The subjects in this study consisted of PCOS and healthy women who visited the Reproductive Medicine Center of Peking University Third Hospital from April to November 2017. PCOS was diagnosed according to the 2003 Rotterdam Criteria, i.e., the presence of two or more of oligo-ovulation and/or anovulation cycles, clinical and/or biochemical signs of hyperandrogenism, and polycystic ovaries after exclusion of other etiologies. The control subjects were selected from women attending the clinic on account of male azoospermia. All of the controls had regular menstrual cycles and normal ovarian morphology, and none exhibited clinical or biochemical hyperandrogenism. All subjects were free of medications known to affect metabolic function or reproductive function within 3 months preceding enrollment. This study was approved by the Reproductive Medicine Ethics Committee of Peking University Third Hospital (2016SZ-027). Informed consent after a detailed explanation of the procedure was acquired from each participant.

Patients underwent in vitro fertilization with the same stimulation protocol as previously described (Hu et al., 2019). Mural granulosa cells from PCOS and control subjects were isolated from the follicular fluid aspirated during oocyte retrieval by Ficoll density gradient centrifugation (Zhao Y. et al., 2015), and then the cells were lysed directly for protein extraction and analysis.

The granulosa cells were collected from 47 PCOS patients and 55 controls. The samples of two groups were mixed separately, and quantitative acetylated proteomics analysis was performed. There were three technical replicates. The sample was sonicated three times on ice using a high-intensity ultrasonic processor (Scientz) in lysis buffer (8 M urea, 1% Protease Inhibitor Cocktail) (Note: For PTM experiments, inhibitors were also added to the lysis buffer, e.g., 3 μM TSA and 50 mM NAM for acetylation). The remaining debris was removed by centrifugation at 12,000 g at 4°C for 10 min. Finally, the supernatant was collected, and the protein concentration was determined with a BCA kit according to the manufacturer’s instructions.

For digestion, the protein solution was reduced with 5 mM dithiothreitol for 30 min at 56°C and alkylated with 11 mM iodoacetamide for 15 min at room temperature in darkness. The protein sample was then diluted by adding 100 mM NH₄HCO₃ to urea concentration less than 2 M. Finally, trypsin was added at a 1:50 trypsin-to-protein mass ratio for the first digestion overnight and a 1:100 trypsin-to-protein mass ratio for a second 4-h digestion.

To enrich lysine acetylation-modified peptides, tryptic peptides dissolved in NETN buffer (100 mM NaCl, 1 mM EDTA, 50 mM Tris–HCl, 0.5% NP-40, pH 8.0) were incubated with prewashed antibody beads (PTM-101, PTM Bio) at 4°C overnight with gentle shaking. Then, the beads were washed four times with NETN buffer and twice with H₂O. The bound peptides were eluted from the beads with 0.1% trifluoroacetic acid. Finally, the eluted fractions were combined and vacuum-dried. For LC-MS/MS analysis, the resulting peptides were desalted with C18 ZipTips (Millipore) according to the manufacturer’s instructions.

The tryptic peptides were dissolved in 0.1% formic acid (solvent A) and directly loaded onto a homemade reversed-phase analytical column (15-cm length, 75 μm i.d.). The gradient was comprised of a solvent B (0.1% formic acid in 98% acetonitrile) increase from 6% to 23% over 26 min, an increase from 23% to 35% over 8 min, a climb to 80% over 3 min, and a hold at 80% for the last 3 min, all at a constant flow rate of 400 nL/min on an EASY-nLC 1000 UPLC system.

The peptides were subjected to an NSI source followed by tandem mass spectrometry (MS/MS) in Q Exactive^TM Plus (Thermo) coupled online to the UPLC. The electrospray voltage applied was 2.0 kV. The m/z scan range was 350 to 1800 for a full scan, and intact peptides were detected in the Orbitrap at a resolution of 70,000. Peptides were then selected for MS/MS using the NCE setting as 28, and the fragments were detected in the Orbitrap at a resolution of 17,500. A data-dependent procedure that alternated between one MS scan followed by 20 MS/MS scans with a 15.0-s dynamic exclusion was used. The automatic gain control (AGC) was set at 5E4.

The labile-free method was used to quantify the acetyl proteomics. The raw mass spectrometric data had been deposited and could be accessed using the following URL: http://msviewer.ucsf.edu/prospector/cgi-bin/msform.cgi?form=msviewer (the search key for the saved data set is q40sd3k4bg).

