and Species-Dependent Inactivation of Glutathione Transferase Zeta

Mar 3, 2000 - 601 Elmwood Avenue, Box 711, Rochester, New York 14642, and Molecular Genetics Group,. Division of Molecular Medicine, John Curtin ...
0 downloads 0 Views 59KB Size
Download PDF
Chem. Res. Toxicol. 2000, 13, 231-236

231

Polymorphism- and Species-Dependent Inactivation of Glutathione Transferase Zeta by Dichloroacetate Huey-Fen Tzeng,†,‡ Anneke C. Blackburn,§,| Philip G. Board,§ and M. W. Anders*,† Department of Pharmacology and Physiology, University of Rochester Medical Center, 601 Elmwood Avenue, Box 711, Rochester, New York 14642, and Molecular Genetics Group, Division of Molecular Medicine, John Curtin School of Medical Research, Australian National University, Canberra ACT 2601, Australia Received October 28, 1999

Glutathione transferase zeta catalyzes the glutathione-dependent oxidation or conjugation of a range of R-haloacids. Repeated administration of dichloroacetate to human subjects increases its plasma elimination half-life, and the activity of glutathione transferase zeta is decreased in rats given dichloroacetate. The objective of the studies presented here was to investigate the kinetics and mechanism of the dichloroacetate-induced decrease in glutathione transferase zeta activity. The rate constants (kinact) for the dichloroacetate-dependent inactivation of glutathione transferase zeta in liver cytosol are in the following order: rat > mouse > human; the half-maximal inhibitory concentration (Kinact) of DCA did not differ among the species that were studied. In contrast to dichloroacetate, chlorofluoroacetate produced much less inactivation of mouse liver glutathione transferase zeta activity. Moreover, the addition of N-acetyl-L-cysteine or potassium cyanide did not fully block the dichloroacetate-induced inactivation of glutathione transferase zeta. The kinact values for the dichloroacetate-induced inactivation of four polymorphic variants of recombinant human glutathione transferase zeta (hGSTZ1-1) were in the following order: variant 1a-1a < 1b-1b ≈ 1c-1c ≈ 1d-1d. The dichloroacetate-induced inactivation of hGSTZ1-1 was irreversible. The binding of radioactivity from [1-14C]dichloroacetate and from [35S]glutathione to recombinant hGSTZ1c-1c was demonstrated, indicating covalent modification of the protein. These results show that dichloroacetate is a mechanism-based inactivator of glutathione transferase zeta and is biotransformed to electrophilic metabolites that covalently modify and, thereby, inactivate the enzyme.

Introduction Glutathione transferase zeta (GSTZ)1 catalyzes the glutathione-dependent oxidation of DCA to glyoxylate (13). The cDNA of hGSTZ1 has been cloned and expressed (1). hGSTZ1-1 is identical with maleylacetoacetate isomerase (4), which catalyzes the isomerization of maleylacetoacetate to fumarylacetoacetate in the tyrosine degradation pathway. Recently, several polymorphic variants of hGSTZ1-1 have been identified (5-7). DCA is a common contaminant in drinking water (8, 9). DCA is carcinogenic in rats and mice (10-13), is weakly mutagenic (14), and alters DNA methylation (15). DCA is also used for the clinical management of congenital lactic acidosis and homozygous familial hypercholesterolemia in humans (16). Repeated administration of DCA to human subjects increases its plasma elimina-

tion half-life (17, 18). Also, the rate of biotransformation of DCA to glyoxylate is decreased in liver cytosol from rats given DCA (19). Recent studies also show that the activity of GSTZ1-1 is decreased in a time- and dosedependent manner in rats given DCA and other fluorinelacking dihaloacetates and that the decrease in GSTZ1-1 activity is accompanied by a decrease in the level of immunoreactive GSTZ protein (20). The objective of the studies presented here was to investigate the mechanism of the glutathione-dependent, DCA-induced inactivation of GSTZ1-1. The rate constants for inactivation of GSTZ1-1 by DCA were determined in human, rat, and mouse liver cytosol and with polymorphic variants of purified recombinant hGSTZ1-1. Finally, the covalent modification of recombinant hGSTZ1c-1c by [1-14C]DCA and [35S]glutathione was demonstrated.

Experimental Procedures * To whom correspondence should be addressed: Department of Pharmacology and Physiology, University of Rochester Medical Center, 601 Elmwood Ave., Box 711, Rochester, NY 14642. Telephone: (716) 275-1681. Fax: (716) 244-9283. E-mail: [email protected]. † University of Rochester Medical Center. ‡ Present address: Department of Zoology, National Taiwan University, 1, Section 4, Roosevelt Road, Taipei 106, Taiwan, Republic of China. § Australian National University. | Present address: Department of Veterinary and Animal Sciences, University of Massachusetts, Amherst, MA 01003-6410. 1 Abbreviations: DCA, dichloroacetate; GSTZ, glutathione transferase zeta; hGSTZ1-1, human glutathione transferase zeta.

