Increased NADH-Oxidase–Mediated Superoxide Production in the Early Stages of Atherosclerosis
Evidence for Involvement of the Renin-Angiotensin System
Background—Angiotensin II activates NAD(P)H-dependent oxidases via AT1-receptor stimulation, the most important vascular source of superoxide (O2·−). The AT1 receptor is upregulated in vitro by low-density lipoprotein. The present study was designed to test whether hypercholesterolemia is associated with increased NAD(P)H-dependent vascular O2·− production and whether AT1-receptor blockade may inhibit this oxidase and in parallel improve endothelial dysfunction.
Methods and Results—Vascular responses were determined by isometric tension studies, and relative rates of vascular O2·− production were determined by use of chemiluminescence with lucigenin, a cypridina luciferin analogue, and electron spin resonance studies. AT1-receptor mRNA was quantified by Northern analysis, and AT1-receptor density was measured by radioligand binding assays. Hypercholesterolemia was associated with impaired endothelium-dependent vasodilation and increased O2·− production in intact vessels. In vessel homogenates, we found a significant activation of NADH-driven O2·− production in both models of hyperlipidemia. Treatment of cholesterol-fed animals with the AT1-receptor antagonist Bay 10-6734 improved endothelial dysfunction, normalized vascular O2·− and NADH-oxidase activity, decreased macrophage infiltration, and reduced early plaque formation. In the setting of hypercholesterolemia, the aortic AT1 receptor mRNA was upregulated to 166±11%, accompanied by a comparable increase in AT1-receptor density.
Conclusions—Hypercholesterolemia is associated with AT1-receptor upregulation, endothelial dysfunction, and increased NADH-dependent vascular O2·− production. The improvement of endothelial dysfunction, inhibition of the oxidase, and reduction of early plaque formation by an AT1-receptor antagonist suggests a crucial role of angiotensin II–mediated O2·− production in the early stage of atherosclerosis.
Clinical and experimental studies have identified a marked attenuation in endothelium-dependent vasodilatation as a characteristic of early stages of atherosclerosis.1 2 In some cases, this is related to enhanced inactivation of endothelium-derived nitric oxide (·NO) by superoxide (O2·−)3 rather than a consequence of decreased NO production.4 Recently, we found that incubation of cultured smooth muscle cells with LDL enhances AT1-receptor expression.5 Activation of the AT1 receptor activates membrane-associated NADH-dependent oxidase,6 7 the predominant source of O2·− in vascular cells.6 8 There is also increased expression of ACE in atherosclerotic lesions,9 which may serve as a source for local production of angiotensin II and ultimately increased stimulation of vascular O2·− production. Indeed, long-term treatment with the ACE inhibitor quinapril markedly improved endothelial vasomotor function in patients with coronary artery disease (Trial on Reversing Endothelial Dysfunction, TREND),10 possibly because of decreased O2·−-mediated inactivation of NO.
On the basis of these considerations, the present study was designed to test whether membrane-bound NADH oxidase is activated in hypercholesterolemia and to examine mechanisms responsible for this phenomenon.
Animals and Protocols
A total of 92 rabbits were studied. The study consisted of 3 protocols.
Ten male New Zealand White rabbits and 10 male Watanabe rabbits (hypercholesterolemia secondary to an LDL-receptor defect; age, 2 to 3 months) were used to measure endothelial function and O2·− in intact vessels and NADH-oxidase activity in vessel homogenates.
Forty New Zealand White rabbits were fed either a normal diet or a diet containing 0.5% cholesterol diet for 8 weeks. Half of these animals received concomitant treatment with a recently developed AT1-receptor antagonist, Bay 10-6734.11 Bay 10-6734 was mixed with the diet to achieve a daily dose of 25 mg · kg−1 · d−1.
Thirty-two New Zealand White rabbits were fed either a control or 0.5% cholesterol diet. Half of these received amlodipine 5 mg · kg−1 · d−1 PO for 8 weeks, mixed with the diet, to test whether a hypotensive agent other than an angiotensin II receptor antagonist affects vascular O2·− production and endothelial function.
On the day of the study, blood was drawn for determination of total cholesterol levels. Thereafter, the animals were given 1000 U heparin IV and sodium pentobarbital to produce death. The chest was rapidly opened and the descending thoracic aorta removed. Aortic rings were suspended in organ chambers as described previously.12 Vasodilator responses were determined after preconstriction with phenylephrine to 30% to 50% of maximal (KCl-induced) tone.
