Anomeric configuration-directed diastereoselective C-C bond formation in vinyl sulfone-modified carbohydrates: a general route to branched-chain sugars.

Article abstract of DOI:10.1021/ol0342047

[structure: see text] This is the first report on the diastereoselective addition of carbon nucleophiles to vinyl sulfone-modified hex-2-enopyranosides and pent-2-enofuranosides. Nucleophiles add to the C-2 position from a direction opposite to that of th

Full text of DOI:10.1021/ol0342047

ORGANIC
LETTERS
2003
Vol. 5, No. 8
1285-1288
Anomeric Configuration-Directed
Diastereoselective C−C Bond Formation
in Vinyl Sulfone-Modified
Carbohydrates: A General Route to
Branched-Chain Sugars
,‡
Aditya K. Sanki,† Cheravakkattu G. Suresh,* Usha D. Falgune,† and
,§
Tanmaya Pathak*
Organic Chemistry DiVision (Synthesis), National Chemical Laboratory,
Pune 411 008, India, DiVision of Biochemical Sciences, National Chemical
Laboratory, Pune 411 008, India, and Department of Chemistry,
Indian Institute of Technology, Kharagpur 721 302, India
tpathak@chem.iitkgp.ernet.in
Received February 5, 2003
ABSTRACT
This is the first report on the diastereoselective addition of carbon nucleophiles to vinyl sulfone-modified hex-2-enopyranosides and pent-2-
enofuranosides. Nucleophiles add to the C-2 position from a direction opposite to that of the disposition of the anomeric methoxy group. This
novel concept of anomeric configuration-directed stereocontrolled carbon−carbon bond formation in vinyl sulfone-modified carbohydrates is
general in nature and has been implemented in the synthesis of new hexopyranosyl and pentofuranosyl branched-chain sugars and densely
functionalized carbohydrates.
Branched-chain sugars, having carbon substituents at the
nonterminal carbon atoms of the chains, are components of
many natural products, especially antibiotics.1,2 Branched-
chain sugars also constitute an important class of function-
alized intermediates, useful for further transformations.2-4
The most common methods for the syntheses of branched-
chain sugars in general and branched-chain sugars having
no substituent at the branching carbon atoms in particular
involve the reduction of alkylidene glycosides, opening of
sugar-derived oxiranes by carbon nucleophiles, or Michael-
type addition of carbon nucleophiles to nitro-alkene sugars
and hex-2-enopyranosid-4- or hex-3-enopyranosid-2-uloses.1,2
The usefulness of nitro-alkene sugars5 and hexenopyranoside
uloses6 in the synthesis of branched-chain sugars is limited
only to pyranose systems and therefore cannot be considered
as being of general utility. The carbohydrate oxiranes often
* Corresponding author.
† Organic Chemistry Division (Synthesis), National Chemical Laboratory.
‡ Division of Biochemical Sciences, National Chemical Laboratory.
§ Indian Institute of Technology.
(1) Yoshimura, J. AdV. Carbohydr. Chem. Biochem. 1984, 42, 69-134.
(2) Ferrier, R. J. Carbohydrate Chemistry: Monosaccharides, Disac-
charides and Specific Oligosaccharides; The Royal Society of Chemistry:
Cambridge, 1968-2001; Vols. 1-32.
(3) Hanessian, S. Total Synthesis of Natural Products: The Chiron
Approach; Pergamon Press: Oxford, 1983.
(4) Witczak, J. Z., Nieforth, K. A., Eds. Carbohydrates in Drug Design;
Marcel Dekker, Inc.: New York, 1997.
10.1021/ol0342047 CCC: $25.00 © 2003 American Chemical Society
Published on Web 03/19/2003

produce a mixture of regioisomers depending on several
factors.7 On the other hand, it is not clear from the available
literature whether the reduction of C-2 alkylidene glycosides
is influenced by the 3-OH protecting group lowering the
stereoselectivity of these reactions as is known for 2-keto
sugars.8 It is therefore obvious that there is a great need for
new and general methodologies for the stereocontrolled
construction of carbon-carbon bonds in carbohydrates.
