Association of Genetic Variations in The PIK3-AKT-mTOR Pathway with Endometriosis Susceptibility: A Preliminary Case-Control Study

Endometriosis is a complex, heterogeneous, chronic inflammatory gynaecologic disease characterised by the presence of endometrial tissue that contains glands and stroma outside the uterine cavity, and mainly located in the pelvic cavity and pelvic organ walls (1). Several genetic and environmental factors play key roles in its development (2, 3). A possible genetic component of endometriosis was first proposed in the 1940s, and this theory was later supported in the 1980s (4). A twin study estimates that about 50% of the risk for endometriosis is inherited (2). Identifying disease-causing gene variants can be done in several ways, and the major challenge is to select a few candidates for further investigation (5). Numerous studies have recently investigated the association between endometriosis and single nucleotide polymorphisms (SNPs), particularly genes involved in inflammation and detoxification, as well as cell adhesion and endocrine signaling pathway gene polymorphisms (6, 7). SNPs are the most common cause of genomic dissimilarity. They denote variations at a specific site in the genome with a frequency of 1-50% in the entire population. SNPs can alter protein structure and function, though they may not cause disease, and thereby affect disease susceptibility, including endometriosis. Therefore, SNPs can be useful for describing at-risk populations (8). We based our analysis on the key role played by the phosphatidylinositol 3-kinase/protein kinase B/mammalian target of rapamycin (PI3K/AKT/mTOR) signaling pathway in endometriosis pathogenesis. This pathway participates in various cellular functions that include cell growth, differentiation, transformation, and survival. Activation of the PI3K/AKT/ mTOR pathway in an oestrogen-dependent manner in the human endometrium has been reported. Results of in vitro and in vivo studies suggest that PI3K/AKT/mTOR hyperactivation causes considerable dysregulation of various cellular functions that may lead to endometriosis (9, 10). Overexpression of the PI3K/AKT/mTOR pathway in eutopic endometrial stromal cells of women with endometriosis was reported (11, 12). The PIK3CA gene is mapped on chromosome 3q26.3 and encodes the p110alpha catalytic subunit of phosphoinositide 3-kinases (PI3Ks) (13). The SNP rs2230461 is a genetic variation that leads to the substitution of isoleucine with methionine at position 391 of the enzyme. The downstream target of PI3K is AKT1. Its gene is located on chromosome 14q32 and plays a crucial role in cancer cell survival, cell cycle progression, and glucose metabolism (13). AKT1 rs1130233 is a synonymous coding variation associated with differential expression of the AKT protein (14, 15). The mTOR gene, located on chromosome 1p36.22, is another important target within the PI3K/AKT/mTOR pathway. Its protein product is a serine/threonine kinase that functions as a downstream effector in the signalling pathway (16). The SNP rs2295080 in the promoter region of the mTOR gene is a protective factor of some cancers through downregulation of endogenous protein expression (17, 18). Although it is suggested that the PI3K/AKT/mTOR pathway might influence progression and different stages of endometriosis (19, 20), no study has investigated relations of the PIK3CA , AKT1 , and mTOR gene polymorphisms with endometriosis risk and prognosis. We intend to elucidate whether these polymorphisms can serve as risk and/ or prognostic markers for endometriosis by conducting a case-control study to investigate the PIK3CA rs2230461, AKT1 rs1130233, and mTOR rs2295080 SNPs on endometriosis risk and clinical stage frequency.

