Addition of deoxyribose to guanine and modified DNA bases by Lactobacillus helveticus trans-N-deoxyribosylase

Article abstract of DOI:10.1021/tx9600661

The use of bacterial trans-N-deoxyribosylase was evaluated as an alternative method for deoxyribosylation in the synthesis of deoxyribonucleosides containing potentially mutagenic adducts. A crude enzyme preparation was isolated from Lactobacillus helveticus and compared to Escherichia coli purine nucleoside phosphorylase. trans-N-deoxyribosylase was more regioselective than purine nucleoside phosphorylase in the deoxyribosylation of Gua at the N9 atom, as compared to N7, as demonstrated by NMR analysis of the product. 5,6,7,9-Tetrahydro-7-acetoxy-9- oxoimidazo[1,2-a]purine was efficiently deoxyribosylated by trans-N- deoxyribosylase but not at all by purine nucleoside phosphorylase. Other substrates for trans-N-deoxyribosylase were N²-(2-oxoethyl)Gua, pyrimido[1,2-a]purin-10(3H)-one, 1,N²-ε-Gua, N²,3ε-Gua, 3,N⁴-(-Cyt, 1,N⁶-ε-Ade, C⁸-methylGua, and C⁸-aminoGua, most of which gave the desired isomer (bond at the nitrogen corresponding to N9 in Gua) in good yield. Neither N⁷-alkylpurines nor C⁸-(arylamino)-substituted guanines were substrates. The approach offers a relatively convenient method of enzymatic preparation of many carcinogen-DNA adducts at the nucleoside level, for either use as standards or incorporation into oligonucleotides. trans-N- deoxyribosylase can also be used to remove deoxyribose from modified deoxyribonucleosides in the presence of excess Cyt.

Full text of DOI:10.1021/tx9600661

1140
Chem. Res. Toxicol. 1996, 9, 1140-1144
Ad d ition of Deoxyr ibose to Gu a n in e a n d Mod ified DNA
Ba ses by La ctoba cillu s h elveticu s
tr a n s-N-Deoxyr ibosyla se
Michael Mu¨ller,† Linda K. Hutchinson, and F. Peter Guengerich*
Department of Biochemistry and Center in Molecular Toxicology, Vanderbilt University School of
Medicine, Nashville, Tennessee 37232
Received April 15, 1996; Revised Manuscript Received J une 12, 1996X
The use of bacterial trans-N-deoxyribosylase was evaluated as an alternative method for
deoxyribosylation in the synthesis of deoxyribonucleosides containing potentially mutagenic
adducts. A crude enzyme preparation was isolated from Lactobacillus helveticus and compared
to Escherichia coli purine nucleoside phosphorylase. trans-N-deoxyribosylase was more
regioselective than purine nucleoside phosphorylase in the deoxyribosylation of Gua at the
N9 atom, as compared to N7, as demonstrated by NMR analysis of the product. 5,6,7,9-
Tetrahydro-7-acetoxy-9-oxoimidazo[1,2-a]purine was efficiently deoxyribosylated by trans-N-
deoxyribosylase but not at all by purine nucleoside phosphorylase. Other substrates for trans-
N-deoxyribosylase were N2-(2-oxoethyl)Gua, pyrimido[1,2-a]purin-10(3H)-one, 1,N2-ꢀ-Gua, N2,3-
ꢀ-Gua, 3,N4-ꢀ-Cyt, 1,N6-ꢀ-Ade, C8-methylGua, and C8-aminoGua, most of which gave the desired
isomer (bond at the nitrogen corresponding to N9 in Gua) in good yield. Neither N7-alkylpurines
nor C8-(arylamino)-substituted guanines were substrates. The approach offers a relatively
convenient method of enzymatic preparation of many carcinogen-DNA adducts at the
nucleoside level, for either use as standards or incorporation into oligonucleotides. trans-N-
deoxyribosylase can also be used to remove deoxyribose from modified deoxyribonucleosides
in the presence of excess Cyt.
Sch em e 1. Rea ction s Ca ta lyzed by P u r in e
Nu cleosid e P h osp h or yla se a n d
In tr od u ction
Many carcinogens either are electrophiles or are bio-
logically transformed to electrophiles that react with
DNA bases to form covalent adducts. This process is
generally considered to be important in chemical car-
cinogenesis (1-3). Understanding the chemistry and
biology related to the roles of these modified DNA bases
in mutagenesis and cancer has been challenging. DNA
bases, nucleosides, and nucleotides appropriately modi-
fied with carcinogens are needed for analysis of biological
samples and for biochemical and physical studies involv-
ing oligonucleotides.
