the journal of biological 264, no. 10, of inc. in u.s.a ... · role of glutathione in copper...

THE JOURNAL OF BIOLOGICAL CHEMISTRY 0 1989 by The American Society for Biochemistry and Molecular Biology, Inc.

Vol. 264, No. 10, Issue of April 5, pp. 5598-5605,1999 Printed in U.S.A.

The Role of Glutathione in Copper Metabolism and Toxicity*

(Received for publication, October 31, 1988)

Jonathan H. Freedman$#, Maria Rosa Ciriolo$((, and Jack Peisach$** From the BZnstitute for Structural and Functional Studies, University City Science Center, Philadelphia, Pennsylvania 19140 and the Departments of $Molecular Pharmacology and **Physiology and Biophysics, Albert Einstein College of Medicine, Bronx, New York 10461

Cellular copper metabolism and the mechanism of resistance to copper toxicity were investigated using a wild type hepatoma cell line (HAC) and a copper-resistant cell line (HACeoo) that accumulates copper and has a highly elevated level of metallothionein (MT). Of the enzymes involved in reactive oxygen metabolism, only glutathionine peroxidase was elevated (3-4-fold) in resistant cells, suggestive of an increase in the cellular flux of hydrogen peroxide.

A majority of the cytoplasmic copper (>60%) was isolated from both cell lines as a GSH complex. Kinetic studies of “Cu uptake showed that GSH bound 67Cu before the metal was complexed by MT. Depletion of cellular GSH with buthionine sulfoximine inhibited the incorporation of 67Cu into MT by greater than 50%. These results support a model of copper metabolism in which the metal is complexed by GSH soon after entering the cell. The complexed metal is then transferred to MT where it is stored. This study also indicates that resistance to metal toxicity in copper-resistant hepatoma cells is due to increases in both cellular GSH and MT. Furthermore, it is suggested that elevated levels of GSH peroxidase allows cells to more efficiently ac- commodate an increased cellular hydrogen peroxide flux that may occur as a consequence of elevated levels of cytoplasmic copper.

Metallothioneins, low molecular weight, cysteine-rich metal-binding proteins, have a central role in hepatic copper metabolism, providing temporary storage for cytoplasmic copper (1-3). Fractionation of cytosols radiolabeled with copper has shown that the metal is initially complexed by a 10-kDa protein (4), with characteristics similar to those of MT,’ then donated to other proteins such as superoxide dismutase and ceruloplasmin (5). The mechanism by which MT accumulates copper and transfers metals to other proteins has not been determined. It has been suggested that cytoplasmic MT is in

Health Service Grants ES-05406 (to J. H. F.) and GM-40168 (to J. * This investigation was supported in part by United States Public

P.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

§ To whom correspondence should be sent: Albert Einstein College of Medicine, Bronx, NY 10461.

11 Present address: Dipartimento di Biologia, I1 Universita di Roma, Via 0. Raimondo, 00173 Roma, Italy.

The abbreviations used are: MT, metallothionein; DEM, diethyl maleate; BSO, L-buthionine-(S,R)-sulfoximine; FBS, fetal bovine serum; GSH, reduced glutathionine; GSSG, oxidized glutathione; GS- Cu, copper-glutathione complex; PBS, phosphate-buffered saline; HEPES, 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid FPLC, fast protein liquid chromatography.

equilibrium with a free copper pool (6). However, free copper would likely be toxic to cells.

Several theories have been proposed to explain the mechanism of cellular copper toxicity. It has been proposed that free copper binds to protein cysteinyl thiol groups thereby inactivating essential enzymes. If oxygen is present, the metal-thiolate bond may be oxidized, irreversibly inactivating enzymes by forming cystine cross-links (7). It has also been suggested that copper interacts with physiologically produced reduced oxygen species, such as superoxide anion and hydrogen peroxide, catalyzing the formation of highly toxic hydroxyl radicals via a Fenton-like reaction (8, 9). Hydroxyl radicals can then go on to inactivate essential enzymes and/ or carry out lipid peroxidation reactions, thereby disrupting membranes and organelles (10).

The production of hydroxyl radicals, associated with the cellular accumulation of copper, may occur soon after the metal enters the cell. Although MT is an efficient copper chelator, the accumulation of MT to a level sufficient to inhibit metal toxicity requires several hours after an initial exposure to metal (11-13). Thus, toxic effects of metals may develop and cells become irreparably damaged before MT can be protective.

It has been proposed that GSH is capable of chelating and detoxifying metals soon after they enter the cell (14-18). This proposal is based on observations that the depletion of GSH potentiates metal toxicity in rats (15), mice (17), and cultured cells (14, 16, 18). The importance of GSH in metal detoxifi- cation is also supported by its role in the removal of toxic oxygen species (8,9). Glutathione is also a substrate for GSH peroxidases, enzymes capable of both removing hydrogen peroxide from the cells and repairing peroxidatively damaged membranes (19).

Since GSH may function as an intracellular metal chelator, as well as a substrate for the removal of toxic oxygen species, the relationship between copper, GSH, and Cu-MT was examined using rat liver hepatoma cell lines resistant to the toxic effects of copper (20). These cells accumulate more copper than the nonresistant parental cell line and also con- tain elevated levels of MT.2 A copper-glutathione complex, which represented the majority of cytoplasmic copper, could be isolated from both resistant and nonresistant cell lines by anaerobic fractionation of cell lysates. Furthermore copper uptake and incorporation of metal into MT, in vivo, is strongly influenced by GSH levels. The results of these studies support a model of copper metabolism in which the metal is rapidly complexed by GSH after entering the cell and is then transferred to MT. Toxic oxygen species, such as hydrogen peroxide, that may be formed from the oxidation of the GS-Cu complex or Cu-MT (22, 23) are efficiently removed by GSH peroxidase. It is proposed that resistance to copper toxicity is

J. H. Freedman and J. Peisach, submitted for publication.

5598

Role of Glutathione in Copper Metabolism and Toxicity 5599

a consequence of increases in GSH, GSH peroxidase, and MT levels.

