Systemic Regulation of Vascular NAD(P)H Oxidase Activity and Nox Isoform Expression in Human Arteries and Veins
http://www.100md.com
《动脉硬化血栓血管生物学》
From the Department of Cardiovascular Medicine (T.J.G., K.M.C.), University of Oxford, and Cardiac Surgery (R.P.), John Radcliffe Hospital, UK; and the Departments of Medicine (T.J.G.), Cardiovascular Surgery (J.S., B.K., P.R.), and Transplantology and Pharmacology (T.J.G., A.J., R.K.), Jagiellonian University School of Medicine, J. Dietl Hospital, Cracow, Poland.
ABSTRACT
Objective— Impaired endothelial function, characterized by nitric oxide scavenging by increased superoxide production, is a hallmark of vascular disease states. However, molecular mechanisms regulating superoxide production in human blood vessels remain poorly defined.
Methods and Results— We compared endothelial function, vascular superoxide production, and the expression of NAD(P)H oxidase subunits in arteries and veins from patients undergoing coronary bypass surgery (n=86). Superoxide release was similar in arteries and veins. Inhibitor studies revealed that the NAD(P)H oxidase system was a quantitatively and proportionately greater source of superoxide in veins, whereas xanthine oxidase also contributed significantly to superoxide production in arteries. Moreover, NAD(P)H oxidase molecular composition differed in veins and arteries; veins expressed more nox2 and p22phox, whereas the relative expression of nox4 was greater in arteries. However, there were strong correlations between p22phox and nox4 expression and between superoxide production, NAD(P)H oxidase activity, and endothelial function in arteries and veins from the same patient.
Conclusions— In individuals with coronary artery disease, changes in vascular superoxide production, endothelial function, and NAD(P)H oxidase activity and expression are related in veins and arteries. These findings highlight the importance of systemic effects on the molecular regulation of the NAD(P)H oxidases in human vascular disease.
Endothelial dysfunction is characterized by increased superoxide production. NAD(P)H oxidase activity and endothelial function are correlated in veins and arteries in coronary artery disease, suggesting regulation by systemic factors. The expression of the NAD(P)H oxidase subunits p22phox and nox4, although different in veins and arteries, are also correlated.
Key Words: endothelium ? oxidant stress ? reactive oxygen species ? nitric oxide ? NAD(P)H oxidase
Introduction
Oxidative stress plays an important role in the pathogenesis of atherosclerosis, hypertension, and other vascular diseases. In particular, overproduction of superoxide anion may be detrimental because of its rapid interaction with nitric oxide (NO), which leads to the loss of NO bioavailability and reduces its anti-atherogenic effects.1 Superoxide also regulates redox-sensitive signaling pathways, acts as a direct vascular smooth muscle cell (VSMC) mitogen, and modulates vessel remodeling and plaque stability.1 Recent studies indicate that patients with endothelial dysfunction in whom arterial superoxide production is increased are at highest risk for vascular morbidity and mortality.2 The sources of vascular superoxide include NAD(P)H oxidases, xanthine oxidase, cyclooxygenases, nitric oxide synthases, or mitochondrial oxidases.1,3,4 In particular, NAD(P)H oxidases have been identified as a major enzyme system involved in the generation of vascular oxidative stress.5 Recent studies have revealed several molecular homologs of the NAD(P)H oxidase large subunit (termed nox-nonphagocytic oxidase). The molecular composition of vascular NAD(P)H oxidases appears to vary in different cell types and at different stages of atherosclerotic plaques.3 However, the molecular regulation of the NAD(P)H oxidases within atherosclerotic plaques may be more relevant to plaque events, such as rupture, rather than reflecting systemic changes related to global disease progression or pathogenesis. Further understanding of the relevance of the NAD(P)H oxidases to human vascular disease pathogenesis requires investigation of the relationships between the molecular regulation of the NAD(P)H oxidases in the vascular wall and established features of disease such as vascular superoxide production and endothelial dysfunction. Importantly, systemic factors, such as oxidized low-density lipoprotein, angiotensin II, pro-inflammatory cytokines, and diabetes, are major contributors to atherosclerotic risk and progression. These systemic factors likely underlie the observed correlations between endothelial function in coronary and peripheral arterial circulations6 and also suggest that functional and molecular markers of vascular disease in arteries may be accompanied by corresponding systemic changes in the venous circulation.
See page 1540
Accordingly, we aimed to characterize the contribution of candidate oxidase systems, in particular the NAD(P)H oxidases, to total superoxide production in paired human arteries and veins from patients with coronary artery disease. Furthermore, we sought to define the quantitative relationships between endothelial function, superoxide production, and the molecular composition of the vascular NAD(P)H oxidases in paired veins and arteries. We find that expression of individual NAD(P)H oxidase subunits are closely correlated in veins and arteries and are, in turn, related to vascular superoxide production, providing evidence for molecular regulation of vascular NAD(P)H oxidases at a systemic level in human atherosclerosis.
Methods
Patients and Blood Vessels
Paired segments of human saphenous vein (HSV) and internal mammary artery (IMA) were obtained from 86 patients undergoing coronary artery bypass graft surgery. Vessels were harvested using a no-touch technique, before surgical distension, and before topical administration of drugs such as papaverine. Patient characteristics are presented in the Table. Collection of tissue specimens was approved by the Local Research Ethics Committees and informed consent was obtained.
Clinical and Demographic Characteristics of Patients
Vascular Superoxide Production
Superoxide production was measured in intact vessel rings and from vascular homgenates using 2 independent assays: by lucigenin-enhanced chemiluminescence (5 μmol/L) and by ferricytochrome c reduction, using previously described and validated methods.4,7 Additionally, superoxide generation was measured in the presence of various oxidase inhibitors using lucigenin at 20 μmol/L. Superoxide production was expressed as relative light units (RLU) per second per mg vessel dry weight.
Isometric Tension Studies
NO-mediated endothelial function was assessed using isometric tension studies in response to acetylcholine, as described previously and expressed as a percentage of the precontracted tension.4 Saphenous vein and internal mammary artery segments were immediately washed and transported to the laboratory in ice-cold Krebs-Henseleit buffer. In vessels that showed endothelial dysfunction (defined as vasorelaxations < median), acetylcholine-mediated vasorelaxations were also measured after pre-incubation with polyethylene glycol (PEG)-SOD (500 IU/mL).
Western Immunoblotting
Portions of vascular homogenate (20 μg protein) were separated by 12% SDS-PAGE and transferred to nitrocellulose membranes. NAD(P)H oxidase subunits were detected using mouse monoclonal antibodies against p67phox or p47phox, (Transduction Laboratories) or by rabbit polyclonal antibodies against p22phox (generously provided by Dr Imajoh-Ohmi, Tokyo, Japan). Bands were visualized by chemiluminescence (Supersignal; Pierce) and quantified using National Institutes of Health Image software.
Quantitative Real-Time Reverse-Transcription Polymerase Chain Reaction
RNA was isolated from snap-frozen segments of saphenous vein and mammary artery using Tri-reagent, repurified using RNA easy kit (Qiagen) with DNAse digestion,3 and quantified using the RiboGreen fluorimetric assay (Molecular Probes). cDNA was synthesized using ImProm Reverse Transcription System (Promega) using random primers. The cDNA synthesized from 20 ng total RNA was used in subsequent quantitative polymerase chain reaction (PCR) using the SYBR Green system (Quiagen) and Rotorgene 3000 fluorescent real-time PCR machine (Corbett Research). Mg2+ concentrations were 1.5 mmol/L for all primers, except nox4 (4 mmol/L). Annealing temperatures were 58°C for all primers except nox4 (68°C). All primers were used as described by Sorescu et al,3 apart from primers for p22phox (forward: CGCTGGCGTCCGGCCTGATCCTCA; reverse: ACGCACAGCCGCCAGTAGGTAGAT). For quantitative reverse-transcription (RT)-PCR, the MLN-51 gene transcript was used to normalize for RT and PCR efficiencies (forward: CAAGAGTGCTGAGGAGTCGG; reverse: TCATTAGCTTCTGATTTCAG),8 although variability between samples was minimal (not shown). Quantification of specific mRNAs was determined relative to standard curves of total RNA isolated from THP1 cells as standards for nox2 and p22phox, or from human microvascular endothelial cells for nox1 or nox4.
Statistical Analysis
Results are expressed as means±SEM with n indicating number of patients. Statistical comparisons between the 2 groups were made using Student t test for independent or dependent samples, Wilcoxon and Mann-Whitney U test or using ANOVA followed by post hoc tests depending on distribution and variance analysis. P<0.05 was considered significant.
