5-Modified-2'-dU and 2'-dC as mutagenic Anti HIV-1 proliferation agents: Synthesis and activity

Article abstract of DOI:10.1021/jm901758f

With the goal of limiting HIV-I proliferation by increasing the mutation rate of the viral genome, we synthesized a series of pyrrolidine nucleoside analogues modified in position 5 of the aglycone moiety but unmodified on the sugar part. The synthetic strategies allow us to prepare the targeted compounds directly from commercially available nucleosides. All compounds were tested for their ability to reduce HIV-I. proliferation in cell culture. Two of them (5-hydroxymethyl-2'-dU (1c) and 5-hydroxymethyl-2'-dC (2c)) displayed a moderate antiviral activity in single passage experiments. The same two compounds plus two additional ones (5-carbamoyl-2'-dU (la) and 5-carbamoylmethyl-2'-dU (1b)) were potent, inhibitors of HIV-1 RT activity in serial passage assays, in which they induced a progressive loss of HIV-1 replication. In addition, viruses collected after seven passages in the presence of 1c and 2c replicated very poorly after withdrawal of these compounds, consistent with the accumulation of deleterious mutations in the HIV-1. genome.

Full text of DOI:10.1021/jm901758f

1534 J. Med. Chem. 2010, 53, 1534–1545
DOI: 10.1021/jm901758f
5-Modified-20-dU and 20-dC as Mutagenic Anti HIV-1 Proliferation Agents: Synthesis and Activity
†
†
‡
‡
§
Yazan El Safadi, Jean-Christophe Paillart, Geraldine Laumond, Anne-Marie Aubertin, Alain Burger, Roland Marquet,*
,†
ꢀ
,†
ꢀ
and Valerie Vivet-Boudou*
†
Architecture et Reꢀactiviteꢀ de l’ARN, Universiteꢀ de Strasbourg, CNRS, IBMC, 15 Rue Reneꢀ Descartes, 67084 Strasbourg Cedex, France,
‡
§
Laboratoire de Virologie, Faculteꢀde Meꢀdecine, INSERM-UMR-778, Universiteꢀde Strasbourg, 67000 Strasbourg, France, and Laboratoire de
Chimie des Moleꢀcules Bioactives et des Aro^mes, Universiteꢀde Nice Sophia Antipolis, CNRS, Institut de Chimie de Nice, Parc Valrose, 06108 Nice
Cedex 2, France
Received July 21, 2009
With the goal of limiting HIV-1 proliferation by increasing the mutation rate of the viral genome, we
synthesized a series of pyrimidine nucleoside analogues modified in position 5 of the aglycone moiety
but unmodified on the sugar part. The synthetic strategies allow us to prepare the targeted compounds
directly from commercially available nucleosides. All compounds were tested for their ability to reduce
HIV-1 proliferation in cell culture. Two of them (5-hydroxymethyl-20-dU (1c) and 5-hydroxymethyl-20-
dC (2c)) displayed a moderate antiviral activity in single passage experiments. The same two compounds
plus two additional ones (5-carbamoyl-20-dU (1a) and 5-carbamoylmethyl-20-dU (1b)) were potent
inhibitors of HIV-1 RT activity in serial passage assays, in which they induced a progressive loss of
HIV-1 replication. In addition, viruses collected after seven passages in the presence of 1c and 2c
replicated very poorly after withdrawal of these compounds, consistent with the accumulation of
deleterious mutations in the HIV-1 genome.
Introduction
was expanding, it rapidly appeared that the emergence and
selection of mutations conferring resistance to these drugs
were a major concern in AIDS treatments.8,9
According to the Joint United Nations Programme on HIV/
AIDS (UNAIDS), in 2007 more than 33 million people were
living with human immunodeficiency virus type 1 (HIV-1a),
the etiological agent of acquired immunodeficiency syndrome
(AIDS).
The rapid emergence of resistance mutations is linked to
the genetic diversity of HIV-1 virions, which is due to the low
fidelity of HIV-1 RT, enhanced by the lack of intrinsic
proofreading activity, the high level of HIV replication, and
the high rate of RT-mediated recombination.10 As a result,
combinations of drugs targeting the same or different viral
targets have been used for more than a decade to control
HIV-1 replication. Highly active antiretroviral therapies
(HAART)11 that associate atleast three antiviral drugs highly
reduced mortality rates. Under HAART, viral loads become
undetectable for several years, but the virus is never elimi-
nated from its host, and it is now widely admitted that a cure
of AIDS will require a strategy different from the existing
ones. In addition, new inhibitors that are active against
HIV-1 multiresistant strains must be developed.
Since the isolation of HIV-1 in 1983,1,2 important efforts
have been expended in developing vaccinal approaches and
antiviral chemotherapies. Whereas the first strategy has not
yet afforded sufficient positive results for use in human
treatments, the second one has provided a panel of 25 drugs
approved by the U.S. Food and Drug Administration (FDA)
for controlling HIV replication. Among them, a fusion in-
hibitor3 and a co-receptor inhibitor4 prevent the viral particles
from entering the cells. Twelve molecules target the viral
reverse transcriptase (RT), a key enzyme in the retroviral life
cycle that catalyzes the conversion of genomic RNA into
double-stranded proviral DNA.5 One drug targets the viral
integrase,6 which is responsible for proviral DNA integration
into cellular DNA. Ten other drugs are protease inhibitors7
that affect maturation of HIV particles by inhibiting Gag and
Gag-Pol precursors processing. As the pool of anti-HIV drugs
Ten years ago, Loeb et al.12 suggested that it should be
possible to push the HIV-1 exceptionally high mutation rate
from a virus self-defense system into a tool for virus destruc-
tion. This idea was based on the work of Chemistry Nobel
Prize Laureate Manfred Eigen, who described the properties
of viral populations with high mutation rates existing as
“quasispecies”. Eigen’s theory predicted that a small increase
inthe mutation ratemay resultina dramaticlossofviability.13
This mechanism is called “lethal mutagenesis” and is by
definition a process that pushes viral populations to extinc-
tion. The mutation rate in RNA viruses and retroviruses is so
high that a slight increase (1.1- to 2.8-fold for RNA viruses
such as vesicular stomatitis virus or poliovirus14 and 13-fold
for spleen necrosis virus15) in the mutation frequency can
drive the viral population to exceed the error threshold for
*To whom correspondence should be addressed. For R.M.: phone,
00 33 (0)3 88 41 70 54; fax, 00 33 (0)3 88 60 22 18; e-mail, r.marquet@
ibmc-cnrs.unistra.fr. For V.V.-B.: phone, 00 33 (0)3 88 41 70 35; fax, 00
33 (0)3 88 60 22 18; e-mail, v.vivet@ibmc-cnrs.unistra.fr.
a Abbreviations: AIDS, acquired immunodeficiency syndrome;
FDA, Food and Drug Administration; HIV-1, human immunodefi-
ciency virus type 1; HAART, highly active antiretroviral therapy; RT,
reverse transcriptase; NRTI, nucleoside reverse transcriptase inhibitors;
CC50, 50% cytotoxic concentration; IC50, 50% inhibitory concentra-
tion; TCID50, 50% tissue culture infective dose; TMEDA, tetramethy-
lethylenediamine.
r
pubs.acs.org/jmc
Published on Web 01/29/2010
2010 American Chemical Society

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 4 1535
Chart 1. Structures of Modified Pyrimidine Nucleoside Analo-
gues Known for Their Mutagenic Properties (A) and the Ones
Prepared and Evaluated in This Study (B)
RNA/DNA hybrids, which are the replication intermediates
of HIV-1 RNA, cannot. Mutagenic nucleoside could be
advantageous when used in combination with chain termina-
tor nucleoside analogues. Studies showed that HIV strains
resistant to chain terminators exhibited no cross-resistance
toward KP-1461 and that viruses cultured in the presence of
KP-1461 displayed increased sensitivity to AZT.23
Within the framework of our research, we designed several
nucleoside analogues that are modified only on the aglycone
moiety in order to remain as potential substrates of HIV-1 RT
after anabolysation into triphosphates, to base-pair with at
least two natural nucleosides, and finally to increase the error
rate in the proviral DNA.25,26 Ambiguous hydrogen-bonding
properties can be obtained by affecting several factors such as
ionization, rotation, or tautomerization of the base moiety.
For instance, a simple rotation of the exocyclic carbamoyl
group of the pseudobase of ribavirin (1,2,4-triazole-3-carbox-
amide) results in two distinct hydrogen-bonding configura-
tions that pair with either uracil or cytosine. Numerous
investigators have demonstrated the mutagenic properties of
this broad-spectrum antiviral nucleoside.18,27-32 Hydroxy-
lated pyrimidine nucleosides are another class of ambiguous
nucleosides. The presence of a hydroxyl group on the C-5
position shifts the tautomeric equilibrium toward the unpre-
ferred or rare form, leading to mispairing and potentially to
mutagenesis. 5-OH-dC was found active against HIV-112
probably because the hydroxyl group increases the imino
tautomer frequency, allowing base-pairing with guanosine
as well as adenosine.33
viability.14,16 In comparison, DNA viruses or bacteria can
accept an 80000-fold increase in mutation frequency17 without
damage to their viability. Even if there is not much theory
underpinning this concept, empirical evidence broadly sup-
ports it. Chemical mutagens have been used to slightly increase
error rates and dramatically reduce viral titers of several
viruses such as vesicular stomatitis virus,14 poliovirus 1,18
foot-and-mouth disease virus,19 and hepatitis C virus.20
In the case of HIV-1, Loeb et al. confirmed this hypothesis
using 5-hydroxy-20-deoxycytidine (5-OH-dC, Chart 1).12
Once transformed by cellular kinases into 5-OH-dCTP, this
nucleotide analogue was incorporated by HIV-1 RT opposite
the template guanosine and mispaired with deoxyadenosine
during subsequent replication of the altered DNA, resulting in
an increase of G to A substitutions. They found that a 2-fold
increase in the mutation rate of the HIV genome after
exposure to 5-OH-dC had a large effect on viral lethality.
More recently, Koronis Pharmaceuticals developed another
mutagenic nucleoside,21-23 KP-1212 (Chart 1). Following
incorporation of KP-1212 monophosphate (administrated
as a prodrug, KP-1461) and several rounds of replication,
the accumulation of mutations exceeded the crucial threshold
beyond which HIV-1 viability is highly reduced, leading to
viral collapse. As for 5-OH-dC, only a few mutations were
needed to induce a large effect on virus proliferation. KP-1212
reached phase 2a trials for clinical development before being
suspended in 2008 in order to conduct new preclinical evalua-
tion studies.
The mechanism of action of viral decay accelerators intro-
ducing mutations in the viral genome is potentially harmful
for mitochondrial and/or genomic DNA, and this risk has to
be kept in mind when developing new mutagenic nucleoside
analogues. For KP-1212, clinical studies have demonstrated
that it induced only moderate adverse effects. Moreover,
genotoxicity assays conducted with KP-1212 revealed very
little mutagenic activity comparable to approved NRTIs.23 It
was demonstrated that human mitochondrial DNA polymer-
ase γ (hDNA pol γ) efficiently incorporated KP-1212-TP
whereas mouse DNA polymerase β (mDNA pol β) did
not.24 On the other hand, KP-1212-MP can be efficiently
excised by the exonuclease activity of hDNA pol γ, limiting by
this way errors in host mitochondrial genome.23 Finally,
mismatches that are present in double-stranded DNA can
be corrected by cellular repair enzymes, while those present in
In this paper, we report the synthesis of pyrimidine nucleo-
side analogues modified on position 5 by a carbamoyl,
carbamoylmethyl, hydroxyl, or hydroxymethyl group
(Chart 1) as well as their anti-HIV-1 activity, showing that
some of them reduce viral replication in cell culture.
Chemistry
The carbamoylnucleoside derivatives 1a,b, 2a,b, and 3 are
unpublished or poorly described and characterized in litera-
ture. On the other hand, hydroxylated nucleosides 1c, 2c, and
1d are known because they were isolated as DNA oxidative
damage products.34,35 For the latter compounds, some synth-
eses were published as well as their ability to form mismatches
with natural nucleosides36-40 but no data were found con-
cerning their use as mutagenic nucleosides for antiviral pur-
poses. Consequently, we focused our efforts on developing
simple and rapid methods to achieve our targeted compounds
starting from commercially available 20-deoxynucleosides
such as 20-deoxyuridine or its 5-halogenated derivatives.
The synthesis of compounds 1a and 2a is described in
Scheme 1 and starts with the commercially available 5-iodo-
20-deoxyuridine. Classical silylation of5-iodo-20-deoxyuridine
was realized with tert-butyldimethylsilyl chloride (TBDMSCl)
in the presence of imidazole in dimethylformamide (DMF) to
give the 30,50-di-O-protected compound 4. This silylated nu-
cleoside 4 was first deprotonated on the N-3 position with
sodium hydride in THF to give the sodium salt of 4, which was
treated with sec-butyllithium (sec-BuLi) in presence of tetra-
methylethylenediamine (TMEDA) for iodine/lithium ex-
change. The mix sodium/lithium intermediate was finally
reacted in situ with ethyl chloroformate to give the C-5 adduct
5 in 67% yield. Protecting position N-3 by forming the sodium
salt allowed an optimum selectivity for the anion formation on
position C-5.41 The derivative 5 was deprotected from its silyl

