Chem. Res. Toxicol. 1994, 7, 503-510
503
Separation of 32P-PostlabeledDNA Adducts of Polycyclic Aromatic Hydrocarbons and Nitrated Polycyclic Aromatic Hydrocarbons by HPLC Leon C. King,*$?Michael George,S J a n e E. Gallagher,? and Joellen Lewtast Health Effects Research Laboratory, US.Environmental Protection Agency, Research Triangle Park, North Carolina 27711, and Integrated Laboratory Systems, Research Triangle Park, North Carolina 27709 Received January 24, 1994'
The 32P-postlabelingassay, thin-layer chromatography, and reverse-phase high-pressure liquid chromatography (HPLC) were used to separate DNA adducts formed from 10polycyclic aromatic hydrocarbons (PAHs) and 6 nitrated polycyclic aromatic hydrocarbons (N02-PAHs). The PAHs included benzoG]fluoranthene, benzo[kl fluoranthene, indeno[l,2,3-cdl pyrene, benzo[a] pyrene, chrysene, 6-methylchrysene,E~methylchrysene, and benz[al anthracene. The N02-PAHs included 1-nitropyrene, 2-nitrofluoranthene, 3-nitrofluoranthene, 1,6-dinitropyrene, 1,3-dinitropyrene, and l,&dinitropyrene. Separation of seven of the major PAH-DNA adducts was achieved by an initial PAH HPLC gradient system. The major N02-PAH-DNA adducts were not all separated from each other using the initial PAH HPLC gradient but were clearly separated from the PAH-DNA adducts. A second N02-PAH HPLC gradient system was developed to separate NO2-PAH-DNA adducts following one-dimensional TLC and HPLC analysis. HPLC profiles of NO2-PAH-DNA adducts were compared using both adduct enhancement versions of the 32P-postlabelingassay to evaluate the use of this technique on HPLC to screen for the presence of NOz-PAH-DNA adducts. To demonstrate the application of these separation methods to a complex mixture of DNA adducts, the chromatographic mobilities of the 32P-postlabeledDNA adduct standards (PAHs and N02-PAHs) were compared with those produced by a complex mixture of polycyclic organic matter (POM) extracted from diesel emission particles. The diesel-derived adducts did not elute with the identical retention time of any of the PAH or N02-PAH standards used in this study. HPLC analyses of the NO2-PAH-derived adducts (butanol extracted) revealed the presence of multiple DNA adducts. HPLC analyses of a nuclease P1 digestion of these xanthine oxidase-derived NO2-PAH-DNA adducts resulted in a significant reduction and in some samples complete loss of adducts when compared to the HPLC profiles of the butanol-extracted samples, suggesting that these DNA adducts are derived from N-substituted aryl compounds. Rat liver S9-mediated metabolism of the described N02-PAH standards did not produce any measurable DNA adducts using the described methodology. The results of this study demonstrate the potential of the 32P-postlabeling assay coupled to HPLC for the separation of both PAH and NO2-PAH-DNA adducts in complex environmental mixtures.
Introduction Humans are exposed to air pollution sources of polycyclic organic matter (POM)l containing both polycyclic aromatic hydrocarbons (PAHs) and nitrated polycyclic aromatic hydrocarbons (N02-PAHs) (1-7). Exposures to POM have historically focused on environmental measurement of PAHs and most particularly benzo [alpyrene (B[a]P) ( 1 ) . N02-PAHs have been identified in extracts of diesel exhaust (3-5), ambient air (2, 6), and other combustion emission particulate matter (7) and have been shown to contribute substantially to the mutagenicity in the Salmonella assay. The importance of exposure and cancer risk of nitro-substituted derivatives relative to the aromatic parent compound is an important unresolved question and has led to studies focusing on their environmentaloccurrence, DNA adduct-forming potential, and carcinogenicity. * Addresscorrespondenceto this author at the Health Effects Research Laboratory, U.S.Environmental Protection Agency, Research Triangle Park, NC 27711. Phone: (919) 541-0720. + U S . Environmental Protection Agency. f Integrated Laboratory Systems. 0 Abstract published in Advance ACS Abstracts, May 15, 1994.
The covalent modification of DNA by chemical carcinogens is believed to be a crucial event in the initiation of mutations and cancer (8-1 0). The detection of covalent adducts formed when genotoxic agents interact with DNA has become increasingly important in the assessment of human exposure to carcinogens (111. 32P-postlabelingis recognized as the most sensitive method for the detection 1 Abbreviations: PAHs, polycyclicaromatic hydrocarbons; NOrPAH, nitrated polycyclic aromatic hydrocarbons;B[alP, benzo[alpyrene;B[al A, benz[alanthracene; 5-MC, 5-methylchrysene; 6-MC, 6-methylchryeene; BulF, benzo~lfluoranthene;B[klF, benzo[klfluoranthene; IP, indenopy-rene1,2-oxide;1-NP,1-nitropyrene;2-NF, 2-nitrofluoranthene;3-NF, 3-nitrofluoranthene; 1,6-DNP, 1,g-dinitropyrene; 1,3-DNP, 1,3-dinitror-7,tpyrene; 1,8-DNP, 1,gdinitropyrene; B[alP-7,8-diol-9,10-epoxide, 8-dihydroxy-t-9,10-epoxy-7,8,9,10-tetrahydrobenzo[alpyrene; B[alA-3,4diol-l,Z-epoxide, r-3,t-4-dihydroxy-t-1,2-epoxy-l,2,3,4-tetrahydrobenz[alanthracene; chrysene-l,Z-dio13,4-epoxide, r-ltt-2-dihydro~3,4-epoxy1,2,3,4-tetrahydrochrysene; B G]F-4,5-diol-6,6a-epoxide, anti-BGlF-4,b dihydrodiol-6,6a-epoxide; B~]F-9,l0-diol-ll,l2-epoxide, anti-BGlP-9,10dihydrodiol-11,lZ-epoxide; B[klF-8,9-diol-l0,11-epo~de,anti-B[klF-8,9 dihydrodiol-l0,ll-epoxide; B[klF-8,9-oxide, benzo[klfluoranthene 8,9oxide; 5-MC-1,2-diol-3,4-epoxide,r-l,t-2-dihydroxy-t-3,4-epoxy-l,2,3,4tetrahydro-5-methylchrysene;&MC-1,2-diol-3,4-epoxide,r-l,t-Bdihydroxyt -3,4-epoxy-l,2,3,Ctetrahydro-&methylchrysene; POM, polycyclic organic matter; EOM, extractable organic matter; CT, calf thymus; PEI-cellulose, poly(ethy1ene imine)-cellulose; MN, micrococcal endonuclease; SPDE, calf spleen phosphodiesterase; RAL, relative adduct level.
