Importance of 3′-hydroxyl group of the nucleosides for the reactivity of thymidine phosphorylase from Escherichia coli

Article abstract of DOI:10.1246/cl.2006.232

Thymidine phosphorylase in phosphate buffer catalyzed the conversion of thymidine to unnatural nucleosides. The 3′-OH, but not the 5′-OH of ribosyl moiety is necessary to be recognized as a substrate. Thus 3′-deoxythymidine could not convert to 5-fluorouracil-2′,3′- dideoxyribose. However, 5′-deoxythymidine was converted to 5-fluorouracil-2′,5′-dideoxyribose. Copyright

Full text of DOI:10.1246/cl.2006.232

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
232
Chemistry Letters Vol.35, No.2 (2006)
Importance of 30-Hydroxyl Group of the Nucleosides for the Reactivity
of Thymidine Phosphorylase from Escherichia coli
Akihiko Hatano,ꢀ Aiko Harano, and Masayuki Kirihara
Department of Materials and Life Science, Shizuoka Institute of Science and Technology,
2200-2 Toyosawa, Fukuroi 437-8555
(Received November 29, 2005; CL-051479; E-mail: a-hatano@ms.sist.ac.jp)
Thymidine phosphorylase in phosphate buffer catalyzed the
conversion of thymidine to unnatural nucleosides. The 30-OH,
but not the 50-OH of ribosyl moiety is necessary to be recognized
as a substrate. Thus 30-deoxythymidine could not convert
to 5-fluorouracil-20,30-dideoxyribose. However, 50-deoxythymi-
dine was converted to 5-fluorouracil-20,50-dideoxyribose.
Unnatural nucleosides are of interest as antiviral and antitu-
mor agents,1 expansion of the genetic alphabet,2–5 and functional
building block molecules of DNA.6,7 These molecules have been
synthesized by a stereoselective coupling reaction of the base
analogue and the protected ribose. These studies have a weak
point because the coupling reactions of the base analogue and
ribosyl moiety with Lewis acid produce mixtures of the 10ꢀ
and ꢁ unnatural nucleoside, and the yields are lowered.
Figure 1. Effect of variation of substrate concentration on the
conversion with thymidine phosphorylase. (a): Effect of changes
in thymidine concentration. 5FU concentration was kept
constant at 5 mM. (b): Effect of changes in 5FU concentration.
Thymidine concentration was kept constant at 5 mM. All
reactions were performed with 5 units of TP in 1 mM phosphate
buffer (pH 7.0) at 35 ꢁC.
Nucleoside phosphorylases8–12 are intracellular metabolic
enzymes that convert each of the corresponding four nucleosides
to 2-deoxy-D-ribose-1ꢀ-phosphate. Thymidine phosphorylase
(TP) catalyzed conversion of thymidine to thymine and 2-de-
oxy-D-ribose-1ꢀ-phosphate,13,14 and it is known that TP is asso-
ciated with tumor-dependent angiogenesis.15–17 We expected
that this enzyme could perform the transglycosylation of thymi-
dine, because the enzymatic reaction was reversible. It is impor-
tant to investigate the substrate specificities of this enzyme to as-
sess if the process is a suitable synthetic method for development
of unnatural nucleosides. Recently, the crystal structure of hu-
man TP was reported in a complex with uracil derivative.18 This
compound did not have a ribosyl moiety, and the ribose binding
pocket was not defined since the iminopyrrolidine moiety had no
OH groups. Schramm suggested a transition state of this enzyme
that was a near-symmetric SN2 nucleophilic displacement of
thymine by inorganic phosphate.19 They proposed that the 50-
OH of the ribosyl moiety may bind to the active-site residues
by hydrogen bonding. In this communication, we studied the
reactivities of thymidine phosphorylase (Thymidine: orthophos-
phate deoxyribosyltransferase, from Escherichia coli, EC
2.4.2.4), the recognized functional groups of the base moiety
and the hydroxyl groups of the ribosyl moiety.
in two steps (Scheme 1, step 1 and step 2) and these are reversi-
ble. Therefore, the produced thymine and the concentration of
phosphate buffer inhibit step 2 under ‘‘one-pot’’ reaction condi-
tions. Figure 1 shows the effects of concentration of substrates
(thymidine (1), 5-fluorouracil (5FU, 4)).20 The conversion and
initial velocity of the produced 5-fluorouracil-20-deoxyribose
(5) increased with increasing concentration of thymidine (the
concentration of 5FU was constant at 5 mM). However, at a
10:1 ratio or higher of thymidine to 5FU the reaction leveled
off. Figure 1b shows the effect of the 5FU concentration. As
5FU increased, the initial velocity decreased, and the conversion
increased. The increasing concentration of 5FU shifted the
Table 1. Effect of the base moiety on the thymidine phospho-
rylase reaction
Base analogue B
6
4
7
8
R
H
F
Cl
Br
I
R1
O
O
O
O
O
O
O
O
O
O
O
S
R2
O
O
O
O
O
O
O
O
O
S
Time/h
2
2
2
2
2
2
2
2
2
2
2
24
Conv./%
First, we looked for the most effective reaction conditions
for each concentration of substrates (thymidine, base analogue,
and phosphate ion). The reaction catalyzed by TP takes place
95
86
87
97
88
67
84
47
9
10
11
12
13
14
15
16
Et
NH2
CF3
NO2
H
CH3
H
no reaction
54
S
S
61
no reaction
Scheme 1. Catalytic reaction of thymidine phosphorylase.
Copyright Ó 2006 The Chemical Society of Japan

