Reversible inactivation of guanylate cyclase by mixed disulfide formation

Highly purified preparations of guanylate cyclase from rat lung were inactivated by several disulfide compounds in a time- and dose-dependent manner. Cystamine and cystine were the most potent disulfides tested, but other compounds which contained the cysteamine moiety (NH2CH2CH2S-), including pantethine and oxidized coenzyme A, were also able to partially inactivate the enzyme. In addition to the decrease in basal activity (measured with either Mg2+-GTP or Mn2+-GTP), disulfide-inhibited enzyme was activated to a lesser extent by nitric oxide. Treatment with dithiothreitol or other reducing agents restored basal activity and increased the level of cGMP production following nitric oxide activation. Control enzyme samples exhibited a single GTP Km of 25 microM or 150 microM with Mn2+ or Mg2+, respectively. However, cystamine-treated enzyme showed these same Km values as well as an additional GTP Km of 2 to 3 microM using either metal ion as cofactor. When [35S]cystine was incubated with purified enzyme, radioactivity was incorporated into the trichloroacetic acid-precipitable protein, and the counts were released following dithiothreitol treatment. In addition, [35S]cystine-labeled enzyme co-migrated with native guanylate cyclase on nondenaturing polyacrylamide gels. These data indicate that mixed disulfides can be formed between guanylate cyclase and certain naturally occurring compounds, and that disulfide formation leads to a reversible loss of enzyme activity.     

Rapid and reversible activation of acetyl CoA hydrolase in intact pineal cells by disulfide exchange

Using intact pinealocytes in suspended cell culture it has been determined that acetyl CoA hydrolase activity can be rapidly increased by treatment with cystamine. Similar results are seen with diacetylcystamine, but not with GSSG, penicillamine disulfide, nor with oxidized DTT. The activation of acetyl CoA hydrolase by cystamine is reversible: after cystamine treatment is terminated, enzyme activity decreases slowly in cell culture. It is also possible to reverse the activation by treating homogenates of cystamine-treated cells with dithiothreitol. These observations are consistent with previous findings indicating that pineal acetyl CoA hydrolase activity can be regulated $\text{via}$ protein thiol: disulfide exchange. The observations presented in this report also indicate that conditions within the cell allow this type of reaction to take place, and raise the possibility that disulfide exchange mechanisms may be physiologically involved in the intracellular regulation of the activity of this and perhaps other enzymes.     

Activation of pineal and brain acetyl-COA hydrolase by cystamine: an apparent case of disulfide exchange.

Rat pineal acetyl-CoA hydrolase was activated about 5-fold by cystamine treatment (30 mM) at pH 6.8 and 10-fold at pH 8.5. Six other disulfides were found to be ineffective or to produce a small activation. Cystamine activation was not reversed when free cystamine was removed, but was reversed by treatment with DTT. Analysis of other tissues indicated acetyl-CoA hydrolase from rat brain, sheep pineal gland and chicken pineal gland could also be activated by cystamine. In contrast, cystamine activation of rat liver acetyl-CoA hydrolase was not seen.     

Study on human erythrocyte thioltransferase: comparative characterization with bovine enzyme and its physiological role under oxidative stress.

Thioltransferase, an enzyme which catalyzes the thiol/disulfide exchange reaction in the presence of GSH, was purified to homogeneity on 15% SDS-PAGE from human (36,000-fold purification) and bovine (23,000-fold) erythrocyte hemolysates. These enzymes had similar properties in their monomeric structures (M(r) = 11,000) and broad specificities for substrates ranging from low-molecular disulfides (S-sulfocysteine, cystamine, and cystine) to protein disulfides (trypsin and insulin). They were highly sensitive to SH-reagents (monoiodoacetic acid and mercuric chloride), but were protected from inactivation by the presence of disulfides (GSSG, cystamine, and cystine). Phosphofructokinase and pyruvate kinase that had been inactivated by disulfides were reactivated effectively by the addition of thioltransferase with GSH. In addition, disulfides in membrane proteins of human erythrocytes that have been oxidatively damaged by diamide treatment were reduced to the SH-free form more effectively by incubation with thioltransferase.     

