Glutathione transferases and cancer

Identification of the electrophilic substrate-binding site of glutathione S-transferase P by photoaffinity labeling

We determined the electrophilic substrate-binding site of rat glutathione S-transferase P (GST-P) by photoaffinity labeling using the photosensitive compound S-[2-(2-fluoro-4-nitrophenoxy)ethyl]glutathione. This photosensitive glutathione analogue inhibited the catalytic activity in a competitive manner against both glutathione and 1-chloro-2,4-dinitrobenzene, a putative electrophilic substrate. The enzyme kinetics indicated that the photoactivatable glutathione analogue was specifically bound at the active site, which consisted of glutathione-binding (G-site) and the electrophilic substrate-binding (H-site) regions. The procedure involved the following steps: S-[2-(2-fluoro-4-nitrophenoxy)ethyl]glutathione was photochemically reacted with a purified recombinant GST-P expressed in Escherichia coli using ultraviolet irradiation for 30 min on ice. After the reaction, only the GST-P complexed with the glutathione analogue was prepared with glutathione-immobilized agarose. The GST-P covalently bound with the analogue was digested with lysyl endopeptidase (Achromobacter protease I), and the peptides were separated by high-performance liquid chromatography. Only a single major peak with appreciable absorbance at 340 nm was observed by peptide mapping. The peptide was collected and analyzed using an automated peptide sequencer (ABI 477A). Amino acid sequence analysis showed that this peptide consisted of seven amino acid residues corresponding to the sequence at positions 122-128 of GST-P (Ala-Leu-Pro-Gly-Xaa-Leu-Lys). No appreciable phenylthiohydantoin-amino acid was detected at the fifth cycle, which indicated that His126 was chemically labeled with the photosensitive glutathione analogue. It was concluded that His126 was one of the amino acid residues forming the electrophilic substrate-binding site of GST-P.

Octopus S-crystallins with endogenous glutathione S-transferase (GST) activity: sequence comparison and evolutionary relationships with authentic GST enzymes

S-Crystallin is a major protein present in the lenses of cephalopods (octopus and squid). To facilitate the cloning of this crystallin gene, cDNA was constructed from the poly(A)+ mRNA of octopus lenses, and amplified by PCR for nucleotide sequencing. Sequencing of 10 of 15 positive clones coding for this crystallin revealed three distinct S-crystallin isoforms with 61-64% identity in nucleotide sequences and 42-58% similarity in amino acid sequences when compared with homologous crystallins in squid lenses. These charge-isomeric crystallins also show between 26 and 33% amino acid sequence identity to four major classes of glutathione S-transferase (GST), a major detoxification enzyme present in most mammalian tissues. For further analysis, expression of one of the S-crystallin cDNAs was carried out in the bacterial expression system pQE-30, and the S-crystallin protein produced in Escherichia coli was purified to homogeneity to determine the enzymic properties. We found that the expressed octopus S-crystallin possessed much lower GST activity than the authentic GSTs from other tissues. Sequence comparison and construction of phylogenetic trees for S-crystallins from squid and octopus lenses and various classes of GSTs revealed that S-crystallins represent a multigene family which is structurally related to Alpha-class GSTs and probably derived from the ancestral GST by gene duplication and subsequent multiple mutational substitutions.

Structural analysis of human alpha-class glutathione transferase A1-1 in the apo-form and in complexes with ethacrynic acid and its glutathione conjugate

BACKGROUND

Glutathione transferases (GSTs) constitute a family of isoenzymes that catalyze the conjugation of the tripeptide glutathione with a wide variety of hydrophobic compounds bearing an electrophilic functional group. Recently, a number of X-ray structures have been reported which have defined both the glutathione- and the substrate-binding sites in these enzymes. The structure of the glutathione-free enzyme from a mammalian source has not, however, been reported previously.

RESULTS

We have solved structures of a human alpha-class GST, isoenzyme A1-1, both in the unliganded form and in complexes with the inhibitor ethacrynic acid and its glutathione conjugate. These structures have been refined to resolutions of 2.5 A, 2.7 A and 2.0 A respectively. Both forms of the inhibitor are clearly present in the associated electron density.

CONCLUSIONS

The major differences among the three structures reported here involve the C-terminal alpha-helix, which is a characteristic of the alpha-class enzyme. This helix forms a lid over the active site when the hydrophobic substrate binding site (H-site) is occupied but it is otherwise disordered. Ethacrynic acid appears to bind in a non-productive mode in the absence of the coenzyme glutathione.

Pentobarbital-induced changes in Drosophila glutathione S-transferase D21 mRNA stability

The Drosophila glutathione S-transferase (gstD) genes are a family of divergently transcribed, intronless genes and pseudogenes. Under control conditions, the steady-state level of gstD1 mRNA is 20-fold higher than that of the gstD21 mRNA despite a lower transcription rate of the gstD1 gene. The GST D1 protein level is four times as abundant as the GST D21 protein. The gstD1 and gstD21 genes responded rapidly to pentobarbital (PB) as changes in mRNA levels were detectable within 30 min of treatment. Maximal induction of gstD1 and gstD21 resulted in 3-fold and 20-fold elevation of their respective mRNA levels. The major mechanism for the increase in gstD1 mRNAs appears to be transcriptional activation. The 2-fold increase in the rate of gstD21 transcription, however, cannot fully account for the 20-fold increase in the steady-state level of gstD21 mRNA. Therefore, post-transcriptional mechanism(s) should also be responsible for the increase of gstD21 mRNA by PB. Because the gstD21 mRNA is relatively unstable under control conditions, induction of the intronless gstD21 mRNA by PB occurs mainly at the level of enhanced mRNA stability. The GST D1 protein level in adult Drosophila was increased approximately 2-fold after PB treatment, whereas the GST D21 level remained relatively the same. Thus, an increase in gstD21 mRNA stability by PB treatment is probably coupled to a regulatory effect at the translational level.

Flow cytometric detection of glutathione S-transferase isoenzymes by quantitative immunofluorescence under nonsaturating conditions

The glutathione (GSH)-glutathione S-transferase (GST) detoxification system is an important element in cellular defence against injurious agents and anticancer drugs. GST isoenzymes may represent biochemical markers of neoplastic transformation, and, possibly, drug resistance is associated with altered GST-isoenzyme levels. The ability to measure GST-isoenzymes in cell populations would be useful for several biological and clinical applications. We have developed an immunofluorescence flow cytometric method for the simultaneous detection of different GST-isoenzymes and of DNA in fixed cells. Due to the impossibility of working under saturating conditions for the anti-GST antibody, a normalizing procedure was developed to permit quantitative analysis of single cells labelled with the anti-GST antibody at high dilution. A theoretical model and experimental data supported the use of this procedure. The method proposed is general and could be applied to other antibodies in order to obtain quantitative data outside saturating conditions. The method was challenged in different applications in order to compare it with other classical techniques. First, we characterized sublines resistant to different anticancer drugs with respect to variations of GST isotypes. In a second application, we studied the intercellular heterogeneity of GST content in mouse renal cells. In addition, GST was determined in aneuploid cells from solid tumor biopsies by separation from diploid cells on the basis of DNA content. Finally, GST distribution during cell-cycle progression was studied in two different cell lines by the biparametric analysis of GST/DNA.

