Analysis of 1,N2-ethenoguanine and 5,6,7,9-tetrahydro-7-hydroxy-9- oxoimidazo[1,2-α]purine in DNA treated with 2-chlorooxirane by high performance liquid chromatography/electrospray mass spectrometry and comparison of amounts to other DNA adducts

Article abstract of DOI:10.1021/tx960124i

High performance liquid chromatography (HPLC)/electrospray mass spectrometry methods were developed for the analysis of 1,N²- etheno(ε)guanine (Gua) and 5,6,7,9-tetrahydro-7-hydroxy-9-oxoimidazo[1,2- α]purine (HO-ethanoGua) [the cyclized form of N²-(2-oxoethyl)Gua] and its deoxyribose derivative in DNA. Evidence was provided for the formation of the latter adduct in DNA treated with 2-chlorooxirane, the reactive product formed from the carcinogen vinyl chloride. Measured levels of HO-ethanoGua and HO-ethanodeoxyguanosine were similar, although the assay for the deoxyribosyl derivative has some technical advantages. 3,N⁴-ε-Deoxycytidine was also estimated in 2-chlorooxirane-treated DNA using HPLC with fluorescence detection. Levels of all known adducts formed from vinyl chloride have now been estimated in DNA treated with 2-chlorooxirane and vary in the order N⁷-(2-oxoethyl)Gua >> 1,N⁶-ε-adenine > HO-ethanoGua > N²,3- ε-Gua > 3,N⁴-ε-cytosine > 1,N²-ε-Gua. Although in vivo adduct levels may not parallel these due to differential stability and rates of repair, analyses of the adducts in DNA treated with 2-chlorooxirane provide a basis for consideration of the biological effects of these adducts.

Full text of DOI:10.1021/tx960124i

242
Chem. Res. Toxicol. 1997, 10, 242-247
An a lysis of 1,N2-Eth en ogu a n in e a n d
5,6,7,9-Tetr a h yd r o-7-h yd r oxy-9-oxoim id a zo[1,2-a ]p u r in e
in DNA Tr ea ted w ith 2-Ch lor ooxir a n e by High
P er for m a n ce Liqu id Ch r om a togr a p h y/Electr osp r a y Ma ss
Sp ectr om etr y a n d Com p a r ison of Am ou n ts to Oth er DNA
Ad d u cts
Michael Mu¨ller,†,‡ Frank J . Belas,§ Ian A. Blair,§ and F. Peter Guengerich*,†
Departments of Biochemistry and Pharmacology and Center in Molecular Toxicology,
Vanderbilt University School of Medicine, Nashville, Tennessee 37232
Received J uly 29, 1996X
High performance liquid chromatography (HPLC)/electrospray mass spectrometry methods
were developed for the analysis of 1,N2-etheno(ꢀ)guanine (Gua) and 5,6,7,9-tetrahydro-7-
hydroxy-9-oxoimidazo[1,2-a]purine (HO-ethanoGua) [the cyclized form of N2-(2-oxoethyl)Gua]
and its deoxyribose derivative in DNA. Evidence was provided for the formation of the latter
adduct in DNA treated with 2-chlorooxirane, the reactive product formed from the carcinogen
vinyl chloride. Measured levels of HO-ethanoGua and HO-ethanodeoxyguanosine were similar,
although the assay for the deoxyribosyl derivative has some technical advantages. 3,N4-ꢀ-
Deoxycytidine was also estimated in 2-chlorooxirane-treated DNA using HPLC with fluores-
cence detection. Levels of all known adducts formed from vinyl chloride have now been
estimated in DNA treated with 2-chlorooxirane and vary in the order N7-(2-oxoethyl)Gua .
1,N6-ꢀ-adenine > HO-ethanoGua > N2,3-ꢀ-Gua > 3,N4-ꢀ-cytosine > 1,N2-ꢀ-Gua. Although in
vivo adduct levels may not parallel these due to differential stability and rates of repair,
analyses of the adducts in DNA treated with 2-chlorooxirane provide a basis for consideration
of the biological effects of these adducts.
Ch a r t 1. Str u ctu r es of E a n d Rela ted DNA Ba ses
In tr od u ction
The industrial monomer vinyl chloride was linked to
the incidence of liver hemangiosarcomas in the early
1970s and shown to cause the same cancer in rodents
(1-3). In vitro and in vivo experiments provide strong
evidence that the tumorigenicity of vinyl chloride is
dependent upon its oxidation to 2-chlorooxirane (2-
chloroethylene oxide) (4, 5). The reaction of 2-chloroox-
irane with DNA, nucleosides, or DNA bases yields N7-
(2-oxoethyl)Gua and at least four etheno (ꢀ)1 adducts:
1,N6-ꢀ-Ade, 3,N4-ꢀ-Cyt, 1,N2-ꢀ-Gua, and N2,3-ꢀ-Gua (Chart
1) (4, 5). The ꢀ adducts can also be derived from the
2-chlorooxirane rearrangement product 2-chloroacetal-
dehyde (6, 7) and from a number of functional equivalents
of these compounds (5, 8-13), several of which are formed
by oxidation of suspect carcinogens. Of particular inter-
est is the finding that some of these ꢀ adducts are present
in tissues of animals and humans not exposed to vinyl
halides or other compounds known to yield ꢀ adducts
(14, 15).
This laboratory previously reported that the reaction
of 2-chlorooxirane with Gua derivatives forms 5,6,7,9-
tetrahydro-7-hydroxy-9-oxoimidazo[1,2-a]purine (HO-
ethanoGua), which is stable and does not dehydrate to
form 1,N2-ꢀ-Gua (Chart 1) (16). However, definitive
evidence as to the formation of this adduct in DNA was
not presented. The lack of characteristic UV absorption
or fluorescence did not allow HPLC quantitation. There-
fore, HPLC/electrospray ionization (ESI) MS assays were
developed for the analysis of HO-ethanoGua and HO-
ethano-dGuo in DNA. Such assays were also utilized to
measure 1,N2-ꢀ-Gua, which has only very weak fluores-
cence at nonalkaline pH (16). HPLC/fluorescence assays
were also done to complete the analysis of 2-chloroox-
irane-derived DNA adducts and permit comparisons of
the amounts of each.
