Donor and recipient plasma follistatin levels are associated with acute GvHD in Blood and Marrow Transplant Clinical Trials Network 0402.

Follistatin is an angiogenic factor elevated in the circulation after allogeneic hematopoietic cell transplantation (HCT). Elevations in follistatin plasma concentrations are associated with the onset of and poor survival after acute GvHD (aGvHD). Using data from the Blood and Marrow Transplant Clinical Trials Network 0402 study (n=247), we sought to further quantify the longitudinal associations between plasma follistatin levels in transplant recipients, as well as baseline HCT donor follistatin levels, and allogeneic HCT outcomes. Higher recipient baseline follistatin levels were predictive of development of aGvHD (P=0.04). High donor follistatin levels were also associated with the incidence of aGvHD (P<0.01). Elevated follistatin levels on day 28 were associated with the onset of grade II-IV aGvHD before day 28, higher 1-year non-relapse mortality (NRM) and lower overall survival. In multivariate analyses, individuals with follistatin levels >1088 pg/mL at day 28 had a 4-fold increased risk for NRM (relative risk (RR)=4.3, 95% confidence interval (CI) 1.9-9.9, P<0.01) and a nearly three-fold increased overall risk for mortality (RR=2.8, 95% CI 1.5-5.2, P<0.01). Given the multiple roles of follistatin in tissue inflammation and repair, and the confirmation that this biomarker is predictive of important HCT outcomes, the pathobiology of these relationships need further study.

Impact of ABO-mismatch on risk of GVHD after umbilical cord blood transplantation.

Recent advances in allogeneic hematopoietic cell transplant (allo-HCT) have led to an increasing use of alternative donors, including banked umbilical cord blood (UCB). Despite these advances, acute GVHD (aGVHD) and chronic GVHD (cGVHD) continue to be the leading causes of early and late transplant-related mortality. ABO-mismatch has been frequently reported as a risk factor for GVHD, however, data in the UCB recipients are limited. We hypothesized that as the lymphocytes in the cord blood are thought to be naive, they will therefore be less likely to mediate GVHD. Therefore, we analyzed the impact of ABO-mismatch on aGVHD and cGVHD in recipients of single and double UCB-HCT. In both univariate and multivariate analyses, presence of ABO-mismatch did not have an impact on aGVHD or cGVHD. Whereas ABO-compatible donors are preferred in recipients of URD-HCT, ABO compatibility generally need not be considered in recipients of UCB-HCT.

Hemodialysis reduces inhibitory effect of plasma ultrafiltrate on LDL oxidation and subsequent endothelial reactions.

Oxidative modification of low-density lipoprotein (LDL) and its deleterious effect on endothelium is implicated in the pathogenesis of atherosclerosis. Endothelium responds to such an insult by upregulating the synthesis of heme oxygenase-1 (HO-1) and ferritin. Endothelial cell damage and dysfunction have been observed in patients with chronic kidney disease (CKD) on maintenance hemodialysis (HD). We studied the effect of low-molecular-weight components of uremic plasma on LDL oxidation and LDL-oxidation-provoked endothelial cell reactions, such as the induction of cytotoxicity and the upregulation of cell-protective HO-1 and ferritin. Plasma ultrafiltrate (molecular weight<5000 Da) from CKD patients on HD or when treated conservatively exhibited a pronounced inhibition on heme-mediated oxidative modification of LDL. Endothelial cell cytotoxicity provoked by LDL oxidation was also attenuated by plasma ultrafiltrate from CKD patients. During HD treatment, a dramatic drop occurred in the retardation of oxidative reactions, and a loss of endothelial cytoprotection exerted by plasma ultrafiltrate was noted. The upregulation of HO-1 and ferritin in response to oxidative stress of LDL was blunted by uremic plasma ultrafiltrate that was released by the end of HD. The decreased antioxidant capacity of ultrafiltrate after HD occurred as a consequence of the intradialytic removal of L-ascorbic acid, uric acid, bilirubin, 3-indoxyl sulfate, indoxyl-beta-D-glucuronide, p-cresol, and phenol. Intradialytic removal of L-ascorbic acid, uric acid, bilirubin, 3-indoxyl sulfate, indoxyl-beta-D-glucuronide, p-cresol, and phenol increases the risk of LDL oxidation and subsequent endothelial cell damage, which underlines the importance of activation of cytoprotective HO-1 and ferritin in endothelium.

