miR‐486 improves fibrotic activity in myocardial infarction by targeting SRSF3/p21‐Mediated cardiac myofibroblast senescence

Abstract The regulation of fibrotic activities is key to improving pathological remodelling post‐myocardial infarction (MI). Currently, in the clinic, safe and curative therapies for cardiac fibrosis and improvement of the pathological fibrotic environment, scar formation and pathological remodelling post‐MI are lacking. Previous studies have shown that miR‐486 is involved in the regulation of fibrosis. However, it is still unclear how miR‐486 functions in post‐MI regeneration. Here, we first demonstrated that miR‐486 targeting SRSF3/p21 mediates the senescence of cardiac myofibroblasts to improve their fibrotic activity, which benefits the regeneration of MI by limiting scar size and post‐MI remodelling. miR‐486‐targeted silencing has high potential as a novel target to improve fibrotic activity, cardiac fibrosis and pathological remodelling.


| INTRODUC TI ON
Myocardial infarction (MI) is the leading cause of death worldwide. 1 Improvement of profibrotic activity and pathological remodelling in ischaemic myocardium mediated by activated cardiac myofibroblasts (CMFs) post-MI are still major challenges. Cardiac fibroblasts (CFs) play an important role in cardiac repair, scar formation, fibrosis and pathological reconstruction post-MI. 2,3 In MI, loss of structural integrity of the myocardium exposes CFs to mechanical stress, accompanied by particular inflammatory effectors, cytokines, hormones and growth factors, inducing CF proliferation, migration to the infarct zone and transdifferentiation into CMFs. In the adult mammalian heart, after MI, due to a very low potency to regenerate cardiomyocyte death, the dead cardiomyocytes are replaced by fibrotic tissue and scar, which is secreted by CMFs. CMFs are responsible for scar tissue formation at every site of cardiomyocyte necrosis and fibrosis in the ischaemic myocardium. 4,5 Accompanied by cardiomyocyte loss and scar formation post-MI, the following response is remodelling of the ischaemic myocardium and eventually incurs hypertrophy and fibrosis of the left ventricular wall. This remodelling process is progressive in the post-MI myocardium and eventually leads to heart failure. 6 However, the mechanism of maintaining physiopathology for fibrotic and antifibrotic environments for cardiac fibrosis and remodelling in ischaemic myocardium post-MI is still unclear.
A recent study demonstrated that p53-mediated senescence of cardiac fibroblasts is crucial to limit cardiac fibrosis via downregulation of cardiac collagen production after myocardial infarction. 7 CMF senescence is an essential requirement for antifibrotic mechanisms, and CCN1, a potent inducer of senescence, was able to reduce perivascular fibrosis by approximately 50%. 8 More recently, it was revealed that CCN1-induced cellular senescence of cardiac fibroblasts was able to promote heart regeneration. 9 The up-to-date progress in the field proposes that cellular senescence of cardiac fibroblasts acts as an essential mechanism for antifibrosis and heart regeneration. However, the critical effectors regulating this essential mechanism to improve the fibrotic environment and cardiac regeneration, which are targeted for CFs and CMFs, are still unclear. Currently, in the clinic, safe and curative therapies for cardiac fibrosis and im-  15,16 In addition, miR-486-5p was shown to be involved in the cellular senescence of human diploid fibroblasts 17 and IgE elevation-mediated pathologic cardiac fibrosis. 18 These findings suggest that miR-486 might be an important endogenous modulator of the antifibrotic environment in fibroblasts and myofibroblasts. However, it is still unclear how miR-486 functions in post-MI regeneration. Accordingly, we hypothesized that miR-486 might act as an important player in suppressing fibrotic activity in MI, which benefits the improvement of cardiac fibrosis, pathological remodelling and regeneration of MI. Indeed, in the present study, we first demonstrated that miR-486 targeting SRSF3/p21 mediates the senescence of cardiac myofibroblasts to improve their fibrotic activity, which benefits the improvement of post-MI cardiac fibrosis, pathological remodelling and regeneration.

| Animals
In present study, two-month-old female Sprague-Dawley (SD) rats (200-250 g) were used. The rats were adapted for 1 week feeding before experimentation. The animals were provided with food and water ad libitum. Animal care, surgery and handling procedures were approved by the Jinan University Animal Care Committee (Approval No. IACUC-20190104-010).

