An alternative pathway to ribonucleoside β-hydroxyphosphonate analogues and related prodrugs

Article abstract of DOI:10.1021/ol402143y

Nucleoside β-(S)-hydroxyphosphonate analogues have recently proven to be interesting bioactive compounds as 5′-nucleotidase inhibitors. These derivatives were obtained in a pyrimidine series through an ex-chiral pool pathway or the stereoselective reducti

Full text of DOI:10.1021/ol402143y

ORGANIC
LETTERS
2013
Vol. 15, No. 18
4778–4781
An Alternative Pathway to Ribonucleoside
β‑Hydroxyphosphonate Analogues and
Related Prodrugs
†
Audrey Hospital, Ma¨ıa Meurillon, Suzanne Peyrottes,* and Christian Perigaud
†
ꢀ
ꢀ
Institut des Biomolecules Max Mousseron (IBMM), UMR5247 CNRS-UM1-UM2,
Nucleosides and Phosphorylated Effectors Team, University Montpellier 2, cc 1705,
ꢁ
place Eugene Bataillon, 34095 Montpellier cedex 5, France
suzanne.peyrottes@univ-montp2.fr
Received July 29, 2013
ABSTRACT
Nucleoside β-(S)-hydroxyphosphonate analogues have recently proven to be interesting bioactive compounds as 50-nucleotidase inhibitors.
These derivatives were obtained in a pyrimidine series through an ex-chiral pool pathway or the stereoselective reduction of a β-ketophosphonate
intermediate. Herein, an original synthesis of these compounds using nucleoside epoxide intermediates, containing either a pyrimidine or a
purine as nucleobase, was explored and allowed the direct synthesis of the corresponding bis S-acyl-2-thioethyl (SATE) prodrugs.
Nucleos(t)ide analogues are an important family of
bioactive compounds widely used as anticancer and anti-
viral drugs.1aꢀc Nevertheless, their clinical use is currently
limited by the appearance of resistance mechanisms
mainly associated with nucleos(t)ide metabolism.1a,d As
an example, the involvement of 50-nucleotidase II (cN-II)
is proposed as a new resistance mechanism to cytotoxic
nucleosides.1d To date, our research interests have focused
on the development of 50-monophosphate nucleoside
mimics containing a nonhydrolyzable PꢀC bond rather
than one with a PꢀO bond, as potential therapeutic agents.2
Thus, we previously reported the synthesis of nucleoside
β-hydroxyphosphonate analogues from the ex-chiral pool3
or by reduction of a β-ketophosphonate intermediate4
(Figure 1). These pathways allowed synthesis of the
desired mononucleotide analogues in the pyrimidine series
whereas the synthesis of purine analogues remained un-
achieved, especially for the guanine derivative. Further-
more, extensive assays were performed to prepare the
corresponding prodrugs, incorporating biolabile protect-
ing groups such as pivaloyloxymethyl (POM) or S-acyl-2-
thioethyl (SATE) moieties, using previously described
methodologies such as the coupling reaction of a β-hydroxy-
phosphonic acid (or its salts) with 2-(S-benzoylthio)ethanol
in the presence of TPSCl or MSNT,5 an alkylation reaction
using pivaloyloxymethyliodide,6 or the transesterification of
† These authors contributed equally.
(1) (a) Jordheim, L. P.; Durantel, D.; Zoulim, F.; Dumontet, C. Nat.
ꢂ
ꢀ
Rev. Drug Discovery 2013, 12, 447. (b) Krecmerova, M. In Herpesviridae -
A Look Into This Unique Family of Viruses; Magel, G. D., Ed.; InTech:
2012. (c) Mathe, C.; Gosselin, G. Antiviral Res. 2006, 71, 276. (d) Jordheim,
L. P.; Chaloin, L. Curr. Med. Chem. 2013, 20, in press.
(3) Gallier, F.; Peyrottes, S.; Perigaud, C. Eur. J. Org. Chem. 2007,
925.
ꢀ
(4) Meurillon, M.; Chaloin, L.; Perigaud, C.; Peyrottes, S. Eur. J.
Org. Chem. 2011, 3794.
(2) (a) Gallier, F.; Alexandre, J. A. C.; El Amri, C.; Deville-Bonne,
D.; Peyrottes, S.; Perigaud, C. ChemMedChem 2011, 6, 1094. (b) Gallier,
F.; Lallemand, P.; Meurillon, M.; Jordheim, L. P.; Dumontet, C.;
(5) (a) Benzaria, S.; Pelicano, H.; Johnson, R.; Maury, G.; Imbach,
J. L.; Aubertin, A. M.; Obert, G.; Gosselin, G. J. Med. Chem. 1996, 39,
4958. (b) Koh, Y. H.; Shim, J. H.; Wu, J. Z.; Zhong, W. D.; Hong, Z.;
Girardet, J. L. J. Med. Chem. 2005, 48, 2867.
ꢀ
Perigaud, C.; Lionne, C.; Peyrottes, S.; Chaloin, L. PLoS Comput. Biol.
2011, 7, e1002295.
r
10.1021/ol402143y
Published on Web 09/09/2013
2013 American Chemical Society

