Impact of obesity on proteomic profiles of follicular fluid-derived small extracellular vesicles: A comparison between PCOS and non-PCOS women.

This study was conducted following approval from the Ethics Committee of the First Affiliated Hospital of Sun Yat-sen University (Ethical Approval NO. [2021]507). The cohort comprised women undergoing IVF or intracytoplasmic sperm injection (ICSI) at our center from November 2021 through November 2022. Participants were categorized into two primary groups: women diagnosed with PCOS and a control group of women with normal ovarian function, both groups matched for age and body mass index (BMI). BMI was calculated as weight divided by height squared (kg/m2). According to the Rotterdam criteria (2003), the diagnosis of PCOS was confirmed in 9 participants, while the control group included 9 subjects undergoing IVF/ICSI for tubal or male factor infertility [ 22 ]. Each primary group was subsequently divided into subcategories based on BMI: lean (BMI < 24 kg/m²) and overweight/obese (BMI ≥ 24 kg/m²), resulting in four groups: PCOS lean (PCOS-Ln), PCOS overweight/obese (PCOS-OW), control lean (CTRL-Ln), and control overweight/obese (CTRL-OW). The BMI cutoff of 24 kg/m² was applied following the World Health Organization (WHO) recommendations and Chinese national standard [ 23 ]. Exclusion criteria included the presence of uterine fibroids, endometriosis, any form of cancer, premature ovarian insufficiency (POI), and other medical conditions potentially affecting follicular development, including endocrine or genetic disorders. The sample size for this study was based on previous proteomic studies and our experience with analyzing small extracellular vesicles [ 24 – 26 ]. To ensure the validity and reliability of the results, we utilized Data-Independent Acquisition (DIA) technology and employed appropriate statistical corrections. Comprehensive clinical data were collected, including height, weight, menstrual cycle characteristics, and hormone levels such as anti-Müllerian hormone (AMH), follicle-stimulating hormone (FSH), luteinizing hormone (LH), estradiol (E2), and testosterone (T). We also recorded IVF/ICSI outcomes, specifically the total number of oocytes retrieved, the number of fertilized oocytes (2PN), transferrable embryos, and instances of good-quality embryos. All patients received a gonadotropin-releasing hormone (GnRH) antagonist protocol for controlled ovarian stimulation. Following the administration of 10,000 IU of human chorionic gonadotropin (hCG) after 34–36 h, follicular fluid (FF) from follicles larger than 18 mm in diameter was collected by transvaginal ultrasound-guided aspiration. To minimize contamination from blood-derived small extracellular vesicles, only the FF from the first aspirated follicle on each side was collected. The samples were placed in sterile centrifuge tubes and centrifuged at 4 °C, 3000 g (relative centrifugal force) for 25 min to remove cellular components. The supernatant was collected and filtered through a 0.22 μm filter to remove any remaining impurities and subsequently stored at -80 °C for future analysis. Standardized laboratory protocols for conventional IVF and ICSI, as outlined by Magli et al., were employed, with ICSI performed exclusively for male factor infertility in accordance with European Society of Human Reproduction and Embryology (ESHRE) guidelines [ 27 , 28 ]. Embryo quality was assessed on Day 3 based on the modified Istanbul consensus, defining good-quality embryos as those with 7–9 cells, 0–20% cytoplasmic fragmentation, and uniform size [ 29 ]. Small extracellular vesicles (sEVs) from FF were isolated by ultracentrifugation. Specifically, 2 ml of frozen FF supernatant, stored at -80 °C, was appropriately diluted and centrifuged at 10,000 g for 20 min at 4 °C to remove larger extracellular vesicular components. The supernatant was then transferred to a new ultracentrifuge tube and subjected to ultracentrifugation at 100,000 g for 90 min at 4°. The supernatant was discarded, and the sEVs pellet was resuspended in 100 µl of sterile phosphate-buffered saline (PBS) and stored at -80 °C for further analysis. A total of 20 µL of sEVs suspension (5 µg/µL) was fixed on a continuous copper grid and negatively stained with a 2% uranyl acetate solution for 1 min, followed by air drying. The samples were observed and imaged using a FEI Tecnai G2 Spirit transmission electron microscope (FEITM) operated at an acceleration voltage of 120 kV. Nanoparticle tracking analysis (NTA) measurements were conducted using a ZetaView instrument. Prior to measurement, the instrument was calibrated using 110 nm polystyrene particles. The samples were diluted with PBS and injected into the sample chamber for analysis. The obtained data were analyzed using ZetaView 8.04.02 SP2 software to determine the size distribution and raw concentration of sEVs. FF-sEVs samples were lysed with RIPA buffer (Beyotime, Shanghai, China) with phosphatase and protease inhibitors, and the lysate was collected and heated with loading buffer (Beyotime, Shanghai, China). After protein denaturation, samples were separated on a 10% SDS-PAGE gel and transferred onto a PVDF membrane (Millipore, Billerica, MA, USA). The membrane was then incubated overnight at 4 °C with primary antibodies against CD9, CD63, TSG101, EMILIN1, LBP, F12, and S100A8 (Proteintech, Wuhan, China), followed by incubation with secondary antibodies. The images were scanned and quantified by densitometric analysis by iBright Imaging System (Thermofisher Scientific, Waltham, MA, USA). The protein concentration of FF-sEVs was determined using the BCA kit (Beyotime, Shanghai, China). For trypsin digestion, the samples were transferred to ultrafiltration tubes and centrifuged at 14000 g for 15 min to remove the permeate. The post-ultrafiltration samples were then equilibrated with DB buffer. The recovered supernatant was treated with DTT reagent and incubated at 56 °C for 1 h, followed by IAM reaction at room temperature for 1 h. For DIA analysis, gradient elution was performed using mobile phase A (0.1% FA in 100% water) and mobile phase B (0.1% FA in 80% ACN). A total of 4 µg of sample was mixed with 0.8 µl of iRT reagent, and half of the mixture was subjected to analysis. The analysis was conducted using an EASY-nLC 1200 nano-UHPLC system in DIA mode, with a spray voltage of 2.1 kV and an ion transfer tube temperature of 320 °C. The scan range was set to m/z (350–1500). The MS1 resolution was set to 60000 (at 200 m/z), and the MS2 resolution was set to 30000 (at 200 m/z) to generate raw MS detection data. For protein identification and quantification, the DIA data was imported into Spectronaut software to generate a DDA library and extract ion pair chromatographic peaks. Peptide identification and quantification were achieved by matching ions and calculating peak areas. Sample retention time correction was performed using the iRT reagent added in the previous step, and the precursor ion q-value cutoff was set to 0.01. The protein quantification results were statistically analyzed using a t-test to determine if the differences were statistically significant ( p   (FC > or FC < [fold change, FC]) were defined as differentially expressed proteins (DEPs). Venn diagrams, volcano plots, and cluster heatmaps were generated using R software. The identified proteins were annotated with Gene Ontology (GO) using the UniProt-GOA database and InterProScan software [ 30 ]. Pathway annotation was performed using the Kyoto Encyclopedia of Genes and Genomes (KEGG) database [ 31 ]. DEPs were subjected to cluster heatmap analysis and enrichment analysis for GO and KEGG pathways. Enriched GO terms and pathways were ranked based on significance ( P  < 0.05). Protein-protein interactions (PPI) were obtained from the Search Tool for the Retrieval of Interacting Genes/Proteins (STRING) database and visualized using Cytoscape software [ 32 , 33 ]. Quantitative data conforming to a normal distribution were expressed as mean ± standard deviation. One-way analysis of variance (ANOVA) was used for comparing multiple groups, with statistical significance defined as p  < 0.05. Statistical comparisons among the groups were performed using Tukey’s multiple comparisons test and Chi-square tests where appropriate, with significance set at p  < 0.05. Correlation analysis was performed using the Pearson correlation coefficient, and a two-sided p  < 0.05 was considered statistically significant. All statistical analyses were performed using GraphPad Prism 8 (GraphPad Software, San Diego, CA, USA) for group comparisons and data visualization, and IBM SPSS Statistics for Windows, version 25.0 (IBM Corp., Armonk, NY, USA) for correlation analyses.

