Oxidative stress in a clonal cell line of neuronal origin: effects of antioxidant enzyme modulation.

The effects of intracellularly generated H2O2 on cell viability, morphology, and biochemical markers of injury have been investigated in a clonal cell line of neuronal origin (140-3, mouse neuroblastoma X rat glioma) as a cell culture model for the role of oxidative stress in the long-term loss of neurons in the brain. The H2O2 was generated from the redox cycling of menadione, or by the oxidation of serotonin catalyzed by monoamine oxidase, to simulate the effect of amine neurotransmitter turnover. Incubation with menadione at concentrations as low as 10 microM for several hours resulted in significant losses of cell viability and altered morphology. Similar effects were evident in the presence of serotonin only after incubation overnight with concentrations > 1 mM. The cytotoxicity of either agent was potentiated by preincubation with specific inhibitors of two enzymes important to cellular antioxidant defenses, 3-amino-1,2,4-triazole for catalase and 1,3-bis(chloromethyl)-1-nitrosourea for glutathione reductase. Activity of another antioxidant enzyme of particular importance to antioxidant defenses in brain, the selenoprotein glutathione peroxidase, was stimulated fourfold by growth of cultures in the presence of sodium selenite as a source of active-site Se for the enzyme. The only effect of the selenite on other functionally coupled antioxidant enzymes was a decrease in activity of superoxide dismutase at concentrations > 200 nM. The selenite substantially protected cells against oxidative stress induced by combinations of menadione, 3-amino-1,2,4-triazole, and 1,3-bis(chloromethyl)-1-nitrosourea, but was only marginally effective with serotonin as a source of oxidative stress. The monoamine oxidase inhibitor pargyline increased cell survival in the presence of serotonin, demonstrating the role of this enzyme in its cytotoxicity. DNA damage (single strand breaks), but not lipid peroxidation, correlated with the cytotoxic effects of menadione.

A comparison of the effects of dietary selenium on selenoprotein expression in rat brain and liver.

In studies with rodents, when dietary supplies of the essential nutrient Se are restricted, in most tissues there are parallel substantial losses of the element and the important antioxidant selenoenzyme glutathione peroxidase (GPx) for which it is a cofactor. In brain, however, there appears to be both a sequestration of Se and a conservation of GPx activity when dietary Se is limited. To further explore the relation between these phenomena, we have undertaken a comparison of the effects of diets low, normal and high in Se on GPx activity, and labeling of selenoproteins following short-term (72 h) in vivo exposure to 75Se, in subcellular fractions from rat brain and liver, the latter serving as a representative tissue which does not retain Se and is depleted of most GPx activity following dietary restriction. Brains and livers from animals on the three diets showed different patterns of response with respect to both GPx activity and retention of the 75Se dose. The low-Se diet (0.006 ppm) substantially reduced GPx activity in liver but not brain, while high levels (1 ppm) did not increase GPx in either tissue relative to a normal (0.1 ppm) intake. The 75Se was retained in brain homogenates and subcellular fractions to the greatest extent by rats on the restricted diet, while in liver, retention was greater in rats fed the normal supplement than in animals on either the low- or high-Se diets. Levels of non-protein-bound 75Se were higher in brain than liver and increased with dietary Se in both tissues. When proteins in brain and liver homogenates and subcellular fractions where separated by one-dimensional SDS-PAGE and exposed to X-ray film, the resulting autoradiograms revealed the existence of seven distinct selenoprotein bands in brain and eight in liver. Different patterns of selenoprotein expression were observed in subcellular fractions isolated from both tissues. Dependence of levels of individual selenoproteins on diet paralleled the effects on 75Se retention. Dietary influences on expression of protein bands tentatively identified as GPx were more pronounced in liver than brain. All of these observations provide further evidence of the unique nature of Se metabolism in brain.

Effects of low selenium diets on antioxidant status and MPTP toxicity in mice.

To investigate the role of chronic oxidative stress in MPTP neurotoxicity, C57BL mice were maintained 6-8 weeks on diets deficient in nutrients essential to cellular antioxidant defenses, selenium (Se) and alpha-tocopherol (vit E), and the effects on tissue antioxidant status and MPTP toxicity were evaluated relative to controls on supplemented diets. Activities of the major antioxidant enzymes, glutathione peroxidase (GPx), catalase, and superoxide dismutase, and levels of malondialdehyde as a marker for oxidative stress, were measured in brain, lung, liver and blood. Caudate depletion of dopamine and its metabolites served as a measure of MPTP neurotoxicity. For mice on the Se deficient diet, levels of the selenoenzyme GPx decreased from 50% in brain to 90% in blood. No compensatory changes in the activities of the other antioxidant enzymes were observed and addition of vit E to the diet did not alter antioxidant enzyme activities or malondialdehyde levels. In animals not treated with MPTP, the Se deficient diet significantly increased malondialdehyde only in liver. No protective effect of the antioxidant supplements against caudate depletion of dopamine and its metabolites were observed. However, malondialdehyde levels were increased in the brains of MPTP treated mice on the low Se diets, suggesting the possibility of secondary oxidative damage to tissues accompanying the destruction of substantia nigra neurons by MPTP.

