The interplay of LncRNA ANRIL and miR‐181b on the inflammation‐relevant coronary artery disease through mediating NF‐κB signalling pathway

Abstract This study was designed to investigate whether ANRIL affected the aetiology of coronary artery disease (CAD) by acting on downstream miR‐181b and NF‐κB signalling. Altogether 327 CAD patients diagnosed by angiography were included, and mice models of CAD were established. Human coronary endothelial cells (HCAECs) and human umbilical vein endothelial cells (HUVECs) were also purchased. In addition, shRNA‐ANRIL, shRNA‐NC, pcDNA3.1‐ANRIL, miR‐181b mimic, miR‐181b inhibitor and miR‐NC were transfected into the cells. The lipopolysaccharides (LPS) and pyrrolidine dithiocarbamate (PDTC) were also added to activate or deactivate NF‐κB signalling. Both highly expressed ANRIL and lowly expressed miR‐181b were associated with CAD population aged over 60 years old, with smoking history, with hypertension and hyperlipidemia, with CHOL H 4.34 mmol/L, TG ≥ 1.93 mmol/L and Hcy ≥ 16.8 μmol/L (all P < 0.05). Besides, IL‐6, IL‐8, NF‐κB, TNF‐α, iNOS, ICAM‐1, VCAM‐1 and COX‐2 expressions observed within AD mice models were all beyond those within NC and sham‐operated groups (P < 0.05). Also VEGF and HSP 70 were highly expressed within AD mice models than within NC and sham‐operated mice (P < 0.05). Transfection of either pcDNA‐ANRIL or miR‐181b inhibitor could significantly fortify HCAECs’ viability and put on their survival rate. At the meantime, the inflammatory factors and vascular‐protective parameters were released to a greater level (P < 0.05). Finally, highly expressed ANRIL also notably bring down miR‐181b expression and raise p50/p65 expressions within HCAECs (P < 0.05). The joint role of ANRIL, miR‐181b and NF‐κB signalling could aid in further treating and diagnosing CAD.

studies mentioning the pathogenesis of atherosclerosis and development of ischaemic heart disease, around 30% of CAD cases failed to be explained by known cardiovascular hazards. 3,4 It was before manifested that CAD could be taken as a consequence of progressive inflammation, 5 yet the exact mechanisms underlying dysfunctions of CAD remained vague.
Long non-coding RNAs (lnc RNAs), a type of transcripts comprising >200 nucleotides, were involved in the aetiology of diverse human disorders through epigenetic, transcriptional and post-transcriptional regulations. 6,7 Interestingly, it was appeared that lncRNAs also participated in the development of cardiovascular diseases, including heart failure, cardiac hypertrophy, cardiac metabolic diseases and myocardial infarction. 8 Among them, lncRNA ANRIL, also named as CDKN2BAS, was ranked as the best replicating genetic risk factor for CAD, 8 and it was speculated to alter expressions of related proteins through RNA interference, gene silencing, chromatin remodelling and DNA methylation. 9 In addition, the function of lncRNAs was often mediated through the regulation of microRNAs (miRNAs), which post-transcriptionally regulate gene expression by binding to the 3′ untranslated region (UTR) of mRNAs. 10 For instance, it was documented that ANRIL modifying miR-181b could boost a series of vicious transformations that were relevant to human vascular inflammation. 11 However, whether ANRIL would impact on miR-181b to regulate the onset or progression of CAD remained unanswered.
The miR-181 pointed out here has been revealed to affect different aspects of cell life activities, including cell proliferation, cell differentiation and cell death. 12,13 Furthermore, miR-181b was discovered to inhibit NF-κB-mediated endothelial cell activation by lowering the expression of importin-α3 (IPOA3), a key protein assisting in translocation of NF-κB from cytoplasm to nucleus. 14 Notwithstanding, the role of miRNA-181b in chronic inflammatory disease, such as atherosclerosis, has not been examined. However, in the vascular endothelium, NF-κB activation induces the expression of pro-inflammatory genes, which seemed as a pivotal parameter for the occurrence and development of atherosclerosis. [15][16][17] All in all, it was hypothesized that ANRIL, miR-181b and NF-κB might function to modify the aetiology of atherosclerosis. In response, this study was designed to investigate whether ANRIL regulated the presence and progression of CAD by acting on downstream miR-181band NF-κB signalling.

