Synthesis of a benzene metabolite adduct, 3''-hydroxyl-1,N2-benzetheno- 2'-deoxyguanosine, and its site-specific incorporation into DNA oligonucleotides

Article abstract of DOI:10.1021/tx960168r

p-Benzoquinone (p-BQ) is a stable metabolite of benzene and a number of other drugs and chemicals. 2'-Deoxyguanosine was allowed to react with p-BQ in aqueous solution (pH 4.5, 7.4, and 9.3). At pH 7.4 and 9.3 one major product was found in low yield; at pH 4.5 no product was detected. In nonaqueous conditions (DMF or DMSO, in the presence of K₂CO₃), an unstable intermediate with two p-BQ moieties was found which slowly converted to the product formed in aqueous solution. These products were isolated by silica gel, column chromatography, or reverse-phase HPLC and characterized by UV, ¹H NMR, FAB-MS, and electrospray MS. The major stable product of the reaction of dG with p-BQ is an exocyclic compound, 3''-hydroxy-1,N²- benzetheno-2'-deoxyguanosine (P-BQ-dG). Incorporation of the adduct into oligonucleotides requires the protection of three hydroxyl groups (C7, 5', 3') and the amine group at N5. The exocyclic hydroxyl and the amine group were protected by acylation after protecting the 5'- and the 3'-hydroxyl groups of the sugar moiety by 4,4'-dimethoxytrityl and a cyanoethyl N,N- diisopropylphosphoramidite group, respectively. This is the first time the fully protected phosphoramidite of p-BQ-dG has been prepared and used in the synthesis of site specifically modified oligonucleotides. After deprotection with 1,3-diazabicyclo[5.4.0]undec-7-ene (DBU), in ethanol, oligomers purified by gel electrophoresis and HPLC. Enzymatic hydrolysis and analysis by HPLC confirmed the presence of p-BQ-dG in the oligomers. These oligomers are now under investigation for their biochemical properties.

Full text of DOI:10.1021/tx960168r

Chem. Res. Toxicol. 1997, 10, 165-171
165
Syn th esis of a Ben zen e Meta bolite Ad d u ct,
3′′-Hyd r oxy-1,N2-ben zeth en o-2′-d eoxygu a n osin e, a n d Its
Site-Sp ecific In cor p or a tion in to DNA Oligon u cleotid es
A. Chenna and B. Singer*
Donner Laboratory, Life Sciences Division, Lawrence Berkeley National Laboratory,
University of California, Berkeley, California 94720
Received September 27, 1996X
p-Benzoquinone (p-BQ) is a stable metabolite of benzene and a number of other drugs and
chemicals. 2′-Deoxyguanosine was allowed to react with p-BQ in aqueous solution (pH 4.5,
7.4, and 9.3). At pH 7.4 and 9.3 one major product was found in low yield; at pH 4.5 no product
was detected. In nonaqueous conditions (DMF or DMSO, in the presence of K2CO3), an unstable
intermediate with two p-BQ moieties was found which slowly converted to the product formed
in aqueous solution. These products were isolated by silica gel, column chromatography, or
reverse-phase HPLC and characterized by UV, 1H NMR, FAB-MS, and electrospray MS. The
major stable product of the reaction of dG with p-BQ is an exocyclic compound, 3′′-hydroxy-
1,N2-benzetheno-2′-deoxyguanosine (p-BQ-dG). Incorporation of the adduct into oligonucleotides
requires the protection of three hydroxyl groups (C7, 5′, 3′) and the amino group at N5. The
exocyclic hydroxyl and the amino group were protected by acylation after protecting the 5′-
and the 3′-hydroxyl groups of the sugar moiety by 4,4′-dimethoxytrityl and a cyanoethyl N,N-
diisopropylphosphoramidite group, respectively. This is the first time the fully protected
phosphoramidite of p-BQ-dG has been prepared and used in the synthesis of site specifically
modified oligonucleotides. After deprotection with 1,3-diazabicyclo[5.4.0]undec-7-ene (DBU),
in ethanol, oligomers purified by gel electrophoresis and HPLC. Enzymatic hydrolysis and
analysis by HPLC confirmed the presence of p-BQ-dG in the oligomers. These oligomers are
now under investigation for their biochemical properties.
Several adducts have been identified from the in vitro
reaction of p-BQ with nucleosides, nucleotides, and DNA
(17-21); however, the relationship between the adduct
structure and biological effects has not yet been estab-
lished. Both dC and dA form single exocyclic benzetheno
adducts (19,20). Multiple adducts have been reported
(17,18) with deoxyguanosine in p-BQ-treated DNA. Only
one adduct has been characterized resulting from reac-
tion of dG with p-BQ at neutrality (17,18). The total
modified bases formed are less than 1/106 bases in DNA
from human bone marrow or HL-60 cells when treated
with p-BQ in vitro (22). With such low levels of reaction
in cells, 32P-postlabeling was the only method of detection
(19-21).
In order to study how p-BQ exerts its biological effects,
it was necessary to develop syntheses of both the dA and
dC adducts and their phosphoramidites. We have re-
cently reported the large-scale synthesis of p-BQ-dC and
p-BQ-dA adducts and the site-specific incorporation of
their phosphoramidates into a defined deoxyoligonucle-
otides (23). In this work, we report the preparative-scale
synthesis of 3′′-hydroxy-1,N2-benzetheno-2′-deoxygua-
nosine (p-BQ-dG) and its phosphoramidite and the suc-
cessful site-specific incorporation into DNA oligonucle-
otides. These oligonucleotides are now being studied for
their effect on replication, mutation, and repair in vitro.
