α-Lipoic Acid Prevents Endothelial Dysfunction in Obese Rats via Activation of AMP-Activated Protein Kinase
Objective— Lipid accumulation in vascular endothelial cells may play an important role in the pathogenesis of atherosclerosis in obese subjects. We showed previously that α-lipoic acid (ALA) activates AMP-activated protein kinase (AMPK) and reduces lipid accumulation in skeletal muscle of obese rats. Here, we investigated whether ALA improves endothelial dysfunction in obese rats by activating AMPK in endothelial cells.
Methods and Results— Endothelium-dependent vascular relaxation was impaired, and the number of apoptotic endothelial cells was higher in the aorta of obese rats compared with control rats. In addition, triglyceride and lipid peroxide levels were higher, and NO synthesis was lower. Administration of ALA improved all of these abnormalities. AMPK activity was lower in aortic endothelium of obese rats, and ALA normalized it. Incubation of human aortic endothelial cells with ALA activated AMPK and protected cells from linoleic acid–induced apoptosis. Dominant-negative AMPK inhibited the antiapoptotic effects of ALA.
Conclusions— Reduced AMPK activation may play an important role in the genesis of endothelial dysfunction in obese rats. ALA improves vascular dysfunction by normalizing lipid metabolism and activating AMPK in endothelial cells.
Central obesity is associated with increased cardiovascular morbidity and mortality.1 Subjects with central obesity show increased lipid accumulation in nonadipose tissues such as muscle, liver, and pancreatic islets.2,3 It has been proposed that increased lipid accumulation in vascular tissue and the consequent increase in oxidative stress may be a missing link between obesity and atherosclerosis.4
α-Lipoic acid (ALA), a naturally occurring short chain fatty acid containing sulfhydryl groups, has potent antioxidant capacities. ALA is an essential cofactor for mitochondrial respiratory enzymes and improves mitochondrial function.5,6 We showed in rodents recently that chronic ALA treatment significantly reduced body weight gain primarily by decreasing food intake, and that this effect was mediated by the effect of ALA to decrease AMP-activated protein kinase (AMPK) activity in the hypothalamus.7 AMPK is a major regulator of cellular energy metabolism.8,9 When activated, AMPK increases glucose uptake and fatty acid oxidation8,9 and decreases lipid accumulation in the tissues.10 In contrast to the effect of ALA to inhibit AMPK in the hypothalamus, ALA activated AMPK in skeletal muscle, resulting in enhanced fatty acid oxidation and reduced lipid accumulation.11 AMPK is expressed in vascular endothelial cells,12 and AMPK dysregulation has been suggested to contribute to endothelial dysfunction.10 We hypothesized that ALA may improve vascular function via activation of AMPK in vascular endothelial cells. In the present study, we examined whether endothelial dysfunction in obese Otsuka Long Evans Tokushima Fatty (OLETF) rats is associated with impaired AMPK activation in vascular endothelial cells and whether these defects can be normalized by ALA treatment.
Materials and Methods
For details regarding animal and cell experiments, measurement of various metabolic parameters, and markers for apoptosis and reactive oxygen species (ROS), please see the online supplement, available at http://atvb.ahajournals.org.
Four-week-old male OLETF rats and their lean controls, Long Evans Tokushima Otsuka (LETO) rats, were obtained from Tokushima Research Institute, Otsuka Pharmaceutical Co. (Tokushima, Japan). All experimental procedures were approved by the institutional animal care and use committee of the Asan Institute for Life Sciences.
Vascular Function Study
Endothelium-dependent and -independent vasorelaxations were measured using an isometric force displacement transducer (Hugo Sachs Elektronik KG D-7806) as described previously.13
Electron Microscopy and TUNEL Staining
Apoptosis in the aortic endothelium was examined by electron microscopy and TUNEL staining.
Human aortic endothelial cells (HAECs) were obtained from BioWhittaker Inc. and maintained in the endothelial basal medium (BioWhittaker Inc.) supplemented with various growth factors and 2% FBS.
