Stereoselective synthesis of 4-C-methyl-2,3,5-tri-O-benzyl-D-ribofuranose and 4-C-methyl-2,3,5-tri-O-benzyl-L-lyxofuranose

Article abstract of DOI:10.1016/S0040-4039(03)00877-3

Sugar intermediates 4-C-methyl-2,3,5-tri-O-benzyl-D-ribofuranose (8b) and 4-C-methyl-2,3,5-tri-O-benzyl-L-lyxofuranose (8a) were synthesized by addition of alkylithium reagents to pentanones 3a,b. The nucleophilic additions proceeded with good stereoselectivity and good yields to give the titled compounds in four steps from perbenzylated methyl D-ribofuranoside and methyl 5′-deoxy-D-ribofuranoside.

Full text of DOI:10.1016/S0040-4039(03)00877-3

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 4109–4112
Stereoselective synthesis of
4-C-methyl-2,3,5-tri-O-benzyl-
D
-ribofuranose and
4-C-methyl-2,3,5-tri-O-benzyl-
L
-lyxofuranose
Serge H. Boyer,* Bheemarao G. Ugarkar and Mark D. Erion
Chemistry Department, Metabasis Therapeutics, Inc., 9390 Towne Centre Dr., San Diego, CA 92121, USA
Received 13 March 2003; revised 27 March 2003; accepted 27 March 2003
Abstract—Sugar intermediates 4-C-methyl-2,3,5-tri-O-benzyl-
nose (8a) were synthesized by addition of alkylithium reagents to pentanones 3a,b. The nucleophilic additions proceeded with
good stereoselectivity and good yields to give the titled compounds in four steps from perbenzylated methyl -ribofuranoside and
methyl 5%-deoxy- -ribofuranoside. © 2003 Elsevier Science Ltd. All rights reserved.
D-ribofuranose (8b) and 4-C-methyl-2,3,5-tri-O-benzyl-L-lyxofura-
D
D
Sugar-modified nucleosides and nucleotides have long
been used against a wide variety of diseases. Drugs such
as AZT, 2%,3%-dideoxy nucleosides and 2%,3%-didehydro-
dideoxy nucleosides are in used to treat HIV,1 arabino-
nucleosides such as fludarabine and cytarabine are used
to treat various leukemias,1b,2 whereas the antisense
oligonucleotide field uses modified sugar nucleosides to
design RNA/DNA probes.3
Although many syntheses of 4-C-modified sugars have
been published,6 in all but Johnson’s synthesis,6d the
C4-substituent was introduced via the aldol-Cannizaro
reaction of the corresponding C5-aldehyde with formal-
dehyde.7 While this procedure is very useful for the
synthesis of bis-(4-hydroxymethyl)-sugar analogs, low
selectivity is generally observed when attempting to
deoxygenate only one of the C4-hydroxymethyls.6e,f In
contrast, Johnson’s synthesis is highly stereoselective
and allows for introduction of a large array of sub-
stituents at the C4 position.6d However, this procedure
involves multiple steps and was judged unsuitable for
the synthesis of the current targets. Herein we describe
Tubercidin analogs with potent adenosine kinase (AK)
inhibition were shown to have anti-epileptic and anti-
inflammatory activities.4 However, such nucleosides
present limitations as therapeutic agents because of
toxicities arising from potential 5%-O-phosphorylation
and subsequent incorporation into the nucleotide pools
resulting in toxic side effects.5 Accordingly, our efforts
focused on designing nucleosides that would not be
phosphorylated. Since efficient phosphorylation is
dependent on sugar conformation and possibly substi-
tutions near the 5%-hydroxyl, we hypothesized that
introduction of a methyl group at the C4%-position of
nucleosides would cause subtle changes in the orienta-
tion of the 5%-hydroxyl group and thereby prevent
phosphorylation while maintaining enzyme inhibition.
To synthesize such nucleosides, suitably protected 4-C-
an alternative stereoselective synthesis of 4-C-methyl-
ribose (1) and 4-C-methyl- -lyxose (2).
D-
L
Upon retro-synthetic analysis of the target compounds,
the introduction of the C4-methyl group was envisioned
to proceed via the nucleophilic addition of a methyl
equivalent to pentanone 3a (Scheme 1). Such addition
could proceed through one of the three chelated transi-
tion states as shown in Figure 1 (Cram’s model, T1–3)
or through a non-chelated transition state (Felkin–
Ahn’s model, T4).8
methyl-D-ribose and 4-C-methyl-L-lyxose were needed
as key intermediates.
* Corresponding author. Tel.: 858-622-5576; fax: 858-622-5573;
e-mail: boyer@mbasis.com
Scheme 1.
0040-4039/03/$ - see front matter © 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0040-4039(03)00877-3

