Rethinking the pathogenesis of endometriosis: Complex interactions of genomic, epigenetic, and environmental factors

Aim Endometriosis is a complex, multifactorial disease. Recent advances in molecular biology underscore that somatic mutations within the epithelial component of the normal endometrium, alongside aberrant epigenetic alterations within endometrial stromal cells, may serve as stimulators for the proliferation of endometriotic tissue within the peritoneal cavity. Nevertheless, pivotal inquiries persist: the deterministic factors driving endometriosis development in certain women while sparing others, notwithstanding comparable experiences of retrograde menstruation. Within this review, we endeavor to synopsize the current understanding of diverse pathophysiologic mechanisms underlying the initiation and progression of endometriosis and delineate avenues for future research.

A literature search without time restriction was conducted utilizing PubMed and Google Scholar.

Given that aberrant clonal expansion stemming from cancer-associated mutations is common in normal endometrial tissue, only endometrial cells harboring mutations imparting proliferative advantages may be selected for survival outside the uterus. Endometriotic cells capable of engendering metabolic plasticity and modulating mitochondrial dynamics, thereby orchestrating responses to hypoxia, oxidative stress, inflammation, hormonal stimuli, and immune surveillance, and adeptly acclimating to their harsh surroundings, stand a chance at viability.

The genesis of endometriosis appears to reflect the evolutionary principles of mutation, selection, clonal expansion, and adaptation to the environment.

Endometriosis affects approximately 10% of women of reproductive age, causing pelvic pain and infertility.1 It is characterized by endometrial-like tissue present outside the uterus. Pelvic endometriosis is believed to arise from retrograde menstruation, the most widely accepted hypothesis.2 Several factors related to hormonal regulation, inflammatory response, immune surveillance, oxidative stress, and hypoxia have been reported as mechanisms underlying the development of endometriosis.3, 4 However, in the last decade, updated findings and valid theories on genetic mutations and epigenetic alterations have been discussed. Recently, cancer-associated mutations in the epithelial component of normal endometrium, deep infiltrating endometriosis,5 and ovarian endometriosis,6 along with aberrant epigenetic modifications in endometrial stromal cells, have been reported by several independent groups.3, 7-11 Aberrant clonal expansion resulting from cancer-associated mutations confers a proliferative advantage in endometrial and endometriotic epithelial cells.6, 11, 12 Conversely, endometriotic stromal cells exhibit specific epigenetic abnormalities rather than nonsynonymous mutations.13 It is known that the expression of many genes related to steroid hormone response, inflammatory response, angiogenesis, and apoptosis is regulated by DNA methylation.3, 14 Monthly bleeding in normal endometrium and endometriotic lesions can lead to oxidative stress and associated DNA mutations and epigenetic abnormalities, which are inevitable until menopause.15 Furthermore, refluxed endometrial tissue is exposed to severe hypoxic stress.16 To survive ectopically, endometrial tissue must regulate oxidative stress, hypoxia, inflammation, and immune surveillance.3, 4 Previous studies have reported mechanisms by which endometriotic cells survive in harsh environments. For instance, recurrent episodes of bleeding and fragments of basal endometrium dislocated into the peritoneal cavity may induce chronic inflammation and affect lesion growth via macrophage recruitment and local estrogen production.17 Macrophage recruitment may represent a host effort to eliminate an intruder,3 contributing to the inflammatory process and adhesion formation.13, 18, 19 Hypoxia-induced mitochondrial fission and fusion turnover may provide a significant fitness advantage by activating mitochondrial quality control mechanisms.20 Mitochondrial dynamics has been reported to maintain energy homeostasis.21 Energy homeostasis orchestrates diverse cellular processes, such as unique metabolic remodeling, biogenesis, respiration, and redox homeostasis.22 Spatiotemporal interactions of genomic, epigenetic, and environmental factors at different stages may lead to the development of endometriosis. However, the association between genetic mutations and epigenetic alterations with the molecular mechanisms of endometriosis development remains unclear. In this review, we summarize the current understanding of the pathogenesis of endometriosis, focusing on how these factors interrelate and contribute to the development of endometriosis, and discuss future research directions.

