Enzymatic synthesis of 2'-deoxyguanosine with nucleoside deoxyribosyltransferase-II.

Article abstract of DOI:10.1271/bbb.67.989

Nucleoside deoxyribosyltransferase-II (NdRT-II) of Lactobacillus helveticus, which catalyzes the transfer of a glycosyl residue from a donor deoxyribonucleoside to an acceptor base, has a broad specificity for the acceptor bases. Six-substituted purines were found to be substrates as acceptor bases for NdRT-II. Using this property of the enzyme, we established a practical procedure for enzymatic synthesis of 2'-deoxyguanosine (dGuo), consisting of the transglycosylation from thymidine to 6-substituted purine (2-amino-6-chloropurine; ACP) instead of natural guanine and the conversion of 2-amino-6-chloropurine-2'-deoxyriboside (ACPdR) to dGuo with bacterial adenosine deaminase. Through the successive reactions, dGuo was synthesized in high yield.

Full text of DOI:10.1271/bbb.67.989

This article was downloaded by: [University of Windsor]
On: 13 November 2014, At: 17:51
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer
House, 37-41 Mortimer Street, London W1T 3JH, UK
Bioscience, Biotechnology, and Biochemistry
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/tbbb20
Enzymatic Synthesis of 2′-Deoxyguanosine with
Nucleoside Deoxyribosyltransferase-II
a
a
a
a
Kiyoshi OKUYAMA , Susumu SHIBUYA , Tomoki HAMAMOTO & Toshitada NOGUCHI
a
Biochemicals Division, Yamasa CorporationChoshi, Chiba 288-0056, Japan
Published online: 22 May 2014.
To cite this article: Kiyoshi OKUYAMA, Susumu SHIBUYA, Tomoki HAMAMOTO & Toshitada NOGUCHI (2003) Enzymatic
Synthesis of 2′-Deoxyguanosine with Nucleoside Deoxyribosyltransferase-II, Bioscience, Biotechnology, and Biochemistry,
67:5, 989-995, DOI: 10.1271/bbb.67.989
To link to this article: http://dx.doi.org/10.1271/bbb.67.989
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of
the Content. Any opinions and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied
upon and should be independently verified with primary sources of information. Taylor and Francis shall
not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other
liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or
arising out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions

Biosci. Biotechnol. Biochem., 67 (5), 989–995, 2003
Enzymatic Synthesis of 2?-Deoxyguanosine with Nucleoside
Deoxyribosyltransferase-II
Kiyoshi OKUYAMA,† Susumu SHIBUYA, Tomoki HAMAMOTO, and Toshitada NOGUCHI
Biochemicals Division, Yamasa Corporation, Choshi, Chiba 288-0056, Japan
Recived October 24, 2002; Accepted January 10, 2003
Nucleoside deoxyribosyltransferase-II (NdRT-II) of
Lactobacillus helveticus, which catalyzes the transfer of
a glycosyl residue from a donor deoxyribonucleoside to
an acceptor base, has a broad speciˆcity for the acceptor
bases. Six-substituted purines were found to be sub-
strates as acceptor bases for NdRT-II. Using this
property of the enzyme, we established a practical
(NdRT; EC 2.4.2.6), have been used to synthesize
nucleosides and nucleoside analogues by transferring
ribosyl residues from a donor to an acceptor base.
Some organisms, notably the lactobacilli, in which
nucleoside phosphorylases activity can be hardly
detected, have appreciable quantities of nucleoside
deoxyribosyltransferase. It is generally considered
that NdRT is involved in an essential step in the
salvage pathway of deoxyribonucleoside biosynthesis
procedure for enzymatic synthesis of 2?-deoxyguanosine
(
dGuo), consisting of the transglycosylation from
thymidine to 6-substituted purine (2-amino-6-chloropu-
rine; ACP) instead of natural guanine and the conver-
for lactobacilli. Two types of
N-deoxyribosyltran-
sferases have been identiˆed in both Lactobacillus
3
)
4,5)
sion
of
2-amino-6-chloropurine-2?-deoxyriboside
helveticus and L. leichmannii
.
Type I (NdRT-I)
(
ACPdR) to dGuo with bacterial adenosine deaminase.
catalyzes the reversible transfer of a deoxyribose
residue between a purine base and a purine nucleo-
side, while type II (NdRT-II) catalyzes the transfer of
a deoxyribose residue between purine or pyrimidine
bases and purine or pyrimidine nucleosides as fol-
lows:
Through the successive reactions, dGuo was synthesized
in high yield.
Key words: nucleoside deoxyribosyltransferase;
2?-
deoxyribonucleoside; 2
-amino-6-chloropurine;
deaminase
?
-deoxyguanosine;
2
adenosine
dRib-Pur
dRib-Pyr
dRib-Pur
＋
Pur
Pyr
Pyr
?
