Enantio-selectivity of human nucleoside monophosphate kinases

Article abstract of DOI:10.1080/15257770701534014

Over recent years, there has been a renewed interest in the development of L-nucleosides as safe and efficacious drugs for the treatment of viral infections. Biological activity of these compounds requires phosphorylation to their triphosphate form, involving nucleoside monophosphate kinases in the second step. In order to characterize the activation pathway of L-nucleosides of the pyrimidine series, we studied the enantio-selectivity of human uridylate-cytidylate and thymidylate kinases. The results showed that these enzymes are only weakly enantio-selective and are thus probably involved in the activation of L-nucleosides in vivo. An activation pathway for telbivudine (L-dT) was therefore proposed. Copyright Taylor & Francis Group, LLC.

Full text of DOI:10.1080/15257770701534014

Nucleosides, Nucleotides, and Nucleic Acids, 26:1375–1379, 2007
Copyright ꢀ Taylor & Francis Group, LLC
C
ISSN: 1525-7770 print / 1532-2335 online
DOI: 10.1080/15257770701534014
ENANTIO-SELECTIVITY OF HUMAN NUCLEOSIDE
MONOPHOSPHATE KINASES
J. A. C. Alexandre
FRE2852-CNRS-Universite´ Paris 6, Paris, France
Laboratoire d’Enzymologie Mole´culaire et Fonctionnelle,
2
B. Roy Laboratoire de Chimie Organique Biomole´culaire de Synthe`se, UMR 5625
2
CNRS-UMII, Universite´ Montpellier II, Montpellier, France
D. Topalis Laboratoire d’Enzymologie Mole´culaire et Fonctionnelle,
2
FRE2852-CNRS-Universite´ Paris 6, Paris, France
C. Pe´rigaud Laboratoire de Chimie Organique Biomole´culaire de Synthe`se, UMR
2
5625 CNRS-UMII, Universite´ Montpellier II, Montpellier, France
D. Deville-Bonne Laboratoire d’Enzymologie Mole´culaire et Fonctionnelle,
2
FRE2852-CNRS-Universite´ Paris 6, Paris, France
Over recent years, there has been a renewed interest in the development of L-nucleosides as safe
2
and efficacious drugs for the treatment of viral infections. Biological activity of these compounds
requires phosphorylation to their triphosphate form, involving nucleoside monophosphate kinases in
the second step. In order to characterize the activation pathway of L-nucleosides of the pyrimidine
series, we studied the enantio-selectivity of human uridylate-cytidylate and thymidylate kinases. The
results showed that these enzymes are only weakly enantio-selective and are thus probably involved in
the activation of L-nucleosides in vivo. An activation pathway for telbivudine (L-dT) was therefore
proposed.
Keywords Structure activity relationship; human NMP kinases; HBV; L-dT
INTRODUCTION
Following FDA approval of lamivudine (3TC) for the treatment of HIV
and HBV, L-nucleosides have become the focus of a number of studies to as-
sess their anti-viral potential.[1] On overall, L-nucleosides are more resistant
These studies were supported by a grant from the Agence nationale pour la Recherche (ANR)
France.
Address correspondence to D. Deville-Bonne, Universite´ Paris 6, FRE 2852-CNRS-Paris 6, Labora-
toire d’Enzymologie T43, 4e, 43-44, 4 place Jussieu, F-75005 Paris, France. E-mail: ddeville@moka.ccr.
jussieu.fr
1375

1376
J. A. C. Alexandre et al.
FIGURE 1 Structure of D- and L-deoxyribonucleoside 5ꢁ-monophosphates (B = thymine, uracil,
cytosine).
to degrading enzymes and show better toxicological profiles than their
D-counterparts.[1] Additionally, derivatives of the L series not only possess
activity against viruses but some of them also act as anti-cancer[2] and anti-
malarial agents.[3] In vivo, activation of L-nucleosides necessitates phospho-
rylation to the triphosphate.[4] While the first phosphorylation step has been
extensively studied, less attention has been drawn to the second step.[4] This
step, i.e., the phosphorylation of nucleosides monophosphate and their ana-
logues into their diphosphate derivatives, involves nucleoside monophos-
phate kinases in both the de novo and salvage pathways.[4,5] Human
cells contain a thymidylate kinase (TMPK) and a uridine monophosphate-
cytidine monophosphate kinase (UMP-CMPK) that are only 21% identical
in sequence but that adopt the same fold characterised by a P loop, a NMP
binding domain and a LID domain.[6] The recent interest in L-nucleosides
of therapeutic potential has raised the need to assess the enantio-selectivity
of NMP kinases. In this study, we therefore examined the possibility to
phosphorylate L-dTMP, L-dUMP and L-dCMP (Figure 1) using hTMPK and
hUMP-CMPK, respectively.
MATERIALS AND METHODS
All D-nucleotides were purchased from Sigma Chemicals (St. Louis, MO,
USA).
L-Nucleoside 5ꢁ-monophosphates (NMP) were prepared from L-
nucleosides following Yoshikawa’s phosphorylation procedure.[7] Nucleo-
sides were dissolved in triethylphosphate and 1.2–2 eq. phosphorus oxychlo-
ride (POCl3) was added. After several hours of stirring at room temperature,
the intermediate phosphorodichloridates were hydrolyzed with triethylam-
monium hydrogencarbonate buffer (TEAB) 1 M pH 7.5 to afford the cor-
responding nucleoside monophosphates. Nucleoside 5ꢁ-monophosphates
were purified on DEAE-Sephadex A-25 (elution: linear gradient of TEAB
pH 7.6 from 10 to 300 mM) followed by chromatography on RP18 (elu-
tion: water to methanol 50%). The triethylammonium counter ions were ex-
changed to sodium by passing the nucleotide solution through a DOWEX-
AG 50WX2-400 column. Yields were 43–80%. The structures and purities

