Synthesis, characterization, and comparative 32P-postlabeling efficiencies of 2,6-dimethylaniline-DNA adducts

Article abstract of DOI:10.1021/tx0002031

2,6-Dimethylaniline (2,6-diMeA) is a ubiquitous environmental pollutant that is used in industry as a synthetic intermediate. It is also found in tobacco smoke and as a major metabolite of lidocaine. Although the potential carcinogenicity of 2,6-diMeA in humans is presently uncertain, this aromatic amine has been classified as a rodent carcinogen. In addition, it is known to form hemoglobin adducts in humans, which indicates a profile of metabolic activation similar to that of typical arylamine carcinogens. Like other aromatic amines, 2,6-diMeA has been shown to yield N-(deoxyguanosin-8-yl)-2,6-dimethylaniline (dG-C8-2,6-diMeA) as a major DNA adduct in vitro. In this study, we show that 2,6-diMeA yields an unusual pattern of DNA adducts. In addition to dG-C8-2,6-diMeA, we have isolated two new adducts, 4-(deoxyguanosin-N²-yl)-2,6-dimethylaniline (dG-N²-2,6-diMeA) and 4-(deoxyguanosin-O⁶-yl)-2,6-dimethylaniline (dG-O⁶-2,6-diMeA), from the reaction of N-acetoxy-2,6-dimethylaniline with deoxyguanosine. A similar reaction conducted with deoxyadenosine yielded 4-(deoxyadenosin-N⁶-yl)-2,6-dimethylaniline (dA-N⁶-2,6-diMeA). All four adducts were detected in DNA reacted with N-acetoxy-2,6-dimethylaniline, with the relative yields being 46% for dA-N⁶-2,6-diMeA, 22% for dG-N²-2,6-diMeA, 20% for dG-O⁶-2,6-diMeA, and 12% for dG-C8-2,6-diMeA. This product profile contrasts markedly with the usual pattern of adducts obtained with aromatic amines, where C8-substituted deoxyguanosine products typically predominate. We further analyzed the kinetics of the T₄ polynucleotide kinase (PNK)-catalyzed phosphorylation of the C8 and N² deoxyguanosine 3′-phosphate adducts from 2,6-diMeA. The kinetic parameters obtained with these two structurally different adducts are compared to those determined with the parent nucleotide (dG3′p), and with (±)-anti-10-(deoxyguanosin-N²-yl)-7,8,9-trihydroxy-7,8,9,10 -tetrahydrobenzo[α]pyrene 3′-phosphate, the major adduct derived from the environmental pollutant benzo[α]pyrene. The results indicate that all the adducts were labeled with lower efficiencies than dG3′p, stressing the likely underestimation of adduct levels in typical ³²P-postlabeling protocols. Nonetheless, the N² adducts derived from 2,6-diMeA and benzo[α]pyrene were both labeled with higher efficiencies than the C8 adduct derived from 2,6-diMeA, with the benzo[α]pyrene adduct being the best substrate for PNK. Thus, the data suggest that N² adducts from dG3′p are intrinsically better substrates than their C8 analogues for PNK, and that bulkier aromatic fragments may favor the enzyme-substrate interaction during the labeling step.

Full text of DOI:10.1021/tx0002031

Chem. Res. Toxicol. 2001, 14, 165-174
165
Syn th esis, Ch a r a cter iza tion , a n d Com p a r a tive
P -P ostla belin g Efficien cies of 2,6-Dim eth yla n ilin e-DNA
Ad d u cts
3
2
†
,‡
,†
Lu ´ı sa L. Gon c¸ alves, Frederick A. Beland,* and M. Matilde Marques*
Centro de Qu ´ı mica Estrutural, Complexo I, Instituto Superior T e´ cnico, Av. Rovisco Pais,
049-001 Lisboa, Portugal, and Division of Biochemical Toxicology,
1
National Center for Toxicological Research, J efferson, Arkansas 72079
Received September 14, 2000
2
,6-Dimethylaniline (2,6-diMeA) is a ubiquitous environmental pollutant that is used in
industry as a synthetic intermediate. It is also found in tobacco smoke and as a major metabolite
of lidocaine. Although the potential carcinogenicity of 2,6-diMeA in humans is presently
uncertain, this aromatic amine has been classified as a rodent carcinogen. In addition, it is
known to form hemoglobin adducts in humans, which indicates a profile of metabolic activation
similar to that of typical arylamine carcinogens. Like other aromatic amines, 2,6-diMeA has
been shown to yield N-(deoxyguanosin-8-yl)-2,6-dimethylaniline (dG-C8-2,6-diMeA) as a major
DNA adduct in vitro. In this study, we show that 2,6-diMeA yields an unusual pattern of DNA
adducts. In addition to dG-C8-2,6-diMeA, we have isolated two new adducts, 4-(deoxyguanosin-
2
2
6
N -yl)-2,6-dimethylaniline (dG-N -2,6-diMeA) and 4-(deoxyguanosin-O -yl)-2,6-dimethylaniline
6
(
dG-O -2,6-diMeA), from the reaction of N-acetoxy-2,6-dimethylaniline with deoxyguanosine.
A similar reaction conducted with deoxyadenosine yielded 4-(deoxyadenosin-N -yl)-2,6-
dimethylaniline (dA-N -2,6-diMeA). All four adducts were detected in DNA reacted with
N-acetoxy-2,6-dimethylaniline, with the relative yields being 46% for dA-N -2,6-diMeA, 22%
for dG-N -2,6-diMeA, 20% for dG-O -2,6-diMeA, and 12% for dG-C8-2,6-diMeA. This product
profile contrasts markedly with the usual pattern of adducts obtained with aromatic amines,
where C8-substituted deoxyguanosine products typically predominate. We further analyzed
6
6
6
2
6
4
the kinetics of the T polynucleotide kinase (PNK)-catalyzed phosphorylation of the C8 and
2
N deoxyguanosine 3′-phosphate adducts from 2,6-diMeA. The kinetic parameters obtained
with these two structurally different adducts are compared to those determined with the
2
parent nucleotide (dG3′p), and with (()-anti-10-(deoxyguanosin-N -yl)-7,8,9-trihydroxy-7,8,9,10-
tetrahydrobenzo[a]pyrene 3′-phosphate, the major adduct derived from the environmental
pollutant benzo[a]pyrene. The results indicate that all the adducts were labeled with lower
3
2
efficiencies than dG3′p, stressing the likely underestimation of adduct levels in typical P-
2
postlabeling protocols. Nonetheless, the N adducts derived from 2,6-diMeA and benzo[a]pyrene
were both labeled with higher efficiencies than the C8 adduct derived from 2,6-diMeA, with2
the benzo[a]pyrene adduct being the best substrate for PNK. Thus, the data suggest that N
adducts from dG3′p are intrinsically better substrates than their C8 analogues for PNK, and
that bulkier aromatic fragments may favor the enzyme-substrate interaction during the
labeling step.
In tr od u ction
Like other aromatic amines (6-8), 2,6-diMeA has the
potential to undergo metabolic activation through cyto-
chrome P450-mediated N-hydroxylation, followed by O-
1
2
,6-Dimethylaniline (2,6-diMeA) is a ubiquitous en-
vironmental pollutant that is used in industry as an
intermediate in the manufacture of several products,
including pesticides, dyestuffs, and synthetic resins (1).
In addition, 2,6-diMeA is present in nanogram amounts
in tobacco smoke (2, 3) and is also a major metabolite of
the potent anesthetic and antiarrythmic drug lidocaine
1
Abbreviations: Ar, arylamine; BPDE, (()-r-7,t-8-dihydroxy-t-9,-
10-epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene; dA, 2′-deoxyadenosine;
6
6
dA-N -2,6-diMeA, 4-(deoxyadenosin-N -yl)-2,6-dimethylaniline; dG, 2′-
deoxyguanosine; dG3′p, 2′-deoxyguanosine 3′-phosphate; dG3′,5′p, 2′-
deoxyguanosine 3′,5′-bisphosphate; dG-C8-Ar, N-(deoxyguanosin-8-
yl)arylamine; dG-C8-2,6-diMeA, N-(deoxyguanosin-8-yl)-2,6-dimethyl-
2
2
aniline; dG-N -2,6-diMeA, 4-(deoxyguanosin-N -yl)-2,6-dimethylaniline;
(
4). Although the potential carcinogenicity of 2,6-diMeA
2
2
dG-N -BPDE,
(()-anti-10-(deoxyguanosin-N -yl)-7,8,9-trihydroxy-
6
in humans is presently uncertain, results from a National
Toxicology Program bioassay have indicated that it is a
rodent carcinogen (5).
7,8,9,10-tetrahydrobenzo[a]pyrene; dG-O -2,6-diMeA, 4-(deoxyguanosin-
O -yl)-2,6-dimethylaniline; dG3′p-C8-4-ABP, N-(deoxyguanosin-8-yl)-
6
4
-aminobiphenyl 3′-phosphate; dG3′p-C8-Ar, N-(deoxyguanosin-8-yl)-
arylamine 3′-phosphate; dG3′p-C8-2,6-diMeA, N-(deoxyguanosin-8-yl)-
2
2
,6-dimethylaniline 3′-phosphate; dG3′p-N -2,6-diMeA, 4-(deoxy-
2 2
guanosin-N -yl)-2,6-dimethylaniline 3′-phosphate; dG3′p-N -BPDE,
(()-anti-10-(deoxyguanosin-N -yl)-7,8,9-trihydroxy-7,8,9,10-tetrahy-
2
*
To whom correspondence should be addressed. F.A.B.: HFT-110,
NCTR, J efferson, AR 72079; telephone, (870) 543-7205; fax, (870) 543-
136; e-mail, fbeland@nctr.fda.gov. M.M.M.: CQE, Complexo I, IST,
Av. Rovisco Pais, 1049-001 Lisboa, Portugal; telephone, 351-21-841-
200; fax, 351-21-846-4455; e-mail, matilde.marques@ist.utl.pt.
