C8-arylguanine and C8-aryladenine formation in calf thymus DNA from arenediazonium ions

Article abstract of DOI:10.1021/tx980179f

Arylhydrazides, arylhydrazines, and N-alkyl-N-arylnitrosamines are metabolized to arenediazonium ions which yield C⁸-arylpurine adducts in calf thymus and cellular DNA. The mechanism of adduct formation has not been fully elucidated. C⁸-Arylguanine adducts likely form from direct aryl radical (Ar(·)) addition to the C⁸ position of guanine. However, the amounts of C⁸-aryladenine adducts measured here are inconsistent with direct radical attack at the C⁸ position of adenine. An intermediate product, an aryltriazene, is likely formed which then decomposes to the C⁸-aryladenine adduct. We have demonstrated that N¹-aryl-N⁸-purinyltriazene adducts are formed from a variety of para-substituted arenediazonium ions with adenine. Decomposition of the N¹-aryl-N³-purinyltriazene, at high pH and elevated temperatures, has been shown to give C⁸-aryladenine derivatives, and a free radical mechanism for this process has been proposed. Here we show that this process can occur under physiological conditions and that the C⁸-aryladenine adduct can be quantitated by HPLC. ESR studies, in which DMPO was used as a spin trap, have been used to demonstrate the intermediacy of aryl radicals during the decomposition of the N¹-aryl-N³-purinyltriazenes and to demonstrate that this process also occurs in calf thymus (ct) DNA treated with arenediazonium ions. These results suggest the involvement of an aryl radical in the formation of the observed DNA adducts. Finally, we have found that the treatment of ct DNA with arenediazonium ions produces a significant amount of depurination. Both the formation of C⁸-arylguanine and C⁸- aryladenine adducts and the generation of apurinic sites may contribute to the genotoxicity of arylhydrazides, arylhydrazines, N-alkyl-N- arylnitrosamines, and arenediazonium ions.

Full text of DOI:10.1021/tx980179f

Chem. Res. Toxicol. 1999, 12, 297-304
297
C8-Ar ylgu a n in e a n d C8-Ar yla d en in e F or m a tion in Ca lf
Th ym u s DNA fr om Ar en ed ia zon iu m Ion s
Peter M. Gannett,*,† J eannine H. Powell,† Ramakrishna Rao,† Xiangling Shi,‡
Terence Lawson,§ Carol Kolar,§ and Bela Toth§
Department of Basic Pharmaceutical Sciences, School of Pharmacy, West Virginia University,
P.O. Box 9530, Morgantown, West Virginia 26506, Pathology and Physiology Research Branch,
National Institute for Occupational Safety and Health, 1095 Willowdale Road, Morgantown,
West Virginia 26505, and Eppley Cancer Research Institute, University of Nebraska Medical Center,
600 South 42nd Street, Omaha, Nebraska 68198
Received J uly 31, 1998
Arylhydrazides, arylhydrazines, and N-alkyl-N-arylnitrosamines are metabolized to arene-
diazonium ions which yield C8-arylpurine adducts in calf thymus and cellular DNA. The
mechanism of adduct formation has not been fully elucidated. C8-Arylguanine adducts likely
form from direct aryl radical (Ar•) addition to the C8 position of guanine. However, the amounts
of C8-aryladenine adducts measured here are inconsistent with direct radical attack at the C8
position of adenine. An intermediate product, an aryltriazene, is likely formed which then
decomposes to the C8-aryladenine adduct. We have demonstrated that N1-aryl-N3-purinyltria-
zene adducts are formed from a variety of para-substituted arenediazonium ions with adenine.
Decomposition of the N1-aryl-N3-purinyltriazene, at high pH and elevated temperatures, has
been shown to give C8-aryladenine derivatives, and a free radical mechanism for this process
has been proposed. Here we show that this process can occur under physiological conditions
and that the C8-aryladenine adduct can be quantitated by HPLC. ESR studies, in which DMPO
was used as a spin trap, have been used to demonstrate the intermediacy of aryl radicals
during the decomposition of the N1-aryl-N3-purinyltriazenes and to demonstrate that this
process also occurs in calf thymus (ct) DNA treated with arenediazonium ions. These results
suggest the involvement of an aryl radical in the formation of the observed DNA adducts.
Finally, we have found that the treatment of ct DNA with arenediazonium ions produces a
significant amount of depurination. Both the formation of C8-arylguanine and C8-aryladenine
adducts and the generation of apurinic sites may contribute to the genotoxicity of arylhy-
drazides, arylhydrazines, N-alkyl-N-arylnitrosamines, and arenediazonium ions.
1) (12). Previously, we reported that C8-arylguanine
adducts (4a -c) are observed in vitro and in cells from
2a -c and that aryl radical formation and C8-arylguanine
adduct formation were directly related (13). Since the C8
position of adenine is also susceptible to radical addition,
the formation of C8-aryladenine adducts seemed to be a
likely possibility. A preliminary study demonstrated that
both C8-(p-methylphenyl)guanine (4a ) and C8-(p-meth-
ylphenyl)adenine (6a ) are formed when calf thymus DNA
(ct DNA) is treated with p-methylbenzene diazonium ion
(2a ) (14). Contrary to what would be expected on the
basis of the reactions of purines with radicals (15, 16),
In tr od u ction
The reaction of free radicals with DNA can produce a
variety of adducts. The reaction of oxygen-centered
radicals has been extensively studied, and adducts of all
four DNA bases are known (1). The reaction of carbon-
centered radicals with DNA has not been extensively
studied. However, the reactions of carbon-centered radi-
cals with DNA are generally more selective, and they
react mainly with guanine at the C8 position (2-5). Other
sites in DNA for adduct formation have been observed,
including the C8 position of adenine (6, 7) and the C4,
C5, or N3 positions of cytosine (8, 9).
