A Myocardial Nox2 Containing NAD(P)H Oxidase Contributes to Oxidative Stress in Human Atrial Fibrillation
http://www.100md.com
循环研究杂志 2005年第9期
The University Department of Cardiovascular Medicine (Y.M.K., T.J.G., Y.H.Z., M.H.Z., K.M.C., B.C.) and Department of Cardiothoracic Surgery (H.K., C.R., R.P.), John Radcliffe Hospital, Oxford, OX3 9DU, UK.
Abstract
Human atrial fibrillation (AF) has been associated with increased atrial oxidative stress. In animal models, inhibition of reactive oxygen species prevents atrial remodeling induced by rapid pacing, suggesting that oxidative stress may play an important role in the pathophysiology of AF. NAD(P)H oxidase is a major source of superoxide in the cardiovascular system; however, whether this enzyme contributes to atrial oxidative stress in AF remains to be elucidated. We investigated the sources of superoxide production (using inhibitors and substrates of a range of oxidases, RT-PCR, immunofluorescence, and immunoblotting) in tissue homogenates and isolated atrial myocytes from the right atrial appendage (RAA) of patients undergoing cardiac surgery (n=54 in sinus rhythm [SR] and 15 in AF). A membrane-bound gp91phox containing NAD(P)H oxidase in atrial myocytes was the main source of atrial superoxide production in SR and in AF. NADPH-stimulated superoxide release from RAA homogenates was significantly increased in patients with AF in the absence of changes in mRNA expression of the p22phox and gp91phox subunits of the NAD(P)H oxidase. In contrast with findings in SR patients, NO synthases (NOSs) contributed significantly to atrial superoxide production in fibrillating atria, suggesting that increased oxidative stress in AF may lead to NOS "uncoupling." These findings indicate that a myocardial NAD(P)H oxidase and, to a lesser extent, dysfunctional NOS contribute significantly to superoxide production in the fibrillating human atrial myocardium and may play an important role in the atrial oxidative injury and electrophysiological remodeling observed in patients with AF.
Key Words: atrial fibrillation humans nitric oxide synthase NAD(P)H oxidase superoxide
Introduction
Atrial fibrillation (AF) promotes its own maintenance by causing tachycardia-induced electrophysiological and structural remodeling of the atrial myocardium.1,2 One potential link between AF, tachycardia, and atrial remodeling is oxidative stress. Consistent with this hypothesis, rapid atrial pacing has been shown to increase myocardial peroxynitrite formation and lead to a shortening of the atrial effective refractory period (ERP), both of which are reversed by treatment with the antioxidant and peroxynitrite decomposition catalyst ascorbate.3 More recently, administration of other agents with antioxidant and anti-inflammatory properties such as statins has proved similarly effective in preventing ERP shortening and the vulnerability to AF in a dog model of rapid atrial pacing.4
The mechanisms underlying oxidative injury in the fibrillating atrial myocardium remain to be elucidated; however, an increasing body of evidence indicates that formation of the oxygen-derived free radical superoxide by NAD(P)H oxidases plays a critical role in the development of a wide range of cardiovascular diseases,5eC7 suggesting that this oxidase system may be an important source of oxidative injury in the fibrillating human atrial myocardium.
In neutrophils, NAD(P)H oxidase consists of a core heterodimer comprising the electron transferring plasma membrane subunits p22phox and gp91phox (or nox2) and 4 cytosolic subunits (p40phox, p47phox, p67phox, and the small G-protein rac1/2), which provide regulatory function.5 In vascular smooth muscle and endothelial cells, gp91phox coexists with other homologues termed nox1 and nox4,8eC10 whereas a gp91phox containing NAD(P)H oxidase is the most common isoform in the human ventricular myocardium.7,11
Here, we investigated the source of atrial superoxide production in right atrial appendage (RAA) homogenates and in isolated atrial myocytes from patients in sinus rhythm (SR) undergoing cardiac surgery. We then examined whether the presence of AF affected the magnitude or source of superoxide production in RAA homogenates obtained from patients with permanent or paroxysmal AF.
Materials and Methods
Patient Characteristics
Experiments designed to characterize the source of superoxide production in right atrial tissue (whole tissue homogenates and isolated intact atrial myocytes) were performed in RAAs obtained from 39 patients in SR who underwent first-time coronary artery bypass surgery with or without mitral/aortic valve replacement.
In addition, we evaluated superoxide production and its sources and the expression of NAD(P)H oxidase subunits in RAA homogenates obtained from 15 patients with a history of AF and 15 patients in SR with no previous history of arrhythmias who were matched for treatment, age, and other known risk factors for cardiovascular disease. The AF group included 6 patients who had permanent AF and 9 who had a history of symptomatic paroxysmal AF or persistent AF12 for which they had previous cardioversion. Of these 9 patients, 4 were in AF at the time of admission. We grouped these patients together because of evidence indicating that paroxysmal AF in humans is associated with electrophysiological and structural changes of the atrial myocardium that are similar to those observed in patients with permanent AF,13eC15 suggesting that intermittent bursts of AF, particularly if frequent, are sufficient to instigate atrial remodeling in humans.
A preoperative left ventricular ejection fraction (LVEF; %) was obtained in all patients, whereas left atrial size (by transthoracic echocardiography) was obtained in 24 (14 in AF and 10 in SR) of the 30 patients included in the matched cohort. Demographic and clinical characteristics of the three groups are shown in Table 1. The study was approved by the local research ethics committee, and all patients gave written informed consent.
Atrial Myocyte Isolation Protocol
Samples of RAA were digested (40 minutes) in isolation buffer supplemented with type I collagenase (2 mg/mL; Worthington) and protease (0.2 mg/mL; Sigma), 1% BSA, and 50 eol/L Ca2+, and then transferred into fresh collagenase solution without protease. After isolation, rod-shaped atrial myocytes were enriched (to 90%) by a two-step centrifugation (3 minutes at 600 rpm) in Kraft-Bruhe medium supplemented with L-glutamate, sodium pyruvate, taurine, creatine, succinic acid, Na2ATP, and -OH butyrate and used within 3 hours for superoxide measurements.
Atrial Superoxide Production
Superoxide production in homogenized RAAs or isolated atrial myocytes was measured by lucigenin-enhanced chemiluminescence using a single-tube luminometer (Berthold FB12) modified to maintain the sample temperature at 37°C.16 Briefly, basal chemiluminescence from RAA homogenate or isolated right atrial myocytes was measured in buffer (2 mL) containing lucigenin (5 eol/L) after reaching equilibrium (7 minutes). Further measurements were taken after the addition of inhibitors or substrates of specific oxidase systems16 (Table 2; n=9 to 14 for each comparison). Superoxide scavengers such as polyethylene glycol-superoxide dismutase (PEG-SOD; 650 U/mL) and tiron (10 mmol/L) were also used.
After application of NADPH (100 eol/L), the chemiluminescence signal in RAA homogenate reached a plateau within 3 minutes; however, in atrial myocytes, the signal continued to increase linearly over time. Hence, in this preparation, measurements do not reflect the maximal NADPH-stimulated superoxide production. Atrial homogenates were separated into soluble (cytosolic) and particulate (membrane-associated) fractions by ultracentrifugation, as described previously.17
Because lucigenin (at concentration 20 eol/L) may be subject to redox cycling,18 we compared basal and NADPH-stimulated superoxide measurements in RAA homogenates by 5 eol/L lucigenin with those obtained by using the luminol analogue L-012 (100 eol/L in the presence of ebselen; 50 eol/L)19,20 and the SOD-inhibitable ferricytochrome c reduction,16,21 respectively. In both cases, measurements were closely correlated with those obtained in parallel by lucigenin-enhanced chemiluminescence (basal measurements r=0.88, P<0.005, n=8; NADPH-stimulated r=0.99, P<0.0001, n=11), in agreement with previous reports indicating that 5 eol/L lucigenin is a sensitive and valid probe for assessing superoxide production.22,23
Oxidative Fluorescent Microtopography Using Dihydroethidium
Superoxide production in tissue sections of human RAA was detected using the fluorescent probe DHE.24 Human RAAs were freshly harvested and frozen in OCT compound. Cryosections (30 e) were incubated in Krebs-HEPES buffer for 30 minutes at 37°C with or without PEG-SOD (1000 U/mL) and with dihydroethidium (DHE; 2 eol/L; Molecular Probes) for another 5 minutes at 37°C in darkness. Images were obtained using a laser confocal microscope (Bio-Rad; MRC 1024) at identical acquisition settings using an excitation wavelength of 488 nm. Fluorescence was detected at 585 nm.