This article not subject to U.S. Copyright. Published 1994 by the American Chemical Society
504 Chem. Res. Toxicol., Vol. 7, No. 4, 1994
of DNA adducts (12) and has been widely used in the analysis of DNA isolated from rodents and humans exposed to complex environmental mixtures (11, 13). Identification of DNA adducts is often not definitive due to variations in TLC migration of DNA adducts and lack of separation of adducts. Therefore, HPLC in conjunction with 32P-postlabelingand TLC has been recently used to improve the separation and identification of specific carcinogen-DNA adducts (14-1 7). Most of these studies involved the analysis of DNA adducts formed by single carcinogens (14-18) or did not achieve complete separation of a mixture of DNA adducts formed by reacting select anti-diol epoxides of PAHs with calf thymus DNA (18). Recently, a reverse-phase HPLC procedure was developed and used in the separation of the major 32P-postlabeled DNA adducts formed following in vitro modification of salmon sperm DNA with 10 different PAH diol epoxides (19). This procedure has also been used to isolate the major benzo[bl fluoranthene-DNA adducts formed following in vivo exposures (20)and benzo[ghilperylene DNA adducts formed following in vivo exposures with mice and in vitro modifications with salmon sperm DNA (21). Most recently, it has been used to identify those PAHs responsible for the formation of coal tar DNA adducts in mouse skin (22) and in the detection of the major DNA adducts of both benzolil fluoranthene and benzo[blfluoranthene in mouse skin (23,24). The objective of the present study was to optimize and improve this methodology for the separation of both PAH- and N02-PAHDNA adducts. Using an in vitro calf thymus DNA model system, DNA adducts were generated using two different activation systems, a microsomal mixed-function oxidase (S9) and a nitroreductase (xanthine oxidase) system. We also evaluated whether differences in the labeling efficiencies as a result of nuclease P1 or butanol extraction enhancement versions of the 32P-postlabelingassay could be used to test for the presence or absence of N02-PAHDNA adducts separated by HPLC. The DNA adducts derived from diesel emissions were applied to these methods to demonstrate the potential application of these procedures for the characterization of complex mixtures of DNA adducts.
Materials and Methods Chemicals. Caution: B[j]F, B[k]F, B[a]A, chrysene, 5-MC, 6-MC, IP, B[a]P, 1-NP, 1,3-DNP, 1,6-DNP, 1,8-DNP, 2-NF, 3-NF, and the derivatives described in this paper have been determined to be carcinogenic to laboratory animals. Therefore, protective clothing should be worn and appropriate safety procedures should be followed when working with these compounds. Calf thymus (CT) DNA was purchased from CalBiochemCorp. (LaJolla, CA). Poly(ethy1eneimine)-cellulose (PEI-Cellulose) TLC plates were prepared as described by Gupta et al. (25) except that PEI solution (50%aqueous) was purchased from Aldrich Chemical Co. (Milwaukee, WI). Micrococcal nuclease (MN) and nuclease P1 were purchased from Sigma Chemical Co. (St. Louis, MO). Calf spleen phosphodiesterase (SPDE) was purchased from Boehringer Mannheim (Indianapolis, IN). T d polynucleotide kinase was from Pharmacia (Piscataway, NJ). [+zP]ATP (>3000 Ci/mmol) in an aqueous solution containing 5 mM 2-mercaptoethanol was obtained from Amersham (Arlington Heights, IL). Flo-Scint I1 was obtained from Radiomatic Instruments & Chemical Co., Inc. (Meriden, CT). Benzo[a]pyrene (B[a]P-7,8diol-9,10-epoxide),benz[al anthracene (B[a]A-3,4-diol-1,2-epoxide), and chrysene (chrysene-l,2-diol3,4-epoxide) were obtained from NCI Chemical Carcinogen Repository (Bethesda, MD). 5-Methylchrysene (5-MC-1,2-diol-
King et al. 3,4-epoxide) and 6-methylchrysene (6-MC-1,2-diol-3,4-epoxide) were a gift from Dr. Stephen Hecht, American Health Foundation (Valhalla, NY). 1-Nitropyrene (1-NP), 2-nitrofluoranthene (2NF), and 3-nitrofluoranthene (3-NF) were obtained from Midwest Research Institute (Kansas City, MO). 1,6-, 1,8-, and 1,3dinitropyrene (DNP) (99% purity) were purchased from Aldrich. The internal UV standard cis-9,10-dihydroxy-9,lO-dihydrophenanthrene was a gift from Dr. David H. Phillips, Haddow Laboratories, The Institute of Cancer Research (Sutton, Surrey, U.K.). Theanti-BLjlF-4,5-dihydrodiol-6,6a-epoxide(BLj]F-4,5diol-6,6a-epoxide)-modifiedCT DNA, anti-BLjlF-9,lO-dihydrodiol-ll,l2-epoxide (BLj]F-9,10-diol-ll,12-epoxide)-modified CT DNA, anti-B [k] F-8,9-dihydrodiol-10,ll-epoxide (B[k]F-8,9diol-l0,11-epoxide)-modifiedCTDNA, benzo[klfluoranthene 8,9oxide (B[k]F-8,9-oxide)-modifiedCT DNA, and indenopyrene 1,a-oxide (1P)-modified mouse skin DNA were a gift from Dr. Eric Weyand, Rutgers, The State University of New Jersey, College of Pharmacy (Piscataway, NJ). All other chemicals were reagent grade and commercially available. Diesel Particles Extract. The POM from a Volkswagen Rabbit's diesel emission particles was isolated by soxhlet extraction of the particles with dichloromethane (26) to obtain the extractable organic matter (EOM) and quantitatively analyzed for individual PAHs and N02-PAHs (27). This diesel sample contains benzo[alpyrene (B[alP) (30pg/g of EOM), benz[alanthracene (B[alA) (130 pg/g of EOM), chrysene and triphenylene (220 pg/g of EOM), IP (70 pglg of EOM), benzoLj1fluoranthene (BLjlF) (90 pglg of EOM), benzo[klfluoranthene (B[k]F) (60 pg/g of EOM), 1-NP (589 pg/g of EOM), 3-NF (1.2 pglg of EOM), and the DNP isomers (0.4-0.6 pg/g of EOM) (27). In Vitro Modification of CT DNA Diol Epoxides of PAH. The procedure used to generate PAH-DNA adducts [B[a]P, B[a]A, chrysene,5-methylchrysene(5-MC)and 6-methylchryaene (6-MC)I was as described (18). Solutions containing 0.05 M TrisHC1 buffer (pH 7.0))CT DNA (1mg/mL), and 662 pM anti-diol epoxide of B[a]P, B[a]A, chrysene, 5-MC, and 6-MC were incubated a t 37 "C for 18 h. The final incubation volume was 12 mL. The PAH-modified CT DNA samples were extracted as described below. Nitrated Polycyclic Aromatic Hydrocarbons. The generation of NOg-PAH-DNA adducts through xanthine oxidasecatalyzed nitroreduction of 1-NP, 2-NF, 3-NF, 1,6-DNP, 1,8DNP, and 1,3-DNPwas essentially as described (27,28). Briefly, solutions containing 50 mM potassium phosphate buffer (pH 5.8), 3.7 mM hypoxanthine, CT DNA (1 mg/mL), and 20 pM 1-NP,2-NF, 3-NF, 1,6-DNP, 1,8-DNP,and 1,3-DNPwere purged of oxygen by allowingthe solutions to sit overnight in an anaerobic chamber (Coy Laboratory Products, Ann Arbor, MI, containing 10% Hz,85% Nz,and 5 % Cog, National Specialty Gases, Raleigh, NC). Xanthine oxidase was placed in the chamber the following morning, purged of oxygen for 2 h, and subsequently added (0.5 unit/mL) to the described solution and incubated anaerobically for 4 h a t 37 "C. The final incubation volume was 2 mL. Samples were removed from the anaerobic chamber and placed in a chemical fume hood, and the reaction was terminated by the addition of 10 mL of ethyl acetate. CT DNA (1 mg/mL) was also incubated with the described individual N02-PAHsfor 4 h at 37 "C, 80 pM each in the presence of Aroclor-inducedrat liver S9 (0.5mg/mL) prepared as previously described (29). The NOz-PAH-modifiedCT DNA samples were extracted as described in the DNA adduct isolation section. In Vitro Modification of CT DNA with a Diesel Particle Extract. CT DNA (1mg/mL) was modified by incubation for 4 h a t 37 "C with a 100 pg/mL solution of the diesel particle extracts (100 pg/mL) in the presence of either Aroclor-induced rat liver S9 (0.5 mg/mL) or xanthine oxidase (0.5 unit/mL) as described (27,28). The final incubation volume was 2 mL. The samples were removed from the anaerobic chamber and placed in a chemical fume hood, and the reaction was terminated by the addition of 1volume of chloroform/isoamyl alcohol/phenol(24: 1:25). The diesel particle extract-modified DNA sample was extracted as described in the DNA adduct isolation section.
HPLC Separation of 32P-Postlabeled DNA Adducts
Isolation of DNA Adducts. The procedure used to isolate the PAH- and NO*-PAH-modifiedCT DNA was as described by Amin et al. (18). The incubates were extracted 3 times with 10 of mL ethyl acetate. The aqueous portion cooled to 0 OC, and 30 mL of cold ethanol was added slowly. The DNA was precipitated, spooled out with a glass rod and washed with 10 mL of ethanol, 10 mL of acetone and 10 mL of ether; it was then dried under a stream of nitrogen at 37 "C. CT DNA modified with the diesel particle extract was isolated as described by Gupta (30). The modified DNA sample residues were resuspended in 2 mL of HPLC-grade water and placed in a -80 OC freezer until time of digestion and adduct analysis as described below. S2P-postlabelingAnalysis. The CT DNAs (12 pg) modified with PAHs and diesel extracts were digested to mononucleotides at 37 "C for 3.5 h with MN and SPDE as described (29). CT DNA (50 pg), modified separately with diesel emission extract, and 1-NP, 3-NF, 2-NF, 1,6-DNP, 1,8-DNP, and 1,3-DNP, in the presence of xanthine oxidase, were digested to mononucleotides at 37 OC for 3.5 h with MN and SPDE as described (31). DNA adducts were enriched with the butanol extraction method for the N02-PAHs and diesel emission extract activated by xanthine oxidase (29)and by nuclease P1 treatment (12)for the PAH diol epoxide and diesel emission extract DNA modified in the presence of S9, as previously described (27). The described samples were incubated with 50 pCi of ATP (Amersham, 3000 Ci/mmol) and 3.5 units of T4 polynucleotide kinase for 30 min, and the total incubates were spotted onto PEI TLC plates. TLC Separation and Preparation of Samples for HPLC. The labeled digests of the PAH diol epoxide-modifiedCT DNA adduct standards were spotted on 10 cm X 10 cm PEI-cellulose sheets and separated using a 2-dimensional TLC system (Dl0 5 ) as described (27). Eluents were D1 (1.0 M NaHzP04, pH 6.8); D2 (2.5 M ammonium formate, pH 3.5); D3 (4 M lithium formate/7 M urea, pH 3.4); D4 (1.07 M LiCl, 0.5 M Tris, 7 M urea, pH 8.0);and D5 (0.5M magnesium chloride). Radioactivity of spots was located by autoradiography and relative adduct level (RAL) quantitated as previously described (27). To prepare samples for HPLC analysis, the protocol was used as described (15,19)with minor modifications(Le., PEI-cellulose layer not scraped off, and the volume of the extracting solvent was 1000 pL). The labeled mixtures were spotted on P E E cellulose TLC plates and developed overnight with 1M sodium phosphate buffer (pH 6.8) (D1 direction only). The major PEIcellulose adduct spots (1 cmz) were cut out and placed in