Trials and Tribulations in the First Three Years in Operation of the SSAMS for Biomedical 14C-AMS at LLNL.

We report on the first several years of operation of our recently installed 250 kV SSAMS at LLNL, purchased to replace our 1-MV AMS system for the measurement of 14C from labeled biochemical samples. We have modified the ion source region to improve ion output. Additionally, the SSAMS required significant software modifications to the data acquisition system in order to accurately measure 14C at the high-count rates typically encountered with labeled biochemical samples. We found that the data can be corrected assuming a nonparalyzable dead time response with a single event dead time of 6 µs. Since operation began, we have measured over 13,000 graphitic unknowns and over 1900 standards with an overall precision of 1.0%. We have optimized our system for the analysis of CO2 gas samples. We compared aliquots of identical samples measured as solid graphite and as liquid drops. Excellent agreement was found between the two, although the average precision of the graphite targets was an order of magnitude better than the liquid drop analysis due to the much larger number of 14C atoms available for measurement.

Mass Spectrometric Characterization of an Acid-Labile Adduct Formed with 2-Amino-1-methyl-6-phenylimidazo[4,5-b]pyridine and Albumin in Humans.

2-Amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) is a carcinogenic heterocyclic aromatic amine formed during the high-temperature cooking of meats. The cytochrome P450-mediated N-hydroxylation of the exocyclic amine group of PhIP produces 2-hydroxyamino-1-methyl-6-phenylimidazo[4,5-b]pyridine, an electrophilic metabolite that forms adducts with DNA and proteins. Previous studies conducted by our laboratory showed that the reaction of N-oxidized PhIP metabolites with human albumin in vitro primarily occurs at the Cys34 residue, to produce an acid-labile linked sulfinamide adduct. On the basis of these findings, we developed a sensitive ultraperformance liquid chromatography-mass spectrometry method to measure acid-labile albumin-PhIP adducts in human volunteers administered a dietary-relevant dose of 14C-labeled PhIP [Dingley, K. H., et al. (1999) Cancer Epidemiol., Biomarkers Prev. 8, 507-512]. Mild acid treatment of albumin (0.1 N HCl, 37 °C for 1 h) or proteolytic digestion with Pronase [50 mM ammonium bicarbonate buffer (pH 8.5) at 37 °C for 18 h] released similar amounts of covalently bound PhIP, which was characterized by multistage scanning and quantified by Orbitrap mass spectrometry. The amount of [14C]PhIP recovered by acid treatment of albumin 24 h following dosing accounted for 7.2-21.3% of the [14C]PhIP bound to albumin based on accelerator mass spectrometry measurements. 2-Amino-1-methyl-6-(5-hydroxy)phenylimidazo[4,5-b]pyridine, a hydrolysis product of the Cys34 S-N linked sulfenamide adduct of PhIP, was not detected in either acid-treated or protease-treated samples. These findings suggest that a portion of the PhIP bound to albumin in vivo probably occurs as an acid-labile sulfinamide adduct formed at the Cys34 residue.

Operation of the "Small" BioAMS Spectrometers at CAMS: Past and Future Prospects.

A summary of results from the solid samples run on our compact 1 MV AMS system over its 13.5 years of operation is presented. On average 7065 samples per year were measured with that average dropping to 3278 samples per year following the deployment of our liquid sample capability. Although the dynamic range of our spectrometer is 4.5 orders in magnitude, most of the measured graphitic samples had 14C/C concentrations between 0.1 and 1 modern. The measurements of our ANU sucrose standard followed a Gaussian distribution with an average of 1.5082 ± 0.0134 modern. The LLNL biomedical AMS program supported many different types of experiments, however, the large majority of samples measured were derived from animal model systems. We have transitioned all of our biomedical AMS measurements to the recently installed 250 kV SSAMS instrument with good agreement compared in measured 14C/C isotopic ratios between sample splits. Finally, we present results from replacement of argon stripping gas with helium in the SSAMS with a 22% improvement in ion transmission through the accelerator and high-energy analyzing magnet.

An Interface for the Direct Coupling of Small Liquid Samples to AMS.

