There is evidence that reactive nitrogen species are implicated in diabetic vascular complications, but their sources and targets remain largely unidentified. In the present study, we aimed to study the roles of endothelial nitric oxide synthase (eNOS) in diabetes. Exposure of isolated bovine coronary arteries to high glucose (30 mmol/l d-glucose) but not to osmotic control mannitol (30 mmol/l) switched angiotensin II–stimulated prostacyclin (PGI2)-dependent relaxation into a persistent vasoconstriction that was sensitive to either indomethacin, a cyclooxygenase inhibitor, or SQ29548, a selective thromboxane receptor antagonist. In parallel, high glucose, but not mannitol, significantly increased superoxide and 3-nitrotyrosine in PGI2 synthase (PGIS). Concurrent administration of polyethylene-glycolated superoxide dismutase (SOD), l-nitroarginine methyl ester, or sepiapterin not only reversed the effects of high glucose on both angiotensin II–induced relaxation and PGI2 release but also abolished high-glucose–enhanced PGIS nitration, as well as its association with eNOS. Furthermore, diabetes significantly suppressed PGIS activity in parallel with increased superoxide and PGIS nitration in the aortas of diabetic C57BL6 mice but had less effect in diabetic mice either lacking eNOS or overexpressing human SOD (hSOD+/+), suggesting an eNOS-dependent PGIS nitration in vivo. We conclude that diabetes increases PGIS nitration in vivo, likely via dysfunctional eNOS.

Endothelial dysfunction is a hallmark of vascular injury in diabetes and is preceded by the development of overt cardiovascular diseases (13). There is an overwhelming mass of evidence demonstrating the development of endothelial dysfunction in animal models of diabetes and in human blood vessels from diabetic patients, as evidenced by increased release of reactive oxygen species, decreased nitric oxide (NO) bioactivity, decreased release of prostacyclin (PGI2), and enhanced endothelial production of vasoconstrictor thromboxane (Tx)A2/prostaglandin (PG)H2 in early stages of diabetes (13). The net effect of endothelial dysfunction is vascular damage, which is responsible for complications in both types of diabetes.

Evidence has accumulated indicating that the generation of reactive oxygen and nitrogen species plays an important role in the etiology of diabetes complications. Our previous studies (4,5) and others (6,7) have shown that exposure of human aortic endothelial cells to high glucose leads to augmented production of superoxide anion (O2.−), which may quench NO, thereby reducing the efficacy of this potent endothelium-derived vasodilator system and generating toxic oxidant species, such as peroxynitrite (ONOO) (4,5). ONOO is a highly reactive species and can initiate both nitrosative and oxidative reactions in vitro and in vivo (810). A characteristic reaction of ONOO is the nitration of protein-bound tyrosine residues to generate 3-nitrotyrosine–positive proteins (810).

PGI2 synthase (PGIS; EC5.3.99.4) catalyzes the conversion of PGH2 into PGI2 (11,12). The limiting step of PGI2 is the formation of the PG endoperoxide, PGH2, by the two rate-limiting cyclooxygenases (COXs), COX-1 and -2. In blood vessels, particularly larger arteries, the predominant PG is PGI2. PGI2 relaxes isolated vascular strips and is a potent endogenous inhibitor for platelet and leukocyte activation. PGI2 not only prevents platelets from sticking together but also disperses existing aggregates in vitro and in the circulation of humans (11,12). The effects of PGI2 are opposed by TxA2, a major product of platelets (10,11). Suppressed PGI2 (1215) and/or increased TxA2 or PGH2 (1619) have been reported in the early stage of diabetes (1,4). For example, PGI2 production by blood vessels from patients with diabetes is depressed and urinary and circulating levels of 6-keto-PGF1α are reduced in patients with proliferative retinopathy (13,14). Furthermore, decreased PGI2 has been linked to platelet hyperaggregability, increased adhesiveness, and increased release of PGH2/TxA2 in diabetic patients (13,14). However, how diabetes leads to an imbalance of TxA2/PGI2 remains elusive.

Our previous studies (2024) had demonstrated that ONOO, when given exogenously or when produced endogenously, is able to induce endothelial dysfunction via a mechanism dependent on the inhibition and nitration of PGIS at tyrosine 430 (25) and consequent stimulation of Tx receptor (TPr) (5,22). In the present study, we hypothesized that high glucose might, via ONOO, alter the balance of PGI2/TxA2 by tyrosine nitration of PGIS. Using isolated bovine coronary arteries (BCAs) and streptozotocin (STZ)-induced diabetic mice, we have, for the first time, demonstrated an endothelial nitric oxide synthase (eNOS)-mediated PGIS nitration in diabetes in vivo.

