Pivotal role of Cu,Zn-superoxide dismutase in endothelium-dependent hyperpolarization

The endothelium plays an important role in maintaining vascular homeostasis by synthesizing and releasing several vasodilating factors, including prostacyclin, NO, and endothelium-derived hyperpolarizing factor (EDHF). We have recently identified that endothelium-derived H₂O₂ is an EDHF in mesenteric arteries of mice and humans and in porcine coronary microvessels. However, the mechanism for the endothelial production of H₂O₂ as an EDHF remains to be elucidated. In this study, we tested our hypothesis that Cu,Zn-superoxide dismutase (Cu,Zn-SOD) plays a pivotal role in endothelium-dependent hyperpolarization, using control and Cu,Zn-SOD–/– mice. In mesenteric arteries, EDHF-mediated relaxations and hyperpolarizations were significantly reduced in Cu,Zn-SOD–/– mice with no inhibitory effect of catalase, while endothelium-independent relaxations and hyperpolarizations were preserved. Endothelial H₂O₂ production also was significantly reduced in Cu,Zn-SOD–/– mice. In Langendorff isolated heart, bradykinin-induced increase in coronary flow was significantly reduced in Cu,Zn-SOD–/– mice, again with no inhibitory effect of catalase. The exogenous SOD mimetic tempol significantly improved EDHF-mediated relaxations and hyperpolarizations and coronary flow response in Cu,Zn-SOD–/– mice. These results prove the novel concept that endothelial Cu,Zn-SOD plays an important role as an “EDHF synthase” in mice, in addition to its classical role to scavenge superoxide anions.

The endothelium plays an important role in maintaining vascular homeostasis by synthesizing and releasing several vasodilators, including prostacyclin, NO, and endothelium-derived hyperpolarizing factor (EDHF) (1, 2). EDHF plays an important role in modulating vascular tone, especially in microvessels (1–3). Since the first reports on the existence of EDHF (4, 5), several candidates for EDHF have been proposed, including cytochrome P450 metabolites (6, 7), potassium ions (8, 9), and gap junctional electrical communication (10, 11). We have recently identified that endothelium-derived H₂O₂ is an EDHF in mesenteric arteries of mice (3) and humans (12) and in porcine coronary microvessels (13). Subsequently other investigators have confirmed the importance of H₂O₂ as an endogenous EDHF in coronary microvessels (14, 15) and piglet pial arterioles (16). However, the mechanism for the endothelial production of H₂O₂ as an endogenous EDHF remains to be elucidated. In endothelial cells, H₂O₂ is mainly produced by superoxide dismutase (SOD) from superoxide anions.

There are three isoforms of SOD that dismutate superoxide anions to H₂O₂. Copper,zinc-SOD (Cu,Zn-SOD, or SOD1) is located in the cytosol and nucleus and to a lesser extent in the mitochondria, and is expressed in almost all cell types (17). Cu,Zn-SOD is reported to dismutate superoxide anions derived from eNOS during NO synthesis (18) and thereby prolongs the half-life of NO (19, 20). Manganese SOD (MnSOD, or SOD2) is located in the mitochondrial matrix and is the primary SOD isoform that dismutates superoxide anions generated by the respiratory chain (21). Extracellular SOD (ecSOD, or SOD3) is secreted extracellularly and is bound to sulfated polysaccharides such as heparan sulfate on cell surfaces (21). Cu,Zn-SOD is a major SOD in blood vessels, accounting for approximately 50–80% of total SOD activity (22, 23). Since we have found that endothelial production of H₂O₂/EDHF is reduced in eNOS-knockout (eNOS^–/–) mice (3), we hypothesized that H₂O₂ is produced by endothelial Cu,Zn-SOD from superoxide anions (mainly derived from eNOS) and acts as an endogenous EDHF. In this study, we examined our hypothesis that endothelial Cu,Zn-SOD plays a pivotal role as an EDHF synthase using Cu,Zn-SOD knockout (Cu,Zn-SOD^–/–) mice.

Characterization of Cu,Zn-SOD–/– mice. There was no significant difference in systolic blood pressure (mmHg) between the Cu,Zn-SOD^–/– mice (117 ± 3) and the control mice (120 ± 4) (n = 17 each). Western blot analysis showed that expression of Cu,Zn-SOD was absent in mesenteric arteries from the Cu,Zn-SOD^–/– mice, whereas the expression of MnSOD or ecSOD was unchanged (Figure 1a). The Cu,Zn-SOD activity in mesenteric arteries, which was examined by a nitroblue tetrazolium method (24), was totally absent in the Cu,Zn-SOD^–/– mice (Figure 1b). By contrast, the activities of glutathione peroxidase and catalase in the liver, both of which dismutate H₂O₂, were unchanged in the knockout mice (Figure 1, c and d, respectively). Western blot analysis for eNOS showed that the extent of eNOS expression was comparable between the two strains (Figure 1e).