1536 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 4
Scheme 1. Synthesis of Compounds 1a and 2aa
El Safadi et al.
a Reagents and conditions: (i) TBDMSCl, imidazole, DMF, room temperature, 60 h, quantitative; (ii) NaH, THF, room temperature, 30 min; then
sec-BuLi, TMEDA, -78 °C, 15 min; then ClCOOEt, -78 °C, 75 min, 67%; (iii) TBAF 3H2O, THF, room temperature, 3 h, 81%; (iv) NH4OH, pyridine,
3
80 °C, 48 h, 50%; (v) TBDMSCl, imidazole, DMF, room temperature, 60 h, 55%; (vi) TPSCl, DMAP, Et3N, CH3CN, room temperature, 20 h; then
NH4OH 25%, room temperature, 3 h, degradation; (vii) TPSCl, DMAP, Et3N, CH3CN, room temperature, 21 h; then NH4OH 25%, room temperature,
4 h, 70%; (viii) TBAF .3H2O, THF, room temperature, 20 h, 67%; (ix) 25% NH4OH, pyridine, 80 °C, 48 h, 20% (2a) and 40% (10).
3
groups by treatment with tetrabutylammonium fluoride trihy-
drate (TBAF 3H2O) in THF. The resulting deprotected com-
potassium cyanide in the presence of 18-crown-6 ether in
DMF at room temperature gave the intermediate 30,50-di-O-
acetyl-6-cyano-20-dU. The “one pot” reaction with an addi-
tional equivalent of potassium cyanide and 18-crown-6 ether
under thermal treatment led to the addition of a second cyano
group (at position 5) and the elimination of the 6-cyano group
affording the more stable 30,50-di-O-acetyl-5-cyano-20-dU.
Performing the first step of the reaction at higher temperature
and/or higher KCN concentrations decreased the reaction
yield. This strategy led to only a partial conversion of the
6-cyano intothe 5-cyano nucleoside and also to the removal of
some protecting groups. 5-Cyano-20-dU derivative 12 was
reacted under the conditions described above for the prepara-
tion of 8 to give the intermediate 13. Removal of the silylated
3
pound 6 was reacted with aqueous ammonia in pyridine to give
compound 1a in a total of four steps and a 27% overall yield
as illustrated in Scheme 1. The same synthetic route but using
5-methoxycarbonyl nucleoside as the intermediate also af-
forded compound 1a but in lower yield (16%). The conversion
of the intermediate to compound 2a was then considered.
However, after protection of the hydroxyls with silyl groups,
attempts to convert compound 7 into the cytidine analogue
with 2,4,6-triisopropylbenzenesulfonyl chloride (TPSCl) in the
presence of dimethylaminopyridine (DMAP) and triethyla-
mine in acetonitrile followed by ammonolysis with NH4OH at
room temperature42 failed because of complete degradation of
the starting material. On the other hand, conversion of com-
pound 5 into the cytidine derivative 8 using the same procedure
was efficient (70% yield). At this stage, the silylated protecting
groups of 8 were eliminated and treatment of compound 9 with
ammonia in pyridine in a sealed tube at 80 °C gave the target
nucleoside derivative 2a in 20% yield along with 5-carboxyl-20-
deoxycytidine 10 (40%).
protecting groups by TBAF 3H2O in THF followed by
3
hydrolysis of the cyano group of 14 with a 1 M sodium
hydroxide solution afforded the target compound 2a in 14%
global yield.
The preparation of compounds 1b, 1d, and 2b started with
20-deoxyuridine (Scheme 3). 5-OH-20-dU 1d was obtained
from20-dU in one step and 54% yield by reaction with bromide
in water followed by in situ treatment with pyridine.40 Wittig
coupling on 1d with carbamoyltriphenylphosphorane44 15
in 1,4-dioxane gave compound 1b in 38% total yield.45 After
silyl protection of the hydroxyl groups of 1b, conversion of the
4-carbonyl function of 16 to the corresponding amino group
was achieved following the procedure of activation/substitu-
tion as previously described for the preparation of compound
Alternatively, 2a can be prepared with a better overall yield,
14% instead of 6%, starting from 5-bromo-20-deoxyuridine
(Scheme 2). The hydroxyl groups of 5-Br-20-dU were pro-
tected to give compound 11 as described for 5-iodo-20-dU.
Cyanation at position 5 was inspired from the methodology of
“double addition elimination” developed by Inoue et al.43 In
the first step, the reaction of 5-Br-20-dU with 2 equiv of

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 4 1537
Scheme 2. Synthesis of Compound 2aa
a Reagents and conditions: (i) TBDMSCl, imidazole, DMF, room temperature, 60 h, 96%; (ii) KCN, 18-crown-6, DMF, room temperature, 60 h;
then KCN, 18-crown-6, DMF, 80 °C, 9 h, 65%; (iii) TPSCl, DMAP, Et3N, CH3CN, room temperature, 21 h; then NH4OH (27%), room temperature,
90 min, 73%; (iv) TBAF 3H2O, THF, room temperature, 4 h, 64%; (v) 1 M NaOH, room temperature, 4 h, 49%.
3
Scheme 3. Synthesis of Compounds 1b, 1c, 1d, 2b, 2c, and 3a
a Reagents and conditions: (i) Br2, H2O; then pyridine, 0 °C, 30 min, and room temperature, 24 h, 54%; (ii) 15, 1,4-dioxane, 100 °C, 4 h, 71%;
(iii) H2CO, Et3N, H2O, 100 °C, 48 h, 40%; (iv) TBDMSCl, imidazole, DMF, room temperature, 60 h, 91% (16), 69% (17) and 90% (20); (v) TPSCl,
DMAP, Et3N, CH3CN, room temperature, 21 h; then NH4OH (27%), room temperature, 90 min, 61% (18) and 2 h, 51% (19); (vi) TBAF 3H2O, THF,
3
room temperature, 2 h, 83% (2b), 54% (2c), and 65% (3); (vii) NaH, THF, room temperature, 30 min; then 2-chloroacetamide, THF, 0 °C to room
temperature, 3 h, 72%.
8. Removal of TBDMS protecting groups by TBAF 3H2O in
3
to furnish compound 1c in 40% yield according to the
methodology described by LaFrancois et al. for the prepara-
tion of isotopically labeled compounds.40 The synthetic
strategy described in Scheme 1 applied to compound 1c
afforded the corresponding cytidine analogue 2c in 19%
overall yield.
THF afforded the 5-carbamoylmethyl-20-dC, 2b (46% in three
steps).
The synthesis of compounds 1c and 2c was realized
straightforwardly as shown in Scheme 3. 20-dU was reacted
with formaldehyde in the presence of triethylamine and water

1538 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 4
El Safadi et al.
Figure 1. Viral inhibition of HIV-1 LAI (Inh.) and cellular toxicity
(Tox.) curves for compounds 1c and 2c in CEM-SS cells in single
passage experiments.
Table 1. Anti-HIV-1 Activity (IC50) and Cytotoxicity (CC50) of
5-Modified 20-dU and 20-dC Derivatives in Cell Cultures on Limited
Time (Single Passage, 5 Days)
HIV-1 LAI/CEM-SS
IC50 (μM)a CC50 (μM)b
>200 >200
HIV-1 IIIB/MT4
IC50 (μM)a
CC50 (μM)b
1a
1b
1c
>200
>200
>130
>200
>200
>200
>200
>200
0.018
>200
>200
130
>200
30
>200
120
1d
2a
2b
2c
>200
>200
>200
24
>200
>200
>200
>200
>200
>1
>200
>200
>200
>200
>200
>1
3
AZT
>200
0.0024
a IC50: 50% inhibitory concentration, which is the concentration
needed to inhibit 50% virus replication in vitro. b CC50: 50% cytotoxic
concentration, which is the concentration required to cause 50% death
of uninfected cells.
Figure 2. RT activity detected in supernatant over serial passage
experiments of HIV-1 LAI in CEM-SS cells. Panel A shows active
molecules (1a, 1b, 1c, and 2c), whereas panel B concerns inactive
compounds (1d, 2a, 2b, and 3). The concentration of all compounds
is 200 μM except for 1c (100 μM).
The final target compound 3 was obtained in three steps
as shown in Scheme 3. Silylation of 20-dU gave compound
20, which was treated successively with NaH in THF and
2-chloroacetamide to give the N-alkylated derivative 21.
Removal of the silyl protecting groups under standard con-
ditions afforded compound 3 in 42% global yield.
supernatant of cell cultures. The dose-effect curves (illustrated
for compounds 1c and 2c in Figure 1, blue curves) were plotted
to afford the IC50 values (Table 1, IC50 entries). Two molecules,
1c and 2c, displayed a moderate effect on HIV-1 replication in
CEM-SS cells (IC50 = 30 and 24 μM, respectively). However,
none of these compounds were found to exhibit anti-HIV
activity on MT-4 cells.
1H and 13C NMR, MS, UV, and elemental analysis affor-
ded analytical data in harmony with the proposed structures
of the above compounds.
Results and Discussion
5-Modified pyrimidine nucleoside derivatives were synthe-
sized in order to test their ability to reduce HIV-1 replication
by lethal mutagenesis. Their activity on HIV-1 replication was
first measured by quantification of (i) the inhibition of virus-
induced cytopathogenicity in infected MT-4 cells and (ii) the
reverse transcriptase (RT) activity associated with virus par-
ticles released from infected CEM-SS cells. The 50% cytotoxic
concentration (CC50) was evaluated in parallel to the 50%
inhibitory concentration (IC50). The dose-effect curves
(illustrated for compounds 1c and 2c in Figure 1, red curves)
were plotted with the computer-generated median effect
to give the CC50 values (Table 1, CC50 entries). In both cell
types the large majority of molecules did not induce toxicity
(CC50 > 200 μM, the higher concentration tested). Only com-
pound 1c was found to be toxic for cells with a CC50
of 120 and 130 μM in CEM-SS and MT-4 cells, respectively.
In parallel, we determined the inhibitory concentrations
(IC50) for each compound by measuring the RT activity in
In order to evaluate the potential mutagenic effect of our
compounds, serial passage experiments were performed in
CEM-SS cells infected with HIV-1 LAI. The CC50 data
allow us to determine the concentrations of our compounds
to be used in this assay. For nontoxic compounds (CC50
>
200 μM) the infected cells were treated with a 200 μM
solution, but for compound 1c (CC50=120 μM) a 100 μM
solution was used in order to limit cellular mortality.
Figure 2 shows the effect of our compounds on HIV
replication over eight sequential passages. An identical
volume of supernatant from the control and the treated
cultures was transferred to uninfected cells at each passage.
In control cultures (Figure 2, purple bars), we observed a
regular increase of RT titer until passage 4 that was most
likely a consequence of the low multiplicity of infection
used for the initial infection (passage 0). For the last four
passages, RT titer showed considerable fluctuations but
kept high values. In contrast, virus replication followed a

