Abstract
Background: Chest pain is a typical presentation in the emergency department (ED), often due to acute coronary syndrome. Accurate and fast identification is crucial. The diagnosis and proper treatment of unstable angina (UA) can reduce the chance of acute myocardial infarction, and reduce high mortality and morbidity rates. However there is a lack of reliable and valid biomarkers in the diagnosis of UA. This study investigated the usefulness of circulating microRNAs (miRNAs) for differentiating UA from non-ischemic chest pain (NICP) in the ED. Methods The expressions of circulating miRNAs in patients with UA were evaluated relative to individuals with NICP (control subjects). Circulating miR-21, miR-25, miR-92a, miR-106b, miR-126* and miR-451 levels were measured in 98 patients with UA and 95 control subjects in the ED. To investigate the underlying functions of miRNAs in UA, bioinformatic analysis of validated miRNAs was conducted. Results: Circulating miRNAs were upregulated in UA compared with the control group. The combination of the modified HEART score (m-HS) and miR-25 (AUC 0.901, NRI 0.096) could better distinguish UA than m-HS alone. Bioinformatic analysis indicated that miRNAs may take part in platelet activation, cGMP-PKG signaling pathways etc. Conclusion: The circulating levels of miRNAs (miR-21, miR-25, miR-106b, miR-126*) are significantly higher in UA patients compared with patients with NICP, and the addition of the m-HS that combined ECG, age, risk factors and troponin is useful to detect or rule out UA. The associated signaling pathways are involved in the pathogenesis of vulnerable plaque.
Keywords: Acute coronary syndrome, unstable angina, microRNAs
Introduction
Chest pain is a common complaint of patients presenting in the emergency department (ED). Patients with chest pain account for 130 million ED visits annually worldwide, and chest pain is the second most frequent reason for ED visits (5.4%) [1]. Data from the General Practice Research Database indicates that ischemic heart disease is the most common cause of death during the 1-year follow-up , with mortality up to 36% [2]. Acute coronary syndrome (ACS) is a leading cause of cardiac chest pain, which includes acute myocardial infarction (AMI) and unstable angina (UA). Patients with suspected ACS are classified into 3 groups, the first collects patients with ACS, and then admitted, the second group with patients with diagnosis of non-ischemic chest pain (NICP), and finally the group with patients with suspected ACS who require observation. Patients in the latter group account for about 50.2% [3].
The assessment of patients with suspected ACS is based on clinical presentation, 12-lead electrocardiogram (ECG), and cardiac troponin [4]. Cardiac troponin is a specific and sensitive biomarker of cardiomyocyte injury. The rule-out algorithm of AMI needs re-test cardiac troponin sometimes [4]. Accurate diagnosis and effective treatment of UA can reduce the chance of AMI and greatly influence the outcome of the patient in the ED. However there is a lack of reliable and valid biomarkers in the diagnosis of UA.
MicroRNAs (miRNAs) are small non-coding RNAs that can regulate the expression of mRNAs by promoting mRNA degradation or inhibiting their function [5]. It is widely recognized that the expressions of specific miRNAs are associated with various cardiovascular diseases [6]. Some studies show a link between miRNAs and chest pain patients. However, these studies focused on an association between miRNAs and AMI patients with chest pain.
This study investigated the usefulness of circulating miRNAs for detecting UA.
Materials and methods
Study population
This study was a prospective observational study, with no intervention or treatment changes. The Ethics Committee of Peking University People’s Hospital, Beijing, China approved the research protocol. Informed consent was obtained from participating patients.
Between October 2015 and July 2016, 193 patients presented with chest pain to the ED of Peking University People’s Hospital and were eligible for the study. To be included in the study, patients were aged ≥ 18 years and agreed to medical examination and blood sample collection. Patients with any of the following were excluded: liver or renal impairment; anemia; stable angina; elevation of troponin and previous history of hematonosis; malignant tumor; heart failure; or autoimmune diseases.
All patients were apportioned to either a UA (n = 98) or NICP (n = 95) control group. UA was defined as typical chest pain (angina occurring at rest and prolonged usually > 20 min; new-onset angina within 1 month; increasing angina of longer duration, more frequent, or lower threshold), without elevation in troponin and with or without ECG changes indicative of ischemia. The NICP patients had chest pain, but not related to myocardial ischemia, and therefore no evidence of ACS, no elevation in troponin, and no ECG ST-elevation or depression.
Data collection
Each patient’s past medical history and chief complaint were collected from the patient and from the electronic medical record. Data regarding age, gender, blood pressure (BP), heart rate (HR), cardiovascular risk factors, blood sample, and ECG results were recorded for all patients.
Visser et al.’s [7] research used the HEART score (history, ECG, age, risk factors, and troponin) for diagnosing ACS in patients with chest pain presenting in the ED. In the present study, a modified HEART score (m-HS) was calculated for each patient. The m-HS consisted of the following components: ECG, age, risk factors, and troponin. The risk factors of a high m-HS score include hypertension, hypercholesterolemia, diabetes mellitus, obesity, cigarette smoking, and positive family history.
