Design, synthesis, and anti-HIV activity of 2′,3′-didehydro-2′,3′-dideoxyuridine (d4U), 2′,3′-dideoxyuridine (ddU) phosphoramidate 'ProTide' derivatives

Article abstract of DOI:10.1016/j.bmcl.2007.04.043

We report the synthesis of 2′,3′-didehydro-2′,3′-dideoxyuridine (d4U) and 2′,3′-dideoxyuridine (ddU) phosphoramidate 'ProTide' derivatives and their evaluation against HIV-1 and HIV-2. In addition, we conducted molecular modeling studies on both d4U and ddU monophosphates to investigate their second phosphorylation process. The findings from the modeling studies provide compelling evidence for the lack of anti-HIV activity of d4U phosphoramidates, in contrast with the corresponding ddU phosphoramidates.

Full text of DOI:10.1016/j.bmcl.2007.04.043

Bioorganic & Medicinal Chemistry Letters 17 (2007) 3666–3669
0
0
0
0
Design, synthesis, and anti-HIV activity of 2 ,3 -didehydro-2 ,3 -
0
0
dideoxyuridine (d4U), 2 ,3 -dideoxyuridine (ddU)
phosphoramidate ‘ProTide’ derivatives
a
a,
a
b
*
Youcef Mehellou, Christopher McGuigan, Andrea Brancale and Jan Balzarini
a
Welsh School of Pharmacy, Cardiff University, King Edward VII Avenue, Cardiff CF10 3XF, UK
b
Rega Institute for Medical Research, Katholieke Universiteit Leuven, B-3000 Leuven, Belgium
Received 30 March 2007; revised 13 April 2007; accepted 15 April 2007
Available online 22 April 2007
0
0
0
0
0
0
Abstract—We report the synthesis of 2 ,3 -didehydro-2 ,3 -dideoxyuridine (d4U) and 2 ,3 -dideoxyuridine (ddU) phosphoramidate
ProTide’ derivatives and their evaluation against HIV-1 and HIV-2. In addition, we conducted molecular modeling studies on both
‘
d4U and ddU monophosphates to investigate their second phosphorylation process. The findings from the modeling studies provide
compelling evidence for the lack of anti-HIV activity of d4U phosphoramidates, in contrast with the corresponding ddU
phosphoramidates.
Ó 2007 Elsevier Ltd. All rights reserved.
0
Nucleoside analogues lacking the 3 -hydroxyl group, for
0
O
N
O
N
O
N
0
0
0
example, 2 ,3 -dideoxy 2 ,3 -didehydrothymidine (d4T)
0
NH
NH
NH
0
and 2 ,3 -dideoxycytidine (ddC), have proved to be effec-
tive in treating the Acquired Immunodeficiency Syn-
HO
O
HO
O
HO
O
O
O
O
1
drome (AIDS)
which is caused by Human
2
Immunodeficiency Virus (HIV). These agents produce
their effects as HIV reverse transcriptase (HIV-RT)
inhibitors and/or DNA chain terminators after being
d4T
d4U
ddU
Figure 1. Structures of d4T, d4U, and ddU.
0
3
converted to their corresponding 5 -triphosphates.
However, the use of these drugs has been limited due
to the emergence of resistance and inherent toxicity as
well as dependence on kinase mediated nucleoside acti-
vation to generate the bioactive triphosphates. Bearing
in mind the success of d4T as an anti-HIV agent and
with the aim of finding alternative anti-HIV drugs with
high potency and acceptable toxicity profiles, we decided
5
for ddU. Thus, we decided to explore these possibilities
further by applying the ‘ProTide’ approach to both d4U
and ddU in order to bypass the first phosphorylation
step and to estimate the effects that phosphoramidate
prodrugs have on the antiviral activity. In addition, we
carried out some molecular modeling investigations to
study the possible second phosphorylation step that is
0
0
to investigate the anti-HIV activity of 2 ,3 -didehydro-
0
0
0
0
0
required for the activation of these agents at their 5 -
monophosphate level.
2
(
,3 -dideoxyuridine (d4U) and 2 ,3 -dideoxyuridine
ddU) (Fig. 1).
In the ‘ProTide’ approach, the phosphate group is
masked to improve the poor membrane permeability
seen when the free nucleotides are used, coupled with
the inherent lability of free monophosphates to dephos-
phorylation. Upon entering the cell, the group masking
the phosphate moiety may undergo enzymatic conver-
sion to release the nucleoside analogue monophosphate,
which may be subsequently phosphorylated to the di-
Both (d4U) and (ddU) are known to have poor anti-
4
HIV activity. For d4U, we hypothesized that this poor
activity is due to its inefficient phosphorylation to its ac-
tive triphosphate form as this was proved to be the case
Keywords: HIV; Phosphoramidates; ddU; ProTide.
*
Corresponding author. Tel./fax: +44 2920 874537; e-mail:
mcguigan@cardiff.ac.uk
6
and triphosphates of the nucleoside analogue.
0
960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.04.043