EXPERIMENTAL PROCEDURES

Materials-All media and media supplements were obtained from Grand Island Biological Co. Rabbit MT-I, xanthine, xanthine oxidase, cytochrome c (type VI), cycloheximide, hydrogen peroxide, and BSO were from Sigma. Cumene hydroperoxide, GSH, GSSG, polyethylene glycol 400, 5,5'-dithiobis(2-nitrobenzoic acid), DEM, and l-chloro- 2,4-dinitrobenzene were obtained from Aldrich. GSSG reductase and NADPH were purchased from Boehringer Mannheim. Imidazole and HEPES, Ultrol grade, were from Mallinckrodt Chemical Works and Calbiochem, respectively. [36S]Cysteine (specific activity >1200 Ci/ mmol) and L-[U-"C]glutamic acid (specific activity >250 mCi/mmol) were obtained from Amersham Corp. Carrier-free ( 6 7 C ~ ) ~ ~ p r i ~ chloride, in 2 N HCl, was prepared at Los Alamos National Laboratory as described previously (5). At the time of shipment the stock solution contained 0.4 p~ [67Cu] with a specific activity of 105 Ci/pmol.

Cells and Tissue Culture-The resistant cell line, HA%, grown in medium supplemented with 600 p M copper, and the wild type cell line, HAC, were cultured as described previously (20). A second resistant cell line, HACmNM, was developed by growing HACm cells in the absence of added copper for a minimum of 3 months. These cells had previously been shown to maintain resistance to copper toxicity and to produce MT at levels similar to those in the resistant parent, even when grown in the absence of added metal for longer than 12 months (20)? The amount of cytoplasmic copper in HACmNM cells, however, decreased to a level similar to that of wild type cells within 2 weeks after removal from the copper-enriched medium (20). All cell lines were maintained in RPMI 1640 in 10.0 mM HEPES buffer, pH 7.4, supplemented with 10% FBS, nonessen- tial amino acids, 2.0 mM glutamine, penicillin G (10 units/ml), and streptomycin (10 pg/ml), unless otherwise noted.

Glutathione Assay-Glutathione was measured in cells grown for 72 +- 4 h. Samples were prepared by washing 4 X 100-mm confluent plates of cells (-2 X 10' cells) three times with PBS (20). Cells were then removed with a rubber policeman and suspended in 10 ml of fresh PBS. Cell number was determined from a 0.5-ml aliquot using a Coulter counter. The remaining cell suspension was centrifuged at 1,000 X g for 10 min and the PBS-washed cell pellet resuspended in 0.5 ml of 10.0 mM HCl. Cells were disrupted by sonication and the lysate centrifuged at 35,000 X g for 15 min. Proteins were precipitated by adding sodium metaphosphoric acid to a final concentration of 5% (w/v), and the precipitate was removed by centrifugation at 14,000 X g for 15 min. The amount of GSH equivalents (GSH and GSSG) in the supernatant was determined as described by Anderson (24) and is expressed as nmol/106 cells.

The export of GSH from cells was determined by measuring the concentration of GSH equivalents in the medium (30 ml) in which cells had been growing for 72 h. Residual cells present were removed by centrifugation at 2,000 X g for 10 min. To precipitate serum proteins, sodium metaphosphoric acid was added to the supernatant to a final concentration of 5% (w/v), and the solution centrifuged at 35,000 X g for 15 min. The supernatant was lyophilized and then redissolved in 2.0 ml of 146 mM sodium phosphate, 6.3 mM EDTA buffer, pH 7.5. Glutathione equivalents were determined as described above and expressed as nmol/mg protein isolated from the cells grown in the medium. Copper added to control solutions, at concentrations similar to those present in cell and media samples, was shown to have no effect on the glutathione assay.

Enzyme Assays-Cells were isolated and lysates prepared as described above, except that PBS-washed cell pellets were resuspended in 50 mM potassium phosphate buffer, pH 7.4, prior to sonication. Superoxide dismutase activity, measured as described by McCord and Fridovich (25), is expressed as units/mg cell protein. One unit of activity is defined as the amount of protein required to inhibit the rate of cytochrome c reduction by 50%. Purified superoxide dismutase, used as a standard in the assay, was prepared from bovine erythro- cytes (26). Catalase activity was measured as described by Luck (27) and is expressed as mol/min/mg protein. GSH peroxidase activity, using hydrogen peroxide or cumene hydroperoxide as substrates, was determined as described by Lawrence and Burk (28). GSSG-reductase activity was assayed according to the method of Bergmeyer et al. (29). Both GSH peroxidase and GSSG reductase activities are expressed in nmol of NADPH oxidized per min/mg protein. GSH-&"transferase activity was determined by the method of Hibig et al. (30) using 1- chloro-2,4-dinitrobenzene as substrate. Enzyme activity is expressed

as pmol of product formed per min/mg protein. Protein concentration was determined by the method of Lowry et al. (31) using bovine serum albumin as a standard.

To assay y-glutamylcysteine synthetase activity, cell homogenates were prepared by suspending PBS-washed cell pellets in 1.0 ml of distilled water. Cells were lysed by 10 strokes of the Teflon pestle in a Potter-Elvehjem glass tissue grinder, and the homogenates were centrifuged for 10 min at 14,000 X g. Aliquots of the supernatant (10, 25, 50, and 100 pl) were added to 0.5 M imidazole HC1 buffer, pH 8.25, to a final volume of 570 pl. Assay reactions were initiated by the addition of 430 pl of the substrate mixture as described by Paniker and Beutler (32). Activity is expressed as pmol of y-["CIglutamyl- cysteine produced per min/mg protein. Radioactivity was measured using an LKB Rackbeta counter.

Uptake of Copper-Wild type and resistant cells were plated onto 96-well multiwell plates (Falcon) and incubated for 48 h under standard conditions (20). To deplete cellular GSH, BSO (0.9 mM) or DEM (0.18 mM) in 0.05% polyethylene glycol 400 was added to PBS-washed cells in fresh medium and then incubated for 8 and 3 h, respectively (33,34). PBS-washed cells were prepared by removing the old medium and then twice rinsing the wells of cells with sterile, 37 "C PBS and adding fresh RPMI 1640 containing 1% FBS? To inhibit protein synthesis PBS-washed cells were incubated with cycloheximide (4.6 pg/ml) for 3 h (35).

Following depletion of GSH or inhibition of protein synthesis, 4 X IO6 cpm of 67Cu (-30 pmol), dissolved in serum-free RPMI 1640, pH 7.4, were added to each well containing cells. These were then incubated for 30 min and 90 min. At the end of the incubation, the uptake medium was removed, cells were washed three times with ice-cold PBS, containing 1.0 mM EDTA, and then lysed by adding 200 pl of 1.0 N sodium hydroxide and incubating for 3 h at room temperature. The 67Cu content was determined by measuring the radioactivity in a 100-pl aliquot of the lysate with an LKB Labgamma counter. Protein concentration was determined in the remaining lysate as described above.

