Obesity Reduces Endometrial Receptivity by Downregulating the Ob-Rb/STAT-3 Signaling Pathway in Women and Female Mice.

This study enrolled 79 women with normal weight (n = 32), overweight (n = 32), and obesity (n = 15). No significant intergroup differences in age were observed ( Fig. 1A ). Mean BMI values were 21.86 ± 2.11 kg/m² in the normal-weight group, 27.76 ± 1.37 kg/m² in the overweight group, and 31.10 ± 2.20 kg/m²in the obese group, demonstrating distinct anthropometric stratification ( Fig 1B ). Serum Ln(leptin) levels were significantly elevated in the overweight (2353.02 ± 189.82 pg/mL) and obese (2392.56 ± 170.52 pg/mL) groups compared to normal-weight controls (2002.99 ± 125.78 pg/mL, Fig. 1C , Table 1 ). Table 1 displays the characteristics of patients in the current study (control group: n = 32, overweight group: n = 32, obese group: n = 15). The BMI, leptin level, Ln(Leptin), and fasting insulin levels were significantly higher in women with overweight and obesity compared with the normal weight group. Women with overweight (37.50%) and obesity (53.33%) had a higher percentage of leptin resistance. No significant inflammatory factors [C-reactive protein, erythrocyte sedimentation rate (ESR), and white blood cell (WBC)] between the 3 groups were found. Leptin and fasting insulin levels were significantly higher in women with obesity. There is a trend that the implantation rate, clinical pregnancy rate, and live birth rate are lower in women with overweight and obesity. A further linear regression model indicated overweight (β = 350.03, 95% confidence interval: 268.97-431.10) and obesity (β = 389.58, 95% confidence interval: 288.11-491.04) were associated with higher Ln(Leptin) (data not shown). We also summarized 2126 patients who underwent their first frozen-thawed embryo transfer in our reproductive medicine center in 2019 (aged ≤ 35 years with high-quality embryos). Compared to the normal-weight group, patients with obesity had significantly longer duration of infertility and higher WBC count, ESR, and fasting insulin levels. However, their clinical pregnancy and live birth rates were significantly lower ( Table 2 ). In the regression model, after adjusting for duration of infertility, WBC, ESR, and fasting insulin, the clinical pregnancy rate in the obese group remained significantly lower ( Table 3 ). This suggests that obesity may exert a detrimental effect on endometrial receptivity independent of metabolic and inflammatory factors. Leptin receptor (Ob-Rb) and pSTAT-3 in secretory endometrium by body weight. (A) Comparison of age among the control group (n = 32), overweight group (n = 32), and obese group (n = 15). (B) Comparison of BMI among the control group (n = 32), overweight group (n = 32), and obese group (n = 15). **** P < .0001. (C) Serum Ln(leptin) levels in women of the control group (n = 32), overweight group (n = 32), and obese group (n = 15). **** P < .0001. (D) The expressions of Ob-Rb and pSTAT-3 immumohistochemical staining in secretory endometrium of women in the control group, overweight group, and obese group. (Scale bar, 200 μm.) (E) Comparison of Ob-Rb expressions in secretory phase endometrium among women in the control group, overweight group, and obese group. * P < .05 (F) Comparison of pSTAT-3 expressions in secretory phase endometrium among women in the control group, overweight group, and obese group. ** P < .01. Abbreviations: BMI, body mass index; ns, not significant. The characteristics of study participants Bold italics of P value indicates P .05. Abbreviations: BMI, body mass index; CPR, C-reactive protein; E2, estradiol; ESR, erythrocyte sedimentation rate; P, progesterone; WBC, white blood cell. a Leptin resistance defined as more than 75th percentile of the study participants' leptin value. Clinical characteristics and pregnancy outcomes of participants who underwent the first embryo transfer with high-quality embryo by weight group, 