N-Nitrosobenzylmethylamine is activated to a DNA benzylating agent in rats

Article abstract of DOI:10.1021/tx9601014

The ability of rat tissues to activate the esophageal carcinogen, N- nitrosobenzylmethylamine (NBzMA), to a DNA benzylating intermediate was investigated. [3-³H]NBzMA was prepared and given to male F344 rats. Tissues were harvested 4 h after treatment, and DNA was isolated. HPLC analysis with radiochemical detection of chemical and enzymatic hydrolysates of DNA from liver and lung revealed the formation of benzyl adducts. Benzyl alcohol, N²- benzylguanine, 3-benzyladenine, N⁶-benzyladenine, and 7-benzylguanine were the major radioactive components in the hydrolysates. An unknown adduct was also observed. The adduct distribution was similar to that observed in [3- ³H]benzylnitroseurea ([3-³H]BzNU)treated calf thymus DNA. However, enzymatic hydrolysates of [3-³H]BzNU-treated DNA also contained significant levels of O⁶-benzyl-2'-deoxyguanosine (O⁶-BzdG). This radioactive adduct disappeared upon incubation of the DNA with a crude preparation of the repair protein, O⁶-alkylguanine-DNA alkyltransferase isolated from rat liver. These data provide evidence that O⁶-BzdG is probably rapidly repaired in vivo. No benzylation of esophageal mucosal DNA was detected. The level of DNA benzylation observed in tissues from [3-³H]NBzMA-treated rats was several orders of magnitude lower than the level of DNA methylation in these same tissues. Therefore, these data indicate that DNA benzylation plays a minor role, if any, in the carcinogenic activity of NBzMA.

Full text of DOI:10.1021/tx9601014

Chem. Res. Toxicol. 1997, 10, 19-26
19
Articles
N-Nitr osoben zylm eth yla m in e Is Activa ted to a DNA
Ben zyla tin g Agen t in Ra ts
Lisa A. Peterson
Division of Chemical Carcinogenesis, American Health Foundation, 1 Dana Road,
Valhalla, New York 10595
Received J une 20, 1996X
The ability of rat tissues to activate the esophageal carcinogen, N-nitrosobenzylmethylamine
(NBzMA), to a DNA benzylating intermediate was investigated. [3-3H]NBzMA was prepared
and given to male F344 rats. Tissues were harvested 4 h after treatment, and DNA was
isolated. HPLC analysis with radiochemical detection of chemical and enzymatic hydrolysates
of DNA from liver and lung revealed the formation of benzyl adducts. Benzyl alcohol,
N2-benzylguanine, 3-benzyladenine, N6-benzyladenine, and 7-benzylguanine were the major
radioactive components in the hydrolysates. An unknown adduct was also observed. The
adduct distribution was similar to that observed in [3-3H]benzylnitrosourea ([3-3H]BzNU)-
treated calf thymus DNA. However, enzymatic hydrolysates of [3-3H]BzNU-treated DNA also
contained significant levels of O6-benzyl-2′-deoxyguanosine (O6-BzdG). This radioactive adduct
disappeared upon incubation of the DNA with a crude preparation of the repair protein,
O6-alkylguanine-DNA alkyltransferase isolated from rat liver. These data provide evidence
that O6-BzdG is probably rapidly repaired in vivo. No benzylation of esophageal mucosal DNA
was detected. The level of DNA benzylation observed in tissues from [3-3H]NBzMA-treated
rats was several orders of magnitude lower than the level of DNA methylation in these same
tissues. Therefore, these data indicate that DNA benzylation plays a minor role, if any, in the
carcinogenic activity of NBzMA.
In tr od u ction
N-Nitrosobenzylmethylamine (NBzMA)1 is a highly
selective esophageal carcinogen in laboratory animals (1-
3). This compound is a possible human esophageal
carcinogen (4, 5). Carcinogenicity is believed to be
initiated by cytochrome P450-catalyzed hydroxylation
adjacent to the nitroso group; this reaction forms me-
tabolites that decompose to DNA reactive compounds.
Since NBzMA is an asymmetric nitrosamine, it has two
potential activation pathways (Figure 1). Benzyl hy-
droxylation leads to a methylating agent, whereas methyl
hydroxylation generates a benzylating agent (6, 7). The
available data support the formation of O6-methylgua-
nine (O6-mG) as the important step in tumor initiation
by this compound. High levels of methyl adducts are
detected in target tissues relative to nontarget tissues
F igu r e 1. Proposed activation pathways of NBzMA and BzNU.
(8-10). Furthermore, O6-mG derived from NBzMA
persists longer in the esophagus relative to nontarget
tissues, such as liver, following a single dose (11, 12). In
addition, mutations detected in activated H-ras and p53
oncogenes (13-15) are consistent with the formation of
O6-mG, not O6-benzylguanine (O6-BzG; 16). Moreover,
the mutational frequency of O6-mG was 2 times higher
than that of O6-BzG in Rat4 TK- cells (16).
The role of the DNA benzylation pathway in NBzMA-
induced esophageal carcinogenesis has not been exten-
sively investigated. The only reported attempt to mea-
sure benzyl adducts was performed in Wistar rats treated
with [methylene-14C]NBzMA (iv, 0.017 nmol/kg) of low
specific activity (11.3 mCi/mmol; 9). No benzyl adducts
were detected. The limits of detection were 0.5-1
benzylated bases in 106 bases. However, liver mi-
crosomes catalyze both benzylation and methylation
pathways as indicated by formaldehyde and benzalde-
hyde formation (6, 17-21). The relative amount of these
two metabolites varied with the concentration of NBzMA
X Abstract published in Advance ACS Abstracts, November 15, 1996.
1
Abbreviations: AGT, O6-alkylguanine-DNA alkyltransferase;
3-BzA, 3-benzyladenine; 7-BzA, 7-benzyladenine; 3-BzdC, 3-benzyl-2′-
deoxycytidine; 7-BzG, 7-benzylguanine; BzNU; N-benzyl-N-nitrosourea;
MNU, N-methyl-N-nitrosourea; N2-BzdG, N2-benzyl-2′-deoxygua-
nosine; N2-BzG, N2-benzylguanine; N6-BzA, N6-benzyladenine; N6-
BzdA, N6-benzyl-2′-deoxyadenosine; 7-mG, 7-methylguanine; NBzMA,
N-nitrosobenzylmethylamine; O6-BzdG, O6-benzyl-2′-deoxyguanosine;
O6-BzG, O6-benzylguanine; O6-mG, O6-methylguanine; SDS, sodium
dodecyl sulfate.