Intracellular Distribution of Copper-The kinetics of intracellular distribution of copper was analyzed in HAC and HACm cells grown on 35-mm plates for 48 h under standard conditions. After this time, cells were washed three times with sterile, 37 "C PBS and then incubated for 0.5 to 12 h in the presence of 8 X lo7 cpm of 67Cu (-0.6 nmol) in RPMI 1640, containing 1% FBS? At the end of the incubation, the uptake medium was removed, the cells were washed three times with ice-cold PBS/EDTA, and 0.5 ml of 10.0 mM HEPES buffer, pH 7.4, was added. After 5 min, cells were removed by gentle pipetting, transferred to a 1.5 ml microcentrifuge tube, and pelleted by centrifugation at 14,000 X g for 10 s. Cell pellets were resuspended in 0.5 ml of HEPES buffer and then lysed by sonication. The lysate was centrifuged for 15 min at 14,000 X g, and the supernatant was collected and then stored at -70 "C.

To determine the intracellular distribution of copper, a 0.5-ml aliquot of the supernatant was fractionated by gel filtration chromatography using a 1.6 X 50-cm Superose 12B FPLC column, equilibrated with 10.0 mM HEPES, 0.15 M sodium chloride buffer, pH 7.4, made anaerobic by purging with ultra-high purity helium (Linde) for a minimum of 30 min. Proteins were eluted at a flow rate of 1.5 ml/ min, 1.0-ml fractions were collected, and the 67Cu content in each fraction was measured. Molecular weights of the material eluted by gel filtration chromatography were determined by comparing their elution volume with those of molecular weight standards.

The amount of 67Cu incorporated into MT, superoxide dismutase, and GS-Cu was determined by integrating the area under peaks in the chromatogram from the normalized elution profile. The chromat- ographic data were normalized to account for differences in the amounts of 67Cu and protein applied to the column. The elution volume for superoxide dismutase was determined by chromatographing unlabeled cell lysates or purified superoxide dismutase and assay- ing aliquots from each fraction for enzyme activity. The elution volumes for Cu-MT and GS-Cu were identified as described below. The percent of T u coeluting with these components was determined from the areas under the respective peaks relative to the total integrated area of the elution profile. The kinetics of copper incorporation into subcellular pools was analyzed by fitting copper concentrations uersus time to a pseudo-first order rate equation by an iterated least squares analysis procedure described previously (20).

The effect of inhibiting protein synthesis or decreasing the cellular

The medium added to HACm cells was not supplemented with additional copper.

5600 Role of Glutathione in Copper Metabolism and Toxicity GSH concentration on copper incorporation into MT was examined. Cells were grown on 35-mm plates as described above. Protein synthesis was inhibited by preincubating cells with cycloheximide (5 pg/ ml) for 3 h, or GSH was depleted by preincubating with BSO for 8 h (1.0 mM), prior to 67Cu addition. Radiolabeled copper (8 X 10’ cpm) was added to each plate and the cells incubated for 90 min. At the end of this incubation, cell lysates were prepared and the proteins fractionated as described above. As controls, the concentration of GSH was measured after treating cells with BSO or cycloheximide. The levels of protein synthesis in cells treated with cycloheximide or BSO, as described above, were determined by labeling cells for 90 min with [3SS]cysteine and measuring the amount of radiolabel incorporated into trichloroacetic acid-precipitable material.

RESULTS

Enzyme Activities of Oxygen-metabolizing Enzymes-The relative activities of superoxide dismutase, catalase, and GSH peroxidase were compared between wild type and resistant cells to ascertain whether elevated copper levels in resistant cells caused an increased flux of toxic oxygen species (36,37). Results presented in Table I show that there was no significant difference in superoxide dismutase or catalase activities between the three cell lines. Glutathione peroxidase activity, with hydrogen peroxide as substrate, however, was increased in the resistant cell lines, suggesting an increase in hydrogen peroxide flux.

Glutathione Content in Wild Type and Resistant Cells-The content of GSH in wild type and resistant cells was determined. Copper-resistant cells contained higher levels of GSH equivalents (GSH and GSSG) than nonresistant cells (Table 11). In addition, HACwNM cells grown in the absence of added copper for 3-6 months contained more GSH than HAC~OO cells maintained in a copper-enriched medium.

Changes in the level of GSH equivalents in the resistant cell lines may be due to a defect in the reduction of GSSG, a change in the conversion of GSH into covalent bound deriv- atives, an increase in GSH synthesis, or a decrease in GSSG efflux (34). To assess which of these activities may be responsible for the changes observed, the activities of GSSG-reductase, GSH-S-transferase, y-glutamylcysteine synthetase, and the level of GSSG efflux from cells were determined. No significant differences in enzyme activities between the three cell lines were observed (Table 11). However, there was three to seven times less glutathione in the medium in which resistant cells were maintained, compared with wild type cells. These results suggest that the elevated level of glutathione found in resistant cells is due to a decrease in glutathione efflux.

Identification of Copper-Glutathione and Copper-Glutathi-

TABLE I Superoxide dismutase, catalase, and glutathione peroxidase activities

in hepatoma cells Enzyme activities for wild type (HAC) and resistant cells main-

tained in copper-enriched medium (HAC,) or nonenriched medium (HACWNM) were determined as described under “Experimental Procedures.” Enzvme activities are means f S.E.

~ ~

Enzyme

Superoxide dismutase ( n = 8) (units/mg protein)

Catalase ( n = 6) (mol/min/ mg protein)

GSH peroxidase ( n = 8) (nmol/min/mg protein)

Cumene hydroperoxide Hydrogen peroxide

Cell line

HAC HAC- HACwNM

28.2 f 5.1 23.3 6.3 31.2 f 6.0

0.32 f 0.10 0.28 f 0.08 0.31 f 0.07

41.1 & 3.9 35.0 f 1.4 41.5 f 2.0 3.9 f 1.7 6.8 f 2.7” 16.1 f 0.7”

Significantly different, p < 0.05, from wild type cells.

TABLE I1 Glutathione content and activities of glutathione metabolizing

enzymes in hepatoma cell lines Glutathione levels and enzyme activities for wild type (HAC) and

resistant cells maintained in copper-enriched medium (HAC,) or nonenriched medium (HACmNM) were determined as described under “Experimental Procedures.” The values are means f S.E.

Cell line

HAC HACw HACmNM

Cellular glutathione ( n = 8) 8.1 f 1.7 21.7 f 9.1“ 37.6 f 10.1’

Media glutathione ( n = 2) 0.49 f 0.09 0.16 & 0.03” 0.04 0.02”

y-Glutamylcysteine synthe- 3.6 & 0.7 4.1 f 0.8 4.3 f 0.8

(nmol/W cells)

(nmol/mg protein)

tase ( n = 6) (nmol/min/ mg protein)

(pmollminlmg protein)

(nmol/min/mg protein)

GSH-S-transferase ( n = 6) 0.38 & 0.13 0.36 f 0.11 0.40 f 0.06

GSSG reductase ( n = 6) 17.9 & 1.7 14.9 f 5.0 20.4 f 4.8

Significantly different, p < 0.05, from wild type cells.

