CDKN2B-AS1 participates in high glucose-induced apoptosis and fibrosis via NOTCH2 through functioning as a miR-98-5p decoy in human podocytes and renal tubular cells

Background Diabetic nephropathy (DN) is the most common causes of end-stage renal disease. Long non-coding RNA cyclin-dependent kinase inhibitor 2B antisense RNA 1 (CDKN2B-AS1) is connected with the development of DN, but the role of CDKN2B-AS1 in DN has not been entirely elucidated. Methods Quantitative real-time polymerase chain reaction (qRT-PCR) was carried out to measure CDKN2B-AS1 and miR-98-5p levels. Cell viability, proliferation, and apoptosis were analyzed with 3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium bromide (MTT) or flow cytometry assays. Protein levels were measured by western blotting. The relationship between CDKN2B-AS1 or notch homolog 2 (NOTCH2) and miR-98-5p was verified via dual-luciferase reporter assay. Results CDKN2B-AS1 and NOTCH2 were upregulated in the serum of DN patients and high glucose-disposed human podocytes (HPCs) and human renal tubular cells (HK-2), whereas miR-98-5p was downregulated. High glucose repressed viability and accelerated apoptosis of HPCs and HK-2 cells. CDKN2B-AS1 knockdown impaired high glucose-induced apoptosis and fibrosis of HPCs and HK-2 cells. Mechanistically, CDKN2B-AS1 sponged miR-98-5p to regulate NOTCH2 expression. Also, CDKN2B-AS1 inhibition-mediated effects on apoptosis and fibrosis of high glucose-disposed HPCs and HK-2 cells were weakened by miR-98-5p inhibitor. Also, NOTCH2 knockdown partly reversed miR-98-5p inhibitor-mediated impacts on apoptosis and fibrosis of high glucose-disposed HPCs and HK-2 cells. Conclusion High glucose-induced CDKN2B-AS1 promoted apoptosis and fibrosis via the TGF-β1 signaling mediated by the miR-98-5p/NOTCH2 axis in HPCs and HK-2 cells. Supplementary Information The online version contains supplementary material available at 10.1186/s13098-021-00725-5.


Introduction
Diabetic nephropathy (DN), a progressive kidney disease caused by diabetes, is characterized by persistent albuminuria and a gradual decline in estimated glomerular filtration rate [1,2]. It is reported that about 30-40% of patients with diabetes may develop DN, and approximately 50% of DN patients tend to develop end-stage renal disease [3]. Renal fibrosis is the main driving force for the occurrence of DN, and hyperglycemia in diabetic patients may trigger renal fibrosis [4]. Therefore, exploring the pathogenesis of DN is important to improve DN.
It has been confirmed that the NOTCH pathway mediates renal fibrosis [20,21]. Notch homolog 2 (NOTCH2) is one of the important receptors in the NOTCH pathway [22]. NOTCH2 was reported to be connected with high glucose-stimulated cardiac fibrosis and epithelial-tomesenchymal in HAVECs [23]. Also, JNK is an upstream effector of NOTCH2 in TGF-β1-mediated renal fibrosis [24]. Furthermore, HG triggered EMT through the Notch2 pathway in NRK-52E cells [25]. However, the regulatory mechanisms associated with NOTCH2 in DN development have not been fully elucidated.
Herein, we reported an accelerative influence of CDKN2B-AS1 on the pathogenesis of DN. Also, we found that CDKN2B-AS1 induced apoptosis and fibrosis through upregulating NOTCH2 via sponging miR-98-5p under high glucose treatment. Therefore, the research provided a novel mechanism to comprehend the pathogenesis of DN.

Subjects
The research was authorized and supervised by the ethics committee of Qilu Hospital, Cheeloo College of Medicine, Shandong University. 30 patients with DN and 30 healthy controls were recruited from Qilu Hospital, Cheeloo College of Medicine, Shandong University. T2D patients with a urine albumin/creatinine ratio > 30 mg/g or estimated glomerular filtration rate (eGFR) < 60 mL/ min/1.73 m 2 were defined to have DN. The participating patients were free of cardiovascular disease, chronic liver disease, or cerebrovascular disease. Participants in this study signed informed consent.

Quantitative real-time polymerase chain reaction (qRT-PCR)
Total RNA was extracted through TRIzol reagent (Life Technologies). Total RNA was reverse-transcribed by PrimeScript RT reagent Kit (Takara, Dalian, China) or miRNA First-Strand Synthesis Kit (Takara). QPCR was conducted through the SYBR Premix Ex Taq (Takara) with specific primers (Table1, β-actin and U6 were utilized as house-keeping genes). Expression levels of CDKN2B-AS1 and miR-98-5p were figured with the 2 − ΔΔCt method.

Cell apoptosis analysis
After collection, digestion, and centrifugation, the cells were re-suspended in binding buffer (1×). Cell apoptosis was analyzed using the Annexin V-fluorescein isothiocyanate (FITC)/PI apoptosis detection kit (KeyGen, Jiangsu, China). Cell fluorescence was analyzed through a FACScan flow cytometry (Beckman Coulter, Brea, CA, USA).

