Abstract
BACKGROUND:
Endothelial dysfunction is considered to be an important factor in the pathogenesis of atherosclerosis. Circular RNAs (circRNAs) have been confirmed to participate in the development of atherosclerosis. Nevertheless, the role and mechanism of circ_0091822 in atherosclerosis have not been studied yet.
METHODS:
The expression of circ_0091822, miR-661 and RAB22A were analyzed by quantitative real-time polymerase chain reaction (qRT-PCR). The levels of interleukin-6 (IL-6) and tumor necrosis factor-α (TNF-α) were estimated by enzyme-linked immunosorbent assay (ELISA). Cell viability was analyzed by Cell Counting Kit-8 (CCK-8) assay, cell proliferation was evaluated by EdU assay, and cell apoptosis was gauged by flow cytometry. Western blot was performed to assess the protein levels of Bax, Cleaved-caspase-3 and RAB22A. The interaction among miR-661 and circ_0091822 or RAB22A was verified by dual-luciferase reporter assay
RESULTS:
Ox-LDL enhanced the expression of circ_0091822 in HUVECs. It also constrained proliferation, promotes apoptosis and inflammation in HUVECs, and down-regulation of circ_0091822 attenuated these effects. Mechanically, circ_0091822 could serve as a sponge of miR-661, miR-661 interference rescued circ_0091822 inhibition-mediated effect on the biological functions in ox-LDL-induced HUVECs. Additionally, RAB22A was a target of miR-661, and its overexpression could partially overturn the negative regulation of miR-661 on ox-LDL-treated HUVECs injury. Importantly, circ_0091822 sponged miR-661 to positively regulate RAB22A expression.
CONCLUSION:
Circ_0091822 contributed to cell injury by targeting miR-661/RAB22A axis in ox-LDL-stimulated HUVECs.
Keywords
Introduction
Atherosclerosis (AS) is one of the most common cardiovascular diseases worldwide [1]. As a chronic inflammatory disease, AS leads to the development of a variety of cardiovascular diseases, and is also one of the prominent causes of high mortality in elderly population [2, 3]. However, the mechanisms of the occurrence and development of AS are not completely understood. Therefore, clarifying the mechanism of its occurrence and development will help to identify effective diagnostic indicators and treatment strategies. Endothelial cells play a crucial role in maintaining vascular homeostasis [4]. Oxidized low-density Lipoprotein (ox-LDL)-evoked injury in endothelial cells is considered to be a vital factor in the pathogenesis of AS [5]. Ox-LDL facilitates endothelial cell damage characterized by abnormal cell growth, apoptosis, inflammation, and angiogenesis, thereby disrupting endothelial integrity and aggravating the development of AS [6]. Studies have reported that ox-LDL promoted endothelial cell apoptosis by enhancing the activity of CPP32-like protease [7]. Ox-LDL could inhibit the activity of human aortic endothelial cells and promote apoptosis and inflammation [8]. Accumulating studies have also identified that ox-LDL-treated endothelial cells are used as a cell model of AS [9, 10]. Hence, it is urgent to identify key regulators of endothelial cell dysfunction.
As a new type of non-coding RNA, circular RNAs (circRNAs) are involved in the regulation of a variety of cell biological process [11]. Because of its special closed loop structure and strong stability, circRNAs may be applied as a potential marker for the diagnosis and treatment of human diseases [12]. With further exploration, researchers found that there were multiple microRNA (miRNA) binding sites in circRNAs, and confirmed that circRNA regulated the expression of target genes by sponging miRNA, and ultimately affected cell biological process [13]. Several studies have confirmed that circRNAs modulated the disease process of AS [14]. For instance, Ji et al. identified that circ_0004104 and Tumor necrosis factor-α-induced protein 8 (TNFAIP8) were upregulated in ox-LDL-exposed endothelial cells, and circ_0004104 silencing weakened the influence of ox-LDL on endothelial cell injury. TNFAIP8 is a member of the TIPE/TNFAIP8 family which modulates cell growth and survival. Mechanically, circ_0004104 promote the expression of TNFAIP8 by sponging miR-100 in endothelial cells [15]. Huang et al. verified that the circUSP36 was highly expressed in ox-LDL-exposed endothelial cells, and circUSP36 retarded the proliferation and migration in ox-LDL-stimulated endothelial cells by targeting miR-637/WNT4 pathway [16]. Peng et al. demonstrated that the circ-USP9X deficiency protected against ox-LDL-evoked cytotoxicity in Human umbilical vein endothelial cells (HUVECs) via sponging miR-599 to regulate the level of CLIC4 [17]. As for circ_0091822, it has been identified to be increased in ox-LDL-stimulated HUVECs. However, the biological function of circ_0091822 in ox-LDL-triggered cytotoxicity in HUVECs has never been reported [18]. The purpose of our work is to uncover the effect of circ_0091822 on the biological functions of ox-LDL-stimulated HUVECs and to demonstrate its underlying molecular mechanism.
