trans-3,4-dihydroxy-anti-1,2-epoxy-1,2,3,4-tetrahydrodibenz[a,j]acridine involvement in dibenz[a,j]acridine DNA adduct formation in mouse skin consistent with Ha-ras mutation patterns in tumors

Article abstract of DOI:10.1021/tx010014y

Dibenz[a,j]acridine (DBA), is a N-heteropolycyclic aromatic environmental carcinogen found in complex combustion mixtures. The major route of DBA metabolic activation is reportedly through the trans-3,4-dihydroxy-3,4-dihydroDBA (DBA-3,4-DHD). The present studies were undertaken to determine the role of trans-3,4-dihydroxy-anti-1,2-epoxy-1,2,3,4-tetrahydroDBA (DBADE) in DBA activation pathway(s), the DNA bases involved in the binding of DBA to DNA, and whether the adducts produced are consistent with the mutation pattern in the Haras gene. DBA (300 μg) or 50 μg synthesized (±)-DBADE was applied to the back of female Hsd:ICR(Br) mice. The mice were sacrificed 48 h later, and skin DNA was isolated, hydrolyzed, and analyzed with ³²P-postlabeling. Of the four adducts produced in vivo, adduct 1 was the major adduct for DBA (>50%) and adduct 2 was the major adduct for DBADE (89%). After the reaction of (±)-DBADE with purine nucleotides or calf thymus (CT) DNA in vitro, 100% of the DBADE-2′-dAMP adducts and 94% of DBADE-CT DNA adducts were chromatographically identical on TLC with adduct 2 and 86% of the DBADE-2′-dGMP adducts were chromatographically consistent with adduct 1 by ³²P-postlabeling. Papillomas were induced on the backs of mice by a single application of 0.2 μmol of DBA followed by twice-weekly application of 12-o-tetra-decanoylphorbol-13-acetate (TPA, 2 μg) for 24-26 weeks. Skin carcinomas were induced by twice weekly applications of DBA (0.1 μmol) on the backs of mice. A to T and G to T transversions were found in codons 12, 13, and 61 of the Ha-ras gene in the treated mouse skin carcinoma and papilloma DNA. The mutational spectra in the Ha-ras gene are consistent with the DNA binding of DBA to dG or dA in vivo. Thus, this research has indicated that DBADE plays an important role in DBA metabolic activation and DNA binding in mouse skin, and an alternative pathway through a bis-dihydrodiol-epoxide of DBA may also be involved.

Full text of DOI:10.1021/tx010014y

Chem. Res. Toxicol. 2001, 14, 871-878
871
tr a n s-3,4-Dih yd r oxy-a n ti-1,2-ep oxy-1,2,3,4-tetr a h yd r od i-
ben z[a ,j]a cr id in e In volvem en t in Diben z[a ,j]a cr id in e
DNA Ad d u ct F or m a tion in Mou se Sk in Con sisten t w ith
Ha -r a s Mu ta tion P a tter n s in Tu m or s
Weiling Xue, J oanne Schneider, Kent Mitchell, Marlene J aeger,
†
Vajira Nanayakkara, Glenn Talaska, and David Warshawsky*
Department of Environmental Health, University of Cincinnati Medical Center, P.O. Box 670056,
Cincinnati, Ohio 45267-0056
Received J anuary 19, 2001
Dibenz[a,j]acridine (DBA), is a N-heteropolycyclic aromatic environmental carcinogen found
in complex combustion mixtures. The major route of DBA metabolic activation is reportedly
through the trans-3,4-dihydroxy-3,4-dihydroDBA (DBA-3,4-DHD). The present studies were
undertaken to determine the role of trans-3,4-dihydroxy-anti-1,2-epoxy-1,2,3,4-tetrahydroDBA
(DBADE) in DBA activation pathway(s), the DNA bases involved in the binding of DBA to
DNA, and whether the adducts produced are consistent with the mutation pattern in the Ha-
ras gene. DBA (300 µg) or 50 µg synthesized (()-DBADE was applied to the back of female
Hsd:ICR(Br) mice. The mice were sacrificed 48 h later, and skin DNA was isolated, hydrolyzed,
and analyzed with 32P-postlabeling. Of the four adducts produced in vivo, adduct 1 was the
major adduct for DBA (>50%) and adduct 2 was the major adduct for DBADE (89%). After the
reaction of (()-DBADE with purine nucleotides or calf thymus (CT) DNA in vitro, 100% of the
DBADE-2′-dAMP adducts and 94% of DBADE-CT DNA adducts were chromatographically
identical on TLC with adduct 2 and 86% of the DBADE-2′-dGMP adducts were chromato-
3
2
graphically consistent with adduct 1 by P-postlabeling. Papillomas were induced on the backs
of mice by a single application of 0.2 µmol of DBA followed by twice-weekly application of
1
2-o-tetra-decanoylphorbol-13-acetate (TPA, 2 µg) for 24-26 weeks. Skin carcinomas were
induced by twice weekly applications of DBA (0.1 µmol) on the backs of mice. A to T and G to
T transversions were found in codons 12, 13, and 61 of the Ha-ras gene in the treated mouse
skin carcinoma and papilloma DNA. The mutational spectra in the Ha-ras gene are consistent
with the DNA binding of DBA to dG or dA in vivo. Thus, this research has indicated that
DBADE plays an important role in DBA metabolic activation and DNA binding in mouse skin,
and an alternative pathway through a bis-dihydrodiol-epoxide of DBA may also be involved.
In tr od u ction
istration (6), a moderate mouse skin tumor initiator (7)
and a skin carcinogen in mice following topical applica-
tion (8). In comparison with DBC and benzo[a]pyrene
1
N-Heteropolycyclic aromatic compounds (NHA) are
formed from the incomplete combustion of organic mate-
rial They are widely distributed in the environment and
are found in tobacco smoke condensate, synthetic fuels,
automobile engine exhaust, and as effluent from the
combustion of coal (1, 2). Many of these NHA are
carcinogenic including dibenz[a,j]acridine (DBA, Figure
), 7H-dibenzo[c,g]carbazole (DBC), dibenz[c,h]acridine,
and dibenz[a,h]acridine, etc. (1-5). DBA has been found
to be a lung carcinogen in mice by subcutaneous admin-
(BaP) in Hsd:ICR(Br) mice, DBA produced 27 carcinomas
in 50 mice relative to 42 and 49 carcinomas for DBC and
BaP, respectively (8). Similar skin tumor data were ob-
tained for DBA, DBC, and BaP in the C3H mouse (9).
