MicroRNA‐495‐3p diminishes doxorubicin‐induced cardiotoxicity through activating AKT

Abstract Doxorubicin (Dox) is a broad‐spectrum antitumour agent; however, its clinical application is impeded due to the cumulative cardiotoxicity. The present study aims to investigate the role and underlying mechanisms of microRNA‐495‐3p (miR‐495‐3p) in Dox‐induced cardiotoxicity. Herein, we found that cardiac miR‐495‐3p expression was significantly decreased in Dox‐treated hearts, and that the miR‐495‐3p agomir could prevent oxidative stress, cell apoptosis, cardiac mass loss, fibrosis and cardiac dysfunction upon Dox stimulation. In contrast, the miR‐495‐3p antagomir dramatically aggravated Dox‐induced cardiotoxicity in mice. Besides, we found that the miR‐495‐3p agomir attenuated, while the miR‐495‐3p antagomir exacerbated Dox‐induced oxidative stress and cellular injury in vitro. Mechanistically, we demonstrated that miR‐495‐3p directly bound to the 3′‐untranslational region of phosphate and tension homology deleted on chromosome ten (PTEN), downregulated PTEN expression and subsequently activated protein kinase B (PKB/AKT) pathway, and that PTEN overexpression or AKT inhibition completely abolished the cardioprotective effects of the miR‐495‐3p agomir. Our study for the first time identify miR‐495‐3p as an endogenous protectant against Dox‐induced cardiotoxicity through activating AKT pathway in vivo and in vitro.

Protein kinase B (PKB/AKT) plays critical roles in cell survival and has been proposed as a promising therapeutic target of various cardiovascular diseases, including Dox-induced cardiotoxicity. [10][11][12] Findings from Zhang et al. showed that AKT activation significantly alleviated cardiomyocyte apoptosis and cardiac dysfunction in Doxtreated mice. In addition, AKT activation can suppress the nuclear export and degradation of nuclear factor-E2-related factor 2 (NRF2), thereby protecting against Dox-induced oxidative stress in the heart. 13 Phosphate and tension homology deleted on chromosome ten (PTEN) is a major negative regulator of AKT phosphorylation and activation, and PTEN upregulation reduces AKT activation and amplifies oxidative damage in Dox-treated hearts. 14,15 Consistently, Johnson et al. 16 demonstrated that PTEN inhibitor significantly reduced apoptosis, cardiac remodelling and dysfunction in Doxtreated mice. In contrast, overexpression of PTEN could exacerbate Dox-induced cardiomyocyte apoptosis and oxidative stress through blocking AKT pathway. 17 These findings identify AKT as a promising therapeutic candidate to treat Dox-induced cardiotoxicity.
MicroRNAs (miRs) function as a class of endogenous negative gene regulators through binding to the 3′-untranslational region (UTR) of target mRNAs. They are implicated in various biological processes, such as cell survival, death, senescence and canceration. [18][19][20] Recent findings have indicated that miRNAs are also involved in regulating oxidative stress and Dox-induced cardiotoxicity. Han et al. 21 found that miR-330-5p contributed to Dox-induced oxidative stress, DNA damage, cardiomyocyte injury and cardiac dysfunction. In addition, we also validated that miR-22 directly bound to the 3′-UTR of silent information regulator 1 (SIRT1), and subsequently aggravated Dox-induced oxidative stress, cardiomyocyte apoptosis and cardiac dysfunction, which were significantly attenuated by the miR-22 antagomir. 9 miR-495-3p is well-studied in human tumours, with high potency to inhibit tumour growth and chemoresistance. [22][23][24] In addition, miR-495-3p is essential for regulating oxidative stress and cell survival. Lin et al. 25 revealed that miR-495-3p upregulation promoted cell proliferation and inhibited apoptosis of tumour necrosis factor-alpha-induced human nucleus pulposus cells. Consistently, miR-495-3p elevation significantly suppressed oxidative stress, endothelial dysfunction and fibrosis in atherosclerotic mouse aortas. 26 The present study aims to investigate the role and molecular mechanisms of miR-495-3p in Dox-induced cardiotoxicity.

