3′-(1,2,3-Triazol-1-yl)-3′-deoxythymidine analogs as substrates for human and Ureaplasma parvum thymidine kinase for structure-activity investigations

Article abstract of DOI:10.1016/j.bmc.2010.03.023

The pathogenic mycoplasma Ureaplasma parvum (Up) causes opportunistic infections and relies on salvage of nucleosides for DNA synthesis and Up thymidine kinase (UpTK) provides the necessary thymidine nucleotides. The anti-HIV compound 3?-azido-3′-deoxythy

Full text of DOI:10.1016/j.bmc.2010.03.023

Bioorganic & Medicinal Chemistry 18 (2010) 3261–3269
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
30-(1,2,3-Triazol-1-yl)-30-deoxythymidine analogs as substrates for human
and Ureaplasma parvum thymidine kinase for structure–activity investigations
c
Jay Lin a, , Vincent Roy b, , Liya Wang a, Li You b, Luigi A. Agrofoglio b, , Dominique Deville-Bonne ,
*
Tamara R. McBrayer d,e, Steven J. Coats e, Raymond F. Schinazi d, Staffan Eriksson a,
*
a Dpt. Anatomy, Physiology and Biochemistry, Veterinary Medical Biochemistry, Swedish University Agricultural Sciences, Uppsala, Sweden
b Institut de Chimie Organique et Analytique, CNRS UMR 6005, Université d’Orléans, 45067 Orléans Cedex 2, France
c Laboratoire d’ Enzymologie Moléculaire et Fonctionnelle, Université Pierre et Marie Curie, Paris, France
d Center for AIDS Research, Lab. Biochem. Pharmacol., Dpt. Pediatrics, Emory University School of Medicine and Veterans Affairs Medical Center, Decatur, GA 30033, USA
e RFS Pharma, LLC, 1860 Montreal Road, Tucker, GA 30084, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The pathogenic mycoplasma Ureaplasma parvum (Up) causes opportunistic infections and relies on
salvage of nucleosides for DNA synthesis and Up thymidine kinase (UpTK) provides the necessary thymi-
Received 12 November 2009
Revised 4 March 2010
Accepted 11 March 2010
Available online 15 March 2010
0
´
dine nucleotides. The anti-HIV compound 3-azido-3 -deoxythymidine (AZT) is a good substrate for TK.
Methods for a rapid and efficient synthesis of new 30- -[1,2,3]triazol-30-deoxythymidine analogs from
a
AZT under Huisgen conditions are described. Thirteen 30-analogues were tested with human cytosolic
thymidine kinase (hTK1) and UpTK. The new analogs showed higher efficiencies (Km/Vmax values) in all
cases with UpTK than with hTK1. Still, hTK1 was preferentially inhibited by 9 out of 10 tested analogs.
Structural models of UpTK and hTK1 were constructed and used to explain the kinetic results. Two dif-
ferent binding modes of the nucleosides within the active sites of both enzymes were suggested with
one predominating in the bacterial enzyme and the other in hTK1. These results will aid future develop-
ment of anti-mycoplasma nucleosides.
Keywords:
Thymidine kinase
Mycoplasma
Nucleoside analogs
Microwave
Click-chemistry
Structure function-relationship
AZT
Ó 2010 Elsevier Ltd. All rights reserved.
Ureaplasma parvum
Huisgen reaction
1. Introduction
be incorporated into by DNA polymerases.1–3 The first phosphory-
lation reaction is often rate limiting and TK1 expression is closely
associated with cell proliferation, with a peak in the S-phase of
mammalian cells (for TK1) followed by degradation during mito-
sis.2,4,5 High TK1 levels are found in malignant cells and TK1 has
been used as a tumor marker in cancer diagnostics.6
20-Deoxynucleoside kinases (dNKs) such as thymidine kinases 1
(TK1) are found in most organisms including viruses, bacteria and
mammals. TK primarily activates thymidine (dThd) and 20-deoxy-
´
uridine (dUrd) by phosphorylation of the 5-OH group, forming a
charged nucleotide, which is trapped in the cell. The nucleoside
monophosphates formed, will after further phosphorylation steps
Nucleoside analogs, mimicking the building blocks of DNA, uti-
lize dNKs as activating enzymes in case of several anticancer and
antiviral prodrugs3–5 such as 3-azido-3 -deoxythymidine (AZT)
used in HIV treatment. Its triphosphate form blocks viral reverse
transcriptase and results in discontinued virus replication.7 AZT
also inhibits the growth of gram negative bacteria like Escherichia
coli.8,9
0
´
Abbreviations: ATP, adenosine triphosphate; AZT, 30-azido-thymidine; BSA,
bovine serum albumin; dAK, deoxyadenosine kinase; CuAAC, copper (I) catalyzed
azide-alkyne cycloaddition; DMSO, dimethyl sulfoxide; dThd, thymidine; DTT,
dithiothreital; dUrd, deoxyuridine; hTK1, human thymidine kinase 1; dThd,
thymidine; Up, Ureaplasma parvum; UpTK, Ureaplasma parvum thymidine kinase;
Vv, Vaccinia virus; TLC, thin layer chromatography; NMR, nuclear magnetic
resonance; HRMS, high resolution mass spectrometry; HPLC, high-performance
The hTK1 family together with the other dNK families are key
enzymes in the salvage pathway, whereas de novo nucleotide syn-
thesis requires ribonucleotide reduction as well as other enzymatic
reactions.3,5 Not all organisms have the de novo pathway and the
mycoplasma Ureaplasma parvum (Up), previously called
Ureaplasma urealyticum, is an example of an organism lacking this
pathway and therefore relies entirely on the salvage pathway.10–12
Up is a free-living self-replicating bacteria containing one of the
liquid chromatography; pdb, protein database file; IC50
,
median inhibitory
concentration.
* Corresponding authors. Tel.: +33 2 3849 4582; fax: +33 2 3851 0728 (L.A.A.);
tel.: +46 7099 08059 (S.E.).
E-mail addresses: luigi.agrofoglio@univ-orleans.fr (L.A. Agrofoglio), staffan.er
iksson@afb.slu.se (S. Eriksson).
These authors equally contributed to this work.
0968-0896/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2010.03.023

3262
J. Lin et al. / Bioorg. Med. Chem. 18 (2010) 3261–3269
smallest genomes known to date.10,12–14 and is found in up to 80%
of all adults.12–14 Up colonizes the urogenital tract and has the abil-
ity to harvest ATP from urea.10,13,14 Up is associated with preg-
nancy complications such as infertility, altered sperm motility
and pneumonia in the neonate. This type of infection could result
in chronic lung disease with severe scaring on lung tissue causing
an increased mortality.14,15 The range of potential antibiotics
against Up is limited because of the absence of a cell wall and lim-
ited biosynthetic pathways compared to other organisms.10–12,16 In
chemical reactions,35 including some alkyne-azide cycloadditions,
with substantial decreases in reaction times. The 1,3-dipolar cyclo-
addition reaction was conducted with microwave activation in a
water/tert-butanol (1:1) solvent mixture at 125 °C catalyzed by
36
Cu(0)/CuSO4 and at room temperature for compounds 7, 10
37
and 11, catalyzed by sodium ascorbate/CuSO4 (Fig. 1).
Both methods generate the in situ Cu(I) catalyst to form the de-
sired compounds with a complete conversion of starting materiel,
in a fully regioselective (1,4) manner, without contamination by
the 1,5-regioisomer. To provide a starting point for a detailed
structure–activity study, diverse alkyne substituents, such as alco-
hol, alkyl, aromatic, and halogen derivatives, were utilized. The
reactions were monitored by TLC.
All compounds 1–13 (Fig. 2) were isolated in high yields, rang-
ing from 81% to 96%. Complete consumption of AZT occurred under
microwave irradiation in 2–120 min reactions, depending on the
substituents, and was followed by 12 h at room temperature. The
regioselectivity of the ligation leading to 1,4-disubstitued-[1,2,3]-
triazole moiety was confirmed by NMR using 1H, 13C long range
correlation spectra (HMBC). It is interesting to note that the reac-
tion leading to compound 1, which has been previously reported
to need 10 h,28c occurred under our optimized conditions within
10 min. Compounds 1, 9, 12 and 13 have already been reported
to be devoid of significant antiviral activity.28 They were included
in our study to investigate the structure–activity relationship with
the key bacterial enzyme UpTK.
