The role of ferredoxin-NADP+ reductase in the concerted cell defense against oxidative damage -- studies using Escherichia coli mutants and cloned plant genes

Ferredoxin-NADP+ reductases (FNR) participate in cellular defense against oxidative damage. Escherichia coli mutants deficient in FNR are abnormally sensitive to methyl viologen and hydrogen peroxide. Tolerance to these oxidants was regained by expression of plant FNR, superoxide dismutase, or catalase genes in the mutant cells. FNR contribution to the concerted defense against viologen toxicity under redox-cycling conditions was similar to that of the two major E. coli superoxide dismutases together, as judged by the phenotypes displayed by relevant mutant strains. However, FNR expression in sodA sodB strains failed to increase their tolerance to viologens, indicating that the FNR target is not the superoxide radical. Sensitivity of FNR-deficient cells to oxidants is related to extensive DNA damage. Incubation of the mutant bacteria with iron chelators or hydroxyl radical scavengers provided significant protection against viologens or peroxide, suggesting that oxidative injury in FNR-deficient cells was mediated by intracellular iron through the formation of hydroxyl radicals in situ. The NADP(H)-dependent activities of the reductase were necessary and sufficient for detoxification, without participation of either ferredoxin or flavodoxin in the process. Possible mechanisms by which FNR may exert its protective role are discussed.     

Inhibition of in vitro replication of the oyster parasite Perkinsus marinus by the natural iron chelators transferrin, lactoferrin, and desferrioxamine

The mammalian iron-binding proteins transferrin and lactoferrin, the bactericidal peptide lactoferricin B, and the bacterial siderophore desferrioxamine were tested for their ability to inhibit the in vitro replication of the oyster parasite Perkinsus marinus. All three chelators were effective in reducing the parasite proliferation in a dose-dependent manner. Lactoferricin B, a peptide of lactoferrin that exhibits bactericidal properties unrelated to iron chelation, had no inhibitory activity on the parasite. When the chelators were partially or completely saturated with the appropriate iron equivalents, their inhibitory effects on the parasite proliferation were diminished or abolished accordingly, confirming that this activity was related to the chelator's capacity for iron sequestration. Our results indicate that the parasite has a strong requirement for soluble iron and its growth rates are correlated with iron availability. We propose that excess iron accumulation in the host Crassostrea virginica promotes parasite proliferation. P. marinus may avoid oxidative damage that would compromise its intracellular survival by exhaustion the host's intracellular selected iron pools required for superoxide and hydroxyl radical production.     

The early HPV16 proteins can regulate mRNA levels of cell cycle genes in human cervical carcinoma cells by p53-independent mechanisms

Abnormal metabolism of metal ions such as zinc may contribute to neuropathology. Complexing zinc could reduce this pathology. Thus, to examine the effectiveness of metal chelating agents in vivo, a model system was used. This involved determining the ability of chelating agents to prevent neuronal death caused by zinc chloride injected into the rat hippocampus. Significant protection against zinc toxicity was obtained with pyrithione, inositol hexakisphosphate, ethylenediamine tetraacetate (EDTA) and N,N,N',N'-tetrakis(2-pyridylmethyl)ethylenediamine (TPEN). The affinity of these agents for zinc varied between 106 M-1 and 1018 M-1. Thus, the affinity for zinc within this range does not appear to be a major factor affecting the ability of chelators to provide neuroprotection. While almost complete protection was found with EDTA and TPEN given simultaneously with zinc chloride, poor protection was obtained if TPEN was given before or after zinc chloride. Other agents either did not protect against zinc-induced neuronal death (zincon), or exacerbated zinc toxicity (BTC-5N and about 40% of rats injected with a combination of zinc chloride and diethylenetriamine pentaacetate [DTPA]). Rats showing increased damage after zinc plus BTC-5N or DTPA suffered wet dog-like shakes (WDS), suggesting that these zinc chelate complexes can induce seizures resulting in seizure-related damage. In contrast, in the 60% of rats treated with zinc chloride and DTPA that had no WDS, there was about an 80% reduction in the size of the zinc-induced lesion. The ability of chelators to cross cell membranes was examined by determining whether Timm's staining for vesicular zinc was reduced following the injection of a chelator into the hippocampus. TPEN and pyrithione reduced Timm's staining for zinc. However, cell permeability was not necessary for a chelator to protect against zinc toxicity.     