S0893-228x(96)00101-4 CCC: $14.00 © 1997 American Chemical Society

20 Chem. Res. Toxicol., Vol. 10, No. 1, 1997
Peterson
(3-Br om oben zyl)m eth ylam in e Hydr och lor ide. The prepa-
ration of (3-bromobenzyl)methylamine hydrochloride was adapted
used and the source of microsomes. In esophageal
microsomes, there was at least 100 times more benzal-
dehyde than formaldehyde formed (17), indicating that
the methylation pathway likely predominates in the
target tissue.
from
a published procedure (31). Trifluoromethyl sulfonic
anhydride (3.2 g, 0.011 mol) in CH2Cl2 (11 mL) was added
dropwise to 3-bromobenzylamine (4.2 g, 0.025 mol) in 35 mL of
CH2Cl2 at 0 °C. The reaction was stirred for 1 h at room
temperature, washed two times with 1.0 N HCI (20 mL) and
three times with water (20 mL), and dried over MgSO4. The
resulting oil was reacted with iodomethane (0.7 mL, 0.011 mol)
in the presence of potassium carbonate (1.6 g, 0.016 mol) in
acetone (25 mL). The acetone was removed under reduced
pressure, and the product was added dropwise as an anhydrous
diethyl ether solution (11 mL) to a stirred suspension of lithium
aluminum hydride (1.4 g, 0.037 mol). The reaction was stirred
for 2 days at room temperature. Then, water (4.4 mL) was
added dropwise, followed by 15% NaOH (4.4 mL) and, finally,
additional water (13 mL). The ether layer was filtered, dried
over MgSO4, and concentrated to yield an oil (1.5 g). HCl in
ether (0.1 N) was added to the oil, and the resulting white
precipitate was collected by filtration and washed with ether
(1.5 g). NMR of the product demonstrated that we had obtained
(3-bromobenzyl)methylamine that contained approximately 20%
benzylmethylamine. This side product resulted presumably
from reductive debromination by lithium aluminum hydride. 1H
NMR (CDCl3) δ 7.5 (s, 1H, C2), 7.4 (d, 1H, C4), 7.3 (m, 1H, C6),
7.2 (dd, 1H, C5) 3.8 (s, 2H, CH2N), and 2.5 (s, 3H, NCH3).
[3-3H]Ben zylm eth yla m in e. (3-Bromobenzyl)methylamine
hydrochloride was converted to its free base and dissolved in
ethanol containing 5% triethylamine. The reductive tritiation
was performed in the presence of tritium gas and 5-10% Pd/
CaCO3 for 1-2 h at atmospheric pressure and room tempera-
ture. The catalyst was removed by filtration. Then 0.1 volume
of dilute HCl was added and the solvent was removed under
reduced pressure. The product was stored in ethanol at -80
°C (sp act., approximately 8.3 Ci/mmol).
[3-3H]NBzMA. [3-3H]Benzylmethylamine hydrochloride (200
mCi, 6.1 mg, 0.0386 mmol) was dissolved in 30% acetic acid (1
mL). Sodium nitrite (16.4 mg, 0.240 mmol) was added. One
and a half hours later, the reaction mixture was extracted with
CH2Cl2 (3 × 3 mL). The organic phase was washed with water,
dried over MgSO4 (anhyd), and concentrated under reduced
pressure. The reaction product was purified using a short silica
gel column eluted with CH2Cl2. The fractions containing
product were combined, dried over MgSO4, and stored at -80
°C in CH2Cl2. The identity and purity were confirmed by
coelution with standards on reverse-phase HPLC (Phenomenex,
Torrance, CA, Bondclone C18 column, 3300 × 3.9 mm) linked
to radioflow detection (FLO-one/Beta, Radiomatics Instruments,
Tampa, FL), using 3% acetic acid (solvent A) and 95% methanol
(solvent B). The column was eluted with a linear gradient from
100% solvent A to 50% solvent A/50% solvent B over 50 min.
Yield: 93.7 mCi (47%, 98% radiochemically pure, sp act., 5.2
mCi/mmol).
[3-3H]BzNU. [3-3H]Benzylamine hydrochloride (22 mCi, 3.6
mg, 25 µmol) was dissolved in 500 µL of water. Potassium
cyanate (4.5 mg, 56 µmol) and 0.1 N H2SO4 (125 µL) were added,
and the reaction was heated at 110 °C with stirring for 15 min.
After the reaction was placed on ice, 5 N H2SO4 (1 mL) was
added. Then an aqueous solution of sodium nitrite (100 µL, 3.75
mg, 55 µmol) was added slowly over 5 min. The reaction was
stirred at 0 °C for 20 min, and then extracted with ether (8 ×
1 mL). The ether layer was dried over MgSO4 and concentrated
to 4 mL. It was then applied to a Florisil column (1.5 × 13 cm).
The product was eluted with ether. The fractions containing
product were pooled, concentrated under reduced pressure, and
dissolved in ethanol (5 mL). The radioactivity coeluted with
cold standard on TLC (20% ethyl acetate in hexane). Yield: 3.3
mCi (15%, sp act., approximately 880 mCi/mmol).
DNA benzylation may be involved in the carcinogenic
activity of NBzMA. The benzylating agent, N-benzyl-N-
nitrosourea (BzNU), is a locally acting carcinogen in
animals (22). Rat liver S9 can activate NBzMA to a
mutagen in TA100 but not TA1535 Salmonella strains
(23). This strain selectivity parallels that observed for
BzNU but not the methylating agent, N-methyl-N-
nitrosourea (MNU; 23). BzNU as well as the benzylating
N-(acetoxymethyl)-N-benzylnitrosamine are mutagenic in
a variety of in vitro mutagenicity assays (23-25). The
mutagenic activity of these benzylating species suggests
that phenylmethanediazohydroxide is capable of alky-
lating DNA. Therefore, if benzylation of DNA occurs in
vivo, it could potentially contribute to the carcinogenic
activity of NBzMA.
DNA adducts generated by benzylating agents have
not been extensively characterized. Reaction of BzNU
with guanosine produced 7-, N2-, and O6-benzylguanosine
(26). This compound also alkylated the N6- and 1-posi-
tions of adenosine (26). In addition, 7-benzylguanine
(7-BzG) has been identified as a liver DNA adduct
isolated from [14C]BzNU-treated Sprague-Dawley rats
(27).
The possibility that NBzMA is activated to a DNA
benzylating agent in vivo was re-examined by treating
rats with [3-3H]NBzMA of high specific activity and
characterizing the adducts present in DNA isolated from
target and nontarget tissues. As described below, DNA
benzylation does occur in rats treated with NBzMA.
However, the levels of DNA benzylation are orders of
magnitude lower than those of DNA methylation. The
adduct distribution in the DNA from these tissues was
compared to that observed in [3-3H]BzNU-treated calf
thymus DNA. The major difference was the absence of
O6-benzyl-2′-deoxyguanosine (O6-BzdG) in the in vivo
benzylated DNA, a likely result of efficient DNA repair.
Exp er im en ta l P r oced u r es
Ca u tion : NBzMA and BzNU are carcinogens in experimental
animals.
Ma ter ia ls. NBzMA was obtained from the NCI Carcinogen
Reference Standard Repository (Midwest Research Institute,
Kansas City, MO). Phenol saturated with 10 mM Tris (pH 8)
and 1 mM EDTA, calf thymus DNA, nuclease P1, phosphodi-
esterase I (type VII), DNase I (type II), proteinase K, RNase K,
and RNase T1 were purchased from Sigma Chemical Co. (St.
