Alpha-Lipoic Acid Protects Human Aortic Endothelial Cells Against H₂O₂-Induced Injury and Inhibits Atherosclerosis in Ovariectomized Low Density Lipoprotein Receptor Knock-Out Mice

Despite considerable therapeutic advances over the past few decades, atherosclerosis remains a leading cause of death worldwide, especially among postmenopausal women [1, 2]. Compared with premenopausal women, postmenopausal women have a higher incidence of cardiovascular disease due to loss of responsiveness to either internal or even externally provided estrogen [3]. It has been postulated that estrogen deficiency contributes to the high morbidity of atherosclerosis in postmenopausal women through a role in numerous atherogenic processes, including inflammation [4, 5], vascular smooth muscle cell proliferation [6], generation of reactive oxygen species (ROS) [7], and apoptosis of vascular endothelial cells [8]. Some studies have demonstrated that hormone replacement therapy (HRT) can lower the risk of cardiovascular disease in postmenopausal women, but only when therapy is initiated in the early stages after menopause [9-12].

Many studies have shown that inflammation and oxidative stress play a central role in the progression of atherosclerosis [2, 13, 14]. Oxidative stress, as a result of excessive generation of intracellular ROS including hydroxyl radicals (OH^–), hydrogen peroxide (H₂O₂), and superoxide anion (O₂^–), is reported to play a key role in the development and progression of atherosclerosis via promoting endothelial dysfunction, inflammation, and lipid peroxidation [15, 16]. The NADPH oxidase (Nox) family of enzymes and their regulatory subunits, including p22phox, p47phox, Noxa1, and p67phox, are important sources of ROS in the vasculature triggered by stress, hormones, and vasoactive agents. ROS is reported to activate NF-κB [17], which plays a key role in various inflammatory and immune responses and is linked to the development of atherosclerosis [18]. In addition, it has been shown that NF-κB is involved in the transcriptional regulation of Nox subunit expression [19, 20]. Therefore, pharmacological agents that could attenuate inflammation and oxidative stress by interrupting the NF-κB pathway should possess anti-atherogenic effects.

Estrogen is directly or indirectly involved in regulating immune responses, cell proliferation, inflammation, and cell adhesion [18]. Estrogen signaling has been shown to reduce the development of atherosclerosis through rapid non-genomic effects and/or the classic pathway by binding to estrogen receptors (ERα and ERβ) [21-23]. Estrogen also has antioxidant properties by inhibiting the expression of Nox and the generation of O₂^–superoxide anion [24]. Furthermore, estrogen exerts its atheroprotective effects by binding to its receptors to generate nitrogen monoxide in endothelial cells, which tends to have a protective function on blood vessels [25-27]. Unfortunately, the expression levels of estrogen receptors decline during the late stage of menopause. As a result, estrogen supplementation at this stage cannot reverse the risk of atherosclerosis [28, 29]. Therefore, a pharmacological agent that could promote the levels of estrogen receptors would benefit postmenopausal women by protecting against atherosclerosis.

Alpha-lipoic acid (ALA), also known as 1, 2-dithiolane-3-pentanoic acid, 1, 2-dithiolane-3-valeric acid, or thioctic acid, is reported to be a potent antioxidant that can eliminate intracellular ROS [30-32]. Several lines of evidence have shown that ALA has anti-inflammatory and anti-obesity effects against the pathological process of atherosclerosis [33-35]. However, the mechanisms of the protective function of ALA and its effects on the development of postmenopausal atherosclerosis remain unknown.

Human aortic endothelial cells (HAECs) were obtained from the Institute of Biochemistry and Cell Biology, Chinese Academy of Sciences (Shanghai, China). Cell culture medium (DMEM, RPMI-1640) and fetal bovine serum (FBS) were purchased from Gibco-BRL Co. (Gaithersburg, MD, USA). ALA (at a purity of > 99.5%) was purchased from Fushilai Medicine & Chemical Co., Ltd. (Changshu, China). The DCFH-DA fluorescent probe was purchased from Beyotime Institute of Biotechnology (Haimen, China). Protein extract assay kits were obtained from KeyGEN Biotechnology (Nanjing, China). Antibodies specific for Bcl-2, caspase-3, ERα, ERβ, ICAM-1, VCAM-1, IκBα, phosphorylated p65 (p-p65), and Nox4 were obtained from Proteintech Group (Wuhan, China), and antibodies specific for p65, vinculin were purchased from Bioworld Technology (Nanjing, China). The NF-κB inhibitor PDTC was obtained from Beyotime (Jiangsu, China), and the Nox4 inhibitor DPI and estrogen receptor inhibitor ICI182, 780 were purchased from Sigma (St. Louis, MO, USA).

