EXPERIMENTAL
PARASITOLOGY
75,47-55 (1992)
Schistosoma mansoni: Single-Step of Glutathione
Purification and Characterization S-Transferase lsoenzyme 4
KATHLEEN A. O'LEARY,'KAREN M. HATHAWAY,ANDJAMES W. TRACY* Departments of Comparative Biosciences and Pharmacology, and the Environmental Toxicology Center, University of Wisconsin, 2015 Linden Drive West, Madison, Wisconsin 53704-1102, U.S.A. O’LEARY, K. A., HATHAWAY, K. M., AND TRACY, J. W. 1992. Schistosoma mnnsoni: Single-step purification and characterization of glutathione S-transferase isoenzyme 4. Experimental Parasitology 75, 47-55. A soluble glutathione S-transferase isoenzyme, designated SmGST-4 was purified to apparent homogeneity in a single step from the cytosol of adult Schistosoma mansoni by selective elution of the enzyme from a glutathione-agarose affinity column using glutathione disulfide. SmGST-4, which comprised about 5% of the bound glutathione S-transferase activity, could be distinguished from the previously characterized glutathione S-transferase isoenzyme family (SmGST-l/2/3) by its unique chromatographic behavior, lower subunit M, (26,000), differences in substrate specificity and inhibitor sensitivity, and a lack of reactivity with antiserum to SmGST-3. The purified isoenzyme catalyzed the conjugation of several model xenobiotics including I-chloro-2,4dinitrobenzene, ethacrynic acid, and transd-phenyl-3-buten-2-one. Like the SmGST-l/2/3 isoenzyme family, SmGST-4 failed to catalyze the conjugation of a model epoxide substrate, 1,2-epoxy-3+nitrophenoxy)propane. Because glutatbione S-transferases from other organisms play a role in protecting cells against the toxic products of lipid peroxidation, SmCST-4 and the members of the SmGST-l/2/3 isoenzyme family were tested for their capacity to reduce cumene hydroperoxide and to catalyze the conjugation of 4-hydroxyalk2-enals. Although all four isoenzymes catalyzed both reactions, the specific activity of SmGST-1, SmGST-2, and SmGST-3 toward cumene hydroperoxide was at least lo-fold greater than that of SmGST-4. In contrast, the latter more effectively conjugated a homologous series of 4-hydroxyalk-2-enal isomers. For example, the specific activity of SmGST-4 toward 4-hydroxynon-2-enal(10.8 pmol . min-’ . mg protein-‘), the isomer formed under physiological conditions, was about S-fold greater than that of SmGST-1 and 1Zfold greater than that of the major isoenzyme SmGST-3. These results are consistent with a multifunctional role for schistosome glutathione S-transferases in protecting the parasite against both xenobiotics and toxic endogenous products of lipid peroxidation. Q 1992 Academic PESS. IIK. INDEX DESCRIPTORS AND ABBREVIATIONS: Trematode; Schistosomn mnnsoni; Glutathione S-transferase (GST; EC 2.5.1.18); Isoenzyme; Affinity chromatography; Substrate specificity; 4-Hydroxyalk-2-enal; Cumene hydroperoxide; H,O,; Glutathione (GSH); Affinity-purified Schistosoma mansoni Glutathione S-transferase (SmGST-AP); Glutathione disulfide (GSSG); Glutathione peroxidase; 1-chloro-2,Cdinitrobenzene (CDNB); Sodium dodecyl sulfate-polyacrylamide-gel electrophoresis (SDS-PAGE).
INTRODUCTION Glutathione S-transferases (EC 2.5.1.18) catalyze the conjugation of both endogenous and xenobiotic electrophiles with the major cellular nucleophile GSH (Mannervik and Danielson 1988). GSH S-trans-
ferases are widely distributed in nature, being found in both vertebrates and invertebrates, including schistosomes (O’Leary and Tracy 1988; Taylor et al. 1988; Henkle et al. 1990; Trottein et al. 1990; Wright et al. 1991). Most organisms contain multiple GSH S-transferase isoenzymes that are characterized by distinct, but often overlapping, substrate specificities that account for their capacity to metabolize structurally diverse chemicals.
’ Present address: McArdle Laboratory for Cancer Research, University of Wisconsin, Madison, WI 53706. * To whom correspondence should be addressed. 47
0014-4894192$5.00 Copyright 0 1992 by Academic Press, Inc. AU rights of reproduction in any form reserved.
48
O'LEARY,HATHAWAY,ANDTRACY
Adult Schistosoma mansoni contain at least five forms of GSH S-transferase. We previously described the purification and characterization of three of them, designated SmGST-1, SmGST-2, and SmGST-3 (O’Leary and Tracy 1988). These isoenzymes, which compose 2-4% of the soluble male worm protein, were purified by a combination of GSH affinity chromatography and high-pressure liquid chromatography. Although SmGST- 1, SmGST-2, and SmGST-3 have the same apparent subunit M, (28,500) and are immunologically crossreactive, they can be distinguished from one another by differences in isoelectric point, catalytic activity, and inhibitor sensitivity (O’Leary and Tracy 1988). They have since been found to compose an isoenzyme family consisting of two homodimers (SmGST-1 and SmGST-3) and the corresponding heterodimer SmGST-2 (O’Leary and Tracy, under review). Another isoenzyme that fails to bind to GSHagarose (O’Leary and Tracy 1988) preferentially catalyzes the conjugation of epoxide substrates and the detoxication of dichlorvos, the active form of the antischistosomal drug metrifonate (O’Leary and Tracy 1991). Besides the three major isoenzymes, affinity-purified SmGST preparations contained a minor form with an apparent subunit M, of 26,000. That isoenzyme was not, however, recovered after high-pressure liquid chromatography (O’Leary and Tracy 1988). We now report that this GSH S-transferase, designated SmGST-4, can be purified in a single step by selective elution from GSH-agarose with GSSG. SmGST-4 can be distinguished from the SmGST-l/2/3 isoenzyme family both by its unique chromatographic behavior and by differences in catalytic specificity, inhibitor sensitivity, apparent subunit M,, and a lack of immunological cross-reactivity. In addition to detoxicating drugs and other foreign chemicals, GSH S-transferases play an important role in protecting
cells against the toxic products of lipid peroxidation. For example, they act as selenium-independent GSH peroxidases catalyzing the reduction of fatty acid and other organic hydroperoxides (Ketterer et al. 1988). Unlike selenium-dependent GSH peroxidase, however, GSH S-transferases do not reduce HzOz. GSH S-transferases also catalyze the conjugation of 4-hydroxyalk-Zenals, chemically reactive aldehyde products of lipid peroxidation that are implicated in the cytotoxicity of reactive oxygen (Esterbauer et al. 1988). Because adult schistosomeslive in an aerobic environment and derive at least one-third of their energy from aerobic catabolism of glucose (Van Oordt et al. 1985), they can be expected to undergo reactive oxygenmediated lipid peroxidation. We therefore tested SmGST-4 and the SmGST-l/2/3 family isoenzymes for their capacity to reduce a model organic hydroperoxide and to conjugate a homologous series of 4-hydroxyalk-Zenal isomers. MATERIALSANDMETHODS Parasites. Female mice (CF, strain, 18-22 g, Sasco, Madison, WI) were injected subcutaneously with 200 freshly shed cercariae of a Puerto Rican strain of S. mansoni. After 42-49 days, parasites were recovered by portal venous perfusion with ice-cold RPM1 medium 1640 (GIBCO, Grand Island, NY) containing 25 mM 2-hydroxyethylpiperazine-N’-2-ethane sulfonate (Research Organics, Cleveland, OH) buffer, pH 7.4 (Siegel and Tracy 1989). Parasites were washed several times with cold medium to remove traces of contaminating host tissue. A sample of adult Schistosoma japonicum, provided by Dr. John Bruce (University of Lowell, Lowell, MA) was stored frozen below - 20°C until use. Enzyme assays. GSH S-transferase activity was routinely measured during enzyme purillcation with I-chloro-2,Cdinitrobenzene (CDNB, 99 + %, Aldrich Chemical, Milwaukee, WI) as described by Habig and Jakoby (1981). The rate of product formation was monitored at 340 nm (c = 9.6 m&-i . cm-‘) using a Gilford Response spectrophotometer (Ciba-Coming Diagnostics, Oberlin, OH) maintained at 25°C. One unit of SmGST activity is defined as the amount of enzyme that catalyzes the formation of 1 kmol of S-(2,4dinitrophenyl)glutathione per minute under the conditions of the assay. Specific activity is expressed
c~~rwxERIzAw0~
OF S.mansoni~~~
