Graphical abstract

   graphic file with name fx1.jpg
   [31]Open in a new tab

Highlights

     * •
       Nav1.5 deficiency enhances cardiac fibroblast activity and
       attenuates miR-452-5p
     * •
       MiR-452-5p regulates fibrogenesis by targeting the TGF-β/SMAD4 axis
     * •
       AAV miR-452-5p reduces cardiac fibrosis and improves heart failure
       in rats
     __________________________________________________________________

   Cardiovascular medicine; Molecular biology; Cell biology

Introduction

   Cardiac arrhythmia affects 33 million people globally and is the
   prevalent cause of myocardial fibrosis leading to heart failure
   (HF).[32]^1 Predominantly, myocardial fibrosis occurs due to
   abnormalities in ion channels and structural organization. The former
   contributes to the development of cardiac fibrosis by affecting the
   initiation and propagation of electrical stimuli, for instance, the
   reduced I[Na] significantly increased collagen deposition in
   SCN5A-knockout mice.[33]^2 SCN5A gene produces α-subunit of the cardiac
   sodium voltage-gated channel (Nav1.5) arbitrates the rapid Na^+ influx,
   which is crucial for the upstroke of action potential and excitation of
   cardiac cells.[34]^3^,[35]^4 Structural changes in Nav1.5 channel
   syndicate a spectrum of arrhythmic disorders that lead to sudden
   cardiac death. Studies reported that the loss of SCN5A gene expression
   in the human and heterozygous SCN5A mouse models favors the tissue
   remodeling vulnerability of the heart, and these structural
   abnormalities are secondary to cardiac sodium
   channelopathies.[36]^5^,[37]^6^,[38]^7^,[39]^8 In addition to this,
   under diminished expression of Nav1.5, distinctive fibrosis has been
   reported in calcineurin-induced murine cardiac hypertrophic ventricles
   and Cx43 knockout mice.[40]^9^,[41]^10 Sodium channel
   dysfunction-dependent onset of cardiac fibrosis and collagen production
   requires the activation of the transforming growth factor β (TGF-β)
   pathway. The inhibition of I[Na] triggered the abnormal upregulation of
   TGF-β receptors,[42]^11 and SCN5A-knockout mice also exhibited
   increased expression of TGF-β transcripts.[43]^12 Under activated
   TGF-β, fibroblasts differentiate into myofibroblasts which foster
   extracellular matrix (ECM) secretion. ECM coupled with paracrine
   factors (TGF-β) and electrical modulators further damages the
   myocyte-fibroblast’s communication, which then leads to irregular
   conduction propagation with more signal blockage.[44]^13 Furthermore,
   increased collagen release/deposition in ECM increases the myocardium’s
   proclivity to cardiac dysfunction and HF.[45]^14^,[46]^15^,[47]^16
   Another study showed that SCN5A decisively contributes to the
   transdifferentiating of human atrial fibroblasts suggesting the
   ineluctable involvement of SCN5A mutation in fibroblast
   activity.[48]^17 Although the reduced expression of SCN5A plays an
   essential role in the development of cardiac fibrosis, there is no
   study on how the defective SCN5A increases the pro-fibrotic TGF-β
   signaling in human cardiac fibroblasts (HCF).

   MicroRNAs (miRNAs) contain 22 nucleotide sequences, crucial as
   profibrotic or anti-fibrotic by binding to 3ˊ untranslated region
   (3ˊUTR) of target mRNA.[49]^18^,[50]^19 The plethora of miRNAs has been
   associated with SCN5A channelopathies and HF, vital regulators in
   myocardial fibrosis by regulating ECM synthesis and cytokines
   secretion.[51]^20 For instance, miR-24 functionally engaged in the
   regulation of SCN5A expression in patients with HF and miR-210-5p
   reduces cardiac fibrosis by interfering with TGF-β type I receptor
   binding in rats.[52]^21^,[53]^22 In addition, miR-452-5p is a putative
   pathological factor, and significantly downregulated in the cardiac
   tissues of patients with hypertrophic cardiomyopathy and upregulated in
   aortic stenosis-induced HF rats.[54]^23^,[55]^24 However, the
   involvement of miR-452-5p in cardiac fibrosis has not been reported
   yet, and no research has been done to explore the regulatory function
   of miR-452-5p in SCN5A-dependent cardiac remodeling.

