LncRNA FAF attenuates hypoxia/ischaemia‐induced pyroptosis via the miR‐185‐5p/PAK2 axis in cardiomyocytes

Abstract Pyroptosis is associated with various cardiovascular diseases. Increasing evidence suggests that long noncoding RNAs (lncRNAs) have been implicated in gene regulation, but how lncRNAs participate in the regulation of pyroptosis in the heart remains largely unknown. In this study, we aimed to explore the antipyroptotic effects of lncRNA FGF9‐associated factor (FAF) in acute myocardial infarction (AMI). The expression patterns of lncRNA FAF, miR‐185‐5p and P21 activated kinase 2 (PAK2) were detected in hypoxia/ischaemia‐induced cardiomyocytes. Hoechst 33342/PI staining, lactate dehydrogenase (LDH) release assay, immunofluorescence and Western blotting were conducted to assay cell pyroptosis. The interaction between lncRNA FAF, miR‐185‐5p and PAK2 was verified by bioinformatics analysis, small RNA sequencing luciferase reporter assay and qRT‐PCR. The expression of LncRNA FAF was downregulated in hypoxic cardiomyocytes and myocardial tissues. Overexpression of lncRNA FAF could attenuate cardiomyocyte pyroptosis, improve cell viability and reduce infarct size during the procession of AMI. Moreover, lncRNA FAF was confirmed as a sponge of miR‐185‐5p and promoted PAK2 expression in cardiomyocytes. Collectively, our findings reveal a novel lncRNA FAF/miR‐185‐5p/PAK2 axis as a crucial regulator in cardiomyocyte pyroptosis, which might be a potential therapeutic target of AMI.

burden. 5,6 AMI mainly results from sudden and sustained occlusion of the coronary artery and leads to the deprivation of oxygen and nutrients within the ischaemic zone. 7 As a result, various types of cell death, such as necrosis, apoptosis, autophagy and pyroptosis, can occur in infarcted myocardium, respectively. 8 Among them, pyroptosis, a highly inflammatory form of programmed cell death, is characterized by cell lysis and the release of inflammatory cytokines. 9 Pyroptosis is initially discovered as an innate immune response to certain pathogen infection in a caspase-1-dependent manner. 10 Inflammasomes, molecular platforms that function as cytosolic sensors, recognize pathogen-associated molecular patterns (PAMPs) as well as danger-associated molecular patterns (DAMPs) and further assemble themselves into inflammasome complexes to trigger pyroptosis via the cleavage of caspase-1. Subsequently, activated caspase-1 hydrolyses gasdermin D (GSDMD) to its N-terminal fragment and converts pro-inflammatory cytokine (pro-IL-18 or pro-IL-1β) to its mature form. Once cleaved, the N-terminus of GSDMD functions as a pore-forming protein thus promoting cell swelling, inflammatory cytokine release, membrane rupture and ultimately pyroptotic cell death. 11,12 NACHT, LRR and PYD domains-containing protein 3 (NLRP3), one of the most studied inflammasomes in the heart, senses multiple host-derived danger signals, triggers sterile inflammatory responses and induces pyroptotic cell death during AMI.
Excessive activation of pyroptosis contributes to the continual loss of cardiomyocytes, increased infarct size, adverse cardiac remodelling and ventricular aneurysm formation following AMI. [13][14][15][16] Therefore, therapies that target the pyroptosis pathway might be valuable strategies for acute myocardial infarction.
Long noncoding RNAs (LncRNAs), over 200 nucleotides in length, generally represent a class of RNA transcripts without protein-coding capability. LncRNAs were first described as transcriptional noise and polymerase II-transcribed byproducts with no biological function. 17 LncRNAs can regulate gene expression by interacting with mRNAs, microRNAs (miRNAs) and RNA-binding proteins. 18,19 Numerous studies have revealed that lncRNAs participate in multiple cardiovascular diseases (CVDs). For instance, our prior study has found that lncRNA Kcan2-AS promotes ventricular arrhythmias by downregulating Kcan2 expression during heart failure. 20 LncRNA CPR has been reported to be involved in cardiomyocyte proliferation and cardiac repair after myocardial injury. 21 LncRNA FGF9-associated factor (FAF), a recently discovered ln-cRNA, is downregulated in infarcted myocardium and inhibits cardiomyocyte apoptosis and cardiac fibrosis through transcriptional regulation of FGF9. 22,23 Nonetheless, the relationship between ln-cRNA FAF and cardiomyocyte pyroptosis is still unclear.
miRNAs are involved in various pathologic events including CVDs. 24 Several miRNAs have been reported to modulate pyroptosis directly or indirectly in CVDs, such as miR-223, miR-1, miR-30d and miR-214. 25 MiR-185-5p was reported to function as a tumour suppressor by inhibiting proliferation and inducing apoptosis in cancers. [26][27][28] Recently, one study has revealed that downregulation of miR-185-5p expression contributes to neovascularization in endothelial cells, which further mitigates the decline in cardiac function in postinfarction mice. 29 Previous studies have suggested that lncRNAs could bind to miRNAs and act as competitive endogenous RNAs (ceRNAs) to regulate cell biological processes, which play a vital role in myocardial infarction. 30,31 However, the exact regulatory mechanism of miR-185-5p remains poorly understood in cardiomyocytes. In the present study, we investigated the potential of lncRNA FAF against cardiomyocyte pyroptosis and AMI.

