Novel antiviral C5-substituted pyrimidine acyclic nucleoside phosphonates selected as human thymidylate kinase substrates

Article abstract of DOI:10.1021/jm1011462

Acyclic nucleoside phosphonates (ANPs) are at the cornerstone of DNA virus and retrovirus therapies. They reach their target, the viral DNA polymerase, after two phosphorylation steps catalyzed by cellular kinases. New pyrimidine ANPs have been synthesize

Full text of DOI:10.1021/jm1011462

222 J. Med. Chem. 2011, 54, 222–232
DOI: 10.1021/jm1011462
Novel Antiviral C5-Substituted Pyrimidine Acyclic Nucleoside Phosphonates Selected as Human
Thymidylate Kinase Substrates
Dimitri Topalis,†,¥ Ugo Pradere,‡,¥ Vincent Roy,‡ Christophe Caillat,§ Ahmed Azzouzi,‡ Julie Broggi,‡ Robert Snoeck,
ꢀ
Graciela Andrei, Jay Lin,^ Staffan Eriksson,^ Julie A. C. Alexandre,† Chahrazade El-Amri,† Dominique Deville-Bonne,†
Philippe Meyer,§ Jan Balzarini, and Luigi A. Agrofoglio*,‡
†
ꢁ
Groupe d’Enzymologie Moleculaire et Fonctionnelle, UR4-UPMC, Universite Pierre et Marie Curie, Sorbonne Universites, case courrier 256, 7,
quai St Bernard, 75252 Paris Cedex 05, France, Institut de Chimie Organique et Analytique, Centre National de Recherche Scientifique Unite
ꢁ
ꢁ
‡
ꢁ
§
ꢁ
ꢁ
ꢁ
Mixte de Recherche 6005, Universite d’Orleans, 45067 Orleans, France, Laboratoire d’Enzymologie et Biochimie Structurales, Centre National
de la Recherche Scientifique UPR 3082, 91198 Gif-sur-Yvette Cedex, France, REGA Institute for Medical Research, Katholieke Universiteit
Leuven, Leuven, Belgium, and ^Department of Veterinary Medical Chemistry, Swedish University of Agricultural Sciences, Box 575,
Biomedical Center, S-751 24 Uppsala, Sweden. ¥Both authors made equal contributions to this work.
Received September 4, 2010
Acyclic nucleoside phosphonates (ANPs) are at the cornerstone of DNA virus and retrovirus therapies.
They reach their target, the viral DNA polymerase, after two phosphorylation steps catalyzed by
cellular kinases. New pyrimidine ANPs have been synthesized with unsaturated acyclic side chains
(prop-2-enyl-, but-2-enyl-, pent-2-enyl-) and different substituents at the C5 position of the uracil
nucleobase. Several derivatives in the but-2-enyl- series 9d and 9e, with (E) but not with (Z)
configuration, were efficient substrates for human thymidine monophosphate (TMP) kinase, but not
for uridine monophosphate-cytosine monophosphate (UMP-CMP) kinase, which is in contrast to
cidofovir. Human TMP kinase was successfully crystallized in a complex with phosphorylated
(E)-thymidine-but-2-enyl phosphonate 9e and ADP. The bis-pivaloyloxymethyl (POM) esters of
(E)-9d and (E)-9e were synthesized and shown to exert activity against herpes virus in vitro (IC50 = 3 μM)
and against varicella zoster virus in vitro (IC50 = 0.19 μM), in contrast to the corresponding inactive (Z)
derivatives. Thus, their antiviral activity correlates with their ability to act as thymidylate kinase substrates.
Introduction
to an unsaturated side chain and described a method of
screening those ANPs for their affinity to act at the acceptor-
binding site of hUCK.11 (Note: PDB ID for the human TMP
kinase complexed to (E)-TbutP and ADP: 2xx3.
In the past ten years, acyclic nucleoside phosphonates (ANPs),
nucleotide analogues with a stable P-C bond, have become
major antiviral nucleotide derivatives; three ANPs (adefovir
[PMEA], tenofovir [(R)-PMPA], and cidofovir [(S)-CDV])
are approved for the treatment of severe viral infections
(Figure 1).1 The only acyclic pyrimidine nucleoside phospho-
nate analogue found to be effective to date is cidofovir (CDV)
and the acyclic 2,4-diaminopyrimidine nucleoside phospho-
nates (i.e., PMEO-DAPym).2 Their activities are based on
the intracellular phosphorylation to their diphosphates
(ANP-PPs), which act as a triphosphate, by two salvage pathway
kinases (nucleoside monophosphate [NMP] and nucleoside
diphosphate [NDP] kinases). Thediphosphorylated ANPsact
as chain terminators of virus DNA polymerase3-6 inhibiting
the viral (i.e., in HIV7 or vaccinia) replication.8,9 All studied
ANPs are slowly phosphorylated by human NMP kinases:
AMPkinases1and2forPMEAand(R)-PMPA,3,10 and human
UMP-CMP kinase (hUCK) for CDV.4,11 The last phosphory-
lation step is performed by several enzymes, including NDP
kinases and creatine kinases.3,4 In order to improve the antiviral
properties of ANPs, we have recently focused on the synthesis
of new pyrimidine ANPs with the phosphorus atom attached
Several novel (E)-3-(N1-uracil)-prop-2-enyl phosphonic
acid (U-proP, 6a-c) derivatives were found to have similar
affinities to the natural 20-deoxyuridine monophosphate (dUMP)
and 20-deoxycytidine monophosphate (dCMP) substrates, and
higher affinity than that of CDV; however, they were not sub-
strates for hUCK.11,12 Thus, we wanted to investigate the
contribution, if any, of the length of the unsaturated chain of
our derivatives (e.g., extending from a prop-2-enyl- to a but-2-
enyl- or pent-2-enyl- side chain) to the efficiency of phosphor-
ylation by human TMP kinase (hTMPK) and hUCK. Our
rational design was based on the replacement of the O-C
bond with a trans-alkene which would allow mimicry of the
three-dimensional geometry provided by the backbone of
PMEA, PMPA, and CDV while maintaining an electronic
contribution similar to that brought by the oxygen atom.
From a chemical viewpoint, a general strategy for the
convenient stereo-controlled synthesis of trans-alkene acyclic
nucleoside phosphonates based on olefin cross metathesis seemed
to be attractive, as the original approach tends to be tedious
and low-yielding. Several new derivatives are efficiently activated
by hTMPK, and the best substrates were converted to bis-
(pivaloyloxymethyl)ester phosphonateprodrugsand foundto
be active against several herpes viruses in cell cultures.
*Corresponding authors. Luigi A. Agrofoglio, Phone þ33 238-494-
582, Fax þ 33 238-471-281, E-mail: Luigi.Agrofoglio@univ-orleans.fr.
Chahrazade El-Amri, Phone þ33-144-276-952, Fax þ331.44.27.51.40,
E-mail: chahrazade.el_amri@upmc.fr.
r
pubs.acs.org/jmc
Published on Web 12/03/2010
2010 American Chemical Society

Article
Journal of Medicinal Chemistry, 2011, Vol. 54, No. 1 223
Figure 1. Structure of bioactive ANP and derivatives studied.
Scheme 1. Synthesis of Prop-2-enyl-, But-2-enyl-, and Pent-2-enyl- Acyclonucleoside Phosphonatesa
a Reagents and conditions: (i) BzCl, pyridine/CH3CN; (ii) crotyl bromide, K2CO3, DMF; (iii) diethyl vinylphosphonate, [Ru] = cat., CH2Cl2, 40 ꢀC;
(iv) dimethyl allylphosphonate, [Ru] = cat., CH2Cl2, 40 ꢀC; (v) diethyl but-3-enylphosphonate, [Ru] = cat., CH2Cl2, 40 ꢀC; (vi) NH3/MeOH;
(vii) TMSBr, CH2Cl2.
Results and Discussion
reaction between 3a-e and dimethyl allylphosphonate cat-
alyzed by the [Ru] = Grubbs II catalyst afforded a chroma-
tographically separable mixture of (E)-but-2-enyl 7a-e and
the Z isomers 10a-e (E/Z, 4:1). Removal of the benzoyl
protecting group by 7 N NH3/MeOH of compounds 7a-e
and 10a-e afforded, respectively, derivatives (E)-8a-e and
(Z)-11a-e. Subsequent cleavage of the phosphonic ester by
TMSBr led to the free but-2-enyl-phosphonic acids (E)-9a-e
and (Z)-12a-e, respectively. Under the same conditions of
CM, 3a-e was treated with diethyl but-3-enylphosphonate
to afford the desired cross-coupling heterodimers (E)-13a-e,
with only traces of (Z)-isomers. Subsequent deprotections
afforded the free pent-2-enyl- phosphonic acid (E)-15a-e. From
NH3/MeOH deprotection 3a-e, the cross-metathesis reaction
between 4a-e and an excess of vinyl diethylphosphonate was
Chemistry. The key step of olefin cross metathesis was
realized between commercial alkenyl phosphonates and
various C5-substituted-N3-benzoylated crotylated uracils
3a-e or unprotected analogues 4a-e to prepare phospho-
nate derivatives 6a-e, 9a-e, 12a-e, and 15a-e, according
toprocedurespreviouslydescribedbyour group(Scheme 1).13
In agreement with the general model reported by Grubbs14
for olefin cross metathesis, the use of crotylated building
blocks promoted heterodimer formation while reducing the
formation of homodimers, contrary to allylated derivatives.
Thus, starting from the appropriate C5-substituted uracil
1a-d and thymine 1e, a regioselective N3-benzoylation15 led
to compounds 2a-e and a subsequent alkylation with crotyl
bromide afforded 3a-e, respectively. A cross metathesis

224 Journal of Medicinal Chemistry, 2011, Vol. 54, No. 1
Topalis et al.
Scheme 2. Synthesis of Bis(POM) Prodrugs of (E) and (Z)
TbutP and 5Br-UbutPa
a Reagents and conditions: (i) chloromethyl pivalate, NaI, CH3CN,
reflux.
performed with Grubbs II catalyst. The desired cross-
coupling heterodimers 5a-e were obtained in moderate to good
yields exclusively as the (E)-isomers. The stereochemistry of
1
3
the olefin was confirmed by H NMR, with a J coupling
constant (∼17.0 Hz) of the olefinic protons. Upon treatment
with TMSBr in CH2Cl2, the free prop-2-enyl- phosphonic
acids 6a-e were isolated in good yields. Finally, for cellular
evaluation of the antiviral activity, the two best acid phos-
phonate substrates of hTMPK 9d,e and 12d,e were converted
to their bis(POM) prodrugs by reaction of dimethyl phos-
phonate derivatives 8d,e and 11d,e with chloromethyl pivalate
and sodium iodide (Scheme 2).16 The following bis(POM)-
ester prodrugs (E)-bis(POM)TbutP 16, (Z)-bis(POM)TbutP
17, (E)-bis(POM)-5Br-UbutP 18,and(Z)-bis(POM)-5Br-UbutP
19 were obtained in moderate (for (Z)-isomers) to good yields
(for (E)-isomers) (Scheme 2).