scintillation vials containing 2.5 mL of ethanol, and the total radioactivity was determined. The ethanol was decanted after counting, and adducted spots were extracted overnight (18 h) in 1mL of ppidinium formate (pH 4.0). Approximately 80-95% of the total radioactivity associated with each spot was detected after the overnight extraction (data not shown). The extracts were transferred to 1.5-mL microcentrifuge vials, and the particulate was sedimented by centrifugation (10 000 rpm, 1min). The samples were reduced to dryness using a Speed-Vac (Savant Instruments, NY), resuspended with 100 pL of methanol/0.5 M sodiumphosphate buffer (pH 2.0) (9l),vortexed, and centrifuged again (10 000 rpm, 1min). The supernatant (75 pL) from each sample was carefully removed and spiked with 4 nmol(3 pL) of the UV marker cis-9,10-dihydroxy-9,lO-dihydrophenanthrene. The volume was adjusted to 100 pL with a mixture of MeOH/0.5 M NaHzP04 buffer (pH 2.0) (9:l) and placed in 100-pL autosampler vials for HPLC analysis. A mixture (100 WLeach) of the predominant a2P-postlabeledPAH-DNA adducts and NOzPAH adducts (1-NP,2-NF,and 3-NF)was also analyzed by HPLC using the PAH gradient as described below. HPLC Analysis of SaP-PostlabeledAdducts. Separation of the 3zP-postlabeled3',5'-bisphosphate adducts was performed on a Varian Model 5060 HPLC system equipped with a Model 100 detector at 260 nm and a Model CDS 401 Vista data system, and a Varian Model 5500 HPLC system equipped with a Model 200 detector at 260 nm and a Model DS-654 data system, Varian Associates, Inc. (Walnut Creek, CA). Both HPLC systems are
Chem. Res. Toxicol., Vol. 7, No. 4, 1994 505 equipped with a Varian Model 9100 autosampler, Varian Associates,Inc. (Walnut Creek, CA), and are used extensively in this study. The extracts of the excised adducts (100 pL) were applied to a 5-pm, 4.6 mm X 250 mm Zorbax phenyl-modified column (MAC-MOD Analytical, Inc., Chadds Ford, PA), and eluted with two linear gradients using the same solvent systems. Solvent A: 0.5 M NaHzHPOI buffer (pH 2.0); solvent B: 90% methanol and 10% solvent A. The first gradient as described by Pfau and Phillips (19) is referred to here as the PAH gradient. The PAH gradient was 0-12.5 min, 10-43% B; 12.5-60 min, 4347% B; 60-80 min, 90% B; 80-85 min, 95% B; 85-90 min, 10% B. The second, N02-PAH gradient was 0-12.5 min, 10-43% B; 12.5-45 min, 43-95% B; 45-50 min, 95% B; 50-65 min, 10% B. The flow rate for either HPLC gradient system was 1 mL/min, and the column was allowed to equilibrate at the initial solvent ratio (90% A10% B) for 5 min before subsequent analysis. Analysis for the presence of szP used in-line flow-through scintillation counters (Model A-250 Flo-One\B, Radiomatic Instruments & Chemical Co., Inc., Tampa, FL, on the Varian Model 5060; and the Model A-500 Flow-One\B, Radiomatic Instruments & Chemical Co., Inc., Tampa, FL, on the Varian Model 5500). Both in-line flow-through scintillation counters are equipped with lead shieldingwhich covers the photomultiplier and flow cell (1.0 mL) and resulted in a 43% (50-24 dpm) reduction in the normal background count. Using the described gradient conditions, the counting efficiency for both counters for detecting 3ZP was 100% (data not shown).
Results PEI-cellulose TLC profiles of t h e 32P-postlabeledPAH diol epoxide-CT DNA adduct standards generated and analyzed in this study are shown in Figure 1(panels a-j). These TLC profiles isolated using a two-dimensional chromatography system indicated the presence of multiple adducts (both major a n d minor). IP-modified mouse skin DNA (Figure 1, panel j) demonstrated t h e presence of only one major adduct near t h e origin. Quantitative analyses of the major DNA adduct spots isolated by TLC, as indicated numerically (spot 1) on each of t h e autoradiographs, demonstrated that t h e major adduct accounted for 58% of the modification of B GIF-4,5-diol-6,6a-epoxide, 97% for B[klF-8,9-oxide, 70% for B[alA-3,4-diol-1,2epoxide, 58% for chrysene-1,2-diol3,4-epoxide, 73 % for BG] F-9,10-diol-11,12-epoxide, 56 % for 5-MC-1,2-diol-3,4epoxide, 83% for B[klF-8,9-diol-10,11-epoxide, 49% for 6-MC-1,2-diol-3,4-epoxide,85% for B[alP-7,8-diol-9,10epoxide, and 100% for IP (data not shown). HPLC analyses (spot 1)for each of t h e major PAH diol epoxide-CT DNA adduct spots are shown in Figure 1. T h e HPLC profiles of each of t h e major adducts with the exception of B[alA-3,4-diol-1,2-epoxide, 5-MC-172-diol3,4-epoxide, 6-MC-1,2-diol-3,4-epoxide, a n d IP resulted in the isolation of a major adduct peak a n d several minor peaks. In contrast t o t h e HPLC adduct profiles, the PEIcellulose TLC profiles showed a similar migration pattern for t h e major DNA adducts of B ~ l F - 9 , 1 0 - d i o l - l l , l 2 epoxide (Figure 1,panel e), B[klF-8,9-diol-10,1l-epoxide (Figure 1, panel g), a n d B~alP-7,8-diol-9,10-epoxide (Figure 1, panel i); notable differences in the retention time of these adducts were achieved by HPLC. Figure 1,panel k, shows a typical HPLC profile (using t h e initial P A H gradient) of a mixture of t h e major P A H diol epoxide-and N02-PAH (1-NP, 2-NF, and 3-NF)-CT DNA adducts. Separation of 7 of t h e major PAH-DNA adducts and one of t h e N02-PAH adducts (1-NP) was achieved following 32P-postlabeling,2-dimensional TLC,
King et al.
