Hydrogen peroxide restrains endothelium-derived nitric oxide bioactivity—Role for iron-dependent oxidative stress

Free Radical Biology & Medicine 41 (2006) 681 – 688 www.elsevier.com/locate/freeradbiomed

Original Contribution

Hydrogen peroxide restrains endothelium-derived nitric oxide bioactivity— Role for iron-dependent oxidative stress Shane R. Thomas ⁎, Eberhard Schulz, John F. Keaney Jr. Evans Memorial Department of Medicine and Whitaker Cardiovascular Institute, Boston University School of Medicine, Boston, Massachusetts 02118, USA Received 23 January 2006; revised 12 April 2006; accepted 10 May 2006 Available online 19 May 2006

Abstract Vascular diseases are characterized by impairment of endothelial-derived nitric oxide (NO) bioactivity and increased vascular levels of hydrogen peroxide (H2O2). Here we examined the implications of H2O2 for agonist-stimulated endothelial NO bioactivity in rabbit aortic rings and cultured porcine aortic endothelial cells (PAEC). Vessels pre-treated with H2O2 exhibited impaired endothelial-dependent relaxation induced by acetylcholine or calcium ionophore. In contrast, H2O2 had no effect on endothelium-independent relaxation induced by a NO donor, indicating a defect in endothelium-derived NO. This defect was not related to eNOS catalytic activity; treatment of PAEC with H2O2 enhanced agonist-stimulated eNOS activity indicated by increased eNOS phosphorylation at Ser-1177 and de-phosphorylation at Thr-495 and enhanced conversion of [3H]-L-arginine to [3H]-L-citrulline that was prevented by inhibitors of Src and phosphatidylinositol-3 kinases. Despite activating eNOS, H2O2 impaired endothelial NO bioactivity indicated by attenuation of the increase in intracellular cGMP in PAEC stimulated with calcium ionophore or NO. The decrease in cGMP was not due to impaired guanylyl cyclase as H2O2 treatment increased cGMP accumulation in response to BAY 41-2272, a NO-independent activator of soluble guanylyl cyclase. At concentrations that impaired endothelial NO bioactivity H2O2 increased intracellular oxidative stress and size of the labile iron pool in PAEC. The increase in oxidative stress was prevented by the free radical scavenger's tempol or tiron and the iron chelator desferrioxamine and these antioxidants reversed the H2O2-induced impairment of NO bioactivity in PAEC. This study shows that despite promoting eNOS activity, H2O2 impairs endothelial NO bioactivity by promoting oxidative inactivation of synthesized NO. The study highlights another way in which oxidative stress may impair NO bioactivity during vascular disease. © 2006 Elsevier Inc. All rights reserved. Keywords: Endothelial dysfunction; Reactive oxygen species; Atherosclerosis; Guanylyl cyclase; Antioxidant; Desferrioxamine; Iron chelator

Introduction Nitric oxide (NO) derived from endothelial nitric oxide synthase (eNOS) represents an important component of vascular homeostasis. Endothelial-derived NO bioactivity is implicated in the regulation of vascular tone [1], arterial pressure [2], platelet aggregation [3], and vascular smooth muscle cell proliferation [4]. Vascular diseases including hypertension, Abbreviations: A23187, calcium ionophore; ACH, acetylcholine; ANP, atrial naturetic peptide; ANOVA, analysis of variance; eNOS, endothelial nitric oxide synthase; DEANO, diethylamine NONOate; DFO, desferrioxamine; DHR, dihydrorhodamine; FBS, fetal bovine serum; H2O2, hydrogen peroxide; PAEC, porcine aortic endothelial cells; PSS, physiologic salt solution; LNAME, L-nitro-arginine methyl ester; ROS, reactive oxygen species. ⁎ Corresponding author. Centre for Vascular Research, University of New South Wales, Sydney, NSW 2052, Australia. E-mail address: [email protected] (S.R. Thomas). 0891-5849/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.freeradbiomed.2006.05.012

diabetes and atherosclerosis are characterized by impaired endothelial-derived NO bioactivity and this may contribute to the clinical events associated with these diseases [5]. Considerable evidence indicates that impaired endothelialderived NO bioactivity during vascular diseases is due, in part, to excess vascular oxidative stress [6]. Diseased blood vessels produce increased amounts of reactive oxygen species, derived primarily from endothelial and smooth muscle cells and detected principally as superoxide anion radical and its dismutation product hydrogen peroxide (H2O2) [7,8]. Superoxide combines readily with NO (k = 1.9 × 1010 M−1 sec−1) [9] to limit endothelial-derived NO bioactivity [10,11]. In contrast, H2O2 does not react with NO and has garnered less attention with regard to modulation of endothelial-derived NO bioactivity. However, recent studies indicate that H2O2 affects eNOS activity and NO production. For example, under basal conditions H2O2 treatment of aortic endothelial cells acutely