Western Blot Analysis of AMPK Phosphorylation and Isoform-Specific AMPK Activity
Phosphorylation and protein levels of AMPK were assayed by Western blot analysis using the antibodies directed against phosphopeptides based on the amino acid sequence surrounding Thr172 of the α-subunit of human AMPK and the antibody against α-subunit of human AMPK (Cell Signaling), respectively. For isoform-specific AMPK activity measurement, we immunoprecipitated lysates of aortic endothelium (200 μg protein each) with specific antibodies (Upstate Biotechnology) against the α1- and α2-catalytic subunits of AMPK bound to protein G–Sepharose beads. Kinase activity was measured using synthetic “SAMS” peptide and [γ-32P]ATP as described.14
Quantification of Cell Apoptosis
Apotosis was measured by ELISA assay (Boehringer Mannheim) and caspase 3, 8, and 9 activity assay using an Apo Alert caspase florescence assay kit (Clontech).
Adenoviral Gene Transfer of Dominant-Negative α1 and α2 AMPK
Plasmid encoding c-Myc–tagged forms of dominant-negative α1 and α2 AMPK were a kind gift from Dr J. Ha (Department of Molecular Biology, Kyung Hee University College of Medicine, Seoul, Korea). Adenoviruses containing β-galactosidase (Ad-β-gal) or mixture of dominant-negative α1 AMPK and α2 AMPK (Ad-DN-AMPK) were added to subconfluent HAECs at a concentration of 10 plaque-forming units (pfu) per cell for 1 hour at 37°C in DMEM without serum, as described previously.15
All data are shown as mean±SEM. Comparisons between 2 groups were analyzed using unpaired Student’s t tests, and among multiple groups by ANOVA, followed by a post hoc analysis using the Tukey’s multiple comparison test (SPSS). A P value <0.05 was considered to be statistically significant.
Compared with LETO rats, OLETF rats had higher body weight, blood pressure, and fasting plasma levels of triglyceride, free fatty acid (FFA), glucose, and insulin. In addition, plasma levels of oxidative stress markers (ie, 8-hydroxydeoxyguanosine and malondialdehyde) were higher and urinary NO excretion was lower in OLETF rats than in LETO rats (Table I, available online at http://atvb.ahajournals.org). ALA treatment completely normalized all of these changes in OLETF rats except for systolic blood pressure, which was substantially lowered but not to the level of LETO rats. These effects of ALA treatment seemed to be mediated by its effect to reduce food intake7 because pair feeding of OLETF rats resulted in exactly the same effects. Plasma adiponectin concentrations in OLETF rats were unaltered by ALA treatment.
Endothelium-Dependent Vascular Relaxation
Compared with LETO rats, OLETF rats showed reduced vasodilation of aortic rings in response to acetylcholine, indicating impaired endothelium-dependent vascular relaxation. The endothelium-dependent vasorelaxation in aortic rings of OLETF rats was substantially improved by ALA treatment (Figure 1A). Vascular relaxation in response to sodium nitroprusside, an NO donor, was not different among groups (Figure 1B), indicating that vascular reactivity to NO is intact in OLETF rats. Pair feeding also significantly improved endothelium-dependent vascular function in OLETF rats, but this improvement was significantly less than the improvement achieved by ALA treatment (Figure 1C).
NO Synthesis and Lipid Contents in Aortic Endothelium
To evaluate whether the changes in endothelium-dependent vasodilation were attributable to changes in NO synthesis, we measured acetylcholine-stimulated NO synthesis in isolated aortas. NO synthesis was significantly lower in aortas of OLETF rats compared with that of LETO rats (Figure 2A). NO synthesis was substantially increased in aortas of OLETF rats by ALA treatment. Pair feeding also significantly improved NO synthesis, but this effect was smaller than that of ALA treatment.
Triglyceride and lipid hydroperoxide levels in aortic endothelial cells of OLETF rats were significantly higher than those in LETO rats (Figure 2B and 2C). These changes were reversed completely by ALA treatment but only partially by pair feeding.
