Skip Navigation

Cardiovascular Research 2007 75(2):349-358; doi:10.1016/j.cardiores.2007.03.030
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary Data
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Selemidis, S.
Right arrow Articles by Drummond, G. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Selemidis, S.
Right arrow Articles by Drummond, G. R.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Copyright © 2007, European Society of Cardiology

Nitric oxide suppresses NADPH oxidase-dependent superoxide production by S-nitrosylation in human endothelial cells

Stavros Selemidisa,b,c, Gregory J. Dustinga,c,d,*, Hitesh Peshavariyaa, Barbara K. Kemp-Harperb,c and Grant R. Drummondb,c

aBernard O’Brien Institute of Microsurgery, University of Melbourne, Fitzroy, Victoria 3065, Australia
bDepartment of Pharmacology, Monash University, Clayton, Victoria 3800, Australia
cCentre for Vascular Health, Monash University, Clayton, Victoria 3800, Australia
dDepartments of Surgery and Pharmacology, University of Melbourne, Victoria 3010, Australia

* Corresponding author. Tissue Engineering, Bernard O'Brien Institute of Microsurgery, University of Melbourne, 42 Fitzroy Street, Fitzroy, Victoria 3065, Australia. Tel.: +61 3 9288 4047; fax: +61 3 9416 0926. dusting{at}unimelb.edu.au

Received 13 November 2006; revised 16 March 2007; accepted 28 March 2007


   Abstract
Top
 Abstract
1. Introduction
 2. Methods
 3. Results
 4. Discussion
 Appendix A. Supplementary data
 Acknowledgments
 References
 
Objective: Endothelial NADPH oxidase is a major source of superoxide in blood vessels and is implicated in the oxidative stress accompanying vascular diseases, including atherosclerosis. Here we investigate the regulation of NADPH oxidase activity by nitric oxide (NO).

Methods: Human cultured microvascular endothelial cells (HMEC-1) were treated with the NO donors, diethylenetriamine (DETA)-NONOate, S-nitroso-N-acetylpenicillamine (SNAP) or sodium nitroprusside (SNP) for 0.5–24 h. Superoxide production was measured by lucigenin chemiluminescence and dihydroethidium fluorescence, while NADPH oxidase subunit expression was measured via Western blotting. S-nitrosylation was assessed using the 2,3-diaminonapthalene (DAN) assay, and via immunoblotting with an anti-nitrosocysteine antibody.

Results: Specific siRNA reduced Nox2 and Nox4 protein expression and markedly decreased superoxide production in HMEC-1. DETA-NONOate (10–300 µmol/L) suppressed superoxide production in HMEC-1 in a concentration- and time-dependent manner, which was not entirely attributable to stoichiometric reaction with NO, for the effect was observed more than 6 h after removing DETA-NONOate from solution. Similarly, sustained attenuation of superoxide production was achieved with SNP (10–100 µmol/L) and SNAP (10–100 µmol/L). The suppressive effect of NO was not dependent on (1) the sGC/cGMP/PKG pathway, (2) peroxynitrite-formation, (3) reduced protein expression of NADPH oxidase subunits or (4) dissociation of NADPH oxidase subunits. Treatment with NO caused S-nitrosylation of the crucial organizer subunit p47phox, and de-nitrosylation with UV light restored superoxide production.

Conclusions: NO causes sustained suppression of NADPH oxidase-dependent superoxide production in human endothelial cells by S-nitrosylation of p47phox. These findings highlight a novel approach by which vascular oxidative stress might be suppressed by NO donors.

KEYWORDS Nitric oxide; Endothelium; NADPH oxidase; Oxygen radicals


   1. Introduction
Top
 Abstract
 1. Introduction
2. Methods
 3. Results
 4. Discussion
 Appendix A. Supplementary data
 Acknowledgments
 References
 
NADPH oxidases are enzymes composed of up to five protein subunits: a flavin-containing catalytic ‘Nox’ subunit, the nature of which varies with different isoforms (Nox1, 2 or 4); a smaller subunit, p22phox, which forms a stable heterodimer with the Nox in cellular membranes; a cytosolic activator protein (p67phox or NoxA1); an organiser protein (p47phox or NoxO1); and a small G-protein, Rac-1 or 2 [1]. When assembled, all isoforms of NADPH oxidase generate ROS (superoxide and/or hydrogen peroxide) by transferring electrons from NADPH to FAD to molecular oxygen, which is bound by a heme moiety on the Nox subunit [1].

Isoforms of NADPH oxidase are expressed by all major cell types of the vascular wall, with Nox2- and Nox4-utilising oxidases being major contributors to endothelial superoxide production, at least during resting conditions [2–5]. Endothelial NADPH oxidases appear to exist as preassembled complexes associated with the cytoskeleton and, when expressed at normal levels, they produce low levels of ROS to influence redox-dependent cell signalling processes including ion channel opening, protein phosphatase and kinase activity and gene expression [3,4]. However, when expression of these and other NADPH oxidase isoforms (e.g. Nox1) are inappropriately up-regulated by pathophysiological stimuli such as oscillatory shear stress, hyperglycaemia, cytokines, growth factors, and/or oxidised lipids, they may become major initiators of oxidative stress causing reduced nitric oxide (NO) bioavailability, upregulation of adhesion molecules and chemokines, and apoptosis [1]. Indeed, we and others have shown that upregulation of endothelial NADPH oxidase-derived ROS is associated with neointima formation/atherogenesis in both animal models [1,6,7] and in humans [8]. Moreover, genetic deletion of the gene encoding p47phox markedly reduced atherosclerotic lesion development in hypercholesterolemic apolipoprotein E-deficient mice, confirming a causal role for NADPH oxidases in atherogenesis [9].