682

S.R. Thomas et al. / Free Radical Biology & Medicine 41 (2006) 681–688

promotes eNOS activity and NO production by inducing changes in eNOS phosphorylation [12] while prolonged H2O2 treatment produces increased eNOS expression [13]. Vascular diseases are characterized by impaired endothelial-dependent relaxation induced by physiological agonists. Therefore, in the current study, we sought to determine the implications of H2O2 for eNOS activity and endothelial-derived NO bioactivity in response to endothelial-dependant agonists. Materials and methods Materials Tempol and BAY 41-2272 were from Alexis Biochemicals. LY294002 and PP2 were from Calbiochem. Polyclonal antibodies against eNOS phosphorylated at Ser-1177 (p-eNOSSer-1177 ) and Thr-495 (p-eNOSThr-495) were from Cell Signaling Technology and Upstate Biotechnology, respectively. Anti-eNOS monoclonal antibody was from Transduction Laboratories. [3H]-L-Arginine (1 mCi, 53 mCi/ mmol) was from New England Nuclear. Diethylamine NONOate (DEANO) and cGMP assay kits were from Cayman. Dihydrorhodamine123, Phen green SK and C11bodipy581/591 were from Molecular Probes. All other reagents were from Sigma.

treatments as outlined in the text. After treatment, C11bodipy581/591 green fluorescence was examined by fluorescent microscopy and images captured using Open Lab software. Intracellular labile iron pool assay The intracellular chelatable or labile iron pool was determined using the fluorescent probe, Phen Green SK [18,19]. PAEC were loaded with 20 μM Phen Green SK diacetate for 30 min in PSS. The cells were then washed to remove excess dye and treated with H2O2 (100 μM) for 20 min prior to washing and sequential addition of up to 2 mM of the cell permeable iron chelator 1,10 phenanthroline for 20 min. After washing the intensity of green fluorescence was examined by fluorescent microscopy and captured using Open Lab software. As phen green SK fluoresces when iron is not bound, the extent of increase in green fluorescence after treatment of cells with 1,10 phenanthroline is proportional to the size of the labile iron pool [18]. Immunoprecipitation and Western blotting Immunoprecipitation and Western blotting of eNOS and its phosphorylation status were performed as previously described [12]. Aortic ring studies

Endothelial cell culture Primary porcine aortic endothelial cells (PAEC) were isolated and cultured as described [12,14] and used for experiments between passages 3–9 and at 1 day post-confluence. For experiments, PAEC were incubated in HEPES-buffered physiologic salt solution (PSS) [12,14]. Assay of eNOS activity and endothelial NO bioactivity Determination of eNOS activity in PAEC was assessed as the conversion of [3H]-L-arginine to [3H]-L-citrulline that was sensitive to L-nitro-arginine methyl ester (L-NAME) [12,15]. Endothelial-derived NO bioactivity was monitored in PAEC by measuring the intracellular accumulation of cGMP [12,15]. Intracellular oxidative stress Intracellular oxidative stress was assessed by measuring dihydrorhodamine123 (DHR) fluorescence. PAEC were loaded for 30 minutes with 10 μM DHR in PSS, washed and then treated as outlined. After treatment, PAEC were washed and removed with a cell scraper in ice-cold PBS. Equal numbers of cells were added to 96 well plates and DHR fluorescence assessed in a fluorescent plate reader (excitation 480 nm, emission 535 nm). Intracellular lipid oxidation was assessed using C11-bodipy581/591, an oxidizable fluorescent lipid probe [16,17]. The C11-bodipy581/591 in its oxidized state emits a green fluorescence that absorbs at 520 nm. PAEC were loaded with C11-bodipy581/591 in M199 containing 1% FBS for 30 minutes. PAEC were washed and incubated in PSS before

Animal studies were approved by the Boston University Medical Center Institutional Animal Care and Use Committee. The thoracic aorta was isolated from New Zealand White rabbits (Pine Acres, VT) sacrificed with pentobarbital (120 mg/kg) via a marginal ear vein. Vessel segments were prepared and suspended in organ chambers as described [20]. After equilibration for 90 minutes, vessels were treated with the indicated dose of H2O2 for 20 min prior to contraction with phenylephrine (1 μM) and assessment of vascular relaxation in response to increasing doses (1 nM–10 μM) of acetylcholine, A23187 or DEANO. Statistics Numerical data are presented as means ± S.E. For parametric data, comparisons among treatment groups were performed with one-way analysis of variance (ANOVA) and appropriate posthoc comparison. Non-parametric data were compared using Kruskal-Wallis one-way ANOVA on ranks with a post hoc Dunnet's comparison. Instances involving only two comparisons were evaluated with a Student's t-test or Mann-Whitney rank sum test as appropriate. Statistical significance was accepted if the null hypothesis was rejected with a P < 0.05. Results H2O2 impairs endothelium-dependent, agonist-stimulated relaxation of aortic rings In accordance with previous studies [12,21], treatment of preconstricted vessel segments with 200 μM H2O2 promoted