* Address correspondence to this author at Department of Biochem-
istry and Center in Molecular Toxicology, Vanderbilt University School
of Medicine, Nashville, TN 37232-0146.
† Department of Biochemistry and Center in Molecular Toxicology.
‡ Present address: Abteilung Arbeits- und Sozialmedizin der Georg-
August-Universita¨t Go¨ttingen, Waldweg 37, 37073 Go¨ttingen, Federal
Republic of Germany.
§ Department of Pharmacology and Center in Molecular Toxicology.
X Abstract published in Advance ACS Abstracts, J anuary 15, 1997.
1 Abbreviations: ESI, electrospray ionization; SRM, selected reaction
monitoring; ꢀ, etheno. HO-ethanoGua denotes the base 5,6,7,9-
tetrahydro-7-hydroxy-9-oxoimidazo[1,2-a]purine, the cyclized form of
N2-(2-oxoethyl)Gua, and the same designation applies for the deoxy-
ribonucleoside. See list of standard abbreviations of this journal (first
issue of year) for nucleic acid bases.
S0893-228x(96)00124-5 CCC: $14.00 © 1997 American Chemical Society

Analysis of 2-Chlorooxirane-Derived DNA Adducts
Chem. Res. Toxicol., Vol. 10, No. 2, 1997 243
eV was employed for the HO-ethanoGua adduct/internal stan-
dard. A collision cell pressure of 2.4 mT was used to induce
fragmentation in the collision cell (Q2), and Ar was used as the
collision gas.
Oth er In str u m en ta tion . 1H-NMR spectra were recorded
in H2O/2H2O using a Bruker AM-400 spectrometer (Bruker,
Billerica, MA) in the Vanderbilt facility. HPLC work involved
the use of a Spectra-Physics 8700 pumping system (Spectra-
Physics, Piscataway, NJ ) and a 10 × 250 mm Beckman 5 µm
Ultrasphere reversed-phase octadecylsilane column (Beckman,
San Ramon, CA).
Mod ifica tion of Ca lf Th ym u s DNA 2-Ch lor ooxir a n e a n d
An a lysis of Gu a Ad d u cts. Samples of DNA (5 mg in 1 mL of
0.1 M N-ethylmorpholine acetate buffer, pH 7.8) were treated
with varying concentrations of 2-chlorooxirane as described
previously (16). DNA was recovered by C2H5OH precipitation
and dried under N2.
Ma ter ia ls a n d Meth od s
Ch em ica ls. Chemicals were purchased from Aldrich (Mil-
waukee, WI) unless indicated otherwise. Stable isotope labeled
materials were obtained from Isotec Inc. (Cincinnati, OH). Calf
thymus DNA was purchased from Sigma Chemical Co. (St.
Louis, MO).
Syn th eses. Warning! 2-Chlorooxirane and 2-chloroacetal-
dehyde are probable human carcinogens and should be handled
in a fume hood with adequate skin and eye protection! 2-Chlo-
rooxirane was prepared by photochemical chlorination of eth-
ylene oxide (17), and the concentration was estimated using
4-(4′-nitrobenzyl)pyridine reagent (18). 3,N4-ꢀ-dCyd was pre-
pared from dCyd by reaction with 2-chloroacetaldehyde (λmax
272 nm) (6, 19). [4,5,8-13C]Gua was synthesized according to
the procedure of Scheller et al. (20). Unlabeled and trilabeled
HO-ethanoGua were prepared from 2-fluorohypoxanthine as
described previously (16). 1,N2-ꢀ-Gua (21) and N2,3-ꢀ-Gua (7)
were obtained following established procedures. Unlabeled
1,N2-ꢀ-dGuo was a gift of Prof. L. J . Marnett of this Center. All
other nucleoside adducts were prepared by enzymatic attach-
ment of the deoxyribose to the base with trans-N-deoxyribosy-
lase (22). Analytical data follow: [2,9,9a-13C]HO-ethanoGua:
MS, m/ z (assignments and relative abundance in parentheses)
For the analysis of bases, 50 ng of [2,9,9a-13C]HO-ethanoGua
and [2,9,9a-13C]1,N2-ꢀ-Gua standards were added to each sample.
The DNA was dissolved in 0.8 mL of 0.10 N HCl and heated for
1 h at 70 °C. After cooling and neutralizing, the apurinic acid
was removed from the samples with Centricon-30 filters (Ami-
con, Lexington, MA). Each sample was divided into three
aliquots, which were subjected to preparative HPLC cleanup.
In the analysis of the Gua derivatives, the HPLC solvent system
consisted of 50 mM NH4HCO2 buffer (pH 5.0) and CH3OH with
a linear gradient of 0-50% CH3OH over 25 min at flow rate of
2.5 mL min-1. The cleanup of the depurinated calf thymus DNA
sample was achieved by holding at 100% 50 mM NH4HCO2 (pH
5.0) for 5 min, followed by a linear gradient of 0-50% CH3OH
1
197 (MH+, 100), 179 (MH+ - H2O, 10); H-NMR (2H2O) δ 3.67
(dd, 1 H, H-6), 4.04 (dd, 1 H, H-6), 6.32 (dd, 1 H, H7), 8.66 (d,
1 H, H-2). [2,9,9a-13C]1,N2-ꢀ-Gua: MS, m/ z 179 (MH+, 100),
155 (MH+, Gua, 8); 1H-NMR (2H2O) δ 7.34 (d, 1 H, H-6), 7.62
(d, 1 H, H-7), 8.1 (d, 1 H, H-2). HO-ethano-dGuo: MS, m/ z
310 (MH+, 100), 194 (MH+ - deoxyribose, 26); 1H-NMR (2H2O)
δ 2.46 (m, 1 H, H-2′), 2.70 (m, 1 H, H-2′′), 3.74 (m, 3 H, H-5′,
H-5′′, H-6), 4.08 (m, 2 H, H-4′, H-6), 4.56 (m, 1 H, H-3′), 6.19 (t,
2 H, H-1′, H-7), 7.91 (s, 1 H, H-2). [2,9,9a-13C]HO-ethano-dGuo:
MS, m/ z 313 (MH+, 100), 197 (MH+ - deoxyribose, 26); 1H-
NMR (2H2O) δ 2.46 (m, 1 H, H-2′), 2.70 (m, 1 H, H-2′′), 3.74 (m,
3 H, H-5′, H-5′′, H-6), 4.08 (m, 2 H, H-4, H-6), 4.56 (m, 1 H,
over 30 min at a flow rate of 2.5 mL min-1
. Absorbance was
monitored at 254 nm. Adduct fractions eluting at tR 14.5-16.7
and 20-22.3 min were pooled, and buffer salts were removed
by repeated lyophilization. Each sample was dissolved in 100
µL of 10 mM NH4CH3CO2 buffer, pH 5.5, and filtered through
a 0.22 µm filter. A 35 µL aliquot was analyzed by HPLC/MS/
MS.
