Site-Specific Antiatherogenic Effect of the Antioxidant Ebselen in the Diabetic Apolipoprotein E–Deficient Mouse
Objective— Recently we showed that lack of the antioxidant enzyme glutathione peroxidase-1 (GPx1) accelerates atherosclerosis and upregulates proatherogenic pathways in diabetic apoE/GPx1-deficient double-knockout mice, thereby establishing GPx1 as an important therapeutic target. In vivo studies now investigate ebselen, a seleno-organic GPx1-mimetic, for its potential to reduce diabetes-associated atherosclerosis.
Methods and Results— Lesions were significantly increased in diabetic apoE−/− aortas (P<0.001) compared with nondiabetic controls after 20 weeks of diabetes. Ebselen-gavage significantly reduced total aortic lesions (P<0.001), with significant regional reductions in the arch (P<0.001), thoracic (P<0.001), and abdominal regions (P<0.05), but not within the aortic sinus of diabetic apoE−/− mice. These reductions were accompanied by significantly lower nitrotyrosine and Nox2 levels, reduced proatherogenic cellularity (macrophages and SMCs), and reduced expression of the proatherogenic mediator RAGE. Within the aortic sinus, ebselen reduced nitrotyrosine, Nox2, and VEGF levels but had no effect on RAGE. Studies in HAECs show that ebselen abrogates H2O2-induced increases in P-IKK, P-JNK, TNF-α, and Nox2.
Conclusions— Ebselen reduces atherosclerotic lesions in most regions of diabetic apoE−/− aorta, except within the aortic sinus, suggesting its effectiveness as a potential antiatherogenic therapy in diabetic-macrovascular disease. Ebselen may elicit its effect via modulation of transcription factors such as NF-κB and AP-1.
Lifestyle factors, together with genetic predisposition, now contribute to a growing number of patients with Diabetes mellitus. This has resulted in a dramatic increase in patients with cardiovascular disease (CVD) given that diabetes is a major risk factor for the development of atherosclerosis.1 Other known risk factors include dyslipidemia, hyperglycemia, hypertension, and obesity. It is also becoming increasingly clear that reactive oxygen species (ROS) generated by the prooxidant diabetic milieu contribute to diabetic complications including atherosclerosis.2 The involvement of ROS has been shown in the upregulation of proatherogenic pathways that include the biochemical process of advanced glycation3 and in particular, the receptor for advanced glycation end-products (RAGE).3 Oxidative stress also affects the expression of other proatherogenic proteins such as the adhesion molecule, VCAM-1, the chemokine MCP-1 and the proangiogenic growth factor (vascular endothelial growth factor [VEGF]).4 Importantly, the heightened state of oxidative stress observed in diabetic patients is postulated to play a role in diabetic complications2 based on strong evidence that oxidation of LDL and other lipids is a key factor required for the initiation of atherosclerosis.
In light of the strong evidence for oxidant-mediated events associated with atherosclerosis, antioxidants such as vitamins E5 and C6 have been tested in several cardiovascular human studies with mostly equivocal results. However, recent clinical studies have suggested an important antiatherogenic role for the antioxidant enzyme glutathione peroxidase-1 (GPx1). A patient cohort with suspected CVD showed an inverse association between GPx1 activity and cardiovascular events, and it was suggested that increasing GPx1 activity might lower the risk of CVD.7 Furthermore, an association was shown between reduced erythrocyte GPx1 activity, elevated plasma levels of homocysteine, and future cardiovascular events.8 A significant correlation has also been shown for a polymorphism in the human GPx1 gene and the risk of coronary artery disease.9 Importantly, in type II diabetic patients, functional variants within the GPx1 gene leading to decreased GPx1 activity, correlate with an increase in carotid artery intima-media thickness10 and coronary artery calcification.11 These clinical findings, although of significant diagnostic value, offer no functional role for GPx1 in atherogenesis.
Recently, we showed the importance of GPx1 in limiting diabetes-associated atherosclerosis (DAA) in apoE−/−GPx1−/− double knockout mice.12 Our findings led us to propose a significant antiatherogenic role for GPx1 in a diabetic milieu. In light of the mostly equivocal results with previous antioxidant trials,5,6 we postulated that compounds with GPx-like activity may provide a more targeted antioxidant approach to limiting DAA.