Gamma-glutamylcysteine synthetase. Interactions of an essential sulfhydryl group

gamma-Glutamylcysteine synthetase (isolated from rat kidney) has one sulfhydryl group that reacts with 5,5'-dithiobis-(2-nitrobenzoate). This single exposed sulfhydryl group is not required for enzyme activity. The enzyme is potently inactivated by cystamine, which apparently interacts with a sulfhydryl group at the active site to form a mixed disulfide. 5,5'-Dithiobis-(2-nitrobenzoate) does not interact with the sulfhydryl group that reacts with cystamine. After the enzyme was 90% inactivated by reaction with cystamine, 3.4 mol of 5,5'-dithiobis-(2-nitrobenzoate) reacted per mol of enzyme, indicating that binding of cystamine exposes sulfhydryl groups which are apparently buried or unreactive in the native enzyme. L-Glutamate (but not D-glutamate or L-alpha-aminobutyrate) protected against inactivation by cystamine. In contrast, ATP enhanced the rate of inactivation by cystamine, and the apparent Km value for this effect is similar to that for ATP in the catalytic reaction. Studies on the structural features of cystamine that facilitate its interaction with the enzyme showed that selenocystamine, monodansylcystamine, and N-[2[2-aminoethyl)-dithio)ethyl]-4-azido-2-nitrobenzeneamine are also good inhibitors. Whereas S-(S-methyl)cysteamine-Sepharose does not interact with the enzyme (Seelig, G. F., and Meister, A. (1982) J. Biol. Chem. 257, 5092-5096), S-(S-methyl)cysteamine is a potent inhibitor; 1 mol of this compound completely inactivated 1 mol of enzyme. In the course of this work, a useful modification of the method for isolating this enzyme from kidney was developed.     

Inhibition of gamma-glutamylcysteine synthetase by cystamine: an approach to a therapy of 5-oxoprolinuria (pyroglutamic aciduria).

γ-Glutamylcysteine synthetase is strongly inhibited by cystamine; thus, 20 μM cystamine inhibited the activity by 50%. Inhibition is rapid and the inhibited enzyme is reactivated by dithiothreitol suggesting that cystamine reacts with an enzyme sulfhydryl group. Inhibition by cystamine is not prevented by MgATP, L-α-aminobutyrate, or L-glutamate suggesting that cystamine may not interact at the active site. Little or no inhibition was observed with N,N′-diacetyl cystamine, L-cystine, glutathione disulfide, 2-hydroxyethyl disulfide, and thioglycolate disulfide, whereas thiocholine disulfide produced moderate inhibition. Cystamine or an inhibitory analog of cystamine might be useful in the therapy of the disease 5-oxoprolinuria in which there is an overproduction of γ-glutamylcysteine.     

Cystamine induces toxicity in hepatocytes through the elevation of cytosolic Ca2+ and the stimulation of a nonlysosomal proteolytic system

Infusion of cystamine into the isolated, perfused rat liver resulted in tissue damage preceded by the formation of cystamine-protein mixed disulfides which were mainly detected in the plasma membrane fraction. Hepatotoxicity was prevented when dithiothreitol was infused after cystamine or when the calcium antagonist, verapamil, was co-infused with the disulfide. In isolated hepatocytes, the formation of cystamine-protein mixed disulfides was associated with an inhibition of plasma membrane Ca2+-ATPase activity and a decreased rate of Ca2+ efflux from the cells. This resulted in intracellular Ca2+ accumulation which was followed by a stimulation of both phospholipid hydrolysis and proteolysis, as indicated by enhanced rates of release of radioactivity from hepatocytes prelabeled with [14C]arachidonate and [14C]valine, respectively. Preincubation of hepatocytes with the calmodulin inhibitor, calmidazolium, or with the phospholipase inhibitors, chlorpromazine and dibucaine, inhibited the stimulation of [14C]arachidonate release by cystamine. However, none of these agents prevented the onset of cystamine toxicity in hepatocytes. In contrast, pretreatment of the cells with antipain or leupeptin, two inhibitors of Ca2+-activated proteases, abolished the stimulation of proteolysis by cystamine and also protected the cells from cystamine toxicity. Our results suggest that the perturbation of intracellular Ca2+ homeostasis by cystamine is caused by the inhibition of Ca2+ efflux associated with the formation of cystamine-protein mixed disulfides in the plasma membrane and that subsequent cytotoxicity results from Ca2+-activation of a nonlysosomal proteolytic system.     