The powerful vasodilator properties of NO donors have been exploited for over a century for the acute management of angina pectoris – a major symptom of coronary artery disease. However, in addition to their effects on vascular tone, NO donors may exert numerous other protective effects on the blood vessel wall, suggesting that their therapeutic potential may extend beyond acute symptomatic treatment, to prevention, and even reversal, of artery disease [10]. These effects include inhibition of lipid oxidation, vascular smooth muscle cell proliferation, inflammatory molecule and chemokine expression and platelet activation and aggregation. More recently, several reports have highlighted that certain NO donors may also reduce superoxide levels in vascular cells both in vitro [11] and in vivo [12,13]. However, it is unclear whether these ROS lowering effects of NO donors were simply due to scavenging of superoxide by NO, inhibition of signal transduction elements upstream of NADPH oxidase, and/or to a direct effect on one/or more NADPH oxidase subunits. Therefore we set out to determine whether treatment of human cells with NO donors affects superoxide production, and if so, whether this involved a specific inhibitory effect at the level of NADPH oxidase.


   2. Methods
Top
 Abstract
 1. Introduction
 2. Methods
3. Results
 4. Discussion
 Appendix A. Supplementary data
 Acknowledgments
 References
 
2.1 Cell culture
Human microvascular endothelial cells (HMEC-1) were cultured in MCDB-131 media (Invitrogen) supplemented with 10% fetal bovine serum (FBS), L-glutamine (10 mmol/L), penicillin/streptomycin (all from Trace Biosciences) and hydrocortisone (50 µmol/L; Sigma) in a CO2/O2 incubator at 37 °C. Cells (passage 10–16) were seeded in a 1:8 ratio and, unless otherwise stated, used 1 day post-confluence.

2.2 siRNA transfection
HMEC-1 were seeded and allowed to grow to ~70–80% confluency in 24-well plates. On day of transfection, three transfection solutions were prepared. The first, control solution, was composed of 200 µL of serum and antibiotic-free MCDB-131. The second and third transfection solutions each contained 1 µL of lipofectamine 2000 (Invitrogen) and either 5 µL of control siRNA (250 nmol/L) or specific siRNA for Nox2 or Nox4 (250 nmol/L) in 200 µL of serum- and antibiotic-free MCDB-131. Control and transfection solutions were mixed gently for 15 min at room temperature and then added to the cells. Following incubation for 5 h, the transfection media was removed from cells and replaced with fresh MCDB-131 with 10% FBS. At 48 h after transfection, lucigenin chemiluminescence, dihydroethidium (DHE) fluorescence or Western blotting was employed.

2.3 NO donor treatment protocols
HMEC-1 were treated with either the slow-release (i.e. half-life >20 h) NO donor, diethylenetriamine (DETA)-NONOate (3–300 µmol/L), or the solvent used to dissolve DETA-NONOate, NaOH (0.01 mol/L), for 0.5 to 24 h. In separate cells, the NO donors, S-nitroso-N-acetylpenicillamine (SNAP; 1–100 µmol/L) or sodium nitroprusside (SNP; 1–100 µmol/L) were applied for 6 h. In all but the initial studies (Fig. 2A), the possibility of stoichiometric removal of superoxide by NO at time of assay was eliminated by washing the NO donor from the cells with Krebs–HEPES (see Supplementary data) at the end of the incubation period.


Figure 2
View larger version (28K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2 Effect of Nox2 and Nox4 siRNA on superoxide production measured by lucigenin chemiluminescence and DHE fluorescence in HMEC-1. (A and E) Representative immunoblots showing Nox2 and Nox4 expression in control HMEC-1 and following transfection with either scrambled siRNA (‘control siRNA’), Nox2 siRNA or Nox4 siRNA. Also shown in (A), as a positive control, is expression of Nox2 and the effect of Nox2 siRNA in U937 monocytes. (B and F) Densitometric analysis of Nox2 and Nox4 expression normalized as a ratio of β-actin expression and then as a percentage of control. (C and G) Superoxide production measured by lucigenin chemiluminescence and (D and H) DHE fluorescence. Values (mean±S.E.M. from 6 experiments) are expressed as a percentage of control. *P<0.05 vs. Control siRNA.

 
2.4 Superoxide detection with lucigenin chemiluminescence and dihydroethidium (DHE) fluorescence in intact HMEC-1
Following NO donor treatment, HMEC-1 were pre-incubated for 30 min with a Krebs–HEPES-based pre-incubation solution containing NADPH (100 µmol/L), to ensure adequate substrate availability for NADPH oxidase, and diethyldithiocarbamate (DETCA; 3 mmol/L), to inactivate endogenous Cu2+/Zn2+ superoxide dismutases (SOD). Some cells were also treated with inhibitor drugs for 1 h prior to exposure to the Krebs–HEPES pre-incubation solution and then for the duration of the pre-incubation. Cells were then exposed to a Krebs–HEPES-based assay solution containing lucigenin (5 µmol/L), NADPH (100 µmol/L) and the appropriate inhibitor. Photon emission, as a measure of superoxide production [14], was recorded from each well every 2 min using a TopCount (Packard) and averaged over 30 min. Cell numbers in each well were evaluated using the Cell Titer Aqueous One Solution Cell Proliferation assay (Promega). Individual data points for each treatment group were derived from the average of six replicates.

In some experiments, superoxide production was measured with DHE fluorescence as we have recently described [5].