Mycoplasma/Ureaplasma two dNKs are present,
a TK and a
dAK,10,11,13,16 making these important medicinal target enzymes,
since these salvage enzymes are essential for DNA precursor syn-
thesis and their inhibition would stop bacterial growth. Up-TK
has previously been crystallized and characterized as strictly
pyrimidine specific.11 Some nucleoside analogs were evaluated as
potential inhibitors of UpTK.11,17 Human TK1 has also been crystal-
lized in parallel to Up-TK revealing a similar 3D structure,17–19 de-
spite the relatively low amino acid similarity (30%). Some
differences in the specificities of Up-TK and hTK1 were detected,
particularly with substitutions at the 5- or 3-position of the base
and 30-position of the sugar.11,17
AZT is well accepted as substrate by the TK1 family11,20 and tria-
zol-derivates have been studied intensively for many years due to
their high reactivity, ease of preparation, anti-microbial, antiviral,
anti-inflammatory and anti-tumor properties.21–27 Some inactive
and nontoxic 30(1,2,3-triazol-1-yl)-30-deoxythymidines28 were al-
ready synthesized but their phosphorylation mechanisms have
not been studied.
´
2.1.1. In vitro activity of the 3-substituted analogs with hTK1
Here we describe the synthesis and an enzymatic characteriza-
tion of new 30-substituted thymidine analogues, starting from AZT
forming [1,2,3]-triazoles via copper(I) catalysed reactions. Pharma-
comodulation has become central to drug discovery and has played
a major role in the search for new treatments of viral diseases.
However, the discovery and process optimization of potential
agents is often slow, expensive and involves complex synthetic
schemes. The ‘click chemistry’ proposed by Sharpless et al.29 has
emerged as a fast and efficient approach to simplify compound
synthesis. The Huisgen 1,3-dipolar cycloaddition of azides and ter-
minal alkynes is one of the best known and powerful click reac-
tions,30 which is compatible with microwave activation and can
offer a rapid method to aid in the drug discovery process.31
Structural models of UpTK and hTK1 in complex with several of
the new analogues based on the crystal structures of both en-
zymes18,19 were built and used in this study to explain the differ-
ences in substrate selectivity and catalytic rates. The result
increase our understanding of the mechanisms of TK enzymes
and may help in future design of selective and efficient anti-Up
agents. Antibiotic resistance in pathogenic bacteria has become a
major public-health risk, therefore alternative antibiotics with
new modes of action are needed and nucleoside analogues have
the potential to be one of these new antibiotics.17,32,33
and UpTK
The phosphorylation of different nucleoside analog by Up-TK
and hTK1 is reported in Table 1. The activity with 100 lM ATP
and dThd was set to 100%. All compounds 1–13 are similarly phos-
phorylated by hTK1 at 100 M, (Table 1).
The efficiency of phosphorylation was in general about two
times higher with UpTK compared to hTK1 and particularly for 6
and 8, 10–13 (Tables 1 and S1). More specifically, 13 was five times
more potent with UpTK than hTK1. Compared to dThd, AZT showed
28% relative activity with UpTK and 47% with hTK1, as previously
reported.17 Enzyme kinetic studies with hTK1 and UpTK were done
using [c-
32P]-ATP transfer assay with varying concentrations of
nucleosides. The kinetic results (Table S1) appeared to follow
Michaelis Menten kinetics (Fig. S1) and the natural substrate dThd
showed as expected the lowest Km and highest Vmax values with
both enzymes.
AZT was a relatively good substrate for both enzymes (50%
compared to dThd). Compounds 4–13 were in all cases more effi-
ciently phosphorylated with UpTK than with hTK1. While com-
pound 6 showed the overall lowest Km values with UpTK, the
largest relative difference was found with 11, 10-times higher with
UpTK than hTK1 (Fig. 3 and Table S1).
However, the overall relative efficiency was low and in the best
cases it reached about 1% of dThd (set to 100%). Compound 6
2. Results and discussion
2.1. Results
O
R
O
NH
sodium ascorbate/CuSO4
NH
or
Compounds containing a [1,2,3]-triazole moiety are often asso-
ciated with interesting biological activities, such as anti-microbial,
antiviral and anti-proliferative effects. Since the discovery of cop-
per(I) as catalyst of Huisgen’s 1,3-dipolar cycloadditions, this reac-
tion has been shown to be highly specific, irreversible,
regioselective and chemo-selective.29b Here we report the synthe-
sis of a several AZT analogues bearing a 1,2,3-triazole moiety at
30-position under microwave-assisted CuAAC.34 Microwave heat-
ing is known as a powerful technique to promote a variety of
O
HO
N
O
Cu(0) / CuSO4, MW
O
HO
N
O
tBuOH/H2O, 125°C
N
N
N
N3
R
1-13
AZT
Figure 1. A scheme summarizing the synthesis of AZT derived triazoles via a
CuAAC.

J. Lin et al. / Bioorg. Med. Chem. 18 (2010) 3261–3269
3263
HO
HO
HO
HO
T
T
T
T
O
O
O
O
N
N
N
N
N
N
N
N
N
N
N
N
O
OH
O
OH
1
2
3
4
O
HO
T
HO
HO
HO
O
T
T
T
O
O
O
N
N
N
N
N
N
N
N
N
N
N
N
Cl
( )4
6
5
7
8
HO
HO
HO
HO
T
T
T
T
HO
T
O
O
O
O
O
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
F
9
10
11
12
13
Figure 2. Synthesized compounds.
Table 1
9, 12–13, consisting of various phenyl-analogues had much higher
Km values (30–40 M) compared with the other compounds. The
Vmax values of UpTK for these compounds were 3–4 times lower
than with AZT and the Km values with hTK1 were in general higher
than UpTK except in case of 13, containing a phenyl-fluorine moi-
Nucleoside analog phosphorylation by UpTK and hTK1. The activity with 100 M ATP
and nucleosides, respectively, are presented with dThd activity set to 100%
Compounds
Relative activity (%)
UpTK
hTK1
ety. The Km values for 13 were 10 M and 29
UpTK, respectively.
lM with hTK1 and
dThd
AZT
1
100
100
27.8 1.2
47.3 0.4
Nda
Nda
2
3
4
5
6
7
8
9
10
11
12
13
4.1 0.2
1.6 0.2
3.9 0.2
6.5 0.1
7.2 0.9
4.7 0.7
8.7 1.1
5.5 0.5
11.6 1.2
12.3 1.8
7.1 1.2
13.7 1.1
2.2 0.1
1.3 0.01
2.4 0.1
2.7 0.2
1.9 0.04
2.2 0.1
2.9 0.1
2.1 0.1
4.8 0.1
2.4 0.1
3.1 0.04
2.6 0.2
´
2.1.2. Inhibition studies with the 3-dThd substituted analogs
The IC50-values of AZT and compounds 4–13 for hTK1 and UpTK
were determined (Table 2).
In case of 12 and 13, the lowest IC50-values were 40–50 M with
hTK1 and 71 M and 211 M with UpTK, respectively. The largest dif-
ferences in IC50-values between on both enzymes were found with
4 (fivefold), 12 (fourfold) and 7 (threefold); in all cases, the lower
IC50-values were observed with hTK1. Adding an extra spacer be-
tween the triazole and phenyl moiety lowered the IC50-value by al-
most threefold (from 9 to 10) in case of UpTK but also to some
extent with hTK1. By extending the spacer without the phenyl
moiety (5–6 and 8), no major changes with UpTK were found,
while with hTK1 the IC50-values decreased 4–3-fold in case of 8
and 5 compared to 6. Taken together, these results reveal that most
of the analogues 4–13 were better inhibitors of hTK1 compared to
The relative activity values are means SEM of three independent determinations.
The specific activities of UpTK and hTK1 are 1140 and 757 nmol dTMP/min per mg,
respectively.
a
Nd: Not determined.
Table 2
The IC50-values (means and SEM from three determinations) were determined with
1
l
M
3H-dThd and additions of nucleoside analogues (20–500
l
M)
Compounds
IC50 (l
M)
UpTK
hTK1
AZT 4
5
6
7
8
12.5 9.2
208 32
189 44
318 59
150 25
3.0 3.7
105 23
311 37
114 53
86 31
Figure 3. Catalytic efficiency phosphorylation of compound 4–13 by hTK1 and
UpTK.