Figure 1

Characterization of Cu,Zn-SOD–/– mice. (a) Western blot analysis for SOD isoforms in mesenteric arteries from control (WT) and Cu,Zn-SOD–/– (KO) mice. Cu,Zn-SOD expression was absent in the Cu,Zn-SOD–/– mice, whereas the expression of MnSOD and ecSOD was unaltered. (b) Cu,Zn-SOD activity of mesenteric arteries from the two strains. Cu,Zn-SOD activity was absent in the Cu,Zn-SOD–/– mice. Comparable activities of glutathione peroxidase (c) and catalase (d) of the liver in the two strains (n = 5 each). (e) Western blot analysis for eNOS expression in mesenteric arteries (n = 5 each). eNOS protein expression was comparable between the two strains.

Endothelium-dependent relaxations. In mesenteric arteries of control mice, endothelium-dependent relaxations in response to ACh were resistant to indomethacin (10^–5 M) or L-NNA (10^–4 M) but were significantly inhibited by KCl (40–60 mM) (Figure 2a). In mesenteric arteries of Cu,Zn-SOD^–/– mice, endothelium-dependent relaxations in response to ACh were slightly attenuated compared with those of control mice, and L-NNA markedly inhibited the relaxations (Figure 2b). Figure 2c shows the relative contributions of prostacyclin, NO, and EDHF to the endothelium-dependent relaxations of mesenteric arteries from both strains when using KCl as an inhibitor of EDHF-mediated responses. In mesenteric arteries, EDHF-mediated relaxations were significantly reduced, while NO-mediated relaxations were significantly increased in Cu,Zn-SOD^–/– mice compared with control mice. The combination of charybdotoxin and apamin also significantly inhibited EDHF-mediated relaxations in both strains (Figure 2, d and e). The EDHF-mediated relaxations, as evaluated by the inhibitory effect of the combination, were also significantly reduced in Cu,Zn-SOD^–/– mice (Figure 2f).

Figure 2

Reduced EDHF-mediated relaxations in Cu,Zn-SOD^–/– mice. Endothelium-dependent relaxations of mesenteric arteries from control WT mice (a) and from Cu,Zn-SOD^–/– mice (b) (n = 7 each) are shown. Indo, indomethacin (10^–5 M); L-NNA, 10^-4 M; KCl, 40–60 mM. In mesenteric arteries of control mice (a), endothelium-dependent relaxations were resistant to indomethacin or L-NNA, whereas in those of Cu,Zn-SOD^–/– mice (b), indomethacin and L-NNA markedly inhibited the relaxations. **P < 0.01, ***P < 0.0001. (c) Relative contribution of prostacyclin (PGI₂), NO, and EDHF to the endothelium-dependent relaxations of mesenteric arteries in response to ACh. In mesenteric arteries from Cu,Zn-SOD^–/– mice, EDHF-mediated relaxations were significantly attenuated and NO-mediated relaxations were significantly enhanced. **P < 0.01, ***P < 0.0001 vs. WT. (d and e) Inhibitory effect of the combination of charybdotoxin (CTx) and apamin (Apm) on EDHF-mediated relaxations. The combination significantly inhibited EDHF-mediated relaxations in both strains. ^#P < 0.0001. (f) Relative contribution of EDHF to the endothelium-dependent relaxations of mesenteric arteries as evaluated by the inhibitory effects of charybdotoxin and apamin. EDHF-mediated relaxations were also significantly reduced in Cu,Zn-SOD^–/– mice. ^##P < 0.01.

Endothelium-independent relaxations. Endothelium-independent relaxations in response to sodium nitroprusside (Figure 3a) and to NS-1619 (Figure 3b) were comparable between the two strains. Furthermore, relaxations of endothelium-denuded mesenteric arteries in response to exogenous H₂O₂ were comparable between the two strains (Figure 3c).

Figure 3

Preserved vascular smooth muscle cell responses in Cu,Zn-SOD^–/– mice. (a) Endothelium-independent relaxations in response to sodium nitroprusside (SNP) of mesenteric arteries from control mice and Cu,Zn-SOD^–/– mice (n = 6–8). (b) Endothelium-independent relaxations in response to NS-1619 of mesenteric arteries from control mice and Cu,Zn-SOD^–/– mice (n = 6–8). Those endothelium-independent relaxations were comparable between the two strains. (c) Exogenous H₂O₂-induced relaxations of mesenteric arteries without endothelium (n = 6 each). H₂O₂ caused a comparable extent of endothelium-independent relaxations in the two strains.

Endothelium-dependent and -independent hyperpolarizations. Electrophysiological recordings of membrane potentials with the microelectrode technique in mesenteric arteries demonstrated that endothelium-dependent hyperpolarizations induced by ACh (10^–5 M) were significantly reduced in the Cu,Zn-SOD^–/– mice compared with control mice (Figure 4a). By contrast, endothelium-independent hyperpolarizations induced by levcromakalim were comparable between the two strains (Figure 4b).