Estimation of Vascular O2·− Production and NADH/NADPH-Oxidase Activity
Vascular O2·− was estimated with lucigenin-enhanced chemiluminescence (lucigenin concentration, 250 μmol/L) as previously described.12 Some vessels from Watanabe rabbits were incubated with either diphenylene iodonium 10 μmol/L, oxypurinol 100 μmol/L, NG-nitro-l-arginine (L-NNA) 10 μmol/L, or rotenone 100 μmol/L to examine the potential role of flavin-containing oxidases, xanthine oxidase, ·NO synthase, and mitochondrial respiration enzymes, respectively. We also examined the effect of the O2·− scavenger Tiron 10 mmol/L on vascular O2·−. To validate data obtained with high lucigenin concentrations, chemiluminescence studies with low concentrations of lucigenin (5 μmol/L)13 and a cypridina luciferin analogue (CLA, 1 μmol/L),14 were performed. We also quantified vascular O2·− with electron spin resonance (ESR) studies using the O2·−-specific spin trap 1-hydroxy-3-carboxy-pyrrolidine.15 Intensities of ESR signals were quantified by measurement of magnitudes of the low-field component of triplet ESR signal of the carboxy-pyrrolidine radical with a dual-probe resonator and an ESR standard probe (Bruker). Settings of the ESR spectrometer (ECS 106, Bruker) were as recently described.14
To measure NADH/NADPH-oxidase activity, thoracic aortas were homogenized as described.12 In some experiments, the endothelium of the vessel was mechanically removed before homogenization.
The macroscopic area detected by positive fat stain and histological parameters of the plaque were determined with a Zeiss/Kontron morphometry unit and KS 400 software. The descending thoracic aorta was cut open and stained for 15 minutes in “Fettrot” solution (Fettrot 7B, No. 1A727, Chroma). The vessels were then washed in 50% ethanol for 10 minutes. The luminal surfaces were photographed, scanned, and digitized. The red-stained area was planimetered and expressed as a percentage of the total luminal surface.
For histochemical analysis, vessels were fixed in 4% neutral buffered formaldehyde and mounted in paraffin blocks. Sections were stained with hematoxylin and eosin and Masson’s trichrome stain. Immunohistochemical staining for macrophages was carried out with the monoclonal antibody RAM 11 (Dako) and the ABC technique with peroxidase/DAB detection by a Vecstatin standard kit (Vector Laboratories). Morphometric analysis of 8 aortic cross sections were performed at three 1-mm cross-sectional steps of each paraffin block and included the percent value of macrophage staining for plaques (% macrophage/stained area to total plaque cross-sectional area).
mRNA Isolation and Northern Analysis
Northern blots were performed as described previously.16 Briefly, aortic segments were homogenized and total RNA was isolated with RNA-clean according to the manufacturer’s protocol. Aliquots (10 μg) were electrophoresed through agarose-formaldehyde gels and transferred onto Hybond N membranes (Amersham). Northern blots were prehybridized for 2 hours at 42°C and then hybridized for 15 hours at 42°C with a random-primed, [32P]dCTP-labeled rat AT1-receptor cDNA probe.
Radioligand Binding Assays
Binding assays on homogenized aortas were performed in 25 mmol/L Tris-HCl (pH 7.4), 5 mmol/L MgCl2, and 100 mmol/L NaCl. Saturation binding assays were conducted with increasing amounts of 125I-labeled angiotensin II or [3H]prazosin (Amersham) as described previously. Total and nonspecific binding points were measured in duplicate. Nonspecific binding was determined in the presence of 10 μmol/L of the AT1-receptor antagonist TCV-116 (Takeda). The samples were incubated for 90 minutes at 22°C, followed by rapid aspiration through Whatman GF/C filters. Samples were counted in a gamma counter.
All chemicals were purchased from Sigma Chemical Co. [32P]dCTP, Hybond N nylon membranes, and 125I-labeled angiotensin II were obtained from Amersham. RNA-Clean was from AGS GmbH.
Results are expressed as mean±SEM. The ED50 value for each experiment was obtained by logit transformation. To compare NADH- and NADPH-driven O2·− production in normal and hypercholesterolemic vessels, 1-way ANOVA was used. Comparisons of vascular responses were performed by multivariate ANOVA. A Scheffé post hoc test was used to examine differences between groups when significance was indicated. Probability values <0.05 were considered significant.
