A potential relationship between MMP-9 rs2250889 and ischemic stroke susceptibility

Purpose Ischemic stroke (IS), a serious cerebrovascular disease, greatly affects people's health and life. Genetic factors are indispensable for the occurrence of IS. As a biomarker for IS, the MMP-9 gene is widely involved in the pathophysiological process of IS. This study attempts to find out the relationship between MMP-9 polymorphisms and IS susceptibility. Methods A total of 700 IS patients and 700 healthy controls were recruited. The single nucleotide polymorphism (SNP) markers of the MMP-9 gene were genotyped by the MassARRAY analyzer. Multifactor dimensionality reduction (MDR) was applied to generate SNP–SNP interaction. Furthermore, the relationship between genetic variations (allele and genotype) of the MMP-9 gene and IS susceptibility was analyzed by calculating odds ratios (ORs) and 95% confidence intervals (CIs). Results Our results demonstrated that rs2250889 could significantly increase the susceptibility to IS in the codominant, dominant, overdominant, and log-additive models (p < 0.05). Further stratification analysis showed that compared with the control group, rs2250889 was associated with IS risk in different case groups (age, female, smoking, and non-drinking) (p < 0.05). Based on MDR analysis, rs2250889 was the best model for predicting IS risk (cross-validation consistency: 10/10, OR = 1.56 (1.26–1.94), p < 0.001). Conclusion Our study preliminarily confirmed that SNP rs2250889 was significantly associated with susceptibility to IS.


. Introduction
Stroke, an acute cerebrovascular disease, is the main cause of death and long-term disability (1). The National Epidemiological Survey of Stroke in China (NESS-China) estimated that the death rate of stroke in China was 149.49 per 100,000, accounting for 1.57 million deaths in 2018 (2). There are two main types of strokes: ischemic stroke (IS) and hemorrhagic stroke (HS). Previous research studies have suggested that IS has become a major disease in China and one of the most important reasons for people's disabilities (3). From this perspective, IS remains one of the biggest challenges that China is facing, and the need for a large number of research studies on the pathogenesis of IS is more than justified.  Although the pathogenesis of IS still has been debated, more and more studies have highlighted that the pathogenesis of IS is regulated by multiple factors, such as smoking, hypertension, diabetes, drinking, and age (4). Recently, a report of Mendelian randomization has identified supported atrial fibrillation (AF), body mass index (BMI), smoking, blood pressure, white matter hyperintensity, and type 2 diabetes mellitus (T2DM) as risk factors of IS (5). In addition, many studies have emphasized the influence of genetic factors on IS, and exactly, single nucleotide polymorphisms (SNPs) in gene have a certain regulatory effect on the occurrence of IS (4,6). For example, ALDH7A1 rs12514417 polymorphism may increase the risk of IS in individuals exposed to alcohol (7). OPG SNP T245G is associated with enhanced IS risk among the Chinese (8). At present, the research on postgenome-wide association study (GWAS) is ongoing, which mainly studies whether genetic variants associated with different disease phenotypes exert their pathogenicity, and it is a means of exploring the relationship between gene polymorphisms and disease risk.
Matrix metalloproteinases (MMPs) are a group of zincdependent endoproteinases that regulate extracellular matrix (ECM) via proteolysis, cell adhesion, and cytokines (9). In addition, MMPs have been proven to participate in a series of biological processes, such as cell proliferation, migration, angiogenesis, and immune response (10,11). MMP family members include MMP-1, -2, -3, -7, -8, -9, -12, -13, MT1-MMP, and MT3-MMP, which are expressed in various vascular tissues and cells (12). The matrix metalloproteinase-9 (MMP-9) gene is located in 20q13.12 and consists of 13 exons and 12 introns. It plays a role in the movement of vascular smooth muscle cells and the instability of atherosclerotic plaque (13,14). At present, a meta-analysis showed that MMP-9 and MMP-12 gene polymorphisms may be risk factors for IS, whereas MMP-1, MMP-2, and MMP-3 were not associated with the risk of IS (15). Moreover, the latest study has found that rs243849 and rs14070 in MMP-2 are significantly related to the risk of IS in the Shaanxi population from China (16). However, no significant association between MMP-9 genetic variants and intracerebral hemorrhage (ICH) susceptibility was reported (17). We hypothesized that MMP-9 gene variants may play different roles in the occurrence of IS. Therefore, our case-control study aimed to investigate the relationship between MMP-9 SNPs and the susceptibility to IS in the Northwest Chinese population to predict potential loci for IS risk.