Mass spectrometric analysis of rat liver cytosolic glutathione S-transferases: modifications are limited to N-terminal processing

Cytosolic glutathione S-transferases (GSTs) from rat livers were purified using an S-hexylglutathione affinity column. The GST subunits were resolved by reverse-phase HPLC and their molecular masses were determined by electrospray mass spectrometry. The major hepatic GSTs detected were subunits 1, 1', 2, 3 and 4, with molecular mass of 25,520, 25,473, 25,188, 25,782 and 25,571 Da respectively. Subunits 6, 7 and 10 are minor components, with molecular mass of 25,551, 23,308 and 25,211 Da respectively. Alternatively, the hepatic GSTs were purified using a glutathione affinity column. Subunits 1, 1', 2, 8 and 10 were eluted from this column with GSSG, the oxidized form of glutathione. Subunit 8 has a molecular mass of 25,553 Da. The remaining proteins on the glutathione affinity column were removed with glutathione and S-hexylglutathione. Subunits 2, 3, 4 and 6 could be detected in the eluate. We could not detect any significant difference in molecular mass between GSTs isolated from male and female rat livers. Cytosolic GSTs were isolated from livers of buthionine sulphoximine-treated female rats for MS analysis. The molecular masses obtained were identical to those determined for the controls.

Retinoids inhibit mammalian glutathione transferases

Affinity-purified cytosolic glutathione transferases from adult female rat liver, adult human liver and human term placenta were used. Of the 8 retinoids tested, all-trans retinoic acid was found to be the most potent inhibitor of placental glutathione transferase. The inhibition was non-competitive and exhibited Ki values of 20 and 41 microM for all-trans retinoic acid in the presence of varying concentrations of 1-chloro-2,4-dinitrobenzene and glutathione, respectively. Micromolar all-trans retinoic acid also caused significant (30-90%) inhibition of rat and human liver glutathione transferases. Taken together, the data suggest that inhibition of glutathione transferase(s) may represent yet another mechanism of retinoid action.

Glutathione S-transferases from larval Manduca sexta midgut: sequence of two cDNAs and enzyme induction

Two glutathione S-transferase (GST) clones from a larval midgut cDNA library of the tobacco hornworm, Manduca sexta were sequenced. The nucleotide sequence of the first clone, M. sexta GST1, encoded a protein of 217 amino acids with a predicted molecular weight of 24,644 and isoelectric point of 4.8. The M. sexta GST1 was 45.9-48.6% identical to GSTs from Musca domestica and several Drosophila species. The M. sexta GST2 cDNA encoded a protein of 203 amino acids with a predicted molecular weight of 23,596 and isoelectric point of 5.5. The M. sexta GST2 shared 44.8-50.0% sequence identity to a second cluster of insect GSTs from M. domestica, D. melanogaster and Anopheles gambiae. GST1 and GST2 were only 24.1% identical in amino acid sequence. The divergence of these two classes of insect GSTs occurred before the radiation of Diptera and Lepidoptera. Northern analysis of the expression of these GSTs showed increased GST1 mRNA levels in midguts of larvae fed diets containing 2-undecanone, or phenobarbital. Midgut and fat body cytosolic GST activities were induced when larvae were fed diets containing 2-tridecanone, 2-undecanone, or phenobarbital. Partial purification of midgut GSTs by size-exclusion and glutathione affinity chromatography resulted in a series of isoelectric focusing bands, with the major one corresponding to the predicted isoelectric point of the M. sexta GST1. In summary, two midgut GSTs have been identified on the basis of cDNA sequence and one of these, GST1, was inducible by dietary chemicals.

ARE- and TRE-mediated regulation of gene expression. Response to xenobiotics and antioxidants

Antioxidant response elements (AREs) containing 12-O-tetradecanoylphorbol-13-acetate response element (TRE) (perfect AP1) and TRE-like (imperfect AP1) elements mediate high basal transcription of the NAD(P)H:quinone oxidoreductase1 (NQO1) and glutathione S-transferase Ya genes in tumor cells and its induction in response to xenobiotics and antioxidants. Mutations in the human NQO1 gene ARE (hARE) revealed the requirement for two TRE or TRE-like elements arranged in inverse orientation at the interval of three base pairs and a GC box for optimal expression and beta-naphthoflavone induction of the NQO1 gene. A single TRE element from the human collagenase gene failed to respond to beta-naphthoflavone. These results demonstrate that ARE (2 x TRE or TRE-like elements)-containing detoxifying enzyme genes and not genes that contain 1 x TRE are responsive to xenobiotics and antioxidants. Bandshift assays showed shifting of a complex of more or less similar mobility with hARE and TRE that could be competed by each other. Mutations in the 3'-TRE of the NQO1 gene hARE eliminated binding of nuclear proteins to the hARE and resulted in the loss of basal and induced expression, indicating that 3'-TRE is the most important element within the hARE. 5'-TRE-like element within the NQO1 gene hARE is required for xenobiotic response but may not bind to the nuclear proteins by itself. The GC box located immediately following the 3'-TRE is required for optimal expression and induction of the NQO1 gene. The comparison of AREs from several different genes indicated the requirement for specific arrangement and spacing of two TRE and TRE-like elements within the AREs.

A glutathione depletion selectively imposed on mu glutathione S-transferase overproducing cells increases nitrogen mustard toxicity

Glutathione (GSH) contributes to the detoxification of anticancer drugs through the operation of specific glutathione S-transferases (GST) and innate, or acquired, overexpression of this enzyme family has been frequently observed in tumor cell lines. In the GMA32 line of Chinese hamster fibroblasts, we showed that GSH starvation produced by exposing cells to buthionine sulfoximine (BSO) increased the toxicity of chlorambucil and melphalan, but not that of N,N'-bis(2-chloroethyl)-N-nitrosourea (BCNU), cisplatine and doxorubicin. This indicates that efficient mechanisms of detoxification using GSH operate for chlorambucil and melphalan, but not for the other drugs in these cells. We then showed that GSH depletion could be selectively and transiently induced in the mu GST overexpressing cell line derived from GMA32, HC474, by exposing cells to substrates specific to the overexpressed isozyme. Exposing cells to such a substrate, trans-stilbene oxide, does not alter the sensibility of GMA32 cells to melphalan and chlorambucil, but increases that of HC474 cells to these drugs, to an extent comparable to that obtained with BSO. This observation highlights the possibility of exploiting GST overexpression, a frequent feature of tumor cells, to selectively sensitize these undesirable cells to anticancer drugs.