Immunoblotting
Protein expression of subunits of NAD(P)H oxidase in RAA homogenates was investigated using rabbit polyclonal antibodies directed against cytosolic p47phox and p67phox subunits (07-001 and 07-002; Upstate) and a rabbit polyclonal antibody directed against the p22phox subunit (a kind gift from Dr Frans B. Wientjes, University College London, United Kingdom) after separation in 10% SDS-PAGE gels and protein transfer to polyvinylidene difluoride membranes. After incubation in primary antibody, antibody binding was visualized using horseradish peroxidase-conjugated anti-rabbit IgG (Promega) chemiluminescence.
Immunolocalization
Freshly isolated human atrial myocytes were concentrated by centrifugation (600 rpm; 3 minutes). Myocytes were plated on laminin (60 e/mL)-coated coverslips and fixed (1.5 hours at room temperature) by applying 50 e蘈 of cooled 4% paraformaldehyde containing 0.2% Triton X-100. Coverslips were washed with PBS three times and then blocked (2 hours at room temperature) with 5% BSA. Cells were incubated with the primary antibody (1 hour at room temperature; rabbit polyclonal anti-p47phox or anti-p67phox Ab, 07-001 and 07-002; Upstate) diluted 1:500 in PBS containing 5% BSA, washed in PBS, then incubated (30 minutes) with 1:50 fluorescence-conjugated goat anti-rabbit IgG (Alexis). Fluorescence images were obtained with a Bio-Rad MRC 1024 scanning confocal microscope.
Conventional and Quantitative Real-Time RT-PCR
Conventional RT-PCR was initially used to evaluate the presence of gp91phox/nox2, its homologues nox1 and nox4, and p22phox in RAA homogenates of patients in SR. Optimal PCR conditions were 100 nmol/L primers for p22phox, gp91phox, nox1, and nox4 (Table 3), 2.5 mmol/L Mg2+ for p22phox, gp91phox, and nox1, and 4.0 mmol/L Mg2+ for nox4. Annealing temperatures were 58°C for all primers except nox4 (68°C) and extension at 72°C. Total RNA extracted from human THP-1 cells (monocytic cell line) served as positive controls for the detection of p22phox and gp91phox, and RNA derived from human EA (endothelial cell line) cells provided positive control for the presence of nox1 and nox4. PCR amplification for p22phox, gp91phox, nox1, and nox4 cDNA generated 341-, 225-, 377-, and 516-bp fragments, respectively, and were resolved on 2% agarose gel.
Quantitative real time RT-PCR (Rotor-Gene 3000; Corbett Research) was used to compare the expression level of the primary transcripts for p22phox and gp91phox/nox2 subunits of NAD(P)H oxidase in patients with AF and their matched controls in SR. Total RNA was extracted from RAA using a double Trizol extraction procedure, with the resulting RNA pellet treated with amplification grade DNase (Invitrogen). Total RNA was then quantified using RiboGreen, adjusted to 25 ng/uL, and then subjected to one-step reverse transcription and PCR amplification (Table 2) using QuantiTect SYBR Green (Qiagen) in the following conditions: 2.5-pmol/L primers for gp91phox and p22phox, 2.5 mmol/L MgCl2, annealing at 60°C, extension at 72°C. Arbitrary copy numbers were analyzed using Rotor-Gene v.5 software (Corbett Research) from standard curves generated from serial dilutions of total RNA extracted from RAAs for human p22phox and gp91phox templates.
Statistical Analysis
Data are expressed as mean±SEM unless specified otherwise. In all cases, n refers to numbers of patients. Nonparametric statistics was used to evaluate some of the responses to inhibitors or substrates of oxidases and the differences in mRNA expression between AF and SR patients. Other comparisons were made by using ANOVA and Fisher’s protected least significant difference post hoc test. A value of P<0.05 was considered statistically significant.
Results
Sources of Superoxide Production in RAA Homogenates and Isolated Myocytes From Patients in SR
Basal superoxide production was determined by lucigenin (5 eol/L)-enhanced chemiluminescence in homogenized RAAs and in intact right atrial myocytes (average 80±8 relative light units [RLU]/mg protein; n=39 patients; and 0.031±0.006 RLU/myocyte; n=16 patients, respectively). Specificity for superoxide was demonstrated by near-abolition of chemiluminescence after coincubation with either PEG-SOD (650 U/mL) or tiron (10 mmol/L; Figure 1A). In addition, basal superoxide production was greatly reduced by the inhibitor of flavin-containing oxidases, diphenyleneiodonium (DPI; 100 eol/L in RAA and 10 eol/L in atrial myocytes; 70%; P<0.005), or by pretreatment with apocynin (80%; P<0.005), a specific inhibitor of NAD(P)H oxidases (Figure 1A). A small reduction (10% to 20%) in basal superoxide production was observed in response to inhibition of cyclooxygenase (indomethacin; P=0.047) in isolated myocytes and of mitochondrial complex I (rotenone; P=0.02) in RAA homogenates. In contrast, inhibition of NO synthase (NOS) with N-nitro-L-arginine methyl ester (L-NAME) tended to increase basal superoxide production (by 25% in RAA homogenates; P=0.35). Xanthine oxidase inhibition with oxypurinol had no effect on basal superoxide production in both preparations.
NADPH-stimulated superoxide release was markedly inhibited by apocynin and DPI in both preparations, whereas oxypurinol, indomethacin, and L-NAME caused very little change (5% to 10% reduction; Figure 1B). Rotenone resulted in a small reduction in NADPH-stimulated superoxide release in both preparations (10% to 20%; P<0.005).
As observed previously in vascular tissue,23 NADPH chemiluminescence was significantly greater (2-fold) than the NADH-evoked signal in RAA homogenates and in atrial myocytes. Further experiments showed that NADPH and NADH were the only substrates that elicited a significant increase in superoxide production in both preparations (Figure 2A).
To further characterize myocardial NAD(P)H oxidase activity, we evaluated NADPH-dependent superoxide production in subcellular fractions of RAA and atrial myocyte homogenates (Figure 2B). Subcellular fractionation by ultracentrifugation into soluble (cytosolic) and particulate (membrane) fractions revealed that >95% of the NADPH-stimulated oxidase activity originated from the particulate fraction in both preparations (P<0.005 versus cytosolic).
Oxidative fluorescent microtopography using the fluorescent probe DHE showed in situ superoxide production in cryosections of RAAs (Figure 3A), which was significantly attenuated after incubation with PEG-SOD (1000 U/mL).
Together, these data indicate that a membrane-associated NAD(P)H oxidase is the main source of superoxide production in RAA homogenates and right atrial myocytes isolated from patients in SR.
Expression of NAD(P)H Oxidase Subunits in Human RAA Homogenates and Isolated Myocytes
To determine whether NAD(P)H oxidase subunits were expressed in the human atrial myocardium, we used a combined approach using antibodies directed against the p22phox, p47phox, and p67phox subunits of the NAD(P)H oxidase in RAA homogenates and isolated atrial myocytes and conventional RT-PCR to detect p22phox and gp91phox mRNA in isolated atrial myocytes.
Figure 3B shows that the p22phox, p47phox, and p67phox subunits are detected in RAA homogenate and atrial myocytes. p22phox, p47phox, and p67phox appeared as single bands, which were aligned to their respective positive control bands (+ human neutrophils). Conventional RT-PCR confirmed the presence of p22phox and gp91phox mRNA in purified human atrial myocytes but did not detect other gp91phox homologues such as nox1 or nox4 (Figure 3C).
p47phox or p67phox immunolocalization in isolated human atrial myocytes showed a diffuse pattern consistent with cytosolic and membrane-bound localization of these NAD(P)H oxidase subunits (Figure 3D). Omission of the primary antibody (negative control) resulted in no staining in 17 of 17 cells from 3 patients. Corresponding phase contrast images of the same cells are shown in the bottom panel of Figure 3D.
Atrial NAD(P)H Oxidase Activity Is Increased in Patients With AF
NADPH-stimulated superoxide production was significantly higher in RAA homogenates derived from AF patients compared with patients in SR (Figure 4A; P=0.02), who were matched for age, LVEF (46±12% in SR and 50±13% in AF; P=0.40), treatment, and known risk factors (Table 1). Basal superoxide production also tended to be higher in the AF group, but this difference did not reach statistical significance (P=0.15; Figure 4A). LVEF was not different between AF and SR patients (Table 1). As expected, atrial size was significantly larger in the AF group (4.05±0.16 cm in SR patients; n=10 versus 5.01±0.17 cm in AF patients; n=14; P=0.0006); however, there were no differences in LVEF (P=0.97), atrial size (P=0.38), or superoxide production (P=0.32) between patients with permanent or paroxysmal AF. Similarly, we did not find a significant correlation between atrial size (range 3.3 to 6 cm) and basal or NADPH-stimulated superoxide production (r2=0.03 for both), suggesting that the latter is unlikely to be a consequence of atrial hypertrophy and remodeling in AF.
