Lack of Influence of the ACE1 Gene I/D Polymorphism on the Formation and Growth of Benign Uterine Leiomyoma in Turkish Patients

Uterine leiomyomas (ULM), are benign tumors of the smooth muscle cells of the myometrium. They represent a common health problem and are estimated to be present in 30-70% of clinically reproductive women. Abnormal angiogenesis and vascular-related growth factors have been suggested to be associated with ULM growth. The angiotensin-I converting enzyme (ACE) is related with several tumors. The aim of this study was to identify possible correlation between ULM and the ACE I/D polymorphism, to evaluate whether the ACE I/D polymorphism could be a marker for early diagnosis and prognosis. ACE I/D was amplified with specific primer sets recognizing genomic DNA from ULM (n=72) and control (n=83) volunteers and amplicons were separated on agarose gels. The observed genotype frequencies were in agreement with Hardy-Weinberg equilibrium (χ 2 =2.162, p=0.339). There was no association between allele frequencies and study groups (χ 2 =0.623; p=0.430 for ACE I allele, χ 2 =0.995; p=0.339 for ACE D allele). In addition, there were no significant differences between ACE I/D polymorphism genotype frequencies and ULM range in size and number (χ 2 =1.760; p=0.415 for fibroid size, χ 2 =0.342; p=0.843 for fibroid number). We conclude that the ACE gene I/D polymorphism is not related with the size or number of ULM fibroids in Turkish women. Thus it cannot be regarded as an early diagnostic parameter nor as a risk estimate for ULM predisposition.


Introduction
Uterine leiomyomas (ULM), are benign tumors of the smooth muscle cells of the myometrium. They represent a common health problem and are estimated to be present in 30-70% of clinically reproductive women (Baird et al., 2003;Ekin et al., 2014;Tal and Segars, 2014). Although, ULM are rarely associated with mortality; they may have shown that ULMs are influenced by genetics risk factors (Edwards et al., 2013), additionally; Hakverdi and colleagues (2013) recently discovered novel chromosome aberrations in ULM patients (Hakverdi et al., 2013).
The aim of this study is to identify a possible correlation between ULM formation risk, fibroid size and number and ACE I/D polymorphism. Consequently to evaluate weather the ACE I/D polymorphism could be suggested as a marker in the early diagnosis and possibly improve quality of life of affected women.

Materials and Methods
To conduct this study we used 2 groups, where healthy women (n=83) and patients (n=72) had pelvic imaging to detect the presence of ULM. The measurement of each fibroid was assessed with 3 perpendicular diameters: length, width and depth. Patients were grouped according to their fibroid number, being one or multiple; the latter including two up to 13 separate fibroids. A second grouping was made by tumor size, with the threshold selected at 5 cm. The presence of pelvic pain and abnormal bleeding was noted, however not all women showed these signs. This study was approved by the Yeditepe University's Ethical Committee and informed, written consent was obtained from all volunteer participants.

DNA isolation and genotyping for ACE I/ D polymorphism
Peripheral blood (10 ml) was collected into EDTA tubes in Departments of Gynaecology and Obstetrics, Istanbul and Yeditepe Universities, . Genomic DNA was extracted from peripheral whole blood using iPrep PureLink gDNA Blood Kit with the iPrep Purification Instrument (Invitrogen, Life Technologies, NY, USA). Consequently, DNA samples were quantified (μg) and qualified (260/280 and 260/230) by the use of Nanodrop. The DNA samples were stored at +4°C until genomic studies were conducted.
Polymerase chain reactions (PCR) were performed with 10 pmol for each primer: forward primer 5'-CTGGAGACCACTCCCATCCTTTCT-3', reverse primer 5'-GATGTGGCCATCTTCGTCAGAT-3' in final volume of 25 μl containing 1.5 µl MgCl2, 0.5 mM of each dNTP (PCR grade, Invitrogen, Life Technologies, NY, USA), and 0.25 unit Taq Polymerase (BIORON GmbH, Germany). Amplifications were carried out in a DNA Thermal Cycler (Verite Thermal Cycler, Applied Biosystems) with an initial denaturation step at 94°C for 5 minutes, fallowed by 30 cycles of denaturation at 94°C for 1 minute, annealing at 58°C for 1 minute and extension at 72°C for 2 minutes, and a final extension at 72°C for 2 minutes as previously described (Rigat et al., 1992). PCR products were separated on 3% agarose gel and DNA was visualized by ethidium bromide staining. The amplification products at 190 and 490 bp represent, the homozygous D and I alleles respectively. Those with both bands present represent the heterozygous ID alleles. Samples with DD genotype, were retyped with a second primer set, 5'-TGGGACCACAGCGCCCGCCACTAC-3' and 5'-TCGCCAGCCCTCCCATGCCCATAA-3', that specifically recognizes the insertion sequence, as described earlier (Ergen et al., 2004). Amplicons were obtained with a similar PCR cycle, with the annealing temperature set at 61°C. A band present at 335 bp was considered to correspond to the I allele, thus those samples were interpreted as heterozygous ID alleles.

