MicroRNA miR-301a is a novel cardiac regulator of Cofilin-2

Calsarcin-1 deficient mice develop dilated cardiomyopathy (DCM) phenotype in pure C57BL/6 genetic background (Cs1-ko) despite severe contractile dysfunction and robust activation of fetal gene program. Here we performed a microRNA microarray to identify the molecular causes of this cardiac phenotype that revealed the dysregulation of several microRNAs including miR-301a, which was highly downregulated in Cs1-ko mice compared to the wild-type littermates. Cofilin-2 (Cfl2) was identified as one of the potential targets of miR-301a using prediction databases, which we validated by luciferase assay and mutation of predicted binding sites. Furthermore, expression of miR-301a contrastingly regulated Cfl2 expression levels in neonatal rat ventricular cardiomyocytes (NRVCM). Along these lines, Cfl2 was significantly upregulated in Cs1-ko mice, indicating the physiological association between miR-301a and Cfl2 in vivo. Mechanistically, we found that Cfl2 activated serum response factor response element (SRF-RE) driven luciferase activity in neonatal rat cardiomyocytes and in C2C12 cells. Similarly, knockdown of miR301a activated, whereas, its overexpression inhibited the SRF-RE driven luciferase activity, further strengthening physiological interaction between miR-301a and Cfl2. Interestingly, the expression of SRF and its target genes was strikingly increased in Cs1-ko suggesting a possible in vivo correlation between expression levels of Cfl2/miR-301a and SRF activation, which needs to be independently validated. In summary, our data demonstrates that miR-301a regulates Cofilin-2 in vitro in NRVCM, and in vivo in Cs1-ko mice. Our findings provide an additional and important layer of Cfl2 regulation, which we believe has an extended role in cardiac signal transduction and dilated cardiomyopathy presumably due to the reported involvement of Cfl2 in these mechanisms.


Materials and methods
Generation and characterization of Calsarcin-1 knock out in C57BL/6 pure background The Calsarcin-1 knockout mice were originally created and characterized in 2004 by Frey et al. in a mixed background [13]. The backcrossing in a pure C57BL/6NCr background was performed by Schoensiegel et al. in 2007 [29]. Primer pairs used for the genotyping were: neo_F: 5'-gat gcg gtg ggc tct atg gct tct gag gc-3', CS1_F: 5'-cag tgt gtt cta tta ccc agg ctg tc-3' and CS1_R: 5'-gtc ctc aca act aat tca tgt aca gat g-3'. All the animal experiments were carried out in strict accordance to the ethical guidelines by MELUR (Ministry of Energy, Agriculture, the Environment and Rural Areas). Mice were given access to the food and water 'ad libitum' and maintained under 12 h dark and light cycle temperature and air controlled rooms.