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 4 1539
Figure 3. Viral replication kinetics. CEM-SS cells were infected with a normalized amount of viruses (equivalent to 25 cpm/mL of RT counts)
issued from passage 3 (A) or 7 (B). Supernatant RT was quantitated over 17 days.
different trend in several cultures treated with our nucleo-
side analogues (Figure 2A).
drug. If our compounds induce lethal mutagenesis, then the
mutations accumulated in the viral genome after a number of
serial passages should exceed the error threshold and drama-
tically reduce viral fitness even after the nucleoside analogue
that produced these mutations has been removed. Alterna-
tively, if our compounds partially inhibit viral replication by
other mechanisms, their effects should disappear after they
have been removed.
In four of the eight treated cultures (Figure 2A, 1a, 1b, 1c,
and 2c), while RT activity was almost equivalent to control
values until passage 4, RT titer considerably and gradually
decreased from passage 5 to 8. RT titers in the supernatant
were almost undetectable at passage 8 for compounds 1b and
1c. The lower HIV-1 replication at passage 4 in cultures
treated with compound 1c might be due to a higher activity
of this compound or/and to its toxicity. The other four
compounds (Figure 2B, 1d, 2a, 2b, and 3) did not show any
significant difference with the untreated control in limiting
replication of HIV-1 after passage 4.
The results obtained with 1a, 1b, 1c, and 2c are consistent
with lethal mutagenesis resulting from a progressive accumu-
lation of mutations in the HIV genome due to the incorpora-
tion of mutagenic analogues during repetitiveinfectionof host
cells.
In order to mimic the in vivo situation, we chose to infect
untreated and treated cultures with identical volumes of
supernatant rather than with identical amounts of virus.
However, by doing so, we cannot exclude that the progressive
decrease in HIV-1 titers observed with compounds 1a, 1b, 1c,
and 2c was due to partial inhibition of the virus replication at
each passage, although the increase of viral replication during
the first four passages seems to argue against this possibility.
In order to distinguish between lethal mutagenesis and partial
inhibition due to other mechanisms, we used a normalized
amount of viruses (based on RT activity) produced in the
presence or absence of nucleoside analogues at the end of
passages 3 and 7 to infect fresh CEM-SS cells in the absence of
Viruses collected after three passages in the absence of
nucleoside analogue or in the presence of 1a, 1b, 1c, 2b, or 2c
all replicated efficiently (Figure 3A). This result was expected,
as these compounds showed no effect on HIV-1 replication
after three serial passages (Figure 2). At the opposite, viruses
collected after seven passages in the presence of compounds
1c and 2c hardly replicated at all in the absence of nucleoside
analogue (Figure 3B). Thus, this experiment unambiguously
demonstrates that 1c and 2c induce lethal mutagenesis of
HIV-1. However, viruses collected after seven passages in
the presence of 1a and 1b replicated as efficiently as viruses
passaged in the absence of nucleoside analogues (Figure 3B),
even though these compounds were active in the serial passage
assay (Figure 2A). Thus, 1a and 1b do not induce lethal
mutagenesis, even though they partially inhibit viral replica-
tion. The mechanism of action of these compounds is presently
unknown. Finally, viruses collected after seven passages in the
presence of 2b also replicated as efficiently as viruses passaged
in the absence of nucleoside analogues (Figure 3B). This result
was expected, since this compound was also inactive in the two
previous assays (Table 1 and Figure 2).
The first relationship between the structure of our com-
pounds and their effects on HIV replication over several

1540 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 4
El Safadi et al.
bonding may stabilize the rare tautomers 1c* and 2c* as
illustrated in Figure 4 and allow base pairing of 1c and 2c
with either A or G. The hydroxymethyl group might also be
directly involved in intermolecular base-pairing, stabilizing
noncanonical base-pairs and favoring mutagenicity.37 How-
ever, additional experiments will have to be conducted to test
these hypotheses.
Conclusion
Among the dU and dC analogues we synthesized and tested
in this study, two displayed anti-HIV-1 activity in single
passage assays. The same two compounds were the most
efficient in the serial passage assay, and viruses collected after
seven consecutive passages in the presence of these nucleoside
analogues failed to replicate after withdrawal of the drugs.
The progressive loss of HIV-1 replication is consistent with
deleterious mutagenesis of HIV-1 induced by these two
compounds. The data presented here are also consistent with
a detailed analysis of the viral RT gene before the first and
after the eighth passage (El Safadi, Paillart, et al., manuscript
in preparation). The two mutagenic compounds are charac-
terized by a hydroxymethyl substituent at position C-5 of dU
and dC. Intramolecular hydrogen bonding or inter-residue
hydrogen bonding between these groups and neighboring
bases in DNA strand might favor misincorporation of these
residues and explain the observed mutagenic activity. Two dU
analogues with carbamoyl and carbamoylmethyl groups at
position C-5 inhibited HIV-1 replication during serial pas-
sages, but this was the result of partial inhibition of viral
replication at each passage rather than accumulation of
deleterious mutations. Additional biochemical studies would
be required to elucidate the mechanism of action of these two
dU analogues.
Figure 4. Base-pairing schemes involving “predominant” (left) and
“rare” (/, right) tautomers for compounds 1c and 2c.
passages reveals three features. (1) Only modification of posi-
tion 5 by a hydroxymethyl group (compounds 1c and 2c)
confers an anti HIV-1 mutagenic activity, albeit compound 1c
shows significant cytotoxicity whereas compound 2c does not
at the highest concentration tested. (2) Surprisingly, modifica-
tion of position 5 of dU by a hydroxyl group (compound 1d)
has no effect, whereas this modification has a positive effect on
the anti-HIV activity in the dC series (5-OH-dC described by
Loeb et al.12). (3) Introduction of carbamoyl or carbamoyl-
methyl modification in both series does not induce mutagenic
activity, even if in the dU series compounds 1a and 1b reduce
HIV proliferation over several passages.
It is difficult to identify the origin of the lack of mutagenic
activity of nucleoside analogues because it may be directly
linked to intrinsic molecule inactivity but also to several
indirect effects such as a poor phosphorylation or a lack of
recognition and incorporation by the viral polymerase. How-
ever, the difference between compounds 1d and 5-OH-dC
anti-HIV activity can be related to the stabilization of ionized
forms or raretautomersthrough the hydroxyl groupinductive
and fields effects. La Francois et al. established a linear
correlation between the σm Hammet values for a substituent
and the pK observed in the dU and dC series.35 Reducing σm
(from electron-attracting to electron-donating groups) in-
creases pKa in the dU series and pKb in the dC series, thus
Experimental Section
Materials and Methods. All chemicals are of pa purity and
purchased from Sigma. Organic solvents were dried and distilled
when necessary. Pyridine and triethylamine were distilled
over KOH. CH2Cl2 was distilled over P2O5. MeOH and THF
were distilled over Na in the presence of benzophenone for
THF. Flash chromatography was performed on silica gel
60 (40-63 μM) from Merck. Thin layer chromatography was
performed on Merck 60F254 coated plates. UV spectra were
recorded on a Nanodrop ND-100 spectrophotometer. pH was
measured on a Schott CG-825 apparatus. Proton and carbon
nuclear magnetic resonance (1H and 13C NMR) were recorded
at 300 MHz on a Bruker Advance spectrometer. Chemical shifts
were reported in ppm relative to tetramethylsilane (TMS), and
signals are given as follows: s (singlet); d (doublet); t (triplet); m
(multiplet). Mass spectra were recorded on Bruker Daltonics
MicroTOF LC spectrometer using positive electron spray ioni-
zation (ESI) mode. Elemental analyses were performed with an
Elementar VARIO ELIII CHN analyzer, and the purity of the
target compounds was more than 95%.
1
increasing the affinity of N-3 for a proton. Moreover a H
NMR study of a broad series of C-5 substituted 20-deoxyur-
idine concluded that electron-withdrawing groups cause a
downfield shift of H-6, whereas electron-donating groups
induce an upfield shift.46 As a result, the effect of C-5-
substituents can be estimated through the value of the H-6
displacement. For dU and compound 1d these values are in
accordance with the ones previously published.35 They indi-
cate that the hydroxyl modification induces an electron
donating effect (7.33 ppm for 1d compared to 7.71 ppm for
20-dU). Whereasthis electron donating effect may enhance the
frequency of the imino form relative to the amino tautomer in
dC series, at the opposite, the stabilization of the keto
tautomer in dU series is unfavorable to multibase pairing
and mutagenicity.
30,50-Di-O-tert-butyldimethylsilyl-5-iodo-20-deoxyuridine (4).
To a solution of 5-iodo-20-deoxyuridine (0.765 g, 2.16 mmol)
in dry DMF (4.5 mL) tert-butyldimethylsilyl chloride (2.10 g,
14 mmol) and imidazole (1.40 g, 20 mmol) were added. The
clear solution was stirred at room temperature for 60 h. Water
(40 mL) was added, the aqueous layer was extracted with EtOAc
(3 ꢀ 50 mL), and the combined organic layers were washed with
brine (3 ꢀ 50 mL), dried over Na2SO4, and concentrated to
dryness. The oily residue was purified by silica gel column
chromatography (0-3% MeOH in CH2Cl2) to give pure com-
Concerning the activity observed with compounds 1c and
2c, it cannot be assigned to the substituent electronic effects,
since the methylene group between the hydroxyl and the
heterocyclic ring reduces dramatically the spread of the OH
effect as confirmed by the 1H NMR H-6 displacements (7.73
and 7.75 ppm compared to 7.71 and 7.70 ppm for dU and
dC, respectively). Alternatively, intramolecular hydrogen
1
pound 4 (1.24 g, 99%) as a white foam. H NMR (DMSO-d6,
300 MHz) δ ppm: 0.1-0.2 (m, 12H, CH3-Si); 0.8-1.0 (2s, 18H,

