Oligofuranosides containing conformationally restricted residues: Synthesis and conformational analysis

Article abstract of DOI:10.1021/jo011127p

The synthesis of a panel of arabinofuranosyl oligosaccharide analogues (5-13) in which one ring is locked into either the E₃ or ^oE conformation is described. The E₃-locked scaffolds 15 and 16 required for the synthesis of 5-10 were prepared in one step from known 1,5-anhydroalditols. A number of routes were explored for the preparation of the ^oE-locked monosaccharide derivative 17 needed for the preparation of 11-13. The successful synthesis of 17 was achieved in 17 steps from D-arabinose. Subsequent analysis of 5-13 by ¹H NMR spectroscopy demonstrated that the locked residue does not exert any detectable influence upon the conformers populated by adjacent conformationally unrestricted furanose rings.

Full text of DOI:10.1021/jo011127p

4150
J . Org. Chem. 2002, 67, 4150-4164
Oligofu r a n osid es Con ta in in g Con for m a tion a lly Restr icted
Resid u es: Syn th esis a n d Con for m a tion a l An a lysis
J ustin B. Houseknecht and Todd L. Lowary*
Department of Chemistry, The Ohio State University, Columbus, Ohio 43210
lowary.2@osu.edu
Received December 5, 2001
The synthesis of a panel of arabinofuranosyl oligosaccharide analogues (5-13) in which one ring
is locked into either the E3 or OE conformation is described. The E3-locked scaffolds 15 and 16
required for the synthesis of 5-10 were prepared in one step from known 1,5-anhydroalditols. A
number of routes were explored for the preparation of the OE-locked monosaccharide derivative 17
needed for the preparation of 11-13. The successful synthesis of 17 was achieved in 17 steps from
1
D-arabinose. Subsequent analysis of 5-13 by H NMR spectroscopy demonstrated that the locked
residue does not exert any detectable influence upon the conformers populated by adjacent
conformationally unrestricted furanose rings.
In tr od u ction
Furanose rings are found throughout nature as com-
ponents of nucleic acids,1 polysaccharides,2 and other
natural products.3 An important structural characteristic
of these and all five-membered rings is their inherent
flexibility. Conformationally unrestricted furanose rings
can adopt a number of envelope (E) and twist (T)
conformers, which can be conveniently depicted on the
pseudorotational wheel (Figure 1),4 where P is the
pseudorotational phase angle.
The standard model used to assess the conformation
of a furanose ring in solution assumes an equilibrium
mixture of two conformers, termed north (N) and south
(S) on the basis of the hemisphere of the wheel in which
they are located.4 For a given ring, determining these
conformers and their populations can be done through
F igu r e 1. Pseudorotational wheel for a D-aldofuranose ring.
3
the measurement of all intracyclic ring J H,H values and
entities. These structural differences in turn modulate
many critical biological recognition events involving DNA
and RNA.1
Over the past decade, there has been increasing
interest in the synthesis of nucleic acid analogues in
which the sugar residues are locked into a single con-
former.7 Research in this area has been motivated largely
by the desire to identify antisense oligonucleotides that
bind tightly to RNA, thus blocking protein translation.8
subsequent analysis of these data with the program
PSEUROT.5 The identities of the conformers populated
at equilibrium are influenced by both the number and
orientation of the ring substituents, as well as stereo-
electronic effects (e.g., the anomeric effect and gauche
effects) involving the ring oxygen.6 It is well-known that
the conformational preferences of the furanose ring
significantly influence the biological activity of molecules
containing them. For example, in nucleic acids, the
A-form and B-form double helix families are differenti-
ated by the conformation of the furanose rings in these
(6) (a) Thibaudeau, C.; Chattopadhyaya, J . Stereoelectronic Effects
in Nucleosides and Nucleotides and their Structural Implications;
Uppsala University Press: Uppsala, Sweden, 1999. (b) Thibaudeau,
C.; Chattopadhyaya, J . Nucleosides Nucleotides 1997, 16, 523. (c)
Plavec, J .; Thibaudeau, C.; Chattopadhyaya, J . Pure Appl. Chem. 1996,
68, 2137. (d) Thibaudeau, C.; Plavec, J .; Chattopadhyaya, J . J . Am.
Chem. Soc. 1994, 116, 8033. (e) Thibaudeau, C.; Plavec, J .; Garg, N.;
Papchikhin, A.; Chattopadhyaya, J . J . Am. Chem. Soc. 1994, 116, 4038.
(f) Plavec, J .; Thibaudeau, C.; Chattopadhyaya, J . J . Am. Chem. Soc.
1994, 116, 6558. (g) Plavec, J .; Tong, W.; Chattopadhyaya, J . J . Am.
Chem. Soc. 1993, 115, 9734.
(7) For recent reviews see: (a) Kværnø, L.; Wengel, J . Chem.
Commun. 2001, 1419. (b) Egli, M. Antisense Nucleic Acid Drug Dev.
1998, 8, 123. (c) Kool, E. T. Chem. Rev. 1997, 97, 1473. (e) Herdewijn,
P. Liebigs Ann. 1996, 1337. Meldgaard, M.; Wengel, J . J . Chem. Soc.,
Perkin Trans. 1 2000, 3539.
(1) (a) Saenger, W. Principles of Nucleic Acid Structure; Spring-
Verlag: Berlin, 1988. (b) Birnbaum, G. I. Shugar, D. Biologically Active
Nucleosides and Nucleotides: Conformational Features and Interaction
with Enzymes. In Topics In Nucleic Acid Structure; Neidle, S., Ed.;
MacMillan Press: London, 1987; p 1.
(2) (a) de Lederkremer, R. M.; Colli, W. Glycobiology 1995, 5, 547.
(b) Brennan, P. J .; Nikaido, H. Annu. Rev. Biochem. 1995, 64, 29.
(3) Selected examples: (a) Macko, V.; Acklin, E.; Hildenbrand, C.;
Weibel, F.; Arigoni, D. Experientia 1983, 39, 343. (b) Komatsu, K.;
Shigemori, H.; Kobayashi, J . J . Org. Chem. 2001, 66, 6189.
(4) Altona, C.; Sundaralingam, M. J . Am. Chem. Soc. 1972, 94, 8205.
(5) (a) PSEUROT 6.2, Gorlaeus Laboratories, University of Leiden,
The Netherlands. (b) de Leeuw, F. A. A. M.; Altona, C. J . Comput.
Chem. 1983, 4, 428. (c) Altona, C. Recl. Trav. Chem. Pays-Bas 1982,
101, 413.
(8) DeMesmaeker, A.; Ha¨ner, R.; Martin, P.; Moser, H. E. Acc. Chem.
Res. 1995, 28, 366.
10.1021/jo011127p CCC: $22.00 © 2002 American Chemical Society
Published on Web 05/15/2002