Like many mutagenic/carcinogenic polycyclic aromatic
hydrocarbons (PAH), NHA require metabolic activation
to electrophilic species in order to bind covalently with
cellular components and to exert their carcinogenic effect.
Consistent with the bay-region theory of PAH carcino-
genesis (10), DBA has been reported to metabolize to a
series of dihydrodiols, phenols and epoxides of which the
trans-3,4-dihydroxy-3,4-dihydrodibenz[a,j]acridine (DBA-
3,4-DHD, Figure 1) is the major product (11-13). These
metabolites are further activated, and trans-3,4-dihy-
droxy-anti-1,2-epoxy-1,2,3,4-tetra-hydroDBA (DBADE,
Figure 1) is predicted to be an ultimate active metabolite
(12).
1
*
To whom correspondence should be addressed.
Department of Molecular Genetics, Biochemistry and Microbiology,
University of Cincinnati.
†
1
Abbreviations: NHA, N-heteropolycyclic aromatic compounds;
DBA, dibenz[a,j]acridine; DBC, dibenzo[c,g]carbazole; PAH, polycyclic
aromatic hydrocarbon, BaP, benzo[a]pyrene; DB[a,h]A, dibenz[a,h]-
anthracene; DB[a,j]A, dibenz[a,j]anthracene; DB[a,l]P, dibenzo[a,l]-
pyrene; DBA-3,4-DHD, trans-3,4-dihydroxy-3,4-dihydrodibenz[a,j]-
acridine; DBA-5,6-DHD, trans-5,6-dihydroxy-5,6-dihydroDBA; DBADE,
trans-3,4-dihydroxy-anti-1,2-epoxy-1,2,3,4-tetrahydro-DBA; (()-D-
BADE, synthetic racemic DBADE; DBA-3,4,8,9-bis-DHD-1,2-epoxide,
,2-epoxy-3,4,8,9-tetrahydroxy-1,2,3,4,8,9-hexahydroDBA; dG, 2′-deox-
yguanosine; dA, 2′-deoxyadenosine; TPA, 12-o-tetra-decanoylphorbol-
3-acetate; EH, epoxide hydrolase; MALDI-TOF-MS, matrix-assisted
laser desorption ionization-time-of-flight mass spectrometry.
In previous studies, DBA and its metabolites were
1
topically applied to the backs of mice (14). DNA was
3
2
isolated and DNA adducts were analyzed by P-postla-
1
beling. DBA produced two distinct adducts (1 and 2) in
1
0.1021/tx010014y CCC: $20.00 © 2001 American Chemical Society
Published on Web 05/26/2001

8
72 Chem. Res. Toxicol., Vol. 14, No. 7, 2001
Xue et al.
pyruvate (type II), sodium adenosine diphosphate, ribonuclease
(Rnase T1, grade IV, 400 000 units/mL), and RNaseA (type IIIa,
7
5 units/mg). Boehringer-Mannheim Biochemicals (Indianapolis,
IN) supplied: calf spleen exonuclease (phosphodiesterase 2
untis/mg), proteinase K (20 untis/mg), L-glycerol-3-phosphate
and â-nicotinamide adenine dinucleotide (NAD grade I, 100%).
Poly-ethyleneimine-cellulose (PEI) thin-layer plastic-backed
sheets were purchased from Alltech Inc. (Waukeegan, IL).
Polynucleotide kinase was obtained from U.S. Biochemical Co.
3
2
(
3 4
Cleveland, OH). Carrier-free [ P]H PO was obtained from ICN
3
2
Biochemicals (Irvine, CA). γ-[ P]ATP was synthesized as
described previously (18). Sep-Pak cartridges were from Waters
(Milford, MA). Fisher Scientific (Pittsburgh, PA) was the source
of all other chemicals used in this work.
Ca u tion : DBA and metabolites are potentially hazardous and
were handled in accordance with NIH Guidelines.
In str u m en ta l An a lysis. HPLC was performed on a Waters
system equipped with model 501 pumps, a WISP 710B autoin-
jector and 484 tunable absorbance detector controlled with the
Millenium data system. Fluorescence excitation, emission and
synchronous spectra in methanol solution were recorded on a
1
Perkin-Elmer MPF-66 Fluorescence Spectrophotometer.
H
NMR spectra were recorded on a Bruker NMR/300 AF spec-
trometer in deuterated dimethyl sulfoxide, and chemical shifts
reported with tetramethylsilane (TMS) as the internal standard
(0 ppm). The mass spectral data of synthesized standards were
recorded on a Kratos MS-890 high-resolution MS (HR-MS)
operated with a DS-90 data system based on Data General
Nova/4 eclipse. A matrix-assisted laser desorption ionization
time-of-flight (MALDI-TOF) MS was employed to obtain MS
data of adducts. An aliquot (0.5 µL) of the matrix solution (R-
cyano-4-hydroxy cinnamic acid in 30% acetonitrile/1% trif-
louroacetic acid) was mixed with the sample in a 0.5 mL vial,
and the resultant mixture was deposited onto the MALDI probe
tip. Multiple sample probe tip was then introduced into the
MALDI/TOF mass spectrometer, and the solvent was allowed
to evaporate under the vacuum conditions. MALDI MS was
obtained on a TOF instrument (VG TOF SPEC-SE model) under
reflectron mode as well as the linear mode. Background pressure
F igu r e 1. Chemical structure of DBA and metabolites and
proposed pathways of DBA metabolicactivation. Only one enan-
tiomer is shown, for each chiral compound.
skin. The same two adducts were seen when DBA-3,4-
DHD was topically applied. The total adduct levels
elicited by the DBA-3,4-DHD was twice that of DBA. In
addition, two different DNA adducts (3 and 4) were seen
in mouse skin when the trans-5,6-dihydroxy-5,6-dihy-
droDBA (DBA-5,6-DHD, Figure 1) was topically applied.
-
7
within the instrument was better than 3 × 10
Torr as
measured by an ion gauge located below the source. The laser
used for ionization was nitrogen laser at 337 nm with a pulse
width of 6 ns and 180 µJ /pulse. Each laser shot produced a full
mass spectrum. The spectra were averages of 20-30 laser shots.
The ion acceleration potential was set at 25 kV. Ion detection
and signal amplification were achieved using micro channel
plate detector-discrete dynode multiplier assembly. The data
were processed and stored on a DEC-300 a-work station.