Table 2 lists the clinical characteristics of all participants. A study of endometriosis-associated symptoms and the number of deliveries revealed significant differences between the endometriosis and control groups. Symptoms of chronic pelvic pain, dysmenorrhoea, and dyspareunia were significantly common among endometriosis patients (P<0.001). However, BMI, menstrual cycle length and/or menstrual bleeding length, and spontaneous abortions did not significantly differ between cases and controls. There were 58.26% of endometriosis patients with stages I and II disease and 41.73% with stages III and IV disease . The SNP primer and genotyping methods SNP; Single nucleotide polymorphism, AS-PCR; Allele-specific polymerase chain reaction, and PCR-RFLP; Polymerase chain reaction-restriction fragment length polymorphism. Characteristic of endometriosis cases and controls Data are presented as mean ± SD or n (%). Pearson’s chi-square (χ2) test, (95% CI) with an unconditional logistic regression model. The bold values are significance <0.05. BMI; Body mass index, CI; Confidence interval, *; A patient can have more than one symptom, and **; According to the revised classification of the American Society of Reproductive Medicine (rASRM). We genotyped three potentially functional PIK3CA, AKT1 , and mTOR SNPs in the endometriosis and control samples to assess potential associations between the SNPs and endometriosis risk. The genotype distributions for all SNPs in healthy women were in accordance with the Hardy-Weinberg equilibrium (HWE). Table 3 shows the association between variants of PIK3CA rs2230461 A>G and endometriosis risk. The GG homozygous mutant genotype had a 2.58-fold increased risk (95% CI: 1.119-5.985, P=0.026). Significant associations were observed for the PIK3CA dominant variant (AA vs. AG+GG), which was correlated with an approximately two-fold higher prevalence of this variant between endometriosis women (OR: 1.881; 95% CI: 1.065-3.322, P=0.029). We observed a significant difference for the PIK3CA gene polymorphism in the recessive model (GG vs. AA+AG), with an approximately 2.5-fold increase for endometriosis (OR: 2.4, P=0.037). In addition, there was a high frequency of mutant allele ‘G’ in endometriosis patients (OR: 1.938; 95% CI: 1.255-3.135, P=0.003, Table 3). The AKT1 rs1130233 mutant AA variant was significantly more frequent in endometriosis patients (OR: 2.870; 95% CI: 1.093-7.535, P=0.032). In the dominant GA and AA combined vs. GG model, there was a significantly higher risk of endometriosis in cases compared to controls (OR: 1.835, 95% CI:1.113-3.025, P=0.017). Women with endometriosis were significantly more likely to have the AKT1 mutant allele (A) than those with the normal allele (OR: 1.671, 95% CI: 1.135- 2.460, P=0.009, Table 3). The mTOR rs2295080 mutant TT genotype was related to a decreased risk for endometriosis (OR: 0.317, 95% CI: 0.108-0.927, P=0.036). However, a lower risk for endometriosis was associated with the mTOR SNP in cases under the dominant genetic model (GT/GG) (OR: 0.539, 95% CI: 0.317-0.951, P=0.032). The results also suggested that the mTOR rs2295080 T allele was more frequent in healthy controls; therefore, it has a probable protective role in the development of endometriosis (OR: 0.529, 95% CI: 0.333-0.839, P=0.006, Table 3). Genotype and allele frequencies of PIK3CA rs2230461, AKT1 rs1130233, and mTOR rs2295080 gene polymorphisms in the control and endometriosis groups Data are presented as n (%). Pearson’s chi-square (χ2) test, (95% CI) with an unconditional logistic regression model. The bold values are significance <0.05. OR; Odds ratio, CI; Confidence interval, and HWE; Hardy-Weinberg equilibrium. Table 4 shows the combined effects of these SNPs and their association with endometriosis. The frequencies of the combined gene polymorphisms of the PIK3CA rs2230461AA, AKT1 rs1130233 GA and mTOR rs2295080 GG and the PIK3CA rs2230461 AG, AKT1 rs1130233 GG and mTOR rs2295080 GG were 2.6- and 2.8-fold significantly higher in endometriosis group compared to the control group (P=0.010 and P=0.047, respectively, Table 4). The combined frequencies of the PIK3CA rs2230461 GG, AKT1 rs1130233 GA and mTOR rs2295080 GG as well as the PIK3CA rs2230461 GG, AKT1 rs1130233 GG and mTOR rs2295080 GG showed an approximately 8- and 9- fold risk for endometriosis compared with the control group. These increases were statistically significant (P=0.014 and P=0.009, respectively). There was an approximately 12-fold increase of the PIK3CA rs2230461 AA, AKT1 rs1130233 AA and mTOR rs2295080 GG