An Updated Systematic Review and Meta-analysis of Association Between Adiponectin Gene Polymorphisms and Coronary Artery Disease

Abstract

Coronary artery disease (CAD) is a significant contributor to global health burden. Adiponectin gene single nucleotide polymorphisms (SNPs) have been associated with CAD susceptibility, but with inconsistent results across the studies. We present, in this study, an updated meta-analysis to discern the genetic susceptibility of adiponectin SNPs in relation to CAD. PubMed and EMBASE databases were used to identify the relevant published articles using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Pooled odds ratios and 95% confidence intervals were generated to assess the strength of the associations. Thirty-five articles with a total of 28,947 participants (mean age 55.3 years, 11,632 cases/17,315 controls, 19,443 males/8353 females, and 1151 persons with unspecified gender data) were included. The dominant, recessive, and additive models were applied. We found that the SNPs +45T>G (rs2241766), −4034A>C (rs822395), and −11391G>A (rs17300539) were linked to CAD development. In addition, +276G>T (rs1501299) SNP was associated with a decreased susceptibility to CAD among Caucasians. We did not find an association between the CAD susceptibility and the −11377C>G (rs266729) SNP. These observations offer new potential genetic biomarker candidates in relation to CAD, and warrant further research in independent world populations.

Introduction

Coronary artery disease (CAD) is one of the common causes of mortality and morbidity worldwide (Vedanthan et al., 2014). As a multifactorial disease, CAD results from various causalities of environmental exposures and genetic factors. Adiponectin, an adipocytokine composed of 244 amino acids, also named as ADIPOQ, is specifically secreted by adipose tissue (Ouchi et al., 2004; Yang et al., 2012). As an insulin-sensitive hormone, adiponectin regulates glucose and lipid metabolism (Francke et al., 2001; Gnacińska et al., 2009). Circulating adiponectin is reported to inhibit expression of vascular inflammatory molecules (Karaduman et al., 2006). Moreover, adiponectin downregulates the proliferation of vascular smooth muscle cells, and maintains destabilization of atherosclerotic plaque (Wassel et al., 2011).

Adiponectin is therefore considered as a preventive factor for CAD (Vaiopoulos et al., 2012; Zhang et al., 2012). In 2009, a genome-wide association study demonstrated that serum adiponectin level was associated with the single nucleotide polymorphisms (SNPs) of adiponectin (3q27) (Richards et al., 2009). Since then, more than 30 original studies and meta-analyses have demonstrated a correlation between adiponectin variants and CAD among different populations (Shen et al., 2015). However, the conclusion so far remains inconsistent.

To further confirm the relationship between adiponectin SNPs and CAD, we conducted the present systematic review and meta-analysis in a large sample of 28,947 participants, focusing on the following five SNPs at adiponectin: +45T>G (rs2241766), +276G>T (rs1501299), −4034A>C (rs822395), −11377C>G (rs266729), and −11391G>A (rs17300539), due to their susceptibilities to CAD and existence of enough publications. The dominant, recessive, and additive models were applied, and the comparisons of allelic frequencies were further included in this study.

Materials and Methods

Search strategy and inclusion criteria

We followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, as utilized in previous systematic reviews published in the Journal (Ameh et al., 2017). The PRISMA Checklist was presented in Supplementary Table S1.

PubMed and EMBASE databases were used to search for eligible studies published before October 31, 2016. The search strategy was based on a combination of the following terms: “coronary artery disease” or “CAD” or “coronary heart disease” or “CHD” or “ischemic heart disease” or “IHD” or “myocardial infarction” or “MI,” “adiponectin gene” or “ADIPOQ” and “polymorphism” or “variant” or “mutation.” Additionally, publications cited in the retrieved articles were screened.

Two authors (H.H. and L.Z.) independently reviewed all retrieved publications and citations, then evaluated the quality of each study with the predefined scale for quality assessment (Supplementary Table S2), which was designed in strict accordance with the scoring form of meta-analysis (Thakkinstian et al., 2005). The inclusion criteria for eligible studies were as follows: (1) case–control studies; (2) genotype distributions and source of participants are available; (3) as to the overlapping reports, the recent data or larger sample result were collected; and (4) English language publications.

Data extraction

Original data were independently extracted by H.H. and L.Z. using standardized data extraction forms. Contested issues in data extraction were resolved by the third colleague (Z.S.). The following information was retrieved from the included studies: author names, the year of publication, the characteristics of participants, method of genotyping, number of individuals, distribution information of genotype, and allelic frequency.

Statistical analyses

All statistical analyses were performed with Stata 14.0 (StataCorp, College Station, TX). The pooled odds ratios (ORs) and 95% confidence intervals (CIs) were used to quantitatively evaluate the relationship between adiponectin SNPs and CAD risk under dominant, additive, and recessive models, and also for allelic frequency contrast. Subgroup analyses were performed based on ethnicity (i.e., Arabian, African, Asian, Caucasian, and South American) and disease background (i.e., CAD, myocardial infarction [MI], and type 2 diabetes mellitus [T2DM]). Z test was applied to identify the statistical significance of pooled ORs. Hardy–Weinberg equilibrium in the control group was assessed with chi-square test (χ²-test).

Cochran's Q-test and I² statistic were used to evaluate heterogeneity across included studies. When p < 0.10 or I² > 50%, the heterogeneity was considered significant, then a fixed-effects meta-analysis was performed. Otherwise, the random-effects meta-analysis was applied. Galbraith plot analysis was utilized to determine the outliers across involved studies. In this analysis, the studies within limit area were considered homogeneous, whereas the ones beyond limit area were outliers. The stability of result in each meta-analysis was detected with sensitivity analysis by removing each individual study sequentially or by omitting the outliers. The significance of publication bias was estimated by funnel plots analysis, and Egger's test was further complemented. A p-value <0.05 was considered representation of statistical significance.