Toxicity of MPTP and structural analogs in clonal cell lines of neuronal origin expressing B type monoamine oxidase activity.

The toxicity of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP), its oxidized metabolite, and two recently synthesized 2'-alkyl derivatives of MPTP (methyl and ethyl), found to be more toxic in vivo in mice, have been compared in two neuroblastoma hybrid cell lines (NCB-20 and 140-3) that express the B form of monoamine oxidase (MAO), as tissue culture models for the mode of action of MPTP in the central nervous system. Unlike previously reported studies with cultured cells of neuronal origin expressing only MAO A, both of these cell lines were sensitive to MPTP. Consistent with the in vivo findings, the 2'-alkyl derivatives were much more toxic than MPTP and comparable to the oxidized metabolite MPP+ in their effects on cell survival and morphology. The cells could be protected against the reduced toxins, but not MPP+, by either the MAO A selective inhibitor, clorgyline or the MAO B selective inhibitor, deprenyl. The effectiveness of the MAO inhibitors in blocking the action of the reduced toxins was consistent with their ability to inhibit MAO activity in the cell cultures, but did not reflect MAO-substrate specificity of the toxins. Inhibitors of serotonin and dopamine uptake, which have been found to protect against MPTP toxicity in vivo, were generally ineffective in the cell cultures, with the exception of a marginal increase in survival of MPP(+)-treated 140-3 cells in the presence of the serotonin uptake inhibitor fluoxetine. These findings are discussed in relation to proposed in vivo mechanisms of MPTP cytotoxicity.

Platelet glutathione peroxidase and monoamine oxidase activity in schizophrenics with CT scan abnormalities: relation to psychosocial variables.

We have previously reported that the activity in platelets of the important antioxidant enzyme glutathione peroxidase (GPx) is inversely correlated with computed tomographic (CT) measures of brain atrophy in a population of patients with chronic schizophrenia, suggesting that low GPx may be a vulnerability factor in those schizophrenic patients with structural brain abnormalities. The significance of this finding has now been explored in a larger clinical population by examining the relation of GPx and CT parameters to psychosocial variables and to the activity of platelet monoamine oxidase (MAO), which has also been reported to be altered in certain schizophrenic populations. In the present study, low platelet GPx and high brain atrophy were found to be associated with DSM-III diagnoses of nonparanoid schizophrenia, a high degree of chronicity, and a predominance of negative symptoms. Contrary to some literature reports, atrophy also correlated with age and length of illness among the schizophrenic patients, although the contribution of these factors was less than that of low GPx, which was itself not age dependent. The ventricle-brain ratio (VBR) and atrophy were highly correlated in a control group of affective disorder patients, but not in the schizophrenic group, where large VBRs were found predominantly in the DSM-III undifferentiated subgroup. The low-GPx/high-atrophy schizophrenic patients had normal platelet MAO levels, and MAO was significantly lower only in the paranoid subgroup, consistent with reported observations. There was no evidence for a neuroleptic-induced effect on either enzyme.

Platelet monoamine oxidase B activity in women with premenstrual syndrome.

Several lines of evidence suggest a strong association between premenstrual syndrome and affective disorder. Similar psychological symptoms, behavioral manifestations, and biochemical etiologies have been reported. We attempted to evaluate the biologic interconnection between premenstrual syndrome and psychiatric disorder by investigating the platelet enzyme, monoamine oxidase B. The activity of this enzyme has been noted to be decreased in affective disorder, alcoholism, and psychiatric vulnerability. Platelet monoamine oxidase B activity, estradiol, and progesterone were assessed throughout one menstrual cycle in 13 women with premenstrual syndrome and 19 control subjects. No significant differences were noted between groups using these parameters. The study indicates that well-screened subjects with premenstrual syndrome are, as evidenced by the parameter of monoamine oxidase B, biochemically similar to normal control subjects.

Interaction of 1-methyl-4-phenylpyridinium ion with human platelets.