We describe the moving wire interface attached to the 1-MV AMS system at LLNL's Center for Accelerator Mass Spectrometry for the analysis of nonvolatile liquid samples as either discrete drops or from the direct output of biochemical separatory instrumentation, such as high-performance liquid chromatography. Discrete samples containing at least a few 10s of nanograms of carbon and as little as 50 zmol 14C can be measured with a 3-5% precision in a few minutes. The dynamic range of our system spans approximately 3 orders in magnitude. Sample to sample memory is minimized by the use of fresh targets for each discrete sample or by minimizing the amount of carbon present in a peak generated by an HPLC containing a significant amount of 14C. Liquid Sample AMS provides a new technology to expand our biomedical AMS program by enabling the capability to measure low-level biochemicals in extremely small samples that would otherwise be inaccessible.

Accelerator mass spectrometry allows for cellular quantification of doxorubicin at femtomolar concentrations.

Accelerator mass spectrometry (AMS) is a highly sensitive analytical methodology used to quantify the content of radioisotopes, such as (14)C, in a sample. The primary goals of this work were to demonstrate the utility of AMS in determining total cellular [(14)C]anthracycline concentrations following administration of doxorubicin (DOX) and to develop a sensitive assay that is superior to high performance liquid chromatography (HPLC) for the quantification of [(14)C]anthracycline at the tumor level. In order to validate the sensitivity of AMS versus HPLC with fluorescence detection, we performed three studies comparing the cellular accumulation of DOX: one in vitro cell line study, and two in vivo xenograft mouse studies. Using AMS, we quantified cellular [(14)C]anthracycline content up to 4 h following in vitro exposure at concentrations ranging from 0.2 pg/ml (345 fM) to 2 microg/ml (3.45 microM) [(14)C]DOX. The results of this study show that, compared to standard fluorescence-based HPLC, the AMS method was over five orders of magnitude more sensitive. Two in vivo studies compared the sensitivity of AMS to HPLC using a nude mouse xenograft model in which breast cancer cells were implanted subcutaneously. After sufficiently large tumors formed, [(14)C]DOX was administered intravenously at two dose levels. Additionally, we tested the AMS method in a nude mouse xenograft model of multidrug resistance (MDR) in which each mouse was implanted with both wild type and MDR+ cells on opposite flanks. The results of the second and third studies showed that [(14)C]anthracycline concentrations were significantly higher in the wild type tumors compared to the MDR+ tumors, consistent with the MDR model. Although this method does not discriminate between parent drug and metabolites, the extreme sensitivity of AMS should facilitate similar studies in humans to establish target site drug delivery and to potentially determine the optimal treatment dose and regimen.

Synthesis and spectroscopic characterization of site-specific 2-amino-1-methyl-6-phenylimidazo.

The aim of the present study is to determine the chemical structure and conformation of DNA adducts formed by incubation of the bioactive form of 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP), N-acetoxy-PhIP, with a single-stranded 11mer oligodeoxyribonucleotide. Using conditions optimized to give the C8-dG-PhIP adduct as the major product, sufficient material was synthesized for NMR solution structure determination. The NMR data indicate that in duplex DNA this adduct exists in equilibrium between two different conformational states. In the main conformer, the covalently bound PhIP molecule intercalates in the helix, whilst in the minor conformation the PhIP ligand is probably solvent exposed. In addition to the C8-dG-PhIP adduct, at least eight polar adducts are found after reaction of N-acetoxy-PhIP with the oligonucleotide. Three of these were purified for further characterization and shown to exhibit lowest energy UV absorption bands in the range 342-347 nm, confirming the presence of PhIP or PhIP derivative. Accurate mass determination of two of the polar adducts by negative ion MALDI-TOF MS revealed ions consistent with a spirobisguanidino-PhIP derivative and a ring-opened adduct. The third adduct, which has the same mass as the C8-dG-PhIP oligonucleotide adduct, may contain PhIP bound to the N2 position of guanine.

Fate and distribution of 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine in mice at a human dietary equivalent dose.

