Reversible inhibition of adenylate cyclase activity of rat brain caudate nucleus by oxidized glutathione

Mixed results with mixed disulfides

A period of research with Helmut Sies in the 1980s is recalled. Our experiments aimed at an in-depth understanding of metabolic changes due to oxidative challenges under near-physiological conditions, i.e. perfused organs. A major focus were alterations of the glutathione and the NADPH/NADP(+) system by different kinds of oxidants, in particular formation of glutathione mixed disulfides with proteins. To analyze mixed disulfides, a test was adapted which is widely used until today. The observations in perfused rat livers let us believe that glutathione-6-phosphate dehydrogenase (G6PDH), i.a. might be activated by glutathionylation. Although we did not succeed to verify this hypothesis for the special case of G6PDH, the regulation of enzyme/protein activities by glutathionylation today is an accepted posttranslational mechanism in redox biology in general. Our early experimental approaches are discussed in the context of present knowledge.

Thioltransferases

A family of small molecular weight proteins with thiol-disulfide exchange activity have been discovered, widely distributed from E. coli to mammalian systems, called thioltransferases or glutaredoxins. There are no substantiated reports of thioltransferases-glutaredoxins in plants; however, partially purified dehydroascorbate reductase from peas had thiol-disulfide exchange catalytic activity using glutathione as reductant and S-sulfocysteine as thiosulfate cosubstrate (unpublished data). Thus, this class of proteins is universally distributed. Based on mutagenesis studies, a sequence of Cys-Pro-Tyr(Phe)-Cys- followed by Arg-Lys- or Lys alone is critical for both the thiol-disulfide exchange reaction and the dehydroascorbate reductase activity. The dithiol-disulfide loop represented by this structure is unique since the cystine closer to the N-terminus has a highly acidic thiol pKa (3.8 as determined for the pig liver enzyme) that contributes to the protein's high S- nucleophilicity. Compared with the microbial enzyme, the mammalian thioltransferases (glutaredoxins) are extended at both N and C termini by 10-12 amino acid residues, including a second pair of cysteines toward the C-terminus with no known special function. Yeast thioltransferase is more like mammalian enzymes in length (106 amino acids) but more like E. coli glutaredoxin in being unblocked at the N-terminus and having only one set of cysteines; that is, at the active center. The three mammalian enzymes, for which sequences are available, are blocked at the N-terminus by an acetyl group linked to alanine with no known special function other than possibly to impart greater cellular turnover stability. A report of carbohydrate (8.6%) content in rat liver thioltransferase has not been verified by more sensitive methods of carbohydrate analysis, nor has carbohydrate been identified in samples of purified glutaredoxin from any source. Thiol transferase and glutaredoxin are two names for the same protein based on similarity of amino acid sequence, immunochemical cross-reactivity, and other enzyme properties. The inability of thioltransferase from some mammalian sources to act as an electron carrier in ribonucleotide reductase systems, whether homologous or heterologous in origin, remains to be explained in future studies.

Inhibition of brain mitochondrial respiration by dopamine: involvement of H(2)O(2) and hydroxyl radicals but not glutathione-protein-mixed disulfides