   This study aimed to investigate how the downregulation of SCN5A gene
   regulates fibrogenesis and clarifies the underlying mechanisms. In the
   current study, we identified miR-452-5p as a key regulator of
   SCN5A-induced cardiac fibrosis via miRNA sequencing. We further
   explored the effect of miR-452-5p on fibroblast differentiation,
   migration, and the potential molecular mechanism of TGF-β signaling
   mediation. Our study provides insight into the SCN5A knockdown-induced
   fibrosis and miR-452-5p as a potential therapeutic target.

Results

SCN5A knockdown promotes fibrosis in human cardiac fibroblasts

   After establishing the SCN5A knockdown cell model, the expression of
   Nav1.5 was measured via western blot. There was less protein expression
   of SCN5A by 50% than control HCF without affecting fibroblast
   morphology ([56]Figures 1A and 1B). To investigate the role of SCN5A
   knockdown in cardiac fibrosis, the expression level of known
   pathological cardiac fibrosis markers was checked in SCN5A knockdown
   and control HCF by western blot. SCN5A knockdown HCF significantly
   increased the protein expressions of pro-Collagen type 1A1, α-SMA, and
   fibronectin ([57]Figure 1C). Fibroblasts' ability to synthesize and
   secrete precursors of fibrillar collagen changes with cardiac
   remodeling.[58]^25 As excessive collagen secretion can lead to
   disproportionate fibrosis, we assessed the difference in collagen
   secretion capacity between SCN5A knockdown HCF and control HCF. The
   soluble collagen type-I level in conditioned media was 4-fold higher in
   SCN5A knockdown HCF compared to control HCF ([59]Figure 1D). These
   results divulge the crucial commitments of SCN5A knockdown in cardiac
   fibrosis.

Figure 1.

   [60]Figure 1
   [61]Open in a new tab

   SCN5A knockdown promotes the expression of fibrogenic signaling

   (A) Upper panel: Knockdown of SCN5A gene in human cardiac fibroblasts
   by targeting SCN5A gene using shRNA lentivirus. Lower panel: fibroblast
   grown after SCN5A gene knockdown and morphology analyzed
   microscopically, Scale bar 350 μm (representative pictures shown).

   (B) The protein expression of Nav1.5 after the knockdown of the SCN5A
   gene in HCF represents a 50% decrease in Nav1.5 protein expression.
   Data are expressed as mean ± SEM, paired t-test, ∗∗p < 0.01, n = 5
   independent experiments.

   (C) Representatives immunoblot and quantitative analysis showing the
   expression of pro-Col1agen 1A1, α-SMA, and fibronectin in SCN5A
   knockdown and control HCF normalized with the internal control group.
   Data are expressed as mean ± SEM, paired t-test, ∗∗p < 0.01,
   ∗∗∗p < 0.001, n = 6 independent experiments.

   (D) There were higher soluble collagen-type 1 levels measured in a
   conditioned medium (serum-free) of SCN5A knockdown HCF than that from
   control cells Data are expressed as mean ± SEM, paired t-test,
   ∗∗∗p < 0.001, n = 6 independent experiments. SCN5A shRNA: SCN5A
   knockdown HCF.

