Association between peroxisome proliferator-activated receptor-alpha, delta, and gamma polymorphisms and risk of coronary heart disease: A case-control study and meta-analysis.

OBJECTIVES: Risk of coronary heart disease (CHD) has been suggested to be associated with polymorphisms of peroxisome proliferator-activated receptors (PPARs), while the results were controversial. We aimed to systematically assess the association between PPAR polymorphisms and CHD risk.
METHODS: A case-control study with 446 subjects was conducted to evaluate the association between CHD risk and C161T polymorphism, which was of our special interest as this polymorphism showed different effects on risks of CHD and acute coronary syndrome (ACS). Meta-analyses were conducted to assess all PPAR polymorphisms. Either a fixed- or a random-effects model was adopted to estimate overall odds ratios (ORs).
RESULTS: In the case-control study, T allele carriers of C161T polymorphism were not significantly associated with CHD risk (Odds ratio (OR) = 0.74, 95% confidence interval (CI) 0.47-1.15, P = 0.19), while T allele carriers showed higher risk of ACS (OR = 1.63, 95% CI 1.00-2.65, P = 0.048). The meta-analysis indicated that compared with CC homozygous, T allele carriers had lower CHD risk (OR = 0.69, 95% CI 0.59-0.82, P < 0.001) but higher ACS risk (OR = 1.43, 95% CI 1.09-1.87, P = 0.010). Three other polymorphisms were also found to be significantly associated with CHD risk under dominant model: PPAR-alpha intron 7G/C polymorphism (CC+GC vs GG, OR 1.42, 95% CI 1.13-1.78, P = 0.003), L162V polymorphism (VV+LV vs LL, OR 0.74, 95% CI 0.56-0.97, P = 0.031), and PPAR-delta +294T/C polymorphism (CC+TC vs TT, OR 1.51, 95% CI 1.12-2.05, P = 0.007).
CONCLUSIONS: The results suggested that PPAR-alpha intron 7G/C and L162V, PPAR-delta +294T/C and PPAR-gamma C161T polymorphisms could affect CHD susceptibility, and C161T polymorphism might have different effects on CHD and ACS.

Introduction

Progression of atherosclerosis plays a critical role in the development of coronary heart disease (CHD), and is influenced by multiple genetic and environmental factors. Peroxisome proliferator-activated receptors (PPARs) are ligand-activated nuclear transcription factors that regulate lipid and carbohydrate metabolism. Three subgroups have been identified: PPAR alpha (PPARA), PPAR delta (PPARD), and PPAR gamma (PPARG). PPARs can alter metabolism by binding to specific elements in the promoter region of target genes. In addition to the influence on lipid and glucose metabolism, PPARs also have many other functions. For instance, PPARG plays an important role in the suppression of inflammation, free radical generation, and smooth muscle cell growth. [ ] Besides, PPARG was found to be expressed in atherosclerotic lesions and macrophage foam cells, suggesting that PPARG may influence atherosclerogenic processes.[ ]

In recent years, there is growing interest in the link between PPAR polymorphisms and CHD risk, including PPARA intron 7G/C, PPARD +294T/C, PPARG P12A, and C161T (the general description of these polymorphisms are shown in Fig. 1). [ ] However, the current results are inconsistent and no concrete conclusion can be drawn regarding the relationship between PPAR polymorphisms and CHD risk. The previous meta-analyses on the association between PPAR polymorphisms and CHD risk evaluated PPARG polymorphisms but not PPARA and PPARD polymorphisms. Besides, the results of these meta-analyses were inconsistent. [ ] For example, Xu et al found no significant association between C161T polymorphism and CHD susceptibility, while another meta-analysis indicated that C161T polymorphism was statistically associated with CHD risk among Chinese. We found that for C161T polymorphism, all the studies reported that T allele was negatively associated with CHD risk.[ ] On the other hand, only 2 studies investigating acute coronary syndrome (ACS) indicated a non-significant positive association between C161T polymorphism and ACS risk (CT+TT vs CC, OR = 1.45, 95% CI 0.91–2.32 for Evangelisti et al ; OR = 1.25, 95% CI 0.79–1.98 for Chao et al ). To further clarify this issue, we conducted a case–control study to further assess the possible different role of C161T polymorphism between CHD and ACS risk. Besides, a meta-analysis was also applied to evaluate all PPAR polymorphisms.