Echocardiography
Echocardiography was carried out on anaesthetized mice with Isoflurane (2.5 ppm) on Vivid 7 Pro Ultrasound System (GE healthcare). The examiner was blinded for the genotype. Mice were killed immediately after the echocardiography by cervical dislocation and organs were harvested, weighed and stored at -80˚C until further processing.
miR-301a regulates Cofilin-2 in the heart Microarray analysis RNA preparation and hybridization. Total RNA from Cs1-ko and wild-type mice was isolated using QIAzol lysis reagent according to the manufacture's instruction (Qiagen). The quality of total RNA was checked by gel analysis using the total RNA Nano chip assay on an Agilent 2100 Bioanalyzer (Agilent Technologies). Only samples with RNA index values >8.5 were selected for expression profiling. RNA concentrations were determined using the NanoDrop spectrophotometer (NanoDrop Technologies). Biotin-labeled cRNA samples for hybridization on Illumina Mouse Sentrix-6 BeadChip arrays (Illumina, Inc.) were prepared according to Illumina's recommended sample labeling procedure based on the modified Eberwine protocol [30]. In brief, 250 ng total RNA was used for complementary DNA (cDNA) synthesis, followed by an amplification/labeling step (in vitro transcription) to synthesize biotinlabeled cRNA according to the MessageAmp II RNA Amplification kit (Ambion, Inc.). The cRNA was column purified according to TotalPrep RNA Amplification Kit, and eluted in 60 μl of water. Quality of cRNA was controlled using the RNA Nano Chip Assay on an Agilent 2100 Bioanalyzer and spectrophotometrically quantified (NanoDrop). Hybridization was performed at 58˚C, in GEX-HCB buffer (Illumina Inc.) at a concentration of 100 ng cRNA/μl, unsealed in a wet chamber for 20h. Spike-in controls for low, medium and highly abundant RNAs were added, as well as mismatch control and biotinylation control oligonucleotides. Microarrays were washed twice in E1BC buffer (Illumina Inc.) at room temperature for 5 minutes. After blocking for 5 min in 4 ml of 1% (wt/vol) Blocker Casein in phosphate buffered saline Hammarsten grade (Pierce Biotechnology), array signals were developed by a 10-min incubation in 2 ml of 1 μg/ml Cy3-streptavidin (Amersham Biosciences) solution and 1% blocking solution. After a final wash in E1BC, the arrays were dried and scanned.
Scanning and data analysis. Microarray scanning was carried out using a Beadstation array scanner, with settings adjusted to a scaling factor of 1 and PMT settings at 430. Data extraction was done for all beads individually, and outliers were removed when > 2.5 MAD (median absolute deviation). All remaining data points were used for the calculation of the mean average signal for a given probe, and standard deviation for each probe was calculated. Scanning data was analyzed by normalization of signals using the quantile normalization algorithm without background subtraction, and differentially regulated genes were defined by calculating the standard deviation differences of a given probe in Cs1-ko vs WT mice comparison. Microarray data is deposited to GEO databank under the accession number GSE100851.

Histology
Mouse hearts were molded into Tissue-Tek Cryomolder (Sakura Finetek), and frozen on dry ice. Cryosections of 7 μm thickness were used for the histology. Lectin staining was carried out using FITC conjugated lectin from Triticum vulgaris (wheat), according to the manufacturer's instructions. Images were captured on BZ-9000 immunofluorescence microscope (Keyence) and cross-sectional area of the cardiomyocytes was analyzed with ImageJ software (version 1.46). The extent of fibrosis was measured by Sirius-red/fast green staining as described earlier [31,32]. Images captured on BZ-9000 Keyence microscope were analyzed by BZ-II Analyzer software to measure the fibrotic area.

Cloning of rat Cofilin-2
Rat Cofilin-2 was cloned using rat heart cDNA and Invitrogen™ Gateway1 cloning technology (all Thermo Fisher Scientific). Primers used were, attB_Cofilin-2_F: 5'-ggg gac aag ttt gta caa aaa agc agg ctt cga agg aga tag aac cat ggc atc tgg agt tac agt gaa tg-3', attB_Cofilin-2_R: 5'-ggg gac cac ttt gta caa gaa agc tgg gtc cta cag tgg ctt tcc ttc cag gga-3'. PCR product was recombined using BP Clonase™ II into the Gateway1 pDONR™221 entry vector which upon sequence confirmation, transferred via LR Clonase™ II to Gateway1 pAd/CMV/V5-DEST destination vectors, for transfection or transduction, respectively Isolation and culture of NRVCM and fibroblasts NRVCMs were isolated as described before [31,33]. In short, 1-2 days old Wistar rats (Charles River) were decapitated to obtain the hearts, which were stored in ice cold ADS buffer (120 mmol/L NaCl, 20 mmol/L HEPES, 8 mmol/L NaH2PO4, 6 mmol/L glucose, 5 mmol/L KCl and 0.8 mmol/L MgSO4 (pH 7.4)). The ventricles were minced with the scissor and digested 4-5 times in sterile ADS buffer containing collagenase type II (0.5 mg/mL, Worthington Biochemical Corporation) and pancreatin (0.6 mg/mL, Sigma-Aldrich) to separate the cells. By performing a gradient centrifugation using Percoll (GE Healthcare), NRVCM were purified and separated from cardiac fibroblasts. The NRVCM were incubated at 37˚C in complete DMEM media (DMEM supplemented with 10% FCS, 100 U/mL penicillin, 100 μg/mL streptomycin and 2 mmol/L L-glutamine (Thermo Fisher Scientific)) for 24h before virus transduction or other downstream treatments. Fibroblast fraction from the gradient step mentioned above was plated in 6x well culture plates at 37˚C for 4 h in complete DMEM medium, followed by media aspiration and washing with PBS to remove floating contaminant cells. Adhered fibroblasts were incubated for further 96 h in complete DMEM at 37˚C.

