Modulation of apoptosis‐related microRNAs following myocardial infarction in fat‐1 transgenic mice vs wild‐type mice

Abstract Background microRNAs (miRNAs) post‐transcriptionally regulate cardiac repair following myocardial infarction (MI). Omega‐3 polyunsaturated fatty acid (ω‐3 PUFAs) may support cardiac healing after MI, but the mechanism is unclear. Methods The fat‐1 transgenic mouse expresses a ω‐3 fatty acid desaturase which converts ω‐6 PUFAs to ω‐3 PUFAs in vivo. MI was induced in fat‐1 transgenic (n = 30) and wild‐type (WT) mice (n = 30) using permanent ligation. Other transgenic and WT mice underwent sham procedure (n = 30 and n = 30, respectively). One week after occlusion, cardiac function was measured by echocardiography and the infarct size was assessed using histology and miRNA microarray profiling. Expression of selected miRNA was confirmed using quantitative real‐time PCR. Results One week following MI, the fat‐1 transgenic myocardium had better cardiac function, a smaller fibrotic area, and fewer apoptotic cardiomyocytes than WT myocardium. Post‐MI profiling showed 33 miRNAs that were significantly up‐regulated, and 35 were down‐regulated, in fat‐1 group compared to the WT group (n = 3 and n = 2 mice, respectively). Among selected apoptosis‐associated miRNAs, 9 miRNAs were up‐regulated (miR‐101a‐3p, miR‐128‐3p,miR‐133a‐5p,miR‐149‐5p,miR‐192‐5p,miR‐1a‐3p,miR‐208a‐3p,miR‐29c‐5p,miR‐30c‐2‐3p), and 3 were down‐regulated (miR‐210‐3p,miR‐21a‐3p,miR‐214‐3p) in fat‐1 transgenic mice compared with WT mice. Kyoto encyclopaedia of genes and genomes (KEGG) pathway analysis indicated likely roles for these miRNAs in MI. Furthermore, Bcl‐2 expression was increased, and caspase‐3 decreased, in infarcted fat‐1 transgenic mouse hearts compared to WT hearts. Conclusions ω‐3 PUFAs may have a protective effect on cardiomyocytes following MI through their modulation of apoptosis‐related miRNAs and target genes.


| INTRODUCTION
Myocardial infarction (MI) is a major cause of morbidity and mortality worldwide. It can lead to loss of cardiomyocytes, left ventricular remodelling, decreased cardiac function, and potentially heart failure. 1 In recent years, evidence has suggested that omega-3 fatty acids (ω-3 PUFAs) have a cardioprotective effect in coronary heart disease and heart failure (HF) [2][3][4][5] ; however, the mechanism this protection is not fully understood.
Following MI, cardiomyocyte apoptosis in the border zone around infarct scars and in the remote zone of the noninfarcted myocardium, 6-8 exacerbates remodelling and aggravates cardiac dysfunction. 9,10 Reversal of cardiomyocyte apoptosis during early stage of MI is crucial for maintaining cardiac function. 11 Studies have shown that ω-3 PUFAs can protect cells against apoptosis; including smooth muscle cells, 12 cardiomyocytes during pressure overload-induced cardiac hypertrophy, 13 and neuronal apoptosis following hyperoxic injury. 14  proliferation, and apoptosis. 15 They are critically involved in heart function in both physiological and pathophysiological conditions. The effect of ω-3 PUFAs on miRNA regulation is not well understood. In this study, we aimed to analyse the miRNA expression profile in the infarcted myocardium of fat-1 transgenic and wild-type (WT) mice to determine whether a lower ω-6 PUFA/ω-3 PUFA ratio can alter the expression of miRNAs that regulate cardiomyocyte apoptosis. We propose that higher tissue ω-3 PUFAs concentrations are cardioprotective effects in MI via the modulation of apoptosis-related miRNA expression.