Eluton Volume(ml1

FIG. 1. Subcellular distribution of glutathione and copper in wild type and copper-resistant cells. HAC (upper panel) and HAC, (lower panel) cell lysates were prepared in 10.0 mM HEPES buffer, pH 7.4, from -2 X lo7 cells as described previously.2 Aliquots (0.5 ml) were fractionated by gel filtration fast protein liquid chromatography (FPLC) using a Superose 12B column, equilibrated with anaerobic 10.0 mM HEPES, 0.15 M NaCl buffer, pH 7.4. Proteins were eluted at a flow rate of 1.0 ml/min. Fractions (1.0 ml) were collected and then assayed for copper (0) or glutathione (A). Copper was measured by graphite furnace atomic absorption spectroscopy from 50-al aliquots diluted to 1.0 ml with distilled-deionized water. Glutathione was determined from 0.5-ml aliquots treated with sodium metaphosphoric acid and then assayed using the GSSG-reductase system as described under “Experimental Procedures.”

one-Metallothionein Complexes-Nonradiolabeled lysates of HAC and HACw cells were fractionated by gel filtration chromatography, and the eluates were assayed for copper by atomic absorption spectroscopy (22), and for glutathione, as described above. The majority of the copper in the HAC cell lysate (64%) and all the glutathione eluted as a single peak between 76 and 85 ml (peak M, = 4,500) (Fig. 1). In the HACm cell lysate, however, copper eluted in two major peaks. The first peak between 67 and 75 ml (peak M, = 13,000), contained 30% of the total copper applied to the column. This material had been identified previously as hepatoma Cu-MT.’ The second peak, between 76 and 85 ml (peak M, = 4,500), contained 63% of the total copper. Glutathione in the HACw cell lysate eluted as a single peak between 77 and 88 ml (peak M, = 3,200), with a shoulder near 4,500, the latter coeluting with one of the copper peaks.

To identify the different forms of glutathione separated in

Role of Glutathione in Coppc 750 I .o *

0 10 20 30 40 50 Fraction Number

FIG. 2. Anion exchange FPLC of copper-metallothionein- containing fractions from HAGOO cells. Metallothionein-containing fractions isolated during gel filtration FPLC (65-75 ml) were pooled and concentrated by ultrafiltration through an Amicon YM2 membrane. The buffer in the concentrate was exchanged with anaerobic 10.0 mM Tris-HC1, pH 8.8.’ The concentrate was applied to an HR 5/5 Mono Q anion exchange FPLC column equilibrated with Tris-HC1 and then washed with 10 ml of Tris-HC1. Proteins were eluted with two linear gradients of NaCl in 10.0 mM Tris-HC1, pH 8.8 (. . .), 35 ml of 0 - 0.2 M followed by 10 ml of 0.2 + 1.0 M. 1-ml fractions were collected and 50-pl aliquots were assayed for copper by atomic absorption spectroscopy and metallothionein content by immunoblot analysis.‘

cell lysates, the elution volumes for GSH, GSSG, and GS-Cu (GSH:Cu, 51) were determined. Reduced glutathione and GSSG elute with M, values of 2200 and 3100, respectively. GS-Cu, however, elutes with an M, value of 4500. These results indicate that the 4.5-kDa material, identified in cell lysates from both wild type and resistant cells, that contained both copper and GSH is a GS-Cu complex.

The stoichiometric ratio of GSH to copper in the GS-Cu complex was determined from the integrated areas of the peaks in the chromatogram. For HAC lysates, the GSH:Cu ratio was 1:1, whereas for HACsoo lysates the ratio was 2:1, assuming that the GSH coeluting with copper (76-85 ml) was complexed to metal.

When Cu-MT-containing fractions from the gel filtration chromatography of HAC, cell lysates were pooled, concentrated, then fractionated by anion exchange chromatography, three copper-containing proteins were isolated (Fig. 2). The first two proteins, eluting at salt concentrations of 75 and 110 mM, were previously identified as hepatoma Cu-MT-I and Cu-MT-11.’ The protein eluting at the highest salt concentration, 640 mM, was also shown to be a Cu-MT, based on its copper content (Fig. 2) and its cross-reactivity with anti-rat liver MT antibody (results not shown). Glutathione also coeluted with this last protein, as determined by its activity with GSSG-reductase for which purified Cu-MT, apo-MT, and rabbit Cd/Zn-MT are inactive as substrate^.^ Based on the presence of GSH in the Cu-MT-containing fractions that eluted at high salt concentration, it is proposed that GSH forms a complex with this protein.

Effect of GSH Depletion and the Inhibition of Protein Syn- thesis on Copper Uptake-The effects of DEM, BSO, and cycloheximide on protein synthesis and GSH content in HAC and HACsm cells were examined (Table 111). Cycloheximide inhibited protein synthesis by greater than 90%, relative to untreated cells, in both cell lines. It also caused significant increases in the level of cellular GSH. Treatment with BSO or DEM reduced the GSH content in both cell lines by 85%. An increase in [35S]cysteine incorporation into proteins was also observed in BSO-treated cells. Treatment with BSO,

‘By anion exchange chromatography GSH, GSSG, and GS-Cu (GSH:Cu, 5x1) eluted at NaCl concentrations of 140, 160, and 420 mM, respectively, well resolved from the Cu-MT that eluted at 640 mM NaCI.

?r Metabolism and Toxicity 5601 TABLE 111

Effect of cycloheximide, L-buthionine-(S,R)-sulfoximine, and diethyl maleate on protein synthesis and glutathione content in

hepatoma cell lines Wild type (HAC) and resistant (HACW) cells were incubated with

5.0 pg/ml of cycloheximide, 1.0 mM BSO, or 0.2 mM DEM in medium containing 1% FBS, as described under “Experimental Procedures.” The level of glutathione equivalents (GSH and GSSG) was then measured, or cells were labeled with [35S]cysteine for 90 min. The level of protein synthesis was determined from the amount of [35S] cysteine incorporated into trichloroacetic acid-precipitable material. Untreated cells were incubated in medium containing 1% FBS without inhibitors. Values are the means of duplicate measurements.