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 4 1541
tBu-Si); 2.1-2.3 (m, 2H, CH2-20 and 200); 3.7-3.9 (m, 3H, CH-40,
CH2-50 and 500); 4.3-4.4 (m, 1H, CH-30); 6.09 (t, 1H, CH-10,
150.87 (C-2); 163.69 (C-4); 171.76 (CO-amide). MS (ESI, pos) m/
z: 294.1 (M þ Naþ). UV (H2O, 30 μM): λmax = 276 nm, ε =
10400 mol-1 L cm-1. Anal. (C10H13N3O6) C, H, N.
3
3
0
0
0
00
JH1 -H2 = JH1 -H2 = 7.2 Hz); 7.97 (s, 1H, CH-6); 11.74 (s, 1H,
3
3
NH-3). 13C NMR (DMSO-d6, 75 MHz) δ ppm: -5.07 and -4.40
(4C, CH3-Si); 18.08 (2C, C(CH3)3); 26.08 and 26.30 (6C, C-
(CH3)3); 40.34 (C-20); 63.05 (C-50); 70.40 (C-30); 72.37 (C-10);
84.95 (C-40); 87.85 (C-5); 144.40 (C-6); 150.46 (C-2); 160.84 (C-4).
MS (ESI, pos) m/z: 581.4 (M þ Hþ).
30,50-Di-O-tert-butyldimethylsilyl-5-ethylcarboxylate-20-deoxy-
cytidine (8). A solution of compound 5 (0.840 g, 1.59 mmol) in
anhydrous acetonitrile (25 mL) was treated with freshly distilled
triethylamine (0.7 mL, 5.03 mmol), DMAP (0.590 g, 4.88 mmol),
and triisopropylbenzenesulfonyl chloride (TPSCl, 1.46 g, 4.43
mmol). The yellow solution rapidly turned red. After the mixture
was stirred for 21 h at room temperature, a 25% aqueous
solution of ammonium hydroxide (35 mL) was added. The
reaction mixture was slightly decolorized, and stirring was
maintained for 1.5 h. Solvents were removed under vacuum,
and the brown residue was partitioned between water (50 mL)
and EtOAc (50 mL). The aqueous layer was extracted with
EtOAc (3 ꢀ 50 mL), and the combined organic layers were
washed with a saturated solution of NaHCO3 (2 ꢀ 50 mL) and
brine (2 ꢀ 50 mL). The organic layer was dried over Na2SO4 and
concentrated to dryness under reduced pressure. Purification by
silica gel column chromatography (0-5% MeOH in CH2Cl2)
gave pure 8 (0.59 g, 70%) as a white foam. 1H NMR (DMSO-d6,
300 MHz) δ ppm: 0.06 and 0.09 (2s, 12H, CH3-Si); 0.84 and 0.89
30,50-Di-O-tert-butyldimethylsilyl-5-ethylcarboxylate-20-deoxy-
uridine (5). To a solution of compound 4 (0.880 g, 1.51 mmol) in
freshly distilled THF (3 mL) was slowly added a suspension
of sodium hydride (0.160 g, 4 mmol) in freshly distilled THF
(6 mL). The reaction mixture was stirred at room temperature
for 30 min, then cooled to -78 °C. N,N0-Tetramethylethylene-
diamine (TMEDA, 0.5 mL, 3 mmol) was added followed by the
dropwise addition of a 1.3 M solution of sec-BuLi in cyclohexane
(2.7 mL, 3.51 mmol). The reaction mixture was stirred for 15 min.
Then freshly distilled ethyl chloroformate (2 mL, 21.1 mmol)
was slowly added, and the solution was stirred for an additional
75 min at -78 °C. Reaction was quenched at low temperature by
adding water (5 mL). The aqueous layer was extracted by EtOAc
(4 ꢀ 20 mL), andcombinedorganic layers werewashedwithbrine
(3 ꢀ 20 mL), dried over Na2SO4, and concentrated to dryness
under reduced pressure. The solid material was purified by silica
gel chromatography (10-25% EtOAc in petroleum ether) to give
t
3
(2s, 18H, Bu-Si); 1.28 (t, 3H, CH3-CH2, JCH3-CH2 = 7.3 Hz)
2.1-2.2 (m, 1H, CH2-20); 2.5-2.6 (m, 1H, CH2-200); 3.7-3.8 (m,
2H, CH2-50 and 500); 3.9-4.0 (m, 1H, CH-40); 4.2-4.3 (m, 2H,
1
3
CH3-CH2);4.3-4.4 (m, 1H, CH-30); 6.16 (t, 1H, CH-10, JH1 -H2
pure compound 5 (0.530 g, 67%) as a white foam. H NMR
0
0
=
3
(DMSO-d6, 300MHz) δ ppm: 0.0-0.1(m, 12H, CH3-Si);0.8-0.9
(2s, 18H, tBu-Si); 1.23 (dd, 3H, CH3-CH2-O, 3J1 = 6.9 Hz, 3J2 =
7.3 Hz); 2.2-2.3 (m, 2H, CH2-20 and 200); 3.7-3.8 (m, 2H, CH3-
0
00
JH1 -H2 = 6.2 Hz); 7.66 and 7.99 (2s, 2H, NH2); 8.45 (s, 1H,
CH-6). 13C NMR (DMSO-d6, 75 MHz) δ ppm: -5.56 and -4.74
(CH3-Si); 15.26 (CH3-CH2); 18.53 and 18.87 (C-(CH3)3); 24.96
and 25.84 (C-(CH3)3); 43.76 (C-20); 63.38 and 63.59 (CH3-CH2
and C-50); 72.97 (C-30); 87.26 (C-10); 90.34 (C-40); 104.06 (C-5);
150.85 (C-6); 156.35 (C-2); 164.95 (CO-ester); 168.58 (C-4). MS
(ESI, pos) m/z: 550.43 (M þ Naþ).
CH2); 3.9-4.0 (m, 1H, CH2-50); 4.1-4.3 (m, 2H, CH-40 and
3
CH2-500); 4.3-4.4 (m, 1H, CH-30); 6.03 (t, 1H, CH-10, JH1 -H2
0
0
=
JH1 -H2 = 6.6 Hz); 8.31 (s, 1H, CH-6); 11.60 (s, 1H, NH-3). 13C
NMR (DMSO-d6, 75 MHz) δ ppm: -5.52 and -4.63 (4C, CH3-
Si); 13.7 (CH3-CH2-O); 18.11 and 18.33 (2C, C(CH3)3); 25.09
and 25.91 (6C, C(CH3)3); 42.18 (C-20); 60,8 (CH3-CH2-O); 63.14
(C-50); 72.72 (C-30); 86.99 (C-10); 88.89 (C-40); 105.04 (C-5);
147.36 (C-6); 149.18 (C-2); 158.99 (C-4); 163.22 (CO-ester). MS
(ESI, pos) m/z: 527.6 (M þ Hþ).
3
0
00
5-Ethylcarboxylate-20-deoxycytidine (9). To a solution of
compound 8 (0.580 g, 1.1 mmol) in THF (8 mL) was added
TBAF 3H2O (1.20 g, 3.8 mmol). The reaction mixture was stirred
3
at room temperature for 16 h. Then solvent was removed and the
crude material was purified by silica gel chromatography (2-10%
MeOH in CH2Cl2) to give pure9 (0.22 g, 67%) as a white solid. 1H
NMR (DMSO-d6, 300 MHz) δ ppm: 1.30 (t, 3H, CH3-CH2,
3JCH3-CH2 = 6.9 Hz) 2.0-2.1 (m, 1H, CH2-20); 2.2-2.4 (m, 1H,
CH2-200); 3.6-3.7 (m, 2H, CH2-50 and 500); 3.8-3.9 (m, 1H, CH-
40); 4.1-4.2 (m, 3H, CH3-CH2 and CH-30); 5.05 (br s, 1H, OH-50);
5-Ethylcarboxylate-20-deoxyuridine (6). To a solution of com-
pound 5 (0.500 g, 0.95 mmol) in THF (8 mL), TBAF 3H2O
3
(0.900 g, 2.85 mmol) was added. The solution was stirred at
room temperature for 3 h. Solvent was removed under reduced
pressure and the brown oily residue was purified by silica gel
column chromatography (0-10% MeOH in CH2Cl2), affording
3
3
5.26 (br s, 1H, OH-30); 6.08 (t, 1H, CH-10, JH1 -H2 = JH1 -H2
=
0
0
0
00
1
pure compound 6 (0.230 g, 81%). H NMR (DMSO-d6, 300
5.9 Hz); 7.65 and 7.95 (2s, 2H, NH2); 8.96 (s, 1H, CH-6). 13C NMR
(DMSO-d6, 75 MHz) δ ppm: 14.96 (CH3-CH2); 43.46 (C-20); 61.57
and 61.68 (CH3-CH2 and C-50); 69.54 (C-30); 84.25 (C-10); 86.95
(C-40); 104.32 (C-5); 149.64 (C-6); 155.38 (C-2); 164.09 (CO-ester);
167.98 (C-4). MS (ESI, pos) m/z: 300.40 (M þ Hþ).
MHz) δ ppm: 1.24 (t, 3H, CH3-CH2-O, 3J = 6.9 Hz); 2.1-2.3
(m, 2H, CH2-20 and 200); 3.5-3.7 (m, 2H, CH2-50 and 500); 3.8-
3.9 (m, 1H, CH-40); 4.17 (q, 2H, CH3-CH2, 3J = 6.9 Hz); 4.2-
4.3 (m, 1H, CH-30); 5.0-5.1 (br s, 1H, OH-50); 5.2-5.3 (br s, 1H,
3
3
OH-30); 6.12 (dd, 1H, CH-10, JH1 -H2 = 6.5, JH1 -H2 = 6.2 Hz);
8.79 (s, 1H, CH-6); 11.53 (s, 1H, NH-3). 13C NMR(DMSO-d6, 75
MHz) δ ppm: 13.9 (CH3-CH2-O); 39.28 (C-20); 60,6 (CH3-CH2-
O); 60.35 (C-50); 69.45 (C-30); 82.18 (C-10); 89.56 (C-40); 102.26
(C-5); 149.32 (C-2); 151.84 (C-6); 159.77 (C-4); 164.14 (CO-ester).
MS (ESI, pos) m/z: 322.3 (M þ Naþ).
5-Carbamoyl-20-deoxycytidine (2a). A solution of compound
9 (0.189 g, 0.63 mmol) in a mixture pyridine-25% ammonia
(1-3 mL) was heated to 80 °C in a sealed tube for 48 h. Solvents
were eliminated under reduced pressure, and crude material was
coevaporated once with water. The resulting white powder was
purified by silica gel chromatography (10-40% MeOH in
CH2Cl2) and pure fractions were pooled, concentrated, and
filtrated through PTFE membrane Minisart SRP 15 (0.20 μm,
Sartorius) unit to give pure compound 2a (0.034 g, 20%) as a
white powder. 1H NMR (DMSO-d6, 300 MHz) δ ppm: 2.1-2.2
(m, 2H, CH2-20 and 200); 3.5-3.7 (m, 2H, CH2-50 and 500);
0
0
0
00
5-Carbamoyl-20-deoxyuridine (1a). A solution of compound 6
(0.220 g, 0.74 mmol) in a pyridine/25% ammonia mixture
(5 mL, 1/4) was heated at 80 °C in a sealed tube for 48 h. After
the mixture was cooled to room temperature, solvents were
removed under vacuum and the crude residue was coevaporated
twice with water and dissolved in methanol for crystallization.
Pure compound 1a was obtained as a white solid (0.099 g, 50%).
1H NMR (DMSO-d6, 300 MHz) δ ppm: 2.0-2.1 (m, 2H, CH2-20
3.7-3.9 (m, 1H, CH-40); 4.1-4.3 (m, 1H, CH-30); 5.08 (t, 1H,
3
OH-50, JOH5 -H5 = 3JOH5 -H5 = 5.4 Hz); 5.22 (d, 1H, OH-30,
0
0
0
00
3
3
JOH3 -H3 = 4.2 Hz); 6.11 (t, 1H, CH-1 , JH1 -H2 = JH1 -H2 =
3
0
0
0
0
0
0
00
and 200); 3.5-3.6 (m, 2H, CH2-50 and 500); 3.7-3.8 (m, 1H,
6.5 Hz); 7.3-7.6 (m, 2H, NH2-4); 7.72 and 8.32 (2s, 2H, NH2-
amide); 8.41 (s, 1H, CH-6). 13C NMR (DMSO-d6, 75 MHz) δ
ppm: 40.20 (C-20); 61.60 (C-50); 70.79 (C-30); 88.78 (C-10); 89.36
(C-40); 99.07 (C-5); 144.75 (C-6); 156.06 (C-2); 163.90 (C-4);
169.22 (CONH2). MS (ESI, pos) m/z: 293.21 (M þ Naþ). UV
3
CH-40); 4.2-4.3 (m, 1H, CH-30); 5.00 (t, 1H, OH-50, JOH5 -H5
0
0
=
JOH5 -H5 = 5.1 Hz); 5.28 (d, 1H, OH-3 , JOH3 -H3 = 4.4); 6.13
3
3
0
0
00
0
0
3
3
(dd, 1H, CH-10, JH1 -H2 = 6.9 Hz, JH1 -H2 = 6.6 Hz); 7.52 and
0
0
0
00
8.12 (2s, 2H, NH2); 8.66 (s, 1H, CH-6); 11.85 (s, 1H, NH-3). 13
C
NMR (DMSO-d6, 75 MHz) δ ppm: 33.58 (C-20); 61,79 (C-50);
(H2O, 30 μM): λmax = 284 nm, ε = 13 500 mol-1 L cm-1
.
3
3
70.79 (C-30); 84.38 (C-10); 87.70 (C-40); 109.36 (C-5); 138.69 (C-6);
Anal. (C10H14N4O5) C, H, N.