3
2
Using more sensitive nuclease P
1
P-postlabeling, all four
DNA adducts were detected following topical administra-
tion of the DBA-3,4-DHD. These data suggested that the
route of DBA activation was through the DBA-3,4-DHD
and formation of the DBADE and another pathway
involving a common secondary metabolite of both DBA-
3
,4-DHD and DBA-5,6-DHD, bis-DHD-epoxide (Figure 1)
similar to dibenz[a,h]anthracene (DB[a,h]A; 15, 16), was
proposed (14). Therefore, we undertook studies to deter-
mine (1) the role of DBADE in metabolic activation
pathways and DNA binding of DBA, (2) the nucleotide
bases involved in binding of DBADE to DNA, and (3)
whether the formation of the DNA adducts are consistent
with the mutation pattern produced in the Ha-ras gene.
In this study, the DBADE used is the chemically syn-
thesized racemic mixture, (()-DBADE.
Syn th esis of (()-DBADE. Following a known method (19)
with modifications (()-DBADE (>95% purity) was synthesized
and the synthesis should be performed just prior to use due to
the instability of the compound. DBA-3,4-DHD (2.1 mg, previ-
ously synthesized in this laboratory according to ref 20), was
dissolved in 3 mL of anhydrous tetrahydrofuran (THF) and 1.2
mg of m-chloro-peroxybenzoic acid (mCPBA, Aldrich Chem. Co.,
Milwaukee, WI) was added. The mixture was stirred at ambient
temperature in the dark, under dry nitrogen for 2 h and an
additional 1.2 mg of mCPBA was added. The reaction was
monitored using HPLC and allowed to run for additional two
to 4 h. Then, the reaction mixture was diluted with 10 mL of
ethyl acetate and washed with 2% triethylamine/water (2 × 5
mL), saturated sodium chloride aqueous solution (2 × 5 mL),
and then dried over sodium carbonate. After evaporation of the
solvent, the product was isolated using HPLC on a 7.8 × 300
mm Waters µPorasil column with an isocratic eluting program
of 7/3 hexane/mixture of ethyl acetate, methanol and triethy-
lamine (3/1/0.02) at a flow rate of 5 mL/min. Fluorescence
Exp er im en ta l P r oced u r es
Ma ter ia ls. DBA (> 99% purity) was synthesized following
the published procedures (17) and recrystallized from benzene,
mp 221-222 °C. The following reagents and enzymes were
purchased from Sigma Chemical Co. (St. Louis, MO): urea,
dithiothreitol, calf thymus DNA, ethylenediamine tetraacetic
acid (EDTA), 2′-deoxyadenosine-3′-monophosphate (3′-dAMP),
2
′-deoxyadenosine-5′-monophosphate (5′-dAMP), 2′-deoxygua-
spectra are shown in Figure 2; HR-MS: 329.1008, calcd
1
nosine-3′-monophosphate (3′-dGMP), 2′-deoxyguanosine-5′-
monophosphate (5′-dGMP), 12-o-tetradecanoyl-phorbol-13-ac-
etate (TPA), phosphatase alkaline (bacteria, type III, 82 units/
mg), micrococcal endonuclease (grade VI, 100 units/mg), sodium
329.1052 for C21
H
15NO
3
; H NMR: 3.83(d, 1H, H2), 3.91 (dd,
1H, H3), 4.55 (dd, 1H, H4), 5.50 (d, 1H, H1), 5.76 (d, 1H, OH3),
5.97 (d, 1H, OH4), 7.8-8.2 (aromatic H), 9.37 (d, 1H, H13), 10.35
(s, 1H, H14).

Dibenz[a,j]acridine DNA Adduction and Ha-ras Mutation
Chem. Res. Toxicol., Vol. 14, No. 7, 2001 873
plastic cages under controlled temperature, humidity and day-
night cycle. Animal chow and water were given ad libitum. DBA
(
300 µg) in 50 µL of acetone or acetone alone were applied to
the shaved interscapular dorsal skin. Animals were sacrificed
8 h after dosing. Skin, liver, and lung were removed and were
4
frozen in liquid nitrogen, stored at -80 °C until analyzed. Nine
animals were treated once with 50 µg of (()-DBADE in 50 µl
acetone or vehicle alone and three animals as a group were
sacrificed at 4, 8, and 24 h as described above.
Extr a ction of DNA fr om Tissu es a n d 32P -P ostla belin g
An a lysis. DNA was extracted from the various tissues using
techniques described earlier (25) and DNA (2 µg at 0.5 µg/µL)
was hydrolyzed to 3′-phosphodeoxynucleotides as described
earlier (14). DBA-DNA adduct levels are maximal when
hydrolyzed at pH 6 for 6 h (26). DNA adducts were analyzed
using a 32P-postlabeling assay as described earlier (27-29).
Nucleotide digests were labeled under adduct intensification
3
2
conditions employing a limiting amount of [ P]ATP relative to
DNA (25, 30). The intensification factor for adducts 1 and 2
under these labeling conditions were 71.7 ( 8.4- and 57.0 (
F igu r e 2. Comparison of fluorescence excitation, emission and
synchronous spectra of synthetic (()-DBADE in methanol (s)
and DBADE-dAMP(- - -) made from reaction of (()-DBADE
with 5′-dAMP, in vitro: ex. scan from wavelength 240 to 380
nm with em. set at 398 nm; em. scan from 380 to 500 nm with
ex. set at 281 nm; synchronous scan from 300 to 450 nm with
the delta wavelength, ∆λ ) 10 nm, and the peak shown at 388
nm.
7
.1-fold, respectively. The intensification factors for adducts 3
and 4 were not determined because the levels were too low for
reliable determination. For TLC separation, 2.6 M lithium
formate, 6 M urea, pH 3.6, was used as D3 solvent, and the D4
solvent was 0.75 M lithium chloride, 0.45 M Tris, 8.2 M urea,
pH 8, as previously described (14). Unadducted nucleotides were
labeled and analyzed on TLC, and adduct spots and background
areas were excised from the TLC plates and the radioactivity
in each adduct was determined using Cerenkov counting. The
level of adduction and relative adduct labeling (〈RAL〉) values
were determined as follows:
Ad d u ction of DBADE w ith DNA or Nu cleotid es in Vitr o.
(
()-DBADE (∼0.1 mg) was dissolved in 0.1 mL of THF mixed
with a solution of either 5′-dGMP (5 mg), 5′-dAMP (5 mg), 3′-
dGMP (1.5 mg), or 3′-dAMP (1.5 mg) in 1 mL of 0.1 M Tris (pH
7
.0) or calf thymus DNA (1.5 mg) in 0.3 mL of water and 0.7
mL of 0.1M Tris (pH 7.0) and then incubated at 37 °C in the
cpmadducts
× 10
cpmunadducted nucleotides
dark for 8-16 h. DBADE adduct formation with the 3′-
7
3
2
〈RAL〉 )
phosphates or DNA were analyzed by P-postlabeling, while
adducts produced by reaction with 5′-phosphates were further
isolated and purified for chemical analysis.
