Initial characterization of the glutamate-cysteine ligase modifier subunit Gclm(-/-) knockout mouse. Novel model system for a severely compromised oxidative stress response

Glutamate-cysteine ligase (GCL) is the rate-limiting enzyme in the GSH biosynthesis pathway. In higher eukaryotes, this enzyme is a heterodimer comprising a catalytic subunit (GCLC) and a modifier subunit (GCLM), which change the catalytic characteristics of the holoenzyme. To define the cellular function of GCLM, we disrupted the mouse Gclm gene to create a null allele. Gclm(-/-) mice are viable and fertile and have no overt phenotype. In liver, lung, pancreas, erythrocytes, and plasma, however, GSH levels in Gclm(-/-) mice were 9-16% of that in Gclm(+/+) littermates. Cysteine levels in Gclm(-/-) mice were 9, 35, and 40% of that in Gclm(+/+) mice in kidney, pancreas, and plasma, respectively, but remained unchanged in the liver and erythrocytes. Comparing the hepatic GCL holoenzyme with GCLC in the genetic absence of GCLM, we found the latter had an approximately 2-fold increase in K(m) for glutamate and a dramatically enhanced sensitivity to GSH inhibition. The major decrease in GSH, combined with diminished GCL activity, rendered Gclm(-/-) fetal fibroblasts strikingly more sensitive to chemical oxidants such as H(2)O(2). We conclude that the Gclm(-/-) mouse represents a model of chronic GSH depletion that will be very useful in evaluating the role of the GCLM subunit and GSH in numerous pathophysiological conditions as well as in environmental toxicity associated with oxidant insult.

Mitochondrial reactive oxygen production is dependent on the aromatic hydrocarbon receptor

2,3,7,8-Tetrachlorodibenzo-p-dioxin (dioxin; TCDD) is a pervasive environmental contaminant that induces hepatic and extrahepatic oxidative stress. We have previously shown that dioxin increases mitochondrial respiration-dependent reactive oxygen production. In the present study we examined the dependence of mitochondrial reactive oxygen production on the aromatic hydrocarbon receptor (AHR), cytochrome P450 1A1 (CYP1A1), and cytochrome P450 1A2 (CYP1A2), proteins believed to be important in dioxin-induced liver toxicity. Congenic Ahr(-/-), Cyp1a1(-/-) and Cyp1a2(-/-) knockout mice, and C57BL/6J inbred mice as their Ahr/Cyp1a1/Cyp1a2(+/+) wild-type (wt) counterparts, were injected intraperitoneally with dioxin (15 microg/kg body weight) or corn-oil vehicle on 3 consecutive days. Liver mitochondria were examined 1 week following the first treatment. The level of mitochondrial H(2)O(2) production in vehicle-treated Ahr(-/-) mice was one fifth that found in vehicle-treated wt mice. Whereas dioxin caused a rise in succinate-stimulated mitochondrial H(2)O(2) production in the wt, Cyp1a1(-/-), and Cyp1a2(-/-) mice, this increase did not occur with the Ahr(-/-) knockout. The lack of H(2)O(2) production in Ahr(-/-) mice was not due to low levels of Mn(2+)-superoxide dismutase (SOD2) as shown by Western immunoblot analysis, nor was it due to high levels of mitochondrial glutathione peroxidase (GPX1) activity. Dioxin decreased mitochondrial aconitase (an enzyme inactivated by superoxide) by 44% in wt mice, by 26% in Cyp1a2(-/-) mice, and by 24% in Cyp1a1(-/-) mice; no change was observed in Ahr(-/-) mice. Dioxin treatment increased mitochondrial glutathione levels in the wt, Cyp1a1(-/-), and Cyp1a2(-/-) mice, but not in Ahr(-/-) mice. These results suggest that both constitutive and dioxin-induced mitochondrial reactive oxygen production is associated with a function of the AHR, and these effects are independent of either CYP1A1 or CYP1A2.

Dioxin exposure is an environmental risk factor for ischemic heart disease

Epidemiologic studies have linked dioxin exposure to increased mortality caused by ischemic heart disease. To test the hypothesis that dioxin exposure may constitute an environmental risk factor for atherosclerosis, we exposed C57BL/6J mice to 5 microg/kg of dioxin daily for 3 d, and measured various molecular and physiological markers of heart disease. Dioxin treatment led to an increase in the urinary excretion of vasoactive eicosanoids and an elevation in the mean tail-cuff blood pressure. In addition, dioxin exposure led to an increase in triglycerides, but not in high-density lipoproteins, in both Apoe(+/+) mice and in hyperlipidemic Apoe(-/- mice. Dioxin exposure also led to an increase in low-density lipoproteins in Apoe(-/-) mice. After treatment, dioxin was associated with low-density lipoprotein particles, which might serve as a vehicle to deliver the compound to atherosclerotic plaques. Dioxin treatment of vascular smooth-muscle cells taken from C57Bl/6J mice resulted in the deregulation of several genes involved in cell proliferation and apoptosis. Subchronic treatment of Apoe(-/-) mice with dioxin (150 ng/kg, three times weekly) for 7 or 26 wk caused a trend toward earlier onset and greater severity of atherosclerotic lesions compared to those of vehicle treated mice. These results suggest that dioxin may increase the incidence of ischemic heart disease by exacerbating its severity.

Induction of cellular oxidative stress by aryl hydrocarbon receptor activation

The aryl hydrocarbon receptor (AHR) has long been associated with the induction of a battery of genes involved in the metabolism of foreign and endogenous compounds. Depending on experimental conditions, AHR can mediate either activation or amelioration of chemical toxicity. For the past decade, evidence has mounted that AHR is associated with a cellular oxidative stress response that must be considered when evaluating the mechanism of action of xenobiotics capable of activating AHR, or capable of metabolic activation by enzymes encoded by genes under control of AHR. In this review, we have evaluated the diverse mechanisms by which AHR generates an oxidative stress response, including inflammation, antioxidant and prooxidant enzymes and cytochrome P450. A review of the regulation of Ahr transcription and functional polymorphisms especially related to oxidative stress is also included. We have carefully avoided placing a value judgment on the degree of toxicity produced by such a response, in view of the realization that an oxidative response is involved in many normal physiological processes. Since the interface between physiological, adaptive and toxicological responses elicited by the AHR-mediated oxidative stress response is not clearly defined, it behooves the researcher to evaluate both toxicological and physiological features of the response.

Decrease in 4-aminobiphenyl-induced methemoglobinemia in Cyp1a2(-/-) knockout mice

Methemoglobin formation, as well as hemoglobin or DNA adducts, are useful biomarkers of occupational exposure to certain arylamines. It has been suggested that, in liver from animals not treated with a cytochrome P450 (CYP) inducer, hepatic CYP1A2 is the major P450 involved in N-hydroxylation. This is the first step in the metabolic activation of many arylamines, such as the human urinary bladder carcinogen 4-aminobiphenyl (ABP). The product of this catalytic step, N-hydroxy-4-ABP, reacts in the blood with oxyhemoglobin to form methemoglobin and nitrosobiphenyl. We therefore examined the role of CYP1A2 in causing methemoglobinemia in ABP-treated Cyp1a2(-/-) knockout mice. Application of ABP (100 micromol/kg body wt) to the skin resulted in a marked depletion in the levels of the hepatic thiols (reduced glutathione and cysteine) after 2 h, which rebounded to basal levels 24 h later, and we found no differences between the Cyp1a2(-/-) and wild-type Cyp1a2(+/+) animals. Unexpectedly, the methemoglobin levels were significantly (p < 0.05) higher in Cyp1a2(-/-) than Cyp1a2(+/+) mice at 2, 7, and 24 h following topical ABP. Treatment with dioxin, 24 h prior to ABP, decreased methemoglobin levels by about half at each of the time points in both the Cyp1a2(-/-) and Cyp1a2(+/+) mice. These data suggest that CYP1A2 does not play a positive role in methemoglobin formation via the activation of ABP; rather, the absence of CYP1A2 enhances ABP-induced methemoglobinemia. Because liver CYP1A2 levels are known to vary more than 60-fold between humans, our findings may be relevant to patients who are exposed to arylamines in the workplace.

Glutamate-cysteine ligase modifier subunit: mouse Gclm gene structure and regulation by agents that cause oxidative stress

Glutamate-cysteine ligase is a heterodimer comprising a modifier (GCLM) and a catalytic (GCLC) subunit. In mouse Hepa-1c1c7 hepatoma cell cultures, we found that tert-butylhydroquinone (tBHQ; 50 microM) induces the GCLM and GCLC mRNAs approximately 10- and approximately 2-fold, respectively, and that these increases primarily reflect de novo transcription. We determined that the mouse Gclm gene has seven exons, spanning 22.3 kb; all exons, intron-exon junctions, and 4.7 kb of 5'-flanking region were sequenced. By RNase protection analysis, we identified two major and several minor transcription start-site clusters over a 300-bp region. The Gclm 5'-flanking region is GC-rich and lacks a canonical TATA box. Transient and stable transfection studies, using luciferase reporter constructs containing incremental Gclm 5'-flanking deletions (4.7-0.5 kb), showed high basal activity but only modest ( approximately 2-fold) inducibility by tBHQ. The only candidate motif for oxidative stress regulation (in the 4.7-kb region we sequenced) is a putative inverted electrophile response element (EPRE) 9 bp upstream from the 5'-most transcription start-site. Site-directed mutagenesis of this -9 EPRE demonstrated minimal (30-40%) decreases in tBHQ induction and no effect on basal activity-suggesting that this EPRE might be necessary but not sufficient. The nuclear erythroid factor-2 (NEF2)-related factor-2 (NRF2) is known to transactivate via EPRE motifs. In the presence of co-transfected NRF cDNA expression vector, however, no increase in Gclm promoter activity was observed. Thus, the endogenous Gclm gene shows robust transcriptional activation by tBHQ in the intact Hepa-1 cell, but reporter constructs containing up to 4.7 kb of promoter (having only the one EPRE at -9) demonstrate a disappointing response, indicating that the major tBHQ-responsive regulatory element of the mouse Gclm gene must exist either further 5'- or 3'-ward of the 4.7-kb region studied.

