The present retrospective and descriptive study included patients who attended the cardiology department at the Hospital Regional de Alta Especialidad-Ciudad Salud (HRAE-CS; Chiapas, Mexico) between March 2010 and December 2012. It included a cohort of 252 patients, 159 male (63.1%) and 93 female (36.9%), with an age range of 22-81 years and a median age of 53.5. In total, 114 patients were diagnosed with primary ACS according to the American College of Cardiology (ACA) criteria (24) and were included in the ACS group under the following criteria: i) Persistent chest pain for >20 min; ii) typical electrocardiographic changes including new pathologic Q waves; iii) ST-segment elevation of >1 mm; and iv) increased plasma levels of CK-MB >2-fold higher than the upper reference limit and/or troponin-T >0.1 µg/ml.

In parallel, 138 healthy individuals were included in the control group according to the ACA criteria (24): i) Individuals without a history of cardiovascular disease; and ii) individuals without evidence of other systemic diseases.

Based on their medical history, no subjects in the control group suffered from ACS or any related pathology. The age and sex of the control and the ACS groups were matched. The selection of patients is shown in the STROBE flow (Fig. 1). The current study was conducted in accordance with The Declaration of Helsinki. All participants provided written informed consent before taking their blood samples. The study was approved by the HRAE-CS Ethics Committee (approval no. 02/2010).

For each patient, 5 ml of blood was drawn into a purple top EDTA tube and stored at 4˚C for a maximum of 1 week to prevent degradation of nucleic acids. DNA was extracted from EDTA-treated venous blood samples using the QIAamp DNA Blood Mini kit (Qiagen GmbH) and stored at -20˚C until use. DNA was quantified in an Eppendorf 6131 Biophotometer (Eppendorf SE) and sample concentrations were adjusted to 50 ng/µl. Genotyping was performed in a StepOne Real-Time PCR System with the TaqMan Genotyping Master Mix (Applied Biosystems; Thermo Fisher Scientific, Inc.). Specific assays for the SNPs 3872C>T [rs1205; genotyping primer: ACTTCCAGTTTGGCTTCTGTCCTCA(C/T)AGTCTCTCTCCATGTGGCAAACAAG], 5237G>A [rs2808630; genotyping primer: AGGCCAGAGGCTGTCTACCAGACTA(G/A)GTATAGTAAGATGCAAGCAACTGAA], 2667C>G [rs1800947; genotyping primer: AGATGGTGTTAATCTCATCTGGTGA(C/G)AGCACAAAGTCCCACATGTTCACAT] and 3006A>C [rs3093066; genotyping primer: ATGTCACTGGCCTCATGCTTTGCAC(A/C)TTACAAAGTGAGTAATGTGTGCTGA] previously associated with cardiovascular disease (CVD) in the CRP gene (25) were purchased from Thermo Fisher Scientific, Inc. with ID C_7479334_10, C_12035003_10, C_11663840_10 and C_27452296_10, respectively. PCR protocol included initial denaturation 1 cycle at 95˚C for 10 min, followed by 40 cycles of denaturation at 95˚C for 15 seg and annealing/extension at 60˚C for 1 min. The genotyping was performed following the manufacturer's protocol. Genotype calls were made upon visualization of the allelic discrimination charts in which the clusters were identified.

Data are presented as mean ± standard deviation (age) or percentages. For inferential statistics, the normal distribution of the data was first determined using the Kolmogorov-Smirnov test and the equality of variances was confirmed using the Levene test. The age of the groups was compared using the Student's unpaired t-test. Fisher's exact test was used to compare categorical values. The association between various cardiovascular risk variables (age, male, diabetes, smoking, alcoholism, arterial hypertension, dyslipidemia, sedentary lifestyle and obesity), or CRP genotypes, with the presence of ACS was determined using odds ratios (OR) and 95% confidence intervals (95% CI) using univariate binary logistic regression analyses. Subsequently, significant predictors were added to the multivariate model and with forward stepwise logistic regression the most parsimonious model was identified. The probability used for the stepwise regression was set at 0.15 for variable input and 0.25 for removal. Analysis involving a two-way interaction (A x B, where A and B are the genotypes 2667CC/CG and 3872CC/CT) was made in a separate multivariate model that did not include genotypes of other polymorphisms. The multivariate analysis was considered pertinent because various metabolic and vascular disorders have been previously reported to be associated with both the exposure variable of interest (CRP genotypes) and the outcome variable (ACS) (26-29). Multivariate logistic regression allowed for the necessary statistical adjustments for probable confounding variables, resulting in an adjusted OR (AdOR). Data were analyzed using SPSS V.21 (IBM Corp.) apart from linkage disequilibrium (LD) and haplotype analyses, which were performed using SNPStats software (30). P<0.05 was considered to indicate a statistically significant difference.