4110
S. H. Boyer et al. / Tetrahedron Letters 44 (2003) 4109–4112
Based on the transition states in Figure 1, the incoming
nucleophile would be expected to add to the re-face of
ketone 3a in transition state T2 whereas, in transition
state T3, the carbanion would add to the si-face giving
rise to the opposite stereochemistry at the 4-carbon. On
the other hand, no selectivity would be expected from
transition state T1 since the 5-carbon is achiral. Due to
the uncertainties associated with the stereochemical
outcome of a nucleophilic addition to ketone 3a via a
chelated transition state, we hypothesized that the non-
chelated addition of a carbanion proceeding through
transition state T4, where the stereochemical outcome is
only dictated by the chirality of the 3-carbon through
steric and electronic factors, should give rise to prod-
ucts with predictable stereochemistry.9
To further test the hypothesis, benzyloxymethyl
lithium12 was added under similar reaction conditions
to ketone 3b10 (Scheme 2). Once again, the reaction
mixture gave two products in a 7.5:1 ratio, which were
isolated by column chromatography. As predicted in
1
this case, the H NMR spectrum of the major isomer
matched with that of the minor isomer 7b11 from the
previous reaction, whereas the spectrum of the minor
isomer matched that of isomer 7a,11 indicating that the
addition took place with the same facial selectivity.
Unambiguous stereochemical assignments were made
by first removing the dithiane protecting group13 and
derivatizing both benzylated furanoses 8a,b11 into
methyl furanosides 10a,b (Scheme 3). Initial attempts to
remove the benzyl protecting groups via hydrogenolysis
resulted in significant decomposition. However this
problem was overcome by protecting the anomeric
hydroxyls of furanoses 8a,b as acetates prior to
hydrogenolysis. Further treatment of the triol interme-
diates 9a,b with 2,2-dimethoxypropane and a catalytic
amount of p-toluenesulfonic acid in CH2Cl2 resulted in
not only the formation of the 2,3-isopropylidene, but
also exchanged the anomeric acetate to produce methyl
Our hypothesis was first tested on ketone 3a.10 Addi-
tion of methyl lithium to a THF solution of ketone 3a
at −78°C produced a mixture of two products (Scheme
2) in a 12/1 ratio, as evidenced by integration of the
methyl resonances at 1.25 ppm (major) and 1.13 ppm
1
(minor) in the H NMR spectrum of the crude reaction
mixture. Both products were isolated by column chro-
matography and were tentatively assigned structure
7a11 (major, 69%) and 7b11 (minor, 8%).
1
furanosides 10a,b.11 Direct comparison of the H and
Figure 1.
Scheme 2.
Scheme 3. Reagents and conditions: (a) CaCO3, MeI, ACN/H2O/THF (2/9/2), reflux; (b) Ac2O, pyridine, DMAP; (c) Pd(OH)2/C,
1 atm H2, EtOAc/AcOH (5/1); (d) CH2Cl2, p-TsOH, Me2C(OMe)2.