Cell line Glutathione [36S]Cysteine and treatment content incorporation

HAC None Cycloheximide BSO DEM

HACm None

BSO Cycloheximide

DEM

nmolfmg protein

19.5 f 1.3 36.0 f 0.4 2.9 f 0.3 2.3 f 0.2

83.0 k 1.0 95.0 f 1.2 12.5 k 0.4 12.5 f 0.7

cpmfmgprotein X

11.1 f 1.7 1.1 f 0.1

18.2 f 2.0 ND“

8.2 f 0.7 0.47 f 0.08 12.6 * 1.4

ND ND, not determined.

TABLE IV Effect of inhibiting protein synthesis or depleting glutathione on

a7Cul uptake in hepatoma cell lines Protein synthesis was inhibited with cycloheximide or glutathione

depleted with BSO or DEM in wild type (HAC) and copper-resistant (HACW) cells as described under “Experimental Procedures.” Cells were then incubated with 67Cu for 30 and 90 min. The amount of 67Cu taken up by the treated cells is expressed relative to the amount of radiolabel in untreated HAC or HACGOO cells, after 30- or 90-min incubations.

Cell line Copper content and treatment 30 min“ 90 min“

HAC None

BSO Cycloheximide

DEM HACm

None

BSO Cycloheximide

DEM Incubation time.

% of control

100 f 9 100 f 7 91 k 6 82 f 5

108 f 12 77 f 8 116 f 13 68 f 3

100 f 4 100 f 8 107 f 8 57 f 12 119 f 14 103 f 10 112 f 7 94 f 9

DEM, or cycloheximide did not affect cell viability within the time period of the experiment, as determined by trypan blue viable staining (20).

Copper content after 30- and 90-min incubations was measured in cells following GSH depletion with BSO or DEM or after inhibition of protein synthesis with cycloheximide (Table IV). No significant differences in 67Cu uptake between treated and untreated cells in either cell line were observed after 30-min incubations. At longer incubation times, BSO, DEM, and cycloheximide pretreatment inhibited T u uptake into HAC cells, while “Cu uptake decreased only when protein synthesis was inhibited in cells.

Effect of Glutathione Depletion or Inhibition of Protein Synthesis on the Distribution of Copper into Subcellular Pools-The gel filtration chromatogram of cell lysates from HAC and HAC600 cells labeled for 90 min showed that eluted in the void volume (32-40 ml, peak M , = 1,000,000)

5602 Role of Glutathione in Copper Metabolism and Toxicity

and coeluted with MT (68-78 ml, peak M, = 13,000) and GSH (78-89 ml, peak M, = 4,500) (Fig. 3). The distribution of ‘j7Cu in wild type and resistant cell lysates, however, was consid- erably different. In HAC cell lysates, 20% of the total radiolabel eluted in the void volume, with 42 and 5% coeluting with MT and GSH, respectively. In HACm cell lysates, 1.4% of the 67Cu eluted in the void volume and 1.8% with GSH, while 82% coeluted with MT. Tu-Labeled superoxide dismutase was also identified in HAC600 cells lysates, eluting between 58 and 62 ml (peak M, = 32,000) and containing less than 1% of the total eluted 67Cu.

Depleting cells of GSH by treatment with BSO caused the level of copper in MT-containing fractions to decrease by 74% in HAC cells and by 48% in HACWO cells, as compared with untreated cells (Fig. 3). This decrease suggests that GSH may function in intracellular copper transport.

The inhibition of protein synthesis in HAC cells reduced the amount of observed in MT fractions by 55%, compared with untreated cells (Fig. 3). In contrast, cycloheximide had no effect on the level of copper in MT fractions of HAC6, cells. An increase in GS-‘j7Cu in the elution profiles of lysates from both cycloheximide-treated cell lines was observed. This change reflects the increase in GSH levels seen when protein synthesis was inhibited (Table 111).

Kinetics of Copper Distribution into Subcellular Pools-The time course for the subcellular distribution and complexation of 67Cu by GSH, MT, and superoxide dismutase was determined by fractionating extracts of HAC and HACm cells that had been incubated with 67Cu for different periods of time. Radiolabeled copper was observed at elution volumes corre- sponding to superoxide dismutase, MT, and GSH in both cell lines.

In HAC cell lysates, 67Cu was observed only as the GS-Cu complex at incubation times less than 1 h (Fig. 4). At incubation times greater than 1 h, radiolabeled copper began to appear in the MT-containing fractions and increased until a steady-state level was reached in which 65% of the total 67Cu

30

2.5 HAC

E 8.0 1

J 30 40 50 60 70 80 90 100

Elutlon Volume (mi)

FIG. 3. Effect of depleting glutathione and inhibiting protein synthesis on the incorporation of copper into metallothionein. Elution profiles of ‘?Cu-labeled HAC (upper panel) and HACw (lower panel) cells treated with BSO ( . . . ) or cycloheximide (- - -) were compared with untreated cells (-). Cells were treated with inhibitors and then incubated in ‘?Cu-containing medium for 90 min. Lysates were prepared and fractionated as described under “Experimental Procedures.” The concentration of radiolabeled copper in each fraction is expressed as cpm of ‘?Cu per pg of the total protein applied to the column. Individual data points were removed from the plots for clarity.

9 3.0 4 + 2.0 \ E e 1.0

0.0

Elution Volume (ml)

Q 1.0

L

0 i TIME (Hrr.)

FIG. 4. Kinetics of the distribution of “Cu into subcellular pools in HAC cells. Wild type cells were incubated with ‘?Cu for 0.5, 1, 2, 4, 8, and 12 h under standard conditions. Cell lysates were prepared and equal amounts of protein fractionated by gel filtration FPLC, as described under “Experimental Procedures.” To illustrate more clearly changes in the subcellular distribution of ‘?Cu with time, elution profiles for 0.5-, 1-, 2-, and 4-h incubations are presented in the upperpanel, and elution profiles for 2-, 4-, 8-, and 12-h incubations are shown in the middle panel. ‘?Cu-Labeled MT and GS eluted between 66 and 76 ml, and 77 and 87 ml, respectively. The amounts of ‘?Cu complexed by MT (A) and GSH (W) as a function of the incubation time are shown in the bottom panel.

applied to the column was bound to MT and 10% was complexed to GSH. These results suggest that after copper enters the cell it is quickly complexed by GSH. The data also implies that copper may be transferred from GSH to MT.

