Association of Major Histocompatibility Complex Class I Related Chain A/B Positive Microparticles with Acute Myocardial Infarction and Disease Severity

Background: Various cell types undergo activation and stress during atherosclerosis resulting in the development of acute myocardial infarction (AMI) in coronary artery disease (CAD). Major histocompatibility complex class I related chain A and B (MICA/B) can be expressed on the surface of activated and stressed cells and released into blood circulation in several forms including microparticles (MICA/B+ MPs) from various cell types. We aimed to investigate the association of these MICA/B+ MPs with the presence of AMI. Fifty-one AMI and 46 age-matched control subjects were recruited. Methods: Levels of MICA/B+ MPs derived from various parent cells including endothelial cells, platelets, monocytes, neutrophils, and T lymphocytes were determined by flow cytometry. Results: The levels and proportion of MICA/B+ MPs from all types of cell origin were significantly increased in AMI patients compared to those of the controls. A multivariate regression model showed an independent association between MICA/B+ MPs and AMI (OR = 11.6; 95% CI = 2.8, 47.3). Interestingly, based on the disease severity, we found that the levels of MICA/B+ MPs were significantly elevated in the ST-segment elevation myocardial infarction (STEMI) compared to the non-STEMI (NSTEMI) patients. Moreover, an independent association of MICA/B+ MPs with the occurrence of STEMI was also demonstrated (OR = 4.1; 95% CI = 1.5, 16.7). Conclusions: These results suggest that MICA/B+ MPs are associated with AMI and disease severity. They may act as mediators contributing to the pathological process of AMI. Alternatively, they are the results of various cell activations contributing to AMI.


Introduction
Major histocompatibility complex class I related chain A and B (MICA/B) are major ligands of the natural killer group 2, member D (NKG2D) activating receptor, which is normally expressed

Blood Samples
Blood samples were obtained from patients within 1-2 h after the onset of AMI. Three milliliters of blood were collected in 3.2% trisodium citrate at a ratio of 9:1 (blood:anticoagulant) and placed at room temperature prior to plasma separation. Platelet-free plasma (PFP) was prepared within 1 h after blood collection by double centrifugation. Platelet-rich plasma (PRP) was first prepared by centrifugation at 2500× g for 15 min at room temperature. The second centrifugation of PRP at 2500× g for 15 min at room temperature was performed to obtain PFP. Subsequently, the plasma was immediately stored at −80 • C until analysis. Three milliliters of blood were collected to prepare serum for measuring blood chemistry parameters and high-sensitivity C reactive protein (hs-CRP). Complete blood count was performed using 3 mL of ethylenediaminetetraacetic acid (EDTA) blood. PFP samples were thawed in a water bath at 37 • C prior to analysis. The experiment was separated into two reactions. In the first reaction, 50 µL of PFP were mixed with 5 µL of Annexin V-PE-Cy7, Alexa Fluor 488 mouse anti-human MICA/B, PE mouse anti-human CD144, APC mouse anti-human CD41a, and 50 µL of diluted Annexin V-binding buffer solution containing calcium, which is essential for binding between Annexin V and PS on MPs. In the second reaction, 50 µL of PFP was mixed with 5 µL of Annexin V-PE-Cy7, Alexa Fluor 488 mouse anti-human MICA/B, PE mouse anti-human CD14, APC mouse anti-human CD66b, and PerCP mouse anti-human CD3, and 50 µL of diluted Annexin V-binding buffer solution. The mixtures were subsequently incubated for 15 min at room temperature in the dark. Subsequently, phosphate buffered saline, pH 7.4, was added to 300 µL of each reaction. The mixtures were performed in a 5-mL polystyrene round-bottom tube (BD Bioscience, San Jose, CA, USA). The 123count eBeads counting beads (Thermo Fisher Scientifics, Waltham, MA, USA) were used to calculate the absolute MP count. Then, the mixtures were immediately analyzed on a FACS Canto II flow cytometer (BD Bioscience, San Jose, CA, USA). The size of MPs was defined using 0.93-µM polystyrene beads (Spherotect, Lake Forest, IL, USA). The levels of MICA/B + MPs were displayed as absolute numbers of MICA/B + MPs from various cell origins ( Figure 1) and were calculated using the following formula.
Absolute microparticle count (particles/(µL))= The result was also shown as the proportion of MICA/B + on each source of MPs, which was used to determine the intensity of cell stress and activation during the disease progression ( Figure 2). The result was also shown as the proportion of MICA/B + on each source of MPs, which was used to determine the intensity of cell stress and activation during the disease progression ( Figure 2).