tr a n s-N-Deoxyr ibosyla se (B ) Ba se)
Strategies for synthesis of such modified deoxyribo-
nucleosides are often most logical when the desired
chemical modifications are done on the base and then
the deoxyribose group is attached. This approach usually
allows for better control of the modification chemistry
because the sugar hydroxyls may react with an electro-
phile, the nucleoside may not be soluble in certain
solvents, or the glycosidic bond may be unstable to
modification conditions. Chemical approaches exist for
attachment of a deoxyribose moiety to a modified base
(4). However, there are inherent problems in controlling
both the regiochemistry (e.g., N7 vs N9 atom of a purine)
and stereoselectivity (e.g., an SN2 attack is needed on a
sugar isomer to generate the desired linkage).
purine nucleoside phosphorylase (EC 2.4.2.1) (Scheme 1).
This enzyme has been used in a number of applications
(5-10), including some in our own group (11, 12).
However, we were totally unsuccessful in achieving
deoxyribosylation of 5,6,7,9-tetrahydro-7-acetoxy-9-ox-
oimidazo[1,2-a]purine, a key intermediate in our strategy
to introduce the vinyl chloride-derived DNA adduct N2-
(2-oxoethyl)Gua1 into oligonucleotides (12). As an alter-
native, we considered the enzyme trans-N-deoxyribosy-
lase (EC 2.4.2.6, nucleoside:purine (pyrimidine) deoxy-
ribosyltransferase) (Scheme 1).
The activity of trans-N-deoxyribosylase in Lactobacillus
was first reported in 1952 (13). The enzyme has been
purified (14, 15) and preparations have been used in
syntheses of 2-aminopurine (16, 17). In contrast to
One approach to attaching deoxyribose to modified
purines has been the use of the enzyme Escherichia coli
* Address correspondence to this author at the Department of
Biochemistry and Center in Molecular Toxicology, Vanderbilt Univer-
sity School of Medicine, Nashville, TN 37232-0146.
1
Abbreviations: NOE, nuclear Overhauser effect; M1G, pyrimido-
† Present address: Abteilung Arbeits-und Sozialmedizin der Georg-
August-Universita¨t Go¨ttingen, Waldweg 37, Go¨ttingen, Federal Re-
public of Germany.
[1,2-a]purin-10(3H)-one. We refer to N2-(2-oxoethyl)Gua and its nucleo-
side using that convention for simplicity, although the predominant
form appears to be the cyclized hemiaminal, 5,6,7,9-tetrahydro-7-
hydroxy-9-oxoimidazo[1,2-a]purine (12).
X Abstract published in Advance ACS Abstracts, September 1, 1996.
S0893-228x(96)00066-5 CCC: $12.00 © 1996 American Chemical Society

DNA Adduct Synthesis with trans-N-Deoxyribosylase
Chem. Res. Toxicol., Vol. 9, No. 7, 1996 1141
Bruker AM-400 spectrometer (Bruker, Billerica, MA) in the
Vanderbilt facility. UV spectra (other than HPLC diode array)
were recorded with a modified OLIS/Cary 14 instrument (On-
Line Instrument Systems, Bogart, GA). Mass spectra were
collected using a Kratos Concept II HH instrument (Kratos,
Manchester, U.K.) with fast atom bombardment ionization and
a mixture of glycerol, Me2SO, and 3-nitrobenzyl alcohol as the
matrix or with an electrospray instrument (Finnigan TSQ 7000,
Finnigan, Sunnyvale, CA).
purine nucleoside phosphorylase, this enzyme also trans-
fers pyrimidines, in a reversible reaction (13, 15, 18).
Other deoxyribonucleosides are cosubstrates, so an aux-
iliary enzyme system is not required for synthetic ap-
plications (Scheme 1). Xanthine, hypoxanthine, N1-
methylhypoxanthine, uric acid, kinetin, 5-methylCyt,
8-azaAde, 6-mercaptopurine, N6-methylaminopurine,
2-ethyl-6-hydroxypurine, 5-fluoroUrd, 2-aminopurine,
and (possibly) 2,6-diaminopurine are reported substrates
for the enzyme (14-18). We have examined the useful-
ness of this system in the formation of several DNA-
carcinogen adducts, as an alternative to purine nucleo-
side phosphorylase. The regioselectivity of coupling has
also been analyzed in many of these reactions.
Assa ys. Incubations with purine nucleoside phosphorylase
(95 µg of protein mL-1) were done as described elsewhere, in
the presence of thymidine phosphorylase (10).
