O 6-Alkylguanine-DNA Alkyltransferase Has Opposing Effects in

742

Chem. Res. Toxicol. 2004, 17, 742-752

O6-Alkylguanine-DNA Alkyltransferase Has Opposing Effects in Modulating the Genotoxicity of Dibromomethane and Bromomethyl Acetate Liping Liu,†,‡ Kevin M. Williams,§,| F. Peter Guengerich,§ and Anthony E. Pegg*,† Department of Cellular and Molecular Physiology, The Pennsylvania State University College of Medicine, Hershey, Pennsylvania 17033, and Department of Biochemistry and Center in Molecular Toxicology, Vanderbilt University School of Medicine, Nashville, Tennessee 37232 Received January 25, 2004

O6-Alkylguanine-DNA alkyltransferase (AGT) is a DNA repair protein that removes O6alkylguanine adducts. The interaction of dibromomethane (CH2Br2) and bromomethyl acetate (BrCH2OAc) with AGT was studied in vitro, and the effect of AGT on their toxicity and mutagenicity was investigated using Escherichia coli strain TRG8 (lacking endogenous AGT) that expressed human AGT or its inactive C145A mutant. Both CH2Br2 and BrCH2OAc reacted with AGT at its cysteine acceptor site, abolishing its DNA repair activity with the latter agent being much more potent. The formation of AGT-Cys145S-CH2OAc by BrCH2OAc was confirmed by mass spectral analysis, but the presumed AGT-Cys145S-CH2Br adduct from CH2Br2 was too unstable for such characterization. In the presence of CH2Br2, AGT was covalently crosslinked to an oligodeoxyribonucleotide, 5′-d(AG)8-3′, but no cross-link was formed by BrCH2OAc. Survival of cells exposed to CH2Br2 was reduced, and the number of mutants was greatly increased when wild-type AGT was present. The cytotoxicity of CH2Br2 was similar to that of BrCH2CH2Br2, but the mutagenicity was about four times less. Virtually all of the AGTmediated mutants induced by CH2Br2 in the rpoB gene were at G:C sites with equal numbers of transitions to A:T and transversions to T:A. In contrast, BrCH2OAc was more than 10-fold less genotoxic than CH2Br2 and the survival of cells exposed to BrCH2OAc was not affected by AGT. The number of mutations (almost all G:C to A:T transitions) induced by BrCH2OAc was slightly reduced by the presence of wild-type AGT and substantially increased by the inactive C145A mutant. These results with CH2Br2 are consistent with a mechanism in which reaction at the active site Cys145 residue followed by attack of AGT-Cys145S-CH2Br at guanine in DNA forms a covalent adduct, which leads to cytotoxicity and to mutagenicity. The results with BrCH2OAc suggest that it reacts directly with DNA to form O6-(CH2OAc)guanine, which, if unrepaired, causes G:C to A:T transitions. Our experiments reveal two novel pathways (direct inactivation of AGT and formation of AGT-Cys145S-CH2-DNA adducts) by which CH2Br2 may cause damage to the genome in addition to the well-recognized pathway involving activation by GSTs.

Introduction Halogenated aliphatic hydrocarbons have been employed extensively as pesticides, solvents, and chemical intermediates, but concerns about their genotoxic effects including mutagenicity and carcinogenicity have curtailed recent use. Dihalomethanes are of interest, particularly dichloromethane (CH2Cl2). The annual U.S. production of CH2Cl2 is currently about 3 × 108 kg (1, 2). High doses of CH2Cl2 have been shown to produce liver and lung tumors in mice (3-6). These and other findings have led to regulation of the limits of human exposure to CH2Cl2 and to considerable debate about the risk of * To whom correspondence should be addressed. Tel: 717-531-8152. Fax: 717-531-5157. E-mail: [email protected]. † The Pennsylvania State University College of Medicine. ‡ Present address: Abramson Family Cancer Research Institute, University of Pennsylvania, 421 Curie Blvd., 438 BRB II/III, Philadelphia, PA 19104. § Vanderbilt University School of Medicine. | Present address: Department of Chemistry, Western Kentucky University, Thompson Complex North Wing 329, 1 Big Red Way, Bowling Green, KY 42101.

this commodity chemical (7, 8). Dibromomethane (CH2Br2) is known to be both mutagenic and carcinogenic (9, 10). CH2Br2 is of interest as a model for CH2Cl2, in the context of mechanistic toxicology (11), and as an environmental contaminant and toxicant itself. CH2Br2 has been detected in drinking water supplies (12). One source is water disinfection procedures, but there is also evidence that CH2Br2 is a natural product and can be formed by peroxidases (13, 14). Several mechanisms for the metabolism and toxicity of CH2Br2 have been proposed. Detoxification via conversion to readily excretible water soluble products can occur either via oxidation by P450s, which generates CO (15), or via conjugation to GSH catalyzed by GST1 enzymes, which ultimately generates formaldehyde (16). Both pathways may also lead to the formation of electrophilic reactants able to cause muta1 Abbreviations: GST, GSH S-transferase; AGT, O6-alkylguanineDNA alkyltransferase; hAGT, human AGT; NER, nucleotide excision repair; MNNG, N-methyl-N′-nitro-N-nitrosoguanidine; SDS, sodium dodecyl sulfate; DMSO, dimethyl sulfoxide; MALDI, matrix-assisted laser desorption ionization MS; TOF, time-of-flight.

10.1021/tx049958o CCC: $27.50 © 2004 American Chemical Society Published on Web 05/05/2004

Alkyltransferase-Mediated Mutagenesis by Dibromomethane

tions in Salmonella typhimurium TA100 (9, 10). The GST-mediated pathway has been studied in more detail, and CH2Br2 was able to induce revertants in S. typhimurium TA1535 expressing mammalian GSTs (17, 18), indicating a reaction at guanine or cytosine. The reaction of the putative GS-CH2Br intermediate results in the formation of DNA adducts. Multiple DNA adducts were derived from the model conjugate S-(1-acetoxymethyl)GSH in vitro with the four nucleosides present in DNA (17, 19). The major product was S-[1-(N2-deoxyguanosinyl)methyl]GSH (95%), and adducts were also formed on the N3 atom of thymine (1-2%), the N4 atom of thymine (1-2%), and the N7 atom of adenine (0.05%) (17, 19, 20). The repair of O6-alkylguanine adducts in DNA by AGT provides an important means of defense against the mutagenic, carcinogenic, and cytotoxic effects of many simple alkylating agents (21-25). AGT repairs alkyl adducts on the O6-position of guanine by transferring them to a Cys residue in the active site of the protein. This repair reaction is highly effective in limiting mutations and lethality caused by simple alkylating agents such as MNNG and N-methyl-N-nitrosourea. It was therefore quite unexpected when it was found that the overexpression of Ogt (one of the two Escherichia coli AGTs) actually increased the toxicity of dibromoalkanes such as CH2Br2 and 1,2-dibromoethane (BrCH2CH2Br) toward E. coli (26). The enhancement of the mutagenicity and toxicity of dibromoalkanes in E. coli has been confirmed and extended to AGTs from other species. These include hAGT and other mammalian AGTs (2729), the E. coli Ada (28), and the S. typhimurium Ogt, although the latter was only weakly active (30). The expression of the E. coli Ada was also reported to increase the killing of mammalian cells by CH2Br2 (31). Recently, we have studied the mechanism of the paradoxical increase in the mutagenicity and toxicity of BrCH2CH2Br by hAGT in detail and have obtained evidence that it results from the reaction of BrCH2CH2Br at the active site of AGT to generate a reactive intermediate at the alkyl acceptor site Cys145 (32, 33). This intermediate leads to a covalent AGT adduct in DNA, which brings about a decreased survival and an increase in the frequency of G:C to A:T transition and G:C to T:A transversion mutations. These studies provide a plausible model for the hAGTmediated increase of BrCH2CH2Br toxicity. In the present work, we have examined the role of AGT in facilitating the genotoxicity of CH2Br2 and compared the results to those observed with BrCH2CH2Br. We have examined the reactivity of hAGT with CH2Br2 and a model substrate bromomethyl acetate (BrCH2OAc) and have measured the effects of hAGT and an inactive mutant C145A on survival and mutations in cells exposed to CH2Br2 and BrCH2OAc. In contrast to the increase in the mutagenicity and toxicity of CH2Br2 that occurs in the presence of active hAGT, the mutagenicity and toxicity of BrCH2OAc were reduced by active hAGT and enhanced by the inactive C145A mutant. A scheme explaining these results and outlining the role(s) of AGT in response to halogenated alkanes is proposed.