Male mice overexpressing human Cu,Zn superoxide dismutase (SOD) (hSOD-TG; TgHS-51), eNOS knockout mice (eNOS−/−), and their littermates, C57BL6 mice, 7 weeks of age, were obtained from The Jackson Laboratory (Bar Harbor, ME). Mice were housed in temperature-controlled cages with a 12-h light/dark cycle and given free access to water and normal diet. Mice were randomly divided into control or diabetic groups. STZ was prepared in 0.05 mol/l citrate buffer, pH 4.5 (50 mg · kg−1 · day−1 for both C57BL6 and eNOS−/− but 70 mg · kg−1 · day−1 for SOD-TG), and was intraperitoneally injected each day for 5 days. Control nondiabetic animals received the citrate buffer (pH 4.5) solvent. The diabetic state was confirmed the following week by glucose determination on tail vein blood. Body weight, water, and food intake were monitored routinely. Two weeks after STZ injection, the mice were killed with inhaled isoflorane. Mice hearts and aortas were removed and immediately frozen in liquid nitrogen. The animal protocol was reviewed and approved by the Institute Animal Care and Use Committee.

l-nitroarginine methyl ester (l-NAME), PGH2, 1S[1α, 2B(5Z),3b,4α]-7-[3-[2(phenylamino) carbonyl] hydrazino methyl [-7-oxabicyclo(2.2.1)hept-2-yl-5-heptenoic acid] (SQ29548), indomethacin, sepiapterin, and the enzyme-linked immunoassay kits for 6-keto-PGF1α, TxB2, PGF2α, and PGE2 were obtained from Cayman Chemicals (Ann Arbor, MI). Protein-A sepharose CL-4B was obtained from Pharmacia. A monoclonal antibody against 3-nitrotyrosine was purchased from Upstate Biotechnology Incorporated (Waltham, MA). Antibodies against eNOS and caveolin-1 were obtained from Transduction Laboratory. Rabbit anti-PGIS antisera were kindly provided by Dr. T. Klein (Altana Pharm, Konstanz, Germany). Secondary antibodies were from Pierce. Enhanced chemiluminescence kits and nitrocellulose membranes (Hybond-C) were purchased from Amersham. Other chemicals, if not indicated, were acquired from Sigma (St. Louis, MO) with highest quality.

Isolation and isometric measurement of tension in BCAs.

BCAs of the left ventricle were isolated and assayed as described previously (2224). Briefly, experiments were started by obtaining, from each spiral, a reference response of vasoconstriction relaxation and prostanoid release after addition of angiotensin II (50 nmol/l) over a time period of 30 min. Subsequently, the spiral was rinsed several times with prewarmed (37°C) Krebs-Ringer buffer and supplemented with normal glucose (5.5 mmol/l), high glucose (30 mmol/l), or mannitol (24.5 mmol/l mannitol plus 5.5 mmol/l glucose). At the times indicated, incubation was terminated by removing glucose-containing medium and rinsed several times with Krebs-Ringer buffer. After the tension returned to the baseline, the vessel was stimulated with the same concentration of angiotensin II. Indomethacin (10−5 mol/l) or nonselective NOS inhibitor, l-NAME (10−4 mol/l), or polyethylene-glycolated (PEG)-SOD (300 units/ml) were added during incubation and supplemented to the organ bath after removing elevated glucose- or mannitol-containing medium, as well as during the second stimulation with angiotensin II. The media from the first and the second stimulation with angiotensin II were collected and stored at −20°C for prostanoid analysis using enzyme-linked immunoassay kits according to the manufacturer’s instructions. During the experiments, care had been taken to avoid any injury to the endothelium. In some experiments, the endothelial layer was deliberately removed after dissection by intraluminal perfusion with 0.5% 3-[(3-cholamidopropyl)-dimethyl-ammonio]-1 propane sulfonate in Krebs-Ringer buffer for 40 s followed by repeated washings with Krebs-Ringer buffer.

Assay of PGIS activity.

PGIS activity was assayed by the stable metabolite of PGI2, 6-keto-PGF1α, after incubating cells with its substrate, PGH2 (10−5 mol/l for 3 min), as described previously (2225).

Detection of aortic O2.−.

Aortic O2.− in isolated mouse aorta was measured by lucigenin chemiluminescence (5 μmol/l). Measurements on intact arteries with lucigenin at 5 μmol/l have been corroborated with electron spin resonance and were not complicated by its redox cycling (26).

High-performance liquid chromatography detection of 3-nitrotyrosine.

