Idebenone Protects against Atherosclerosis in Apolipoprotein E-Deficient Mice Via Activation of the SIRT3-SOD2-mtROS Pathway

Atherosclerosis, a chronic disease of the arteries, results from pathological processes including the accumulation and aggregation of oxidized low-density lipoprotein (oxLDL) in the vessel walls, development of neointima, formation of a fibrous cap, and migration of immune cells to damaged vascular endothelium. Recent studies have shown that mitochondrial dysfunction is closely associated with the development and progression of atherosclerosis. Idebenone, a short-chain benzoquinone similar in structure to coenzyme Q10, can effectively clear oxygen free radicals as an electron carrier and antioxidant. In the present study, we aim to investigate weather idebenone protects against atherosclerosis in apolipoprotein E-deficient (apoE−/−) mice. apoE−/− mice receiving a high-fat diet (HFD) were treated with idebenone for 16 weeks. A total of 60 mice were randomized into the following four groups: (1) HFD, (2) HFD and low-dose idebenone (100 mg/kg/d), (3) HFD and medium-dose idebenone (200 mg/kg/d), and (4) HFD and high-dose (400 mg/kg/d). Proteomic analysis was performed between the HFD and idebenone-high-dose group. Plaque analysis was carried out by histological and immunohistochemical staining. Western blot, TUNEL staining, and MitoSOX assays were performed in human umbilical vein endothelial cells (HUVECs) to investigate the SIRT3-SOD2-mtROS pathway. Histological and morphological analysis demonstrated that idebenone significantly reduced plaque burden and plaque size. Idebenone treatment effectively stabilized the atherosclerotic plaques. In mice treated with idebenone, 351 up-regulated and 379 down-regulated proteins were found to be significantly altered in proteomic analysis. In particular, the expression of SIRT3, SOD2, and NLRP3 was significantly regulated in the idebenone treatment groups compared with the HFD group both in vivo and in vitro. We further confirmed that idebenone protected against endothelial cell damage and inhibited the production of mitochondrial reactive oxygen species (mtROS) in cholesterol-treated HUVECs. We demonstrated that idebenone acted as a mitochondrial protective agent by inhibiting the activation of NLPR3 via the SIRT3-SOD2-mtROS pathway. Idebenone may be a promising therapy for patients with atherosclerosis by improving mitochondrial dysfunction and inhibiting oxidative stress.


Introduction
Atherosclerosis is the leading cause of morbidity and mortality in most developing and developed countries globally [1]. As a type of chronic inflammatory disease, atherosclerosis results from pathological processes involving aggregation of oxidized lowdensity lipoprotein (oxLDL) in the walls of blood vessels, formation of a fibrous cap, development of intimal hyperplasia, and migration of immune cells to damaged vascular endothelium [2]. Mitochondrial dysfunction has been increasingly associated with the evolution of atherosclerotic lesions due to the increased production of reactive oxygen species (ROS) and subsequent oxidative damage and impaired mitochondrial oxidative phosphorylation [3,4]. Vasquez-Trincado et al. revealed that patients with atherosclerosis exhibited higher levels of mitochondrial DNA damage in aortic walls as compared with non-atherosclerotic controls [5]. Further studies indicated that mitochondrial DNA damage led to mitochondrial dysfunction through distinct physiological mechanisms such as mitochondrial autophagy and apoptosis, directly contributing to the formation and progression of atherosclerosis [6]. Therefore, mitochondrion is an important player in the initiation and progression of atherosclerosis.
Mitochondrial oxidative damage resulting from excessive mitochondrial ROS (mtROS) contributes to the formation of fatty plaques in the arteries, also known as atherogenesis [7,8]. As a class of small molecules, mtROS plays an essential role in many biological processes essential for humans. Excess mtROS can be scavenged by the antioxidant defense system, which includes superoxide dismutase (SOD), catalase, and glutathione peroxidase [9,10]. However, 1-2% of electrons leak out and produce superoxide free radicals, including mtROS, during the oxidative phosphorylation process [11]. Excessive production of mtROS leads to the overoxidation of lipids and proteins, and has a significant impact on the functioning of cells [12]. Excess production of mtROS also induces mitochondrial dysfunction, endothelial cell damage, and oxLDL aggregation in smooth muscle and vascular endothelial cells, thereby accelerating the progression of atherosclerosis [3,13]. Moreover, mtROS contributes to the inflammatory response by increasing the expression of certain inflammatory and adhesion factors [14,15]. Hence, mtROS plays an important role in the pathophysiology of atherosclerosis by oxidation of cholesterol and inflammatory response of vascular endothelial cells.
Idebenone is a synthetic quinone structurally similar to coenzyme Q10 (CoQ10). However, it has fewer and shorter lipophilic tails than CoQ10. It is a novel antioxidant that effectively functions in hypoxic microenvironments [16]. By preventing lipid peroxidation and removing oxygen-derived free radicals, idebenone protects cell membranes and mitochondria from oxidative damage [17]. Therefore, there are great social health benefits to investigating whether idebenone can delay the development and progression of atherogenesis.
In a previous study, we showed that idebenone could protect cells from mitochondrial damage and cell apoptosis caused by ox-LDL via GSK-3β/β-catenin signaling in human umbilical vein endothelial cells (HUVECs) [18]. We found that idebenone reduced endothelial cell injury caused by oxidative stress. In the present study, we aim to investigate whether idebenone can protect against atherosclerosis in apolipoprotein E-deficient (apoE−/−) mice. We further perform functional analyses to explore the potential biological mechanisms. Our findings indicate that idebenone acts as a mitochondrial protective agent to mitigate the process of initiation and progression of atherosclerosis by inhibiting NLPR3 activation through the SIRT3-SOD2-mtROS signaling pathway. Idebenone may thus be a promising agent in the prophylaxis and treatment of atherosclerosis.