AMPK Phosphorylation and Activities in Aortic Endothelium
AMPK phosphorylation was lower in aortic endothelium of OLETF rats compared with that of LETO rats, and ALA treatment completely normalized it (Figure 2D). α1 and α2 AMPK activities were also lower in aortic endothelium of OLETF rats compared with those of LETO rats. ALA treatment increased α1 (Figure 2E) and α2 (Figure I, available online at http://atvb.ahajournals.org) AMPK activities. Interestingly, pair feeding had no effect on these measurements despite its effects on blood metabolic parameters, which were identical to those of ALA treatment (Table I). These data suggest that the effect of ALA treatment on AMPK phosphorylation and activities might be a direct effect of ALA, independent of the effects of ALA on body weight and metabolic parameters.
Effects of Acute Administration of ALA on AMPK Activation and Endothelial Function
To further demonstrate direct effects ALA on AMPK and vascular function, we administrated ALA to 15-week-old OLETF rats (n=12) by an intraperitoneal injection (75 mg/kg body weight). AMPK phosphorylation in aortic endothelium increased 2.3-fold 30 minutes after the ALA administration and remained elevated thereafter until 240 minutes (Figure 2F). Acute administration of ALA also improved endothelium-dependent vascular relaxation, increased NO production, and decreased lipid hydroperoxide levels in the isolated aorta (Figure II, available online at http://atvb.ahajournals.org).
Endothelial Cell Apoptosis
Many in vitro studies have suggested that endothelial cell apoptosis may be an important early event in the pathogenesis of atherosclerosis.16 Examination by electron microscopy revealed that lumenal walls of aortas of LETO rats were well covered with endothelial cells (Figure 3A). In contrast, in OLETF rats, multiple foci of endothelial cell loss were noticed on lumenal walls of the aorta, and the subendothelial matrix was exposed to the lumen (Figure 3B). Many of the endothelial cells showed apoptotic profiles with nuclear fragmentation and disconnection from neighboring endothelial cells and from underlying subendothelial matrix (Figure 3C). The ultrastructural characteristics of aortas of ALA-treated OLETF rats were between those of LETO rats and of OLEFT rats. Intercellular gap junctions and interdigitations with surrounding tissue were relatively well preserved (Figure 3D). TUNEL staining (Figure 3E) revealed that the number of apoptotic endothelial cells was significantly higher in aortas of OLETF rats than in those of LETO rats, and that ALA treatment of OLETF rats substantially lowered it (Figure 3F). Pair feeding also lowered the number of apoptotic endothelial cells, but this effect was significantly smaller than the effect of ALA treatment (Figure 3F).
AMPK and Apoptosis in Cultured Endothelial Cells
It was suggested that endothelial dysfunction in metabolic syndrome is mediated, at least in part, by elevated circulating FFA levels.17 Incubation of HAECs with 300 μmol/L linoleic acid led to a 2.5-fold increase in apoptosis (Figure 4A), which was accompanied by increases in caspase-3, caspase-9 (a measure of mitochondrial pathway), and caspase-8 (a measure of death receptor signaling pathway) activities (Figure 4B through 4D). The linoleic acid–induced increases in apoptosis and caspase activities were all inhibited by 0.5 mmol/L ALA. Linoleic acid decreased AMPK phosphorylation in HAECs, and ALA substantially reversed this decrease (Figure 4E). ALA increased AMPK phosphorylation rapidly (ie, within 30 minutes; Figure 4F) in HAECs as observed in vivo (Figure 2F).
Effects of Dominant-Negative α1 and α2 AMPK
Mixture of dominant-negative α1 and α2 AMPK nearly completely reversed the effects of ALA against the changes in intracellular triglyceride, ROS generation, and apoptosis induced by linoleic acid, indicating that these effects of ALA were mediated by AMPK activation (Figure 5A through 5D).
Effects of ALA on NAD(P)H Oxidase Activity and Mitochondrial Membrane Potential
Major sites of intracellular ROS generation are mitochondria and cell membrane NAD(P)H oxidase.18 The mitochondrial membrane potential increased rapidly and peaked at 30 minutes after exposure to linoleic acid compared with Ad-β-gal–treated control cells. The membrane potential then decreased and became lower than that of the control cells. ALA treatment nearly completely prevented these changes in membrane potential. On the other hand, pretreatment with Ad-DN-AMPK substantially reversed the effects of ALA on membrane potential (Figure 6A). Similarly, linoleic acid increased NAD(P)H oxidase activity, and ALA decreased it. The effect of ALA on NAD(P)H oxidase activity was prevented by Ad-DN-AMPK (Figure 6B).
