Advances in Clinical and Experimental Medicine

Title abbreviation: Adv Clin Exp Med
JCR Impact Factor (IF) – 2.1
5-Year Impact Factor – 2.2
Scopus CiteScore – 3.4
Index Copernicus  – 168.52; MEiN – 140 pts

ISSN 1899–5276 (print)
ISSN 2451-2680 (online)
Periodicity – monthly

Download original text (EN)

Advances in Clinical and Experimental Medicine

Ahead of print

doi: 10.17219/acem/168242

Publication type: original article

Language: English

License: Creative Commons Attribution 3.0 Unported (CC BY 3.0)

Download citation:

  • BIBTEX (JabRef, Mendeley)
  • RIS (Papers, Reference Manager, RefWorks, Zotero)

Cite as:


Li G, Tang X, Tang H. Circular RNA ANKIB1 alleviates hypoxia-induced cardiomyocyte injury by modulating miR-452-5p/SLC7A11 axis [published online as ahead of print on August 7, 2023]. Adv Clin Exp Med. 2024. doi:10.17219/acem/168242

Circular RNA ANKIB1 alleviates hypoxia-induced cardiomyocyte injury by modulating miR-452-5p/SLC7A11 axis

Gang Li1,D, Xiaolei Tang2,C, Huaping Tang1,A,B

1 Department of Geriatric Medicine, Maanshan People’s Hospital, China

2 Department of Clinical Laboratory, The Second Affiliated Hospital of Wannan Medical College, Wuhu, China

Graphical abstract


Graphical abstracts

Abstract

Background. Acute myocardial infarction (AMI) is a common cardiovascular disease worldwide. Circular RNAs (circRNAs) have been shown to exert essential roles in the progression of AMI. However, it remains unclear whether circANKIB1 protects cardiomyocytes from hypoxia-induced injury.

Objectives. The aim of the study was to elucidate the function and mechanisms of circANKIB1 in AMI.

Materials and methods. The expression of RNA was estimated using a quantitative real-time polymerase chain reaction (qPCR) assay, and the level of protein was determined with the use of western blot analysis. Methyl thiazolyl tetrazolium (MTT) assay was introduced to test cell viability, and a terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay was used to detect apoptosis. The relative levels of ferrous ion (Fe2+), reactive oxygen species (ROS) and malondialdehyde (MDA) were measured with their corresponding detection kits. The potential target of circANKIB1 and miR-452-5p was predicted using the StarBase database and verified by employing a dual luciferase reporter assay.

Results. This study showed a significant decrease in circANKIB1 in hypoxia-treated H9c2 cells. Hypoxic exposure significantly reduced the viability of H9c2 cells and the expression of GPX4, and increased the content of Fe2+, ROS and MDA. These effects were reversed by the overexpression of circANKIB1. Additionally, miR-452-5p was found to be a direct target of circANKIB1, and the miR-452-5p mimic significantly eliminated the protective effect of circANKIB1 overexpression in hypoxia-induced cells. In addition, miR-452-5p could bind to SLC7A11 and negatively regulate its expression. The knockdown of SLC7A11 abolished the effect of circANKIB1 overexpression on hypoxia-induced cardiomyocyte injury.

Conclusions. This investigation revealed for the first time that circANKIB1 regulated signaling of the miR-452-5p/SLC7A11 axis, thereby ameliorating hypoxia-induced cardiomyocyte injury. These findings suggest that circANKIB1 might be a useful adjunct in the treatment of AMI.

Key words: cardiomyocyte, SLC7A11, acute myocardial infarction, miR-452-5p, circANKIB1

Background

Acute myocardial infarction (AMI), as a cardiovascular disease, is associated with cardiomyocyte injury caused by acute ischemia and hypoxia. Acute myocardial infarction causes severe damage to the myocardium, resulting in high mortality.1, 2 Previous findings suggest that AMI could induce oxidative stress in cardiomyocytes, ultimately leading to their dysfunction.3, 4 Exploring the molecular network of cardiomyocyte injury and revealing target genes that protect them from acute ischemia and hypoxia are essential for the treatment of AMI patients.

Circular RNA (circRNA) is highly stable and displays a novel circular structure without 5’ or 3’ ends.5, 6, 7 CircRNAs are involved in the regulation of various intracellular activities8 and are considered markers in a variety of human diseases. Additionally, scientists have demonstrated that multiple circRNAs play an important role in the progression of AMI.9, 10 For example, Ren et al. demonstrated that circ_0023461 knockdown mitigated hypoxia-induced cardiomyocyte injury by mediating the miR-370-3p/PDE4D signaling axis.11 Wang et al. discovered that circUBXN7 improved hypoxia/reoxygenation (H/R)-induced apoptosis and the inflammatory responses in cardiomyocytes by modulating the miR-622/MCL1 axis signaling.12 The circRNA ankyrin repeats and IBR domain containing 1 (circANKIB1) have been reported to be a splicing product of ANKIB1 mRNA that regulate the development of multiple cells.13, 14 Mao et al. revealed that circANKIB1 was closely related to the process of peripheral nerve injury.15 However, the role of circANKIB1 in hypoxia-induced cardiomyocyte injury remains unclear.

MicroRNAs (miRNAs) are another widely studied non-coding RNAs that have been implicated in diverse biological functions, such as cell proliferation.16 Current investigations have verified that circRNAs could serve as competitive endogenous RNA (ceRNAs) of miRNAs to influence the expression of downstream mRNA.17 Here, we examined the connection between circANKIB1 and miR-452-5p and explored the effects of circANKIB1/miR-452-5p in hypoxia-induced cardiomyocyte injury.

Iron accumulation has been reported to increase the risk of cardiovascular disease (CVD).18 Ferroptosis is a recently discovered mode of cell death characterized by iron-dependent lipid peroxides increased into the toxic range.19, 20 In addition, Fang et al. reported that ferroptosis can modulate ischemia–reperfusion-induced cardiomyopathy.19 Therefore, revealing the regulators of ferroptosis is crucial for finding the appropriate therapeutic measures for AMI. The SLC7A11 is a multichannel transmembrane protein that prevents the development of ferroptosis by increasing glutathione (GSH) synthesis and reducing the accumulation of lipid oxide.21, 22 Our previous study found that SLC7A11 is an important potential target for miR-452-5p. However, whether circANKIB1/miR-452-5p mediates ferroptosis of cardiomyocytes under hypoxia conditions by SLC7A11 is yet to be explored.

