Acacetin exerts antioxidant potential against atherosclerosis through Nrf2 pathway in apoE−/− Mice

Oxidative stress has a considerable influence on endothelial cell dysfunction and atherosclerosis. Acacetin, an anti‐inflammatory and antiarrhythmic, is frequently used in the treatment of myocarditis, albeit its role in managing atherosclerosis is currently unclear. Thus, we evaluated the regulatory effects of acacetin in maintaining endothelial cell function and further investigated whether the flavonoid could attenuate atherosclerosis in apolipoprotein E deficiency (apoE−/−) mice. Different concentrations of acacetin were tested on EA.hy926 cells, either induced or non‐induced by human oxidized low‐density lipoprotein (oxLDL), to clarify its influence on cell viability, cellular reactive oxidative stress (ROS) level, apoptotic ratios and other regulatory effects. In vivo, apoE−/− mice were fed either a Western diet or a chow diet. Acacetin pro‐drug (15 mg/kg) was injected subcutaneously two times a day for 12 weeks. The effects of acacetin on the atherosclerotic process, plasma inflammatory factors and lipid metabolism were also investigated. Acacetin significantly increased EA.hy926 cell viability by reducing the ratios of apoptotic and necrotic cells at 3 μmol/L. Moreover, 3 μmol/L acacetin clearly decreased ROS levels and enhanced reductase protein expression through MsrA and Nrf2 pathway through phosphorylation of Nrf2 and degradation of Keap1. In vivo, acacetin treatment remarkably attenuated atherosclerosis by increasing reductase levels in circulation and aortic roots, decreasing plasma inflammatory factor levels as well as accelerating lipid metabolism in Western diet‐fed apoE−/− mice. Our findings demonstrate the anti‐oxidative and anti‐atherosclerotic effects of acacetin, in turn suggesting its potential therapeutic value in atherosclerotic‐related cardiovascular diseases (CVD).


| INTRODUC TI ON
Atherosclerosis-related cardiovascular disease (CVD) is a leading cause of mortality in developed and developing countries.
Atherosclerosis is a chronic inflammatory condition characterized by dyslipidaemia and oxidative stress. 1 Low-density lipoprotein (LDL) is considered to be a key molecular in every stage of atherosclerosis. Specifically, it is the oxidized LDL (oxLDL) particles, as opposed to normal LDL, that have a pathogenic influence, activating the vascular intima and subsequently initiating atherosclerosis. In addition to their lipid-lowering effects in clinical therapies, statins, according to many studies, possess anti-oxidative ef- Acacetin, a natural flavone widely distributed in plant pigments, has been shown by many studies to have multiple beneficial biological effects in cancers, 9,10 cardiac remodelling, 11 microbial infections, 12 inflammation 13 and oxidative stress. 14 In human umbilical vein endothelial (HUVEC) cells, acacetin inhibited E-selectin expression through the p38/MAPK pathway and activation of the nuclear factor NF-κB. 15 Acacetin has also demonstrated an ability to down-regulate inflammatory iNOS and COX-2 gene expression in RAW264.7 cells by inhibiting the activation of NF-κB through interfering with the PI3K/Akt/IKK and MAPK pathways. 16 Moreover, our previous study found that AMPK-mediated Nrf2 activation through acacetin is involved in cardiomyocyte protection against hypoxia/ reoxygenation injury by its anti-oxidative, anti-inflammatory and anti-apoptotic effects. 17 Collectively, this evidence clearly illustrates the involvement of acacetin in oxidative stress and inflammation-associated conditions. Methionine, a thiol amino acid, is not only an initiation amino acid but also a sensitive target for oxidants, giving it vital roles in some critical signalling pathways. Met is easily oxidized to MetO, which is reduced back to Met exclusively by the intracellular methionine sulphoxide reductase (Msr) system. Specifically, MsrA, an enzyme involved in this system, reduces methionine-S-sulfoxide (MetSO) to Met. 18 We previously demonstrated that exogenous reconstructed MsrA protein plays a protective role against oxidative stress as well as inflammation in RAW264.7 cells, and attenuates the atherosclerotic process in western diet-fed apoE deficiency (apoE −/− ) mice. 19 However, it is currently still unclear precisely how MsrA exerts protection against oxidative stress.
The goal of this study is to investigate whether acacetin can protect against oxidative stress in humans and attenuate atherosclerosis in apoE −/− mice. Moreover, we will attempt to reveal the mechanisms underlying the role of acacetin in the MsrA-and Nrf2-related pathways, aiming to provide evidence of its potential therapeutic role in atherosclerosis-related CVD.