Demographic and clinical parameters of the patients with ACS and control group are presented in Table I. In total, 252 participants were included in the present study. Male individuals were the majority in both groups, with 55.8% in the control and 71.9% in the ACS group. Significant differences were found between the group control and ACS in age, sex, T2DM, smoking, alcohol, HBP, dyslipidemia and sedentary lifestyle (P<0.05). On the other hand, the obesity incidence was not found significantly different (P=0.876). Except for T2DM, the aforementioned variables are considered risk factors for ACS.

Allele and genotype frequencies of the four polymorphisms in the patients with ACS and controls were in Hardy-Weinberg equilibrium (P>0.05; data not shown). The frequency of the genotypes 2667CC/CG, 3872CC/CT, and 3006AA/AC were significantly higher in the group of patients with ACS, compared with the control group (26.1 vs. 57.0%, P<0.001; 66.7 vs. 80.7%, P=0.015; and 8.0 vs. 16.7%, P=0.049; respectively). The genotype 5237GG/GA did not show a statistical significance between groups (P=0.467; Table II). Furthermore, the analysis involved a two-way interaction between the 2667^*3872 polymorphisms, showing that the combination of genotypes 2667CC/CG^*3872CC/CT was significantly higher in the group with ASC (43.0%) compared with the control group (18.1%) (P<0.001; Table II).

The univariate logistic regression analysis showed that age, maleness, T2DM, smoking, alcohol, HBP, dyslipidemia and sedentary lifestyle were factors associated with ACS (Table III). The multivariate analysis reveals that only age, T2DM, alcohol, HBP and sedentary lifestyle had an association with ACS (Table III). The univariate model reveals that genotypes 2667CC/CG, 3872CC/CT and 3006AA/AC were ACS-associated (OR=3.75, 95% CI: 2.21-6.39; OR=2.09, 95% CI: 1.16-3.75; and OR=2.3, 95% CI: 1.04-5.08, respectively; Table III). The multivariate model showed that genotypes 2667CC/CG (AdOR=4.82, 95% CI: 1.69-13.72) and 3872CCGG/CTGA (AdOR=3.78, 95% CI: 1.1-12.92) remain associated to ACS independently of other clinical variables or genotypes analyzed; while the 3006AA/AC genotype lost relevance and was not included in the multivariate model. The interaction of the 2667CC/CG and 3872CC/CT genotypes increased the risk for ACS in the univariate and multivariate model (OR=3.40, 95% CI: 1.92-6.02; and AdOR=4.14, 95% CI: 1.49-11.52, respectively). However, the probability of developing ACS generated by this combination of genotypes was not higher than that generated by the 2667CC/CG genotype alone (Table III).