HAECs were cultured in DMEM containing 10% FBS at 37°C in 5% CO₂. After reaching 60% confluence, HAECs were treated with different concentrations of ALA (75, 150, 300μM) for 24 h before treatment with H₂O₂ (1 mM) for 4 h. ALA was dissolved in dimethylsulfoxide.

HAECs or tissues were collected for protein extraction and detection of Bcl-2, caspase-3, ERα, ERβ, ICAM-1, VCAM-1, IκBα, p-p65, Nox4, total p65, and vinculin using a protein extraction kit (KeyGEN Biotechnology). HAECs were treated with different concentrations of ALA (75, 150, 300μM) for 24 h before treatment with H₂O₂ (1 mM) for 4 h. Mice were fasted overnight and sacrificed, aortas were dissected and separated from fatty tissue, and aortic protein was extracted using a protein extraction kit containing protease inhibitor and protein phosphatase inhibitor cocktail. Proteins (20-30 μg per lane) were loaded on an 8-12% sodium dodecyl sulfate polyacrylamide electrophoresis gel, electrophoresed at 100 V for 2 h in gel running buffer for separation, and subsequently transferred to polyvinylidene fluoride (PVDF) membranes. Thereafter, membranes were blocked using 5% fat-free milk dissolved in Tris-buffered saline with Tween-20 (TBS-T) for 2 h at room temperature and then incubated at 4°C overnight with primary antibodies against Bcl-2 (Proteintech Group, Wuhan, China, 1: 750 dilution), caspase-3 (Proteintech Group, 1: 750 dilution), cleaved caspase-3 (Proteintech Group, 1: 750 dilution), ERα (Proteintech Group, 1: 500 dilution), ERβ (Proteintech Group, 1: 1000 dilution), ICAM-1 (Proteintech Group, 1: 500 dilution), VCAM-1 (Proteintech Group, 1: 500 dilution), IκBα (Proteintech Group, 1: 750 dilution), p-p65 (Proteintech Group, 1: 750 dilution), Nox4 (Proteintech Group, 1: 750 dilution and Affinity Biosciences, 1: 750 dilution), total p65 (Bioworld Technology, 1: 1000 dilution), and vinculin (Bioworld Technology, 1: 5000 dilution). After a series of rinses, PVDF membranes were incubated with a peroxidase-conjugated secondary antibody (Bioworld Technology, 1: 5000 dilution) in TBS-T for 2 h at room temperature. Finally, proteins were visualized using a chemiluminescence method (Gel Doc^TMXR, Bio-Rad, Hercules, CA, USA) and the protein levels were evaluated semiquantitatively using Image Lab software.

At indicated times of differentiation, cultured cells were washed with phosphate buffered saline (PBS) twice and fixed with 4% paraformaldehyde/PBS for 10 min at room temperature. Then, cells were permeabilized with 0.5% Triton-X/PBS for 10 min. After washing with PBS, cells were blocked with 2% bovine serum albumin and incubated overnight at 4˚C with primary antibodies, including ERα (Proteintech Group, 1: 50 dilution), and p65 (Bioworld Technology, 1: 50 dilution). Secondary antibodies including Alexa Fluor 488-conjugated anti-rabbit IgG (goat, 1: 100, Thermo Fisher Scientific, Waltham, MA, USA) were used according to the manufacturer’s instructions. The samples were incubated at room temperature for 1 h. After washing with PBS and counterstaining with DAPI, samples were mounted with Prolong Gold antifade solution (Thermo Fisher Scientific). Cells were visualized and photomicrographs were obtained using an upright fluorescence microscope (Olympus, Center Valley, PA, USA). Quantification of fluorescence intensity was performed using ImageJ. At least 50 cells from three different areas per chamber were measured.

