LncRNA SUMO1P3 Aggravates Doxorubicin-Induced Cardiomyocyte Apoptosis by Targeting miR-93-5p / Bin1

Background/Aims Non-coding RNA plays a critical role in myocardial apoptosis induced by doxorubicin (DOX). However, the specic function of Long noncoding RNA (lncRNA) small ubiquitin-like modier 1 pseudogene 3 (SUMO1P3) is unclear. The purpose of this study was to determine the role of lncRNA SUMO1P3 in myocardial apoptosis induced by DOX. Methods QRT-PCR were used to detect the expression levels of SUMO1P3 and miR-93-5p in DOX-treated primary cardiomyocytes and rat models. QRT-PCR and Western blot were used to detect the expression levels of Bin1 in DOX-treated primary cardiomyocytes and rat models. The relationship between SUMO1P3, miR-93-5p and Bin1 was analyzed using bioinformatics analysis and Luciferase reporter assay. The effects of DOX on the viability and apoptosis of cardiomyocytes were evaluated by ow cytometry and CCK-8. The effects of SUMO1P3 on cardiomyocyte apoptosis were analyzed by TUNEL staining and echocardiography. si-SUMO1P3 on primary cardiomyocyte apoptosis (P < 0.01). that miR-93-5p Bin1.


Introduction
Doxorubicin (DOX) is a clinically effective and widely used antitumor drug [1,2]. However, its clinical application is limited by its cardiotoxicity [3]. The cardiotoxicity of DOX can be divided into acute toxicity and chronic toxicity according to the course of the disease [4]. Acute toxicity is often manifested as acute multiple organ failure induced by a small dose of DOX. However, the chronic toxicity of DOX is more common in the clinic [5]. Because of the strong a nity between DOX and cardiomyocytes, DOX is easy to accumulate in the heart and produce cardiotoxicity. Long term use of drugs causes damage to the function of myocardial cells, leads to irreversible myocardial damage and dysfunction, eventually induces dose-dependent and time-dependent congestive heart failure [6]. The end-stage heart failure for cancer patients seriously affects the survival rate [7]. The mechanism has always been a research hotspot, but the exact mechanism needs to be further explored. At present, the main view is that the toxicity of DOX is mainly re ected in the induction of oxidative stress and lipid peroxidation in cardiomyocytes, which affect the synthesis of nucleic acid and protein [8]. Among them, cardiomyocyte apoptosis may be a critical role of DOX-induced cardiotoxicity [9]. DOX can lead to cardiac dysfunction, and eventually develop into severe heart failure or even death. Free radical injury, mitochondrial injury, abnormal energy metabolism, calcium overload and apoptosis are all involved in DOX induced myocardial injury. At present, it is considered that the cardiotoxicity of DOX may be the result of many factors, which are related to each other and promote the development of cardiotoxicity [10]. Therefore, it is urgent to develop new prevention strategies for DOX-induced cardiotoxicity.
Long noncoding RNAs (lncRNAs) are involved in cardiac development and closely related to heart disease [11,12]. LncRNAs are RNAs that are longer than 200 bp and do not have protein encoding functions [13].
Given the critical roles of LncRNAs in many important regulatory processes, such as genomic imprinting, transcriptional activation, transcriptional interference, chromatin modi cation, this suggests a tremendous in uence of LncRNAs in cardiovascular diseases [14]. Recent studies of lncRNAs in the cardiac system have mainly focused on its role in heart injury and remodeling, leukocyte and in ammation [15]. For example, it is found that lncRNA GASL1 is downregulated in chronic heart failure (CHF). The overexpression of lncRNA GASL1 may improve CHF by inhibiting the inactivation of TGF-β1 and inhibiting cardiomyocyte apoptosis [16]. In particular, lncRNA SUMO1P3 is a recently discovered new lncRNA and has been indicated as an oncogenic lncRNA in bladder cancer and breast cancer [17,18].
However, there haven't been any studies on the biological function of SUMO1P3 on DOX induced cardiac cells.
Accumulating evidence demonstrates that lncRNAs serve as competing endogenous RNAs (ceRNAs) through competitive binding to miRNAs [19]. MiRNA is widely involved in cell differentiation, disease, repair and apoptosis. A variety of miRNAs can be expressed in mammalian cardiac tissue, which play a critical role in the progress of cardiac diseases, including cardiac hypertrophy, arrhythmia, myocardial infarction, myocardial brosis, myocardial cell apoptosis, heart failure and myocardial apoptosis [20][21][22].
In particular, recent studies have found that miR-93-5p plays an important role in the pathological processes of various cancer via regulating the expression of related genes [23]. For example, miR-93-5p inhibits the cellular migration of breast cancer cells [24] and promotes gastric cancer metastasis [25].
Previous study has shown that miR-93 inhibits the oxygen-glucose deprivation/reoxygenation-induced the cardiomyocyte apoptosis [26]. However, the role of miR-93-5p in DOX-induced cardiomyocyte apoptosis and its target are not clear. Based on the prediction results of Starbase v2.0, one of the putative targets of miR-93-5p is Bin1. Bin1 is a N-terminal binding protein of c-myc protein, which is the only ligand protein with anti-cancer function [27]. It has been shown that Bin1 expression is low or absent in breast cancer, liver cancer, bladder cancer and colon cancer [28,29].
Based on the aforementioned literature, the interactions between lncRNAs, miRNA and Bin1 are critical regulators in cell cycle progression and relevant diseases. Thus, we hypothesize that lncRNA SUMO1P3 has protective effect on DOX-induced myocardial injury through the miR-93-5p/Bin1 axis. This provides a theoretical basis for expanding the biological role of SUMO1P3 and clinical protection of DOX-induced myocardial injury.

