Electron Spin Resonance Detection of Hydrogen Peroxide as an Endothelium-Derived Hyperpolarizing Factor in Porcine Coronary Microvessels
Objective— Endothelium-derived hyperpolarizing factor (EDHF) plays an important role in modulating vascular tone, especially in microvessels, although its nature has yet to be elucidated. This study was designed to examine whether hydrogen peroxide (H2O2) is an EDHF in porcine coronary microvessels with use of an electron spin resonance (ESR) method to directly detect H2O2 production from the endothelium.
Methods and Results— Isometric tension and membrane-potential recordings demonstrated that bradykinin and substance P caused EDHF-mediated relaxations and hyperpolarizations of porcine coronary microvessels in the presence of indomethacin and Nω-nitro-l-arginine. The contribution of H2O2 to the EDHF-mediated responses was demonstrated by the inhibitory effect of catalase and by the relaxing and hyperpolarizing effects of exogenous H2O2. Endothelial production of H2O2 was quantified in bradykinin- or substance P–stimulated intact blood vessels by ESR spectroscopy. Tiron, a superoxide scavenger that facilitates H2O2 formation, enhanced bradykinin-induced production of H2O2, as well as the EDHF-mediated relaxations and hyperpolarizations. By contrast, cytochrome P-450 inhibitors (sulfaphenazole or 17-octadecynoic acid) or a gap junction inhibitor (18α-glycyrrhetinic acid) failed to inhibit the EDHF-mediated relaxations. Involvement of endothelium-derived K+ was not evident in experiments with ouabain plus Ba2+ or exogenous K+.
Conclusion— These results provide ESR evidence that H2O2 is an EDHF in porcine coronary microvessels.
The endothelium plays an important role in maintaining vascular homeostasis by synthesizing and releasing several vasodilating factors, including prostacyclin, nitric oxide (NO), and the still-unidentified endothelium-derived hyperpolarizing factor (EDHF). Since the first reports of the existence of EDHF,1,2⇓ several candidates for EDHF have been proposed, including cytochrome P-450 metabolites,3,4⇓ endothelium-derived K+,5,6⇓ and gap-junctional electrical communication between endothelial cells and smooth muscle cells.7,8⇓ We and others have recently demonstrated that endothelium-derived hydrogen peroxide (H2O2) is an EDHF of mouse and human mesenteric arteries,9,10⇓ piglet pial arterioles,11 and human and canine coronary microvessels (in flow-induced and autoregulatory relaxation, respectively).12,13⇓ However, it remains to be determined whether H2O2 is also an EDHF in agonist-induced relaxation of the coronary microvessels, in which EDHF plays an important role in modulating vascular tone. Moreover, endothelial production of H2O2/EDHF remains to be demonstrated by methods other than conventional tension or membrane-potential recordings. The present study was thus designed to examine whether H2O2 plays a role as an EDHF in porcine coronary microvessels by using an electron spin resonance (ESR) method to directly detect endothelial production of H2O2.
This study was reviewed by the Committee on Ethics in Animal Experiments of the Kyushu University and was carried out according to the Guidelines for Animal Experiments of the Kyushu University and of the Japanese Government.
Animals and Tissue Preparation
We used a total of 36 domestic male pigs (Nihon Crea, Tokyo, Japan; 2 to 3 months old and weighing 25 to 30 kg). Animals were humanely killed with a lethal dose of pentobarbital sodium, and then the right ventricular free wall was carefully removed. Epicardial right coronary arteries and distal coronary microvessels (250 to 300 μm in diameter) were excised from the right ventricular free wall and carefully cleaned of adherent perivascular connective tissue under a microscope. The microvessels were cut into 1.2-mm rings for organ chamber experiments.
