Silenced SOX2‐OT alleviates ventricular arrhythmia associated with heart failure by inhibiting NLRP3 expression via regulating miR‐2355‐3p

Nucleotide‐binding oligomerization domain‐like receptor family pyrin domain containing 3 (NLRP3) inflammasomes are the most important factors in ventricular arrhythmia associated with heart failure (VA‐HF). However, how the relationship between lncRNA and NLRP3 inflammasomes is regulated in VA‐HF has not been investigated in detail. Thus, we aimed to determine the effects of SOX2‐overlapping transcripts (SOX2‐OT) by targeting NLRP3 in rats with VA‐HF.

region of NLRP3. Inhibiting miR-2355-3p reversed the effect of SOX2-OT in rats with VA-HF.
arrhythmias, heart failure, inflammasome, lncRNA, microRNAs 1 | BACKGROUND Chronic heart failure (HF) is the end-stage manifestation of cardiovascular disease, which is mainly caused by hypertension, cardiomyopathy, as well as coronary and valvular heart diseases. 1 The rate of sudden death among patients with HF is six-to ninefold that of the general population, and 50%-70% of sudden cardiac deaths are due to malignant ventricular arrhythmias (VA) including tachycardia and fibrillation. 2,3 The main cause of death among patients with heart disease is VA, which needs to be resolved through appropriate investigation.
Inflammation is the most important factor in the induction of cardiomyocyte necrosis, because, with subsequent electrical and structural remodeling, it leads to VA. C-reactive protein, interleukin (IL)-1β, IL-2, IL-6, IL-8, and tissue necrosis factor (TNF-α), are inflammatory markers that become increased in VA, and C-reactive protein, IL-6, and white blood cells are independently associated with the incidence of HF. [4][5][6] Transforming growth factor-β (TGF-β) promotes extracellular matrix remodeling and cardiac collagen expression that contribute to diastolic dysfunction. 7 IL-1β aggravates VA induced by ischemia. 8 Inhibiting the expression of TGF-β, IL-1β, IL-6, IL-18, and TNF-α can reduce VA in ischemic HF, which plays a role in anti-VA. [9][10][11] Nucleotide-binding oligomerization domain-like receptor family pyrin domain containing 3 (NLRP3) inflammasomes were recently discovered as inflammatory response regulators that consist of NLRP3, ASC, and caspase-1. Activating NLRP3 inflammasomes can promote the release of IL-1β and IL-18, which triggers a cascade of downstream inflammation that eventually leads to local or systemic inflammation. 12 Abnormal activation of NLRP3 inflammasomes can lead to cardiac inflammation and heart dysfunction. 13,14 The activation of NLRP3 inflammasomes increases IL-1β release to promote myocardial inflammation, systolic dysfunction, and VA. 15,16 Additionally, silencing NLRP3 inflammasomes can reduce inflammation and improve cardiac dysfunction. 17 These findings show that NLRP3 inflammasomes comprise a novel mediator in patients with ventricular arrhythmia associated with heart failure (VA-HF). However, the regulatory mechanism of NLRP3 inflammasome activation in VA-HF remains unknown.
Long noncoding ribonucleic acids (lncRNA) play important functional roles in cardiovascular diseases. The lncRNA LIPCAR is a biomarker of cardiac remodeling that also independently predicts the death of patients with HF. 18 The expression of lncRNA Heat2 is significantly increased in the blood of patients with HF, and it regulates inflammatory and immune functions. 19 Downregulated lncRNA GASL1 promotes cardiomyocyte apoptosis via TGF-β1 activation. 20 The lncRNA SOX2-overlapping transcripts (SOX2-OT), is transcribed in the same orientation as an SRY-box containing gene 2 that is embedded in an intron of the SOX2-OT gene. The expression of SOX2-OT is significantly increased in HF. 21 However, the expression and functions of SOX2-OT in VA-HF remain unclear.
We constructed a rat model of VA-HF to determine SOX2-OT expression as well as the mechanism through which silencing SOX2-OT improves heart failure by inhibiting NLRP3 expression.