Methods
Peripheral venous plasma samples were collected during patient admission for detection of miRNAs (< 12 hours after symptom onset). The expressions of miRNAs were quantified in individual plasma samples from all patients using real-time PCR. If a patient was initially considered to have UA but experienced increased troponin later, the patient was excluded from the study.
Extraction of total RNA from plasma
The total RNA was extracted from plasma using a miRNeasy Mini Kit (Qiagen) in accordance with the manufacturer’s instructions. The RNA was used for reverse transcription (RT) reactions.
MiRNA measurements
The cDNA (i.e., the RT reaction products) was prepared using a TaqMan MicroRNA Reverse Transcription Kit in accordance with the introduction of the recommended protocol. The RT reaction was performed at 16°C for 30 min, and then 42°C for 30 min and 85°C for 5 min.
The RT reaction products were combined with the TagMan Universal PCR Master Mix in accordance with the instructions for the measurement of miRNAs. Subsequently, miRNAs were measured by real-time PCR using the cycling conditions 95°C for 10 min; and 40 cycles of 95°C for 15 s and 60°C for 1 min. Cel-miR-39 was used for normalization. The level of miRNA was expressed as 2-(CT[miRNA]-CT[miR-39]), and CT is the cycle threshold.
Bioinformatic analysis of miRNAs
The target genes of the miRNAs were assessed using three databases (miRanda, miRWalk and TargetScan). Enrichment analysis was performed to find the signaling pathways that were associated with the target genes, using DAVID (Database for Annotation, Visualization, and Integrated Discovery) bioinformatics resources and KEGG (Kyoto Encyclopedia of Genes and Genomes). Interactions between the miRNAs and signaling pathways were visualized (P < 0.05) using Cytoscape software (Version 3.4.0).
Statistical analysis
Continuous variables that were normally distributed are presented as mean ± standard deviation, and comparisons were performed with Student’s t-test. Continuous variables that were not normally distributed are presented as median ± interquartile range, and comparisons between groups employed the Wilcoxon rank-sum test. Categorical variables are presented as percentages, and the chi-squared test was used for the comparisons.
Spearman’s correlation analysis was performed to investigate associations be-tween miRNAs and risk factors. Multivariate logistic regression analysis was used to study the miRNA expressions after adjustments for risk factors. The receiver operating characteristic (ROC) curve was analyzed to evaluate the ability of miRNAs and the m-HS to detect UA. The net reclassification improvement (NRI) index was evaluated to determine the incremental influence of miRNAs for detecting UA.
Results
Characteristics of study participants
The study included 193 patients with an average age of 59.53 years, and 50.8% were men (Table 1). All participants were apportioned to either a UA group (n = 98) or a control group (NICP group, n = 95). Between the 2 groups there were differences in age, hypertension, hyperlipidemia, diabetes mellitus, and red blood cells. The red blood cell levels were within the normal range in the participants of both two groups.
Table 1.
Baseline characteristics of patients with UA and NICP*
| Total population | NICP | UA | P | |
|---|---|---|---|---|
| Subjects, n | 193 | 95 | 98 | — |
| Age, y | 59.53 ± 17.38 | 51.01 ± 17.84 | 68.14 ± 11.82 | 0.000 |
| Male, % | 50.80 | 50.50 | 51.00 | 0.945 |
| SBP, mmHg | 143.68 ± 21.51 | 141.09 ± 22.08 | 146.09+20.83 | 0.165 |
| DBP, mmHg | 77.67 ± 14.62 | 79.17 ± 15.39 | 76.27 ± 13.83 | 0.237 |
| Heart rate, bpm | 79.61 ± 16.31 | 80.93 ± 13.95 | 78.30 ± 18.35 | 0.329 |
| BMI, kg/m2 | 24.70 ± 3.65 | 24.68 ± 3.66 | 24.71 ± 3.67 | 0.962 |
| Hypertension, % | 57.7 | 39.2 | 76.0 | 0.000 |
| Hyperlipidemia, % | 43.6 | 28.4 | 58.7 | 0.000 |
| Diabetes mellitus, % | 27.5 | 13.5 | 41.3 | 0.000 |
| Current smoking, % | 19.6 | 16.2 | 23.0 | 0.300 |
| White blood cell, ×109/L | 7.15 ± 2.38 | 7.11 ± 2.18 | 7.19 ± 2.57 | 0.845 |
| Red blood cell, ×1012/L | 4.52 ± 0.55 | 4.63 ± 0.53 | 4.43 ± 0.56 | 0.043 |
| Blood platelet, ×109/L | 225.69 ± 60.81 | 233.74 ± 55.16 | 218.47 ± 65.01 | 0.155 |
| Cr, µmol/L | 68.71 ± 19.99 | 65.33 ± 13.25 | 70.58 ± 22.76 | 0.208 |
| eGFR, mL/min/1.73 m2 | 88.59 ± 18.92 | 93.38 ± 18.41 | 85.93 ± 18.81 | 0.058 |
| ALT, µ/L | 17.00 (13.00, 30.50) | 19.50 (13.00, 27.25) | 17.00 (12.00, 32.00) | 0.865 |
| AST, µ/L | 20.00 (17.00, 25.50) | 19.50 (16.50, 24.25) | 20.00 (17.00, 27.00) | 0.602 |
| TC, mmol/L | 4.39 ± 1.14 | 4.78 ± 1.13 | 4.28 ± 1.13 | 0.183 |
| TG, mmol/L | 1.67 ± 0.94 | 1.32 ± 0.65 | 1.77 ± 0.99 | 0.142 |
| HDL-C, mmol/L | 1.19 ± 0.32 | 1.33 ± 0.28 | 1.15 ± 0.32 | 0.099 |
| LDL-C, mmol/L | 2.61 ± 0.84 | 2.94 ± 0.91 | 2.51 ± 0.80 | 0.128 |
| D-dimer, ng/mL | 105.00 (48.75, 222.25) | 97.50 (47.25, 154.25) | 118.50 (49.50, 238.75) | 0.349 |
| Cardiac troponin, ng/mL | 0.00 (0.00, 0.00) | 0.00 (0.00, 0.00) | 0.00 (0.00, 0.00) | 1.000 |
| CK-MB, ng/mL | 0.80 (0.00, 2.33) | 0.00 (0.00, 1.80) | 1.30 (0.00, 2.50) | 0.052 |
| Myo, ng/mL | 46.02 ± 3.10 | 44.51 ± 2.98 | 47.21 ± 3.21 | 0.694 |
Data are shown as mean ± standard deviation, percentages (%) or median (interquartile range).