Primary Cardiomyocyte Cultures
Postnatal Wistar rats younger than 12 h were chosen and disinfected using 75% ethanol. The hearts were separated and soaked in 1% PBS to remove the blood. Then, the ventricles were selected and cut into 1 mm 3 blocks. The tissue blocks were digested with 0.25% trypsin-EDTA (Gibco, USA) at 37 °C for 8 min, and the upper layer suspension was discarded. Then, the remaining tissues were continued to be digested with 0.25% trypsin-EDTA at 37 °C for 8 min, and the upper layer suspension was collected. Then, the action of trypsin was terminated by the addition of the same amount of culture medium (10% fetal bovine serum, 1% L-glutamine, 1% penicillin-streptomycin solution and 88% DMEM). The remaining tissues were continued to be digested 3-5 times using the aforementioned steps until all the tissues were digested. The collected suspension was centrifuged at 1,000 rpm for 10 min. The cells were plated at a cell density of 5 × 10 5 cells/mL on the 0.1 mg/mL poly-D-lysine-coated 15-mm confocal dishes.
Afterwards, the medium was changed every two days. The experimental results con rmed that the isolated cells expressed cadriac-α-acitn, and successfully established the model of cultured cardiomyocytes in vitro.
Cell culture HEK293T cells for the luciferase reporter assay were purchased from the Institute of Biochemistry and Cell Biology (Shanghai). All cells were cultured in DMEM Medium (Gibco, USA) supplementing with 10% fetal bovine serum (FBS) (Invitrogen), and 1% penicillin/streptomycin at 37°C with 5% CO 2 .

In vivo studies
All animal studies were approved by First A liated Hospital of Xi'an Jiaotong University Animal Protection and Committee. DOX at a dose of 8 mg/kg was injected intraperitoneally into the rats at 3 week intervals (cumulative dose of 24 mg/kg). Rats in the control group (sham operation group, n = 10) were injected intraperitoneally with saline. The parameters of echocardiography were analyzed by VEVO 770 (VisualSonics Inc, Toronto, Canada). LVFS (left ventricular fractional shortening) and LVEF (Left ventricular ejection fraction) were analyzed by linear sensor (Minuo, Shenzhen, China).