S. H. Boyer et al. / Tetrahedron Letters 44 (2003) 4109–4112
4111
Scheme 4.
13C NMR spectra for both isopropylidene-protected
methyl furanosides with the spectrum of compound
10b, synthesized independently as an intermediate in
4. Wiesner, J. B.; Ugarkar, B. G.; Castellino, A. J.;
Barankiewicz, J.; Dumas, D. P.; Gruber, H. E.; Foster,
A. C.; Erion, M. D. J. Pharmacol. Exp. Ther. 1999, 289,
1669.
the total synthesis of S-(4%-methyladenosyl)-L-homocys-
teine,6d confirmed the original assignment of the stereo-
chemistry of the 4-carbon. Further evidence confirming
the structures of 10a and 10b came from the solvent
effect studies conducted during the protection of the 2-
and 3-hydroxyls of intermediates 9a,b. It was noticed
that in DMF, triol 9a (Scheme 4) gave a new product,
which was assigned structure 11 based on 1H NMR
studies, whereas 9b gave 10b, confirming the lyxo-
configuration for intermediate 9a.
5. Davies, L. P.; Jamieson, D. D.; Baird-Lambert, J. A.;
Kazlauskas, R. Biochem. Pharmacol. 1984, 33, 347.
6. (a) Rajwanshi, V. K.; Kumar, R.; Hansen, M. K.; Wen-
gel, J. J. Chem. Soc., Perkin Trans. 1 1999, 1407; (b)
Bjo¨rsne, M.; Classon, B.; Kers, I.; Samuelsson, B. Bioorg.
Med. Chem. Lett. 1995, 5, 43; (c) McVinish, L. M.;
Rizzacasa, M. A. Tetrahedron Lett. 1994, 35, 923; (d)
Johnson, C. R.; Esker, J. L.; Van Zandt, M. C. J. Org.
Chem. 1994, 59, 5854; (e) Waga, T.; Nishizaki, T.;
Miyakawa, I.; Ohrui, H.; Meguro, H. Biosci. Biotech.
Biochem. 1993, 57, 1433; (f) Tam, T. F.; Fraser-Reid, B.
Can. J. Chem. 1979, 57, 2818; (g) Youssefyeh, R. D.;
Verheyden, J. P. H.; Moffat, J. G. J. Org. Chem. 1979,
44, 1301 and references cited therein.
7. Schaffer, R. J. Am. Chem. Soc. 1959, 81, 5452.
8. (a) Bartlet, P. A. Tetrahedron 1980, 36, 3; (b) Eliel, E. L.
Asymmetric Synthesis; Morrison, J. D., Ed.; Academic
Press: New York, 1983; Vol. 2, Part A, p. 125; (c) Reetz,
M. T. Angew. Chem., Int. Ed. Engl. 1984, 23, 556; (d)
Jurczak, J.; Pikul, S.; Bauer, T. Tetrahedron 1986, 42,
447.
In conclusion, this work demonstrates the generality
and the good stereoselectivity of the non-chelation con-
trolled addition of alkyllithium reagents to ketones 3a,b
for the stereoselective synthesis of 4-C-substituted sugar
analogs.
Acknowledgements
We thank Professor C. R. Johnson for kindly providing
1
us the H and 13C NMR spectra of compound 10b.
9. For non-chelation-controlled addition of organolithium
reagents to related systems, see: (a) Mukai, C.; Mohar-
ram, S. M.; Azukizawa, S.; Hanaoka, M. J. Org. Chem.
1997, 62, 8095; (b) Brus, W.; Horns, S.; Redlich, H.
Synthesis 1995, 335; (c) Nakatani, K.; Arai, K.;
Hirayama, N.; Matsuda, F.; Terashima, S. Tetrahedron
1992, 48, 633; (d) Nakatani, K.; Arai, K.; Hirayama, N.;
Matsuda, F.; Terashima, S. Tetrahedron Lett. 1990, 31,
2323; (e) Hikota, M.; Tone, H.; Horita, K.; Yonemitsu,
O. J. Org. Chem. 1990, 55, 7; (f) Tone, H.; Nishi, T.;
Oikawa, Y.; Hikota, M.; Yonemitsu, O. Tetrahedron
Lett. 1987, 28, 4569; (g) Wilcox, C. S.; Gaudino, J. J. J.
Am. Chem. Soc. 1986, 108, 3102.
References
1. (a) Johnston, M. I.; Hoth, D. F. Science 1993, 260, 1286;
(b) Prisbe, E. J.; Maag, H.; Verheyden, J. P. H.;
Rydzewsky, R. M. In Nucleosides and Nucleotides as
Antitumor and Antiviral Agents; Chu, C. K.; Baker, D.
C., Eds.; Plenum Press: New York, 1993; p. 101.
2. Gutheil, J.; Kearns, K. In The Chemotherapy Source
Book, 2nd Ed.; Perry, M. C., Ed.; Williams & Wilkins:
Baltimore, MD, 1996; p. 317.
3. (a) Uhlmann, E.; Peyman, A. Chem. Rev. 1990, 90, 544;
(b) Milligan, J. F.; Matteucci, M. D.; Martin, J. C. J.
Med. Chem. 1993, 36, 1923; (c) Antisense Research and
Applications; Crooke, S. T.; Lebleu, B., Eds.; CRC Press:
Boca Raton, FL, 1993; (d) Antisense Therapeutics;
Agrawal, S., Ed.; Humana Press: Totowa, NJ, 1996; (e)
Thuong, N. T.; He´le`ne, C. Angew. Chem., Int. Ed. Engl.
1993, 32, 666.
10. Ketone 3a was synthesized from methyl 2,3,5-tri-O-ben-
zyl-D-ribofuranoside (4a, Barker, R.; Fletcher, H. G., Jr.
J. Org. Chem. 1961, 26, 4605) in the following two-step
sequence. Ketone 3b was synthesized in a similar fashion
from methyl 5-deoxy-2,3-di-O-benzyl-D-ribofuranoside
(4b, Chen, X.; Schneller, S. W. J. Med. Chem. 1993, 36,
3727–3730).