⇔
⇔
⇔
dRib-Pur
?
＋
Pur
＋
?
dRib-Pyr
?
＋
Pyr
Recently there is pressing worldwide demand for
-deoxyribonucleosides as the raw materials for
antisense drugs. Natural 2 -deoxyribonucleosides
＋
dRib-Pyr
＋
Pur
2?
?
where dRib, Pur, and Pyr represent 2-deoxy- -
D
have been mainly prepared and supplied from
hydrolysate of DNA. However, this procedure in-
cludes an uncertain supply of DNA from natural
source (ˆsh sperms) and complicated isolation proc-
esses. Therefore, another procedure using chemical
ribose, a purine base, and a pyrimidine base, respec-
tively.
Recently, we have successfully cloned the L. hel-
6
)
veticus NdRT-II gene and been trying to develop a
process for the production of deoxyribonucleosides
by NdRT-II overproduced in E. coli cells. Using this
enzyme, dAdo and dCyd were conˆrmed to be
e‹ciently synthesized from dThd or dUrd and corre-
sponding bases.
and or enzymatic reactions has been expected to take
W
the place of hydrolysis of DNA. Thymidine (dThd)
and 2?-deoxyuridine (dUrd) can be readily synthe-
sized at a moderate price by chemical methods. On
the contrary, other 2 -deoxyadenosine (dAdo), 2
deoxycytidine (dCyd), and -deoxyguanosine
dGuo) can hardly be synthesized by the same way.
Although many attempts to enzymatically synthesize
?
?-
On the contrary, dGuo was scarcely synthesized
from dThd or dUrd using guanine as a base acceptor
because of the low solubility of guanine. Thus, using
soluble 6-substituted purine bases, several guanosine
derivatives have been tested to be synthesized from
dUrd by bacterial conversion due to their nucleoside
2?
(
1
)
the latter deoxyribonucleosides have been reported,
but a practical method for producing these deox-
yribonucleosides has not been developed.
7
–9)
phosphorylase activities followed by hydrolysis of
Two main classes of enzymes, nucleoside phos-
an amino group or a halogen at position 6 of the base
2
)
10)
phorylases and nucleoside deoxyribosyltransferases
by chemical and or enzymatic reaction with mam-
W
†
To whom correspondence should be addressed. Fax:
81-479-22-9865; E-mail: okuyama yamasa.com
＋ ＠
Abbreviations: DAP, 2,6-diaminopurine; ACP, 2-amino-6-chloropurine; HPLC, high performance liquid chromatography; IPTG,
isopropyl- -thiogalactopyranoside; NdRT, nucleoside deoxyribosyltransferase; Pu-NPase, purine nucleoside phosphorylase; Py-NPase,
pyrimidine nucleoside phosphorylase
b-D

9
90
K. OKUYAMA et al
.
7
,11–13)
malian adenosine deaminase.
However, this
tion at 10,000
um sulfate was gradually added until the salt concen-
tration was 45 of saturation. The precipitates were
×
g for 20 min at 49C. Solid ammoni-
procedure contains degradation of deoxyribonucleo-
sides by phosphorolysis activity of nucleoside phos-
phorylases, and the troublesome preparation process
of mammalian adenosine deaminase. In this paper,
we describe the development of a practical method
for enzymatic synthesis of dGuo by the combination
of transglycosylation with NdRT-II from thymidine
to a 2-amino-6-substituted purine, and the hydrolysis
reaction with bacterial adenosine deaminase.
z
collected by centrifugation, dissolved in TE buŠer,
and dialyzed against TE buŠer. The dialyzed solution
was put on a DE52 (Whatman) column (
13 cm) equilibrated with TE buŠer. Elution was done
with a linear gradient of 0 to 0.3 NaCl (1,380 ml).
q4.3
×
M
The active fractions were pooled and concentrated to
3 ml by centrifugation with Centriprep-10 (Amicon).
The concentrated solution was put on a Sepharose
Materials and Methods
S-300 (Pharmacia Biotechnology) column (
q
1.6
×
9
0 cm) equilibrated with TE buŠer containing 0.1
M
Chemicals
diaminopurine-2
chloropurine were purchased from Sigma Aldrich
USA). Other nucleosides and bases were prepared
by Yamasa Corporation. Adenosine deaminase (EC
.5.4.4, from calf intestine) was the product of
.
2,6-diaminopurine (DAP), 2,6-
NaCl (0.4 ml min). The active fractions, in which
W
?
-deoxyriboside, and 2-amino-6-
neither cytidine deoxycytidine deaminase activity
nor adenosine deaminase activity was detected, were
W
(
combined (12.8 ml; 36 units mg-protein) and stored
W
9
at 4 C.