| Cell culture
Human endothelial cell line EA.hy926 cells were obtained from ATCC.

| Western blot analysis
Harvested cells or frozen tissues were lysed by RIPA with 1% protease and phosphatase inhibitors (Roche) for Western blot assay.
The appropriate amount of proteins was loaded and separated by 10% or 12% SDS-PAGE and transformed onto PVDF membrane.
Protein expression was detected by primary antibodies followed by HRP-conjugated secondary antibodies. Signals were detected using an enhanced chemiluminescence kit (ECL, GE Healthcare) and captured by a chemiluminescence detection system (FlouChem E). The band densitometry was analysed by Image J software (NIH).

| Determination of basic biochemical parameters in Western diet-fed mice
After 12 weeks of experimental procedures, blood samples were collected from mice after overnight fasting by retro-orbital venous plexus puncture. Plasma was immediately separated by centrifugation at 1000× g for 10 minutes at 4°C. Total cholesterol (TC), triglyceride (TG), high-density lipoprotein cholesterol (HDL-C), lowdensity lipoprotein cholesterol (LDL-C) and apolipoprotein AI (apoAI) levels were measured by enzymatic colorimetric methods using Mind Bioengineering kits. The remaining plasma was used for determination of IL-6, IL-10, TNFα and MCP-1 levels by ELISA kits according to the manufacturer's instructions. Frozen mouse livers were lysed by RIPA with 1% proteinase inhibitors for Western blot.

| Histochemical and immunohistochemistry of atherosclerotic lesions
Mice were subcutaneously injected with acacetin or normal saline two times every day for 12 weeks. Mice were made to fast after the last injection and then sacrificed. The aortic roots were embedded in OCT (Sakura, USA) and quickly frozen horizontally to −20℃. Eight micrometers serial sections of the aortic root were collected on 10 slides.
For atherosclerosis analysis, the entirety of the aorta was fixed in 4% paraformaldehyde, opened longitudinally, and then analysed en face. The aortic root slides were determined by Oil Red O (ORO) staining and quantification by Image J software as described previously. 19 For plaque component analysis, immunohistochemistry was carried out using Abcam's IHC staining protocol for frozen sections.
Briefly, the aortic root slides were fixed in precooled acetone and then stained with anti-iNOS, anti-CD206, anti-pNrf2 S40 , anti-Nrf2 and anti-MsrA primary antibodies followed by donkey anti-rabbit (Alexa Fluor® 488) or anti-mouse (Alexa Fluor® 647) secondary antibodies, respectively. Images were captured using the Leica SP8 fluorescent microscope.

| Statistical analysis
Data are presented as mean ± SEM. Statistical analyses were performed using Onaway ANOVA between groups. Differences were considered to be significant at P < .05.