The general procedure used for deoxyribosylation of bases
involved a mixture of 5 or 25 µg of partially purified trans-N-
deoxyribosylase protein (vide supra) and 5.3 mM dCyd (as the
deoxyribose donor) in a total volume of 1.0 mL of 0.1-0.5 M
potassium 2-(N-morpholino)ethanesulfonate buffer (pH 6.0) and
containing 0.3-0.5 mM Gua or a modified base, dissolved either
in 50-100 µL of 0.1-1 N HCl or Me2SO. Incubations proceeded
for 3 or 15 h (under Ar), and products were analyzed by HPLC.
In some cases the pH was adjusted or the scale of the reaction
was increased (for preparative work).
Exp er im en ta l P r oced u r es
En zym es. E. coli purine nucleoside phosphorylase and
thymidine phosphorylase (EC 2.4.2.4) were purchased from
Sigma Chemical Co. (St. Louis, MO).
trans-N-Deoxyribosylase was partially purified from Lacto-
bacillus helveticus (from American Type Culture Collection,
Rockville, MD) using the procedure described by Uerkvitz (15).
Cells were grown in stationary culture on Difco Lactobacillus
medium (Difco, Detroit, MI) containing 20 µM dThd, and the
enzyme was fractionated through the heat and ammonium
sulfate precipitation steps exactly as described (15). The sample
was dialyzed and protein concentration was estimated using a
bicinchoninic acid assay according to the supplier’s instructions
(Pierce Chemical Co., Rockford, IL). The preparation derived
from 1 L of bacteria contained 8 mg of protein.
The synthesis of N2-(2-oxoethyl)dGuo on a semipreparative
scale is presented as an example. N2-(2-Oxoethyl)Gua (10 mg,
52 µmol) (12) was dissolved in 0.5 mL of 0.1 N HCl and diluted
in 10 mL of 0.5 M potassium 2-(N-morpholino)ethanesulfonate
buffer (pH 6.0) containing 180 mg (790 µmol) of dCyd. The
mixture was incubated with 250 µg of the trans-N-deoxyribo-
sylase enzyme preparation for 15 h at 37 °C under an Ar
atmosphere (to prevent bacterial growth). The mixture was
filtered through a 0.22 µm filter, and aliquots were injected onto
a semipreparative HPLC column as described above. The
product, eluting at tR 14.1 min (both N9 and N7 addition
products), was collected and the salts were removed by repeated
lyophilization to yield 15.3 mg of N2-(2-oxoethyl)dGuo [cyclized
form (12)]: 95% yield, ratio of N9:N7 products 95:5 as deter-
Ch em ica ls. Gua, dGuo, dCyd, dThd, N7-methylGua, N7-
methylAde, Cyt, 3,N4-ꢀ-Cyt, and 1,N6-ꢀ-Ade were purchased
from Sigma and used without further purification. [4,5,8-13C3]-
Gua was synthesized as described by Scheller et al. (19). N2-
(2-Oxoethyl)Gua was prepared from 2-fluorohypoxanthine as
described elsewhere (12) and treated with a mixture of pyridine
and (CH3CO)2O at room temperature to form 5,6,7,9-tetrahydro-
7-acetoxy-9-oxoimidazo[1,2-a]purine in quantitative yield: fast
atom bombardment MS, m/ z (assignment and relative abun-
1
mined by H NMR; UV λmax 280 nm (12); 1H NMR (N9 coupled
product) (2H2O) δ 2.46 (m, 1 H, H-2′a), 2.70 (m, H-2′b), 3.74 (m,
3 H, H-5′a, H-5′b, H-6a), 4.08 (m, 2 H, H-4′, H-6b), 4.56 (m, 1 H,
H-3′), 6.19 (t, 2 H, H-1′, H-7), 7.91 (s, 1 H, H-2); electrospray
mass spectrometry, m/ z 310 (MH+, 100), 194 (MH+ - deoxyri-
bose, 26).
dance in parentheses) 236 (MH+, 58), 176 [MH+ - CH3CO2
-
H2O (ꢀ-Gua), 20]; 1H-NMR ([2H6]Me2SO) δ 2.03 (s, 3 H, CH3-
CO), 3.51 (d, 1 H, H-6), 3.91 (dd, 1 H, H-6), 6.91 (d, 1 H, H-7),
7.68 (s, 1 H, NH), 8.14 (s, 1 H, H-2), 12.53 (s, 1 H, NH). C8-
(N-Anilino)Gua, C8-[N-(2-aminofluorenyl)]Gua, and C8-[N-(4-
aminobiphenyl)]Gua (all labeled with 3H in the aromatic rings)
were gifts of Dr. F. F. Kadlubar, National Center for Toxicology
Research, J efferson, AR. 1,N2-ꢀ-Gua and N2,3-ꢀ-Gua and their
deoxyribonucleosides were prepared as described previously (12,
20). C8-MethylGua was prepared by acid hydrolysis (70 °C, 1
N aq HCl, 1 h) of C8-methylGuo (21). The product precipitated
on cooling to ambient temperature and was recovered by
centrifugation and washed with H2O; analytical HPLC indicated
that the recovered base was homogeneous: fast atom bombard-
ment MS, m/ z 166 (MH+, 100). The same procedure was used
to prepare C8-aminoGua from C8-aminoGuo (Sigma): UV (H2O,
pH 7) λmax 291 nm (22); fast atom bombardment MS, m/ z 167
(MH+, 100).