9
200
3
99
2
10
11
12
13
73 19
120 31
211 21
71 13
78 36
60 23
53 23
44 10
showed 1.6% relative efficiency with UpTK and was the best phos-
phorylated bulky 30-substitution, followed by 12–13. Compounds

3264
J. Lin et al. / Bioorg. Med. Chem. 18 (2010) 3261–3269
UpTK. However, compound 13 was unique because it was the best
hTK1-Vv group, but at least the corresponding residues were
inhibitor and one of the best substrate for both enzymes.
hydrophobic (Fig. 4). In hTK1, Gly176 is responsible for a hydrogen
´
bond with the 3-oxygen of the sugar. Glu98 is the catalytic base
´
2.1.3. Model construction of nucleoside analogs in TK1 active
site
assisting in the phosphoryl transfer from ATP to 5-OH and this res-
idue is conserved in all organisms. UpTK is the only TK having a
cysteine, Cys183 in the lasso loop (Fig. 4 and S3), and this may
be used as a target in anti-UpTK agent development. Differences
in the TK structures were observed when hTK1-dTTP and UpTK-
dTTP complexes were compared and superimposed with PyMOL
(Fig. 5A). The longer UpTK lasso loop section seen in the sequence
alignment ( Fig. 4), extended the hTK1 lasso loop starting from
Lys166 to Asp176 (Fig. 5A).
Analogue 13 which had the lowest IC50-values with both en-
zymes was modelled into the active sites of the hTK1 (1W4R)
and UpTK (2UZ3) structures. These TK structures were chosen be-
cause they had visible lasso loops and P-hairpins. The base and su-
gar position of 13 fitted well with the available crystal structures of
hTK1/UpTK in complex with dTTP17–19 (Fig. 5B and C and S3). All
the relevant enzyme–residues interactions were almost preserved
(indicated by triangles in Fig. 4).
Multiple sequence alignment of human, Vaccinia virus (Vv) and
several bacterial TK sequences are shown in Figure 4. The sequence
identity was about 29% between hTK1 and UpTK. Phylogenetic
analyses32,33,39 suggested that hTK1 is in the same subgroup as
TKs of gram positive bacteria, while gram negative bacteria form
another group with regard to the P-loop and Mg2+ binding-motif
sequences. The amino acid composition of the P-b hairpin differs
between the TKs, e.g., a serine rich region at position 62–64 was
observed in hTK1, whereas only Ser54 is found in UpTK (Fig. 4).
The lasso loop is another important secondary structure ele-
ment with high sequence diversity, containing inserts and short
deletions. Bacterial TKs have more extended loops in this region
than hTK1 and VvTK (Fig. 4). Gram negative bacteria contain a con-
served Gln169 (using the E. coli numbering), which probably form
main chain hydrogen bonds with the N3 position of the nucleoside
base just like Ile178 in UpTK.17 The E. coli TK also has a conserved
hydrophobic Val171 which probable interact with the 2-oxygen
position of the base, like the corresponding Lys 180 in UpTK.17
These features were less clear in the gram positive bacteria-
Conformation A was found with 13 modelled into the active site
´
of UpTK. The 3 triazole-ring adopted a conformation (A) where it
interacted with Gly182 and thus it was in a bent positioned be-
tween the lasso loop and P-hairpin (Fig. 5C and S3B). In conforma-
Figure 4. Comparison of amino acid sequence alignment of human, Vaccinia virus (Vv) and several bacterial TK.

J. Lin et al. / Bioorg. Med. Chem. 18 (2010) 3261–3269
3265
Figure 5. dTTP, compound 6 and 13 docked in the active site of HTK1 (grey) and UpTK (turquoise).
tion A, the cyclic triazole-phenyl moiety is positioned between the
hydrophobic backbone of Gly182 and side chain of Arg53, however
it is also partly exposed to solvent and there were no interactions
with the other active site residues (Fig. 5C and S3B). Modelling 13
into the hTK1 structure produced another conformation (B) and in
tion in one step, starting from unprotected AZT. Some 1,2,3-tria-
zole AZT analogues have previously been synthesized, and were
found to lack significant activity against a set of viruses (HIV, Vac-
cinia virus, Cowpox virus, etc.).28 The lack of activity of these ana-
logues may be due to a failure to be phosphorylated by the
necessary cellular or viral kinases. That is why in this study, a
variety of different chemical structures have been attached to the
30-position of AZT, to gain a better understanding of the struc-
ture–activity relationship of hTK1 and UpTK.
´
this case the 3-triazole-phenyl moiety was instead shifted 90°,
going straight through the P-hairpin and retained interactions with
Gly182. Here the triazole phenyl-fluorine atom may form a main
chain hydrogen bond with either Ser63 or Ser64 (Fig. 5B ant S3D)
and apparently Tyr61 makes this possible by extending the hairpin.
The corresponding UpTK Ser54 is positioned 3.30 Å or 4.15 Å from
hTK1 Ser63 or 64, respectively (Fig. 5A and S4), and seems also
available for this interaction. However, the UpTK Arg53 side chain
might prevent this, because of a clash with the triazole phenyl moi-
ety (Fig. S4A). This may explain the hTK1 preference for conforma-
tion B rather than A, and the low Km-value for 13.
U. parvum lacks the novo dNTP biosynthesis and relies on sal-
vage for dNTP biosynthesis. Thus, UpTK is needed to provide dTTP
for DNA repair and replication. Previous studies11,17,18 also demon-
strated that UpTK is a promising drug target for antibiotic develop-
ment. TK sequence alignment from various microorganisms and
structural alignment of hTK1 and UpTK revealed that small differ-
ences exist between the enzymes, especially in the P-b hairpin and
lasso loop. Here hTK1 and Vv-TK with shorter lasso loop appeared
to form one subgroup, while the bacterial TKs form another. The
UpTK lasso loop was the only TK in this alignment containing a
Cys183 with unknown functional role. Gerland et al.40 synthesised
Modelling compound 6 in the active site of UpTK revealed that
´
the long 3 hydrophobic triazole pentyl carbon substitution was lo-
cated along the cleft between lasso loop and P-b hairpin. The short-
er side chain of Val179 of the lasso loop and different located
Pro48, in comparison with corresponding hTK1 amino acids
(Glu173 and Tyr55) makes the cleft more hydrophobic and larger
in UpTK (Fig. 5A and C and S5). This might explain the higher rel-
ative efficiency for compound 6 in UpTK than hTK1.
´
dThd nucleoside analogues harbouring 3-disulfide substitutions,
which were shown to be activated by hTK1 and they had promising
anti-HIV activity. These types of analogs could be a starting point
for targeting the Cys183 in UpTK.
´
Substitutions in the 3 of the sugar and the 3 and 5 position of
base could increase the selectivity for bacterial kinases.17 30-AZT
3. Discussion
´
is well tolerated by TK1 and served as a start point for 3 triazole
analogs synthesised and tested here.
Both hTK1 and UpTK were able to accept many 3-analogs. Com-
pounds 10 and 13 exhibit similar properties as substrate or inhib-
The copper(I) azide-alkyne 1,3-dipolar cycloaddition of AZT
gave easy access to a small library of compounds. The Huisgen
cycloaddition reaction was performed under microwave irradia-
´

3266
J. Lin et al. / Bioorg. Med. Chem. 18 (2010) 3261–3269
itors, but compound 13 was found to be the best substrate with
hTK1 and the next best with UpTK. In the latter case compound 6
was the most efficient substrate. The other tested analogs achieved
The minimal activity with hTK1 may be one explanation for the
lack of biological activity observed previously with this type of
analogues.28 However, when a 3-triazole dThd triphosphate was
´
3–10-fold higher activities with UpTK than with hTK1. The IC50
-
synthesized it could inhibit the HIV-RT activity in vitro.28a Informa-
tion presented here may help in the design of new triazole
compounds of medical interest in anti-Ureaplasma and other
anti-microbial therapies with minimal effects on host cell
proliferation.
values for these analogues revealed that hTK1 was usually more
inhibited than UpTK (Table 2). Thus, the capacity to serve as accep-
tor was not correlated to the inhibitory effect and this fact has been
observed previously.17 However, none of the analogs were very po-
tent inhibitors but compounds 6, 9, 12 and 13 exhibited among the
best kinetic results and were subjected to antibacterial activity
tests. All synthesized analogs have been tested for their inhibitory
activity on Up growth as well as on uninfected phytohemaggluti-
nin-stimulated primary human peripheral blood mononuclear
(PBM), a T-lymphoblastoid cell line (CEM), or African green mon-
key kidney (Vero) cells.38 None of these analogs displayed toxic ef-
fects neither inhibition of Up growth at the concentrations tested.
Compound 13 was modelled in the active site of UpTK and hTK1
5. Experimental section
5.1. Chemistry section
For the synthesis of 30-thymidine derivatives, commercially
available chemicals were reagent grade and used as received. The
starting material for the triazoles was AZT, which was prepared
as described previously.42 The microwave was a Biotage AB Initia-
tor EXP EU with a maximum power of 300 W. The vials used in the
microwave were Emrys™ process vials 0.5–2 mL. The reactions
were monitored by thin-layer chromatography (TLC) analysis
using silica gel plates (kieselgel 60 F254, E. Merck). Silica gel (Merck
´
and our analysis suggested that the 3 substitutions of 1–13 may be
positioned in two different conformations; conformation A and B.
