Translocation of intracellular glutathione to membrane-bound γ-glutamyl transpeptidase as a discrete step in the γ-glutamyl cycle: Glutathionuria after inhibition of transpeptidase

Several inhibitors of gamma-glutamyl transpeptidase in vitro [L-serine plus borate, 6-diazo-5-oxo-L-norleucine, and L- and D-gamma-glutamyl-(o-carboxy)phenylhydrazide] are active in vivo, as indicated by their effect in decreasing the conversion of administered D-gamma-glutamyl-L-alpha-amino[(14)C]butyrate to respiratory (14)CO(2) in mice. The hydrazides (both L and D isomers) are the most potent inhibitors in vitro and in vivo. Inhibition of gamma-glutamyl transpeptidase in vivo by the hydrazides is accompanied by extensive glutahionuria. The evidence suggests that a substantial fraction of the urinary glutathione arises from the kidney. The findings support the view that renal intracellular glutathione is normally translocated to the membrane-bound gamma-glutamyl transpeptidase as a separate step in the gamma-glutamyl cycle. Studies on in vivo inhibition of glutathione synthesis and of gamma-glutamyl transpeptidase provide direct evidence that glutathione is normally translocated from tissues to the blood plasma and that the turnover of plasma glutathione is relatively high. The data suggest that the low but significant steady-state level of glutathione in the plasma reflects synthesis of glutathione (predominantly in the liver) and its utilization by gamma-glutamyl transpeptidase (predominantly in the kidney). Thus, glutathione synthesized in cells that have transpeptidase may be translocated to and used by the membrane-bound enzyme, whereas glutathione synthesized in cells that lack the transpeptidase may be transported via the plasma to transpeptidase located on the membranes of other cells.     

Excretion of cysteine and gamma-glutamylcysteine moieties in human and experimental animal gamma-glutamyl transpeptidase deficiency.

Animals treated with potent gamma-glutamyl transpeptidase inhibitors and a patient with severe gamma-glutamyl transpeptidase deficiency excrete much larger than normal amounts of glutathione, gamma-glutamylcysteine, and cysteine in their urine; these compounds were found in disulfide forms. The findings indicate that the metabolic function of gamma-glutamyl transpeptidase is associated with the metabolism or transport (or both) of cysteine, gamma-glutamylcysteine, and glutathione, and that gamma-glutamylcysteine is a physiological substrate of the enzyme. The occurrence of gamma-glutamylcysteine in urine and other considerations suggest that this dipeptide is formed as an extracellular metabolite of glutathione in addition to its recognized role as an intrcellular precursor of glutathione. The dipeptide may be formed by a pathway involving transpeptidation or by cleavage of the Cys-Gly bond of glutathione. In the course of this work it was found that the mixed disulfide between glutathione and gamma-glutamylcysteine is a good substrate of glutathione reductase.     

Transport of gamma-glutamyl amino acids: role of glutathione and gamma-glutamyl transpeptidase.

This work relates to the hypothesis that one of the mechanisms that mediates amino acid translocation across cell membranes involves the action of membrane-bound gamma-glutamyl transpeptidase on intracellular glutathione and extracellular amino acids to form gamma-glutamyl amino acids. According to this idea, the latter are translocated into the cell where the gamma-glutamyl moiety is removed to yield free amino acids. Previous studies in this laboratory showed that intracellular glutathione is translocated out of many cells. We have now directly examined the transport of gamma-glutamyl amino acids into tissues in the mouse by use of the model substrate L-gamma-glutamyl-L-[14C]methionine sulfone. Of 11 tissues examined, only the kidney showed strong and preferential uptake of the substrate. A substantial amount of the administered L-gamma-glutamyl-L-[14C]methionine sulfone was found intact in the kidney; the total uptake of this compound was greater (by about 2-fold) than that of free L-methionine sulfone. Studies with a number of other gamma-glutamyl amino acids and gamma-glutamyl compounds indicate that the kidney has a relatively specific transport system for gamma-glutamyl amino acids. Small but significant amounts of gamma-glutamylmethionine sulfone were found in the liver and pancreas, suggesting that other tissues may also have this system. Transport of gamma-glutamylmethionine sulfone into the kidney was inhibited by inhibitors of glutathione synthesis and of gamma-glutamyl transpeptidase. The results suggest that both the transpeptidase and glutathione may be involved in transport of gamma-glutamyl amino acids.     

Evidence that the gamma-glutamyl cycle functions in vivo using intracellular glutathione: effects of amino acids and selective inhibition of enzymes.