The clinical characteristics of participants diagnosed with PCOS in relation to their BMI, matched by age and BMI with established non-PCOS control group. Participants were categorized into the following groups: PCOS overweight/obese (PCOS-OW) and PCOS lean (PCOS-Ln), along with control overweight/obese (CTRL-OW) and control lean (CTRL-Ln). Table  1 presents detailed data on the differences observed among these groups, which revealed significant variations in several hormonal parameters: basal FSH levels ( p  = 0.006), basal LH levels ( p  = 0.044), the LH to FSH ratio ( p  = 0.044), basal testosterone levels ( p  = 0.034), and AMH levels ( p  = 0.007). Regarding reproductive outcomes, there were no statistically significant differences across all groups in terms of the number of oocytes retrieved and 2PN oocytes, as well as the rates of 2PN and transferrable embryos ( p  > 0.05). However, a trend towards lower good-quality embryo production was noted in the PCOS groups compared to the non-PCOS control groups, particularly among overweight/obese subjects, suggesting that obesity may exert a subtle yet detrimental impact on embryo quality, meriting further exploration in the context of IVF/ICSI outcomes. These findings enhance understanding of the clinical implications of obesity in women with PCOS and set the stage for subsequent in-depth proteomic analyses. Table 1 Clinical characteristics and IVF/ICSI laboratory outcomes in PCOS/non-PCOS patients Variable PCOS-Ln: PCOS lean ( N  = 4) PCOS-OW: PCOS Overweight/obese ( N  = 5) Ctrl-Ln: Control lean ( N  = 5) Ctrl-OW: Control Overweight/obese ( N  = 4) P  value (x 2 ) Age (years) 27.25(3.30) 28.40(2.70) 31.60(5.23) 29.25(4.11) 0.425 BMI (kg/m 2 ) 21.70(0.93) 27.20(2.10) 21.80(1.66) 26.87(2.31) 0.001 a Basal FSH (IU/L) 5.03(0.71) 5.03(1.31) 7.58(1.08) 6.00(0.84) 0.006 b Basal LH (IU/L) 10.09(6.04) 8.77(3.80) 4.24(0.95) 3.71(1.50) 0.044 c LH/FSH 2.03(1.32) 1.88(1.00) 0.56(0.12) 0.63(0.26) 0.029 c Basal testosterone (ng/mL) 0.66(0.36) 0.42(0.08) 0.27(0.05) 0.33(0.07) 0.034 d Basal estrogen (ng/mL) 55.50(6.56) 37.00(8.89) 66.36(45.26) 48.75(23.16) 0.412 PRL (ng/mL) 14.45(6.25) 21.76(19.52) 14.65(6.25) 14.23(4.09) 0.696 AMH (ng/mL) 12.10(4.52) 6.90(5.05) 3.31(1.37) 2.71(0.85) 0.007 e Duration of stimulation (days) 8.00(0.82) 9.60(1.95) 11.20(2.17) 11.75(2.63) 0.0698 Gonadotropin dosage (IU/L) 1134.0(266.40) 1968.0(636.20) 2730.0(836.20) 3013.0(1160.0) 0.017 f Fertilization method: * 0.528 (2.22)  IVF 2(50%) 2(40%) 4(80%) 3(75%)  ICSI 2(50%) 3(60%) 1(20%) 1(25%) Number of oocytes retrieved 26.25(20.66) 17.00(7.97) 10.8(5.54) 16.50(7.05) 0.289 Number of 2PN oocytes 15.75(15.69) 11.00(6.60) 7.20(2.59) 12.50(6.25) 0.541 Rate of 2PN oocytes in retrieved oocytes (%) 60.02(20.53) 63.40(17.17) 73.06(23.32) 72.61(15.78) 0.693 Number of transferrable embryos 7.00(6.98) 5.60(3.05) 4.60(4.04) 9.00(5.35) 0.586 Stage of transferrable embryos: * 0.164 (5.12)  Cleavage embryo 2(7.1%) 4(14.3%) 7(30.4%) 6(16.7%)  Blastocyst 26(92.9%) 24(85.7%) 16(69.6%) 30(83.3%) Rate of transferrable embryos in retrieved oocytes (%) 26.23(14.56) 34.13(13.08) 50.95(33.44) 50.14(16.89) 0.296 Rate of transferrable embryos in 2PN oocytes (%) 43.84(28.56) 53.30(17.03) 63.10(38.50) 67.40(12.32) 0.599 Number of good-quality embryos 4.00(3.37) 4.20(4.60) 3.00(2.55) 7.50(7.00) 0.526 Rate of good-quality embryos in retrieved oocytes (%) 25.28(22.08) 20.35(16.62) 36.31(32.34) 36.56(29.15) 0.713 Rate of good-quality embryos in 2PN oocytes (%) 42.99(34.42) 31.59(22.84) 45.43(36.84) 46.60(38.53) 0.891 Data presented as the mean (SD). When overall p  values < 0.05, data analyzed using the Tukey’s multiple comparisons test. Data* presented as as n (%) and analyzed using the Chi-square test. BMI, body mass index; FSH, follicle-stimulating hormone; LH, luteinizing hormone; PRL, prolactin; AMH, Anti-Müllerian hormone; IVF, in vitro fertilization; ICSI, intracytoplasmic sperm injection a PCOS-Ln versus PCOS-OW = 0.003, PCOS-Ln versus Ctrl-OW = 0.007, PCOS-OW versus Ctrl-Ln = 0.002, Ctrl-Ln versus Ctrl-OW = 0.005 b PCOS-Ln versus Ctrl-Ln = 0.012, PCOS-OW versus Ctrl-Ln = 0.008 c PCOS-Ln versus PCOS-OW > 0.05, PCOS-Ln versus Ctrl-Ln > 0.05, PCOS-Ln versus Ctrl-OW > 0.05, PCOS-OW versus Ctrl-Ln > 0.05, PCOS-OW versus Ctrl-OW > 0.05, Ctrl-Ln versus Ctrl-OW > 0.05 d PCOS-Ln versus Ctrl-Ln = 0.028 e PCOS-Ln versus Ctrl-Ln = 0.011, PCOS-Ln versus Ctrl-OW = 0.010 f PCOS-Ln versus Ctrl-Ln = 0.040, PCOS-Ln versus Ctrl-OW = 0.021 Clinical characteristics and IVF/ICSI laboratory outcomes in PCOS/non-PCOS patients Data presented as the mean (SD). When overall p  values < 0.05, data analyzed using the Tukey’s multiple comparisons test. Data* presented as as n (%) and analyzed using the Chi-square test. BMI, body mass index; FSH, follicle-stimulating hormone; LH, luteinizing hormone; PRL, prolactin; AMH, Anti-Müllerian hormone; IVF, in vitro fertilization; ICSI, intracytoplasmic sperm injection a PCOS-Ln versus PCOS-OW = 0.003, PCOS-Ln versus Ctrl-OW = 0.007, PCOS-OW versus Ctrl-Ln = 0.002, Ctrl-Ln versus Ctrl-OW = 0.005 b PCOS-Ln versus Ctrl-Ln = 0.012, PCOS-OW versus Ctrl-Ln = 0.008 c PCOS-Ln versus PCOS-OW > 0.05, PCOS-Ln versus Ctrl-Ln > 0.05, PCOS-Ln versus Ctrl-OW > 0.05, PCOS-OW versus Ctrl-Ln > 0.05, PCOS-OW versus Ctrl-OW > 0.05, Ctrl-Ln versus Ctrl-OW > 0.05 d PCOS-Ln versus Ctrl-Ln = 0.028 e PCOS-Ln versus Ctrl-Ln = 0.011, PCOS-Ln versus Ctrl-OW = 0.010 f PCOS-Ln versus Ctrl-Ln = 0.040, PCOS-Ln versus Ctrl-OW = 0.021 To comprehensively characterize the FF-sEVs, the morphology was assessed using transmission electron microscopy (TEM), revealing a predominance of round to oval vesicles (Fig.  1 A). Subsequent nanoparticle tracking analysis (NTA) confirmed that the diameters of FF-sEVs across all study groups ranged from 100 to 150 nm, adhering to the established dimensions for small extracellular vesicles (Fig.  1 B). The mean concentration of FF-sEVs in FF ranged from approximately 1.3 × 10⁹ to 2.5 × 10⁹ particles/mL in the PCOS group, and from 1.2 × 10⁹ to 1.5 × 10⁹ particles/mL in the control group (Fig.  1 B). The purity of the isolated FF-sEVs was confirmed through Western blot analysis, which detected the sEVs markers CD9, CD63, and TSG101 (Fig.  1 C). Notably, the protein concentration of sEVs in the FF of overweight and obese PCOS patients was significantly elevated compared to that of lean PCOS patients ( p  = 0.010, Fig.  1 D). In contrast, no significant difference in sEV protein concentration was observed between overweight/obese and lean women in the control cohort ( p  > 0.05, Fig.  1 D). Furthermore, Pearson correlation analysis demonstrated a significant positive correlation