Practical synthetic routes to carbon-substituted nucleosides

Article abstract of DOI:10.1016/j.tetlet.2011.09.074

Robust practical routes to three different carbon-substituted nucleosides are described. Short synthetic procedures for preparing 1-homonucleosides, 2'-C-methyl-carbanucleosides and 2'-C-methyl-cyclopro-pyl/cyclopentyl fused bicyclic ribose analogues are described starting from readily available building blocks. Crown Copyright

Full text of DOI:10.1016/j.tetlet.2011.09.074

Tetrahedron Letters 52 (2011) 6415–6419
Contents lists available at SciVerse ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Practical synthetic routes to carbon-substituted nucleosides
David C. Pryde a, Donald S. Middleton a, Peter T. Stephenson a, Philip Wainwright b, , Adrian Maddaford b,
⇑
Xiurong Zhang b, David Leese b, Rebecca Glen b, James Hart b, Neil Forrest b, Thierry Guyot a
a WorldWide Medicinal Chemistry, Pfizer Global Research and Development, Ramsgate Road, Sandwich CT13 9NJ, United Kingdom
b Peakdale Molecular Ltd, Chapel-en-le-Frith, Derbyshire SK23 0PG, United Kingdom
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 1 June 2011
Revised 31 August 2011
Accepted 16 September 2011
Available online 24 September 2011
Robust practical routes to three different carbon-substituted nucleosides are described. Short synthetic
procedures for preparing 10-homonucleosides, 20-C-methyl-carbanucleosides and 20-C-methyl-cyclopro-
pyl/cyclopentyl fused bicyclic ribose analogues are described starting from readily available building
blocks.
Crown Copyright Ó 2011 Published by Elsevier Ltd. All rights reserved.
Keywords:
Nucleosides
Carbanucleosides
Homonucleosides
Sugar synthesis
Nucleosides represent a useful class of both anti-viral and anti-
cancer agents.1 The nucleoside first requires activation by cellular
kinases to the corresponding triphosphate, which can then act as
either a competitive inhibitor of viral and/or cellular nucleic acids
or a pseudo-substrate which is incorporated into a growing nucleic
acid strand. Due to many nucleosides and/or their triphosphates
showing poor selectivity and unacceptable toxicity, much research
has gone into finding more specific and less toxic analogues.2 This
has been attempted through modifications to either the base or
the sugar component of the nucleosides. The latter strategy has
enjoyed some considerable success, with many examples now of
clinically useful nucleoside analogues such as acyclovir3 1, emtri-
citabine4 2 and NM2835 3 (Fig. 1).
the sugar through activation of a substituted methylene group at
the 10-position (viz. 8), or de novo construction of the base from
a 10-aminomethyl intermediate (viz. amine 9). Amine 9 can clearly
be accessed directly from 7.7
Typically, the requisite methylene-substituted grouping in 8
and/or 9 is derived from a ribose analogue, for example, a glycosyl
halide 7, and a C–N synthon such as a nitrile group. These methods
introduce a mixture of diastereoisomers at the anomeric centre,
and typically use benzyl protecting groups for the ribose hydroxyl
groups. Ideally, we wished to access the building blocks 8 and 9 in
diastereoisomerically pure form and be able to use simple acid-
labile protecting groups.
We initially looked into known methods of synthesising ribose
analogues in which the 10-position featured a hydroxymethyl
group. Several syntheses of such substrates have been reported,
for example the epoxide opening route shown in Scheme 1.6b By
In this Letter, we wish to describe novel syntheses of several
carbon-substituted nucleoside analogues (Fig. 2).
In particular, we describe routes to 10-homonucleosides 4 in
which an extra methylene has been inserted between the base
and the 10-position, 20-C-Me-carbanucleosides
5
and 20-C-Me
substituted versions of cyclopropyl/cyclopentyl fused ribose ana-
logues 6.
HO
S
The routes described herein represent some diverse and practi-
cally useful chemistry to access a range of substituted nucleoside
structures.
N
F
N
HO
N
O
NH2
O
N
N
HO
NH2
10-Homonucleosides of general structure 10 are well known in
the literature.6 Their synthesis generally can be achieved in one
of the two ways shown in Figure 3. Firstly, base attachment to
NH2
N
1
O
O
2
N
N
HO
O
HO
OH
3
⇑
Corresponding author.
E-mail address: phil.wainwright@peakdale.co.uk (P. Wainwright).
Figure 1. Clinically precedented nucleoside analogues.