Selective reduction of the disulfide bonds of ovine placental lactogen

Reduction and carbamidomethylation of two of the three disulfide bridges of ovine placental lactogen was accomplished by the use of 20-fold molar excess of dithiothreitol over protein disulfide content. The derivative retained its binding capacity to somatogenic as well as lactogenic rat liver receptors, although the latter was somewhat diminished. The two disulfide bonds exposed to the reducing agent are those located near the carboxy- and amino-terminus, while the larger loop remained intact after reduction. This behaviour is similar to that of bovine growth hormone, where the larger loop was also more resistant to reduction.     

Thiol/disulfide redox equilibrium and kinetic behavior of chicken liver fatty acid synthase

Chicken liver fatty acid synthase is rapidly inactivated and cross-linked at pH 7.2 and 8.0 by incubation with low concentrations of common biological disulfides including glutathione disulfide, coenzyme A disulfide, and glutathione-coenzyme A-mixed disulfide. Glutathione disulfide inactivation of the enzyme is accompanied by the oxidation of a total of 4-5 enzyme thiols per monomer. Only one glutathione equivalent is incorporated per monomer as a protein-mixed disulfide, and its rate of incorporation is significantly slower than the rate of inactivation. The formation of protein-SS-protein disulfides results in significant cross-linking of enzyme subunits. The inactive enzyme is rapidly and completely reactivated, and the cross-linking is completely reversed by incubation of the enzyme with thiols (10-20 mM) including dithiothreitol, mercaptoethanol, and glutathione. In a glutathione redox buffer (GSH + GSSG), disulfide bond formation comes to equilibrium. The enzyme activity at equilibrium is dependent both on the ratio of glutathione to glutathione disulfide and on the total glutathione concentration. The equilibrium constant for the redox equilibration of fatty acid synthase in a glutathione redox buffer is 15 mM (Ered + GSSG in equilibrium Eox + 2GSH). The formation of at least one protein-protein disulfide per monomer dominates the redox properties of the enzyme while the formation of one protein-mixed disulfide with glutathione (Kmixed = 0.45) has little effect on activity. The oxidation equilibrium constant suggests that there would be no significant cycling between the reduced and the oxidized enzyme in response to likely physiological variations in the hepatic glutathione status. The possibility that changes in the concentration of cellular glutathione may act as a mechanism for metabolic control of other enzymes is discussed.     

Irreversible inactivation of calcium-dependent proteinases from rat liver by biological disulfides

The effect of low-molecular-mass biological disulfides and their related reduced compounds on the activity of two calcium-dependent neutral proteinases (calpains) from rat liver has been investigated. L-Cystine and L-cystamine bring about the inactivation of both enzymes, while the related reduced compounds L-cysteine and L-cysteamine are without effect. Calpain II is more sensitive to the inactivating effect of glutathione disulfide in comparison with calpain I. The inactivation rates of both calpains depend on the concentration of glutathione disulfide. Reduced glutathione, added at physiological concentration (5 mM), neither affects the proteinase activities nor protects the enzymes from the inactivating effect of glutathione disulfide. The enzymes inactivated by biological disulfides cannot be restored by a large excess of a reducing thiolic compound (dithiothreitol). It is suggested that calcium-dependent proteinases might be inactivated also in vivo by enhanced level of glutathione disulfide.     