In str u m en ta l An a lysis. HPLC analyses and preparative
work were done with reversed-phase octadecylsilane columns
(Phenomenex Partisil 5 ODS (3), 4.6 × 250 mm, 5 µm, Phe-
nomenex, Torrance, CA, orsfor larger sampless10 × 250 mm
Beckman Ultrasphere, 5 µm, Beckman, San Ramon, CA) con-
nected to a Spectra-Physics 8700 pumping system (Thermo-
Separation Products, Piscataway, NJ ), with the effluent passing
through a Hewlett Packard 1040A diode array detector (Hewlett
Packard, Palo Alto, CA). The separations used here all involved
increasing gradients of CH3OH in 50 mM NH4HCO2 (pH 5.0),
usually 0-50% CH3OH over 25 min, with a flow rate of 1.0 mL
min-1 (2.5 mL min-1 for preparative work) and monitoring
absorbance at 254 nm or the λmax of the product. 1H NMR
spectra were recorded in [2H6]Me2SO or H2O/2H2O using a
Resu lts a n d Discu ssion
P r ep a r a tion of tr a n s-N-Deoxyr ibosyla se. L. hel-
veticus cells were grown in the presence of dThd, and
trans-N-deoxyribosylase was partially purified as de-
scribed (15). In our experience, the enzyme activity is
stable for at least 6 months at 4 °C. The specific activity
of the preparation we made, using the heat and am-
monium sulfate precipitation steps, was 6.7 µmol of dGuo
formed h-1 (mg of protein)-1, which compared favorably
with an activity of 1.0 (same units) prepared by an
alternate streptomycin sulfate precipitation method (17)
and kindly provided by Dr. L. C. Sowers, City of Hope
Medical Center, Duarte, CA. trans-N-Deoxyribosylase
can be further purified by DEAE chromatography and
crystallization (15); however, we have not observed any
reactions with the bases or nucleosides that would
interfere with the use of the crude enzyme preparation
for most routine preparative work (except possibly in the
case of 1,N6-ꢀ-Ade, vide infra).
In studies on deoxyribosylation of the Gua, the extent
of nucleoside formation in a 3 h period was linear up to
7 µg of protein mL-1 (results not shown). We saw little
difference in rates over the pH range of 6-8, and the
addition of 5% Me2SO (v/v) to the reaction did not change
the rate. Our findings on the pH dependence are more
in line with those of Kanda and Takag (18) than Beck
and Levin (14), who reported a much sharper profile.

1142 Chem. Res. Toxicol., Vol. 9, No. 7, 1996
Mu¨ller et al.
F igu r e 2. Deoxyribosylation of 5,6,7,9-tetrahydro-7-acetoxy-
9-oxoimidazo[1,2-a]purine (structure shown) using (A) purine
nucleoside phosphorylase and (B) trans-N-deoxyribosylase. The
reactions were done for 18 h using either 93 µg (A) or 50 µg (B)
(protein, mL-1) of the designated enzyme fraction, and the
products were analyzed by HPLC. The diastereomeric products
1 and 2 are denoted in part B. (The split peaks for dThd and
dCyd are due to an overload of absorbance for the detector.)
Deoxyr ibosylation of 5,6,7,9-Tetr ah ydr o-7-acetoxy-
9-oxoim id a zo[1,2-a ]p u r in e. In the course of work with
N2-(2-oxoethyl)Gua, a DNA adduct formed from reaction
of dGuo or DNA with 2-chlorooxirane (12), we found that
purine nucleoside phosphorylase could catalyze the deox-
yribosylation of N2-(2-oxoethyl)Gua but not the O-acetyl
derivative, 5,6,7,9-tetrahydro-7-acetoxy-9-oxoimidazo[1,2-
a]purine. Other lots of the enzyme, a preparation of
recombinant enzyme (provided by Burroughs-Wellcome
through Prof. L. J . Marnett of this department), and calf
spleen purine nucleoside phosphorylase (Sigma) were all
unable to catalyze this reaction. However, trans-N-
deoxyribosylase efficiently converted the base to a nu-
cleoside (Figure 2).
Two peaks of equal size were detected as products.
Both had nearly identical UV, mass, and 1H NMR
spectra, and they do not appear to be positional isomers.