Dioxin increases reactive oxygen production in mouse liver mitochondria

Dioxin (2,3,7,8-tetrachlorodibenzo-p-dioxin; TCDD) causes an oxidative stress response in liver and several extrahepatic tissues. The subcellular sources and underlying mechanisms of dioxin-induced reactive oxygen, however, are not well understood. In this study, we examined whether mitochondria, organelles that consume the majority of cellular oxygen, might be a source of dioxin-induced reactive oxygen. Female C57BL/6 mice were treated with dioxin (15 microg/kg body wt ip) on 3 consecutive days, and liver mitochondria were examined at 1, 4, and 8 weeks after the first treatment. Mitochondrial aconitase activity, an enzyme inactivated by superoxide, was decreased by 44% at 1 week, 22% at 4 weeks, and returned to control levels at 8 weeks. Dioxin elevated succinate-stimulated mitochondrial H2O2 production twofold at 1 and 4 weeks; H2O2 production remained significantly elevated at 8 weeks. The enhanced H2O2 production was due to neither increased Mn-superoxide dismutase activity nor decreased mitochondrial glutathione peroxidase activity. Dioxin treatment augmented mitochondrial glutathione, but not glutathione disulfide levels, a result that might be explained by increased mitochondrial glutathione reductase activity. Liver ATP levels were significantly lowered at 1 and 4 weeks, the peak times of mitochondrial reactive oxygen production. Increased dioxin-stimulated reactive oxygen at 1 and 4 weeks did not appear to be related to the observed decrease in cytochrome oxidase activity, since State 3 and State 4 respiration were not diminished. To our knowledge, this is the first report to show that dioxin increases mitochondrial respiration-dependent reactive oxygen production, which may play an important role in dioxin-induced toxicity and disease.

Benzo[a]pyrene-induced toxicity: paradoxical protection in Cyp1a1(-/-) knockout mice having increased hepatic BaP-DNA adduct levels

Previous studies have shown that cytochrome P450 1A1 (CYP1A1), CYP1B1, and prostaglandin-endoperoxide synthase (PTGS2) are inducible by benzo[a]pyrene (BaP) and 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD, dioxin), and all three metabolize BaP to reactive DNA-binding intermediates and excreted products. Because these three enzymes show differing patterns of basal levels, inducibility, and tissue-specific expression, animal studies are necessary to delineate the role of CYP1A1 in BaP-mediated toxicity. In mice receiving large daily doses of BaP (500 mg/kg i.p.), Cyp1a1(-/-) knockout mice are protected by surviving longer than Cyp1a1(+/-) heterozygotes. We found that a single 500 mg/kg dose of BaP induces hepatic CYP1A1 mRNA, protein, and enzyme activity in Cyp1a1(+/-) but not in Cyp1a1(-/-) mice; TCDD pretreatment increases further the CYP1A1 in Cyp1a1(+/-) but not Cyp1a1(-/-) mice. Although a single 500 mg/kg dose of BaP was toxic to Cyp1a1(+/-) mice (serum liver enzyme elevated about 2-fold above control levels at 48 h), Cyp1a1(-/-) mice displayed no hepatotoxicity. Unexpectedly, we found 4-fold higher BaP-DNA adduct levels in Cyp1a1(-/-) than in Cyp1a1(+/-) mice; TCDD pretreatment lowered the levels of BaP-DNA adducts in both genotypes, suggesting the involvement of other TCDD-inducible detoxification enzymes. BaP was cleared from the blood much faster in Cyp1a1(+/-) than Cyp1a1(-/-) mice. Our results suggest that absence of the CYP1A1 enzyme protects the intact animal from BaP-mediated liver toxicity and death, by decreasing the formation of large amounts of toxic metabolites, whereas much slower metabolic clearance of BaP in Cyp1a1(-/-) mice leads to greater formation of BaP-DNA adducts.

Glutathione protects chemokine-scavenging and antioxidative defense functions in human RBCs

Oxidant stress, in vivo or in vitro, is known to induce oxidative changes in human red blood cells (RBCs). Our objective was to examine the effect of augmenting RBC glutathione (GSH) synthesis on 1) degenerative protein loss and 2) RBC chemokine- and free radical-scavenging functions in the oxidatively stressed human RBCs by using banked RBCs as a model. Packed RBCs were stored up to 84 days at 1-6 degrees C in Adsol or in the experimental additive solution (Adsol fortified with glutamine, glycine, and N-acetyl-L-cysteine). Supplementing the conventional additive with GSH precursor amino acids improved RBC GSH synthesis and maintenance. The rise in RBC gamma-glutamylcysteine ligase activity was directly proportional to the GSH content and inversely proportional to extracellular homocysteine concentration, methemoglobin formation, and losses of the RBC proteins band 3, band 4.1, band 4.2, glyceraldehyde-3-phosphate dehydrogenase, and Duffy antigen (P < 0.01). Reduced loss of Duffy antigen correlated well with a decrease in chemokine RANTES (regulated upon activation, normal T-cell expressed, and secreted) concentration. We conclude that the concomitant loss of GSH and proteins in oxidatively stressed RBCs can compromise RBC scavenging function. Upregulating GSH synthesis can protect RBC scavenging (free radical and chemokine) function. These results have implications not only in a transfusion setting but also in conditions like diabetes and sickle cell anemia, in which RBCs are subjected to chronic/acute oxidant stresses.

The micronutrient indole-3-carbinol: implications for disease and chemoprevention

This review provides a historical perspective for the development of indole-3-carbinol (I-3-C) as a chemopreventive or therapeutic agent. Early experiments in animal models clearly showed that feeding cruciferous vegetables reduced the incidence of chemical carcinogenesis. Excitement was generated by the finding that these vegetables contained a high content of indole-containing compounds, and I-3-C could by itself inhibit neoplasia. The mechanism of action was linked primarily to the ability of I-3-C and derived substances to induce mixed-function oxidases and phase II antioxidant enzymes by binding and activating the aryl hydrocarbon receptor. Most of the literature on chemoprotection by dietary indole compounds relates to this mechanism of action. Other mechanisms, however, are notable for this class of compounds, including their ability to act as radical and electrophile scavengers; the various ascorbate conjugates of I-3-C (ascorbigens) may be important in this regard. Exciting recent findings have demonstrated that I-3-C and its reaction products can affect cellular signaling pathways, regulate the cell cycle, and decrease tumor cell properties related to metastasis. It does not appear that I-3-C per se is the primary active compound in chemoprotection or chemoprevention. Rather, I-3-C and ascorbate provide the parent compounds for the formation of a myriad of nonenzymatic reaction products that have strong biological potency. We conclude with our thoughts regarding the current status and future directions for the use of I-3-C and related compounds.

Knockout of the mouse glutamate cysteine ligase catalytic subunit (Gclc) gene: embryonic lethal when homozygous, and proposed model for moderate glutathione deficiency when heterozygous

The biosynthesis of reduced glutathione (GSH) is carried out by the enzymes gamma-glutamylcysteine synthetase (GCL) and GSH synthetase. GCL is the rate-limiting step and represents a heterodimeric enzyme comprised of a catalytic subunit (GCLC) and a ("regulatory"), or modifier, subunit (GCLM). The nonhomologous Gclc and Gclm genes are located on mouse chromosomes 9 and 3, respectively. GCLC owns the catalytic activity, whereas GCLM enhances the enzyme activity by lowering the K(m) for glutamate and increasing the K(i) to GSH inhibition. Humans have been identified with one or two defective GCLC alleles and show low GSH levels. As an initial first step toward understanding the role of GSH in cellular redox homeostasis, we have targeted a disruption of the mouse Gclc gene. The Gclc(-/-) homozygous knockout animal dies before gestational day 13, whereas the Gclc(+/-) heterozygote is viable and fertile. The Gclc(+/-) mouse exhibits a gene-dose decrease in the GCLC protein and GCL activity, but only about a 20% diminution in GSH levels and a compensatory increase of approximately 30% in ascorbate-as compared with that in Gclc(+/+) wild-type littermates. These data show a reciprocal action between falling GSH concentrations and rising ascorbate levels. Therefore, the Gclc(+/-) mouse may be a useful genetic model for mild endogenous oxidative stress.