| Preparation and culture of CMFs
CFs were isolated from the ventricles of 2-month-old female SD rats as published reported. 19 For details, see Appendix.
For details, see Appendix.

| β -galactosidase staining
The prepared cells or frozen sections were used for senescence β-gal staining using a β-galactosidase staining kit (C0602; Beyotime). For details, see Appendix.

| β -galactosidase fluorescence assay
After prepared CMFs were treated with miR-486 for 48 hours, cellular senescence was analysed using the ImaGene Green™ C 12 FDG lacZ Gene Expression Kit (I-2904; Invitrogen). For details, see Appendix.

| Cell proliferation assay
A CCK-8 assay (CK04; Dojindo) was performed to analyse the proliferation of CMFs after miR-486 treatment. For details, see Appendix.

| Apoptosis assay
The Annexin Alexa Fluor 488/PI kit (FXP022-100, 4A Biotech) and subsequent flow cytometry analysis were applied to investigate the apoptosis of miR-486 treated CMFs. For details, see Appendix.

| Cell cycle assay
PI/RNase Staining Buffer (550,825; BD) was applied to analyse the cell cycle of CMFs after miR-486 treatment for 72 hours. For details, see Appendix.

| Western blot analysis
The radioimmunoprecipitation assay (RIPA) buffer (cat. no. P0013C; Beyotime) containing a protease inhibitor cocktail (cat. no. W2200s; Cwbio) was applied to prepare the CMFs lysates. Please note that the Western blot band of GAPDH in Figure

| Prediction of miR-486 targeted genes
TargetScan (http://www.targe tscan.org) was used to predict the potential target genes of rno-miR-486. FirePlex Discovery Engine (https://www.firef lybio.com/ portal/search) was used to verify whether or not a gene is already reported as target gene for miR-486.

| Induction of MI and intramyocardial injection
Two-month-old female SD rats were applied to establish MI. For details, see Appendix.

| Echocardiography
Transthoracic echocardiograms were performed to analyse the cardiac function in the experimental rats. The ejection fraction (EF) was calculated using the area-length method. 20 For details, see Appendix.

| Histological analysis
The MI extent was analysed at the level of the mid-papillary heart muscles and scored following Masson's trichrome staining. The infarct size, with linear approximations to account for area gaps in histology, was expressed as a percentage of the total LV myocardial area as our and others previously described. 20,21 The area of cardiomyocytes was evaluated by Wheat Germ Agglutinin (WGA) staining.
For details, see Appendix.

| Immunohistochemistry staining
The immunostaining for vWF, a marker of endothelial cells, was applied to measure small blood vessel density in the infarct zone and border zone. For details, see Appendix.

| Treadmill test
A treadmill (SA101C, Sansbio) was used to measure the endurance and physical fitness. For details, see Appendix.

| Statistical analysis
An independent samples t-test was performed using GraphPad prism software to determine the P-values in repeated experiments.
All values are expressed as the mean ± standard deviation (S.Dev).
p < 0.05 was set as statistically significant differences.

| miR-486 promotes cellular senescence and apoptosis and inhibits proliferation in CMFs in vitro
To investigate the functional role of miR-486 in cardiac myofibroblasts (CMFs), CMFs were generated by TGFβ-induced transdifferentiation from cardiac fibroblasts (Vimentin + ; CFs), which were isolated from rat ventricles. The transdifferentiated CMFs were confirmed by the expression of the α-SMA and Col1a1 genes and immunofluorescence of Vimentin-EGFP and α-SMA-Cy3 ( Figure 1A, B; p < 0.05). The results showed that miR-486 was expressed in both CFs and CMFs ( Figure 1C). Furthermore, overexpression of miR-486 in CMFs was able to promote cellular senescence, which was confirmed by a significant increase in the density of β-gal-positive cells in the miR-486-treated group compared to the mimic NC-treated group via β-gal histological staining and cytometry quantitative assays ( Figure 1D, E; p < 0.05). It was revealed that miR-486 overexpression mediated CMF senescence in a dose-dependent manner, and 100 nmol/L was a suitable dose ( Figure S1A). Therefore, 100 nmol/L miR-486 was applied in the in vitro study. In addition, the CCK-8 assay indicated that the proliferation and survival rates of the miR-486-treated CMFs were significantly lower than those of the mimic NC-treated CMFs ( Figure 1F; p < 0.05). Flow cytometry analysis of apoptosis showed that the number of apoptotic cells in the miR-486-treated CMFs was significantly higher than that in the mimic NC-treated CMFs ( Figure 1G; p < 0.05) in a dose-dependent manner ( Figure S1B). Furthermore, flow cytometry analysis of the cell cycle showed that overexpression of miR-486 in CMFs was able to promote cell cycle inhibition at the S and G2/M phases ( Figure 1H; p < 0.05).
Double staining analysis of cellular senescence and apoptosis in miR-486-treated CMFs was further performed to determine whether miR-486 overexpression in CMFs mediates both senescence and apoptosis simultaneously. It was found that miR-486mediated cellular senescence and apoptosis coexisting in cells

