Arylamine N-Acetyltransferases: Characterization of the Substrate

Aug 3, 2007 - Lisa F Potts , Alex C Cambon , Owen A Ross , Rosa Rademakers , Dennis W Dickson , Ryan J Uitti , Zbigniew K Wszolek , Shesh N Rai ...
0 downloads 0 Views 607KB Size
Download PDF
1300

Chem. Res. Toxicol. 2007, 20, 1300–1308

Arylamine N-Acetyltransferases: Characterization of the Substrate Specificities and Molecular Interactions of Environmental Arylamines with Human NAT1 and NAT2 Li Liu,† Annette Von Vett,† Naixia Zhang,‡ Kylie J. Walters,‡ Carston R. Wagner,† and Patrick E. Hanna*,† Department of Medicinal Chemistry, UniVersity of Minnesota, 308 HarVard Street SE, Minneapolis, Minnesota 55455, and Department of Biochemistry, Molecular Biology and Biophysics, UniVersity of Minnesota, Minneapolis, Minnesota 55455 ReceiVed May 11, 2007

Arylamine N-acetyltransferases (NATs) are phase II xenobiotic metabolism enzymes that catalyze the detoxification of arylamines by N-acetylation and the bioactivation of N-arylhydroxylamines by O-acetylation. Endogenous and recombinant mammalian NATs with high specific activities are difficult to obtain in substantial quantities and in a state of homogeneity. This paper describes the overexpression of human wild-type NAT2 as a dihydrofolate reductase fusion protein containing a TEV protease-sensitive linker. Treatment of the partially purified fusion protein with TEV protease, followed by chromatographic purification, afforded 2.8 mg of homogeneous NAT2 from 2 L of cell culture. The kinetic specificity constants (kcat/Km) for N-acetylation of arylamine environmental contaminants, some of which are associated with bladder cancer risk, were determined with NAT2 and NAT1. The NAT1/NAT2 ratio of the specificity constants varied almost 1000-fold for monosubstituted and disubstituted alkylanilines containing methyl and ethyl ring substituents. 2-Alkyl substituents depressed N-acetylation rates but were more detrimental to catalysis by NAT1 than by NAT2. 3-Alkyl groups caused substrates to be preferentially N-acetylated by NAT2, and both 4-methyl- and 4-ethylaniline were better substrates for NAT1 than NAT2. NMR-based models were used to analyze the NAT binding site interactions of the alkylanilines. The selectivity of NAT1 for acetylation of 4-alkylanilines appears to be due to binding of the substituents to V216, which is replaced by S216 in NAT2. The contribution of 3-alkyl substituents to NAT2 substrate selectivity is attributed to multiple bonding interactions with F93, whereas a single bonding interaction occurs with V93 in NAT1. Unfavorable steric clashes between 2-methyl substituents and F125 of NAT1 may account for the selective NAT2-mediated N-acetylation of 2-alkylanilines; F125 is replaced by S125 in NAT2. These results provide insight into the structural basis for the substrate specificity of two NATs that play major roles in the biotransformation of genotoxic environmental arylamines. Introduction 1

Acetyl CoA:aromatic amine N-acetyltransferases (NATs, EC 2.3.1.5) are a family of polymorphic enzymes that catalyze several pharmacologically and toxicologically significant reactions (1, 2). The most common of the NAT-catalyzed reactions is the AcCoA-dependent N-acetylation of primary arylamines (aromatic amines), a phase II biotransformation process in which an arylamine (ArNH2) is converted to an arylamide (ArNH* To whom correspondence should be addressed. E-mail: [email protected]. † Department of Medicinal Chemistry. ‡ Department of Biochemistry, Molecular Biology and Biophysics. 1 Abbreviations: AcCoA, acetyl coenzyme A; 2-AF, 2-aminofluorene; 4-ABP, 4-aminobiphenyl; 4-AABP, 4-acetylaminobiphenyl; ANL, aniline; DEAE, diethylaminoethyl; DHFR, dihydrofolate reductase; DMSO, dimethyl sulfoxide; DTT, dithiothreitol; 2,3-DMA, 2,3-dimethylaniline; 2,4-DMA, 2,4-dimethylaniline; 2,5-DMA, 2,5-dimethylaniline; 2,6-DMA, 2,6-dimethylaniline; 3,4-DMA, 3,4-dimethylaniline; 3,5-DMA, 3,5-dimethylaniline; 2-EA, 2-ethylaniline; 3-EA, 3-ethylaniline; 4-EA, 4-ethylaniline; EDTA, ethylenediaminetetraacetic acid; ESI, electrospray ionization; IPTG, isopropyl D-thiogalactopyranoside; 2-MA, 2-methylaniline; 4-MA, 4-methylaniline; MALDI-TOF, matrix-assisted laser desorption ionization time-offlight; MOPS, 3-(N-morpholino)propanesulfonic acid; MTX, methotrexate; NAT, arylamine N-acetyltransferase; PABA, p-aminobenzoic acid; PAS, p-aminosalicylic acid; PNP, p-nitrophenol; PNPA, p-nitrophenyl acetate; Q-TOF, quadrupole time-of-flight; SMZ, sulfamethazine; TEV, tobacco etch virus; TMP, trimethoprim.

COCH3). N-Acetylation of the monoamino arylamines to which humans are commonly exposed is considered a detoxification reaction because acetylation decreases substrate availability for metabolic hydroxylation of the primary amino group, a process that frequently leads to the production of toxic and/or carcinogenic N-arylhydroxylamines (ArNHOH) (3, 4). NATs also catalyze the formation of N-acetoxyarylamines (ArNHOCOCH3) by O-acetylation of N-arylhydroxylamines and by using Narylhydroxamic acids (ArNOHCOCH3) as substrates in an AcCoA-independent N,O-acetyl transfer reaction (5–7). NAcetoxyarylamines are unstable, electrophilic molecules that form covalent bonds with nucleophilic functional groups on DNA and are believed to be responsible, at least in part, for initiation of arylamine-induced carcinogenesis (8). Humans express two NATs, NAT1 and NAT2, which have 81% identical sequences, and both of which exhibit genetic polymorphism (1). Although some arylamine substrates are acetylated by both NAT1 and NAT2, each of the NATs exhibits specificity with regard to other substrates. For example, PABA and PAS are acetylated selectively by NAT1, whereas SMZ is preferentially acetylated by NAT2 (9). Because of the lack of a reliable source of homogeneous NATs, substrate specificity analyses, as well as bioactivation studies, typically have been conducted with either partially purified or unpurified cytosolic

10.1021/tx7001614 CCC: $37.00  2007 American Chemical Society Published on Web 08/03/2007

Substrate Specificity of NATs

fractions from mammalian cells or with NATs expressed in cytosols of bacterial or mammalian cells (9–11). A number of reports establish that purification of NATs from mammalian tissues consistently results in a low recovery of activity and incompletely purified protein (12–17). Similarly, purification of a recombinant human NAT1 afforded a low yield of protein with low specific activity (18). Our laboratory has developed novel protocols for the overexpression and purification of several mammalian NATs (19–21). The availability of homogeneous recombinant NATs with high specific activities has made it possible to elucidate the catalytic mechanism of hamster NAT2, to determine the details of the self-catalyzed inactivation of hamster NAT1, hamster NAT2, and human NAT1 by carcinogenic N-arylhydroxamic acids, to investigate the effects of an NAT polymorphism on aggregation and ubiquitylation, and to conduct a NMR structural analysis of hamster NAT2 (22–27). The protocols also have been used to produce hamster NAT2 and partially purified human NAT1 for high-throughput screening of potential substrates (28). In this paper, we report the details of the first protocol for the production of homogeneous human NAT2 (NAT2 4) in milligram quantities. The availability of human “wild-type” NATs (NAT1 4 and NAT2 4) has made it possible to determine kinetic specificity constants for acetylation of a group of environmental arylamines for which human exposure is welldocumented, and for some of which an association with bladder cancer risk has been determined. A NMR-based model is used to explain key differences in substrate specificity and isoform selectivity. These results provide new insight into the structural and molecular determinants of detoxification of a group of environmentally ubiquitous arylamines via N-acetylation.