S.R. Thomas et al. / Free Radical Biology & Medicine 41 (2006) 681–688

relaxation that was significantly inhibited by L-NAME (Fig. 1A). In contrast, arterial segments pre-treated with H2O2 (50– 200 μM) exhibited impaired endothelium-dependent relaxation in response to acetylcholine (Fig. 1B) or calcium ionophore, A23187 (Fig. 1C). There was no significant attenuation of endothelium-independent relaxation in response to exogenous NO generated by DEANO (Fig. 1D). These results indicate that H2O2 pre-treatment impairs agonist-stimulated, endotheliumdependent but not endothelium-independent vessel relaxation. H2O2 promotes eNOS activity in aortic endothelial cells Since impaired endothelium-derived NO bioactivity may reflect reduced production of NO, we next examined the effect of H2O2 on agonist-stimulated eNOS activity in cultured endothelial cells. Consistent with our previous study [12], H2O2 dose-dependently increased basal eNOS activity as determined by the conversion of [3H]-L-arginine to [3H]-L-citrulline (Fig. 2A). Endothelial cells treated with H2O2 also demonstrated enhanced A23187-induced eNOS activity (Fig. 2A). The activity of eNOS is subject to control through changes in the phosphorylation state of certain amino

683

acid residues including Ser-1177 or Thr-495, the phosphorylation of which act to stimulate or inhibit enzyme activity, respectively [22]. Consistent with increased eNOS catalytic activity, PAEC treatment with H2O2 and A23187 together increased eNOS Ser-1177 phosphorylation and Thr-495 dephosphorylation to a greater extent than treatment with either agent alone (Fig. 2C). The ability of H2O2 to enhance A23187-induced eNOS activity was attenuated by PP2 or LY294002, inhibitors of Src or phosphatidylinositol-3 kinases, respectively (Fig. 2B). PP2 and LY294002 did not affect eNOS activity stimulated with A23187 alone (Fig. 2B). These data indicate that H2O2 promotes calcium ionophore-stimulated eNOS catalytic activity through a Src and phosphatidylinositol-3 kinase-dependent mechanism. H2O2 attenuates NO bioactivity in PAECs To determine the implications of H2O2 for endothelial cell NO bioactivity, we measured the intracellular accumulation of cGMP in PAEC as a function of H2O2 and A23187. Treatment with H2O2 alone increased PAEC basal cGMP content (Fig. 3) that is inhibited by L-NAME [12].

Fig. 1. Hydrogen Peroxide inhibits endothelium-derived NO bioactivity. Rabbit thoracic aortic segments were suspended in organ chambers. (A) Vessels were contracted with 1 μM phenylephrine in the presence or absence of 300 μM L-NAME and relaxation assessed in response to 200 μM H2O2. Data represent the mean ± S.E. of 9 independent experiments. (B–D) Vessels were incubated in the absence ( ) or presence of 50 (▾), 100 (▴) or 200 μM (●) H2O2 for 20 minutes prior to being contracted with phenylephrine. The extent of vessel relaxation was then assessed in response to (B) acetylcholine, (C) A23187 or (D) DEANO. Data represent the mean ± S.E. of 6–8 independent experiments; *P < 0.05 vs no H2O2 by two-way ANOVA and post hoc Dunnett's Test.

▪

684

S.R. Thomas et al. / Free Radical Biology & Medicine 41 (2006) 681–688

cyclase isoforms, respectively. We found that H2O2 pretreatment had either no affect or enhanced the response to atrial naturetic peptide (Fig. 4A) or BAY 41-2272 (Fig. 4B), respectively indicating that H2O2 impairs NO bioactivity in endothelial cells without inhibition of particulate or soluble guanylyl cyclase. H2O2 promotes iron-dependent oxidative stress in PAEC H2O2 could limit endothelial NO bioactivity indirectly by promoting secondary oxidative reactions that inactivate NO. Consistent with this proposal we found that H 2O2, at concentrations that impaired NO bioactivity, promoted intracellular oxidative stress in PAEC as indicated by the increase in dihydrorhodamine (DHR) fluorescence (Fig. 5A) or cellular lipid peroxidation, assessed by measuring C11-bodipy581/591 fluorescence (Fig. 5B) [16]. The H2O2-mediated increase in these indices of intracellular oxidative stress was inhibited by the free radical scavenger's tempol or tiron and the iron chelator desferrioxamine (DFO) (Fig. 5). Consistent with a role of intracellular iron for H2O2-induced oxidative stress we found that the oxidant increased the size of the intracellular labile iron pool as assessed by the Phen Green SK fluorescence assay [18]. Thus, a greater increase in Phen green SK fluorescence upon