Materials. Dichloroacetic acid (>99% pure) was obtained from Aldrich Chemical Co. (Milwaukee, WI). Bromofluoroacetic acid (97% pure) and chlorofluoroacetic acid (99% pure) were prepared as described previously (3). [14C]DCA (57 mCi/mmol) was purchased from American Radiolabeled Chemicals (St. Louis, MO). [35S]Glutathione (286.86 Ci/mmol) was purchased from NEN Life Science Products (Boston, MA). Glutathione and ampicillin were obtained from Sigma Chemical Co. (St. Louis, MO). Isopropyl thio-β-D-galactoside, kanamycin, and LB medium were purchased from Gibco-BRL (Bethesda, MD). 14C-

10.1021/tx990175q CCC: $19.00 © 2000 American Chemical Society Published on Web 03/03/2000

232

Chem. Res. Toxicol., Vol. 13, No. 4, 2000

labeled protein molecular weight markers and Hyperfilm MP were obtained from Amersham Pharmacia Biotech (Piscataway, NJ). The expression plasmids for recombinant hGSTZ1 polymorphic variants were prepared as previously described (7). Liver cytosolic fractions were prepared from male, Fischer 344 rats (200-225 g, Charles River Laboratories, Wilmington, MA); male, B6C3F1 mice (4 weeks old, Charles River Laboratories); and male, human liver (International Institute for the Advancement of Medicine, Scranton, PA), as described previously (3). Protein assay reagents and SDS-PAGE molecular weight standards (low range) were purchased from Bio-Rad (Hercules, CA). Expression of Recombinant hGSTZ1-1 Variants. Escherichia coli M15[pREP4] cells (Qiagen Inc., Valencia, CA) were transformed with plasmids and plated on LB agarose plates containing 100 µg of ampicillin/mL and 25 µg of kanamycin/ mL. A single, isolated colony was selected and used to inoculate 4 mL of LB medium that contained 100 µg of ampicillin/mL and 25 µg of kanamycin/mL. The starter culture was shaken at 37 °C until the absorbance at 600 nm reached 0.2 AU. The starter culture was used to inoculate 320 mL of LB medium that contained 100 µg of ampicillin/mL and 25 µg of kanamycin/mL and was incubated until the absorbance at 600 nm reached 0.8-1 AU, at which time 50 µL of 1 M isopropyl thio-β-Dgalactoside was added. The culture was shaken overnight at 37 °C, and the cells were harvested by centrifugation at 6000g for 15 min at 4 °C and were stored at -20 °C until they were used. Purification of Recombinant hGSTZ1-1 Variants. The recombinant variants of hGSTZ contain an N-terminal His‚Tag sequence, which allows purification with a nickel affinity column. The cells from 160 mL of culture were suspended in 8 mL of binding buffer [50 mM sodium phosphate buffer (pH 7.4), 0.3 M sodium chloride, and 10 mM imidazole], and the suspension was sonicated with three 10-s bursts at 80% power with the micro tip of a Branson model 450 sonifier (Danbury, CT). The lysate was centrifuged at 39000g for 20 min. The supernatant was filtered through a 0.45 µm filter (Costar, Cambridge, MA), and the filtrate was loaded onto a polypropylene column (1 cm × 6.5 cm) containing 1.5 mL of His‚Bind resin (Novagen, Madison, WI) that had been equilibrated with binding buffer. The column was eluted with 15 mL of binding buffer and then with 10 mL of wash buffer [50 mM sodium phosphate (pH 7.4), 0.3 M sodium chloride, and 60 mM imidazole]. The protein was eluted with 10 mL of elution buffer [50 mM sodium phosphate (pH 7.4), 0.3 M sodium chloride, and 1 M imidazole], and 0.5 mL fractions were collected. The concentration of protein in each fraction was determined with the Bio-Rad protein assay solution with bovine serum albumin as the standard. The activity of the protein in each fraction with DCA as the substrate was measured as published previously (3). Fractions containing high protein concentrations and high activities were combined, dialyzed overnight against 20 mM potassium phosphate, 50 mM sodium chloride, 0.5 mM EDTA, 1.5 mM DTT, and 10% glycerol (pH 7.0) and stored at -20 °C. Kinetics of GSTZ1-1 Inactivation. Liver cytosol (0.67 mg/ mL) was incubated with 30-500 µM DCA in the presence of 5 mM glutathione at 37 °C for 4-55 min in 0.1 M potassium phosphate (pH 7.4). Proteins were recovered by three concentration-dilution cycles with Centricon-10 concentrators (YM-10, 10 kDa molecular mass cutoff; Millipore, Bedford, MA). The biotransformation of DCA to glyoxylate with the recovered proteins from liver cytosol was assessed as described previously (2). The ratio of measured activity (A) to A0, the rate in incubation mixtures that contained glutathione but no DCA, was fit to the equation ln(A/A0) ) -kobst, where t is the incubation time. The inactivation rate constant (kinact) and the half-maximal inhibitory concentration (Kinact) were determined from the equation (21)