Defective endothelium-dependent vascular relaxation is an early event in the development of atherosclerosis.19 We confirmed that endothelium-dependent vascular relaxation is impaired in obese OLETF rats20,21 and found that this impairment is associated with increased lipid accumulation and apoptosis and decreased NO synthesis and AMPK activities in endothelial cells. All of these alterations in endothelial cells and vascular dysfunction were substantially improved by ALA treatment.
Our previous study demonstrated strong anorexic effects of ALA in normal and obese rodents.7 In the present study, ALA reduced food intake (data not shown) and normalized body weight in OLETF rats. In addition, all of the metabolic changes (FFA, triglycerides, insulin, etc) observed in OLETF rats were normalized by ALA. The normalization of metabolic parameters seemed to be attributable to the reduction in food intake because it also occurred in pair-fed, untreated OLETF rats; the metabolic parameters were identical between ALA-treated and pair-fed OLETF rats. Pair feeding (or normalization of metabolic parameters) alone also improved vascular function and reduced the alterations (lipid accumulation, apoptosis, NO synthesis, etc) in endothelial cells of OLETF rats. However, these effects were only about a half of the effects of ALA treatment, suggesting that ALA improved vascular dysfunction in OLETF rats in part by reducing food intake but also by exerting additional effects. In regard to this, it is of extreme interest that decreased AMPK activities in endothelial cells of OLETF rats was completely normalized by ALA but little affected by pair feeding. In addition, intraperitoneal injection of ALA increased AMPK phosphorylation in aortic endothelial cells within 30 minutes and improved vascular dysfunction in OLETF rats. These data suggest that ALA directly activates AMPK in endothelial cells and that this effect leads to the improvement of vascular dysfunction in OLETF rats beyond that with pair feeding or metabolic normalization. However, to prove the causal relationship between AMPK activation and improvement of vascular function, it would be necessary to demonstrate the effect of selective inactivation of AMPK in aortic endothelial cells, as was done in in vitro experiment.
In cultured HAECs, linoleic acid decreased AMPK phosphorylation and increased triglyceride accumulation, apoptosis, and ROS generation. ALA prevented the linoleic acid–induced decrease in AMPK phosphorylation, and this effect was associated with normalization of triglyceride level, apoptosis, and ROS generation in the presence of linoleic acid. Furthermore, expression of dominant-negative AMPK in these cells led to inhibition of ALA effects. Together, these data suggest that AMPK activity in endothelial cells is an important regulator of endothelial function.
A key feature of defective endothelium-dependent vasodilation is reduced bioavailability of NO.19 Consistent with previous studies in other animal models of obesity and diabetes,22 aortas isolated from OLETF rats displayed attenuated vascular responses to acetylcholine but not to the direct-acting exogenous NO donor sodium nitroprusside. Urinary NO excretion and acetylcholine-stimulated NO synthesis in aortic endothelium were significantly reduced in OLETF rats compared with LETO rats. These results suggest that vascular reactivity to NO is intact, but NO bioavailability is reduced in OLETF rats. NO bioavailability is determined as the balance between NO production and removal. Evidence from experimental animals with diabetes suggests that the most likely mechanism underlying reduced NO bioavailability is inactivation of NO by oxygen-derived free radicals.23 Bakker et al4 proposed that increased availability of lipid (ie, triglyceride and long chain fatty acyl coenzyme A [LCAC]) in vascular endothelial cells could cause oxidative stress by affecting the mitochondrial respiratory chain. Supporting this hypothesis, the present study demonstrates that intracellular contents of lipid (ie, triglycerides) and the oxidative stress marker (ie, lipid peroxide) were elevated in aortic endothelium of OLETF rats.
