The association between Matrix Metallo-proteinases-9 (MMP-9) gene family polymorphisms and risk of Coronary Artery Disease (CAD): a systematic review and meta-analysis

Background We performed a systematic review and meta-analysis of the Matrix metalloproteinases (MMP)-9 (C1562T), MMP-9 (R279Q), MMP-9 (P574R) and MMP-9 (R668Q) polymorphisms and risk of Coronary Artery Disease (CAD). Methods After a systematic literature search, pooled odds ratio (OR) and their corresponding 95% confidence interval (CI) were used to evaluate the strength of the association. Results We identified 40 studies with 11,792 cases and 8280 controls for C1562T, 7 case-control studies with 5525 cases and 2497 controls for R279Q, 2 studies with 1272 cases and 785 controls for P574R, and 2 studies with 1272 cases and 785 controls for R668Q. MMP-9 (C1562T) polymorphism was associated with increased risk of CAD under dominant model (OR = 1.41, P < 0.001), recessive model (OR = 1.59, P < 0.001), allelic model (OR = 1.38, P < 0.001), TT vs. CC model (OR = 1.70, P < 0.001), and CT vs. CC model (OR = 1.35, P < 0.001). Moreover, the subgroup analysis based on the continent of the study populations in this SNP indicated strong significant association in Asians but not in Europeans. Subgroup analysis was not performed in Africa, America and Oceania, due to lack of sufficient data. Conclusions Our meta-analysis revealed that MMP-9 (C1562T) SNP conferred a susceptibility risk for CAD in the overall analysis and Asian population. The overall analysis and subgroup analysis of the other three SNPs reject the association between MMP-9 polymorphisms and the risk of CAD. Although the results should interpret with caution because of small sample size of included studies in these three SNPs.


Background
Coronary artery disease (CAD) is a worldwide medical problem that is the leading cause of death in both developed and developing countries, especially in older people [1,2]. Several studies have shown that the traditional risk factors, such as blood lipid, diabetes, hypertension, obesity play crucial roles in the initiation and perpetuation of CAD. However, it is nowadays accepted that genetic component has an essential role in the development of CAD [3][4][5][6]. Researches have suggested that family aggregation of CAD is not unusual, and genetic association investigations revealed that the average heritability of CAD is more than 50% [5,7]. Epidemiological studies have found many genetic variants especially single-nucleotide polymorphism (SNP) in association with an increased risk of CAD [8]. The exact mechanism underlying the influence of polymorphism on the pathogenesis of CAD is not fully understood. Nevertheless, polymorphisms in numerous genes involved in inflammation, metabolism of lipid and glucose, blood clotting, and homocysteine may affect susceptibility to CAD [9,10].
. This enzyme is involved in the degradation of extracellular matrix (ECM) components, such as type IV collagen, which is involved in the neovascularization, angiogenesis, inflammatory processes, and development of atherosclerosis.
MMPs are zinc containing enzymes that belong to a neutral protease family. Among the MMP family, MMP-9 is the most important enzyme of this class that is produced by the cells in the vascular wall. Moreover, inflammatory immune cells, such as neutrophils, monocytes as well as endothelial cells and vascular smooth muscle cells (VSMCs) generate MMP-9 [11]. This enzyme is involved in the degradation of extracellular matrix (ECM) components, such as type IV collagen, which is involved in the neovascularization, angiogenesis, inflammatory processes, and development of atherosclerosis [12,13]. Several studies have shown that the levels of MMPs and their matrix-degrading activity are raised in exposed areas of atherosclerotic plaques, or following acute coronary syndrome [13,14]. As a result, it is rational to hypothesize that genetic defects resulting in the overexpression of activated MMPs play a crucial role in the pathogenesis of coronary artery disease (CAD).
Several studies have shown that MMP-9 family polymorphisms might be associated with the risk of CAD [16][17][18]. However, the results are inconsistent. For example, Mahmoodi et al. conducted a case-control study to investigate the association between -1562C > T genetic polymorphism and susceptibility to CAD. But, genotype and allele frequencies of MMP9 -1562C > T polymorphism were similar between CAD patients and controls (P > 0.05) [19]. However, Rodriguez-Perez et al. demonstrated that MMP9 (1562 C > T) allele and the CT genotype were associated with the risk of developing myocardial infraction (MI) [20]. The causes for these controversial results probably due to small sample sizes, different ethnicity, patient selection, clinical heterogeneity, low statistical power, or a combination of these factors. Therefore, we performed this meta-analysis to evaluate whether MMP-9 gene family polymorphisms play a role in CAD susceptibility.