In HACs, cells, 85% of the 67Cu was complexed by MT at the earliest incubation time examined (Fig. 5). As the cellular concentration of ‘jlCu increased, the amount of bound to MT increased proportionally, remaining as 85 f 2% of the total radiolabeled copper present (Fig. 5). In HAC6, cell lysates less than 2% of the total radiolabel present appeared in the GS-Cu fractions. Radiolabeled copper was also complexed to superoxide dismutase, increasing from 1 to 4% of the total 67Cu during the course of the experiment. V u - Labeled superoxide dismutase was also observed in HAC cells but could not be quantitated due to the poor signal-to-noise ratio in that region of the chromatogram.

The rates of uptake and incorporation into Cu-MT was determined. Copper uptake reached steady-state in the

cells with a tzh of 0.27 h, while the tlh for 67Cu incorporation into MT was 0.28 h. These values suggest that once the metal is taken up by the resistant cells, it is quickly complexed to MT. Steady-state in either uptake or incorporation into MT was not reached in wild type cells during the course of the experiment.

Role of Glutathiom in Copper Metabolism and Toxicity 5603

1.25 t

1 .o VI I

2 0.75

2 0.50 E

0.25

- X

Q U

- . -

30 40 50 60 70 80 90 100

0.0

Elution Volume (ml)

2 4 6 8 10 12 TIME (Hrs)

FIG. 5. Kinetics of the distribution of “Cu into subcellular pools in HACew cells. Samples from resistant cells were prepared and fractionated as described under “Experimental Procedures” (upper panel). The effects of incubation time on the absolute amounts of 67Cu in the lysate (0) and complexed by MT (W) are presented in the lower panel. The amount of 67Cu bound to superoxide dismutase (A), which eluted between 58 and 62 ml, is also presented, at 25 times greater than the scale.

DISCUSSION

Copper Toxicity-Although several mechanisms have been proposed to describe cellular copper toxicity, it is primarily a result of the metal’s ability to catalyze the formation of toxic oxygen species through a series of redox reactions (8-10). Superoxide anion is produced during normal cellular oxygen metabolism, or it may be formed by the oxidation of GS-Cu or Cu-MT (22, 23). It can then be reduced to hydrogen peroxide, either by the action of superoxide dismutase or by spontaneous dismutation. Hydrogen peroxide has been shown to react with Cu(1) to form highly toxic hydroxyl radicals (8, 9). An elevation in the activity of hydrogen peroxide metabolizing GSH peroxidase is believed to be an indication of increased cellular hydrogen peroxide flux (36, 37). Increases in GSH peroxidase activity have been observed previously in human cells resistant to adriamycin, a drug that produces oxygen radical species including hydrogen peroxide, and in yeast exposed to copper (36-38). An increase in GSH peroxidase activity in copper-resistant hepatoma cells, as compared with HAC cells (Table 11), suggests an elevated flux of hydrogen peroxide in resistant cells. Hydrogen peroxide levels may be enhanced due to higher levels of GS-Cu, Cu-MT, or glutathione radicals in these cells.

The importance of GSH in cellular resistance to transition metal toxicity has been attributed to its involvement in the repair of metal-induced cellular damage (15). A decrease in GSH levels when cells were exposed to metal was reported to be the result of GSH utilization by GSH peroxidase during the enzymatic repair of lipid peroxides (39, 40). Copper-

resistant hepatoma cells, however, show a significant increase in total glutathione levels (Table 11). A similar increase was previously observed in rats (15) and cadmium-/zinc-resistant ovarian carcinoma cells (14) exposed to metal, and in adriamycin-resistant cells (36, 37). It has been suggested that increased GSH peroxidase activity may cause cells to develop higher steady-state levels of GSH (36, 37). This may also, in part, be responsible for elevated GSH levels in copper-resistant hepatoma cells. Copper-resistant cells accumulate glutathione through a decrease in glutathione efflux and not by an increase in GSH synthesis (Table 11). In contrast, elevated GSH levels in kidneys from rats exposed to mercury and zinc were caused by increases in GSH synthesis (15).

Copper Metabolism-When HAC cells are depleted of GSH by treatment with either BSO or DEM, 67Cu uptake is inhibited (Table IV), suggesting that GSH is involved in metal uptake (Table 111). GSH depletion in HACsw cells, however, has no effect on uptake. In these cells, sufficient GSH may remain after BSO or DEM treatment to still function in copper uptake. In lung carcinoma cells it was shown that cadmium uptake was unaffected by BSO treatment (41). In those cells, however, cadmium uptake was inhibited by DEM. Since DEM is a nonspecific thiol blocking agent, the authors (41) suggested that a thiol-containing protein, and not GSH, was directly involved in metal transport.

Inhibition of protein synthesis in wild type and resistant hepatoma cells causes a decrease of 67Cu uptake (Table IV). In contrast, it was reported that rat hepatocytes and HepG2 cells treated with cycloheximide accumulated more copper than did nontreated cells ( 5 , 42). This accumulation was attributed to a decrease in the synthesis of proteins responsible for copper transport into bile. Since rat hepatoma cell lines in culture lack the bile export system, the amount of metal accumulated is not influenced by this form of copper efflux. The results with hepatoma cell lines suggest that both protein synthesis and GSH are necessary for copper intake.

In addition to serving as a substrate for GSH peroxidase, GSH is shown to function as an intracellular copper chelator. Germann and Lerch (43) hypothesized that GSH may be a copper chelator based on the isolation of GSSG from Neuro- spora crassa cell lysates and the presence of copper in the breakthrough fraction during the purification of Cu-MT by reverse-phase high pressure liquid chromatography. In the mechanism describing cellular copper metabolism presented in Fig. 6, it is proposed that either Cu(1) is taken up by cells or Cu(I1) taken up is quickly reduced to Cu(1). The cuprous ion is then chelated by GSH forming a GS-Cu(1) complex. GS-Cu(1) and (GS)Z-Cu(I) complexes have been tentatively identified in HAC and HACm cell lysates.

The difficulty in isolating GS-Cu arises from the aerobic lability of copper-thiol complexes (22, 44) and the ability of thiol-containing reducing agents and copper-chelating buffers to dissociate these complexes. Purifications of Cu-MT from copper-loaded cells are commonly performed in Tris or phosphate buffers, both of which can chelate copper (log K values for Cu(I1) with Tris and phosphate are 4.0 and 4.7, respectively) (45). Thiol-reducing agents, such as 2-mercaptoethanol or dithiothreitol, are frequently added to the buffers (46,47).* Under these purification conditions, any GS-Cu present in cell lysates would likely dissociate. By chromatographing hepatoma cell extracts under anaerobic conditions, using nonche- lating buffers and without thiol-reducing agents, an intact GS-Cu complex is isolated. If, however, Tris buffer is substi- tuted for HEPES buffer during gel filtration chromatography, or dithiothreitol is added, the GS-Cu complex is not detected (results not shown).