Results
Sources of Superoxide Production in Human Veins and Arteries
Quantification of basal superoxide release from intact vessel rings measured ex vivo using low-concentration lucigenin chemiluminescence (5 μmol/L) revealed no statistically significant differences between saphenous veins and mammary artery segments (16.8±1.7 versus 14.3±1.9 RLU/second per mg; n=41). Pre-incubation with SOD reduced chemiluminescence by almost 70% and addition of tiron by 80% in both arteries and veins, confirming specificity for superoxide. Superoxide production was also demonstrated by SOD-inhibitable dihydroethidium fluorescence in vessel sections, revealing superoxide production from endothelium, media, and adventitia (Figure I, available online at http://atvb.ahajournals.org). Systematic investigation of potential sources of superoxide revealed that in HSV and IMA superoxide production was greatly inhibited by flavin oxidase inhibitor diphenyliodonium (50 μmol/L) and by apocynin (500 μmol/L). The inhibitory effect of apocynin was more pronounced in veins than in arteries (80% in HSV versus 60% in IMA; P<0.05). In contrast, oxypurinol had almost no effect on superoxide production in veins but had a moderate inhibitory effect in arteries, suggesting a role for xanthine oxidase in superoxide production. Cyclooxygenase inhibition by indomethacin resulted in modest inhibition of superoxide production in veins and arteries. Inhibitory effects of diphenyliodonium, oxypurinol, and indomethacin were concentration-dependent (Table I, available online at http://atvb.ahajournals.org). Other inhibitors showed no significant effects on superoxide release in either veins or arteries. As previously described, we observed that the nitric oxide synthase inhibitor L-NMMA tended to increase net superoxide release in arteries, suggesting a scavenging effect of arterial nitric oxide production on superoxide release.9 These findings suggest that NAD(P)H oxidase(s) are a major source of superoxide production in veins and arteries. However, in mammary arteries also, xanthine oxidase is an additional important source.
Effects of Superoxide on Nitric Oxide Bioavailability in Human Arteries and Veins
NO synthase inhibition with L-NAME increased basal superoxide release, indicating superoxide NO scavenging. This effect was significant in arteries but not in veins (Figure 1). To address the importance of superoxide production for endothelial dysfunction in human vessels, we measured changes in acetylcholine-dependent vasorelaxations in response to pre-incubation with PEG-SOD. We found that PEG-SOD (500 U/mL) improved vasorelaxations, more markedly in IMA than in HSV (Figure II, available online at http://atvb.ahajournals.org).
Figure 1. Sources of vascular superoxide generation in human saphenous veins and mammary arteries. Superoxide production was determined by lucigenin-enhanced chemiluminescence (20 μmol/L lucigenin) in saphenous vein (HSV) and internal mammary artery (IMA) segments (n=21). Vessels were incubated for 30 minutes before and during superoxide determination with oxidase inhibitors: diphenyleneiodonium (50 μmol/L), apocynin (500 μmol/L), oxypurinol(100 μmol/L), rotenone (100 μmol/L), or L-NMMA (100 μmol/L). Oxygen consumption was measured to ensure that mitochondrial respiration was inhibited by rotenone. Superoxide generation was expressed as RLU/mg dry weight (mean±SEM). *P<0.05 versus basal; **P<0.01 versus basal; ?P<0.01 versus HSV.
NAD(P)H Oxidase Protein and Activity in Human Veins and Arteries
Because NAD(P)H oxidases appear to play a principal role in superoxide production in human arteries and veins, we next sought to investigate the presence of the NAD(P)H oxidase protein subunits and enzyme activity in vessel protein extracts. Western blotting revealed that both membrane-associated (p22phox) and cytoplasmic subunits (p67phox and p47phox) were more abundant in saphenous veins than in mammary arteries (Figure 2). In line with these findings, measurements of NAD(P)H oxidase activity in vascular homogenates using SOD-inhibitable ferricytochrome c reduction assay showed greater activity in HSV (Figure 2). To investigate the contribution of PKC signaling, known to be important in NAD(P)H oxidase activation, we used the PKC inhibitor chelerythrine. PKC inhibition significantly decreased basal and NADPH-stimulated superoxide production (Table II, available online at http://atvb.ahajournals.org) and abolished the difference in NADPH-stimulated oxidase activity between veins and arteries.
Figure 2. NAD(P)H oxidase protein subunit levels and activity in human saphenous veins and mammary arteries. A, Western blots showing NAD(P)H oxidase subunits, p22phox, p67phox, and p47phox in saphenous vein (HSV) and IMA. -Actin band shows equal loading of protein. B, Densitometric analysis of bands in HSV and IMA (mean±SEM; paired n=8), expressed in relation to actin. C, Superoxide production measured from vascular homogenates in response to 100 μmol/L NADPH or NADH measured using SOD-inhibitable ferricytochrome c reduction assay. Bars show mean±SEM (n=10 patients). *P<0.01 versus HSV.
Systemic Relationships Between NAD(P)H Oxidase Activities and Endothelial Function in Human Arteries and Veins
There was marked variability in basal vascular superoxide production and in maximally stimulated NAD(P)H oxidase activity in saphenous veins and in mammary arteries from patients with coronary artery disease. However, in paired arteries and veins from individual subjects basal and stimulated superoxide production were significantly correlated (Figure 3A and 3B). Similarly, NO-mediated endothelial function, measured by vasorelaxations to ACh, revealed a strong correlation between arteries and veins (Figure 3C). These related findings in veins and arteries from individual patients suggest that vascular superoxide production and nitric oxide-mediated endothelial function are regulated by systemic factors in patients with coronary artery disease.
Figure 3. Relationships between human arteries and veins in vascular superoxide production and endothelial function. Basal superoxide (O2–) production (A), maximal NAD(P)H oxidase activity (B), and endothelial function (C) were determined in paired segments of IMA and HSV. Superoxide production and NAD(P)H oxidase activity in intact vascular rings were measured using lucigenin-enhanced chemiluminescence (5 μmol/L) and endothelial function was assessed using maximal acetylcholine-dependent vasorelaxations (ACh max); n=31 patients.
Molecular Composition of Vascular NAD(P)H Oxidase in Human Arteries and Veins
To investigate potential differences in the molecular composition of NAD(P)H oxidases between veins and arteries, expression of NAD(P)H oxidase subunit mRNA was analyzed using quantitative fluorescent RT-PCR (Figure 4). The expression of p22phox and nox2 (gp91phox) mRNA was greater in saphenous veins than in mammary arteries, in keeping with our earlier observations of increased p22phox protein levels in veins. Nox4 mRNA was readily detected in both veins and arteries, with a trend toward higher levels in arteries. Nox1 mRNA was not detected in veins or arteries except at very low levels in vessels from 2 patients. Finally, no expression of macrophage colony-stimulating factor (CSF) receptor (MCSFR) mRNA was detected in either saphenous veins or mammary arteries, suggesting that the presence of NAD(P)H oxidase subunit mRNA in the vascular wall was not significantly influenced by infiltrating leukocytes. These findings suggest that human blood vessels express NAD(P)H oxidases that are based principally on nox2 and nox4. In saphenous veins, a nox2 oxidase appears to predominate, and levels of p22phox expression are also higher, whereas a nox4 NAD(P)H oxidase is relatively more abundant in mammary arteries.
Figure 4. Molecular composition of NAD(P)H oxidases in saphenous veins and mammary arteries. A, RT-PCR of NADPH oxidase subunits of HSV and IMA. Equivalent RNA amounts (20 ng) were used. Inflammatory cell content was assessed by MCSFR mRNA expression. THP-1 RNA (nox2; p22phox; MCSFR) or hMEC RNA (nox1, nox4) were used as positive control. B, Quantitative RT-PCR of NADPH oxidase subunits of HSV and IMA (n=10 pairs). Results were normalized to mRNA levels of the housekeeping gene MLN-51. *P<0.05 versus HSV.
In view of our previous observation of strong relationships between superoxide production and endothelial function in veins and arteries, we next investigated relationships between the levels of NAD(P)H oxidase subunit mRNA expression in veins and arteries from individual patients. We observed a strong correlation between p22phox mRNA levels and nox4 levels in veins and arteries (Figure 5). In contrast, nox2 (gp91phox) mRNA levels showed no correlation between veins and arteries from the same patient.