Reaction of C5-Substituted-Uracil ANP with Human
hTMPK. The ability of new unsaturated ANPs (bearing a
uracil, thymine, cytosine, or C5-halogeno-uracil moiety) to
be phosphorylated by kinases present in human cells was
measured first using an enzymatic in vitro assay with re-
combinant hTMPK.17 TMPK was found associated with cell
proliferation resulting in an expression peak in the S-phase of
the cell cycle.18 The recombinant forms of this “housekeep-
ing” enzyme was already described and its substrate specific-
ity and catalytic potency determined toward a large panel
of substrates including antiviral compounds has been de-
scribed.17,19-21 Recently, a putative mitochondrial hTMPK
has been reported.22 However, it was not considered here,
since it may be only a minor contributor to nucleoside
analogue activation in cells.
Figure 2 represents the saturation curves for (Z)- and
(E)-5-halouracil-but-2-enyl phosphonates as substrates with
hTMPK in the presence of saturating ATP. All substrates
show a hyperbolic saturation curve, indicating that they are
all true Michaelis-Menten substrates. The rate constants of
the (E)-isomers were better than those for the (Z)-deriva-
tives, indicating that hTMPK worked better on butenyl-
substrates with a trans configuration, as shown for UbutP 9a
and 12a, 5F-UbutP 9b and 12b, 5Cl-UbutP 9c and 12c, and
5Br-UbutP 9d and 12d (Figure 2). This confirmed our
hypothesis that replacement of O-C bond with trans-alkene
would allow us to mimic the three-dimensional geometry
provided by the backbone of ANP.
Thus, on the basis of saturation curves which indicate that
only the (E)-isomers of the butenyl series are substrates for
hTMPK, only the (E)-isomers of the newly obtained ANP
were compared to natural 20-deoxyribonucleotides mono-
phosphate for their enzymatic properties (kcat, KM, and cata-
lytic efficiency kcat/KM) toward human TMP kinase (Table 1).
Figure 2. Treatment of human TMP kinase with (Z)- and (E)-5-
halouracil-but-2-enyl-phosphonates. hTMPK saturationcurveswere
plotted as a function of UbutP derivatives as substrates: (E)-UbutP
9a (2), (Z)-UbutP 12a (4), (E)-5F-UbutP 9b (9), (Z)-5F-UbutP 12b
(0), (E)-5Br-UbutP 9d (b), (Z)-5Br-UbutP 12d (O), and (E)-5Cl-
UbutP 9c (þ). The initial rate of each acceptor phosphorylation was
measured in the presence of 1 mM ATP and 5 mM Mg2þ using the
standard coupled assay. The quantity of enzyme used in each
experiment was 0.56 μM for (E)-5Br-UbutP (9d); 1.4 μM for
(E)-5F-UbutP (9b), (E)-5Cl-UbutP (9c), and (E)-UbutP (9a);
2.8 μM for (Z)-UbutP (12a) and (Z)-5Br-UbutP (12d), 8.4 μM for
(Z)-5F-UbutP (12b). KM and kcat values were obtained by fitting to a
hyperbole (results shown in Table 1 for (E) stereoisomers). KM and
k
cat values were 0.55 mM and 0.12 s-1 for (Z)-UbutP 12a; 0.3 mM
and 0.037 s-1 for (Z)-5F-UbutP 12b; 0.4 mM and 0.2 s-1 for (Z)-5Br-
UbutP 12d.
Michaelis constants KM are rather similar to KMTMP within
one order of magnitude. The presence of a halogen at the C5
position enhanced both the affinity and the turnover number.
The best substrates for hTMPK in the (E)-butenyl series
were the 5Br-UbutP 9d and 5Me-UbutP (TbutP) 9e, as their
catalytic efficiencies (kcat/KM) were only found to be 3-4
times lower than the natural substrate TMP. Both (E)-prop-
2-enyl- and (E)-pent-2-enyl analogues were less efficient
substrates than their butenyl counterparts. The (E)-5Cl-UproP
6c was slowly phosphorylated by hTMPK at low concen-
trations11 and (E)-TproP ((E)-5Me-UproP, 6e) was slightly
better. The (E)-pentenyl- series was found to be more efficient
with the best substrate being (E)-TpenP 15e (Table 1, Figure 3).
The histogram provided in Figure 3 clearly shows that
(E)-5Br-UbutP 9d and (E)-TbutP 9e are phosphorylated by
hTMPK at efficiencies similar to natural substrates, dUMP
and TMP.
Crystal Structure of (E)-TbutP (9e) in hTMPK Active Site.
The hTMPK fold was found to be highly similar to the
adenylate kinase fold, with a CORE domain containing the
ATP binding site, and two mobile domains, the NMP and
LID domains, closing up upon substrate binding.21 To
establish and understand the structural basis for (E)-TbutP
9e recognition and phosphorylation by the hTMPK, the
structure of the hTMPK/(E)-TbutP 9e co-crystal was solved,
and details of the crystallographic data analysis are available
(Supporting Information Table 1). The structure showed a
phosphorylated TbutP molecule 9e (TbutP-P) at the accep-
tor site of the enzyme and an ADP molecule at the donor site
(Figure 4A). Since the crystals grew in the presence of TbutP

Article
Journal of Medicinal Chemistry, 2011, Vol. 54, No. 1 225
Table 1. Enzymatic Parameters for [E]-C5-Substituted Uracil Acyclic Phosphonates with Human TMP Kinase
R
acronym
K
M (mM)
kcat (s-1
)
kcat/KM (M-1 s-1
)
A- Propenyl
B- Butenyl
C-Pentenyl
references
H-
F-
6a (E)-UproPa
6b (E)-5F-UproP
6c (E)-5Cl-UproP
6d (E)-5Br-UproP
6f (E)-TproP
nd
nd
nd
nd
1
nd
nd
nda
nd
Cl-
Br-
CH3-
H-
nd
nd
80a
nda
600
0.6
9a (E)-UbutP
9b (E)-5F-UbutP
9c (E)-5Cl-UbutP
9d (E)-5Br-UbutP
9e (E)-TbutP
1.7
1
0.48
0.73
0.50
0.90
0.84
0.074
0.09
0.1
280
730
F-
Cl-
Br-
CH3-
H-
0.13
0.02
0.024
2.5
4100
45 000
35 000
30
15a (E)-UpenP
15b (E)-5F-UpenP
15c (E)-5Cl-UpenP
15d (E)-5Br-UpenP
15e (E)-TpenP
dUMP
F-
1.9
47
660
Cl-
Br-
CH3-
0.15
0.16
0.10
0.17b
0.02b
0.01
0.058
0.32
4.8b
3b
360
3200
28 000b
150 000b
120 000
1000c
nd
dTMP
5Br-dUMP
1.2
AZTMP
cidofovir
nd
nd
a Ref 11. b Ref 17. c Ref 21. nd: not detectable, standard error 15%.
Figure 3. Histogram of human TMP kinase phosphotransfer effi-
ciency with synthesized (E)-unsaturated ANP. The kinetic param-
eters of pure recombinant hTMPK were measured at the steady
state. Catalytic efficiencies were calculated from the ratio kcat/KM
and expressed as M-1 s-1. Oy axis: blue, propenyl series; red,
butenyl series; yellow, pentenyl series; and green, 20-deoxyriboses
(standard). Ox axis: substitutions at C5 position: H, F, Cl, Br, and
CH3.
Figure 4. Crystal structure of human TMP kinase complexed to
(E)-TbutP 9e and ADP. (A) X-ray crystal structure of the (E)-TbutP
(9e) molecule bound to the human TMP kinase enzyme (cartoon
with cyan helices and pink loops). (E)-TbutP is phosphorylated and
the phosphate donor site is occupied by ADP (atom type color) and
Mg2þ (green). The structure of the natural substrates dTDP and
ADP (gray) bound to the enzyme are superimposed (PDB ID 1e2g).
(B) Key active site residues involved in TButP-P recognition are
labeled, and the polar interactions made with the nucleotide analog
are shown by red dashes.
and ATP, the phosphoryl transfer must have taken place
during crystallization.
We compared TbutP-P with the natural substrate TDP
from the hTMPK/TDP/ADP structure.21 The (E)-but-2-
enyl side chain moiety mimics the conformation of the
C10-O40-C40-C50 atoms from the 20-deoxyribose in TDP,
confirming our hypothesis. The phosphorylated TbutP-P
derivative was slightly shifted toward the phosphate site so
that its phosphonate and phosphate are almost superim-
posed with the R and β-phosphate of TDP. As a consequence
of the lack of a 30-hydroxyl, the interaction with Asp15 is lost,
and the enzyme LID conformation is rearranged in such a
manner that Arg150 is reoriented and occupies this leaving
space. Consequently, Arg150 together with His148 strongly
interacts with the TbutP-P phosphonate and phosphate
(Figure 4B), allowing a productive binding of TbutP-P and
ADP in hTMPK active site.
C5-Substituted-Uracil ANP as Potential Ligands of hUCK,
hTK1, and hTK2. As mentioned above, the synthesized
C5-substituted-uracil unsaturated ANPs were assayed with
different human kinases compatible with their structures
(hUCK, thymidine kinases 1 and 2) allowing their recogni-
tion by the enzymes. All uracil unsaturated ANPs were not
substrates of hUCK, regardless of the unsaturated acyclic

226 Journal of Medicinal Chemistry, 2011, Vol. 54, No. 1
Topalis et al.
Table 2. Antiviral Activities and Cytotoxicities in HEL Cell Cultures of Bis(POM) Prodrugs of (E) and (Z) TbutP 16 and 17 and 5Br-UbutP 18 and 19
EC50a (μM)
HSV-1 HSV-2 HSV-1 TK- Vaccinia virus
HCMV
VZV
b
compound
CC50 MCCc (KOS)
35 g200 3.1
35 g200 >40
34 200 20
70 >200 >200
>250 >250 0.04
(G)
6.1
>40
(KOS ACVr)
(AD-169 and Davis) TKþ (OKA) TK- (07/1)
[E]-bis(POM)-TbutP (16)
[Z]-bis(POM)-TbutP (17)
[E]-bis(POM)-5Br-UbutP (18)
[Z]-bis(POM)-5Br-UbutP (19)
Brivudin
9.2
>40
19
30
>40
33
>40
>200
>40
0.41
>40
0.19
>40
16
45
30
20
97
15
110
>200
250
90
>200
>250
1.5
6.0
>100
0.012
0.044
3.3d
>100
Cidofovir
Ganciclovir
280 >250
164 >100
3.5
0.08
2.0
0.05
2.5
3.0
1.2
5.2
0.070
62d
a Effective concentration (μM), required to reduce virus-induced cytopathogenicity by 50%. b Cytostatic concentration (μM), required to inhibit HEL
cell proliferation by 50%. c Minimum cytotoxic concentration (μM), required to cause a microscopically detectable alteration of normal cell
morphology. d Data for acyclovir (μM).
chain, but inhibitors for some of them (e.g., 9c), (Supporting
Information Figure 1). This is in contrast to CDV, which is
slowly processed to CDV-phosphate by hUCK.4,11 Spectro-
scopic experiments using circular dichroism also indicate
that the new phosphonates described here do not induce the
conformation change induced by dCMP and cidofovir
(Supporting Information Table 2).
vaccinia virus (VV) in HEL cell cultures. All derivatives were
devoid of antiviral activity except when the phosphonate
group was protected by POM groups (Table 2). This lack of
activity confirms the low biodisponibility of acid phospho-
nates and the need of a biolabile protecting group to reveal
antiviral activities.