In conformation A, the plane of the triazole-phenyl rings was posi-
tioned between the hydrophobic part of lasso loop Gly182 and P-b
hairpin Arg53. The terminal part of the phenol with the hydrophilic
fluorine atom was probably exposed to solvent. The absence of
specific amino acid interactions apparently results in similar Km-
values as seen with compounds 6, 9, 12 and 13 (Table S1).
Kieselgel 60, 15–40 lm) was used for flash chromatography. Com-
pounds were visualized by UV irradiation and by spraying with
2.5% phosphomolybdic acid in 95% EtOH, followed by charring at
150 °C. The 1H and 13C NMR spectra were recorded on a Brucker
AVANCE DPX 250 Fourier transform spectrometer at 400 MHz for
1H and 100 MHz for 13C, respectively, using tetramethylsilane as
the internal standard, unless otherwise stated. Chemical shifts
are given in ppm and signals are reported as s (singlet), d (doublet),
t (triplet), q (quartet), and m (multiplet). High Resolution Mass
spectra were performed by the Mass Spectrometry Center of Emory
University (Atlanta, USA) and Blaise Pascal University (Aubière,
France). The purity was determined by semi-preparative HPLC
Compound 13 was also modelled into the active site of hTK1;
here in conformation B, the triazole-phenyl-fluorine stretches from
the lasso loop through the P-b hairpin. In this case, there are two
additional serines 63–64, which provide potential hydrogen donor
interactions with the fluorine atom of 13. The Tyr61 seems to be
responsible for making this interaction possible; because it is
extending the hTK1 hairpin enough for the triazole-phenyl moiety
to avoid a clash with Arg60 (Fig. 5B and S4B). This may explain its
apparent high affinity for hTK1 and in the absence of this fluorine
atom; the Km-values were 3–4-fold higher. The extra tyrosine is
not observed in UpTK and the Arg53 may therefore block the tria-
zole-phenyl moiety interactions with serine 54 (Fig. S4A), which
explains that the most likely conformation in UpTK is A. In the case
(Hypersil 100; C-18; 5 lm); with an appropriate gradient of aceto-
nitrile/H2O; purity of key target compounds was >95%.
5.1.1. 30-(4-Hydroxymethyl-1,2,3-triazol-1-yl)-30-
deoxythymidine (1)
´
30-Azido-30-deoxythymidine (50 mg, 0.19 mmol) was dissolved
in mix solvent tert-BuOH/H2O 1/1 (2 mL), Cu(0) (0.8 equiv), CuSO4
of hTK1, the 3 substitution in conformation A may sterically clash
against the Arg60 side chain (Fig. S4B) explaining that the most
likely conformation in hTK1 is B.
(1 M) 38 lL and propargyl alcohol (0.2 mmol). The mixture was
Modelled compound 6 revealed additional amino acid differ-
ences between UpTK (Val179 and Pro48) and hTK1 (Glu173 and
Tyr55), located in either the lasso loop or P-b hairpin (Fig. 5A and
D and S5), possibly explaining why compound 6 is a better sub-
strate for UpTK than hTK1.
Thus, each enzyme most likely has a preference for one of two
substrate analog conformations and the active site of UpTK seems
to be more flexible than that of hTK1. This fact may be further
exploited in the design of more selective inhibitors that may block
DNA precursor synthesis or in substrates that would be selectively
activated and incorporated in bacterial cells.
stirred under microwave irradiation at 125 °C during 10 min. The
solvent was removed under reduced pressure and the crude resi-
due purified using flash chromatography with an elution gradient
AcOEt to AcOEt/MeOH. Yield: 80% (49 mg) as white solid. CAS reg-
istration: 127479-69-0. 1H NMR (CD3OD) d 1.91 (s, 3H, CH3), 2.75
(m, 1H, H-20b), 2.91 (m, 1H, H-20a), 3.77 (dd, 1H, J = 2.8, 12.3 Hz,
H-50b), 3.91 (dd, 1H, J = 2.8, 12.3 Hz, H-50a), 4.37 (dt, 1H, J = 2.8,
5.4 Hz, H-40), 4.69 (s, 2H, CH2), 5.44 (dt, 1H, J = 5.4, 8.2 Hz H-30),
6.48 (t, J = 6.3 Hz, 1H, H-10), 7.91 (s, 1H, H-6), 8.07 (s, 1H, CH@);
13C NMR (CD3OD) 12.5(CH3), 39.1(C-20), 61.0(CH2–OH), 62.1(C-30),
62.8(C-50),
86.4(C-40),
86.7(C-10),
100.1(C-5),
111.7(C@),
127.3(C@), 138.3(C-6), 152.4(C-2), 166.4(C-4). HRMS (M++H)
4. Conclusion
324.13000, calcd for C13H18N5O5 324.13025.
An efficient and rapid procedure to synthesize new 30-
a-
5.1.2. 30-(4-(2-Hydroxy-ethyl)-1,2,3-triazol-1-yl)-30-
[1,2,3]triazol-30-deoxythymidine analogs is presented and their
biological testing involved kinetic, structural and modelling analy-
ses with the key activation enzyme TK1, of human and bacterial
origin. The results as well as those of others41 revealed the impor-
tance of detailed knowledge of the structure–activity relationships
in the TK1 enzyme family, especially in the absence of determined
substrate–enzyme complex structures. An important difference in
structure and function between the human and bacterial TK1 en-
zymes, related to sequence diversity in the active sites, was found.
deoxythymidine (2)
Compound 2 was prepared as described for 1 starting from AZT
(50 mg, 0.19 mmol) and 3-but-1-yn-1-ol (0.2 mmol) under micro-
wave during 10 min, yield: 82% (52 mg) as white solid.
mp = 218–220 °C, 1H NMR (CD3OD) : 1.91(d, 3H, J = 1.0 Hz, CH3),
2.75(dd, 1H, J = 6.4, 8.8, 14.8 Hz, H-20b), 2.88–2.94 (m, 3H, H-20a
and CH2), 3.76 (dd, 1H, J = 2.9, 12.3 Hz, H-50b), 3.81 (m, 2H, CH2OH),
3.91 (dd, 1H, J = 2.9, 12.3 Hz, H-50a), 4.36 (dt, 1H, J = 2.9, 5.6 Hz, H-
40), 5.40(dt, 1H, J = 5.6, 8.3 Hz, H-30), 6.48(t, 1H, J = 6.3 Hz, H-10),

J. Lin et al. / Bioorg. Med. Chem. 18 (2010) 3261–3269
3267
7.91(d, 1H, J = 1.0 Hz, H-6), 7.94 (s, 1H, CH@); 13C NMR (CD3OD) :
12.5(CH3), 29.9(CH2), 39.0(C-20), 60.9(C-30), 62.0(CH2–OH),
62.1(C-50), 86.4(C-40), 86.7(C-10), 111.7(C-5), 123.8(CH@), 138.3(C-
6), 152.3(C-2), 166.4(C-4). HRMS (M++H) 338.14548, calcd for
C14H20N5O5 338.14590. Purity HPLC 99.2%, tR = 3.7 min acetoni-
trile/H2O (20:80, vol/vol).
(dd, 1H, J = 3.1, 12.3 Hz, H-50b), 3.89 (dd, 1H, J = 3.1, 12.3 Hz,
H-50a), 4.35 (dt, 1H, J = 3.0, 5.8 Hz, H-40), 5.39 (dt, 1H, J = 5.8,
8.5 Hz, H-30), 6.47 (t, 1H, J = 6.4 Hz, H-10), 7.87 (s, 1H, CH@), 7.91
(d, 1H, J = 1.0 Hz, H-6); NMR 13C (CD3OD) 12.5(T-CH3), 13.9(CH3),
23.7(CH2), 28.9(CH2), 39.0(C-20), 60.8(C-30), 62.1(C-50), 86.4(C-40),
86.7(C-10), 111.7 (C-5), 122.8(C@), 138.3(C-6), 149.6(C@),
152.3(C-2), 166.4(C-4). Purity HPLC 95.1%, tR = 32.4 min acetoni-
trile/H2O (15:85, vol/vol).
