Phosphorylation of isocarbostyril- and difluorophenyl-nucleoside thymidine mimics by the human deoxynucleoside kinases

Article abstract of DOI:10.1081/NCN-200040634

The thymidine mimics isocarbostyril nucleosides and difluorophenyl nucleosides were tested as deoxynucleoside kinase substrates using recombinant human cytosolic thymidine kinase (TK1) and deoxycytidine kinase (dCK), and mitochondrial thymidine kinase (TK2) and deoxyguanosine kinase (dGK). The isocarbostyril nucleoside compound 1-(2-deoxy-β-D-ribofuranosyl)- isocarbostyril (EN1) was a poor substrate with all the enzymes. The phosphorylation rates of EN1 with TK1 and TK2 were <1% relative to Thd, where as the phosphorylation rates for EN1 were 1.4% and 1.1% with dCK and dGK relative to dCyd and dGuo, respectively. The analogue 1-(2-deoxy-β-D- ribofuranosyl)-7-iodoisocarbostyril (EN2) showed poor relative-phosphorylation efficiencies (k_cat/K_m) with both TK1 and dGK, but not with TK2. The k_cat/K_m value for EN2 with TK2 was 12.6% relative to that for Thd. Of the difluorophenyl nucleosides, 5-(1′-(2′-deoxy-β-D-ribofuranosyl))-2,4-difluorotoluene (JW1) and 1-(1′-(2′-deoxy-β-D-ribofuranosyl))-2,4-difluoro-5- iodobenzene (JW2) were substrates for TK1 with phosphorylation efficiencies of about 5% relative to that for Thd. Both analogues were considerably more efficient substrates for TK2, with k_cat/K_m values of 45% relative to that for Thd. 2,5-Difluoro-4-[1-(2-deoxy-β-L-ribofuranosyl)]- aniline (JW5), a L-nucleoside mimic, was phosphorylated up to 15% as efficiently as deoxycytidine by dCK. These data provide a possible explanation for the previously reported lack of cytotoxicity of the isocarbostyril- and difluorophenyl nucleosides, but potential mitochondrial effects of EN2, JW1 and JW2 should be further investigated.

Full text of DOI:10.1081/NCN-200040634

This article was downloaded by: [New York University]
On: 29 November 2013, At: 04:34
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
Nucleosides, Nucleotides and Nucleic Acids
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/lncn20
Phosphorylation of Isocarbostyril‐ and
Difluorophenyl‐Nucleoside Thymidine Mimics by the
Human Deoxynucleoside Kinases
a
a
b
b
Ashraf Said Al‐Madhoun , Staffan Eriksson , Zhi‐Xian Wang , Ebrahim Naimi , Edward E.
b
b
Knaus & Leonard I. Wiebe
a
Department of Cardiology, Vascular Biology Laboratory , University of Ottawa Heart
Institute , Ottawa, KIY 4W7, Canada
b
Faculty of Pharmacy and Pharmaceutical Sciences , University of Alberta , Edmonton,
Alberta, Canada
Published online: 17 Aug 2006.
To cite this article: Ashraf Said Al‐Madhoun , Staffan Eriksson , Zhi‐Xian Wang , Ebrahim Naimi , Edward E. Knaus &
Leonard I. Wiebe (2004) Phosphorylation of Isocarbostyril‐ and Difluorophenyl‐Nucleoside Thymidine Mimics by the Human
Deoxynucleoside Kinases, Nucleosides, Nucleotides and Nucleic Acids, 23:12, 1865-1874, DOI: 10.1081/NCN-200040634
To link to this article: http://dx.doi.org/10.1081/NCN-200040634
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

NUCLEOSIDES, NUCLEOTIDES & NUCLEIC ACIDS
Vol. 23, No. 12, pp. 1865–1874, 2004
Phosphorylation of Isocarbostyril- and
Difluorophenyl-Nucleoside Thymidine
Mimics by the Human
Deoxynucleoside Kinases
1
Ashraf Said Al-Madhoun, Staffan Eriksson, Zhi-Xian Wang,
1,*
2
2
Ebrahim Naimi, Edward E. Knaus, and Leonard I. Wiebe
2
2
1
Division of Cardiology, Vascular Biology Laboratory, University of Ottawa
Heart Institute, Ottawa, Canada
2
Faculty of Pharmacy and Pharmaceutical Sciences, University of Alberta,
Edmonton, Alberta, Canada
ABSTRACT
The thymidine mimics isocarbostyril nucleosides and difluorophenyl nucleosides
were tested as deoxynucleoside kinase substrates using recombinant human cytosolic
thymidine kinase (TK1) and deoxycytidine kinase (dCK), and mitochondrial thy-
midine kinase (TK2) and deoxyguanosine kinase (dGK). The isocarbostyril nucleoside
compound 1-(2-deoxy-b-D-ribofuranosyl)-isocarbostyril (EN1) was a poor substrate
with all the enzymes. The phosphorylation rates of EN1 with TK1 and TK2 were
<
1% relative to Thd, where as the phosphorylation rates for EN1 were 1.4% and
.1% with dCK and dGK relative to dCyd and dGuo, respectively. The analogue
1
1
-(2-deoxy-b-D-ribofuranosyl)-7-iodoisocarbostyril (EN2) showed poor relative-
phosphorylation efficiencies (kcat/K ) with both TK1 and dGK, but not with TK2.
