Direct synthesis of β-mannans. A hexameric [→3)-β-D-man-(1→4)-β-D-man-(1]3 subunit of the antigenic polysaccharides from Leptospira biflexa and the octameric (1→2)-linked β-D-mannan of the Candida albicans phospholipomannan. X-ray crystal struc

Full text of DOI:10.1021/ja015985e

5826
J. Am. Chem. Soc. 2001, 123, 5826-5828
mannans derived from C. albicans phosphopeptidomannans
induce TNF1-R synthesis from cells of the macrophage lineage
and bind to macrophage cell membranes.8 The octamer 2 was
chosen for synthesis as the shortest of a series of phospho-
lipomannans isolated recently and shown to have strong TNF-R
inducing properties in vivo and in vitro.8d,9 The two syntheses
together illustrate the ease with which the â-1,2, â-1,3, and â-1,4
mannobiose linkages may now be incorporated into complex
oligosaccharides.
Direct Synthesis of â-Mannans. A Hexameric
[f3)-â-D-Man-(1f4)-â-D-Man-(1]3 Subunit of the
Antigenic Polysaccharides from Leptospira biflexa
and the Octameric (1f2)-Linked â-D-Mannan of the
Candida albicans Phospholipomannan. X-ray Crystal
Structure of a Protected Tetramer
David Crich,*,† Hongmei Li,† Qingjia Yao,† Donald J. Wink,†
Roger D. Sommer,‡ and Arnold L. Rheingold‡
The synthesis of 1 began with donor 3,10 which was readily
prepared from phenyl 1-thio-R-D-mannoside by conversion to the
4,6-O-p-methoxybenzylidene derivative,11 followed by dibenzy-
lation and oxidation3a,e to a single sulfoxide. Donor 4 was obtained
from diol 53a by selective protection of the equatorial alcohol by
treatment with Bu2SnO,12 then PMB chloride, followed by
benzylation of the residual hydroxy group in 6, and, eventually,
oxidation of 7. Donor 3 was then converted to the methyl
glycoside 8 by activation with Tf2O at -78 °C in CH2Cl2 in the
presence of 2,4,6-tri-tert-butylpyrimidine (TTBP)3d followed by
addition of methanol. The p-methoxybenzylidene group was
removed with camphorsulfonic acid (CSA) in methanol, giving
9, regioselective monobenzylation of which, with Bu2SnO and
benzyl bromide, afforded the acceptor 10.
Department of Chemistry, UniVersity of Illinois at Chicago
845 West Taylor Street, Chicago, Illinois 60607-7061
Department of Chemistry and Biochemistry
UniVersity of Delaware, Academy Street
Newark, Delaware 19716
ReceiVed April 10, 2001
ReVised Manuscript ReceiVed May 3, 2001
Protocols for the reliable, efficient, diastereoselective synthesis
of â-mannopyranosides have only been developed within the past
decade. Among the methods now available,1,2 the direct ones
developed in this laboratory from 4,6-O-benzylidene protected
thiomannosides and their sulfoxides3 are arguably optimal, owing
to the combination of high yield, excellent diastereoselectivity,
and the ease of operation that they offer. We now turn to the
application of this methodology, an off-shoot of the sulfoxide
glycosylation method,4 to the synthesis of â-mannans and select
as first proving grounds a mixed hexasaccharide (1) and a
homogeneous octamer (2).
The mannobiose [f3)-â-D-Man-(1f4)-â-D-Man-(1] was re-
cently characterized as the main repeat unit of the antigenic
polysaccharides from Leptospira biflexa serovar patoc strain Patoc
I.5 This particular mannan was chosen as target, aside from its
potential importance as a genus-specific leptospiral antigen,
because of its intriguing, alternating â-(1f3)-â-(1f4)-configu-
ration: the hexamer was selected as being sufficiently long to
prove the chemistry. Oligomeric â-1,2-linked mannans such as 2
are found in the C. albicans cell wall phosphopeptidomannan.6
They are immunogenic and elicit specific antibodies in both
humans and animals.7 Furthermore, it has been shown that â-1,2-
Standard activation of 4 with Tf2O and TTBP in CH2Cl2 at
-78 °C, followed by the addition of 10 gave the disaccharide 11
in 88% yield as a separable 11.6/1 â/R mixture,13 a typical
selectivity for the â-mannosylation of the somewhat hindered
glucopyranose 4-OH.3c Exposure of 11 to DDQ14 then afforded
the alcohol 12 in 83% yield (Scheme 1).
† University of Illinois at Chicago.
Standard activation of donor 3 followed by addition of 12
afforded the trisaccharide 13 as a 9.0/1 â/R mixture. Selective
‡ University of Delaware.