For the analysis of nucleosides, 1.0 mg of 2-chlorooxirane-
treated DNA was dried and mixed with 10 ng of [2,9,9a-13C]-
HO-ethano-dGuo and 10 ng of [2,9,9a-13C]-1,N2-ꢀ-dGuo. DNA
was digested according to the method of Chaudhary et al. (23).
The (1.0 mL) digests were filtered through a 0.22 µm filter, and
250 µL aliquots were analyzed by HPLC/MS/MS.
H-3′), 6.19 (t,
2 H, H-1′, H-7), 7.92 (dd, 1 H, H-2).
[2,9,9a-13C]1,N2-ꢀ-dGuo: MS, m/ z 295 (MH+, 100), 317 (M +
Na+, 58), 179 (MH+ - deoxyribose, 26); H-NMR (2H2O) δ 2.51
1
(m, 1 H, H-2′), 2.82 (m, 1 H, H-2′′), 3.80 (m, 2 H, H-5′, H-5′′),
4.14 (m, 1 H, H-4′), 4.66 (m, 1 H, H-3′), 6.45 (t, 1 H, H-1′), 7.35
(d, 1 H, H-6), 7.64 (d, 1 H, H-7), 8.16 (dd, 1 H, H-2). N2,3-ꢀ-
dGuo: MS, m/ z 292 (MH+, 100), 176 (MH+ - deoxyribose, 24);
UV (pH 5) λmax 215, 262 nm.
Continuous flow injections of DNA adduct standards (base
or nucleoside, each 100 ng/µL) were done to establish MS
operating conditions using a 100 µL syringe pump at a flow rate
Ma ss Sp ectr om etr ic Con d ition s. Mass spectrometry stud-
ies were conducted using a Finnigan TSQ 7000 triple quadrupole
mass spectrometer (Finnigan, San J ose, CA) operating in the
positive ion mode with an electrospray needle voltage of 4.5 kV.
N2 was used as the sheath gas (60 psi) to assist with nebuliza-
tion and as the auxiliary gas (15 psi) to assist with desolvation.
The stainless steel capillary was heated to 220 and 200 °C,
respectively, to provide optimal desolvation, and the ESI
interface and mass spectrometer parameters were optimized to
obtain maximum sensitivity. The tube lens and the heated
capillary were operated at 74.5 and 20.0 V, respectively, and
the electron multiplier was set at 1700 V. Selected reaction
monitoring (SRM) experiments were conducted by monitoring
the m/ z 176 f m/ z 121 transition for the 1,N2-ꢀ-Gua adduct
(m/ z 179 f 123 transition for the internal standard)2 and the
m/ z 194 f m/ z 176 transition for HO-ethanoGua (m/ z 197 f
m/ z 179 transition for the internal standard). For the 1,N2-ꢀ-
Gua adduct, the protonated molecular ion at m/ z 176 (m/ z 179
for the internal standard) was selected in quadrupole 1 (Q1) of
the mass spectrometer, and the product ion at m/ z 121 (m/ z
123 for the internal standard) was monitored in quadrupole 3
(Q3). Similarly, for the HO-ethanoGua adduct, the protonated
molecular ion at m/ z 194 (m/ z 197 for internal standard) was
selected in Q1 of the mass spectrometer, and the product ion at
m/ z 176 (m/ z 179 for the internal standard) was monitored in
Q3. During SRM studies, a collision offset energy of -28 eV
was used for the 1,N2-ꢀ-Gua adduct/internal standard and -22
of 5 µL min-1
. Initial sensitivity was investigated with loop
injections of appropriate dilutions, g1 ng per 35 µL injection
volume (DNA base adduct standards) and g1 ng per 25 µL
injection volume (nucleoside DNA adduct standards). Both of
the HPLC methods involved the use of a Hewlett-Packard 1090
HPLC pumping system, connected to a Phenomenex Partisil
ODS-3 reversed phase column (3.2 × 250 mm, 5 µm, Phenom-
enex, Torrance, CA). The flow rate was 0.25 mL min-1
. The
solvent system consisted of 10 mM NH4CH3CO2 buffer, pH 5.5
(A), and 0.05% CH3CO2H in CH3OH, v/v (B). Separation of DNA
bases was achieved with the following gradient: 0 min (90% A,
10% B), 30 min (60% A, 40% B), 35 min (30% A, 70% B), 45 min
(90% A, 10% B), 50 min (90% A, 10% B). Separation of
nucleosides involved a different gradient: 0 min (100% A, 0%
B), 7 min (100% A, 0% B), 37 min (50% A, 50% B), 45 min (30%
A, 70% B), 55 min (100% A, 0% B), 70 min (100% A, 0% B).