Materials and Methods
This study was approved by the Brain Science Research Institute and the Ethics Committee of Qilu Hospital of Shandong University (Jinan, China).

Drug and Solution Preparation
Idebenone powder was obtained from Qilu Pharmaceutical Co., Ltd. (Shandong, China). The idebenone, with purity > 98%, was dissolved in absolute ethyl alcohol to prepare a stock solution of 50 mM for the in vitro experiments. For in vivo studies, idebenone was dissolved in corn oil for daily intragastric administration. High-fat diet (HFD) containing 0.25% cholesterol/15% cocoa butter was purchased from Beijing Keao Xieli Limited Company (Beijing, China).

Animal and Tissue Preparation
A total of 60 apoE−/− mice (male, 8 weeks old) were obtained from Beijing Vital River Laboratory Animal Technology Co., Ltd. (Beijing, China). The mice were kept in a controlled environment with a constant temperature of 22 ± 2°C and humidity of 50-60%. The mice were subjected to a 12-h light-dark cycle and had free access to food and water. The animal experimental protocol complied with the Animal Management Rules of the Chinese Ministry of Health and was approved by the Animal Care Committee of Shandong University.
All the mice were given a HFD. One week after starting the HFD, the mice were randomly divided into four groups (n = 15 per group): HFD, HFD with low-dose idebenone (IDE-L, 100 mg/kg/d), HFD with medium-dose idebenone (IDE-M, 200 mg/kg/d), and HFD with high-dose idebenone (IDE-H, 400 mg/kg/d). Idebenone was given by intragastric administration. Sixteen weeks later, all animals were weighed before the following experiments. The mice were subjected to fasting for 4 h before blood was drawn for lipid analysis and tissue assays. After blood sampling, the mice were anesthetized by 10% chloral hydrate (0.001 ml/g) and euthanized by 10% chloral hydrate (0.01 ml/g). In each group, the aorta, brain, and liver tissues from seven mice were extracted, flash-frozen with liquid nitrogen, and stored at −80°C for future experiments. Tissues from the other eight mice were fixed for immunohistochemistry and Oil Red O staining. After the above procedures, the cardiovascular system was rapidly perfused with 0.9% saline, followed by 4% paraformaldehyde. The aortic roots were embedded in optimal cutting temperature (OCT) compound for cryosectioning. Serial cross-sections of 5-μm thickness were obtained and stored at −20°C for future immunohistochemistry studies. Frozen tissue sections used for analyses were determined by blinded and unconscious selection to avoid bias.

Cell Culture
HUVECs were obtained from the umbilical cords of healthy mothers following their deliveries. The freshly isolated HUVECs from the human umbilical cord were cultured with ECM supplemented with 10% FBS, 1% penicillin/streptomycin, and endothelial cell growth supplement (ECGS) in a humidified incubator at 37°C with 5% CO 2 . The HUVECs displayed a cobblestone morphology and stained positively for the common von Willebrand factor (vWF) endothelial cell marker. The cells were passaged four to eight times before experiments. The effects of 0.2 μM idebenone alone on HUVECs was tested before the experiment. When the HUVECs reached approximately 50-60% confluence, they were pretreated with ECM with or without 0.2 μM idebenone for 3 h [18], followed by exposure to 10 μM cholesterol with or without 0.2 μM idebenone for an additional 24 h.
siRNA Assay SIRT3 siRNAs (human, sc-61555), along with control siRNA (sc-44230) and siRNA transfection reagent (sc-29528), were purchased from Santa Cruz Biotechnology (Dallas, TX, USA). Endothelial cells were transfected with 100 pmol/L siRNA according to the manufacturer's protocol. Forty-eight hours after transfection, the HUVECs were pretreated with or without 0.2 μM idebenone for 3 h, followed by exposure to 10 μM cholesterol with 0.2 μM idebenone for an additional 24 h. Cells were then harvested and subjected to further analyses.

