1. IntroductionMetabolic syndrome is a complex of interrelated risk factors forcardiovascular disease (CVD) and diabetes. These factorsinclude dysglycemia, raised blood pressure, elevated triglyceride(TG) levels, decreased high-density lipoprotein cholesterol(HDL-C) levels, and obesity (particularly central adiposity). Theassociations and clustering of these factors have been knownfor decades. Identification of these risk factors has clearlydemonstrated that the syndrome is common and that it has arising prevalence worldwide that relates largely to increasingobesity and sedentary lifestyles. As a result, metabolic syndromeis now both a public health and clinical issue 1.There is widespread agreement that atherosclerotic cardiovasculardisease (ASCVD) is the major clinical outcome ofmetabolic syndrome; all the metabolic risk factors predisposepatients to the development of ASCVD 1,2. Metabolicsyndrome is also associated with an elevated risk for type 2diabetes 3. When diabetes develops, it serves as a powerfulrisk factor for the development of ASCVD in and of itself 2.On the other hand, diet plays an important role in the developmentof metabolic syndrome; and a low-fat 4 or fruit anddairy diet 5 is associated significantly with decreased riskfor metabolic syndrome.Estrogen exerts many protective effects on the cardiovascularsystem, mainly through estrogen receptor alpha (ER-?),including direct effects on vascular tissues and systemiceffects. Estrogen can cause vasodilatation through ERs in botha rapid, non-genomic and a genomic manner 6.ER belongs to the nuclear hormone receptor superfamilyand acts as a ligand-activated transcription factor 7. Thehuman ESR1 gene is located on the long arm of chromosome 6(6q25.1) and contains 8 exons separated by 7 intronic regions8. Several polymorphisms of the ESR1 gene, both singlenucleotide polymorphisms (SNPs) and tandem repeats, havebeen identified. In candidate gene association studies, themost extensively investigated SNPs of the ESR1 gene havebeen, XbaI c.454-351 A/G: XbaI restriction site (rs9340799) andPvuII c.454-397 T/C: PvuII restriction site (rs2234693), whichare both in the first intron 9.The XbaI and PvuII SNPs of the ESR1 gene were found to beassociated with various diseases and metabolic factors,including cardiovascular disorders 10,11, venous thromboembolism12,13, abdominal obesity 14, blood pressure 15,lipoprotein metabolism and serum lipid levels 9. The aim ofthe current study was to investigate the association of ESR1polymorphisms (XbaI and PvuII) with metabolic syndrome andits related phenotypes in females.A case control study was conducted on 300 female subjects.They were divided into 150 healthy subjects and 150metabolic syndrome patients. All subjects were Egyptians ofthe same ethnic group. Patients were selected from theOutpatient Clinic of the Ismailia General Hospital. Diagnosisof metabolic syndrome was done according to the ThirdReport of the National Cholesterol Education Program’s AdultTreatment Panel (ATPIII) with some modifications as previouslydescribed 4. Metabolic syndrome is diagnosed by thepresence of any three or more of the following factors: waistcircumference ?88 cm in women; fasting blood glucose(FBG) ? 100 mg/dL or known diabetes; serum TG ? 150 mg/dL;HDL-C < 50 mg/dL in women; and blood pressure ? 130/85 mmHg or treated hypertension.No smokers were included in the study groups. Subjectssuffering from heart failure, myocardial infarction, diabetesmellitus type I, any type of cancer, renal failure or chronic liverdisease were excluded. Pregnant or lactating women andthose receiving hormone replacement therapy or contraceptivemedications were also excluded. All participants hadregular menstruation.The present study was conducted according to theprinciples of the Declaration of Helsinki, and all the participantsprovided written informed consent following a protocolapproved by the Faculty of Pharmacy, Suez CanalUniversity Research Ethics Committee. Body mass index(BMI), waist circumference, and systolic and diastolic bloodpressure were measured for all the subjects.Peripheral blood was drawn after a 12 h fast and a portion wascollected with EDTA and used for DNA extraction. The serumwas separated from the remaining portion of the bloodsamples and used for assessment of the following parameters:Glucose homeostasis traits: FBG was measured by theenzymatic colorimetricmethod (Biodiagnostic, Egypt) and fastingserum insulin was measured by enzyme linked immunesorbent assay (Monobind). The homeostasismodel assessmentof insulin resistance (HOMA-IR) 16 and the quantitativeinsulin sensitivity check index (QUICKI) 17 were calculated.Lipid profile: TG, total cholesterol and HDL-C were measuredby enzymatic colorimetric methods (Biodiagnostic,Egypt). Low-density lipoprotein cholesterol (LDL-C) was calculated18.Genomic DNA was isolated from 300 ?L of whole bloodcollected in EDTA anticoagulated tubes using the Wizardgenomic DNA purification kit (Promega, Madison) according tothe manufacturer's instructions. The XbaI and PvuII polymorphismswithin the ESR1 gene were detected usingpolymerase chain reaction-based restriction fragment lengthpolymorphism (PCR-RFLP). A fragment of 1300 base pairs (bp)was amplified by a forward primer (5?CTGCCACCCTATCTGTATCTTTTCCTATTCTCC-3?) and a reverse primer (5?