As observed in patients in SR, basal superoxide production in RAA homogenates from AF patients was inhibited by apocynin (81%; P=0.005) and rotenone (40%; P=0.02; Figure 4B; P=0.55 and 0.33, respectively, for comparison with the apocynin- and rotenone-induced inhibition in SR patients). Interestingly, in fibrillating RAA, pretreatment with L-NAME (but not with D-NAME; 1 mmol/L; n=3; data not shown) significantly reduced basal superoxide production by 40% (P=0.01 versus control and P=0.002 versus SR), indicating that dysfunctional NOS may become a significant source of basal superoxide production in fibrillating atria. Similar results were obtained by using NG-monomethyl-L-arginine (L-NMMA; 1 mmol/L; n=3; data not shown).
Pretreatment with apocynin significantly decreased NADPH-stimulated superoxide release (90%; P=0.005), and an 20% reduction in NADPH-stimulated superoxide release was also observed in response to L-NAME (P=0.008 versus control and P=0.07 versus SR; Figure 4B). Rotenone, oxypurinol, and antimycin-A (10 eol/L) had no significant effect.
Quantification of p22phox and gp91phox mRNA in RAA from patients with AF and matched controls in SR using real-time RT-PCR showed no difference in the expression of these subunits between groups (n=15 patients per group, arbitrary copy numbers; p22phox 328.5±71.4 in AF versus 280.1±37.5 in SR; gp91phox 139.9±51 in AF versus 138.6±27.1 in SR; P=NS; Figure 5).
Discussion
The main findings of this study are as follows: (1) a membrane-bound gp91phox/nox2 containing NAD(P)H oxidase is the main source of superoxide production in human atrial myocytes; (2) NAD(P)H-dependent superoxide production from RAA homogenates is increased in patients with a history of AF in the absence of changes in the mRNA expression of the p22phox and gp91phox/nox2 subunits; and (3) in atrial tissue from patients with AF, NOS inhibition resulted in a significant reduction in lucigenin-enhanced chemiluminescence, suggesting that NOS may be "uncoupled" and contribute to superoxide production in fibrillating atria.
These findings suggest that a myocardial NAD(P)H oxidase and, to a lesser extent, NOS and mitochondrial oxidases may play an important role in the atrial oxidative injury and electrophysiological remodeling observed in patients with AF.
Characterization of the Human Atrial NAD(P)H Oxidase
Our findings indicate that a membrane-bound gp91phox NAD(P)H oxidase is present in human right atrial myocytes, where it constitutes the main source of superoxide production. This is evidenced by the observation that RAA superoxide production is significantly increased by the addition of NADPH or NADH but not by application of substrates of other oxidase systems. Similarly, basal and NADPH-stimulated superoxide production (the latter resulting almost exclusively from the particulate fraction) were nearly abolished by apocynin and DPI but not by inhibitors of other oxidases. Pretreatment of RAA homogenates with rotenone and antimycin-A revealed a trend toward a reduction in basal superoxide production, suggesting that the contribution of mitochondrial oxidases to basal superoxide production in the human atrial myocardium is small. We also provide evidence of atrial myocyte-specific expression of the NAD(P)H oxidase subunits p22phox, p47phox, and p67phox at the protein level and of p22phox and gp91phox/nox2 (but not nox1 or nox4) at the messenger level.
Together, our findings indicate that human atrial myocytes have the capability of producing superoxide through a membrane-bound gp91phox/nox2 containing NAD(P)H oxidase.
Myocardial Superoxide Production, Oxidative Stress, and AF
Increased atrial superoxide production may have important implications in the ionic remodeling process, which promotes AF maintenance. Enhanced myocardial superoxide production can decrease NO bioavailability by scavenging NO (to form the potent oxidant peroxynitrite) and by "uncoupling" NOS activity,25,26 which in turn may lead to an increase in myocardial oxygen consumption,27 -adrenergic responsiveness,28,29 and thrombogenic risk.30
Carnes et al3 reported an increase in atrial 3-nitrotyrosine content (a marker of peroxynitrite formation in vivo) in a canine model of rapid atrial pacing that was associated with an abbreviation of the atrial ERP. Interestingly, administration of the antioxidant ascorbate attenuated atrial peroxynitrite formation and electrophysiological remodeling in these animals. Similarly, statins have been shown recently to reduce angiotensin II-stimulated (but not basal) NAD(P)H oxidase activity in human RAA31 and to attenuate ERP reduction and AF inducibility in dogs exposed to rapid atrial pacing.4 Together, these findings suggest the presence of a causal relationship between pro-oxidative cellular redox state, atrial ionic remodeling, and AF.
Our findings clearly show that NAD(P)H oxidase is the main source of atrial superoxide production in AF and in SR; however, they also indicate that the contribution of NOS to myocardial superoxide production is significantly increased in the fibrillating atrial myocardium. This is evidenced by a reduction in basal- and NADPH-stimulated superoxide production after pretreatment with L-NAME or L-NMMA in RAA homogenates from AF patients. These data differ significantly from those obtained in patients in SR, in whom L-NMMA tended to increase lucigenin chemiluminescence.
It is now well established that NOSs can release superoxide when deprived of their critical cofactor tetrahydropbiopterin (BH4) or of their substrate L-arginine.32 Under these conditions (referred to as "NOS uncoupling"), electron flow results in a reduction of molecular oxygen at the prosthetic heme site of the enzyme rather than in NO synthesis. In the presence of increased oxidative stress, oxidation of BH4 can uncouple NOS to generate reactive oxygen species.26,33 The notion that this may occur in the fibrillating atrial myocardium is supported by the recent findings of Cai et al,30 who showed that eNOS expression in the left atrial appendage in a porcine model of rapid pacing-induced AF did not differ from control animals despite a documented reduction in NO formation at this site.
We observed that the capacity of NAD(P)H oxidase to generate superoxide was significantly higher in patients with AF compared with matched patients in SR. Enhanced NAD(P)H oxidase activity can either be attributable to increased expression or post-translational modifications of the oxidase subunits. Real-time quantitative RT-PCR revealed no difference in p22phox or gp91phox mRNA between AF and SR groups, suggesting that the increase in NAD(P)H oxidase activity observed in AF may be attributable to the latter mechanism. Indeed, a key feature of cardiovascular NAD(P)H oxidase is its responsiveness to hormones, hemodynamic forces, and local metabolic changes.5 In particular, vascular NAD(P)H oxidase activity is known to be greatly increased by angiotensin II via phosphorylation of p47phox.34 In agreement with these findings, we found that angiotensin II potently stimulates atrial NAD(P)H oxidase activity (Kim et al, unpublished observations, 2004) which, like its vascular counterpart, is markedly inhibited by chelerythrine, indicating that these effects are at least in part protein kinase C dependent. Together, these findings suggest that the activation of the renin-angiotensin system may be an important underlying mechanism for the increased NAD(P)H oxidase activity and ensuing atrial oxidative stress observed in human AF.
Limitations
The findings of our study should be interpreted in the context of their limitations. Because regional differences in endocardial NO production have been reported in a porcine model of AF induced by rapid right atrial pacing,30 caution should be exerted in extrapolating our findings to the whole atrial myocardium. However, in humans, AF has been shown to cause similar changes in size, function,35 and ion channel protein expression14 in both atria, suggesting that remodeling may be a diffuse phenomenon.
To our knowledge, our study provides the first evidence of NOS uncoupling in the human myocardium; however, detailed investigation of the mechanisms responsible for this phenomenon was hampered by limited tissue availability because the incidence of AF in patients undergoing first-time elective coronary revascularization was <10%. Because all our SR controls patients underwent coronary artery bypass surgery, we elected to recruit AF patients from a similar cohort to match the two groups for risk factors (eg, age, atherosclerosis, and diabetes mellitus) and treatment agents (eg, statins, angiotensin-converting enzyme inhibitors, or angiotensin II receptor blockers) that are known to affect myocardial/endothelial superoxide production.9,31
Conclusions
Increasing evidence indicates that NAD(P)H oxidase may play an important role in the myocardial response to stress or injury. Our findings indicate that the primary source of superoxide production in the human atrial myocardium is an NAD(P)H oxidase in patients in SR and in those with AF. However, in the fibrillating myocardium, NOS contributes significantly to basal- and NADPH-stimulated superoxide release, suggesting that increased oxidative stress in this condition may lead to NOS uncoupling with potentially important implications on myocardial function and thrombogenesis.
Acknowledgments
Y.M.K. was in receipt of a Wellcome Trust Cardiovascular Research Initiative Scholarship, a Clarendon Award, and Mary Goodger Award from the University of Oxford. B.C., K.M.C., and Y.H.Z. are funded by the British Heart Foundation and B.C. by the Garfield Weston Trust. We are grateful to Dr J. Khoo for his help with dihydroethidium oxidative fluorescent microtopography.
References
Wijffels MCEF, Kirchhof CJHJ, Dorland R, Allessie MA. Atrial fibrillation begets atrial fibrillation: a study in awake chronically instrumented goats. Circulation. 1995; 92: 1954eC1968.