The kcat/Km value for EN2 with TK2 was 12.6% relative to that for Thd. Of the
m
*
Correspondence: Staffan Eriksson, Division of Cardiology, Vascular Biology Laboratory,
University of Ottawa Heart Institute, Ottawa, KIY 4W7, Canada; E-mail: Staffan.Eriksson@
vmk.slu.se.
1
865
DOI: 10.1081/NCN-200040634
1525-7770 (Print); 1532-2335 (Online)
www.dekker.com
Copyright D 2004 by Marcel Dekker, Inc.

1866
Al-Madhoun et al.
difluorophenyl nucleosides, 5-(1’-(2’-deoxy-b-D-ribofuranosyl))-2,4-difluorotoluene
JW1) and 1-(1’-(2’-deoxy-b-D-ribofuranosyl))-2,4-difluoro-5-iodobenzene (JW2)
(
were substrates for TK1 with phosphorylation efficiencies of about 5% relative to
that for Thd. Both analogues were considerably more efficient substrates for TK2,
with kcat/K values of 45% relative to that for Thd. 2,5-Difluoro-4-[1-(2-deoxy-b-L-
m
ribofuranosyl)]-aniline (JW5), a L-nucleoside mimic, was phosphorylated up to 15%
as efficiently as deoxycytidine by dCK. These data provide a possible explanation
for the previously reported lack of cytotoxicity of the isocarbostyril- and difluoro-
phenyl nucleosides, but potential mitochondrial effects of EN2, JW1 and JW2
should be further investigated.
Key Words: Deoxynucleoside kinases; Thymidine kinase; Deoxycytidine kinase;
Thymidine analogues.
INTRODUCTION
Deoxyribonucleoside kinases catalyze the rate limiting phosphorylation of deoxy-
nucleosides into their corresponding monophosphate forms, utilizing ribonucleoside
triphosphates as phosphor donors. Four deoxyribonucleoside kinases have been char-
acterized in mammalian cells, two of which, thymidine kinase (TK1) and deoxycytidine
kinase (dCK), are found in the cytosol, whereas thymidine kinase (TK2) and deoxy-
[1–4]
guanosine kinase (dGK) are predominantly localized in the mitochondria.
and TK2 are pyrimidine specific, phosphorylating thymidine (Thd) and deoxyuridine
dUrd) to their monophosphate derivatives. In addition, TK2 uses deoxycytidine (dCyd) as
Both TK1
(
a substrate. dCK has a broad substrate specificity, phosphorylating dCyd, dGuo and dAdo,
whereas, the mitochondrial dGK is purine-specific, and phosphorylates dGuo and dAdo
(for review, see Ref. [1]).
Deoxynucleoside kinases play a crucial role in chemotherapeutic treatment of cancer
and viral infections. These enzymes catalyze the rate-limiting phosphorylation of the
nucleoside-analogue pro-drugs into their cytotoxic phosphorylated forms. Interestingly,
elevated levels of deoxynucleoside kinases are detected in proliferating cells such as
cancer cells. These properties recruited great interest in the synthesis and biological
[5]
evaluation of nucleoside analogues as possible candidate for cancer and viral therapy.