The carbon (C7) attached to the acetoxy group is chiral,
and the addition of chiral deoxyribose generates diaster-
eomers. In support of this view, the two products had
opposite CD spectra (data not presented). The CD
spectra of both disappeared upon treatment with 0.1 N
NaOH at ambient temperature, consistent with the view
that after hydrolysis of the acetoxy group epimerization
occurred at the C7 atom of the 5,6,7,9-tetrahydro-7-
hydroxy-9-oxoimidazo[1,2-a]purine through opening to
N2-(2-oxoethyl)Gua and reclosing of the ring.
F igu r e 1. 1H NMR spectra of dGuo prepared from [4,5,8-13C3]-
Gua by (A) trans-N-deoxyribosylase and (B) purine nucleoside
phosphorylase. Spectra were recorded in 2H2O/H2O. In part B,
the assignments of the 3′ and 4′ protons are indicated for the
two forms with deoxyribose attached at the N7 and N9 atoms.
Asterisks (*) denote 13C atoms. The “prime” and “double prime”
denote the anomeric hydrogens at the 2′ and 5′ positions of the
deoxyribose.
Deoxyr ibosyla tion of Gu a . In the course of prepar-
ing [4,5,8-13C3]dGuo by deoxyribosylation of the base with
purine nucleoside phosphorylase, we found that a sig-
nificant fraction of the product contained the deoxyribose
moiety attached to the Gua N7 atom instead of the N9
atom, a problem noted previously in the synthesis of
deoxyribosylpyrimido[1,2-a]purin-10(3H)-one (M1G) with
this enzyme (10). The two products can be separated by
HPLC using a shallow gradient, after initial HPLC
fractionation with a sharper gradient, but the need for
the second step limits the usefulness of the procedure.
The crude dGuo peak was collected from the initial
HPLC from reactions done with either trans-N-deoxyri-
bosylase or purine nucleoside phosphorylase and ana-
1
lyzed by H NMR (Figure 1). In the case of the product
formed by trans-N-deoxyribosylase (Figure 1A), the spec-
trum is consistent with the presence of only dGuo. The
spectrum of the product obtained with purine nucleoside
phosphorylase (Figure 1B) contains ∼40% of the isomer
with the deoxyribose attached at the N7 atom, as best
demonstrated by the shifts of the 3′ and 4′ sugar protons
(23). (In both cases the H-2 proton is split by the geminal
13C.) The yield of total “dGuo” (N9 and N7 linkage) was
essentially quantitative with both enzymes under these
conditions.
Deoxyr ibosyla tion of 1,N2-E-Gu a by tr a n s-N-Deox-
yr ibosyla se. The above experiments with Gua and
5,6,7,9-tetrahydro-7-acetoxy-9-oxoimidazo[1,2-a]purine in-
dicated considerable regioselectivity in deoxyribosylation
of bases, at least Gua derivatives, by trans-N-deoxyribo-
sylase. 1,N2-ꢀ-Gua was examined as a substrate. The
reaction proceeded to 86% total deoxyribosylation under
the conditions used (25 µg of protein mL-1, overnight at
37 °C). HPLC analysis indicated the presence of a minor

DNA Adduct Synthesis with trans-N-Deoxyribosylase
Chem. Res. Toxicol., Vol. 9, No. 7, 1996 1143
Ta ble 1. Deoxyr ibosyla tion of Ba ses w ith
tr a n s-N-Deoxyr ibosyla se
combined yield
(total N9 and N7 coupling)a
ratio
coupling at
5 µg of protein
25 µg of
N9 and N7
base
3 h
15 h protein, 15 h
atomsa
Gua
>95
100:0
95:5
100:0
N2-(2-oxoethyl)Gua
5,6,7,9-tetrahydro-7-
acetoxy-9-oxoimid-
azo[1,2-a]purine
M1G
36
88
>95
90
11
15
1
58
64
5
16
66
68
85
97
86
14
49
85:15
86:14
50:50
92:8
c
c
c
c
1,N2-ꢀ-Gua
N2,3-ꢀ-Gua (pH 6.0)
N2,3-ꢀ-Gua (pH 8.0)
3,N4-ꢀ-Cytb
2
12
23
40
79b
1,N6-ꢀ-Ade
94d
91
F igu r e 3. 1H NMR spectra of 1,N2-ꢀ-dGuo formed by trans-
N-deoxyribosylase. (A) 1-Dimensional spectrum (2H2O/H2O
mixture). (B) NOE difference spectrum. Irradiation was at H-1′.
The “prime” and “double prime” denote the anomeric hydrogens
at the 2′ and 5′ positions of the deoxyribose.