Oligofuranoside Synthesis and Conformational Analysis
J . Org. Chem., Vol. 67, No. 12, 2002 4151
F igu r e 2. Conformers populated by oligosaccharide fragments of mycobacterial arabinogalactan and lipoarabinomannan.
The rationale for the preparation of oligonucleotides with
conformationally restricted furanose residues is that
these compounds will be preorganized to bind to comple-
mentary sequences of RNA and hence the entropic
penalty associated with duplex formation will be dimin-
ished. Many classes of such analogues have now been
synthesized, and some have significantly greater affinity
for RNA than the native oligonucleotide parent struc-
ture.7 A more recent motivation for work in this area has
been the realization that conformationally locking or
biasing a nucleotide often substantially influences its
recognition by various processing enzymes.9
Restricting the conformation of a furanose ring is
generally achieved by one of two methods. The first is
through the attachment of covalent tethers that lock the
ring into a single conformation.7c,9c The second is by
replacement of the hydroxyl groups at C-2 or C-3 with
other functionalities. When the substitution involves the
replacement of the OH with a strongly electronegative
group (e.g., fluorine), conformers in which these substit-
uents are oriented pseudoaxially are favored due to an
attractive gauche interaction with the ring oxygen.6
Recently, we have probed the conformation of the
furanose rings in oligosaccharides containing arabino-
furanose residues.10 These investigations were carried out
by applying methods developed for the conformational
analysis of nucleosides to the oligosaccharides of interest.
In this earlier study, we described the ring conformers
populated by methyl R-D-arabinofuranoside (1) and oli-
gosaccharides 2-4 (Figure 2). These glycans are frag-
ments of two polysaccharides, arabinogalactan and li-
poarabinomannan, which are present in the cell wall of
Mycobacterium tuberculosis and other mycobacteria.11 In
other investigations, we12 and others13 have shown that
small oligosaccharides, including 2-4, are substrates for
the arabinosyltransferases that are involved in the
biosynthesis of mycobacterial arabinogalactan and li-
poarabinomannan. Our conformational investigations are
directed at understanding the structural motifs present
in these oligosaccharides, as we anticipate that an
appreciation of the conformation of these glycans will
facilitate the development of potential inhibitors of
mycobacterial arabinosyltransferases. Such compounds
are of current interest as anti-mycobacterial agents.14
To probe the conformational equilibrium of each ring
3
in 1-4, we measured the J H,H values between the ring
protons and carried out PSEUROT analyses. A summary
of these results is discussed below and presented in
Figure 2.15 In aqueous solution, monosaccharide 1 exists
as an equilibrium of an approximately 2:1 ratio of E4 (N)
2
and T3 (S) conformers. For disaccharide 2, the identities
of the conformers and their populations for each ring are
essentially the same as those of the monosaccharide. For
disaccharide 3 and trisaccharide 4, the nonreducing end
monosaccharide residues (B and/or C) also adopt confor-
mations similar to those of the monosaccharide. However,
the conformational equilibrium of the reducing end ring
(residue A) in 3 and 4 is altered relative to that of 1; the
identify of the northern conformer changes to one inter-
mediate between OT4 and OE, while the southern con-
2
former remains at T3.
(11) Crick, D. C.; Mahapatra, S.; Brennan, P. J . Glycobiology 2001,
11, 107R.
(12) Ayers, J . D.; Lowary, T. L.; Morehouse, C. B.; Besra, G. S.
Bioorg. Med. Chem. Lett. 1998, 8, 437.
(13) Lee, R. E.; Brennan, P. J .; Besra, G. S. Glycobiology 1997, 7,
1121.
(9) (a) Ford, H., J r.; Dai, F.; Mu, L.; Siddiqui, M. A.; Nicklaus, M.
C.; Anderson, L. Marquez, V. E.; Barchi, J . J ., J r. Biochemistry 2000,
39, 2581. (b) Wang, P.; Brank, A. S.; Banavali, N. K.; Nicklaus, M. C.;
Marquez, V. E.; Christman, J . K.; MacKerell, A. D., J r. J . Am. Chem.
Soc. 2000, 122, 12422. (c) Mu, L.; Sarafianos, S. G.; Nicklaus, M. C.;
Russ, P.; Siddiqui, M. A.; Ford, H., J r; Mitsuya, H.; Le, R.; Kodama,
E.; Meier, C.; Knispel, T.; Anderson, L.; Barchi, J . J ., J r.; Marquez, V.
E. Biochemistry 2000, 39, 11205. (d) Kim, H. S.; Ravi, R. G.; Marquez,
V. E.; Maddileti, S.; Wihlborg, A.-K.; Erlinge, D.; Malmsjo¨, M.; Boyer,
J . L.; Harden, T. K.; J acobsen, K. A. J . Med. Chem. 2002, 45, 208.
(10) D’Souza, F. W.; Ayers, J . D.; McCarren, P. R.; Lowary, T. L. J .
Am. Chem. Soc. 2000, 122, 1251.
(14) (a) Reynolds, R. C.; Bansal, N.; Rose, J .; Friedrich, J .; Suling,
W. J .; Maddry, J . A. Carbohydr. Res. 1999, 317, 164. (b) Maddry, J .
A.; Suling, W. J .; Reynolds, R. C. Res. Microbiol. 1996, 147, 106. (c)
Bouix, C.; Bisseret, P.; Eustache, J . Tetrahedron Lett. 1998, 39, 825.
(d) Maddry, J . A.; Bansal, N.; Bermudez, L. E.; Comber, R. N.; Orme,
I. M.; Suling, W. J .; Wilson, L. N.; Reynolds, R. C. Bioorg. Med. Chem.
Lett. 1998, 8, 237. (e) Ha¨usler, H.; Kawakami, R. P.; Mlaker, E.; Severn,
W. B.; Stu¨tz, A. E. Bioorg. Med. Chem. Lett. 2001, 11, 1679.
(15) These values differ from those originally published by us in ref
10. The values provided in Figure
2 reflect the populations as
calculated by PSEUROT using the most updated generalized Karplus
equation (GKE) available. Our earlier studies employed an older GKE.