506 Chem. Res. Toxicol., Vol. 7, No. 4, 1994 B[ klF-8,9-OXIDE
B~j1F-4.5-DlOLG,6A-EPOXIDE
B[alA-3,4-DlOL-1,2-EPOXIDE
:::: JK-=--w1000 0
, ,
,-,-._,,
,. l O M J O 4 0 5 0 6 0 7 0 8 0 9 0
0
,
,,
,
, ,
, , , , ,
0
0
10
40
20 30
CHRYSENE-1,2-DIOL3,J-EPOXlDE
0
60 70 80 90
50
1 0 2 0 3 0 4 0 5 0 6 0 7 0 8 0 9 0
0
5-MC-lV2-D10L-3,4-EPOXIDE
BIjIF-9~10-DIOL11,12-EPOXIDE
25000 20000
L
15000
300000
10000
200000
5000
100000
140 105 70
o 0
10
20 30 40
50 60 70 80 90
BlkJF-8,9-D10LIO,I 1-EPOXIDE 5000
4000 3000 2000 low 'Oo0 0
0 0
l O M J O 4 0 5 0 6 7 0 8 0 9 0 0
0
0 1 0 X I ~ 4 0 5 0 6 0 7 0 8 0 9 0
II,, ,
,
9
h
,
a io
,
, ,,
,,
,
,
,
,, , ,
, , , ,,
so
30
40
,
,
, ,
900 so0
60
70
80
90
vi
loo0
,
"I
A
0 0
60 70 80 90
1
, , , ,
,
,,-1
1 0 2 0 J O 4 0 5 O W 7 0 8 0 9 0
MIXTURE OF PAH AND NITRATED PAH DNA ADDUCTS I . BljIF-4.5-6,6A-EPOXlDE 2. I-NP
1000
50
i
J-~J., ,
20 30 40
10 20
IO. 3-NF (SPOT I )
3 B l s l A J , ~ - D I O M , h A - E P O X l D E 11, 3-NF (SPOT 2)
700
600 500
4
400
300 200
100
0
l----rT
0
10 20 30
40
50
--l
60 70 80 90
0
10 20
30
40
50
Time (min)
60 70
80
Chgsene-1,2-DIO~3,J-EPOXlDE
5. B[j~F-9,10-DIOL11,12-EPOXlDE 12. IP-I,2-OXIDE 6 . 5-MC-1,2-DIOL3,4-EPOXIDE + 3-NF (SPOT 3) + B[KlF-8,9-DIOLIO. II-EPOXIDE + 2-NF 7. 6-MC-I.2-DIOL3,J-EPOXlDE 8. BlalP-7,8-DIOL9,lCEPOXlDE .. 90 9. 3-NF(SPOTI)
Figure 1. HPLC analyses of the major 32P-postlabeledPAH diol epoxide-DNA adducts (spot 1)and a mixture of each PAH adduct spot with the major adduct spots of three N02-PAHs (1-NP, 2-NF, and 3-NP). The 32P-postlabeleddigests of each of the above were separated by PEI-cellulose TLC (Dl-D5), and the major DNA adducts were analyzed by HPLC as described in the Materials and detected Methods section. The arrows indicate the position of elution of a marker of cis-9,10-dihydroxy-9,lO-dihydrophenanthrene by its UV absorbance a t 260 nm. Panels a-j are HPLC profiles and autoradiographs of TLC maps of 32P-postlabeledDNA adducts. Panel k is a mixture of the major isolated adduct spots of both PAH and three N02-PAH adduct standards. The number in the autoradiographs indicates the position of the major adduct. The films were exposed for 3 min for the PAHs and 30 min for the N02-PAHs a t room temperature.
and HPLC analysis. The major adducts of 5-methylchrysene-1,Bdiol 3,4-epoxide and B[klF-8,9-diol-10,11epoxide coeluted at 41 min; 2-NF and 3-NF coeluted (80 min) using the initial PAH HPLC gradient. The lateeluting adduct peaks of the NO2-PAH-DNA adducts suggested that this more hydrophobic characteristic may be useful in determining whether N02-PAHs were in complex environmental mixtures. Figure 2 shows the HPLC profiles of the diesel emission extract of CT DNA activated in the presence of Aroclor-induced rat liver S9 (panel 2a) and xanthine oxidase (panel 2b) and the described NO2-PAHs activated in the presence of xanthine oxidase (panels 3b-8b) using the PAH HPLC gradient. The HPLC profile of the diesel extract-modified CT DNA in the presence of S9 and adducts enriched by nuclease P1 digestion indicates the presence of a predominant adduct peak at 40 min and two minor peaks at 42 and 60 min. A comparison of the HPLC profile of the mixture of isolated PAH standards (Figure 1, panel k) suggests that only a minor amount of the B[alP-7,8-diol-9,10epoxide adduct was present; the major adduct@) eluted in the region of 5-MC-1,2-diol-3,4-epoxideand B[klF-8,9-
diol-l0,ll-epoxide adducts. When xanthine oxidase was substituted in the activation media for Aroclor-induced rat liver S9 to enhance the formation of N02-PAH adducts followed by the butanol-enhanced postlabeling method, a markedly different HPLC profile was observed (Figure 2, panels 2b-8b). The major adducts of the diesel extract and the N02-PAHs indicated a similar chromatographic pattern (late-elutingpeaks), with the predominant adducts eluting at 82 min for the diesel extract-modified CT DNA and 1,3-DNP, 80 min for 3-NF, and 83 min for 2-NF, 1,8DNP, and 1,6-DNP. Minor adducts at 43 and 60 min for the VW diesel extract, 59,76, and 82 for 3-NF, and 85 min for 1,6- and 1,8-DNP were also detected. Having verified our hypothesis that the initial gradient could be used to detect both PAHs and N02-PAHs in a sample simultaneously, but because of the poor resolution of the NOzPAH-CT DNA adducts, we developed a new HPLC gradient specifically for the analysis of N02-PAHs. We also evaluated whether differences in the labeling efficiences as a result of the nuclease P1 or butanol adduct enhancement versions of the 32P-postlabelingassay could be used to test for the presence of NO2-PAH-DNA adducts.
Chem. Res. Toricol., Vol. 7, No. 4, 1994 507
HPLC Separation of S2P-Postlabeled DNA Adducts CONTROL CALF
THYMUS DNA
VW
2n
3h 3-NITROF'LUORANTHENE
DIESEL
4200 3500 2800 2100 1400 0 1 0 2 0 3 0 4 0 5 0 8 0 7 0 8 0 9 0
240 180 120 80
0
l b CONTROL CALF THYMUS DNA
10 20 30 40
2b
50 60
70 80 90
VW DIESEL
0
9 ..