Organ Chamber Experiments
Experiments were performed in 37°C Krebs’ solution bubbled with 95% O2 and 5% CO2. Isometric tension was recorded in isolated arterial rings contracted with prostaglandin F2α (3 to 10×10−6 mol/L) or KCl (40 to 60 mmol/L). The extent of contraction was adjusted to 50% to 70% of the contractions induced by 62 mmol/L KCl.9 Endothelium-dependent relaxations to bradykinin and substance P and endothelium-independent relaxations to sodium nitroprusside (SNP), H2O2, and K+ were examined. The contributions of vasodilator prostaglandins, NO, and EDHF to bradykinin-induced endothelium-dependent relaxations were determined by the inhibitory effect of indomethacin (10−5 mol/L), Nω-nitro-l-arginine (l-NNA, 10−4 mol/L), and KCl (40 to 60 mmol/L), respectively.9
To examine the nature of EDHF and the mechanism of EDHF-mediated relaxations, the following drugs were used: apamin (10−6 mol/L, an inhibitor of small conductance Ca2+-activated K+ [KCa] channels), charybdotoxin (10−7 mol/L, an inhibitor of large and intermediate conductance KCa channels), tetrabutylammonium (TBA, 10−3 mol/L, a nonspecific inhibitor of KCa channels), catalase (6250 U/mL, an enzyme that dismutates H2O2),9 3-amino-1,2,4-triazole (aminotriazole, 5×10−2 mol/L, an inhibitor of catalase),9 4,5-dihydroxy-1,3-benzene disulfonic acid (Tiron, 10−3 mol/L, which reacts with superoxide anions to form H2O2),14 polyethylene glycol (PEG)–conjugated superoxide dismutase (SOD, 1000 U/mL), sulfaphenazole (10−5 mol/L, a specific inhibitor of cytochrome P-450 epoxygenase),4 17-octadecynoic acid (17-ODYA, 3×10−6 mol/L, a nonselective inhibitor of cytochrome P-450),15,16⇓ 18α-glycyrrhetinic acid (18α-GA, 10−4 mol/L, an inhibitor of gap junctions),15,16⇓ ouabain (10−3 mol/L, an inhibitor of the Na/K pump), and Ba2+ (3×10−5 mol/L, an inhibitor of inward rectifier K+ channels [KIR channels]).5 Catalase and PEG-SOD were applied to organ chambers 2 hours before precontraction and subsequent observation of vasorelaxations. Other inhibitors were applied to organ chambers 30 minutes before precontraction and subsequent observation of vasorelaxations.
Rings of porcine coronary microvessels were placed in experimental chambers, perfused with 37°C Krebs’ solution containing indomethacin (10−5 mol/L) and l-NNA (10−4 mol/L), and bubbled with 95% O2 and 5% CO2. A fine, glass capillary microelectrode was impaled into a smooth muscle cell from the adventitial side in coronary microvessels. Changes in membrane potentials produced by bradykinin, levcromakalim, and H2O2 were continuously recorded.9
ESR Measurement of H2O2
To increase the amount of H2O2, epicardial coronary arteries as well as microvessels were used in the ESR experiments. Isolated porcine coronary microvessels and epicardial arteries (3.0 mm in diameter and 1.0 cm in length) were incubated in 500 μL Krebs’ solution containing indomethacin (10−5 mol/L), l-NNA (10−4 mol/L), and aminotriazole (10−3 mol/L) with or without Tiron (10−2 mol/L) or PEG-SOD (1000 U/mL), which was bubbled with 95% O2 and 5% CO2. In some experiments, the vascular endothelium was removed with a cotton swab. After 30 minutes of incubation, some arteries were stimulated with bradykinin (10−6 mol/L) for 10 minutes, and then the perfusate was collected and centrifuged in centrifuge tubes (Centricon YM-3, Millipore Co) for 90 minutes at 4°C. The filtrates were added to 1-hydroxy-2,2,5,5,-tetramethyl-3-imidazoline-3-oxide (5×10−4 mol/L), p-acetamidophenol (5×10−5 mol/L), and horseradish peroxidase 0.1 mg/mL)17 and incubated for 60 seconds at 37°C, and then ESR measurements were performed with an X-band (9.45-GHz) ESR spectrometer (JES-RE-1X, Jeol Ltd). In this system, H2O2 was detected as a specific waveform of a stable nitroxide radical, and relative signal intensity was determined in each experiment with a manganese marker as a control. H2O2 concentration was determined from the relative ESR signal–H2O2 concentration relation obtained from preliminary experiments. Response to bradykinin was determined as the fold increase in H2O2 signal from baseline signal intensity under control conditions.