Recently, the syntheses of several unnatural pyrimidine analogues as potential
anticancer and antiviral therapeutic agents have been reported (Fig. 1). A group of
compounds (EN1 and EN2) containing the isocarbostyril moiety (and its 7-I-derivative)
[6]
as a replacement for the thymine base were designed. Earlier, McMinn et al. reported
that the triphosphate derivative of 7-propynylisocarbostyril nucleoside mimics is a
[7]
template for replication by the Klenow fragment (KF) of E. coli DNA polymerase. In
synthetic DNA, this compound is efficiently forms a stable self-pair and the unnatural
self-pairing resulted in an insufficient polymerase chain elongation, which could lead to
[7]
cytotoxic effects in rapidly proliferating cancer cells.
A second group of nonpolar hydrophobic isosteres of pyrimidine nucleoside
[8]
[9]
mimics, based on the pioneering study of Moran et al. has also been synthesized.
These compounds were characterized by a 2,4-difluorotoluene moiety as an analogue to
thymine in Thd, where F is the isosteric replacement for O, and C–H groups replaces

Thymidine Mimics by Human Deoxynucleoside Kinases
1867
Figure 1. Chemical structures of thymidine and the nucleoside mimics EN1, EN2 and JW1–JW5.
N–H groups (compound JW1, [9]). In 1-(2-deoxy-b-D-ribofuranosyl)-2,4-difluoro-5-
iodobenzene (JW2, 9), the CH group was replaced with I, analogous to iodo-dUrd. The
3
2
,4-difluorophenyl-5-iodo moiety was also attached to an acyclic 2’-deoxyribose ring
similar to that found in ganciclovir (GCV) (JW3, [10]). Based on the same principle,
-fluoro-4-[1-(2-deoxy-b-D-ribofuranosyl)] aniline (JW4) and 2,5-difluoro-4-[1-(2-
deoxy-b-L-ribofuranosyl)]-aniline (JW5) were synthesized as dCyd mimics. In
3
[
10]
JW4,
whereas JW5 was designed to mimic L-dCyd.
these unusual pyrimidine mimics relate to the work of Kool and his collaborators,
who found that the triphosphate derivative of the 2,4-difluoro-5-methylphenyl isostere
dFTP) is a substrate for DNA polymerases, e.g. Klenow fragment. Steady-state
the cytosine moiety was replaced by a 4-(3-fluoroaniline) ring system,
[11]
The rationale behind the synthesis of
[8,12]
(
measurements indicated that dFTP base pair with dATP at high efficiency and fidelity.
dFTP is a non polar molecule that forms a weak hydrogen bonds with dATP sufficient
[13]
to destabilize DNA duplexes.
The 2,4-difluoro-5-methylphenyl isostere and its
derivatives could also be promising candidates for cancer and viral chemotherapy
agents. Pharmacokinetic and metabolism studies of JW2 in rats present a picture of
high bioavailability from oral doses, dose-dependent pharmacokinetics, hepatic
[14–16]
conversion to glucuronide and sulphate conjugates and metabolic deiodination.
We now report studies designed to provide information on the mechanisms of
action of these unnatural nucleoside analogues. The objectives of this study were to
provide insights into the failure of these compounds to exhibit any biological activity,
and to develop new ideas to design mimics, which might serve as antitumor and
antiviral drugs.

1868
Al-Madhoun et al.
EXPERIMENTAL METHODS
Materials
[6]
Nucleoside mimics, the isocarbostyril deoxynucleosides EN1 and EN2,
difluorophenyl deoxynucleosides JW1, JW2, JW3, JW4 and JW5, were synthesized
and
[
9–11,17]
using literature methods.
ST. Louis, MO, USA.
Natural nucleosides were from Sigma–Aldrich Corp.,
Expression and Purification of the Recombinant Human Deoxynucleoside Kinases:
The recombinant enzymes TK1, TK2, dCK and dGK were expressed and purified from
[2–4]
bacterial expression systems according to procedures described previously.
enzymes were at least 95% pure as judged by SDS-PAGE.