3
Roche Diagnostics (Germany). Other chemicals used
were the best available commercial products.
Nucleoside phosphorylases. Purine nucleoside
phosphorylase II (Pu-NPase II; EC 2.4.2.1) and
pyrimidine nucleoside phosphorylase (Py-NPase; EC
2.4.2.2) were produced in E. coli JM105 carrying an
expression plasmid, pTrc-punA and pTrc-pyn, re-
spectively. Cultivation and preparation of enzyme
Microorganism, plasmid, and medium
. E. coli
JM105 and JM109 were used as host cells for plas-
1
4)
mid-transformation. Plasmid pDR-540
carrying
1
6,17)
the tac promoter was purchased from Pharmacia
LKB Biotechnology. The culture medium contained
solutions were done as described previously.
1
0 g peptone, 5 g yeast extract, and 5 g NaCl in dis-
E. coli adenosine deaminase. Adenosine
deaminase was produced in JM105[pDR-add] with
tilled water to make the total volume of 1 l, and was
adjusted to pH 7.0 with NaOH.
IPTG (0.1 m
um with 100
M
) induction in 500 ml of culture medi-
W
m
g ml of ampicillin. For protein puriˆ-
Cloning of the E. coli adenosine deaminase gene
.
cation, cells were harvested after 4 h of induction at
37 C and suspended in standard buŠer containing
20 m Tris-HCl (pH 8.0), 2 m CaCl , 0.05 m
EDTA, and 20 (v v) ethylene glycol. The cells were
1
5)
The adenosine deaminase gene (add
)
of E. coli was
9
cloned by using PCR with oligonucleotides primer-1
M
M
2
M
(
5
?
-TAGGATCCACCATGATTGATACCACCC-
TG-3 ) and primer-2 (5 -AAGGATCCTTTGTGCC-
TTGCATGAAAAAT-3 ) and E. coli C600 chromo-
some DNA as a template. The parameters were 25
cycles with three steps (1 min at 94 C, 2 min at 55 C,
min at 72
z
W
?
?
?
disrupted by sonication, and the cell-free extract was
recovered by removal of cell debris with centrifuga-
tion at 10,000
extract was put on a DE52 column (
equilibrated with standard buŠer. Elution was done
with a linear gradient of 50 to 250 m NaCl (600 ml).
×
g
for 20 min at 4
9
q
C. The crude
9
9
2.2 15 cm)
×
3
9
C). The resulting PCR product was
digested with the restriction enzyme BamHI, puri-
M
ˆed, and ligated to an expression vector, pDR540
The active fractions were pooled and diluted with an
equal volume of standard Tris buŠer. The diluted
enzyme solution was put on a DE 52 column (2 ml)
equilibrated with standard buŠer. Elution was done
(
Pharmacia LKB Biotechnology) digested with
BamHI. The ligation mixture was used to transform
E. coli JM105, and the obtained transformant was
named JM105[pDR-add].
with standard buŠer containing 250 m
4.5 ml). The eluted enzyme fraction was put on a
Sepharose S-300 column ( 1.6 90 cm) equilibrated
NaCl
M
NaCl
(
Enzyme preparations. All operations were done at
q
×
4
9
C unless otherwise indicated.
with standard Tris buŠer containing 0.1
M
NdRT-II. The NdRT-II was produced in E. coli
(1.6 ml min). The active fractions having no detecta-
W
7
)
JM109 carrying pTrc-T2F4 with IPTG (0.1 m
induction in 500 ml of culture medium with
00 g ml of ampicillin. For protein puriˆcation,
cells were harvested after 20 h of induction at 25
and suspended in TE buŠer consisting 10 m Tris-
HCl (pH 8.0) and 1 m EDTA. The cells were dis-
M
)
ble deoxyguanosine-degradation activity were com-
bined (20 ml; 206.9 units mg-protein) and stored at
W
1
m
4
9
C.
W
9C
M
Protein assay. Protein concentration was meas-
M
ured with the Bio-Rad protein assay dye reagent (Bio-
Rad Laboratories) using bovine serum albumin as a
standard.
rupted by sonication, and the cell-free extract was
recovered by removal of cell debris with centrifuga-

Enzymatic Synthesis of 2?-Deoxyguanosine
991
Enzyme assay. NdRT-II activity was measured as
follows: the reaction mixture (0.2 ml) contained
mole of thymidine, 1 mole of cytosine, 4 moles
Table 1. Substrate Speciˆcity of NdRT-II
Reaction mixtures (5 ml) containing 1 or 50 m
M
deoxyribose
donor, 1 or 50 m
M
base acceptor, and NdRT-II (2.38 units mmole
W
1
m
m
m
of substrate) were incubated at 40
9
C at pH 6.4 with stirring.
of sodium citrate (pH 6.0), and the enzyme. The
reaction was started by addition of the enzyme, done
Results are expressed as the concentration of products and reac-
tion e‹ciency as the molar yield of transglycosylation at equilibri-
um after 20 h of eaction.