Treatment of postprostatectomy urinary incontinence with behavioral methods

Neurodegeneration is characterized by a marked accumulation of iron in the affected brain regions. The reason for this is still unknown. In this article we review the available data on the possible involvement of iron and mediated oxidative stress in the aetiology of Parkinson's disease and related disorders. Iron chelators, if they effectively prevent radical formation, have great therapeutic potential against ischaemia/reperfusion, rheumatoid arthritis, and anthracycline toxicity, which are most likely free radical-mediated. The efficacy of the best established chelating drug desferal in neurodegenerative disease is limited due to its high cerebro- and oculotoxicity. New bioactive chelating agents are currently being developed, among them are oxidative stress activatable iron chelators which are most likely less toxic and can flexibly respond to an increase of free radical formation in the cell.     

Right to left differences in the ankle joint complex range of motion

In most eukaryotic cells, synthesis of the iron storage protein, ferritin is regulated by iron levels and redox conditions. Proper iron storage is important to protect against damaging iron-catalysed free radical reactions. Although iron-catalysed reactions are believed to contribute to oxidative damage and cataractogenesis, little is known about iron storage in the lens. In this study, ferritin concentration was measured in cultured canine lens epithelial cells. Baseline ferritin concentration ranged from 76-163 ng (mg protein)-1; cells cultured in low-iron media had significantly lower ferritin levels than cells cultured in iron-supplemented media. Addition of a large excess of iron as hemin resulted in an eight-fold increase in ferritin concentration. The iron chelator, Desferal, significantly decreased ferritin concentration. The reducing agent dithiothreitol decreased the hemin-induced increase in ferritin levels, but not baseline levels. In contrast, ascorbic acid induced a large increase in ferritin content. Other studies have shown that induction of ferritin synthesis can protect against oxidative damage. Regulation of ferritin levels may represent a mechanism by which the lens epithelium is protected from oxidative damage. In vivo, epithelial cells are normally exposed to much lower iron concentrations than the cultured lens epithelial cells in this study. However, in pathological circumstances, the iron content and redox state of the aqueous humor is dramatically altered and may affect the steady state levels of ferritin within the lens. This remains to be determined.     

Stability of the peritoneal membrane in long-term peritoneal dialysis patients

It is now well established that the formation of free radicals and oxidative stress-induced neuronal cell death can be involved in various neurodegenerative disorders, including Alzheimer's disease and Parkinson's disease. The pineal hormone melatonin has been suggested to be a neuroprotective antioxidant. To better understand the molecular mechanism of this activity, we compared the ability of melatonin and its precursor, N-acetyl-serotonin (normelatonin), to protect human neuroblastoma SK-N-MC cells and primary cerebellar granular neurons against oxidative stress. We found that normelatonin and melatonin have differential neuroprotective effects depending on the neuronal cell type. Normelatonin was more protective against hydrogen peroxide (H2O2) and glutamate-induced cell death in SK-N-MC cells compared to melatonin which was more effective to protect primary cerebellar granular neurons against the toxicity of H2O2, glutamate and N-methyl-D-aspartate when compared to normelatonin. At the molecular level, we tested the capacity of normelatonin and melatonin to inhibit the oxidative stress-induced NF-kappaB activation in both neuronal systems. Whereas normelatonin was more potent in the suppression of the activation of NF-kappaB by H2O2 in SK-N-MC cells compared to melatonin, no apparent differences in the extent of suppression could be detected in primary neurons. Normelatonin's and melatonin's neuroprotective activity in SK-N-MC neuroblastoma cells may be mediated by the suppression of NF-kappaB activation.     

Aromatic hydroxylation in animal models of diabetes mellitus

Although the involvement of oxidative stress is well documented in the diabetic state, the individual active oxygen species generated have not been demonstrated in animal models of diabetes currently used. Since streptozotocin-induced diabetes mellitus in animals still serves as an animal model of diabetes mellitus, but streptozotocin induces diabetes and generates oxidative stress per se, we decided to study whether aromatic hydroxylation reflecting hydroxyl radical attack was found in three animal models of diabetes mellitus without streptozotocin induction or in streptozotocin-induced diabetes only. For this purpose, we compared lipid peroxidation, aromatic hydroxylation of phenylalanine, glycoxidation in genetically determined diabetic mouse strains db/db and kk, and the diabetic BB rat to these parameters in the streptozotocin-treated rat. Kidney malondialdehyde concentrations, reflecting lipid peroxidation, pentosidine, and Nepsilon-caboxymethyllysine concentrations, reflecting glycoxidation, were significantly elevated in all diabetic groups as compared to their nondiabetic mates. Aromatic hydroxylation was significantly elevated in the streptozotocin-induced diabetic state exclusively. We conclude that biochemical, pathophysiological, and treatment studies in the streptozotocin model of diabetes mellitus may be confounded by the presence of products, reactions, and tissue damage generated by aromatic hydroxylation reflecting hydroxyl radical attack. We suggest it is not the diabetic state but streptozotocin that generates the hydroxyl radical, as reflected by aromatic hydroxylation in this model.     