0040-4039/$ - see front matter Crown Copyright Ó 2011 Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2011.09.074

6416
D. C. Pryde et al. / Tetrahedron Letters 52 (2011) 6415–6419
H2N
O
O
OH
N
CO2Me
HO
HO
CO2Me
NH2
N
N
N
a
HO
O
O
Ph3P
O
O
N
N
N
N
14
15
HO
OH
O
5
4
HO
b
NH2
O
HO
OH
N
O
OH
N
TBDPSO
HO
CO2Me
TBDPSO
TBDPSO
O
c
HO
OH
O
O
6
O
O
16
17
18
Figure 2. Target carbon-substituted nucleosides.
d,e
O
O
O
OH
f
HO
TBDPSO
X
Displace
with base
C-N
O
O
synthon
Base
8
HO
OH
O
O
19
O
O
X= suitable leaving group
Hal
HO
HO
g
7
HO
OH
10
HO
OH
De novo
base
O
O
O
HO
NH2
construction
OTs
TBDPSO
h,i
Base
HO
9
HO
OH
20
O
O
HO
4A, 4U, 4C, 4G
Base= A, U, C, G
OH
Figure 3. Literature strategies towards 10-homonucleosides.
Scheme 2. A robust synthetic route to all four 10-homonucleosides. Reagents and
conditions: (a) Ph3PCHCO2Me, MeCN, reflux, 3 h; (b) TBDPSCl, imidazole, dichlo-
romethane, rt, 73% over two steps; (c) DIBAL-H, dichloromethane, ꢀ78 °C?rt, 24 h,
87%; (d) MsCl, dichloromethane, 1 h, 0 °C?rt, 99%; (e) KOtBu, THF, 2 h, rt, 63%; (f)
OsO4, NMO, acetone, rt, 4 h then NaIO4, NaBH4, EtOAc, 3 h, rt, 41%; (g) TsCl, Et3N,
dichloromethane, reflux, 4 h, 94%; (h) nucleobase, LiH, DMSO, 90 °C, 5–16 h, 36–
60%, then (i) deprotection, 40–80% (see Supplementary data for more detail).
O
O
OH
OH
OH
TrO
TrO
TrO
mcpba
O
very low yield
Ph3P=
88%
O
O
O
O
O
13
11
12
Scheme 1. Attempted synthesis of 10-homologated nucleoside core. Reagents and
conditions: (a) methyltriphenylphosphonium bromide, nBuLi, THF, ꢀ78 °C?rt,
16 h; (b) m-CPBA, DCE, reflux.
O
O
O
O
N
N
HO
HO
H
N
N
HO
OH
HO
OH
incorporating an acetonide function at the 20/30-position it was
hoped to bias the diastereoselectivity of the epoxide-opening reac-
tion, but in our hands we were not able to secure acceptable yields
of any desired products, possibly as a consequence of this choice of
protecting groups. Derivatisation of the known ribose analogue 11
and then addition of m-chloroperbenzoic acid (m-CPBA) to the ace-
tonide-protected ribose analogue 12 gave a very low yield of the
desired homonucleoside in what was a very messy reaction.
We then investigated ways of increasing the effectiveness of
this type of ring-closure through olefin activation (Scheme 2).
Treatment of the known protected ribose 14 with methoxycarbon-
ylmethylene triphenylphosphorane gave the ester 15 as a single
diastereoisomer according to the method of Moffat.8 While this
strategy introduced an extra carbon atom at the anomeric centre,
by fixing the stereochemistry at the anomeric position at this early
stage, we felt this to be acceptable. Alcohol protection to give 16
and functional group interconversion through alcohol 17 and olefin
18 led to the 10-homoribose analogue 19. Activation of the
hydroxymethyl group as its tosylate 20 and displacement with
all four of the nucleobases proceeded well.
NH2
NH2
<0.1%
4C
0.6%
Figure 4. NOE data for compound 4C.
Confirmation of the cis-stereochemistry was confirmed by the
NOE analysis on 4C (Fig. 4).
Full experimental detail is provided in the Supplementary data.
This completed a practically useful synthesis of all four ribose
homonucleosides as single diastereoisomers.