Ade, adenine; Cyt, cytosine; Gua, guanine; dThd, thymidine;
dAdo, deoxyadenosine; dCyd, deoxycytidine; dGuo, deoxyguano-
sine.
at 40
9C , and stopped by addition of 0.2 ml of 0.1
M
NaOH. The amount of deoxycytidine was measured
by HPLC. One unit of enzyme was deˆned as the
amount of enzyme forming 1 mmole of deoxycytidine
per min under these conditions.
Deoxyribosyl
donor
Base
Product
(m
Molar yield
acceptor
M
)
(z)
Pu-NPase II and Py-NPase activities were meas-
1
6,17)
ured as described previously.
dThd
Ade (1 m
Cyt (1 m
M
)
dAdo 0.83
dCyd 0.49
dAdo 0.28
83.3
48.8
27.7
Adenosine deaminase activity was measured as
follows: the reaction mixture (5 ml) contained
M
)
(1 mM)
Gua (1 m
M
)
1
50
pH 8.2), and the enzyme. The reaction was started
by addition of the enzyme, done at 37 C, and
stopped by addition of an equal volume of 0.1
mmoles of adenosine, 250 mmoles of Tris-HCl
Ade (50 m
M
)
dAdo 43.5
dCyd 24.9
dGuo 0.53
87.0
49.8
1.1
(
dThd
(50 mM)
Cyt (50 m
M)
9
Gua (50 m
M
)
M
NaOH and boiling for 2 min. The increase of inosine
was measured by HPLC. One unit of enzyme was
deˆned as the amount of enzyme forming a
inosine per min under these conditions.
m
mole of
done with 50 units of bacterial adenosine deaminase
at 37 C by adjusting the pH to 7.8 with 1 HCl or
NaOH. The reaction was stopped by addition of
an equal volume of 0.1 NaOH and boiling for
2 min. The amount of dGuo was measured by
HPLC.
9
N
1
N
Transglycosylation reactions. Five ml of reaction
mixtures were contained 1 or 50 m of deoxyribosyl
donor, and 1 or 50 m of base acceptor were adjust-
ed pH 6.0–6.3 with HCl or NaOH. The reaction was
started by addition of NdRT-II (2.38 units mmole-
N
M
M
Analysis. Quantitative determination of nucleo-
W
substrate) and done at 40
reaction was done by mixing with an equal volume of
.01 NaOH and boiling for 2 min. Samples were as-
9C . Termination of the
sides and bases was done by HPLC with a YMC
packed column AQ302 (4.6
cal, Japan) at 20
for ACPdR and ACP at 300 nm. The mobile phase
×
150 mm, Eishin chemi-
0
N
9
C with detection at 260 nm, except
sayed by HPLC and the extent of reaction was esti-
mated as the percentage of deoxyribosyl transferred
from donor to acceptor. The extent of reactions after
was 50 m
M
KH
2
4
z
PO containing 7 acetonitrile, and
the ‰ow rate was 0.6 ml min.
W
1
7 h and 20 h were practically identical.
Results
Hydrolysis of deoxyribonucleosides with adeno-
sine deaminase from calf intestine and E. coli. One
Enzymatic synthesis of dAdo, dCyd, and dGuo
with NdRT-II
ml of reaction mixtures were contained 10 m
deoxyribonucleoside (dAdo, DAPdR, or ACPdR),
00 m of Tris-HCl (pH 8.0), and the enzyme. The
reaction was started by addition of calf intestine or
E. coli ADA (1.5 units ml-reacton mixture) and
M
of
Table 1 shows the transglycosylation e‹ciencies at
equilibrium using thymidine as a deoxyribosyl donor.
dAdo and dCyd were e‹ciently synthesized from
dThd. In the case of dAdo synthesis, the reaction
mixtures contained three to four times more dAdo
than dThd at equilibrium. When the dCyd synthesis
reaction reached equilibrium, concentrations of
dCyd and dThd were approximately equal. In the
synthesis of both dAdo and dCyd, increase of sub-
strate concentration (to 50 m ) did not change the
initial reaction rate or extent of reaction.
On the other hand, the yield of dGuo was lower
than those of dAdo and dCyd, possibly because of
1
M
W
done at 37
9
C. Termination of the reaction was done
by mixing with an equal volume of 0.01
N
NaOH and
boiling for 2 min. The decrease of substrate was
measured by HPLC.