Fig. 2. Hydrogen Peroxide promotes eNOS catalytic activity in aortic endothelial cells. (A) PAEC were treated with the indicated concentrations of H2O2 for 15 min prior to stimulation with A23187 (1 μM) or vehicle. (B) PAEC were incubated with 10 μM LY294002 or 20 μM PP2 for 30 min prior to treatment with 100 μM H2O2 for a further 20 min. Cells were then stimulated with A23187 (1 μM) for 15 min and then processed for determination of [3H]-Larginine to [3H]-L-citrulline conversion. Data are presented as the mean ± S.E. of 4–6 independent experiments; *P < 0.05 vs no H2O2. (C) PAEC were treated with 100 μM H2O2 for 15 min prior to treatment with A23187 (1 μM) for a further 15 min. After treatment the level of eNOS phosphorylated at Ser-1177 or Thr-495 or total eNOS was assessed by immunoblotting. The Western blot shown is representative of 4 independent experiments.

In contrast, H2O2 pre-treatment attenuated the increase in intracellular cGMP in response to A23187 (Fig. 3A) or exogenous NO produced by DEANO (Fig. 3B), indicating the H2O2 impaired NO bioactivity in PAEC. To determine if the inhibitory effect of H2O2 on NO bioactivity in PAEC reflected diminished guanylyl cyclase activity, we next tested the effect of H2O2 on cGMP increases in response to atrial naturetic peptide or BAY 41-2272, NOindependent activators of particulate or soluble guanylyl

Fig. 3. H2O2 impairs NO bioactivity in aortic endothelial cells. PAEC were exposed to the indicated concentrations of H2O2 for 15 min prior to treatment with (A) A23187 or (B) DEANO and determination of intracellular cGMP content. Data is presented as mean ± S.E. of 4–9 independent experiments. *P < 0.05 vs no H2O2, †P < 0.05 vs A23187 alone, ‡P < 0.05 vs DEANO alone.

S.R. Thomas et al. / Free Radical Biology & Medicine 41 (2006) 681–688

685

Discussion Diseased blood vessels produce increased levels of H2O2, an oxidant that has important implications for vascular disease [8]. For example, over-expression of the H2O2-detoxifying enzyme catalase in apolipoprotein E gene knockout mice attenuates the extent of atherosclerosis indicative of a pro-atherogenic role of the oxidant [23]. Also, mice over-expressing catalase in the vascular wall exhibit decreased blood pressure indicative of a vasoconstrictive role of H2O2 [24]. Vascular diseases are characterized by impaired endothelial-derived NO bioactivity

Fig. 4. Effect of H2O2 on guanylyl cyclase activity in PAEC. PAEC were exposed to the indicated concentrations of H2O2 for 15 min followed by treatment with (A) 100 nM ANP or (B) 500 nM BAY 41-2272 for a further 3 min after which the intracellular cGMP content was determined. Data is presented as mean ± S.E. of 4–5 independent experiments.

treatment with the cell permeable iron chelator 1, 10phenanthroline was noted in H2O2 treated compared to nontreated PAEC (Fig. 5C). Antioxidants reverse H2O2-induced impairment of NO bioactivity in PAEC We next investigated the effects of tempol, tiron and DFO on NO bioactivity as a function of H2O2 treatment in PAEC. Tempol and tiron significantly reversed the H2O2-induced impairment of cGMP accumulation in response to either A23187 or DEANO (Figs. 6A and B). With DFO, we found that overnight pre-treatment with the iron chelator potentiated the increase in cGMP in response to DEANO alone (Fig. 7A), indicative of the presence of an intracellular iron pool capable of interfering with basal endothelial NO bioactivity. Importantly, overnight DFO pre-treatment also significantly reversed the impairment of NO bioactivity in PAEC pre-treated with H2O2 (Fig. 7B). In contrast to overnight DFO incubations required for the iron chelator to enter the cell, short-term DFO incubation (<1 h) or treatment with the non-cell permeable transition metal chelator DTPA had no effect on NO bioactivity in PAEC in the absence or presence of H2O2 (data not shown), indicating that intracellular, but not extra-cellular iron is important for the actions of DFO.

Fig. 5. H2O2-induced oxidative stress in PAEC and its inhibition by antioxidants. (A) DHR or (B) C11-bodipy loaded PAEC pre-treated with 500 μM tempol, 5 mM tiron or 200 μM DFO were exposed to 100 μM H2O2 for 20 min, followed by measurement of DHR or green C11-bodipy fluorescence. Data is presented as (A) mean ± S.E. of 4 independent experiments or (B) a representative image of 4 independent experiments. (C) PAEC were loaded with 20 μM Phen Green SK for 30 min prior to exposure to 100 μM H2O2 for 20 min, followed by addition of 2 mM 1,10-phenanthroline. After 30 min the intensity of green fluorescence in (I) control or (II) H2O2 treated cells was compared to that exhibited after 1, 10phenanthroline treatment of (III) control or (IV) H2O2 treated cells. Images shown are representative of 4 independent experiments.

