CircSLC8A1 and circNFIX can be used as auxiliary diagnostic markers for sudden cardiac death caused by acute ischemic heart disease

Abstract

Sudden cardiac death (SCD) caused by acute ischemic heart disease (IHD) is a major cause of sudden death worldwide. Circular RNAs (circRNAs) are abundant in the heart and play important roles in cardiovascular diseases, but the role of circRNAs as biomarkers in the forensic diagnosis of SCD caused by acute IHD remains poorly characterized. To investigate the potential of two heart-enriched circRNAs, circNFIX and circSLC8A1, we explored the expression of these two circRNAs in different kinds of commonly used IHD models, and further verified their expressions in forensic autopsy cases. The results from both the IHD rat and H9c2 cell models revealed that circSlc8a1 level was upregulated, while the circNfix level was elevated in the early stage of ischemia and subsequently downregulated. The time-dependent expression patterns of the two circRNAs suggested their potential as SCD biomarkers. In autopsy cases, the results showed that the expression of these two circRNAs in the myocardium with acute IHD-related SCDs corresponded to the observations in the ischemic models. Further analysis related to myocardial ischemia indicated that circSLC8A1 showed high sensitivity and specificity for myocardial infarction and was positively correlated with creatine kinase MB in pericardial fluid. Downregulated circNFIX level could indicate the ischemic myocardial damage, and it was negatively correlated with the coronary artery stenosis grade. The combination of circSLC8A1 and circNFIX had better performance to discriminate IHD-related SCDs. The results suggested that circSLC8A1 and circNFIX may be used as auxiliary diagnostic markers for SCD caused by acute IHD in forensic medicine.

Figure 1

Electrocardiogram (ECG) tracings of rats in acute IHD models. The vertical axis of each ECG indicates voltage (mV), with 0.5 mV for all grids. The horizontal axis of each ECG indicates time (s), with 0.2 s for all four grids. (A) Normal rat ECG tracing. (B–D) ECG tracings of rats with ventricular arrhythmias (VAs) induced by BaCl₂ solution: (B) supraventricular tachycardia (SVT), (C) ventricular tachycardia (VT), and (D) ventricular fibrillation (VF). (E) The ST segment elevation in rats after coronary artery ligation (CAL). (F) The ST segment elevation and pathological Q wave were shown in ECG tracings after continuous injection of isoproterenol (ISO) solution for 2 consecutive days.

Figure 2

Haematoxylin–eosin (H-E) staining of the rat myocardium. (A) Normal rat left ventricular myocardium. (B) The left ventricular myocardium showed myocardial enhanced eosinophil staining (arrows) after 10 min of arrhythmia. (C) Enhanced eosinophil staining (arrow) was shown in the left myocardium after 10 min of arrhythmia in rat. (D) Myocardial wavy-like (arrow) was shown after 30 min of arrhythmia in rats. (E) Myocardial necrosis (arrow) with haemorrhage and contraction bans was shown after 60 min of CAL. (F) Coagulation necrosis with loss of nuclei and striations, interstitial infiltrate of neutrophils was shown after 2 consecutive-days injections of ISO solution.

Figure 3

Verification of circSlc8a1 and circNfix. (A) CircSlc8a1 and circNfix are present in the myocardial tissues of rats. Clear single bands were amplified from the cDNA of rat myocardial tissue by divergent primers (◄►) and convergent primers (►◄), but could not be amplified by divergent primers from gDNA. Full-length gels are presented in Supplementary Fig. S1. online (B) Total RNA was incubated with RNase R or buffer only (Mock). After digestion, circSlc8a1, circNfix, and Gapdh mRNA was analysed by qRT-PCR. Differences between mock and RNase R were analyzed using student’s t-test, *P < 0.05, n = 3. (C) Sequence analysis validated the circular junction of the two circRNAs. (D) The CT values of circSlc8a1 and circNfix were determined by qRT-PCR at the corresponding timepoints. Differences between other timepoints group and 0-day group of each indicators were analyzed using one-way analysis of variance (ANOVA); post hoc analyses were performed using Dunnett’s multiple comparison test, *P < 0.05, n = 5.