| Experimental model of Dox-induced cardiotoxicity
All experimental procedures were approved by our hospital and also in accordance with the Guidelines for the Care and Use of Laboratory Animals (NIH Publication, revised 1996). To generate Dox-induced cardiotoxicity, male C57BL/6 mice (8-10-week-old) were randomly assigned to reduplicative intraperitoneal injections of Dox (4 mg/kg) weekly for 4 consecutive weeks according to previous studies by us and the others, whereas the control mice were treated with an equal volume of saline. 9,27 To overexpress or inhibit miR-495-3p, mice were intravenously treated with the agomir, antagomir or respective controls of miR-495-3p once 2 days for 10 times from the second day of Dox injection. Briefly, mice were fixed with the tails disinfected using 75% ethanol solution. Then, the agomir, antagomir or respective controls of miR-495-3p were injected from the tail vein, and special attention was paid to avoid liquid leakage. To inhibit AKT in vivo, mice were intraperitoneally injected with AKTi (20 mg/kg/day) for 14 days before sacrificed according to a previous study. 13 The structure of AKTi was disclosed in the official website of Sigma https:// www.sigma aldri ch.cn/CN/zh/produ ct/sigma/ a6730. To specifically overexpress PTEN in the myocardium, mice received a single injection of NC or PTEN carried by AAV9 from the tail vein at a concentration of 1 × 10 11 viral genome per mouse at 4th week before Dox treatment. 28 To observe the survival rate, mice were maintained for 6 weeks after the last Dox injection.

| Determination of cardiac function
Cardiac function was performed by echocardiography and cardiac catheter according to previous studies. 9,29,30 In brief, mice were anaesthetized with 2% isoflurane and then subjected to the echocardiographic assessment using a Vevo ® 2100 Imaging System (Visual Sonics) equipped with a 40 MHz MicroScan transducer (model MS-550D). The functional parameters were calculated from at least five consecutive cardiac cycles. In addition, pressurevolume loops were captured by a SPR-839 microtip cardiac catheter (Millar Instruments), and analysed with the LabChart 7 software (ADInstruments) to evaluate haemodynamic parameters.

| Western blot
Total proteins were extracted from the left ventricles of murine hearts or cultured cells using the RIPA lysis buffer as previously described. 31,32 Next, 20 μg total proteins were separated by 10% SDS-PAGE, transferred onto PVDF membranes, blocked with 5% skim milk at room temperature and incubated with indicating primary antibodies at 4°C overnight. On the second day, the membranes were probed with horse radish peroxidase (HRP)-conjugated secondary antibodies and visualized with an electrochemiluminescence reagent. The bands were analysed using an Image Lab software and normalized to GAPDH.

| Quantitative real-time PCR
Total RNA was extracted using Trizol reagent, and reverse transcription was performed with the Prime Script RT Master Mix (Takara) and random or oligo (dT) primer. Then, quantitative real-time PCR was conducted using a SYBR ® Premix Ex Taq TM kit (Takara) with GAPDH and U6 used as internal controls for mRNA and miRNA respectively. 31,33,34 The primer sequences were listed as follows:

| Collagen content measurements
Total collagen content in the left ventricles was determined by measuring the level of hydroxyproline, a major component of collagen, as previously described. 35 Briefly, left ventricles were homogenized and hydrolysed in NaOH at 120°C for 1 h, and then neutralized by concentrated HCl. The lysates were then centrifuged at 10,000 g for 5 min to remove the insoluble debris, and then the supernatants were incubated with chloramine T and oxidation buffer according to the manufacturer's instructions at room temperature for 20 min.
Then, the mixture was reacted with Developer at 37°C for 5 min, and DMAB concentrate solution 65°C for 45 min, and the absorbance was measured at 560 nm.