1
We have studied the reaction of DNA with aryl radicals
formed from arylhydrazines. The formation of aryl radi-
cals from arylhydrazines is a two-step process involving
(i) oxidative metabolism of arylhydrazines (1a -c)1 to
arenediazonium ions (2a -c) (10, 11) and (ii) reduction,
with the loss of nitrogen, to aryl radicals (3a -c, Scheme
Abbreviations: DMPO, 5,5-dimethyl-1-pyrroline N-oxide; DMPO-
OH, 2-hydroxyl-5,5-dimethylpyrrolidine-1-oxyl; aN, nitrogen hyperfine
coupling constant; aH, hydrogen hyperfine coupling constant; DMPO-
Ar, 2-aryl-5,5-dimethylpyrrolidine-1-oxyl; 1a , (p-methylphenyl)hydra-
zine; 1b, (p-hydroxymethylphenyl)hydrazine; 1c, p-hydrazinobenzoic
acid; 2a , p-methyldiazonium ion; 2b, (p-hydroxymethyl)benzenedia-
zonium ion; 2c, (p-carboxymethyl)benzenediazonium ion; 3a , p-meth-
ylphenyl radical; 3b, p-hydroxymethylphenyl radical; 3c, p-carbox-
yphenyl radical; 4a , C8-(p-methylphenyl)guanine; 4b, C8-(p-hydroxy-
methylphenyl)guanine; 4c, C8-(p-carboxyphenyl)guanine; 5a , N1-(p-
methylphenyl)-N3-purinyltriazene; 5b, N1-(p-hydroxymethylphenyl)-
N3-purinyltriazene; 5c, N1-(p-carboxyphenyl)-N3-purinyltriazene; 6a ,
C8-(p-methylphenyl)adenine; 6b, C8-(p-hydroxymethylphenyl)adenine;
6c, C8-(p-carboxyphenyl)adenine; 7a , 2-(p-methylphenyl)-5,5-dimeth-
ylpyrrolidine-1-oxyl; 7b, 2-(p-hydroxymethylphenyl)-5,5-dimethylpyr-
rolidine-1-oxyl; 7c, 2-(p-carboxymethylphenyl)-5,5-dimethylpyrrolidine-
1-oxyl; ct DNA, calf thymus DNA.
* To whom all correspondence should be directed: School of
Pharmacy, West Virginia University, P.O. Box 9530, Morgantown, WV
26506-9530. Phone: (304) 293-1480. Fax: (304) 293-2576. E-mail:
pgannet@wvnvm.wvnet.edu.
† West Virginia University.
‡ National Institute for Occupational Safety and Health.
§ University of Nebraska Medical Center.
10.1021/tx980179f CCC: $18.00 © 1999 American Chemical Society
Published on Web 02/13/1999

298 Chem. Res. Toxicol., Vol. 12, No. 3, 1999
Gannett et al.
at the role of C8-arylguanine, C8-aryladenine, and N1-
aryl-N3-purinyltriazenes in the genotoxicity of these
compounds.
Sch em e 1
Exp er im en ta l P r oced u r es
Ca u tion : Arylhydrazines and arenediazonium ions are geno-
toxins and should be handled accordingly. Arenediazonium ions,
especially when completely dry, are shock sensitive and may
detonate.
Gen er a l P r oced u r es. Arylamines, arylhydrazines, 5,5-
dimethyl-1-pyrroline N-oxide (DMPO), phosphate buffer (100
mM, pH 7.4), adenine, guanine, and nitrosyl hydrogen sulfate
were purchased from Aldrich Chemical Co. (Milwaukee, WI).
Calf thymus DNA (ct DNA) was obtained from Sigma (St. Louis,
MO). Other inorganic reagents and solvents were obtained from
Fisher Scientific (Pittsburgh, PA) and used as received unless
otherwise noted. Arenediazonium ions were prepared by the
method of Piercy and Ward (20) from the corresponding ary-
lamines and nitrosyl hydrogen sulfate in acetic acid. Both
p-aminobenzyl alcohol and methoxy p-aminobenzyl ether were
prepared as previously described (12). DMPO was freed of
paramagnetic impurities by the method of Floyd (21). ESR
Sch em e 2
spectra were measured on
a Varian E-9 instrument with
aqueous solutions contained in a quartz flat cell at room
temperature. Values for aN and aH were obtained by computer
simulation using the program PEST which is available from
NIEHS (http://EPR.niehs.nih.gov/pest.html). NMR spectra were
measured on a Varian Gemini 300 broad-band spectrometer
equipped with a VT probe/controller.
Syn th esis of N1-Ar yl-N3-p u r in yltr ia zen es, C8-Ar yla d -
en in e Ad d u cts, a n d C8-Ar ylgu a n in e Ad d u cts. The C8-
arylguanine adducts were synthesized as previously described
(13) as were the N1-aryl-N3-purinyltriazenes (18). The C8-
aryladenine derivatives were prepared by the procedure of Chin
et al. (17) starting from the N1-aryl-N3-purinyltriazenes. In
general, the N1-aryl-N3-purinyltriazenes (1 mmol) were added
to water (20 mL), and the pH was adjusted to 10 with 1 N
NaOH. The resulting solution was then heated to 90 °C for 72
h and cooled to room temperature, and the pH was adjusted to
7 with 1 N HCl. The precipitate which forms was isolated by
filtration, washed thoroughly with chloroform and water (20 mL,
each), and dried in vacuo. The product was then recrystallized
from 1 N HCl.
adenine adduct 6a predominated. The results suggested
to us that the mechanism of formation of the C8-
aryladenine adduct is more complicated than direct
addition of an aryl radical to the C8 position of adenine
(Scheme 2, pathway a).
A plausible alternative mechanism to direct aryl radi-
cal addition involves the formation of an intermediate
N1-aryl-N3-purinyltriazene (5) from the reaction of ad-
enine with 2 (Scheme 2, pathway b). The reaction of
arenediazonium ions with adenine is known to give 5 (17,
18). Heating 5 in the presence of base results in the
formation of 6. It is not known whether the process 2 f
5 f 6 can occur in DNA under in vitro or in vivo
conditions (i.e., physiological pH and temperature). In-
direct evidence for triazene formation in ct DNA (2 f 5)
and in vivo was reported by Koepke et al. (19). However,
direct evidence for the formation of triazenes such as 5
in DNA has not been possible, possibly due to their
sensitivity to acid, base, and heat.
Here, we report the results of several studies which
support the proposal that C8-aryladenine adducts that
form in DNA exposed to para-substituted arenediazo-
nium ions 2 form from the triazene intermediate 5. HPLC
studies have been conducted that show that, in addition
to C8-arylguanine adducts, C8-aryladenine adducts form
in DNA. NMR studies regarding the decomposition of 5
show that 6 is the major product. The HPLC studies also
show that depurination accompanies C8-arylguanine and
C8-aryladenine adduct formation to a significant extent.