In tr od u ction
Humans are exposed to benzene from many environ-
mental sources, including automobile exhaust and ciga-
rette smoke, and in industries such as rubber manufac-
ture, and crude oil, gasoline, and chemical manufacturing.
Benzene causes acute leukemia in humans (1) and bone
marrow toxicity (2). p-Benzoquinone (p-BQ)1 is a stable
metabolites of benzene (3-6), as well as of a number of
drugs and chemicals (7-9). Benzene must be metabo-
lized to exert its toxic and leukemogenic effects (10-12).
The initial step in one pathway is oxidation of benzene
by cytochrome P450 to form benzene oxide, which further
yields phenol, and then hydroquinone which in turn is
oxidized to p-BQ. The latter accumulates in the bone
marrow (13,14), which is the site of the observed myelo-
toxic and leukemogenic effects (12,15).
p-BQ is one of a large family of R,â-unsaturated
carbonyl compounds including acrolein, crotonaldehydes,
malondialdehyde, and methyl vinyl ketone. In most
cases, both mono- and diadducts are formed when these
compounds are reacted with deoxyguanosine via Michael
addition reaction. Shapiro et al. (16) reported the forma-
tion of more than one product resulting from the reaction
of the R,â-unsaturated carbonyl compounds glyoxal and
acrolein with 2′-deoxyguanosine.
* To whom correspondence and reprint requests should be addressed
at: Donner Laboratory, University of California, Berkeley, CA 94720.
Telephone: 510-642-0637. Fax: 510-486-6488.
Ma ter ia ls a n d Meth od s
X Abstract published in Advance ACS Abstracts, J anuary 1, 1997.
Ch em ica ls a n d Rea gen ts. Ca u tion : p-BQ is highly toxic
and should be handled carefully. Many organic solvents used
in this work should be stored and used in a well-ventilated hood.
2′-Deoxyguanosine (dG) was purchased from Sigma Chemical
Co. (St. Louis, MO). p-BQ, DMSO, 4,4′-dimethoxytrityl chloride
1
Abbreviations: p-BQ, p-benzoquinone; dG, 2′-deoxyguanosine;
DBU, 1,3-diazabicyclo[5.4.0]undec-7-ene; BAP, bacterial alkaline phos-
phatase; SVDE, snake venom phosphodiesterase; FAB, fast-atom
bombardment; DMTCl, 4,4′-dimethoxytrityl chloride; DMF, N,N-
dimethylformamide.
S0893-228x(96)00168-3 CCC: $14.00 © 1997 American Chemical Society

166 Chem. Res. Toxicol., Vol. 10, No. 2, 1997
Chenna and Singer
(DMTCl), pyridine, triethylamine, 2-cyanoethyl N,N-diisopropyl-
chlorophosphoramidite, isobutyric anhydride, silica gel, toluene,
hexane, and 1,3-diazabicyclo[5.4.0]undec-7-ene (DBU), were
from Aldrich Chemical Co. (Milwaukee, WI). Methanol, meth-
ylene chloride, benzene, acetonitrile, and ethyl acetate were from
J . T. Baker Inc. (Phillipsburg, NJ ). Potassium carbonate,
phosphorus pentoxide (P2O5), and thin-layer chromatography
(TLC) silica plates were purchased from EM Science (Gibbston,
NJ ), and calcium hydride was from Fluka Chemika (Ronkonko-
ma, NY). Bacterial alkaline phosphatase (BAP) and snake
venom phosphodiesterase (SVDE) were from Pharmacia Biotech
Inc. (Alameda, CA). The PAC phosphoramidates were pur-
chased from Pharmacia (Piscataway, NJ ). All solvents were
dried by distillation over CaH2.
(the product decomposed to a black material). UV: product (1)
(pH 6.5) λmax 233, 272, 328 nm; (pH 12) λmax 251, 291, 361 nm;
(pH 1) λmax 235, 271, 284, 329 nm. These data are similar to
those reported by J owa et al. (17). FAB/MS (positive ion): m/z
380 (15) (M + Na)+, 358 (30) MH+, 264 (12) (BH + Na)+, 242
+
(100) BH2
.
1H NMR (DMSO-d6, 400 MHz): δ 9.485 (S, 1H,
exch, HO-7), 8.124 (S, 1H, H-2), 8.022 (d, 1H, J ) 2.3 Hz, H-6),
7.230 (d, 1H, J ) 8.6 Hz, H-9), 6.866 (d, 1H, J ) 2.35 Hz, H-8),
6.531 (′S, 1H, exch, HN-5), 6.257 (t, 1H, J ) 6.5 Hz, H-1′), 5.329
(S, 1H, exch, HO-3′), 5.011 (S, 1H, exch, HO-5′), 4.369 (S, 1H,
H-4′), 3.849 (m, 1H, H-3′), 3.518 (m, 2H, H-5′), 2.608 (m, 1H,
H-2′), 2.270 (m, 1H, H-2′′).
The above reaction was repeated in DMF/K2CO3 and pro-
duced the same products as shown by the monitoring of the
reaction by TLC and HPLC (Figure 1). In another reaction, the
base, K2CO3, was replaced by NaOH pellets in DMSO and gave
the major product and only traces of the intermediate product
throughout the reaction course.
Proton (400 MHz) NMR spectra were recorded in DMSO-d6
as solvent, and D2O exchanges were carried out to assign
exchangeable protons, unless otherwise indicated. NMR spectra
were recorded using a Brucker AM400 spectrometer and are
reported in parts per million (ppm) relative to an internal
standard of tetramethylsilane. 31P NMR spectra were recorded
on a Bruker AM400 spectrometer, and the chemical shifts are
reported relative to an external standard of phosphoric acid.