Results

Eligible studies

The initial search returned 191 articles. Among them 156 publications were excluded for the following reasons: duplicated data, studies of animals, investigating irrelevant polymorphisms, not referring to CAD, not original studies, and no essential data. As outlined in Figure 1 and Table 1, we included 35 articles (Al-Daghri et al., 2011; Antonopoulos et al., 2013; Bacci et al., 2004; Boumaiza et al., 2011; Chang et al., 2009; Cheung et al., 2014; Chiodini et al., 2010; De Caterina et al., 2011; Du et al., 2016; Esteghamati et al., 2012; Filippi et al., 2005; Foucan et al., 2010; Gable et al., 2007; Gui et al., 2012; Hegener et al., 2006; Hoefle et al., 2007; Jung et al., 2006; Katakami et al., 2012; Lacquemant et al., 2004; Mofarrah et al., 2016; Mohammadzadeh et al., 2016; Oguri et al., 2009; Ohashi et al., 2004; Oliveira et al., 2012; Persson et al., 2010; Pischon et al., 2007; Prior et al., 2009, 2011; Qi et al., 2005; Rizk et al., 2012; Rodriguez-Rodriguez et al., 2011; Sabouri et al., 2011; Zhang et al., 2016; Zhong et al., 2010; and Zhou et al., 2011). A total of 28,947 participants (mean age of 55.3 years, 11,632 cases/17,315 controls, 19,443 males/8353 females, and 1151 participants with unspecified gender data) were involved in the present study. We enrolled 11,720 more participants than previously published meta-analyses on this related topic.

FIG. 1.

Flow chart for the process of searching and including studies.

Table 1.

Characteristics of Included Studies in the Meta-Analysis.

							Cases			Controls
Author	Year	Country	Ethnicity	Diseases	Baseline status	Source of controls	n	Age	Gender (male)	n	Age	Gender (male)
Al-Daghri	2011	Saudi Arabia	Arabian	CAD	T2DM	Hospital-based	123	69.4 ± 9.8	74	298	60.7 ± 11.6	207
Antonopoulos	2012	Greece	Caucasian	CAD	NA	Hospital-based	462	64.9 ± 0.4	395	132	63.4 ± 0.7	109
Bacci	2004	Italy	Caucasian	CAD	NA	Hospital-based	142	64 ± 8	91	234	60 ± 8	101
Boumaiza	2011	Tunisia	Arabian	CAD	NA	Hospital-based	212	60.6 ± 10.6	141	104	59.4 ± 11.9	58
Chang	2009	China	Asian	CAD	NA	Hospital-based	600	63.7 ± 12.1	470	718	51.1 ± 10.4	383
Cheung	2014	China	Asian	CAD	NA	Population-based	184	56.3 ± 10.9	111	2012	44.9 ± 11.7	915
Chiodini	2010	Italy	Caucasian	MI	NA	Hospital-based	503	56.5 ± 4.5	449	503	54.7 ± 5.2	482
De Caterina	2011	Italy	Caucasian	MI	NA	Population-based	1864	39.6 ± 4.8	1655	1864	39.7 ± 4.9	1655
Du	2016	China	Asian	CAD	NFL	Hospital-based	246	61.5 ± 10.3	127	304	61.3 ± 9.4	126
Esteghamati	2012	Iran	Arabian	CAD	T2DM	Hospital-based	114	61.1 ± 10.0	44	117	51.1 ± 9.0	71
Filippi	2005	Italy	Caucasian	CAD	NA	Population-based	580	60.3 ± 10.0	446	466	50.9 ± 14.4	231
Foucan	2010	Caribbean	African	CAD	T2DM	Hospital-based	57	68 ± 10	29	159	63 ± 12	57
Gable	2007	United Kingdom	Caucasian	MI	NA	Population-based	530	56.6 ± 3.6	530	568	56.0 ± 3.4	568
Gui	2012	China	Asian	CAD	NA	Hospital-based	438	60 ± 10	321	443	60 ± 10	299
Hegener	2006	United States	Caucasian	MI	NA	Population-based	341	60.1 ± 0.5	341	341	60.2 ± 0.5	341
Hoefle	2007	Austria	Caucasian	CAD	NA	Hospital-based	277	61 ± 10	277	125	61 ± 10	125
Jung	2006	Korea	Asian	CAD	NA	Hospital-based	88	60.4 ± 10.2	63	68	53.4 ± 11.2	34
Katakami	2012	Japan	Asian	CAD	T2DM	Hospital-based	213	58.1 ± 6.0	141	2424	54.4 ± 8.0	1471
Lacquemant	2004	France & Switzerland	Caucasian	CAD	T2DM	Hospital-based	162	—	—	315	—	—
Mofarrah	2016	Iran	Arabian	CAD	T2DM	Hospital-based	152	59.2 ± 8.9	95	72	55.6 ± 7.5	19
Mohammadzadeh	2016	Iran	Arabian	CAD	T2DM	Hospital-based	100	57.3 ± 7.5	52	100	52.2 ± 9.4	41
Oguri	2009	Japan	Asian	CAD	T2DM	Hospital-based	773	64.8 ± 10.0	597	1114	68.3 ± 8.9	567
Ohashi	2004	Japan	Asian	CAD	NA	Hospital-based	383	63.0 ± 0.4	270	368	62.3 ± 0.6	240
Oliveira	2012	Brazil	South American	CAD	T2DM	Hospital-based	450	61 ± 10	275	153	56 ± 10	67
Persson	2010	Sweden	Caucasian	MI	NA	Hospital-based	244	54 (49–57)	216	244	54(49–57)	216
Pischon	2007	United States	Caucasian	CAD	NA	Hospital-based	497	65.2 ± 8.3	266	989	65.2 ± 8.3	532
Prior	2009	United Kingdom	Caucasian	CAD	T2DM	Hospital-based	155	68.6 ± 10.2	104	611	61.1 ± 14.0	361
Prior	2011	United Kingdom	Caucasian	CAD	T1DM	Hospital-based	85	71.0 ± 7.6	46	298	68.2 ± 9.2	128
Qi	2005	United States	Caucasian	CAD	T2DM	Hospital-based	239	59.6 ± 7.2	239	640	55.0 ± 8.6	640
Rizk	2012	Qatar	Arabian	CAD	ACS	Hospital-based	142	53.5 (51.0–54.6)	131	122	35.0 (26.4–33.6)	106
Rodríguez-Rodríguez	2011	Spain	Caucasian	CAD	RA	Hospital-based	119	—	—	555	—	—
Sabouri	2011	Iran	Arabian	CAD	NA	Hospital-based	329	58.4 ± 10.6	211	106	47.6 ± 9.4	60
Zhang	2016	China	Asian	CAD	NA	Hospital-based	306	45.0 ± 6.1	227	412	46.7 ± 6.3	160
Zhong	2010	China	Asian	CAD	NA	Hospital-based	198	60.6 ± 10.1	107	237	54.5 ± 10.2	109
Zhou	2011	China	Asian	CAD	NA	Hospital-based	358	58.9 ± 5.4	358	65	48.9 ± 4.3	65