dimethylphosphonate protecting groups.7 However, none
of these were satisfactory.
Thus, we sought to investigate an original synthetic
strategy aiming to obtain the β-hydroxyphosphonate
nucleoside from a nucleoside epoxide intermediate, which
could be opened by an appropriate phosphite entity giving
rise to the expected products.
efficient reagents for the silylation step, promoting the use
of toluene and/or dichloromethane as solvent. This study
also revealed that, for an improved reaction yield, the use
of all reagents in excess was required.
Scheme 1. Synthesis of the Osidic β-(S)-Hydroxyphosphonate
Analogue 3
Figure 1. Key steps toward the synthesis of nucleoside
β-hydroxyphosphonates.
Although there are abundant literature reports of the
ring-opening reactions of epoxides by various nucleo-
philes,8 only a few examples describe the use of trialkyl-
phosphites and/or dialkyl phosphite salts.9 Among the
last, most of the reactions involved strong bases9a,10 and
harsh reaction conditions incompatible with the chemical
stability of biolabile groups. Thus, the methodology
developed herein allows the regioselective formation of
β-hydroxyphosphonate nucleosides under mild condi-
tions, compatible with the stability of the future prodrug
moiety. It starts from a nucleoside epoxide intermediate
and a silylated bis(alkyl)phosphite9a in the presence of a
Table 1. Ring-Opening Reaction of Epoxide 2 with Diethyl
Phosphite
reagents
silylation
step:
BF3
3
solvent (EtO)2P(O)H N,O-BSA
Et2O solvent yielda
(equiv) for 2 (%)
and t
(equiv)
(equiv)
1 toluene, Δ
2 toluene, rt
3 CH3CN, Δ
4 toluene, Δ
5 toluene, Δ
6 toluene, Δ
7 CH2Cl2, Δ
8 toluene, rt
3
3
3
3
3
6
6
6
3.2
3
3
3
3
3
6
6
6
toluene 44
toluene 14
CH3CN
3.2
3.2
0
Lewis acid such as BF3 etherate.
3
3.2
CH2Cl2 37
CH3CN 46
CH2Cl2 88
CH2Cl2 79
CH2Cl2 40
At first, we examined the key ring-opening reaction
using a model substrate (Scheme 1). For this purpose,
the epoxide 2 was synthesized from commercially available
1,2:5,6-di-O-isopropylidene-D-allofuranose in three steps,
including an intramolecular Mitsunobu reaction with
retention of configuration.11
3.2
6.4
6.4
(TMSCl) 12
(NEt3) 12
(TMSCl) 12
(NEt3) 12
9 toluene, rt
6
6
CH2Cl2 98b
Optimization of the experimental conditions (selected
data are reported in Table 1) showed that either N,O-
bis(trimethyl)silyl acetamide (N,O-BSA, entries 6 and 7)
or trimethylsilyl chloride (TMSCl, entries 8 and 9) are
a Isolatedyield ofderivative 3 after silica gelcolumn chromatography.
b Tedious filtration of the reaction mixture containing the silylated
phosphite was performed before addition to the epoxide, to eliminate
the triethylammonium salts. Δ means that the reaction was carried out
under reflux conditions.
(6) (a) Ryu, Y. H.; Scott, A. I. Org. Lett. 2003, 5, 4713. (b) Wiemer,
A. J.; Yu, J. S.; Shull, L. W.; Barney, R. J.; Wasko, B. M.; Lamb, K. M.;
Hohl, R. J.; Wiemer, D. F. Bioorg. Med. Chem. 2008, 16, 3652.
(7) (a) Hwang, Y. S.; Cole, P. A. Org. Lett. 2004, 6, 1555.
(b) Vepsalainen, J. J. Tetrahedron Lett. 1999, 40, 8491.
Following this, the applicability of the ring-opening
reaction to an epoxy-nucleoside intermediate was investi-
gated in the uracil series (Scheme 2). Peracetylation of the
1,2:5,6-di-O-isopropylidene-D-allofuranose was performed
as previously reported12 giving rise to compound 4, as a
(8) Schneider, C. Synthesis 2006, 3919.
(9) (a) Li, Z. G.; Racha, S.; Dan, L.; Elsubbagh, H.; Abushanab, E.
J. Org. Chem. 1993, 58, 5779. (b) Sardarian, A. R.; Shahsavari Fard, Z.
Synth. Commun. 2007, 37 (1ꢀ3), 289. (c) Sobhani, S.; Vafaee, A.
Tetrahedron 2009, 65, 7691.
(10) (a) Bichlmaier, I.; Kurkela, M.; Joshi, T.; Siiskonen, A.; Rueffer,
T.; Lang, H.; Suchanova, B.; Vahermo, M.; Finel, M.; Yli-Kauhaluoma,
J. J. Med. Chem. 2007, 50, 2655. (b) Giannessi, F.; Chiodi, P.; Marzi, M.;
Minetti, P.; Pessotto, P.; De Angelis, F.; Tassoni, E.; Conti, R.; Giorgi,
F.; Mabilia, M.; Dell’Uomo, N.; Muck, S.; Tinti, M. O.; Carminati, P.;
Arduini, A. J. Med. Chem. 2001, 44, 2383.
mixture of R/β anomers, in 88% yield. Uracil was con-
13
densed with 4 under Vorbruggen conditions leading to
€
(12) Guillerm, G.; Muzard, M.; Glapski, C.; Pilard, S. Bioorg. Med.
Chem. Lett. 2004, 14, 5803.
(13) Vorbruggen, H.; Krolikiewicz, K.; Bennua, B. Chem. Ber./Recl.
(11) Hanaya, T.; Sugiyama, K.; Kawamoto, H.; Yamamoto, H.
Carbohydr. Res. 2003, 338, 1641.
1981, 114, 1234.
Org. Lett., Vol. 15, No. 18, 2013
4779