Activation of transcription factors activator protein-1 and nuclear factor-kappaB by 2,3,7,8-tetrachlorodibenzo-p-dioxin

2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD; dioxin), the prototype agonist of the aromatic hydrocarbon (Ah) receptor, is a potent tumor promoter as well as a complete liver carcinogen that produces an oxidative stress response in rodents and in cultured cell lines. It has been proposed that TCDD promotes neoplastic transformation through oxidative signal transduction pathways, which results in activation of immediate-early response transcription factors. To set the stage for a test of this hypothesis, we evaluated the effect of TCDD treatment on the activation of several transcription factors, including those in the nuclear factor-kappaB (NF-kappaB) and activator protein-1 (AP-1) families, which are activated by changes in the redox state of cells. In an extension of prior results, we found that TCDD treatment produced a sustained overexpression of AP-1 for at least 72 hr in wild-type mouse hepatoma Hepa-1 cells, but not in the Ah receptor-deficient derivative c35 or in cytochrome P450-1A1 (CYP1A1)-negative c37 cells. In addition, TCDD treatment caused a significant increase in the DNA binding activity of NF-kappaB, but not in the activities of the other transcription factors tested. AP-1 and NF-kappaB activation were blocked by the thiol antioxidant N-acetylcysteine and by nordihydroguaiaretic acid, an antioxidant and lipooxygenase inhibitor and an inhibitor of the epoxygenase activity of CYP1A1, and did not take place in c35, c37, or in Ah nuclear translator-deficient c4 cells. Hence, sustained activation of these two transcription factors by TCDD is likely to result from a CYP1A1-dependent and Ah receptor complex-dependent oxidative signal. Electrophoretic mobility supershift analyses with specific antibodies showed that most of the increase in NF-kappaB binding activity could be accounted for by increases in p50/p50 complexes. Since these complexes are known to repress NF-kappaB-dependent gene transcription, our results delineate a second molecular mechanism, in addition to the recently found block of tumor necrosis factor-alpha-mediated p50/p65 activation, that may be responsible for the immunosuppresive effects of TCDD.

Determining glutathione and glutathione disulfide using the fluorescence probe o-phthalaldehyde

Because of the importance of glutathione (GSH) and glutathione disulfide (GSSG) in cellular signal transduction, gene regulation, redox regulation, and biochemical homeostasis, accurate determination of cellular glutathione levels is critical. Several procedures have been developed, but many suffer from overestimating GSSG or from cellular substances interfering or competing with GSH determination. Assays based on HPLC, with enzymatic reduction of GSSG by glutathione reductase and NADPH, appear to be valid but are limited in sample throughput and availability of equipment. The fluorescence probe o-phthalaldehyde (OPA, phthalic dicarboxaldehyde) reacts with GSH and has a high quantum yield, yet its use has been limited due to unidentified interfering and fluorescence-quenching substances in liver. This paper describes assay conditions under which these limitations are avoided. By using a phosphate-buffered assay at lower pH, interference with nonspecific reactants is minimal. Since enzymatic reduction is not possible due to the reaction of OPA with NAD(P)H and other stronger reducing agents, leading to an overestimation of GSSG levels, dithionite was used to reduce GSSG. High sample throughput combined with sensitive (20-pmol limit of detection) and accurate determination of GSH and GSSG using OPA is achievable with any monochromatographic spectrofluorometer. Sample preparation and storage conditions are described that return the same levels of GSH and GSSG for at least 4 weeks.

Targeted knockout of Cyp1a1 gene does not alter hepatic constitutive expression of other genes in the mouse [Ah] battery

Using the Cre-lox system, we have generated a cytochrome P450 1A1 Cyp1a1(-/-) knockout mouse by deletion of the translated portions of the Cyp1a1 gene. These mice are viable and demonstrate no obvious phenotype, compared with wild-type littermates. As a first step toward characterizing genes that might be expected to compensate for loss of CYP1A1, constitutive expression of [Ah] gene battery members was examined. In a cultured hepatoma CYP1A1 metabolism-deficient mutant line that does not express Cyp1a2, we have previously shown that constitutive transcriptional up-regulation of other [Ah] gene battery members occurs; these results are consistent with the elevation of a putative endogenous ligand (EL) for the Ah receptor that is a substrate for CYP1A1. The [Ah] battery includes Cyp1a2, NAD(P)H:quinone oxidoreductase (Nqo1), and three other Phase II genes. Examining mRNA, protein, and enzyme activity, we demonstrate that the absence of CYP1A1 has no effect on the hepatic constitutive expression of Cyp1a2 or Nqo1. We postulate that CYP1A1 and CYP1A2 might have overlapping substrate specificity for metabolism of the EL, such that basal CYP1A2 in the liver can compensate for the loss of CYP1A1.

Inhibition of CYP1A1 enzyme activity in mouse hepatoma cell culture by soybean isoflavones

The mechanisms by which soybean- and soybean isoflavone-enriched diets inhibit carcinogenesis are not known. We found that the isoflavones genistin and daidzin, and their respective aglucone forms daidzein and genistein, block 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD; dioxin)-induced CYP1A1 enzyme activity. This inhibition is correlated with the capacity of the isoflavones to prevent CYP1A1-mediated covalent binding of benzo[a]pyrene (BaP) metabolites to DNA. We further evaluated daidzein and genistein, believed to be the active forms of the isoflavones, for the mechanism of the inhibitory process. Although daidzein and genistein appear structurally similar to known aromatic hydrocarbon receptor (AHR) agonists and antagonists, gel mobility shift assays indicated that the isoflavones do not inhibit dioxin-induced activation of the AHR or the accumulation of CYP1A1 mRNA, suggesting that the isoflavones do not act at the transcriptional level. We therefore evaluated the isoflavones for direct effects on the CYP1A1 enzyme. Daidzein and genistein non-competitive with the CYP1A1 substrate BaP for microsomal BaP hydroxylation, with apparent Ki values of 325 microM and 140 microM, respectively. The extent of CYP1A1 inhibition increases with time of preincubation at 37 degrees C, but not at 4 degrees C, in the presence of isoflavone plus NADPH; after 60 min preincubation the inhibition remains non-competitive, with apparent Ki values of 55 microM and 50 microM, respectively. Inhibition is neither prevented nor reversed by the thiol antioxidant dithiothreitol, nor by the iron chelator deferoxamine. Repeated washing of the microsomes does not reverse the inhibition. The dependency on NADPH, temperature and time for inhibition of CYP1A1 suggests that metabolism of either isoflavone or molecular oxygen to reactive species is required. Isoflavone-mediated inhibition of CYP1A1 activity may contribute to the mechanism by which these soybean isoflavones protect against carcinogenesis.

Ozone-induced acute lung injury: genetic analysis of F(2) mice generated from A/J and C57BL/6J strains

Acute lung injury (or acute respiratory distress syndrome) is a devastating and often lethal condition. This complex disease (trait) may be associated with numerous candidate genes. To discern the major gene(s) controlling mortality from acute lung injury, two inbred mouse strains displaying contrasting survival times to 10 parts/million ozone were identified. A/J (A) mice were sensitive [6.6 +/- 1 (SE) h] and C57BL/6J (B) were resistant (20.6 +/- 1 h). The designation for these phenotypes was 13 h, a point that clearly separated their survival time distributions. Our prior segregation studies suggested that survival time to ozone-induced acute lung injury was a quantitative trait, and genetic analysis identified three linked loci [acute lung injury-1, -2, and -3 (Ali1-3, respectively)]. In this report, acute lung injury in A or B mice was characterized histologically and by measuring lung wet-to-dry weight ratios at death. Ozone produced comparable effects in both strains. To further delineate genetic loci associated with reduced survival, a genomewide scan was performed with F(2) mice generated from the A and B strains. The results strengthen and extend our initial findings and firmly establish that Ali1 on mouse chromosome 11 has significant linkage to this phenotype. Ali3 was suggestive of linkage, supporting previous recombinant inbred analysis, whereas Ali2 showed no linkage. Together, our findings support the fact that several genes, including Ali1 and Ali3, control susceptibility to death after acute lung injury. Identification of these loci should allow a more focused effort to determine the key events leading to mortality after oxidant-induced acute lung injury.

Regulation of gene expression by reactive oxygen

Reactive oxygen intermediates are produced in all aerobic organisms during respiration and exist in the cell in a balance with biochemical antioxidants. Excess reactive oxygen resulting from exposure to environmental oxidants, toxicants, and heavy metals perturbs cellular redox balance and disrupts normal biological functions. The resulting imbalance may be detrimental to the organism and contribute to the pathogenesis of disease and aging. To counteract the oxidant effects and to restore a state of redox balance, cells must reset critical homeostatic parameters. Changes associated with oxidative damage and with restoration of cellular homeostasis often lead to activation or silencing of genes encoding regulatory transcription factors, antioxidant defense enzymes, and structural proteins. In this review, we examine the sources and generation of free radicals and oxidative stress in biological systems and the mechanisms used by reactive oxygen to modulate signal transduction cascades and redirect gene expression.