| METHODS
All animal experiments were performed in accordance with the National Institutes of Health Guidelines on the Use of Laboratory Animals and were approved by the Animal Care and Welfare Committee at Guangxi Medical University (Approval No. 201711060).

| Model induction and experimental design
SPF Male Sprague-Dawley rats (n = 48; weight, 240 ± 10 g) purchased from Experimental animal center of Guangxi Medical University (Nanning City, China) were housed at 23°C ± 2°C with 55% ± 5% humidity and 12:12 h equivalent day-night cycles. The rats had free access to food and water. After adaptation feeding for 1 week, the rats were randomly assigned to sham (n = 6) and HF (n = 40) groups. We established models of HF by aortic coarctation as described. 22 Briefly, after shaving the left side of the chest, the rats were anesthetized by 3% pentobarbital sodium (30 mg/kg intraperitoneally) inhalation. Heart failure was established by surgical aortic coarctation, then the rats were administered with penicillin (100,000 IU/d intramuscularly) for 3 days. Chronic heart failure was complete in the models after 4 weeks. The sham group underwent a similar surgical procedure without aortic coarctation. The HF model was stimulated with a 0.8 mA constant-current for 2 ms in the left cervical sympathetic ganglion to construct a VA model. Rats that developed ventricular tachycardia and fibrillation were assigned to a VA-HF group (n = 32), whereas those that did not were assigned to a VA-HF control group (n = 8). The VA-HF model rats were randomly assigned to VA-HF, small interfering negative control (si-NC), si-SOX2-OT, si-SOX2-OT + negative control sequence (anti-miR-NC), and si-SOX2-OT + anti-miR-miR-2355-3p groups (n = 6 per group). Each group was processed as follows: 100 μg si-NC, si-SOX2-OT, si-SOX2-OT and anti-miR-NC, and si-SOX2-OT and anti-miR-2355-3p were mixed with 50 μl of Entranster™-in vivo (animal in vivo transfection reagent, specially used for animals in vivo transfection, can directly transfer RNA into the animal body, No. 18668-11-1, Engreen Biosystem Co., Ltd.), and injected into the tail veins of the rats every 3 days for 4 weeks. 22 All animals were housed in the same room and the treatments were carried out in the order before grouping. Only therapists aware of the group allocation at the different stages of the experiment. All animals were killed at 6 weeks thereafter by an overdose of pentobarbital sodium (120 mg/kg). The following characteristics were considered humane endpoints to the study that required immediate intervention: Infection at the surgical site, rapid weight loss (>20% body weight loss), becoming cachectic, self-harming, biting or aggressive, and difficulty eating, drinking, or moving around freely.

| Tissue processing
The ventricular chamber was flushed with 0.9% sodium chloride chilled at 4°C to remove blood. Myocardial tissues were immediately cut into 2 to 3-mm wide sections and fixed in 4% paraformaldehyde for 4 h for hematoxylin-eosin staining (H&E staining), Masson staining, Picro Sirius Red staining, and apoptotic cell detection. Portions of the tissues were placed in cryotubes and stored under liquid nitrogen for subsequent determination of messenger RNA and related protein expression.

| Staining with H&E, Masson trichrome, and Picro Sirius Red
All outcome measures assessed according to the pathology results of H&E, Masson trichrome, and Picro Sirius Red. The ventricular chambers were washed with PBS for 30 min, dehydrated in a graded series of 50%, 70%, 90%, and 100% ethanol for 2 h each, cleared in xylene overnight, then infiltrated with paraffin. Tissue blocks were trimmed and sliced into 5-µm sections that were stained with H&E, (ab245880, Abcam plc.) Masson trichrome (ab150669, Abcam plc.), and Picro Sirius Red (ab150681, Abcam plc.) using respective kits as described by the manufacturer. Cardiomyocyte cross-sectional areas, changes in myocardial fibrosis, and the distribution of collagen I were assessed in 10 fields per sample by light microscopy (Olympus) and analyzed using Image-Pro Plus 6.0 software (Media Cybernetics, Inc.). Masson trichrome stained collagen fibers and cardiomyocytes blue and red, respectively, and Picro Sirius Red stained collagen I red.