Conversion of deoxyribonucleosides in trans-
glycosylation mixture to deoxyguanosine with adeno-
M
sine deaminase. Ten
ml of the resultant transglycosylation reaction mix-
tures described in Table 4 (containing 48.2 m
DAPdR and 46.4 m ACPdR) to inactivate NdRT-
II, and 100 l of 1 Tris-HCl buŠer (pH 7.5) was
added to each of the solutions. The pH of the solu-
tion was adjusted to 7.8 with 0.1 HCl for the
adenosine deaminase reaction. The reaction was
m
l of 20
z
NaOH were added to
1
M
the low solubility of guanine (0.1 m
40 C). Especially, the molar yield of reaction at
50 m substrates was extremely lower than that at
1 m substrates. These results demonstrate that gua-
M
in water at
M
9
m
M
M
M
N
nine is not available for the NdRT-II reaction.
dUrd was also conˆrmed to be available as a deox-

9
92
K. OKUYAMA et al
.
Table 2. Enzymatic Synthesis of DAPdR and ACPdR from
Table 3. Comparison of Synthesis of DAPdR and ACPdR with
Thymidine
NdRT-II and NPases
Fifty m
M
thymidine and 50 m
M
base acceptor were incubated at
Fifty or 100 m
M
donor and 50 m
M
acceptor were incubated for
C or 0.385
W
4
0
9
C with 0.02 units ml NdRT-II with stirring. Results are ex-
16 h with stirring with 0.385 units ml of NdRT-II at 409
W
pressed as the molar yield of glycosyltransfer from thymidine to
-deoxyribosides. DAP, 2,6-diaminopurine; ACP, 2, -amino-6-
units ml of Py-NPase and 0.77 units ml of Pu-NPase II at 50
9C
W
W
2?
in the presence of 50 m
M
potassium-phosphate buŠer (pH 7.0).
chloropurine.
Results are expressed as the concentration of product at equilibri-
um. dThd, thymidine; DAP, 2,6-diaminopurine; DAPdR, 2,6-di-
Molar yield of reaction at
20 h 42 h
aminopurine-2
ACPdR, 2-amino-6-chloropurine-2
nucleoside phosphorylases.
?
-deoxyriboside; ACP, 2-amino-6-chloropurine;
Base
acceptor
?
-deoxyriboside; NPases,
3
h
DAP
ACP
21.2
18.5
z
z
55.9
59.0
z
z
70.0
66.8
z
z
Substrate (m
M
)
Product (mM)
Enzyme
dThd
DAP
ACP
DAPdR
ACPdR
5
5
00
0
0
50
—
50
—
—
50
—
50
39.9
—
48.2
—
—
31.5
—
NdRT-II
yribose donor for synthesis of dAdo and dCyd. The
reaction using dUrd resulted in similar e‹ciencies to
those using dThd (data not shown). However dGuo
could hardly be produced from dUrd and guanine.
1
100
46.4
5
5
00
00
0
0
50
—
50
—
—
50
—
50
39.0
—
46.3
—
—
26.8
—
NPases
1
1
39.1
Enzymatic production of DAPdR and ACPdR
Solubility of substrates is important for enzymatic
reactions. However, guanine is insoluble in water at
neutral pH and the normal temperature for the en-
zyme reaction. On the contrary, 6-substituted gua-
nine derivatives such as 2,6-diaminopurine (DAP)
and 2-amino-6-chloropurine (ACP) are more soluble
than guanine. By the application of nucleoside phos-
rine ribonucleoside, 2
?
3?
-dideoxypurine nucleosides,
4
,8,9,18)
and so on.
We investigated the availability of
bacterial ADA for conversion of DAPdR and
ACPdR to dGuo, because bacterial enzymes can be
more easily prepared than mammalian enzymes.
We tested the hydrolysis reaction of puriˆed
DAPdR and ACPdR with E. coli adenosine
deaminase (ADA) comparing mammalian (calf intes-
tine) ADA. E. coli ADA was prepared from E. coli
cells carrying pDR-add and ADA activity in the cell-
phorylases (NPases) using Enterobactor aerogenes
,
DAPdR was produced from deoxyuridine and DAP,
following a deamination reaction of DAPdR for
7)
producing dGuo. Therefore we next attempted the
transglycosylation reactions with NdRT-II instead of
NPases using 6-substituted purine as a deoxyribose
acceptor instead of guanine, and found that DAPdR
or ACPdR was e‹ciently synthesized from dThd
free extract was found to be 1,698 units ml. ACPdR
W
was puriˆed by using ion-exchange chromatography
from the transglycosylation reaction mixture and
used as a substrate for ADA. Figure 1 shows the time
course of hydrolysis of dAdo, DAPdR, and ACPdR
by the same units of ADA from calf intestine (A) and
E. coli (B). Mammalian ADA could e‹ciently con-
vert both DAPdR and ACPdR into dGuo. The
(
Table 2).
