Elevated lipoprotein(a) and genetic polymorphisms in the LPA gene may predict cardiovascular events

Elevated lipoprotein(a) [Lp(a)] is a risk factor for coronary heart disease (CHD), but there are few studies on the prediction of future cardiovascular events by Lp(a) and its LPA single nucleotide polymorphisms (SNPs). The aim of this study was to investigate whether elevated Lp(a) and its SNPs can predict cardiovascular events. We evaluated whether Lp(a) and LPA SNPs rs6415084 and rs12194138 were associated with the incidence rate and severity of CHD. All participants were followed up for 5 years. Elevated Lp(a) is an independent risk factor for the risk and severity of CHD (CHD group vs. control group: OR = 1.793, 95% CI: 1.053–2.882, p = 0.043; multiple-vessel disease group vs. single-vessel disease group: OR = 1.941, 95% CI: 1.113–3.242, p = 0.027; high GS group vs. low GS group: OR = 2.641, 95% CI: 1.102–7.436, p = 0.040). Both LPA SNPs were risk factors for CHD, and were positively associated with the severity of CHD (LPA SNPs rs6415084: CHD group vs. control group: OR = 1.577, 95% CI: 1.105–1.989, p = 0.004; multiple-vessel disease group vs. single-vessel disease group: OR = 1.613, 95% CI: 1.076–2.641, p = 0.030; high GS group vs. low GS group: OR = 1.580, 95% CI: 1.088–2.429, p = 0.024; LPA SNPs rs12194138: CHD group vs. control group: OR = 1.475, 95% CI: 1.040–3.002, p = 0.035; multiple-vessel disease group vs. single-vessel disease group: OR = 2.274, 95% CI: 1.060–5.148, p = 0.038; high GS group vs. low GS group: OR = 2.067, 95% CI: 1.101–4.647, p = 0.021). After 5 years of follow-up, elevated Lp(a) and LPA SNPs rs6415084 and rs12194138 can independently predict cardiovascular events. The increase of serum Lp(a) and LPA SNPs rs6415084 and rs12194138 are associated with increased prevalence and severity of CHD, and can independently predict cardiovascular events.