Finally, ESR spin-trapping experiments designed to
elucidate the mechanism of the aryladenyltriazene rear-
rangement (17) show that aryl radicals are formed during
the decomposition of 5 and suggest that the rearrange-
ment of 5 to 6 also occurs in DNA. Therefore, studies
regarding the genotoxicity of arylhydrazides, arylhydra-
zines, and N-aryl-N-alkylnitrosamines should be directed
(1) C8-(p-Meth ylp h en yl)a d en in e (6a ): 1H NMR (DMSO-
d6) δ (ppm) 2.371 (3H, s, CH3), 7.286 (2H, bs, NH2), 7.349 (2H,
d, J ) 7.5 Hz, ArH), 8.024 (2H, d, J ) 7.5 Hz, ArH), 8.137 (1H,
s, Ade-C2H); 13C NMR (DMSO-d6) δ (ppm) 21.92 (Ar-CH3), 119.5
(C-5), 127.2 (Ar-3,5), 127.8 (Ar-1), 130.5 (Ar-2,6), 140.9 (C-8),
150.1 (Ar-4), 152.3 (C-2), 153.0 (C-4), 155.3 (C-6); IR (KBr) 3385,
3206, 1694, 1671, 1488, 1104 cm-1; UV (H2O, pH 1) λmax (log ꢀ)
305 nm (4.09).
1
(2) C8-(p-Hyd r oxym eth ylp h en yl)a d en in e (6b): H NMR
(DMSO-d6) δ (ppm) 4.567 (2H, s, CH2OH), 7.2 (2H, bs, NH2),
7.470 (2H, d, J ) 7.9 Hz, Ar-H), 8.091 (2H, d, J ) 7.9 Hz, Ar-
H), 8.180 (1H, s, Ade-C2H); 13C NMR (DMSO-d6) δ (ppm) 64.20
(Ar-CH2OH), 119.0 (C-5), 126.8 (Ar-3,5), 129.0 (Ar-2,6), 139.8
(Ar-1), 140.0 (C-8), 148.2 (Ar-4), 153.2 (C-2), 154.0 (C-4), 156.4
(C-6); IR (KBr) 3559, 3160, 1690, 1485, 1191 cm-1; UV (H2O,
pH 1) λmax (log ꢀ) 296 nm (3.97).
(3) C8-(p-Ca r boxyp h en yl)a d en in e (6c): 1H NMR (DMSO-
d6) δ (ppm) 7.129 (2H, bs, NH2), 8.158 (2H, d, J ) 8.2 Hz, ArH),
8.338 (2H, d, J ) 8.2 Hz, ArH), 8.608 (1H, s, C2H); 13C NMR
(DMSO-d6) δ (ppm) 117.6 (C-5), 128.5 (Ar-3,5), 129.5 (Ar-2,6),
129.8 (Ar-1), 141.5 (C-8), 142.5 (Ar-4), 152.1 (C-2), 153.3 (C-4),
155.5 (C-6), 169.0 (Ar-CO2H); IR (KBr) 2600-3600, 1663, 1491,
1201 cm-1; UV (H2O, pH 1) λmax (log ꢀ) 308 nm (4.04).
HP LC An a lysis. The HPLC system consisted of a Waters
Associates 510 chromatography pump, an Alltech Versapak
reverse-phase C-18 10 µm column, a Shimadzu RF-535 fluores-
cence detector, and an HP 3390A integrator. Two different sets
of chromatographic conditions were used. For conditions A, the

C8-Aryladenine Adducts from Arenediazonium Ions
Chem. Res. Toxicol., Vol. 12, No. 3, 1999 299
mobile phase was 40% MeOH/100 mM citrate buffer (pH 4.8)
at a flow rate of 0.8 mL/min (detection, an excitation wavelength
of 295 nm and an emission wavelength of 375 nm). For
conditions B, the mobile phase was 30% MeOH/100 mM citrate
buffer (pH 4.8) at a flow rate of 0.2 mL/min (an excitation
wavelength of 295 nm and an emission wavelength of 375 nm).
The C8-arylguanine (4a ) and C8-aryladenine (6a ) adducts were
chromatographed using conditions A, and the retention times
for 4a and 6a were 15.7 and 37.0 min, respectively (retention
times for guanine and adenine are 2.4 and 3.4 min, respectively,
under conditions A). The remaining C8-arylpurine adducts were
chromatographed using conditions B. The retention times of the
adducts were as follows: 4b, 34.8 min; 6b, 72 min; 4c, 22.8 min;
and 6c, 41.0 min (retention times for guanine and adenine were
15.0 and 19.9 min, respectively). The limit of detection for C8-
arylpurine adducts was approximately 10 pg.
Dia zon iu m Ion Tr ea tm en t of ct DNA for HP LC An a ly-
sis. To a solution of ct DNA (5 mg/mL, 2 mL) in phosphate buffer
(20 mM, pH 7.4) was added an aqueous solution of the
arenediazonium ion (200 mM), and the resulting solution was
incubated at 37 °C for 24 h. The solutions were then cooled in
ice, and the DNA was precipitated by adding 1 volume of ice-
cold isopropyl alcohol. The DNA was then washed with 70%
ethanol and quantitated fluorimetrically (22). DNA was hydro-
lyzed by adding 50 µL of 1 N HCl to 450 µL of the arenediazo-
nium ion-treated DNA and heating for 1 h at 70 °C. The samples
were cooled, centrifuged, filtered, and stored at 4 °C prior to
HPLC analysis.
Sch em e 3
Ta ble 1. C8-Ar ylp u r in e Ad d u ct F or m a tion in ct DNAa
C8-arylguanine C8-aryladenine
arenediazonium ion
adduct (4)
adduct (6)
2a (X ) CH3) in ct DNA
930
2100
330
4700
820
-
1200
2300
1600
5900
210
2a (X ) CH3) in supernatant
2b (X ) CH2OH) in ct DNA
2b (X ) CH2OH) in supernatant
2c (X ) CO2H) in ct DNA
2c (X ) CO2H) in supernatant
920
a
Values given are in millimoles per mole of guanine in the case
of 4 and adenine in the case of 6 and are the average of three
runs.