Figure 4

Endothelium-dependent and -independent hyperpolarizations of mesenteric arteries. (a) Endothelium-dependent hyperpolarizations of mesenteric arteries in response to ACh (10^–5 M) were significantly attenuated in the Cu,Zn-SOD^–/– mice (n = 4–5). **P < 0.01 vs. WT. (b) Endothelium-independent hyperpolarizations of mesenteric arteries in response to levcromakalim (10^–5 M) were comparable between the two strains (n = 4–5).

Inhibitory effect of catalase on EDHF-mediated responses. In the presence of indomethacin (10^–5 M) and L-NNA (10^–4 M), catalase significantly inhibited EDHF-mediated relaxations of mesenteric arteries from control mice (Figure 5a). By contrast, catalase had no inhibitory effect on the EDHF-mediated relaxations of mesenteric arteries from Cu,Zn-SOD^–/– mice (Figure 5b). Catalase significantly inhibited the hyperpolarization response in control mice (Figure 5c) but not in Cu,Zn-SOD^–/– mice (Figure 5d).

Figure 5

Inhibitory effect of catalase on EDHF-mediated relaxations and hyperpolarizations of mesenteric arteries. (a) In mesenteric arteries from control mice, catalase (6,250 U/ml) in the presence of indomethacin and L-NNA significantly inhibited EDHF-mediated relaxations (n = 4 each). ***P < 0.0001. (b) In mesenteric arteries from Cu,Zn-SOD^–/– mice, catalase (6,250 U/ml) had no inhibitory effect on the remaining EDHF-mediated relaxations (n = 4 each). (c) Catalase significantly inhibited endothelium-dependent hyperpolarizations in response to ACh (10^–5 M) of mesenteric arteries from control mice (n = 3). *P < 0.05. (d) Catalase had no effect on endothelium-dependent hyperpolarization of mesenteric arteries from Cu,Zn-SOD^–/– mice (n = 3).

ACh-induced H₂O₂ production by endothelial cells. In the control mice, ACh significantly increased the fluorescence intensity in endothelial cells (Figure 6, a and b), while in the Cu,Zn-SOD^–/– mice, ACh-induced fluorescence intensity was significantly reduced compared with the control mice (Figure 6c). Figure 6d summarizes the relative fluorescence intensity in response to ACh in both strains.

Figure 6

ACh-induced production of H₂O₂ by the endothelium. H₂O₂ production by the endothelium was detected using confocal microscopy with DCF. (a) Basal fluorescence intensity. (b) Application of ACh significantly increased the fluorescence intensity in endothelial cells in a control mouse. (c) ACh-induced increase in fluorescence intensity was significantly reduced in Cu,Zn-SOD^–/– mice. (d) Summary of ACh-induced increase in fluorescence intensity in endothelial cells. ACh significantly increased the fluorescence intensity in control mice but not in Cu,Zn-SOD^–/– mice (n = 5 each). ^#P < 0.0001.

Coronary flow response in isolated heart. In the presence of indomethacin (10^–5 M) and L-NNA (10^–4 M), bradykinin-induced increase in coronary flow was 201% ± 9% of baseline in the control mice, whereas in the Cu,Zn-SOD^–/– mice, the response was significantly reduced at 151% ± 7% (Figure 7a). Catalase significantly inhibited the coronary flow response in the control mice (153% ± 7%) but not in the Cu,Zn-SOD^–/– mice (Figure 7a). Thus, the inhibitory effect of catalase on the coronary flow response was markedly reduced in the Cu,Zn-SOD^–/– mice compared with the control mice (Figure 7b).

Figure 7

Reduced coronary flow increase in Cu,Zn-SOD^–/– mice. (a) Coronary flow increase in response to bradykinin (BK) (10^–6 M) in isolated heart was significantly attenuated in the Cu,Zn-SOD^–/– mice compared with the control mice (n = 6 each). Catalase significantly inhibited the coronary flow increase in control mice, but not in Cu,Zn-SOD^–/– mice. *P < 0.05 vs. WT control. (b) The inhibitory effect of catalase on bradykinin-induced increase in coronary flow was absent in Cu,Zn-SOD^–/– mice. *P < 0.05 vs. WT.

Effect of SOD mimetic supplementation on EDHF-mediated relaxations in Cu,Zn-SOD–/– mice. Since SOD is a large molecule that penetrates the endothelium with difficulty, we used a cell-permeable SOD mimetic, tempol (27). In the presence of indomethacin (10^–5 M) and L-NNA (10^–4 M), tempol did not significantly improve EDHF-mediated relaxations of mesenteric arteries from control mice, and the relaxations were again significantly inhibited by catalase (Figure 8a). By contrast, in mesenteric arteries from Cu,Zn-SOD^–/– mice, tempol significantly improved the EDHF-mediated relaxations, and the enhancing effect of tempol was abolished by catalase (Figure 8b). Membrane potential recordings demonstrated that in control mice, tempol did not enhance endothelium-dependent hyperpolarizations, whereas in Cu,Zn-SOD^–/– mice, it significantly enhanced the responses (Figure 8c). Importantly, this enhancing effect of tempol was totally abolished by catalase. In isolated heart studies, tempol did not enhance the bradykinin-induced increase in coronary flow in the control mice, but significantly improved the response in the Cu,Zn-SOD^–/– mice (Figure 8d). In a separate experiment, the enhancing effect of tempol was abolished by catalase (control, 144% ± 8% vs. tempol plus catalase, 147% ± 9%, n = 3).

