Implication of FSHB rs10835638 variant in endometriosis in Brazilian women

Endometriosis is an estrogen-dependent, chronic, inflammatory, debilitating disease associated with pelvic pain and infertility. It is characterized by the presence of ectopic endometrial glands and stroma, predominantly but not exclusively in the pelvic compartment. It affects 5%-10% of women of reproductive age ( 1 - 3 ) and can severely compromise their quality of life. ( 4 ) Moreover, endometriosis is observed in 50%-80% of women with pelvic pain and in up to 50% of women with infertility. ( 5 ) Estrogens fuel ectopic endometrial growth. In addition to the production of estradiol by the ovaries, local production of estradiol in endometriotic lesions is mediated by steroidogenic enzymes, including aromatase. ( 6 ) Ectopic endometrial tissue overexpresses estrogen receptor beta (ER-beta, encoded by ESR2 ), resulting in the suppression of estrogen receptor alpha (ER-alpha, encoded by ESR1 ), leading to diminished ER-alpha-mediated induction of the progesterone receptor. ( 7 ) Ultimately, ER-beta promotes cell survival and perpetuates inflammation. At the same time as estrogen overproduction and ER overexpression, progesterone resistance occurs in endometriotic tissue, impeding the modulation of genes involved in decidualization, cell cycle regulation, and estrogen response inhibition. ( 8 ) Progesterone is a potent antagonist of estrogen-induced proliferation in the endometrium and acts as an anti-inflammatory agent. ( 3 , 9 ) Several theories have been proposed with regards to the possible factors responsible for the development of endometriosis. However, none of the existing theories are able to explain all of the implantation sites and symptoms of the disease. Retrograde menstruation is the pathogenic hypothesis supported by the most robust evidence regarding the mechanism of endometrial peritoneal implants. ( 8 , 10 ) Endometriosis is heritable and is influenced by multiple genetic and environmental factors. Genetic factors contribute to about half of the variation in the endometriosis risk, which has an estimated hereditability of 51%, ( 2 , 11 ) although no distinct inheritance pattern has been established for the disease. Several studies have highlighted the association between single nucleotide variants (SNVs) and an increased risk of endometriosis in populations of different ethnic origins. ( 12 - 15 ) In a UK Biobank cross-sectional study, the T allele of the FSHB c.-211G>T polymorphism (rs10835638) was associated with detrimental effects on fertility and protective effects against endometriosis. ( 16 ) The β-subunit of the follicle-stimulating hormone (FSH), encoded by the FSHB gene, ensures the binding specificity of the hormone to FSH receptors, enabling the hormone to stimulate the growth of ovarian follicles and production of estrogens. ( 17 , 18 ) The abovementioned SNV in the promoter region of the FSHB gene reportedly influences assisted reproductive treatments as well as the concentrations of FSH, luteinizing hormone (LH), and progesterone. ( 18 )

This case-control study included 326 fertile and infertile women with endometriosis and a Control Group of 482 fertile women without endometriosis. The study design was approved by the independent Research Ethics Committee of Centro Universitário FMABC (CAAE: 13974913.4.0000.0082; # 310.094). Each patient enrolled in this study signed an informed consent form for all procedures and for biological sample and data collection and analysis for research purposes. The endometriosis group comprised 326 women (103 fertile and 223 infertile) who underwent in vitro fertilization (IVF) treatment at the Human Reproduction and Genetics Center of Centro Universitário FMABC , Santo André, Brazil between 2014 to 2019. Inclusion criteria were: age ≤38 years; diagnosis of endometriosis confirmed by laparoscopy or laparotomy and histology, classified according to the revised American Society for Reproductive Medicine (rASRM) score; ( 19 ) the presence of both ovaries without morphological abnormalities; ovulatory cycles lasting 25 to 35 days; and no evidence of endocrine diseases, such as hyperprolactinemia, thyroid dysfunction, or polycystic ovary syndrome. Women with a previous ovarian surgery, those who underwent chemo/radiotherapy, and those with endometrial polyps, hydrosalpinx, or submucosal and/or intramural fibroids were excluded from the study. The investigation of the cause of infertility included hormonal and biochemical profiling, testing for sexually transmitted diseases, imaging examinations, investigation of genetic and/or immunological abnormalities, hysterosalpingography, hysteroscopy, laparoscopy, and semen analysis of the partner. ( 20 , 21 ) We considered women who had undergone laparoscopy/laparotomy to treat endometriosis and who had failed to achieve pregnancy spontaneously or by IVF treatment after surgery within a maximum period of 12 months, as infertile. The Control Group comprised 482 fertile women who had undergone permanent contraception by laparoscopy. The absence of endometriosis was confirmed by inspection of the pelvic cavity during the procedure, and none of these women had a personal and/or familial history of endometriosis, autoimmune diseases, or cancer. Whole-blood samples were collected in EDTA-containing tubes. DNA was extracted from peripheral lymphocytes by using a standard salting-out method. ( 22 ) Genotyping of the NG_008144.1 :g.4790G>T (c.-211G>T, rs10835638) variant of the FSHB gene was performed using the TaqMan system and real-time PCR. Assays were performed using primers, probes, and Master Mix (Thermo Fisher Scientific, Waltham, MA, USA), with 25ng of DNA per reaction. The PCR conditions were as follows: 50 cycles of denaturation at 95°C for 15 seconds, and annealing/extension at 60°C for 90 seconds. Statistical analyses were performed using the R programming language (R Core Team, 2018). ( 23 ) The allele and genotype frequencies were compared between groups by using the χ 2 test, as well as the estimation of Hardy-Weinberg equilibrium to compare the genetic models (dominant or recessive models) of the SNV. Odds ratios and 95% confidence intervals (CIs) were calculated in relation to the presence of the reference genotype by using a logistic regression model. Statistical significance was set at p<0.05.