We assessed the role of circ_0091822 in AS development via measuring cell inflammatory response, viability, proliferation and apoptosis in ox-LDL-treated HUVECs. Subsequently, the downstream target genes of circ_0091822 were predicted and verified to elucidate the working mechanism of circ_0091822.
Materials and methods
Tissue samples
27 patients with AS and 23 healthy volunteers who have no history of cardiovascular disease were collected at Luodian Hospital, Baoshan District. Each patient signed written informed consent before collecting blood samples. The protocols were approved by the Ethics Committee of Luodian Hospital, Baoshan District.
Cell culture and treatment
HUVECs were acquired from Procell (Wuhan, China), and cultured with Dulbecco’s modified Eagle medium (DMEM, Solarbio, Beijing, China) complete medium. Oxidized low density lipoprotein (ox-LDL) was used to establish injury model. HUVECs of ∼60% confluence were incubated with various concentrations of ox-LDL (0, 20, 40, 80μg/mL) for 24 h. In subsequent functional assay, cells were stimulated with ox-LDL (40μg/mL).
Quantitative real-time polymerase chain reaction (qRT-PCR)
HUVECs were lysed and total RNA was extracted by TRIzol reagent (Invitrogen, Carlsbad, CA, USA). The purity and concentration of total RNA were measured by NanoDrop-1000 apparatus. RT-PCR kit (Invitrogen) and miScript Reverse Transcription Kit (Qiagen, Hilden, Germany) were utilized to conduct reverse transcription experiment. After that, qRT-PCR was performed by using the SYBR Green (Invitrogen). Glyceraldehyde 3-phosphate dehydrogenase (GADPH) or U6 served as reference genes. The data was analyzed by 2–ΔΔCt method. Primer sequences were exhibited in Table 1.
Cell transfection
The circ_0091822 small interfering RNA (si-circ_0091822) and corresponding negative control (NC) (si-NC), and overexpression plasmid, miR-661 mimic (miR-661) and matching NC (miR-NC), miR-661 inhibitor (anti-miR-661) and matching NC (anti-miR-NC), RAB22A overexpression vector pcDNA3.1-RAB22A (RAB22A) and matching NC pcDNA vectors (pcDNA). They were all synthesized from GenePharma (Shanghai, China). Lipofectamine 3000 (Invitrogen) was used to perform cell transfection.
Cell viability
After various transfections, HUVECs (approximately 3×103 cells/well) were cultured in a 96-well culture plates and then were stimulated with different dose of ox-LDL. Next, theCell Counting Kit-8 (CCK-8) (Solarbio) solution (10μL) was added to incubate with cells. Finally, microplate reader (Bio-Rad Laboratories, Richmond, CA) was used to detect the absorbance value at 490 nm.
5-ethynyl-2’-deoxyuridine (EdU) assay
Cell proliferation was estimated via measuring EdU positive cells. In brief, HUVECs were seeded into the 6-well plates (5×104/well) overnight, followed by ox-LDL exposure and transfection for 24 h. cells were labeled with 50μM EdU solution for 2 h according to the EdU Detection Kit (RiboBio, Guangzhou, China), Following being fixed and permeabilized, cell nucleus was stained using DAPI. The EdU-positive cells were determined by a fluorescence microscope.