Abbreviations: BMI, body mass index; DBP, diastolic blood pressure; eGFR, estimated glomerular filtration rate; NICP, non-ischemic chest pain; SBP, systolic blood pressure; UA, unstable angina.
Levels of candidate miRNAs
Six miRNAs (miR-21, miR-25, miR-92a, miR-106b, miR-126* and miR-451) were selected for validation, based on relevant literature. Real-time PCR was applied to evaluate the levels of miRNAs. The plasma levels of miR-21, miR-25, miR-92a, miR-106b, miR-126* and miR-451 were higher in the UA group than in the control group (Figure 1; Table 2).
Figure 1.
Plasma miRNAs levels in the UA and control group (NICP). The histograms show the expression levels of miR-21, miR-25, miR-92a, miR-106b, miR-126* and miR-451 measured by real-time PCR. UA, unstable angina; NICP, non-ischemic chest pain; *P < 0.05.
Table 2.
Expression levels of plasma miRNAs in UA relative to control
| UA/control ratio | P | |
|---|---|---|
| miR-21 | 2.020 | 0.000 |
| miR-25 | 2.097 | 0.006 |
| miR-92a | 1.611 | 0.000 |
| miR-106b | 1.540 | 0.000 |
| miR-126-5p | 1.716 | 0.000 |
| miR-451 | 1.678 | 0.000 |
Associations between the miRNAs and cardiovascular risk factors
To identify whether the miRNAs were associated with the known cardiovascular risk factors (age, hypertension, hyperlipidemia, diabetes and smoking), Spearman’s rank correlation coefficients between the levels of miRNAs and risk factors were calculated (Table 3). The results indicated that miRNAs correlated with each other, while the miRNAs were independent from other risk factors such as age, hypertension, hyperlipidemia, diabetes and smoking.
Table 3.
Spearman’s correlation analysis between the miRNAs and cardiac risk factors
| miR-21 | miR-25 | miR-92a | miR-106b | miR-126* | miR-451 | |
|---|---|---|---|---|---|---|
| miR-21 | — | — | — | — | — | — |
| miR-25 | 0.758* | — | — | — | — | — |
| miR-92a | 0.761* | 0.952* | — | — | — | — |
| miR-106b | 0.717* | 0.881* | 0.864* | — | — | — |
| miR-126* | 0.867* | 0.680* | 0.699* | 0.742* | — | — |
| miR-451 | 0.647* | 0.856* | 0.879* | 0.899* | 0.643* | — |
| Age | 0.163 | 0.050 | 0.023 | -0.067 | 0.074 | -0.022 |
| Hypertension | 0.125 | 0.015 | 0.009 | -0.008 | 0.063 | 0.008 |
| Hyperlipidemia | 0.050 | 0.097 | 0.096 | 0.015 | 0.111 | 0.044 |
| Diabetes mellitus | 0.094 | 0.084 | 0.068 | 0.030 | 0.055 | 0.115 |
| Current smoking | 0.008 | 0.060 | 0.036 | 0.090 | 0.021 | 0.054 |
P < 0.05.
r = Spearman’s correlation coefficient.
Multivariate logistic regression analysis
Multivariate logistic regression analysis was adjusted for age, hypertension, hyperlipidemia, and diabetes (Table 4). After the adjustment, the levels of miRNA (miR-21, miR-25, miR-92a, miR-106b, miR-126* and miR-451) were still independently associated with UA. The odds ratios for the six miRNAs ranged from 2.036 to 6.354 after adjustment.
Table 4.