[DCA] kobs ) kinact [DCA] + Kinact

Tzeng et al. The half-life was determined from the equation t1/2 ) ln 2/kobs. To determine whether exogenous nucleophiles would block the DCA-induced inactivation of GSTZ1-1, mouse liver cytosol was incubated with 500 µM DCA and 5 mM glutathione in the presence of 10 mM N-acetyl-L-cysteine or 10 mM potassium cyanide at 37 °C for 4-30 min in 0.1 M potassium phosphate (pH 7.4). Protein recovery, the biotransformation of DCA to glyoxylate, and data analysis were carried out as described above. Mouse liver cytosol was incubated with 500 µM chlorofluoroacetate in the presence of 5 mM glutathione at 37 °C for 10-30 min in 0.1 M potassium phosphate (pH 7.4). Protein recovery, the biotransformation of DCA to glyoxylate, and data analysis were carried out as described above. Purified recombinant hGSTZ1-1 variants (1 µg/mL) were incubated with 0.5 mM DCA and 1 mM glutathione at 37 °C for 3.5-40 min in 0.1 M potassium phosphate (pH 7.4) with bovine serum albumin (100 µg/mL). Proteins were recovered from incubation mixtures as described above, and the remaining recombinant hGSTZ1-1 activity was determined by assessing the biotransformation of 0.5 mM chlorofluoroacetate to glyoxylate in the presence of 1 mM glutathione and 1.1 µg/mL recombinant hGSTZ1-1 for 20 min at 37 °C; glyoxylate formation was assessed as described previously (2). Data analysis was carried out as described above. Covalent Modification of Recombinant hGSTZ1c-1c by [1-14C]DCA and [35S]Glutathione. To determine if the DCAinduced inactivation of GSTZ1-1 was associated with covalent modification of the enzyme, the DCA-dependent binding of 35S from [35S]glutathione to recombinant hGSTZ1c-1c and the glutathione-dependent binding of 14C from [1-14C]DCA to recombinant hGSTZ1c-1c were studied. To assess the glutathionedependent binding of 14C from [1-14C]DCA to the enzyme, recombinant hGSTZ1c-1c (1 µg/mL) was incubated with 25 µM [14C]DCA and 475 µM DCA in the absence or presence of 1 mM glutathione in 5.4 mL of 0.1 M potassium phosphate buffer (pH 7.4) at 37 °C for 4 h. Additional enzyme (1 µg/mL) was added to the reaction mixture every 20 min, and additional DCA (1.4 µM [14C]DCA and 26.8 µM DCA) was added after incubation for 1, 2, and 3 h. The protein was recovered as described above, and the recovered protein (∼2.5 µg) was analyzed by SDS-PAGE on a 12% Tris-HCl Ready gel (Bio-Rad) according to the manufacturer’s instructions. After electrophoresis, the gel was stained with Coomassie Blue R-250, destained, and soaked in 1 M sodium salicylate aqueous solution for 30 min. The gel was dried (Bio-Rad model 583 gel dryer) with a gradient cycle at 80 °C for 2 h and exposed to Hyperfilm MP at -80 °C for 7 days. To assess the DCA-dependent binding of 35S from [35S]glutathione to recombinant hGSTZ1c-1c, the enzyme was incubated with 11.9 nM [35S]glutathione and 500 µM glutathione in the absence or presence of 1 mM DCA in 5.4 mL of 0.1 M potassium phosphate buffer (pH 7.4) at 37 °C for 4 h. Additional enzyme (1 µg/mL) was added to the reaction mixture every 20 min. The protein was recovered and analyzed as described above.

Results Kinetics of GSTZ1-1 Inactivation. Incubation of liver cytosol or purified recombinant hGSTZ1-1 with DCA alone, with glutathione alone, or with glutathione and glyoxylate, the product of GSTZ1-1-catalyzed biotransformation of DCA, did not result in a significant decrease in the activity of GSTZ1-1 (data not shown). In contrast, a time- and concentration-dependent decrease in GSTZ1-1 activity was observed when human liver cytosol was incubated with both glutathione and DCA (Figure 1), indicating that DCA inactivated GSTZ1-1. When mouse liver cytosol was incubated with both DCA and glutathione, a nonlinear decrease in GSTZ1-1 activity was observed (Figure 2).

Inactivation of Glutathione Transferase Zeta

Chem. Res. Toxicol., Vol. 13, No. 4, 2000 233 Table 1. Kinetic Constants for the Glutathione-Dependent, DCA-Induced Inactivation of GSTZ1-1 in Mouse, Rat, and Human Liver Cytosola

Figure 1. Time- and concentration-dependent inactivation of GSTZ1-1 activity in human liver cytosol by DCA. Human liver cytosol was incubated with 5 mM glutathione and 30 (b), 50 (9), 100 (2), or 250 µM DCA (0) in 0.1 M potassium phosphate buffer (pH 7.4) at 37 °C for 10, 25, 40, and 55 min, and activity was measured as described in Experimental Procedures. The slope of a plot of ln(A/A0) vs t gives kobs, which was plotted against the DCA concentration (inset).