Differential expression of miR-452-5p, functional enrichment and validation
in SCN5A knockdown HCF

   We performed hypothesis-free micro-RNA sequencing to identify the
   possible mechanistic link between the loss of SCN5A and the
   upregulation of fibrotic genes in HCF. The transcriptomic analysis
   revealed that a total of 410 microRNAs were expressed in SCN5A
   knockdown HCF, among which 53 were significantly expressed as shown in
   the heatmap ([62]Figure S1). A Total of 13 miRNAs were differentially
   expressed including three miRNAs markedly upregulated and ten were
   significantly downregulated ([63]Figure 2A). Next, we performed target
   prediction of upregulated and downregulated miRNAs using multiple tools
   and identified most common genes predicted by at least two tools
   ([64]Table S1). These genes were then subjected to GO analysis.
   ([65]Figure 2B). The KEGG pathway enrichment analysis revealed the most
   enriched pathways were TGF-β signaling pathway (path:hsa04350),
   cellular senescence (path:hsa04218), hippo signaling pathway
   (path:hsa04390), diabetic cardiomyopathy (path:hsa05415), and relaxin
   signaling pathway (path:hsa04926) ([66]Figure 2C). Furthermore, the
   differentially expressed miRNAs were verified through stem-loop
   quantitative real-time PCR. Overall, the top three upregulated and the
   top two downregulated miRNAs in SCN5A knockdown HCF were selected for
   validation ([67]Figure 2D). The expression of examined miRNAs showed
   partial agreement with the RNA-seq data. miR-34a-5p and miR-199b-5p
   were significantly upregulated, whereas miR-335-5p and miR-452-5p were
   strikingly lower in SCN5A knockdown HCF, with miR-452-5p being the
   significantly lowest expressed in SCN5A knockdown HCF. Consequently, we
   selected miR-452-5p for further functional analysis, and the miRNA-mRNA
   network was constructed using datasets comprising miRNA-target
   gene-binding data and miRNA-mRNA interactions using Cytoscape 3.9.1
   ([68]Figure 2E). The network analysis unveiled a regulatory abundance
   of miR-452-5p, with numerous target genes displaying enrichment in
   TGF-β signaling pathway and genes related to ECM interaction.
   Specifically, SMAD2, SMAD4, TGF-βRI, and TGF-βRII were notably direct
   mediators of the TGF-β pathway. Further studies were performed to
   determine the role of miR-452-5p in SCN5A knockdown-induced fibrosis.

Figure 2.

   [69]Figure 2
   [70]Open in a new tab

   Maladaptive TGF-β signaling is upregulated in SCN5A knockdown HCF

   (A) Differentially expressed miRs against their enrichment score in
   SCN5A knockdown HCF. Data plotted against enrichment score, p values
   determined via DESeq2 R package.

   (B) GO enrichment of differentially expressed miRs target genes
   (including UP/Downregulated miRs) are presented, top 10 significantly
   enriched GO term (biological process, cellular component, and molecular
   function) branches are presented. The GO terms were plotted against
   fold enrichment values and -log p-values (Fisher’s exact test,
   p < 0.05).

   (C) Top 5 up-regulated (red bars) and top 3 down-regulated (blue bars)
   significantly enriched pathways in KEGG pathway analysis of
   differentially expressed genes plotted against -log FDR values. Their
   corresponding p-values calculated by the Fisher’s exact test are
   mentioned in parentheses in front of the bars.

   (D) qRT-PCR analysis of micro-RNAs (hsa-miR-1307-5p, hsa-miR-34a-5p,
   hsa-miR-199b-5p, hsa-miR-335, has-miR-452-5p) identified by small RNA
   sequencing analysis in SCN5A knockdown and control HCF. Shrek green
   represents upregulated miRNAs, blue bars represent downregulated miRNAs
   in SCN5A knockdown HCF, while black bars represent control cells. Data
   are expressed as mean ± SEM, one-way ANOVA, ∗p < 0.05, ∗∗∗p < 0.001,
   n = 7 independent biological repeats. SCN5A shRNA: SCN5A knockdown HCF.
   miR-NTC: miR negative control.