Western Blot Analysis
Western blot was performed according to a previously published protocol [18]. Briefly, HUVECs and aorta tissues were lysed using radioimmunoprecipitation assay (RIPA) lysis buffer. The proteins were separated on 8-15% sodium dodecyl sulfate-polyacrylamide gel by electrophoresis, and then transferred to methanol-activated polyvinylidene fluoride membranes. The membranes were blocked in 5% skim milk for 60 min at 37°C and incubated with primary antibodies overnight at 4°C. The membranes were then washed three times with TBST and incubated with horseradish peroxidase (HRP) secondary antibodies for 60 min. Protein bands were visualized with Millipore ECL Plus reagent and imaged on a Tanon 5500 Imaging Analysis System. The intensity of the bands was quantified using ImageJ software.

Histopathological and Immunohistochemical Analysis
The en face aorta was stained with Oil Red O to assess the overall burden and distribution of atherosclerosis as previously described [19]. The aortic roots, which are predilection sites for atherosclerosis, were cut into 5-μm-thick cross-sections and stained with hematoxylin and eosin (H&E), Oil Red O, Sirius Red, and Masson's trichrome. The sections were incubated with primary antibodies against CD68, smooth muscle cells (SMCs), and tumor necrosis factor-α (TNF-α). Immunohistochemical staining was performed in accordance with standard protocols of the Shandong Province Joint Key Laboratory of Translational Cardiovascular Medicine of Qilu Hospital. In brief, the aortic root sections were incubated with 3% hydrogen peroxide for reducing endogenous peroxidase activity and blocked with 5% goat serum for 30 min at room temperature. Next, the sections were incubated with primary antibodies overnight at 4°C, followed by washing three times with phosphate-buffered saline (PBS). The sections were then incubated with the secondary antibodies for 60 min at room temperature. A 3′3-diaminobenzidine (DAB) kit was used for detection. The primary antibodies were anti-CD68 (1:150,), α-SMA (1:200), and TNF-α (1:200). Stained sections were analyzed using Image-Pro Plus version 6.0 software (Media Cybernetics, Silver Springs, MD, USA). The en face analysis of the aorta sections was performed as previously described [19]. Each field was counted three times, and the average value was calculated as the final result. All measurements were performed under the same conditions to ensure consistency.

Proteomic Analysis
The tissues between the aortic root and the abdominal aorta were collected from the HFD and IDE-H groups (Fig. S1). Collected tissues were sonicated three times on ice using a high-intensity ultrasonic processor (Ningbo Scientz Biotechnology Co., Ltd., Zhejiang, China) in lysis buffer consisting of 8 M urea and 1% protease inhibitor cocktail. The samples were then centrifuged at 12,000×g for 10 min at 4°C, after which the supernatant was collected and the protein concentration was determined using the BCA kit. Next, the protein solution was reduced with 5 mM dithiothreitol for 30 min at 56°C and alkylated with 11 mM iodoacetamide for 15 min at room temperature in darkness. The protein sample was then diluted by adding 100 mM triethylamine-carbonic acid buffer (TEAB). Trypsin was added at a 1:50 trypsin-to-protein mass ratio for the first digestion overnight and then 1:100 trypsin-to-protein mass ratio for a second 4-h digestion. The peptide was then desalted with the Strata-X C18 solid-phase extraction (SPE) column (Phenomenex, Torrance, CA, USA) and dried in a vacuum. The peptide was further reconstituted in 0.5 M TEAB and processed with a TMT kit/iTRAQ kit. Briefly, one unit of the TMT/iTRAQ reagent was thawed and reconstituted in acetonitrile. The peptide mixtures were then incubated for 2 h at room temperature and then pooled, desalted, and dried by vacuum centrifugation.
The quantitative values of proteins in multiple replicates were obtained by a protein quantification experiment, the first step of which is to determine the expression ratio of each protein between two samples. We calculated the average value in multiple replicates, and then used the ratio of the average values between two samples as the final expression ratio. The second step is to calculate the significance p value of differentially expressed proteins between two samples by t test. We first performed log2 transform on the relative quantitative value of each sample in multiple replicates (making the data conform to the normal distribution), and the p value was then calculated using a two-sample two-tailed t test. Proteins with p values < 0.05 and expression ratios > 1.3 were regarded as up-regulated, while proteins with p values < 0.05 and expression ratios < 1/1.3 were regarded as down-regulated.