Figure 2. Time-dependent inactivation of GSTZ1-1 activity in mouse liver cytosol by DCA and chlorofluoroacetate. Mouse liver cytosol was incubated with 5 mM glutathione and 0.5 mM DCA (b) or 0.5 mM chlorofluoroacetate (0) in 0.1 M potassium phosphate buffer (pH 7.4) at 37 °C for 10, 20, 30, and 40 min (DCA only). For chlorofluoroacetate, the slope of a plot of ln(A/ A0) vs t gives kobs. The half-life for inactivation (t1/2) was calculated from the equation t1/2 ) ln 2/kobs.

Although previous studies indicated that both bromofluoroacetate and chlorofluoroacetate are substrates for purified rat liver GSTZ1-1 (3), the studies presented here showed that bromofluoroacetate undergoes a rapid, glutathione-dependent, and nonenzymatic conversion to glyoxylate (data not shown). To determine whether chlorofluoroacetate inactivates GSTZ1-1, mouse liver cytosol was incubated with 0.5 mM chlorofluoroacetate in the presence of 5 mM glutathione at 37 °C. After incubation for 30 min, DCA produced 88% inactivation, whereas chlorofluoroacetate produced 13% inactivation (Figure 2). The kinetic constants for the glutathione-dependent, DCA-induced inactivation of GSTZ1-1 in mouse, rat, and human liver cytosol were determined (Table 1). The relative rates of DCA-induced inactivation of GSTZ1-1 in liver cytosol were in the following order: rat > mouse > human. The half-life for the DCA-induced inactivation of human GSTZ1-1 was 4-fold longer than that of rat and mouse. The Kinact did not differ among the rat, mouse, or human GSTZ1-1. The mechanism of inactivation of GSTZ1-1 by DCA may involve the formation of an intermediate electrophilic species that reacts with and, thereby, inactivates the enzyme. If this mechanism applies, high concentrations of exogenous nucleophiles should not block the DCA-induced inactivation of GSTZ1-1. Hence, GSTZ1-1 was incubated with 10 mM potassium cyanide or 10 mM N-acetyl-L-cysteine. N-Acetyl-L-cysteine (10 mM) reduced the level of DCA-induced inactivation of GSTZ1-1 by

species

kinact (min-1)

half-life (min)

Kinact (µM)

mouse rat human

0.105 ( 0.005b 0.127 ( 0.002 0.032 ( 0.002

6.61 5.44 22

41 ( 7c 43 ( 2 38 ( 5

a Mouse, rat, or human liver cytosol (0.67 mg/mL) was incubated with 30-500 µM DCA and 5 mM glutathione at 37 °C for 4-55 min in 0.1 M potassium phosphate buffer (pH 7.4). Protein recoveries and GSTZ1-1 activities were as described in Experimental Procedures. kinact and Kinact were determined by plotting kobs vs the DCA concentration, and the half-life was calculated from the equation t1/2 ) ln 2/kinact, as described in Experimental Procedures. Data are shown as means ( SD (n ) 3). b All means are significantly different from one another (p < 0.05, ANOVA). c Means do not differ significantly from one another (p > 0.05, ANOVA).

Table 2. Rate Constants for the Glutathione-Dependent, DCA-Induced Inactivation of Recombinant hGSTZ1-1 Polymorphic Variantsa variant

kinact (min-1)

half-life (min)

hGSTZ1a-1a hGSTZ1b-1b hGSTZ1c-1c hGSTZ1d-1d

(3.0 ( 0.1) × 10-2 (7.2 ( 0.2) × 10-2 (6.9 ( 0.2) × 10-2 (7.3 ( 0.2) × 10-2

23 ( 1b 9.6 ( 0.3 10.1 ( 0.5 9.5 ( 0.3

a Polymorphic variants of purified recombinant hGSTZ1-1 (1 µg/ mL) were incubated with 0.5 mM DCA, 1 mM glutathione, and bovine serum albumin (100 µg/mL) at 37 °C for 3.5-40 min in 0.1 M potassium phosphate buffer (pH 7.4). Protein recoveries, GSTZ1-1 activities, and data analyses were as described in Experimental Procedures. Data are shown as means ( SD (n ) 3). b Significantly different from variants 1b-1b, 1c-1c, and 1d-1d (p < 0.05, ANOVA).