Adenovirus gene delivery
Recombinant adenovirus containing mouse SUMO1P3 shRNA (Ad-SUMO1P3-shRNA), 7 days before DOX induction, adenovirus was intratracheally dripped into rats. Control adenovirus (Ad-GFP) was injected into the control group. MiR-93-5p mimic and its NC were purchased from GenePharma (Shanghai, China). 50 μg miR-93-5p mimic or its negative control was dissolved in 50 μl of sterile double distilled water, and 50 μl of glucose solution was added. Then, the tail vein of rats was injected with 200 μ l working solution.

RNA extraction and quantitative real-time PCR
Total RNA in tissues and cells was extracted using TRIzol reagent (Biosntech, Beijing, China). The quality of RNA was analyzed using NanoDrop 1000 (Thermo Fisher Scienti c, Inc., Waltham, MA, USA). SYBR-Green (Takara Biotechnology, Co., Lt. (Dalian, China) was used in by qRT-PCR. Ampli cation was performed by ABI 7,500 real-time PCR system. A qScript microRNA cDNA synthesis kit (Quantabio, Beverly, CA, USA) was used for cDNA synthesis. The expression levels of SUMO1P3 and miR-93-5p were calculated using the 2 -△△CT method. The expression levels of miRNA were standardized by U6. The expression levels of lncRNA were standardized by GAPDH. The primers used for qRT-PCR analysis were listed in Table 1 (n=3). The cells were harvested and washed with 1 × PBS, and then we used 2 × SDS loading buffer to lyse cells. The lysates were boiled at 95℃ for 10 min. The solution was subject to centrifuge at 12,000 rpm for 1 min. About 60 ug of total protein (10 -15 ul) was loaded onto SDS-PAGE gel and resolved at 120 V for 0.5-1 h. After that, the proteins were transferred to PVDF membrane at 300 mA for 2-3 h. The membrane was blocked with 5% non-fat milk in 1 × TBST for 1 h at room temperature, and then the membrane was incubated with proper primary antibodies at 4℃, overnight. The following day, the membrane was washed with 1 × TBST for 3 times, 10 min each time. The membrane was incubated with secondary antibody at room temperature for 1 h. Finally, the membrane was incubated with ECL and then exposed using Bio-Rad ChemiDoc Touch Imaging System. The following antibodies were used in this study: anti-Bin1 (1: 1000, Youliante, Shanghai, China) and anti-GAPDH antibodies (1: 1000, Youliante, Shanghai, China) overnight. An anti-rabbit secondary antibody (1: 1000, Youliante, Shanghai, China).
Results were visualized with using the Supersignal West Dura Substrate (Pierce). After that, the relative protein expression was expressed as the ratio of the gray value of the target band / GAPDH band.

TUNEL analysis
The TUNEL method was used to analyze the apoptotic index of myocardial cells. The nuclei of positive apoptotic cells were brown-yellow granules. Five 400 high-power elds were randomly observed. The apoptotic index (AI), AI (%) = (number of apoptotic positive nuclei / total counted nuclei) × 100%, was calculated.

Statistical methods
Data are expressed as mean ± SEM. Statistical analysis of the results was performed with GraphPad Prism version 5.0 (GraphPadSoftware Inc., San Diego, CA, USA). Student's t-test was used for two group comparison. One-way ANOVA followed by Dunnet t-test was used for multiple group comparison. P < 0.05 was considered statistically signi cant.

Results
SUMO1P3 was highly expressed in doxorubicin stimulated primary cardiomyocyte As shown in Fig. 1A and 1B, DOX gradually reduced the cell viability of primary cardiomyocyte cells with the increase in the dose and duration (P < 0.01). Then, the expression level of SUMO1P3 was analyzed in the cells after DOX treatment. The expression level of SUMO1P3 was increased signi cantly with the increase of the dose and duration of DOX action (P < 0.01, Fig.1C and 1D). These results suggest that SUMO1P3 plays a key role in myocardial cell apoptosis induced by DOX.