The function of the gamma-glutamyl cycle was explored in in vivo studies in which amino acids and specific inhibitors of cycle enzymes (gamma-glutamyl transpeptidase, gamma-glutamyl cyclotransferase, gamma-glutamylcysteine synthetase, and 5-oxoprolinase) were administered to mice. The findings, which show that the gamma-glutamyl cycle functions in vivo, support the conclusion that gamma-glutamyl amino acids formed by gamma-glutamyl transpeptidase from externally supplied amino acids and intracellular glutathione are translocated into the cell and thus indicate that there is a significant physiological connection between the metabolism of glutathione and the transport of amino acids.     

Translocation of glutathione from lymphoid cells that have markedly different gamma-glutamyl transpeptidase activities.

Translocation of intracellular glutathione to the medium was studied in lymphoid cells (grown in tissue culture) that have very high, very low, or intermediate levels of membrane-bound gamma-glutamyl transpeptidase, in the absence and presence of various inhibitors of this enzyme. The data show that glutathione is translocated to the medium by all of the cell lines studied, but that glutathione does not accumulate in the medium unless the cellular transpeptidase activity is either very low or substantially inhibited. Translocation of glutathione does not seem to be directly related to the activity of gamma-glutamyl transpeptidase. The present and previous [Griffith, O.W. & Meister, A. (1979) Proc. Natl. Acad. Sci. USA 76, 268--272] findings suggest that translocation of intracellular glutathione is a general property of many mammalian cells. Glutathione exported from cells that have membrane-bound transpeptidase may be recovered by the cell in the form of transpeptidation or degradation products. Translocation of glutathione may also reflect operation of a rather general mechanism that protects and maintains the integrity of cell membranes.     

Conversion of glutathione to glutathione disulfide by cell membrane-bound oxidase activity.

An apparently specific glutathione oxidase activity is present in renal cortex, epididymal caput, jejunal villus tip cells, choroid plexus, and retina (but not in liver). The activity is membrane-bound and is localized on the luminal surface of the brush border membranes of the kidney and jejunum. The distribution and localization of the oxidase are similar to those of gamma-glutamyl transpeptidase, suggesting that there is a significant relationship among the translocation of intracellular glutathione, the extracellular oxidation of glutathione to glutathione disulfide, and the reactions of the gamma-glutamyl cycle. Thus, both glutathione present in the blood plasma and intracellular glutathione translocated to the cell surface are accessible to oxidation and transpeptidation. Acceptor substrates of the transpeptidase (e.g., L amino acids) promote transpeptidation and decrease oxidation of glutathione. Conversion of glutathione to glutathione disulfide is followed by utilization of the latter compound by gamma-glutamyl transpeptidase and dipeptidase. Although intracellular oxidation of glutathione to glutathione disulfide is readily reversed by the action of glutathione reductase, glutathione disulfide formed extracellularly cannot be reduced; instead, it undergoes hydrolytic and transpeptidation reactions leading to gamma-glutamyl amino acid and amino acid products which may be recovered by being transported into the cell.     

Intrahepatic transport and utilization of biliary glutathione and its metabolites.

Glutathione transported by hepatocytes into the bile canaliculi is metabolized by the actions of gamma-glutamyl transpeptidase and dipeptidase located on the biliary ductular epithelium. This pathway is revealed by the finding of high levels of cyst(e)inylglycine, gamma-glutamylglutathione, gamma-glutamylcyst(e)ine, glutamate, glycine, and cyst(e)ine in bile, by studies in which intrahepatic metabolism of glutathione was inhibited by administration of a potent inhibitor of gamma-glutamyl transpeptidase and by experiments in which glutathione synthesis was inhibited. Canalicular transport of glutathione, as estimated from totals of metabolites found, is much greater than the glutathione found in bile. Glutathione and glutathione metabolites found in bile increase with age, in association with an increase in hepatic glutathione. In younger rats there is apparent uptake of cysteine and glycine moieties that may reflect uptake of cysteinylglycine at the ductular level. This intrahepatic pathway of glutathione transport and metabolism, which resembles that which occurs in the kidney, seems to function as a cellular protective mechanism in the processing of glutathione conjugates and as a recovery system for cysteine moieties.     

Formation of gamma-glutamycyst(e)ine in vivo is catalyzed by gamma-glutamyl transpeptidase.