Experimental Procedures Materials. CH2Br2, BrCH2OAc, and BrCH2CH2Br were purchased from Aldrich Chemical Co. (Milwaukee, WI) and used without further treatment. Isopropyl β-D-thiogalactopyranoside,

Chem. Res. Toxicol., Vol. 17, No. 6, 2004 743 ampicillin, kanamycin, and reagents for the bacterial cultures that were used were obtained from Sigma Chemical Co. (St. Louis, MO). The oligodeoxyribonucleotide 5′-(AG)8-3′ was synthesized by Invitrogen (Gaithersburg, MD). Adenosine 5′-[γ-35S]thiotriphosphate triethylammonium salt ([γ-35S]ATPγS) was purchased from Amersham/Pharmacia (Piscataway, NJ). T4 polynucleotide kinase was obtained from New England Biolabs (Beverly, MA). Recombinant hAGT-His6 and C145S-His6 were purified by Co2+ affinity chromatography using a BioCad Sprint Perfusion system (PerSeptive Biosystems, Framingham, MA) as described previously (34). The recombinant P140K and G160R mutants were purified in a similar way, but the constructs that were used had an MRGSH(H)6GS- sequence at the N-terminal of the protein instead of the C-terminal (H)6 replacement tag (35, 36). Bacterial Strains and Plasmids. TRG8 cells were derived from the GWR109 strain by selection for cells with an altered cell membrane that allows for greater permeability to exogenous chemicals (36). The pIN vector was used for expression of the wild-type hAGT or its C145A mutant. TRG8 cells were transformed with these plasmids or the pIN vector alone by standard electroporation procedures (29). TRG8 cells were grown in Luria-Bertani (LB) medium under ampicillin and kanamycin (50 µg/mL) selection (32). Measurement of CH2Br2 and BrCH2OAc Toxicity in E. coli. TRG8 cells transformed with the pIN expression vector, the wild-type hAGT, or the C145A mutant vector were grown as described previously (32). Aliquots (0.5 mL) of the cell suspension were treated with 0-0.1 mM CH2Br2, 0-0.3 mM BrCH2OAc, or DMSO vehicle at 37 °C for 0-90 min with constant shaking. The cells were plated on M9 minimal media plates supplemented with 40 µg/mL histidine for survival rate measurements and on plates lacking histidine to obtain His+ revertants (32). The cytotoxicity of CH2Br2 or BrCH2OAc was expressed as the percentage of surviving cells (cell colonies that formed in the presence of histidine) in comparison with vehicletreated cells. The mutagenicity of these two agents was determined as the number of His+ revertants (cells that grew in the absence of histidine) per 108 survivors. Alkyltransferase Activity Measurement. Wild-type hAGT (50 ng) was incubated with 0-5 mM BrCH2CH2Br or CH2Br2 or 0-0.2 mM BrCH2OAc for 0-30 min at 37 °C in a buffer containing 50 mM Tris-HCl (pH 7.6), 0.1 mM EDTA, and 20 µg/mL hemocyanin. The residual activity of the hAGT protein was examined using an assay that measured the removal of [3H]-labeled methyl adducts from O6-guanine in calf thymus DNA (37, 38). Results were expressed as the percentage of AGT activity remaining in treated samples as compared to DMSO controls. Analysis of Covalent Binding of AGT to DNA by Gel Electrophoresis. The oligodeoxyribonucleotide 5′-(AG)8-3′ was labeled at the 5′-end with [γ-35S]ATPγS using T4 polynucleotide kinase as described previously (33). Mixtures of unlabeled (20 pmol) and labeled oligonucleotide (6.5 pmol) were incubated at 37 °C for 60 min with 20 mM CH2Br2 or BrCH2OAc. These mixtures also contained 2 µg of purified recombinant wild-type hAGT, C145S hAGT, or spermine synthase protein. The reaction (15 µL) was carried out in a buffer containing 50 mM Tris (pH 7.6) and 0.1 mM EDTA. A 1.5 µL aliquot of 10% SDS solution (w/v) was added to terminate the reaction. The mixtures were then resolved on a 12.5% (w/v) SDS polyacrylamide gel. The gel was dried and visualized using a Molecular Dynamics PhosphorImager SI system and analyzed using ImageQuant application software (32). MALDI-TOF Analysis. Wild-type, C145S, P140K, or G160R mutants of hAGT protein (2 µg) were incubated with 5 mM BrCH2OAc (diluted from a 500 mM stock in CH3CN) in 50 mM diisopropylethylammonium acetate (pH 6.7) buffer. Immediately after the they were mixed, 0.7 µL of the sample was spotted on a MALDI plate along with 0.7 µL of sinapinic acid matrix solution (10 mg/mL in 50:50 CH3CN:2% CF3CO2H, v/v). The plate was placed under vacuum in a desiccator to rapidly

744

Chem. Res. Toxicol., Vol. 17, No. 6, 2004

Liu et al.