intermediate 5, as a mixture of R/β products (ratio of 1/3).
At this stage, the two anomers could not be separated, and
the acetyl protecting groups were replaced by isopropyli-
dene groups using a two-step procedure, thus allowing an
efficient separation of the R and β anomers of nucleoside 6
by silica gel chromatography. A selective removal of the
hydroxyl protecting groups from the 50- and 60-positions led
to diol 7 as a precursor of the required epoxide. Treatment
of compound 7 under Mitsunobu conditions was per-
formed, but due to purification issues it was only isolated
in poor yield. Therefore, the hydroxyl in position 60 of the
diol was selectively mesylated.14 Intramolecular nucleophi-
lic substitution then gave rise to 8 in 47% yield (Scheme 2).
A first attempt of the ring-opening reaction was carried out
using previously optimized conditions (Table 1, entry 6).
Unfortunately, we only observed formation of the 2,50-
anhydro derivative 9 through an SN2 process (Scheme 2).
To avoid this side reaction, the N-3 position of uracil was
protected by a tert-butoxycarbonyl (Boc) group,15 which
was cleavable under similar conditions to those of the
20,30-isopropylidene protecting group. The fully protected
compound 10 was opened with silylated diethyl phosphite
under conditions analogous to some of those reported for
the opening of 2 in Table 1. When the conditions of Table 1,
entries 6 and 9 were applied, the expected β-(S)-hydroxy-
phosphonate 11 was obtained in 60% yield and 65% yield,
respectively. Thus, both procedures behaved similarly with
nucleoside epoxides as substrates. Finally, the protecting
groups were removed with aqueous trifluoroacetic acid,
giving rise to derivative 12 quantitatively.
groups was carried out under acidic conditions. Finally,
hydrolysis of the di(ethyl)phosphonoester was accom-
plished using trimethylsilyl bromide in DMF, with the
phosphonic acids 16aꢀc being isolated as sodium salts.
Scheme 2. Synthesis of the β-(S)-Hydroxy(diethyl)phosphonate
Analogue of Uridine 12 through a Ring Opening Reaction
With this synthetic route for the uracil containing
derivative established, we extended it to the preparation
of the corresponding analogues in the cytosine, hypox-
anthine, and guanine series. The first goalwas to synthesize
the epoxide intermediates (Scheme 3, compounds 13aꢀc).
As for compound 10, similar synthetic pathways were
developed (see SI for details), starting with the condensa-
tion of 6-chloro-purine or 2-amino-6-chloro-purine with
compound 4 under Saneyoshi16 or Vorbrugen conditions,
Scheme 3. Synthesis of β-(S)-Hydroxyphosphonate Derivatives
16aꢀc
13
€
respectively. The corresponding cytosine starting material was
obtained from the uracil-peracetylated analogue 5, via a two-
step procedure involving the use of Lawesson’s reagent.17
Thus, the ring-opening reactions of epoxides 13aꢀc
in the presence of silylated diethyl phosphite and of
BF3 etherate were performed under the previously defined
3
conditions, affording the desired β-(S)-hydroxyphosphonate
14aꢀc in lower yields than those for the uracil analogue.
We hypothesized the lower stability of the starting material
(formation of side-products was observed during the
reaction) as well as the influence of steric hindrance on the
β-face of the nucleoside, especially for the purine-containing
derivatives. Removal of the sugar and nucleobase protecting
(14) Hioki, H.; Yoshio, S.; Motosue, M.; Oshita, Y.; Nakamura, Y.;
Mishima, D.; Fukuyama, Y.; Kodama, M.; Ueda, K.; Katsu, T. Org.
Lett. 2004, 6, 961.
(15) Dini, C.; Drochon, N.; Guillot, J. C.; Mauvais, P.; Walter, P.;
Aszodi, J. Bioorg. Med. Chem. Lett. 2001, 11, 533.
(16) Saneyoshi, M.; Satoh, E. Chem. Pharm. Bull. 1979, 27, 2518.
(17) Chambert, S.; Gautier-Luneau, I.; Fontecave, M.; Decout, J. L.
J. Org. Chem. 2000, 65, 249.
4780
Org. Lett., Vol. 15, No. 18, 2013