Statistical Analysis
Data were analyzed using SPSS version 17 (SPSS Inc., Chicago, IL, USA). Normal distribution of continuous variables was tested by the one-sample Kolmogorov-Smirnov test. Continuous values that were not in normal distribution were logarithm transformed before mean comparison. Independent-sample t-test was used to compare the means of each variable between two groups. Differences of proportions between two groups were tested using a Chi-square test. Binary logistic regression analysis including univariate and multivariate analysis adjusted for potential confounders was applied to determine the association between MICA/B + MPs with AMI and disease severity. The

Statistical Analysis
Data were analyzed using SPSS version 17 (SPSS Inc., Chicago, IL, USA). Normal distribution of continuous variables was tested by the one-sample Kolmogorov-Smirnov test. Continuous values that were not in normal distribution were logarithm transformed before mean comparison. Independent-sample t-test was used to compare the means of each variable between two groups. Differences of proportions between two groups were tested using a Chi-square test. Binary logistic regression analysis including univariate and multivariate analysis adjusted for potential confounders was applied to determine the association between MICA/B + MPs with AMI and disease severity. The relationship between continuous variables was determined by Spearman correlation. Statistical significance was defined as p < 0.05.

Subject Characteristics
Demographic data of the study population are shown in Table 1. Ages of both groups were not significantly different whereas the proportions of male gender, diabetes mellitus, hypertension, dyslipidemia, smoking, and hs-CRP > 3 mg/L were significantly higher in patients than in the controls. Significant elevation of fasting blood sugar and hs-CRP was found in the AMI group whereas decreased levels of diastolic blood pressure, total cholesterol, and high-density lipoprotein cholesterol were observed. Demographic data of AMI patients are shown in Table 2. All parameters were not significantly different between NSTEMI and STEMI patients except the proportion of hypertension was higher in STEMI patients. The proportions of medications that can influence MP levels were also analyzed and revealed no significant differences between NSTEMI and STEMI patients ( Table 3). In addition, the effect of statin and angiotensin converting enzyme inhibitor/angiotensin receptor blocker (ACEI/ARB) on MICA/B + MP levels were evaluated. The results showed that MICA/B + MP levels between the patients who had taken and not taken those drugs were not significantly different ((statin vs. without statin, 21 Values are mean ± SD or n (%), FBS and hs-CRP are expressed as geometric mean ± geometric standard deviation. BMI: Body mass index; SBP: Systolic blood pressure; DBP: Diastolic blood pressure; LDL-C: Low density lipoprotein-cholesterol; HDL-C: High density lipoprotein-cholesterol; FBS: Fasting blood sugar; hs-CRP: High sensitivity-C reactive protein.

MICA/B + MPs' Analysis
Levels of MICA/B + MPs from all cell origins were significantly increased in the patients compared to those in the controls (Figures 1 and 3). Similarly, the proportions of MICA/B + MPs in each type of MPs were elevated in the patient group ( Figure 2). A multivariate regression model adjusted for the traditional cardiovascular risk factors including male gender, diabetes mellitus, hypertension, dyslipidemia, and smoking [18] showed an independent association of hs-CRP and MICA/B + MPs with AMI ( Figure 4). Interestingly, this association was increased after combining with hs-CRP [odds ratio (OR) = 30.2; 95% confidence interval (CI) = 4.5, 202.0]. In addition, MICA/B + MPs that originated from all parent cells were positively correlated with hs-CRP (Table 4). When the patients were divided into two subgroups according to the disease severity, the levels of MICA/B + MPs that originated from all cell types were increased in patients with STEMI compared to the NSTEMI group ( Figure 5). A multivariate regression model adjusted for hypertension, which was only a confounder associated with STEMI in this study, revealed the independent association of MICA/B + MPs with the STEMI occurrence ( Figure 6). The proportion of MICA/B expression on TMPs was significantly increased compared to other MPs (Figure 7). All types of MICA/B + MPs were positively correlated with high sensitivity-troponin T (hs-TnT) levels, a necrotic marker for myocardial infarction (Figure 8).        odds ratios included male gender, diabetes mellitus, hypertension, dyslipidemia, and smoking. MICA/B + MPs and hs-CRP were independently associated with AMI. * The level at 3 mg/L was used as the cutoff. ** The 50th percentile was used as the cutoff to distinguish high and low levels.