Figure 5. Relationships between NAD(P)H oxidase subunit mRNA expression in arteries and veins. Quantitative RT-PCR of NADPH oxidase subunit mRNA (p22phox, nox2, and nox4) of HSV and IMA (n=10 patients) was performed as described in Methods; 20 ng of total RNA was used and results were normalized to expression of the housekeeping gene MLN-51. Relationships were analyzed using Pearson correlation analysis.
Discussion
In this study, we find that NAD(P)H oxidases are important sources of superoxide production in human veins and arteries. The NAD(P)H oxidase system is quantitatively and proportionately a greater source of superoxide in veins, whereas xanthine oxidase appears to additionally contribute substantially to superoxide production in arteries. We find that increased vascular NAD(P)H oxidase activity is associated with increased protein levels of p22phox, p47phox, and p67phox, and increased p22phox and nox2 (gp91phox) mRNA expression. The NAD(P)H oxidase is predominantly nox2-based in saphenous veins, whereas a nox4-based oxidase appears proportionately more important in mammary arteries. Despite these differences in the functional and molecular characteristics of superoxide production between veins and arteries, several key findings strongly suggest that oxidative stress and endothelial dysfunction are regulated by similar systemic factors in human atherosclerosis. First, both superoxide production and endothelial dysfunction are significantly correlated in arteries and veins from individual patients. Second, the expression of both p22phox and nox4 mRNA are strikingly correlated in arteries and veins. Finally, protein kinase C signaling has a key role in regulating superoxide production in veins and arteries.
The NAD(P)H oxidases are multicomponent enzymes composed of membrane-associated proteins and cytosolic subunits and expressed in endothelial smooth muscle cells (SMC), and adventitial cells.5 In the phagocytic-type NAD(P)H oxidase, the membrane-associated proteins gp91phox and p22phox compose the flavocytochrome b558 complex, which forms the catalytic subunit of the oxidase. The cytosolic subunits, including p47phox, p67phox, and the G-protein Rac, provide regulatory function. Recently, homologues of the NAD(P)H oxidase gp91phox(nox2), termed nox1 and nox4, have been identified in VSMC and may underlie biologically important differences in NAD(P)H oxidase regulation and activity in the vascular wall in the development of atherosclerosis.3,10,11 Nox homologs may be differentially associated with various vascular disease phenotypes. Changes in nox1 expression directly alter cell proliferation in culture,10 and the treatment of SMC with angiotensin II or PDGF upregulates nox1 while downregulating nox4.12 In the rat carotid model of vascular injury, the expression of nox1, nox2, and p22phox is elevated early after injury, whereas nox4 increases later,11 coinciding with a reduction in the rate of SMC proliferation. Indeed, nox2 appears to play a particularly important role in the proliferative response, as demonstrated by inhibition of neointimal hyperplasia by specific peptide inhibition of nox2-containing NAD(P)H oxidases.13 Similarly, a nox2-containing NAD(P)H oxidase mediates the hypertrophic phenotype of VSMC in response to angiotensin II.14 These findings suggest that although nox1 and nox2 are involved in acute response to injury or to angiotensin II stimulation, nox4 is involved in maintaining the quiescent phenotype.11 The importance of nox1 in human vessels is less clear, because both our study and previous studies in human coronary arteries3 show only very low levels of expression. The previous observation of increased nox2 expression after vascular injury are interesting in the light of our observations that both p22phox and nox2 are more abundant in saphenous veins than in internal mammary arteries. The marked difference in the nox2/nox4 ratio could at least in part account for the difference in susceptibility to smooth muscle intimal hyperplasia leading to adverse vein graft remodeling and accelerated atherosclerosis in vein grafts,15 whereas mammary artery grafts are not susceptible to atherosclerosis.
We estimated the relative contribution of the individual vascular wall segments (endothelium-media-adventitia) to total superoxide production in human vessels. Semi-quantitative analysis of dihydroethidium fluorescence showed that endothelium accounts for approximately one quarter of total superoxide in the vessel wall in human veins and arteries, which is in agreement with our previous study in vascular homogenates.4 The media appeared to be a more important relative contributor in IMA, whereas the adventitia was the dominant source in veins.
Our study adds further insights into the relationships between nox isoform expression and increased oxidative stress in human atherosclerosis. Azumi et al were the first to show the presence of a p22phox-based NAD(P)H oxidase in human coronary artery atherosclerotic plaque.16 Sorescu et al very elegantly demonstrated that nox2 and p22phox are greatly increased with the progression of human atherosclerotic plaques in coronary arteries, in part related to inflammatory cell infiltration, whereas nox4 was increased in early lesions and decreased in very severe lesions.3 A recent study has additionally shown that nox4 expression is increased by oscillatory versus pulsatile flow, which may be particularly relevant to the development of atherosclerotic plaques in regions of turbulent flow. Importantly, nox4 expression in this model coincided with increased oxidative stress and low-density lipoprotein oxidation.17 Azumi et al showed a direct spatial relationship between NAD(P)H oxidase-generated oxidative stress and oxidized low-density lipoprotein in atherectomy specimens of human atherosclerotic plaque that were increased in samples from patients with unstable angina.18 The importance of NAD(P)H oxidase for plaque stability is further emphasized by the finding that the shoulder region of plaques is a particularly intense area of reactive oxygen species production, in association with p22phox and Nox2 expression.3
The differences in NAD(P)H subunit molecular composition, between arteries and veins, could also reflect the differences in the vascular cells that contribute to superoxide production. Although all layers of human artery and vein produce superoxide,9 the relative importance of endothelium, SMC, and adventitial cells may be subject to differential regulation. In veins, the predominance of nox2 expression suggests major contributions from the endothelium and adventitia, because these contain nox2-based oxidases.19 Adventitial superoxide production from a nox2-containing NAD(P)H oxidase directly contributes to endothelial dysfunction by NO scavenging.20 In human arteries, our observation of increased nox4 expression suggests that SMC could play a critical role.3,11 However, recent data show that human microvascular SMC express nox2 in response to angiotensin II stimulation, mediated by a c-Src pathway.21 This clearly illustrates that nox isoform expression in human vascular cells is regulated in a complex manner that can vary with cell type in different vessels and in response to different pathophysiologic stimuli.
Apart from the NAD(P)H oxidases, xanthine oxidoreductase may be an additional source of vascular superoxide.22,23 Our study confirms the earlier findings of Spiekermann et al by showing that xanthine oxidase contributes to superoxide production, but principally in human arteries rather than in veins. Arachidonic acid metabolism may also contribute to vascular superoxide, as indicated by inhibition by indomethacin, although this effect may reflect the importance of arachidonic acid in NAD(P)H oxidase activation.5
Although previous studies have emphasized the differences between endothelial function and oxidative stress in human veins and arteries,24,25 we have focused on potential relationships that suggest systemic regulatory factors. We find that NO bioactivity and oxidative stress, despite wide variations between individual patients, are closely correlated between veins and arteries. Whereas sources of superoxide anion in arteries and veins are not identical, they are clearly subject to systemic factors, such as diabetes, hypercholesterolemia, or angiotensin II.6,21,26,27 We have observed that nox4 expression was most closely correlated between arteries and veins. Although nox2 was expressed in veins and arteries, there was no correlation between the two. Nox1, in turn, was expressed at only low levels in a small number of patients, but when detected it was present in arteries and veins. It is important to point out that mRNA levels do not necessarily correspond directly to NAD(P)H oxidase activity, because of possible differences between mRNA levels and protein levels, and because of complex regulatory interactions between individual oxidase subunits. Nevertheless, the strong correlation of mRNA levels of p22phox and nox4 between human arteries and veins shows that systemic regulation of the NAD(P)H oxidase system in humans is evident at the molecular level.
There are multiple factors that could affect endothelial function and oxidative stress parameters in a systemic fashion. In a previous study, we found that clinical risk factors such hypercholesterolemia and diabetes are most strongly associated with NAD(P)H oxidase activity4 and reduced NO bioavailability (Table III, available online at http://atvb.ahajournals.org). Genetic factors could also affect vascular NAD(P)H oxidase activity at a systemic level.28
Our findings add further weight to recent evidence suggesting that the PKC pathway is an important regulator of NADPH oxidase activity in the vascular wall.7,29 Recently described, orally active, specific PKC-? inhibitors reverse the acute deterioration in endothelial function observed after a hyperglycemic challenge in humans.30 Our present data add further weight to suggest that the beneficial effects of PKC inhibition in human vascular disease may not be restricted to patients with diabetes mellitus.