as units per milligram of protein. Activity of individual SmGST isoenzymes toward 1,2-chloro-4nitrobenzene (Pfaltz and Bauer, Stamford, CT), ethacrynic acid (Sigma Chemical, St. Louis, MO), 1,2-epoxy-3-(pnitrophenoxy)propane (Sigma), and rrans-Cphenyl3buten-2-one (Aldrich) was similarly assayed at the appropriate wavelengths (Habig and Jakoby 1981). A homologous series of 4-hydroxyalk-2-enals (C,, C,, C,, and Cl1 isomers) was generously provided by Professor Hermann Esterbauer (Universitit Graz, Austria). The rate of 4-hydroxyalk-2-enal conjugation with GSH was measured spectrophotometrically at 224 nm (Alin et al. 1985) using E = 13.75 rn&-’ ’ cm-’ (Esterbauer and Weger 1967). GSH peroxidase activity of SmGST’s toward H,O, and cumene hydroperoxide (Sigma) was determined by coupling peroxide-dependent GSSG formation to GSSG reductase (Type III, Sigma) catalyzed NADPH oxidation (Paglia and Valentine 1967). Appropriate controls were included to ensure that NADPH oxidation was both protein and peroxide dependent. One unit of GSH peroxidase activity is defined as the amount of enzyme that catalyzes the oxidation of 1 nmol of NADPH per minute under the conditions of the assay. Specific activity is expressed as units per milligram protein. Commercially prepared GSH peroxidase from bovine erythrocytes (Sigma) was used to validate the enzyme assay. Several prototypic inhibitors, including hematin (Sigma), bromosulfophthalein (Sigma), indocyanine green (Pfaltz and Bauer), triethyltin bromide, and triphenyltin chloride (Alpha Products, Danvers, MA) were tested for their ability to inhibit CDNB conjugation. The inhibition constant I,, is defined as the concentration of inhibitor required to produce 50% inhibition under the assay conditions defined for the experiment and was interpolated from graphs of fractional enzyme velocity as a function of the log,, of the inhibitor concentration (Yalcin et al. 1983).Protein concentrations were estimated by a dye-binding method (Bradford 1976)using crystalline bovine serum albumin (Miles Laboratories, Elkhart, IN) as the reference protein. Enzyme purification. All operations were carried out at OXC, unless otherwise noted. Adult schistosomes (1 to 2 g wet wt) were homogenized in 4 vol of 10 mM potassium phosphate buffer, pH 7.0, containing 1 mM Na,EDTA. The homogenate was centrifuged for 60 min at 47,800g. The supematant was decanted and centrifuged for 90 min at lO5,OOOg.Both pellets were discarded. The resulting cytosolic fraction was applied to a 1.5 x 6.0-cm column of S-linked GSHagarose (Sigma), equilibrated with 10 mM potassium phosphate buffer, pH 7.0. Unbound material was washed through the column with 5 bed vol of equilibration buffer. The column was next eluted with 5 bed vol of 10 mM potassium phosphate buffer, pH 7.4,
S-TRANSFERME ISOENZYME 4
49
containing 5 mM GSSG (Sigma). The remaining SmGST’s were eluted with 50 mM Tris-HCl buffer, pH 9.6, containing 5 mM GSH. SmGST-1, SmGST-2, and SmGST-3 were subsequently resolved by highpressure liquid chromatography (O’Leary and Tracy 1988). Column fractions containing enzyme activity were pooled and concentrated in a stirred ultrafiltration cell fitted with a YM-10 membrane (Amicon, Danvers, MA). Puritied SmGST’s were stored at - 20°C in a pH 7.0 buffer containing 10 nu&f potassium phosphate, 30% (v/v) glycerol, 1 mM Na,EDTA and 1 mM dithiothreitol (Research Organics). Electrophoresis. SDS-PAGE was carried out in 0.75-mm thick 12.5% vertical slab gels according to the method of Laemmli (1970). Protein samples were mixed with an equal volume of 60 mM Tris-HCl, 5% (v/v) 2-mercaptoethanol, 3% (w/v) SDS, 20% (v/v) glycerol, and 0.1% (w/v) bromophenol blue tracking dye, pH 6.8, and immediately boiled for 3 min. A mixture reference proteins (soybean trypsin inhibitor, M, 20,000; bovine erythrocyte carbonic anhydrase, M, 29,000; rabbit muscle aldolase, M, 40,000; and bovine serum albumin, M, 67,008; Sigma) was prepared in parallel. Gels were stained for protein with 0.25% (w/ v) Coomassie blue R-250 in 25% (v/v) 2-propanol/lO% (v/v) acetic acid. After overnight staining, gels were destained in 10% (v/v) acetic acid. Immunoblotting. After SDS-PAGE separation, gels were electroblotted onto nitrocellulose (Schleicher and Schuell, Keene, NH) for 3 hr at 250 mA. Filters were blocked overnight in a solution of 1.5% (w/v) Carnation Nonfat Dry Milk in 10 mM Tris-HCl buffer, pH 8.0, containing 0.15 M NaCl and 0.05% (v/v) Tween 20 (TBST). Duplicate filters were incubated for 2 hr in a 1:2O,UOO (v/v) dilution in TBST of primary rabbit antiserum elicited (O’Leary and Tracy 1988) against either affinity purified SmGST or an electrophoretically pure sample of SmGST-3 (anti-SmGST3). Filters were washed three times with TBST, followed by a 30-min incubation in a 1:7500(v/v) dilution of goat anti-rabbit IgG-alkaline phosphatase conjugate (Promega, Madison, WI). After washing three times with TBST, filter were exposed to a solution of 1.65 mg . ml-’ nitro blue tetrazolium (Promega) and 0.83 . ml-’ 5-bromo-4-chloro-3-indolyl phosphate zomega) in 100 mM Tris-HCl buffer, pH 9.5, containing 0.15 M NaCl and 5 mM MgCl,. The reaction was stopped after color formation (typically 1 to 3 min) by transferring the falters to 20 mM Tris-HCI buffer, pH 8.0, containing 5 mM Na,EDTA.
RESULTS
When the cytosolic fraction of adult S. munsoni was applied to thiol-linked GSHagarose, about 90% of the CDNBconjugating activity was retained on the
50
O’LEARY,
HATHAWAY,
column. We observed that about 10%of the bound GSH S-transferase activity could be selectively eluted from the column with a neutral phosphate buffer containing 5 mM GSSG, resulting in a 45 to 50-fold increase in specific enzyme activity (Table I), Analysis of the GSSG-eluted fraction by SDSPAGE showed a single polypeptide with an apparent M, = 26,000 (Fig. lA, lane 4). Isoelectric focusing in polyacrylamide under nondenaturing conditions also revealed a single protein species with an isoelectric point of approximately 7.4 (not illustrated). This GSH S-transferase isoenzyme was designated SmGST-4. The remainder of the bound GSH S-transferase activity was eluted with an alkaline buffer containing 5 mM GSH. The latter fraction, which could be resolved by chromatofocusing into the previously characterized major isoenzymes SmGST-1, SmGST-2, and SmGST-3 (O’Leary and Tracy 1988), still contained a small amount (l-2%) of M, 26,000polypeptide that could not be eluted from GSHagarose with GSSG, regardless of the volume of elution buffer used. Comparisons of substrate specificity demonstrated that SmGST-4 could be distinguished from the previously purified SmGST isoenzymes (Table II). For example, SmGST-4 was about 1.5-fold more active toward the prototypic aryl halide substrate CDNB, but 1.4- to 2.2-fold less active toward another aryl halide 1,Zdichloro-6 nitrobenzene compared with the SmGST-II
AND
TRACY
2/3 isoenzyme family. SmGST-4 was 2- to 12-fold more active toward ethacrynic acid than the other isoenzymes, whereas its specific activity toward the a&unsaturated carbonyl substrate truns-4-phenyl3-buten2-one was intermediate between that of the other isoenzymes. Like the SmGST-l/2/3 isoenzyme family, SmGST-4 did not catalyze the conjugation of a model epoxide substrate 1,2-epoxy-3-(p-nitrophenyl)propane. In addition to catalyzing the conjugation of xenobiotics, many GSH S-transferases display selenium-independentGSH peroxidase activity and thus catalyze the reduction of organic hydroperoxides, but not H,Oz (Ketterer et al. 1988).Consistent with this idea, SmGST’s failed to catalyze the reduction of H,O, (Table II). All four isoenzymes did, however, catalyze the reduction of cumene hydroperoxide, a model organic hydroperoxide. Although SmGST-1, SmGST-2, and SmGST-3 showed very similar specific activities toward cumene hydroperoxide, SmGST-4 was 13- to 15-fold less active toward that substrate (Table II). Because GSH S-transferases seem to play a key role in protecting cells against the toxicity of chemically reactive aldehydie lipid peroxidation products (Esterbauer et al. 1988), the four SmGST isoenzymes were tested for their ability to catalyze the conjugation of a homologous series of 4-hydroxyalk-Zenal isomers (Table III). Members of the SmGST-l/2/3 family showed
TABLE I Purification of SmGST-4 from Adult S. mansoni
Step Cytosol GSH-agarose GSSG elution GSH elution
Protein concentration (mg . ml-‘) 3.02 0.05 0.61
Total protein b-m) 27 0.04 0.76
Specific” activity (units . mg-I) 0.95 48.4 20.1
Total activity (units)
Fold purification
Yield cm
19.4
-
100
1.9 15.3
50 21
(1Determined with CDNB as the xenobiotic substrate. One unit = 1 pmol product formed . min-‘.