Dioxin causes a sustained oxidative stress response in the mouse

Dioxin (2,3,7,8-tetrachlorodibenzo-p-dioxin; TCDD) is the prototype for environmental agonists of the aromatic hydrocarbon receptor (AHR) that are known to produce multiple adverse effects in laboratory animals as well as humans. Although not directly genotoxic, dioxin is known to increase transformation and mutations in mammalian cell culture and to cause an exaggerated oxidative stress response in the female rat. In humans and mice, however, dioxin-mediated oxidative stress appears to be more subtle, causing a response that has been poorly characterized. Using the female C57BL/6J inbred mouse, we show here that intraperitoneal treatment of 5 micrograms TCDD per kilogram on 3 consecutive days produces a striking, prolonged oxidative stress response: hepatic oxidized glutathione levels increase 2-fold within 1 week, and these effects persist for at least 8 weeks despite no further dioxin treatment. Urinary levels of 8-hydroxydeoxyguanosine--a product of DNA base oxidation and subsequent excision repair--remain elevated about 20-fold at 8 weeks after dioxin treatment, consistent with chronic and potentially promutagenic DNA base damage. These results demonstrate that dioxin exposure does produce a sustained oxidative stress response in the mouse.

Sustained increase in intracellular free calcium and activation of cyclooxygenase-2 expression in mouse hepatoma cells treated with dioxin

2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) is a non-genotoxic environmental pollutant that causes multiple adverse effects in experimental animals and in humans. We show here that TCDD treatment of mouse hepatoma cells causes a rapid mobilization of intracellular calcium both in wild type Hepa-1 cells and in its c2 variant, a cell line that has highly reduced levels of functional aromatic hydrocarbon (Ah) receptor (AHR). In wild type cells, but not in the c2 variant, TCDD treatment leads to a sustained elevation of cytosolic free calcium. TCDD also induces elevated levels of cyclooxygenase-2 (COX-2) mRNA in wild type and in c37, a CYP1A1-deficient cell line, but not in c2 cells. Induction of Cox-2 is in fact dependent on the presence of a functional Ah receptor, since it can be blocked by antisense oligonucleotides to Ah receptor mRNA. Most likely as a consequence of Cox-2 induction, we find a significant increase in the level of 12-hydroxyheptadecatrienoic acid (12-HHT) secreted from TCDD-treated Hepa-1 cells. In addition, we observe elevated levels of 6-keto prostaglandin F1alpha in c2 cells and high levels of secreted prostaglandin F2alpha in c2, c37 and c4, the variant cell line lacking aromatic hydrocarbon nuclear translocator protein. These data suggest that Cox-2 activation by TCDD leads to the release of prostaglandins, eicosanoids and other mediators which may have an important role in the biological and toxic effects of TCDD.

Genetic analysis of ozone-induced acute lung injury in sensitive and resistant strains of mice

Epidemiological studies have found air pollution to be associated with excessive mortality, particularly death from respiratory and cardiovascular causes. Interpretation of these findings is controversial, however, because toxicological mechanisms controlling mortality are uncertain. Susceptibility to many air pollutants entails an oxidative stress response. Accordingly, the best-characterized oxidant air pollutant is ozone, which causes direct oxidative damage of lung biomolecules. An underlying characteristic derived from clinical and epidemiological studies of healthy and asthmatic individuals of all ages is marked variability in the respiratory effects of ozone. This susceptibility difference among humans suggests that genetic determinants may control predisposition to the harmful effects of ozone. Mice also vary considerably in their response to ozone. Moreover, ozone-induced differences in strain responses indicate that susceptibility in mice can be genetically determined. Therefore, we used inbred mice to investigate the genetic determinants of acute lung injury. Recombinant inbred (RI) strains derived from A/J (A) mice (sensitive) and C57BL/6J (B) mice (resistant) showed a continuous phenotypic pattern, suggesting a multigenic trait. Quantitative trait locus and RI analyses suggested three major loci linked to ozone susceptibility. Differences in phenotype ratios among the reciprocal back-crosses were consistent with parental imprinting. These findings implicate various genetic and epigenetic factors in individual susceptibility to air pollution.

Genetic differences in alcohol drinking preference between inbred strains of mice

Genetic factors are known to influence the preference for drinking alcohol-in humans as well as certain inbred strains of laboratory animals. Here we examined the possible role of the aromatic hydrocarbon receptor (AHR) in alcohol-preferring C57BL/6J (B6, high-affinity AHR) and alcohol-avoiding DBA/2J (D2, low-affinity AHR) inbred mouse strains, and in the two congenic lines B6.D2-Ahrd (> 99% B6 genome with the D2 low-affinity AHR) and D2.B6-Ahrb-1 (> 99% D2 genome with the B6 high-affinity AHR). This laboratory had previously shown an association between resistance to intraperitoneal ethanol-induced toxicity and the high-affinity AHR. Offering the choice between drinking water and 10% ethanol, we found that alcohol preference is three- to four-fold greater in B6 than D2 mice, as well as three- to four-fold greater in B6.D2-Ahrd than D2.B6-Ahrb-1 mice-indicating that alcohol preference is AHR-independent. The prototype AHR agonist 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD; dioxin) did not affect the rates of chronic alcohol consumption in B6 or D2 mice, suggesting that dioxin-inducible metabolism does not play a major role in alcohol drinking preference. In B6 mice, we found that oral treatment with the aldehyde dehydrogenase (ALDH) inhibitor disulfiram decreased alcohol preference by 50%, whereas oral treatment of the catalase inhibitor 3-amino-1,2,4-triazole increased alcohol drinking preference by 15-20%. Although liver and brain ALDH activities were both significantly higher in D2 than B6, these activities were not related to alcohol consumption. Hepatic and brain catalase activities, on the other hand, were two- to three-fold higher in D2 and D2.B6-Ahrb-1 mice, compared with that in B6 and B6.D2-Ahrd. Furthermore, brain acetaldehyde levels were inversely related to the quantity of alcohol voluntarily consumed. We conclude that the alcohol drinking preference between the B6 and D2 inbred mouse strains is independent of the Ah receptor-but is genetically determined, in part, by the level of brain catalase activity which, in turn, regulates brain acetaldehyde concentrations.

Effects of glutathione and pH on the oxidation of biomarkers of cellular oxidative stress

Cellular oxidative stress is associated with such pathological conditions as arteriosclerosis, inflammatory diseases and cancer. The oxidation of the biomarkers. 2',7'-dichlorofluorescin (DCFH), 2-deoxyribose, and lipid peroxidation are often used to assess the status of oxidative stress in cells and tissues. Since high levels of reduced glutathione (GSH) and acidic conditions have been associated with diminished chemical lethality, we evaluated the influence of these parameters on the cellular response to oxidative stress. We used a cultured hepatocyte line (ch/ch cells) that is susceptible to oxidative toxicity. A hydroxyl radical-generating system consisting of H2O2, ascorbate and iron produced a pH-dependent lethality, with complete cell killing at pH 7.4 and none at pH 6.8. Lethality correlated with the depletion of intracellular GSH, and with an increase in DNA fragmentation. The influence of GSH and pH was assessed for DCFH and 2-deoxyribose oxidation, and for lipid peroxidation. The oxidation of DCFH and 2-deoxyribose was inhibited by GSH, with about 4-fold greater inhibition efficacy at pH 6.8 than at pH 7.4 [IC50 values (microM GSH) for pH 6.8 and 7.4, respectively: DCFH = 7 and 30; 2-deoxyribose = 125 and 490]. GSH did not affect lipid peroxidation at either pH, even at a high intracellular concentration of 10 mM. We conclude: 1) GSH is not inhibiting DCFH and 2-deoxyribose oxidation by simply quenching reactive oxygen (hydroxyl radical or perferryl oxygen), since GSH did not inhibit lipid peroxidation: 2) the protonated form GSH is more likely to be the inhibitory species rather than GS-, since even in the simple cell-free systems lower pH inhibited biomarker oxidation; and; 3) hydroxyl radical may not be the primary intracellular oxidant of DCFH, since intracellular GSH concentrations are typically 10- to 100-fold higher than the IC50 values for GSH inhibiting reactive oxygen-mediated DCFH oxidation.

Characterization of novel indenoindoles. Part I. Structure-activity relationships in different model systems of lipid peroxidation

Structure-activity relationships are presented for some representative compounds from a novel series of potent inhibitors of lipid peroxidation. The compounds are indenoindole derivatives with oxidation potentials in organic solvents of between 0.2 and 1.5 V. Two of these compounds, cis-5,5a,6,10b-tetrahydro-9-methoxy-7-methylindeno[2,1-b]indole (H 290/51) with an oxidation potential of 0.32 V and cis-4b,5,9b,10- tetrahydro-8-methoxy-6-methylindeno[1,2-b]indole (H 290/30) with an oxidation potential of 0.30 V, have been tested more extensively and compared with reference compounds in several pharmacological models of lipid peroxidation. The inhibitory potencies (pIC50) of the compounds in respect to Fe/Ascorbate-induced production of thiobarbituric acid-reactive substances (TBARS) in a suspension of purified soybean lecithin were calculated. These data are 8.2 for H 290/51; 8.0 for H 290/30; 5.6 for vitamin E; and 6.6 for butylated hydroxytoluene (BHT). In isolated rat renal tissue subjected to hypoxia and reoxygenation, the potency for inhibition of TBARS formation is 6.9 for H 290/51, 6.9 for H 290/30, and <5 for vitamin E. In oxidative modification of low-density lipoproteins (LDL) induced by mouse peritoneal macrophages, the corresponding pIC50 values for TBARS inhibition for each compound are: 8.7, 8.3, <5, and 6.9, respectively. It is concluded that the synthetic indenoindoles are potent antioxidants. The results suggest that indenoindoles such as H 290/51 and H 290/30 could be useful as therapeutic agents in pathophysiological situations where lipid peroxidation plays an important role.

