MBNL1-AS1 Promotes Hypoxia-Induced Myocardial Infarction via the miR-132-3p/RAB14/CAMTA1 Axis

Background Mounting evidence have indicated that long noncoding RNA (lncRNA) muscleblind like splicing regulator 1 antisense RNA 1 (MBNL1-AS1) play a crucial regulatory role in cardiovascular disease, myocardial infarction (MI) included. In this research, we sought to probe into the biological function and potential mechanism of MBNL1-AS1 in MI. Methods Cardiomyocytes were treated under hypoxic conditions for 0–12 h. Functional assays including CCK-8 and flow cytometry were performed to assess hypoxia-stimulated cardiomyocyte viability and apoptosis, respectively. Moreover, bioinformatics analysis and mechanical assays were conducted to reveal the competitive endogenous RNA (ceRNA) mechanism of MBNL1-AS1. Results The upregulation of MBNL1-AS1 was found in hypoxia-stimulated cardiomyocytes. Functionally, the downregulation of MBNL1-AS1 dramatically promoted hypoxia-induced cardiomyocyte viability and inhibited apoptosis. Mechanistically, miR-132-3p bound to MBNL1-AS1 in hypoxia-induced cardiomyocytes, and miR-132-3p directly targeted RAB14, member RAS oncogene family (RAB14) and calmodulin binding transcription activator 1 (CAMTA1). Furthermore, MBNL1-AS1 upregulates the expression of RAB14 and CAMTA1 in hypoxia-stimulated cardiomyocytes via targeting miR-132-3p. Conclusions The current study revealed the critical role of the MBNL1-AS1/miR-132-3p/RAB14/CAMTA1 axis in MI, indicating MBNL1-AS1 as an innovative therapeutic target for MI.


Introduction
As a kind of ischemic heart disease, myocardial infarction (MI) is accompanied with high morbidity and mortality [1,2]. Existing medicines and therapeutic methods can only retard the progression of the disease [3,4]. As a consequence, it is urgent to explore and develop effective treatment strategies to prevent MI.
Noncoding RNAs (ncRNAs), a type of RNAs lack of the potential of protein-coding, were initially recognized as "junk DNAs" [5,6]. Increasing researches have proved that about 98% of the human genome are ncRNAs, which have modulatory functions and can effectively couple back into a broader communication network [7,8]. As the two main types of ncRNAs, long ncRNAs (lncRNAs) and microRNAs (miR-NAs) have achieved extensive concern. Accumulating evidences have demonstrated that lncRNAs and miRNAs are important regulators in MI. Moreover, lncRNAs may act as competitive endogenous RNAs (ceRNA) to sequester miRNAs in MI development. At the same time, miRNAs usually function in MI through binding to the 3′-untranslated region (3′-UTR) of mRNAs, which affect the translation of mRNAs.
lncRNA MBNL1-AS1 has been widely documented to be implicated in multiple cancers, such as nonsmall cell lung cancer [9], bladder cancer, and retinoblastoma. Given the fact that the functions of MBNL1-AS1 in cancers have been well-studied, some studies have been devoted to exploring the role of MBNL1-AS1 in noncancerous diseases. A study proposed by Li et al. has indicated that MBNL1-AS1 targets KCNMA1 to enhance sevoflurane-pretreated ischemia-   Oxidative Medicine and Cellular Longevity reperfusion injury [10]. However, the role of MBNL1-AS1 in MI remains be investigated. In this research, to reveal the function of MBNL1-AS1 in MI, we constructed an in vitro model of MI in H9c2 cells treated with hypoxia. Moreover, we explored the interaction between MBNL1-AS1 and miRNA as well as the downstream genes. Our study might provide a promising prospect for MI treatment.

Material and Methods
2.1. Cell Culture and Treatment. The rat embryonic ventricular cardiomyocyte H9c2 and human embryonic kidney cell (HEK293T) were procured from American Type Culture Collection. Both H9c2 and HEK293T cells were maintained in DMEM (A4192101, Gibco, Rockville, MD, USA) containing 10% fetal bovine serum with 5% CO 2 at 37°C. To mimic MI, H9c2 cells were maintained in a hypoxia incubator containing 1% O 2 , 5% CO 2 , and 94% N 2 . . Three independent experiments were performed. Cardiomyocytes were cultured with 10 μl CCK-8 solution procured from Dojindo (Gaithersburg, MD, USA) in 96-well plates. The absorbance was measured at 450 nm.