about 40%, whereas potassium cyanide afforded no protection (data not shown). DCA-Induced Inactivation of Polymorphic Variants of Recombinant hGSTZ1-1. The cDNA of hGSTZ1 has been cloned (1), and several polymorphic variants have been identified and expressed in E. coli (5, 6). The biotransformation of DCA to glyoxylate catalyzed by recombinant hGSTZ1a-1a is 3.6-fold faster than that catalyzed by recombinant hGSTZ1b-1b, -1c-1c, and -1d-1d (7). One possible explanation for this difference between variant 1a-1a and the others is that the DCAinduced inactivation of recombinant hGSTZ1a-1a was much slower than that of the others. To test this hypothesis, the four recombinant hGSTZ1-1 polymorphic variants were incubated with 0.5 mM DCA and 1 mM glutathione at 37 °C, and the rate constants for the DCAinduced inactivation were determined (Table 2). Recombinant hGSTZ1a-1a showed a significantly longer halflife for inactivation than the other polymorphic variants. Also, the activity of recombinant hGSTZ1a-1a could not be restored by dialysis of the inactivated protein against 0.1 M potassium phosphate buffer (pH 7.4) overnight at 4 °C (data not shown). Covalent Modification of Recombinant hGSTZ1c1c by [1-14C]DCA and [35S]Glutathione. To test the hypothesis that the DCA-induced inactivation of recombinant hGSTZ1c-1c was associated with covalent modification, the enzyme was incubated with [14C]DCA in the absence or presence of glutathione or with [35S]glutathione in the absence or presence of DCA followed by SDS-PAGE and fluorography (Figure 3). As shown in Figure 3, the inactivated recombinant hGSTZ1c-1c had a molecular mass similar to the unmodified enzyme. The inactivated recombinant hGSTZ1c-1c was labeled with

234

Chem. Res. Toxicol., Vol. 13, No. 4, 2000

Figure 3. SDS-PAGE and fluorography of covalently modified recombinant hGSTZ1c-1c. Recombinant hGSTZ1c-1c was incubated with [1-14C]DCA alone (lane 1), with [1-14C]DCA and glutathione (lane 2), with [35S]glutathione alone (lane 3), or with [35S]glutathione and DCA (lane 4). Proteins were recovered and analyzed by SDS-PAGE. The gel was stained with Coomassie Blue R-250 (A) or visualized by fluorography (B), as described in Experimental Procedures.

Tzeng et al.

14C

Figure 4. Proposed mechanism for the GSTZ1-1-catalyzed biotransformation of DCA to glyoxylate and for the DCAdependent inactivation of GSTZ1-1: 1, DCA; 2, S-(R-chlorocarboxymethyl)glutathione; 3, carbonium-sulfonium intermediate; 4, glyoxylate; 5, covalently modified and inactivated GSTZ1-1; GSH, glutathione.

The objective of the studies presented here was to investigate the kinetics and mechanism of the DCAinduced decrease in GSTZ1-1 activity. The rate constants for inactivation (kinact) for the DCA-induced inactivation of GSTZ measured in liver cytosol were in the following order: rat > mouse . human (Table 1). In contrast, no significant species differences in the half-maximal inhibitory concentration (Kinact) for the DCA-induced inactivation of GSTZ1-1 were observed (Table 1). The observation that DCA inactivates GSTZ1-1 affords an explanation for the prolonged half-life of DCA in humans seen after repeated administration of DCA (17, 18). Similarly, the data also explain the decreased level of GSTZ1-1catalyzed oxidation of DCA to glyoxylate in liver cytosol from rats given DCA (19). In contrast to the present finding that DCA reduced GSTZ1-1 activity in human liver cytosol, Cornett et al. (22) recently reported that DCA failed to reduce GSTZ1-1 activity in human liver cytosol, although DCA-induced inactivation was seen in rat liver cytosol. The basis for these contrasting findings is not apparent, but the data presented herein indicate that all known polymorphic variants of hGSTZ1-1 are inactivated by DCA (Table 2). Compared with DCA, chlorofluoroacetate produced much less glutathione-dependent inactivation of mouse liver GSTZ1-1; the half-life for inactivation of GSTZ1-1 chlorofluoroacetate was 235 min (Figure 2), whereas the half-life for DCA-induced inactivation was 6.6 min (Table 1). Bromofluoroacetate, but not chlorofluoroacetate, undergoes rapid glutathione-dependent conversion to glyoxylate in potassium phosphate buffer (pH 7.4) at 37 °C, which precludes its use as a substrate in assaying GSTZ1-1 activity. In contrast, chlorofluoroacetate undergoes little nonenzymatic reaction and produces little inactivation of GSTZ1-1 and is, therefore, a more suitable substrate than DCA for assaying GSTZ1-1 activity. The DCA-induced inactivation of GSTZ1-1 activity in mouse liver cytosol was not linear with time (Figure 2). This time-dependent non-pseudo-first-order inactivation of GSTZ1-1 activity in mouse liver cytosol by DCA may be explained by the biotransformation of DCA to glyoxylate, which reduces the concentration of DCA and, thereby, slows the rate of inactivation (21). Also, incubation of GSTZ1-1 with DCA or with glyoxylate alone did not inactivate the enzyme.