Fast-atom bombardment (FAB) spectra were obtained on a
VG70 SE Instrument, and glycerin or thioglycerin was used as
a matrix. Electrospray mass spectra were obtained on a VG
Bio-Q Instruments mass spectrometer. Loop injections of the
sample were made into solvent (50:50, acetonitrile/water) flow-
ing into the electrospray source (4 mL/min). Analytical and
semipreparative HPLC were conducted using a Hewlett Packard
1050 photodiode array detector and quaternary gradient pump
(solvent delivery system).
The reaction of dG with p-BQ was also performed in aqueous
solutions, at pH 4.5, 7.4, and 9.3 using potassium phosphate
buffers. HPLC analysis of these reactions after 24 h showed
that the reaction at pH 7.4 and 9.3 produced the same stable
product as in nonaqueous solutions (DMSO/K2CO3) and only
traces of the unstable intermediate, but at pH 4.5 no reaction
was detected.
Sm a ll-Sca le Rea ction of 2′-Deoxygu a n osin e w ith p-BQ.
2′-Deoxyguanosine (40 mg, 0.145 mmol) was dissolved in 2 mL
of dry DMF, and potassium carbonate (38 mg, 0.28 mmol, 2
equiv) was added and stirred for 20 min at room temperature
and then p-benzoquinone (18 mg, 0.168 mmol, 1.2 equiv)
dissolved in 1 mL of dry DMF was added dropwise. The reaction
color changed from colorless to yellow to green to dark. The
reaction mixture was allowed to continue stirring at room
temperature while it was monitored by HPLC at increasing
times (5 min, 30 min, 1 h, 2.5 h, 3.5 h, 4.5 h, 5.5 h, 6.3 h, 12 h,
22.5 h) by using system 3. HPLC analysis showed the formation
of an intermediate product which slowly converted to the stable
desired product overnight (Figure 1). Analytical data for the
intermediate is as follows. UV: unstable intermediate (pH 6.5)
λmax 238, 273, 334 nm; (pH 12) λmax 250, 298, 366 nm; (pH 1)
HP LC. Solvent systems included solvent A (acetonitrile),
solvent B (triethylammonium acetate; 0.1 M, pH 7.0), and
solvent C (potassium phosphate buffer; 0.01 M, pH 4.5). System
1: After gel electrophoresis, oligonucleotides were purified using
a C18 column (150 × 3.9 mm, 5 µm, Millipore Corp.). The initial
concentration of 5% solvent A, 95% solvent B was maintained
for 10 min, and then solvent A was increased linearly to 29%
over 20 min at a flow rate of 1 mL/min. System 2: Analysis of
the enzyme digest of the oligonucleotides was performed with
a supelcosil LC-18-DB column (2.5 × 0.46 cm, 5 µm, Supelco,
Inc.) and 0% solvent A, 100% solvent C. Solvent A was linearly
increased to 12% over 30 min, then to 40% over 15 min where
it was held for 10 min at a flow rate of 1 mL/min. System 3:
The progress of the reaction of p-BQ with dG was monitored
using a Supelco C18 reverse-phase semipreparative (250 × 10
mm, 5 µm) and 0% solvent A, 100% H2O. Solvent A was
increased linearly to 30% over 30 min at a flow rate of 2 mL/
min, and then to 70% over 15 min at room temperature.
λ
max 240, 286, 284, 333 nm. Electrospray MS (positive ion): m/z
486 (74) (M + Na)+; 464 (20) MH+; 370 (48) (BH + Na)+; 348
(100) BH2+. Electrospray MS (negative Ion): m/z 462 (100) (M
- H)+.
5′-O-(4,4′-Dim eth oxytr ityl- 3′′-h yd r oxy-1,N2-ben zeth en o-
2′-d eoxygu a n osin e (2). Compound (1) (50 mg, 0.14 mmol) was
coevaporated from dry pyridine three times and then dissolved
in 5 mL of dry pyridine and treated with dimethoxytrityl
chloride (DMTCl) (142 mg, 0.42 mmol, 3 equiv) at room
temperature with stirring. After 6 h, TLC on silica gel (10%
MeOH + CH2Cl2) showed that all the starting material was
consumed and a new product was formed which was less polar
than the starting material. The reaction was stopped, and the
pyridine was reduced to about 1 mL, then 2 mL of methylene
chloride was added, and the resulting mixture was applied to a
silica gel column (2 × 18 cm) equilibrated with CH2Cl2/MeOH/
Et3N (97:2:1), as eluant. About 200 mL of this solvent was run
on the column to remove any material of low polarity, as well
as any traces of pyridine. The polarity of the solvent was then
increased by using CH2Cl2/MeOH/Et3N (89:10:1). After analyz-
ing all fractions by TLC, the fractions containing the product
were collected and evaporated to give an off-white foamy product
Ultraviolet spectra were recorded on
a Hitachi U-2000
spectrophotometer using 0.5 cm cuvettes. TLC was performed
on EM5735/7 silica gel 60, F254 plates. Column chromatography
was performed using silica gel 60 with elution under pressure.
