Comparison of chromatin accessibility remodeling of granulosa cells in patients with endometrioma or pelvic/tubal infertility

Purpose To elucidatethe epigenetic alteration associated with impaired oogenesis in endometrioma using multi-omic approaches.

ATAC-seq was performed on the granulosa cells (GCs) of 6 patients (3 with endometrioma and 3 without). Follicular samples from another 20 patients (10 with endometrioma and 10 without) were collected for mRNA-seq analysis of GCs and extracellular vesicles (EVs) of follicular fluid. qRT-PCR validated candidate genes in GCs from 44 newly enrolled patients (19 with endometrioma and 25 without). mRNA abundance was compared with the Mann–Whitney test. Pearson’s correlation analyzed relationships between candidate genes and oocyte parameters.

Chromatin accessibility and gene expression profiles of GCs from endometrioma patients differed significantly from the pelvic/tubal infertility group. RNA-seq revealed most differentially expressed genes were downregulated (6216/7325) and enriched in the cellular localization pathway. Multi-omics analyses identified 22 significantly downregulated genes in the GCs of endometrioma patients, including PPIF (P < 0.0001) and VEGFA (P = 0.0148). Both genes were further confirmed by qRT-PCR. PPIF (r = 0.46, p = 0.043) and VEGFA (r = 0.45, p = 0.048) correlated with the total number of retrieved oocytes.

GC chromatin remodeling may disrupt GC and EV transcriptomes, interfering with somatic cell-oocyte communication and leading to compromised oogenesis in endometrioma patients. Supplementary Information The online version contains supplementary material available at 10.1007/s10815-024-03302-7.

Endometrioma, Granulosa cells, Extracellular vesicles, Compromised oogenesis, Multi-omics data

Endometriosis is a prevalent gynecological illness that affects around 10% of women of reproductive age [1, 2]. A total of 30–50% of patients with endometriosis attend fertility clinics due to infertility [3, 4]. Endometrioma, which refers to the presence of endometrial tissue in the ovary [5, 6], may impede oogenesis by disrupting follicle health [7, 8]. Patients with endometrioma usually harvest fewer and poorer oocytes during ovarian stimulation cycles, which is associated with unsatisfactory pregnancy outcomes [9, 10]. Data from single-cell RNA sequencing revealed that oocytes from patients with endometrioma, regardless of whether they originated from unaffected ovaries, displayed a distinct transcriptomic signature in comparison with oocytes from healthy egg donors [11]. Moreover, the oocytes retrieved from the patients with endometrioma are more likely to have morphological abnormalities and lower mitochondrial content, which could be attributed to reduced p450 aromatase activity, increased reactive oxygen species (ROS), and disrupted follicular microenvironment [12, 13]. However, the mechanism by which endometrioma impairs oogenesis remains unknown. Epigenetic regulations, including DNA methylation, histone methylation, and chromatin accessibility, have been proven to be crucial for oogenesis [14]. For instance, our prior research demonstrated that an aberrant profile of PIWI-interacting RNAs, which were engaged in the epigenetic regulation of genes that form the oocyte extracellular matrix, was associated with recurrent oocyte maturation failure [15]. Beginning with primary follicles, DNA methylation was gradually established in the oocytes of growing follicles, which is closely associated with oocyte development [16, 17]. In addition, previous studies revealed that H3K4me3 histone methylation remarkably increased in antral stage oocytes and thereafter [18–20]. Also, studies discovered that extensive chromatin regions stayed open prior to zygotic genome activation (ZGA), which may contribute to the reprogramming of early human embryos [21, 22]. A recent study reported that Snf2h, as a chromatin remodeling factor, could modulate the transcription of oocyte meiotic genes by increasing the accessibility of chromatin around their promoters [23]. Given the significance of epigenetic regulation in chromatin organization, it is of great interest to investigate the chromatin accessibility landscapes of granulosa cells (GCs) in patients with endometrioma. Assay for Transposase-Accessible Chromatin with high-throughput sequencing (ATAC-seq) is a powerful method for detecting genome-wide chromatin accessibility [24]. This method employs the Tn5 transposase to insert sequencing adapters into all accessible regions of the genome, where the DNA sequences are cleaved, amplified, and sequenced utilizing next-generation sequencing (NGS) [25, 26]. The sequencing reads can then be used to identify the genomic regions with increased accessibility and infer the elements harboring transcription factors (TFs) and nucleosome position sites [27]. Therefore, ATAC-seq has been extensively employed to elucidate the pathogenic mechanisms underlying human diseases [28, 29]. Furthermore, extracellular vesicles (EVs) are lipid-bound vesicles that are secreted by cells into the extracellular space. EVs can be classified into three subtypes: microvesicles (MVs), exosomes, and apoptotic bodies based on their biogenesis, size, composition, and function. The EV cargos, which contain lipids, nucleic acids, and proteins, have been identified as a key mediator of cell-to-cell communication and thus are closely linked to regulating the cellular microenvironment [30–32]. Therefore, studying EVs would aid in comprehending microenvironment interaction and identifying novel diagnostic biomarkers for human diseases [33–35]. The intrafollicular microenvironment, consisting of GCs, follicular fluid (FF), and oocytes, is critical to oogenesis. FF-derived EVs serve as a bridge between GCs and oocytes, delivering molecular signals and cytokines during intercellular communication. More importantly, EVs can transport the nutrients or cytokines released by GCs to nurture oocyte development [36]. As a result, we may examine the healthy state of follicles by detecting the contents of EV cargos. Previous studies also reported that EV micro-RNA (miRNA) profiles in the FF of women with polycystic ovary syndrome (PCOS) and premature ovarian insufficiency (POI) were considerably altered [37]. Nevertheless, most studies have only examined the miRNA landscapes in FF-derived EVs and the mRNA transcriptome of FF EVs from patients with endometrioma is still lacking. In the current study, multi-omics methods, including GC ATAC sequencing (ATAC-seq), GC mRNA sequencing (mRNA-seq), and EV mRNA-seq, were performed on various follicular compositions from patients with endometrioma in order to investigate the mechanism of compromised oogenesis.