Drugs and Solution
The ionic composition of the Krebs’ solution (mmol/L) was as follows: Na+ 144, K+ 5.9, Mg2+ 1.2, Ca2+ 2.5, H2PO4− 1.2, HCO3− 24, Cl− 129.7, and glucose 5.5. Bradykinin, indomethacin, l-NNA, catalase (No. C-40), PEG-SOD, SNP, 18α-GA, 17-ODYA, apamin, charybdotoxin, TBA, BaCl2, and ouabain were obtained from Sigma Chemical Co. Catalase from another supplier (No. 106836, Roche Diagnostics Co) was also used. 1-Hydroxy-2,2,5,5,-tetramethyl-3-imidazoline-3-oxide was obtained from Acros Organics, and p-acetamidophenol and horseradish peroxidase were obtained from Wako Pure Chemical Industries. Levcromakalim was a gift from SmithKline Beecham Pharmaceuticals (Philadelphia, Pa). Indomethacin was dissolved in 10−2 mol/L Na2CO3. 18α-GA and 17-ODYA were dissolved in dimethyl sulfoxide. Sulfaphenazole was dissolved in dimethyl sulfoxide or ethanol, and in both cases, identical results were obtained. Levcromakalim was dissolved in 90% ethanol, and all other drugs were dissolved in distilled water. The solvents used did not affect the mechanical or electrical responses at their final bath concentrations.
Data are shown as mean±SEM. Dose-response curves were analyzed by 2-way ANOVA, followed by Scheffé’s post hoc test for multiple comparisons. Other values were analyzed by paired Student’s t test or 1-way ANOVA, according to propriety. P<0.05 was considered statistically significant.
EDHF Responses of Porcine Coronary Microvessels
Bradykinin (10−10 to 10−6 mol/L) elicited concentration-dependent relaxations of porcine coronary microvessels, which were almost resistant to indomethacin or indomethacin plus l-NNA, but were highly sensitive to the combined treatment with indomethacin, l-NNA, and KCl (n=5, Figure 1A). In the presence of KCl alone, bradykinin still elicited significant relaxations, which were partially sensitive to additional treatment with indomethacin, whereas the remaining relaxations could be attributed to NO (n=6, Figure 1B). Relaxations resistant to indomethacin and l-NNA were abolished by charybdotoxin plus apamin, KCa channel inhibitors (maximal relaxation 84±3% vs 13±7% in the absence and presence of charybdotoxin/apamin, respectively; n=6, P<0.05), suggesting that those relaxations are mediated by EDHF. Bradykinin elicited concentration-dependent membrane hyperpolarizations of vascular smooth muscle cells in the presence of indomethacin and l-NNA (Figure 1C). These results suggest a primary role for EDHF in bradykinin-induced relaxations of porcine coronary microvessels, with a small contribution from NO and vasodilator prostaglandins. Substance P is another vasodilating agonist in porcine coronary arteries.18,19⇓ Substance P (10−12 to 10−6 mol/L) also elicited indomethacin- and l-NNA–resistant relaxations, which were significantly inhibited by additional treatment with KCl (maximal relaxation 36±12% and 14±7% in the absence and presence of KCl, respectively; n=6, P<0.05), suggesting the involvement of EDHF in these relaxations.
Inhibitory Effects of Catalase on Vascular Responses
To examine the role of H2O2 in EDHF-mediated relaxations, we first examined the inhibitory effect of catalase, which selectively dismutates H2O2 into oxygen and water.9,10⇓ Catalase (Sigma No. C-40 used throughout this study) markedly inhibited the EDHF-mediated relaxations elicited by bradykinin (n=5, P<0.05; Figure 2A). Catalase from another supplier (Roche) also significantly inhibited the EDHF-mediated relaxations (maximal relaxation 83±5% vs 64±9% in the absence and presence of catalase, respectively; n=6, P<0.05), although the extent of inhibition was smaller than that of Sigma catalase. We found that catalase requires a 2-hour incubation period to exert a significant inhibitory effect on relaxation (data not shown). By contrast, catalase did not affect NO-mediated relaxations in the presence of indomethacin and KCl in response to bradykinin, excluding its inhibitory effect on NO-mediated responses (n=6; Figure 2A). Catalase also significantly inhibited bradykinin (10−7 mol/L)-induced hyperpolarizations (n=5; Figure 2B left). To further clarify the specificity of catalase, aminotriazole was simultaneously applied, which inhibits the peroxide-binding site of catalase and thus, enzymatically inactivates the enzyme.9 When inactivated with aminotriazole, catalase lost its inhibitory effect on EDHF-mediated hyperpolarizations (n=3; Figure 2B middle). Furthermore, catalase did not affect hyperpolarizations to levcromakalim, a direct K+ channel opener (n=3; Figure 2B right). Catalase also significantly inhibited EDHF-mediated relaxations in response to substance P in the presence of indomethacin and l-NNA (maximal relaxation 36±12% vs 10±6% in the absence and presence of catalase, respectively; n=6, P<0.05).