Instituto Superior T e´ cnico.
drobenzo[a]pyrene 3′-phosphate; 2,6-diMeA, 2,6-dimethylaniline; DMF,
N,N-dimethylformamide; dR, 2′-deoxyribosyl; dT3′p, thymidine 3′-
phosphate; ESI, electrospray ionization; N-AcO-2,6-diMeA, N-acetoxy-
2,6-dimethylaniline; N-OH-2,6-diMeA, N-hydroxy-2,6-dimethylaniline;
7
9
†
PEI, polyethyleneimine; PNK, T
adduct labeling.
4
polynucleotide kinase; RAL, relative
‡
National Center for Toxicological Research.
1
0.1021/tx0002031 CCC: $20.00 © 2001 American Chemical Society
Published on Web 01/05/2001

1
66 Chem. Res. Toxicol., Vol. 14, No. 2, 2001
Gon c¸ alves et al.
esterification to a reactive derivative capable of forming
DNA adducts. Indirect evidence for the formation of
N-hydroxyarylamines in vivo can be obtained from the
detection of hemoglobin adducts, which are formed by a
sequence involving oxidation of the N-hydroxyarylamines
to nitrosoarenes, covalent binding of the nitrosoarenes
to a cysteine SH residue in the â-chain of hemoglobin,
and rearrangement of the intermediate species to a
sulfinamide (9). This redox process occurs with concomi-
tant production of methemoglobin (10, 11). Hemoglobin
adducts from 2,6-diMeA have been found in cigarette
smokers and in nonsmokers with no known exposure to
this arylamine (12) and are also present at significant
levels in patients treated with therapeutic doses of
lidocaine (13). These findings, together with the fact that
lidocaine has been reported to induce severe methemo-
globinemia in some patients (14, 15), indicate that
humans have the ability to metabolize 2,6-diMeA to
N-hydroxy-2,6-dimethylaniline (N-OH-2,6-diMeA).
Aniline derivatives bearing two alkyl substituents
ortho to the nitrogen atom have been reported to be
nongenotoxic (16), and it has been hypothesized that
steric hindrance in the vicinity of the nitrogen could
prevent metabolic activation. 2,6-diMeA is only a weak
mutagen in Salmonella typhimurium TA100 in the
presence of S9 (17); however, its N-hydroxylated metabo-
lite, N-OH-2,6-diMeA, is the most mutagenic of the entire
series of N-(hydroxymethyl)-, N-(hydroxydimethyl)-, and
N-(hydroxyethyl)aniline derivatives in the same tester
strain in the absence of S9 (18, 19). These data show that
N-OH-2,6-diMeA has genotoxic properties and suggest
that it may be involved in the potential carcinogenicity
of 2,6-diMeA.
and report the isolation and characterization of two new
2
6
adducts from dG through the exocyclic N and O atoms,
6
and one new adduct from dA through the exocyclic N
atom. Possible implications of this uncommon adduct
pattern in the biological response elicited by 2,6-diMeA
are discussed. Furthermore, in view of the current
relevance of the 32P-postlabeling methodology for the
detection of carcinogen-DNA adducts, and of the uncer-
tainties regarding the influence of structural factors on
the relative labeling efficiencies of specific adducts (26-
29), we have also examined the kinetics of the T
4
polynucleotide kinase (PNK)-catalyzed phosphorylation
2
of the C8 and N deoxyguanosine 3′-phosphate adducts
from 2,6-diMeA. The kinetic parameters obtained with
these two structurally different adducts are compared to
those determined with the parent nucleotide, deoxygua-
nosine 3′-phosphate (dG3′p), and also to those we re-
ported previously using other N-(deoxyguanosin-8-yl)-
arylamine 3′-phosphate (dG3′p-C8-Ar) adducts (29). For
comparative purposes, the kinetics of phosphorylation of
2
(()-anti-10-(deoxyguanosin-N -yl)-7,8,9-trihydroxy-7,8,9,-
2
10-tetrahydrobenzo[a]pyrene 3′-phosphate (dG3′p-N -
BPDE), the major adduct (30) derived from the environ-
mental pollutant benzo[a]pyrene, was also analyzed.
Ma ter ia ls a n d Meth od s
Ca u tion : N-OH-2,6-diMeA and N-AcO-2,6-diMeA are poten-
tially carcinogenic. They should be handled with protective
3
2
clothing in a well-ventilated fumehood. Exposure to P should
be kept as low as possible, by working in a confined laboratory
area, with protective clothing, plexiglass shielding, Geiger
counters, and body dosimeters. Waste materials must be dis-
carded according to appropriate safety procedures.
Aromatic amine carcinogens are known to yield
N-(deoxyguanosin-8-yl)arylamines (dG-C8-Ar) as the ma-
jor persistent adducts in vivo, although minor adducts
Ch em ica ls. 2′-Deoxyguanosine (dG) was acquired from
Amersham U.S. Biochemical (Cleveland, OH). 2-Nitro-m-xylene
was obtained from Aldrich Chemical Co. (Milwaukee, WI). (()-
r-7,t-8-Dihydroxy-t-9,10-epoxy-7,8,9,10-tetrahydrobenzo[a]py-
rene (BPDE) was purchased from ChemSyn Laboratories (Len-
exa, KS). ATP was acquired from Boehringer Manheim Gmbh
2
6
through the N and O atoms of deoxyguanosine (dG) or
6
the C8 and N atoms of deoxyadenosine (dA) may also
be present in some instances (6-8). Although DNA
adducts from single-ring aromatic amines have yet to be
identified in vivo, several in vitro studies have confirmed
that these arylamines can form such adducts. The most
common strategy for the synthesis of DNA adduct
standards from aniline derivatives involves the reaction
of dG, dG nucleotides, or DNA with N-acyloxyanilines
3
2
(Indianapolis, IN). Carrier-free [γ- P]ATP was purchased from
ICN Pharmaceuticals (Costa Mesa, CA) and diluted 500-1000-
fold before being used in the labeling reactions. PNK (90 090
units/mg; 35 000 mg/mmol of monomer) was acquired from
Amersham U.S. Biochemical at a concentration of 30 units/µL.
Dilutions of the stock solution were performed as needed. All
other reagents and enzymes were purchased from Sigma
Chemical Co. (St. Louis, MO).
(19-23). These procedures, which mimic the metabolic
activation pathway for arylamines, give dG-C8-Ar ad-
ducts as the major products, but the yields are typically
very low, precluding the isolation of presumed minor
adducts. More recently developed strategies, involving
palladium-catalyzed C-N bond formation (24) or the use
of arylnitrenes as adduct precursors (25), have the
potential to become suitable routes for certain types of
minor arylamine adducts, but their use has yet to become
generalized.
Previous reports (19, 23) have shown that, like other
N-acyloxyarylamines, N-acetoxy-2,6-diMeA (N-AcO-2,6-
diMeA) yields N-(deoxyguanosin-8-yl)-2,6-dimethyl-
aniline (dG-C8-2,6-diMeA) as a major DNA adduct in
vitro. However, we have observed that the reaction profile
is considerably more complex than with other alkyl-
aniline derivatives, and that the isolation of additional
putative adducts requires thorough and efficient separa-
tion from a variety of hydrophilic solvolysis products. We
now provide evidence that N-AcO-2,6-diMeA yields an
unusual mixture of DNA adducts in comparable yields,
In str u m en ta tion . Melting temperatures were measured
with a Leica Galen III hot-stage apparatus and are uncorrected.
Reversed-phase HPLC analyses and separations were con-
ducted using a µBondapak C18 column (0.39 cm × 30 cm, Waters
Associates, Milford, MA), either on a Varian system consisting
of a Star 9012 ternary gradient pump and a Polychrom 9065
diode array spectrophotometric detector (Varian, Inc., Palo Alto,
CA), equipped with a Rheodyne injector (Rheodyne, Cotati, CA),
or on a Waters Associates system consisting of two model 510
pumps and a model 660 automated gradient controller, equipped
with a Rheodyne injector and a Hewlett-Packard 1050 diode
array spectrophotometric detector. The peaks were monitored
at 280 nm.
UV spectra were recorded with either a Beckman DU-40 or
a Shimadzu 1202 UV/vis spectrophotometer.
NMR spectra were obtained at 23 °C on a Varian Unity 300
1
spectrometer, operating at 300 MHz ( H spectra) or 75.43 MHz
13
(
2 6
C spectra). The adduct samples were dissolved in Me SO-d .
The proton chemical shifts were referenced to the residual
Me SO proton resonance (2.49 ppm) and the carbon chemical
2
1
2
shifts to the Me SO carbon resonance (39.50 ppm). H NMR

2
,6-Dimethylaniline-DNA Adducts
Chem. Res. Toxicol., Vol. 14, No. 2, 2001 167
assignments were based on comparisons with the spectra of the
parent nucleosides and of 2,6-diMeA, combined with homo-
nuclear decoupling experiments, chemical exchange of the labile
partially obscured by the solvent resonance), 62.01 (C5′), 71.14
(C3′), 83.69 (C1′), 87.91 (C4′), 117.92 (C5), 121.46 (ArC3,5), 122.11
(ArC2,6), 127.94 (ArC
4 1
), 137.03 (C8), 140.85 (ArC ), 150.66 (C4),
protons with D
(
2
O, and observations of nuclear Overhauser effect
155.21 (C2/C6), 156.36 (C6/C2); MS (ESI) (relative intensity)
NOE) enhancement patterns. 13C NMR assignments were
m/z 409 [(M + Na) , 18%], 387 (MH , 100%), 271 (MH
+
+
+
2
- dR,
based on published data for the parent deoxynucleosides dG and
dA, and for polysubstituted aniline derivatives (31), combined
with the use of a distortionless enhancement by polarization
transfer (DEPT) sequence, employing a pulse angle of 135°, and
with analysis of the C- H coupling patterns in heteronuclear
coupled spectra.