Experiments on the DCA-induced inactivation of recombinant hGSTZ1-1 polymorphic variants showed that recombinant hGSTZ1a-1a was inactivated more slowly than the other variants. Recombinant hGSTZ1a-1a differs from variants 1b-1b, 1c-1c, and 1d-1d by replacement of arginine at amino acid residue 42 with glycine. Comparison of the sequence homology of hGSTZ1 with other cytosolic GSTs indicates that amino acid residue 42 is located peripheral to the active site and not believed to be directly involved in catalysis but may, for example, affect substrate binding (7). The DCA-induced inactivation of GSTZ1-1 was associated with the covalent modification of the protein (Figure 3). Labeling with 14C from [1-14C]DCA was glutathionedependent, and that with 35S from [35S]glutathione was DCA-dependent. Also, activity could not be restored by overnight dialysis, indicating that the DCA-induced inactivation of GSTZ1-1 is irreversible. In addition, N-acetyl-L-cysteine, but not potassium cyanide, afforded some protection against the DCA-induced inactivation of GSTZ1-1 in mouse liver cytosol. This may indicate that nucleophilic amino acid residues, such as cysteine or lysine, may be modified by the DCA-induced inactivation of the enzyme. These data indicate that DCA is a mechanism-based inactivator of GSTZ1-1. Previous studies on the GSTZ11-catalyzed biotransformation of R-haloacids, including dihaloacetates, [2H]dihaloacetates, and 2-halopropionates, have defined a reaction mechanism that suffices to explain the mechanism of the DCA-induced inactivation of GSTZ1-1 and the covalent modification of the enzyme by DCA and glutathione (3, 23) (Figure 4). In this mechanism, a GSTZ1-1-catalyzed SN2 reaction results in the displacement of chloride by glutathione to give S-(R-chlorocarboxymethyl)glutathione 2. R-Chlorosulfide 2 may lose chloride to give carbonium-sulfonium intermediate 3. Intermediate 3 may undergo SN1-type hydrolysis to give glyoxylate 4 and glutathione (24) or may react with a nucleophilic residue in GSTZ1-1 to give covalently modified and inactivated enzyme 5. This mechanism accounts for the observation that DCA both is converted to glyoxylate and inactivates the enzyme and that the covalently modified GSTZ1-1 contains radioactivity from both [1-14C]DCA and [35S]glutathione. Presuming that intermediate 3 reacts with a sulfhydryl group in GSTZ1-1, dithioacetal 5 would be formed. The stability of the protein-bound dithioacetal 5 was not explored in the studies presented here, although dithioacetals may be reduced by reaction with thiols (25). GSTZ1-1, which is identical with maleylacetoacetate isomerase (EC 5.2.1.2) (4), catalyzes both the biotrans-

from [1-14C]DCA in the presence of glutathione and with 35S from [35S]glutathione in the presence of DCA. Neither 14C-labeled nor 35S-labeled protein was observed when the enzyme was incubated with [1-14C]DCA alone or with [35S]glutathione alone.

Discussion

Inactivation of Glutathione Transferase Zeta

formation of R-haloacids and the isomerization of maleylacetoacetate to fumarylacetoacetate. Hereditary tyrosinemia type 1 is an autosomal recessive inherited disease characterized by a deficiency of fumarylacetoacetate hydrolase (EC 3.7.1.2), which catalyzes the hydrolysis of fumarylacetoacetate to fumarate and acetoacetate (26). Hereditary tyrosinemia type 1 is also characterized by the development of hepatocellular carcinomas in a significant fraction of affected individuals (26). Fumarylacetoacetate, which accumulates in hereditary tyrosinemia type 1, is mutagenic in Chinese hamster V79 cells, but maleylacetoacetate and succinylacetone are not mutagenic (27, 28). If the accumulation of fumarylacetoacetate is associated with the development of hepatocellular carcinomas in hereditary tyrosinemia type 1, agents that prevent the formation and accumulation of fumarylacetoacetate may be of therapeutic benefit. 2-(2Nitro-4-trifluoromethylbenzoyl)-1,3-cyclohexanedione, which inhibits 4-hydroxyphenylpyruvate dioxygenase and thereby prevents the formation of fumarylacetoacetate, has been used experimentally and with favorable results in the treatment of hereditary tyrosinemia type 1 (29); it also abolishes neonatal lethality in fumarylacetoacetate hydrolase null mice (30, 31). The data presented herein show that DCA efficiently inactivates liver hGSTZ1-1. DCA is used for the clinical management of congenital lactic acidosis and is tolerated well by treated patients (16, 32). Although, as noted above, DCA is carcinogenic in rodents, more than 40 patient years of cumulative DCA exposure have been recorded with no evidence that DCA causes neoplasia in humans (33). Hence, DCA may reduce the formation of fumarylacetoacetate and may, therefore, merit consideration for trial in the clinical management of hereditary tyrosinemia type 1. The finding that DCA inactivates GSTZ1-1 has implications for estimating the effects of acute therapeutic and chronic human exposure to DCA and for interpreting the results of chronic toxicity studies in laboratory animals. As indicated above, DCA is hepatocarcinogenic in rats and mice (10-13). DCA-induced inactivation of GSTZ1-1 may increase the concentration of DCA in liver, which may have implications for the tumorigenicity of DCA. The kinetics of inactivation reported herein could be incorporated into physiologically based pharmacokinetic models to simulate more accurately the internal dose of DCA in chronic toxicity experiments in laboratory animals. Finally, the observed differences among the polymorphic variants of hGSTZ1-1 may indicate that subpopulations may differ in their susceptibility to DCA-induced toxicity. Further studies on the mechanisms of the DCAinduced modification of recombinant hGSTZ1-1 and of the GSTZ1-1-catalyzed biotransformation of R-haloacids are warranted as are studies on the possible use of DCA in the management of hereditary tyrosinemia type 1.