The most frequent haplotype in the population was H1 T/A/G/C, with 29.15% of occurrence, followed by H2 C/A/G/C with 28.46%, H3 C/A/C/C with 11.92% and H4 T/A/C/C with 10.68%; together, they represent 80.21% of the total haplotypes. Due to the frequency of the remaining haplotypes in the population, only four were analyzed. Haplotypes numbers H2, H3, and H4 showed an association with risk of ACS (OR=14.86, 95% CI: 2.62-84.15, P=0.0026; OR=29.98, 95% CI: 4.38-205.34, P=7x10^-4 and OR=100.56, 95% CI: 4.64-2177.88, P=0.0037 respectively) after adjusting for age, sex and other clinical variables (T2DM, smoking, alcohol, hypertension, dyslipidemia, sedentary lifestyle and obesity), as shown in Table IV. The polymorphism 3006 exhibited a strong LD with 3872 and 2667 (with D 0.0143, D'=0.4556, P=0.006; and D=0.0184, D'=0.4402, P<0.001, respectively), as shown in Fig. 2. The genotype 3006AA/AC rs3093066 was associated with ACS in the univariate model. This polymorphism was previously associated with CRP levels in Caucasian American and African American young adults (25). However, this polymorphism does not seem to be an independent risk factor for the genesis of ACS in the population in the present study, since other clinical or genetic factors studied here make it lose relevance in a multivariate analysis. A probable explanation for this fact is that polymorphism 3006 exhibited a strong LD with 3872 and 2667 (with D=0.0143, D'=0.4556, P<0.05; and D=0.0184, D'=0.4402, P<0.05, respectively). This means that the alleles of the 3006 polymorphisms are nonrandomly associated with the alleles at nearby polymorphic sites 3872 and 2667. Therefore, it is likely that the 3006AA/AC genotype is not an independent risk factor, but rather a ‘linked or predicted’ factor by the genotypes 2667CC/CG and 3872CC/CT, an aspect that was detected in the multivariate model.

ACS is a spectrum of myocardial ischemic disorders, including myocardial infarction with ST-segment elevation, myocardial infarction with non-ST segment elevation and unstable angina, with a multifactorial origin. In addition, previous works have linked genetic polymorphisms with the development of ACS (13,14,31). The present study analyzed the association between four polymorphisms in the promoter region of the CRP gene and ACS in a group of individuals from Southern Mexico. In the present study variables such as age, maleness, T2DM, alcohol consumption, smoking, HBP, dyslipidemia and sedentary lifestyle showed significant differences between the groups (Table I), a result which is consistent with previous reports (16,32-34). Hypertension, dyslipidemia and smoking are the three modifiable risk factors most strongly and independently associated with coronary heart disease (CHD) (35). Recently, a study showed that factors such as age 70 years old, femaleness sex, HBP, T2DM and hypertension are more frequent in patients with combined endpoint (cardiovascular mortality, mortality from stroke, myocardial infarction and stroke/transient ischemic attack in 10-year follow-up) in comparison with the control group in German patients (P<0.05) (36). DM is linked with comorbidities such as CAD (37) and represents 25-30% of the patients with ACS. Patients with ACS and DM, present poorer outcomes regarding CV morbidity and mortality, compared with individuals with ACS but not with DM. In addition, T2DM prolongs hospitalization time in patients with ACS and disruption of the normal glucose homeostasis has been found in ~70-75% of patients with CAD (38).

Few studies have been conducted in the Hispanic population to establish an association between gene polymorphisms and ACS. In the present study, the genotype 3872CC/CT (rs1205) was found to be a risk factor for ACS (AdOR=3.78; 95% CI: 1.11-12.92, P=0.034). The only study conducted in Mexico with this SNP did not show any association with ACS in individuals from Western Mexico; however, the allele T is associated with lower CRP concentration (38). 3872CT has been associated with low levels of CRP (33,39-42) myocardial infarction (33) and CHD (12), but other studies have not shown an association with CHD (40) or aortic stenosis (30). The current results reported that the 5237GG/AA genotype did not show an association with ACS (OR=1.04; 95% CI: 1.60-1.79, P=0.884); which agrees with a previous report (12).

The 2667CC/CG genotypes (rs1800947) were found to be a risk factor for ACS in the present study population (AdOR=4.82, 95%CI: 1.69-13.72, P=0.003). These polymorphisms have been associated with CHD (43), coronary artery disease (44), myocardial infarction (45), cerebrovascular ischemia-related disease (46), cardiovascular disease mortality (23) and other diseases such as diabetes (47), prostatic (48) and colorectal cancer (49) and microangiopathic stroke (50). In addition, other investigations have not found an association (33,15,51,52). Allele G from rs1800947 has been associated with low CRP levels in coronary artery disease (44), myocardial infarction (53) and adverse cardiovascular (54).