-TCTTTCTCTGCCACCCTGGCGTCGATTATCTGA-3?) 9. PCRamplifications were performed in a total volume of 50 ?Lcontaining 250 ng genomic DNA, 25 ?L Go Taq® Green MasterMix, 2× (Promega, Madison) and 0.2 ?mol/L of each primer.Thermal cycling conditions were as follows: an initialdenaturation step at 95 °C for 5 min; followed by 35 cycles ofdenaturation at 94 °C for 30 s, annealing at 53 °C for 40 s andextension at 72 °C for 90 s; and ending with a final extensionat 72 °C for 7 min. Thermal cycling was performed in anEppendorf Mastercycler® machine (Eppendorf, Hamburg,Germany). The PCR products, which contained a part of intron1 and exon 2 of the ESR1 gene, were digested with XbaI andPvuII restriction enzymes (Promega, Madison) at 37 °C for 4 h.Digestion with XbaI, produced two fragments, 900 + 400 bp (Aallele) while undigested product produced a single 1300 bpfragment (G allele) and digestion with PvuII produced twofragments 850 + 450 bp (T allele) while undigested productproduced a single 1300 bp fragment (C allele). The cleavageproducts were electrophoresed on a 1% agarose gel.Differences in baseline characteristics between metabolicsyndrome patients and the control group were assessedthrough the student t-test. Genotype and allele frequenciesas well as other qualitative variables were analyzed with thechi-square test. Additionally, the chi-square test was used todetermine whether the genotype distribution was consistentwith Hardy–Weinberg equilibrium expectations. Differencesin component traits of metabolic syndrome based on genotypewere evaluated in the patient group using analysis ofvariance techniques. Analysis was performed using SPSS,version 17.0. All data are presented as means ± standarddeviations (SD) except where otherwise stated. A value ofp < 0.05 was considered statistically significant.3.1. Demographic and clinical variablesThe demographic, clinical and biochemical parameters of themetabolic syndromepatients and their age-matched controls aresummarized in Table 1. In comparison with the control group,patients showed a significant increase in serum triglycerides,total cholesterol, LDL-cholesterol, fasting blood glucose andfasting serum insulin levels. Patients had a significantly higherinsulin resistance than controls indicated by a significantincrease of HOMA-IR and a significant decrease of QUICKI. Onthe other hand, serum HDL-cholesterol levels were significantlydecreased in the metabolic syndrome patients as compared tothe control subjects. Additionally, BMI, waist circumference,systolic and diastolic BP were higher in patients than in controls.3.2. Genotypes and allele frequenciesThe genotype distribution and the allele frequencies amongthe different groups are shown in Table 2. All genotypedistributions were compatible with Hardy–Weinberg equilibrium.The minor G allele of the XbaI polymorphism had asignificantly higher frequency in metabolic syndrome patientsthan in the control subjects (OR = 1.6, 95% CI = 1.174–2.238, p = 0.003). The distribution of the AG, and GG genotypeof XbaI was 55% and 30% among patients and 55.3% and 18%among the controls, respectively. These differences in thegenotype distribution of the XbaI polymorphism were statisticallysignificant. Regarding the PvuII polymorphism, thefrequencies of the T and C alleles were 48% and 52% amongpatients and 42% and 57.6% among controls, respectively.Statistical analysis showed that the PvuII genotypes and allelefrequencies did not significantly differ between the patientand control groups.3.3. Influence of Gene polymorphisms on metabolicsyndrome traitsWe also investigated the intergenotypic variations in measuresof adiposity, glucose homeostasis traits and lipid levelswith respect to the XbaI polymorphism (Table 3). Carriers ofthe G allele in the homozygous or heterozygous form (GG orAG genotypes) were associated with higher BMI, waistcircumference, systolic and diastolic BP, FBG, fasting seruminsulin, HOMA-IR, total cholesterol and LDL-C. On the otherhand, the subjects with the GG or AG genotype had lowerQUICKI compared to the subjects with the AA genotype.Regarding the PvuII polymorphism, carriers of the minor Callele in the homozygous or heterozygous form (CC or TCgenotypes) had higher diastolic BP, high FBG, fasting seruminsulin, LDL-C, HOMA-IR and lower QUICKI compared tosubjects with the TT genotype (Supplementary material).4. DiscussionMetabolic syndrome is a risk factor for coronary heart diseasesas well as diabetes, fatty liver and several cancers. Theprevalence of metabolic syndrome in women appears to beincreasing, particularly in those of childbearing age 19.In the present study, we assessed the association of ESR1polymorphisms (XbaI and PvuII) with metabolic syndromeand its related phenotypes. Our data suggest that the XbaIpolymorphism within ESR1 has significant correlation withmetabolic syndrome. The results of the present study indicatethat the PvuII polymorphism is not associated with risk ofmetabolic syndrome. Therefore, our results lend support tothe notion that some genetic variants of ESR1 may be candidatesfor