NMR a n d ESR Decom p osition of N1-Ar yl-N3-p u r in yl-
tr ia zen es. Solutions of the N1-aryl-N3-purinyltriazenes (1 mg/
mL) were prepared in phosphate buffer (10 mM, apparent pH
of 7.4, D2O/phosphate buffer for NMR experiments) and ana-
lyzed immediately after preparation. For the ESR experiments,
DMPO (final concentration of 1 mM) was also added and the
samples were heated to the indicated temperature for 30 min
and cooled to room temperature, and the ESR spectrum was
recorded. A fresh sample was prepared for measurements at
each temperature. For NMR experiments, the sample was
heated in the probe and spectral measurements were taken at
regular time intervals over a 2 h period.
Dia zon iu m Ion Tr ea tm en t of ct DNA for ESR Decom -
p osition Stu d ies. Calf thymus DNA (50 mg) was dissolved in
phosphate buffer (50 mL, 100 mM, pH 7.4) overnight; to this
solution was added the arenediazonium hydrogen sulfate (0.2
mmol) in phosphate buffer (5 mL). The reaction mixture was
incubated for 30 min at 37 °C and cooled to 4 °C and the DNA
precipitated by the addition of absolute ethanol (25 mL). The
DNA was isolated by centrifugation, redissolved in 25 mL of
distilled water, and precipitated by the addition of absolute
ethanol (12 mL), and this procedure was repeated twice more.
The presence of arenediazonium ion was assayed by addition
of 1-naphthol (2 mg) and 6 N HCl (1 mL) and then HPLC
analysis (C-18 reverse phase, 2:3 MeOH/H2O, at a flow rate of
1 mL/min). The presence of arenediazonium ion would be
confirmed if the 4-hydroxyazonaphthalene derivative were
detected. Following the DNA isolation procedure described
above, this derivative was never detected. The isolated DNA
was dried overnight in a vacuum desiccator. Solutions for ESR
were made up by dissolving the DNA in phosphate buffer (1
mg/mL, 100 mM, pH 7.4).
F igu r e 1. HPLC of C8-arylpurine adducts obtained from a
sample of ct DNA (2 mg/mL) treated with 200 µM 2a . Peaks
are guanine (2.4 min), adenine (3.4 min), unknown (7.7 min),
4a (15.7 min), and 6a (37 min). HPLC conditions were as
follows: mobile phase of 40% MeOH/60% 20 mM citrate buffer
(pH 4.8), flow rate of 0.8 mL/min, Versapak C-18 column,
fluorometer settings, excitation at 295 nm and emission at 375
nm.
thesized from the N1-aryl-N3-purinyltriazenes (5a -c) by
heating in base (pH 10) at 90 °C. Final purification of
the C8-aryladenine adducts was effected by recrystalli-
zation. The overall synthetic route for the preparation
of 6a -c is shown in Scheme 3.
HP LC An a lysis of C8-Ar ylp u r in e Ad d u cts fr om ct
DNA. ct DNA treated with the arenediazonium ions
2a -c was hydrolyzed and analyzed for the formation of
both C8-arylguanine (4a -c) and C8-aryladenine adducts
(6a -c). The results of this analysis are reported in Table
1, and a representative chromatogram is shown in Figure
1. Due to the large difference in the polarity of p-
methylphenyl adducts 4a and 6a relative to those of the
p-hydroxymethyl and p-carboxy adducts 4b,c and 6b,c,
two different mobile phases were used as described in
Experimental Procedures. Cavalieri and co-workers (23,
24) reported that depurination occurs when radical
cations of polyaromatic hydrocarbons react at the C8
Resu lts
Syn th esis. A series of N1-aryl-N3-purinyltriazenes
(5a -c) were prepared by the reaction of the correspond-
ing arenediazonium ion (2, prepared from the arylamine
and nitrous acid) and adenine as previously reported (12).
The product was isolated by filtration and used directly.
Attempts to purify 5a -c only resulted in a less pure
product due to decomposition. However, NMR analysis
indicated that the products, as isolated, were >95% pure
in all cases. C8-Aryladenine adducts (6a -c) were syn-

300 Chem. Res. Toxicol., Vol. 12, No. 3, 1999
Gannett et al.
F igu r e 2. 1H NMR spectra of (a) 5a at 25 °C, (b) the
decomposition reaction mixture of 5a after 10 min at 75 °C, and
(c) 6a at 25 °C. In spectrum b, the integrals give a 6a :5a ratio
of 1:4.7. All spectra were obtained in phosphate buffer (10 mM,
apparent pH of 7.4). Resonances were assigned by analogy to
analogous compounds (17).
position of guanine. The process being studied here is
quite similar, and therefore, the supernatant obtained
after isolation of arenediazonium ion-treated DNA was
also examined for the presence of both 4 and 6. Both
adducts were detected in the supernatant, and the
amount of each adduct is listed in Table 1. Remarkably,
the concentration of the C8-arylpurine adducts found in
the supernatant typically exceeded what was found in
the ct DNA.
F igu r e 3. ESR spectra obtained after thermolysis of the
triazene 5a at (a) 25, (b) 40, (c) 55, and (d) 75 °C. All samples
initially contained 5a (0.1 mM) and DMPO (1 mM) in sodium
phosphate buffer (pH 7.4, 10 mM). The six major lines in all
spectra correspond to 7a (aN ) 15.5 G, aH ) 24.0 G). The minor
lines in all spectra except spectrum d are due to DMPO-H (aN
) 16.6 G, aH ) 22.6 G). In spectrum d, additional lines appear
and are due to DMPO-OH (aN ) 15.0 G, aH ) 15.0 G).
Instrument settings were as follows: receiver gain, 105; micro-
wave power, 20 mW; modulation amplitude, 1 G; sweep width,
100 G; sweep time, 240 s; and time constant, 0.25 s.
NMR of N1-Ar yl-N3-p u r in yltr ia zen e Decom p osi-
tion . The decomposition of 5 at pH 7.4 was monitored
by NMR. Representative spectra for 5a are shown in
Figure 2. Analogous spectra were obtained for 5b,c. In
Figure 2a-c are shown the NMR spectra of (a) 5a , (b)
the decomposition reaction mixture of 5a after 10 min
at pH 7.4 and 75 °C, and (c) 6a , respectively. The data
demonstrate that 5a rearranges to 6a . No other major
product was observed after decomposition of 5a was
complete, although at least one minor product, toluene
(8a , <10%), could be detected at the end of the reaction.