between sEV protein concentration and BMI within the PCOS group ( p  = 0.006), whereas the control group exhibited an increasing trend without statistical significance ( p  > 0.05, Fig.  1 E). Fig. 1 Characterization and BMI-related protein concentration of follicular fluid-derived small extracellular vesicles. ( A ) Transmission electron microscopy (TEM) of follicular fluid-derived small extracellular vesicles (FF-sEVs) among four groups: PCOS-OW, PCOS-Ln, CTRL-OW, and CTRL-Ln groups. The magnification of the TEM images was 75,000x, with a scale bar of 200 nm. ( B ) Nanoparticle tracking analysis (NTA) of FF-sEVs among four groups using a ZetaView instrument. The size distribution and raw concentration of sEVs were determined by ZetaView 8.04.02 SP2 software. ( C ) Western blot analysis of FF-sEVs among four groups. ( D ) FF-sEVs protein concentration among four groups. ( E ) Correlation between sEVs protein concentration in FF and BMI. * p  < 0.05, ** p  < 0.01 Characterization and BMI-related protein concentration of follicular fluid-derived small extracellular vesicles. ( A ) Transmission electron microscopy (TEM) of follicular fluid-derived small extracellular vesicles (FF-sEVs) among four groups: PCOS-OW, PCOS-Ln, CTRL-OW, and CTRL-Ln groups. The magnification of the TEM images was 75,000x, with a scale bar of 200 nm. ( B ) Nanoparticle tracking analysis (NTA) of FF-sEVs among four groups using a ZetaView instrument. The size distribution and raw concentration of sEVs were determined by ZetaView 8.04.02 SP2 software. ( C ) Western blot analysis of FF-sEVs among four groups. ( D ) FF-sEVs protein concentration among four groups. ( E ) Correlation between sEVs protein concentration in FF and BMI. * p  < 0.05, ** p  < 0.01 The proteomic profiles of FF-sEVs were meticulously examined using data-independent acquisition (DIA) proteomic sequencing in patients diagnosed with PCOS and in non-PCOS control women, leading to the identification of a total of 4,797 proteins. To elucidate the impact of obesity on the protein landscape, differential expression analyses were performed among PCOS overweight/obese versus lean groups, and control overweight/obese versus lean groups. These analyses revealed 180 significantly upregulated and 256 downregulated proteins in the PCOS-OW group compared to the PCOS-Ln group (Supplementary table S1 and S2 ). In contrast, the CTRL-OW versus CTRL-Ln comparison identified 106 significantly upregulated and 67 downregulated proteins (Supplementary table S3 and S4 ). The Venn diagrams presented in Fig.  2 A and B illustrate the unique (proteins differentially expressed only in one comparison group) and shared (proteins differentially expressed in both comparison groups) upregulated (Fig.  2 A) and downregulated (Fig.  2 B) differentially expressed proteins (DEPs) among the major groups. Notably, our findings emphasize that overweight/obesity exerts a more substantial influence on the alteration of DEPs in women with PCOS as opposed to their non-PCOS counterparts, regardless of whether these proteins were upregulated or downregulated. Differentially expressed proteins within both the PCOS and control comparison groups are visualized in volcano plots (Fig.  2 C and D). In these plots, blue dots represent proteins with a P -value of less than 0.05, green dots represent proteins with a fold change greater than 2, and red dots highlight proteins simultaneously meeting both criteria (fold change greater than 2 and P -value less than 0.05). Selected red dots are annotated with gene names to indicate the most notable DEPs. Additionally, heatmaps in Fig.  2 E and F illustrate the expression patterns of both upregulated and downregulated DEPs within the two groups. The results from both heatmap analyses indicate a substantial number of differentially expressed proteins, whose expression profiles effectively discriminate between the samples in each group (Fig.  2 E and F). Fig. 2 Differential proteomic analysis of follicular fluid-derived small extracellular vesicles in PCOS and non-PCOS women. ( A ) Venn diagram illustrating the unique (proteins differentially expressed only in one comparison group) and shared (proteins differentially expressed in both comparison groups) counts of upregulated DEPs identified in the comparison of PCOS-OW versus PCOS-Ln and CTRL-OW versus CTRL-Ln. ( B ) Venn diagram depicting the unique and shared counts of downregulated DEPs identified in the comparison of PCOS-OW versus PCOS-Ln and CTRL-OW versus CTRL-Ln. ( C ) Volcano plot presenting the DEPs in the PCOS-OW versus PCOS-Ln comparison. Blue dots represent proteins with a P -value of less than 0.05, green dots represent proteins with a fold change greater than 2, and red dots highlight proteins simultaneously meeting both criteria (fold change greater than 2 and P -value less than 0.05). Selected red dots are annotated with gene names to indicate the most notable DEPs. ( D ) Volcano plot illustrating the DEPs identified in the CTRL-OW versus CTRL-Ln comparison, with the same color scheme and statistical thresholds applied. ( E ) Heatmap clustering of the DEPs in the PCOS-OW versus PCOS-Ln comparison group. ( F ) Heatmap clustering of the DEPs in the CTRL-OW versus CTRL-Ln comparison group Differential proteomic analysis of follicular fluid-derived small extracellular vesicles in PCOS and non-PCOS women. ( A ) Venn diagram illustrating the unique (proteins differentially expressed only in one comparison group) and shared (proteins differentially expressed in both comparison groups) counts of upregulated DEPs identified in the comparison of PCOS-OW versus PCOS-Ln and CTRL-OW versus CTRL-Ln. ( B ) Venn diagram depicting the unique and shared counts of downregulated DEPs identified in the comparison of PCOS-OW versus PCOS-Ln and CTRL-OW versus CTRL-Ln. ( C ) Volcano plot presenting the DEPs in the PCOS-OW versus PCOS-Ln comparison. Blue dots represent proteins with a P -value of less than 0.05, green dots represent proteins with a fold change greater than 2, and red dots highlight proteins simultaneously meeting both criteria (fold change greater than 2 and P -value less than 0.05). Selected red dots are annotated with gene names to indicate the most notable DEPs. ( D ) Volcano plot illustrating the DEPs identified in the CTRL-OW versus CTRL-Ln comparison, with the same color scheme and statistical thresholds applied. ( E ) Heatmap clustering of the DEPs in the PCOS-OW versus PCOS-Ln comparison group. ( F ) Heatmap clustering of the DEPs in the CTRL-OW versus CTRL-Ln comparison group GO analysis was used to categorize the protein classification of DEPs in two comparison groups (Fig.  3 A and B). In the comparison between PCOS-OW and PCOS-Ln groups, functions associated with intracellular organelles, nucleosome assembly, and various binding activities were enriched, suggesting potential effects of obesity on intracellular structure and function (Fig.  3 A). The comparison between CTRL-OW and CTRL-Ln groups revealed enrichment of functions related to carbohydrate derivative metabolic process and fructose-bisphosphate aldolase activity, potentially indicating metabolic changes and enzyme activity regulation in obesity (Fig.  3 B). The KEGG pathway analysis of the proteomic profiles of FF-sEVs revealed distinct functional alterations and metabolic changes associated with obesity (Fig.  3 C and D). Notably, the IL-17 signaling pathway ( p  = 0.032) and protein processing in the endoplasmic reticulum ( p  = 0.040) were identified as significantly impacted in the comparison between the PCOS-OW and PCOS-Ln groups, suggesting that obesity modifies immune responses and protein processing within the FF microenvironment (Fig.  3 C). These findings indicate that obesity in women with PCOS may substantially disrupt functional pathways that are crucial for reproductive health. Conversely, the analysis of the CTRL-OW versus CTRL-Ln groups revealed alterations primarily in metabolic pathways (Fig.  3 D). The NF-kappa B signaling pathway emerged as particularly noteworthy ( p  = 0.027), implicating inflammation in metabolic regulation. Additionally, pathways related to fructose and mannose metabolism, and glycolysis/gluconeogenesis were characterized, underscoring the metabolic adaptations associated with obesity in non-PCOS women, potentially affecting their overall health and reproductive outcomes. The juxtaposition of functional changes in the PCOS group versus metabolic alterations in the non-PCOS control group elucidates the differential impact of obesity on FF-sEVs and underscores the complexity of the interplay between obesity and reproductive health. Fig. 3 Functional enrichment and pathway analysis of differentially expressed proteins in follicular fluid-derived small extracellular vesicles in PCOS and non-PCOS women. ( A ) Gene Ontology (GO) enrichment analysis of DEPs in the comparison between PCOS-OW and PCOS-Ln groups, indicating enrichment in intracellular organelles, nucleosome assembly, and binding activities. ( B ) GO enrichment analysis of DEPs in the comparison between CTRL-OW and CTRL-Ln groups, highlighting changes related to carbohydrate derivative metabolic processes and fructose-bisphosphate aldolase activity. ( C ) Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis of DEPs in the PCOS-OW versus PCOS-Ln comparison. ( D ) KEGG pathway analysis of DEPs in the CTRL-OW versus CTRL-Ln comparison. Dot sizes represent the counts of DEPs associated with corresponding pathways, while dot colors indicate the -log10-transformed statistical p  values Functional enrichment and pathway analysis of differentially expressed proteins in follicular fluid-derived small extracellular vesicles in PCOS and non-PCOS women. ( A ) Gene Ontology (GO) enrichment analysis of DEPs in the comparison between PCOS-OW and PCOS-Ln groups, indicating enrichment in intracellular organelles, nucleosome assembly, and binding activities. ( B ) GO enrichment analysis of DEPs in the comparison between CTRL-OW and CTRL-Ln groups, highlighting changes related to carbohydrate derivative metabolic processes and fructose-bisphosphate aldolase activity. ( C ) Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis of DEPs in the PCOS-OW versus PCOS-Ln comparison. ( D ) KEGG pathway analysis of DEPs in the CTRL-OW versus CTRL-Ln comparison. Dot sizes represent the counts of DEPs associated with corresponding pathways, while dot colors indicate the -log10-transformed statistical p  values The Protein-Protein Interaction (PPI) analysis revealed obesity-associated differences in the proteomic profiles of FF-sEVs between PCOS and non-PCOS women, with distinct structural and functional patterns (Fig.  4 A and B). In PCOS-OW women, dysregulation of structural proteins—particularly keratins (KRT13, KRT14) and collagen alpha-2(V) chain—suggested compromised follicular integrity, while cytoplasmic actin (ACTB) alterations may impair cellular motility and oocyte microenvironment remodeling. Key enzymatic differences emerged: PCOS-OW women exhibited elevated protein disulfide-isomerase (PDI), indicating heightened endoplasmic reticulum stress, alongside steroidogenic enzymes (hydroxy-delta-5-steroid dehydrogenase) that directly promote hyperandrogenism. In contrast, CTRL-OW women primarily adjusted metabolic enzymes like fructose-bisphosphate aldolase (glycolytic adaptation) and 7-dehydrocholesterol reductase (cholesterol homeostasis). Hormonal regulation showed distinct patterns: PCOS-OW samples displayed upregulated sex hormone-binding globulin (SHBG), coupled with increased hydroxy-delta-5-steroid dehydrogenase (HSD3B2) that potentiates androgen synthesis. Conversely, the CTRL-OW group showed downregulated angiotensinogen, possibly indicating altered renin-angiotensin system regulation in non-PCOS obesity. Transport protein alterations further differentiated the groups: PCOS-OW women showed vitamin D-binding protein (DBP) alterations that may impair vitamin D signaling—a recognized PCOS modifier—and LDL receptor-related protein 1 (LRP1) abnormalities suggesting dyslipidemia involvement, whereas CTRL-OW samples upregulated lipopolysaccharide-binding protein, possibly influencing immune modulation. Notably, the PCOS-OW proteome was dominated by inflammatory effectors: elevated C-reactive protein (CRP) and haptoglobin (HP) marked systemic inflammation, upregulated complement components (complement component 1 receptor, complement component 4B) activated immune cascades, and increased S100 Calcium-binding Protein A8 (S100A8) amplified sterile inflammation. Histone dysregulation (histones H2A, H3, and H4) further suggested obesity-driven epigenetic modifications in PCOS. In comparison, CTRL-OW samples showed elevated inter-α-trypsin inhibitor heavy chain 4 (ITIH4), an extracellular matrix stabilizer that may counteract metabolic stress. Collectively, these findings delineate how obesity exacerbates PCOS-specific pathologies—via steroidogenic disruption, Endoplasmic reticulum (ER) stress, and chronic inflammation—while driving distinct metabolic and vascular adaptations in non-PCOS women. Fig. 4 Protein-Protein Interaction networks of differentially expressed proteins in follicular fluid-derived small extracellular vesicles in PCOS and non-PCOS women. ( A ) Protein-Protein Interaction (PPI) network