1542 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 4
El Safadi et al.
5-Carboxyl-20-deoxycytidine (10). Compound 10 (0.069 g,
40%) was obtained as a side product during the transformation
of ethyl carboxylate 9 into carboxamide derivative 2a. 1H NMR
(D2O, 300 MHz) δ ppm: 2.1-2.2 (m, 1H, CH2-20); 2.2-2.4 (m,
88.86 (C-40); 106.02 (C-5); 115.42 (CN); 150.73 (C-6); 154.21
(C-2); 164.08 (C-4). MS (ESI, pos) m/z: 480.48 (M þ Hþ).
5-Cyano-20-deoxycytidine (14). Removal of silyl protecting
groups from compound 13 (0.950 g, 1.98 mmol) was achieved by
1H, CH2-200); 3.6-3.8 (m, 2H, CH2-50 and 500); 3.9-4.0 (m, 1H,
reaction with TBAF 3H2O (2.19 g, 6.93 mmol) in THF (15 mL)
3
3
CH-40); 4.2-4.4 (m, 1H, CH-30); 6.11 (t, 1H, CH-10, JH1 -H2
as described for compound 5. Purification by silica gel column
chromatography (0-12% MeOH in CH2Cl2) and crystallization
from MeOH afforded compound 14 (0.32 g, 64%) as a pale-yellow
solid. 1H NMR (DMSO-d6, 300 MHz) δ ppm: 2.0-2.1 (m, 1H,
CH2-20); 2.1-2.3 (m, 1H, CH2-200); 3.5-3.7 (m, 2H, CH2-50 and
0
0
=
JH1 -H2 = 6.6 Hz); 8.36 (s, 1H, CH-6). 13C NMR (D2O, 75
MHz) δ ppm: 39.63 (C-20); 61.21 (C-50); 70.45 (C-30); 86.40
(C-10); 86.77 (C-40); 103.08 (C-5); 146.25 (C-6); 156.67 (C-2); 164.83
(C-4); 170.66 (COOH). MS (ESI, pos) m/z: 272.5 (M þ Hþ).
Alternative to the Preparation of Compound 2a. 30,50-Di-O-
tert-butyldimethylsilyl-5-bromo-20-deoxyuridine (11). To a solu-
tion of 5-bromo-20-deoxyuridine (3.99 g, 12.98 mmol) in dry
DMF (26 mL), tert-butyldimethylsilyl chloride (11.80 g, 77.88
mmol) and imidazole (7.00 g, 103.8 mmol) were added. The clear
solution was stirred at room temperature for 60 h and treated as
described for compound 4. Pure compound 11 (6.69 g, 96%) was
3
0
00
500); 3.7-3.8 (m, 1H, CH-40); 4.1-4.2 (m, 1H, CH-30); 5.5-5.8
3
(m, 2H, OH-30 and OH-50); 6.01 (t, 1H, CH-10, JH1 -H2
0
0
=
3
0
00
JH1 -H2 = 6.0 Hz); 7.2-7.6 (2 br s, 2H, NH2); 8.71 (s, 1H,
CH-6). 13C NMR (DMSO-d6, 75 MHz) δ ppm: 41.42 (C-20); 60.89
(C-50); 79.15 (C-30); 86.71 (C-10); 88.20 (C-40); 106.38 (C-5); 115.34
(CN); 150.83 (C-6); 153.03 (C-2); 162.82 (C-4). MS (ESI, pos) m/z:
275.19 (M þ Naþ). UV (H2O, 30 μM): λmax = 281 nm, ε = 9300
mol-1 L cm-1. Results are in accordance with 1H NMR and UV
1
obtained as a white foam. H NMR (DMSO-d6, 300 MHz) δ
3
3
t
ppm: 0.1-0.2 (m, 12H, CH3-Si); 0.88 and 0.91 (2s, 18H, Bu-
data published by Lahoud et al.47
Si); 2.1-2.3 (m, 2H, CH2-20 and 200); 3.7-4.0 (m, 3H, CH-40,
CH2-50 and 500); 4.3-4.4 (m, 1H, CH-30); 6.10 (dd, 1H, CH-10,
5-Carbamoyl-20-deoxycytidine (2a). Compound 14 (0.210 g,
0.83 mmol) in water (60 mL) was treated with a 1 M NaOH
solution (6 mL). After 4 h of being stirred at room temperature,
the reaction mixture was neutralized with Dowex 50WX8
hydrogen form resin. Purification by silica gel column chroma-
tography (10-30% MeOH in CH2Cl2) followed by filtration
through PTFE membrane Minisart SRP 15 (0.20 μm, Sartorius)
unit afforded pure compound 2a (0.11 g, 49%) as a white foam.
All characteristics are in accordance with the ones previously
described.
3
3
0
0
0
00
JH1 -H2 = 6.5 Hz, JH1 -H2 = 6.9 Hz); 7.99 (s, 1H, CH-6);
11.86 (s, 1H, NH-3). 13C NMR (DMSO-d6, 75 MHz) δ ppm:
-4.50 and -4.36 (4C, CH3-Si); 18.10 (2C, C(CH3)3); 26.07 and
26.32 (6C, C(CH3)3); 40.34 (C-20); 62.38 (C-50); 70.68 (C-30);
86.15 (C-10); 87.62 (C-40); 88.06 (C-5); 149.26 (C-6); 149.37 (C-
2); 160.54 (C-4). MS (ESI, pos) m/z: 534.68 (M þ Hþ).
30,50-Di-O-tert-butyldimethylsilyl-5-cyano-20-deoxyuridine (12).
To a solution of compound 11 (3.60 g, 6.73 mmol) in dry DMF
(14 mL) were added KCN (0.880 g, 13.13 mmol) and 18-crown-6
(3.20 g, 12.12 mmol). The clear solution was stirred at room
temperature for 60 h. Then additional amounts of KCN (0.15 g,
2.24 mmol) and 18-crown-6 (0.43 g, 1.63 mmol) were added, and
the solution was heated at 85 °C for 9 h. The orange reaction
mixture was allowed to cool to room temperature, diluted with a
EtOAc/H2O (20 mL, v/v, 50/50) mixture, and quickly neutralized
with a 1 M HCl solution. Caution: This step should be done under
a well-ventilated fume hood, and neutralization should be con-
ducted with care to avoid acidification of the solution and the
concomitant formation of the highly toxic HCN gas. Layers were
separated, and the aqueous layer was extracted with AcOEt
(4 ꢀ 50 mL). The combined extracts were washed with brine
(2 ꢀ 50 mL), driedover Na2SO4, and concentrated to dryness. The
residue was purified by silica gel column chromatography (0-3%
MeOH in CH2Cl2) to give 12 (2.10 g, 65%) as a white foam. 1H
NMR (DMSO-d6, 300 MHz) δ ppm: 0.1-0.2 (m, 12H, CH3-Si);
0.88 and 0.91 (2s, 18H, tBu-Si); 2.1-2.4 (m, 2H, CH2-20 and 200);
5-Hydroxy-20-deoxyuridine (1d). Starting from 20-deoxyuri-
dine (3.47 g, 15.2 mmol), 5-hydroxy-20-deoxyuridine 1d (2.00 g,
54%) was prepared as previously described by La Francois et
al.40 1H NMR and MS data are in accordance with the data
previously published. 1H NMR (D2O, 300 MHz) δ ppm:
2.0-2.1 (m, 2H, CH2-20 and 200); 3.5-3.6 (m, 2H, CH2-50 and
500); 3.7-3.8 (m, 1H, CH-40); 4.2-4.3 (m, 1H, CH-30); 5.00 (dd,
3
1H, OH-50, JOH5 -H5 = 5.1 Hz, 3JOH5 -H5 = 4.8 Hz); 5.22 (d,
0
0
0
00
3
3
1H, OH-30, JOH3 -H3 = 4.0 Hz); 6.19 (dd, 1H, CH-1 , JH1 -H2
=
0
0
0
0
0
6.6 Hz, 3JH1 -H2 = 7.3 Hz); 7.33 (s, 1H, CH-6); 8.67 (s, 1H, OH-
5); 11.43 (s,1H, NH-3). 13C NMR (D2O, 75 MHz) δ ppm: 41.43
(C-20); 62.65 (C-50); 71.54 (C-30); 86.06 (C-10); 88.93 (C-40); 123.33
(C-6); 132.65 (C-5); 151.85 (C-2); 162.07 (C-4). MS (ESI, pos) m/
z: 267.36 (M þ Naþ). UV (H2O, 30 μM): λmax = 278 nm, ε =
7600 mol-1 L cm-1. Anal. (C9H12N2O6) C, H, N.
0
00
3
3
Carbamoyltriphenylphosphorane (15) was prepared as de-
scribed by Trippett et al.44 in 66% yield. 1H NMR (CDCl3,
300 MHz) δ ppm: 1.92 (s, 1H, CH vinyl); 7.2-7.7 (m, 15H, CH
phenyl). 13C NMR (CDCl3, 75 MHz) δ ppm: 128.45 (CH vinyl);
128.60 (9C, CH-m and CH-p phenyl); 132.00 and 132.13 (6C,
CH-o aromatic); 133.19 (3C, C aromatic); 172.50 (CO amide).
31P NMR (CDCl3, 300 MHz) δ ppm: 3.04 (1P).
3.7-3.8 (m, 1H, CH2-50); 3.8-4.0 (m, 2H, CH-40, CH2-500);
3
4.3-4.4 (m, 1H, CH-30); 6.02 (dd, 1H, CH-10, JH1 -H2 = 6.0
0
0
Hz, 3JH1 -H2 = 6.1 Hz); 8.44 (s, 1H, CH-6); 12.07 (s, 1H, NH-3).
13C NMR (DMSO-d6, 75 MHz) δ ppm: -4.65 and -4.42 (4C,
CH3-Si); 18.11 (2C, C(CH3)3); 26.09 and 26.34 (6C, C(CH3)3);
40.32 (C-20); 62.77 (C-50); 70.54 (C-30); 82.95 (C-10); 85.37 (C-40);
88.06 (C-5); 115.53 (CN); 150.28 (C-2); 150.95 (C-6); 160.73 (C-4).
MS (ESI, pos) m/z: 481.52 (M þ Hþ).
0
00
5-Carbamoylmethyl-20-deoxyuridine (1b). To a solution of 5-
hydroxy-20-deoxyuridine 1d (1.65 g, 6.76 mmol) in 1,4-dioxane
(37 mL) was added carbamoyltriphenylphosphorane 15 (4.30 g,
13.48 mmol). The yellow solution was stirred for 4 h at 100 °C.
Solvent was eliminated under reduced pressure and the crude
material was crystallized from methanol to give pure compound
30,50-Di-O-tert-butyldimethylsilyl-5-cyano-20-deoxycytidine (13).
A solution of compound 12 (1.90 g, 3.95 mmol) in anhydrous
acetonitrile(50mL) wastreatedwithfreshlydistilledtriethylamine
(2.2 mL, 8.6 mmol), DMAP (2.01 g, 16.6 mmol), and tri-isopro-
pylbenzenesulfonyl chloride (TPSCl, 5.51 g, 18.2 mmol), followed
by ammonolysis as for compound 8. The residue was purified by
silica gel chromatography (0-10% MeOH in CH2Cl2) to give 13
(1.37 g, 72%) as a white foam. 1H NMR (DMSO-d6, 300 MHz) δ
ppm: 0.0-0.1 (m, 12H, CH3-Si); 0.88 and 0.90 (2s, 18H, tBu-Si);
1
1b (1.22 g, 71%) as a white solid. H NMR (DMSO-d6, 300
MHz) δ ppm: 2.0-2.1 (m, 2H, CH2-20 and 200); 3.03 (s, 2H, C5-
CH2); 3.5-3.6 (m, 2H, CH2-50 and 500); 3.7-3.8 (m, 1H, CH-40);
3
4.2-4.3 (m, 1H, CH-30); 4.99 (dd, 1H, OH-50, JOH5 -H5 = 5.2
0
0
3
Hz, 3JOH5 -H5 = 5.1 Hz); 5.24 (d, 1H, OH-30, JOH3 -H3 = 4.2
0
00
0
0
3
3
Hz); 6.17 (dd, 1H, CH-10, JH1 -H2 = 6.7 Hz, JH1 -H2 = 7.0
Hz); 6.85 and 7.28 (2s, 2H, NH2); 7.73 (s, 1H, CH-6); 11.31 (s,
1H, NH-3). 13C NMR (DMSO-d6, 75 MHz) δ ppm: 32.98 (C5-
CH2); 37.86 (C-20); 61.03 (C-50); 70.38 (C-30); 85.22 (C-10); 86.64
(C-40); 108.79 (C-5); 140.09 (C-6); 151.43 (C-2); 165.69 (C-4);
175.46 (CO-amide). MS (ESI, pos) m/z: 308.34 (M þ Naþ). UV
(H2O, 30 μM): λmax = 266 nm, ε = 7500 mol-1 L cm-1. Anal.
0
0
0
00
2.1-2.3 (m, 2H, CH2-20 and 200); 3.7-3.9 (m, 3H, CH-40, CH2-50
3
and 500); 4.3-4.4 (m, 1H, CH-30); 6.02 (t, 1H, CH-10, JH1 -H2
0
0
=
3
0
00
JH1 -H2 = 6.4 Hz); 7.55 and 8.03 (2s, 2H, NH2); 8.40 (s, 1H,
CH-6). 13C NMR (DMSO-d6, 75 MHz) δ ppm: -5.52 and -4.64
(CH3-Si); 18.11 and 18.35 (C-(CH3)3); 25.06 and 25.92
(C-(CH3)3); 41.97 (C-20); 62.76 (C-50); 72.72 (C-30); 86.59 (C-10);
3
3
(C11H15N3O6) C, H, N.