686

S.R. Thomas et al. / Free Radical Biology & Medicine 41 (2006) 681–688

tion during vascular disease by catalyzing secondary oxidative reactions capable of scavenging synthesized NO. Our findings that H2O2 impairs agonist-stimulated increases of cGMP in endothelial cells are consistent with a recent study by Jaimes et al. [25]. The current study provides new insights into how the oxidant impairs endothelial NO bioactivity by demonstrating that it is not due to decreased eNOS or guanylyl cyclase activities but the result of secondary oxidative reactions involving iron and/or the generation of free-radicals capable of inactivating already synthesized NO. Peroxide is a versatile oxidant species capable of participating in a variety of redox reactions. As such, while the current study indicates that H2O2 impairs agonist-stimulated endothelial-derived NO bioactivity through catalyzing oxidative reactions involving labile iron and generation of free radicals the precise nature of the oxidant reactions responsible is

Fig. 6. Antioxidants reverse H2O2-induced impairment of NO bioactivity in PAEC. PAEC were incubated in the absence or presence of (A) tempol (500 μM) or (B) tiron (5 mM). Cells were then exposed to the 75 μM H2O2 for 15 min prior to stimulation with 1 μM A23187 or 10 nM DEANO and determination of intracellular cGMP. Data for H2O2 treated cells is expressed as a percentage of the cGMP increase afforded by treatment with A23187 or DEANO alone or A23187 or DEANO in the presence of the respective antioxidant (% control). Data represents the mean ± S.E. of 4 independent experiments.

in response to endothelial-dependent agonists. The current study examined the implications of H2O2 for endothelial-derived NO bioactivity under agonist-stimulated conditions. We found that H2O2 impairs agonist-stimulated endothelial-derived NO bioactivity by promoting oxidative inactivation of synthesized NO as indicated by the following: (i) H2O2 pre-treatment attenuated endothelium-dependent, but not endothelial-independent, relaxation of aortic rings indicating that the oxidant induced a defect specific for the endothelium, (ii) H2O2 decreased endothelial NO bioactivity in cultured PAEC as indicated by an attenuated increase in the intracellular content of cGMP in response to the eNOS-agonist A23187 or authentic NO, and (iii) the H2O2induced impairment of endothelial NO bioactivity and increase in intracellular oxidative stress were both significantly reversed by the free radical scavengers tempol or tiron and the ironchelator, DFO. This study therefore supports the contention that elevated vascular H2O2 could contribute to endothelial dysfunc-

Fig. 7. The iron chelator desferrioxamine improves NO bioactivity in PAEC. PAEC were pre-incubated overnight with 200 μM DFO in complete culture medium. PAEC were then (A) non-treated or (B) treated with 75 μM H2O2 for 15 min prior to assessment of the increase in the intracellular cGMP content in response to 10 nM DEANO. Data are expressed as (A) a percentage of the increase in cGMP afforded by DEANO alone or (B) as a percentage of the cGMP increase afforded by DEANO alone or DEANO in the presence of DFO (% control) in H2O2 treated cells. Data represents the mean ± S.E. of 4 independent experiments.

S.R. Thomas et al. / Free Radical Biology & Medicine 41 (2006) 681–688

potentially complex and requires further investigation. Candidate free radicals include, superoxide anion or lipid peroxyl radicals, reactive species capable of scavenging NO to limit its bioactivity [9,26]. A role for superoxide anion is supported by the actions of tempol and tiron, agents that act as SOD mimetics. A plausible role for lipid peroxyl radicals stems from our data showing that H2O2 at concentrations that impair NO bioactivity also induced lipid oxidation that was inhibited by tempol, tiron and DFO. Thus, the pleotropic redox reactions of both peroxide and iron dictate that further investigation will be needed to precisely define the reactive species involved. It is becoming increasingly apparent that intracellular irondependent signaling and oxidative reactions are important for endothelial function and phenotype. For example, intracellular iron is important for the increased expression of endothelial adhesion molecules in response to pro-inflammatory cytokines [27] and intracellular oxidative stress induced by H2O2 in endothelial cells [28]. The current study also highlights the importance of intracellular iron for the increase in endothelial oxidative stress induced by H2O2 and that this is accompanied by a rapid increase in the intracellular levels of labile iron. The mechanism(s) by which H2O2 increases the labile iron pool in endothelial cells is currently unknown. Consistent with our endothelial cell studies, a recent report using a T lymphocyte cell line showed that H2O2 treatment of these cells also induced a rapid increase in the labile iron pool that was important for intracellular oxidative reactions [29]. This study further proposed that the increase in labile iron is the result of transient lysosomal rupture and the resultant release of redox-active iron [29]. Alternatively, increased labile iron pool could involve a perturbation of mitochondrial function as has been proposed for endothelial cells chronically exposed to NO [19]. Further studies are necessary to elucidate the mechanism(s) responsible for the increase in labile iron pool in endothelial cells exposed to H2O2. Together, these studies indicate that increases in the size of the labile iron pool can represent an important response of cells to oxidative and NO-mediated stress. The current study identifies endothelial intracellular iron status as an important determinant of endothelial NO bioactivity. Thus, chelation of intracellular iron not only improved endothelial NO bioactivity in PAEC under basal conditions but also significantly reversed the impairment in NO bioactivity and the increase in intracellular oxidative stress induced by H2O2. These data are consistent with observations that vascular iron is important for endothelial dysfunction in cardiovascular disease patients. Duffy and colleagues reported intra-arterial infusion of DFO improved endothelium-dependent, but not endothelialindependent, vasodilatation in human patients with coronary artery disease [30]. These clinical data support a recent study using non-invasive, electron paramagnetic resonance spectroscopy that reported a significantly increased content of lowmolecular weight, labile iron complexes in human atherosclerotic arteries [31]. In light of these clinical studies, our data suggest that the increase in vascular H2O2 content during atherosclerosis may, in part, be responsible for increased vascular labile iron that in turn has important implications for endothelialderived NO bioactivity and hence endothelial dysfunction.