Carbocyclic ribose analogues have been synthesised in a variety
of ways, including methods to build up the cyclopentane ring using
ring closing metathesis and constructing a cylopentane ring start-
ing from a ribose analogue precursor. As mentioned above, ribose
systems that contain substitution, for example methyl group sub-
stitution at the 20-position, have been the subject of considerable
research efforts as potential treatments for HCV infection.5,10 The
insertion of a 20-C-methyl group favours the 30-endo northern con-
formation. Methods for the stereospecific introduction of a 20-C-Me
group into a carbanucleoside are much less well precedented.11
Liao et al.12 recently described
methyl-neplanocin derivatives, however we developed alternative
route to this class of nucleosides (Scheme 3).
a
novel route to 20-b-C-
In the majority of cases the desired base was used directly in the
displacement reaction, that is, adenine, cytosine and uracil, but in
the case of the target guanine base, displacement was carried out
with 2-amino-6-chloropurine followed by the conversion of chlo-
rine into oxygen with 2-mercaptoethanol prior to deprotection.9
We chose to start from the chiral pool, in which the stereo-
chemistry of the 20/30 centres was already fixed, and the required

D. C. Pryde et al. / Tetrahedron Letters 52 (2011) 6415–6419
6417
O
O
O
O
O
TBDMSO
O
O
a,b
c
HO
O
O
O
O
HO
OH
21
22
23
d
O
N3
OMs
TBDMSO
TBDMSO
TBDMSO
g
e,f
O
O
O
O
O
O
26
25
24
h,i
N
NHCHO
Cl
H
N
N
TBDMSO
TBDMSO
Cl
j
N
N
N
N
O
O
O
O
27
28
k,l
N
N
NH2
NH2
N
N
m
HO
HO
N
N
N
N
HO
HO
OH
OH
5
29
+ 4'-diastereoisomer
Scheme 3. Synthetic route to an adenine substituted 20-C-Me-carbanucleoside. Reagents and conditions: (a) acetone, H2SO4, rt, 100%; (b) Dess–Martin periodinane,
dichloromethane, rt, 84%; (c) TBDMSCl, MeCN, Et3N, rt, 44%; (d) tBuOK, tBuLi, BrCH2PPh3, THF, ꢀ78 °C, 64%; (e) NaBH4, CeCl3, EtOH, rt, 50%; (f) MsCl, dichloromethane, Et3N, rt;
(g) NaN3, DMF, rt, 46% over two steps; (h) PPh3, H2O, THF, rt, 91%; (i) 4,6-dichloro-5-(N-formyl)-aminopyrimidine, dioxane, Et3N, reflux, 99%; (j) diethoxymethyl acetate,
120 °C, 51%; (k) ammonia, MeOH, 100 °C, 84%; (l) 90% aqueous TFA, rt, 82%; (m) 1 atm H2, 5% Pd(C), MeOH, rt, 25%.
20-C-Me group was already installed. In particular, 20-C-Me-
ribonolactone13 21 was used as a versatile and readily available
starting material.
The ribonolactone 21 was protected as its acetonide, oxidised to
the aldehyde 22 and then converted to the silyl enol ether 23 in
good overall yield according to the method of Shiozaki et al.14
Treatment of 23 with bromomethyl triphenylphosphine gave the
cyclopentenone 24 which was selectively reduced12 using a mix-
ture of NaBH4 and CeCl3 and then mesylated to give 25. Azide dis-
placement proceeded smoothly to yield 26, from which the
adenine ring of 28 was built up via the pyrimidine analogue 27.
Chloride displacement and a global deprotection gave the cyclo-
pentene nucleoside 29. The final step that remained was the
hydrogenation of the olefin. A simple hydrogen pressure heteroge-
neous reduction using palladised charcoal gave an approximately
1:1 mixture of diastereoisomers at the 40-centre of the target 5.
While this lack of selectivity was disappointing, we have not as
yet explored any of the myriad methods that could offer a more
selective reduction at this final step, or indeed attempting the
hydrogenation at an earlier step in which the ring is much more
sterically crowded by protecting groups which would greatly im-
prove this final step in the synthesis.
OH
O
O
steps
TBDPSO
HO
O
O
OH
HO
30
31
steps
OTBDPS
Another way to bias a ribose analogue towards the northern 30-
endo conformation is to fuse the cyclopentane ring to a cyclopro-
pane ring at the 40/60 positions.15 We were interested in combining
this structural feature with the 20-C-Me insertion described above
to access target compounds of structure 6.
TBDPSO
OH
steps
TBDPSO
O
OH
HO
OH
33
32
Few methods of making targets of this type are known in the lit-
erature. Lee et al.16 described a method of preparing the adenine
analogue 34 in 2006 as shown in Figure 5.
steps
N
This method started from the commercially available ribonolac-
tone 30, and followed a similar route to that described in Scheme 3
above in that a cyclopentenol 31 was also targeted, and the olefin
used to insert the cyclopropane ring in 32 using a Simmons–Smith
reaction. Protecting group manipulation led to 33, from which a
cyclic sulfate 34 was constructed, and then used as a leaving group
to insert the adenine base and yield the final target nucleoside 35
after deprotection.