Aortic proteins were isolated and hydrolyzed by pronase digestion (72 h at 56°C). 3-nitrotyrosine was quantified by a high-performance liquid chromatography/ultraviolet detector coupled with electrochemical detections, as previously described (2224).

Immunoprecipation and Western blots.

Immunoprecipitation and Western blots were performed as described previously (2125).

Quantification of Western blot.

The intensity (area × density) of the individual bands on Western blots was measured by densitometry (Model GS-700, Imaging Densitometer; Bio-Rad). The background was subtracted from the calculated area.

Statistical analysis.

Results were analyzed by using a two-way ANOVA. Values are expressed as means ± SD for the number of assays. A P value <0.05 was considered statistically significant.

Hyperglycemia blunts PGI2-dependent relaxation and releases PGH2-mediated vasospasm in BCAs.

We have shown (2224) that in BCAs, PGI2 is responsible for ∼80% of angiotensin II–induced relaxation, whereas NO accounts for ∼20%. Stimulation of BCAs with angiotensin II (50 nmol/l) triggered a rapid rise of tension (constriction) followed by a PGI2-mediated relaxation (2224). Thus, BCAs appeared to be an ideal system for studying the effects of high glucose on PGI2-dependent relaxation.

Using this well-characterized system, we next investigated whether high glucose mimicked the effects of ONOO in vivo. Freshly isolated BCAs were exposed to normal glucose (control, 5 mmol/l d-glucose), high glucose (30 mmol/l d-glucose), or osmotic control (5 mmol/l d-glucose plus mannitol) for 0.5–4 h. As shown in Fig. 1B, high glucose did not alter angiotensin II–triggered vasoconstriction. Similar to the vessels exposed to submicromolar concentrations of ONOO (22), exposure of BCAs to high glucose for 4 h significantly impaired angiotensin II–induced relaxation (Fig. 1A and B) and subsequently triggered a sustained vasoconstriction after a transient and small relaxation (Fig. 1B).

We next investigated if high-glucose–impaired relaxation was due to a reduction of PGI2, the major vasodilator in BCA. Short exposure of BCAs to high glucose (<30 min) slightly increased the release of 6-keto-PGF1α (data not shown). In parallel with impaired relaxation, prolonged exposure (1–4 h) of BCAs to high glucose significantly inhibited angiotensin II–induced PGI2 release (Fig. 1A). In contrast, exposure of BCAs to mannitol for 4 h did not inhibit angiotensin II–induced PGI2 release (Fig. 1A).

We next determined if the second vasospasm caused by high glucose was due to an overproduction of PGH2, which acts upon TPr (27,28). As shown in Fig. 1B, concurrent administration of SQ29548 (10 μmol/l), a potent TxA2/PGH2 receptor antagonist, and high glucose abolished high-glucose–induced vasospasms. Furthermore, both indomethacin, a COX inhibitor, and SQ29548 restored angiotensin II–induced relaxation without altering PGI2 reactivity. However, CGS13080, a potent TxA2 synthase inhibitor, had no effect (Fig. 1C). Taken together, these results suggest that high glucose increased the release of COX-derived products, likely PGH2, resulting in vasospasm.

Our previous studies demonstrated that tyrosine nitration and inactivation of PGIS leads to an accumulation of PGH2, resulting in vasospasm by TPr activation in isolated BCAs exposed to either ONOO (22) or hypoxia reoxygenation (23). Although both indomethacin and TPr blockade were effective in preventing vasospasm in vascular rings exposed to high glucose indicating PGH2 (Fig. 1B and C), a quantification of this precursor is necessary because other prostanoid and hydroxyeicosatetraenoic acids are capable of stimulating TPr.

PGH2 is an unstable prostaglandin that is converted into PGF2α by mild reducing agents such as SnCl2 (22,23). Therefore, to estimate the PGH2 released, we calculated the difference between the PGF2α peak of the samples from cells in the presence of 200 μg/mL SnCl2. As depicted in Fig. 1D, the levels of PGF2α were significantly elevated in BCAs exposed to high glucose compared with BCAs with either 5 mmol/l glucose or mannitol, which indicated that high glucose increased the release of PGH2 in isolated BCAs.

Together with other PGs and Tx, both TxA2 and PGI2 are produced by the two rate-limiting COXs, COX-1 and -2, which form the PG endoperoxide PGG2. Therefore, we next determined if the reduction of PGI2 by high glucose was due to reduction of either COX-1 or -2. As shown in Fig. 2A, neither high d-glucose nor mannitol altered the levels of COX-1. Conversely, high glucose but not mannitol increased the detection of COX-2 (Fig. 2A). Furthermore, exposure of BCAs to high glucose significantly increased PGE2, whereas the levels of both PGF2α and TxB2 remained unaffected (Fig. 2B). This suggested that high glucose might increase the levels and activity of COX-2. Thus, high-glucose–induced PGI2 inhibition was not due to a reduction of either COX-1 or -2.