Experimental Procedures Caution: 4-ABP and alkylanilines should be handled in accordance with NIH Guidelines for the Laboratory Use of Chemical Carcinogens (29). Materials and Methods. Recombinant human NAT1 was overexpressed and purified as described previously (21). pAnisidine, 4-aminobiphenyl (4-ABP), 3-(N-morpholino)propanesulfonic acid (MOPS), p-aminobenzoic acid (PABA) (sodium salt), p-aminosalicylic acid (PAS), p-nitrophenyl acetate (PNPA), methotrexate (MTX) agarose, lysozyme, ampicillin, chloramphenicol, and isopropyl β-D-thiogalactopyranoside (IPTG) were purchased from Sigma (St. Louis, MO). Other arylamines were from Aldrich (Milwaukee, WI) or Acros Organics (Morris Plains, NJ). Trimethoprim (TMP) was purchased from MP Biomedicals (Irvine, CA). Competent BL-21 Codon Plus RIL Escherichia coli cells were purchased from Stratagene (La Jolla, CA), and supercompetent E. coli DH5R cells were from Invitrogen (Carlsbad, CA). Restriction enzymes were purchased from New England Biolabs (Ipswich, MA). DEAE-Sepharose Fast Flow anion-exchange resin was obtained from Amersham Pharmacia (Piscataway, NJ). Dialysis tubing (Spectra/Por membrane, molecular weight cutoff of 12000–14000) was from Spectrum Laboratories (Rancho Dominguez, CA). Oligodeoxynucleotides were synthesized by the MicroChemical Facility of the University of Minnesota, and DNA sequences were determined by the BioMedical Genomics Center of the University of Minnesota. Protein concentrations were determined by the method of Bradford (30). Spectrophotometric data were acquired on a Varian Cary 50 UV–vis spectrophotometer. All buffers were degassed under vacuum, and all incubations were conducted under aerobic conditions. The concentrations of all solutions reported as percentages are volume per volume. Proteins were concentrated with Amicon ultrafiltration cells (model 202, 52, or 12). Kinetic data were analyzed with the JMP IN software suite

Chem. Res. Toxicol., Vol. 20, No. 9, 2007 1301 (SAS Institute, Inc.), and the statistical significance of differences between means was assessed with a Student’s t test. Construction of Plasmid pPH90D. To allow use of the XhoI and XbaI restriction cloning sites in the pPH70D plasmid (19), a silent mutation was incorporated in the coding region of human nat2*4 to eliminate an XhoI restriction site by using the QuickChange site-directed mutagenesis kit (Stratagene). The oligonucleotide primer was 5′-G TGG CAG CCG CTA GAA TTA ATT TCT GGG-3′. The coding region of human nat2*4 was amplified by PCR with the forward primer 5′-GCGACGTCTAGAATGGACATTGAAGCATATTTTG-3′, which includes an XhoI restriction site, and the reverse primer 5′-CGCTCGAGCGCGCTAAATAGTAAGGGATCCATCACCAGG-3′, which includes an XbaI restriction site. The PCR was conducted with 2 units of Taq polymerase, 1 µg of genomic DNA as a template, 200 µM dNTP, and 30 pmol each of the primers. The reaction mixtures were heated at 94 °C for 2 min, followed by 30 cycles for 30 s at 94 °C, 45 s at 50 °C, and 2 min at 72 °C. To ensure complete extension, the reaction mixture was incubated at 72 °C for 10 min after the 30 cycles. The PCR products were purified with a QIAquick PCR purification kit (Qiagen). The purified PCR product and the plasmid vector pPH70D were digested with XhoI and XbaI separately. The digestion products were isolated with a QIAquick gel extraction kit (Qiagen). The vector DNA was ligated with the human nat2*4 insert with T4 ligase (Invitrogen). Supercompetent E. coli DH5R cells were transformed with the ligation mix according to the supplier’s protocol, and ampicillinresistant plasmid-containing clones were selected on LB agar plates. The plasmids from positive colonies were isolated with the Wizard Plus SV Miniprep DNA purification system. The cloned sequence was confirmed by automated DNA sequencing, and the plasmid was designated pPH90D. Construction of Plasmid pPH100D. The oligonucleotide insert for the TEV protease cleavage linker encodes a space arm region and a TEV protease cleavage site, Asp-Tyr-Asp-Ile-Pro-Thr-ThrGlu-Asn-Leu-Tyr-Phe-Gln-Gly. The single-stranded DNA (CTAGGTGATTACGATATCCCGACAACGGAAAACCTGTATTTTCAGGGCC and TCGAGGCCCTGAAAATACAGGTTTTCCGTTGTCGGGATATCGTAATCAC) was annealed by being heated to 65 °C and the mixture allowed to cool slowly to room temperature. Plamid pPH90D was digested with AvrII and XhoI and purified with a QIAquick gel extraction kit (Qiagen). The vector was ligated with the annealed double-stranded DNA insert with T4 ligase. Supercompetent E. coli DH5R cells were transformed and selected as described above. The constructed plasmid was designated pPH100D; the sequence was confirmed by automated DNA sequencing. Expression and Purification of Human Recombinant NAT2. Competent BL-21 Codon Plus RIL E. coli cells were transformed with plasmid pPH100D according to the supplier’s protocol. The cells were grown at 37 °C in TB to an OD600 of 1.2. IPTG (final concentration of 100 µM) was added after the culture had been quickly cooled to 17 °C. The growth was continued for 16 h at 17 °C. The cells were harvested by centrifugation at 5000g for 15 min at 4 °C. The cell pellets were frozen at -80 °C or on dry ice. All purification steps were performed at 4 °C in degassed buffers. The cell pellets from 2 L of BL-21 (RIL)/pPH100D culture were lysed as reported previously (21). The soluble fraction of the bacterial cell lysate was acidified to pH 6.5 by dialysis against 6 L (three 2 L portions) of PE buffer [20 mM potassium phosphate (pH 6.5), 1 mM EDTA, and 1 mM DTT]. The acidified soluble fraction was applied to a methotrexate (MTX) affinity column (10 mm × 200 mm) that had been equilibrated with PE buffer (pH 6.5) and presaturated by treatment with the soluble fraction from transformed cells. The column was washed with 1 L of high-salt PE buffer [1 M NaCl (pH 6.5)], followed by elution with 150 mL of trimethoprim (TMP) buffer [300 µM TMP, 1 M NaCl, 20 mM potassium phosphate (pH 9.0), 1 mM EDTA, and 1 mM DTT]. Fractions (9 mL) were collected at a flow rate of 1 mL/min. The absorbance at 280 nm was determined for each fraction, and an aliquot (1 µL) of each fraction was assayed for transacetylation