| Analysis of serum cardiac injury biomarkers
Serum samples were collected from mice at 3 days after the last Dox injection and then were used to detect the levels of cTnT and LDH to determine cardiac injury according to the manufacturer's instructions.

| Oxidative stress detection
Reactive oxygen species generation was detected using a DCFH-DA method as previously described. 36

| Cell apoptosis quantification
TdT-mediated dUTP Nick End Labelling (TUNEL)+ cells were detected by a commercial kit as we previously described, and the percent of TUNEL+nuclei was calculated as the apoptotic index. 9 In addition, caspase-3 activity was determined to evaluate cell apoptosis according to our previous study. 9 Briefly, fresh cell lysates were prepared and centrifuged at 10,000 g for 1 min to obtain the cell-free supernatants, which were then incubated with DVED-pNA (200 μmol/L) at 37°C for 2 h and detected at 400 nm.

| Cell culture and treatments
H9C2 cells were purchased from ATCC and cultured in DMEM containing 10% FBS as we previously described. 9 To overexpress or inhibit miR-495-3p, cells were transfected with the agomir, antagomir and respective controls of miR-495-3p at a concentration of 50 nmol/L using Lipofectamine™ 3000 Transfection Reagent for 24 h, and then stimulated with or without Dox (1 μmol/L) for an additional 24 h. 38 To inhibit AKT, cells were treated with AKTi (1 μmol/L) for 30 min. 13 To overexpress PTEN, cells were infected with adenovirus-carried PTEN at a multiple of infection of 30.

| Cell viability and injury
Cell viability was determined by the CCK-8 method as previously described. 39,40 In brief, stimulated cells were incubated with the CCK-8 solution at 37°C for 30 min, and then the absorbance was detected at 450 nm with a reference wavelength of 650 nm. LDH releases to the medium were detected using a commercial kit according to the manufacturer's instructions to evaluate cell injury.

| Luciferase reporter assay
Wild type (WT) or mutant (MUT) PTEN 3′-UTR were amplified from the genomic DNA and then cloned into the psi-CHECK2 luciferase reporter plasmid (Promega), which were then co-transfected with the miR-495-3p agomir or control to HEK293T cells. 48 h after transfection, cells were lysed, and the luciferase activity was determined by a dual luciferase reporter system (Promega). 9,41,42

| Statistical analysis
All results were expressed as the mean ± standard deviation and analysed by SPSS 23.0 software. Two-tailed Student's t-test was conducted to compare differences between two groups, while comparisons among three or more groups were performed by one-way ANOVA, followed by Tukey post hoc test. Survival rate was evaluated by the Kaplan-Meier method and survival curves were compared using the Mantel-Cox log-rank test. p < 0.05 was considered statistically significant.
As shown in Figure 1B, injections with the miR-495-3p agomir at a dose of 10 nmol/g/day completely restored cardiac miR-495-3p expression upon Dox stimulation; therefore, we used this dose for further study. As expected, Dox injection caused cardiac dysfunction in mice, as evidenced by the decreased ejection fraction (EF), stroke volume (SV), stroke work (SW) and ±dP/dt, which were significantly alleviated by the miR-495-3p agomir ( Figure 1C-E). In addition, treatment with the miR-495-3p agomir also prevented the loss of cardiac mass in Dox-treated mice, as evidenced by the increased heart weight (HW)/tibial length (TL) ( Figure 1F). Fibrosis is a key feature of Dox-induced cardiotoxicity and increases cardiac stiffness. 43 Intriguingly, the miR-495-3p agomir significantly suppressed cardiac fibrosis upon Dox stimulation ( Figure 1G,H). Yet, no alteration of heart rate or blood pressure was found in the miR-495-3p agomirtreated mice with or without Dox injection ( Figure 1I,J). Meanwhile, the elevated levels of serum cTnT and LDH were also reduced by the miR-495-3p agomir in Dox-treated mice ( Figure 1K). Moreover, Dox-caused mortality in mice was largely prevented with the miR-495-3p agomir (55% in Dox+miR-495-3p agomir control vs. 85% in Dox+miR-495-3p agomir). Taken together, our data reveal that the miR-495-3p agomir alleviates Dox-induced cardiotoxicity in mice.
More importantly, we observed that the miR-495-3p antagomirtreated mice all died within 5 weeks after the last Dox injection, indicating a higher mortality rate caused by the miR-495-3p antagomir.