Figure 1. Effects of CH2Br2 and BrCH2OAc on the survival and mutation of E. coli TRG8 cells. TRG8 strains transformed with either the pIN vector, the wild-type hAGT, or the C145A mutant were treated for 90 min with 0-0.1 mM CH2Br2, BrCH2CH2Br, or 0-0.3 mM BrCH2OAc. Cells were diluted and plated for the determination of survival and mutation frequencies as described under the Experimental Procedures. (A) Cytotoxicity and (B) mutagenicity following exposure to CH2Br2. The toxicity of BrCH2CH2Br in cells expressing wild-type hAGT, which has been published previously (32, 33), is also included. (C,D) Effects of exposure to BrCH2OAc. Percentage survival represents the number of colonies from bacteria exposed to CH2Br2 or BrCH2OAc divided by that of bacteria exposed to the DMSO vehicle. Mutation frequencies were calculated as the number of his revertants that grew on M9 minimal plates lacking histidine over the 108 survivors on M9 plates supplemented with histidine. Plots indicate the average ( SD of four measurements. evaporate the solvent. Each experiment was repeated at least four times, and the data from all runs were compiled and averaged. MALDI-TOF spectra were recorded over a range of m/z 100061 639 on a Perseptives Voyager Elite instrument (Applied Biosystems, Framingham, MA) in the positive linear mode. The accelerating voltage, grid voltage, and guide wire voltages were 250 000, 93, and 0.2%, respectively. Ubiquitin was used as an external reference. Analysis of Mutants in the rpoB Gene of E. coli. Mutations in the rpoB gene in TRG8 cells transformed with wild-type hAGT, C145A hAGT mutant, or an empty pIN vector were examined following exposure to CH2Br2 and BrCH2OAc. Briefly, cells were treated with 0.035 or 0.1 mM CH2Br2, with 1 or 3 mM BrCH2OAc, or with DMSO vehicle for 90 min. Rifampicin resistant mutants were derived by plating onto LB media plates supplemented with 100 µg/mL rifampicin. The mutation frequency of the rpoB gene in TRG8 cells was expressed as the number of rpoB mutants per 108 survivors, after adjusting to the level of surviving cells (cells grown on LB plates lacking rifampicin). A section of the rpoB gene from rifampicin resistant colonies was amplified by PCR and sequenced (33).

Results Effect of AGT Expression on Toxicity and Mutagenicity of CH2Br2 in E. coli. To examine the role of hAGT on the cellular response to CH2Br2 and BrCH2OAc, we utilized the uvrABC NER proficient E. coli TRG8 strain, which lacks endogenous AGT (36). Survival and mutations causing a reversion to histidine independence

were measured after exposure to CH2Br2 or BrCH2OAc. TRG8 cells that expressed hAGT or the inactive C145A mutant hAGT or no AGT were compared in their responses. The basic hypothesis was that AGT might increase the toxicity of CH2Br2 by reacting with it at Cys145 to form a reactive intermediate, AGT-Cys145SCH2Br, in a manner similar to the activation of BrCH2CH2Br by AGT (32, 33). BrCH2OAc was used for comparison with CH2Br2 because S-(1-acetoxymethyl)GSH has been successfully applied as a model for studying the reaction of the short-lived S-(1-bromomethyl)GSH intermediate formed by the interaction of GSH and CH2Br2 with nucleosides and DNA (19), and it was expected that the AGT-Cys145S-CH2Br intermediate would also be highly unstable. The presence of active hAGT increased both the toxicity and the mutagenicity of CH2Br2 in a dose-dependent manner, but the C145A mutant had no effect (Figure 1A,B). There was no increase in toxicity, and only a marginal number of mutations in cells lacking hAGT showed that AGT activity is a key factor in mediating the toxic effects of CH2Br2 in the dose range tested (up to 0.1 mM CH2Br2 for 90 min). CH2Br2 was only slightly less effective than BrCH2CH2Br in reducing the survival of cells expressing hAGT but was significantly less potent in inducing mutants. Thus, treatment with 0.05 mM CH2Br2 for 90 min caused a decrease in viability of 75% and resulted in approximately 140 histidine revertants per


Chem. Res. Toxicol., Vol. 17, No. 6, 2004 745

Figure 2. Effects of CH2Br2 and BrCH2OAc on hAGT activity. The hAGT protein was exposed to 0-5 mM CH2Br2 (A), 0-150 µM BrCH2OAc (C), or DMSO vehicle for 30 min. Results for exposure to 5 mM CH2Br2 and 150 µM BrCH2OAc for 0-30 min are shown in panels B and D, respectively. The percentage of alkyltransferase activity remaining was determined by dividing the activity following the treatment by alkyltransferase activity at 0 min (n ) 4).

108 survivors, whereas BrCH2CH2Br caused a similar reduction in viability but more than 600 mutants. In contrast, the toxicity and mutagenicity of BrCH2OAc were not increased by the presence of active hAGT (Figure 1C,D). The reduction of survival required the addition of 1-3 mM BrCH2OAc, and there was no effect of hAGT. The presence of the C145A mutant hAGT actually slightly increased the loss of viability. A more pronounced difference was observed among the three cell types in terms of mutant induction. While the levels of His revertants increased as the BrCH2OAc concentration rose in all three strains, BrCH2OAc was markedly more mutagenic in cells expressing the C145A mutant than in the other two strains. Exposure to 3 mM BrCH2OAc caused approximately 47 000 His revertants per 108 survivors in this strain, 18-fold more than in the control TRG8 cells and 52-fold more than in cells expressing wild-type hAGT (Figure 1D). The expression of wild-type hAGT significantly reduced the mutagenicity of BrCH2OAc over the whole dose range. The differential effects of hAGT and C145A on the toxicity of CH2Br2 and BrCH2OAc clearly indicate that AGT plays distinct roles in mediating the toxicity of these two compounds. Reaction of CH2Br2 and BrCH2OAc with hAGT in Vitro. To assess the reactivity of CH2Br2 and BrCH2OAc with hAGT, the AGT protein was incubated with varying concentrations of these agents for 0-30 min at pH 7.6 and the hAGT activity was monitored. As shown in Figure 2A,B, CH2Br2 inactivated the hAGT protein in a dose- and time-dependent manner, although it was less potent than BrCH2CH2Br. Eighty percent of the hAGT activity was lost following a 30 min incubation with 2 mM BrCH2CH2Br, while only 38% was inactivated by 2

mM CH2Br2. BrCH2OAc was a very potent inactivator of hAGT (Figure 2C,D) with an 80% reduction produced by a 30 min exposure to 0.1 mM BrCH2OAc. Structures of Adducts Formed by hAGT and CH2Br2 or BrCH2OAc. MALDI-TOF was performed to analyze the structures of reaction products of hAGT with CH2Br2 or BrCH2OAc. Because of the unstable nature of these reaction products, mass analysis was attempted for unreacted hAGT, wild-type hAGT, or hAGT mutants incubated with CH2Br2 or BrCH2OAc without trypsin digestion. These recombinant proteins have the six residues at the C terminus replaced with His residues for ease of purification. This alteration does not affect the properties of the AGT (34). Unreacted wild-type AGT had a MALDI signal with an m/z of 21 890 (Figure 3). The mass spectra were not changed when wild-type hAGT was incubated with CH2Br2 for 1 or 10 min prior to analysis (data not shown). Likewise, the spectra did not change when the sample was spotted immediately after mixing. The lack of an observable hAGT and CH2Br2 conjugate may be due to the instability of the conjugate. Alternatively, the instrument may not have been sensitive enough to detect the extremely low level of residual product. To determine whether intermediate stability was a problem, we analyzed the reaction product of hAGT and BrCH2OAc, a reactive analogue of CH2Br2 that should produce a more stable reaction intermediate. Because GSCH2OAc had previously been shown to have a halflife of only 11 s at pH 8.0 (19), we performed MALDI experiments with BrCH2OAc at a slightly lower pH and with shorter reaction times. We utilized a diisopropylethylammonium acetate buffer (pH 6.7), which is unable

746


Liu et al.

Figure 3. Mass spectral analysis of wild-type hAGT and the C145S hAGT mutant after incubation with BrCH2OAc. Partial MALDI spectra (top to bottom) of wild-type hAGT protein, wildtype hAGT incubated with BrCH2OAc for 1 min, C145S hAGT, and C145S mutant incubated with BrCH2OAc for 1 min. Data were subjected to a 13 point Gaussian smoothing function.