Figure 8

Effect of supplementation of SOD mimetic on EDHF-mediated relaxations in Cu,Zn-SOD^–/– mice. (a) In the presence of indomethacin and L-NNA, the SOD mimetic tempol (1 mM) did not improve the EDHF-mediated relaxations of mesenteric arteries from control mice, which were significantly inhibited by catalase (6,250 U/ml) (n = 6 each). (b) In mesenteric arteries from Cu,Zn-SOD^–/– mice, tempol (1 mM) in the presence of indomethacin and L-NNA significantly enhanced EDHF-mediated relaxations, an effect that was abolished by catalase (6,250 U/ml) (n = 5 each). ***P < 0.0001. (c) Tempol did not improve hyperpolarizations in control mice but did so in Cu,Zn-SOD–/– mice, and that effect was abolished by simultaneous administration of catalase (6,250 U/ml). (d) In isolated heart, tempol did not improve the coronary flow responses in control mice, but did so in Cu,Zn-SOD^–/– mice. *P < 0.05.

The novel finding of this study is that Cu,Zn-SOD plays a pivotal role in EDHF synthesis in both the mesenteric and the coronary microvessels in mice, further supporting our hypothesis that H₂O₂ is an EDHF. Although the role of Cu,Zn-SOD has been emphasized only for its effect of scavenging superoxide anions (19, 20), its biological role in maintaining vascular homeostasis remains to be fully elucidated. The present study demonstrates a novel role of Cu,Zn-SOD as an “EDHF synthase” in maintaining vascular homeostasis.

Characterization of Cu,Zn-SOD–/– mice. In Cu,Zn-SOD^–/– mice, the expression of Cu,Zn-SOD protein was absent, while there was no upregulation of the other two SODs (MnSOD and ecSOD). Furthermore, the activities of glutathione peroxidase and catalase were comparable between the control and the Cu,Zn-SOD^–/– mice, suggesting that these endogenous H₂O₂-scavenging enzymes are not involved in the reduced EDHF responses in Cu,Zn-SOD^–/– mice. Regarding the other two isoforms of SOD, a role for MnSOD in synthesizing EDHF/H₂O₂ is less likely because MnSOD is located in mitochondria (21) and thus does not seem to be involved in EDHF synthesis. We have also ruled out the possible role of ecSOD, because EDHF-mediated responses were not altered by heparin (unpublished observations).

Reduced EDHF-mediated responses in Cu,Zn-SOD–/– mice. Upregulated NO compensates the reduced contribution of EDHF to endothelium-dependent relaxations in Cu,Zn-SOD^–/– mice, confirming the importance of endothelial Cu,Zn-SOD as an EDHF synthase. EDHF-mediated relaxations were predominant in small mesenteric arteries, a finding consistent with our previous studies (2, 3, 30). EDHF-mediated relaxations, as evaluated by the combination of charybdotoxin and apamin, also were reduced in Cu,Zn-SOD^–/– mice, further confirming the reduced EDHF response in those mice. Both the expression and the activity of Cu,Zn-SOD are enhanced by fluid shear stress in endothelial cells but not in smooth muscle cells in the human aorta (31). This finding is consistent with our present study because shear stress stimulates the synthesis of EDHF by the endothelium (32).

In the carotid artery of Cu,Zn-SOD^–/– mice, NO-mediated relaxations were attenuated and contractile responses of vascular smooth muscle to serotonin and phenylephrine were enhanced (23). In the present study, however, NO-mediated relaxations were enhanced to partially compensate for the endothelium-dependent relaxations in mesenteric arteries. It is unlikely that upregulated eNOS is involved in this compensation because eNOS expression was unaltered in Cu,Zn-SOD^–/– mice. Although the mechanism for the upregulation of NO-mediated responses remains to be elucidated, this could explain, at least in part, the absence of hypertension in those mice.

Several abnormalities have been reported in the cardiovascular system of Cu,Zn-SOD^–/– mice. In those mice, myocardial ischemia-reperfusion injury was enhanced (33), as was cerebral injury after transient ischemia (34). It remains to be elucidated to what extent Cu,Zn-SOD contributes to those abnormalities as an EDHF synthase in Cu,Zn-SOD^–/– mice.

