To the Editor: Myocardial ischemia/reperfusion injury (MI/RI) often causes death, dysfunction, or damage to the cardiomyocytes, and thus more effective clinical treatments are required.[1] MicroRNAs (miRNAs) bind to target mRNAs to regulate post-transcriptional gene expression, with important effects on MI/RI by regulating critical molecular signaling pathways.[2] Previous studies have shown that flavonoids have beneficial effects on cardiac function and the outcomes after cardiac operations. Luteolin is a type of flavonoid, and high dietary intake of luteolin-rich food decreases the myocardial infarction area.[3] However, the transcriptional mechanism that allows luteolin to modulate gene expression and protect against MI/RI remains unknown. In this study, we screened the miRNAs with the greatest effects in MI/RI models after luteolin pretreatment. We identified the target gene of the miRNAs and explored the mechanisms associated with the protective effects of luteolin against MI/RI both in vitro and in vivo. All the experiments were performed in accordance with guidelines and approved by the Animal Ethics Committee of the Xuzhou Medical University (No. CMCACUC2014-04-027).

Luteolin pretreatment can protect against MI/RI. The luteolin concentrations were 7.5 μmol/L for neonatal rat cardiomyocytes (NRCs) and 16.0 μmol/L for H9c2 cells according to detection by the methyl thiazoleterazolium assay [Supplementary Figure 1A, https://links.lww.com/CM9/B188]. Lactate dehydrogenase (LDH) detection and apoptosis rate assessments using the Terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay showed that luteolin pretreatment decreased the LDH levels and inhibited cell apoptosis (all P < 0.01; Supplementary Figure 1B, C, https://links.lww.com/CM9/B188). Furthermore, cleaved caspase-3 was activated after simulated I/R injury, but activation was decreased by luteolin pretreatment (all P < 0.01; Supplementary Figure 1D, https://links.lww.com/CM9/B188).

MiRNAs are potential therapeutic targets in various diseases.[4] Eight miRNAs (miRNA-1, -21, -23a, -24, -133a, -199a-3p, -214, and -378) were searched for in MEDLINE, PubMed Central, and the Human microRNA Disease Database. The reverse transcription polymerase chain reaction (RT-PCR) results showed that miRNA-1, -23a, -24, -133a, and -378 were upregulated in perfused rat hearts after MI/RI, but that luteolin pretreatment downregulated miRNA-23a and upregulated miR-378. MiRNA-1, -21, -23a, and -24 were upregulated in the simulate I/R (sI/R) model in NRCs, but luteolin pretreatment only downregulated miRNA-23a. Thus, miRNA-23a was selected for further investigation [Supplementary Figure 2, https://links.lww.com/CM9/B188].

Our in vitro experiments showed that miRNA-23a was upregulated after sI/R, but downregulated by luteolin pretreatment (all P < 0.01; Supplementary Figure 1E, https://links.lww.com/CM9/B188). The miRNA-23a mimic aggravated sI/R injury and activated the cleavage of caspase-3, whereas the miRNA-23a inhibitor had the opposite effect (all P < 0.01; Supplementary Figure 1F, G, https://links.lww.com/CM9/B188). Overexpression of miRNA-23a abolished the protective effect of luteolin and the inactivation effect on cleaved caspase-3 (all P > 0.05; Supplementary Figure 1H, I, https://links.lww.com/CM9/B188).

Tissue factor pathway inhibitor 2 (TFPI2) acts as a tumor suppressor and modulates cell growth, apoptosis, and invasion.[5] TFPI2 is upregulated by ischemic preconditioning in rats, thereby suggesting that it may alleviate apoptosis. Using TargetScan, PicTar, and miRDB, we found that Tfpi2 is a putative target gene of miRNA-23a. Real-time quantitative PCR (qPCR) and Western blotting showed that the miRNA-23a mimic did not affect the mRNA level of Tfpi2 but decreased its protein expression level, whereas the miRNA-23a inhibitor had the opposite effect [Supplementary Figure 3A, B, https://links.lww.com/CM9/B188]. Luciferase reporter assays showed that overexpression of miRNA-23a decreased the luciferase activity (P < 0.001), whereas knockdown of miRNA-23a increased the luciferase activity (P < 0.01). Moreover, induced mutation of the predicted binding site of miRNA-23a abolished these effects (all P > 0.05; Supplementary Figure 3C, https://links.lww.com/CM9/B188).

Western blotting analysis for the luteolin group showed that TFPI2 was upregulated in both NRCs and H9c2 cells (all P < 0.001, vs. sI/R group; Supplementary Figure 1J, https://links.lww.com/CM9/B188). Knockdown of TFPI2 with siRNA partly abolished the protective effects of luteolin [Supplementary Figure 1K,L, https://links.lww.com/CM9/B188]. After pretreatment with the miRNA-23a mimic, the TFPI2 protein and cleaved caspase-3 levels were almost the same with or without luteolin pretreatment (P > 0.05; Supplementary Figure 1M, https://links.lww.com/CM9/B188). Overall, these results indicate that miRNA-23a is involved in apoptosis during MI/RI, and that Tfpi2 is a target gene of miRNA-23a.