(1) (a) Barresi, F.; Hindsgaul, O. Can. J. Chem. 1994, 72, 1447. (b) Stork,
G.; La Clair, J. J. J. Am. Chem. Soc. 1996, 118, 247. (c) Lemanski, G.; Ziegler,
T. Tetrahedron 2000, 56, 563. (d) Ito, Y.; Ogawa, T. Angew. Chem., Int. Ed.
Engl. 1994, 33, 1765. (e) Lichtenthaler, F. W.; Schneider-Adams, T. J. Org.
Chem. 1994, 59, 6728. (f) Weingart, R.; Schmidt, R. R. Tetrahedron Lett.
2000, 41, 8753. (g) Hodosi, G.; Kova´c, P. J. Am. Chem. Soc. 1997, 119, 2335.
(2) Reviews: (a) Pozsgay, V. In Carbohydrates in Chemistry and Biology;
Ernst, B., Hart, G. W., Sinay¨, P., Eds.; Wiley-VCH: Weinheim, 2000; Vol.
1, p 319. (b) Barresi, F.; Hindsgaul, O. In Modern Methods in Carbohydrate
Synthesis; Khan, S. H., O’Neill, R. A., Eds.; Harwood Academic Publishers:
Amsterdam, The Netherlands, 1996; p 251.
(8) (a) Jouault, T.; Lepage, G.; Bernigauda, A.; Trinel, P. A.; Fradin, C.;
Wieruszeski, J.; Strecker, G.; Poulain, D. Infect. Immun. 1995, 63, 2378. (b)
Li, R. K.; Cutler, J. E. J. Biol. Chem. 1993, 268, 18293. (c) Fradin, C.; Jouault,
T.; Mallet, A.; Mallet, J. M.; Camus, D.; Sinay¨, P.; Poulain, D. J. Leukocyte
Biol. 1996, 60, 81. (d) Trinel, P.-A.; Plancke, Y.; Gerold, P.; Jouault, T.;
Delplace, F.; Schwarz, R. T.; Strecker, G.; Poulain, D. J. Biol. Chem. 1999,
274, 30520.
(9) While the present work was in progress the synthesis of a tetrameric
version of 2 was described: Nitz, M.; Purse, B. W.; Bundle, D. R. Org. Lett.
2000, 2, 2939.
(10) The choice of phenyl thioglycoside sulfoxides in the synthesis of 1
and of ethyl thioglycoside sulfoxides in that of 2 was purely arbitrary. In
general both function equally well in â-mannosylation reactions.3a
(11) Johansson, R.; Samuelsson, B. J. Chem. Soc., Perkin Trans. 1 1984,
2371.
(12) David, S. In PreparatiVe Carbohydrate Chemistry; Hanessian, S., Ed.;
Dekker: New York, 1997; p 69.
(3) (a) Crich, D.; Sun, S. Tetrahedron 1998, 54, 8321. (b) Crich, D.; Sun,
S. J. Am. Chem. Soc. 1997, 119, 11217. (c) Crich, D.; Dai, Z. Tetrahedron
1999, 55, 1569. (d) Crich, D.; Smith, M.; Yao, Q.; Picione, J. Synthesis 2001,
323. (e) Crich, D.; Mataka, J.; Sun, S.; Lam, K.-C.; Rheingold, A. R.; Wink,
D. J. J. Chem. Soc., Chem. Commun. 1998, 2763.
(4) (a) Kahne, D.; Walker, S.; Cheng, Y.; Engen, D. V. J. Am. Chem. Soc.
1989, 111, 6881.
(5) Matsuo, K.; Isogai, E.; Araki, Y. Carbohydr. Res. 2000, 328, 517.
(6) Shibata, N.; Ichikawa, T.; Tojo, M.; Takahashi, M.; Ito, N.; Okubo,
Y.; Suzuki, S. Arch. Biochem. Biophys. 1985, 243, 338.
(7) (a) Faille, C.; Michalski, J. C.; Strecker, G.; Mackenzie, D. W. R.;
Camus, D.; Poulain, D. Infect. Immun. 1990, 58, 3537. (b) Shibata, N.; Arai,
M.; Haga, E.; Kikuchi, T.; Najima, M.; Satoh, T.; Kobayashi, H.; Suzuki, S.
Infect. Immun. 1992, 60, 4100. (c) Tojo, M.; Shibata, N.; Kobayashi, M.;
Mikami, T.; Suzuki, M.; Suzuki, S. Clin. Chem. 1988, 34, 539. (d) Poulain,
D.; Faille, C.; Delaunoy, C.; Jacquinot, P. M.; Trinel, P. A.; Camus, D. Infect.
Immun. 1993, 61, 1164. (e) De Bernardis, F.; Boccanera, M.; Adriani, D.;
Spreghini, E.; Santoni, G.; Cassone, A. Infect. Immun. 1997, 65, 3399. (f)
Caesar-TonThat, T.-C.; Cutler, J. E. Infect. Immun. 1997, 65, 5354.