10 79
51
CHA~A~-~ER~~AT~ON OF S. mansoni GSHS-TRAN~FERASE I~OEN~YME4 Mr
x 10-3 67
-
1 1 2 3 4 5 1 2 34 5 2 3 4 5 FIG. 1. SDS-PAGE analysis and immunoblotting of schistosome GSH S-transferases. Triplicate protein samples were separated on 12.5% acrylamide gels under reducing conditions in the presence of SDS. (A) One gel was stained for protein with Coomassie blue R-250. The other gels were electroblotted onto nitrocellulose and probed with either (B) a 1:20,000 (v/v) dilution of rabbit antiSmGST-AP, or (C) a 1:20,000(v/v) dilution of rabbit anti-SmGST-3. Binding of anti-SmGST antibodies was detected using a goat anti-rabbit IgG-alkaline phosphatase conjugate and nitro blue tetrazoliuml 5-bromo-4~chloro-3-indolyl phosphate substrate solution. Lane 1, Unfractionated S. mansoni cytosol (40 kg); lane 2, SmGST-AP (4 kg); lane 3, Purified SmGST-3 (2 pg); lane 4, SmGST-4 (1.3 wg) selectively eluted from GSH-agarose with GSSG; lane 5, unfractionated S. japonicum cytosol(30 kg).
similar specific activities toward all isomers tested. In each case SmGST-2 showed activity intermediate between that of SmGST-1 and SmGST-3. In contrast, SmGST-4 was from 4- to 25fold more active toward each of the 4-hydroxyalk-2enals tested than were members of the SmGST-l/2/3 isoenzyme family. In particular the activity of SmGST-4 toward 4-hydroxynon-Zenal, the isomer that is formed
under physiological conditions, was nearly S-fold greater than SmGST-1 and 1Zfold more active than the major isoenzyme SmGST-3. Several inhibitors that have been used to characterize other SmGST’s were tested for their ability to inhibit the CDNBconjugating activity of SmGST-4. The Iso values, interpolated from graphs of activity remaining as a function of inhibitor concen-
TABLE II Substrate Specificity of Purified S. mansoni GSH S-Transferases Specific activity” Substrate I-Chloro-2,4-dinitrobenzene 1,2-Dichloro-4-nitrobenzene Ethacrynic acid trans-4-Phenyl-3-buten-2-one 1,2-Epoxy-3-(p-nitrophenoxy)propane W, Cumene hydroperoxide
SmGST- 1
SmGST-2
SmGST-3
SmGST-4
306 136 3.6’
32’ 186 1.96 0.32’ 06 0 0.45
26” 116 0.536 0.16b 06 0 0.39
44 8 6.5 0.29 0 0 0.03
F” 0 0.39
LISpecific activity is expressed as micromole of product formed . min- ’ . mg protein-‘. ’ For comparison, data taken from O’Leary and Tracy (1988).
52
O’LEARY,
HATHAWAY,
Conjugation of CHydroxyalk-2-enals
AND TRACY
TABLE III by Purified 5. munsoni GSH S-Transferases Specific activity”
Substrate
SmGST-I
SmGST-2
SmGST-3
SmGST-4
4-Hydroxypent-Zenal (C,) 4-Hydroxyhex-2-enal (Cd CHydroxynon-Zenal (C,) 4-Hydroxyundec-2-enal (C, r)
2.7 1.2 2.4 2.0
1.4 0.5 1.9 1.7
0.8 0.5 0.9 1.4
14.7 12.2 10.8 11.9
a Specific activity is expressed as micromoles of product formed * min- ’ . mg protein- ‘.
tration (Yalcin ef al. 1983) are shown in Table IV. Triphenyltin chloride was the most potent inhibitor of SmGST-4 with an I,, value about lOO-fold lower than the other inhibitors tested. Another organotin compound, triethyltin bromide, was about IOfold less potent than the triphenyl compound. Compared with SmGST-l/2/3 isoenzyme family, SmGST-4 was 50- to 500-fold less sensitive to inhibition by either hematin or bromosulfophthalein and about 6-fold less sensitive to inhibition by indocyanine green. Rabbit antisera elicited against SmGST preparations (O’Leary and Tracy 1988) were used to compare the immunological relationship of SmGST-4 to other SmGST’s (Fig. 1). Triplicate protein samples were separated by SDS-PAGE. One gel was stained for protein (Fig. lA), while the other two were immunoblotted with either rabbit anti-affinity purified SmGST (Fig. 1B) or rabbit anti-SmGST-3 (Fig. 1C). Antiaffinity purified SmGST recognized two
species (M, 28,500 and M, 26,000) in both the unfractionated cytosol and a sample of affinity purified SmGST, purified by conventional GSH elution (Fig. lB, lanes I and 2). This antiserum recognized a single species (M, 28,500) in purified SmGST-3 (Fig. lB, lane 3) and a single species (M, 26,000) in purified SmGST-4 prepared by selective GSSG elution (Fig. lB, lane 4). Although anti-SmGST-3 recognized the SmGST-l/2/3 family in unfractionated S. mansoni cytosol and in affinity-purified SmGST, it did not recognize purified SmGST-4 (Fig. lC, lane 4). Because an immunological relationship has been noted previously between SmGST’s and GSH S-transferases from S. juponicum (Tiu et al. 1988), a sample of unfractionated S. juponicum cytosol was also analyzed for reactivity against the two antisera. Anti-affinity purified SmGST reacted with two species in S. juponicum (Fig. lB, lane 5) that had the same apparent M, (28,500 and 26,000, respectively) as the SmGST’s. Anti-SmGST-3 recognized the
TABLE IV Inhibition of Purified S. mansoni GSH S-Transferases Iso values (uma Inhibitor
SmGST-lb
SmGST-2b
SmGST-3b
SmGST-4
Bromosulfophthalein Hematin Indocyanine green Triphenyltin chloride Triethyltin bromide
0.05 0.02 2.7 0.18 11.0
0.05 0.09 2.7 0.24 5.3
0.03 0.20 3.3 0.36 8.6
12 11 20 0.16 15
LIMeasured with CDNB as the xenobiotic substrate at 25°C. b For comparison, data taken from O’Leary and Tracy (1988).
CHARACTERIZATION 0~ S. mansoniOs~
S-T~~N~FE~~SE I~OENZYME 4
53
M, 28,500 species, but not the M, 26,000 more striking differences can be seen when comparing catalytic activity toward cumene hydroperoxide and 4-hydroxyalk2-enals (Tables II and III). Whereas SmGST-4 was substantially less active toward the organic hydroperoxide, it was clearly more effective at catalyzing the conDISCUSSION jugation of the series of reactive c.u,BAt least five GSH S-transferases are ex- unsaturated aldehydes. Inhibitor studies pressed in adult S. mansoni. Earlier we re- also revealed differences between SmGST-4 and the SmGST-l/2/3 family (Taported the purification and characterization of three major GSH S-transferases, desig- ble IV). Although the I,, values for organnated SmGST- 1, SmGST-2, and SmGST-3 otin compounds were comparable for all (O’Leary and Tracy 1988). These isoen- four isoenzymes, SmGST-4 was 50- to 500zymes, which compose 24% of the total fold less sensitive to inhibition by hematin cytosolic protein, represent an isoenzyme or bromosulfophthalein than the major family consisting of two homodimers isoenzymes. In contrast to SmGST-1, (SmGST-1 and SmGST-3) and the corre- SmGST-2, and SmGST-3 that have the sponding heterodimer SmGST-2 (O’Leary same electrophoretic mobility by SDSand Tracy, under review). As reported PAGE (apparent subunit M, 28,500) and that share major immunological determihere, another isoenzyme, designated SmGST-4, can be selectively eluted from nants (O’Leary and Tracy 1988), SmGST-4 GSH-agarose with GSSG. Based on pro- displays a faster electrophoretic mobility tein recovery, SmGST-4 compose about (M, 26,000) and does not react with antise5% of the total affinity-purified SmGST rum to SmGST-3 (Fig. lC, lane 4). Finally, fraction. At least one additional form of the SmGST-4 can be distinguished from the enzyme that fails to bind to the affinity ma- fifth isoenzyme both by its affinity for trix has been detected, but not isolated due GSH-agarose and by its lack of catalytic to its instability (O’Leary and Tracy 1988). activity toward 1,2-epoxy-3-(p-nitroThat isoenzyme preferentially catalyzes the phenoxy)propane. The molecular basis for selective elution conjugation of the model epoxide 1,2epoxy-3-(p-nitrophenoxy)propane and the of SmGST-4 from GSH-agarose by GSSG detoxication of dichlorvos, the active form is unknown. However, Meyer et al. (1989) of the antischistosomal drug metrifonate found that certain rat GSH S-transferases (isoenzymes l-l and 8-8) could likewise be (O’Leary and Tracy 1991). selectively eluted from GSH-agarose with In addition to its unique chromatographic behavior on GSH-agarose, SmGST-4 can GSSG. Interestingly, rat isoenzyme 8-8 and be readily distinguished from the SmGST- SmGST-4 seem to share some catalytic l/2/3 isoenzyme family based on differ- similarities. Like SmGST-4, isoenzyme 8-8 ences in the pattern of substrate specificity, is quite active toward ethacrynic acid and apparent subunit M,, and a lack of immu- of all rat GSH S-transferases it is most efnological cross-reactivity. For example, fective catalyst of 4-hydroxyalk-Zenal conSmGST-4 shows the highest specific activ- jugation (Jensson et al. 1986). Additional ity toward the prototypic aryl halide sub- studies will be necessary, however, to destrate CDNB, but the lowest activity to- termine whether SmGST-4 and isoenzyme ward another aryl halide 1,2-dichloro-48-8 have other properties in common. nitrobenzene (Table II). It is the isoenzyme The reason for the abundance and commost active toward ethacrynic acid. Even plexity of GSH S-transferases in adult S. species in S. japonicum (Fig. lC, lane 5). No reaction was observed with either antiserum at the dilution employed against proteins that failed to bind to the GSH affinity column (data not shown).