Michael-type addition reaction of carbon nucleophiles to
vinyl sulfone-modified carbohydrates should be considered
as an efficient route for the synthesis of branched-chain
sugars mainly because almost all carbohydrates, either in
pyranose or furanose form, could be converted to their vinyl
sulfone derivatives very easily.9-11 Moreover, the product
of the reaction carrying a sulfone functionality has the
potential to undergo a wide variety of transformations.12,13
We observed that a nucleophile attacked the C-2′ position
of the 2′-enesulfone nucleoside exclusively from the R-face
of the pent-2′-enofuranosyl moiety of a â-nucleoside.14 This
observation led us to investigate the effect of the anomeric
configuration on the stereochemical outcome of the addition
of nucleophiles to vinyl sulfone-modified carbohydrates
(Scheme 1).
on the stereochemical outcome of the reactions was not
obvious. The addition of primary amines to 1r and 1â
exclusively produced C-2 equatorial (gluco) products. Sec-
ondary amines, on reactions with 1â, produced only gluco
derivatives but, with 1r, produced a mixture in which gluco
was still the predominant isomer.11b On the other hand,
sterically bulky tertbutylamine reacted only with 1â (and
not 1r) at elevated temperatures to produce the gluco
derivative in high yield.15
We report here for the first time that in contrast to our
earlier observation,11b,15 1r and 1â on reaction with carbon
nucleophiles generated products that manifested fully the
directing effect of the anomeric configurations. In both cases,
the carbanion added to the planar olefinic systems of 1r and
1â from a direction opposite to that of the disposition of the
anomeric methoxy group.
Thus, the nucleophile generated from CH3NO2 and NaOMe
reacted with 1r to produce a single isomer 2r (60%).
Similarly, the nucleophile generated from dimethylmalonate
and sodium hydride produced exclusively 3r (82%) (Scheme
2). It should be noted that in contrast to the R-D-manno
Scheme 2
Scheme 1
Although amines added in diastereoselective fashion to
1r and 1â, the directing effect of the anomeric configuration
configuration of the product obtained from the reaction of
carbon nucleophiles with 2,3-dideoxy-3-C-nitro-R-D-erythro-
hex-2-enopyranoside,16 1r produced 2r having three axially
disposed functional groups on three consecutive carbon atoms
(R-D-altro configuration) of a six-membered system.
(5) (a) Baer, H. H.; Ong, K. S. Can. J. Chem. 1968, 46, 2511-2517. (b)
Paulsen, H.; Grewe, W. Chem. Ber. 1974, 107, 3013-3019. (c) Sakakibara,
T.; Sudoh, R. J. Chem. Soc., Chem. Commun. 1974, 69-70. (d) Sakakibara,
T.; Sudoh, R. Carbohydr. Res. 1976, 50, 191-196. (e) Sakakibara, T.; Seta,
A.; Tachimori, Y.; Takamoto, T.; Sudoh, R. Chem. Lett. 1977, 1365-1366.
(f) Baer, H. H.; Hanna, Z. S. Carbohydr. Res. 1980, 85, 136-142. (g)
Sakakibara, T.; Nakagawa, T. Carbohydr. Res. 1987, 163, 239-246. (h)
Sakakibara, T.; Takaide, A.; Sata, A. Carbohydr. Res. 1992, 226, 271-
278. (i) Sakakibara, T.; Tokuda, K.; Hayakawa, T.; Seta, A. Carbohydr.
Res. 2000, 327, 489-496.
(9) Collins, P. M.; Ferrier, R. J. Monosaccharides: Their Chemistry and
Their Roles in Natural Products; John Wiley & Sons: Chichester, UK,
1996.
(10) Simpkins, N. S. Sulphones in Organic Synthesis; Pergamon Press:
Oxford, 1993.
(6) (a) Paulsen, H.; Koebernick, W. Carbohydr. Res. 1977, 56, 53-66.
(b) Paulsen, H.; Koebernick, W. Carbohydr. Res. 1977, 110, 2127-2145.
(d) Yunker, M. B.; Plaumann, D. E.; Fraser-Reid, B. Can. J. Chem. 1977,
55, 4002-4009. (e) Shafizadeh, F.; Ward, D. D.; Pang, D. Carbohydr. Res.
1982, 102, 217-230. (f) Nakahara, Y.; Fujita, A.; Beppu, K.; Ogawa, T.
Tetrahedron 1986, 42, 6465-6476. (g) The problems of using carbohydrate
enones as precursors have been discussed in ref 12.