The bis(POM) derivative of (E)-TbutP 16 showed pro-
nounced inhibitory activity against HSV-1, HSV-2, and
VZV (EC50 = 2.5-6.1 μM). Interestingly, the compound
showed comparable antiviral activity against TK- deficient
virus strains with a decreased inhibitory effect (EC50 = 9.2 μM)
against the TK- deficient HSV-1 strain, but an excellent
The ability of the most promising hTMPK substrates to
inhibit hUCK activity was investigated. (E)-TbutP 9e and
(E)-5Cl-UbutP 9c were both found to be poor inhibitors
(KI = 1.2 mM and 0.25 mM, respectively). Surprisingly,
9c was found to bind to both sites (dCMP and ATP sites)
of hUCK, thus preventing any substrate activity (Supporting
Information Figure 1A,B). The binding to the ATP site was
confirmed by fluorescence competition assay with N-methyl
anthraniloyl ATP (Mant-ATP) as probe.11,23 Compound 9c
readily displaced Mant-ATP (appKD = 0.17 mM). CDP-
β-(N0-methylanthraniloylaminobutyl)-phosphoramidate
(MABA-CDP) was also displaced by 9c, indicating that it
also binds to the acceptor site (dCMP site) with an apparent
equilibrium dissociation constant KD = 0.16 mM.11 The
affinity constants measured by fluorophore displacement
were in the same range as the KI obtained from kinetic
inhibition data, any differences being presumably due to
the bulky Mant-moiety. It is important to note that CDV,
PMEA, and PMPA did not displace Mant-ATP from
hUCK, showing that they do not bind to the ATP site.
Similarly, the new acyclic compounds also did not displace
Mant-ATP from the ATP site of hTMPK.
Several acyclic nucleoside analogues have been previously
shown to inhibit mitochondrial thymidine kinase (TK).24 So,
the possible inhibition of human recombinant TK1 (cytosolic)
and TK2 (mitochondrial) by (E)-5Cl-UbutP 9c has been
investigated. At the concentration of 0.3 mM, 9c proved to
inhibit thymidine phosphorylation (1 μM) catalyzed by TK1
and TK2 (ATP 1 mM and thymidine 0.3 mM) by less than
15% activity. Altogether, these data rule out a strong inter-
action of the C5-substituted-uracil-unsaturated ANP with
TK1 and TK2, as well as with hUCK. They most likely will
not be recognized by these enzymes in the cell context, as the
cellular concentration of ANP rarely reaches a concentration
such as 0.3 mM.
increase in activity against the VZV TK- strain with EC50
=
0.19 μM. There was moderate antiviral activity against VV
replication, but no activity against HCMV. Also, (E)-bis-
(POM)-5Br-UbutP 18 proved to be inhibitory against
HSV-1, HSV-2, VZV, and VV, but not HCMV (EC50=16-
33 μM) and fully kept activity against HSV-1 TK- and VZV
TK-. The compounds were not significantly cytotoxic at
200 μM, but slightly cytostatic at 34-70 μM against HEL cell
proliferation. In contrast with the (E) isomer of bis(POM)-
TbutP 16 and bis(POM)-UbutP 18, the corresponding (Z)
isomers 17 and 19 exhibited little or no antiviral activity.
The (E)-bis(POM)-TbutP derivative 16 proved most in-
hibitory against HSV-1, HSV-2, and VZV. Presumably, in
analogy to bis(POM)-PMEA, the compound was rapidly
taken up by the virus-infected cells and intracellularly con-
verted to the free phosphonate derivative after hydrolysis of
the POM prodrug moieties by cellular esterases. In this way,
it circumvented the HSV-1 TK dependency of pyrimidine
nucleoside analogues such as brivudin (BVDU) for further
conversion to its active metabolite. The independence of
cellular TK activity was testified by the pronounced antiviral
activity of the compound against mutant TK-deficient HSV-
1 TK- and VZV TK-. As is also the case for other phos-
phonates such as PMPA and CDV, the nature of the isomers
[(R) or (S) for PMPA/CDV or (E) or (Z) for bis(POM)-
TbutP] is important for its eventual antiviral activity. The
higher antiviral activity of the (E) enantiomers 16 and 18
compared to the corresponding (Z) isomers 17 and 19 clearly
correlated with their ability to be phosphorylated by
hTMPK. This is an argument in favor of a mode of action
involving phosphorylation of the compounds in the host cells
in order to target virus DNA polymerase. It is remarkable
that Hostetler’s group recently reported the antiviral activity
of similar pentenyl derivatives of uracil and thymine against
several DNA viruses and hepatitis B virus, but in (Z) con-
figuration, when assayed as hexadecyloxy-propyl esters to
increase their penetration into cells.25 Indeed, (E)-TbutP
Antiviral Activities of Selected Novel Pyrimidine Unsatu-
rated ANP in HEL Cell Cultures. The newly synthesized
compounds were evaluated against a variety of DNA viruses
including herpes simplex virus type 1 (HSV-1, strain KOS),
HSV-2 (strain G), thymidine kinase (TK)-deficient HSV-1
TK-, varicella zoster virus (VZV, strain OKA), TK-deficient
VZV TK- (07/1), human cytomegalovirus (HCMV), and

Article
Journal of Medicinal Chemistry, 2011, Vol. 54, No. 1 227
recorded on 250 or 400 MHz spectrometers, 13C NMR on 62.9
or 100 MHz spectrometers, and 31P on 162 MHz spectrometer at
room temperature, using deuterated solvents as internal stan-
dard. Chemical shifts are given in ppm and multiplicities are
reported as s (singlet), d (doublet), t (triplet), q (quartet),
qi (quintuplet), or m (multiplet). Assignments of NMR spectra
follow standard nucleosides nomenclature: nucleoside bases are
numbered from N-1 to C-6 and acyclic chains are numbered
from C-10 to C-50. Similar conventions apply for the correspond-
ing hydrogen atoms. High-resolution mass spectrometry was
performed by the Mass Spectrometry Center of Blaise Pascal
ꢀ
University (Aubiere, France). The purity was determined by
semipreparative HPLC (Hypersil 100; C-18; 5 μm); with an
appropriate gradient of acetonitrile/H2O; purity of key target
compounds was >95%.
The following products are known products or previously
reported by our group:14 N3-Benzoyluracil (2a), CAS registration
2775-87-3; N3-Benzoyl-5-fluorouracil (2b), CAS registration
61251-77-2; N3-Benzoyl-5-chlorouracil (2c), CAS registration
214687-03-3; N3-Benzoyl-5-bromouracil (2d), CAS registra-
tion 206762-91-6; N3-Benzoylthymine (2e), CAS registration
4330-20-5; N3-Benzoyl-N1-crotyluracil (3a), CAS registration
1067228-36-7; N3-Benzoyl-N1-crotyl-5-chlorouracil (3c),
CAS registration 1023339-53-8; N3-Benzoyl-N1-crotyl-5-bro-
mouracil (3d), CAS registration 1023339-58-3; N3-Benzoyl-
N1-crotylthymine (3e), CAS registration 882659-57-6;
N1-Crotyluracil (4a), CAS registration 852998-43-7; N1-[(E)-
3-Diethoxyphosphonyl-prop-2-enyl]-uracil (5a), CAS registra-
tion 1023339-16-3; N1-[(E)-3-Diethoxyphosphonyl-prop-2-
enyl]-5-chlorouracil (5c), CAS registration 1023339-72-1; N1-
[(E)-3-Diethoxyphosphonyl-prop-2-enyl]-5-bromouracil (5d),
CAS registration 1023339-74-3; N1-[(E)-3-Dihydroxypho-
sphonyl-prop-2-enyl]-uracil (6a), CAS registration 1023339-
19-6; N1-[(E)-3-Dihydroxyphosphonyl-prop-2-enyl]-5-chloro-
uracil (6c), CAS registration 1023339-96-9; N1-[(E)-3-Dihy-
droxyphosphonyl-prop-2-enyl]-5-bromouracil (6d), CAS regis-
tration 1023339-98-1; N3-Benzoyl-N1-[(E)-4-dimethoxyphos-
phonylbut-2-enyl]-uracil (7a), CAS registration 1023339-30-1;
N3-Benzoyl-N1-[(E)-4-dimethoxyphosphonylbut-2-enyl]-5-chloro-
uracil (7c), CAS registration 1023339-83-4; N3-Benzoyl-N1-[(E)-
4-dimethoxyphosphonylbut-2-enyl]-5-bromouracil (7d), CAS
registration 1023339-86-7; N1-[(E)-4-Dihydroxyphosphonyl-but-
2-enyl]-uracil (9a), CAS registration 1023339-32-3; N1-[(E)-
4-Dihydroxyphosphonyl-but-2-enyl]-5-chlorouracil (9c), CAS
registration 1023340-06-8; N1-[(E)-4-Dihydroxyphosphonyl-
but-2-enyl]-5-bromouracil (9d), CAS registration 1023340-07-9;
N3-Benzoyl-N1-[(Z)-4-dimethoxyphosphonylbut-2-enyl]-uracil
(10a), CAS registration 1023339-30-1; N3-Benzoyl-N1-[(Z)-4-
dimethoxyphosphonyl-but-2-enyl]-5-chlorouracil (10c), CAS
registration 1023340-16-0; N3-Benzoyl-N1-[(Z)-4-dimethoxy-
phosphonyl-but-2-enyl]-5-bromouracil (10d), CAS registration
1023340-17-1; N1-[(Z)-4-Dihydroxyphosphonyl-but-2-enyl]-
uracil (12a), CAS registration 1023339-34-5; N1-[(Z)-4-Dihydrox-
yphosphonyl-but-2-enyl]-5-chlorouracil (12c), CAS registration
1173147-85-7.
Figure 5. Structural comparison of (E)-TbutP and (Z)-TpentP as
nonclassical bioisosters of TMP.
with a 4-carbon acyclic moiety (butenyl series) in trans
configuration and (Z)-TpentP in cis configuration both
mimic TMP and can be considered as nonclassical bioisos-
ters for hTMPK. They both adopt a similar conformation
to the natural substrate TMP (Figure 5): the side chain of
(E)-TbutP mimic atoms C10-O40-C40-C50 of TMP (skeleton
in red), as shown from the crystal structure, whereas the side
chain of (Z)-TpentP mimic atoms C10-C20-C30-C40-C50 of
TMP (skeleton in blue). (Z)-TpenP and (Z)-UpenP are then
likely efficient substrates for hTMPK.
Conclusion
In the present study, from several new compound acyclo-
nucleoside phosphonates, the (E)-thymidine-but-2-enyl phos-
phonate 9e was markedly phosphorylated by recombinant
hTMPK with efficiency reaching more than twenty times that
of AZT-monophosphate. We have demonstrated by X-ray
analysis of (E)-TbutP 9e in hTMPK that the (E)-unsaturated
chain moiety mimics the conformation of the C10-040-C40-C50
atoms from the 20-deoxyribose in TDP The bis(POM) pro-
drug of (E)-TbutP 16 presented a potent antiviral activity in
vitro against several herpes viruses, namely, herpes simplex
viruses 1 and 2 and varicella zoster virus. Its C5-substituted
analogues were also substrates for hTMPK with efficiencies
strongly dependent on the nature of the C5-substituent but
with weaker antiviral activities. The (E)-bis(POM)-TbutP
16 derivative should be considered a novel antiviral lead
compound for further optimization of both activation by
cellular and viral kinases and antiviral potencies. Particularly,
interactions of the ANPs with the ATP site of human UMP-
CMP kinase have to be considered and avoided in the design
of future effective compounds.
