Associations of the PON1 rs854560 polymorphism with plasma lipid levels: a meta-analysis

Background Previous studies have investigated the associations of paraoxonase 1 (PON1) rs854560 polymorphism with plasma lipid levels, but the results are inconclusive. This meta-analysis aimed to clarify the associations of the rs854560 polymorphism with plasma lipid levels. Methods A comprehensive search of the literature was carried out by using the databases which include Medline, Google Scholar, Web of Science, Embase, Cochrane Library, China National Knowledge Infrastructure (CNKI), Wanfang and VIP database up till August 2018. The pooled standardized mean difference (SMD) with 95% confidence interval (CI) was used to assess the differences in lipid levels between the genotypes. Begg’s funnel plots and Egger’s test were used to examine the publication bias. Results A total of 41 studies (22,844 subjects) were identified for the associations of rs854560 polymorphism with plasma lipid levels. The M carriers had lower levels of high-density lipoprotein Cholesterol (HDL-C) (SMD = − 0.15, 95% CI = − 0.23--0.07, P < 0.01) and apolipoprotein A-I (APOA1) (SMD = − 0.67, 95% CI = − 0.93--0.41, P < 0.01) than the non-carriers. Subgroup analysis by ethnicity revealed that the effect on HDL level was significant in Caucasians and the subjects of other ethnic origins. No publication bias was detected in this meta-analysis. Conclusions The meta-analysis suggests that the PON1 rs854560 polymorphism is associated with a lower HDL-C level in Caucasians and subjects of other ethnic origins. Electronic supplementary material The online version of this article (10.1186/s12944-018-0924-0) contains supplementary material, which is available to authorized users.


Introduction
Coronary heart disease (CHD) is the leading cause of death in China and most of the developed countries [1]. While the interactions between genetic and environmental factors determine the pathogenesis of this disease, the report shows that dyslipidemia ranks as one of the most important risk factors accounting for at least 50% of the population attributable risk [2]. Dyslipidemia is characterized by elevated levels of triglycerides (TG), total cholesterol (TC) and low-density lipoprotein cholesterol (LDL-C) as well as reduced levels of high-density lipoprotein cholesterol (HDL-C) in circulation. Over the years, several investigations have been carried out on the genetic polymorphism/mutation that affects plasma lipid levels. The results, however, have been inconclusive primarily due to small sample size, ethnicity and difference in health conditions. High-density lipoprotein (HDL) has been shown to play a critical role in reverse cholesterol transport (RCT) by eliciting cholesterol efflux from macrophage foam cells which prevents the progression of atherosclerotic lesions and induces the regression of existing plaques [3]. HDL-C has also been primarily associated with a protein, apolipoprotein A-I (APOA1) and the over-expression of APOA1 has equally been found to reduce atherosclerosis in mice. Along the same line, Paraoxonase 1 (PON1), an enzyme that hydrolyses aryl esters, phosphate esters and lactones has been shown to be associated with apoA-1 and HDL-C. According to Akbas et al. [4] HDL-associated PON1 is considered to be a major anti-atherosclerosis component of HDL, as it inhibits the oxidation of low-density lipoprotein (LDL) and promotes cholesterol efflux from macrophage foam cells [4][5][6].
The PON1 gene is located on the long arm of human chromosome 7 (7q21-22), containing 9 exons and 8 introns. Although some studies have demonstrated that PON1 gene rs854560 polymorphism was significantly associated with CHD [7][8][9], whether this polymorphism is associated with dyslipidemia remains to be examined. A number of researches have investigated the associations of this polymorphism with plasma lipid levels, but the results were inconsistent and inconclusive. In some of these studies, the M allele of the rs854560 polymorphism was reported to be significantly associated with higher levels of TG [10][11][12], TC [13,14], LDL-C [13][14][15], and lower levels of HDL-C [14,16] and apoA-I [16] while reports from other studies differ [17][18][19][20][21]. Hence, we conducted this meta-analysis to clarify the associations of the PON1 rs854560 polymorphism and the different plasma lipid levels by using a larger sample size and to put into consideration ethnicity and disease condition particularly atherosclerosis.

Literature search
A comprehensive search of the literature was carried out by using databases which include Medline, Google Scholar, Web of Science, Embase, Cochrane Library, China National Knowledge Infrastructure (CNKI), Wanfang and VIP database (up till August 2018). The terms "paraoxonase 1" or "PON1"; "polymorphism" or "mutation" or "variant" or "SNP" or "rs854560" or "L55 M"; "blood lipid" or "serum lipid" or "lipids" were used for the search. The variables were limited to TG, TC, LDL-C, HDL-C and APOA1. The languages of the articles were limited to English and Chinese. All references cited in the included articles were reviewed to check for published works that were not indexed by Medline, Google Scholar, Web of Science, Embase, Cochrane Library, CNKI, Wanfang and VIP database.

Inclusion criteria
Studies that fulfill the following criteria were included: (1) studies in which the mean serum lipid values and standard deviations (SD) or standard errors (SE) by the rs854560 genotypes were available; (2) data reported on at least one of the five variables (TG, TC, LDL-C, HDL-C and APOA1); (3) data reported on fasting lipid variables; (4) pre-intervention baseline data that were used for interventional studies. Reports with incomplete data, studies based on pedigree data, case reports, review articles, abstracts and animal studies were excluded from the meta-analysis.