Plasma Lipid Profile
The total plasma cholesterol in Watanabe rabbits averaged 603±45 mg/dL and was therefore significantly higher than in controls (32±3 mg/dL). In cholesterol-fed rabbits (0.5% cholesterol for 8 weeks), the total cholesterol level was significantly increased (1362±92 versus 32±3 mg/dL).
Effects of Hyperlipidemia on Vasodilator Responses to Acetylcholine and Nitroglycerin
Compared with control vessels, maximal relaxations to the endothelium-dependent vasodilator acetylcholine were significantly impaired in vessels from Watanabe rabbits, whereas the sensitivity, as reflected by the ED50, was not altered (Table 1⇓). The sensitivity and maximal relaxations to the endothelium-independent vasodilator nitroglycerin were comparable in both groups (Table 1⇓).
Effects of Hyperlipidemia on Vasoconstriction to Angiotensin II and Phenylephrine
Constrictions to angiotensin II were enhanced in vessels from Watanabe rabbits compared with control vessels. In contrast, vasoconstrictor responses to phenylephrine were similar between control and Watanabe rabbits (Figure 1⇓).
O2·− Production by Aortas From Watanabe and Control Rabbits
Rates of O2·− production, estimated by lucigenin-enhanced chemiluminescence, were increased by ≈2-fold in aortic segments from Watanabe rabbits compared with controls. This increase was abolished by denudation of the endothelium (Figure 2A⇓). Diphenylene iodonium (DPI), the radical scavenger Tiron, and oxypurinol, an inhibitor of xanthine oxidase, markedly inhibited O2·− in tissue from hyperlipidemic animals, whereas L-NNA and rotenone had no effect (Figure 2B⇓). The effect of DPI was significantly greater than that of oxypurinol, suggesting that an enzyme other than xanthine oxidase was contributing to the signal (Figure 2B⇓).
Effect of Hyperlipidemia on Vascular NADH-Dependent Oxidase Activity in Watanabe Rabbits
NADH oxidase activity, assessed by addition of NADH to the vascular homogenates, was significantly increased in Watanabe rabbits compared with controls. In contrast, NADPH-oxidase activity was similar in the 2 groups (Figure 3A⇓). Endothelial removal decreased NADH-oxidase activity from 5.5±0.5 to 3.5±0.3 nmol O2·− · mg−1 · min−1 (P<0.05) in hypercholesterolemic vessels. NADH-dependent activity in both control and hyperlipidemic animals was located predominantly (>90%) in the particulate fraction (Figure 3B⇓).
Effects of AT1-Receptor Blockade on Plasma Lipid Profile, Endothelial Function, and Vascular O2·− in Cholesterol-Fed Animals
AT1-receptor blockade had no effect on cholesterol levels in either control (32±3 versus 28±3 mg/dL) or hyperlipidemic (1362±92 versus 1312±121 mg/dL) animals. In cholesterol-fed rabbits, endothelium-dependent vasodilation was reduced, whereas endothelium-independent vasorelaxation was preserved (Figure 4A⇓ and 4B⇓). Furthermore, a significant activation of NADH-dependent, membrane-associated oxidase (Figure 5A⇓ and 5B⇓) was observed. AT1-receptor blockade (25 mg · kg−1 · d−1 PO) for 8 weeks inhibited sensitivity and potency of angiotensin II with respect to control vessels and hypercholesterolemic animals, compatible with a sufficient blockade of the AT1 receptor (Table 2⇓). In cholesterol-fed animals, AT1-receptor blockade reduced O2·− production in intact rings and inhibited NADH-oxidase activity (Figure 5A⇓ and 5B⇓). This reduction in NADH-oxidase activity was associated with a marked improvement in endothelium-dependent vasodilation in cholesterol-fed animals, whereas responses to nitroglycerin were comparable in all 4 groups (Figure 4A⇓ and 4B⇓, Table 3⇓). Short-term incubation of aortas from hyperlipidemic animals with the AT1-receptor antagonist Bay 10-6734 did not inhibit O2·− production, indicating that the antagonist has no intrinsic antioxidant properties (hyperlipidemic, 2.12±0.20 versus hyperlipidemic+Bay, 1.98±0.18 pmol · mg−1 · min−1).