The mean age of the endometriosis group was 32.8±2.8 years; 39.9% had minimal/mild disease, and 60.1% had moderate/severe endometriosis. Moreover, 68.4% of women in this group were considered infertile. The Control Group had a mean age of 31.5±4.6 years. Genotype and allele frequencies of the FSHB rs10835638 variant in women with endometriosis compared to those in controls are summarized in table 1 . The genetic models of the variant are summarized in table 2 . Only four women with endometriosis and seven in the Control Group presented with the homozygous variant TT genotype. The genotype frequency of the FSHB variant was in Hardy-Weinberg equilibrium in all studied groups. Table 1 Genotype and allelic frequencies of the variant follicle stimulating hormone subunit beta rs10835638: G>T in women with endometriosis and controls Population n Genotypes Alleles OR (95%CI) HWE GG GT p value TT p value G T p value Endometriosis Overall 326 259 (79.5) 63 (19.3) 0.060 4 (1.2) 0.862 581 (89.1) 71 (10.9) 0.125 1.30 (0.93-1.81) 0.998 Minimal/Mild 130 96 (73.9) 31 (23.8) 0.007 3 (2.3) 0.388 223 (85.8) 37 (14.2) 0.006 1.76 (1.16-2.67) 0.995 Moderate/Severe 196 163 (83.1) 32 (16.4) 0.536 1 (0.5) 0.314 358 (91.3) 34 (8.7) 0.974 1.01 (0.66-1.53) 0.980 Fertile Overall 103 81 (78.6) 20 (19.4) 0.182 2 (2) 0.656 182 (88.3) 24 (11.7) 0.169 1.40 (0.86-2.27) 0.847 Minimal/Mild 49 35 (71.5) 12 (24.5) 0.046 2 (4.0) 0.112 82 (83.7) 16 (16.3) 0.012 2.07 (1.16-3.70) 0.913 Moderate/Severe 54 46 (85.2) 8 (14.8) 0.956 0 (0.0) 0.373 100 (92.6) 8 (7.4) 0.670 0.85 (0.40-1.81) 0.951 Infertile Overall 223 178 (79.8) 43 (19.3) 0.099 2 (0.9) 0.592 399 (89.5) 47 (10.5) 0.244 1.25 (0.86-1.82) 0.945 Minimal/Mild 81 61 (75.3) 19 (23.5) 0.036 1 (1.2) 0.964 141 (87.0) 21 (13.0) 0.076 1.58 (0.95-2.63) 0.989 Moderate/Severe 142 117 (82.4) 24 (16.9) 0.467 1 (0.7) 0.505 258 (90.8) 26 (9.2) 0.774 1.07 (0.67-1.70) 0.998 Controls   482 406 (84.2) 69 (14.3) -- 7 (1.5) -- 881 (91.4) 83 (8.6) -- -- 0.980 OR: odds ratio; 95%CI: condidence interval; HWE: hardy-weinberg equilibrium. OR: odds ratio; 95%CI: condidence interval; HWE: hardy-weinberg equilibrium. Table 2 Genetic models of the variant rs10835638: G>T of the follicle stimulating hormone subunit beta gene in women with ndometriosis and controls Population n Dominant model Recessive model GG+GT TT p value OR (95%CI) GG GT+TT p value OR (95%CI) Endometriosis 326 322 4 0.785 0.84 (0.24-2.90) 259 67 0.080 1.38 (0.96-1.99) Minimal/Mild 130 127 3 0.494 1.60 (0.41-6.29) 96 34 0.006 1.89 (1.19-30.00) Moderate/Severe 196 195 1 0.303 0.35 (0.04-2.85) 163 33 0.073 1.08 (0.69-1.69) Fertile 103 101 2 0.714 1.34 (0.28-6.56) 81 22 0.167 1.45 (0.85-2.47) Infertile 223 221 2 0.541 0.61 (0.13-2.98) 178 45 0.148 1.35 (0.90-2.03) Controls 482 475 7   Reference 406 76   Reference OR: odds ratio; 95%CI: condidence interval. OR: odds ratio; 95%CI: condidence interval. The genotype and allele frequencies of the FSHB rs10835638 variant did not differ between women with and those without endometriosis. Upon subgroup analysis of fertile and infertile women with endometriosis, these frequencies also did not differ. However, the subgroup with minimal/mild endometriosis had a higher frequency of the GT genotype than the Control Group (23.8% versus 14.3%, p=0.007). The frequency of the GT genotype was also higher in the subgroup with minimal/mild endometriosis in both fertile (24.5% versus 14.3%, p=0.046) and infertile women (23.5% versus 14.3%, p=0.036) than that in the Control Group. Regarding the genetic models, the T allele was more common in women with minimal/mild endometriosis than that in the Control Group in the recessive model (p=0.006, 95%CI=1.89 (1.19-3.00). In the table 3 we presented the frequency of the variant allele T of the FSHB rs10835638:G>T SNV in different population. The data was extracted from dbSNP (Database of Single Nucleotide Variants) (https://www.ncbi.nlm.nih.gov/snp/rs10835638) based on the ALFA project (Allele Frequency Aggregator) that presents allele frequency pooled data from dbSNP and summarized results from dbGaP (Database of Genotypes and Phenotypes), which are developed databases and hosted by the of National Center for Biotechnology Information (NCBI) in collaboration with the National Human Genome Research Institute (NHGRI). dbSNP is a free public archive that presents genetic variation within and between different species, while dbGAP contains the results of over 1,200 studies, over 2 million individuals and hundreds of millions of variants, along with thousands of phenotypes and data of molecular assays. The data was also extracted from the ABraOM (Brazilian Online Mutations Archive) repository that contains genomic variants obtained from the exome and complete genome sequencing of SABE (a longitudinal study on Health, Well-Being and Aging) in a census sample of 1,171 elderly people in the city of São Paulo, Brazil. The data presented in table 3 makes clear the diffenrence of allele frequency according to population studied, varying from 0% in African and Latin American to 10.9% in Brazilian people. Table 3 Variant allele frequency in different populations according to the ALFA and ABraOM project Population SNV rs10835638:G>T n VAF Global 17118 6.5 European 12502 8.1 African 3118 2.1 African american 3010 2.1 Others † 108 0.0 Asian 138 1.4 East Asian 114 1.8 South Asian 88 1.0 Other§ 24 0.0 Latin American 1 ‡ 104 0.0 Latin American 2 # 456 0.0 ABraOM 1,171 10.8 Present study     Endometriosis 326 10.9 Control 482 8.6 † Other Africans: individuals of African descent; § Other Asians: Individuals of Asian descent; ‡ Latino-Americans 1: Latin American individuals with Afro-Caribbean ancestry; # Latin Americans 2: Latin American individuals of primarily European and Native American ancestry. ALFA projetc: allele frequency aggregator; ABraOM: Brazilian Online Mutations Archive; VAF: variant allele frequency; n: number of genotyped individuals. † Other Africans: individuals of African descent; § Other Asians: Individuals of Asian descent; ‡ Latino-Americans 1: Latin American individuals with Afro-Caribbean ancestry; # Latin Americans 2: Latin American individuals of primarily European and Native American ancestry. ALFA projetc: allele frequency aggregator; ABraOM: Brazilian Online Mutations Archive; VAF: variant allele frequency; n: number of genotyped individuals.

To investigate the possible relationship between the FSHB gene variant (c.-211G>T, rs10835638) and the development and/or progression of endometriosis.