Cell Viability Assay
The CCK-8 assay was used to evaluate cell viability. Briefly, 3 × 10 3 HUVECs were seeded into 96-well culture plates for 24 h. Next, cells were treated with or without 0.2 μM idebenone for 3 h, and were then exposed to 10 μM cholesterol with or without 0.2 μM idebenone for an additional 24 h, after which 10 μl of the CCK-8 assay solution was added to the cells for 2 h at 37°C. The cell viability was detected using the EnSpire ® Multimode Plate Reader (PerkinElmer, Waltham, MA, USA) by recording the optical density at 450 nm. Each experiment was performed in triplicate.

JC-1 (MitoScreen) and MitoSOX Assays
The depolarization of the mitochondrial membrane potential (MMP) was determined using the JC-1 kit. The HUVECs were seeded in 96-well plates at a density of 3 × 10 3 per well for 24 h. The next day, cells were treated with or without 0.2 μM idebenone for 3 h, and were then exposed to 10 μM cholesterol with or without 0.2 μM idebenone for an additional 24 h, after which the cells were washed three times with PBS. For the MitoScreen assay, 100 μl of the JC-1 working solution was added. After incubation at 37°C for 15 min, cells were washed twice with the JC-1 staining solution and then were assessed on a microplate reader. To perform the MitoSOX red stain assay, cells were incubated with 5 μM MitoSOX red and MitoTracker at 37°C for 15 min. A fluorescence microscope was used to detect the signals with excitation and emission wavelengths of 510 and 580 nm, respectively (SP8, Leica Biosystems, Wetzlar, Germany).

Measurement of MDA and SOD Activity
HUVECs from three groups (control group, 10 μM cholesterol group, 10 μM cholesterol + 0.2 μM idebenone group) were washed with PBS and then lysed in lysis buffer. The lysates were collected and centrifuged at 12,000×g for 15 min at 4°C. Total proteins were then extracted with RIPA lysis buffer, and the protein concentrations were determined by the BCA method. The supernatants were subsequently collected and the concentration of malondialdehyde (MDA) was examined using the MDA assay kit, which indirectly represents lipid

Apoptosis Assay
Cell apoptosis assay was assessed using DAPI/terminal deoxynucleotidyl transferase (TdT)-mediated dUTP nickend labeling (TUNEL) staining. The treated cells were fixed with 4% paraformaldehyde for 30 min, incubated with 0.3% Triton X100 for 5 min, and then stained with DAPI or TUNEL working solution for 5 min. Next, samples were washed with PBS for 5 min three times, and then observed under a fluorescence microscope (Leica Microsystems, Wetzlar, Germany).

ELISA
The supernatants from each group of HUVECs and mice aorta tissue were centrifuged. The assays were conducted utilizing commercial enzyme-linked immunosorbent assay (ELISA) kits (Beyotime, Jiangsu, China) according to the manufacturer's instructions. A microplate absorbance reader (PerkinElmer, Waltham, MA, USA) was employed to measure the absorbance at 450 nm. Standard curves were applied to calculate the analyte concentrations.

Statistical Analysis
Numerical data are expressed as means ± SD based on three or more independent experiments. Differences between groups were determined and analyzed using one-way analysis of variance (ANOVA) followed by the Tukey post hoc test with the SPSS 16.0 software package (SPSS Inc., Chicago, IL, USA). P values < 0.05 were considered to be statistically significant.

Results
Body Weight, Glucose Concentration, and Serum Lipid Profile Levels Are Not Altered by Idebenone Treatment Table 1 shows the body weight, blood lipid analyses, and blood glucose levels of mice from this study. The statistical analysis revealed no significant differences in blood lipid levels, including triglycerides (TG), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), or lowdensity lipoprotein cholesterol (LDL-C), among the four groups. In addition, idebenone treatment did not cause any significant changes in body weight or blood glucose levels as compared with the HFD group.

Idebenone Suppresses the Formation of Atherosclerotic Plaque in apoE−/− Mice
To investigate the effect of idebenone in atherosclerosis, three different concentrations of idebenone were intragastrically injected in the apoE−/− mice. The en face lesion analysis showed that, compared with the HFD group, the relative lesion area was significantly reduced in all idebenone treatment groups, especially in the IDE-H group (Fig. 1a and c). When compared with the HFD group, the cross-sectional plaque area of the aortic sinus was significantly reduced in the idebenone treatment groups, especially in the IDE-H group ( Fig. 1b and d). Taken together, our study results demonstrate that in apoE−/− mice, idebenone attenuated the formation of atherosclerotic plaques in the entire aorta and aortic root.