| ROS assays
The generation ROS was measured using a dihydroethidium (DHE) probe (Beyotime). Briefly, cardiomyocytes were separated and incubated with the DHE (50 μM) probe for 30 min at 37°C. Levels of ROS were analyzed using a FACSCalibur flow cytometry system (Becton Dickinson & Co., D) with an argon-ion laser tuned to 485 nm.

| Dual-luciferase reporter gene assays
MicroRNA (miRNA) sponged by SOX2-OT were predicted using StarBase 3.0, 23 and miRNA that regulate the 3′-untranslated region (UTR) of NLRP3 were predicted using TargetScan (http://www.targetscan.org/). Wild-type (WT) or mutant SOX2-OT and the WT or mutant 3′-UTR of NLRP3 were cloned into the 3′-end of the firefly luciferase gene of the psi-CHECK2 vector. Thereafter, 30 ng of either WT or mutant constructs plasmid were cotransfected with 50 nM of either miR-2355-3p mimics or negative control into HEK 293 T cells derived from the HEK 293 cell line that expresses a mutant version of the SV40 large T antigen. Luciferase activity was measured 48 h later using a dualluciferase assay kit (Promega). The activity of Renilla luciferase was normalized against that of firefly luciferase.

| Statistical analysis
All data are shown as means ± standard deviation and were statistically analyzed using SPSS 19.0 software (IBM Corp.). Differences among the six groups which are normally distributed data were evaluated using one-way analyses of variance, followed by Tukey posthoc tests. Nonnormally distributed count data were analyzed using the rank-sum test. Values with p < .05 were considered statistically significant.

| SOX2-OT expression and NLRP3 inflammasomes are increased in rats with VA-HF
The results showed the expression of SOX2-OT was significantly higher in the HF than the sham group, and further enhanced compared with the HF and VA-HF-control groups ( Figure 1A). We also found that NLRP3, ASC, caspase-1, IL-1β, and TGF-β1 expression was significantly higher in the HF than in the sham group. Compared with the HF and VA-HF-control groups, the expression of NLRP3, ASC, caspase-1, IL-1β, and TGF-β1 was further enhanced in the VA-HF group ( Figure 1B).

| Silenced SOX2-OT inhibited NLRP3 inflammasomes, ROS levels, and histopathological changes in the myocardia of rats with VA-HF
We silenced SOX2-OT expression to determine its function in rats with VA-HF. The expression of SOX2-OT was significantly reduced in the ventricular chambers after rats were injected with si-SOX2-OT ( Figure 2A). Staining with H&E showed swollen and necrotic cardiomyocytes in the VA-HF group, an edematous myocardial cell gap, cytoplasmic vacuolization, and broken, dissolved, and disordered myocardial fibers in the si-NC group. In contrast, the number of cardiomyocytes was basically normal, muscle fibers were properly arranged, and the degree of necrosis among cardiomyocytes was relatively mild after injecting si-SOX2-OT into rats with VA-HF ( Figure 2B). Masson and Picro Sirius Red staining revealed obviously increased fibrosis in the si-NC group that was improved after injecting si-SOX2-OT into rats with VA-HF ( Figure 2C,D). The expression of NLRP3, ASC, caspase-1, IL-1β, and TGF-β1 and the ROS levels were obviously inhibited in the si-SOX2-OT, compared with the si-NC group ( Figure 2E,F).