Participants Before initiating the study, approval from IRB (no. 201717) was obtained following a review by the scientific research ethics committee of Sun Yat-sen Memorial Hospital. We confirm that all experiments were performed in accordance with the Declaration of Helsinki. All the patients have signed an informed consent before donating their abandoned GCs and FF for research purposes. The following criteria were taken to recruit patients with endometrioma: (1) endometrioma identified by ultrasonography or laparoscopy; (2) age < 35 years; (3) BMI ≥ 18 kg/m2 and < 22 kg/m2; (4) received less than two in vitro fertilization (IVF) cycles. The pelvic/tubal infertility group was chosen based on the following criteria: (1) normal ovarian reserve (AMH ≥ 1.2 ng/ml, AFC ≥ 10, FSH < 10 IU/L); (2) pelvic or fallopian tubal infertility indicated by HSG or laparoscopy; (3) age < 35 years; (4) BMI ≥ 18 kg/m2 and < 22 kg/m2; (5) IVF cycles (less than two cycles). The exclusion criteria for both groups were as follows: (1) ovary surgery or chemotherapy; (2) polycystic ovary syndrome (PCOS); (3) severe hydrosalpinx (≥ 3 cm); (4) other complications such as insulin resistance, abnormal thyroid function, antiphospholipid syndrome, systemic lupus erythematosus, and genetic abnormalities; (5) sperm abnormality. Thirteen patients with endometrioma were enrolled in the case group (3 cases for ATAC-seq, 10 additional cases for GC mRNA-seq and EV mRNA-seq). In addition, the pelvic/tubal infertility group consisted of thirteen infertile patients (three for ATAC-seq, ten for GC mRNA-seq and EV mRNA-seq). ATAC-seq and mRNA-seq were performed on GCs from various patients. Controlled ovarian stimulation All the patients received a long-acting GnRH agonist (1.25 mg) for pituitary desensitization on day 20 of the cycle, followed by gonadotrophin injections for ovarian hyperstimulation. The dosage of gonadotrophin was determined by ovarian response and adjusted by ultrasound and serum sex hormone testing at intervals. When at least two follicles reached 18 mm in diameter, 10,000 IU human chorionic gonadotropin (hCG) was administrated to trigger oocyte maturation. GCs and FF collection The oocyte retrieval surgery was performed 36–38 h after hCG triggering. The dominant follicles were punctured with a 17-gauge needle to collect the clear FF, which was stored at − 80 °C to isolate EVs later. The GCs were collected by puncturing follicles under ultrasonographic guidance. The GCs of patients with bilateral endometrioma were mixed together, whereas the GCs of patients with unilateral endometrioma were collected separately. After RBC lysis with buffer (Invitrogen, 00–4333-57), the GCs were purified using density gradient centrifugation with Percoll (Sigma, P1644) and then kept at − 80 °C for future use [38, 39]. The morphological characteristics of GCs were assessed using Wright-Giemsa staining to evaluate their purity under a light microscope. In vitro fertilization and embryo culture Briefly, semen samples were processed via density gradient centrifugation (40%/80% PureSperm, Nidacon) at 300 g for 20 min. The resulting pellet was washed twice in G-IVF PLUS medium (Vitrolife) and adjusted to a concentration of 1 × 106 sperm/mL. Cumulus-oocyte complexes were denuded using 80 IU/mL hyaluronidase (SAGE) followed by mechanical pipetting. Mature (MII) oocytes were identified and cultured in a G-IVF PLUS medium. For insemination, approximately 50,000 motile sperm were added to each oocyte. Gametes were co-incubated at 37 °C with 6% CO2 and 5% O2 for 16–18 h, with fertilization confirmed by the presence of two pronuclei. Zygotes were cultured in the G1-PLUS medium (Vitrolife) for 48 h before being transferred to the G2-PLUS medium (Vitrolife) for extended culture. Embryos were assessed daily for developmental progression, and on day 5, high-quality blastocysts (≥ 3 BB according to Gardner’s criteria) were selected for transfer or cryopreservation [40, 41]. ATAC-seq We utilized the Epi™ ATAC-seq kit (Epibiotek, EPI20180202) to establish GC ATAC libraries, per the manufacturer’s instructions. After cell counting, approximately 50,000 GCs were pretreated with DNase for 30 min at 37 °C to remove free-floating DNA and digest DNA from dead cells. Cells were centrifuged for 5 min at 500 r.c.f. and then resuspended in 50 µL cold lysis buffer. After cell lysis, 1 mL of ATAC-seq RSB was added. Then, cellular nuclei were centrifuged for 10 min at 500 r.c.f. before being resuspended in 50 µL of transposition mix. Transposition reactions were incubated at 37 °C for 30 min in a thermomixer with shaking at 1000 r.p.m. Reactions were cleaned up with Zymo DNA Clean and Concentrator 5 columns. Finally, the DNA was resuspended in 25 µL of 2 × HiFi PCR mix and 1 µL of each Nextera i5 primer (N5xx) and Nextera i7 primer (N7xx). We performed PCR amplification reactions according to the following conditions: 72 °C for 5 min, 98 °C for 1 min, 14 cycles at 98 °C for 15 s, 60 °C for 30 s, and 72 °C for 1 min. After PCR, 35 µL of Epi™ DNA Clean Beads (Epibiotek, R1809) were added for size selection. The supernatant was transferred to a new tube, and 10 µL of fresh