These values represent the number of adducts in 107 unad-
ducted nucleotides under carrier-free conditions and should be
divided by the intensification factors to estimate the unbiased
adduct levels.
Isola tion a n d An a lysis of Nu cleotid e Ad d u cts. Unmodi-
fied nucleotides were removed from the 5′-phosphates reaction
mixtures through HPLC using a Whatman Partisil 10 ODS-2
column (4.6 × 250 mm) and eluted with 100% water for 12 min
and 100% methanol for 18 min with a flow rate of 1 mL/min.
Fractions from 20 to 26 min were collected and evaporated to
dryness under nitrogen at 30 °C. The residues from DBADE-
dAMP adduction were redissolved in 1 mL of 0.1 M Tris (pH,
Tu m or In d u ction a n d Ha -r a s An a lysis. Papillomas were
induced by a single topical application of DBA (0.2 µmol) to the
shaved backs of 12 female Hsd:ICR(Br) mice followed by twice
weekly applications of TPA (2 µg) for 24-26 weeks. Six TPA
and six no treatment animals were used as controls. Carcinomas
were induced in eight female Hsd:ICR(Br) mice by twice weekly
topical applications of DBA (0.1 µmol) until tumor development.
Six acetone control animals and six no treatment animals were
used as controls. Twelve papillomas and eight carcinomas were
removed and DNA extracted by a nonorganic method (Gentra
Systems). The DNA was screened for Ha-ras codon 12, 13, and
9
.0) and incubated with 0.1 mL of phosphotase alkaline solution
(25 units/ml) in 0.02 M Tris (pH, 7.0) at 37 °C for 24 h to convert
the nucleotide adducts to nucleoside adducts (21, 22). The
digested mixtures were purified on a preconditioned Sep-Pak
cartridge as previously described (23). DBADE-dA adducts
were separated by HPLC using a gradient program: 40%
methanol/water for 4 min, increasing to 60% methanol over 15
min, isocratic for 6 min, increasing to 100% methanol over 5
min and maintained for 8 min at 100% methanol, with a flow
rate of 1 mL/min. To isolate adducts on HPLC, several runs were
repeated and six fractions were collected. After fluorescence
spectroscopic analyses, each fraction was evaporated to dryness
under nitrogen and then acetylated using acetic anhydride/
pyridine procedures described previously (24) to give the
6
1 mutations using a modified enriched polymerase chain
reaction (EPCR) method (31). This method, which eliminates
interference of K-ras, N-ras, and pseudogenes by using an intron
based primer on the 3′ end, is a four-step reaction that includes
initial amplification, intermediate digestion, secondary ampli-
fication and final digestion. The final digestion products were
resolved on 3% MetaPhor agarose-1X TBE gels. Positive bands
were isolated, reamplified, purified (Wizard PCR Preps, Prome-
ga) and reversed sequenced (fmol Sequencing System, Promega).
Sequencing autoradiograms were visualized on Kodak Xomat
AR film.
R
completely acetylated derivatives. Fractions at t 29.2 and 29.8
min were identified as the DBADE-adducts (in their pentaacetyl
+
form) using MALDI-TOF MS, (M + H) 791 for C41
38 6 11
H N O .
For DBADE-dGMP adduction, the reaction mixture, after
removal of unmodified nucleotides and evaporation of solvent,
was redissolved in 0.3 mL of 30% methanol/water and directly
separated on HPLC with the same solvent gradient program
and seven fractions were collected for analysis. MALDI-TOF MS
Resu lts
It has been proposed that DBADE is an ultimate active
metabolite of DBA (12) responsible for DNA binding and
carcinogenicity. In this study, the covalent adducts of (()-
DBADE with dAMP and dGMP were isolated from in
vitro reaction and identified using fluorescence spectral
and mass spectral analysis. For DBADE-5′-dAMP ad-
ducts, the phosphate group was enzymatically hydrolyzed
R
showed that the fraction at t 32.2 min. was the DBADE-dGMP
+
adduct, (M + H) 677 for C31
29 6
H N O10P.
An im a l Tr ea tm en t w ith DBA or DBADE. Female Hsd:
ICR (Br) mice were purchased from Harlan Sprague Dawley
(
Indianapolis, IN) at 4-6 weeks of age and allowed to ac-
climatize for two weeks. Animals were housed individually in

8
74 Chem. Res. Toxicol., Vol. 14, No. 7, 2001
Xue et al.
and the ensuing dA adduct mixtures were separated by
HPLC to remove the major byproducts1,2,3,4-tetrol of
DBA. Fluorescence spectra of each fraction were recorded
to confirm the DBADE moiety of adducts (Figure 2). To
facilitate the chemical stability and vaporizability, the
adduct fractions were then acetylated for MS analysis.
Due to the low yield of the adduction and the small scale
of the reaction, the acetylated adducts were analyzed
using MALDI-TOF MS. That confirmed the formation of
the adduct(s) similar to other PAH-diol-epoxides (21), the
binding was through the 1,2-epoxide-ring opening in
DBADE and the resultant benzylic carbon, i.e., C1, to
link with nucleophilic site(s) of the dA residue to form
the covalent bonded adduct(s), most likely, 2,3,4-trihy-
droxy-1,2,3,4-tetrahydroDBA-1-dA. However, the binding
site of Ade and the stereo-preference of the ring opening
of the epoxide need to be further characterized by NMR
although reportedly, the amino group of Ade and Gua is
the site to be linked by dihydrodiol-epoxides for PAHs
(32). By virtue of the MALDI-TOF MS (33), the mass
spectral data of the DBADE-5′-dGMP adduct(s) were
obtained without hydrolyzing the phosphate group and
acetylating the hydroxyls in adducts that would be
required by the conventional electron impact MS (21).
The MALDI-TOF mass spectral data showed that the
chemical identity of DBADE-5′-dGMP adduct(s) was
similar to the dA adduct(s), through the covalent linkage
of C1 in DBADE with dG residue. The reaction rate of
DBADE with dG was found to be slower and at lower
yield than dA adduction.
To study DNA adducts in vivo, five mice were treated
topically with 300 µg of DBA, a 3-fold higher dose than
the earlier study (100 µg, ref 14), and sacrificed 48 h later,
and skin DNA was isolated and analyzed using 32P-
postlabeling. Four adducts were observed in the skin
DNA (Figure 3A), and the relative percentage level of the
four adducts is reported in Table 1. At least half of the
total adduct level was found in adduct 1. The DNA adduct
patterns of (()-DBADE in vivo were determined by
topically applying 50 µg of (()-DBADE to nine mice, and
three animals each were sacrificed 4, 8, and 24 h later.