H₂O₂ plays an important role in many blood vessels under physiological conditions as an endogenous relaxing and hyperpolarizing factor. Indeed, H₂O₂ causes relaxations and hyperpolarizations (3, 12, 13, 16, 35, 36). We were recently able to demonstrate the endothelial production of H₂O₂ as an endogenous EDHF in porcine coronary microvessels by an electron spin resonance method (13). It has recently been reported that endothelial production of superoxide anions from eNOS and the subsequent synthesis of H₂O₂ are increased in the aorta of deoxycorticosterone acetate–salt hypertensive rats and that this eNOS-derived H₂O₂ is involved in endothelium-dependent relaxations in response to calcium ionophore A23187 (37). The increase in endothelial H₂O₂ production was due to oxidation of tetrahydrobiopterin that resulted in eNOS uncoupling (37). However, endothelial production of superoxide anions from eNOS is noted even under physiological conditions without deficiencies of tetrahydrobiopterin (18). Thus, we consider that endothelial production of superoxide anions by eNOS and subsequent synthesis of H₂O₂ by Cu,Zn-SOD are physiological phenomena and play a pivotal role in maintaining vascular homeostasis.

Preserved vascular smooth muscle responses in Cu,Zn-SOD–/– mice. Endothelium-independent relaxations in response to sodium nitroprusside, NS-1619, and exogenous H₂O₂ were comparable between the two strains and endothelium-independent hyperpolarizations induced by levcromakalim also were unchanged in Cu,Zn-SOD^–/– mice, suggesting that relaxation properties of vascular smooth muscle cells were preserved in those mice.

Decreased inhibitory effect of catalase on EDHF-mediated relaxations and hyperpolarizations in Cu,Zn-SOD–/– mice. In mesenteric arteries from control mice, pretreatment with catalase significantly inhibited EDHF-mediated relaxations, suggesting that those relaxations were mediated by H₂O₂ in control mice, a finding consistent with our previous studies (3). By contrast, in mesenteric arteries from Cu,Zn-SOD^–/– mice, the inhibitory effect of catalase was absent, suggesting that the contribution of H₂O₂ to EDHF-mediated relaxations is markedly reduced in those mice. Catalase also significantly inhibited endothelium-dependent hyperpolarization in control mice, while it had no inhibitory effect in Cu,Zn-SOD^–/– mice. These results further support our hypothesis that endothelial Cu,Zn-SOD plays a pivotal role in the production of H₂O₂/EDHF.

Decreased formation of ACh-induced H₂O₂ in Cu,Zn-SOD–/– mice. Microscopy of DCF-stained vascular endothelium demonstrated that ACh-induced fluorescence intensity was significantly reduced in Cu,Zn-SOD^–/– mice compared with control mice. This suggests that ACh-induced endothelial production of H₂O₂ was significantly decreased in Cu,Zn-SOD^–/– mice, further confirming the importance of Cu,Zn-SOD in endothelial synthesis of H₂O₂/EDHF.

Reduced coronary flow response in Cu,Zn-SOD–/– mice. In Cu,Zn-SOD^–/– mice, EDHF-mediated increase in coronary flow was also significantly reduced, and the remaining coronary flow was no longer sensitive to catalase. These results indicate that catalase-sensitive, H₂O₂-mediated relaxations of coronary microvessels were almost absent in Cu,Zn-SOD^–/– mice, confirming the importance of Cu,Zn-SOD in the coronary circulation.

Improvement of EDHF-mediated responses by an SOD mimetic in Cu,Zn-SOD–/– mice. Since SOD is a large molecule and penetrates the endothelium only with difficulty, we used the cell-permeable SOD mimetic tempol. Tempol is a stable nitroxide and has been reported to act as an SOD mimetic and to dismutate superoxide into H₂O₂ (27, 38). In the present study, tempol significantly improved the EDHF-mediated relaxations and hyperpolarizations of mesenteric arteries and also significantly improved the coronary flow response to bradykinin in Cu,Zn-SOD^–/– mice, further confirming the pivotal role of the enzyme in endothelium-dependent hyperpolarization. Catalase completely inhibited this effect of tempol, indicating that the effect of tempol was mediated by H₂O₂. By contrast, tempol had no augmenting effect on the EDHF-mediated relaxations and hyperpolarizations of mesenteric arteries and coronary flow responses in control mice, suggesting that a sufficient amount of SOD is present in those mice.