| Gas chromatography analysis of FAs
The profile of FAs in the heart was determined by gas chromatography as described previously. 16,17 In brief,~10 mg heart tissue was ground into powder under liquid nitrogen, then centrifuged for 5 minutes at 100 g. The sediment was retained for FA methylation by 14% boron trifluoride (BF3)-methanol reagent (Sigma-Aldrich, St Louis, MO, USA) at 100°C for 1 hour. FA methyl esters were analysed with the Agilent HP6890N gas chromatography system equipped with a flame ionization detector (Agilent, Palo Alto, CA, USA).FA peaks were identified by comparing their relative retention times with commercial mixed standards (Nu-Chek Prep, Elysian, MN, USA), and the area percentages for all resolved peaks were analysed using GC Chemstation software. FA mass was determined by comparing areas of various analysed FAs to that of a fixed concentration of external standard. All mice were heparinized (300 U), anesthetized with ip, sodium pentobarbital (40 mg/kg), and ventilated using a small animal ventilator at a frequency of 120/min and tidal volume of 0.8 mL. Body temperature was maintained at 37°C using a heating pad while each chest was surgically opened along the midline. A standard limb lead ECG continuously recorded while a left thoracotomy was performed at the 4th rib and a segment of saline-soaked 7-0 sutures was looped around the left anterior descending (LAD) coronary artery, near its origin from the left coronary artery. The ends of the suture were passed through a 1-inch segment of double-barrelled polyethylene tubing. The ends of the tubing were rounded so that a smooth surface abutted the coronary artery.

| Mouse MI model
In mice from the MI groups, the tubing was gently placed next to the heart and secured with 3 suture knots, with the last knot up against the ends of the tubing. The tubing and suture assembly were then externalized and the thorax closed. Occlusion was accomplished by withdrawing the distal knot 2 mm away from the tubing end. This permits reproducible occlusion of the LAD without tearing the artery.
Successful occlusion was confirmed by the increase in the amplitude of the R wave of lead I during the first few seconds of each occlusion, and elevation of ST segment of lead II. Mice from the sham group underwent the same procedure but the LAD was not occluded.

| Echocardiography
Post-MI left ventricular function was measured after 1 week using the Visual SonicsVevo ® 2100 small animal ultrasound system

| Histology
Following the post-MI echocardiography, mice were humanely killed.
The hearts were harvested and 6 from each group were fixed in 4% paraformaldehyde. Fixed hearts were sectioned into 5-μm-thick sections with longitudinal cutting, then stained with haematoxylin and eosin or Masson's Trichrome.

| RNA isolation
The infarcted parts of myocardium, the border zone of infarct scars, and the remote zone of noninfarcted myocardium from infarcted fat-1 transgenic mice (n = 3) and infarcted WT mice (n = 3) were used for RNA (50%) and protein extraction (50%).
Total RNA was isolated using kits according to manufacturer's instructions. For detecting the microRNA expression, miRcute miRNA isolation kit, miRcute miRNA First-Strand cDNA synthesis kit and miRcute miRNA qPCR detection kit (SYBR Green) were used (all from Vazyme Biotech Co. Ltd, Nanjing, China). For detecting mRNA, TRIzol RNA (Invitrogen, Carlsbad, CA, USA), Thermo-Script ™ RT-PCR system and a Fast Start universal SYBR green master (ROX) were used. RNA quality and integrity were determined using a 2100 bioanalyzer (Agilent Technologies Inc) and RNA LabChip ® kits (Agilent Technologies Inc).

| miRNA arrays
Small RNAs were isolated from the total RNA and labelled with Values < 0.01 were set to 0.01 and each measurement was divided by the 75th percentile of all measurements from the same samples. miRNAs whose expression differed by at least 2fold between fat-1 transgenic MI and WT MI groups were selected for further investigation.

| Quantitative real-time PCR analysis
Following MI, total cellular RNA was extracted using Trizol reagent

| Effect of a low ω-6 PUFA/ω-3 PUFA ratio on post-MI cardiac function
To determine whether ω-3 PUFAs have a cardioprotective effect on the ischaemic myocardium, cardiac function was analysed by echocardiography, 1 week after MI. Fat-1 transgenic mice showed a significantly increased FS, LVEF, and a decreased left ventricular end-diastolic dimension (LVEDd) compared with WT mice( Figure 1B).
Masson's Trichrome staining showed a significantly smaller fibrotic area in fat-1 transgenic hearts compared with WT hearts, post-MI ( Figure 1D). Figure 1C also shows that there were fewer apoptotic cardiomyocytes in the myocardium.

| A low ω-6PUFA/ω-3 PUFA ratio has antiapoptotic effects in the infarcted myocardium
C-caspase 3 is a key factor in the apoptosis pathway and its expression level is positively correlated with apoptosis. C-caspase 3 protein, detected using immunocytochemistry, was lower in cardiac tissue from fat-1 transgenic mice compared with WT mice (Figure 2A). The mRNA expression of caspase-3 (determined by quantitative PCR) showed a comparable result and was statistically significant (Figure 2B).
The amount of Bax, Bcl-2, and PCNA protein were also determined by immunohistochemical staining. Following MI, fat-1 transgenic mice had more of Bcl-2 protein in their cardiac tissue than WT mice ( Figure 2B).