Acknowledgment. This work was supported by National Institutes of Environmental Health Sciences Grant ES03127 to M.W.A. We thank Mr. Wayne Anderson for preparation of the rat liver cytosol, Ms. Daria Krenitsky for technical assistance, and Ms. Gail Johnson for her assistance in preparing the manuscript.

References (1) Board, P. G., Baker, R. T., Chelvanayagam, G., and Jermiin, L. S. (1997) Zeta, a novel class of glutathione transferases in a range of species from plants to humans. Biochem. J. 328, 929-935.

Chem. Res. Toxicol., Vol. 13, No. 4, 2000 235 (2) Tong, Z., Board, P. G., and Anders, M. W. (1998) Glutathione transferase Zeta catalyzes the oxygenation of the carcinogen dichloroacetic acid to glyoxylic acid. Biochem. J. 331, 371-374. (3) Tong, Z., Board, P. G., and Anders, M. W. (1998) Glutathione transferase Zeta-catalyzed biotransformation of dichloroacetic acid and other R-haloacids. Chem. Res. Toxicol. 11, 1332-1338. (4) Ferna´ndez-Can˜o´n, J. M., and Pen˜alva, M. A. (1998) Characterization of a fungal maleylacetoacetate isomerase gene and identification of its human homologue. J. Biol. Chem. 273, 329-337. (5) Board, P., Blackburn, A., Jermiin, L. S., and Chelvanayagam, G. (1998) Polymorphism of phase II enzymes: identification of new enzymes and polymorphic variants by database analysis. Toxicol. Lett. 102-103, 149-154. (6) Blackburn, A. C., Woollatt, E., Sutherland, G. R., and Board, P. G. (1998) Characterization and chromosome location of the gene GSTZ1 encoding the human Zeta class glutathione transferase and maleylacetoacetate isomerase. Cytogenet. Cell Genet. 83, 109-114. (7) Blackburn, A. C., Tzeng, H.-F., Anders, M. W., and Board, P. G. (1999) Discovery of a functional polymorphism in human glutathione transferase zeta by expressed sequence tag database analysis. Pharmacogenetics (in press). (8) Krasner, S. W., McGuire, M. J., Jacangelo, J. G., Patania, N. L., Reagan, K. M., and Aieta, E. M. (1989) The occurrence of disinfection by-products in U.S. drinking water. J. Am. Water Works Assoc. 81, 41-53. (9) Weisel, C. P., Kim, H., Haltmeier, P., and Klotz, J. B. (1998) Exposure estimates to disinfection by-products of chlorinated drinking water. Environ. Health Perspect. 107, 103-110. (10) Herren-Freund, S. L., Pereira, M. A., Khoury, M. D., and Olson, G. (1987) The carcinogenicity of trichloroethylene and its metabolites, trichloroacetic acid and dichloroacetic acid, in mouse liver. Toxicol. Appl. Pharmacol. 90, 183-189. (11) Bull, R. J., Sanchez, I. M., Nelson, M. A., Larson, J. L., and Lansing, A. J. (1990) Liver tumor induction in B6C3F1 mice by dichloroacetate and trichloroacetate. Toxicology 63, 341-359. (12) DeAngelo, A. B., Daniel, F. B., Stober, J. A., and Olson, G. R. (1991) The carcinogenicity of dichloroacetic acid in the male B6C3F1 mouse. Fundam. Appl. Toxicol. 16, 337-347. (13) DeAngelo, A. B., Daniel, F. B., Most, B. M., and Olson, G. R. (1996) The carcinogenicity of dichloroacetic acid in the male Fischer 344 rat. Toxicology 114, 207-221. (14) Harrington-Brock, K., Doerr, C. L., and Moore, M. M. (1998) Mutagenicity of three disinfection by-products: di- and trichloroacetic acid and chloral hydrate in L5178Y/TK+/- -3.7.2C mouse lymphoma cells. Mutat. Res. 413, 265-276. (15) Tao, L., Kramer, P. M., Ge, R., and Pereira, M. A. (1998) Effect of dichloroacetic acid and trichloroacetic acid on DNA methylation in liver and tumors of female B6C3F1 mice. Toxicol. Sci. 43, 139-144. (16) Stacpoole, P. W., Henderson, G. N., Yan, Z. M., and James, M. O. (1998) Clinical pharmacology and toxicology of dichloroacetate. Environ. Health Perspect. 106 (Suppl. 4), 989-994. (17) Curry, S. H., Chu, P.-I., Baumgartner, T. G., and Stackpoole, P. W. (1985) Plasma concentrations and metabolic effects of intravenous sodium dichloroacetate. Clin. Pharmacol. Ther. 37, 89-93. (18) Curry, S. H., Lorenz, A., Chu, P.-I., Limacher, M., and Stacpoole, P. W. (1991) Disposition and pharmacodynamics of dichloroacetate (DCA) and oxalate following oral DCA doses. Biopharm. Drug Dispos. 12, 375-390. (19) James, M. O., Cornett, R., Yan, Z., Henderson, G. N., and Stacpoole, P. W. (1997) Glutathione-dependent conversion to glyoxylate, a major metabolite of dichloroacetate biotransformation in hepatic cytosol from humans and rats, is reduced in dichloroacetate-treated rats. Drug Metab. Dispos. 25, 1223-1227. (20) Anderson, W. B., Board, P. G., Gargano, B., and Anders, M. W. (1999) Inactivation of glutathione transferase zeta by dichloroacetic acid and other fluorine-lacking R-haloalkanoic acids. Chem. Res. Toxicol. 12, 1144-1149. (21) Silverman, R. B. (1995) Mechanism-based enzyme inactivators. Methods Enzymol. 249, 240-283. (22) Cornett, R., James, M. O., Henderson, G. N., Cheung, J., Shroads, A. L., and Stacpoole, P. W. (1999) Inhibition of glutathione S-transferase ζ and tyrosine metabolism by dichloroacetate: a potential unifying mechanism for its altered biotransformation and toxicity. Biochem. Biophys. Res. Commun. 262, 752-756. (23) Wempe, M. F., Anderson, W. B., Tzeng, H.-F., Board, P. G., and Anders, M. W. (1999) Glutathione transferase zeta-catalyzed biotransformation of deuterated dihaloacetic acids. Biochem. Biophys. Res. Commun. 261, 779-783.