Objectives

We hypothesized that circANKIB1 plays a key role in hypoxia-induced cardiomyocyte injury by regulating SLC7A11-associated ferroptosis, which would provide a promising therapeutic network for the treatment of AMI.

Materials and methods

Cell culture

Rat H9c2 cardiomyocytes (BNCC337726) were obtained from the BeNa Culture Collection (Beijing, China) and maintained in Dulbecco’s modified Eagle’s medium (DMEM; SH30022.LS; Hyclone, Logan, USA) with 10% fetal bovine serum (FBS; F8318; Sigma-Aldrich, St. Louis, USA) and 1× Penicillin-Streptomycin (P1400; Solarbio, Beijing, China) at 37°C and 5% CO2.23

Hypoxia model

The hypoxia model was completed as described in the previous report.11 The H9c2 cells were subjected to hypoxic conditions (O2:CO2:N2 = 1:5:94) for 24 h, 48 h and 72 h, and then cultured under normoxia conditions (O2:CO2:N2 = 21:5:74) for an additional 6 h to simulate AMI in vitro. The H9c2 cells cultured for 30 h under normoxia conditions (O2:CO2:N2 = 21:5:74) were considered control cells.

Cell transfection

The vector with circANKIB1 (oe-circANKIB1) and the empty vector (vector) were obtained from Hunan Fenghui Biotechnology Co. Ltd. (Hunan, China). Small interfering RNA (siRNA) targeting the SLC7A11 sequence (si-SLC7A11) and the corresponding scrambled control (si-NC) were synthesized by Sangon Biotech (Shanghai, China). The miR-452-5p mimic/inhibitor and its control (mimic/inhibitor)-NC were obtained from Guangzhou Ruibo Biotechnology Co. Ltd. (Guangzhou, China). Lipofectamine 2000 Transfection Reagent (L7800; Solarbio) was used to transfect all vectors (4 µg) and oligonucleotides (50 nM) into H9c2 cells.24

qPCR assay

Total RNA was extracted from H9c2 cells using Trizol reagent (Invitrogen, Waltham, USA). NanoDrop 2000 nucleic acid analyzer (Thermo Fisher Scientific, Waltham, USA) was used to detect the concentration and purity of RNA to confirm that the A260/A280 ratio was 1.9–2.1, according to the Minimum Information for Publication of Quantitative Real-Time PCR Experiments (MIQE) guidelines.25 The RNA was reverse-transcribed into complementary DNA (cDNA) utilizing the Evo M-MLV RT Master Mix (AG11706; Accurate Biology, Hunan, China). Then, a quantitative polymerase chain reaction (qPCR) analysis was performed using the qPCR kit (RK02001; BioMarker Technologies, Beijing, China) on the QuantStudio 5 RT fluorescence qPCR instrument system (BJ005277; ABI, Hunan, China). The following parameters were used for qPCR: 1 cycle at 98°C for 3 min, followed by 40 cycles for 15 s at 94°C, 30 s at 60°C and 1 min at 72°C. The U6 served as the internal reference for miR-452-5p, while glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was the internal reference for circANKIB1. The 2−ΔΔCt method was utilized to estimate the levels of miR-452-5p and circANKIB1.23 All primers (synthesized by BGI Group, Shenzhen, China) are shown in Table 1.

RNase R assay

The extracted RNA was treated with 1 U/μg RNase R (14606ES72; YESEN, Shanghai, China) for 30 min, followed by quantitative real-time polymerase chain reaction (qPCR) analysis to detect the expression of circANKIB1 and liner ANKIB1. A sample of non-processed RNA was considered the control group.26

Subcellular localization assay

A PARIS kit (AM1921; Invitrogen) was used to isolate cytoplasmic and nuclear RNA from H9c2 cells. The qPCR was then performed to determine the content of circANKIB1 distributed in the cytoplasm or nucleus in H9c2 cells. The GAPDH and U6 served as the housekeeping genes for the cytoplasm and nucleus, respectively.27

Methyl thiazolyl tetrazolium (MTT) assay

Cell viability was tested by employing the MTT Cell Proliferation Assay Kit (40206ES76; YESEN). Briefly, the transfected H9c2 cells were cultured overnight in 96-well plates (3×103 cells/well), and cell transfection was performed for 48 h. Then 10 μL of MTT was added to each well for 4 h. The optical density (OD) was recorded at 570 nm with a microplate reader (SpectraMax Mini; Molecular Devices Ltd., Shanghai, China).28

Reactive oxygen species and malondialdehyde detection

For the detection of reactive oxygen species (ROS), cells were incubated with a 2’-7’-dichlorofluorescin diacetate (DCFH-DA) probe (10 μM; Beyotime Biotechnology, Shanghai, China) for 30 min in darkness at room temperature. Then, the cells were rinsed twice with phosphate-buffered saline (PBS) and imaged with a fluorescent microscope (Leica DM1000; Leica Camera, Wetzlar, Germany). The fluorescence intensity was examined using a fluorescent microplate reader (Epoch2; BioTek, Vermont, USA; excitation/emission 488/525 nm). The level of ROS and malondialdehyde (MDA) in H9c2 cells was estimated using the ROS assay kit (CA1410; Solarbio) or the MDA assay kit (BC0025; Solarbio), respectively.29

Iron detection

The ferrous ion (Fe2+) in H9c2 cells was monitored using a Fe2+ assay kit (BC5415; Solarbio), following the manufacturer’s protocol. Briefly, supernatants of conditioned media from cells were placed onto 96-well plates and then incubated with a 5-microliter iron reducer at 25°C for 30 min to detect total iron content. Then, in a dark environment, samples were incubated with 100 μL of the iron probe for 60 min at 25°C. Finally, the OD was measured at 594 nm with a microplate reader (SpectraMax Mini; Molecular Devices Ltd.).29

Luciferase activity

The circANKIB1 or SLC7A11 fragments containing wild-type (WT) or mutant-type (MUT) binding sites with miR-452-5p were cloned into pmirGLO vectors (VT1439; YouBio, China). Four luciferase vectors (circANKIB1-WT, circANKIB1-MUT, SLC7A11-WT, and SLC7A11-MUT) were then obtained. The H9c2 cells (4×105 cells/well) were transfected with miR-452-5p mimic/NC (25 nM final concentration) and circANKIB1-WT/circANKIB1-MUT (2 µg) or SLC7A11-WT/SLC7A11-MUT (2 µg) using Lipofectamine 2000 Transfection Reagent (L7800; Solarbio). Then, the luciferase activity of Renilla (normalized as the control) and Firefly was tested with a Dual-Lucy Assay Kit (D0010; Solarbio) on a microplate reader (SpectraMax Mini; Molecular Devices Ltd.).23