Cell Counting
2.3. Flow Cytometry. The experiment was performed thrice independently. The FITC-annexin V/PI detection kit procured from Biosea Biotechnology (Beijing, China) was applied as per the user guide. Cardiomyocytes were collected and resuspended in 6-well plates, and then dyed with FITCannexin V and PI, and assessed by the cytometry procured from Beckman Coulter (Fort Collins, CO, USA).

Quantitative
Real-Time Polymerase Chain Reaction (RT-qPCR). Three experiments were performed independently. Total RNA was extracted from hypoxia-induced cardiomyocytes using the Trizol reagent procured from Life Technologies Corporation (Carlsbad, USA). After the application of First Strand cDNA Synthesis Kit (GeneCopoeia, USA), qPCR was implemented using Takara SYBR® PrimeScript™ PCR kit (11736051, Invitrogen, Carlsbad, CA, USA). Calculation of gene expression was performed using 2 −ΔΔCt method, with GAPDH or U6 as the internal control.            Oxidative Medicine and Cellular Longevity were treated with MS2-tagged MBNL1-AS1 to acquire miR-NAs related to MBNL1-AS1. The RIP assay was conducted as previously described [11]. The abundance of miRNAs was measured by RT-qPCR. 2.11. Statistical Analysis. Data of experimental results was presented as mean ± standard deviation and processed by Graphpad Prism 8 software (GraphPad Software, San Diego, United States). Statistical analyses were implemented with SPSS 19.0 software (IBM, Stanford University, United States). Statistical differences were tested by Student's t test or one-way analysis of variance (ANOVA). Statistical differences were considered significant when P < 0:05.

Upregulation of MBNL1-AS1 in Hypoxia-Induced H9c2
Cells. Firstly, H9c2 cells were subjected to the exposure of hypoxia for 0-12 h, and the viability of H9c2 cells was declined in a time dependent manner (Figure 1(a)). Since cell viability was reduced to about 50% after 8 h of hypoxia treatment, 8 h was chosen as a hypoxia-stimulating condition to be applied in the subsequent assays. Besides, the results from flow cytometry manifested that the apoptosis rate of H9c2 cells was promoted with increasing duration of hypoxia treatment (Figure 1(b)). This phenomenon was further obtained from western blot analysis. As indicated in Figure 1(c), the levels of proapoptotic proteins (Bax and cleaved caspase-3) were significantly elevated whereas that of Bcl-2 (antiapoptotic protein) was decreased as the hypoxia treatment time increased. These data collectively suggested that the MI in vitro model was successfully established. Then, we assessed the expression of MBNL1-AS1 in H9c2 cells subjected to hypoxic treatment using RT-qPCR. The results indicated that MBNL1-AS1 expression was elevated with increasing duration of hypoxia treatment (Figure 1(d)). In addition, the results of subcellular fractionation and FISH assays indicated that MBNL1-AS1 was majorly distributed in the cytoplasm of both H9c2 and hypoxia-induced H9c2 cells (Figures 1(e) and 1(f)), suggesting that MBNL1-AS1 might exert critical functions in hypoxia-induced H9c2 cells via posttranscriptional regulation.

MBNL1-AS1 Depletion Increases the Viability and
Decreases the Apoptosis of Hypoxia-Induced H9c2 Cells. To seek the functional impacts of MBNL1-AS1 in hypoxiainduced H9c2 cells, two specific shRNAs targeting MBNL1-AS1 were transfected into cells to silence MBNL1-AS1 expression. Then, RT-qPCR analysis verified the   (Figure 2(a)). The results of functional assays revealed that the viability of hypoxia-induced H9c2 cells was increased when MBNL1-AS1 was knocked down (Figure 2(b)). Meanwhile, MBNL1-AS1 depletion diminished the apoptosis rate of H9c2 cells under hypoxia treatment (Figure 2(c)). Additionally, the expression of apoptosis-linked proteins was detected by western blot analysis. The results showed that the expression of Bax and cleaved caspase-3 were decreased in hypoxia-induced H9c2 cells after MBNL1-AS1 deletion but that of Bcl-2 was enhanced (Figure 2(d)).