Acknowledgments
This work was supported by a Wellcome Trust International Research Development Award (to T.J.G. and K.M.C.) and by grants from the British Heart Foundation.
References
Cai H, Harrison DG. Endothelial dysfunction in cardiovascular diseases: the role of oxidant stress. Circ Res. 2000; 87: 840–844.
Heitzer T, Schlinzig T, Krohn K, Meinertz T, Munzel T. Endothelial dysfunction, oxidative stress, and risk of cardiovascular events in patients with coronary artery disease. Circulation. 2001; 104: 2673–2678.
Sorescu D, Weiss D, Lassegue B, Clempus RE, Szocs K, Sorescu GP, Valppu L, Quinn MT, Lambeth JD, Vega JD, Taylor WR, Griendling KK. Superoxide production and expression of nox family proteins in human atherosclerosis. Circulation. 2002; 105: 1429–1435.
Guzik TJ, West NEJ, Black E, McDonald D, Ratnatunga C, Pillai R, Channon KM. Vascular superoxide production by NAD(P)H oxidase: association with endothelial dysfunction and clinical risk factors. Circ Res. 2000; 86: e85–e90.
Griendling KK, Sorescu D, Ushio-Fukai M. NAD(P)H oxidase: role in cardiovascular biology and disease. Circ Res. 2000; 86: 494–501.
Anderson TJ, Uehata A, Gerhard MD, Meredith IT, Knab S, Delagrange D, Lieberman EH, Ganz P, Creager MA, Yeung AC, Selwyn AP. Close relation of endothelial function in the human coronary and peripheral circulations. J Am Coll Cardiol. 1995; 26: 1235–1241.
Guzik TJ, Mussa S, Gastaldi D, Sadowski J, Ratnatunga C, Pillai R, Channon KM. Mechanisms of increased vascular superoxide production in human diabetes mellitus: Role of NAD(P)H oxidase and endothelial nitric oxide synthase. Circulation. 2002; 105: 1656–1662.
Hamalainen HK, Tubman JC, Vikman S, Kyrola T, Ylikoski E, Warrington JA, Lahesmaa R. Identification and validation of endogenous reference genes for expression profiling of T helper cell differentiation by quantitative real-time RT-PCR. Anal Biochem. 2001; 299: 63–70.
Guzik TJ, West N, Pillai R, Taggart D, Channon KM. Nitric oxide modulates superoxide release and peroxynitrite formation in human blood vessels. Hypertension. 2002; 39: 1088–1094.
Suh YA, Arnold RS, Lassegue B, Shi J, Xu X, Sorescu D, Chung AB, Griendling KK, Lambeth JD. Cell transformation by the superoxide-generating oxidase Mox1. Nature. 1999; 401: 79–82.
Szocs K, Lassegue B, Sorescu D, Hilenski LL, Valppu L, Couse TL, Wilcox JN, Quinn MT, Lambeth JD, Griendling KK. Upregulation of Nox-based NAD(P)H oxidases in restenosis after carotid injury. Arterioscler Thromb Vasc Biol. 2002; 22: 21–27.
Lassegue B, Sorescu D, Szocs K, Yin Q, Akers M, Zhang Y, Grant SL, Lambeth JD, Griendling KK. Novel gp91(phox) homologues in vascular smooth muscle cells : nox1 mediates angiotensin II-induced superoxide formation and redox-sensitive signaling pathways. Circ Res. 2001; 88: 888–894.
Jacobson GM, Dourron HM, Liu J, Carretero OA, Reddy DJ, Andrzejewski T, Pagano PJ. Novel NAD (P)H oxidase inhibitor suppresses angioplasty-induced superoxide and neointimal hyperplasia of rat carotid artery. Circ Res. 2003; 92: 637–643.
Cifuentes ME, Pagano PJ. c-Src and smooth muscle NAD(P)H oxidase: assembling a path to hypertrophy. Arterioscler Thromb Vasc Biol. 2003; 23: 919–921.
West NEJ, Guzik TJ, Black E, Channon KM. Enhanced superoxide production in experimental venous bypass graft intimal hyperplasia: role of NAD(P)H oxidase. Atheroscler Thromb Vasc Biol. 2001; 21: 189–194.
Azumi H, Inoue N, Takeshita S, Rikitake Y, Kawashima S, Hayashi Y, Itoh H, Yokoyama M. Expression of NADH/NADPH oxidase p22phox in human coronary arteries. Circulation. 1999; 100: 1494–1498.
Hwang J, Ing MH, Salazar A, Lassegue B, Griendling K, Navab M, Sevanian A, Hsiai TK. Pulsatile Versus Oscillatory Shear Stress Regulates NADPH Oxidase Subunit Expression. Implication for Native LDL Oxidation. Circ Res. 2003; 93: 1225–1232.
Azumi H, Inoue N, Ohashi Y, Terashima M, Mori T, Fujita H, Awano K, Kobayashi K, Maeda K, Hata K, Shinke T, Kobayashi S, Hirata K, Kawashima S, Itabe H, Hayashi Y, Imajoh-Ohmi S, Itoh H, Yokoyama M. Superoxide generation in directional coronary atherectomy specimens of patients with angina pectoris: important role of NAD(P)H oxidase. Arterioscler Thromb Vasc Biol. 2002; 22: 1838–1844.
Bayraktutan U, Blayney L, Shah AM. Molecular characterization and localization of the NAD(P)H oxidase components gp91-phox and p22-phox in endothelial cells. Arterioscler Thromb Vasc Biol. 2000; 20: 1903–1911.
Rey FE, Li XC, Carretero OA, Garvin JL, Pagano PJ. Perivascular superoxide anion contributes to impairment of endothelium-dependent relaxation: role of gp91(phox). Circulation. 2002; 106: 2497–2502.
Touyz RM, Chen X, Tabet F, Yao G, He G, Quinn MT, Pagano PJ, Schiffrin EL. Expression of a functionally active gp91phox-containing neutrophil-type NAD(P)H oxidase in smooth muscle cells from human resistance arteries: regulation by angiotensin II. Circ Res. 2002; 90: 1205–1213.
Spiekermann S, Landmesser U, Dikalov S, Bredt M, Gamez G, Tatge H, Reepschlager N, Hornig B, Drexler H, Harrison DG. Electron Spin Resonance Characterization of Vascular Xanthine and NAD(P)H Oxidase Activity in Patients With Coronary Artery Disease: Relation to Endothelium-Dependent Vasodilation. Circulation. 2003; 107: 1383–1389.
Landmesser U, Spiekermann S, Dikalov S, Tatge H, Wilke R, Kohler C, Harrison DG, Hornig B, Drexler H. Vascular oxidative stress and endothelial dysfunction in patients with chronic heart failure: role of xanthine-oxidase and extracellular superoxide dismutase. Circulation. 2002; 106: 3073–3078.
Huraux C, Makita T, Kurz S, Yamaguchi K, Szlam F, Tarpey MM, Wilcox JN, Harrison DG, Levy JH. Superoxide production, risk factors, and endothelium-dependent relaxations in human internal mammary arteries. Circulation. 1999; 99: 53–59.
Berry C, Hamilton CA, Brosnan MJ, Magill FG, Berg GA, McMurray JJ, Dominiczak AF. Investigation into the sources of superoxide in human blood vessels: angiotensin II increases superoxide production in human internal mammary arteries. Circulation. 2000; 101: 2206–2212.
Bonetti PO, Lerman LO, Lerman A. Endothelial dysfunction: a marker of atherosclerotic risk. Arterioscler Thromb Vasc Biol. 2003; 23: 168–175.
Neunteufl T, Katzenschlager R, Hassan A, Klaar U, Schwarzacher S, Glogar D, Bauer P, Weidinger F. Systemic endothelial dysfunction is related to the extent and severity of coronary artery disease. Atherosclerosis. 1997; 129: 111–118.
Guzik TJ, West NEJ, Black E, McDonald D, Ratnatunga C, Pillai R, Channon KM. Functional effect of the C242T polymorphism in the NAD(P)H oxidase p22phox subunit on vascular superoxide production in atherosclerosis. Circulation. 2000; 102: 1744–1747.
Hink U, Li H, Mollnau H, Oelze M, Matheis E, Hartmann M, Skatchkov M, Thaiss F, Stahl RA, Warnholtz A, Meinertz T, Griendling K, Harrison DG, Forstermann U, Munzel T. Mechanisms underlying endothelial dysfunction in diabetes mellitus. Circ Res. 2001; 88: E14–E22.