44%).
6
6
4-(Deoxygu a n osin -O
-yl)-2,6-d im et h yla n ilin e (d G-O -
2,6-d iMeA, 3): η 0.3%; mp >300 °C dec; UV (methanol) λ
max
1
6
237 (log ꢀ ) 4.0), 286 nm (log ꢀ ) 3.8); H NMR (DMSO-d ) δ
1
3
1
2.08 (6H, s, ArCH ), 2.24 (1H, m, H2′′), 2.57 (1H, m, H2′), 3.52
3
(2H, m, H5′,5′′), 3.82 (1H, m, H4′), 4.35 (1H, m, H3′), 4.44 (2H,
Positive ion electrospray ionization (ESI) mass spectra were
recorded on a Finnigan TSQ-7000 spectrometer, using flow
injection analysis in 50% methanol containing 0.1% ammonium
formate (pH 3.5), at a flow rate of 0.1 mL/min.
Liquid scintillation counting was performed on a Packard Tri-
Carb model 1600TR instrument (Packard Instrument Co.,
Meriden, CT), using Ultima Gold (Packard) as the scintillation
fluid. TLC plates were exposed to intensifying screens at room
temperature for variable periods of time (1-2 h) in Kodak X
cassettes. Visualization and quantitation of the chromatograms
were subsequently conducted with a phosphorimager (Storm 860
Imager, Molecular Dynamics, Sunnyvale, CA).
s, exchanged with D
exchanged with D O, 5′-OH), 5.27 (1H, d, J ) 2.7 Hz, exchanged
with D O, 3′-OH), 6.21 (1H, t, J ) 6.3 Hz, H1′), 6.35 (2H, s,
exchanged with D
H8); 13C NMR (DMSO-d
(C5′), 70.81 (C3′), 82.81 (C1′), 87.65 (C4′), 114.12 (C5), 120.75
(ArC3,5), 121.60 (ArC2,6), 128.38 (ArC ), 138.31 (C8), 142.25
(ArC ), 154.30 (C4), 159.88 (C2/C6), 160.66 (C6/C2); MS (ESI)
2 2
O, ArNH ), 4.98 (1H, t, J ) 5.1 Hz,
2
2
2
2
O, N H
2
), 6.65 (2H, s, ArH3,5), 8.14 (1H, s,
), 39.37 (C2′), 61.76
6
) δ 17.90 (ArCH
3
4
1
+
+
(relative intensity) m/z 409 [(M + Na) , 16%], 387 (MH , 100%),
+
271 (MH
2
- dR, 93%).
The following adduct was isolated from dA.
6
6
4
-(Deoxya d en osin -N -yl)-2,6-d im eth yla n ilin e (d A-N -2,6-
Syn th esis of Nu cleosid e Ad d u cts. The nucleoside adducts
from 2,6-diMeA were prepared by reacting dG or dA with
N-AcO-2,6-diMeA, essentially as described for other alkylaniline
adducts (19, 22). Briefly, a 200 mM solution of the nucleoside
in a 1/2 mixture of 10 mM sodium citrate (pH 6.5) and DMF
was prepared, purged with argon, and kept at 0-5 °C. To this
solution were added aliquots of 2 M N-AcO-2,6-diMeA in THF,
which had been generated from N-OH-2,6-diMeA and acetyl
cyanide (19) and kept at -30 °C. Following addition of 5 molar
equiv of N-AcO-2,6-diMeA over a period of 2 h, the reaction
mixture was maintained under argon and stirred overnight at
room temperature. Upon evaporation to dryness, the mixture
was resuspended in water, and the lower-polarity solvolysis
products were removed by extraction with methylene chloride.
The adducts were subsequently partitioned into n-butanol; the
n-butanol was evaporated, and the residue was dissolved in a
small volume of methanol and separated by column chroma-
tography on Sephadex LH-20 (Pharmacia/PL Biochemicals,
Piscataway, NJ ), using a 0 to 100% step gradient (in 5% steps)
of aqueous methanol, followed by fractionation of the 15-50%
MeOH eluates using RP18 modified silica (E. Merck, Darmstadt,
Germany). The adducts derived from dG were finally purified
by reversed-phase HPLC, using a 20 min linear gradient of 5
to 50% aqueous methanol, followed by a 5 min linear gradient
to 70% methanol.
d iMeA, 4): η 2%; mp >220 °C dec; UV (methanol) λ
max
306 nm
1
(log ꢀ ) 4.2); H NMR (DMSO-d ) δ 2.08 (6H, s, ArCH ), 2.26
6
3
(1H, m, H2′′), 2.74 (1H, m, H2′), 3.60 (2H, m, H5′,5′′), 3.88 (1H,
m, H4′), 4.38 (2H, exchanged with D O, ArNH ), 4.42 (1H, m,
2
2
H3′), 5.25 (1H, bs, exchanged with D O, OH), 5.36 (1H, bs,
2
exchanged with D O, OH), 6.36 (1H, t, J ) 6.8 Hz, H1′), 7.20
2
(2H, s, ArH3,5), 8.24 (1H, s, H8), 8.40 (1H, s, H2), 9.33 (1H, s,
6
13
exchanged with D O, N H); C NMR (DMSO-d ) δ 18.25
2
6
(ArCH ), 39.67 (C2′), 62.01 (C5′), 71.12 (C3′), 84.29 (C1′), 88.15
3
(C4′), 119.96 (C5), 121.0 (ArC2,6), 122.23 (ArC3,5), 128.05 (ArC4),
140.03 (C8), 140.82 (ArC ), 148.79 (C4), 152.49 (C2), 152.76 (C6);
1
+
+
MS (ESI) (relative intensity) m/z 371 (MH , 100%), 255 (MH
2
- dR, 44%).
Syn th esis of d G3′p Ad d u cts. (1) 2,6-Dim eth yla n ilin e
Ad d u cts. A solution of dG3′p (10 mg) in 10 mM sodium citrate
(pH 6.0, 300 µL) was allowed to react overnight with N-AcO-
2
,6-diMeA (1.2 molar equiv in 200 µL of THF), generated as
described above. The THF was evaporated; water was added to
a final volume of 1 mL, and the solvolysis products were
extracted with diethyl ether (5 × 500 µL). The adducts were
separated by reversed-phase HPLC, using a 30 min linear
gradient of 5 to 50% acetonitrile in 100 mM ammonium acetate
(pH 5.7). The isolated adducts were N-(deoxyguanosin-8-yl)-2,6-
dimethylaniline 3′-phosphate (dG3′p-C8-2,6-diMeA) and 4-(deoxy-
2
2
guanosin-N -yl)-2,6-dimethylaniline 3′-phosphate (dG3′p-N -2,6-
diMeA). For characterization purposes, aliquots of both adducts
were treated with alkaline phosphatase for 3 h at 37 °C. In both
instances, the UV spectra and HPLC retention times of the
dephosphorylated products were identical to those of the cor-
responding dG adducts.
The following adducts were isolated from dG.
N-(Deoxygu a n osin -8-yl)-2,6-d im eth yla n ilin e (d G-C8-2,6-
d iMeA, 1): η 1%; mp >285 °C dec; UV (methanol) λmax 269 nm
1
(
log ꢀ ) 4.2); H NMR (DMSO-d
6
) δ 2.02 (1H, m, H2′′), 2.13 (6H,
s, ArCH
H4′), 4.39 (1H, m, H3′), 5.32 (1H, bs, exchanged with D
OH), 5.56 (1H, bs, exchanged with D O, 5′-OH), 6.32 (1H, m,
H1′), 6.33 (2H, bs, exchanged with D
ArH3,4,5), 7.84 (1H, s, exchanged with D
s, exchanged with D
O, N1H); 13C NMR (DMSO-d
), 37.83 (C2′), 61.30 (C5′), 71.46 (C3′), 82.78 (C1′), 87.17
C4′), 112.57 (C5), 125.40 (ArC ), 127.81 (ArC3,5), 135.01 (ArC2,6),
3
), 2.71 (1H, m, H2′), 3.67 (2H, m, H5′,5′′), 3.91 (1H, m,
O, 3′-
(2) (()-a n ti-10-(Deoxygu a n osin -N -yl)-7,8,9-tr ih yd r oxy-
2
2
2
7,8,9,10-tetr a h yd r oben zo[a ]p yr en e 3′-P h osp h a te (d G3′p -
2
2
2
O, N H
O, ArNH), 10.56 (1H,
) δ 18.12
2
), 7.04 (3H, m,
N -BP DE). An 11 mM solution of (()-r-7,t-8-dihydroxy-t-9,10-
2
epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene (BPDE) in THF (300
µL) was added to an 8-fold molar excess of dG3′p, dissolved in
the same volume of 10 mM sodium citrate (pH 6.0), and the
mixture was stirred overnight at room temperature. Following
evaporation of the THF, water was added to a final volume of
1 mL, and the unbound BPDE derivatives were removed by
2
6
(ArCH
3
(
4
1
1
2
37.32 (C8/ArC
1
), 145.85 (ArC
1
/C8), 149.89 (C4), 152.41 (C2),
+
55.52 (C6); MS (ESI) (relative intensity) m/z 387 (MH , 8%),
+
2
71 (MH
4
2
- dR, 100%).