Ebselen [2-phenyl-1,2-benzisoselenazol-3[2H]-one] is a synthetic lipid soluble seleno-organic low molecular weight compound with strong antioxidant activity.13 It possesses GPx-like activity,14 particularly its ability to scavenge peroxynitrite radicals.15 Indeed, the mechanism of the GPx reaction by ebselen is proposed to be kinetically identical to that of the GPx enzyme.14 Previous studies support a role for ebselen in reducing various oxidative stress–mediated pathologies, including cerebral infarction,16 the protection of the endothelium in stroke-prone hypertensive rats,17 and cardiac dysfunction in a model of chronic iron overload.18 Of dual relevance, ebselen reduced oxidative damage of proteins and partially restored endothelial dysfunction in Zucker diabetic rats.19 Importantly, ebselen decreased arterial lesions in a superoxide-driven noninflammatory transgenic murine model consistent with a potential role for ebselen in reducing atherosclerosis.20 However, no study has directly evaluated the effect of ebselen on DAA.
Accordingly, the aims of this study were (1) to evaluate the effects of ebselen on the development of DAA and (2) to determine whether ebselen affects known proatherogenic pathways implicated in this process. Finally, potential mechanism/s involved in the antiatherogenic effects of ebselen were investigated in human aortic endothelial cells (HAECs).
Materials and Methods
Experimental Design, Assessment of Lesions, and Immunohistochemical Analysis
For expanded Materials and Methods, please see the supplemental materials (available online at http://atvb.ahajournals.org). 8-week old male C57Bl/J6 apoE−/− mice were rendered diabetic with streptozotocin and divided into ebselen-gavaged and nongavaged groups. Ebselen, dissolved in 5% CM-cellulose, was gavaged twice daily at 10 mg/kg body weight starting at 10 weeks of age. Dosage (20 mg/kg/d) was based on previously published data in mice.18 Sham-injected male C57Bl/J6 apoE−/− mice served as nondiabetic controls. A further group of diabetic mice received 5% CM-cellulose as a vehicle control. Animals were maintained for 10 (10-week study) and 20 weeks (20-week study) for gene expression and en face aortic lesion analysis respectively. Lesions were quantitated within the aortic sinus at the 10 and 20-week time points. For a detailed assessment of atherosclerotic lesions within the aorta and the aortic sinus region, and for details on blood sampling, plasma biochemistry and tissue collection, please see the supplemental materials. For immunohistochemical analysis, aortic tissue was embedded in paraffin and cross-sections prepared from the arch, thoracic, and abdominal areas. Hearts, cut at a plane parallel to the tips of the atria, were fixed in paraformaldehyde and frozen in OCT for lesion assessment within the sinus region. Paraffin sections of aortas and frozen aortic sinus sections were stained for receptors for advanced glycation endproducts (RAGE), nitrotyrosine, VEGF, Nox2, α-smooth muscle actin (α-SMA), and F4/80 to detect macrophages. Primary and secondary antibodies and dilutions are described in supplemental Table I. For immunohistochemical methods please see the supplemental materials. Images were visualized under light microscopy and quantitated using Image Pro-Plus 6.0. On average, 3 sections of each region were assessed per mouse for aortic lesions, whereas an average of 5 cross-sections per mouse were analyzed within the aortic sinus. Results were calculated as percentage positively stained tissue in the media or sinus region and expressed relative to nondiabetic apoE−/− mice that were arbitrarily assigned a value of 1.
Quantitative Reverse-Transcription Polymerase Chain Reaction
Total RNA extraction, preparation of DNA-free RNA, RNA reverse transcription, and gene expression of VEGF, RAGE, Nox2, α-SMA, superoxide dismutase-1 (SOD-1), GPx1, and catalase by qRT-PCR were as described previously12 (please see supplemental Table II, which describes probes and primers).
Immunoblot Analysis of Individual Aortas
Aortas were homogenized in RIPA buffer together with protease inhibitor and cytosolic fractions obtained after 2 sequential centrifugations at 1500g and 10 000g. Concentrations were determined using a BCA Protein Assay Kit and equivalent amounts of protein loaded per lane. Primary and secondary antibodies for SOD1, GPx1, catalase, and β-actin are described in supplemental Table I. Proteins were detected by chemiluminescence and band intensities determined by densitometry.
In Vitro Experiments in Human Aortic Endothelial Cells
HAECs (passage 4 to 8) seeded into flat bottom dishes precoated with gelatin, were maintained in endothelial cell media. For growth conditions please see the supplemental materials. Cells were pretreated with 0.03 μmol/L ebselen for 30 minutes after which 100 μmol/L hydrogen peroxide was added for an additional for 30 minutes in the presence of ebselen. In separate experiments, cells were either left untreated or treated with 100 μmol/L H2O2 for 30 minutes; 0.03 μmol/L DMSO (ebselen vehicle control) or 0.03 μmol/L ebselen alone for 60 minutes. For description of the detection of Phospho-IKKαser180βser181 (P-IKK), IκB, phospho-JNK, TNF-α and Nox2 proteins on crude lysates, please see the supplemental materials. Antibodies against P-IKK, IκB, phospho-JNK, TNF-α, and Nox2 are described in supplemental Table I.