2.5 Superoxide detection with lucigenin chemiluminescence in HMEC-1 homogenates
Following NO donor treatment, 100-mm dishes of confluent HMEC-1 were washed in ice-cold phosphate-buffered saline (PBS; see Supplementary data) and lysed by incubation in a high-sucrose HEPES buffer (see Supplementary data) for 20 min at 4 °C. Protein concentration was then determined using a kit (Bio-Rad). Some homogenates (500 µL aliquots) were then placed 30 cm from a 4.9-W output, UV-C lamp for 5 min at 4 °C prior to superoxide detection, to disrupt S–NO bonds. Other homogenates were exposed to antagonists in PBS buffer for 1 h prior to superoxide detection. Finally, 10 µg of protein was added to an assay solution containing lucigenin (5 µmol/L) and NADPH (100 µmol/L) and photon emission was monitored as described above.

2.6 Whole-cell protein extraction for immunoprecipitation and Western blotting
Following NO donor treatment, 60-mm dishes of confluent HMEC-1 were washed with cold PBS and incubated for 60 min at 4 °C in lysis buffer (see Supplementary data). The lysate was then centrifuged for 20 min (14,000 rpm at 4 °C), and the resulting supernatant retained.

2.7 Immunoprecipitation
Whole-cell protein (500 µg) was exposed to rabbit IgG (10 µg) for 1 h at 4 °C and then ‘precleared’ using protein-G agarose beads. Following centrifugation, the supernatant was collected and incubated overnight at 4 °C with 5 µg of either anti-p22phox antibody or anti-p47phox antibody. An equal volume of protein G agarose was then added and the samples incubated for a further 1 h at 4 °C. The suspension was then centrifuged and the subsequent pellet containing protein G agarose beads and immunoprecipitated proteins was washed with chilled PBS.

2.8 Western blotting
Whole-cell or immunoprecipitated proteins (40 µg) were separated by SDS–PAGE and immunoblotted with primary antibodies against Nox2 (rabbit polyclonal, 1:1000 dilution, Upstate), Nox4 (goat polyclonal, 1:1000 dilution, Santa Cruz), p22phox (rabbit polyclonal, 1:1000 dilution, Santa Cruz), Rac-1 (mouse monoclonal, 1:1000 dilution, Transduction Laboratories), p47phox (mouse monoclonal, 1:1000 dilution, Santa Cruz) or nitrosocysteine (mouse monoclonal, 1:1000 dilution, AG Scientific) and appropriate secondary antibodies conjugated to horseradish peroxidase. Immunoreactive bands were visualised using ECL (Amersham).

2.9 Detection of S-nitrosylated proteins with the 2,3-diaminonapthalene (DAN) assay
P22phox protein was immunoprecipitated from control and NO donor-treated HMEC-1. Given that p22phox is common to all NADPH oxidase isoforms (and therefore associates directly or indirectly with all known NADPH oxidase subunits), immunoprecipitation with a p22phox antibody afforded a method of separating NADPH oxidase subunits from other cellular proteins that might become S-nitrosylated. The pellet containing the antibody/antigen complex and protein G beads was washed with PBS and exposed to high glycine (pH 3) buffer to separate the protein G from the antibody/antigen complex. The supernatant obtained from the washes was pooled and then divided into two equal aliquots. All aliquots were incubated with DAN (100 µmol/L) and one aliquot from each treatment group was subsequently exposed to HgCl2 (10 µmol/L). Samples were kept on ice in the dark for 30 min before being exposed to 0.1 mol/L NaOH. Fluorescence of the triazole produced by the reaction of released NO with DAN (excitation 375 nm; emission 450 nm) was then measured from each aliquot with a Polarstar microplate reader. The fluorescence intensity generated in the absence of HgCl2 (background) was then subtracted from the fluorescence intensity generated in the presence of HgCl2. Fluorescence intensity above background was normalised to the amount of protein.

2.10 Drugs and their suppliers
See Supplementary data.

2.11 Statistical analysis
Results are expressed as mean±standard error of the mean (S.E.M.). Statistical comparisons were made using Student's unpaired t-test or one-way analysis of variance with Tukey–Kramer's post hoc test. P<0.05 was considered significant.


   3. Results
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
4. Discussion
 Appendix A. Supplementary data
 Acknowledgments
 References
 
3.1 Superoxide detection with chemiluminescence
NADPH-dependent chemiluminescence in intact HMEC-1 was attenuated by 50% and 90%, respectively, by exogenous SOD (600 U/mL) and the cell permeable superoxide scavenger, MnTMPyP (25 µmol/L; Fig. 1A). In endothelial cell homogenates, SOD appeared to be more effective at reducing NADPH-dependent chemiluminescence (i.e. by 75%), presumably due to its better access to intracellular sites of superoxide production (Supplementary Fig. 1B), while the cell permeable ROS scavenger, tiron, abolished the signal. These observations confirm that the lucigenin-enhanced chemiluminescence signal in response to NADPH in both intact and homogenised endothelial cells is due to superoxide.


Figure 1
View larger version (22K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1 Superoxide production in HMEC-1 measured by lucigenin chemiluminescence. HMEC-1 were treated with NADPH (100 µmol/L) and diethyldithiocarbamate (3 mmol/L) for 45 min in the absence (control) or presence of (A) superoxide dismutase (SOD; 600 U/ml) or MnTMPyP (25 µmol/L); (B) diphenyleneiodonium (DPI; 100 nmol/L) or apocynin (1 mmol/L); and (C) NG-nitro-L-arginine methyl ester (L-NAME; 100 µmol/L), indomethacin (Indo; 3 µmol/L), 17-octadecynoic acid (17-OD; 100 µmol/L), allopurinol (Allop; 100 µmol/L) or rotenone (ROT; 1 µmol/L) prior to superoxide assay. Values (mean±S.E.M. from 6 experiments) are expressed as a percentage of control. *P<0.05 vs. control.