Western blotting analysis

The total protein of H9c2 cells was collected using radioimmunoprecipitation assay (RIPA) lysis buffer (R0020; Solarbio) and isolated on a sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), and then transferred to polyvinylidene fluoride (PVDF) membranes (YA1701; Solarbio). The membranes were treated using Tris-buffered saline (TBS) containing 3% skim milk (T476445-10EA; Aladdin, Shanghai, China) at 4°C for 1 h and then incubated with the primary antibodies, including anti-GPX4 (1:1000, 14432-1-AP; Proteintech, Rosemont, USA), anti-SLC7A11 (1:1000, 26864-1-AP; Proteintech) and anti-β-actin (1:1000, ab8227; Abcam, Cambridge, UK). Finally, the membrane was treated with the secondary antibody (1:2000, goat anti-rabbit IgG antibody, bs-0295G; Bioss, Beijing, China) for 2 h and enhanced chemiluminescence (ECL) detection solutions (HR0340, E266188; Aladdin) to observe the protein bands.24

TUNEL assay

Cells were fixed in 4% paraformaldehyde for 30 min, followed by incubation with 0.3% Triton X-100 for 5 min. Then, the cells were incubated using terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) detection buffer (Beyotime Biotechnology) for 1 h. The 4’,6-diamidino-2-phenylindole (DAPI) was applied to counterstain the nuclei. The cells were observed under a fluorescence microscope (Leica DM1000; Leica Camera), and the TUNEL-positive cells were counted.30

Statistical analyses

The analysis was conducted using GraphPad Prism v. 8.0 software (GraphPad Software, San Diego, USA), and all data were presented as mean ± standard deviation (M ±SD). Student’s t-test or one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test were used to analyze differences between 2 groups or multiple groups, respectively. The normality of the data and the homogeneity of variance between groups were tested using the Shapiro–Wilk test and Levene’s test, respectively. The value of p > 0.05 indicated that the assumption of normality of data and homogeneity of variance were consistent, and further parameter testing could be performed. Statistical significance was set at p < 0.05.13 The results of the statistical analyses are presented in Table 2 and Table 3.

Results

Overexpression of circANKIB1 ameliorates hypoxia-induced cardiomyocyte injury and ferroptosis

The expression levels of HIF1α, which is an indicator of hypoxia,31 were detected in H9c2 cells following hypoxia treatment. The expression of HIF1α was significantly increased in H9c2 cells after hypoxia treatment for 24 h, 48 h and 72 h compared with normoxic H9c2 cells (Figure 1A). The results indicated that the hypoxia-induced cardiomyocyte model was successfully established. As hypoxia treatment over a long duration may result in irreversible damage to H9c2 cells, 24 h timepoint was selected for subsequent experiments (Figure 1A). We demonstrated that circANKIB1 expression was significantly decreased in a time-dependent manner under hypoxic conditions compared to normoxic H9c2 cells (Figure 1B). To reveal the properties of circANKIB1, we carried out an RNase R assay and discovered that circANKIB1 effectively blocked the degradation of RNase R, implying that circANKIB1 was more stable than liner ANKIB1 (Figure 1C). Furthermore, subcellular localization assays demonstrated that circANKIB1 was localized in the cytoplasm of H9c2 cardiomyocytes more than in the nucleus (Figure 1D). These data indicated that circANKIB1 was a stable circRNA that might be involved in the progression of hypoxia-induced cardiomyocyte injury.

Overexpression of circANKIB1 inhibited hypoxia-induced cardiomyocyte injury and ferroptosis

To investigate the effect of circANKIB1 on hypoxia-triggered cardiomyocyte injury, we performed overexpression experiments in H9c2 cells (Figure 2A). Cell viability was estimated using the MTT assay, which showed that hypoxia suppressed the viability of H9c2 cells, which was notably restored by circANKIB1 overexpression (Figure 2B). Interestingly, the level of iron was significantly reduced in H9c2 cells with circANKIB1 overexpression compared with control cells, which implied that circANKIB1 was likely to be associated with the progression of ferroptosis (Figure 2C). Notably, the accumulation of ROS and lipid peroxidation (MDA) is considered an important factor in ferroptosis.32 Therefore, the levels of ROS and MDA in H9c2 cells after the overexpression of circANKIB1 were detected. The data demonstrated that hypoxia dramatically increased the accumulation of ROS and MDA, which was effectively reversed by the overexpression of circANKIB1 (Figure 2D,E). The GPX4, an inhibitor of ferroptosis,33 was also investigated. After H/R treatment, GPX4 was decreased in H9c2 cells, whereas overexpressed circANKIB1 significantly upregulated GPX4 (Figure 2F). Taken together, these data suggest that the overexpression of circANKIB1 inhibited ferroptosis from alleviating hypoxia-induced cardiomyocyte injury.

miR-452-5p negatively interacts with circANKIB1 in hypoxic H9c2 cells

We used the starBase database (http://starbase.sysu.edu.cn/) to predict the target miRNA of circANKIB1. We found that miR-452-5p contains binding sequences of circANKIB1 (Figure 3A). The miR-452-5p mimics were obtained to upregulate miR-452-5p expression in H9c2 cells, and functional analyses were performed (Figure 3B). The luciferase assay showed that miR-452-5p mimics decreased the luciferase activity of circANKIB1-WT but not circANKIB1-MUT (Figure 3C). Additionally, the expression level of miR-452-5p was significantly upregulated in hypoxia-treated cells, compared with normoxic H9c2 cells (Figure 3D). Furthermore, to observe the regulatory influence of circANKIB1 on miR-452-5p expression, H9c2 cells were transfected with oe-circANKIB1 or a vector, followed by the examination of miR-452-5p expression using qPCR. The results demonstrated a decrease in miR-452-5p expression after the overexpression of circANKIB1 in H9c2 cells under both normoxic and hypoxic conditions (Figure 3E).

miR-452-5p mimic reversed the regulation of circANKIB1 on hypoxia-induced cardiomyocyte injury and ferroptosis

To verify whether miR-452-5p was associated with the protective effect of circANKIB1 in cardiomyocyte injury, we transfected a miR-452-5p mimic into H9c2 cells with circANKIB1 overexpression and performed a series of functional experiments. The MTT assay indicated that the miR-452-5p mimic effectively counteracted circANKIB1 overexpression regarding the viability of hypoxia-treated H9c2 cells (Figure 4A). Furthermore, the inhibitory effect of overexpressed circANKIB1 on iron, ROS and MDA accumulation was notably reversed by miR-452-5p mimics in hypoxia-induced H9c2 cells (Figure 4B–D). Moreover, the overexpression of circANKIB1 elevated the expression of GPX4 in hypoxia-induced H9c2 cells, and this effect was significantly reduced by the miR-452-5p mimic (Figure 4E). These data imply that circANKIB1 exerts its protective role in cardiomyocytes by mediating the expression of miR-452-5p.