Measurements of lipoprotein(a) and other biomarkers.
Blood samples were collected in the morning after at least 12 h of fasting. All measurements were performed within 6 h. Lp(a) in the serum samples was measured using latex enhanced immunoturbidimetry. The Lp(a) detection kit (Roche Inc., Germany) was used to determine the precipitation at 800/660 nm using latex particles coated with anti Lp(a) antibody to allow agglutination with human lipoprotein. FPG, homocysteine (HCY), hypersensitive C-reactive protein (hs-CRP), serum lipid profiles, including triglycerides (TG), total cholesterol (TC), LDL-C, and high density lipoprotein cholesterol (HDL-C), were analyzed with a Beckman AU5832 analyzer (Beckman Coulter Inc., USA). Apolipoproteins A-1 (apoA1) and B (apoB) were measured by immunoturbidimetry (Daiichi Pure Chemicals Co., Ltd., Tokyo). Direct quantitative analysis of small dense low-density lipoprotein cholesterol (sdLDL-C) assay was done using sdLDL-C reagent kits (Denka Seiken Co., Ltd. Japan). The HbA1c was determined with highperformance liquid chromatography (Trinity Biotech Inc., USA). DNA extraction and genotyping. According to the manufacturer's instructions, genomic DNA was extracted from 3.5 ml EDTA anticoagulant blood collection vessel (Becton, Dickinson and company, USA) using puregene DNA separation kit (TianGen Biotech, Beijing, China). The incidence of LPA variant rs6415084 or rs12194138 was determined by gene sequencing (TsingKe Biological Technology, Beijing, China). All Sanger sequencing data shall be visually inspected by three experienced operators in our laboratory.  Table 1. Baseline characteristics of all subjects. Data are reported as means ± SD or n(%), median (interquartile ranges). SD: Standard deviation. Statistical analysis was performed with the student t test or Mann-Whitney U test and with Chi-square test for categorical variables. BMI Body mass index, HbA1c Hemoglobin A1c, apoB Apolipoprotein B, apoA1 Apolipoprotein A1, HDL-C High density lipoprotein cholesterol, LDL-C Low density lipoprotein cholesterol, sdLDL-C Small dense low-density lipoprotein cholesterol, Hs-CRP Hypersensitive C-reactive protein, HCY Homocysteine, Lp(a) Lipoprotein(a). www.nature.com/scientificreports/ The serum Lp (a) level and the prevalence of LPA SNPs were associated with the severity of CHD. The patients with CHD were then classified into single-vessel (n = 287), two-vessel (n = 310), and multiple-vessel disease (n = 504) groups (Table 2). We found a significant higher in the Lp(a) levels in multiple-vessel diseases group [31.40 (41.32)  The effect of two SNPs on serum Lp(a) levels in Chinese Han people. The power values of SNPs rs6415084 and rs12194138 were 99% and 90%. The allele frequencies of SNPs rs6415084 and rs12194138 were 7.59% and 2.93%. The frequencies of the two tested variants did not deviate significantly in all subjects from the Hardy-Weinberg equilibrium: rs6415084, F = 2.630, p = 0.105; rs12194138, F = 2.502, p = 0.113. Table 4 shows the relationship between LPA SNPs and serum Lp(a) levels. In LPA SNP rs6415084, the Lp(a) levels of rs6415084 (CC) genotype, rs6415084 (CT) genotype and rs6415084 (TT) genotype were different in all participants and CHD group (p < 0.001). In the control group, the Lp(a) levels of rs6415084 (CT/TT) genotype were higher than those of rs6415084 (CC) genotype (p < 0.001). In LPA SNP rs12194138, the Lp(a) levels of rs12194138 (AT/TT) genotype were higher than that of rs12194138 (AA) genotype in different groups (p < 0.001; p = 0.009; p < 0.001).
Serum Lp (a) levels and LPA SNPs were associated with the risk of CHD.. In order to explore whether Lp(a) and LPA SNPs increase the risk of CHD, we conducted univariate and multivariate logistic regres- Table 2. Baseline characteristics of patients with multi vessel coronary artery disease. Data are reported as means ± SD or n (%), median (interquartile ranges). SD: Standard deviation. Statistical analysis was performed with the ANOVA or Kruskal-Wall test and and with Chi-square test for categorical variables. BMI Body mass index, HbA1c Hemoglobin A1c, apoB Apolipoprotein B, apoA1 Apolipoprotein A1, HDL-C High density lipoprotein cholesterol, LDL-C Low density lipoprotein cholesterol, sdLDL-C Small dense low-density lipoprotein cholesterol, Hs-CRP Hypersensitive C-reactive protein, HCY Homocysteine, Lp(a) Lipoprotein(a). *p < 0.05 compared with the 1 vessel group. # p < 0.05 compared with the 2 vessels group. www.nature.com/scientificreports/    Table 6. There were significant differences in genotype (rs6415084: CC vs. CT/TT; rs12194138: AA vs. AT/TT) and allele frequency (rs6415084: C vs. T; rs12194138: A vs. T) between CHD group and control group, multi vessel disease group and single vessel disease group, high GS group and low GS group (p < 0.05 for both). After adjusting for age, sex, BMI, diabetes, hypertension, smoking, consumers of alcohol, FPG, HbA1c, ApoB, ApoA1, TC, TG, HDL-C, hs-CRP and HCY were adjusted by multiple logistic regression analysis, it was found that both LPA SNPs rs6415084 (CT/TT ) and rs12194138 (AT/TT) were risk factors for CHD, and were positively associated with the severity of CHD (LPA SNPs rs6415084: CHD group vs. control group: OR = 1.577, 95% CI:   Tables 7 and 8, the total incidence of MACEs in the high Lp(a) group was significantly higher than that in the low Lp(a) group, both in the control group and the CHD group. These data indicate that Lp(a) level is associated with MACEs of all subjects. Further study of LPA SNPs and cardiovascular outcomes showed that MACEs were significantly higher in individuals with LPA SNPs rs6415084 (CT/TT) and rs12194138 (AT/TT) genotype carriers than CC and AA genotype carriers in all subjects (Fig. 2B,C) (HR = 2.499, 95% CI: 1.783-3.502, p < 0.001; HR = 2.565, 95% CI: 1.545-4.257, p = 0.001). In the control group, the results were consistent with the test results of all subjects (Fig. 3B,C) (HR = 1.949, 95% CI: 1.099-3.456, p = 0.022; HR = 3.087, 95% CI: 1.238-7.698, p = 0.016). The results were the same in the CHD group (Fig. 4B,C) (HR = 2.441, 95% CI: 1.628-3.661, p < 0.001; HR = 2.298, 95% CI: 1.254-4.211, p = 0.007). As shown in Tables 9 and 10, the total incidence of MACEs in LPA SNPs rs6415084 (CT/ TT) and rs12194138 (AT/TT) genotype carriers was significantly higher than that in CC and AA genotype carriers, and there were differences in the non-fatal MI and hospitalized unstable angina of SNPs rs6415084 (CT/TT vs. CC) in the CHD group in addition to the total incidence. These results indicate that the LPA SNPs rs6415084 (CT/TT) and rs12194138 (AT/TT) genotype are related to the occurrence of cardiovascular events in the future.