ESR of N1-Ar yl-N3-p u r in yltr ia zen e Decom p osi-
tion . The decomposition of 5 in the presence of the spin
trap DMPO was studied by ESR after heating it to four
temperatures (25, 40, 55, and 75 °C) and then cooling to
room temperature. A representative set of ESR spectra
for 5a are shown in Figure 3. Analogous sets of ESR
spectra were obtained for 5b,c. The hyperfine coupling
constants measured (aN and aH) were in agreement with
those previously reported for the DMPO-Ar spin adducts
(7a -c) (12).
appears to be DMPO-OH, on the basis of the hyperfine
coupling constants obtained by simulation of the spec-
trum (aN ) 15.0 G, aH ) 15.0 G) (25).
ESR of th e Dia zon iu m Ion -Tr ea ted ct DNA De-
com p osition . ct DNA was treated with the arenediazo-
nium ions 2a -c and the DNA then isolated and dissolved
in phosphate buffer (10 mM, pH 7.4). DMPO was then
added (final DMPO concentration of 1.0 mM) as for the
arylpurinyltriazenes. ESR spectra were then obtained
after heating at different temperatures and cooling to
room temperature. ESR spectra for ct DNA treated with
2a are shown in Figure 4. Similar spectra were recorded
for 2b,c. The hyperfine coupling constants were in
agreement with those previously reported for DMPO-Ar
spin adducts 7a -c (12) and those derived by thermal
decomposition of triazenes 5a -c. The ESR spectra, with
the exception of the relative intensity, were nearly
identical with those obtained from the N1-aryl-N3-puri-
nyltriazenes, over the temperature range of 25-55 °C.
The spectra obtained at 75 °C showed the formation of
DMPO-OH as in the case of 5a -c, though the ratio of
DMPO-OH to 7 was significantly greater for the ct DNA
samples. In addition, the signal-to-noise ratio for the ct
DNA spectra obtained at 75 °C was significantly lower
A second radical species was trapped for all of the
triazenes studied, and the hyperfine coupling constants
observed for this radical are in agreement with those
reported for hydrogen radical (aN ) 16.6 G, aH ) 22.6 G)
(25). Finally, at 75 °C, a third radical was trapped which

C8-Aryladenine Adducts from Arenediazonium Ions
Chem. Res. Toxicol., Vol. 12, No. 3, 1999 301
ascorbic acid, and iron(II) (13), as well as enzymes that
can mediate reductive processes (14). One-electron reduc-
tion of arenediazonium ions produces aryl radicals, and
the ease of this reduction is proportional to the presence
and nature of the substituents on the aryl ring (12, 28).
In some cases, such as with 2b or 2c, water or DNA is
sufficient to reduce arenediazonium ions to aryl radicals.
In other cases, such as 2a , stronger reducing agents are
required such as glutathione or iron(II) (13).
Our investigations of arylhydrazine and arenediazo-
nium ion genotoxicity have largely focused on the aryl
radical intermediate. This work has shown that arylhy-
drazines are oxidized to arenediazonium ions, which are
then converted to aryl radicals in vitro and in cellular
systems (13, 29). Their reduction can be mediated by
typical cellular reductants such as ascorbate, glutathione,
or NADPH (30, 31). As noted, even water may serve as
the reductant in certain cases. Once formed, the aryl
radicals may react with DNA and cause single-strand
breaks, DNA cross-links, and C8-arylguanine adduct
formation (12, 13). The formation of the C8-arylguanine
adducts may play a significant role in the genotoxicity
of arylhydrazine and arenediazonium ion genotoxicity.
In this study, the formation of C8-aryladenine adducts
in DNA was investigated. Two possible routes to these
adducts have been considered here (Scheme 2, pathways
a and b). The first of these routes (Scheme 2, pathway a)
is analogous to the mechanism proposed for C8-arylgua-
nine adduct formation (Scheme 1). The C8-arylguanine
adduct is proposed to form by first the addition of the
aryl radical to guanine, followed by oxidation of the
intermediate C8-arylguanyl radical to 4 following depro-
tonation. The second route is by addition of 2 to N6 of
adenine to form the N1-aryl-N3-purinyltriazene 5. This
adduct may then decompose to form an aryl radical which
can then add to the C8 position of adenine as in the
former route to produce 6 (Scheme 2, pathway b) as
proposed by Chin et al. (17).
F igu r e 4. ESR spectra obtained upon thermolysis of isolated
ct DNA after treatment with 2a at (a) 25, (b) 40, (c) 55, and (d)
75 °C. All samples initially contained 1 mg of treated ct DNA
and DMPO (1 mM) in sodium phosphate buffer (pH 7.4, 10 mM).
The six major lines in all spectra correspond to 7a (aN ) 15.5
G, aH ) 24.0 G). The minor lines in all spectra except spectrum
d are due to DMPO-H (aN ) 16.6 G, aH ) 22.6 G). In spectrum
d, additional lines appear and are due to DMPO-OH (aN ) 15.0
G, aH ) 15.0 G). Instrument settings are the same as those
described in the legend of Figure 1.
The reaction of radicals with purines in DNA is usually
explained by the first mechanism (either Scheme 1 or
Scheme 2, pathway a), the direct addition of the radical
to the C8 position of the purine. This may occur for 6;
however, if direct addition were the only mechanism
responsible for the formation of 6, we would expect the
ratio of 4 to 6 to be greater than 1. For example, the
addition of hydroxyl radical to DNA purines has been
extensively studied, and the ratio of C8-oxoguanine to C8-
oxoadenine, over a wide range of conditions, is always
greater than 1 and is typically approximately 3 (15, 16).
In contrast, the HPLC data recorded here showed the
total amount (adduct in DNA and depurinated adenine
adduct) of the C8-aryladenine adduct always exceeded the
total amount of the C8-arylguanine adduct. The indi-
vidual amounts of adduct in ct DNA and in the super-
natant mirrored this trend for arenediazonium ions 2a
and 2b. Only in the case of 2c was the amount of C8-
arylguanine adduct contained in the ct DNA greater than
the amount of C8-aryladenine remaining in the ct DNA.