Nattel S. New ideas about atrial fibrillation 50 years on. Nature. 2002; 415: 219eC226.
Carnes CA, Chung MK, Nakayama T, Nakayama H, Baliga RS, Piao S, Kanderian A, Pavia S, Hamlin RL, McCarthy PM, Bauer JA, Van Wagoner DR. Ascorbate attenuates atrial pacing-induced peroxynitrite formation and electrical remodeling and decreases the incidence of postoperative atrial fibrillation. Circ Res. 2001; 89: 32eeC38e.
Shiroshita-Takeshita A, Schram G, Lavoie J, Nattel S. Effect of simvastatin and antioxidant vitamins on atrial fibrillation promotion by atrial-tachycardia remodeling in dogs. Circulation. 2004; 110: 2313eC2319.
Griendling KK, Sorescu D, Ushio Fukai M. NAD(P)H oxidase: role in cardiovascular biology and disease. Circ Res. 2000; 86: 494eC501.
Li J-M, Gall NP, Grieve DJ, Chen M, Shah AM. Activation of NADPH oxidase during progression of cardiac hypertrophy to failure. Hypertension. 2002; 40: 477eC484.
Heymes C, Bendall JK, Ratajczak P, Cave AC, Samuel J-L, Hasenfuss G, Shah AM. Increased myocardial NADPH oxidase activity in human heart failure. J Am Coll Cardiol. 2003; 41: 2164eC2171.
Lassegue B, Sorescu D, Szocs K, Yin Q, Akers M, Zhang Y, Grant SL, Lambeth JD, Griendling KK. Novel gp91phox homologues in vascular smooth muscle cells: nox1 mediates angiotensin II-induced superoxide formation and redox-sensitive signaling pathways. Circ Res. 2001; 88: 888eC894.
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: 1429eC1435.
Ago T, Kitazono T, Ooboshi H, Iyama T, Han YH, Takada J, Wakisaka M, Ibayashi S, Utsumi H, Iida M. Nox4 as the major catalytic component of an endothelial NAD(P)H oxidase. Circulation. 2004; 109: 227eC233.
Krijnen PAJ, Meischl C, Hack CE, Meijer CJLM, Visser CA, Roos D, Niessen HWM. Increased Nox2 expression in human cardiomyocytes after acute myocardial infarction. J Clin Pathol. 2003; 56: 194eC199.
McNamara RL, Brass LM, Drozda JP Jr, Go AS, Halperin JL, Kerr CR, Levy S, Malenka DJ, Mittal S, Pelosi F Jr, Rosenberg Y, Stryer D, Wyse DG, Radford MJ, Goff DC Jr, Grover FL, Heidenreich PA, Peterson ED, Redberg RF. ACC/AHA key data elements and definitions for measuring the clinical management and outcomes of patients with atrial fibrillation: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Data Standards. Circulation. 2004; 109: 3223eC3243.
Sparks PB, Jayaprakash S, Vohra JK, Kalman JM. Electrical remodeling of the atria associated with paroxysmal and chronic atrial flutter. Circulation. 2000; 102: 1807eC1813.
Brundel BJJM, Van Gelder IC, Henning RH, Tuinenburg AE, Wietses M, Grandjean JG, Wilde AAM, Van Gilst WH, Crijns HJGM. Alterations in potassium channel gene expression in atria of patients with persistent and paroxysmal atrial fibrillation: differential regulation of protein and mRNA levels for K+ channels. J Am Coll Cardiol. 2001; 37: 926eC932.
Brundel BJJM, Ausma J, van Gelder IC, Van Der Want JJL, van Gilst WH, Crijns HJGM, Henning RH. Activation of proteolysis by calpains and structural changes in human paroxysmal and persistent atrial fibrillation. Cardiovasc Res. 2002; 54: 380eC389.
Guzik TJ, West NE, 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: E85eCE90.
Mohazzab H, Kaminski PM, Agarwal R, Wolin MS. Potential role of a membrane-bound NADH oxidoreductase in nitric oxide release and arterial relaxation to nitroprusside. Circ Res. 1999; 84: 220eC228.
Vasquez-Vivar J, Hogg N, Pritchard Jr KA, Martasek P, Kalyanaraman B. Superoxide anion formation from lucigenin: an electron spin resonance spin-trapping study. FEBS Lett. 1997; 403: 127eC130.
Nishinaka Y, Aramaki Y, Yoshida H, Masuya H, Sugawara T, Ichimori YA. New sensitive chemiluminescence probe, L-012, for measuring the production of superoxide anion by cells. Biochem Biophys Res Commun. 1993; 193: 554eC559.
Daiber A, Oelze M, August M, Wendt M, Sydow K, Wieboldt H, Kleschyov AL, Munzel T. Detection of superoxide and peroxynitrite in model systems and mitochondria by the luminol analogue L-012. Free Radic Res. 2004; 38: 259eC269.
Pritchard KA Jr, Groszek L, Smalley DM, Sessa WC, Wu M, Villalon P, Wolin MS, Stemerman MB. Native low-density lipoprotein increases endothelial cell nitric oxide synthase generation of superoxide anion. Circ Res. 1995; 77: 510eC518.
Skatchkov MP, Sperling D, Hink U, Mulsch A, Harrison DG, Sindermann I, Meinertz T, Munzel T. Validation of lucigenin as a chemiluminescent probe to monitor vascular superoxide as well as basal vascular nitric oxide production. Biochem Biophys Res Commun. 1999; 254: 319eC324.
Li J-M, Shah AM. Differential NADPH- versus NADH-dependent superoxide production by phagocyte-type endothelial cell NADPH oxidase. Cardiovasc Res. 2001; 52: 477eC486.
Zhao H, Kalivendi S, Zhang H, Joseph J, Nithipatikom K, Vasquez-Vivar J, Kalyanaraman B. Superoxide reacts with hydroethidine but forms a fluorescent product that is distinctly different from ethidium: potential implications in intracellular fluorescence detection of superoxide. Free Radic Biol Med. 2003; 34: 1359eC1368.
Mollnau H, Wendt M, Szocs K, Lassegue B, Schulz E, Oelze M, Li H, Bodenschatz M, August M, Kleschyov AL, Tsilimingas N, Walter U, Forstermann U, Meinertz T, Griendling K, Munzel T. Effects of angiotensin II infusion on the expression and function of NAD(P)H oxidase and components of nitric oxide/cGMP signaling. Circ Res. 2002; 90: 58eeC65e.
Landmesser U, Dikalov S, Price SR, McCann L, Fukai T, Holland SM, Mitch WE, Harrison DG. Oxidation of tetrahydrobiopterin leads to uncoupling of endothelial cell nitric oxide synthase in hypertension. J Clin Invest. 2003; 111: 1201eC1209.
Adler A, Messina E, Sherman B, Wang Z, Huang H, Linke A, Hintze TH. NAD(P)H oxidase-generated superoxide anion accounts for reduced control of myocardial O2 consumption by NO in old Fischer 344 rats. Am J Physiol Heart Circ Physiol. 2003; 285: H1015eCH1022.
Balligand JL. Regulation of cardiac beta-adrenergic response by nitric oxide. Cardiovasc Res. 1999; 43: 607eC620.
Sears CE, Ashley EA, Casadei B. Nitric oxide control of cardiac function: is neuronal nitric oxide synthase a key component Philos Trans R Soc Lond B. 2004; 359: 1021eC1044.
Cai H, Li Z, Goette A, Mera F, Honeycutt C, Feterik K, Wilcox JN, Dudley SC Jr, Harrison DG, Langberg JJ. Downregulation of endocardial nitric oxide synthase expression and nitric oxide production in atrial fibrillation: potential mechanisms for atrial thrombosis and stroke. Circulation. 2002; 106: 2854eC2858.
Maack C, Kartes T, Kilter H, Schafers H-J, Nickenig G, Bohm M, Laufs U. Oxygen free radical release in human failing myocardium is associated with increased activity of Rac1-GTPase and represents a target for statin treatment. Circulation. 2003; 108: 1567eC1574.
Vasquez-Vivar J, Kalyanaraman B, Martasek P. The role of tetrahydrobiopterin in superoxide generation from eNOS: enzymology and physiological implications. Free Radic Res. 2003; 37: 121eC127.
Alp NJ, Mussa S, Khoo J, Cai S, Guzik T, Jefferson A, Goh N, Rockett KA, Channon KM. Tetrahydrobiopterin-dependent preservation of nitric oxide-mediated endothelial function in diabetes by targeted transgenic GTP-cyclohydrolase I overexpression. J Clin Invest. 2003; 112: 725eC735.
Li J-M, Shah AM. Mechanism of endothelial cell NADPH oxidase activation by angiotensin II. Role of the p47phox subunit. J Biol Chem. 2003; 278: 12094eC12100.
