Alternative synthetic routes to 2′,3′-didehydro-2′,3′-dideoxy-5-hydroxymethyluridine

Article abstract of DOI:10.1016/j.tetlet.2006.09.083

Alternative syntheses for the nucleoside analogue 2′,3′-didehydro-2′,3′-dideoxy-5-hydroxymethyluridine starting from 5-methyluridine and uridine are described.

Full text of DOI:10.1016/j.tetlet.2006.09.083

Tetrahedron Letters 47 (2006) 8361–8363
Alternative synthetic routes to
20,30-didehydro-20,30-dideoxy-5-hydroxymethyluridine
Raymond Chung and Karen S. Anderson*
Yale School of Medicine, Department of Pharmacology, 333 Cedar Street, New Haven, CT 06520, United States
Received 11 August 2006; revised 11 September 2006; accepted 15 September 2006
Available online 9 October 2006
Abstract—Alternative syntheses for the nucleoside analogue 20,30-didehydro-20,30-dideoxy-5-hydroxymethyluridine starting from 5-
methyluridine and uridine are described.
Ó 2006 Elsevier Ltd. All rights reserved.
Several 20,30-dideoxynucleosides and 20,30-didehydro-
20,30-dideoxynucleosides have been used for the treat-
ment of patients with human immunodeficiency virus
(HIV). These nucleosides, in their 50-triphosphate form,
inhibit the polymerization of proviral DNA facilitated
by HIV reverse transcriptase. 20,30-Dideoxynucleosides
that are FDA-approved for clinical use include 20,30-
dideoxycytidine (ddC), 20,30-dideoxyinosine (ddI) and
30-azido-30-deoxythymidine (AZT). 20,30-Didehydro-20,30-
dideoxythymidine (d4T) has also been FDA-approved
for the treatment of HIV. However, resistance to these
nucleoside RT inhibitors (NRTIs) has become a major
obstacle in HIV therapy and the development of new
NRTIs is paramount.
tant intermediate towards potential inhibitors contain-
ing analogues of d4T. In our laboratory, alternative
syntheses of 20,30-didehydro-20,30-dideoxy-5-hydroxy-
methyluridine (1) have been developed to provide
straightforward synthesis of 1 without the need for
special equipment or reagents and allows for a wider
range of transformations that can be carried out on
the 5-hydroxymethyl substituent.
The synthesis of hydroxymethyluridine 1 was initiated
with commercially available uridine (2). First, the 20
and 30 hydroxyls of uridine were protected as an aceto-
nide by reaction with 2,2-dimethoxypropane in the pres-
ence of p-toluensulfonic acid using a modified literature
procedure.4 The resulting alcohol 3 was subjected to
hydroxymethylation conditions (0.5 M KOH, paraform-
aldehyde, 60 °C) described by Gavriliu et al.1 to give
the bis alcohol product 4. At this point, the purification
of 4 proved to be a challenge due to the co-elution of a
side product of similar polarity during flash column
chromatography. As a result, compound 4 was only
purified on a short silica gel column to remove most
of the impurities present. Bis alcohol 4 was treated with
50% acetic acid solution to yield intermediate 5, which
was not purified. Instead, the primary hydroxyl groups
of crude 5 were protected as t-butyldiphenylsilyl ethers,
resulting in compound 6 in a moderate yield. The free
vicinal diol was converted to the corresponding bisxan-
thate (7) by reaction with 5 N NaOH and carbon disul-
fide, followed by the addition of methyl iodide. The
reaction of 7 under radical deoxygenation conditions
(Bu3SnH, AIBN, toluene, reflux) enabled the formation
of 20,30-didehydro-20,30-dideoxynucleoside 8 in a 77%
isolated yield.5 Finally, cleavage of the silyl ethers with
20,30-Didehydro-20,30-dideoxy-5-hydroxymethyluridine (1)
is a compound similar in structure to d4T. In contrast
to d4T, 20,30-didehydro-20,30-dideoxy-5-hydroxymethyl-
uridine possesses a pendant hydroxyl moiety off of the
5-methyl substituent on the pyrimidine base. Gavriliu
et al.1 and Renoud-Grappin et al.2 have synthesized
20,30-didehydro-20,30-dideoxy-5-hydroxymethyluridine
and exploited the hydroxyl group as a handle for further
chemical transformation in studies directed towards the
synthesis of new d4T-derived anti-HIV compounds.
Pugazhenthi et al.3 have also solved the X-ray crystal
structure of this compound.
It is apparent from previously published work1,2 as well
as ongoing work in our laboratory that 20,30-didehydro-
20,30-dideoxy-5-hydroxymethyluridine (1) is an impor-
*
Corresponding author. E-mail: Karen.anderson@yale.edu
0040-4039/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2006.09.083