Statistical analysis
The data were expressed as mean ± standard deviation, which was derived from 3 replicate experiments. Graph-Pad Prism 6.0 software was utilized for statistical analysis. Differences were deemed significant if P < 0.05. Student's t test was used to analyze the differences between two groups. One-way variance analysis with Turkey's test Table 1 Primer sequences for qRT-PCR was utilized for the comparison of the differences among more groups.

CDKN2B-AS1 was upregulated in DN and high glucose-stimulated HPCs and HK-2 cells
Considering the abnormal expression of CDKN2B-AS1 in DN, we detected the expression level of CDKN2B-AS1 in serums from 30 DN patients and 30 normal controls. QRT-PCR manifested that CDKN2B-AS1 expression levels were increased in the serum of DN patients compared to the control group (Fig. 1A). Subsequently, we assessed the viability of HPCs and HK-2 cells treated with different concentrations of glucose.
MTT assay presented that high glucose (30 mM and 40 mM) led to a decrease in the viability of HPCs and HK-2 cells (Fig. 1B, C). And the HPCs and HK-2 cells treated with 30 mM glucose were chose for subsequent analysis. We observed that CDKN2B-AS1 expression levels were elevated in high glucose-treated HPCs and HK-2 cells (Fig. 1D). These results indicated that CDKN2B-AS1 might be involved in the development of DN.

CDKN2B-AS1 regulated apoptosis and fibrosis of HPCs and HK-2 cells under high glucose treatment
Given that the upregulation of CDKN2B-AS1 in DN and high glucose-disposed HPCs and HK-2 cells, we further investigated the role of CDKN2B-AS1 in DN through loss-of-function experiments. Compared to the control groups, CDKN2B-AS1 was overexpressed in HPCs and HK-2 cells after transfection with CDKN2B-AS1 under high glucose treatment and decreased in HPCs and HK-2 cells after transfection with si-CDKN2B-AS1 under high glucose treatment ( Fig. 2A, B). Moreover, CDKN2B-AS1 elevation aggravated proliferation inhibition and apoptosis of high glucose-stimulated HPCs and HK-2 cells, but CDKN2B-AS1 downregulation impaired proliferation inhibition and apoptosis of high glucosestimulated HPCs and HK-2 cells (Fig. 2C-F). Western blotting displayed that CDKN2B-AS1 elevation resulted in a decrease in Bcl-2 protein levels and an increase in Bax protein levels in high glucose-disposed HPCs and HK-2 cells, while CDKN2B-AS1 silencing played an opposing impact (Fig. 2G, H). In addition, CDKN2B-AS1 overexpression elevation protein levels of TGF-β1, FN, Col.I in high glucose-disposed HPCs and HK-2 cells, but CDKN2B-AS1 silencing decreased protein levels of TGF-β1, FN, Col.I in high glucose-disposed HPCs and HK-2 cells. Collectively, these findings demonstrated that CDKN2B-AS1 regulated apoptosis and fibrosis of HPCs and HK-2 cells under high glucose treatment.

CDKN2B-AS1 was identified as a sponge for miR-98-5p
To explore the underlying molecular mechanism of CDKN2B-AS1 in DN, we predicted the miRNAs that might interact with CDKN2B-AS1 through using the starBase database. MiR-98-5p was discovered to possess a complementary sequence with CDKN2B-AS1 (Fig. 3A). Subsequently, we performed the dual-luciferase reporter assay to verify this prediction. The results exhibited that miR-98-5p mimic repressed the luciferase intensity in HPCs and HK-2 cells with a luciferase reporter containing WT-CDKN2B-AS1, while there was no overt difference in the luciferase reporter containing MUT-CDKN2B-AS1 (Fig. 3B, C). And miR-98-5p expression was decreased in HPCs and HK-2 cells after transfection with miR-98-5p inhibitor (Fig. 3D). Also, CDKN2B-AS1 silencing elevated miR-98-5p expression in HPCs and HK-2 cells, but this suppression was reversed by miR-98-5p downregulation (Fig. 3E, F). Moreover, miR-98-5p was downregulated in the serum of DN patients relative to the normal controls (Fig. 3G). There was a marked reduction in miR-98-5p expression in high glucose-treated HPCs and HK-2 cells (Fig. 3H). These results suggested that CDKN2B-AS1 acted as a miR-98-5p sponge.

NOTCH2 was a downstream target of miR-98-5p
We further predicted the underlying targets of miR-98-5p with the starBase database. And the results presented that NOTCH2 had the complementary base fragment with miR-98-5p (Fig. 5A). The luciferase activity of luciferase reporter with WT-NOTCH2-3ʹUntranslated Regions (UTR) was decreased by miR-98-5p mimic in HPCs and HK-2 cells, while the luciferase intensity of luciferase reporter with MUT-NOTCH2-3ʹUTR did not change (Fig. 5B, C). And the protein levels of NOTCH2 in HPCs and HK-2 cells were markedly restrained after transfection with si-NOTCH2 compared to the control si-NC (Fig. 5D). Moreover, miR-98-5p inhibitor elevated NOTCH2 protein levels in HPCs and HK-2 cells, while this elevation was weakened by NOTCH2 silencing (Fig. 5E, F). Furthermore, NOTCH2 protein levels were also increased in the serum of DN patients and high glucose-treated HPCs and HK-2 cells (Fig. 5G, H). Also, the level of activated NOTCH2 protein was significantly increased under HG conditions (Additional file 1: Figure  S1). Collectively, these results indicated that NOTCH2 served as a target of miR-98-5p.