Examination of the downstream mediators responsible for inhibition of mitochondrial respiration by dopamine (DA) was investigated. Consistent with findings reported by others, exposure of rat brain mitochondria to 0.5 mm DA for 15 min at 30 degrees C inhibited pyruvate/glutamate/malate-supported state-3 respiration by 20%. Inhibition was prevented in the presence of pargyline and clorgyline demonstrating that mitochondrial inhibition arose from products formed following MAO metabolism and could include hydrogen peroxide (H(2) O(2) ), hydroxyl radical, oxidized glutathione (GSSG) or glutathione-protein mixed disulfides (PrSSG). As with DA, direct incubation of intact mitochondria with H(2) O(2) (100 microm) significantly inhibited state-3 respiration. In contrast, incubation with GSSG (1 mm) had no effect on O(2) consumption. Exposure of mitochondria to 1 mm GSSG resulted in a 3.3-fold increase in PrSSG formation compared with 1.4- and 1.5-fold increases in the presence of 100 microm H(2) O(2) or 0.5 mm DA, respectively, suggesting a dissociation between PrSSG formation and effects on respiration. The lack of inhibition of respiration by GSSG could not be accounted for by inadequate delivery of GSSG into mitochondria as increases in PrSSG levels in both membrane-bound (2-fold) and intramatrix (3.5-fold) protein compartments were observed. Furthermore, GSSG was without effect on electron transport chain activities in freeze-thawed brain mitochondria or in pig heart electron transport particles (ETP). In contrast, H(2) O(2) showed differential effects on inhibition of respiration supported by different substrates with a sensitivity of succinate > pyruvate/malate > glutamate/malate. NADH oxidase and succinate oxidase activities in freeze-thawed mitochondria were inhibited with IC(50) approximately 2-3-fold higher than in intact mitochondria. ETPs, however, were relatively insensitive to H(2) O(2). Co-administration of desferrioxamine with H(2) O(2) had no effect on complex I-associated inhibition in intact mitochondria, but attenuated inhibition of rotenone-sensitive NADH oxidase activity by 70% in freeze-thawed mitochondria. The results show that DA-associated inhibition of respiration is dependent on MAO and that H(2) O(2) and its downstream hydroxyl radical rather than increased GSSG and subsequent PrSSG formation mediate the effects.

Glutathione alters the mode of calcium-mediated regulation of adenylyl cyclase in membranes from mouse brain

We examined the effect of sulfhydryl compounds on the regulation of adenylyl cyclase by calcium in mouse cerebrum membranes. Isoproterenol-stimulated adenylyl cyclase (IP-AC) activity in the membranes was increased by addition of the optimum concentrations of calcium/calmodulin. However, in the presence of 0.01-0.07 mM glutathione (GSH), calcium/calmodulin inhibited the activity. At high concentrations of GSH (1-10 mM), the IP-AC activity was stimulated by calcium/calmodulin to a greater extent than that in the control (no GSH). Cysteine at less than 1.7 mM induced a similar inhibition of the IP-AC activity, but dithiothreitol did not. The activity of IP-AC measured in the absence of calmodulin decreased when calcium levels were greater than 300 microM. GSH at 0.05 mM enhanced the calcium-dependent inhibition (22% inhibition by 200 microM calcium), while 10 mM GSH lowered it. Calmodulin itself had no significant effect on the IP-AC activity, irrespective of the concentrations of GSH involved. It caused a small increase in the IP-AC activity that had been reduced by the presence of calcium and GSH. These results indicate that the redox status of sulfhydryls in adenylyl cyclase plays an important role in the calcium-mediated regulation of the enzyme. The enzyme becomes much more sensitive to the calcium-dependent inhibition after partial reduction of the sulfhydryls via the particular mode of reactin.

Effect of oxidized glutathione on intestinal mitochondria and brush border membrane

Oxidative stress is associated with the formation of oxidized glutathione (GSSG) in the cells, which can form mixed disulfide with proteins leading to alteration of their function. The present study looks at the effect of in vitro exposure of GSSG on intestinal mitochondria and brush border membrane (BBM). Incubation with 1 mM GSSG increased the protein bound GSH in mitochondria by 15-fold. This was associated with loss of activity of certain mitochondrial enzymes such as succinic dehydrogenase, isocitrate dehydrogenase, total ATPase and NADH dehydrogenase whereas NADH oxidase was not affected. A similar treatment of BBMV with GSSG increased the protein bound GSH by 4.7-fold without altering its enzyme activity. Exposure to GSSG had no effect on the Na(+)-dependent glucose transport by BBMV. These studies suggest that GSSG formed during oxidative stress may modify thiol groups in proteins by forming mixed disulfides leading to functional alteration of certain cellular proteins.