SUMO1P3 knockdown inhibited myocardial apoptosis
To further analyze whether SUMO1P3 was involved in adriamycin-induced apoptosis of myocardial cells. DOX signi cantly inhibited the activation of primary cardiomyocytes and induced the apoptosis of primary cardiomyocytes contrasted with the Control group (P < 0.01). The 2-ΔΔCt values of si-NC and si-SUMO1P3 was (18.19 ± 7.94) and (33.42 ± 2.64) respectively. Contrasted with the DOX-treated group, the viability of primary cardiomyocytes was signi cantly raised, and the apoptosis of primary cardiomyocytes was inhibited after DOX treatment in the si-SUMO1P3 group (P < 0.01, Fig. 2A and 2B). As shown in Fig. 2C, compared to the control group, DOX treatment signi cantly increased the expression levels of Bax and C-Caspase-3 protein (P < 0.01) and inhibited the expression level of Bcl-2 (P < 0.01) in the si-NC group. Contrasted with the DOX-treated group, transfection with si-SUMO1P3 signi cantly inhibited the Dox-induced Bax and C-Caspase-3 protein expression (P < 0.01), and restored the Bcl-2 protein expression level in primary cardiomyocytes (P < 0.01, Fig. 2C). These results indicate that si-SUMO1P3 inhibits the apoptosis of primary cardiomyocytes.

SUMO1P3 functions as a ceRNA of miR-93-5p in primary cardiomyocytes
Next, the potential mechanism of SUMO1P3 regulating DOX-induced cytotoxicity was explored. Starbase v2.0 online prediction tool was used and predicted that MiR-93-5p was a potential target of SUMO1P3 (Fig.3A). To con rm whether miR-93-5p directly binds to SUMO1P3, we performed a dual-luciferase reporter assay. SUMO1P3 containing the wild type or mutant type putative miR-93-5p binding site was inserted into a reporter vector that was cotransfected with miR-93-5p mimics into cells. The results demonstrated that miR-93-5p overexpression signi cantly reduced the luciferase activity of a reporter vector containing the wild type SUMO1P3 (P < 0.01) (Fig. 3B). However, miR-93-5p overexpression showed no obvious effect on the luciferase activity of the reporter vector containing the mutant SUMO1P3 (Fig. 3B). As shown in Fig. 3C, the expression level of SUMO1P3 was reduced in the si-SUMO1P3 group while increased in the SUMO1P3 mimic group, indicating high e ciency of transfection. Compared to the si-NC group, the expression level of miR-93-5p was signi cantly increased in the si-SUMO1P3 group (P < 0.05), while signi cantly reduced after the addition of SUMO1P3 (P < 0.05) (Fig.   3C). In addition, the expression level of miR-93-5p in the DOX group was signi cantly reduced contrasted with the Control group (P < 0.05, Fig. 3D). In summary, these results demonstrate that SUMO1P3 targets and negatively regulates the expression of miR-93-5p.
Transfection of miR-93-5p mimic signi cantly increased the expression level of miR-93-5p, which was in turn reduced by co-transfection of SUMO1P3 (Fig. 4C). This result was consistent with previous data in Fig. 3C. Compared with the NC group, the expression level of Bin1 was signi cantly decreased in the miR-93-5p mimic group (P < 0.01), while it was signi cantly increased in the SUMO1P3 overexpression group (P < 0.01). Co-transfection of SUMO1P3 with miR-93-5p reversed SUMO1P3-induced expression of Bin1 (P < 0.01, Fig. 4C). The 2 -ΔΔCt values of miR-NC and miR-93-5p mimic was (15.04 ± 4.58) and (27.32 ± 5.71), respectively. As shown in Fig. 4C and 4D, contrasted with the miR-NC group, the expression level of Bin1 mRNA and protein was signi cantly reduced in the miR-93-5p mimic group (P < 0.01), and the expression levels of Bin1 gene and protein were signi cantly increased in the SUMO1P3 overexpression group (P < 0.01) (Fig. 4C-4D). Co-transfection of SUMO1P3 with miR-93-5p reversed the SUMO1P3induced expression of Bin1 (P < 0.01, Fig. 4C) (P < 0.01) (Fig. 4C-4D). In addition, the expression level of Bin1 in the DOX treatment group was signi cantly raised compared with the Control group (P < 0.01, Fig.  4E). In summary, these results demonstrate that SUMO1P3 increases the expression of Bin1 by acting as a sponge for miR-93-5p.
SUMO1P3 knockdown inhibited myocardial apoptosis via miR-93-5p/ Bin1 axis in vitro Next, the mechanism of action of SUMO1P3 in adriamycin-induced cardiomyocyte apoptosis was further analyzed. DOX signi cantly inhibited the proliferation of primary cardiomyocytes contrasted with the control group (P< 0.01), and cell viability of primary cardiomyocytes was signi cantly raised in the si-SUMO1P3 group, miR-93-5p mimic group, and si-Bin1 group (P < 0.01), while co-transfection of si-SUMO1P3 with miR-93-5p or si-SUMO1P3 and si-Bin1 reversed the effect of si-SUMO1P3 on the viability of primary cardiomyocytes (P < 0.01, Fig. 5A and 5B). As shown in Fig.5C and 5D, DOX signi cantly promoted the apoptosis of primary cardiomyocytes (P < 0.01). Apoptosis of primary myocardial cells was signi cantly inhibited in the si-SUMO1P3 group, miR-93-5p overexpression group and si-Bin1 group (P < 0.01), while co-transfection of si-SUMO1P3 with miR-93-5p or si-SUMO1P3 and si-Bin1 reversed the effect of si-SUMO1P3 on primary cardiomyocyte apoptosis (P < 0.01). These results indicate that SUMO1P3 inhibites primary cardiomyocyte apoptosis through miR-93-5p / Bin1.