687

Wolin and Burke indicated that H2O2 can induce endothelialindependent arterial relaxations by stimulating guanylyl cyclase activity in vascular smooth muscle cells and the resultant increase in intracellular cGMP [32]. Our current and previous [12] study indicates that H2O2 treatment of endothelial cells also increases basal cGMP. However, in contrast to smooth muscle cells the increase in endothelial cell cGMP content is not due to direct activation of guanylyl cyclase but due to eNOS-derived NO [12]. The current study also indicates that H2O2 pretreatment of endothelial cells primes soluble guanylyl cyclase for activation by BAY 41-2272, a NO-independent activator of the enzyme. The mechanism by which H2O2 primes soluble guanylyl cyclase in endothelial cells remains to be examined but may relate to a recent study showing that NO and BAY 41-2272 act synergistically to activate the enzyme [33]. Therefore, in the presence of BAY 41-2272, H2O2 may enhance soluble guanylyl cyclase activity by inducing increases in eNOS-derived NO. Tempol exhibits beneficial actions on endothelial dysfunction in animal models of hypertension [34] and diabetes [35]. We found that tempol reversed the impairment of endothelialderived NO bioactivity induced by H2O2. Tempol is commonly used as a SOD mimetic and hence its beneficial activities are frequently ascribed to superoxide scavenging. However, tempol is a cell-permeable stable nitroxide radical that efficiently scavenges reactive free radicals and reduces ferryl iron species, actions that may underlie the ability of tempol to attenuate H2O2induced cell death [36] or impairment of endothelial NO bioactivity (current study). It is likely that the findings from the current in vitro study using low micromolar concentrations of H2O2 have important implications for endothelial function under both physiological and patho-physiological conditions. Increasing evidence indicates that H2O2 is an important cell signaling molecule [7,37] and regulator of vascular tone [24]. Thus, H2O2 exhibits vasoconstrictive or vaso-relaxant properties depending on its concentration and blood vessel species, anatomical origin or size. The current study focused on the implications of H2O2 for NO bioactivity in arterial endothelial cells. We have found that while H2O2 promotes cell signaling pathways that result in changes in eNOS phosphorylation at Ser-1177 and Thr-495, enzyme activation and increased NO production, the oxidant induces in parallel oxidative reactions in endothelial cells that act to impair NO bioavailability. Importantly, the propensity for H2O2 to catalyze oxidative reactions in endothelial cells may increase under pathological settings such as atherosclerosis where vascular concentrations of small molecular weight iron complexes increase [31]. Therefore, the current study highlights that inhibition of H2O2-catalyzed oxidative reactions in endothelial cells (e.g., through the chelation of labile iron) represents a potential intervention strategy aimed at preserving endothelial-derived NO bioactivity and hence endothelial function during cardiovascular disease. Acknowledgments We thank Nikhiel Rau, Ana Sharma and Hiu Xiao for excellent technical assistance. This work was supported by a CJ

688

S.R. Thomas et al. / Free Radical Biology & Medicine 41 (2006) 681–688

Martin Postdoctoral Research Fellowship 007158 from the Australian National Health and Medical Research Council (to S. R.T) and grants HL67206 and HL68758 from the National Institute of Health (to J.F.K).