The
Phosphoryl-Transfer Assay with Recombinant Deoxynucleoside Kinases: Natural
substrates (Thd, dCyd, dGuo and dUrd) and the nucleoside analogues were dissolved in
DMSO to produce stock solutions of various concentrations (100–200 mM). The
[
18]
assays were carried out as described previously
reaction mixture contained 100 mM nucleoside, 100 mM ATP, 0.03 mM [gÀ P] ATP
Amersham Pharmacia Biotech, IL, USA), 50 mM Tris-HCl, pH 7.6, 5 mM MgCl , 125
with minor modifications. The
3
2
(
2
mM KCl, 10 mM DTT and 0.5 mg/ml bovine serum albumin (BSA). The reaction
mixture was incubated at 37°C for 20 min in the presence of 50 ng of enzyme.
Following the incubation period, the enzyme was heat inactivated for 2 min at 95°C.
The reaction mixture was centrifuged and a 1-ml portion was spotted on PEI-cellulose
TLC plates (Merck). These were placed overnight in a solvent system containing
isobutyric acid: ammonium hydroxide: water (66:1:33). The radiolabeled spots were
visualized by a phospho-imager (Fuji Film, Science Lab., Image Gauge V3.3).
Kinetic Analysis
The kinetic parameters were determined by non-linear regression analysis using the
Data
[19]
Michaelis-Menten equation at a substrate concentrations range of 1–100 mM.
were analyzed by the Sigma Plot Enzyme Kinetic Module version 1.1 (SPSS Inc.).
RESULTS
Specificity of dN Kinases with Nucleoside Analogues
Thymidine and deoxycytidine analogues were evaluated at a fixed concentration
(
100 mM) with recombinant human deoxynucleoside kinases highly purified from
[2–4]
bacterial expression systems.
Phosphoryl transfer assays indicated that substitution
of the thymine moiety by an unnatural isocarbostyril group (EN1, EN2; Fig. 1) resulted
in new properties that are not favored by the human deoxynucleoside kinases. The
recombinant human thymidine kinases (TK1 and TK2) did not accept EN1 as a
substrate, but, EN2, the 7-I-derivative of EN1, showed some capacity to undergo
phosphorylation compared to Thd as a reference (Table 1). The relative phosphor-
ylations of EN2 were 1.8 and 6.9% with TK1 and TK2, respectively (Table 1).

Thymidine Mimics by Human Deoxynucleoside Kinases
1869
Table 1. Phosphorylation of natural and mimetic nucleosides by recombinant human
nucleoside kinases.
Enzyme
Compound
TK1
<0.1
TK2
<0.1
6.9 ± 0.7
dCK
dGK
EN1
EN2
JW1
JW2
JW3
JW4
JW5
dThd
dUrd
dCyd
dGuo
1.4 ± 0.3
<0.1
<0.1
<0.1
<0.1
<0.1
14.6 ± 0.8
–
1.1 ± 0.4
4.4 ± 0.2
<0.1
<0.1
<0.1
<0.1
<0.1
–
1.8 ± 0.8
6.0 ± 0.9
9.9 ± 1.3
<0.1
90.4 ± 1.4
73.6 ± 0.7
5.1 ± 0.2
<0.1
<0.1
<0.1
<0.1
100
–
100
77
–
100
–
–
–
–
100
–
–
–
Phosphoryl transfer assay was performed as described in Experimental Methods, the
concentrations of the compounds were 100 mM and the phosphate donor ATP was 100 mM.
The values and the standard deviations are the percent phosphorylation relative to the natural
substrate for each enzyme.
Although the isocarbostyril group mimics, to some extent, the pyrimidine ring, EN1
was poorly phosphorylated by dCK and dGK. The 7-I-substitution (EN2) was a
marginally better dGK substrate, with a relative phosphorylation of 4.4% compared to
dGuo as a reference (Table 1).
The 2,4-difluorophenyl isosteres of pyrimidine nucleosides (JW1–JW3) showed
variable substrate specificity with the human recombinant kinases with respect to their
configuration, the nature of the sugar ring and the sub-group substitutions within the
moiety (Fig. 1). These 2,4-difluorophenyl pseudo-bases thymidine mimics were also
not phosphorylated by the recombinant kinases dCK and dGK. On the other hand, JW1
and JW2 were substrates for the recombinant human thymidine kinases. TK1 formed
low but significant levels of phosphorylated products with JW1 and JW2 (6 and 9.9%,
respectively, relative to Thd). Interestingly, JW1 and JW2 were phosphorylated by
recombinant TK2 at relatively high rates (90.4 and 73.6%, respectively) in relation to
Thd as a reference. No activity (less than 0.1%) was observed with JW3, the acyclic-
sugar derivative of JW2, using recombinant TK1. Low levels of phosphorylation
products with recombinant TK2 were detected for JW3, but with roughly 14-fold lower
compared to JW2 (5.1% with respect to Thd as a reference).