Search strategy and selection criteria A comprehensive narrative review was performed using the PubMed and Google Scholar databases to identify pertinent studies without temporal limitations, employing the keywords delineated in Table 1. The search terms were combined using Boolean operators AND and OR. Additionally, a manual reference search of published articles was conducted. Included studies comprised original research publications in English and reference lists from review articles. Duplicated studies, literature irrelevant to the research topic, and non-English publications were excluded. | Search mode | The keyword and search term combinations | |---|---| | Search term 1 | Endometriosis OR Peritoneal endometriosis OR Ovarian endometriosis OR Deep infiltrating endometriosis | | Search term 2 | Epigenetic alterations OR Epigenetic mutations | | Search term 3 | Genetic mutations OR Cancer-associated mutations OR somatic mutations | | Search term 4 | Metabolic plasticity OR Metabolic flexibility | | Search term 5 | Mitochondrial dynamics OR Mitochondrial fusion OR Mitochondrial fission | | Search | Search term 1 AND Search term 2 | | Search term 1 AND Search term 3 | | | Search term 1 AND Search term 4 | | | Search term 1 AND Search term 5 |

The origin of endometriosis Endometriosis occurs spontaneously only in menstruating women and female non-human primates.23 Several theories have been proposed to elucidate its pathogenesis,24 including retrograde menstruation,2 coelomic metaplasia, extrauterine-sourced stem cells (stem cell dysfunction), Müllerian rest induction (embryonic rest), and hematogenous or lymphatic spread.3, 25, 26 The retrograde menstruation theory is among the most widely accepted.2 However, the presence of endometriosis in regions beyond the pelvic area and in women who have undergone hysterectomy indicates that retrograde menstruation is not the sole pathway for its development.24, 27, 28 The stem cell theory suggests that stem cells may originate from the uterine endometrium or bone marrow.28-30 Endometriotic implants comprise distinct populations, characterized by a small number of epithelial cells and a larger number of stromal cells.13 Given that cancer-associated mutations in normal endometrial and endometriotic epithelial cells are similar, it is posited that endometriotic epithelial cells could regenerate from the normal endometrium.6, 9 Conversely, with the advent of new techniques, it has been reported that the origins of epithelial and stromal components differ, suggesting that stromal cells might be perpetually recruited from sources other than endometrial or endometriotic epithelial cells.9, 10 In this review, we examine the pathogenesis and progression of peritoneal endometriosis, primarily grounded in the theory of retrograde menstruation. Somatic mutations in endometriosis Numerous studies have aimed to identify the genes implicated in the development of endometriosis.31 Relatives of individuals with endometriosis are more likely to have the condition compared to relatives of those without the disease.31-33 Twin studies have also shown a higher concordance rate for endometriosis in monozygotic twins than in dizygotic twins.34 However, family-based or case–control genetic association analyses have not yet demonstrated statistical significance.35 Furthermore, endometriosis increases susceptibility to infections, allergies, psychiatric conditions, various autoimmune diseases such as rheumatoid arthritis, systemic lupus erythematosus, and multiple sclerosis, as well as adverse obstetric outcomes, including placenta previa, placental abruption, postpartum hemorrhage, preeclampsia, amniotic fluid abnormality, preterm labor, premature rupture of membranes, metabolic syndrome, coronary heart disease, ovarian cancer, and breast cancer.36, 37 Additionally, genome-wide association studies have identified genetic variants associated with endometriosis risk, including genes related to estrogen, progesterone, and androgen receptors, cytochrome P450, steroidogenesis, inflammation, immune response, angiogenesis, matrix remodeling, DNA repair, cell adhesion, cell growth, cell cycle regulation, metabolism regulation, oxidative stress, and p53 signaling.13, 14, 38-41 However, studies suggest the absence of germline mutations or causative somatic mutations responsible for the endometriosis phenotype.13 Next, over the past decade, somatic epithelial mutations in both normal endometrium and endometriosis have garnered increasing attention. Recent advancements in exome sequencing have identified somatic epithelial mutations not only in deep infiltrating endometriosis,5 ovarian endometriosis,6 and iatrogenic endometriosis,42 but also in normal endometrium.5-7 A comprehensive review of cancer-associated mutations in endometriosis was reported by Sun-Wei Guo in 2020.10 Historically, an X-chromosome inactivation pattern analysis conducted in 2003 demonstrated that