We used a colorimetric assay kit (Beyotime, Nanjing, China) to measure lactic dehydrogenase (LDH), malondialdehyde (MDA), superoxide dismutase (SOD), and L-glutathione (GSH) release. HAECs were cultured in 6-well plates at a density of 1x105 per well and treated with different concentrations of ALA (75, 150, 300μM) for 24 h before stimulating with H₂O₂ (1mM) for 4 h.

Biomarkers of oxidative stress and antioxidants in serum, including LDH, MDA, SOD, and GSH, were detected using the abovementioned assay kits. Biomarkers of lipid metabolism in serum, including total cholesterol (CHO), triglycerides (TG), high-density lipoprotein cholesterol (HDL-C), and low-density lipoprotein cholesterol (LDL-C), were measured using assay kits (Beyotime, Nanjing, China).

THP-1 cells (a human acute monocytic leukemia cell line) were labeled with the fluorescent probe BCECF-AM (Beyotime, Jiangsu, China) then incubated with confluent HAECs for 1 h at 37oC. Non-adherent THP-1 cells were removed gently with PBS. Images of adherent THP-1 cell were obtained using fluorescence microscopy (Olympus; x100 magnification).

The level of ROS was detected using the fluorescent probe, 2, 7-dichlorofluorescein diacetate (DCFH-DA). A final concentration of 10μM DCFH-DA was added to the HAECs, and cells were washed twice with serum-free medium, and then incubated for 20 min at 37oC in the dark. The relative levels of fluorescence were quantified using a FACScalibur Flow Cytometer (Becton, Dickinson and Company, Franklin Lakes, NJ, USA).

Cell apoptosis in plaques was examined using TUNEL ApopTag Fluorescein in Situ Apoptosis Detection Kit (Biotool, Houston, TX, USA). Freshly cut cryosections were fixed with 1% paraformaldehyde in PBS for 10 min at room temperature, rinsed with PBS, and permeabilized with a mixture of ethanol and acetic acid (2: 1) for 5 min at -20°C. Slides were incubated with reaction buffer containing the TdT enzyme for 1 h at 37°C in a humidified chamber and then incubated with a FITC-conjugated anti-digoxigenin antibody for 30 min at room temperature. Finally, slides were mounted in a DAPI/antifade-containing medium. Slides were observed and photographed with an Olympus BX61 fluorescence microscope.

Successive sections were collected on the same slide and at least 15 slides from 5 consecutive mouse thoracic aortas per area per mouse were examined. Aortas were dissected and separated from fatty tissue. After Oil-Red-O staining, we used Image-Pro Plus analysis to measure the proportion of lipid deposition. Mouse aortas were fixed in 10% formalin, embedded in paraffin, and sectioned (5 μm thickness). Sections were treated with xylene to remove the paraffin and were rehydrated, incubated with 3% H₂O₂ for 10 min, and then washed three times with PBS. Next, the sections were blocked with serum for 30 min and incubated with primary antibodies overnight against Nox4 (Proteintech Group, 1: 100 dilution), p-p65 (Proteintech Group, 1: 100 dilution), ICAM-1 (Proteintech Group, 1: 50 dilution), VCAM-1 (Proteintech Group, 1: 50 dilution), and CD68 (Abcam, Cambridge, MA, USA 1: 200 dilution). Images were captured using Leica Microsystems.

Elastin in the murine arterial wall was stained using Gomrori’s aldehyde-fuchsin staining method, with an elastic fiber staining kit (Maixin Bio, Fuzhou, China). Briefly, after deparaffinization and rehydration, sections were incubated for 5 min in Lugol’s iodine solution, washed with PBS and incubated with sodium thiosulphate for 5 min. After washing with PBS and 70% ethanol, sections were incubated with aldehydefuchsin for 10 min and acid Orange G for seconds.