Discussion
Recently, the role of Lp(a) in cardiovascular diseases has attracted more and more attention 15,19,20 . The association between Lp(a) and CHD, which is independent of traditional cardiovascular risk factors, has been known for many years 4,21 . It is based on findings mainly from studies of healthy participants in the general population and investigations of patients with CHD 22 . This study demonstrates that Lp(a) level and the prevalence of LPA SNPs are associated with the risk and severity of CHD, and LPA SNPs rs6415084 and rs12194138 are significantly associated with serum Lp(a) levels. That the increase of Lp(a) level was the key variable to predict CHD risk, and the role of LPA SNPs rs6415084, rs12194138 and Lp(a) level in predicting CHD risk was clarified. Most Table 6. Relationship between two SNPs and coronary heart disease and its severity. Crude OR was determined by χ 2 test, cases versus control subjects. Adjusted OR was obtained on multivariate logistic regression after controlling for age, sex, body mass index, diabetes, hypertension, smoking, consumers of alcohol, hemoglobin A1c, fasting plasma glucose, apolipoprotein B, apolipoprotein A1, total cholesterol, triglycerides, high density lipoprotein cholesterol, hypersensitive C-reactive protein, and homocysteine. M = C and m = T for single nucleotide polymorphism (SNP) rs6415084; M = A and m = T for single nucleotide polymorphism (SNP) rs12194138. CHD Coronary heart disease, CI Confidence interval. www.nature.com/scientificreports/ of the previous data on LPA and Lp(a) were studied in Caucasian populations in Europe [23][24][25] . However, there are few studies on the relationship between LPA and Lp(a) and the risk of coronary heart disease in Chinese Han population. At the same time, our study also predicted the occurrence of MACEs between rs6415084 and rs12194138 in Chinese Han population, so as to better study the role of Lp(a) level and LPA SNPs in predicting future MACEs of different nationalities. We demonstrated that the rs6415084 and rs12194138 polymorphisms were significantly more frequent in patients with CHD than in healthy subjects. A higher risk of CHD was observed for the rs6415084 CT/TT and the rs12194138 AT/TT heterozygotes and homozygotes polymorphism carriers. The study of Lee et al. 15 , showed that the LPA SNPs was associated with the size of apolipoprotein (a) isoforms and the serum level of Lp(a) in different ethnicity. Clarke et al. 3 showed that LPA SNPs rs10455872 and rs3798220 were strongly associated with increased level of Lp(a), a reduced copy number in LPA, and a small Lp(a) size. Lanktree et al. 14  Consistent with the role of Lp(a) in predicting future MACEs, it can be concluded that higher Lp(a) levels are primarily important to CHD risk prediction 27,28 . The increase of plasma Lp(a) level can promote thrombosis, and there is a correlation between Lp(a) and atherosclerotic stenosis. The same results are presented in some recent research reports 29 .
At present, most of the studies on LPA SNPs focus on the risk of CHD and atherosclerosis 30 . The prediction of future MACEs by SNPs is only seen in a few articles, and most of them focus on the European population 31 . Therefore, our study of the prediction of MACEs by two SNPs rs6415084 and rs12194138 in Chinese Han population, in order to better study the role of Lp(a) level and LPA SNPs in predicting future MACEs in different ethnicity. Our results are similar to those predicted by Gudbjartsson et al. 1 , for patients with diabetes mellitus with CHD in Iceland. CHD is a multifactorial disease, the combination of genetic variation and environmental factors may lead to phenotypic variation 32 . In our study, elevated Lp(a) level is an independent MACEs predictor; www.nature.com/scientificreports/ and some specific LPA SNPs variations may cause the increase of serum Lp(a), which also explains that SNPs rs6415084 and rs12194138 are independent MACEs predictors. Some studies have shown that LPA SNPs variation can increase Lp(a) level, but there is no direct correlation between LPA SNPs and MACEs 3,33 . The reason for this difference was not fully understood, but the discrepancy between the results of the studies may be caused by a variety of confounding factors, such as different population characteristics, study design, disease status, or confounding variables. Therefore, we used a large number of Han Chinese in this study. We not only found that serum Lp(a) levels and LPA SNPs variants were associated with the risk and severity of coronary heart disease, but they were independent predictors of MACEs. Serum Lp(a) level is largely determined by the variation of LPA in many populations. Genetic variation of LPA is directly related to the risk of cardiovascular disease. The reason and mechanism of LPA variants rs6415084 and rs12194138 and Lp(a) level increase remains unclear 2,14 . The mechanism of increased Lp (a) lipoprotein level increasing the risk of coronary heart disease is unclear, which may involve LDL lipoprotein cholesterol 34 ,  www.nature.com/scientificreports/  www.nature.com/scientificreports/ inhibition of plasminogen to plasmin 35 , inhibition of tissue factor expression 36 , or carrying pro-inflammatory oxidized phospholipids 37 . From our results, we found that the level of Lp(a) is positively associated with the risk and severity of CHD. Besides, the present study has several limitations. First, only two centers were involved in the research, which might have led to selective biases in the data results, and some of the conclusions should be verified in larger multicenter studies. Second, we have only studied two SNPs and the power value of SNPs rs12194138 is low. In our future work, we will increase more SNPs and sample size. In addition, CAG results lack of centralized core laboratory for angiography analysis. However, our data still provide the incidence rate and severity of CHD increase with the increase of serum Lp(a) level. In addition, the increased serum Lp(a) level and SNPs rs6415084 and rs12194138 variants will also increase the risk of cardiovascular events in the future.

Conclusion
In conclusion, our data support the association of elevated serum Lp(a) levels and LPA SNPs rs6415084, rs3798220 variants with the risk and severity of coronary heart disease, and the prediction of future cardiovascular events.

Data availability
The datasets used and/or analyzed during the current study are not publicly available but are available from the corresponding author on reasonable request.