Knock‐out of MicroRNA 145 impairs cardiac fibroblast function and wound healing post‐myocardial infarction

Abstract Prevention of infarct scar thinning and dilatation and stimulation of scar contracture can prevent progressive heart failure. Since microRNA 145 (miR‐145) plays an important role in cardiac fibroblast response to wound healing and cardiac repair after an myocardial infarction (MI), using a miR‐145 knock‐out (KO) mouse model, we evaluated contribution of down‐regulation of miR‐145 to cardiac fibroblast and myofibroblast function during adverse cardiac remodelling. Cardiac function decreased more and the infarct size was larger in miR‐145 KO than that in WT mice after MI and this phenomenon was accompanied by a decrease in cardiac fibroblast‐to‐myofibroblast differentiation. Quantification of collagen I and α‐SMA protein levels as well as wound contraction revealed that transdifferentiation of cardiac fibroblasts into myofibroblasts was lower in KO than WT mice. In vitro restoration of miR‐145 induced more differentiation of fibroblasts to myofibroblasts and this effect involved the target genes Klf4 and myocardin. MiR‐145 contributes to infarct scar contraction in the heart and the absence of miR‐145 contributes to dysfunction of cardiac fibroblast, resulting in greater infarct thinning and dilatation. Augmentation of miR‐145 could be an attractive target to prevent adverse cardiac remodelling after MI by enhancing the phenotypic switch of cardiac fibroblasts to myofibroblasts.


| INTRODUC TI ON
Myocardial infarction (MI) remains the leading cause of death worldwide. Adverse cardiac remodelling after MI resulting from infarct thinning and expansion contributes to ventricular dilatation and heart failure which can result in disability and death. [1][2][3] The mechanisms underlying cardiac remodelling are complex, but the contractile insufficiency and ventricular dilation remain the most important contributors to adverse outcomes. 4 Augmentation of cardiac contractile function may improve outcomes, but clinical translation of this strategy has not been successful after an MI. 5,6 However, a number of new technologies continue to be evaluated, such as improving calcium handling. 1 Other therapeutic interventions have also been evaluated, including stem cell therapy to improve wound healing and to prevent ventricular remodelling, 5,[7][8][9][10] and altering molecular pathways to reduce the extent of cardiomyocyte necrosis. 11,12 Although several discoveries are promising, none have become standards of clinical care and many patients continue to suffer from congestive heart failure after MI despite medical therapy.
Novel strategies to prevent cardiac dilatation and dysfunction are required to complement existing treatments for reverse the trend of increasing morbidity and mortality from heart failure.
Cardiomyocytes loss as a result of ischaemia initiates a series of responses to repair the infarct and heal the scar. Cardiac fibroblasts play an important role in wound healing after MI. 13,14 The fibroblasts are a dynamic cell type and can differentiate into myofibroblasts in response to injury, to mediate wound healing and tissue repair. [15][16][17] As myofibroblasts, they synthesize intracellular contractile proteins, such as α-SMA and they generate increased levels of extracellular proteins, such as collagen. These activities contribute to reparative fibrosis and the formation of a stable scar which prevents ventricular rupture and may limit infarct expansion and ventricular dilatation. 15 We believe that promoting differentiation of fibroblasts to myofibroblasts might be an effective strategy to stimulate infarct scar contracture, preserve ventricular morphology after injury and prevent progressive heart failure.
MicroRNA 145 (miR-145) regulates the proliferation and differentiation of a variety of cells, 18,19 including endothelial and smooth muscle cells (SMCs) in the vascular system. [20][21][22][23] In vascular SMCs, deletion of MiR-143/145 resulted in severe reduction in the number of contractile SMCs. 24 In the current study, using a miR-145 knockout (KO) mouse model, we investigated the effects of miR-145 deletion on myofibroblast formation during adverse cardiac remodelling after an MI and evaluated potential mechanisms.