3
2
Then, the DNA was isolated and P-postlabled with the
same techniques as those used for DBA assays. The
adduct data are reported in Table 1 and Figure 3B. The
majority of the adduction of (()-DBADE to DNA was
found in adduct 2 (89%), which was different from the
results observed for DBA in vivo. The highest level of the
adduct 2 was observed at 8 h (Figure 4). To verify the
DNA bases involved in binding in vivo, 3′-dAMP, 3′-
dGMP, and CT-DNA were incubated in vitro with (()-
DBADE individually, and each reaction mixture was
directly analyzed using 32P-postlabeling (Table 1). Chro-
matographic comparison of the P-postlabeling TLC
maps with those obtained from in vivo experiments
indicated that adduct 1 was DBADE bound to dG (86%),
adduct 2 was DBADE bound to dA (100%), and for CT
DNA, DBADE was almost exclusively bound to dA (94%)
forming adduct 2. Adducts 3 and 4 were not seen in any
of these three DBADE reactions. It would appear from
these data that there is another pathway other than
DBADE in the activation of DBA to produce metabolite-
F igu r e 3. Autoradiogram resulting from 32P-postlabeling of
mouse skin DNA adducts formed with a single topical applica-
tion of (A) 300 µg of DBA and (B) 50 µg of (()-DBADE.
(Table 2 and Figures 5 and 6). Four of eight carcinomas,
produced in eight mice, were found to have mutations in
the Ha-ras gene. One mutation was an Af T transver-
sion in the second base of codon 61 similar to DBC and
DMBA (34, 35), a second mutation was a G f T
transversion in the second base of codon 12 similar to
BaP (36) and a third mutation was a G f T transversion
in the first base of codon 13. These three mutations are
in accordance with the DBA-DNA binding studies, which
indicate that both dG and dA are involved in binding of
DBA. The fourth mutation was a G f T transversion in
the first base of codon 13 that was consistent with a
spontaneous mutation (37). The acetone control had a
spontaneous skin carcinoma with a mutation in the first
base of the 13th codon, a G f C transversion (37). Six of
12 papillomas, produced in 12 mice were found to have
3
2
(s) that bind to DNA to form adducts 3 and 4.
Mutations in the Ha-ras gene of DBA-induced mouse
skin tumors produced by two different carcinogenesis
protocols were determined using EPCR screening method
(31, 34) and confirmed by DNA sequencing analysis

Dibenz[a,j]acridine DNA Adduction and Ha-ras Mutation
Chem. Res. Toxicol., Vol. 14, No. 7, 2001 875
Ta ble 1. Rela tive P er cen ta ge Levels of DBA-DNA Ad d u cts in Vitr o a n d in Vivo
adduct 1
adduct 2
adduction
DBA in vivo
DBADE in vivo
DBADE-DNA in vitro
DBADE-dGp in vitro
DBADE-dAp in vitro
(dG adduct)
(dA adduct)
adduct 3
adduct 4
50.3 ( 13.2
4.7 ( 2.5
89.3 ( 9.5
94.0 ( 6.7
14f
24.6 ( 4.1
4.6 ( 4.6
20.4 ( 11.5
4.4 ( 5.5
b
1.7 ( 1.0
c
6
0
0
0
0
0
0
d
86.0 ( 6.0
e
0
100
a
b
DBA, 300 µg applied topically to backs of mice and sacrificed 48 h later. (()-DBADE, 50 µg applied topically to backs of mice and
sacrificed 8 h later. (()-DBADE incubated with calf thymus DNA for 16 h. (()-DBADE incubated with 3′-dGMP for 12 h. e (()-DBADE
incubated with 3′-dAMP for 12 h, only one adduct seen. May not be chromatographically identical to adduct 2.
c
d
f
produced DBADE from DBA in vivo would probably be
one of the factors that lead to their distinct relative
reactivity with dA and dG. The synthesis and resolution
of stereoisomers of DBADE will be important in further
investigation of the activation mechanism of DBA.
At lower dose of DBA (100 µg), adducts 1 and 2 were
almost exclusively formed (14). We now found that higher
dose of DBA (300 µg) resulted in increased levels of
adducts 3 and 4 (Table 1). It may be that at higher doses,
the primary activation pathway to form DBADE-dG
(adduct 1) and DBADE-dA (adduct 2) becomes saturated
leading to an alternative pathway to produce adducts 3
and 4 (see Figure 1). They are most likely the corre-
sponding dG and dA adducts of a common active species
from the metabolism of DBA-5,6-DHD and DBA-3,4-
DHD, the 1,2-epoxy-3,4,8,9-tetrahydroxy-1,2,3,4,8,9-
hexahydroDBA (DBA-3,4,8,9-bis-DHD-1,2-epoxide, Fig-
ure 1), because (1) their relative positions on TLC are
very similar to adducts 1 and 2 except that they are more
polar, (2) they are not formed to an appreciable extent
when DBADE is applied in vivo or in vitro, suggesting
that most DBADE binding with DNA occurs before the
DBA-5,6-epoxide and subsequent dihydrodiol can be
formed, and (3) the precursor of the DBA-3,4,8,9-bis-
DHD-1,2-epoxide, DBA-3,4,8,9-bis-DHD has been found
in DBA metabolism (12) and it could be formed either
through DBA-3,4-DHD or DBA-5,6-DHD by monooxyge-
nase and epoxide-hydrolase (EH) in P450. These data
support the hypothesis that bay-region 1,2-epoxides are
important in the activation of DBA to genotoxic species.
In addition, a novel mechanism for the formation of
NHA-DNA adducts, involving the bis-DHD-epoxides, is
consistent with that has been described for some PAHs,
such as DB[a,h]A (15, 16). However, definitive proof that
a bis-DHD-epoxide of this type derived from DBA binds
to DNA should await synthesis and analytical charac-
terization of this bis-DHD-epoxide and its DNA adducts
formed.
F igu r e 4. Relative adduct labeling 〈RAL〉 value of DNA adducts
formed over 4, 8, and 24 h, following a single topical application
of 50 µg of (()-DBADE to mouse skin. Black bars, adduct 1;
white-black dots, adduct 2; slanted lines, adduct 3; cross-
sectioned, adduct 4.
mutations in the Ha-ras gene. One mutation was a G fT
transversion in the second base of codon 13 similar to
BaP (36), and three mutations were A f T transversions
in the second base of codon 61 similar to 3-methylcholan-
threne (38). The remaining two mutations were C f A
transversions in the first base of codon 61 which could
be due to a G f T transversion on the non-coding strand.