These studies indicate that gamma-glutamylcyst(e)ine, found in the urine of a patient with gamma-glutamyl transpeptidase deficiency and also in the urine of experimental animals injected with glutathione or with inhibitors of gamma-glutamyl transpeptidase, is formed by the action of gamma-glutamyltranspeptidase. The evidence demonstrates that transpeptidation between glutathione and cystine occurs in vivo and also that this reaction constitutes a significant physiological function of the enzyme. The appearance of large amounts of gamma-glutamylcyst(e)ine in the urine seems to reflect an inhibitory effect of glutathione on the transport of gamma-glutamylcyst(e)ine into cells. The findings also indicate that conversion of glutathione to gamma-glutamylcysteine by hydrolytic cleavage of the COOH-terminal glycine moiety of glutathione (or analogous cleavage of glutathione disulfide) is not a quantitatively significant pathway. The results reported here show that gamma-glutamyl transpeptidase activity is not completely absent in a patient found to have a deficiency of this enzyme and that the activity of the enzyme is not abolished in experimental animals treated with potent gamma-glutamyl transpeptidase inhibitors.     

The γ-Glutamyl Cycle: A Possible Transport System for Amino Acids

Evidence is presented that rat kidney contains enzymes that catalyze the synthesis and utilization of glutathione; these reactions, which involve the uptake and release of amino acids from gamma-glutamyl linkage, constitute a cyclical process which is termed "the gamma-glutamyl cycle." The gamma-glutamyl cycle has properties that fulfill the requirements of an amino acid transport system. Thus, gamma-glutamyl transpeptidase may function in translocation and gamma-glutamylcysteine synthetase and glutathione synthetase may catalyze energy-requiring "recovery" steps in transport. These and other considerations suggest that glutathione serves a carrier function in amino acid transport.     

High resistance to cisplatin in human ovarian cancer cell lines is associated with marked increase of glutathione synthesis.

Exposure of human ovarian tumor cell lines to cisplatin led to development of cell lines that exhibited increasing degrees of drug resistance, which were closely correlated with increase of the levels of cellular glutathione. Cell lines were obtained that showed 30- to 1000-fold increases in resistance; these cells also had strikingly increased (13- to 50-fold) levels of glutathione as compared with the drug-sensitive cells of origin. These levels of resistance to cisplatin and the cellular glutathione levels are substantially greater than previously reported. Very high cisplatin resistance was associated with enhanced expression of mRNAs for gamma-glutamylcysteine synthetase and gamma-glutamyl transpeptidase; immunoblots showed increase of gamma-glutamylcysteine synthetase but not of glutathione synthetase. Glutathione S-transferase activity was unaffected, as determined with chlorodinitrobenzene as a substrate. These studies suggest the potential value of examining regulation of glutathione synthesis as an indicator of clinical prognosis. The highly resistant cell lines are proving useful for studying the multiple mechanisms by which tumor cells acquire drug- and radiation-resistance.     

Glutathione deficiency leads to mitochondrial damage in brain.

Glutathione deficiency induced in newborn rats by giving buthionine sulfoximine, a selective inhibitor of gamma-glutamylcysteine synthetase, led to markedly decreased cerebral cortex glutathione levels and striking enlargement and degeneration of the mitochondria. These effects were prevented by giving glutathione monoethyl ester, which relieved the glutathione deficiency, but such effects were not prevented by giving glutathione, indicating that glutathione is not appreciably taken up by the cerebral cortex. Some of the oxygen used by mitochondria is known to be converted to hydrogen peroxide. We suggest that in glutathione deficiency, hydrogen peroxide accumulates and damages mitochondria. Glutathione, thus, has an essential function in mitochondria under normal physiological conditions. Observations on turnover and utilization of brain glutathione in newborn, preweaning, and adult rats show that (i) some glutathione turns over rapidly (t 1/2, approximately 30 min in adults, approximately 8 min in newborns), (ii) several pools of glutathione probably exist, and (iii) brain utilizes plasma glutathione, probably by gamma-glutamyl transpeptidase-initiated pathways that account for some, but not all, of the turnover; thus, there is recovery or transport of cysteine moieties. These studies provide an animal model for the human diseases involving glutathione deficiency and are relevant to oxidative phenomena that occur in the newborn.     

Interconversion of leukotrienes catalyzed by purified gamma-glutamyl transpeptidase: concomitant formation of leukotriene D4 and gamma-glutamyl amino acids.