 
NADPH-dependent superoxide production by HMEC-1 was abolished by the flavin inhibitor, DPI (100 nmol/L), and markedly reduced by the NADPH oxidase inhibitor, apocynin (1 mmol/L, Fig. 1B; see also Supplementary Fig. 1B). In contrast, L-NAME (100 µmol/L), indomethacin (3 µmol/L), 17-OD (3 µmol/L), allopurinol (100 µmol/L) and rotenone (1 µmol/L) had no effect on superoxide production in intact HMEC-1 (Fig. 1C) or in HMEC-1 homogenates (Supplementary Fig. 1A) indicating that neither eNOS, cyclo-oxygenase, cytochrome P450, xanthine oxidase nor the mitochondrial respiratory complex are significant sources of superoxide in endothelial cells under these conditions.

Transfection of HMEC-1 with Nox2 siRNA significantly reduced expression of Nox2 protein (Fig. 2A and B) but had no effect on Nox4 expression (data not shown). Conversely, transfection with Nox4 siRNA markedly reduced the expression of Nox4 protein (Fig. 2E and F) and had no effect on Nox2 protein expression. Importantly, transfection of HMEC-1 with a scrambled control siRNA had no effect on either Nox2 or Nox4 expression (Fig. 2A and E). Knockdown of either Nox2 or Nox4 protein by siRNA caused a significant ~50% and ~25% reduction in superoxide production, respectively, compared to the control siRNA (Fig. 2C, D, G and H).

3.2 NO inhibits NADPH oxidase-dependent superoxide production
Acute DETA-NONOate (100 µmol/L) exposure caused a significant ~50% reduction in chemiluminescence compared to control cells, most likely reflecting removal of superoxide by direct reaction with NO (Fig. 3A). Longer-term incubation with DETA-NONOate caused further reductions in superoxide levels with maximum reduction achieved after 6 h of treatment (Fig. 3A). Importantly, direct stoichiometric interaction between superoxide and NO could not account entirely for the time-dependent reduction of superoxide levels caused by DETA-NONOate, for a similar time-dependent reduction in superoxide was observed when DETA-NONOate was washed from the cells prior to assay (Fig. 3B).


Figure 3
View larger version (31K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3 Effect of NO donors on superoxide production measured by lucigenin (5 µmol/L) chemiluminescence and DHE fluorescence in HMEC-1. HMEC-1 were either untreated (Control) or pre-treated for 0.5 h to 24 h with DETA-NONOate (100 µmol/L) prior to assay for superoxide. In (A) DETA-NONOate was present during assay for superoxide whereas in (B) it was removed. (C) HMEC-1 were pre-treated for 6 h with varying concentrations of DETA-NONOate or (D) sodium nitroprusside (SNP) or S-nitroso-N-acetylpenicilamine (SNAP), prior to washout of the NO donor and assaying for superoxide. In (E) HMEC-1 were either untreated (Control) or treated with DETA-NONOate (100 µmol/L) for 6 h prior to washout of DETA-NONOate, and incubation for a further 0 to 24 h. Superoxide was then measured with lucigenin. In (F) HMEC-1 were untreated (Control) or treated for 6 h with either SNP (30, 100 µmol/L) or DETA-NONOate (100, 300 µmol/L), or for 1 h with either SOD (600 U/ml) or the superoxide scavenger, MnTMPyP (25 µmol/L), prior to assay with DHE fluorescence. Values (mean±S.E.M. from 5–6 experiments) are expressed as a percentage of control. *P<0.05 vs. control. #P<0.05 vs. acute treatment.

 
Treatment of HMEC-1 with DETA-NONOate (3–300 µmol/L) for 6 h followed by washout prior to assay resulted in a concentration-dependent reduction in superoxide production (Fig. 3C). Importantly, the amount of NO liberated by DETA-NONOate over this concentration range (Supplementary Fig. 2) is representative of the physiological range of NO concentrations observed in endothelial cells at rest (i.e. <50 nM) and following stimulation with endothelium-dependent relaxing agonists and statins (i.e. 250–1000 nM) [15].

Finally, SNP (1–100 µmol/L) and SNAP (1–100 µmol/L) reduced NADPH oxidase-stimulated superoxide production in HMEC-1 following a 6-h exposure period and washout prior to assay (Fig. 3D).

3.3 Sustained suppression of NADPH oxidase activity by NO
In this series of experiments, HMEC-1 were treated with DETA-NONOate (100 µmol/L) for 6 h and then assayed for superoxide 0 to 24 h later, following washout of DETA-NONOate. Superoxide production was attenuated 60–70% when measured immediately after washout of DETA-NONOate and this effect was sustained for at least 6 h (Fig. 3E). 24 h after removing DETA-NONOate from the solution, superoxide levels had returned to control levels (Fig. 3E).

To ensure that the above effects of NO donors on superoxide production were not merely an artefact of the ROS detection technique used, the effect of DETA-NONOate and SNP on superoxide production was also assessed by DHE fluorescence. Exposure of HMEC-1 to DHE for 30 min resulted in a significant increase in fluorescence. Pre-treatment of HMEC-1 with a cell membrane-permeable superoxide scavenger, MnTMPyP (25 µmol/L), for 30 min prior to DHE exposure nearly abolished the fluorescent signal, whereas exogenous native SOD (600 U/mL) had no effect (Fig. 3F). These findings confirm that DHE detects superoxide produced intracellularly [5]. A 6-h exposure of HMEC-1 to either DETA-NONOate (100–300 µmol/L) or SNP (30; 100 µmol/L), followed by washout of the NO donor for 30 min, caused a significant reduction in DHE fluorescence (Fig. 3F).