Selective inhibition of adenylate cyclase in bovine cortex by quinones: a novel cellular substrate for quinone cytotoxicity

Quinones are widely distributed substances of often potential toxicological significance. On the other hand, cyclic AMP is known to promote a cell-survival response and to retard apoptosis [Berridge, Tan and Hilton (1993) Exp. Hematol. 21, 269-276]. Therefore the effects of quinones on adenylate cyclase were tested. Adenylate cyclase is rapidly inhibited by quinones, with IC50 values of 40-45 microM for p-benzoquinone (BQ) or 200 microM for dichlorophenol-indophenol (DCIP), with 2-substituted quinones being inactive. Membrane solubilization decreases the IC50 values for BQ and DCIP to 18 microM and 40 microM respectively. The inhibition is not affected by GTP, GDP or analogues, or by cholera and pertussis toxins; therefore it is not mediated by a G-protein or the activation of a defined receptor. Further, the inhibition stoichiometrically competes with forskolin activation of adenylate cyclase, equimolar concentrations of quinone and forskolin restoring the enzyme activity to its basal value. Reduction of BQ with sodium dithionite stoichiometrically prevents the inhibition of adenylate cyclase; in turn, oxidation of hydroquinone with ferricyanide fully restores it, indicating that the oxidized state of the quinone is required for inhibition. In addition, BQ is cytotoxic in vivo on HepG2, a human hepatocellular carcinoma cell line, but the effect can be prevented with forskolin. In plasma membranes, BQ tightly binds only one major and two minor proteins; these BQ-binding proteins were purified by means of labelling with [14C]BQ followed by PAGE under native conditions. Together these observations indicate that the action of quinone can be traced to targeting a limited number of proteins at the plasma membrane in a highly selective way and to affecting key enzymes such as adenylate cyclase.

Effect of oxidative stress on receptors and signal transmission

Reactive oxygen metabolites affect binding of ligands to membrane receptors and also coupling of receptors to G-proteins and effector enzymes. Peroxidation of membrane lipids may lead to a lowered receptor density and also will alter the viscosity of the plasma membrane, which affects receptor coupling. Reactive oxygen species may also interact with thiol/disulfide moieties on receptor proteins or on other factors in the receptor system, which is responsible for alterations in receptor binding or coupling. Moreover, lipid peroxidation is associated with the phospholipase A2 pathway, which might indirectly affect receptor function. Moreover, oxidative stress may lead to a disturbance in cellular Ca(2+)-homeostasis. This might be related to an effect on Ca(2+)-mobilizing receptors, but there is also evidence for a decreased Ca(2+)-sequestration by ATPases. In addition, peroxidation of membrane lipids increases membrane permeability to Ca2+. Finally, reactive oxygen species interfere with actions of nitric oxide, thus affecting another pharmacological messenger system.

Agonist activity of a novel compound, 1-[3-(3,4-methylenedioxyphenoxy)propyl]-4-phenyl piperazine (BP-554), at central 5-HT1A receptors

We used an in vitro radioligand receptor binding assay with rat cerebral cortex, hippocampus and striatum membrane preparations to show that 1-[3-(3,4-methylenedioxyphenoxy)propyl]-4-phenyl piperazine (BP-554) had much higher affinity for 5-HT1A recognition sites than for 5-HT1-non-A, 5-HT2, benzodiazepine, dopamine D-2 and alpha 2-adrenergic recognition sites. The compound inhibited the activity of forskolin-stimulated adenylate cyclase in rat hippocampal membranes. Intraperitoneal injection of BP-554 to mice decreased the concentration of only 5-hydroxy-indoleacetic acid of the amines and their metabolites in the brain and decreased the accumulation of 5-hydroxytryptophan in the brain after decarboxylase inhibition by 3-hydroxybenzylhydrazine. Furthermore, the administration of BP-554 caused hypothermia and increased serum corticosterone levels in mice. The observed effects of BP-554 were similar to those of 8-hydroxy-2-(di-n-propylamino)tetralin. These results suggest that BP-554 acts as a selective 5-HT1A receptor agonist in vivo.