O
NH2
N
TBDMSO
O
S
HO
O
N
O
O
N
OH
HO
34
35
Figure 5. The route by Lee et al. route to 20-C-methyl cyclopropyl fused
cyclopentanes.

6418
D. C. Pryde et al. / Tetrahedron Letters 52 (2011) 6415–6419
NHBz
OH
N
TBDMSO
a,b
N
HO
O
O
O
HO
OH
37
36
c
NHBz
NHBz
N
N
N
N
d,e
O
O
Si
Si
O
O
O
OH
O
OH
O
O
Si
Si
39
38
f
NH2
NHBz
N
N
g
N
N
HO
HO
O
O
HO
OH
HO
OH
40
6
Scheme 4. An expeditious route to 20-C-Me cyclopentane/cyclopropane fused carbanucleosides. Reagents and conditions: (a) 4-N-benzoyl-cytosine, DIAD, PPh3, THF, 0 °C,
70%; (b) AcOH, water, 70 °C, 64%; (c) tetra-isopropyl-dichloro-siloxane, imidazole, DMF, 76%; (d) Dess–Martin periodinane, dichloromethane, rt, 82%; (e) MeLi, THF, ꢀ78 °C,
69%; (f) TBAF, THF, rt, 82%; (g) MeOH, NH3, rt, 32%.
0.2%
Cl
OH
N
NH2
N
TBDMSO
TBDMSO
a
HO
N
H
N
N
N
O
O
O
O
36
41
HO
OH
43
3.4%
b
Figure 6. NOE data for compound 43.
NH
N
NH2
N
TBDMSO
c
lar Mitsunobu conditions to give the coupled material 41 in good
yield. Ammonia displacement to 42 and deprotection gave a
straightforward alternative synthesis18 of the final nucleoside 43.
Confirmation of the cis-stereochemistry was provided by the
NOE analysis on 43 (Fig. 6).
Throughout this Letter, we have described robust synthetic
routes to install carbon substitution in a range of different loca-
tions around a ribose, or the ribose ring analogues. The methodol-
ogy described herein could be adapted to further expand the range
of substitution available to create yet more nucleoside diversity.
HO
N
N
N
N
O
O
HO
OH
43
42
Scheme 5. An expeditious route to cyclopentane/cyclopropane fused carbanucle-
osides. Reagents and conditions: (a) 4-chloro-7H-pyrrolo[2,3-a]pyrimidine, DIAD,
PPh3, THF, 0 °C, 16 h, 79%; (b) MeOH, NH3, 100 °C, 24 days, 90%; (c) AcCl, MeOH, rt,
5 h, 85%.
Our initial route sought to cyclopropanate intermediates similar
to 24 and 25 as a simple entry into the desired ring system. All
these attempts failed, presumably due to steric encumbrance of
the olefin. Our next route initially followed a very similar path to
that of Lee. The cyclopentane 3617 was built using the same
chemistry, but at this stage our route diverged in order to shorten
and simplify the synthesis. We chose to insert the base directly
onto 36 which proceeded well to give the nucleoside analogue
37. The 30 and 50 hydroxys were then protected using a tetra-
isopropyl-disiloxane group that favours the protection of 1,3-diols.
This created a single free alcohol group in 33 that was oxidised and
methylated to give a single diastereoisomer 39. Siloxane deprotec-
tion gave 40, and final benzoyl deprotection gave the final target 6.
The alcohol intermediate 36 is of course a highly versatile inter-
mediate, in which the first step illustrated in Scheme 4 could easily
incorporate any number of base analogues to generate a diverse set
of nucleosides.
Supplementary data
Supplementary data (experimental procedures and analytical
data) associated with this article can be found, in the online ver-
sion, at doi:10.1016/j.tetlet.2011.09.074.
References and notes
1. Chemistry of Nucleosides and Nucleotides; Townsend, L. B., Ed.; Plenum Press:
New York, 1994; Vol. 3,.
2. For a recent review, see: Mathé, C.; Périgaud, C. Eur. J. Org. Chem. 2008, 1489.
3. Gnann, J. W., Jr.; Barton, N. H.; Whitley, R. J. Pharmacotherapy. 1983, 3, 275.
4. Saag, M. S. Clin. Infect. Dis. 2006, 42, 126.
5. Pierra, C.; Amador, A.; Benzaria, S.; Cretton-Scott, E.; D’Amours, M.; Mao, J.;
Mathieu, S.; Moussa, A.; Bridges, E. G.; Standring, D. N.; Sommadossi, J.-P.;
Storer, R.; Gosselin, G. J. Med. Chem. 2006, 49, 6614.