   (E) Rectangular nodes represent targets mRNA and Diamond nodes
   represent miRNAs. The network displays mRNAs with the highest
   predictive score. Edges with arrows indicate the inhibitory effect on
   targeted mRNA. Light blue nodes show mRNAs involved in the TGF-β
   signaling pathway. Green nodes illustrate genes related to ECM
   interaction. Blue and purple nodes show the genes involved in the FGF
   receptor signaling pathway and positive regulation of cell
   proliferation respectively. Gray nodes are for genes without specific
   interpretation. GO: Gene Ontology, KEGG: Kyoto Encyclopedia of Genes
   and Genome.

miR-452-5p attenuated TGF-β signaling in SCN5A knockdown HCF

   Based on the bioinformatic study, we assessed TGF-β expression in SCN5A
   knockdown and control HCF. Our western blot and qRT-PCR analysis
   revealed a significant increase in both TGF-β protein expression and
   transcripts in SCN5A knockdown HCF compared to the control group
   ([71]Figures 3A and 3B). Furthermore, the ELISA consolidated the 4-fold
   increase in secretory form of TGF-β in conditioned media (24 h) from
   SCN5A knockdown HCF as compared to control ([72]Figure 3C). To validate
   the functional relevance of miR-452-5p in TGF-β signaling, we treated
   the SCN5A knockdown HCF with miR-452-5p-mimic (10 nM) or miR negative
   control (miR-NTC) (10 nM) to increase miR-452-5p expression
   ([73]Figure 3D), and measured the TGF-βRI, TGF-βRII expression levels
   because these are TGF-β receptors and their activation leads to
   phosphorylation of downstream signaling mediators, such as SMAD2 and
   SMAD3 (canonical TGF-β signaling pathway),[74]^26 as predicted in
   miRNA-mRNA network. Interestingly, we found that TGF-β1 transcripts and
   expression levels of TGF-β1, TGF-βRI, and TGF-βRII, were significantly
   suppressed by the miR-452-5p mimic than the miR-NTC group which was
   initially activated in SCN5A knockdown HCF. In line with these results,
   the phospho-SMAD2/3 expression was noticeably reduced in the miR-452-5p
   mimic-treated group whereas the total smad2/3 remained unchanged
   ([75]Figures 3E and 3F). These findings suggest that miR-452-5p
   actively repressed the TGF-β signaling, which contributed to fibrosis.

Figure 3.

   [76]Figure 3
   [77]Open in a new tab

   miR-452-5p mimic regulates canonical TGF-β signaling pathway in SCN5A
   knockdown HCF

   (A) Representative immunoblot showing the expression of TGF-β1
   expression in SCN5A knockdown and control HCF normalized with the
   internal control group. Data are expressed as mean ± SEM, paired
   t-test, ∗∗∗p < 0.001, n = 6 independent experiments.

   (B) The expression of TGF-β1 mRNA measured via qRT-PCR in SCN5A
   knockdown and control HCF normalized with the internal control. Data
   are expressed as mean ± SEM, paired t-test, ∗∗∗p < 0.001, n = 6
   independent experiments.

   (C) ELISA was used to measure the total secreted TGF-β in a conditioned
   medium (serum-free) of SCN5A knockdown HCF than that from control
   cells. Data are expressed as mean ± SEM, paired t-test, ∗∗∗∗p < 0.0001,
   n = 6 independent experiments.

   (D) MiR-452-5p mimic was used to increase the expression of miR-452-5p
   in SCN5A knockdown HCF and miR-452-5p expression was measured through
   qRT-PCR. Data are expressed as mean ± SEM, paired t-test, ∗∗∗p < 0.001,
   n = 6 independent experiments.

   (E) TGF-β1 mRNA measured in SCN5A knockdown HCF treated with miR-452-5p
   and miR-NTC and control HCF via qRT-PCR. Data are expressed as mean ±
   SEM, paired t-test, ∗∗p < 0.01, ∗∗∗∗p < 0.0001, n = 6 independent
   experiments.

