In total, 477 IS patients (316 males and 161 females) and 493 control subjects (325 males and 168 females) were recruited. There were no significant differences between patients and controls in terms of gender (p = 0.898). The mean age was 64.13 ± 10.82 years for the patients with IS and 60.05 ± 6.56 years for the control subjects. Significant differences were also found in age distribution (p < 0.001), suggesting that age may have an effect on the etiology of IS. The total protein, serum uric acid, blood glucose, bilirubin, triglyceride, hemoglobin, cholesterol and low-density lipoprotein levels in the IS patients were significantly different from those noted in the healthy control subjects. The clinical characteristics of the patients were described in Table 1.
Seven SNPs in CYP3A4 and CYP11A1 were successfully genotyped, and the average variant call rate was 99.6%. Detailed information and potential function of candidate SNPs were listed in Table 2. These intronic SNPs were associated with the regulation of promoter and/or enhancer histones, changed motifs, and selected eQTL hits, suggesting they might exert biology functions in silico. MAF of all SNPs was higher than 5% of the study population. All SNPs were in HWE among the controls (p > 0.05).
The allele and genotype frequency distributions of the SNPs and their association with IS susceptibility were shown in Table 3 and Supplementary Table 2. CYP3A4 SNPs rs3735451 and rs4646440 were associated with reduced susceptibility of IS (Table 3). We found that individuals carrying rs3735451-C allele had a decreased risk of IS in allele model (OR = 0.81, 95% CI: 0.66-0.98, p = 0.039), genotype model (OR = 0.74, 95% CI: 0.57-0.97, p = 0.029), dominant model (OR = 0.73, 95% CI: 0.56-0.95, p = 0.018) and additive model (rs3735451 OR = 0.78, 95% CI: 0.63-0.96, p = 0.019), respectively. With rs4646440 GG genotype as reference, the presence of the GA genotype was associated with a significantly decreased risk of IS after adjustment for age and gender (GA vs. GG, OR = 0.72, 95% CI: 0.55-0.95, p = 0.021; GA-AA vs. GG, OR = 0.72, 95% CI: 0.55-0.94, p = 0.017, Table 3). Furthermore, rs4646440 polymorphism also might reduce the susceptibility to IS under additive model (OR = 0.77, 95% CI: 0.61-0.97, p = 0.024). Nevertheless, other polymorphisms in CYP3A4 and CYP11A1 did not relate to IS susceptibility (Supplementary Table 2).
We further analyzed whether the genotypic effects on IS risk were dependent on gender (Table 4). We found that CYP3A4 rs4646440 was associated with a decreased risk under the additive model (OR = 0.74, 95% CI: 0.56-0.98, p = 0.038), and showed a marginal p value in allele model (OR = 0.76, 95% CI: 0.58-1.00, p = 0.050) among males, which indicated insufficient evidence for claiming an association. CYP11A1 rs12912592 polymorphism also showed significant risk-increasing effects in the heterozygote model (OR = 1.58, 95% CI: 1.04-2.42, p = 0.034), and dominant model (OR = 1.56, 95% CI: 1.02-2.37, p = 0.039).
In the stratification of age, CYP3A4 SNPs rs3735451 and rs4646440 were associated with the susceptibility to IS at age 61 years (Table 5). For rs3735451, the C allele carriers had a decreased risk of IS (OR = 0.63, 95% CI: 0.41-0.96, p = 0.031 for CT vs. TT genotypes; OR = 0.65, 95% CI: 0.43-0.97, p = 0.036 for CT-CC vs. TT genotypes) after adjusting for age and gender. For rs4646440, we found that the A allele was significantly associated with a reduced risk of IS (GA vs.GG, OR = 0.57, 95% CI: 0.37-0.88, p = 0.012; and GA-AA vs.GG OR = 0.60, 95% CI: 0.40-0.91, p = 0.017). Among the population under the age of 61, we found that CYP3A4 rs4646437, CYP11A1 rs12912592 and rs28681535 were associated with IS risk. CYP3A4 rs4646437 and CYP11A1 rs28681535 polymorphisms were significantly associated with decreased risk for IS (rs4646437, OR = 0.59, 95% CI: 0.37-0.95, p = 0.029; and rs28681535, OR = 0.66, 95% CI: 0.45-0.98, p = 0.038). Additionally, the carriers of the T allele at CYP11A1 rs12912592 appeared to have a higher risk of IS (T vs G, OR = 1.64, 95% CI: 1.04-2.61, p = 0.043; GT vs GG, OR = 1.84, 95% CI: 1.11-3.05, p = 0.017 and GT-TT vs GG, OR = 1.89, 95% CI: 1.15-3.12, p = 0.013).
We next performed haplotype analyses, and the results showed that CYP3A4 rs4646440 was in strong linkage disequilibrium (LD) with rs35564277. Additionally, three CYP11A1 SNPs (rs1484215, rs12912592, and rs28681535) were in strong LD, as shown in Figure 1. We found that CYP3A4 GT haplotype conferred an increased risk of IS after adjusted by age and gender (OR = 1.29, 95% CI: 1.02-1.62, p = 0.033, Table 6).
Furthermore, we also assessed the association of the selected SNPs and clinical variables in patients (Table 7). Significant association was observed between the genotypes of the CYP3A4 SNPs rs3735451 and rs4646440 and the levels of total protein (p = 0.021 and p = 0.043, respectively). A significant association of CYP11A1 rs12912592 polymorphism with total bilirubin was identified (p = 0.025). Besides, the TT genotype of CYP11A1 rs28681535 was higher high-density lipoprotein cholesterol level than GT genotype and GG genotype (p = 0.027). However, there was no difference in the remaining lipid parameters among the genotypes of the selected SNPs (p > 0.05 for all).