Impaired Endothelial Function and Microvascular Asymmetrical Dimethylarginine in Angiotensin II–Infused Rats
Effects of Tempol
Angiotensin (Ang) II causes endothelial dysfunction, which is associated with cardiovascular risk. We investigated the hypothesis that Ang II increases microvascular reactive oxygen species and asymmetrical dimethylarginine and switches endothelial function from vasodilator to vasoconstrictor pathways. Acetylcholine-induced endothelium-dependent responses of mesenteric resistance arterioles were assessed in a myograph and vascular NO and reactive oxygen species by fluorescent probes in groups (n=6) of male rats infused for 14 days with Ang II (200 ng/kg per minute) or given a sham infusion. Additional groups of Ang or sham-infused rats were given oral Tempol (2 mmol · L−1). Ang II infusion increased mean blood pressure (119±5 versus 89±7 mm Hg; P<0.005) and plasma malondialdehyde (0.57±0.02 versus 0.37±0.05 μmol · L−1; P<0.035) and decreased maximal endothelium-dependent relaxation (18±5% versus 54±6%; P<0.005) and hyperpolarizing (19±3% versus 29±3%; P<0.05) responses and NO activity (0.9±0.1 versus 1.6±0.2 U; P<0.01) yet enhanced endothelium-dependent contraction responses (23±5% versus 5±5%; P<0.05) and reactive oxygen species production (0.82±0.05 versus 0.15±0.03 U; P<0.01). Ang II decreased the expression of dimethylarginine dimethylaminohydrolase 2 and increased asymmetrical dimethylarginine in vessels (450±50 versus 260±35 pmol/mg of protein; P<0.01) but not plasma. Tempol prevented any significant changes with Ang II. In conclusion, Ang redirected endothelial responses from relaxation to contraction, reduced vascular NO, and increased asymmetrical dimethylarginine. These effects were dependent on reactive oxygen species and could, therefore, be targeted with effective antioxidant therapy.
Normal vessels display endothelium-dependent relaxation responses to acetylcholine (Ach). The endothelium-dependent relaxation factor (EDRF) is mediated by endothelial NO synthase (NOS). The mediators of the endothelium-dependent hyperpolarizing factor (EDHF) vary by species and vascular bed but can include hydrogen peroxide (H2O2),1 epoxyeicosatrienoic acids (EETs),2 and electromechanical coupling.3 These cause activation of calcium-dependent potassium channels on vascular smooth muscle cells (VSMCs) that can be blocked by a combination of apamin and charybdotoxin.2 Some vessels also display a cyclooxygenase (COX)-dependent relaxation response attributable to prostacyclin, although this was not apparent in rat afferent arterioles.2 An endothelium-dependent contracting factor (EDCF) occurs in models with increased reactive oxygen species (ROS). The EDCF in renal afferent arterioles from rabbits infused with angiotensin (Ang) II entailed a vasoconstrictor prostaglandin generated by COX that activated thromboxane prostanoid receptors (TP-Rs) on VSMCs.4
Ang II generates ROS in blood vessels that can bioinactivate NO.4 However, NOS-dependent EDRF responses of resistance arterioles are not consistently reduced4,5 by Ang infusion. Moreover, ROS can generate an EDCF.4,6 However, it is not clear whether prolonged inhibition of ROS can prevent the defects in endothelium-dependent relaxations and prevent the EDCF responses of resistance vessels from animals infused with Ang II.
Increasing evidence suggests that an endogenous inhibitor of NOS, asymmetrical dimethylarginine (ADMA), and its metabolism by dimethylarginine dimethylaminohydrolase (DDAH)7 may regulate vascular NO. Ang II infusion can increase circulating levels of ADMA.8,9 However, it is unclear whether increased ADMA levels in plasma or tissues can be prevented by an effective antioxidant. We reported recently that culture of VSMCs with Ang II increased cellular, but not medium, levels of ADMA.10 To resolve these issues, we assessed the 3 major endothelium-dependent pathways and related these responses to vascular NO and ROS activities and plasma and tissue ADMA concentrations in mesenteric resistance arteries in rats during a slow pressor infusion of Ang II. Other groups of sham and Ang II–infused rats received the antioxidant drug Tempol throughout.11
These experiments were designed to test the hypothesis that ROS, generated in microvascular resistance vessels during prolonged Ang II–induced hypertension, impairs EDRF, EDHF, and NO activity and increases microvascular ADMA and EDCF. We selected Tempol, which is a redox-cycling nitroxide, to reduce ROS and enhance vascular NO.11 These experiments are significant because restoration of endothelial function and NO and abrogation of EDCF, ROS, and ADMA would reverse some of the earliest manifestations of hypertension and vascular disease.12
Male Sprague-Dawley rats (200 to 220 g; Taconics Laboratory) were maintained on tap water and standard chow (Na+ 0.4 g · 100 g−1) under constant humidity and temperature and 12-hour light-dark cycles. The protocols were approved by Georgetown University Institutional Animal Care and Use Committee. Details of methods appear in the online Data Supplement (please see http://hyper.ahajournals.org).