repeated extractions with diethyl ether. The dG3′p-N -BPDE
-(Deoxygu a n osin -N -yl)-2,6-d im et h yla n ilin e (d G-N2-
2
was isolated by reversed-phase HPLC using a 40 min nonlinear
gradient (Waters curve 2) of 20 to 56% methanol in 100 mM
ammonium acetate (pH 5.7). Upon treatment with alkaline
phosphatase, the HPLC retention time and UV spectrum of the
adduct were identical to those of (()-anti-10-(deoxyguanosin-
2
2
2
,6-d iMeA, 2): η 0.5%; mp >160 °C dec; UV (methanol) λmax
1
84 nm (log ꢀ ) 4.0); H NMR (DMSO-d
6 3
) δ 2.08 (6H, s, ArCH ),
.22 (1H, m, H2′′), 2.64 (1H, m, H2′), 3.44 (2H, m, H5′,5′′), 3.79
(
D
1H, m, H4′), 4.29 (1H, m, H3′), 4.6-5.1 (1H, bs, exchanged with
2
2
O, OH), 5.1-5.4 (1H, bs, exchanged with D O, OH), 6.12 (1H,
2
N -yl)-7,8,9-trihydroxy-7,8,9,10-tetrahydrobenzo[a]pyrene ob-
t, J ) 6.9 Hz, H1′), 7.00 (2H, s, ArH3,5), 7.95 (1H, s, H8), 8.26
tained by reacting BPDE with dG. The concentration of dG3′p-
2
2
(1H, s, exchanged with D
2
O, N H), 10.34 (1H, s, exchanged with
N -BPDE was determined using the extinction coefficient
1
3
D
2
O, N1H); C NMR (DMSO-d ) δ 18.34 (ArCH
6
3
), ∼39.5 (C2′,
reported by Pulkrabek et al. (32).

1
68 Chem. Res. Toxicol., Vol. 14, No. 2, 2001
Gon c¸ alves et al.
DNA Mod ifica tion . The modification of DNA with 2,6-
diMeA was conducted by reacting salmon testes DNA with
N-AcO-2,6-diMeA as described previously (19). Upon standard
enzymatic hydrolysis of the DNA to nucleosides (33), the HPLC
retention times and UV spectra of the 2,6-diMeA-DNA adducts
were compared to those of the corresponding dG and dA adducts.
Kin etic Stu d ies. (1) P h osp h or yla tion Rea ction s. The
effect of the substrate concentration on the PNK-catalyzed 32P-
postlabeling of dG3′p adducts was studied for dG3′p-C8-2,6-
2
2
diMeA, dG3′p-N -2,6-diMeA, and dG3′p-N -BPDE using the
methodology described by Mourato et al. (29). Briefly, aliquots
(
5 µL) of a labeling solution containing 600 milliunits/µL (50
fmol/µL) PNK and 955 µM ATP in 80 mM bicine-NaOH, 40 mM
MgCl , 40 mM dithiothreitol, and 4 mM spermidine (pH 9.4)
buffer were added to 15 µL of various substrate solutions (20-
000 pmol, 1.3-267 µM). The mixtures were vortexed, centri-
2
4
fuged for 5-10 s, and incubated at 37 °C for 5 min. Each reaction
was quenched by mixing with 2 µL of 673 µM thymidine 3′-
phosphate (dT3′p) and 30 milliunits of apyrase (1.5 µL), and
the mixtures were further incubated for 30 min at 37 °C. In
each instance, the adduct concentrations were determined by
UV spectroscopy, using the extinction coefficients of the corre-
sponding nucleoside adducts (vide supra).
(
2) Ch r om a togr a p h y. Aliquots of each mixture (15 µL) were
spotted on Macherey-Nagel polyethyleneimine cellulose (PEI-
cellulose) TLC plates (20 cm in length), purchased from Alltech
Associates, Inc. (Deerfield, IL), that had been washed with
deionized water and dried prior to use. The elution conditions
were as follows.
F igu r e 1. Structures of the adducts isolated from the reaction
of N-AcO-2,6-diMeA and nucleosides.
(
i) d G3′,5′p -C8-2,6-d iMeA. The elution was conducted onto
a Whatman #1 paper wick (direction D1) with 2.5 M sodium
phosphate (pH 6.0) and then in the opposite direction (D2) with
to be dG-C8-2,6-diMeA (1, Figure 1), the product of C8
substitution through the arylamine nitrogen. This ad-
duct, which has been characterized previously (19, 23),
eluted in the 40-50% methanol fraction from Sephadex
LH-20 and was obtained in 1% yield. However, as
indicated above, N-AcO-2,6-diMeA yielded a complex
profile upon reaction with dG. In contrast to what we
observed with other methyl- and dimethylaniline deriva-
tives (19, 22), additional putative adducts were present
in levels comparable to that of dG-C8-2,6-diMeA. These
adducts had to be thoroughly separated from hydrophilic
side products, stemming from hydrolysis of N-AcO-2,6-
diMeA. Following partitioning into n-butanol, we were
able to isolate two additional dG adducts, which eluted
in the 30-40% methanol fraction from Sephadex LH-20.
The purification of both species required further chro-
matographic separation on RP18 modified silica, followed
by HPLC. The new adducts (Figure 1) were characterized
1
.8 M lithium formate and 4.2 M urea (pH 3.5).
2
(
ii) d G3′,5′p -N -2,6-d iMeA. The elution was conducted onto
a Whatman #1 paper wick (direction D1) with 900 mM sodium
phosphate (pH 6.8) and then in direction D2 with 1.8 M lithium
formate and 4.2 M urea (pH 3.5).
2
(
iii) d G3′,5′p -N -BP DE. The elution was conducted onto a
Whatman #1 paper wick (direction D1) with 900 mM sodium
phosphate (pH 6.8) and then in direction D2 with 3.6 M lithium
formate and 8.5 M urea (pH 3.5).
3
2
(
3) Ca lcu la tion s. The specific activity of the [γ- P]ATP in
each labeling mixture was established by liquid scintillation
counting of diluted aliquots, whose concentration had been
determined by UV spectroscopy. Following elution and prior to
exposure of each TLC plate to the intensifying screen, an aliquot
of the labeling mixture was spotted on the plate to serve as a
3
2
quantitation standard. Corrections for P decay were performed
whenever necessary. The kinetic parameters (Vmax and K ) were
m
calculated using the weighted least-squares fitting program
Leonora (34).
2
2
as 4-(deoxyguanosin-N -yl)-2,6-dimethylaniline (dG-N -
6
2
,6-diMeA, 2, 0.5%) and 4-(deoxyguanosin-O -yl)-2,6-
Resu lts
6
dimethylaniline (dG-O -2,6-diMeA, 3, 0.3%) on the basis
Ch a r a cter iza tion of th e 2,6-d iMeA Ad d u cts. One
of the goals of this work was to elucidate the structures
of several putative 2,6-diMeA DNA adducts that are
produced from N-acyloxy-2,6-diMeA species but have
eluded isolation and characterization due to the complex-
ity of the reaction mixtures. Toward this purpose, N-AcO-
of spectroscopic evidence.
(1) d G-N -2,6-d iMeA (2). The ESI mass spectrum of
2
adduct 2 in the positive ion mode indicated the presence
of the protonated (m/z 387) molecular ion, as expected
from an adduct of dG with 2,6-diMeA. In addition, the
detection of an intense ion at m/z 271 is consistent with
loss of the deoxyribosyl fragment from the parent ion.
The UV spectrum of adduct 2 (Figure 2) had a major
absorption band at 284 nm, which represents a batho-
chromic shift of ca. 30 nm compared to dG, presumably
due to significant conjugation between the arylamine and
dG chromophores. This observation was the first sug-
gestion that substitution did not occur through the
arylamine nitrogen, since the steric hindrance caused by
the presence of the two ortho methyl groups should
decrease the ability of the arylamine nitrogen to act as
an electron donor. For instance, while the UV absorption
2
,6-diMeA (19) was used as the electrophilic synthon to
generate 2,6-diMeA adducts from dG, dG3′p, dA, and
DNA. The adducts formed upon reaction with the nucleo-
sides or dG3′p were isolated by chromatography following
solvent extractions to remove the less polar solvolysis
products of N-AcO-2,6-diMeA. The adducts formed upon
reaction with DNA were identified after enzymatic hy-
drolysis of the modified DNA to nucleosides.
Ad d u cts fr om th e Rea ction w ith d G. In agreement
with previous reports (19, 23), the major adduct from the
reaction between dG and N-AcO-2,6-diMeA was found

2
,6-Dimethylaniline-DNA Adducts
Chem. Res. Toxicol., Vol. 14, No. 2, 2001 169
2
dG-N -2,6-diMeA, the UV spectrum of this adduct (Figure
) had two absorption bands with similar intensities,
2
centered at 237 and 286 nm, which suggested that two
relatively independent chromophores were present in the
molecule. A similar UV spectral pattern has been re-
6
ported recently for an O -substituted adduct formed by
reaction of dG with the arylnitrene generated from
1
2
-methylnitrobenzene (25). The H NMR spectrum of 3
showed all the expected nonexchangeable sugar reso-
nances, as well as a singlet at 8.14 ppm, assigned to the
guanine H8 (Table 1). A six-proton singlet at 2.08 ppm
and an aromatic pattern of two equivalent protons at 6.65
ppm were consistent with substitution in the 2,6-diMeA
ring, para to the amino group. Four sets of labile protons
were observed. Two of these, at 4.98 and 5.27 ppm,
accounted for one proton each and were assigned to the
two sugar hydroxyls, 5′-OH and 3′-OH, respectively, on
the basis of their coupling patterns. The two additional
sets of exchangeable signals, at 4.44 and 6.35 ppm,
accounted for two protons each. This suggests the pres-
2
ence of two exocyclic NH groups, one in the arylamine
fragment and the other in the dG moiety. Furthermore,
no downfield resonance consistent with the presence of
a dG imino proton was observed. Taken together, these
6
data suggest substitution through the O atom of dG.