Data were analyzed using 1-way analysis of variance (ANOVA) using GraphPad Prism 5 (GraphPad Software Inc). Student-Newman-Keuls Multiple Comparison determined comparisons of group means. P<0.05 was considered statistically significant. Results are expressed as mean±SEM.
Statement of Responsibility
The authors had full access to the data and take responsibility for its integrity. All authors have read and agree to the manuscript as written.
Phenotypic Assessment of Ebselen-Gavaged ApoE−/− Mice and Metabolic Parameters After 20 Weeks of Diabetes
Body weights and metabolic parameters at the conclusion of the 20-week study are shown in supplemental Table III. Nongavaged, ebselen-gavaged, and cellulose-gavaged diabetic mice have significantly lower body weights compared with nondiabetic controls (P<0.001). However, the ≈35% diabetes-induced weight loss of the nongavaged diabetic group was significantly reduced to ≈23% after ebselen-gavage (P<0.05). Diabetes was associated with a significant increase in glycohemoglobin, plasma glucose, total cholesterol, HDL, and LDL (P<0.001) compared with nondiabetic counterparts. Gavage with either ebselen or cellulose did not affect these parameters in diabetic mice. Based on these findings, we conclude that ebselen does not affect glucose or lipid pathways of diabetic apoE−/− mice, in agreement with other studies where ebselen had no effect on metabolic control.19
Assessment of Aortic Atherosclerotic Lesions
En Face Analysis of Aortic Tree
After 20 weeks of diabetes, total aortic plaque was significantly increased in apoE−/− aortas (P<0.001; Figure 1). Regional plaque evaluation showed highly significant increases in the arch, thoracic, and abdominal regions in nongavaged diabetic mice compared with nondiabetic controls (P<0.001). Ebselen reduced total aortic plaque by ≈50% in diabetic mice compared with nongavaged diabetic aortas (P<0.001), with significant regional reductions in plaque area of ≈50% in the arch (P<0.001), thoracic (P<0.001), and abdominal aorta (P<0.05) of diabetic mice compared with nongavaged diabetic aortas. Gavaging diabetic mice with cellulose had no effect on total (Figure 1) or regional plaque area (data not shown), indicating that the effects seen with ebselen are drug-related.
Aortic Sinus Region
After 10 weeks of diabetes, there was a small albeit nonsignificant increase in lesion size within the aortic sinus of diabetic mice compared with nondiabetic controls (Figure 2). However, there was no significant difference in lesion size between nongavaged and ebselen-gavaged diabetic mice. Analysis at an even earlier time-point (5 weeks of diabetes) gave similar results (data not shown). After 20 weeks of diabetes, there was a significant increase in lesion size of nongavaged diabetic mice compared with nondiabetic controls (P<0.01). Ebselen had no effect on lesion size in diabetic mice (P>0.05). Similarly, cellulose-gavage had no effect on diabetic lesion size compared with nongavaged diabetic aortas (data not shown).
Effect of Ebselen on Nitrotyrosine and Nox2 in Aortic Tissue
Immunostaining for nitrotyrosine was detected predominantly in the endothelial cell–containing intimal layer in control mice (Figure 3). After 10 weeks of diabetes, nitrotyrosine staining of apoE−/− aorta was significantly increased ≈2.5-fold when compared with nondiabetic controls (P<0.001) with increased staining observed primarily in the smooth muscle–rich medial layer but also in the intimal layer. Ebselen reduced nitrotyrosine levels by ≈50% in diabetic aortas compared with nongavaged diabetic aortas (P<0.01). Cellulose-gavage of diabetic mice had no significant effect on aortic nitrotyrosine levels compared with nongavaged diabetic aortas (data not shown). Similarly, after 10 weeks of ebselen-gavage, nitrotyrosine was reduced by ≈60% within diabetic aortic sinuses compared with nongavaged diabetic sinuses (P<0.001). To assess whether protection by ebselen is evident earlier in the progression toward larger lesion development within the sinus, we also analyzed the level of nitrotyrosine in 5-week-old diabetic mice. We found, in agreement with the 10-week data, that ebselen reduced nitrotyrosine staining within the diabetic aortic sinus compared with nongavaged diabetic mice (diabetic 1±0.2 versus diabetic+ebselen 0.6±0.1, P=0.05, n=4 to 6 aorta per group).