Three sets each of 4 groups of Sprague-Dawley rats were prepared (6 per group were used). Rats were anesthetized with sodium pentobarbital (50 mg · kg−1) for subcutaneous insertion of a sham minipump (sham) or an osmotic minipump (model 202, Alzato) at the dorsum of the neck. Group 1 had a sham infusion. Group 2 received Ang II (Peninsula Laboratory) at 200 ng · kg−1 · min−1 for 14 days. Group 3 received a sham minipump and oral Tempol (2 mmol · L−1 of water). Group 4 received an Ang II minipump and Tempol. On the experimental day, rats were anesthetized with inhaled isoflurane (3% balance with oxygen). The carotid artery was cannulated for measurement of mean arterial pressure (model DPM-1B, Biotek). Blood was withdrawn, and the plasma was stored at −80°C. Thereafter, rats were euthanized by exsanguination, the abdomen was opened, and mesenteric arteries were isolated and studied in a myograph or snap frozen and stored at −80°C for analysis of ADMA or protein.
EDRF/NO, EDHF, and EDCF Responses of Mesenteric Resistance Arterioles
The mesenteric vessels were isolated, mounted in a Mulvany-Halpern myograft (JP Trading, Science Park), studied as described in detail previously,13 described in the online Data Supplement, and summarized in Figure S1 (see online Data Supplement).
Fluorescence Detection of Ach-Induced NO and ROS in Mesenteric Resistance Arterioles
These followed methods13 described in details in the online Data Supplement.
Protein Expression Studies
The protein expression in lysates of isolated mesenteric resistance arterioles was studied as described previously.13
Plasma malondialdehyde was measured by high-performance capillary electrophoresis-micellar electrokinetic chromatography (see the online Data Supplement).
l-Arginine, Asymmetrical, and Symmetrical Dimethylarginine Concentrations in Plasma and Tissue
For details, see the online Data Supplement.
Values are expressed as mean±SEM. The concentration-response curves to Ach were analyzed using nonlinear regression of sigmoidal concentration-response curves (GraphPad Prism), which were used to calculate the EC50. Other values were analyzed by ANOVA followed by a post hoc test for multiple comparisons (GraphPad Prism). A value of P<0.05 was considered statistically significant.
Mean arterial pressure under anesthesia and plasma malondialdehyde were increased by Ang II, but this was prevented in rats receiving Tempol. There were no changes in plasma arginine, ADMA, or symmetrical dimethylarginine (SDMA; Table 1).
Figure S1 depicts how endothelium-dependent responses were assessed (see Figure S1). Responses to Ach are shown in the Figure and EC50 and maximal values in Table 2. Compared with sham-infused rats, those infused with Ang II had a 57% reduction in maximal relaxation with Ach and a 67% reduction in EDRF but a more modest 34% reduction in EDRF. Ang II enhanced EDCF responses by 4-fold. Ang II infusion also reduced the sensitivity of the Ach and EDRF responses but increased the sensitivity of the EDCF responses and did not change the sensitivity of EDHF responses. There were no significant effects of Ang II infusion on these responses in rats administered Tempol (Table 2).
Further data for mesenteric resistance arterioles are shown in Table 3. Ach-induced DAFFM (4-amino-5-methylamino-2′,7′-difluorescein) fluorescence was reduced by 41% in vessels from Ang-infused rats, whereas Ach-induced Tempo-9AC-ROS fluorescence was increased 5-fold (Table 3). Both of these changes were prevented in rats given Tempol. Tissue concentrations of l-arginine and SDMA in mesenteric vessels were unaffected by Ang II infusion, but ADMA was increased significantly by 73%. This was prevented by Tempol, which led to a significant increase in tissue l-arginine in rats infused with Ang II. The expression of DDAH-2 protein was reduced by Ang II infusion but was prevented in rats given Tempol.