C8-methylGua
C8-aminoGua
58e
<1f
<1f
C8-(N-anilino)Gua
C8-[N-(4-aminobiphenyl)]-
Gua
<1
<1
<1
<1
C8-[N-(2-aminofluorenyl)]- <1
<1
<1f
Gua
N7-methylGua
N7-methylAde
<1
<1
<1
<1
<1f
<1f
a
The N7/N9 nomenclature indicates the atoms in the basic
purine nucleus. In some derivatives the numbering system
b
changes (e.g., ꢀ-Gua, 3,N4-ꢀ-Cyt). 3,N4-ꢀ-dCyd: MS (fast atom
bombardment), m/ z 274 (M + Na, 28), 252 (MH+, 54), 223 (M -
28, 100). c A single HPLC peak was seen, but the regioselectivity
d
was not examined by NMR. The product 1,N6-ꢀ-dAdo had the
correct retention time as an authentic sample (Sigma), and its
identity was further verified by its characteristic UV (λmax 275,
266 nm at pH 7) and fluorescence spectra (emission maximum at
418 nm with excitation at 225 nm) (25). e UV λmax 288, 255 nm
(ratio 0.6, pH 5) (22). f No new peaks were found (with the
approximate expected tR values); further, none of the characteristic
Cyt was formed from dCyd in these reactions.
NMR, and mass spectrometry. The apparent fraction
linked at the N9 atom instead of N7 (using the number-
ing system for Gua) was estimated from 1H NMR
measurements, using the chemical shifts of the sugar
protons as a guide (e.g., Figures 1 and 4). Different
enzyme concentrations and reaction times were used.
The reactions with N2-(2-oxoethyl)Gua and M1G pro-
ceeded in high yield and high, but not complete, regiose-
lectivity.
F igu r e 4. 1H NMR of the minor product of the deoxyribosy-
lation of 1,N2-ꢀ-Gua with trans-N-deoxyribosylase. The product
was separated by HPLC and eluted immediately following 1,N2-
ꢀ-dGuo. The structure is consistent with coupling of the deox-
yribose to the N7 atom of 1,N2-ꢀ-Gua. (A) 1-Dimensional
spectrum (2H2O/H2O/Me2SO mixture). (B) NOE difference spec-
trum, with irradiation at H-1′. The “prime” and “double prime”
denote the anomeric hydrogens at the 2′ and 5′ positions of the
deoxyribose.
component eluting just after 1,N2-ꢀ-dGuo. The major
peak showed the expected UV spectra diagnostic of 1,N2-
ꢀ-(d)Guo, including the strong shift in alkali (24). NMR
and nuclear Overhauser effect (NOE) analysis indicated
proximity of the H-1′ proton to H-2′ and H-2, as expected
(Figure 3).
We considered the possibility that coupling of the
deoxyribose could have occurred at either the N1 or N5
atom (corresponding to N7 and N2 of Gua, respectively).
The minor product was separated and analyzed by NMR
spectroscopy (Figure 4). The spectrum indicated some
1,N2-ꢀ-dGuo as a contaminant, but the major component
had additional signals at δ 6.6 and 8.5. When the 1′
proton was irradiated, NOE changes were seen in the
signals assigned as H-2 and H-2′. Thus, the structure
of this compound involves linkage of the deoxyribose to
the N1 atom of 1,N2-ꢀ-Gua. Linkage at the N5 atom is
not an option, because the NOE would have been
expected at H-6 instead of H-2.
The reaction with N2,3-ꢀ-Gua did not work very well
at pH 6.0 but was considerably better at pH 8.0, in terms
of total yield and apparent regiochemistry. This result
can be rationalized by the instability of the glycosidic
bond of N2,3-ꢀ-dGuo at acid pH (26).2
trans-N-Deoxyribosylase also transfers pyrimidines
(15), and 3,N4-ꢀ-Cyt was a substrate.
1,N6-ꢀ-Ade was a good substrate, although an uniden-
tified peak, of unknown origin, eluted after the product
1,N6-ꢀ-dAdo and it is not clear if this is a degradation
product. Neither the UV nor the fluorescence spectra
suggested any relationship to 1,N6-ꢀ-dAdo.
C8-MethylGua was deoxyribosylated in high yield, with
no extra HPLC peaks. However, the C8-Gua arylamine
adducts were not utilized by the enzyme. This could be
due to steric bulk, even with the anilino derivative, or
2 The difference between the N9:N7 coupling ratios of 86:14 for 1,N2-
ꢀ-dGuo and 92:8 for N2,3-ꢀ-dGuo (pH 8.0) is not considered to be very
great. Steric hindrance is not the only determinant of the glycosydic
instability of N2,3-ꢀ-dGuo. The H-5 proton does show an NOE with
the H-1′ proton (12), but space-filling models indicate that the sugar
is easily accommodated, and H-1′ proximity to the H-5 atom is
consistent with an anti conformation. Also, the glycosydic stability of
N2,3-ꢀ-dGuo is influenced by pH, 5′-phosphorylation, and 2′-hydroxy-
lation (ribose substitution) (26).