beads was added to capture fragments ranging from 250 to 350 bp. Libraries were quantified using the Bioptic Qsep100 Analyzer and sequenced to generate paired-end reads with lengths of 2 × 150 bp. Trimmomatic was used to remove low-quality reads (quality < 15), and 150 bp paired-end ATAC-seq reads were mapped to the reference human genome (GRCh38/hg38 build) by using Bowtie2. Only uniquely mapped reads were sorted using SAMtools and retained for subsequent analysis. Significant peak regions of each sample were identified using MACS2. The genomic region of the peak was annotated with the “ChIPseeker” R package [42], and the peaks of differential chromatin accessibility were detected with the “DiffBind” R package [43]. EV extraction and validation The FF was centrifugated (10,000 g, 60 min, 4 °C) to discard the debris and then filtered with a 0.22 µM filter (Millipore, GSWP04700) to remove the large suspended particles. The filtered supernatant was transferred to a centrifuge bottle and centrifuged at ultra-high speed (120,000 g) for 120 min at 4 °C. Then, the pellet was gently washed with PBS and centrifuged again, then the supernatant was carefully removed before resuspending the EV pellet with PBS. We measured the size and concentration of EVs with Nanosight NS300 (Malvern). Besides, an electron microscope was employed to confirm the morphological feature of EVs. Furthermore, we used a western blot to verify the expression of CD9 and CD63, the classic EV biomarkers. Briefly, the protein was extracted from EVs before protease and phosphatase inhibitors were added to prevent the degrading of samples. The protein was separated by electrophoresis in the SDS-PAGE gel and transferred to a blotting membrane. Moreover, 3% BSA was used to block the non-specifical antibodies for 1 h. After blocking non-specific antibodies, the membrane was incubated with the primary antibody overnight at 4 °C then incubated with the secondary antibody for 1 h at room temperature. The protein bands were visualized with electrochemiluminescence (ECL) reagents. In addition, the antibody information was listed as follows: both the primary antibodies of CD9 (System Biosciences, EXOAB-CD9A-1) and CD63 (System Biosciences, EXOAB-63A-1) were rabbit anti-human and diluted by 1:1000; both the secondary antibodies (System Biosciences, 180,202–001) were HRP conjugated goat anti-rabbit and diluted by 1:5000. GC and EV mRNA sequencing The total RNAs of GCs and EVs were extracted with Trizol (Invitrogen, 15,595,026). We constructed the mRNA libraries using VAHTSTM Stranded mRNA-seq Library Prep Kit (Vazyme Biotech, NR612), following the instructions provided by the manufacturer. Bioptic Qsep100 Analyzer was applied to check the quality of the constructed libraries. The mRNA-seq was performed on the Illumina platform (NovaSeq 6000) to generate 150 bp paired-end reads, which were then mapped to the human reference genome (GRCh38/hg38 build) with Hisat2. Then, HTSeq v0.6.6 was used to detect and summarize the distribution of reads on each chromosome and expression statistics. In the end, the gene expression levels were presented as TPM values (expected number of reads per kilobase of transcript sequence per million base pairs sequenced). Differentially expressed gene analysis The significantly differentially expressed (DE) genes were identified by an adjusted p-value threshold of 1 using the DEGseq2 software. Co-expression networks and sub-modules were constructed with WGCNA_1.70–3 [44]. Key regulators in the transcriptional regulatory networks were investigated with a TRRUST online tool (https://www.grnpedia.org/trrust/) [45]. Finally, we performed a hierarchical clustering analysis with the R package “gplots” based on the transcripts per million (TPM) values of the DE genes. The intersected DE genes and the enrichment of KEGG pathways were mapped with KOBAS 3.0 software (http://www.genome.jp/kegg). All the sequencing data can be obtained in the NCBI database (BioProject accession: PRJNA806789). Quantitative RT-PCR validation The total RNA of GCs was extracted using Trizol (Invitrogen, USA) from 44 newly enrolled patients (13 with bilateral endometrioma, 6 with unilateral endometrioma, and 25 without endometrioma). The SuperScript III First-Strand Synthesis System (Invitrogen, USA) was utilized for cDNA synthesis, followed by qPCR amplification with SYBR® Green qPCR supermixes (Bio-Rad, USA). Bio-Rad CTX96 real-time PCR detection system was applied for a real-time PCR reaction. The PCR primers of candidate genes were presented in Table S1, and their relative levels were calculated using the 2−ΔΔCt algorithm. The GAPDH was chosen as a housekeeping gene due to its stable expression in human GCs. The PCR reaction of each biological sample was triply repeated. Statistical analysis We used Student’s t-test to compare the patients’ demographic data, and the mRNA abundance of candidate genes was compared with the Mann–Whitney test. Pearson’s correlation was conducted to analyze the relationships between the candidate genes and oocyte parameters. GraphPad Prism v8.4 was used for statistical analysis, with p < 0.05 considered statistically significant.