2-Amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) is a heterocyclic amine rodent carcinogen that is found at the ppb level in cooked meat. Most laboratory studies are at 10(4)-10(7)-fold greater concentrations than actual ingested human doses. We report the first study of the bioavailability and fate of this heterocyclic amine at a human dietary equivalent dose using the high sensitivity offered by accelerator mass spectrometry. [2-14C]PhIP was administered to C57BL/6 male mice (41 ng/kg) by gavage. Tissues and excreta were collected over the subsequent 96 h. One hundred % of the administered dose was excreted in urine (90%) and feces (10%) over the length of the study. Absorption of the radiocarbon-tagged PhIP from the gastrointestinal tract was rapid, with radiocarbon levels peaking in the whole blood and urine within 1 h of exposure. Fecal 14C levels peaked at 12 h. Tissue levels peaked by 3 h with the highest concentrations of radiolabel in the intestine, stomach, and liver, followed by the kidney, pancreas, lung, and spleen. Low levels of 14C from PhIP (0.01-0.04% of the administered dose) could be detected in the tissues 48-96 h after exposure, possibly due to covalent binding to protein or DNA. The calculated half-life of PhIP at this dose was 1.14 h. This study is the first example of how accelerator mass spectrometry can be used to gather biological information about carcinogenic compounds at environmental levels of exposure.

Chemical analysis, prevention, and low-level dosimetry of heterocyclic amines from cooked food.

Potent mutagenic and carcinogenic heterocyclic amines are produced from heated food derived from muscle. These compounds are present at part-per-billion levels and consist primarily of the amino-imidazoazaarene class of chemicals. Additional mutagens present in the meat are not as clearly characterized. Commercial fried-beef patties (hamburgers) have low levels of 2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline (MeIQx) and 2-amino-3,4,8-trimethylimidazo[4,5-f]quinoxaline (4,8-DiMeIQx), 0.1-0.68 ng/g meat for MeIQx and slightly lower for 4,8-DiMeIQx. The formation of these heterocyclic amines can be reduced by microwave pretreatment of meat, with the resulting liquid being poured off before frying. The Ames/Salmonella mutagenic activity was reduced to 5-10% of that of non-microwave-treated samples. MeIQx and DiMeIQx concentrations were reduced to 12% and 50% of levels in the non-microwave-treated samples, respectively. MeIQx adducts, as measured by accelerator mass spectrometry, were found to be linear with doses from 5 mg/kg to 500 ng/kg. Linear DNA binding at low doses is important for assuming linear risk estimation from the high animal-feeding doses causing cancer to the low human-dietary exposures. Extrapolating from the rodent TD50 dose to humans gives a maximum credible risk from consumption of heterocyclic amines of approximately 1/1000 exposed individuals.

DNA and protein adduct formation in the colon and blood of humans after exposure to a dietary-relevant dose of 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine.

Epidemiology studies have indicated that certain dietary components, including well-cooked meat, are risk determinants for colon cancer. Cooked meat can contain significant quantities of heterocyclic aromatic amines (HCAs), which have been established as carcinogens in laboratory animals. 2-Amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) is usually the most mass-abundant HCA, with concentrations up to 480 ppb. We used accelerator mass spectrometry to establish whether DNA and protein adducts can be detected in humans exposed to a quantity of PhIP comparable with levels of exposure that occur in the diet. Five human volunteers were administered a dietary-relevant dose of [14C]PhIP (70-84 microg) 48-72 h before surgery for removal of colon tumors. Blood samples were collected at various time points, and albumin, hemoglobin, and WBC DNA were extracted for analysis by accelerator mass spectrometry. Tissue samples were collected during surgery and used to assess either tissue available doses of [14C]PhIP or adduct levels. The results of this study show: (a) PhIP is activated to a form that will bind to albumin, hemoglobin, and WBC DNA in peripheral blood. WBC DNA adducts were unstable and declined substantially over 24 h; (b) PhIP is bioavailable to the colon, with levels in normal tissue in the range 42-122 pg PhIP/g tissue; and (c) PhIP binds to both protein and DNA in the colon. DNA adduct levels in the normal tissue were 35-135 adducts/10(12) nucleotides, which was significantly lower than tumor tissue. The results of this study demonstrate that PhIP is bioavailable to the human colon following defined dietary-relevant doses and forms DNA and protein adducts.

Bioanalytical applications of accelerator mass spectrometry for pharmaceutical research.