ONOO-dependent tyrosine nitration of PGI2 synthase.

We next determined if high glucose increased PGIS nitration in BCAs. To this end, PGIS was first immunoprecipitated with a polyclonal antibody against PGIS and then Western blotted with a monoclonal antibody against 3-nitrotyrosine. Compared with a weak staining in BCAs exposed to 5 mmol/l glucose or mannitol, high glucose significantly increased PGIS nitration (Fig. 3A and B).

To further identify the source(s) of O2.− and ONOO, we determined the levels of PGIS nitration in presence of PEG-SOD; apocynin, a selective inhibitor for NAD(P)H oxidase assembly (29); allupurinol, a potent inhibitor for xanthine oxidase (30); or l-NAME, a nonselective inhibitor for NOS. As shown in Fig. 3A and B, concurrent administration of both l-NAME and PEG-SOD significantly suppressed high-glucose–enhanced PGIS nitration, suggesting that ONOO was involved in high-glucose–enhanced PGIS nitration. In contrast, neither allopurinol nor apocynin altered high-glucose–enhanced PGIS nitration (Fig. 3A and B), implying that eNOS might be responsible for high-glucose–enhanced PGIS nitration.

Identification of eNOS as the source of O2.−.

We next determined the source of O2.− and ONOO in BCAs exposed to high glucose. As shown in Fig. 3C, high glucose but not mannitol significantly increased the release of O2.−. Cocurrent administration of apocynin, allopurinol, or rotenone did not alter high-glucose–enhanced O2.− production. In contrast, either tiron, a potent O2.− scavenger, or l-NAME abolished high-glucose–enhanced O2.− (Fig. 3C). Theses results coincided with the selective inhibition of l-NAME on high-glucose–enhanced PGIS nitration (Fig. 3A and B), confirming that eNOS might be a source of O2.− and ONOO.

Effects of elevated glucose on PGIS nitration and O2.− in endothelium-denuded vessels.

Further evidence for eNOS-derived ONOO came from endothelium-denuded BCAs. To establish the contributions of eNOS in high-glucose–enhanced oxidant stress and PGIS nitration, endothelium was mechanically removed to delete eNOS located within endothelium. Removal of the endothelium blunted angiotensin II–induced PGI2 release (−91 ± 4%). In addition, exposure of endothelium-denuded vessels to high glucose for up to 4 h altered neither angiotensin II–stimulated PGI2 nor O2.− release (data not shown), implying that intact endothelium was required for high-glucose–enhanced O2.− and PGIS nitration.

Supplementation of sepiapterin attenuates high-glucose–induced PGIS nitration.

There is evidence that under conditions such as lack of its essential cofactor, tetrahydrobiopterin (BH4), eNOS releases O2.− instead of NO (eNOS uncoupling) (31,32). Sepiapterin is a precursor of BH4 and can be converted into BH4 through a savage pathway (33,34). We next determined if supplementation of sepiapterin altered high-glucose–enhanced oxidant stress and PGIS nitration. As expected, concurrent administration of sepiapterin, which had no effect on angiotensin II–triggered vasorelaxation in BCAs exposed to 5 mmol/l glucose, not only abolished high-glucose–impaired vasorelaxation and PGI2 release (Fig. 4A) but also prevented high-glucose–enhanced PGIS nitration (Fig. 4B and C). Together, these results indicated that high glucose, via eNOS, increased PGIS nitration by reducing the levels of BH4, the essential cofactor of eNOS.

High glucose increases coimmunoprecipitation of caveolin-1, eNOS, and PGIS.

We next investigated why eNOS-derived ONOO targeted PGIS. PGIS is mainly localized in endoplasmic reticulum, but there is evidence that PGIS is colocalized with caveolin-1 within caveolae (35). We first examined if high glucose increased the translocation of PGIS from endoplasmic reticulum into caveolae, where eNOS is localized (36,37). Exposure of BCAs to either high glucose or chemically synthetized ONOO (50 μmol/l) significantly increased the association of caveolin-1 with PGIS (Fig. 5A and B). The same blots were stripped and Western blotted against eNOS. As expected, both high glucose and ONOO simultaneously increased the association of eNOS with caveolin-1 (Fig. 5A and B). These results indicated that ONOO generated by high glucose increased the colocalization of both PGIS and eNOS within caveolae.