5.1.3. 30-(4-Methoxymethyl-1,2,3-triazol-1-yl)-30-
deoxythymidine (3)
Compound 3 was prepared as described for 1 starting from AZT
(50 mg, 0.19 mmol) and methyl propargyl ether (0.2 mmol) under
microwave during 10 min, yield: 90% (57 mg) as white solid. mp:
208–210 °C. 1H NMR (CD3OD) 1.90 (d, 3H, J = 1.2 Hz, T-CH3), 2.75
(ddd, 1H, J = 6.0, 8.4, 14.2 Hz, H-20b), 2.91 (ddd, 1H, J = 5.6, 6.8,
14.2 Hz,, H-20a), 3.38 (s, 3H, OCH3), 3.77 (dd, 1H, J = 3.0, 12.3 Hz,
H-50b), 3.91 (dd, 1H, J = 3.0, 12.3 Hz, H-50a), 4.36 (dt, 1H, J = 3.0,
5.6 Hz, H-40), 4.54 (s, 2H, –CH2O), 5.44 (dt, 1H, J = 5.2, 8.4 Hz,
H-30), 6.48 (t, 1H, J = 6.4 Hz, H-10), 7.91 (d, 1H, J = 1.2 Hz, H-6),
5.1.7. 30-(4-tert-Butyl-1,2,3-triazol-1-yl)-30-deoxythymidine (7)
AZT (100 mg, 0.374 mmol) and 3,3-dimethyl-1-butyne
(0.412 mmol) were suspended in a 1:1 mixture of water and tert-
butanol (4 mL). Then, sodium ascorbate (0.1 equiv) and CuSO4
(0.01 equiv) were added and the reaction mixture was stirred until
the complete disappear of starting material. The solvent was re-
moved under reduced pressure and the crude residue purified
using flash chromatography with an elution gradient AcOEt/EtOH
to AcOEt/MeOH. Yield: 82% (107 mg) as white solid. mp:
138–140 °C. 1H NMR (CDCl3) d 1.35 (s, 9H, CH3 tert-but), 1.93 (s,
3H, T-CH3), 2.90–2.95 (m, 2H, H-20), 3.79 (dd, 1H, J = 2.4, 12.4 Hz,
H-50b), 4.02 (dd, 1H, J = 2.2, 12.4 Hz, H-50a), 4.43 (dt, 1H, J = 2.2,
4.8 Hz, H-40), 5.39 (dt, 1H, J = 5.7, 8.2 Hz, H-30), 6.22 (t, 1H,
J = 6.6 Hz, H-10), 7.36 (s, 1H, CH@), 7.43 (d, 1H, J = 1.0 Hz, H-6);
8.92 (s, 1H, NH), NMR 13C (CDCl3) : 12.4(T-CH3), 30.3(3 CH3),
30.8(C–CH3), 37.5(C-20), 60.0(C-30), 61.7(C-50), 85.3(C-40), 89.0
(C-10), 111.3 (C-5), 118.5(C@), 137.9(C-6), 150.3(C-2), 158.1 (C@),
163.5(C-4) Purity HPLC 99.4%, tR = 22.6 min; acetonitrile/H2O
(20:80, vol/vol).
13
8.11 (s, 1H, CH@). C NMR (CD3OD) 12.5 (CH3), 39.0 (C-20), 58.2
(CH2–O), 61.1 (C-30), 62.1 (C-50), 66.3 (O–CH3), 86.4 (C-40), 86.7
(C-10), 111.7 (C-5), 124.8 (C@), 138.3 (C-6), 152.3 (C-2), 166.3
(C-4). HRMS (M++H) 338.14551, calcd for C14H20N5O5 338.14590.
Purity HPLC 99.6%, tR = 4.7 min acetonitrile/H2O (20:80, vol/vol).
5.1.4. 30-(4-Methylpropionate-1,2,3-triazole-1-yl)-30-
deoxythymidine (4)
Compound 4 was prepared as described for 1 starting from AZT
(50 mg, 0.19 mmol) and propargyl acetate (0.2 mmol) under
microwave during 2 min. Yield: 81% (56 mg) as white solid. mp:
128–130 °C. 1H NMR (CD3OD) 1.90 (s, 3H, CH3), 2.05 (s, 3H, OAc),
2.70 (ddd, 1H, J = 6.4, 8.8, 14.4 Hz,, H-20b), 2.90 (ddd, 1H, J = 5.2,
6.8, 14.4 Hz, H-2’a), 3.77 (dd, 1H, J = 2.8, 12.3 Hz, H-50b), 3.91 (dd,
1H, J = 2.8, 12.3 Hz, H-50a), 4.36 (dt, 1H, J = 2.8, 5.6 Hz, H-40), 5.18
(s, 2H, CH2), 5.43 (dt, 1H, J = 5.6, 8.5 Hz, H-30), 6.48 (t, 1H
J = 6.3 Hz, H-10), 7.90 (s, 1H, H-6), 8.15 (s, 1H, CH@); 13C NMR
(CD3OD) 12.5 (CH3), 20.6 (COCH3), 39.0 (C-20), 58.2 (CH2-OAc),
61.1 (C-30), 62.1 (C-50), 86.3 (C-40), 86.7 (C-10), 111.7 (C-5), 125.6
(C@), 138.2 (C-6), 140.7 (C@), 152.3 (C-2), 166.3 (C-4), 172.3
(C@O). HRMS (M++H) 366.14056, calcd for C15H20N5O6
366.14081. Purity HPLC 99.9%, tR = 11.5 min acetonitrile/H2O
(15:85, vol/vol).
5.1.8. 30-(4-Pentyl-1,2,3-triazol-1-yl)-30-deoxythymidine (8)
Compound 7 was prepared as described for 1 starting from AZT
(50 mg, 0.19 mmol) and 1-heptyne (0.2 mmol) under microwave
during 100 min, yield: 91% (63 mg) as white solid. Mp = 132–
134 °C. 1H NMR (CD3OD) d: 0.91 (t, 3H, J = 7.4 Hz, CH3e), 1.37 (m,
4H, CH2b et CH2c), 1.68 (t, 2H, J = 7.2 Hz, CH2a), 1.91 (s, 3H,
T-CH3), 2.67–2.78 (m, 3H, H-20b et CH2d), 2.88 (m, 1H, H-20a),
3.77 (dd, 1H, J = 3.1, 12.3 Hz, H-50b), 3.91 (dd, 1H, J = 3.1, 12.3 Hz,
H-50a), 4.35 (dt, 1H, J = 3.0, 5.6 Hz, H-40), 5.39 (dt, 1H, J = 5.6,
8.5 Hz, H-30), 6.48 (t, 1H, J = 6.3 Hz, H-10), 7.89 (s, 1H, H-6), 7.91
(s, 1H, CH@); 13C NMR (CD3OD) 12.5(CH3), 14.3(CH3e), 23.4, 26.3,
30.2, 32.5(4CH2), 39.0(C-20), 60.8(C-30), 62.1(C-50), 86.4(C-40),
86.7(C-10), 111.7(C-5), 122.8(C@), 138.3(C-6), 149.5(C@), 152.3(C-
2), 166.4(C-4). HRMS (M++H) 364.19748, calcd for C17H26N5O4
364.19793. Purity HPLC 95.2%, tR = 16.65 min acetonitrile/H2O
(30:70, vol/vol).
5.1.5. 30-(4-(3-Chloro-propyl)-1,2,3-triazol-1-yl)-30-
deoxythymidine (5)
Compound 5 was prepared as described for 1 starting from AZT
(50 mg, 0.19 mmol) and 5-chloropentyne (0.2 mmol) under micro-
wave during 30 min, yield: 78% (55 mg) as white solid. mp = 203–
205 °C. 1H NMR (CD3OD) d 1.91 (d, 3H, J = 1.0 Hz, T-CH3), 2.13 (m,
2H, CH2b), 2.71 (m, 1H, H-20b), 2.88 (t, 2H, J = 7.2 Hz, CH2a), 2.90
(m, 1H, H-20a), 3.61 (t, 2H, J = 6.3 Hz, CH2c), 3.76 (dd, 1H, J = 3.0,
12.3 Hz, H-50b), 3.89 (dd, 1H, J = 3.0, 12.3 Hz, H-50a), 4.35 (dt, 1H,
J = 3.0, 5.8 Hz, H-40), 5.39 (dt, 1H, J = 5.5, 8.5 Hz, H-30), 6.47 (t, 1H,
J = 6.4 Hz, H-10), 7.90 (br d, 1H, J = 1.0 Hz, H-6), 7.92 (s, 1H, CH@);
13C NMR (CD3OD) 12.5(T-CH3), 23.5(CH2), 33.24(CH2), 39.0(C-20),
44.8(CH2), 60.9(C-30), 62.1(C-50), 86.4(C-40), 86.7(C-10), 111.6(C-5),
123.2(C@), 138.3(C-6), 148.0(C@), 152.3(C-2), 166.4(C-4). HRMS
(M++H) 370.12758, calcd for C15H2134Cl1N5O4 370.12766. Purity
HPLC 99.0%, tR = 18.45 min; acetonitrile/H2O (20:80, vol/vol).
