Knockdown of C3G Mitigates Post-Infarct Remodeling Via Regulation of ERK1/2, Bcl-2 and Bax in Rat Myocardial Cells

Cardiomyocyte-specic knockout of pro-survival integrin β1 subunit and its downstream components have been demonstrated to aggravate remodeling after myocardial infarction (MI). However, as a component of integrin pathway, it is unclear whether knockdown of pro-survival C3G (rap guanine nucleotide exchange factor 1) in cardiac myocytes and broblasts could have effect on myocardial remodeling. A rat model of MI was established by ligation of left anterior descending coronary artery. Infarcted myocardium and its border zones in Sprague-Dawley rats were transiently infected with C3G knockout lentivirus via local injection to knockdown C3G in the myocardium. Twelve weeks after injection with the lentiviruses, cardiomyocytic apoptosis and collagen in surviving myocardium, and left ventricle (LV) end-diastolic diameter were decreased, whereas LV weight / body weight ratio and LV ejection fraction were increased in MI group via down-regulation of pro-survival C3G, phosphorylated (p) ERK1/2 (phosphorylated extracellular regulated kinase 1/2) and Bcl-2 (B-cell lymphoma-2), and up-regulation of pro-apoptotic Bax in the surviving myocardium. On the other hand, treatment with the lentiviruses was found to delete C3G and diminish cell proliferation in vitro cardiac myocytic and broblastic cell lines respectively via down-regulation of p-ERK1/2 and Bcl-2, and up-regulation of Bax. In conlusion, knockdown of pro-survival C3G in myocardium may mitigate surviving myocardial remodeling after MI, possibly through regulation of p-ERK1/2, Bcl-2 and Bax in vivo cardiac myocytes and broblasts.


Introduction
Myocardial remodeling, ischemic cardiomyopathy and left ventricular dysfunction are related to the increased risk of adverse cardiovascular events and mortality after myocardial infarction (MI) [1].
Pathological changes in the heart are determined by the improperly over-activated pro-survival genes in vivo cardiac myocytes and broblasts in the surviving myocardium around the infarcted lesion [2] . Cardiomyocyte-speci c knockout of pro-survival integrin β1 subunit and its downstream molecules such as focal adhesion kinase (FAK) have been found to exacerbate myocardial remodeling after MI via their aggravation of apoptosis in the cardiomyocytes [3][4][5]. Pro-survival adaptor protein CrkL (v-crk avian sarcoma virus CT-10 oncogene homolog (Crk)-like) can bind to C3G (rap guanine nucleotide exchange factor 1), which acts as a guanine nucleotide exchanger in integrin pathway [3]. Our previous study has shown that knockdown of CrkL can worsen cell apoptosis, which was induced by hypoxia / reoxygenation (H/R) in vitro cardiomyocytic cell line [6]. The effects of CrkL are inferred to be mediated by its downstream pro-survival molecules such as C3G exchange factor. In addition, C3G protein has been shown to express in vitro cardiomyocytic cell line, and its elevated expression was con rmed in the surviving myocardium after MI [7,8]; and knockdown of pro-survival C3G can aggravate H/R-induced apoptosis and pathological processes in vitro cardiomyocytic cell line [7]. However, it is still unclear whether knockdown of pro-survival C3G in myocardium may have effect on the process of surviving myocardial remodeling after MI via regulation of p-ERK1/2 (phosphorylated extracellular regulated kinase 1/2), Bcl-2 (B-cell lymphoma-2) and Bax in cardiac myocytes and broblasts.

Construction of C3G Knockout Recombinant Lentiviruses
The candidate sequences knockouting speci cally all 13 variant mRNA transcripts of rat C3G gene were designed (http://www.e-crisp.org/E-CRISP/) [7] and cloned respectively into pLenti-Cas-Guide lentiviral vector (Origene Co., Ltd, Rockville MD, USA). RT-PCR (reverse transcription polymerase chain reaction) and protein immunoblot screening experiments revealed that the recombinant vector containing the targeting sequence was the most effective one to knockout C3G gene in the rat-derived H9C2 cardiac myoblastic cell line (purchased from American Type Culture Collection, Manassas, VA, USA), and was used in subsequent experiments to knockout rat endogenous C3G mRNA ( Table 1). The non-target (NT) was also cloned into the pLenti-Cas-Guide lentiviral vector, and formed a NT recombinant lentiviral vector, which was used for negative control ( Table 1).
The C3G knockout (KO) and NT recombinant vectors were packaged into CRISPR (clustered regularly interspaced short palindromic repeats) / Cas9 recombinant lentiviruses (x 10 7 TU/ml) respectively in 293T cells through using Lenti-vpak Lentiviral Packaging Kit, and selected respectively. DNA sequencing experiment revealed that the sequence knocking out C3G gene was successfully inserted into the recombinant lentivirus.