. . Research participants
The study participants, including 700 IS patients and 700 healthy controls, were consecutively recruited from Xi'an No. 3 Hospital from August 20 to December 22. The case group consisted of IS patients, who were admitted to the hospital within 72 h after the symptom appeared and had focal neurological deficit symptoms lasting for more than 24 h. According to the diagnostic criteria of the World Health Organization (WHO), all IS patients were confirmed by at least two independent neurologists using brain computed tomography (CT) scans, magnetic resonance imaging (MRI), and blood routine tests. Participants in the control group were normal and healthy, without other complicated cerebrovascular diseases, tumors, hypertension, diabetes, and other medical histories. All study participants were from the Northwest of China and provided written informed consent.

. . Primer design
By retrieving the NCBI (http://www.ncbi.nlm.nih.gov/omim/) and GeneBank databases combined with the MassArray analysis software, the primer sequence was designed. These primers were sent to Shangon Biotechnology Co., Ltd (Shanghai, China) to synthesize for the polymerase chain reaction (PCR) process of MMP-9 SNPs (Supplementary Table 1).

. . Gene selecting and SNP genotyping
The physical position of MMP-9 on chromosome 20:46008908-46016561 was obtained through the e!GRCh37 (http://asia. ensembl.org/Homo_sapiens/Info/Index) database, and the table of MMP-9 variants was downloaded. SNPs were selected based on Hardy-Weinberg equilibrium (HWE) > 0.01, MAF > 0.05, and Min Genotype > 75% using Haploview software. We further combined MassARRAY primer design software, HWE > 0.05, MAF > 0.05, and the call rate > 95% in our study population, for the selection of SNPs. Among the remaining SNPs, three candidate . /fneur. . SNPs (rs2250889, rs17577, and rs13925) in MMP-9 have previously been reported to be associated with stroke risk (18, 19) but have not been studied in a northwest Chinese Han population. Hence, this study was conducted to explore the correlation between these SNPs and IS risk in the northwest Chinese Han population. The dbSNP, Haploreg, and RegulomeDB databases were used for predicting the potential function of these SNPs. Approximately 5 ml of peripheral blood samples were collected from each participant and placed in tubes containing ethylenediaminetetraacetic acid (EDTA) in a −80 • C refrigerator. Subsequently, genomic DNA was extracted from blood samples using the GoldMag nanoparticles method (Gold Mag Co. Ltd., Xi'an, China) according to the manufacturer's instructions.

. . Statistical analysis
All statistical analyses in our study were performed using SPSS (v.22). The differences in demographic characteristics between IS patients and controls were estimated by the chi-square test for categorical variables and the student's t-test for continuous variables. HWE was assessed using the χ 2 test with one degree of freedom. The strength of all associations between genetic variants in MMP-9 and IS risk was evaluated by logistic regression models with odds ratios (ORs) with 95% confidence intervals (CIs). Statistical analysis was performed using the online SNPStats (v.9.4) and PLINK (v.1.09) analysis software. Logistic regression analysis was introduced to study the effect of each characteristic on the relationship between MMP-9 variants and IS risk. The credibility of the significant associations was assessed by falsepositive report probability (FPRP) analysis, with 0.2 as an FPRP threshold and 0.1 as a prior probability (20,21). Multivariate dimension reduction (MDR) software was utilized to further evaluate SNP-SNP interaction and predict the best model related to IS risk (22). All p-values are two-sided, with a p-value of < 0.05 indicating statistical significance, whereas the corrected pvalue of < 0.05/3 was considered statistically significant after the Bonferroni correction.

. . Study information about the case-control group
The detailed demographic characteristics of the case and control groups that were included are shown in Table 1. A total of 1,400 subjects were selected, including 700 IS patients and 700 healthy controls. There were no significant differences between the two groups in gender, age, smoking status, and drinking status (p > 0.05).