Beckman JA, Goldfine AB, Gordon MB, Garrett LA, Creager MA. Inhibition of Protein Kinase C beta prevents impaired endothelium-dependent vasodilation caused by hyperglycemia in humans. Circ Res. 2002; 90: 107–111.(Tomasz J. Guzik; Jerzy Sa)
ABSTRACT
Objective— Impaired endothelial function, characterized by nitric oxide scavenging by increased superoxide production, is a hallmark of vascular disease states. However, molecular mechanisms regulating superoxide production in human blood vessels remain poorly defined.
Methods and Results— We compared endothelial function, vascular superoxide production, and the expression of NAD(P)H oxidase subunits in arteries and veins from patients undergoing coronary bypass surgery (n=86). Superoxide release was similar in arteries and veins. Inhibitor studies revealed that the NAD(P)H oxidase system was a quantitatively and proportionately greater source of superoxide in veins, whereas xanthine oxidase also contributed significantly to superoxide production in arteries. Moreover, NAD(P)H oxidase molecular composition differed in veins and arteries; veins expressed more nox2 and p22phox, whereas the relative expression of nox4 was greater in arteries. However, there were strong correlations between p22phox and nox4 expression and between superoxide production, NAD(P)H oxidase activity, and endothelial function in arteries and veins from the same patient.
Conclusions— In individuals with coronary artery disease, changes in vascular superoxide production, endothelial function, and NAD(P)H oxidase activity and expression are related in veins and arteries. These findings highlight the importance of systemic effects on the molecular regulation of the NAD(P)H oxidases in human vascular disease.
Endothelial dysfunction is characterized by increased superoxide production. NAD(P)H oxidase activity and endothelial function are correlated in veins and arteries in coronary artery disease, suggesting regulation by systemic factors. The expression of the NAD(P)H oxidase subunits p22phox and nox4, although different in veins and arteries, are also correlated.
Key Words: endothelium ? oxidant stress ? reactive oxygen species ? nitric oxide ? NAD(P)H oxidase
Introduction
Oxidative stress plays an important role in the pathogenesis of atherosclerosis, hypertension, and other vascular diseases. In particular, overproduction of superoxide anion may be detrimental because of its rapid interaction with nitric oxide (NO), which leads to the loss of NO bioavailability and reduces its anti-atherogenic effects.1 Superoxide also regulates redox-sensitive signaling pathways, acts as a direct vascular smooth muscle cell (VSMC) mitogen, and modulates vessel remodeling and plaque stability.1 Recent studies indicate that patients with endothelial dysfunction in whom arterial superoxide production is increased are at highest risk for vascular morbidity and mortality.2 The sources of vascular superoxide include NAD(P)H oxidases, xanthine oxidase, cyclooxygenases, nitric oxide synthases, or mitochondrial oxidases.1,3,4 In particular, NAD(P)H oxidases have been identified as a major enzyme system involved in the generation of vascular oxidative stress.5 Recent studies have revealed several molecular homologs of the NAD(P)H oxidase large subunit (termed nox-nonphagocytic oxidase). The molecular composition of vascular NAD(P)H oxidases appears to vary in different cell types and at different stages of atherosclerotic plaques.3 However, the molecular regulation of the NAD(P)H oxidases within atherosclerotic plaques may be more relevant to plaque events, such as rupture, rather than reflecting systemic changes related to global disease progression or pathogenesis. Further understanding of the relevance of the NAD(P)H oxidases to human vascular disease pathogenesis requires investigation of the relationships between the molecular regulation of the NAD(P)H oxidases in the vascular wall and established features of disease such as vascular superoxide production and endothelial dysfunction. Importantly, systemic factors, such as oxidized low-density lipoprotein, angiotensin II, pro-inflammatory cytokines, and diabetes, are major contributors to atherosclerotic risk and progression. These systemic factors likely underlie the observed correlations between endothelial function in coronary and peripheral arterial circulations6 and also suggest that functional and molecular markers of vascular disease in arteries may be accompanied by corresponding systemic changes in the venous circulation.
See page 1540
Accordingly, we aimed to characterize the contribution of candidate oxidase systems, in particular the NAD(P)H oxidases, to total superoxide production in paired human arteries and veins from patients with coronary artery disease. Furthermore, we sought to define the quantitative relationships between endothelial function, superoxide production, and the molecular composition of the vascular NAD(P)H oxidases in paired veins and arteries. We find that expression of individual NAD(P)H oxidase subunits are closely correlated in veins and arteries and are, in turn, related to vascular superoxide production, providing evidence for molecular regulation of vascular NAD(P)H oxidases at a systemic level in human atherosclerosis.
Methods
Patients and Blood Vessels
Paired segments of human saphenous vein (HSV) and internal mammary artery (IMA) were obtained from 86 patients undergoing coronary artery bypass graft surgery. Vessels were harvested using a no-touch technique, before surgical distension, and before topical administration of drugs such as papaverine. Patient characteristics are presented in the Table. Collection of tissue specimens was approved by the Local Research Ethics Committees and informed consent was obtained.
Clinical and Demographic Characteristics of Patients
Vascular Superoxide Production
Superoxide production was measured in intact vessel rings and from vascular homgenates using 2 independent assays: by lucigenin-enhanced chemiluminescence (5 μmol/L) and by ferricytochrome c reduction, using previously described and validated methods.4,7 Additionally, superoxide generation was measured in the presence of various oxidase inhibitors using lucigenin at 20 μmol/L. Superoxide production was expressed as relative light units (RLU) per second per mg vessel dry weight.
Isometric Tension Studies
NO-mediated endothelial function was assessed using isometric tension studies in response to acetylcholine, as described previously and expressed as a percentage of the precontracted tension.4 Saphenous vein and internal mammary artery segments were immediately washed and transported to the laboratory in ice-cold Krebs-Henseleit buffer. In vessels that showed endothelial dysfunction (defined as vasorelaxations < median), acetylcholine-mediated vasorelaxations were also measured after pre-incubation with polyethylene glycol (PEG)-SOD (500 IU/mL).
Western Immunoblotting
Portions of vascular homogenate (20 μg protein) were separated by 12% SDS-PAGE and transferred to nitrocellulose membranes. NAD(P)H oxidase subunits were detected using mouse monoclonal antibodies against p67phox or p47phox, (Transduction Laboratories) or by rabbit polyclonal antibodies against p22phox (generously provided by Dr Imajoh-Ohmi, Tokyo, Japan). Bands were visualized by chemiluminescence (Supersignal; Pierce) and quantified using National Institutes of Health Image software.
Quantitative Real-Time Reverse-Transcription Polymerase Chain Reaction
RNA was isolated from snap-frozen segments of saphenous vein and mammary artery using Tri-reagent, repurified using RNA easy kit (Qiagen) with DNAse digestion,3 and quantified using the RiboGreen fluorimetric assay (Molecular Probes). cDNA was synthesized using ImProm Reverse Transcription System (Promega) using random primers. The cDNA synthesized from 20 ng total RNA was used in subsequent quantitative polymerase chain reaction (PCR) using the SYBR Green system (Quiagen) and Rotorgene 3000 fluorescent real-time PCR machine (Corbett Research). Mg2+ concentrations were 1.5 mmol/L for all primers, except nox4 (4 mmol/L). Annealing temperatures were 58°C for all primers except nox4 (68°C). All primers were used as described by Sorescu et al,3 apart from primers for p22phox (forward: CGCTGGCGTCCGGCCTGATCCTCA; reverse: ACGCACAGCCGCCAGTAGGTAGAT). For quantitative reverse-transcription (RT)-PCR, the MLN-51 gene transcript was used to normalize for RT and PCR efficiencies (forward: CAAGAGTGCTGAGGAGTCGG; reverse: TCATTAGCTTCTGATTTCAG),8 although variability between samples was minimal (not shown). Quantification of specific mRNAs was determined relative to standard curves of total RNA isolated from THP1 cells as standards for nox2 and p22phox, or from human microvascular endothelial cells for nox1 or nox4.
Statistical Analysis
Results are expressed as means±SEM with n indicating number of patients. Statistical comparisons between the 2 groups were made using Student t test for independent or dependent samples, Wilcoxon and Mann-Whitney U test or using ANOVA followed by post hoc tests depending on distribution and variance analysis. P<0.05 was considered significant.