| Experimental animals
All animal procedures were approved by the University Health Network Animal Care Committee. All experiments were carried out according to the Guide for the Care and Use of Laboratory Animals (NIH, 8th Edition, 2011). MiR-145 KO mice were generated as previously described 25 and maintained in a C57BL/6 background. Briefly, miR-145 was deleted in mice, by introducing loxP sites for Cre-mediated recombination at the regions flanking the pre-miR coding regions for homologous recombination. The targeting strategy deleted the 70-bp pre-miR stem-loop sequence of miR-145, or a genomic region encompassing miR-145, and replaced it with a neomycin resistance cassette flanked by loxP sites. The neomyocin resistance cassette was removed by breeding these mice with mice expressing a ubiquitously-expressed CAG-Cre transgene. Breeding the offspring of these crosses generated homozygous mutants with a ubiquitous knock-out of miR-145.
WT littermates were also from a C57BL/6 background.
The left coronary artery of mice was permanently ligated to induce an anterior MI. During this procedure, mice were incubated and ventilated with 2% isoflurane (Pharmaceutical Partners of Canada Inc). The infarct size in each mouse was estimated immediately after coronary artery ligation by visual identification of the tissue discoloration (loss of the normal pink colour and accumulation of a blue colour). Mice with infarct sizes between 30% and 35% of the left ventricular free wall were used in the following experiments. WT littermates receiving the same procedure, though without ligation of the left coronary artery, were used as sham control.
Cardiac function was measured with echocardiography before and 7, 14, 21 and 28 days after MI and with a pressure-volume catheter 28 days after MI. The following parameters were calculated by echocardiography: Left ventricular internal systolic dimension (LVIDs), left ventricular internal diastolic dimension (LVIDd) and percentage of fractional shortening (fractional shortening %) = (LVIDd-LVIDs)/ LVIDd*100. Pressure-volume analysis was used to determine ejection fraction, dP/dt, tau and left ventricular (LV) volumes. At the end of the study (28 days after MI), the hearts were arrested and fixed at physiologic pressures. The LV was serially sectioned into five 1 mm rings along the longitudinal axis, representing cross-sections of that region of the heart. The scar area was calculated by computerized planimetry (Image-Pro Plus, Media Cybernetics) of digital images, consisting of three Masson's trichrome-stained serial LV sections from the apex. The scar area of each heart section was measured by scar tissue length (epicardial scar + endocardial scar lengths)/2 × 1 mm of section thickness. Total scar area was calculated as the sum of the scar area of each left ventricle cross-section ring, divided by the sum of the left ventricular free wall area (epicardial + endocardial lengths)/2 × 1 mm thickness of each ring section. Scar thickness was measured from the second ring of the serial LV sections from the apex, with an average of three measurements, one at the middle, and two at the edges of the scar region within that section. Remote myocardial wall thickness was measured from the interventricular septum of the third ring in the LV serial sections from the apex, with an average of three measurements, one at the middle, and two at the edges of the remote region within that section. The Masson's trichrome staining was conducted to show the fibrotic tissue and the scar area in both WT and miR-145 KO mice. The heart tissues were collected at 3,7,14 and 28 days after MI for miRNA detection. For immunostaining, hearts were collected and fixed at 7 days after MI at physiological ventricular pressures to preserve ventricular geometry.

| Total RNA extraction and real-time qPCR
Total RNA was extracted from mouse heart tissue or cells with Trizol reagent (Sigma-Aldrich, T9424) according to the manufacturer's instructions. Expression of miR-145 was evaluated by the TaqMan microRNA Assay kit (Thermo Fisher scientific, 4366596 for Taqman microRNA reverse transcription and 4324018 for quantitative PCR). U6 snRNA was used as a control. Total RNA was converted to cDNA and the expression of myocardin was detected by real-time qPCR with the fast SYBR Green master mix kit. GAPDH was served as the loading control. Primer pairs specific for mouse myocardin A (sense primer: 5′-CTTCTCTCCCCCAGCTTCCA-3′; antisense primer: 5′-CTTGGGCTTTTGGGACAAGG-3′) and GAPDH (sense primer: 5′-AGAACATCATCCCTGCATCC-3′; antisense primer: 5′-CACATTGGGGGTAGGAACAC-3′). The primers of miR-145 and U6 were purchased from Thermo Fisher scientific. Relative expression was calculated using the 2 -ΔΔCt method.
Transfection with the miR-145 mimic (5 nmol/L) was performed using HiPerFect transfection reagent (Qiagen) following the manufacturer's instructions. The cells were harvested 48 hours after transfection and used for the following experiments.

| Collagen gel contraction assay
Collagen type I from rat tail was obtained from Corning (Cat#: 354249). Cardiac fibroblasts, which were isolated from WT or miR-145 KO mice, transfected with 5 nmol/L miR-145 mimic or scrambled miRNA were mixed with the neutralized collagen solution. The mixture with cardiac fibroblasts or no cells (blank) were placed into a 24-well plate and allowed to form gels at room temperature in the hood. Then culture medium was added, and the plate was placed into the incubator for 24 hours. Gel size was measured and the percentage of gel shrinkage was calculated.