normal human endometrial glands originate monoclonally.11, 43 In 2014, whole-exome sequencing revealed numerous nonsynonymous single-nucleotide variants in both ectopic and matched eutopic endometrial epithelial samples from patients with endometriosis, as well as in normal endometrial epithelial samples from controls.7 Moreover, in 2017, several nonsynonymous somatic mutations were identified in the epithelial component of deep invasive endometriosis in 19 (79%) out of 24 patients.5 Additionally, cancer-associated driver mutations (e.g., ARID1A, PIK3CA, KRAS, PPP2R1A), commonly found in ovarian clear cell carcinoma, were identified in 26% of deep infiltrating endometriosis tissues.5 A 2018 report indicated that 19 mutations were identified within the epithelial component in six patients with either ovarian endometriosis or deep infiltrating endometriosis, but not in stromal cells.44 Around the same time, numerous somatic mutations (e.g., PIK3CA, KRAS) were identified in a significant number of epithelial cells from ovarian endometriosis, control eutopic endometrium, and normal endometrium.6 Collectively, somatic mutations were detected in a significant proportion of endometriotic lesions, with cancer-related mutations found in 26% to over 40% of cases, irrespective of the endometriosis subtype. This includes 28% in iatrogenic endometriosis (IE),42 over 40% in ovarian endometriosis (OE),6 and 26%5 to 36% in extraovarian deep endometriosis (DE).42 Other studies45, 46 also confirmed the presence of somatic driver mutations in normal endometrium. This suggests that the prevalence of cancer-associated mutations in normal endometrium indicates a significant fitness advantage for endometriosis phenotypes, thereby promoting clonal expansion.6, 10, 47 Although the small number of cases examined makes it difficult to draw definitive conclusions, many of these mutations spontaneously accumulate in endometrial epithelial cells early in life46 and appear to increase with age.12 Finally, it has been reported that cancer-associated mutations have also been identified in healthy human tissues other than endometrial tissue. With the advent of next-generation sequencing and technological advances, it has become widely recognized that replicative aging can lead to the occurrence of cancer-associated mutations in tissues that appear to be healthy.10, 12, 48, 49 For instance, normal tissues, including blood, colon, skin, esophagus, and endometrium, frequently accumulate cancer-associated mutations (i.e., nonsynonymous mutations) throughout life.11, 12 The types of mutations vary across tissues, reflecting the spectrum of driver gene mutations identified in cancers specific to each tissue.12 For example, mutations in the TP53, APC, and KRAS genes are commonly found in ulcerative colitis, while mutations in the APC gene are not observed in endometriosis.50 Somatic mutations in specific cancer-associated genes lead to clonal expansion as individuals age and can induce different phenotypes of benign diseases, such as ulcerative colitis and Barrett's esophagus. We believe the same is true for endometriosis.12 However, even if driver gene mutations are present, endometriosis itself is a benign disease and not cancer. DNA mutations in the normal endometrium, resulting from environmental factors such as bleeding, are an inevitable part of life. The presence of unique cancer-associated mutations in separate endometrial glands underscores the diversity in the genomic structure of the endometrial epithelium.6, 9 Once the intracavitary endometrium is refluxed into the peritoneal cavity, only those cells that provide a proliferative advantage to the glands in a hypoxic environment are likely to implant as clonal populations (Figure 1). Indeed, KRAS mutations suppress apoptosis,51 potentially leading to increased cell survival in ectopic endometrium. ARID1A mutations suppress transforming growth factor-β (TGF-β) signaling52 and downregulate the expression of progesterone receptors,53 contributing to the pathophysiological characteristics of endometriosis. These cancer-associated mutations also play a significant role in fibrogenesis.10, 54 Further growth advantage and transformation into endometriosis likely require adaptation through clonal expansion to the microenvironment, which is not present in normal endometrial tissue.55 This concept is similar to the clonal evolution of tumor cell populations proposed approximately 50 years ago.56 In addition to clonal expansion, adaptation to local estrogen production, the inflammatory phenotype of refluxed endometrium, immune surveillance, oxidative stress, and hypoxia in the peritoneal microenvironment may play a critical role in the transformation of normal endometrium into endometriotic lesions. Epigenetic alterations in endometriosis In 2014, Li et al.7 identified that normal endometrial epithelial cells in patients with endometriosis frequently harbor somatic