Inhibition of let-7c Regulates Cardiac Regeneration after Cryoinjury in Adult Zebrafish

The let-7c family of micro-RNAs (miRNAs) is expressed during embryonic development and plays an important role in cell differentiation. We have investigated the role of let-7c in heart regeneration after injury in adult zebrafish. let-7c antagomir or scramble injections were given at one day after cryoinjury (1 dpi). Tissue samples were collected at 7 dpi, 14 dpi and 28 dpi and cardiac function was assessed before cryoinjury, 1 dpi, 7 dpi, 14 dpi and 28 dpi. Inhibition of let-7c increased the rate of fibrinolysis, increased the number of proliferating cell nuclear antigen (PCNA) positive cardiomyocytes at 7 dpi and increased the expression of the epicardial marker raldh2 at 7 dpi. Additionally, cardiac function measured with echocardiography recovered slightly more rapidly after inhibition of let-7c. These results reveal a beneficial role of let-7c inhibition in adult zebrafish heart regeneration.


Introduction
Cardiovascular diseases are the leading cause of death globally [1]. Cardiac remodeling after myocardial infarction in mammals leads to scar formation and hypertrophy [2,3]. Adult zebrafish hearts regenerate completely without scar formation after cryoinjury [4,5], making zebrafish a useful animal model for studying cardiac regeneration [6][7][8]. Cell-lineage-tracing studies with zebrafish have shown complete regeneration after amputation through proliferation of pre-existing cardiomyocytes [9][10][11][12], unlike in mammals [3]. Epicardial-derived cells undergo epithelial-mesenchymal transition to form coronary smooth muscle cells and fibroblasts [11,13,14]. The epicardium plays a crucial role in zebrafish heart regeneration [13,14], as epicardial markers such as tbx18, tcf21 and raldh2 are re-expressed after cryoinjury, suggesting that tissue injury reactivates developmental genes [15,16]. We used the cryoinjury method to simulate myocardial infarction. Compared to amputation, cryoinjury more closely mimics the physiological responses associated with myocardial infarction, including cell death, inflammation and scarring [4,5,17]. Cell death and inflammation in the infarct area is followed by simultaneous proliferation of endocardium, epicardium and cardiomyocytes [4,5]. During the first few weeks, fibrin, collagen and other extracellular matrix proteins accumulate in the injury area [4,5,18,19]. The adult zebrafish heart completely regenerates and scar tissue is replaced by functional cardiac tissue, in approximately two months following cryoinjury [11].
Novel therapies are required for improving cardiac remodeling and function after injury. MiRNAs are short, noncoding RNA molecules that act as negative regulators of gene expression by inhibiting mRNA translation or promoting mRNA degradation. They were first studied in model organisms, such as Caenorhabditis elegans and Drosophila [20,21]. MiRNAs are involved in cell proliferation, apoptosis, hematopoiesis and oncogenesis [22]. The let-7 family of miRNAs is expressed during embryonic development [23]. let-7c is ubiquitously expressed in the whole body in mice [24]. let-7c is required for cell growth and differentiation, and its sequence and function is highly conserved [25]. let-7c is upregulated in human cardiac maturation in vitro. Overexpression of the let-7 family of miRNAs matures human embryonic stem cell-derived cardiomyocytes, increases cell size, sarcomere length and contraction [26,27]. let-7 miRNAs are upregulated in mouse and human heart disease samples [28,29]. Inhibition of let-7c improves cardiac remodeling and function after infarction in mice [30]. Another study in mice reported increased recruitment of epicardial cells after let-7c inhibition [23]. let-7a/b overexpression reduces collagen I expression, cardiac fibroblast proliferation and migration in primary cardiac fibroblast cultures from rats [31].
The aim of our study is to elucidate the role of let-7c in cardiac regeneration after cryoinjury in adult zebrafish.

Zebrafish
Zebrafish aged 6-18 months were obtained from the breeding line of a Turku strain maintained in the zebrafish core facility at the University of Helsinki [32]. They were housed at 28 • C with a 14:10 h light:dark cycle. The permit for zebrafish experiments was obtained from the Regional State Administrative Agency for Southern Finland in agreement with the ethical guidelines of the European convention (ESAVI/2988/04.10.07/2014 and ESAVI/4131/04.10.07/2017).

Cryoinjury
Two hundred and forty wild type adult zebrafish were anesthetized with 0.03% tricaine and a small incision was made on the ventral side of the fish for direct access to the heart. Then, a metal probe cooled in liquid nitrogen was applied to the ventricle, resulting in death of approximately 20% of ventricular cardiomyocytes. Sham operations were performed similarly with a metal probe at room temperature.