Y. Mehellou et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3666–3669
3667
d4U was prepared via the same procedure adopted by
Table 1. The structures of potential anti-HIV ProTides
7
0
0
0
McGuigan et al. Thus, 2 -deoxyuridine was first 3 ,5 -
dimesylated and then reacted with aqueous sodium
hydroxide to afford 2 in a 52% yield. Treatment of 2
O
O
N
NH
NH
O
O
P
NH
0
0
with sodium hydride and DMF gave the 2 ,3 -didehy-
0
R1
O
P
NH
O
N
O
R1
O
O
O
O
O
0
dro-2 ,3 -dideoxyuridine (d4U) in a moderate yield,
3%. Hydrogenation of d4U using 10% palladium on
carbon yielded ddU in a 69% yield (Scheme 1).
4
HC CH3
HC CH3
O
O
R2
O
R2 O
1a-d
2a-d
In our previous work on nucleoside analogue phospho-
ramidates, particularly d4T, we found that phenyl-L-ala-
nine ester phosphoramidates were efficacious in
enhancing the anti-HIV activity of d4T. More recently
9
Congiatu et al. reported the superiority of naphthyl-L-
alanine esters in improving the anticancer activity profile
of BVDU compared to the phenyl analogues. Thus, we
decided to make a small family of phenyl and naphthyl-
alanine ester phosphoramidates of both d4U and ddU,
and then examine their anti-HIV activity. The selection
of d4U and ddU phosphoramidates synthesized for this
study is summarized in Table 1.
Compound
R
1
R
2
1
1
1
1
a
b
c
Phenyl
Phenyl
Phenyl
Me
Et
8
t-Bu
Bn
d
1-Naphthyl
Phenyl
Phenyl
2a
2b
2c
Me
Et
Phenyl
1-Naphthyl
t-Bu
Bn
2
d
phosphate and/or third, the triphosphate of d4U being
inactive as an inhibitor of HIV RT.
The ‘ProTide’ phosphoramidates were synthesized
1
according to the previously reported synthetic routes.
0
Aryl phosphochloridates were prepared by the reaction
of phenyl/naphthyl dichlorophosphates with the appro-
priate amino acid ester hydrochloride. The resulting
phosphorochloridates were allowed to react with the
nucleoside analogue in THF and t-BuMgCl to give
the target phosphoramidates in moderate yields
By contrast, although the nucleoside analogue ddU was
similarly found not to possess any anti-HIV activity, its
phosphoramidates exerted moderate activities (Table 3).
This indicated that using phosphoramidates to bypass
the first phosphorylation step of ddU resulted in a boost
of anti-HIV activity. This highlights the success of phos-
phoramidates as a tool for the intracellular delivery of
nucleoside analogue monophosphates. Therefore, we
concluded that the first phosphorylation step is the rea-
son for the poor anti-HIV activity of ddU. As for d4U,
none of the tested ddU phosphoramidates exerted sig-
nificant toxicity (Table 3).
(
Scheme 2).
The synthesized nucleosides and ProTides were tested
against both HIV-1 and HIV-2 with data shown in
Tables 2 and 3, and d4T (Stavudine) included as a posi-
tive control. The biological data for d4U derivatives re-
vealed that neither of the parent nucleosides nor any of
the phosphoramidate derivatives was found to be signif-
icantly active against HIV-1 or HIV-2 (Table 2),
although the naphthyl benzyl compound (1d) showed
slight activity. This highlighted the inability of the par-
ent nucleoside to inhibit HIV which is in agreement with
It is also of note that a significant SAR emerged for the
ddU ProTides. The t-butyl ester 2c was inactive, while
the methyl and ethyl esters showed some activity. The
9
naphthyl benzyl ester 2d was most active of all.
4
the observation made by Balzarini et al. In addition,
As a result of the differences in activities between d4U
and ddU as well as the failure of d4U phosphoramidates
to improve the anti-HIV activity unlike that for ddU, we
decided to conduct molecular modeling studies investi-
gating the second phosphorylation step. This was done
the biological data indicated the failure in general of
the tested phosphoramidate derivatives to improve the
anti-HIV activity of d4U. All the tested compounds,
including d4U, were shown to exert some toxicity. This
poor anti-HIV activity of d4U and its phosphorami-
dates could be attributed to three reasons. First, the
inability of the d4U ProTides to efficiently deliver d4U
monophosphate into cells. Second, the failure of d4U-
monophosphate to be further phosphorylated to its tri-
1
1
by docking both d4U and ddU monophosphates into
the active site of thymidylate kinase, which was shown
to be the enzyme responsible for the phosphorylation
of numerous clinically proven antiviral nucleosides. In
this study, we used thymidine monophosphate as refer-
O
O
O
O
NH
NH
NH
NH
HO
N
O
i, ii
O
HO
N
O
N
O
HO
N
O
O
iii
iv
O
O
OH
O
2
d4U
ddU
Scheme 1. Synthetic route to d4U and ddU. Reagents and conditions: (i) MeSO
iv) H , 10% Pd/C, overnight.
2
Cl, Pyr, rt; (ii) aq. NaOH, MeOH, reflux; (iii) NaH, DMF, 100 °C;
(
2