However, in this particular case, depurination of the C8-
arylguanine adduct was not observed while depurination
of the corresponding C8-aryladenine adduct was observed.
The fact that the C8-aryladenine adduct formed to a
significantly greater extent than did the C8-arylguanine
adduct suggests that 6 may form either by an alternate
route or by multiple routes.
than that for the spectrum obtained from the N1-aryl-
N3-purinyltriazenes at this temperature. The observation
of DMPO-Ar ESR spectra and the similarity of the data
obtained here to that obtained with the N1-aryl-N3-
purinyltriazenes strongly support the formation of N1-
aryl-N3-purinyltriazene adducts in ct DNA.
Discu ssion
Arylhydrazines, including those described here, are
known genotoxins (10, 11, 26). The metabolism of aryl-
hydrazines is mediated by a variety of enzymes, including
cytochrome P450 and prostaglandin (H) synthase (10, 11).
Oxidative metabolism of arylhydrazines by these en-
zymes first produces an arenediazonium ion (2) (10). This
species is not as reactive as its alkyl counterpart and
persists (>24 h at pH 7.4 and 25 °C) (12). Arenediazo-
nium ions can be trapped by nucleophilic reagents such
as adenine or guanine and result in the formation of
triazenes or azo compounds, respectively (17, 27). Also,
unlike alkyldiazonium ions, the nitrogen portion of the
molecule is retained in these latter products.
A second process which may be either chemical or
metabolic is reduction of the intermediate arenediazo-
nium ion. There are many cellular reductants capable of
reducing arenediazonium ions, including glutathione,

302 Chem. Res. Toxicol., Vol. 12, No. 3, 1999
Gannett et al.
Sch em e 4
F igu r e 5. Possible conformations of the N1-aryl-N3-purinyl-
triazenes 5.
modified version of the mechanism shown in Scheme 2,
pathway b. The new proposed mechanism is shown in
Scheme 4.
Previous studies by our group (18) and others (19, 27)
have shown that adenine can react with arenediazonium
ions, yielding product 6. Triazenes such as 6 are unstable
toward acid, base, and heat. Chin et al. (17) demonstrated
that triazenes such as 5 rearrange to 6 upon heating in
the presence of base (90 °C, pH 10, 48 h). However, it is
not known whether this reaction could occur under
physiological conditions. Here, we heated triazenes 6 at
pH 7.4 at 25-75 °C and determined that this reaction
occurs. Thus, it is likely that the rearrangement of 5 to
6 occurs in DNA.
The mechanism shown in Scheme 4 involves the
dissociation of 5 to adenine and 2 following protonation.
Compound 2 can then be reduced and, following loss of
nitrogen, produces an aryl radical (A, Scheme 4). The
addition of this radical to adenine then proceeds as in
Scheme 2, pathway a. Alternatively, 2 may dissociate and
react with N7 of adenine to give the N7-2+ adduct (B,
Scheme 4). This intermediate is likely to have a very
weak glycosidic bond and could explain the amount of
depurination observed. The N7-2+ adduct may then be
reduced to yield an aryl radical and then follow Scheme
2, pathway a. The mechanistic pathway and intermedi-
ates A or B, shown in Scheme 4, are also consistent with
the data obtained in this study and cannot be ruled out
at this time.
The mechanism of the rearrangement of 5 to 6 has not
been investigated, and it may be important with regard
to the observed purine adduct ratios. The reaction in
Scheme 2, pathway b, was proposed by Chin et al. (17)
on the basis of the observed product. Here we prepared
the triazenes 5 and monitored their decomposition by
NMR and ESR. When a radical reaction is monitored by
NMR, in some cases certain peaks appear as emissions
(negative direction relative to normal peaks) while others
appear to show an absorption increase compared to
normal peaks (32). Such observations are indicative of a
radical process and are referred to as chemically induced
dynamic nuclear polarization (CIDNP). Failure to observe
CIDNP does not mean the process does not involve a
radical species. Unfortunately, no CIDNP was observed
in the NMR experiments conducted here, and only the
conversion of 5 to 6 was seen.
DNA treated with arenediazonium ions form C8-
aryladenine adducts, as demonstrated here, and probably
N1-aryl-N3-purinyltriazenes 5. The first evidence for the
formation of 5 in DNA was obtained by Koepke et al. (19),
who (i) treated DNA with benzene diazonium ion, (ii)
treated DNA with sodium borohydride to reduce 5 to
6-hydrazinopurine, (iii) carried out DNA hydrolysis, and
finally (iv) assayed by HPLC for 6-hydrazinopurine which
was detected in vitro and in vivo. Our data further
support the formation of this adduct in DNA. Further-
more, our ESR data of the decomposition of arenediazo-
nium ion-treated DNA show that this adduct can decom-
pose and generate aryl radicals. Since the formation of
this radical would occur near the C8 position of adenine,
relative to a radical formed in the bulk solution, its
addition to this position would be highly favored, espe-
cially if the adenine was not adjacent to guanine. In the
latter case, the aryl radical could add to the C8 position
of either adenine or guanine.
The ESR experiments demonstrate that aryl radicals
are produced during the decomposition of 5 since aryl-
DMPO spin adducts were observed. Remarkably, the
ESR data show that this process can occur at 25 °C, albeit
slowly. The formation of spin adducts must be interpreted
with caution, since the observation of an aryl-DMPO spin
adduct only indicates aryl radical formation. Their
formation may only be a minor process, and their fate
(in the absence of DMPO) can only be inferred from other
data. However, in combination with the NMR data and
the structure of the adducts that are formed in DNA, the
addition of this radical to the C8 position of adenine is a
plausible fate for the intermediate aryl radical.
The conformation about the C6N position of 5 has not
been extensively studied, though we have made some
preliminary studies in this regard. These preliminary
studies suggest that the preferred conformation for 5 is
with the aryltriazene portion of the molecule anti to the
N7 position of adenine (34) (5-a n ti) as shown in Figure
5. However, this apparently does not prohibit reaction
of the intermediate aryl radical, formed during the
decomposition of 5, at the C8 position of adenine given
the synthetic utility of this reaction and the NMR studies
reported here. 5, when contained in DNA, may be in an
alternate conformation and have the aryl triazene portion
syn to the N7 position of adenine. Steric considerations
suggest that the preferred conformation of the aryltria-
zene would be the syn conformation 5-syn . Under these
The apparent instability of the aryltriazenes studied
here is remarkable. Triazenes are known to be somewhat
unstable, especially in the presence of acid or base.