Limitations of the study. Several limitations of the present study should be mentioned. First, we did not demonstrate the endothelial source of superoxide anions that are dismutated into H₂O₂. Since EDHF-mediated relaxations are significantly attenuated in eNOS^–/– mice, it is possible that eNOS is one of the important sources of superoxide anions (3). Other possible sources of superoxide anions and the details of the interaction between eNOS and other enzymes remain to be examined in future studies. Second, while we were able to demonstrate the reduced H₂O₂ production by the endothelium in Cu,Zn-SOD^–/– mice using confocal microscopy with DCF, we were unable to quantitatively measure endothelial H₂O₂ production because a sufficient amount of mesenteric microvessels was not available in mice. However, we have recently demonstrated with an electron spin resonance method that endothelial cells produce H₂O₂, at least in micromolar quantities, in porcine coronary microvessels under physiological conditions (13). Third, the mechanisms for the compensatory role of NO in Cu,Zn-SOD^–/– mice remain to be elucidated. Fourth, we did not elucidate the mechanisms for the remaining EDHF-mediated responses in Cu,Zn-SOD–/– mice. Fifth, although we demonstrated the importance of Cu,Zn-SOD as an EDHF synthase in mesenteric arteries in vitro and in the coronary circulation ex vivo, the importance of the enzyme in EDHF-mediated responses in vivo remains to be examined.

Clinical implications. EDHF plays an important role in human arteries, especially in microvessels (39, 40). We have recently observed that SOD mimetics also improve EDHF-mediated relaxations and hyperpolarizations in human mesenteric arteries (unpublished observations). It has been reported that abnormality in Cu,Zn-SOD processing and mutation of genes encoding Cu,Zn-SOD is involved in the pathogenesis of several diseases, including diabetes mellitus, hypertension, amyotrophic lateral sclerosis, Down syndrome, and malignant breast diseases (41–44). It has also been reported that Cu,Zn-SOD is important for ovary function (45) and hearing function (46). EDHF-mediated relaxations are impaired in several disease states, including hypercholesterolemia and aging (39), which may promote the progression of vascular disease, especially at microvascular levels. It remains to be examined to what extent Cu,Zn-SOD contributes to the above-mentioned physiological and pathophysiological conditions as an EDHF synthase.

In summary, the present study provides the novel concept that endothelial Cu,Zn-SOD plays a pivotal role as an EDHF synthase.

We thank K. Takeshige of Kyushu University Graduate School of Medical Sciences and H. Utsumi of Kyushu University Graduate School of Pharmaceutical Sciences for valuable comments on our work, and I. Kunihiro, M. Motoishi, and Y. Matsuo for excellent technical assistance in this study. This work was supported in part by grants-in-aid (07307010, 10877111, 12877114, and 13670724) from the Japanese Ministry of Education, Science, Sports, and Culture (Tokyo, Japan) and from the Mitsui Life Social Welfare Foundation (Tokyo, Japan).

Conflict of interest: The authors have declared that no conflict of interest exists.

Nonstandard abbreviations used: endothelium-derived hyperpolarizing factor (EDHF); superoxide dismutase (SOD); copper,zinc-SOD (Cu,Zn-SOD); manganese SOD (MnSOD); extracellular SOD (ecSOD); prostaglandin F_2α (PGF_2α); acetylcholine (ACh); calcium-activated potassium channels (KCa channels); N^ω-nitro-l-arginine (L-NNA); 2′,7′-dichlorodihydrofluorescein diacetate (DCF).