FIGURE
Association results of MMP-gene polymorphism rs with stroke risk. SNPs, single nucleotide polymorphisms; CI, confidence interval; MAF, minor allele frequency; OR, odds ratio. p a : Student's t-test is used; p < . indicates statistical significance.

. . Results of MMP-gene polymorphism and IS risk
Three SNPs (rs2250889, rs17577, and rs13925) of MMP-9 were successfully screened, and the basic information about them was demonstrated ( Table 2). Database analysis presented that the potential functions of these SNPs might be related to promoter/histone marks, DNAse, protein-bound motifs changed, eQTL, and transcription factor binding. In addition, the association between three SNPs in MMP-9 and IS susceptibility was analyzed under five genetic models (codominant, dominant, recessive, overdominant, and log-additive), as illustrated in Figure 1

. . Stratification analysis by demographic characteristics
The stratified analysis by gender (male and female), age, drinking, and smoking was conducted to further analyze the correlation between selected SNPs and IS risk, as shown in Table 3. The effect of rs2250889 on the risk of IS varied obviously with age. More precisely, rs2250889 played a risk-increasing role in After Bonferroni correction, rs2250889 was associated with an increased IS risk in the subgroup aged ≤ 51 years (codominant, dominant, and overdominant), women (codominant, dominant, and overdominant), smokers (codominant, dominant, and overdominant), and non-drinkers (codominant, dominant, overdominant, and log-additive), respectively. Unfortunately, there were no significant associations between other SNPs (rs17577, rs13925) and IS risk in terms of age, gender, smoking, and drinking (Supplementary Table 3).

. . Stratification analysis by complications
The relationship between MMP-9 SNPs and IS risk was also investigated in the subgroups of IS patients     with/without hypertension or diabetes, as shown in Table 4 and Supplementary . . FPRP analysis of the positive findings FPRP analysis was performed to confirm whether the significant findings deserved attention (Table 5). At the prior probability level of 0.01, the significant association of MMP-9 rs2250889 with IS risk remained noteworthy in the overall analysis (FPRP = 0.074, 0.125, and 0.070). Moreover, the association of MMP-9 rs2250889 with IS risk was also positive in the subgroup .
aged ≤ 55 years, women, smokers, and non-drinkers at the prior probability level of 0.01.

. . MDR analysis of MMP-SNPs
The multi-factor dimensionality reduction (MDR) analysis was carried out to find the relationship between the SNP-SNPs interactions in IS occurrence. The significant combinations of variables were selected based on the entropy measure to assess the information gain (IG) associated with attribute interaction. The patterns of entropy recapitulate the main and/or interaction effects of the paired combination of each attribute. As demonstrated in Figure 2, rs17577 and rs13925 had a synergistic effect with the IG value of 0.16%, and rs2250889 and rs17577 had strong redundancy with the IG value related to IS of −0.05%. More importantly, we found that rs2250889 was the best model for predicting IS risk in Table 6 (Cross-Validation Consistency: 10/10, OR= 1.48(1.20-1.83), p = 0.0003).