mutations, including modifications to histone H3 lysine 4, leading to epigenetic changes. Epigenetic modifications encompass chemical changes to DNA or histone proteins that regulate gene activity without altering the DNA sequence itself.57, 58 Research indicates that differentially methylated genes are associated with the overproduction of steroid hormones, regulation of inflammatory responses, angiogenesis, cell proliferation, differentiation, apoptosis, and evasion of immune surveillance.3, 59, 60 Epigenetic abnormalities are characteristic of stromal cells, influencing gene expression.6, 9, 13 Consequently, the endometrial cell population consists of epithelial cells with nonsynonymous mutations and stromal cells with epigenetic abnormalities.13 Epigenetic alterations, such as DNA methylation, histone modification, and non-coding RNA, in endometriosis are summarized below13, 59, 61 (Figure 2). DNA methylation patterns are non-random, displaying increased methylation at chromosome ends in endometriotic stromal cells.62 Aberrant DNA methylation affects the expression of various genes, including homeobox A10 (HOXA10), estrogen receptor-beta (ERβ; ESR2), estrogen receptor-α (ERα; ESR1), progesterone receptor (PGR), and aromatase (CYP19A1), as well as GATA binding protein 2 and 6 (GATA2/6), splicing factor 1 (SF1), and nuclear factor kappa B (NF-κB).13, 28, 63-67 Notably, CpG islands in the ESR2 promoter region are hypomethylated, whereas those in the ESR1 and PGR promoter regions are hypermethylated within endometriotic stromal cells.24, 68 These epigenetic modifications are believed to contribute to progesterone resistance, a hallmark of endometriosis.69 Epigenetically altered stromal cells have also developed additional adaptations to thrive in the pelvic environment.13 Second, post-translational modifications of histones, including acetylation and methylation, are crucial in regulating DNA accessibility.14 Dynamic changes in histone modifications are observed during decidualization; however, the patterns of H3 and H4 acetylation and methylation differ between women with and without endometriosis.70, 71 For example, aberrant epigenetic modifications can downregulate the expression of insulin like growth factor binding protein 1 (IGFBP1) and prolactin (PRL), mediated by H3K27 acetylation (H3K27ac), potentially leading to impaired decidualization.72 Third, unlike DNA methylation and histone modifications, non-coding RNAs (ncRNAs) regulate gene expression, cell physiology, and cellular functions at the post-transcriptional level in response to environmental signals.73 MicroRNAs (miRNAs) are known for their extensive influence on genes related to hormones, cytokines, chemokines, oxidative stress, inflammation, hypoxia, and immune responses in endometriosis.13, 59, 74 Finally, factors that influence the epigenetic process include environmental stimuli.14 DNA methyltransferases (DNMTs) are enzymes that catalyze the transfer of methyl groups to DNA, thereby manipulating epigenetic information which is critical in determining cell fate.14 For instance, in the placenta, DNMT activity is regulated by microenvironmental stimuli such as hypoxia, inflammation, and steroidogenic pathways.75 Thus, changes in the microenvironment can lead to the epigenetic reprogramming of endometriotic cells. The interplay between epigenetic mechanisms and endometriosis-specific alterations in the microenvironment is complex and interdependent. The molecular basis for the pathophysiology of endometriosis The prevailing theory regarding the development of endometriotic lesions posits that menstrual blood retrogrades through the fallopian tubes into the peritoneal cavity, a phenomenon known as retrograde menstruation. This may result in the implantation of exfoliated endometrial cells possessing cancer-associated mutations and stem cell-like properties onto the peritoneum.6, 9, 69 In contrast, most other cells lacking these genetic mutations fail to implant and are likely to perish. Given that glandular epithelial cells in both normal endometrium and endometriosis share analogous cancer-related mutations,6, 9 it is plausible that ectopic endometrial epithelial cells originate from intracavitary endometrial epithelial stem cells, rather than from bone marrow mesenchymal stem cells. Moreover, the transdifferentiation of epithelial cells into stromal cells appears improbable due to the distinct gene mutation profiles exhibited by these cell types.6, 9 Endometriotic lesions contain a higher proportion of stromal cells compared to epithelial cells,13 and these stromal cells may originate from endometrial stromal cells or bone marrow-derived mesenchymal stem cells. However, the precise mechanism by which some exfoliated endometrial cells progress to endometriosis remains elusive. This subsection delineates the current understanding of endometriosis development and pathophysiology, encompassing the molecular aspects