Significant remodeling of chromatin accessibility was found in GCs of endometrioma As shown in Fig. 1a, the GC ATAC-seq data was presented with three biological replicates in endometrioma and pelvic/tubal infertility groups, and the intersected peaks were 3832 and 2071 in each group. A total of 61.35% of identified peaks in the pelvic/tubal infertility group were located in promoter regions (≤ 1 kb), whereas 77.89% in the endometrioma group (Fig. 1b, c). Across the entire human genome, ATAC-seq identified 1820 regions with significant alterations in chromatin accessibility, with 844 exhibiting higher chromatin accessibility and 976 displaying lower chromatin accessibility (Fig. 1d). Furthermore, genes adjacent to those regions were subjected to the analysis of pathway enrichment and TF binding, revealing that the cell-cycle pathway was the most over-represented pathway (Fig. 1e); c-MYC was the most enriched TF with lower chromatin accessibility (Fig. 1f), and Fosl2 was the most enriched TF with higher chromatin accessibility (Fig. 1g). RNA-seq data indicated that organelle homeostasis of GCs was disrupted in endometrioma Following a good cluster indicated by PCA analysis, the DE genes analysis identified 6216 downregulated genes and 1109 upregulated genes in the endometrioma group (Fig. 2a, b). The KEGG pathways of DE genes were primarily enriched in the organization of organelles and biosynthetic process (Fig. 2c, d), implying that metabolic and organelle networks in GCs were dysfunctional. Furthermore, to investigate the potential TF hubs in transcriptional regulatory networks, our results revealed a MYC-associated zinc finger protein (MAZ) as the top key regulator (Fig. 2e). In addition, given a large number of genes showed significant changes in expression, we investigated whether those DE genes, from a global view, were regulated as a co-expression network. Four co-expression modules were identified, and most DE genes were enriched in modules I and III (Fig. 2f). In module I, the enriched pathways were correlated with the establishment of localization in cell, cellular localization, response to organic substances, mRNA catabolic process, and RNA catabolic process (Fig. 2g). The enriched pathways in module III were associated with cellular localization, establishment of localization in cell, ion transport, cellular protein localization, and granulocyte activation (Fig. 2h). Multi-omics analysis and qRT-PCR validation of candidate genes The co-expression network analysis revealed that the cellular localization pathway, related to protein complex or organelle transportation within a cell, was remarkably enriched. Therefore, we looked into whether EVs play a role in disease progression. First, the EV was identified with sizes ranging from 50 to 450 nM and an average size of 212.6 ± 3.5 nm, with the most abundant EVs being around 200 nM (Fig. 3a). Second, the EV morphological hallmarks were shown in Fig. 3b, and EV was further confirmed by molecular markers CD9 (~ 28 kDa) and CD63 (~ 53 kDa), as shown in Fig. 3c. Third, DE mRNAs in EV were discovered (Fig. 3d) and integrated with ATAC-seq and RNA-seq data of GCs. Finally, 22 downregulated genes (Fig. 3e) were detected, whereas no upregulated gene was intersected (Fig. 3f). ERG, MCF2L, RFT1, KIF26B, ITIH5, MYO16, IRF4, CDH4, PRKCZ, PNPLA7, SERINC2, TLL2, MN1, PPIF, MORC4, FAM53A, KNDC1, RBFOX1, PKN1, SUSD4, THBS2, and VEGFA were among the genes that were downregulated. According to the TPM values in mRNA-seq data, five candidate genes, including RFT1, ITIH5, PPIF, MORC4, and VEGFA, were stably expressed in the GCs of patients with endometrioma. Furthermore, quantitative RT-PCR validation revealed that PPIF (p < 0.001) and VEGFA (p = 0.0148) were significantly downregulated in GCs from the patients with endometrioma. Besides that, the expression levels of ITIH5 and RTF1 tended to be lower in the samples, but the differences were not statistically significant (Fig. 3g). Correlation between candidate genes and oocyte parameters The patients’ demographic data are presented in Table S2. The expression levels of PPIF (r = 0.46, p = 0.043) and VEGFA (r = 0.45, p = 0.048) were found to be significantly associated with the total number of retrieved oocytes. In addition, both PPIF (r = 0.43, p = 0.056) and VEGFA (r = 0.43, p = 0.056) appeared to be related to the number of blastocysts, but the relationships were not significantly different (Table 1). Table 1. | Oocyte parameters | Statistics | PPIF | VEGFA | |---|---|---|---| | Total oocyte no | p-value | 0.043* | 0.048* | | Pearson r | 0.46 | 0.45 | | | Mature oocyte no | p-value | 0.08 | 0.077 | | Pearson r | 0.40 | 0.41 | | | 2PN zygote no | p-value | 0.08 | 0.082 | | Pearson r | 0.40 | 0.30 | | | Usable embryo no | p-value | 0.47 | 0.45 | | Pearson r | 0.17 | 0.18 | | | Good embryo no | p-value | 0.57 | 0.43 | | Pearson r | 0.14 | 0.19 | | | Blastocyst no | p-value | 0.056 | 0.056 | | Pearson r | 0.43 | 0.43 |

During IVF cycles, patients with endometrioma usually have poor therapeutic outcomes due to diminished oocyte quantity or quality. However, only a few studies have looked into the communication between the GCs and oocytes in patients with endometrioma. As cargo carriers, EVs can coordinate the interaction of the somatic cells and oocytes by delivering signaling molecules and cytokines [36, 37]. Our study hypothesizes that endometrioma may compromise oogenesis by impairing the GC function and its communication with oocytes. By comparing with patient with pelvic/tubal infertility, we present the chromatin accessibility profiles of GCs from patients with endometrioma, reveal its contribution to the changes in mRNA expressed in GCs and EVs, and comprehensively display the follicular transcriptomic alterations associated with compromised oogenesis. Moreover, the candidate genes were found to be associated with oocyte parameters, but further experiments will be required to validate the cause-and-effect relationship between those genes and oogenesis. The majority of detected peaks in ATAC-seq data were located within the promoter region, indicating that chromatin accessibility remodeling, as the epigenetic regulation, plays a vital role in the transcriptional process. In general, the genes with differential chromatin accessibility are linked to pathways such as adherence junction, RAP1, Hippo, Wnt, and TGF-β, which have been identified as critical pathways in modulating folliculogenesis and oogenesis [46, 47]. Notably, enriched motif sequences for differential chromatin accessibility regions were found to be associated with cellular proliferation, differentiation, and migration. C-MYC serves as a “master regulator” of cellular metabolism and proliferation, and its suppression can result in cancer cells’ growth arrest and apoptosis, similar to the compromised oogenesis experienced by patients with endometrioma [48, 49]. In addition, HIF-1a, as