When uptake of the Parkinson's syndrome inducing neurotoxin MPTP (1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine) and its major brain metabolite MPP+ (1-methyl-4-phenylpyridinium ion) by human platelets were compared in platelet rich plasma, a much higher rate was observed for the metabolite. The uptake process was saturable (Km = 6.8 microM; Vmax = 0.064 nmole/min/mg protein) and could be blocked by inhibitors of serotonin uptake. The accumulation of MPP+ by the platelets was accompanied by a decrease in intracellular ATP and an inhibition of mitochondrial state 3 respiration. These findings are consistent with earlier reports of the effect of MPP+ on isolated mitochondria as a potential cytotoxic mechanism, but also demonstrate that the dopamine uptake system is not the only means by which this metabolite can be efficiently transported into cells.

Glutathione peroxidase and CT scan abnormalities in schizophrenia.

The search for morphological clues to the etiology of schizophrenia has led to widespread application of computed tomography (CT) scans in the examination of patients. These investigations have resulted in numerous reports over the past several years of brain atrophy and increased ventricle-brain ratios (VBR), suggestive of neuronal tissue damage, associated with the disorder. Altered activity of cellular antioxidant systems have been implicated in the neuronal cell loss that is associated with degenerative diseases of the central nervous system (CNS), but this phenomenon has not been investigated with respect to functional disorders like schizophrenia. A search for such a relationship in schizophrenics with evidence of brain atrophy has been initiated by measuring the activity of the important antioxidant enzyme glutathione peroxidase (GPx) in blood samples from a population of chronic schizophrenics and age- and sex-matched nonschizophrenic mental patients as controls. A strong negative correlation has been found between GPx activity in both isolated platelets and erythrocytes and CT scan measures of brain atrophy and VBR in the schizophrenics, but not in the control population, which exhibited comparable CT scan abnormalities. These observations suggest a unique relationship of GPx to the mechanism of tissue damage in the schizophrenics.

A comparison of platelet monoamine oxidase activity and phosphatidylserine content between chronic paranoid schizophrenics and normal controls.

Mitochondrial monoamine oxidase (MAO), type B, has been implicated in the etiology of schizophrenia. We have found the phospholipid, phosphatidylserine (PS) to be a highly specific inhibitor of MAO-B, which has led us to postulate that the PS-MAO interaction might offer a basis for the lower MAO levels observed in platelets from certain schizophrenic populations. In this study we compared platelet MAO activity with phospholipid composition in a group of normals and chronic paranoid schizophrenics. The phospholipids in platelets and erythrocytes were extracted and separated by high-performance liquid chromatography into major classes phosphatidylcholine (PC), phosphatidylethanolamine, lysophosphatidylethanolamine, phosphatidylinositol and PS. The paranoid subjects showed statistically significantly lower MAO activity as well as higher mean levels of PS and lower levels of PC in both platelets and erythrocytes, consistent with our hypothesis. The Ca2+-stimulated synthesis of serine-lipid in platelets was also monitored by incorporation of radioisotope into lipid extracts from 14C-labelled serine substrate, and no significant differences were found between subjects groups with respect to this parameter.

Phosphatidylserine inhibition of monoamine oxidase in platelets of schizophrenics.

Phosphatidylserine (PS) has recently been reported to be a specific inhibitor of B-type monoamine oxidase (MAO-B). The effect of added PS liposomes on platelet MAO-B activity was examined in two schizophrenic groups (paranoid and a mixture of residual/undifferentiated) and in normal controls. PS was a potent partial-mixed-uncompetitive inhibitor of the platelet enzyme, whereas other phospholipids tested were without effect. The PS concentration required for 50% inhibition was significantly higher for the paranoid relative to the residual/undifferentiated group and controls. This correlated with a lower mean basal MAO-B activity in this group. Hill plots, as a measure of the concentration dependence of PS sensitivity, however, revealed a similarity between the two schizophrenic groups in regard to inhibitor binding properties. Mean Hill coefficients for both groups were significantly different from the controls. The results were consistent with an in vivo role for PS as an allosteric regulator of platelet MAO-B.

Photoinactivation of B-type monoamine oxidase by a 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine metabolite.

The reaction of the neurotoxin MPTP (1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine) with monoamine oxidase from a variety of tissues including rat and monkey brain, bovine liver, and human placenta and platelets was found to yield, as a primary product, a reactive photosensitive substance with an absorbance maximum at 345 nm which is not the cation 1-methyl-4-phenylpyridinium ion previously reported as a monoamine oxidase-MPTP metabolite in vivo and in vitro. Our results suggest that the 1-methyl-4-phenyl-pyridinium ion is probably only generated in subsequent nonenzymatic transformations of this reactive monoamine oxidase metabolite. This substance was found to specifically inactivate the B-form of monoamine oxidase by a photo-induced mechanism and to react directly with NADPH and dopamine. Properties of the metabolite and potential significance of its reactions to MPTP neurotoxicity are discussed.