Modulation of cellular chemoresistance in keratinocytes by activation of different oncogenes

Response to chemotherapeutic agents in malignant tumors depends on many factors, most of which are as yet unknown. We investigated the correlation between the activation of different oncogenes and protein-kinase-C (PKC) modulation, and the cytotoxicity of some of the most widely used anti-cancer drugs. We transformed the murine keratinocyte cell line PAM 212, with different oncogenes (v-H-ras, v-myc and adenovirus E1a) and a mutant p53 suppressor gene (mp53). The cytotoxic effect of cisplatin (CDDP), doxorubicin (DOX) and vincristine (VCR), together with the concomitant action of modulators of PKC, TPA and staurosporine were evaluated by the crystal-violet method, thymidine incorporation and flow cytometry. We report that (a) the oncogene v-H-ras induces resistance to CDDP (> 50%), DOX (> 25%) and VCR (> 20%); (b) the E1a oncogene induces only resistance to VCR (> 40%) and marked sensitivity to CDDP and DOX; (c) the mp53 oncogene induces more resistance to VCR and insignificant resistance to the other drugs; and (d) activation of PKC by TPA increases the resistance to VCR and DOX in cells transformed by the v-H-ras, while it significantly increases the lethality with CDDP of the E1a-transformed cells. Staurosporine increases the cytoxicity of all the drugs, especially in the E1a-transformed keratinocytes. In the flow-cytometry analysis, the percentage of BUdR incorporation was related to sensitivity to anti-cancer drugs.

A molecular genetic approach for the identification of essential residues in human glutathione S-transferase function in Escherichia coli

The common substrate for glutathione S-transferases (GSTs), 1-chloro-2,4-dinitrobenzene (CDNB), is an inhibitor of Escherichia coli growth. This growth inhibition by CDNB is enhanced when E. coli expresses a functional GST. Cells under growth inhibition have reduced intracellular GSH levels and form filaments when they resume growth. Based on this differential growth inhibition by CDNB we have developed a simple procedure to select for null-mutants of a human GST in E. coli. Null mutations in the human GST gene from hydroxylamine mutagenesis or oligonucleotide-directed mutagenesis can be selected for on agar plates containing CDNB after transformation. The molecular nature of each mutation can be identified by DNA sequence analysis of the mutant GST gene. We have identified three essential amino acid residues in an alpha class human GST at Glu31, Glu96, and Gly97. Single substitution at each of these residues, E31K, E96K, G97D, resulted in mutant GST proteins with loss of CDNB conjugation activity and failure in binding to the S-hexyl GSH affinity matrix. In contrast, a mutant GST (Y8F) resulting from substitution of the conserved tyrosine near the N terminus has much reduced CDNB conjugation activity but was still capable of binding to the S-hexyl GSH-agarose. Additional mutant GSTs with substitutions at position 96 (E96F, E96Y) and 97 (G97P, G97T, G97S) resulted in changes in both Km and kcat to different extents. The in vitro CDNB conjugation activity of the purified mutant enzymes correlate negatively with the plating efficiencies of strains encoding them in the presence of CDNB. Based on the x-ray structure model of human GST 1-1, two of these residues are involved in salt bridges (Arg19-Glu31, Arg68-Glu96) and the third Gly97 is in the middle of the helix alpha 4. Our results provide evidence in vivo that Tyr8, Gly97, and the two salt bridges are important for GST structure-function. This molecular genetic approach for the identification of essential amino acids in GSTs should be applicable to any GSTs with CDNB conjugation activity. It should also complement the x-ray crystallographic approach in understanding the structure and function of GSTs.

Effect of ovariectomy and sex hormone replacement on glutathione and glutathione-related enzymes in rat hepatocarcinogenesis

The effects of ovariectomy and hormone replacement in control and carcinogen treated female rats were investigated by measuring whole blood and liver glutathione (WGSH, HGSH), glutathione S-transferase (GST), glutathione peroxidase (GPx), and glutathione reductase (GRx) and histological evaluation. Hepatocarcinogenesis was induced by diethylnitrosamine and 2-acetylaminofluorene. In control rats not receiving carcinogen, ovariectomy significantly increased the GST and GRx activities. Replacement with either estrogen or progesterone reduced the GST activities to below intact female values whereas replacement of both hormones together brought the GST activities to that of intact females. GRx activities were brought to intact female values by replacement with estrogen or progesterone, either singly or in combination. Neither ovariectomy nor sex hormone/s replacement influenced the levels of WGSH, HGSH and GPx activities. Carcinogen administration to intact rats increased all the parameters measured. Ovariectomized rats treated with carcinogen showed lower GPx and GRx activities at 2 mths. However, replacement with either progesterone or combined estrogen and progesterone increased GPx and GRx activities to original values. On the other hand GST and GPx activities in ovariectomized rats which had carcinogen treatment were lower than intact rats after 5 mths. Replacement with hormones either singly or both brought GST and GPx activities up to intact rat levels receiving carcinogen. The levels of WGSH, HGSH and GRx activities (5 mths) in carcinogen treated rats were not influenced by ovariectomy and/or hormone/s replacement. The results from this study suggested that ovariectomy reduced the severity of hepatocarcinogenesis which was restored by sex hormone/s replacement.

Soybean isoflavone extract suppresses early but not later promotion of hepatocarcinogenesis by phenobarbital in female rat liver

The antioxidant and anticarcinogenic activities of soybean isoflavone extracts were investigated in female F344/rats. Diethylnitrosamine (DEN, 15 mg/kg body wt) as a cancer initiator was injected intraperitoneally into 120 female F344/N rats at 10 days of age, and at weaning, phenobarbital (PB, 500 mg/kg diet) was fed to one-half of the rats. Soybean isoflavones were extracted in acetone-0.1 N HCl and analyzed by high-performance liquid chromatography, and two levels of soybean isoflavones (920 and 1,840 mumol/kg diet) were fed during PB treatment for 3 and 11 months. Control rats were fed diets without PB and with or without isoflavones. The effect of soybean isoflavone extract on hepatic glutathione peroxidase was measured, and development of gamma-glutamyltransferase (GGT)-positive (GGT+) and placental glutathione transferase (PGST)-positive (PGST+) altered hepatic foci (AHF) was analyzed by computerized stereology. Soybean isoflavone extract providing 920 or 1,840 mumol/kg diet normalized total heptic glutathione peroxidase activity, which was suppressed about 17% by PB (p < 0.05), and both doses of isoflavone extract suppressed PB promotion of hepatocarcinogenesis, decreasing the volume occupied by GGT+ and PGST+ AHF (p < 0.05) after three months. After 11 months of PB promotion, isoflavone extract at 920 mumol/kg diet decreased PGST+ AHF compared with the PB-fed group, but neither dose of isoflavone extract suppressed development of GGT+ AHF compared with the group fed PB alone. Furthermore the control group fed isoflavone extract at 1,840 mumol/kg diet showed greater development of GGT+ and PGST+ AHF than the group fed the basal diet alone. Therefore soybean isoflavones may be anticarcinogenic, but their margin of safety is relatively narrow, with a cancer-promoting dose of 1,840 mumol/kg in female F344/N rats initiated with DEN.