Effects of AT1-Receptor Blockade on Macrophage Infiltration and Plaque Area in Cholesterol-Fed Animals
The macroscopic fat-stained area was reduced in cholesterol-fed animals treated with the AT1-receptor blocker compared with cholesterol-fed animals without AT1-receptor blockade (5.3±1.4% versus 28.6±7.5%, P<0.05, Figure 6⇓). Histologically, in animals treated with Bay 10-6734, macrophage infiltration averaged 1±0.2% compared with 58.8±15% in plaques from cholesterol-fed animals without receptor blockade.
Effects of Therapy With Amlodipine on Endothelial Function and Vascular O2·− Production in Cholesterol-Fed Animals
In marked contrast to the effect of the angiotensin II receptor blockade, treatment of control rabbits with amlodipine for 8 weeks significantly increased NADH-induced O2·− production (from 4.3±0.3 to 8.2±0.2 nmol · mg−1 · min−1) but did not alter NADH-oxidase activity in aortas from cholesterol-fed animals (7.2±0.2 versus 7.1±0.8 nmol · mg−1 · min−1; P<0.05 for all). This increase in O2·− did not alter responses to nitroglycerin but did decrease responses to acetylcholine (Table 4⇓). Endothelium-dependent vasodilation was not modified by amlodipine in hypercholesterolemic animals.
Lucigenin Validation Experiments
It has recently been shown that high concentrations of lucigenin can produce a redox cycle with flavin-containing enzymes, leading to artifactual increases in estimates of superoxide production.17 18 This has not been found to occur with low concentrations of lucigenin (5 μmol/L).13 Experiments with 5 μmol/L lucigenin revealed a significant increase in vessels from hypercholesterolemic Watanabe rabbits (980±107 counts · mg−1 · min−1, n=5) and cholesterol-fed New Zealand White rabbits (1058±90 counts · mg−1 · min−1, n=5) compared with control (524±39 counts · mg−1 · min−1, n=5). Differences of similar magnitude were observed with the chemiluminescence probe CLA (control, 9529±489 counts · mg−1 · min−1; Watanabe, 27 405±3285 counts · mg−1 · min−1; cholesterol-fed animals, 21 159±2225 counts · mg−1 · min−1; P<0.05). Likewise, ESR studies also indicated an increase in vascular superoxide production in hypercholesterolemic vessels (Figure 7⇓). In summary, with several independent methods, we found a 2- to 4-fold increase in vascular O2·− production in vessels from hypercholesterolemic animals compared with vessels from controls.
Effect of Hyperlipidemia on Vascular AT1-Receptor Density and Expression
Saturation binding assays on vessel membranes from Watanabe and New Zealand White rabbits using 125I-labeled angiotensin II revealed a significant increase in AT1-receptor density without changes in binding affinity (Figure 8⇓). Likewise, AT1-receptor mRNA levels were markedly enhanced in hypercholesterolemic Watanabe rabbit vessels compared with control animals. GAPDH mRNA levels were similarly expressed in both normocholesterolemic and hypercholesterolemic rabbits (Figures 9⇓ and 10⇓).
In this study, we demonstrate that hypercholesterolemia is associated with AT1-receptor upregulation and increased vascular O2·− production secondary to an activation of the vascular NADH oxidase. AT1-receptor blockade normalized the activity of the oxidase, reduced plaque area and macrophage infiltration, and in parallel improved endothelial dysfunction. These findings suggest a pathogenic role for the renin-angiotensin system in early stages of atherosclerosis.
Although numerous epidemiological studies have shown that elevated levels of LDL are associated with the onset of hypertension and atherosclerosis (for review, see Reference 1919 ), the underlying mechanisms remain unclear. ACE inhibition has been shown to promote regression and even prevention of atherosclerosis, suggesting a link between atherosclerosis and the renin-angiotensin system.20
Endothelial Dysfunction and O2·− Production in Hypercholesterolemia
Our present findings are in accord with previous observations3 showing that hypercholesterolemia is associated with endothelial dysfunction and increased vascular O2·− production. In hyperlipidemic Watanabe rabbits, vascular O2·− was increased ≈2-fold compared with controls. Incubation of vessels from Watanabe rabbits with diphenylene iodonium and oxypurinol inhibited O2·− production, whereas L-NNA and rotenone were ineffective, suggesting that the flavoprotein containing xanthine oxidase rather than ·NO synthase or mitochondrial NADH dehydrogenase is a significant source of O2·− in these vessels. This result is consistent with recent observations suggesting that vascular and/or circulating xanthine oxidase is a significant source of O2·− in cholesterol-fed rabbits.3 21 Because the effects of DPI on steady-state O2·− production were significantly greater than that of oxypurinol, an additional flavin-containing oxidase (such as NADH/NADPH oxidase) probably also serves as a source of O2·− in hypercholesterolemic vessels. NADH/NADPH oxidase has recently been demonstrated to be the predominant O2·− source in both endothelial and smooth muscle cells.6 8 Direct measurement of NADH/NADPH-oxidase activity in homogenates of aortas from controls, hyperlipidemic Watanabe rabbits, and cholesterol-fed rabbits confirmed this observation. In both models of hypercholesterolemia, NADH-oxidase activity was increased compared with controls.