We next studied the effects of sepiapterin on high-glucose–enhanced association of eNOS with PGIS. BCAs were exposed to high glucose with or without sepiapterin. After incubation, eNOS was immunoprecipiated and Western blotted with either PGIS or eNOS. As expected, high glucose drastically increased the association of eNOS with PGIS (Fig. 5C and D). Interestingly, concurrent administration of sepiapterin attenuated high-glucose–enhanced association of PGIS with eNOS (Fig. 5C and D).

eNOS-dependent tyrosine nitration of PGIS in STZ-induced diabetes.

It was interesting to investigate if diabetes increased PGIS nitration in vivo and, if so, the contributions of eNOS-derived oxidants. To this end, three types of animals, mice overexpressing hSOD (hSOD+/+, scavenging O2.− to prevent ONOO), eNOS-KO mice (lack of eNOS-derived NO to prevent ONOO), and their littermates, C57BL6 mice, were made diabetic by STZ injections. Mice with hSOD+/+ appeared resistant to STZ; therefore, a higher dose of STZ was given to hSOD+/+ mice to achieve comparable levels of serum glucose. Two weeks after STZ injections, serum glucose levels in STZ-injected mice were significantly elevated compared with mice given vehicle (451 ± 31 vs. 129 ± 11 mg/dl for C57BL6, 437 ± 21 vs. 113 ± 8 mg/dl for hSOD+/+, and 461 ± 34 vs. 135 ± 15 mg/dl for eNOS−/−; P < 0.001, n = 9–11). The levels of serum glucose were not significantly different in all STZ-injected mice. In all three groups, the body weights of diabetic animals were significantly lower (11%) than those of nondiabetic controls (17.0 ± 0.6 vs. 19.1 ± 1.1; P < 0.05, n = 10). The diabetic animals had similar levels in heart weight (72.4 ± 3.1 vs. 71.1 ± 0.9 mg) compared with untreated diabetic mice.

As shown in Fig. 6A, diabetes increased O2.− in isolated arteries from diabetic mice. O2.− release was significantly elevated in diabetic C57BL6 mice aortas compared with those in the sham-treated mice (Fig. 6A). In hSOD mice, the basal aortic O2.− release was significantly lower than those in C57BL6 mice (Fig. 6A). Basal aortic O2.− was elevated in eNOS−/− mice compared with C57BL6 mice, which might be due to lack of NO. Interestingly, diabetes increased O2.− levels in hSOD+/+ mice to a much less significant extent than in diabetic C57BL6 mice (Fig. 6A). Unlike in C57BL6 mice, diabetes did not increase O2.− in eNOS−/− mice. O2.− release in diabetic eNOS−/− mice was significantly less than that in diabetic C57BL6 mice (Fig. 6A).

Concomitant production of both O2.− and NO results in the formation of reactive nitrogen such as ONOO. Since ONOO has a half-life of <1 s at physiological conditions, we next assayed the contents of 3-nitrotyrosine in mouse aortic homogenates, a footprint of reactive nitrogen species including ONOO. As shown in Fig. 6B, diabetes significantly increased the levels of 3-nitrotyrosine in C57BL6 but significantly less in both eNOS−/− and SOD+/+ mice. Since neither NO nor O2.− alone causes 3-nitrotyrosine formation in vivo, these results implied that diabetes enhances 3-nitrotyrosine, likely via the formation of ONOO.

We next determined if ONOO formed in diabetes altered PGIS activity by nitrating the enzyme. As shown in Fig. 7A, diabetes significantly suppressed PGIS activity, whereas overexpression of either hSOD or eNOS−/− prevented diabetes-induced PGIS inhibition. In parallel, diabetes drastically increased PGIS nitration (Fig. 7B and C) in STZ-injected C57BL6 mice but to a lesser extent in mice either lacking eNOS−/− or overexpressing hSOD (Fig. 7B and C). These results further support the notion that diabetes increases PGIS nitration in vivo, likely via ONOO.