combined gene polymorphisms in the endometriosis women compared to the healthy controls, which was significant (P=0.028, Table 4). No statistically significant association was observed between other genotypic combinations in the endometriosis women compared with the control group. The combined effects of PIK3CA rs2230461, AKT1 rs1130233 and mTOR rs2295080 gene polymorphisms on endometriosis risk Pearson’s chi-square (χ2) test, (95% CI) with an unconditional logistic regression model. The bold values are significance <0.05. OR; Odds ratio and CI; Confidence interval. Patients with stages I and II (minimal/mild) and stages III and IV (moderate/severe) endometriosis were studied separately. We noted a high prevalence of mutant alleles of PIK3CA rs2230461 and AKT1 rs1130233 SNPs in patients with stages III and IV disease. According to Table 5, moderate/severe endometriosis risk was about five times greater in patients with PIK3CA rs2230461 (OR: 4.800, 95% CI: 2.171-10.611, P<0.001) and more than two times greater in patients with AKT1 rs1130233 (OR: 2.674, 95% CI: 1.261-5.670, P=0.010, Table 5). These two mutant alleles may be high-risk alleles that could have a special effect on endometriosis severity. However, the mTOR variant rs2295080 mutant allele was not predominant in moderate/ severe cases (OR: 0.337 95% CI: 0.132-0.859, P=0.022). Association between PIK3CA rs2230461, AKT1 rs1130233, and mTOR rs2295080 gene polymorphisms and clinical stage frequency in endometriosis patients Data are presented as n (%). Pearson’s chi-square (χ2 ) test, (95% CI) with an unconditional logistic regression model. The bold values are significance < 0.05. SNP; Single nucleotide polymorphism, OR; Odds ratio, and CI; Confidence interval.

This is the first case-control study to investigate the impact of PI3K/AKT/mTOR SNPs on the development of endometriosis in an Iranian population. Genotyping analysis of these three SNPs revealed significant risk associations between carriers of the mutant alleles in the endometriosis group compared with healthy controls. In particular, the SNPs PIK3CA rs2230461 and AKT1 rs1130233 lead to missense mutations and to a more than approximately two-fold increase in disease risk in homozygous mutation. The rs2230461 SNP results in a missense mutation, which likely alters the function of the PI3K enzyme. Changes in PI3K activity can lead to abnormal cellular proliferation and survival, which are central to endometriosis development (19). The rs1130233 SNP results in a missense mutation in the AKT1 protein, potentially altering its kinase activity. This can affect numerous downstream processes, including cell growth and apoptosis, and contribute to the pathological environment observed in endometriosis (19). However, the results were different for mTOR variants, and the risk of developing the disease was reduced by half or less in carriers of the mutant alleles. This specific SNP in the mTOR gene is a genetic variation that can potentially influence the expression or function of the mTOR protein and could influence cellular processes in a way that protects against disease development. In addition, given mTOR's role in autophagy, a reduction in mTOR activity due to the polymorphism could enhance autophagy, which is a cellular process known to be protective against various diseases by removing damaged organelles and proteins. It is possible that changes in mTOR signaling can alter cellular metabolism. A polymorphism that reduces mTOR activity might improve metabolic homeostasis and reduce disease risk (23). In addition, analysis of the association between the PIK3CA rs2230461, AKT1 rs1130233, and mTOR rs2295080 gene polymorphisms with clinical stage of endometriosis confirmed the previous results. There was a statistical association between PIK3CA rs2230461 and AKT1 rs1130233 polymorphisms and moderate/severe (stages III and IV) endometriosis. Dysregulation of the PI3K/AKT pathway might lead to larger and more invasive lesions characteristic of stages III and IV (19). The pathway also intersects with inflammatory signaling. Enhanced or aberrant PI3K/AKT signaling due to these polymorphisms may exacerbate inflammation, a key feature of severe endometriosis (11, 12). In addition, mTOR , downstream of AKT , regulates angiogenesis. Increased angiogenesis can support the growth and maintenance of endometriotic lesions, particularly in severe cases (19). In the case of the mTOR rs2295080 SNP, the mutant allele was more frequent in stages I and II of the disease. Its protective effects might prevent disease progression