N3-Benzoyl-5-fluoro-N1-crotyluracil (3b). To a DMF (12 mL)
solution of N3-benzoyl 5-fluoro-uracil (1.0 g) were added
K2CO3 (1.05 equiv) and crotyl bromide (1.05 equiv) under argon
atmosphere. After 1 h stirring at room temperature, the reaction
mixture was diluted with EtOAc and then washed with an
aqueous saturated NH4Cl solution. The aqueous phases were
extracted with EtOAc, and the combined organic layers were
dried over MgSO4, filtered, and concentrated under reduced
pressure. The residue was purified by silica gel column chroma-
tography (petroleum ether/EtOAc 5:5). HRMS (ESI): m/z calcd
for C15H13FN2O3Na (Mþ þ Na) 311.0796, found 311.0792. 1H
NMR (400 MHz, CDCl3): δ 7.92 (d, J = 8.0 Hz, 2H), 7.64
(t, J = 8.0 Hz, 1H), 7.51 (t, J = 8.0 Hz, 2H), 7.35 (d, 1H), 5.96-
5.74 (m, 1H), 5.59-5.42 (m, 1H), 4.40 (d, J = 6.5 Hz, 0.4H), 4.29
(d, J = 6.5 Hz, 1.6H), 1.78 (d, J = 6.5 Hz, 3H).
Experimental Section
Chemistry. Natural nucleotides were purchased from Sigma
Chemicals (St. Louis, MO, USA) and Mant-ATP from Jena
Biosciences (Jena, Germany). MABA-TDP and MABA-CDP
were kindly synthesized by Dr. S. Pochet (Institut Pasteur, Paris)
as described.11,26 Commercially available chemicals were used
as received. CH2Cl2 was distilled from CaH2 prior to use.
Reactions were monitored by thin-layer chromatography
(TLC) analysis using silica gel plates (Kieselgel 60 F254). Com-
pounds were visualized by UV irradiation and/or spraying with
water solution of potassium permanganate KMnO4, followed
by charring at 150 ꢀC. Column chromatography was performed
1
on silica gel 60 M (0.040-0.063 mm). H NMR spectra were

228 Journal of Medicinal Chemistry, 2011, Vol. 54, No. 1
Topalis et al.
General Procedure for Deprotection of N3-Benzoylated-N1-
crotyluracil Analogues. The debenzoylation of N3-benzoyl-N1-
crotyl-C5-substituted uracil (100 mg) was carried out at 4 ꢀC
overnight in a methanol solution 7 N of ammonia (5 mL). After
evaporation of all the volatiles, the residue was purified by
chromatography on silica gel (petroleum ether/EtOAc 50:50) to
yield the desired compound, as a white solid. Following this
general procedure, analogues 4a-e were prepared and charac-
terized as follows.
[MþH]þ 393.1216, found 393.1232. 1H NMR (400 MHz,
CDCl3): δ 7.87 (m, 2H), 7.61 (m, 1H), 7.46 (m, 2H), 7.09 (d,
J = 1.2 Hz, 1H), 5.64-5.79 (m, 2H), 4.29 (m, 2H), 3.71 (d, J =
10.9 Hz, 6H), 2.61 (dd, J = 5.9, 21.4 Hz, 2H), 1.91 (d, J = 1.2
Hz, 3H). 31P NMR (162 MHz, CDCl3): δ 28.552.
N3-Benzoyl-N1-[(Z)-4-dimethoxyphosphonyl-but-2-enyl]-5-fluoro-
uracil (10b). Yield: 17%. HRMS (ESI): m/z calcd for C17H18-
N2O6FNaP [MþH]þ 419.0784, found 419.0763. 1H NMR (400
MHz, CDCl3): δ 7.82-7.90 (m, 3H), 7.63 (tt, J = 2.0, 12.4 Hz,
1H), 7.51 (t, J = 12.4 Hz, 2H), 5.60-5.81 (m, 2H), 4.41 (dd, J =
6.0, 10.8 Hz, 2H), 3.72 (d, J = 17.6 Hz, 6H), 2.69 (dd, J = 11.2,
36.0 Hz, 2H). 31P NMR (162 MHz, CDCl3): δ 34.32.
N1-Crotyl-5-fluorouracil (4b). Yield: 89% (major/minor ratio =
70:30). Mp = 90 ꢀC. 1H NMR (400 MHz, CDCl3): δ 10.96 (s,
1H), 7.35 (d, J = 5.7 Hz, 1H major), 7.27 (d, J = 5.7 Hz, 1H
minor), 5.89-5.77 (m, 1H), 5.54-5.45 (m, 1H), 4.40 (d, J =
7.2 Hz, 2H minor), 4.30 (d, J = 6.6 Hz, 2H major), 1.77-1.73
(m, 3H).
N3-Benzoyl-N1-[(Z)-4-dimethoxyphosphonylbut-2-enyl]-thymine
(10e). Yield: 13%. HRMS (ESI): m/z calcd for C18H21N2O6NaP
[MþH]þ 415.1035, found 415.1040. 1H NMR (250 MHz,
CDCl3): δ 7.90 (dd, J = 1.2, 8.2 Hz, 2H), 7.62 (m, 1H), 7.47
(t, J = 8.2 Hz, 2H), 7.34 (d, J = 0.8 Hz, 1H), 5.63-5.79 (m, 2H),
4.43 (m, 2H), 3.74 (d, J = 10.9 Hz, 6H), 2.74 (dd, J = 6.9, 22.3
Hz, 2H), 1.93 (s, 3H). 31P NMR (162 MHz, CDCl3): δ 28.557.
N3-Benzoyl-N1-[(E)-5-diethoxyphosphonyl-pent-2-enyl]-uracil
(13a). Yield: 86%. HRMS (ESI): m/z calcd for C20H25N2O6NaP
[MþH]þ 443.1348, found 443.1352. 1H NMR (250 MHz,
CDCl3): δ 7.91 (d, J = 7.5 Hz, 2H), 7.64 (t, J = 7.5 Hz, 1H),
7.48 (t, J = 7.5 Hz, 2H), 7.30 (d, J = 8.0 Hz, 1H), 5.88-5.76 (m,
1H), 5.80 (d, J = 8.0 Hz, 1H), 5.57 (dt, J = 6.4, 15.2 Hz, 1H),
4.31 (d, J = 6.4 Hz, 2H), 4.15-4.02 (m, 4H), 2.44-2.35 (m, 2H),
1.86-1.77 (m, 2H), 1.31 (t, J = 6.8 Hz, 6H). 31P NMR (162
MHz, CDCl3): δ 30.57.
N1-Crotyl-5-chlorouracil (4c). Yield: 99% (major/minor ratio =
70:30). Mp = 142 ꢀC. 1H NMR (250 MHz, CDCl3): δ 11.43 (s,
1H), 7.49 (s, 1H major), 7.39 (s, 1H minor), 5.93-5.73 (m, 1H),
5.56-5.28 (m, 1H), 4.40 (d, J = 7.5 Hz, 2H minor), 4.28 (d, J =
7.5 Hz, 2H major), 1.79-1.74 (m, 3H). HRMS (ESI): m/z calcd
for [MþH]þ C8H9ClN2O2 200.6240, found 200.6242.
N1-Crotyl-5-bromouracil (4d). Yield: 99% (major/minor ratio =
70:30). Mp = 144 ꢀC. 1H NMR (250 MHz, CDCl3): δ 8.96 (s,
1H), 7.51 (s, 1H major), 7.48 (s, 1H minor), 5.90-5.75 (m, 1H),
5.56-5.43 (m, 1H), 4.42 (d, J = 7.2 Hz, 2H minor), 4.30 (d, J =
6.6 Hz, 2H major), 1.78-1.75 (m, 3H). MS (ESI): m/z calcd for
[MþH]þ C8H9BrN2O2 245.0753, found 245.0757.
N1-Crotylthymine (4e). Yield: 92% (major/minor ratio =
70:30). Mp = 110 ꢀC. 1H NMR (400 MHz, CDCl3): δ 10.48 (s,
1H), 7.04 (dd, J = 5.6, 1.1 Hz, 1H), 5.78-5-69 (m, 1H major),
5.78-5-69 (m, 1H minor), 5.56-5.49 (m, 1H major), 5.45-5.39
(m, 1H minor), 4.57 (d, J = 6.7 Hz, 2H minor), 4.30 (d, J = 6.1
Hz, 2H major), 1.91 (s, 3H), 1.79 (d, J = 6.4 Hz, 3H major), 1.65
(d, J = 6.1 Hz, 3H minor).
N3-Benzoyl-N1-[(E)-5-diethoxyphosphonyl-pent-2-enyl]-5-fluoro-
uracil (13b). Yield: 78%. HRMS (ESI): m/z calcd for C20H24-
FN2O6NaP [MþH]þ 461.1254, found 461.1243. 1H NMR
(250 MHz, CDCl3): δ 7.95 (d, J = 7.2 Hz, 2H), 7.70 (t, J =
7.6 Hz, 1H), 7.54 (t, J = 7.6 Hz, 2H), 7.48 (d, J = 5.6 Hz, 1H),
5.82 (dt, J = 6.4, 15.2 Hz, 1H), 5.52 (dt, J = 5.2, 16.8 Hz, 1H),
4.34 (d, J = 6.4 Hz, 2H), 4.08-4.20 (m, 4H), 2.38-2.55 (m, 2H),
182-1.92 (m, 2H), 1.27 (t, J = 7.2 Hz, 6H). 31P NMR (162
MHz, CDCl3): δ 30.43.
General Procedure for Cross Coupling Metathesis.14. A mix-
ture of N3-benzoyl-N1-crotyl-C5-substituted uracil, an appro-
priate alkenyl phosphonate [diethyl vinylphosphonate (4 equiv),
dimethyl allylphosphonate (4 equiv), or diethyl but-2-enylpho-
sphonate (2 equiv)] and [Ru] = Grubbs second generation
catalyst (5% mol) was stirred in dry CH2Cl2 (5-10 mL) at
40 ꢀC for 16 h under positive pressure of dry argon. After evapo-
ration of all volatiles, the residue was purified by chromatogra-
phy on silica gel with an elution gradient of EtOAc/MeOH to
yield the desired compound as an oil. Following this general
procedure, analogues 5a-e, 7a-e, 10a-e, and 13a-e were
prepared and characterized as follows.
N3-Benzoyl-N1-[(E)-5-diethoxyphosphonyl-pent-2-enyl]-5-chloro-
uracil (13c). Yield: 69%. HRMS (ESI): m/z calcd for C20H24-
ClN2O6NaP [MþH]þ 477.0958, found 477.0937. 1H NMR (250
MHz, CDCl3): δ 7.89 (d, J = 7.5 Hz, 2H), 7.65 (t, J = 7.5 Hz,
1H), 7.57 (s, 1H), 7.53-7.45 (m, 2H), 5.86 (dt, J = 6.5, 15.0 Hz,
1H), 5.55 (dt, J = 6.5, 15.0 Hz, 1H), 4.31 (d, J = 6.5 Hz, 2H),
4.14-4.02 (m, 4H), 2.46-2.32 (m, 2H), 1.88-1.75 (m, 2H), 1.30
(t, J = 7.0 Hz, 6H). 31P NMR (162 MHz, CDCl3): δ 33.53.