An antibiotic regimen for the treatment of active Crohn''s disease: a randomized, controlled clinical trial of metronidazole plus ciprofloxacin

Potassium channel openers (KCOs) such as diazoxide and levochromakalim can protect cardiac myocytes against ischemic injury and neurons against excitotoxic injury, presumably because of their ability to hyperpolarize the plasma membrane and reduce calcium influx. We now report that diazoxide, levocromakalim (LCC), and to a lesser extent pinacidil, protect cultured rat hippocampal neurons against oxidative injury induced by exposure to FeSO4 and amyloid beta-peptide (A beta). Imaging studies of intracellular peroxide levels revealed that KCOs suppressed the generation of peroxides induced by FeSO4 and A beta. KCOs were effective in protecting neurons against oxidative insults in the presence of the K+ channel blockers glibenclimide and 4-aminopyridine indicating that their protective mechanism involved actions in addition to activation of K+ channels. The data suggest that KCOs may be of therapeutic value in an array of neurodegenerative disorders that involve oxidative injury.     

Left ventricular regional systolic motion in patients with right ventricular pressure overload

The highly reactive and cytotoxic hydroxyl radical (OH) was found by electrochemical detection to be produced in reactions involving hydrogen peroxide (H2O2) and the nitric oxide (NO) donor diethylamine- NO complex. Using aromatic hydroxylation of salicylate as a specific indicator of OH, three salicylate hydroxylation products were identified; catechol, 2,3- and 2,5-dihydroxybenzoic acid. Four additional compounds were detected but not identified. The interactions of H2O2 and NO represent a biologically feasible reaction mechanism that can account for OH-induced damage in cellular environments where transition metal ions are unavailable for participation in the superoxide-mediated Fenton reaction. The ability of the NO/H2O2 complex to generate OH independently of iron or other transition metals provides a new focus for studies concerned with the origin of tissue-specific damage caused by oxygen-derived species.     

Calcium vs. iron-mediated processes in hydrogen peroxide toxicity to L929 cells: effects of glucose

H2O2 toxicity was studied in L929 cells in the presence and absence of glucose. The data obtained in the absence of glucose suggest a Ca2+-dependent mechanism of cell injury. No evidence was found for any involvement of iron in the process. In particular, cell injury was unaffected by the intracellular iron chelators 2,2'-dipyridyl and deferoxamine or by the hydroxyl radical scavengers DMSO and DMPO. On the other hand, the intracellular Ca2+ chelator BAPTA/AM provided significant protection. The cytosolic Ca2+ level rapidly and consistently increased after H2O2 addition, prior to visible bleb formation and loss of cell viability. Additionally, GSH not only prevented cell death but also significantly decreased cytosolic calcium accumulation. In the presence of glucose, however, Ca2+ does not seem to play any role in H2O2 toxicity. Cell death is now mainly mediated by iron: the iron chelators and hydroxyl radical scavengers prevented cell injury, the increase in cytosolic Ca2+ was significantly less pronounced, and BAPTA/AM did not exert any protection under these conditions. Hence, the metabolic state of the L929 cells, as given by the availability of glucose, decisively determines the biochemical mechanism of H2O2 cell injury.     