With both the bases and nucleosides, simultaneous SRM
experiments were designed. Basic parameters were established
in collision-induced dissociation experiments. Precursor ions
(the MH+ ions of the DNA adducts) were generated in the ESI
source and focused (quadrupole 1). These ions were dissociated
in a collision cell (quadrupole 2) yielding defined product ions
which were analyzed (quadrupole 3). The optimal collisional
offset voltages were investigated to maximize the yields of
product ions: HO-ethanoGua, -22 mV; 1,N2-ꢀ-Gua, -28 mV;
N2,3-ꢀ-Gua, -36 mV; HO-ethano-dGuo, N2,3-ꢀ-dGuo, and 1,N2-
ꢀ-dGuo, -18 mV. HPLC methods were devised for the separa-
2 With 1,N2-ꢀ-Gua, one of the labeled carbons is lost in the ionization
process.

244 Chem. Res. Toxicol., Vol. 10, No. 2, 1997
Mu¨ller et al.
tion of the adducts from one another and from N2,3-ꢀ-Gua (DNA
base adduct assay) or N2,3-ꢀ-dGuo (nucleoside assay). The latter
adduct is known to be formed from 2-chlorooxirane in the same
reaction and has an identical m/ z with 1,N2-ꢀ-Gua or 1,N2-ꢀ-
dGuo. To test the selectivity of the base adduct method, 5.6 ng
of N2,3-ꢀ-Gua was analyzed. The adduct eluted ∼1 min ahead
of the 1,N2-ꢀ-Gua (tR 20.48 min compared to 21.46 min).
Although the m/ z of product ions are identical for both deriva-
tives, there was a significant difference in the collisional offset
voltage of -36 eV for the N2,3-ꢀ-Gua versus -28 eV for the 1,N2-
ꢀ-Gua. This resulted in a decrease of sensitivity for the detection
of the N2,3-ꢀ-Gua, which gave only a response twice background
under conditions optimized to reproducibly quantitate a 10-fold
smaller amount of 1,N2-ꢀ-Gua. Comparable results were ob-
tained with N2,3-ꢀ-dGuo subjected to the nucleoside method.
Mod ifica tion of Ca lf Th ym u s DNA w ith 2-Ch lor oox-
ir a n e a n d An a lysis of 3,N4-E-d Cyd . Samples of DNA (10 mg
in 2.0 mL of 0.1 M potassium phosphate buffer) were treated
with varying amounts of 2-chlorooxirane for 15 min at 23 °C
(16). DNA was precipitated by the addition of 10 mL of cold
C2H5OH (-20 °C), collected by centrifugation ((3 × 103)g, 10
min), washed with 5 mL of C2H5OH, dried under N2, and
dissolved in 1.5 mL of 50 mM Tris-HCl buffer (pH 6.8) contain-
ing 10 mM MgCl2. Enzymatic digestion was done at 37 °C as
described elsewhere (24), for 3 h in the DNase I step and
overnight (under Ar) with phosphodiesterase and acid phos-
phatase.
Aliquots (0.20-1.0 mL) were injected onto the 10 × 250 mm
Beckman Ultrasphere octadecylsilane HPLC column, and the
column was eluted using a gradient of increasing CH3OH in 50
mM NH4HCO2 (pH 5.5) (2.7% CH3OH for 5 min, then 2.7-27%
over 15 min, 27-36% over 4 min, 36% for 3 min, and then return
to 2.7%). The flow rate was 4.0 mL min-1, and the effluent was
mixed on-line with 75 mM aqueous HClO4, pumped at 2.0 mL
min-1 (LDC Model III pump, Milton-Roy, Rivera Beach, FL;
Anspec TC-50 mixer, Fiatron Laboratory Systems, Oconomowoc,
WI), and passed through a fluorescence detector (Varian 9070,
Varian, Walnut Creek, CA, excitation 290 nm, emission 340 nm).
The mixture had a pH of 3.1, sufficient to give maximum
fluorescence of 3,N4-ꢀ-dCyd (25) but high enough to suppress
much of the fluorescence of the natural nucleosides. Estimates
of 3,N4-ꢀ-dCyd were made by comparison with external stan-
dards, calibrated using ꢀ272 ) 11.7 mM cm-1 (25).
F igu r e 1. ESI precursor and product spectra of HO-etha-
noGua. A 0.10 µg µL-1 solution of HO-ethanoGua, dissolved in
a 70:30 (v/v) mixture of (i) 10 mM NH4CH3CO2 buffer, pH 5.5,
and (ii) 0.05% CH3CO2H in CH3OH (v/v), was infused at a rate
of 20 µL min-1 for 1 min along with additional mobile phase
(same as solution used to dissolve sample, 230 µL min-1). The
signal was averaged, and the scan time was 1 scan s-1
Precursor spectrum; (B) product spectrum.
. (A)
Resu lts a n d Discu ssion
An a lysis of H O-et h a n oGu a a n d 1,N2-E-Gu a b y
HP LC/MS/MS. Assays were developed using stable
isotope internal standards (+3 mass units). The de-
scribed SRM parameters were used to establish standard
curves for m/ z 176/179 to 121/1232 in the case of 1,N2-
ꢀ-Gua and 194/197 to 176/179 transitions in the case of
HO-ethanoGua (Figure 1). The lower limit of quantita-
tion of standards was approximately 3 pmol for each.
Meth od Develop m en t. A full scan mass spectrum
of the HO-ethanoGua adduct (unlabeled) is shown in
Figure 1A. The predominant ion in the spectrum is m/ z
194, the [MH]+ ion of the adduct. The product ion
spectrum for the HO-ethanoGua adduct (unlabeled) is
shown as Figure 1B. The base peak at m/ z 176 results
from the loss of a molecule of H2O from the protonated
molecular species, m/ z 194. Similar full scan and
product ion mass spectra were obtained for the labeled
internal standard except that the corresponding ions
were, as expected, 3 daltons greater than for the unla-
beled compound.
For the unlabeled 1,N2-ꢀ-Gua adduct (data not shown),
the full scan mass spectrum revealed a predominant ion
at m/ z 176, corresponding to the [MH]+ of this adduct.