Further support for the identification of adduct 3 as dG-
O -2,6-diMeA stemmed from the observation of downfield
6
F igu r e 2. UV spectra of the isolated 2,6-diMeA adducts,
2
recorded in methanol: (a) dG-C8-2,6-diMeA (1), (b) dG-N -2,6-
shifts (3-5 ppm) of the C2 and C6 resonances, compared
to the same resonances in dG (31), suggesting full
aromatization of the purine ring system. Similar down-
field shifts of the C2 and C6 resonances have been
observed in other O -substituted derivatives of dG (25,
35).
6
6
diMeA (2), (c) dG-O -2,6-diMeA (3), and (d) dA-N -2,6-diMeA
4).
(
of dG-C8-2,6-diMeA had a bathochromic shift of only 15
nm compared to dG, shifts of up to 30 nm have been
observed for other dG-C8-Ar adducts without ortho
6
1
substituents in the arylamine moiety (19, 22). The H
Ad d u cts fr om th e Rea ction w ith d G3′p . The HPLC
profile from the reaction of N-AcO-2,6-diMeA with dG3′p
was virtually identical to that obtained with dG. The
NMR spectrum of adduct 2 provided conclusive structural
proof. Thus (Table 1), a six-proton singlet at 2.08 ppm
and a two-proton nonexchangeable singlet at 7.00 ppm
were fully consistent with substitution in the aromatic
ring, para to the arylamine nitrogen. The observation of
a one-proton singlet at 7.95 ppm, assigned to guanine
H8, further indicated that substitution must have oc-
curred through one of the dG heteroatoms. With the
2
major adducts, dG3′p-C8-2,6-diMeA and dG3′p-N -2,6-
diMeA, were isolated by reversed-phase HPLC. Both had
UV spectra identical to those of the corresponding dG
adducts, into which they were converted upon treatment
with alkaline phosphatase. On the basis of the UV
6
pattern of the reaction products, an O -substituted ad-
6
exception of the arylamine NH
2
, presumably obscured by
duct, dG3′p-O -2,6-diMeA, was also present in the reac-
the water resonance, all the expected exchangeable
signals were detected. Of these, two downfield one-proton
signals, at 8.26 and 10.34 ppm, were essential for
structure elucidation (Figure 3). The latter was assigned
tion mixture. However, this adduct was formed in low
yield and was not isolated.
Ad d u ct fr om th e Rea ction w ith d A (d A-N -2,6-
d iMeA, 4). The HPLC profile from the reaction of N-AcO-
2,6-diMeA with dA indicated the presence of one putative
major adduct, obtained in 2% yield, which eluted in the
30-40% methanol fraction from Sephadex LH-20
6
to the guanine imino proton. Both the location (ca. 2 ppm
2
downfield from the N -H
2
protons in dG) and the intensity
of the signal at 8.26 ppm were consistent with substitu-
1
tion through the exocyclic nitrogen of dG. The H NMR
The ESI mass spectrum of this species in the positive
ion mode was consistent with formation of a dA adduct
from 2,6-diMeA. Thus, the parent peak (m/z 371) cor-
responded to the protonated molecular ion, and a promi-
nent ion at m/z 255 resulted from loss of the deoxyribosyl
fragment. The UV spectrum of adduct 4 had one major
absorption band at 306 nm (Figure 2), which represents
a bathochromic shift of ca. 50 nm compared to dA,
suggesting the occurrence of extensive conjugation be-
spectrum further indicated the presence of all the ex-
pected deoxyribose resonances, and the 13C NMR spec-
trum confirmed the presence of all the expected carbon
resonances. Taken together, the spectroscopic data for
adduct 2 are fully compatible with the formation of a
covalent bond between the N2 atom of dG and the C4
atom of 2,6-diMeA.
6
(
2) d G-O -2,6-d iMeA (3). The ESI mass spectrum of
1
adduct 3 in the positive ion mode indicated the presence
of the protonated (m/z 387) and sodiated (m/z 409)
molecular ions expected from an adduct of dG with 2,6-
diMeA. Like adduct 2, an intense ion at m/z 271,
consistent with loss of the deoxyribosyl fragment from
the parent ion, was detected. In contrast with a single
major absorption band observed for dG-C8-2,6-diMeA and
tween the arylamine and dA chromophores. The H and
1
3
C NMR spectra of this adduct are shown in Figure 4
(panels a and b, respectively). As indicated in Figure 4b,
all the expected carbon resonances were present. The
detection of one singlet due to two equivalent methyl
groups (2.08 ppm) and one singlet due to two equivalent
1
aromatic protons (7.20 ppm) in the H NMR spectrum of

1
70 Chem. Res. Toxicol., Vol. 14, No. 2, 2001
Gon c¸ alves et al.
Ta ble 1. P r oton Ch em ica l Sh ifts Releva n t for Str u ctu r a l Elu cid a tion of th e 2,6-d iMeA Ad d u cts
δ (ppm)a
2
6
6
dG-C8-2,6-diMeA (1)
dG-N -2,6-diMeA (2)
dG-O -2,6-diMeA (3)
dA-N -2,6-diMeA (4)
CH3
ArH
2.13 (s, 6H)
7.04 (m, 3H)
7.84 (s, 1H)
6.33 (s, 2H)
-
2.08 (s, 6H)
7.00 (s, 2H)
NDc
2.08 (s, 6H)
6.65 (s, 2H)
4.44 (s, 2H)
6.35 (s, 2H)
-
2.08 (s, 6H)
7.20 (s, 2H)
4.38 (s, 2H)
-
9.33 (s, 1H)
8.24 (s, 1H)
8.40 (s, 1H)
-
b
ArNH/ArNH2
2
2
b
N -H/N -H2
8.26 (s, 1H)
-
6
b
N -H
H8
H2
N1H
-
7.95 (s, 1H)
-
8.14 (s, 1H)
-
-
b
10.56 (s, 1H)
10.34 (s, 1H)
-
a
Recorded in DMSO-d6. b Exchanged with D2O. c ND, not detected.
two singlets at 8.24 and 8.40 ppm, assigned to adenine
H8 and H2, respectively. Finally, the presence of a one-
proton exchangeable signal at 9.33 ppm (Figure 4a), ca.
6
2
ppm downfield from the N -H
2
resonance in dA (not
shown), strongly indicated the existence of a bond
between the exocyclic nitrogen of dA and the arylamine
C4. Thus, the identification of adduct 4 as 4-(deoxy-
6
6
adenosin-N -yl)-2,6-dimethylaniline (dA-N -2,6-diMeA) is
fully consistent with the spectroscopic data.
Ad d u cts fr om th e Rea ction w ith DNA. Like other
N-acetoxyarylamines, N-AcO-2,6-diMeA has been shown
to yield the C8-substituted dG derivative, dG-C8-2,6-
diMeA, as a major adduct when reacted with DNA in
vitro (19). Following solvent extractions and enzymatic
hydrolysis of the DNA to nucleosides, we compared the
UV profiles and HPLC retention times of the reaction
products with those of the new adduct standards obtained
from the reaction of N-AcO-2,6-diMeA with dG and dA
F igu r e 3. Downfield region (6-12 ppm) of the 1H NMR
2
spectrum of dG-N -2,6-diMeA (2), recorded in DMSO-d
6
. Full
chemical shift assignments are given in Materials and Methods.
(not shown). When the extinction coefficients determined
for each of the four 2,6-diMeA adduct standards were
taken into consideration (cf. Materials and Methods), we
concluded that the relative yields were approximately
6
2
4
6% for dA-N -2,6-diMeA, 22% for dG-N -2,6-diMeA, 20%
6
for dG-O -2,6-diMeA, and 12% for dG-C8-2,6-diMeA. The
fact that the dA derivative is the adduct formed in
greatest proportion upon reaction with DNA is consistent
with the relative yields obtained in the reactions with
the nucleosides. In addition, it should be noted that
although dG-C8-2,6-diMeA was the major adduct ob-
tained from the reaction of N-AcO-2,6-diMeA with dG,
2
6
the lower yields of the dG-N -2,6-diMeA and dG-O -2,6-
diMeA adducts isolated in that reaction are likely to have
resulted from the laborious purification procedure. This
product profile contrasts markedly with the usual pre-
dominance of dG-C8-Ar adducts when other alkylated
aniline derivatives are reacted with DNA (19, 22).
32P -P ostla belin g Stu d ies. With the aim of evaluating
the contribution of structural factors to the efficiency of
3
2
P-postlabeling of specific DNA adducts, we conducted
a comparative kinetic analysis of the PNK-catalyzed
phosphorylation reaction, using 2′-deoxyguanosine 3′-
phosphate (dG3′p) and the dG3′p adducts from 2,6-diMeA
1
2
F igu r e 4. H NMR spectrum (a) and proton broadband-
(dG3′p-C8-2,6-diMeA and dG3′p-N -2,6-diMeA). For com-
1
3
6
2
decoupled C NMR spectrum (b) of dA-N -2,6-diMeA (4),
recorded in DMSO-d . Full chemical shift assignments are given
6
in Materials and Methods. S represents the DMSO resonance
and W the water resonance.
parative purposes, dG3′p-N -BPDE, the dG3′p adduct
from BPDE, was included in the study.
