FormalPara What does this study add to the clinical work

This systematic review on the main genes related to the PCOS physiopathology lead to a deeper understanding of the PCOS pathogenesis, and have the potential to orientate more precise diagnoses, and also help to establish more effective treatment protocols.

Introduction

Polycystic ovary syndrome (PCOS) is an endocrinological disease that affects 6–15% of women of reproductive age [1]. This syndrome has a combination of symptoms, such as hyperandrogenism, menstrual irregularities, metabolic syndrome, infertility, acne, and obesity [2]. According to the Rotterdam Consensus Workshop criteria for PCOS [3], it is necessary to identify at least two of the following characteristics: (i) clinical or biochemical signs of hyperandrogenism, (ii) oligo or anovulation, and (iii) presence of polycystic ovaries.

In addition to the heterogeneity of clinical signs and symptoms, PCOS has an etiology that is not fully understood. Thus, the study of this disease is quite challenging [1], being the object of an increasing number of studies over the years (Fig. 1). Among the possible causes of the development of PCOS, the excess production of androgens and insulin resistance have been identified as the main factors in the etiology of the disease [4].

Fig. 1
figure 1

Survey of the number of articles on polycystic ovary syndrome (PCOS), using the MEDLINE database. Over the years, it is possible to notice the increase in the number of articles on PCOS, pointing to a greater number of researches, knowledge and relevance of the subject. This chart is created in late June 2022

Full size image

Genetic factors also play an important role in the etiology of the disease, as alterations in gene transcription or genetic polymorphisms can cause serious transcriptional alterations related to PCOS. According to Ajmal et al. [5], the genes that encode androgen receptors, luteinizing hormone (LH), follicle-stimulating hormone (FSH), and leptin receptors are the most likely to be involved in the pathophysiology of the disease.

Epigenetics, which deals with processes associated with changes in the expression pattern of genes without changing the DNA nucleotide sequence [6], is a branch of genetics that is increasingly associated with the pathogenesis of PCOS [7]. Epigenetic processes include DNA cytosine methylation, modifications of histone proteins present in the nucleosome, and mechanisms mediated by noncoding RNA [8]. DNA methylation involves the addition of a methyl group on carbon 5 of cytosine through the action of DNA methyltransferases [9, 10]. Much of this methylation occurs at CpG sites, which are groups of dinucleotides, resulting in chromatin condensation. Therefore, hypermethylated DNA regions hinder gene transcription and cause gene silencing [9]. In histones, several covalent modifications can occur, such as acetylation, methylation, phosphorylation, and ubiquitination [10, 11], which change the conformation and accessibility of chromatin in different ways [11]. Noncoding RNAs, in turn, are transcribed from RNAs that do not code for proteins but can, for example, interact with histone-modifying complexes or DNA methyltransferases to regulate gene expression [12].

In addition to being current, the relationship between epigenetics and PCOS is quite relevant, as changes in gene expression can generate important phenotypic changes, such as hyperandrogenism [7, 13,14,15]. Thus, the discovery and investigation of genes undergoing epigenetic alterations in tissues affected by the pathology may lead to more effective therapies for the treatment of women with PCOS [7, 13,14,15]. However, given the clinical heterogeneity of PCOS associated with the complex gene expression pathways involved in this disease, there are still gaps to be filled regarding its etiology. Therefore, this study aimed to review the main genes involved in the pathophysiology of PCOS and DNA methylation associated with the expression of these genes.

Methods

This study is a systematic review that used the updated Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) recommendation [16] (Fig. 2). The study's guiding question was: “What is the influence of DNA methylation on the main genes involved in PCOS?”.

Fig. 2
figure 2

Flowchart of the systematic review of manuscripts related to epigenetic alterations of genes involved in polycystic ovary syndrome (PCOS), based on the PRISMA recommendation. The records are identified by searching the MEDLINE database of articles published until December 2021. After applying the exclusion criteria, 77 manuscripts are selected. After a complete reading of 43 articles, 23 articles are in accordance with the study proposal and, therefore, are included in this review

Full size image

The search was carried out until June 2022 in the Medical Literature Analysis and Retrieval System Online (MEDLINE) database. The descriptors used for the search were (polycystic ovary syndrome) AND (genes) AND (epigenetics) based on the MeSH descriptors.

We included papers that met our inclusion criteria: primary and original articles, written in English, found through our entry terms. All papers included must have been published until June 2022. Our exclusion criteria were papers written in other language than English, literature reviews, and articles that used animals to perform the experiments or that focused in other epigenetics mechanism instead of DNA methylation.