| miR-495-3p agomir inhibits oxidative damage and apoptosis in Dox-treated hearts
Oxidative stress plays critical roles in the initiation and progression of Dox-induced cardiotoxicity. 2,29 Accordingly, we detected a significant increase of ROS accumulation in the heart with Dox treatment, which was reduced by injections of the miR-495-3p agomir ( Figure 3A

| miR-495-3p antagomir exacerbates oxidative damage and apoptosis in Dox-treated hearts
In contrast with the antioxidant role of the miR-495-3p agomir, Doxinduced ROS generation was further amplified by the miR-495-3p antagomir, as evidenced by the increased ROS, MDA, 3-NT and 8-OHdG levels ( Figure 4A,B). Accordingly, treatment with the miR-495-3p antagomir also suppressed the activities of total SOD and CAT upon Dox injection ( Figure 4C). Meanwhile, mice treated with the miR-495-3p antagomir displayed lower NRF2 expression and activity than those treated with the antagomir control upon Dox injection ( Figure 4D,E). In addition, Dox-induced apoptosis of cardiac cells was also aggravated by the miR-495-3p antagomir, as evidenced by the increased TUNEL+nuclei, caspase-3 activity and decreased Bcl-2/Bax mRNA level ( Figure 4F-H). Collectively, we demonstrate that the miR-495-3p antagomir exacerbates oxidative damage and apoptosis in Dox-treated hearts.

| miR-495-3p modulates Dox-induced oxidative stress and cellular injury in vitro
Then, we evaluated the role of miR-495-3p in Dox-treated H9C2 cells in vitro, and the efficiency was presented in Figure 5A. Consistent

| miR-495-3p agomir prevents Dox-induced cardiotoxicity through activating AKT in vivo and in vitro
Next, we explored the underlying mechanisms that mediate the cardioprotective effects of the miR-495-3p agomir in vivo and in vitro.
AKT plays critical roles in cell survival and oxidative stress, and activating AKT is sufficient to alleviate Dox-induced cardiotoxicity. 13 Interestingly, we found that Dox-induced suppression on AKT phosphorylation was prevented by the miR-495-3p agomir, but further aggravated by the miR-495-3p antagomir ( Figure 6A,B). To determine whether the miR-495-3p agomir prevented Dox-induced cardiotoxicity through activating AKT, mice were treated with AKTi to inhibit AKT activity as previously described. As shown in Figure 6C

| miR-495-3p agomir activates AKT through downregulating PTEN
Finally, we tried to unveil the molecular basis through which the Quantitative results of caspase-3 activity in the heart. N = 6 per group. All results were expressed as the mean ± standard deviation and p < 0.05 was considered statistically significant activation and cardioprotection, H9C2 cells were infected with adenovirus to overexpress PTEN in vitro ( Figure 7F). As shown in Figure 7G, PTEN overexpression completely abolished the miR-495-3p agomir-induced AKT activation upon Dox treatment in H9C2 cells. Accordingly, the decreased oxidative stress and cellular injury in the miR-495-3p agomir-treated cells were also prevented in those with PTEN overexpression, as determined by the increased ROS generation, LDH releases and decreased cell viability ( Figure 7H-J). In addition, mice were also injected with AAV9 vectors carrying either PTEN or NC to specifically overexpress PTEN in the myocardium ( Figure 7K). As shown in Figure 7L,M, the miR-495-3p agomir significantly reduced Dox-induced oxidative stress and cell apoptosis in the heart, but failed to do so in those with PTEN overexpression. Accordingly, the miR-495-3p agomir-induced improvement of cardiac injury, cardiac mass loss and fibrotic remodelling was also blocked in PTEN-overexpressed mice ( Figure 7N-P). In addition, PTEN overexpression significantly abrogated functional restoration in the miR-495-3p agomir-treated mice upon Dox injection ( Figure 7Q). Overall, our findings determine that the miR-495-3p agomir activates AKT through downregulating PTEN.