54
O'LEARY,HATHAWAY,AND
mansoni is not fully understood. In mammals, these enzymes are multifunctional, catalyzing the metabolism of both endogenous and xenobiotic compounds and serving as intracellular binding/transport proteins for ligands such as bilirubin and hematin (Mannervik and Danielson 1988). Clearly, SmGST’s are able to catalyze the detoxication of xenobiotics, including at least one clinically used drug. Results of experiments reported here demonstrate that SmGST’s also catalyze the detoxication of endogenous products of lipid peroxidation. Because of their location within the host’s portal venous system proximal to the liver, S. mansoni are exposed to a wide variety of potentially toxic chemicals. The bloodstream is an aerobic environment and adult worms do carry out oxidative metabolism (Van Oordt et al. 1985).Both of these conditions result in exposure of the organism to reactive forms of oxygen, such as 02-* and H202, leading to peroxidation of membrane lipids (Esterbauer et al. 1988). Furthermore, various host effector cells generate reactive oxygen species resulting in peroxidative damage and cytotoxicity. Adult schistosomes are generally resistant to such immune attack and they do contain antioxidant enzymes. Simurda et al. (1988) cloned an S. mansoni Cu/Zn superoxide dismutase that catalyzes the metabolism of 02-* to H,Oz. Mkoji et al. (1988) reported the activities of several antioxidant enzymes including selenium-dependentglutathione peroxidase that catalyzes the reduction of H*O,. As shown here, SmGST’s catalyze the reduction of organic hydroperoxides, but not H,Oz, consistent with their role as selenium-independentglutathione peroxidases. Furthermore, SmGST’s, particularly SmGST-4, catalyze the conjugation with GSH of 4-hydroxyalk2-enals, one class of reactive lipid peroxidation byproducts that are implicated in the cytotoxicity of reactive oxygen (Esterbauer et al. 1988).When taken together these data support the idea that schistosome GSH S-transferases are multifunctional en-
TRACY
zymes. It seemslikely that additional catalytic and binding functions of these schistosome proteins await discovery. Identical cDNA’s encoding S. mansoni GSH S-transferases with a subunit M, of 26,000 (Sm26) have been isolated by two laboratories (Henkle et al. 1990;Trottein et al. 1990) and the immunological relationship between Sm26 and the corresponding S. japonicum GSH S-transferase (Sj26) has been examined (Tiu et al. 1988). Because the focus of those studies has been only on the potential of GSH S-transferasesas vaccine antigens, other biochemical properties of the recombinant Sm26 proteins have been largely ignored. This makes direct comparisons of the catalytic properties of SmGST-4 and recombinant Sm26’s impossible. However, the identical subunit M, together with immunological cross-reactivity between i!4, 26,000 proteins in S. mansoni and S. juponicum cytosol in our work (Fig. lB, lanes 1 and 5) and that of Tiu et al. (1988), make it quite likely that SmGST-4 corresponds to one of the Sm26 gene products. Recently, Wright et al. (1991)isolated another Sm26 cDNA that encodes a distinct, but very similar S. mansoni GSH S-transferase to one previously isolated (Henkle et al. 1990). Based on Southern blot analysis of genomic S. munsoni DNA, Wright et al. (1991) suggested that there may be two distinct Sm26 genes. Although GSSG elution of GSH-agarose yielded a single catalytically active isoenzyme, SmGST-4, we noted that a trace of M, 26,000 polypeptide remained bound to the column, regardless of the volume of GSSG elution buffer used. This minor form of GSH S-transferase may correspond to the second Sm26 gene product. ACKNOWLEDGMENTS We thank Heather Gange for her technical assistance with 4-hydroxyalk-2-enal conjugation experiments. This work was supported by NIH Grant AI22520 from the United States-Japan Cooperative Medical Sciences Program. K.A.O. and K.M.H. were supported in part by NIH Training Grant ES07015
CHAR~TERI~ATI~N
OF S. munsoni GSH S-TRANSFERASE
This is contribution 252 of the Environmental Toxicology Center, University of Wisconsin-Madison.
REFERENCES ALIN, P., DANIELSON, H., AND MANNERVIK, B. 1985. 4-Hydroxyalk-Zenals are substrates for glutathione transferase. FEBS Letters 179, 267-270. BRADFORD, M. M. 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of proteindye binding. Analytical Biochemistry 72, 248-254. ESTERBAUER, H., AND WEGER, W. 1967. Uber die Wirkung von Aldehyden auf gesunde und maligna Zellen, 3. Mitteilung: Synthese von homologen 4-Hydroxy-2-alkenalen, II. Monatshefte Chemie 98, 1994-2000. ESTERBAUER, H., ZOLLNER, H., AND SCHAUR, R. J. 1988. Hydroxyalkenals: Cytotoxic products of lipid peroxidation. ISI Atlas of Science: Biochemistry 1, 311-317. HABIG, W. H., AND JAKOBY, W. B. 1981. Assays for differentiation of glutathione S-transferases. In “Methods in Enzymology” (W. B. Jakoby, Ed.), Vol. 77, pp. 398-405. Academic Press, New York. HENKLE, K. J., DAVERN, K. M., WRIGHT, M. D., RAMOS, A. J., AND MITCHELL, G. F. 1990. Comparison of the cloned genes of the 26- and 28kilodalton glutathione S-transferases of Schistosoma japonicum and Schistosoma mansoni. Molecular Biochemistry and Parasitology 40, 23-34. JENSSON, H., GUTHENBERG, C., ALIN, P., AND MANNERVIK, B. 1986. Rat glutathione transferase 8-8, an enzyme efficiently detoxifying 4-hydroxyalk-2enals. FEBS Letters 203, 207-209. KETTERER, B., MEYER, D. J. AND CLARK, A. G. 1988. Soluble glutathion transferase isoenzymes. In “Glutathione Conjugation: Mechanisms and Biological Significance” (H. Sies and B. Ketterer, Eds.), pp. 73-135. Academic Press, London. LAEMMLI, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680-685. MANNERVIK, B., AND DANIELSON, U. H. 1988. Glutathione transferases. Structure and catalytic activity. CRC Critical Reviews in Biochemistry 23, 283337. MEYER, D. J., LALOR, E., COLES, B., KISPERT, A., ALIN, P., MANNERVIK, B., AND KETTERER, B. 1989. A single-step purification and h.p.1.c. analysis of glutathione transferase 8-8 in rat tissues. Biochemical Journal 260, 785-788. MKOJI, G. M., SMITH, J. M., AND PRICHARD, R. K. 1988. Antioxidant systems in Schistosoma mansoni: Correlation between susceptibility to oxidant killing and the levels of scavengers of hydrogen peroxide and oxygen free radicals. International Journal for Parasitology 18, 661-666.
ISOENZYME
4
55
O’LEARY, K. A., AND TRACY, J. W. 1988. Purification of three cytosolic glutathione S-transferases from adult Schistosoma mansoni. Archives of Biochemistry and Biophysics 264, 1-12. O’LEARY, K. A., AND TRACY, J. W. 1991. Schistosoma mansoni: Glutathione S-transferase-catalyzed detoxication of dichlorvos. Experimental Parasitoiogy72, 3.55-361. PAGLIA, D. E., AND VALENTINE, W. N. 1967. Studies on the quantitative and qualitative characterization of erythrocyte glutathione peroxidase. Journal of Laboratory and Clinical Medicine IO, 158-169. SIEGEL, D. A., AND TRACY, J. W. 1989. Schistosoma mansoni: Influence of the female parasite on glutathione biosynthesis in the male. Experimental Parasitology 69, 116124. SIMURDA, M. C., VAN KEULEN, H., REKOSH, D. M., AND Lo VERDE, P. T. 1988. Schistosoma mansoni: Identification and analysis of an mRNA and a gene encoding superoxide dismutase (Cu/Zn). Experimental Parasitology 67, 73-84. TAYLOR, J. B., VIDAL, A., TORPIER, G., MEYER, D. J., ROITSCH, C., BALLOUL, J. M., SOUTHAN, C., SONDERMEYER, P., PEMBLE, S., CAPRON, A., AND KETTERER, B. 1988. The glutathione transferase activity and tissue distribution of a cloned Mr28K protective antigen of Schistosoma mansoni. EMBO Journal 7, 465-472. TIU, W. D., DAVERN, K. M., WRIGHT, M. D., BOARD, P. G., AND MITCHELL, G. F. 1988. Molecular and serological characteristics of the glutathione S-transferases of Schistosoma japonicum and Schistosoma mansoni. Parasite Immunology 10,
693-706. TROT~EIN, F., KIENY, M. P., VERWAERDE, C., TORPIER, G., PIERCE, R. J., BALLOUL, J.-M., SCHMITT, D., LECOCQ, J.-P., AND CAPRON, A. 1990. Molecular cloning and tissue distribution of a 26kilodalton Schistosoma mansoni glutathione S-transferase. Molecular Biochemistry and Parasitology 41,3544. VAN OORDT, B. E. P., VAN DEN HEUVEL, J. M., TIELENS, A. G. M., AND VAN DEN BERGH, S. G. 1985. The energy production of the adult Schistosoma mansoni is for a large part aerobic. Molecular Biochemistry and Parasitology 16, 117-126. WRIGHT, M. D., HARRISON, R. A., MELDER, A. M., NEWPORT, G. R., AND MITCHELL, G. F. 1991. Another 26-kilodalton glutathione S-transferase of Schistosoma mansoni. Molecular Biochemistry and Parasitology 49, 177-180. YAL~IN, S., JENSSON, H., AND MANNERVIK, B. 1983. A set of inhibitors for discrimination between the basic isozymes of glutathione transferase in rat liver. Biochemical and Biophysical Research Communications 114,829-834. Received 21 January April 1992.