Forty 16-week-old Ldlr^-/- female mice with a C57BL/6 background were purchased from Vital River Laboratory Animal Technology Co., Ltd. (Beijing, China). All experimental procedures were performed in accordance with the Institutional Animal Care and Use Committee of the Center for Laboratory Animal Sciences, Dalian Medical University, China. The mice were randomly divided into five groups: a sham surgery group (SHAM); a SHAM+high fat diet (HFD) group, and three ovariectomized (OVX) groups; the OVX mice were then randomly divided into three groups of eight animals per group: (1) OVX+HFD group; (2) OVX+ALA 70 mg/kg; (3) OVX+ALA 210 mg/kg. Mice in the three OVX groups underwent ovariectomy during the first week of the experiment, and 1 week later, the two OVX+ALA groups were treated with ALA (dissolved in 0.5% carboxymethylcellulose [CMC-Na], 3.75 ml/kg body weight per day) by gavage once daily. Mice in the SHAM and SHAM+HFD group were treated with vehicle (0.5% CMC-Na, 3.75 ml/kg body weight) by gavage once daily. All mice except those in the SHAM group were fed an HFD. The feeding regimens and ALA gavage continued for 16 weeks. The body weight of each mouse was monitored weekly. After 16 weeks, the mice were fasted overnight and sacrificed, and blood samples were obtained from the retro-orbital plexus, and centrifuged at 1200 g for 10 min at 4oC. Aortas were dissected and separated from fatty tissue. After Oil-Red-O staining, we used Image-Pro Plus to measure the proportion of lipid deposition. Subsequently, the serum was frozen at -80˚C for later use.

Results are expressed as the mean ± SEM or SD. The in vitro assay was performed at least three times. Statistical analysis was performed using SPSS 13.0 statistical software (SPSS lnc., Chicago, IL, USA). Data were compared between groups using a Student t-test or one-way ANOVA analysis of variance, followed by the least significant difference post-test. Statistical significance was set at P < 0.05.

The Ldlr^-/- mouse is a widely used animal model of atherosclerosis because it demonstrates a sensitive response to oxidative stress and inflammation and naturally develops atherosclerosis [36]. To simulate the development of atherosclerotic lesions in postmenopausal women, we performed OVX in 16-week-old Ldlr^-/- female mice and subsequently fed them a HFD. To evaluate the effects of ALA treatment on the development of atherosclerosis in the OVX Ldlr^-/- mouse model and/or induced by a HFD, these mice received ALA by oral gavage from the second week after OVX. We observed more body weight gain in the OVX Ldlr^-/- group fed a HFD than in the control SHAM+HFD group without OVX; weight gain in both of these groups was substantially higher than in the SHAM group fed normal chow without OVX (Fig. 1A). However, ALA treatment markedly reduced the body weight of the OVX Ldlr^-/- group on a HFD, especially after a prolonged period of ALA administration (Fig. 1A). The Ldlr^-/- mice that underwent OVX and received a HFD developed markedly more severe atherosclerotic lesions with fatty streaks and dots covering the thoracic aorta and aortic arch/sinus than did the Ldlr^-/- SHAM+HFD group (Fig. 1B). Microscopically, the ALA treated group exhibited drastic reductions in plaque size and lipid deposition (Fig. 1C). In addition, histological sections showed that ALA treatment significantly attenuated arterial wall thickness, protecting the internal elastic fibers from fragmenting and disrupting (Fig. 1D). Taken together, these results demonstrate that ALA prevented the development of atherosclerosis induced by OVX and HFD.

Next, we evaluated the effects of ALA on elevated serum lipid levels, one of the risk factors for atherosclerosis. Mice in the OVX+HFD group had higher serum TG, CHO, and LDL-C levels but lower levels of HDL-C compared with the SHAM+HFD group, suggesting that menopause played a role in hyperlipidemia. However, the administration of ALA significantly decreased the levels of TG, CHO, and LDL-C while elevating the level of HDL-C (Fig. 2A-D). These results demonstrate that the hyperlipidemia induced by HFD and OVX was prevented by ALA.

Accumulating evidence suggests that oxidative stress plays an important role in the process of atherosclerosis. Therefore, we measured several biomarkers of oxidative damage and antioxidants, including LDH, a representative marker of oxidative injury, MDA, a marker of lipid-peroxidation [37], and antioxidant enzymes SOD and GSH [38] in the Ldlr^-/- animal model. We found that OVX and HFD suppressed the levels of SOD and GSH activity while elevating the levels of LDH and MDA in Ldlr^-/- mice, suggesting menopause and an HFD contributed to oxidative stress. Of note, ALA treatment significantly decreased the levels of biomarkers of oxidative stress but enhanced the levels of antioxidant enzymes in mouse serum (Fig. 2E-H). Taken together, our data indicate that ALA administration prevented the development of atherosclerosis by inhibiting hyperlipidemia and oxidative stress in HFD/OVX Ldlr^-/- mice.