| DISCUSS ION
Chemotherapy is one of the major approaches to treat human cancers, and Dox, belonging to the family of anthracyclines antibiotics, has attracted extensive interests due to its broad-spectrum antitumour capacities. Unfortunately, the clinical application of Dox is extremely impeded because of its cumulative cardiotoxicity. 45,46 Dexrazoxane is the only FDA-approved cardioprotectant to treat Dox-induced cardiotoxicity; however, long-term use of high doses of dexrazoxane can lead to severe hepatotoxicity and even second malignancy. 47 Therefore, it is of great significance to find novel therapeutic agents to replace dexrazoxane. In the present study, Oxidative stress is a key feature of Dox-induced cardiotoxicity, and contributes to cell loss and cardiac dysfunction. It has been reported that Dox is specifically and abundantly accumulated in the heart, and exhibits high affinity to cardiolipin. The formation of Dox-cardiolipin complex anchors Dox to the inner membrane of mitochondria that disturbs normal electron transport chain and promotes ROS generation in the heart. 48 In addition, the redoxcycling of Dox generates massive semiquinone free radicals and exacerbates oxidative stress. 49 Iron is also essential for Doxinduced ROS generation; however, the free iron level within most cells is lower than the threshold to couple with Dox to the extent necessary to cause cardiotoxicity. Emerging studies have identified that Dox interferes the proteins that sequester and bind intracellular iron, thereby increasing the accumulation of iron inside the mitochondria and promoting ROS amplification through a Fenton reaction. 7 Excessive free radicals induce oxidative damage to biomacromolecules and destroy cellular physiology. Moreover, the All results were expressed as the mean ± standard deviation and p < 0.05 was considered statistically significant less active antioxidant defence and negligible regenerative capability of the heart further exacerbate Dox-induced cardiotoxicity. 8 Accordingly, previous studies by us and the others have found that suppressing oxidative stress is sufficient to prevent Dox-induced cardiac injury and dysfunction. 9,46 miRNAs belong to the family of small noncosding RNAs and participate in various biological processes through negatively regulating gene expression. 50,51 Dysregulated miRNAs have been identified in the pathogenesis of Dox-induced cardiotoxicity, and even can be regarded as serum biomarkers for cardiac injury in Dox-treated patients. 52,53 We previously demonstrated that miR-22 was elevated in murine hearts with chronic Dox injection, and that miR-22 inhibition reduced Dox-induced oxidative stress, cardiomyocyte apoptosis and cardiac dysfunction in vivo and in vitro. 9 miR-495-3p expression is altered in various human tumours, and exhibits high potency to inhibit tumour growth and chemoresistance. 22,23 Herein, we determined that treatment with the miR-495-3p agomir could alleviate Dox-induced cardiotoxicity in vivo and in vitro. Based on these findings, we reasonably define miR-495-3p as an effective and safe target to treat Doxinduced cardiotoxicity. Of course, there are some limitations of the present study. First, the primary cellular source of miR-495-3p and its targeted cells have not been completely determined in this study. Second, the exact mechanisms mediating the cardioprotection of miR-495-3p except AKT pathway remains unknown. Third, whether the miR-495-3p agomir can sensitize human tumour cells to Dox chemotherapy needs further investigation.
In summary, we demonstrate that miR-495-3p diminishes Doxinduced cardiotoxicity through activating AKT pathway, and targeting miR-495-3p may provide novel cardioprotective approaches for cancer patients receiving anthracycline chemotherapy.

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 research data are not shared.