Matrix metalloproteinase‐9 gene polymorphisms are associated with ischemic stroke severity and early neurologic deterioration in patients with atrial fibrillation

Abstract Objectives The mechanisms of ischemic stroke severity and early neurologic deterioration (END) are not fully understood. The aim of the present study was to investigate the association of six variants in MMP‐9 gene with ischemic stroke severity and the risk for END in ischemic stroke (IS) patients with atrial fibrillation (AF). Methods This was a multi‐center, prospective, observational study of 615 acute IS patients with AF admitted to six participating hospitals between June 2016 and October 2017. Ischemic stroke severity was assessed using the National Institutes of Health Stroke Scale (NIHSS) score on admission. END was defined as an increase of four or more points in NIHSS within 10 days of admission. Six variants of MMP‐9 gene were examined using mass spectrometry. Results Among the 615 enrolled patients, 112 (18.2%) patients presented with moderate or severe stroke (NIHSS score ≥16), and 108 (17.6%) patients suffered from END within 10 days of admission. Multiple logistic analysis showed that prestroke antiplatelet therapy, prestroke anticoagulant therapy, rs3918242 CT/TT, and rs3787268 AG/GG were independent predictors for stroke severity. Cox proportional hazard regression revealed that diabetes mellitus, prestroke antiplatelet therapy, prestroke anticoagulant therapy, rs1056628 AC/CC, and rs3918242 CT/TT were independently associated with the risk of END. Conclusions The incidence of moderate or severe stroke and END was very common in acute IS patients with AF. MMP‐9 polymorphisms were independently associated with severe stroke and higher risk of END, and prestroke antithrombotic treatment was associated with less severe stroke and lower risk of END in patients with AF.


| BACKG ROU N D
Stroke is a major cause of death and adult disability (Luengo-Fernandez et al., 2013), and there are approximately 3-4 million new stroke cases every year in China, with roughly 80% being ischemic strokes (IS) (Guan et al., 2017;Jiang et al., 2006;Wang et al., 2007).
Atrial fibrillation (AF) is an independent risk factor for IS, and accounts for 10%-25% of all ischemic strokes (Meschia et al., 2014;Sposato et al., 2015). Recanalization of occluded vessels after embolic stroke is the only therapeutic intervention available to treat acute IS (Powers et al., 2018). However, the rate of recanalization treatment is very low (Wang et al., 2007).
Initial ischemic stroke severity is associated with post-stroke hemorrhagic transformation (HT), mortality and functional outcomes (Huang, Weng, Su, & Lin, 2018;Lee, Lee, & Jung, 2018). Compared with other stroke subtypes, IS patients with AF are known to be related to stroke severity and stroke outcomes (Park et al., 2016).
Stroke severity is dependent on the type of stroke, duration of vessel occlusion, and is also influenced by several toxic mechanisms, most identified in experimental animal models of stroke (Endres et al., 2008). Brain repair involves mechanisms differentially activated in space and time, including inflammation, brain remodeling and relearning of activated neural networks (Pekna & Pekny, 2012;Wieloch & Nikolich, 2006). Genetic factors impact on the innate cellular mechanisms and environmental factors, affecting the severity of brain damage and the subsequent functional outcomes (Cramer & Procaccio, 2012).