of DEPs from the comparison of PCOS-OW versus PCOS-Ln groups, emphasizing interactions among inflammation-related proteins (C-reactive protein, haptoglobin, vitamin D-binding protein, S100 Calcium-binding Protein A8), coagulation and complement pathway components (complement component 1 receptor, complement component 4B), and markers of oxidative and ER stress, underscoring obesity-induced inflammatory and stress responses. ( B ) PPI network of DEPs from the comparison of CTRL-OW versus CTRL-Ln groups, showing clusters associated with metabolic reprogramming and anti-inflammatory proteins (inter-α-trypsin inhibitor heavy chain 4), as well as the renin-angiotensin system (angiotensinogen), reflecting distinct obesity-induced molecular adaptations. In both PPI networks, nodes represent DEPs and lines indicate potential interactions. The color gradient ranges from red to blue, reflecting the significance of interactions Protein-Protein Interaction networks of differentially expressed proteins in follicular fluid-derived small extracellular vesicles in PCOS and non-PCOS women. ( A ) Protein-Protein Interaction (PPI) network of DEPs from the comparison of PCOS-OW versus PCOS-Ln groups, emphasizing interactions among inflammation-related proteins (C-reactive protein, haptoglobin, vitamin D-binding protein, S100 Calcium-binding Protein A8), coagulation and complement pathway components (complement component 1 receptor, complement component 4B), and markers of oxidative and ER stress, underscoring obesity-induced inflammatory and stress responses. ( B ) PPI network of DEPs from the comparison of CTRL-OW versus CTRL-Ln groups, showing clusters associated with metabolic reprogramming and anti-inflammatory proteins (inter-α-trypsin inhibitor heavy chain 4), as well as the renin-angiotensin system (angiotensinogen), reflecting distinct obesity-induced molecular adaptations. In both PPI networks, nodes represent DEPs and lines indicate potential interactions. The color gradient ranges from red to blue, reflecting the significance of interactions Correlation analysis using the Pearson correlation coefficient was conducted to evaluate the relationship between DEPs, clinical characteristics, and IVF/ICSI outcomes across the PCOS-OW and PCOS-Ln groups, as well as the CTRL-OW and CTRL-Ln groups (Fig.  5 ). The values of the Pearson correlation coefficient range from − 1 to + 1, with + 1 indicating perfect positive correlation, -1 indicating perfect negative correlation, and 0 representing no correlation. In the PCOS cohort, several DEPs showed significant correlations with BMI and reproductive outcomes. For instance, proteins such as CRP (0.63) and S100A8 (0.63) demonstrated positive correlations with BMI, indicating that higher expression levels of these inflammatory proteins are associated with increased obesity. Notably, S100A8 and complement component 4B (C4B) was negatively correlated with the rates of transferable (-0.81, -0.55) and good-quality embryos (-0.52, -0.22), suggesting that elevated levels may compromise embryo viability in the context of obesity-related reproductive challenges. Conversely, in the control group, proteins like Heat Shock Protein 90 Beta Family Member 1 (HSP90B1) exhibited a significant positive correlation with the rates of 2PN oocytes retrieved (0.91) and good-quality embryos (0.69), indicating a potential protective effect on reproductive outcomes. Additionally, Vascular Cell Adhesion Molecule 1 (VCAM1) showed a negative correlation with 2PN oocytes retrieved (-0.74) and transferable embryos (-0.75), reinforcing the notion that certain DEPs may modulate the adverse impacts of obesity on fertility. These findings highlight the intricate relationships between obesity, proteomic alterations, and fertility outcomes, warranting further investigation into their implications for reproductive health. Fig. 5 Correlation of differentially expressed proteins with clinical characteristics and IVF/ICSI outcomes in PCOS and non-PCOS women Correlation of differentially expressed proteins with clinical characteristics and IVF/ICSI outcomes in PCOS and non-PCOS women This figure presents a correlation analysis using the Pearson correlation coefficient to evaluate the relationships between DEPs, clinical characteristics, and IVF/ICSI outcomes in PCOS and non-PCOS patients. The heatmap colors indicate the correlation strength; red indicates positive correlations, while blue indicate negative correlations. The values range from − 1 to + 1, with + 1 indicating perfect positive correlation, -1 indicating perfect negative correlation, and 0 representing no correlation. Key findings include strong positive correlations of inflammatory proteins (CRP, S100A8) with BMI in the PCOS group, and significant negative correlations of S100A8 and C4B with transferable and good-quality embryos. In contrast, in the control group, HSP90B1 positively correlates with 2PN oocytes and good-quality embryos, suggesting potential modulatory roles of these proteins in reproductive success. Western blot analysis was performed to validate the expression levels of selected DEPs identified in the context of our comparative analyses (Fig.  6 A). Elastin Microfibril Interface Located Protein 1 (EMILIN1), prominently featured in the CTRL-OW versus CTRL-Ln groups, revealed a significant difference ( p  = 0.0151), indicating that obesity has a notable effect on EMILIN1 expression in the CTRL cohort (Fig.  6 B). Meanwhile, Lipopolysaccharide-binding protein (LBP) also exhibited a significant increase when comparing CTRL- OW to CTRL- Ln ( p  = 0.0076), reinforcing its role in the metabolic alterations associated with obesity (Fig.  6 C). Additionally, Coagulation Factor XII (F12) exhibited significant changes when comparing CTRL-OW to CTRL-Ln ( p  = 0.0385), indicating its potential in distinguishing between the CTRL-OW and CTRL-Ln groups (Fig.  6 D). Furthermore, the pro-inflammatory marker S100A8 showed a substantial and highly significant difference between PCOS-Ln and PCOS-OW groups ( p  < 0.001), highlighting its critical involvement in the obesity-related pathophysiology of PCOS (Fig.  6 E). Collectively, these results underscore the differential impact of obesity on the expression of specific proteins across the studied groups. These DEPs may offer promising avenues for further investigation into their functional roles and potential as biomarkers for IVF/ICSI outcomes. Fig. 6 Detection of specific differentially expressed proteins in follicular fluid-derived small extracellular vesicles in PCOS and non-PCOS women. ( A ) Representative Western blot graphs illustrating the expression levels of selected