Mass spectrometric acquisition conditions were similar
to those described for the HO-ethanoGua adduct. The
product ion spectrum (data not shown) for the 1,N2-ꢀ-
Gua adduct (unlabeled) yielded a product ion at m/ z 121
resulting from the loss of C2N2H3 from the protonated
molecular species, m/ z 176. For the labeled 1,N2-ꢀ-Gua
adduct, the product ion at m/ z 123 resulted from the loss
of CC*N2H3 (C* indicating a label in the case of the
internal standard) from the protonated molecular species,
m/ z 179.
In the analysis of DNA treated with 2-chlorooxirane,
direct injection of depurinated samples onto HPLC/MS/
MS did not lead to useful data due to the suppression of
ionization by the presence of large amounts of Ade and
Gua resulting from acid depurination (results not pre-
sented). Prior HPLC workup removed the bulk of these
natural bases and provided considerable enrichment for
the DNA adducts. With this approach, the lowest detect-
able level of HO-ethanoGua was 24 pmol (µmol of Gua)-1
,
equivalent to 6 pmol (µmol of DNA base)-1 (Figure 2).
The lowest detectable level of 1,N2-ꢀ-Gua was ∼4 pmol
(µmol of Gua)-1, or 1 pmol (µmol of DNA base)-1
.
An a lysis of HO-eth a n o-d Gu o by HP LC/MS/MS.
Analysis of nucleosides was considered in order to
improve the assay of modified bases. There are two
inherent advantages to this approach. One is that the
problem of ionization suppression of bases due to depu-
rination is avoided and the need for preliminary HPLC

Analysis of 2-Chlorooxirane-Derived DNA Adducts
Chem. Res. Toxicol., Vol. 10, No. 2, 1997 245
F igu r e 2. Analysis of 1,N2-ꢀ-Gua and HO-ethanoGua from
DNA treated with 2-chlorooxirane. (A) Standard curves for
1,N2-ꢀ-Gua (O) and HO-ethanoGua (0) were prepared using
HPLC/SRM/MS/MS by injecting 17.5 ng of [2,9,9a-13C]1,N2-ꢀ-
Gua and 17.5 ng of [2,9,9a-13C]HO-ethanoGua plus the indicated
amounts of unlabeled adducts, in 35 µL. The ratios are of the
M/M + 3 signals for 1,N2-ꢀ-Gua (176 to 121 and 179 to 123
transitions)2 and N2-(2-oxoethylGua) (194 to 176 and 197 to 179
transitions). (B) The HPLC/SRM/MS/MS assay was used to
estimate the levels of 1,N2-ꢀ-Gua (O) and HO-ethanoGua (0) in
DNA treated with the indicated concentration of 2-chloroox-
irane.
F igu r e 4. Formation of HO-ethano-dGuo in DNA treated with
2-chlorooxirane. The method shown in Figure 2 was utilized.
The inset shows a standard curve prepared by injecting mixtures
of 2.5 ng (8 pmol) of [2,9,9a-13C]HO-ethano-dGuo along with the
indicated amounts of unlabeled HO-ethano-dGuo.
F igu r e 3. HPLC of HO-ethano-dGuo. HPLC/SRM/MS/MS was
done monitoring the 310 to 194 or the 313 to 197 transition (loss
of deoxyribose). A standard mixture of (A) 0.40 pmol of
unlabeled and (B) 8 pmol of [2,9,9a-13C]HO-ethano-dGuo was
injected and analyzed in the respective channels. Part C shows
the 310 to 194 transition trace for HPLC/SRM/MS/MS of
digested DNA that had been treated with 2-chlorooxirane (10
mM), with a section of the time scale expanded.
F igu r e 5. Formation of 3,N4-ꢀ-dCyd in DNA treated with
2-chlorooxirane. Levels were estimated by HPLC/fluorescence
as described after treatment of calf thymus DNA (5 mg mL-1
with the indicated concentrations of 2-chlorooxirane.
)
is circumvented. The other is that the short tR of HO-
ethanoGua (16) is avoided and the concentration of CH3-
OH in the HPLC mobile phase is higher when the
compound enters the mass spectrometer. We also changed
the SRM mode to monitor the loss of the 2′-deoxyribose
moiety from the protonated molecular ion, i.e., m/ z 310/
194 and 313/197 (Figure 3).
development of a reversed-phase HPLC system using a
pH gradient of 3.5-7.0, with increasing CH3OH. How-
ever, we found that the 3,N4-ꢀ-dCyd resolution was
attenuated at the lower pH. An alternate system was
used in which the nucleosides were separated at pH 5.5
(24) and the pH of the effluent was lowered by post-
column mixing prior to on-line fluorescence analysis. The
limit of detection using this method was ∼20 pmol of
3,N4-ꢀ-dCyd. The level of this adduct formed by treat-
ment of DNA (5 mg mL-1) with 10 mM 2-chlorooxirane
was ∼7 pmol (µmol of DNA base)-1 (Figure 5).
The system worked well for HO-ethano-dGuo, with a
lower limit of quantitation of ∼0.4 pmol. With 2-chlo-
rooxirane-treated DNA samples, adduct levels could be
determined as low as 10 pmol of HO-ethano-dGuo (µmol
of dGuo)-1 or 2.5 pmol (µmol of DNA base)-1 (Figure 4).
This approach is somewhat more sensitive than that of
the analysis of bases (vide supra) and also avoids the
prior HPLC step. Moreover, the estimated levels of HO-
ethanoGua adducts from the two methods are very
similar.
Com p a r ison of Levels of Ad d u cts. Levels of all
known DNA adducts derived from vinyl chloride have
now been estimated in DNA treated with 2-chlorooxirane,
either in this report or in earlier reports from this
laboratory (Table 1)3 (16, 27). Generally, the level of
adducts has been shown to be proportional to the
concentration of 2-chlorooxirane. The levels of adducts
formed with 10 mM 2-chlorooxirane are presented for
comparative purposes.