Oxidation of hemoglobin by lipid hydroperoxide associated with low-density lipoprotein (LDL) and increased cytotoxic effect by LDL oxidation in heme oxygenase-1 (HO-1) deficiency.

Heme-catalyzed oxidation of low-density lipoprotein (LDL) is one of the relevant mechanisms involved in LDL modification. We previously revealed a substantial oxidation of plasma hemoglobin to methemoglobin and a subsequent heme-catalyzed LDL oxidation generating moieties toxic to endothelium in heme oxygenase-1 (HO-1)-deficiency in human. Drawing upon our previous observation we posited a pathway for oxidation of plasma hemoglobin in the HO-1-deficient child involving LDL-associated lipid hydroperoxide. In support, LDL-associated lipid hydroperoxide oxidized ferrohemoglobin to methemoglobin--known to readily release its heme moieties--in a dose-dependent manner. Repeated heme exposure of the child s LDL further increased its lipid hydroperoxide content within min leading to additional cytotoxic effect on endothelium. Both cytotoxicity and HO-1 inducing ability of the oxidized LDL were strongly dependent on its lipid hydroperoxide content. We wondered if cells of the HO-1-deficient patient were prone to oxidative damage arising from heme-mediated oxidation of LDL. Indeed, we found elevated cytotoxicity induced by heme-catalyzed oxidation of LDL in lymphoblastoid cells derived from the HO-1-deficient patient. We conclude that oxidation of hemoglobin to methemoglobin by LDL-associated lipid hydroperoxide and increased sensitivity of cells of the HO-1-deficient child to stress of oxidized LDL might contribute to the vascular disorders reported earlier.

Overview of infections and cardiovascular diseases.

Microbes have been proposed as inciting agents of tissue injury and inflammation, both of which underlie the pathogenesis of atherosclerosis. Viruses, including the herpes simplex virus and cytomegalovirus, as well as bacteria such as Chlamydia pneumoniae, have been implicated in the process. In vitro, these agents promote a proinflammatory and a procoagulant phenotype in vascular cells. Viruses augment cell accumulation through alterations of apoptosis. Infectious agents may play a role in pathogenesis of atherosclerosis by triggering an autoimmune response due to microbial molecular mimicry. It is unlikely that a single agent is the sole cause or modulator of this heterogeneous disease. Contradictory epidemiological studies may be reconciled with a new construct suggesting that multiple pathogens infecting an individual in aggregate may promote an inflammatory and procoagulant environment that underlies the pathogenesis of atherosclerosis.

Microbes, inflammation and atherosclerosis: will old pathology lessons guide new therapies?

Although attractive, the microbial pathogenesis theory for atherosclerosis remains unproven. Over the last century, microbiologists have invoked fulfillment of Koch's postulates to determine pathogen causality. Certainly a multifactorial disease process such as atherosclerosis unlikely will be due to a single microbial agent, an agent when transferred to another host, will always induce atherosclerosis. Conflicting epidemiological data also do not support a single causative agent. However, as presented here, considerable in vitro, animal, and human epidemiological data support the plausibility that infectious agents can promote a proinflammatory, procoagulant and proatherogenic environment in the vessel wall. Microbial genes and molecules can catalyze these processes and foil normal cellular events. But, must intact microbes enter the vessel wall or can microbial molecules incite immune responses from afar? A new focus on pathogen-induced auto-immunity toward vasculature has been presented. For example, microbes contain molecules that mimic host cellular components (55). An immune response to a pathogen may cross react with vessel wall cellular structures. This immune response enhanced by infection may lead to high levels of cross reacting auto-antibodies or auto-aggressive T-cells. Epstein has championed the concept of pathogen burden in support of this auto-immune theory (56). Individuals infected with multiple pathogens such as HSV-1, HSV-2, CMV, Helicobacter pylori, and Hepatitis A, have high C-reative protein levels (markers of inflammation) and the greatest relative risk for coronary artery disease (57). Thus, pathogens might contribute to the atherosclerotic process by promoting inflammatory responses. It is this author's view that microbes and inflammation do play a role in the pathogenesis of atherosclerosis (58). Infection may contribute to the process promoting vessel wall injury initiated by oxidized lipids, smoking derived oxidants, hypertensive shear or diabetes glyoxidized molecules. Inflammation and immune reactions in response to infection can exacerbate and act synergistically with all of the aforementioned vasculotoxic moieties. Continued investigations in the 21st century will determine if vaccines, antibiotics, anti-inflammatory agents or immunosuppressants will alter the picture the early 19th century pathologists observed under their monocular microscopes.