Activity of melphalan in combination with the glutathione transferase inhibitor sulfasalazine

Glutathione (GSH) transferases (GST), a family of detoxification enzyme proteins, are suggested to play an important role in tumor cell resistance to melphalan. The GST-activity inhibitor ethacrynic acid has been shown to increase the antitumor activity of melphalan in vitro as well as in vivo. In this study we determined the activity and toxicity of melphalan in combination with another GST-activity inhibitor, sulfasalazine, an agent used to treat ulcerative colitis. We entered 37 previously treated patients with advanced cancer of different histologies on sulfasalazine given at the individually calculated maximum tolerated dose (MTD) and melphalan given at doses beginning at 20 mg/m2. The main toxicity arising from this combination was nausea and vomiting, whereas increased myelosuppression was not observed. A partial response was seen in 2/4 of the ovarian cancer patients only. Plasma sulfasalazine levels varied between 2.5 and 47.1 micrograms/ml. Although reductions in GSH/GST levels were observed in peripheral mononuclear cells of certain patients following sulfasalazine treatment, there was no correlation between the extent of reduction and the plasma sulfasalazine level. A larger patient population must be studied to determine the usefulness of this combination.

Disease heterogeneity: does it impact our ability to detect dietary associations with breast cancer?

It is generally believed that breast cancer is a multistage process and that multiple and varying genetic events occur on the pathway to disease. We hypothesize that disease heterogeneity has an impact on our ability to identify risk factors. If a genetic alteration occurred in 50% of cases and a risk factor was associated only with that specific alteration, a risk estimate of 1.6 would be detected rather than the true risk estimate of 2.5 if analyses had been limited to those cases with the genetic alteration. Based on the literature we know that many genetic alterations occur in less than 50% of breast tumors. Thus, if environmental factors are related to some, but not all genetic alterations, we are decreasing our ability to identify potentially important risk factors. We therefore hypothesize that identification of dietary factors associated with breast cancer has been hampered by our inability to identify and capture the unique disease pathways which exist and contribute to the heterogeneity of common cancers such as breast cancer.

Kinetic analysis of the intracellular conjugation of monochlorobimane by IC-21 murine macrophage glutathione-S-transferase

Monochlorobimane (MCB) reacts with glutathione (GSH) in a reaction catalyzed by the glutathione-S-transferase (GST) isozymes. The diffusion of MCB through cell membranes is rapid and the fluorescence conjugates are relatively insensitive to quenching and to pH effects, and are expelled slowly from the cell, allowing the rate of fluorescence increase to be used to probe the dynamics of the intracellular reaction. Using low-light microscopic cytometry to monitor the initial rates of fluorescence increase for the GST-catalyzed reaction within IC-21 macrophages yields Vmax = 8.4 x 10(-16) mol s-1 cell-1 and KMCBm = 65 microM. Combining these data with an integrated Michaelis analysis of the reaction course yields KIP approximately 1.5 x 10(-5) M, and KmGSH approximately 3.0 x 10(-4) M (at [MCB] = 50 microM). The values of Vmax and KMCBm for the cell-free (extracellular) GST-catalyzed conjugation reaction are 1.2 x 10(-18) mol s-1 cell-1 and 3.1 microM, respectively. The values of Vmax for the intra- and extracellular conjugation reactions differ by 700-fold, suggesting the presence of an intracellular activator for this enzyme system.

Modification of glutathione S-transferase 3-3 mutants with 2-(S-glutathionyl)-3,5,6-trichloro-1,4-benzoquinone. Identification of the C-terminal tryptic fragment as part of the H-site and evidence that 2-(S-glutathionyl)-3,5,6-trichloro-1,4-benzoquinone is not specific for cysteine labelling

A triple mutant of rat liver glutathione S-transferase 3-3 that has all three cysteine residues replaced with serine (CallS) and a quadruple mutant with a Tyr-115 to phenylalanine substitution on CallS (CallSY115F) were reacted with 2-(S-glutathionyl)-3,5,6-trichloro-1,4-benzoquinone (GS-1,4-TCBQ). The modified proteins were analysed on a triple-quadrupole mass spectrometer equipped with an electrospray ionization source. At an enzyme: GS-1,4-TCBQ ratio of 1:10, the enzymes were modified at multiple sites. Covalent attachment of a single inhibitor on to the protein was achieved by lowering the enzyme: GS-1,4-TCBQ ratio to 1:1. Results from m.s. analyses suggest that the inhibitor on the CallSY115F mutant exists as a glutathionyl dichlorobenzoquinone derivative. The modifiers of the CallS mutants are glutathionyl monochlorobenzoquinone derivatives. Therefore, GS-1,4-TCBQ reacts at a single site on CallSY115F, but probably cross-links two regions on wild-type and CallS mutant. To confirm our observation, CallS was modified with 1-chloro2,4-dinitrobenzene, which specifically labels Tyr-115, before reacting with GS-1,4-TCBQ. The inhibitor formed a glutathionyl dichlorobenzoquinone adduct on the dinitrophenyl-CallS mutant. In addition, the benzoquinone derivative on the protein can be partially removed by 1-chloro-2,4-dinitrobenzene. Peptide mapping and sequencing analysis of the GS-1,4-TCBQ-modified CallS mutant revealed that the C-terminal 16-amino-acid fragment is labelled. Molecular modelling suggests the C(5) and C(6) on the benzoquinone ring of the inhibitor interact with the oxygen atoms of Tyr-115 and Ser-209 respectively.

Modifications of the anti-oxidant metabolism during proliferation and differentiation of colon tumor cell lines

The anti-oxidant metabolism was studied at different times after sub-culture in 2 colon cell lines previously characterized for their growth and differentiation properties. The HT29 cell line is mainly composed of proliferative and undifferentiative cells, while the derived 5-fluorouracil (FUra)-adapted cells undergo growth-dependent differentiation, which is complete at post-confluence. In the 2 cell lines, all the anti-oxidant parameters studied appeared to be related to proliferation, with increased activity of superoxide dismutase (SOD) 1 and 2, catalase (CAT), glutathione peroxidase (GPX), glutathione reductase (GSR), and glutathione transferase (GST), and decreased glucose-6-phosphate dehydrogenase (G6PD) activity and glutathione content, in parallel with slowing down of proliferation. At post-confluence, these metabolic parameters remained stable, except for GPX activity, which continued to increase, and CAT activity, which decreased. The amounts of SOD1, SOD2 and CAT immunoreactive proteins, estimated by Western blotting, appeared to be correlated to their respective enzymatic activities. SOD1, CAT and GST activity and glutathione content, which remained at similar levels in the 2 cell lines for all times studied, appeared unrelated to the differentiation process. GSR and GPX activity, which was lower in FUra-adapted than in parental cells only at post-confluence, could be considered as markers of differentiated cells. The higher SOD2 and lower G6PD activity observed in FUra-resistant cell in comparison with parental cells at all times after sub-culture could be characteristic both of differentiative and of differentiated cells. Interestingly, cytogenetics have previously indicated that deletions of the long arm of chromosome 6, which carry the gene for SOD2, were frequently observed in parental but not in FUra-adapted cells. These results demonstrate that modifications of the anti-oxidant metabolism occur in relation with proliferation and differentiation, and suggest a particular role for SOD2 in these cellular processes.