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Chemistry Letters Vol.35, No.2 (2006)
233
possess the 30-OH, did not react with 5FU to form 5-fluoroura-
cil-20,30-dideoxyribose under the same conditions as above.
Compounds 20 and 21, in which the 30-OH was replaced with
SH and NH2, respectively, showed a low conversion, and also
5FU did not react with 30ꢁ-OH-thymidine 19. These substrates
most likely did not form hydrogen bonds with amino acids in
the binding pocket of TP. Therefore, it can be concluded that
the 30ꢀ-hydroxyl group is very important for the recognition
of the ribosyl substrate. Whereas the ribosyl 20-OH retarded
the catalytic reaction by TP, 5-fluorouridine was obtained in
82% yield from uridine 22.
Table 2. Effect of the hydroxyl group of the ribosyl moiety on
the thymidine phosphorylase reaction
Substrate
R3
R4
R5
Time/h
Conv./%
1
17
18
19
20
21
22
H
H
H
H
H
OH
OH
H
OH(ꢀ)
SH
NH2
OH
OH
H
2
3
86
85
<1
In conclusion, thymidine phosophorylase effectively cata-
lyzed the reaction replacing the nucleobase of thymidine. We
conclude that this enzyme recognized the 30-OH of thymidine
to fix the substrate.
OH
OH
OH
OH
OH
24
24
24
24
24
no reaction
<1
<1
H
OH
82
This work was supported partly by the Astellas Foundation
for Research on Medicinal Resources.
equilibrium and inhibited the production of 5-fluorouracil-20-
deoxyribose. The increasing concentration of phosphate ions
inhibited the catalytic 5-fluorouracil conversion of thymidine
to compound 5. We decided to use the 10:1 ratio of thymidine
(50 mM) to 5FU (5 mM) and 1 mM phosphate buffer (pH 7.0).21
Table 1 shows the substrate specificities of the base moiety
in the TP-catalyzed reaction. Each of the 5-substituted uracil
compounds (halogen: 4 and 7–9, ethyl: 10, amino: 11, and tri-
fluoromethyl: 12) reacted with thymidine to convert the corre-
sponding nucleoside in good yield. However, 5-nitrouracil 13
did not react with thymidine. 2-S-Substituted uracil (2-thiouracil
14 and 2-thiothymine 15) in which oxygen at the 2 position of
the base was replaced with sulfur reacted with thymidine to give
54 and 61%, respectively, of the corresponding nucleoside. Gago
et al. predicted by molecular dynamic simulations and quantum
chemical calculation that replacing the oxygen at position 2 of
the pyrimidine base with sulfur should accelerate the reaction
rate because of activation of the 2 position of uridine by
His-85 of E. coli TP.22 These authors studied the cleavage of
the glycosidic bond of thymidine and 2-thiothymidine to base
and 2-deoxy-D-ribose-1ꢀ-phosphate. We could not observe the
acceleration by 2-thiothymine of the transfer reaction by TP.
2,4-Dithiouracil 16 did not dissolve in phosphate buffer, and
the reaction did not occur.
References and Notes
1
2
3
4
5
6
R. Garg, S. Gupta, H. Gao, M. Babu, A. Debnath, C. Hansch, Chem.
Rev. 1999, 99, 3523.
M. J. Lutz, H. A. Held, M. Hottiger, S. A. Hubscher, Nucleic Acids
Res. 1996, 24, 1308.
¨
D. L. McMinn, A. K. Ogawa, Y. Wu, J. Liu, P. G. Schultz, F. E.
Romesberg, J. Am. Chem. Soc. 1999, 121, 11585.
T. Mitsui, A. Kitamura, M. Kimoto, T. To, A. Sato, I. Hirao, S.
Yokoyama, J. Am. Chem. Soc. 2003, 125, 5298.
H. Liu, J. Gao, S. R. Lynch, L. Y. Maynard, D. Saito, E. T. Kool,
Science 2003, 302, 868.