5.1.9. 30-(4-Phenyl-1,2,3-triazol-1-yl)-30-deoxythymidine (9)
Compound 9 was prepared as described for 1 starting from AZT
(50 mg, 0.19 mmol) and phenylacetylene (0.2 mmol), under micro-
wave during 20 min, yield: 84% (59 mg) as white solid. CAS regis-
tration: 127728-29-4. 1H NMR (CD3OD) 1.91(s, 3H, T-CH3), 2.78
(m, 1H, H-20b), 2.97 (m, 1H, H-20a), 3.82 (dd, 1H, J = 2.9, 12.2 Hz,
H-50b), 3.94 (dd, J = 2.9, 12.2 Hz, 1H, H-50a), 4.43 (dt, 1H, J = 2.9,
5.6 Hz, H-40), 5.48 (dt, 1H, J = 5.6, 8.5 Hz, H-30), 6.52 (t, J = 6.6 Hz,
1H, H-10), 7.34–7.46 (m, 3H, Harom), 7.83 (d, 2H, J = 7.8 Hz, Harom),
7.93 (s, 1H, H-6), 8.47(s, 1H, CH@); 13C NMR (CD3OD) 12.5(CH3),
39.1(C-20), 61.1(C-30), 62.2(C-50), 86.4(C-40), 86.7(C-10), 111.7(C-5),
121.9(C@), 126.7, 129.5, 130.0(6 Carom), 134.1(C@), 138.3(C-6),
152.3(C-2), 166.3(C-4). HRMS (M++H) 370.15055, calcd for
C18H20N5O4 370.15098. Purity HPLC 98.9%, tR = 10.0 min acetoni-
trile/H2O (30:70, vol/vol).
5.1.6. 30-(4-Propyl-1,2,3-triazol-1-yl)-30-deoxythymidine (6)
Compound 6 was prepared as described for 1 starting from AZT
(50 mg, 0.19 mmol) and 1-pentyne (0.2 mmol) under microwave
during 120 min, yield: 85% (54 mg) as white solid. mp: 210–
212 °C. 1H NMR (CD3OD) 0.97 (t, 3H, J = 7.3 Hz, CH3c), 1.69 (six,
2H, J = 7.3 Hz, CH2b), 1.91 (d, 3H, J = 1.0 Hz, T-CH3), 2.68 (t, 2H,
J = 7.2 Hz, CH2a), 2.71 (m, 1H, H-20b), 2.90 (m, 1H, H-20a), 3.76
5.1.10. 30-(4-Benzyl-1,2,3-triazol-1-yl)-30-deoxythymidine (10)
AZT
(100 mg,
0.374 mmol)
and
3-phenyl-1-propyne
(0.412 mmol) were suspended in a 1:1 mixture of water and tert-
butanol (4 mL). Then sodium ascorbate (0.1 equiv) and CuSO4

3268
J. Lin et al. / Bioorg. Med. Chem. 18 (2010) 3261–3269
(0.01 equiv) were added and the reaction mixture was stirred until
the complete disappear of starting material. The solvent was re-
moved under reduced pressure and the crude residue purified
using flash chromatography with an elution gradient AcOEt/EtOH
to AcOEt/MeOH. Yield: 84% (120 mg) as white solid: mp = 120–
122 °C. 1H NMR (CD3OD) 1.89 (s, 3H, T-CH3), 2.69 (ddd, 1H,
J = 6.4, 8.6, 14.2 Hz, H20b), 2.87 (dt, 1H, J = 6.0, 14.2 Hz, H20a),
3.73 (dd, 1H, J = 3.2, 12.4 Hz, H50b), 3.88 (dd, J = 3.2, 12.4 Hz, 1H,
H50a), 4.05 (s, 2H, CH2–Ph), 4.33 (dt, 1H, J = 3.2, 6.0 Hz, H40), 5.36
(dt, 1H, J = 6.0, 8.4 Hz, H30), 6.45 (t, 1H, J = 6.4 Hz, H10), 7.17–7.30
(m, 5H, Harom), 7.83(s, 1H, CH@); 7.88 (s, 1H, H6). 13C NMR (CD3OD)
12.5(CH3), 32.6(CH2), 39.0(C-20), 60.9(C-30), 62.1(C-50), 86.4(C-40),
86.7(C-10), 111.7(C-5), 123.5(C@), 127.6(2 Carom), 126.5, 129.6,
129.7 (3 Carom), 138.3(C-6), 140.3(Carom), 148.8(C@), 152.3(C-2),
166.4(C-4). Purity HPLC 95.9%, tR = 10.0 min acetonitrile/H2O
(30:70, vol/vol).
5.46 (dt, 1H, J = 5.3, 8.4 Hz, H-30), 6.50 (t, J = 6.2 Hz, 1H, H-10), 7.14
(m, 2H, Harom), 7.83 (m, 2H, Harom), 7.90 (d, 1H, J = 1.0 Hz, H-6), 8.42
13
(s, 1H, CH@); C NMR (CD3OD) 12.5(CH3), 39.1(C-20), 61.1(C-30),
62.2(C-50), 86.4(C-40), 86.7(C-10), 111.7(C-5), 116.8(d, J = 21.3 Hz, 2
Carom), 121.8(C@), 123.9(Carom), 128.7(d, J = 7.3 Hz,
2 Carom),
138.3(C-6), 148.2, (C@), 164.2(d, J = 234.6 Hz, Carom), 152.3(C-2),
166.2(C-4), HRMS (M++H) 388.14105, calcd for C18H19FN5O4
388.14211. Purity HPLC 98.48%, tR = 14.13 min acetonitrile/H2O
(30:70, vol/vol).
5.2. Biology section
The none-radioactive-labelled substances dThd/dUrd were from
(Sigma Aldrich) and radiolabeled substance
[c-
32P]-ATP
(3000 Ci mmolꢀ1) was from Perkin–Elmer. All compounds 1–13
(Fig. 2) were diluted in DMSO and the nucleosides dThd and AZT
in sterile H2O.
5.1.11. 30-(4-(2-Phenyl-ethyl)-1,2,3-triazole-1-yl)-20-
deoxythymidine (11)
5.3. Enzyme relative activity, kinetics and IC50 determination
AZT
(100 mg,
0.374 mmol)
and
4-phenyl-1-butyne
(0.412 mmol) were suspended in a 1:1 mixture of water and tert-
butanol (4 mL). Then sodium ascorbate (0.1 equiv) and CuSO4
(0.01 equiv) were added and the reaction mixture was stirred until
the complete disappear of starting materiel. The solvent was re-
moved under reduced pressure and the crude residue purified
using flash chromatography with an elution gradient AcOEt/EtOH
to AcOEt/MeOH. Yield: 86% (127 mg) as white solid mp = 150–
152 °C. 1H NMR (CD3OD) d 1.90 (s, 3H, T-CH3), 2.69 (ddd, 1H,
J = 6.0, 8.4, 14.4 Hz, H20b), 2.85 (m, 1H, H20a), 2.95–3.04 (m, 4H, 2
CH2), 3.71 (dd, 1H, J = 3.0, 12.0 Hz, H50b), 3.87 (dd, J = 3.0,
12.0 Hz, 1H, H50a), 4.29 (dt, 1H, J = 3.0, 6.0 Hz, H40), 5.34 (dt, 1H,
J = 6.0, 8.8 Hz, H30), 6.44 (t, 1H, J = 6.4 Hz, H10), 7.15 (m, 3H, Harom),
7.24 (m, 2H, Harom), 7.72(s, 1H, CH@); 7.89 (s, 1H, H6). 13C NMR
(CD3OD) 12.5(CH3), 28.4(CH2), 36.5(CH2), 39.0(C-20), 60.7(C-30),
62.0(C-50), 86.4(C-40), 86.7(C-10), 111.7(C-5), 123.2(C@), 127.6(2
Carom), 126.4, 129.5, 129.6 (3 Carom), 138.3(C-6), 142.3(Carom),
148.6(C@), 152.3(C-2), 166.4(C-4). Purity HPLC 99.6%, tR = 14.6 min
acetonitrile/H2O (30:70, vol/vol).