3
668
Y. Mehellou et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3666–3669
Table 3. Anti-HIV data of ddU and its ProTides
ence when analyzing the docking results of both d4U
and ddU monophosphates. The resulting structures for
the two nucleotide analogues were ranked according to
the relative position of the uracil base with the corre-
sponding position of the thymine base in the crystal
structure of TMP, and only the results with a RMSD
Compound
EC50 (lM)
CEM/0
CC50 (lM)
CEM/0
HIV-1
HIV-2
2
2
2
2
a
b
c
36.7 ± 11.5
43.3 ± 20.8
>250
P50
93.3 ± 32.1
>250
121 ± 1.4
>250
>250
˚
values below 1.0 A were further analyzed. The best re-
sults, shown in Figure 2, revealed that the phosphorus
group of ddU-monophosphate was placed in a very sim-
ilar position to that of the naturally occurring thymidine
monophosphate. This therefore permits a good second
phosphorylation of ddU monophosphate, and thus al-
lows the activation of inactive ddU on ProTide bypass
of the limiting initial phosphorylation.
d
15.0 ± 0.0
>250
0.65
20.0 ± 7.1
>250
0.77
96.9 ± 0.85
>250
174
ddU
d4T
EC50: The effective concentration (lM) required to protect CEM cells
against the cytopathogenicity of HIV by 50%. CC50: The cytostatic
concentration (lM) required to reduce CEM cell viability by 50%.
However, by contrast the phosphorus group of d4U
monophosphate was placed significantly far from that
of thymidine monophosphate. In fact, it was placed fur-
ther out from the active site of thymidylate kinase and,
more importantly, closer to the ATP molecule. As the
monophosphate group bears oxygen atoms that are
negatively charged like that of ATP, a putative repulsion
could occur between d4U monophosphate and ATP
which would most likely limit the second phosphoryla-
CH3
O
P
O
P
Et3N
O
R1
Cl
R2
NH3+Cl-
R1
Cl
Cl
O
NH
CH CH3
O
R2
O
THF
ddU
tBuMgCl
THF
d4U
t
BuMgCl
O
N
O
NH
NH
O
P
O
P
R1
O
O
O
R1
O
O
N
O
O
O
NH
NH
HC CH3
HC CH3
O
O
R2
O
R2 O
2
a-d
1a-d
Figure 2. Docking results of both ddUMP (bottom) and d4UMP (top)
with TMP in thymidylate kinase.
Scheme 2. General synthetic scheme of d4U/ddU ProTides.
tion step of d4U monophosphate if not prevented it at
all. Thus, even a successful ProTide bypass of the initial
phosphorylation of d4U would not be able to confer
antiviral activity on this nucleoside. Thus, the conclu-
sions of these modeling studies on the monophosphates
of ddU and d4U are entirely consistent with the differen-
tial outcomes of the ProTide approach in each case.
Table 2. Anti-HIV data of d4U and its ‘ProTide’ phosphoramidates
Compound
EC50 (lM)
CEM/0
CC50 (lM)
CEM/0
HIV-1
HIV-2
1
1
1
1
a
b
c
>50
>50
>250
25
>10
>50
>250
25
96.6 ± 13.3
212 ± 53.9
>250
In conclusion, we have synthesized and studied the anti-
HIV activity of d4U, ddU, and some of their phospho-
ramidate ProTides. The results showed that neither d4U
nor ddU possessed significant anti-HIV activity. As for
their phosphoramidates, d4U phosphoramidates did
not exert significant anti-HIV activity, whereas ddU
phosphoramidates showed markedly improved anti-
HIV activity. This indicated that the first phosphorylation
d
100
207 ± 60.8
174
d4U
d4T
>50
0.65
>50
0.77
EC50: The effective concentration (lM) required to protect CEM cells
against the cytopathogenicity of HIV by 50%. CC50: The cytostatic
concentration (lM) required to reduce CEM cell viability by 50%.