3.4 No role for guanylate cyclase or peroxynitrite
Neither the soluble GC inhibitor, ODQ, nor the selective PKG inhibitor, KT5823, had any effect on NADPH oxidase-dependent superoxide production in HMEC-1 (data not shown). Moreover, neither inhibitor reversed the attenuation by DETA-NONOate (100 µmol/L, 6 h) of NADPH oxidase-dependent superoxide production (see Supplementary Fig. 3). Uric acid also failed to reverse the inhibition of superoxide production by DETA-NONOate thereby excluding a role for peroxynitrite in attenuation of NADPH oxidase activity.

3.5 Effect of NO on expression and co-association of NADPH oxidase subunits
Chronic DETA-NONOate (100 µmol/L) exposure failed to affect protein expression of Nox2, Nox4, p47phox or the G-protein, Rac in HMEC-1 (see Supplementary Fig. 4) indicating that attenuation of NADPH oxidase activity was not due to down-regulation of expression of any of these subunits.

We next examined the effect of DETA-NONOate (100 µmol/L) on association of p47phox with p22phox as a measure of NADPH oxidase activity. Whole-cell protein was immuno-precipitated with p22phox antibody, separated by SDS–PAGE, and transferred to PVDF membranes. Immunoblotting of these membranes with p47phox antibody revealed a single protein band with a molecular weight of ~47 kDa (see Supplementary Fig. 5). Exposure of HMEC-1 to DETA-NONOate (100 µmol/L, 6 h) prior to homogenisation did not significantly affect the amount of p47phox associated with p22phox, nor did it affect the expression of p22phox (Supplementary Fig. 5). Parallel experiments were performed in which the p47phox antibody was used for immunoprecipitation, and a Nox2 antibody for immunoblotting and we similarly found no effect of DETA-NONOate on association of these subunits (data not shown).

3.6 S-Nitrosylation of p47phox
To assess for a possible role of S-nitrosylation, we examined the effects of exposure of HMEC-1 homogenates to UV light, a previously used method of disrupting S–NO bonds [16], on the ability of NO donors to inhibit superoxide production. UV light exposure did not affect superoxide production in untreated cells, but completely reversed the attenuation of superoxide production following 6 h of pre-treatment with SNAP (100 µM; Fig. 4A).


Figure 4
View larger version (32K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4 S-Nitrosylation of NADPH oxidase by NO donors in HMEC-1. (A) Effect of UV light (5 min) exposure of HMEC-1 homogenates on superoxide production measured by lucigenin chemiluminescence following 6 h treatment of cells with SNAP (100 µmol/L). (B) After immunoprecipitation with p22phox antibody, S-nitrosylation was detected by use of DAN fluorescence in untreated HMEC-1 (control) and HMEC-1 treated with either DETA-NONOate (100 µmol/L) for 24 h or SNAP (100 µmol/L) for 6 h. In (A) and (B), values are mean±S.E.M. from 3–4 experiments and are expressed as a percentage of untreated HMEC-1 (control). (C) Proteins immunoprecipitated with p47phox antibody were separated by electrophoresis and then immunoblotted, firstly with anti-nitrosocysteine antibody, and secondly with p47phox antibody. (D) Quantification of protein expression using densitometric analysis. Values (mean±S.E.M. from 8 experiments) indicate the ratio of the optical densities of S–NO to p47phox. *P<0.05 vs. control.

 
Both DETA-NONOate (100 µmol/L for 24-h exposure) and SNAP (100 µmol/L for 6-h exposure) caused a significant increase in DAN fluorescence above controls (Fig. 4B), indicating that p22phox and/or its associated proteins are S-nitrosylated by chronic exposure to NO.

Finally, to examine whether p47phox is S-nitrosylated by DETA-NONOate, we measured S-nitrosocysteine immunoreactivity in blots generated from protein that was immunoprecipitated with the p47phox antibody (Fig. 4D). The S-nitrosocysteine antibody revealed protein bands at ~47 kDa, the reactivity of which was significantly (~30%) higher in samples derived from cells that were exposed to DETA-NONOate (100 µM) for 6 h and 24 h (Fig. 4C and D).


   4. Discussion
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
Appendix A. Supplementary data
 Acknowledgments
 References
 
Here we present the first evidence that prolonged exposure of human endothelial cells to NO donors causes sustained suppression of superoxide production via inhibition of NADPH oxidase activity through S-nitrosylation of p47phox. The implication of these findings is that NO (derived from NO donor drugs and potentially endogenous enzymatic sources) regulates NADPH oxidase activity directly, and thereby might also modulate oxidative stress in vascular disease.

Over the past decade, substantial evidence has accumulated suggesting that some NO donors, and especially those belonging to the organic nitrate class (e.g. glyceryl trinitrate), may increase vascular superoxide production and thereby exacerbate oxidative stress in diseased blood vessels [17]. The mechanism(s) by which organic nitrates elevate vascular ROS levels are yet to be fully elucidated but appear to involve upregulation of Nox2 expression, activation of endothelial protein kinase C, and uncoupling of eNOS [17]. Hence, our finding that newer generation NO donors actually decrease vascular superoxide production, highlights the fact that this (and potentially other) adverse effect only applies to certain organic nitrates (and possibly SIN-1) and not to the NO donor class per se.

