RANKL is a new Epigenetic Biomarker for the Vasomotor Symptom During Menopause.

During menopausal transition, decreased level of estrogen brings a number of physiological problems and hormonal changes. In this study, promoter methylation of RANKL and FSHR genes were identified in 30 premenopausal and 35 postmenopausal women using methylation-specific high resolution melting (MS-HRM) analysis. The statistical analyses and their association with patient characteristics were performed by Pearson χ2 and Fisher's exact test (p <0.05). The methylated RANKL gene was detected in 16 postmenopausal cases, and 12 (75.0%) of the RANKL methylated cases had hot flashes (p = 0.024). The methylated FSHR gene was detected in 18 postmenopausal cases, and 13 (75.0%) of the methylated cases had hot flashes (p = 0.028). In vitro studies demonstrated the association between RANKL expression, FSH level and hot flashes in the mouse. Although lack of epigenetic studies in this field proves our results crucial and therefore, our results showed magnitude of epigenetic profiles of Turkish Cypriot post-menopausal women. This was the first study which has investigated the RANKL and FSHR methylation and their relationship with hot flashes in postmenopausal women.

Introduction

The first menstrual period and onset of menopause are the key milestones of female reproductive ageing [1]. Menopause occurs when the follicle pool in the ovaries has become exhausted and insufficient to maintain menstrual cycles [2]. Genetic and environmental factors are the major determinants of the ending and starting timing of these periods [3].

An increasing number of studies tried to identify the genetic background of reproductive ageing and disorders of reproduction [4, 5, 6, 7, 8]. Candidate-gene, genome-wide approach, linkage mapping and association studies tried to highlight genes that are related with the ending time of menstrual cycle. Genome-wide association studies (GWAS) described polymorphic loci in GNRH1, HMCES, ZCCHC27, ZNF518A, NRB2, LY6G5C and BAG69 genes in European [7] and Japanese [9] populations, which was linked with menopause. Day et al. [10] reported the PT-PRD, PTPRF, PTPRJ, PTPRK, PTPRS and PTPRZ1 genes to be related to the age of menarche. Epigenetic age is another most popular topic in different types of diseases and recent studies identified statistically significant association with breast cancer susceptibility and increased epigenetic age [10]. Levine et al. [11] used different tissue types to be able to demonstrate the interaction between epigenetic aging and age at menopause. They showed increased epigenetic age in blood samples and this was associated with earlier menopause. The association between single nucleotide polymorphism (SNP), rs11668344 and age of menopause was also identified by Levine et al. [11].

An increasing number of studies tried to identify the genetic background of age of menopause and bone mass differences in menopause. The CYP19A [12], HDC [13], ERα [14, 15, 16, 17], FSHR [18] gene polymorphisms are associated with hot flashes and bone mass during the menopause. In addition, until now, the interaction between epigenetics and hot flashes has not been reported in postmenopausal cases.

The RANKL gene is one of the members of the tumor necrosis factor (TNF) family [19,20] and RANKL/RANK genes are essential regulators of bone remodeling [21], lymph node organogenesis [22] and formation of a lactating mammary gland [23]. These genes are also expressed in the main region of thermoregulation [22] and have an important role for thermoregulation [11]. Hanada et al. [17] tested central RANKL injection in mice and rats and this triggered a severe fever response.

The follicle-stimulating hormone (FSH) is another important hormone in female reproduction which acts via binding to the FSH receptor (FSHR). The interaction between FSHR gene mutations and reproduction in the female has been described by researchers [24, 25, 26]. Woad et al. [27] identified mutations that were associated with ovarian dysgenesis (NM_000145.3: c.566C>T) and FSHR function (NM_000145.3: c.1043C>G and NM_000145.3: c.1555C>A). Sun et al. [28], reported that FSH is able to directly stimulate osteoclast formation and metabolic activities via its receptor (FSHR). In vitro studies showed that FSHR null mice models demonstrate that FSH is required for postmenopausal bone loss [26, 27, 28]. Until now, genetic studies in menopause has been limited and researchers started to analyze a pool of different study populations under a large consortium to increase their power. These studies are based on the assumption of common variants or mutations underlying common bone mineral density, age at menarche and menopause. An increased number of studies in epigenetics helped to identify the genetic background of diseases, such as cancer and aging. In this study, instead of identification of structural genetic variants in menopause, we identified the epigenetic background of the hot flashes in menopause. Here, we reported methylation analysis of FSHR and RANKL genes in patients from Cyprus and identification of an epigenetic background of hot flashes during the menopause.

Materials and methods

The FSHR and RANKL promoter methylation was analyzed in 35 postmenopausal and 30 premenopausal women. The study was performed in accordance with the Declaration of Helsinki and approved by the Research Ethics Committee of the Near East University, Nicosia, Turkish Republic of Northern Cyprus. Written informed consent was obtained from all subjects participating in the study.