Molecular modeling parameters predict antioxidant efficacy of 3-indolyl compounds

Many dietary constituents, such as indole-3-carbinol, are chemoprotective in toxicity and carcinogenicity bioassays. Indole-3-carbinol and related congeners appear to protect partly via radical and electrophile scavenging. To develop better chemoprotective indoles with lower intrinsic toxicity, we performed molecular graphic and quantum-mechanical analyses of model indolyl compounds to ascertain the determinant molecular features for antioxidant activity. We examined eight structurally related 3-indolyl compounds for relationships between antioxidation potential (using in vitro lipid peroxidation assays) and electronic, polar, and steric parameters, including bond dissociation energies, bond lengths, dipole moments, electronic charge densities, and molecular size parameters. Electronic features of the 3-methylene carbon and 1-nitrogen were not predictive of antioxidative potency due to extensive charge delocalization of the cation radical following electron abstraction from the nitrogen. Antioxidant efficacy of 3-indolyl compounds was most strongly predicted by molecular size parameters and by the energy of electron abstraction as calculated from the difference in heat of formation between the parent compound and its cation radical. A highly predictive multiple linear regression correlation model (r2 = 0.97) was obtained using the parameters of heat of formation, molecular weight, log P, and diplole moment.

The Y-box motif mediates redox-dependent transcriptional activation in mouse cells

We show here that the OxyR response element (ORE) in the bacterial oxyR promoter can also function as a redox-dependent enhancer in mammalian cells. Fusion of ORE to an SV40 basal promoter driving chloramphenicol acetyltransferase (CAT) expression confers H2O2 inducibility to expression of the cat gene in mouse Hepa-1 hepatoma cells. Nuclear extracts from these cells contain DNA-binding proteins that specifically interact with ORE DNA, cannot be completed by cognate oligonucleotides to AP-1 or NF kappa B, and are constitutively expressed, since treatment with H2O2 causes no detectable changes in binding activity or DNA-protein interaction. Recombinant cDNA clones that express ORE-binding proteins were isolated from a mouse hepatoma expression library and found to be representatives of two different members of the murine Y-box family of transcription factors. Canonical Y-box and ORE oligonucleotides compete with each other for binding to Y-box proteins in gel shift assays and antibodies to FRGY2, a Xenopus Y-box protein, supershift both Y-box and ORE DNA-protein complexes. In addition, antisense oligonucleotides to mouse YB-1 mRNA abolish induction of ORE-mediated cat expression by H2O2, and luciferase reporter constructs containing ORE, or the Y-box from the human MHC class II HLA-DQ gene, exhibit identical dose-dependent H2O2 inducibilities, which can be abolished by addition of 2-mercaptoethanol to the culture medium. These results suggest that the Y-box proteins may be an integral component of a eukaryotic redox signaling pathway.

Response of [Ah] battery genes to compounds that protect against menadione toxicity

We have studied the response of genes in the dioxin-inducible [Ah] battery to three compounds that protect mouse hepatoma cells (Hepa-1c7c7 wild-type, wt) against menadione toxicity. Pretreatment of wt cells with 25 microM 5,10-dihydroindenol[1,2-b]indole (DHII), 25 microM tert-butylhydroquinone (tBHO) or 10 microM menadione itself, generated substantial protection against toxicity produced by subsequent menadione exposure. The gene response was examined in wt cells, and three mutant lines: CYP1A1 metabolism-deficient (c37 or P1-); nuclear translocation-impaired (c4 or nt-); and AHR-deficient (c2 or r-, containing < 10% of normal functional receptor levels). DHII treatment of wt cells for 12 hr markedly elevated the enzyme activities and mRNA levels of genes in the [Ah] battery: aryl hydrocarbon hydroxylase (Cyp1a1), NAD(P)H:menadione oxidoreductase (Nmol), cytosolic aldehyde dehydrogenase class 3 (Ahd4), and UDP-glucuronosyltransferase form 1*06 (Ugt1*06). Treatment of the c4 and c2 cells with DHII failed to induce mRNA levels of the genes, indicating that induction of the [Ah] gene battery by DHII is aromatic hydrocarbon receptor (AHR)-mediated. On the other hand, neither tBHO nor menadione caused increases in CYPlAl mRNA, but tBHQ significantly enhanced the NMO1, AHD4, and UGT1*06 mRNA levels in all three mutant cell lines. In conclusion, we expect one or more putative electrophile response elements (EpRE), previously found in the regulatory regions of the murine Nmol, Ahd4, and ugt1*06 genes, to be functional in responding to phenolic antioxidants.

Enzyme induction by L-buthionine (S,R)-sulfoximine in cultured mouse hepatoma cells

Induction of Phase II enzymes of the [Ah] gene battery by L-buthionine (S,R)-sulfoximine (BSO) and other agents was examined in mouse hepatoma Hepa-1c1c7 cells. BSO, a nonelectrophilic inhibitor of gamma-glutamylcysteine synthetase (GCS), is routinely used to examine the toxicological implications of GSH depletion. Exposure to BSO for 24 h produced a 75-85% depletion of GSH levels, proportional to the inhibition of GCS activity, as well as small increases in the UDP-glucuronosyltransferase (UGT, 60%) and glutathione transferase (GST, 30%) enzyme activities in Hepa-1 wild-type (wt) cells. However, for the NAD(P)H:menadione oxidoreductase (NMO1) and cytosolic aldehyde dehydrogenase class 3 (AHD4) enzyme activities, BSO produced larger increases (110% and 170%, respectively). The mechanisms of NMO1 and AHD4 induction were examined further. In Hepa-1 wt cells, NMO1 and AHD4 activities were increased by the aromatic hydrocarbon inducer 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and by the electrophile tert-butylhydroquinone (tBHQ), known inducing agents for these enzymes. However, NMO1 and AHD4 were induced in Ah receptor nuclear translocation-defective mutant (c4) cells by BSO and tBHQ, but not by TCDD, suggesting that the induction by BSO and tBHQ is not Ah receptor-mediated. In wt cells, N-acetylcysteine produced a concentration-dependent increase in intracellular cysteine levels, but not GSH levels, in the absence or presence of BSO. Furthermore, N-acetylcysteine had no effect on NMO1 activity under any conditions examined, suggesting that GSH levels per se, rather than change in overall thiol status, might be mediating increased NMO1 activity.(ABSTRACT TRUNCATED AT 250 WORDS)

Retardation of benzo[a]pyrene-induced epidermal tumor formation by the potent antioxidant 4b,5,9b,10-tetrahydroindeno[1,2-b]indole

The ability of the potent antioxidant, 4b,5,9b,10-tetrahydroindeno[1,2-b]indole (THII), to inhibit tumor formation by topically-applied benzo[a]pyrene was evaluated using a complete carcinogenicity mouse skin bioassay. THII was administered by direct application to the skin, in the food or through the drinking water. In each case, THII increased the average time until the appearance of tumors by 4 weeks, and also decreased the total number of tumors compared with benzo[a]pyrene alone. These protective effects corresponded with the ability of THII to inhibit benzo[a]pyrene- or 12-O-tetradecanoylphorbol-13-acetate-induced epidermal ornithine decarboxylase activity, a biomarker of tissue proliferation in skin of the treated animals. This is the first report of an antioxidant administered in food or water inhibiting chemically induced skin carcinogenesis.

Regulation of [Ah] gene battery enzymes and glutathione levels by 5,10-dihydroindeno[1,2-b]indole in mouse hepatoma cell lines

The murine aromatic hydrocarbon ([Ah]) gene battery consists of at least six genes that code for two functionalizing (Phase I) enzymes and four non-functionalizing (Phase II) enzymes. These enzymes are induced by compounds such as aromatic hydrocarbons and 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) that bind to the cytosolic Ah receptor protein. Studies in rodents indicate that certain enzymes of this battery, namely cytochrome P4501A1 (CYP1A1), UDP-glucuronosyltransferase (UGT1*06) and NAD(P)H: quinone acceptor oxidoreductase (NMO1) are induced by the synthetic antioxidant 5,10-dihydroindeno[1,2-b]indole (DHII). The induction of [Ah] gene battery enzymes and the levels of reduced glutathione (GSH) were examined in mouse Hepa-1c1c7 hepatoma wild-type cells (wt), a CYP1A1 metabolism-deficient mutant (c37) and an Ah receptor nuclear translocation-defective mutant (c4). DHII and TCDD increased the activities of ethoxyresorufin O-deethylase, an indicator of CYP1A1 activity, as well as NMO1, UGT1*06, cytosolic aldehyde dehydrogenase class 3 and glutathione S-transferase form A1 in wt cells, but had little or no induction effect in c37 or c4 cells. DHII and TCDD differed in their effects on GSH levels; while DHII increased GSH levels 3-fold in wt, but not at all in c37 or c4 cells, TCDD had no effect on GSH levels in any cell type. However, GSH levels were enhanced in both wt and c4 cells by tert-butyl hydroquinone (TBHQ). L-Buthionine S,R-sulfoximine, an inhibitor of gamma-glutamylcysteine synthetase, prevented DHII-induced increases in wt cell GSH. The increase in GSH levels occurred after 8 h, while the induction of enzymes occurred within 4 h. The induction of the higher GSH levels in wt cells by DHII and TBHQ correlated with increases in intracellular levels of the GSH precursor thiol cysteine, as well as with increased activities of gamma-glutamylcysteine synthetase, the rate-limiting enzyme of GSH synthesis. However, TBHQ-mediated GSH increases in c4 cells were accompanied by increased gamma-glutamylcysteine synthetase activity with no change in intracellular cysteine concentration. The results suggest that DHII induction of [Ah] gene battery enzymes requires a functional Ah receptor, but not the functional gene product CYP1A1. Furthermore, metabolism, possibly via CYP1A1, appears to be required for DHII to enhance intracellular levels of cysteine and GCS activity that result in higher GSH levels.