Using a combination of pharmacological inhibitors to exclude other sources of vascular ROS, and gene silencing approaches to knockdown expression of NADPH oxidase subunits, we confirmed that Nox2- and Nox4-containing NADPH oxidases are major sources of superoxide in endothelial cells under basal culture conditions [2–4]. It has previously been demonstrated that NO inhibits NADPH oxidase-dependent superoxide production in neutrophils [18–20]. We have now shown that NO donor drugs similarly suppress NADPH oxidase-dependent superoxide production in endothelial cells, with maximal inhibition occurring after 6 h of treatment. On one level, this reduction of superoxide levels by NO might be expected, given that NO reacts with superoxide in a near-diffusion limited manner to produce peroxynitrite. Indeed, when HMEC-1 were exposed to DETA-NONOate immediately before superoxide detection, significantly lower levels of superoxide were measured, presumably reflecting stoichiometric removal. However, two separate observations in the present study suggest that stoichiometric removal of superoxide by NO alone cannot explain the inhibitory effects of NO donors on endothelial ROS production. First, given that the reaction between NO and superoxide is near-diffusion limited, if stoichiometric removal of superoxide alone was responsible for the effects of DETA-NONOate on ROS levels, one would expect the time-course to maximum inhibition by DETA-NONOate, to reflect the time it takes for the NO concentration within the pre-incubation/assay solution to reach steady-state levels. Following dissolution of DETA-NONOate in physiological buffer at 37 °C, maximum concentrations of NO are achieved within 20 min [21] – a far shorter period than the 6-h time-course of inhibition by DETA-NONOate observed here. Second, we showed that a large component of the inhibitory effect of NO donors persisted even following their washout. In the case of DETA-NONOate, suppression of NADPH oxidase-dependent superoxide levels persisted for at least 6 h after the washout.

It was previously shown that chronic treatment with NO donors inhibits cytokine- and LPS-induced superoxide production and Nox2 protein expression in porcine cultured aortic endothelial cells [11]. Although the time-course of inhibition by NO donors on basal endothelial NADPH oxidase activity reported here (>6 h) is potentially consistent with an action at the level of gene expression, we were unable to demonstrate any effect of long-term DETA-NONOate on protein levels of Nox2, or indeed any other NADPH oxidase subunits (Nox4, Rac-1, p22phox or p47phox) in HMEC-1. This suggests that while NO has no effect on the signal transduction pathways responsible for maintenance of NADPH oxidase expression under basal conditions, it may block upstream signalling elements responsible for the upregulation of NADPH oxidase during pathophysiological conditions.

An alternative explanation for how the NO donors inhibited basal superoxide production in the present study is via post-translational mechanism(s). It is now established that NO can chemically modify certain functional groups on proteins and that this represents a major mechanism by which NO modulates cellular function. The best characterised of these interactions are (1) S-nitrosylation in which NO and/or higher nitrogen oxides (N2O3) nitrosate (–NO) specific reactive cysteine residues; and (2) tyrosine nitration, resulting from the conversion of NO into peroxynitrite and subsequent addition of a nitro group (–NO2) onto one of the ortho carbons on the aromatic ring of tyrosine residues [22,23]. While failure of the peroxynitrite scavenger, urate [24], in the present study appears to rule out a role for peroxynitrite-dependent nitration of critical tyrosine residues on any of the NADPH oxidase subunits, a number of lines of evidence support a role for S-nitrosylation:

1. Theoretical examination of the primary amino acid structure of the human p47phox subunit reveals that this subunit contains 4 cysteine residues, namely, C98, C111, C196 and C378. The latter three of these cysteine residues are conserved across species and one of these, C378, is nestled amongst an amino acid environment that is predicted to be favourable for S-nitrosylation to occur [22,23,25].
2. S-Nitrosylation-dependent protein modification by NO does not involve cGMP or the PKG pathway [22,23,25], which is consistent with our observations of a lack of effect of ODQ and KT5823 on suppression of NADPH oxidase-derived superoxide by NO.
3. In the present study, brief exposure of HMEC-1 to UV light, a commonly used procedure for disrupting S–NO bonds [16], reversed the NO-dependent suppression of superoxide production.
4. Using the DAN fluorescent assay for detection of S-nitrosylated proteins [22], we demonstrated that p22phox-associated proteins from NO treated HMEC-1 are S-nitrosylated to a greater degree than the untreated controls.
5. Finally, our data with a nitrosocysteine-specific antibody suggests that NO preferentially nitrosylates the p47phox subunit.

Given that p47phox is the major organizer protein for Nox2-containing oxidases, and that Nox4 activity is unlikely to be regulated by any of the currently identified organizer (i.e. p47phox and NoxO1) or activator (p67phox and NoxA1) subunits [1], our findings provide functional evidence for regulation of the constitutive endothelial Nox2-containing NADPH oxidase expressed in endothelial cells by S-nitrosylation.