After 10 weeks of diabetes, aortic Nox2 mRNA levels increased ≈7-fold when compared with nondiabetic controls (P<0.001; supplemental Figure I). In control mice, Nox2 immunostaining was seen particularly in the intimal layer, whereas in diabetic mice there was a marked increase in Nox2 immunostaining both in the intimal and medial layers. Ebselen reduced Nox2 mRNA and protein levels by ≈50% in diabetic aortas compared with nongavaged diabetic aortas (P<0.01). Similarly, after 10 weeks of ebselen-gavage, Nox2 protein levels were reduced by ≈70% within diabetic aortic sinuses compared with nongavaged diabetic sinuses (P<0.01). Cellulose-gavage of diabetic mice had no significant effect on aortic Nox2 mRNA and protein levels compared with nongavaged diabetic aortas (data not shown).
Effect of Ebselen on Proatherogenic Markers
Receptor for Advanced Glycation End Products
After 10 weeks of diabetes, apoE−/− aortas showed ≈4- and 1.6-fold increases in RAGE mRNA (P<0.05) and protein levels, respectively (P<0.01), when compared with nondiabetic controls (supplemental Figure II). Aortic RAGE protein expression was clearly seen in the endothelial cell–containing intimal layer in control mice, whereas in diabetic mice, RAGE expression was more widespread with a marked increase in the aortic medial layer. Ebselen significantly reduced these increases back to basal levels at both the mRNA (P<0.05) and protein level (P<0.01) in diabetic apoE−/− mice. Cellulose-gavage had no significant effect on diabetic RAGE protein levels compared with nongavaged diabetic aortas (data not shown). RAGE protein was significantly increased in the aortic sinus after 10 weeks of diabetes compared with nondiabetic controls (P<0.05). Ebselen did not affect RAGE protein in diabetic sinuses compared with nongavaged diabetic sinuses (P>0.05). As seen after 10 weeks of diabetes, immunohistochemical analysis of 5-week-old diabetic sinuses showed that ebselen did not significantly affect RAGE protein levels at this earlier time point (diabetic aortic sinus 1±0.3 versus diabetic sinus + ebselen 1.0±0.2, P>0.05, n=4 to 6 aortas per group).
Vascular Endothelial Growth Factor
After 10 weeks of diabetes, apoE−/− aortas showed ≈4- and 2.5-fold increases in VEGF mRNA (P<0.05) and protein levels, respectively (P<0.01), when compared with nondiabetic controls (supplemental Figure III). Ebselen significantly attenuated the diabetic-induced increase of VEGF at both the mRNA (P<0.05) and protein level (P<0.05) back to levels similar to nondiabetic apoE−/− mice. Increased VEGF expression was observed in both intimal and medial layers of the diabetic aorta, which was attenuated in both layers of the aorta after ebselen treatment. Diabetic VEGF protein levels were unaffected by cellulose-gavage compared with nongavaged diabetic aortas (data not shown). Within the sinus region, VEGF levels were slightly reduced after 10 weeks of diabetes, although this was not significant (P>0.05). After 10 weeks of ebselen-gavage, VEGF levels were significantly reduced compared with nongavaged diabetic aortas (P<0.05).
Effect of Ebselen on Plaque and Vessel Wall Cellularity
After 10 weeks of diabetes, ebselen significantly reduced both mRNA and protein levels of α-SMA in diabetic apoE−/− aortas (P<0.001 for both; supplemental Figure IV). Within the sinus region, after 10 weeks of ebselen-gavage, α-SMA levels were significantly reduced compared with nongavaged diabetic aortas (P<0.01). Cellulose gavage had no effect on diabetic α-SMA protein levels (data not shown).
After 10 weeks of diabetes, ebselen significantly reduced F4/80 staining, a macrophage marker, in diabetic apoE−/− aortas (P<0.05; supplemental Figure V). Within the sinus region, F4/80 staining was confined to the plaque area only. Ebselen reduced F4/80 staining within the plaques of diabetic sinuses (P<0.05). Diabetic F4/80 levels were unaffected by cellulose-gavage at all sites investigated (data not shown).
Effect of Ebselen on Antioxidants
No significant difference in SOD1 mRNA was detected in aortas after 10 weeks of diabetes when compared with nondiabetic aortas (Figure 4A). Ebselen-gavage did not affect SOD1 mRNA levels. Similarly, after 5 weeks (Figure 4D) and 10 weeks (data not shown) of diabetes, SOD1 protein levels did not differ significantly between the 3 groups.
Diabetes did not affect GPx1 mRNA or protein levels of apoE−/− mice (P>0.05) after 5 and 10 weeks of diabetes (Figure 4B and 4E). GPx1 mRNA expression showed a trend toward increased expression in ebselen-gavaged diabetic aortas, albeit nonsignificant (P=0.065), whereas GPx1 protein levels were significantly increased ≈2-fold (P<0.01) after ebselen-gavage of diabetic mice compared with nondiabetic aortas.