1992; accepted with revision
30
PARASITOLOGY
75,47-55 (1992)
Schistosoma mansoni: Single-Step of Glutathione
Purification and Characterization S-Transferase lsoenzyme 4
KATHLEEN A. O'LEARY,'KAREN M. HATHAWAY,ANDJAMES W. TRACY* Departments of Comparative Biosciences and Pharmacology, and the Environmental Toxicology Center, University of Wisconsin, 2015 Linden Drive West, Madison, Wisconsin 53704-1102, U.S.A. O’LEARY, K. A., HATHAWAY, K. M., AND TRACY, J. W. 1992. Schistosoma mnnsoni: Single-step purification and characterization of glutathione S-transferase isoenzyme 4. Experimental Parasitology 75, 47-55. A soluble glutathione S-transferase isoenzyme, designated SmGST-4 was purified to apparent homogeneity in a single step from the cytosol of adult Schistosoma mansoni by selective elution of the enzyme from a glutathione-agarose affinity column using glutathione disulfide. SmGST-4, which comprised about 5% of the bound glutathione S-transferase activity, could be distinguished from the previously characterized glutathione S-transferase isoenzyme family (SmGST-l/2/3) by its unique chromatographic behavior, lower subunit M, (26,000), differences in substrate specificity and inhibitor sensitivity, and a lack of reactivity with antiserum to SmGST-3. The purified isoenzyme catalyzed the conjugation of several model xenobiotics including I-chloro-2,4dinitrobenzene, ethacrynic acid, and transd-phenyl-3-buten-2-one. Like the SmGST-l/2/3 isoenzyme family, SmGST-4 failed to catalyze the conjugation of a model epoxide substrate, 1,2-epoxy-3+nitrophenoxy)propane. Because glutatbione S-transferases from other organisms play a role in protecting cells against the toxic products of lipid peroxidation, SmCST-4 and the members of the SmGST-l/2/3 isoenzyme family were tested for their capacity to reduce cumene hydroperoxide and to catalyze the conjugation of 4-hydroxyalk2-enals. Although all four isoenzymes catalyzed both reactions, the specific activity of SmGST-1, SmGST-2, and SmGST-3 toward cumene hydroperoxide was at least lo-fold greater than that of SmGST-4. In contrast, the latter more effectively conjugated a homologous series of 4-hydroxyalk-2-enal isomers. For example, the specific activity of SmGST-4 toward 4-hydroxynon-2-enal(10.8 pmol . min-’ . mg protein-‘), the isomer formed under physiological conditions, was about S-fold greater than that of SmGST-1 and 1Zfold greater than that of the major isoenzyme SmGST-3. These results are consistent with a multifunctional role for schistosome glutathione S-transferases in protecting the parasite against both xenobiotics and toxic endogenous products of lipid peroxidation. Q 1992 Academic PESS. IIK. INDEX DESCRIPTORS AND ABBREVIATIONS: Trematode; Schistosomn mnnsoni; Glutathione S-transferase (GST; EC 2.5.1.18); Isoenzyme; Affinity chromatography; Substrate specificity; 4-Hydroxyalk-2-enal; Cumene hydroperoxide; H,O,; Glutathione (GSH); Affinity-purified Schistosoma mansoni Glutathione S-transferase (SmGST-AP); Glutathione disulfide (GSSG); Glutathione peroxidase; 1-chloro-2,Cdinitrobenzene (CDNB); Sodium dodecyl sulfate-polyacrylamide-gel electrophoresis (SDS-PAGE).
INTRODUCTION Glutathione S-transferases (EC 2.5.1.18) catalyze the conjugation of both endogenous and xenobiotic electrophiles with the major cellular nucleophile GSH (Mannervik and Danielson 1988). GSH S-trans-
ferases are widely distributed in nature, being found in both vertebrates and invertebrates, including schistosomes (O’Leary and Tracy 1988; Taylor et al. 1988; Henkle et al. 1990; Trottein et al. 1990; Wright et al. 1991). Most organisms contain multiple GSH S-transferase isoenzymes that are characterized by distinct, but often overlapping, substrate specificities that account for their capacity to metabolize structurally diverse chemicals.
’ Present address: McArdle Laboratory for Cancer Research, University of Wisconsin, Madison, WI 53706. * To whom correspondence should be addressed. 47
0014-4894192$5.00 Copyright 0 1992 by Academic Press, Inc. AU rights of reproduction in any form reserved.
48
O'LEARY,HATHAWAY,ANDTRACY
Adult Schistosoma mansoni contain at least five forms of GSH S-transferase. We previously described the purification and characterization of three of them, designated SmGST-1, SmGST-2, and SmGST-3 (O’Leary and Tracy 1988). These isoenzymes, which compose 2-4% of the soluble male worm protein, were purified by a combination of GSH affinity chromatography and high-pressure liquid chromatography. Although SmGST- 1, SmGST-2, and SmGST-3 have the same apparent subunit M, (28,500) and are immunologically crossreactive, they can be distinguished from one another by differences in isoelectric point, catalytic activity, and inhibitor sensitivity (O’Leary and Tracy 1988). They have since been found to compose an isoenzyme family consisting of two homodimers (SmGST-1 and SmGST-3) and the corresponding heterodimer SmGST-2 (O’Leary and Tracy, under review). Another isoenzyme that fails to bind to GSHagarose (O’Leary and Tracy 1988) preferentially catalyzes the conjugation of epoxide substrates and the detoxication of dichlorvos, the active form of the antischistosomal drug metrifonate (O’Leary and Tracy 1991). Besides the three major isoenzymes, affinity-purified SmGST preparations contained a minor form with an apparent subunit M, of 26,000. That isoenzyme was not, however, recovered after high-pressure liquid chromatography (O’Leary and Tracy 1988). We now report that this GSH S-transferase, designated SmGST-4, can be purified in a single step by selective elution from GSH-agarose with GSSG. SmGST-4 can be distinguished from the SmGST-l/2/3 isoenzyme family both by its unique chromatographic behavior and by differences in catalytic specificity, inhibitor sensitivity, apparent subunit M,, and a lack of immunological cross-reactivity. In addition to detoxicating drugs and other foreign chemicals, GSH S-transferases play an important role in protecting
cells against the toxic products of lipid peroxidation. For example, they act as selenium-independent GSH peroxidases catalyzing the reduction of fatty acid and other organic hydroperoxides (Ketterer et al. 1988). Unlike selenium-dependent GSH peroxidase, however, GSH S-transferases do not reduce HzOz. GSH S-transferases also catalyze the conjugation of 4-hydroxyalk-Zenals, chemically reactive aldehyde products of lipid peroxidation that are implicated in the cytotoxicity of reactive oxygen (Esterbauer et al. 1988). Because adult schistosomeslive in an aerobic environment and derive at least one-third of their energy from aerobic catabolism of glucose (Van Oordt et al. 1985), they can be expected to undergo reactive oxygenmediated lipid peroxidation. We therefore tested SmGST-4 and the SmGST-l/2/3 family isoenzymes for their capacity to reduce a model organic hydroperoxide and to conjugate a homologous series of 4-hydroxyalk-Zenal isomers. MATERIALSANDMETHODS Parasites. Female mice (CF, strain, 18-22 g, Sasco, Madison, WI) were injected subcutaneously with 200 freshly shed cercariae of a Puerto Rican strain of S. mansoni. After 42-49 days, parasites were recovered by portal venous perfusion with ice-cold RPM1 medium 1640 (GIBCO, Grand Island, NY) containing 25 mM 2-hydroxyethylpiperazine-N’-2-ethane sulfonate (Research Organics, Cleveland, OH) buffer, pH 7.4 (Siegel and Tracy 1989). Parasites were washed several times with cold medium to remove traces of contaminating host tissue. A sample of adult Schistosoma japonicum, provided by Dr. John Bruce (University of Lowell, Lowell, MA) was stored frozen below - 20°C until use. Enzyme assays. GSH S-transferase activity was routinely measured during enzyme purillcation with I-chloro-2,Cdinitrobenzene (CDNB, 99 + %, Aldrich Chemical, Milwaukee, WI) as described by Habig and Jakoby (1981). The rate of product formation was monitored at 340 nm (c = 9.6 m&-i . cm-‘) using a Gilford Response spectrophotometer (Ciba-Coming Diagnostics, Oberlin, OH) maintained at 25°C. One unit of SmGST activity is defined as the amount of enzyme that catalyzes the formation of 1 kmol of S-(2,4dinitrophenyl)glutathione per minute under the conditions of the assay. Specific activity is expressed
c~~rwxERIzAw0~
OF S.mansoni~~~