Effect of Exogenous H2O2
We then examined whether exogenously applied H2O2 could mimic the EDHF-mediated relaxations and hyperpolarizations of coronary microvessels without the endothelium. H2O2 (10−8 to 10−4 mol/L) caused concentration-dependent relaxations that were inhibited by either the KCa channel inhibitor TBA or high extracellular K+ (n=4 to 6; Figure 3A). H2O2 (10−5 to 10−4 mol/L) also elicited significant hyperpolarizations of blood vessels without the endothelium (n=7; Figure 3B).
ESR Detection of Endothelial H2O2 Production
It is mandatory to directly detect endothelial production of H2O2 to conclude that H2O2 is an EDHF. Because direct measurement of H2O2 is difficult, especially in living tissues, we performed ESR spectrometry of H2O2 by using p-acetamidophenol, 1-hydroxy-2,2,5,5,-tetramethyl-3-imidazoline-3-oxide, and horseradish peroxidase to form stable nitroxide radicals as an H2O2 indicator.17 To increase the amount of H2O2, epicardial coronary arteries as well as microvessels were used in the ESR experiments. H2O2 production was detected as an ESR signal of nitroxide radicals in perfusates of porcine coronary arteries. Bradykinin elicited an increase in H2O2 production, which was highly sensitive to catalase (500 U/mL) (Figure 4A). Bradykinin significantly increased H2O2 production (n=5, P<0.05; Figure 4B). The mean concentration of H2O2 in the perfusate of the arteries on stimulation with bradykinin was 1.5±0.2×10−6 mol/L (n=5). Removal of the endothelium abolished the H2O2 signal, confirming that H2O2 production in response to bradykinin is endothelium dependent (n=5; Figure 4B). An increase in the substance P–induced H2O2 signal was noted but insignificant (1.2±0.7-fold increase from baseline, n=5).
Effect of Tiron and PEG-SOD on H2O2 Production and Vascular Responses
To characterize endothelial H2O2 production, Tiron (10−3 mol/L, a cell-permeable SOD mimetic) and PEG-SOD were applied to porcine coronary microvessels. Because Tiron and SOD react with superoxide anions to form H2O2,14,20⇓ they are expected to enhance agonist-induced H2O2 production from the endothelium if H2O2 is converted from superoxide. Indeed, Tiron significantly enhanced bradykinin-induced H2O2 production from porcine coronary arteries (n=5, P<0.05; Figure 4B). By contrast, PEG-SOD (1000 U/mL) failed to enhance H2O2 production from porcine coronary arteries (n=5; data not shown). Pretreatment with Tiron also significantly enhanced the bradykinin-induced, EDHF-mediated relaxations (n=5, P<0.05; Figure 5A) and hyperpolarizations (n=6, P<0.05; Figure 5B) in the presence of indomethacin and l-NNA. However, Tiron did not affect endothelium-independent relaxations to either SNP (10−10 to 10−6 mol/L; maximal relaxation 97±1% vs 96±2% in the absence and presence, respectively, of Tiron; n=6) or levcromakalim (10−8 to 10−5 mol/L; maximal relaxation 60±13% vs 50±16% in the absence and presence of Tiron, respectively; n=4). By contrast, pretreatment with PEG-SOD (1000 U/mL) failed to enhance EDHF-mediated relaxations to bradykinin in the presence of indomethacin and l-NNA (maximal relaxation 85±4% vs 81±3% in the absence and presence of PEG-SOD, respectively; n=4).