Recombinant Up-TK and hTK1 enzymes were expressed and
purified as described previously.11,18 Phosphate transfer assay with
100 M [c-
32P] ATP was used to determine TK activities as previ-
ously described with some modifications.11,16,17 In short, 100 M
of dThd and dThd-analogues with 5 ng Up-TK or 10 ng hTK1 were
used. The mixtures were incubated for 15 min for UpTK and 20 min
for hTK1 at 37 °C and heat-inactivated at 95 °C for 2 min. The rel-
ative activity was expressed using the activity with dThd as
100%. The kinetic parameters were determined with different sub-
strate concentrations for example, dThd 2–20 M for 0.5 ng UpTK
and 2–50 M for 1 ng hTK1. The standard reaction contain
10–500 M dThd-analogues with 10 ng UpTK or -hTK1, 50 mM
Tris/HCl pH 7.6, 5 mM MgCl2, 100 lM [c-
32P]-ATP, 0.5 mg/ml BSA
and 10 mM DTT. The final DMSO concentration did not exceed
5% in the reaction solution. The reactions were performed for
20 min at 37 °C and then heat-inactivated at 95 °C for 2 min. Two
microlitres of the reaction solutions were spotted on PEI-TLC
plates, which were developed in 33 mL dH2O, 66 ml iso-butyric
acid and 1 mL NH4. Quantification was done by phosphoimaging
analysis (Fujifilm Image Gauge, version 3.3). The enzyme kinetic
parameters Km, Vmax, and Vmax/Km were calculated by using the
Michaelis Menten equation with the SigmaPlot Enzyme Kinetic
Module version 2.1 (SPSS Science, Chicago, IL). The mean and stan-
dard error of mean (SEM) for relative activity %, Km and Vmax values
were calculated from three independent determinations.
The inhibitory concentration at 50% (IC50) were determined
by using DE-81 filter paper technique and the [3H]-dThd assay.
The standard reaction contained 2 mM MgCl2, 5 mM ATP, fixed
1 M [3H]-dThd, 5 mM DTT, 0,5 ng UpTK and 4 ng hTK1 to achieve
similar and comparable activity. Nucleoside analog concentra-
tions of 10–500 M (1–20 M for AZT) competing with 1 M [3H]-
dThd in total volume of 50 L were used. The reaction tempera-
ture was 25 °C for both enzymes in order to keep the total con-
version of substrate to product below 25%. The filters with 10 L
fractions of the reaction solutions were washed three times in
10 mM ammonium formate and once in water. The reaction
products were eluted with 0.5 mL 0.1 M HCl/0.2 M KCl and
radioactivity was determined by liquid scintillation counting
(Beckman). Relative inhibition was set to 100% for dThd with
no nucleoside analogs and IC50 was estimated by the Biofitdata
tool (Change Bioscience) http://www.changbioscience.com/stat/
ec50.html.
5.1.12. 30-(4-(2-Pyridine)-1,2,3-triazole-1-yl)-20-
deoxythymidine (12)
Compound 12 was prepared as described for 1 starting from
AZT (50 mg, 0.19 mmol) and 2-ethynylpyridine (0.2 mmol), under
microwave during 10 min, yield: 86% (60 mg) as white solid. CAS
registration: 127479-76-9. 1H NMR (CD3OD) d 1.91 (s, 3H, T-
CH3), 2.79 (m, 1H, H-20b), 2.97 (m, 1H, H-20a), 3.82 (dd, 1H,
J = 2.9, 12.2 Hz, H-50b), 3.94 (dd, J = 2.9, 12.2 Hz, 1H, H-50a), 4.41
(dt, 1H, J = 2.9, 5.6 Hz, H-40), 5.52 (dt, 1H, J = 5.6, 8.5 Hz, H-30),
6.53 (t, J = 6.6 Hz, 1H, H-10), 7.36 (m, 1H, Hpyr), 7.91 (m, 2H, Hpyr),
8.10 (d, 1H, J = 8.1 Hz, Hpyr), 8.57 (br s, 2H, CH@ and H-6). 13C
NMR (CD3OD)
:
12.5(CH3), 39.1(C-20), 61.3(C-30), 62.1(C-50),
86.4(C-40), 86.7(C-10), 111.7(C-5), 121.9(C@), 120.7, 124.0, 136.2,
148.8, 158.3 (Carom), 138.3(C-6), 150.9(C-2), 166.4(C-4). HRMS
(M++H) 371.14584, calcd for C17H19N6O4 371.14678. Purity HPLC
99.2%, tR = 5.53 min acetonitrile/H2O (30:70, vol/vol).
5.1.13. 30-(4-(4-Fluorophenyl)-1,2,3-triazole-1-yl)-20-
deoxythymidine (13)
Compound 13 was prepared as described for 1 starting from AZT
(50 mg, 0.19 mmol) and 1-ethynyl-4-fluorobenzene (0.2 mmol), un-
der microwave during 20 min, yield: 96% (71 mg) as white solid. CAS
registration: 127479-75-8. 1H NMR (CD3OD) d 1.88 (d, 3H, J = 1.0 Hz,
T-CH3), 2.78 (ddd, 1H, J = 6.0, 8.4, 14.2 Hz, H20b), 2.94 (ddd, 1H,
J = 5.6, 6.8, 14.2 Hz, H20a), 3.82 (dd, 1H, J = 2.9, 12.2 Hz, H-50b), 3.93
(dd, J = 2.9, 12.2 Hz, 1H, H-50a), 4.44 (dt, 1H, J = 2.9, 5.3 Hz, H-40),
One unit of kinase activity was defined as the formation of
1 nmol deoxyribonucleoside 50-monophosphate per milligram pro-
tein per minute.

J. Lin et al. / Bioorg. Med. Chem. 18 (2010) 3261–3269
3269
7. Camerman, A.; Mastropaolo, D.; Camerman, N. Proc. Natl. Acad. Sci. U.S.A. 1987,
84, 8239.
5.4. Sequence and structure sources
8. Elwell, L. P.; Ferone, R.; Freeman, G. A.; Fyfe, J. A.; Hill, J. A.; Ray, P. H.; Richards,
C. A.; Singer, S. C.; Knick, V. B.; Rideout, J. L., et al Antimicrob. Agents Chemother.
1987, 31, 274.
9. Wang, J.; Neuhard, J.; Eriksson, S. Antimicrob. Agents Chemother. 1998, 42, 2620.
10. Glass, J. I.; Lefkowitz, E. J.; Glass, J. S.; Heiner, C. R.; Chen, E. Y.; Cassell, G. H.
Nature 2000, 407, 757.
11. Carnrot, C.; Wehelie, R.; Eriksson, S.; Bolske, G.; Wang, L. Mol. Microbiol. 2003,
50, 771.
12. Bizarro, C. V.; Schuck, D. C. Gen. Mol. Biol. 2007, 30, 190.
13. Pollack, J. D. Trends Microbiol. 2001, 9, 169.
14. Hudson, M. M.; Talbot, M. D. Int. J. STD AIDS 1997, 8, 546.
15. Waites, K. B.; Katz, B.; Schelonka, R. L. Clin. Microbiol. Rev. 2005, 18, 757.
16. Wang, L.; Westberg, J.; Bolske, G.; Eriksson, S. Mol. Microbiol. 2001, 42, 1065.
17. Kosinska, U.; Carnrot, C.; Eriksson, S.; Wang, L.; Eklund, H. FEBS J. 2005, 272,
6365.
TK sequences from human (P04183), gram positive sequences;
U. parvum (Q9PPP5), Bacillus antracis (Q81JX0), and gram negative
bacteria; E. coli (P23331), Yersinia pestis (Q8ZEJ1), V. virus (P68563)
were downloaded from protein knowledgebase (http://www.uni-
prot.org/) and pasted in ClustalW2 (http://www.ebi.ac.uk/Tools/
clustalw2/index.html) tool according to instructions with default
settings. Monomers with crystal structures for; hTK1 1XBT (chain
G) and UpTK 2UZ3 (chain B) were superimposed using the PyMOL
graphic system (DeLano Scientific).
5.5. Computational studies of human TK1 and Ureaplasma TK
18. Welin, M.; Kosinska, U.; Mikkelsen, N. E.; Carnrot, C.; Zhu, C.; Wang, L.;
Eriksson, S.; Munch-Petersen, B.; Eklund, H. Proc. Natl. Acad. Sci. U.S.A. 2004,
101, 17970.
19. Birringer, M. S.; Claus, M. T.; Folkers, G.; Kloer, D. P.; Schulz, G. E.; Scapozza, L.
FEBS Lett. 2005, 579, 1376.
20. Munch-Petersen, B.; Cloos, L.; Tyrsted, G.; Eriksson, S. J. Biol. Chem. 1991, 266,
9032.