Figure 4. Mass spectral analysis of P140K and G160R hAGT mutants after incubation with BrCH2OAc. Partial MALDI spectra (top to bottom) of P140K and G160R incubated with BrCH2OAc for 1 min. Data were subjected to a 13 point Gaussian smoothing function.

to react with BrCH2OAc. This buffer is also volatile enough not to interfere with the MALDI analysis. Because of the rapid nature of the experiments, we were unable to desalt the samples prior to acquisition; as a result, the observed peaks were broad. Wild-type AGT had a MALDI signal with an average m/z of 21 895 (Figure 3), which is slightly higher than the predicted mass of 21 876. This higher mass may be due to the presence of trace amounts of sodium (∆amu ) +22). BrCH2OAc was added to wild-type AGT, and the sample was spotted and dried immediately. The MALDI signal shifted to an m/z of 21 966, a change that was statistically significant by t-test analysis (P value ) 0.0001). When the C145S mutant of AGT was examined, a MALDI signal with an average m/z of 21 875 was observed. In contrast to the wild-type spectrum, the signal shifted only slightly to 21 884 when BrCH2OAc was added (Figure 3). This was a statistically insignificant shift (P ) 0.6045). Two other hAGT mutants, P140K and G160R, were also tested for BrCH2OAc reaction. Both of these mutants had an additional 12 amino acid sequence, MRGSH(H)6GS- at the N-terminal of the protein instead of the C-terminal (H)6 replacement tag. Therefore, they have masses larger than the carboxyl terminal His6-tagged wild-type and C145S mutant (35, 36). The P140K mutant had an m/z of 23 085, which shifted to 23 153 upon BrCH2OAc addition. The G160R mutant had an m/z of 23 149, which shifted to 23 230 upon BrCH2OAc addition (Figure 4). Because of the broadness of the MALDI signals, we were typically unable to resolve individual signals for the unmodified and adducted proteins. Thus, we would expect that the calculated mass would be a weighted average of the unmodified and adducted proteins. The C145S mutant showed no change in the MALDI spectrum ∼1 min after BrCH2OAc addition, indicating the absence of adduct formation. In contrast, the wild-type, P140K,

and G160R mutants showed similar increases in mass upon BrCH2OAc addition (+71, 67, and 82, respectively). All of these were reasonably close to the value of 72 mass units that would be expected for -CH2OAc adduct formation. These mass increases suggest that there is a displacement of the Br atom by the AGT Cys145. Demonstration of hAGT-DNA Cross-Linking by CH2Br2. The results described above suggest that CH2Br2 and BrCH2OAc react with hAGT to form an adduct at Cys145. We have shown that Br2CH2CH2Br also interacts with Cys145 and that the resulting half mustard is then able to react with DNA to form a covalent hAGT-DNA adduct (32, 33). Polyacrylamide gel electrophoresis under denaturing conditions was used to test if similar adducts could be formed by the reaction of the AGT complexes from CH2Br2 and BrCH2OAc with the oligodeoxyribonucleotide 5′-d(AG)8-3′. As shown in Figure 5A, incubation with CH2Br2 resulted in the formation of wild-type hAGT-DNA adducts at levels higher than those observed for BrCH2CH2Br. With both dihaloalkanes, there were two distinct complexes (complex 1 and 2), consisting of one and two hAGT proteins, respectively. CH2Br2 did not cause the covalent attachment of spermine synthase, used as a control protein, to DNA indicating that formation of hAGT-DNA adducts is specific for hAGT. Mutation of the C145 in the C145S protein completely prevented conjugation, indicating a critical role for AGT activity and the Cys145 residue in the reaction (Figure 5A). BrCH2OAc, on the other hand, did not produce any detectable amounts of hAGT-DNA conjugates with either wild-type or Cys145Ser hAGT (Figure 5B). This result is consistent with the lack of hAGT-promoted BrCH2OAc mutagenicity or toxicity. The hAGT-CH2Br2 complex was quite unstable, and the ability to link hAGT covalently to DNA was lost rapidly when the hAGT and CH2Br2 were incubated prior to the addition of the 5′-d(AG)8-3′ (Figure 6). After a 10


Figure 5. Effects of CH2Br2 and BrCH2OAc on the formation of cross-links between hAGT and DNA. Wild-type hAGT, the C145S mutant, or spermine synthase (S.S.) (2 µg) was incubated with [35S]5′-(AG)8-3′ at 37 °C for 60 min in the presence or absence of 20 mM BrCH2CH2Br, CH2Br2, or BrCH2OAc. The reaction mixtures were separated by electrophoresis with a 12.5% SDS polyacrylamide gel and visualized with a PhosphorImager system and ImagerQuant application software.

Figure 6. Instability of hAGT adducts formed with CH2Br2 or BrCH2CH2Br. Wild-type hAGT was incubated with 20 mM BrCH2CH2Br or CH2Br2 for 0-30 min followed by the addition of [35S]5′-(AG)8-3′ for 60 min. Conjugation between hAGT and DNA was analyzed as described for Figure 5.

min preincubation, the amount of hAGT-DNA complexes formed as shown by the percentages of the 5′-d(AG)8-3′ in complex forms decreased to 31%. The decomposition was faster than that of the intermediate formed by the reaction of hAGT and BrCH2CH2Br2 (Figure 6). Determination of AGT-Modulated Mutation Spectra Caused by CH2Br2 and BrCH2OAc. The types of mutations caused by CH2Br2 and BrCH2OAc in E. coli expressing hAGT were examined in a forward mutation assay to rifampicin resistance. This resistance results from changes in the rpoB gene sequence that encodes the β subunit of RNA polymerase II and has been widely used as a marker to examine the spectra of mutants arising


Figure 7. Effect of hAGT on mutation frequency of the rpoB gene by treatment with CH2Br2 and BrCH2OAc. E. coli TRG8 cells transformed with pIN wild-type hAGT or C145A hAGT were exposed to 0-0.1 mM CH2Br2 or 0-3 mM BrCH2OAc for 90 min. Rifampicin resistant mutants were obtained on LB plates containing 100 µg/mL rifampicin. The mutation frequency of the rpoB gene was corrected with the number of survivors following the treatment.

endogenously or through induction by exogenous reagents (39, 40). As expected from the results in the histidine reversion assay of Figure 1, the frequency of rifampicin resistant (Rifr) mutants produced by CH2Br2 was increased about 100-fold by the expression of hAGT (Figure 7). Sequence analysis of the mutants (Table 1) indicated that the spectrum of mutations induced by CH2Br2 in cells expressing hAGT was quite different from that in cells lacking hAGT. Mutations observed in cells lacking hAGT were nearly equivalent amounts (each 2025% of the total) of A:T to T:A and A:T to C:G transversions and of G:C to T:A transversions and G:C to A:T transitions (Table 1). In contrast in the hAGT-expressing cells, 90% of the mutants occurred at G:C sites with G:C to A:T transitions and G:C to T:A transversions accounting for approximately 48 and 40% of total mutants (Table 1). The exposure of cells to the solvent DMSO did not result in any significant increase in mutations in the rpoB gene irrespective of the hAGT status. The spectra of background mutants in TRG8 cells did not reveal any specific induction of G:C to A:T or G:C to T:A mutations (results not shown) and were similar to those reported for a recent large study (40). In cells treated with BrCH2OAc, G:C to A:T transition mutations were predominant in both cells expressing wild-type AGT (80%) and in cells expressing the C145A mutant (93%) (Table 2).