Oxidation pathways for the intracellular probe 2',7'-dichlorofluorescein

The oxidation of 2',7'-dichlorofluorescin (DCFH) to a fluorescent product is currently used to evaluate oxidant stress in cells. However, there is considerable uncertainty as to the enzymatic and nonenzymatic pathways that may result in DCFH oxidation. Iron/hydrogen peroxide-induced DCFH oxidation was inhibited by catalase or by the hydroxyl radical scavenger dimethylsulfoxide; however, superoxide dismutase (SOD) had no effect on DCFH oxidation. The formation of hydroxyl radical (indicated by the oxidation of salicylic acid to 2,3-dihydroxybenzoic acid) was proportional to DCFH oxidation, suggesting that the hydroxyl radical is responsible for the iron/peroxide-mediated oxidation of DCFH. Utilizing a superoxide generating system consisting of hypoxanthine/xanthine oxidase, oxidation of DCFH was unaffected by SOD, catalase or desferoxamine, and stimulated by removing hypoxanthine from the reaction mixture. In contrast, SOD or elimination of hypoxanthine abolished superoxide formation. In addition, potassium superoxide did not support the oxidation of DCFH. Thus, superoxide is not involved in DCFH oxidation. Boiling xanthine oxidase eliminated its concentration-dependent oxidation of 1 microM DCFH, indicating that xanthine oxidase can enzymatically utilize DCFH as a high affinity substrate. Kinetic studies of the oxidation of DCFH by xanthine oxidase indicated a Km(app) of 0.62 microM. Hypoxanthine competed with DCFH with a Ki(app) of 1.03 mM. These studies suggest that DCFH oxidation may be a useful indicator of oxidative stress. However, other types of cellular damage may produce DCFH oxidation.(ABSTRACT TRUNCATED AT 250 WORDS)

The role of thiols in mitochondrial susceptibility to iron and tert-butyl hydroperoxide-mediated toxicity in cultured mouse hepatocytes

Cultured hepatocytes derived from the newborn mutant c14CoS/c14CoS mouse (14CoS/14CoS cells) have 3-fold higher levels of reduced glutathione (GSH) and greater resistance to menadione toxicity than hepatocytes derived from the wild-type cch/cch mouse (ch/ch cells). Therefore, we used these cell lines to examine mechanisms of oxidative stress produced by iron and tert-butyl hydroperoxide (TBHP). Both cell types were resistant to 25 microM Fe2+ toxicity in the absence of added TBHP. However, in the presence of Fe2+, striking differences in susceptibility to TBHP toxicity between the cell types were observed. With 25 microM Fe2+, ch/ch cells showed TBHP concentration-dependent toxicity, with total lethality at 500 microM; in contrast, 14CoS/14CoS cells were completely resistant to the lethal effects of this concentration of TBHP. Concentration-dependent TBHP-mediated increases in cytosolic Ca2+, pH, and GSSG/GSH ratios, and decreases in GSH levels, were evident in ch/ch cells. 14CoS/14CoS cells exhibited concentration-dependent TBHP-mediated changes in GSH and GSSG/GSH ratios, but cytosolic Ca2+ and pH remained at control levels. Mitochondrial GSH pools were also diminished by TBHP, although there was no selective depletion; mitochondrial GSH remained at about 14% of total cellular GSH. Both cell types exhibited the same time-dependent decrease in plasma membrane protein thiols and a time-dependent increase in plasma membrane protein carbonyls. However, only ch/ch cells displayed a time-dependent depletion of mitochondrial protein thiols, concomitant with an increase in mitochondrial protein carbonyls, while 14CoS/14CoS cells were resistant to such changes. All of the effects produced by TBHP were prevented by desferoxamine.(ABSTRACT TRUNCATED AT 250 WORDS)

Mechanisms of protection from menadione toxicity by 5,10-dihydroindeno[1,2,-b]indole in a sensitive and resistant mouse hepatocyte line

Established cell lines derived from newborn livers of c14CoS/c14CoS and cch/cch mice have been shown to be genetically resistant (14CoS/14CoS cells) or susceptible (ch/ch cells) to menadione toxicity. These differences are due in part to relatively higher levels of reduced glutathione (GSH) and NAD(P)H:menadione oxidoreductase (NMO1) activity in the 14CoS/14CoS cells. The indolic membrane-stabilizing antioxidant 5,10-dihydroindeno[1,2-b]indole (DHII) was shown previously to protect against various hepatotoxicants in vivo and in primary rat hepatocytes. This report describes how the 14CoS/14CoS and ch/ch cell lines provide a valuable experimental system to distinguish the mechanism of chemoprotection by DHII from menadione toxicity. The addition of 25 microM DHII produced a time-dependent decrease in menadione-mediated cell death in 14CoS/14CoS cells, with little effect on ch/ch cell viability. The maximum protective effect occurred at 24 hr, although the concentration of DHII remained constant for 48 hr. The protective effect of DHII correlated with enhanced glutathione levels (234% increase at 24hr), as well as induction of four enzymes involved in the detoxification and excretion of menadione: NAD(P)H:menadione oxidoreductase (NMO1, quinone reductase), glutathione reductase, glutathione transferase (GST1A1), and UDP glucuronosyltransferase (UGT1*06), with 24-hr maximum induction of 707, 201, 171 and 198%, respectively. Other biotransformation enzymes not directly involved in menadione metabolism (glutathione peroxidase, cytochromes P4501A1 and P4501A2, copper-, zinc-dependent superoxide dismutase, and NADPH cytochrome c oxidoreductase) were not induced by DHII. Menadione-stimulated superoxide production was inhibited 50% by DHII only in 14CoS/14CoS cells, and the inhibition required 24-hr preincubation. Pretreatment with DHII also protected both cell types against the menadione-mediated depletion of GSH, and the increase in percent (oxidized glutathione GSSG), an indicator of oxidative stress. These results suggest that DHII does not protect against menadione toxicity by virtue of its antioxidant or membrane-stabilizing properties. Rather, it acts by inducing a protective enzyme profile that migates redox cycling and facilitates excretion of menadione.

Menadione toxicity in two mouse liver established cell lines having striking genetic differences in quinone reductase activity and glutathione concentrations

Established cell lines derived from newborn livers of c14CoS/c14CoS and cch/cch mice were examined for differences in menadione toxicity. The 14CoS/14CoS cells exhibit 10-fold higher NAD(P)H:menadione oxidoreductase (NMO1) activity and 3-fold greater concentrations of reduced glutathione (GSH) than the ch/ch cells. In 14CoS/14CoS cells there are also 50% to 3-fold increases in glutathione transferase (GSTA1), UDP glucuronosyltransferase, and the copper, zinc-dependent superoxide dismutase activities. Catalase activity, on the other hand, is six times lower in the 14CoS/14CoS than the ch/ch line. The 14CoS/14CoS cells are two to four times more resistant to menadione killing than ch/ch cells. At concentrations of dicumarol that completely block NMO1 and GSTA1 activities, the 14CoS/14CoS cells show more than twice as much resistance to menadione toxicity than the ch/ch cells. Although superoxide formation is three times higher in untreated 14CoS/14CoS than ch/ch cells, menadione-induced superoxide formation is greater in the dying ch/ch than in the 14CoS/14CoS cells. Cellular resistance to menadione toxicity is correlated with intracellular GSH levels, rather than with the percentage of oxidized glutathione; cytotoxicity is not observed as long as GSH concentrations are sufficiently high (about 5-8 nmol/mg protein). For menadione, the results are consistent with a dominant role of GSH depletion in mediating toxicity and support a protective role for NMO1 activity. This report demonstrates the usefulness of these cell lines as a model system to study mechanisms of oxidative chemically induced toxicity, as well as to understand how intracellular levels of GSH are regulated.