   (F) Immunoblots and quantitative analysis of TGF-β1, TGF-βRI, TGF-βRII,
   total SMAD 2/3, and phospho-SMAD 2/3 in SCN5A knockdown HCF as compared
   to control and SCN5A knockdown HCF treated without or with miR-NTC and
   miR-452-5p mimic. n = 6 experiments. Data are expressed as mean ± SEM,
   paired t-test, ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, n = 6 independent
   experiments. SCN5A shRNA: SCN5A knockdown HCF. miR-NTC: miR negative
   control.

miR-452-5p inhibited fibrosis-related genes, differentiation, and migration
of SCN5A knockdown HCF but not restore sodium current (I[Na])

   To assess the anti-fibrotic effect of miR-452-5p on SCN5A knockdown
   HCF, we performed immunoblot analysis of SCN5A knockdown HCF treated
   with miR-452-5p mimic (10 nM) and miR-NTC (10 nM). Our results showed a
   significant decrease in pro-Collagen type 1A1 and fibronectin levels in
   the miR-452-mimic treated SCN5A knockdown HCF compared to the miR-NTC
   group ([78]Figure 4A). In accordance, the concentration of soluble
   collagen type-I in the medium of miR-452-5p mimic treated SCN5A
   knockdown HCF reflected the downregulation at the protein level, unlike
   the miR-NTC group ([79]Figure 4B). Furthermore, our immunofluorescence
   analyses indicated that SCN5A knockdown dramatically promoted the
   formation of differentiation-related stress fibers, whereas the
   miR-452-5p mimic treatment showed a marked decrease in α-SMA fibers as
   compared to the miR-NTC group consistent with the α-SMA protein
   expression in miR-452-5p mimic treated SCN5A knockdown HCF
   ([80]Figures 4C and 4D). Increased myofibroblast differentiation is
   accompanied by fibroblast cell-migration. To corroborate the role of
   miR-452-5p in fibroblast migration, we treated SCN5A knockdown HCF with
   either miR-452-5p mimic or miR-NTC and assessed migration. The wound
   healing assay showed that SCN5A knockdown enhanced fibroblast
   migration, while treatment with the miR-452-5p mimic clearly restrained
   the migration of SCN5A knockdown HCF as compared to the miR-NTC group
   ([81]Figure 4E). We also investigated whether miR-452-5p could reverse
   the decreased sodium current (I[Na]) and late sodium current
   (I[Na-Late]) resulting from knockdown of SCN5A gene in HCF. Our
   findings indicate that miR-452-5p was unable to restore the decreased
   sodium currents ([82]Figure 5), suggesting that the anti-fibrotic
   potential of miR-452-5p is exclusively molecular in nature but does not
   have any effect on I[Na] under the loss of SCN5A activity.

Figure 4.

   [83]Figure 4
   [84]Open in a new tab

   MiR-452-5p mimic restored the fibrogenic phenotype in SCN5A knockdown
   HCF

   (A) Immunoblot and quantitative analysis of the expression in protein
   levels of pro-Collagen type 1A1 and fibronectin in control, and SCN5A
   knockdown HCF, miR-NTC, and miR-452-5p-mimic groups. Data are expressed
   as mean ± SEM, paired t-test, ∗∗p < 0.01, n = 6 independent
   experiments).

   (B) Soluble collagen-type1 level measured in conditioned medium from
   SCN5A knockdown HCF as compared to control and SCN5A knockdown HCF
   treated without or with miR-NTC and miR-452-5p mimic. Data are
   expressed as mean ± SEM, paired t-test, ∗p < 0.05, ∗∗p < 0.01, n = 5
   independent experiments.