Nucleoside phosphorylases can catalyze not only
the transfer of both ribosyl and deoxyribosyl residues
but also the phosphorolysis of both nucleoside and
deoxyribonucleoside contrary to NdRT-II that only
catalyzes the transfer a deoxyribosyl residue to a base
acceptor. We compared the transglycosylation
e‹ciency of NdRT-II and that of the nucleoside
phosphorylases, Pu-NPase II and Py-NPase
hydrolysis rate of DAPdR and ACPdR (t1W2
and 8.8 min) were about 4.5 and 2.5 times slower
than dAdo (t1 W2 2.7 min), respectively. On the other
＝
17.2
＝
hand, E. coli ADA showed a diŠerent substrate
speciˆcity. Although dAdo was hydrolyzed at a simi-
lar rate with mammalian ADA and bacterial ADA,
DAPdR and ACPdR were hydrolyzed very slowly
(
Table 3). Yields of DAPdR or ACPdR were im-
proved when the molar concentration of dThd was
doubled against that of a base acceptor. When DAP
were used as a base acceptor, NdRT-II or a pair of
Pu-NPase II and Py-NPase equally synthesized
DAPdR. On the other hand, ACPdR was synthesized
more e‹ciently by NdRT-II than by NPases.
(t1 2
＝
206, 216 min) with bacterial ADA. However,
W
the bacterial ADA could hydrolyze both DAPdR and
ACPdR to dGuo. At 20 h of reaction, both deox-
yribonucleosides had been completely hydrolyzed
(
Fig. 1C).
Hydrolysis of DAPdR and ACPdR with mam-
malian and bacterial adenosine deaminase
It is well known that mammalian adenosine
deaminase (ADA) has a broad substrate speciˆcity,
and mammalian ADA has been used for hydrolysis
of adenosine analogues such as DAPdR, 6-chloropu-
Production of dGuo via DAPdR or ACPdR by
combining the reactions of transglycosylation and
hydrolysis
From the results of transglycosylation and ADA-
hydrolysis reactions, dGuo was expected to be syn-

Enzymatic Synthesis of 2?-Deoxyguanosine
993
(7.5 to 8.0) and NdRT-II has little activity at pH over
8.0.
After transglycosylation, NdRT-II in the reaction
mixtures was inactivated by alkaline treatment with
NaOH. The pH was then adjusted to the optimum
pH for adenosine deaminase, 7.8 with Tris-HCl
buŠer and 0.1
N
HCl.
Table 4 shows dGuo production from DAP or
ACP through sequential reactions with NdRT-II and
E. coli ADA. Molar yields against base acceptors and
dThd were over 90
z and over 45z, respectively.
Discussion
In the transglycosylation test, the reaction e‹cien-
cies of NdRT-II diŠer with the kinds and concentra-
tions of substrates. Unlike guanine, 6-substituted
purine bases, DAP and ACP, could be e‹cient sub-
strates for NdRT-II reaction. Although the molar
yields of dGuo synthesis using DAP or ACP were
almost the same, ACP should be suitable for practi-
cal production of dGuo because ACP can be pre-
pared at lower cost than DAP.
Trans-
ATCC 10697) has been reported to have trans-
glycosylation activity but DAP was inactive as a
N-deoxyribosylase of Lactobacillus lactis
(
1
9)
deoxyribosyl acceptor for this enzyme. However,
we show that NdRT-II of L. helveticus could transfer
a deoxyribose from dThd to DAP. Further studies
about the molecular structure of both enzymes will
demonstrate the amino acid residue(s) determining
the deoxyribosyl acceptor speciˆcity.
DAPdR and ACPdR were synthesized with NdRT-
II using equal molar concentrations of deoxyribosyl
donor and acceptor with a molar yield of 79.7
z
and
6
2.9 , respectively. When the concentration of a
z
deoxyribosyl donor and that of a base acceptor were
at the ratio of 2:1, the molar yield against base accep-
tors could be over 90
z under these conditions. The
more dThd increased, the more a molar yield of
ACPdR from ACP was increased. However, ACP
did not completely accepted deoxyribosyl residues
under these conditions even though dThd and ACP
were in the ratio 3:1 (data not shown). Residual dThd
can be recovered from the reaction mixture and
reused as a deoxyribosyl donor for another reaction.
In the transglycosylation reaction of DAPdR or
ACPdR production from dThd by a nucleoside phos-
phorylase, two types of NPases are needed as
follows: phosphorolysis of dThd with Py-NPase and
transferring phosphate group of deoxyribose-1-
phosphate to DAP or ACP with Pu-NPase II.