Reversible inactivation of Saccharomyces cerevisiae glutathione reductase under reducing conditions

Glutathione reductase from Saccharomyces cerevisiae was rapidly inactivated following aerobic incubation with NADPH, NADH, and several other reductants, in a time- and temperature-dependent process. The inactivation had already reached 50% when the NADPH concentration reached that of the glutathione reductase subunit. The inactivation was very marked at pH values below 5.5 and over 7, while only a slight activity decrease was noticed at pH values between these two values. After elimination of excess NADPH the enzyme remained inactive for at least 4 h. The enzyme was protected against redox inactivation by low concentrations of GSSG, ferricyanide, GSH, or dithiothreitol, and high concentrations of NAD(P)+; oxidized glutathione effectively protected the enzyme at concentrations even lower than GSH. The inactive enzyme was efficiently reactivated after incubation with GSSG, ferricyanide, GSH, or dithiothreitol, whether NADPH was present or not. The reactivation with GSH was rapid even at 0 degree C, whereas the optimum temperature for reactivation with GSSG was 30 degrees C. A tentative model for the redox interconversion, involving an erroneous intramolecular disulfide bridge, is put forward.     

Pyrimidine nucleoside monophosphate kinase from rat bone marrow cells: chromatographic, electrophoretic, and sedimentation behavior of active and inactive enzyme forms

This paper describes the study of a highly purified pyrimidine nucleoside monophosphate kinase from rat bone marrow cells. Short-term storage (24 h at 4 degrees C) of the purified enzyme in the absence of dithiothreitol, a sulfhydryl reducing agent, led to considerable losses of enzyme activity. Most of the lost activity could be regained, however, by incubating the enzyme with 50 mM dithiothreitol. Enzyme stabilization by dithiothreitol and reactivation by dithiothreitol were enhanced in the presence of phosphate buffer. Severe enzyme inhibition was produced by micromolar concentrations of sulfhydryl group reagents. Chromatographic, electrofocusing, and sucrose gradient centrifugation experiments revealed that the enzyme has a molecular weight of about 26,000, an isoelectric point of 4.7, and a sedimentation coefficient of 2.5. These experiments were also carried out with enzyme preparations which had been almost completely inactivated by means of dialysis to remove dithiothreitol. Enzyme preparations of this type displayed at least one additional enzyme form. This form(s) was inactive but capable of being partially reactivated by dithiothreitol. The inactive form(s) exhibited the same apparent molecular weight as the native enzyme but possessed a higher isoelectric point (5.7). A working hypothesis was presented which states (1) that inactive enzyme forms arise because of disulfide bond formation, (2) that enzyme sulfhydryl groups are less susceptible to oxidation in the presence of phosphate buffer, and (3) that enzyme reactivation by dithiothreitol results from the regeneration of critical enzyme sulfhydryls.     

Multiple oxidation products of sulfhydryl groups near the active site of thiolase I from porcine heart

The inactivation of porcine heart thiolase I with the disulfide reagents 5,5'-dithiobis(2-nitrobenzoate) (DTNB) and 2,2- and 4,4-dithiopyridine in 0.2 M phosphate buffer, pH 7.5, follows second-order kinetics with rate constants of 2.2 X 10(2), 25 X 10(2), and 5.8 X 10(2) M-1 min-1, respectively. Stoichiometric concentrations of the thiol-oxidizing reagent diethyl azodicarboxylate inactivate thiolase in less than 1 min at pH 7.5. The presence of saturating concentrations of the substrate acetoacetyl coenzyme A or the formation of the acetyl enzyme (a normal catalytic intermediate) results in a significant protection against the inactivation of thiolase by DTNB, 2,2-dithiopyridine, and diethyl azodicarboxylate. All five sulfhydryl residues of native thiolase react with either of the dipyridyl disulfides, but only the equivalent of 3.2 residues react with DTNB even at high concentrations and prolonged incubation times. The reaction of thiolase with DTNB leads to the formation of 1.0-1.4 mol of intrachain disulfide and 0.65 mol of mixed disulfides. After inactivation of thiolase with an equimolar concentration of diethyl azodicarboxylate, 1.2 mol of intrachain disulfide per subunit is found. No cross-linking between the subunits occurs as a result of the reaction of thiolase with DTNB or diethyl azodicarboxylate. The DTNB-inactivated enzyme can be reactivated with excess dithiothreitol while the diethyl azodicarboxylate inactivated enzyme is totally resistant to reactivation by dithiothreitol. There appear to be at least two different ways of forming inactive, oxidized enzyme products depending on the oxidant used, suggesting the possibility of multiple sulfhydryl groups at or near the active site.     