Three-dimensional structure of Schistosoma japonicum glutathione S-transferase fused with a six-amino acid conserved neutralizing epitope of gp41 from HIV

The 3-dimensional crystal structure of glutathione S-transferase (GST) of Schistosoma japonicum (Sj) fused with a conserved neutralizing epitope on gp41 (glycoprotein, 41 kDa) of human immunodeficiency virus type 1 (HIV-1) (Muster T et al., 1993, J Virol 67:6642-6647) was determined at 2.5 A resolution. The structure of the 3-3 isozyme rat GST of the mu gene class (Ji X, Zhang P, Armstrong RN, Gilliland GL, 1992, Biochemistry 31:10169-10184) was used as a molecular replacement model. The structure consists of a 4-stranded beta-sheet and 3 alpha-helices in domain 1 and 5 alpha-helices in domain 2. The space group of the Sj GST crystal is P4(3)2(1)2, with unit cell dimensions of a = b = 94.7 A, and c = 58.1 A. The crystal has 1 GST monomer per asymmetric unit, and 2 monomers that form an active dimer are related by crystallographic 2-fold symmetry. In the binding site, the ordered structure of reduced glutathione is observed. The gp41 peptide (Glu-Leu-Asp-Lys-Trp-Ala) fused to the C-terminus of Sj GST forms a loop stabilized by symmetry-related GSTs. The Sj GST structure is compared with previously determined GST structures of mammalian gene classes mu, alpha, and pi. Conserved amino acid residues among the 4 GSTs that are important for hydrophobic and hydrophilic interactions for dimer association and glutathione binding are discussed.

Allele-specific monoclonal antibodies against glutathione S-transferase Mu1-1

IgG1 class mouse monoclonal antibodies (MAbs) were produced against human glutathione S-transferase Mu1-1 (GSTMu1-1). Eight MAbs of 16 are able to recognize only the native form of the enzyme; 4 MAbs bind to native and denaturated enzyme, and the remaining 4 can bind only to partially denatured antigen in direct ELISA or Western blot. The antibodies recognizing the native form of the enzyme bind to six different epitopes. Three overlapping epitopes are responsible for specific binding of MAbs to different allelic variants of GSTMu1-1. Three allele-specific antibodies, 2E1, 11F12, and 7D11, bind to GSTM1a monomer and the other two, 1H8 and 3H10, recognize GSTM1b monomer.

Cytidine methylation of regulatory sequences near the pi-class glutathione S-transferase gene accompanies human prostatic carcinogenesis

Hypermethylation of regulatory sequences at the locus of the pi-class glutathione S-transferase gene GSTP1 was detected in 20 of 20 human prostatic carcinoma tissue specimens studied but not in normal tissues or prostatic tissues exhibiting benign hyperplasia. In addition, a striking decrease in GSTP1 expression was found to accompany human prostatic carcinogenesis. Immunohistochemical staining with anti-GSTP1 antibodies failed to detect the enzyme in 88 of 91 prostatic carcinomas analyzed. In vitro, GSTP1 expression was limited to human prostatic cancer cell lines containing GSTP1 alleles with hypomethylated promoter sequences; a human prostatic cancer cell line containing only hypermethylated GSTP1 promoter sequences did not express GSTP1 mRNA or polypeptides. Methylation of cytidine nucleotides in GSTP1 regulatory sequences constitutes the most common genomic alteration yet described for human prostate cancer.

Human antioxidant-response-element-mediated regulation of type 1 NAD(P)H:quinone oxidoreductase gene expression. Effect of sulfhydryl modifying agents

Human antioxidant-response element (hARE) containing two copies of the AP1/AP1-like elements arranged as inverse repeat is known to mediate basal and beta-naphthoflavone-induced transcription of the type 1 NAD(P)H:quinone oxidoreductase (NQO1) gene. Band-shift assays revealed that beta-naphthoflavone increased binding of nuclear proteins at the hARE. Super shift assays identified Jun-D and c-Fos proteins in the band-shift complexes observed with control and beta-naphthoflavone-treated Hepa-1 nuclear extracts. Hepa-1 cells stably transformed with hARE-tk-chloramphenicol acetyl transferase (CAT) recombinant plasmid were used to demonstrate that, in addition to beta-naphthoflavone, a variety of antioxidants, tumor promoters and hydrogen peroxide (H2O2) also increased expression of hARE-mediated CAT gene. beta-naphthoflavone induction of the CAT gene expression in Hepa-1 cells was found insensitive to inhibitors of protein kinase C and tyrosine kinases. However, binding of regulatory proteins at the hARE and the CAT gene expression in Hepa-1 cells were increased by dithiothreitol, 2-mercaptoethanol and diamide. Treatment of the Hepa-1 cells with N-ethylmaleimide reduced binding of proteins at the hARE and interfered with expression and beta-naphthoflavone induction of the CAT gene. These results suggested a role of sulfhydryl modification of hARE binding (Jun and Fos) proteins which mediate basal and induced expression of the NQO1 gene. We also report that in-vitro-translated products of the proto-oncogenes, Jun and Fos, bind to the hARE in band-shift assays. The incubation of Jun and Fos proteins with small amounts of nuclear extract from dimethylsulfoxide-treated (control) or beta-naphthoflavone treated Hepa-1 cells prior to band-shift assays increased the binding of Jun and Fos proteins to the hARE. Interestingly, the increase in binding of Jun and Fos proteins to the hARE was more prominent with beta-naphthoflavone-treated nuclear extract as compared to the control nuclear extract. In addition, incubation of control nuclear extract with beta-naphthoflavone, microsomes and NADPH increased the binding of Jun and Fos proteins to the hARE. Evidence from in vitro studies indicate the presence of unknown nuclear factor(s) that receive signals from metabolites of beta-naphthoflavone and modulate Jun and Fos binding to the AP1 site contained within the hARE.

A novel type of glutathione S-transferase in Onchocerca volvulus

Onchocerca volvulus is a pathogenic human filarial parasite which, like other helminth parasites, is capable of evading the host's immune responses by a variety of defense mechanisms which are likely to include the detoxification and repair mechanisms of the enzyme glutathione S-transferase (GST). In this study, we show that one of the previously described GSTs from O. volvulus appears to possess the characteristics of a secreted enzyme. When the complete O. volvulus GST1 (OvGST1) sequence presented here is compared with those of other GSTs, 50 additional residues at the N terminus are observed, the first 25 showing characteristics of a signal peptide. This is consistent with the N-terminal sequence data on the native mature enzyme which begins at amino acid 26, based on the deduced protein sequence from the cDNA. The native protein, without the signal peptide sequence, possesses a 24-amino-acid extension not present in other GSTs. The deduced amino acid sequence of the OvGST1 cDNA clone was shown to possess four potential N-glycosylation sites. Digestion of O. volvulus homogenate with endoglycosidase, followed by detection of OvGST1 with specific antibody, indicated that the enzyme possesses at least two N-linked oligosaccharide chains. Gel filtration of the Escherichia coli-produced recombinant OvGST1 showed that it is enzymatically active as a nonglycosylated dimer. OvGST1 is found in the media surrounding adult worms maintained in culture, indicating that, in vitro, this enzyme is released from the worm. The strongest immunostaining for OvGST1 was observed in the outer cellular covering of the adult worm body, the syncytial hypodermis, especially in the interchordal hypodermis, where the peripheral membrane forms a series of lamellae which run into the outer zone of the hypodermal cytoplasm.