To the best of our knowledge, this is the first study in which the relationship between the FSHB rs10835638 variant and the development and/or progression of endometriosis was investigated in Brazilian women. The homozygous TT genotype was very rare in this study (a prevalence of 1.4%), detected in only four women in the endometriosis group and seven in the Control Group. The T allele frequency was 10.9% in the endometriosis group and 8.6% in the Control Group, similar to that observed in the European population (14.5%) and higher than that detected in the African population (2.8%). ( 24 ) In a study conducted by Polyzos et al., ( 25 ) the prevalence of the TT genotype was 0.5% among a study sample that included women from Europe and Asia; it occurred only in European participants. In a cross-sectional study by the UK Biobank, that included white British individuals aged 40-69 years, FSHB T allele of the rs10835638 variant was tested for associations with mainly female reproductive phenotypes in 119958 individuals (up to 63350 women and 56608 men). Also a GWAS was conducted in 9534 individuals to identify genetic variants associated with length of menstrual cycle. The results showed that the T allele was associated with longer menstrual cycles and increased age at menopause. Moreover, although it showed detrimental effects on fertility, it was protective against endometriosis. ( 16 ) On the other hand, in the presente study a case-control analysis was performed to investigate the possible relationship between the FSHB variant and the development and/or progression of endometriosis in 808 women who had the the pelvic cavity inspectioned to confirm or exclude endometriosis. The genotypes and allelic frequencies of the FSHB variant were not different between women with and without endometriosis, even when the group was subdivided into fertile and infertile. However, considering the stage of the disease, the group with minimal/mild endometriosis had a higher frequency of the GT genotype, regardless of fertility status, and the T allele was more frequent in women with minimal or mild endometriosis than in the Control Group in the recessive model. This is in contrast to the study by Ruth et al. ( 16 ) which indicated harmful effects of the T variant allele on fertility and protective effects against endometriosis. The results of the present study suggest that the T allele is associated with the development of minimal/mild endometriosis. So, the patients selection was different and also the genetic background of the population studied, as well as the methodological approach. These factors together may be responsible for the discordant findings between the two studies. FSHB encodes the hormone-specific βsubunit of FSH, a key promoter of ovarian follicle growth and estrogen production. The rs10835638 variant is located in the promoter region of the gene, where the transcription factor LHX3 binds, regulating gene expression. In vitro , transcriptional activity is reportedly reduced by 50% in cells carrying the T allele. ( 26 ) In the study by Schüring et al., ( 18 ) which included 365 women with normal ovulation who underwent IVF treatment owing to male infertility, the T allele was associated with higher FSH and LH concentrations and lower progesterone concentrations. Busch et al. ( 27 ) observed that, in healthy peripubertal girls, the presence of the T allele was negatively associated with the number of follicles. Although the rs10835638 variant potentially decreases the serum FSH concentration, which could alter estrogen production as well as follicular cell growth and dominance, the results of the present study indicate that the genotype and allele frequency of this variant did not differ between fertile and infertile women with endometriosis. However, women with minimal/mild endometriosis had a higher frequency of the T allele than women without endometriosis, regardless of fertility. The association between endometriosis and infertility is well established, but the mechanisms remain unclear. Severe infiltrative endometriosis is associated with impairment of fertility; however, the mechanism by which such a small lesion, observed during laparoscopy, leads to infertility is unknown. Recently, Kasapoglu et al. ( 28 ) confirmed a faster decline in anti-Müllerian hormone (AMH) concentrations in women with endometriomas who did not undergo surgical or medical therapy than that in healthy women. The median percentage decline in serum AMH concentration after six months of follow-up was 26.4% in the endometrioma group and 7.4% in the Control Group. Moreover, oocyte quality may be affected in women with endometriosis. A study in which women with and those without endometriosis who partook in a donation program were compared revealed that oocytes from donors with endometriosis were associated with significantly lower implantation and pregnancy rates compared with donors with tubal infertility, suggesting an alteration in oocyte quality in women with endometriosis, decreasing implantability. ( 29 ) In contrast, Hong et al. ( 30 ) compared IVF results between women with a decreased ovarian reserve after surgery for endometrioma and women with a decreased ovarian reserve who did not undergo ovarian surgery; they discovered no difference in FSH or AMH concentrations or antral follicle count in response to controlled ovarian stimulation, nor in assisted reproduction outcomes. Bianco et al. ( 31 ) evaluated the rs10835638 variant and hormonal profile and reproduction outcomes of 213 women with endometriosis; they discovered that women carrying the T allele had significantly higher serum LH concentrations, but that their assisted reproduction outcomes did not differ from those without it. As the pathogenesis of endometriosis is highly complex, involving both genetic background and environmental conditions, conflicting results have complicated the interpretation of the relevant data. An advantage of our study was the selection of a homogeneous patient cohort and a Control Group with laparoscopic examination of all patients. With regard to the study limitations, only one FSHB variant was was investigated, fertility data were considered, but not hormonal profiles. Moreover, this variant may have a small effect, and its frequency may vary according to the study population.

Our results suggest that the FSHB rs10835638 variant is associated with the development of minimal/mild endometriosis in Brazilian women. When approaching a characteristic as complex as endometriosis, the characterization of genotypes that can provide data on predisposition or prognosis are relevant and desirable, even considering the methodological difficulties involved.