| miR-486 activates the p21 cellular senescence pathway and inhibits the expression of fibrotic genes in CMFs in vitro
The pathway and functional effects related to miR-486-mediated cellular senescence in CMFs were investigated. The qPCR results revealed that miR-486 overexpression in CMFs increased the expression of the well-recognized cellular senescence-regulating genes p21, p53 and p16 at the mRNA level (

| miR-486-targeted silencing of SRSF3 expression induces p21 activation-mediated cellular senescence in CMFs
The mechanism by which miR-486 promotes cell senescence via activation of p21 was further investigated. A total of 115 potential target genes were predicted by the TargetScan database. The top five predicted potential target genes of miR-486 by TargetScan are listed in Figure 3A. Among them, SRSF3, an effector of mRNA alternative splicing, [27][28][29] was selected for further investigation because it ranked first and was reported to be involved in cellular senescence regulation, 27 p < 0.05). In addition, the results of the dual-luciferase assay showed that SRSF3 can be regulated by miR-486 through its 3'UTR and confirmed that SRSF3 is a novel target gene of miR-486 ( Figure 3D) that can be silenced by miR-486 ( Figure 3B, C). Furthermore, the effect of SRSF3 on cellular senescence was investigated and showed that SRSF3 silencing was able to significantly increase the density of β-gal-positive cells ( Figure 3E; p < 0.05).
The possible effects of the SRSF3 gene on p21 expression and fibrosis effector genes in CMFs were further investigated. We first confirmed that transfection of siSRSF3 in CMFs significantly downregulated the expression of SRSF3 at both the mRNA and protein levels ( Figure 4A, B). Therefore, the effect of SRSF3 silencing was next studied. We first conducted RIP-qPCR assays to evaluate the binding capacity between SRSF3 and p21. The results of anti-SRSF3 antibody-RIP-qPCR revealed that SRSF3 was able to bind with p21 in CMFs ( Figure 4C). Furthermore, p21 expression in siSRSF3-treated CMFs at the mRNA and protein levels was significantly upregulated compared to that in siNC-treated CMFs, but p16 and p53 were not ( Figure 4D Taken together, the results verify that in CMFs, miR-486 targets and silences SRSF3. SRSF3 inhibits p21 gene expression through a direct binding mechanism. Therefore, miR-486-targeted silencing of SRSF3 leads to p21 activation of cellular senescence, resulting in downregulation of the expression of fibrosis effector genes in CMFs.