Lipoprotein oxidation and measurement of thiobarbituric acid reacting substances formation in a single microtiter plate: its use for evaluation of antioxidants

Transition metals catalyze free radical-mediated oxidation of lipids and lipoproteins. This process is currently studied because of its potential relevance to pathological processes like atherosclerosis. Formation of thiobarbituric acid-reacting substances from polyenoic fatty acids is frequently used to follow oxidation of lipids and plasma lipoproteins. We describe here how Cu(II)- and Fe(III)-catalyzed oxidation of human low density lipoprotein or soy bean phospholipids and the photometric evaluation of the thiobarbituric acid-reaching substances formed can be conducted in the same 96-well microtiter plate. The procedure showed a correlation of 0.98 with conventional two-stage fluorimetric and spectrophotometric methods and also showed better reproducibility. The plate method can handle up to one plate per hour with considerably less labor than the test tube assays. The plate procedure required small volumes of diluted samples of lipoproteins lipids and reagents. The method was suitable for testing the concentration-dependent antioxidant potency of substances like probucol, butylated hydroxytoluene, and alpha-tocopherol. The method can also be used to follow the kinetics of oxidation of lipoproteins.

Evaluation of iron binding and peroxide-mediated toxicity in rat hepatocytes

A novel assay was developed to determine subnanomolar amounts of Fenton-reactive iron (FRI) in biological tissues. FRI represents that pool of iron that is redox active and capable of participating in a model Fenton reaction. The FRI was used to identify a kinetically-defined cellular iron binding site. This site displays positive cooperativity, with apparent kinetic constants of Kd = 10.6 microM, Bmax = 20.7 nmol/mg protein, and the Hill coefficient = 1.4. After addition of exogenous ferrous ammonium sulfate to hepatocytes, binding occurred within a few seconds and was stable for at least an hour. Free extracellular iron, but not bound iron, stimulated lipid peroxidation in hepatocytes. In contrast, bound but not free iron produced a concentration-dependent increase in tert-butyl hydroperoxide (TBH)-mediated toxicity, suggesting the toxicological relevance of bound, rather than free iron. Furthermore, the hydroxyl radical scavengers mannitol and 2-deoxyribose inhibited Fe2/TBH-mediated lipid peroxidation, but not cell killing, suggesting that hydroxyl radical may not be involved in the critical toxic event. The divalent cations Mn2+ and Co2+ inhibited iron-mediated hepatocyte killing in the presence of TBH, but only if added prior to Fe2+. Mn2+, but not Co2+, inhibited Fe(2+)-mediated lipid peroxidation regardless of the order of addition. These results indicate the existence of a specific, kinetically-defined cellular iron binding site. Such binding is involved in peroxide-mediated toxicity, but independent of lipid peroxidation. The specific nature of this site and involvement with other forms of chemical intoxication or cellular iron homeostasis are unknown.

Menadione-mediated membrane fluidity alterations and oxidative damage in rat hepatocytes

Menadione toxicity in isolated rat hepatocytes was mitigated by the antioxidant 4b,5,9b,10-tetrahydroindeno[1,2-b]indole at low concentrations (less than 100 microM), but not at high concentrations (greater than 200 microM) of menadione. When hepatocytes were incubated with menadione, there was a time-dependent and concentration-dependent inhibition of lipid peroxidation in intact cells, as well as an increase in the antioxidative potency of acetone extracts, suggesting that metabolites of menadione could inhibit oxidative stress, and that at high menadione concentrations a different mechanism was involved in cytotoxicity. A possible mechanism was suggested by the ability of acetone extracts from hepatocytes that had been incubated with menadione to increase osmotic fragility in red blood cells. This increase correlated with an increase in membrane fluidity in red blood cells, determined by flourescence polarization using the membrane probe 1,6-diphenyl-1,3,5-hexatriene. At 200 microM menadione, an increase in membrane fluidity was also observed in hepatocytes. The thiol dithiothreitol protected hepatocytes from 50 microM menadione toxicity, but not from greater than or equal to 100 microM menadione. The results suggest that while oxidative stress and arylation may be the critical mechanisms of toxicity at low menadione concentrations, at higher concentrations another mechanism such as enhanced membrane fluidity is operative.

"Oxidative stress" response in liver of an untreated newborn mouse having a 1.2-centimorgan deletion on chromosome 7

The c14CoS/c14CoS mouse has a homozygous deletion of about 1.2 cM on chromosome 7 that includes the albino (c) locus. The untreated 14CoS/14CoS newborn has been reported to exhibit a marked transcriptional activation of the hepatic NAD(P)H:menadione oxidoreductase (Nmo-1; DT diaphorase; quinone reductase; azo dye reductase) gene, as well as elevated UDP glucuronosyl-transferase (UGT1*06) and glutathione transferase (GT1) activities, when compared with the cch/cch wild-type and the cch/c14CoS heterozygote. We show here that the newborn hepatic activities of seven enzymes that play a role in the oxidative stress response--NMO1, UGT1*06, GT1, copper-zinc superoxide dismutase, glutathione peroxidase, glutathione reductase, and glucose-6-phosphate dehydrogenase--are increased 1.5- to 25-fold in 14CoS/14CoS, as compared with ch/ch and ch/14CoS mice. The activities of four additional enzymes having no known association with the oxidative stress response--benzo[a]pyrene hydroxylase (CYP1A1, cytochrome P(1)450), acetanilide 4-hydroxylase (CYP1A2, cytochrome P(3)450), lactate dehydrogenase (LDH), and NADPH-cytochrome c reductase--are not significantly different among the three genotypes. These data suggest that there exists an "oxidative stress" response in the untreated 14CoS/14CoS newborn. We postulate that a chromosome 7 regulatory gene, which we have named Nmo-1n, might encode a trans-acting negative effector of the Nmo-1 gene, and genes corresponding to the other elevated enzymic activities described above. When both copies of Nmo-1n are deleted, as is the case in 14CoS/14CoS mice, a battery of genes involved in oxidative stress is released from negative control and becomes activated--despite the absence of any apparent oxidative insult by foreign chemicals.

Relationship of membrane fluidity, chemoprotection, and the intrinsic toxicity of butylated hydroxytoluene

In isolated rat hepatocytes, many chemicals elicit toxicity which is inhibitable by antioxidants such as butylated hydroxytoluene (BHT). Although BHT protection is evident at concentrations of less than about 50 nmol/mg protein, higher concentrations exhibit intrinsic concentration-dependent toxicity, which involves mitochondrial dysfunction. We evaluated the possibility that both chemoprotection and intrinsic toxicity could be explained by a common mechanism involving alterations in the physical properties of cellular membranes. In the red blood cell (RBC) osmotic fragility assay, BHT at less than 60 nmol/mg protein protected against osmotic fragility; however, BHT at higher concentrations enhanced osmotic fragility such that total osmolysis occurred at 135 nmol/mg. The BHT-mediated alterations in osmotic fragility correlated with changes in membrane fluidity, determined by fluorescence polarization of the hydrophobic probe 1,6-diphenyl-1,3,5-hexatriene. Protection from osmolysis correlated with decreased fluidity, while enhanced RBC fragility correlated with increased fluidity. In rat hepatocyte suspensions, high BHT concentrations also permeabilized the plasma and mitochondrial membranes to enzyme leakage, and these effects were accompanied by enhanced membrane fluidity. Although other mechanisms may be operative, alterations in membrane fluidity appear to be, in part, responsible for the observed chemoprotective effects at low concentrations, and intrinsic toxicity at higher concentrations of BHT.

Chemoprotective and hepatic enzyme induction properties of indole and indenoindole antioxidants in rats

Three indole antioxidants were compared for their efficacy to inhibit lipid peroxidation, prevent chemical hepatotoxicity and induce enzyme systems involved in the biotransformation of xenobiotics. The dietary indolyl compound indole-3-carbinol (I-3-C), and the synthetic compounds 5,10-dihydroindeno[1,2-b]-indole (DHII) and 4b,5,9b,10-tetrahydroindeno[1,2-b]indole (THII) inhibited carbon tetrachloride (CCl4)-initiated lipid peroxidation in rat-liver microsomes, with the order of efficacy THII greater than DHII = butylated hydroxytoluene (BHT) much greater than I-3-C. Each of the indole compounds protected isolated rat hepatocytes against toxicity by CCl4, N-methyl-N'-nitro-N-nitrosoguanidine and methylmethanesulphonate (THII congruent to DHII much greater than I-3-C). In vivo administration of the indole compounds 1 hr before treatment with CCl4 protected against hepatotoxicity (THII greater than DHII greater than I-3-C). For the enzyme induction studies, phenobarbital and beta-naphthoflavone were used as standards, with corn-oil vehicle controls. The compounds were administered by gavage at 50 mg/kg body weight/day for 10 days. I-3-C produced increases in levels of hepatic cytochromes P-450 and ethoxyresorufin O-deethylase (EROD) activity, as well as in UDP-glucuronosyl transferase (UDPGT), glutathione S-transferase (GST), glutathione reductase (GSSG-Red) and quinone reductase. I-3-C produced decreased glutathione peroxidase (GSH-Px) and superoxide dismutase (SOD) activities. DHII produced increases in EROD, UDPGT, GST, GSSG-Red and quinone reductase, with decreases in NDMA-demethylase and GSH-Px activities. The only observed effect of THII was a modest induction of EROD activity. After treatment with the indole compounds for 10 days, I-3-C enhanced, while DHII diminished, CCl4-mediated 24-hr hepatotoxicity in rats. We conclude that DHII and THII are suitable candidates to develop further as potential chemoprotective and therapeutic agents for use in humans to treat disorders involving free radicals. THII has the greater radical scavenging efficacy, whereas DHII has the greater capacity to induce many different antioxidative enzymes.