Y. Mehellou et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3666–3669
3669
step is the one responsible for the inability of ddU to ex-
ert any significant anti-HIV activity. Regarding d4U, we
have shown that bypassing the first phosphorylation
step does not improve its anti-HIV activity and using
molecular modeling, we have suggested that the ineffi-
cient second phosphorylation step may in part be
responsible for the poor anti-HIV activity of d4U. Over-
all, we are currently synthesizing more ddU phospho-
ramidates with different amino acids and esters to
explore whether we can tune its anti-HIV activity to
sub-micromolar levels.
3. (a) Balzarini, J.; Herdewijn, P.; De Clercq, E. J. Biol.
Chem. 1989, 264, 6127; (b) Gao, W. Y.; Agbaria, R.;
Driscoll, J. S.; Mitsuya, H. J. Biol. Chem. 1994, 269,
1
2639.
4
. Balzarini, J.; Kang, G.; Dalal, M.; Herdewijn, P.; De Clercq,
E.; Broder, S.; Johns, D. G. Mol. Pharmacol. 1987, 32, 162.
. Hao, Z.; Cooney, D. A.; Farquhar, D.; Perno, C. F.;
Zhang, K.; Masood, R.; Wilson, Y.; Hartman, N. R.;
Balzarini, J.; Johns, D. Mol. Pharmacol. 1989, 37, 157.
5
6. Cahard, D.; McGuigan, C.; Balzarini, J. Mini-Rev. Med.
Chem. 2004, 4, 371.
7. McGuigan, C.; Pathirana, R. N.; Snoeck, R.; Andrei, G.;
De Clercq, E.; Balzarini, J. J. Med. Chem. 2003, 47, 1847.
8
. McGuigan, C.; Cahard, D.; Sheeka, H. M.; De Clercq, E.;
Balzarini, J. J. Med. Chem. 1996, 39, 1748.
Acknowledgments
9
. Congiatu, C.; Brancale, A.; Mason, M. D.; Jiang, W. G.;
McGuigan, C. J. Med. Chem. 2006, 49, 452.
The authors are grateful to Mrs. Ann Absillis for excel-
lent technical assistance. We also thank Helen Murphy
for excellent secretarial assistance.
10. (a) McGuigan, C.; Pathirana, R. N.; Balzarini, J.; De
Clercq, E. J. Med. Chem. 1993, 36, 1048; (b) McGuigan,
C.; Cahard, D.; Sheeka, H. M.; De Clercq, E.; Balzarini, J.
J. Med. Chem. 1996, 39, 1748; (c) McGuigan, C.; Tsang,
H. W.; Cahard, D.; Turner, K.; Velazquez, S.; Salgado,
A.; Bidois, L.; Naesens, L.; De Clercq, E.; Balzarini, J.
Antiviral Res. 1997, 35, 195.
References and notes
1
1. Docking studies were performed using the FlexX module
of SYBYL7.2 (Tripos Inc., 1699 South Hanley Rd., St.
Louis, MO 63144, USA. http://www.tripos.com) and the
results were analyzed using MOE 2006.08 (Molecular
Operating Environment (MOE 2006.08). Chemical Com-
puting Group, Inc., Montreal, Que., Canada. http://
www.chemcomp.com).
1
2
. De Clercq, E. J. Med. Chem. 1995, 38, 2491.
. (a) Barre-Sinoussi, F.; Cherman, J. C.; Rey, R.; Nugeyre,
M. T.; Chamaret, S.; Gruest, J.; Dauguet, C.; Axler-Blin,
C.; V e´ zinet-Brun, F.; Rouzioux, C.; Rozenbaum, W.;
Montagnier, L. Science 1983, 220, 868; (b) Border, S.;
Gallo, R. C. New Eng. J. Med. 1984, 311, 1292.