SLC7A11 is a candidate target of miR-452-5p in hypoxia-treated H9c2 cells

A further analysis predicted that miR-452-5p would be able to bind to the 3’UTR of SLC7A11 (Figure 5A). Results collected from a dual-luciferase reporter gene assay suggested that the miR-452-5p mimic decreased luciferase activity in SLC7A11-WT, while the activity of SLC7A11-MUT was not affected (Figure 5B). Furthermore, the expression level of SLC7A11 was significantly decreased by hypoxia treatment in a time-dependent manner compared with normoxic H9c2 cells (Figure 5C). The qPCR and western blotting analyses showed that hypoxia reduced SLC7A11 mRNA and protein expression, which was effectively restored by a miR-452-5p inhibitor (Figure 5D,E). Moreover, to observe the regulatory influence of circANKIB1 on SLC7A11 expression, H9c2 cells were transfected with oe-circANKIB1 or vector, followed by an examination of SLC7A11 expression using western blot. The results highlighted that after the overexpression of circANKIB1 in H9c2 cells, the expression of SLC7A11 was elevated, and this could be effectively restored by the miR-452-5p inhibitor (Figure 5F).

Silencing SLC7A11 reversed the decreased cell apoptosis effect of circANKIB1 overexpression on hypoxia-induced cardiomyocyte injury

To confirm the role of SLC7A11 in hypoxia-treated cardiomyocyte injury, we constructed a vector with a SLC7A11 knockdown for rescue experiments (Figure 6A). Cell viability restoration through circANKIB1 overexpression was significantly reduced by SLC7A11 silencing in hypoxic-treated H9c2 cardiomyocytes (Figure 6B). Then, we explored the effect of SLC7A11 downregulation in H9c2 cells with various treatments using si-SLC7A11. The number of TUNEL-postitive cells with transfected with si-SLC7A11 was markedly increased when compared with circANKIB1 overexpression transfection in hypoxia-induced H9c2 cells (Figure 6C,D).

Downregulation of SLC7A11 reversed the decreased ferroptosis effect of circANKIB1 overexpression on hypoxia-induced H9c2 cardiomyocytes

The suppressive roles of overexpressed circANKIB1 on the accumulation of iron, ROS and MDA were effectively counteracted by SLC7A11 knockdown (Figure 7A–C). In addition, the decreased Fe2+ level in circANKIB1 overexpression in H9c2 cells was markedly elevated after the transfection with si-SLC7A11 (Figure 7D). Furthermore, the overexpression of circANKIB1 increased GPX4 expression in H9c2 cells under hypoxic conditions, which was notably reversed by SLC7A11 knockdown (Figure 7E). Collectively, our findings demonstrated that circANKIB1 protected cardiomyocytes from hypoxia-induced injury by mediating the miR-452-5p/SLC7A11 axis.

Discussion

Our research revealed that circANKIB1 was a stable circRNA that was mainly distributed in the cytoplasm and decreased in hypoxia-treated H9c2 cells. The overexpression of circANKIB1 significantly inhibited the progression of ferroptosis and protected H9c2 cardiomyocytes from hypoxia-induced injury. In addition, miR-452-5p is considered a direct target of circANKIB1, and a miR-452-5p mimic notably reversed the effect of circANKIB1 overexpression on ferroptosis and injury in cardiomyocytes induced by hypoxia. Moreover, SLC7A11 was a potential target for miR-452-5p, and SLC7A11 silencing abolished the protective role of circANKIB1 against hypoxia-induced cardiomyocyte injury.

Increasing evidence has shown that circRNAs exert vital roles in human CVDs.9, 34, 35 For example, circRbms1 is highly expressed in AMI, and the loss of circRbms1 effectively blocked H2O2-related cardiomyocyte apoptosis and ROS accumulation by mediating the miR-92a/BCL2L11 axis.36 Therefore, targeting circRNAs may be a promising approach for the treatment of AMI. In our previous study, we established a hypoxia model on H9c2 cardiomyocytes to simulate AMI in vitro, and tested the expression of several circRNAs associated with organ injury. The results suggested that hypoxia treatment increased the level of circANKIB1 in H9c2 cells. The circANKIB1 has been reported to promote Schwann cell proliferation, thereby exerting a protective role in peripheral nerve injury.15 However, whether circANKIB1 affects hypoxia-induced cardiomyocytes remains unclear. This work demonstrated that circANKIB1 was mainly localized in the cytoplasm of H9c2 cardiomyocytes and that the overexpression of circANKIB1 effectively restored the viability of H9c2 cardiomyocytes that was reduced by hypoxia exposure.

Ferroptosis is a new cell death pattern, different from apoptosis, autophagy and pyrodeath, which is induced by iron-dependent MDA,37 during which iron levels are notably elevated. An overload of ferrous ions causes ROS accumulation, which triggers the production of lipid peroxides such as MDA.38 Furthermore, GPX4 acts as an effective terminator of ferroptosis, capable of converting toxic lipids into non-toxic lipids.33 Accumulating evidence highlights that ferroptosis is tightly correlated to the occurrence of CVDs, including AMI.37 Interestingly, the present study revealed that the overexpression of circANKIB1 decreased the level of iron, ROS and MDA, but increased GPX4 expression in hypoxia-induced H9c2 cells. All these data indicated that circANKIB1 suppressed ferroptosis in our in vitro model.