Enantio-Selectivity
1377
TABLE 1 Catalytic efficiencies of the human UMP-CMP and TMP kinases for the natural nucleoside
monophosphates and their corresponding L-enantiomers
Enzyme Substrate
Km (mM)
V (U.mg−1
)
kcat (s−1
)
kcat/Km (M−1s−1
)
Ki (mM)
D-UMP
D-dUMP
hUMP/ L-dUMP
0.05 0.01
1.3 0.3
0.70 0.09
0.020 0.005
0.9 0.1
350 30
130 10
8.0 0.7
0.37 0.02
130 10
2.8 × 106
6.2 × 103
5.3 × 102
6.5 × 106
8 × 104
1.5 0.4
18
1
0.92 0.04
350 30
198 13
CMPK
D-CMP
D-dCMP
L-dCMP
0.5 0.1
80
5
0.73 0.06
2.00 0.05
0.82 0.02
1 × 103
hTMPK D-dTMP
L-dTMP
0.024 0.002
0.38 0.07
6.8 0.2
0.86 0.07
3.0 0.1
0.34 0.03
1.2 × 105
9 × 102
5
1
of L-NMP were assessed by nuclear magnetic resonance (1H, 13C, 31P), fast-
atom-bombardment MS, UV spectroscopy, HPLC and polarimetry. In some
cases, it was necessary to separate nucleoside 5ꢁ-monophosphate from traces
of nucleoside 3ꢁ-monophosphate. This was done by HPLC using Hypercarb,
a porous graphic carbon stationary phase.[8]
After transformation of E. coli Rosetta(DE3)pLysS cells with the appro-
priate expression plasmid, recombinant hTMPK and hUMP-CMPK were
produced and purified as already reported.[5] The NMP kinases activity was
mesured as previously described.[5]
RESULTS
The reaction of recombinant human UMP-CMPK for L-dCMP and
L-dUMP was detected using a coupled spectrophotometric assay.[5] Hu-
man UMP-CMPK showed an apparent affinity (Km) to L-dUMP and L-dCMP
slightly better than that to D-dUMP and D-dCMP, i.e., 0.70 mM and 0.73 mM
compared to 1.3 mM and 0.9 mM, respectively (Table 1). The L-derivatives
were however much slower substrates than the natural nucleotides with a
20-fold decrease in velocity from D-dUMP to L-dUMP and a 100-fold de-
crease from D-dCMP to L-dCMP. Similarly, the reaction rate of hTMPK for
L-dTMP was 7-fold lower than for D-dTMP. Excess of L-dTMP inhibited the
enzyme whereas no inhibition was observed with the natural enantiomer.
Inhibition occurred at 1.2 mM of L-dTMP leading to an inhibition constant
of 5 mM (data not shown). The Michaelis constant of hTMPK for L-dTMP
was 0.38 mM compared to 0.025 mM for D-dTMP. The catalytic efficiency for
L-dTMP was thus approximately 100-fold lower than for D-dTMP (Table 1).
DISCUSSION
From the results presented here it appears that recombinant hTMPK
and hUMP-CMPK both showed relaxed specificity. The activities for the