1302 Chem. Res. Toxicol., Vol. 20, No. 9, 2007 activity. The fractions containing the highest levels of activity were combined and concentrated to 2 mg/mL of protein. To separate the fusion protein from the truncated fusion protein and other contaminating proteins, the protein solution from the MTX column was dialyzed against 6 L (three 2 L portions) of PE buffer (pH 7.4) and loaded onto a DEAE anion-exchange column (25 mm × 200 mm) which was packed with Fast Flow DEAE resin and had been equilibrated with PE buffer (pH 7.4). The column was washed with 200 mL of PE buffer (pH 7.4), followed by a 0 to 0.4 M NaCl gradient in PE buffer (pH 7.4, 720 mL). Fractions (9 mL) were collected at a flow rate of 1 mL/min. The fractions with the highest levels of transacetylation activity were combined and concentrated to 1.5 mg/mL of protein. The concentrated solution was dialyzed against 6 L (three 2 L portions) of TEV protease cleavage buffer [50 mM Tris-HCl (pH 8.0), 0.5 mM EDTA, and 1 mM DTT]. Human NAT2 was cleaved from the fusion protein by a 12 h incubation with recombinant TEV protease (350 units/mg of fusion protein) at 4 °C. The solution was dialyzed against PE buffer (pH 7.4) and loaded onto a second DEAE column. The column was washed with 200 mL of PE buffer (pH 7.4), followed by a 0 to 0.2 M NaCl gradient in PE buffer (pH 7.4). Fractions (9 mL) were collected at a flow rate of 1 mL/min, and the fractions with the highest transacetylation activity were combined and concentrated to 1 mg/mL of protein. The second DEAE chromatography step resulted in the purification of human NAT2 to homogeneity according to SDS–PAGE and Nano-ESI-Q-TOF MS. The homogeneous recombinant protein was stored at -80 °C as a 10% glycerol/PE (pH 7.4) solution. NAT2 Activity Assay. The assay mixture contained NAT2 (96 ng/mL, 2.84 nM), PNPA (2 mM) as the acetyl donor, p-anisidine (1 mM) as the acetyl acceptor, and MOPS buffer [100 mM (pH 7.0), 150 mM NaCl, and 0.1 mM DTT] in a final volume of 500 µL. The reaction was initiated by addition of PNPA dissolved in DMSO (5 µL). The final concentration of DMSO was 1%. Incubations were conducted at either 23 or 37 °C in acryl cuvettes in a Varian Cary 50 UV–vis spectrophotometer equipped with a circulation water bath. The reaction rates were determined by monitoring the increase in absorbance at 400 nm, due to the formation of p-nitrophenol (PNP) (ε400nm ) 9400 M-1 cm-1). The results were corrected for nonenzymatic PNPA hydrolysis by conducting the reaction in the absence of enzyme. The specific activities were expressed as micromoles of product formed per milligram of protein per minute. NAT1 Activity Assay. The assay was conducted as described for NAT2, except the final concentration of NAT1 was 0.5 µg/mL (14.6 nM) and PABA (0.5 mM) was used as the acetyl acceptor. Substrate Specificities of Human NAT1 and NAT2. In a final volume of 500 µL, purified NAT1 or NAT2 was incubated with PNPA (2 mM) and various concentrations of arylamine substrates in MOPS buffer [100 mM (pH 7.0), 150 mM NaCl, and 0.1 mM DTT]. Protein concentrations were adjusted to obtain reaction rates that were linear with time during the incubation period. The reactions were initiated by addition of PNPA dissolved in DMSO. The final concentration of DMSO was 1%. The initial velocities of the reactions were determined at 37 °C as described for the NAT activity assays. The nonenzymatic hydrolysis of PNPA was assessed by conducting incubations in the absence of enzyme. Substrate concentrations for NAT1 were as follows: 3000 µM SMZ, 40–350 µM PABA, 25–400 µM PAS, 20–1000 µM 4-ABP, 20–600 µM 2-AF, 200–10000 µM ANL, 100–1000 µM 4-MA, 100–1000 µM 4-EA, 50–1000 µM 3,4-DMA, 70–2000 µM 3,5-DMA, 80–1500 µM 3-EA, 500–10000 µM 2-MA, 300–7000 µM 2,3-DMA, 300–7000 µM 2,4-DMA, 3000–50000 µM 2,5-DMA, 50000 µM 2,6-DMA, and 20000 µM 2-EA. Substrate concentrations for NAT2 were as follows: 400–8000 µM SMZ, 4000–60000 µM PABA, 2000–12000 µM PAS, 50–1000 µM 4-ABP, 50–600 µM 2-AF, 800–10000 µM ANL, 500–15000 µM 4-MA, 500–8000 µM 4-EA, 100–1200 µM 3,4-DMA, 50–1000 µM 3,5-DMA, 250–8000 µM 3-EA, 2000–15000 µM 2-MA, 1000–8000 µM 2,3-DMA, 600–10000 µM 2,4-DMA, 200–10000 µM 2,5-DMA, 50000 µM 2,6-DMA, and 200–8000 µM 2-EA.

Liu et al. Nano-ESI-Q-TOF MS of Purified Recombinant Human NAT2. Human NAT2 samples were desalted with Millipore C18 ZipTips according to the supplier’s instructions for peptides in solution with a high concentration of salt. ESI mass spectra were obtained with a QSTAR Pulsar I quadrupole time-of-flight (Q-TOF) mass spectrometer (ABI, Foster City, CA) equipped with a nanoESI source (Protana Engineering). The ESI voltage was 1000 V; the TOF region acceleration voltage was 4 kV, and the injection pulse repetition rate was 6.0 kHz. Mass spectra were the average of 300 scans collected in the positive ion mode. Protein zero-charge mass was acquired by deconvolution of the series of multiply charged protein peaks from m/z 700 to 3000 with the Bayesian Reconstruct tool in the BioAnalyst software package (ABI). Modeling of Arylamines into the Catalytic Cavities of Human NAT1 and NAT2. The lowest-energy model structure of the human NAT1–PABA complex obtained as previously described was used to produce models for human NAT1 complexed with 4-MA, 3,5-DMA, and 3-EA (27). The substrates were generated by using the BUILDER module in Insight II, and their aryl ring carbon atoms were superimposed onto those of PABA. Ligplot was used to analyze the interactions between human NAT1 and the bound substrate (31). Similarly, the lowest-energy model structure of the human NAT1–PABA complex obtained as described was used to generate the structural model of human NAT2 complexed with 3,5-DMA (27). In particular, the MODELER module of Insight II was used to produce a model structure of human NAT2 based on that of human NAT1 complexed with PABA. The backbone atoms of this NAT2 structure were superimposed onto those of human NAT1 in its structural complex with 3,5-DMA to produce the NAT2–3,5-DMA structure. Ligplot was used to analyze the interactions between human NAT2 and the bound substrate (31).

Results Cloning, Expression, and Purification of Human NAT2. The protocols for production of homogeneous mammalian NATs previously reported by our laboratory involve the construction of plasmids that encode a FLAG mutant DHFR–NAT protein, which contains a thrombin-sensitive linker between DHFR and NAT (19–21). After partial purification of the fusion proteins, NATs are liberated by thrombin-catalyzed hydrolysis of the linker and are readily purified to homogeneity. Substantial modification of the protocols was required for the overexpression and purification of human NAT2. Plasmid pPH70D contains two restriction sites, KpnI and XhoI, upstream of the hamster NAT2 gene (19). Both restriction sites also are present in the human NAT2 coding region, which necessitated the incorporation of a silent mutation, T401 to G, to permit the use of the XhoI site in the construction of plasmid pPH90D. Construction of plasmid pPH90D was accomplished by replacement of the hamster NAT2 gene of pPH70D with nat2*4. Expression of the FLAG mutant DHFR–NAT2 4 fusion protein in E. coli was successful, but incubation of the fusion protein with thrombin for 19 h resulted in a loss of 25% of the NAT activity (data not shown). Hamster and human NATs contain a potential thrombin cleavage site at the R64-G65 pair. Hamster NAT1 and NAT2, as well as human NAT1, however, were successfully cleaved from the fusion proteins by treatment with thrombin (19–21). Human NAT2 contains a second potential thrombin cleavage site (R156-G157), which is not present in the other NATs and which likely is responsible for the loss of specific activity upon incubation with thrombin. Plasmid pPH100D was constructed by replacement of the thrombin-sensitive linker of pPH90D with a TEV proteasesensitive linker. It was found that the presence of the TEV protease cleavage linker increased the level of expression of the soluble NAT2 fusion protein approximately 2-fold and that

Substrate Specificity of NATs

Chem. Res. Toxicol., Vol. 20, No. 9, 2007 1303

Figure 1. SDS–PAGE analysis of human NAT2 purification: lane A, molecular mass markers; lane B, BL21 (RIL)/pPH100D cell lysate (without IPTG induction); lane C, BL21 (RIL)/pPH100D cell lysate (with IPTG induction); lane D, partially purified fusion protein after the MTX column; lane E, partially purified fusion protein after the first DEAE column; lane F, protein mixture after TEV protease cleavage; and lane G, purified human NAT2 after the second DEAE column.

Table 1. Purification of Human NAT2a

purification step

anisidine/ PNPA total specific activity total activity (µmol/ % fold protein (µmol (mg) mg–1 min–1) min) recovery purification

cell lysate 1560 MTX column 62 first DEAE column 32 second DEAE 2.8 column a

1.8 31 73 689

2980 2570 2340 1920

100 86 78 64

1.0 17 40 373

These data were obtained from 2 L of cell culture.

incubation of the fusion protein with TEV protease yielded human NAT2 without loss of activity. Conditions for expression of the soluble fusion protein were optimized and are described in Experimental Procedures. We previously reported that purification of a mutant DHFR–hamster NAT2 fusion protein by MTX affinity chromatography resulted in loss of more than 90% of the NAT2 activity because of an apparent irreversible binding of hamster NAT2 to MTX (19). In this study, it again was found that recovery of the DHFR–human NAT2 fusion protein after MTX affinity chromatography was very low unless the column had been previously treated with bacterial cell cytosol in which the fusion protein had been expressed. Thus, passage of cell cytosol containing the recombinant fusion protein through the cytosoltreated MTX affinity column, followed by elution with TMP, afforded a 17-fold purification of the 60 kDa fusion protein (Figure 1, lane D), and the treated column could be used repeatedly in subsequent purifications. The efficiency of the TEV protease-catalyzed cleavage of the DHFR–NAT2 fusion protein was enhanced by further partial purification of the fusion protein by DEAE anion exchange chromatography (Figure 1, lane E; Table 1). After incubation with TEV protease, the 60 kDa fusion protein band was no longer present, and the mixture consisted primarily of NAT2 (approximately 33 kDa) and FLAG-DHFR (20 kDa) (Figure 1, lane F). Passage of the reaction mixture through a final DEAE column yielded homogeneous human recombinant NAT2 in the amount of 2.8 mg from 2 L of cell culture (Table 1). The deconvoluted Nano-ESI-Q-TOF mass spectrum of purified

Figure 2. Structures of arylamines (aromatic amines) used in this study.