1 0 2 0 3 0 4 0 5 0 8 0 7 0 8 0 9 0
4b 2-NITROFLUORANTEENE
1400
MOT
840
700 560 420 280 140
E
g:
0
0
1 0 2 0 3 0 4 0 5 0 6 0 7 0 8 0 9 0
10 20
30 40
50
60 70 80 90
_. 1.3-DINITROPYRENE
1-NlTROPYRENE
0
1 0 2 0 3 0 4 0 5 0 6 0 7 0 8 0 9 0
o
1200 1080
gso
840 720 800
480 360 240 1x)
0
0
1 0 2 0 3 0 4 0 5 0 8 0 7 0 8 0 9 0
0
12000 10500 4500 3000 1500
0 10 20 30 40 50 80 70 80 90 Time (min)
Figure 2. HPLC profies of the major DNA adducta of VW diesel extract, 2-NF, 3-NF 1,3-DNP, l,&DNP, and 1,6-DNP separated by PEI-cellulose TLC (D1 direction only); the excised adduct spots were analyzed by HPLC using the initial gradient for PAHs as described in the Materials and Methods section. Panels l a and 2a show control CT DNA or diesel extract adducta derived following incubation with S9 and nuclease P1 treatment. Panels lb-8b are profiles of the control CT DNA, diesel extract, and select NOp-PAH derived following xanthine oxidase incubation and butanol extraction. The arrows indicate the position of elution of a marker of cis-9,10-dihydroxy-9,lO-dihydrophenanthrenedetected by ita W absorbance at 260 nm. Modified CT DNA (50 pg; xanthine oxidase-catalyzed nitroreduction) for the diesel extract and each of the described N02-PAH samples were digested, treated as described (29)and used for a more thorough HPLC analysis of isolated adduct spots following a one-dimensionalTLC using D1. Figure 3 shows the result of the HPLC analysis using the new N02-PAH gradient following the in vitro modification of CT DNA with diesel emission extract, 1-NP, 2-NF 1,3-DNP, 1,6-DNP, and l,&DNP. HPLC analysis of the diesel adducts (enriched by the butanol extraction method before 32P-postlabeling)using the new HPLC N02-PAH gradient revealed six component peaks, two closely eluting predominant adducts eluting at 35 and 36 min and three minor adducts at 33, 38, and 39 min. Figure 3, panel 2b, shows the HPLC profile of a composite sample of the diesel adducts enriched by the nuclease P1 method. Only one adduct peak at 36 min was observed. Panels 3a-8a of Figure 3 show the HPLC adduct profile of select N02-PAH adducts enriched by the butanol extraction method before 32P-postlabeling. When compared to the HPLC profiles of the diesel-derived NO2PAH adducts, neither 1-NP,which is present in the highest concentration in this diesel sample, the dinitropyrenes, nor 3-NF produced adduct profiles with a similar retention
time. Only the adducts of 2-NF (34 and 36 min) indicated a similar retention. Figure 3, panels 3b-8b, shows the HPLC adduct profiles of select N02-PAH adducts enriched by the nuclease P1 method before 32P-postlabeling. When compared to the HPLC profiles of the butanol-extracted diesel-derived N02-PAH adducts, these adducts were diminishedfollowingnuclease P1 digestion. Only a minor amount of the adducts of 2-NF (36 min) and 3-NF CT DNA adducts (41 min) were present after nuclease P1 digestion. A minor adduct at 36-37 min was also common to all the N02-PAH standards. Experimentswere also performed to determinethe DNA adducts produced by the individual N02-PAH standards and CT DNA incubated in the presence of Aroclor-induced rat liver S9. The samples were subjected to both adduct enhancement procedures and analyzed by HPLC using the N02-PAH HPLC gradient. No measurable adduct peaks were detected in any of the NOrPAH-DNA standards. Experiments were also conducted to estimate the detection limit for 32Pradioactivity as a measurable peak using the in-line flow-throughscintillator detector coupled to HPLC. Select concentrations of the BLilF-9,lO-diol11,12-epoxide-modified CT DNA standard were treated
King et al.
508 Chem. Res. Toxicol., Vol. 7, No. 4, 1994 VW DIESEL
CONTROL CALF THYMUS DNA
1-NITROPYRENE
1,3-DINITROPYRENE
9
la
i
I
250000 300000 200000 150000 j 100000 4
j
84
.-
7n
56 42 28 14 0
A
3a
8ow0.
j(,,,jc
250000 7 200000
60wO
150000 100000 50000
4a
,,,,,
0 0
10 20
30
40
0
50 60 70
56 42 28 14
6000 5400
A 20 30
40
50
0
10 20 30 40 50
VW DIESEL d -
CONTROL CALF THYMUS DNA
lb
10 20 30 40 50 60 70
€0 70
0
1-NITROPYRENE
13-DINITROPYRENE 6000 5400 4800 4200 3600 3000 2400 1800 1200
3b
3000 2400 1800 1200 600
10 20 30 40 50 60 70
2400 im
600 0
€0 70
0
3-NITROFLUORANTHENE
2-NITROFLUORANTHENE
196-DINITROPYRENE
10 20 30 40 50 60 70
1,8-DINITROPYRENE
I400000 1200000 1000000 800000 600000 400000 200000
3ow
27W 24w 21 W
1800 15W 1200 900 6Ml
m
0
0
0
10
20 30
40 50 W
15W
6000 5400 4800 1 4200 3600 3000 2400 1800 1200 600 0 0
10
20
30 40
0
l96-DINlTROPYRENE
6b
900
800
, , , , , , , , , , , , ,
3w 0 0
10 20
30
40
50 W
70
10 20 30 40 50 60 70
1,&DINITROPYRENE 6000
7b
L,,
1200
50 60 70
10 20 30 40 50 60 70
3-NITROFLUORANTHENE
2-NITROFLUORANTHENE 24w 21 W 1800
0
0
70
7
8b
9
x
10 20 30 40 50 60 70
0
10
M
30
1200 40
50
60 70
0
10 20 30 40 50 60 70
Time (min)
Figure 3. HPLC profiles of the major DNA adducts of VW diesel extract, 2-NF, 3-NF 1,3-DNP, 1,s-DNP,and 1,8DNP separated
by PEI-cellulose TLC (D1 direction only); the excised adduct spots were analyzed by HPLC using the new gradient for N02-PAHs as described in the Materials and Methods section.Panels la-Sa are profiles of the control CT DNA, diesel extract,and select N02-PAH derived following xanthine oxidase incubation and butanol extraction. Panels lb-8b are profiles of the control CT DNA, diesel extract, and select N02-PAH derived following xanthine oxidase incubation and nuclease P1 treatment. The arrows indicate the position of elution of a marker of cis-9,1O-dihydroxy-g,lO-dihydrophenanthrene detected by its UV absorbance at 260 nm.
and analyzed as described. The results indicate an ability to detect a level of 0.05 fmol/pg of DNA (RAL = 1.7/108) using our current procedures (data not shown).