ERs can activate several cellular kinases by inducing a “rapid” non-nuclear signaling cascade independent of estrogen [36]. We found that protein levels of ERα and ERβ were markedly enhanced after treatment with ALA in a dose- and time-dependent manner (Fig. 3A-C). We next confirmed this observation with an immunofluorescent assay, in which ALA reversed the down-regulation of ERα expression induced by H₂O₂ treatment (Fig. 3D). To further validate our findings, we examined the expression of ER in Ldlr^-/- mice. We found that the expression of ERα and ERβ decreased significantly in HFD/ OVX Ldlr^-/- mice. However, ALA treatment significantly elevated the expression of estrogen receptors in a dose-dependent manner, which is consistent with our in vitro findings (Fig. 3E-F). These results suggest that ALA may up-regulate the expression of ERα and ERβ that was diminished during the development of atherosclerosis.

ALA prevented H₂O₂-induced oxidative stress through down-regulating the expression of Nox4 and p22phox.

Oxidative stress is mainly derived from excessive generation of ROS, which plays a key role in the progression of atherosclerosis [2, 13, 14]. To evaluate the effects of ALA on eliminating intracellular ROS, we measured changes in ROS levels in response to ALA treatment by flow cytometry and fluorescence microscopy. We found that intracellular ROS levels were substantially reduced following ALA treatment, suggesting that ALA eliminated the excessive generation of intracellular ROS stimulated by H₂O₂ (Fig. 4A-B). We next examined the protein levels of Nox4 and p22phox, two key components in ROS-generation, and found that ALA treatment markedly abrogated Nox4 and p22phox expression induced by H₂O₂ (Fig. 4C-D). In the Ldlr^-/- mouse model, the expression of Nox4 in the SHAM+HFD and OVX+HFD groups was remarkably elevated compared to the SHAM group. However, ALA administration significantly decreased the expression of Nox4 in OVX+HFD Ldlr^-/- mice in a dose-dependent manner (Fig. 4E). Immunohistochemical staining showed similar results as the western blot analysis (Fig. 4F). These findings suggest that ALA might exert its anti-oxidative effects by suppressing the expression of ROS-generating enzymes such as Nox4 and p22phox.

The ROS-NF-κB signaling pathway plays a crucial role in activating inflammatory and immune responses that contribute to the pathogenesis of atherosclerosis [17]. It has been reported that IkB phosphorylation and its subsequent ubiquitin-mediated proteasomal degradation translocate NF-kB dimers to the nucleus, leading to NF-kB activation and induction of NF-kB- dependent transcription [39]. To investigate the effect of ALA on H₂O₂-induced NF-κB activation in HAECs, we examined the protein levels of IκBα and p-p65. We found that ALA markedly suppressed the phosphorylation of p65 and the degradation of cytoplasmic IκBα compared with the H₂O₂ treatment group (Fig. 5A-B). To further evaluate the effects of ALA on NF-κB activation, we examined the cellular localization of its transcriptionally active subunit p65 in an immunofluorescence assay. We found that exogenous H₂O₂ treatment markedly stimulated p65 translocation into the nucleus, while ALA treatment led to the removal of p65 out of the nucleus in a dose-dependent manner (Fig. 5C). Consistent with in vitro findings, p-p65 was markedly increased in the OVX group compared with the SHAM group, while ALA treatment markedly reversed this change (Fig. 5D-E). These results demonstrated that oxidative stress injury induced by H₂O₂ or HFD/OVX could activate the NF-κB signaling pathway that contributes to the pathogenesis of atherosclerosis, whereas ALA could effectively inhibit activation of this pathway, providing a protective effect against atherosclerosis.