For data screening and extraction, a table was filled with the quantification of the following data: author, year of publication, cited genes, methods, objectives, main genes addressed in each study, and type of epigenetic alteration involved in the expression of these genes when epigenetically altered. To be included for data screening and extraction, the paper must have analyzed the DNA of women with and without PCOS. Our primary outcome measure was the genes differentially expressed in these two groups, and this difference had to be explained by the methylation levels. Then, articles were grouped based on the similarity of the genes and their metabolic role. The data selection and extraction were done and reviewed by two authors independently. Subsequently, a third author reviewed the results and pointed out suggestions. Only the papers that met the inclusion criteria were added to this systematic review, and any inter-researched disagreement was resolved among the authors. The risk of bias in included papers was assessed by the Newcastle–Ottawa Scale (NOS) [17], with modifications (Supplementary Table 1). The NOS measures the quality of nonrandomized studies based on the selection of the study groups, the comparability of the groups, and the ascertainment of either the exposure or outcome of interest for case–control or cohort studies, respectively, to be used in a systematic review [17]. This scale allowed us to evaluate all the articles with the same tool, as not all of them were case–control studies. The criteria adopted were: (1) adequate definition of cases; (2) selection of controls; (3) control for important factor; (4) explicit DNA tissue extracted reported; and (5) significant statistic difference between PCOS vs control for DNA methylation levels.

Results

In total, we identified 77 articles. After screening based on the title or abstract, 34 studies were excluded. Among these, 19 were reviews, five focused on miRNA modifications, and 10 did not focus on epigenetic modifications. Forty-three articles were read in full, and 20 papers were excluded because studied epigenetic alterations in a non-human population. Finally, 23 were eligible to answer the central question of this review. We summarized this process in a flow-diagram (Fig. 2). The studies appraisal was qualitatively done by stratifying methodological characteristics of each study, i.e., the type of study, the size of the included population, the presence of a control group, the genes involved, and their methylation and expression levels. Then, the quality of the assessment of the included studies was measured by NOS (Supplementary Table 1).

Among the included articles, the main genes affected by PCOS were identified and are detailed in Table 1. The genes for androgen receptors and those related to the regulation of ovulation and metabolism were the most recurrent.

Table 1 List of the main genes affected in polycystic ovary syndrome (PCOS), according to a literature review
Full size table

Of the 23 selected articles, 18 confirmed the epigenetic influence on PCOS-related genes. Table 2 lists those with greater relevance to the epigenetic alterations described by the authors, with genes identified in more than one study. The epigenetic alterations presented were only in DNA, specifically CpG sites (Fig. 3), which could be hypomethylated or hypermethylated depending on the gene. Therefore, no alterations in histones were observed. In the other articles (n = 5), although the direct influence of epigenetics on the genes involved in PCOS was not found, no work contradicted the existence of this influence [18,19,20,21,22]. Among the articles that confirmed the relationship between epigenetics and PCOS genes, ten articles identified epigenetic alterations in genes related to insulin resistance [15, 23,24,25,26,27,28,29,30,31]. In total, 14 genes were identified: growth hormone releasing hormone receptor (GHRHR), peroxisome proliferator activated receptor gamma (PPARG), resistin (RETN), nicotinamide phosphoribosyltransferase (NAMPT), brain derived neurotrophic factor (BDNF), insulin receptor substrate 1 (IRS1), paired box 6 (PAX6), insulin receptor (INSR), G protein subunit alpha 11 (GNA11), MLX interacting protein like (MLXIPL), syntaxin binding protein 5L (STXBP5L), leptin (LEP), estrogen receptor 1 (ESR1), and lysophosphatidylcholine acyltransferase 1 (LPCAT1). Other studies also identified genes related to hyperandrogenism: luteinizing hormone/choriogonadotropin receptor (LHCGR), CL2 interacting protein 3 (BNIP3), GHRHR, and tumor necrosis factor (TNF) [23, 28, 32, 33].

Table 2 Epigenetic alterations of genes were found in women with polycystic ovary syndrome (PCOS), according to a literature review
Full size table
Fig. 3
figure 3

DNA methylation. According to epigenetic modifications, regions of chromatin can be transcriptionally silenced (chromatin condensation) (A) or activated (chromatin decondensation) (B). The genes involved in PCOS present epigenetic modifications in DNA, leading to its hypermethylation (A) or hypomethylation (B), which are involved in gene silencing and activation, respectively. DNA methylation occurs through the conversion of cytosine to 5-methylcytosine through DNA methyltransferase (DNMT). Created by BioRender.com

Full size image

Discussion

Given the high number of genes cited in the articles, many authors have not discussed their relationship with PCOS in depth. Wang et al. [36], for example, analyzed an extensive number of genes, 54 in all, to test their correlation with PCOS. However, not all genes showed a direct relationship between methylation and regulation of gene expression or function, or their function had not been well elucidated by methylation [36].