Role of Cytochrome P-450 Epoxygenases, K+, and Gap Junctions in EDHF-Mediated Responses
It has been reported that cytochrome P-450 epoxygenases are involved in the synthesis of EDHF in epicardial porcine coronary arteries, because EDHF-mediated responses were inhibited by sulfaphenazole, an inhibitor of cytochrome P-450 epoxygenase.4 In the present study, we used sulfaphenazole and 17-ODYA, another inhibitor of cytochrome P-450,15,16⇓ to examine the role of cytochrome P-450 epoxygenases in the EDHF-mediated responses in porcine coronary microvessels. Importantly, neither inhibitor affected the EDHF-mediated relaxations (n=5 to 6; Figure 6A). It has also been reported that K+ released from the endothelium is an EDHF in rat hepatic arteries.5 This notion was based on the observations that K+ hyperpolarized and relaxed vascular smooth muscle by activating Na/K pumps and inward rectifier potassium channels (KIR channels) and that Ba2+ plus ouabain, inhibitors of KIR channels and Na/K pumps respectively, inhibited the EDHF-mediated responses.5 In the present study, although Ba2+ (3×10−5 mol/L) plus ouabain (10−3 mol/L) shifted the concentration-response curve to bradykinin to the right in the presence of indomethacin and l-NNA in porcine coronary microvessels, the combination failed to affect maximal vasodilator responses (n=4; Figure 6B). Exogenously applied K+ (5 to 20 mmol/L) caused only contractions in the arteries studied (n=3; data not shown). To examine the role of gap junctions in the EDHF-mediated responses, 18α-GA7 was used. 18α-GA (10−4 mol/L) failed to inhibit the EDHF-mediated relaxations of porcine coronary microvessels (n=5; Figure 6C).
H2O2 as a Primary EDHF in Porcine Coronary Microvessels
In porcine coronary microvessels, bradykinin elicited relaxations attributable to prostaglandins, NO, and EDHF. The contribution of EDHF to bradykinin-induced relaxation is dominant compared with that of NO or prostacyclin. As a primary relaxing factor in response to bradykinin, EDHF appears to compensate for the relaxations after inhibition by NO and prostaglandins. By contrast, the extent of EDHF-mediated relaxation in response to substance P is relatively small. In the present study, catalase markedly inhibited the bradykinin-induced, EDHF-mediated relaxations and hyperpolarizations of porcine coronary microvessels. Although previous reports failed to find an inhibitory effect of catalase on EDHF-mediated responses in porcine coronary arteries,18,21⇓ recent investigations successfully utilized catalase to inhibit EDHF-mediated vascular responses.11,12⇓ The different results with catalase may be caused by the difference in vessel size and/or experimental conditions, including the catalase preparations. Indeed, in the present study, a longer (≈2 hours) incubation period was required to observe the inhibitory effect of catalase. Similarly, myeloperoxidase, another large molecule, also needs 2 to 3 hours to accumulate in the subendothelial space by way of transcytosis.22 Thus, it is not surprising that exogenous catalase, a large molecule, also needs a longer incubation period to penetrate into the myoendothelial space. The difference in the effect of different catalase preparations can also be caused by the different efficiencies of penetration into the myoendothelial space. In the present study, the specificity of catalase was confirmed by the following findings: (1) catalase did not affect endogenous. NO-mediated relaxations; (2) enzymatically inactivated catalase lost its inhibitory effect on EDHF-mediated hyperpolarizations; (3) catalase did not affect levcromakalim-induced, direct hyperpolarizations of vascular smooth muscle; and (4) catalases from different suppliers significantly inhibited EDHF-mediated responses. Exogenously applied H2O2 mimicked the EDHF-mediated relaxations and hyperpolarizations of blood vessels without the endothelium. H2O2-induced relaxations were sensitive to a KCa channel inhibitor or high extracellular K+, suggesting that those relaxations are mainly caused by hyperpolarization of the smooth muscle through activation of KCa channels. These functional data also indicate that H2O2 plays a primary role as an EDHF in porcine coronary microvessels.
