Endometriosis is a common gynecologic disease with an estimated frequency of 5–10% among the female population of reproductive age. This disorder is characterized by inflammation, but the pathogenesis of the disease remains unclear. A stepwise process of accumulation of genetic mutations or epigenetic alterations may contribute to the development of endometriosis under the influence of environmental factors such as inflammatory responses (1). Endometriosis is reported to be associated with the frequent up- and downregulation of disease susceptibility genes in several pathways including cytokines, inflammation, immune, oxidative stress, detoxification, hormone receptors, metabolism, matrix remodeling, adhesion molecules, growth factors, cell cycle regulation, signaling, oncogenes, and transcriptional regulation (2). The marked regulation of disease may be associated with genetic alterations (1). The common mechanisms leading to loss- or gain-of-function have a causal relationship to genomic instability, including microsatellite instability (MSI), chromosomal instability (CIN), loss of heterozygosity (LOH), single-nucleotide polymorphism (SNP), gene mutations, and mitochondrial DNA (mtDNA) mutations (3,4). In addition, evidence has emerged indicating that a specific gene has been shown to regulate its neighbor genes by epigenetic mechanisms (5). Insights have emerged from various lines of evidence, including that endometriosis may be an epigenetic disease (6).

This review focused on the relationship between genomic instability, gene mutations and epigenetic alterations associated with increased risk of the development of endometriosis.

A computerized literature search was performed to identify relevant studies reported in the English language. We searched MEDLINE electronic databases (http://www.ncbi.nlm.nih.gov/sites/entrez) published between January 1966 and December 2013, combining the keywords ‘endometriosis’, ‘genetic’, ‘epigenetic’, and ‘environment’. Various combinations of the terms were used, depending on the database searched. Each gene was also linked to the NCBI Entrez Gene pages (http://www.ncbi.nlm.nih.gov/sites/entrez). In addition, references in each article were searched to identify potentially missed studies.

Transtubal retrograde flow and hemolysis occurring during menstruation result in the accumulation of high levels of pro-oxidant factors, such as heme and iron, into the peritoneal cavity (7). Free heme catalyzes oxidative reactions, but impairs lipid bilayers of mitochondria. Mitochondria are a major source for reactive oxygen species (ROS) production. Heme also acts as a pro-inflammatory molecule, leading to cellular injury and DNA damage (2). Furthermore, iron overload initiates a Fenton chemical reaction, which causes oxidative stress and large-scale genomic alterations (2). Persistent oxidative stress induces destruction of the peritoneal mesothelium followed by the overexpression of inflammatory cytokines and adhesion molecules for ectopic endometrial cells, which may cause the development of endometriosis (7). Increased ROS generation, secondary to the influx of heme and iron during retrograde menstruation, have been found in endometriosis (8,9). Oxidative signals from the microenvironment may be required for the development, maintenance and progression of endometriotic precursor lesions. Therefore, endometriosis has been considered to be associated with a chronic inflammatory state leading to pro-inflammatory cytokine excess by oxidative stress (10).

It is well known that the fine-tuning of alterations in oxidant and antioxidant pathways as well as endogenous redox regulators have been reported in the endometrium, serum, or peritoneal fluid (8). Antioxidant defense enzymes provide protection against oxidative DNA damage from carcinogen-specific mutations. It is likely that endometriosis is inherited in a polygenic manner (11). Therefore, inherited sequence variations in specific genes that encode inflammatory and antioxidant defense proteins may alter disease predisposition and thus individual susceptibility to endometriosis.

Endometriosis is a benign disease, however, it shares some features with malignancy, and has been associated with an increased risk of malignant tumors, including epithelial ovarian carcinomas (endometrioid adenocarcinoma and clear cell carcinoma), other Müllerian-type tumors (Müllerian-type mucinous borderline tumor and serous borderline tumor) and sarcomas such as adenosarcoma and endometrial stromal sarcoma in the female pelvic cavity (12–14). The precise cellular and molecular mechanisms leading to endometriosis-associated ovarian carcinogenesis recently became more evident (15–19). Excessive iron accumulation in the pelvic cavity or endometriotic cysts leads to increased oxidative stress and inflammation. Abundant iron-induced ROS is thought to be mutagenic, and chronic exposure of ectopic endometrium to this microenvironment facilitates the accumulation of somatic mutations, which can cause non-regulated mitotic division, growth and migration, similar to malignant mechanisms (3,10,18). This microenvironment is a possible link between endometriosis and tumor development (3). Environment-gene interactions may persistently occur in endometriosis as well as in endometriosis-related carcinogenesis.