Enzyme regulation by biological disulfides

More than a dozen enzymes have been found to be activated or inhibited in vitro by disulfide-exchange between the protein and small-molecule disulfides. Accordingly, thiol/disulfide ratio changes in vivo may be of great importance in the regulation of cellular metabolism. An awareness of this regulatory mechanism in both host cells and parasites, coupled with information on the presence or absence of key enzymes, may lead to rational drug design against certain diseases involving thiol intermediates, including trypanosomiasis.

A disbalance between beta-adrenergic and muscarinic responses caused by hydrogen peroxide in rat airways in vitro

The effect of hydrogen peroxide on adrenergic and muscarinic responses of rat airway smooth muscle was studied. The trachea muscle and the lung parenchymal strip were contracted with methacholine and relaxed with (-)-isoprenaline. Recording of three (-)-isoprenaline curves on the trachea muscle and the lung parenchymal strip was followed by treatment for 30 min with hydrogen peroxide (H2O2) (1mM) after which a new dose response curve for (-)-isoprenaline was constructed. Using the trachea muscle this treatment with H2O2 resulted in a decrease of 61% of the maximum contraction by methacholine compared with the control and a complete inhibition of the relaxation by (-)-isoprenaline. In the lung parenchymal strip preparation we found, after the same treatment no reduction of the contraction by methacholine and 61% reduction of the relaxation by (-)-isoprenaline, compared with the control. The results demonstrate that the adrenergic response in rat airways is more susceptible to hydrogen peroxide than the muscarinic response.

Formation and reduction of glutathione-protein mixed disulfides during oxidative stress. A study with isolated hepatocytes and menadione (2-methyl-1,4-naphthoquinone)

Incubation of isolated rat hepatocytes with menadione (2-methyl-1,4-naphthoquinone) resulted in a dose-dependent depletion of intracellular reduced glutathione (GSH), most of which was oxidized to glutathione disulfide (GSSG). Menadione metabolism was also associated with a dose- and time-dependent inhibition of glutathione reductase, impairing the regeneration of GSH from GSSG produced during menadione-induced oxidative stress. Inhibition of glutathione reductase by pretreatment of hepatocytes with 1,3-bis(2-chloroethyl)-1-nitrosourea (BCNU) greatly potentiated both GSH depletion and GSSG formation during the metabolism of low concentrations of menadione. Concomitant with GSH oxidation, mixed disulfides between glutathione and protein thiols were formed. The amount of mixed disulfides produced and the kinetics of their formation were dependent on both the intracellular GSH/GSSG ratio and the activity of glutathione reductase. The mixed disulfides were mainly recovered in the cytosolic fraction and, to a lesser extent, in the microsomal and mitochondrial fractions. The removal of glutathione from protein mixed disulfides formed in hepatocytes exposed to oxidative stress was dependent on GSH and/or cysteine and appeared to occur predominantly via a thiol-disulfide exchange mechanism. However, incubation of the microsomal fraction from menadione-treated hepatocytes with purified glutathione reductase in the presence of NADPH also resulted in the reduction of a significant portion of the glutathione-protein mixed disulfides present in this fraction. Our results suggest that the formation of glutathione-protein mixed disulfides occurs as a result of increased GSSG formation and inhibition of glutathione reductase activity during menadione metabolism in hepatocytes.

Glutathione cycle activity and pyridine nucleotide levels in oxidant-induced injury of cells