ESR Detection of Endothelial H2O2 Production
It is important to detect a reasonable amount of H2O2 released from the endothelium. In the present study, ESR measurement demonstrated that endothelial cells of porcine coronary microvessels produce a significant amount of H2O2 on stimulation by bradykinin. There was a difference between the H2O2 concentration detected in the perfusate from blood vessels and that required to cause vascular relaxations/hyperpolarizations. It is conceivable that the concentration of endogenous H2O2 detected in the perfusate is lower than that in myoendothelial space where EDHF works, whereas the concentration of exogenous H2O2 required to cause vascular relaxations may be higher than that in the myoendothelial space because exogenous H2O2 rapidly reacts with endogenous peroxidases. Endothelial cells have a capacity to produce superoxide anions from several intracellular sources, including endothelial NO synthase, cyclooxygenase, lipoxygenase, cytochrome P-450 enzymes, and NAD(P)H oxidases.23 In the present study, H2O2 formation can be attributed to superoxide-producing enzymes, because Tiron facilitated H2O2 production as well as EDHF-mediated vascular responses, although PEG-SOD affected neither H2O2 production nor EDHF-mediated relaxations. The different effect of Tiron and PEG-SOD might be caused by differences in cell permeability, although long-term treatment with PEG-SOD has been reported to augment endothelium-dependent vascular responses in a previous study.20 Among the possible sources of superoxide anions, we have demonstrated that endothelial NO synthase is a major source of H2O2 in mouse mesenteric arteries.9 However, the source of superoxide and H2O2 in porcine coronary arteries remains to be determined in a future study.
In the present study, the extent of EDHF-mediated relaxations to substance P was smaller than that to bradykinin (maximal relaxation 36% vs 91% for substance P– and bradykinin-induced relaxations, respectively). Under these experimental conditions, catalase still inhibited the EDHF-mediated relaxations to substance P, although H2O2 production in response to substance P was not statistically significant in the ESR experiments, probably because the sensitivity of the ESR measurement was not high enough to detect H2O2 production in response to substance P. Concerning other stimuli, a recent report suggests that H2O2 also plays a primary role as an EDHF in flow-induced vasodilatations in human coronary microvessels.12 Further studies are needed to clarify whether endothelial production of H2O2 as an EDHF can be generalized to various vascular beds and stimuli, including endothelial agonists and physiological stimuli.
Other Candidates for EDHF
Epoxyeicosatrienoic acids, major products of cytochrome P-450 epoxygenase, have been proposed to be EDHFs in some vascular beds, including epicardial porcine coronary arteries.3,4,24⇓⇓ In the present study, we concluded that cytochrome P-450 metabolism may not be involved in the EDHF-mediated responses of porcine coronary microvessels because sulfaphenazole, a specific inhibitor of cytochrome P-450 epoxygenase,4 did not affect the EDHF-mediated relaxations or hyperpolarizations. The difference in vessel size may account for the different conclusion between a previous report4 and the present study. 17-ODYA, another inhibitor of cytochrome P-450,15,16⇓ also failed to inhibit the EDHF-mediated relaxations or hyperpolarizations, a consistent finding with a previous report that studied epicardial porcine coronary arteries.21
K+ released from the endothelium has also been suggested to be an EDHF in epicardial porcine coronary arteries.6 We attempted to inhibit Na/K pumps and KIR channels, proposed targets of endothelium-derived K+, by using ouabain plus Ba2+.5 Although the combination of ouabain plus Ba2+ shifted the dose-response curve to bradykinin to the right, the combination did not significantly affect the maximal responses to bradykinin. Ouabain-induced membrane depolarization may decrease bradykinin-induced Ca2+ influx into the endothelium, which is essential for the production of both NO and EDHF.25 We have observed that ouabain (10−3 mol/L) inhibits NO production in epicardial porcine coronary microvessels (authors’ unpublished observations). Thus, the specificity of ouabain plus Ba2+ is questionable in the present preparations. Furthermore, exogenously applied K+ elicited only contractions and depolarizations of the smooth muscle under the same experimental condition where bradykinin-induced, EDHF-mediated relaxations were consistently observed. Thus, involvement of K+ was not evident in porcine coronary microvessels in the present study.
It has also been proposed that intercellular electrical communication through gap junctions could contribute to endothelium-dependent hyperpolarizations/relaxations of epicardial porcine coronary arteries.19 In the present study, 18α-GA failed to inhibit the EDHF-mediated relaxations in porcine coronary microvessels. Thus, gap junctions may not be required to activate the H2O2-hyperpolarization pathway in porcine coronary microvessels.
