SREBF1c and SREBF2 gene polymorphisms are associated with acute coronary syndrome and blood lipid levels in Mexican population

Aim It has recently been reported that the sterol regulatory element-binding transcription factors (SREBF-1c, and -2) contribute to the variation in the plasma lipids levels, which have an important role in the atherosclerotic plaque development. The aim of the present study was to evaluate whether the SREBF1c and SREBF2 gene single nucleotide polymorphisms (SNPs) are associated with plasma lipids levels and ACS susceptibility in a case-control association study. Material and methods A case-control study was carried out in 625 patients with ACS (82% men and 18% women, with a mean age of 57.97 ± 10.5 years) and 700 healthy controls (66% men and 34% women, with a mean age of 54.37 ± 7.65 years). The sample size was calculated for a statistical power of 80%. We genotyped three SREBF1c (rs2297508, rs11656665 and rs11868035) and three SREBF2 (rs2267439, rs2267443, and rs2228314) gene polymorphisms by 5’ exonuclease TaqMan assays. The associations were evaluated by logistic regression under the co-dominant, dominant, recessive, over-dominant and additive inheritance models. The contribution of the genotypes on the plasma lipids levels was evaluated by Student’s t-test. Results Under different models, the SREBF1c rs2297508 (OR = 1.50, pCRes = 0.03), SREBF1c rs11656665 (OR = 1.35, pCDom = 0.02 and OR = 1.26, pCAdd = 0.02) and SREBF2 rs2228314 (OR = 1.78, pCRes = 0.03, OR = 1.27, pCAdd = 0.04) SNPs were associated with higher risk of ACS. On the other hand, the SREBF1c rs11868035 SNP was associated with lower risk of ACS (OR = 0.49, pCCo-dom = 0.001, OR = 0.66, pCDom = 0.003, OR = 0.57, PRes = 0.003 and OR = 0.71, pCAdd = 0.001). There was a statistically significant association of both SREBF1c rs11656665 and rs11868035 polymorphisms with plasma triglyceride levels. Conclusions In summary, our data suggest the association of the SREBF1c and SREBF2 SNPs with risk of developing ACS and with triglyceride levels in a Mexican population.


Introduction
The acute coronary syndrome (ACS) is an important consequence of both atherosclerosis and abnormalities in lipid levels, constitutes a worldwide public health problem. Generally, the transition of a stable coronary atherosclerotic lesion into a ruptured or eroded plaque produces the clinical manifestations of the acute coronary syndrome [1,2].
On the other hand, SREBF-2 [encoded by the sterol regulatory element binding transcription factor-2 (SREBF-2) gene, located in chromosome 22q13.2] is preferentially involved in the regulation of cholesterol metabolism [14,15]. In this context, we postulated that SREBF-1 and SREBF-2 genes could be new candidates in the development of ACS due to their involvement in the regulation of the lipid biosynthesis; these molecules play an important role in the development of atherosclerotic plaque [1,2,5,6].
susceptibility of developing ACS. Furthermore, we evaluated whether these polymorphisms were associated with plasma lipids levels in a Mexican population sample.