N3-Benzoyl-N1-[(E)-5-diethoxyphosphonyl-pent-2-enyl]-5-
bromouracil (13d). Yield: 80%. HRMS (ESI): m/z calcd for
C20H24BrN2NaO6P [MþH]þ 521.0480, found 521.0483. 1H
NMR (400 MHz, CDCl3): δ 7.87 (d, J = 7.2 Hz, 2H), 7.67
(s, 1H), 7.63 (t, J = 7.2 Hz, 1H), 7.46 (t, J = 7.5 Hz, 2H), 5.85
(dt, J = 6.4, 15.2 Hz, 1H), 5.53 (dt, J = 6.4, 15.2 Hz, 1H), 4.29
(d, J = 6.4 Hz, 2H), 4.13-4.00 (m, 4H), 2.41-2.32 (m, 2H),
1.84-1.76 (m, 2H), 1.29 (t, J = 7.0 Hz, 6H). 31P NMR (162
MHz, CDCl3): δ 30.49.
N1-[(E)-3-Diethoxyphosphonyl-prop-2-enyl]-5-fluorouracil (5b).
Yield: 47%. HRMS (ESI): m/z calcd for C11H17N2O5FP
[MþH]þ 307.0859, found 307.0850. 1H NMR (250 MHz,
CDCl3): δ 7.80 (d, 1H, J = 6.2 Hz, H6), 6.73 (ddt, 1H, J =
3
2
0
0
4.8, 17.5, and 22.0 Hz, H ), 5.88 (tt, 1H, J = 1.8, 17.5 Hz, H ),
1
0
4.50 (m, 2H, H ), 4.04-4.11 (m, 4H, O-CH2-CH3), 1.31 (t, 6H
J = 7.0 Hz, O-CH2-CH3). 31P NMR (161.9 MHz, CDCl3): δ
30.37.
N1-[(E)-3-Diethoxyphosphonyl-prop-2-enyl]-thymine (5e). Yield:
54%. HRMS (ESI): m/z calcd for C12H20N2O5P [MþH]þ
N3-Benzoyl-N1-[(E)-5-diethoxyphosphonyl-pent-2-enyl]-thymine
(13e). Yield: 64%. HRMS (ESI): m/z calcd for C21H27N2NaO6P
[MþH]þ 457.1504, found 457.1501. 1H NMR (250 MHz,
CDCl3): δ 7.90 (d, J = 7.6 Hz, 2H), 7.63 (t, J = 7.6 Hz, 1H),
7.48 (t, J = 7.6 Hz, 2H), 7.12 (s, 1H), 5.73 (dt, J = 6.8, 15.2 Hz,
1H), 4.30 (d, J = 6.4 Hz, 2H), 5.52 (dt, J = 6.4, 15.2 Hz, 1H),
4.03-4.17 (m, 4H), 2.34-2.78 (m, 2H), 1.95 (s, 3H), 1.78-1.89
(m, 2H), 1.32 (t, J = 7.2 Hz, 6H). 31P NMR (162 MHz, CDCl3):
δ 30.63.
General Procedure for the Deprotection of Benzoyl Group.
Benzoylated compounds 7a-e, 10a-e, and 13a-e, respectively,
were stirred in methanolic ammonia (7 N, 100 equiv) at room
temperature during 14 h. The solvent was removed under
reduced pressure, and the crude residue was purified, using flash
chromatography with elution gradient of AcOEt/MeOH to
yield the desired compounds as amorphous solid. Following
1
303.1110, found 303.1109. H NMR (400 MHz, CDCl3): δ 9.31
(s, 1H), 6.89 (d, J = 1.2 Hz, 1H), 6.68 (ddt, J = 4.8, 17.2, and 21.6
Hz, 1H), 5.74 (tt, J=1.8, 17.6 Hz, 1H), 4.44 (ddd, J=1.8, 2.8, and
4.8 Hz, 2H), 4.02-4.10 (m, 4H), 1.86 (d, J = 1.2 Hz, 3H), 1.29 (t,
J = 7.2 Hz, 6H). 31P NMR (162 MHz, CDCl3): δ 26.18.
N3-Benzoyl-N1-[(E)-4-dimethoxyphosphonylbut-2-enyl]-5-fluoro-
uracil (7b). Yield: 72%. HRMS (ESI): m/z calcd for C17H18-
N2O6FNaP [MþH]þ 419.0784, found 419.0769. 1H NMR (400
MHz, CDCl3): δ 7.94-8.00 (m, 3H), 7.63 (t, J = 7.2 Hz, 1H),
7.51 (t, J = 8.0 Hz, 2H), 5.72-5.85 (m, 2H), 3.74 (m, 2H), 3.72
(d, J = 11.2 Hz, 6H), 2.74 (dd, J = 6.0, 22.0 Hz, 2H). 31P NMR
(162 MHz, CDCl3): δ 34.25.
N3-Benzoyl-N1-[(E)-4-dimethoxyphosphonylbut-2-enyl]-thymine
(7e). Yield: 60%. HRMS (ESI): m/z calcd for C18H22N2O6P

Article
Journal of Medicinal Chemistry, 2011, Vol. 54, No. 1 229
this general procedure, analogues 8a-e, 11a-e, and 14a-e were
prepared and characterized as follows.
1.90 (m, 2H), 1.30 (t, J = 7.2 Hz, 6H). 31P NMR (162 MHz,
CDCl3): δ 30.83.
N1-[(E)-4-Dimethoxyphosphonylbut-2-enyl]-uracil (8a). Yield:
89%. 1H NMR (400 MHz, CDCl3): δ 9.55 (s, 1H), 7.13 (d, J =
8.0 Hz, 1H), 5.68-5.62 (m, 3H), 4.28 (m, 2H), 3.68 (d, J = 11.0
Hz, 6H), 2.65 (m, 2H). 31P NMR (162 MHz, CDCl3): δ 29.28.
N1-[(E)-4-Dimethoxyphosphonylbut-2-enyl]-5-fluorouracil (8b).
Yield: 92%. HRMS (ESI): m/z calcd for C10H14N2FNaO5P
[MþH]þ 315.0522, found 315.0540. 1H NMR (400 MHz,
CDCl3): δ 9.55 (s, 1H), 7.28 (d, J = 5.6 Hz, 1H), 5.81-5.64
(m, 2H), 4.31 (m, 2H), 3.75 (d, J = 10.8 Hz, 6H), 2.65 (dd, J =
6.8, 22.0 Hz, 2H). 31P NMR (162 MHz, CDCl3): δ 28.53.
N1-[(E)-4-Dimethoxyphosphonylbut-2-enyl]-5-chlorouracil (8c).
Yield: 90%. 1H NMR (400 MHz, CDCl3): δ 9.80 (s, 1H), 7.42 (d,
1H), 5.83-5.64 (m, 2H), 4.31 (m, 2H), 3.75 (d, J = 10.8 Hz, 6H),
2.66 (dd, J = 7.2, 22.0 Hz, 2H). 31P NMR (162 MHz, CDCl3): δ
28.55.
N1-[(E)-5-Diethoxyphosphonyl-pent-2-enyl]-5-chlorouracil (14c).
Yield: 86%. HRMS (ESI): m/z calcd for C13H20ClN2O5NaP
[MþNa]þ 373.0696, found 373.0690. 1H NMR (250 MHz,
CDCl3): δ 7.41 (s, 1H), 5.80 (dt, J = 6.4, 15.2 Hz, 1H), 5.53
(dt, J = 6.8, 15.2 Hz, 1H), 4.27 (d, J = 6.4 Hz, 2H), 4.40-4.16
(m, 4H), 2.30-2.69 (m, 2H), 1.75-1.99 (m, 2H), 1.28 (t, J = 7.2
Hz, 6H). 31P NMR (162 MHz, CDCl3): δ 30.66.
N1-[(E)-5-Diethoxyphosphonyl-pent-2-enyl]-5-bromouracil (14d).
Yield: 76%. HRMS (ESI): m/z calcd for C13H20BrN2O5NaP
[MþNa]þ) 417.0191, found 417.0181. 1H NMR (250 MHz,
CDCl3): δ 8.69 (br s, 1H), 7.54 (s, 1H), 5.85 (dt, J = 6.4, 15.2
Hz, 1H), 5.55 (dt, J = 6.4, 15.2 Hz, 1H), 4.31 (d, J = 6.4 Hz,
2H), 4.12-4.04 (m, 4H), 2.45-2.37 (m, 2H), 1.88-1.80 (m, 2H),
1.32 (t, J = 7.0 Hz, 6H). 31P NMR (162 MHz, CDCl3): δ 30.49.
N1-[(E)-5-Diethoxyphosphonyl-pent-2-enyl]-thymine (14e). Yield:
87%. HRMS (ESI): m/z calcd for C14H23N2O5NaP [MþNa]þ
353.1242, found 353.1231. 1H NMR (250 MHz, CDCl3): δ 9.90 (br
s, 1H), 7.00 (s, 1H), 5.72 (dt, J=6.3, 15.5 Hz, 1H), 5.51 (dt, J=6.3,
15.5 Hz, 1H), 4.25 (d, J = 6.3 Hz, 2H), 4.19-4.00 (m, 4H),
2.63-2.31 (m, 2H), 1.94-1.72 (m, 2H), 1.88 (s, 3H), 1.29 (t, J =
7.0 Hz, 6H). 31P NMR (162 MHz, CDCl3): δ 30.81.
General Procedure for the Deprotection of Phosphonate Die-
sters. To a CH2Cl2 (5 mL) solution of phosphonate diester,
TMSBr (6 equiv) was added and stirred at room temperature
until starting material disappeared on TLC. MeOH (5 mL) was
added and evaporated with heating (ca. 60 ꢀC). MeOH (5 mL)
was added again, and this procedure was repeated three times.
The residue was extracted with ultrapure H2O and CH2Cl2, and
the inorganic phase was evaporated to yield the expected
compound as an amorphous solid. Following this general
procedure, analogues 6a-e, 9a-e, 12a-e, 15a-e were prepared
and characterized as follows.
N1-[(E)-4-Dimethoxyphosphonylbut-2-enyl]-5-bromouracil (8d).
Yield: 94%. HRMS (ESI): m/z calcd for C10H15N2O5PBr
[MþH]þ 352.9902, found 352.9894. 1H NMR (400 MHz,
CDCl3): δ 8.96 (s, 1H), 7.52 (s, 1H), 5.83-5.65 (m, 2H), 4.35
(dd, J = 4.4, 5.6 Hz, 2H), 3.77 (d, J = 10.8 Hz, 6H), 2.59 (dd,
J = 7.2, 22.0 Hz, 2H). 31P NMR (162 MHz, CDCl3): δ 28.37.
N1-[(E)-4-Dimethoxyphosphonyl-but-2-enyl]-thymine (8e). Yield:
95%. HRMS (ESI): m/z calcd for C11H18N2O5P [MþH]þ
289.0953, found 289.0942. 1H NMR (400 MHz, CDCl3): δ 10.03
(s,1H),7.04(d,J=4.9 Hz, 1H), 5.71-5.73 (m, 2H), 4.50 (t, J=4.8
Hz, 2H), 4.09 (d, J = 10.3 Hz, 6H), 2.59 (dd, J = 6.0, 22.2 Hz, 2H),
1.90 (s, 3H). 31P NMR (162 MHz, CDCl3): δ 29.55.
N1-[(Z)-4-Dimethoxyphosphonylbut-2-enyl]-uracil (11a). Yield:
92%. 1H NMR (400 MHz, CDCl3): δ 8.63 (s, 1H), 7.45 (d, J =
8.0 Hz, 1H), 5.65-5.75 (m, 3H), 4.44 (dd, J = 4.0, 6.4 Hz, 2H),
3.76 (d, J = 11.2 Hz, 6H), 2.65 (dd, J = 7.2, 22.4 Hz, 2H).
31P NMR (162 MHz, CDCl3): δ 28.53.