Multivariate logistic regression analysis for the risk of UA*
| B | SE | Wald | P value | OR | 95% CI | |
|---|---|---|---|---|---|---|
| miR-21 | 1.849 | 0.504 | 13.436 | 0.000 | 6.354 | 2.364-17.080 |
| miR-25 | 0.889 | 0.318 | 7.784 | 0.005 | 2.432 | 1.303-4.539 |
| miR-92a | 0.748 | 0.297 | 6.326 | 0.012 | 2.112 | 1.179-3.781 |
| miR-106b | 0.848 | 0.285 | 8.873 | 0.003 | 2.334 | 1.336-4.077 |
| miR-126* | 1.004 | 0.373 | 7.237 | 0.007 | 2.728 | 1.313-5.669 |
| miR-451 | 0.711 | 0.277 | 6.585 | 0.010 | 2.036 | 1.183-3.503 |
The model included age, hypertension, hyperlipidemia, and diabetes.
OR = odds ratio, CI = confidence interval.
The diagnostic accuracy of the miRNAs
ROC curve analyses were performed to evaluate the diagnostic accuracy of the miRNAs (Tables 5, 6). The results showed that the area under the ROC curve (AUC) for miR-21 was 0.744 (P < 0.05). Combining miR-21 and miR-106b increased the power, with an AUC of 0.759 (Figure 2).
Table 5.
ROC curve analysis of the miRNAs
| AUC | P | 95% CI | Cut-off | Sensitivity | Specificity | |
|---|---|---|---|---|---|---|
| miR-21 | 0.744 | 0.000 | 0.666-0.822 | 1.62 | 0.506 | 0.960 |
| miR-25 | 0.649 | 0.000 | 0.571-0.727 | 1.64 | 0.412 | 0.872 |
| miR-92a | 0.620 | 0.004 | 0.541-0.699 | 1.91 | 0.286 | 0.958 |
| miR-106b | 0.573 | 0.081 | 0.490-0.655 | 2.18 | 0.265 | 0.989 |
| miR-126-5p | 0.653 | 0.000 | 0.574-0.732 | 2.03 | 0.298 | 1.000 |
| miR-451 | 0.581 | 0.055 | 0.498-0.663 | 2.03 | 0.278 | 0.989 |
Table 6.
ROC curve analysis of the combinations of miRNAs
| AUC | P | 95% CI | Cut-off | Sensitivity | Specificity | ||
|---|---|---|---|---|---|---|---|
| miR-21 | 0.744 | 0.000 | 0.666-0.822 | 1.62 | 0.506 | 0.960 | |
| +miR-25 | 0.747 | 0.000 | 0.669-0.826 | 1.60 | 0.513 | 0.973 | |
| +miR-92a | 0.749 | 0.000 | 0.671-0.826 | 1.49 | 0.494 | 0.973 | |
| +miR-106b | 0.759 | 0.000 | 0.684-0.835 | 1.30 | 0.506 | 0.960 | |
| +miR-126-5p | 0.738 | 0.000 | 0.657-0.818 | 1.31 | 0.480 | 0.972 | |
| +miR-451 | 0.742 | 0.000 | 0.663-0.821 | 1.63 | 0.513 | 0.946 | |
| miR-25 | 0.649 | 0.000 | 0.571-0.727 | 1.64 | 0.412 | 0.872 | |
| +miR-92a | 0.654 | 0.000 | 0.576-0.732 | 0.91 | 0.412 | 0.915 | |
| +miR-106b | 0.655 | 0.000 | 0.578-0.733 | 1.39 | 0.299 | 0.979 | |
| +miR-126-5p | 0.674 | 0.000 | 0.597-0.752 | 4.20 | 0.538 | 0.780 | |
| +miR-451 | 0.655 | 0.000 | 0.577-0.733 | 1.60 | 0.344 | 0.946 | |
| miR-92a | 0.620 | 0.004 | 0.541-0.699 | 1.91 | 0.286 | 0.958 | |
| +miR-106b | 0.616 | 0.005 | 0.536-0.695 | 1.80 | 0.408 | 0.863 | |
| +miR-126-5p | 0.656 | 0.000 | 0.578-0.734 | 11.11 | 0.298 | 0.989 | |
| +miR-451 | 0.614 | 0.006 | 0.535-0.694 | 2.33 | 0.316 | 0.947 | |
| miR-106b | 0.573 | 0.081 | 0.490-0.655 | 2.18 | 0.265 | 0.989 | |
| +miR-126-5p | 0.657 | 0.000 | 0.579-0.736 | 1.61 | 0.319 | 0.989 | |
| +miR-451 | 0.586 | 0.040 | 0.504-0.668 | 5.78 | 0.289 | 0.989 | |
| miR-126-5p | 0.653 | 0.000 | 0.574-0.732 | 2.03 | 0.298 | 1.000 | |
| +miR-451 | 0.656 | 0.000 | 0.577-0.735 | 2.08 | 0.298 | 1.000 | |
| miR-451 | 0.581 | 0.055 | 0.498-0.663 | 2.03 | 0.278 | 0.989 |
Figure 2.
ROC curve of miRNAs. AUCs: miR-21, 0.744 (0.666-0.822); miR-25, 0.649 (0.571-0.727); miR-106b, 0.573 (0.490-0.655); miR-21+miR-25, 0.747 (0.669-0.826); miR-21+miR-106b, 0.759 (0.684-0.835).