Figure 4

Expressions of circSlc8a1 and circNfix in the myocardial tissues of IHD rats. (A, B) CircSlc8a1 and circNfix expression in the myocardial tissues of rats after ventricular arrhythmia, n = 6. (C, D) CircSlc8a1 and circNfix expression in the myocardial tissues of rats after coronary artery ligation (CAL), n = 6. (E, F) CircSlc8a1 and circNfix expression in the myocardial tissues of rats at 48 h after the injection of ISO solution, n = 7. (G, H) Expressions of circSlc8a1 and circNfix in H9c2 cells treated with ischemia-hypoxia, n = 3. Differences between IHD treated group and saline group were analyzed using student’s t-test. Differences between other timepoints group and 0-min group of models were analyzed using one-way analysis of variance (ANOVA); post hoc analyses were performed using Dunnett’s multiple comparison test. *P < 0.05 (vs saline group), ^#P < 0.05 (vs 0-min group).

Figure 5

Relationship between circSLC8A1 and circNFIX quantification and the cause of death. Numbers in each group: control = 18; non-MI = 29; MI = 18. (A–D) Expression of circSLC8A1 in myocardium. The circRNA level in the whole ventricular wall was calculated based on the mean ΔCT value for each ventricular wall. (E–H) Expression of circNFIX in myocardium. (I) ROC curves of circSLC8A1. (J) ROC curves of circNFIX. (K) ROC curves of the combination of circSLC8A1 and circNFIX. Differences between groups were analyzed using Mann–Whitney U-test. * P < 0.05.

Figure 6

Correlation between circRNA levels and coronary stenosis grades. (A, B) Total circSLC8A1 and circNFIX levels in cardiac tissues supplied by three coronary arteries (numbers in each group: control = 54; 0 = 36; I—II = 34; III = 37; IV = 34). The circRNA level in cardiac tissues was calculated based on the mean ΔCT value for each coronary artery in controls. Differences between groups were analyzed using one-way analysis of variance (ANOVA); post hoc analyses were performed using Dunnett’s multiple comparison test, *P < 0.05. (C, D) Spearman plots of circSLC8A1 and circNFIX levels and coronary stenosis grades.

Figure 7

Relationships between the two circRNAs and traditional biomarkers in the pericardial fluid. CK-MB, creatine kinase MB; cTnI, cardiac troponin I; NT-proBNP, N-terminal pro-B-type natriuretic peptide. (A) CircSLC8A1 expression in cardiac tissues from decedents with different levels of CK-MB. Differences between groups were analyzed using Mann–Whitney U-test. * P < 0.05. (B) Scatter plot for the IHD group with CK-MB concentration on the X axis and circSLC8A1 expression on the Y axis. (C) CircSLC8A1 expression in cardiac tissues from decedents with different levels of cTnI and NT-proBNP in the pericardial fluid, Differences between groups were analyzed using Mann–Whitney U-test. * P < 0.05. (E–G) CircNFIX expression in cardiac tissues from decedents with different levels of CK-MB, cTnI and NT-proBNP in the pericardial fluid. Differences between groups were analyzed using Mann–Whitney U-test. * P < 0.05. (H) Venn diagram of biomarkers and their relationships. (I, J) Expression of circSLC8A1 and circNFIX in cardiac tissues from decedents with elevated levels of the indicated number of biomarkers (CK-MB, cTnI and NT-proBNP) in the pericardial fluid. ^△ 0 or 1 biomarker whose levels were elevated. Differences between groups were analyzed using one-way analysis of variance (ANOVA); post hoc analyses were performed using Dunnett’s multiple comparison test, *P < 0.05.