6. (a) Bobek, M.; Farkas, J. Collect. Czech. Chem. Commun. 1969, 34, 1684; (b)
Hossain, N.; Blaton, N.; Peeters, O.; Rozenski, J.; Herdewijn, P. Tetrahedron 1996,
52, 5563; (c) Scremin, C. L.; Boal, J. H.; Wilk, A.; Phillips, L. R.; Beaucage, S. L.
As an illustration (Scheme 5), 36 was treated with the commer-
cially available 4-chloro-7H-pyrrolo[2,3-a]pyrimidine under simi-

D. C. Pryde et al. / Tetrahedron Letters 52 (2011) 6415–6419
6419
Biorg. Med. Chem. Lett. 1996, 6, 207; (d) Winkley, M. W. Carbohydr. Res. 1973, 31,
245.
12. Liao, X.; Butora, G.; Olsen, D. B.; Carroll, S. S.; McMasters, D. R.; Leone, J. F.;
Stahlhut, M.; Doss, G. A.; Yang, L.; MacCross, M. Tetrahedron Lett. 2008, 49,
4149.
13. Booth, K. V.; da Cruz, F. P.; Hotchkiss, D. J.; Jenkinson, S. F.; Jones, N. A.;
Weymouth-Wilson, A. C.; Clarkson, R.; Heinz, T.; Fleet, G. W. J. Tetrahedron:
Asymmetry 2008, 19, 2417.
14. Shiozaki, M.; Arai, M.; Kobayashi, Y.; Kasuya, A.; Miyamoto, S.; Furukawa, Y.;
Takayama, T.; Haruyama, H. J. Org. Chem. 1994, 59, 4450.
15. Marquez, V. E.; Choi, Y.; Comin, M. J.; Russ, P.; George, C.; Huleihel, M.; Ben-
Kasus, T.; Agbaria, R. J. Am. Chem. Soc. 2005, 127, 15145.
7. For some elegant, more recent methods towards 10-homonucleosides, see: (a)
Busca, P.; McCort, I.; Prangé, T.; Merrer, Y. L. Eur. J. Org. Chem 2006, 2403; (b)
Lamberth, C. Org. Prep. Proc. Int. 2002, 34, 149.
8. Ohrui, H.; Jones, G. H.; Moffatt, J. G.; Maddox, L. M.; Christensen, A. T.; Byram, S.
K. J. Am. Chem. Soc. 1975, 97, 4602.
9. Jeong, L. S.; Schinazi, R. F.; Beach, W.; Kim, H. O.; Nampalli, S.;
Shanmuganathan, K.; Alves, A. J.; McMillan, A.; Chu, C. K.; Mathis, R. J. Med.
Chem. 1993, 36, 181.
10. Eldrup, A. B.; Prhavc, M.; Brooks, J.; Bhat, B.; Prakash, T. P.; Song, Q.; Bera, S.;
Bhat, N.; Dande, P.; Cook, P. D.; Bennett, F.; Carroll, S. S.; Ball, R. G.; Bosserman,
M.; Burlein, C.; Colwell, L. F.; Fay, J. F.; Flores, O. A.; Getty, K.; LaFemina, R. L.;
Leone, J.; MacCoss, M.; McMasters, D. R.; Tomassini, J. E.; Von Langen, D.;
Wolanski, B.; Olsen, D. B. J. Med. Chem. 2004, 47, 5284.
16. (a) Lee, J. A.; Kim, H. O.; Tosh, D. K.; Moon, H. R.; Kim, S.; Jeong, L. S. Org. Lett.
2006, 8, 5081; (b) Jeong, L. S.; Lee, J. A.; Kim, H. O.; Tosh, D. K.; Moon, H. R.; Lee,
S. J.; Lee, K. M.; Sheen, Y. Y.; Moon, W. C. Nucleosides, Nucleotides, Nucleic Acids
2007, 26, 1021.
17. Michel, B. Y.; Strazewski, P. Chem. Eur. J. 2009, 15, 6244.
18. Jacobsen, K. A.; Mackman, R. L.; Joshi, B. V. Chem. Abstr. 2006, 145, 272000.
11. Li, H.; Lee, W.; Yoo, J. C.; Hong, J. H. Bull. Korean Chem. Soc. 2009, 30, 2039.