Characteristics of the study population
The sample size calculation for unmatched cases and controls study, with a power of 80% and an alpha error of 0.05, showed that the total sample size needed to carry out this study was 528 Mexican mestizo individuals (264 patients with ACS and 264 control individuals) (http:// www.openepi.com/SampleSize/SSCC.html). In this study, we included 1325 Mexican mestizo individuals (625 patients with ACS and 700 healthy controls unmatched by age or gender). From July 2007 to July 2017, we recruited 625 patients with ACS (82% men and 18% women, with a mean age of 57.97 ± 10.5 years) who were referred to the Instituto Nacional de Cardiologia Ignacio Chavez. From this patient population, 501 were diagnosed with myocardial infarction and 124 with unstable angina. ACS was diagnosed based on clinical characteristics, electrocardiographic changes and biochemical markers of cardiac necrosis (creatinine kinase isoenzymes, creatinine phosphokinase, or troponin I above the normal upper limit). The diagnosis of ACS was made according to guidelines from the European Society of Cardiology (ESC) and American College of Cardiology (ACC) [22,23]. The exclusion criteria were (1) patients with clear inflammatory pathologies on admission, such as infection established by clinical, laboratory or image investigations, and (2) patients with an autoimmune disease or cancer previously diagnosed or documented during their hospitalization. Moreover, we included 700 apparently healthy controls (66% men and 34% women, with a mean age of 54.37 ± 7.65 years) without a family history of premature coronary artery disease (pCAD) or atherosclerosis. The control group was recruited from June 2009 to June 2013 from blood bank donors and with the assistance of brochures posted in social service centers. These control subjects were nested in the cohort of the Genetics of Atherosclerosis Disease (GEA) Mexican study. The exclusion criteria included the use of anti-inflammatory drugs at the time of the study, congestive heart failure, and liver, renal, thyroid or oncological disease. In addition, the control subjects had a coronary calcium score of zero determined by computed tomography, indicating the absence of subclinical atherosclerosis in them [24]. All the included subjects were ethnically matched and considered Mexican mestizos only if they were at least third generation Mexicans born in the country.

Ethics
The study complies with the Declaration of Helsinki. Ethical approval was provided by the Ethics and Research commission of Instituto Nacional de Cardiologia Ignacio Chavez-registration number: 17CI09012010. Written informed consent was obtained from all participants.

Laboratory analyses
After a 12-h overnight fast, EDTA blood samples were drawn and centrifuged within 15 min after collection; plasma was separated into aliquots and immediately analyzed or frozen at −80˚C until analysis. Cholesterol and triglycerides plasma concentrations were determined by enzymatic/colorimetric assays (Randox Laboratories, UK). The phosphotungstic acid-Mg 2+ method was used to determine HDL-C concentrations. LDL-C was estimated in samples with a triglyceride level lower than 400 mg/dl, using the modified Friedewald formula [25]. Plasma lipid concentrations were determined within 24-h after blood sample collection. We followed the National Cholesterol Education Project (NCEP) Adult Treatment Panel (ATP III) guidelines and thus defined dyslipidemia with the following levels: cholesterol > 200 mg/dl, LDL-C > 130 mg/dl, HDL-C < 40 mg/dl, and triglyceride > 150 mg/dl. (http://www.nhlbi. nih.gov/guidelines/cholesterol/atp3_rpt.htm).

Statistical analysis
Continuous variables that are not normally distributed, such as age, body mass index (BMI), blood pressure, glucose, total cholesterol, HDL-C, LDL-C, were analyzed using the Mann-Whitney U test. Categorical variables, such as gender, hypertension, T2DM, and smoking habit were analyzed with the chi-squared test or Fisher's exact test. Using co-dominant, dominant, recessive, over-dominant, and additive models, we analyzed the association of the polymorphisms with ACS by logistic regression. Multiple logistic models were built to identify the variables that best explained the risk of developing ACS. Models were constructed including one variable at the time, while final models included variables with biological relevance, statistical significance or both. Confounding bias was accepted when changes in estimated odds ratios (ORs) were equal to or greater than 10%. When a principal effect model was reached, effect modification was also tested and interaction terms were constructed between the polymorphisms and different variables; the terms were included in the model when the significance of the p-value was greater than or equal to 0.05. All p-values were corrected (pC) by the Bonferroni test. The values of pC < 0.05 were considered statistically significant, and all odds ratios (OR) are presented with 95% confidence intervals. The occurrence of the ACS in our population was based in the OR values: (a) OR = 1 does not affect the odds of developing ACS, (b) OR>1 is associated with higher odds of developing ACS, and (c) OR<1 is associated with lower odds of developing ACS. The linkage disequilibrium analysis (LD, D") of the polymorphisms and the haplotypes construction was performed with Haploview version 4.1 (Broad Institute of Massachusetts Institute of Technology and Harvard University, Cambridge, MA, USA). We used the chi-squared test to evaluate Hardy-Weinberg equilibrium (HWE). Student's t-test was used for establishing the contributions of the genotypes on the plasma lipids levels. Values are expressed as means ± SD, and statistical significance was set at p < 0.05. The analysis of data was performed with SPSS version 18.0 (SPSS, Chicago, Il) statistical package. The statistical power to detect an association with ACS was 0.80. We used the QUANTO software [http://biostats.usc.edu/software] to estimate the association.