SUMO1P3 knockdown improved myocardial function through regulating miR-93-5p in vivo
Four weeks after the rst injection of DOX in rats, organ samples were collected and cardiac function was measured by echocardiography. As shown in Fig.6A and 6B, si-SUMO1P3 signi cantly inhibited the expression levels of SUMO1P3 and Bin1 (P <0.01), while si-SUMO1P3 signi cantly increased the expression level of miR-93-5p in mouse heart tissues (P <0.01), co-transfection of si-SUMO1P3 with miR-93-5p reversed the effects of si-SUMO1P3 on SUMO1P3, Bin1 and miR-93-5p expression levels (P < 0.01). Contrasted with the Control group, LVEF and LVFS scores were signi cantly reduced after DOX treatment (P < 0.01), while LVEF and LVFS scores were signi cantly raised in the si-SUMO1P3 group and miR-93-5p overexpression group (P < 0.01). Co-transfection of si-SUMO1P3 with miR-93-5p was able to reverse the effect of si-SUMO1P3 on LVEF and LVFS scores (P < 0.01, Fig. 6C and 6D). As The TUNEL staining results showed that the apoptosis rate in the DOX group was signi cantly increased (P < 0.01), the apoptosis rate was signi cantly reduced in the si-SUMO1P3 group and the miR-93-5p overexpression group (P < 0.01), while co-transfection of si-SUMO1P3 and miR-93-5p reversed the effect of si-SUMO1P3 on the apoptosis rate (P < 0.01, Fig. 6E). These results indicate that si-SUMO1P3 relieves myocardial apoptosis via miR-93-5p/ Bin1 axis.