Echocardiography
Echocardiography was performed with Vevo 2100 and 50 MHz ultrasound probe (FUJIFILM VisualSonics, Toronto, ON, Canada). The fish were first anesthetized in 0.03% tricaine for one minute, followed by anesthesia in 0.015% tricaine during echocardiography. B-mode videos were recorded before cryoinjury, 1 dpi, 7 dpi, 14 dpi, 21 dpi and 28 dpi. Recorded B-mode videos were analyzed with Visual Sonics software as described previously [33,34]. Briefly, during five consecutive heartbeats, systolic and diastolic ventricular volumes were determined and the epicardial edge was manually marked; diastolic and systolic lengths of the apical image long axis (L) and its diastolic and systolic area (A) are measured ( Figure 6A). Diastolic and systolic ventricular volumes (V) are then calculated using the formula:

Histology
Acid fuchsin orange G-stain (AFOG) was used for measuring infarct size and fibrosis as described previously [16]. Briefly, paraffin sections were dewaxed and post-fixed in Bouin's solution for 60 min at 60 • C in an incubator, stained with Weigert's iron hematoxylin and AFOG-solution. The infarct region was measured using Image J 1.43u software (National Institutes of Health, Bethesda, MD, USA) and the percentage of the infarct size relative to the ventricle size was calculated. Fibrin and collagen percentages were also measured similarly from the infarct region.

Immunohistochemistry
Formation of new cardiomyocytes was analyzed using BRDU labelling or PCNA staining. Dissected adult zebrafish hearts were fixed in 10% formalin overnight, dehydrated in ethanol and embedded in paraffin. Sections (5 µm) were deparaffinized, rehydrated and antigens retrieved by heating in citrate buffer, pH 6.0. Sections were incubated with anti-mouse myosin heavy chain (MHC) (Merck Millipore, Billerica, MA, USA), anti-rat BRDU (Thermofisher, Waltham, MA, USA), anti-rabbit mef-2 (Santa Cruz, California, USA) and anti-mouse PCNA (Cell Signalling, Danvers, MA, USA) at 4 • C overnight, washed with PBS followed by incubation with AlexaFluor-488-conjugated goat anti-mouse, AlexaFluor-594-conjugated goat anti-rabbit and AlexaFluor-594-conjugated goat anti-rat secondary antibodies at room temperature for 1 h. DAPI (4 ,6-Diamidino-2-Phenylindole, Dihydrochloride) (Molecular Probes, Eugene, OR, USA) was used for labelling nuclei. Samples were imaged with Leica DM 4500B fluorescence microscope (Leica microsystems, Wetzlar, Germany) and Image J 1.43u software was used for counting proliferating cardiomyocytes. PCNA positive cardiomyocytes were counted from the whole ventricle, PCNA positive and mef-2 positive cells were considered new cardiomyocytes. BRDU positive cardiomyocytes were counted from border area of infarct, BRDU positive and MHC positive cells were considered new cardiomyocytes. PCNA positive or BRDU positive cardiomyocytes were expressed as positive cardiomyocytes/mm 2 , as described previously [36].

Statistical Analysis
Data were analyzed with nonparametric Kruskall-Wallis test followed by Mann-Whitney U test or two-tailed Student's t-test. A value of p < 0.05 was considered statistically significant. Results are shown as mean ± SEM, * p ≤ 0.05, ** p ≤ 0.005, *** p ≤ 0.0005.

Expression of pai-1
pai-1 expression was quantified to study the increased rate of fibrinolysis observed in the let-7c antagomir fish. pai-1 expression remained similar. However, there was a trend for lower relative pai-1 expression in the heart at 28 dpi in let-7c antagomir fish (0.02 ± 0.01, p = 0.10) compared to scramble fish (0.04 ± 0.01) (Figure 3).

Expression of pai-1
pai-1 expression was quantified to study the increased rate of fibrinolysis observed in the let-7c antagomir fish. pai-1 expression remained similar. However, there was a trend for lower relative pai-1 expression in the heart at 28 dpi in let-7c antagomir fish (0.02 ± 0.01, p = 0.10) compared to scramble fish (0.04 ± 0.01) (Figure 3).

Expression of pai-1
pai-1 expression was quantified to study the increased rate of fibrinolysis observed in the let-7c antagomir fish. pai-1 expression remained similar. However, there was a trend for lower relative pai-1 expression in the heart at 28 dpi in let-7c antagomir fish (0.02 ± 0.01, p = 0.10) compared to scramble fish (0.04 ± 0.01) (Figure 3).