Louis, MO). Alkaline phosphatase was purchased from
Boehringer Mannheim (Indianapolis, IN). [3-3H]Benzylamine
hydrochloride was purchased from Amersham (Arlington Heights,
IL). All other chemicals were obtained from Aldrich Chemical
Co. (Milwaukee, WI). 3-Benzyladenine (3-BzA), 7-benzylad-
enine (7-BzA), 3-benzyl-2′-deoxycytidine (3-BzdC), N2-benzyl-
2′-deoxyguanosine(N2-BzdG), N6-benzyl-2′-deoxyadenosine (N6-
BzdA) standards were a generous gift from Dr. Robert Moschel,
NCI-FCRDC, Frederick, MD. N2-Benzylguanine (N2-BzG) and
N6-benzyladenine (N6-BzA) were generated by heating the
corresponding 2′-deoxynucleoside in 0.1 N HCl. O6-BzG and O6-
BzdG were synthesized according to literature procedures (28,
29). 7-BzG was produced by reacting benzyl bromide with 2′-
deoxyguanosine in DMF: 1H NMR (DMSO-d6) δ 10.75 (s, N1-
H), 8.1 (s, C8-H), 7.4 (m, aromatic H), 6.1 (br s, 2-NH2), and 5.4
(s, CH2). Rat liver O6-alkylguanine-DNA alkyltransferase
(AGT) was prepared as previously described (30).
[3-3H]BzNU-Tr ea ted DNA. [3-3H]BzNU (3.3 mCi, 0.00375
mmol) was reacted with calf thymus DNA (50 mg) in 50 mM
sodium cacodylate (pH 6.0) containing 1 mM EDTA (10 mL) at
room temperature overnight. The DNA was precipitated from
the solution by the addition of 2 M NaCl (2 mL) and ice cold

Benzylation of DNA in Vivo and in Vitro
Chem. Res. Toxicol., Vol. 10, No. 1, 1997 21
ethanol (15 mL). It was redissolved in 100 mM sodium
phosphate (pH 7.0, 2.3 mg/mL) and dialyzed against the same
buffer overnight at 4 °C. Following the addition of 2 M NaCl
(3 mL) and 20 mL of ice cold ethanol, the DNA was isolated,
washed with ice cold ethanol several times, dried under an N2
stream, and stored at -20 °C.
An im a l Tr ea tm en t. Twenty male F344 rats (140-165 g,
Charles River) were treated with [3-3H]NBzMA (2.5 mg/kg, 1860
mCi/mol) in saline (sc). NBzMA was soluble in saline at the
concentration used (7.3 mM). The animals were sacrificed 4 h
after injection. Lungs and liver were removed and frozen at
-80 °C for DNA isolation. Esophageal mucosae were stripped
from the esophagus using forceps prior to storage at -80 °C.
The lungs were pooled in four groups of five, and the esophageal
mucosae were grouped into two groups of ten.
37 °C. HPLC analysis of these hydrolysates was achieved on a
C18 column (Phenomenex Bondclone) linked to UV detection.
They were eluted with isocratic 12.5 mM citric acid, 25 mM
sodium acetate, 30 mM NaOH, and 10 mM acetic acid (pH 5.1)
containing 5% methanol. A comparison of the guanosine levels
to 2′-deoxyguanosine levels demonstrated less than 8% RNA
contamination.
DNA Hyd r olysis Con d ition s. DNA (3-5 mg) was dissolved
in 10 mM sodium cacodylate (pH 7.0, 1-1.5 mL) and heated to
100 °C for 30 min. This treatment releases positively charged
purines such as 7-BzG, 3-BzA, and 7-BzA from the DNA. After
cooling on ice, 1/10 volume of 1 N HCl was added to precipitate
the DNA. Following centrifugation, the supernatant was
removed for HPLC analysis. The pellet was heated in 0.8 N
HCl for 5-6 h. The volume of the 0.8 N HCl hydrolysate
equaled the final volume of the neutral thermal hydrolysate
after addition of 1 N HCl. Some of the DNA was directly
hydrolyzed with 0.8 N HCl (0.2-2.7 mg in 0.5-1 mL) for 5-6
h at 80 °C. All samples were neutralized to pH 7 prior to
analysis by HPLC.
DNA Isola tion . Several DNA isolation methods were at-
tempted (32-34). A modification of a published combined
Kirby-Marmur method (32, 33) was chosen since it yielded the
cleanest DNA and was more successfully adapted to smaller
amounts of tissue, such as esophageal mucosae. The tissue was
minced and homogenized in 1% sodium dodecyl sulfate (SDS)
containing 1 mM EDTA (10 mL/g of tissue). The esophageal
mucosae were frozen and crushed or powdered before they were
homogenized in the SDS solution. Proteinase K (0.5 mg/g of
tissue) was added, and the mixture was incubated at room
temperature for 30 min. Then 1 M Tris-HCl (pH 7.4, 0.5 mL/
10 mL of solution) was added. The homogenate was extracted
for 5 min with 1 volume of phenol (saturated with 10 mM Tris,
pH 8, and 1 mM EDTA) containing 0.1% (v/v) cresol (phenol
reagent). Since benzyl adducts can be unstable in acid (35),
buffer-washed phenol was used to ensure neutral pH. Following
centrifugation at 14000g for 15 min, the aqueous phase was
removed and extracted with an equal volume of phenol reagent-
chloroform-isoamyl alcohol (25:24:1) for 3 min. This mixture
was centrifuged for 15 min. The resulting aqueous phase was
further extracted with 1 volume of chloroform-isoamyl alcohol
(24:1) for 3 min. This mixture was centrifuged for 15 min. The
aqueous phase was removed and made 0.5 M in NaCl. The DNA
was precipitated from this solution by the slow addition of ice
cold ethanol (1 volume). The esophageal DNA was not precipi-
tated at this point. Lung and liver DNA were harvested,
washed with 80% ethanol, and redissolved in 150 mM NaCl,
15 mM sodium citrate (pH 7.0), and 1 mM EDTA (2 mL/g of
tissue). This mixture was incubated with RNase A (200 µL/g
of a 50 µg of RNase A/mL of 0.3 M NaCl solution) and RNase
T1 (100 units/g of tissue) for 30 min at room temperature. This
solution was extracted with an equal volume of chloroform-
isoamyl alcohol (24:1) and centrifuged for 15 min. The aqueous
solution was extracted as described until there was no discern-
ible interface (2-3 times). DNA was then precipitated from this
solution with ice cold ethanol. At this point, the lung and
esophageal DNA was washed extensively with 80% ethanol
followed by absolute ethanol. The DNA was dried under an N2
stream and stored desiccated at -20 °C.