Methods
We followed a protocol based on observational studies in epidemiology (MOOSE) guidelines [21], and results were reported based on Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guideline [22]. This article does not contain any studies with human participants performed by any of the authors.

Systematic search strategy
An exhaustive systematic search was conducted through electronic databases (Scopus, Medline) retrieving all potential publications considered the association between MMP-9 family gene polymorphism and susceptibility to CAD. All publications from inception to February 2020 were included (the search was updated before manuscript submission). The combination of key words and Mesh (Medical Subject Headings) terms were as follow: ("matrix metalloproteinase" [Mesh] OR "MMP" OR "gelatinase") AND ("coronary heart disease" OR "CAD" OR "coronary syndrome" OR "ischemic heart disease" OR "vascular disease" OR "myocardial infarction" OR "MI" OR "atherosclerosis" OR "arteriosclerosis" OR "coronary stenosis" OR "coronary disease" OR "CHD" OR "angina") AND ("single nucleotide polymorphism" OR "SNP" OR "polymorphisms" OR "mutation" OR "variation"). The references of review articles were crosschecked to find potential publications. Only human studies and English language publications were considered.

Inclusion and exclusion criteria
We screened retrieval publications according to following inclusion criteria: 1) observational studies (cohort or case-control design); 2) publications considered the association between MMP-9 family gene polymorphism (C1562T, R279Q, P574R and R668Q) and susceptibility to CAD; 3) publications reporting sufficient data to extract or calculate risk estimates with 95% CI; 4) publications reporting numbers or genotype frequencies in cases and healthy controls. Duplicates, book chapters, letters to editor, animal study, case reports, review articles, and studies with repetitive subjects all were excluded. The application of these criteria recognized 40, 7, 2, and 2 eligible studies for C1562T SNP, R279Q SNP, P574R SNP and R668Q SNP, respectively.

Data extraction
Two authors screened the literature and extracted data independently according to the inclusion and exclusion criteria. The following data was extracted: the first author's name, journal and year of publication, country of origin, ethnicity, number of subjects in the case and control groups, mean or range of age, genotyping method, genotype counts in the case and control group.

Quality assessment
The quality of eligible studies was assessed by using the Newcastle-Ottawa Scale (NOS) [23]. Studies were scored based on three main components: selection, comparability, and ascertainment of outcome. This scale ranges from 1 to 9 stare and studies with scores 0-3, 4-6 or 7-9 were of low, moderate, or high-quality, respectively.

Statistical analysis
In the current meta-analysis, the strengths of association between MMP-9 family gene polymorphism and the risk of CAD was estimated via the OR and 95% CI in five genetic models: dominant model, recessive model, allelic model, homozygote contrast, and heterozygotes contrast. The potential heterogeneity was evaluated by the Q test and the I2 test [23]. According to these two test, if Q had a P value less than 0.1 and I2 exceed 50%, the   [24,25]. Additionally, risk of publication bias was examined by funnel plot, Egger's weighted regression test and Begg's rank correlation test (P < 0.05 was regarded as statistically significant publication bias) [26,27]. Besides, quality assessment of genotype data in case control studies was

Study characteristics
The search and screening process workflow is shown in Fig. 1

Meta-analysis of MMP-9 (C1562T) and risk of CAD
A total of 40 studies with 11,792 cases and 8280 controls were included in quantitative synthesis of the association between MMP-9 (C1562T) polymorphism and CAD susceptibility [16-20, 28-49, 51-61, 63, 64]. Among included studies, 26 studies were carried out in Asian countries, 10 studies were in European countries, 2 studies were in African countries, one study in America, and one in Oceania. Since there were only two studies for Africans and one study for American and Oceania, we excluded them from subgroup analysis. The pooled OR divulged a strong positive association between MMP-9 (C1562T) polymorphism and risk of CAD and announced this SNP as a risk factor for CAD.  (Fig. 2). No statistically significant association was observed in Europeans.
The stratification of studies also performed based on type of CAD, including acute coronary syndrome (ACS) and stable angina. The findings demonstrated a statistically significant association between MMP9 (C1562T) polymorphism and stable angina susceptibility across all genotype model. However, the positive association be-  Table 3.
Meta-analysis of MMP-9 (R279Q) and risk of CAD There were 7 case-control studies with 5525 cases and 2497 controls concerning MMP-9 (R279Q) polymorphism and risk of CAD [18,38,42,47,48,50,62]. Of those, 4 studies were performed in Asians and 3 studies were in Europeans. The pooled results indicated a negative, but not significant, association between MMP-9 (R279Q) gene polymorphism and CAD risk under all genotype models for the overall population and subgroup analysis (Fig. 3). The results of pooled ORs, heterogeneity tests and publication bias tests in different analysis models are shown in Table 3.