DEPs. ( B ) Quantitative analysis of Elastin Microfibril Interface Located Protein 1 (EMILIN1) expression levels. ( C ) Quantitative analysis of Lipopolysaccharide-binding protein (LBP) expression. ( D ) Quantitative analysis of Coagulation Factor XII (F12) expression. ( E ) Quantitative analysis of S100 Calcium-binding Protein A8 (S100A8) expression. These results highlighting the impact of obesity on specific protein expression profiles and all statistical data are presented as means ± standard deviation of two-tailed unpaired Student’s t -tests. * p  < 0.05, ** p  < 0.01, *** p  < 0.001 Detection of specific differentially expressed proteins in follicular fluid-derived small extracellular vesicles in PCOS and non-PCOS women. ( A ) Representative Western blot graphs illustrating the expression levels of selected DEPs. ( B ) Quantitative analysis of Elastin Microfibril Interface Located Protein 1 (EMILIN1) expression levels. ( C ) Quantitative analysis of Lipopolysaccharide-binding protein (LBP) expression. ( D ) Quantitative analysis of Coagulation Factor XII (F12) expression. ( E ) Quantitative analysis of S100 Calcium-binding Protein A8 (S100A8) expression. These results highlighting the impact of obesity on specific protein expression profiles and all statistical data are presented as means ± standard deviation of two-tailed unpaired Student’s t -tests. * p  < 0.05, ** p  < 0.01, *** p  < 0.001

Polycystic ovary syndrome (PCOS) is a prevalent endocrine and metabolic disorder affecting women of reproductive age, characterized by heterogeneity in its pathological and clinical profiles [ 1 ]. It primarily manifests through persistent ovulatory dysfunction, anovulation, hyperandrogenism, and the presence of polycystic ovaries [ 2 ]. Women diagnosed with PCOS often present with obesity, insulin resistance, and fertility challenges, elevating their risk for metabolic syndrome and long-term complications such as type 2 diabetes, non-alcoholic fatty liver disease, and hypertension [ 1 – 3 ]. Additionally, PCOS patients frequently experience reduced fertility, with alterations in oocyte competence being significant contributors to reproductive impairment [ 4 ]. Notably, the quality of oocytes harvested during assisted reproductive techniques, such as in vitro fertilization (IVF), is often suboptimal among women with PCOS, indicating a need for further investigation into the underlying mechanisms that influence both oocyte quality and embryonic development potential [ 5 , 6 ]. Obesity serves as a critical comorbidity in PCOS, is a chronic metabolic disorder caused by the interaction of genetic and environmental factors [ 7 , 8 ]. The prevalence of obesity in women with PCOS is alarming, reported to range between 50% and 80%, which is threefold higher than women without this condition [ 8 ]. Obesity exacerbates the long-term metabolic abnormalities prevalent in PCOS, influencing reproductive outcomes negatively [ 7 , 8 ]. Specifically, within the realm of assisted reproductive technology (ART), excess body weight has been shown to correlate with poorer IVF outcomes, impeding ovarian response, reducing oocyte yield and quality, and diminishing live birth rates [ 9 , 10 ]. Despite these associations, there exists a significant gap in the literature detailing the mechanistic pathways through which obesity affects IVF outcomes in women with PCOS. Additionally, treatment options for obese PCOS patients remain limited, necessitating urgent research to elucidate these mechanisms and develop effective strategies. Within the follicular development context, the follicular fluid (FF) environment is crucial for oocyte maturation and developmental competence [ 11 ]. FF is a metabolically active microenvironment that contains hormones, metabolites, and extracellular vesicles (EVs), facilitating communication between somatic cells and oocytes through paracrine signaling and other intercellular communication pathways [ 12 , 13 ]. Notably, EVs, including small extracellular vesicles (sEVs), are recognized as promising biomarkers in reproductive health, as they transport proteins, lipids, and genetic material between cells, influencing various physiological processes, including follicle development and oocyte quality [ 14 – 16 ]. In women with PCOS, the composition and functionality of FF-sEVs have been linked to dysregulated hormonal and metabolic pathways, potentially exacerbating infertility challenges [ 17 – 19 ]. Abnormalities in the protein and miRNA profiles of sEVs in PCOS patients have been documented, with implications for ovarian function and metabolic disturbances [ 18 , 20 ]. For instance, altered expression of EV proteins associated with inflammation, steroidogenesis, and cell signaling pathways has been observed, suggesting that these vesicles may play a crucial role in mediating the pathological effects of the syndrome [ 19 , 20 ]. However, a notable gap exists in current research regarding the specific impact of obesity on the protein profiles of FF-sEVs in both PCOS and non-PCOS populations. The intersection of obesity and PCOS presents unique challenges, as obesity is known to worsen the metabolic abnormalities associated with PCOS, potentially leading to further alterations in the composition and activity of FF-sEVs [ 21 ]. This raises the question of how obesity may influence the molecular signatures within sEVs, potentially contributing to the adverse reproductive outcomes frequently observed in obese women with PCOS. Understanding how obesity modulates FF-sEVs protein profiles is critical, as it may reveal novel biomarkers and therapeutic targets that could enhance fertility outcomes in this patient population. In light of these considerations, this study aims to address the critical knowledge gap regarding the impact of obesity on the proteomic composition of FF-sEVs. Utilizing data-independent acquisition (DIA) technology for comprehensive proteomic analysis, we will employ bioinformatics approaches—including Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), and Protein-Protein Interaction (PPI) analyses—to elucidate the molecular mechanisms by which obesity influences the FF microenvironment and how these alterations correlate with IVF outcomes. This exploration seeks to deepen our understanding of the interplay between obesity and PCOS while identifying potential biomarkers and therapeutic targets that may improve reproductive health in women adversely affected by obesity. Given the ongoing challenges faced by women with PCOS, especially those who are obese, our findings may contribute valuable insights for developing more effective fertility management strategies.