(13) The assignment of stereochemistry in all coupling reactions en route
to 1 and 2 was facilitated by our earlier observation3a of the characteristic
upfield chemical shift (δ 3.1-3.3) of the mannose H-5 resonance in 4,6-O-
benzylidene protected â-mannosides. The corresponding â-mannosides have
no such characteristic signal with all ring proton chemical shifts being normal.
The assignments were supported by the isolation of the minor R-anomer, which
was always lacking in the unusual upfield â-mannoside region. Further
confirmation was obtained, resolution permitting, by measurement of the
1
characteristic JCH anomeric coupling constants: Bock, K.; Pedersen, C. J.
Chem. Soc., Perkin Trans. 2 1974, 293.
(14) Stauch, T.; Greilich, U.; Schmidt, R. R. Liebigs 1995, 2101.
10.1021/ja015985e CCC: $20.00 © 2001 American Chemical Society
Published on Web 05/22/2001

Communications to the Editor
J. Am. Chem. Soc., Vol. 123, No. 24, 2001 5827
Scheme 1. Typical Formation of a â-1f4 Linkagea
Turning to octamer 2 we reasoned that it could be accessed by
a two-step iterative protocol involving â-selective coupling to
mannose O2, followed by regioselective deprotection of O2 in
the newly introduced residue, and iteration.
Thioglycoside 233c,10 was converted to the 2-O-PMB ether 24
and thence to the diastereomerically pure sulfoxide 25.3e Cyclo-
hexanol was arbitrarily chosen to cap the reducing end, and,
following activation of 25 with Tf2O in the presence of TTBP in
CH2Cl2 at -78 °C, was coupled up to give 26 in excellent yield
a Conditions: (a) (i) Tf2O, TTBP, -78 °C, (ii) 10, 81% â, 7% R; (b)
DDQ, 83%.
Scheme 2. Typical Formation of a â-1f3 Linkagea
a Conditions: (a) (i) Tf2O, TTBP, CH2Cl2, -78 °C, (ii) 12, 72% â,
8% R; (b) 80% HOAc, 70%; (c) Bu2SnO, BnBr, 87%.
and selectivity. Removal of the PMB group with DDQ then
provided alcohol 27. The first iteration provided disaccharide 28,
which in turn gave acceptor 29. Subsequent steps proceeded more
or less uneventfully with the yields and stereoselectivities
presented in Table 1, ultimately leading to the octasaccharide 40
and, after removal of the PMB group, the alcohol 41.
removal of the p-methoxybenzylidene group, in the presence of
the benzylidene group, was achieved in 70% yield by exposure
to 80% acetic acid at room temperature.15 Regioselective monobenz-
ylation of diol 14 was accomplished uneventfully, as for the
conversion of 9 to 10, by treatment first with Bu2SnO and then
with benzyl bromide giving 15 in 87% yield (Scheme 2). At this
stage each of the various protocols required for oligomer synthesis
had been successfully implemented with the result that completion
of the synthesis was a matter of careful iteration.
Accordingly, coupling of donor 4 with acceptor 15 gave the
tetramer 16 in 70% yield as a separable 7.8/1 â/R mixture and
removal of the PMB group provided the next acceptor 17 in 72%
yield. Coupling of 17 to the activated donor 3 afforded the
pentamer 18 as a 9.0/1 â/R mixture in 80% yield (Scheme 3).
Acidolysis of the p-methoxybenzylidene group of 18 gave the
diol 19 and recovered substrate in 50 and 39% yields, respectively
(Scheme 3). Prolonged exposure of 18 to the cleavage conditions
resulted in the onset of benzylidene cleavage; it was therefore
found expeditious to stop the reaction before completion and
recycle the substrate. Treatment of diol 19 with Bu2SnO and BnBr
in the usual manner afforded the acceptor 20 in 62% yield and
set the stage for the final coupling. This was accomplished in the
standard manner with the sulfoxide donor 213a to give the
hexasaccharide 22 in 70% yield as 9.0/1 â/R mixture. Final
deprotection of 22 was achieved by hydrogenolysis (3 atm) over
Pd/C giving 1, with a correct ESI MS, in 85% yield (Scheme 3).
The yields and selectivities for the first three linkages in 2 are
consistent with those reported previously from our laboratory for
the formation of simple â-mannosides.3a They do, however, fall
off at the level of introduction of the third and fourth units before
settling down to around 80% and 4-5:1, respectively, thereafter.
The early yields and selectivites are certainly superior to the those
in the earlier synthesis of the corresponding tetramer using the
alternative ulosyl bromide coupling.9 It seems apparent that longer
oligomers should be accessible by this methodology if yields and
selectivities of the type in the lower half of Table 1 can be
tolerated.