Intrinsic acute toxicity and hepatic enzyme inducing properties of the chemoprotectants indole-3-carbinol and 5,10-dihydroindeno[1,2-b]indole in mice

Indole-3-carbinol (I-3-C) and 5,10-dihydroindeno[1,2-b]indole (DHII) have been shown to be protective against carbon tetrachloride and other chemicals that cause hepatic toxicity. In part, this protection appears to be afforded by the ability of these compounds to act as antioxidants, with DHII having much the greater efficacy. In order to understand the mechanisms of chemoprotection, as well as the potential for therapeutic and pharmaceutical use in humans, the antioxidants I-3-C and DHII were examined for their intrinsic acute toxicity, and their hepatic enzyme inducing properties in mice. The results were compared with those of the well characterized agent phenobarbital. Following treatment by gavage for 10 days with 50 mg compound/kg body weight, I-3-C produced modest (10-50%) increases in hepatic cytochrome P-450, aminopyrine N-demethylase, UDP-glucuronosyl transferase (UDPGT) and glutathione S-transferase (GST), and a four-fold increase in NAD(P)H: (quinone acceptor) oxidoreductase (quinone reductase) activity. DHII did not alter oxidative enzyme activities, but increased GST and UDPGT by about 50%, and quinone reductase over five-fold. In the acute toxicity studies, DHII produced no observable 24-hr acute toxicity up to 4 g/kg body weight, except for a slight decrease in haematocrit. However, I-3-C exhibited a dose-dependent toxicity above 100 mg/kg body weight, including a decrease in hepatic reduced glutathione after 2 hr and severe neurological toxicity, and the release of liver enzymes to the plasma at 24 hr. We conclude, on the basis of the superior antioxidation efficacy of DHII, its enzyme-inducing properties, and intrinsic toxicity, that DHII or cogeners thereof have great potential as chemoprotective or therapeutic agents. However, I-3-C does not have such potential.

Mechanisms of chemical mediated cytotoxicity and chemoprotection in isolated rat hepatocytes

Although N-methyl-N'-nitro-N-nitrosoguanidine (MNNG) and methylmethanesulfonate (MMS) cause injury and malondialdehyde formation in rat hepatocytes, MNNG toxicity is much more sensitive to inhibition by antioxidants. In order to quantify the relationship between toxicity and antioxidation potential, we compared 14 antioxidants that protected against MNNG and MMS toxicity. Chemoprotection was quantified as the concentration that delayed by 1 h the decline in trypan blue exclusion to less than or equal to 50%. While chemoprotection against MNNG and antioxidant efficacy were directly related (R = 0.86), chemoprotection against MMS and antioxidant efficacy were unrelated (R = 0.37). Since we hypothesized that protection against MMS involved stabilization of membranes, the capacity of the 14 compounds to stabilize membranes in an unrelated system (i.e. prevention of erythrocyte osmotic rupture) was assayed. Chemoprotection against both MNNG and MMS correlated with reduced RBC fragility (R = 0.97 and 0.70, respectively). One of the better protecting compounds, 4b,5,9b,10-tetrahydroindeno[1,2-b]indole, was also protective against hepatocellular toxicity mediated by acetaminophen, carbon tetrachloride and tert-butyl hydroperoxide, suggesting a fundamental basis in the mechanism of chemoprotection. We propose that methylating agents and perhaps other chemical toxicants destabilize cellular membranes resulting in hepatocellular injury. For MNNG, radical mediated events may result in membrane destabilization; for MMS, membranes are destabilized without concurrent radical events. The current studies provide a basis for future work to determine structure-activity relationships of chemoprotective agents, examine protection mechanisms, and develop better protective compounds.

Importance of protein thiols during N-methyl-N'-nitro-N-nitrosoguanidine toxicity in primary rat hepatocytes

N-Methyl-N'-nitro-N-nitrosoguanidine (MNNG), a potent toxicant in isolated rat hepatocytes, was evaluated for its mechanism of cytotoxicity. This direct acting toxicant generates an alkylating carbonium ion that covalently binds to cell macromolecules, depletes nonprotein thiols (NPT), and subsequently kills cells. In this study MNNG depleted protein thiols (PT) in a two-phase process. The first phase (about 30% depletion) occurred rapidly, in parallel with the depletion of NPT. After a plateau, a second phase of PT depletion occurred 5-8 min prior to cell death. Indole-3-carbinol (I-3-C), added prior to MNNG, did not alter the depletion of NPT nor the first phase of PT depletion. However, cell killing was substantially retarded and was still immediately preceded by the second phase of PT depletion. The addition of o-phenanthroline or 5,10-dihydroindeno[1,2-b]indole (DHII) prior to MNNG did not alter the first phase of PT depletion, but partially protected (about 30%) against the depletion of NPT. However, o-phenanthroline or DHII completely protected against the MNNG-induced loss of cell viability and the second-phase depletion of PT. When DHII was added after MNNG and prior to the expected second phase of PT depletion, that depletion was markedly depressed, as was the subsequent loss of cell viability. We conclude that MNNG covalent binding, depletion of NPT, and first-phase depletion of PT may be necessary, but insufficient to kill cells. We propose that rapid depletion of cellular antioxidants predisposes the cell to oxidative stress and that oxygen toxicity is responsible for the second-phase depletion of PT and the final cytotoxic events. The fact that the second-phase depletion of PT is required for and immediately precedes cell death suggests the importance of critical but as yet unidentified target thiol proteins in MNNG hepatotoxicity.

Toxicity of methylating agents in isolated hepatocytes

To investigate the pathogenesis of hepatotoxicity by methylating agents, we exposed isolated hepatocytes to N-nitrosodimethylamine (NDMA), N-methyl-N'-nitro-N nitrosoguanidine (MNNG), N-methyl-N-nitrosourea (MNU), or methyl methanesulfonate (MMS). Although NDMA is a potent in vivo hepatotoxicant in rats, no evidence of hepatocyte injury, measured by the leakage of lactate dehydrogenase (LDH) activity into the medium, was observed following exposure to a 1-100 mM concentration of either NDMA or MNU. In contrast, exposure of hepatocytes to MMS or MNNG resulted in greater than or equal to 90% LDH release. These differences in toxicity were not related to the extent of covalent binding to hepatocytes. Following MMS or MNNG, but not MNU or NDMA exposure, a significant rise in the generation of thiobarbiturate (TBA)-reactive species was observed. When hepatocytes were exposed to the antioxidant promethazine prior to the addition of MMS or MNNG, the formation of TBA-reactive species was inhibited completely. Although promethazine blocked MNNG-mediated cell injury, the antioxidant had no effect on MMS intoxication. These data suggest that methylating agents can cause hepatotoxicity by more than a single mechanism. For MNNG, lipid peroxidation may be involved in the pathogenesis of acute hepatotoxicity.

Protection against carbon tetrachloride hepatotoxicity by 5,10-dihydroindeno[1,2-b]indole, a potent inhibitor of lipid peroxidation

The influence of 5,10-dihydroindeno[1,2-b]indole (indenoindole) on carbon tetrachloride (CCl4)-mediated hepatotoxicity and lipid peroxidation were examined. Indenoindole (25 mg/kg body weight) ameliorated the increase in liver enzymes appearing in the plasma 24 hr after CCl4 administration, with about a 63% reduction for alanine transaminase, 56% for ornithine transcarbamylase and 84% for alkaline phosphatase. Indenoindole also partially prevented, in a dose-dependent fashion, the decrease in hepatic cytochromes P-450, total tissue reducing equivalents and hepatic ascorbate levels resulting 4 hr after CCl4 administration. In a homogeneous chemical system consisting of purified soybean phospholipid substrate in chlorobenzene, azobisisobutyronitrile-initiated lipid peroxidation was inhibited by indeno-indole, with 50% inhibition occurring at about 17 microM. Inhibition by indenoindole of iron-ascorbate-initiated lipid peroxidation in aqueous buffer containing phospholipid vesicles was about tenfold more efficient, with 50% inhibition occurring at about 1.5 microM. Presumably, this was due to the increased concentration of indenoindole in the membrane of the phospholipid vesicle. The efficiency of inhibition of lipid peroxidation was in the order of indenoindole = butylated hydroxytoluene (BHT) greater than alpha-tocopherol much greater than indole greater than indene. These 50% inhibition values of lipid peroxidation for these compounds were similar in an assay system composed of NADPH-fortified mouse-liver microsomes initiated with CCl4. For indenoindole, the 50% inhibition value (1.3 microM) was more than two orders of magnitude less than the spectral binding constant for indenoindole to mouse-liver cytochrome P-450 (Kd = 236 microM), implying that the partial inhibition of metabolic activation of CCl4 was not responsible for the inhibition of lipid peroxidation observed with indenoindole in this system. It appears that indenoindole may trap reactive radicals and inhibit lipid peroxidation in vitro. Regardless of whether inhibition is at the level of scavenging CCl4 metabolite radicals, or lipid radicals in membranes, radical trapping provides a plausible mechanism by which this compound inhibited CCl4 hepatotoxicity.