1. Mombouli, JV, Vanhoutte, PM. Endothelium-derived hyperpolarizing factor(s): updating the unknown. Trends Pharmacol. Sci. 1997. 18:252-256.
   View this article via: PubMed Google Scholar
2. Shimokawa, H. Primary endothelial dysfunction: atherosclerosis. J. Mol. Cell. Cardiol. 1999. 31:23-37.
   View this article via: PubMed CrossRef Google Scholar
3. Matoba, T, et al. Hydrogen peroxide is an endothelium-derived hyperpolarizing factor in mice. J. Clin. Invest. 2000. 106:1521-1530.
   View this article via: JCI PubMed CrossRef Google Scholar
4. Feletou, M, Vanhoutte, PM. Endothelium-dependent hyperpolarization of canine coronary smooth muscle. Br. J. Pharmacol. 1988. 93:515-524.
   View this article via: PubMed Google Scholar
5. Chen, G, Suzuki, H, Weston, AH. Acetylcholine releases endothelium-derived hyperpolarizing factor and EDRF from rat blood vessels. Br. J. Pharmacol. 1988. 95:1165-1174.
   View this article via: PubMed Google Scholar
6. 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.
   View this article via: PubMed Google Scholar
7. Fisslthaler, B, et al. Cytochrome P450 2C is an EDHF synthase in coronary arteries. Nature. 1999. 401:493-497.
   View this article via: PubMed CrossRef Google Scholar
8. Edwards, G, Dora, KA, Gardener, MJ, Garland, CJ, Weston, AH. K+ is an endothelium-derived hyperpolarizing factor in rat arteries. Nature. 1998. 396:269-272.
   View this article via: PubMed CrossRef Google Scholar
9. Beny, JL, Schaad, O. An evaluation of potassium ions as endothelium-derived hyperpolarizing factor in porcine coronary arteries. Br. J. Pharmacol. 2000. 131:965-973.
   View this article via: PubMed CrossRef Google Scholar
10. Taylor, HJ, Chaytor, AT, Evans, WH, Griffith, TM. Inhibition of the gap junctional component of endothelium-dependent relaxations in rabbit iliac artery by 18-alpha glycyrrhetinic acid. Br. J. Pharmacol. 1998. 125:1-3.
    View this article via: PubMed CrossRef Google Scholar
11. Yamamoto, Y, Fukuta, H, Nakahira, Y, Suzuki, H. Blockade by 18beta-glycyrrhetinic acid of intercellular electrical coupling in guinea-pig arterioles. J. Physiol. 1998. 511:501-508.
    View this article via: PubMed CrossRef Google Scholar
12. Matoba, T, et al. Hydrogen peroxide is an endothelium-derived hyperpolarizing factor in human mesenteric arteries. Biochem. Biophys. Res. Commun. 2002. 290:909-913.
    View this article via: PubMed CrossRef Google Scholar
13. Matoba, T, et al. Electron spin resonance detection of hydrogen peroxide as an endothelium-derived hyperpolarizing factor in porcine coronary microvessels. Arterioscler. Thromb. Vasc. Biol. 2003. 23:1224-1230.
    View this article via: PubMed CrossRef Google Scholar
14. Miura, H, et al. Role for hydrogen peroxide in flow-induced dilation of human coronary arterioles. Circ. Res. 2003. 92:e31-e40.
    View this article via: PubMed CrossRef Google Scholar
15. Yada, T, et al. Hydrogen peroxide, an endogenous endothelium-derived hyperpolarizing factor, plays an important role in coronary autoregulation in vivo. Circulation. 2003. 107:1040-1045.
    View this article via: PubMed CrossRef Google Scholar
16. Lacza, Z, et al. 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.
    View this article via: PubMed Google Scholar
17. Okado-Matsumoto, A, Fridovich, I. Subcellular distribution of superoxide dismutases (SOD) in rat liver: Cu,Zn-SOD in mitochondria. J. Biol. Chem. 2001. 276:38388-38393.
    View this article via: PubMed CrossRef Google Scholar
18. Stuehr, D, Pou, S, Rosen, GM. Oxygen reduction by nitric-oxide synthases. J. Biol. Chem. 2001. 276:14533-14536.
    View this article via: PubMed CrossRef Google Scholar
19. Rubanyi, GM, Vanhoutte, PM. Superoxide anions and hyperoxia inactivate endothelium-derived relaxing factor. Am. J. Physiol. 1986. 250:H822-H827.
    View this article via: PubMed Google Scholar
20. Mugge, A, Elwell, JH, Peterson, TE, Harrison, DG. Release of intact endothelium-derived relaxing factor depends on endothelial superoxide dismutase activity. Am. J. Physiol. 1991. 260:C219-C225.
    View this article via: PubMed Google Scholar
21. McIntyre, M, Bohr, DF, Dominiczak, AF. Endothelial function in hypertension: the role of superoxide anion. Hypertension. 1999. 34:539-545.
    View this article via: PubMed Google Scholar
22. Fukai, T, et al. Regulation of the vascular extracellular superoxide dismutase by nitric oxide and exercise training. J. Clin. Invest. 2000. 105:1631-1639.
    View this article via: JCI PubMed CrossRef Google Scholar
23. Didion, SP, et al. Increased superoxide and vascular dysfunction in CuZnSOD-deficient mice. Circ. Res. 2002. 91:938-944.
    View this article via: PubMed CrossRef Google Scholar
24. Beauchamp, C, Fridovich, I. Superoxide dismutase: improved assays and an assay applicable to acrylamide gels. Anal. Biochem. 1971. 44:276-287.
    View this article via: PubMed CrossRef Google Scholar
25. Wheeler, CR, Salzman, JA, Elsayed, NM, Omaye, ST, Korte (Jr), DW. Automated assays for superoxide dismutase, catalase, glutathione peroxidase, and glutathione reductase activity. Anal. Biochem. 1990. 184:193-199.