Cell surface area measurement
Immunofluorescence staining and cell surface area measurements were performed as detailed earlier [34]. Briefly, cardiomyocytes cultured on coverslips in 12x well culture plates, either transduced with Cfl2 overexpression/knockdown adenoviral particles, or transfected with miR-301a mimic/inhibitor, were washed 2x with PBS and fixed with 4% paraformaldehyde for miR-301a regulates Cofilin-2 in the heart 10 min. Fixed cells were washed 2x with PBS followed by a common step of permeabilization and blocking with 0.1% Triton X-100 in 2.5% BSA for 1 h at room temperature. Cells were then incubated for 1 h with primary anti-α-actinin antibody (1:200; Sigma-Aldrich), 5x washes with PBS, followed by the incubation with respective secondary antibody conjugated to Cy3 (Dianova) and DAPI for nuclear staining. After washings with PBS, coverslips were mounted on glass-slides using Fluoromount (Biozol). Immunofluorescence images were captured using BZ-9000 microscope (Keyence). Cell surface area was measured using HybridCell-Count module BZ-II Analyzer software (Keyence).

Generation of miR-301a predicted binding site mutants and luciferase assay
Murine genomic DNA was used as a template for cloning 3' untranslated regions (3'UTR) of the genes of interest into pmirGLO vector (Promega). Mutants of the predicted binding sites in the 3'UTR of Cofilin-2 were generated by QuikChange1 II site-directed mutagenesis kit (Agilent Technologies) by replacing bases 2'-6' of the seed sequence (5'GCACT-3' to 5'-TAC AG-3). C2C12 cells (24-well format, 30.000 cells/well) were co-transfected with 20ng pmiR-GLO vector per well and mimics (20pmol/well, mirVana™ miRNA Mimic, Negative Control #1 and miR-301a-3p, Thermo Fisher Scientific) 24h post cell seeding and incubated for additional 48h with one change of media. Cells were then lysed using passive lysis buffer provided with the Dual-Glo1 Luciferase Assay System (Promega), and chemiluminescence was measured using the Infinite M200Pro (Tecan) in a 96-well format.

SRF-reporter assay
The luciferase reporter assay was carried out in 24-well format in C2C12 Cells (ATCC1) or in 12-well format in NRVCMs. As an indicator for the serum response factor (SRF) activity the pGL4.34[luc2P/SRF-RE/Hygro] vector by Promega (20ng/well) was used, which contains a SRF responsive element promotor prior to the firefly luciferase gene. For normalization, Renilla luciferase containing pGL4.74[hRluc/TK] vector by Promega (5ng/well) was used. These vectors were co-transfected with either Cofilin-2 or miR-301a knockdown (Negative Control siRNA, Qiagen1; Cofilin 2 siRNA (m): sc-37026, Santa Cruz Biotechnology), or overexpression constructs (n = 6) using Lipofectamine1 2000 Transfection Reagent (Thermo Fisher Scientific) 24h after seeding the C2C12 cells. Cells were cultured for 48h, lysed and luciferase assay was performed as mentioned above for 3'UTR mutants. Reporter assays in the NRVCM were performed using SRF-RE driven firefly luciferase adenovirus construct as described in [32]. Adenovirus encoding Cfl2 or a synthetic microRNA specifically targeting Cfl2 was used for the overexpression or knockdown of Cfl2, respectively.

Statistical analysis
The error bars represent the standard error of the mean (SEM), unless stated otherwise. The statistical analysis was carried out using either two-tailed student's t-test when comparing two groups. Equal distribution of the cell size measurement data was tested by Shapiro-Wilk test, and samples were compared by Kruskal-Wallis test (one-way ANOVA on ranks). P-values 0.05 were considered as statistically significant.