The following conditions yielded the best results for the
enzymatic hydrolysis of the benzylated DNA. DNA (3-5 mg)
was incubated with DNase I (72 units/mg of DNA) in 10 mM
Tris (pH 7.4)/5 mM MgCl2 for 2 h at 37 °C (total volume: 2-5
mL). Then, phosphodiesterase I (0.1 unit/mg of DNA) was added
and the incubation was continued at 37 °C for another 2 h. At
that time, alkaline phosphatase (600 units) was added. After
1 h at 37 °C, the hydrolysates were cooled on ice, spiked with
unlabeled benzyl DNA standards, and filtered using Centrifuge
Micropartition Systems, MW cutoff 30 000 (Amicon, Beverly,
MA). The hydrolysates were frozen at -20 °C until analysis.
The HPLC analyses were performed using from 1.8 to 4.5 mL
of the hydrolysates.
In cu ba tion s of [3-3H]BzNU-Tr ea ted DNA w ith AGT.
[3-3H]BzNU-treated DNA (1 mg) was incubated with semipu-
rified rat liver AGT (7 pmol) in HEPES (pH 7.8) containing 1
mM EDTA and 1 mM dithiothreitol (total volume: 2.5 mL) for
30 min prior to enzymatic degradation of the DNA as described
above. Controls were conducted in the absence of the AGT
preparation.
HP LC Con d ition s. The DNA hydrolysates were analyzed
using reverse-phase HPLC linked to radioflow detection (FLO-
one/Beta, Radiomatics Instruments, Tampa, FL). The mixtures
were separated on a Phenomenex Bondclone C18 column (300
× 3.9 mm) using one of two gradient systems. System 1 used
20 mM sodium phosphate (pH 4.5, solvent A) and 95% methanol
(solvent B). After 5 min at 100% A, a linear gradient was run
from 100% solvent A to 40% solvent A/60% solvent B over 60
min. System 2 employed 20 mM sodium phosphate (pH 7.0
solvent C) and solvent B. After a 5 min isocratic period at 80%
solvent C/20% solvent B, a 60 min linear gradient to 40% solvent
C/60% solvent B was used to elute the hydrolysates. When a 2
mL HPLC sample loop was utilized, 2-3 injections of the sample
were made during the initial isocratic portion of the gradient.
The multiple injections affected only the BzOH peak. When a
5 mL sample loop was used, the separation between 3-BzA and
the unknown adduct was poor.
The liver DNA was dissolved in distilled water (0.75 mL/g of
tissue). An equal volume of phosphate reagent (2.5 M K2HPO4
containing 1.6% phosphoric acid) was added. Addition of a
similar volume of 2-methoxyethanol removed polysaccharides.
The mixture was centrifuged at 20000g for 30 min at 4 °C.
Addition of an equal volume of 1% cetyltrimethylammonium
bromide in water precipitated the DNA from the upper layer.
The DNA was washed three times with cold deionized water,
followed by two washes with 2% sodium acetate in 70% ethanol.
Then it was allowed to stand in 2% sodium acetate in 70%
ethanol for at least 15 min at room temperature, to convert the
DNA to its sodium salt. It was then washed extensively with
absolute ethanol, dried under an N2 stream, and stored desic-
cated at -20 °C until analysis.
Meth yla tion Levels. The amounts of 7-methylguanine (7-
mG) and O6-mG in lung and liver DNA were quantitated using
published methods (36). Levels of 7-mG, O6-mG, and guanine
were determined by HPLC analysis with fluorescence detection
(LC 240 fluorescence detector; Perkin Elmer, Norwalk, CT). The
hydrolysates were separated using two Partisil 10 SCX columns
in tandem (Phenomenex, Torrance, CA). Neutral thermal
hydrolysates were eluted with 100 mM ammonium phosphate
(pH 2.2), and mild acid hydrolysates were eluted with the same
buffer plus 10% methanol (flow: 1 mL/min). Quantitation was
achieved using standard curves prepared for each analysis.
Minimum detection limits were 1.0 pmol of O6-mG and 4.6 pmol
of 7-mG.
Samples of this DNA (200 µg) were hydrolyzed with nuclease
P1 (5 units) in 10 mM Tris-HCl (pH 7.0, 200 µL) containing 0.5
M sodium acetate (pH 5.1, 10 µL) for 30 min at 37 °C. Alkaline
phosphatase (5 units) and 0.4 M Tris-HCl (pH 7.5, 80 µL) were
added, and the hydrolysis was continued for another hour at

22 Chem. Res. Toxicol., Vol. 10, No. 1, 1997
Peterson
F igu r e 2. HPLC chromatograms obtained for benzyl DNA adduct standards obtained using a 20 mM sodium phosphate-methanol
gradient. A ) pH 4.5 and B ) pH 7.0. See Experimental Procedures for details.
F igu r e 3. Representative radiograms of neutral thermal (A and C) and subsequent 0.8 N HCl (B and D) hydrolysates of lung (A
and B) and liver (C and D) DNA isolated from [3-3H]NBzMA-treated rats. The hydrolysates were eluted from a C18 column using
20 mM sodium phosphate (pH 4.5) with a methanol gradient. See Experimental Procedures for details.
Ta ble 1. Qu a n tita tion of BzOH a n d Ben zyla ted Ad d u cts in DNA Isola ted fr om [3-3H]NBzMA-Tr ea ted Ra tsa
levels of BzOH and adducts (pmol/µmol of guanine)b
tissue
liver
hydrolysis conditions
BzOH
3-BzA
7-BzG
N2-BzGb
N6-BzAb
O6-BzGb
enzyme (n ) 3)
0.7 ( 0.2
1.0 ( 0.2
0.2 ( 0.1
0.3 ( 0.1
0.7 ( 0.2
0.1 ( 0.03
0.3
0.2 ( 0.1
nd
ndc
0.06 ( 0.03
nd
<0.03
0.03 ( 0.01
nd
0.3 ( 0.03
nd
0.2 ( 0.1
0.1 ( 0.03
nd
0.1 ( 0.03
nd
0.1 ( 0.03
e0.03 ( 0.03
nd
e0.07 ( 0.00
neutral thermal (n ) 4)
0.8 N HCld (n ) 3)
enzyme (n ) 3)
nd
nd
lung
0.1
e0.03 ( 0.01
neutral thermal (n ) 3)
0.8 N HCl (n ) 3)
0.1 ( 0.1
nd
nd
nd
0.1 ( 0.02
0.03 (2); nd (1)
a
Twenty rats were treated with [3-3H]NBzMA (2.5 mg/kg, 1860 mCi/mmol) in saline (sc). The rats were sacrificed at 4 h after treatment.
b
See Experimental Procedures for details. In the enzyme hydrolysates, these adducts were detected as the 2′-deoxyguanosine or
2′-deoxyadenosine derivatives and are expressed as pmol/µmol of 2′-deoxyguanosine. c nd ) not detected. The detection limits were 150
d
dpm, approximately 0.03 ( 0.01 pmol/µmol of guanine except where noted. The pellet obtained from the neutral thermal hydroylsate
was heated in 0.8 N HCl for 5-6 h. See Experimental Procedures for details.
sates contained three major radioactive peaks, coeluting
with standards for BzOH, N2-BzG, and N6-BzA (Figure
3B,D). Adduct levels are displayed in Table 1.