The ovarian granulosa cells were lysed with TNE buffer (50 mM Tris–HCl, pH 7.5, 100 mM NaCl, 1 mM EDTA, and 0.5% NP-40). One microgram of antibody was added and incubated at 4°C for 12 h with gentle shaking. The protein A magnetic beads were added and incubated at 4°C for 1 h. Then, the beads were washed with TNE buffer five times and boiled at 100°C for 5 min. The granulosa cell samples of 3 PCOS patients and 3 controls were tested by western blotting with an anti-acetylated antibody (PTM-105, PTM Biolabs) to verify the changes of protein lysine acetylation, and the granulosa cell samples of 25 PCOS patients were examined to analyze the correlation of the lysine acetylation of ACAT1 protein to the clinical outcomes. Antibodies included PGK1 (17811-1-AP, Proteintech), PGAM1 (ab129191, Abcam), GAPDH (ab181602, Abcam), and ACAT1 (#44276, Cell Signaling Technology).

The proteomic data were searched by MaxQuant database (v.1.5.2.8). The parameters were SwissProt Human (20130 sequences). We used a subcellular localization predication software, WoLF PSORT, to predict subcellular localization. The STRING database system was used to construct a protein–protein interaction network. Cluster membership was visualized by a heat map using the “heatmap.2” function from the “gplots” R-package. The Gene Ontology (GO) annotation proteome was derived from the UniProt-GOA database (http://www.ebi.ac.uk/GOA/). The Kyoto Encyclopedia of Genes and Genomes (KEGG) database was used to annotate the protein pathways. A corrected p-value < 0.05 was considered significant. In motif analysis, Soft motif-x was used to analyze the model of sequences constituted with amino acids in specific positions of modify-21-mers (10 amino acids upstream and downstream of the site) in all protein sequences. Correlation analysis of acetylation level and clinical results was based on Pearson correlation analysis. ImageJ was used to analyze the grayscale of Western blotting.

We collected the ovarian granulosa cells from 47 PCOS patients and 55 non-PCOS controls. The baseline characteristics of the PCOS and control groups are described in Table 1. These two groups had similar ages and BMIs. The women in the PCOS group had much longer menstrual cycles than those in the control group. In addition, there was a broad spectrum of endocrine and metabolic changes in the PCOS group compared with the control group, including obviously increased triglyceride and low-density lipoprotein (LDL) levels, reduced high-density lipoprotein (HDL) levels, marked increases in the serum concentrations of androgens including testosterone and androstenedione, and an increased ratio of luteinizing hormone (LH) to follicle-stimulating hormone (FSH). These changes were consistent with the clinical phenotypes of PCOS.

To comprehensively detect the changes in protein lysine acetylation in ovarian granulosa cells, a quantitative acetylated proteomics analysis was carried out according to the flowchart shown in Figure 1A. The granulosa cell samples in 47 PCOS patients and 55 non-PCOS controls were mixed separately. We first measured the lysine acetylation levels in the total protein sample of the two groups and found widespread protein lysine acetylation in the granulosa cells (Figure 1B). The quantitative profile of protein lysine acetylation by LC-MS/MS was performed in triplicate granulosa cells from the PCOS and control groups. The results showed that there were 1398 acetylated sites and 804 acetylated proteins overlapped between the two groups (Figure 1C). Pearson correlation analysis indicated that three technical parallel replicates had a good repeatability (Supplementary Figure 1), and the length of peptides agreed with the quality of tryptic peptides (Supplementary Figure 2).

In total, 410 sites on 333 proteins were identified differentially acetylated when the differential expression multiples were more than 1.5 between PCOS and control groups, of which 334 sites on 265 proteins had increased levels of acetylation and 76 sites on 68 proteins had decreased levels of acetylation in PCOS group (Figures 2A,B and Supplementary Table 1). The subcellular locations of these differentially acetylated proteins were extensively distributed in the cytoplasm, nucleus, and mitochondria (Figure 2C). The characteristics of specific acetylated peptides were evaluated by motif analysis (Supplementary Figure 3). Phenylalanine (F), histidine (H), serine (S), and asparagine (N) were prone to appear in the positions flanking the lysine acetylation sites. Moreover, the quantitative analysis of the top differentially acetylated proteins was shown in triplicate granulosa cells from PCOS and control groups. The clustering results revealed that proteins with significant differences in acetylation levels in the PCOS group consisted of many enzymes in metabolic regulation (Figure 2D). To better understand the function of lysine acetylation, the differentially acetylated proteins were analyzed using a protein–protein interaction network. Three highly connected clusters were identified, including carbon metabolism, ribosome, and spliceosome (Figure 2E), which indicated that lysine acetylation occurred in various metabolic complexes.