| Cell migration assay
Cardiac fibroblasts from different groups grown to confluence in 35 mm plates were scratched with a sterile pipette tip, washed twice with PBS and incubated in serum-free medium at 37°C for 24 hours. Pictures were taken at 40× magnification, and the migration rate was calculated using ImageJ software (NIH). Then the cells were fixed and stained, and the percentage of polarized and α-SMA + polarized cells was calculated.

| Immunofluorescence staining
Paraffin-embedded mouse heart slices or cardiac fibroblasts were fixed with 2% paraformaldehyde in PBS for 20 minutes at room temperature. The fixed cells or slices were permeabilizated with 0.2% Triton X-100 in PBS and blocked for 1 hour in PBS containing 10% bovine serum albumin. They were then incubated with primary antibody α-SMA (Cat#: A2547, 1:800, Sigma) overnight at 4°C.

| Statistical analyses
All values were expressed as mean ± SD. Analyses were performed with SPSS software. Student's t test was used for 2-group comparisons. Comparisons among three or more groups were analysed using one-way analysis of variance (ANOVA) or two-way ANOVA with repeated measures over time, followed by Tukey post hoc tests.
Differences were considered statistically significant at P < .05.

| Cardiac function decreased more in miR-145 KO than WT mice
To evaluate effect of miR-145 on cardiac remodelling and function,

| Increased infarct scar and decreased myofibroblast formation in miR-145 KO mice
After functional analysis, the hearts of both KO and WT mice were

| Knock-out of miR-145 impairs the functional conversion of cardiac fibroblasts to myofibroblasts
To further confirm our in vivo finding, we harvested cardiac fibro-

| Klf4 and myocardin are the targets of miR-145 involved in the regulation of cardiac fibroblast differentiation to myofibroblasts
Since myocardin and Klf4 are regulators for SMC-specific contractile protein expression, 26 we postulated that KO of miR-145 may inhibit the transdifferentiation of cardiac fibroblasts into contractile myofibroblasts through the same target genes Klf4 and myocardin as in SMCs.
As expected, Klf4 protein expression level was significantly higher in cardiac fibroblasts of KO mice than those in WT mice, implying releasing the inhibition on Klf4 after knock-out miR-145 ( Figure 6A). in cardiac fibroblasts of miR-145 knockout (KO) mice treated with a scrambled miRNA was significantly lower than in wild type (WT) mice treated with a scrambled miRNA. Treatment with a miR-145 mimic partially restored the expression of collagen I and α-SMA in KO cardiac fibroblasts. Cardiac fibroblast-mediated collagen gel contraction (detected by a collagen gel contraction assay) in KO treated with a scrambled miRNA was significantly less than in WT mice treated with a scrambled miRNA. Treatment with a miR-145 mimic partially restored the ability of KO cardiac fibroblasts to contract the collagen gel (C). *P < .05, **P < .01, n = 3/group for A and B, n = 4/ group for C decreased Klf4 and in turn increased myocardin expression which eventually promoted the transdifferentiation of cardiac fibroblasts into more contractile myofibroblasts.

| D ISCUSS I ON
Accumulating evidence indicates that preventing infarct scar thinning and dilatation may prevent cardiac remodelling and reduce left ventricular contractile dysfunction and improve the prognosis after MI. 5,7,8 However, clinically relevant treatments to prevent heart failure are limited. Novel strategies which complement existing therapy are needed to stem the increasing number of patients with heart failure. In this study, we identified miR-145 as a novel target for preventing ventricular dilation after MI. We demonstrated that miR-145 was decreased in both the scar and border regions 3 days post-MI in WT mice. We further demonstrated that cardiac function decreased more and the infarct size was larger but thinner in miR-145 KO than that in WT mice. The increased scar thinning and dilatation in miR-145 KO mice were associated with the thickening of the LV remote region, owing to compensatory cardiac hypertrophy. These results suggested that miR-145 may participate in cardiac repair after MI.
Previous studies have shown that miR-145 represses the proliferation and promotes the differentiation of multiple cell types, including cancer cells, stem cells, endothelial cells and SMC etc. 20,[26][27][28][29] More importantly, studies have demonstrated the regulation of SMC

CO N FLI C T O F I NTE R E S T
The authors confirm that there are no conflicts of 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.