as units per milligram of protein. Activity of individual SmGST isoenzymes toward 1,2-chloro-4nitrobenzene (Pfaltz and Bauer, Stamford, CT), ethacrynic acid (Sigma Chemical, St. Louis, MO), 1,2-epoxy-3-(pnitrophenoxy)propane (Sigma), and rrans-Cphenyl3buten-2-one (Aldrich) was similarly assayed at the appropriate wavelengths (Habig and Jakoby 1981). A homologous series of 4-hydroxyalk-2-enals (C,, C,, C,, and Cl1 isomers) was generously provided by Professor Hermann Esterbauer (Universitit Graz, Austria). The rate of 4-hydroxyalk-2-enal conjugation with GSH was measured spectrophotometrically at 224 nm (Alin et al. 1985) using E = 13.75 rn&-’ ’ cm-’ (Esterbauer and Weger 1967). GSH peroxidase activity of SmGST’s toward H,O, and cumene hydroperoxide (Sigma) was determined by coupling peroxide-dependent GSSG formation to GSSG reductase (Type III, Sigma) catalyzed NADPH oxidation (Paglia and Valentine 1967). Appropriate controls were included to ensure that NADPH oxidation was both protein and peroxide dependent. One unit of GSH peroxidase activity is defined as the amount of enzyme that catalyzes the oxidation of 1 nmol of NADPH per minute under the conditions of the assay. Specific activity is expressed as units per milligram protein. Commercially prepared GSH peroxidase from bovine erythrocytes (Sigma) was used to validate the enzyme assay. Several prototypic inhibitors, including hematin (Sigma), bromosulfophthalein (Sigma), indocyanine green (Pfaltz and Bauer), triethyltin bromide, and triphenyltin chloride (Alpha Products, Danvers, MA) were tested for their ability to inhibit CDNB conjugation. The inhibition constant I,, is defined as the concentration of inhibitor required to produce 50% inhibition under the assay conditions defined for the experiment and was interpolated from graphs of fractional enzyme velocity as a function of the log,, of the inhibitor concentration (Yalcin et al. 1983).Protein concentrations were estimated by a dye-binding method (Bradford 1976)using crystalline bovine serum albumin (Miles Laboratories, Elkhart, IN) as the reference protein. Enzyme purification. All operations were carried out at OXC, unless otherwise noted. Adult schistosomes (1 to 2 g wet wt) were homogenized in 4 vol of 10 mM potassium phosphate buffer, pH 7.0, containing 1 mM Na,EDTA. The homogenate was centrifuged for 60 min at 47,800g. The supematant was decanted and centrifuged for 90 min at lO5,OOOg.Both pellets were discarded. The resulting cytosolic fraction was applied to a 1.5 x 6.0-cm column of S-linked GSHagarose (Sigma), equilibrated with 10 mM potassium phosphate buffer, pH 7.0. Unbound material was washed through the column with 5 bed vol of equilibration buffer. The column was next eluted with 5 bed vol of 10 mM potassium phosphate buffer, pH 7.4,
S-TRANSFERME ISOENZYME 4
49
containing 5 mM GSSG (Sigma). The remaining SmGST’s were eluted with 50 mM Tris-HCl buffer, pH 9.6, containing 5 mM GSH. SmGST-1, SmGST-2, and SmGST-3 were subsequently resolved by highpressure liquid chromatography (O’Leary and Tracy 1988). Column fractions containing enzyme activity were pooled and concentrated in a stirred ultrafiltration cell fitted with a YM-10 membrane (Amicon, Danvers, MA). Puritied SmGST’s were stored at - 20°C in a pH 7.0 buffer containing 10 nu&f potassium phosphate, 30% (v/v) glycerol, 1 mM Na,EDTA and 1 mM dithiothreitol (Research Organics). Electrophoresis. SDS-PAGE was carried out in 0.75-mm thick 12.5% vertical slab gels according to the method of Laemmli (1970). Protein samples were mixed with an equal volume of 60 mM Tris-HCl, 5% (v/v) 2-mercaptoethanol, 3% (w/v) SDS, 20% (v/v) glycerol, and 0.1% (w/v) bromophenol blue tracking dye, pH 6.8, and immediately boiled for 3 min. A mixture reference proteins (soybean trypsin inhibitor, M, 20,000; bovine erythrocyte carbonic anhydrase, M, 29,000; rabbit muscle aldolase, M, 40,000; and bovine serum albumin, M, 67,008; Sigma) was prepared in parallel. Gels were stained for protein with 0.25% (w/ v) Coomassie blue R-250 in 25% (v/v) 2-propanol/lO% (v/v) acetic acid. After overnight staining, gels were destained in 10% (v/v) acetic acid. Immunoblotting. After SDS-PAGE separation, gels were electroblotted onto nitrocellulose (Schleicher and Schuell, Keene, NH) for 3 hr at 250 mA. Filters were blocked overnight in a solution of 1.5% (w/v) Carnation Nonfat Dry Milk in 10 mM Tris-HCl buffer, pH 8.0, containing 0.15 M NaCl and 0.05% (v/v) Tween 20 (TBST). Duplicate filters were incubated for 2 hr in a 1:2O,UOO (v/v) dilution in TBST of primary rabbit antiserum elicited (O’Leary and Tracy 1988) against either affinity purified SmGST or an electrophoretically pure sample of SmGST-3 (anti-SmGST3). Filters were washed three times with TBST, followed by a 30-min incubation in a 1:7500(v/v) dilution of goat anti-rabbit IgG-alkaline phosphatase conjugate (Promega, Madison, WI). After washing three times with TBST, filter were exposed to a solution of 1.65 mg . ml-’ nitro blue tetrazolium (Promega) and 0.83 . ml-’ 5-bromo-4-chloro-3-indolyl phosphate zomega) in 100 mM Tris-HCl buffer, pH 9.5, containing 0.15 M NaCl and 5 mM MgCl,. The reaction was stopped after color formation (typically 1 to 3 min) by transferring the falters to 20 mM Tris-HCI buffer, pH 8.0, containing 5 mM Na,EDTA.
RESULTS
When the cytosolic fraction of adult S. munsoni was applied to thiol-linked GSHagarose, about 90% of the CDNBconjugating activity was retained on the
50
O’LEARY,
HATHAWAY,
column. We observed that about 10%of the bound GSH S-transferase activity could be selectively eluted from the column with a neutral phosphate buffer containing 5 mM GSSG, resulting in a 45 to 50-fold increase in specific enzyme activity (Table I), Analysis of the GSSG-eluted fraction by SDSPAGE showed a single polypeptide with an apparent M, = 26,000 (Fig. lA, lane 4). Isoelectric focusing in polyacrylamide under nondenaturing conditions also revealed a single protein species with an isoelectric point of approximately 7.4 (not illustrated). This GSH S-transferase isoenzyme was designated SmGST-4. The remainder of the bound GSH S-transferase activity was eluted with an alkaline buffer containing 5 mM GSH. The latter fraction, which could be resolved by chromatofocusing into the previously characterized major isoenzymes SmGST-1, SmGST-2, and SmGST-3 (O’Leary and Tracy 1988), still contained a small amount (l-2%) of M, 26,000polypeptide that could not be eluted from GSHagarose with GSSG, regardless of the volume of elution buffer used. Comparisons of substrate specificity demonstrated that SmGST-4 could be distinguished from the previously purified SmGST isoenzymes (Table II). For example, SmGST-4 was about 1.5-fold more active toward the prototypic aryl halide substrate CDNB, but 1.4- to 2.2-fold less active toward another aryl halide 1,Zdichloro-6 nitrobenzene compared with the SmGST-II
AND
TRACY
2/3 isoenzyme family. SmGST-4 was 2- to 12-fold more active toward ethacrynic acid than the other isoenzymes, whereas its specific activity toward the a&unsaturated carbonyl substrate truns-4-phenyl3-buten2-one was intermediate between that of the other isoenzymes. Like the SmGST-l/2/3 isoenzyme family, SmGST-4 did not catalyze the conjugation of a model epoxide substrate 1,2-epoxy-3-(p-nitrophenyl)propane. In addition to catalyzing the conjugation of xenobiotics, many GSH S-transferases display selenium-independentGSH peroxidase activity and thus catalyze the reduction of organic hydroperoxides, but not H,Oz (Ketterer et al. 1988).Consistent with this idea, SmGST’s failed to catalyze the reduction of H,O, (Table II). All four isoenzymes did, however, catalyze the reduction of cumene hydroperoxide, a model organic hydroperoxide. Although SmGST-1, SmGST-2, and SmGST-3 showed very similar specific activities toward cumene hydroperoxide, SmGST-4 was 13- to 15-fold less active toward that substrate (Table II). Because GSH S-transferases seem to play a key role in protecting cells against the toxicity of chemically reactive aldehydie lipid peroxidation products (Esterbauer et al. 1988), the four SmGST isoenzymes were tested for their ability to catalyze the conjugation of a homologous series of 4-hydroxyalk-Zenal isomers (Table III). Members of the SmGST-l/2/3 family showed
TABLE I Purification of SmGST-4 from Adult S. mansoni
Step Cytosol GSH-agarose GSSG elution GSH elution
Protein concentration (mg . ml-‘) 3.02 0.05 0.61
Total protein b-m) 27 0.04 0.76
Specific” activity (units . mg-I) 0.95 48.4 20.1
Total activity (units)
Fold purification
Yield cm
19.4
-
100
1.9 15.3
50 21
(1Determined with CDNB as the xenobiotic substrate. One unit = 1 pmol product formed . min-‘.