Results
Calsarcin-1 deficient mouse in pure C57BL/6 genetic background displays dilated cardiomyopathy phenotype We have earlier reported that mice with Calsarcin-1 null mutation in a mixed genetic background do not exhibit any overt basal cardiac hypertrophy phenotype, however show accelerated hypertrophic cardiomyopathy in response to pathological biomechanical stress [13]. We then back-crossed these mice for more than 10 generations to obtain Calsarcin-1 deficiency in a pure C57BL/6 genetic background (henceforth these mice will be referred as Cs1-ko). Using echocardiography analysis we found that Cs1-ko mice showed severely reduced fractional shortening ( Fig 1A) and intra-ventricular diameter (Fig 1B), whereas, left ventricular end-diastolic diameter was significantly increased (Fig 1C). Surprisingly however, there was no difference between heart weight to body weight (Fig 1D), and heart weight to tibia length ratios (Fig 1E) in Cs1-ko mice. Moreover, in line with heart weight to body weight ratios, cardiomyocyte cell surface area was also unaltered between both the genotypes (Fig 1F and 1G). Furthermore, lack of Calsarcin-1 did not increase fibrosis as evident from the unaltered fibrosis and unchanged expression of fibrosis markers, Collagen I and III (Fig 1H-1J). Altogether, these data suggests that Cs1-ko mice in pure C57BL/6 background displays strict dilated cardiomyopathy phenotype.

MicroRNA miR-301a is downregulated in Cs1-ko mice
To understand the molecular causes of dilated cardiomyopathy phenotype of Cs1-ko mice, we performed comparative microRNA microarray analysis of Cs1-ko with wild-type mouse heart. Microarray data revealed that several microRNAs were dysregulated in Cs1-ko mice (S1 Table), including miR301a, which was maximally downregulated, whereas, miR-298 was highly upregulated in Cs1-ko mouse hearts (schematically depicted in Fig 2A). We further miR-301a regulates Cofilin-2 in the heart validated the expression of miR-301a and miR-298 by quantitative real-time PCR (qPCR) in independent set of mouse cohort to confirm its downregulation (Fig 2B and 2C).
miR-301a targets Cofilin-2 in cardiomyocytes and in Cs1-ko mice Next, we used online prediction databases to identify possible miR-301a targets which resulted into hundreds of putative targets. We selected few of the targets including Cofilin-2 (Cfl2), Activin A Receptor Type 1 (ACVR1), Quaking (Qk), and Chloride Voltage-Gated Channel-3 (CLCN3) for further validation using pmirGLO Dual-Luciferase miRNA Target Expression Vector and assay system. Luciferase activity was found to be reduced only in Cfl2 construct (Fig 3A) which led us evaluate its possible miR-301a binding sites in details. We found four putative miR-301a binding sites in 3'UTR of Cfl2 depicted in S1A Fig which we mutated by site directed mutagenesis and studied for the validation of miR-301a binding. Mutation in two of the four putative binding sites (binding sites at position 370 and 1030 of the 3'UTR) prevented the reduction in luciferase activity by miR-301a overexpression suggesting that these two binding sites are responsible for miR-301a effect on Cfl2 expression (Fig 3B). We then confirmed if miR-301a targets Cfl2 in vitro in neonatal rat ventricular cardiomyocytes (NRVCM). As anticipated, overexpression of miR-301a reduced while its knockdown increased Cfl2 expression determined by immunoblotting (Fig 3C-3E). Finally, we found a strong in vivo correlation between downregulation of miR-301a to the upregulation of Cfl2 in Cs1-ko mice (Fig 3F-3H), suggesting a possible physiological importance of miR-301a in regulating Cfl2 in the heart. determined by quantitative real-time PCR. The statistical analysis was carried out using two-tailed student's t-test. *: p<0.05, †: p<0.01, ‡: p<0.001, n.s.: non-significant. https://doi.org/10.1371/journal.pone.0183901.g001