The interplay between obesity and PCOS presents a complex challenge within reproductive medicine, particularly regarding the proteomic alterations observed in the follicular development microenvironment, which affects the composition of FF-sEVs. Importantly, prior to our study, a significant research gap existed regarding the specific effects of obesity on the protein composition of sEVs in both PCOS and non-PCOS women, thereby underscoring the significance of our findings. Our proteomic analysis reveals that obesity profoundly alters the expression levels of proteins within sEVs, particularly in the PCOS cohort, emphasizing a marked divergence from non-PCOS counterparts. The upregulation of inflammatory markers such as S100A8 in the PCOS-OW group suggests a dysregulated inflammatory response leading to compromised reproductive outcomes. This contrasts with the more metabolic and inflammatory adaptations observed in non-PCOS women, where the inflammatory markers exhibited an anti-inflammatory state. Notably, our results implicate significant pathways such as Interleukin-17 (IL-17) signaling and protein processing in the endoplasmic reticulum as central to the mechanistic pathway by which obesity impacts reproductive health in PCOS. These observations indicate that the intricate interplay of inflammation, oxidative stress, and metabolic dysregulation in the follicular microenvironment serves as a critical juncture affecting oocyte quality and IVF success rates. Thus, tailored therapeutic strategies that address these unique molecular alterations may be essential in enhancing reproductive outcomes for obese women with PCOS, paving the way for more effective fertility treatments in this population. Our findings indicated significant upregulation of haptoglobin (HP), vitamin D-binding protein (DBP), and S100 Calcium-binding Protein A8 (S100A8) within the protein-protein interaction network of FF-sEVs in obese PCOS patients, underscoring the critical role of these inflammatory markers. Haptoglobin, an inflammatory factor synthesized mainly in the liver, serves multiple functions, including antioxidant activity and immune modulation [ 34 ]. Elevated serum HP levels during inflammation, along with proteomic studies identifying HP as a potential biomarker for PCOS, further illustrate its significance, as PCOS patients exhibit increased HP and HP β-chain abundance compared to controls [ 35 – 37 ]. Moreover, since HP is expressed in adipose tissue, obesity is likely to further elevate serum and follicular levels of HP, reinforcing our results [ 38 ]. The connection between obesity in PCOS and an elevated risk of ovarian hyperstimulation syndrome (OHSS) highlights the relevance of inflammation, with HP previously reported as a predictor of OHSS, indicating its role in this serious complication. Notably, prior randomized controlled trials have demonstrated that α-1 haptoglobin fragment in embryo culture media may aid in identifying non-viable embryos, suggesting that increased HP levels could mediate cellular damage affecting embryonic development, thereby influencing IVF outcomes [ 38 ]. DBP plays a crucial role in enhancing neutrophil chemoattractant activity during inflammation, which may indirectly impact oocyte development by modulating immune cell trafficking and inflammation within the ovarian microenvironment [ 39 ]. The formation of DBP-actin complexes is indicative of tissue damage and may amplify chemotactic signaling, releasing pro-inflammatory molecules such as S100A8/A9 (40). These proteins, part of the S100 calcium-binding family, are elevated in various metabolic inflammatory conditions, including obesity and PCOS [ 40 ]. Within FF-sEVs, S100A8/A9 significantly enhance inflammation and disrupt steroidogenesis via activation of the nuclear factor-kappa B (NF-κB) signaling pathway [ 26 ]. This pathway is a key regulator of immune responses and has been shown to play an essential role in regulating granulosa cell function [ 26 ]. Activation of NF-κB in the ovarian microenvironment, particularly within granulosa cells, induces a cascade of inflammatory responses that culminate in the formation of the NOD-like receptor protein 3 (NLRP3) inflammasome. The NLRP3 inflammasome, when activated, impairs granulosa cell proliferation and disrupts folliculogenesis, negatively affecting oocyte and follicular development [ 41 ]. This provides mechanistic insight into how the inflammation induced by elevated S100A8/A9 and NF-κB activation can directly impact oocyte quality, reducing the likelihood of successful fertilization and implantation during IVF. Recent transcriptomic studies further confirm the heightened inflammatory state in women with PCOS, particularly involving innate immune activation and metabolic dysfunction, reinforcing the central role of inflammation in the pathophysiology of this disorder [ 42 ]. Additionally, while our proteomic analysis did not detect significant differential expression of vitamin D receptor (VDR) in FF-sEVs, emerging evidence suggests that VDR polymorphisms may contribute to PCOS pathogenesis through indirect mechanisms. A recent meta-analysis demonstrated that specific VDR variants are associated with elevated testosterone levels and altered AMH synthesis, potentially disrupting ovarian steroidogenesis [ 43 ]. This genetic predisposition, combined with DBP-mediated inflammatory cascades, may create a feed-forward loop exacerbating hyperandrogenism and follicular dysfunction in PCOS. Our results highlight a pro-inflammatory state in FF-sEVs in PCOS, suggesting that inflammation is a key factor exacerbating reductions in oocyte and embryonic developmental potential associated with obesity. In women with PCOS, obesity significantly increases cardiovascular risks, including hypertension and dyslipidemia, particularly among those aged 25 to 34 [ 44 ]. Complement activation has been shown to play a role in various disease processes associated with PCOS, including diabetes and its complications, as well as cardiovascular diseases [ 45 , 46 ]. However, the regulatory mechanisms underlying complement dysregulation in PCOS remain unclear. Previous studies have indicated that the alternative complement pathway is upregulated in non-obese PCOS patients and further exacerbated in obese individuals, highlighting the detrimental impact of obesity on complement dysregulation in PCOS, consistent with our findings [ 47 ]. Complement proteins were found to be involved in the signaling networks of the proteomic profile of FF-sEVs in obese PCOS patients, emphasizing the potential of targeting complement dysregulation to mitigate complications such as metabolic syndrome and cardiovascular diseases. Additionally, PCOS increases the risk of venous thromboembolism (VTE), with cohort studies indicating a 1.5-fold increase in risk among affected women [ 48 ]. Evidence suggests that obese women with PCOS exhibit a hypercoagulable state, with changes in coagulation factors aligning with obesity profiles [ 49 ]. ​The coagulation pathway’s involvement extends beyond cardiovascular implications; it may also influence the ovarian environment [ 49 ]. Dysregulation of coagulation factors in the FF could lead to microthrombi formation or altered blood flow within the follicle, potentially impairing nutrient and oxygen delivery to the developing oocyte [ 48 – 50 ]. Such perturbations might compromise oocyte maturation and quality, thereby affecting fertilization potential and embryo development [ 50 ]. Thus, the impacts of coagulation and complement pathways on cardiovascular health warrant careful attention in obese PCOS patients, as certain injuries such as cardiovascular events or infections may be amplified by obesity [ 51 ]. Understanding these interactions is crucial for preventing long-term cardiovascular disease and metabolic syndrome. Balanced metabolism is crucial for optimal oocyte quality, as metabolic fluctuations within the follicular microenvironment can alter follicular cell function and subsequently impact oocyte developmental potential [ 52 ]. Our data demonstrated significant enrichment in metabolic pathways related to obesity and PCOS. Lipid metabolism plays a vital role in regulating granulosa cell functions, and its disruptions may lead to cellular apoptosis [ 53 ]. Lipotoxicity could contribute to oocyte organelle damage in obesity, characterized by increased fatty acid synthesis and inhibited oxidation [ 54 ]. When dietary fatty acid accumulation in adipocytes exceeds their storage capacity, free fatty acids deposit in non-adipose tissues, causing harm. Elevated circulating free fatty acids in obese women can induce oxidative stress through increased reactive oxygen species (ROS) production, damaging non-adipose cells, instigating mitochondrial and ER stress, and ultimately promoting apoptosis [ 55 ]. Notably, high levels of free fatty acids in FF correlate with abnormal oocyte morphology [ 54 ]. In PCOS patients, oxidative capacity in FF is heightened, with pro-oxidative forces surpassing antioxidant defenses. Although total antioxidant capacity (TAC) remains stable, slight increases in antioxidant markers suggest a compensatory mechanism to maintain redox homeostasis, which may be exacerbated by obesity [ 56 ]. Research shows that both diet-induced and genetically induced obesity negatively affect oocyte meiotic maturation, spindle morphology, and polarity, leading to oxidative stress, mitochondrial