| Establishment of mice models and sample collection
The SD rats of clean grade (male; 7-week; 220-250 g) were provided by the experimental animal centre of the 455th Hospital of Chinese People's Liberation Army. The rats were anaesthetized with usage of 50 mg/kg pentobarbital sodium. Their four limbs were subcutaneously inserted with needle electrodes, and changes of electrocardiogram were monitored. After tracheal intubation, the rats' breath was controlled at 90 times/min, and tidal volume was set as 3-4 mL. The rats were disinfected and operated with thoracotomy before their pericardium was cut. After 30-minute successful ligation, the artery clamp was loosen, and reperfusion was sustained for 90 minutes to cause ischaemia reperfusion injury.
About 10 minutes later, ligation was considered as successful if STsegment elevation was observed within electrocardiogram, and those without ST-segment elevation was sifted out. The cardiac tissues of rats were taken out, and the ischaemic myocardium within left ventricle of the heart was clipped and was frozen within liquid nitrogen at −80°C.

| Determination of cardiac function
Sacculus was inserted through atrioventricular valve, and pressure transducer was connected with the computer. Then left ventricular systolic pressure (LVSP), maximum change rate of left intraventricular pressure (±dp/dtmax) and coronary blood-flow volume (CF) were recorded.

| Cell culture
Human coronary endothelial cells (HCAECs) and human umbilical vein endothelial cells (HUVECs) were purchased from the American Type Culture Collection (ATCC). HUVECs and HCAECs were cultured within endothelial cell growth medium (ScienCell, USA) that was supplemented with 5% (v/v) FBS, 1% endothelial cell growth supplement (ECGS) and 1% penicillin streptomycin. The HUVECs and HCAECs within 3-7 passages were applied for the following experiments.

| Colony formation assay
The transfected cells were seeded at 1 × 104 within culture dishes of a 35-mm diameter. The cells stained with trypan blue were observed and counted in triplicate within 6 weeks. Then cells were dissociated and were suspended in the medium containing 0.3% agar. The colonies exceeding 0.5 mm in diameter were counted after 14 days.

| Cell proliferation assay
All the procedures were carried through in line with the instructions of CCK-8 kit (Dojindo Laboratories, Kumamoto, Japan). After 72-hour transfection, 2 × 10 4 cells (100 μL) were inoculated within each well of 96-well culture plate. About 3-4 hours later when cells grew against the wall, 100 μL RPMI 1640 complete medium and 10 μL CCK-8 were added. Then, the cells were cultured in 5% CO 2 at 37°C for 2 hours.
The D450 values would be determined with a microplate reader (infinite M200, Tecan, Austria) at the wavelength of 450 nm.

| Cell apoptosis assay
When cells were re-suspended within 500 μL binding buffer, 5 μL

| Western blotting
We prepared RIPA lysis buffer to extract proteins from tissues and cells, and Bradford method was employed to measure the protein concentration. Subsequently, equal amounts (ie, 30 μg) of proteins were electrophoresed on the 6% or 10% polyacrylamide gel, and the isolated proteins were then transferred onto the polyvinylidene

| Dual luciferase reporter gene assay
The ANRIL fragments that covered specific miR-181b binding sites were cloned into the pmirGLO dual luciferase expression vector (Promega, Madison, WI, USA) to construct reporter vectors named as pmirGLO-ANRIL-Wt. The same vector including miR-181b mutational sites within the ANRIL sequence was called as pmirGLO-ANRIL-Mut.
The HCAECs that have been transfected with miR-181b and miR-NC were again transfected with pmirGLO-ANRIL-WT or pmirGLO-ANRIL -MUT. Similarly, certain NF-κB fragments that contained miR-181b binding sites were amplified by PCR and were then cloned into pmir-GLO dual luciferase expression vector to form NF-κB-Wt. The NF-κB-Mut was produced in a manner same to NF-κB-Wt, except that the miR-181b binding sites within NF-κB were mutated. When cell confluency reached 40% -50%, the cells that have been transfected with miR-181b mimics or miR-NC were, respectively, transfected with NF-κB-Wt, NF-κB-Mut or pRL-TK reporter vector. Then, a luciferase reporter assay was performed based on the dual-luciferase reporter gene detection system (Promega).

| Statistical analyses
All the statistical analyses were performed with the aid of GraphPad Prism software (GraphPad Prism Software Inc., San Diego, USA). Student's t test or one-way ANOVA test was employed to compare quantitative data (mean ± SD), and chi-square test was applied for evaluation of enumeration data. A P value < 0.05 was considered as a mark of significant differences.