748


Liu et al.

Table 1. Mutations in the rpoB Gene in E. coli Cells with and without hAGT Expression (0.1 mM CH2Br2) base change A:T to T:A A:T to C:G A:T to G:C G:C to A:T

G:C to T:A

A:T to T:A

G:C to A:T

G:C to T:A

codon change

amino acid

mutation

(a) no hAGT expression (total ) 20) CAG 513 QfL CTG CAG 513 QfP CCG CAG 513 QfR CGG TCC 531 SfF TTC CCT 564 PfL CTT TCT 512 SfY TAT CAC 526 HfN AAC CGT 529 RfL CTT (b) hAGT expression (total )27) CAG 513 CTG ATC 572 TTC GAC 516 AAC CAC 526 TAC CGT 529 CAT CGT 529 AAT TCC 531 TTC GAC 516 TAC CGT 529 CTT CGT 529 AGT CGT 529 AAT TCC 531 TAC

n

Table 2. Mutations in rpoB Gene in E. coli Cells with and without hAGT Expression (0.1 mM BrCH2OAc) base change

codon change

amino acid no.

mutation

QfL

2

(a) wild-type hAGT expression (total ) 15) A:T to T:A CAC 526 HfL CTC A:T to G:C CAG 513 QfR AGG G:C to A:T GAC 516 DfN AAC CAC 526 HfY TAC CGT 529 RfC TGT CGT 529 RfH CAT TCC 531 SfF TTC CCT 564 PfL CTT G:C to T:A CGT 529 RfL AGT

IfF

1

G:C to A:T

DfN

3

HfY

6

RfH

1

RfN

1

SfF

2

DfY

3

RfL

3

RfS

1

RfN

1

Sf Y

3

4 5 1 3 2 2 2 1

Discussion Our results confirm the original observations of Abril et al. (26-28, 31) that the DNA repair protein AGT paradoxically increases the mutagenicity and toxicity of CH2Br2. Although these investigations suggested that hAGT-induced mutations were entirely (for BrCH2CH2Br) and substantially (for CH2Br2) dependent on the absence of a functional NER, we are able to detect hAGTmediated toxicity of both BrCH2CH2Br (32) and CH2Br2 (the present study) in an NER proficient E. coli strain. Our results are consistent with the scheme shown in Scheme 1 in which CH2Br2 reacts at the active site Cys145 residue of AGT and the resulting AGT-Cys145SCH2Br then reacts with guanine in DNA to form a covalent adduct, which leads to cytotoxicity and to mutagenicity. The facile reaction of CH2Br2 with Cys145 is likely to be due to the very low pKa of this residue (41). The Cys145 residue is part of an extensive hydrogen-bonding network in hAGT protein that involves His146, Arg 147, Glu172, and a water molecule (42). This environment renders Cys145 highly reactive toward alkyl groups on the O6-position of guanine and allows it to carry out its normal repair reaction when bound to DNA containing O6-methyguanine. However, electrophiles unrelated to

(b) C145A hAGT expression (total ) 13) GAC 516 DfN AAC CAC 526 HfY TAC CGT 529 RfH CAT TCC 531 SfP TTC CCT 564 PfL CTT A:T to G:C TCT 512 SfP CCT

n 1 1 2 1 4 2 1 2 1

1 2 6 2 1 1

alkylated DNA such as CH2Br2 and BrCH2OAc can also react readily at this site. The AGT-Cys145S-CH2Br formed in this reaction with CH2Br2 would be expected to be highly unstable. The results of the MS analysis and the experiment shown in Figure 7 are consistent with this. However, if DNA is available, the DNA binding function of AGT would cause the formation of a noncovalent DNA:AGT-Cys145S-CH2Br complex and this would lead to the generation of AGTCys145S-CH2-DNA adducts. It should be noted that Cys145 is buried in a deep binding pocket in hAGT. The normal repair process therefore requires a significant distortion of the DNA that is brought about by the AGT protein (42, 43). In a reaction that requires residues Arg128 and Tyr114, hAGT causes the target nucleoside to be “flipped” out of the helix and into the binding pocket in close proximity to Cys145 (42, 44, 45). The specificity of this flipping is not fully understood, but there appears to be only relatively little preference for O6-methyguanine over guanine itself although, of course, no repair reaction occurs when guanine is recognized. Because the reactive -CH2Br side chain can only extend a short distance into the binding pocket of the AGT-Cys145S-CH2Br complex, it is likely that the reaction is restricted to nucleosides that can be accepted in the binding pocket to place a suitable electrophilic site within this limited distance. This proposed stalling/idling at guanine residues would account for the very high percentage of mutations in the CH2Br2-treated AGT-expressing cells that occur at G:C pairs (Table 1). The G:C to T:A transversions that were observed and accounted for 40% of the total are likely to be due to formation of a labile guanine-N7 adduct and subsequent depurination because apurinic sites cause such mutations



Scheme 1. Role of AGT in the Toxicity and Mutagenesis of CH2Br2 and BrCH2OAc

(46, 47); we have identified such an N7 adduct in cells treated with BrCH2CH2Br (33). At present, we do not know which adduct causes the G:C to A:T transitions, which account for 48% of the total. Reaction at either the O6- or the N2-atom of guanine is possible, but the resulting protein DNA cross-link would be stable. It is unclear whether further modification of this large adduct is needed to allow copying by bypass polymerases to allow DNA synthesis to continue (48-50). The results obtained with BrCH2OAc are in striking contrast to those with CH2Br2, but they can also be explained by the scheme shown in Scheme 1. BrCH2OAc is very active in reacting with Cys145 of the AGT; the combination of the high reactivity of this Cys residue and that of the BrCH2OAc causes the rapid loss of the ability of AGT to repair DNA, shown in Figure 2C,D. The resulting AGT-Cys145S-CH2OAC is unable to form a covalent DNA adduct after noncovalent binding to DNA (Figure 5B); therefore, no increase in mutagenicity or toxicity results. Although the model conjugate GSCH2OAc is capable of reacting with nucleosides and DNA (19, 20), acetate appears to be a poor leaving group as compared with the halides, even Cl and F. For instance, in earlier work (51), we reported that S-(2-acetoxyethyl)GSH was quite stable in water and was only marginally mutagenic in S. typhimurium, although S-(2-fluoroethyl)GSH was quite mutagenic. The t1/2 of GSCH2OAc was 11 s in water (neutral pH), which appears to be far longer that that of a corresponding RSCH2Cl compound, CH3SCH2Cl (52). The 11 s t1/2 would argue that a compound of the nature RSCH2OAc should be reactive. We do not understand the chemistry of the acetate completely. Earlier, we noted that GSCH2OAc does not react with p-nitrobenzylpyridine (17) and S-(2-acetoxyethyl)GSH yields only a weak reaction with this model electrophile