Oxidative stress has been also implicated to play a major role in cellular apoptosis.16 The present study demonstrates that the number of apoptotic cells was significantly increased in aortic endothelium of OLETF rats. In cultured HAECs, linoleic acid increased intracellular triglyceride, oxidative stress, and apoptosis. Thus, increased circulating levels of FFA and triglycerides in obese animals may increase oxidative stress and apoptosis in vascular endothelial cells. This may further contribute to the decrease in NO bioavailability and vascular dysfunction (Figure III, available online at http://atvb.ahajournals.org).
We are the first to show that AMPK phosphorylation was decreased in aortic endothelium of obese rats and to suggest that it could be causally related to vascular dysfunction in obese animals. The mechanism by which this change occurs is unclear. However, because AMPK functions as a fuel sensor in the cell and is activated when cellular energy is depleted, it is conceivable that surplus cellular energy in association with high plasma FFA or glucose levels is responsible for reduced AMPK activity in aortic endothelium of OLETF rats. Consistent with this idea, we demonstrated that incubation of HAECs with linoleic acid significantly decreased AMPK activation. Reduced AMPK activity would decrease fatty acid oxidation by activating acetyl CoA carboxylase and increasing intracellular malonyl coenzyme A levels.24 Thus, increased FFA flux from the circulation and reduced FFA oxidation in aortic endothelial cells could lead to excessive accumulation of triglyceride and LCAC. This may be an early event that leads to a cascade of increased ROS generation, increased apoptosis, and decreased NO bioavailability in endothelial cells and vascular dysfunction, as discussed above.
ALA is well known as an antioxidant. ALA directly scavenges free radicals, chelates transition metal ion (eg, iron and copper), increases cytosolic glutathione and vitamin C levels, and prevents toxicities associated with their loss.5,6 In addition, the present study demonstrates that antioxidant action of ALA is, at least in part, mediated by its effect on AMPK. Major sites of intracellular ROS generation are mitochondria and cell membrane NAD(P)H oxidase.18 The mitochondrial respiratory chain generates ROS when the electrochemical gradient between the mitochondrial inner membrane is high and the rate of electron transport is limited.25 Consistent with this concept, recent studies from our group26,27 and others28 have shown that high glucose or linoleic acid leads to significant increases in mitochondrial membrane potential (hyperpolarization) and ROS generation. In the present study, ALA treatment nearly completely prevented hyperpolarization induced by linoleic acid, and pretreatment with Ad-DN-AMPK reversed the effect of ALA. As stated above, LCAC is known to impair the flow of electrons through the electron transfer chain,4 and the reduction of LCAC levels by AMPK activation would be responsible for improvement in electron transfer and reduction in ROS generation.
We also found that linoleic acid increases and ALA decreases NADPH oxidase activity. Again, DN-AMPK reversed the effect of ALA, implicating that this effect is AMPK mediated. The mechanism by which AMPK activation decreases NAD(P)H oxidase activity is presently unknown but may be attributable to its effect on the cellular NADH/NAD+ redox state.29 It can be assumed that linoleic acid–induced mitochondrial membrane hyperpolarization, which reflects impairment of mitochondrial electron transfer and respiration, would increase NAD(P)H oxidase activity by increasing the cytosolic NADH:NAD+ ratio.29 Conversely, normalization of mitochondrial membrane potential by AMPK activation would decrease NAD(P)H oxidase activity by reducing the cytosolic NADH:NAD+ ratio.
In conclusion, our results suggest that reduced AMPK activity in endothelial cells may play an important role in the genesis of vascular dysfunction in obese rats, and that ALA may improve vascular dysfunction in obese rats by activating AMPK in endothelial cells. This study provides a rationale for the therapeutic use of ALA for vascular dysfunction in obese subjects.
This study was supported by the National Research Laboratory grant from the Ministry of Science and Technology (M1040000000804J000000810) and grants from the Korea Ministry of Health and Welfare (02-PJ1-PG10-20708-0007) and the Asan Institute for Life Sciences, (01-122, 01-279, 05-006) Republic of Korea.
W.J.L. and I.K.L. contributed equally to this work.
- Received March 7, 2005.
- Accepted September 7, 2005.