Hydroxyguanidines inhibit peroxynitrite-induced oxidation

Hydroxyguanidines (OHGs), including the endogenously formed NG-hydroxy-L-arginine (OH-arg), can react with nitric oxide (NO) and nitrogen oxides (NOx) in vitro. Therefore, we have tested OHGs and related compounds for their ability to scavenge peroxynitrite and to protect against peroxynitrite-induced oxidative processes in cells. Hydroxyguanidine, NG-hydroxy-L-arginine and other N-substituted OHGs, dose-dependently inhibited the in vitro oxidation of dihydrorhodamine (DHR) by peroxynitrite (PN), with similar or better efficacy than glutathione or cysteine. Amidoximes, aminoguanidines and O-substituted OHGs were less effective, and guanidines were without effect. In contrast to their effects on DHR oxidation, OHGs exerted only minimal inhibitory effects on the hydroxylation of benzoate by PN, suggesting that OHGs do not react with the activated isomer of peroxynitrous acid. Selected compounds were tested for protection against PN-induced suppression of mitochondrial respiration and protein oxidation in cultured J774 murine macrophages. Aminoguanidines afforded some protection against the effects of PN, but substituted-phenyl OHGs were considerably more effective. Analysis of the products of the reaction of 4-methoxybenzyl-OHG with PN showed rapid formation of nitrosated derivatives, as well as 4-methoxybenzylcyanamide and a small amount of 4-methoxybenzylurea. Nitric oxide and nitrous oxide were also evolved, but indirectly, arising from the decomposition of one of the nitrosation products. The current results demonstrate that hydroxyguanidines react with PN to protect cells against PN-mediated injury and may be more effective than the endogenous antioxidants cysteine and glutathione.     

The effect of a synthetic neuromelanin on yield of free hydroxyl radicals generated in model systems

Neuromelanin is an amorphous pigment of the catecholamine origin that accumulates in certain dopaminergic neurons of the substantia nigra of human brain. In Parkinson's disease, there appears to be selective degeneration of the most heavily pigmented neurons of the substantia nigra, and this process has been linked to the presence of neuromelanin. It has been postulated that neuromelanin could increase the risk of oxidative stress reactions. On the other hand, melanin is usually considered to be an efficient antioxidant. Here we analyze experimental conditions that stimulate, or inhibit, antioxidant properties of neuromelanin. Using electron spin resonance (ESR)--spin trapping technique and salicylate hydroxylation assay, we monitored the formation of free hydroxyl radicals generated by a Fenton system in the presence of varying concentration of dopamine-melanin, a synthetic model for neuromelanin. Our data clearly indicate that the antioxidant action of neuromelanin is predominantly due to its ability to sequester redox-active metal ions such as iron. Using direct ESR spectroscopy, we have shown that ferric complexes with neuromelanin are resistant to reduction by mild biological reductants such as ascorbate. We have demonstrated that dopamine-melanin saturated with ferric ions, could enhance the formation of free hydroxyl radicals by redox activation of the ions. Thus, under the conditions that stimulate the release of accumulated metal ions, neuromelanin may actually become an efficient prooxidant. It is conceivable that neuromelanin, which normally is able to protect pigmented dopaminergic neurons against metal-ion related toxicity, could under extreme conditions have a cytotoxic role.     

Selection of a new generation of orally active alpha-ketohydroxypyridine iron chelators intended for use in the treatment of iron overload

The prospect of selecting oral alpha-ketohydroxypyridine chelators intended for clinical use in iron overload has been examined using several animal models of efficacy and toxicity. Studies using iron dextran-loaded mice labelled with 59Fe have shown that only the 1-substituted methyl, ethyl, (n)propyl, allyl, cyclopropyl, 2'-methoxyethyl, 3'-ethoxypropyl, or 2-methyl- or 2-ethyl-3-hydroxypyrid-4-one chelators were orally effective in increasing iron (59Fe) excretion by comparison to intraperitoneally administered desferrioxamine at the same dose (250 mg/kg). In contrast, chelators containing -H, mono- or dihydroxyalkyl and diethoxyethyl 1-substituents caused very little or no increase in iron (59Fe) excretion by the oral or intraperitoneal routes. In vitro studies using ferritin and haemosiderin have shown that equivalent iron release took place with both groups of chelators irrespective of their in vivo effects. In most cases there was no correlation between the n-octanol/water partition coefficient (Kpar) and iron removal efficacy but positive correlation between the lipophilicity and acute or subacute toxicity of these chelators in rats. The most toxic chelator in the chronic toxicity studies in rats was the lipophilic 1,2-diethyl-3-hydroxypyrid-4-one (EL1NEt). The most effective chelator in increasing iron excretion in mice and rabbits was 1-allyl-2-methyl-3-hydroxypyrid-4-one (L1NAII), and the chelator with the highest safety margin in mice and rats was 1,2-dimethyl-3-hydroxypyrid-4-one (L1). Overall the oral effectiveness in increasing iron excretion by these chelators in animals does not appear to be related to their lipophilicity or their ability to mobilise polynuclear iron in vitro but rather to other properties possibly related to their rate of biotransformation and excretion.     