Early neurological deterioration (END) is fairly common in pa-
tients with acute IS, and is associated with increased morbidity and mortality (Vahidy et al., 2014;Yi, Han, Zhou, Lin, & Liu, 2016). It is very important to investigate the mechanisms of END, and prevent this detrimental event. Although our previous studies have shown that cytochrome P450 (CYP) genetic variants, CYP metabolite levels, antiplatelet drug resistance, and dual therapy with clopidogrel and aspirin are associated with END risk Yi, Lin, Li, Zhou, & Han, 2017;, the underlying mechanisms of END remain unclear.
Furthermore, previous studies did not investigate the incidence and possible mechanisms of END in acute IS patients with AF.
Matrix metalloproteinases (MMPs), well-known inflammatory mediators, belong to a family of structurally related zinc-binding proteolytic enzymes that are widely distributed in human tissues.
MMP-9 protein is involved in inflammatory responses trigged by ischemia, and it plays a key role in BBB destruction and neuronal damage (Barr et al., 2010). MMP-9 inhibitors can decrease the BBB destruction in the experimental stroke model (Lu et al., 2009). The MMP-9 gene is located on chromosome 20q12.2-13.1.
Previous studies have demonstrated that MMP-9 activity is controlled by single nucleotide polymorphisms (SNPs) of MMP-9 gene (Blankenberg et al., 2003). Several reports have shown that MMP-9 polymorphisms are associated with carotid atherosclerosis and increased IS risk (Lin et al., 2012;Yuan et al., 2013). However, there are few studies to evaluate the association of MMP-9 gene polymorphisms with initial ischemic stroke severity and the risk for END IS patients with AF.
With this background we aimed to investigate whether MMP-9 gene polymorphisms influence initial ischemic stroke severity and the risk of END in acute IS patients with AF in the Chinese population. This study was expected to provide novel insight into the mechanisms for stroke severity and END, and prevent this detrimental event.

| Study populations
Between June 2016 and October 2017, we consecutively enrolled the first-ever IS patients with known history of AF or atrial flutter within 48 hr of stroke onset. All enrolled patients had relevant cerebral lesions on diffusion weighted magnetic resonance imaging. (d) patients who were receiving unfractionated heparin, low-molecular-weight heparin within 7 days of stroke onset; (e) intravenous thrombolytic therapy or intra-arterial catheter-based treatment; (f) hypoxia, fever, or any relevant hemodynamic compromise on admission; (g) severe cardiovascular, liver, or renal disease; and (h) unwillingness to participate in this study. All enrolled patients received standard therapies based on standard guidelines (Kernan et al., 2014;Powers et al., 2018).
The following data were recorded: (a) age, sex, history of hypertension and diabetes mellitus, dyslipidemia, smoking, and coronary artery disease or myocardial infarction; (b) fasting total plasma cholesterol (TC), triglycerides (TG), low-density lipoprotein cholesterol (LDL-C), fasting glucose, and hemoglobin A1c. Dyslipidemia was defined as TC > 200 mg/dl, TG > 180 mg/dl or use of lipid-lowering medication Yi, Lin, Li, Zhou, et al., 2017); (c) prestroke medication and in-hospital treatment. Prestroke antithrombotic treatment, including antiplatelet therapy, warfarin, or non-vitamin K antagonist oral anticoagulants (NOACs), was defined as documentation of patients receiving an antithrombotic agent within 7 days before their index stroke onset.
Whole blood (3 ml) from each patient was drawn from an arm vein into a sterile tube containing ethylenediaminetetraacetic acid and genomic DNA was extracted using a modified phenol/chloroform method and purified using a UNIQ-10 kit (Sangon Biotech Co., Ltd., Shanghai, China). The genotyping of the six SNPs was performed using the matrix-assisted laser desorption/ionization time of flight mass spectrometry method as previously described Yi, Lin, Li, Zhou, et al., 2017;. For each variant, subjects were dichotomized a priori into two groups based on its known effect on enzymatic function according to the literature and with the use of established common-consensus star allele nomenclature .

| Initial stroke severity and assessment of END
For each patient, National Institutes of Health Stroke Scale (NIHSS) assessment was performed by a member of the stroke team on admission, and subsequently on a daily basis throughout the period of hospitalization. An additional NIHSS assessment was performed whenever examination deteriorated. Upon notification of deterioration, a stroke team member reassessed the patient and performed the additional NIHSS evaluation. Initial ischemic stroke severity was assessed using NIHSS score on admission, and patients with an NIHSS score ≥ 16 were classified as having a moderate or severe stroke (Adams et al., 1999;Powers et al., 2018). According to our previous studies (Yi, Chi, Wang, Zhang, & Lin, 2015;Yi, Lin, Wang, Zhang, & Chi, 2014), END was defined as an increase of four or more points in NIHSS within 10 days of admission, while excluding a new infarct in another vascular territory or HT of infarct.