The circRNAs are regarded as the sponge of miRNAs, regulating their functional role.39 Bioinformatics analysis illustrated that miR-452-5p was a potential miRNA target of circANKIB1, which was confirmed with luciferase activity that validated the interaction between circANKIB1 and miR-452-5p. The miR-452-5p is a vital miRNA that participates in various intracellular activities, including tumorigenesis40 and chronic contractile injury.41 The present study demonstrated a significant upregulation of miR-452-5p in H9c2 cardiomyocytes after hypoxia exposure. The overexpression of miR-452-5p effectively eliminated the effect of circANKIB1 on hypoxia-induced H9c2 cell injury and ferroptosis, indicating that circANKIB1 protected cardiomyocytes from hypoxia by sponging miR-452-5p.

Recent studies have confirmed that miRNAs can block the expression of target genes by binding to the 3’UTR of mRNAs.42 This study showed that SLC7A11 was a direct target of miR-452-5p. The SLC7A11 has previously been reported to inhibit oxidative stress and maintain GSH levels, thereby eliminating the occurrence of ferroptosis.43 Emerging evidence showed that SLC7A11 plays a positive role in protecting cardiomyocytes.43 Our study showed that hypoxia significantly reduced SLC7A11 expression in cardiomyocytes, which was notably counteracted by miR-452-5p inhibitors or circANKIB1 overexpression. Additionally, the SLC7A11 knockdown effectively eliminated the protective effect of circANKIB1 overexpression on cardiomyocyte injury and hypoxia-induced ferroptosis.

Limitations

There are some limitations to this study. First, there no animal experiments were conducted under the AMI model to verify whether circANKIB1 exerts cardioprotective effects in vivo. In addition, human and mouse cardiomyocytes need to be introduced to further validate these findings. In terms of data analysis, the Shapiro–Wilk test and Levene’s test have very low power at such a small sample size. In future studies, we will increase the sample size to ensure that the results of the statistical analysis are more accurate.

Conclusions

In conclusion, this study revealed that circANKIB1 alleviates hypoxia-induced cardiomyocyte injury and ferroptosis by modulating miR-452-5p/SLC7A11 signaling, providing a potential therapeutic target for patients with AMI.

Tables


Table 1. Quantitative real-time polymerase chain reaction (qPCR) primers

Genes

Primers (forward)

Primers (reverse)

circANKIB1

5’-AGACCGCAGACATGCTCC-3’

5’-AGTCCCTAATATCCTATTCATTCCA-3’

miR-452-5p

5’-GCGCAACTGTTTGCAGAG-3’

5’-GTGCAGGGTCCGAGGT-3’

SLC7A11

5’-GCTGACACTCGTGCTATT-3’

5’-ATTCTGGAGGTCTTTGGT-3’

U6

5’-CTCGCTTCGGCAGCACA-3’

5’-AACGCTTCACGAATTTGCGT-3’

GAPDH

5’-GGGAGCCAAAAGGGTCAT-3’

5’-GAGTCCTTCCACGATACCAA-3’

Table 2. The results of t-test and ANOVA

Figure

Method

F (DFn, DFd)

t

df

pSk

pL

Figure 1A

ANOVA

F (3,8) = 230.4

>0.1

0.6704

Figure 1B

ANOVA

F (3,8) = 123.3

>0.1

0.9921

Figure 1C

ANOVA

F (1,8) = 494.8

>0.1

0.8131

Figure 1D

ANOVA

F (2,12) = 939.9

>0.1

0.7471

Figure 2A

Student’s t-test

57.88

4

>0.1

0.9988

Figure 2B

ANOVA

F (3,8) = 148.2

>0.1

0.9846

Figure 2C

ANOVA

F (3,8) = 230.3

>0.1

0.9163

Figure 2D

ANOVA

F (3,8) = 358.5

>0.1

0.9535

Figure 2E

ANOVA

F (3,8) = 342.0

>0.1

0.9003

Figure 2F

ANOVA

F (3,8) = 131.7

>0.1

0.9024

Figure 3B

Student’s t-test

35.18

4

>0.1

0.9968

Figure 3C

ANOVA

F (1,8) = 168.1

>0.1

0.6943

Figure 3D

ANOVA

F (3,8) = 294.2

>0.1

0.7502

Figure 3E

ANOVA

F (3,8) = 164.1

>0.1

0.9279

Figure 4A

ANOVA

F (3,8) = 202.4

>0.1

0.7200

Figure 4B

ANOVA

F (3,8) = 427.5

>0.1

0.9888

Figure 4C

ANOVA

F (3,8) = 291.3

>0.1

0.9603

Figure 4D

ANOVA

F (3,8) = 240.9

>0.1

0.8582

Figure 4E

ANOVA

F (3,8) = 147.3

>0.1

0.9645

Figure 5B

ANOVA

F (1,8) = 197.4

>0.1

0.6893

Figure 5C

ANOVA

F (3,8) = 60.41

>0.1

0.9097

Figure 5D

ANOVA

F (3,8) = 57.63

>0.1

0.9688

Figure 5E

ANOVA

F (3,8) = 164.7

>0.1

0.9932

Figure 5F

ANOVA

F (3,8) = 215.5

>0.1

0.8702

Figure 6A

Student’s t-test

13.63

4

>0.1

0.8026

Figure 6B

ANOVA

F (3,8) = 234.7

>0.1

0.8439

Figure 6D

ANOVA

F (3,8) = 182.8

>0.1

0.8641

Figure 7B

ANOVA

F (3,8) = 242.3

>0.1

0.6751

Figure 7C

ANOVA

F (3,8) = 87.08

>0.1

0.4217

Figure 7D

ANOVA

F (3,8) = 126.6

>0.1

0.3352

Figure 7E

ANOVA

F (3,8) = 91.43

>0.1

0.9008

ANOVA – analysis of variance; df – degrees of freedom; t – Student’s t-test results; pSK – Shapiro–Wilk test; pL – Levene’s test; DFd – degrees of freedom denominator; DFn – degrees of freedom numerator.
Table 3. Tukey’s post hoc test results of ANOVA