NPases also catalyze phosphorolysis of residual
dThd and synthesized DAPdR or ACPdR indepen-
dently of transglycosylation. On the other hand,
NdRT-II can catalyze transglycosylation between a
pyrimidine deoxyribonucleoside and a purine base,
and doesn't catalyze cleavage of a glycosidic linkage
Fig. 1. Time Course of Hydrolysis of Deoxyribonucleosides with
Adenosine Deaminase from Calf Intestine and E. coli
A total of 1 ml reaction mixture containing 10 m
yribonucleoside, 100 m Tris-HCl (pH 7.0), and 1.5 units ml of
adenosine deaminase from calf intestine (A) and E. coli (B and
C) were incubated at 37 C. ( ) ACPdR, ( ) 2 -deoxyadeno-
sine, ( ) DAPdR.
.
M
deox-
W
M
9

$
?

thesized via DAPdR or ACPdR using DAP or ACP
in one pot (Fig. 2). However, the transglycosylation
with NdRT-II and hydrolysis with ADA should not
be done at the same time, because guanine was gener-
ated from dGuo by a reverse reaction of NdRT-II in
the presence of thymine. Therefore we attempted to
synthesize dGuo through sequential reactions in one
pot by controlling the pH since the optimum pH of
NdRT-II (6.0 to 7.0) was diŠerent from that of ADA

9
94
K. OKUYAMA et al
.
Fig. 2. Enzymatic Synthesis of Deoxyguanosine Via 2-Amino-6-substituted purine-2?-deoxyriboside by Combining the Transglycosylation
Reaction with NdRT-II and Hydrolysis with Adenosine Deaminase.
Table 4. Synthesis of dGuo from dThd and 6-Substituted Purine
zyme. It is clear that calf intestine ADA is more
e‹cient for synthesizing dGuo from DAPdR and
ACPdR than E. coli ADA. However, E. coli enzyme
can completely convert DAPdR or ACPdR to dGuo
though it takes a long time, and the enzyme can be
easily prepared in large amounts by ampliˆcation of
the gene. Accordingly, the use of E. coli ADA is
available for practical production of dGuo.
In a one-pot synthesis of dGuo consisting of trans-
glycosylation and hydrolysis reactions, it is necessary
to repress the transglycosylation activity in the sec-
ond reaction, because residual activity of NdRT-II
diminishes the yield of dGuo. NdRT-II is suitable for
the sequential reaction system of dGuo synthesis,
because NdRT-II is hardly active at the high pH for
the ADA reaction compared to nucleoside phos-
phorylases. ADA-hydrolysis can be done without
decrease of dGuo even in the presence of thymine by
controlling the pH.
Bases
One hundred m
ml of NdRT-II were incubated at 40
glycosylation, 10 l of 20 NaOH and 100
pH 7.5) were added to 1 ml of transglycosylation mixtures. The
reaction mixtures were adjusted at pH 7.8 and incubated with
0 units of E. coli adenosine deaminase at 40 C for 13 h with stir-
ring. Results are expressed as moles of products in each reaction
mixtures. DAPdR, 2,6-diaminopurine-2 -deoxyriboside; ACPdR,
-amino-6-chloropurine-2 -deoxyriboside; dGuo, 2 -deoxyguano-
sine.
M M
dThd, 50 m base acceptor, and 0.385 units
W
9
C for 16 h. After trans-
m
z
m
l of 1 Tris-HCl
M
(
5
9
m
?
2
?
?
Substrate
Transglycosylation
Hydrolysis
Base
(
m
moles) Product moles) Product
(
m
(
m
moles)
acceptor
DAP
ACP
50
50
DAPdR
ACPdR
48.2
47.3
dGuo
dGuo
47.5
46.2
between a deoxyribonucleoside without an acceptor
base. Actually ACPdR was synthesized more
e‹ciently by NdRT-II than a pair of Pu-NPase II
and Py-NPase in the transglycosylation reactions.
Furthermore, the amount of residual dThd after the
reaction with NdRT-II was 5–20 percent higher than
that with NPases. Therefore we concluded that the
use of NdRT-II is better than that of NPases for
practical production of deoxyribonucleosides.
As described in this paper, we have established the
practical procedures for synthesis of dAdo, dGuo,
and dCyd from dThd and corresponding bases with
NdRT-II. In addition, we have already conˆrmed
that crude enzyme solutions prepared from adeno-
sine deaminase-deˆcient or cytidine deaminase-
deˆcient recombinant E. coli cells are available for
synthesis of three kinds of deoxynucleosides from
dThd. Deoxyribonucleosides will be produced and
supplied by chemical and enzymatic procedures
independent of DNA hydrolysis in the near future.
The substrate speciˆcity of ADA was quite diŠer-
ent between a bacterial enzyme and a mammalian en-

Enzymatic Synthesis of 2?-Deoxyguanosine
995
References
11) Giri, I., Bolon, P. J., and Chu, C. K., Synthesis of
-C-alfa-methyl-2 ,3 -dideoxynucleosides. Nucleo-
sides Nucleotides, 15, 183–204 (1996).