| miR-486 expression is downregulated in the infarct zone, and its overexpression improves fibrotic pathology, fibrosis and pathological remodelling in post-MI hearts, which benefits the regeneration of MI in vivo
Having identified the function and signalling pathways of miR- In parallel, in comparison with the control group, the AAV9-miR-486-EGFP-treated group had a significantly thicker left ventricular anterior wall (LVAW) in the end-systolic phase ( Figure 5D-D1,2) and end-diastolic phase ( Figure 5D The echocardiography analysis (D1) showed that the left ventricular anterior wall in the end-systolic phase (LVAWs, D2) and the left ventricular anterior wall in the end-diastolic phase (LVAWd, D3) of the AAV9-miR-486-EGFP-treated group were significantly thicker than those of the AAV9-NC-treated group. The left ventricular end-systolic diameter (LVESD, D4) of the AAV9-miR-486-treated group was significantly smaller than that of the AAV9-NC-treated group, but the difference in the left ventricular end-diastolic diameter (LVEDD, D5) between the AAV9-miR-486-EGFP-treated group and the AAV9-NC-treated group was not statistically significant. The ejection fraction (EF, D6) and fractional shortening (FS, D7) of the AAV9-miR-486-EGFP-treated group were significantly higher than those of the AAV9-NC-treated group. n = 10. (E) The analysis of Masson's trichrome staining (E1) showed that the collagen area of the infarct zone (CAIZ) in the AAV9-miR-486-EGFPtreated group was significantly smaller than that of the AAV9-NC-treated group (E2). The thickness of the infarcted myocardium of the left ventricle (TIM, E3) and the wall thickness of the border zone of the left ventricle (WTBZ, E4) of the AAV9-miR-486-EGFP-treated group were significantly larger than those of the AAV9-NC-treated group. The infarct size of the AAV9-miR-486-EGFP-treated group was significantly smaller than that of the AAV9-NC-treated group (E5). n = 10. (F) The treadmill test revealed that the running endurance time of the AAV9-miR-486-EGFP-treated group was significantly longer than that of the AAV9-NC-treated group. n = 8 and 10 for the AAV9-NC and AAV9-miR-486 groups. (G) WGA staining of the cardiomyocyte area showed that the cardiomyocyte area of the AAV9-miR-486-EGFP-treated group was significantly smaller than that of the AAV9-NC-treated group in both the infarct zone and border zone.