10 79
51
CHA~A~-~ER~~AT~ON OF S. mansoni GSHS-TRAN~FERASE I~OEN~YME4 Mr
x 10-3 67
-
1 1 2 3 4 5 1 2 34 5 2 3 4 5 FIG. 1. SDS-PAGE analysis and immunoblotting of schistosome GSH S-transferases. Triplicate protein samples were separated on 12.5% acrylamide gels under reducing conditions in the presence of SDS. (A) One gel was stained for protein with Coomassie blue R-250. The other gels were electroblotted onto nitrocellulose and probed with either (B) a 1:20,000 (v/v) dilution of rabbit antiSmGST-AP, or (C) a 1:20,000(v/v) dilution of rabbit anti-SmGST-3. Binding of anti-SmGST antibodies was detected using a goat anti-rabbit IgG-alkaline phosphatase conjugate and nitro blue tetrazoliuml 5-bromo-4~chloro-3-indolyl phosphate substrate solution. Lane 1, Unfractionated S. mansoni cytosol (40 kg); lane 2, SmGST-AP (4 kg); lane 3, Purified SmGST-3 (2 pg); lane 4, SmGST-4 (1.3 wg) selectively eluted from GSH-agarose with GSSG; lane 5, unfractionated S. japonicum cytosol(30 kg).
similar specific activities toward all isomers tested. In each case SmGST-2 showed activity intermediate between that of SmGST-1 and SmGST-3. In contrast, SmGST-4 was from 4- to 25fold more active toward each of the 4-hydroxyalk-2enals tested than were members of the SmGST-l/2/3 isoenzyme family. In particular the activity of SmGST-4 toward 4-hydroxynon-Zenal, the isomer that is formed
under physiological conditions, was nearly S-fold greater than SmGST-1 and 1Zfold more active than the major isoenzyme SmGST-3. Several inhibitors that have been used to characterize other SmGST’s were tested for their ability to inhibit the CDNBconjugating activity of SmGST-4. The Iso values, interpolated from graphs of activity remaining as a function of inhibitor concen-
TABLE II Substrate Specificity of Purified S. mansoni GSH S-Transferases Specific activity” Substrate I-Chloro-2,4-dinitrobenzene 1,2-Dichloro-4-nitrobenzene Ethacrynic acid trans-4-Phenyl-3-buten-2-one 1,2-Epoxy-3-(p-nitrophenoxy)propane W, Cumene hydroperoxide
SmGST- 1
SmGST-2
SmGST-3
SmGST-4
306 136 3.6’
32’ 186 1.96 0.32’ 06 0 0.45
26” 116 0.536 0.16b 06 0 0.39
44 8 6.5 0.29 0 0 0.03
F” 0 0.39
LISpecific activity is expressed as micromole of product formed . min- ’ . mg protein-‘. ’ For comparison, data taken from O’Leary and Tracy (1988).
52
O’LEARY,
HATHAWAY,
Conjugation of CHydroxyalk-2-enals
AND TRACY
TABLE III by Purified 5. munsoni GSH S-Transferases Specific activity”
Substrate
SmGST-I
SmGST-2
SmGST-3
SmGST-4
4-Hydroxypent-Zenal (C,) 4-Hydroxyhex-2-enal (Cd CHydroxynon-Zenal (C,) 4-Hydroxyundec-2-enal (C, r)
2.7 1.2 2.4 2.0
1.4 0.5 1.9 1.7
0.8 0.5 0.9 1.4
14.7 12.2 10.8 11.9
a Specific activity is expressed as micromoles of product formed * min- ’ . mg protein- ‘.
tration (Yalcin ef al. 1983) are shown in Table IV. Triphenyltin chloride was the most potent inhibitor of SmGST-4 with an I,, value about lOO-fold lower than the other inhibitors tested. Another organotin compound, triethyltin bromide, was about IOfold less potent than the triphenyl compound. Compared with SmGST-l/2/3 isoenzyme family, SmGST-4 was 50- to 500-fold less sensitive to inhibition by either hematin or bromosulfophthalein and about 6-fold less sensitive to inhibition by indocyanine green. Rabbit antisera elicited against SmGST preparations (O’Leary and Tracy 1988) were used to compare the immunological relationship of SmGST-4 to other SmGST’s (Fig. 1). Triplicate protein samples were separated by SDS-PAGE. One gel was stained for protein (Fig. lA), while the other two were immunoblotted with either rabbit anti-affinity purified SmGST (Fig. 1B) or rabbit anti-SmGST-3 (Fig. 1C). Antiaffinity purified SmGST recognized two
species (M, 28,500 and M, 26,000) in both the unfractionated cytosol and a sample of affinity purified SmGST, purified by conventional GSH elution (Fig. lB, lanes I and 2). This antiserum recognized a single species (M, 28,500) in purified SmGST-3 (Fig. lB, lane 3) and a single species (M, 26,000) in purified SmGST-4 prepared by selective GSSG elution (Fig. lB, lane 4). Although anti-SmGST-3 recognized the SmGST-l/2/3 family in unfractionated S. mansoni cytosol and in affinity-purified SmGST, it did not recognize purified SmGST-4 (Fig. lC, lane 4). Because an immunological relationship has been noted previously between SmGST’s and GSH S-transferases from S. juponicum (Tiu et al. 1988), a sample of unfractionated S. juponicum cytosol was also analyzed for reactivity against the two antisera. Anti-affinity purified SmGST reacted with two species in S. juponicum (Fig. lB, lane 5) that had the same apparent M, (28,500 and 26,000, respectively) as the SmGST’s. Anti-SmGST-3 recognized the
TABLE IV Inhibition of Purified S. mansoni GSH S-Transferases Iso values (uma Inhibitor
SmGST-lb
SmGST-2b
SmGST-3b
SmGST-4
Bromosulfophthalein Hematin Indocyanine green Triphenyltin chloride Triethyltin bromide
0.05 0.02 2.7 0.18 11.0
0.05 0.09 2.7 0.24 5.3
0.03 0.20 3.3 0.36 8.6
12 11 20 0.16 15
LIMeasured with CDNB as the xenobiotic substrate at 25°C. b For comparison, data taken from O’Leary and Tracy (1988).
CHARACTERIZATION 0~ S. mansoniOs~
S-T~~N~FE~~SE I~OENZYME 4
53
M, 28,500 species, but not the M, 26,000 more striking differences can be seen when comparing catalytic activity toward cumene hydroperoxide and 4-hydroxyalk2-enals (Tables II and III). Whereas SmGST-4 was substantially less active toward the organic hydroperoxide, it was clearly more effective at catalyzing the conDISCUSSION jugation of the series of reactive c.u,BAt least five GSH S-transferases are ex- unsaturated aldehydes. Inhibitor studies pressed in adult S. mansoni. Earlier we re- also revealed differences between SmGST-4 and the SmGST-l/2/3 family (Taported the purification and characterization of three major GSH S-transferases, desig- ble IV). Although the I,, values for organnated SmGST- 1, SmGST-2, and SmGST-3 otin compounds were comparable for all (O’Leary and Tracy 1988). These isoen- four isoenzymes, SmGST-4 was 50- to 500zymes, which compose 24% of the total fold less sensitive to inhibition by hematin cytosolic protein, represent an isoenzyme or bromosulfophthalein than the major family consisting of two homodimers isoenzymes. In contrast to SmGST-1, (SmGST-1 and SmGST-3) and the corre- SmGST-2, and SmGST-3 that have the sponding heterodimer SmGST-2 (O’Leary same electrophoretic mobility by SDSand Tracy, under review). As reported PAGE (apparent subunit M, 28,500) and that share major immunological determihere, another isoenzyme, designated SmGST-4, can be selectively eluted from nants (O’Leary and Tracy 1988), SmGST-4 GSH-agarose with GSSG. Based on pro- displays a faster electrophoretic mobility tein recovery, SmGST-4 compose about (M, 26,000) and does not react with antise5% of the total affinity-purified SmGST rum to SmGST-3 (Fig. lC, lane 4). Finally, fraction. At least one additional form of the SmGST-4 can be distinguished from the enzyme that fails to bind to the affinity ma- fifth isoenzyme both by its affinity for trix has been detected, but not isolated due GSH-agarose and by its lack of catalytic to its instability (O’Leary and Tracy 1988). activity toward 1,2-epoxy-3-(p-nitroThat isoenzyme preferentially catalyzes the phenoxy)propane. The molecular basis for selective elution conjugation of the model epoxide 1,2epoxy-3-(p-nitrophenoxy)propane and the of SmGST-4 from GSH-agarose by GSSG detoxication of dichlorvos, the active form is unknown. However, Meyer et al. (1989) of the antischistosomal drug metrifonate found that certain rat GSH S-transferases (isoenzymes l-l and 8-8) could likewise be (O’Leary and Tracy 1991). selectively eluted from GSH-agarose with In addition to its unique chromatographic behavior on GSH-agarose, SmGST-4 can GSSG. Interestingly, rat isoenzyme 8-8 and be readily distinguished from the SmGST- SmGST-4 seem to share some catalytic l/2/3 isoenzyme family based on differ- similarities. Like SmGST-4, isoenzyme 8-8 ences in the pattern of substrate specificity, is quite active toward ethacrynic acid and apparent subunit M,, and a lack of immu- of all rat GSH S-transferases it is most efnological cross-reactivity. For example, fective catalyst of 4-hydroxyalk-Zenal conSmGST-4 shows the highest specific activ- jugation (Jensson et al. 1986). Additional ity toward the prototypic aryl halide sub- studies will be necessary, however, to destrate CDNB, but the lowest activity to- termine whether SmGST-4 and isoenzyme ward another aryl halide 1,2-dichloro-48-8 have other properties in common. nitrobenzene (Table II). It is the isoenzyme The reason for the abundance and commost active toward ethacrynic acid. Even plexity of GSH S-transferases in adult S. species in S. japonicum (Fig. lC, lane 5). No reaction was observed with either antiserum at the dilution employed against proteins that failed to bind to the GSH affinity column (data not shown).