Lucigenin Validation Experiments
The majority of data regarding O2·− formation in vessels and vessel homogenates are based on measurements of the intensity of lucigenin-enhanced chemiluminescence.7 12 The validity of these data was recently questioned because lucigenin itself increases formation of O2·− in the presence of enzymes that are capable of providing a 1-electron reduction of lucigenin2+, such as ·NO synthase and xanthine/xanthine oxidase.17 18 We therefore performed experiments using low concentrations of lucigenin (5 μmol/L), which have recently been shown not to produce additional O2·−.13 Measurements were also made using CLA-enhanced chemiluminescence14 and ESR spectroscopy with a spin trap recently shown to detect O2·− with a high sensitivity and specificity.15 All 3 approaches demonstrated a 2- to 4-fold increase in vascular O2·− in vessels from hypercholesterolemic animals compared with vessels from controls. These data therefore confirm the initial observation from our group that in the setting of hypercholesterolemia, vascular O2·− levels are increased3 and validate our lucigenin measurements.
Hypercholesterolemia and Angiotensin II Receptor Expression
To gain insights into how hypercholesterolemia increases oxidase activity, AT1-receptor gene expression was assessed. Recent studies have shown that incubation of vascular smooth muscle cells with LDL increases expression of the angiotensin II subtype AT1.22 Similar results were obtained in cholesterol-fed animals.16 The present investigation illustrates that an upregulation of vascular AT1-receptor expression also occurs in Watanabe rabbit aortas. It is likely that increased smooth muscle levels of the AT1 receptor underlie the increase in constrictor responses to angiotensin II found in these vessels.
AT1-receptor activation has been shown to increase the activity of NADH/NADPH oxidase in smooth muscle6 and endothelial cells.23 There is a growing body of evidence that in hypercholesterolemia, the vascular renin-angiotensin system is activated. Uehara et al24 reported a significant correlation between chymase-like activity (a major non–ACE-dependent angiotensin II generation pathway) in internal mammary artery and serum total cholesterol levels. Moreover, it was demonstrated that O2·− activates vascular ACE.25 Of note, atherosclerotic lesions contain large amounts of angiotensin I–converting enzyme,9 and activated macrophages in atherosclerotic plaques can produce angiotensin II.26 In aggregate, together with our present studies, these data strongly suggest a key role of the renin-angiotensin system in atherosclerosis.
Mechanisms of ACE Inhibitor–Induced Improvement in Endothelial Dysfunction
Endothelial dysfunction in hypercholesterolemic animals has been shown to be improved by ACE inhibitors.20 Some of this benefit was diminished by a bradykinin antagonist. It was therefore hypothesized that inhibition of bradykinin breakdown (and subsequent high concentrations of ·NO and endothelium-derived hyperpolarizing factor) rather than inhibition of angiotensin II formation itself27 was important in this protective effect. Our present studies indicate that improvement in endothelium-dependent vasodilation and vascular O2·− production can be achieved independently of bradykinin preservation.
Recently, Mancini et al10 showed that treatment of patients with coronary artery disease with an ACE inhibitor markedly improved coronary vasomotor function. The TREND study demonstrated that quinapril 40 mg/d given for 6 months markedly improved acetylcholine-provoked vasoconstriction. The authors speculated that one of the basic mechanisms responsible for this improvement may be inhibition of angiotensin II–sensitive, NAD(P)H-dependent O2·−-producing enzymes, resulting in a reduction of ·NO inactivation. Our results support this concept and provide possible molecular mechanisms involved in this phenomenon. Indeed, in the present study, AT1-receptor blockade inhibited NADH-oxidase activity and in parallel improved endothelial dysfunction in cholesterol-fed animals.