Discussion
Persistent proteinuria with or without decreased glomerular filtration rate has been used to define DN [26]. In this study, T2D patients with a urine albumin/creatinine ratio > 30 mg/g or estimated glomerular filtration rate (eGFR) < 60 mL/min/1.73 m 2 were defined to have DN. In addition, the inclusion of DN patients without renal biopsy was a major limitation of the study. Podocytes, which constitute the glomerular filtration barrier, have limited regeneration and repair capabilities [27]. Renal tubular injury is an important manifestation of DN [28]. Studies have confirmed that the deregulation of lncRNAs is closely related to the progress of DN [8]. Report of Zhang et al. revealed that lncRNA MALAT1 was overexpressed in high glucose-treated HK-2 cells, resulting in accelerating cell epithelial-to-mesenchymal transition and injury [29]. Another report pointed out that lncRNA MALAT1 increased SIRT1 expression via targeting miR-9, thus alleviating podocyte damage via boosting cell viability and repressing cell apoptosis under high glucose treatment [30]. Lv et al. manifested that lncRNA GAS5 silencing mitigated high glucose-induced viability suppression and apoptosis acceleration of HK-2 cells via sponging miR-27a [28]. In the current study, CDKN2B-AS1 was upregulated in the serum of DN patients and high glucose-treated HPCs and HK-2 cells. Moreover, CDKN2B-AS1 silencing elevated cell viability and decreased cell apoptosis in HPCs and HK-2 cells under high glucose treatment. Thomas et al. demonstrated that CDKN2B-AS1 downregulation protected decreased urine albumin levels and urine volume in diabetic mice [31]. A recent research indicated that CDKN2B-AS1 knockdown inhibited extracellular matrix accumulation and proliferation of high glucose-treated HGMC cells through repressing HMGA2 expression by adsorbing miR-424-5p [16].
TGF-β1 is considered to be the main regulator of profibrosis [32]. Increasing evidence has demonstrated that the TGF-β1 signaling exerts a vital role in DN pathogenesis [33][34][35][36]. Moreover, TGF-β1 can contribute to glomerular filtration disorder, fibrosis, and sclerosis [37]. Also, Sitagliptin can block the TGF-beta1/Smad pathway, thus ameliorating diabetic nephropathy [38]. Herein, CDKN2B-AS1 silencing decreased protein levels of TGF-β1, FN and Col.I in high glucose-disposed HPCs and HK-2 cells, indicating that CDKN2B-AS1 silencing decreased the fibrosis of HPCs and HK-2 under high glucose treatment. Thus, we concluded that high glucoseinduced apoptosis and fibrosis of HPCs and HK-2 were partly dependent on CDKN2B-AS1. LncRNAs usually exert their roles through acting as a sponge for miRNAs in DN [16,28]. A previous study revealed that miR-98-5p repressed human endothelial cell growth through targeting cyclinD2 [39]. Another research reported that miR-98-5p mitigated renal fibrosis and epithelial-to-mesenchymal via modulating HMGA2 expression in DN [19]. Herein, miR-98-5p was downregulated in the serum of DN patients and high glucose-treated HPCs and HK-2 cells. CDKN2B-AS1 was validated as a sponge for miR-98-5p, and the impacts of CDKN2B-AS1 inhibition on proliferation, apoptosis, and fibrosis of high glucose-treated HPCs and HK-2 cells were overturned by miR-98-5p inhibitor. Thus, we concluded that CDKN2B-AS1 played its influence on high glucose-treated HPCs and HK-2 cells via sponging miR-98-5p.
In conclusion, high glucose-mediated CDKN2B-AS1 elevated NOTCH2 expression via adsorbing miR-98-5p, leading to facilitating cell apoptosis and fibrosis in HPCs and HK-2 cells. The study offered a novel mechanism by which CDKN2B-AS1 participated in the pathogenesis of DN. Fig. 6 MiR-98-5p regulated apoptosis and fibrosis of high glucose-treated HPCs and HK-2 cells by targeting NOTCH2. A-H HPCs and HK-2 cells were transfected with inhibitor NC, miR-98-5p inhibitor, miR-98-5p inhibitor + si-NC, or miR-98-5p inhibitor + si-NOTCH2 and then treated with high glucose. A-F The proliferation, apoptosis, and apoptosis-related proteins of HPCs and HK-2 cells under high glucose treatment were evaluated via MTT assay, flow cytometry assay, and western blotting. G, H Protein levels of TGF-β1, FN, and Col.I in HPCs and HK-2 cells under high glucose treatment were measured by western blot analysis. *P < 0.05