It remains to be determined as to how S-nitrosylation of p47phox interferes with NADPH oxidase activity. It has previously been suggested that NO donors and related S-nitrosothiols inhibit the activity of the neutrophil NADPH oxidase complex by interacting directly with either the membrane or cytosolic components of the complex and preventing their subsequent association [18–20]. For example, Park demonstrated that when membrane fractions were pre-treated with S-nitrosothiol and then combined with a cytosolic fraction from resting neutrophils to reconstitute NADPH oxidase, both translocation of the cytosolic NADPH oxidase components (p47phox and p67phox) to the plasma membrane fraction and superoxide generation were inhibited [20]. However, such a mechanism is unlikely to explain the effects of the NO donors on resting endothelial NADPH oxidase activity described here. Whereas the neutrophil NADPH oxidase complex at rest is dissociated and generates very little superoxide, the endothelial isoforms of NADPH oxidase exist as constitutively active, pre-assembled complexes containing either Nox2 [4] or Nox4 [2] as the catalytic domain. Thus, NO generated by NO donors would not have the opportunity to prevent assembly of these pre-assembled endothelial NADPH oxidase complexes. Moreover, we provide evidence that the NO donors did not cause disassembly of the endothelial NADPH oxidase complexes by the demonstration that chronic incubation of HMEC-1 with DETA-NONOate failed to disrupt the association of p47phox to either p22phox or Nox2.

Interestingly, C378 is positioned in close proximity to one of the major phosphorylation sites on p47phox, namely Ser379, raising the possibility of significant cross-talk between these two sites. Yet while phosphorylation of p47phox appears to decrease the sensitivity of C378 to alkylation by N-ethylmaleimide [26], the converse was not true, i.e. N-ethylmaleimide treatment did not prevent phosphorylation of p47phox [27]. Whether, S-nitrosylation (as opposed to alkylation) of C378 inhibits p47phox phosphorylation remains to be determined.

In conclusion, we have demonstrated that the NO donor, DETA-NONOate, at concentrations that generate physiological levels of NO, suppresses basal NADPH oxidase-dependent superoxide production in human endothelial cells. Given that NADPH oxidases are likely to be the major source of vascular ROS during normal physiology, our findings may suggest a role for endogenous NO in regulation of basal superoxide production. Moreover, it has recently been demonstrated that upregulation of certain isoforms of NADPH oxidase in the early stages of vascular disease gives rise to ‘kindling’ radicals, that ultimately cause depletion of tetrahydrobiopterin and uncoupling of eNOS, as well as oxidative conversion of xanthine dehydrogenase into xanthine oxidase, turning these enzymes into additional sources of ROS [28]. One would predict that suppression of NADPH oxidase activity by NO donors will therefore not only reduce ROS production by this enzyme in vascular disease, but also by these other, normally dormant, oxidases. Hence our findings, together with previous reports [11–13,29], highlight the potential of newer generation NO donors (i.e. those not belonging to the organic nitrate class) as treatments, not only for the symptoms of vascular disease (e.g. angina), but also of one of the major underlying causes of it, namely oxidative stress.

Time for primary review 20 days


   Appendix A. Supplementary data
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
 Appendix A. Supplementary data
Acknowledgments
 References
 
Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.cardiores.2007.03.030.


   Acknowledgments
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
 Appendix A. Supplementary data
 Acknowledgments
References
 
Supported by the National Health and Medical Research Council (NHMRC) of Australia. Fellowships from the NHMRC supported SS (232324), GJD (400303), HP (327401) and GRD (465109). GRD was also supported by a Monash University Category I Fellowship.


   References
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
 Appendix A. Supplementary data
 Acknowledgments
 References
 