Although the response of standard 1,N2-ꢀ-dGuo was
very similar to that of HO-ethano-dGuo, we were unsuc-
cessful in utilizing this for the analysis of DNA samples
due to apparent coeluting contaminants.
The rank of adducts formed shows some similarity to
the chemical reactivity of 2-halooxiranes with nucleo-
Estim a tion of 3,N4-E-d Cyd in 2-Ch lor ooxir a n e-
Tr ea ted DNA. All of the known ꢀ adducts of DNA bases
have been quantified in calf thymus DNA treated with
2-chlorooxirane, with the exception of 3,N4-ꢀ-dCyd. 3,N4-
ꢀ-dCyd is fluorescent, but the intensity is attenuated
above pH 3.5 (25). Bedell et al. (26) reported the
3
In an earlier publication (16; Figure 1), the plot symbols for 1,N2-
ꢀ-Gua and N2,3-ꢀ-Gua had been inadvertently reversed. Corrected
values are shown in Table 1 with the level of N2,3-ꢀ-Gua > 1,N2-ꢀ-
Gua.

246 Chem. Res. Toxicol., Vol. 10, No. 2, 1997
Ta ble 1. DNA Ad d u cts F or m ed w ith 2-Ch lor ooxir a n e
Mu¨ller et al.
concn formed with 10 mM
2-chlorooxirane, pmol
(µmol of DNA base)-1
adducta
method
ref
N7-(2-oxoethyl)Gua
10 000
500
24
HPLC/fluorescence
HPLC/fluorescence
HPLC/MS/MS
(27)
(27)
1,N6-ꢀ-dAdo
HO-ethanoGua
HO-ethano-dGuo
N2,3-ꢀ-Gua
this work
this work
(16, 27)3
this work
(16)3
29
16
HPLC/MS/MS
HPLC/fluorescence
HPLC/fluorescence
HPLC/fluorescence
HPLC/MS/MS
3,N4-ꢀ-dCyd
7
1,N2-ꢀ-Gua
∼1b
2.5
this work
a
b
The name of the adduct indicates whether the base or deoxyribonucleoside was assayed. In other work, a 15 mg mL-1 concentration
of herring sperm DNA was used instead of 5 mg of calf thymus DNA mL-1 in ref 27 and this work, and a value of 6.5 pmol (µmol of DNA
base)-1 was obtained.
sides. E.g., the highest yield of an ꢀ adduct formed from
alkaline 2,2-dibromoethanol was for 1,N6-ꢀ-Ado (16, 19),
and the yield of N7-(2-oxoethyl)Gua was very high (16).
HO-ethanoGuo appeared to be readily formed from Guo
(16), although the initial characterization was done only
with nucleosides and not DNA. The order of levels will
probably not match that seen in vivo, due to differential
stability and rates of enzymatic repair. For instance,
Swenberg and his associates have reported differences
in the persistence of ꢀ adducts in rat liver DNA (28).
However, all known ꢀ adducts are substrates for a human
glycosylase (29), and details of overall human repair rates
are unknown. No information is available about in vivo
levels of some of the adducts, particularly HO-ethanoGua
and 1,N2-ꢀ-Gua.
(HO-ethanoGua and 1,N2-ꢀ-Gua) in DNA prepared from
unexposed animals and humans. Although more labori-
ous, GC/MS assays with derivatives (e.g., use of halogens
and chemical ionization MS) might prove to be more
appropriate. However, the available new results lead to
a comparison of the levels of the six known DNA adducts
formed by treatment with 2-chlorooxirane: N7-(2-oxoet-
hyl)Gua . 1,N6-ꢀ-Ade > HO-ethanoGua > N2,3-ꢀ-Gua >
3,N4-ꢀ-Cyt > 1,N2-ꢀ-Gua (Table 1). This pattern should
be considered in the interpretation of previous and future
investigations of the mutagenicity of these adducts and
their relevance to the carcinogenicity of vinyl chloride (4),
as well as other procarcinogens that generate these
adducts (8-11, 38).
Con sid er a tion s of Biologica l Activity. Some in-
formation is available regarding the miscoding potentials
and mutagenicity of several of the adducts. Relatively
simple in vitro assays suggest that the major adduct, N7-
(2-oxoethyl)Gua, is not miscoding (30). 1,N6-ꢀ-Ade has
been inserted into a vector and, with proper attention to
a chemical artifact of the adduct, the mutagenicity of this
adduct in bacteria seems to be low (31). N2,3-ꢀ-Gua has
been reported to be highly miscoding when examined in
bacterial systems (32, 33). However, even semiquanti-
tative estimates of mutagenicity and misincorporation
are difficult to make because this adduct has not been
successfully inserted into a defined oligonucleotide, due
to instability (34, 35), and mutagenesis experiments are
dependent upon either (in vivo) incorporation of N2,3-ꢀ-
dGTP into vectors, from which mutations are scored, or
else (in vitro) incorporation of N2,3-dGTP into oligonucle-
otides in polymerase extension experiments (32, 33).
Some conflicting results have been reported for 3,N4-ꢀ-
Cyt. One report indicates a high bacterial mutagenicity
for 3,N4-ꢀ-Cyt (36); two others have much lower bacterial
mutagenicity (31, 37). However, one of the latter groups
has also found much higher mutagenicity of the same
adduct in a mammalian system (37). It should be
emphasized that the major adducts, N7-(2-oxoethylGua)
and 1,N6-ꢀ-Ade, do not appear to be mutagenic, but the
amounts of these relative to the minor products is an
issue and even weak mutagenicity (per adduct) might
prove to be important.
Ack n ow led gm en t. This work was supported in part
by U.S. Publich Health Service (USPHS) Grants R35
CA44353 and P30 ES00267. M.M. was supported in part
by a fellowship from the Deutsche Forschungsgemein-
schaft.
Refer en ces
(1) Gedigk, P., Mu¨ller, R., and Bechtelsheimer, H. (1975) Morphology
of liver damage among polyvinyl chloride production workers. A
report on 51 cases. Ann. N.Y. Acad. Sci. 246, 278-285.