236

Chem. Res. Toxicol., Vol. 13, No. 4, 2000

(24) Brodwell, F. G., Cooper, G. D., and Morita, H. (1957) The hydrolysis of chloromethyl aryl sulfides. J. Am. Chem. Soc. 79, 376-378. (25) Peach, M. E. (1974) Thiols as nucleophiles. In The Chemistry of the Thiol Group (Patai, S., Ed.) pp 722-784, Wiley, New York. (26) Mitchell, G. A., Lambert, M., and Tanquay, R. M. (1995) Hypertyrosinemia. In The Metabolic and Molecular Basis of Inherited Disease (Scriver, C. R., Beaudet, A. L., Sly, W., and Vallee, D., Eds.) pp 1077-1106, McGraw-Hill, New York. (27) Tanguay, R. M., Jorquera, R., Poudrier, J., and St-Louis, M. (1996) Tyrosine and its catabolites: from disease to cancer. Acta Biochim. Pol. 43, 209-216. (28) Jorquera, R., and Tanguay, R. M. (1997) The mutagenicity of the tyrosine metabolite, fumarylacetoacetate, is enhanced by glutathione depletion. Biochem. Biophys. Res. Commun. 232, 42-48. (29) Lindstedt, S., Holme, E., Lock, E. A., Hjalmarson, O., and Strandvik, B. (1992) Treatment of hereditary tyrosinaemia type

Tzeng et al.

(30)

(31)

(32)

(33)

I by inhibition of 4-hydroxyphenylpyruvate dioxygenase. Lancet 340, 813-817. Grompe, M., Lindstedt, S., Al-Dhalimy, M., Kennaway, N. G., Papaconstantinou, J., Torres-Ramos, C. A., Ou, C.-N., and Finegold, M. (1995) Pharmacological correction of neonatal lethal hepatic dysfunction in a murine model of hereditary tyrosinaemia type I. Nat. Genet. 10, 453-460. Grompe, M., Overturf, K., al-Dhalimy, M., and Finegold, M. (1998) Therapeutic trials in the murine model of hereditary tyrosinaemia type I: a progress report. J. Inherited Metab. Dis. 21, 518-531. Stacpoole, P. W., Barnes, C. L., Hurbanis, M. D., Cannon, S. L., and Kerr, D. S. (1997) Treatment of congenital lactic acidosis with dichloroacetate. Arch. Dis. Child. 77, 535-541. Stacpoole, P. W., Henderson, G. N., Yan, Y., Cornett, R., and James, M. O. (1998) Pharmacokinetics, metabolism, and toxicology of dichloroacetate. Drug Metab. Rev. 30, 499-539.

TX990175Q