Resu lts
In Vivo [3-3H]NBzMA Exp er im en ts. Standards of
benzyl alcohol (BzOH) and the expected benzyl DNA
adducts were separated from one another on a C18
column using one of two mobile phase systems (Figure
2). HPLC analysis with radiochemical detection of the
neutral thermal hydrolysates of lung and liver DNA from
[3-3H]NBzMA-treated rats indicated the presence of four
radioactive species. Radioactivity coeluted with BzOH,
3-benzyladenine (3-BzA), and 7-BzG standards (Figure
3A,C). The radioactive peak eluting just after 3-BzA in
the pH 4.5 mobile phase did not coelute with any of the
standards. The radiograms of the strong acid hydroly-
No radioactivity coeluted with O6-BzG, a previously
reported benzyl guanine adduct (26). The absence of this
adduct in the acid hydrolysate may have resulted from
its instability in acidic solutions; it decomposes to BzOH
and guanine (35). In order to detect this adduct, the DNA
was also subjected to enzymatic hydrolysis. Radiograms
obtained for these enzyme hydrolysates (Figure 4) con-
tained radioactive peaks that coeluted with BzOH, 3-BzA,
N2-BzdG, and N6-BzdA. There was radioactivity that
migrated in the region of O6-BzdG; however, it did not

Benzylation of DNA in Vivo and in Vitro
Chem. Res. Toxicol., Vol. 10, No. 1, 1997 23
F igu r e 4. Representative radiograms of enzymatic hydroly-
sates from lung (A) and liver (B) DNA isolated from [3-3H]-
NBzMA-treated rats. The hydrolysates were eluted from a C18
column using 20 mM sodium phosphate (pH 4.5) with
methanol gradient. See Experimental Procedures for details.
a
coelute with the standard. The enzyme hydrolysates also
contained a radioactive peak that had a similar retention
time as the unknown peak in the neutral thermal
hydrolysates.
A comparison of the adduct levels in enzymatic hy-
drolysates and chemical hydrolysates demonstrated that
these two methods yielded similar results with the
following exceptions (Table 1). Levels of BzOH were
higher in chemical hydrolysates than in enzymatic hy-
drolysates. The amount of radioactivity eluting in the
region of O6-BzdG did not account for the difference.
There were also measurable levels of 7-BzG in the neutral
thermal hydrolysates but not in the enzyme hydrolysates.
The levels of 7-mG in liver and lung DNA (120 ( 30 and
250 ( 70 pmol/µmol of guanine, respectively; n ) 3) were
several orders of magnitude higher than the levels of total
benzylation in these tissues (4.3 ( 1.2 and 1.9 ( 0.7 pmol
of benzylation/µmol of guanine, respectively). O6-mG was
also detected in lung DNA mild acid hydrolysates (46 (
28 pmol/µmol of guanine; n ) 3). No detectable O6-mG
was found in liver DNA hydrolysates. Similar results
have been reported and ascribed to efficient O6-mG repair
by the liver (37).
Isolation of esophageal DNA yielded small amounts
(0.4 mg). In order to ensure adduct detection, this DNA
was subjected to direct strong acid hydrolysis (0.8 N HCl,
80 °C for 6 h) prior to HPLC analysis. Preliminary
studies were conducted with similar quantitities of lung
DNA (0.3-0.6 mg). BzOH was the major radioactive
species present in the lung DNA hydrolysates; this
compound represented one-third of the total binding (0.5
( 0.1 pmol of BzOH/µmol of guanine and 1.5 ( 0.2 pmol
of total benzyl adducts/µmol of guanine). The total
amount of [3H] binding to the esophageal DNA was in
the same range as the lung DNA (1.8 pmol/µmol of
guanine). However, none of this radioactivity coeluted
with BzOH or the benzyl adduct standards (data not
shown). The limit of detection of BzOH in these samples
was 0.08 pmol/µmol of guanine. Therefore, the radioac-
tivity is probably not covalently bound to the DNA. The
small amounts of esophageal DNA prevented measure-
ment of DNA methylation. Using the same dose, Siglin
et al. reported approximately 50 pmol of O6-mG/µmol of
guanine in F344 rat esophageal mucosal DNA 24 h
F igu r e 5. Representative radiograms of enzymatic (A), neutral
thermal (B), and subsequent 0.8 N HCl hydrolysates (C) of
[3-3H]BzNU-treated calf thymus DNA. The hydrolysates were
eluted from a C18 column using 20 mM sodium phosphate (pH
7.0) with a methanol gradient. See Experimental Procedures
for details.
following NBzMA treatment (37).
In Vitr o [3H]BzNU Exp er im en ts. In order to de-
termine relative adduct yields in the absence of DNA
repair, adduct levels were determined in [3H]BzNU-
treated calf thymus DNA. Radiograms of the chemical
and enzymatic hydrolysates were obtained using the pH
7 mobile phase and are displayed in Figure 5. The
adduct levels are listed in Table 2. These hydrolysates
contained radioactive peaks that coeluted with BzOH,
7-BzG, N2-BzG, 3-BzA, and N6-BzA. The radioactive
peak coeluting with BzOH was extractable by methylene
chloride. The unknown that eluted at 45 min using the
pH 4.5 HPLC system elutes at 24 min with the pH 7.0
mobile phase. The enzymatic hydrolysates of this DNA
also contained a significant peak that coeluted with
O6-BzdG. This radioactive peak is not present in hy-
drolysates obtained from [3H]BzNU-treated DNA that
had been incubated with a crude preparation of rat liver
AGT. AGT repairs O6-benzylguanine (38). Interestingly,
the unknown that eluted at 24 min also disappeared with
this treatment.
Discu ssion
The data presented in this report represent the first
comprehensive study of DNA benzyl adducts formed by
benzylating agents both in vivo and in vitro. Male F344
rats were injected with [3-3H]NBzMA (2.5 mg/kg in
saline, sc) and sacrificed 4 h after treatment. This dose
of NBzMA produces esophageal tumors in rats when
administered in a multiple dose regimen (14). The 4 h
time point was chosen in order to determine the initial

24 Chem. Res. Toxicol., Vol. 10, No. 1, 1997
Ta ble 2. Qu a n tita tion of BzOH a n d Ben zyla ted Ad d u cts in [3-3H]BzNU-Tr ea ted DNA
levels of BzOH and adducts (pmol/µmol of guanine)a
Peterson
hydrolysis conditions
BzOH
3-BzA
7-BzG
N2-BzGa
N6-BzAa
O6-BzGa
enzyme (n ) 4)
4.8 ( 0.7
10.7 ( 1.9
6.8 ( 3.6
4.4 ( 0.9
4.5 ( 1.0
nd
1.7 ( 0.7
2.4 ( 1.0
nd
5.0 ( 0.8
ndb
3.8 ( 1.4
2.7 ( 0.5
nd
2.0 ( 0.9
2.8 ( 0.2
neutral thermal (n ) 3)
nd
nd
0.8 N HClc (n ) 3)
a
In the enzyme hydrolysates, these adducts were detected as the 2′-deoxyguanosine or 2′-deoxyadenosine derivatives and are expressed
as pmol/µmol of 2′-deoxyguanosine. nd ) not detected; limits of detection were 150 dpm. c The pellet obtained from the neutral thermal
hydrolysate was heated in 0.8 N HCl for 5-6 h. The solution was neutralized prior to analysis by HPLC. See Experimental Procedures
for details.