Meta-analysis of MMP-9 (P574R) and risk of CAD
For MMP-9 (P574R) SNP, two studies with 1272 case and 785 controls were included for quantitative analysis [47,50]. Studies were carried out in China and India. The results of overall population reject any association between MMP-9 (P574R) SNP and risk of CAD across all genotype models. The results of pooled ORs, heterogeneity tests and publication bias tests in different analysis models are shown in Table 3.

Meta-analysis of MMP-9 (R668Q) and risk of CAD
Two studies with 1272 case and 785 controls were included for quantitative analysis for MMP-9 (R668Q) SNP [47,50]. Studies were carried out in China and India. There was no evidence of significant association between MMP-9 (R668Q) SNP and risk of CAD under all genotype models. The results of pooled ORs, heterogeneity tests and publication bias tests in different analysis models are shown in Table 3.

Publication bias and heterogeneity
In this study, we used Egger's regression test, Begg's adjusted rank correlation test and visual examination of the funnel plot (just for C1562T and R279Q) to measure publication bias (Fig. 4). Overall, no significant publication bias was detected. Besides, the estimation of heterogeneity by I 2 and Q test was significant in some models (Table 3).

Sensitivity analysis
The leave-one-out method was used in the sensitivity analysis to explore the effect of individual data on the pooled ORs (just for C1562T and R279Q). The significance of ORs was not altered through omitting any single study, indicating that our results were statistically robust (Fig. 5).