ACS, acute coronary syndrome; CAD, coronary artery disease; MI, myocardial infarction; RA, rheumatoid arthritis; NFL, nonalcoholic fatty liver disease; T2DM, type 2 diabetes mellitus; T1DM, type 1 diabetes mellitus; NA, no reported baseline disease.

The detailed information of ethnicity of the subjects, source of individuals, and genotype frequency were listed in Table 1 and Supplementary Table S3. Among the 28,947 participants in the 35 studies, 8536 (in 11 studies) were CAD patients with concomitant T2DM, 550 (in 1 study) were CAD patients with nonalcoholic fatty liver disease, 674 (in 1 study) were CAD patients with rheumatoid arthritis, and 383 (in 1 study) were CAD patients with T1DM. For ethnic group distributions, 2306 (in 8 studies) were Arabians, 11,906 (in 11 studies) were Asians, 14,132 (in 15 studies) were Caucasians, and 603 (in 1 study) were South Americans.

Associations between +45T>G polymorphism and CAD

We included 23 studies published on the relationship between +45T>G polymorphism and CAD risk. Significant associations were found between +45T>G polymorphism and CAD under both allelic frequency contrast (OR = 1.22, 95% CI = 1.06–1.41, p = 0.007) and dominant model analysis (OR = 1.21, 95% CI = 1.04–1.39, p = 0.011). On the contrary, no significant associations were found under additive model (OR = 1.21, 95% CI = 0.88–1.67, p = 0.242) or recessive model (OR = 1.16, 95% CI = 0.86–1.56, p = 0.344) (Fig. 2 and Table 2).

FIG. 2.

Forest plot of the association between +45T>G (rs2241766) polymorphism and CAD under allelic frequency contrast. Each line represents one original study. The gray box shows the weight for each study. The black diamond shows the pooled OR and 95% CI. 95% CI, 95% confidence interval; CAD, coronary artery disease; OR, odds ratio.

To eliminate the heterogeneity across the included studies, we conducted subgroup analyses based on ethnic groups. Consequently, significant associations were observed among the Arabian populations, with the ORs (95% CIs) of 1.65 (1.12–2.42) for allelic frequency contrast, 2.35 (1.26–4.39) for additive model, 1.56 (1.03–2.35) for dominant model, and 2.29 (1.30–4.03) for recessive model analysis (Table 3).

Table 2.

Associations Between Adiponectin Gene Polymorphisms and Coronary Artery Disease

SNPs	Genetic comparison	No. of studies	No. of subjects	I² (%)	Q (p)	Statistical model	OR [95% CI]	Z (p)
+45T>G (rs2241766)	Allelic frequency contrast	23	13,933	75.4	85.49 (0.000)	Ran	1.22 [1.06, 1.41]	2.72 (0.007)
	Additive model	23	13,933	61.3	54.25 (0.000)	Ran	1.21 [0.88, 1.67]	1.17 (0.242)
	Dominant model	23	14,812	66.8	66.17 (0.000)	Ran	1.21 [1.04, 1.39]	2.54 (0.011)
	Recessive model	23	13,933	57.1	48.92 (0.001)	Ran	1.16 [0.86, 1.56]	0.95 (0.344)
+276G>T (rs1501299)	Allelic frequency contrast	22	17,533	68.2	66.02 (0.000)	Ran	0.97 [0.88, 1.07]	0.65 (0.513)
	Additive model	22	17,533	58.9	51.05 (0.000)	Ran	0.87 [0.71, 1.07]	1.29 (0.196)
	Dominant model	22	17,533	64.9	59.79 (0.000)	Ran	0.99 [0.88, 1.12]	0.09 (0.930)
	Recessive model	22	17,533	50.0	42.03 (0.004)	Ran	0.87 [0.72, 1.04]	1.53 (0.125)
−4034A>C (rs822395)	Allelic frequency contrast	7	9911	4.1	6.26 (0.395)	Fix	1.04 [0.97, 1.11]	0.99 (0.324)
	Additive model	7	9911	40.5	10.08 (0.121)	Fix	1.19 [1.01, 1.40]	2.05 (0.040)
	Dominant model	7	9911	0	4.42 (0.620)	Fix	1.00 [0.91, 1.09]	0.01 (0.989)
	Recessive model	7	9911	49.9	11.97 (0.063)	Ran	1.24 [0.92, 1.67]	1.42 (0.157)
−11377C>G (rs266729)	Allelic frequency contrast	16	16,965	68.7	47.90 (0.000)	Ran	1.10 [0.99, 1.22]	1.69 (0.091)
	Additive model	16	16,965	49.4	29.65 (0.013)	Ran	1.19 [0.95, 1.49]	1.53 (0.126)
	Dominant model	16	16,965	64.7	42.54 (0.000)	Ran	1.16 [0.99, 1.26]	1.77 (0.077)
	Recessive model	16	16,965	38.4	24.35 (0.055)	Ran	1.15 [0.94, 1.40]	1.36 (0.173)
−11391G>A (rs17300539)	Allelic frequency contrast	4	3089	0	1.96 (0.58)	Fix	0.96 [0.80, 1.16]	0.41 (0.685)
	Additive model	4	3089	0	0.50 (0.919)	Fix	2.49 [1.05, 5.92]	2.06 (0.039)
	Dominant model	4	3089	0	1.71 (0.635)	Fix	0.90 [0.74, 1.10]	1.00 (0.319)
	Recessive model	4	3089	0	0.47 (0.924)	Fix	2.56 [1.07, 6.08]	2.12 (0.034)

95% CI, 95% confidence interval; Fix, fixed-effects meta-analysis model; OR, odds ratio; Ran, random-effects meta-analysis model; SNP, single nucleotide polymorphism.