[20]

References

[21]

[1] Lefroy, D. C.; Crake, T.; Uren, N. G.; Davies, G. J.; Maseri, A. Effect of inhibition of nitric oxide synthesis on epicardial coronary artery caliber and coronary blood flow in humans. Circulation 88:43–54; 1993. [2] Rees, D. D.; Palmer, R. M.; Moncada, S. Role of endothelium-derived nitric oxide in the regulation of blood pressure. Proc. Natl. Acad. Sci. U. S. A. 86:3375–3378; 1989. [3] Azuma, H.; Ishikawa, M.; Sekizaki, S. Endothelium-dependent inhibition of platelet aggregation. Br. J. Pharmacol. 88:411–415; 1986. [4] Garg, U. C.; Hassid, A. Nitric oxide-generating vasodilators and 8-bromocGMP inhibit mitogenesis and proliferation of cultured rat vascular smooth muscle cells. J. Clin. Invest. 83:1774–1777; 1989. [5] Widlansky, M. E.; Gokce, N.; Keaney, J. F. Jr., Vita, J. A. The clinical implicationsofendothelialdysfunction.J.Am.Coll.Cardiol.42:1149–1160; 2003. [6] Thomas, S. R.; Chen, K.; Keaney, J. F. Jr. Oxidative stress and endothelial nitric oxide bioactivity. Antioxid. Redox Signal. 5:181–194; 2003. [7] Griendling, K. K.; Sorescu, D.; Ushio-Fukai, M. NAD(P)H oxidase: role in cardiovascular biology and disease. Circ. Res. 86:494–501; 2000. [8] Miura, H.; Bosnjak, J. J.; Ning, G.; Saito, T.; Miura, M.; Gutterman, D. D. Role for hydrogen peroxide in flow-induced dilation of human coronary arterioles. Circ. Res. 92:e31–e40; 2003. [9] Kissner, R.; Nauser, T.; Bugnon, P.; Lye, P. G.; Koppenol, W. H. Formation and properties of peroxynitrite as studied by laser flash photolysis, highpressure stopped-flow technique, and pulse radiolysis. Chem. Res. Toxicol. 10:1285–1292; 1997. [10] Ohara, Y.; Peterson, T. E.; Harrison, D. G. Hypercholesterolemia increases endothelial superoxide anion production. J. Clin. Invest. 91:2546–2551; 1993. [11] Keaney, J. F. Jr., Xu, A.; Cunningham, D.; Jackson, T.; Frei, B.; Vita, J. A. Dietary probucol preserves endothelial function in cholesterol-fed rabbits by limiting vascular oxidative stress and superoxide generation. J. Clin. Invest. 95:2520–2529; 1995. [12] Thomas, S. R.; Chen, K.; Keaney, J. F. Jr., Hydrogen peroxide activates endothelial nitric oxide synthase through coordinated phosphorylation and dephosphorylation via a phosphoinositide 3-kinase-dependent signaling pathway. J. Biol. Chem. 277:6017–6024; 2002. [13] Drummond, G. R.; Cai, H.; Davis, M. E.; Ramasamy, S.; Harrison, D. G. Transcriptional and posttranscriptional regulation of endothelial nitric oxide synthase expression by hydrogen peroxide. Circ. Res. 86:347–354; 2000. [14] Huang, A.; Vita, J. A.; Venema, R. C.; Keaney, J. F. Jr., Ascorbic acid enhances endothelial nitric oxide synthase activity by increasing intracellular tetrahydrobiopterin. J. Biol. Chem. 275:17399–17406; 2000. [15] Huang, A.; Thomas, S. R.; Keaney, J. F. Jr., Measurements of redox control of nitric oxide bioavailability. Methods Enzymol. 359:209–216; 2002. [16] Pap, E. H.; Drummen, G. P.; Winter, V. J.; Kooij, T. W.; Rijken, P.; Wirtz, K. W.; Op den Kamp, J. A.; Hage, W. J.; Post, J. A. Ratiofluorescence microscopy of lipid oxidation in living cells using C11BODIPY(581/591). FEBS Lett. 453:278–282; 1999. [17] Drummen, G. P.; van Liebergen, L. C.; Op den Kamp, J. A.; Post, J. A. C11-BODIPY(581/591), an oxidation-sensitive fluorescent lipid peroxidation probe: (micro)spectroscopic characterization and validation of methodology. Free Radic. Biol. Med. 33:473–490; 2002. [18] Petrat, F.; de, G. H.; Rauen, U. Subcellular distribution of chelatable iron: a laser scanning microscopic study in isolated hepatocytes and liver endothelial cells. Biochem. J. 356:61–69; 2001. [19] Ramachandran, A.; Ceaser, E.; rley-Usmar, V. M. Chronic exposure to nitric oxide alters the free iron pool in endothelial cells: role of

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

[33]

[34]

[35]

[36]

