International Journal of

Introduction
Polycystic ovary syndrome (PCOS) is the most common women ̍s disorder that according to the latest reports of the PCOS Consensus Workshop Group, is affecting 6-10% of the population and according to the National Institutes of Health (NIH) criteria, affects 15% of reproductive aged women (1). It also causes 91% of anovulatory infertility in women (2). It is characterized by hyper-androgenism, oligomenorrhea or amenorrhea, anovulatory cycle and presence of multiple cysts of ovaries on the ultrasound examination. On the other hand, obesity and high levels of luteinizing hormone (LH) and insulin resistance are other symptoms of this disease (3-5). Restoring ovulation is the aim of PCOS treatment. In this regard, drugs used in PCOS treatment is aromatase inhibitors (AIs) like letrozol, which is co-administered with gonodotrapins to improve ovulation and clomiphene citrate (CC), which is the first treatment to stimulate ovulation of PCOS patients who are undergoing Intrauterine Insemination (IUI). In the second stage of treatment, recombinant Follicle-stimulating hormone (rFSH) and metformin as insulin sensitizers are prescribed. Different diagnosis of PCOS markers and also different responses to treatment because of individual pharmacogenetic aspects, make PCOS a heterogeneous syndrome (3).
As mentioned above, patients with PCOS have an abnormality in androgen biosynthesis resulting in hyperandrogenemia (3, 6). The first step in androgen biosynthesis is transporting cholesterol through the StAR protein from outer to the inner mitochondrial membrane (7, 8). Gonads, placenta, adrenal glands and central nervous system are places that steroid hormones are synthesized. Regardless of the places that steroid hormones synthetize, a common precursor of all steroid hormones is cholesterol. The delivery of cellular cholesterol from the outer to the inner mitochondrial membrane, where the cytochrome p450 side chain cleavage enzyme exists, is the first and rate limiting step of steroidogenesis (9). Several proteins are candidates for the delivery of cholesterol, but StAR protein is the best candidate to do this role (7, 9). Pol et al discovered a mitochondrial protein classified as phosphoproteins, which had 30KDa weight (10), and it was purified, sequenced and expressed in 1994 as a new protein named StAR (9,11).
StAR protein has 284 amino acids (12) and it has been cloned and highly conserved in many species such as mammals, birds, amphibians, and fish (8). Human adrenal cortex, ovary, testis and kidney were tissues that StAR mRNA has been found in them and steroidogenic factor 1, GATA-4, insulin-like growth factor (IGF) -I and GATA enhancer binding protein have been known as regulators of StAR protein (7). StAR gene is located on the short arm of chromosome 8 in the region 8p11.2. Its pseudo gene is located on chromosome 13 (13) with the length of 8 kbp, containing 7 exons and 6 introns (14). The N- terminal with 62 amino acids as a mitochondrial targeting sequence and a site that transfers cholesterol across the mitochondrial membrane as the StART domain consisting of 210 amino acids are mainly two functional domains of StAR protein (11, 12).
Since StAR has a critical role in binding to cholesterol, its facility to transfer it stimulates and initiates the process of steroidogenesis (6, 12). On the other hand, findings indicated that PCOS patients have high levels of steroid hormones so StAR may result in the abnormality of steroidogenesis found in the PCOS patients. Investigations indicated that variations located between exons 5 and 7 or affected the StAR- related lipid transfer (StART) domain were the most mutations that are seen in StAR gene (11).
Therefore in this study, the most significant single nucleotide polymorphisms (SNPs) of this gene (Table I), were investigated in PCOS patients who were undergone IUI treatment.
 
Materials and methods
 
In this case control study, 125 Iranian women from 2008-2012 referred to Royan Institiute, Tehran, Iran, in three groups were included group 1) women with PCOS (n= 45), group 2) healthy women (n= 40) who had at least one child with normal fertilization that have been chosen from healthy volunteered females, and group 3) male factor/unexplained infertile women (n= 40) as a drug response control group in order to evaluate the association of chosen SNPs with drug response in comparison with PCOS patients.
Women in groups 1 and 3 have been undergone intrauterine insemination (IUI) during 2008-2012 at Royan Infertility Institute, and were administered clomiphene citrate/letrozole, 100 mg/day from the days three to seven of their menstrual cycle according to the ovulation induction protocol and were prescribed gonadotropin according to their ovarian response.
PCOS women with a history of ovary cutter, using other medication in past 3 months, and older than 35 years were excluded from the study. Our study was approved by the Ethics Committee of Royan Institute, Tehran, Iran, and all cases signed an informed consent.
 
Clinical data
Clinical data, such as body mass index (BMI), age, and follicles’ number and diameter in 10^th day of menstruation and therapy cycles were collected from the patient’s records.
The comparison was done between two groups (groups 1 and 3) and group 2 which was chosen just for detecting. The SNPs was omitted in this comparison. To study the similarities between these two groups (1 and 3), affecting and potentially confounding factors was investigated. It is worth mentioning that these variables were not confounded (Table II (.
 