   (C) Immunohistochemistry shows increased expression of α-SMA in SCN5A
   knockdown HCF and this expression was significantly reduced upon the
   treatment of miR-452-5p mimic as compared to both control and miR-NTC
   groups. α-SMA (green), DAPI (blue). Scale bar: 90 μm.

   (D) Immunoblots and quantitative analysis of α-SMA in SCN5A knockdown
   HCF as compared to control and SCN5A knockdown HCF, miR-NTC, and
   miR-452-5p-mimic groups. Data are expressed as mean ± SEM, paired
   t-test, ∗∗p < 0.01. ∗∗∗p < 0.001, n = 6 independent experiments.

   (E) Representative images of migration at 0 h and 10 h post-wounding.
   Scale bar: 300 μm. Graph showing the cell migration distances (μm) of
   control, SCN5A knockdown HCF, miR-NTC, and miR-452-5p mimic treated
   groups. Data are expressed as mean ± SEM, paired t-test, ∗∗∗p < 0.001,
   n = 10 independent experiments. SCN5A shRNA: SCN5A knockdown HCF.
   miR-NTC: miR negative control.

Figure 5.

   [85]Figure 5
   [86]Open in a new tab

   Effect of miR-452-5p on the sodium current (I[Na]) and late sodium
   current (I[Na-Late]) in SCN5A knockdown HCF

   (A) Representative current traces and current-voltage (I–V)
   relationships of I[Na] from control HCF (n = 9), SCN5A knockdown HCF
   (n = 10), and SCN5A knockdown HCF transfected with miR-452-5p mimic
   (n = 10). Data are expressed as mean ± SEM, unpaired t-test, ∗p < 0.05
   Control versus SCN5A knockdown HCF, #p < 0.05, ##p < 0.01 Control
   versus miR-452-5p mimic-treated SCN5A knockdown HCF.

   (B) Representative trace and average data of I[Na-Late] from control
   HCF (n = 9), SCN5A knockdown HCF (n = 7), and miR-452-5p mimic
   transfected SCN5A knockdown HCF (n = 11). Data are represented mean ±
   SEM, unpaired t-test, ∗∗p < 0.01,∗p < 0.05.

MiR-452-5p directly targets SMAD4 to mitigate fibrogenesis in SCN5A knockdown
human cardiac fibroblasts

   To determine the mechanism through which miR-452-5p regulates the TGF-β
   signaling pathway, analysis using miRanda, predicted pivotal binding
   sites between miR-452-5p and the SMAD4 mRNA duplex ([87]Figure S2). In
   addition, the prediction of binding sites in the 3ˊUTR sequence
   substantiates that among miR-452-5p target genes, 3ˊUTR of SMAD4
   contains the most and multiple conserved complementary sites for the
   seed region of miR-452-5p, corroborating the recruitment of SMAD4 via
   miR-452-5p ([88]Figure 6A). To verify this hypothesis, we evaluated the
   expression of SMAD4 in our SCN5A knockdown cell model. The western blot
   results disclosed that the SMAD4 expression was significantly higher in
   SCN5A knockdown HCF ([89]Figure 6B). To identify the substantial
   binding and influence of miR-452-5p on target, we performed 3′UTR
   reporter assay ([90]Figure 6C). Overexpression of miR-452-5p
   dramatically amortized the luciferase activity of the SMAD4 3ˊUTR
   construct and diminished SMAD4 protein level ([91]Figures 6D and 6E),
   indicating that miR-452-5p may interfere with the nuclear translocation
   of SMAD4. To confirm this, we performed the nuclear and cytosolic
   fractionation assay from SCN5A knockdown HCF, miR-452-5p mimic, and
   miR-NTC treated HCF. As shown in [92]Figure 6F, SMAD4 was translocated
   into the nucleus of SCN5A knockdown HCF as compared to the control,
   while the nuclear accumulation of SMAD4 was attenuated in the
   miR-452-5p mimic group, when compared with the miR-NTC group.

Figure 6.