Cytomegalovirus inhibits p53 nuclear localization signal function.

Endothelial cells (EC) infected with the VHL strain of cytomegalovirus (CMV) are resistant to p53-mediated apoptosis, which may be relevant to EC dysfunction and atherogenesis. This resistance to apoptosis may be mediated by cytoplasmic sequestration of p53, which functions only in the nucleus. We explored the hypothesis that CMV sequesters p53 in the cytoplasm by blocking p53 nuclear localization signal (NLS) function. We transfected VHL CMV infected EC with recombinant p53 NLSI conjugated with chicken muscle pyruvate kinase (PK) plasmid. NLSI is responsible for 90% of p53 nuclear localization, and PK is not normally translocated to the nucleus after cytoplasmic production. Thus it cannot be localized in the nucleus without the assistance of the artificial NLSI. A double-labeling immunofluorescence staining method was used to identify the localization of p53 NLSI-conjugated PK in CMV-infected EC. We found that CMV infection sequesters PK and p53 in the cytoplasm by blocking NLSI function. This inactivation of NLSI function is dependent upon infection stage; it occurs only in the early and late phases and not the immediate early phase of infection. These findings may be relevant to endothelial dysfunction and initiation of atherogenesis. Our study also suggests a novel mechanism of the p53 inactivation by virus, which may be important for atherogenesis and tumorgenesis.

Heme protein-induced chronic renal inflammation: suppressive effect of induced heme oxygenase-1.

BACKGROUND: Heme oxygenase (HO) is the rate-limiting enzyme in the degradation of heme; its inducible isozyme, HO-1, protects against acute heme protein-induced nephrotoxicity and other forms of acute tissue injury. This study examines the induction of HO-1 in the kidney chronically inflamed by heme proteins and the functional significance of such an induction of HO-1. METHODS: Studies were undertaken in a patient with chronic tubulointerstitial disease in the setting of paroxysmal nocturnal hemoglobinuria (PNH), in a rat model of chronic tubulointerstitial nephropathy caused by repetitive exposure to heme proteins, and in genetically engineered mice deficient in HO-1 (HO-1 -/-) in which hemoglobin was repetitively administered. RESULTS: The kidney in PNH evinces robust induction of HO-1 in renal tubules in the setting of chronic inflammation. The heme protein-enriched urine from this patient, but not urine from a healthy control subject, induced expression of HO-1 in renal tubular epithelial cells (LLC-PK1 cells). A similar induction of HO-1 and related findings are recapitulated in a rat model of chronic inflammation induced by repetitive exposure to heme proteins. Additionally, in the rat, the administration of heme proteins induces monocyte chemoattractant protein (MCP-1). The functional significance of HO-1 so induced was uncovered in the HO-1 knockout mouse: Repeated administration of hemoglobin to HO-1 +/+ and HO-1 -/- mice led to intense interstitial cellular inflammation in HO-1 -/- mice accompanied by striking up-regulation of MCP-1 and activation of one of its stimulators, nuclear factor-kappaB (NF-kappaB). These findings were not observed in similarly treated HO-1 +/+ mice or in vehicle-treated HO-1 -/- and HO-1 +/+ mice. CONCLUSION: We conclude that up-regulation of HO-1 occurs in the kidney in humans and rats repetitively exposed to heme proteins. Such up-regulation represents an anti-inflammatory response since the genetic deficiency of HO-1 markedly increases activation of NF-kappaB, MCP-1 expression, and tubulointerstitial cellular inflammation.

Activated monocytes in sickle cell disease: potential role in the activation of vascular endothelium and vaso-occlusion.