Flow cytometry
HUVECs (5×104 cells/well) were cultured in 6-well plates for 48 h, followed by collecting the cell suspensions, and the cell suspensions were stained with 5μL of Annexin V-fluorescein isothiocyanate (FITC) and 5μL of propidium iodide (PI) (Solarbio) for 15 min in the dark. Later on, cell apoptosis was monitored by flow cytometer.
Enzyme-linked immunosorbent assay (ELISA)
ELISA Kits (Invitrogen) were used to assess the levels of Interleukin-6 (IL-6) and Tumor necrosis factor-α (TNF-α). After HUVECs were incubated with ox-LDL for 48 hours, the supernatant was collected and the contents of IL-6 and TNF-α were measured. The optical density value at 450 nm was determined by the microplate reader (BD Bioscience; San Jose, CA, USA).
Western blot
Total protein was acquired by cell lysis buffer (Beyotime, Shanghai, China). Next, 10% SDS-PAGE was used to separate protein and then blotted on polyvinylidene fluoride (PVDF) membranes (Beyotime). After blocked with 5% non-fat milk for 1 h at room temperature, the membranes were washed with tris buffered saline tween (TBST) buffer and then were labeled with primary antibodies against Bax (ab32503, 1:1000) Cleaved-caspase 3 (ab32042, 1:500), RAB22A (ab137093, 1:1000) and GAPDH (ab8245, 1:1000), followed by the protein bands were incubated with secondary antibody (ab150077, 1:3000) for 1 h at room temperature. All antibodies were acquired from Abcam (Cambridge, MA, USA). Protein bands were observed via using the enhanced chemiluminescence (Beyotime).
Dual-luciferase reporter assay
The wild-type or mutant sequences of circ_0091822 or RAB22A 3’UTR containing the binging sites of miR-661 were constructed into pmirGLO vector (Promega, Madison, WI, USA). The miR-NC or miR-661 and the reporter plasmid were co-transfected into HUVECs. Subsequently, that luciferase activity was measured by the Dual-Luciferase Reporter Assay Kit (Solarbio).
Statistical analysis
All results were assessed by GraphPad Prism 7.0. Data were displayed as mean±standard deviation. Student’s t-test or one-way ANOVA was used to analyze statistical. Pearson’s correlation analysis was utilized to analyze correlations. P-value <0.05 meant significant.
Results
Ox-LDL contributed to cell injury in HUVECs
We first investigated the effect of ox-LDL on cell damage in HUVECs. As shown in Fig. 1A–D, ox-LDL treatment incurred a strikingly increase in the level of IL-6 and TNF-α (Fig. 1A) and a markedly inhibition in cell viability and proliferation (Fig. 1B-C) as well as a notably elevation in cell apoptosis rate in HUVECs (Fig. 1D). In addition, we also examined the effect of ox-LDL on pro-apoptosis-related protein, and found ox-LDL stimulation enhanced the protein expression of Bax and Cleaved-caspase-3 in HUVECs (Fig. 1E). These data identified that ox-LDL induced the cell injury process in HUVECs.

Ox-LDL triggered HUVECs cytotoxicity. (A) IL-6 and TNF-α level was assessed by ELISA. (B) cell viability, (C) cell proliferation and (D) cell apoptosis were monitored in HUVECs stimulated with various concentrations ox-LDL (20, 40 and 80μg/mL) by CCK-8 assay, EdU assay, and flow cytometry, respectively. (E) The protein expression of Bax and Cleaved-caspase-3 were gauged by western blot. *P < 0.05, **P < 0.01, ***P < 0.001 or ***P < 0.0001.
We detected the abundance of circ_0091822 in AS patients and ox-LDL-treated HUVECs. As exhibited in Fig. 2A, circ_0091822 level was evidently elevated in AS serum samples than that in normal controls. Likewise, circ_0091822 expression was conspicuously up-regulated in ox-LDL-stimulated HUVECs (Fig. 2B). To demonstrate the circular characteristic of circ_0091822, RNase R assay was conducted and the results displayed that circ_0091822 could efficiently resist to RNase R (Fig. 2C).