| Acacetin inhibited high oxLDL-induced cell death and oxidation
MTT method was used to determine the optimal concentration of high oxLDL for the following studies. The cytotoxicity of high oxLDL F I G U R E 2 Regulatory effects of acacetin on anti-oxidative stress-related reductases as well as the abolishment of its protective effects against oxLDL in EA.hy926 cells with the silencing of MsrA or Nrf2. (A-G) EA.hy926 cell MsrA, Nrf2/Keap1-related protein, and SIRT1 expression levels without (control) or with oxLDL stimulation in the absence (oxLDL) or presence of 0.3, 1, or 3 μmol/L acacetin were measured by Western blot. (H-J) Apoptosis of EA.hy926 cells was measured by flow cytometry. Cells were transfected with scrambled siRNA, MsrA siRNA, Nrf2 siRNA or SIRT1 siRNA for 48 h, and then subjected to oxLDL treatment in the absence (control) or presence of 3 μmol/L acacetin. (K-L) Intracellular ROS levels were measured by flow cytometry. Cells were transfected with scrambled siRNA, MsrA siRNA or Nrf2 siRNA for 48 h, and then subjected to oxLDL treatment for 15 min in the absence (control) or presence of 3 μmol/L acacetin for 24 h. n = 5 for each group, P # < .05, P ## < .01 vs control group; P * < .05, P ** < .01, P *** < .001 vs oxLDL-stimulated group was tested on EA.hy926 cells, revealing that high oxLDL, unlike normal oxLDL, had stronger cytotoxicity that induced much more cell death after 24 hours incubation ( Figure S1A). We chose 5 μg/mL high oxLDL, which caused nearly 28% cell death, as the final study concentration. Pretreatment with acacetin for 4 hours significantly protected against the oxLDL-mediated reduction of cell viability ( Figure S1B). To determine the deep effects of acacetin on cell death, cells were treated with the same procedures as in the MTT study.
In addition to this, we also measured the intracellular ROS levels. Cells were pretreated with acacetin for 24 hours followed by 20 μmol/L DCFH-DA and 50 μg/mL high oxLDL for 15 minutes.
As shown in Figure 1G,H, high oxLDL treatment resulted in a high intracellular ROS level (the fluorescence intensity nearly 2.36 folds of the control group). In contrast, ROS production was markedly reduced in cells pretreated with 3μM acacetin (decreased to 1.4 folds of control). Moreover, the oxidase protein and MDA were also reduced by different concentrations of acacetin ( Figure S1C,D).
These results in turn indicate that acacetin could protect EA.hy926 cells from high oxLDL-induced cell death by reducing intracellular ROS levels.

| Acacetin enhanced cellular antioxidative defence through increasing oxidoreductases expression
To investigate how acacetin exerts its anti-oxidative stress effects, EA.hy926 cells were treated with acacetin at different concentrations for 24 hours. Three micrometers acacetin significantly increased reductase MsrA, Nrf2, Nrf2 downstream HO-1, and CAT protein expression at the cellular basal level ( Figure S1E-S1I). These results suggest that acacetin may enhance basal anti-oxidative defences.
Furthermore, in order to ensure the anti-oxidative effects of acacetin under oxidative stress conditions, cells were pretreated with acacetin at different concentrations for 4 hours followed by 5 μg/ mL high oxLDL stimulation for 20 hours. Surprisingly, we found that MsrA protein expression level was also significantly increased at 3 μmol/L acacetin (Figure 2A,B). Moreover, acacetin also increased Nrf2, downstream HO-1, Trx ( Figure 2C-E) and SIRT1 protein expression levels ( Figure 2F), but did not change pAMPK Thr172 / tAMPK levels ( Figure S1N). Interestingly, we also showed that 3 μmol/L acacetin remarkably decreased oxLDL-induced Keap1 expression level ( Figure 2G). The Nrf2/Keap1 system is a defence mechanism used to preserve cellular homeostasis, and Nrf2 is regarded as a master regulator of the oxidative stress response. 20 As such, these results preliminarily confirmed that acacetin can exert anti-oxidative effects through the Nrf2 pathway but not the AMPK pathway. Acacetin also slightly up-regulated CAT, SOD1 and SOD2 levels (Figure S1J-S1M).
Collectively, these data suggest that acacetin can potentiate cellular anti-oxidative defence through the up-regulation of MsrA, Nrf2, HO-1, Trx and SIRT1 to reduce intracellular ROS levels, which in turn indicates that acacetin may be involved in Nrf2/Keap1 or MsrArelated pathways.

| Acacetin enhanced cellular anti-oxidative effects through the MsrA-Nrf2/Keap1 pathway
In order to confirm whether acacetin exerts its anti-oxidative stress effects through Nrf2/Keap1 or some other pathway, we silenced cellular nrf2, msra, sirt1 gene expression using siRNAs. We displayed that the influence of acacetin on not only easing apoptosis and necrosis ( Figure Figure 3C), the SIRT1 level was significantly increased after acacetin treatment ( Figure 3D) which had the same tendency as the scrambled group. Interestingly, Keap1 expression in the scrambled group was remarkably contrary to that when Nrf2 expression was reduced ( Figure 3E). Moreover, when MsrA was silenced, aside from the lack of MsrA increase after either oxLDL stimulation or pretreatment with acacetin followed by oxLDL ( Figure 3C