2019 Bold italics of P value indicates P .05. Abbreviations: BMI, body mass index; ESR, erythrocyte sedimentation rate; WBC, white blood cell. The associations of obesity and pregnancy outcomes for patients undergoing embryo transfer, 2019 Bold figure indicates P < .05. Abbreviations: CI, confidence interval; OR, odds ratio. The expression of Ob-Rb decreased in succession from the normal group (33.75 ± 2.22%) to the overweight (20.87 ± 2.22%) and obese groups (6.50 ± 7.05%) detected by immunohistochemical staining that was observed as a specific brown-yellow stain primarily located in the cytoplasm ( Fig. 1D and 1E ). The expression of pSTAT-3 in the midsecretory endometrium also decreased successively in normal-weight women (23.75 ± 2.21%), overweight women (13.87 ± 2.94%), and women with obesity (10.83 ± 2.02%) ( Fig. 1D and 1F ) and significantly differed between normal-weight and overweight/obese groups. Further Western blotting indicated the expressions of endometrial receptivity factors LIF and VEGF were downregulated in overweight and obese women compared to the normal-weight controls, with a more pronounced and statistically significant reduction in the obese group ( Fig. 2A-2C ). Differential expression of LIF and VEGF in secretory endometrium across body weight groups. (A) Representative Western blot analysis of LIF and VEGF protein levels. (B) Quantitative analysis of LIF expression. *** P < .001. (C) Quantitative analysis of VEGF expression. ** P < .01. Abbreviations: LIF, leukemia inhibitory factor; VEGF, vascular endothelial growth factor. An obese mouse model was established using a classic DIO method ( Fig. 3A ) with a success rate of 60%. The weights of the mice differed among the following groups, with the obese group (n = 24) being the heaviest at 30.07 ± 4.88 g, followed by the leptin resensitization group (n = 24) at 24.02 ± 2.65 g, the normal-weight group (n = 20) at 23.70 ± 2.74 g, and the Ob-Rb inhibitor group (n = 20) at 21.59 ± 1.55 g ( Fig. 3B ). Serum leptin levels in the 4 mice groups are shown in Fig. 3C with the obese leptin-resistant group at 14.97 ± 10.70 ng/mL, leptin resensitization group at 1.37 ± 0.70 ng/mL, normal-weight group at 1.23 ± 0.75 ng/mL, and Ob-Rb inhibitor group at 0.61 ± 0.55 ng/mL. Weight, leptin, and fertility outcomes in mouse endometrial treatment groups. (A) Comparison chart of body weight in different treatment groups of mice. *** P < .001; **** P < .0001. (B) Statistical comparison chart of serum leptin levels in different treatment groups of mice. **** P < .0001. (C) Comparison chart of pregnancy rates in different treatment groups of mice. * P < .05; ** P < .01. (D) Comparison chart of the number of embryos in different treatment groups of mice. (E) Representative image of the number of embryos on the seventh day of pregnancy in different treatment groups of mice. The pregnancy rate in the normal-weight group was 90% and decreased to 50% in the obese leptin-resistant mice model, whereas it improved to 80% in those who underwent dietary weight control in the leptin resensitization group ( Table 4 and Fig. 3D ). Mice receiving the Ob-Rb inhibitor also exhibited a significantly reduced pregnancy rate of 40%, which was similar to that in obese mice ( Table 4 and Fig. 3D ). The average numbers of blastocyst in the 4 mice groups were further recorded ( Table 4 ) and are displayed in Fig. 3E and 3F; these decreased in the obese leptin-resistant group (3.35 ± 3.95) and the Ob-Rb inhibitor group (3.80 ± 4.96), whereas they increased in the leptin resensitization group (6.40 ± 4.25) and normal control group (9.00 ± 3.52). Pregnancy outcomes in the 4 groups of female mice The expression of mice Ob-Rb in secretory endometrium among the 4 mice models is shown in Fig. 4A and 4B . It was the highest (25.59 ± 1.32%) in the normal-weight group, followed by the leptin-resensitized group (24.53 ± 2.51%), and it decreased in the Ob-Rb inhibitor group (19.58 ± 3.97%) and was lowest in the obese leptin resistance group (14.06 ± 3.79%). However, the expression of Ob-Rb in the inhibitor group was not statistically lower compared with the normal-weight group, perhaps because the short-term application of the antagonist does not significantly change its expression but inhibits its function. Leptin receptor and pSTAT-3 in mouse endometrial treatment groups. (A) Immunohistochemical analysis images showing the expression levels of leptin receptors and pSTAT-3 in the endometrium of mice in different treatment groups. (Scale bar, 200 μm.) (B) Quantification analysis of leptin receptor expression levels in the endometrial tissue of mice in different treatment groups. * P < .05; ** P < .01. (C) Quantification analysis of pSTAT-3 expression levels in the endometrial tissue of mice in different treatment groups. ** P < .01; *** P < .001; **** P < .0001. The expression of pSTAT-3 in the endometrium of mice in the normal-weight group was the highest (24.92 ± 1.74%), whereas it was the lowest in the leptin resistance group (11.06 ± 1.34%). The expression of pSTAT-3 in the leptin-resensitized group increased to 20.87 ± 1.51%, and it decreased to 12.70 ± 1.94% in the Ob-Rb inhibitor group ( Fig. 4A and 4C ). The expressions of LIF and VEGF in the obesity leptin resistance group decreased compared with those in the normal-weight group ( Fig. 5A-5C ). After diet restriction, endometrial receptivity was increased in the resensitized group as suggested by the increased expression of LIF and VEGF ( Fig. 5A-5C ). The expressions of LIF and VEGF protein in the Ob-Rb inhibitor group were decreased compared with those in the normal-weight and leptin-resensitized groups ( Fig. 5A-5C ). Effects of experimental treatments on LIF and VEGF expression in mouse endometrium. (A) Representative Western blots of LIF and VEGF protein levels. (B) Quantification analysis of LIF expression. * P < .05; ** P < .01. (C) Quantification analysis of VEGF expression. * P < .05. Abbreviations: LIF, leukemia inhibitory factor; VEGF, vascular endothelial growth factor.

For the first time, this study indicates that obesity reduces endometrial receptivity by downregulating the Ob-Rb/STAT-3 signaling pathway and contributes to the current knowledge of how obesity affects endometrial receptivity. Endometrial receptivity is impaired in women with obesity [ 9 ]. Although higher ovulation rates and high-quality embryos can be obtained clinically through ovulation induction and in vitro fertilization therapy in women with obesity, the rates of embryo implantation, clinical pregnancy, and live birth are still lower than those in women with normal weight [ 26 , 27 ]. Obesity-associated endometrial infertility has become an intractable clinical problem. In the current study, we verified the association between obesity, elevated serum leptin levels, deceased Ob-Rb/STAT-3 signaling, and impaired endometrial receptivity. We confirmed that the serum leptin levels were higher in women with overweight and obesity. However, in our study, the serum leptin concentrations of women with overweight and obesity were somewhat lower than that reported in the literature [ 28 ]. It is possible that regional, cohort-specific variations, the different menstruation phase of leptin measured [ 29 ], or the previous infertility treatment may contribute to these differences. STAT-3 plays a role in multiple aspects of the body, including the entire pregnancy process. The activation of STAT-3 can regulate the expression of cell cycle protein CyclinD1 and plays an important regulatory role in cell growth, proliferation, transformation, and lipid metabolism [ 30 ]. Mouse experiments have demonstrated that STAT-3 and its downstream regulatory pathways will control the reorganization of uterine epithelial junctions, stromal proliferation, and differentiation, all of which are key determinants for successful embryo implantation [ 30 ]. Also, STAT-3 has been suggested to be associated with endometrial stromal cell decidualization, angiogenesis, and vascular remodeling in our previous study [ 31 ]. The Ob/STAT-3 signaling pathway was first examined in the endometrium, and we proved that it serves as a crucial molecular bridge between obesity and endometrial dysfunction. We further established a DIO mouse model, an Ob-Rb inhibitor mouse model, and a leptin resensitization mouse model to provide a mechanistic framework of how obesity disrupts endometrial receptivity and to test potential therapies. The DIO hyperleptinemia mouse model was successfully built by validation of leptin resistance (serum leptin 2-to 3-fold higher in obese mice) to confirm leptin resistance as a cause of endometrial dysfunction. The leptin resensitization mouse model by caloric restriction [ 20 , 24 ], which mimicked clinic weight loss interventions, demonstrated that weight loss alone can restore Ob-Rb/STAT-3 signaling and endometrial function. For further validation, we used Ob-Rb inhibitor Allo-aca, a selective Ob-Rb antagonist, to precisely target Ob-Rb and test if it is associated with STAT-3 signaling in mice. These mice models combined reveal the pivotal role of Ob/STAT-3 signaling in the mechanism of obesity-associated decreased endometrial receptivity and offer new perspectives for therapeutic interventions. Strategies aimed at restoring leptin sensitivity or bypassing Ob-Rb signaling defects may improve reproductive outcomes in obese women. Controlling BMI (ie, reducing body weight) is an effective and noninvasive method to improve endometrial receptivity in patients with obesity [ 32 , 33 ]. Research suggested that a 5% weight loss can significantly improve serum hormonal levels such as free testosterone, LH, and insulin, as well as leptin levels [ 34 , 35 ]. In our study, we showed that restoring leptin sensitivity could improve Ob-Rb/STAT-3 signaling and finally benefit endometrial receptivity and pregnancy outcomes in obese mice. Therefore, weight-loss strategies including dietary approaches, exercise modalities, and combined interventions for women with obesity and fertility problems are recommended in clinical practice as they not only improve ovarian function and oocyte quality but also contribute to a better endometrial environment. This study has several limitations. First, we did not investigate the expression of Ob-Rb across the menstrual cycle to examine its potential specific association with the window of implantation. Previous studies have reported that leptin levels were peak at mid-cycle [ 36 ] and that the expression of Ob-Rb was highest in the early-secretory phase [ 17 ]. Further research is needed to draw the map of the expression of serum leptin and Ob-Rb within the whole menstrual cycle to confirm its association with fertility in the midsecretory phase. Second, our DIO mouse model primarily illuminates the pathophysiology driven by caloric excess and may not fully represent infertility stemming from strong genetic predispositions or other specific environmental factors in humans. Human obesity involves complex genetic, epigenetic, and metabolic factors beyond diet alone. However, the high-fat diet-induced mouse model used here is a well-established model that is widely used in the literature to explore obesity-associated hyperleptinemia and leptin resistance [ 37 ], although it cannot represent all human obesity etiologies. Future research should employ more diverse models to capture the complexity of human obesity and its associated infertility.

Leptin resistance in women with obesity-induced Ob-Rb/STAT-3 signaling dysfunction is a central mechanism for obesity-associated impaired endometrial receptivity. Therapeutic approaches including weight loss or novel fertility treatments targeting leptin resistance/Ob-Rb expression that restore Ob-Rb/STAT-3 signaling may offer new hope for improving fertility in women with obesity.