The experimental conditions for this type of study have
been described (29). Briefly, an incubation time of 5 min
was chosen on the basis of our previous results (29) with
dG3′p and N-(deoxyguanosin-8-yl)-4-aminobiphenyl 3′-
phosphate (dG3′p-C8-4-ABP), which indicated that the
linear region of the time course curve spanned ap-
proximately 10 min regardless of the substrate. A PNK
concentration of 150 milliunits/µL (12 nM) was selected
adduct 4 (Figure 4a and Table 1) indicated that, like dG-
2
6
N -2,6-diMeA and dG-O -2,6-diMeA, substitution oc-
curred through the C4 atom of 2,6-diMeA. This was
further confirmed by the presence of an exchangeable
two-proton singlet at 4.38 ppm, assigned to the arylamine
2
NH . All the sugar resonances were observed, as were

2
,6-Dimethylaniline-DNA Adducts
Chem. Res. Toxicol., Vol. 14, No. 2, 2001 171
trations selected to ensure that the Vmax plateau was
reached. The results indicated that Michaelis-Menten
kinetics was followed for all the substrates. As shown in
Table 2, the apparent PNK turnover number (kcat ) 160
-1
-1
-1
min ) and specificity constant (kcat/K ) 19 µM min )
m
obtained with dG3′p were substantially higher than those
determined with any of the dG3′p-adducted substrates.
This is fully consistent with our previous results using a
series of dG3′p-C8-Ar adducts and strongly suggests that
3
2
P-postlabeling of adducted nucleotides is less efficient
than that of dG3′p. In addition, it should be noted that
the PNK specificity constant determined with dG3′p in
this study is approximately 3-fold higher than the one
we reported previously (29), which demonstrates the
importance of conducting the kinetic measurements at
a higher ATP concentration.
Despite a consistently lower efficiency of labeling of all
the adducts compared to that of the parent nucleotide,
significant differences were apparent depending on the
adduct structures. Thus, while the enzyme specificity
constant obtained with dG3′p-C8-2,6-diMeA (kcat/K
m
) 1.1
-
1
-1
µM min ) was comparable to that previously deter-
-
1
-1
mined with dG3′p-C8-4-ABP [kcat/K
m
) 1.4 µM min
2
(
29)], the labeling efficiency of dG3′p-N -2,6-diMeA [kcat/
-1
-1
K
m
) 2.4 µM min (Table 2)] was approximately 2-fold
2
higher. This result suggests that N adducts from dG3′p
are intrinsically better substrates than their C8 ana-
logues for PNK. The enzyme specificity constant (kcat/K
m
2
-
1
-1
)
7.9 µM
min ) determined with the bulkier N
2
adduct, dG3′p-N -BPDE, is consistent with this inter-
pretation. In addition, since the BPDE derivative was the
adduct labeled with greatest efficiency, the results fur-
ther imply that the bulky aromatic moiety may play a
role in the binding of the substrate to the active site of
the enzyme.
Discu ssion
Mu ltip licity of 2,6-d iMeA Ad d u cts. In this study,
we have shown that N-AcO-2,6-diMeA, a metabolically
plausible electrophilic derivative of 2,6-diMeA, forms an
unusual pattern of DNA adducts. Unlike N-acyloxy
derivatives of other alkylanilines, which give dG-C8-Ar
adducts as the predominant products of reaction with
DNA in vitro (19, 22), N-AcO-2,6-diMeA yielded three
F igu r e 5. Variation of the initial velocity of phosphorylation
as a function of substrate concentration, in the presence of 240
µM ATP and 150 milliunits/µL PNK, using (a) dG3′p-C8-2,6-
diMeA, (b) dG3′p-N -2,6-diMeA, and (c) dG3′p-N -BPDE as
substrates. The incubation conditions are outlined in Materials
and Methods.
2
2
because this enzyme concentration falls within the range
of PNK concentrations (100-1000 milliunits/µL) typically
used in 32P-postlabeling protocols (26, 27) and is suf-
ficiently low to prevent the accumulation of bisphosphate
products, thus ensuring stationary state conditions (29).
The ATP concentration was set at 240 µM because our
previous study (29) indicated that this was a near-
saturating concentration, which allowed the determina-
tion of satisfactory kinetic data while preventing exces-
sive exposure to 32P. Since our previously reported kinetic
parameters for the parent nucleotide, dG3′p, were ob-
tained with a lower ATP concentration (101 µM), we also
analyzed the labeling of this substrate at 240 µM ATP.
The experimental curves displaying the initial velocity
of phosphorylation as a function of substrate concentra-
tion are shown in Figure 5 for each of the three dG3′p
adducts. The calculated kinetic parameters are listed in
Table 2. In all instances, a greater number of data points
was collected in the region of low substrate concentration
for better accuracy in defining the experimental curve
2
major dG adducts, dG-C8-2,6-diMeA, dG-N -2,6-diMeA,
6
6
and dG-O -2,6-diMeA, and one major dA adduct, dA-N -
2,6-diMeA. Furthermore, the dA adduct was the product
formed in the greatest proportion upon reaction with
DNA, while dG-C8-2,6-diMeA was the minor adduct of
the four. This unique behavior also contrasts markedly
with the characteristic pattern of adduct formation from
known aromatic amine carcinogens, which tend to yield
dG-C8-Ar adducts as the major lesions in vivo (6-8).
Nonetheless, exceptions to this typical behavior have
been reported; for instance, N-hydroxy-1-naphthylamine
predominantly forms dG adducts through the exocyclic
6
2
6
O atom (36), while N -dG and N -dA species have been
characterized as minor adducts from 2-naphthylamine
(37).
The peculiar multiplicity of adduct structures obtained
from N-AcO-2,6-diMeA reveals a low selectivity of the
reactive intermediate toward specific nucleophilic sites
in the purine bases. Extensive work on the mechanisms
of hydrolysis of N-acyloxyarylamines and N-acyl-N-
acyloxyarylamines (38-41, and references therein) has
(34). The substrate concentrations ranged from 1.0 to 200
µM in the incubation mixtures, with the higher concen-

1
72 Chem. Res. Toxicol., Vol. 14, No. 2, 2001
Gon c¸ alves et al.
Ta ble 2. Kin etic P a r a m eter s Obta in ed for th e P h osp h or yla tion of d G3′p a n d d G3′p Ad d u cts
Vmax
kcat
kcat/Km
substratea
dG3′p
Km (µM)b
(µM min-1)b
(min-1)c
(µM-1 min-1)d
8.4 ( 0.93
15.9 ( 2.5
3.4 ( 1.6
1.9 ( 0.12
0.20 ( 0.016
0.096 ( 0.013
0.60 ( 0.069
160
17
8.2
51
19
dG3′p-C8-2,6-diMeA
1.1
2.4
7.9
2
dG3′p-N -2,6-diMeA
2
dG3′p-N -BPDE
6.5 ( 1.4
a
Substrate concentrations ranged from 1.0 to 200 µM in the incubation mixtures. The ATP concentration was 240 µM. b Calculated for
the Michaelis-Menten equation using the weighted least-squares (relative weights) fitting program Leonora (34). Both Km and Vmax are
c
apparent values and are presented ( the standard error. Apparent turnover number, defined as Vmax/[E]0. The PNK concentration in
the incubation mixture was 150 milliunits/µL (12 nM). d Apparent specificity constant.
indicated that these reactions proceed by N-O bond
heterolysis, leading to a nitrenium/carbenium ion inter-
mediate as the cationic component of a tight ion pair,
which may undergo diffusion to a solvent-separated ion
pair. In general, trapping of the cationic species by
external nucleophiles or internal return of the carboxy-
late anion, followed by nucleophilic attack on a resulting
cyclohexadiene derivative, has been invoked to rational-
ize the structures of solvolysis products, including those
obtained from carboxylate esters of N-OH-2,6-diMeA (40,
used methodologies for the detection and quantitation of
carcinogen-DNA adducts (26-28). However, the preci-
sion of measured adduct levels is highly dependent on a
number of assumptions, including the complete digestion
of the DNA, the total recovery of the adducts during the
enrichment and chromatography steps, and an equiva-
lent efficiency of conversion of normal and adducted
nucleotide 3′-phosphates to their corresponding 3′,5′-
bisphosphates. Despite potential problems with these
assumptions, these premises underlie the widespread use
of the relative adduct labeling (RAL) approach for
quantitative purposes (42). Although some reports have
suggested that certain aromatic and bulky adducts may
be better substrates than normal nucleotides for PNK
(43, 44), a substantial body of evidence has indicated that
a variety of structurally different adducts are labeled
with low efficiencies (reviewed in ref 29). Arylamine-
DNA adduct levels, in particular, seem to be consistently
underestimated by RAL, as compared to other DNA
adduct detection methodologies (45, 46). When the results
of our previous comparative kinetic study of the PNK-
4
1). The observed substitution in the arylamine ring for
2
6
three of the adducts (dG-N -2,6-diMeA, dG-O -2,6-diMeA,
and dA-N -2,6-diMeA), along with the unusual formation
6
of a predominant dA adduct upon reaction with DNA, is
fully consistent with the involvement of a nitrenium/
carbenium ion intermediate in the rate-limiting step. The
occurrence of aryl ring substitution para to the arylamine
nitrogen instead of the more common ortho position (6-
8
), which was prevented by the presence of the two ortho
methyl substituents, further supports this interpretation.
It seems plausible that the steric hindrance imposed on
the arylamine nitrogen by the two ortho methyl substit-
uents may have been the factor determining a decreased
extent of aryl N-substitution (i.e., formation of dG-C8-
32
catalyzed P-postlabeling of dG3′p and a series of dG3′p-
C8-Ar adducts (29) are taken into account, this under-
estimation appears to be due, in part, to the lower
efficiency of labeling of these adducts compared to that
of the normal nucleotide. In particular, we were able to
demonstrate that optimal labeling of dG3′p-C8-Ar ad-
ducts required ATP concentrations higher than those
2
,6-diMeA), while favoring nucleophilic attack at the less
hindered para position. Thus, the ability of 2,6-diMeA
to yield multiple adducts may stem, for the most part,
from the steric constraints in the vicinity of the nitrogen
atom.