Deoxyr ibosyla tion of Oth er Ba ses by tr a n s-N-
Deoxyr ibosyla se. Several bases were tested as sub-
strates for trans-N-deoxyribosylase (Table 1). The prod-
ucts were either available as standards, have been
characterized above, or else were analyzed by UV, 1H

1144 Chem. Res. Toxicol., Vol. 9, No. 7, 1996
Mu¨ller et al.
(3) Guengerich, F. P. (1992) Metabolic activation of carcinogens.
Pharmacol. Ther. 54, 17-61.
(4) Garner, P., and Ramakanth, S. (1988) A regiocontrolled synthesis
of N7- and N9-guanine nucleosides. J . Org. Chem. 53, 1294-1298.
(5) Krenitsky, T. A., Koszalka, G. W., and Tuttle, J . V. (1981) Purine
nucleoside synthesis, an efficient method employing nucleoside
phosphorylases. Biochemistry 20, 3615-3621.
(6) Krenitsky, T. A., Rideout, J . L., Chao, E. Y., Koszalka, G. W.,
Gurney, F., Crouch, R. C., Cohn, N. K., Wolberg, G., and Vinegar,
R. (1986) Imidazo[4,5-c]pyridines (3-deazapurines) and their
nucleosides as immunosuppressive and antiinflammatory agents.
J . Med. Chem. 29, 138-143.
(7) Holy, A., and Votruba, I. (1987) Facile preparation of purine and
pyrimidine 2-deoxy-â-D-ribonucleosides by biotransformation on
encapsulated cells. Nucleic Acids Symp. Ser. 18, 69-72.
(8) Massefki, W., J r., Redfield, A., Sarman, U. D., Bannerji, A., and
Roy, S. (1990) [7-15N]Guanosine-labeled oligonucleotides as nuclear
magnetic resonance probes for protein-nucleic acid interaction
in the major groove. J . Am. Chem. Soc. 112, 5351-5353.
(9) Gaffney, B. L., Kung, P-P., and J ones, R. A. (1990) Nitrogen-15-
labeled deoxynucleosides. 2. Synthesis of [7-15N]-labeled deoxy-
adenosine, deoxyguanosine, and related deoxynucleosides. J . Am.
Chem. Soc. 112, 6748-6749.
(10) Chapeau, M.-C., and Marnett, L. J . (1991) Enzymatic synthesis
of purine deoxynucleoside adducts. Chem. Res. Toxicol. 4, 636-
638.
(11) Humphreys, W. G., and Guengerich, F. P. (1991) Structure of
formamidopyrimidine adducts as determined by NMR using
specifically 15N-labeled guanosine. Chem. Res. Toxicol. 4, 632-
636.
(12) 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.
(13) McNutt, W. S. (1952) The enzymatically catalysed transfer of the
deoxyribosyl group from one purine or pyrimidine to another.
Biochem. J . 50, 384-397.
(14) Beck, W. S. and Levin, M. (1963) Purification, kinetics, and
repression control of bacterial trans-N-deoxyribosylase. J . Biol.
Chem. 238, 702-709.
(15) Uerkvitz, W. (1971) trans-N-deoxyribosylase from Lactobacillus
helveticus: crystallization and properties. Eur. J . Biochem. 23,
387-395.
(16) Bessman, M. J ., Muzyczka, N., Goodman, M. F., and Schnaar, R.
L. (1974) Studies on the biochemical basis of spontaneous muta-
tion. II. The incorporation of a base and its analogue into DNA
by wild-type, mutator and antimutator DNA polymerases. J . Mol.
Biol. 88, 409-421.
(17) Sowers, L. C., Mhaskar, D. N., Khwaja, T. A., and Goodman, M.
F. (1989) Preparation of imino and amino N-15 enriched 2-ami-
nopurine deoxynucleoside. Nucleosides Nucleotides 8, 23-34.
(18) Kanda, M., and Takag, Y. (1959) Purification and properties of a
bacterial deoxyribose transferase. J . Biochem. 46, 725-732.
(19) 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.
(20) Guengerich, F. P. (1992) Roles of the vinyl chloride oxidation
products 2-chlorooxirane and 2-chloroacetaldehyde in the in vitro
formation of etheno adducts of nucleic acid bases. Chem. Res.
Toxicol. 5, 2-5.