Figure

Method

pS

pT1

pT2

pT3

Figure 1A

ANOVA

0.0001

0.0002

0.0001

0.0001

Figure 1B

ANOVA

0.0001

0.0045

0.0001

0.0001

Figure 1C

ANOVA

0.0001

0.0665

0.0001

Figure 1D

ANOVA

0.0001

0.0001

0.0001

0.0001

Figure 2B

ANOVA

0.0001

0.0001

0.0001

Figure 2C

ANOVA

0.0001

0.0001

0.0001

Figure 2D

ANOVA

0.0001

0.0001

0.0001

Figure 2E

ANOVA

0.0001

0.0001

0.0001

Figure 2F

ANOVA

0.0001

0.0001

0.0001

Figure 3C

ANOVA

0.0001

0.9425

0.0001

Figure 3D

ANOVA

0.0001

0.0011

0.0001

0.0001

Figure 3E

ANOVA

0.0001

0.0001

0.0001

Figure 4A

ANOVA

0.0001

0.0001

0.0001

0.0001

Figure 4B

ANOVA

0.0001

0.0001

0.0001

0.0001

Figure 4C

ANOVA

0.0001

0.0001

0.0001

0.0001

Figure 4D

ANOVA

0.0001

0.0001

0.0001

0.0001

Figure 4E

ANOVA

0.0001

0.0001

0.0001

0.0001

Figure 5B

ANOVA

0.0001

0.0001

0.7740

Figure 5C

ANOVA

0.0001

0.0193

0.0002

0.0001

Figure 5D

ANOVA

0.0001

0.0001

0.0001

Figure 5E

ANOVA

0.0001

0.0001

0.0001

Figure 5F

ANOVA

0.0001

0.0001

0.0001

Figure 6B

ANOVA

0.0001

0.0001

0.0001

0.0001

Figure 6D

ANOVA

0.0001

0.0001

0.0001

0.0001

Figure 7B

ANOVA

0.0001

0.0001

0.0001

0.0001

Figure 7C

ANOVA

0.0001

0.0001

0.0001

0.0002

Figure 7D

ANOVA

0.0001

0.0001

0.0001

0.0001

Figure 7E

ANOVA

0.0001

0.0001

0.0001

0.0001

ANOVA – analysis of variance; pS – p-value summary of ANOVA; PT – Tukey’s post hoc test results. The p-values of the comparison results between the different groups were divided into PT1/2/3, respectively.

Figures


Fig. 1. The overexpression of circANKIB1 ameliorates hypoxia-induced cardiomyocyte injury and ferroptosis. A. Hypoxia-inducible factor (HIF)1α mRNA expression level in H9c2 cells was increased after hypoxia treatment for 24 h, 48 h or 72 h tested with quantitative real-time polymerase chain reaction (qPCR) assay (p = 0.001, n = 3, analysis of variance (ANOVA) test); B. CircANKIB1 expression level was decreased in H9c2 cells after hypoxia treatment for 0 h, 24 h, 48 h, or 72 h (p = 0.001, n = 3, ANOVA test); C. RNase R assay and qPCR assay were used to test the expression level of circANKIB1 or liner and it was found that there was no significant change in circANKIB1 expression, while liner ANKIB1 expression was decreased (p = 0.001, n = 3, ANOVA test); D. The distribution of circANKIB1 was confirmed using subcellular localization assay (p = 0.001, n = 3, ANOVA test)
Fig. 2. The overexpression of circANKIB1 improved hypoxia-induced cardiomyocyte injury and ferroptosis. A. The expression level of circANKIB1 was increased and tested with quantitative real-time polymerase chain reaction (qPCR) assay (p = 0.001, n = 3, Student’s t-test); B. Methyl thiazolyl tetrazolium (MTT) assay monitored that cell viability was suppressed by hypoxia and could be restored by circANKIB1 overexpression (p = 0.001, n = 3, analysis of variance (ANOVA) test); C–E. The reduced expression of ferrous ion (Fe2+) (C), reactive oxygen species (ROS) (D) and malondialdehyde (MDA) (E) was analyzed using corresponding kits (p = 0.001, n = 3, ANOVA test); F. The increased GPX4 expression was examined with western blotting analysis (n = 3, ANOVA test)
Fig. 3. MiR-452-5p negatively interacts with circANKIB1 in hypoxic H9c2 cells. A. The binding sequences between circANKIB1 and miR-452-5p were forecast using the starBase database; B. The increased miR-452-5p expression was tested with quantitative real-time polymerase chain reaction (qPCR) assay (p = 0.001, n = 3, Student’s t-test); C. The interaction between circANKIB1 and miR-452-5p was verified using luciferase activity assay (p = 0.001, n = 3, analysis of variance (ANOVA) test); D. MiR-452-5p expression was increased in H9c2 cells after hypoxia treatment for 24 h, 48 h or 72 h (p = 0.001, n = 3, ANOVA test); E. MiR-452-5p expression levels in hypoxia-treated H9c2 cells transfected with oe-circANKIB1 were decreased and tested with qPCR assay (p = 0.001, n = 3, ANOVA test)
Fig. 4. MiR-452-5p mimic reversed the regulation of circANKIB1 on hypoxia-induced cardiomyocyte injury and ferroptosis. A. Methyl thiazolyl tetrazolium (MTT) assay confirmed that cell viability induced in H9c2 cells by co-transfection with circANKIB1 overexpression was effectively counteracted by miR-452-5p mimic (p = 0.001, n = 3, analysis of variance (ANOVA) test); B–D. Corresponding kits showed that the reduced levels of ferrous ion (Fe2+) (B), reactive oxygen species (ROS) (C) and malondialdehyde (MDA) (D) were effectively counteracted by miR-452-5p mimic (p = 0.001, n = 3, ANOVA test); E. Western blotting analysis showed that GPX4 expression elevated by the overexpression of circANKIB1 was eliminated by the miR-452-5p mimic (p = 0.001, n = 3, ANOVA test)
Fig. 5. SLC7A11 is a candidate target of miR-452-5p in hypoxia-treated H9c2 cells. A. The binding sites between miR-452-5p and SLC7A11 were forecast via the starBase database; B. The interaction between miR-452-5p and SLC7A11 was verified using luciferase activity assay (p = 0.001, n = 3, analysis of variance (ANOVA) test); C. Decreased SLC7A11 expression level in H9c2 cells after hypoxia treatment for 24 h, 48 h or 72 h was tested with quantitative real-time polymerase chain reaction (qPCR) assay (p = 0.001, n = 3, ANOVA test); D,E. Western blotting analysis and qPCR showed that hypoxia reduced SLC7A11 mRNA and protein expression, which was effectively restored by miR-452-5p inhibitor (p = 0.001, n = 3, ANOVA test); F. The increased protein level of SLC7A11 expression in hypoxia-treated H9c2 cells transfected with circANKIB1 overexpression was effectively restored by the miR-452-5p inhibitor and tested using western blotting analysis (p = 0.001, n = 3, ANOVA test)
Fig. 6. SLC7A11 silencing reversed the decreased apoptotic effects of circANKIB1 overexpression on hypoxia-induced cardiomyocyte injury. A. The decreased protein level of SLC7A11 was estimated via western blotting analysis (p = 0.001, n = 3, Student’s t-test); B. Methyl thiazolyl tetrazolium (MTT) assay revealed that cell viability restored through circANKIB1 overexpression was significantly reduced by SLC7A11 silencing (p = 0.001, n = 3, analysis of variance (ANOVA) test); C,D. Terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay indicated that cell apoptosis was markedly increased when compared with circANKIB1 overexpression transfection in hypoxia-induced H9c2 cells (p = 0.001, n = 3, ANOVA test)
Fig. 7. Downregulation of SLC7A11 reversed the decreased ferroptosis effect of circANKIB1 overexpression on hypoxia-induced H9c2 cardiomyocytes.
A–D. The levels of reactive oxygen species (ROS) (A,B), malondialdehyde (MDA) (C) and ferrous ion (Fe2+) (D) were markedly decreased after transfection with si-SLC7A11 when tested using corresponding kits (p = 0.001, n = 3, analysis of variance (ANOVA) test); E. GPX4 expression increased by circANKIB1 overexpression and was effectively counteracted by SLC7A11 knockdown, as analyzed using western blotting analysis (p = 0.001, n = 3, ANOVA test)