The adhesion molecules ICAM-1 and VCAM-1 that are secreted by activated endothelial cells play a key role in initiating atherosclerosis and development of fatty streak lesions [40]. To determine whether Nox4, ERs, and NF-κB modulate ICAM-1 and VCAM-1 to promote atherosclerosis, we measured the expression of ICAM-1 and VCAM-1 in HAECs induced by H of ICAM-1 and VCAM-1 induced by H₂O₂ was abrogated by the Nox4 inhibitor DPI and the NF-kB inhibitor PDTC, respectively, but was intensified in response to treatment with the estrogen receptor antagonist ICI182, 780 (Fig. 6A-C). These findings suggest that Nox4 and NF-kB played a positive role in promoting atherosclerosis while ER functioned as a negative regulator in the pathogenesis of atherosclerosis. Next, we investigated the effects of ALA on ICAM-1 and VCAM-1 by treating H₂O₂-induced HAECs with ALA as a single agent or with each of ALA+DPI, ALA+ICI, and ALA+PDTC. As expected, we found that ALA alone or in combination with PDI, ICI, or PDTC significantly suppressed the expression levels of ICAM-1 and VCAM-1 in H 2O2-induced HAECs (Fig. 6A-C), indicating that ALA had protective effects against atherosclerosis by inhibiting the expression of ICAM-1 and VCAM-1. These results suggest that estrogen receptors might play a role in preventing postmenopausal atherosclerosis. In contrast, Nox4 and NF-kB promoted atherosclerosis. 2O2 in the presence of Nox4, ERs, and NF-κB inhibitors. We found the increased expression.

As shown above, ROS could activate the NF-κB signaling pathway while suppressing estrogen receptors expression. As a primary source of ROS, Nox4 is highly expressed in the endothelium and interacts with NF-κB and ERs to form a ternary complex in the regulation of H₂O₂-induced oxidative stress. We next explored the relationships among Nox4, NF-kB, and ERs in response to oxidative stress. We found that PDTC down-regulated H₂O₂-induced Nox4 overexpression while up-regulating ERα expression upon H₂O₂ stimulation (Fig. 6D-E). In addition, as expected, DPI markedly reduced p-p65 levels and promoted ERα expression upon H 2O2-induction (Fig. 6F-G). However, ICI182, 780 treatment significantly increased the levels of Nox4 and p-p65 in response to H₂O₂ stimulation (Fig. 6H-I), indicating that inhibition of ERs activated the NF-κB pathway and stimulated Nox4 overexpression. Immunofluorescence staining showed that ICI182, 780 treatment resulted in the translocation of p65 into the nucleus in HAECs upon H₂O₂ induction, but that ALA treatment blocked this process, with p65 remaining in the cytoplasm (Fig. 6J). Based on our findings, we propose that Nox4, NF-κB, and ERs might interact to form a ternary complex with ROS to modulate the atherosclerotic process.

It has been reported that endothelial injury and apoptosis lead to lipid accumulation and inflammation which contribute to atherosclerotic plaque formation [41]. In postmenopausal women, estrogen deficiency promotes caspase-3 dependent apoptosis [42]. To investigate the effects of ALA on endothelial apoptosis, we measured the protein levels of cleaved caspase-3 and Bcl-2 under oxidative stress. We found that ALA treatment significantly reduced the expression levels of cleaved caspase-3 and enhanced Bcl-2 expression in HAECs upon H₂O₂ induction (Fig. 7A-B). In an vivo study, we showed that OVX Ldlr^-/- mice receiving HFD developed atherosclerosis with a significant increase in cellular apoptosis as evidence by TUNEL staining. Notably, ALA treatment markedly inhibited apoptosis in the arterial wall of HFD/OVX Ldlr^-/- mice in a dose dependent manner (Fig. 7C). Together, these findings indicate that ALA exerted its anti-apoptotic functions by up-regulating prosurvival Bcl-2 protein levels and down-regulating pro-apoptotic cleaved caspase-3 levels in arterial endothelial cells.