Results
Sources of Superoxide Production in Human Veins and Arteries
Quantification of basal superoxide release from intact vessel rings measured ex vivo using low-concentration lucigenin chemiluminescence (5 μmol/L) revealed no statistically significant differences between saphenous veins and mammary artery segments (16.8±1.7 versus 14.3±1.9 RLU/second per mg; n=41). Pre-incubation with SOD reduced chemiluminescence by almost 70% and addition of tiron by 80% in both arteries and veins, confirming specificity for superoxide. Superoxide production was also demonstrated by SOD-inhibitable dihydroethidium fluorescence in vessel sections, revealing superoxide production from endothelium, media, and adventitia (Figure I, available online at http://atvb.ahajournals.org). Systematic investigation of potential sources of superoxide revealed that in HSV and IMA superoxide production was greatly inhibited by flavin oxidase inhibitor diphenyliodonium (50 μmol/L) and by apocynin (500 μmol/L). The inhibitory effect of apocynin was more pronounced in veins than in arteries (80% in HSV versus 60% in IMA; P<0.05). In contrast, oxypurinol had almost no effect on superoxide production in veins but had a moderate inhibitory effect in arteries, suggesting a role for xanthine oxidase in superoxide production. Cyclooxygenase inhibition by indomethacin resulted in modest inhibition of superoxide production in veins and arteries. Inhibitory effects of diphenyliodonium, oxypurinol, and indomethacin were concentration-dependent (Table I, available online at http://atvb.ahajournals.org). Other inhibitors showed no significant effects on superoxide release in either veins or arteries. As previously described, we observed that the nitric oxide synthase inhibitor L-NMMA tended to increase net superoxide release in arteries, suggesting a scavenging effect of arterial nitric oxide production on superoxide release.9 These findings suggest that NAD(P)H oxidase(s) are a major source of superoxide production in veins and arteries. However, in mammary arteries also, xanthine oxidase is an additional important source.
Effects of Superoxide on Nitric Oxide Bioavailability in Human Arteries and Veins
NO synthase inhibition with L-NAME increased basal superoxide release, indicating superoxide NO scavenging. This effect was significant in arteries but not in veins (Figure 1). To address the importance of superoxide production for endothelial dysfunction in human vessels, we measured changes in acetylcholine-dependent vasorelaxations in response to pre-incubation with PEG-SOD. We found that PEG-SOD (500 U/mL) improved vasorelaxations, more markedly in IMA than in HSV (Figure II, available online at http://atvb.ahajournals.org).
Figure 1. Sources of vascular superoxide generation in human saphenous veins and mammary arteries. Superoxide production was determined by lucigenin-enhanced chemiluminescence (20 μmol/L lucigenin) in saphenous vein (HSV) and internal mammary artery (IMA) segments (n=21). Vessels were incubated for 30 minutes before and during superoxide determination with oxidase inhibitors: diphenyleneiodonium (50 μmol/L), apocynin (500 μmol/L), oxypurinol(100 μmol/L), rotenone (100 μmol/L), or L-NMMA (100 μmol/L). Oxygen consumption was measured to ensure that mitochondrial respiration was inhibited by rotenone. Superoxide generation was expressed as RLU/mg dry weight (mean±SEM). *P<0.05 versus basal; **P<0.01 versus basal; ?P<0.01 versus HSV.
NAD(P)H Oxidase Protein and Activity in Human Veins and Arteries
Because NAD(P)H oxidases appear to play a principal role in superoxide production in human arteries and veins, we next sought to investigate the presence of the NAD(P)H oxidase protein subunits and enzyme activity in vessel protein extracts. Western blotting revealed that both membrane-associated (p22phox) and cytoplasmic subunits (p67phox and p47phox) were more abundant in saphenous veins than in mammary arteries (Figure 2). In line with these findings, measurements of NAD(P)H oxidase activity in vascular homogenates using SOD-inhibitable ferricytochrome c reduction assay showed greater activity in HSV (Figure 2). To investigate the contribution of PKC signaling, known to be important in NAD(P)H oxidase activation, we used the PKC inhibitor chelerythrine. PKC inhibition significantly decreased basal and NADPH-stimulated superoxide production (Table II, available online at http://atvb.ahajournals.org) and abolished the difference in NADPH-stimulated oxidase activity between veins and arteries.
Figure 2. NAD(P)H oxidase protein subunit levels and activity in human saphenous veins and mammary arteries. A, Western blots showing NAD(P)H oxidase subunits, p22phox, p67phox, and p47phox in saphenous vein (HSV) and IMA. -Actin band shows equal loading of protein. B, Densitometric analysis of bands in HSV and IMA (mean±SEM; paired n=8), expressed in relation to actin. C, Superoxide production measured from vascular homogenates in response to 100 μmol/L NADPH or NADH measured using SOD-inhibitable ferricytochrome c reduction assay. Bars show mean±SEM (n=10 patients). *P<0.01 versus HSV.
Systemic Relationships Between NAD(P)H Oxidase Activities and Endothelial Function in Human Arteries and Veins
There was marked variability in basal vascular superoxide production and in maximally stimulated NAD(P)H oxidase activity in saphenous veins and in mammary arteries from patients with coronary artery disease. However, in paired arteries and veins from individual subjects basal and stimulated superoxide production were significantly correlated (Figure 3A and 3B). Similarly, NO-mediated endothelial function, measured by vasorelaxations to ACh, revealed a strong correlation between arteries and veins (Figure 3C). These related findings in veins and arteries from individual patients suggest that vascular superoxide production and nitric oxide-mediated endothelial function are regulated by systemic factors in patients with coronary artery disease.
Figure 3. Relationships between human arteries and veins in vascular superoxide production and endothelial function. Basal superoxide (O2–) production (A), maximal NAD(P)H oxidase activity (B), and endothelial function (C) were determined in paired segments of IMA and HSV. Superoxide production and NAD(P)H oxidase activity in intact vascular rings were measured using lucigenin-enhanced chemiluminescence (5 μmol/L) and endothelial function was assessed using maximal acetylcholine-dependent vasorelaxations (ACh max); n=31 patients.
Molecular Composition of Vascular NAD(P)H Oxidase in Human Arteries and Veins
To investigate potential differences in the molecular composition of NAD(P)H oxidases between veins and arteries, expression of NAD(P)H oxidase subunit mRNA was analyzed using quantitative fluorescent RT-PCR (Figure 4). The expression of p22phox and nox2 (gp91phox) mRNA was greater in saphenous veins than in mammary arteries, in keeping with our earlier observations of increased p22phox protein levels in veins. Nox4 mRNA was readily detected in both veins and arteries, with a trend toward higher levels in arteries. Nox1 mRNA was not detected in veins or arteries except at very low levels in vessels from 2 patients. Finally, no expression of macrophage colony-stimulating factor (CSF) receptor (MCSFR) mRNA was detected in either saphenous veins or mammary arteries, suggesting that the presence of NAD(P)H oxidase subunit mRNA in the vascular wall was not significantly influenced by infiltrating leukocytes. These findings suggest that human blood vessels express NAD(P)H oxidases that are based principally on nox2 and nox4. In saphenous veins, a nox2 oxidase appears to predominate, and levels of p22phox expression are also higher, whereas a nox4 NAD(P)H oxidase is relatively more abundant in mammary arteries.
Figure 4. Molecular composition of NAD(P)H oxidases in saphenous veins and mammary arteries. A, RT-PCR of NADPH oxidase subunits of HSV and IMA. Equivalent RNA amounts (20 ng) were used. Inflammatory cell content was assessed by MCSFR mRNA expression. THP-1 RNA (nox2; p22phox; MCSFR) or hMEC RNA (nox1, nox4) were used as positive control. B, Quantitative RT-PCR of NADPH oxidase subunits of HSV and IMA (n=10 pairs). Results were normalized to mRNA levels of the housekeeping gene MLN-51. *P<0.05 versus HSV.
In view of our previous observation of strong relationships between superoxide production and endothelial function in veins and arteries, we next investigated relationships between the levels of NAD(P)H oxidase subunit mRNA expression in veins and arteries from individual patients. We observed a strong correlation between p22phox mRNA levels and nox4 levels in veins and arteries (Figure 5). In contrast, nox2 (gp91phox) mRNA levels showed no correlation between veins and arteries from the same patient.
Figure 5. Relationships between NAD(P)H oxidase subunit mRNA expression in arteries and veins. Quantitative RT-PCR of NADPH oxidase subunit mRNA (p22phox, nox2, and nox4) of HSV and IMA (n=10 patients) was performed as described in Methods; 20 ng of total RNA was used and results were normalized to expression of the housekeeping gene MLN-51. Relationships were analyzed using Pearson correlation analysis.