Table 3.

Subgroup Analyses Based on the Accompanied Disease and Ethnicity of Subjects

		Population			Disease
SNPs	Genetic comparison	Asian	Caucasian	Arabian	CAD	MI	T2DM
+45T>G (rs2241766)	Allelic frequency contrast	1.12 (0.88–1.42)	1.01 (0.89–1.15)	1.65 (1.12–2.42)^*	1.10 (0.92–1.32)	1.00 (0.87–1.15)	1.60 (1.14–2.25)^*
	Additive model	1.07 (0.68–1.67)	0.72 (0.50–1.05)	2.35 (1.26–4.39)^*	0.97 (0.69–1.36)	0.71 (0.42–1.23)	2.48 (1.23–4.98)^*
	Dominant model	1.18 (0.88–1.58)	1.03 (0.89–1.19)	1.56 (1.03–2.35)^*	1.1 (0.92–1.40)	1.04 (0.88–1.22)	1.42 (1.01–1.98)^*
	Recessive model	0.96 (0.69–1.35)	0.71 (0.49–1.03)	2.29 (1.30–4.03)^*	0.96 (0.69–1.32)	0.70 (0.40–1.23)	2.27 (1.17–4.42)^*
+276G>T (rs1501299)	Allelic frequency contrast	1.06 (0.84–1.35)	0.95 (0.86–1.05)	0.92 (0.66–1.29)	1.04 (0.91–1.19)	0.90 (0.78–1.02)	0.92 (0.74–1.14)
	Additive model	1.18 (0.81–1.73)	0.77 (0.59–1.01)	0.85 (0.47–1.51)	1.02 (0.78–1.35)	0.73 (0.43–1.24)	0.75 (0.52–1.09)
	Dominant model	1.07 (0.79–1.45)	1.00 (0.88–1.12)	0.92 (0.61–1.39)	1.08 (0.91–1.28)	0.92 (0.80–1.07)	0.94 (0.71–1.24)
	Recessive model	1.16 (0.90–1.51)	0.76 (0.58–0.99)^*	0.87 (0.55–1.39)	1.00 (0.78–1.26)	0.74 (0.43–1.27)	0.76 (0.57–1.03)
−4034A>C (rs822395)	Allelic frequency contrast	0.94 (0.78–1.13)	1.05 (0.98–1.14)	—	1.04 (0.95–1.14)	—	1.02 (0.98–1.08)
	Additive model	0.97 (0.48–1.94)	1.20 (1.02–1.43)^*	—	1.17 (1.11–1.95)^*	—	1.01 (0.74–1.36)
	Dominant model	0.92 (0.75–1.13)	1.02 (0.92–1.13)	—	0.98 (0.90–1.07)	—	0.97 (0.87–1.08)
	Recessive model	1.07 (0.33–3.42)	1.25 (0.92–1.71)	—	1.60 (1.19–2.13)	—	1.05 (0.73–1.51)
−11377C>G (rs266729)	Allelic frequency contrast	1.22 (0.93–1.59)	1.04 (0.95–1.14)	—	1.07 (0.87–1.32)	1.06 (0.96–1.17)	1.09 (0.88–1.34)
	Additive model	1.53 (0.97–2.40)	1.05 (0.84–1.32)	—	1.24 (0.75–2.05)	1.09 (0.89–1.33)	1.22 (0.95–1.49)
	Dominant model	1.24 (0.90–1.71)	1.05 (0.95–1.17)	—	1.05 (0.84–1.31)	1.07 (0.93–1.23)	1.10 (0.99–1.26)
	Recessive model	1.42 (0.98–2.06)	1.04 (0.84–1.28)	—	1.20 (0.76–1.90)	1.07 (0.88–1.31)	1.21 (0.94–1.40)
−11391G>A (rs17300539)	Allelic frequency contrast	—	0.95 (0.78–1.16)	—	—	0.94 (0.78–1.14)	—
	Additive model	—	2.34 (0.94–5.81)	—	—	2.02 (0.69–5.90)	—
	Dominant model	—	0.90 (0.73–1.11)	—	—	0.91 (0.76–1.10)	—
	Recessive model	—	2.41 (0.97–5.97)	—	—	2.08 (0.71–6.08)	—

CAD, cases were patients of coronary artery disease without accompanied disease; MI, cases were patients of myocardial infarction without accompanied disease; T2DM, the baseline status of subject was type 2 diabetes mellitus.

The associations were statistically significant (p < 0.05).

Furthermore, we performed subgroup analyses based on the disease background. The results showed that +45T>G polymorphism was significantly associated with CAD among T2DM patients under allelic frequency contrast (OR = 1.60, 95% CI = 1.14–2.25), and under additive (OR = 2.48, 95% CI = 1.23–4.98), dominant (OR = 1.42, 95% CI = 1.01–1.98), and recessive (OR = 2.27, 95% CI = 1.17–4.42) models, respectively (Table 3).