NAT2 exhibited a molecular mass of 33 844.0 Da, which agrees well with the theoretical mass of 33 841.79 Da, including the three additional amino acids (Gly-Leu-Glu) at the N-terminus of the recombinant protein. Substrate Specificities and Arylamine–NAT Binding Models. The substrate specificities of human NAT1 and NAT2 were determined by using a fixed concentration of PNPA as the acetyl donor, as previously described (21). The kinetic constants, kcat,app and Km,app, were calculated by using nonlinear regression analysis to fit the initial velocities of the N-acetylation reactions to the following equation: kobs ) kcat,app[ArNH2]/ (Km,app + [ArNH2]). Because NAT catalysis occurs through a ping-pong bi-bi kinetic mechanism, the kcat,app/Km,app ratio, determined from this analysis, is the specificity constant (kcat/ Km) of the enzyme for the arylamine substrate and is independent of kinetic characteristics of the acetyl donor (21, 23). The kcat,app and Km,app values determined under the conditions of these experiments may differ from the corresponding values determined with physiologically relevant concentrations of AcCoA as the acetyl donor. The values of the kcat,app/Km,app ratios derived from this data are, however, valid quantitative descriptors of the specificities of the NATs for catalyzing the N-acetylation of the arylamines (21). The arylamines investigated as NAT substrates are shown in Figure 2, and the specificity constants are reported in Table 2. Initially, kinetic experiments were conducted with SMZ, PAS, PABA, 4-ABP, and 2-AF to confirm that the substrate specificity of the homogeneous recombinant NAT2 was consistent with the published substrate specificities of human NATs. Sulfamethazine (SMZ), a known NAT2 substrate, was N-acetylated by recombinant NAT2, whereas the rate of acetylation of SMZ by NAT1 was very low. Also consistent with the reported substrate specificities of NAT2 and NAT1 was the very low efficiency of NAT2, compared to that of NAT1, in catalyzing the N-acetylation of PABA and PAS (9, 11, 12). The arylamine carcinogens, 4-ABP and 2-AF, were substrates for both NAT2 and NAT1, with NAT1 catalyzing the reactions somewhat more effectively than NAT2, as previously reported (Table 2) (11). Two principal objectives of this research were to determine the relative substrate specificities of human NAT1 and NAT2 for N-acetylation of aniline and a series of ring-substituted alkylanilines, all of which are confirmed environmental con-

1304 Chem. Res. Toxicol., Vol. 20, No. 9, 2007

Liu et al.

Table 2. Substrate Specificities of Human NAT1 and Human NAT2a NAT1

NAT2

compound

Km,app (µM)

kcat,app (s-1)

kcat/Km (s-1 mM-1)

Km,app (µM)

kcat,app (s-1)

kcat/Km (s-1 mM-1)

[kcat/Km(NAT1)]/ [kcat/Km(NAT2)]

SMZ PABA PAS 4-ABP 2-AF ANL 4-MA 4-EA 3,4-DMA 3,5-DMA 3-EA 2-MA 2,3-DMA 2,4-DMA 2,5-DMA 2,6-DMA 2-EA

NDb 84 ( 7.9 59 ( 9.2 191 ( 13 109 ( 11 2490 ( 140 483 ( 28 205 ( 20 352 ( 21 742 ( 27 576 ( 27 2320 ( 170 1110 ( 47 2060 ( 130 6730 ( 880 NDc NDd

NDb 298 ( 10 591 ( 32 243 ( 5.4 449 ( 17 281 ( 5.1 303 ( 8.2 430 ( 14 461 ( 12 308 ( 5.2 310 ( 7.0 12 ( 0.33 27 ( 0.37 67 ( 1.7 5.7 ( 0.23 NDc NDd

3550 10000 1270 4120 113 627 2100 1310 415 538 5.2 24 32 0.85

5390 ( 520 152000 ( 39100 5390 ( 530 486 ( 23 286 ( 38 11200 ( 1790 11800 ( 1290 3270 ( 390 688 ( 30 280 ( 21 1320 ( 150 7180 ( 130 5530 ( 590 5030 ( 610 3120 ( 305 NDc 2110 ( 180

387 ( 20 38 ( 7.5 27 ( 1.1 256 ( 5.6 759 ( 46 711 ( 70 1600 ( 120 700 ( 34 800 ( 18 1220 ( 37 1960 ( 68 111 ( 8.1 256 ( 16 661 ( 39 313 ( 12 NDc 61 ( 5.0

72 0.25 5.0 527 2650 63 136 214 1160 4360 1490 15 46 131 100

14200 2000 2.41 1.55 1.79 4.61 9.80 1.13 0.10 0.36 0.35 0.52 0.24 0.01

a

29

b

Results are expressed as the means ((standard deviation) of three experiments. ND, not determined. At a SMZ concentration of 3 mM, the rate of N-acetylation was 7.1 µmol mg–1 min–1. The limit of detection was 2–3 nmol. The lowest quantifiable rate was defined as an absorbance change of 0.1 min–1. c Not determined. N-Acetylation was not detectable at a 2,6-DMA concentration of 50 mM. d Not determined. The rate of N-acetylation of 2-EA (20 mM) by NAT1 was 2.2 µmol mg–1 min–1.

Figure 3. Model structure of human NAT1 bound to 4-methylaniline (4-MA). Panel A illustrates the position of 4-MA in the protein’s interior with secondary structural elements and the bonds of the side chain heavy atoms of the catalytic triad, C68, H107, and D122, as well as V216 and 4-MA. (B) An expanded view is provided to illustrate the contacts between human NAT1 and 4-MA. Carbon, nitrogen, oxygen, sulfur, and hydrogen atoms are colored white, blue, red, yellow, and pink, respectively, and a hydrophobic contact with V216 is highlighted.

taminants, and to analyze key substrate–activity differences with regard to the structural characteristics of both the substrates and the enzymes (32–35). Aniline (Figure 2) was a slightly better substrate for NAT1 than NAT2, whereas the 4-methyl group of 4-MA (p-toluidine) and the 4-ethyl group of 4-EA caused a shift in the substrate specificity substantially toward NAT1. The selectivity of NAT1 over NAT2 for N-acetylation of 4-MA had been previously observed by Hein and co-workers (11). To gain insight into the selectivity of NAT1 for N-acetylation of 4-MA, the compound was modeled into the catalytic cavity of NAT1 as described in Experimental Procedures. As shown in Figure 3, the 4-methyl group of 4-MA appears to engage in favorable nonbonded contacts with a side chain methyl group of V216. This favorable hydrophobic interaction is not available with NAT2, which contains S216, rather than V216. 3,4-DMA was N-acetylated efficiently, and with similar specificity constants, by both NAT1 and NAT2 (Table 2). Thus, the addition of a methyl group to the 3-position of 4-MA caused a 9-fold increase in the NAT2 specificity constant and a 2-fold increase in the NAT1 specificity constant, indicating that the 3-methyl group contributes favorable binding interactions with both NATs, although the effect is greater for NAT2. The influence of 3-alkyl substituents on the N-acetylation of the ring-