5604

Corrrm

Es- 1 L- 1 cxl(I,-Ur

FIG. 6. Schematic diagram of cellular copper metabolism. The oxidation state of copper transported into cells is unknown, therefore both Cu(I1) and Cu(1) are presented. The diagram outlines a mechanism for the transfer of copper from GSH to MT. It also indicates that when copper is needed for the synthesis of other metalloenzymes, such as superoxide dismutase and cytochrome oxidase, it can be obtained from the GS-Cu(1) pool. The cellular mechanisms to remove reactive oxygen species, generated from the oxidation of the GS-Cu(1) complex or from glutathione radical (GS’), is also presented. MT, metallothionein; SOD, superoxide dismutase; GSH-Px, GSH peroxidase.

When wild type cells are incubated with T u for 30 and 60 min, the majority of the radiolabel is complexed to GSH, and 67Cu-labeled MT is not detected (Fig. 4). After longer incubation times, however, there is a decrease in the levels of GS- 67Cu, with a concomitant increase in 67Cu-labeled MT. These results suggest that copper is transferred from GSH to MT. It should be noted that the concentration of 67Cu used in these experiments is 1000 times lower than the cellular glutathione level. Therefore 67Cu is not saturating the GSH pool prior to its complexation by MT. Alternatively, the late appearance of “Cu-labeled MT could reflect the time required to accumulate MT after induction by copper. This is unlikely because the concentration of “Cu in the uptake medium is four times lower than the amount of copper normally present in HAC growth medium and would not induce de nouo MT synthesis.

Depleting HAC and HACeoo cells of GSH causes a decrease of 67Cu incorporation into MT (Fig. 3). These results further support the involvement of GSH in copper complexation by MT. The difference in the magnitude of the effect of GSH depletion on “Cu incorporation into MT between HAC and HAC, cells (74 uersus 48%, respectively) may be due to the higher concentration of GSH remaining in the resistant cells following BSO treatment (Table 111). Although BSO treatment depletes 85% of the GSH in HACs, cells, relative to untreated HAC, cells, the amount of GSH remaining is only 40% lower than that in untreated HAC cells.

It has been reported that copper incorporation into MT decreased in cycloheximide-treated rat hepatocytes and HepG2 cells, suggesting that copper is complexed concomitantly with MT synthesis (5, 42, 48). A similar result is obtained with HAC cells; however, there is only a 55% decrease in 67Cu incorporation into MT (Fig. 3), whereas protein synthesis is inhibited 90% (Table 111). Since the half-life of MT in HAC cells‘ is greater than 20 h, the decrease in 67Cu incorporation is not due to a loss of MT by protein degradation. Inhibiting protein synthesis in HACsoo cells by 95% has no effect on T!u incorporation into MT. This lack of effect

J. H. Freedman, unpublished observation.

may be due to the higher steady-state concentration of MT maintained in the resistant cells, relative to HAC cells, even when de nouo MT synthesis is inhibited. These results indicate that copper is complexed by previously synthesized MT. It is also suggested that MT can chelate copper concomitantly with protein synthesis, because cycloheximide-treated HAC cells show a significant decrease in the level of “Cu-labeled MT (Fig. 3).

Based on the appearance of GS-67C~ prior to the formation of 67Cu-labeled MT, the reduction of 67Cu incorporation into MT when GSH is depleted, and that the inhibition of protein synthesis in HAC, cells does not affect “Cu complexation by MT, it is proposed that copper is transferred from GSH to MT, possibly via a GS-Cu-MT intermediate (Fig. 6). Gluta- thione may also function to reduce Cu(I1) to Cu(I), preventing the oxidation of the cysteine residues in MT. This would allow all MT cysteine residues to be used exclusively for metal chelation, as described previously (1). Glutathione-bound copper may bind to vacant sites in the protein and/or exchange with MT-bound copper. Other models of cellular copper metabolism proposed the presence of a “free copper” pool, from which MT complexed metal (6). In the present model (Fig. 6) free copper levels are minimal, with the metal largely being complexed to thiol-containing compounds.

It is also proposed that there is an equilibrium between GS- Cu and Cu-MT (Fig. 6). The detection of low levels of GS- “Cu in HAC, cell lysates ( ~ 2 % ) may be due to the high concentration of MT, which shifts the equilibrium such that most of the metal is bound to the protein. Steady-state levels of “Cu-labeled MT and GS-“Cu are reached in HAC cells in greater than 6 h. In contrast, steady-state levels in HACw cells are reached in less than 30 min. Again, this difference may be related to the higher concentration of MT in the resistant cells.

The model for copper incorporation into MT, via a GS-Cu complex (Fig. 6), also suggests that GSH is able to remove copper from MT. It was shown by Bremner and co-workers (49) that the half-life for Cu-MT, in uiuo, is 12-17 h. In contrast, Cu-MT in uitro was resistant to degradation by

Role of Glutathione in Coppe

lysosomal enzymes, unless some of the metal was removed from the protein (50, 51). The authors suggested that an endogenous copper chelator might remove metal from Cu-MT in the cell before lysosomal enzymes can degrade the protein (52). Glutathione may function as such a chelator.

Mechanism of Resistance-Resistance to copper toxicity in hepatoma cell lines is believed to be due to elevated levels of GSH, MT, and GSH peroxidase, relative to nonresistant cells (Tables I and 11). High concentrations of GSH allow resistant cells to efficiently remove potentially toxic, free cytoplasmic copper by quickly complexing the metal. The elevated MT concentration increases the metal-binding capacity of the cell and may increase the rate of copper transfer from GS-Cu. Finally, increased GSH peroxidase activity inhibits the production of hydroxyl radical by metabolizing hydrogen peroxide. Thus, resistance to copper toxicity has developed by an amplification of the cells’ ability to both sequester copper and to prevent oxidative cellular damage.

Although resistance to copper toxicity in hepatoma cells was induced by adding a single agent, copper, to the growth medium (ZO), the levels of three different components, GSH, MT, and GSH peroxidase, are amplified. The elevated levels of these components are related to increased levels of cellular copper. The amplification of MT genes in response to metal exposure is well documented (21). An increase in cytoplasmic copper can stimulate hydrogen peroxide production which could lead to an increase in GSH peroxidase. Cellular levels of GSH increase in resistant cells because it is not removed from the cell (Table 11). It is possible that complexation of GSH as GS-Cu or GS-Cu-MT inhibits its efflux.

1.

2.

3. 4. 5.

6. 7.

8.

9.

10.

11.

12.

13.

14.

15.