Associations between +276G>T polymorphism and CAD

We combined 22 eligible studies to analyze the associations between +276G>T polymorphism and CAD. Consequently, no significant associations were found under allelic frequency contrast (OR = 0.97, 95% CI = 0.88–1.07, p = 0.513), additive (OR = 0.87, 95% CI = 0.71–1.07, p = 0.193), dominant (OR = 0.99, 95% CI = 0.88–1.12, p = 0.930), and recessive models (OR = 0.87, 95% CI = 0.72–1.04, p = 0.125) (Fig. 3 and Table 2). Moreover, subgroup analyses did not return any significant association for other populations, except for Caucasians under the recessive model analysis (OR = 0.76, 95% CI = 0.58–0.99).

FIG. 3.

Forest plot of the association between +276G>T (rs1501299) polymorphism and CAD under recessive model analysis. Each line represents one original study. The gray box shows the weight for each study. The black diamond shows the pooled OR and 95% CI.

Associations between −4034A>C polymorphism and CAD

Seven studies matched the designated inclusion criteria for estimating the relationship between −4034A>C polymorphism and CAD. The combining results from these studies demonstrated a significant association under the additive model analysis, with the pooled OR of 1.19 (95% CI = 1.01–1.40, p = 0.040) (Fig. 4 and Table 2). Meanwhile, heterogeneity test across the seven studies showed no statistical significance. Subgroup analyses elucidated that −4034A>C polymorphism was significantly associated with CAD among Caucasians under the additive model (OR = 1.20, 95% CI = 1.02–1.43) (Table 3).

FIG. 4.

Forest plot of the association between −4034A>C (rs822395) polymorphism and CAD under additive model analysis. Each line represents one original study. The gray box shows the weight for each study. The black diamond shows the pooled OR and 95% CI.

Associations between −11377C>G polymorphism and CAD

Sixteen studies on the association between the −11377C>G polymorphism and CAD were enrolled in this meta-analysis (Fig. 5 and Tables 2 and 3). No significant associations were found under additive (OR = 1.19, 95% CI = 0.95–1.49, p = 0.126), dominant (OR = 1.12, 95% CI = 0.99–1.26, p = 0.077), recessive (OR = 1.15, 95% CI = 0.94–1.40, p = 0.173) model analyses, and allelic frequency contrast (OR = 1.09, 95% CI = 0.99–1.22, p = 0.091). Further subgroup analyses indicated no significant association either (Table 3).

FIG. 5.

Forest plot of the association between −11377C>G (rs266729) polymorphism and CAD under allelic frequency contrast. Each line represents one original study. The gray box shows the weight for each study. The black diamond shows the pooled OR and 95% CI.

Associations between −11391G>A polymorphism and CAD

The synthesized results were statistically significant under additive model (OR = 2.14, 95% CI = 1.05–5.92, p = 0.039) (Fig. 6) and recessive model (OR = 2.56, 95% CI = 1.07–6.08, p = 0.034). However, the allelic frequency contrast and dominant model analysis did not return significant results (Table 2), and no significant heterogeneity was observed across four studies included.

FIG. 6.

Forest plot of the association between −11391G>A (rs17300539) polymorphism and CAD under additive model analysis. Each line represents one original study. The gray box shows the weight for each study. The black diamond shows the pooled OR and 95% CI.

Publication bias and sensitivity analysis

Funnel plot analysis, followed by Egger's test was implemented to detect potential publication bias. As shown in Supplementary Figure S1, the statistically significant publication bias was found in +45T>G polymorphism analyses.

Galbraith plot analysis was carried out to identify the outliers across eligible studies. Two studies on −11377C>G (Du et al., 2016; Hoefle et al., 2007), three studies on +276G>T (Filippi et al., 2005; Gui et al., 2012; Mohammadzadeh et al., 2016), and six studies on +45T>G (Al-Daghri et al., 2011; Cheung et al., 2014; Esteghamati et al., 2012; Foucan et al., 2010; Mofarrah et al., 2016; Rizk et al., 2012) were considered as outliers (Supplementary Figs. S2–S4).

Furthermore, we conducted sensitivity analysis by sequential removal of each study and excluding studies exhibited as outliers in Galbraith plot analysis. No significant change was found in meta-analyses of +45T>G, +276G>T, and −11377C>G polymorphisms. However, the results of −4034A>C polymorphism changed significantly when we excluded two studies conducted among the Caucasians (De Caterina et al., 2011; Pischon et al., 2007) (Supplementary Fig. S5).

Discussion

Adiponectin is expressed in adipocytes (Hoefle et al., 2007), and regulates the metabolism of glucose, energy storage, and fatty acid (Nakano et al., 1996; Sun et al., 2012). Although previous studies have extensively investigated an association between the five common adiponectin SNPs (+45T>G, +276G>T, −4034A>C, −11377C>G, −11391G>A) and CAD, the past literature was not uniformly consistent.

Based on combined analyses of 28,947 participants, our study suggested that +45T>G, −4034A>C, and −11391G>A polymorphisms are associated with CAD risk, +276G>T polymorphism is associated with a decreased risk of CAD among Caucasians, and −11377C>G polymorphisms are not related to CAD susceptibility.

The +45T>G SNP is a common synonymous mutation located at exon 2 of adiponectin. It has been proven in clinical investigation that adiponectin mRNA expression is relatively higher in G allelic carriers (Chang et al., 2009). This SNP has also been reported to be associated with higher serum adiponectin level and insulin resistance (Mackawy, 2013). Our study indicated that +45T>G polymorphism is a susceptibility factor for CAD. Subgroup analyses illuminated that +45T>G polymorphism is associated with CAD risk among Arabians, as well as in T2DM patients. This can be explained with the strong association of +45T>G polymorphism with insulin resistance and obesity (Zhou et al., 2014). Although a previous meta-analysis (Zhou et al., 2012) had reported a protective association of +45T>G polymorphism with CAD, our study presented similar result with two meta-analyses based on Asians and Caucasians (Zhang et al., 2012; Zhou et al., 2014).