Discussion
Our study reveals that inhibition of let-7c results in an increased rate of fibrinolysis, increased number of proliferating cardiomyocytes, higher expression of the epicardial cell marker raldh2 and a faster rate of improvement of cardiac function.
Initially two different doses of antagomir (20 mg/kg and 100 mg/kg) were tested and 20 mg/kg was chosen, as it was sufficient for inhibition of let-7c. A single injection of the let-7c antagomir at 20 mg/kg downregulated let-7c for several weeks in mice [37,38]. We found a single injection of let-7c antagomir to result in let-7c silencing in the cryoinjured fish in all organs tested; the heart, the liver, and the kidney. Variation of the qPCR results was highest in the heart samples, which may be partly explained by the very small size of the heart compared to that of the kidney and the liver. The variation of expression together with the small n = 3−5 resulted in not reaching statistical significance in all the comparisons. let-7c expression was at 7 dpi 21-fold, at 14 dpi 11-fold, and at 28 dpi 11-fold in the cryoinjured scramble hearts compared to the cryoinjured antagomir hearts, reaching statistical significance at 7 dpi, a time-point where cardiomyocyte proliferation and epicardial activation were found to be increased with let-7c silencing.
Fibrin deposition and scar formation after cryoinjury are essential for stimulating cardiomyocyte proliferation and cardiac remodeling in zebrafish [19]. In our study we observed decreased fibrin in the infarct area at 28 dpi after let-7c inhibition. We further looked at fibrinolysis and found that pai-1 shows a trend of lower expression in let-7c antagomir fish hearts at 28 dpi. pai-1 is the main inhibitor of plasmin, a serine protease, which dissolves many blood proteins including fibrin [39]. Adult zebrafish hearts completely regenerate after cryoinjury, unlike mammal hearts [4,5]. Injury stimulates expression of the developmental genes raldh2, tbx18 and tcf21 in the epicardium [13,15,16]. Activated epicardium is a hallmark of heart regeneration in adult zebrafish [13,14]. Hence, we quantified the expression of raldh2, an epicardial marker and found increased expression at 7 dpi in the let-7c antagomir fish hearts. Retinoic acid signaling plays an important developmental role in the embryonic zebrafish heart [40]. raldh2 oxidizes retinaldehyde to retinoic acid and regulates tissue levels of retinoic acid. In the adult zebrafish heart, raldh2 expression is upregulated after cryoinjury and inhibition of raldh2 affects cardiomyocyte proliferation [13][14][15]. We examined cardiomyocyte proliferation by PCNA staining and found an increase in the number of proliferating cardiomyoctes in let-7c antagomir fish hearts at 7 dpi, in agreement with increased raldh2 expression at 7 dpi. At later time-points cardiomyocyte proliferation was similar between let-7c antagomir and scramble fish. Our results are in agreement with previous studies, which have found similar amounts of PCNA positive cardiomyocytes in the heart after cryoinjury, with cardiomyocyte proliferation peaking during the first week after cryoinjury [5,41]. Additionally, we used BRDU baths to detect proliferating cardiomyocytes. However, we found that administration of BRDU by fish water to be inefficient in getting BRDU into the heart, as we observed a weak BRDU signal in stainings. After activation of epicardium, epithelial-mesenchymal transition gives rise to fibroblasts [13][14][15]. Fibroblasts are essential for regeneration; one study has reported that genetic ablation of fibroblasts negatively affects cardiac remodeling [19]. However, we found that expression of fibroblast associated genes col12a1a and postna to remain similar in let-7c antagomir and scramble fish hearts.
Echocardiography is the method-of-choice for studying cardiac function. However, due to the small size of the adult zebrafish (3-4 cm) and the very small size of the ventricle (750 µm), obtaining reproducible and reliable data is challenging [42]. The zebrafish ventricle is highly trabeculated, thus determination of the endocardial border is difficult [43]. We found that FVS measured from the epicardial borders of the ventricle is the most reliable method to quantify cardiac function in zebrafish. We found the recovery of cardiac function after cryoinjury essentially similar in let-7c antagomir and scramble fish. However, more proliferating cardiomyocytes in the let-7c antagomir fish at 7 dpi may reduce cardiac contractile function, as they are still too immature to contribute to it at this early time-point. Cardiomyocyte proliferation peaks at 7 dpi during heart regeneration in zebrafish, followed by restoration of myocardium and removal of fibrotic tissue at 14 dpi and later time-points [44,45]. Indeed, the recovery is statistically faster in fish receiving let-7c antagomir at 14 dpi ( Figure 6).
Studies in mice show beneficial effects of silencing let-7c on cardiac recovery; one-study reports improved cardiac function with increased recruitment of epicardial cells [23]. Another study reports reduced fibrosis and improved cardiac function [30]. Consistently, we show let-7c inhibition to increase the number of proliferating cardiomyocytes, increase the expression of the epicardial cell marker raldh2 and to accelerate the rate of recovery of cardiac function.
In conclusion, our study reports a favorable response to inhibition of let-7c in heart regeneration after cryoinjury in adult zebrafish. We show increased cardiomyocyte proliferation and epicardial activation at 7 dpi following inhibition of let-7c, as well as faster recovery of cardiac function and faster fibrinolysis. On the basis of these results, further studies on inhibition of let-7c to improve cardiac function are warranted.