These mutations, in accordance with the DBA-DNA
binding studies, indicate that both dG and dA are
involved in binding of DBA.
Discu ssion
The formation of covalent DNA adducts in vivo is
usually regarded as the initiating step in the carcinogenic
process. Identification and characterization of the DNA
adducts formed would provide useful information in
understanding the activation mechanism of carcinogens.
When DBA is applied topically in vivo, four adducts are
formed and the major adduct is adduct 1. Of the four
adducts observed, adducts 1 and 2 match chromato-
graphically the dG and dA adducts of DBADE, respec-
tively. On the other hand, (()-DBADE applied topically
in vivo or in vitro with CT DNA, results predominantly
in binding to dA (adduct 2). In addition, chemical
reactions of (()-DBADE with dAMP and dGMP also show
the preference to react with dA at a higher yield than
dG adduct. Presently, there is no clear understanding of
factors that direct dihydrodiol epoxides to dA or dG
residues in DNA. Some studies on PAH, however, have
indicated that the product distribution between the two
purine-nucleosides in DNA was dependent on the con-
figuration of the dihydrodiol epoxides and the planarity
of the hydrocarbons (39). Therefore, the different steric
structures (configurational isomers) of the chemically
synthesized racemic DBADE and the enzymatically
Mutations in papillomas and carcinomas as a result
of exposure to DBA occur in codons 12, 13, and 61 of the
Ha-ras gene. These transversions occur at the Gua and
Ade bases, which are consistent with the formation of
DNA adducts at dG and dA residues (Tables 1 and 2).
These data are similar to that reported for DB[a,l]P in
which lung tumors in the strain A/J mice following
exposure to this PAH produced a mixture of both codon
12 and codon 61 Ki-ras mutations involving the first and
second base of each codon (40). The predominant muta-
tions were G f T transversions in the first base of codon
12, A f G transitions in the second base of codon 12 and
A f T transversions in the second and third bases of
codon 61 consistent with DNA adduct profiles even
though the major adduct was identified as the product
of the DB[a,l]P-diol-epoxide with a dA residue (40). It

8
76 Chem. Res. Toxicol., Vol. 14, No. 7, 2001
Xue et al.
Ta ble 2. Resu lts of Ha -Ra s Mu ta tion by EP CR Scr een in g a n d Sequ en cin g of DBA-In d u ced /TP A-P r om oted Sk in
P a p illom a s a n d DBA-In d u ced Sk in Ca r cin om a s
compd
DBA/TPA
tumor type
papilloma
codon
13
mutations
amino acid
1-GGC f GTC
3-CAA f CTA
2-CAA f AAA
Gly f Val
Gln f Leu
Gln f Lys
6
6
1
1
acetone/TPA control
no treatment control
none
none
carcinoma
82 weeks
12
13
13
61
13
GGA f GTA
GGA f GTA
GGC f TGC
CAA f CTA
GGC f CGC
Gly f Val
Gly f Arg
Gly f CYS
Gln f Leu
Gly f Arg
8
8
7
2 weeks
1 weeks
5 weeks
acetone control
no treatment control
carcinoma
none
89 weeks
F igu r e 5. EPCR screening of DBA induced mouse skin tumor DNA for Ha-ras codon 12, 13, and 61 mutations. (A) Codon 12: lane
, negative carcinoma; lane 2, positive carcinoma; 3, negative papilloma; 4, H O; 5, negative control; 6, positive control; lane m, 25
bp marker. (B) Codon 13: lane 1, negative carcinoma; lane 2, positive carcinoma; lane 3, negative papilloma; lane 4, positive papilloma;
lane 5, H O; lane 6, negative control; lane 7, positive control; lane m, 25 bp marker. (C) Codon 61: lane 1, negative carcinoma; lane
, positive carcinoma; lane 3, negative papilloma; lane 4, positive papilloma; lane 5, H O; lane 6, negative control; lane 7, positive
1
2
2
2
2
control; lane m, 25 bp marker.
also has been shown that BaP-induced papillomas had
mostly Ha-ras mutations (70%) in codons 12 and 13 and
(43), which is consistent with major adduction at dA
residues (44). In another study, DB[a,j]A, DB[a,j]A-diol-
epoxide, and 7,14-dimethyl-DB[a,j]A all produced papil-
lomas with Ha-ras mutations in codon 61 in the second
base even though adduction of the first two compounds
appear to involve both dA and dG residues (45). It should
be noted, however, that dibenz[c,h]acridine, an NHA,
induced papillomas that showed exclusively Ha-ras codon
61 mutations although the adduction pattern is not
known (35). These data indicate that the relationship of
induced adducts of both bases with the mutation spectra
is compound specific.
2
0% mutations in codon 61. The major adduct was
identified as the product of bay-region diol-epoxide with
a dG residue (41). On the basis of the data for DBA, it
would be apparent that NHA or PAH that induce adducts
at both bases, induce both types of mutation. In contrast,
7
,12-dimethyl-benz[a]anthracene induced skin carcino-
mas had exclusively Ha-ras codon 61 mutations with
major adduction resulting extensively from the diol-
epoxide and both dG and dA residues (35, 38, 42). In
addition, 3,4-dihydroxy-1,2-epoxide-1,2,3,4-tetrahydro-
benzo[c]phenanthrene induced mouse skin tumors pro-
duced significant number of mutations in the codon 61
This work has shown that the environmental carcino-
gen DBA is activated and covalently binds to mainly dG

Dibenz[a,j]acridine DNA Adduction and Ha-ras Mutation
Chem. Res. Toxicol., Vol. 14, No. 7, 2001 877
(
Harvey, R. G., Ed) pp 63-84, American Chemical Society,
Washington, DC.
(
11) Warshawsky, D., Hollingsworth, L., Reilman, R., and Stong, D.
B. (1985) The metabolism of dibenz[a,j]-acridine in the isolated
perfused lung. Cancer Lett. 28, 317-326.
(
12) Gill, G. H., Duke, C. C., Rosario, C. A., Ryan, A. J ., and Holder,
G. M. (1986) Dibenz[a,j]acridine metabolism: identification of in
vitro products formed by liver microsomes from 3-methylcholan-
threne pretreated rats. Carcinogenesis 7, 1371-1378.
13) Gill, G. H., Duke, C. C., Ryan, A. J ., and Holder, G. M. (1987)
Dibenz[a,j]acridine: distribution of metabolites formed by liver
and lung from control and pretreated rats. Carcinogenesis 8, 425-
(
(
(
(
4
31.