Presence of an expressed beta-tubulin gene (TUBB) in the HLA class I region may provide the genetic basis for HLA-linked microtubule dysfunction

Cyclophosphamide (CP) is associated with significant pulmonary toxicity; however, the mechanism of toxicity is unknown. An in vitro endothelial model of injury was developed to assess the direct toxic effects of CP, CP derivatives and CP metabolites on cultured endothelial cells. Injury to 51Cr-labeled bovine artery pulmonary endothelial (BPAE) cells was quantified by the release of 51Cr from BPAE cells incubated for 18 h with injury expressed as a cytotoxic index. Because CP activation and metabolism occurs primarily in liver, assays assessing CP effects were conducted in the presence of an hepatic microsomal enzyme system. Upon activation, CP produces 4-hydroxycyclophosphamide, acrolein (ACR) and the alkylating metabolite, phosphoramide mustard. Nonactivated CP demonstrated no toxicity to BPAE cells within 18 h; whereas, activated CP induced significant BPAE cell injury in a concentration-dependent manner. Specific metabolites of CP 4-hydroxycyclophosphamide and ACR were markedly more toxic to BPAE cells than phosphoramide mustard. Sulfhydryl-rich compounds, S-2-(3-aminopropylamino)ethylphosphoric acid (WR-2721) and N-acetylcysteine, significantly reduced 4-hydroxycyclophosphamide- and ACR-induced injury but had no significant protective effect against phosphoramide mustard-induced toxicity. These studies suggest 1) CP is not metabolized within pulmonary artery endothelial cells, 2) ACR may be the principal CP metabolite involved in mediating direct injury to pulmonary artery endothelial cells and 3) sulfhydryl-rich agents may be effective in reducing CP-induced damage to critical endothelial cell barriers.     

Inactivation of dehydratase [4Fe-4S] clusters and disruption of iron homeostasis upon cell exposure to peroxynitrite

Phagocytes produce both nitric oxide and superoxide as components of the oxidative defense against pathogens. Neither molecule is likely at physiological concentrations to kill cells. However, two of their reaction products, hydrogen peroxide and peroxynitrite, are strong oxidants, cell-permeant, and toxic. Hydrogen peroxide generates oxidative DNA damage, while the primary mechanism of toxicity of peroxynitrite has not yet been determined. Recent in vitro studies indicated that peroxynitrite is capable of oxidizing the [4Fe-4S] clusters of a family of dehydratases (Hausladen, A., and Fridovich, I. (1994) J. Biol. Chem. 269, 29405-29408; Castro, L., Rodriguez, M., and Radi, R. (1994) J. Biol. Chem. 269, 29409-29415). We demonstrate here that peroxynitrite at 1% of its lethal dose almost fully inactivated the labile dehydratases in Escherichia coli. The rate at which peroxynitrite inactivated the clusters substantially exceeded the rate at which it oxidized thiols or spontaneously decomposed. These results suggest that these dehydratases may be primary targets of peroxynitrite in vivo. Another consequence of the cluster damage was the release of 100 microM iron into the cytosol. During phagocytosis, this intracellular free iron could increase lethal DNA damage by hydrogen peroxide or protein modification by additional peroxynitrite. In response to peroxynitrite challenges, E. coli rapidly sequestered the intracellular free iron using an undefined scavenging system. The iron-sulfur clusters were more gradually repaired by a process that drew iron from its iron-storage proteins. These are likely to be critical events in the struggle between phagocyte and pathogen.     

Evaluation of new iron chelators for clinical use

Evaluation of new chelators for clinical use is limited by the availability of models which will predict the therapeutic safety margin of chelators in iron-overloaded humans such as those with thalassaemia major. Animal models show significant differences with respect to the relative toxicity of different chelators compared with human. These differences can be ascribed to several factors: differences in iron metabolism between different species, human metabolism being significantly more conservative than in rodents or nonhuman primates; differences in drug metabolism between different species which are often difficult to predict from first principles, and difficulties in obtaining iron-overloaded models that are truly representative of transfusional iron overload clinically. These differences have been highlighted by clinical studies on hydroxypyridinone iron chelators such as 1,2-dimethyl-3-hydroxypyridin-4-one (L1, CP20, deferiprone) and 1,2-diethyl-3-hydroxypyridin-4-one (CP94). New tissue culture approaches towards understanding the mechanisms of neutropenia, cytostasis and apoptosis induced by chelators as well as the relative rates of inhibition of non-haem-iron-containing enzymes such as ribonucleotide reductase are predicted to identify chelators with a higher therapeutic safety margin.     