(2) Buffler, P. A., Wood, S., Eifler, C., Suarez, L., and Kilian, D. J .
(1979) Mortality experience of workers in
a vinyl chloride
monomer production plant. J . Occup. Med. 21, 195-203.
(3) Nicholson, W. J ., Hammond, E. C., Seidman, H., and Selikoff, I.
J . (1975) Mortality experience of a cohort of vinyl chloride-
polyvinyl chloride workers. Ann. N.Y. Acad. Sci. 246, 225-230.
(4) Bolt, H. M. (1988) Roles of etheno-DNA adducts in tumorigenicity
of olefins. Crit. Rev. Toxicol. 18, 299-309.
(5) Guengerich, F. P., Min, K. S., Persmark, M., Kim, M. S.,
Humphreys, W. G., and Cmarik, J . L. (1994) Dihaloalkanes and
polyhaloalkenes. In DNA Adducts: Identification and Signifi-
cance (Hemminki, K., Dipple, A., Shuker, D. E. G., Kadlubar, F.
F., Segerba¨ck, D., and Bartsch, H., Eds.) IARC Scientific Publica-
tions No. 125, pp 57-72, IARC Publications, Lyon.
(6) Barrio, J . R., Secrist, J . A., III, and Leonard, N. J . (1972)
Fluorescent adenosine and cytidine derivatives. Biochem. Bio-
phys. Res. Commun. 46, 597-604.
(7) Sattsangi, P. D., Leonard, N. J ., and Frihart, C. R. (1977) 1,N2-
Ethenoguanine and N2,3-ethenoguanine. Synthesis and com-
parison of the electronic spectral properties of these linear and
angular triheterocycles related to the Y bases. J . Org. Chem. 42,
3292-3296.
(8) Singer, B., and Bartsch, H. (1986) The Role of Cyclic Nucleic Acid
Adducts in Carcinogenesis and Mutagenesis, IARC Scientific
Publications, Lyon.
Con clu sion s. We have developed HPLC/SRM/MS/MS
assays for HO-ethanoGua and 1,N2-ꢀ-Gua in DNA. The
work provides evidence that the former adduct is actually
formed in DNA treated with 2-chlorooxirane. The HPLC/
SRM/MS/MS assays developed here are useful but for
several technical reasons not sensitive enough to do
assays on DNA of animals treated in vivo with vinyl
halides or to determine the levels of these two adducts
(9) Leithauser, M. T., Liem, A., Stewart, B. C., Miller, E. C., and
Miller, J . A. (1990) 1,N6-Ethenoadenosine formation, mutagenicity
and murine tumor induction as indicators of the generation of
an electrophilic epoxide metabolite of the closely related carcino-
gens ethyl carbamate (urethane) and vinyl carbamate. Carcino-
genesis 11, 463-473.
(10) Guengerich, F. P., Geiger, L. E., Hogy, L. L., and Wright, P. L.
(1981) In vitro metabolism of acrylonitrile to 2-cyanoethylene

Analysis of 2-Chlorooxirane-Derived DNA Adducts
Chem. Res. Toxicol., Vol. 10, No. 2, 1997 247
oxide, reaction with glutathione, and irreversible binding to
proteins and nucleic acids. Cancer Res. 41, 4925-4933.
(11) Kronberg, L., Sjo¨holm, R., and Karlsson, S. (1992) Formation of
3,N4-ethenocytidine, 1,N6-ethenoadenosine, and 1,N2-ethenogua-
nosine in reactions of mucochloric acid with nucleosides. Chem.
Res. Toxicol. 5, 852-855.
(25) Barrio, J . R., Sattsangi, P. D., Gruber, B. A., Dammann, L. G.,
and Leonard, N. J . (1976) Species responsible for the fluorescence
of 3,N4-ethenocytidine. J . Am. Chem. Soc. 98, 7408-7414.
(26) Bedell, M. A., Dyroff, M. C., Doerjer, G., and Swenberg, J . A.
(1986) Quantitation of etheno adducts by fluorescence detection.
In The Role of Cyclic Nucleic Acid Adducts in Carcinogenesis and
Mutagenesis (Singer, B., and Bartsch, H., Eds.) pp 425-436, IARC
Scientific Publications, Lyon.
(27) Guengerich, F. P. (1992) Roles of the vinyl chloride oxidation
products 1-chlorooxirane and 2-chloroacetaldehyde in the in vitro
formation of etheno adducts of nucleic acid bases. Chem. Res.
Toxicol. 5, 2-5.
(28) Swenberg, J . A., Fedtke, N., Ciroussel, F., Barbin, A., and Bartsch,
H. (1992) Etheno adducts formed in DNA of vinyl chloride-exposed
rats are highly persistent in liver. Carcinogenesis 13, 727-729.
(29) Dosanjh, M. K., Chenna, A., Kim, E., Fraenkel-Conrat, H.,
Samson, L., and Singer, B. (1994) All four known cyclic adducts
formed in DNA by the vinyl chloride metabolite chloroacetalde-
hyde are released by a human DNA glycosylase. Proc. Natl. Acad.
Sci. U.S.A. 91, 1024-1028.
(30) Barbin, A., Laib, R. J ., and Bartsch, H. (1985) Lack of miscoding
properties of 7-(2-oxoethyl)guanine, the major vinyl chloride-
DNA adduct. Cancer Res. 45, 2440-2444.
(31) Basu, A. K., Wood, M. L., Niedernhofer, L. J ., Ramos, L. A., and
Essigmann, J . M. (1993) Mutagenic and genotoxic effects of three
vinyl chloride-induced DNA lesions: 1,N6-ethenoadenine, 3,N4-
ethenocytosine, and 4-amino-5-(imidazol-2-yl)imidazole. Bio-
chemistry 32, 12793-12801.