| Acacetin attenuated atherosclerosis in Western diet-fed apoE -/mice through the activation of Nrf2 and MsrA in the lesions
ApoE −/− mice were subcutaneously injected with normal saline or acacetin and fed a Chow (blank group) or Western diet for 12 weeks.
The body weights (at various time-points) of mice showed no difference between the control and acacetin-treated groups, but spleen/ Mice were fed a Western-type diet and injected subcutaneously with normal saline or acacetin for 12 wk.
Abbreviations: TC, total cholesterol; TG, triglycerides. # Is statistically significant vs blank group. Micrographs were captured at ×200 magnification. n = 5-9 of each group, P * < .05, P ** < .01 vs control group body weight ratios (at the endpoint) were significantly lower in acacetin-treated mice (Table 1). After 12 weeks, plasma TC, TG, HDL-C and LDL-C levels were measured. TC and LDL-C levels were not different between control and acacetin-treated groups, albeit TG levels were significantly higher in the latter. Interestingly, HDL-C and apoAI levels were markedly increased in acacetin-treated mice ( Table 1).
The impact of acacetin injection on the development of atherosclerosis in apoE -/mice was assessed. Representative atherosclerotic lesions in en face images and cross-sections of aortic roots stained with ORO are shown in Figure 4A,B En face analysis of pinned-out aortas revealed that the atherosclerotic lesion percentage area in acacetin-injected mice (8.19% ± 0.92%) was significantly reduced compared to that of in the control group (11.04% ± 1.04%, P < .05, Figure 4D), especially in the arch region (20.25% ± 2.28% vs 25.64% ± 1.42%, P < .05, Figure 4C). In addition to this, the lipid staining area in the aortic root lesion of acacetin-treated mice (0.13 ± 0.01 mm 2 ) was 26.7% (P < .05) smaller than that of in control mice (0.18 ± 0.02 mm 2 , Figure 4E). We uncovered a higher expression level of pNrf2 S40 in both the arterial wall and plaque but Nrf2 mainly in the plaque of acacetin-injected mice compared to those in control groups, indicating that acacetin could successfully activate Nrf2 in aortic cells ( Figure 4F).
We also found that MsrA expression recovered mainly in the arterial wall of acacetin-treated mice ( Figure 4G).

| Acacetin ameliorated oxidative stress and inflammatory in apoE −/− mice
Anti-serum amyloid A and PON1 are HDL apolipoproteins: whilst the former has pro-atherogenic activities, 22 the latter is instead considered to be atheroprotective and decreases after inflammatory stimuli. 23,24 We found that the plasma levels of SAA and PON1 were significantly decreased and increased, respectively, in acacetin-treated mice ( Figure 5A-C). Furthermore, reverse cholesterol transport (RCT) is a process that facilitates cholesterol transport from peripheral organs back to the liver to regulate excessive systemic cholesterol levels. The cholesterol efflux mediated via ABCA1 and ABCG1 outside the cells whilst HDL-C can be taken up by SR-BI for degradation of HDL. 25 We found that ABCA1, SR-BI and ABCG1 protein levels were up-regulated in acacetin-injected mice ( Figure 5D,G), which indicates that it may accelerate RCT in the liver. Meanwhile, liver CAT protein expression level was also significantly increased in acacetin-injected mice relative to that of in controls (Fig. S2A,E). However, Nrf2 and the other oxidoreductases, MsrA, and PON1 expression levels showed F I G U R E 5 Acacetin ameliorated oxidative and inflammatory stress in the circulation and liver of apoE −/− mice. (A-C) Plasma SAA and PON1 levels were detected by Western blot. (D-G) ABCA1, ABCG1, and SR-BI levels in the liver were determined by Western blot. n = 5-9 of each group, P # < .05, P ## < .01 vs blank group; P * < 0.05, P *** < 0.001 vs control group no observable differences between the control and acacetin-treated groups (Fig. S2B,D). These data accordingly suggest that acacetin plays an anti-oxidative stress role in the liver.
The inflammatory process in the atherosclerotic-burdened artery may lead to increased blood levels of pro-inflammatory cytokines, including but not limited to IL-6 and TNFα. 26 In this study, plasma samples were diluted to the appropriate ratio and measured using the ELISA method. As shown in Table 1