Fig 2. MicroRNA miR-301a is downregulated in Cs1-ko mice.
Microarray analyses were performed on Illumina Mouse Sentrix-6 BeadChip arrays (Illumina, Inc.) using total RNA isolated from Calsarcin knockout (Cs1-ko) and wild-type (WT) mice. Microarray scanning was done using a Beadstation array scanner and analyzed by normalization of the signals using the quantile normalization algorithm without background subtraction. Differentially regulated microRNAs were defined by calculating the standard deviation differences of a given probe in Cs1-ko and WT genotypes. (A) Bar graph presenting few selected dysregulated microRNAs in Cs1-ko mice compared to WT mice. MiR-301a was identified the most downregulated microRNA, whereas, miR-298 was highly upregulated (N = 4 each), which was confirmed in independent cohort by quantitative real-time PCR for miR-301a (B), and miR-298 (C) (N = 5 (WT), and 6 (Cs1-ko)). Statistical analysis was carried out using two-tailed student's t-test. *: p<0.05, †: p<0.01.
https://doi.org/10.1371/journal.pone.0183901.g002 Expression of miR-301a and Cfl2 is higher in cardiomyocytes than fibroblasts Tissue distribution pattern for miR-301a determined by qPCR revealed its ubiquitous expression in the heart, brain, skeletal muscle, etc. (S2 Fig). To discriminate the expression of miR-301a and Cfl2 in cardiac major cell types, we studied their expression levels in isolated fibroblasts and cardiomyocytes from neonatal rat ventricles. We found that the expression of both miR-301a and Cfl2 was higher (>2 fold) in cardiomyocytes compared to fibroblasts (Fig 4A-4C). Expression of α-actinin and vimentin were used as markers to determine the purity of cardiomyocytes and fibroblasts, respectively (Fig 4D-4F).

miR-301a and Cfl2 oppositely regulates Rho-mediated SRF signaling but not cellular hypertrophy
The ADF/cofilin family proteins are actin-binding that are actively involved in actin remodeling. RhoA, a small Rho family GTPase and an activator of serum response factor (SRF) signaling has also been shown to play an essential role in the control of myocardial fibrosis by regulating cofilins [35]. We therefore hypothesized that Cfl2 plays an essential role in RhoA-SRF activation and miR-301a will negatively affect this activation by regulating Cfl2 expression. To test this hypothesis, we performed SRF-response element (SRF-RE) driven firefly luciferase activity assay by either overexpressing or knocking down Cfl2/miR-301a in C2C12 cells. Overexpression of Cfl2 alone did not regulate the luciferase activity (Fig 5A). Surprisingly however, co-expression of Cfl2 and RhoA (both constitutively active and native RhoA) dramatically increased the activation of luciferase reporter (Fig 5A). In contrast, knockdown of Cfl2 strongly attenuated the luciferase activity not only at basal level, but also inhibited the RhoA-mediated SRF-RE activation (Fig 5B) suggesting that Cfl2 is necessary and sufficient for the SRF-signaling activation through RhoA. In contrast, downregulation of miR-301a resulted in the activation, whereas, its overexpression significantly blunted the basal as well as RhoAmediated activation of SRF-RE signaling (Fig 5C and 5D). Importantly, strong effect on RhoA-mediated SRF activation via Cfl2 overexpression observed in C2C12 cells was consistent in cardiomyocytes as well (Fig 5E). Knockdown of Cfl2 also significantly blunted the activation of luciferase activity, both at baseline as well as in the presence of RhoA (Fig 5F). Similarly, inhibition of miR-301a expression effectively accelerated SRF-signaling at basal level (Fig 5G), whereas, its overexpression significantly abrogated the RhoA-mediated activation of SRF activity ( Fig 5H). Finally, we evaluated if Cfl2/miR-301a influences cellular hypertrophy by measuring cell surface area. To our surprise, neither overexpression nor knockdown of Cfl2 or miR-301a affected the cell size in neonatal rat cardiomyocytes (Fig 5I-5L).