. Discussion
Ischemic stroke is a type of disease that causes serious harm to the human body and cognition. It is particularly important to pay attention to clinical treatment and diagnosis. In our experiments, 1,400 subjects were recruited and statistically analyzed to report the association between MMP-9 SNPs (rs2250889, rs17577, and rs13925) and IS susceptibility. In the overall analysis, only rs2250889 was significantly associated with IS susceptibility. In a stratified analysis, rs2250889 could increase the risk of IS in subjects of different ages, women, smokers, and non-drinkers. Moreover, rs2250889 might be a risk factor for IS patients with hypertension and IS patients without diabetes. MDR showed that rs2250889 was the best model for predicting IS risk. Taken together, our findings are the first to discuss the relationship between MMP-9 rs2250889 and IS susceptibility and to evaluate the diagnostic significance of this SNP. Rs3749966, a missense variant [R (Arg) > P (Pro)], might be associated with promoter/histone marks, DNAse, proteinbound, motifs changed, eQTL, and transcription factor binding. Therefore, we speculated that MMP-9 rs2250889 might influence IS occurrence by regulating the expression level of MMP-9, which certainly requires experimental verification.
As a member of the MMP family, MMP-9 can regulate the balance of ECM and participate in multiple important biological processes. It has been reported that MMP-9 is closely related to intracranial aneurysms, atherosclerosis, ischemic brain injury, and other diseases (23). Recently, studies have found that MMP-9 polymorphism is associated with intracranial aneurysm (24). More studies have detected the expression of MMP-9 in the serum and brain tissues of IS patients and found that it is significantly high (25,26). In addition, the TT genotype of MMP-9-1562C/T has been proven to be a risk factor for hemorrhagic complications after thrombolytic therapy for acute IS (AIS) (27). More importantly, in a case-control study, MMP-9 SNP rs3918242 was associated with the increased risk of IS in the Asian population, but there was no such finding in the Caucasian population (28). Similarly, Fan et al. have indicated that MMP-9 rs17576 is associated with an increased IS risk in the Han Hakka population (29). Finally, our study concluded that MMP-9 rs2250889 significantly increased the susceptibility to IS in the northwest Chinese population. This shows that the SNPs of MMP-9 are different in different races.
Age and sex exhibit complex effects on IS risk and pathophysiology. Aging is the most important unmodifiable risk factor for IS. Compared with young stroke patients, the mortality and morbidity of elderly stroke patients are higher, and functional recovery is slower. Of note, the role of gender in IS risk is affected by age (30). Some studies have explored the impact of SNPs at different ages on IS risk in people of different ages. For instance, PITX2 rs6817105 significantly increased the risk of stroke in people aged over 65 years (31). In this study, MMP-9 rs2250889 was related to the increased risk of IS in subgroups of different ages. This may be due to differences in MMP-9 SNPs. Certainly, we would not rule out a trend of getting younger of IS patients. From the perspective of gender, a recent study has revealed that young women (18-45 years old) may have a 44% higher risk of IS than their male counterparts (32). Our research also revealed that SNP rs2250889 could promote the occurrence of IS in women. Tobacco smoking can worsen IS prognosis by increasing blood-brain barrier permeability and accelerating the formation of cerebral edema (33). Drinking frequency may be also tied to the risk of IS (34). In terms of hierarchical analysis of other characteristics (smoking and drinking), similar to some studies, the genotypes rs3793917 and rs2672587 have a cumulative risk of IS in smokers (8). Our study also concluded that rs2250889 would increase the susceptibility to IS in smokers. More interestingly, the study found that HTRA1 rs2268350 was significantly associated with IS in drinkers (8). In contrast, in this study, rs2250889 was related to the increased susceptibility to IS in non-drinkers, suggesting that rs2250889 may be a new locus on MMP-9 for predicting IS risk.
Comorbidities are a hallmark of IS, which both increase the incidence of IS and worsen the prognosis. Hypertension  is common in IS patients and the most important modifiable risk factor for IS (35). Another potential risk factor for stroke pathogenesis is diabetes mellitus, which has been shown to be associated with increased mortality in stroke (36). In this study, compared with healthy controls, rs2250889 was related to the risk of IS in IS patients with hypertension, without hypertension, and without diabetes, respectively. Although we have discussed the results described above, several potential limitations in this study should be noted. First, our sample size is small, so we need to expand the sample size to further verify our experiment. Second, the result of this research is not generalizable as it was found in a specific group of the Chinese Han population. Third, our experiment preliminarily determined the impact of MMP-9 on the susceptibility to IS.
In subsequent experiments, we will further verify the role of MMP-9 in the occurrence of IS through the quantitative realtime (qPCR) and Western blot (WB) experiments. Finally, IS is a complex disease. Further analysis of the impact of MMP-9 on the pathogenesis of IS in cell and animal models is worthy of attention.

. Conclusion
In summary, our research preliminarily determined the correlation between MMP-9 gene polymorphisms (rs2250889, rs17577, and rs13925) and IS risk, especially SNP rs2250889, which provides a theoretical basis for studying the impact of MMP-9 on the pathogenesis of IS.

Data availability statement
The data presented in the study are deposited in the zenodo repository (https://zenodo.org/), accession number 8078657.

Ethics statement
The studies involving human participants were reviewed and approved by Xi'an Third Hospital (Ethics Approval No.: SYXSLL-2019-034). The patients/participants provided their written informed consent to participate in this study.