Discussion
The present study is the first report of the association between MICA/B + MPs considered as NKG2D ligands and AMI. MICA/B + MPs from diverse cell origins in AMI patients were significantly elevated compared to the disease controls. Moreover, MICA/B + MPs were an independent risk factor for AMI, and MICA/B + MPs was more strongly associated with AMI than hs-CRP ( Figure 4B,C). Synergic effect was demonstrated when MICA/B + MPs was combined with hs-CRP. Accordingly, it was clearly established that MICA/B + MPs were independently associated with the occurrence of AMI. In addition, the proportions of MICA/B + MPs in all cell types of MPs were elevated in AMI. This suggested that the degree of cell activation leading to MICA/B expression in AMI was greater than that in the controls. In addition, all types of MICA/B + MPs were positively correlated with hs-CRP, which is a biological marker of inflammatory diseases including atherosclerosis. It should be noted that the major MPs were from three sources, i.e., endothelial cells, monocytes, and neutrophils, suggesting the major contributions of activations of these cells in AMI.

Discussion
The present study is the first report of the association between MICA/B + MPs considered as NKG2D ligands and AMI. MICA/B + MPs from diverse cell origins in AMI patients were significantly elevated compared to the disease controls. Moreover, MICA/B + MPs were an independent risk factor for AMI, and MICA/B + MPs was more strongly associated with AMI than hs-CRP ( Figure 4B,C). Synergic effect was demonstrated when MICA/B + MPs was combined with hs-CRP. Accordingly, it was clearly established that MICA/B + MPs were independently associated with the occurrence of AMI. In addition, the proportions of MICA/B + MPs in all cell types of MPs were elevated in AMI. This suggested that the degree of cell activation leading to MICA/B expression in AMI was greater than that in the controls. In addition, all types of MICA/B + MPs were positively correlated with hs-CRP, which is a biological marker of inflammatory diseases including atherosclerosis. It should be noted that the major MPs were from three sources, i.e., endothelial cells, monocytes, and neutrophils, suggesting the major contributions of activations of these cells in AMI.
Atherosclerosis is the underlying feature of CAD and AMI and has been speculated as the most common stress-related disease. Classically, inflammation plays a critical role in the initiation and progression of stress-related diseases. Proinflammatory substances induced by stress include CRP, interleukin (IL) -6, and tumor necrosis factor α [8]. Increases in inflammatory substances have been demonstrated in AMI patients [19]. Similarly, high levels of CRP were found among AMI patients in the present study. Atherosclerotic lesions from early disease stages to advanced lesions contain MPs that probably result from apoptosis, activation, and stress of cells in the plaque [20]. As a result, under conditions of chronic activation and stress in various cells, MICA/B can be presented on the cell surface because NKG2D ligands including MICA/B are induced in response to stress stimuli [21]. Consequently, MPs that express MICA/B are released from activated and stress cells during all stages of the disease. Therefore, MICA/B + MPs may be markers of cell activation and stress. Moreover, they might be contributors to inflammation during atherosclerosis progression resulting from its role in stimulation of the NKG2D system.
Accordingly, the results of the present study revealed that MICA/B was present on all cell types and was subsequently released from cell surfaces as MICA/B + MPs. Interestingly, the proportion of MICA/B expression on TMPs was significantly increased compared to other MPs (p < 0.001), albeit the amount of MICA/B + TMPs was the smallest, suggesting the potency of T cells in the process. Alternatively, MICA/B + TMPs may be the results of acquiring of MICA/B binding to NKG2D at the immune synapses in T cell activation, similar to NK cells [22]. At present, it is well known that T lymphocytes are the major players of adaptive immunity participating in atherosclerosis leading to AMI [23]. An increase in T cell activation has been shown in AMI. T helper 1 (Th1) cells are the major cell subset of T cells contributing to atherosclerosis [24]. Production of a large amount of interferon (IFN) γ is the crucial feature of Th1 activation found in AMI. Moreover, CD4 T cells from AMI patients display an increased response to T cell receptor stimulation [25]. Unfortunately, we did not classify the MP sources of the T cell subpopulation in this study. Thus, we do not know the contribution of T cell subpopulations producing MPs in AMI. Expectedly, experimental targets for T lymphocyte modulation are required to relieve the inflammatory response in atherosclerosis [26]. Hence, these evidences confirm that T cell activation in AMI encourages MICA/B expression in the cells.