54
O'LEARY,HATHAWAY,AND
mansoni is not fully understood. In mammals, these enzymes are multifunctional, catalyzing the metabolism of both endogenous and xenobiotic compounds and serving as intracellular binding/transport proteins for ligands such as bilirubin and hematin (Mannervik and Danielson 1988). Clearly, SmGST’s are able to catalyze the detoxication of xenobiotics, including at least one clinically used drug. Results of experiments reported here demonstrate that SmGST’s also catalyze the detoxication of endogenous products of lipid peroxidation. Because of their location within the host’s portal venous system proximal to the liver, S. mansoni are exposed to a wide variety of potentially toxic chemicals. The bloodstream is an aerobic environment and adult worms do carry out oxidative metabolism (Van Oordt et al. 1985).Both of these conditions result in exposure of the organism to reactive forms of oxygen, such as 02-* and H202, leading to peroxidation of membrane lipids (Esterbauer et al. 1988). Furthermore, various host effector cells generate reactive oxygen species resulting in peroxidative damage and cytotoxicity. Adult schistosomes are generally resistant to such immune attack and they do contain antioxidant enzymes. Simurda et al. (1988) cloned an S. mansoni Cu/Zn superoxide dismutase that catalyzes the metabolism of 02-* to H,Oz. Mkoji et al. (1988) reported the activities of several antioxidant enzymes including selenium-dependentglutathione peroxidase that catalyzes the reduction of H*O,. As shown here, SmGST’s catalyze the reduction of organic hydroperoxides, but not H,Oz, consistent with their role as selenium-independentglutathione peroxidases. Furthermore, SmGST’s, particularly SmGST-4, catalyze the conjugation with GSH of 4-hydroxyalk2-enals, one class of reactive lipid peroxidation byproducts that are implicated in the cytotoxicity of reactive oxygen (Esterbauer et al. 1988).When taken together these data support the idea that schistosome GSH S-transferases are multifunctional en-
TRACY
zymes. It seemslikely that additional catalytic and binding functions of these schistosome proteins await discovery. Identical cDNA’s encoding S. mansoni GSH S-transferases with a subunit M, of 26,000 (Sm26) have been isolated by two laboratories (Henkle et al. 1990;Trottein et al. 1990) and the immunological relationship between Sm26 and the corresponding S. japonicum GSH S-transferase (Sj26) has been examined (Tiu et al. 1988). Because the focus of those studies has been only on the potential of GSH S-transferasesas vaccine antigens, other biochemical properties of the recombinant Sm26 proteins have been largely ignored. This makes direct comparisons of the catalytic properties of SmGST-4 and recombinant Sm26’s impossible. However, the identical subunit M, together with immunological cross-reactivity between i!4, 26,000 proteins in S. mansoni and S. juponicum cytosol in our work (Fig. lB, lanes 1 and 5) and that of Tiu et al. (1988), make it quite likely that SmGST-4 corresponds to one of the Sm26 gene products. Recently, Wright et al. (1991)isolated another Sm26 cDNA that encodes a distinct, but very similar S. mansoni GSH S-transferase to one previously isolated (Henkle et al. 1990). Based on Southern blot analysis of genomic S. munsoni DNA, Wright et al. (1991) suggested that there may be two distinct Sm26 genes. Although GSSG elution of GSH-agarose yielded a single catalytically active isoenzyme, SmGST-4, we noted that a trace of M, 26,000 polypeptide remained bound to the column, regardless of the volume of GSSG elution buffer used. This minor form of GSH S-transferase may correspond to the second Sm26 gene product. ACKNOWLEDGMENTS We thank Heather Gange for her technical assistance with 4-hydroxyalk-2-enal conjugation experiments. This work was supported by NIH Grant AI22520 from the United States-Japan Cooperative Medical Sciences Program. K.A.O. and K.M.H. were supported in part by NIH Training Grant ES07015
CHAR~TERI~ATI~N
OF S. munsoni GSH S-TRANSFERASE
This is contribution 252 of the Environmental Toxicology Center, University of Wisconsin-Madison.
REFERENCES ALIN, P., DANIELSON, H., AND MANNERVIK, B. 1985. 4-Hydroxyalk-Zenals are substrates for glutathione transferase. FEBS Letters 179, 267-270. BRADFORD, M. M. 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of proteindye binding. Analytical Biochemistry 72, 248-254. ESTERBAUER, H., AND WEGER, W. 1967. Uber die Wirkung von Aldehyden auf gesunde und maligna Zellen, 3. Mitteilung: Synthese von homologen 4-Hydroxy-2-alkenalen, II. Monatshefte Chemie 98, 1994-2000. ESTERBAUER, H., ZOLLNER, H., AND SCHAUR, R. J. 1988. Hydroxyalkenals: Cytotoxic products of lipid peroxidation. ISI Atlas of Science: Biochemistry 1, 311-317. HABIG, W. H., AND JAKOBY, W. B. 1981. Assays for differentiation of glutathione S-transferases. In “Methods in Enzymology” (W. B. Jakoby, Ed.), Vol. 77, pp. 398-405. Academic Press, New York. HENKLE, K. J., DAVERN, K. M., WRIGHT, M. D., RAMOS, A. J., AND MITCHELL, G. F. 1990. Comparison of the cloned genes of the 26- and 28kilodalton glutathione S-transferases of Schistosoma japonicum and Schistosoma mansoni. Molecular Biochemistry and Parasitology 40, 23-34. JENSSON, H., GUTHENBERG, C., ALIN, P., AND MANNERVIK, B. 1986. Rat glutathione transferase 8-8, an enzyme efficiently detoxifying 4-hydroxyalk-2enals. FEBS Letters 203, 207-209. KETTERER, B., MEYER, D. J. AND CLARK, A. G. 1988. Soluble glutathion transferase isoenzymes. In “Glutathione Conjugation: Mechanisms and Biological Significance” (H. Sies and B. Ketterer, Eds.), pp. 73-135. Academic Press, London. LAEMMLI, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680-685. MANNERVIK, B., AND DANIELSON, U. H. 1988. Glutathione transferases. Structure and catalytic activity. CRC Critical Reviews in Biochemistry 23, 283337. MEYER, D. J., LALOR, E., COLES, B., KISPERT, A., ALIN, P., MANNERVIK, B., AND KETTERER, B. 1989. A single-step purification and h.p.1.c. analysis of glutathione transferase 8-8 in rat tissues. Biochemical Journal 260, 785-788. MKOJI, G. M., SMITH, J. M., AND PRICHARD, R. K. 1988. Antioxidant systems in Schistosoma mansoni: Correlation between susceptibility to oxidant killing and the levels of scavengers of hydrogen peroxide and oxygen free radicals. International Journal for Parasitology 18, 661-666.
ISOENZYME
4
55
O’LEARY, K. A., AND TRACY, J. W. 1988. Purification of three cytosolic glutathione S-transferases from adult Schistosoma mansoni. Archives of Biochemistry and Biophysics 264, 1-12. O’LEARY, K. A., AND TRACY, J. W. 1991. Schistosoma mansoni: Glutathione S-transferase-catalyzed detoxication of dichlorvos. Experimental Parasitoiogy72, 3.55-361. PAGLIA, D. E., AND VALENTINE, W. N. 1967. Studies on the quantitative and qualitative characterization of erythrocyte glutathione peroxidase. Journal of Laboratory and Clinical Medicine IO, 158-169. SIEGEL, D. A., AND TRACY, J. W. 1989. Schistosoma mansoni: Influence of the female parasite on glutathione biosynthesis in the male. Experimental Parasitology 69, 116124. SIMURDA, M. C., VAN KEULEN, H., REKOSH, D. M., AND Lo VERDE, P. T. 1988. Schistosoma mansoni: Identification and analysis of an mRNA and a gene encoding superoxide dismutase (Cu/Zn). Experimental Parasitology 67, 73-84. TAYLOR, J. B., VIDAL, A., TORPIER, G., MEYER, D. J., ROITSCH, C., BALLOUL, J. M., SOUTHAN, C., SONDERMEYER, P., PEMBLE, S., CAPRON, A., AND KETTERER, B. 1988. The glutathione transferase activity and tissue distribution of a cloned Mr28K protective antigen of Schistosoma mansoni. EMBO Journal 7, 465-472. TIU, W. D., DAVERN, K. M., WRIGHT, M. D., BOARD, P. G., AND MITCHELL, G. F. 1988. Molecular and serological characteristics of the glutathione S-transferases of Schistosoma japonicum and Schistosoma mansoni. Parasite Immunology 10,
693-706. TROT~EIN, F., KIENY, M. P., VERWAERDE, C., TORPIER, G., PIERCE, R. J., BALLOUL, J.-M., SCHMITT, D., LECOCQ, J.-P., AND CAPRON, A. 1990. Molecular cloning and tissue distribution of a 26kilodalton Schistosoma mansoni glutathione S-transferase. Molecular Biochemistry and Parasitology 41,3544. VAN OORDT, B. E. P., VAN DEN HEUVEL, J. M., TIELENS, A. G. M., AND VAN DEN BERGH, S. G. 1985. The energy production of the adult Schistosoma mansoni is for a large part aerobic. Molecular Biochemistry and Parasitology 16, 117-126. WRIGHT, M. D., HARRISON, R. A., MELDER, A. M., NEWPORT, G. R., AND MITCHELL, G. F. 1991. Another 26-kilodalton glutathione S-transferase of Schistosoma mansoni. Molecular Biochemistry and Parasitology 49, 177-180. YAL~IN, S., JENSSON, H., AND MANNERVIK, B. 1983. A set of inhibitors for discrimination between the basic isozymes of glutathione transferase in rat liver. Biochemical and Biophysical Research Communications 114,829-834. Received 21 January April 1992.
1992; accepted with revision
30