The genotype 3006AA/AC rs3093066 was associated with ACS in the univariate model. This polymorphism was previously associated with CRP levels in European American and African American young adults (25). However, this polymorphism does not seem to be an independent risk factor for the genesis of ACS in the present population, since other clinical or genetic factors studied here make it lose relevance in a multivariate analysis. A probable explanation for this is that polymorphism 3006 exhibited a strong LD with 3872 and 2667 (with D=0.0143, D'=0.4556, P<0.05; and D=0.0184, D'=0.4402, P<0.05, respectively). This meant that the alleles of the 3006 polymorphism were non-randomly associated with the alleles at nearby polymorphic sites 3872 and 2667. Therefore, it is likely that the 3006AA/AC genotype is not an independent risk factor, but rather a ‘linked or predicted’ factor by the genotypes 2667CC/CG and 3872CC/CT, an aspect that was detected in the multivariate model. The LD between polymorphism 3006 with polymorphisms 3872 and 2667 may reflect the natural selection history, gene conversion and mutation history of the analyzed population (from Chiapas, southeast of Mexico) (55), which although it can be classified as mestizo (individuals of mixed ancestry with a European and an indigenous background), has a strong Mayan ethnological and linguistic affiliations (56). Future analyses will be necessary in this respect. On the other hand, the present results showed that haplotype numbers H2 (C/A/G/C), H3 (C/A/C/C) and H4 (T/A/C/C) showed an association with risk ACS. Similar results have been shown in the US population (12).

Polymorphisms in the CRP promoter region would increase the risk of ACS by favoring an elevation of this protein, which is a clear sign of systemic inflammation (57,58). The expression of the CRP gene is modulated by several SNPs in the promoter region and these SNPs affects changes in gene promoter activity, RNA structure or turnover, transcription factor binding and transcriptional activity (25,59). The mechanisms by which SNP 2667 affects the expression of CRP are unclear, so it remains possible that 2667 is in strong LD with a polymorphism in a distal regulatory element not within the region scanned for polymorphisms (25). However, a study showed that the minor alleles of the 2667 and 3872 polymorphisms linked with CRP haplotypes are associated with decreased promoter activity (23). As aforementioned, several SNPs have been associated with CHD and with CRP levels. CRP is an inflammatory marker associated with cardiovascular disease, T2DM and other pathology. It is probable that CRP genotypes affect CRP synthesis and influence the onset of clinical CVD or ACS (23).

Due to the aforementioned factors, it is hypothesized that the number of subjects included in the present study may be a limitation. However, the results shown can be considered reliable, considering that it has previously been postulated that it is necessary to have ≤10 individuals with the event of interest for each predictor variable included in logistic regression models (60) to improve the performance of the model and reduce the risk of false positive results. In the multivariate logistic regression model performed in the present study (Table III), eight predictor variables were included in the multivariate model, analyzing 114 individuals with ACS and 138 subjects in the control group. Therefore, the present study complied with the rule of having ≤10 events per variable. However, some reports have suggested that this rule is controversial (61,62). Another limitation was that the serum levels of CRP were not determined in the participants, so it would be relevant for future research to quantify this protein to evaluate whether the risk conferred by certain genotypes is directly related to the elevation of CRP.

In conclusion, four polymorphisms in the promoter region of the CRP gene in association with ACS were analyzed in the present study. Variables such as age, T2DM, alcohol, HBP and sedentary lifestyle were associated with ACS. To the best of the authors' knowledge, this work provided the first evidence of the association of CRP genotypes 2667CC/CG and 3872CC/CT with ACS in Mexico. These genotypes were evidenced as risk factors independent of other clinical risk factors in the multivariate analysis. It is possible that a set of polymorphisms, along with environmental factors, contribute to the development of chronic degenerative diseases, modulating the increased risk for heart disease. However, further studies are required to ascertain whether these polymorphisms participate in the development of ACS.