2-Amino-3,8-dimethylimidazo-[4,5-f]quinoxaline-induced DNA adduct formation and mutagenesis in DNA repair-deficient Chinese hamster ovary cells expressing human cytochrome P4501A1 and rapid or slow acetylator N-acetyltransferase 2

Abstract

2-Amino-3,8-dimethylimidazo-[4,5-f]quinoxaline (MeIQx) is one of the most potent and abundant mutagens in the western diet. Bioactivation includes N-hydroxylation catalyzed by cytochrome P450s followed by O-acetylation catalyzed by N-acetyltransferase 2 (NAT2). In humans, NAT2*4 allele is associated with rapid acetylator phenotype, whereas NAT2*5B allele is associated with slow acetylator phenotype. We hypothesized that rapid acetylator phenotype predisposes humans to DNA damage and mutagenesis from MeIQx. Nucleotide excision repair-deficient Chinese hamster ovary cells were constructed by stable transfection of human cytochrome P4501A1 (CYP1A1) and a single copy of either NAT2*4 (rapid acetylator) or NAT2*5B (slow acetylator) alleles. CYP1A1 and NAT2 catalytic activities were undetectable in untransfected Chinese hamster ovary cell lines. CYP1A1 activity did not differ significantly (P > 0.05) among the CYP1A1-transfected cell lines. Cells transfected with NAT2*4 had 20-fold significantly higher levels of sulfamethazine N-acetyltransferase (P = 0.0001) and 6-fold higher levels of N-hydroxy-MeIQx O-acetyltransferase (P = 0.0093) catalytic activity than cells transfected with NAT2*5B. Only cells transfected with both CYP1A1 and NAT2*4 showed concentration-dependent cytotoxicity and hypoxanthine phosphoribosyl transferase mutagenesis following MeIQx treatment. Deoxyguanosine-C8-MeIQx was the primary DNA adduct formed and levels were dose dependent in each cell line and in the following order: untransfected < transfected with CYP1A1 < transfected with CYP1A1 and NAT2*5B < transfected with CYP1A1 and NAT2*4. MeIQx DNA adduct levels were significantly higher (P < 0.001) in CYP1A1/NAT2*4 than CYP1A1/NAT2*5B cells at all concentrations of MeIQx tested. MeIQx-induced DNA adduct levels correlated very highly (r2 = 0.88) with MeIQx-induced mutants. These results strongly support extrahepatic activation of MeIQx by CYP1A1 and a robust effect of human NAT2 genetic polymorphism on MeIQx-induced DNA adducts and mutagenesis. The results provide laboratory-based support for epidemiologic studies reporting higher frequency of heterocyclic amine-related cancers in rapid NAT2 acetylators.

Figure 1

EROD activities in UV5 CHO cell lines. Each bar represents Mean ± S.E.M. for three experiments. EROD activity in UV5 cells was below the level of detection (< 0.2 pmoles/min/million cells). No significant (p>0.05) differences in EROD activity among the CYP1A1-transfected CHO cells were observed.

Figure 2

SMZ N-acetyltransferase (top panel) and N-hydroxy-MeIQx O-acetyltransferase (bottom) activities in cell lysates of CYP1A1/NAT2*4- and CYP1A1/NAT2*5B-transfected CHO cells. Activity in the UV5 and UV5/CYP1A1 cell lines were below the level of detection for SMZ NAT (< 20 pmol/min/mg). Low but detectable levels of N-hydroxy-MeIQx activation detected in the UV5 and the UV5/CYP1A1 cell lines were subtracted from the experimental measurements in the NAT2-transfected cells. *Both SMZ NAT and N-hydroxy-MeIQx OAT activities were significantly (p < 0.01) higher in CYP1A1/NAT2*4- than CYP1A1/NAT2*5B-transfected CHO cells.

Figure 3

MeIQx-induced cytotoxicity in UV5 CHO cell lines. Percent survival on the ordinate is plotted versus MeIQx treatment concentration on the abscissa. Each data point represents Mean ± S.E.M. for three experiments. MeIQx-induced cytotoxicity was significantly greater (p<0.05) in CYP1A1/NAT2*4- than CYP1A1/NAT2*5B-transfected CHO cells at 3 μM.

Figure 4

MeIQx-induced hprt mutants in UV5 CHO cell lines. MeIQx-induced hprt mutants are plotted on the ordinate versus MeIQx treatment concentration on the abscissa. Each data point represents Mean ± S.E.M. for three experiments. *MeIQx-induced hprt mutants were concentration dependent only in the CYP1A1/NAT2*4- transfected CHO cells.

Figure 5

Structure and electrospray ionization spectra of dG-C8-MeIQx (top two panels) and dG-C8-MeIQx-D3 (bottom two panels) showing [M+H]⁺ = 479 and 482, respectively. Collision induced dissociation fragmentation of dG-C8-MeIQx and d3 dG-C8-MeIQx at −50 V collision energy shows fragmentation of the MeIQx and guanosine moieties dominated by the major fragment produced from loss of deoxyribose: the aglycone ion of m/z 363 or 366 for dG-C8-MeIQx and dG-C8-MeIQx-D3, respectively. The collision energy was −25 V for MRM transitions used during quantitation.

Figure 6

Chromatograms of representative HPLC-tandem mass spectrometry traces of dG-C8-MeIQx (top panel) and dG-C8-MeIQx-D3 (bottom panel) of DNA from CHO cells treated with MeIQx and spiked with dG-C8-MeIQx-D3. Elution time in minutes is plotted on the abscissae. Multiple reaction monitoring (MRM) was used to measure the [M+H]⁺ to [(M-116) + H]⁺ (loss of deoxyribose) mass transition. The dG-C8-MeIQx adduct was monitored using the transition from m/z 479 to m/z 363 (top) and the deuterated internal standard (dG-C8-MeIQx-D3) was monitored using the transition from m/z 482 to m/z 366 (bottom).

Figure 7

MeIQx-induced dG-C8-MeIQx adduct levels in UV5 CHO cell lines. Adduct levels are plotted on the ordinate versus MeIQx treatment concentration on the abscissa. Each data point represents Mean ± S.E.M. for three experiments (the S.E.M. sometimes falls within the symbol). *dG-C8-MeIQx adducts were significantly higher (p<0.001) in CYP1A1/NAT2*4- than CYP1A1/NAT2*5B-transfected CHO cells at all concentrations tested.