Circ_0091822 was upregulated in AS serum samples and ox-LDL-induced HUVECs. (A) The level of circ_0091822 in AS serum samples and healthy control was tested by qRT-PCR. (B) QRT-PCR was employed to calculate the circ_0091822 expression in ox-LDL-treated HUVECs. (C) The stability of circ_0091822 was monitored via RNase R assay. **P < 0.01, ***P < 0.001 or ****P < 0.0001.
To study the function of circ_0091822 in ox-LDL-induced HUVECs, HUVECs were transfected with si-circ_0091822. The results demonstrated that circ_0091822 knockdown could significantly reverse the increase of circ_0091822 level induced by ox-LDL (Fig. 3A). In the study of cell function, we found that ox-LDL could notably expedite cell inflammation and inhibit cell proliferation, while promote cell apoptosis, which was effectively rescued by circ_0091822 silencing (Fig. 3B-E). Additionally, the expression of apoptosis-related markers (Bax and Cleaved-caspase-3) was detected by western blot assay to further identify the effects of circ_0091822 on the apoptosis of HUVECs. As displayed in Fig. 3F, ox-LDL exposure enhanced the expression of Bax and Cleaved-caspase-3, and circ_0091822 inhibition partially overturned this effect. Taken together, ox-LDL exposure induced the damage of HUVECs via elevating circ_0091822.

Circ_0091822 silencing attenuated ox-LDL-stimulated damage in HUVECs. HUVECs were transfected with si-circ_0091822 or si-NC before ox-LDL treatment. (A) Circ_0091822 expression was estimated by qRT-PCR when HUVECs were transfected with si-circ_0091822. (B) The concentrations of IL-6 and TNF-α were measured by ELISA. (C-D) MTT assay and EdU assay were employed to investigate the cell growth. (E) Flow cytometry was applied to calculate apoptotic cells. (F) Western blot was executed to examine the levels of Bax and Cleaved-caspase-3 in ox-LDL-treated HUVECs. **P < 0.01, ***P < 0.001 or ****P < 0.0001.
To explore the mechanism of circ_0091822 in AS, the circinteractome software was applied to predict the targeted miRNAs of circ_0091822. As shown in Fig. 4A, the binding sites between circ_0091822 and miR-661 were displayed. The overexpression efficiency of miR-661 was estimated and results revealed that miR-661 level was markedly higher in miR-661 group than that in control group (Fig. 4B). Then, dual-luciferase reporter assay was executed to further demonstrate the targeting relationship among them. Our results uncovered that miR-661 mimic could restrain the luciferase activity of the WT-circ_0091822 group, but had no effect on that in the MUT-circ_0091822 group (Fig. 4C). Moreover, we found that miR-661 was specially decreased in AS patients and ox-LDL-treated HUVECs (Fig. 4D-E). Correlation analysis revealed that the level miR-661 was negatively association with circ_0091822 expression in AS patients (Fig. 4F). Collectively, the results suggested that miR-661 was a target of circ_0091822.

Circ_0091822 acted as a sponge of miR-661. (A) The sequences of miR-661 in circ_0091822 were shown. (B) Transfection efficiency of miR-661 was estimated via qRT-PCR. (C) The binding among miR-661 and circ_0091822 was verified by dual-luciferase reporter assay. (D) The expression of miR-661 in AS patients and healthy controls was assessed through qRT-PCR. (E) QRT-PCR was performed to determine the miR-661 level in HUVECs treated with different concentrations of ox-LDL. (F) The correlation between miR-661 and circ_0091822 was evaluated by Pearson’s correlation analysis. **P < 0.01, ***P < 0.001 or ****P < 0.0001.