Ch em ica l Syn th eses. P r ep a r a tion of 3′′-Hyd r oxy-1,N2-
ben zeth en o-2′-d eoxygu a n osin e (p-BQ-d G) (1). dG (500 mg,
1.87 mmol) was dissolved in 20 mL of dry DMSO or DMF and
very fine potassium carbonate (316 mg, 2.28 mmol) was added
at room temperature. After 15 min, p-BQ (414 mg, 3.82 mmol)
was added in a single dose at room temperature and stirred for
7 h. TLC showed the formation of two products and some
unreacted starting material. A further amount of a very fine
K2CO3 (316 mg, 2.28 mmol) and p-BQ (207 mg, 1.41 mmol) was
added and stirred for another 15 h. HPLC analysis showed only
traces of starting material, one major product, and traces of an
unstable intermediate. TLC on silica gel (MeOH/CH2Cl2; 15:
85) showed the formation of a less polar product (Rf 0.30) and
some very polar material in the bottom of the TLC which was
attributed to decomposition. The reaction mixture was absorbed
on silica gel (5-25 µm), then applied to a silica column (18 × 4
cm of silica), and eluted with 15% methanol in methylene
chloride. After analysis of all fractions by TLC, the fractions
containing the desired product were evaporated to dryness to
yield a light brown solid material (156 mg, 28%): mp: 275 °C
(72 mg, 78%): Rf (10% MeOH in CH2Cl2) 0.37; mp 135-137 °C.
FAB/MS (positive ion): m/z 660 (42) MH+, 242 (16) BH2
+
.
1H
NMR (DMSO-d6, 400 MHz): δ 9.569 (S, 1H, exch, HO-7), 8.012
(S, 1H, H-2), 7.301 (d, 1H, J ) 7.3 Hz, H-6), 7.178 (m, 9H, 8
aromatic H + H-9), 6.891 (d, 1H, J ) 11 Hz, H-8), 6.7609 (m,
5H, aromatic H), 6.295 (t, 1H, J ) 7.2 Hz, H-1′), 5.437 (d, 1H,
J ) 4.7 Hz, exch, HO-3′), 4.406 (m, 1H, H-4′), 3.946 (m, 1H,
H-3′), 3.701 (m, 2H, H-5′), 3.649 (2s, 6H, OCH3), 2.608 (m, 1H,
H-2′), 2.270 (m, 1H, H-2′′).
5′-O-(4,4′-Dim eth oxytr ityl)-3′-O-[(d iisop r op yla m in o)(2-
cya n oeth oxy)p h osp h in o]- 3′′-h yd r oxy-1,N2-ben zeth en o-2′-

Synthesis of 1,N2-p-BQ-dG Adduct
Chem. Res. Toxicol., Vol. 10, No. 2, 1997 167
d eoxygu a n osin e (3). Compound 2 (125 mg, 0.19 mmol) was
dried over P2O5 and NaOH pellets in a vacuum desiccator for
24 h and then was coevaporated with a mixture of dry CH2Cl2
and dry benzene, prior to reaction. To a mixture of compound
2 and dry Et3N (1 mL to aid the solubility) in 12 mL of dry CH2-
Cl2 under N2 was added 2′-cyanoethyl N,N-diisopropylchloro-
phosphoramidite (53 mg, 36 mL, 0.226 mmol, 1.2 equiv)
dropwise over 1 min at -72 °C and with vigorous stirring. The
reaction was stirred at -72 °C for a further 20 min. The
temperature was then slowly increased to room temperature
and stirred for another 45 min and then analyzed by TLC (CH2-
Cl2/MeOH, 90:10) prerun in CH2Cl2CH3CO2C2H5Et3N (60:40:
5), which showed that all the starting material was consumed
and a new less polar product was formed. The reaction was
stopped by adding 1 mL of dry methanol dropwise at -20 °C,
and after about 3 min, all the solvent was evaporated under
vacuum, without using any heat, to give a yellowish solid
material. This material was dissolved in dry CH2Cl2 and
washed twice with a 5% solution of NaHCO3 and then once with
a saturated solution of NaCl, dried over Na2SO4, filtered and
evaporated under vacuum, and coevaporated twice in a mixture
of dry CH2Cl2 and dry benzene to give a light brown gum product
(113 mg, 69%). TLC analysis of this product on silica gel (CH2-
Cl2/MeOH; 90:10) gave Rf ) 0.27 and 31P NMR (CDCl3, 400
MHz) δ 145.857 with reference to phosphoric acid as the
external standard. The 31P signals for the two diastereoisomers
which comprise phosphoramidite 3 were unresolved.
fraction from the cartridges containing the desired oligomer
together with the failure sequences were concentrated and
further purified by 20% polyacrylamide gel electrophoresis and
HPLC. The following DNA oligomers were made.
1: 5′-CCGCTAXCGGGTACCGAGCTCGAAT-3′
2: 5′-XCGGGTACCGAGCTCGAAT-3′
X ) p-BQ-dG
The defined oligonucleotides (0.5 A260 units) were dissolved
in 44 mL of freshly deionized water, 0.8 mL of 1 M MgCl2, and
3.5 mL of 0.5 M Tris-HCl buffer (pH 7.5) and digested with
snake venom phosphodiesterase (2.5 mL) and bacterial alkaline
phosphatase (4.0 mL) at 37 °C for 16 h. The digested mixture
was analyzed on a C-18 reverse-phase HPLC Supelcosil column
(250 × 4.5 mm) using system 2. Quantitation of the deoxy-
nucleosides was on the basis of integration of the peak areas at
260 nm.