substituted anilines was also reflected by the results obtained with 3,5-DMA and 3-EA, both of which are N-acetylated by NAT1 and NAT2 and both of which are preferential substrates for NAT2. In particular, 3,5-DMA had a 10-fold greater specificity constant with NAT2 than with NAT1 (Table 2). Model structures of NAT1 and NAT2 complexed with 3,5DMA are shown in panels A and B of Figure 4, respectively. In NAT1 (Figure 4A), favorable bonding interactions can occur between one of the methyl groups and residues S215 and L209, while the other methyl group displays probable bonding with F217, F125, and V93. With NAT2 (Figure 4B), the bonding interactions of one of the methyl groups with S215 and L209 are similar to those in NAT1 (Figure 4A), and attractive interactions appear to exist between F217 and both the 3-methyl and 5-methyl substituents. The interaction with F125, which plays a role in the complexation of 3,5-DMA with NAT1 (Figure 4A), cannot occur in the active site of NAT2 due to the replacement of F125 with S125. The loss of the F125 bonding with 3,5-DMA, however, appears to be compensated by interactions with F93 that are not available in NAT1, which contains V93, rather than F93. As illustrated in Figure 4B, the model structure of human NAT2 bound with 3,5-DMA indicates that the aromatic ring of F93 forms several nonbond contacts with a methyl group of 3,5-DMA. The additional hydrophobic interactions between 3,5-DMA and residues in the binding cavity of NAT2 are thought to contribute to the 10-fold higher specificity constant of NAT2, compared to that of NAT1. The model structures of human NAT1 bound to 3-EA in two different orientations are shown in panels C and D of Figure 4. In this model, the 3-ethyl group engages in attractive interactions with the same hydrophobic amino acid residues as the methyl groups of 3,5-DMA (Figure 4A,B), but there are a greater number of noncontact bonds that can occur with the larger and more hydrophobic ethyl group than with a methyl substituent. NAT2, however, exhibits an almost 3-fold selectivity over NAT1 for N-acetylation of 3-EA, a result that may be attributed, at least in part, to the additional bonding interactions that are likely to occur between the 3-ethyl group of 3-EA and F93 of NAT2. The presence of a 2-methyl substituent in the alkylaniline structures caused a substantial lowering of the kcat/Km values

Substrate Specificity of NATs

Chem. Res. Toxicol., Vol. 20, No. 9, 2007 1305

Figure 4. (A) Expanded view of 3,5-dimethylaniline (3,5-DMA) bound to human NAT1. (B) Expanded view of 3,5-DMA bound to human NAT2. (C and D) Model structures of 3-ethylaniline (3-EA) bound to NAT1 in two different orientations. The labeling and color scheme are the same as in Figure 3. The lengths of hydrophobic contacts are represented by the black lines.

obtained with both NAT1 and NAT2, as shown in Table 2 for 2-MA, 2,3-DMA, 2,4-DMA, and 2,5-DMA. For example, the specificity constant of 2-MA was 22-fold lower than that of ANL with NAT1 and 4-fold lower with NAT2. 2,6-DMA was not a substrate for either NAT1 or NAT2, and the N-acetylation of 2-EA was catalyzed by NAT2, but not by NAT1. The finding that the kcat/Km for N-acetylation of 2-MA (o-toluidine) by NAT2 is 3-fold greater than that obtained with NAT1 is consistent with the 3-fold difference in specific activity obtained with NATs expressed in bacterial cell lysates (11). The effect of 2-alkyl substituents on NAT-catalyzed Nacetylation of alkylanilines can result from a combination of steric hindrance to the development of the transition state for formation of the tetrahedral intermediate between the enzymebound thioacetyl ester (Cys68-COCH3) and the arylamine substrate and any unfavorable steric clashes that occur between the 2-alkyl substituent and amino acid side chains in the NAT binding sites (23). The NMR-based structural modeling results of Zhang et al. (27) demonstrated that unfavorable interactions of the 2-methyl group of 2-MA with F125 in NAT1 may hinder binding of 2-MA in the catalytic cavity. The smaller S125 residue of NAT2 causes less hindrance to complex formation with 2-MA and contributes to the higher specificity constant for N-acetylation of 2-MA by NAT2 (Table 2) (27). All of the 2-substituted alkylanilines were preferentially acetylated by NAT2 except 2,6-DMA, which was not a substrate for either NAT (Table 2). 2,4-DMA was the best substrate for NAT1 among the anilines that contained a 2-methyl substituent, indicating the influence of bonding between the 4-methyl group and V216 of NAT1 (Figure 3), but the specificity constant for 2,4-DMA with NAT2 was 4-fold greater than that with NAT1, which appears to reflect the more favorable interaction of the 2-methyl group of 2,4-DMA in the active site of NAT2. The specificity constant for 2,5-DMA, obtained with NAT2, is more than 100-fold greater than the specificity constant exhibited by NAT1. The relative specificities of NAT2 and NAT1 for N-acetylation of 2,5-DMA also can be rationalized by the steric clash between the 2-methyl group and F125 of NAT1 and the several favorable noncontact bonds formed between the 5-methyl group and F93 of NAT2 (Figure 4B).

The relevance of the more favorable binding of 2-alkylanilines in the active site of NAT2, compared to NAT1, is emphasized by the results obtained with 2-EA, which was N-acetylated by NAT2, but not by NAT1 (Table 2). The steric effect of a 2-alkyl substituent on the formation of the tetrahedral intermediate in the catalytic step of N-acetylation would be expected to be similar, if not identical, for reactions catalyzed by both NAT1 and NAT2 (23). Therefore, the failure of NAT1 to catalyze the N-acetylation of 2-EA is most likely attributable to highly unfavorable steric interactions with F125, as previously shown in the model of 2-MA bound to NAT1 (27).

Discussion Arylamine N-acetyltransferases (NATs) catalyze several important phase II biotransformation reactions, including the N-acetylation of arylamines and hydrazides, as well as the O-acetylation of N-arylhydroxylamines (1, 2). The molecular and structural characteristics of human NATs that influence their capacities for N-acetylation of environmental arylamines to which humans are commonly exposed are not well understood. The execution of studies aimed at elucidating the details of interactions of NAT with toxic and carcinogenic arylamines and their metabolites requires reliable access to homogeneous NATs with high specific activities. The outstanding progress in overexpression and purification of prokaryotic NATs notwithstanding, the development of effective protocols for production of large quantities of purified human recombinant NATs remains a challenge (36–38). We reported the overexpression and purification of human NAT1 as a DHFR fusion protein with a thrombin-sensitive linker that permitted cleavage of the fusion protein and facile purification of NAT1 in a yield of 4 mg/L of E. coli cell culture (21). The expression of a soluble DHFR–NAT2 fusion protein containing a thrombin-sensitive linker was also successful, but attempts to free the NAT2 from the fusion protein by treatment with thrombin consistently afforded unsatisfactory yields of recombinant NAT2. As NAT2 contains a potential thrombin cleavage site (R156G157) that is not present in NAT1, plasmid pPH100D, which encodes a TEV proteasesensitive linker, was constructed. After partial purification, the overexpressed fusion protein was cleaved with TEV protease,

1306 Chem. Res. Toxicol., Vol. 20, No. 9, 2007

and the NAT2 was purified by ion exchange chromatography (Table 1). This is the only published method for the production of milligram quantities of homogenous human recombinant NAT2 with high specific activity. Arylamines (aromatic amines) are ubiquitous environmental contaminants, some of which are carcinogens, and several of which are associated with bladder cancer risk in humans (32, 45). Aniline and the arylamine carcinogen 2-MA (otoluidine) have been detected in human milk from both nonsmokers and smokers of tobacco, and numerous arylamine–hemoglobin adducts have been analyzed and quantified as biomarkers of environmental exposure (32, 35, 40). The principal source of human exposure to arylamines in the environment is believed to be tobacco smoke (41–44). Other exposure sources include the workplace and, possibly, the use of hair dyes (34, 39, 45). The metabolic transformation of therapeutic agents also can result in exposure to arylamines; patients who were treated with the local anesthetics lidocaine and prilocaine exhibited significant increases in the level of 2,6DMA–hemoglobin adducts and 2-MA–hemoglobin adducts, respectively (46, 47). Gan and co-workers analyzed the arylamine–hemoglobin adducts formed by nine alkylanilines in a case-control study and found that the hemoglobin adduct levels of 2,6-DMA, 3,5DMA, and 3-EA were independently correlated with bladder cancer risk in both smokers and nonsmokers (32). Because NAT-catalyzed N-acetylation is the major biotransformation reaction involved in the disposition and detoxification of many arylamines, we determined the specificity constants with human NAT1 and NAT2 for the nine alkylanilines studied by Gan and co-workers, as well as for ANL, 2-MA, and 4-MA, each of which is a prominent component of tobacco smoke (32, 44). The data presented in Table 2 reflect several striking effects of small alkyl substituents on the relative abilities of NAT1 and NAT2 to acetylate substituted anilines. One pronounced effect is the influence of 3-alkyl substituents on NAT2 selectivity. Whereas 4-MA is a preferential substrate for NAT1, the incorporation of a meta-methyl group, as in 3,4-DMA, increases the kcat/Km value with NAT2 by almost 9-fold over that observed for acetylation of 4-MA (Table 2). 3,5-DMA, which contains two meta-methyl substituents, exhibits a specificity constant with NAT2 that is 10-fold greater than the value obtained with NAT1. As shown in Figure 4B, favorable bonding interactions between small meta-alkyl substituents and F93 in the binding pocket of NAT2 appear to contribute substantially to the selectivity of NAT2 for N-acetylation of the meta-substituted alkylanilines. Site-directed mutagenesis of NAT1 revealed that F125 is important for substrate binding and contributes to substrate selectivity (48). The NMR-based model of 4-ABP complexed with acetylated NAT1 illustrated the role of F125 bonding with the 4-phenyl substituent (27). The models of 3,5-DMA and 3-EA in the active site of NAT1 (Figure 4A,C) serve to emphasize the contributions of favorable interaction between F125 and the hydrophobic ring substituents of the two alkylanilines. The more numerous and more favorable interactions of the alkyl groups of 3,5-DMA and 3-EA with F93 in the NAT2 binding pocket (Figure 4B), however, appear to be important factors in the preferential NAT2-catalyzed N-acetylation of the 3-alkylanilines. The proposed contribution of F93 to substrate binding in the catalytic cavity of NAT2 is consistent with the results of Savulescu et al. (49), who found that docking 2-AF to a threedimensional structural model of NAT2 indicated favorable hydrophobic contacts with F93.