We evaluated whether a reduced response to NO, H2O2, or EETs could account for the reduced EDRF or EDHF responses of vessels from Ang-infused rats. The maximal relaxation to the NO donor, sodium nitroprusside, in vessels from sham-infused rats (10−4 m; 97±8%) was not significantly different (94±6%) in vessels from Ang II–infused rats. H2O2 elicited a modest relaxation of 9±3% at 10−5 mol · L−1 and a marked relaxation at 10−4 mol · L−1. However, these responses were similar in vessels from sham- and Ang-infused rats (Figure S2A). EETs did not elicit a significant relaxation even at 10−5 mol · L−1 (Figure S2B). Therefore, differences in responsiveness to NO, H2O2, or EETs could not explain the reduced relaxation responses from Ang-infused rats.
To assess the role of vasodilator prostaglandins in EDRF responses, vessels from sham-infused rats were incubated with 5×10−5 mol · L−1 of indomethacin (to block COX-1 and -2). Relaxation responses to 10−4 m Ach after vehicle (84±8%) were unchanged after inhibition of COX with 5×10−5 mol · L−1 of indomethacin (79±6%) but were abolished by endothelium removal (1±3%). Therefore, indomethacin was not used in subsequent relaxation studies.
The mediation of EDHF was studied in vessels from sham-infused rats with the use of specific inhibitors (Table 4). EDHF response to 10−4 m Ach with vehicle was 29±3%. This was not significantly blunted by catalase to metabolize H2O2 or by 14,15-epoxyeicosa-5-(Z)-enoic acid to block EETs14 in concentrations shown to be effective in isolated vessels2,15 but was abolished by endothelium removal. Thus, the EDHF responses depended on the endothelium but were not mediated significantly by H2O2 or EETs.
The EDCF response of vessels from Ang-infused rats (Table 5) was not reduced significantly by incubation with parecoxib to block COX-2 or with OKY-046 to block thromboxane synthase. However, the response was inhibited similarly by SC-560 to block COX-1 or SQ-29 548 to block TP-Rs and was abolished by endothelium removal. These drug concentrations were fully effective in isolated vessels.4 Thus, EDCF responses depended on the endothelium and COX-1 products that activated TP-Rs.
Concentrations of l-arginine, ADMA, and SDMA in homogenates of aorta, kidney, and liver are shown in Table S1. Tissue levels of l-arginine and SDMA were unaffected by Ang II infusion, but levels of ADMA in all 3 of the tissues were increased. These effects were prevented in rats given Tempol.
The main new findings are that a prolonged slow pressor infusion of Ang II into rats changed endothelial function of mesenteric resistance vessels by diminishing vasodilator and increasing vasoconstrictor pathways. This was accompanied by a reduced endothelial release of microvascular NO and increased ROS and ADMA. Ang infusion downregulated the vascular expression of DDAH-2. These effects of Ang II were prevented by 2 weeks of oral Tempol administration. Plasma ADMA did not predict vascular, renal, or hepatic tissue levels of ADMA.
We confirmed previous findings that oral Tempol prevented the increased blood pressure with a slow pressor infusion of Ang II.11,16 Vascular ROS with Ang II are derived from NADPH oxidase activity17 and a reduction in extracellular superoxide dismutase expression.18 Tempol prevented superoxide anion generation by Ang II in VSMCs19 and inhibited lipid peroxidation and microvascular ROS in response to Ang II in the present study. We selected a slow pressor rate of Ang II infusion in this study because high rates of Ang II infusion caused renal damage that reduced DDAH activity.8
A reduction in EDRF in microvessels of rabbits infused with Ang II was improved by bath addition of Tempol.20 A reduced EDRF has been related to bioinactivation of NO by ROS.11 This is consistent with our findings that Tempol administration prevented the reduction in microvascular EDRF and NO and the increase in ROS in rats infused with Ang II.