In the present study, we have, for the first time, demonstrated eNOS-dependent PGIS nitration in isolated BCAs when stimulated with angiotensin II in diabetes in vivo. The evidence can be summarized as follows. First, high glucose significantly impaired PGI2-dependent relaxation in isolated BCAs and increased O2.− release after a 1- to 4-h exposure. In addition, high glucose, but not mannitol, significantly increased PGIS nitration, although the levels of PGIS expression were not changed. Second, removal of endothelium (depletion of eNOS) prevented high-glucose–induced O2.− release and PGIS nitration, suggesting that the oxidant was mainly from vascular endothelium. Third, our previous studies (2224) have demonstrated that this enzyme is tyrosine nitrated by low concentrations of exogenous ONOO. In this study, we found that either inhibition of eNOS with l-NAME or scavenging O2.− with SOD attenuated high-glucose–enhanced PGIS nitration, which implied that reactive nitrogen species, likely ONOO, are involved in high-glucose–enhanced PGIS nitration. Fourth, inhibition of eNOS by l-NAME abolished both high-glucose–enhanced O2.− and PGIS nitration, suggesting that eNOS might be a source of ONOO. Fifth, sepiapterin restored PGI2-dependent relaxation by preventing high-glucose–enhanced PGIS nitration. This lends support to eNOS becoming the source of O2.− and ONOO in BCAs exposed to high glucose. Finally, using STZ-induced diabetic mice, we found that both diabetic eNOS−/− mice and hSOD+/+ mice exhibited less PGIS nitration and released less O2.− compared with their diabetic littermates. Conversely, PGIS activity was significantly preserved in both eNOS−/− and hSOD+/+ mice. Thus, tyrosine nitration of PGIS is most likely mediated by ONOO formed endogenously from NO and O2.− in diabetes. Although peroxidases such as myeloperoxidase have been reported to cause protein nitration in vivo (38,39), the role of peroxidase-catalyzed PGIS nitration is less likely, since the hSOD+/+ mice significantly attenuated diabetes-enhanced PGIS nitration and inhibition. Thus, peroxidase-catalyzed PGIS nitration is less likely because overexpression of hSOD, which provides hydrogen peroxide (H2O2) to facilitate peroxidase-catalyzed 3-nitrotyrosine formation, should enhance instead of decrease PGIS nitration in STZ-injected mice. Although we cannot totally exclude the possibility of peroxidase-catalyzed PGIS nitration, our data strongly suggest that ONOO is likely to be responsible for the increased PGIS nitration caused by diabetes in vivo.

Another important finding we have presented is that inactivation of PGIS in diabetes results in a consequent TPr activation through its cumulative substrate, PGH2, on isolated vessels ex vivo (Fig. 1D). This consequent TPr stimulation triggers vasospasm. In addition, there is evidence that TPr causes platelet aggregation and pathological changes, such as increased apoptosis and abnormal expression of adhesion molecules in endothelial cells, that are opposed by PGI2. Since TXA2 promotes and PGI2 prevents the initiation and progression of atherogenesis, our results unveil a novel mechanism by which diabetes causes atherosclerosis, i.e., diabetes generates uncontrolled O2.−, resulting in an increased destruction of NO and a concomitant formation of a highly cytotoxic oxidant, ONOO, that triggers nitration and inhibition of PGIS and results in consequent TPr stimulation. Thus, the present study, for the first time, provides evidence that diabetes via ONOO may contribute to the functional defects of the endothelium in pathological situations, not only by a lack of the vasorelaxants NO and PGI2 but even more directly by causing accumulation of the proatherothrombotic PGH2. Indeed, infusion of oxygen radical scavengers (SOD, vitamin C, deferoxamine, etc.) improves cold pressor–induced vasorelaxation and abolishes acetylcholine-induced paradoxical vasoconstriction, suggesting that inactivation of NO by reactive oxygen species and subsequent formation of ONOO contribute to some extent to the abnormal vasomotion observed in diabetic patients (1620). In addition, our results might also provide a potential explanation for the paradoxic effects of endothelium-dependent vasorelaxants, such as acetylcholine, that trigger vasoconstriction in human atherosclerotic arteries. Even individuals with significant atherosclerotic risk factors but without clinically manifested atherosclerosis have a decreased vasodilator response in parallel with higher production of vasoconstricting PGs. Such abnormal responses are normalized by inhibition of COX. Thus, the nitration of PGIS we have described in the present study may contribute to the initiation and progression of vascular complications in diabetes as a result of the downregulation of protective actions of both PGI2 and NO and because nonmetabolized PGH2 tips the balance toward platelet aggregation, atheroma accumulation, and thrombus formation. The mechanism of diabetes via ONOO formation for which we have presented evidence may serve as an explanation for the observed endothelial dysfunction, since ONOO-dependent tyrosine nitration of PGIS helps to unify and explain several previously proposed pathogenic abnormalities in diabetes, including 1) decreased availability of NO, 2) decreased PGI2, 3) expression of vasoconstrictors PGH2/TxA2, and 4) increased free radicals (Fig. 8).

H.N. and J.W. contributed equally to this work.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

This work was supported by grants from the National Institutes of Health (HL079584, HL07439, and HL080499) and a Research Award from the American Diabetes Association.