Figure 1

Meta-analysis of PPARD+294T/C polymorphism. PPARD = peroxisome proliferator-activated receptor delta.

Methods

This case–control study and meta-analysis was designed, conducted, and reported according to STROBE, MOOSE, and PRISMA statements. [ ]

Case–control study

Study subjects

We enrolled 281 consecutive CHD patients in this study, including 137 ACS patients. All subjects were documented by angiography and were recruited from May 2011 to June 2012 at Department of Cardiology, The First People's Hospital of Hangzhou. CHD was defined as at least one stenosis of >50% diagnosed in a major coronary artery. Diagnosis of ACS was established according to American College of Cardiology Foundation/American Heart Association. A total of 165 healthy subjects were recruited in this study as controls.

Ethics statement

This study was approved by Ethics Committee of The First People's Hospital of Hangzhou (IRB approval number: HUM00081230), and all participants were given and signed the written informed consent form.

Genotyping

Genomic DNA was extracted from peripheral lymphocytes using a whole blood DNA isolation kit (TIANamp Genomic DNA kit, Tiangen, China). Genotyping of the C161T polymorphism at exon 6 of the PPAR-gamma gene was performed by polymerase chain reaction (PCR), PCR kit was provided by TAKARA, Japan. The forward primer was 5′-CAA GAC AAC CTG CTA CAA GC-3′ and reverse primer was 5′-TCC TTG TAG ATC TCC TGC AG-3′. The amplification was performed in a 20 μL volume containing 100 ng DNA, 20 pmol of each primer, 1.5 mmol/L MgCl2, 50 mmol/L KCl, 25 μmol/L dNTP, and 1 Unit Taq polymerase. Samples were subjected to denaturing at 95°C for 5 minutes followed by 30 cycles of 95°C for 30 seconds, 58°C for 1 minute, 72°C for 1 minute. The final thermal cycle was at 72 °C for 5 minutes. All PCR products were sent to sequencing for results.

Statistical analyses

The data were analyzed using the SPSS software (Version 13.0; SPSS, Chicago, IL). Since the number of the TT homozygous patients was small, allelic variants were dichotomized into T carriers (TT and CT) and CC homozygous. The statistical significance was defined as P < 0.05.

Meta-analysis

Ethics review board approval was not required for the meta-analysis process, since no animal experiment or direct human trial was conducted in this section.

Literature search and selection criteria

PUBMED and EMBASE (up to October 2015) were searched to identify eligible studies using key words relating to CHD disease (“coronary heart disease” OR “coronary artery disease” OR “myocardial infarction (MI)” OR “acute coronary syndrome” OR “ischemic heart disease (IHD)” OR “cardiovascular disease” OR “CHD” OR “CAD” OR “MI” OR “ACS” OR “IHD”), PPAR (“peroxisome proliferator-activated receptor” OR “PPAR” OR “PPAR alpha” OR “PPAR delta” OR “PPAR gamma”), and polymorphism (“polymorphism” OR “variant” OR “SNP” OR “mutation”). References of relevant articles were also scanned for potentially missing studies. Titles and abstracts were scanned and then full papers were reviewed. Articles published in English and Chinese were retrieved. The retrieved studies were carefully examined to exclude potential duplicates or overlapping data.

Articles were included if they met all the following criteria: study should evaluate the association between PPAR polymorphism and cardiovascular diseases, including CHD, MI, ACS, and other IHD; odds ratio (OR) estimates and their 95% confidence intervals (95% CI) were given or sufficient data were available to calculate these numbers; and each polymorphism should be analyzed by at least 2 studies.

Our original case–control study met all the inclusion criteria and would also be included in the meta-analysis.

Data extraction

Two reviewers (Drs YQ and PL) independently extracted study characteristics using standardized forms, and the following information was extracted from each study: first author, year of publication, country, study design, phenotype (disease), genotype of cases and controls, consistency of genotype frequencies with Hardy–Weinberg equilibrium (HWE). Discrepancies were resolved by a third investigator. The minor allele frequency (MAF) was calculated using the extracted genotype.