The maximal EDHF response was also reduced by prolonged Ang II infusion, but the effect was smaller than on EDRF, and the EC50 response was not changed. Mediation of EDHF varies by vessel and species21 but can include H2O21 and EETs.2 However, the impaired EDHF responses in this study in mesenteric arterioles of rats were independent of H2O2 or EET generation or action. Thus, neither H2O2 nor EETs caused significant vasorelaxation except at very high H2O2 concentrations, and the addition of catalase to metabolize H2O2 or the EET antagonist, 14,15-epoxyeicosa-5-(Z)-enoic acid, to the bath did not alter EDHF responses. The mediator of EDHF was not identified in this study.
EDHF is upregulated in blood vessels of endothelial NOS knockout mice.22 We confirm that Ang II–infused rats have impaired EDRF responses in mesenteric arterioles.23 Kang et al5 reported that Ang II reduced the sensitivity of mesenteric vessels to Ach, as in the present study, but reduced the maximal response only after blockade of NOS. They proposed that NOS was upregulated in Ang II–infused rats to compensate for a defect in K+-channel–mediated relaxation. However, in our study, the maximal EDRF response and the Ach-induced NO activity were reduced in vessels from Ang II–infused rats. The difference may relate to the lower rate of Ang II infusion.
An endothelium-dependent contraction is apparent in many models of hypertension.4,24 EDCF responses entailed an endothelial COX-dependent production of a factor under the influence of ROS6 that activates TP-Rs on VSMCs.4 The contractile responses of vessels from Ang II–infused rats in the present study were prevented by blockade of COX-1, TP-Rs, or endothelium removal, but blockade of TxA2 (thromboxane A2) synthase had no significant effect. The present study did not identify the COX-1 product responsible, but prostaglandin endoperoxides have been implicated. The vascular signaling of Ang II contractile response via TP-Rs in this study may underlie the finding that TP-R knockout mice have an impaired pressor and renal vasoconstrictor response to infused Ang II.25
The kidney26 and the liver27 clear plasma ADMA, and the liver clears l-arginine. Avid uptake by these organs may account for the many-fold higher arginine concentrations in the liver and ADMA concentrations in the kidney, relative to the other organs sampled. DDAH-2 is the principal isoform expressed in the rat mesenteric resistance arterioles13 and DDAH-1 in the renal proximal tubules.28 The downregulation by Ang II of DDAH-2 in cultured endothelial cells29 and in mesenteric resistance arterioles in this study and of DDAH-1 in the kidney cortex in a previous study28 could, therefore, have reduced the metabolism of ADMA in these organs and contributed to the increased tissue levels of ADMA. Indeed, Tempol prevented the Ang-induced reduction in DDAH-2 expression in the mesenteric vessels, which could have contributed to the prevention of an Ang II–induced increase in vascular ADMA in Ang-infused rats receiving Tempol. DDAH activity was not assessed in these studies. However, Tain and Baylis30 reported that directly increasing ROS in the kidney cortex decreased DDAH activity. Because Tempol prevented the Ang II–induced increased plasma malondialdehyde and vascular ROS activity, this relates the decreased tissue DDAH-2 expression and increased tissue ADMA to oxidative stress. However, tissue ADMA is also increased by upregulation of protein arginine methyltransferases and by downregulation of cationic amino acid transporters, which could have contributed to the increased tissue ADMA in this study.7 We reported recently that incubation of VSMCs with Ang II increased NADPH oxidase and reduced DDAH and cationic amino acid transporter activity and increased cellular, but not medium, levels of ADMA.10 The increase in cellular ADMA was prevented by Tempol or blockade of Ang type 1 receptors. Further studies in VSMCs transfected with p22phox to increase NADPH oxidase directly demonstrated reduced DDAH-1 and -2 expression and activity, increased protein arginine methyltransferase 3 expression and activity, decreased cationic amino acid transporter 2A expression and activity, and increased cellular ADMA concentrations. Thus, ROS can direct cellular metabolic pathways to increase cellular ADMA concentrations, but a reduction in cationic amino acid transporter activity may restrain ADMA export and limit accumulation of ADMA in the extracellular fluid or plasma.