(51). It is possible that AGT-Cys145-S-CH2OAc, the putative conjugate, is also more prone to hydrolysis than nucleophilic attack. The ability of the C145A hAGT mutant to increase the mutagenicity and toxicity of BrCH2OAc is quite similar to the ability of this mutant to enhance the response to MNNG (53). This effect is due to the ability of the C145A mutant to bind to O6-methylguanine formed in DNA by MNNG and prevent its repair by NER (53). We therefore suggest that BrCH2OAc forms O6-(CH2OAc)guanine adducts in DNA and that these are recognized by C145A AGT and protected in a similar way. The magnitude of C145A enhancement on response to BrCH2OAc is greater than that seen with MNNG. This could be due to a tighter binding of the C145A mutant to O6-(CH2OAc)guanine than to O6-methylguanine. A more likely explanation is that there is a greater role for NER in repair of the O6(CH2OAc)guanine adducts because O6-methylguanine, which causes little distortion of the DNA, is poorly recognized by NER (53, 54). The modest effect of wildtype hAGT in reducing the mutations caused by BrCH2OAc may be due to the fact that BrCH2OAc is such a potent direct inactivator of hAGT (Figure 2). Another possibility is that the O6-(CH2OAc)guanine is only slowly repaired. It has been reported that O6-(carboxymethyl)guanine was not repaired by E. coli Ogt under conditions where all O6-methylguanine was lost (55). The inactivated AGT-Cys145S-CH2COAc would not be able to protect the O6-(CH2OAc)guanine by NER in the same way as the C145A mutant because AGT alkylated at Cys145 is rapidly degraded in vivo (56, 57). The finding that virtually all of the rifampicin resistant mutants caused by BrCH2OAc are G:C to A:T transitions (Table 2) is consistent with the fact that they are caused by a O6(CH2OAc)guanine adduct because small adducts on the

750


guanine O6 are well-known as a source of such mutations (58). Our view is that the reaction of BrCH2OAc with DNA is a direct one, based on the lack of enhancement of mutagenicity by AGT (Figure 2) and the known electrophilic properties of the reagent. Although acetate appears to be a much poorer leaving group than Br, the presence of acetate on a methyl seems to make the single Br a better leaving group, presumably due to the ability of the ether oxygen to donate charge (despite the carbonyl). Our results support a mechanism for AGT-mediated activation of CH2Br2 that is broadly similar to that for activation of BrCH2CH2Br, but there are several noteworthy differences. First, CH2Br2 was less potent in reacting directly with AGT (Figure 2A). This might be expected since the proximity of the second bromine atom would decrease the reactivity of the first. However, the AGT-Cys145S-CH2Br conjugate, derived from CH2Br2, is more reactive than the episulfonium ion, AGT-Cys145S+(CH2CH2), derived from BrCH2CH2Br (Figures 5 and 6). These two factors, lesser reactivity of CH2Br2 with AGT but greater reactivity of the resulting complex, may partially cancel out in the cell where much of the AGT is DNA-bound, but the instability of the AGT-Cys145S-CH2Br may cause decomposition before the reaction with DNA. As shown in Figure 1B, CH2Br2 was significantly less mutagenic than BrCH2CH2Br in cells expressing hAGT, but the ratio of G:C to A:T transitions to G:C to T:A transversions was 1.2, whereas this ratio was 3.8 after exposure to BrCH2CH2Br. This suggests that the formation of N7-guanine adducts, which depurinate to give the apurinic sites leading to transversions, occurs as a higher fraction of the total adducts with the activated AGT derived from CH2Br2. Because of the complexity of this labile system, we do not yet have direct evidence for the existence of depurination or an N7-guanyl adduct. However, in our previous work with BrCH2CH2Br (33), we were able to identify the guanine base covalently linked to the AGT active site peptide (residues 136-147) through an ethylene bridge. A similar conjugate resulting from the reaction with CH2Br2 would not be expected to be stable through the trypsin cleavage procedure because of the instability of the putative N-CH2-S linkage. Although all AGTs tested from various species were effective to some extent in enhancing the mutagenicity and toxicity of BrCH2CH2Br and CH2Br2, differences in potency have been reported (27, 28, 30). We have not investigated this question since all of our experiments were carried out with hAGT, but it could be due to either a difference in the ability to interact at the Cys acceptor site or an altered stability of the reactive intermediate. Steric factors allowing access of the dihaloalkanes to the acceptor site are unlikely to play a major role. The P140K and G160R hAGT mutants, which are resistant to inhibition by O6-benzylguanine by restricting access of the inhibitor to the active site pocket of the protein (35, 36), were able to react readily with BrCH2OAc (Figure 4). The relatively low stability of the AGT-Cys145S-CH2Br intermediate (Figure 6) may account for the lack of hAGT-CH2Br2 conjugates in MALDI analysis. On the basis of the model of reaction of the hAGT intermediate with DNA and/or water, this instability would be expected for AGT-Cys145S-CH2Br since it cannot undergo cyclic transformation to an episulfonium ion in a way similar to the AGT-Cys145S-CH2CH2Br generated from

Liu et al.

BrCH2CH2Br (32). The reaction product of GSH and CH2Br2 transforms into GSCH2OH, which easily breaks down into GSH and HCHO (16, 19). While a similar reaction may occur in vitro with isolated hAGT, it is unlikely that an intact hAGT is restored by hydrolysis in vivo since alkylated hAGT is rapidly degraded (56, 57). The ability of haloalkanes to react with hAGT at the Cys acceptor site and thus inactivate the protein, which is demonstrated in the current study and in our previous work (32, 33), therefore, should not be ignored as an additional potential source of genotoxicity of these agents since AGT is needed to mitigate the effects of exposure to endogenous and exogenous alkylating agents. In summary, we have demonstrated that AGT is a target for the binding of the CH2Br2 or BrCH2OAc due to the reaction of the highly reactive Cys145 residue that forms the alkyl acceptor site of this protein. This reaction inactivates AGT, and such inactivation would impair the ability of the cell to repair alkylated DNA. The AGTCys145S-CH2Br resulting from the reaction with CH2Br2 is able to react with guanine residues in DNA causing mutations and reducing survival. Our experiments therefore reveal two novel pathways by which exposure to CH2Br2 may cause damage to the genome in addition to the well-recognized pathway involving activation by GSTs.

Acknowledgment. This work was supported in part by U.S. Public Health Service Grants R01 CA18137 (A.E.P.), R01 ES10546, P30 ES00267 (F.P.G.), and T32 ES07028 (K.M.W.). We thank M. L. Manier and D. L. Hachey of the Vanderbilt Mass Spectrometry Resource Facility for their assistance.