a key regulator for orchestrating cellular adaptation to low oxygen and nutrient-deprived environments, was also significantly enriched, suggesting that the decreased transcription of HIF-1a may lessen the stress response of GCs to the lesion [50, 51]. Additionally, SREBP1a and SREBP2 are linked to cholesterol and lipid production, implying that suppression of these two genes could affect the synthesis and homeostasis of steroidal hormones in the ovary [52, 53]. ETV4 is an essential stimulator of endometrial cancer cells because it is a principle component of estrogen signaling [54, 55]. Surprisingly, its chromatin accessibility is significantly increased in endometrioma GCs, so it is of great interest to explore the involvement of ETV4 in the pathology of endometrioma. Given the significance of GCs in folliculogenesis, the GC transcriptome is intimately linked to ovarian function. In general, downregulated genes are the most prevalent differentially expressed genes in endometrioma GCs, and their function is linked to mitochondrial function, metabolic process, and inflammatory response. In patients with endometrioma, a suppressed inflammatory response of follicles may result in irregular ovulation or luteinized unruptured follicle syndrome (LUFS) [56, 57]. The upregulated genes are primarily enriched in pathways such as organelle organization and mitotic cell-cycle pathways. Moreover, a MYC-associated zinc finger protein (MAZ) was identified as a potential TF hub for regulating differentially expressed genes, like CLCNKA and MMP9 [58, 59]. This finding is consistent with the ATAC-seq data, which revealed MYC as the highest hit of motif sequences. In the future, it will be interesting to investigate the role of MYC-related TFs in disease progression. Co-expression modules were constructed to further analyze the transcriptional regulatory network, and most genes showed co-expression patterns in modules I and III. Notably, genes in both modules were enriched in the cellular localization pathway, implying that protein localization and organelle organization may be disturbed in the GCs of patients with endometrioma. It will be interesting to explore the impact of altered protein localization and organelle organization on the communication between GCs and oocytes. In our study, the size of EVs ranges from 50 to 450 nM, which is consistent with previous findings [30, 31]. Furthermore, the spheric structure was confirmed as EV using both an electron microscope and classic molecular makers. The tendency that more genes were inhibited in EVs, despite variations in mRNA abundance, corresponds to the situation in GCs, indicating that endometrioma may get involved in impairing somatic cell-oocyte communication. Only 22 downregulated and no upregulated genes were discovered after intersecting multi-omics data. The first explanation might be that EV mRNA-seq can only detect secreted nucleic acids, limiting the number of intersected genes. Another possibility is that most DE genes are downregulated because the GCs from patients with endometrioma have lower chromatic accessibility than those from the pelvic/tubal infertility group. Importantly, these 22 downregulated genes are linked to biological processes such as extracellular matrix stabilization, organelle transportation, mitochondrial function, and cellular proliferation, implying their roles in endometrioma pathogenesis. PPIF belongs to a member of the peptidyl-prolyl cis–trans isomerase (PPIase) family, which catalyzes the cis–trans isomerization of proline imidic peptide bonds in oligopeptides and accelerates the folding of proteins. In addition, the abnormally expressed PPIF, a component of the mitochondrial permeability transition pore in the inner mitochondrial membrane, may be linked to the activation of apoptotic and necrotic cell death [60, 61]. VEGFA promotes angiogenesis by inducing the proliferation and migration of vascular endothelial cells. The VEGFA knock-out mouse model showed that VEGFA deficiency in GCs can cause subfertility by halting follicular growth and impeding ovulation, demonstrating the significance of VEGFA in GC development [62, 63]. The expression levels of PPIF and VEGFA were significantly associated with the number of retrieved oocytes and appeared to be positively related to the number of blastocysts. Additionally, ITIH5 encodes a heavy chain component of the inter-alpha-trypsin inhibitor (ITI) family members, stabilizes the extracellular matrix, and prevents tumor metastasis [64, 65]. MORC4, functioning as an ATPase, is associated with acute and chronic pancreatitis and inflammatory bowel disorders [66, 67], but the relationship between MORC4 and endometrioma is unclear. It would be intriguing to explore the relationships between candidate genes and oocyte variables with a larger sample size. This study discovers that a considerable remodeling in GC chromatin accessibility may result in transcriptomic disruption of GCs and EVs in follicular fluid, impeding somatic cell-oocyte communication and leading to compromised oogenesis in patients with endometrioma. Since the chromatin accessibility of ETV4, an essential stimulator of endometrial cancer cells, is significantly enhanced in the endometrioma GCs, it is of great interest to explore the role of ETV4 in endometrioma pathology. Moreover, because MAZ was identified as a potential TF hub for regulating cell differentiation and growth genes, it will be intriguing to investigate the role of MYC-related TFs in endometriosis progression. The candidate genes could be potential markers associated with oocyte development, although more samples are needed to validate the findings. The major strength of this study is that a multi-omics approach was utilized to present the chromatin accessibility profiles of GCs from patients with endometrioma, reveal its contribution to the changes in mRNA expressed in GCs and EVs, and comprehensively display the follicular transcriptomic alterations associated with compromised oogenesis. The limitation of this study is that a larger number of patient samples will be required to investigate the associations between the candidate genes and oocyte parameters. Additionally, differential expression in genes may not necessarily result in observable alterations in patients, so further study will be needed to validate the cause-and-effect relationship between genes and oogenesis. In comparison with the patient with pelvic/tubal infertility, a considerable remodeling in GC chromatin accessibility may result in transcriptomic disruption of GCs and EVs in follicular fluid, which may impede somatic cell-oocyte communication and lead to compromised oogenesis in patients with endometrioma. It would be intriguing to investigate the relationship between the candidate genes and oocyte clinical parameters in the future. Supplementary Information Below is the link to the electronic supplementary material.