Discussion
In this study, we find that NAD(P)H oxidases are important sources of superoxide production in human veins and arteries. The NAD(P)H oxidase system is quantitatively and proportionately a greater source of superoxide in veins, whereas xanthine oxidase appears to additionally contribute substantially to superoxide production in arteries. We find that increased vascular NAD(P)H oxidase activity is associated with increased protein levels of p22phox, p47phox, and p67phox, and increased p22phox and nox2 (gp91phox) mRNA expression. The NAD(P)H oxidase is predominantly nox2-based in saphenous veins, whereas a nox4-based oxidase appears proportionately more important in mammary arteries. Despite these differences in the functional and molecular characteristics of superoxide production between veins and arteries, several key findings strongly suggest that oxidative stress and endothelial dysfunction are regulated by similar systemic factors in human atherosclerosis. First, both superoxide production and endothelial dysfunction are significantly correlated in arteries and veins from individual patients. Second, the expression of both p22phox and nox4 mRNA are strikingly correlated in arteries and veins. Finally, protein kinase C signaling has a key role in regulating superoxide production in veins and arteries.
The NAD(P)H oxidases are multicomponent enzymes composed of membrane-associated proteins and cytosolic subunits and expressed in endothelial smooth muscle cells (SMC), and adventitial cells.5 In the phagocytic-type NAD(P)H oxidase, the membrane-associated proteins gp91phox and p22phox compose the flavocytochrome b558 complex, which forms the catalytic subunit of the oxidase. The cytosolic subunits, including p47phox, p67phox, and the G-protein Rac, provide regulatory function. Recently, homologues of the NAD(P)H oxidase gp91phox(nox2), termed nox1 and nox4, have been identified in VSMC and may underlie biologically important differences in NAD(P)H oxidase regulation and activity in the vascular wall in the development of atherosclerosis.3,10,11 Nox homologs may be differentially associated with various vascular disease phenotypes. Changes in nox1 expression directly alter cell proliferation in culture,10 and the treatment of SMC with angiotensin II or PDGF upregulates nox1 while downregulating nox4.12 In the rat carotid model of vascular injury, the expression of nox1, nox2, and p22phox is elevated early after injury, whereas nox4 increases later,11 coinciding with a reduction in the rate of SMC proliferation. Indeed, nox2 appears to play a particularly important role in the proliferative response, as demonstrated by inhibition of neointimal hyperplasia by specific peptide inhibition of nox2-containing NAD(P)H oxidases.13 Similarly, a nox2-containing NAD(P)H oxidase mediates the hypertrophic phenotype of VSMC in response to angiotensin II.14 These findings suggest that although nox1 and nox2 are involved in acute response to injury or to angiotensin II stimulation, nox4 is involved in maintaining the quiescent phenotype.11 The importance of nox1 in human vessels is less clear, because both our study and previous studies in human coronary arteries3 show only very low levels of expression. The previous observation of increased nox2 expression after vascular injury are interesting in the light of our observations that both p22phox and nox2 are more abundant in saphenous veins than in internal mammary arteries. The marked difference in the nox2/nox4 ratio could at least in part account for the difference in susceptibility to smooth muscle intimal hyperplasia leading to adverse vein graft remodeling and accelerated atherosclerosis in vein grafts,15 whereas mammary artery grafts are not susceptible to atherosclerosis.
We estimated the relative contribution of the individual vascular wall segments (endothelium-media-adventitia) to total superoxide production in human vessels. Semi-quantitative analysis of dihydroethidium fluorescence showed that endothelium accounts for approximately one quarter of total superoxide in the vessel wall in human veins and arteries, which is in agreement with our previous study in vascular homogenates.4 The media appeared to be a more important relative contributor in IMA, whereas the adventitia was the dominant source in veins.
Our study adds further insights into the relationships between nox isoform expression and increased oxidative stress in human atherosclerosis. Azumi et al were the first to show the presence of a p22phox-based NAD(P)H oxidase in human coronary artery atherosclerotic plaque.16 Sorescu et al very elegantly demonstrated that nox2 and p22phox are greatly increased with the progression of human atherosclerotic plaques in coronary arteries, in part related to inflammatory cell infiltration, whereas nox4 was increased in early lesions and decreased in very severe lesions.3 A recent study has additionally shown that nox4 expression is increased by oscillatory versus pulsatile flow, which may be particularly relevant to the development of atherosclerotic plaques in regions of turbulent flow. Importantly, nox4 expression in this model coincided with increased oxidative stress and low-density lipoprotein oxidation.17 Azumi et al showed a direct spatial relationship between NAD(P)H oxidase-generated oxidative stress and oxidized low-density lipoprotein in atherectomy specimens of human atherosclerotic plaque that were increased in samples from patients with unstable angina.18 The importance of NAD(P)H oxidase for plaque stability is further emphasized by the finding that the shoulder region of plaques is a particularly intense area of reactive oxygen species production, in association with p22phox and Nox2 expression.3
The differences in NAD(P)H subunit molecular composition, between arteries and veins, could also reflect the differences in the vascular cells that contribute to superoxide production. Although all layers of human artery and vein produce superoxide,9 the relative importance of endothelium, SMC, and adventitial cells may be subject to differential regulation. In veins, the predominance of nox2 expression suggests major contributions from the endothelium and adventitia, because these contain nox2-based oxidases.19 Adventitial superoxide production from a nox2-containing NAD(P)H oxidase directly contributes to endothelial dysfunction by NO scavenging.20 In human arteries, our observation of increased nox4 expression suggests that SMC could play a critical role.3,11 However, recent data show that human microvascular SMC express nox2 in response to angiotensin II stimulation, mediated by a c-Src pathway.21 This clearly illustrates that nox isoform expression in human vascular cells is regulated in a complex manner that can vary with cell type in different vessels and in response to different pathophysiologic stimuli.
Apart from the NAD(P)H oxidases, xanthine oxidoreductase may be an additional source of vascular superoxide.22,23 Our study confirms the earlier findings of Spiekermann et al by showing that xanthine oxidase contributes to superoxide production, but principally in human arteries rather than in veins. Arachidonic acid metabolism may also contribute to vascular superoxide, as indicated by inhibition by indomethacin, although this effect may reflect the importance of arachidonic acid in NAD(P)H oxidase activation.5
Although previous studies have emphasized the differences between endothelial function and oxidative stress in human veins and arteries,24,25 we have focused on potential relationships that suggest systemic regulatory factors. We find that NO bioactivity and oxidative stress, despite wide variations between individual patients, are closely correlated between veins and arteries. Whereas sources of superoxide anion in arteries and veins are not identical, they are clearly subject to systemic factors, such as diabetes, hypercholesterolemia, or angiotensin II.6,21,26,27 We have observed that nox4 expression was most closely correlated between arteries and veins. Although nox2 was expressed in veins and arteries, there was no correlation between the two. Nox1, in turn, was expressed at only low levels in a small number of patients, but when detected it was present in arteries and veins. It is important to point out that mRNA levels do not necessarily correspond directly to NAD(P)H oxidase activity, because of possible differences between mRNA levels and protein levels, and because of complex regulatory interactions between individual oxidase subunits. Nevertheless, the strong correlation of mRNA levels of p22phox and nox4 between human arteries and veins shows that systemic regulation of the NAD(P)H oxidase system in humans is evident at the molecular level.
There are multiple factors that could affect endothelial function and oxidative stress parameters in a systemic fashion. In a previous study, we found that clinical risk factors such hypercholesterolemia and diabetes are most strongly associated with NAD(P)H oxidase activity4 and reduced NO bioavailability (Table III, available online at http://atvb.ahajournals.org). Genetic factors could also affect vascular NAD(P)H oxidase activity at a systemic level.28
Our findings add further weight to recent evidence suggesting that the PKC pathway is an important regulator of NADPH oxidase activity in the vascular wall.7,29 Recently described, orally active, specific PKC-? inhibitors reverse the acute deterioration in endothelial function observed after a hyperglycemic challenge in humans.30 Our present data add further weight to suggest that the beneficial effects of PKC inhibition in human vascular disease may not be restricted to patients with diabetes mellitus.
Acknowledgments
This work was supported by a Wellcome Trust International Research Development Award (to T.J.G. and K.M.C.) and by grants from the British Heart Foundation.
References
Cai H, Harrison DG. Endothelial dysfunction in cardiovascular diseases: the role of oxidant stress. Circ Res. 2000; 87: 840–844.