   [93]Figure 6
   [94]Open in a new tab

   SMAD4: a direct target of miR-452-5p

   (A) Sequence was predicted from the online database (Target Scan) and
   complementary targets of 3ˊUTR of human SMAD4 gene with hsa-miR-452-5p.

   (B) Quantitative analysis and immunoblot representing the protein
   expression level of SMAD4 in SCN5A knockdown and control HCF. Data are
   expressed as mean ± SEM, paired t-test, ∗∗p < 0.01, n = 5 independent
   experiments.

   (C) Profile of pEZX-MT06 plasmid for dual-luciferase reporter assay to
   detect the interaction of miR-452-5p and 3ˊUTR of SMAD4 in SCN5A
   knockdown vs. control HCF and miR-NTC vs. miR-452-5P mimic in SCN5A
   knockdown HCF.

   (D) The luciferase activity was assessed in SCN5A knockdown and control
   HCF co-transfected without or with miR-NTC and miR-452-5p mimic
   (10 nM). Data are expressed as mean ± SEM, paired t-test, ∗∗p < 0.01,
   n = 5 independent experiments.

   (E and F) Representative immunoblots and quantitative analysis showing
   the expression of SMAD4 in whole cell lysate, nuclear, and cytoplasmic
   fractions from SCN5A knockdown HCF as compared to control and miR-NTC
   vs. miR-452-5p mimic groups revealed that SMAD4 nuclear translocation
   blocked by miR-452-5p mimic. Lamin B and GAPDH indicate nuclear and
   cytoplasmic fractions, respectively. Data are represented as mean ±
   SEM, paired t-test. ∗∗p < 0.01, ∗∗∗p < 0.001, n = 5–6 experiments.
   SCN5A shRNA: SCN5A knockdown HCF. miR-NTC: miR negative control, fLuc:
   firefly luciferase gene, rLuc: Renilla luciferase gene, CMV-promotor:
   cytomegalovirus promotor, Amp^R: ampicillin resistance gene, ori:
   origin of replication.

Treatment with miR-452-5p mimic mitigates fibrosis in isoproterenol-induced
HF

   HF is a common heart disease with enhanced cardiac fibrosis and
   acquired down-regulation of Nav1.5.[95]^21^,[96]^27^,[97]^28^,[98]^29
   To examine whether miR-452-5p could rescue the fibrogenic phenotype
   under HF conditions, we generated HF rat model by subcutaneous
   injection of isoproterenol (100 mg/kg) at the age of 10 weeks. After
   2 weeks rats were given the intravenous injection (tail vein) of
   AAV9-miR-452-5p (3.0 x10^10 genome copies per rat, once a week for
   2 weeks) and saline as negative control once a week. After 2 weeks,
   rats were sacrificed, and the heart and the serum were collected
   ([99]Figure 7A). We found that AAV miR452 markedly recovered the
   fibrotic area in the left ventricular region of the heart as well as
   the heart weight/body weight ratio ([100]Figures 7B and 7C). Improved
   cardiac function including restored mean blood pressure, ejection
   fraction, fraction shortening, left ventricular diameter both in
   systole and diastole, and thickness of intraventricular septum was
   observed in AAV miR452 rats compared to HF rats ([101]Figure 7D). As
   shown in [102]Figure 8A, the expression of Nav1.5 in the ventricular
   tissues of HF rats were significantly reduced. We also identified the
   downregulation of miR-452-5p in the HF rats, and it was increased after
   the delivery of miR-452-5p mimic via AAV9 ([103]Figure 8B). Consistent
   with in vitro findings, over-expression of miR452-5p reversed fibrosis
   in AAV miR452 group ([104]Figure 8C). Impaired TGF-β signaling in HF
   was significantly retrieved after treatment with AAV-miR452, SMAD4
   expression was also attenuated in AAV9 miR-452 group ([105]Figure 8D).
   We conducted liver and kidney function tests at the end of the
   experiments to examine the potential toxicity of the viral vector.
   There were similar ALT, AST, and BUN in different groups, which implies
   the non-toxicity of AAV9 ([106]Figure S3). Collectively, these data
   illustrate the notable significance of in vivo administered miR-452 in
   ameliorating the pathophysiology of the HF animal model.