The unnatural deoxycytidine mimic analogue, 2-fluoro-4-aminophenyl (JW4) did
not show any detectable phosphorylation with any of the four human deoxynucleoside
kinases studied, whereas the b-L-deoxycytidine mimic (JW5) was significantly
phosphorylated by recombinant dCK (14.6% relative to dCyd). This observation is in
agreement with the previously reported stereo-specificity of dCK, which shows that
dCK have a higher affinity for b-L-isomers than for the b-D-isomers, although the
[21]
catalytic activity is higher for substrates in the natural configuration.

1870
Al-Madhoun et al.
Kinetic Properties of the Analogues
The biochemical analysis of the interactions of deoxynucleoside kinases with these
Thd mimics was extended to include kinetics parameters, which gives a better
understanding of the nucleoside/enzyme interaction. The kinetic parameters with
recombinant TK1 and TK2 were determined using the phosphoryl transfer assay, in
which the formation of radioactive monophosphate products was monitored using
3
2
[
gÀ P]-ATP as a phosphate donor.
The apparent Km values of the isocarbostyril nucleoside compound 1-(2-deoxy-b-
D-ribofuranosyl)-7-iodoisocarbostyril (EN2) with TK1 and TK2 was 38- and 4- fold
higher than Thd, respectively. Thus, the calculated relative phosphorylation efficiencies
were 0.8% and 12.6% relative to Thd using both TK1 and TK2, respectively. When
dGK was used, EN2 showed 9 fold higher apparent Km and 10 fold lower kcat values
relative to dGuo, as a result the k /K relative efficiency was 100 fold lower than that
cat
m
of dGuo.
The difference in the capacity of TK1 and TK2 to use JW1 and JW2, as substrates
was apparent when the Km and kcat values were determined. The Km values for JW1
and JW2 with TK1 were 2- and 6- fold higher than those with TK2 (Table 2). The
relative efficiency, k /K , for JW1 and JW2 with TK1 were 3.2% and 4.5%, relative
cat
m
to that for Thd. In the case of TK2, JW1 had 3.5 fold higher apparent K and two fold
m
higher kcat values relative to that for Thd, resulting in an over all efficiency of 44.6%
relative to Thd. JW2 showed a two-fold higher apparent K value and no differences in
m
the kcat value compared to that for Thd; i.e. the k /K value was 44.1% of that for Thd
cat
m
(Table 2).
Table 2. Michaelis–Menten kinetic parameters for phosphorylation of nucleoside mimics by the
recombinant human deoxynucleoside kinases.
kcat/K Relative to
m
natural nucleoside
À1
Nucleoside
Enzyme
K
m
(mM)
k
cat (s
)
dThd
EN2
JW1
JW2
dThd
EN2
JW1
JW2
dGuo
EN2
dCyd
JW5
TK1
TK1
TK1
TK1
TK2
TK2
TK2
TK2
dGK
dGK
dCK
dCK
2.3 ± 0.5
88.8 ± 27
147.4 ± 47
259.9 ± 75
18.2 ± 0.8
74.4 ± 4.6
69.9 ± 6.3
41.1 ± 4.1
2.0 ± 0.71
18.8 ± 3.3
1.0 ± 0.3
3.5 ± 0.3
1.2 ± 0.2
7.7 ± 1.3
100
0.8 ± 0.1
3.2 ± 0.5
4.5 ± 0.7
100
19.0 ± 3.2
0.31 ± 0.03
0.16 ± 0.08
0.53 ± 0.02
0.31 ± 0.01
0.07 ± 0.01
0.007 ± 0.005
0.11 ± 0.02
0.02 ± 0.01
12.6 ± 2.4
44.6 ± 1.0
44.1 ± 0.4
100
1.0 ± 0.04
100
13.1 ± 1
1.3 ± 0.6
The parameters were calculated using Michaelis–Menten equation. The concentration of the
compounds was varied from 1 to 100 mM. The kcat values were calculated using the equation
V
max = kcatÂ[E], where [E] is the total enzyme concentration and is based on one active
site/monomer.