  1. Bedard K., Krause K.H. The NOX family of ROS-generating NADPH oxidases: physiology and pathophysiology. Physiol Rev (2007) 87:245–313.[Abstract/Free Full Text]
  2. Ago T., Kitazono T., Ooboshi H., Iyama T., Han Y.H., Takada J., et al. Nox4 as the major catalytic component of an endothelial NAD(P)H oxidase. Circulation (2004) 109:227–233.[Abstract/Free Full Text]
  3. Li J.M., Shah A.M. Intracellular localization and preassembly of the NADPH oxidase complex in cultured endothelial cells. J Biol Chem (2002) 277:19952–19960.[Abstract/Free Full Text]
  4. Ray R., Shah A.M. NADPH oxidases and endothelial cell function. Clin Sci (2005) 109:217–226.[CrossRef][ISI][Medline]
  5. Peshavariya HM, Dusting GJ, Selemidis S. Analysis of dihydroethidium fluorescence for the detection of intracellular and extracellular superoxide produced by NADPH oxidase. Free Radic Res, [Electronic publication ahead of print]. http://dx.doi.org/10.1080/10715760701297354.
  6. Paravicini T.M., Gulluyan L.M., Dusting G.J., Drummond G.R. Increased NADPH oxidase activity, gp91phox expression, and endothelium-dependent vasorelaxation during neointima formation in rabbits. Circ Res (2002) 91:54–61.[Abstract/Free Full Text]
  7. Warnholtz A., Nickenig G., Schulz E., Macharzina R., Brasen J.H., Skatchkov M., et al. Increased NADH-oxidase mediated superoxide production in the early stages of atherosclerosis: evidence for involvement of the renin–angiotensin system. Circulation (1999) 99:2027–2033.[Abstract/Free Full Text]
  8. Guzik T.J., Sadowski J., Guzik B., Jopek A., Kapelak B., Przybylowski P., et al. Coronary artery superoxide production and Nox isoform expression in human coronary artery disease. Arterioscler Thromb Vasc Biol (2006) 26:333–339.[Abstract/Free Full Text]
  9. Barry-Lane P.A., Patterson C., van der Merwe M., Hu Z., Holland S.M., Yeh E.T., et al. p47phox is required for atherosclerotic lesion progression in ApoE(–/–) mice. J Clin Invest (2001) 108:1513–1522.[CrossRef][ISI][Medline]
  10. Herman A.G., Moncada S. Therapeutic potential of nitric oxide donors in the prevention and treatment of atherosclerosis. Eur Heart J (2005) 26:1945–1955.[Abstract/Free Full Text]
  11. Muzaffar S., Shukla N., Angelini G., Jeremy J.Y. Nitroaspirins and morpholinosydnonimine but not aspirin inhibit the formation of superoxide and the expression of gp91phox induced by endotoxin and cytokines in pig pulmonary artery vascular smooth muscle cells and endothelial cells. Circulation (2004) 110:1140–1147.[Abstract/Free Full Text]
  12. Kojda G.H.A., Noack E. Effects of nonintermittent treatment of rabbits with pentaerythritol tetranitrate on vascular reactivity and superoxide production. Eur J Pharmacol (1998) 355:23–31.[CrossRef][ISI][Medline]
  13. Müller S., König I., Meyer W., Kojda G. Inhibition of vascular oxidative stress in hypercholesterolemia by eccentric isosorbide mononitrate. J Am Coll Cardiol (2004) 44:624–631.[Abstract/Free Full Text]
  14. Dikalov S., Griendling K.K., Harrison D.G. Measurement of reactive oxygen species in cardiovascular studies. Hypertension (2007) 49:717–727.[Free Full Text]
  15. Kalinowski L., Dobrucki L.W., Brovkovych V., Malinski T. Increased nitric oxide bioavailability in endothelial cells contributes to the pleiotropic effect of cerivastatin. Circulation (2002) 105:933–938.[Abstract/Free Full Text]
  16. Beltran B.O.A., Clementi E., Moncada S. Oxidative stress and S-nitrosylation of proteins in cells. Br J Pharmacol (2000) 129:953–960.[CrossRef][ISI][Medline]
  17. Munzel T., Daiber A., Mulsch A. Explaining the phenomenon of nitrate tolerance. Circ Res (2005) 97:618–628.[Abstract/Free Full Text]
  18. Fujii H.I.K., Hoshiai K., Nakazawa H. Nitric oxide inactivates NADPH oxidase in pig neutrophils by inhibiting its assembling process. J Biol Chem (1997) 272:32773–32778.[Abstract/Free Full Text]
  19. Clancy R.M., Leszczynska-Piziak J., Abramson S.B. Nitric oxide, an endothelial cell relaxation factor, inhibits neutrophil superoxide anion production via a direct action on the NADPH oxidase. J Clin Invest (1992) 90:1116–1121.[ISI][Medline]
  20. Park J.W. Attenuation of p47phox and p67phox membrane translocation as the inhibitory mechanism of S-nitrosothiol on the respiratory burst oxidase in human neutrophils. Biochem Biophys Res Commun (1996) 220:31–35.[CrossRef][ISI][Medline]
  21. Bebok Z., Varga K., Hicks J.K., Venglarik C.J., Kovacs T., Chen L., et al. Reactive oxygen nitrogen species decrease cystic fibrosis transmembrane conductance regulator expression and cAMP-mediated Cl-secretion in airway epithelia. J Biol Chem (2002) 277:43041–43049.[Abstract/Free Full Text]
  22. Ascenzi P., Colasanti M., Persichini T., Muolo M., Polticelli F., Venturini G., et al. Re-evaluation of amino acid sequence and structural consensus rules for cysteine–nitric oxide reactivity. Biol Chem (2000) 381:623–627.[CrossRef][ISI][Medline]
  23. Gow A.J., Farkouh C.R., Munson D.A., Posencheg M.A., Ischiropoulos H. Biological significance of nitric oxide-mediated protein modifications. Am J Physiol Lung Cell Mol Physiol (2004) 287:L262–L268.[Abstract/Free Full Text]
  24. Kuzkaya N., Weissmann N., Harrison D.G., Dikalov S. Interactions of peroxynitrite with uric acid in the presence of ascorbate and thiols: implications for uncoupling endothelial nitric oxide synthase. Biochem Pharmacol (2005) 70:343–354.[CrossRef][ISI][Medline]
  25. Hess D.T., Matsumoto A., Kim S.O., Marshall H.E., Stamler J.S. Protein S-nitrosylation: purview and parameters. Nat Rev Mol Cell Biol (2005) 6:150–166.[CrossRef][ISI][Medline]
  26. Park J.W., Babior B.M. Activation of the leukocyte NADPH oxidase subunit p47phox by protein kinase C. A phosphorylation-dependent change in the conformation of the C-terminal end of p47phox. Biochemistry (1997) 36:7474–7480.[CrossRef][ISI][Medline]
  27. Park J.W., Park H.S., Lee S.M. Possible target components for the inhibitory effect of N-ethylmaleimide on the activation of neutrophil NADPH oxidase. Biochem Mol Biol Int (1998) 45:699–707.[ISI][Medline]
  28. Mueller C.F.H., Laude K., McNally J.S., Harrison D.G. Redox mechanisms in blood vessels. Arterioscler Thromb Vasc Biol (2005) 25:274–278.[Abstract/Free Full Text]
  29. Jiang F., Roberts S.J., Datla S.R., Dusting G.J. Nitric oxide modulates NADPH oxidase function via heme oxygenase-1 in human endothelial cells. Hypertension (2006) 48:950–957.[Abstract/Free Full Text]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 IOVSHome page
Y. Saito, A. Uppal, G. Byfield, S. Budd, and M. E. Hartnett
Activated NAD(P)H Oxidase from Supplemental Oxygen Induces Neovascularization Independent of VEGF in Retinopathy of Prematurity Model
Invest. Ophthalmol. Vis. Sci., April 1, 2008; 49(4): 1591 - 1598.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary Data
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Selemidis, S.
Right arrow Articles by Drummond, G. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Selemidis, S.
Right arrow Articles by Drummond, G. R.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?