All participants were genetically unrelated postmenopausal females. Subjects with an unnatural menopause, took medications such as anxiolytics, antidepressants, exogenous hormone, women who have serious disease or mental retardation, smoking, alcohol usage and have a weight loss therapy, food allergies, heart disease history, insulin-dependent diabetes, diabetes mellitus type 2 (DMT2), kidney or liver disease, were excluded. Individuals matching the menopause criteria for >1 year, were recruited [29].

Genomic DNA was extracted from blood samples according to the All Prep DNA/RNA/Protein isolation kit (Qiagen GmbH, Hilden, Germany), and its quantity was measured using a NanoDrop ND-1000 Spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA) at the Near East University Research Center of Experimental Health Sciences (DESAM), Nicosia, Turkish Republic of Northern Cyprus.

Determination of . To determine the RANKL and FSHR methylation status, first bisulfite modification reaction was applied, and in this reaction, unmethylated cytosine residues were converted to uracil by bisulfite treatment of 1.3 μg DNA using the EpiThet Bisulfite Kit (Qiagen GmbH) according to the manufacturer’s protocol. After conversion, DNA was eluted in buffer (Qiagen GmbH) to a final concentration of 30 ng/μL. Universal methylated and unmethylated DNA (EpiThet Control DNA Set; Qiagen GmbH) were used as methylated and unmethylated controls. We used Rotor Gene Q for methylation-sensitive high resolution melting (MS-HRM) analysis (Qiagen Ltd., Manchester, UK) to detect the methylation status of our samples. Primers were designed according to the EpiTect® HRM™ PCR (HRM-polymerase chain reaction) Handbook (Qiagen GmbH). We used comparable amounts of template genomic DNA for all samples resulting in cycle threshold (CT) values below 30 and differing by no more than three CT values as described previously [29].

Statistical Analyses. The statistical analyses and their associations with patient characteristics were performed by Pearson’s χ2 test and two-tailed Fisher’s exact test. Calculations were performed using the Statistical Package for the Social Sciences SPSS® 16.0 software (SPSS Demo for Mac, Chicago, IL, USA), with a statistical significance of p <0.05.

Results

The mean (± SD) age of the 30 premenopause patients was 33.5 years (33.5 ± 6.9) and for the postmenopause patients 56.7 (56.7 ± 4.9).

The . In the study group: RANKL promoter was methylated in 16 (45.7%) postmenopausal women and 19 (54.3%) were unmethylated (Table 1). In 12 (75.0%) of the 16 RANKL methylated postmenopausal cases had hot flashes. There was a statistically significant association identified between two groups (p = 0.024) (Table 2).

Table 1

Methylation status of RANKL and FSHR genes in pre- and postmenopausal women.

Genes	Postmenopause	Premenopause	Positive Hot Flashes History	Negative Hot Flashes History
RANKL (M) n	16 (45.7%)	10 (33.3%)	12 (75.0%)	4 (25.0%)
RANKL (U) n	19 (54.3%)	20 (66.7%)	7 (36.8%)	12 (63.2%)
FSHR(M) n	18 (51.4%)	20 (66.7%)	13 (72.2%)	5 (27.8%)
FSHR (U) n	17 (48.6%)	10 (33.3%)	6 (35.3%)	11 (64.7%)

M: methylated; U: unmethylated.

Table 2

Methylation status of RANKL and FSHR genes and postmenopausal women who have hot flashes.

Genes	Positive Hot Flashes History	Negative Hot Flashes History	p Value
RANKL (M) n	12 (75.0%)	4 (25.0%)	0.024
RANKL (U) n	7 (36.8%)	12 (63.2%)
FSHR(M) n	13 (72.2%)	5 (27.8%)	0.028
FSHR (U) n	6 (35.3%)	11 (64.7%)

M: methylated; U: unmethylated.

In the control group, 10 (33.3%) of the samples were methylated and 20 (66.7%) of the samples were unmethylated (Table 1). There was no statistically significant association between menopause and methylation of the RANKL promotor (p >0.05) (Figure 1).

Figure 1

Methylated RANKL patient. The RANKL unmethylated control is purple, methylated control is red. Patient #50 was methylated and patient #52 was unmethylated.

The . The FSHR gene was methylated in 18 post-menopausal cases and unmethylated in 17 postmenopausal cases. In the control group, FSHR genes were methylated in 20 cases and unmethylated in 10 cases. There was no statistically significant association between two group (p >0.05) (Table 2).

Thirteen (75.0%) of the 18 FSHR methylated post-menopausal cases had hot flashes. Statistically significant association was identified between the two groups (p = 0.028) (Table 2).

Discussion

Several studies focused on identification of interactions between methylation and cancer. Ambatipudi et al. [2], worked with epigenetic ageing and they showed accelerated epigenetic age was associated with CpG island methylation and breast cancer susceptibility in postmenopausal women [2]. Epigenetic-based studies in postmenopausal women is limited and the largest part of the studies were related to mutational screening. To the best of our knowledge, this study is the first study which shows the epigenetic alterations in menopause and the epigenetic background of hot flashes in postmenopausal women.