Accelerator mass spectrometry (AMS) is a mass spectrometric method for quantifying isotopes. It has had great impact in the geosciences and is now being applied in the biomedical fields. AMS measures radioisotopes such as 14C, 3H, 41Ca, and 36Cl, and others, with attomole sensitivity and high precision. Its use is allowing absorption, distribution, metabolism and elimination studies, as well as detailed pharmacokinetics, to be carried out directly in humans with very low chemical or radiological hazard. It is used in combination with standard separation methodologies, such as chromatography, in identification of metabolites and molecular targets for both toxicants and pharmacologic agents. AMS allows the use of very low specific activity chemicals (< 1 mCi/mmol), creating opportunities to use compounds not available in a high specific activity form, such as those that must be biosynthesized, produced in combinatorial libraries, or made through inefficient synthesis. AMS is allowing studies to be carried out with agents having low bioavailability, low systemic distributions, or high toxicity where administered doses must be kept low (<1 microg/kg). It may have uses in tests for idiosyncratic metabolism, drug interaction, or individual susceptibility, among others. The ability to use very low chemical doses, low radiological doses, small samples and conduct multiple dose studies may help move drug candidates into humans faster and safer than before. The uses of AMS are growing and its potential for drug development is only now beginning to be realized.

Macromolecular adduct formation and metabolism of heterocyclic amines in humans and rodents at low doses.

2-Amino-3,8-dimethylimidazo[4,5-f]quinoxaline (MeIQx) and 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) are heterocyclic amines formed during the cooking of meat and fish. Both are genotoxic in a number of test systems and are carcinogenic in rats and mice. Human exposure to these compounds via dietary sources has been estimated to be under 1 microg/kg body wt. per day, although most laboratory animal studies have been conducted at doses in excess of 10 mg/kg body wt. per day. We are using accelerator mass spectrometry (AMS), a tool for measuring isotopes with attomole sensitivity, to study the dosimetry of protein and DNA adduct formation by low doses of MeIQx and PhIP in rodents and comparing the adduct levels to those formed in humans. The results of these studies show: 1, protein and DNA adduct levels in rodents are dose-dependent; 2, adduct levels in human tissues and blood are generally greater than in rodents administered equivalent doses; and 3, metabolite profiles differ substantially between humans and rodents for both MeIQx and PhIP, with more N-hydroxylation (bioactivation) and less ring oxidation (detoxification) in humans. These data suggest that rodent models do not accurately represent the human response to heterocyclic amine exposure.

In vivo human metabolism of [2-14C]2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP).

To better understand the interactions of the pathways of activation and detoxification on the metabolism of the putative carcinogen, PhIP, we administered a dose of 70-84 microg [2-14C] PhIP (17.5 [microCi 14C) 48-72 h before scheduled colon surgery. Blood and urine collected for the next 48-72 h was evaluated by linear accelerator mass spectroscopy (AMS) and scintillation counting LC-MS to identify specific PhIP metabolites. The thermostable phenol sulfotransferase (SULT1A1) phenotype was correlated with the 4'-PhIP-SO4 levels in the urine at 0-4 h (R = 0.86, P = 0.059). The CYP1A2 activity had a negative correlation with PhIP serum levels at 1 h (R = 0.94, P = 0.06) and a positive correlation with urine N-OH-PhIP levels at 0-4 h (R = 0.85, P = 0.15). This low level radioisotope method of determining the influence of phenotype on metabolism will significantly improve our understanding of the interrelationships of these pathways and provide a critical foundation for the development of individual risk assessment.

Comparative biotransformation studies of MeIQx and PhIP in animal models and humans.

MeIQx and PhIP are putative carcinogenic heterocyclic amines formed during the cooking of meat and fish. Using accelerator mass spectrometry, we have investigated the metabolism and macromolecule binding of 14C-labelled MeIQx and PhIP in human cancer patients compared to the rat. Following oral administration of MeIQx and PhIP, more DNA adducts were formed in human colon tissue compared with rats. Differences were also observed between rats and humans in the metabolite profile and urine excretion for these compounds. These results suggest humans metabolise heterocyclic amines differently to laboratory rodents and question their use as models of human risk.

Low-level biological dosimetry of heterocyclic amine carcinogens isolated from cooked food.

The bioavailability and the bioreactivity of the carcinogenic heterocyclic amine [2-14C]2-amino-1-methyl-6-phenyl-imidazo[4,5-b]pyridine (PhIP) have been investigated at a dose approximating that likely from the human diet by accelerator mass spectrometry (AMS). [2-14C]PhIP was administered to mice at a dose equivalent ot the consumption of two 100 g beef patties (41 ng/kg). The biological half-life of PhIP was 1 hr, with 90% of the dose being excreted via the urine. Peak tissue PhIP concentrations were reached within 3 hr, with the highest levels in the tissues of the gastrointestinal tract, followed by the liver, kidney, pancreas, and thymus. Since the detection limit by AMS is dependent on the natural abundance of 14C, we have achieved further increases in sensitivity by producing mice that have 20% of the natural abundance of 14C. Use of these 14C-depleted animals allows measurements to be made near the natural level of exposure for many environmental carcinogens. PhIP-DNA adduct levels have also been measured by 32P-postlabeling at doses of 1.0, 10, and 20 mg/kg. The highest adduct levels were found in the pancreas, thymus, heart, and liver and increased linearly with dose. The principal adducts are derived from guanine.