(7) (a) Rajanbabu, T. V.; Nugent, W. A. J. Am. Chem. Soc. 1994, 116,
986-997. (b) Benefice-Malouet, S.; Coe, P. L.; Walker, R. T. Carbohydr.
Res. 1992, 229, 293-305.
(8) (a) Kerekgyarto, J.; Rako, J.; Agoston, K.; Gyemant, G.; Szurmai,
Z. Eur. J. Org. Chem. 2000, 3931-3935. (b) Lichtenthaler, F. W.;
Schneider-Adams, T. J. Org. Chem. 1994, 59, 6728-6734. (c) Lichtenthaler,
F. W.; Kaji, E.; Weprek, S. J. Org. Chem. 1985, 50, 3505-3515.
(11) (a) Takai, I.; Yamamoto, A.; Ishido, Y.; Sakakibara, T.; Yagi, E.
Carbohydr. Res. 1991, 220, 195-207. (b) Ravindran, B.; Sakthivel, K.;
Suresh, C. G.; Pathak, T. J. Org. Chem. 2000, 65, 2637-2641.
(12) Ravindran, B.; Deshpande, S. G.; Pathak, T. Tetrahedron, 2001,
57, 1093-1098.
(13) For a review on desulfonylation reactions, see: Najera, C.; Yus,
M. Tetrahedron 1999, 55, 10547-10658.
(14) (a) Wu, J. C.; Pathak, T.; Tong, W.; Vial, J. M.; Remaud, G.;
Chattopadhyaya, J. Tetrahedron 1988, 44, 6705-6722. (b) Bera, S.;
Langley, G. J.; Pathak, T. J. Org. Chem. 1998, 63, 1754-1760.
(15) Suresh, C. G.; Ravindran, B.; Pathak, T.; Narasimha Rao, K.;
Sasidhar Prasad, J. S.; Lokanath, N. K. Carbohydr. Res. 2002, 337, 1507-
1512.
(16) Sakakibara, T.; Sudoh. R. J. Org. Chem. 1975, 40, 2823-2825.
1286
Org. Lett., Vol. 5, No. 8, 2003

Scheme 3
The nucleophile generated from CH3NO2 and NaOMe
reacted with 1â to produce a single isomer 2â (56%).
Similarly, the nucleophile generated from dimethylmalonate
and sodium hydride produced 3â (98%) (Scheme 3).
Compound 1â produced the expected11b,15 â-D-gluco ana-
logues 2â and 3â. Elucidation of crystal structures of 2r/
3r and 2â/3â unambiguously established the configurations
at positions C-2 and C-3 of these compounds (Figures 1 and
2).17,18
Figure 2. X-ray structure of 3â.
5â (55%), respectively. Dimethylmalonate adducts 6r (74%)
and 6â (87%) were also synthesized in a diastereoselective
fashion (Schemes 4 and 5).
Scheme 4
It should be noted that the Michael addition of various
nucleophiles to related furanose systems such as 1-(5-O-trityl-
2,3-dideoxy-3-toluenesulfonyl-â-D-glyceropent-2-enofurano-
syl) uracil and the corresponding adenine derivative generated
trans-(xylo) products exclusively in most cases; a few of these
reactions have produced a mixture of cis-(ribo) and trans-
Figure 1. X-ray structure of 3r.
To establish the general pattern of the directing effect of
the anomeric configurations, anomerically pure pentofura-
noses 4r and 4â were synthesized in multisteps from
D-xylose.19 Compounds 4r and 4â, on reaction with car-
banions generated from CH3NO2/NaOMe and CH3NO2/
tBuO-K+, produced branched-chain sugars 5r (73%) and
Scheme 5
(17) Crystal data. Compound 3r: C25H28O10S, M ) 520.53, orthorhom-
bic, a ) 12.130(8), b ) 14.187(9), c ) 14.515(9) Å, U ) 2498(3) Å3, T
) 295(2) K, space group P212121 (no. 19), Z ) 4, µ(Mo KR) ) 0.186
mm-1, 14 406 reflections measured, 5551 unique (Rint ) 0.066), which were
used in all calculations. The final wR(F2) was 0.112 (all data). Compound
3â: C25H28O10S, M ) 520.53, monoclinic, a ) 5.6780(5), b ) 16.5197-
(15), c ) 13.5563(12) Å, â ) 95.212(2)°, U ) 1266.3(2) Å3, T ) 295(2)
K, space group P21 (no. 4), Z ) 2, µ(Mo KR) ) 0.184 mm-1, 7436
reflections measured, 4949 unique (Rint ) 0.015), which were used in all
calculations. The final wR(F2) was 0.0943 (all data).