Article
30,50-Di-O-tert-butyldimethylsilyl-5-carbamoylmethyl-20-deoxy-
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 4 1543
(20 mL) at reflux for 48 h to give, after silica gel chromatography
(4-14% MeOH in CH2Cl2), filtration through PTFE mem-
brane Minisart SRP 15 (0.20 μm, Sartorius) unit, and crystal-
lization from ethanol, compound 1c (0.45 g, 40%). The 1H
NMR data were identical with the ones previously published.40
1H NMR (DMSO-d6, 300 MHz) δ ppm: 2.0-2.1 (m, 2H, CH2-20
and 200); 3.5-3.6 (m, 2H, CH2-50 and 500); 3.8-3.9 (m, 1H, CH-
40); 4.1-4.2 (m, 2H, CH2-OH); 4.2-4.3 (m, 1H, CH-30); 4.9-5.0
(m, 2H, OH-50 and 5-CH2-OH); 5.2-5.3 (m, 1H, OH-30); 6.22
uridine (16). To a solution of compound 1b (0.90 g, 3.33 mmol) in
dry DMF (7 mL) were added imidazole (1.90 g, 27.9 mmol) and
tert-butyldimethylsilyl chloride (3.09 g, 20.5 mmol). The reaction
mixture was stirred at room temperature for 60 h. Water (50 mL)
was added, and the aqueous layer was extracted with EtOAc
(3 ꢀ 50 mL). The combined organic layers were washed with brine
(3 ꢀ 50 mL), dried over Na2SO4, and concentrated under vacuum.
The oilyresidue was purified by silica gel chromatography (0-3%
MeOH in CH2Cl2) to give pure compound 16 (1.55 g, 91%) as a
white foam. 1H NMR (CDCl3, 300 MHz) δ ppm: 0.08 and 0.09
(2s, 12H, CH3-Si); 0.89 and 0.90 (2s, 18H, tBu-Si); 2.1-2.2(m,2H,
CH2-20 and 200);3.02(s, 2H, C5-CH2);3.7-3.8 (m, 2H, CH2-50 and
500); 3.8-3.9 (m, 1H, CH-40); 4.3-4.4 (m, 1H, CH-30); 6.16 (dd,
3
3
(dd, 1H, CH-10, JH1 -H2 = 6.4 Hz, JH1 -H2 = 7.0 Hz); 7.73
(s, 1H, CH-6); 11.31 (s, 1H, NH-3). 13C NMR (DMSO-d6,
75 MHz) δ ppm: 45.96 (C-20); 61.82 (C5-CH2); 61.90 (C-50);
70.99 (C-30); 84.35 (C-10); 87.74 (C-40); 114.71 (C-5); 137.21
(C-6); 150.80 (C-2); 163.08 (C-4). MS (ESI, pos) m/z: 281.54
(M þ Naþ). UV (H2O, 30 μM): λmax = 264 nm, ε = 8600
mol-1 L cm-1. Anal. (C10H14N2O6) C, H, N.
0
0
0
00
3
3
1H, CH-10, JH1 -H2 = 6.3 Hz, JH1 -H2 = 7.5 Hz); 6.85 and 7.28
(2s, 2H, NH2); 7.50 (s, 1H, CH-6); 11.37 (s, 1H, NH-3). 13C NMR
(CDCl3, 75 MHz) δ ppm: -5.04 and -4.75 (4C, CH3-Si); 18.05
and 18.35 (2C, (CH3)3C-Si); 25.73 and 25.88 (6C, (CH3)3C-Si);
33.08 (C5-CH2); 42.18 (C-20); 63.07 (C-50); 72.84 (C-30); 86.94 (C-
10); 88.94 (C-40); 109.07 (C-5); 139.42 (C-6); 151.13 (C-2); 164.56
(C-4); 172.48 (CO-amide). MS (ESI, pos) m/z: 513.5 (M þ Hþ).
30,50-Di-O-tert-butyldimethylsilyl-5-carbamoylmethyl-20-deoxy-
cytidine (18). To a solution of compound 16 (1.50 g, 2.92 mmol)
in anhydrous acetonitrile (50 mL) were added DMAP (1.10 g,
9.1 mmol), freshlydistilledtriethylamine (1.22mL, 8.8mmol), and
triisopropylbenzenesulfonyl chloride (2.75 g, 9.1 mmol). The
solution was stirred at room temperature for 21 h before a 25%
solution of ammonium hydroxide in water (70 mL) was added.
After the mixture was stirred for an additional 1.5 h at room
temperature, solvents were removed under vacuum. The crude
material was dissolved in a water/EtOAc (100 mL, 1/1) mixture.
Layers were separated, and the aqueous was extracted with
EtOAc (3 ꢀ 50 mL). The combined organic layers were washed
with a saturated solution of NaHCO3 (3 ꢀ 50 mL) and then with
brine (3 ꢀ 50 mL), dried over Na2SO4, and concentrated to
dryness. A purification by silica gel chromatography afforded
compound 18 (0.91 g, 61%). 1H NMR (DMSO-d6, 300 MHz) δ
ppm: 0.07 and 0.08 (2s, 12H, CH3-Si); 0.88 and 0.89 (2s, 18H, tBu-
Si); 2.0-2.2 (m, 2H, CH2-20 and 200); 3.05 (s, 2H, C5-CH2); 3.6-
3.8 (m, 2H, CH2-50 and 500); 3.8-4.0 (m, 1H, CH-40); 4.4-4.5 (m,
0
0
0
00
3
3
5-Hydroxymethyl-20-deoxycytidine (2c). Compound 2c was
prepared following the strategy used to convert compound 1b
into 2b and described by LaFrancois et al. for the preparation of
labeled nucleoside analogues of compound 2c.40 5-Methylhy-
droxy-20-deoxyuridine 1c (1.80 g, 7 mmol), TBDMSCl (10.5 g,
70 mmol), and imidazole (4.80 g, 70 mmol) in dry DMF (14 mL)
were stirred for 60 h at room temperature. Purification by silica
gel column chromatography (5-15% AcOEt in petroleum
ether) afforded compound 17 as a white foam (2.9 g, 69%). 1H
NMR (DMSO-d6, 300 MHz) δ ppm: 0.1-0.2 (m, 18H, 6 ꢀ CH3-
Si); 0.8-1.0 (m, 27H, 3(CH3)3); 2.0-2.1 (m, 2H, CH2-20 and 200);
3.6-3.7 (m, 2H, CH2-50 and 500); 3.7-3.8 (m, 1H, CH-40);
4.3-4.4 (m, 3H, CH2-O and CH-30); 6.17 (dd, 1H, CH-10,
3
3
0
0
0
00
JH1 -H2 = 6.4 Hz, JH1 -H2 = 7.0 Hz); 7.43 (s, 1H, CH-6);
11.34 (s, 1H, NH-3).
Compound 17 (2.90 g, 4.83 mmol), DMAP (1.75 g, 14.5
mmol), Et3N (2 mL, 14.5 mmol), and triisopropylbenzenesulfo-
nyl chloride (4.60 g, 15.2 mmol) in anhydrous acetonitrile
(83 mL) were reacted for 21 h at room temperature. A 25%
ammonia solution (115 mL) was added, and the reaction was
pursued for 4 h at room temperature. Extractions with EtOAc
and purification by silica gel column chromatography (0-2.5%
1
MeOH in CH2Cl2) afforded compound 19 (1.49 g, 51%). H
NMR (DMSO-d6, 300 MHz) δ ppm: 0.0-0.1 (m, 18H, 6 ꢀ
CH3-Si); 0.8-0.9 (m, 27H, 3(CH3)3); 1.9-2.2 (2 m, 1H, CH2-20
and 200); 3.6-3.8 (m, 3H, CH-40, CH2-50 and 500); 4.3-4.4 (m,
1H, CH-30); 4.4-4.5 (m, 1H, CH2-O); 6.18 (dd, 1H, CH-10,
3
3
1H, CH-30); 6.18 (dd, 1H, CH-10, JH1 -H2 = 6.2 Hz, JH1 -H2
=
0
0
0
00
7.4 Hz); 6.97 (br s, 2H, NH2-4); 7.02 and 7.36 (2s, 2H, NH2-
amide); 7.41 (s, 1H, CH-6).
3
3
5-Carbamoylmethyl-20-deoxycytidine (2b). To a solution of
compound 18 (0.870 g, 1.70 mmol) in THF (20 mL) was added
TBAF 3H2O (2.68 g, 8.5 mmol). The solution was stirred at
JH1 -H2 = 5.9 Hz, JH1 -H2 = 7.9 Hz); 6.5-6.6 and 7.4-7.5
0
0
0
00
(2 m, 2H, NH2); 7.47 (s, 1H, CH-6).
Treatment of 19 (1.48 g, 2.47 mmol) with TBAF 3H2O
3
3
room temperature for 2 h. Solvent was removed under reduced
pressure, and the crude material was purified by silica gel
column chromatography (0-20% MeOH in CH2Cl2). Pure
fractions were pooled, concentrated, and filtrated through
PTFE membrane Minisart SRP 15 (0.20 μm, Sartorius) unit.
Compound 2b (0.40 g, 83% yield) was obtained by crystal-
lization from absolute ethanol. 1H NMR (DMSO-d6, 300 MHz)
δ ppm: 1.9-2.1 (2 m, 2H, CH2-20 and 200); 3.08 (s, 2H, C5-CH2);
(7.78 g, 24.7 mmol) in THF (18 mL) followed by a purification
by silica gel column chromatography (10-30% MeOH in
CH2Cl2), a filtration through PTFE membrane Minisart SRP
15 (0.20 μm, Sartorius) unit, and a crystallization from ethanol
afforded compound 2c (0.34 g, 54%). 1H NMR (DMSO-d6,
300 MHz) δ ppm: 1.9-2.1 (2 m, 2H, CH2-20 and 200); 3.5-3.6 (m,
2H, CH2-50 and 500); 3.7-3.8 (m, 1H, CH-40); 4.1-4.2 (m, 3H,
CH2-OH and CH-30); 4.9-5.0 (m, 2H, OH-50 and 5-CH2-OH);
3
3.5-3.6 (m, 2H, CH2-50 and 500); 3.7-3.8 (m, 1H, CH-40);
5.20 (d, 1H, OH-30, JH3 -OH = 4.1 Hz); 6.16 (dd, 1H, CH-10,
0
3
3
3
4.1-4.3 (m, 1H, CH-30); 4.95 (t, 1H, OH-50, JOH5 -H5
=
JH1 -H2 = 6.6 Hz, JH1 -H2 = 6.9 Hz); 6.6-6.7 and 7.3-7.4
0
0
0
0
0
00
3
3
JOH5 -H5 = 5.5 Hz); 5.20 (d, 1H, OH-30, JOH3 -H3 = 4.2
(2 m, 2H, NH2); 7.75 (s, 1H, CH-6). 13C NMR (DMSO-d6,
75 MHz) δ ppm: 40.74 (C-20); 57.93 (C5-CH2); 61.85 (C-50);
70.90 (C-30); 85.15 (C-10); 87.58 (C-40); 106.38 (C-5); 139.38
(C-6); 155.42 (C-2); 164.84 (C-4). MS (ESI, pos) m/z: 280.21
(M þ Naþ). UV (H2O, 30 μM): λmax = 274 nm, ε = 8700
mol-1 L cm-1. Anal. (C10H15N3O5) C, H, N. Data are in
0
00
0
0
3
3
Hz); 6.15 (dd, 1H, CH-10, JH1 -H2 = 6.5 Hz, JH1 -H2 = 7.0
Hz); 6.6-6.9 (m, 1H, NH2-4); 7.04 (s, 1H, NH2-amide); 7.1-7.3
(m, 1H, NH2-4); 7.39 (s, 1H, NH2-amide); 7.61 (s, 1H, CH-6).
13C NMR (DMSO-d6, 75 MHz) δ ppm: 30.43 (C5-CH2); 39.35
(C-20); 60.96 (C-50); 70.18 (C-30); 86.19 (C-10); 86.55 (C-40);
101.79 (C-5); 139.58 (C-6); 157.02 (C-2); 166.28 (C-4); 174.88
(CO-amide). MS (ESI, pos) m/z: 307.43 (M þ Naþ). UV (H2O,
30 μM): λmax = 277 nm, ε = 7800 mol-1 L cm-1. Anal. (C11-
0
0
0
00
3
3
accordance with the ones published for the same compound
isolated after oxidation of 5-methyl-20-deoxycytidine by mena-
dione.48
3
3
H16N4O5) C, H, N.
30,50-Di-O-tert-butyldimethylsilyl-20-deoxyuridine (20). As de-
scribed for 4, 20-deoxyuridine (0.950 g, 4.17 mmol) was treated
with TBDMSCl (3.78 g, 25 mmol) and imidazole (2.23 g, 33.3
mmol) in DMF (8 mL) at room temperature for 60 h to give,
after purification by silica gel chromatography, compound 20
5-Hydroxymethyl-20-deoxyuridine (1c). Compound 1c was
obtained from commercially available 20-deoxyuridine.40 20-
Deoxyuridine (1.0 g, 4.39 mmol) in water (10 mL) was reacted
with triethylamine (8 mL) and 37% aqueous formaldehyde