The +276G>T polymorphism, sitting at intron 2 of adiponectin is also reported to regulate both adiponectin level and insulin resistance, as well as reduce oxidative stress (Fumeron et al., 2004; Jang et al., 2005; Menzaghi et al., 2002). In the previous meta-analyses, +276G>T polymorphism was dominated to be a preventive factor for CAD in general populations (Zhou et al., 2014) and among peoples with T2DM background (Zhang et al., 2012). This study showed no significant association between +276G>T polymorphism and CAD, except for Caucasians when recessive model analysis was performed.

Also, −4034A>C SNP has been considered a candidate gene for CAD, and is related to an increased susceptibility of fatal CAD and non-fatal MI (Zhong et al., 2010). In this meta-analysis, the seven studies included in the additive model analysis showed that −4034A>C SNP is a risk factor for CAD.

The polymorphism of −11377C>G, located in the promoter region of adiponectin, has been related to obesity among healthy adult population (Park et al., 2011). Two meta-analyses indicated the significant association of −11377C>G SNP with an increased risk of cardiovascular disease (Zhang et al., 2012; Zhou et al., 2012). However, we did not observe any significant association in the present study, which was consistent to a study of ischemic stroke patients (Shen et al., 2015).

The −11391G>A SNP is another adiponectin gene variant located in the promoter region of the adiponectin gene, which decreases the risk of insulin resistance (Chiodini et al., 2010). Four eligible studies reported the association of −11391G>A SNP with CAD. Nonetheless, no meta-analysis referred to −11391G>A polymorphism with CAD risk. Our study suggested that −11391G>A polymorphism is a risk factor for CAD.

Regarding ethnic stratification, which is recognized as one of the most important covariate in meta-analysis, we found disagreements between subgroups of ethnicities. Among the Asian populations, the five adiponectin SNPs examined in this study were not associated with CAD. However, a significant association of CAD with +45T>G polymorphism among Arabians as well as a −4034A>C and +276G>T polymorphisms among Caucasians were found in this study. These findings may probably be explained by the differences in genetic backgrounds, environmental factors, and lifestyles of the populations (Zhao et al., 2016; Zhou et al., 2014). Besides, the genetic diversity is owing to the flip-flop phenomenon, a phenomenon resulting from an interlocus correlation with the variant at another locus through linkage disequilibrium, which has been reported to be a contributing factor (Zhang et al., 2011).

There are some limitations of our study which are noteworthy and are common in meta-analyses of genetic polymorphisms and disease risks. First, the searched database was limited to English language publications. Second, only four original studies were found in terms of −11391G>A polymorphism, thus the potential publication bias on this locus could not be excluded thoroughly.

Conclusions

The current meta-analysis suggests that the polymorphisms of +45T>G (under dominant model), −4034A>C (under additive model), and −11391G>A (under both additive model and recessive model) are susceptible SNPs to CAD. The polymorphism of +276G>T is associated with a reduced susceptibility for CAD among the Caucasians (under recessive model), but not in other populations. Meanwhile, −11377C>G polymorphism is not related to CAD risk. These observations offer new potential genetic biomarker candidates in relation to CAD and warrant further analyses in independent world populations.

Footnotes

Acknowledgments

This study was supported by the National Natural Science Foundation of China (Grant: NSFC, No. 81202170). The authors thank Enoch Odame Anto, School of Medical and Health Sciences, for his English editing on this article.

Author Disclosure Statement

The authors declare that no competing financial interests exist.

Abbreviations Used

References

Al-Daghri

, Al-Attas

, Alokail

, Alkharfy

, and Hussain

. (2011). Adiponectin gene variants and the risk of coronary artery disease in patients with type 2 diabetes. Mol Biol Rep, 38, 3703–3708.

Ameh

, Okpechi

, Dandara

, and Kengne

. (2017). Association between telomere length, chronic kidney disease, and renal traits. A Systematic Review. OMICS, 21, 143–155.

Antonopoulos

, Tousoulis

, Antoniades

, et al. (2013). Genetic variability on adiponectin gene affects myocardial infarction risk: The role of endothelial dysfunction. Int J Cardiol, 168, 326–330.

Bacci

, Menzaghi

, Ercolino

, et al. (2004). The +276 G/T single nucleotide polymorphism of the adiponectin gene is associated with coronary artery disease in type 2 diabetic patients. Diabetes Care, 27, 2015–2020.

Boumaiza

, Omezzine

, Rejeb

, et al. (2011). Single-nucleotide polymorphisms at the adiponectin locus and risk of coronary artery disease in Tunisian coronaries. J Cardiovasc Med (Hagerstown), 12, 619–624.

Chang

, Jiang

, et al. (2009). Interaction of ADIPOQ gene polymorphisms with blood pressure and plasma cholesterol level on the risk of coronary artery disease. Circ J, 73, 1934–1938.

Cheung

, Hui

, Cheung

, et al. (2014). Adiponectin gene variants and the risk of coronary heart disease: A 16 year longitudinal study. Eur J Endocrinol, 171, 107–115.

Chiodini

, Specchia

, Gori

, et al. (2010). Adiponectin gene polymorphisms and their effect on the risk of myocardial infarction and type 2 diabetes: An association study in an Italian population. Ther Adv Cardiovasc Dis, 4, 223–230.

De Caterina

, Talmud

, Merlini

, et al. (2011). Strong association of the APOA5 − 1131T>C gene variant and early-onset acute myocardial infarction. Atherosclerosis, 214, 397–403.

10.

, Lu

, Liu

, Dong

, Xuan

, and Xin

. (2016). Association of adiponectin gene polymorphisms with the risk of coronary artery disease in patients with nonalcoholic fatty liver disease in a Chinese Han population. Hepat Mon, 16, e37388.

11.

Esteghamati

, Mansournia

, Nakhjavani

, Mansournia

, Nikzamir

, and Abbasi

. (2012). Association of +45(T/G) and +276(G/T) polymorphisms in the adiponectin gene with coronary artery disease in a population of Iranian patients with type 2 diabetes. Mol Biol Rep, 39, 3791–3797.

12.