Sickle cell anemia is characterized by painful vaso-occlusive crises. It is hypothesized that monocytes are activated in sickle cell disease and can enhance vaso-occlusion by activating endothelium. To test this hypothesis, human umbilical vein endothelial cells (HUVEC) and human microvascular endothelial cells (MVEC) with sickle and normal mononuclear leukocytes were incubated, and endothelial activation was measured. Endothelial cells incubated with sickle mononuclear leukocytes were more activated than those incubated with normal mononuclear leukocytes, as judged by the increased endothelial expression of adhesion molecules and tissue factor and the adhesion of polymorphonuclear leukocytes (PMNL). Monocytes, not lymphocytes or platelets, were the mononuclear cells responsible for activating endothelial cells. Sickle monocytes triggered endothelial nuclear factor-kappa B (NF-kappaB) nuclear translocation. Cell-to-cell contact of monocytes and endothelium enhanced, but was not required for, activation. Antibodies to tumor necrosis factor-alpha (TNF-alpha) and interleukin-1-beta (IL-1beta) blocked activation of the endothelium by monocytes. Peripheral blood monocytes from patients with sickle cell disease had 34% more IL-1beta (P =.002) and 139% more TNF-alpha (P =.002) per cell than normal monocytes. Sixty percent of sickle monocytes expressed the adhesion molecule ligand CD11b on their surfaces compared with only 20% of normal monocytes (P =.002). Serum C-reactive protein, a marker of systemic inflammation, was increased 12-fold in sickle serum than in normal serum (P =.003). These results demonstrate that sickle monocytes are activated and can, in turn, activate endothelial cells. It is speculated that vascular inflammation, marked by activated monocytes and endothelium, plays a significant role in the pathophysiology of vaso-occlusion in sickle cell anemia.

Reperfusion injury pathophysiology in sickle transgenic mice.

Reperfusion of tissues after interruption of their vascular supply causes free-radical generation that leads to tissue damage, a scenario referred to as "reperfusion injury." Because sickle disease involves repeated transient ischemic episodes, we sought evidence for excessive free-radical generation in sickle transgenic mice. Compared with normal mice, sickle mice at ambient air had a higher ethane excretion (marker of lipid peroxidation) and greater conversion of salicylic acid to 2,3-dihydroxybenzoic acid (marker of hydroxyl radical generation). During hypoxia (11% O(2)), only sickle mice converted tissue xanthine dehydrogenase to oxidase. Only the sickle mice exhibited a further increase in ethane excretion during restitution of normal oxygen tension after 2 hours of hypoxia. Only the sickle mice showed abnormal activation of nuclear factor-kappaB after exposure to hypoxia-reoxygenation. Allopurinol, a potential therapeutic agent, decreased ethane excretion in the sickle mice. Thus, sickle transgenic mice exhibit biochemical footprints consistent with excessive free-radical generation even at ambient air and following a transient induction of enhanced sickling. We suggest that reperfusion injury physiology may contribute to the evolution of the chronic organ damage characteristic of sickle cell disease. If so, novel therapeutic approaches might be of value.

Human cytomegalovirus immediate early proteins upregulate endothelial p53 function.

Infected endothelial cells are found to be resistant to apoptosis possibly mediated by p53 cytoplasmic sequestration. We explored whether the immediate early 84 kDa protein (IE84) of cytomegalovirus (CMV) is responsible for p53 cytoplasmic sequestration. The endothelial cells were transfected with plasmids containing IE1 and 2 coding regions which are known to synthesize IE84 and 72 proteins. Our study found that p53 expression was significantly elevated in endothelial cells transfected with IE1 and 2 plasmids. However, p53 was only found in the nucleus rather than sequestered in the cytoplasm. We have demonstrated that IE84 and 72 are not responsible for p53 dysfunction caused by CMV infection, rather they upregulate p53 function and promote endothelial apoptosis.

Ferriporphyrins and endothelium: a 2-edged sword-promotion of oxidation and induction of cytoprotectants.

Heme arginate infusions blunt the symptoms of patients with acute intermittent porphyria without evidence of the vascular or thrombotic side effects reported for hematin. To provide a rationale for heme arginate's safety, the present study examined the effects of various ferriporphyrins to sensitize human endothelial cells to free radical injury and to induce heme oxygenase and ferritin expression. Heme arginate, unlike hematin, did not amplify oxidant-induced cytotoxicity mediated by hydrogen peroxide (5.3 +/- 2.4 versus 62.3 +/- 5.3% (51)Cr release, P <.0001) or by activated neutrophils (14.4 +/- 2.9 versus 41.1 +/- 6.0%, P <.0001). Nevertheless, heme arginate efficiently entered endothelial cells similarly to hematin, since both markedly induced heme oxygenase mRNA (more than 20-fold increase) and enzyme activity. Even with efficient permeation, endothelial cell ferritin content was only minimally increased by heme arginate compared with a 10-fold induction by hematin; presumably less free iron was derived from heme arginate despite up-regulation of heme oxygenase. Hematin is potentially vasculopathic by its marked catalysis of oxidation of low-density lipoprotein (LDL) to endothelial-toxic moieties. Heme arginate was significantly less catalytic. Heme arginate-conditioned LDL was less than half as cytotoxic to endothelial cells as hematin-conditioned LDL (P <.004). It is concluded that heme arginate may be less vasculotoxic than hematin since it is an effective heme oxygenase gene regulator but a less efficient free-radical catalyst.