(18) All these data will be reported and discussed in detail in a full paper.
(19) Sanki, A. K.; Pathak, T. Synlett 2002, 1241-1244.
Org. Lett., Vol. 5, No. 8, 2003
1287

(xylo) products, although the latter were overwhelmingly the
major components. The stereoselectivity of protonation at
C-3 after the addition of nucleophiles to C-2 of vinyl sufone
modified furanosides can be attributed to steric interactions
between the nucleophiles attached at C-2 and the toluene-
sulfonyl group at C-3.14a From the steric consideration, it is
therefore expected that 4r and 4â, on reaction with nucleo-
philes, will produce arabino and xylo derivatives, respec-
tively. In fact, the structures of 5r (Scheme 4) and 5â
(Scheme 5) have been established unambiguously with the
help of X-ray diffraction of single crystals and were found
to be in line with the expected configurations at C-2 and
C-3.18
earlier11b,12 as well as 2r/3r and 2â/3â (Schemes 2 and 3)
can be predicted to a great extent by comparing the spectral
data of our compounds with those of the products obtained
from the addition of nucleophiles to nitro-alkene pyrano-
sides.5,16
To highlight the usefulness of the branched-chain sugars
generated so far with the help of our method, 3â was
deprotected under acidic conditions to 7â in high yield.
Mesylation of the crude dihydroxy compound 7â furnished
a densely functionalized branched-chain sugar 8â (Anal.
Calcd for C19H24O11S2: C, 46.33; H, 4.90. Found: C, 46.33;
H, 5.07) (Scheme 6). This novel Michael acceptor 8â is a
Although the structures of compounds 2r/3r, 2â/3â, and
5r/5â have been established unambiguously with the help
of X-ray diffraction of single crystals,18 the multiplicities of
the anomeric protons20 of these products are to a great extent
indicative of the configurations at positions C-2 and C-3 as
well. For example, the H-1/H-2 coupling constant (J1,2)
values of authentic methyl R-D-arabinofuranosides range
between 0 and 3.0 Hz.19,21a,b Those of authentic methyl R-D-
ribofuranosides19,21a,c range between 4 and 4.9 Hz, and those
for methyl R-D-xylofuranosides,19,21a,b,d-f on the other hand,
always range between 4 and 4.7 Hz. Excluding the possibility
of any lyxo derivative formation for steric reasons,14a the J1,2
values of 5r (0.0 Hz) and 6r (0.0 Hz) surely indicated the
presence of an arabino configuration in these molecules. On
the other hand, it is evident from the available data that the
J1,2 values of authentic methyl â-D-xylofuranosides19,21a,d,f,g
range between 0 and 2.3 Hz and those of methyl â-D-
ribofuranosides19,21a,c are close to 0.0 Hz. Although the J1,2
values of 5â (3.5 Hz) and 6â (3.9 Hz) were higher than those
of the authentic xylofuranosides, these values were surely
indicative of the absence of a ribo configuration in these
compounds. On the other hand, the configurations at C-2
and C-3 of many of the pyranosides reported by us
Scheme 6
ready intermediate for nucleophilic attack at C-4 as well as
C-6 by external nucleophiles for the synthesis of extensively
modified carbohydrates.
25.5
In summary, a facile route for the synthesis of new
branched chain sugars has been designed by utilizing the
directing effects of the anomeric configuration of easily
accessible vinyl sulfone-modified carbohydrates for the first
time. Studies on the further functionalization of the branched
chain sugars and the application of this methodology in the
synthesis of higher sugars, unnaturally linked oligosaccha-
rides, and acyclic synthons are currently in progress.22
(20) Selected analytical data. Compound 2r: mp 188-189 °C; [R]D
+26.4 (c 0.880, CHCl3); 1H NMR δ 4.66 (s, 1 H, H-1), 5.40 (s, 1 H, PhCH).