Resu lts a n d Discu ssion
Syn th esis of 3′′-Hydr oxy-1,N2-ben zeth en o-2′-deoxy-
gu a n osin e (p-BQ-d G, 1). In 1986, J owa et al. (17)
reported the formation and characterization of the exo-
cyclic adduct, p-BQ-dG, formed from reaction of dG with
p-BQ in aqueous solution at pH 7.2. However, the yield
was low (∼1%) (18). The same adduct was also detected
in DNA reacted with p-BQ, but to such a low extent that
it was necessary to use a 14C-labeled reagent for detec-
tion.
Our objective in this work was to synthesize oligo-
nucleotides containing a site-specific p-BQ-dG adduct (1)
using phosphoramidite chemistry and to use DNA oligo-
mers for biochemical studies. This could be achieved only
by optimizing the conditions of the reaction of p-BQ with
dG to obtain a large amount of the product, p-BQ-dG,
and convert it to the fully protected phosphoramidite (4).
The reaction of dG with p-BQ was performed in
aqueous solutions and at different pHs in order to find
the best conditions to maximize the yield of the major
product. The reaction in aqueous solution at both pH
7.4 and 9.3 produced the same product as reported by
J owa (17), as well as traces of other products which were
not studied. However, no reaction was detected at pH
4.5. The major product had the same UV spectrum as
reported by J owa, but the yield of the desired compound
at both pH 7.4 and 9.3 was poor (<2%) and not suitable
for our purposes. Therefore, the reaction was repeated
in nonaqueous conditions, either in dry DMSO or DMF
in the presence of K2CO3 or NaOH pellets. Under these
conditions, substantial amounts of product (about 50%)
were obtained, as shown by HPLC.
The reaction of dG with p-BQ in DMSO/K2CO3 first
produced an unstable product which was slowly con-
verted to the final stable product (p-BQ-dG). This
reaction, when monitored by HPLC or TLC over a period
of time, clearly showed the conversion of the intermediate
which had a longer retention time (32-33 min) to the
final product (Figure 1), which was eluted at 31 min. The
UV spectrum of the unstable intermediate does not
resemble that of the final product; the final product has
a significant trough at about 250 nm and a peak at about
270 nm, whereas the intermediates have a broad shoul-
der from 260 to 275 nm, in addition to red-shifted
maxima (Figure 2).
Acyla tion of 5′-O-(4,4′-Dim eth oxytr ityl)-3′-O-[(d iisop r o-
p yla m in o)(2-cya n oeth oxy)p h osp h in o]- 3′′-h yd r oxy-1,N2-
ben zeth en o-2′-d eoxygu a n osin e (3). A 75 mg sample of
compound 3 was coevaporated three times from dry pyridine
under a very strong vacuum and then was dissolved in 5 mL of
dry pyridine, and 100 mL of isobutyric anhydride was added
dropwise at ambient temperature with stirring. After the
mixture was stirred for 24 h, TLC on silica gel (CH2Cl2/MeOH;
90:10) showed that the starting material was fully consumed
and gave a product of very low polarity at Rf 0.83. The reaction
was stopped by evaporating all pyridine and coevaporated from
dry benzene twice. The product was dissolved in 1 mL toluene
and added dropwise to dry hexane at -42 °C (CH3CN/CO3) to
precipitate. The product was centrifuged for 2 min, washed with
fresh hexane at -42 °C, and again centrifuged for 2 min. The
hexane was removed and the product was dissolved in dry
pyridine and coevaporated three times from dry C6H6 to give a
yellow gum product (71 mg, 81%). This product was dried over
P2O5 and NaOH pellets in a vacuum desiccator overnight at
room temperature and used to synthesize the DNA oligomers.
31PNMR: (CDCl3), δ 148.003, this signal for the two diastere-
oisomers which represent phosphoramidite 4 was unresolved.
Solid -P h a se Syn th esis of th e Oligon u cleotid es. Synthe-
sis of the oligonucleotides was done on an Applied Biosystem
394 automated DNA synthesizer on a 1 µmol scale using
phosphoramidite chemistry on a controlled pore glass (CPG)
support with an aminopropyl succinate linker. Phenoxyacetyl
phosphoramidites (PAC) were used to synthesize the DNA
oligomers. The synthesis followed standard protocol of the DNA
synthesizer for phosphoramidite chemistry, except for the step
in which the modified deoxynucleoside was inserted. In this
step, the time of coupling was increased to 30 min, in order to
maximize the coupling efficiency (96%). The oligonucleotides
were recovered as the 5′-dimethoxytritylated derivatives and
deprotected under nonaqueous conditions with 1,8-diazobicyclclo-
[5.4.0]undec-7-ene (DBU) in ethanol (200 proof). The resin was
treated with 2 mL of DBU/ethanol (10:90) under anhydrous
conditions with occasional shaking. After 24 h at room tem-
perature, the resin was removed by centrifugation, and the
pellet washed with ethanol twice (0.5 mL). The combined
supernatant and washings were evaporated under reduced
pressure to remove all ethanol. Water was added (3 mL), and
immediately the pH was adjusted to 7.0 by adding 180 µL of a
solution of 50% acetic acid and stirring. This was desalted and
detritylated by 2% TFA in water using two Sep-Pack Cartridges
(Waters Associates) as described by the supplier. The eluted
This intermediate showed two peaks very close to each
other by HPLC (Figure 1). When either peak was
collected and immediately reanalyzed under the same

168 Chem. Res. Toxicol., Vol. 10, No. 2, 1997
Chenna and Singer
The MS spectrum suggests that the molecule carries two
molecules of p-BQ on the dG base, with removal of a
molecule of water (dG + 2p-BQ - H2O) ) 463.