(32) Singer, B., Kusmierek, J . T., Folkman, W., Chavez, F., and
Dosanjh, M. K. (1991) Evidence for the mutagenic potential of
the vinyl chloride induced adduct, N2,3-etheno-deoxyguanosine,
using a site-directed kinetic assay. Carcinogenesis 12, 745-747.
(33) Cheng, K. C., Preston, B. D., Cahill, D. S., Dosanjh, M. K., Singer,
B., and Loeb, L. A. (1991) The vinyl chloride DNA derivative, N2,3-
ethenoguanine, produces G f A transitions in Escherichia coli.
Proc. Natl. Sci. U.S.A. 88, 9974-9978.
(12) Sodum, R. S., and Chung, F. L. (1988) 1,N2-Ethenodeoxyguanosine
as a potential marker for DNA adducts formation by trans-4-
hydroxy-2-nonenal. Cancer Res. 48, 320-323.
(13) Okazaki, O., Persmark, M., and Guengerich, F. P. (1993) N-Nitro-
so-N-methylvinylamine: reaction of the epoxide with guanyl and
adenyl moieties to yield adducts derived from both parts of the
molecule. Chem. Res. Toxicol. 6, 168-173.
(14) Fedtke, N., Boucheron, J . A., Walker, V. E., and Swenberg, J . A.
(1990) Vinyl chloride-induced DNA adducts. II. Formation and
persistence of 7-(2′-oxoethyl)guanine and N2,3-ethenoguanine in
rat tissue DNA. Carcinogenesis 11, 1287-1292.
(15) El Ghissassi, F., Barbin, A., Nair, J ., and Bartsch, H. (1995)
Formation of 1,N6-ethenoadenine and 3,N4-ethenocytosine by lipid
peroxidation products and nucleic acid bases. Chem. Res. Toxicol.
8, 278-283.
(16) Guengerich, F. P., Persmark, M., and Humphreys, W. G. (1993)
Formation of 1,N2- and N2,3-ethenoguanine derivatives from
2-halooxiranes: isotopic labeling studies and formation of a
hemiaminal derivative of N2-(2-oxoethyl)guanine. Chem. Res.
Toxicol. 6, 635-648.
(17) Walling, C., and Fredricks, P. S. (1962) Positive halogen com-
pounds. IV. Radical reactions of chlorine and t-butyl hypochlorite
with some small ring compounds. J . Am. Chem. Soc. 84, 3326-
3331.
(18) Guengerich, F. P., Crawford, W. M., J r., and Watanabe, P. G.
(1979) Activation of vinyl chloride to covalently bound metabo-
lites: roles of 2-chloroethylene oxide and 2-chloroacetaldehyde.
Biochemistry 18, 5177-5182.
(19) Guengerich, F. P., and Raney, V. M. (1992) Formation of etheno
adducts of adenosine and cytidine from 1-halooxiranes. Evidence
for a mechanism involving initial reaction with the endocyclic
nitrogens. J . Am. Chem. Soc. 114, 1074-1080.
(20) Scheller, N., Sangaiah, R., Ranasinghe, A., Amarnath, V., Gold,
A., and Swenberg, J . A. (1995) Synthesis of [4,5,6,8-13C4]guanine,
a reagent for the production of internal standards of guanyl DNA
adducts. Chem. Res. Toxicol. 8, 333-337.
(34) Kusmierek, J . T., Folkman, W., and Singer, B. (1989) Synthesis
of N2,3-ethenodeoxyguanosine, N2,3-deoxyguanosine 5′-phosphate,
and N2,3-ethenodeoxyguanosine 5′-triphosphate. Stability of the
glycosyl bond in the monomer and in poly(dG,ꢀdG-dC). Chem.
Res. Toxicol. 2, 230-233.
(35) Khazanchi, R., Yu, P. L., and J ohnson, F. (1993) N2,3-Etheno-2′-
deoxyguanosine [8,9-dihydro-9-oxo-2′-deoxy-3-â-D-ribofuranosyl-
imidazo[2,1-b]purine]: a practical synthesis and characterization.
J . Org. Chem. 58, 2552-2556.
(21) Shapiro, R., Cohen, B. I., Shiuey, S. J ., and Maurer, H. (1969)
On the reaction of guanine with glyoxal, pyruvaldehyde, and
kethoxal, and the structure of the acylguanines. A new synthesis
of N2-alkylguanines. Biochemistry 8, 238-245.
(22) Mu¨ller, M., Hutchinson, L. K., and Guengerich, F. P. (1996)
Addition of deoxyribose to guanine and modified DNA bases by
Lactobacillus helveticus trans-N-deoxyribosylase. Chem. Res.
Toxicol. 9, 1140-1144.
(23) Chaudhary, A. K., Nokubo, M., Marnett, L. J ., and Blair, I. A.
(1994) Analysis of the malondialdehyde-2′-deoxyguanosine adduct
in rat liver DNA by gas chromatography/electron capture negative
chemical ionization mass spectrometry. Biol. Mass Spectrom. 23,
457-464.
(24) Ciroussel, F., Barbin, A., Eberle, G., and Bartsch, H. (1990)
Investigations on the relationship between DNA ethenobase
adduct levels in several organs of vinyl chloride-exposed rats and
cancer susceptibility. Biochem. Pharmacol. 39, 1109-1113.
(36) Simha, D., Palejwala, V. A., and Humayun, M. Z. (1991) Mech-
anisms of mutagenesis by exocyclic DNA adducts. Construction
and in vitro template characteristics of an oligonucleotide bearing
a single site-specific ethenocytosine. Biochemistry 30, 8727-8735.
(37) Moriya, M., Zhang, W., J ohnson, F., and Grollman, A. P. (1994)
Mutagenic potency of exocyclic DNA adducts: marked differences
between Escherichia coli and simian kidney cells. Proc. Natl.
Acad. Sci. U.S.A. 91, 11899-11903.
(38) Leonard, N. J . (1992) Etheno-bridged nucleotides in structural
diagnosis and carcinogenesis. Chemtracts: Biochemistry and
Molecular Biology 3, 273-297.
TX960124I