Filippi

, Sentinelli

, Romeo

, et al. (2005). The adiponectin gene SNP +276G> T associates with early-onset coronary artery disease and with lower levels of adiponectin in younger coronary artery disease patients (age <or = 50 years). J Mol Med (Berl), 83, 711–719.

13.

Foucan

, Ezourhi

, Maimaitiming

, et al. (2010). Adiponectin multimers and ADIPOQ T45G in coronary artery disease in Caribbean type 2 diabetic subjects of African descent. Obesity (Silver Spring), 18, 1466–1468.

14.

Francke

, Manraj

, Lacquemant

, et al. (2001). A genome-wide scan for coronary heart disease suggests in Indo-Mauritians a susceptibility locus on chromosome 16p13 and replicates linkage with the metabolic syndrome on 3q27. Hum Mol Genet, 10, 2751–2765.

15.

Fumeron

, Aubert

, Siddiq

, et al. (2004). Epidemiologic Data on the Insulin Resistance Syndrome (DESIR) Study Group. Adiponectin gene polymorphisms and adiponectin levels are independently associated with the development of hyperglycemia during a 3-year period: The epidemiologic data on the insulin resistance syndrome prospective study. Diabetes, 53, 1150–1157.

16.

Gable

, Matin

, Whittall

, et al. (2007). Common adiponectin gene variants show different effects on risk of cardiovascular disease and type 2 diabetes in European subjects. Ann Hum Genet, 71, 453–466.

17.

Gnacińska

, Małgorzewicz

, Stojek

, Łysiak-Szydłowska

, and Sworczak

. (2009). Role of adipokines in complications related to obesity: A review. Adv Med Sci, 54, 150–157.

18.

Gui

, Li

, Jiang

, Gao

, Lu

, and Gao

. (2012). Association of the adiponectin gene rs1501299 G<T variant, serum adiponectin levels, and the risk of coronary artery disease in a Chinese population. Diabetes Res Clin Pract, 97, 499–504.

19.

Hegener

, Lee

, Cook

, Ridker

, and Zee

. (2006). Association of adiponectin gene variations with risk of incident myocardial infarction and ischemic stroke: A nested case-control study. Clin Chem, 52, 2021–2027.

20.

Hoefle

, Muendlein

, Saely

, et al. (2007). The −11377 C>G promoter variant of the adiponectin gene, prevalence of coronary atherosclerosis, and incidence of vascular events in men. Thromb Haemost, 97, 451–457.

21.

Jang

, Lee

, Chae

, et al. (2005). Association of the 276G->T polymorphism of the adiponectin gene with cardiovascular disease risk factors in nondiabetic Koreans. Am J Clin Nutr, 82, 760–767.

22.

Jung

, Rhee

, Kim

, et al. (2006). Associations between two single nucleotide polymorphisms of adiponectin gene and coronary artery diseases. Endocr J, 53, 671–677.

23.

Karaduman

, Sengul

, Oktenli

, et al. (2006). Tissue levels of adiponectin, tumour necrosis factor-alpha, soluble intercellular adhesion molecule-1 and heart-type fatty acid-binding protein in human coronary atherosclerotic plaques. Clin Endocrinol (Oxf), 64, 196–202.

24.

Katakami

, Kaneto

, Matsuoka

, et al. (2012). Adiponectin G276T gene polymorphism is associated with cardiovascular disease in Japanese patients with type 2 diabetes. Atherosclerosis, 220, 437–442.

25.

Lacquemant

, Froguel

, Lobbens

, Izzo

, Dina

, and Ruiz

. (2004). The adiponectin gene SNP +45 is associated with coronary artery disease in type 2 (non-insulindependent) diabetes mellitus. Diabet Med, 21, 776–781.

26.

Mackawy

. (2013). Association of +45T>G adiponectin gene polymorphism with insulin resistance in non-diabetic Saudi women. Gene, 530, 158–163.

27.

Menzaghi

, Ercolino

, Paola

, et al. (2002). A haplotype at the adiponectin locus is associated with obesity and other features of the insulin resistance syndrome. Diabetes, 51, 2306–2312.

28.

Mofarrah

, Ziaee

, Pilehvar-Soltanahmadi

, Zarghami

, Boroumand

, and Zarghami

. (2016). Association of KALRN, ADIPOQ, and FTO gene polymorphism in type 2 diabetic patients with coronary artery disease: Possible predisposing markers. Coron Artery Dis, 27, 490–496.

29.

Mohammadzadeh

, Ghaffari

, Heibar

, and Bazyar

. (2016). Association of two common single nucleotide polymorphisms (+45T/G and +276G/T) of ADIPOQ gene with coronary artery disease in type 2 diabetic patients. Iran Biomed J, 20, 152–160.

30.

Nakano

, Tobe

, Choi-Miura

, Mazda

, and Tomita

. (1996). Isolation and characterization of GBP28, a novel gelatin-binding protein purified from human plasma. J Biochem, 120, 803–812.

31.

Oguri

, Kato

, Yokoi

, et al. (2009). Association of genetic variants with myocardial infarction in Japanese individuals with metabolic syndrome. Atherosclerosis, 206, 486–493.

32.

Ohashi

, Ouchi

, Kihara

, et al. (2004). Adiponectin I164T mutation is associated with the metabolic syndrome and coronary artery disease. J Am Coll Cardiol, 43, 1195–1200.

33.

Oliveira

, Saddi-Rosa

, Crispim

, et al. (2012). Association of ADIPOQ variants, total and high molecular weight adiponectin levels with coronary artery disease in diabetic and non-diabetic Brazilian subjects. J Diabetes Complications, 26, 94–98.

34.

Ouchi

, Kobayashi

, Kihara

, et al. (2004). Adiponectin stimulates angiogenesis by promoting cross-talk between AMP-activated protein kinase and Akt signaling in endothelial cells. Biol Chem, 279, 1304–1309.

35.

Park

, Park

, and Jee

. (2011). ADIPOQ gene variants associated with susceptibility to obesity and low serum adiponectin levels in healthy Koreans. Epidemiol Health, 33, e2011003.