The 50% reduction in DDAH-2 expression in mesenteric vessels with Ang II in this study was similar to that produced by in vivo gene silencing,13 which impaired EDRF responses and NO bioactivity in mesenteric arterioles substantially. Moreover, Torondel et al31 reported that overexpression of DDAH-1 or -2 in vascular endothelial cells reduced ADMA concentrations, enhanced in vitro vascular NO production, and conferred resistance to administration of ADMA in DDAH heterozygote mice. The 70% increase in ADMA and 33% reduction in the arginine:ADMA ratio in mesenteric vessels of Ang-infused rats should contribute to the reduced EDRF responses, because this would reduce the substrate:inhibitor ratio for NOS. Ang II increased the mesenteric vascular arginine levels in rats given Tempol, which also could have contributed to improved EDRF responses in Tempol-treated rats, but the reason for this was unclear. Moreover, the increase in superoxide in the blood vessels during Ang II infusion should have enhanced NO bioinactivation.11 The finding that the Ang II-induced functional defects, reduced NO and DDAH-2 expressions, and increased tissue levels of ADMA and ROS were all prevented by Tempol implies that an increase in ROS was an upstream event that impaired endothelial responses by coordinating these pathways for endothelial dysfunction.
Infusions of Ang II have variable effects on plasma ADMA. Thus, prolonged Ang II infusions into mice doubled plasma ADMA in one study,9 whereas in others Ang II did not change plasma ADMA,8,32 as in the present study, except at high rates of Ang II infusion that caused renal damage and reduced renal DDAH expression.8 We find that tissue ADMA can increase during a slow pressor Ang II infusion despite unchanged plasma levels.
Endothelial dysfunction or plasma markers of ROS11 or ADMA33 predict future cardiovascular events or death in high-risk patients. The present study shows that tissue ADMA may not always be reflected by plasma levels but that plasma malondialdehyde was a valid predictor of vascular ROS. A recent review concluded that the mechanisms that regulate the balances between NO and EDCF and the processes that transform the endothelium from a protective organ to a source of vasoconstriction and proaggregatory and promitogenic responses in human hypertension are important but remain to be determined.24 The present findings that Tempol given to Ang II–infused rats prevented the defective endothelial relaxation responses and NO bioactivity and the enhanced ROS, endothelial contractions, and ADMA suggest that vascular ROS is an important component of this transformation of endothelial function. Thus, maneuvers to reduce ROS could have therapeutic potential.11
We thank Sigrid de Jong for expert technical assistance and Emily Wing Kam Chan for preparing and editing the article.
Sources of Funding
The work described in this review was supported by research grants to C.S.W. from the National Institute of Diabetes and Digestive and Kidney Diseases (DK-049870 and DK-036079); from the National Heart, Lung, and Blood Institute (HL-68686); by a training grant to Z.L. (T32-DK-059274); by funds from the George E. Schreiner Chair of Nephrology; and by grants to J.R.F. from the National Institutes of Health (GM31278) and the Robert A. Welch Foundation.
- Received May 26, 2010.
- Revision received June 16, 2010.
- Accepted August 19, 2010.
Figueroa XF, Isakson BE, Duling BR. Vascular gap junctions in hypertension. Hypertension. 2006; 48: 804–811.
Wang D, Chabrashvili T, Wilcox CS. Enhanced contractility of renal afferent arterioles from angiotensin-infused rabbits: roles of oxidative stress, thromboxane-prostanoid receptors and endothelium. Circ Res. 2004; 94: 1436–1442.
Kang KT, Sullivan JC, Sasser JM, Imig JD, Pollock JS. Novel nitric oxide synthase–dependent mechanism of vasorelaxation in small arteries from hypertensive rats. Hypertension. 2007; 49: 893–901.
Sasser JM, Moningka NC, Cunningham MW Jr, Croker B, Baylis C. Asymmetric dimethylarginine in angiotensin II-induced hypertension. Am J Physiol Regul Integr Comp Physiol. 2010; 298: R740–R746.
Hasegawa K, Wakino S, Tatematsu S, Yoshioka K, Homma K, Sugano N, Kimoto M, Hayashi K, Itoh H. Role of asymmetric dimethylarginine in vascular injury in transgenic mice overexpressing dimethylarginie dimethylaminohydrolase 2. Circ Res. 2007; 101: e2–e10.
Luo Z, Teerlink T, Griendling K, Aslam S, Welch WJ, Wilcox CS. Angiotensin II and NADPH oxidase increase ADMA in vascular smooth muscle cells. Hypertension. 2010; 56: 498–504.