A total of 32 women with overweight [25 kg/m 2 ≤ body mass index (BMI) < 30 kg/m 2 ] and 15 women with obesity (BMI ≥ 30 kg/m 2 ) of childbearing age (aged 20-40 years) were included from August 2021 to December 2022 at the Reproductive Medicine Center of The First Affiliated Hospital of Anhui Medical University. These women were nulligravid and diagnosed with female infertility. Meanwhile, 32 women with normal weight and male factor infertility (18.5 kg/m 2 ≤ BMI < 25 kg/m 2 ) were included as controls. All the participants were prepared for in vitro fertilization. To minimize variation from physiological fluctuations in hormone receptor expression, all endometrial samples were specifically obtained during the midsecretory phase (LH +5-7 [ 22 ], days 20-23), thus ensuring analysis at a comparable and critical functional stage (the window of implantation). Midsecretory endometrial samples were collected using pipe-suction curettage (LILYCLEANER, Shanghai, China), and fasting blood samples were collected. Patients diagnosed with inflammatory diseases, autoimmune diseases, or other conditions that significantly impact endometrial receptivity were excluded from the study. A thorough endometrial ultrasonographic examination was performed to determine the endometrial thickness and to exclude any intrauterine abnormalities within the uterine cavity before the endometrial biopsy including endometrial polyps, hydrosalpinx, uterine malformations, submucosal fibroids, uterine adhesions, and other factors that may interfere embryo implantation. Written informed consent was obtained from all the participants. The study was approved by the Medical Ethics Committee of The First Affiliated Hospital of Anhui Medical University (5101367). Four-week-old female C57BL/6J mice were purchased from the Experimental Animal Center of Anhui Medical University. After 1 week of acclimatization, female mice were randomly divided into a high-fat diet group (80 mice) [diet-induced obesity (DIO) mouse model] and a normal diet group (40 mice). All animals were kept in a specific pathogen-free environment with an ambient temperature of around 25 °C and a humidity of around 50%, alternating between light and darkness daily, and received adequate food and water. All animal experiments were performed in accordance with the requirements of the Experimental Animal Ethics of Anhui Medical University (20211255). DIO is likely the most frequent animal model used to investigate leptin resistance [ 23 ] with feeding a high-fat diet (60% kcal/fat, Xietong Pharmaceutical Biotechnology Co., Ltd., Jiangsu, China), whereas the normal diet group received standard feed. The DIO model was successfully constructed when the body weight of the mice was 20% higher than that of the mice in the normal feeding group. The leptin resensitization model was constructed by dietary restriction as previously described [ 20 , 24 ]. Briefly, we performed 7 weeks of dietary restriction on DIO mice with fasting and switched to low-fat diets (10% kcal/fat, Xietong Pharmaceutical Biotechnology Co., Ltd.), reducing weight to normal levels; this served as the leptin resensitization group. Allo-aca (MedChemExpress, New Jersey, USA), a leptin peptide mimic, is a potent and specific Ob-Rb antagonist that can block leptin signaling and action in various in vitro and in vivo models [ 25 ]. Normal-weight mice were intraperitoneally injected with Ob-Rb inhibitor Allo-aca at 0.1 mg/kg body weight once a day for 10 days and served as Ob-Rb inhibitor mouse models; age-matched controls were administered equivalent volumes of sterile saline (0.9% NaCl) following the same regimen. Female mice were cohoused with fertile male mice at a 2:1 female-to-male ratio. Vaginal plug checks were conducted every morning, with the presence of a plug designated as gestational day (GD) 0.5. For GD5 processing, females underwent terminal blood collection via ocular exsanguination before euthanasia. Excised uterine horns were transversely divided into 3 segments. Middle segments were Paraformaldehyde-fixed (4%, 24 hours, 4 °C) for immunohistochemistry; adjacent segments were flash-frozen (LN₂) and stored at −80 °C for proteomic assays. For GD7 assessment, pregnant dams were euthanized following blood collection. Uteri were imaged ex vivo for pregnancy parameter quantification (maternal weight, pregnancy rate, blastocyst enumeration). Following blood collection from human participants and experimental animals, samples were processed identically. Whole blood underwent clotting at room temperature for 2 hours, followed by incubation at 4 °C for 1 hour. Serum was isolated by centrifugation at 3000 × g for 10 minutes, aliquoted, and stored at −80 °C