Effects of anti-rheumatic gold salts on NF-kappa B mobilization and tumour necrosis factor-alpha (TNF-alpha)-induced neutrophil-dependent cytotoxicity for human endothelial cells.

We have previously shown that the gold-containing disease-modifying anti-rheumatic drugs, auranofin (AF) and gold sodium aurothiomalate (GSTM) reduce human umbilical vein endothelial cell (HUVEC) adhesion molecule expression and neutrophil (PMN) adherence. AF diminishes E-selectin and intercellular adhesion molecule-1 (ICAM-1) on cytokine-activated HUVEC, while GSTM decreases only E-selectin. Since tight adhesion is critical for PMN to damage EC, we tested whether these drugs modulated human PMN-mediated injury to TNF-alpha-activated HUVEC in vitro (as measured by 51Cr release). Here we show that TNF-alpha caused a prominent PMN-mediated cytotoxicity that was dose-dependently reduced when AF and GSTM were added to the assay system. We also found that a potent inhibitor of NF-kappaB, pyrrolidine dithiocarbamate (PDTC) in a dose-dependent manner impaired TNF-alpha-induced cytotoxicity, indicating a role of NF-kappaB activation in cytokine-induced endothelial injury. To examine the effects of AF and GSTM on TNF-alpha-induced NF-kappaB activation this was measured in HUVEC nuclear extracts by an electrophoretic mobility shift assay. AF, but not GSTM, decreased TNF-alpha-induced NF-kappaB activation in HUVEC. Thus, in this in vitro model of vasculitis, AF and GSTM dose dependently reduced TNF-alpha-mediated neutrophil-dependent cytotoxicity for HUVEC, and AF, but not GSTM, inhibited NF-kappaB mobilization, thereby providing possible mechanisms for effects of AF and GSTM.

Low-density lipoprotein susceptibility to oxidation and cytotoxicity to endothelium in sickle cell anemia.

Patients with sickle-cell anemia exhibit pro-oxidative metabolic perturbations. We hypothesize that because of chronic oxidative stress, plasma low-density lipoprotein (LDL) from patients with sickle-cell anemia is more susceptible to oxidation. To test this hypothesis, LDL susceptibility to copper-mediated oxidation was measured in 24 patients with sickle-cell anemia and 48 control subjects. Sickle-cell LDL was more susceptible to oxidation than control LDL, measured by a 22% shorter mean lag time between LDL exposure to CuSO4 and conjugated diene formation (97 vs 124 minutes; P = .023). LDL vitamin E, iron, heme, and cholesterol ester hydroperoxide (CEOOH) levels were also measured. LDL vitamin E levels were significantly lower in patients with sickle-cell anemia compared with control subjects (1.8 vs 2.9 mol/mol LDL; P = .025), but there was no correlation with lag time. Pro-oxidant heme and iron levels were the same in sickle-cell and control LDL. LDL CEOOHs were not significantly different in sickle and control LDL (3.1 vs 1.2 mmol/mol of LDL unesterified cholesterol, P = .15), but LDL CEOOH levels were inversely correlated with lag times in patients with sickle-cell anemia (r2 = 0.38; P = .018). The cytotoxicity of partially oxidized LDL to porcine aortic endothelial cells was inversely correlated with lag times (r2 = 0.48; P = .001). These preliminary data suggest that increased LDL susceptibility to oxidation could be a marker of oxidant stress and vasculopathy in patients with sickle-cell anemia.

Intercellular adhesion molecule-1 expression in endothelial cells is activated by cytomegalovirus immediate early proteins.