Characteristics of the study population
Demographic, clinical and biochemical characteristics of the ACS patients and healthy controls are shown in Table 1. There were significant differences between the ACS patients and healthy controls. Compared to healthy controls, the ACS patients had a higher frequency of T2DM, hypertension, dyslipidemia, and smoking habit. Conversely, the total cholesterol and LDL-C levels of ACS patients were lower than those of the control group; this effect may be due to their treatment with statins.

Allele and genotype frequencies
Genotype frequencies in the polymorphic sites were in HWE. The frequencies of the SREBF2 IVSI C8407T (rs2267439) and SREBF2 IVSI2 A1667G (rs2267443) polymorphisms were similar in ACS patients and healthy controls. Nonetheless, four SNPs [SREBF-1c 3'UTR A30225G (rs11868035), SREBF-1c 3'UTR G30009C (rs2297508), SREBF-1c IVS1 G954A (rs11656665), and SREBF2 G1784C Gly595Ala (rs2228314)] were associated with the presence of ACS ( Table 2). Considering that the whole group of patients and controls were not age and sex matched, this analyses was adjusted by these variables. In order to corroborate the effect of these variables, we also performed a sub-analysis including a group of patients and controls age and sex matched. This sub-analysis showed similar results than those obtained with the whole group; the SREBF-1c 3'UTR A30225G (rs11868035), SREBF-1c 3'UTR G30009C (rs2297508), and SREBF-1c IVS1 G954A (rs11656665) SNPs were associated with the presence of ACS (S1 Table). Under a recessive model, the G allele of the SREBF-1c 3'UTR G30009C (rs2297508) SNP was associated with a higher risk of ACS

Association of polymorphisms with plasma lipids levels
We determined the contributions of these SNPs in plasma lipids levels, such as total cholesterol, HDL-C, LDL-C, triglycerides, and LDL-C / HDL-C ratio. For this analysis, we only selected the group of healthy individuals who fulfilled the above-mentioned inclusion criteria for control subjects. We did not include the analysis of lipids plasma levels in patients with ACS, because in the setting of the coronary syndrome, these levels may be altered by the use of the anti-dyslipidemic or anti-inflammatory drugs. Comorbidities, such as insulin resistance/ T2DM, hypertension, and inflammatory processes can also mask the real impact of SREBF1c and SREBF2 polymorphisms on plasma lipids.

Functional prediction
The functional prediction analysis showed that the presence of the G allele of the SREBF1c 3'UTR G30009C (rs2297508) polymorphism potentially produces a binding motif for Srp55 protein. The analysis also revealed that the G allele of the SREBF2 G1784C Gly595Ala (rs2228314) polymorphism may generate binding motifs for SF2/ASF protein. In contrast, the analysis of the SREBF1c 3'UTR G30009C (rs2297508), SREBF1c IVS1 G954A (rs11656665), SREBF2 IVSI C8407T (rs2267439), and SREBF2 IVSI2 A1667G (rs2267443) polymorphisms did not exhibit evidence of potentially functional motifs.