to more severe stages. Our recent study has shown that certain variations in important genes may make individuals more susceptible to endometriosis (25). Endometriosis is an inflammatory disease where endometrial-like tissue colonises outside the uterine cavity after an increase in the proliferative, invasive, and migratory potential of phenotypically normal endometrial cells such that it has cancer-like features from the beginning of its development. Among the various signaling pathways that control cell proliferation, the PI3K/AKT/mTOR pathway is under investigation as a possible component of the pathogenic mechanism that leads to endometriosis (26). Migration of normal endometrial cells to the peritoneal surface is regulated by activating the PI3K/AKT/mTOR pathway in endometriotic cells by oestrogens and growth factors (27). Its abnormal activation is a common event in endometriosis, and this pathway may play a role in the development of endometriosis. The antiapoptotic and proliferative effects of the PI3K/AKT/mTOR pathway in endometriosis have been reported (28). Therefore, this pathway has become an attractive therapeutic target in endometriosis, and recent studies have attempted to evaluate the effects of the PI3K/ AKT/mTOR signaling pathway on disease progression. The role of SNPs in the PI3K/AKT/mTOR pathway and susceptibility to various diseases has been studied (16- 18). Some of these reported data supports our findings (21). PIK3CA rs2230461 is localized in the C2 domain and does not affect the binding site of the protein product. Although 7% of PIK3CA mutations are located in the C2 domain, it is possible that all mutations far from the binding site could affect the function of the protein and alter its dynamic behaviour (29). Qiu et al. (30) showed that the oncogenic properties of PIK3CA played a role in the development of human head and neck cancers, especially those that affect the pharynx. A study conducted in Iran examined the frequency of the PIK3CA rs2230461 SNP in stage III breast cancer patients and a control group. The results showed a significantly higher occurrence of the mutant genotype among stage III cases compared to the control group (31). Several studies reported the association of AKT rs1130233 polymorphisms located on chromosome 14 in exon 8 with increased disease risk, including bladder cancer (20), schizophrenia (32), gastric cancer (33), weight loss (34), nasopharyngeal carcinoma (35), and pancreatic cancer (36). A recent study examined the impact of mTOR (rs2295080) and AKT1 (rs1130233) gene variations on the likelihood of developing papillary thyroid cancer. Similar to our findings, their results indicate that rs2295080 and rs1130233 are protective and risk factors for disease development, respectively (37). Various studies have shown that mTOR 2295080 (T > G) variants have a positive or negative association with diseases. A meta-analysis found that a significant association of mTOR rs2295080 with reduced oesophageal squamous cell carcinoma risk under the homogenous (GG vs.TT) model (38). Piao et al. (33) reported an association between rs2295080 and gastric cancer risk in the Chinese population. Unlike, the results of recent meta-analysis that have revealed the mTOR 2295080 gene polymorphism as a genetic risk factor for cancer susceptibility and their carcinogenesis and clinical outcomes (39), Zhao et al. (40) had observed the protective effect of this SNP on breast cancer development. A meta-analysis reported a significant association between the mTOR rs2295080 G allele in the dominant model and decreased risk of genitourinary cancers, along with an increased risk of acute leukaemia in the recessive model (17). Findings from these studies contribute to the identification of potential markers for a genetic screening test that can be used to detect diseases like endometriosis. A genetic test that examines biomarkers in the endometrium and a thorough review of the family's medical history might identify women who have a higher likelihood of developing endometriosis. The use of these gene polymorphisms in diagnostic modalities in primary care for symptomatic patients might reduce the time to diagnose endometriosis. A limitation of this study is that it focuses on an Iranian cohort. The findings may not be generalisable to other populations because of potential differences in genetic backgrounds, environmental factors, and disease prevalence. Further studies that enrol diverse populations are needed to confirm the role of these gene polymorphisms in endometriosis susceptibility.