Liu et al.

Two additional and prominent effects of ring substituents resulted in the preferential N-acetylation of 4-MA and 4-EA by NAT1 and the selective N-acetylation of the 2-methyl- and 2-ethyl-substitued anilines by NAT2. The selectivity of NAT1 for 4-MA and 4-EA is proposed to be due, in large part, to favorable interactions of the 4-alkyl groups with V216 (Figure 3). A model (not shown) of 4-EA bound to NAT1 indicates possible bonding interactions of the methyl group of the 4-substituent with the carbonyl carbon of S215, with C2 and C3 of S215, and with the methyl group of T103 and C6 of K100. These interactions, which are not available to 4-MA, may contribute to the ability of NAT1 to N-acetylate 4-EA more readily than it N-acetylates 4-MA. K100, T103, and S215 are conserved in NAT2; the preferential NAT2-catalyzed Nacetylation of 4-EA, compared to that of 4-MA and ANL, may be facilitated by interaction of the methyl group of the 4-ethyl substituent with these residues. The selectivity of NAT2 for acetylation of 2-MA, 2,3-DMA, 2,4-DMA, 2,5-DMA, and 2-EA is attributed to a more favorable accommodation of the 2-substituents by the NAT2 binding cavity, which contains a S125 residue, rather than the sterically bulky F125 of NAT1 (27). Following the completion of this study, a crystal structure of F125S NAT1 was deposited in the RCSB Protein Data Bank (entry 2ija). The crystal structure is highly consistent with the NMR-based model used in this investigation, especially with regard to the key residues proposed to interact with the arylamine substrates. Because of the predicted role of residue 125 in determining the relative abilities of wild-type NAT1 and NAT2 to N-acetylate the 2-alkylanilines, it will be of interest to evaluate the effect of the F125S mutation on the acetylation of these substrates. Koshland suggested that the kcat/Km value be considered a “performance constant” because it represents enzymatic performance with regard to binding a substrate and catalyzing its conversion to product(s) (50). The ratios of NAT1 to NAT2 performance constants with the environmental and carcinogenic arylamines investigated in this study vary by almost 1000-fold, indicating that subtle structural changes in the substrates have a marked influence on the ability of the NAT isoforms to catalyze their N-acetylation. Because of the differential tissue distribution of NATs, the substrate selectivity of NAT1 and NAT2 will influence the metabolic disposition of environmental arylamines with regard to either detoxification by N-acetylation or bioactivation by O-acetylation of N-hydroxyarylamines. Although it has been observed that expression of NAT1 and NAT2 mRNA occurs in numerous tissues, NAT2 activity is usually either very low or not detectable in tissues other than liver or the gastrointestinal tract; NAT1 activity, in contrast, is readily measurable in most tissues (51–64). Several studies indicate that NAT2 deficiency in humans increases the risk of arylamine-induced bladder cancer and is associated with enhanced levels of 4-ABP–hemoglobin adducts in tobacco smokers (65–67). It is not, however, to be expected that either the metabolic disposition or the toxicological and carcinogenic risk associated with exposure to arylamines will necessarily be dependent only on acetylation capacity. For example, Sugimori et al. found that clearance of 2-AF by NAT1/2 knockout mice was compromised by 4-fold, whereas the clearance of 4-ABP was unaffected (68). It has been suggested that arylamines containing a single aromatic ring, such as the alkylanilines, and those having two aromatic rings (e.g., 4-ABP) are predominantly metabolized by different pathways and that interindividual differences in those pathways account for part of the interindividual differences in the formation of

Substrate Specificity of NATs

arylamine–hemoglobin adducts (69). The results of this study serve to emphasize that, in addition to interindividual variations in the levels of activity of various xenobiotic metabolizing enzymes, and in addition to major structural differences such as the number of aromatic rings in the substrates, relatively modest structural differences in substrate structures can have a very substantial impact on their abilities to undergo biotransformation by such enzymes. The importance of building a comprehensive understanding of the structural characteristics of environmental arylamines and the biotransformation enzymes responsible for their activation and detoxification is exemplified by the recent work of Skipper et al., who demonstrated that administration of small doses of 2,6-DMA, 3,5-DMA, and 3-EA to mice resulted in DNA adduct formation in bladder, colon, liver, kidney, and lung (70). The results of this investigation provide insight into the molecular and structural features of the binding cavities of human NAT1 and human NAT2 that determine the propensity of the NAT isoforms to detoxify genotoxic environmental alkylanilines by N-acetylation and also provide progress toward development of an understanding of the molecular details of arylamine metabolism.

Chem. Res. Toxicol., Vol. 20, No. 9, 2007 1307

(12)

(13) (14)

(15) (16)

(17)

(18)

Acknowledgment. We thank Professor David Hein for the gift of NAT2*4 cDNA. The assistance and advice of Dr. Haiqing Wang and Dr. Tsui-fen Chou are gratefully acknowledged. We also thank Professor Rory Remmel for helpful discussions. The mass spectrometry data were obtained with the assistance of Dr. Sudha Marimanikkuppam of the Mass Spectrometry Consortium for the Life Sciences, University of Minnesota. This research was supported in part by an Academic Health Center Faculty Research Development Grant, by a Developmental Grant for Drug Design and Discovery from the Department of Medicinal Chemistry, and by a Melendy Summer Research Scholarship (A.M.V.V.).

References (1) Josephy, P. D., and Mannervik, B. (2006) Molecular Toxicology, 2nd ed., pp 426–447, Oxford University Press, New York. (2) Hanna, P. E. (1994) N-Acetyltransferases, O-acetyltransferases and N,O-acetyltransferases: Enzymology and bioactivation. AdV. Pharmacol. 27, 401–430. (3) Hanna, P. E. (1996) Metabolic activation and detoxification of arylamines. Curr. Med. Chem. 3, 195–210. (4) Kim, D., and Guengerich, F. P. (2005) Cytochrome P450 activation of arylamines and heterocyclic amines. Annu. ReV. Pharmacol. Toxicol. 45, 27–49. (5) Flammang, T. J., and Kadlubar, F. F. (1986) Acetyl coenzyme A-dependent metabolic activation of N-hydroxy-3,2-dimethyl-4-aminobiphenyl and several carcinogenic N-hydroxy arylamines in relation to tissue and species differences, other acyl donors, and arylhydroxamic acid-dependent acyltransferases. Carcinogenesis 7, 919–926. (6) Bartsch, H., Dworkin, M., Miller, J. A., and Miller, E. C. (1972) Electrophilic N-acetoxyaminoarenes derived from carcinogenic Nhydroxy-N-acetylaminofluorenes by enzymatic deacetylation and transacetylation in liver. Biochim. Biophys. Acta 286, 272–298. (7) Glowinski, I. B., Weber, W. W., Fysh, J. M., Vaught, J. B., and King, C. M. (1980) Evidence that arylhydroxamic acid N,O-acyltransferase and the genetically polymorphic N-acetyltransferase are properties of the same enzyme in rabbit liver. J. Biol. Chem. 255, 7883–7890. (8) Beland, F. A., and Kadlubar, F. F. (1990) Metabolic activation and DNA adducts of aromatic amines and nitroaromatic hydrocarbons. Handb. Exp. Pharmacol. 94/I, 267–325. (9) Grant, D. M., Blum, M., and Meyer, U. A. (1991) Monomorphic and polymorphic human arylamine N-acetyltransferases: A comparison of liver isozymes and expressed products of two cloned genes. Mol. Pharmacol. 39, 184–191. (10) Minchin, R. F. (1995) Acetylation of p-aminobenzoylglutamate, a folic acid catabolite, by recombinant human arylamine N-acetyltransferase and U937 cells. Biochem. J. 307, 1–3. (11) Hein, D. W., Doll, M. A., Rustan, T. D., Gray, K., Feng, Y., Ferguson, R. J., and Grant, D. M. (1993) Metabolic activation and deactivation