| Statistical analysis
We speculated that a sample of 600 patients would provide 80% power to detect a relative risk increment of 10% in the percentage of END in patients carrying variant genotypes, with a two-sided type I error of 0.05, assuming an event rate of 15% in patients carrying wild-type genotypes.
Baseline clinical characteristics were compared using Student t test (continuous variables) and χ 2 test (categorical variables) between patients with NIHSS score ≥16 and <16 or between patients with and without END. Deviation of Hardy-Weinberg equilibrium for genotype frequencies was also analyzed using goodness-of-fit χ 2 -test Zhang et al., 2018). Difference of genotype frequencies was compared using χ 2 -test between patients with NIHSS score ≥16 and <16 or between patients with and without END. Multivariable logistic regression analysis was employed to investigate independent predictors for initial stroke severity, and reported as odds ratio (OR) with 95% confidence interval (CI).
Cox proportional-hazards model was used to assess the risk factors for END, and reported as hazard ratio (HR) with 95% CI. Variables entered the models were traditional risk factors (such as age, sex, hypertension and diabetes mellitus) and the variables showed a significant association (p < 0.1) with stroke severity or END on univariate analysis.
All statistical analyses were performed using SPSS 16.0 (SPSS Inc., Chicago, IL). All tests were two sided, and a p value less than 0.05 was considered statistically significant.