Methods of DNA adduct determination and their application to testing compounds for genotoxicity.

At the International Workshop on Genotoxicity Test Procedures (IWGTP) held in Washington, DC (March 25-26, 1999), a working group considered the uses of DNA adduct determination methods for testing compounds for genotoxicity. When a drug or chemical displays an unusual or inconsistent combination of positive and negative results in in vitro and in vivo genotoxicity assays and/or in carcinogenicity experiments, investigations into whether or not DNA adducts are formed may be helpful in assessing whether or not the test compound is a genotoxin. DNA adduct determinations can be carried out using radiolabeled compounds and measuring radioactive decay (scintillation counting) or isotope ratios (accelerator mass spectrometry) in the isolated DNA. With unlabeled compounds adducts may be measured by (32)P-postlabeling analysis of the DNA, or by physicochemical methods including mass spectrometry, fluorescence spectroscopy, or electrochemical detection, or by immunochemical methods. Each of these approaches has different strengths and limitations, influenced by sensitivity, cost, time, and interpretation of results. The design of DNA binding studies needs to be on a case-by-case basis, depending on the compound's profile of activity. DNA purity becomes increasingly important the more sensitive, and less chemically specific, the assay. While there may be adduct levels at which there is no observable biological effect, there are at present insufficient data on which to set a threshold level for biological significance.

Dose-dependent binding of trichloroethylene to hepatic DNA and protein at low doses in mice.

Trichloroethylene (TCE) is a widely used industrial chemical and a low level contaminant of surface and ground water in industrialized areas. It is weakly mutagenic in several test systems and carcinogenic in rodents. However, the mechanism for its carcinogenicity is not known. We investigated the binding of [1,2-14C]TCE ([14C]TCE) to liver DNA and proteins in male B6C3F1 mice at doses more relevant to humans than used previously. The time course for the binding was studied in animals dosed with 4.1 micrograms [14C]TCE/kg body weight (b.w.) and sacrificed between 0.5 and 120 h after i.p. injection. A dose response study was carried out in mice given [14C]TCE at doses between 2 micrograms/kg and 200 mg/kg b.w. and sacrificed 2 h post-treatment. [14C]TCE associated with the DNA and protein extracts was measured using accelerator mass spectrometry. The highest level of protein binding (2.4 ng/g protein) was observed 1 h after the treatment followed by a rapid decline, indicating pronounced instability of the adducts and/or rapid turnover of liver proteins. DNA binding was biphasic with the first peak (75 pg/g DNA) at 4 h. However, the highest binding (120 pg/g DNA) was found between 24 and 72 h after the treatment. Dose response curves were linear for both protein and DNA binding. The binding of TCE metabolites to DNA was ca. 100-fold lower than to proteins when calculated per unit weight of macromolecules and when measured 2 h post-exposure. This study shows that TCE metabolites bind to DNA and proteins in a dose-dependent manner in liver, one of the target organs for its tumorigenicity. Thus, protein and DNA adduct formation should be considered as a factor in the tumorigenesis of TCE.

Quantitation of benzo[a]pyrene-DNA adducts by postlabeling with 14C-acetic anhydride and accelerator mass spectrometry.