Discussion
In our study, a similar distribution of IVSI C8407T (rs2267439) and IVSI2 A1667G (rs2267443) SNPs of the SREBF2 gene were observed in ACS patients and healthy controls. Nonetheless, we also found that the presence of the G allele of SREBF2 G1784C Gly595Ala (rs2228314) SNP was associated with a risk of developing ACS. Contrary to our findings, Robinet et al. reported that the C allele of the G1784C (rs2228314) SNP is associated with development of intima-media thickness and early-stage carotid atherosclerosis in a French population [17]. In same way, Miserez et al. reported that the C allele of G1784C (rs2228314) SNP contributes to the development of hypercholesterolemia and increased plasma cholesterol levels in Swiss and Israeli populations [7]. In contrast to these data, studies in Asian populations report a lack of association of the G1784C Gly595Ala (rs2228314) SNP with premature coronary artery disease and coronary heart disease [10,29]. Additionally, the haplotype analysis showed that the "TGG" haplotype, composed of IVSI C8407T (rs2267439), G1784C Gly595Ala (rs2228314), and IVSI2 A1667G (rs2267443) SNPs, was associated with the risk of developing ACS. In contrast, the "TCG" haplotype showed a lower frequency in ACS patients. As can be seen, the "TGG" haplotype is differentiated by the presence of the G allele of the G1784C Gly595Ala (rs2228314) polymorphism; the G allele marks the risk haplotype, while the C allele characterizes the protective haplotype. Of note, in the SNP independent analysis, we found an association between the G allele and the presence of ACS; this allele has thus an important role in the development of the disease. On the other hand, we found an association of the A allele of the SREBF-1c gene IVS1 G954A (rs11656665) SNP with a higher risk of developing ACS in the study population. A meta-analysis showed that the A allele of SREBF-1c IVS1 G954A (rs11656665) SNP is associated with lower adiponectin levels [18]. This meta-analysis is in accord with our findings since hypoadiponectinemia may contribute to a higher cardiovascular risk. In addition, our results demonstrated that the 3'UTR A30225G (rs11868035) and 3'UTR G30009C (rs2297508) SNPs of the SREBF-1c gene play an important role in the presence of ACS. To the best of our knowledge, this study is the first to describe the association between these polymorphisms with the presence of ACS. In this context, the association of these SNPs with several inflammatory diseases in different populations is controversial. In agreement with our findings, Felder et al. reported that G allele of the SREBF-1c 3'UTR G30009C SNP is associated with a risk of developing T2DM in a French population [9]. Similarly, in a Danish population, Grarup et al. reported in a meta-analysis study that the G allele is associated with a higher risk of T2DM [30]. However, Peng et al. studied a Chinese Han population and reported that the 3'UTR A30225G and 3'UTR G30009C polymorphisms were not associated with risk of non-alcoholic fatty liver disease (NAFLD) [19]. With regard to the 3'UTR A30225G (rs11868035) polymorphism, we found that the G allele was associated with lower risk of developing ACS. In contrast, Liu et al. reported that G allele of the 3'UTR A30225G (rs11868035) SNP in a Chinese population increased the risk of developing T2DM (OR = 1.76) [31]. In line with our findings, studies in Caucasian populations have shown that the A allele of the SREBF-1c 3'UTR A30225G (rs11868035) SNP is associated with a higher risk of developing T2DM [30,32,33]. In addition, we found that the AGG and AGA haplotypes were associated with increased risk of developing ACS. However, as far as we know, this study is first that describes haplotypes between these polymorphisms.
Moreover, in our study, the association of the SREBF2 G1784C (rs2228314) and SREBF-1c 3'UTR A30225G (rs11868035) polymorphisms with the presence ACS were positive, but controversial with other populations. We think that the association of the SREBF1c and SREBF2 polymorphisms could be due to the classical cardiovascular risk factors, and the environmental factors such as diet, exercise and lifestyle that have an important role in the development of the inflammatory diseases [19,33], as well as, to the fact that the allelic distribution of these polymorphisms varies according to the ethnic origin of the study populations. In this context, data obtained from the National Center for Biotechnology Information, showed that Caucasian population presents a higher frequency of the G allele of the 3'UTR A30225G (rs11868035) SNP (72%) when compared