In the ROC curve analysis for discriminating the UA patients, the AUC for the m-HS alone was 0.874 (95% CI 0.817-0.930). When miR-25 was added to the m-HS, the AUC increased to 0.901 (95% CI 0.853-0.949), and the NRI was positive (0.096). A significant positive NRI (0.079) was also observed by adding miR-126* to m-HS (Table 7; Figure 3).
Table 7.
Reclassification for the presence of UA after addition of miRNAs to the m-HS
| AUC | P | 95% CI | Cut-off | Sensitivity | Specificity | NRI | ||
|---|---|---|---|---|---|---|---|---|
| m-HS | 0.874 | 0.000 | 0.817-0.930 | 3.50 | 0.803 | 0.860 | — | |
| +miR-21 | 0.928 | 0.000 | 0.885-0.970 | 5.43 | 0.884 | 0.829 | -0.120 | |
| +miR-25 | 0.901 | 0.000 | 0.853-0.949 | 4.24 | 0.920 | 0.800 | 0.096 | |
| +miR-92a | 0.895 | 0.000 | 0.846-0.945 | 4.12 | 0.882 | 0.814 | 0.065 | |
| +miR-106b | 0.900 | 0.000 | 0.852-0.948 | 4.11 | 0.895 | 0.837 | 0.138 | |
| +miR-126-5p | 0.911 | 0.000 | 0.866-0.955 | 4.70 | 0.861 | 0.869 | 0.079 | |
| +miR-451 | 0.900 | 0.000 | 0.851-0.949 | 4.09 | 0.867 | 0.847 | 0.084 | |
| +miR-21+miR-25 | 0.914 | 0.000 | 0.866-0.962 | 5.46 | 0.870 | 0.826 | -0.132 | |
| +miR-21+miR-92a | 0.928 | 0.000 | 0.886-0.970 | 5.57 | 0.870 | 0.871 | -0.050 | |
| +miR-21+miR-106b | 0.928 | 0.000 | 0.885-0.970 | 5.53 | 0.853 | 0.870 | -0.088 |
UA, unstable angina; m-HS, modified HEART score; NRI, net reclassification improvement.
Figure 3.
ROC curves of m-HS. AUCs: m-HS, 0.874 (0.817-0.930); m-HS+miR-25, 0.901 (0.853-0.949); m-HS+miR-126*, 0.911 (0.866-0.955).
Bioinformatic analysis of miRNAs
Five miRNAs (miR-21, miR-25, miR-92a, miR-106b and miR-126*) had common target genes in the 3 databases (miRanda, miRWalk and TargetScan). Consequently, using the DAVID bioinformatics resource, signaling pathways (P < 0.05) associated with target genes in the KEGG database were found, and combined with the function of the signaling pathways, we found that the targeted genes take part in platelet activation, cGMP-PKG, AMPK, PI3K-Akt, FoxO, and MAPK pathways (Table 8). Finally, the Cytoscape software revealed miRNA-target gene (Figure 4) and miRNA-target gene-signaling pathway interaction networks (Figure 5).
Table 8.
Enriched KEGG pathways associated with target genes of miRNAs
| Target gene | Signaling pathway | P | |
|---|---|---|---|
| miR-21 | ATP2B4, CREB5, GTF2I, etc. | cGMP-PKG | 0.0026 |
| FASLG, GNG12, MKNK2, etc. | MAPK | 0.0024 | |
| ATM, FASLG, SOS2, etc. | FoxO | 0.00042 | |
| HMGCR, PFKFB2, CREB5, etc. | AMPK | 0.028 | |
| miR-25 | ADCY3, ADCY9, COL1A2, etc. | Platelet activation | 0.029 |
| ATP1B2, ATP2B4, ATP2A2, etc. | cGMP-PKG | 0.003 | |
| MCL1, BCL2, FASLG, etc. | PI3K-Akt | 0.0042 | |
| HMGCR, RAB14, CREB1, etc. | AMPK | 0.0079 | |
| ATP1B2, ATP2B4, BCL2, etc. | Adrenergic signaling in cardiomyocytes | 0.0037 | |
| miR-92a | ATP1B2, ATP2B1, IRS2, etc. | cGMP-PKG | 0.00058 |
| MCL1, BCL2L11, BCL2, etc. | PI3K-Akt | 0.0042 | |
| HMGCR, PFKFB2, INSR, etc. | AMPK | 0.0017 | |
| BCL2L11, FASLG, KLF2, etc. | FoxO | 0.0026 | |
| ATP1B2, ATP2B1, ATP2B4, etc. | Adrenergic in cardiomyocytes | 0.028 | |
| miR-106b | PDPK1, AKT3, ATM, etc. | FoxO | 0.000079 |
| miR-126-5p | PDPK1, AKT3, MCL1, etc. | PI3K-Akt | 0.000059 |
Figure 4.
The miRNAs-target gene interaction network visualized by Cytoscape software. The yellow nodes indicate miRNAs. The brown nodes indicate target genes. The lines indicate interactions between the miRNAs and target genes.
Figure 5.
The miRNAs-signaling pathway interaction network visualized by Cytoscape software. The yellow nodes indicate miRNAs. The blue nodes indicate signaling pathways. The lines indicate target genes. The target genes establish linkages between miRNAs and signaling pathways.