| D ISCUSS I ON
With this multi-center, prospective, observational study, we investigated the possible role of MMP-9 gene polymorphisms in initial stroke severity and END risk in IS patients with AF. The results showed that the proportion of patients presenting with moderate or severe stroke was 18.2%, and 17.6% of patients suffered from END. Prestroke antiplatelet therapy, prestroke anticoagulant therapy, rs3918242 CT/TT, and rs3787268 AG/GG were independent predictive factors for initial stroke severity by multivariable logistic regression. Cox proportional hazards model revealed that diabetes mellitus, prestroke antiplatelet therapy, prestroke anticoagulant therapy, rs1056628 AC/CC, and rs3918242 CT/TT were independently associated with the risk of END.
Platelet activation may increase atherogenesis and promote blood vessel wall injuries, and has a crucial role in thrombus extension and ischemic damage (Fateh-Moghadam et al., 2005). Thus, the current guideline recommends oral administration of aspirin within 48 hr of stroke onset in patients with acute IS to reduce mortality and unfavorable outcomes (Powers et al., 2018). However, the effect of prestroke aspirin use on initial stroke severity or END is controversial. One recent study demonstrated that strokes were less severe in patients already taking aspirin (Park et al., 2016), whereas some others have not shown an association between prestroke aspirin use and initial stroke severity (Kim et al., 2010). We previously reported that prestroke concomitant statin and aspirin use was associated with lower END and platelet activity in IS patients with atherothrombosis or small artery disease (Yi, Han, Wang, Zhou, & Lin, 2017). The results of our study show that prestroke antiplatelet therapy may reduce initial stroke severity and the incidence of END in IS patients with AF. Aspirin may benefit IS patients by improving the microcirculation in the ischemic penumbra through inhibition of platelet-derived vasoconstrictors, such as thromboxane A2 (Rosenblum & El-Sabban, 1977). In addition, aspirin may also limit clot size, extent of thrombosis, and subsequent embolism (Joseph, Han, Tsering, Grunfeld, & Welch, 1992). Furthermore, aspirin has neuroprotective and antiinflammatory effects, these may be other beneficial mechanisms of aspirin (Kuhn, Muller, Bttner, & Gerlach, 1995).
AF is a potentially treatable risk factor for IS. Numerous studies have revealed that warfarin or NOACs can reduce IS risk (Patel et al., 2011). Based on these data, the current guideline recommends warfarin or NOACs for stroke prevention in high-risk patients with AF (Meschia et al., 2014). However, oral anticoagulants are especially underused in patients with known nonvalvular AF in China . Several studies have demonstrated that preceding oral anticoagulants are associated with reduced initial stroke severity on admission and reduced disability or death at discharge in IS patients with AF (O'Donnell et al., 2006;Xian et al., 2017). These results were consistent with our study. Thus, preceding antithrombotic treatment is very important for reducing ischemic stroke severity and END risk in the high-risk patients with AF.
The most interesting finding in this study was that the variants of rs3918242 and rs3787268 in MMP-9 gene were independent predictors for initial stroke severity, and rs1056628 AC/CC and rs3918242 CT/TT were independently associated with higher risk of END after adjusting for covariates. To the best of our knowledge, this study is the first to investigate the associations of variants in MMP-9 gene with initial stroke severity and END risk in IS patients with AF. However, the pathophysiological mechanisms of these variants in MMP-9 gene effect on initial stroke severity and END risk remain unclear.
TA B L E 2 Multivariable logistic regression analysis of independent predictors for initial stroke severity (NIHSS score ≥16)   & Yang, 2012). MMP-9 is also involved in the generation of free radicals that result in atherosclerotic lesions and plaque rupture in carotid atherosclerotic plaques (Heo et al., 2011). The high concentration of MMP-9 in plasma or MMP-9 mRNA concentration within the acute phase of IS increases the risk of HT (Montaner et al., 2001), and is a predictor of poor outcomes and mortality in patients with IS (Graham et al., 2012). MMP-9 gene polymorphisms encode and regulate the transcription of MMP-9 protein, and are associated with concentration of MMP-9 in plasma (Blankenberg et al., 2003).
Several studies have investigated the association of MMP-9 gene polymorphisms with carotid atherosclerosis and the risk of IS or HT (Ho et al., 2015;Lin et al., 2012;Nie, Wang, & Tang, 2014;Yuan et al., 2013;Zhang, Cao, Xu, Li, & Xu, 2015). The rs3918242 polymorphism in the MMP-9 promoter may reduce the rate of transcription by inhibiting protein binding, which downregulates MMP-9 expression, and is associated with carotid atherosclerosis and increased IS risk (Lin et al., 2012 polymorphisms were associated with risk of spontaneous intracerebral hemorrhage (Ho et al., 2015). In this study, we found that the variants of rs3918242 and rs3787268 in MMP-9 gene were independent predictors for initial stroke severity, and rs1056628 AC/CC and rs3918242 CT/TT were independently associated with the higher risk of END after adjusting for covariates in IS patients with AF. Thus, we reason that the variants in MMP-9 gene could potentially provide these individuals with higher concentration of MMP-9, result in BBB destruction, apoptosis, and neuronal damage, thereby increasing the risk of END and stroke severity in these patients. However, further studies are needed to validate our findings in the future.
Despite novel findings, several limitations were noted in this study. First, the concentration of MMP-9 in plasma was not measured in this study, which may otherwise provide additional functional information to support our hypothesis. Second, although we genotyped multiple known functional variants of the MMP-9 gene, some rare functional variants were not investigated in this population. Moreover, as previously mentioned, oxygen free radical damage, inflammation reactions, MMP-2, MMP-3 may also play a role in BBB destruction and neuronal damage (Candelario-Jalil et al., 2009;Fatar et al., 2005;Heo et al., 2011;Hill et al., 2012;Kurzepa et al., 2014).
We did not measure the variants in oxygen free radical relevant genes, inflammation relevant genes, and MMP-2 and MMP-3 genes in this study. Thus, future studies involving a larger set of genetic variants must be conducted to evaluate the full extent of relevant genetic variants effect on stroke severity and END risk. Third, due to the limited sample size and six-center study, our findings must be confirmed with larger, multi-center studies. Finally, the main aim of this study was to investigate the association of MMP-9 gene polymorphisms with initial stroke severity and the risk of END in IS patients with AF. The other stroke subtypes were excluded in this study. Thus, we did not know whether there was an effect of MMP-9 gene polymorphisms on stroke severity and END risk in other stroke subtypes.

| CON CLUS ION
The incidence of moderate or severe stroke and END was very common in acute IS patients with AF. MMP-9 polymorphisms were independently associated with severe stroke and higher risk of END, and prestroke antithrombotic treatment was associated with less severe stroke and lower risk of END in patients with AF. Our findings may be useful to insight into the complex pathogenesis of stroke severity and END, prevent END, and decrease stroke severity in acute IS patients with AF. However, further studies are needed to confirm our findings in the future.

ACK N OWLED G M ENTS
This study was supported in part by grants from the Sichuan

CO N FLI C T O F I NTE R E S T
The authors declare that they have no competing interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.