The GO functional classification showed the enrichment trends of the differentially acetylated proteins in the biological process and molecular function categories. The proteins with significant changes in acetylation level in the PCOS group were mainly involved in the metabolic processes and immune regulation (Figures 3A,B). We further performed a function-enrichment analysis based on the KEGG database. Most notably, differentially acetylated proteins were significantly enriched in the metabolic pathways such as carbon metabolism, tryptophan metabolism, glycolysis/gluconeogenesis, and fatty acid degradation, and molecular signaling pathways regulated glucose and lipid metabolism such as HIF-1 and PPAR signaling pathways (Figure 3C and Supplementary Table 2). To further confirm the correlations of protein functions with the alteration level of acetylation modification, we divided the differentially modified proteins into four quantified parts according to their differential expression multiples: Q1 (0 < ratio ≤ 1/2), Q2 (1/2 < ratio ≤ 1/1.5), Q3 (1.5 < ratio ≤ 2), and Q4 (ratio > 2) (Figure 4A). We then conducted KEGG enrichment analyses for each Q group and performed a cluster analysis (Figure 4B). The enrichment analyses showed that the differentially acetylated proteins were mainly enriched in the branched-chain amino acid degradation and fatty acid degradation/metabolism in the Q1 group, in the glycolysis/gluconeogenesis in the Q2 group, in the TCA cycle and carbon metabolism in the Q3 group, and inositol phosphate metabolism in the Q4 group.

Since several metabolic pathways exhibited acetylation changes by KEGG analysis, we narrowed our investigation to the subset of enzymes involved in glycolysis, fatty acid degradation, TCA cycle, and amino acid metabolism, to examine whether lysine acetylation level of key enzymes influenced the metabolism homeostasis of granulosa cells. The enzymes with significant changes of acetylation level in the PCOS group were highlighted in a metabolic pathway diagram (Figure 5A). In glycolysis, five enzymes involved in the reactions for the conversion of glucose-6-phosphate to pyruvate were hypoacetylated in PCOS granulosa cells, including glyceraldehyde-3-phosphate dehydrogenase (GAPDH), phosphoglycerate kinase 1 (PGK1), phosphoglycerate mutase 1 (PGAM1), triosephosphate isomerase 1 (TPI1), and alpha-enolase 1 (ENO1). The dihydrolipoyllysine-residue acetyltransferase component of pyruvate dehydrogenase complex (DLAT), which catalyzed the conversion of pyruvate to acetyl-CoA, was hyperacetylated in the PCOS group. Two enzymes that catalyzed the fatty acid degradation and produced acyl-CoA, long-chain fatty acid CoA ligase 4 (ACSL4) and peroxisomal acyl-CoA oxidase1 (ACOX1), were hypoacetylated in the PCOS group. Besides, the acetylation levels of several key enzymes in TCA cycle and amino acid metabolism (tryptophan metabolism, branched-chain amino acid degradation, and aspartate and glutamate metabolism) were significantly changed in the PCOS group. Acetyl-CoA linked the glycolysis, fatty acid degradation, and amino acid metabolism to the TCA cycle, and these metabolic pathways had in common a single enzyme, acetyl-CoA acetyltransferase 1 (ACAT1). ACAT1 catalyzed a reversible reaction of the cleavage of acetoacetyl-CoA into two acetyl-CoAs. The lysine acetylation level of ACAT1 was notably increased at the specific site of Lys-174 in the PCOS group. In addition, co-immunoprecipitation and Western blotting analysis further verified the changes in protein acetylation levels of four enzymes in PCOS ovarian granulosa cells compared to control cells, ACAT1, GAPDH, PGK1, and PGAM1 (Figure 5B).

The changes in the lysine acetylation of enzymes might affect their activities and regulated metabolic reactions in PCOS. Metabolic disorders of granulosa cells interfere with the metabolic homeostasis in the microenvironment of follicles, thereby affecting oocyte maturation and development potential. Since ACAT1 was an enzyme central to metabolic pathways, we further analyzed whether the changes of ACAT1 acetylation influence the process of oocyte maturation and embryo development. We collected the granulosa cell samples of 25 PCOS patients, and examined the ACAT1 acetylation level of each sample (Figure 6A). The acetylation level of ACAT1 Lys-174 was normalized with immunoprecipitated ACAT1 protein level by ImageJ. When the ratio of the acetylation level of ACAT1 Lys-174 to the expression level of ACAT1 protein was more than 2, it was classified as a hyperacetylation subgroup, while the hypoacetylation subgroup was defined when the ratio was less than 2. The clinical outcomes of these two subgroups were further analyzed. As shown in Table 2, the PCOS women with enhanced ACAT1 acetylation in their granulosa cells had a markedly reduced 2PN fertilization rate (56.5% vs. 69.3%, respectively), cleavage rate (56.2% vs. 69.3%, respectively), and rate of transferable embryos (52.5% vs. 91.7%, respectively). At the same time, there was a significantly negative correlation of ACAT1 acetylation levels in PCOS granulosa cells with 2PN fertilization rate and rate of transferable embryos (Figures 6B,C), indicating that increased ACAT1 acetylation in granulosa cells affected the ovarian metabolic microenvironment and inhibited oocyte quality and embryo development efficiency.