of hormone production, inflammatory responses, immune function, oxidative stress, and hypoxia. Local production of estrogen The initial adhesion of endometrial tissue to the peritoneal surface may induce inflammation and local estrogen production via tissue injury and repair (TIAR) mechanisms17 (Figure 3). Additionally, estrogen production facilitates the recruitment of macrophages to the peritoneum and the release of proinflammatory cytokines such as tumor necrosis factor-α (TNF-α) and interleukin-1β (IL-1β).19 These cytokines activate NF-kB, leading to the expression of vascular endothelial growth factor (VEGF)76 and the activation of the anti-apoptotic gene BCL2 apoptosis regulator (BCL2).19, 77 Such mechanisms protect endometrial cells from macrophage-induced destruction and promote angiogenesis, enabling their survival at ectopic sites. Furthermore, estradiol in endometriotic stromal cells induces cyclooxygenase-2 (COX-2) expression via ERβ, resulting in an overproduction of prostaglandin E2 (PGE2).13, 78 PGE2 binds to the EP receptor on the cell membrane, triggering adenylate cyclase and boosting cyclic AMP (cAMP) production. This increase in cAMP leads to the activation of protein kinase A (PKA), which subsequently activates transcription factors. These activated transcription factors then bind to the promoter regions of genes for aromatase and steroidogenic acute regulatory protein (StAR), enhancing their expression. This overproduction is crucial for the continuous synthesis of estradiol.13, 24, 79 Consequently, a feedback loop emerges where estrogen increases inflammation by upregulating the expression of COX-2 and PGE2, which in turn stimulates endometriotic cell proliferation via estrogen production.69 Moreover, both eutopic endometrial tissue and endometriotic lesions are characterized by their proliferative80 and antiapoptotic properties,81 which are influenced by estrogen. Estrogen also activates the C-X-C motif chemokine ligand 12 (CXCL12)-C-X-C motif chemokine receptor 4 (CXCR4) pathway, attracting bone marrow-derived mesenchymal stem cells to the endometriotic lesions.82, 83 Hence, estrogen is believed to be pivotal in the initial stages of endometriosis, facilitating the transformation of refluxed endometrium into endometriotic cells and their survival in the hostile peritoneal environment. Inflammatory phenotype of refluxed endometrium Endometriosis is consistently associated with chronic inflammation, dependent on the host's ability to recognize and eliminate ectopic endometrial tissue as a foreign entity. Repeated episodes of bleeding also activate inflammatory and oxidative stress pathways, which can lead to increased adhesion formation13 (Figure 4). Consequently, endometriotic tissue may either become fibrotic or evade elimination, proliferating as active lesions.3 The severe adhesions, scarring, and fibrosis may exacerbate pain symptoms.3 Hence, inflammation, a critical mechanism in endometriosis, triggers tissue remodeling, adhesion formation, and fibrosis, which can ultimately lead to pain and infertility.13 Why do some women develop endometriosis while others do not, despite experiencing similar inflammation? Women prone to developing endometriosis may be more susceptible to inflammation. A recent review on endometriosis and endobiota was reported by Uzuner et al. in 2023.31 Microbial dysbiosis is characterized by a decrease in beneficial probiotics and an increase in pathogenic microbes. Gut or reproductive tract microbiome dysbiosis has been reported in basic studies using mouse and non-human primate endometriosis animal models, as well as in clinical studies in women with and without endometriosis.31, 84-86 In fact, research has explored the association between the microbiome and inflammation in endometriosis patients, suggesting that these patients often have altered microbiomes.87 Animal models of endometriosis support that not only does the gut microbiota influence the development of endometriotic lesions, but also that the presence of endometriotic lesions causes an imbalance in the gut microbiota. Transfer of fecal microbes from mice with endometriosis leads to lesion regrowth, and antibiotics reduce lesion size.88 Furthermore, microbiota imbalance (i.e., dysbiosis) in the gut, peritoneal fluid, and female reproductive tract, particularly the prevalence of bacteria that produce estrogen deconjugating enzymes, may lead to estrobolomic changes, triggering the development of endometriosis.31 Dysbiosis might predispose individuals to developing endometriosis. Immune surveillance and its dysregulation Macrophages eliminate physiological quantities of refluxed menstrual debris, including ectopic cells and tissues.19 In patients with endometriosis, recurrent menstruation results in the inadequate clearance of menstrual debris, leading to alterations in the types and quantities of immunocompetent cells, along with variations in the patterns and