We appreciate Epibiotek (Guangzhou, China) for performing the ATAC-seq and mRNA-seq. Funding This work was supported by the National Natural Science Foundation of China grant 82171642 (ZQX); National Natural Science Foundation of China grant 81671523 (YL); National Natural Science Foundation of China grant 32170612 (CWC); Natural Science Funding of Guangdong Province grant 4210015092 (YL); Natural Science Funding of Guangdong Province grant 2017A030313895 (YL); Funding of Yat-sen Scholarship for Young Scientist (YL); Guangdong Science and Technology Department grant 2020B121200600180. Data availability All sequencing data presented in this study can be found in Sequence Read Archive (SRA) with accession no. PRJNA806789. Code availability All analysis codes used in this study can be obtained by application to the authors. Declarations Ethics approval This study has been reviewed by the scientific research ethics committee of Sun Yat-sen Memorial Hospital and approved by the institutional review board of Sun Yat-sen Memorial Hospital (no. 201717). All experiments related to human tissue were performed in accordance with the Declaration of Helsinki. Consent to participate All the patients have signed an informed consent before donating their abandoned GCs and FF. Conflict of interest The authors declare no competing interests. Footnotes Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Songbang Ou, Xuedan Jiao and Yi Li contributed equally to this work. Contributor Information Qingxue Zhang, Email: [email protected]. Chunwei Cao, Email: [email protected]. Lina Wei, Email: [email protected].

- 1.Zondervan KT, Becker CM, Missmer SA. Endometriosis. N Engl J Med. 2020;382(13):1244–56. [DOI] [PubMed] [Google Scholar] - 2.Nnoaham KE, et al. Impact of endometriosis on quality of life and work productivity: a multicenter study across ten countries. Fertil Steril. 2011;96(2):366-373 e8. [DOI] [PMC free article] [PubMed] [Google Scholar] - 3.Yilmaz Hanege B, GulerCekic S, Ata B. Endometrioma and ovarian reserve: effects of endometriomata per se and its surgical treatment on the ovarian reserve. Facts Views Vis Obgyn. 2019;11(2):151–7. [PMC free article] [PubMed] [Google Scholar] - 4.Macer ML, Taylor HS. Endometriosis and infertility: a review of the pathogenesis and treatment of endometriosis-associated infertility. Obstet Gynecol Clin North Am. 2012;39(4):535–49. [DOI] [PMC free article] [PubMed] [Google Scholar] - 5.Kaponis A, et al. Current treatment of endometrioma. Obstet Gynecol Surv. 2015;70(3):183–95. [DOI] [PubMed] [Google Scholar] - 6.Galczynski K, et al. Ovarian endometrioma - a possible finding in adolescent girls and young women: a mini-review. J Ovarian Res. 2019;12(1):104. [DOI] [PMC free article] [PubMed] [Google Scholar] - 7.Giacomini E, et al. Characteristics of follicular fluid in ovaries with endometriomas. Eur J Obstet Gynecol Reprod Biol. 2017;209:34–8. [DOI] [PubMed] [Google Scholar] - 8.Sanchez AM, et al. The distinguishing cellular and molecular features of the endometriotic ovarian cyst: from pathophysiology to the potential endometrioma-mediated damage to the ovary. Hum Reprod Update. 2014;20(2):217–30. [DOI] [PubMed] [Google Scholar] - 9.Shah DK. Diminished ovarian reserve and endometriosis: insult upon injury. Semin Reprod Med. 2013;31(2):144–9. [DOI] [PubMed] [Google Scholar] - 10.Seyhan A, Ata B, Uncu G. The impact of endometriosis and its treatment on ovarian reserve. Semin Reprod Med. 2015;33(6):422–8. [DOI] [PubMed] [Google Scholar] - 11.Ferrero H, et al. Single-cell RNA sequencing of oocytes from ovarian endometriosis patients reveals a differential transcriptomic profile associated with lower quality. Hum Reprod. 2019;34(7):1302–12. [DOI] [PubMed] [Google Scholar] - 12.Sanchez AM, et al. Is the oocyte quality affected by endometriosis? A review of the literature. J Ovarian Res. 2017;10(1):43. [DOI] [PMC free article] [PubMed] [Google Scholar] - 13.Xu B, et al. Oocyte quality is decreased in women with minimal or mild endometriosis. Sci Rep. 2015;5:10779. [DOI] [PMC free article] [PubMed] [Google Scholar] - 14.Ben Maamar M, Nilsson EE, Skinner MK. Epigenetic transgenerational inheritance, gametogenesis and germline development dagger. Biol Reprod. 2021;105(3):570–92. [DOI] [PMC free article] [PubMed] [Google Scholar] - 15.Li Y, et al. Abnormal PIWI-interacting RNA profile and its association with the deformed extracellular matrix of oocytes from recurrent oocyte maturation arrest patients. Fertil Steril. 2021;115(5):1318–26. [DOI] [PubMed] [Google Scholar] - 16.Tomizawa S, Nowacka-Woszuk J, Kelsey G. DNA methylation establishment during oocyte growth: mechanisms and significance. Int J Dev Biol. 2012;56(10–12):867–75. [DOI] [PubMed] [Google Scholar] - 17.Sendzikaite G, Kelsey G. The role and mechanisms of DNA methylation in the oocyte. Essays Biochem. 