| Cell isolation, culture and establishment of hypoxia/ischaemia
Neonatal rat cardiomyocytes (NRCM) were isolated from 1-to 3-day-old Sprague-Dawley rats through enzyme digestion as previously described. 32 After isolation, NRCM were transferred to a cell incubator and cultured in DMEM supplemented with 10% HS, 5% FBS and 1% penicillin-streptomycin for 24 hours. For the establishment of hypoxia/ischaemia models in vitro, NRCM were incubated with the hypoxic solution as previously described 16 and then transferred to a hypoxia chamber. Hypoxic stimulation was maintained at 37°C with 5% CO 2 and 1% O 2 . After 8 hours, the cells were collected for further experiments. NRCM were transfected with Adv-FAF, si-FAF, OE-PAK2, miR-185-5p mimics and inhibitors, respectively, for 12 hours following the recommendations from the manufacturer.  The animal experiments were divided into two parts. In the first part of the study, rats were anaesthetized and kept ventilated through an animal ventilator. Rat models of MI were established by surgical obstruction of the left anterior descending coronary artery (LAD). For determination of the effect of lncRNA FAF overexpression on rats with MI, Adv-FAF or Adv-EGFP adenovirus (1 × 10 10 /100 µl) were intramyocardially injected into the ligation area after ligation of the LAD. For the sham group, rats also received similar surgical procedures but with no ligation. One week later, echocardiography was performed on all rats. Then, heart tissues were collected for further experiments.

| Quantitative real-time polymerase chain reaction
Total RNA was extracted from NRCM and rat heart tissues by Then, the cells were stained with the proper secondary antibody.
DAPI (Sigma, USA) was applied to label the nucleus. The staining results were acquired via fluorescence microscopy (Carl Zeiss, Germany).

| LDH release assay
The LDH release in the supernatants was determined by an LDH cytotoxicity assay kit (Beyotime, Shanghai, China). Cells were cultured in 96-well plates. Subsequently, the LDH detection agent was added following the manufacturer's instructions. The absorbance of the supernatants was determined at 490 nm, and LDH release was calculated by a standard curve.

| TTC staining
The infarct size was measured by triphenyl tetrazolium chloride (TTC) staining. Hearts were washed with normal saline and sliced into 2 mm coronary sections. Subsequently, the slices were incubated in 1% TTC at 37°C for 20 min and fixed in 10% formalin to measure the infarct area. The surviving tissue turned deep red, whereas the infarcted tissue bleached.

| Echocardiography
For assessment of the cardiac structure and function, rats were anaesthetized and examined by a transthoracic echocardiography detection system (Visuasonics, Canada) as previously described. 23 Left ventricular ejection fraction (LVEF) and left ventricular fractional shortening (LVFS) were automatically measured by a microcomputer of the echocardiography system.

| Scanning electron microscopy
Neonatal rat cardiomyocytes were harvested and sent to the Electron Microscopy Centre of Nanjing Medical University for further preparations. Samples were viewed under Scanning Electron Microscopy (SEM) (JEOL JSM-7900F SEM system).

| Transmission electron microscopy
Fresh myocardium was cut into fragments of 1 mm sections and fixed in 4% glutaraldehyde and 1% osmic acid overnight at 4°C overnight.
Then, the samples were sent to the Electron Microscopy Centre of Nanjing Medical University for further preparations. The ultra-

structure of myocardium was examined by Transmission Electron
Microscopy (TEM) (JEOL JEM-1400Flash TEM system).

| Statistical analysis
Statistical analyses were accomplished by using SPSS 19.0 software (IBM, USA). All experimental results are displayed as mean ± SD. The two-tailed Student's t-test was performed to determine the difference between two groups, while the one-way ANOVA followed by Tukey's test was applied for multiple comparisons. p < 0.05 indicated a significant difference. Graphs were generated by GraphPad Prism 8.0 (GraphPad Software, USA).