(19)

(20)

(21)

(22)

(23) (24)

(25)

(26)

(27)

(28)

(29) (30) (31)

of arylamine carcinogens by recombinant human NAT1 and polymorphic NAT2 acetyltransferases. Carcinogenesis 14, 1633–1638. Andres, H. H., Vogel, R. S., Tarr, G. E., Johnson, L., and Weber, W. W. (1987) Purification, physicochemical, and kinetic properties of liver acetyl-CoA: Arylamine N-acetyltransferase from rapid acetylator rabbits. Mol. Pharmacol. 31, 446–456. Grant, D. M., Lottspeich, F., and Meyer, U. A. (1989) Evidence for two closely related isozymes of arylamine N-acetyltransferase in human liver. FEBS Lett. 244, 203–207. Ozawa, S., Abu-Zeid, M., Kawakubo, Y., Toyama, S., Yamazoe, Y., and Kato, R. (1990) Monomorphic and polymorphic isozymes of arylamine N-acetyltransferases in hamster liver: Purification of the isozymes and genetic basis of N-acetylation polymorphism. Carcinogenesis 11, 2137–2144. Cheon, H. G., Boteju, L. W., and Hanna, P. E. (1992) Affinity alkylation of hamster hepatic arylamine N-acetyltransferases: Isolation of a modified cysteine residue. Mol. Pharmacol. 42, 82–83. Trinidad, A., Hein, D. W., Rustan, T. D., Ferguson, R. J., Miller, L. S., Bucher, K. D., Kirlin, W. G., Ogolla, F., and Andrews, A. F. (1990) Purification of hepatic polymorphic arylamine N-acetyltransferase from homozygous rapid acetylator inbred hamster: Identity with polymorphic N-hydroxyarylamine-O-acetyltransferase. Cancer Res. 50, 7942– 7949. Mattano, S. S., Land, S., King, C. M., and Weber, W. W. (1989) Purification and biochemical characterization of hepatic arylamine N-acetyltransferase from rapid and slow acetylator mice: Identity with arylhydroxamic acid N,O-acyltransferase and N-hydroxyarylamine O-acetyltransferase. Mol. Pharmacol. 35, 599–609. Ward, A., Sumers, M. J., and Sim, E. (1995) Purification of recombinant human N-acetyltransferase type 1 (NAT1) expressed in E. coli and characterization of its potential role in folate metabolism. Biochem. Pharmacol. 49, 1759–1767. Sticha, K. R. K., Sieg, C. A., Bergstrom, C. P., Hanna, P. E., and Wagner, C. R. (1997) Overexpression and large scale purification of recombinant hamster polymorphic arylamine N-acetyltransferase as a dihydrofolate reductase fusion protein. Protein Expression Purif. 10, 141–153. Sticha, K. R. K., Bergstrom, C. P., Wagner, C. R., and Hanna, P. E. (1998) Characterization of hamster recombinant monomorphic and polymorphic arylamine N-acetyltransferases. Bioactivation and mechanism based inactivation studies with N-hydroxy-2-acetylaminofluorene. Biochem. Pharmacol. 56, 47–59. Wang, H., Vath, G. M., Kawamura, A., Bates, C. A., Sim, E., Hanna, P. E., and Wagner, C. R. (2005) Overexpression, purification, and characterization of recombinant human arylamine N-acetyltransferase 1. Protein J. 24, 65–77. Wang, H., Vath, G. M., Gleason, K. J., Hanna, P. E., and Wagner, C. R. (2004) Probing the mechanism of hamster arylamine Nacetyltransferase 2 acetylation by active site modification, site-directed mutagenesis, and pre-steady state and steady-state kinetic studies. Biochemistry 43, 8234–8246. Wang, H., Liu, L., Hanna, P. E., and Wagner, C. R. (2005) Catalytic mechanism of hamster arylamine N-acetyltransferase 2. Biochemistry 44, 11295–11306. Guo, Z., Wagner, C. R., and Hanna, P. E. (2004) Mass spectrometric investigation of the mechanism of inactivation of hamster arylamine N-acetyltransferase 1 by N-hydroxy-2-acetylaminofluorene. Chem. Res. Toxicol. 17, 275–286. Wang, H., Wagner, C. R., and Hanna, P. E. (2005) Irreversible inactivation of arylamine N-acetyltransferases in the presence of N-hydroxy-4-acetylaminobiphenyl: A comparison of human and hamster enzymes. Chem. Res. Toxicol. 18, 183–197. Liu, F., Zhang, N., Zhou, X., Hanna, P. E., Wagner, C. R., Koepp, D. M., and Walters, K. J. (2006) Arylamine N-acetyltransferase aggregation and constitutive ubiquitylation. J. Mol. Biol. 361, 482– 492. Zhang, N., Liu, L., Liu, F., Wagner, C. R., Hanna, P. E., and Walters, K. J. (2006) NMR-based model reveals the structural determinants of mammalian arylamine N-acetyltransferase substrate specificity. J. Mol. Biol. 363, 188–200. Kawamura, A., Graham, J., Mushtag, A., Tsiftsoglou, S. A., Vath, G. M., Hanna, P. E., Wagner, C. R., and Sim, E. (2005) Eukaryotic arylamine N-acetyltransferase. Investigation of substrate specificity by high-throughput screening. Biochem. Pharmacol. 69, 347–359. NIH Guidelines for the Laboratory Use of Chemical Carcinogens (1981) NIH Publication 81-2385, U.S. Government Printing Office, Washington, DC. Bradford, M. M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 75, 248–254. Wallace, A. C., Laskowski, R. A., and Thornton, J. M. (1995) LIGPLOT: A program to generate schematic diagrams of proteinligand interactions. Protein Eng. 8, 127–134.