Escudero E, San Mauro M, Laugle C. Bilateral atrial function after chemical cardioversion of atrial fibrillation with amiodarone: an echo-Doppler study. J Am Soc Echocardiogr. 1998; 11: 365eC371.(Young M. Kim, Tomasz J. G)
Abstract
Human atrial fibrillation (AF) has been associated with increased atrial oxidative stress. In animal models, inhibition of reactive oxygen species prevents atrial remodeling induced by rapid pacing, suggesting that oxidative stress may play an important role in the pathophysiology of AF. NAD(P)H oxidase is a major source of superoxide in the cardiovascular system; however, whether this enzyme contributes to atrial oxidative stress in AF remains to be elucidated. We investigated the sources of superoxide production (using inhibitors and substrates of a range of oxidases, RT-PCR, immunofluorescence, and immunoblotting) in tissue homogenates and isolated atrial myocytes from the right atrial appendage (RAA) of patients undergoing cardiac surgery (n=54 in sinus rhythm [SR] and 15 in AF). A membrane-bound gp91phox containing NAD(P)H oxidase in atrial myocytes was the main source of atrial superoxide production in SR and in AF. NADPH-stimulated superoxide release from RAA homogenates was significantly increased in patients with AF in the absence of changes in mRNA expression of the p22phox and gp91phox subunits of the NAD(P)H oxidase. In contrast with findings in SR patients, NO synthases (NOSs) contributed significantly to atrial superoxide production in fibrillating atria, suggesting that increased oxidative stress in AF may lead to NOS "uncoupling." These findings indicate that a myocardial NAD(P)H oxidase and, to a lesser extent, dysfunctional NOS contribute significantly to superoxide production in the fibrillating human atrial myocardium and may play an important role in the atrial oxidative injury and electrophysiological remodeling observed in patients with AF.
Key Words: atrial fibrillation humans nitric oxide synthase NAD(P)H oxidase superoxide
Introduction
Atrial fibrillation (AF) promotes its own maintenance by causing tachycardia-induced electrophysiological and structural remodeling of the atrial myocardium.1,2 One potential link between AF, tachycardia, and atrial remodeling is oxidative stress. Consistent with this hypothesis, rapid atrial pacing has been shown to increase myocardial peroxynitrite formation and lead to a shortening of the atrial effective refractory period (ERP), both of which are reversed by treatment with the antioxidant and peroxynitrite decomposition catalyst ascorbate.3 More recently, administration of other agents with antioxidant and anti-inflammatory properties such as statins has proved similarly effective in preventing ERP shortening and the vulnerability to AF in a dog model of rapid atrial pacing.4
The mechanisms underlying oxidative injury in the fibrillating atrial myocardium remain to be elucidated; however, an increasing body of evidence indicates that formation of the oxygen-derived free radical superoxide by NAD(P)H oxidases plays a critical role in the development of a wide range of cardiovascular diseases,5eC7 suggesting that this oxidase system may be an important source of oxidative injury in the fibrillating human atrial myocardium.
In neutrophils, NAD(P)H oxidase consists of a core heterodimer comprising the electron transferring plasma membrane subunits p22phox and gp91phox (or nox2) and 4 cytosolic subunits (p40phox, p47phox, p67phox, and the small G-protein rac1/2), which provide regulatory function.5 In vascular smooth muscle and endothelial cells, gp91phox coexists with other homologues termed nox1 and nox4,8eC10 whereas a gp91phox containing NAD(P)H oxidase is the most common isoform in the human ventricular myocardium.7,11
Here, we investigated the source of atrial superoxide production in right atrial appendage (RAA) homogenates and in isolated atrial myocytes from patients in sinus rhythm (SR) undergoing cardiac surgery. We then examined whether the presence of AF affected the magnitude or source of superoxide production in RAA homogenates obtained from patients with permanent or paroxysmal AF.
Materials and Methods
Patient Characteristics
Experiments designed to characterize the source of superoxide production in right atrial tissue (whole tissue homogenates and isolated intact atrial myocytes) were performed in RAAs obtained from 39 patients in SR who underwent first-time coronary artery bypass surgery with or without mitral/aortic valve replacement.
In addition, we evaluated superoxide production and its sources and the expression of NAD(P)H oxidase subunits in RAA homogenates obtained from 15 patients with a history of AF and 15 patients in SR with no previous history of arrhythmias who were matched for treatment, age, and other known risk factors for cardiovascular disease. The AF group included 6 patients who had permanent AF and 9 who had a history of symptomatic paroxysmal AF or persistent AF12 for which they had previous cardioversion. Of these 9 patients, 4 were in AF at the time of admission. We grouped these patients together because of evidence indicating that paroxysmal AF in humans is associated with electrophysiological and structural changes of the atrial myocardium that are similar to those observed in patients with permanent AF,13eC15 suggesting that intermittent bursts of AF, particularly if frequent, are sufficient to instigate atrial remodeling in humans.
A preoperative left ventricular ejection fraction (LVEF; %) was obtained in all patients, whereas left atrial size (by transthoracic echocardiography) was obtained in 24 (14 in AF and 10 in SR) of the 30 patients included in the matched cohort. Demographic and clinical characteristics of the three groups are shown in Table 1. The study was approved by the local research ethics committee, and all patients gave written informed consent.
Atrial Myocyte Isolation Protocol
Samples of RAA were digested (40 minutes) in isolation buffer supplemented with type I collagenase (2 mg/mL; Worthington) and protease (0.2 mg/mL; Sigma), 1% BSA, and 50 eol/L Ca2+, and then transferred into fresh collagenase solution without protease. After isolation, rod-shaped atrial myocytes were enriched (to 90%) by a two-step centrifugation (3 minutes at 600 rpm) in Kraft-Bruhe medium supplemented with L-glutamate, sodium pyruvate, taurine, creatine, succinic acid, Na2ATP, and -OH butyrate and used within 3 hours for superoxide measurements.
Atrial Superoxide Production
Superoxide production in homogenized RAAs or isolated atrial myocytes was measured by lucigenin-enhanced chemiluminescence using a single-tube luminometer (Berthold FB12) modified to maintain the sample temperature at 37°C.16 Briefly, basal chemiluminescence from RAA homogenate or isolated right atrial myocytes was measured in buffer (2 mL) containing lucigenin (5 eol/L) after reaching equilibrium (7 minutes). Further measurements were taken after the addition of inhibitors or substrates of specific oxidase systems16 (Table 2; n=9 to 14 for each comparison). Superoxide scavengers such as polyethylene glycol-superoxide dismutase (PEG-SOD; 650 U/mL) and tiron (10 mmol/L) were also used.
After application of NADPH (100 eol/L), the chemiluminescence signal in RAA homogenate reached a plateau within 3 minutes; however, in atrial myocytes, the signal continued to increase linearly over time. Hence, in this preparation, measurements do not reflect the maximal NADPH-stimulated superoxide production. Atrial homogenates were separated into soluble (cytosolic) and particulate (membrane-associated) fractions by ultracentrifugation, as described previously.17
Because lucigenin (at concentration 20 eol/L) may be subject to redox cycling,18 we compared basal and NADPH-stimulated superoxide measurements in RAA homogenates by 5 eol/L lucigenin with those obtained by using the luminol analogue L-012 (100 eol/L in the presence of ebselen; 50 eol/L)19,20 and the SOD-inhibitable ferricytochrome c reduction,16,21 respectively. In both cases, measurements were closely correlated with those obtained in parallel by lucigenin-enhanced chemiluminescence (basal measurements r=0.88, P<0.005, n=8; NADPH-stimulated r=0.99, P<0.0001, n=11), in agreement with previous reports indicating that 5 eol/L lucigenin is a sensitive and valid probe for assessing superoxide production.22,23
Oxidative Fluorescent Microtopography Using Dihydroethidium
Superoxide production in tissue sections of human RAA was detected using the fluorescent probe DHE.24 Human RAAs were freshly harvested and frozen in OCT compound. Cryosections (30 e) were incubated in Krebs-HEPES buffer for 30 minutes at 37°C with or without PEG-SOD (1000 U/mL) and with dihydroethidium (DHE; 2 eol/L; Molecular Probes) for another 5 minutes at 37°C in darkness. Images were obtained using a laser confocal microscope (Bio-Rad; MRC 1024) at identical acquisition settings using an excitation wavelength of 488 nm. Fluorescence was detected at 585 nm.
Immunoblotting
Protein expression of subunits of NAD(P)H oxidase in RAA homogenates was investigated using rabbit polyclonal antibodies directed against cytosolic p47phox and p67phox subunits (07-001 and 07-002; Upstate) and a rabbit polyclonal antibody directed against the p22phox subunit (a kind gift from Dr Frans B. Wientjes, University College London, United Kingdom) after separation in 10% SDS-PAGE gels and protein transfer to polyvinylidene difluoride membranes. After incubation in primary antibody, antibody binding was visualized using horseradish peroxidase-conjugated anti-rabbit IgG (Promega) chemiluminescence.