| Infarcted myocardium miRNA profile
Three infarcted myocardium samples from fat-1 transgenic mice and 2 from WT mice were used the miRNA array. The sample from 1 WT mouse was excluded because of poor RNA quality.
Sixty-eight miRNAs were differentially expressed between the transgenic and WT groups, including 33 that were up-regulated and 35 that were down-regulated (more than a 2-fold change; P < 0.05).
In unsupervised hierarchical clustering analysis, normalized microarray expression for the 68 miRNAs showing differential expression in the five infarcted myocardium samples were used to generate a heat map (Figure 3).

| Analysis of infarcted myocardium miRNAs by quantitative real-time PCR
Following the selection strategy, fifteen miRNAs were chosen to be verified by quantitative real-time PCR (Table 2).

| The time-course trend of miR-210 expression and its target gene
To test the time course of miR-210 expression, expression level of miR-210 in heart tissue was measured at 3, 7 and 14 days after myocardial infarction. It showed that miR-210 was up-regulated in the fat-1 mice as compared to WT mice at week-1 post-MI ( Figure 5A). The protein level of CASPASE8AP2, a target gene of miRNA-210, was also measured.
Western blotting results showed that the protein level of Caspase8ap2 in fat-1 group is lower than in WT group. CASPASE8AP2 is known to promote cardiomyocyte apoptosis, thus the reduction of CASPA-SE8AP2 level may prevent cardiomyocyte apoptosis.
T A B L E 2 microRNAs with apoptosis characteristics identified from the microarray analysis of infarcted myocardium from fat-1 transgenic (n = 3) and wild-type mice (n = 2). (>2-fold change)  We identified 9 apoptotic-related miRNAs that had a higher expression in infarcted myocardium from fat-1 transgenic mice than in WT mice, while 3 had a lower expression. Each is believed to be cardiac-specific or cardiac-enriched miRNAs involved in differentiation of heart myocytes and maintaining the functionality/survival of cardiac muscle cells. 22 The activity and function of these miRNAs are strictly regulated to ensure proper cardiac contractility and conduction. In pathological conditions such as MI, deregulation of these miRNAs may lead to myocardial apoptosis, necrosis, fibrosis and other destructive processes, eventually lead to cardiac arrhythmia, hypoxia, ischaemia, left ventricular dilatation. Modulation of specific miRNAs could serve as potential therapeutic approaches for MI. 23 In this study, we saw that fat-1 transgenic mice (which have a lower ratio of ω-6 PUFA/ω-3 PUFAs in the myocardium) are partially protected from MI-induced cardiomyocyte apoptosis. This was correlated with a specific miRNA expression profile. miRNA expression profiles play an essential role in apoptotic-related signalling pathways such as phosphoinositide 3-kinase (PI3K), Bcl-2 and caspase pathways. The Bcl-2 family of proteins includes anti-apoptotic members like Bcl-2 and Bcl-x L along with pro-apoptotic member Bax, and together regulate mitochondrial integrity during an apoptotic insult.

Gene name Fold change P-value Regulation
MI-related cardiomyocyte apoptosis has been shown to involve increased expression of Bax and decreased levels of Bcl-2. 24 Our results provide early evidence that lowering the ω-6/ω-3 PUFAs ratio can influence apoptosis at the mitochondrial level of by restoring the anti-apoptotic balance of Bcl-2. Similarly, another study has found that ω-3 PUFAs are anti-apoptotic in developing cerebellum through Bcl-2 and MAPK pathways. 25 One of the key events in apoptosis is activation of caspase-3, and its expression is positively correlated with apoptosis. 26  Overall, our study suggests that a lower ω-6 PUFA/ω-3 PUFA ratio in the heart can alter the miRNA, mRNA, and ultimately the protein expression profile in the murine infarcted myocardium. Our results suggest that this profile is anti-apoptotic and results in fewer cardiomyocyte deaths, less fibrosis and better cardiac function. Further research into the effects of the identified miRNA may provide future therapeutic targets to reduce the cardiac damage, morbidity, and mortality caused by MI.

CONFLI CT OF INTEREST
None declared.