Hot-spot p53 mutants interact specifically with two cellular proteins during progression of the cell cycle

Inactivation of both alleles of the p53 gene is commonly found in human cancers. In contrast to mutations of the retinoblastoma gene, certain altered forms of p53 gain growth-promoting functions. To explore the mechanisms underlying this gain of function, we have identified two nuclear proteins, with molecular masses of 42 and 38 kDa, respectively, that are specifically associated with p53 mutated within the simian virus 40 T-antigen-binding domain, "hot spots" found in many human tumors. These mutants transactivate the multiple-drug resistance gene promoter and cause cells to grow to higher density. Both the mutated p53 complex with p42 and p38 increase when cells enter S phase of the cell cycle but decrease in G1 and M phases, suggesting that they may have a role in promoting cell growth.

P-glycoprotein-mediated multidrug resistance in normal and neoplastic hematopoietic cells

The multidrug transporter, P-glycoprotein (P-gp), is expressed by CD34-positive bone marrow cells, which include hematopoietic stem cells, and in other cells in the bone marrow and peripheral blood, including some lymphoid cells. Multidrug resistance mediated by P-gp appears to be a major impediment to successful treatment of acute myeloid leukemias and multiple myelomas. However, the impact of P-gp expression on prognosis has to be confirmed in several other hematopoietic neoplasms. The role of P-gp in normal and malignant hematopoiesis and clinical attempts to circumvent multidrug resistance in hematopoietic malignancies are reviewed. The recent transduction of the MDR1 gene into murine hematopoietic cells, which protects them from toxic effects of chemotherapy, suggests that MDR1 gene therapy may help prevent myelosuppression following chemotherapy.

Immunohistochemical localization of glutathione S-transferases in sarcomas

The glutathione S-transferases (GSTs) are a multi-gene family of enzymes involved in detoxifying electrophilic compounds and the expression of these enzymes in tumours has been proposed as one important mechanism of anti-cancer drug resistance. In this study, the localization of the major cytoplasmic forms of GST has been studied in soft tissue sarcomas by immunohistochemistry. The alpha, mu, and pi forms of GST were identified in 59, 68, and 51 per cent of tumours, respectively. In addition, GST pi immunoreactivity was consistently identified in fibroblasts in adjacent non-neoplastic tissue. The presence of specific forms of GST in soft tissue sarcomas may contribute to the drug resistance frequently observed in these tumours.

Hot-spot p53 mutants interact specifically with two cellular proteins during progression of the cell cycle

Inactivation of both alleles of the p53 gene is commonly found in human cancers. In contrast to mutations of the retinoblastoma gene, certain altered forms of p53 gain growth-promoting functions. To explore the mechanisms underlying this gain of function, we have identified two nuclear proteins, with molecular masses of 42 and 38 kDa, respectively, that are specifically associated with p53 mutated within the simian virus 40 T-antigen-binding domain, "hot spots" found in many human tumors. These mutants transactivate the multiple-drug resistance gene promoter and cause cells to grow to higher density. Both the mutated p53 complex with p42 and p38 increase when cells enter S phase of the cell cycle but decrease in G1 and M phases, suggesting that they may have a role in promoting cell growth.

Comparative study of intracellular calcium and adenosine 3',5'-cyclic monophosphate levels in human breast carcinoma cells sensitive or resistant to Adriamycin: contribution to reversion of chemoresistance

Multidrug resistance (MDR) corresponds to the cross-over resistance of tumour cells to structurally unrelated cytotoxic chemotherapeutic drugs. One of the mechanisms causing this resistance is the enhanced expression of a transmembrane drug efflux pump P-glycoprotein (P-170). Reversal of P-glycoprotein-associated MDR has received much attention in recent years. In experimental cell lines, P-170 and the glutathione redox cycle seem to contribute to this phenomenon; P-170 may be inactivated by calcium and calmodulin antagonists and the glutathione redox cycle altered by buthionine sulphoximine (BSO). Treatment of human MCF-7 breast cancer cells with chemosensitizers (CS), such as verapamil, trifluoperazine or BSO, for 72 hr resulted in an enhanced sensitization of cells to Adriamycin, trifluoperazine being the most potent compound in the reversion of chemoresistance. In these Adriamycin sensitive or resistant cells, treated or not by the CS, the possible role of calcium and cyclic adenosine monophosphate (cAMP) in mediating the reversion of chemoresistance to Adriamycin was investigated. It was found that intracellular calcium was approximately 2-fold higher in resistant than in sensitive cells, the opposite was true for cAMP. Modifications in calcium and cAMP levels were observed in MCF-7 resistant cells after treatment with verapamil and BSO; trifluoperazine had no effect on these two parameters. These results seemed to rule out any implication of calcium and cAMP levels in the contribution of these three chemosensitizers in the mechanisms of reversion of chemoresistance to Adriamycin.

Modulation of cisplatin cytotoxicity by sulphasalazine

The efficacy of cisplatin [cis-diamminedichloroplatinum (II); DDP] is hampered by acquired or de novo resistance of malignant cells to its cytotoxic effects. We have previously reported that cisplatin resistance parallels glutathione S-transferase (GST) activity in several human small-cell lung cancer cell lines. In the presently described studies, we used sulphasalazine, an inhibitor of GSTs, to evaluate the relative role of GSTs in mediating cisplatin resistance in two human small-cell lung cancer cell lines, NCI H-69 and H-2496. The H-69 cell line, which contained relatively higher GST activity than the H-2496 cell line (317 +/- 7 vs 9 +/- 1 mU mg-1 protein respectively), also displayed a greater degree of cisplatin resistance (IC50 values of 25.0 +/- 3.9 vs 4.5 +/- 1.0 microM respectively). Western blot and Northern blot analyses of purified GSTs revealed the expression of only the pi-class GST in both cell lines. Sulphasalazine inhibited the purified GSTs (IC50 of 10 microM for H-69 and 12 microM for H-2496) from both lines in a competitive manner with similar Ki values (6.5 and 7.9 microM for the H-69 and H-2496 cell lines respectively). Cytotoxicity studies revealed that sulphasalazine increased the cytotoxicity of cisplatin towards both cell lines. Isobologram analysis showed that sulphasalazine synergistically enhanced the cytotoxicity of cisplatin towards both cell lines, the magnitude of synergy being remarkably higher in H-69 cells than in H-2496 cells. Our studies indicate that clinically achievable concentrations of sulphasalazine may be useful in modulating cisplatin resistance in malignancies with increased GST-pi content.