| D ISCUSS I ON
Atherosclerosis is no longer thought to be a chronic inflammatory disease characterized solely by dyslipidaemia, but rather by oxida- The Nrf2/Keap1 complex is a potent transcriptional activator that plays a central role in the induction of many cytoprotective genes in response to electrophilic and oxidative stress. 37 Induction of HO-1 and Trx in a Nrf2-dependent manner has also been experimentally demonstrated. 38  without altering plasma lipid levels. In addition to this, injection of acacetin also raised mouse plasma levels of PON1 and IL-10 and reduced levels of pro-inflammatory factors, SAA, IL-6 and TNFα. PON1 is an important anti-oxidation enzyme that is synthesized in the liver and secreted into the plasma wherein it associates with HDL particles. SAA is likewise also produced in the liver, and its expression is correspondingly increased in response to IL-6 and TNFα. 24 However, whilst the expression levels of CAT and SOD2 similarly increased in the liver under pro-inflammatory factor stimulation, PON1 levels remained unchanged, possibly due to its secretion to the plasma. All these results suggested that acacetin confers an anti-atherogenic benefit through accelerating lipid metabolism, anti-oxidation along with anti-inflammation mechanisms in the circulation and liver.
In our results, both the spleen/body weight ratios and the altered fate of macrophages are changed by acacetin. We speculate that, as the most important immune organ, spleen weight may represent strong immunity with increasing monocytes and enhancing phagocytosis effects (M1 macrophage-like). YUNG-LUEN SHIH and colleagues found that bufalin increased the body weight, but reduced liver and spleen weights, and reduced CD3, CD16 and Mac-3 cell markers.
They finally conclude that bufalin may modulate immune responses not only through increasing monocyte (CD11b) population and T-and B-cell proliferation, but also by increasing macrophage phagocytosis in leukaemic mice in vivo. 42 Marco Busnelli and colleagues found that fenretinide, a synthetic retinoid derivative, could induce spleen abnormally enlarged and markedly increased atherosclerotic lesions at the aortic arch, thoracic and abdominal aorta of fenretinide-treated mice, just the similar results in our control (western diet) group. 43 In our present study, we found that mice treated with acacetin had lower spleen/body weight ratio and it may be because acacetin regulates spleen immunity function, decreasing monocytes or promoting monocyte-macrophage to M2 differentiation.
Interestingly, acacetin-treated mice had a higher body weight than that of in the control group, a finding in stark contrast to that of Liou et al's, showing that in high fat diet-fed obese mice, acacetin significantly reduced body weight. 44 Burke et al recently also reported that citrus flavonoids supplementation to a high fat, cholesterol-containing diet protected against obesity. These effects may be related to reserve existing obesity, adipocyte size and number through enhanced energy expenditure and increased hepatic fatty acid oxidation. 45 Potential reasons our differing results include variations in mice type as well as intravenous administrative methods.

| CON CLUS ION
In conclusion, our present study demonstrated that the natural flavone acacetin promotes not only a significant reduction in cellular apoptosis through anti-oxidative stress effects via the MsrA-Nrf2/ Keap1 pathway in vitro, but also halts atherogenesis through accelerating lipid metabolism as well as the anti-oxidation and anti-inflammatory capacity of Western diet-fed apoE -/mice. It may therefore serve as a potential drug candidate for prevention and treatment atherosclerosis-related CVD.

ACK N OWLED G EM ENTS
We are grateful to professor Gui-Rong Li, Dr Cui-Lian Dai and associate researcher Dianyu Dong from Xiamen Cardiovascular Hospital of Xiamen University for giving advice and editing the English text of a draft of this manuscript.

CO N FLI C T O F I NTE R E S T S
The authors have no conflicts of interest to declare. Project administration (lead); Resources (lead).

E TH I C S A PPROVA L A N D CO N S E NT TO PA RTI CI PATE
The Xiamen University Ethics Committee approved the protocols according to the Helsinki Declaration.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data and materials in this study are available on request from the authors.