Discussion
Early detection of ACS in the ED is crucial for improving the quality of patient care and reducing medical costs. For this reason, an association between the overexpression of some miRNAs and myocardial infarction has been the subject of several studies. Some research has indicated a link between miRNAs and patients with cardiogenic chest pain in the ED [8]. For example, levels of miR-221-3p were elevated in AMI patients who were admitted to the ED with chest pain [9], and miR-423-5p and miR-30d levels were also elevated in ST-elevation myocardial infarction (STEMI) patients with chest pain on presentation to the ED [10]. Circulating miR-133a was upregulated in UA and AMI patients with chest pain admitted to the ED [8]. In UA and MI patients, miR-486 and miR-92a prevailed in cases with high levels of high-density lipoprotein cholesterol, and their levels discriminate between vulnerable and stable coronary artery disease patients [11]. Serum miRNA-210 and miRNA-499 were associated with non-STEMI and UA patients with chest pain in the ED [12]; miR-122-5p may be a useful biomarker for diagnosis of AMI patients [13]. Levels of miR-208b, miR-320a, and miR-499 were significantly higher in AMI patients [14].
There has been less interest and research that focused on miRNAs in UA, although early diagnosis and treatment is essential to avoid myocardial infarction and related complications. The present study draws attention to the role of six miRNAs and their elevation in UA. This may be essential in distinguishing patients with UA from those with NICP.
Previous studies have confirmed that miRNAs are involved in the different pathologies of coronary heart disease [6,15]. The expression of miR-21 is upregulated after injury of large vessels, and recovery to normal levels can prevent restenosis [16]. Overexpression of miR-25 can protect cardiomyocytes from oxidative damage [17]. The miR-106b~25 cluster participates in the regulation of post-ischemic neovascularization [18]. Hypoxia can activate of miR-92a in endothelial cells that can block angiogenesis [19]. MiR-126 is also rich in endothelial cells that can influence susceptibility to atherosclerosis by regulating endothelial cell function [20]. miR-451 can regulate the proliferation and differentiation of erythrocytes under oxidative stress [21].
In the present study, to evaluate the underlying function of miRNAs in UA, a complete bioinformatic analysis was performed of validated miRNAs. It may be that miRNAs are involved in multiple signaling pathways by regulating multiple target genes, and the interrelations of miRNAs and signaling pathways are complex. MiRNAs may take part in platelet activation, cGMP-PKG, AMPK, PI3K-Akt, FoxO, and the MAPK signaling pathways. The cGMP-PKG pathway has a cardioprotective role in ischemia and reperfusion injury [22]. Nitrous oxide and its secondary messenger cGMP are important in the process [22]. The AMPK signaling pathway can promote cardiac protection against ischemia [23,24]. The PI3K-Akt signaling pathway may underlie thrombosis in ACS [25]. The activity of FoxO is important in maintaining endothelial quiescence [26].
Results of the present study show that levels of miR-21, miR-25, miR-92a, miR-106b, miR-126* and miR-451 are elevated in patients with UA. Compared with the NICP control group, the levels of the six miRNAs remained higher after adjustment for the risk factors (age, hypertension, hyperlipidemia, and diabetes). Moreover, this study highlights how the combination of these miRNAs and m-HS can increase the specificity and sensitivity for detecting patients with UA. The HEART score has been used to estimate the risk of patients with chest pain in the ED [27], and the HEART score can facilitate accurate diagnosis and improve therapeutic choices [7] [28]. The score system has five components: past medical history, ECG, age, risk factors, and troponin [7]. In the present study, objective index (ECG, age, risk factors, and troponin) were selected to define an m-HS. The AUC of m-HS and miR-25 were 0.874 and 0.649 respectively. The combination of miR-25 and m-HS (AUC 0.901, NRI 0.096) can better differentiate UA than m-HS alone.
In summary, this study provides evidence that the expression profiles of circulating miRNAs may have an essential role in the diagnosis of UA among patients with chest pain in the ED. This evidence, which is stronger in association with m-HS, is an interesting starting point that needs further research and validation, and is an important advancement toward providing better management of chest pain of patients in the ED.
Acknowledgements
This work was supported by the National Natural Science Foundation of China (No. 81470473, 81600340), the Capital Health Research and Development of Special Fund (No. 2016-2-4083), and Beijing Science and Technology Major Project (No. D141100003014002).
Disclosure of conflict of interest
None.