Cell adhesion molecules such as ICAM-1 and VCAM-1 play an important role in initiation and development of fatty streak lesions in the arterial wall [43]. As shown in Fig. 7D-E, H₂O₂ stimulation significantly increased the expression of ICAM-1 and VCAM-1 in HAECs. However, ALA treatment markedly and dose-dependently attenuated the expression of ICAM-1 and VCAM-1 (Fig. 7D-E). In the monocyte adhesion assay, H₂O₂ induction resulted in a striking increase in THP-1 adhesion to HAECs (Fig. 7F), while ALA treatment markedly suppressed this effect in a dose-dependent manner. Consistently, immunohistochemical staining showed that OVX+HFD Ldlr^-/- mice exhibited significantly higher expression of ICAM-1 and VCAM-1 than the SHAM or SHAM+HFD groups, whereas ALA administration substantially reduced the expression of ICAM-1 and VCAM-1 in OVX+HFD Ldlr^-/- mice in a dose-dependent manner (Fig. 7G). Accumulating evidence suggests that macrophages play an essential role in the formation and development of atherosclerotic plaques [44]. Here we used CD68 staining, a representative marker of activation of the mononuclear-phagocyte system, to show the presence of macrophages in our in vivo atherosclerotic model. We found that ALA treatment significantly inhibited atherosclerosis-associated macrophages in the area of atherosclerotic plaques of OVX+HFD Ldlr^-/- mice in a dose-dependent manner (Fig. 7H). Collectively, these data suggest that ALA can block the activation of monocyte adhesion through suppressing ICAM-1 and VCAM-1 expression and the mononuclear-phagocyte system to prevent the pathogenesis of atherosclerosis.

LDH is a stable enzyme normally present in the cytosol of most eukaryotic cells but rapidly releases into the supernatant upon cell death due to damage of the plasma membrane [45]. We found that ALA treatment significantly inhibited LDH activity of HAECs induced by H₂O₂ stimulation (Fig. 8A). To further evaluate the antioxidative effects of ALA, we measured the levels of MDA, SOD and GSH in HAECs in response to H₂O₂ stimulation. We showed that ALA treatment significantly suppressed MDA levels while elevating SOD activity and GSH levels in HAECs during oxidative injury (Fig. 8B-D). Taken together, our findings suggest that ALA provided antioxidant protection and prevention against oxidative damage-induced endothelial cell death.

Atherosclerosis is a complex pathological process modulated by multiple factors such as inflammation, oxidative stress, and apoptosis [46-48]. It has become one of the most common chronic vascular diseases worldwide and is an important public health concern in modern society, especially for postmenopausal women. It is well known that postmenopausal women exhibit a higher incidence of cardiovascular disease than premenopausal women due to loss of responsiveness to either internal or even externally provided estrogen [3]. HRT in post-menopausal women for the purpose of primary or secondary prevention of cardiovascular diseases is not effective, and is associated with an increase in the risk of developing stroke and venous thromboembolic events [49]. Estrogen may have a beneficial effect if initiated early during menopause, when a woman’s arteries are likely to be relatively healthy, but it has a deleterious effect if started in late menopause, when the arteries are more likely to exhibit signs of atherosclerotic disease [50]. ALA, a natural antioxidant synthesized in the mitochondria of liver and other tissues [51], has multiple anti-inflammatory and anti-oxidative effects. In this study, we found that ALA stimulated dose- and time-dependent expression of estrogen receptors. Consequently, ALA may be used as a naturally occurring free radical scavenger to replace HRT for postmenopausal women.

It is well established that the NF-κB signaling pathway has a role in a number of physiological process, involving immune inflammatory responses, oxidative stress, cell adhesion, differentiation, and apoptosis [52-54], many of which contribute to atherosclerosis. NF-κB can be activated by oxidative stress and inflammation, but on the other hand, activation of NF-κB up-regulates the genes modulating oxidative stress, inflammation, cell proliferation, and apoptosis, providing a feedback control [55]. Therefore, activation of NF-kB plays an essential role in the pathogenesis of atherosclerosis. In this study, we showed that ALA treatment inhibited NF-kB activation upon H₂O₂ induction in vitro and OVX/HFD in vivo by suppressing phosphorylation of p65 and the degradation of cytoplasmic IκBα, suggesting that ALA prevents atherosclerosis, at least in part, through inhibiting the NF-κB signaling pathway.

ROS is the main product of oxidative stress, which has a role in multiple pathological process in the development of cardiovascular diseases [56]. It was reported that estrogen deficiency led to an increase in ROS generation, implying that postmenopausal women may have a greater risk of oxidative stress [7]. Many cardiovascular diseases are known to be associated with elevated ROS levels. A major source of cellular ROS is the Nox family of enzymes, comprising seven members [57]. Nox4 is the major isoform that is constitutively active and expressed in endothelial cells and vascular smooth muscle cells [58]. Nox4-derived ROS production is considered to be regulated mainly through alterations in Nox4 expression levels [59]. Nox4 colocalizes with the integral membrane protein p22phox which is necessary for Nox4 activity [60]. In this study, we showed that ALA treatment significantly reduced the over-expression of Nox4 and p22phox in response to H₂O₂ stimulation. Moreover, we demonstrated that ALA diminished the excessive generation of intracellular ROS induced by H₂O₂. Therefore, ALA may modulate the ROS-generating enzymes to exert its potent anti-oxidative effects and provide protection against atherosclerosis.