Discussion
Up until now, numerous investigations have been performed to address the association between SNPs of the MMP gene family, including MMP-9 (C1562T), MMP-9 (R279Q), MMP-9 (P574R), and MMP-9 (R668Q) polymorphisms and risk of CAD. The findings of these investigations have sometimes been in accordance with  Different human and animal experiments have suggested that there is an elevated level of MMP-9 in the atherosclerotic arteries in comparison to health controls. It was also shown that MMP-9 is predominantly active in the lipid core margin of the atherosclerotic plaques, the shoulder regions of plaque, and in regions with active formation of microvessels. As a consequence, it appears that MMP-9 plays a critical role in the stability/instability of the coronary artery plaques and development of myocardial infarction during CAD [65]. In addition, researchers have observed that in mice knock-out for the MMP9 gene, migration potency of the VSMCs as well as atherosclerosis lesions were reduced in comparison to the wild-type animals [66]. According to clinical observations, upregulation of MMP-9 exhibited a correlation with instability of the atherosclerosis plaque and premature CAD development [67]. Based on the prospective studies, serum levels of MMP-9 could confer a tool to estimate the mortality risk during the cardiovascular diseases [68].
Zhang et al. in the 214 meta-analysis, by including 26 studies containing 12,776 cases and 6371 controls, indicated that MMP-9 (C1562T) polymorphism was not associated with the risk of CAD in the overall results [69]. However, they reported that MMP-9 (C1562T) SNP is involved in the decrease susceptibility to CAD in Asian population. In 2016, a metaanalysis was conducted on 10 case-control studies to assess the possible relationship between the MMP-9 (C1562T) SNP and CAD in the Chinese Han population. This study indicated that all genetic comparisons of the MMP-9 (C1562T) SNP increased the risk of CAD in the Chinese Han population [70]. In the current meta-analysis, association between SNPs of the MMP-9 gene family, including MMP-9 (C1562T), MMP-9 (R279Q), MMP-9 (P574R), and MMP-9 (R668Q) polymorphisms and risk of CAD was evaluated. Our literature search led to identification and inclusion of 40 studies with 11,792 cases and 8280 controls for MMP-9 (C1562T) SNP, 7 case-control studies with 5525 cases and 2497 controls for MMP-9 (R279Q) SNP, 2 studies with 1272 case and 785 controls for MMP-9 (P574R) SNP, and 2 studies with 1272 case and 785 controls for MMP-9 (R668Q) SNP. Therefore, this is the most comprehensive metaanalysis of MMP-9 gene family polymorphisms and risk of CAD to date (March 2020). Our analysis revealed that MMP-9 (C1562T) polymorphism increased the risk of CAD in the overall analysis under dominant (OR = 1.41), recessive (OR = 1.59), allelic (OR = 1.38), homozygous TT vs. CC (OR = 1.70), and heterozygous CT vs. CC (OR = 1.35) models. In contrast to Zhang et al. [69] study, we noticed that MMP-9 (C1562T) polymorphism increased the susceptibility of CAD risk in the Asian population under all genotyping models; dominant (OR = 1.47), recessive (OR = 2.06), allelic (OR = 1.45), homozygous TT vs. CC (OR = 2.42), and heterozygous CT vs. CC (OR = 1.39) models. However, other three polymorphisms of the MM9 gene, including MMP-9 (R279Q), MMP-9 (P574R), and MMP-9 (R668Q) polymorphisms, were not associated with CAD risk. Regulatory mechanisms at the transcriptional level is involved in the modulation of MMP-9 expression. The MMP-9 (C1562T) SNP is harbored within the 9 bp sequence GCGCAC/TGCC (− 1567 → − 1559), which is considered as a regulatory element of the gene and confers a site for binding of molecules involved in the inhibition of transcription [71]. It was found that an alteration in the binding site structure by substitution of the MMP-9-1562 C allele with − 1562 T allele led to decreased binding potential of the proteins involved in the inhibition of transcription to the DNA sequence [30]. As a result, MMP-9 (C1562T) SNP plays a role in orchestrating the transcription activity of MMP-9 and, hence, modulate the susceptibility risk to several diseases. Therefore, we analyzed the available data to gain a wide understanding of this SNP in case of CAD. We noticed that T allele representation was increased in all models of MMP-9 (C1562T) SNP comparison in the overall analysis, including dominant model (OR = 1.41, 95% CI = 1.23-1.61, P < 0.001), recessive model (OR = 1.59, 95% CI = 1.29-1.96, P < 0.001), allelic model (OR = 1.38, 95% CI = 1.23-1.55, P < 0.001), TT vs. CC model (OR = 1.70, 95% CI = 1.35-2.13, P < 0.001), and CT vs. CC model (OR = 1.35, 95% CI = 1.18-1.54, P < 0.001), which was associated with an increased risk of CAD significantly. Upregulation of MMP-9 may be involved in the CAD development by multiple approaches, including increased proliferation and migration of VSMCs, remodeling of the injured vascular cells, and enhancing the plaque instability and rupture (that leads to the development of thrombosis), eventuating in myocardial infarction and CAD [72].
In spite of an attempt to perform the most comprehensive meta-analysis of the MMP9 gene SNPs and the risk of CAD, a number of limitations and caveats of this meta-analysis study should be taken into consideration. First, the number of studies and sample size for MMP-9 (R279Q), MMP-9 (P574R), and MMP-9 (R668Q) polymorphisms in this meta-analysis was relatively small to conclude a valid report of the association of these SNPs and CAD risk. Second, we searched for the articles published in only the English language and a number of potential studies might be omitted. Third, this metaanalysis was based on a crude analysis of the genetic polymorphisms, and the adjusting the analysis by gender, age, and other environmental factors were not implemented. Fourth, we detected some degrees of heterogeneity for the analyzed SNPs, that might stem from difference in genetic stratification and ethnicity, diversity in the environmental factors in different populations, and the detection methods.

Conclusion
Taken all the evidence into conclusion, this was the most comprehensive evaluation of the four MMP9 gene SNPs in association with CAD. We reported that MMP-9 (C1562T) SNP conferred a susceptibility risk for CAD in the overall analysis and Asian population. That notwithstanding, other three polymorphisms were not associated with disease risk, probably due to little sample size. Hence, we warrant further studies with respect to evaluation of other MMP9 gene SNPs in association with CAD. Furthermore, the role of other factors, such as age, gender, environmental contributing factors as well as other MMP9 gene variations in the analyses ahead will hopefully shed further light on the bona fide association of MMP9 gene polymorphisms and risk of CAD susceptibility.