N1-[(Z)-4-Dimethoxyphosphonylbut-2-enyl]-5-fluorouracil (11b).
Yield: 89%. HRMS (ESI): m/z calcd for C10H14N2FO6P
[MþH]þ 293.0713, found 293.0703. 1H NMR (400 MHz,
CDCl3): δ 9.78 (s, 1H), 7.65 (d, J = 6.0 Hz,1H), 5.79-5.62
(m, 2H), 4.43 (m, 2H), 3.78 (d, J = 10.8 Hz, 6H), 2.74 (dd,
J = 8.0, 22.8 Hz, 2H). 31P NMR (162 MHz, CDCl3): δ 28.43.
N1-[(Z)-4-Dimethoxyphosphonylbut-2-enyl]-5-chlorouracil (11c).
Yield: 90%. 1H NMR (400 MHz, CDCl3): δ 8.67 (s, 1H), 7.67 (s,
1H), 5.83-5.63 (m, 2H), 4.46 (dd, J = 3.6, 6.4 Hz, 2H), 3.78 (d,
J = 12.0 Hz, 6H), 2.75 (dd, J = 8.0, 22.8 Hz, 2H). 31P NMR
(162 MHz, CDCl3): δ 28.22.
N1-[(E)-3-Dihydroxyphosphonyl-prop-2-enyl]-5-fluoro-uracil (6b).
Yield: 94%. HRMS (ESI): m/z calcd for C7H9FN2O5P [MþH]þ
251.0233, found 251.0251. 1H NMR (400 MHz, CD3OD): δ 7.82
(d, J = 6.0 Hz, 1H), 6.64 (ddt, J = 4.8, 17.6, 22.0 Hz, 1H), 5.94
(t, J = 17.6 Hz, 1H), 4.49 (m, 2H). 31P NMR (161.9 MHz,
CD3OD): δ 13.82. Purity >99%. HPLC tR = 2 min 15 s; aceto-
nitrile/H2O (5:95).
N1-[(E)-3-Dihydroxyphosphonyl-prop-2-enyl]-thymine (6e). Yield:
85%. HRMS (ESI): m/z calcd for C8H12N2O5P [MþH]þ 247.0484,
found 247.0475. 1H NMR (400 MHz, CD3OD): δ 7.39 (s, 1H), 6.73
(ddt, J = 4.5, 17.5, 22.0 Hz, 1H), 5.85 (t, J = 17.5 Hz, 1H), 4.49
(m, 2H), 1.87 (s, 3H). 31P NMR (161.9 MHz, CD3OD): δ 13.99.
Purity >97%. HPLC tR = 2 min 8 s; acetonitrile/H2O (5:95).
N1-[(E)-4-Dihydroxyphosphonyl-but-2-enyl]-5-fluorouracil (9b).
Yield: 95%. HRMS (ESI): m/z calcd for C8H10N2O5NaPF
[MþNa]þ 287.0209, found 287.0197. 1H NMR (250 MHz,
CD3OD): δ 7.77 (d, J = 6.3 Hz, 1H), 5.67-5.86 (m, 2H), 4.32
(m, 2H), 2.63 (dd, J = 6.0, 21.5 Hz, 2H0). 31P NMR (161.9 MHz,
CD3OD): δ 24.15. Purity >98%. HPLC tR = 2 min 7 s; aceto-
nitrile/H2O (5:95).
N1-[(Z)-4-Dimethoxyphosphonyl-but-2-enyl]-5-bromouracil (11d).
HRMS (ESI): m/z calcd for C10H15N2O5PBr [MþH]þ 352.9902,
found 352.9891. 1H NMR (400 MHz, CD3OD): δ 8.03 (s, 1H),
5.69-5.84 (m, 2H), 4.45 (dd, J = 3.0, 5.3 Hz, 2H), 3.77 (d, J =
10.9 Hz, 6H), 2.59 (dd, J = 6.8, 23.0 Hz, 2H). 31P NMR (162
MHz, CD3OD): δ 30.36.
N1-[(Z)-4-Dimethoxyphosphonyl-but-2-enyl]-thymine (11e). Yield:
95%. HRMS (ESI): m/z calcd for C11H18N2O5P [MþH]þ 289.0953,
found 289.0939. 1H NMR (400 MHz, CD3OD): δ 7.44 (d, J = 0.8
Hz, 1H), 5.64-5.76(m,2H),4.42(m,2H),3.77(d,J=10.8 Hz, 6H),
2.93 (dd, J = 7.2, 22.8 Hz, 2H), 1.90 (d, J = 0.8 Hz, 3H). 31P NMR
(162 MHz, CD3OD): δ 34.46.
N1-[(E)-4-Dihydroxyphosphonyl-but-2-enyl]-thymine (9e). Yield:
87%. HRMS (ESI): m/z calcd for C9H14N2O5P [MþH]þ
261.0640, found 261.0639. 1H NMR (400 MHz, CD3OD): δ 7.28
(s, 1H), 5.57-5.70 (m, 2H), 4.30 (m, 2H), 2.49 (m, 2H), 1.77 (s,
3H). 31P NMR (161.9 MHz, CD3OD): δ 23.58. Purity >99%.
HPLC tR = 2 min 10 s; acetonitrile/H2O (10:90).
N1-[(E)-5-Diethoxyphosphonyl-pent-2-enyl]-uracil (14a). Yield:
93%. HRMS (ESI): m/z calcd for C13H21N2O5NaP [MþNa]þ
339.1086, found 339.1070. 1H NMR (250 MHz, CDCl3): δ 9.30
(br s, 1H), 7.53 (d, J = 8.0 Hz, 1H), 5.80 (dt, J = 6.4, 15.6 Hz,
1H), 5.63 (d, J = 8.0 Hz, 1H), 5.61 (dt, J = 6.0, 15.6 Hz, 1H0),
4.29 (d, J = 6.4 Hz, 2H), 2.58-2.30 (m, 2H), 1.89-1.75 (m, 2H).
31P NMR (162 MHz, CDCl3): δ 30.74.
N1-[(Z)-4-Dihydroxyphosphonyl-but-2-enyl]-5-fluorouracil (12b).
Yield: 96%. HRMS (ESI): m/z calcd for C8H10N2O5NaPF
[MþNa]þ 287.0209, found 287.0204. 1H NMR (250 MHz,
CD3OD): δ 7.85 (d, J = 6.3 Hz, 1H), 5.62-5.85 (m, 2H), 4.42
(dd, J = 3.4, 6.7 Hz, 2H), 2.78 (dd, J = 7.3, 22.0 Hz, 2H).
31P NMR (161.9 MHz, CD3OD): δ 24.16. Purity>99%. HPLC
tR = 2 min; acetonitrile/H2O (5:95, v/v).
N1-[(E)-5-Diethoxyphosphonyl-pent-2-enyl]-5-fluorouracil (14b).
Yield: 86%. HRMS (ESI): m/z calcd for C13H20FN2O5NaP
[MþNa]þ 357.0992, found 357.1009. 1H NMR (250 MHz,
CDCl3): δ 10.33 (br s, 1H), 7.32 (d, J = 5.6 Hz, 1H), 5.82 (dt,
J = 6.6, 15.2 Hz, 1H), 5.52 (dt, J = 6.8, 15.2 Hz, 1H), 4.27 (d,
J = 6.8 Hz, 2H), 4.03-4.17 (m, 4H), 2.31-2.62 (m, 2H), 1.78-
N1-[(Z)-4-Dihydroxyphosphonyl-but-2-enyl]-5-bromouracil (12d).
HRMS (ESI): m/z calcd for C8H11N2O5PBr [MþH]þ 324.9589,
found 324.9585. 1H NMR (400 MHz, CD3OD): δ 8.05 (s, 1H),

230 Journal of Medicinal Chemistry, 2011, Vol. 54, No. 1
Topalis et al.
5.65-5.75 (m, 2H), 4.47 (dd, J = 3.2, 6.0 Hz, 2), 2.76 (dd, J =
8.0, 22.4 Hz, 2H). 31P NMR (162 MHz, CD3OD): δ 23.66.
Purity >98%. HPLC tR = 2 min 4 s; acetonitrile/H2O (5:95).
N1-[(Z)-4-Dihydroxyphosphonyl-but-2-enyl]-thymine (12e). Yield:
89%. HRMS (ESI): m/z calcd for C9H14N2O5P [MþH]þ 261.0632,
found 261.0639. 1H NMR (400 MHz, CD3OD): δ 7.46 (d, J = 1.2
Hz, 1H), 5.61-5.80 (m, 2H), 4.43 (m, 2H), 2.49 (m, 2H), 1.86 (d,
J = 1.2 Hz, 3H). 31P NMR (161.9 MHz, CD3OD): δ 23.71. Purity
>99%. HPLC tR = 2 min 20 s; acetonitrile/H2O (5:95).
N1-[(E)-5-Dihydroxyphosphonyl-pent-2-enyl]-uracil (15a). Yield:
78%. HRMS (ESI): m/z calcd for C9H12N2O5P [M - H]-
259.0484, found 259.0496. 1H NMR (400 MHz, CD3OD): δ 7.53
(d, J=8.0 Hz, 1H), 5.80 (dt, J=6.4, 15.6 Hz, 1H), 5.63 (d, J=8.0
Hz, 1H), 5.61 (dt, J = 6.0, 15.6 Hz, 1H), 4.29 (d, J = 6.4 Hz, 2H),
2.58-2.30 (m, 2H), 1.89-1.75 (m, 2H). 31P NMR (161.9 MHz,
CD3OD): δ 30.17. Purity >98%. HPLC tR = 2 min 20 s; aceto-
nitrile/H2O (5:95).
[MþNa]þ 511.1821, found 511.1828. 1H NMR (400 MHz,
CD3OD): δ 7.46 (d, J = 1.2 Hz, 1H), 5.78-5.60 (m, 7H), 4.40
(dd, J = 3.8, 5.5 Hz, 2H), 3.01 (dd, J = 7.8, 23.2 Hz, 2H), 1.88
(d, J = 1.1 Hz, 3H), 1.24 (d, J = 3.2 Hz, 18H). 31P NMR (162
MHz, CD3OD): δ 27.6. Purity >96%. HPLC tR = 5 min 44 s;
acetonitrile/H2O (60:40).
N1-[(E)-4-Bis(pivaloyloxymethyl)phosphinyl-2-butenyl]-5-bromo-
uracil (18). Yield: 78%. HRMS (ESI): m/z calcd for C20H30-
N2O9NaPBr [MþNa]þ 575.0770, found 575.0764. 1H NMR
(400 MHz, CD3OD): δ 7.98 (s, 1H), 5.85-5.76 (m, 1H), 5.75-
5.62 (m, 5H), 4.36 (t, J = 5.1 Hz, 2H), 2.82 (dd, J = 6.9, 22.5 Hz,
2H), 1.23 (s, 18H). 31P NMR (162 MHz, CD3OD): δ 27.1.
Purity >96%. HPLC tR = 6 min 13 s; acetonitrile/H2O (60:40).
N1-[(Z)-4-Bis(pivaloyloxymethyl)phosphinyl-2-butenyl]-5-bromo-
uracil (19). HRMS (ESI): m/z calcd for C20H30N2O9NaPBr
[MþNa]þ 575.0770, found 575.0778. 1H NMR (400 MHz,
CD3OD): δ 8.06 (s, 1H), 5.70-5.52 (m, 6H), 4.45 (dd, J =
3.7, 6.8 Hz, 2H), 3.01 (dd, J = 23.4, 7.8 Hz, 2H), 1.24 (s, 18H).