Proteases as probes of mitochondrial monoamine oxidase topography in situ.

Selective inactivation of the multiple forms of mitochondrial monoamine oxidase (MAO) by proteases in intact and hypotonically disrupted rat liver mitochondria has been used to examine the question of differential membrane orientations of the A and B enzymes. Proteases used as probes included trypsin, beta-chymotrypsin, and the extracellular protease of Staphylococcus aureus, chosen for their different amino acid specificities. With all three proteases, no changes in the relative rates of MAO-A and MAO-B inactivation were observed after disruption of the mitochondria. Trypsin and beta-chymotrypsin gave much faster rates of MAO-A inactivation in both intact and disrupted mitochondria. The selective effect of trypsin on MAO-A was also confirmed in human placental mitochondria, which possess only A-type activity. The effectiveness of hypotonicity in disrupting the outer membrane of the mitochondria was shown by rapid protease inactivation of an intermembrane space marker enzyme, adenylate kinase (EC 2.7.4.3). Contrary to some recent reports in the literature, these findings strongly suggest that the MAO-A and MAO-B multiple-form catalytic activities do not reside on opposite faces of the membrane.

Investigations of the mechanism of selective inhibition of type B mitochondrial monoamine oxidase by phosphatidylserine.

Liposomes of phosphatidylserine (PS) were found to inhibit strongly the B-form of membrane bound monoamine oxidase (MAO) isolated from rat and bovine liver, while having no effect on the rat liver A-form. Use of 14C-liposomes demonstrated high levels of PS association with the membrane, which could not be removed by extensive washing with high ionic strength buffers. The inhibition of MAO-B was not reversed on further perturbation of the membrane by chaotropic agents, sonication, or treatment with additional liposome preparations of phosphatidylcholine or phosphatidylinositol. Partial reversal of the inhibition was found when the PS-treated bovine liver membrane was solubilized with the detergent octyl glucoside. PS, however, had no effect on a solubilized preparation of bovine liver MAO. These results suggest a specific interaction between MAO and PS rather than an indirect effect of bulk changes in membrane properties, but an intact membrane was, nevertheless, required to mediate the inhibition. Comparison of the decreases in apparent levels of MAO-B in rat liver mitochondrial membranes that were calculated from changes in relative catalytic activities with A and B specific substrates or changes in sensitivity to A-form specific reversible and irreversible inhibitors, all showed good quantitative correlation. Lineweaver-Burk plots of the effect of PS incorporation into bovine liver mitochondrial membranes on MAO oxidation of phenylethylamine exhibited the expected pattern for a noncompetitive inhibitor acting on a ping-pong mechanism bireactant enzyme. On the basis of these results, a possible in vivo role for the acidic phospholipids in regulating apparent levels of MAO from one tissue to another and/or in response to environmental effects is proposed.

Selective effects on catalysis by the multiple forms of monoamine oxidase produced by interactions of acidic phospholipids with mitochondrial membranes.

The effect of acidic phospholipids on the A and B multiple forms of membrane-bound mitochondrial monoamine oxidase has been investigated by incubating liposomes with isolated rat liver mitochondrial outer membrane preparations at lipid:protein ratios of 0.01 to 1. A strong inhibition of monoamine oxidase B was observed with phosphatidylserine and a moderate activation of monoamine oxidase A with phosphatidylinositol, while cardiolipin had no significant effect on either form. The specificity of phosphatidylserine inhibition for monoamine oxidase B was also confirmed in mitochondrial outer membrane isolated from tissues containing exclusively the A or B form of the enzyme (human placenta and bovine liver). Levels of incorporation were comparable for all the phospholipids and tissues studied and could not account for the different effects observed. Inhibition of monoamine oxidase B was found to be similar in an intact mitochondria preparation to that observed in the isolated outer membrane. A recent report of activation of both monoamine oxidase forms in delipidated whole mitochondria by the acidic phospholipids was re-examined and found to involve release of monoamine oxidase from the mitochondria. The details of the effects of phosphatidylserine and phosphatidylinositol on membrane-bound monoamine oxidase are consistent with the concept of the multiple forms as two distinct peptides, and suggest a second possible mode of in vivo regulation of substrate specificity.