Idebenone Treatment Increases the Stability of Atherosclerotic Plaques in apoE−/− Mice
To further assess whether idebenone could alter the components of atherosclerotic plaques, histological and immunohistochemical staining of the aortic root were performed. As compared with the HFD group, the levels of lipids and macrophages in the atherosclerotic plaques were significantly lower in the idebenone treatment groups, especially in the IDE-H group (Fig. 2a, e, h, and k). However, the collagen content in the plaques was significantly higher in the treatment groups (Fig. 2b, c, d, l, and j). Similarly, the intraplaque content of smooth muscle cells was significantly increased in the three treatment groups (Fig. 2f and l). We also confirmed that idebenone suppressed the expression of TNF-α in a dosedependent manner (Fig. 2g and m). Thus, these findings demonstrate that idebenone altered the composition of atherosclerotic plaques in a dose-dependent manner, suggesting that idebenone can effectively stabilize intravascular plaques and inhibit the development and progression of atherosclerosis.

Proteomic Analysis between the HFD and the IDE-H Groups
To further understand how idebenone protects against atherosclerosis and the mechanisms involved in this process, proteomic analysis was performed between the HFD (n = 3) and IDE-H (n = 3, concentration = 400 mg/kg/d) groups, as the IDE-H group displayed the maximum effectiveness against HFD. As shown in Fig. 3a, 351 up-regulated and 379 downregulated proteins were found to be significantly altered. The differentially expressed proteins were located primarily in the mitochondria, extracellular matrix, and cytoplasm (Fig. 3b). A directed acyclic graph (DAG) of each Gene Ontology (GO) category for differentially expressed proteins is shown in Fig.  3c. Statistical distribution charts of the differentially expressed proteins under each GO category (second level) are shown in Fig. 3d. GO enrichment bubble plots of the differentially expressed proteins in the three categories and comprehensive heat maps for the cluster analysis are shown in Figs. S2-7. We divided the differentially expressed proteins into four parts, Q1 to Q4, according to their differential expression ratios.

Through GO and KEGG [Kyoto Encyclopedia of Genes and
Genomes] enrichment analysis, we found that idebenone ameliorated the biological processes of oxidative stress and inflammatory reaction (Fig. S2-7). Idebenone is mainly involved in biological processes associated with respiratory electron transport chain, oxidoreduction coenzyme metabolism, and the like, which can reduce oxidative stress. In addition, the change in IL-1β and caspase-1 expression demonstrated that inflammatory reaction also plays an important role in this process. The Clusters of Orthologous Groups of proteins (COG)/euKaryotic Orthologous Groups (KOG) functional classification chart of these proteins shows that they are primarily involved in five processes, including energy production and conversion, amino acid transport and metabolism, lipid transport and metabolism, general function prediction, and signal transduction (Fig. S8). As idebenone is a mitochondrial protectant and antioxidant, greater attention was given to the differentially expressed proteins associated with mitochondrial function. Thus, we further verified the expression of some proteins associated with both atherosclerosis and mitochondrial oxidative stress. The samples used for western blot were the same as those used for the proteomic analysis to ensure consistency between the studies. Surprisingly, three of them, SIRT3, SOD2, and NLRP3, were shown to be differentially expressed by western blot (Fig. 3e-h). It is known that excess mtROS can induce NLRP3-related inflammatory responses. However, SOD2, which is primarily activated by the deacetylation of specific conserved lysine residues and catalyzed by SIRT3, plays an essential role in the clearance of the mtROS. The information regarding these three proteins is provided in Table 2. Hence, the above results demonstrate that idebenone inhibited the initiation and progression of atherosclerosis likely through modulation of the SIRT3-SOD2-mtROS signaling pathway.