Heitzer T, Schlinzig T, Krohn K, Meinertz T, Munzel T. Endothelial dysfunction, oxidative stress, and risk of cardiovascular events in patients with coronary artery disease. Circulation. 2001; 104: 2673–2678.
Sorescu D, Weiss D, Lassegue B, Clempus RE, Szocs K, Sorescu GP, Valppu L, Quinn MT, Lambeth JD, Vega JD, Taylor WR, Griendling KK. Superoxide production and expression of nox family proteins in human atherosclerosis. Circulation. 2002; 105: 1429–1435.
Guzik TJ, West NEJ, Black E, McDonald D, Ratnatunga C, Pillai R, Channon KM. Vascular superoxide production by NAD(P)H oxidase: association with endothelial dysfunction and clinical risk factors. Circ Res. 2000; 86: e85–e90.
Griendling KK, Sorescu D, Ushio-Fukai M. NAD(P)H oxidase: role in cardiovascular biology and disease. Circ Res. 2000; 86: 494–501.
Anderson TJ, Uehata A, Gerhard MD, Meredith IT, Knab S, Delagrange D, Lieberman EH, Ganz P, Creager MA, Yeung AC, Selwyn AP. Close relation of endothelial function in the human coronary and peripheral circulations. J Am Coll Cardiol. 1995; 26: 1235–1241.
Guzik TJ, Mussa S, Gastaldi D, Sadowski J, Ratnatunga C, Pillai R, Channon KM. Mechanisms of increased vascular superoxide production in human diabetes mellitus: Role of NAD(P)H oxidase and endothelial nitric oxide synthase. Circulation. 2002; 105: 1656–1662.
Hamalainen HK, Tubman JC, Vikman S, Kyrola T, Ylikoski E, Warrington JA, Lahesmaa R. Identification and validation of endogenous reference genes for expression profiling of T helper cell differentiation by quantitative real-time RT-PCR. Anal Biochem. 2001; 299: 63–70.
Guzik TJ, West N, Pillai R, Taggart D, Channon KM. Nitric oxide modulates superoxide release and peroxynitrite formation in human blood vessels. Hypertension. 2002; 39: 1088–1094.
Suh YA, Arnold RS, Lassegue B, Shi J, Xu X, Sorescu D, Chung AB, Griendling KK, Lambeth JD. Cell transformation by the superoxide-generating oxidase Mox1. Nature. 1999; 401: 79–82.
Szocs K, Lassegue B, Sorescu D, Hilenski LL, Valppu L, Couse TL, Wilcox JN, Quinn MT, Lambeth JD, Griendling KK. Upregulation of Nox-based NAD(P)H oxidases in restenosis after carotid injury. Arterioscler Thromb Vasc Biol. 2002; 22: 21–27.
Lassegue B, Sorescu D, Szocs K, Yin Q, Akers M, Zhang Y, Grant SL, Lambeth JD, Griendling KK. Novel gp91(phox) homologues in vascular smooth muscle cells : nox1 mediates angiotensin II-induced superoxide formation and redox-sensitive signaling pathways. Circ Res. 2001; 88: 888–894.
Jacobson GM, Dourron HM, Liu J, Carretero OA, Reddy DJ, Andrzejewski T, Pagano PJ. Novel NAD (P)H oxidase inhibitor suppresses angioplasty-induced superoxide and neointimal hyperplasia of rat carotid artery. Circ Res. 2003; 92: 637–643.
Cifuentes ME, Pagano PJ. c-Src and smooth muscle NAD(P)H oxidase: assembling a path to hypertrophy. Arterioscler Thromb Vasc Biol. 2003; 23: 919–921.
West NEJ, Guzik TJ, Black E, Channon KM. Enhanced superoxide production in experimental venous bypass graft intimal hyperplasia: role of NAD(P)H oxidase. Atheroscler Thromb Vasc Biol. 2001; 21: 189–194.
Azumi H, Inoue N, Takeshita S, Rikitake Y, Kawashima S, Hayashi Y, Itoh H, Yokoyama M. Expression of NADH/NADPH oxidase p22phox in human coronary arteries. Circulation. 1999; 100: 1494–1498.
Hwang J, Ing MH, Salazar A, Lassegue B, Griendling K, Navab M, Sevanian A, Hsiai TK. Pulsatile Versus Oscillatory Shear Stress Regulates NADPH Oxidase Subunit Expression. Implication for Native LDL Oxidation. Circ Res. 2003; 93: 1225–1232.
Azumi H, Inoue N, Ohashi Y, Terashima M, Mori T, Fujita H, Awano K, Kobayashi K, Maeda K, Hata K, Shinke T, Kobayashi S, Hirata K, Kawashima S, Itabe H, Hayashi Y, Imajoh-Ohmi S, Itoh H, Yokoyama M. Superoxide generation in directional coronary atherectomy specimens of patients with angina pectoris: important role of NAD(P)H oxidase. Arterioscler Thromb Vasc Biol. 2002; 22: 1838–1844.
Bayraktutan U, Blayney L, Shah AM. Molecular characterization and localization of the NAD(P)H oxidase components gp91-phox and p22-phox in endothelial cells. Arterioscler Thromb Vasc Biol. 2000; 20: 1903–1911.
Rey FE, Li XC, Carretero OA, Garvin JL, Pagano PJ. Perivascular superoxide anion contributes to impairment of endothelium-dependent relaxation: role of gp91(phox). Circulation. 2002; 106: 2497–2502.
Touyz RM, Chen X, Tabet F, Yao G, He G, Quinn MT, Pagano PJ, Schiffrin EL. Expression of a functionally active gp91phox-containing neutrophil-type NAD(P)H oxidase in smooth muscle cells from human resistance arteries: regulation by angiotensin II. Circ Res. 2002; 90: 1205–1213.
Spiekermann S, Landmesser U, Dikalov S, Bredt M, Gamez G, Tatge H, Reepschlager N, Hornig B, Drexler H, Harrison DG. Electron Spin Resonance Characterization of Vascular Xanthine and NAD(P)H Oxidase Activity in Patients With Coronary Artery Disease: Relation to Endothelium-Dependent Vasodilation. Circulation. 2003; 107: 1383–1389.
Landmesser U, Spiekermann S, Dikalov S, Tatge H, Wilke R, Kohler C, Harrison DG, Hornig B, Drexler H. Vascular oxidative stress and endothelial dysfunction in patients with chronic heart failure: role of xanthine-oxidase and extracellular superoxide dismutase. Circulation. 2002; 106: 3073–3078.
Huraux C, Makita T, Kurz S, Yamaguchi K, Szlam F, Tarpey MM, Wilcox JN, Harrison DG, Levy JH. Superoxide production, risk factors, and endothelium-dependent relaxations in human internal mammary arteries. Circulation. 1999; 99: 53–59.
Berry C, Hamilton CA, Brosnan MJ, Magill FG, Berg GA, McMurray JJ, Dominiczak AF. Investigation into the sources of superoxide in human blood vessels: angiotensin II increases superoxide production in human internal mammary arteries. Circulation. 2000; 101: 2206–2212.
Bonetti PO, Lerman LO, Lerman A. Endothelial dysfunction: a marker of atherosclerotic risk. Arterioscler Thromb Vasc Biol. 2003; 23: 168–175.
Neunteufl T, Katzenschlager R, Hassan A, Klaar U, Schwarzacher S, Glogar D, Bauer P, Weidinger F. Systemic endothelial dysfunction is related to the extent and severity of coronary artery disease. Atherosclerosis. 1997; 129: 111–118.
Guzik TJ, West NEJ, Black E, McDonald D, Ratnatunga C, Pillai R, Channon KM. Functional effect of the C242T polymorphism in the NAD(P)H oxidase p22phox subunit on vascular superoxide production in atherosclerosis. Circulation. 2000; 102: 1744–1747.
Hink U, Li H, Mollnau H, Oelze M, Matheis E, Hartmann M, Skatchkov M, Thaiss F, Stahl RA, Warnholtz A, Meinertz T, Griendling K, Harrison DG, Forstermann U, Munzel T. Mechanisms underlying endothelial dysfunction in diabetes mellitus. Circ Res. 2001; 88: E14–E22.
Beckman JA, Goldfine AB, Gordon MB, Garrett LA, Creager MA. Inhibition of Protein Kinase C beta prevents impaired endothelium-dependent vasodilation caused by hyperglycemia in humans. Circ Res. 2002; 90: 107–111.(Tomasz J. Guzik; Jerzy Sa)