[37]

mitochondrial respiratory complexes and heat shock proteins. Proc. Natl. Acad. Sci. U. S. A. 101:384–389; 2004. Keaney, J. F. Jr., Guo, Y.; Cunningham, D.; Shwaery, G. T.; Xu, A.; Vita, J. A. Vascular incorporation of a-tocopherol prevents endothelial dysfunction due to oxidized LDL by inhibiting protein kinase C stimulation. J. Clin. Invest. 98:386–394; 1996. Zembowicz, A.; Hatchett, R. J.; Jakubowski, A. M.; Gryglewski, R. J. Involvement of nitric oxide in the endothelium-dependent relaxation induced by hydrogen peroxide in rabbit aorta. Br. J. Pharmacol. 110:151–158; 1993. Fulton, D.; Gratton, J. P.; Sessa, W. C. Post-translational control of endothelial nitric oxide synthase: why isn't calcium/calmodulin enough? J. Pharmacol. Exp. Ther. 299:818–824; 2001. Yang, H.; Roberts, L. J.; Shi, M. J.; Zhou, L. C.; Ballard, B. R.; Richardson, A.; Guo, Z. M. Retardation of atherosclerosis by overexpression of catalase or both Cu/Zn-superoxide dismutase and catalase in mice lacking apolipoprotein E. Circ. Res. 95:1075–1081; 2004. Suvorava, T.; Lauer, N.; Kumpf, S.; Jacob, R.; Meyer, W.; Kojda, G. Endogenous vascular hydrogen peroxide regulates arteriolar tension in vivo. Circulation 112:2487–2495; 2005. Jaimes, E. A.; Sweeney, C.; Raij, L. Effects of the reactive oxygen species hydrogen peroxide and hypochlorite on endothelial nitric oxide production. Hypertension 38:877–883; 2001. O'Donnell, V. B.; Chumley, P. H.; Hogg, N.; Bloodsworth, A.; DarleyUsmar, V. M.; Freeman, B. A. Nitric oxide inhibition of lipid peroxidation: kinetics of reaction with lipid peroxyl radicals and comparison with alphatocopherol. Biochemistry 36:15216–15223; 1997. Zhang, W. J.; Frei, B. Intracellular metal ion chelators inhibit TNFalphainduced SP-1 activation and adhesion molecule expression in human aortic endothelial cells. Free Radic. Biol. Med. 34:674–682; 2003. Tampo, Y.; Kotamraju, S.; Chitambar, C. R.; Kalivendi, S. V.; Keszler, A.; Joseph, J.; Kalyanaraman, B. Oxidative stress-induced iron signaling is responsible for peroxide-dependent oxidation of dichlorodihydrofluorescein in endothelial cells: role of transferrin receptor-dependent iron uptake in apoptosis. Circ. Res. 92:56–63; 2003. Tenopoulou, M.; Doulias, P. T.; Barbouti, A.; Brunk, U.; Galaris, D. Role of compartmentalized redox-active iron in hydrogen peroxide-induced DNA damage and apoptosis. Biochem. J. 387:703–710; 2005. Duffy, S. J.; Biegelsen, E. S.; Holbrook, M.; Russel, J. D.; Gokce, N.; Keaney, J. F. Jr., Vita, J. A. Iron chelation improves endothelial function in patients with coronary artery disease. Circulation 103:2799–2804; 2001. Stadler, N.; Lindner, R. A.; Davies, M. J. Direct detection and quantification of transition metal ions in human atherosclerotic plaques: evidence for the presence of elevated levels of iron and copper. Arterioscler. Thromb. Vasc. Biol. 24:949–954; 2004. Burke, T. M.; Wolin, M. S. Hydrogen peroxide elicits pulmonary arterial relaxation and guanylate cyclase activation. Am. J. Physiol. 252: H721–H732; 1987. Schmidt, P. M.; Schramm, M.; Schroder, H.; Wunder, F.; Stasch, J. P. Identification of residues crucially involved in the binding of the heme moiety of soluble guanylate cyclase. J. Biol. Chem. 279:3025–3032; 2004. Schnackenberg, C. G.; Welch, W. J.; Wilcox, C. S. Normalization of blood pressure and renal vascular resistance in SHR with a membrane-permeable superoxide dismutase mimetic: role of nitric oxide. Hypertension 32:59–64; 1998. Nassar, T.; Kadery, B.; Lotan, C.; Da'as, N.; Kleinman, Y.; Haj-Yehia, A. Effects of the superoxide dismutase-mimetic compound tempol on endothelial dysfunction in streptozotocin-induced diabetic rats. Eur. J. Pharmacol. 436:111–118; 2002. Hahn, S. M.; Mitchell, J. B.; Shacter, E. Tempol inhibits neutrophil and hydrogen peroxide-mediated DNA damage. Free Radic. Biol. Med. 23:879–884; 1997. Chen, K.; Thomas, S. R.; Keaney, J. F. Jr., Beyond LDL oxidation: ROS in vascular signal transduction. Free Radic. Biol. Med. 35:117–132; 2003.