Discussion The 32P-postlabelingassay has been widely used in the analysis of DNA adducts isolated from exposures to complex environmental mixtures as reviewed by Beach and Gupta (11). Several recent publications have demonstrated the use of HPLC coupled to in-line flow-through radioactivity detectors to isolate and identify PAH-DNA adducts resulting from exposures to both single chemicals and a coal tar mixture (18-24). We have modified a recently published HPLC procedure (19)to separate both PAH- and NO2-PAH-DNA adducts. The chromatographic mobilities of these 32P-postlabeledDNA standards (PAHs and N02-PAHs) were compared with those produced by a complex mixture of POM extracted from diesel emission particles. We also coupled this HPLC methodology to the diagnostic procedures of Gallagher et al. (27') to metabolize selectively the PAHs (S9) and NO2PAHs (xanthine oxidase-catalyzed nitroreduction) followed by two adduct enhancement procedures (nuclease
P1 digestion vs butanol extraction) to screen for the presence of N02-PAH-DNA adducts. The HPLC methods described here successfully separated most of the PAH and N02-PAH adducts. The PAH gradient separated the 4-5-ring PAH-DNA adduct standards within a 30-50-min chromatographic region. IP, a 6-ring PAH, eluted outside this region (79 min). Several of the N02-PAH-DNA adduct standards comigrated in late-eluting peaks, indicating their less polar character, requiring a high methanol content for elution. These results suggest that N02-PAH-DNA adducts could initially be separated from PAH-DNA adducts in a mixture, but a different gradient must be used to completely resolve the N02-PAHs adducts from each other. The N02-PAH gradient developed for this purpose separates the NO2PAH adducts from each other in a chromatographicregion between 30 and 50 min except for the 1-NP and 1,3-DNP adducts. The PAH-DNA adduct standards (chrysene1,2-diol 3,4-epoxide, B GI F-9,10 diol-11,12-epoxide, and B[alP-7,8-diol-9,10-epoxide) eluted between 25 and 30 min using the N02-PAH gradient, thereby overlapping with the adducts of 1-NP and 1,3-DNP (data not shown). Again, these results demonstrate the less polar character of the N02-PAH constituents, which required a higher
HPLC Separation of 32P-Postlabeled DNA Adducts methanol content for elution (74-87% vs 23-39% for the PAH adducts). The diagnostic selective metabolism (S9 vs xanthine oxidase) and adduct enhancement procedures (nuclease P1vs butanol extraction) applied to these HPLC gradients showed a dramatic loss of N02-PAH-DNA adducts after nuclease P1 treatment. N02-PAHs treated with S9 and CT DNA under the same conditions that produced PAHDNA adducts did not result in detectable DNA adducts by HPLC (data not shown). These data are consistent with published reports demonstrating that several arylamine-derived DNA adducts bound to the C8 position of guanine are sensitive to the monophosphatase activity of nuclease P1, with significant losses in labeling efficiences ranging from 65% to 99% (27, 32, 33). A complex mixture of diesel-derived DNA adducts presumably containing both PAH and NOrPAH-DNA adducts (27)was used to evaluate the utility of these HPLC methods. Using TLC after 32P-postlabeling,DNA adducts were found to comigrate with B[alP-DNA adducts. However, the low B[alP content of these emissions led to the previous speculation that B[alP could not account for this adduct (34). The major adduct@)derived from the diesel extract after S9 incubation eluted at 40 min using the PAH gradient and did not elute with a retention time that matched any of the PAH standards used in this study. No measurable adduct peaks were observed in the area where the major adduct of B[a]P-7,8-diol-9,10-epoxide eluted (48 min). This agrees with our previous report that skin painting of diesel extracts resulted in an unusually high DNA adduct level per microgram of B[a]P in diesel samples compared to other POM due to coelution of other DNA adducts with B b I P using TLC (34). Hughes et al. (22) have also demonstrated that multiple PAH-DNA adducts coelute on TLC and may be better separated by HPLC. The diesel-derived DNA adducts formed from xanthine oxidase and separated on the PAH HPLC gradient did result in several late-eluting DNA adducts (79 and 82 min) that could be derived from N02-PAHs such as 1,3-DNP (82 min). The N02-PAH HPLC gradient further separated these DNA adducts into at least four peaks that were dramatically reduced followingnuclease P1treatment. The chromatographic properties on HPLC and reduction by nuclease P1 treatment suggest that these are N02-PAHDNA adducts. When the HPLC profile of these DNA adducts was compared to the profiles of the N02-PAH standards, neither 1-NP, which is present in the highest concentration in this diesel sample, the dinitropyrenes, nor 3-NF produced adduct profiles with an identical chromatographic mobility. Although additional studies will be needed to identify definitively these diesel-derived DNA adducts, this study demonstrates the utility of the 32P-postlabelingassay coupled to HPLC for the separation and tentative identification of both PAH- and N02-PAHDNA adducts derived from exposure to complex environmental mixtures. Acknowledgment. The authors wish to express their sincere thanks to Drs. Stephen Hecht and Shantu Amin for the generous gift of the 5- and 6-MC-1,2-diol-3,4epoxide standards, to Dr. David H. Phillips for the internal UV standard (cis-9,10-dihydroxy-9,lO-dihydrophenanthrene), and to Dr. Eric Weyand for the benzoG1fluoranthene, benzo[kl fluoranthene, and indenopyrene 1,2-oxide DNA adduct standards.