Diabetes did not affect catalase mRNA or protein levels of apoE−/− mice (P>0.05) after 5 and 10 weeks of diabetes (Figure 4C and 4F). Catalase mRNA and protein levels were significantly increased approximately 6-fold and 2-fold, respectively (P<0.05 for both), after ebselen treatment of diabetic apoE−/− aortas compared with nondiabetic aortas. Cellulose-gavage of diabetic mice had no significant effect on any of the antioxidants investigated (data not shown).
In Vitro Analysis of Ebselen in HAECs
Initial experiments established optimal treatment time with H2O2 (30 minutes, data not shown) and dosage (100 μmol/L; supplemental Figure VI) for increased expression of P-IKK. Pretreatment with a range of ebselen concentrations (0.03 μmol/L-0.1 μmol/L, data not shown) for 30 minutes followed by treatment with 100 μmol/L H2O2 significantly reduced the H2O2-mediated increase in the phosphorylation of IKK. Figure 5A shows the reduction in IKK phosphorylation after pretreatment with 0.03 μmol/L ebselen in the presence of 100 μmol/L H2O2 (P<0.05). Hydrogen peroxide treatment resulted in a significant decrease in the levels of the inhibitory subunit IκB (supplemental Figure VII), in agreement with the known degradation of IκB in response to H2O2,21 whereas ebselen abrogated this H2O2-mediated decrease in IκB. In addition, 0.03 μmol/L ebselen significantly reduced the H2O2-mediated increase in TNF-α (Figure 5B; P<0.01), Nox2 (Figure 5C; P<0.05), and significantly reduced JNK phosphorylation (Figure 5D; P<0.001).
This study has shown that administration of the synthetic antioxidant, ebselen, to diabetic apoE-deficient mice attenuates lesion formation in most regions of the aorta, suggesting that ebselen is an effective antiatherogenic agent against diabetic macrovascular disease. This was accompanied by a reduction in oxidative stress as reflected by reduced nitrotyrosine levels as well as a reduction in the Nox2 subunit of NADPH oxidase (Nox), an enzyme implicated in the generation of vascular ROS. Diabetes was associated with a significant increase in staining within the aortic media for both nitrotyrosine and Nox2, which was attenuated by ebselen. Similar effects were also seen within the intimal layer, most likely a result of increased endothelial cell expression of nitrotyrosine and Nox2 in response to diabetes, which was again attenuated by ebselen. Furthermore, the cellularity associated with a proatherosclerotic phenotype (α-smooth muscle cells and macrophages) was reduced by ebselen in association with a reduction in expression of proatherosclerotic mediators RAGE and VEGF in both the intimal and medial layers of the aorta. However, this study has also revealed that ebselen elicits its antiatherogenic effects in a site-specific manner because ebselen did not affect lesions within the aortic sinus. The results of this study support a growing list of studies where site-specific effects of modulators of atherosclerosis have been observed.22 For example, in a study by Witting et al,23 the lipid-lowering antioxidant probucol decreased lesion formation in most aortic regions but resulted in increased lesion formation within the aortic sinus. Indeed, it has been postulated that parts of the aorta behave differently because of local hemodynamic factors such as low shear stress, turbulence, oscillating flow, and inherent properties of the vessel wall.22,23 It is therefore feasible that such factors or the availability of ebselen within the sinus may have prevented or masked its efficacy in the diabetic apoE−/− sinus in this study. However, importantly, our data showing a reduction in atherosclerotic plaque in most regions of the aorta after 20 weeks of diabetes support the notion of Blankenberg et al7 that bolstering GPx-like activity reduces atherosclerosis. This is in agreement with our recent data where lack of GPx1 greatly accelerated plaque deposition in the aorta of apoE−/−GPx1−/− mice.12 Our current study therefore shows for the first time that an antioxidant with GPx-like activity14 reduces atherosclerosis in a diabetic setting.