1308 Chem. Res. Toxicol., Vol. 20, No. 9, 2007 (32) Gan, J., Skipper, P. L., Gago-Dominquez, M., Arakawa, K., Ross, R. K., Yu, M. C., and Tannenbaum, S. R. (2004) Alkylanilinehemoglobin adducts and risk of non-smoking related bladder cancer. J. Natl. Cancer Inst. 96, 1425–1431. (33) Palmiotto, G., Pieraccini, G., Moneti, G., and Dolara, P. (2001) Determination of the levels of aromatic amines in indoor and outdoor air in Italy. Chemosphere 43, 355–361. (34) Markowitz, S. B., and Levin, K. (2004) Continued epidemic of bladder cancer in workers exposed to ortho-toluidine in a chemical factory. J. Occup. EnViron. Med. 46, 154–160. (35) DeBruin, L. S., Pawliszyn, J. B., and Josephy, P. D. (1999) Detection of monocyclic aromatic amines, possible mammary carcinogens, in human milk. Chem. Res. Toxicol. 12, 78–82. (36) Sinclair, J. C., Delgoda, R., Noble, M. E. M., Jarmin, S., Goh, N. K., and Sim, E. (1998) Purification, characterization, and crystallization of an N-hydroxyarylamine O-acetyltransferase from Salmonella typhimurium. Protein Expression Purif. 12, 371–380. (37) Sandy, J., Mushtaq, A., Kawamara, A., Sinclair, J. C., Sim, E., and Noble, M. (2002) The structure of arylamine N-acetyltransferase from Mycobacterium smegamatis: An enzyme which inactivates the antitubercular drug, isoniazid. J. Mol. Biol. 318, 1071–1083. (38) Westwood, I. M., Holton, S. J., Rodrigues-Lima, F., Dupret, J. M., Bhakta, S., Noble, M. E. M., and Sim, E. (2005) Expression, purification, characterization and structure of Pseudomonas aeruginosa arylamine N-acetyltransferase. Biochem. J. 385, 605–612. (39) Talaska, G. (2003) Aromatic amines and human urinary bladder cancer: Exposure sources and epidemiology. J. EnViron. Sci. Health C21, 29– 43. (40) Talaska, G., and Al-Zoughool, M. (2003) Aromatic amines and biomarkers of human exposure. J. EnViron. Sci. Health C21, 133– 164. (41) Patrianakos, C., and Hoffman, D. (1979) Chemical studies on tobacco smoke LXIV. On the analysis of aromatic amines in tobacco smoke. J. Anal. Toxicol. 3, 150–154. (42) Bryant, M. S., Vineis, P., Skipper, P. L., and Tannenbaum, S. R. (1988) Hemoglobin adducts of aromatic amines: Associations with smoking status and type of tobacco. Proc. Natl. Acad. Sci. U.S.A. 85, 9788– 9791. (43) Maclure, M., Katz, R. B.-A., Bryant, M. S., Skipper, P. L., and Tannenbaum, S. R. (1989) Elevated blood levels of carcinogens in passive smokers. Am. J. Public Health 79, 1383–1384. (44) Smith, C. J., Dooly, G. L., and Moldoveanu, S. C. (2003) New technique using solid-phase extraction for the analysis of aromatic amines in mainstream cigarette smoke. J. Chromatogr., A 991, 99– 107. (45) Turesky, R. J., Freeman, J. P., Holland, R. D., Nestorick, D. M., Miller, D. W., Ratnasinghe, D. L., and Kadlubar, F. F. (2003) Identification of aminobiphenyl derivatives in commercial hair dyes. Chem. Res. Toxicol. 16, 1162–1173. (46) Bryant, M. S., Simmons, H. F., Harrell, R. E., and Hinson, J. A. (1994) 2,6-Dimethylaniline-hemoglobin adducts from lidocaine in humans. Carcinogenesis 15, 2287–2290. (47) Gaber, K., Harreus, U. A., Matthias, C., Kleinsasser, N. H., and Richter, E. (2007) Hemoglobin adducts of the human bladder carcinogen o-toluidine after treatment with the local anesthetic prilocaine. Toxicology 229, 157–164. (48) Goodfellow, G. H., Dupret, J. M., and Grant, D. M. (2000) Identification of amino acids imparting acceptor substrate selectivity to human arylamine acetyltransferases NAT1 and NAT2. Biochem. J. 348, 159– 166. (49) Savulescu, M. R., Mushtaq, A., and Josephy, P. D. (2005) Screening and characterizing human NAT2 variants. Methods Enzymol. 400, 192– 215. (50) Koshland, D. E., Jr. (2002) The application and usefulness of the ratio kcat/Km. Bioorg. Chem. 30, 211–213. (51) Windmill, K. F., Gaedigk, A., de la M. Hall, P., Samaratunga, H., Grant, D. M., and McManus, M. E. (2000) Localization of Nacetyltransferases NAT1 and NAT2 in human tissues. Toxicol. Sci. 54, 19–29. (52) Stanley, L. A., Coroneos, E., Cuff, R., Hickman, D., Ward, A., and Sim, E. (1996) Immunochemical detection of arylamine N-acetyl-

Liu et al.

(53)

(54) (55) (56) (57) (58) (59)

(60) (61)

(62)

(63)

(64) (65) (66)

(67)

(68) (69)

(70)

transferase in normal and neoplastic bladder. J. Histochem. Cytochem. 44, 1059–1067. Rodrigues-Lima, F., Cooper, R. N., Goudeau, B., Atmane, N., Chamagne, A.-M., Butler-Browne, G., Sim, E., Vicart, P., and Dupret, J.-M. (2003) Skeletal muscles express the xenobiotic-metabolizing enzyme arylamine N-acetyltransferase. J. Histochem. Cytochem. 51, 784–796. Cribb, A. E., Grant, D. M., Miller, M. A., and Spielberg, S. P. (1991) Expression of monomorphic arylamine N-acetyltransferase (NAT1) in human leukocytes. J. Pharmacol. Exp. Ther. 259, 1241–1246. Lawson, T., and Kolar, C. (2002) Human prostate epithelial cells metabolize chemicals of dietary origin to mutagens. Cancer Lett. 175, 141–146. Smelt, V. A., Mardon, H. J., Redman, C. W. G., and Sim, E. (1997) Acetylation of arylamines by the placenta. Eur. J. Drug Metab. Pharmacokinet. 22, 403–408. Derewlany, L. O., Knie, B., and Koren, G. (1994) Arylamine N-acetyltransferase activity of the human placenta. J. Pharmacol. Exp. Ther. 269, 756–760. Kawakubo, Y., Merk, H. F., Al Masoudi, T., Sieben, S., and Blomeke, B. (2000) N-Acetylation of paraphenylenediamine in human skin and keratinocytes. J. Pharmacol. Exp. Ther. 292, 150–155. Dairou, J., Malecaze, F., Dupret, J. M., and Rodrigues-Lima, F. (2005) The xenobiotic-metabolizing enzymes arylamine N-acetyltransferases in human lens epithelial cells: Inactivation by cellular oxidants and UVB-induced oxidative stress. Mol. Pharmacol. 67, 1299–1306. Geylan, Y. S., Dizbay, S., and Guray, T. (2001) Arylamine Nacetyltransferase activities in human breast cancer tissues. Neoplasma 48, 108–111. Sadrieh, N., Davis, C. D., and Snyderwine, E. G. (1996) NAcetyltransferase expression and metabolic activation of the foodderived heterocyclic amines in the human mammary gland. Cancer Res. 56, 2683–2687. Williams, J. A., Stone, E. M., Fakis, G., Johnson, N., Cordell, J. A., Meinl, W., Glatt, H., Sim, E., and Philips, D. H. (2001) NAcetyltransferases, sulfotransferases and heterocyclic amine activation in the breast. Pharmacogenetics 11, 373–388. Rodriguez, J. W., Kirlin, W. G., Ferguson, R. J., Doll, M. A., Gray, K., Rustan, T. D., Lee, M. E., Kemp, K., Urso, P., and Hein, D. W. (1993) Human acetylator genotype: Relationship to colorectal cancer incidence and arylamine N-acetyltransferase expression in colon cytosol. Arch. Toxicol. 67, 445–452. Hickman, D., Pope, J., Patil, S. D., Fakis, G., Smelt, V., Stanley, L. A., Payton, M., Unadkat, J. D., and Sim, E. (1998) Expression of arylamine N-acetyltransferase in human intestine. Gut 42, 402–409. Hein, D. W. (2006) N-Acetyltransferase 2 genetic polymorphism: Effects of carcinogen and haplotype on urinary bladder cancer risk. Oncogene 25, 1649–1658. Vineis, P., Caporaso, N., Tannenbaum, S. R., Skipper, P. L., Glogowski, J., Bartsch, H., Coda, M., Talaska, G., and Kadlubar, F. (1990) Acetylation phenotype, carcinogen-hemoglobin adducts, and cigarette smoking. Cancer Res. 50, 3002–3004. Probst-Hensch, N. M., Bell, D. A., Watson, M. A., Skipper, P. L., Tannenbaum, S. R., Chan, K. K., Ross, R. K., and Yu, M. C. (2000) N-Acetyltransferase 2 phenotype but not NAT1*10 genotype affects aminobiphenyl-hemoglobin adduct levels. Cancer Epidemiol., Biomarkers PreV. 9, 619–623. Sugamori, K. S., Brenneman, D., and Grant, D. M. (2006) In vivo and in vitro metabolism of arylamine procarcinogens in acetyltransferase-deficient mice. Drug Metab. Dispos. 34, 1697–1702. Ronco, G., Vineis, P., Bryant, M. S., Skipper, P. L., and Tannenbaum, S. R. (1990) Hemoglobin adducts formed by aromatic amines in smokers: Sources of inter-individual variability. Br. J. Cancer 61, 534– 537. Skipper, P. L., Trudel, L. J., Kensler, T. W., Groopman, J. D., Egner, P. A., Liberman, R. G., Wogan, G. N., and Tanncnbaum, S. R. (2006) DNA adduct formation by 2,6-dimethyl-, 3,5-dimethyl-, and 3-ethylaniline in vivo in mice. Chem. Res. Toxicol. 19, 1086–1090.

TX7001614