One of the main genes reported was the INSR gene. The promoter regions of the INSR gene were reported by Yu et al. [24] as hypomethylated in the ovarian tissue of women with PCOS compared with women without PCOS, resulting in a greater expression of the INSR gene. Jones et al. [23], in turn, verified that the INSR gene was expressed more in cumulus cells of ovarian follicles of obese women with PCOS than in non-obese women with PCOS. They presented another relevant finding that the INSR gene is hypermethylated, i.e., less expressed in metabolic tissues, such as skeletal muscle tissue, of obese women with PCOS [23]. This finding was used to confirm the existing theory of selective insulin sensitivity, since ovarian tissue is not resistant to insulin while skeletal muscle is [23].

Of the other genes that also underwent epigenetic alterations and were related to PCOS and insulin resistance, the ESR1 gene, an androgen receptor, was found to be hypomethylated in women with PCOS compared to the control group (women without PCOS), which was overexpressed [25]. The authors associated overexpression of the ESR1 gene with overexpression of lipid kinases related to the development of insulin resistance, which may be a possible explanation for the glucotoxic environment in women with PCOS [25]. Another important gene, IRS1, which is a gene that plays a central role in insulin signaling and a relationship with type 2 diabetes mellitus, was shown to be altered based on the body mass index of women [15]. The authors showed that overweight or obese women with PCOS had reduced IRS1 expression compared to others in the cohort [15]. Similarly, another gene directly correlated with type 2 diabetes mellitus, PPARG, one of the targets of pharmacological drugs used to control plasma glucose levels, was found to be hypermethylated and, therefore, less expressed in women with PCOS [15]. When studying the PPARG coactivator 1 alpha (PPARGC1A), a PPARG coactivator [37], Zhao et al. observed that women with PCOS also had this hypermethylated gene compared to healthy control women [38], indicating an epigenetic orchestration of genes with integrated functions.

A review of the articles also pointed to epigenetic alterations in genes related to the regulation of ovulation, such as the LHCGR gene, which encodes the LH and chorionic gonadotropin (CG) receptor. In two studies, elevated expression of this gene was found in women diagnosed with PCOS compared to women without PCOS due to its hypomethylation [23, 32].

Other studies have highlighted epigenetic alterations in the RAB5B, member RAS oncogene family (RAB5B), which encodes a member of the Ras-related GTPase superfamily responsible for the transport of intracellular vesicles and endosome formation related to diseases, such as obesity and type 1 diabetes mellitus [15, 23]. Jones et al. [23] reported that RAB5B was much less expressed in women with PCOS than in the control group of healthy women. Consistent with these results, Kokosar et al. [15] showed significantly lower mRNA expression of the RAB5B gene in the adipose tissue of women with PCOS compared to healthy women without PCOS.

Other genes that have undergone epigenetic alterations are BNIP3, GHRHR, and TNF, which have been correlated with the appearance of the clinical characteristics of hyperandrogenism [28, 33]. A study of BNIP3 gene, responsible for participating in the metabolism of lipid precursors in the biosynthesis of androgens, showed that hypomethylation of this gene's promoter correlated with a higher expression of BNIP3 in women with PCOS [33]. However, the discussion of the study contradicts the results presented. The authors state that a lower expression of BNIP3, related to gene hypermethylation instead of a higher expression, probably results in an excess of lipids, which contributes to hyperandrogenism [33]. Sagvekar et al. [28] observed hypomethylation and, therefore, an overexpression of GHRHR, responsible for regulating the release of somatotropin (GH) in the ovary, in granulosa cells of women with PCOS. According to the authors, this increased expression may be an indirect mediator of androgen excess in PCOS, as high levels of GH increase the sensitivity of developing ovarian follicles to gonadotropins [39]. Sagvekar et al. [28] also related the hypermethylation of the TNF gene, whose protein is responsible for suppressing the expression of LHCGR induced by FSH [40], with an indirect contribution of hyperandrogenism in PCOS [28].

In addition, Jiao et al. also reported the possibility that women with PCOS are more prone to developing cancer [41]. Patients with irregular menstruation and PCOS generally have hypomethylated global DNA in their ovarian tissues, a common feature in cancer tissues [41]. According to the authors, hormone levels in an irregular menstrual cycle are atypical, which may be a starting point for future studies that correlate cancer development in women with PCOS and hormonal changes. In the same study, the BRCA1 DNA repair associated (BRCA1) was altered in the ovarian tissues of women with PCOS at three local points (c154C, c1337G, and c2566T) [41]. It was not possible to relate these alterations to ovarian cancer, but possibly to the progression of breast cancer [41], which is an extremely relevant biomarker for future studies that have a relationship between the development of PCOS and breast cancer [41].

Upon reviewing the articles, it can be stated that epigenetic DNA methylation pathways affect the expression of the main genes involved in the etiology of PCOS. Among the various genes reported, INSR, LHCGR, and RAB5B were identified as fundamental to understanding the disease. The INSR and LHCGR genes were hypomethylated, while the RAB5B gene was hypermethylated in women with PCOS.