The above research has confirmed that circ_0091822 acted as a sponge of miR-661. Next, rescue experiments were performed to confirm whether circ_0091822 regulated ox-LDL-evoked damage in HUVEC through regulating miR-661. The results showed that circ_0091822 silencing elevated the level of miR-661 in ox-LDL-induced HUVECs, while this effect was partially abolished after anti-miR-661 introduction (Fig. 5A). Next, we evaluated the inflammatory response, proliferation and apoptosis abilities of HUVECs, the decreased effect of circ_0091822 inhibition on IL-6 and TNF-α level were remarkably rescued by anti-miR-661 in HUVECs (Fig. 5B). Furthermore, anti-miR-661 restrained the viability and proliferation induced by si-circ_0091822 and reinforced the apoptosis blocked by si-circ_0091822 in ox-LDL-induced HUVECs (Fig. 5C-F). Under ox-LDL exposure, miR-661 deficiency elevated the expression of Bax and Cleaved-caspase-3 in circ_0091822-silenced HUVECs (Fig. 5G). All data verified that circ_0091822 regulated ox-LDL-induced injury in HUVECs by targeting miR-661.

Circ_0091822 silencing weakened ox-LDL-induced damage in HUVECs by increasing miR-661. (A-G) HUVECs were transfected with si-NC, si-circ_0091822, si-circ_0091822 + anti-miR-NC or si-circ_0091822 + anti-miR-661, and then treated with ox-LDL. (A) Circ_0091822 expression was measured through qRT-qPCR. (B) The levels of IL-6 and TNF-α were estimated via ELISA analysis. (C-D) Cell proliferation was evaluated by using CCK-8 and EdU assays. (E-G) Cell apoptosis and apoptosis-related protein (Bax and Cleaved-caspase-3) were analyzed through flow cytometry and western blot analysis, respectively. *P < 0.05, **P < 0.01, ***P < 0.001 or ****P < 0.0001.
Targetscan software was used to predict the targets of miR-611 and the results showed that the 3’UTR of RAB22A could interact with miR-661 (Fig. 6A). MiR-661 was found to bind with WT-RAB22A 3’UTR but not MUT-RAB22A 3’UTR in HUVECs by conducting dual-luciferase reporter assay (Fig. 6B). The mRNA expression of RAB22A was increased in AS serum samples compared with the control (Fig. 6C), and there was a negative relationship in the expression among RAB22A and miR-661 (Fig. 6D). Western blot identified that RAB22A protein level was upregulated in a dose-dependent manner in ox-LDL-induced HUVECs (Fig. 6E). Importantly, we uncovered that circ_0091822 inhibition could retard RAB22A expression, and this influence could be reversed by miR-661 inhibitor.

RAB22A was a target for miR-661. (A) The binding sites between RAB22A 3’UTR and miR-661 were predicted through targetscan software. (B) The targeting relationship was confirmed by dual-luciferase reporter assay. (C) RAB22A mRNA level was tested via using qRT-qPCR in AS patients and healthy controls. (D) Pearson’s correlation analysis was carried out to analyze the relationship among the levels of RAB22A and miR-661. (E) RAB22A protein level was gauged through western blot in ox-LDL-induced HUVECs. (H) The protein level of RAB22A was estimated by using western blot in ox-LDL-induced HUVECs transfected with si-circ_0091822 and anti-miR-661. ***P < 0.001, ****P < 0.0001.
A key question was whether miR-661 regulated the expression of RAB22A and affected the biological function HUVECs. As expected, ox-LDL treatment increased RAB22A expression, and miR-661 overexpression decreased RAB22A expression in ox-LDL-induced HUVECs. Nevertheless, the enforced expression of RAB22A largely regained RAB22A expression (Fig. 7A). Moreover, the addition of miR-661 led to a striking suppression in IL-6 and TNF-α production (Fig. 7B), a remarkable enhancement in cell viability and cell proliferation (Fig. 7C-D), a prominent decrease on cell apoptosis, as well as a evident decrease in expression of pro-apoptotic protein, in ox-LDL-treated HUVECs (Fig. 7E-G). Nevertheless, these influences were partially overturned by the restored RAB22A level. To sum up, these effects of miR-611 overexpression were notably abolished by the restored RAB22A level in ox-LDL-induced HUVECs.

MiR-661 treatment retarded ox-LDL-induced injury in HUVECs via targeting RAB22A. (A-G) HUVECs were transfected with miR-NC, miR-661, miR-661 + pcDNA or miR-661 + RAB22A, and then treated with ox-LDL. (A) Western blot analysis was carried out to test the expression of RAB22A. ELISA analysis was implemented to evaluate the levels of IL-6 and TNF-α (B), CCK-8 assay (C), EdU assay (D), flow cytometry (E-F) were conducted to measure the cell viability, proliferation and cell apoptosis, respectively. (G) The protein expression of Bax and Cleaved-caspase-3 was examined via using western blot analysis. *P < 0.05, **P < 0.01, ***P < 0.001 or ****P < 0.0001.