31% yield, respectively. The 20,30-isopropylidene and Boc
protecting groups were removed under acid conditions as
previously described.
Scheme 4. Synthesis of the Bis(phenyl)SATE Prodrugs of
β-(S)-Hydroxyphosphonate Derivative in the Uracil and
Cytosine Series
In conclusion, the objective of this work, namely the
description of a new strategy for the synthesis of ribonu-
cleoside β-hydroxyphosphonate analogues and of a bis-
(phenylSATE) prodrug of a cN-II inhibitor of interest, has
been met. It is based on the regioselective ring-opening of a
nucleoside epoxide by a required phosphite under mild
conditions, owing to the potential instability of the biola-
bile prodrug under nucleophilic and/or basic conditions.
This synthetic pathway was first validated and optimized
for monosaccharide epoxide before being applied to
epoxy-nucleosides. It provided access to a novel β-(S)-
hydroxyphosphonate analogue of guanosine, which was
not available from a previously reported synthetic ap-
proach, as well as bis(phenylSATE) prodrugs of the uracil
and cytosine series.
Acknowledgment. This work was supported by institu-
tional funds from the Institut National du Cancer (INCa,
Project No. 2010-200) and Agence Nationale de la
Recherche (ANR Programme Blanc 2011-SIMI-7, Project
acronym cN-II Focus). M.M. and A.H. are grateful for
their doctoral fellowships to the Centre National de la
As we previously reported that, within the β-hydroxy-
phosphonate ribonucleoside series, the cytosine-containing
analogue was found to be a more potent inhibitor of
50-nucleotidase II (cN-II),2b we applied our strategy to the
synthesis of the corresponding bis(phenylSATE) prodrug
in a pyrimidine series (compounds 19 and 20, Scheme 4).
The ring-opening reaction was performed using bis-
(phenylSATE) phosphite (synthesized beforehand from
2-(S-benzoylthio)ethanol and PCl3 in the presence of
triethylamine and then in situ hydrolysis of the trialkylphos-
phite by silica, in 87% yield over the two steps).18 Thus, the
fully protected prodrugs 17 and 18 were isolated in 33 and
ꢀ
Recherche Scientifique & Region Languedoc-Roussillon
and the Ministere National de l’Enseignement, de la
ꢁ
Recherche et des Technologies, respectively. Thanks to
Mrs. M.-C. Bergogne (IBMM) for editing the manuscript.
Supporting Information Available. Experimental pro-
cedure, Scheme S1, and spectral data for final com-
pounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
(18) Briggs, A. D.; Camplo, M.; Freeman, S.; Lundstrom, J.; Pring,
B. G. Eur. J. Pharm. Sci. 1997, 5, 199.
The authors declare no competing financial interest.
Org. Lett., Vol. 15, No. 18, 2013
4781