4152 J . Org. Chem., Vol. 67, No. 12, 2002
Houseknecht and Lowary
F igu r e 3. Synthetic targets 5-14. The rings have been lettered to facilitate comparison with 2-4.
With an understanding of the low-energy conformers
available to each ring in 2-4, we next became interested
in synthesizing analogues of these glycans in which one
ring was locked into a single conformer that approxi-
mated the N and S minimum-energy structures. On the
basis of the previous successes with nucleic acids and
nucleosides described above, it was our hope that this
modification would provide substrates with improved
affinity for the arabinosyltransferases by minimizing the
entropic penalty necessary for binding.16 We also were
interested in understanding what effect, if any, biasing
one of the rings had on the conformation of adjacent
residues. Previous studies on oligonucleotides containing
nucleoside residues with conformationally locked or
biased furanose rings have shown that these modified
residues do influence the conformation of unrestricted
sugar residues of adjacent nucleosides in both single- and
double-stranded systems.17
In this paper, we describe the synthesis of analogues
of 2-4 in which the reducing end residue (residue A) has
been replaced with a monosaccharide locked into a
conformation that approximates either the N or S con-
former present in the parent oligosaccharides. We have
also carried out PSEUROT analyses to determine whether
this modification of residue A influences the conforma-
tions of rings B and C relative to those of 2-4. The
targets we chose to synthesize are shown in Figure 3. In
5-10 we predicted that residue A would be locked in the
2T3 (S) conformation, while in oligosaccharides 11-14 this
ring is predicted to adopt an OE (N) geometry.
Resu lts a n d Discu ssion
Com p u ta tion a l In vestiga tion s of Lock ed Sca f-
fold s. Before proceeding with the synthesis of the targets,
we carried out a series of molecular mechanics and ab
initio calculations to determine whether the conforma-
(16) The conformation of the ring bound by these arabinosyltrans-
ferases is unknown. Clearly, locking the ring into a conformation other
than the one bound by the enzyme will likely prevent recognition of
the substrate. Another important, but fundamentally different, issue
is conformational flexibility about the glycosidic linkages. Previous
efforts to enhance protein-carbohydrate interaction through the
conformational restriction of glycosidic linkages in oligosaccharides
have not resulted in significant increases in affinity: (a) Bundle, D.
R.; Alibe´s, R.; Nilar, S.; Otter, A.; Warwas, M.; Zhang, P. J . Am. Chem.
Soc. 1998, 120, 5317. (b) Navarre, N.; Amiot, N.; van Oijen, A.; Imberty,
A.; Poveda, A.; J imenez-Barbero J .; Cooper, A.; Nutley, M. A.; Boons,
G. J . Chem.sEur. J . 1999, 5, 2281.
(17) (a) Petersen, M.; Nielsen, C. B.; Nielsen, K. E.; J ensen, G. A.;
Bondensgaard, K.; Singh, S. K.; Rajwanshi, V. K.; Koshkin, A. A.; Dahl,
B. M.; Wengel, J . J acobsen, J . P. J . Mol. Recognit. 2000, 13, 44. (b)
Nielsen, K. E.; Singh, S. K.; Wengel, J . J acobsen, J . P. Bioconjugate
Chem. 2000, 11, 228. (c) J orgensen, L. B.; Nielsen, P.; Wengel, J .
J acobsen, J . P. J . Biomol. Struct. Dyn. 2000, 18, 45. (d) Nielsen, C. B.;
Singh, S. K.; Wengel, J . J acobsen, J . P. J . Biomol. Struct. Dyn. 1999,
17, 175. (e) Ikeda, H.; Fernandez, R.; Wilk, A.; Barchi, J . J ., J r.; Huang,
X.; Marquez, V. E. Nucleic Acids Res. 1998, 26, 2237. (f) Pradeepkumar,
P. I.; Zamaratski, E.; Fo¨ldesi, A.; Chattopadhyaya, J . J . Chem. Soc.,
Perkin Trans. 2 2001, 402.