A
comparison of the UV spectra of the intermediate and
the stable adduct gives an indication of substitution on
the N-1 and N2 of the dG by p-BQ, but we do not know
where the second p-BQ reacts with dG. However we
speculate that the second p-BQ molecule is either on the
N-7 or C-8 or alternately cyclizes to form a N-7-C-8 ring
as suggested by Shapiro et al. for acrolein reacted with
dG (16). Due to the unstable nature of this adduct, it
was not possible to use 1H NMR to obtain further
information about the structure or mechanism of the
interconversion.
In order to obtain sufficient product for making the
fully protected phosphoramidite of p-BQ-dG, the reaction
was scaled up to react 500 mg of dG with p-BQ, which
after chromatography on silica gel, yielded 28% of a light
brown solid pure product 1 (Scheme 1), which was
analyzed by TLC, HPLC, 1H NMR, UV, and FAB/MS (see
Materials and Methods). The 1H NMR of the product
showed, in particular, four exchangeable protons, two on
the 3′- and 5′-hydroxyl groups of the sugar moiety, the
third at 9.465 ppm as a singlet for the 7-hydroxyl group
on the phenolic ring on the base, and the fourth one at
the NH-5 of the base (at 6.531 ppm). The FAB/MS of
product 1 showed an intense natriated molecule at m/z
380 (M + Na)+ and protonated molecular ion MH+ at m/z
358, an (BH + Na)+ peak at m/z 264 and a (BH2)+ peak
at m/z 242. These data are consistent with the addition
of one molecule of p-BQ to the base moiety followed by
dehydration to form the stable adduct (1). The UV
spectra of this product 1 at different pHs was similar to
the one reported (17).
Syn th esis of th e F u lly P r otected P h osp h or a m id -
ite of p-BQ-d G(4). Compound 1 has three hydroxyl
groups and an amino group. In order to incorporate this
adduct in oligonucleotides, all four reactive groups must
be protected. There is a hydroxyl group at position 7 and
an amino group at position 5 of the exocyclic base moiety,
as well as the usual two hydroxy groups (3′- and 5′-OH)
on the sugar moiety. The hydroxyl group (7-OH) and the
amino group (NH-5) on the base of p-BQ-dG would
definitely interfere in the synthesis of the oligonucleotides
by branching. Therefore, they must be protected with
protecting group(s), which later can be easily removed
in the last step of the deprotection of oligonucleotides.
p-BQ-dG (1) was converted to its 5′-O-DMT product in
78% yield, which was achieved according to standard
conditions (24,25). Phosphitylation of 5′-O-DMT-p-BQ-
dG (2) with 2-cyanoethyl N,N-diisopropylchlorophos-
phoramidite (26) yielded the phosphoramidite (3) (69%
yield). By controlling the amount of the phosphitylating
reagent and the temperature (-70 °C) only the 3′-OH was
successfully reacted, with only traces of a product which
was phosphitylated on both hydroxyl group (3′-OH and
7-OH). These two derivatives could be separated on the
basis of polarity. Compound 3 contains a free phenolic
hydroxyl group and an amino group which may lead to
branching in the DNA synthesizer if not protected.
Therefore, derivative 3 was acylated by using isobutyric
anhydride in pyridine at room temperature (27). The
reaction, monitored by TLC on silica gel indicated the
formation of a new product which was less polar than
the starting material 3. This product 4 was purified by
precipitation in dry hexane at -42 °C, which gave a
yellow gum product (81%) which was coevaporated twice
F igu r e 1. HPLC analysis of products of reactions of dG with
p-BQ at 30 min and 22 h. The reaction was performed in DMF/
K2CO3 at room temperature. (Top) 30 min reaction: peak I (∼19
min) corresponds to the starting material, dG. peak II (∼26 min)
is p-BQ; peak III (∼31 min) is the stable final adduct; peaks IV
and V between 32 and 33 min are the two unstable intermedi-
ates. (Bottom) 22 h reaction: peak I is dG; only traces of p-BQ
remain (peak II); the peak at ∼31 min (peak III) is the stable
p-BQ-dG product; the doublet at 32-33 min consists of two
interconvertable forms of the unstable intermediates (peaks IV,
V). When either of the doublets is collected and immediately
reanalyzed in the same HPLC system, the same two peaks (32-
33 min) result (data not shown).
F igu r e 2. UV spectra of p-BQ-dG (peak III) and the intermedi-
ate product (peaks IV and V). The arrows indicate the spectrum
of each. Note that the three maxima for p-BQ-dG are shifted
∼5 nm.
HPLC conditions, there were again two peaks from each
of the single peaks. This suggests that there is a single
product in equilibrium in two forms, which are presumed
to be isomers. The intermediate product was isolated by
HPLC and analyzed immediately by electrospray MS
both in positive and negative ion. In the positive mode,
the spectrum showed a protonated molecular ion at 464
for MH+ and a protonated base BH2 at 348 (Figure 3).
+
The negative ion electrospray showed (M - H)- at 462.

Synthesis of 1,N2-p-BQ-dG Adduct
Chem. Res. Toxicol., Vol. 10, No. 2, 1997 169
F igu r e 3. Electrospray mass spectrum of the unstable intermediate (combined peaks IV and V) in Figure 1.