Chem. Res. Toxicol., Vol. 7, No. 4, 1994 509
References (1) Holberg, B., and Ahlborg, U., Eds. (1983)Proceedings of symposium: mutagenicity and carcinogenicity of air pollutants.Environ. Health Perspect. 47,l-339. (2) h e y , J., Zielineka, B., Atkinson, R., and Winer, A. M. (1987) Polycyclic aromatic hydrocarbon and nitroarene concentrations in ambient air during awintertime high-NO.episode in the Loa Angela basin. Atmos. Environ. 21, 1437-1444. (3) IARC (1989) International Agency for Research on Cancer. Monographon theEvaluationof the CarcinogenicRiskof Chemicals to Humans, Vol. 46: Diesel and Gasoline Engine Exhausts and Some Nitroarenes, IARC, Lyon. (4) Nishioka, M. G., Pederson, B., and Lewtas, J. (1982)Comparison of nitro-PNA content and mutagenicity of diesel emissions. In Polynuclear Aromatic Hydrocarbons: Physical and Biological Chemistry (Cooke, M., Dennis, A. J., and Fisher, G. L., Eds.) pp 603-613,Battelle Press, Columbus, OH. (5) Pederson, T. C., and Si&, J. S. (1986)The role of nitroaromatic compounds in the direct action mutagenicity of diesel particulate extracts. J. Appl. Toxicol. 1, 54-60. (6) Lewtas, J.,and Nishioka, M. G. (1991)Nitroarenes: Their detection, mutagenicity in the environment. In Nitroarene Occurrences, Metabolism and Biological Impact (Howard, R. C., Hecht, S. S., and Beland, F. A., Eds.) pp 61-72,Plenum Press, New York. (7) Gibson, T. L. (1982) Nitroderivatives of polynuclear aromatic hydrocarbons in airborne and source particulate matter. Atmos. Environ. 16,2037-2040. (8) Miller, E. C. (1978)Some current perspectives on chemical carcinogensin humans and experimental animels: Presidential Address. Cancer Res. 38, 1479-1496. (9) Miller, E. C. and Miller, J. A. (1981)Searches for ultimate chemical carcinogensand their reaction with cellular macromolecules. Cancer 47,2327-2345. (10) Perera, F. P. (1988)The significance of DNA and protein adducts in human biomonitoring studies. Mutat. Res. 205,225-269. (11)Beach, A. C., and Gupta, R. C. (1992)Human biomonitoring and the 32P-postlabeling assay. Carcinogenesis 13,1053-1074. (12) Reddy, M. V., and Randerath, K. (1986)Nuclease P1-mediated enhancement of sensitivity of s2P-postlabellingtest for structurally diverse DNA adducts. Carcinogenesis 7, 1543-1551. (13)Watson, W. P. (1987)Post-radiolabelling for detecting DNA damage. Mutagenesis 2, 319-331. (14) Dietrich, M. W., Hopkins, W. E., 11, Asbury, K. J., and Ridley, W. P. (1987)Liquid chromatographic characterization of the deoxyribonucleosides-5’-phosphatesand deoxyribonucleoside3’,5’-bisphosphates obtained by S2P-postlabeling.Chromatographia 24,545-551. (15) Weyand, E. H., Rice, J. E., and LaVoie, E. J. (1987)alp-Postlabeling analysis of DNA adducts from non-alternant PAH using thin-layer and high performance liquid chromatography. Cancer Lett. 37,257266. (16) Levy, G. N. and Weber, W. W. (1988) HPLC analysis of 32Ppostlabeled DNA-2-aminofluorene adducts. In Carcinogenic and Mutagenic Responses to Aromatic Amines and Nitroarenes (King, C. M., Romano, L. J., and Scheutzle D., Eds.) pp 283-287,Elsevier Science Publishing Co., New York. (17) Gorelick, N. J., and Wogan, G. N. (1989) Fluoranthene-DNA adducts: identification and quantification by an HPLCJZPpostlabeling method. Carcinogenesis 10,1567-1577. (18) Amin, S., Misra, B., Desai, D., Huie, K., and Hecht. S. S. (1989) Chromatographicconditionsfor separation of 3T-labeled phosphates of major polynuclear aromatic hydrocarbon-deoxyribonucleoside adducts. Carcinogenesis 10, 1971-1974. (19) Pfau, W., and Phillips, D. H. (1991)Improved reversed-phase highperformance liquid chromatographic separation of s*P-labelled nucleoside 3’,5’-bisphosphate adducts or polycyclic aromatic hydrocarbons. J. Chromatogr. 570,65-76. (20) Pfau, W., Hughes, N. C., Grover, P. L., and Phillips, D. H. (1992) HPLC separation of 32P-postlabelled benzo[blfluoranthene-DNA adducts. Cancer Lett. 65,159-167. (21) Hughes, N. C., and Phillips, D. H. (1993)a2P-postlabellinganalysis of the covalent binding of benzo[ghilperylene to DNA in vivo and in vitro. Carcinogenesis 14,127-133. (22) Hughes, N. C., Pfau, W., Hewer, A., Jacob, J., Grimmer, G., and Phillips, D. H. (1993) Covalent binding of polycyclic aromatic hydrocarbon components of coal tar to DNA in mouse skin. Carcinogenesis 14, 135-144. (23) Weyand, E. H., Bryla, P., Wu, Y., He, Z.-M., and LaVoie, E. J. (1993) Detection of the major DNA adducts of benzoGlfluoranthene in mouse skin: Nonclassicaldihydrodiol epoxides.Chem. Res. Toxicol. 6,117-124. (24) Weyand, E.H., Cai, Z.-W., Wu,Y., Rice, J. E., He, Z.-M., andLaVoie, E. J. (1993) Detection of the major DNA adducts of benzo[blfluoranthene in mouse skin: role of phenolic dihydrodiole. Chem. Res. Toxicol. 6,568-577.
510 Chem. Res. Toxicol., Vol. 7, No. 4, 1994 Gupta, R. C., Reddy, M. V., and Randerath, K. (1982)82p-postlabeling analysis of non-radioactive aromatic carcinogen DNA adducts. Carcinogenesis 3,1081-1092. Le-, J., Bradow,R. L., Jungers,R. H., Harris,B. D.,and Zweiinger, R. B. (1981)Mutagenic and carcinogenicpotency of extracts of diesel and related environmental emiseions: study design, sample generation, collection and preparation. Environ. Znt. 5,383-387. Gallagher, J. E.,Kohan,M. J., George, M. H., and Lewtas, J. (1991) Improvement in the diagnostic potential of *T-postlabelinganalysis demonstrated by the selective formation and comparative analysis of nitrated-PAH-derived adducts arisingfrom dieaelparticle extracts. Carcinogenesis 12, 1685-1691. Howard, P. C., and Beland, F. A. (1982)Xanthineoxidaee catalyzed binding of 1-nitropyrenetoDNA. Biochem. Biophys. Res. Commun. 104,727-732. Maron, D. M., and Ames, B. N. (1983) Revised methods of the Salmonella mutagenicity test. Mutot. Res. 113, 173-213.
King et al. (30)Gupta,R. C. (1984)Nonrandom bindingofthe carcinogenN-hydroxy2-acetylaminofluorene to repetitive sequences in rat liver DNA in vivo. R o e . Natl. Acod. Sci. U.S.A.81, 6943-6947. (31) Gupta,R. C. (1985)EnhancedsensitivityofagP-postlabelinganalysis of aromatic carcinogen-DNA adducts. Cancer Res. 45,5656-5662. (32) Gupta, R. C., and Early, K. (1988) SZP-Postlabelingassay commative recoveries of structurally diverse DNA adducts in the various enhancement procedures. Carcinogenesis 9, 1687-1693. (33) Gallagher, J. E., Jackson, M. A., George, M. H., Lewtas, J., and Robertson, I. C. G.(1989) Differences in detection of DNA adducts in the apP-postlabeling away aftar either 1-butanol or nuclease P1 treatment. Cancer Lett. 45, 5656-5662. (34) Gallagher,J. E., Jackson, M. A., George,M. H., and Lewtas,J. (1990) Dose related differences in DNA adduct level in rodent tissues following skin application of complex mixtures from air pollution sources. Carcinogenesis 11, 63-68.