2019;63(6):691–705. [DOI] [PMC free article] [PubMed] [Google Scholar] - 18.Sha QQ, Zhang J, Fan HY. Function and regulation of histone H3 lysine-4 methylation during oocyte meiosis and maternal-to-zygotic transition. Front Cell Dev Biol. 2020;8: 597498. [DOI] [PMC free article] [PubMed] [Google Scholar] - 19.Yu C, et al. CFP1 regulates histone H3K4 trimethylation and developmental potential in mouse oocytes. Cell Rep. 2017;20(5):1161–72. [DOI] [PubMed] [Google Scholar] - 20.Sha QQ, et al. CFP1 coordinates histone H3 lysine-4 trimethylation and meiotic cell cycle progression in mouse oocytes. Nat Commun. 2018;9(1):3477. [DOI] [PMC free article] [PubMed] [Google Scholar] - 21.Miyamoto K, et al. Chromatin accessibility impacts transcriptional reprogramming in oocytes. Cell Rep. 2018;24(2):304–11. [DOI] [PMC free article] [PubMed] [Google Scholar] - 22.Gu C, et al. Integrative single-cell analysis of transcriptome, DNA methylome and chromatin accessibility in mouse oocytes. Cell Res. 2019;29(2):110–23. [DOI] [PMC free article] [PubMed] [Google Scholar] - 23.Zhang C, et al. The chromatin remodeler Snf2h is essential for oocyte meiotic cell cycle progression. Genes Dev. 2020;34(3–4):166–78. [DOI] [PMC free article] [PubMed] [Google Scholar] - 24.Buenrostro JD, et al. Transposition of native chromatin for fast and sensitive epigenomic profiling of open chromatin, DNA-binding proteins and nucleosome position. Nat Methods. 2013;10(12):1213–8. [DOI] [PMC free article] [PubMed] [Google Scholar] - 25.Bajic M, Maher KA, Deal RB. Identification of open chromatin regions in plant genomes using ATAC-seq. Methods Mol Biol. 2018;1675:183–201. [DOI] [PMC free article] [PubMed] [Google Scholar] - 26.Buenrostro JD, et al. ATAC-seq: a method for assaying chromatin accessibility genome-wide. Curr Protoc Mol Biol. 2015;109:21 29 1-21 29 9. [DOI] [PMC free article] [PubMed] [Google Scholar] - 27.Picelli S, et al. Tn5 transposase and tagmentation procedures for massively scaled sequencing projects. Genome Res. 2014;24(12):2033–40. [DOI] [PMC free article] [PubMed] [Google Scholar] - 28.Li Z, et al. Identification of transcription factor binding sites using ATAC-seq. Genome Biol. 2019;20(1):45. [DOI] [PMC free article] [PubMed] [Google Scholar] - 29.Hendrickson DG, et al. Simultaneous profiling of DNA accessibility and gene expression dynamics with ATAC-seq and RNA-seq. Methods Mol Biol. 2018;1819:317–33. [DOI] [PubMed] [Google Scholar] - 30.Doyle LM, Wang MZ. Overview of extracellular vesicles, their origin, composition, purpose, and methods for exosome isolation and analysis. Cells. 2019;8(7):727. [DOI] [PMC free article] [PubMed] [Google Scholar] - 31.Borges FT, Reis LA, Schor N. Extracellular vesicles: structure, function, and potential clinical uses in renal diseases. Braz J Med Biol Res. 2013;46(10):824–30. [DOI] [PMC free article] [PubMed] [Google Scholar] - 32.Bebelman MP, et al. Biogenesis and function of extracellular vesicles in cancer. Pharmacol Ther. 2018;188:1–11. [DOI] [PubMed] [Google Scholar] - 33.Miranda KC, et al. Nucleic acids within urinary exosomes/microvesicles are potential biomarkers for renal disease. Kidney Int. 2010;78(2):191–9. [DOI] [PMC free article] [PubMed] [Google Scholar] - 34.Wang J, et al. Plasma exosomes as novel biomarker for the early diagnosis of gastric cancer. Cancer Biomark. 2018;21(4):805–12. [DOI] [PMC free article] [PubMed] [Google Scholar] - 35.Chen IH, et al. Phosphoproteins in extracellular vesicles as candidate markers for breast cancer. Proc Natl Acad Sci U S A. 2017;114(12):3175–80. [DOI] [PMC free article] [PubMed] [Google Scholar] - 36.Dumesic DA, et al. Oocyte environment: follicular fluid and cumulus cells are critical for oocyte health. Fertil Steril. 2015;103(2):303–16. [DOI] [PubMed] [Google Scholar] - 37.Andronico F, et al. Extracellular vesicles in human oogenesis and implantation. Int J Mol Sci. 2019;20(9):2162. [DOI] [PMC free article] [PubMed] [Google Scholar] - 38.Quinn MC, et al. Purification of granulosa cells from human ovarian follicular fluid using granulosa cell aggregates. Reprod Fertil Dev. 2006;18(5):501–8. [DOI] [PubMed] [Google Scholar] - 39.Nottola SA, et al. Ultrastructural characteristics of human granulosa cells in a coculture system for in vitro fertilization. Microsc Res Tech. 2006;69(6):508–16. [DOI] [PubMed] [Google Scholar] - 40.Alpha Scientists in Reproductive Medicine and ESHRE Special Interest Group of Embryology. The Istanbul consensus workshop on embryo assessment: proceedings of an expert meeting. Hum Reprod. 2011;26(6):1270–83. 10.1093/humrep/der037. [DOI] [PubMed] - 41.Machtinger R, Racowsky C. Morphological systems of human embryo assessment and clinical evidence. Reprod Biomed Online. 2013;26(3):210–21. [DOI] [PubMed] [Google Scholar] - 42.Yu G, Wang LG, He QY. ChIPseeker: an R/bioconductor package for ChIP peak annotation, comparison and visualization. Bioinformatics. 2015;31(14):2382–3. [DOI] [PubMed] [Google Scholar] - 43.Wu DY, et al. Identifying differential transcription factor binding in ChIP-seq. Front Genet. 2015;6:169. [DOI] [PMC free article] [PubMed] [Google Scholar] - 44.Langfelder P, Horvath S. WGCNA: an R package for weighted correlation network analysis. BMC Bioinformatics. 