References (43)

  1. Benjamin EJ, Virani SS, Callaway CW, et al. Heart Disease and Stroke Statistics – 2018 update: A report from the American Heart Association. Circulation. 2018;137(12):e67–e492. doi:10.1161/CIR.0000000000000558
  2. Anderson JL, Morrow DA. Acute myocardial infarction. N Engl J Med. 2017;376(21):2053–2064. doi:10.1056/NEJMra1606915
  3. Tang Q, Li MY, Su YF, et al. Absence of miR-223-3p ameliorates hypoxia-induced injury through repressing cardiomyocyte apoptosis and oxidative stress by targeting KLF15. Eur J Pharmacol. 2018;841:67–74. doi:10.1016/j.ejphar.2018.10.014
  4. Huang L, Guo B, Liu S, Miao C, Li Y. Inhibition of the lncRNA Gpr19 attenuates ischemia–reperfusion injury after acute myocardial infarction by inhibiting apoptosis and oxidative stress via the miR-324-5p/Mtfr1 axis. IUBMB Life. 2020;72(3):373–383. doi:10.1002/iub.2187
  5. Ebbesen KK, Kjems J, Hansen TB. Circular RNAs: Identification, biogenesis and function. Biochim Biophys Acta. 2016;1859(1):163–168. doi:10.1016/j.bbagrm.2015.07.007
  6. Li X, Yang L, Chen LL. The biogenesis, functions, and challenges of circular RNAs. Mol Cell. 2018;71(3):428–442. doi:10.1016/j.molcel.2018.06.034
  7. Patop IL, Wüst S, Kadener S. Past, present, and future of circRNAs. EMBO J. 2019;38(16):e100836. doi:10.15252/embj.2018100836
  8. Gao JL, Chen G, He HQ, Wang J. CircRNA as a new field in human disease research [in Chinese]. Zhongguo Zhong Yao Za Zhi. 2018;43(3):457–462. doi:10.19540/j.cnki.cjcmm.20171106.012
  9. Altesha M, Ni T, Khan A, Liu K, Zheng X. Circular RNA in cardiovascular disease. J Cell Physiol. 2019;234(5):5588–5600. doi:10.1002/jcp.27384
  10. Wang Y, Liu B. Circular RNA in diseased heart. Cells. 2020;9(5):1240. doi:10.3390/cells9051240
  11. Ren K, Li B, Jiang L, et al. Circ_0023461 silencing protects cardiomyocytes from hypoxia-induced dysfunction through targeting Mir-370-3p/Pde4d signaling. Oxid Med Cell Longev. 2021;2021:8379962. doi:10.1155/2021/8379962
  12. Wang S, Cheng Z, Chen X, Lu G, Zhu X, Xu G. CircUBXN7 mitigates H/R-induced cell apoptosis and inflammatory response through the miR-622-MCL1 axis. Am J Transl Res. 2021;13(8):8711–8727. PMID:34539989.
  13. Tang J, Duan G, Wang Y, Wang B, Li W, Zhu Z. Circular RNA_ANKIB1 accelerates chemo-resistance of osteosarcoma via binding microRNA-26b-5p and modulating enhancer of zeste homolog 2. Bioengineered. 2022;13(3):7351–7366. doi:10.1080/21655979.2022.2037869
  14. Du YX, Guo LX, Pan HS, Liang YM, Li X. Circ_ANKIB1 stabilizes the regulation of miR-19b on SOCS3/STAT3 pathway to promote osteosarcoma cell growth and invasion. Hum Cell. 2020;33(1):252–260. doi:10.1007/s13577-019-00298-6
  15. Mao S, Zhang S, Zhou S, et al. A Schwann cell-enriched circular RNA circ-Ankib1 regulates Schwann cell proliferation following peripheral nerve injury. FASEB J. 2019;33(11):12409–12424. doi:10.1096/fj.201900965R
  16. Mirahmadi Y, Nabavi R, Taheri F, et al. MicroRNAs as biomarkers for early diagnosis, prognosis, and therapeutic targeting of ovarian cancer. J Oncol. 2021;2021:3408937. doi:10.1155/2021/3408937
  17. Ma X, Liu C, Gao C, et al. circRNA-associated ceRNA network construction reveals the circRNAs involved in the progression and prognosis of breast cancer. J Cell Physiol. 2020;235(4):3973–3983. doi:10.1002/jcp.29291
  18. Kobayashi M, Suhara T, Baba Y, Kawasaki NK, Higa JK, Matsui T. Pathological roles of iron in cardiovascular disease. Curr Drug Targets. 2018;19(9):1068–1076. doi:10.2174/1389450119666180605112235
  19. Fang X, Wang H, Han D, et al. Ferroptosis as a target for protection against cardiomyopathy. Proc Natl Acad Sci U S A. 2019;116(7):2672–2680. doi:10.1073/pnas.1821022116
  20. Stockwell BR, Friedmann Angeli JP, Bayir H, et al. Ferroptosis: A regulated cell death nexus linking metabolism, redox biology, and disease. Cell. 2017;171(2):273–285. doi:10.1016/j.cell.2017.09.021
  21. Zheng J, Conrad M. The metabolic underpinnings of ferroptosis. Cell Metab. 2020;32(6):920–937. doi:10.1016/j.cmet.2020.10.011
  22. Koppula P, Zhuang L, Gan B. Cystine transporter SLC7A11/xCT in cancer: Ferroptosis, nutrient dependency, and cancer therapy. Protein Cell. 2021;12(8):599–620. doi:10.1007/s13238-020-00789-5
  23. Zhou Y, Li X, Zhao D, Li X, Dai J. Long non coding RNA MEG3 knockdown alleviates hypoxia induced injury in rat cardiomyocytes via the miR 325 3p/TRPV4 axis. Mol Med Rep. 2021;23(1):18. doi:10.3892/mmr.2020.11656