14) Talaska, G., Roh, J ., Schamer, M., Reilman, R., Xue, W., and
3
2
Warshawsky, D. (1995) P- Postlabeling analysis of dibenz[a,j]-
acridine DNA adducts in mice: identification of proximate
metabolites. Chem.-Biol. Interact. 95, 161-174.
15) Fuchs, J ., Micoch, J ., Platt, K. L., and Oesch, F. (1993) Charac-
terization of highly polar bis-dihydrodiol epoxide-DNA adducts
formed after metabolic activation of dibenz[a,h]-anthracene.
Carcinogenesis 14, 863-867.
16) Carmichael, P. L., Platt, K. L, She, M. N., Lecoq, S., Oesch, F.,
Phillips, D. H., and. Grover, P. L. (1993) Evidence for the
involvement of a bis-diol-epoxide in the metabolic activation of
dibenz[a,h]anthracene to DNA-binding species in mouse skin.
Cancer Res. 53, 944-948.
F igu r e 6. Autoradiogram of reverse cycle sequencing of a Ha-
ras codon 61 EPCR positive papilloma DNA induced by a single
topical application of 0.2 µmol of DBA, showing CAA f CTA.
as well as dA residues in mouse skin DNA via DBADE
and an alternative pathway that through DBA-3,4,8,9-
bis-DHD-1,2-epoxide to bind to dG and dA, is also
involved. The mutational spectra are consistent with the
DNA binding of DBA. Further investigation of the
metabolic activation of DBA will require the preparation
and application of the DBADE stereoisomers and the bis-
DHD-epoxide metabolite(s).
(
17) Blout, E. R., and Corley, R. S. (1947) The reactions of â-naphthol,
â-naphthylamine and formaldehyde III, The dibenzacridine prod-
ucts. J . Am. Chem. Soc. 69, 763-768.
(
18) Talaska, G., Au, W. W., Ward, J . B., Randerath, K., and Legator,
M. S. (1987) The correlation between DNA adducts and chromo-
somal aberrations in the target organ of benzidine exposed
partially hepatetomized mice. Carcinogenesis, 8, 1899-1905.
19) Sayer, J . M., Lehr, R. E., Kumar, S., Yagi, H., Yeh, H. J . C.,
Holder, G. M., Duke, C. C., Silverton, J . V., Gibson, C., and J erina,
D. M. (1990) Comparative solvolytic reactivity of bay-region diol
epoxide derived from dibenz[a,j]anthracene and dibenzacridines.
J . Am. Chem. Soc. 112, 1177-1185.
(
Ack n ow led gm en t. Drs. K. J ayasimhulu and E.
Brooks of the Department of Chemistry, University of
Cincinnati kindly provided MS and NMR analysis. The
authors appreciate Dr M. Miller’s help in preparation of
the electronic files. This work is supported by NIEHS
Grants 5R01-ES04203, 2P01-ES05652, 5T32-ES07250,
and 1P30-ES06096.
(20) Rosario, C. A., Holder, G. M., and Duke, C. C. (1987) Synthesis
of potential metabolite of dibenz[a,j]acridine: dihydrodiols and
phenols. J . Org. Chem. 52, 1064-1072.
(
21) Sheng, S. C., Hilton, B. D., Roman, J . M., and Dipple, A. (1989)
DNA adducts from carcinogenic and noncarcinogenic enantiomers
of benzo[a]pyrene dihydrodiol epoxide. Chem. Res. Toxicol. 2,
334-340.
Refer en ces
(
22) Canella, K., Peltonen, K., and Dipple, A. (1991) Identification of
(
(
(
(
1) Warshawsky, D. (1992) Environmental sources, carcinogenicity,
mutagenicity, metabolism and DNA binding of nitrogen and
sulfur heterocyclic aromatics. Environ. Carcinogen. Ecotoxicol.
Rev. C10, 1-71.
2) Warshawsky, D., Talaska, G., Xue, W., and Schneider, J . (1996)
Comparative carcinogenicity, metabolism, and DNA binding of
(
+) and (-) anti benzo[a]-pyrene dihydrodiol epoxide-nucleic acid
32
adducts by the P-postlabeling assay. Carcinogenesis 12, 1109-
1
114.
(
23) Bigger, C. A. H., Sawicki, J . T., Blake, D. M., Raymond, L. G.,
and Dipple, A. (1983) Products of binding of 7,12-dimethylbenz-
[
a]anthracene to DNA in mouse skin. Cancer Res. 43, 5647-5651.
24) Xue, W., Schneider, J ., J ayasimhulu, K., and Warshawsky, D.
1993) Acetylation of phenolic derivatives of 7H-dibenzo[c,g]-
7
H-dibenzo[c,g]carbazole and dibenz[a,j]acridine. Crit. Rev. Toxi-
(
col. 26, 213-249.
(
3) IARC (1987) Overall evaluation of carcinogenicity: an updating
of IARC monographs volumes 1 to 42. Monographs on the
Evaluation of the Carcinogenic Risk of Chemicals to Humans,
Suppl. 7, Lyon France.
carbazole: Identification and quantitation of major metabolites
by rat liver microsomes. Chem. Res. Toxicol. 6, 345-350.
(
25) Randerath, E., Agrawal, H. P., Weaver, J . A., Bordelon, C. B.,
3
2
and Randerath, K. (1985) P-Postlabeling analysis of DNA
adducts persisting for up to 42 weeks in the skin, epidermis and
dermis of mice treated topically with 7,12-dimethyl benz[a]-
anthracene. Carcinogenesis 6, 1117-1126.
4) IARC (1983) Polynuclear aromatic compounds, Part 1, chemical,
environmental and experi-mental data. Monographs on the
Evaluation of the Carcinogenic Risk of Chemicals to Humans,-
Vol. 32, Lyon, France.
(
(
(
26) Talaska, G., Roh, J . H., and Getek, T. (1992) 32P-Postlabeling and
mass spectrometric methods for analysis of bulky, polyaromatic
carcinogen-DNA adducts in humans. J . Chromatogr. 580, 293-
(5) IARC (1973) Certain polycyclic aromatic hydrocarbons and het-
erocyclic compounds. Monographs on the Evaluation of the
Carcinogenic Risk of Chemicals to Humans,Vol. 3, Lyon, France.
(6) Andervont, H. B., and Shimkin, M. R. (1940) Biologic testing of
carcinogens, II. Pulmonary tumor induction techniques. J . Natl.
Cancer Inst. 1, 225-239.
(
3
23.