Idebenone Activates the SIRT3-SOD2-mtROS Pathway and Inhibits NLRP3-Induced Inflammatory Responses in apoE−/− Mice
The involvement of the SIRT3-SOD2-mtROS pathway was further verified in the apoE−/− mice. The protein levels of SIRT3, FOXO3A, and SOD2 were detected by western blot in aortic tissue samples. We found that idebenone significantly up-regulated the expression of SIRT3 (Fig. 4a-b), FOXO3A ( Fig. 4a and c), and SOD2 ( Fig. 4a and d)   and triggering an inflammatory response [20]. As expected, both NLRP3 and caspase1-p20 were significantly downregulated in the aortic tissues of apoE−/− mice receiving different concentration of idebenone (Fig. 4a, e, and f). Downregulation of IL-1β was also found in the idebenone treatment groups (Fig. 4a and g). In addition, idebenone suppressed the expression of NLRP3 in the plaques in a dosedependent manner (Fig. 4h and k). The expression of SIRT3 ( Fig. 4h and I) and SOD2 (Fig. 4h and j) in the plaques was significantly higher in the idebenone treatment groups than in the HFD group. Expression of IL-1β was also decreased after the mice were treated with idebenone as shown by ELISA assay (Fig. S9). These findings indicate that the protective effects of idebenone against atherosclerotic plaques are dependent on the activation of the SIRT3-SOD2-mtROS pathway and the alleviation of NLRP3 inflammasome-mediated inflammatory responses.

Idebenone Protects against Endothelial Cell Damage and Inhibits the Production of mtROS in Cholesterol-Mediated HUVECs
MitoSOX red is a highly selective fluorescent probe for the detection of mtROS generated within the mitochondria. Our results revealed that, the fluorescent intensity of mtROS in the mitochondria was significantly decreased in HUVECs pretreated with idebenone ( Fig. 5a and b), confirming that idebenone inhibit endothelial mtROS generation. We next investigated the mechanisms responsible for the protective effects of idebenone against endothelial cell damage in HUVECs. As shown in Fig. 5c-e, 10 μM cholesterol significantly decreased the expression of Bcl-2 and increased the expression of Bax, whereas idebenone treatment produced the opposite effects. Caspase-3 is a key factor in the initiation of apoptosis. As shown in Fig. 5c-g, although cholesterol significantly increased the protein levels of caspase-3 and cleaved caspase-3, idebenone pretreatment suppressed their expression. The effects of idebenone on cell viability was next examined using the CCK-8 assay. We first tested the effects of idebenone at 0.2 μM alone on cell viability and ROS production. HUVECs were treated with 0.2 μM idebenone for 24 h before CCK-8 and MitoSOX assays. No significant differences in cell viability or MitoSOX staining were found in HUVECs treated with idebenone alone when compared with normal control (Fig. S10). Next, we tested the effects of idebenone induced by cholesterol. After 24 h of incubation, HUVECs with 10 μM of cholesterol exhibited noticeable changes in cell shrinkage, suspension, and retraction as compared with the control group. However, 3 h of pretreatment with 0.2 μM idebenone significantly increased cell viability (Fig. 5h). These results indicate that idebenone effectively prevented cholesterol-induced damage to endothelial cells.
The JC-1 dye is commonly used to detect mitochondrial potential, as it preferentially enters the mitochondria due to its highly negative mitochondrial membrane permeabilization (MMP). Mitochondrial membrane potential in the HUVECs was next detected with the JC-1 kit after cholesterol and idebenone co-treatment for 24 h. The ratio of red/green fluorescence is dependent on the membrane potential. As JC-1 accumulates in polarized mitochondria with negative MMP, the dye shows red fluorescence. However, when MMP is reduced, JC-1 aggregates turn into monomers and the fluorescence appears green. Our results showed that pretreatment with idebenone significantly increased the ratio of aggregates to monomers when compared with cells treated with cholesterol alone (Fig. 5I). Moreover, incubation of HUVECs with 10 μM cholesterol for 24 h caused a significant increase in MDA content and markedly decreased SOD2 activity. However, pretreatment with 0.2 μM idebenone for 3 h markedly attenuated the effects caused by cholesterol ( Fig. 5j and k).