These findings cannot be attributed to cholesterol-lowering effects, because treatment with the AT1-receptor blocker had no effect on this parameter. It is also unlikely that the AT1-receptor antagonist had direct radical scavenging effects, because short-term incubation of intact segments with Bay 10-6734 did not reduce the lucigenin signal. The inhibitory effects of Bay 10-6734 on vascular O2·− production were also not secondary to nonspecific vasodilator effects,28 because the use of the vasodilator amlodipine activated, rather than inhibiting, NADH-mediated O2·− production in vascular tissue. We cannot exclude, however, that changes in endothelial function and/or reduction in NADH-oxidase activity is mediated in part by ongoing activation of alternative receptors such as the AT IV or AT2·− subtype, which is particularly manifest in the context of AT1 blockade.29
The reduction in vascular O2·− production caused by AT-receptor blockade was associated with attenuation of plaque formation and macrophage infiltration in cholesterol-fed animals. These findings are in line with recent observations demonstrating that lipid peroxidation and plaque formation are significantly reduced in apolipoprotein E–deficient mice treated with losartan.30
A major mechanism for reduction in atherosclerosis caused by AT1-receptor blockade relates to the observed reduction in vessel macrophages. It is unlikely, however, that macrophages are the source of the increased vascular O2·−, because the increase in oxidase activity in vessels from hypercholesterolemic animals was NADH-dependent, whereas macrophages utilize NADPH as a substrate to produce O2·−.
The present study provides novel information concerning O2·− sources in the early stages of atherosclerosis. It is evident that in hyperlipidemic animals, in addition to vascular and/or circulating xanthine oxidase, NADH oxidase represents a major vascular source of O2·−. Because hypercholesterolemia is associated with increased expression of the AT1 receptor and AT1-receptor blockade improves endothelial dysfunction, it is tempting to speculate that the observed increased activity of NADH oxidase in hypercholesterolemia is at least in part secondary to increased local angiotensin II generation. Together with previous data, our findings suggest that the renin-angiotensin system plays an important role in both the initiation and acceleration of the atherosclerotic process and that inhibition of the renin-angiotensin system may have benefit in treatment of this disease.
This work was supported by the Deutsche Forschungsgemeinschaft (Mu 1079/2-1, Ni 397/2-1) and in part by a vascular biology grant from the William Harvey Institute, London.
- Received September 12, 1998.
- Revision received November 19, 1998.
- Accepted December 17, 1998.
- Copyright © 1999 by American Heart Association
Freiman PC, Mitchell GC, Heistad DD, Armstrong ML, Harrison DG. Atherosclerosis impairs endothelium-dependent vascular relaxation to acetylcholine and thrombin in primates. Circ Res. 1986;58:783–789.
Zeiher AM, Drexler H, Wollschlager H, Just H. Endothelial dysfunction of the coronary microvasculature is associated with coronary blood flow regulation in patients with early atherosclerosis. Circulation. 1991;84:1984–1992.
Ohara Y, Peterson TE, Harrison DG. Hypercholesterolemia increases endothelial superoxide anion production. J Clin Invest. 1993;91:2546–2551.
Minor RLJ, Myers PR, Guerra RJ, Bates JN, Harrison DG. Diet-induced atherosclerosis increases the release of nitrogen oxides from rabbit aorta. J Clin Invest. 1990;86:2109–2116.
Nickenig G, Sachinidis A, Michaelsen F, Bohm M, Seewald S, Vetter H. Upregulation of vascular angiotensin II receptor gene expression by low-density lipoprotein in vascular smooth muscle cells. Circulation. 1997;95:473–478.
Griendling KK, Minieri CA, Ollerenshaw JD, Alexander RW. Angiotensin II stimulates NADH and NADPH oxidase activity in cultured vascular smooth muscle cells. Circ Res. 1994;74:1141–1148.
Rajagopalan S, Kurz S, Munzel T, Tarpey M, Freeman BA, Griendling KK, Harrison DG. Angiotensin II-mediated hypertension in the rat increases vascular superoxide production via membrane NADH/NADPH oxidase activation: contribution to alterations of vasomotor tone. J Clin Invest. 1996;97:1916–1923.
Mohazzab-H KM, Kaminski PM, Wolin MS. NADH oxidoreductase is a major source of superoxide anion in bovine coronary endothelium. Am J Physiol. 1994;266:H2568–H2572.