levels of cytokines and chemokines derived from inflammation in the peritoneal fluid.19, 89 Macrophages, pivotal in immune function, undergo functional polarization into M1 and M2 phenotypes, associated with pro-inflammatory and anti-inflammatory activities, respectively24 (Figure 4). In the initial stages of endometriosis, M1 macrophages, which possess the capability to eradicate defective cells, predominate during the pro-inflammatory phase, while M2 macrophages, which serve an immunosuppressive role, become prevalent in the tissue repair and regeneration phase at later stages.90 M1 macrophages detect tissue damage induced by hypoxia and iron overload and augment the production of inflammatory cytokines and growth factors, including VEGF, TNF-α, and interleukin-8 (IL-8), through heightened activation of the NF-κB pathway.91, 92 These cytokines contribute to estrogen-induced inflammation and facilitate the proliferation of endometrial implants.91, 92 Over time, M2 macrophages contribute to further tissue repair and fibrogenesis by producing anti-inflammatory cytokines such as interleukin-10 (IL-10) and TGF-β.19, 93, 94 An excessive shift in the Th1/Th2 balance toward Th2 is fundamental to the immunoregulation of endometriosis, potentially accounting for the immune dysregulation, leading to compromised immunosurveillance.95, 96 Inadequate immune surveillance permits endometriosis to sustain its ability to survive and proliferate. Thus, endometriosis is considered a chronic condition wherein macrophages treat endometriotic lesions as wounds and foreign entities, continuously triggering inflammation, immune responses, angiogenesis, and fibrosis. Oxidative stress-mediated regulatory network Reactive oxygen species (ROS) are primarily generated from endogenous sources such as mitochondria and NADPH oxidases (NOXs), as well as exogenous sources like recurrent bleeding episodes in endometriosis97 (Figure 5). ROS activate NF-κB and phosphatidylinositol-4,5-bisphosphate 3-kinase (PI3K) to enhance downstream signaling.97 The NF-κB pathway contributes to the pathogenesis of endometriosis by modulating immune and inflammatory responses.97 The PI3K/AKT pathway activates mechanistic target of rapamycin kinase (mTOR), which inhibits autophagy and apoptosis in endometriotic cells.97, 98 While endogenous ROS sources are ubiquitous among living organisms, the correlation between monthly recurrent hemorrhages and oxidative stress is a distinctive feature of endometriosis pathology. Although iron is essential to multiple functions including oxygen transport, electron transport, mitochondrial function, and DNA replication and repair, recurrent bleeding in normal endometrium and endometriotic lesions precipitates pathological events, notably oxidative stress.15 An excess of free intracellular iron fosters the generation of deleterious hydroxyl radicals, disrupts redox homeostasis, and induces lipid peroxidation (ferroptosis).99, 100 In endometriosis, this redox imbalance may be driven by red blood cells and their toxic derivatives—hemoglobin, heme, and iron—accumulating in the peritoneal cavity, endometriotic cysts, or tissues.101 Notably, the iron concentration in endometriotic cysts is remarkably high (244.4 ± 204.9 mg/L [median ± SD]).102 Autoxidation and the Fenton reaction generate excessive ROS, such as superoxide (O2-) and hydroxyl radicals (∙OH), respectively.99 Both normal and endometriotic endometrial tissues can endure adverse physiological and pathological conditions that can damage DNA, because the associated oncogenic mutations and epigenetic modifications may enhance cell proliferation and survival (see Sections 3.2 and 3.3).10, 99 Furthermore, it has been reported that macrophages protect endometriotic cells from oxidative damage through a crosstalk mechanism.103 There is mounting evidence suggesting that chronic inflammation induced by oxidative stress may underlie chronic diseases including cancer and diabetes, and such persistent oxidative stress may also contribute to the development of endometriosis.104 Hypoxia-mediated regulatory network This subsection elucidates how hypoxia facilitates the implantation, proliferation, and progression of ectopic lesions. Refluxed and transplanted endometrial tissues face severe hypoxic stress, potentially leading to cell death.16 Endometriotic cells endeavor to mitigate ROS production and circumvent cell death by reprogramming their energy metabolism from oxidative phosphorylation (OXPHOS) to glycolysis22 (Figure 6). It has been established that the upregulation of 6-phosphofructo-2-kinase/fructose-2,6-biphosphatase 3 (PFKFB3), a pivotal enzyme in glycolysis, fosters the progression of endometriosis through glycolytic activation.105 Consequently, endometriotic cells modify cellular metabolism toward aerobic glycolysis, actively generating pyruvate and lactate, thereby promoting cellular proliferation.20, 106 Indeed, elevated levels of