However, many triazenes can be distilled at temperatures
above 100 °C, and triazene decomposition reactions
usually are conducted at temperatures in excess of 200
°C (33). This fact, coupled with the lack of CIDNP during
the NMR decomposition of 5, forces us to consider a

C8-Aryladenine Adducts from Arenediazonium Ions
Chem. Res. Toxicol., Vol. 12, No. 3, 1999 303
(9) Zady, M. F., and Wong, J . L. (1980) Unusual competition between
nitrogen and carbon methylation of nucleosides by methyl radical
in various aqueous media. J . Org. Chem. 45, 2373-2377.
(10) Lawson, T., and Chauhan, Y. (1985) Metabolism of arylhydrazines
by mouse liver mixed-function oxidases in vitro. J . Agric. Food
Chem. 33, 218-219.
(11) Lawson, T. (1987) Metabolism of arylhydrazines by cytochrome
P-450 mixed function oxidases and prostaglandin(H)synthase
from mouse lungs. Cancer Lett. 34, 193-200.
(12) Lawson, T., Gannett, P. M., Yau, W.-M., Dalal, N. S., and Toth,
B. (1995) Different patterns of mutagenicity of arenediazonium
ions in V79 cells and Salmonella typhimurium TA 102: Evidence
for different mechanisms of action. J . Agric. Food Chem. 43,
2627-2635.
(13) Gannett, P. M., Lawson, T., Miller, M., Thakkar, D. D., Lord, J .
W., Yau, W.-M., and Toth, B. (1996) 8-Arylguanine adducts from
arenediazonium ions and DNA. Chem.-Biol. Interact. 101, 1-25.
(14) Gannett, P. M., Yau, W.-M., Lawson, T., Lord, J ., Kolar, C., and
Toth, B. (1997) Formation of C8-aryladenine adducts from are-
nediazonium ions and DNA. American Association for Cancer
Research, Proceedings 38, A2233.
(15) Fuciarelli, A. F., Wegher, B. J ., Gajewski, E., Dizdaroglu, M., and
Blakely, W. F. (1989) Quantitative measurement of radiation-
induced base products in DNA using gas chromatography-mass
spectrometry. Radiat. Res. 119, 219-231.
(16) Aruoma, O. I., Halliwell, B., Gajewski, E., and Dizdaroglu, M.
(1989) Damage to the bases in DNA induced by hydrogen peroxide
and ferric ion chelates. J . Biol. Chem. 264, 20509-20512.
circumstances, decomposition will lead to aryl radical
generation in close proximity to the C8 position of
adenine, rendering this reaction favorable. Alternatively,
the preferred conformation could be the 5-a n ti since the
phenyl ring of 5 could stack between the bases, though
this might cause significant local distortions of the DNA
helix. The relative stabilities of these two conformations
of 5 in DNA remain to be explored.
Arylhydrazines are metabolically converted to arene-
diazonium ions and then to aryl radicals in vitro and in
cellular systems, ultimately forming C8-arylguanine ad-
ducts. Here, we have shown that C8-aryladenine adducts
also form in DNA. The preponderance of the adenine
adducts over the guanine adducts provides additional
support that alternative or additional processes are
involved in the formation of the C8-aryladenine adduct.
The data presented here suggest that C8-aryladenine
formation occurs in a multistep process that includes the
formation of the triazene 5, the subsequent decomposition
of 5 to an aryl radical directly or through several steps,
and finally addition of the aryl radical to adenine. These
results are relevant to elucidating arylhydrazine geno-
toxicity. They also have broader application, since a key
intermediate in the mechanism, the arenediazonium ion,
is also produced from other environmentally significant
compounds, including triazenes (35, 36), azo dyes (37, 38),
and N-aryl-N-alkylnitrosamines (19). Finally, questions
about the mechanism of 5 f 6 (Scheme 2, pathway b,
versus Scheme 4), conformational aspects of 5 relevant
to the formation of C8-aryladenine adducts, and the
biological consequences of C8-arylpurine adduct formation
remain, and we are currently exploring these questions.
(17) Chin, A., Hung, M.-H., and Stock, L. M. (1981) Reactions of
benzenediazonium ions with adenine and its derivatives. J . Org.
Chem. 46, 2203-2207.
(18) Gannett, P. M., Lawson, T., and Toth, B. (1987) Anomalous
reaction of adenine and 4-(hydroxymethyl)benzenediazonium ion.
Tetrahedron Lett. 28, 1015-1018.
(19) Koepke, S. R., Kroeger-Koepke, M. B., and Michejda, C. J . (1990)
Evidence for an unstable DNA adduct from N-nitroso-N-methy-
laniline. Chem. Res. Toxicol. 3, 17-20.
(20) Piercey, M. R., and Ward, E. R. (1962) The preparation of pure
solid diazonium sulphates. J . Chem. Soc., 3841-3843.
(21) Floyd, R. A., Lewis, C. A., and Wong, P. K. (1984) High-pressure
liquid chromatography-electrochemical detection of oxygen free
radicals. Methods Enzymol. 105, 231-237.
(22) Zubroff, J ., and Sarma, D. S. R. (1976) A nonradioactive method
for measuring DNA damage and its repair in nonproliferating
tissues. Anal. Biochem. 70, 387-396.
(23) RamaKrishna, N. V. S., Gao, f., Padmavathi, N. S., Cavalieri, E.,
Rogan, E. G., Cerny, R. L., and Gross, M. L. (1992) Model Adducts
of benzo[a]pyrene and nucleosides formed from its radical cation
and diol epoxide. Chem. Res. Toxicol. 5, 293-302.
Ack n ow led gm en t. We thank the National Institutes
of Health for their support of this work (Grant CA 31611),
the National Institute of Occupational Safety and Health
for providing us access to its ESR facilities, and the
American Foundation for Pharmaceutical Education and
the Arlen and Louise Swiger Doctoral Fellowship for their
financial support to J .H.P.