until analysis. Serum leptin concentrations were quantified using a species-specific human (RRID: AB_3665146, https://scicrunch.org/resolver/AB_3665146 ) and mouse Leptin (RRID:AB_3717948, https://scicrunch.org/resolver/AB_3717948 ) ELISA Kit according to the manufacturer's protocol. Fixed uterine tissues were dehydrated in graded ethanols (50-100%), cleared in xylene, and paraffin embedded. Serial 5 μm sections were mounted on poly-L-lysine-coated slides. After drying at 60 °C for 1 hour, sections were deparaffinized in xylene (1 hour) and rehydrated through descending ethanol concentrations. Antigen retrieval was performed in 0.01 M citrate buffer (pH 6.0) using microwave irradiation (100 °C, 20 minutes) followed by 30 minutes of cooling. Sections were permeabilized with 3% Triton X-100 in PBS (pH 7.4) for 30 minutes prior to blocking. Immunohistochemical experiments were conducted according to the instructions of the pv-9001 polink-2plus® polymer horseradish peroxidase (HRP) kit (ZSGBBio, Beijing, China). Endogenous peroxidase activity was blocked with a solution containing 3% hydrogen peroxide for 20 minutes. The sections were then rinsed thrice in PBS (pH 7.2-7.4) for 5 minutes. Subsequently, the sections were immersed in 0.01 M citric acid buffer (pH 6.0), and the temperature was maintained at 90 to 100 °C using a microwave oven for 20 minutes. The sections were then rinsed again with PBS for 5 minutes. Following this, the sections were blocked with normal goat serum for 20 minutes at room temperature before incubating with primary antibodies: leptin receptor (RRID:AB_10901772, https://scicrunch.org/resolver/AB_10901772 ) and pSTAT-3 (RRID:AB_3713209, https://scicrunch.org/resolver/AB_3713209 ) overnight at 4 °C. The following day, sections were incubated at room temperature for 30 minutes with a Polymer helper. After washing thrice with PBS, the sections were treated with a secondary anti-rabbit IgG-HRP antibody for 30 minutes. The sections were then developed with diaminobenzidine and nuclear staining with hematoxylin. They were immersed in double-distilled water for 5 minutes, dehydrated in an alcohol gradient, and mounted with a neutral resin. An IgG-negative control was used for each antibody. Section imaging was performed using an upright fluorescence microscope (Olympus BX53, Tokyo, Japan). Three individuals per group were randomly selected, with 1 representative field of view imaged per tissue section. Positive staining intensity was quantified blindly using ImageJ software (National Institutes of Health, Bethesda, MD, USA). Tissues were lysed in radio-immunoprecipitation assay buffer (Beyotime, Shanghai, China) supplemented with 1 mM phenylmethanesulfonylfluoride (Beyotime) and phosphatase inhibitor cocktail to prevent protein degradation and dephosphorylation. The concentrations of the extracted proteins were quantified using a BCA Protein Assay Kit (Beyotime). Subsequently, approximately 30 μg of total protein from each sample was resolved by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and then transferred onto a polyvinylidene difluoride membrane (Millipore, Billerica, MA, USA). Following transfer, nonspecific binding sites on the membrane were blocked for 2 hours at room temperature using 5% skim milk in Tris-buffered saline containing 0.1%Tween 20 (TBST). The membranes were incubated overnight at 4 °C with primary antibodies against leukemia inhibitory factor (LIF; RRID: AB_3083551, https://scicrunch.org/resolver/AB_3083551 ), vascular endothelial growth factor (VEGF; RRID: AB_2212642, https://scicrunch.org/resolver/AB_2212642 ), and glyceraldehyde-3-phosphate dehydrogenase (RRID: AB_2862549, https://scicrunch.org/resolver/AB_2862549 ). After washing with TBST, the membranes were incubated with HRP-conjugated secondary antibodies with goat anti-rabbit (RRID: AB_3712952, https://scicrunch.org/resolver/AB_3712952 ) at room temperature for 1 hour. After washing again with TBST, immunoblots were developed using an enhanced chemiluminescence detection system (Thermo Scientific), and images were captured. Protein band intensities were quantified using the Scion Image software (Scion Corporation, Bethesda, MD, USA). All analyses were performed using STATA software, version 14.2 (Stata Corp., College Station, TX, USA) and GraphPad Prism. Human serum leptin levels were natural logarithm (ln)-transformed prior to statistical analysis to normalize their distribution. Data are presented as mean ± SD. Comparisons between 2 groups were assessed using an unpaired Student's t -test. Multiple group comparisons were evaluated by 1-way ANOVA followed by Bonferroni post hoc correction, with statistical significance defined as P < .05.