Artemia salina as a test organism for measuring superoxide-mediated toxicity

The purpose of this study was to examine the possibility of using Artemia salina as a test organism in the search for compounds having the ability to protect against superoxide-mediated toxicity. The basic procedure for the assay using Artemia salina was performed as described in previous literature, with minor modifications. We found that Artemia salina are extremely sensitive to menadione bisulfite, a compound whose toxicity is probably mediated by intracellular superoxide generation. Desferrioxamine (desferal), a compound with known protective effects, was shown to display dramatic protective activity in our system. We also observed that an inhibitor of endogenous superoxide dismutase (SOD) activity increased the toxicity of menadione toward Artemia salina. In conclusion, this simple, inexpensive, and convenient assay could be a valuable addition to a screening effort in the search for compounds that will be protective against damage by superoxide or other active oxygen species.     

The rationale, evolution and clinical application of the self-ligating bracket

The binding of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) with the aryl hydrocarbon (AH) receptor and subsequent changes in gene expression have been studied intensively, but the mechanisms by which these lead to toxicity are unclear. We investigated the influence of iron, previously implicated in TCDD-induced hepatic porphyria, in mice with alleles of Ahr that encode receptors with varied affinity for TCDD. The administration of iron to Ahrb-1 C57BL/6J (AH-responsive) mice before a single dose of TCDD (75 micrograms/kg) markedly potentiated not only the hepatic porphyria but also general hepatocellular damage and elevation of plasma hepatic enzymes. The formation of hydroxylated and peroxylated derivatives of uroporphyrins formed from uroporphyrinogen and the induction of a mu-glutathione transferase (GST) were consistent with the operation of an oxidative mechanism. In a comparison of C57BL/6J mice with Ahrb-2 BALB/c (AH-responsive) and Ahrd SWR and DBA/2 (AH-nonresponsive) mice, iron overcame the weak hepatic porphyria and toxicity responses in BALB/c and SWR strains but not in DBA/2. CYP1A isoforms are strongly implicated in the mechanism of porphyria, but activities were lowered by 20-30% with iron treatment, and a comparison of levels between strains did not fully account for the resistance of DBA/2 mice. Studies with the use of gel shift assays and cytosolic aconitase of the capacity of the iron regulatory protein controlling the translation of some iron metabolism proteins showed a significant difference between C57BL/6J and DBA/2 mice after the administration of TCDD. We conclude that iron potentiates both the hepatic porphyria and toxicity of TCDD in susceptible mice in an oxidative process with disturbance of iron regulatory protein capacity. Iron even overcomes the AH-nonresponsive Ahrd allele in the SWR strain but not in DBA/2 mice, which remain resistant.     

Heat shock provides delayed protection against oxidative injury in cultured human umbilical vein endothelial cells

During both mild and severe ischemia, vascular endothelial cells lining large and small vessels of the ischemic organ are exposed to oxygen-derived free radicals resulting in oxidative damage to the organ. Heat shock has been shown to induce thermotolerance and also protect against ischemic injury, possibly via increased synthesis of heat shock proteins (HSPs). We hypothesized that heat shock preconditioning may protect human endothelial cells against oxidative damage. Cultured human umbilical vein endothelial cells (HUVEC) were subjected to heat shock (42 degrees C, 1 h) and allowed to recover for 2 or 20 h, at which times the cells were oxidatively stressed for 1 h by exposing them to 100-200 mumol/l of hydrogen peroxide (H2O2). Cellular damage was assessed immediately and 18 h later by morphology and release of lactate dehydrogenase (LDH). No protection of HUVEC was seen using the 2-hour recovery interval, but a significant protection (P < 0.05) was observed after the 20-hour delay. Northern blot analysis at 1 and 2 h after heating showed induction of HSP-70 mRNA. Western blot analysis demonstrated a significant increase in HSP-72 protein after 2 as well as 20 h of recovery from heat shock, although the amounts of protein at the two times were not significantly different. Furthermore, no differences in the activity of the antioxidant enzyme catalase were observed between heated and unheated HUVEC at 2 and 20 h after heat preconditioning. Thus, heat shock preconditioning induces delayed protection against oxidative injury in HUVEC, and the mechanism of protection appears to involve more than the expression of HSP-72 or activity of catalase.