glycolysis-associated molecules (solute carrier family 16 member 7 [SLC16A7; MCT2], pyruvate dehydrogenase kinase 1 [PDK1], Glucose transporter 1 [GLUT1], hexokinase 2 [HK2], TGFβ, hypoxia-inducible factor-1α [HIF-1α]) have been observed in peritoneal endometriosis among adolescents.106 Increased HK2 expression may also activate the pentose phosphate pathway (PPP), which diverges from glycolysis, producing nicotinamide adenine dinucleotide phosphate (NADPH), ribose, and other metabolites essential for DNA repair and antioxidant responses, indicative of the Warburg effect.107 Mechanistically, through upregulation of HIF-1 and its downstream target TGF-β, hypoxia induces the overexpression of genes encoding glucose transporters, PDK1, and lactate dehydrogenase kinase A (LDHA), while downregulating pyruvate dehydrogenase (PDH) expression.22, 108 This suggests that endometriotic cells strive to minimize ROS production by reprogramming their energy metabolism from OXPHOS to glycolysis via TGF-β overexpression.109 Thus, endometriotic cells acquire a metabolic phenotype that is distinct from endometrial cells. Moreover, this adaptation mechanism is intricately linked to mitochondrial dynamics, which govern fission and fusion processes, thereby modulating their morphology and function. Mitochondria serve as a pivotal regulatory hub in physiological processes, encompassing biogenesis, metabolism, redox homeostasis, stress response, and organelle turnover.110 Mitochondrial dynamics represent tightly regulated mechanisms essential for supporting remarkable phenotypic plasticity required for adaptation to dynamic environments (e.g., hypoxia and nutrient deprivation),111 cellular energy metabolism (e.g., aerobic glycolysis and OXPHOS),112 and pathophysiological scenarios (e.g., mitochondrial quality control and cellular senescence).113 Mitochondria alter their morphology through membrane fusion and fission, directly impacting metabolism and biogenesis. Mitochondrial fusion (e.g., mitofusin [MFN] 1 and 2) and fission (e.g., dynamin-related protein 1 [DRP1]) are governed by evolutionarily conserved dynamin-like GTPases (Guanosine-Triphosphate hydrolase).114 Mitochondrial fusion enhances mitochondrial respiration, membrane potential, and adenosine 5′ triphosphate (ATP) levels, ensuring bioenergetic efficiency, genomic stability, and cell survival.115 Over time, ROS generated as byproducts of cellular metabolism via mitochondrial OXPHOS accumulate, potentially leading to oxidative stress. Therefore, endometriotic cells rely on mitochondrial dynamics to maintain metabolic plasticity, ensure mitochondrial quality control, and regulate redox homeostasis.20 Additionally, ROS, inflammation, and hypoxia can induce epigenetic modifications, enabling the acquisition of diverse survival strategies.16 Collectively, the intricate regulatory network involving mitochondrial dynamics and epigenetic regulation, triggered by hypoxia, equips endometriotic cells with essential survival strategies in challenging environments.16 We posit that only endometriotic cells that have adapted to low-oxygen conditions through regulated mitochondrial dynamics can thrive and proliferate.

In this review, we summarize the current understanding of the development and pathophysiology of endometriosis, emphasizing the molecular mechanisms of genetic mutations, epigenetic alterations, inflammation triggered by excess estrogen, evasion from immune surveillance, adaptation mechanisms linked to the complex interplay between hypoxia and oxidative stress within the microenvironment, regulation of mitochondrial dynamics, and remodeling of metabolic networks, and discuss future research perspectives. The prevailing theory is retrograde menstruation, and with this framework in mind, we compiled our findings (Figure 7). Recent genetic and epigenetic studies have progressively elucidated the pathophysiology of the disease.14 Given that cancer-associated mutations are prevalent in normal endometrial tissue and confer a proliferative advantage, only shed endometrial cells with aberrant clonal expansion may be selected for survival.12 This constitutes the first hurdle for the shed endometrial cells (Figure 7①). However, endometrial tissue refluxed into the peritoneal cavity is chronically exposed to lower oxygen tension and oxidative stress, leading to the death of many cells. Particularly, estrogen (e.g., COX-2/PGE2/Aromatase pathway) and persistent inflammation (e.g., NF-kB/TNF-α/IL-1β pathway) are thought to be critical initial steps for the refluxed endometrium to survive in the peritoneal environment (Figure 3). Some endometrial tissue may adapt to its microenvironment and progress to endometriosis. This constitutes the second hurdle for the attached endometrial cells (Figure 7②). The macrophage-mediated foreign body response triggered by endometriotic lesions causes persistent inflammation and fibrosis (Figure 4). Chronic inflammation is