References
- 1.Rybicki FJ, Udelson JE, Peacock WF, Goldhaber SZ, Isselbacher EM, Kazerooni E, Kontos MC, Litt H, Woodard PK. 2015 ACR/ACC/AHA/AATS/ACEP/ASNC/NASCI/SAEM/SCCT/SCMR/SCPC/SNMMI/STR/STS appropriate utilization of cardiovascular imaging in emergency department patients with chest pain: a joint document of the American College of Radiology Appropriateness Criteria Committee and the American College of Cardiology Appropriate Use Criteria Task Force. J Am Coll Cardiol. 2016;67:853–79. doi: 10.1016/j.jacc.2015.09.011. [DOI] [PubMed] [Google Scholar]
- 2.Ruigómez A, Rodríguez LA, Wallander MA, Johansson S, Jones R. Chest pain in general practice: incidence, comorbidity and mortality. Fam Pract. 2006;23:167–74. doi: 10.1093/fampra/cmi124. [DOI] [PubMed] [Google Scholar]
- 3.Skinner JS, Smeeth L, Kendall JM, Adams PC, Timmis A Chest Pain Guideline Development Group. NICE guidance: chest pain of recent onset: assessment and diagnosis of recent onset chest pain or discomfort of suspected cardiac origin. Heart. 2010;96:974–8. doi: 10.1136/hrt.2009.190066. [DOI] [PubMed] [Google Scholar]
- 4.Roffi M, Patrono C, Collet JP, Mueller C, Valgimigli M, Andreotti F, Bax JJ, Borger MA, Brotons C, Chew DP, Gencer B, Hasenfuss G, Kjeldsen K, Lancellotti P, Landmesser U, Mehilli J, Mukherjee D, Storey RF, Windecker S, Baumgartner H, Gaemperli O, Achenbach S, Agewall S, Badimon L, Baigent C, Bueno H, Bugiardini R, Carerj S, Casselman F, Cuisset T, Erol Ç, Fitzsimons D, Halle M, Hamm C, Hildick-Smith D, Huber K, Iliodromitis E, James S, Lewis BS, Lip GY, Piepoli MF, Richter D, Rosemann T, Sechtem U, Steg PG, Vrints C, Luis Zamorano J Management of Acute Coronary Syndromes in Patients Presenting without Persistent ST-Segment Elevation of the European Society of Cardiology. 2015 ESC Guidelines for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation: task force for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation of the European Society of cardiology (ESC) Eur Heart J. 2016;37:267–315. doi: 10.1093/eurheartj/ehv320. [DOI] [PubMed] [Google Scholar]
- 5.Ambros V. The functions of animal microRNAs. Nature. 2004;431:350–5. doi: 10.1038/nature02871. [DOI] [PubMed] [Google Scholar]
- 6.Small EM, Olson EN. Pervasive roles of microRNAs in cardiovascular biology. Nature. 2011;469:336–42. doi: 10.1038/nature09783. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Visser A, Wolthuis A, Breedveld R, ter Avest E. HEART score and clinical gestalt have similar diagnostic accuracy for diagnosing ACS in an unselected population of patients with chest pain presenting in the ED. Emerg Med J. 2015;32:595–600. doi: 10.1136/emermed-2014-203798. [DOI] [PubMed] [Google Scholar]
- 8.Ke-Gang J, Zhi-Wei L, Xin Z, Jing W, Ping S, Xue-Jing H, Hong-Xia T, Xin T, Xiao-Cheng L. Evaluating diagnostic and prognostic value of plasma miRNA133a in acute chest pain patients undergoing coronary angiography. Medicine (Baltimore) 2016;95:e3412. doi: 10.1097/MD.0000000000003412. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Coskunpinar E, Cakmak HA, Kalkan AK, Tiryakioglu NO, Erturk M, Ongen Z. Circulating miR-221-3p as a novel marker for early prediction of acute myocardial infarction. Gene. 2016;591:90–6. doi: 10.1016/j.gene.2016.06.059. [DOI] [PubMed] [Google Scholar]
- 10.Eryılmaz U, Akgüllü Ç, Beşer N, Yıldız Ö, Kurt Ömürlü İ, Bozdoğan B. Circulating microRNAs in patients with ST-elevation myocardial infarction. Anatol J Cardiol. 2016;16:392–6. doi: 10.5152/AnatolJCardiol.2015.6603. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Niculescu LS, Simionescu N, Sanda GM, Carnuta MG, Stancu CS, Popescu AC, Popescu MR, Vlad A, Dimulescu DR, Simionescu M, Sima AV. MiR-486 and miR-92a identified in circulating HDL discriminate between stable and vulnerable coronary artery disease patients. PLoS One. 2015;10:e0140958. doi: 10.1371/journal.pone.0140958. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Shalaby SM, El-Shal AS, Shoukry A, Khedr MH, Abdelraheim N. Serum miRNA-499 and miRNA-210: a potential role in early diagnosis of acute coronary syndrome. IUBMB Life. 2016;68:673–82. doi: 10.1002/iub.1529. [DOI] [PubMed] [Google Scholar]
- 13.Yao XL, Lu XL, Yan CY, Wan QL, Cheng GC, Li YM. Circulating miR-122-5p as a potential novel biomarker for diagnosis of acute myocardial infarction. Int J Clin Exp Pathol. 2015;8:16014–9. [PMC free article] [PubMed] [Google Scholar]