32
used in current P-postlabeling protocols (29). In the
The possible implications of the formation of multiple
,6-diMeA-DNA adducts to the biological response elic-
study presented here, we have used the same methodol-
ogy to investigate whether the site of attachment of the
same arylamine fragment to the nucleotide (dG3′p)
affects the labeling efficiency. Toward this purpose, we
conducted a comparative kinetic analysis of the PNK-
catalyzed phosphorylation of dG3′p-C8-2,6-diMeA and
2
ited by this arylamine have yet to be established. As
indicated above, there is substantial evidence (12, 13)
that 2,6-diMeA undergoes activation in vivo through
N-hydroxylation. N-Hydroxyalkylanilines substituted ortho
to the nitrogen tend to be more mutagenic than their
analogues not bearing ortho substituents (18, 19), which
has been interpreted as implying that ortho-substituted
dG-C8-Ar adducts may have higher intrinsic mutage-
nicities (19). However, while N-OH-2,6-diMeA is the most
mutagenic of the entire series of N-(hydroxymethyl)-,
N-(hydroxydimethyl)-, and N-(hydroxyethyl)anilines (18,
2
dG3′p-N -2,6-diMeA. Under the same experimental con-
ditions, both adducts were labeled with much lower
efficiencies (18- and 8-fold lower, respectively) than
dG3′p, as indicated by the corresponding apparent speci-
ficity constants [kcat/Km (Table 2)]. This observation
suggests that, under kinetically controlled conditions,
normal nucleotides will be labeled preferentially com-
2
1
9), this property does not seem to result from a higher
efficiency of formation of the C8-substituted dG adduct
19, 23). Instead, it appears that one or more of the
pared to dG3′p-N -Ar adducts, which is similar to our
previous observation with dG3′p-C8-Ar adducts (29).
Nonetheless, the fact that the labeling efficiency of dG3′p-
(
additional 2,6-diMeA adducts may contribute signifi-
cantly to the increased mutagenicity of N-OH-2,6-diMeA
compared to other N-hydroxyalkylanilines. While specific
adducts are expected to have characteristic mutagenici-
ties and mutational spectra, the relative contributions
of each adduct to the toxicological effects induced by 2,6-
diMeA are presently unknown.
N -2,6-diMeA was approximately 2-fold higher than that
2
of dG3′p-C8-2,6-diMeA confirms that the site of attach-
ment of the aromatic fragment to the nucleotide affects
the specificity of PNK toward the adducted substrate. In
addition, it should be noted that all the dG3′p-C8-Ar
adducts analyzed in our previous work (29) were labeled
2
with efficiencies lower than that of dG3′p-N -2,6-diMeA,
3
2
2
P -P ostla belin g Stu d ies. Due to its high sensitivity
and versatility, 32P-postlabeling is one of the most widely
which suggests that dG3′p-N adducts may be intrinsi-
cally better substrates for PNK than dG3′p-C8 adducts.

2
,6-Dimethylaniline-DNA Adducts
Chem. Res. Toxicol., Vol. 14, No. 2, 2001 173
This is further substantiated by the results obtained with
the bulky dG3′p-N -BPDE adduct. Despite being labeled
Carcinogenesis and Mutagenesis I (Cooper, C. S., and Grover, P.
L., Eds.) Vol. 94/I, pp 267-325, Springer-Verlag, Heidelberg,
Germany.
2
with lower efficiency than the parent nucleotide, this
adduct showed the best labeling efficiency of all the
adducts that were tested, both in this study and in our
previous report (29), as measured by the corresponding
apparent specificity constant (Table 2). The reasons
underlying this increased affinity of the bulkier sub-
(
7) Kadlubar, F. F. (1994) DNA adducts of carcinogenic aromatic
amines. In DNA Adducts: Identification and Biological Signifi-
cance (Hemminki, K., Dipple, A., Shuker, D. E. G., Kadlubar, F.
F., Segerb a¨ ck, D., and Bartsch, H., Eds.) IARC Scientific Publica-
tion No. 125, pp 199-216, International Agency for Research on
Cancer, Lyon, France.
(8) Delclos, K. B., and Kadlubar, F. F. (1997) Carcinogenic aromatic
amines and amides. In Comprehensive Toxicology, Volume 12:
Chemical Carcinogens and Anticarcinogens (Bowden, G. T., and
Fischer, S. M., Eds.) pp 141-170, Elsevier Science, Ltd., Oxford,
U.K.
(9) Ringe, D., Turesky, R. J ., Skipper, P. L., and Tannenbaum, S. R.
(1988) Structure of the single stable hemoglobin adduct formed
by 4-aminobiphenyl in vivo. Chem. Res. Toxicol. 1, 22-24.
10) McLean, S., Starmer, G. A., and Thomas, J . (1969) Methaemo-
globin formation by aromatic amines. J . Pharm. Pharmacol. 21,
441-450.
(11) Harrison, J . H., J r., and J ollow, D. J . (1987) Contribution of
aniline metabolites to aniline-induced methemoglobinemia. Mol.
Pharmacol. 32, 423-431.
2
strate, and of N adducts in general, toward the enzyme
are not clear. Nonetheless, it is conceivable that the
location of the aromatic fragment in the adduct may favor
noncovalent stacking interactions with amino acid resi-
dues in the active site of the enzyme. We should also note
that the labeling efficiency obtained for dG3′p-C8-2,6-
diMeA was on the same order of magnitude previously
determined for dG3′p-C8-4-ABP, but 5-25-fold higher
than that obtained for other dG3′p-C8-Ar adducts from
alkylanilines (29). This result was somewhat surprising,
since we have observed consistently lower labeling ef-
ficiencies with adducts derived from ortho-substituted
anilines. Although the reasons for the better labeling of
dG3′p-C8-2,6-diMeA are obscure, it is possible that the
steric hindrance in the vicinity of the arylamine nitrogen
may enhance the stability of the adduct under the harsh
alkaline conditions used for the labeling reaction.
(
(
12) Bryant, M. S., Vineis, P., Skipper, P. L., and Tannenbaum, S. R.
1988) Hemoglobin adducts of aromatic amines: associations with
smoking status and type of tobacco. Proc. Natl. Acad. Sci. U.S.A.
5, 9788-9791.
13) Bryant, M. S., Simmons, H. F., Harrell, R. E., and Hinson, J . A.
1994) 2,6-Dimethylaniline-hemoglobin adducts from lidocaine in
humans. Carcinogenesis 15, 2287-2290.
(
8
(
(
(14) Burne, D., and Doughty, A. (1964) Methaemoglobinaemia follow-
In conclusion, this 32P-postlabeling study confirms our
ing lignocaine, Lancet ii, 971.
(
15) Weiss, L. D., Generalovich, T., Heller, M. B., Paris, P. M., Stewart,
R. D., Kaplan, R. M., and Thompson, D. R. (1987) Methemoglobin
levels following intravenous lidocaine administration. Ann. Emerg.
Med. 16, 323-325.
(16) Zimmer, D., Mazurek, J ., Petzold, G., and Bhuyan, B. K. (1980)
Bacterial mutagenicity and mammalian cell DNA damage by
several substituted anilines. Mutat. Res. 77, 317-326.
previous observation that under kinetically controlled
conditions, normal nucleotides are labeled preferentially
to adducted substrates, which may lead to the underes-
3
2
timation of adduct levels in typical P-postlabeling
experiments. Nonetheless, the results demonstrate that
compared to their C8-substituted analogues, aromatic
adducts through the N2 atom of dG are labeled with
higher efficiencies.
(17) Kugler-Steigmeier, M. E., Friederich, U., Graf, U., Lutz, W. K.,
Maier, P., and Schlatter, Ch. (1989) Genotoxicity of aniline
derivatives in various short-term tests. Mutat. Res. 211, 279-
2
89.
(
18) Nohmi, T., Yoshikawa, K., Nakadate, M., Miyata, R., and Ishidate,
M., J r. (1984) Mutations in Salmonella typhimurium and inac-
tivation of Bacillus subtilis transforming DNA induced by phen-
ylhydroxylamine derivatives. Mutat. Res. 136, 159-168.
Ack n ow led gm en t. This work was supported in part
by the NATO (Collaborative Research Grant 960165).
L.L.G. received financial assistance from FCT (Portugal)
and the Luso-American Foundation for Development
(
19) Marques, M. M., Mourato, L. L. G., Amorim, M. T., Santos, M.
A., Melchior, W. B., J r., and Beland, F. A. (1997) Effect of
substitution site upon the oxidation potentials of alkylanilines,
the mutagenicities of N-hydroxyalkylanilines, and the conforma-
tions of alkylaniline-DNA adducts. Chem. Res. Toxicol. 10, 1266-
(FLAD). We thank J . Pat Freeman for the mass spectra
and Cindy Hartwick for helping to prepare the manu-
script.
1
274.
(
20) Famulok, M., and Boche, G. (1989) Formation of N-(deoxygua-
nosin-8-yl)aniline in the in vitro reaction of N-acetoxyaniline with
deoxyguanosine and DNA. Angew. Chem., Int. Ed. 28, 468-469.
Refer en ces
(
1) International Agency for Research on Cancer (1993) 2,6-Di-
methylaniline (2,6-xylidine). IARC Monographs on the Evaluation
of Carcinogenic Risks to Humans. Occupational Exposures of
Hairdressers and Barbers and Personal Use of Hair Colourants;
Some Hair Dyes, Cosmetic Colourants, Industrial Dyestuffs and
Aromatic Amines, Vol. 57, pp 323-335, International Agency for
Research on Cancer, Lyon, France.