Circulating MICA/B can be presented as membrane-bound and soluble forms. Therefore, MICA/B + MPs may be included in the soluble MIC. Previous studies have shown that soluble MIC was associated with the occurrence and severity of diseases including cancer and inflammatory diseases [27][28][29]. Hence, MICA/B + MPs, which are supposed to be included in the detection of soluble MIC in a serum sample, were found to be associated with AMI and the severity of disease in this study.
Previously, extensive studies on NKG2D ligands were performed in cancer along with a few in inflammatory diseases including Crohn s disease and atherosclerosis [4,30]. Two reports confirming the association between MICA/B and atherosclerosis have been published. Firstly, Xia et al. [9] investigated the association between NKG2D ligands and diabetes mellitus and showed that type 2 diabetic patients had significantly higher levels of soluble MICA compared to healthy controls. Furthermore, they also revealed the presence of MICA/B in atherosclerotic aortic plaques of humans, where the ligands were expressed on macrophages and endothelial cells. Similarly, MICA/B + MMPs and EMPs were abundant in this study. Interestingly, the NKG2D system was correlated with the development and progression of atherosclerotic plaques in mouse models. In addition, the study demonstrated that inhibition of NKG2D function could relieve inflammation by reducing the levels of many cytokines including IL-6, IFN-γ, and IL-1β. Secondly, Ikeshita et al. [5] demonstrated that MICA/B appeared in atherosclerotic plaque and was expressed on the surface of macrophages. Interestingly, the study showed that MICA/B could be specifically present in advanced atherosclerotic lesions including large necrotic cores, thin fibrous caps, or hemorrhagic lesions but not in stable plaques.
One of our major findings notes that MICA/B + MPs from all cell origins were significantly increased in STEMI compared to NSTEMI patients, and that these parameters were independent risk factors for the occurrence of STEMI. Hypertension was the only risk factor used in the adjusted model because a significant difference in hypertension was observed between the NSTEMI and STEMI groups among the study subjects. It has been reported that levels of MPs were decreased by the effect of some vascular disease medications including statin, ACEI/ARB, aspirin, and clopidogrel [31]. Although the study subjects had taken these drugs, the proportion of NSTEMI and STEMI patients who were treated with these medications was not significantly different. Accordingly, the effect of medication on MP levels was not considered in this study.
In addition, MICA/B + MPs were positively correlated with hs-TnT levels. It was thus confirmed that patients with STEMI had more inflammatory conditions than NSTEMI. Proinflammatory mediators including hs-CRP, IL-6, and granzyme B were significantly elevated in STEMI compared to NSTEMI [32,33]. Similarly, the present study demonstrated higher levels of hs-CRP in STEMI patients compared to NSTEMI patients but did not reach a significant level (p = 0.091). Interestingly, we found that MICA/B + MPs from all cell types were positively correlated with the degree of AMI severity determined by the hs-TnT levels. The correlation between hs-TnT and inflammatory status in AMI examined by hs-CRP concentration has been investigated [34]. It is well known that TnT isoforms are extremely specific and sensitive to cardiac myocytes. Therefore, detection of hs-TnT in the blood stream is a highly specific marker for cardiac damage, resulting in different severities of myocardial injury in the AMI. Likewise, our results confirmed the correlation between hs-TnT and hs-CRP levels among AMI patients (r = 0.399, p = 0.005). Inflammatory status in AMI was found to be associated with MICA/B expression in various cells and correlated with the disease severity. Consequently, elevation of MICA/B + MPs could reflect the intensity of inflammation, leading to more severe disease. Nevertheless, the main limitation of this study is a cross-sectional study design that could not demonstrate a utilization of MICA/B + MPs as the early predictor of AMI symptom in CAD patients. A retrospective study in patients with stable CAD should be carried out.

Conclusions
To our knowledge, this is the first report investigating the association of MICA/B expressed on MPs from various cell origins in AMI. The results suggest that MICA/B + MPs are associated with AMI and disease severity. Accordingly, MICA/B + MPs could be mediators contributing to the pathological process of AMI. Alternatively, MICA/B + MPs could be the results of cell activations and apoptosis in AMI. Nevertheless, they could represent inflammatory markers indicating the severity of AMI. As a result, MICA/B + MPs could be further applied as an indicator of vulnerable plaques resulting in the symptoms of AMI and disease severity.