27) Lu, L. J ., Disher, R. M., Reddy, M. V., and Randerath, K. (1986)
3
2
P-Postlabeling assay in mice of transplacental DNA damage
induced by the environmental carcinogens Safrole, 4-amino-
biphenyl and benzo[a]pyrene. Cancer Res. 46, 3046-3054.
28) Gupta, R. C., Reddy, M. V., and Randerath, K. (1982) 32P-
Postlabeling analysis of non-radioactive aromatic carcinogen-DNA
adducts. Carcinogenesis 3, 1081-1092.
7) Warshawsky, D., Barkley, W., Miller, M. L., LaDow, K., and
Andringa, A. (1992) Comparative tumor-initiating ability of 7H-
dibenzo[c,g]carbazole and dibenz[a,j]acridine in mouse skin.
Toxicology 71, 233-243.
8) Warshawsky, D., Barkley, W., Miller, M. L., LaDow, K., and
Andringa, A. (1994) Carcinogenicity of 7H-dibenzo[c,g]carbazole,
dibenz[a,j]acridine and benzo[a]pyrene in mouse skin and liver
following topical application. Toxicology 93, 135-149.
9) Warshawsky, D., and Barkley, W. (1987) Comparative carcino-
genic potencies of 7H-dibenzo-[c,g]carbazole, dibenz[a,j]acridine
and benzo[a]pyrene in mouse skin. Cancer Lett. 37, 334-337.
(
(29) Reddy, M. V., Gupta, R. C., Randerath, E., and Randerath, K.
3
2
(1984) P-Postlabeling test for covalent DNA binding of chemicals
in vivo: application of a variety of aromatic carcinogens and
methylating agents. Carcinogenesis, 5, 231-243.
(30) Randerath, K., Reddy, M. V., and Gupta, R. C. (1981) 32P-
Postlabeling test for DNA damage. Proc. Natl. Acad. Sci. U.S.A.
78, 6126-6129.
(31) Mitchell, K. R., and Warshawsky, D. (1998) Mutational analysis
using enriched PCR and cycle sequencing. BioTechniques 24,
1028-1031.
(
(
10) Lehr, R. E., Kumar, S, Levin, W., Wood, A. W., Chang, R. L.,
Conney, A. H., Yagi, H., Sayer, J . M., and J erina, D. M. (1985)
The bay region theory of polycyclic aromatic hydrocarbon carcino-
genesis. In Polycyclic Aromatic Hydrocarbons and Carcinogenesis

8
78 Chem. Res. Toxicol., Vol. 14, No. 7, 2001
Xue et al.
(
32) Dipple, A. (1994) In DNA Adducts: Identification and Biological
Significance (Hemminki, K., Dipple, A., Segerb a¨ ck, D., Kadlubar,
F. F., Shuker, D., and Bartsch, H., Eds.) pp 107-129, IARS
Scientific Publications, Lyon, France.
(40) Prahalad, A. K., Ross, J . A., Nelson, G. B., Roop, B. C., King, L.
C., Nesnow, S., and Mass, M. J . (1997) Dibenzo[a,l]pyrene-induced
DNA adduction, tumorigenicity, and Ki-ras oncogenemutations
in strain A/J mouse lung. Carcinogenesis 18, 1955-1963.
(
(
(
33) Hillenkamp, F., Karas, M., Bearis, R. C., and Chait, B. T. (1991)
Matrix-Assisted Laser Desorption Ionization Mass Spectrometry
of Biopolymers. Anal. Chem. 63, 1193A-1203A.
34) Mitchell, K. R., and Warshawsky, D. (1999) Frequent Ha-ras
mutations in murine skin and liver tumors induced by 7H-
dibenzo[c,g]carbazole. Mol. Carcinog. 25, 107-112.
(
41) Colapietro, A. M., Goodell, A. L., and Smart, R. C. (1993)
Characterization of benzo[a]pyrene-initiated mouse skin papil-
lomas for Ha-ras mutations and protein kinase C levels. Car-
cinogenesis 14, 2289-2295.
(
42) Cheng, S. C., Prakash, A. S., Piggot, M. A., Hilton, B. D., Roman,
J . M., Lee, H., Harvey, R. G., and Dipple, A. (1988) Characteriza-
tion of 7,12-dimethylbenz[a]anthracene-adenine nucleosidead-
ducts. Chem. Res. Toxicol. 1, 216-221.
35) Bizub, D., Wood, A. W., and Skalka, A. N. (1986) Mutagenesis of
the Ha-ras oncogene in mouse skin tumors induced by polycyclic
aromatic hydrocarbons. Proc. Natl. Acad. Sci. U.S.A. 83, 6048-
6
052.
(
36) Ross, J . A., and Nesnow, S. (1999) Polycyclic aromatic hydrocar-
bons: correlations between DNA adducts and ras oncogene
mutations. Mutat. Res. 424, 155-166.
(43) Ronai, Z. A., Gradia, S., El-Bayoumy, A. K., Amin, S., and Hecht,
S. S. (1994) Contrastingincidence of ras mutations in rat mam-
mary and mouse skin tumors induced by anti-benzo[c]phenan-
threne-3,4-diol-1,2-epoxide. Carcinogenesis 15, 2113-2116.
(
37) DiGiovanni, J ., Beltran, L., Rupp, A., Harvey, R. G., and Gill, R.
D. (1993) Further analysis of c-Ha-ras mutations in papillomas
initiated by several polycyclic aromatic hydrocarbons and Papil-
lomas from uninitiated, promoter-treated skin in SENCAR mice.
Mol. Carcinog. 8, 272-279.
(
44) Agarwal, R., Canella, K. A., Yagi, H., J erina, D. M., and Dipple,
A. (1996) Benzo[c]phenathrene-DNA adducts in mouse epidermis
in relation to the tumorigenicities of four configurational isomeric
3
,4-dihydrodiol 1,2-epoxide. Chem. Res. Toxicol. 9, 586-592.
(
38) Brown, K., Buchmann, A., and Balmain, A. (1990) Carcinogen-
induced mutations in the mousec-Ha-ras gene provides evidence
of multiple pathways for tumor progression. Proc. Natl. Acad. Sci.
U.S.A. 87, 538-542.
39) Szeliga, J ., and Dipple, A. (1998) DNA adduct formation by
polycyclic aromatic hydrocarbondihydrodiol epoxide. Chem. Res.
Toxicol. 11, 1-11.
(45) Gill, R. D., Beltran, L., Nettikumara, A. N., Harvey, R. G.,
Kootstra, A., and DiGiovanni, J . (1992), Analysis of point muta-
tions in murine c-Ha-ras of skin tumors initiated with dibenz [a,j]-
anthracene and derivatives. Mol. Carcinog. 6, 53-59.
(
TX010014Y