References (1) Agency for Toxic Substances & Disease Registry (1993) Toxicological Profile for Methylene Chloride, U. S. Department of Health & Human Services, Atlanta. (2) Kari, F. W., Foley, J. F., Seilkop, S. K., Maronpot, R. R., and Anderson, M. W. (1993) Effect of varying exposure regimens on methylene chloride-induced lung and liver tumors in female B6C3F1 mice. Carcinogenesis 14, 819-826. (3) Burek, J. D., Nitschke, K. D., Bell, T. J., Wackerle, D. L., Childs, R. C., Beyer, J. D., Dittenber, D. A., Rampy, L. W., and McKenna, M. J. (1984) Methylene chloride: A 2 year inhalation toxicity and oncogenicity study in rats and hamsters. Fundam. Appl. Toxicol. 4, 30-47. (4) Andersen, M. E., Clewell, H. J., III, Gargas, M. L., Smith, F. A., and Reitz, R. H. (1987) Physiologically based pharmacokinetics and the risk assessment process for methylene chloride. Toxicol. Appl. Pharmacol. 87, 185-205. (5) Anderson, M. W., and Maronpot, R. R. (1993) Methylene chlorideinduced tumorigenesis. Carcinogenesis 14, 787-788. (6) Nitschke, K. D., Burek, J. D., Bell, T. J., Kociba, R. J., Rampy, L. W., and McKenna, M. J. (1988) Methylene chloride: a 2-year inhalation toxicity and oncogenicity study in rats. Fundam. Appl. Toxicol. 11, 48-59. (7) Hanson, D. (1997) Methylene chloride exposure sharply limited. Chem. Eng. News 75, 7. (8) Hanson, D. (1991) OSHA to cut methylene chloride exposure limits. Chem. Eng. News 69, 8. (9) van Bladeren, P. J., Breimer, D. D., Rotteveel-Smijs, G. M. T., and Mohn, G. R. (1980) Mutagenic activation of dibromomethane and diiodomethane by mammalian microsomes and glutathione S-transferases. Mutat. Res. 74, 341-346. (10) Osterman-Golkar, S., Hussain, S., Walles, S., Anderstam, B., and Sigvardsson, K. (1983) Chemical reactivity and mutagenicity of some dihalomethanes. Chem.-Biol. Interact. 46, 121-130. (11) Jepson, G. W., and McDougal, J. N. (1997) Physiologically based modeling of nonsteady-state dermal absorption of halogenated methanes from an aqueous solution. Toxicol. Appl. Pharmacol. 144, 315-324.

Alkyltransferase-Mediated Mutagenesis by Dibromomethane (12) Komsta, E., Chu, I., Secours, V. E., Valli, V. E., and Villeneuve, D. C. (1988) Results of a short-term toxicity study for three organic chemicals found in Niagara River drinking water. Bull. Environ. Contam. Toxicol. 41, 515-522. (13) Beissner, R. S., Guilford, W. J., Coates, R. M., and Hager, L. P. (1981) Synthesis of brominated heptanones and bromoform by a bromoperoxidase of marine origin. Biochemistry 20, 3724-3731. (14) Goodwin, K. D., Schaefer, J. K., and Oremland, R. S. (1998) Bacterial oxidation of dibromomethane and methylbromide in natural waters and enrichment cultures. Appl. Environ. Microbiol. 1998, 4629-4636. (15) Kubic, V. L., and Anders, M. W. (1978) Metabolism of dihalomethanes to carbon monoxide- -III. Studies on the mechanism of the reaction. Biochem. Pharmacol. 27, 2349-2355. (16) Ahmed, A. E., and Anders, M. W. (1978) Metabolism of dihalomethanes to formaldehyde and inorganic halide- -II. Studies on the mechanism of the reaction. Biochem. Pharmacol. 27, 20212025. (17) Thier, R., Taylor, J. B., Pemble, S. E., Humphreys, W. G., Persmark, M., Ketterer, B., and Guengerich, F. P. (1993) Expression of mammalian glutathione S-transferase 5-5 in Salmonella typhimurium TA1535 leads to base-pair mutations upon exposure to dihalomethanes. Proc. Natl. Acad. Sci. U.S.A. 90, 85768580. (18) Wheeler, J. B., Stourman, N. V., Thier, R., Dommermuth, A., Vuilleumier, S., Rose, J. A., Armstrong, R. N., and Guengerich, F. P. (2001) Conjugation of haloalkanes by bacterial and mammalian glutathione transferases: mono- and dihalomethanes. Chem. Res. Toxicol. 14, 1118-1127. (19) Marsch, G. A., Mundkowski, R. G., Morris, B. J., Manier, M. L., Hartman, M. K., and Guengerich, F. P. (2001) Characterization of nucleoside and DNA adducts formed by S-(1-acetoxymethyl)glutathione and implications for dihalomethane-glutathione conjugates. Chem. Res. Toxicol. 14, 600-608. (20) Marsch, G. A., Botta, S., Martin, M. V., McCormick, W. A., and Guengerich, F. P. (2004) Formation and mass spectrometric analysis of DNA and nucleoside adducts by S-(1-acetoxymethyl)flutathione and by glutathione S-transferase-mediated activation of dihalomethanes. Chem. Res. Toxicol. 17, 45-54. (21) Lindahl, T., Sedgwick, B., Sekiguchi, M., and Nakabeppu, Y. (1988) Regulation and expression of the adaptive response to alkylating agents. Annu. Rev. Biochem. 57, 133-157. (22) Samson, L. (1992) The suicidal DNA repair methyltransferases of microbes. Mol. Microbiol. 6, 825-831. (23) Pegg, A. E., Dolan, M. E., and Moschel, R. C. (1995) Structure, function and inhibition of O6-alkylguanine-DNA alkyltransferase. Prog. Nucleic Acid Res. Mol. Biol. 51, 167-223. (24) Pegg, A. E. (2000) Repair of O6-alkylguanine by alkyltransferases. Mutat. Res. 462, 83-100. (25) Margison, G. P., and Santiba´n˜ez-Koref, M. F. (2002) O6-Alkylguanine-DNA alkyltransferase: role in carcinogenesis and chemotherapy. BioEssays 24, 255-266. (26) Abril, N., Luque-Romero, F. L., Prito-Alamo, M. J., Margison, G. P., and Pueyo, C. (1995) Ogt alkyltransferase enhances dibromoalkane mutagenicity in excision repair-deficient Escherichia coli K-12. Mol. Carcinog. 12, 110-117. (27) Abril, N., Luque-Romero, L., Prieto-Alamo, M.-J., Rafferty, J. A., Margison, G. P., and Pueyo, C. (1997) Bacterial and mammalian DNA alkyltransferases sensitize Escherichia coli to the lethal and mutagenic effects of dibromoalkanes. Carcinogenesis 18, 18831888. (28) Abril, N., Luque-Romero, L., Christians, F. C., Encell, L. P., Loeb, L., and Pueyo, C. (1999) Human O6-alkylguanine-DNA alkyltransferase: protection against alkylating agents and sensitization to dibromoalkanes. Carcinogenesis 20, 2089-2094. (29) Liu, H., Xu-Welliver, M., and Pegg, A. E. (2000) The role of human O6-alkylguanine-DNA alkyltransferase in promoting 1,2-dibromoethane-induced genotoxicity in Escherichia coli. Mutat. Res. 452, 1-10. (30) Abril, N., Luque-Romero, F. L., Yamada, M., Nohmi, T., and Pueyo, C. (2001) The effectiveness of the O6-alkylguanine-DNA alkyltransferase encoded by the ogtST gene from S. typhimurium in protection against alkylating drugs, resistance to O6-benzylguanine and sensitisation to dibromoalkane genotoxicity. Mutat. Res. 497, 111-121. (31) Abril, N., and Margison, G. P. (1999) Mammalian cells expressing Escherichia coli O6-alkylguanine-DNA alkyltransferases are hypersenstivie to dibromoalkanes. Chem. Res. Toxicol. 12, 544551. (32) Liu, L., Pegg, A. E., Williams, K. M., and Guengerich, F. P. (2002) Paradoxical enhancement of the toxicity of 1,2-dibromoethane by O6-alkylguanine-DNA alkyltransferase. J. Biol. Chem. 277, 3792037928.