The authors acknowledge Dr. T. Klein for kindly providing an antibody against human PGIS. The also thank Melissa Guzman for editorial assistance.

1
Zou MH, Ullrich V, Cohen R: Peroxynitrite and vascular endothelial dysfunction in diabetes mellitus.
Endothelium
11
:
89
–97,
2004
2
Ruderman NB, Williamson JR, Brownlee M: Glucose and diabetic vascular disease.
FASEB J
6
:
2905
–2914,
1992
3
Cohen RA: Role of nitric oxide in diabetic complications.
Am J Ther
12
:
499
–502,
2005
4
Brownlee M: Biochemistry and molecular cell biology of diabetic complications.
Nature
414
:
813
–820,
2001
5
Zou MH, Shi C, Cohen RA: High glucose via peroxynitrite causes tyrosine nitration and inactivation of prostacyclin synthase that is associated with thromboxane/prostaglandin H2 receptor–mediated apoptosis and adhesion molecule expression in cultured human aortic endothelial cells.
Diabetes
51
:
198
–203,
2002
6
Zou MH, Shi S, Cohen RA: Oxidation by peroxynitrite of zinc-thiolate cluster of eNOS uncouples the enzyme in diabetes.
J Clin Invest
109
:
817
–826,
2002
7
Beckman JS, Beckman TW, Chen J, Marshall PA, Freeman BA: Apparent hydroxyl radical production by peroxynitrite: implications for endothelial injury from nitric oxide and superoxide.
Proc Natl Acad Sci U S A
87
:
1620
–1624,
1990
8
Beckman JS, Chen J, Ischiropoulos H, Crow JP: Oxidative chemistry of peroxynitrite.
Methods Enzymol
233
:
229
–240,
1994
9
Beckman JS, Koppenol WH: Nitric oxide, superoxide, and peroxynitrite: the good, the bad, and the ugly.
Am J Physiol
271
:
C1424
–C1437,
1996
10
Moncada S, Vane JR: Pharmacology and endogenous roles of prostaglandin endoperoxides, thromboxane A2 and prostacyclin.
Pharmacol Rev
30
:
292
–331,
1979
11
Moncada S: Biological importance of prostacyclin.
Br J Pharmacol
76
:
3
–31,
1982
12
Dollery CT, Friedman LA, Hensby CN, Kohner E, Lewis PJ, Porta M, Webster J: Circulating prostacyclin may be reduced in diabetes.
Lancet
ii
:
1365
,
1979
13
Johnson M, Harrison HE, Raftery AT, Elder JB: Vascular prostacyclin may be reduced in diabetes in man.
Lancet
i
:
325
–326,
1979
14
Silberbauer K, Schernthaner G, Sinzinger H, Piza-Katzer H, Winter M: Decreased vascular prostacyclin in juvenile-onset diabetes.
N Engl J Med
300
:
366
–367,
1979
15
Silberbauer K, Clopath P, Sinzinger H, Schernathaner G: Effect of experimentally induced diabetes on swine vascular prostacyclin (PGI2) synthesis.
Artery
8
:
30
–36,
1980
16
Tesfamarian B, Brown ML, Cohen RA: Elevated glucose impairs endothelium-dependent relaxation by activating protein kinase C.
J Clin Invest
87
:
1643
–1648,
1991
17
Shimizu K, Muramatsu M, Kakegawa Y, Asano H, Toki Y, Miyazaki Y, Okumura K, Hashimoto H, Ito T: Role of prostaglandin H2 as an endothelium-derived contracting factor in diabetic state.
Diabetes
42
:
1246
–1252,
1993
18
Dai FX, Diederich A, Skopec J, Diederich D: Diabetes-induced endothelial dysfunction in streptozotocin-treated rats: role of prostaglandin endoperoxides and free radicals.
J Am Soc Nephrol
4
:
1327
–1336,
1993
19
Morinelli TA, Tempel GE, Jaffa AA, Silva RH, Naka M, Folger W, Halushka PV: Thromboxane A2/prostaglandin H2 receptors in streptozotocin-induced diabetes: effects of insulin therapy in the rat.
Prostaglandins
45
:
427
–438,
1993
20
Zou MH, Ullrich V: Peroxynitrite formed by simultaneous generation of nitric oxide and superoxide selectively inhibits bovine aortic prostacyclin synthase.
FEBS Lett
382
:
101
–104,
1996
21
Zou MH, Martin C, Ullrich V: Tyrosine nitration as a mechanism of selective inactivation of prostacyclin synthase by peroxynitrite.
Biol Chem
378
:
707
–713,
1997
22