| DISCUSSION
Ventricular arrhythmia associated with heart failure is involved in high morbidity and mortality rates. 24 Here, we found increased SOX2-OT expression and NLRP3 inflammasome levels in rats with VA-HF. Silencing SOX2-OT reduced ROS levels and the numbers of NLRP3 inflammasomes and alleviated histopathological lesions and cardiomyocyte fibrosis in rats with VA-HF. We also found that miR-2355-3p, which is a target of SOX2-OT, can reverse the effect of SOX2-OT in rats with VA-HF.
The regulatory role played by SOX2-OT in human diseases including bipolar disorder and schizophrenia is important. 25,26 Furthermore, SOX2-OT can predict poor overall and disease-free survival, and correlates with tumor development and metastasis; thus, it serves as a novel diagnostic marker and plays oncogenic roles in many cancers. 27,28 The expression of SOX2-OT gradually increases in healthy persons, and in patients with end-, and non-end-stage HF. 29 We found that SOX2-OT expression gradually increased in rats that were healthy, and in those with HF and VA-HF. The findings were similar to those of a previous study. Additionally, silenced SOX2-OT expression reduced the degree of cardiomyocyte necrosis and fibrosis and alleviated cardiac dysfunction in rats with VA-HF.
A major new finding of this study is that silencing SOX2-OT inhibited NLRP3, ASC, caspase-1, IL-1β, and TGF-β1 expression. The inflammatory markers, IL-1β and TGF-β1 promote extracellular matrix remodeling and cardiac collagen expression and aggravate VA induced by ischemia. 4,7,8 Caspase-1 NLRP3 and ASC are components of NLRP3 inflammasomes that can cause cardiac inflammation and cardiac dysfunction via IL-1β and TGF-β1 release. [13][14][15][16] Inhibiting TGF-β and IL-1β expression can alleviate cardiac dysfunction in ischemic VA-HF. [9][10][11] The present study found that NLRP3 inflammasomes and the expression of IL-1β and TGF-β1 gradually increased in healthy rats, and in rats with HF and VA-HF. These findings were in line with previous results. The overproduction of ROS is a key influencing factor that can induce heart failure, cardiac arrhythmias, and other pathological states. 30 Levels of ROS are significantly increased in myocytes incubated with IL-1β and TNF-α. 31 We found here that silencing SOX2-OT reduced ROS levels that were increased in rats with VA-HF. These results showed that SOX2-OT can regulate the NLRP3 inflammasome/IL-1β/ROS signaling pathway.
The relationship between lncRNA and NLRP3 has been discussed. The lncRNA MEG3/miR-223/NLRP3 axis in atherosclerosis promotes endothelial cell pyroptosis. 32 The lncRNA MALAT1/miR-22/NLRP3 promotes high levels of glucose-induced human endothelial cells pyroptosis. 33 The LINC00339/miR-22-3p/NLRP3 axis in kidney stones induced by calcium oxalate promotes renal tubular epithelial pyroptosis. 34 These results showed that lncRNA regulate NLRP3 expression by sponging miRNA. Others have also found that SOX2-OT can sponge miRNA such as miR-654 and miR-369-3p and that miR-146b-5p facilitates cancer cell proliferation and migration. [35][36][37] Here, we found that the target miRNA of SOX2-OT and NLRP3 is miR-2355-3p. The expression of SOX2-OT gradually decreased in healthy rats, and in rats with HF and VA-HF. We also found that miR-2355-3p reversed the effects of SOX2-OT on the degree of cardiomyocyte necrosis and fibrosis, the expression of NLRP3, ASC, caspase-1, IL-1β, and TGF-β1, and levels of ROS. These results suggested that silencing SOX2-OT inhibited NLRP3 expression by sponging miR-2355-3p, then inhibiting IL-1β and TGF-β1 expression and reducing ROS levels, resulting in alleviated cardiac dysfunction in rats with VA-HF.

| CONCLUSIONS
The present findings revealed that silencing SOX2-OT alleviated cardiac dysfunction by reducing the activation of NLRP3 inflammasomes. MicroR-93 might serve as a novel therapeutic target for the prevention of VA-HF.