(21) Humphreys, W. G., Kadlubar, F. F., and Guengerich, F. P. (1992)
Mechanism of C8 alkylation of guanyl residues by activated
arylamines: evidence for a role of initial attack at the N7 atom.
Proc. Natl. Acad. Sci. U.S.A. 89, 8278-8282.
(22) Stoeckler, J . D., Cambor, C., Kuhns, V., Chu, S.-H., and Parks,
R. E., J r. (1982) Inhibitors of purine nucleoside phosphorylase.
C(8) and C(5′) substitutions. Biochem. Pharmacol. 31, 163-171.
(23) Chapeau, M.-C. Doctoral Thesis, Wayne State University, 1992.
(24) Sattsangi, P. D., Leonard, N. J ., and Frihart, C. R. (1977) 1,N2-
Ethenoguanine and N2,3-ethenoguanine. Synthesis and compari-
son of the electronic spectral properties of these linear and
angular triheterocycles related to the Y bases. J . Org. Chem. 42,
3292-3296.
due to electrostatic effects. In order to consider the
possibility of an electronic effect, we examined C8-
aminoGua, which has been reported to be deoxyribosy-
lated by mammalian purine nucleoside phosphorylase
(22). This base was also deoxyribosylated by trans-N-
deoxyribosylase (58% yield, Table 1). Apparently the
bulk of even C8-(N-anilino)Gua is a problem for purine
nucleoside phosphorylases (10) and trans-N-deoxyribo-
sylase.
Neither N7-methylGua nor N7-methylAde was a sub-
strate. The addition of a sugar at the N9 atom requires
an electronically different transition state, since the
product bears a positive charge.
Use of tr a n s-N-Deoxyr ib osyla se in R em ova l of
Deoxyr ibose. In all of the studies thus far, the object
has been to add deoxyribose to a modified base. The
reaction of trans-N-deoxyribosylase is one in which an
equilibrium is reached between two bases (Scheme 1),
and we considered the possibility that there might be
situations in which it would be desirable to use an
enzymatic reaction in removing the deoxyribose to gener-
ate a base, without the use of acid or heat. The deox-
yribosylation of C8-methyl dGuo was considered as an
example. The general reaction conditions were used (i.e.,
pH 6.0, 0.5 mM C8-methyl dGuo, and a 10-fold molar
excess of Cyt). In the presence of 5 µg of trans-N-
deoxyribosylase mL-1, yields of C8-methyl Gua were 26%
and 86% after 3 and 20 h, respectively, at 37 °C. With
25 µg of protein mL-1, the yield was 92% after 20 h.
Con clu sion s. L. helveticus trans-N-deoxyribosylase
is a useful enzyme for attaching deoxyribose to normal
and modified purines and pyrimidines. The bacterial
preparation is relatively easy to obtain and is stable.
Comparisons of purines examined previously with purine
nucleoside phosphorylase indicate comparable yields and
regioselectivity (10). In some cases (e.g., Gua) regiose-
lectivity appears better (Figure 1). The lack of regiose-
lectivity may be a thermodynamic rather than a kinetic
factor, which could be exacerbated in efforts to increase
yields. A dramatic example of the broader specificity of
trans-N-deoxyribosylase was seen in the case of the
deoxyribosylation of 5,6,7,9-tetrahydro-7-acetoxy-9-ox-
oimidazo[1,2-a]purine, a requisite intermediate in the
synthesis of oligonucleotides modified with N2-(2-oxo-
ethyl)Gua.
Ack n ow led gm en t. This work was supported in part
by NIH Grants CA44353 and ES00267. M.M. was the
recipient of a postdoctoral fellowship from the Deutsche
Forschungsgemeinschaft. We thank Dr. R. Reddy and
Prof. L. J . Marnett for providing M1G, deoxyribosyl-M1G,
and recombinant E. coli purine nucleoside phosphorylase,
Dr. W. G. Humphreys for synthesizing C8-methylGuo, Dr.
F. F. Kadlubar for providing the C8-arylamine Gua
adducts, M. Voehler for assistance with some of the NMR
studies, and particularly Dr. L. C. Sowers for suggesting
the use of trans-N-deoxyribosylase with some of these
adducts and providing an initial enzyme preparation and
L. helveticus cells.
(25) Barrio, J . R., Secrist, J . A., III, and Leonard, N. J . (1972)
Fluorescent adenosine and cytidine derivatives. Biochem. Biophys.
Res. Commun. 46, 597-604.
(26) 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.
Refer en ces
(1) Miller, E. C., and Miller, J . A. (1981) Searches for ultimate
chemical carcinogens and their reactions with cellular macro-
molecules. Cancer 47, 2327-2345.
(2) Heidelberger, C. (1975) Chemical carcinogenesis. Annu. Rev.
Biochem. 44, 79-121.
TX9600661