SNPs and polymorphism genotyping analysis
Selected polymorphisms were chosen from the NCBI site (http://www.ncbi.nlm.nih.gov/clinicalsnp/) (14) and based on gene alternation of StAR deficiency disorders (Table I).
Genomic DNA samples were collected from the Royan DNA Bank, which had been isolated from peripheral blood by salting out method. Primer pairs were designed for each of the three studied exons (Table III).
Primer 3 software (http://primer3.ut.ee/) was used to design all primers, that were set in specific temperatures. Genotyping of 7 chosen polymorphisms of StAR gene were studied by Polymerase Chain Reaction-Restriction Fragment Length Polymorphism (PCR-RFLP).
Polymerase chain reaction (PCR) was used to amplify genomic DNA in a 50 μl reaction volume containing 30 ng genomic DNA, 1-1.5Mm MgCl2, 0.2 Mm dNTP, 0.4 pMol of each primer and 0.6 unit/μl Tag DNA polymerase enzyme (Cinnagene) in the 1X PCR buffer. For exon 6, 25μl Ampliqon Tag DNA polymerase, 2X master mix with 1.5 Mm MgCl2 were used. The PCR was performed using Eppendorf gradient thermal cycler as follows: an initial denaturation at 94^°C for 4 min, followed by 35 cycles of denaturation at 94^°C for 30 seconds, annealing at a defined temperature (Table III) for 30 seconds, and extension stage at 72^°C for 60 seconds with a final extension at 72^°C for 10 min. All PCR products were visualized under UV light using ethidium bromide staining by 1.2-2% agarose gel electrophoresis. Related restriction enzyme (RFLP) was done on PCR products to identify SNPs. Seven enzymes were used. NmuCI, SphI and HinP1I for exon 5 (figures1-3, respectively), AIuI and AIwNI for exon 7 (figure 6 and 7, respectively) all were purchased from Fermentase fastdigest. 8 μl for exon 5 and 10 μl for exon 7 PCR products were incubated with one unit of each determined enzyme in 37^°C at least for 5 minutes separately. For exon 6 EciI and BtgI (figure 4 and 5 respectively) enzymes were provided from New England Biolab (England). 1 unit EciI and 3 units BtgI enzyme were needed to digest 12 μl PCR products in 37^°C during 1 hour. The fragments were separated by 2% agarose gel electrophoresis and visualized by ethidium bromide staining. The size of fragments achieved has been shown in table IV.
 
Statistical analysis
Mean ± SEM (SD) and frequency (proportion) were used to describe cases and controls quantitative and qualitative clinical characteristics, respectively. For quantitative treatment response (diameter and number of follicles) independent sample t test was used to determine the significant difference between group 1 and 3. Group 2 was excluded from this comparison of the diameter and number of follicules. For investigating the potential confounding variables in the comparison of group1 and 3, all baseline characteristics were compared between these two groups  using t-test or Chi-square test where appropriated. All statistical analysis was performed using Statistical Package for the Social Sciences, version 16.0, SPSS Inc., Chicago, Illinois, USA (SPSS software) and p-value ≤ 0.05 were considered statistically significant.
For treatment response (follicular diameter) Chi-square and independent sample t-test were used.
 
Results
 
Clinical data
The demographic, clinical and endocrinological characteristics of participants are showed in table II. There were no significant differences in mean age and BMI between two groups (Table II).
 
Pharmacogenetic characteristics
The results of consuming clomiphene citrate in group 1 and group 3 (two treated groups) showed that 64% of group 1 and 35% of group 3 had resistance to this drug, which showed a statistically significant difference between them in usage of Clomiphene Citrate and Letrozol (p= 0.01). On the other hand, data have suggested that no significant difference was seen between the effect of Letrozol and Clomiphene Citrate on treatment response, according to the two parameters; number (p=0.86, p=0.76, respectively) and diameter of follicles (p=0.73, p= 0.30) on the 10^th day of menstruation.
 
Genotype analysis
The RFLP results for the SNPs located in exon 5 and 7 indicated that these polymorphisms were in normal status in three groups, which means no heterozygous or homozygous forms of SNPs were observed. For the polymorphism Arg 217 Thr (located in exon 6), no heterozygous form was seen. The digested fragments of PCR products using Btg enzyme showed a normal length on the electrophoresis gel for this SNP (Table IV) and there was no association between PCOS treatment and this polymorphism (p=0.12). EciI enzyme was used for genotyping Ala 218 Val polymorphism (rs137852689) located in exon 6.
A number of the samples in all three groups indicated the heterozygote form of this SNP (seven in group 1, one in group 2, and five in group 3).
It must be mentioned that no homozygocity of this SNP was seen. Although the genotype distributions of SNP rs137852689 (CTT) versing wild type (CCT) in PCOS patients (7% vs. 93.4%) were different from normal ovulatory women (2.5% vs. 97.5%) and fertile control group (12.5 % vs. 87.5%), no significant association was seen (p=0.12) (Figure 8).
Allele frequencies were observed as follows: the normal allele (C) was 92% in 45 PCOS patients, 93% in 40 fertile control subjects and 98% in 40 normal ovulatory women, and the SNP allele (T) was 8%, 7%, and 2%, respectively.