Data extraction
The studies that do not meet the inclusion criteria were excluded after being reviewed independently by two reviewers. All data were double-checked, compared after extraction and disagreements between reviewers were discussed and resolved. From each paper, the following information was collected: first author's name, year of publication, average age, country, ethnicity, gender, health condition and the mean of serum lipid levels and SD by genotypes. If a paper's data were unconvincing, we tried to contact the correspondent author       by e-mail. All the information was extracted using a standardized data collection form.

Data analysis
All statistical tests were two-sided and conducted by the STATA software package (Version 10, Stata Corporation, College Station, TX). P-value smaller than 0.05 for any test or model was considered to be statistically significant. Standardized mean difference (SMD) with 95% confidence interval (CI) was used for the meta-analysis. A fixed effect model (a Mantel-Haenszel method) was used to evaluate the results when heterogeneity among studies investigated by Cochrane Q statistic was not significant (I 2 ≤ 50%, P > 0.05). Otherwise, the random effect model (DerSimonian and Laird) was used [22]. Where there was significant heterogeneity among studies, we performed the Galbraith plot to detect potential sources of heterogeneity.
Since the results of most of the included studies were reported in a dominant way (LM + MM vs LL), a dominant model was employed to ensure adequate statistical power. Subgroup analyses were performed by ethnicity, gender and health condition. The ethnic subgroup was defined as Caucasian, Asian, and subjects of other ethnic origins. Health condition subgroup was defined as CHD patients, type 2 diabetes mellitus (T2DM) patients and healthy/control subjects. When data were presented for more than one subpopulation (e.g., female or male subjects, the subjects with CHD or T2DM, the subjects from different ethnicities) in one article, each subpopulation was treated as a separate comparison. Hardy-Weinberg equilibrium (HWE) was assessed by Fisher's exact test and a P-value < 0.05 was considered statistically significant. Possible publication bias was tested by Begg's funnel plots and Egger's test using P < 0.05 to indicate the presence of potential publication bias.

Result
Selection and characteristics of studies   The references for the studies included in the present meta-analysis are listed in Additional file 1. The characteristics of the studies included in the lipid association analysis are summarized in Additional file 2: Table S1. The plasma lipid levels by the genotypes of the rs854560 polymorphism are presented in Additional file 2: Table S2. From the subgroup analyses stratified by the characteristics of the subjects, significant associations of the rs854560 polymorphism with lower levels of HDL-C (SMD = − 0.11, 95% CI = − 0.20--0.02, P = 0.02) and higher levels of LDL-C (SMD = 0.06, 95% CI = 0.01-0.12, P = 0.03) were detected in Caucasians. The rs854560 polymorphism was significantly associated with lower levels of HDL-C (SMD = − 0.32, 95% CI = − 0.52--0.13, P < 0.01) and APOA1 (SMD = − 4.46, 95% CI = − 6.69--2.24, P < 0.01) in the subjects of other ethnic origins. The rs854560 polymorphism was also significantly associated with higher levels of TG (SMD = 0.22, 95% CI = 0.01-0.41, P = 0.05) in Asians. When health status was taken into account, the significant association of the rs854560 polymorphism with lower levels of HDL-C was detected in the T2DM patients, the healthy/control subjects and the case-control subjects (Table 1).

Publication bias test
The Begg's funnel plot and Egger's test were used to evaluate the publication bias in the literature. In the present study, Begg's funnel plot showed no publication bias, and this was confirmed by Egger's test (P = 0.43 for TG, 0.57 for TC, 0.56 for LDL-C, 0.98 for HDL-C, and 0.16 for APOA1).

Discussion
To the best of our knowledge, this is the first time that the associations of the PON1 rs854560 polymorphism with serum lipid levels are investigated. The present meta-analysis suggests that M allele of the rs854560 polymorphism is   associated with lower levels of HDL-C and APOA1 in the total population. A number of case-control studies [7][8][9] demonstrated that M allele of the rs854560 polymorphism had a promoting role for CHD risk. In combination with our findings, it is possible that the association of rs854560 polymorphism with a higher risk of CHD is mediated by the decreased levels of HDL-C and APOA1 caused by M allele of the rs854560 polymorphism. Subgroup analyses by ethnicity, gender and health condition were performed since they might be important variables in determining associative risk with lipid levels. For example, the present meta-analysis indicated that ethnicity might modulate the associations of the rs854560 polymorphism with HDL-C levels since the strong significant associations only existed in Caucasians (Table 1). A recent meta-analysis [23] revealed that M carriers of the rs854560 polymorphism had a higher risk of CHD than the non-carriers in the populations involved Caucasians. In combination with our findings, it is possible that the association between M allele of the rs854560 polymorphism and a higher risk of CHD in Caucasians was mediated by decreased HDL-C and APOA1 levels.
The PON1 rs854560 polymorphism may result in markedly reduced levels of HDL-C and APOA1 by affecting PON1 activity [11]. In 2014, Abessolo et al. performed a study on T2DM patients to assess the relationships between rs854560 polymorphism, PON1 activity and plasma lipid levels [16]. The results showed that the PON1 rs854560 polymorphism was significantly associated with decreased levels of HDL-C and APOA1 possibly via decreased serum PON1 activity. One possible explanation for this association is that the reduced PON1 enzyme activity might reduce the capacity of PON1-mediated inhibition of LDL oxidation [24,25], which leads to increased levels of plasma ox-LDL (oxidized low-density lipoprotein) [26] and decreased levels of plasma HDL-C [27] and APOA1 [28]. This may also explain our findings.   Fig. 6 Forest plot of the meta-analysis between the PON1 rs854560 polymorphism and plasma low-density lipoprotein cholesterol (LDL-C) levels