Statistical analysis
Frequency and statistical analysis were performed with SPSS 20.0. Data were presented as mean±standard deviation (SD) or as proportions. A p<0.05 was defined as statistically significant. Expected and observed frequencies of genotypes and alleles were compared with Chi-Square analysis and Fisher's exact tests. Nominal values were analyzed with Student-T test.
The sample size was calculated with the assumption that the prevalence of ULM is 20-30% of the population. To provide 80% power with a 0.05 alpha error, a minimum of 70 participants was found to be adequate, thus our study population (n=72) was readily in the correct power.

Results
A total of 72 patients with ULM and 83 controls were recruited for this study. The mean age of patients and healthy controls were 36.64±9.63 and 39.18±11.98 years, respectively. No significant difference was found between patients and controls in terms of median age (p=0.297).
Products amplified with the first and second primer sets were analysed using 3% agarose gel electrophoresis (Figure 1 and 2). Products at 190 and 490 bp obtained with the first primer set, correspond to the homozygous D and I alleles respectively, those with both bands correspond to heterozygote genotype. Subjects with DD genotype were retyped with a second primer set, where a 325 bp product correspond to the I alleles and genotyped as heterozygote ID.
The genotype and allele frequencies of ACE gene polymorphism among study and control groups are shown in Table 1. The observed genotype frequencies of ACE gene polymorphism in study groups were in agreement with Hardy-Weinberg equilibrium (χ 2 =2.162, p=0.339). Similarly, there was no association between allele frequencies and study groups (χ 2 =0.623; p=0.430 for ACE I allele, χ 2 =0.995; p=0.339 for ACE D allele).