10
Oxidative Medicine and Cellular Longevity CAMTA1 and RAB14 were upregulated in H9c2 cells in a time-independent manner (Figure 4(b)). Besides, we silenced CAMTA1 and RAB14 expression and found that knockdown of CAMTA1 and RAB14 significantly elevated the viability and inhibited the apoptosis of H9c2 cells under hypoxia treatment ( Figure S2A-S2D). To test whether MBNL1-AS1, miR-132-3p, CAMTA1, and RAB14 coexisted in the RNA-induced silencing complex (RISC), RIP assay was performed using Ago2 antibody. It was revealed that these four RNAs were largely abundant in anti-Ago2 groups relative to anti-IgG groups (Figure 4(c)). Moreover, we severally predicted the binding sites of miR-132-3p on RAB14 3 ′ UTR and CAMTA1 3 ′ UTR, and found that miR-132-3p overexpression led to a reduction on the luciferase activities of RAB14 3 ′ UTR-WT and CAMTA1 3′UTR-WT in H9c2/Hypoxia and HEK293T cells, while barely influenced those of RAB14 3′UTR-MUT and CAMTA1 3′UTR-MUT (Figures 4(d) and 4(e)). In addition, we revealed that MBNL1-AS1 silencing markedly declined the expression of RAB14 and CAMTA1, while this effect was counteracted by miR-132-3p silencing (Figures 4(f) and 4(g)).

Discussion
A large amount of lncRNAs have been identified to be associated with multiple kinds of cardiovascular diseases, MI included. lncRNA CAIF represses autophagy and improves MI via p53-mediated myocardin transcription [12]. lncRNA H19 inhibits MI-induced myocardial injury through regulating KDM3A [13]. lncRNA Gm2691 impedes apoptosis and inflammatory response after MI via the PI3K/Akt signaling pathway [14]. In this research, we found that lncRNA MBNL1-AS1, a wide coverage gene in cancer progression, was significantly highly expressed in H9c2 cells under hyp-oxia treatment. Moreover, functional assays proved that MBNL1-AS1 silencing significantly promoted the viability and restrained the apoptosis of hypoxia-induced H9c2 cells. These findings were in line with a previous study about ischemia-reperfusion injury [10], suggesting that lncRNA MBNL1-AS1 might contribute to MI progression.
Recent studies have revealed that lncRNAs interact with miRNAs and exert important functions in the development of diseases. Dysregulation of miRNAs was extensively reported in the progression of MI. miRNA-145 represses MI-induced apoptosis via Akt3/mTOR signaling pathway [15]. miRNA-488-3p impedes acute MI-induced cardiomyocyte apoptosis via modulating ZNF791 [16]. miR-124 influences cardiomyocyte apoptosis and MI via targeting Dhcr24 [17]. Moreover, the lncRNA-miRNA interaction in MI is emerged as an interest topic. In this study, we focused on the relation between MBNL1-AS1 and miRNAs, with the purpose of revealing a potential mechanism of which MBNL1-AS1 promoted MI. We found that MBNL1-AS1 directly bound to miR-132-3p. As reported previously, miR-132-3p is involved in ischemic myocardial injury by interacting with lncRNA TUG1 [18]. In our study, we further showed that MBNL1-AS1 regulates MI in hypoxiainduced H9c2 cells via targeting miR-132-3p. Additionally, miRNAs combines with their target mRNAs to block the mRNA translation. miR-132-3p has been registered to directly target SOX11 to inhibit mantle cell lymphoma progression [19]. In our research, we found that RAB14 and CAMTA1 were targeted by miR-132-3p. Moreover, we validated that MBNL1-AS1 served as a sponge for miR-132-3p, and subsequent regulated RAB14 and CAMTA1 expressions.

Conclusions
To sum up, we demonstrated that lncRNA MBNL1-AS1 promoted hypoxia-induced MI in H9c2 cells via the regulation of miR-132-3p/RAB14/CAMTA1 axis. Our study might provide an innovative target for the treatment of MI. However, our study also exist some limitations. First, our study did not explore the upregulation mechanism of MBNL1-AS1 in hypoxia-treated H9c2 cells. Moreover, our study did not investigate whether MBNL1-AS1 is involved in MI via the modulation of certain pathways. All of these deficiencies will be improved in the near future.