Thymidine Mimics by Human Deoxynucleoside Kinases
1871
The kinetic parameters for JW5, determined over a 1–100 mM concentration range,
showed an apparent K value of 1.3 mM, similar to that for dCyd, whereas the apparent
m
kcat was very low, about 5.5 fold lower than that for dCyd. The phosphorylation ef-
ficiency (k /K ) was 13.1% relative to that for dCyd.
cat
m
DISCUSSION
A group of unnatural pyrimidine analogues (Fig. 1) with diverse structures have
been evaluated for their potential as substrates for the human deoxynucleoside kinases
in phosphoryl transfer assays using recombinant enzyme preparations. The biochemical
properties of this group of compounds were determined to understand their substrate
activity relationships and to evaluate such biochemical approaches to identify and
design new anticancer drugs.
Earlier, these novel classes of isocarbostyril-thymine substitutes (EN-1 and EN-2)
and 2,4-difluorophenyl pseudo-bases exhibited weak cytotoxic effects.
[6,9–11,17]
The
concentration of these compounds that killed 50% of several different cancer cell lines
À3 À5
CC ) were in the range of 10 to 10 M, which is high as compared to cytotoxic
50
(
À11
À12
[6]
nucleosides like 5-fluoro-2’-deoxyuridine (CC50 = 10
to 10
M). The current
findings provide a possible explanation for the low toxicity, as we observed that the
isocarbostyril-thymine mimics are phosphorylated to their 5’-monophosphate deriva-
tives at low rates by the deoxynucleoside kinases. Although the mitochondrial enzymes
(
TK2 and dGK) showed low activity with EN-2 as a substrate, kinetic analysis
indicated that EN2 is a substrate for TK2 but not for dGK. The phosphorylation
efficiency of EN2 with TK2 was 12.6% relative to Thd, indicating a possible biological
role for EN2 in the mitochondrial DNA metabolism. On the other hand, the 2,4-
difluorophenyl nucleosides were relatively poor substrates for the cytosolic enzymes
and thus are probably not activated in proliferating cells. Exceptions were shown for
compounds JW1 and JW2, which were efficiently phosphorylated by TK2. However,
the transport of these analogues into the mitochondria is unknown but it is possible that
they may give mitochondrial toxicity by affecting the dNTP pools and/or DNA
replication. Although the triphosphate derivative of the 2,4-difluoro-5-methylphenyl
isostere (dFTP) is a substrate for the Klenow fragment of E. coli DNA polymerase
[
8,12,13]
I,
it is possible that the 2,4-difluoro-5-methylphenyl-nucleoside is not phos-
phorylated into di- and tri-phosphate under physiological conditions.
It is well established that the substrate specificities of TK1 and the mitochondrial
TK2 differ significantly, and also that the substrate tolerance with TK2 is more
[
20]
flexible.
Thus, TK1 and TK2 can accept minor modifications at the thymine base
[1]
without loss of substrate activity. In accordance, our results indicate that thymidine
kinases can tolerate to some extent the isosteric replacement of halogens at positions 2
and 4. In addition, the existence of a methyl group (JW1) or iodine (JW2) at the 2,4-
difluorotoluene moiety mimics modifications at the 5-position of the thymine base.
These substitutions are still accepted by the thymidine kinases. In fact, TK1 and TK2 are
reported to tolerate substitutions at the 5-position that are in size and physicochemical
[1]
properties comparable to a methyl or ethyl group. On the other hand, the existence of
intact sugar moiety is critical for TK1 activity, but not for TK2 activity, which can to
some extent tolerate acyclic compounds as in the case with compound JW3.

1872
Al-Madhoun et al.
In case of TK2, the substitution of a bulky isocarbostyril group (EN1) resulted in
poor substrate characteristics. However, the existence of the iodine atom at the 5-
position (EN2) that has the potential to interact via charge or hydrogen bonds with the
enzyme resulted in a relatively effective substrate.