Inhibition of mammalian hepatic glutathione S-transferases by acetylenic fatty acids

Micromolar concentrations of 2 classical inhibitors of lipoxygenase, 5,8,11-eicosatriynoic acid (ETI) and 5,8,11,14-eicosatetraynoic acid (ETYA), were found to cause a significant inhibition of the mixture of isozymes of affinity purified rat and human liver glutathione S-transferase (GST) with activity towards 1-chloro-2,4-dinitrobenzene (CDNB). ETI was a more potent inhibitor of both rat and human liver GST than ETYA. Analysis of kinetic data suggested noncompetitive inhibition of human liver GST by ETI towards reduced glutathione and CDNB. ETI also inhibited the hepatic GST activity towards 1,2-dichloro-4-nitrobenzene, p-nitrobenzyl chloride and 4-nitropyridine N-oxide.

Cytochrome P450 expression in oesophageal cancer

The cytochrome P450 superfamily of enzymes play a central part in the metabolism of carcinogens and anti-cancer drugs. The expression, cellular localisation, and distribution of different forms of P450 and the functionally associated enzymes epoxide hydrolase and glutathione S-transferases have been investigated in oesophageal cancer and non-neoplastic oesophageal tissue using immunohistochemistry. Expression of the different enzymes was confined to epithelial cells in both non-neoplastic samples and tumour samples except the CYP3A was also identified in mast cells and glutathione S-transferase pi was present in chronic inflammatory cells. CYP1A was present in a small percentage of non-neoplastic samples but both CYP2C and CYP3A were absent. Epoxide hydrolase was present in half of the non-neoplastic samples and the different classes of glutathione S-transferase were present in a low number of samples. In carcinomas CYP1A, CYP3A, epoxide hydrolase, and glutathione S-transferase pi were expressed in at least 60% of samples. The expression of glutathione S-transferases alpha and mu were significantly less in adenocarcinoma compared with squamous carcinoma.

Purification, molecular cloning and heterologous expression of a glutathione S-transferase from the Australian sheep blowfly (Lucilia cuprina)

Three glutathione S-transferases from Lucilia cuprina (Australian sheep blowfly) pupae were purified by affinity chromatography and anion-exchange chromatography. One isoenzyme was composed of M(r)-24,800 subunits, and two isoenzymes had subunits of M(r) 23,900. The M(r)-23,900 subunits showed immunological identity and were immunologically distinct from the M(r)-24,800 subunits. All three enzymes were active with the substrate 1-chloro-2,4-dinitrobenzene and had low activity with 1,2-dichloro-4-nitrobenzene. A cDNA clone encoding a M(r)-23,900 subunit (LuGST1) was isolated and sequenced. The sequence has close similarities (> 81%) to that of GSTs from the fruitfly Drosophila melanogaster and Musca domestica (housefly). The deduced amino acid sequence of the Lu GST1 subunit showed no significant similarity to that of the mammalian GSTs to the Alpha, Mu and Pi classes, but shows some similarity (33%) over the first 100 residues with the rat subunit 12 Theta-class GST. Southern blots of genomic DNA hybridized with the LuGST1 cDNA identified many hybridizing fragments. Taken together, these data indicated that the L. cuprina genome contains multiple glutathione S-transferase genes.

ocs element promoter sequences are activated by auxin and salicylic acid in Arabidopsis

ocs elements are a group of promoter elements that have been exploited by two distinct groups of plant pathogens, Agrobacterium and certain viruses, to express genes in plants. We examined the activity of single and multiple ocs elements linked to a minimal plant promoter and the uidA reporter gene in transgenic Arabidopsis. beta-Glucuronidase activity was detected only in root tips and in callus tissue after auxin treatment. A more sensitive assay revealed that auxin treatment also increased ocs element activity in aerial parts of the plant, although the absolute levels of ocs element activity were greater in roots. The response of ocs elements to exogenous auxin began within 1 h. Salicylic acid, a disease-resistance signal in plants, also increased ocs element activity in both roots and aerial parts of the plant. The question of whether the induction in ocs element activity is mediated through auxin and/or salicylic acid signal transduction pathways or is part of a more general stress response is discussed.

Interaction of ebselen with glutathione S-transferase and papain in vitro

The interaction of ebselen(2-phenyl-1,2-benzisoselenazol-3(2H)-one) with rat liver cytosolic glutathione S-transferases (GSTs) and the plant cysteine protease, papain, was studied as cysteine residues are important for the activity of these enzymes. The capacity of GST 1-2 and 3-4 for ebselen binding is similar (1.5 mol ebselen/mol GST isozyme), while GST 2-2 and GST 7-7 bind 0.3 and more than 2.0 mol ebselen/mol GST isozyme, respectively. Ebselen does not bind to N-ethylmaleimide-treated GST, and its binding to GST is prevented by 5 mM thiols. Ebselen irreversibly inactivates the different GST isozymes with a second order rate constant ranging from 20 to 2250 M-1 sec-1 for the different subunits. GST inhibition by ebselen is partially restored by 5 mM thiols. Ebselen binds to untreated papain and to cysteine-treated papain at a ratio of about 0.1 and 0.75 mol ebselen/mol papain, respectively. Ebselen does not bind to N-ethylmaleimide-treated papain, and its binding to papain is interfered with by added thiols. Papain is inactivated by ebselen with a second order rate constant of 1800 M-1 sec-1 in the absence of thiols. However, in the presence of GSH, 2-mercaptoethanol or sodium borohydride, ebselen exerts an activating effect on papain. The binding of ebselen by a seleno-sulfide bond to cysteine residues of GSTs and papain leads to their inactivation.

X-ray crystal structures of cytosolic glutathione S-transferases. Implications for protein architecture, substrate recognition and catalytic function

Crystal structures of cytosolic glutathione S-transferases (EC 2.5.1.18), complexed with glutathione or its analogues, are reviewed. The atomic models define protein architectural relationships between the different gene classes in the superfamily, and reveal the molecular basis for substrate binding at the two adjacent subsites of the active site. Considerable progress has been made in understanding the mechanism whereby the thiol group of glutathione is destabilized (lowering its pKa) at the active site, a rate-enhancement strategy shared by the soluble glutathione S-transferases.

Crystallization and preliminary X-ray diffraction studies of a glutathione S-transferase from the Australian sheep blowfly, Lucilia cuprina

Crystals of a glutathione S-transferase from the Australian sheep blowfly Lucilia cuprina have been grown from ammonium sulphate by the hanging drop vapour diffusion method. Successful crystallization required the presence of the inhibitor S-hexylglutathione. The crystals belong to the tetragonal space group P4(1)22 (or P4(3)22) with cell dimensions of a = b = 88.1 A and c = 66.9 A. They contain one monomer in the asymmetric unit and diffract beyond 2.8 A resolution.