to Mexican Mestizos and Asian population that present a lower frequency of the G allele (41% and 16%, respectively). Concerning the G1784C (rs2228314) SNP the Mexican Mestizos presents a lower frequency of the G allele (29%) when compared to Asian and Caucasian populations (80% and 77%, respectively) (https://www.ncbi.nlm.nih.gov/ variation/tools/1000genomes/). Owing to the specific genetic characteristics of the Mexican population, we consider that additional studies are needed in a larger number of individuals and in other populations with different ethnicities. This research could help define the true role of SREBF1c and SREBF2 polymorphisms as risk or protective markers in the development of ACS and others cardiovascular events.
We further determined the effect of the SREBF2 and SREBF-1c gene polymorphisms on plasma lipid levels using genotype groups. In this context, the analysis of the G1784C Gly595Ala (rs2228314) SNP did not show changes in the plasma lipid levels. The G and T alleles of the IVSI2 A1667G (rs2267443) and IVS1 C8407T (rs2267439) SNPs were not associated with risk of developing ACS. Nonetheless, we found an association of these two SNPs with not only higher cholesterol and glucose plasma concentrations, but also higher systolic blood pressure (p < 0.05). In contrast to our data, Galavi et al. reported that the IVSI2 A1667G (rs2267443) and IVS1 C8407T (rs2267439) SNPs were not associated with plasma lipids levels in T2DM patients [21]. On the other hand, Grarup et al. reported that the G allele of the 3'UTR G30009C (rs2297508) SNP was associated with higher cholesterol and glucose plasma levels. In addition, the authors showed that the A allele of 3'UTR A30225G (rs11868035) was associated with higher cholesterol plasma levels [30]. By the same token, Musso et al. reported that the A allele of SREBF-1c 3'UTR A30225G (rs11868035) was associated with higher triglyceride and LDL-C plasma levels [33]. However, in this study population, we found that the SREBF-1c 3'UTR G30009C (rs2297508) SNP was not related with a change of plasma lipid levels. In addition, we found that the G allele of the SREBF-1c 3'UTR A30225G (rs11868035) was associated with lower triglyceride levels. In the same way, Peng et al. reported that the G allele of the SREBF-1c 3'UTR A30225G (rs11868035) was associated with lower triglyceride levels in non-alcoholic fatty liver disease [19]. Additionally, using bioinformatics tools we determined the potential effect of the SREBF2 and SREBF-1c gene polymorphisms associated with ACS. The analysis of the SREBF1c 3'UTR G30009C (rs2297508), SREBF1c IVS1 G954A (rs11656665), SREBF2 IVSI C8407T (rs2267439), and SREBF2 IVSI2 A1667G (rs2267443) polymorphisms did not exhibit evidence of potential functional motifs. Nonetheless, the analysis of the SREBF2 G1784C Gly595Ala (rs2228314) SNP showed that the G allele generates a binding site for the SRp55 protein. In the same way, the analysis of the SREBF1c 3'UTR G30009C (rs2297508) polymorphism showed that the G allele generates a binding site for the SF2/ASF proteins. These proteins have multiple functions in the pre-mRNA splicing process, as well as in the regulation of alternative splicing [34,35]. In our opinion, the specific correlation of these polymorphisms with plasma lipid levels deserves to be addressed in future studies.

Limitations
Unfortunately, in our study was not possible to match patients and controls by age and gender, considering this, the analysis was adjusted by these variables. On the other hand, a sub-analysis in a groups age and gender matched was done. In this case, the results were similar to those reported in the whole group of studied individual. Moreover, our results in Mexicans should be reproduced in other ethnic groups to have a better definition about the contribution of the SREBF1c and SREBF2 gene polymorphisms on the lipid profile as well as their role as possible markers of risk or protection against the development of ACS.
In summary, this study demonstrated that, in the Mexican population, three polymorphisms were associated with a high risk of developing ACS: the SREBF-1c 3'UTR G30009C (rs2297508) and SREBF-1c IVS1 G954A (rs11656665) polymorphisms of the SREBF1c gene, and SREBF2 G1784C Gly595Ala (rs2228314) of the SREBF2 gene. On the other hand, the SREBF-1c 3'UTR A30225G (rs11868035) polymorphism of the SREBF1c gene was associated with a lower risk of developing ACS. There was also a statistically significant association of both SREBF-1c IVS1 G954A (rs11656665) and SREBF-1c 3'UTR A30225G (rs11868035) polymorphisms with lower triglyceride levels.