Cell Culture in Vitro and Treatment with C3G Knockout Lentivirus and Hypoxia
Primary cardiac broblastic cell line (ScienCell Research Laboratories, Carlsbad, CA, USA) and H9C2 cardiac myoblastic cell line were cultured as described previously [7,9]. An equal number of the cells (1 x 10 5 ) were infected with C3G KO and NT recombinant lentiviruses respectively. The cells were divided into C3G KO and NT groups, and subjected to treatment with hypoxia [7,9]. The cells were nally randomly divided into NT, C3G KO, NT + hypoxia and C3G KO + hypoxia groups. The cells were washed with PBS, and harvested at indicated time points respectively for experimental analysis by RT-PCR (reverse transcription-polymerase chain reaction), Western immunoblot, MTT [3-(4,5)-dimethylthiahiazo(-z-y1)-3,5di-phenytetrazoliumromide] assay and ow cytometry.

RT-PCR Analysis for C3G mRNA
Total RNA was puri ed, quanti ed and reversely transcribed into cDNA as described previously [7,9]. The generated cDNA was ampli ed using speci c primers of rat C3G and GAPDH (glyceraldehyde-3phosphate dehydrogenase) genes respectively (Table 1). RT-PCR products were then analyzed.

MTT Assay for Cell Viability in Vitro
At the designated time points (24 h, 72 h) after infection, MTT working solution was added to cells in each well plate. Then subsequent performances were similar as described previously [7,9]. The cell proliferative rate was counted as (absorbance at 72 h -absorbance at 24 h) / absorbance at 24 h, and expressed as a percentage.

Flow Cytometry Analysis for Cell Apoptosis in Vitro
Seventy-two hours after infection, the apoptotic and necrotic cells were labeled respectively with annexin V uorescein isothiocyanate (FITC) and propidium iodide (PI). The labeled cells were then sorted by flow cytometry as described previously [7,9]. The cell apoptotic rates were calculated as the percentage of the number of the annexin V FITC positive and PI negative cells over the number of the total cells.

Myocardial Infarction Model and Experimental Protocol
Sixty Sprague-Dawley adult male rats (body weight (BW), 267.1 ± 19.2 g) were divided randomly into sham-operation (sham) + NT (n = 15), Sham + C3G KO (n = 15), MI + NT (n = 15) and MI + C3G KO (n = 15) groups respectively. MI model and sham-operation were established as described previously [8]. After establishment of MI model and sham-operation, surviving rats were as follows: in sham + NT CRISPR/Cas9 (n = 14), sham + C3G CRISPR/Cas9 (n = 13), MI + NT CRISPR/Cas9 (n = 11) and MI + C3G CRISPR/Cas9 (n = 10) groups. For each injection point, C3G KO or NT lentiviruses (25 μl) were then injected to myocardium at the following points; 3 points located in the infarcted central zones, and another 8 points in the border zones in each rat with MI or at the corresponding injection points in each reference rat in the above 4 groups respectively ( Fig. 3(d)). The chest was then closed, and the rats were treated as described previously [8]. The animals were supplied by the Laboratory Animal Center of Chongqing Medical University, Chongqing, China. All the animal experiments were approved by our institute laboratory animal committee.

Echocardiographic Measurements
Transthoracic doppler echocardiographic study was performed as described previously [8]. One week and 12 weeks after treatment with C3G KO or NT lentiviruses, the number of surviving rats were shown in each group (Table 2). Interventricular septal thickness (IVST), left ventricle (LV) posterior wall thickness (LVPWT), end-diastolic diameter (LVEDD) and ejection fraction (LVEF) were measured from the M-mode tracings of echocardiography Vivid 7 with a 3.5-MHz sector scan probe (General Electric Medical Systems, Milwaukee, WI, USA). The sonographer, who collected the data, was kept unaware of which treatments the rats were assigned to.