(21) Meier, C., and Boche, G. (1990) N-Aryl-O-(R-aminoacyl)hydroxyl-
amine: Modellreaktionen mit Desoxyguanosin, Guanosin und 5′-
Guanosinmonophosphat zur Aktivierung monocyclischer aroma-
tischer Amine (z.B. Phenacetin) zu ultimaten Carcinogenen
[N-Aryl-O-(R-aminoacyl)hydroxylamines: Model reactions with
deoxyguanosine, guanosine, and 5′-guanosine monophosphate for
the activation of monocyclic aromatic amines (e.g. phenacetin) into
ultimate carcinogens]. Chem. Ber. 123, 1699-1705.
(
2) Patrianakos, C., and Hoffmann, D. (1979) Chemical studies on
tobacco smoke LXIV. On the analysis of aromatic amines in
cigarette smoke. J . Anal. Toxicol. 3, 150-154.
(
22) Marques, M. M., Mourato, L. L. G., Santos, M. A., and Beland,
F. A. (1996) Synthesis, characterization, and conformational
analysis of DNA adducts from methylated anilines present in
tobacco smoke. Chem. Res. Toxicol. 9, 99-108.
(
3) Luceri, F., Pieraccini, G., Moneti, G., and Dolara, P. (1993)
Primary aromatic amines from side-stream cigarette smoke are
common contaminants of indoor air. Toxicol. Ind. Health 9, 405-
(
23) Beyerbach, A., Farmer, P. B., and Sabbioni, G. (1996) Synthesis
and analysis of DNA adducts of arylamines. Biomarkers 1, 9-20.
24) Lakshman, M. K., Keeler, J . C., Hilmer, J . H., and Martin, J . Q.
4
13.
(
(
4) Keenaghan, J . B., and Boyes, R. N. (1972) The tissue distribution,
metabolism and excretion of lidocaine in rats, guinea pigs, dogs
and man. J . Pharmacol. Exp. Ther. 180, 454-463.
(
1999) Palladium-catalyzed C-N bond formation: facile and
6
general synthesis of N -aryl 2′-deoxyadenosine analogues. J . Am.
Chem. Soc. 121, 6090-6091.
(5) U.S. National Toxicology Program (1990) Toxicology and carcino-
genesis studies of 2,6-xylidine (2,6-dimethylaniline) (CAS No. 87-
(25) Branco, P. S., Antunes, A. M. M., Marques, M. M., Chiarelli, M.
P., Lobo, A. M., and Prabhakar, S. (1999) New syntheses of DNA
adducts from methylated anilines present in tobacco smoke.
Chem. Res. Toxicol. 12, 1223-1233.
6
2-7) in Charles River CD rats (feed studies), Technical Report
Series No. 278, National Toxicology Program, Research Triangle
Park, NC.
(
26) Beach, A. C., and Gupta, R. C. (1992) Human biomonitoring and
3
2
(
6) Beland, F. A., and Kadlubar, F. F. (1990) Metabolic activation
and DNA adducts of aromatic amines and nitroaromatic hydro-
carbons. In Handbook of Experimental Pharmacology. Chemical
the P-postlabeling assay. Carcinogenesis 13, 1053-1074.
(27) Phillips, D. H., and Castegnaro, M. (1999) Standardization and
validation of DNA adduct postlabelling methods: report of

1
74 Chem. Res. Toxicol., Vol. 14, No. 2, 2001
Gon c¸ alves et al.
interlaboratory trials and production of recommended protocols.
Mutagenesis 14, 301-315.
(38) Novak, M., VandeWater, A. J ., Brown, A. J ., Sanzenbacher, S.
A., Hunt, L. A., Kolb, B. A., and Brooks, M. E. (1999) Hydrolysis
reactions of N-sulfonatooxy-N-acetyl-1-aminonaphthalene and
N-sulfonatooxy-N-acetyl-2-aminonaphthalene: limited correlation
of nitrenium ion azide/solvent selectivities with mutagenicities
of the corresponding amines. J . Org. Chem. 64, 6023-6031.
(39) Novak, M., and Lin, J . (1999) Correlation of azide/solvent
selectivities for nitrenium ions with ab initio hydration energies:
understanding the kinetic lability of nitrenium ions in aqueous
solution. J . Org. Chem. 64, 6032-6040.
(
(
(
28) Phillips, D. H., Farmer, P. B., Beland, F. A., Nath, R. G., Poirier,
M. C., Reddy, M. V., and Turteltaub, K. W. (2000) Methods of
DNA adduct determination and their application to testing
compounds for genotoxicity. Environ. Mol. Mutagen. 35, 222-233.
3
2
29) Mourato, L. L. G., Beland, F. A., and Marques, M. M. (1999) P-
Postlabeling of N-(deoxyguanosin-8-yl)arylamine adducts: a com-
parative study of labeling efficiencies. Chem. Res. Toxicol. 12,
6
61-669.
30) Dipple, A. (1994) Reactions of polycyclic aromatic hydrocarbons
with DNA. In DNA Adducts: Identification and Biological
Significance (Hemminki, K., Dipple, A., Shuker, D. E. G., Kad-
lubar, F. F., Segerb a¨ ck, D., and Bartsch, H., Eds.) IARC Scientific
Publication No. 125, pp 107-129, International Agency for
Research on Cancer, Lyon, France.
(40) Fishbein, J . C., and McClelland, R. A. (1995) Cyclohexadienes
from the rearrangement of O-aroyl-N-acetyl-N-(2,6-dimethylphen-
yl)hydroxylamines. Reaction in aqueous solution to meta- and
para-substituted 2,6-dimethylacetanilides. J . Chem. Soc., Perkin
Trans. 2, 653-662.
(41) Fishbein, J . C., and McClelland, R. A. (1995) Nature of nitrenium:
carboxylate ion pair intermediates in the hydrolysis of O-aroyl-
N-acetyl-N-(2,6-dimethylphenyl)hydroxylamines. J . Chem. Soc.,
Perkin Trans. 2, 663-671.
(
31) Breitmaier, E., and Voelter, W. (1987) Carbon-13 NMR Spectros-
copy. High-Resolution Methods and Applications in Organic
Chemistry and Biochemistry, 3rd ed., pp 1-515, VCH, Weinheim,
Germany.
(42) Reddy, M. V., Gupta, R. C., Randerath, E., and Randerath, K.
(32) Pulkrabek, P., Leffler, S., Weinstein, I. B., and Grunberger, D.
(1984) 32P-Postlabeling test for covalent DNA binding of chemicals
(1977) Conformation of DNA modified with a dihydrodiol epoxide
in vivo: application to a variety of aromatic carcinogens and
methylating agents. Carcinogenesis 5, 231-243.
derivative of benzo[a]pyrene. Biochemistry 16, 3127-3132.
33) Heflich, R. H., Morris, S. M., Beranek, D. T., McGarrity, L. J .,
Chen, J . J ., and Beland, F. A. (1986) Relationships between the
DNA adducts and the mutations and sister-chromatid exchanges
produced in Chinese hamster ovary cells by N-hydroxy-2-ami-
nofluorene, N-hydroxy-N′-acetylbenzidine and 1-nitrosopyrene.
Mutagenesis 1, 201-206.
(
(43) Randerath, E., Agrawal, H. P., Weaver, J . A., Bordelon, C. B.,
and Randerath, K. (1985) 32P-Postlabeling analysis of DNA
adducts persisting for up to 42 weeks in the skin, epidermis and
dermis of mice treated topically with 7,12-dimethylbenz[a]an-
thracene. Carcinogenesis 6, 1117-1126.
(44) Reddy, M. V., and Randerath, K. (1987) 32P-Analysis of DNA
(
34) Cornish-Bowden, A. (1995) Analysis of Enzyme Kinetic Data, pp
adducts in somatic and reproductive tissues of rats treated with
1
-198, Oxford University Press, New York.
the anticancer antibiotic, mitomycin C. Mutat. Res. 179, 75-88.
(
35) Edwards, C., Boche, G., Steinbrecher, T., and Scheer, S. (1997)
Synthesis of 2-substituted 2′-deoxyguanosines and 6-O-allylgua-
nines via the activation of C-2 by a trifluoromethanesulfonate
group. J . Chem. Soc., Perkin Trans. 1, 1887-1893.
(
45) Phillips, D. H., and Castegnaro, M. (1993) Results of an inter-
32
laboratory trial of P-postlabelling. In Postlabelling Methods for
Detection of DNA Adducts (Phillips, D. H., Castegnaro, M., and
Bartsch, H., Eds.) IARC Scientific Publication No. 124, pp 35-
(
36) Kadlubar, F. F., Miller, J . A., and Miller, E. C. (1978) Guanyl
4
9, International Agency for Research on Cancer, Lyon, France.
6
6
O -arylamination and O -arylation of DNA by the carcinogen
N-hydroxy-1-naphthylamine. Cancer Res. 38, 3628-3638.
(
46) Beland, F. A., Doerge, D. R., Churchwell, M. I., Poirier, M. C.,
Schoket, B., and Marques, M. M. (1999) Synthesis, characteriza-
tion, and quantitation of a 4-aminobiphenyl-DNA adduct stan-
dard. Chem. Res. Toxicol. 12, 68-77.
(37) Kadlubar, F. F., Unruh, L. E., Beland, F. A., Straub, K. M., and
Evans, F. E. (1980) In vitro reaction of the carcinogen, N-hydroxy-
2
2
-naphthylamine, with DNA at the C-8 and N atoms of guanine
6
and at the N atom of adenine. Carcinogenesis 1, 139-150.
TX0002031