Sch em e 1. P r ep a r a tion of th e P h osp h or a m id ite of p-BQ-d G Ad d u ct 1
in dry benzene and kept under vacuum before being used
to make the desired oligonucleotides. This fully protected
compound 4 was used to synthesize two different oligo-
nucleotides: a 19-mer and a 25-mer. The phenoxyacetyl
phosphoramidites (PAC) were used to synthesize these
DNA oligomers and followed the standard phosphora-
midite chemistry. The coupling efficiency of p-BQ-dG
phosphoramidite was 96%. The oligonucleotides were
then deprotected under nonaqueous conditions by using
10% DBU in ethanol at room temperature for 24 h (28)
and then purified by 20% polyacrylamide gel electro-
phoresis followed by HPLC. DBU was used to deprotect
these oligomers instead of ammonia, since p-BQ-dG was
not stable in 28% ammonia. In addition, the use of the

170 Chem. Res. Toxicol., Vol. 10, No. 2, 1997
Chenna and Singer
protection and deprotection after incorporation into
defined oligonucleotides. This new synthesis of a p-BQ-
dG containing oligonucleotides will enable us to extend
our biochemical studies of repair of p-BQ-dC and p-BQ-
dA in defined sequences (23,29) to the only other stable
p-BQ adducts formed under physiological conditions.
Such oligonucleotides will also be used for structural
studies.
Ack n ow led gm en t. This work was supported by
Grant No. ES07368 from the National Institutes of
Health to B.S. and was administered by the Lawrence
Berkeley Laboratory under Department of Energy Con-
tract No. DE-AC03-76SF00098. We are grateful to M.
Haggard for the automated synthesis of oligonucleotides
containing site specifically modified nucleosides.
Refer en ces
(1) IARC Monographs on the Evaluation of the Carcinogenic Risks
to Humans. (1987). Suppl. 7, International Agency for Research
on Cancer, Lyon, France.
(2) Goldstein, B. D. (1977) Benzene toxicity: a critical evaluation:
hematotoxicity in humans. J . Toxicol. Environ. Health Suppl. 2,
69-105.
(3) Guengerich, F. P., and MacDonald, T. L. (1990) Mechanisms of
cytochrome P-450 catalysis. FASEB J . 4, 2453-2459.
(4) Guengerich, F. P., Kim, D. H., and Iwasaki, M. (1991) Role of
human cytochrome P-450 IIE1 in the oxidation of many low
molecular weight cancer suspects. Chem. Res. Toxicol. 4, 168-
179.
(5) Huff, J . E., Haseman, J . K., DeMarini, D. M., Eustis, S., Maronpot,
R. R., Peters, A. C., Persing, R. L., Chrisp, C. E., and J acobs, A.
C. (1989) Multiple-site carcinogenicity of benzene in Fischer 344
rats and B6C3F1 mice. Environ. Health Perspect. 82, 125-163.
(6) Snyder, R., J owa, L., Witz, G., Kalf, G., and Rushmore, T. (1987)
Formation of reactive metabolites from benzene. Arch. Toxicol.
60, 61-64.
F igu r e 4. HPLC profile of 2′-deoxynucleosides obtained as a
result of enzymatic digestion of oligonucleotide 1, which contains
p-BQ-dG (see Materials and Methods). HPLC was performed
using system 3. The retention times of the deoxynucleosides are
dC, 11.4 min; dG, 15.3 min; dT, 17.0 min; dA, 20.2 min; and
p-BQ-dG, 36.5 min. The UV spectrum of p-BQ-dG is shown in
the inset.
PAC protecting group allowed the deprotection of DNA
oligomers at room temperature and in a shorter time (24
h) than the standard phosphoramidite, which requires
several days.
(7) Zheng, J ., and Hanzlik, R. P. (1992) Dihydroxylated mercapturic
acid metabolites of bromobenzene. Chem. Res. Toxicol. 5, 561-
567.
The composition of the oligomers was confirmed after
enzymatic digestion of the DNA oligomers to nucleosides,
followed by HPLC analysis. The enzyme hydrolysis
showed that the modified base survived the conditions
used in the DNA synthesizer and the deprotection
procedure (Figure 4).
(8) Pascoe, G. A., Calleman, C. J ., and Baille, T. A. (1988) Identifica-
tion of S-(2,5-dihydroxyphenyl)-cysteine and S-(2,5-dihydroxy-
phenyl)-N-acetyl-cysteine as urinary metabolites of acetami-
nophen in the mouse. Evidence for p-benzoquinone as a reactive
intermediate in acetaminophen metabolism. Chem.-Biol. Interact.
68, 85-98.
(9) McDonald, T. A., Waidyanatha, S., and Rappaport, S. M. (1993)
Measurement of adducts of benzoquinone with hemoglobin and
albumin. Carcinogenesis 14, 1927-1932.
(10) Andrews, L. S., Lee, E. W., Witmer, C. M., Kocsis, J . J ., and
Snyder, R. (1977) Effects of toluene on the metabolism, disposition
and hemopoietic toxicity of [3H]benzene. Biochem. Pharmacol. 26,
293-300.
(11) Sammett, D., Lee, E. W., Kocsis, J . J ., and Snyder, R. (1979)
Partial hepatectomy reduces both metabolism and toxicity of
benzene. J . Toxicol. Environ. Health 5, 785-792.
(12) Kalf, G. F. (1987) Recent advances in the metabolism and toxicity
of benzene. Crit. Rev. Toxicol. 18, 141-159.
(13) Rickert, D. E., Baker, T. S., Bus, J . S., Barrow, C. S., and Irons,
R. D. (1979) Benzene disposition in the rat after exposure by
inhalation. Toxicol. Appl. Pharmacol. 49, 417-423.
(14) Greenlee, W. F., Gross, E. A., and Irons, R. D. (1981) Relationship
between benzene toxicity and the disposition of 14C-labelled
benzene metabolites in the rat. Chem. Biol. Interact. 74, 55-62.