Immunolocalization
Freshly isolated human atrial myocytes were concentrated by centrifugation (600 rpm; 3 minutes). Myocytes were plated on laminin (60 e/mL)-coated coverslips and fixed (1.5 hours at room temperature) by applying 50 e蘈 of cooled 4% paraformaldehyde containing 0.2% Triton X-100. Coverslips were washed with PBS three times and then blocked (2 hours at room temperature) with 5% BSA. Cells were incubated with the primary antibody (1 hour at room temperature; rabbit polyclonal anti-p47phox or anti-p67phox Ab, 07-001 and 07-002; Upstate) diluted 1:500 in PBS containing 5% BSA, washed in PBS, then incubated (30 minutes) with 1:50 fluorescence-conjugated goat anti-rabbit IgG (Alexis). Fluorescence images were obtained with a Bio-Rad MRC 1024 scanning confocal microscope.
Conventional and Quantitative Real-Time RT-PCR
Conventional RT-PCR was initially used to evaluate the presence of gp91phox/nox2, its homologues nox1 and nox4, and p22phox in RAA homogenates of patients in SR. Optimal PCR conditions were 100 nmol/L primers for p22phox, gp91phox, nox1, and nox4 (Table 3), 2.5 mmol/L Mg2+ for p22phox, gp91phox, and nox1, and 4.0 mmol/L Mg2+ for nox4. Annealing temperatures were 58°C for all primers except nox4 (68°C) and extension at 72°C. Total RNA extracted from human THP-1 cells (monocytic cell line) served as positive controls for the detection of p22phox and gp91phox, and RNA derived from human EA (endothelial cell line) cells provided positive control for the presence of nox1 and nox4. PCR amplification for p22phox, gp91phox, nox1, and nox4 cDNA generated 341-, 225-, 377-, and 516-bp fragments, respectively, and were resolved on 2% agarose gel.
Quantitative real time RT-PCR (Rotor-Gene 3000; Corbett Research) was used to compare the expression level of the primary transcripts for p22phox and gp91phox/nox2 subunits of NAD(P)H oxidase in patients with AF and their matched controls in SR. Total RNA was extracted from RAA using a double Trizol extraction procedure, with the resulting RNA pellet treated with amplification grade DNase (Invitrogen). Total RNA was then quantified using RiboGreen, adjusted to 25 ng/uL, and then subjected to one-step reverse transcription and PCR amplification (Table 2) using QuantiTect SYBR Green (Qiagen) in the following conditions: 2.5-pmol/L primers for gp91phox and p22phox, 2.5 mmol/L MgCl2, annealing at 60°C, extension at 72°C. Arbitrary copy numbers were analyzed using Rotor-Gene v.5 software (Corbett Research) from standard curves generated from serial dilutions of total RNA extracted from RAAs for human p22phox and gp91phox templates.
Statistical Analysis
Data are expressed as mean±SEM unless specified otherwise. In all cases, n refers to numbers of patients. Nonparametric statistics was used to evaluate some of the responses to inhibitors or substrates of oxidases and the differences in mRNA expression between AF and SR patients. Other comparisons were made by using ANOVA and Fisher’s protected least significant difference post hoc test. A value of P<0.05 was considered statistically significant.
Results
Sources of Superoxide Production in RAA Homogenates and Isolated Myocytes From Patients in SR
Basal superoxide production was determined by lucigenin (5 eol/L)-enhanced chemiluminescence in homogenized RAAs and in intact right atrial myocytes (average 80±8 relative light units [RLU]/mg protein; n=39 patients; and 0.031±0.006 RLU/myocyte; n=16 patients, respectively). Specificity for superoxide was demonstrated by near-abolition of chemiluminescence after coincubation with either PEG-SOD (650 U/mL) or tiron (10 mmol/L; Figure 1A). In addition, basal superoxide production was greatly reduced by the inhibitor of flavin-containing oxidases, diphenyleneiodonium (DPI; 100 eol/L in RAA and 10 eol/L in atrial myocytes; 70%; P<0.005), or by pretreatment with apocynin (80%; P<0.005), a specific inhibitor of NAD(P)H oxidases (Figure 1A). A small reduction (10% to 20%) in basal superoxide production was observed in response to inhibition of cyclooxygenase (indomethacin; P=0.047) in isolated myocytes and of mitochondrial complex I (rotenone; P=0.02) in RAA homogenates. In contrast, inhibition of NO synthase (NOS) with N-nitro-L-arginine methyl ester (L-NAME) tended to increase basal superoxide production (by 25% in RAA homogenates; P=0.35). Xanthine oxidase inhibition with oxypurinol had no effect on basal superoxide production in both preparations.
NADPH-stimulated superoxide release was markedly inhibited by apocynin and DPI in both preparations, whereas oxypurinol, indomethacin, and L-NAME caused very little change (5% to 10% reduction; Figure 1B). Rotenone resulted in a small reduction in NADPH-stimulated superoxide release in both preparations (10% to 20%; P<0.005).
As observed previously in vascular tissue,23 NADPH chemiluminescence was significantly greater (2-fold) than the NADH-evoked signal in RAA homogenates and in atrial myocytes. Further experiments showed that NADPH and NADH were the only substrates that elicited a significant increase in superoxide production in both preparations (Figure 2A).
To further characterize myocardial NAD(P)H oxidase activity, we evaluated NADPH-dependent superoxide production in subcellular fractions of RAA and atrial myocyte homogenates (Figure 2B). Subcellular fractionation by ultracentrifugation into soluble (cytosolic) and particulate (membrane) fractions revealed that >95% of the NADPH-stimulated oxidase activity originated from the particulate fraction in both preparations (P<0.005 versus cytosolic).
Oxidative fluorescent microtopography using the fluorescent probe DHE showed in situ superoxide production in cryosections of RAAs (Figure 3A), which was significantly attenuated after incubation with PEG-SOD (1000 U/mL).
Together, these data indicate that a membrane-associated NAD(P)H oxidase is the main source of superoxide production in RAA homogenates and right atrial myocytes isolated from patients in SR.
Expression of NAD(P)H Oxidase Subunits in Human RAA Homogenates and Isolated Myocytes
To determine whether NAD(P)H oxidase subunits were expressed in the human atrial myocardium, we used a combined approach using antibodies directed against the p22phox, p47phox, and p67phox subunits of the NAD(P)H oxidase in RAA homogenates and isolated atrial myocytes and conventional RT-PCR to detect p22phox and gp91phox mRNA in isolated atrial myocytes.
Figure 3B shows that the p22phox, p47phox, and p67phox subunits are detected in RAA homogenate and atrial myocytes. p22phox, p47phox, and p67phox appeared as single bands, which were aligned to their respective positive control bands (+ human neutrophils). Conventional RT-PCR confirmed the presence of p22phox and gp91phox mRNA in purified human atrial myocytes but did not detect other gp91phox homologues such as nox1 or nox4 (Figure 3C).
p47phox or p67phox immunolocalization in isolated human atrial myocytes showed a diffuse pattern consistent with cytosolic and membrane-bound localization of these NAD(P)H oxidase subunits (Figure 3D). Omission of the primary antibody (negative control) resulted in no staining in 17 of 17 cells from 3 patients. Corresponding phase contrast images of the same cells are shown in the bottom panel of Figure 3D.
Atrial NAD(P)H Oxidase Activity Is Increased in Patients With AF
NADPH-stimulated superoxide production was significantly higher in RAA homogenates derived from AF patients compared with patients in SR (Figure 4A; P=0.02), who were matched for age, LVEF (46±12% in SR and 50±13% in AF; P=0.40), treatment, and known risk factors (Table 1). Basal superoxide production also tended to be higher in the AF group, but this difference did not reach statistical significance (P=0.15; Figure 4A). LVEF was not different between AF and SR patients (Table 1). As expected, atrial size was significantly larger in the AF group (4.05±0.16 cm in SR patients; n=10 versus 5.01±0.17 cm in AF patients; n=14; P=0.0006); however, there were no differences in LVEF (P=0.97), atrial size (P=0.38), or superoxide production (P=0.32) between patients with permanent or paroxysmal AF. Similarly, we did not find a significant correlation between atrial size (range 3.3 to 6 cm) and basal or NADPH-stimulated superoxide production (r2=0.03 for both), suggesting that the latter is unlikely to be a consequence of atrial hypertrophy and remodeling in AF.
As observed in patients in SR, basal superoxide production in RAA homogenates from AF patients was inhibited by apocynin (81%; P=0.005) and rotenone (40%; P=0.02; Figure 4B; P=0.55 and 0.33, respectively, for comparison with the apocynin- and rotenone-induced inhibition in SR patients). Interestingly, in fibrillating RAA, pretreatment with L-NAME (but not with D-NAME; 1 mmol/L; n=3; data not shown) significantly reduced basal superoxide production by 40% (P=0.01 versus control and P=0.002 versus SR), indicating that dysfunctional NOS may become a significant source of basal superoxide production in fibrillating atria. Similar results were obtained by using NG-monomethyl-L-arginine (L-NMMA; 1 mmol/L; n=3; data not shown).