1544 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 4
El Safadi et al.
(1.7 g, 90% yield). 1H NMR (DMSO-d6, 300 MHz) δ ppm:
Cell Toxicity. The cytotoxicity of the drugs was evaluated in
parallel by incubating uninfected cells in the presence of ranging
concentrations of antiviral products. Cell viability was mea-
sured by means of a colorimetric reaction based on the capacity
of mitochondrial dehydrogenase of living cells to reduce 3-(4,5-
dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT)
into formazan, the quantity of which was determined by the
optical density at 540 nm.51 The 50% cytotoxic concentration
(CC50) was calculated with the program used for the determina-
tion of IC50.
Serial Passage Experiments. Stock preparations of HIV-1
LAI was added at a multiplicity of 0.01-100 μL aliquots of
RPMI medium containing 2.5 ꢀ 105 CEM-SS cells. After 4 h at
37 °C, the infected cells were washed twice with PBS (without
MgCl2) and resuspended in 1 mL of medium with or without the
different deoxynucleoside analogues (200 μM, except com-
pound 1c, 5CH2OH-20dU, which was used at 100 μM) in a
48-well plate. After 4 days, the indicated amount of supernatant
was transferred to fresh CEM-SS cells. This procedure was
iterated for eight cycles; aliquots of both cells and supernatants
were frozen at each passage. Virus production was monitored by
measuring RT activity in culture supernatants as described
above. All experiments were carried out in duplicate.
t
0.0-0.1 (m, 12H, CH3-Si); 0.89 and 0.90 (2s, 18H, Bu-Si);
2.1-2.3 (m, 2H, CH2-20 and 200); 3.6-3.9 (m, 3H, CH-40,
CH2-50 and 500); 4.3-4.4 (m, 1H, CH-30); 5.54 (d, 1H, CH-5,
3
3
3
JH5-H6 = 8.0 Hz); 6.13 (t, 1H, CH-10, JH1 -H2 = JH1 -H2
=
0
0
0
00
3
6.5 Hz); 7.71 (d, 1H, CH-6, JH5-H6 = 8.0 Hz); 11.34 (s, 1H,
NH-3). 13C NMR (DMSO-d6, 75 MHz) δ ppm: -5.05
and -4.36 (4C, CH3-Si); 18.06 (2C, C(CH3)3); 26.10 and
26.27 (6C, C(CH3)3); 40.65 (C-20); 62.86 (C-50); 70.84 (C-30);
73.58 (C-10); 85.48 (C-40); 103.28 (C-5); 137.92 (C-6); 150.64
(C-2); 163.36 (C-4). MS (ESI, pos) m/z: 457.85 (M þ Hþ).
30,50-Di-O-tert-butyldimethylsilyl-3-N-carbamoylmethyl-20-deo-
xyuridine (21). To a solution of compound 20 (1.60 g, 3.50 mmol)
in freshly distilled THF (10 mL) was slowly added a suspension of
sodium hydride in freshly distilled THF (0.280 g, 7 mmol, 10 mL).
The mixture was stirred for 30 min at room temperature, then
cooled to 0 °C before adding the chloroacetamide in suspension in
freshly distilled THF (3.70 g, 40 mmol, 30 mL). The reaction
mixture was allowed to warm to room temperature and stirred for
an additional 3 h before the reaction was stopped by the addition
of water (10 mL). The aqueous layer was extracted with EtOAc
(4 ꢀ 50 mL), and the combined organic layer was washed with
brine (3 ꢀ 50 mL), dried with Na2SO4, and concentrated to
dryness. Purification by silica gel column chromatography
(0-5% MeOH in CH2Cl2) afforded compound 21 (1.29 g, 72%
yield). 1H NMR (DMSO-d6, 300 MHz) δ ppm: 0.85 and 0.10 (2s,
Viral Replication Kinetics. The kinetics of viral replication was
determined with CEM-SS cells. CEM-SS cells (2.5 ꢀ 105) were
infected with a normalized low dose (equivalent to 25 cpm/mL
RT activity) of HIV-1 viruses issued from passages 1, 3, and 7 for
1 h at 37 °C. Cells were washed out twice with PBS and incubated
in 1 mL of RPMI at 37 °C in a 48-well plate in the absence of
nucleoside analogues. Virus growth was measured over 17 days
by RT activity in culture supernatant as described above. The
culture medium was changed at days 2, 4, 6, 8, 10, 13, and 15.
t
12H, CH3-Si); 0.89 and 0.90 (2s, 18H, Bu-Si); 2.1-2.3 (m, 2H,
CH2-20 and 200); 3.7-3.9 (m, 3H, CH-40, CH2-50 and 500); 4.33 (s,
2H, N3-CH2); 4.3-4.4 (m, 1H, CH-30); 5.76 (d, 1H, CH-5,
3
3
JH5-H6 = 8.1 Hz); 6.16 (dd, 1H, CH-10, JH1 -H2 = 6.4 Hz,
0
0
3
0
00
JH1 -H2 = 6.5 Hz); 7.11 and 7.55 (2s, 2H, NH2); 7.79 (d, 1H, CH-
6, JH5-H6 = 8.1 Hz). 13C NMR (DMSO-d6, 75 MHz) δ ppm:
-4.14 and -4.60 (4C, CH3-Si); 18.01 and 18.46 (2C, C(CH3)3);
26.00 and 26.23 (6C, C(CH3)3); 40.34 (C-20); 42,40 (N3-CH2);
62.76 (C-50); 71.67 (C-30); 83.57 (C-10); 87.20 (C-40); 101.36 (C-5);
139.26 (C-6); 150.91 (C-2); 162.10 (C-4); 168,72 (CO amide). MS
(ESI, pos) m/z: 514.81 (M þ Hþ).
3
Acknowledgment. The CEM-SS cells were obtained from
P. Nara through the AIDS Research and Reference Reagent
Program, Division of AIDS, NIAID, NIH. We thank ANRS,
SIDACTION, and Strasbourg University for financial sup-
port.
3-N-Carbamoylmethyl-20-deoxyuridine (3). Elimination of si-
lyl protecting groups from 21 (1.26 g, 2.45 mmol) by the
procedure previously described for compound 2b afforded the
target compound 3 (0.450 g, 65% yield). 1H NMR (DMSO-d6,
300 MHz) δ ppm: 2.0-2.2 (m, 2H, CH2-20 and 200); 3.5-3.6 (m,
2H, CH2-50 and 500); 3.7-3.9 (m, 1H, CH-40); 4.2-4.3 (m, 1H,
Supporting Information Available: Elemental analysis data
for target compounds. This material is available free of charge
via the Internet at http://pubs.acs.org.
3
CH-30); 4.33 (s, 2H, N3-CH2); 5.09 (t, 1H, OH-50, JH5 -OH
=
0
3
3
JH5 -OH = 5.0 Hz); 5.32 (d, 1H, OH-30, JH3 -OH = 3.9 Hz);
References
00
0
3
5.78 (d, 1H, CH-5, JH5-H6 = 8.4 Hz); 6.15 (t, 1H, CH-10,
(1) Barre-Sinoussi, F.; Cherman, J. C.; Rey, F.; Nugeyre, H. T.;
Charmet, S.; Gruest, J.; Dauget, C.; Axler-Blim, C.; Vezinet-Brun,
F.; Rouxioux, C.; Rozenbaum, W.; Montagnier, L. Isolation of a
T-lymphotropic retrovirus from a patient at risk for acquired
immune deficiency syndrome (AIDS). Science 1983, 220, 868–871.
(2) Gallo, R. C.; Sarin, P. S.; Gelmann, E. P.; Robert-Guroff, M.;
Richardson, E.; Kalyanaraman, V. S.; Mann, D.; Sidhu, G. D.;
Stahl, R. E.; Zolla-Pazner, S.; Leibowitch, J.; Popovic, M. Isolation
of human T-cell leukemia virus in acquired immune deficiency
syndrome (AIDS). Science 1983, 220, 865–867.
3
3
0
0
0
00
JH1 -H2 = JH1 -H2 = 6.7 Hz); 7.09 and 7.57 (2s, 2H, NH2);
7.93 (d, 1H, CH-6, 3JH5-H6 = 8.4 Hz). 13C NMR (DMSO-d6, 75
MHz) δ ppm: 40.31 (C-20); 42,86 (N3-CH2); 61.64 (C-50); 70.73
(C-30); 85.63 (C-10); 87.96 (C-40); 101.29 (C-5); 139.66 (C-6);
150.95 (C-2); 162.21 (C-4); 168,81 (CO amide). MS (ESI, pos)
m/z: 308.36 (M þ Naþ). UV (H2O, 30 μM): λmax = 263 nm, ε =
7900 mol-1 L cm-1. Anal. (C11H15N3O6) C, H, N.
3
3
Anti-HIV Activity. The cell cultures of MT-4 and CEM-SS
cells were maintained at 37 °C in a 5% CO2 atmosphere in RPMI
1640 medium supplemented with 10% fetal calf serum (FCS)
containing penicillin, streptomycin, and glutamine. MT-4 cells
were infected with HIV-1 IIIB isolate at 5 TCID50 (tissue culture
infective dose) and CEM-SS cells with HIV-1 LAI isolate at
20 TCID50. These viruses are two closely related isolates differ-
ing by their cytopathogenicity. After 30 min of adsorption,
the infected cells were washed to remove unadsorbed virus
particles and resuspended in RPMI at the final concentration
of 105 cells/mL, in the presence of various dilutions of the test
compounds. After 5 days, virus production was determined by
measuring RT activity in culture medium as already described.49
The 50% inhibitory concentration (IC50) was derived from the
computer-generated median effect plot of the dose-effect
data.50 AZT was used as a positive control (IC50 = 2.4 nM).
The entire experiment was repeated twice.
(3) Wild, C.; Greenwell, T.; Matthews, T. A synthetic peptide from
HIV-1 gp41 is a potent inhibitor of virus-mediated cell-cell fusion.
AIDS Res. Hum. Retroviruses 1993, 9, 1051–1053.
(4) Emmelkamp, J. M.; Rockstroh, J. K. CCR5 antagonists: compar-
ison of efficacy, side effects, pharmacokinetics and interactions;
review of the literature. Eur. J. Med. Res. 2007, 12, 409–417.
(5) El Safadi, Y.; Vivet-Boudou, V.; Marquet, R. HIV-1 reverse
transcriptase inhibitors. Appl. Microbiol. Biotechnol. 2007, 75,
723–737.
(6) Markowitz, M.; Morales-Ramirez, J. O.; Nguyen, B. Y.; Kovacs,
C. M.; Steigbigel, R. T.; Cooper, D. A.; Liporace, R.; Schwartz, R.;
Isaacs, R.; Gilde, L. R.; Wenning, L.; Zhao, J.; Teppler, H.
Antiretroviral activity, pharmacokinetics, and tolerability of
MK-0518, a novel inhibitor of HIV-1 integrase, dosed as mono-
therapy for 10 days in treatment-naive HIV-1-infected individuals.
J. Acquir. Immune Defic. Syndr. 2006, 43, 509–515.
(7) Bhat, M. A.; Siddiqui, N.; Khan, S. A.; Ahmad, M. Non-peptide
protease inhibitors as novel antiHIV agents. Indian J. Pharm. Sci.
2006, 68, 549–555.