Mutations of the FSHR gene analyzed in different groups, such as premature ovarian failure (POF) [30, 31, 32, 33, 34, 35, 36, 37], low number of oocytes [36], age of menopause [37], infertility [38], migraine [39], bone mineral density and bone turnover [18]. Woad et al. [27] searched FSHR variants in exons 7 and 10 in patients who suffered with POF. They identified a novel heterozygous NM_ 000145.3:c.1411A>T variant in FSHR exon 10 in a family who have a history of POF and elevated FSH concentration [27]. The FSHR mutation and POF interaction has been studied by several researchers [30, 31, 32, 33, 34, 40].

Rendlina et al. [34] found that the FSHR rs6166 polymorphism significantly influenced bone mineral density (BMD) in postmenopausal women. Cordts et al. [8] concluded that the NM_000145.3: c.919G>A polymorphism in the FSHR gene was associated with premature ovarian insufficiency (POI) development and could be used as a screening marker in patients with ovarian failure [8].

Our study is the first study based on current literature that has shown the interaction between FSHR promoter methylation and hot flashes in postmenopausal patients. The methylated FSHR gene detected in 18 postmenopausal cases and 13 (75.0%) of the methylated cases had hot flashes. There was statistically significant association between the two groups (p = 0.028). This was the first study which showed the interactions between FSHR methylation and hot flashes in postmenopausal women.

The RANKL gene is important for bone homeostasis and related biological processes [41]. In fact, the RANKL gene has been linked as a candidate gene for osteoporosis susceptibility and also confirmed by GWAS, as a susceptibility locus regulating BMD [42]. Single nucleotide polymorphisms of the RANKL promoter showed association with BMD [41,43,44]. Mencej et al. [25] showed that promoter polymorphisms of the RANKL gene was related with elevated transcriptional activity of the gene and correlated with increased RANKL protein rate. This caused imbalances in bone metabolism and was related with low BMD [44]. Shang et al. [40] analyzed SNPs of the RANK and RANKL genes in a Chinese female population and showed that the RANKL polymorphisms related with BMD in the femoral neck in peri- and postmenopausal Chinese women [41]. Researchers tried to identify the importance of genetic variations in peri- and postmenopausal women and BMD. The importance of SNPs in BMD has been shown by different researchers [27,38,45,46].

Delgado-Calle et al. [12] showed that increased expression of RANKL during osteoporotic fractures but did not show statistically significant association between methylation and osteoporotic fractures. In vitro studies demonstrated the association between RANKL expression and thermoregulation during postmenopause and also homozygous NM _003701.3: c.508A>G mutation in the RANKL gene was associated with impaired fever response [47].

Hanada et al. [16] demonstrated that central RANKL injection in mice and rats triggered severe fever and they mapped RANK during the fever response on astrocytes, on the other hand, when they applied high doses of RANKL in the intraperitoneal this did not cause any changes in body temperature or in activity. They highlighted RANKL as an important fever inducer in the central nervous system [16]. Hanada et al. [16] concluded that RANKL/RANK regulates female body temperature and they also demonstrated that ovariectomy-induced changes in core body temperatures occurred in RANK knockout female mice. Hanada et al. [16,17] identified that genetical inactivation of RANK in the brain resulted in an altered physiological thermoregulation in female mice, which, at least in part, appears to be regulated by ovarian sex hormones.

In this study, we detected RANKL promotor methylation and hot flashes in 12 (75.0%) postmenopausal women (p = 0.024) (Table 2). Researchers highlighted that RANKL used a COX2-PGE(2)/EP3R pathway for thermo-regulation and induction of fever. We suggested that increased methylation of RANKL directly effects the expression on the gene and this causes abnormal fluctuation of temperature on postmenopause patients. At the same time, researchers highlighted that deficiency of RANKL/RANK expression is related to osteoporosis and irregular resorption of bone. In accordance with the literature, we can conclude DNA methylation is another mechanism which decreased gene expression and increased irregular resorption of bone but this hypothesis also needs confirmation on bone cells. Here, consistent with the Hanada et al. [16,17] findings, we demonstrated significant association between the RANKL gene methylation and hot flashes in postmenopausal women. On the other hand, we have found significant differences in the gene methylation profiles of the postmenopausal and premenopausal women, including genes that their epigenetic changes have not yet been associated with postmenopausal women. Our findings may provide further insight into the process of postmenopausal changes and gene methylation status. This was the first study that investigated the epigenetic changes of RANKL and FSHR genes in postmenopause and their relation with hot flashes.

Conclusions. In conclusion, we found significant association between the RANKL and FSHR gene methylation and hot flashes in postmenopausal women. Future studies, in larger samples of postmenopausal women, should focus on the study of different gene methylations, and this will help to clarify potential effects of gene methylation in menopause.