21. Katritzky, A. R.; Zhang, Y.; Singh, S. K. Heterocycles 2003, 60, 1225. and
references cited therein.
The structures of the B chain of UpTK (2UZ3) and chain A of
hTK1 (1W4R) were saved in separate pdb-file. The hTK1 (1XBT)
structure were not used for modelling because residues connected
to serine 64 of the P-hairpin loop was disordered. The Mg2+ atom
was absent in 1W4R, but the ion coordinates could be added to this
structure by superimposing both structure (0.158 Å for 157 C
a
22. Chen, M. D.; Lu, S. J.; Yuag, G. P.; Yang, S. Y.; Du, X. L. Heterocycl. Commun. 2000,
6, 421.
atoms) with each other in PyMOL graphic system (DeLano Scien-
tific). The analog structures in 2D format were created by the Mar-
vin sketch program from the Marvin beans package (http://
www.chemaxon.com/). They were also saved in Smile format and
further converted to pdb format at ‘Online SMILIES translator’
(http://cactus.nci.nih.gov/services/translate/). Modelling of the
best accepted triazole-nucleoside, substrate and inhibitor com-
pound 13 was used for modelled by the Arguslab43 program
(www.arguslab.com). The binding site was defined from the coor-
dinates of the ligand main chain bonds (dTTP) in the original PDB
files of hTK1 (1W4R), respectively UpTK (2UZ3). The modelling set-
tings were set to ‘high’ and ‘flexible ligand docking.’ The chosen
conformations were then saved in pdb format and visualized with
PyMOL.
23. Alvarez, R.; Velazquez, S.; San-Felix, A.; Aquaro, S.; De Clercq, E.; Perno, C. F.;
Karlsson, A.; Balzarini, J.; Camarasa, M. J. Med. Chem. 1994, 37, 4185.
24. Manfredini, S.; Vicentini, C. B.; Manfrini, M.; Bianchi, N.; Rutigliano, C.;
Mischiati, C.; Gambari, R. Bioorg. Med. Chem. 2000, 8, 2343.
25. Markov, V. I.; Polyakov, E. V. Organic Chemistry in Ukrine. 2005, ARKIVOC (viii),
89.
26. Broggi, J.; Joubert, N.; Aucagne, V.; Zevaco, T.; Berteina-Raboin, S.; Nolan, S. P.;
Agrofoglio, L. A. Nucleosides Nucleotides Nucleic Acids 2007, 26, 779.
27. Saito, Y.; Escuret, V.; Durantel, D.; Zoulim, F.; Schinazi, R. F.; Agrofoglio, L. A.
Bioorg. Med. Chem. 2003, 11, 3633.
28. (a) Haebich, D.; Barth, W.; Roesner, M. Heterocycles 1989, 29, 2083; (b)
Wigerinck, P.; Van Aerschot, A.; Claes, P.; Balzarini, J.; De Clercq, E.; Herdewijn,
P. J. Heterocycl. Chem. 1989, 26, 1635; (c) Zhou, L.; Amer, A.; Korn, M.; Burda, R.;
Balzarini, J.; De Clercq, E.; Kern, A. R.; Torrence, P. F. Antiviral Chem. Chemother.
2005, 16, 375.
29. (a) Kolb, H. C.; Sharpless, K. B. Drug Discovery Today 2003, 8, 1128; (b) Kolb, H.
C.; Finn, M. G.; Sharpless, K. B. Angew. Chem., Int. Ed. 2001, 40, 2004.
30. Huisgen, R. Angew. Chem., Int. Ed. Engl. 1963, 75, 742.
31. (a) Piralli, T.; Pagliai, F.; Mercurio, C.; Boggio, R.; Canonico, P. L.; Sorba, G.; Tron,
G. C.; Genazzani, A. J. Comb. Chem. 2008, 10, 624; For the use of azide- alkyne
1,3-dipolar huisgen cycloaddition in nucleoside chemistry see: (b) Amblard, F.;
Cho, J. H.; Shinazi, R. F. Chem. Rev. 2009, 109, 4207.
32. Sandrini, M. P.; Clausen, A. R.; On, S. L.; Aarestrup, F. M.; Munch-Petersen, B.;
Piskur, J. J. Antimicrob. Chemother. 2007, 60, 510.
33. Sandrini, M. P.; Shannon, O.; Clausen, A. R.; Bjorck, L.; Piskur, J. Antimicrob.
Agents Chemother. 2007, 51, 2726.
34. For microwave-assisted syntheses of nucleosides see: Roy, V.; Pradère, U.;
Agrofoglio, L. A. Future Med. Chem. 2010, 2, 177–192.
35. (a) Loupy, A. Microwaves in Organic Synthesis;; Wiley-VCH: Weinheim,
Germany, 2002. and references cited therein; (b) Perez-Balderas, F.; Ortega-
Munoz, M.; Morales-Sanfrutos, J.; Hernandez-Mateo, F.; Calvo-Flores, F. G.;
Calvo-Asin, J. A.; Isac-Garcia, J.; Santoyo-Gonzalez, F. Org. Lett. 2003, 5, 1951;
(c) Singh, P. N. D.; Muthukrishnan, S.; Murthy, R. S.; Klima, R. F.; Mandel, S. M.;
Hawk, M.; Yarbrough, N.; Gudmundsdóttir, A. D. Tetrahedron Lett. 2003, 44,
9169; (d) Khanetskyy, B.; Dallinger, D.; Kappe, C. O. J. Comb. Chem. 2004, 6, 884;
(e) Rijkers, D. T. S.; Esse, W. V.; Merks, R.; Brouwer, A. J.; Jacobs, H. J. F.; Pieters,
R. J.; Liskamp, M. J. Chem. Commun. 2005, 36, 4581.
36. Appukkuttan, P.; Dehaen, W.; Fokin, V. V.; Van der Eycken, E. Org. Lett. 2004, 6,
4223.
37. Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.; Sharpless, K. B. Angew. Chem., Int. Ed.
2002, 41, 2596.
Acknowledgments
This work was supported by grants from the Swedish Research
Council for the Environment, Agricultural Sciences and Spatial Plan-
ning (FORMAS), and the Swedish Research Council to S.E. UMPC-
Paris 6 and the Centre National de la Recherche Scientifique (Unité
Propre de Recherche 3082 and Formation de Recherche en Evolution
2852), AgenceNationaledeRechercheGrantANR-05-BLAN-0368-02
(to L.A.A. and D.D.-B.). This work was also supported in part by NIH
Grant 4R37-AI-025899, 5R37-AI-041980, 5P30-AI-50409 (CFAR),
and by the Department of Veterans Affairs to R.F.S.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmc.2010.03.023.
References and notes
38. Stuyver, L. J.; Lostia, S.; Adams, M.; Matthew, J.; Pai, B. S.; Grier, J.; Tharnish, P.;
Choi, Y.; Chong, Y.; Choo, H.; Chu, C. K.; Otto, M. J.; Schinazi, R. F. Antimicrob.
Agents Chemother. 2002, 46, 3854.
39. Sandrini, M. P.; Clausen, A. R.; Munch-Petersen, B.; Piskur, J. Nucleosides
Nucleotides Nucleic Acids 2006, 25, 1153.
40. Gerland, B.; Desire, J.; Balzarini, J.; Decout, J. L. Bioorg. Med. Chem. 2008, 16,
6824.
41. Carnrot, C.; Wang, L.; Topalis, D.; Eriksson, S. Protein Sci. 2008, 17, 1486.
42. Czernecki, S.; Valéry, J. M. Synthesis 1991, 3, 239–240.
43. Joy, S.; Nair, P. S.; Hariharan, R.; Pillai, M. R. In Silico Biol. 2006, 6, 601–605.
1. Eriksson, S.; Munch-Petersen, B.; Johansson, K.; Eklund, H. Cell. Mol. Life Sci.
2002, 59, 1327.
2. Johansson, K.; Ramaswamy, S.; Ljungcrantz, C.; Knecht, W.; Piskur, J.; Munch-
Petersen, B.; Eriksson, S.; Eklund, H. Nat. Struct. Biol. 2001, 8, 616.
3. Johansson, N. G.; Eriksson, S. Acta Biochim. Pol. 1996, 43, 143.
4. Sherley, J. L.; Kelly, T. J. J. Biol. Chem. 1988, 263, 8350.
5. Al-Madhoun, A. S.; Tjarks, W.; Eriksson, S. Mini-Rev. Med. Chem. 2004, 4, 341.
6. Hallek, M.; Wanders, L.; Strohmeyer, S.; Emmerich, B. Ann. Hematol. 1992, 65, 1.