Exposure of target cells to a bolus of H2O2 induced cell lysis after a latent period of several hours, which was prevented only when the H2O2 was removed within the first 30 min of injury by addition of catalase. This indicated that early metabolic events take place that are important in the fate of the cell exposed to oxidants. In this study, we described two early and independent events of H2O2-induced injury in P388D1 macrophagelike tumor cells: activation of the glutathione cycle and depletion of cellular NAD. Glutathione cycle and hexose monophosphate shunt (HMPS) were activated within seconds after the addition of H2O2. High HMPS activity maintained glutathione that was largely reduced. However, when HMPS activity was inhibited--by glucose depletion or by incubation at 4 degrees C--glutathione remained in the oxidized state. Total pyridine nucleotide levels were diminished when cells were exposed to H2O2, and the breakdown product, nicotinamide, was recovered in the extracellular medium. Intracellular NAD levels fell by 80% within 20 min of exposure of cells to H2O2. The loss of NADP(H) and stimulation of the HMPS could be prevented when the glutathione cycle was inhibited by either blocking glutathione synthesis with buthionine sulfoximine (BSO) or by inhibiting glutathione reductase with (1,3-bis) 2 chlorethyl-1-nitrosourea. The loss of NAD developed independently of glutathione cycle and HMPS activity, as it also occurred in BSO-treated cells.

Increased dihydrofolate reductase activity in methotrexate-resistant human promyelocytic-leukaemia (HL-60) cells. Lack of correlation between increased activity and overproduction

Methotrexate(MTX)-resistant human promyelocytic-leukaemia cells (HL-60) derived from MTX-sensitive cells have a 20-fold increase in dihydrofolate reductase (DHFR) activity as compared with the sensitive cells. This increase is not associated with a concomitant increase in DHFR protein as determined by sodium dodecyl sulphate/polyacrylamide-gel electrophoresis and by immunological methods using mouse anti-DHFR antibody. The rate of DHFR synthesis is similar in both cell lines. Furthermore, both the sensitive and resistant cells have similar amounts of RNA hybridizing to a DHFR complementary-DNA probe, correlating well with the lack of increase in DHFR protein. DHFR-gene dosages were similar in both types of cells. We conclude that the 20-fold increase in DHFR activity present in these MTX-resistant cells is not due to the overproduction of DHFR but due to the expression of a more active form of the enzyme.

Modulation by unsaturated fatty acids of norepinephrine- and adenosine-induced formation of cyclic AMP in brain slices

The effect of linoleic acid on the formation of cyclic AMP in the slices of guinea pig cerebral cortex was examined. Treatment of the slices with linoleic acid resulted in an increase of basal and of norepinephrine-stimulated formation of cyclic AMP. The stimulatory effect on the basal level of cyclic AMP was not specific for linoleic acid: the potency of the fatty acid was related to the magnitude of unsaturation. In contrast, the enhancement of norepinephrine-stimulated formation of cyclic AMP seemed relatively specific for linoleic acid and arachidonic acid. Linoleic acid markedly enhanced the stimulated formation of cyclic AMP by histamine and adenosine, as well that by norepinephrine, without affecting that by excitatory amino acids and veratridine. Theophylline, adenosine deaminase, and 2'-deoxyadenosine antagonized the effect of linoleic acid. Linoleic acid enhanced the maximum responses to norepinephrine and adenosine without altering the ED50 values for these agonists. When linoleic acid-treated slices were washed with Krebs-Ringer containing defatted bovine serum albumin, both enhancement of the response to norepinephrine and the amount of [14C]linoleic acid incorporated in a free form significantly diminished.

Assay of adenylate cyclase in rat brain homogenates and lysed turkey erythrocytes: two different methods give different results with brain but not with erythrocytes

Experiments showed that the cyclic AMP (cAMP) fraction isolated by alumina or Dowex 50/alumina chromatography from brain adenylate cyclase reaction mixtures contained 32P radioactivity 10-12 times in excess of that which could be accounted for by determination of cAMP using binding assays. No such discrepancy was found when lysed turkey erythrocytes were assayed. This indicated that special precautions must be taken for the purification of 32PcAMP from brain adenylate cyclase assays due to the formation of 32P-labelled contaminants.