31P NMR (100 MHz, CD3OD) δ 27.5. Purity >95%. HPLC
N1-[(E)-5-Dihydroxyphosphonyl-pent-2-enyl]-5-fluorouracil (15b).
Yield: 90%. HRMS (ESI): m/z calcd for C9H13FN2O5P
[MþH]þ 279.0546, found 279.0550. 1H NMR (400 MHz,
CD3OD): δ 7.78 (d, J = 6.4 Hz, 1H), 5.86 (dt, J = 6.8, 15.2
Hz, 1H), 5.61 (dt, J = 6.4, 15.2 Hz, 1H), 4.30 (d, J = 6.4 Hz,
2H), 2.31-2.62 (m, 2H), 1.78-1.91 (m, 2H). 31P NMR (161.9
MHz, CD3OD): δ 29.65. Purity>99%. HPLC tR = 2 min 16 s;
acetonitrile/H2O (5:95).
t
R = 7 min 1 s; acetonitrile/H2O (60:40).
Protein Purification and Enzymatic Assays. The recombinant
hTMPK and hUCK were prepared as described.19,27 The activ-
ities of hTMPK and hUCK were followed by the coupled
spectrophotometric assay.17,19 Assays were carried out at 37 ꢀC
in a Tris-HCl 50 mM pH = 7.4 buffer containing 50 mM KCl,
5 mM MgCl2, 1 mM ATP, 0.2 mM NADH, 1 mM phosphoenol-
pyruvate, 1 mM DTT, 4 U.μL-1 pyruvate kinase, and 4 U.μL-1
lactate dehydrogenase (total volume: 130 μL). The reaction was
started by adding the (d)NMP or (d)NMP analogue, and the
decrease in absorbance at 340 nm was measured. The kinase
concentrations were 4 nM to 8 μM in order to measure initial
rates below 0.2 ΔA/min. For inhibition studies, the inhibitor was
added after the enzyme, and then the reaction was started. All
experiments were done at least twice.
N1-[(E)-5-Dihydroxyphosphonyl-pent-2-enyl]-5-chlorouracil (15c).
Yield: 85%. HRMS (ESI): m/z calcd for C9H13N2ClO5P
[MþH]þ 295.0251, found 295.0248. 1H NMR (400 MHz,
CD3OD): δ 7.88 (s, 1H) 5.86 (dt, J = 6.4, 15.2 Hz, 1H), 2.32-
2.62 (m, 2H), 5.62 (dt, J = 6.0, 15.6 Hz, 1H), 4.32 (d, J = 6.0 Hz,
2H), 1.75-1.88 (m, 2H). 31P NMR (161.9 MHz, CD3OD): δ
29.32. Purity >95%. HPLC tR = 2 min 34 s; acetonitrile/H2O
(5:95).
N1-[(E)-5-Dihydroxyphosphonyl-pent-2-enyl]-5-bromouracil
(15d). Yield: 84%. HRMS (ESI): m/z calcd for C9H11BrN2O5P
[M-H]- 336.9589, found 336.9577. 1H NMR (400 MHz,
CD3OD): δ 7.97 (s, 1H), 5.86 (dt, J = 6.6, 15.2 Hz, 1H), 5.63
(dt, J = 6.4, 15.2 Hz, 1H), 4.29 (d, J = 6.0 Hz, 2H), 2.62-2.32
(m, 2H), 1.89-1.78 (m, 2H). 31P NMR (161.9 MHz, CD3OD): δ
29.31. Purity >98%. HPLC tR = 2 min 29 s; acetonitrile/H2O
(5:95).
Thymidine kinase assays were done as described28 at 37 ꢀC
during 15 min, in the presence of 5 mM ATP and 1 μM [3H]-T in
a Tris-HCl 50 mM pH = 7.6 buffer containing 2 mM MgCl2,
0.5 mg/mL bovine serum albumin, and 5 mM DTT. The assays
were done in triplicate. The addition of 0.3 mM 5-Cl-UbutP (9d)
decreased TK1 (4 ng)28 and TK2 (1 ng)29 activity by 11% and
13%, respectively (standard deviations 1% and 2.6%).
N1-[(E)-5-Dihydroxyphosphonyl-pent-2-enyl]-thymine (15e). Yield:
90%. HRMS (ESI): m/z calcd for C10H14N2O5P [M-H]- 273.0640,
found 273.0633. 1H NMR (400 MHz, CD3OD): δ 7.41 (d, J = 0.8
Hz, 1H), 5.80 (dt, J = 6.3, 15.4 Hz, 1H), 5.61 (dt, J = 6.0, 15.4 Hz,
1H), 4.29 (d, J = 6.0 Hz, 2H), 2.63-2.30 (m, 2H), 1.90-1.80
(m, 2H), 1.88 (s, 3H). 31P NMR (161.9 MHz, CD3OD): δ 29.41.
Purity >98%. HPLC tR = 2 min 27 s; acetonitrile/H2O (5:95).
General Procedure for the Synthesis of Bis(Pom)-allylphos-
phonates. To a ACN (18 mL) solution of C5-substituted
dimethyl allylphosphonate (8d,e and 11d,e) (2.6 g, 17.3 mmo1)
and anhydrous sodium iodide (5.2 g, 34.6 mmol), chloromethyl
pivalate (6.58 g, 43.3 mmol) was added. This solution was stirred
at reflux for 48 h under positive pressure of dry Ar. After cooling,
170 mL of diethyl ether was added to this mixture and washed by
35 mL of water. The organic layer was dried on magnesium
sulfate, evaporated, and purified by chromatography on silica gel
(EtOAc/EP 1:4) to give pure Bis-(POM) allylphosphonate as
slightly yellow oil. Following this general procedure, analogues
16-19 were prepared and characterized as follows.
Crystal Structure Determination. The hTMPK was equili-
brated by dialysis against 50 mM Tris-HCl pH 8.0 buffer
containing 200 mM KCl, 0.5 mM DTT. The crystal screening
was done on a Cartesian robot using Quiagen screens in 96 well
vapor diffusion plates (Greiner). Final crystals were grown at
20 ꢀC mixing 1 μL of 20 mg/mL hTMPK, 8 mM (E)-TbutP (9e),
10 mM ATP, 5 mM MgCl2, and 1 μL of 14% (w/v) PEG 3350,
0.1 M Hepes pH 7.5. Crystals were soaked in the reservoir
solution supplemented with glycerol to a final concentration of
20% (w/v) and frozen at 100 K in liquid nitrogen. Data were
collected on the beamline PROXIMA-1 of SOLEIL synchro-
˚
tron (Gif-sur-Yvette, France) at a wavelength of 0.98 A.
The data reduction and scaling were done using XDS
package.30 The structure was solved by molecular replacement
using Phaser31 and the pdb entry 1e2g as the search model. The
space group is C2 with one molecule per asymmetric unit. The
structure was rebuilt using Coot and refined with Phenix.32,33
Statistics for data reduction and structure refinement are pre-
sented in Supporting Information Table 1. The structure of
human TMP kinase in complex with (E)-TbutP (9e) is being
deposited in the Protein Data Bank.
Antiviral Assays. The herpes and vaccinia virus assays were
based on inhibition of virus-induced cytopathicity in HEL cells
[herpes simplex virus type 1 (HSV-1) (KOS), HSV-2 (G), HSV-1
TK- (KOS acyclovir resistant, ACVr) and vaccinia virus (VV)].
Confluent cell cultures in microtiter 96-well plates were inocu-
lated with 100 CCID50 of virus (1 CCID50 being the virus dose to
infect 50% of the cell cultures). After a 1 h virus adsorption
period, residual virus was removed, and the cell cultures were
N1-[(E)-4-Bis(pivaloyloxymethyl)phosphinyl-buten-2-yl]-thymine
(16). Yield: 47%. HRMS (ESI): m/z calcd for C21H33N2O9NaP
[MþNa]þ 511.1821, found 511.1837. 1H NMR (400 MHz,
CD3OD): δ 7.37 (d, J = 0.9 Hz, 1H), 5.85-5.74 (ddt, J = 5.1,
15.6 Hz, 1H), 5.66 (m, 5H), 4.32 (t, J = 5.2 Hz, 2H), 2.82 (dd,
J = 7.2, 22.5, Hz, 2H), 1.87 (s, 3H) 1.23 (s, 18H). 31P NMR
(161.9 MHz, CD3OD): δ 27.3. Purity >96%. HPLC tR = 5 min
12 s; acetonitrile/H2O (60:40).
N1-[(Z)-4-Bis(pivaloyloxymethyl)phosphinyl-2-butenyl]-thymine
(17). Yield: 46%. HRMS (ESI): m/z calcd for C21H33N2O9NaP

Article
Journal of Medicinal Chemistry, 2011, Vol. 54, No. 1 231
incubated in the presence of varying concentrations (200, 40,
8 μM) of the tested compounds. Viral cytopathicity was recorded
as soon as it reached completion in the control virus-infected cell
cultures that were not treated with the test compounds.
Confluent human embryonic lung (HEL) fibroblasts were
grown in 96-well microtiter plates and infected with the human
cytomegalovirus (HCMV) strains Davis and AD-169 at 100
PFU per well. After a 2 h incubation period, residual virus was
removed and the infected cells were further incubated with
medium containing different concentrations of the test com-
pounds (in duplicate). After incubation for 7 days at 37 ꢀC,
virus-induced cytopathogenicity was monitored microscopi-
cally after ethanol fixation and staining with Giemsa. Antiviral
activity was expressed as the EC50 or compound concentration
required to reduce virus-induced cytopathogenicity by 50%.
EC50 values were calculated from graphic plots of the percentage
of cytopathogenicity as a function of concentration of the
compounds.
The laboratory wild-type VZV strain OKA and the thymidine
kinase-deficient VZV strain 07/1 were used. Confluent HEL
cells grown in 96 well microtiter plates were inoculated with
VZV at an input of 20 PFU per well. After a 2 h incubation
period, residual virus was removed and varying concentrations
of the test compounds were added (in duplicate). Antiviral
activity was expressed as the 50% effective concentration re-
quired to reduce viral plaque formation after 5 days by 50% as
compared with untreated controls.
Cytotoxicity Assays. Cytotoxicity measurements were based
on the inhibition of HEL cell growth. HEL cells were seeded
at a rate of 5 ꢀ 103 cells/well into 96 well microtiter plates and
allowed to proliferate for 24 h. Then, medium containing
different concentrations of the test compounds was added. After
3 days of incubation at 37 ꢀC, the cell number was determined
with a Coulter counter. The 50% cytostatic concentration
(CC50) was calculated as the compound concentration required
reducing cell growth by 50% relative to the number of cells in the
untreated controls. CC50 values were estimated from graphic
plots of the number of cells (percentage of control) as a function
of the concentration of the test compounds. Cytotoxicity was
also expressed as the minimum cytotoxic concentration (MCC)
or the compound concentration that causes a microscopically
detectable alteration of cell morphology.
herpes-, and poxvirus infections: the phosphonate bridge. Biochem.
Pharmacol. 2007, 73, 911–922.
(2) Balzarini, J.; Pannecouque, C.; De Clercq, E.; Aquaro, S.; Perno,
C. F.; Egberink, H.; Holy, A. Antiretrovirus activityof a novel class
of acyclic pyrimidine nucleoside phosphonates. Antimicrob. Agents
Chemother. 2002, 46, 2185–2193.