Diet F, Pratt RE, Berry GJ, Momose N, Gibbons GH, Dzau VJ. Increased accumulation of tissue ACE in human atherosclerotic coronary artery disease. Circulation. 1996;94:2756–2767.
Mancini GB, Henry GC, Macaya C, O’Neill BJ, Pucillo AL, Carere RG, Wargovich TJ, Mudra H, Luscher TF, Klibaner MI, Haber HE, Uprichard AC, Pepine CJ, Pitt B. Angiotensin-converting enzyme inhibition with quinapril improves endothelial vasomotor dysfunction in patients with coronary artery disease: the TREND (Trial on Reversing ENdothelial Dysfunction) Study [see comments]. Circulation. 1996;94:258–265.
Böhm M, Zolk O, Flesch M, Schiffer F, Schnabel P, Stasch JP, Knorr A. Effects of angiotensin II type 1 receptor blockade and angiotensin-converting enzyme inhibition on cardiac β-adrenergic signal transduction. Hypertension. 1998;31:747–754.
Li Y, Zhu H, Kuppusamy P, Roubaud V, Zweier JL, Trush MA. Validation of lucigenin (bis-N-methylacridinium) as a chemilumigenic probe for detecting superoxide anion radical production by enzymatic and cellular systems. J Biol Chem. 1998;273:2015–2023.
Dikalov S, Skatchkov M, Bassenge E. Spin trapping of superoxide radicals and peroxynitrite by 1-hydroxy-3-carboxy-pyrrolidine and 1-hydroxy-2,2,6,6-tetramethyl-4-oxo-piperidine and the stability of corresponding nitroxyl radicals towards biological reductants. Biochem Biophys Res Commun. 1997;231:701–704.
Nickenig G, Jung O, Strehlow K, Zolk O, Linz W, Scholkens BA, Böhm M. Hypercholesterolemia is associated with enhanced angiotensin AT1-receptor expression. Am J Physiol. 1997;272:H2701–H2707.
Becker RH, Wiemer G, Linz W. Preservation of endothelial function by ramipril in rabbits on a long-term atherogenic diet. J Cardiovasc Pharmacol. 1991;18:S110–S115.
White CR, Darley-Usmar V, Berrington WR, McAdams M, Gore JZ, Thompson JA, Parks DA, Tarpey MM, Freeman BA. Circulating plasma xanthine oxidase contributes to vascular dysfunction in hypercholesterolemic rabbits. Proc Natl Acad Sci U S A. 1996;93:8745–8749.
Nickenig G, Sachinidis A, Seewald S, Böhm M, Vetter H. Influence of oxidized low-density lipoprotein on vascular angiotensin II receptor expression. J Hypertens. 1997;15(suppl):S27–S30.
Lang D, Shakesby AC, Mosfer S, Lewis MJ. Angiotensin II upregulates NADH/NADPH oxidase mediated superoxide anion production by guinea pig coronary microvascular endothelial cells. Circulation. 1997;96(suppl I):I-44. Abstract.
Uehara Y, Kinoshita A, Nakayama S, Urata H, Noda K, Sasaguri M, Ideishi T, Kimura M, Arakawa K. Significant positive correlation between chymase-like activity of the human internal mammary artery and serum cholesterol levels. Circulation. 1997;96(suppl I):I-763. Abstract.
Usui M, Egashira K, Tomita H, Katoh M. Superoxide is involved in the activation of vascular angiotensin-converting enzyme in rats induced by chronic inhibition of nitric oxide synthesis. Circulation. 1997;96(suppl I):I-489. Abstract.
Potter DD, Sobey CG, Tompkins PK, Rossen JD, Heistad DD. Evidence that macrophages in atherosclerotic lesions contain angiotensin II. Circulation. 1998;98:800–807.
Farhy RD, Carretero OA, Ho KL, Scicli AG. Role of kinins and nitric oxide in the effects of angiotensin converting enzyme inhibitors on neointima formation. Circ Res. 1993;72:1202–1210.
De Keulenaer GW, Chappell DC, Ishizaka N, Nerem RM, Alexander RW, Griendling KK. Oscillatory and steady laminar shear stress differentially affect human endothelial redox state: role of a superoxide-producing NADH oxidase. Circ Res. 1998;82:1094–1101.
Kerins DM, Hao Q, Vaughan DE. Angiotensin induction of PAI-1 expression in endothelial cells is mediated by the hexapeptide angiotensin IV. J Clin Invest. 1995;96:2515–2520.