Relative expression in peripheral blood
The Expressions of lncRNA ANRIL and miR-181b were Compared between Coronary Artery Disease (CAD) Tissues/Cells and Normal Tissues/Cells. A, ANRIL and miR-181b expressions were compared between CAD tissues and normal tissues. *P < 0.05 when compared to normal tissues. B, ANRIL and miR-181b expressions were compared between HUVECs and HCAECs. *P < 0.05 when compared to HUVECs. C, The survival rates of CAD patients with differentially expressed ANRIL and miR-181b were compared. D, ANRIL expressions were negatively correlated with miR-181b expressions within CAD tissues 3 | RESULTS

| Association of ANRIL and miR-181b expressions with baseline characteristics of CAD patients
Founded on the percentage of coronary artery stenosis, the CAD patients were categorized into LAD1 group (81%-100%, severe), LAD2 group (51%-80%, moderate), LAD 3 group (30%-50%, mild) and healthy group. According to Figure 1A, up-regulated ANRIL expression and down-regulated miR-181b expression were observed within CAD patients when compared to the healthy group (P < 0.05). It was further revealed that ANRIL expression increased with the rising degree of LAD (P < 0.05), yet miR-181b expression followed a tendency opposite to ANRIL (P < 0.05). Similar to the results of tissues, ANRIL was expressed more within HCAECs than within HUVECs, and miR-181b expression within HCAECs appeared lower than within HUVECs (P < 0.05)( Figure 1B). More than that, both highly expressed ANRIL and lowly expressed miR-181b were associated with CAD population aged over 60 years old, with smok-  Figure 1C). At the same time, Spearman's correlation analysis unfolded that miR-181b expressions were negatively correlated with ANRIL expressions among the CAD patients investigated (rs = −0.534, P < 0.001) ( Figure 1D).

| Variation of cardiac functions within CAD mice models
When compared to NC group and sham-operated group, the CAD mice models were observed with remarkably decreased LVSP, +dp/dtmax and -dp/dtmax (all P < 0.05). Following an analogous trend, CAD mice models exhibited a significantly inhibited CF in comparison with NC group and sham-operated group (P < 0.05) (Figure 2A). However, there was no significant distinction of LVSP, +dp/dtmax, -dp/dtmax NC and CF within NC group from those within sham-operated group (P > 0.05).
The results of CCK8 assay and colony formation assay both showed that the viability and survival status of HCAECs were more vigorous within miR-NC+NF-κB activator group than within miR-NC group (P < 0.05), suggesting that NF-κB suppressed the miR- apoptosis. What's more, miR-NC+NF-κB activator group was determined with lowly expressed E-cadherin and highly expressed N-cadherin and vimentin, when compared to miR-NC group (P < 0.05).
LncRNAs usually modulated disease progression through functioning on coding or non-coding genes to boost or hinder the regulatory network within cellular processes. 38,39 This investigation held that the regulatory role of ANRIL in CAD was played via regulation of miR-181b (Figures 3-6), which has been a focus among CAD studies. In particular, plasma miR-181b was under-expressed remarkably within CAD patients than within normal subjects. 14 Furthermore, it was mirrored that decreased miR-181b restrained vascular remodelling by activating TGF-β/pSmadD2/3 pathway, 40 and addition of angiotensin II into cardiac fibroblasts could significantly down-regulate miR-181b expression. 41 In addition, it has been illuminated that miR-181b might suppress breast cancer cells' capacity of metastasis by targeting CXCL1 and CXCL2, a couple of inflammatory cytokines. [42][43][44] He et al 45  With respect to inflammatory reactions, it was displayed that MiR-181b could target Importin-3(IPOA3) to restrain nuclear transfer of NF-κB, thereby down-regulating VCAM-1 expression and relieving inflammatory responses. 14 Moreover, after intravenous injection of miR-181b mimic into atherosclerosis mice, the convergence of macrophages, CD 4+ T cells and other inflammatory cells were held up, along with decreased expressions of inflammatory markers. 11 Similar to this, our study verified that miR-181b could induce inflammatory reactions both in vivo and in vitro, which was specifically manifested as that the expressions of IL-6, IL-8, TNF-α, iNOS, ICAM-1, VCAM-1, COX-2 were down-regulated ( Figures 5 and 8).
The NF-κB signalling mentioned above not only resulted in incremental apoptosis and declined migration of cancer cells, 48  Lastly, curves were not fitted to assess if ANRIL and miR-181b could act as biomarkers for early-diagnosis of CAD.