We also used adenovirus containing miR-23a or Tfpi2 to explore the effects on apoptosis, infarct size, and left ventricular function in MI/RI models after luteolin pretreatment in vivo. The TFPI2 protein levels decreased in the MI/RI group (P < 0.001, vs. sham group) but increased in the luteolin + MI/RI group (P < 0.05, vs. MI/RI group). Compared with the MI/RI group, the TFPI2 protein levels decreased in the Ad-miR-23a + MI/RI, Ad-miR-23a + MI/RI + luteolin Ad-Tfpi2 shRNA + MI/RI, and Ad-Tfpi2 shRNA + MI/RI + luteolin groups (all P < 0.05), but increased in the Ad-miR-23a antago + MI/RI and Ad-miR-23a antago + MI/RI + luteolin groups (all P < 0.05; Supplementary Figure 4A, https://links.lww.com/CM9/B188).

Western blotting analysis showed that Bax expression was upregulated in the Ad-miR-23a + MI/RI group whereas Bcl-2 expression was downregulated (P < 0.001 and P < 0.05, vs. the MI/RI group, respectively), but the opposite trend was observed in the Ad-miR-23a antago + MI/RI group. Compared with the MI/RI group, Bax and cleaved caspase-3 expression were upregulated in the Ad-Tfpi2 shRNA + MI/RI group, whereas Bcl-2 and caspase-3 expression were downregulated (all P < 0.05; Supplementary Figure 4B, https://links.lww.com/CM9/B188).

MI/RI increased the myocardial infarct size (P < 0.001) but it decreased under luteolin pretreatment (P < 0.001, vs. MI/RI group). The myocardial infarct sizes were larger in the Ad-miR-23a + MI/RI and Ad-Tfpi2 shRNA + MI/RI groups (P < 0.01 and P < 0.05, respectively), but smaller in the Ad-miR-23a antago + MI/RI group (P < 0.001, vs. MI/RI group; Supplementary Figure 1N, https://links.lww.com/CM9/B188).

The hemodynamic indices comprising left ventricular systolic pressure, +dp/pt, and –dp/pt decreased after MI/RI injury, but these indices improved under luteolin pretreatment (all P < 0.05, vs. MI/RI group). Compared with the MI/RI group, these indices decreased in the Ad-miRNA-23a + MI/RI and Ad-Tfpi2 shRNA + MI/RI groups, but increased in the Ad-miRNA-23a antago + MI/RI group (all P < 0.05). The changes in left ventricular end-diastolic pressure exhibited the opposite trend [Supplementary Figure 4C, https://links.lww.com/CM9/B188].

In conclusion, this study identified the molecular mechanism responsible for the association between miRNA-23a and MI/RI. Luteolin pretreatment protected against MI/RI via miRNA-23a-mediated upregulation of TFPI2 both in vitro and in vivo. These findings further elucidate the mechanisms associated with cardiac I/R injury and the effective protection provided by luteolin.

Funding

This study was supported by grants from the National Natural Science Foundation of China (No. 81570326), Natural Science Foundation of Jiangsu Province (Nos. BK20141139, BK20190988), Jiangsu Commission of Health Foundation (No. H2018005), and Key Research and Development program of Xuzhou (No. KC20097).

Conflicts of interest

None.

References 1. Heusch G. Myocardial ischaemia-reperfusion injury and cardioprotection in perspective. Nat Rev Cardiol 2020; 17:773–789. doi: 10.1038/s41569-020-0403-y. 2. Carbonell T, Gomes AV. MicroRNAs in the regulation of cellular redox status and its implications in myocardial ischemia-reperfusion injury. Redox biol 2020; 36:101607doi: 10.1016/j.redox.2020.101607. 3. Luo Y, Shang P, Li D. Luteolin: a flavonoid that has multiple cardio-protective effects and its molecular mechanisms. Front Pharmacol 2017; 8:692doi: 10.3389/fphar.2017.00692. 4. Shi XJ, Jin Y, Xu WM, Shen Q, Li J, Chen K. MicroRNA-23a reduces lipopolysaccharide-induced cellular apoptosis and inflammatory cytokine production through Rho-associated kinase 1/sirtuin-1/nuclear factor-kappa B crosstalk. Chin Med J 2021; 134:829–839. doi: 10.1097/cm9.0000000000001369. 5. Glöckner SC, Dhir M, Yi JM, McGarvey KE, Van Neste L, Louwagie J, et al. Methylation of TFPI2 in stool DNA: a potential novel biomarker for the detection of colorectal cancer. Cancer Res 2009; 69:4691–4699. doi: 10.1158/0008-5472.Can-08-0142.