Quantitation of carcinogen-DNA adducts provides an estimate of the biologically effective dose of a chemical carcinogen reaching the target tissue. In order to improve exposure-assessment and cancer risk estimates, we are developing an ultrasensitive procedure for the detection of carcinogen-DNA adducts. The method is based upon postlabeling of carcinogen-DNA adducts by acetylation with 14C-acetic anhydride combined with quantitation of 14C by accelerator mass spectrometry (AMS). For this purpose, adducts of benzo[a]pyrene-r-7,t-8-dihydrodiol-t-9,10-epoxide (BPDE) with DNA and deoxyguanosine (dG) were synthesized. The most promutagenic adduct of BPDE, 7R,8S,9R-trihydroxy-10S-(N(2)-deoxyguanosyl)-7,8,9, 10-tetrahydrobenzo[a]pyrene (BPdG), was HPLC purified and structurally characterized. Postlabeling of the BPdG adduct with acetic anhydride yielded a major product with a greater than 60% yield. The postlabeled adduct was identified by liquid chromatography-mass spectrometry as pentakis(acetyl) BPdG (AcBPdG). Postlabeling of the BPdG adduct with 14C-acetic anhydride yielded a major product coeluting with an AcBPdG standard. Quantitation of the 14C-postlabeled adduct by AMS promises to allow detection of attomolar amounts of adducts. The method is now being optimized and validated for use in human samples.

Comparisons of the binding of [14C]radiolabelled tamoxifen or toremifene to rat DNA using accelerator mass spectrometry.

Tamoxifen, widely used as adjuvant therapy in the treatment of breast cancer, is now undergoing trials as a cancer chemopreventative agent. Previous work has shown an association between 32P-postlabelled adducts in rat liver DNA and the development of liver tumours. With the use of accelerator mass spectrometry, [14C]tamoxifen was shown to bind to liver DNA of female rats in a dose-dependent manner and was linear over 0.1-1 mg/kg, compatible with the therapeutic dose used in women (20 mg/person per day). Radiolabel could also be detected in extrahepatic organs, including reproductive and GI-tract, where levels were about 18 and 46%, respectively those seen in liver. Following enzymatic hydrolysis of liver DNA, normal nucleotides by HPLC showed < 2% incorporation of the [14C]radioactivity while > 80% appeared as non-polar products. In contrast, when animals were given an equivalent dose of [14C]toremifene, binding to DNA was an order of magnitude lower than that seen with tamoxifen and no evidence of non-polar adducted nucleotides following HPLC. However, in vitro, using human, rat or mouse liver microsomal preparations, NADPH-dependent binding of both toremifene and tamoxifen to calf thymus DNA could be demonstrated, suggesting that under favourable circumstances toremifene is capable of undergoing conversion to reactive intermediates.

Application of accelerated mass spectrometry (AMS) in DNA adduct quantification and identification.

DNA adducts are nucleotide bases that have been covalently modified by reactive electrophilic chemical intermediates, and have been extensively researched for their role in mutagenesis and carcinogenesis. However, many DNA adduct measurement techniques have difficulty in the quantification of adducts at realistic human exposure levels. We are using the extremely sensitive analytical technique of accelerator mass spectrometry (AMS) to study adducts either at low dose or directly in humans. AMS is a technique for measuring isotope ratios with high selectivity, attomole sensitivity (10(-18) mol) and precision of 0.5-10%, depending on isotope level and preparation method. This sensitivity and precision is being used to study the dose-response, toxicokinetics, and toxicodynamics of DNA adduct formation and removal following administration of very low doses of chemicals.

Role of sulfation and acetylation in the activation of 2-hydroxyamino-1-methyl-6-phenylimidazo[4,5-b]pyridine to intermediates which bind DNA.

Mutagenic activity associated with amino-imidazoazaarene food-derived mutagens such as 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) appears to be dependent upon N-hydroxylation, though additional metabolic pathways may be involved in the production of the ultimate reactive intermediate which covalently binds DNA. We have evaluated the ability of 2-hydroxyamino-1-methyl-6-phenylimidazo[4,5-b]pyridine (N-hydroxy-PhIP) to bind DNA in vitro and have determined which secondary metabolic pathways are involved in the production of electrophilic intermediates. Incubation of DNA with 10 microM N-hydroxy-PhIP alone or with mouse-liver cytosol did not result in detectable adduct formation. Addition of 3'-phosphoadenosine 5'-phosphosulfate or acetyl coenzyme A to cytosolic incubations containing N-hydroxy-PhIP resulted in DNA adducts which could be detected by 32P-postlabeling at levels of 594 and 30 fmoles/micrograms DNA, respectively. Addition of 3'-phosphoadenosine 5'-phosphosulfate and to a lesser extent acetyl coenzyme A to cytosolic incubations also increased the rate of degradation of the unstable N-hydroxy-PhIP intermediate. These data suggest that both sulfation- and acetylation-dependent metabolic pathways may be important in the mammalian genotoxic actions of PhIP.