Idebenone Inhibits Cholesterol-Induced Injury of HUVECs by Modulating SIRT3 Activation and NLRP3-Related Inflammatory Reactions
We next investigated whether idebenone could protect endothelial cells from damage through activation of the SIRT3-SOD2 pathway and inhibition of NLRP3-related inflammatory reactions in vitro. By TUNEL assay, we found that 0.2 μM idebenone pretreatment significantly attenuated cholesterolinduced apoptosis (Fig. 6a and b). In parallel, the protective effect of idebenone on apoptosis was inhibited when SIRT3 was knocked down (Fig. 6a and b).
The downregulation of NLRP3 ( Fig. 6c and g) and upregulation of SIRT3 (Fig. 6a and d), FOXO3A (Fig. 6a and e), and SOD2 ( Fig. 6a and f) were detected in HUVECs pretreated with idebenone. Caspase-1 is a critical factor in inflammatory response and apoptosis. We demonstrated that cholesterol significantly increased the protein levels of caspase1-p20 in cultured HUVECs. However, caspase1-p20 was suppressed by idebenone ( Fig. 6c and h). Decreased expression of IL-1β was also detected in the idebenone treatment groups (Fig. 6a, l and Fig. S9). However, knockdown of SIRT3 with si-SIRT3 led to decreased expression of FOXO3A ( Fig. 6c and e) and SOD2 (Fig. 6c and f) and increased endothelial cell apoptosis as demonstrated by increased levels of NLRP3 ( Fig. 6c and g) and caspase1-p20 ( Fig. 6c and h), further proving that idebenone exerted protective effects against endothelial cell damage via activation of the SIRT3-SOD2 pathway. Thus, our in vitro findings were consistent with in vivo studies showing that several processes, including activation of the SIRT3-SOD2 mitochondrial pathway and inhibition of the associated inflammatory responses, may play a vital role in the protection of cholesterol-induced injury in HUVECs after idebenone treatment.