b
Ta ble 3. Rela tive Ra tios of BzOH a n d Ben zyl Ad d u ct F or m a tion a
hydrolysis
conditions
tissue
liver
BzOH
3-BzA
7-BzG
N2-BzGb
N6-BzAb
O6-BzGb
enzyme
2.3
6.0
3.0
8.0
1.0
4.6
1.0
1.0
1.0
1.0
0.9
1.1
<0.1
0.3
<0.3
<0.3
0.3
1.0
1.0
1.0
1.0
1.0
1.0
0.3
0.5
e0.3
e0.3
0.5
<0.2
<0.2
e0.3
<0.3
0.6
chemicalc
enzyme
lung
chemicalc
enzyme
CT DNA
chemicalc
0.6
0.5
<0.05
a
b
Relative to N2-BzG. In the enzyme hydrolysates, these adducts were detected as the 2′-deoxyguanosine or 2′-deoxyadenosine
derivatives. c Adduct levels in the neutral thermal and 0.8 N HCl hydrolysates were combined.
adduct distribution in the absence of significant DNA
repair. Previous studies indicated that NBzMA is rapidly
metabolized (7, 8, 39). DNA was isolated from the
esophageal mucosa since the mucosa is the primary site
of NBzMA activation in the esophagus (12, 17).
O6-position of guanine. These results demonstrate that
O6-BzdG is probably rapidly repaired in vivo.
These data are consistent with previous studies ex-
amining the reaction products between BzNU and nu-
cleosides (26, 40). Alkylation of guanosine by BzNU
under aqueous conditions generated 7-, O6-, and N2-BzG
in a ratio of 0.5:1.2:1. Reactions between adenosine and
BzNU generated predominately N6-BzA with a small
amount of 1-BzA (10:1; 26). The authors did not report
the formation of 3-BzA. In the nucleoside reactions, the
benzylating agent was more reactive with the exocyclic
sites than ring nitrogens (26). The relative ratios of 7-,
O6-, and N2-guanine adducts in [3-3H]BzNU-treated DNA
were 0.6:0.6:1. Benzylation of guanine’s exocyclic sites
predominates as predicted by the previous study. The
ratio of 3-BzA to N6-BzA in [3-3H]BzNU-treated DNA was
1:0.5. Unlike the nucleoside reactions, alkylation of the
ring nitrogen of adenosine in DNA predominates. In this
current study, 1-BzA was not detected (data not shown).
It is unlikely that this adduct underwent a Dimoth
rearrangement to 6-BzdA, since studies with 1-benzyl-
adenosine indicate that this is a minor reaction (41).
These results demonstrate that NBzMA is metabolized
to a DNA benzylating species in vivo as evidenced by the
detection of benzyl adducts in lung and liver DNA. The
levels of DNA benzylation are several orders of magni-
tude less than the levels of DNA methylation in the same
tissue. The major adducts detected were N2-BzG, 3-BzA,
N6-BzA, and 7-BzG. BzOH is the primary radioactive
species present in the hydrolysates. [3-3H]BzNU reacts
with calf thymus DNA to form the same adducts. In this
case, O6-BzdG was clearly detected. An unknown adduct
was present in enzymatic and neutral thermal hydroly-
sates of both [3-3H]BzNU-treated DNA and DNA from
[3-3H]NBzMA-treated rat tissues.
The source of BzOH in the hydrolysates is unknown.
It could result from decomposition of one or more labile
adducts under the hydrolysis conditions. Consistently,
BzOH levels were higher in the acid hydrolysates than
in the enzyme hydrolysates. Benzyl phosphate adducts
likely decompose to BzOH under the acid hydrolysis
conditions. Alternatively, positively charged adducts
such as 7-benzyl-2′-deoxyguanosine, 1-benzyl-2′-deoxy-
adenosine, and 3-benzyl-2′-deoxyadenosine might also
decompose partially to BzOH under our hydrolysis condi-
tions. O6-BzdG is another possible source of BzOH in
the acid hydrolysates. The possibility that BzOH is
noncovalently associated with the DNA cannot be com-
pletely excluded.
A comparison of the total levels of guanine adducts to
adenine adducts suggests that the benzylating agent
reacts equally well with both of these bases. These
observations demonstrate that the benzylating agent
contains chemical properties intermediate between a
simple alkylating agent such as methanediazohydroxide
and arylalkylating agents such as the epoxides of styrene
oxide and benzo[a]pyrene (26, 42). According to the rules
proposed by Dipple (42), the adduct distributions predict
that the benzylating intermediate likely reacts with DNA
with significant SN1 character where the positive charge
is somewhat delocalized as originally hypothesized (26).
The relative adduct levels formed in vivo and in vitro
were similar, with some significant differences (Table 3).
The hydrolysates of DNA from NBzMA-treated rat tis-
sues contained less 7-BzG relative to other benzyl ad-
ducts than in the [3-3H]BzNU-treated DNA hydrolysates.
Enzyme hydrolysates of the DNA isolated from NBzMA-
treated rat tissues also contained more BzOH than those
of the in vitro treated DNA. Little, if any, O6-BzdG was
observed in the DNA from NBzMA-treated rats, whereas
this adduct was a substantial DNA adduct in the BzNU-
treated DNA. A crude preparation of the repair protein
AGT was able to remove the benzyl group from the
DNA benzylation is lower than DNA methylation in
all tissues studied. These studies appear to indicate that
the tissues activate more NBzMA to a DNA methylating
agent than to a DNA benzylating species. However, in
vitro metabolism studies with F344 rat liver microsomes
predict that more of the benzylating species is formed
relative to the methylating species in the liver (21).
Therefore, it is possible that the methylating metabolite
is more DNA reactive than the benzylating metabolite.
The benzylating agent should have a shorter half-life

Benzylation of DNA in Vivo and in Vitro
Chem. Res. Toxicol., Vol. 10, No. 1, 1997 25
(8) Hodgson, R. M., Wiessler, M., and Kleihues, P. (1980) Preferential
methylation of target organ DNA by the oesophageal carcinogen
N-nitrosomethylbenzylamine. Carcinogenesis 1, 861-866.
(9) Hodgson, R. M., Schweinsberg, F., Wiessler, M., and Kleihues,
P. (1982) Mechanism of esophageal tumor induction in rats by
N-nitrosomethylbenzylamine and its ring-methylated analog
N-nitrosomethyl(4-methylbenzyl)amine. Cancer Res. 42, 2836-
2840.
(10) Ludeke, B. I., Domine´, F., Ohgaki, H., and Kleihues, P. (1992)
Modulation of N-nitrosomethylbenzylamine bioactivation by di-
allyl sulfide in vivo. Carcinogenesis 13, 2467-2470.
(11) Scherer, E., van den Berg, T., Vermeulen, E., Winterwerp, H. H.