To investigate the mechanisms whereby ebselen reduces atherosclerosis in a diabetic milieu, its role as an antioxidant and its role in a number of cellular and molecular processes linked to DAA were examined in our in vivo model. To determine the effect of ebselen on cellularity, we assessed the gene and protein expression of α-smooth muscle actin reflecting the presence of smooth muscle cells, as well as assessing F4/80 immunostaining, a marker of macrophage infiltration. In both instances, ebselen reduced α-smooth muscle actin as well as F4/80 immunostaining within the diabetic aorta, with reductions also seen within the aortic sinus region. Our data are in agreement with other studies where ebselen reduced proliferation of vascular smooth muscle cells,24 and limited inflammatory cell involvement.25
As an indicator of oxidative stress, we investigated peroxynitrite-mediated damage of proteins by assessing nitrotyrosine content. In agreement with our previous findings,12 the present study demonstrates that diabetes is associated with increased nitrotyrosine levels in the aortic wall and sinus region. Importantly, ebselen reduced nitrotyrosine levels back to nondiabetic levels, supporting the observations of Brodsky et al19 where ebselen decreased nitrotyrosine in Zucker diabetic rats, albeit a model not associated with atherosclerosis. Our data therefore highlight the importance of ebselen in its role as a peroxynitrite reductase and strengthens the notion that ebselen functions in a similar fashion to GPx1.15 In addition, we assessed vascular Nox2 gene and protein expression because Nox is a major source of superoxide production in vascular tissue26 and increased expression of Nox2 has been linked to increased oxidative stress in diabetic apoE−/− mice.27 Indeed, in this study ebselen downregulated this important oxidant-generating system in the vasculature. Although ebselen has previously been shown to inactivate Nox through direct interaction and binding to the protein,28 our study reveals an additional ability by ebselen to reduce expression of an important protein subunit of the Nox enzyme complex, most likely at the transcriptional level. This effect occurred both within the aorta as well as specifically within the aortic sinus region. Thus, our data suggest that ebselen prevents ROS formation via (1) a reduced expression of certain subunits of the Nox enzyme and (2) in limiting ROS-mediated damage via its ability to act as a ROS scavenger. Given the known antioxidant properties of ebselen13,14 and to further probe the ability of ebselen to mediate ROS scavenging, we also investigated whether ebselen had any direct effect on the endogenous antioxidant pathway.29 Ebselen did not affect the major superoxide-removing antioxidant, SOD1, but increased GPx1 protein levels within diabetic aortas. Interestingly, ebselen also increased catalase levels, and this may be an additional mechanism whereby ebselen reduces oxidative stress. Our data has therefore shown that in its capacity to reduce oxidative stress, ebselen exerts a positive effect on genes such as GPx1 and catalase while reducing the expression of prooxidant genes such as Nox2. For a detailed discussion on the functional duality of ebselen in enhancing expression of some genes but suppressing others, please see the supplemental materials.
Diabetes-associated atherosclerosis is accompanied by increased transcription and translation of important proatherogenic mediators such as RAGE, which has been strongly implicated in the pathogenesis of diabetes by amplifying inflammatory and immune responses.3 This may be partly attributable to advanced glycation end-products (AGEs) upregulating RAGE.3 Importantly, Mukherjee et al30 showed that tumor necrosis factor–α (TNF-α) induces RAGE expression in HUVECs via activation of Nox to generate ROS. In agreement with published data,3 this study has shown an increase of RAGE at both the transcriptional and translational level in all regions of diabetic apoE−/− aorta. We show that Ebselen attenuates RAGE expression within diabetic vessel walls. It is tempting to speculate that ebselen, through its antioxidant function to reduce ROS such as peroxynitrite, affects proatherogenic pathways such as RAGE. However, this relationship did not hold true for the aortic sinus where RAGE levels were unaffected by ebselen, despite reductions in nitrotyrosine. It may be that additional factors known to affect this region22 influence RAGE expression independent of peroxynitrite-mediated oxidative stress, and that upregulation of RAGE prevents ebselen-mediated reduction of lesions within the aortic sinus. It should also be noted that although RAGE inhibition has been reported to reduce nitrotyrosine levels,31 one cannot assume the converse where changes in ROS, and in particular decreased nitrotyrosine levels, would automatically lead to RAGE depletion. In the context of the presumed mode of action of ebselen, the reduction in nitrotyrosine in the aortic sinus without a concomitant decrease in RAGE expression is most likely attributable to the known peroxynitrite reductase activity of ebselen, which appears to be independent of the RAGE pathway at this site.