Pretreatment with apocynin significantly decreased NADPH-stimulated superoxide release (90%; P=0.005), and an 20% reduction in NADPH-stimulated superoxide release was also observed in response to L-NAME (P=0.008 versus control and P=0.07 versus SR; Figure 4B). Rotenone, oxypurinol, and antimycin-A (10 eol/L) had no significant effect.
Quantification of p22phox and gp91phox mRNA in RAA from patients with AF and matched controls in SR using real-time RT-PCR showed no difference in the expression of these subunits between groups (n=15 patients per group, arbitrary copy numbers; p22phox 328.5±71.4 in AF versus 280.1±37.5 in SR; gp91phox 139.9±51 in AF versus 138.6±27.1 in SR; P=NS; Figure 5).
Discussion
The main findings of this study are as follows: (1) a membrane-bound gp91phox/nox2 containing NAD(P)H oxidase is the main source of superoxide production in human atrial myocytes; (2) NAD(P)H-dependent superoxide production from RAA homogenates is increased in patients with a history of AF in the absence of changes in the mRNA expression of the p22phox and gp91phox/nox2 subunits; and (3) in atrial tissue from patients with AF, NOS inhibition resulted in a significant reduction in lucigenin-enhanced chemiluminescence, suggesting that NOS may be "uncoupled" and contribute to superoxide production in fibrillating atria.
These findings suggest that a myocardial NAD(P)H oxidase and, to a lesser extent, NOS and mitochondrial oxidases may play an important role in the atrial oxidative injury and electrophysiological remodeling observed in patients with AF.
Characterization of the Human Atrial NAD(P)H Oxidase
Our findings indicate that a membrane-bound gp91phox NAD(P)H oxidase is present in human right atrial myocytes, where it constitutes the main source of superoxide production. This is evidenced by the observation that RAA superoxide production is significantly increased by the addition of NADPH or NADH but not by application of substrates of other oxidase systems. Similarly, basal and NADPH-stimulated superoxide production (the latter resulting almost exclusively from the particulate fraction) were nearly abolished by apocynin and DPI but not by inhibitors of other oxidases. Pretreatment of RAA homogenates with rotenone and antimycin-A revealed a trend toward a reduction in basal superoxide production, suggesting that the contribution of mitochondrial oxidases to basal superoxide production in the human atrial myocardium is small. We also provide evidence of atrial myocyte-specific expression of the NAD(P)H oxidase subunits p22phox, p47phox, and p67phox at the protein level and of p22phox and gp91phox/nox2 (but not nox1 or nox4) at the messenger level.
Together, our findings indicate that human atrial myocytes have the capability of producing superoxide through a membrane-bound gp91phox/nox2 containing NAD(P)H oxidase.
Myocardial Superoxide Production, Oxidative Stress, and AF
Increased atrial superoxide production may have important implications in the ionic remodeling process, which promotes AF maintenance. Enhanced myocardial superoxide production can decrease NO bioavailability by scavenging NO (to form the potent oxidant peroxynitrite) and by "uncoupling" NOS activity,25,26 which in turn may lead to an increase in myocardial oxygen consumption,27 -adrenergic responsiveness,28,29 and thrombogenic risk.30
Carnes et al3 reported an increase in atrial 3-nitrotyrosine content (a marker of peroxynitrite formation in vivo) in a canine model of rapid atrial pacing that was associated with an abbreviation of the atrial ERP. Interestingly, administration of the antioxidant ascorbate attenuated atrial peroxynitrite formation and electrophysiological remodeling in these animals. Similarly, statins have been shown recently to reduce angiotensin II-stimulated (but not basal) NAD(P)H oxidase activity in human RAA31 and to attenuate ERP reduction and AF inducibility in dogs exposed to rapid atrial pacing.4 Together, these findings suggest the presence of a causal relationship between pro-oxidative cellular redox state, atrial ionic remodeling, and AF.
Our findings clearly show that NAD(P)H oxidase is the main source of atrial superoxide production in AF and in SR; however, they also indicate that the contribution of NOS to myocardial superoxide production is significantly increased in the fibrillating atrial myocardium. This is evidenced by a reduction in basal- and NADPH-stimulated superoxide production after pretreatment with L-NAME or L-NMMA in RAA homogenates from AF patients. These data differ significantly from those obtained in patients in SR, in whom L-NMMA tended to increase lucigenin chemiluminescence.
It is now well established that NOSs can release superoxide when deprived of their critical cofactor tetrahydropbiopterin (BH4) or of their substrate L-arginine.32 Under these conditions (referred to as "NOS uncoupling"), electron flow results in a reduction of molecular oxygen at the prosthetic heme site of the enzyme rather than in NO synthesis. In the presence of increased oxidative stress, oxidation of BH4 can uncouple NOS to generate reactive oxygen species.26,33 The notion that this may occur in the fibrillating atrial myocardium is supported by the recent findings of Cai et al,30 who showed that eNOS expression in the left atrial appendage in a porcine model of rapid pacing-induced AF did not differ from control animals despite a documented reduction in NO formation at this site.
We observed that the capacity of NAD(P)H oxidase to generate superoxide was significantly higher in patients with AF compared with matched patients in SR. Enhanced NAD(P)H oxidase activity can either be attributable to increased expression or post-translational modifications of the oxidase subunits. Real-time quantitative RT-PCR revealed no difference in p22phox or gp91phox mRNA between AF and SR groups, suggesting that the increase in NAD(P)H oxidase activity observed in AF may be attributable to the latter mechanism. Indeed, a key feature of cardiovascular NAD(P)H oxidase is its responsiveness to hormones, hemodynamic forces, and local metabolic changes.5 In particular, vascular NAD(P)H oxidase activity is known to be greatly increased by angiotensin II via phosphorylation of p47phox.34 In agreement with these findings, we found that angiotensin II potently stimulates atrial NAD(P)H oxidase activity (Kim et al, unpublished observations, 2004) which, like its vascular counterpart, is markedly inhibited by chelerythrine, indicating that these effects are at least in part protein kinase C dependent. Together, these findings suggest that the activation of the renin-angiotensin system may be an important underlying mechanism for the increased NAD(P)H oxidase activity and ensuing atrial oxidative stress observed in human AF.
Limitations
The findings of our study should be interpreted in the context of their limitations. Because regional differences in endocardial NO production have been reported in a porcine model of AF induced by rapid right atrial pacing,30 caution should be exerted in extrapolating our findings to the whole atrial myocardium. However, in humans, AF has been shown to cause similar changes in size, function,35 and ion channel protein expression14 in both atria, suggesting that remodeling may be a diffuse phenomenon.
To our knowledge, our study provides the first evidence of NOS uncoupling in the human myocardium; however, detailed investigation of the mechanisms responsible for this phenomenon was hampered by limited tissue availability because the incidence of AF in patients undergoing first-time elective coronary revascularization was <10%. Because all our SR controls patients underwent coronary artery bypass surgery, we elected to recruit AF patients from a similar cohort to match the two groups for risk factors (eg, age, atherosclerosis, and diabetes mellitus) and treatment agents (eg, statins, angiotensin-converting enzyme inhibitors, or angiotensin II receptor blockers) that are known to affect myocardial/endothelial superoxide production.9,31
Conclusions
Increasing evidence indicates that NAD(P)H oxidase may play an important role in the myocardial response to stress or injury. Our findings indicate that the primary source of superoxide production in the human atrial myocardium is an NAD(P)H oxidase in patients in SR and in those with AF. However, in the fibrillating myocardium, NOS contributes significantly to basal- and NADPH-stimulated superoxide release, suggesting that increased oxidative stress in this condition may lead to NOS uncoupling with potentially important implications on myocardial function and thrombogenesis.
Acknowledgments
Y.M.K. was in receipt of a Wellcome Trust Cardiovascular Research Initiative Scholarship, a Clarendon Award, and Mary Goodger Award from the University of Oxford. B.C., K.M.C., and Y.H.Z. are funded by the British Heart Foundation and B.C. by the Garfield Weston Trust. We are grateful to Dr J. Khoo for his help with dihydroethidium oxidative fluorescent microtopography.
References
Wijffels MCEF, Kirchhof CJHJ, Dorland R, Allessie MA. Atrial fibrillation begets atrial fibrillation: a study in awake chronically instrumented goats. Circulation. 1995; 92: 1954eC1968.
Nattel S. New ideas about atrial fibrillation 50 years on. Nature. 2002; 415: 219eC226.