K., and den Engelse, L. (1989) Immunocytochemical analysis of
O6-alkylguanine shows tissue specific formation in and removal
from esophageal and liver DNA in rats treated with methylbenz-
ylnitrosamine, dimethylnitrosamine, diethylnitrosamine and eth-
ylnitrosourea. Cancer Lett. 46, 21-29.
(12) van Benthem, J ., Wild, C. P., Vermeulen, E., den Engelse, L., and
Scherer, E. (1991) Immunocytochemical localization of DNA
adducts induced by a single dose of N-nitroso-N-methylbenzyl-
amine in target and non-target tissues of tumor formation in the
rat. Carcinogenesis 12, 1831-1837.
(13) Barch, D. H., J acoby, R. F., Brasitus, T. A., Radosevich, J . A.,
Carney, W. P., and Iannaccone, P. M. (1991) Incidence of Harvey
ras oncogene point mutations and their expression in methyl-
benzylnitrosamine-induced esophageal tumorigenesis. Carcino-
genesis 12, 2373-2377.
(14) Wang, Y., You, M., Reynolds, S. H., Stoner, G. D., and Anderson,
M. W. (1990) Mutational activation of the cellular Harvey ras
oncogene in rat esophageal papillomas induced by methylben-
zylnitrosamine. Cancer Res. 50, 1591-1595.
(15) Lozano, J .-C., Nakazawa, H., Cros, M.-P., Cabral, R., and Ya-
masaki, H. (1994) G f A mutations in p53 and Ha-ras genes in
esophageal papillomas induced by N-nitrosomethylbenzylamine
in two strains of rats. Mol.Carcinog. 9, 33-39.
than the methylating species and, therefore, is less likely
alkylate DNA (43-45). Consistently, MNU is apparently
at least 10-fold more reactive than BzNU with calf
thymus DNA under comparable in vitro reaction condi-
tions.2 In addition, in vivo DNA benzylation levels may
be further reduced by enzyme mediated detoxification of
the benzylating agent. Glutathione can partially sup-
press the mutagenic activity of the benzylation pathway
catalyzed by F344 rat liver S9 (23).
No detectable levels of benzyl adducts were observed
in the esophageal DNA. This is consistent with the low
rate of demethylation of NBzMA observed in esophageal
microsomes (17). However, DNA isolated from cultured
esophageal mucosa incubated with [3-3H]NBzMA did
release low levels of [3H]BzOH upon strong acid hydroly-
sis, demonstrating that the esophagus may be capable
of activating NBzMA to a benzylating agent to a small
extent.3 BzOH was a minor metabolite of NBzMA under
the conditions studied. The esophagus may also have
enhanced means of detoxifying the benzylating interme-
diate. Independent of the reason, the levels of benzyla-
tion in the esophagus were lower than those detected in
the nontarget tissues, lung and liver. This result differs
from published studies indicating that levels of DNA
methylation are higher in the esophagus than the lung
or liver of F344 rats 6 h following a subcutaneous dose
of NBzMA (3.5 mg/kg; 10). Therefore, these data indicate
that DNA benzylation likely plays a minor role, if any,
in the carcinogenic activity of NBzMA.
(16) Mitra, G., Pauly, G. T., Kumar, R., Pei, G. K., Hughes, S. H.,
Moschel, R. C., and Barbacid, M. (1989) Molecular analysis of
O6-substituted guanine-induced mutagenesis of ras oncogenes.
Proc. Natl. Acad. Sci. U.S.A. 86, 8650-8654.
(17) Labuc, G. E., and Archer, M. C. (1982) Esophageal and hepatic
microsomal metabolism of N-nitrosomethylbenzylamine and
N-nitrosodimethylamine in the rat. Cancer Res. 42, 3181-3186.
(18) Kawanishi, T., Ohno, Y., Takahashi, A., Takanaka, A., Kasuya,
Y., and Omori, Y. (1985) Relation between hepatic microsomal
metabolism of N-nitrosamines and cytochrome P-450 species.
Biochem. Pharmacol. 34, 919-924.
Ack n ow led gm en t. The author would like to thank
Dr. Robert Moschel for supplying benzyl DNA adduct
standards, Dr. Guo Nie for measuring methylation dam-
age in DNA, Ms. Xiao-Keng Liu for preparing [3-3H]-
BzNU and [3-3H]BzNU-treated DNA, and the Research
Animal Facility at AHF for animal treatment and tissue
harvest. The Research Animal Facility is partially
supported by National Cancer Institute Center Support
Grant CA-17613. This study was supported by Grant
CA-59887 from the National Cancer Institute.
(19) Kawanishi, T., Ohno, Y., Takahashi, A., Takanaka, A., Kasuya,
Y., and Omori, Y. (1985) Substrate concentration dependency of
N-nitrosodimethylamine and N-nitrosomethylbenzylamine me-
tabolism in rat liver. Carcinogenesis 6, 823-828.
(20) Lee, M., Ishizaki, H., Brady, J . F., and Yang, C. S. (1989)
Substrate specificity and alkyl group selectivity in the metabolism
of N-nitrosodialkylamines. Cancer Res. 49, 1470-1474.
(21) Farrelly, J . G., and Stewart, M. L. (1982) The metabolism of a
series of methylalkylnitrosamines. Carcinogenesis 3, 1299-1302.
(22) Ivankovic, S. (1978) Carcinogene wirkung von N-benzyl-N-ni-
trosoharnstoff (BzNH) an BD-ratten (Carcinogenic effects of
N-benzyl-N-nitrosourea in BD rats). Z. Krebsforsch. 91, 63-68.
(23) Lin, D. X., Malaveille, C., Park, S. S., Gelboin, H. V., and Bartsch,
H. (1990) Contribution of DNA methylation and benzylation to
N-nitroso-N-benzylmethylamine-induced mutagenesis in bacteria:
effects of rat liver cytochrome P450 isozymes and glutathione
transferases. Carcinogenesis 11, 1653-1658.
Refer en ces
(1) Druckey, H., Preussmann, R., Ivankovic, S., and Schma¨hl, D.
(1967) Organotrope carcinogene wirkung bei 65 verschiedenen
N-nitrosoverbindungen an BD-ratten (Organotropic carcinogenic
effects of 65 different N-nitroso compounds in BD rats). Z.
Krebsforsch. 69, 103-201.
(2) Stinson, S. F., Squire, R. A., and Sporn, M. B. (1978) Pathology
of oesophageal neoplasms and associated proliferative lesions
induced in rats by N-methyl-N-benzyl-nitrosamine. J . Natl.
Cancer Inst. 61, 1471-1475.
(3) Sander, J ., and Schweinsberg, F. (1973) Tumorinduktion bei
Ma¨usen durch N-Methylbenzyl-nitrosamin in niedriger Dosierung
(Tumor induction in mice following low doses of methylbenzylni-
trosamine). Z. Krebsforsch. 79, 157-161.