Figure 7.

   [107]Figure 7
   [108]Open in a new tab

   Effect of AAV miR452 on cardiac function in isoproterenol-induced HF

   (A) Schematic illustration of in vivo induction of HF via subcutaneous
   injection of isoproterenol 100 mg/kg (once a week for 2 weeks) and
   treatment with AAV miR452 (3.0 x10^10 genome copies per rat via tail
   vein, once a week for 2 weeks), control rats received normal saline for
   2 weeks.

   (B) Cardiac pictures from control, HF, and AAV miR452 groups.

   (C) The ratio of heart to body weight (mg/g) in control, HF, and AAV
   miR452 groups. Data are presented as mean ± SEM, one-way ANOVA followed
   by Tukey’s multiple comparison test, ∗p < 0.05, ∗∗p < 0.01, n = 5-6
   independent experiments.

   (D) Upper Panel: Representative M-mode echocardiographic images. Lower
   Panel: Averaged data presenting the results of mean blood pressure (BP,
   mmHg), percentage of ejection fraction (EF%), left ventricle fractional
   shortening (FS%), left ventricular internal diameter in systole (LVIDs,
   mm), left ventricular internal diameter in diastole (LVIDd, mm), and
   intraventricular septal diameter (IVSd, mm) in control, HF, and AAV
   miR452 groups. Data are represented as mean ± SEM, one-way ANOVA
   followed by Tukey’s multiple comparison test, ∗p < 0.05, ∗∗p < 0.01,
   ∗∗∗p < 0.001, n = 5–6 independent experiments.

Figure 8.

   [109]Figure 8
   [110]Open in a new tab

   Systemic delivery of AAV miR452 ameliorates fibrosis in
   isoproterenol-induced HF rats

   (A) Nav1.5 protein expression in left ventricular tissues of HF and
   control rats. Data are expressed as mean ± SEM, paired t-test,
   ∗∗p < 0.01, n = 3 independent experiments.

   (B) MiR-452-5p mimic delivery through AAV9 increased the miR-452
   expression in left ventricular tissues which was initially reduced
   after HF development as compared to control, measured via qRT-PCR. Data
   are expressed as mean ± SEM, one-way ANOVA followed by Tukey’s multiple
   comparison test, ∗∗∗p < 0.001, n = 5–6 independent experiments.

   (C) Immunoblots and quantitative analysis of pro-collagen type 1a1,
   fibronectin, α-SMA in LV tissues of HF, and AAV miR452 compared with
   control. Data are expressed as mean ± SEM, one-way ANOVA followed by
   Tukey’s multiple comparison test, ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001,
   n = 5–6 independent experiments.

   (D) Immunoblots and quantitative analysis of TGF-β signaling including
   TGF-β1, TGF-βRI, RII, p-SMAD2/3, SMAD4 in LV tissues of HF (induced by
   subcutaneous injection of isoproterenol 100 mg/kg once a week for
   2 weeks) and HF rats treated with AAV miR452 (3.0 x10^10 genome copies
   per rat via tail vein, once a week for 2 weeks) compared with control
   (received normal saline). Data are represented as mean ± SEM, one-way
   ANOVA followed by Tukey’s multiple comparison test, ∗p < 0.05,
   ∗∗p < 0.01, ∗∗∗p < 0.001, n = 5–6 independent experiments.

Discussion

   Cardiac fibrosis provokes pathological changes thereby promoting
   arrhythmia and HF.[111]^30 Despite the quintessential advancement in
   the acknowledgment of cardiac fibrosis and SCN5A channelopathies, the