2008;9:559. [DOI] [PMC free article] [PubMed] [Google Scholar] - 45.Han H, et al. TRRUST: a reference database of human transcriptional regulatory interactions. Sci Rep. 2015;5:11432. [DOI] [PMC free article] [PubMed] [Google Scholar] - 46.Hsueh AJ, et al. Intraovarian control of early folliculogenesis. Endocr Rev. 2015;36(1):1–24. [DOI] [PMC free article] [PubMed] [Google Scholar] - 47.Zhang Y, et al. Transcriptome landscape of human folliculogenesis reveals oocyte and granulosa cell interactions. Mol Cell. 2018;72(6):1021-1034 e4. [DOI] [PubMed] [Google Scholar] - 48.Ahmadi SE, et al. MYC: a multipurpose oncogene with prognostic and therapeutic implications in blood malignancies. J Hematol Oncol. 2021;14(1):121. [DOI] [PMC free article] [PubMed] [Google Scholar] - 49.Xianjun F, et al. Momordin Ic induces G0/1 phase arrest and apoptosis in colon cancer cells by suppressing SENP1/c-MYC signaling pathway. J Pharmacol Sci. 2021;146(4):249–58. [DOI] [PubMed] [Google Scholar] - 50.Wen L, et al. Phenylethanol glycosides from herba cistanche improve the hypoxic tumor microenvironment and enhance the effects of oxaliplatin via the HIF1alpha signaling pathway. Mol Med Rep. 2021;24(1):1. [DOI] [PMC free article] [PubMed] [Google Scholar] - 51.Song H, et al. HIF-1alpha-mediated telomerase reverse transcriptase activation inducing autophagy through mammalian target of rapamycin promotes papillary thyroid carcinoma progression during hypoxia stress. Thyroid. 2021;31(2):233–46. [DOI] [PubMed] [Google Scholar] - 52.Sanchez-Alvarez M, et al. Signaling networks converge on TORC1-SREBP activity to promote endoplasmic reticulum homeostasis. PLoS ONE. 2014;9(7): e101164. [DOI] [PMC free article] [PubMed] [Google Scholar] - 53.Xie XJ, et al. CDK8-Cyclin C mediates nutritional regulation of developmental transitions through the ecdysone receptor in Drosophila. PLoS Biol. 2015;13(7): e1002207. [DOI] [PMC free article] [PubMed] [Google Scholar] - 54.Rodriguez AC, et al. ETV4 is necessary for estrogen signaling and growth in endometrial cancer cells. Cancer Res. 2020;80(6):1234–45. [DOI] [PMC free article] [PubMed] [Google Scholar] - 55.Fonseca AS, et al. ETV4 plays a role on the primary events during the adenoma-adenocarcinoma progression in colorectal cancer. BMC Cancer. 2021;21(1):207. [DOI] [PMC free article] [PubMed] [Google Scholar] - 56.Holtz G, et al. Luteinized unruptured follicle syndrome in mild endometriosis. Assessment with biochemical parameters. J Reprod Med. 1985;30(9):643–5. [PubMed] [Google Scholar] - 57.Mio Y, et al. Luteinized unruptured follicle in the early stages of endometriosis as a cause of unexplained infertility. Am J Obstet Gynecol. 1992;167(1):271–3. [DOI] [PubMed] [Google Scholar] - 58.Xiao T, Li X, Felsenfeld G. The Myc-associated zinc finger protein (MAZ) works together with CTCF to control cohesin positioning and genome organization. Proc Natl Acad Sci U S A. 2021;118(7):e2023127118. [DOI] [PMC free article] [PubMed] [Google Scholar] - 59.Luo W, et al. MYC associated zinc finger protein promotes the invasion and metastasis of hepatocellular carcinoma by inducing epithelial mesenchymal transition. Oncotarget. 2016;7(52):86420–32. [DOI] [PMC free article] [PubMed] [Google Scholar] - 60.Javadov S, et al. Mitochondrial permeability transition in cardiac ischemia-reperfusion: whether cyclophilin D is a viable target for cardioprotection? Cell Mol Life Sci. 2017;74(15):2795–813. [DOI] [PMC free article] [PubMed] [Google Scholar] - 61.Xiao A, et al. The cyclophilin D/Drp1 axis regulates mitochondrial fission contributing to oxidative stress-induced mitochondrial dysfunctions in SH-SY5Y cells. Biochem Biophys Res Commun. 2017;483(1):765–71. [DOI] [PubMed] [Google Scholar] - 62.Sargent KM, et al. Loss of vascular endothelial growth factor A (VEGFA) isoforms in granulosa cells using pDmrt-1-Cre or Amhr2-Cre reduces fertility by arresting follicular development and by reducing litter size in female mice. PLoS ONE. 2015;10(2): e0116332. [DOI] [PMC free article] [PubMed] [Google Scholar] - 63.Rico C, et al. HIF1 activity in granulosa cells is required for FSH-regulated Vegfa expression and follicle survival in mice. Biol Reprod. 2014;90(6):135. [DOI] [PubMed] [Google Scholar] - 64.Rose M, et al. The ECM modulator ITIH5 affects cell adhesion, motility and chemotherapeutic response of basal/squamous-like (BASQ) bladder cancer cells. Cells. 2021;10(5):1038. [DOI] [PMC free article] [PubMed] [Google Scholar] - 65.Liu J, et al. ITIH5, a p53-responsive gene, inhibits the growth and metastasis of melanoma cells by downregulating the transcriptional activity of KLF4. Cell Death Dis. 2021;12(5):438. [DOI] [PMC free article] [PubMed] [Google Scholar] - 66.Yang Z, et al. MORC4 is a novel breast cancer oncogene regulated by miR-193b-3p. J Cell Biochem. 2019;120(3):4634–43. [DOI] [PubMed] [Google Scholar] - 67.Tencer AH, et al. Molecular mechanism of the MORC4 ATPase activation. Nat Commun. 2020;11(1):5466. [DOI] [PMC free article] [PubMed] [Google Scholar] Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Supplementary Materials Data Availability Statement All sequencing data presented in this study can be found in Sequence Read Archive (SRA) with accession no. PRJNA806789. All analysis codes used in this study can be obtained by application to the authors.