known to play a role in the development of cancer, diabetes, and endometriosis.104 Over time, endometriotic lesions may further acquire cancer-associated gene mutations via alterations in intrinsic and extrinsic factors, such as replicative aging and oxidative stress.10 Additionally, some endometriotic cells rewire their energy metabolism to sustain ATP production and support growth (Figure 6). This constitutes the third hurdle for early-stage endometriotic cells (Figure 7③). Activation of PDK1 and inactivation of PDH promotes a metabolic shift from mitochondrial OXPHOS to aerobic glycolysis. A metabolic shift toward glycolysis through mitochondrial fission further enables adaptation to low oxygen environments. Specific molecules (e.g., MFN1, MFN2, and DRP1) extensively reprogram mitochondrial dynamics, altering fusion and fission to support survival in harsh environments (Figure 6). Maintenance of redox homeostasis, alterations of energy metabolism, and regulation of mitochondrial dynamics are critical for endometriotic survival. This constitutes the final hurdle for established endometriotic cells (Figure 7④). The current understanding of endometriosis pathogenesis is still limited, but the initiation and progression of endometriosis appears to mimic the Darwinian logic of mutation, selection, clonal expansion, and adaptation to the environment.56 Finally, we discuss future directions for endometriosis research. Clonal expansions of both endometrial and endometriotic cells are common in women, especially in aging humans. Since endometriotic lesions have already undergone such clonal expansion, targeting cancer-associated genes and epigenetic alterations may not be effective therapeutically. To treat established endometriotic lesions, it is essential to precisely understand the expression patterns and profiles of molecules related to hormone production, inflammatory responses, immune evasion, oxidative stress, and hypoxia. Rapid and accurate analysis of target molecules will allow prediction of vital regulatory molecules and pathways involved in the growth and survival of endometriotic lesions. Furthermore, to reflect the dynamic state of a cell, transcribed genes must be measured over time. For example, to clarify the hormonal environment, it may be necessary to measure specific genes such as ERβ, ERα, PGR, COX-2, PGE2, and aromatase. Identifying inflammation-related genes (e.g., NF-kB, TNF-α, IL-1β, IL-10) may help predict innate and adaptive immune responses or tissue repair and fibrosis. Both metabolic remodeling and the adaptation of mitochondrial dynamics to an ever-changing environment have been recognized as hallmarks of the development and progression of endometriosis by adjusting the respiratory capacity of mitochondria. The identification of metabolic remodeling pathways relies on glycolysis-related genes, including HIF-1, PDK1, PDH, and HK2. It is essential to identify the expression of specific genes (MFN1, MFN2, DRP1) to understand alterations in mitochondrial dynamics. These two pathways are believed to be mechanisms underlying the stress adaptation of endometriotic cells in response to a hostile environment. Exploiting the vulnerabilities of endometriotic cells is critical for the discovery and development of non-hormonal drugs. Custom targeted enrichment panels are a well-established technique to simultaneously investigate all genetic causes.116 This panel can be used to discover targetable genes and customize medical care to an individual's needs. For example, liquid biopsy, a minimally invasive approach, provides the opportunity to detect molecular-level changes in real-time. The implementation of these methodologies generates opportunities for elucidating pathophysiology and identifying novel biomarkers, as well as offering enhanced tools for patient stratification, diagnosis, and optimized treatment. In summary, a deeper understanding of the interactions between endometriotic cells and the microenvironment will guide future research. AUTHOR CONTRIBUTIONS Conception and design, Hiroshi Kobayashi. Acquisition of data, Shogo Imanaka, Chiharu Yoshimoto, and Sho Matsubara. Analysis and interpretation of data, Hiroshi Kobayashi and Hiroshi Shigetomi. Drafting of the manuscript, Hiroshi Kobayashi. Critical revision of the manuscript for important intellectual content, Shogo Imanaka, Chiharu Yoshimoto, Sho Matsubara and Hiroshi Shigetomi. Statistical analysis, none. Obtaining funding, none. Administrative technical or material support, Hiroshi Kobayashi. Supervision, none. All authors have read and agreed to the published version of the manuscript. ACKNOWLEDGMENTS Figure was created by Toyomi Kobayashi (Ms.Clinic MayOne, Kashihara, Japan); https://www.mscl-mayone.com/ (accessed on June 13, 2024). CONFLICT OF INTEREST STATEMENT The authors declared no potential conflicts of interest. DATA AVAILABILITY STATEMENT No new data were created or analyzed in this study.