Regulation of rat liver microsomal glutathione S-transferase activity by thiol/disulfide exchange

The regulation of purified glutathione S-transferase from rat liver microsomes was studied by examining the effects of various sulfhydryl reagents on enzyme activity with 1-chloro-2,4-dinitrobenzene as the substrate. Diamide (4 mM), cystamine (5 mM), and N-ethylmaleimide (1 mM) increased the microsomal glutathione S-transferase activity by 3-, 2-, and 10-fold, respectively, in absence of glutathione; glutathione disulfide had no effect. In presence of glutathione, microsomal glutathione S-transferase activity was increased 10-fold by diamide (0.5 mM), but the activation of the transferase by N-ethylmaleimide or cystamine was only slightly affected by presence of glutathione. The activation of microsomal glutathione S-transferase by diamide or cystamine was reversed by the addition of dithiothreitol. Glutathione disulfide increased microsomal glutathione S-transferase activity only when membrane-bound enzyme was used. These results indicate that microsomal glutathione S-transferase activity may be regulated by reversible thiol/disulfide exchange and that mixed disulfide formation of the microsomal glutathione S-transferase with glutathione disulfide may be catalyzed enzymatically in vivo.     

Purification and some properties of bovine liver cytosol thioltransferase

A cytosol thioltransferase was purified 37,000-fold from bovine liver by essentially the same procedure as reported for rat liver enzyme by Axelsson et al. [1978) Biochemistry 17, 2978-2984). The purified enzyme appears to be homogeneous on sodium dodecyl sulfate (SDS)-gel electrophoresis and has a molecular weight (Mr) of 11,000, an isoelectric point (pI) of 8.1, and an optimum pH with S-sulfocysteine and GSH as substrates of 8.5. It is specific for disulfides including L-cystine, S-sulfocysteine, ribonuclease A, trypsin, soybean kunitz trypsin inhibitor, soybean Bowman Birk trypsin inhibitor and insulin, and converts Bowman Birk trypsin inhibitor to an inactive form. The enzyme does not act as a protein : disulfide isomerase, as measured by reactivation of "scramble" ribonuclease and Kunitz soybean trypsin inhibitor. Thioltransferase activity was found in cytosol of various bovine tissues.     

Action of thrombin on ovine, bovine and human pituitary growth hormones.

1. This communication reports the action of bovine thrombin on ovine, bovine and human growth hormones. Thrombin cleavage was shown to be restricted to a single homologous peptide bond in all three growth hormones (at sequence positions 133--134 of the ovine and bovine hormones). 2. Ovine growth hormone was the most sensitive to the action of thrombin, bovine growth hormone was attacked to a relatively less extent, and human growth hormone was the most resistant to the enzyme. 3. After reduction and carbamidomethylation of the disulfide bonds in thrombin modified ovine growth hormone, the two fragments (residues 1--133 and 134--191) were isolated. The large NH2-terminal thrombin fragment of the hormone (residues 1--133) was found to be inactive in the rat tibia test, whereas a tryptic fragment (residues 96--133) isolated in an independent way gave measurable responses.     