Statistical analysis

Dominant genetic model was adopted to pool ORs of PPAR polymorphisms first, as dominant genetic model was applied in most of the studies included. Then recessive and additive models were also applied. The significance of the pooled OR was determined by Z test (P < 0.05 was considered to be significant). The extent of heterogeneity across studies was checked using the χ2 test and I 2 test (I 2 test quantifies the proportion of total variation across studies due to heterogeneity rather than chance); P ≤ 0.10 in combination with I 2  > 50% indicates significant heterogeneity. The OR was pooled by applying a fixed-effects model or a random-effects model according to heterogeneity. When P > 0.10, OR of each study was pooled by the fixed-effects model, otherwise the random-effects model was applied. Subgroup analyses were adopted to assess C161T and P12A polymorphisms as they were evaluated by sufficient studies. Funnel plots were constructed and Begg and Egger tests were used to assess publication bias, and P ≤ 0.10 was considered to be significant. All analyses were conducted using the Stata software (version 11.0; StataCorp, College Station, TX).

Results

The basic characteristics of the cases and controls are shown in Table 1. The mean age was 57.3 ± 11.5 years in ACS group and was higher than control group (P < 0.05), while no significant difference was found in mean ages in CHD group and control group. No statistical difference in sex was found in case groups (CHD group and ACS group) and control group. As expected, both CHD group and ACS group had a higher prevalence of smoking habit, hypertension, diabetes, and higher levels of total-cholesterol, LDL-cholesterol, and triglycerides. We did not found significant difference in HDL-cholesterol level between case groups and control group.

Table 1

Basic characteristics of participants in the case–control study.

The genotype distributions and allele frequencies of C161T polymorphism in cases and controls are shown in Table 2. Genotype frequencies in control group were in HWE (P = 0.53). Compared with CC homozygous, T allele carriers (CT+TT) were found to increase ACS risk (OR = 1.63, 95% CI 1.00–2.65, P = 0.048), while no significant association between C161T polymorphism and CHD risk was found (OR = 0.74, 95% CI 0.47–1.15, P = 0.19).

Table 2

C161T polymorphism genotype distributions and allele frequencies in cases and controls.

The Meta-analysis

A total of 38 eligible articles (comprising 40 studies) were finally extracted from databases and our own case–control study was also included for the meta-analysis.[ ] The selection process is detailed in Fig. 2. The characteristics of the included studies were presented in Supplementary Table 1. Among the 38 articles, 18 articles evaluated PPARG P12A polymorphism, and risk of CHD; 8 articles investigated PPARG C161T polymorphism; 5 articles reported PPARD +294T/C polymorphism; 4 articles assessed PPARA intron 7G/C polymorphism; and the number of studies assessing PPARA L162V, PPARG −681C/G, −689C/T, and C1431T was 5, 3, 3, and 5, respectively. All studies followed HWE, except 1 study assessing intron 7G/C polymorphism.

Figure 2

Flow diagram of study selection process.

Association between PPARG C161T polymorphism and CHD risk

Seven studies (including our own study) containing 3089 participants (1921 cases and 1168 controls) evaluated the association between C161T polymorphism and risk of CHD. The pooled OR was 0.69 (95% CI 0.59–0.82, P < 0.001) under dominant model (CT+TT vs CC), indicating a significant association between C161T polymorphism and CHD risk (Fig. 3A). No significant heterogeneity was found (I 2 = 0%, P = 0.635) (Fig. 3A).

Figure 3

Meta-analysis of PPARG C161T polymorphism. (A) Forest plot of the association between C161T polymorphism and coronary heart disease (CHD) risk. (B) Forest plot of the association between C161T polymorphism and acute coronary syndrome (ACS) risk. PPARG = peroxisome proliferator-activated receptor gamma.

Three studies (including our own study) with 1091 participants (485 cases and 606 controls) assessed association between C161T polymorphism and ACS risk. Compared with CC homozygous, T allele carriers showed significantly higher ACS risk (OR = 1.43, 95% CI 1.09–1.87, P = 0.010) without significant heterogeneity (I 2 = 0%, P = 0.737) (Fig. 3B).

For all the nine studies (including our own study) focused on the association between C161T polymorphism and CHD risk (including ACS), 3878 participants (2269 cases and 1609 controls) were included. An inverse association was found between T allele and CHD risk (pooled OR = 0.80, 95% CI 0.64–0.99, P = 0.041) (Table 3). Results of subgroup analyses are shown in Table 4.