Mechanism of H2O2-Induced Hyperpolarization and Relaxation
In the present study, EDHF-mediated responses were inhibited by charybdotoxin plus apamin or TBA, KCa channel inhibitors. Recently, it has been suggested that the combination of charybdotoxin plus apamin inhibits the endothelium from synthesizing EDHF.9,19⇓ If so, it is unknown at the present time which mechanism is involved in the EDHF-mediated hyperpolarization at the smooth muscle level. In the present study, H2O2 elicited hyperpolarizations and relaxations of vascular smooth muscle that were sensitive to a KCa channel inhibitor or high extracellular K+. It is conceivable that H2O2 activates KCa channels in smooth muscle to cause hyperpolarizations, as reported previously.26,27⇓ Indeed, H2O2 is known to activates KCa channels by direct modulation of the channels, by a cGMP-dependent mechanism,28 or by a lipoxygenase-mediated mechanism29 in porcine coronary arteries. The precise mechanism by which H2O2 activates KCa channels remains to be clarified in future studies.
In summary, the present study provides direct evidence for endothelial production of H2O2 with use of an ESR method. Further studies are needed to clarify the physiological and pathological significance of the novel role of endothelium-derived H2O2 as an EDHF.
This work was supported in part by grants-in-aid (No. 07307010, 10877111, 12877114) from the Japanese Ministry of Education, Science, Sports and Culture, Tokyo, Japan. We wish to thank Dr Mikio Nakashima, Saga Medical University, for valuable comments on our work.
Consulting Editor for this article was Peter Libby, MD, Brigham and Women’s Hospital, Boston, Mass.
- Received February 27, 2003.
- Accepted May 13, 2003.
- ↵Rosolowsky M, Campbell WB. Role of PGI2 and epoxyeicosatrienoic acids in relaxation of bovine coronary arteries to arachidonic acid. Am J Physiol. 1993; 264: H327–H335.
- ↵Lacza Z, Puskar M, Kis B, Perciaccante JV, Miller AW, Busija DW. Hydrogen peroxide acts as an EDHF in the piglet pial vasculature in response to bradykinin. Am J Physiol Heart Circ Physiol. 2002; 283: H406–H411.
- ↵Miura H, Bosnjak JJ, Ning G, Saito T, Miura M, Gutterman DD. Role for hydrogen peroxide in flow-induced dilation of human coronary arterioles. Circ Res. 2003; 92: e31–e40.
- ↵Yada T, Shimokawa H, Hiramatsu O, Kajita T, Shigeto F, Goto M, Ogasawara Y, Kajiya F. Hydrogen peroxide, an endogenous endothelium-derived hyperpolarizing factor, plays an important role in coronary autoregulation in vivo. Circulation. 2003; 107: 1040–1045.
- ↵Matsuo T, Shinzawa H, Togashi H, Aoki M, Sugahara K, Saito K, Saito T, Takahashi T, Yamaguchi I, Aoyama M, Kamada H. Highly sensitive hepatitis B surface antigen detection by measuring stable nitroxide radical formation with ESR spectroscopy. Free Radic Biol Med. 1998; 25: 929–935.
- ↵Mugge A, Elwell JH, Peterson TE, Hofmeyer TG, Heistad DD, Harrison DG. Chronic treatment with polyethylene-glycolated superoxide dismutase partially restores endothelium-dependent vascular relaxations in cholesterol-fed rabbits. Circ Res. 1991; 69: 1293–1300.
- ↵Baldus S, Eiserich JP, Mani A, Castro L, Figueroa M, Chumley P, Ma W, Tousson A, White CR, Bullard DC, Brennan ML, Lusis AJ, Moore KP, Freeman BA. Endothelial transcytosis of myeloperoxidase confers specificity to vascular ECM proteins as targets of tyrosine nitration. J Clin Invest. 2001; 108: 1759–1770.
- ↵Barlow RS, White RE. Hydrogen peroxide relaxes porcine coronary arteries by stimulating BKCa channel activity. Am J Physiol. 1998; 275: H1283–H1289.
- ↵Burke TM, Wolin MS. Hydrogen peroxide elicits pulmonary arterial relaxation and guanylate cyclase activation. Am J Physiol. 1987; 252: H721–H732.