Calsarcin-1 deficiency upregulates SRF in mouse heart
Due to the significant effect of Cfl2 and miR-301a on RhoA/SRF-signaling in C2C12 cells/ NRVCM, and observed upregulation of Cfl2 and downregulation of miR-301a in Cs1-ko mice, these binding sites act as targets for miR-301a binding. Mutations in binding sites 370 and 1030 resulted in loss of luciferase activation, clearly suggesting that these two sites act as binding sites for miR-301a (N = 4). (C) Immunoblot showing that the overexpression of miR-301a mimic downregulated (N = 3 each, original uncropped blots are shown in S1B & S1C Fig, respective densitometry is shown as a bar graph in D), whereas, overexpression of miR-301a inhibitor upregulated (respective densitometry is shown as a bar graph in E) the protein levels of Cfl2. (F) Cfl2 was found upregulated in Cs1-ko mice as depicted in an immunoblot at protein (N = 3 (WT), 4 (Cs1-ko), original uncropped blots are shown in S1D Fig, respective densitometry is shown as a bar  graph in G), and at transcript level determined by immunoblotting and quantitative real-time PCR (H), respectively. All experiments are repeated at least two times and the statistical analysis was carried out using two-tailed student's t-test. *: p<0.05, †: p<0.01, n.s.: non-significant.
we determined the SRF and RhoA levels in Cs1-ko mice. We found considerable increase in the expression of both SRF and RhoA in Cs1-ko mice compared to the respective wild-type littermates (Fig 6A-6C). Increased levels of SRF in the heart is known to cause robust activation of SRF signaling, and in the absence of other stimuli, SRF upregulation is sufficient to cause cardiomyopathy [17]. Along these lines, the expression of fetal genes nppa, nppb, and myh7, which are also direct targets of SRF transcription factor were dramatically increased in the heart of Cs1-ko mice (Fig 6D-6F). Furthermore, actc1 was also highly upregulated (Fig 6G), which is a bona fide target of SRF. Taken together, these data suggests an increased activation of SRF-signaling in Cs1-ko mice. miR-301a regulates Cofilin-2 in the heart Fig 5. miR-301a and Cfl2 oppositely regulates Rho-mediated SRF signaling. Cfl2 or miR-301a were either overexpressed or knocked-down using respective vectors, mimic, or inhibitor transfection in C2C12 cells together with a luciferase construct carrying SRF-RE driven firefly luciferase. Overexpression of Cfl2 additively increased the luciferase activation by either constitutively RhoA (A), whereas, its siRNA led to inhibition of luciferase activation, basal, as well as in presence of RhoA (B). Knockdown of miR-301a also increased the luciferase activation (C), and the overexpression of miR-301a mimic significantly blunted the luciferase activity (D). For SRF-gene reporter assays in NRVCM, Cfl2 was overexpressed using adenovirus encoding rat Cfl2, whereas, its knockdown was achieved using adenovirus encoding synthetic microRNA specifically targeting Cfl2. Expression of miR-301a was modulated in NRVCM same as in C2C12 cells. Like in C2C12 cells, Overexpression of Cfl2 in NRVCM also exhibited positive effect on the activation of luciferase activity (E); knockdown of Cfl2 on the other hand significantly inhibited the activation of SRF-RE driven luciferase activity (F). Consistently, altered expression of miR-301a by treating NRVCM with miR-301a inhibitor (G) or mimic (H) oppositely affected the luciferase activation. Increased expression of Cfl2 in NRVCM did not alter cell surface area (I). Knockdown of miR-301a also led to no effect on cell size (J). Similarly, siRNA mediated knockdown of Cfl2 (K) or overexpression of miR-301a too did not alter cell surface area (L). N>500 for cell size measurements, and N = 6 luciferase assays. All experiments have been repeated at least twice. The statistical analysis was carried out using two-tailed student's t-test. *: p<0.05, ‡: p<0.001, n.s.: non-significant.