  24. Tan J, Pan W, Chen H, et al. Circ_0124644 serves as a ceRNA for Mir-590-3p to promote hypoxia-induced cardiomyocytes injury via regulating Sox4. Front Genet. 2021;12:667724. doi:10.3389/fgene.2021.667724
  25. Bustin SA, Benes V, Garson JA, et al. The MIQE guidelines: Minimum information for publication of quantitative real-time PCR experiments. Clin Chem. 2009;55(4):611–622. doi:10.1373/clinchem.2008.112797
  26. Liu B, Guo K. CircRbms1 knockdown alleviates hypoxia-induced cardiomyocyte injury via regulating the miR-742-3p/FOXO1 axis. Cell Mol Biol Lett. 2022;27(1):31. doi:10.1186/s11658-022-00330-y
  27. Zhang Y, Li Z, Wang J, Chen H, He R, Wu H. CircTRRAP knockdown has cardioprotective function in cardiomyocytes via the signal regulation of mir-370-3p/PAWR axis. Cardiovasc Ther. 2022;2022:7125602. doi:10.1155/2022/7125602
  28. Lan Z, Wang T, Zhang L, Jiang Z, Zou X. CircSLC8A1 exacerbates hypoxia-induced myocardial injury via interacting with Mir-214-5p to upregulate TEAD1 expression. Int Heart J. 2022;63(3):591–601. doi:10.1536/ihj.21-547
  29. Dai R, Yang X, He W, Su Q, Deng X, Li J. LncRNA AC005332.7 Inhibited ferroptosis to alleviate acute myocardial infarction through regulating mir-331-3p/CCND2 axis. Korean Circ J. 2023;53(3):151–167. doi:10.4070/kcj.2022.0242
  30. Zhang Z, Yang W, Ma F, et al. Enhancing the chemotherapy effect of Apatinib on gastric cancer by co-treating with salidroside to reprogram the tumor hypoxia micro-environment and induce cell apoptosis. Drug Deliv. 2020;27(1):691–702. doi:10.1080/10717544.2020.1754528
  31. Kietzmann T, Samoylenko A, Roth U, Jungermann K. Hypoxia-inducible factor-1 and hypoxia response elements mediate the induction of plasminogen activator inhibitor-1 gene expression by insulin in primary rat hepatocytes. Blood. 2003;101(3):907–914. doi:10.1182/blood-2002-06-1693
  32. Ma S, Sun L, Wu W, Wu J, Sun Z, Ren J. USP22 protects against myocardial ischemia–reperfusion injury via the SIRT1-p53/SLC7A11-dependent inhibition of ferroptosis-induced cardiomyocyte death. Front Physiol. 2020;11:551318. doi:10.3389/fphys.2020.551318
  33. Yang WS, SriRamaratnam R, Welsch ME, et al. Regulation of ferroptotic cancer cell death by GPX4. Cell. 2014;156(1–2):317–331. doi:10.1016/j.cell.2013.12.010
  34. Gong X, Wu G, Zeng C. Role of circular RNAs in cardiovascular diseases. Exp Biol Med (Maywood). 2019;244(2):73–82. doi:10.1177/1535370218822988
  35. Lin F, Zhao G, Chen Z, et al. circRNA miRNA association for coronary heart disease. Mol Med Report. 2019;19(4):2527–2536. doi:10.3892/mmr.2019.9905
  36. Jin L, Zhang Y, Jiang Y, Tan M, Liu C. Circular RNA Rbms1 inhibited the development of myocardial ischemia–reperfusion injury by regulating miR-92a/BCL2L11 signaling pathway. Bioengineered. 2022;13(2):3082–3092. doi:10.1080/21655979.2022.2025696
  37. Li M, Wang ZW, Fang LJ, Cheng SQ, Wang X, Liu NF. Programmed cell death in atherosclerosis and vascular calcification. Cell Death Dis. 2022;13(5):467. doi:10.1038/s41419-022-04923-5
  38. Wang M, Liu CY, Wang T, et al. (+)−Clausenamide protects against drug-induced liver injury by inhibiting hepatocyte ferroptosis. Cell Death Dis. 2020;11(9):781. doi:10.1038/s41419-020-02961-5
  39. Kulcheski FR, Christoff AP, Margis R. Circular RNAs are miRNA sponges and can be used as a new class of biomarker. J Biotechnol. 2016;238:42–51. doi:10.1016/j.jbiotec.2016.09.011
  40. Lin X, Han L, Gu C, et al. MiR-452-5p promotes colorectal cancer progression by regulating an ERK/MAPK positive feedback loop. Aging (Albany NY). 2021;13(5):7608–7626. doi:10.18632/aging.202657
  41. Tian Y, Sun L, Qi T. Long noncoding RNA GAS5 ameliorates chronic constriction injury induced neuropathic pain in rats by modulation of the miR-452-5p/CELF2 axis. Can J Physiol Pharmacol. 2020;98(12):870–877. doi:10.1139/cjpp-2020-0036
  42. Fabian MR, Sonenberg N, Filipowicz W. Regulation of mRNA translation and stability by microRNAs. Annu Rev Biochem. 2010;79:351–379. doi:10.1146/annurev-biochem-060308-103103
  43. Li Y, Yan J, Zhao Q, Zhang Y, Zhang Y. ATF3 promotes ferroptosis in sorafenib-induced cardiotoxicity by suppressing Slc7a11 expression. Front Pharmacol. 2022;13:904314. doi:10.3389/fphar.2022.90431