The reversible conversion of leukotriene C4 to leukotriene D4 and of the latter to leukotriene E4 were studied with highly purified homogeneous preparations of gamma-glutamyl transpeptidase, dipeptidase, and aminopeptidase M. The conversion of leukotriene C4 to leukotriene D4, catalyzed by gamma-glutamyl transpeptidase, is significantly more rapid when carried out in the presence of an amino acid mixture a closely approximating that found in blood plasma and is accompanied by gamma-glutamyl amino acid formation. Because gamma-glutamyl transpeptidase is bound to the external surface of cell membranes and thus is readily accessible to plasma amino acids, it appears that conversion of leukotriene C4 to leukotriene D4 under physiological conditions is coupled with the formation of gamma-glutamyl amino acids. The apparent Km value for leukotriene C4 in this reaction is about 6 X 10(-6) M, a value close to that found for glutathione. Conversion of leukotriene D4 to leukotriene C4 is effectively catalyzed by gamma-glutamyl transpeptidase in the presence of relatively low concentrations of glutathione. The conversion of leukotriene D4 to leukotriene E4 is catalyzed much more rapidly by renal dipeptidase than by renal aminopeptidase M. Incubation of leukotriene E4 with gamma-glutamyl transpeptidase and glutathione leads to formation of a compound with the properties of gamma-glutamyl leukotriene E4; this reaction is analogous to that shown previously in which gamma-glutamyl cystine is formed by transpeptidation between glutathione and cystine.     

Inhibition of gamma-glutamyl transpeptidase and induction of glutathionuria by gamma-glutamyl amino acids.

Treatment of mice with various gamma-glutamyl amino acids leads to marked urinary excretion of glutathione and other gamma-glutamyl compounds. There is good correlation between the affinity of gamma-glutamyl transpeptidase for various gamma-glutamyl amino acids and the extent of glutathionuria. The findings indicate that the administered gamma-glutamyl compounds effectively compete with glutathione (exported from kidney cells and present in the glomerular filtrate) for the enzyme. The administration of certain gamma-glutamyl amino acids appears to be a specific and nontoxic procedure for in vivo inhibition of gamma-glutamyl transpeptidase that may be useful in experimental work on glutathione metabolism and function and also for treatment of certain toxicities and for modulation of the metabolism of endogenous glutathione conjugates.     

Effects of age on rat glutathione metabolism.

The authors hypothesized that rat plasma or tissue glutathione metabolism could change with age due to possible decreases in glutathione-related enzyme activities. To test this hypothesis, the authors measured plasma and tissue concentrations of glutathione and glutathione-related enzymes. Animals were 3 months, 12 months, or 24 months old at the time of experiments. Plasma glutathione was found to be significantly increased in both the 12-month-old and 24-month-old groups compared to the 3-month-old rats. Tissue enzyme measurements showed no significant differences between the groups in lung or liver glutathione peroxidase or glutathione S-transferase. gamma-Glutamyl transpeptidase activity was significantly decreased in kidney and lung with aging. Decreases in tissue gamma-glutamyl transpeptidase activity occur with age; this may contribute to increases in plasma glutathione concentrations.     

Radioprotection of human lymphoid cells by exogenously supplied glutathione is mediated by gamma-glutamyl transpeptidase.

Human lymphoid cells depleted of glutathione by treatment with buthionine sulfoximine, a specific inhibitor of gamma-glutamylcysteine synthetase, may be partially repleted by adding glutathione in the medium. The mechanism of repletion involves the action of gamma-glutamyl transpeptidase on exogenous glutathione, transport of products of glutathione metabolism, and intracellular synthesis of glutathione. Lymphoid cells, previously shown to export glutathione at rates proportional to intracellular glutathione levels, do not take up intact glutathione to an appreciable extent, even under conditions of marked glutathione deficiency. The role of glutathione in radioprotection was examined by subjecting cells to gamma-radiation after modification of cellular glutathione levels. Glutathione-depleted cells exhibited increased radiosensitivity under aerobic conditions, as compared to the nondepleted controls. Partial repletion of cellular glutathione prior to irradiation led to radiosensitivity comparable to nondepleted controls. Cells were not protected by suspension in media containing glutathione just prior to irradiation; thus, protection appears to require intracellular glutathione.     

Studies on gamma-glutamyl transpeptidase in human and rabbit erythrocytes.

Gamma-glutamyl transpeptidase transfers the gamma-glutamyl moiety of glutathione to a variety of acceptor amino acids. Through the operation of the gamma-glutamyl-cyclotransferase cycle, this enzyme has been implicated in the transport of amino acids into cells, especially the cells of the proximal tubules of kidney. It has been reported to be present in rabbit erythrocytes. However, using white cell-free preparations, we have not been able to demonstrate the presence of gamma-glutamyl transpeptidase in human or rabbit erythrocytes either by measuring the utilization of GSH or by following the formation of the product. 14C-L-methionine was used as acceptor amino acid, and the formation of gamma-glutamyl-14C-L-methionine was followed. Using similar conditions, we have been able to demonstrate the presence of gamma-glutamyl transpeptidase in human and rabbit leukocytes and in human kidney. In contrast to a previous report, we were unable to find the accumulation of 5-oxoproline, an intermediate of the gamma-glutamyl-cyclotransferase pathway in human red cells incubated in Krebs-Ringer solution. Immunologic studies demonstrated that human red cell membranes contained no protein antigenically similar to kidney gamma-glutamyl transpeptidase. Thus our studies indicated that in human and rabbit erythrocytes, the gamma-glutamyl transpeptidase-cyclotransferase pathway was not operative.     