Table 3

Association between PPAR polymorphisms and CHD risk under dominant model.

Table 4

Subgroup analysis results of C161T and P12A polymorphisms.

Association between PPARA intron 7G/C, L162V, and PPARD +294T/C polymorphisms and CHD risk

A total of 4 studies assessed intron 7G/C polymorphism, while L162V polymorphism was evaluated by 4 articles including 5 studies. Under dominant model, patients with C allele had a significantly higher risk of CHD (CC+GC vs GG, OR = 1.42, 95% CI 1.13–1.78, P = 0.003) (Fig. 4A), while an inverse association between V allele in L162V polymorphism and CHD risk was suggested (VV+LV vs LL, OR = 0.74, 95% CI 0.56–0.97, P = 0.031) (Fig. 4B).

Figure 4

Meta-analysis of PPARA intron 7G/C and L162V polymorphisms. (A) Forest plot of the association between intron 7G/C polymorphism and CHD risk. (B) Forest plot of the association between L162V polymorphism and CHD risk. PPARA = peroxisome proliferator-activated receptor alpha.

For +294T/C polymorphism, 5 studies including 1301 cases and 1873 controls were included, and the pooled analysis indicated a significant association with CHD risk (CC+TC vs TT, OR = 1.51, 95% CI 1.12–2.05, P = 0.007) (Fig. 1).

Other PPAR polymorphisms and CHD risk

We also assessed the association between PPARG P12A, −681C/G, −689C/T, and C1431T polymorphisms and CHD risk. No significant association with CHD risk was found under dominant model, as shown in Table 3 and Supplementary Figure 1.

Results of recessive and additive models for all the PPARA, PPARD, and PPARG polymorphisms are shown in Supplementary Table 2. No Publication bias was observed in the meta-analyses.

Discussion

Progression of atherosclerosis is the main reason of CHD, and is influenced by numerous factors such as high plasma LDL-C concentration, blood glucose level, inflammation, and oxidant stress. Meanwhile, many of the above factors have been proven to be closely related to PPAR polymorphisms.[ ] Thus, much attention has been paid to evaluate the link between PPAR polymorphisms and CHD risk. We conducted a case–control study to evaluate the association between C161T polymorphism and CHD risk, indicating that C161T polymorphism was not significantly associated with total CHD risk but T allele carriers showed higher ACS risk. In the meta-analysis, a total of 40 studies were included and 8 polymorphisms in PPARA, PPARD, and PPARG were assessed, as shown in Fig. 5. The meta-analysis results of C161T polymorphism indicated T allele carriers had lower CHD risk but higher ACS risk. Moreover, PPARA intron 7G/C, L162V, and PPARD +294T/C polymorphisms also affected CHD susceptibility.

Figure 5

General descriptions of PPAR polymorphisms. BMI = body mass index, PPAR = peroxisome proliferator-activated receptor. ∗ Stands for polymorphisms significantly associated with risk of coronary heart disease (CHD).

When only ACS was assessed, both our case–control study and the meta-analysis indicated a statistically positive association with C161T polymorphism. Interestingly, the meta-analysis of C161T polymorphism showed an inverse association with total CHD risk. The results might indicate that C161T polymorphism has different effects on stable CHD and ACS. The association between C161T polymorphism and CHD risk might involve multiple mechanisms. First, C161T polymorphism might influence CHD risk by modulating adipose metabolism in CHD patients with diabetes. Wan et al reported that T allele carriers had significantly lower triglyceride levels than CC homozygote carriers in CAD patients with type 2 diabetes mellitus (T2DM). Second, C161T polymorphism might influence CHD risk by altering inflammatory cytokines. CC homozygotes seemed to have higher MMP-9 and TNF-α levels compared with T allele, which may partially explain why CC homozygotes are more susceptible to CHD. However, the 161T allele was also associated with an increased plasma leptin level, which is a marker of inflammation and might be involved in the genesis of acute myocardial infarction. [ ] Third, it was found that for gene–gene interaction between C161T polymorphism and apolipoprotein E (ApoE), compared with CT carriers, CC homozygotes accentuated the cholesterol difference between apoE4 carriers and non-apoE4 carriers. Interaction between C161T polymorphism and P12A polymorphism was also reported, and it has been shown that in diabetic CHD patients, serum triglyceride and VLDL-C levels increased in the order of P12P-CC<P12P-CT<P12A-CC<P12A-CT, suggesting the favorable effects of P12P genotype in lowering triglyceride level. There are 3 kinds of mRNAs produced by PPARG gene transcription, namely PPARG1, PPARG2, and PPARG3. C161T polymorphism occurs in all 3 mRNAs, the effects of this polymorphism might be diverse, and C161T polymorphism might have both pro- and anti-atherosclerotic effects. This may partially explain the different effects of T allele on ACS and non-ACS CHD patients. However, it should be noted that ACS and non-ACS CHD patients are not quite distinguishable and the non-ACS patients may have an ACS attack in their later lives. So the results should be interpreted cautiously, and more studies, especially prospective studies with long follow-up time, are warranted to further validate this phenomenon. Besides, the underlying molecular mechanisms remain unclear and are warranted to be investigated.