Discussion
Cofilin-2, a member of the ADF/cofilins family proteins that regulates actin dynamics, is essential for maintaining actin filament length in muscle sarcomere [26]. Through a microRNA microarray we identified miR-301a is significantly downregulated in Calsarcin-1 deficient mice that present DCM phenotype in a pure C57BL/6 background. Functional characterization of this microRNA in cardiac perspective revealed that miR-301a targets and regulates Cfl2 in vitro in neonatal rat cardiomyocytes, and in vivo in Cs1-ko mice. Furthermore, our in vitro data indicated that miR-301a attenuates RhoA-mediated activation of SRF signaling via targeting Cfl2 in vitro without affecting cellular hypertrophy. Importantly, RhoA, SRF and its target genes were strongly upregulated in Cs1-ko mice where Cfl2 is upregulated and miR-301a is downregulated, suggesting a possible involvement of Cfl2 in SRF activation in vivo.
Natural genetic diversity in human population determines the extent of disease phenotype, including cardiac diseases, suggesting that genetic background also plays an important role in pathology. Similarly, increasing number of studies suggests the importance of selecting an appropriate mouse genetic background for cardiac evaluations [36,37]. For example, ApoE knockout mouse displays severe atherosclerotic phenotype in C57BL/6 compared to FVB/J, whereas, MLP knockout has dramatically increased heart failure rate in the 129/Sv than C57BL/6 genetic background [38,39]. Similarly, Lygate et al. ascertained that mitochondrial miR-301a regulates Cofilin-2 in the heart creatine kinase knockout mice do not display any cardiac phenotype in pure C57BL/6 genetic background [40], which earlier reportedly led to LV dysfunction and hypertrophy [41]. In congruence with these reports, we previously found that Calsarcin-1 knockout mice do not exhibit cardiac hypertrophy phenotype at baseline in mixed genetic background notwithstanding striking upregulation of fetal genes and contractile dysfunction; however, these mice displayed strict DCM phenotype, devoid of hypertrophy, even when back-crossed for more than 10 generations to obtain the desired mutation in pure C57BL/6 background. Therefore, to identify the molecular causes behind DCM phenotype despite lack of hypertrophy, we performed microRNA microarray which resulted in identification of several microRNAs that were differentially regulated in Cs1-ko mice.
MiR-301a, the most-downregulated microRNA in our screen has previously been associated strongly with many human cancers including prostate cancer, malignant melanoma, osteosarcoma, etc. [61][62][63][64]. Although miR-301a is significantly expressed in the heart and other tissues, no cardiac role of this ubiquitously expressed microRNA is known yet. Moreover, we here found that miR-301a was highly expressed in isolated cardiomyocyte compared to fibroblasts, suggesting a cell-type specific function for this microRNA. Most interestingly, we discovered Cofilin-2 as one of the putative targets of miR-301a through microRNA target database search, which we further validated through series of experiments including luciferase assays, site directed mutagenesis of possible binding sites, and by manipulation of miR-301a expressions in neonatal rat cardiomyocytes. To strengthen these in vitro findings, we found a strong inverse correlation between Cfl2 and miR-301a expression in Cs1-ko mice.
Cfl2 belongs to the family of actin severing proteins, primarily expressed in muscle, and also in the brain and liver [23], and is critical for the maintenance of sarcomeric actin dynamics and length [26,27]. However, our data indicates that Cfl2 overexpression does not significantly alter the cell surface area of isolated NRVCM. Ablation of miR-301a resulted into similar effects as observed with the Cfl2 overexpression, pertaining to the increased levels of Cfl2 upon miR-301a knockdown. We also found that Cfl2 increases the RhoA-mediated SRF activation, whereas, miR-301a upregulation is sufficient to antagonize these effects (Fig 7). Moreover, increased expression of SRF/RhoA in Cs1-ko mice and target genes of SRF suggesting activation of SRF signaling in these mice. Of note, Cs1-ko mice do not exhibit any signs of hypertrophy notwithstanding upregulation of fetal gene program and contractile dysfunction. Similar findings were observed when we modulated the expression of Cfl2 or miR-301a. Overexpression of Cfl2 or knockdown of miR-301a though resulted in the activation of SRF signaling, neither of these treatments caused hypertrophy. These similar in vitro findings points towards possible involvement of Cfl2-miR301a in the pathophysiology of Cs1-ko mice, at least partially. Recently, increased expression and phosphorylation of Cfl2 has been linked with DCM and myocardial aggregates [28]. Surprisingly, deletion of Cfl2 in mice also led to progressive muscle degeneration and appearance of sarcoplasmic protein aggregates [65]. Both these findings highlight the importance of Cfl2 in muscle and heart pathophysiology. Here, we show that miR-301a regulates the expression as well as physiological function of Cfl2 in cultured cardiomyocytes, which needs additional in vivo evaluations and validation by gainand loss-of-function studies. We therefore propose to explore the possibility of use of miR-301a manipulations for therapeutic intervention to target cardiac disorders caused due to deregulation of Cfl2. Original uncropped blots are shown for Fig 3C (B, C), and 3F (D). (DOCX)