An assessment of gene polymorphisms in young South African Indians with coronary artery disease and the effect of atorvastan in vitro.
The global burden of heart disease increases every year. It has been estimated that by the year 2020, coronary artery disease (CAD) will be the number one cause of death worldwide. Indian populations throughout the world have the highest prevalence of CAD and early onset of the disease compared to other ethnic groups. Glutathione S-transferases (GSTs) detoxify environmental agents which influence the onset and progression of disease. Dysfunctional detoxification enzymes are responsible for prolonged exposure to reactive molecules and can contribute to endothelial damage, an underlying factor in CAD. Uncoupling proteins (UCPs) 2 and 3 play an important role in the regulation of oxidative stress which contributes to chronic inflammation. Coronary artery disease is a chronic inflammatory disorder characterized by elevated levels of C-reactive protein (CRP) and pro-inflammatory cytokines such as interleukin 6 (IL-6). Polymorphisms of these genes have been linked to CAD and other chronic diseases. Statins, metabolised in the liver, are the most commonly used drug to control atherosclerosis progression in CAD patients. The pleiotropic effects of statins have been attributed to both favourable and adverse outcomes in CAD patients particularly related to myopathy and hepatotoxicity. All patients (n=102) recruited into this study were South African Indian males. A corresponding age-, gender- and ethnicity-matched control group (n=100) was also recruited. The frequency of the GSTM1 +/0, GSTP1 A105/G105, IL6 -174G/C and CRP -390C/A/T genotypes was assessed by polymerase chain reaction (PCR) and PCR restriction fragment length polymorphism (PCR-RFLP). For the in vitro study, the biological effect of atorvastatin on HepG2 cells was assessed. The metabolic activity, cytotoxicity, oxidative stress and nitric oxide production was assessed by the ATP, lactate dehydrogenase (LDH), thiobarbituric acid reactive substance (TBARS) and Griess assays, respectively. The profile of 84 microRNA (miRNA) species was evaluated using the miRNA Pathway Finder PCR SuperArray. The predicted targets of up-regulated miRNAs were determined using the online software, Targetscan. The mRNA levels of guanidinoacetoacetate (GAMT), arginine glycine aminotransferase (AGAT) and spermine oxidase (SMO) were determined using quantitative PCR. Western blotting was used to determine GAMT and phosphorylated p53 levels in treated cells. The GSTM1 0/0 and GSTP1 A105/A105 genotypes occurred at higher frequencies in CAD patients compared with the control group (36% vs. 18% and 65% vs. 48%, respectively). A significant association with CAD was observed in GSTM1 0/0 (odds ratio (OR)=2.593; 95% confidence interval (CI) 1.353 - 4.971; p=0.0043) and GSTP1 A105/A105 OR=0.6011; 95% CI 0.3803 - 0.9503; p=0.0377). We found a significant association between smoking and CAD; the presence of either of the respective genotypes together with smoking increased the CAD risk (GSTP1 A105 relative risk (RR)=1.382; 95% CI 0.958 - 1.994; p=0.0987 and GSTM1 null RR=1.725; 95% CI 1.044 - 2.851; p=0.0221). The UCP2 -866G/A and UCP3 -55C/C genotypes occurred at highest frequency in CAD patients (59% vs. 52% and 66% vs. controls: 63% respectively) and did not influence the risk of CAD. Homozygous UCP3 -55T/T genotype was associated with highest fasting glucose (11.87±3.7mmol/L vs. C/C:6.11±0.27mmol/L and C/T:6.48±0.57mmol/L, p=0.0025), HbA1c (10.05±2.57% vs. C/C:6.44±0.21% and C/T:6.76±0.35%, p=0.0006) and triglycerides (6.47±1.7mmol/Lvs. C/C:2.33±0.17mmol/L and C/T:2.06±0.25mmol/L, p<0.0001) in CAD patients. A significant association between the G allele of the IL6 -174 polymorphism and non-diabetic CAD patients was found (p=0.0431 odds ratio: 1.307, 95% CI: 1.047-1.632). A significant association with the C allele of the -390 CRP triallelic variants and CAD (p=0.021 odds ratio: 1.75, 95% CI: 1.109-2.778) was also found using a contingency of the C allele vs. the minor A and T allele frequencies. The strength of the association of the C allele with non- diabetic CAD subjects was much higher (p=0.0048 odds ratio: 2.634, 95% CI: 1.350-5.138). Circulating median levels of IL-6 (0.9 (0.90, 0.91) pg/ml and 0.9 (0.87, 0.92) pg/ml) and CRP (5.65 (1.9, 8.2) mg/l and 2.90 (1.93, 8.35) mg/l) were similar between CAD patients and controls, respectively. A similar finding was observed between controls and non-diabetic CAD subjects. Levels of IL-6 and CRP in CAD subjects were not significantly influenced by polymorphic variants of IL-6 and CRP. In the control group, the level of IL-6 was significantly influenced by the IL6 -174 G allele (p=0.0002) and the CRP -390 C allele (p=0.0416), where subjects with the homozygous GG (0.9 (0.9, 1,78) pg/ml) and CC (0.9 (0.9, 0.95) pg/ml) genotype had higher levels than the C allele carriers (0.9 (0.64, 0.91) pg/ml) or A and T carriers (0.9 (0.69, 0.91) pg/ml) combined. The lowest measure of proliferation/metabolism in HepG2 cells was observed at 20μM atorvastatin, with 82±9.8% viability. The level of cytotoxicity was increased in statin treated cells from 0.95±0.02 units to 1.11±0.03 units (p=0.001) and malondialdehyde levels was reduced from 0.133±0.003 units to 0.126±0.005 units (p=0.009) whilst nitrite levels were elevated (0.0312±0.003 units vs. control: 0.027±0.001 units, p=0.044). MicroRNAs most significantly upregulated by atorvastatin included miR-302a-3p (3.05-fold), miR-302c-3p (3.61-fold), miR-124-3p (3.90-fold) and miR-222-3p (4.4-fold); miR-19a-3p, miR-101-3p and let-7g were downregulated (3.63-fold, 2.92-fold, 2.81-fold, respectively). A list of miRNA targets identified included those with a role in metabolism and inflammation. The miR-124a specifically targets the mRNA of GAMT and SMO.