Histology and Collagen Volume Fraction Examination
Twelve weeks after sham-operation and MI, BW was measured again; all rats were then euthanized after the echocardiography. The heart was excised rapidly, and the LV was separated from other parts of the heart and weighed. The LV/BW ratio was calculated. Histological slice was obtained and stained with hematoxylin and eosin and picrosirius red respectively. Then infarction size and collagen volume fraction (CVF) were measured as described previously [8]. The myocardium in remote zone around the infarcted lesion scar in LV was frozen in liquid nitrogen, and stored in the -80°C freezer for Western immunoblot examination.

Detection of Cardiomyocytic Apoptosis in Myocardium
According to the instruction of TUNEL (terminal deoxynucleotidyl transferase-mediated dUTP nick endlabeling) in situ cell death detection kit (No. 11684817910, F. Hoffmann-La Roche Co. Ltd., CH-4070 Basel, Switzerland), the slice was incubated in TUNEL reaction mixture and converter-peroxidase. After the incubation period, the slice was added with diaminobenzidine substrate. The cardiomyocyte with brown nucleus was de ned as apoptotic cell, while the cell with blue nucleus was de ned as normal cell.
The apoptotic rate was calculated as the ratio of apoptotic to total cell count of cardiomyocytes in each eld, and an averaged apoptotic rate was calculated [10].

Statistical Analysis
Data is presented as mean ± standard deviation (SD). All analyses were performed using the SPSS22.0 statistical software. The statistical signi cance between the groups was evaluated by one-way ANOVA, then, in case of signi cance, by a two-side Tukey test for multiple comparisons. A value of p < 0.05 was considered statistically signi cant.

C3G was Deleted in Vitro Cardiac Myoblastic and Fibroblastic Cell Lines
C3G mRNA and protein in vitro cardiac myoblastic and broblastic cell lines were deleted in C3G KO and C3G KO + Hypoxia groups (All P < 0.05, Fig. 1(a, b), 2(a, b)). C3G mRNA and protein were diminished in NT + Hypoxia group as compared with NT group (All P < 0.05, Fig. 1(a, b), 2(a, b)).
After Deletion of C3G, Decreased p-ERK1/2, Bcl-2 and Increased Bax Proteins were Found in Vitro Cardiac

Myoblastic and Fibroblastic Cell Lines
Decreased p-ERK1/2 and Bcl-2 and increased Bax protein expression in vitro cardiac myoblastic and broblastic cell lines were detected after hypoxia both in C3G KO and NT groups (All P < 0.05, Fig. 1(b), 2(b)), while lower Bcl-2 and p-ERK1/2 and higher Bax protein expression in the cardiac myoblastic and broblastic cell lines were found in C3G KO group than in NT group, either treated with or without hypoxia (All P < 0.05, Fig. 1(b), 2(b)).
After Deletion of C3G, Reduced Cell Proliferation and Elevated Apoptosis were Con rmed in Vitro Cardiac Myoblastic and Fibroblastic Cell Lines Hypoxia inhibited cell proliferation and promoted apoptosis in vitro cardiac myoblastic and broblastic cell lines (All P < 0.05), as shown in Figure 1(c, d) and 2(c, d). C3G knockout group, in addition, exhibited reduction in number of viable cells and an elevated number of apoptotic cells, which were either treated with or without hypoxia (All P < 0.05, Fig. 1(c, d), 2(c, d)).
Knockdown of C3G Leaded to Decreased p-ERK1/2, Bcl-2 and Increased Bax Protein Expression in Surviving Myocardium After MI Twelve weeks after MI, the increase in C3G, p-ERK1/2 and Bcl-2 and decline in Bax protein expression in the surviving myocardium were documented in MI + NT and MI + C3G KO groups as compared to Sham + NT and Sham + C3G KO groups respectively (All P < 0.05, Fig. 3(a)). However, there were decrease in C3G, Bcl-2 and p-ERK1/2 and increase in Bax protein expression in myocardium in Sham + C3G KO and MI + C3G KO groups as compared to Sham + NT and MI + NT groups respectively (All P < 0.05, Fig. 3(a)).