| DISCUSS ION
In the present study, our in vitro studies demonstrated that overexpression of miR-486 in CMFs induced cellular senescence and apoptosis and decreased proliferation and cell cycle inhibition at the S and G2/M phases. In addition, miR-486 overexpression was able to induce more senescent CMFs than apoptotic CMFs. Most miR-486-treated senescent CMFs did not undergo apoptosis simultaneously, which is consistent with the well-established knowledge that senescent cells are usually resistant to apoptosis. 30 (C) Schematic diagram of the molecular mechanism by which miR-486/SRSF3/p21 targets senescence in CMFs to improve MI regeneration. In the infarct zone of MI, the expression of miR-486 was downregulated, but the expression of SRSF3 was upregulated. miR-486 targets silencing of SRSF3, and upregulation of p21 mediates the senescence of cardiac myofibroblasts to improve their fibrotic activity and limit scar size and post-MI remodelling, which benefits the regeneration of MI PTEN/PI3K/AKT signalling pathway. 29 miR-486-5p targeting PTEN to activate PI3K/AKT signalling protected against cardiomyocyte apoptosis mediated by coronary microembolization. 32 In addition, studies of the in vitro rat embryonic ventricular cardiomyocyte cell line H9c2, revealed that miR-486 targeting the silencing of NDRG2 to inactivate JNK/C-Jun and NF-κB signalling improves hypoxiainduced damage to cardiomyocytes. 33 More recently, miR-486-5p found in stem cell-secreted microvesicles was reported to increase angiogenesis and regeneration of MI via fibroblast MMP19-VEGFA cleavage signalling in mice and nonhuman primates. 34 In this study, we also verified that overexpression of miR-486 in MI was able to promote cardiac angiogenesis in both the infarct zone and border zone. Thus, findings from our group and others suggest that in addition to targeting SRSF3/p21 to mediate the senescence of CMFs, miR-486 also has the potential to target cardiomyocytes and endothelial Recent progress has demonstrated that cellular senescence is an important physiopathological process, and it has been documented to play different roles in various diseases, such as fibrotic pulmonary disease and liver fibrosis. 35,36 One of the important advancements in the study of cellular senescence is that triggering cellular senescence plays a critical role in promoting regeneration and limiting fibrosis in tissues and organs. 8,9,[37][38][39][40] It has been demonstrated that in cutaneous wound healing, CCN1 is dynamically expressed at sites of inflammation and wound healing. 41 CCN1 is able to induce cellular senescence of fibroblasts by targeting integrin α6β1 and heparan sulphate proteoglycans, which induce activation of the p16INK4a/ pRb senescence pathway. 38 In addition, Trp53-mediated cardiac fibroblast senescence limits cardiac fibrosis in adult MI hearts. 7 Apical resection-induced cardiomyocyte CCN1 secretion results in fibroblast senescence, which enhances the proliferation of cardiomyocytes and decreases cardiac fibrosis to improve neonatal heart regeneration. 9 Furthermore, the cellular senescence of CMFs was identified as an essential antifibrotic mechanism in the adult myocardium. 8 In contrast to current knowledge, the present study proposes The results of the present study also showed that the expression of p53 and p16 was upregulated in miR-486-treated CMFs but not in siSRSF3-treated CMFs. This result indicated that the therapeutic effects of miR-486 treatment in MI, such as promoting cardiac angiogenesis and increasing the expression of p53 and p16, cannot be attributed to SRSF3 as the target gene. In fact, the existence of multiple target genes is a well-identified mechanism for miRNA-mediated gene silencing. 42 Indeed, miR-486 was reported to connect myostatin and the IGF-1/Akt/mTOR pathway to regulate the size of skeletal muscle 43 and target TGF-β2-Smad2/3 signallingmediated suppression of epithelial cell proliferation. 44 In addition, miR-486-5p was found to target IGF1/PI3K/AKT signalling to limit the proliferation and collagen production of hypertrophic scar fibroblasts. 16 The mechanism and pathway underlying miR-486-mediated cardiac angiogenesis and increased expression of p53 and p16 in the present study need to be studied in-depth in future investigations.
Serine/arginine-rich proteins (SR proteins) belong to the RNA binding protein family, which plays an important role in pre-mRNA constitutive and alternative splicing. Among twelve SRSFs (named SRSF1-SRSF12) that have been identified in humans, SRSF3 is the smallest. 45 The SRSF3 protein (64 amino acids and approximately 19 kDa) consists of two domains, namely an RNA recognition domain (RRM; N-terminal domain) and an arginine/serine domain (RS; C-terminal domain). 45 The RS domain acts to target SRSF3 to nuclear speckles. 46 SRSR3 identifies the CUC(U/G)UCY splicing enhancer sequence to conduct RNA alternative splicing. 47 SRSF3 was reported to be involved in alternative splicing of p53 mRNA towards the p53β isoform to induce cellular senescence, 27 heart development and MI. 48 Recently, it was identified that decreased SRSF3 expression results in preferential usage of the proximal poly(A) site of mRNA (shortening the 3'UTR) in human and mouse cells to produce more protein, which induces cellular senescence by increasing the production of senescence-associated genes (PTEN, PIAS1 and DNMT3A). 28 Here, we further revealed that SRSF3 is a novel target gene of miR-486. miR-486 is able to mediate the targeted silencing of SRSF3 accompanied by upregulation of p21 in CMFs in vitro and in the infarct zone and border zone of MI. Furthermore, RIP-qPCR results showed that the SRSF3 protein is able to bind with p21. In view of the well-established functions of SRSF3 in alternative splicing and knockdown of SRSF3 expression resulting in preferential usage of shortened 3'UTRs of mRNAs that produce more protein, our findings suggested that SRSF3 might utilize a similar mechanism to trigger cellular senescence via p21 upregulation when miR-486-targeted silencing of SRSF3 occurs. The exact molecular mechanism by which SRSF3 induces p21 upregulation-mediated cellular senescence needs to be investigated in-depth in future studies. In addition, it is well established that in the p53/p21-mediated cellular senescence pathway, p53 is an upstream effector for the upregulation of p21. 49,50 However, even though our in vitro study showed that miR-486 overexpression in CMFs was able to upregulate p21, p53 and p16 expression, the results of the siSRSF3 in vitro treatment for CMFs and in vivo study did not show the effect of p53 and p16 upregulation upon overexpression of miR-486 in the myocardium.
This suggests that miR-486-targeted silencing of SRSF3 and p21 upregulation mediates the senescence of CMFs, which might be due to direct binding of p21 but not to activation of p53 as its upstream signalling molecule.
In summary, a novel molecular mechanism was uncovered in the current study, in which miR-486 improved fibrotic activity, pathological remodelling and scar size via an SRSF3 gene silencing-mediated increase in p21 signalling to induce cellular senescence of CMFs, resulting in beneficial effects for the improvement of MI regeneration. Therefore, the pathway by which miR-486 targets SRSF3/ p21 to mediate the senescence of CMFs is a potential therapeutic target to improve the fibrotic activity and pathological fibrosis and remodelling seen in ischaemic myocardium, such as MI, to promote increased regeneration.

CO N FLI C T O F I NTE R E S T
The authors declare no conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.