(24) Chakravarti, D., Pelling, J . C., Cavalieri, E. L., and Rogan, E. G.
(1995) Relating aromatic hydrocarbon-induced DNA adducts and
c-H-ras mutations in mouse skin papillomas: The role of apurinic
sites. Proc. Natl. Acad. Sci. U.S.A. 92, 10422-10426.
(25) Buettner, G. R., and Need, M. J . (1985) Hydrogen peroxide and
hydroxyl free radical production by hematoporphyrin derivative,
ascorbate and light. Cancer Lett. 25, 297-304.
(26) Toth, B., Tompa, A., and Patil, K. (1977) Tumorigenic effect of
4-methylphenylhydrazine hydrochloride in Swiss mice. Z. Kreb-
sforsch. 89, 245-252.
(27) Hung, M.-H., and Stock, L. M. (1982) Reactions of benzenedia-
zonium ions with guanine and its derivatives. J . Org. Chem. 47,
448-453.
(28) Elofson, R. M., and Gadallah, F. F. (1969) Substituent effects in
the polarography of aromatic diazonium salts. J . Org. Chem. 34,
854-857.
(29) Gannett, P. M., Shi, X., Lawson, T., Kolar, C., and Toth, B. (1997)
Aryl radical formation during the metabolism of arylhydrazines
by microsomes. Chem. Res. Toxicol. 10, 1372-1377.
(30) Yasui, S., Nakamura, K., and Ohno, A. (1984) Reduction by a
model of NAD(P)H. 45. Mechanism for the dediazoniation of
arenediazonium salts initiated by one-electron transfer from an
NAD(P)H model. J . Org. Chem. 49, 878-882.
(31) Reszka, K. J ., and Chignell, C. F. (1995) One-electron reduction
of arenediazonium compounds by physiological electron donors
generates aryl radicals. An EPR and spin trapping investigation.
Chem.-Biol. Interact. 96, 223-234.
Refer en ces
(1) Breen, A. P., and Murphy, J . A. (1995) Reactions of oxyl radicals
with DNA. Free Radical Biol. Med. 18, 1033-1077.
(2) Augusto, O., Cavalieri, E. L., Rogan, E. G., RamaKrishna, N. V.
S., and Kolar, C. (1990) Formation of 8-methylguanine as a result
of DNA alkylation by methyl radicals generated during horserad-
ish peroxidase-catalyzed oxidation of methylhydrazine. J . Biol.
Chem. 265, 22093-22096.
(3) Augusto, O. (1993) Alkylation and cleavage of DNA by carbon-
centered radical metabolites. Free Radical Biol. Med. 15, 329-
336.
(4) Hiramoto, K., Masae, M., Kato, T., and Kikugawa, K. (1995) DNA
strand breaking by the carbon-centered radical generated from
4-(hydroxymethyl)benzene diazonium salt, a carcinogen in mush-
room Agaricus bisporus. Chem.-Biol. Interact. 94, 21-36.
(5) Rogan, E. G., Cavalieri, E. L., Tibbels, S. R., Cremonesi, P.,
Warner, C. D., Nagel, D. L., Tomer, K. B., Cerny, R. L., and Gross,
M. L. (1988) Synthesis and identification of benzo[a]pyrene-
guanine nucleoside adducts formed by electrochemical oxidation
and by horseradish peroxidase catalyzed reaction of benzo[a]-
pyrene with DNA. J . Am. Chem. Soc. 110, 4023-4029.
(6) Leonov, D., and Elad, D. (1974) Ultraviolet- and γ-ray-induced
reactions of nucleic acid constituents. Reactions of purines with
ethers and dioxolane. J . Org. Chem. 39, 1470-1473.
(7) Kang, J . O., Gallagher, K. S., and Cohen, G. (1993) Methylation
of RNA purine bases by methyl radicals. Arch. Biochem. Biophys.
306, 178-182.
(32) Fossey, J ., Lefort, D., and Sorba, J . (1995) Free Radicals in
Organic Chemistry, pp 13-16, J ohn Wiley & Sons, New York.
(33) Rondestvedt, C. S., J r., and Davis, S. J . (1957) 1-Aryl-3,3-
(8) Liang, G., Gannett, P., and Gold, B. (1995) The use of 2-hydro-
peroxytetrahydrofuran as a reagent to sequence cytosine and to
probe non-Watson-Crick DNA structures. Nucleic Acids Res. 23,
713-719.
dialkyltriazenes as tumor inhibitors. J . Org. Chem. 22, 200-203.

304 Chem. Res. Toxicol., Vol. 12, No. 3, 1999
Gannett et al.
(34) Iwahashi, H., Sugeta, H., and Kyogoku, Y. (1982) Detection of
separated amino proton resonance signals of adenine derivatives
at low temperature and its application to estimation of population
of the adenine-uracil dimers in solution. Biochemistry 21, 631-
638.
(37) Stiborava, M., Asfaw, B., Anzenbacher, P., Leseticky, L., and
Hodek, P. (1988) The first identification of the benzenediazonium
ion formation from a non-aminoazo dye, 1-phenylazo-2-hydrox-
ynaphthalene (Sudan I) by microsomes of rat livers. Cancer Lett.
40, 319-326.
(35) Carvalho, E., Iley, J ., de J esus Perry, M., and Rosa, E. (1998)
Triazene drug metabolites: Part 15. Synthesis and plasma
hydrolysis of anticancer triazenes containing amino acid carriers.
Pharm. Res. 15, 931-935.
(36) Smith, R. H., J r., Mehl, A. F., Shantz, D. L., J r., Chmurny, G.
N., and Michejda, C. J . (1988) Novel cross-linking alkylating
agents, 1-(2-chloroethyl)-3-methyl-3-acyltriazenes. J . Org. Chem.
53, 1467-1471.
(38) Stiborava, M., Asfaw, B., Anzenbacher, P., and Hodek, P. (1988)
A new way to carcinogenicity of azo dyes: The benzenediazonium
ion formed from
a non-aminoazo dye, 1-phenylazo-2-hydrox-
ynaphthalene (Sudan I) by microsomal enzymes binds to deox-
yguanosine residues of DNA. Cancer Lett. 40, 327-333.
TX980179F