12) Nair, V., and Sells, T. B., Enzymatic synthesis of 2
-dideoxyguanosine. Synlett, 10, 753–754 (1991).
2
?
? ?
1
)
Hanrahan, J. R., and Hutchinson, D. W., The
enzymatic synthesis of antiviral agents. J.
Biotechnol., 23, 193–210 (1992).
?,
3
?
2
)
Bzowska, A., Kulikowska, E., and Shugar, D., Prop-
erties of purine nucleoside phosphorylase (PNP) of
mammalian and bacterial origin. Z. Naturforsch.,
13) Marquez, V. E., Tseng, C. K. H., Mitsuya, H., Aoki,
S., Kelley, J. A., Ford, H., Roth, J. S., Broder, S.,
and Driscoll, J. S., Acid-stable 2?-‰uoro purine
dideoxynucleosides as active agents against HIV. J.
Med. Chem., 33, 978–985 (1990).
4
5c, 59–70 (1990).
Cardinaud, R., Nucleoside deoxyribosyltransferase
from Lactobacillus helveticus Methods Enzymol.,
1, 446–455 (1978).
3
4
)
)
.
14) Debouck, C., Riccio, A., Schumperli, D.,
McKenney, K., JeŠers, J., Hughes, C., Rosenberg,
M., Heusterspreute, M., Brunel, F., and Davison, J.,
Structure of the galactokinase gene of Escherichia
coli, the last (?) gene of the gal operon. Nucleic Acids
Res., 13, 1841–1853 (1985).
15) Chang, Z., Nygaard, P., Chinault, A. C., and
Kellems, R. E., Deduced amino acid sequence of Es-
cherichia coli adenosine deaminase reveals
evolutionarily conserved amino acid residues: impli-
cations for catalytic function. Biochemistry, 30,
2273–2280 (1991).
16) Hamamoto, T., Noguchi, T., and Midorikawa, Y.,
Cloning of purine nucleoside phosphorylase II gene
from Bacillus stearothermophilus TH 6–2 and charac-
terization of its gene product. Biosci. Biotechnol.
Biochem., 61, 276–280 (1997).
17) Okuyama, K., Hamamoto, T., Noguchi, T., and
Midorikawa, Y., Molecular cloning and expression of
the pyrimidine nucleoside phosphorylase gene from
Bacillus stearothermophilus TH 6–2. Biosci.
Biotechnol. Biochem., 60, 1655–1659 (1996).
5
Fischer, W., Vasilouiu, R., and Brendel, M.,
Enzymatic procedure for the synthesis of base
modiˆed
2
?
,
3?-dideoxynucleosides. DECEMA
Biotechnology Conference, 3, 183–187 (1989).
Betbeder, D., Heath, C. M., and Huntchinson, D.
W., Enzymatic synthesis of dideoxyribonucleosides.
Nucleosides Nucleotides, 10, 465–468 (1991).
Okuyama, K., and Noguchi, T., Molecular cloning
and expression of the nucleoside deoxyribosyltran-
5
6
)
)
sferase-II gene from Lactobacillus helveticus
Biotechnol. Biochem., 64, 2243–2245 (2000).
Yokozeki, K., and Tsuji, T., A novel enzymatic
method for the production of purine-2 -deox-
yribonucleosides. J. Mol. Cat. B: Enzymatic, 10,
07–213 (2000).
. Biosci.
7
8
9
)
)
)
?
2
Krenitsky, T. A., Koszalka, G. W., and Tuttle, J. V.,
Purine nucleoside synthesis, an e‹cient method
employing nucleoside phosphorylases. Biochemistry
0, 3615–3621 (1981).
,
2
Shirasaka, T., Murakami, K., Ford, Jr., H., Kelley,
J. A., Yoshioka, H., Kojima, E., Aoki, S., Broder,
S., and Mitsuya, H., Lipophilic halogenated con-
18) Tritsch, G. L., Paolini, N. S., and Bielat, K., Adeno-
sine deaminase activity associated with phagocytic
vacuoles. Cytochemical demonstration by electron
microscopy. Histochem., 82, 281–285 (1985).
geners of 2?, 3?-dideoxypurine nucleosides active
against human immunodeˆciency virus in vitro
Natl. Acad. Sci. USA, 87, 9426–9430 (1990).
. Proc.
19) Beck, W. S., and Levin, M., Puriˆcation, kinetics,
1
0) Shealy, Y. F., and O'Dell, C. A., Carbocyclic analogs
of purine 2 -deoxyribofuranosides. U. S. Patent
543255 (Sep. 24, 1985).
and repression control of bacterial trans-N-deox-
?
yribosylase. J. Biol. Chem., 28, 702–709 (1962).
4