REFERENCES Kagi, J . H. R., and Kojima, Y. (1987) Exper. Suppl. (Basel) 52,

Dunn, M. A., Blalock, T. L., and Cousins, R. J. (1987) Proc. Soc.

Bremner, I. (1987) Exper. Suppl. (Basel) 52,81-107 Terao, T., and Owen, C. A. (1973) Am. J. Physiol. 224, 682-686 Stockert, R. J., Grushoff, P. S., Morell, A. G., Bentley, G. E.,

O’Brien, H. A., Scheinberg, I. H., and Sternlieb, I. (1986) Hepatology 6,60-64

25-61

Exp. Biol. Med. 185, 107-119

Evens, G. W. (1973) Physiol. Rev. 53, 535-570 Nakamura, M., and Yamazaki, I. (1972) Biochim. Biophys. Acta

Samuni, A., Chevion, M., and Czapski, G. (1981) J. Biol. Chem.

Goldstein, S., and Czapski, G. (1986) J. Free Radicals Biol. &

Stacey, N. H., and Klaassen, C. D. (1981) J. Toxicol. Environ.

Kobayshi, S., Okada, T., and Kimura, M. (1985) Chem.-Biol.

Kobayshi, S., Imano, M., and Kimura, M. (1985) Chem.-Biol.

Drunam, D. M., and Palmiter, R. D. (1981) J. Biol. Chem. 256,

Andrews, P. A., Murphy, M. P., and Howell, S. B. (1987) Cancer

Fukino, H., Hirai, M., Hsueh, Y. M., Moriyasu, S., and Yamane,

267,249-257

256,12632-12635

Med. 2,3-11

Health 7, 139-147

Interact. 55, 347-356

Interact. 52, 319-334

5712-5716

Chemother. Pharmacol. 19, 149-154

Y. (1986) J. Toxicol. Environ. Health 19, 75-89

‘r Metabolism and Toxicity 5605

16. Kang, Y.-T., and Enger, M. D. (1988) Toxicology 48,93-101 17. Sinehal. R. K.. Anderson. M. E., and Meister, A. (1987) FASEB

J: 1,220-223

Chem.-Biol. Interact. 65.1-14

. .

18. Ochi, T., Otsuka, F., Takahashi, K., and Ohsawa, M. (1988)

19. Chance, B., Sies, H., and Bovers, A. (1979) Physiol. Rev. 59,527-

20. Freedman, J. H., Weiner, R. J., and Peisach, J. (1986) J. Biol.

21. Palmiter, R. D. (1987) Exper. Suppl. (Basel) 52,63-80 22. Saez, G., Thornalley, P. J., Hill, H. A. O., Hems, R., and Bannis-

23. Minkel, D. T., Poulsen, K., Weilgus, S., Shaw, C. F., and Petering,

24. Anderson, M. E. (1985) Methods Enzymol. 113,548-555 25. McCord, J. M., and Fridovich, I. (1969) J. Biol. Chem. 244,6049-

6055 26. McCord, J. M., andFridovich, I. (1969) J. Biol. Chem. 244,6049-

6055 27. Luck, H. (1963) in Methods of Enzymatic Analysis (Bergmeyer,

H. U., ed) 2nd Ed., pp. 885-888, Academic Press, New York 28. Lawrence, R. A., and Burk, R. F. (1976) Biochem. Biophys. Res.

Commun. 71,952-958 29. Bergmeyer, H. U., Gawehn, K., and Grossl, M. (1974) in Methods

of Enzymatic Analysis (Bergmeyer, H. U., ed) Vol. 1, pp. 425- 522, Academic Press, New York

30. Habig, W. H., Pabst, M. J., and Jakoby, W. B. (1974) J. Biol. Chem. 249, 7130-7139

31. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193, 265-275

32. Paniker, N. V., andBeutler, E. (1972) J. Lab. Clin. Med. 80,481- 487

33. Dethmers, J. K., and Meister, A. (1981) Proc. Natl. Acad. Sci.

34. Reed, D. J., Brodie, A. E., and Meredith, M. J. (1983) in Functions of Glutathione: Ph.ysiologica1, Toxicological, and Clinical Aspects

605

Chem. 261,11840-11848

ter, J. V. (1982) Biochim. Biophys. Acta 719,24-31

D. H. (1980) Biochem. J. 191, 475-485

U. S. A. 78,7492-7496

35.

36.

37.

38.

39.

40. 41. 42.

43. 44.

45.

46.

47. 48. 49.

50. 51.

52.

(Larsson, A., Orrenius, S . , HolmgrenYA., Mammervik, B.; eds) pp. 39-49, Raven Press, New York

Agostini, C., and Muci, F. (1979) Experientiu (Basel) 35, 518- 519

Sinha, B. K., Katki, A. G., Batist, G., Cowan, K. H., and Myers, C . E. (1987) Biochemistry 26,3776-3781

Kramer, R. A., Zakher, J., and Kim, G. (1988) Science 241,694- 697

Galiazzo, F., Schiesser, A., and Rotilio, G. (1987) Biochem. Bio- phys. Res. Commun. 147,1200-1205

Stacey, N. H., and Klaassen, C. D. (1981) Toxicol. Appl. Phar- macol. 58, 211-220

Muller, L. (1986) Toxicol. Lett. (Amst.) 30, 259-265 Kang, Y.-J., and Enger, M. D. (1988) FASEB J. 2, A1820 Weiner, A. L., and Cousins, R. J. (1980) Biochirn. Biophys. Acta

Germann, U. A., and Lerch, K. (1987) Biochem. J. 245,479-484 Karlin, K. D., and Zubieta, J. (1979) Znorg. Perspect. Biol. Med.

Hogeldt, E., and Perrin, D. D. (1979) Stability Constants of Metal- Ion Complexes, Parts A and B, Pergamon Press, New York

Riordan, J. R., and Richards, V. (1980) J. Biol. Chem. 265,5380- 5383

Bremner, I., and Young, B. W. (1976) Biochem. J. 157,517-520 Bremner, I., and Davies, N. T. (1976) Br. J. Nutr. 36,101-112 Bremner, I., Hoekstra, W. G., Davies, N. T., and Young, B. W.

Mehra, R. K., and Bremner, I. (1985) Biochem. J. 227,903-908 Weiser, U., Mutter, W., and Hartmann, H.-J. (1986) FEBS Lett.

Bremner, I. (1987) J. Nutr. 117, 19-29

629,113-125

2,127-149

(1978) Biochem. J. 174,883-892

197,258-262

the journal of biological 264, no. 10, of inc. in u.s.a ... · role of glutathione in copper...

Documents