In contrast, dCK exhibits a high specificity toward pyrimidine nucleosides, with
[
22]
little flexibility toward modification at the cytosine base.
enzyme, phosphorylating dGuo, dAdo, and dIno, but not Thd or its derivatives.
dGK is a purine-specific
[1,2]
In summary, the isocarbostyril- and difluorophenyl- 2’-deoxyribonucleosides were
generally poor substrates for the recombinant human enzymes TK1, TK2, dCK and
dGK. Exceptions were noticed for the difluorophenyl nucleosides JW1 and JW2, which
were moderate substrates of TK1, and JW2, which was a good substrate for the
mitochondrial enzyme TK2. On the other hand, JW1 and JW2 are most likely weak
competitive substrates in the presence of Thd (data not shown). These results, in
combination with the previously reported weak cytotoxicity of this group of compounds
and their kinase by-pass pro-drugs (for JW2), do not support a biochemical model
involving either bioactivation through phosphorylation, nor of incorporation into cel-
lular DNA under cell culture conditions. Surprisingly, the isocarbostyril nucleoside
EN2 showed good substrate specificity with TK2, indicating a possible biological role
in mitochondrial cytotoxicity. Thus, this study encouraged further investigations of the
cellular effects of EN2.
ACKNOWLEDGMENTS
We are grateful for the Swedish Research Council for financial support of this
research, and for Dr. Liya Wang for scientific comments. This work was supported in
part through the Canadian Institutes for Health Research (Grant 1440).
REFERENCES
1.
2.
3.
Arn e´ r, E.S.; Eriksson, S. Mammalian deoxyribonucleoside kinases. Pharmacol.
Ther. 1995, 67, 155–186.
Wang, L.; Hellman, U.; Eriksson, S. Cloning and expression of human mito-
chondrial deoxyguanosine kinase cDNA. FEBS Lett. 1996, 390, 39–43.
Wang, L.; Munch-Petersen, B.; Sj o¨ berg, A.H.; Hellman, U.; Bergman, T.; J o¨ rnvall,
H.; Eriksson, S. Human thymidine kinase 2: molecular cloning and characterization
of the enzyme activity with antiviral and cytostatic nucleoside substrates. FEBS
Lett. 1999, 443, 170–174.
4
.
Usova, E.V.; Eriksson, S. The effects of high salt concentrations on the regulation of
the substrate specificity of human recombinant deoxycytidine kinase. Eur. J.
Biochem. 1997, 248, 762–766.
5
.
.
Naesens, L.; De Clercq, E. Recent developments in Herpesvirus therapy. Herpes
2001, 8, 12–16.
Naimi, E.; Duan, W.; Wiebe, L.; Knaus, E. Synthesis of unnatural 7-substitued-1-(2-
6
deoxy-b-D-ribofuranosyl) isocarbostyrils: ‘‘thymidine replacement’’ analogues of

Thymidine Mimics by Human Deoxynucleoside Kinases
1873
deoxythymidine for evaluation as antiviral and anticancer agents. Nucleosides
Nucleotides Nucleic Acids 2001, 20, 1533–1553.
7.
McMinn, D.L.; Ogawa, A.K.; Wu, Y.; Liu, J.; Schultz, P.G.; Romesberg, F.E.
Efforts toward expansion of the genetic alphabet: DNA polymerase recognition of a
highly stable, self-pairing hydrophobic base. J. Am. Chem. Soc. 1999, 121, 11585–
11586.
8
.
.
Moran, S.; Ren, R.X.; Kool, E.T. Thymidine triphosphate shape analog lacking
Watson–Crick pairing ability is replicated with high sequence selectivity. Proc.
Natl. Acad. Sci. U. S. A. 1997, 94, 10506–10511.
9
Wang, Z.; Duan, W.; Wiebe, L.; Balzarini, J.; De Clercq, E.; Knaus, E. Synthesis of
1
-(2-deoxy-b-D-ribofuranosyl)-2,4-difluoro-5-substituted-benzene thymidine
mimics, some related a-anomers, and their evaluation as antiviral and anticancer
agents. Nucleosides Nucleotides Nucleic Acids 2001, 20, 11–40.
1
0. Wang, Z.; Wiebe, L.; Balzarini, J.; De Clercq, E.; Knaus, E. Syntheses of 4-[1-(2-
deoxy-b-D-ribofuranosyl)]- derivatives of 2-substituted-5-fluoroaniline: ‘‘Cytosine
replacement’’ analogs of deoxycytidine for evaluation as anticancer and anti-
human immunodeficiency virus (anti-HIV) agents. Can. J. Chem. 2000, 78, 1081–
1
088.