Glutathionuria: gamma-glutamyl transpeptidase deficiency

A mentally retarded young woman with severe behaviour problems was found to excrete large amounts of glutathione due to a generalized gamma-glutamyl transpeptidase deficiency. As in the only other case described in detail, plasma levels and renal reabsorption of the amino acids were normal. In the parents' urine, plasma and leukocytes, enzyme activity was normal but in their cultured fibroblasts it was below the minimum for the control range. An autosomal recessive mode of inheritance is suggested. The implications of these findings for possible role of the gamma-glutamyl cycle in amino acid transport are briefly discussed.     

Glutathione and gamma-glutamyl cycle enzymes in crypt and villus tip cells of rat jejunal mucosa.

Villus tip cells and crypt cells of rat jejunal mucosa were separated by the planning procedure of Imondi et al. and were studied with respect to their activities of the enzymes of the gamma-glutamyl cycle and glutathione content. The villus tip cells exhibit much higher gamma-glutamyl transpeptidase activities than do the crypt cells: thus, gamma-glutamyl trnaspeptidase appears to be a villus-specific enzyme. gamma-Glutamyl cyclotransferase and the enzymes required for glutathione synthesis are not specifically localized to either the crypt or villus tip cells but are present in both. The crypt cells have a high concentration of glutathione (4-5 mM) comparable to the levels found in liver and kidney; in contrast, the villus tip cells have much lower concentrations. On fasting, the glutathione concentration decreased markedly in both villus tip and crypt cells; feeding of protein, but not of sucrose, led to increased glutathione concentrations. The migration of cells from the undifferentiated crypt cell region to the villus tip is associated with structural and biochemical changes that equip the cell for its mature functional activities, which include transport. The present findings indicate that such cellular differentiation and migration is associated with a marked increase in gamma-glutamyl transpeptidase activity and in the utilization of glutathione.     

Accumulation of 5-Oxoproline in Mouse Tissues After Inhibition of 5-Oxoprolinase and Administration of Amino Acids: Evidence for Function of the γ-Glutamyl Cycle

5-Oxoprolinase catalyzes the conversion of 5-oxo-L-proline (L-pyroglutamate, L-2-pyrrolidone-5-carboxylate) to L-glutamate with concomitant stoichiometric cleavage of ATP to ADP and inorganic orthophosphate. In this reaction, a step in the gamma-glutamyl cycle, 5-oxoproline (formed by the action of gamma-glutamylcyclotransferase on gamma-glutamyl amino acids, which are in turn formed by transpeptidation of amino acids with glutathione), is made available for glutathione synthesis. When mice are injected with L-2-imidazolidone-4-carboxylate, a competitive inhibitor of 5-oxoprolinase, they accumulate 5-oxoproline in their tissues (kidney, liver, brain, and eye) and excrete it in the urine. Mice given the inhibitor together with one of several L-amino acids accumulate and excrete much more 5-oxoproline than when they are given the inhibitor alone. Such augmentation of 5-oxoproline accumulation offers evidence for the function of the gamma-glutamyl cycle in vivo and supports the view that 5-oxoproline is a quantitatively significant metabolite.     

Transport and direct utilization of gamma-glutamylcyst(e)ine for glutathione synthesis.

Administration of gamma-glutamylcystine or of gamma-glutamylcysteine disulfide to mice leads to significantly increased levels of glutathione in the kidney as compared to controls given glutamate plus cysteine (or cystinylbisglycine). Studies with gamma-glutamylcystine selectively labeled with 35S in either the internal or external S atom indicate preferential utilization of the gamma-glutamylcysteine moiety of this compound for glutathione synthesis. Mice depleted of glutathione by treatment with buthionine sulfoximine do not significantly use the disulfides gamma-glutamylcystine or gamma-glutamylcysteine disulfide but do use gamma-glutamylcysteine for glutathione synthesis. These findings suggest a pathway in which gamma-glutamylcystine, formed by transpeptidation between glutathione and cystine, is transported and reduced by transhydrogenation with glutathione to cysteine and gamma-glutamylcysteine; the latter is used directly for glutathione synthesis. The findings show transport of gamma-glutamyl amino acids, indicate an alternative pathway of glutathione synthesis, and demonstrate a means of increasing kidney glutathione levels.