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 4 1545
(8) Larder, B. A. Mechanisms of HIV-1 drug resistance. AIDS 2001,
15, S27–S34.
(9) Vivet-Boudou, V.; Didierjean, J.; Isel, C.; Marquet, R. Nucleoside
and nucleotide inhibitors of HIV-1 replication. Cell. Mol. Life Sci.
2006, 63, 163–86.
(32) Day, C. W.; Smee, D. F.; Julander, J. G.; Yamshchikov, V. F.;
Sidwell, R. W.; Morrey, J. D. Error-prone replication of West Nile
virus caused by ribavirin. Antiviral Res. 2005, 67, 38–45.
(33) Suen, W.; Spiro, T. G.; Sowers, L. C.; Fresco, J. R. Identification
by UV resonance Raman spectroscopy of an imino tautomer of
5-hydroxy-20-deoxycytidine, a powerful base analog transition
mutagen with a much higher unfavored tautomer frequency than
that of the natural residue 20-deoxycytidine. Proc. Natl. Acad. Sci.
U.S.A. 1999, 96, 4500–4505.
(34) Morningstar, M. L.; Kreutzer, D. A.; Essigmann, J. Synthesis of
oligonucleotides containing two putatively mutagenic DNA le-
sions: 5-hydroxy-20-deoxyuridine and 5-hydroxy-20-deoxycytidine.
Chem. Res. Toxicol. 1997, 10, 1345–1350.
(10) Smith, R. A.; Loeb, L. A.; Preston, B. D. Lethal mutagenesis of
HIV. Virus Res. 2005, 107, 215–228.
(11) Murphy, E. L.; Collier, A. C.; Kalish, L. A.; Assmann, S. F.; Para,
M. F.; Flanigan, T. P.; Kumar, P. N.; Mintz, L.; Wallach, F. R.;
Nemo, G. J. Highly active antiretroviral therapy decreases mor-
tality and morbidity in patients with advanced HIV disease. Ann.
Intern. Med. 2001, 135, 17–26.
(12) Loeb, L.; Essigmann, J.; Kazazi, F.; Zhang, J.; Rose, K.; Mullins, J.
Lethal mutagenesis of HIV with mutagenic nucleoside analogs.
Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 1492–1497.
(13) Eigen, M.; Schuster, P. The hypercycle. A principle of natural self-
organization. Part A: Emergence of the hypercycle. Naturwis-
senschaften 1977, 64, 541–565.
(14) Holland, J. J.; Domingo, E.; de La Torre, J. C.; Steinhauer, D. A.
Mutation frequencies at defined single codon sites in vesicular
stomatitis virus and poliovirus can be increased only slightly by
chemical mutagenesis. J. Virol. 1990, 64, 3960–3962.
(35) LaFrancois, C. J.; Jang, Y. H.; Cagin, T.; Goddard, W. A.; Sowers,
L. C. Conformation and proton configuration of pyrimidine
deoxynucleoside oxidation damage products in water. Chem.
Res. Toxicol. 2000, 13, 462–470.
(36) Brossmer, R.; Roehm, E. Synthesis of 5-hydroxymethyl-20-deox-
ycytidine. Angew. Chem. 1963, 75, 1105.
(37) Mellac, S.; Fazakerley, G. V.; Sowers, L. C. Structures of base pairs
with 5-(hydroxymethyl)-20-deoxyuridine in DNA determined by
NMR spectroscopy. Biochemistry 1993, 32, 7779–7786.
(38) Purmal, A. A.; Kow, Y. W.; Wallace, S. S. Major oxidative
products of cytosine, 5-hydroxycytosine and 5-hydroxyuracil,
exhibit sequence context-dependent mispairing in vitro. Nucleic
Acids Res. 1994, 22, 72–78.
(39) Chaung, W.; Boorstein, R. J. Molecular spectrum of mutations
induced by 5-hydroxymethyl-20-deoxyuridine in (CHO)-PL61
cells. Mutat. Res. 1997, 373, 125–137.
(40) LaFrancois, C. J.; Fujimoto, J.; Sowers, L. C. Synthesis and
characterization of isotopically enriched pyrimidine deoxynucleo-
side oxidation damage products. Chem. Res. Toxicol. 1998, 11,
75–83.
(41) Aso, M.; Kaneko, T.; Nakamura, M.; Koga, N.; Suemune, H. A
simple and efficient method for the synthesis of 5-substituted
20-deoxyuridine nucleosides using metal-halogen exchange reac-
tion of 5-iodo-deoxyuridine sodium salt. Chem. Commun. 2003, 9,
1094–1095.
(42) Li, S. N.; Piccirilli, J. A. Synthesis of the phosphoramidite deriva-
tives of 20-deoxy-20-C-R-methylcytidine and 20-deoxy-20-C-R-hy-
droxymethylcytidine: analogues for chemical dissection of RNA’s
20-hydroxyl group. J. Org. Chem. 2004, 69, 4751–4759.
(43) Inoue, H.; Ueda, T. Synthesis of 6-cyano and 5-cyano-uridines and
their derivatives. Chem. Pharm. Bull. 1978, 26, 2657–2663.
(44) Trippett, S.; Walker, D. M. Reaction of Wittig reagents with
phenyl isocyanate. J. Chem. Soc. 1959, 3874–3876.
(15) Pathak, V. K.; Temin, H. N. 5-Azacytidine and RNA secondary
structure increase the retrovirus mutation rate. J. Virol. 1992, 66,
3093–3100.
(16) Pathak, V. K.; Temin, H. N. Broad spectrum of in vivo forward
mutations, hypermutations and mutational hotspots in a retroviral
shuttle vector after a single replication cycle: deletions and deletions
with insertions. Proc. Natl. Acad. Sci. U.S.A. 1990, 87, 6024–6028.
(17) Camps, M.; Naukkarinen, J.; Johnson, B. P.; Loeb, L. A. Targeted
gene evolution in Escherichia coli using a highly error-prone DNA
polymerase I. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 9727–9732.
(18) Crotty, S.; Cameron, C. E.; Andino, R. RNA virus error cata-
strophe: direct molecular test by using ribavirin. Proc. Natl. Acad.
Sci. U.S.A. 2001, 98, 6895–6900.
(19) Sierra, S.; Davila, M.; Lowenstein, P. R.; Domingo, E. Response of
foot-and-mouth disease virus to increased mutagenesis: influence of
viral load and fitness in loss infectivity. J. Virol. 2000, 74, 8316–8323.
(20) Zhou, S.; Liu, R.; Baroudy, B. M.; Malcolm, B. A.; Reyes, G. R.
The effect of ribavirin and IMPDH inhibitors on hepatitis C virus
subgenomic replicon RNA. Virology 2003, 310, 333–342.
(21) Daifuku, R. Stealth nucleosides: mode of action and potential use
in the treatment of viral diseases. BioDrugs 2003, 17, 169–177.
(22) Anderson, J. P.; Daifuku, R.; Loeb, L. A. Viral error catastrophe
by mutagenic nucleosides. Annu. Rev. Microbiol. 2004, 58, 183–205.
(23) Harris, K.; Brabant, W.; Styrchak, S.; Gall, A.; Daifuku, R. KP-
1212/1461, a nucleoside designed for the treatment of HIV by viral
mutagenesis. Antiviral Res. 2005, 67, 1–9.
(24) Murakami, E.; Basavapathruni, A.; Bradley, W.; Anderson, C.
Mechanism of action of a novel viral mutagenic covert nucleotide:
molecular interactions with HIV-1 reverse transcriptase and host
cell DNA polymerases. Antiviral Res. 2005, 67, 10–17.
(25) El Safadi, Y.; Marquet, R.; Aubertin, A.-M.; Vivet-Boudou, V.
Synthesis and primary evaluation of novel HIV-1 inhibitors.
Nucleosides, Nucleotides Nucleic Acids 2007, 26, 1161–1165.
(26) Vivet-Boudou, V.; Paillart, J. C.; Burger, A.; Marquet, R. In search
of new inhibitors of HIV-1 replication: synthesis and study of 1-(2-
deoxy-β-D-ribofuranosyl)-1,2,4-triazole-3-carboxamide as a selec-
tive viral mutagenic agent. Nucleosides, Nucleotides Nucleic Acids
2007, 26, 743–746.
(45) Hirota, K.; Suematsu, M.; Kuwabara, Y.; Asao, T.; Senda, S.
Novel C-C bond formation at the 5-position of uracils. Facile
synthesis of 5-methoxycarbonylmethyluridine and 5-carbamoyl-
methyl-uridine, minor component nucleosides derived form
transfer ribonuclease. J. Chem. Soc., Chem. Comun. 1981, 623–
625.
1
(46) Poznanski, J.; Felczak, K.; Kulikowski, T.; Remin, M. H NMR
conformational study of antiherpetic C5-substituted 20-deoxyur-
idines: insight into the nature of structure-activity relationships.
Biochem. Biophys. Res. Commun. 2000, 272, 64–74.
(47) Lahoud, G.; Timoshchuk, V.; Lebedev, A.; de Vega, M.; Salas, M.;
Arar, K.; Hou, Y.-M.; Gamper, H. Enzymatic synthesis of struc-
ture-free DNA with pseudo-complementary properties. Nucleic
Acids Res. 2008, 36, 3409–3419.
(27) Domingo, E. Viruses at the edge of adaptation. Virology 2000, 270,
251–253.
(28) Crotty, S.; Maag, D.; Arnold, J. J.; Zhong, W.; Lau, J. Y. N.; Hong,
Z.; Andino, R.; Cameron, C. E. The broad-spectrum antiviral
ribonucleoside ribavirin is an RNA virus mutagen. Nat. Med. 2000,
6, 1375–1379.
(29) Contreras, A.; Hiasa, Y.; He, W.; Terella, A.; Schmidt, E.; Chung,
R. Viral RNA mutations are region specific and increased by
ribavirin in a full-length hepatitis C virus replication system.
J. Virol. 2002, 76, 8505–8517.
(30) Lanford, R. E.; Chavez, D.; Guerra, B.; Lau, J. Y. N.; Hong, Z.;
Brasky, K. M.; Beames, B. Ribavirin induces error-prone replica-
tion of GB virus B in primary Tamarian hepatocytes. J. Virol. 2001,
75, 8074–8081.
(48) Bienvenu, C.; Wagner, J. R.; Cadet, J. Photosensitized oxidation of
5-methyl-20-deoxycytidine by 2-methyl-1,4-naphthoquinone: char-
acterization of 5-(hydroperoxymethyl)-20-deoxycytidine and stable
methyl group oxidation products. J. Am. Chem. Soc. 1996, 118,
11406–11411.
(49) Moog, C.; Wick, A.; Le Ber, P.; Kirn, A.; Aubertin, A.-M. Bicyclic
imidazo derivatives, a new class of highly selective inhibitors for the
human immunodeficiency virus type 1. Antiviral Res. 1994, 24, 275–
288.
(50) Chou, J.; Chou, T. Dose Effect Analysis with Microcomputers:
Quantitation of ED50, LD50, Synergism, Antagomism, Low-Dose
Risk, Receptor Binding and Enzyme Kinetics. In Computer Soft-
ware for Apple II Series and IBM-PC and Instruction Manual;
Elsevier-Biosoft: Cambridge, U.K., 1985; pp 19-25.
(51) Mossmann, T. Rapid colorimetric assay for cellular growth and
survival: application to proliferation and cytotoxicity assays.
J. Immunol. Methods 1983, 65, 55–63.
(31) Severson, W. E.; Schmaljohn, C. S.; Javadian, A.; Jonsson, C. B.
Ribavirin causes error catastrophe during hantaan virus replica-
tion. J. Virol. 2003, 77, 481–488.