dysfunction, disrupted DNA and histone methylation, and oocyte apoptosis [ 57 ]. Histones are essential for genome stability and gene regulation; however, the intricate interactions within the histone family in our data require further investigation [ 58 ]. As oxidative stress escalates, the accumulation of misfolded proteins triggers the unfolded protein response (UPR), aimed at restoring the cell’s protein-folding capacity [ 59 ]. Research has identified the roles of ER stress and UPR signaling pathways in the maturation of oocytes and pre-implantation embryos, underscoring their importance in reproductive outcomes [ 60 ]. Furthermore, impaired ER function adversely impacts fertilization potential, highlighting the necessity of maintaining protein homeostasis [ 61 ]. The protein disulfide isomerase (PDI) family, particularly PDI, plays a pivotal role in protein folding within the ER and can also modulate various biological processes upon secretion [ 62 ]. Our results indicated that PDI was particularly upregulated in the obese PCOS group, serving as a key protein in the protein-protein interaction network, which underscores its crucial role in this context. While direct evidence linking PDI overexpression to compromised oocyte quality is limited, studies have shown that increased ER stress and altered PDI expression in aging ovaries are associated with reduced oocyte quality [ 62 ]. Additionally, antioxidants such as melatonin have shown promise in alleviating ER stress through UPR regulation, improving oocyte maturation and embryo development [ 63 ]. Given the interactions among metabolic dysfunction, oxidative stress, and increased ER stress under obesity, multifaceted approaches are essential for improving IVF/ICSI laboratory outcomes in PCOS patients. The restoration of lipid metabolic balance, regulation of oxidative stress, and enhancement of ER function may collectively benefit the developmental potential of oocytes and embryos in both in vivo and ex vivo culture systems. Previous studies have demonstrated that women accompanied by obesity experience significantly decreased oocyte maturation and fertilization rates during IVF [ 7 ]. This phenomenon is often correlated with smaller oocyte size, resulting in higher cycle cancellation rates and an increased proportion of poor-quality embryos [ 7 ]. The pentose phosphate pathway (PPP) plays a pivotal role in supporting oocyte nuclear and cytoplasmic maturation [ 64 ]. Metabolic pathways involving fructose and mannose, particularly fructose, are closely associated with obesity, leading to increased lipogenesis, dyslipidemia, and reduced insulin sensitivity [ 65 ]. Fructose-1,6-bisphosphate aldolase, a crucial enzyme in fructose metabolism, demonstrates downregulation in the context of obesity-induced metabolic reprogramming, reflecting an adaptation to high-fructose and high-fat environments that may ultimately influence energy metabolism. Our findings indicated significant enrichment of the PPP and altered expressions of key enzymes within fructose and mannose metabolic pathways in the obese non-PCOS group. This underscores the potential metabolic reprogramming that obesity induces in the follicular microenvironment, subsequently affecting energy balance during oocyte maturation and, hence, oocyte quality and embryo developmental potential. Additionally, Inter-α-trypsin inhibitor heavy chain 4 (ITIH4), a controversial inflammatory marker associated with recurrent miscarriage, may contribute to maintaining the balance between Th1 and Th2 responses while supporting anti-inflammatory processes [ 66 ]. Lipopolysaccharide-binding protein (LBP), an acute-phase protein that elevates during inflammation, binds to LPS and enhances its presentation to CD14-positive cells, activating the Toll-like receptor 4 (TLR4) pathway and inciting pro-inflammatory responses [ 67 ]. We observed downregulation of LBP and upregulation of ITIH4 in the obese non-PCOS group, suggesting that inflammatory proteins in FF-sEVs may be in an anti-inflammatory state, potentially counteracting chronic low-grade inflammation induced by obesity and alleviating its suppressive effects on oocyte maturation. Furthermore, the renin-angiotensin system (RAS) in the ovaries plays a critical role in follicular development and ovulation [ 68 ]. We found significant enrichment of renin secretion and the renin-angiotensin system, along with increased expression of angiotensinogen (AGT) in obese subjects, indicating that obesity may alter ovarian RAS functionality, potentially impacting oocyte ovulation capacity. Our findings not only provide novel insights into the reproductive challenges faced by obese women undergoing ART but also lay the groundwork for developing targeted interventions to address fertility issues in this population. We acknowledge several important limitations that must be carefully considered when interpreting our findings. Primarily, the relatively sample size inevitably restricts the statistical power of the study and may limit the generalizability of our results. While we employed a meticulous matching process based on key confounding variables such as age and BMI to mitigate inter-patient variability, the inherent risk of bias remains a critical concern. This risk is accentuated by the small cohort of women diagnosed with PCOS, which might not fully represent the heterogeneity of the PCOS population. Furthermore, the strict exclusion criteria, while essential for maintaining study integrity, may inadvertently limit the applicability of our findings to a wider clinical context. Additionally, due to sample size limitations and resource constraints, certain potentially relevant metabolic variables—such as vitamin D levels—were not assessed in this study. It is also important to emphasize that although our proteomic analysis identified key proteins and pathways potentially implicated in reproductive outcomes, functional validation experiments were not performed. To address these limitations, future research will prioritize selecting critical candidate proteins for in-depth functional studies to verify their mechanistic roles in oocyte competence and embryo development. Furthermore, future research should prioritize multicenter studies with larger, more diverse populations to validate these findings, incorporate expanded metabolic profiling, and comprehensively elucidate the dynamic molecular pathways involved. Finally, integrating longitudinal studies with multi-omics approaches and functional assays will further clarify the impact of obesity on oocyte quality and IVF/ICSI outcomes, ultimately guiding the development of targeted therapeutic strategies for women with PCOS.

In conclusion, our investigation into obesity’s impact on FF-sEVs proteomics reveals significant differences between PCOS and non-PCOS women, highlighting a complex interplay of mechanisms. In PCOS, obesity exacerbates a cycle of enhanced inflammation, coagulopathy, and oxidative stress, adversely affecting oocyte development, maturation, and quality. This interplay compromises embryonic potential and significantly reduces IVF success rates. In contrast, non-PCOS women experience metabolic reprogramming within the follicular microenvironment, influencing energy balance during oocyte maturation. Anti-inflammatory proteins in FF-sEVs may mitigate the chronic low-grade inflammation associated with obesity, offering some protection to oocyte quality. These findings emphasize the need for tailored fertility management in obese women, recognizing the differing mechanisms at play. Current PCOS treatments remain suboptimal, lacking effective strategies to enhance oocyte quality during IVF/ICSI. Educating patients on the importance of weight loss prior to ovarian stimulation is crucial. Furthermore, additional research into embryo culture specific to PCOS patients is necessary, including functional validation of key sEVs proteins to clarify their roles and therapeutic potential. A multi-faceted approach addressing inflammation, oxidative stress, endoplasmic reticulum distress, and metabolic health may support healthier embryo development, ultimately improving IVF outcomes and enhancing pregnancy success rates.

Below is the link to the electronic supplementary material. Additional file 1: Supplementary Table 1: Differentially expressed proteins of PCOS_OW versus PCOS_Ln (Upregulated Proteins) Additional file 1: Supplementary Table 1: Differentially expressed proteins of PCOS_OW versus PCOS_Ln (Upregulated Proteins) Additional file 2: Supplementary Table 2: Differentially expressed proteins of PCOS_OW versus PCOS_Ln (Downregulated Proteins) Additional file 2: Supplementary Table 2: Differentially expressed proteins of PCOS_OW versus PCOS_Ln (Downregulated Proteins) Additional file 3: Supplementary Table 3: Differentially expressed proteins of CTRL_OW versus CTRL_Ln (Upregulated Proteins) Additional file 3: Supplementary Table 3: Differentially expressed proteins of CTRL_OW versus CTRL_Ln (Upregulated Proteins) Additional file 4: Supplementary Table 4: Differentially expressed proteins of CTRL_OW versus CTRL_Ln (Downregulated Proteins) Additional file 4: Supplementary Table 4: Differentially expressed proteins of CTRL_OW versus CTRL_Ln (Downregulated Proteins) Additional file 5: Supplementary Figure 1: Full uncropped Gels and Blots image(s) Additional file 5: Supplementary Figure 1: Full uncropped Gels and Blots image(s)