Endometriosis is considered a genetic disease. This disorder is aggregated in families and individuals with an affected twin, suggesting that some subjects may have a genetic predisposition to developing endometriosis (37). Genetic background inherited from parents may confer susceptibility to endometriosis. The SNP represents a variation in the DNA sequence when a single nucleotide differs in an individual. Genome-wide SNP analysis data (37) have provided valuable insights into the genetic basis of complex traits to identify common and rare variants underlying complex diseases.

Specific SNP alterations of genes and their highly interconnected genes have been previously identified (Table I). Table I shows genetic polymorphisms and their haplotype in selected functional category lists. The genetic polymorphisms in each gene significantly appear to differ in relation to endometriosis risk. The biological categories include inflammation and immune response, oxidative stress and detoxification, hormone receptors and metabolism, matrix remodeling, adhesion molecules, growth factors, cell cycle regulation, signaling and oncogenes, transcriptional regulation, human leukocyte antigens and microRNA regulation (Tables I and II). Endometriosis undergoes a variety of molecular changes depending on the ability to survive, attach, grow, and invade. Many molecular events involved in endometriosis pathogenesis contribute to its development and maintenance. Genes in the category of inflammation showed that endometriosis is characterized by an imbalance between the oxidative and antioxidative arms of the inflammatory system, resulting in the over production of proinflammatory cytokines, oxidative stress and detoxification molecules (7,8). Since the results of some genes have been inconsistent, genetic polymorphism data are considered to be limited and conflicting. A majority of association studies are based on very simple models including one SNP or haplotype and small sample sizes. Thus, the evidence of an association between genetic polymorphisms in a single gene and endometriosis risk may be weak.

Epithelial ovarian carcinomas have been divided into at least five subgroups: high-grade serous, endometrioid, clear cell, mucinous, and low-grade serous (38). Endometriosis is associated with an increased risk of developing ovarian endometrioid and clear cell carcinoma (14). In this section, we mainly focus on genetic alterations of atypical endometriosis and endometriosis-associated ovarian carcinomas (EAOC). EAOC carcinogenesis generally follows a gradual and stepwise accumulation of genetic mutations under the influence of chronic inflammation and hyperestrogenism for clear cell carcinoma and endometrioid adenocarcinoma, respectively (13). We describe specific genetic/genomic alterations that are aberrantly expressed in solitary endometriosis, endometriosis distant from ovarian carcinomas, contiguous endometriosis associated with ovarian carcinomas, and ovarian carcinomas. Endometriosis susceptibility genes are defined as specific genes with a higher frequency of chromosomal aberrations, somatic gain- or loss-of-function mutations, or hypermethylation in solitary endometriosis than in normal endometrium. These genes are highly sensitive to microenvironmental changes, particularly to alterations in the inflammatory milieu, and in the induction of (epi)genomic changes in endometriosis precursor lesions, which may eventually lead to endometriosis. These genes may only be involved in the development and maintenance of endometriosis. Genes responsible for tumor promotion are dervied from genes with further aberrations required for promotion to a premalignant state in contiguous endometriosis associated with ovarian carcinomas than in endometriosis distant from ovarian carcinomas or solitary endometriosis. Genes responsible for the malignant transformation of endometriosis and cancer progression are involved in malignant transformation in ovarian carcinomas as compared to contiguous endometriosis associated with ovarian carcinomas. Tables III and IV show (epi)genetically relevant information of endometriosis susceptibility genes, genes responsible for tumor promotion and genes responsible for the malignant transformation of endometriosis and cancer progression.

Somatic mitochondrial DNA (mtDNA) mutations have been regarded as a hallmark of neoplasms and chronic inflammatory diseases such as aging, neurodegenerative disease and endometriosis (77). mtDNA is highly vulnerable to mutagenesis through the production of ROS. Specific mtDNA mutations also increase ROS overproduction and enhance tumor progression. Several types of mtDNA alterations, including point mutations, deletions, insertions and copy number changes, have been associated with carcinogenesis (78). Findings of previous studies have demonstrated the possible association between mtDNA polymorphisms and susceptibility to endometriosis, including A189G, A13603G, 310C insertion, T16189C polymorphisms, 189G/310TC/16189C haplotype, and 5,335-bp deletion (77,79–81). Therefore, mtDNA genetic alterations may exhibit risk of endometriosis development. No evidence has emerged indicating that these mtDNA mutations are functional and pathogenic.