(24) Kohda, K. H., Anjo, T., Nochizuki, M., Kawazoe, Y., and Okada,
M. (1982) Mutagenicity of R-acetoxy derivatives of N-methyl-N-
benzylnitrosamine and N,N-dibenzylnitrosamine in V79 cells: A
comparison between methylating and benzylating mutagens.
Gann 73, 517-521.
(4) Yang, C. S. (1980) Research on esophageal cancer in China:
a
(25) Wiessler, M., Romuruen, K., and Pool, B. L. (1983) Biological
activity of benzylating N-nitroso compounds. Models of activated
N-nitrosomethylbenzylamine. Carcinogenesis 4, 867-871.
(26) Moschel, R. C., Hudgins, W. R., and Dipple, A. (1980) Aralkylation
of guanosine by the carcinogen N-nitroso-N-benzylurea. J . Org.
Chem. 45, 533-535.
(27) Su, C. M., Brash, D. E., Chang, M. J . W., Hart, R. W., and
D’Ambrosio, S. M. (1983) Induction of single-strand breaks plus
alkali-labile bonds by N-nitrosoureas in rat tissues in vivo:
ethylnitrosourea versus benzylnitrosourea. Mutat. Res. 108, 1-12.
(28) Dolan, M. E., Moschel, R. C., and Pegg, A. E. (1990) Depletion of
mammalian O6-alkylguanine-DNA alkyltransferase activity by
O6-benzylguanine provides a means to evaluate the role of this
protein in protection against carcinogenic and therapeutic alkyl-
ating agents. Proc. Natl. Acad. Sci. U.S.A. 87, 5368-5372.
(29) Pauly, G. T., Powers, M., Pei, G. K., and Moschel, R. C. (1988)
Synthesis and properties of H-ras DNA sequences containing
review. Cancer Res. 40, 2633-2644.
(5) Lu, S., Chiu, S. X., Yang, W. X., Hu, X. N., Guo, L. P., and Li, F.
M. (1991) Relevance of N-nitrosamines to oesophageal cancer in
China. In Relevance to Human Cancer of N-Nitroso Compounds,
Tobacco Smoke and Mycotoxins (O’Neill, I. K., Chen, J ., and
Bartsch, H., Eds.) pp 11-17, International Agency for Research
on Cancer, Lyon, France.
(6) Yoo, J .-S., Guengerich, F. P., and Yang, C. S. (1988) Metabolism
of N-nitrosodialkylamines by human liver microsomes. Cancer
Res. 48, 1499-1504.
(7) Kraft, P. L., Skipper, P. L., and Tannenbaum, S. R. (1980) In vivo
metabolism and whole-blood clearance of N-nitrosomethylbenzyl-
amine in the rat. Cancer Res. 40, 2740-2742.
2
X. K. Liu, T. E. Spratt, and L. A. Peterson, unpublished results.
L. A. Peterson and S. E. Murphy, unpublished results.
3

26 Chem. Res. Toxicol., Vol. 10, No. 1, 1997
Peterson
O6-substituted 2′-deoxyguanosine residues at the first, second, or
both positions of codon 12. Chem. Res. Toxicol. 1, 391-397.
(30) Peterson, L. A., Liu, X.-K., and Hecht, S. S. (1993) Pyridyloxobutyl
DNA adducts inhibit the repair of O6-methylguanine. Cancer Res.
53, 2780-2785.
single and repeated adminstration of N-nitrosomethylbenzyl-
amine. Carcinogenesis 17, 1135-1140.
(38) Pegg, A. E., Boosalis, M., Samson, L., Moschel, R. C., Byers, T.
L., Swenn, K., and Dolan, M. E. (1993) Mechanisms of inactivation
of human O6-alkylguanine-DNA alkyltransferase by O6-ben-
zylguanine. Biochemistry 32, 11998-12006.
(39) Kraft, P. L., and Tannenbaum, S. R. (1980) Distribution of
N-nitrosomethylbenzylamine evaluated by whole-body radioau-
tography and densitometry. Cancer Res. 40, 1921-1927.
(40) Moschel, R. C., Hudgins, W. R., and Dipple, A. (1986) Reactivity
effects on site selectivity in nucleoside aralkylation: A model for
the factors influencing the sites of carcinogen-nucleic acid
interactions. J . Org. Chem. 51, 4180-4185.
(41) Moschel, R. C., Hudgins, W. R., and Dipple, A. (1979) Selectivity
in nucleoside alkylation and aralkylation in relation to chemical
carcinogenesis. J . Org. Chem. 44, 3324-3328.
(42) Dipple, A. (1995) DNA adducts of chemical carcinogenesis.
Carcinogenesis 16, 437-441.
(43) McGarrity, J . F., and Smyth, T. (1980) Hydrolysis of diazo-
methane: kinetics and mechanism. J . Am. Chem. Soc. 102, 7303-
7308.
(44) Finneman, J . I., and Fishbein, J . C. (1995) Mechanisms of benzyl
group transfer in the decay of (E)-arylmethanediazoates and
aryldiazomethanes in aqueous solutions. J . Am. Chem. Soc. 117,
4228-4239.
(45) Chiang, Y., and Kresge, A. J . (1985) Mechanism of hydration of
simple olefins in aqueous solution. cis- and trans-cyclooctene. J .
Am. Chem. Soc. 107, 6363-6367.
(31) Hendrickson, J . B., Bergeron, R., and Sternbach, D. D. (1975) New
“Gabriel” synthesis of amines. Tetrahedron 31, 2517-2521.
(32) Gupta, R. C. (1984) Nonrandom binding of the carcinogen
N-hydroxy-2-acetylaminofluorene to repetitive sequences of rat
liver DNA in vivo. Proc. Natl. Acad. Sci. U.S.A. 81, 6943-6947.
(33) Swann, P. F., and Magee, P. N. (1968) Nitrosamine-induced
carcinogenesis. The alkylation of nucleic acids of the rat by
N-methyl-N-nitrosourea, dimethylnitrosamine, dimethylsulphate,
and methyl methanesulphonate. Biochem. J . 110, 39-47.
(34) Hecht, S. S., Spratt, T. E., and Trushin, N. (1988) Evidence for
4-(3-pyridyl)-4-oxobutylation of DNA in F344 rats treated with
the tobacco specific nitrosamines 4-(methylnitrosamino)-1-(3-
pyridyl)-1-butanone and N′-nitrosonornicotine. Carcinogenesis 9,
161-165.
(35) Moschel, R. C., Hudgins, W. R., and Dipple, A. (1984) Substituent-
induced effects on the stability of benzylated guanosines: Model
systems for the factors influencing the stability of carcinogen-
modified nucleic acids. J . Org. Chem. 49, 363-372.
(36) Hecht, S. S., Trushin, N., Castonguay, A., and Rivenson, A. (1986)
Comparative tumorigenicity and DNA methylation in F344 rats
by 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone and N-ni-
trosodimethylamine. Cancer Res. 46, 498-502.
(37) Siglin, J . C., Morse, M. A., Schut, H. A. J ., Geil, R. G., Conran, P.
B., and Stoner, G. D. (1996) O6-methylguanine levels and histo-
pathological changes in the rat esophagus and liver following
TX9601014