Carnes CA, Chung MK, Nakayama T, Nakayama H, Baliga RS, Piao S, Kanderian A, Pavia S, Hamlin RL, McCarthy PM, Bauer JA, Van Wagoner DR. Ascorbate attenuates atrial pacing-induced peroxynitrite formation and electrical remodeling and decreases the incidence of postoperative atrial fibrillation. Circ Res. 2001; 89: 32eeC38e.
Shiroshita-Takeshita A, Schram G, Lavoie J, Nattel S. Effect of simvastatin and antioxidant vitamins on atrial fibrillation promotion by atrial-tachycardia remodeling in dogs. Circulation. 2004; 110: 2313eC2319.
Griendling KK, Sorescu D, Ushio Fukai M. NAD(P)H oxidase: role in cardiovascular biology and disease. Circ Res. 2000; 86: 494eC501.
Li J-M, Gall NP, Grieve DJ, Chen M, Shah AM. Activation of NADPH oxidase during progression of cardiac hypertrophy to failure. Hypertension. 2002; 40: 477eC484.
Heymes C, Bendall JK, Ratajczak P, Cave AC, Samuel J-L, Hasenfuss G, Shah AM. Increased myocardial NADPH oxidase activity in human heart failure. J Am Coll Cardiol. 2003; 41: 2164eC2171.
Lassegue B, Sorescu D, Szocs K, Yin Q, Akers M, Zhang Y, Grant SL, Lambeth JD, Griendling KK. Novel gp91phox homologues in vascular smooth muscle cells: nox1 mediates angiotensin II-induced superoxide formation and redox-sensitive signaling pathways. Circ Res. 2001; 88: 888eC894.
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: 1429eC1435.
Ago T, Kitazono T, Ooboshi H, Iyama T, Han YH, Takada J, Wakisaka M, Ibayashi S, Utsumi H, Iida M. Nox4 as the major catalytic component of an endothelial NAD(P)H oxidase. Circulation. 2004; 109: 227eC233.
Krijnen PAJ, Meischl C, Hack CE, Meijer CJLM, Visser CA, Roos D, Niessen HWM. Increased Nox2 expression in human cardiomyocytes after acute myocardial infarction. J Clin Pathol. 2003; 56: 194eC199.
McNamara RL, Brass LM, Drozda JP Jr, Go AS, Halperin JL, Kerr CR, Levy S, Malenka DJ, Mittal S, Pelosi F Jr, Rosenberg Y, Stryer D, Wyse DG, Radford MJ, Goff DC Jr, Grover FL, Heidenreich PA, Peterson ED, Redberg RF. ACC/AHA key data elements and definitions for measuring the clinical management and outcomes of patients with atrial fibrillation: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Data Standards. Circulation. 2004; 109: 3223eC3243.
Sparks PB, Jayaprakash S, Vohra JK, Kalman JM. Electrical remodeling of the atria associated with paroxysmal and chronic atrial flutter. Circulation. 2000; 102: 1807eC1813.
Brundel BJJM, Van Gelder IC, Henning RH, Tuinenburg AE, Wietses M, Grandjean JG, Wilde AAM, Van Gilst WH, Crijns HJGM. Alterations in potassium channel gene expression in atria of patients with persistent and paroxysmal atrial fibrillation: differential regulation of protein and mRNA levels for K+ channels. J Am Coll Cardiol. 2001; 37: 926eC932.
Brundel BJJM, Ausma J, van Gelder IC, Van Der Want JJL, van Gilst WH, Crijns HJGM, Henning RH. Activation of proteolysis by calpains and structural changes in human paroxysmal and persistent atrial fibrillation. Cardiovasc Res. 2002; 54: 380eC389.
Guzik TJ, West NE, 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: E85eCE90.
Mohazzab H, Kaminski PM, Agarwal R, Wolin MS. Potential role of a membrane-bound NADH oxidoreductase in nitric oxide release and arterial relaxation to nitroprusside. Circ Res. 1999; 84: 220eC228.
Vasquez-Vivar J, Hogg N, Pritchard Jr KA, Martasek P, Kalyanaraman B. Superoxide anion formation from lucigenin: an electron spin resonance spin-trapping study. FEBS Lett. 1997; 403: 127eC130.
Nishinaka Y, Aramaki Y, Yoshida H, Masuya H, Sugawara T, Ichimori YA. New sensitive chemiluminescence probe, L-012, for measuring the production of superoxide anion by cells. Biochem Biophys Res Commun. 1993; 193: 554eC559.
Daiber A, Oelze M, August M, Wendt M, Sydow K, Wieboldt H, Kleschyov AL, Munzel T. Detection of superoxide and peroxynitrite in model systems and mitochondria by the luminol analogue L-012. Free Radic Res. 2004; 38: 259eC269.
Pritchard KA Jr, Groszek L, Smalley DM, Sessa WC, Wu M, Villalon P, Wolin MS, Stemerman MB. Native low-density lipoprotein increases endothelial cell nitric oxide synthase generation of superoxide anion. Circ Res. 1995; 77: 510eC518.
Skatchkov MP, Sperling D, Hink U, Mulsch A, Harrison DG, Sindermann I, Meinertz T, Munzel T. Validation of lucigenin as a chemiluminescent probe to monitor vascular superoxide as well as basal vascular nitric oxide production. Biochem Biophys Res Commun. 1999; 254: 319eC324.
Li J-M, Shah AM. Differential NADPH- versus NADH-dependent superoxide production by phagocyte-type endothelial cell NADPH oxidase. Cardiovasc Res. 2001; 52: 477eC486.
Zhao H, Kalivendi S, Zhang H, Joseph J, Nithipatikom K, Vasquez-Vivar J, Kalyanaraman B. Superoxide reacts with hydroethidine but forms a fluorescent product that is distinctly different from ethidium: potential implications in intracellular fluorescence detection of superoxide. Free Radic Biol Med. 2003; 34: 1359eC1368.
Mollnau H, Wendt M, Szocs K, Lassegue B, Schulz E, Oelze M, Li H, Bodenschatz M, August M, Kleschyov AL, Tsilimingas N, Walter U, Forstermann U, Meinertz T, Griendling K, Munzel T. Effects of angiotensin II infusion on the expression and function of NAD(P)H oxidase and components of nitric oxide/cGMP signaling. Circ Res. 2002; 90: 58eeC65e.
Landmesser U, Dikalov S, Price SR, McCann L, Fukai T, Holland SM, Mitch WE, Harrison DG. Oxidation of tetrahydrobiopterin leads to uncoupling of endothelial cell nitric oxide synthase in hypertension. J Clin Invest. 2003; 111: 1201eC1209.
Adler A, Messina E, Sherman B, Wang Z, Huang H, Linke A, Hintze TH. NAD(P)H oxidase-generated superoxide anion accounts for reduced control of myocardial O2 consumption by NO in old Fischer 344 rats. Am J Physiol Heart Circ Physiol. 2003; 285: H1015eCH1022.
Balligand JL. Regulation of cardiac beta-adrenergic response by nitric oxide. Cardiovasc Res. 1999; 43: 607eC620.
Sears CE, Ashley EA, Casadei B. Nitric oxide control of cardiac function: is neuronal nitric oxide synthase a key component Philos Trans R Soc Lond B. 2004; 359: 1021eC1044.
Cai H, Li Z, Goette A, Mera F, Honeycutt C, Feterik K, Wilcox JN, Dudley SC Jr, Harrison DG, Langberg JJ. Downregulation of endocardial nitric oxide synthase expression and nitric oxide production in atrial fibrillation: potential mechanisms for atrial thrombosis and stroke. Circulation. 2002; 106: 2854eC2858.
Maack C, Kartes T, Kilter H, Schafers H-J, Nickenig G, Bohm M, Laufs U. Oxygen free radical release in human failing myocardium is associated with increased activity of Rac1-GTPase and represents a target for statin treatment. Circulation. 2003; 108: 1567eC1574.
Vasquez-Vivar J, Kalyanaraman B, Martasek P. The role of tetrahydrobiopterin in superoxide generation from eNOS: enzymology and physiological implications. Free Radic Res. 2003; 37: 121eC127.
Alp NJ, Mussa S, Khoo J, Cai S, Guzik T, Jefferson A, Goh N, Rockett KA, Channon KM. Tetrahydrobiopterin-dependent preservation of nitric oxide-mediated endothelial function in diabetes by targeted transgenic GTP-cyclohydrolase I overexpression. J Clin Invest. 2003; 112: 725eC735.
Li J-M, Shah AM. Mechanism of endothelial cell NADPH oxidase activation by angiotensin II. Role of the p47phox subunit. J Biol Chem. 2003; 278: 12094eC12100.
Escudero E, San Mauro M, Laugle C. Bilateral atrial function after chemical cardioversion of atrial fibrillation with amiodarone: an echo-Doppler study. J Am Soc Echocardiogr. 1998; 11: 365eC371.(Young M. Kim, Tomasz J. G)