Glutathione mixed disulfide inhibitors of the human placental NADP-linked 15-hydroxyprostaglandin dehydrogenase

Six glutathione-containing inhibitors of the human NADP-linked 15-hydroxyprostaglandin dehydrogenase have been isolated from placental homogenates. Glutathione disulfide is one of these inhibitors. Although the structures of the other five have not been fully elucidated, all are disulfides. Studies with these compounds and with other mixed disulfides have shown that the glutathione mixed disulfides of beta-mercaptopyruvate, mercaptoacetate, and beta-mercaptolactate are more effective inhibitors of the enzyme than are the glutathione-containing mixed disulfides isolated from placental homogenates. beta-Mercaptolactate is particularly noteworthy because of its low Ki (0.13 microM). The results reported here suggest that the activity of the prostaglandin dehydrogenase may be influenced in vivo by various glutathione mixed disulfides.     

Assessment of lysyl oxidase variants by urea gel electrophoresis: evidence against disulfide isomers as bases of the enzyme heterogeneity

Methods for the copurification and rapid assessment of the protein profiles corresponding to the multiple variants of bovine aortic lysyl oxidase are described. The individual variants do not resolve from each other by electrophoresis in sodium dodecyl sulfate but are resolved by gel electrophoresis in 8 M urea, thus providing a new method for their detection independent of enzyme assay. Alkylation of the purified mixture of the variants with iodoacetamide after reduction with dithiothreitol identified three disulfides per 32,000-Da monomer. Urea gel electrophoresis revealed that the heterogeneity of lysyl oxidase persists after reduction and alkylation, indicating that disulfide isomers are not the bases of the enzyme heterogeneity.     

Thiol/disulfide redox equilibrium between glutathione and glycogen debranching enzyme (amylo-1,6-glucosidase/4-alpha-glucanotransferase) from rabbit muscle

Rabbit skeletal muscle glycogen debranching enzyme is inactivated in a kinetically biphasic manner by GSSG at pH 8.0. The rapid phase results in the loss of 30% activity, while the slower phase leads to total enzyme inactivation. Both the glucosidase and the transferase activities of the enzyme are inhibited by GSSG. The inactivation by disulfides is fully and rapidly reversed in a biphasic manner by reduction with excess reduced dithiothreitol or GSH. After a fast initial recovery of 70% of the initial activity, the remaining 30% of the activity is recovered more slowly. Equilibration of the enzyme with a redox buffer of GSH and GSSG shows a monophasic equilibration of the activity. The ratio of GSH/GSSG where the enzyme is 50% active (R0.5) is 0.06 +/- 0.03. The R0.5 does not vary significantly with the total concentration of glutathione species suggesting formation of protein-SSG mixed disulfides. The ratios of the observed second-order rate constants for GSSG inactivation and GSH reactivation do not lead to a correct value of the observed thiol/disulfide oxidation equilibrium constant. Although the enzyme has sulfhydryl groups, the oxidation of which leads to activity changes, the kinetic and thermodynamic resistance to oxidation suggests that the enzyme is not likely to be subject to regulation by thiol/disulfide exchange in vivo.     

Flavonoids: potent inhibitors of arachidonate 5-lipoxygenase

Various flavonoids were found to be relatively selective inhibitors of arachidonate 5-lipoxygenase which initiates the biosynthesis of leukotrienes with the activity of slow reacting substance of anaphylaxis. Cirsiliol (3',4',5-trihydroxy-6,7-dimethoxyflavone) was most potent, and the enzyme partially purified from rat basophilic leukemia cells was inhibited by 97% at a concentration of 10 microM (IC50, about 0.1 microM). 12-Lipoxygenases from bovine platelets and porcine leukocytes were also inhibited but at higher concentrations (IC50, about 1 microM), and fatty acid cyclooxygenase purified from bovine vesicular gland was scarcely affected. The compound at 10 microM suppressed by 99% the immunological release of slow reacting substance of anaphylaxis from passively sensitized guinea pig lung (IC50, about 0.4 microM).