metabolic syndrome. Although both the XbaI andPvuII polymorphisms are intronic, they may affect ESR1 expression.The G allele of XbaI was reported to be associatedwith decreased transcription of the ESR1 gene. On the otherhand, the PvuII gene polymorphism may affect the splicing ofthe ER mRNA, resulting in an alteration of protein expression20. It is also possible that these polymorphisms are linked toother polymorphisms in the ESR1 gene that are relevant forprotein expression 7.The constellation of metabolic abnormalities includesglucose intolerance, insulin resistance, central obesity, dyslipidemia,and hypertension, which are all well documentedrisk factors for CVD 21,22. Abdominal obesity and the hypertriglyceridemicwaist phenotype are considered major componentsof CVD 23. The XbaI SNP in the ESR1 gene wasassociated with measures of adiposity in our population,specifically BMI and waist circumference. This result was inagreement with a previous report 14. Estrogen maintains fatdistribution through up-regulation of a number of antilipolytic?2A-adrenergic receptors in the subcutaneous fat depot.Moreover, Toth et al. found that low estrogen levels are associatedwith increased abdominal and subcutaneous fat 24.Both endogenous and exogenous estrogens can exerteffects on lipoprotein metabolism 6. Estrogen is known toinduce mRNA for the LDL receptor in the liver, thereby increasingthe number of LDL receptors and reducing serumLDL-c levels. Our results showed that carriers of the minoralleles of XbaI and PvuII gene polymorphisms, either in homozygousor heterozygous form, were associated with high totalcholesterol and LDL-c levels. A previous report supports ourfindings 25. Moreover, in the present study, the minor allelesof the XbaI polymorphism of ESR1 exhibited a significant associationwith both systolic and diastolic blood pressure. Theseresults can be explained by the short and long term effects ofestrogen on the blood vessels. Estrogen itself increases endothelialnitric oxide synthase activity, which causes rapidvasodilatation. In contrast, a lack of functional ER lowers thevascular expression of inducible nitric oxide synthase. Additionally,estrogen has long-term vasodilation effects that aremediated through activation of transcription factors andthrough ER-? (encoded by the ESR1 gene) which is found inthe vascular endothelium and smooth muscle cells 26.Association between ESR1 polymorphisms and the developmentof Type 2 diabetes mellitus (T2DM) had been previouslyreported. A study of a Caucasian sample showed thatthe frequency of the GG genotype of the XbaI polymorphismwas significantly higher among diabetic patients 27. Anotherstudy on African–American families reported that the PvuII Callele was found to be associated with a reduced insulinsensitivity index (Si) 28. In agreement with these reports, thepresent study showed that homozygous carriers of the minorallele of XbaI and PvuII polymorphisms, GG and CC genotypesrespectively, had significantly higher FBG levels, fastingserum insulin levels, increased HOMA-IR values and decreasedQUICKI values compared to major allele carriers.ESR1 polymorphisms influence many molecular mechanismsof T2DM 20. Estrogen has been implicated in thesuppression of hepatic glucose production and the protectionof ?-cell function and insulin secretion under conditions ofoxidative stress 29. In addition, ER-? can co-modulate withinsulin receptor substrate 1 (IRS1) to affect insulin signalingand action 30. A previous report confirmed the associationbetween the C allele of the PvuII polymorphism and theabsence of expression of the ESR1 gene 31.In conclusion, we used a case–control study to demonstratethe association between the XbaI and PvuII ESR1 geneticvariants and metabolic syndrome and its related phenotypes.Only the XbaI gene polymorphism showed a positive correlationwith metabolic syndrome. However, both minor alleles ofXbaI and PvuII were positively associated with total cholesterol,LDL-C, insulin resistance, and diastolic blood pressure. Only theminor allele of XbaI showed a positive association withmeasures of adiposity in our population, specifically BMI andwaist circumference. A strength of our study was the homogeneityof our population sample, which reduced bias; patientswere ethnically homogenous, well phenotyped, and from thesame geographic location (Suez Canal Area). Additionally,controls were well-matched to patients by age, sex and ethnicbackground in order to reduce the risk of population stratificationthat can produce errors in the results of case controlstudies. However, we acknowledge that the findings presentedhere are preliminary because of our relatively smallsample size.Moreover, the study cohort was limited to middleagedfemales of Egyptian descent, which may prevent thegeneralization of our results to elderly females and femalesof other ethnic backgrounds. Further studies on a large cohortare needed to extend our findings and to determine the molecularmechanisms by which ESR1 gene polymorphisms influencecomponent traits of metabolic syndrome.

Post Author: admin


I'm Irvin!

Would you like to get a custom essay? How about receiving a customized one?

Check it out