Several studies also indicated the relationship between C161T polymorphism and severity of CHD. Wang et al reported a frequency of T allele carriers (CT+TT) of 21.3% for CHD patients with 1 diseased vessel, 28.1% for patients with 2 diseased vessels, and 31.8% for those with 3 diseased vessels (P = 0.004), suggesting that CHD in patients with T allele tends to be more severe. Besides, Wan et al indicated that T allele carriers were at a higher risk of severe stenosis in CHD plus type 2 diabetes mellitus patients, indicating diabetes might modulate the effects of C161T polymorphism on CHD severity. However, there were no significant association between PPARG C1431T, P12A polymorphisms and number of diseased vessels in CHD patients.[ ] Moreover, ACS includes unstable ST or non-ST elevation MI and unstable angina, which are quite different from each other. Thus, it would be interesting and important to include factors such as the cardiac enzymes, left ventricular systolic function, and Killip class in the future analysis.

Three other polymorphisms, PPARA intron 7G/C, L162V, and PPARD +294T/C, also showed significant associations with CHD risk. Our study did not support a significant association between PPARG P12A, −681C/G, −689C/T, and C1431T polymorphism and CHD risk. Interestingly, in the subgroup analysis of P12A polymorphism, we found that in Asian populations, P12A polymorphism was shown to be associated with CHD susceptibility (as shown in Table 4). However, only 2 studies evaluated P12A polymorphism in Asian populations, thus this conclusion should be considered carefully.

Several meta-analyses assessed the association between PPARG polymorphisms and CHD risk and only C161T, P12A, C1431T polymorphisms were evaluated, [ ] while the current study is the first meta-analysis evaluating PPARA intron 7G/C, L162V, and PPARD +294T/C. A total of 8 polymorphisms were assessed and 4 of them were found to be significantly associated with CHD risk. For C161T polymorphism, results of previous meta-analyses were inconsistent: Xu et al found no significant association between C161T polymorphism and CHD risk, Wu et al indicated C161T polymorphism was only associated with CHD risk among Chinese and T allele was associated with reduced CHD risk. Another study reported that T allele of C161T polymorphism was associated with increased CHD risk (CT+TT vs CC, pooled OR = 1.182, 95% CI 1.023–1.341), but most of the studies included reported an opposite trend, furthermore, we found that one study included in that meta-analysis actually assessed C1431T polymorphism (not C161T) which leads to the ambiguous results. Our study adds additional evidence of the effect size of the C161T polymorphism on CHD and ACS risk, and found that C161T polymorphism might have different effects on CHD and ACS susceptibility. No publication bias was found in this study, and subgroup analyses were performed to assess the role of some polymorphisms in special populations or under certain conditions.

The current analysis also has several limitations. First, most of the included studies were from Asia, Europe, and USA, so the conclusions may not be true for other ethnic groups. Second, significant interstudy heterogeneity was found and could not be completely explained when assessing some polymorphisms. Third, gene–gene interaction and gene–environment interaction were not assessed because of insufficient data, though they were important in the genesis of CHD.

To conclude, our study supported that PPARA intron 7G/C and L162V, PPARD +294T/C, and PPARG C161T polymorphisms could affect the susceptibility to CHD, and C161T polymorphism might have different effects on CHD and ACS. The current study provided further insight into the etiology of CHD, and might be helpful to identify those with higher risk of CHD. Larger and well-designed studies of different ethnic populations are warranted to confirm our findings.