BACKGROUND: Human cytomegalovirus (HCMV) infection is associated with allograft vasculopathy and rejection. One potential mechanism is vascular injury from immunologically mediated processes. HCMV infection has been shown to increase the constitutive expression of intercellular adhesion molecule-1 (ICAM-1). The objective of this study was to determine the molecular basis of HCMV enhanced ICAM-1 gene expression in endothelial cells using human umbilical vein endothelial cells (HUVECs) as a model. METHODS: HUVECS were infected with HCMV virus and the level of ICAM-1 mRNA determined over time. HUVECS were then transiently transfected with plasmid constructs containing ICAM-1 and HCMV immediate early (IE) gene sequences and the effect of IE proteins on ICAM-1 promoter expression determined. Antibodies to cytokines known to be affected by HCMV IE proteins or to modulate ICAM-1 expression were added to determine if cytokines were mediating ICAM-1 expression. RESULTS: Infection of HUVECs with HCMV resulted in a rapid rise in ICAM-1 mRNA levels, suggesting that the viral IE proteins were involved in gene activation. The HCMV IE1 and IE2 proteins synergistically activated both transfected and endogenous ICAM-1 gene expression. The addition of antibodies to interleukin-1, tumor necrosis factor-a, transforming growth factor-beta, or interleukin-6 had no effect on the IE protein-mediated increase in ICAM-1 expression. Deletion analysis of the ICAM-1 gene promoter revealed that a minimum of 370 base pairs of 5' flanking sequences was required for maximal transactivation by IE proteins; mutation analysis showed that an NFkappaB site at base pairs -187 to -178 was not required for promoter activation. CONCLUSIONS: These results demonstrate that HCMV regulates the heterologous ICAM-1 gene promoter in endothelial cells not via cellular cytokine production, but rather by a direct effect of IE proteins, and supports a model in which HCMV IE gene products interact with ICAM-1 promoter elements to increase gene expression.

Effects of viral activation of the vessel wall on inflammation and thrombosis.

Herpes simplex virus type 1 and cytomegalovirus alter the phenotype of the endothelium in vitro from anticoagulant to procoagulant, thereby promoting the adherence of neutrophils and platelets to the endothelium. Virus infection of the endothelium induces the expression of viral glycoproteins and adhesion molecules, which promote neutrophil and monocyte adhesion. Herpes simplex infection of the endothelium promotes prothrombinase assembly, allowing more efficient thrombin generation. Excess thrombin generation causes translocation of P-selectin. Viral infection also induces the procoagulant molecule, tissue factor, in endothelial cells. These enhanced procoagulant effects are associated with the loss of anticoagulants, including thrombomodulin, prostacyclin and tissue plasminogen activator. These studies support the speculation that virus infection in vivo promotes vascular injury and thrombosis, which may contribute to disease states such as atherosclerosis.

Superoxide-dependent cerebrovascular effects of homocysteine.

Recent evidence indicates that elevated plasma levels of homocysteine are a risk factor for ischemic cerebrovascular diseases. However, little is known about cerebrovascular effects of homocysteine. Homocysteine could impair cerebrovascular function by metal-catalyzed production of activated oxygen species. We studied whether homocysteine, in the presence of Cu2+, alters reactivity of cerebral circulation and, if so, whether this effect depends on O-2 generation. In halothane-anesthetized rats the parietal cortex was exposed and superfused with Ringer solution. Cerebrocortical blood flow (CBF) was monitored by a laser-Doppler probe. With Ringer solution superfusion, CBF increased with hypercapnia (+134 +/- 7%; PCO2 = 50-60 mmHg) and topical application of 10 microM ACh (+35 +/- 3%), the NO donor S-nitroso-N-acetylpenicillamine (SNAP, 500 microM; +66 +/- 6%), or 1 mM papaverine (+100 +/- 6%; n = 5). Superfusion with 40 microM Cu2+ alone did not perturb resting CBF or responses to hypercapnia, ACh, SNAP, or papaverine (P > 0.05, n = 5). However, superfusion of homocysteine-Cu2+ reduced resting CBF (-28 +/- 4%) and attenuated (P < 0.05) responses to hypercapnia (-31 +/- 9%), ACh (-73 +/- 6%), or SNAP (-48 +/- 4%), but not papaverine. The effect was observed only at 1 mM homocysteine. Cerebrovascular effects of homocysteine-Cu2+ were prevented by coadministration of superoxide dismutase (SOD; 1,000 U/ml; n = 5). SOD alone did not affect resting CBF or CBF reactivity (n = 5). The observation that homocysteine-Cu2+ attenuates the response to hypercapnia, ACh, and SNAP, but not the NO-independent vasodilator papaverine, suggests that homocysteine-Cu2+ selectively impairs NO-related cerebrovascular responses. The fact that SOD prevents such impairment indicates that the effect of homocysteine is O-2 dependent. The data support the conclusion that O-2, generated by the reaction of homocysteine with Cu2+, inhibits NO-related cerebrovascular responses by scavenging NO, perhaps through peroxynitrite formation. O-2-mediated scavenging of NO might be one of the mechanisms by which hyperhomocysteinemia predisposes to cerebrovascular diseases.