Discussion
Doxorubicin (DOX) is often used to treat hematological malignancies and solid tumors [30]. However, it has been found in clinical studies that doxorubicin has a strong a nity with cardiomyocytes [31,32].
Although cardiomyocytes cultured in vitro lack the persuasiveness of animal experiments in the observation of end events, they reduce bias factors and are more conducive to the study of cell mechanism. These experiments explored the potential mechanisms of DOX cardiotoxicity by establishing cardiomyocyte apoptosis model in vitro.
The role of long-chain non-coding RNA (lnc RNA) has attracted widespread attention [33]. LncRNAs have been reported in nervous system disorders, metabolic diseases, reproductive development and cardiovascular diseases [34]. Recent study has con rmed that lncRNA may play an important role in cardiac regeneration and repair as a potential target of treatment [35]. For example, studies have found that LncRNA Carl inhibits mitochondrial ssion and cardiomyocyte apoptosis and reduces ischemiareperfusion injury [36]. So far, the mechanism of lncRNA on DOX-induced cardiotoxicity is still largely unknown. LncRNA SUMO1P3 is a recently discovered lncRNA, and studies has found that SUMO1P3 is oncogenic. Zhang et al. found that SUMO1P3 was signi cantly up-regulated in bladder cancer tissues, and knockdown of SUMO1P3 inhibited bladder cancer cell proliferation / migration inhibition and induced bladder cancer cell apoptosis [37]. Similarly, Liu et al. fount that SUMO1P3 expression was higher in breast cancer tissues and the high levels of SUMO1P3 expression associated signi cantly with tumor progression and poor survival of breast cancer patients. Moreover, knockdown of SUMO1P3 suppressed proliferation, migration and invasion of breast cancer cells [18]. Altogether, these studies suggest that SUMO1P3 functions as an oncogenic lncRNA in some cancer cells. In our present study, we found that with the increase of the dose and duration of DOX, the expression level of SUMO1P3 was raised. Moreover, si-SUMO1P3 inhibited the apoptosis of primary cardiomyocytes.
A subset of miRNAs has been shown to be highly and speci cally expressed in the heart muscle, including miRNA-378, miRNA-208, miRNA-499 and miRNA-22 [38,39]. For example, it has been found that overexpression of miRNA-133a in cardiomyocytes signi cantly reduces H 2 O 2 induced cardiomyocyte apoptosis, indicating that miRNA plays a critical role in the mechanism of anti-cardiomyocyte apoptosis [40]. miR-93-5p has different roles in different diseases [41]. For example, miR-93-5p enhances the growth of HUVECs [42]. Our present study nds that miR-93-5p was a potential target of SUMO1P3. The expression level of miR-93-5p in the DOX group was signi cantly reduced. The expression level of miR-93-5p in the si-SUMO1P3 group was reduced, while the expression levels of LVEF and LVFS were raised, and the apoptosis rate was reduced in the si-SUMO1P3 group and miR-93-5p group. Co-transfection of si-SUMO1P3 with miR-93-5p reversed the effect of si-SUMO1P3 on LVEF, LVFS score and apoptosis. These results demonstrated SUMO1P3 exerted their biological functions by miR-93-5p.
Bin1 is reported to bind to the Myc binding domain (MBD) site at the N-terminal of c-MYC [43], which is a ligand protein with anti-cancer characteristics and has the characteristics of tumor suppressor gene, and is likely to participate in the downregulation of cell growth [44]. Bin1 is widely expressed in the normal body cells [45]. The proliferation, metastasis and invasion of malignant cells to surrounding tissues may be caused by the absence of Bin1. The overexpression of Bin1 inhibits the growth or in ltration of tumor cells, apoptosis and malignant transformation [46]. Bin1 plays a part in in the occurrence of many diseases, including cardiovascular and cerebrovascular diseases, etc. [47]. This study found that Bin1 is a potential target of miR-93-5p. Viability of primary cardiomyocytes in the si-SUMO1P3 group, miR-93-5p mimic group, and si-Bin1 group was raised, and apoptosis was reduced. Co-transfection of si-SUMO1P3 with miR-93-5p or si-SUMO1P3 and si-Bin1 reversed the effects of si-SUMO1P3 on the viability and apoptosis of primary cardiomyocytes. These results indicate that SUMO1P3 inhibits primary cardiomyocyte apoptosis through miR-93-5p/Bin1.

Conclusion
Si-SUMO1P3 protected myocardial cells from DOX-induced damage through miR-93-5p/Bin1, which is of great signi cance for the treatment of DOX -induced heart damage.

Declarations
The data availability statement: The data that support the ndings of this study are available on request from the corresponding author, Aiqun Ma. The data are not publicly available due to their containing information that could compromise the privacy of research participants.
Author's Contribution: Aiqun Ma designed the study. Xuefeng Lin carried out experiments and wrote the manuscipt, Aiqun Ma revised the paper,Xuefeng Lin collected patient specimens and related information and contributed to analysing the data. All authors reviewed the results and approved the nal version of the manuscript.

Compliance with Ethical Standards
Funding: Not applicable.
Ethical approval: This article does not contain any studies with human participants or animals performed by any of the authors.