- 14.Devaux Y, Mueller M, Haaf P, Goretti E, Twerenbold R, Zangrando J, Vausort M, Reichlin T, Wildi K, Moehring B, Wagner DR, Mueller C. Diagnostic and prognostic value of circulating microRNAs in patients with acute chest pain. J Intern Med. 2015;277:260–71. doi: 10.1111/joim.12183. [DOI] [PubMed] [Google Scholar]
- 15.Wang R, Dong LD, Meng XB, Shi Q, Sun WY. Unique MicroRNA signatures associated with early coronary atherosclerotic plaques. Biochem Biophys Res Commun. 2015;464:574–9. doi: 10.1016/j.bbrc.2015.07.010. [DOI] [PubMed] [Google Scholar]
- 16.Ji R, Cheng Y, Yue J, Yang J, Liu X, Chen H, Dean DB, Zhang C. MicroRNA expression signature and antisense-mediated depletion reveal an essential role of microRNA in vascular neointimal lesion formation. Circ Res. 2007;100:1579–1588. doi: 10.1161/CIRCRESAHA.106.141986. [DOI] [PubMed] [Google Scholar]
- 17.Pan L, Huang BJ, Ma XE, Wang SY, Feng J, Lv F, Liu Y, Liu Y, Li CM, Liang DD, Li J, Xu L, Chen YH. MiR-25 protects cardiomyocytes against oxidative damage by targeting the mitochondrial calcium uniporter. Int J Mol Sci. 2015;16:5420–5433. doi: 10.3390/ijms16035420. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Semo J, Sharir R, Afek A, Avivi C, Barshack I, Maysel-Auslender S, Krelin Y, Kain D, Entin-Meer M, Keren G, George J. The 106b~25 microRNA cluster is essential for neovascularization after hindlimb ischaemia in mice. Eur Heart J. 2014;35:3212–23. doi: 10.1093/eurheartj/eht041. [DOI] [PubMed] [Google Scholar]
- 19.Bonauer A, Carmona G, Iwasaki M, Mione M, Koyanagi M, Fischer A, Burchfield J, Fox H, Doebele C, Ohtani K, Chavakis E, Potente M, Tjwa M, Urbich C, Zeiher AM, Dimmeler S. MicroRNA-92a controls angiogenesis and functional recovery of ischemic tissues in mice. Science. 2009;324:1710–3. doi: 10.1126/science.1174381. [DOI] [PubMed] [Google Scholar]
- 20.Harris TA, Yamakuchi M, Ferlito M, Mendell JT, Lowenstein CJ. MicroRNA-126 regulates endothelial expression of vascular cell adhesion molecule 1. Proc Natl Acad Sci U S A. 2008;105:1516–21. doi: 10.1073/pnas.0707493105. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Rasmussen KD, Simmini S, Abreu-Goodger C, Bartonicek N, Di Giacomo M, Bilbao-Cortes D, Horos R, Von Lindern M, Enright AJ, O’Carroll D. The miR-144/451 locus is required for erythroid homeostasis. J Exp Med. 2010;207:1351–8. doi: 10.1084/jem.20100458. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Fiedler B, Feil R, Hofmann F, Willenbockel C, Drexler H, Smolenski A, Lohmann SM, Wollert KC. cGMP-dependent protein kinase type I inhibits TAB1-p38 mitogen-activated protein kinase apoptosis signaling in cardiac myocytes. J Biol Chem. 2006;281:32831–40. doi: 10.1074/jbc.M603416200. [DOI] [PubMed] [Google Scholar]
- 23.Yang H, Sun W, Quan N, Wang L, Chu D, Cates C, Liu Q, Zheng Y, Li J. Cardioprotective actions of Notch1 against myocardial infarction via LKB1-dependent AMPK signaling pathway. Biochem Pharmacol. 2016;108:47–57. doi: 10.1016/j.bcp.2016.03.019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Baron SJ, Li J, Russell RR 3rd, Neumann D, Miller EJ, Tuerk R, Wallimann T, Hurley RL, Witters LA, Young LH. Dual mechanisms regulating AMPK kinase action in the ischemic heart. Circ Res. 2005;96:337–45. doi: 10.1161/01.RES.0000155723.53868.d2. [DOI] [PubMed] [Google Scholar]
- 25.Wang B, Peng Y, Dong J, Lin J, Wu C, Su Y, Fang H, Wang L, Huang K, Li D. Human platelets express functional thymic stromal lymphopoietin receptors: a potential role in platelet activation in acute coronary syndrome. Cell Physiol Biochem. 2013;32:1741–50. doi: 10.1159/000356608. [DOI] [PubMed] [Google Scholar]
- 26.Oellerich MF, Potente M. FOXOs and sirtuins in vascular growth, maintenance, and aging. Circ Res. 2012;110:1238–51. doi: 10.1161/CIRCRESAHA.111.246488. [DOI] [PubMed] [Google Scholar]
- 27.Backus BE, Six AJ, Kelder JC, Bosschaert MA, Mast EG, Mosterd A, Veldkamp RF, Wardeh AJ, Tio R, Braam R, Monnink SH, van Tooren R, Mast TP, van den Akker F, Cramer MJ, Poldervaart JM, Hoes AW, Doevendans PA. A prospective validation of the HEART score for chest pain patients at the emergency department. Int J Cardiol. 2013;168:2153–8. doi: 10.1016/j.ijcard.2013.01.255. [DOI] [PubMed] [Google Scholar]
- 28.Six AJ, Backus BE, Kelder JC. Chest pain in the emergency room: value of the HEART score. Neth Heart J. 2008;16:191–6. doi: 10.1007/BF03086144. [DOI] [PMC free article] [PubMed] [Google Scholar]