Wilcox CS, Pearlman A. Chemistry and antihypertensive effects of tempol and other nitroxides. Pharm Rev. 2008; 60: 418–469.
Wang D, Strandgaard S, Iversen J, Wilcox CS. Asymmetric dimethylarginine, oxidative stress, and vascular nitric oxide synthase in essential hypertension. Am J Physiol Regul Integr Comp Physiol. 2009; 296: R195–R200.
Wang D, Gill P, Chabrashvili T, Onozato ML, Raggio J, Mendonca M, Dennehy K, Li M, Modlinger P, Leiper J, Vallance P, Adler O, Leone A, Tojo A, Welch WJ, Wilcox CS. Isoform-specific regulation by N(G),N(G)-dimethylarginine dimethylaminohydrolase of rat serum asymmetric dimethylarginine and vascular endothelium-derived relaxing factor/NO. Circ Res. 2007; 101: 627–635.
Gauthier KM, Deeter C, Krishna UM, Reddy YK, Bondlela M, Falck JR, Campbell WB. 14,15-Epoxyeicosa-5(Z)-enoic acid: a selective epoxyeicosatrienoic acid antagonist that inhibits endothelium-dependent hyperpolarization and relaxation in coronary arteries. Circ Res. 2002; 90: 1028–1036.
Chen Y, Pearlman A, Luo Z, Wilcox CS. Hydrogen peroxide mediates a transient vasorelaxation with tempol during oxidative stress. Am J Physiol. 2007; 293: H2085–H2092.
Kawada N, Imai E, Karber A, Welch WJ, Wilcox CS. A mouse model of angiotensin II slow pressor response: role of oxidative stress. J Am Soc Nephrol. 2002; 13: 2860–2868.
Laursen JB, Rajagopalan S, Galis Z, Tarpey M, Freeman BA, Harrison DG. Role of superoxide in angiotensin II-induced but not catecholamine-induced hypertension. Circulation. 1997; 95: 588–593.
Welch WJ, Chabrashvili T, Solis G, Chen Y, Gill P, Aslam S, Wang X, Ji H, Sandberg K, Jose P, Wilcox CS. Role of extracellular superoxide dismutase in the mouse angiotensin slow pressor response. Hypertension. 2006; 48: 934–941.
Wang D, Chen Y, Chabrashvili T, Aslam S, Borrego L, Umans J, Wilcox CS. Role of oxidative stress in endothelial dysfunction and enhanced responses to Ang II of afferent arterioles from rabbits infused with Ang II. J Am Soc Nephrol. 2003; 14: 2783–2789.
Feletou M, Vanhoutte PM. Endothelium-derived hyperpolarizing factor: where are we now? Arterioscler Thromb Vasc Biol. 2006; 26: 1215–1225.
Huang A, Sun D, Smith CJ, Connetta JA, Shesely EG, Koller A, Kaley G. In eNOS knockout mice skeletal muscle arteriolar dilation to acetylcholine is mediated by EDHF. Am J Physiol. 2000; 278: H762–H768.
Diep QN, El Mabrouk M, Cohn JS, Endemann D, Amiri F, Virdis A, Neves MF, Schiffrin EL. Structure, endothelial function, cell growth, and inflammation in blood vessels of angiotensin II-infused rats: role of peroxisome proliferator-activated receptor-γ. Circulation. 2002; 105: 2296–2302.
Nijveldt RJ, van Leeuwen PAM, van Guldener C, Stehouwer CDA, Rauwerda JA, Teerlink T. Net renal extraction of asymmetrical (ADMA) and symmetrical (SMDA) dimethylarginine in fasting humans. Nephrol Dial Tranplant. 2002; 17: 1999–2002.
Torondel B, Nandi M, Kelly P, Wojciak-Stothard B, Fleming I, Leiper J. Adenoviral-mediated overexpression of DDAH improves vascular tone regulation. Vasc Med. 2010; 15: 205–213.
Jacobi J, Maas R, Cordasic N, Koch K, Schmieder RE, Boger RH, Hilgers KF. Role of asymmetric dimethylarginine for angiotensin II-induced target organ damage in mice. Am J Physiol Heart Circ Physiol. 2008; 294: H1058–H1066.