(3) Votruba, I.; Bernaerts, R.; Sakuma, T.; De Clercq, E.; Merta, A.;
Rosenberg, I.; Holy, A. Intracellular phosphorylation of broad-
spectrum anti-DNA virus agent (S)-9-(3-hydroxy-2-phosphonyl-
methoxypropyl)adenine and inhibition of viral DNA synthesis.
Mol. Pharmacol. 1987, 32, 524–529.
(4) Cihlar, T.; Chen, M. S. Identification of enzymes catalyzing two-
step phosphorylation of cidofovir and the effect of cytomegalovirus
infection on their activities in host cells. Mol. Pharmacol. 1996, 50,
1502–1510.
(5) Robbins, B. L.; Greenhaw, J.; Connelly, M. C.; Fridland, A.
Metabolic pathways for activation of the antiviral agent 9-(2-
phosphonylmethoxyethyl)adenine in human lymphoid cells. Anti-
microb. Agents Chemother. 1995, 39, 2304–2308.
(6) Koch, K.; Chen, Y.; Feng, J. Y.; Borroto-Esoda, K.; Deville-
Bonne, D.; Gallois-Montbrun, S.; Janin, J.; Morera, S. Nucleosides
Nucleotides Nucleic Acids 2009, 28, 776–792.
(7) Tuske, S.; Sarafianos, S. G.; Clark, A. D., Jr.; Ding, J.; Naeger,
L. K.; White, K. L.; Miller, M. D.; Gibbs, C. S.; Boyer, P. L.; Clark,
P.; Wang, G.; Gaffney, B. L.; Jones, R. A.; Jerina, D. M.; Hughes,
S. H.; Arnold, E. Structures of HIV-1 RT-DNA complexes before
and after incorporation of the anti-AIDS drug tenofovir. Nat.
Struct. Mol. Biol. 2004, 11, 469–474.
(8) Magee, W. C.; Hostetler, K. Y.; Evans, D. H. Mechanism of
inhibition of vaccinia virus DNA polymerase by cidofovir diphos-
phate. Antimicrob. Agents Chemother. 2005, 49, 3153–3162.
(9) Magee, W. C.; Aldern, K. A.; Hostetler, K. Y.; Evans, D. H.
Cidofovir and (S)-9-[3-hydroxy-(2-phosphonomethoxy)propyl]-
adenine are highly effective inhibitors of vaccinia virus DNA
polymerase when incorporated into the template strand. Antimi-
crob. Agents Chemother. 2008, 52, 586–597.
(10) Topalis, D.; Alvarez, K.; Barral, K.; Munier-Lehmann, H.;
ꢁ
Schneider, B.; Veron, M.; Guerreiro, C.; Mulard, L.; El-Amri,
C.; Canard, B.; Deville-Bonne, D. Acyclic phosphonate nucleo-
tides and human adenylate kinases: impact of a borano group on
alpha-P position. Nucleosides Nucleotides Nucleic Acids 2008, 27,
319–331.
(11) Topalis, D.; Kumamoto, H.; Amaya Velasco, M. F.; Dugue, L.;
Haouz, A.; Alexandre, J. A.; Gallois-Montbrun, S.; Alzari, P. M.;
Pochet, S.; Agrofoglio, L. A.; Deville-Bonne, D. Nucleotide bind-
ing to human UMP-CMP kinase using fluorescent derivatives a
screening based on affinity for the UMP-CMP binding site. FEBS
J. 2007, 274, 3704–3714.
(12) Topalis, D.; Broggi, J.; Alexandre, J. A.; Pradere, U.; Roy, V.;
Berteina-Raboin, S.; Agrofoglio, L. A.; Deville-Bonne, D. Evalua-
tion of C-5 Substituted uracil acyclic phosphonates as substrates or
inhibitors for dTMP and UMP-CMP kinases and potential anti-
pox derivatives. Antivir. Res. 2008, 78, A65.
(13) Kumamoto, H.; Broggi, J.; Topalis, D.; Pradere, U.; Roy, V.;
Berteina-Raboin, S.; Nolan, S. P.; Deville-Bonne, D.; Snoeck, R.;
Garin, D.; Agrofoglio, L. Preparation of acyclo nucleoside phos-
phonate analogues based-on cross-metathesis. Tetrahedron 2008,
64, 3517–3526.
(14) Chatterjee, A. K.; Choi, T.-L.; Sanders, D. P.; Grubbs, R. H. A
general model for selectivity in olefin cross metathesis. J. Am.
Chem. Soc. 2003, 125, 11360–11370.
(15) Cruickshank, K. A.; Jiricny, J.; Reese, C. B. The benzoylation of
uracil and thymine. Tetrahedron Lett. 1984, 25, 681–684.
(16) Vepsalainen, J. J. Bisphosphonate prodrugs: a new synthetic
strategy to tetraacyloxymethyl esters of methylenebisphospho-
nates. Tetrahedron Lett. 1999, 40, 8491–8493.
(17) Alexandre, J. A.; Roy, B.; Topalis, D.; Pochet, S.; Perigaud, C.;
Deville-Bonne, D. Enantioselectivity of human AMP, dTMP and
UMP-CMP kinases. Nucleic Acid Res. 2007, 35 (14), 4895–4904.
(18) Huang, S. H.; Tang, A.; Drisco, B.; Zhang, S. Q.; Seeger, R.; Li, C.;
Jong, A. Human dTMP kinase: gene expression and enzymatic
activity coinciding with cell cycle progression and cell growth. DNA
Cell Biol. 1994, 13, 461–471.
Acknowledgment. These studies were supported by Uni-
ꢁ
versite Pierre et Marie Curie (UPMC) the Centre National de
Recherche Scientifique (Unite Mixte de Recherche 6005, For-
ꢁ
ꢁ
mation de Recherche en Evolution 2852 and Unite Propre de
Recherche 3082), the Agence Nationale de Recherche (grant
ANR-05-BLAN-0368-02), the Concerted Actions (GOA 05/19)
of the K. U. Leuven and the Fondation pour la Recherche
ꢁ
Medicale (P. Meyer and C. Caillat). We thank Michele Reboud
(UPMC) for laboratory facilities, Sylvie Pochet and Laurence
ꢀ
ꢁ
Dugue (Institut Pasteur, France) for synthesizing the MABA
derivatives, and Leentje Persoons, Lies Van den Heurck, Anita
Camps, and Steven Carmans (Rega Institute, Leuven, Belgium)
for expert assistance in cellular assays.
Supporting Information Available: Structural data from
X-ray of human thymidilate kinase (SI-Table 1). Circular
dichroism spectroscopic data obtained with hUCK and some
selected ANPs compared to natural substrates (SI-Table 2).
Inhibition of human UCK by [E]-5Cl-UbutP (9c) (SI-Figure 1).
This material is available free of charge via the Internet at http://
pubs.acs.org.
ꢁ
(19) Pochet, S.; Dugue, L.; Labesse, G.; Delepierre, M.; Munier-
Lehmann, H. Comparative study of purine and pyrimidine nucleo-
side analogues acting on the thymidylate kinases of Mycobacterium
tuberculosis and of humans. ChemBioChem 2003, 4, 742–747.
(20) Ostermann, N.; Schlichting, I.; Brundiers, R.; Konrad, M.;
Reinstein, J.; Veit, T.; Goody, R. S.; Lavie, A. Insights into the
phosphoryltransfer mechanism of human thymidylate kinase
gained from crystal structures of enzyme complexes along the
reaction coordinate. Structure 2000, 8, 629–642.
References
(1) De Clercq, E. Acyclic nucleoside phosphonates: past, present and
future. Bridging chemistry to HIV, HBV, HCV, HPV, adeno-,

232 Journal of Medicinal Chemistry, 2011, Vol. 54, No. 1
Topalis et al.
(21) Deville-Bonne, D.; El-Amri, C.; Meyer, P.; Chen, Y.; Agrofoglio,
L. A.; Janin, J. Human and viral nucleoside/nucleotide kinases
involved in antiviral drug activation: Structural and catalytic
properties. Antivir. Res. 2010, 86, 101–120.
(22) Chen, Y. L.; Lin, D. W.; Chang, Z. F. Identification of a putative
human mitochondrial thymidine monophosphate kinase asso-
ciated with monocytic/macrophage terminal differentiation. Genes
Cells 2008, 13, 679–689.
(27) Pasti, C.; Gallois-Montbrun, S.; Munier-Lehmann, H.; Veron, 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.
(28) Kosinska, U.; Carnrot, C.; Eriksson, S.; Wang, L.; Eklund, H.
Structure of the substrate complex of thymidine kinase from
Ureaplasma urealyticum and investigations of possible drug tar-
gets for the enzyme. FEBS J. 2005, 272, 6365–6372.
(23) Hiratsuka, T. J. Affinity labeling of the myosin ATPase with
ribose-modified fluorescent nucleotides and vanadate. Biochemis-
try 1984, 96, 147–154.
(29) Wang, L.; Munch-Petersen, B.; Herrstrom Sjoberg, A.; Hellman,
U.; Bergman, T.; Jornvall, H.; Eriksson, S. Human thymidine
kinase 2: molecular cloning and characterisation of the enzyme
activity with antiviral and cytostatic nucleoside substrates. FEBS
Lett. 1999, 443, 170–174.
ꢁ
(24) Hernandez, A.-I.; Balzarini, J.; Karlsson, A.; Camarasa, M.-J.; Perez-
ꢁ
Perez, M.-J. Acyclic nucleotide analogues as novel inhibitors of human
mitochondrial thymidine kinase. J. Med. Chem. 2002, 45, 4254–4263.
(25) (a) Choo, H.; Beadle, J.; Chong, Y.; Trahan, J.; Hostetler, K.
Synthesis of the 5-phosphono-pent-2-en-1-yl nucleosides: A new
class of antiviral acyclic nucleoside phosphonates. Bioorg. Med.
Chem. 2007, 15, 1771–1779. (b) Choo, H.; Beadle, J.; Kern, E.;
Prichard, M.; KA., K.; Hartline, C.; Trahan, J.; Aldern, K.; Korba, B.;
Hostetler, K. Antiviral activities of novel 5-phosphono-pent-2-en-1-yl
nucleosides and their alkoxyalkyl phosphonoesters. Antimicrob.
Agents Chemother. 2007, 51, 611–615.
(30) Kabsch, B. Automatic processing of rotation diffraction data from
crystals of initially unknown symmetry and cell constants. J. Appl.
Crystallogr. 1993, 26, 795–800.
(31) Storoni, L. C.; McCoy, A. J.; Read, R. J. Likelihood-enhanced fast
rotation functions. Acta Crystallogr., Sect. D 2004, 60, 432–438.
(32) Emsley, P.; Cowtan, K. Coot: model-building tools for molecular
graphics. Acta Crystallogr., Sect. D 2004, 60, 2126–2132.
(33) Adams, P. D.; Grosse-Kunstleve, R. W.; Hung, L.-W.; Ioerger,
T. R.; McCoy, A. J.; Moriarty, N. W.; Read, R. J.; Sacchettini,
J. C.; Sauter, N. K.; Terwilliger, T. C. PHENIX: building new
software for automated crystallographic structure determination.
Acta Crystallogr., Sect. D 2002, 58, 1948–1954.
(26) Topalis, D.; Collinet, B.; Gasse, C.; Dugue, L.; Balzarini, J.;
Pochet, S.; Deville-Bonne, D. Substrate specificity of vaccinia virus
thymidylate kinase. FEBS J. 2005, 272, 6254–6265.