Knockdown of C3G Caused Reduction of Cardiomyocytic Apoptosis and Collagen in Surviving Myocardium After MI
There were increase in CVF and cardiomyocytic apoptosis in the surviving myocardium in MI + NT and MI + C3G KO groups as compared to Sham + NT and Sham + C3G KO groups respectively (All P < 0.05, Fig.  3(b, c)), while the CVF and rate of cardiomyocytic apoptosis in the myocardium in MI + C3G KO group were less than that in MI + NT group (Both P < 0.05, Fig. 3(b, c)).
Knockdown of C3G Improved Cardiac Structural Remodeling and Function After MI LVEDD was increased, while LV/BW ratio and LVEF were decreased in MI + NT and MI + C3G KO groups as compared to Sham + NT and Sham + C3G KO groups 12 weeks after MI, respectively (All P < 0.05, Table 2, Fig. 4(a, b)); While smaller LVEDD and elevated LV/BW ratio and LVEF were found in MI + C3G KO group compared to MI + NT group (All P < 0.05, Table 2, Fig. 4(a, b)).

Discussion
We demonstrated that cardiomyocytic apoptosis, collagen deposition in the surviving myocardium after MI and LV end-diastolic diameter were decreased; While LV weight / body weight ratio and LVEF were increased with C3G knockout lentiviral treatment via down-regulation of pro-survival molecules including C3G, p-ERK1/2 and Bcl-2, and up-regulation of pro-apoptotic Bax expression level in the surviving myocardium. However, in our study, knockout of C3G was discovered to delete C3G and diminish cellular proliferation in vitro cardiac myocytic and broblastic cell lines respectively via down-regulation of p-ERK1/2 and Bcl-2, and up-regulation of Bax protein expression. Knockdown of pro-survival C3G in myocardium may mitigate surviving myocardial remodeling after MI, possibly via down-regulation of p-ERK1/2, Bcl-2 and up-regulation of Bax in vivo cardiac myocytes and broblasts in the surviving myocardium.
With the reparative brosis of infarcted myocardium, the remodeling of the surviving myocardium in the border and remote zones around the infarcted lesion is associated with detrimental processes of LV function and poor prognosis. Terminally differentiated cardiomyocyte in vivo makes up the most (≈90%) of cardiac mass, although it is usually unable to proliferate. Cardiac broblast constitutes the majority (≈70%) of cells in normal cardiac muscle, because it can proliferate [1]. Therefore, the over-proliferated cardiac broblasts are very important for the progressive myocardial remodeling after MI [1].
The surviving myocardial remodeling results from global activation of sympathetic nervous and reninangiotensin-aldosterone systems (RAAS) in the compensatory responses induced by MI [11,12]. After MI, the activated autonomic nervous system and RAAS have been proved to promote cell survival both in vivo cardiomyocytes and cardiac broblasts via up-regulation of pro-survival gene expression [13,14]. The over-activation of sympathetic nervous system and RAAS result in ongoing cardiac broblast proliferation, myocardial brosis, and consequently increase the apoptosis in vivo cardiomyocytes, leading to the progressive surviving myocardial remodeling after MI [1].
In contrast, experimental studies have con rmed that globally non-cardiac selective blockage of the activated sympathetic nervous system and RAAS is able to relieve apoptosis of cardiomyocytes and other harmful process in the surviving myocardium after MI [15], although inhibition of the activated autonomic nervous system and RAAS is considered to result in apoptosis in vivo cardiomyocytes and cardiac broblasts via up-regulation of pro-apoptotic gene expression [15].
A previous study has proved that cardiomyocyte-speci c knockout of a pro-apoptotic gene attenuated myocardial remodeling induced by pressure overload or MI [2]; However, systemic knockout of the same pro-apoptotic gene can augment LV remodeling process [2].