Discussion
According to previous studies, oxLDL and cholesterol can induce the production of ROS, which leads to mitochondrial dysfunction, cell apoptosis, and oxidative stress in the arteries [7]. ROS production occurs primarily in the mitochondria, and mitochondrial dysfunction leads to excessive ROS production, which results in vascular endothelial cell dysfunction and atherosclerosis [21,22]. Ballinger et al. demonstrated that the accumulation of mtROS and DNA damage can increase the progression of atherosclerotic lesions in human arterial specimens and mouse models of atherosclerosis [23]. In addition, Yu et al. showed that mtDNA damage in circulating cells and vessel walls was associated with an increased risk of developing atherosclerosis [24]. Thus, mitochondria play a vital role in the association of atherosclerosis with oxidative stress, inflammatory responses, and endothelial dysfunction. Hence, the early treatment of mitochondrial dysfunction is of vital importance in the prevention and treatment of atherosclerosis.
Most mitochondrial protective agents are known to modulate mitochondrial dysfunction and prevent atherogenesis. Mitoquidone is a new type of ROS scavenger that acts primarily on mtROS and can effectively diminish the formation of free radicals without affecting oxidative phosphorylation or mitochondrial function. In mouse models, mitoquidone was shown to decrease cell proliferation and macrophage content in arterial plaques, while also inhibiting certain characteristics of metabolic syndrome [25]. As another ROS scavenger, idebenone is a short-chain benzoquinone with a structure similar to that of CoQ10 [26]. The pharmacological properties of idebenone have been shown to be effective in treating neurodegenerative diseases such as Alzheimer's disease, Parkinson's disease, and Friedreich's ataxia [27]. In our previous study, we demonstrated that idebenone could prevent oxLDL-induced mitochondrial dysfunction in vascular endothelial cells [18]. In the present study, we further confirmed that idebenone could effectively inhibit the development and progression of atherosclerosis both in vivo and in vitro.
Idebenone functions as an antioxidant and electron carrier, which allows for clearance of oxygen free radicals [28]. Idebenone was previously shown to inhibit complex I in the electron transport chain (ETC), while also supplying idebenol for electron transfer to complex III [29,30]. In addition, some studies have reported that idebenone-dependent metabolic pathways can transfer equal amounts of energy directly from the cytoplasm into the mitochondrial respiratory chain (MRC) [31]. Since idebenone can easily cross the blood-brain barrier, it can be readily broken down into its metabolic products, including QS-10, QS-6, and QS-4 [32]. QS-10 is an electron acceptor with a strong affinity for NQO1, which is a key factor in protecting against oxidative stress and preventing the development of atherosclerosis [33]. Additionally, the pretreatment of cells with idebenone can attenuate the production of pro-inflammatory factors by inhibiting the MAPK and NF-κB signaling pathways, two important pathways in atherosclerosis [34]. Idebenone is also a known PPARα ligand that can protect endothelial cells and stabilize fatty plaques through various mechanisms [26], and has an excellent safety profile in the clinic [34].
Our study demonstrated that idebenone could inhibit the development and progression of atherosclerosis in apoE−/− mice. Morphological and histological analyses revealed a reduction in the plaque burden of apoE−/− mice treated with idebenone as compared with the HFD group. The intravascular plaque size was significantly reduced as well, and the composition of the plaque was stabilized in the idebenone treatment groups.
Our proteomic analyses further elucidated the molecular mechanisms responsible for the protective effect of idebenone. We found that some pro-atherosclerosis proteins and pathways such as the initiation of inflammatory response, activation of immune response, and mtROSmediated mitochondrial dysfunction were inhibited by idebenone. We determined that idebenone functioned by increasing SIRT3 and SOD2 expression, which subsequently inhibited endothelial mtROS generation and NLRP3-related inflammatory response. Based on these results, we propose that idebenone inhibits the development of atherosclerosis via two mechanisms. First, oxLDL and mtROS activate the NLRP3 inflammasome [35]. As a mitochondrial protectant, idebenone may deliver leaked electrons and reduce the generation of mtROS and subsequent NLRP3 inflammasome, while producing a systemic antioxidant effect by minimizing lipid peroxidation. Secondly, idebenone can increase intracellular SIRT3 levels, which can alter the acetylation of mitochondrial antioxidant enzymes including glutathione peroxidase, isocitrate dehydrogenase 2, and SOD2. SIRT3 can also regulate the expression of SOD2 by activating the transcription factor FOXO3A, which affects its ability to scavenge mtROS and suppress NLRP3 inflammasome formation [36].
SIRTs are a family of sirtuins, which are NAD + -dependent histone deacetylases [37]. Activation of SIRTs appears to have beneficial effects for anti-oxygenation and lipid metabolism [38]. Sirtuin3 (SIRT3), one of the antioxidants of the Sir2 family, plays an essential role in the regulation of mitochondrial metabolism via removal of excess mtROS and activation of specific metabolic enzymes in the mitochondria [39]. SIRT3 is a substrate involved in a variety of biological processes including mitochondrial membrane potential maintenance, electron transport chain flux, mitochondrial dynamics, energy metabolism, and mtROS production and clearance [20]. Dysfunctional mitochondria may lead to the development of certain atherosclerotic diseases, such as myocardial infarction, through the inhibition of SIRT3 [40]. SIRT3 can deacetylate a number of mitochondrial-related proteins, including SOD2, which can limit the accumulation of mtROS in the mitochondria [41]. SIRT3 directly combines with and deacetylates SOD2, thereby strengthening the activity of SOD2 and subsequently affecting mtROS homeostatic function [42]. Furthermore, excess mtROS can stimulate the formation of the NLRP3 inflammasome, which triggers a series of inflammatory reactions [43,44]. The NLRP3-related inflammasome is a protein complex consisting of the inactive procaspase-1, adaptor protein apoptotic speck-like protein, and the pattern recognition receptor NLRP3. The NLRP3 inflammasome can activate IL-1β cytokines and participate in the inflammatory response and natural immunity [45]. The NLRP3 inflammasome is a vital inflammation signaling molecule that can be activated by various molecular mechanisms, including mtROS. As such, mtROS may stimulate the assembly of the NLRP3 inflammasome complex, thereby activating caspase-1 and stimulating the secretion of inflammatory cytokines such as IL-1β and IL-18 [46,47]. In the carotid atherosclerotic plaques of patients undergoing endarterectomy, there is increased expression of NLRP3, caspase1-p20, IL-1β, and IL-18, all of which are higher in vulnerable plaques than stabilized plaques [48]. Inflammatory cells play a leading role in early atherosclerotic lesions, as they accelerate the formation of plaques through effector molecules, which results in the activation of inflammation and induction of cardiovascular disease [49,50]. The physiological effects of the SIRT3-SOD2-NLRP3 pathway were previously shown in dysfunctional endothelial cells of the aorta, providing a potential direction in the prevention of cardiovascular diseases associated with atherosclerosis [39].
The production of intracellular oxygen free radicals arises mainly from mitochondria. In our experiments, we evaluated mtROS generation using MitoSOX red, which can differentiate the generation of ROS from the cytoplasm and mitochondria. As expected, idebenone treatment significantly inhibited the generation of mtROS. In addition, idebenone mitigated mtROS production as a specific mtROS scavenger [51], therefore inhibiting mtROS-induced activation of the NLRP3 inflammasome and ultimately attenuating the intra-plaque inflammatory reaction. Thus, activation of the NLRP3 inflammasome in vascular endotheliocytes by cholesterol is a key molecular event leading to endothelial cell injury. Endothelial cell dysfunction through the development of atherosclerosis may be mediated through NLRP3 inflammasomedriven signaling pathways. Our results demonstrate that idebenone attenuated the development of atherosclerotic plaques by inhibiting NLRP3-inflammatory pathways trigged by cholesterol-induced mtROS overproduction, providing theoretical support for the development of potential future drug targets (Fig. 7). Considering that idebenone has been used in clinical trials without reports of serious adverse effects, we believe that idebenone represents a promising agent for the prevention of atherosclerosis. Additional randomized controlled clinical trials are needed to confirm our findings.