    View this article via: PubMed CrossRef Google Scholar
26. Akaike, T, et al. Antagonistic action of imidazolineoxyl N-oxides against endothelium-derived relaxing factor/.NO through a radical reaction. Biochemistry. 1993. 32:827-832.
    View this article via: PubMed CrossRef Google Scholar
27. Krishna, MC, Grahame, DA, Samuni, A, Mitchell, JB, Russo, A. Oxoammonium cation intermediate in the nitroxide-catalyzed dismutation of superoxide. Proc. Natl. Acad. Sci. U. S. A. 1992. 89:5537-5541.
    View this article via: PubMed CrossRef Google Scholar
28. Headrick, JP, Gauthier, NS, Morrison, R, Matherne, GP. Cardioprotection by K(ATP) channels in wild-type hearts and hearts overexpressing A(1)-adenosine receptors. Am. J. Physiol. Heart Circ. Physiol. 2000. 279:H1690-H1697.
    View this article via: PubMed Google Scholar
29. Talukder, MA, et al. Targeted deletion of adenosine A(3) receptors augments adenosine-induced coronary flow in isolated mouse heart. Am. J. Physiol. Heart Circ. Physiol. 2002. 282:H2183-H2189.
    View this article via: PubMed Google Scholar
30. Shimokawa, H, et al. The importance of the hyperpolarizing mechanism increases as the vessel size decreases in endothelium-dependent relaxations in rat mesenteric circulation. J. Cardiovasc. Pharmacol. 1996. 28:703-711.
    View this article via: PubMed CrossRef Google Scholar
31. Inoue, N, Ramasamy, S, Fukai, T, Nerem, RM, Harrison, DG. Shear stress modulates expression of Cu/Zn superoxide dismutase in human aortic endothelial cells. Circ. Res. 1996. 79:32-37.
    View this article via: PubMed Google Scholar
32. Takamura, Y, et al. Important role of endothelium-derived hyperpolarizing factor in shear stress–induced endothelium-dependent relaxations in the rat mesenteric artery. J. Cardiovasc. Pharmacol. 1999. 34:381-387.
    View this article via: PubMed CrossRef Google Scholar
33. Yoshida, T, Maulik, N, Engelman, RM, Ho, YS, Das, DK. Targeted disruption of the mouse Sod I gene makes the hearts vulnerable to ischemic reperfusion injury. Circ. Res. 2000. 86:264-269.
    View this article via: PubMed Google Scholar
34. Kawase, M, et al. Exacerbation of delayed cell injury after transient global ischemia in mutant mice with CuZn superoxide dismutase deficiency. Stroke. 1999. 30:1962-1968.
    View this article via: PubMed Google Scholar
35. Rubanyi, GM, Vanhoutte, PM. Oxygen-derived free radicals, endothelium, and responsiveness of vascular smooth muscle. Am. J. Physiol. 1986. 250:H815-H821.
    View this article via: PubMed Google Scholar
36. Beny, JL, von der Weid, PY. Hydrogen peroxide: an endogenous smooth muscle cell hyperpolarizing factor. Biochem. Biophys. Res. Commun. 1991. 176:378-384.
    View this article via: PubMed CrossRef Google Scholar
37. Landmesser, U, et al. Oxidation of tetrahydrobiopterin leads to uncoupling of endothelial cell nitric oxide synthase in hypertension. J. Clin. Invest. 2003. 111:1201-1209. doi:10.1172/JCI200314172.
    View this article via: JCI PubMed Google Scholar
38. Krishna, MC, et al. Do nitroxide antioxidants act as scavengers of O2-. or as SOD mimics? J. Biol. Chem. 1996. 271:26026-26031.
    View this article via: PubMed CrossRef Google Scholar
39. Urakami-Harasawa, L, Shimokawa, H, Nakashima, M, Egashira, K, Takeshita, A. Importance of endothelium-derived hyperpolarizing factor in human arteries. J. Clin. Invest. 1997. 100:2793-2799.
    View this article via: JCI PubMed CrossRef Google Scholar
40. Inokuchi, K, et al. Role of endothelium-derived hyperpolarizing factor in human forearm circulation. Hypertension. 2003. 42:919-924.
    View this article via: PubMed CrossRef Google Scholar
41. Uchimura, K, et al. Changes in superoxide dismutase activities and concentrations and myeloperoxidase activities in leukocytes from patients with diabetes mellitus. J. Diabetes Complications. 1999. 13:264-270.
    View this article via: PubMed CrossRef Google Scholar
42. Noor, R, Mittal, S, Iqbal, J. Superoxide dismutase—applications and relevance to human diseases. Med. Sci. Monit. 2002. 8:RA210-RA215.
    View this article via: PubMed Google Scholar
43. Wu, R, Millette, E, Wu, L, de Champlain, J. Enhanced superoxide anion formation in vascular tissues from spontaneously hypertensive and desoxycorticosterone acetate-salt hypertensive rats. J. Hypertens. 2001. 19:741-748.
    View this article via: PubMed CrossRef Google Scholar
44. Druzhyna, N, Nair, RG, LeDoux, SP, Wilson, GL. Defective repair of oxidative damage in mitochondrial DNA in Down’s syndrome. Mutat. Res. 1998. 409:81-89.
    View this article via: PubMed Google Scholar
45. Matzuk, MM, Dionne, L, Guo, Q, Kumar, TR, Lebovitz, RM. Ovarian function in superoxide dismutase 1 and 2 knockout mice. Endocrinology. 1998. 139:4008-4011.
    View this article via: PubMed CrossRef Google Scholar
46. McFadden, SL, Ding, D, Reaume, AG, Flood, DG, Salvi, RJ. Age-related cochlear hair cell loss is enhanced in mice lacking copper/zinc superoxide dismutase. Neurobiol. Aging. 1999. 20:1-8.
    View this article via: PubMed CrossRef Google Scholar