AS is a complex process involving vascular injury and atherogenesis [3]. Abnormal proliferation, apoptosis and migration of HUVECs lead to the development of AS [19]. Therefore, maintaining the function of HUVECs is of great significance for the clinical treatment of AS. CircRNAs were found to participate in the development of AS. For instance, circUSP36 was elevated in HUVECs stimulated by ox-LDL, And ox-LDL expedited cell cycle arrest, apoptosis and inflammation injury, and impeded cell migration and invasion, while these effects were reversed by circUSP36 silencing [20]. Chen et al. demonstrated that circ-BANP level was increased in HUVECs induced by ox-LDL, and circ-BANP knockdown reinforced cell growth, metastasis, and angiogenesis, and hindered cell inflammation in ox-LDL-induced HUVEC, indicating that interference of circ-BANP could abate ox-LDL-evoked injury in HUVEC [21]. Our work identified that circ_0091822 level was increased in AS patients and ox-LDL-induced HUVECs, which was in line with the previous study [18]. Cell function experiments demonstrated that knockdown of circ_0091822 dramatically expedited proliferation, while remarkably hampered the inflammation and apoptosis in ox-LDL-evoked dysfunction in HUVEC. These results proved that circ_0091822 inhibition might retard the progress of AS.
Previous studies have shown that circRNAs may act as a sponge of miRNAs, and competitively binding and constraining the activity and function of miRNAs [22]. MiR-661 has been identified as a key participator in the pathogenesis of a variety of tumors and AS [23]. MiR-661 overexpression constrained breast cancer cell proliferation, metastasis and glycolysis [24]. And miR-661 also blocked ovarian cancer cell proliferation migration and invasion [25]. Sun et al. proposed that miR-661 was remarkably reduced in proliferative vascular smooth muscle cell (VSMC), and miR-661 could inhibit the proliferation and migration of VSMC, thereby inhibiting the occurrence of cardiovascular disease [23]. Likewise, we also verified a decreased miR-661 in AS patients and ox-LDL-treated HUVEC, and elevation of miR-661 receded ox-LDL-mediated HUVECs dysfunction. Importantly, miR-661 inhibition mitigated the protective effects of si-circ_0091822 on HUVECs damage under ox-LDL treatment. Thus, si-circ_0091822 confined ox-LDL-induced HUVECs dysfunction through targeting miR-661.
RAB22A was demonstrated to interact with miR-661 in HUVECs. RAB22A plays an important role in multiple cancers and participates in a variety of cell signaling pathways [26, 27]. Additionally, RAB22A is involved in the regulation of AS progression. Gao et al. reported that RAB22A augmented cell viability, metastasis, and inhibited apoptosis and inflammatory damage in ox-LDL-treated HUVECs [28]. Tang et al. clarified that miR-654-3p hampered the inflammatory responses and accelerated cholesterol efflux rate in ox-LDL-induced macrophage by targeting RAB22A, illustrating the potential roles of RAB22A in AS development [29]. In this study, we demonstrated that RAB22A was elevated in ox-LDL-induced HUVECs, and RAB22A overexpression abrogated the function of miR-661 in ox-LDL-induced damage in HUVECs. In addition, we also verified circ_0091822 could sponge miR-661 to regulate RAB22A expression in ox-LDL-mediated HUVECs.
In summary, our work identified that circ_0091822/miR-661/RAB22A network was related to AS development. Since circ_0091822 was significantly overexpressed in AS patient and ox-LDL-induced cell model and affected the disease progression of AS, circ_0091822 may serve as a new biomarker or therapeutic target in clinical studies. Repressing of circ_0091822 might be effective therapeutic strategy for AS.
Footnotes
Acknowledgments
None
Disclosure of interest
The authors declare that they have no financial conflicts of interest.
Funding
None