| LncRNA FAF improved cardiac function and suppressed pyroptosis in rats with MI
Our previous study found that expression of lncRNA FAF was downregulated in the ischaemic heart and associated with cardiomyocyte apoptosis. 22 To investigate the therapeutic efficacy of lncRNA FAF and potential antipyroptotic function in vivo, an adenovirus with FAF-overexpressing vector was synthesized and packaged in HEK293 cells. Then, we overexpressed lncRNA FAF in rats with MI by intramyocardially injecting Adv-FAF. After one week, qRT-PCR was conducted to assess the overexpression efficiency of lncRNA FAF ( Figure 1A). Cardiac function was evaluated by echocardiography one week after MI surgery. The echocardiographic parameters of LVFS and LVEF were increased in MI + Adv-FAF group compared with MI + Adv-EGFP group ( Figure 1B).
Consistently, TTC staining analysis showed that the infarct size in the MI + Adv-FAF group was smaller than that in the MI and MI + Adv-EGFP groups ( Figure 1C). Transmission electron microscopy was used to assess the myocardial injury at the ultrastructure level. As shown in Figure 1D, the TEM results of myocardium in MI and MI + Adv-EGFP group showed severe myofilaments lysis,

| LncRNA FAF suppressed hypoxia/ischaemiainduced cardiomyocyte pyroptosis
To further investigate the effects of lncRNA FAF on cardiomyocyte pyroptosis, NRCM were subjected to hypoxia/ischaemia conditions for 8 hours to induce pyroptosis ( Figure S1). Western blot analysis showed that the expression levels of pyroptosis-related proteins were significantly increased in the hypoxia/ischaemia group, which indicated the activation of pyroptosis under hypoxia/ischaemia (Figure 2A). Consistent with our previous study, 22 downregulation of lncRNA FAF was detected in the hypoxia/ischaemia injured cardiomyocytes by qRT-PCR ( Figure 2B). NRCM were further transfected with Adv-FAF or Adv-EGFP, following by hypoxia/ischaemia stimulus for 8 hours. qRT-PCR results indicated that lncRNA FAF was markedly upregulated in the Adv-FAF group The results of LDH release detection indicated that lncRNA FAF overexpression enhanced cell viability in the hypoxia/ischaemia injured cardiomyocytes ( Figure 2F). Finally, we examined the regulatory effects of lncRNA FAF on the expression of pyroptosisrelated proteins. The protein levels of NLRP3, cleaved caspase-1,

GSDMD-N, IL-1β and IL-18 decreased pronouncedly in the Adv-
FAF group compared with the Adv-EGFP group under hypoxia/ ischaemia ( Figure 2G). In addition, we performed immunofluorescence staining to detect NLRP3 activity in primary cardiomyocytes. Compared with transfection of negative control adenovirus, transfection of FAF-overexpressing adenovirus suppressed the activation of the NLRP3 inflammasome in hypoxic cardiomyocytes ( Figure 2H). Collectively, these findings suggested that overexpression of lncRNA FAF could protect cardiomyocytes against cardiomyocyte pyroptosis.

| Interfering with lncRNA FAF expression aggravated hypoxia/ischaemia-induced pyroptosis in cardiomyocytes
Knowing that lncRNA FAF positively regulates pyroptosis in cardiomyocytes, we next examined whether downregulation of lncRNA Western blot analysis revealed that pyroptosis-related protein levels were significantly increased in the hypoxia/ischaemia injured cardiomyocytes, which was further worsened by transfection of si-FAF ( Figure 3D). Immunofluorescence staining measurements showed that si-FAF transfection promoted the activation of the NLRP3 inflammasome in hypoxia/ischaemia injured cardiomyocytes, compared with the other two groups ( Figure 3E). Taken together, these results indicated that interfering with lncRNA FAF expression aggravated pyroptosis in hypoxia/ischaemia injured cardiomyocytes.

| Expression profile of miRNAs in the FAFoverexpressing NRCM and validation of potential target miRNAs
To further investigate how lncRNA FAF regulates pyroptosis under hypoxia, we performed small RNA sequencing to detect differentially expressed miRNAs in the FAF-overexpressing cardiomyocytes. The differential expression profile is presented as a heat map ( Figure 4A), and the screening criteria were absolute fold change >1.5, and adjusted P-values <0.05. From the miRNA expression profile, there were a total of 140 differentially expressed miRNAs between the Adv-FAF and Adv-EGFP cardiomyocytes.
Compared with the Adv-EGFP group, 75 miRNAs had upregulated expression and 65 had downregulated expression. As previously described, lncRNA FAF is characterized as a cytoplasmic lncRNA, 22 which may exert its regulatory function on pyroptosis by acting as a ceRNA for target miRNA. 25 Therefore, in this particular scenario, we crosschecked the miRNAs downregulated in small RNA-seq and CVDs that have been reported. On top of that, target miRNAs should contain putative binding sites for lncRNA FAF. A total of 6 candidate miRNAs met these criteria. qRT-PCR was further conducted to verify the candidate miRNAs. Among all the candidates, miR-185-5p showed the most significant downregulation in Adv-FAF cardiomyocytes ( Figure 4B). Thus, miR-185-5p was selected for further analysis.