Beyond genetic/genomic alterations, the development of endometriosis is also influenced by epigenetic mechanisms. Accumulating evidence suggest various epigenetic aberrations in endometriosis (6,82). Epigenetic alterations reported in endometriosis thus far include the genomic DNA methylation of progesterone receptor (PGR)-B, E-cadherin (CDH1), homeobox A10 (HOXA10) (83), estrogen receptor-β (ESR2), aromatase (CYP19A1) (84), histone deacetylase inhibition (HDACi) (82), CDKN2A/B (85), IGFBP-1 (83), leukemia inhibitory factor (LIF) (83) and DNA-methyltransferase (DNMTs) (86). Downregulated genes are associated with embryogenesis (the downstream targets of HOXA10), growth factors (IGF and IGFBP) and immuno-endocrine behavior [prolactin (PRL)], interleukin-11 (IL-11), leukemia inhibitory factor (LIF), transforming growth factor (TGF)-β, FK506 binding protein 4, 59 kDa (FKBP4), cyclooxygenase (COX)-2, prostaglandins (PGs), forkhead box O1 (FOXO1) and CCAAT/enhancer binding protein (C/EBP), β (C/EBPβ) (6,82,84–94). Target genes are important for the embryogenesis and decidualization process, which includes hormonal regulation, cytokine expression and transcription factors (89,95). A previous study has shed new light on the overlapping epigenetic signatures between the development of endometriosis and insufficient decidualization process (89). Large-scale epigenetic silencing of decidualization-related genes might play important roles in the development of endometriosis (95).

Endometriosis has been successfully identified a novel gene-environment interaction (96). Previous studies have described a positive relationship and inverse association between endometriosis risk and social, environmental and biological factors, as well as their interactions (97–107). Factors contributing to an increased risk are low birth weight, a multiple gestation, exposure to diethylstilbestrol in utero (97), overweight during late childhood (106), level of indoor exposure to passive smoking during childhood, experiencing food deprivation during World War II, walking activity at 8–15 years of age, exposure to pet animals, living in a farm for ≥3 consecutive months during childhood (100), a flight attendant, service station attendant, or health worker, a nurse (105), night shift work (107), alcohol consumption (98), cutaneous melanoma, skin sensitivity to sun exposure, nevi, freckles (108), pigmentary traits, family history of melanoma, periodontal disease (107), and caesarean section (99). On the other hand, a decreased risk of endometriosis is associated with the factors such as menarcheal age (100), increasing body size during childhood and early adulthood (102), body mass index, long-chain omega-3 fatty acid consumption (101), and in utero cigarette exposure (108). Results of those studies suggest that specific adverse exposures throughout fetal life, in early life, or during childhood or adolescence may influence the risk of endometriosis (100). Evidence of endometriosis risk of dioxin is not sufficient and remains limited (103). Based on insufficient data, it is currently not clear whether each factor is a true characteristic of women who develop endometriosis (106).

Endometriosis is a chronic inflammatory disease with genetic, epigenetic and environmental background (109). Firstly, independent analysis of many cohorts have suggested genetic/epigenetic alterations such as SNPs, copy number variation, loss of heterozygosity, and promoter methylation on the development of endometriosis (Table I). Polymorphic variants of the specific alleles were found to exhibit a significant positive or inverse association between the risk for endometriosis compared to the controls. Genome-wide gene expression profiling studies (88) showed that differentially regulated (ectopic-to-eutopic) genes in endometriosis were classified into several functional categories, including inflammation and immune response, cell cycle regulation, cytokine and growth factor signaling, endocrine function, matrix remodeling, cell adhesion, DNA damage and detoxification, regulation of glucose and lipid metabolism and transcription factors (88). These data allow us to hypothesize that the previously reported endometriosis susceptibility genes (88) tend to overlap those with genetic polymorphisms analyzed in this study (Tables I and II).