Discussion
Here in we have compared the effects of the ACE I/ D polymorphism in ULM patients and healthy controls in a Turkish cohort. The benign ULMs represent the most common reason of hysterectomies in women and the estimated cost of ULM associated care has been reported as 34 billion dollars in the US only (Cardozo et al., 2012), representing a considerable burden on the healthcare system (Catherino et al., 2013). Although, a similar estimate was not reported for the European population, this number could potentially lower in the white race. However, it is well documented that the black race is more susceptible to develop ULMs compared to the white (Baird et al., 2003). It is reported that the disease is seen in 30-70% of reproductively active women and the risk diminishes post-menopause, thus exposure to estrogen significantly increases the risk of tumor formation (Baird et al., 2003;Catherino et al., 2013). Early age-at-menarche is also reported as one of the indications, as it provides a longer exposure to estrogen (Flake et al., 2003). Despite the implication of several etiologies for UML pathology, specific pathways and potential gene interactions are yet to be discovered. Heritability studies conducted in several European populations showed that 26-69% of ULMs were due to genetic factors (Edwards et al., 2013).
The first genome-wide association study for ULM, was conducted in the Japanese population by Cha and colleagues, where they discovered 11 single nucleotide polymorphisms (SNPs) in 3 different chromosomal regions (Cha et al., 2011). Edwards and coworkers (2013) replicated the results from the Japanese study in the European American population and found a strong association between ULM risk and BET1L and TNRC6B genes (Edwards et al., 2013). Enzymes and their respective genes involved in the estrogen metabolic pathway, such as COMT, CYP1A1 and CYP1B1 have been investigated in the Han Chinese population (Shen et al., 2014). It was concluded that COMT and CYP1A1 had protective effects on the formation UMLs. Ateş and colleagues investigated the COMT Val158Met polymorphism and revealed its relevance to the formation of large fibroids. However, it was not attributed with the risk of ULM formation in a Turkish cohort (Ateş et al., 2013). Additionally, chromosomal structural aberrations have recently been reported in ULM patients, including, deletions, breaks and fragilities in several chromosomes (Hakverdi et al., 2013). Thus, it is well recognized by the scientific community, that the most common benign fibroid formation of the women reproductive tract depends on a genetic background.
The female reproductive organs, specifically the uterus undergoes cyclic physiological angiogenesis (Herr et al., 2013;Tal and Segars, 2014). Angiogenesis and sprouting of new blood vessels from existing ones is critical for the development of ULM (Risau, 1997;DiLieto et al., 2005;Tal and Segars, 2014). It has been previously stated that angionesis and vascular-related factors are involved in the formation and growth of ULM (DiLieto et al., 2005). Leiomyomas contain abnormal vascularization compared to the normal surrounding myometrium, such as that the vascular density of fibroids contains an avascular core, however they are enclosed in a dense vascular capsule (Tal and Segars, 2014).
The ACE gene variations have been implicated in several chronic diseases, including coronary heart diseases (Alvarez et al., 2000), hypertension (Henskens et al., 2003), renal diabetes (Ergen et al., 2004) and recently in panic attack disorder (Gulec et al., 2014). The homozygous DD is related to serum and tissue ACE levels, and evidence suggests that the D allele is associated with elevated plasma and serum ACE levels (Rigat et al., 1990). The ACE or kinase II enzyme is part renin-angiotensin system (RAS), which leads to hypertension by vasoconstriction as well as salt retention (Timmermans et al., 1993), thus regulating body fluid homeostasis. The RAS system has been characterized as a circulating hormonal system (Dinh et al., 2001). However, not many studies investigated the relevance of the RAS and the uterus (Herr et al., 2013).
In terms of cancer development, angiogenesis is essential for tumor progression and proliferation (Tal and Segars, 2014), and ACE activity has been related with tumor growth (Fishchuk and Gorovenko, 2013). Such Values are reported as number of patients (percentage of the total group); n: number of individuals; *p values less than 0.05 denoted statistical significance as that ACE inhibitors and angiotensin receptor blockers have been shown to suppress tumor growth (Mc Menamin et al., 2012). In relation to cancer, ACE1 gene I/D polymorphism has been widely investigated in Caucaisians and Asian populations. Meta-analysis reports conducted with the Chinese population reveals a lack of association between ACE1 I/D polymorphism and breast (Li et al., 2014), lung ), nasopharyngheal (Li et al., 2011 cancers and polycystic ovary syndrome (Jia et al., 2013). However an emphasis on racial differences is noticeable in some of the meta-analysis studies that further conducted subgroup analysis stratified by ethnicity (Li et al., 2011;Jia et al., 2013). A recent meta-analysis from Zhang et al. (2014), involving several cancer types, including breast, lung, colorectal, gastric and prostate cancers demonstrated that ACE1 I/D polymorphism was related to Caucasians but not to Asians (Zhang et al., 2014). In contrast, Wei et al. (2014) reported that the D allele was significantly associated with increased lymph node metastasis and advanced clinical stage for gastric cancer in the Chinese population (Wei et al., 2014). Furthermore, the D allele was related with increased risk of metastasis in colorectal cancer patients in Chinese subjects (Liu et al., 2011). Thus, discrepancies exist for different types of cancers. The Dutch population revealed increased risk for breast cancer development in subjects with the DD genotype (van der Knaap et al., 2008). The DD genotype was found to be associated with prolonged patient survival in favor for women with colorectal cancer in a German study (Rocken et al., 2007). The D allele was associated with an elevated risk in the formation of breast cancer in Ukrainian women (Fishchuk and Gorovenko, 2013). The I allele was found to be protective in women subjects with basal cell carcinoma compared to men (Yapijakis et al., 2013). One possible explanation for a mostly carcinogenic D allele might be due to its amplifier affects on ACE gene expression, increasing angiotensin II production, which in turn leads to vasoconstriction, high blood pressure, cell proliferation and neovascularization (Fernandez et al., 1985;Yapijakis et al., 2013).
Beyond the formation of tumors, it is well known that angiotensin 2 receptor is highly abundant in the myometrium (Matsumoto et al., 1996;Dinh et al., 2001) and disappears during pregnancy (Matsumoto et al., 1996). Suggesting a possible role for ACE during the physiological and pathological angiogenesis of the myometrium. The ACE polymorphism was investigated in women with endometrial cancer, which is one of the most common gynecological malignancies in women of 50 to 70 years of age (Freitas Silva et al., 2004). Another study was conducted in Taiwanese women with endometriosis and ULM (Hsieh et al., 2007). In contradiction to most studies, both studies discovered that the presence of the I allele had a higher risk for developing endometrial cancer, endometriosis and leiomyoma (Freitas Silva et al., 2004;Hsieh et al., 2007). Based on the evidence described above, here in we investigated the role of ACE1 I/D polymorphism on the most common pelvic benign tumor encountered in women worldwide. Contrary to must studies resulting with an unfavorable D allele; our study suggest that the ACE polymorphism has no effect on the formation of uterine fibroids, in respect to their size (≤5cm<) and their number (one or multiple) in Turkish women. Additionally, the polymorphism had no effect on the risk of developing ULM in the Turkish cohort, and cannot be regarded as an early diagnostic parameter for ULM predisposition. This could be due to the benign nature of the ULM, however, the I allele was found detrimental in endometrial cancer in the Spanish population (Freitas Silva et al., 2004). Furthermore, most studies investigated the polymorphism with aggressive cancer types, such as gastric (Wei et al., 2014), breast (Fishchuk andGorovenko, 2013), oral (Vairaktaris et al., 2009), and colorectal cancers (Röcken et al., 2007;Liu et al., 2011), as well as basal cell carcinoma (Yapijakis et al., 2013). This divergence could be due to the vascularization difference between benign ULM and other cancer types. Although our results indicate that the ACE polymorphism is not related to ULM in Turkish women, racial effect might alter the result; hence, the discrepancy between our results and the Taiwanese study could be related to ethnical variations.