For the first time, we reported that rs2230461 ( PIK3CA ) and rs1130233 ( AKT1 ) variations are risk factors for endometriosis. We observed that these polymorphisms had a significant relation with the III/IV stages of the disease; therefore, we propose that the selected SNPs may be involved in endometriosis development. On the other hand, the rs2295080 ( mTOR ) SNPs could be a protective factor against endometriosis in a subset of the Iranian population. Validation of our results in a larger population of diverse ethnicities would be necessary to study the role of PI3K/ AKT/mTOR variants in disease development. Knowledge of the genetics of endometriosis can help healthcare professionals determine the underlying biological pathways that contribute to endometriosis, understand aetiology of this disease, define genetically heterogeneous subtypes, identify those at higher risk, and determine different treatments. These findings provide insights into the PI3K/ AKT/mTOR pathway's role in disease progression and help guide future research on early diagnosis, risk stratification, and potential therapeutic targets.

We used an online sample size calculator (https://clincalc. com/stats/SampleSize.aspx) to determine the sample size with a P1 of 47% that represented the frequency of the mutant allele in the controls and a P2 of 22% that indicated the frequency of the mutant allele in the cases. P1 and P2 were determined according to previously reported percentages (21). The study power was set at 80% and a 95% confidence interval (CI) with an obtained critical value of <2.0. The minimum sample size was adjusted for a total of 220 subjects. The study was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of the Zahedan University of Medical Sciences (IR.ZAUMS. REC.1401.080), Zahedan, Iran. All participants provided written informed consent for study participation. Peripheral blood samples were collected from 127 endometriosis patients (age: 26-45 years) and 125 healthy women (control group) matched for age and body mass idex (BMI) who referred to the Department of Obstetrics and Gynaecology at Ali-ibn-Abi Taleb Educational Hospital, Zahedan, Iran. The control group participants had no medical history of endometriosis. Individuals were excluded if they were not Iranian, withdrew consent, or had previous medical histories of chronic pelvic pain, dysmenorrhoea, or dyspareunia. Endometriosis was confirmed by clinicians and pathologists according to the results of laparoscopy or laparotomy and other complementary diagnostic tests. Following the diagnosis, histological examination confirmed the presence of an endometrial gland and/or stroma in the lesions. The endometriosis was staged according to the revised classification of the American Society of Reproductive Medicine (rASRM) (22). For subgroup analysis, the 127 endometriosis patients were divided into two groups – stages I and II (n=74, minimal/mild) and stages III and IV (n=53, moderate/severe). Genomic DNA was isolated from peripheral blood leukocytes using a commercial kit (DynaBio, Takapouzist, Iran) according to the manufacturer's instructions. The DNA was diluted in 100 µL of deionised water and stored at -20˚C until needed. Initially, the allelic discrimination of the PIK3CA rs2230461, AKT1 rs1130233, and mTOR rs2295080 polymorphisms were analysed by allele-specific polymerase chain reaction (AS-PCR) and polymerase chain reactionrestriction fragment length polymorphism (PCR-RFLP) according to a modified protocol (20, 22–24). The primer sequences, genotyping method, and fragment sizes are listed in Table 1. The genotypes were determined by separating the PCR products (digested product for mTOR SNP) with 3% agarose gel electrophoresis, and the results were visualised by safe stain (Sinaclon, Iran). In order to ensure the reliability of the genotyping results, we randomly chose 20 samples (15 homozygotes and five heterozygotes) and re-genotyped them using the same method. Statistical analyses were conducted using the statistical software package SPSS 21.0 (SPSS, Inc., Chicago, IL, USA). The student’s t test and chi-square test were conducted to compare demographic characteristics between patients and controls. The allele or genotype frequencies in each group were counted as the number of incidences of an individual allele divided by the total number of alleles. In order to determine significant differences in genetic distributions between cases and controls, both allele and genotype frequencies were compared using Pearson’s chi-square test. Odds ratio (OR) and 95% CIs were calculated to determine the risk of endometriosis associated with the given genotypes. Statistical significance was set at P<0.05.