Zou M, Jendral M, Ullrich V: Peroxynitrite induces prostaglandin endoperoxide dependent-vasospasm in bovine coronary arteries via nitration of prostacyclin synthase.
Br J Pharmacol
126
:
1283
–1292,
1999
23
Zou MH, Bachschmid M: Hypoxia-reoxygenation causes coronary vasospasm via tyrosine nitration of prostacyclin synthase.
J Exp Med
190
:
135
–139,
1999
24
Davis B, Zou MH: CD40Ligand (CD40L)-dependent tyrosine nitration of prostacyclin synthase in vivo.
Circulation
112
:
2184
–2192,
2005
25
Schmidt P, Youhnovski N, Daiber A, Balan A, Arsic M, Bachschmid M, Przybylski M, Ullrich V: Specific nitration at tyrosine 430 revealed by high resolution mass spectrometry as basis for redox regulation of bovine prostacyclin synthase.
J Biol Chem
278
:
12813
–12819,
2003
26
Skatchkov MP, Sperling D, Hink U, Mulsch A, Harrison DG, Sindermann I, Meinertz T, Munzel T: Validation of lucigenin as a chemiluminescent probe to monitor vascular superoxide as well as basal vascular nitric oxide production.
Biochem Biophys Res Commun
254
:
319
–324,
1999
27
Mais DE, Saussy DL, Chaikouni A, Kochel PJ, Knapp DR, Hamanaka N, Halushka PV: Pharmacological characterization of human and canine thromboxane A2/prostaglandin H2 receptors in platelets and blood vessels: evidence for different receptors.
J Pharmacol Exp Ther
233
:
418
–424,
1985
28
Ogletree M, Harris DN, Greenberg R, Haslanger MF, Nakane M: Pharmacological actions of SQ29548, a novel selective thromboxane antagonist.
J Pharmacol Exp Ther
234
:
435
–441,
1985
29
Griendling KK, Sorescu D, Ushio-Fukai M: NAD(P)H oxidase: role in cardiovascular biology and disease.
Circ Res
86
:
494
–501,
2000
30
Hande KR, Chabner BA: A competitive protein binding assay for allopurinol and oxipurinol.
Anal Biochem
101
:
26
–33,
1980
31
Vasquez-Vivar J, Kalyanaraman B, Martasek P: The role of tetrahydrobiopterin in superoxide generation from eNOS: enzymology and physiological implications.
Free Radic Res
37
:
121
–127,
2003
32
Alp NJ, Channon KM: Regulation of endothelial nitric oxide synthase by tetrahydrobiopterin in vascular disease.
Arterioscler Thromb Vasc Biol
24
:
413
–420,
2004
33
Baker TA, Milstien S, Katusic ZS: Effect of vitamin C on the availability of tetrahydrobiopterin in human endothelial cells.
J Cardiovasc Pharmacol
37
:
333
–338,
2001
34
Mata-Greenwood E, Jenkins C, Farrow KN, Konduri GG, Russell JA, Lakshminrusimha S, Black SM, Steinhorn RH: eNOS function is developmentally regulated: uncoupling of eNOS occurs postnatally.
Am J Physiol Lung Cell Mol Physiol
290
:
L232
–L241,
2006
35
Spisni E, Griffoni C, Santi S, Riccio M, Marulli R, Bartolini G, Toni M, Ullrich V, Tomasi V: Colocalization prostacyclin (PGI2) synthase–caveolin-1 in endothelial cells and new roles for PGI2 in angiogenesis.
Exp Cell Res
266
:
31
–43,
2001
36
Everson WV, Smart EJ: Influence of caveolin, cholesterol, and lipoproteins on nitric oxide synthase: implications for vascular disease.
Trends Cardiovasc Med
11
:
246
–250,
2001
37
Shaul PW: Regulation of endothelial nitric oxide synthase: location, location, location.
Annu Rev Physiol
64
:
749
–774,
2002
38
Eiserich JP, Hristova M, Cross CE, Jones AD, Freeman BA, Halliwell B, van der Villet A: Formation of nitric oxide-derived inflammatory oxidants by myeloperoxidase in neutrophils.
Nature
391
:
391
–397,
1998
39
van der Vliet A, Eiserich JP, O’Neill CA, Halliwell B, Cross CE: Tyrosine modification by reactive nitrogen species: a closer look.
Arch Biochem Biophys
319
:
341
–349,
1995
40
Cosentino F, Lüscher TF: Endothelial dysfunction in diabetes mellitus.
J Cardiovasc Pharmacol
32
:
S54
–S61,
1998