To further explore the mechanisms implicated in DAA, we investigated VEGF, a growth factor involved in angiogenesis and upregulated in DAA.12 VEGF stimulates the proliferation and migration of endothelial cells4 thereby promoting lesion progression.20 In this study, VEGF was increased in the diabetic vessel wall at both the transcriptional and translational level, consistent with our previous results.12 However, VEGF was not increased within the diabetic aortic sinus suggesting that atherogenesis may be independent of this growth factor at this site. Importantly, ebselen reduced VEGF expression at all sites investigated, including the sinus region. Our findings concur with previously published in vitro studies where ebselen reduced VEGF levels in smooth muscle cells via a ROS-related mechanism.20
Our in vitro analysis of HAECs suggests that one of the key mechanisms whereby ebselen confers its antiatherogenic effects is via modulation of the transcription factor NF-κB. Pretreatment with ebselen reduced the increased H2O2-mediated phosphorylation of the IκB-kinase (IKK) complex on critical activatory residues. Because IKK is a key regulator of NF-κB activation,32 it is predicted that by reducing the phosphorylation of IKK, ebselen plays a pivotal role in anchoring NF-κB in the cytoplasm thereby preventing the activation of proinflammatory genes. In addition, we show that ebselen prevents the ROS-mediated degradation of the NF-κB inhibitory subunit IκB, providing further proof that ebselen modulates NF-κB activity. Therefore it is likely that ebselen affects downstream cellular targets regulated by NF-κB. To explore this notion, we investigated the effects of ebselen on Nox2 expression in HAECs, because Nox2 is known to be regulated by NF-κB,33 and in our animal model, ebselen was associated with decreased Nox2 gene and protein expression. Our in vitro data showing reduced Nox2 protein levels after ebselen treatment does indeed lend support for this idea and supports our in vivo findings. We also explored the consequences of ebselen pretreatment on the cytokine TNF-α, because TNF-α is an important diabetes-associated proinflammatory mediator and is involved in the activation of NF-κB.34 Our in vitro data showed that the H2O2-induced upregulation of TNF-α was reduced by ebselen. Finally, our in vitro analysis has also shown that ebselen blocks the H2O2-mediated phosphorylation of c-Jun–N-terminal kinase (JNK), a kinase involved in the activation of the transcription factor AP-1. This is important given that several laboratories suggest a role for JNK in TNFα-mediated endothelial cell activation35 and in particular through interactions of AP-1 with NF-κB,36 and because others have suggested that the effect of ebselen on JNK may be cell type specific.37,38 Taken together, our results with ebselen have implications not only for inflammatory genes known to be regulated by these pathways,39 but also on the proatherosclerotic pathway itself, because inflammatory events are integrally linked with atherosclerosis.
In summary, this study has shown that ebselen reduces DAA throughout the aortic tree of diabetic apoE−/− mice but not within the aortic sinus, thereby suggesting a site-specific antiatherogenic effect of ebselen, as has been reported for other putative antiatherosclerotic agents.22 These site-specific observations raise interesting questions about local factors that contribute to disease, and suggest that within the sinus region, atherogenesis and oxidative stress may not be associated. Finally, it cannot be excluded that ebselen elicits it antiatherogenic effects via additional properties that have been previously ascribed to ebselen, such as its inhibitory effect on 15-lipoxygenase.13 Importantly, the finding that ebselen inhibits atherosclerosis in the majority of the aorta in apoE−/− mice indicates that this antioxidant may be a useful antiatherogenic therapeutic in DAA.
Sources of Funding
J.d.H. is a recipient of a research grant from Merck Research Laboratories and a grant from the National Health and Medical Research Council of Australia (#526656). This work is supported by a Juvenile Diabetes Research Foundation grant, awarded to M.E.C.
Dr Ismail Kola is a Senior Vice-President for Schering-Plough. S.P. has substantial revenues from drugs that treat dyslipidaemias; however, no activities fall within the specific area covered by this article. The remaining authors report no conflicts.
Received April 9, 2008; revision accepted March 18, 2009.
Baynes JW. Role of oxidative stress in development of complications in diabetes. Diabetes. 1991; 40: 405–412.
Mayer-Davis EJ, Monaco JH, Marshall JA, Rushing J, Juhaeri. Vitamin C intake and cardiovascular disease risk factors in persons with non-insulin-dependent diabetes mellitus. From the Insulin Resistance Atherosclerosis Study and the San Luis Valley Diabetes Study. Prev Med. 1997; 26: 277–283.
Schnabel R, Lackner KJ, Rupprecht HJ, Espinola-Klein C, Torzewski M, Lubos E, Bickel C, Cambien F, Tiret L, Munzel T, Blankenberg S. Glutathione peroxidase-1 and homocysteine for cardiovascular risk prediction: results from the AtheroGene study. J Am Coll Cardiol. 2005; 45: 1631–1637.
Hamanishi T, Furuta H, Kato H, Doi A, Tamai M, Shimomura H, Sakagashira S, Nishi M, Sasaki H, Sanke T, Nanjo K. Functional variants in the glutathione peroxidase-1 (GPx-1) gene are associated with increased intima-media thickness of carotid arteries and risk of macrovascular diseases in japanese type 2 diabetic patients. Diabetes. 2004; 53: 2455–2460.
Nemoto M, Nishimura R, Sasaki T, Hiki Y, Miyashita Y, Nishioka M, Fujimoto K, Sakuma T, Ohashi T, Fukuda K, Eto Y, Tajima N. Genetic association of glutathione peroxidase-1 with coronary artery calcification in type 2 diabetes: a case control study with multi-slice computed tomography. Cardiovasc Diabetol. 2007; 6: 23.