1. Wang, Z.; Wiebe, L.; Balzarini, J.; De Clercq, E.; Knaus, E. Chiral synthesis of
-[1-(2-deoxy-b-L-ribofuranosyl)]- derivatives of 2-substituted 5-fluoroaniline:
‘Cytosine replacement’’ analogues of deoxy-b-L-cytidine. J. Org. Chem. 2000,
5, 9214–9219.
1
4
‘
6
1
2. Moran, S.; Ren, R.R.-X.; Rumney, S.; Kool, E.T. Difluorotoluene, a nonpolar
isostere for thymine, codes specifically and efficiently for adenine in DNA
replication. J. Am. Chem. Soc. 1997, 119, 2056–2057.
13. Lai, J.S.; Qu, J.; Kool, E.T. Fluorinated DNA bases as probes of electrostatic effects
in DNA base stacking. Angew. Chem., Int. Ed. 2003, 42, 5973–5977.
1
4. Khalili, P.; Naimi, E.; Knaus, E.E.; Wiebe, L.I. Pharmacokinetics and metabolism
of the novel synthetic C-nucleoside, 1-(2-Deoxy-b-D-ribofuranosyl)-2,4-difluoro-5-
iodobenzene: a potential mimic of 5-Iodo-2’-deoxyuridine. Biopharm. Drug Dispos.
2003, 23, 105–113.
1
1
1
5. Khalili, P.; Naimi, E.; Sun, W.Y.; Knaus, E.E.; Wiebe, L.I. Biochemical and
pharmacokinetic evaluation of a novel nitric oxide donor pyrimidine nucleoside
hybrid drug as a potential anticancer/antiviral agent. Eur. J. Pharm. Sci. 2003, 19,
305–313.
6. Khalili, P.; Naimi, E.; Sun, W.Y.; Knaus, E.E.; Wiebe, L.I. Dose-dependent
pharmacokinetics of 1-(2-Deoxy-b-D-ribofuranosyl)-2,4-difluoro-5-iodobenzene: a
potential mimic of 5-Iodo-2’-deoxyuridine. Biopharm. Drug Dispos. 2003, 24,
3
85–395.
7. Wang, Z.; Duan, W.; Wiebe, L.; Balzarini, J.; De Clercq, E.; Knaus, E. Synthesis of
-[(2-hydroxyethoxy)methyl]- and 1-[(1,3-dihydroxy-2-propoxy)methyl]- deriva-
1
tives of 5-substituted-2,4-difluorobenzene: unnatural acyclo thymidine mimics for
evaluation as anticancer and antiviral agents. Nucleosides Nucleotides Nucleic
Acids 2000, 19, 1397–1411.
1
8. Al-Madhoun, A.S.; Johnsamuel, J.; Yan, J.; Ji, W.; Wang, J.; Zhuo, J.; Lunato, A.J.;
Woollard, J.E.; Hawk, A.E.; Cosquer, G.Y.; Blue, T.E.; Eriksson, S.; Tjarks, W.
Synthesis of a small library of 3-(carboranylalkyl)thymidines and their biological

1874
Al-Madhoun et al.
evaluation as substrates for human thymidine kinases 1 and 2. J. Med. Chem. 2002,
45, 4018–4028.
1
2
2
2
9. Al-Madhoun, A.; Johnsamuel, J.; Barth, R.F.; Tjarks, W.; Eriksson, S. Evaluation of
thymidine kinase 1 substrates as new candidate for boron neutron capture therapy.
Cancer Res. 2004, 64, 6280–6286.
0. Al-Madhoun, A.; Tjarks, W.; Eriksson, S. The role of thymidine kinases in the
activation of pyrimidine nucleoside analogues. Mini-Rev. Med. Chem. 2004, 4,
341–350.
1. Wang, J.; Choudhury, D.; Chattopadhyaya, J.; Eriksson, S. Stereoisomeric se-
lectivity of human deoxyribonucleoside kinases. Biochemistry 1999, 38, 16993–
16999.
2. Eriksson, S.; Wang, J. Substrate specificities of mitochondrial thymidine kinase and
cytosolic deoxycytidine kinase against 5-aryl substituted pyrimidine-2’-deoxyribose
analogues. Nucleosides Nucleotides 1995, 14, 507–510.
Received April 12, 2004
Accepted August 25, 2004