Human glutathione S-transferases

1. Multiple forms of glutathione S-transferase (GST) isoenzymes present in human tissues are dimers of subunits belonging to three distinct gene families namely alpha, mu and pi. Only the subunits within each class hybridize to give active dimers. 2. These subunits are differentially expressed in a tissue-specific manner and the composition of glutathione S-transferases in various tissues differs significantly. 3. Minor GST subunits not belonging to these three classes are also present in some tissues. 4. An ortholog of rat GST 8-8 and mouse mGSTA4-4 is selectively expressed in some human tissues including bladder, brain, heart, liver, and pancreas. This isoenzyme designated as GST 5.8 expresses several fold higher activity towards 4-hydroxy-2,3-trans-nonenal as compared to the routinely used substrate 1-chloro-2,4-dinitrobenzene.

Adenosine triphosphate-dependent transport of doxorubicin, daunomycin, and vinblastine in human tissues by a mechanism distinct from the P-glycoprotein

Previous studies have demonstrated that a human glutathione conjugate transporter, designated as dinitrophenyl-S-glutathione ATPase (DNP-SG ATPase), catalyzed ATP hydrolysis in the presence of several amphiphilic compounds other than glutathione conjugates (Singhal, S. S., R. Sharma, S. Gupta, H. Ahmad, P. Zimniak, A. Radominska, R. Lester, and Y. C. Awasthi. 1991. FEBS [Fed. Eur. Biochem. Soc.] Lett. 281:255-257). We now demonstrate that DNP-SG ATPase purified from human lung and erythrocyte membranes catalyzed the hydrolysis of ATP in the presence of doxorubicin and its metabolites. Doxorubicin-stimulated ATP hydrolysis by DNP-SG ATPase was saturable with respect to doxorubicin (Km 1.2 and 2.8 microM for the lung and erythrocyte enzymes, respectively). Antibodies against DNP-SG ATPase immunoprecipitated the ATP hydrolyzing activity stimulated by doxorubicin, its metabolites, and glutathione conjugates. Inside our vesicles prepared from erythrocyte membranes took up doxorubicin, daunomycin, and vinblastine in an ATP-dependent manner. The uptake was linear with respect to time and vesicle protein, was dependent on ATP and magnesium, was inhibited by heavy metal salts or by heating the vesicles, and was sensitive to both osmolarity and orientation of the vesicles. The transport had an activation energy of 13 kcal/mol, was saturable with respect to both doxorubicin and ATP (Km values of 1.8 microM and 1.9 mM, respectively), and was competitively inhibited by glutathione conjugates as well as by a number of amphiphiles such as daunomycin or vinblastine. Transport was diminished upon coating the vesicles with antibodies against DNP-SG ATPase. Incorporation of increasing amounts of purified DNP-SG ATPase into the vesicles resulted in a linear increase in transport of doxorubicin. These studies demonstrated for the first time that a membrane protein that catalyzed the transport of anionic amphiphilic molecules such as glutathione conjugates could also mediate the transport of weakly cationic antitumor antibiotic, doxorubicin. Notably, the Km of transport was in the range of doxorubicin concentration achievable in human serum after intravenous dosing of doxorubicin.

A novel glutathione S-transferase isozyme similar to GST 8-8 of rat and mGSTA4-4 (GST 5.7) of mouse is selectively expressed in human tissues

A mouse glutathione S-transferase (GST) isozyme designated as GST 5.7 or mGSTA4-4 belongs to a distinct subclass of the alpha-class isozymes of GST. It is characterized by kinetic properties intermediate between the alpha- and pi-classes of GSTs. We have recently cloned and expressed this isozyme (rec-mGSTA4-4) in E. coli and have reported its complete primary sequence (Zimniak, P., et al. (1992) FEBS Lett., 313, 173-176). Using antibodies raised against the homogenous rec-mGSTA4-4 expressed in E. coli, we now demonstrate that an ortholog of this isozyme was selectively expressed in various human tissues. The human ortholog of mGST A4-4 purified from liver had a pI value of 5.8 and constituted approx. 1.7% of total GST protein of human liver. Similar to other alpha-class GSTs, the N-terminus of this isozyme (GST 5.8) was also blocked. CNBr digestion of the enzyme yielded two major fragments with M(r) values of 12 kDa and 6 kDa. The sequences of these two fragments showed identities in 16 out of 20 residues and 17 out of 20 residues with the corresponding sequences of its mouse ortholog (mGSTA4-4), and showed significant homologies with the rat and chicken orthologs, GST 8-8 and GST CL3. Human liver GST 5.8 showed more than an order of magnitude higher activity towards t-4-hydroxy-2-nonenal as compared to 1-chloro-2,4-dinitrobenzene. This isozyme also expressed glutathione-peroxidase activity towards fatty acid, as well as phospholipid hydroperoxides suggesting its role in protection mechanisms against the toxicants generated during lipid peroxidation. Western blot analysis of human tissues revealed that this GST isozyme was selectively expressed in human liver, pancreas, heart, brain and bladder tissues, but absent in lung, skeletal muscle, spleen and colon.

Characterization of quinone reductase, glutathione and glutathione S-transferase in human myeloid cell lines: induction by 1,2-dithiole-3-thione and effects on hydroquinone-induced cytotoxicity

In this study, we have characterized quinone reductase (QR), glutathione (GSH), glutathione S-transferase (GST) and their induction by a chemoprotector, 1,2-dithiole-3-thione (D3T), in the human myeloid cell lines ML-1 and HL-60. In addition, we also examined the toxicity of hydroquinone (HQ), a benzene metabolite, to these two cell lines. Both of the cell lines contain a basal level of cellular GSH, which is similar in the two cell lines. Although ML-1 cells contain much higher QR specific activity than HL-60 cells, which are relatively QR deficient, the GST specific activity of ML-1 cells is 1.8 times less than that of HL-60 cells. Immunoblot experiments showed that the GST in these two cell lines is GST pi. In addition, HL-60 cells exhibit 4.5 times more myeloperoxidase specific activity than ML-1 cells. Inclusion of D3T in the cultures could induce significant increases in cellular GSH content and QR activity, but not GST activity in either cell line. Treatment with HQ caused both inhibition of cell proliferation and loss of cell viability in these two myeloid cell lines. HQ treatment also resulted in a significant depletion of cellular GSH, which preceded the loss of cell viability. Pretreatment of both cell lines with buthionine sulfoximine, an inhibitor of GSH biosynthesis, markedly increased HQ-induced toxicity. In contrast, the presence of dicumarol, a QR inhibitor, failed to potentiate HQ-induced toxicity in ML-1 cells. On the other hand, pretreatment of these two myeloid cell lines with D3T significantly protected against HQ-induced inhibition of cell proliferation and cell death. Therefore, the above results suggest that GSH but not QR is an important factor involved in the toxicodynamics of HQ in these myeloid cells.