4112
S. H. Boyer et al. / Tetrahedron Letters 44 (2003) 4109–4112
11. Selected data for compound 7a: 1H NMR (CDCl3, 200
MHz) l 7.45–7.20 (m, 15H), 5.46 (d, J=10.1 Hz, 1H),
4.90–4.37 (m, 7H), 4.12 (dd, J=7.1, 1.8 Hz, 1H), 3.75 (d,
J=7.1 Hz, 1H), 3.53 (s, 1H), 3.50 (d, J=8.6 Hz, 1H),
3.41 (d, J=8.6 Hz, 1H), 2.90–2.40 (m, 4H), 2.15–1.80 (m,
0.3H), 3.96 (d, J=5 Hz, 0.3H), 3.81 (d, J=5 Hz, 0.7H),
3.57 (br d, J=7 Hz, 0.7H), 3.33 (d, J=10 Hz, 0.7H), 3.22
(s, 0.6H), 3.17 (d, J=10 Hz, 0.7H), 1.33 (s, 0.9H), 1.29 (s,
2.1H). Compound 10a: 1H NMR (CDCl3, 200 MHz) l
4.95 (s, 1H), 4.69 (d, J=6 Hz, 1H), 4.52 (d, J=6 Hz,
1H), 3.75–3.65 (m, 2H), 3.32 (s, 3H), 2.45 (m, 1H), 1.48
(s, 3H), 1.31 (s, 3H), 1.3 (s, 3H); 13C NMR (CDCl3, 125
MHz) l 112.8, 108.9, 86.7, 86.5, 85.6, 67.0, 54.7, 26.2,
1
2H), 1.25 (s, 3H). Compound 7b: H NMR (CDCl3, 200
MHz) l 7.50–7.20 (m, 15H), 5.18 (d, J=10.5 Hz, 1H),
4.86–4.48 (m, 6H), 3.97 (s, 2H), 3.65 (br s, 1H), 3.55 (d,
J=9.9 Hz, 1H), 3.30 (d, J=9.9 Hz, 1H), 2.90–2.55 (m,
1
24.6, 23.5. Compound 10b: H NMR (CDCl3, 200 MHz)
1
l 4.90 (s, 1H), 4.68 (d, J=6 Hz, 1H), 4.64 (d, J=6 Hz,
1H), 3.67 (dd, J=11, 2.9 Hz, 1H), 3.55 (dd, J=12.1, 2.9
Hz, 1H), 3.45 (t, J=11 Hz, 1H), 3.42 (s, 3H), 1.48 (s,
3H), 1.32 (s, 3H), 1.26 (s, 3H); 13C NMR (CDCl3, 125
MHz) l 112.2, 109.7, 90.6, 87.3, 82.9, 69.4, 55.7, 26.5,
24.9, 18.2.
4H), 2.20–1.80 (m, 2H), 1.13 (s, 3H). Compound 8a: H
NMR (CDCl3, 200 MHz) l 7.45–7.15 (m, 15H), 5.48–
5.42 (m, 0.5H), 5.28 (dd, J=11.2, 4 Hz, 0.5H), 4.88–4.32
(m, 6H), 4.10 (t, J=4 Hz, 0.5H), 4.02 (d, J=5.1 Hz,
0.5H), 3.97 (dd, J=7.1, 2 Hz, 0.5H), 3.90–3.74 (m, 1.5H),
3.64 (d, J=10.2 Hz, 1H), 3.46 (d, J=9.1 Hz, 0.5H), 3.75
(d, J=3 Hz, 0.5H), 1.48 (s, 1.5H), 1.28 (s, 1.5H). Com-
pound 8b: 1H NMR (CDCl3) l 7.40–7.10 (m, 15H),
5.28–5.22 (m, 0.3H), 5.16 (br d, J=7 Hz, 0.7H), 4.80–
4.28 (m, 6H), 4.24 (d, J=5 Hz, 0.7H), 4.16–4.00 (m,
12. Ireland, E. R.; Smith, M. G. J. Am. Chem. Soc. 1988,
110, 854 and references cited therein.
13. Trost, B. M.; Preckel, M.; Leichter, L. M. J. Am. Chem.
Soc. 1975, 97, 2224.