Poulter N. Global risk of cardiovascular disease. Heart. 2003; 89 (suppl 2): ii2–ii5.
Kelley DE, Goodpaster BH. Skeletal muscle triglyceride. An aspect of regional adiposity and insulin resistance. Diabetes Care. 2001; 24: 933–941.
Unger RH, Zhou YT Lipotoxicity of beta-cells in obesity and in other causes of fatty acid spillover. Diabetes. 2001; 50 (suppl 1): S118–S121.
Winder WW. Energy-sensing and signaling by AMP-activated protein kinase in skeletal muscle. J Appl Physiol. 2001; 91: 1017–1028.
McGarry JD. Banting lecture 2001: dysregulation of fatty acid metabolism in the etiology of type 2 diabetes. Diabetes. 2002; 51: 7–18.
Shinozaki K, Nishio Y, Okamura T, Yoshida Y, Maegawa H, Kojima H, Masada M, Toda N, Kikkawa R, Kashiwagi A. Oral administration of tetrahydrobiopterin prevents endothelial dysfunction and vascular oxidative stress in the aortas of insulin-resistant rats. Circ Res. 2000; 87: 566–573.
Hayashi T, Hirshman MF, Fujii N, Habinowski SA, Witters LA, Goodyear LJ. Metabolic stress and altered glucose transport: activation of AMP-activated protein kinase as a unifying coupling mechanism. Diabetes. 2000; 49: 527–531.
Park JY, Takahara N, Gabriele A, Chou E, Naruse K, Suzuma K, Yamauchi T, Ha SW, Meier M, Rhodes CJ, King GL. Induction of endothelin-1 expression by glucose: an effect of protein kinase C activation. Diabetes. 2000; 49: 1239–1248.
Schafer M, Schafer C, Ewald N, Piper HM, Noll T. Role of redox signaling in the autonomous proliferative response of endothelial cells to hypoxia. Circ Res. 2003; 92: 1010–1015.
Sakamoto S, Minami K, Niwa Y, Ohnaka M, Nakaya Y, Mizuno A, Kuwajima M, Shima K. Effect of exercise training and food restriction on endothelium-dependent relaxation in the Otsuka Long-Evans Tokushima Fatty rat, a model of spontaneous NIDDM. Diabetes. 1998; 47: 82–86.
Ruderman NB, Saha AK, Vavvas D, Witters LA. Malonyl-CoA, fuel sensing, and insulin resistance. Am J Physiol. 1999; 276: E1–E18.
Arsenijevic D, Onuma H, Pecqueur C, Raimbault S, Manning BS, Miroux B, Couplan E, Alves-Guerra MC, Goubern M, Surwit R, Bouillaud F, Richard D, Collins S, Ricquier D. Disruption of the uncoupling protein-2 gene in mice reveals a role in immunity and reactive oxygen species production. Nat Genet. 2000; 26: 435–439.
Park JY, Park KG, Kim HJ, Kang HG, Ahn JD, Kim HS, Kim YM, Son SM, Kim IJ, Kim YK, Kim CD, Lee KU, Lee IK. The effects of the overexpression of recombinant uncoupling protein 2 on proliferation, migration and plasminogen activator inhibitor 1 expression in human vascular smooth muscle cells. Diabetologia. 2005; 48: 1022–1028.
Lee KU, Lee IK, Han J, Song DK, Kim YM, Song HS, Kim HS, Lee WJ, Koh EH, Song KH, Han SM, Kim MS, Park IS, Park JY. Effects of recombinant adenovirus-mediated uncoupling protein 2 overexpression on endothelial function and apoptosis. Cir Res. 2005; 96: 1200–1207.
Russell JW, Golovoy D, Vincent AM, Mahendru P, Olzmann JA, Mentzer A, Feldman EL. High glucose-induced oxidative stress and mitochondrial dysfunction in neurons. FASEB J. 2002; 16: 1738–1748.
Bassenge E, Sommer O, Schwemmer M, Bunger R. Antioxidant pyruvate inhibits cardiac formation of reactive oxygen species through changes in redox state. Am J Physiol Heart Circ Physiol. 2000; 279: H2431–H2438.