36.

Persson

, Lindberg

, Gustafsson

, Eriksson

, Paulsson-Berne

, and Lundman

. (2010). Low plasma adiponectin concentration is associated with myocardial infarction in young individuals. J Intern Med, 268, 194–205.

37.

Pischon

, Pai

, Manson

, et al. (2007). Single nucleotide polymorphisms at the adiponectin locus and risk of coronary heart disease in men and women. Obesity (Silver Spring), 15, 2051–2060.

38.

Prior

, Gable

, Cooper

, et al. (2009). Association between the adiponectin promoter rs266729 gene variant and oxidative stress in patients with diabetes mellitus. Eur Heart J, 30, 1263–1269.

39.

Prior

, Jones

, Gill

, et al. (2011). Association of the adiponectin rs266729 C>G variant and coronary heart disease in the low risk ‘Golden Years’ type 1 diabetes cohort. Diabetes Res Clin Pract, 91, e71–e74.

40.

, Li

, Rimm

, et al. (2005). The +276 polymorphism of the APM1 gene plasma adiponectin concentration, and cardiovascular risk in diabetic men. Diabetes, 54, 1607–1610.

41.

Richards

, Waterworth

, O'Rahilly

, et al. (2009). A genome-wide association study reveals variants in ARL15 that influence adiponectin levels. PLoS Genet, 5, 1000768.

42.

Rizk

, El-Menyar

, Marei

, et al. (2012). Association of adiponectin gene polymorphism (+T45G) with acute coronary syndrome and circulating adiponectin levels. Angiology, 64, 257–265.

43.

Rodriguez-Rodriguez

, Garcia-Bermudez

, Gonzalez-Juanatey

, et al. (2011). Lack of association between ADIPOQ rs266729 and ADIPOQ rs1501299 polymorphisms and cardiovascular disease in rheumatoid arthritis patients. Tissue Antigens, 77, 74–78.

44.

Sabouri

, Ghayour-Mobarhan

, Moohebati

, et al. (2011). Association between 45T/G polymorphism of adiponectin gene and coronary artery disease in an Iranian population. ScientificWorldJournal, 11, 93–101.

45.

Shen

, Xing

, and Chen

. (2015). Adiponectin gene polymorphisms contributes to ischemic stroke risk: A meta-analysis. J Renin Angiotensin Aldosterone Syst, 16, 178–184.

46.

Sun

, Li

, Wei

, Tong

, Zheng

, and Guo

. (2012). Recessive protective effect of ADIPOQ rs1501299 on cardiovascular diseases with type 2 diabetes: A meta-analysis. Mol Cell Endocrinol, 349, 162–169.

47.

Thakkinstian

, McEvoy

, Minelli

, et al. (2005). Systematic review and meta-analysis of the association between {beta}2-adrenoceptor polymorphisms and asthma: A HuGE review. Am J Epidemiol, 162, 201–211.

48.

Vaiopoulos

, Marinou

, Christodoulides

, and Koutsilieris

. (2012). The role of adiponectin in human vascular physiology. Int J Cardiol, 155, 188–193.

49.

Vedanthan

, Seligman

, and Fuster

. (2014). Global perspective on acute coronary syndrome: A burden on the young and poor. Circ Res, 114, 1959–1975.

50.

Wassel

, Pankow

, Rasmussen-Torvik

, et al. (2011). Associations of SNPs in ADIPOQ and subclinical cardiovascular disease in the multi-ethnic study of atherosclerosis (MESA). Obesity (Silver Spring), 19, 840–847.

51.

Yang

, Zhang

, Ding

, Wang

, Lei

, and Hu

. (2012). Association of ADIPOQ gene polymorphisms and coronary artery disease risk: A meta-analysis based on 12465 subjects. Thromb Res, 130, 58–64.

52.

Zhang

, Li

, and Xu

. (2012). A meta-analysis of the association of adiponectin gene polymorphisms with coronary heart disease in Chinese Han population. Clin Endocrinol (Oxf), 76, 358–364.

53.

Zhang

, Mo

, Hao

, and Gu

. (2012). Association between polymorphisms in the adiponectin gene and cardiovascular disease: A meta-analysis. BMC Med Genet, 28, 13–40.

54.

Zhang

, Dai

, Bian

, et al. (2011). Association of the cell death-inducing DNA fragmentation factor alpha-like effector A (CIDEA) gene V115F (G/T) polymorphism with phenotypes of metabolic syndrome in a Chinese population. Diabetes Res Clin Pract, 91, 233–238.

55.

Zhang

, Peng

, and Lyu

. (2016). Association between genetic variants in the adiponectin gene and premature myocardial infarction. Zhonghua Xin Xue Guan Bing Za Zhi, 44, 577–582.

56.

Zhao

, Mamatyusupu

, Wang

, et al. (2016). The Uyghur population and genetic susceptibility to type 2 diabetes: Potential role for variants in CAPN10, APM1 and FUT6 genes. J Cell Mol Med, 20, 2138–2147.

57.

Zhong

, Zhen

, Qi

, and Genshan

. (2010). A lack of association between adiponectin polymorphisms and coronary artery disease in a Chinese population. Genet Mol Biol, 33, 428–433.

58.

Zhou

, Jin

, Yao

, Duan

, Wang

, and Liu

. (2014). Association between the adiponectin +45T>G genotype and risk of cardiovascular disease: A meta-analysis. Heart Lung Circ, 23, 159–165.

59.

Zhou

, Xi

, Wei

, et al. (2012). Association between adiponectin gene polymorphisms and coronary artery disease across different populations. Thromb Res, 130, 52–57.

60.

Zhou

, Zhuo

, Wang

, Li

, and Ye

. (2011). Correlation of adiponectin gene SNP +45T/G polymorphism and premature coronary heart disease of Han ethnic males. Zhonghua Yi Xue Za Zhi, 91, 3413–3416.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.10 MB

0.05 MB

0.10 MB

0.03 MB

0.02 MB

0.03 MB