1378
J. A. C. Alexandre et al.
L-derivatives were however lower than those for the corresponding D-
enantiomers with a decrease in catalytic efficiency within the 10–100-fold
range. A previous study reported on the phosphorylation of L-deoxycytidine
monophosphate bearing modifications on the sugar ring or the cytidine
base.[9] The study showed that a substitution at the 5 position of the base
had a negative impact on the catalytic efficiency. It also demonstrated that
the replacement of C-2ꢁ of the ribose by a sulphur atom improved the cat-
alytic efficiency contrary to the replacement by an oxygen atom. Among
the L-nucleotides tested only 3TCMP was almost as good as the natural sub-
strate, dCMP. However, this study did not evaluate the enantio-selectivity of
the enzyme as L-dCMP was not tested. Our study therefore concludes on the
relaxed enantio-selectivity of hUMP-CMP kinase.
Among the L-nucleosides of therapeutic potential, L-dT, the enantiomer
of thymidine, is currently in phase III clinical trials.[1] This compound pos-
sesses specific anti-HBV activity and is hoped to minimise the risk of resis-
tance development as, contrary to the nucleoside analogues accepted for
the treatment of HBV, it targets the positive strand of HBV DNA. Addition-
ally, L-dT exhibits an exceptional safety profile with no serious adverse ef-
fects observed in patients. Biological activity of L-dT requires phosphoryla-
tion to its triphosphate form. Although phosphorylation to the triphosphate
form has been demonstrated in HepG2 cells, no phosphorylation pathway
for L-dT has been characterised so far. In the light of the present and pre-
vious studies, an activation pathway for L-dT was proposed. The first step
is thought to involve either the mitochondrial thymidine kinase (dTK2) or
the cytosolic deoxycytidine kinase (dCK) as the cytosolic thymidine kinase
(dTK1) was found to be strictly enantio-selective.[10] The second phosphate
is most likely added by thymidylate kinase (TMPK) as we showed in this
study that this enzyme accepts L-dTMP as acceptor substrate. Finally, the
triphosphate form is probably produced by phosphoglycerate kinase (PGK)
as NDPK was found not to accept L-derivatives.[11] L-dT therefore follows an
activation pathway similar to L-FMAU, another anti-HBV agent in phase III
clinical trials.[12] Interestingly, for both compounds, the efficiency of phos-
phorylation by hTMPK was only 0.9–1% that of D-dTMP, reflecting a prefer-
ence of the active site for the natural configuration.
REFERENCES
1. Mathe´, C.; Gosselin, G. L-Nucleoside enantiomers as antivirals drugs: A mini-review. Antiviral Res.
2006, 71, 276–281.
2. Cheng, Y.-C. Potential use of antiviral L(-)nucleoside analogues for the prevention and treatment of
viral associated cancers. Cancer Letters 2001, 162, 1181–1187.
3. Gero, A.M.; Dunn, C.G.; Brown, D.M.; Pulenthiran, K.; Gorovits, E.L.; Bakos, T.; Weis, A.L. New
malaria chemotherapy developed by utilization of a unique parasite transport system. Current Phar-
maceutical Design 2003, 9, 867–877.
4. Maury, G. The enantioselectivity of enzymes involved in current antiviral therapy using nucleoside
analogues: A new strategy? Antivir. Chem. & Chemother. 2000, 11, 165–190.

Enantio-Selectivity
1379
5. Pasti, C.; Gallois-Montbrun, S.; Munier-Lehmann, H.; Ve´ron, M.; Gilles, A.-M.; Deville-Bonne, D.
Reaction of human UMP-CMP kinase with natural and analog substrates. Eur. J. Biochem. 2003, 270,
1784–1790.
6. Segura-Pena, D.; Sekulic, N.; Ort, S.; Konrad, M.; Lavie, A. Substrate-induced conformational
changes in human UMP-CMP kinase J. Biol. Chem. 2004, 279, 33882–33889.
7. Yoshikawa, M.; Kato, T.; Takenishi, T. A novel method for phosphorylation of nucleosides to 5ꢁ-
nucleotides. Tetrahedron Letters 1967, 8, 5065–5068.
8. Xing, J.; Apedo, A.; Tymiak, A.; Zhao, N. Liquid chromatographic analysis of nucleosides and their
mono-, di- and triphosphates using porous graphitic carbon stationary phase coupled with electro-
spray mass spectrometry. Rapid Communication in Mass Spectrometry 2004, 18, 1599–1606.
9. Liou, J.-Y.; Dutschman, G.E.; Lam, W.; Jiang, Z.; Cheng, Y.-C. Characterization of human UMP/CMP
kinase and its phosphorylation of D- and L-form deoxycytidine analogue monophosphates. Cancer
Research 2002, 62, 1624–1631.
10. Munch-Petersen, B.; Cloos, L.; Tyrsted, G.; Eriksson, S. Diverging substrate specificity of pure human
thymidine kinases 1 and 2 against antiviral dideoxynucleosides. J. Biol. Chem. 1991, 266, 9032–9038.
11. Gallois-Montbrun, S.; Faraj, A.; Seclaman, E.; Sommadossi, J.-P.; Deville-Bonne, D.; Ve´ron, M. Broad
specificity of human phosphoglycerate kinase for antiviral nucleoside analogs. Biochem. Pharmacol.
2004, 68, 1749–1756.
12. Hu, R.; Li, L.; Degre`ve, B.; Dutschman, G.E.; Lam, W.; Cheng, Y.-C. Behavior of thymidylate ki-
nase toward monophosphate metabolites and its role in the metabolism of 1-(2ꢁ-deoxy-2ꢁ-fluoro-β-
L-arabinofuranosyl)-5-methyluracil (Clevudine) and 2ꢁ,3ꢁ-didehydro-2ꢁ,3ꢁ-dideoxythymidine in cells.
Antimicrob. Agents Chemother. 2005, 49, 2044–2049.