Endothelial cell apoptosis plays an important role in the formation and progression of atherosclerotic lesions [61]. Estrogen deficiency has been reported to facilitate caspase-3-dependent apoptosis in postmenopausal women [8]. In the present study, we demonstrated that ALA treatment was sufficient to reverse H₂O₂-induced apoptosis and cellular injury in HAECs, and OVX/HFD induced apoptosis in atherosclerotic lesions of Ldlr^-/- mice. As such, one of the molecular mechanisms responsible for the atheroprotective effects of ALA likely involves repression of cellular apoptotic pathway during the pathogenesis of atherosclerosis.

ICAM-1 and VCAM-1 are endothelial adhesion molecules that play a key role in monocyte accumulation/adhesion to the endothelium and are important at all stages of atherogenesis. Here, we clearly showed that ALA suppressed the expression of ICAM-1 and VCAM-1 both in vivo and in vitro, implying that ALA could disturb one of the key elements in the pathophysiology of atherosclerotic disease, monocyte adhesion, through inhibiting ICAM-1 and VCAM-1 expression.

A significant body of evidence has shown that postmenopausal women have a higher incidence of atherosclerosis than premenopausal women [3]. Among multiple speculations and hypotheses, estradiol is considered to play a central role [3]. In the present study, we showed that ALA treatment up-regulated the protein levels of estrogen receptors both in vivo and in vitro. We also showed that H₂O₂ stimulation down-regulated both ERα and ERβ expression in HAECs. Meanwhile, ERs expressions were decreased in atherosclerotic pathological tissue induced by HFD and/or OVX. Our results are in line with previous work showing that ERs expressions are modulated by oxidative stress and inflammation in cardiovascular disease [62]. Importantly, we found that ALA treatment markedly up-regulated estrogen receptor protein levels both in vivo and in vitro, suggesting that ALA exerts atheroprotective effects, at least partly, by promoting the expression of estrogen receptors.

It has previously been reported that NF-κB and Nox4 together up-regulate indoxyl sulfate-induced angiotensinogen expression in proximal tubular cells [63]. In the present study, Nox4 overexpression induced by H₂O₂ was down-regulated by the NF-κB inhibitor PDTC, suggesting that NF-κB activation elevated the expression of Nox4. Alterations of p-p65 induced by H₂O₂ were reversed by the Nox4 inhibitor DPI, indicating that Nox4 promoted NF-κB activation. The decreased expression of ERα and ERβ induced by H₂O₂ was restored by PDTC and DPI, suggesting that Nox4 and NF-κB inhibited the expression of ERs. Taken together, Nox4 and NF-κB likely cooperate to inhibit the expression of estrogen receptors, and jointly promote the development of atherosclerosis. ALA can abolish such an inhibitory effect by up-regulating the expression of ERs, consequently activating several cellular kinases by inducing a “rapid” non-nuclear signaling cascade independent of estrogen.

In summary, our findings suggest that estrogen receptors “communicate” with NF-κB and Nox4 in the form of mutual restraint to participate in the regulation of initiation and development of atherosclerosis. ALA, a potent natural antioxidant which can enhance the expression of estrogen receptors and inhibit the activation of NF-κB signaling, may be harnessed as a potential therapeutic candidate to treat atherosclerosis in postmenopausal women.

This study was supported in part by the Grant from the National Natural Science Foundation of China (No. 81372853 and No. 81572586 to Pixu Liu; No. 81273508 to Huijun Sun; No. 81600037 to Tingting Shen); Liaoning Provincial Climbing Scholars Supporting Program of China (2012, to Pixu Liu).

The authors declare to have no competing financial or commercial interests that could potentially lead to a conflict of interests.

D. Shen, L. Tian and T. Shen contributed equally to this work.