| LncRNA FAF and PAK2 are direct targets of miR-185-5p in NRCM
RNAhybrid software was used to predict the potential target of  Figure 5G and H, miR-185-5p mimic transfection significantly reduced luciferase activities in wt-PAK2 and wt-lncRNA FAF cells compared with NC-mimic group, whereas transfection of miR-185-5p mimic did not change the luciferase activities in mut-PAK2 and mut-lncRNA FAF group. Overall, these findings revealed that FAF and PAK2 are direct targets of miR-185-5p in NRCM.

| LncRNA FAF exerted antipyroptotic effects by targeting miR-185-5p/PAK2
It has been reported that regulation of miR-185-5p and PAK2 could exert cardioprotective effects. On that basis, we further examined the functional relationship between lncRNA FAF and miR-185-5p/ PAK2 in hypoxia/ischmia-treated NRCM and ischaemic heart. In vitro, upregulation of miR-185-5p expression was detected in the hypoxia/ischaemia injury group, while downregulation of PAK2 expression was found in cardiomyocytes exposed to hypoxia and ischaemia ( Figure 6A,B). Overexpressing lncRNA FAF upregulated PAK2 protein levels under hypoxic conditions, which was reversed by transfection of miR-185-5p mimics ( Figure 6C). In addition, we examined the relationship between lncRNA FAF and PAK1, a close member of PAK2, under hypoxia. The data showed that the expression level of PAK1 was decreased in NRCM subjected to hypoxia/ischaemia condition, while overexpression of lncRNA FAF had no effect on PAK1 in NRCM ( Figure S2). The expression of pyroptosis-related proteins was evaluated by Western blot. As shown in Figure 6D,E, transfection of the miR-185-5p inhibitor contributed to the mitigation of cell pyroptosis. In contrast, miR-185-5p mimic aggravated cell pyroptosis under hypoxia-ischaemia.
It should be noted that both miR-185-5p inhibitor and PAK2 expression plasmid fail to further enhance the antipyroptotic effects in FAF-overexpressed NRCM ( Figure S4). These data sug-

| DISCUSS ION
Pyroptosis is considered as inflammasome-activated programmed cell death in response to various pathogens and host danger signals. 39 Emerging studies have revealed that pyroptosis plays an important role in the pathological process of MI and understanding its regulatory mechanism is urgently needed for developing new treatment strategies for MI. 40 Previously, Li et al. 41 found that overexpression of GDF11 repressed cell pyroptosis and alleviated

F I G U R E 3
Interfering with lncRNA FAF expression aggravated hypoxia/ischaemia-induced pyroptosis in cardiomyocytes. NRCM were transfected with si-FAF and si-NC, respectively, and subsequently exposed to hypoxia/ischaemia. LncRNAs exert biological function through various mechanisms, including chromatin organization, epigenetic modification, transcriptional and translation regulation. 19,43 In our previous study, lncRNA FAF exerts its biological function by regulating gene transcription in nucleus. 22,23 It is worth noting that lncRNAFAF is characterized as a cytoplasmic lncRNA, which tends to modulate gene expression by acting as a miRNA sponge in cytoplasm. In this study, we further investigated the potential molecular mechanism of lncRNA FAF acting as a miRNA sponge. We utilized small RNA sequencing and luciferase reporter assays to determine the candidate miRNAs.
MiR-185-5p expression was found to be significantly downregulated in small RNA profiles, which was further validated by qRT-PCR.
Luciferase reporter assays were applied to confirm that lncRNA FAF could sponge miR-185-5p and further upregulate PAK2 expression.
Collectively, we demonstrated that miR-185-p/PAK2 is a novel target of lncRNA FAF in NRCM.
MiR-185-5p has been reported to participate in the process of cell death in response to various cell stresses, including hypoxia/ ischaemia injury. 27,28,44 A recent study indicated that inhibition of miR-185-5p in endothelial cells contributed to the recovery of heart function after MI. 29 However, the exact mechanism by which miR-  49 In the present study, we also found that lncRNA FAF could regulate caspase-7 activity under hypoxia/ischaemia in NRCM ( Figure S3). However, more efforts are needed to understand its mechanism.
In summary, our study investigated the role of lncRNA FAF and its mechanism on cardiomyocyte pyroptosis. In both in vivo and in vitro experiments, we demonstrated that lncRNA FAF could alleviate cardiomyocyte pyroptosis, enhance cell viability and improve cardiac function through sponging miR-185-5p and thus to promote PAK2 expression. Our work may shed new light on the therapeutic potential of lncRNA FAF for future AMI treatment.

ACK N OWLED G M ENTS
This work was supported by grants from the National Natural

CO N FLI C T S O F I NTE R E S T
There are no conflicts of interest. Project administration (lead); Resources (lead); Supervision (equal).

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.