K. Tanaka, A. Tengeiji, T. Kato, N. Toyama, M. Shiro, M. Shionoya,
Science 2003, 299, 1212.
E. T. Kool, Chem. Rev. 1997, 97, 1473.
7
8
9
J. Holguin, R. Cardinaud, Eur. J. Biochem. 1975, 54, 505.
M.-C. Hung, K. Hatfield, A. W. Roetker, J. A. Montgomery, R. L.
Blakley, Biochem. Pharmacol. 1981, 30, 2663.
10 D. Betbeder, D. W. Hutchinson, A. O. Richards, Nucleic Acids Res.
1989, 17, 4217.
11 H. Komatsu, T. Araki, Tetrahedron Lett. 2003, 44, 2899.
12 Y. Avraham, N. Grossowicz, J. Yashphe, Biochim. Biophys. Acta 1990,
1040, 287.
13 M. Schwartz, Methods Enzymol. 1978, 51, 442.
14 T. Utagawa, J. Mol. Catal. B: Enzym. 1999, 6, 215.
15 S. Matsushita, T. Nitanda, T. Furukawa, T. Sumizawa, A. Tani, K.
Nishimoto, S. Akiba, K. Miyadera, S. Fukushima, Y. Yamada, Cancer
Res. 1999, 59, 1911.
16 A. Moghaddam, H. T. Zhang, T. P. Fan, D. E. Hu, V. C. Lees, H.
Turley, S. B. Fox, K. C. Gatter, A. L. Harris, Proc. Natl. Acad. Sci.
U.S.A. 1995, 92, 998.
17 K. A. Hotchkiss, A. W. Ashton, R. S. Klein, M. L. Lenzi, G. H. Shu,
E. L. Schwartz, Cancer Res. 2003, 63, 527.
18 R. A. Norman, S. T. Barry, M. Bate, J. Breed, J. G. Colls, R. J. Ernill,
R. W. A. Luke, C. A. Minshull, M. S. B. MacMiken, E. J. McCall,
H. H. J. McMiken, D. S. Paterson, D. Timms, J. T. Tucker, R. A.
Pauptit, Structure 2003, 12, 75.
19 a) M. R. Brick, V. L. Schramm, J. Am. Chem. Soc. 2004, 126, 2447.
b) M. R. Brick, V. L. Schramm, J. Am. Chem. Soc. 2004, 126, 6882.
20 Typical procedure for reaction: To 10 mL of 0.1 M phosphate buffer
(pH 7.0), were added thymidine (0.5 mmol, 121 mg), 5-fluorouracil
(0.05 mmol, 6.5 mg), and thymidine phosphorylase (SIGMA, 5 units,
from Escherichia coli, EC 2.4.2.4). The reaction mixture was stirred
at 35 ꢁC for 2 h. The reaction was monitored by HPLC.
21 Each conversion was assayed with a C-18 column (250–4.6 mm)
HPLC at a flow rate of 0.5 mL minꢂ1. The mobile phase was 0.10
mM phosphate buffer (pH 6.8). The UV detector was set at 260 nm
and the column was operated at 35 ꢁC.
However, not all pyrimidine base analogues were converted
to the corresponded nucleosides. For example, the aza com-
pounds, whose C-6 was replaced by a nitrogen atom, did not pro-
duce the nucleosides having a base analogue of an aza compound
(azauracil and azathymine). Moreover, the compounds in which
C-6 was substituted by methyl or keto groups could not be sub-
strates (5,6-dimethyluracil, 5,6-dihydrouracil, and barbituric
acid). We suggest that the C-6 position of the base substrate ster-
ically hindered the nucleophilic attack to form an SN2 transition
state.
Table 2 shows the effect of the hydroxyl group of the ribosyl
moiety on the catalytic reaction. To release the 50-hydroxyl
group, 50-deoxythymidine 17 was reacted with 5-fluorouracil
for 3 h to form 50-deoxy-5-fluorouridine in a yield of 85%. This
enzymatic reaction was nearly as effective as that of thymidine.
Apparently, the 50-OH group is not required to recognize the
substrate for TP. In contrast to the 50-OH reaction, the effect
of 30-OH was drastic. 30-Deoxythymidine 18, which does not
´
22 J. Mendieta, S. Martın-Santamarıa, E.-M. Priego, J. Balzarini, M.-J.
Camarasa, M.-J. Perez-Perez, F. Gago, Biochemistry 2004, 43, 405.
´
´
´
Published on the web (Advance View) January 21, 2006; DOI 10.1246/cl.2006.232