine lavage fluid.

Fig 1. Changes in the estrous cycle of PCOS rats (Wright’s staining, scare bar = 100 μm).

(a–d) Vaginal smears corresponding to different phases of the estrous cycle: (a) Proestrus, (P) was dominated by expanded oval-shaped nucleated epithelial cells (black arrow); (b) Estrus (E), estrus was dominated by irregularly shaped anucleate keratinized cells(red arrow); (c) Metestrus (M), metestrus was characterized by leukocytes(green arrow), nucleated epithelial cells (black arrow) and keratinized cells(red arrow) in comparable quantities; (d) Diestrus (D), diestrus was mainly dominated by leukocytes(green arrow).

Fig 2. Morphological changes in ovarian tissue of PCOS rats.

(a, c) Control group, the oocyte structure was intact and clear, the granulosa cells were multilayered, tightly and neatly arranged, and multiple corpus luteum and follicles at all levels were visible; (b, d) Model group, a large number of cystic follicles were visible in the ovary, the number of oocytes and corpus luteum was significantly reduced, and the number of granulosa cell layers was reduced and loosely arranged. The lower panels (HE staining, scare bar = 200μm) show magnified views of the boxed regions in the upper panels (HE staining, scare bar = 500μm).

Fig 3. Effect of aspirin on the number of blastocysts implanted in PCOS rats during the window of implantation.

(n = 6). **P ＜ 0.01 compared with the control group, and ^#P ＜ 0.05compared with the model group.

Fig 4. Effect of aspirin on serum E₂ and P levels in PCOS rats during the window of implantation.

(n = 4). (a) Serum E₂ level; (b) Serum P level. **P ＜ 0.01 compared with the control group, and ^##P ＜ 0.01 compared with the model group.

Fig 5. Effect of aspirin on endometrial histomorphometry in PCOS rats during the window of implantation.

(n = 5). (HE staining, scare bar = 200μm). (a) Control group; (b) Model group; (c) Aspirin group.

Fig 6. Effect of aspirin on endometrial histomorphometry in PCOS rats during the window of implantation.

(n = 5). (HE staining, scare bar = 50μm). (a) Control group, the endometrial stroma is loose, with synchronous development of glands and stroma, abundant glands, and a rich vascular supply; (b) Model group, endometrial glandular hypoplasia is observed, characterized by small glandular lumina and a reduced number of glands, along with a decrease in vascularity; (c) Aspirin group, endometrial development shows improvement, manifested by larger glandular lumina, loose stroma, and an increased number of glands and vessels within the stroma. Black arrows indicate glands, and red arrows indicate blood vessels.

Fig 7. Effect of aspirin on endometrial histomorphology in PCOS rats during the window of implantation.

(n = 5). (a) Endometrial thickness; (b) Number of endometrial glands; (c) Number of endometrial blood vessels. **P ＜ 0.01 compared with the control group, and ^##P ＜ 0.01 compared with the model group.

Fig 8. Effect of aspirin on the expression of pinopodes in PCOS rats during the window of implantation.

(n = 3). (a, d) Control group, the endometrial architecture appeared normal and intact, exhibiting a dense and uniform distribution of microvilli and numerous pinopodes were observed, evenly distributed across the surface; (b, e) Model group, the endometrial structure showed significant disruption, characterized by sparse and irregularly distributed microvilli and the number of pinopodes was markedly reduced, with the majority exhibiting pronounced wrinkling and collapse; (c, f) Aspirin group, the endometrial structure was slightly damaged, with abundant and evenly distributed microvilli and a large number of pinopodes, but some of them were slightly wrinkled. The lower panels (scare bar = 10μm) magnify the boxed regions in the upper panels (scare bar = 50μm). Red arrows indicate pinopodes.

Fig 9. Multivariate statistical analysis diagram.

(a, b) PLS-DA diagrams. The separation pattern was satisfactory, with significant differences in principal components among the groups, indicating that the metabolic profiles underwent significant alterations between different groups. The ellipses represent the 95% confidence intervals. (c, d) 200 × permutation test of 7-fold cross-validation. The PLS-DA model was validated by a 200 × permutation test, and the results were R²Y = 0.90, Q² = -0.69 < 0 and R²Y = 0.93, Q² = -0.66 < 0. The model is well established, there is no over-fitting. (e, f) Volcano plots. The volcano plots showed the changing trend of the differential metabolites expression in the Model group vs. Control group and Aspirin group vs. Model group. Each dot represents a specific metabolite, and the size of the dot indicates the VIP value. Red dots represent upregulated metabolites; green dots represent down-regulated metabolites. (g, h) Heat maps of metabolites. In the form of a heat map, shows a full range of metabolites in uterine lavage fluid samples from the Model group vs. Control group and Aspirin group vs. Model group and the relative expression magnitude of metabolites in each sample group. Red indicates increased biomarker expression; blue indicates decreased biomarker expression. (i) Venn diagram of metabolites. Five common cross metabolites were screened by intersecting the metabolites in the Model group vs. Control group and Aspirin group vs. Model group. (j, k) Metabolite pathways. Large size and red color represent major pathway enrichment and high pathway effect values, respectively. Statistically significant pathways (P < 0.05) have been labelled. (j) Model group vs. Control group; (k) Aspirin group vs. Model group.