Many susceptibility genes have been reported as candidate genes for the development of endometriosis. However, a majority of genes are not key drivers of somatic expansion, but likely candidate modifiers that bridge inflammation, detoxification, growth and immune escape to license eutopic and ectopic outgrowth. Epistatic modifier genes are known to participate in a wide range of essential processes: one such mechanism is inflammation and oxidative stress (2,7,8,10,110,111). This finding supports the previous hypothesis that iron-induced oxidative stress and detoxification seems to play a key role in the development of endometriosis (2). Many modifier genes are considered to complement the actions of causative genes and play a significant role in variable phenotypic expression of the disease. Although genetic alterations inherited from parents confers susceptibility to endometriosis, wide variations in the penetrance of gene mutations may reflect the genetic background of the phenotypic diversity. Variable penetrance reflects the action of modifier genes. Even particular mutations or their variant transcripts associated with disease onset may fail to cause endometriosis, due to reduced or incomplete penetrance. Despite the identification of mutations associated with the development of endometriosis, the precise functional genetic alterations remain poorly understood.

Secondly, the epigenetic disruption of gene expression also plays an important role in the development of endometriosis through interaction with environmental changes. The ‘thrifty phenotype hypothesis’ demonstrated that maternal diet during fetal development has many epigenetic implications, which affect the offspring’s risk factors for obesity during childhood and adulthood, and even in subsequent generations (112). Similar adverse effects may be seen in other aspects of biological functions such as endometriosis. Low birthweight and multiple pregnancy are associated with subsequent endometriosis risk (97). Specific adverse environmental exposures in fetal and neonatal life, in childhood or adolescence may influence the risk of endometriosis (100). A recent study showed that environmental changes in utero such as maternal dietary energy intakes or prenatal exposures induce altered epigenetic regulation in the offspring affecting the expression of specific modifier genes that are mainly associated with endometrial decidualization processes (89,95). Epigenetic alterations may be associated with altered tissue function in fetal endometrium and influence later-life disease. The developmental origins of health and disease (DOHaD) approach may be used to elucidate the pathogenesis and epigenetic alterations of endometriosis. If gene mutations associated with endometrial decidualization are susceptible to epigenetic alterations, they have subsequent effects on disease mechanisms, such as impaired decidualization and endometriosis (89,95). The demonstration of such a sequence of genetic and epigenetic events has been shown for disease processes such as obesity, metabolic syndrome and type 2 diabetes, cardiovascular disease, cancer and possibly endometriosis. Gene-environment interactions can promote the acquisition of epigenetic alterations, genetic mutations and a different profile of gene expression. However, the precise link between epigenetics and disease is missing. The regulation of these processes in which the individuals more predisposed to endometriosis remain to be elucidated.

Mounting evidence suggests that women with endometriosis have a higher risk for ovarian cancer. In this study, ovarian cancer susceptibility genes have been defined as candidate genes responsible for malignant transformation (from endometriosis to atypical endometriosis) and candidate genes responsible for cancer progression (from atypical endometriosis to ovarian cancer) (Tables III and IV). A majority of genes function as genes responsible for malignant transformation.

Environmental factors including iron, redox and inflammatory modifications may originate from retrograde menstruation and accumulate in endometriotic lesions. Iron is an extremely reactive transition metal and generates hydroxyl radicals via a Fenton reaction (2,3,7–10,18). It is well known that iron is involved in a wide range of oxidative stress, and iron accumulation introduces point mutations as well as DNA single and double-strand breaks (110). Iron overload can also cause genetic and epigenetic changes, including DNA hypermethylation and chromatin remodeling, which lead to genomic instability and a significant increase in cancer risk (2,109,110,113). Iron contributes to carcinogenesis via three major processes: step one, by generating iron-mediated oxidative stress (genetic/epigenetic changes); step two, by promoting DNA mutagenesis, histone modification, chromatin remodeling (EAOC initiation); and step three, by enhancing genome instability (cancer promotion and progression) (111).

In conclusion, genetic and genomic factors have been unable to explain the full etiology of endometriosis. It is tempting to hypothesis that there are at least three distinct phases of the development of endometriosis: the initial wave of genetic background inherited from parents; followed by epigenetic modifications in the female offspring; and the iron overload, which is subject to dynamic modulation later in life. Stress in utero or during adolescence may compromise the future oxidative stress response to an iron insult. The present study may provide new insights into the potential mechanisms by which microenvironmental changes such as iron overload induces endometriosis and enhances endometriosis-associated carcinogenesis. Future investigations should focus on how such epigenetic changes continue to regulate risk of endometriosis from infancy through to adulthood. For example, hypermethylation of the decidualization-related genes in fetal life may cause a decrease in expression, and have a direct impact on uterine endometrial functions such as decidualization, thus influencing risk of endometriosis and infertility later in life. Of note, specific (epi)genetic signatures have led to emerging efforts to apply the knowledge to early detection, diagnosis and development of molecularly targeted therapy.