8362
R. Chung, K. S. Anderson / Tetrahedron Letters 47 (2006) 8361–8363
O
N
O
O
N
HN
HN
HN
OH
2,2-dimethoxypropane,
pTSA.H2O
paraformaldehyde,
KOH (aq.)
60oC (31%)
O
O
N
O
acetone (73%)
O
O
O
OH
OH
O
O
HO
HO
HO
O
O
2
3
4
O
N
O
HN
OH
HN
OTBDPS
TBDPSCl,
imid., DMAP
NaOH, CS2,
MeI (41%)
50% AcOH (aq.),
reflux
O
O
N
DMF (40%, 2 steps)
O
O
OH
OH
OH
OH
HO
TBDPSO
5
6
O
O
N
O
HN
OTBDPS
HN
OTBDPS
HN
OH
Bu3SnH, AIBN
TBAF
O
N
O
O
N
toluene, reflux (77%)
THF (quantitative)
O
OCS2Me
OCS2Me
O
O
TBDPSO
TBDPSO
HO
7
8
1
Scheme 1.
O
O
O
HN
HN
HN
TBDPSCl,
imid., DMAP
2,2-dimethoxypropane,
pTSA.H2O
O
N
O
N
O
N
acetone (58%)
DMF(86%)
O
O
O
OH
O
O
HO
HO
Br
TBDPSO
OH
O
O
9
10
11
O
N
O
HN
OH
NBS,
benzoyl peroxide
HN
TBDPSCl, imid.,
DMAP
sat. NaHCO3 (aq.)
THF (31%, 2 steps)
O
O
N
CCl4, reflux
DMF (24%)
O
O
O
O
TBDPSO
TBDPSO
O
O
12
13
O
N
HN
OTBDPS
50% AcOH (aq.)
reflux
O
O
6
O
TBDPSO
O
14
Scheme 2.

R. Chung, K. S. Anderson / Tetrahedron Letters 47 (2006) 8361–8363
8363
tetrabutylammonium fluoride (TBAF) resulted in the
formation of 20,30-didehydro-20,30-dideoxy-5-hydroxy-
methyluridine (1) in a quantitative yield (Scheme 1).
The structure of 1 was confirmed by comparing our
X-ray crystallographic data to a published crystal
structure.3
new NRTIs as its 5-hydroxymethyl functionality and
allows chemists to perform chemical transformations
to vary substituents and explore diversity at that loca-
tion in the molecule. The utility of this molecule has
already been demonstrated by other research groups1,2
and these alternative routes present various possibilities
for chemistry that may not have been possible in previ-
ous synthetic endeavors.
Alternatively, a synthesis of compound 1 starting with
commercially available 5-methyluridine (9) was also
developed. Starting with compound 9, the 20 and 30
hydroxyls were protected as an acetonide to yield com-
pound 10 in a moderate yield (58%). The 50 hydroxyl
of 10 was immediately protected as the t-butyldiphenyl-
silyl ether to give compound 11 in an 86% isolated yield.
Following a published procedure,6 the 5-methyl substi-
tuent was halogenated in the presence of N-bromosuc-
cinimide and benzoyl peroxide in boiling CCl4 to give
an intermediate bromide 12. Due to instability concerns,
bromide 12 was not purified. Compound 12 was imme-
diately subjected to reaction with saturated sodium
bicarbonate in aqueous solution. THF was added to
facilitate dissolution of compound 12. The desired
product 13 was then reacted with TBDPSCl to protect
the free hydroxyl to give compound 14. The reaction of
the doubly protected 14 with 50% acetic acid cleaved the
acetonide to give compound 6 without cleavage of
the silyl ethers (Scheme 2). It is noteworthy to mention
that it is advantageous to start with 5-methyluridine
because it allows for the manipulation of the 5-hydroxy-
methyl substituent without reaction at the 50 hydroxyl.
Abdel-Rahman and El Ashri has been able to synthesize
compound 13 directly from compound 3. However, their
method required the use of microwave irradiation to
affect the hydroxymethylation step.7
Supplementary data
Experimental procedures and spectral data for com-
pounds 1, 3, 4, 6–8, 10, 11, 13 and 14 are available with
this article. Supplementary data associated with this
article can be found, in the online version, at doi:10.1016/
j.tetlet.2006.09.083.
Acknowledgements
We would like to thank Dr. Christopher Incarvito of the
Yale University X-Ray Crystallographic Facility for the
X-ray crystallographic analysis of compound 1 and Dr.
Guangxiu Dai for helpful discussions. This work was
supported by NIH GM49551 to K.S.A.
References and notes
1. Gavriliu, D.; Fossey, C.; Fontaine, G.; Benzaria, S.; Ciurea,
A.; Delbederi, Z.; Lelong, B.; Laduree, D.; Aubertin, A.
M.; Kirn, A. Nucleosides Nucleotides Nucleic Acids 2000,
19, 1017–1031.
2. Renoud-Grappin, M.; Fossey, C.; Fontaine, G.; Laduree,
D.; Aubertin, A. M.; Kirn, A. Antiviral Chem. Chemother.
1998, 9, 205–223.
3. Pugazhenthi, Umarani; Delbaere, Louis T. J.; Kumar,
Sashi V. P.; Stuart, Allan L.; Gupta, Sagar V. Acta
Crystallogr., Sect. C: Crystal Struct. Commun. 1994, C50,
1262–1265.
4. Winans, K. A.; Bertozzi, C. R. Chem. Biol. 2002, 9, 113–129.
5. Chu, C. K.; Bhadti, V. S.; Doboszewski, B.; Gu, Z. P.;
Kosugi, Y.; Pullaiah, K. C.; Van Roey, P. J. Org. Chem.
1989, 54, 2217–2225.
In conclusion, we have developed alternate synthetic
routes to 20,30-didehydro-20,30-dideoxy-5-hydroxy-
methyluridine (1). The first route takes the advantage
of t-butyldiphenylsilyl protecting groups to aid in chro-
matographic purification as well as utilizes standard
organic transformations to facilitate ease of synthesis.
In addition, the second route allows the manipulation
of the hydroxyl at the 5 position of the base without
reaction of the 50 hydroxyl, opening the door to general
alcohol transformations at that position. This com-
pound presents an opportunity for the development of
6. Piao, D.-Y. Ph.D. thesis, Brown University, RI, 2000.
7. Abdel-Rahman, A.-H.; El Ashry, E. H. Synlett 2002, 12,
2043–2044.