(15) Eastmond, D. A., Smith, M. T., Ruzo, L. O., and Ross, D. (1986)
Metabolic activation of phenol by human myeloperoxidase and
horseradish peroxidase. Mol. Pharmacol. 30, 674-679.
(16) Shapiro, R., Sodum, R. S., Everett, D. W., and Kundu, S. K. (1986)
Reactions of nucleosides with glyoxal and acrolein. The Role of
Cyclic Nucleic Acid Adducts in Carcinogenesis and Mutagenesis
(B. Singer and H. Bartsch, Eds.) pp 165-173. International
Agency for Research on Cancer, Lyon, France.
Su m m a r y
The reaction of dG with p-BQ proceeded with high yield
in DMF or DMSO with K2CO3 or NaOH pellets as the
base catalyst. In contrast, in neutral or basic aqueous
solution, the yield of p-BQ-dG was <2% and no product
was detected at pH 4.5. Although only a single final
stable product was obtained, a small amount of an
unstable intermediate could be isolated which separated
on HPLC into two interconvertable products which are
apparent isomers.
This intermediate is formed under physiological condi-
tions, as well as nonaqueous and both the final product
and intermediate have significantly different UV spectra.
However, electrospray MS indicates that the intermedi-
ate contains two p-BQ molecules. From the comparison
of the two UV spectra we can conclude that one p-BQ
forms an exocyclic ring between the 1- and N2-positions
of dG while the other p-BQ could be on either the N7
and/or C-8 of dG. This speculation is supported by
Shapiro’s finding that, in general, R,â-unsaturated car-
bonyl compounds can form bicyclic 1,N2:7,8-diadducts
(16).
(17) J owa, L., Winkle, S., Kalf, G., Witz, G., and Snyder, R. (1986)
Deoxyguanosine adducts formed from benzoquinone and hydro-
quinone. Adv. Exp. Med. Biol. 197, 825-832.
(18) J owa, L., Witz, G., Snyder, R., Winkle, S., and Kalf, G. F. (1990)
Synthesis and characterization of deoxyguanosine-benzoquinone
adducts. J . Appl. Toxicol. 10, 47-54.
The conversion of p-BQ-dG into the phosphoramidite
was successfully accomplished in good yield with both
(19) Pongracz, K., Kaur, S., Burlingame, A. L., and Bodell, W. J . (1990)
Detection of (3′-hydroxy)-3,N4-benzetheno-2′-deoxycytidine-3′-

Synthesis of 1,N2-p-BQ-dG Adduct
Chem. Res. Toxicol., Vol. 10, No. 2, 1997 171
phosphate by 32P-postlabeling of DNA reacted with p-benzo-
quinone. Carcinogenesis 11, 1469-1472.
(20) Pongracz, K., and Bodell, W. J . (1991) Detection of 3′-hydroxy-
1,N6-benzetheno-2′-deoxyadenosine 3′-phosphate by 32P-postla-
beling of DNA reacted with p-benzoquinone. Chem. Res. Toxicol.
4, 199-202.
ribonucleosides and new syntheses of deoxyribonucleoside-3′
phosphates. J . Am. Chem. Soc. 85, 3821-3827.
(25) Chaudhary, S. K., and Hernandez, O. (1979) 4-Dimethylamino-
pyridine: an efficient and selective catalyst for the silylation of
alcohols. Tetrahedrons Lett. 2, 99-102.
(26) Bodepudi, V., Shibutani, S., and J ohnson, F. (1992) Synthesis of
2′-deoxy-7,8-dihydro-8-oxoguanosine and 2′-deoxy-7,8-dihydro-8-
oxoadenosine and their incorporation into oligomeric DNA. Chem.
Res. Toxicol. 5, 608-617.
(27) Schmidt, O. T., Voigt, H., Puff, W., and Ko¨ster, R. (1956)
Benzyla¨ther der Elligasa¨ure und Hexaoxy-diphenylsa¨ure. J ustus
Liebigs Ann. Chem. 586, 165-179.
(28) Pharmacia-LKB. Biotechnology. (1989) Analects 17, 6-7.
(29) Chenna, A., Hang, B., Rydberg, B., Kim, E., Pongracz, K., Bodell,
and W. J ., Singer, B. (1995) The benzene metabolite p-benzo-
quinone forms adducts with DNA bases that are excised by a
repair activity from human cells that differs from an ethenoad-
enine glycosylase. Proc. Natl. Acad. Sci. U.S.A. 92, 5890-5894.
(21) Pongracz, K., and Bodell, W. J . (1996) Synthesis of N2-(4-
hydroxyphenyl)-2′-deoxyguanosine 3′-phosphate: Comparison by
32P-postlabeling with the DNA adduct formed in HL-60 cells
treated with hydroquinone. Chem. Res. Toxicol. 9. 593-598.
(22) Bodell, W. J ., Levay, G., and Pongracz, K. (1993) Investigation of
benzene-DNA adducts and their detection in human bone marrow.
Environ. Health Perspect. 99, 241-4.
(23) Chenna, A., and Singer, B. (1995) Large scale synthesis of
p-benzoquinone-2′-deoxycytidine and p-benzoquinone-2′-deoxy-
adenosine adducts and their site-specific incorporation into DNA
oligonucleotides. Chem. Res. Toxicol. 8, 865-874.
(24) Schaller, H., Weimann, G., Lerch, B., and Khorana, H. G. (1963)
Studies on polynucleotides. XXIV. The stepwise synthesis of
specific deoxyribopolynucleotides. Protected derivatives of deoxy-
TX960168R