Chem. Res. Toxicol., Vol. 17, No. 6, 2004 751 (33) Liu, L., Hachey, D. L., Valadez, G., Williams, K. M., Guengerich, F. P., Loktionova, N. A., Kanugula, S., and Pegg, A. E. (2004) Characterization of a mutagenic DNA adduct formed from 1,2dibromoethane by O6-alkylguanine-DNA alkyltransferase. J. Biol. Chem. 279, 4250-4259. (34) Liu, L., Xu-Welliver, M., Kanugula, S., and Pegg, A. E. (2002) Inactivation and degradation of O6-alkylguanine-DNA alkyltransferase after reaction with nitric oxide. Cancer Res. 62, 3037-3043. (35) Edara, S., Kanugula, S., Goodtzova, K., and Pegg, A. E. (1996) Resistance of the human O6-alkylguanine-DNA alkyltransferase containing arginine at codon 160 to inactivation by O6-benzylguanine. Cancer Res. 56, 5571-5575. (36) Xu-Welliver, M., Kanugula, S., and Pegg, A. E. (1998) Isolation of human O6-alkylguanine-DNA alkyltransferase mutants highly resistant to inactivation by O6-benzylguanine. Cancer Res. 58, 1936-1945. (37) Dolan, M. E., Moschel, R. C., and Pegg, A. E. (1990) Depletion of mammalian O6-alkylguanine-DNA alkyltransferase activity by O6-benzylguanine provides a means to evaluate the role of this protein in protection against carcinogenic and therapeutic alkylating agents. Proc. Natl. Acad. Sci. U.S.A. 87, 5368-5372. (38) Dolan, M. E., Pegg, A. E., Dumenco, L. L., Moschel, R. C., and Gerson, S. L. (1991) Comparison of the inactivation of mammalian and bacterial O6-alkylguanine-DNA alkyltransferases by O6-benzylguanine. Carcinogenesis 12, 2305-2310. (39) Severinov, K., Soushko, M., Goldfarb, A., and Nikiforov, V. (1993) Rifampicin region revisited. New rifampicin-resistant and streptolydigin-resistant mutants in the beta subunit of Escherichia coli RNA polymerase. J. Biol. Chem. 268, 14820-14825. (40) Garibyan, L., Huang, T., Kim, M., Wolff, E., Nguyen, A., Nguyen, T., Diep, A., Hu, K., Iverson, A., Yang, H., and Miller, J. H. (2003) Use of the rpoB gene to determine the specificity of base substitution mutations on the Escherichia coli chromosome. DNA Repair 2, 593-608. (41) Guengerich, F. P., Fang, Q., Liu, L., Hachey, D. L., and Pegg, A. E. (2003) O6-Alkylguanine-DNA alkyltransferase: Low pKa and reactivity of cysteine 145. Biochemistry 42, 10965-10970. (42) Daniels, D. S., Mol, C. D., Arval, A. S., Kanugula, S., Pegg, A. E., and Tainer, J. A. (2000) Active and alkylated human AGT structures: a novel zinc site, inhibitor and extrahelical binding. DNA damage reversal revealed by mutants and structures of active and alkylated human AGT. EMBO J. 19, 1719-1730. (43) Daniels, D. S., and Tainer, J. A. (2000) Conserved structural motifs governing the stoichiometric repair of alkylated DNA by O6-alkylguanine-DNA alkyltransferase. Mutat. Res. 460, 151163. (44) Goodtzova, K., Kanugula, S., Edara, S., and Pegg, A. E. (1998) Investigation of the role of tyrosine-114 in the activity of human O6-alkylguanine-DNA alkyltransferase. Biochemistry 37, 1248912495. (45) Kanugula, S., Goodtzova, K., Edara, S., and Pegg, A. E. (1995) Alteration of arginine-128 to alanine abolishes the ability of human O6-alkylguanine-DNA alkyltransferase to repair methylated DNA but has no effect on its reaction with O6-benzylguanine. Biochemistry 34, 7113-7119. (46) Sagher, D., and Strauss, B. (1983) Insertion of nucleotides opposite apurinic/apyrimidinic sites in deoxyribonucleic acid during in vitro synthesis: uniqueness of adenine nucleotides. Biochemistry 22, 4518-4526. (47) Kunkel, T. A. (1984) Mutational specificity of depurination. Proc. Natl. Acad. Sci. U.S.A. 81, 1494-1498. (48) Shcherbakova, P. V., Bebenek, K., and Kunkel, T. A. (2003) Functions of eukaryotic DNA polymerases. Sage KE re3, 1-11. (49) Prakash, S., and Prakash, L. (2002) Translesion DNA synthesis in eukaryotes: a one- or two-polymerase affair. Genes Dev. 16, 1872-1883. (50) Friedberg, E. C., Wagner, R., and Radman, M. (2002) Specialized DNA polymerases, cellular survival, and the genesis of mutations. Science 296, 1627-1630. (51) Wheeler, J. B., Stourman, N. V., Armstrong, R. N., and Guengerich, F. P. (2001) Conjugation of haloalkanes by bacterial and mammalian glutathione transferases: mono- and vicinal dihaloethanes. Chem. Res. Toxicol. 14, 1107-1117. (52) Stourman, N. V., Rose, J. H., Vuilleumier, S., and Armstrong, R. N. (2003) Catalytic mechanism of dichloromethane dehalogenase from Methylophilus sp. strain DM11. Biochemistry 42, 1104811056. (53) Edara, S., Kanugula, S., and Pegg, A. E. (1999) Expression of the inactive C145A mutant human O6-alkylguanine-DNA alkyltransferase in E. coli increases cell killing and mutations by N-methylN′-nitro-N-nitrosoguanidine. Carcinogenesis 20, 103-108. (54) Samson, L., Thomale, J., and Rajewsky, M. F. (1988) Alternative pathways for the in vivo repair of O6-alkylguanine and O4-

752


alkylthymine in Escherichia coli: the adaptive response and nucleotide excision repair. EMBO J. 7, 2261-2267. (55) Shuker, D. E., and Margison, G. P. (1997) Nitrosated glycine derivatives as a potential source of O6-methylguanine in DNA. Cancer Res. 57, 366-369. (56) Srivenugopal, K. S., Yuan, X. H., Friedman, H. S., and Ali-Osman, F. (1996) Ubiquitination-dependent proteolysis of O6-methylguanine-DNA alkyltransferase in human and murine tumor cells following inactivation with O6-benzylguanine or 1,3-bis(2-chloroethyl)-1-nitrosourea. Biochemistry 35, 1328-1334.

Liu et al. (57) Xu-Welliver, M., and Pegg, A. E. (2002) Ubiquitin-mediated degradation of alkylated O6-alkylguanine-DNA alkyltransferase. Carcinogenesis 23, 823-830. (58) Horsfall, M. J., Gordon, A. J. E., Burns, P. A., Zielenska, M., van der Vliet, G. M. E., and Glickman, B. W. (1990) Mutational specificity of alkylating agents and the influence of DNA repair. Environ. Mol. Mutagen. 15, 107-122.

TX049958O

O 6-Alkylguanine-DNA Alkyltransferase Has Opposing Effects in

Recommend Documents