Lewis P, Stefanovic N, Pete J, Calkin AC, Giunti S, Thallas-Bonke V, Jandeleit-Dahm KA, Allen TJ, Kola I, Cooper ME, de Haan JB. Lack of the antioxidant enzyme glutathione peroxidase-1 accelerates atherosclerosis in diabetic apolipoprotein E-deficient mice. Circulation. 2007; 115: 2178–2187.
Sies H, Sharov VS, Klotz LO, Briviba K. Glutathione peroxidase protects against peroxynitrite-mediated oxidations. A new function for selenoproteins as peroxynitrite reductase. J Biol Chem. 1997; 272: 27812–27817.
Yamaguchi T, Sano K, Takakura K, Saito I, Shinohara Y, Asano T, Yasuhara H. Ebselen in acute ischemic stroke: a placebo-controlled, double-blind clinical trial. Ebselen Study Group. Stroke. 1998; 29: 12–17.
Davis MT, Bartfay WJ. Ebselen decreases oxygen free radical production and iron concentrations in the hearts of chronically iron-overloaded mice. Biol Res Nurs. 2004; 6: 37–45.
Brodsky SV, Gealekman O, Chen J, Zhang F, Togashi N, Crabtree M, Gross SS, Nasjletti A, Goligorsky MS. Prevention and reversal of premature endothelial cell senescence and vasculopathy in obesity-induced diabetes by ebselen. Circ Res. 2004; 94: 377–384.
Khatri JJ, Johnson C, Magid R, Lessner SM, Laude KM, Dikalov SI, Harrison DG, Sung HJ, Rong Y, Galis ZS. Vascular oxidant stress enhances progression and angiogenesis of experimental atheroma. Circulation. 2004; 109: 520–525.
VanderLaan PA, Reardon CA, Getz GS. Site specificity of atherosclerosis: site-selective responses to atherosclerotic modulators. Arterioscler Thromb Vasc Biol. 2004; 24: 12–22.
Menshikov M, Plekhanova O, Cai H, Chalupsky K, Parfyonova Y, Bashtrikov P, Tkachuk V, Berk BC. Urokinase plasminogen activator stimulates vascular smooth muscle cell proliferation via redox-dependent pathways. Arterioscler Thromb Vasc Biol. 2006; 26: 801–807.
Zhao R, Masayasu H, Holmgren A. Ebselen: a substrate for human thioredoxin reductase strongly stimulating its hydroperoxide reductase activity and a superfast thioredoxin oxidant. Proc Natl Acad Sci U S A. 2002; 99: 8579–8584.
De Haan JB, Crack PJ, Flentjar N, Iannello RC, Hertzog PJ, Kola I. An imbalance in antioxidant defense affects cellular function: the pathophysiological consequences of a reduction in antioxidant defense in the glutathione peroxidase-1 (Gpx1) knockout mouse. Redox Rep. 2003; 8: 69–79.
Bucciarelli LG, Wendt T, Qu W, Lu Y, Lalla E, Rong LL, Goova MT, Moser B, Kislinger T, Lee DC, Kashyap Y, Stern DM, Schmidt AM. RAGE blockade stabilizes established atherosclerosis in diabetic apolipoprotein E-null mice. Circulation. 2002; 106: 2827–2835.
Anrather J, Racchumi G, Iadecola C. NF-kappaB regulates phagocytic NADPH oxidase by inducing the expression of gp91phox. J Biol Chem. 2006; 281: 5657–5667.
Hacker H, Karin M. Regulation and function of IKK and IKK-related kinases. Sci STKE. 2006; 2006: re13.
Min W, Pober JS. TNF initiates E-selectin transcription in human endothelial cells through parallel TRAF-NF-kappa B and TRAF-RAC/CDC42-JNK-c-Jun/ATF2 pathways. J Immunol. 1997; 159: 3508–3518.
Read MA, Whitley MZ, Gupta S, Pierce JW, Best J, Davis RJ, Collins T. Tumor necrosis factor alpha-induced E-selectin expression is activated by the nuclear factor-kappaB and c-JUN N-terminal kinase/p38 mitogen-activated protein kinase pathways. J Biol Chem. 1997; 272: 2753–2761.
Yoshizumi M, Kogame T, Suzaki Y, Fujita Y, Kyaw M, Kirima K, Ishizawa K, Tsuchiya K, Kagami S, Tamaki T. Ebselen attenuates oxidative stress-induced apoptosis via the inhibition of the c-Jun N-terminal kinase and activator protein-1 signalling pathway in PC12 cells. Br J Pharmacol. 2002; 136: 1023–1032.