Scheme 3. Completion of the Synthesis of 1a
a Conditions: (a) (i) Tf2O, TTBP, CH2Cl2, -78 °C, (ii) 15, 62% â, 8% R; (b) DDQ, 72%; (c) (i) 3, Tf2O, TTBP, CH2Cl2, -78 °C, (ii) 17, 72% â,
8% R; (d) 80% HOAc, 50%; (e) Bu2SnO, BnBr, 62%; (f) (i) 21, Tf2O, TTBP, CH2Cl2, -78 °C, (ii) 20, 63% â, 7% R; (g) H2, Pd/C, MeOH, 85%.

5828 J. Am. Chem. Soc., Vol. 123, No. 24, 2001
Communications to the Editor
Table 1. â-1f2 Linkage Yields and Selectivities in the Synthesis
of 2
Table 2. Glycosidic Bond Torsion Angles in Tetrasaccharide 33a
â-glycoside R-glycoside
freed alcohol
(% yield)
acceptor
(% yield)
(% yield)
â/R ratio
C6H11OH
26-â (77)
28-â (94)
30-â (89)
32-â (77)
34-â (69)
36-â (68)
38-â (64)
40-â (64)
26-r (0)
28-r (0)
30-r (9)
32-r (20)
34-r (16)
36-r (15)
38-r (13)
40-r (14)
â only
â only
9.9/1
3.9/1
4.3/1
4.5/1
4.9/1
4.5/1
27 (85)
29 (97)
31 (91)
33 (85)
35 (80)
37 (85)
39 (81)
41 (71)
27
29
31
33
35
37
39
Cyc-a
a-b
b-c
c-d
æ (H1-C1-O-C2′)
ψ (C1-O-C2′-H2′)
43.7
-15.3
44.4
10.7
18.3
-8.7
31.3
13.6
a Where a and d are the sugars at the reducing and nonreducing ends
of the chain, respectively.
mannans and the structure presented is protected, the parallels
are close. Structure 33 is a collapsed irregular helix with
approximately three residues per turn with the calculations
predicting between 2 and 4.
Although all linkages are relatively hindered, it is at the stage
of introduction of the fourth one that severe hindrance from distant
residues first comes into play.17 All subsequent glycosidic bond
forming reactions will suffer from the same type of interaction if
the collapsed helical structure is maintained, as is expected. The
falloff in anomeric selectivity with chain length in the synthesis
of 1 (Schemes 1-3) is much less marked, consistent with the
more open, less hindered nature of â-1,3 and â-1,4 linkages.
Table 2 shows the æ and ψ torsion angles in 33, from which
it is apparent that all linkages are nonideal and that the helix is
disordered. The partially eclipsed bonds presumably arise from
minimization of steric interactions between noncontiguous resi-
dues, which are obviously more important in the protected
mannan. The variation in torsion angles accounts for the irregular-
ity of the helical structure.
Deprotection of 41 was achieved by hydrogenolysis (3 atm)
over Pd/C in methanol for 3 days giving the pure octamer 2, with
a correct ESI MS, in 89% yield. Although we have not yet
conducted any detailed conformational analysis of 2, its 1H NMR
spectrum at 500 MHz exhibits 7 distinct anomeric protons, which
serves to illustrate the irregular nature of its predicted crumpled
helical conformation, aside from any obvious fraying at the ends.
Figure 1. X-ray crystallographic structure of 33 with all nonessential
groups removed for clarity and rings 1 and 4 emphasized.
Tetrasaccharide 33 gave suitable crystals for X-ray analysis
(Figure 1). Examination of this structure provides a possible
explanation for the fall off in yield and selectivity midway through
the synthesis. The tetramer approximates to a compact helical
structure with the fourth residue obliquely above the first as judged
from the distance (5.68 Å) between the centroids of planes defined
by C2-C3-C5-O5 in the first and fourth residues as well as
the angle (34.1°) between those planes. Early computational
studies predicted that 1,2-â-mannans would have crumpled
conformations to alleviate steric interactions between remote
residues.16 Although the calculations were for fully deprotected
Acknowledgment. We thank the NIH (GM 57335) for support of
this work.
Supporting Information Available: 1H and 13C NMR for 1, 2, 8-20,
22, 26-37, 39, and 41 (PDF), the complete structure of 33 (PDB), and
tables of bond angles and lengths for 33 (PDF). This material is available
free of charge via the Internet at http://pubs.acs.org.
JA015985E
(16) Rees, D. A.; Scott, W. E. J. Chem. Soc. B 1971, 469.
(17) It is noteworthy that the introduction of the fourth residue proved the
most difficult in Bundle’s synthesis of the tetramer.9
(15) Smith, M.; Rammler, D. H.; Goldberg, I. H.; Khorana, H. G. J. Am.
Chem. Soc. 1962, 84, 430.