Effect of diamide and reduced glutathione on the elevated levels of cyclic AMP in rat pancreatic islets exposed to glucose, p-chloromercuribenzoate and aminophylline

In rat pancreatic islets the effects of diamide, which has been shown to decrease islet levels of reduced glutathione (GSH), and of exogenous GSH were investigated on cyclic AMP as increased by glucose, p-chloromercuribenzoate, and aminophylline. In addition the effect of diamide on islet ATP level, low Km and high Km phosphodiesterases was studied. Diamide (0.1 mM) inhibited the increase of cyclic AMP (cAMP) in response to glucose (16.7 mM), and p-chloro-mercuribenzoate (1 mM) in the presence of 5.6 mM glucose. No inhibitory effect of diamide could be demonstrated when cAMP was raised by 10 mM aminophylline in the presence of 5.6 mM glucose. The glucose (27.7 mM) stimulated increase of cAMP was further augmented by GSH (0.4 mM) whereas GSH in the presence of 5.6 mM glucose had no such effect. Diamide neither affected islet high Km nor low Km cAMP-phosphodiesterases. Diamide (0.1 mM) as used in this study did not affect islet AMP levels. A concentration dependent decrease of ATP was observed, however, with higher concentrations of diamide (0.25, 0.5 and 1.0 mM). It is suggested that the accumulation of islet cAMP in response to glucose and para-chloromercuribenzoate depends on the redox state of islet thiols. Since thiol oxidant diamide neither affected cAMP-phosphodiesterase activities nor inhibited aminophylline induced accumulation of cAMP in the presence of low glucose the possibility is raised that in pancreatic islets the formation of cAMP rather than its degradation depends on the redox state of islet thiols.

Potentiation of the insulin-releasing capacity of tolbutamide by thiols: studies on the isolated perfused pancreas

It has been suggested that the islet thiol redox status plays a role in the regulation of beta-cell sensitivity in response to insulin secretagogues. Employing the isolated perfused rat pancreas, the effect of reduced glutathione (1 mM) and L-cysteine (5 mM) on insulin release induced by tolbutamide (0.2 mg/ml), glucose (5.6 and 11.1 mM) and tolbutamide (0.1 mg/ml) in the presence of 5.6 mM glucose was studied. In the absence of glucose or in the presence of 5.6 mM of glucose neither glutathione nor L-cysteine stimulated the release of insulin. Reduced glutathione potentiated the secretion induced by glucose (11.1 mM) during the first and the second phase. L-Cysteine potentiated only the first phase of glucose-induced insulin release, whereas the second phase was depressed. Both of the tested thiols potentiated the insulin secretory action of either tolbutamide (0.2 mg/ml) alone or tolbutamide (01. mg/ml) in the presence of glucose (5.6 mM). The data suggest that supplementation of thiols to the pancreatic beta-cells perse cannot initiate the insulin secretory process. It is also suggested that GSH and L-cysteine increase the sensitivity of beta-cells to the stimulatory action of tolbutamide and/or glucose.

Regional distribution of glutathione reductase in the adult rat brain

Glutathione reductase was measured in 10 selected areas of the adult rat brain. Activitives varied over a 2.5-fold range. The caudate-putamen and cortical areas contained the highest activities. Several gray matter structures contained intermediate activities. The substantia nigra and the corpus callosum, a white matter structure, contained the lowest activities. The caudate-putamen, a catecholamine rich area, may produce H2O2 via MAO and would require glutathione reductase to recycle glutathione oxidized by glutathione peroxidase. However, the distribution of glutathione reductase activity also suggests a role in reducing oxidants formed during tissue respiration.

Role of cytoplasmic thioltransferase in cellular regulation by thiol-disulphide interchange

Cytoplasmic thioltransferase purified from rat liver [Axelsson, Eriksson & Mannervik (1978) Biochemistry 17, 2978--2984] catalyses the formation and decomposition of mixed disulphides of proteins and glutathione. The enzyme was found to catalyse the reversible thiol-disulphide interchange between glutathione disulphide and a crude thiol-containing protein fraction from rat liver. This finding indicates a role of the thioltransferase in the regulation of the 'glutathione status' of the cell. Specifically, it was found that thioltransferase catalyses the reactivation of pyruvate kinase from rat liver that had previously been inactivated by glutathione disulphide. It is suggested that thioltransferase may have a general role in regulatory processes involving thiol-disulphide interchange.