Integrin pathway links extracellular matrix substances to cytoskeleton in cells, and senses mechanical stretch, and translates it into intracellular biochemical signaling [3,16]. Cardiomyocyte-speci c deletion of pro-survival integrin β1 subunit and its downstream molecules such as FAK may increase apoptosis in vivo cardiomyocytes and aggravate cardiomyopathy induced by MI, isoproterenol or pressure overload [4,5]. Knockout of cardiomyocyte-speci c pro-survival FAK or integrin-linked kinase (ILK) even spontaneously results in dilated cardiomyopathy with reduced LV function [17,18]. However, a previous study has demonstrated that globally systemic inhibition of pro-survival FAK attenuated cardiomyopathy and preserved cardiac function in post-MI [19]. Therefore, knockdown of the integrin pathway pro-survival component C3G in myocardium may have a protective effect against myocardial remodeling after MI.
Pro-survival C3G protein is a family member of guanine nucleotide exchange factor in the integrin pathway. After stimulation by signals through integrins, C3G binds with N-terminal Src homology 3 domain of its upstream adaptor CrkL, and subsequently activates its downstream GTPases such as Rac1 [20]. On the basis of existing evidences, pro-survival C3G can inhibit cell apoptosis in vitro cardiomyocytes via up-regulation of its downstream pro-survival AKT, p-ERK1/2, Bcl-2 and downregulation of pro-apoptotic Bax [7,9]. In addition, our current experimental study has replenished that knockout of pro-survival C3G by a speci c knockout lentivirus can promote cell apoptosis in vitro cardiac broblastic cell line via down-regulation of p-ERK1/2, Bcl-2, and up-regulation of Bax. A previous study has con rmed that pro-survival expression of C3G, p-ERK1/2 and Bcl-2 were up-regulated, while proapoptotic Bax expression was down-regulated in the surviving myocardium after MI [8]. These molecular expression changes possibly resulted from the global activation of sympathetic nervous system and RAAS induced by MI. However, for the rst time, we have documented that knockdown of pro-survival C3G in myocardium mitigated cardiomyocytic apoptosis, myocardial brosis and remodeling around the infarcted lesion, possibly via down-regulation of pro-survival p-ERK1/2, Bcl-2, and up-regulation of proapoptotic Bax mainly in vivo cardiac broblasts in the surviving myocardium.
The underlying mechanisms for the mitigation of the surviving myocardial remodeling after MI by knockdown of pro-survival C3G in myocardium are as follows: Cardiac broblast in vivo may play a major role in cell survival in vivo cardiomyocytes after MI [1]; While the proliferative cardiac broblasts in vivo may be more susceptible to the treatment with C3G knockdown lentivirus than non-proliferative cardiomyocytes in vivo. Knockdown of pro-survival C3G may mainly depress the proliferation of cardiac broblasts in the myocardium, and alleviate myocardial brosis, and consequently result in inhibition of cardiomyocytic apoptosis and myocardial remodeling after MI, although knockdown of C3G may lead to cellular apoptosis both in vivo cardiac broblasts and myocytes in the surviving myocardium. Appropriate-intensity transient knockdown of pro-survival C3G, which may mainly inhibit cardiac broblast proliferation, could be developed as a novel potential therapeutic measure to relieve the surviving myocardial remodeling, brosis, ischemic cardiomyopathy and cardiac dysfunction. Moreover, we have used experimental results in vitro H9C2 cardiac myoblasts to interpret mechanisms for cardiac structural phenotypic changes in vivo. As considering that the H9C2 cell has some different phenotypes from adult primary cardiomyocyte, the conclusions made in this study should be taken cautiously.
In this study, the recombinant lentivirus can knockout C3G in vitro cardiac myocytic and broblastic cell lines. However, it can not knockout all of C3G in the myocardium, because C3G knockout lentivirus was administered just transiently in the myocardium. Nevertheless, it can knockdown part of C3G in the myocardium and positively affect the myocardial remodeling. The effects of permanently systemic C3G knockout on surviving myocardial remodeling after MI are still unclear. The role of pro-survival C3G in myocardial remodeling remains to be rede ned in the future via establishment of animal transgenic model with systemic knockout of C3G.

Conclusion
Our current data demonstrate that knockdown of pro-survival C3G in myocardium mitigates surviving myocardial remodeling after MI, possibly via down-regulation of p-ERK1/2, Bcl-2 and up-regulation of Bax in vivo cardiac myocytes and broblasts in the surviving myocardium.  Tables   Table 1 The