Anal. Calcd for C21H23O8NS: C, 56.11; H, 5.15; N, 3.11. Found: C, 56.31;
H, 4.96; N, 3.27. Compound 2â: mp 149-150 °C; [R]D25.5 -64.4 (c 0.977,
1
CHCl3); H NMR δ 4.69 (d, J ) 8.3 Hz, 1 H, H-1), 5.31 (s, 1 H, PhCH).
Anal. Calcd for C21H23O8NS: C, 56.11; H, 5.15; N, 3.11. Found: C, 56.29;
H, 5.20; N, 3.21. Compound 3r: mp 239-240 °C; [R]D27.5 +18.3 (c 1.00,
CHCl3); IR (Nujol, cm-1); 1H NMR δ 4.58 (s, 1 H, H-1), 5.38 (s, 1 H,
PhCH). Anal. Calcd for C25H28O10S: C, 57.68; H, 5.41. Found: C, 57.64;
H, 5.27. Compound 3â: mp 98-99 °C; [R]D27.5 -98.2 (c 0.9, CHCl3); 1H
NMR δ 5.17 (d, J ) 7.8 Hz, 1 H, H-1), 5.21 (s, 1 H, PhCH). Anal. Calcd
for C25H28O10S: C, 57.68; H, 5.41. Found: C, 57.63; H, 5.04. Compound
26.5
1
5r: mp 88-89 °C; [R]D
+58.8 (c 1.01, CHCl3); H NMR δ 4.91 (s, 1
Acknowledgment. This work has been supported by a
research grant from the Department of Science and Technol-
ogy, New Delhi, India. A.K.S. thanks the Council of
Scientific and Industrial Research, New Delhi, India, for a
fellowship.
H, H-1). Anal. Calcd for C21H25O7NS: C, 57.91; H, 5.78; N, 3.21; S, 7.36.
Found: C, 57.89; H, 5.67; N, 3.01; S, 7.57. Compound 5â: mp 125-126
°C; [R]D26.5 +14.3 (c 1.017, CHCl3); 1H NMR δ 5.02 (d, J ) 3.5 Hz, 1 H,
H-1). Anal. Calcd for C21H25O7NS: C, 57.91; H, 5.78; N, 3.21; S, 7.36.
Found: C, 58.03; H, 5.91; N, 3.14; S, 7.56. Compound 6r: mp 91-92
27.5
1
°C; [R]D
+51.4 (c 0.90, CHCl3); H NMR δ 4.96 (s, 1 H, H-1). Anal.
Calcd for C25H30O9S: C, 59.27; H, 5.96; S, 6.33. Found: C, 59.16; H,
6.64; S, 6.43. Compound 6â: mp 99-100 °C; [R]27.5 +9.4 (c 1.00, CHCl3);
1H NMR δ 5.16 (d, J ) 3.9 Hz, 1 H, H-1). Anal. Cald for C25H30O9S: C,
59.27; H, 5.96. Found: C, 59.04; H, 5.94.
OL0342047
(21) (a) Kawana, M.; Kuzuhara, H.; Emoto, S. Bull. Chem. Soc. Jpn.
1981, 54, 1492-1504. (b) Montgomery, J. N.; Thorpe, M. C.; Clayton, S.
D.; Thomas, H. J. Carbohydr. Res. 1974, 32, 404-407. (c) Kawana, M.;
Koresawa, T.; Kuzuhara, H. Bull. Chem. Soc. Jpn. 1983, 56, 1095-1100.
(d) Yamashita, A.; Rosowsky, A. J. Org. Chem. 1976, 41, 3422-3425. (e)
Bock, K.; Pederson, C. Carbohydr. Res. 1979, 73, 85-91. (f) Liptak, A.;
Neszmelyi, A.; Kovac, P.; Hirsch, J. Tetrahedron 1981, 37, 2379-2382.
(g) Casini, G.; Goodman, L. J. Am. Chem. Soc. 1964, 86, 1427-1431.
(22) It is not clear at present whether the difference in mode of addition
of amines and carbon nucleophiles to 1r is a result of thermodynamic versus
kinetic control alone. In general, any discussion on the diastereoselectivity
of addition of nucleophiles to complicated systems such as 1r/1â or 4r/
4â should also take into account other factors such as, electrostatic
interactions, stereoelectronic control, steric hindrance, and hydrogen bond-
ing.
1288
Org. Lett., Vol. 5, No. 8, 2003