Oligofuranoside Synthesis and Conformational Analysis
J . Org. Chem., Vol. 67, No. 12, 2002 4153
Ch a r t 1
F igu r e 4. B3LYP/6-31+G**//HF/6-31G* global minima of
15-17.
Sch em e 1a
tionally locked monosaccharides do adopt the structures
predicted above. To this end, a Monte Carlo search of the
locked “scaffolds” 15-17 (Chart 1) was carried out, and
the low-energy structures were then subjected to higher
level calculations. As expected, the Monte Carlo searches
identified only a few conformers for each molecule. All
of the conformers within 4 kcal/mol of the global minima
were then optimized at the HF/6-31G* level of theory,
and then B3LYP/6-31+G** single-point energies were
determined for each resulting conformer. The furan ring
in 15 and 16 is locked in the E3 ring conformation, which
a
Reagents and conditions: (a) PPh3, I2, pyridine, rt; (b) Ac2O,
pyridine, rt; (c) NaOCH3, CH3OH, rt, 63% (three steps); (d) PPh3,
I2, pyridine, rt; (e) Ac2O, pyridine, rt; (f) NaOCH3, CH3OH, rt, 83%
(three steps).
2
is very similar to the T3 we predicted. The pyran ring
1
in 15 and 16 is locked in the C4 conformation (see Chart
1 for atom numbers). The furanose ring in the N-locked
scaffold 17 was found to exist solely in the predicted OE
ring conformation. The B3LYP/6-31+G**//HF/6-31G*
global minima of 15-17 are shown in Figure 4. The other
low-energy conformers of 15-17 found by the protocol
described above differed from the global minima only in
the orientation about O-H or exocyclic C-C bonds. The
conformations of the rings were unchanged in comparison
to those of the lowest energy structures.
In a previous study,18 we explored the conformation of
the 2,3-anhydrofuranoside moiety of disaccharide 14 by
ab initio and density functional theory methods. As
expected19 the furanose ring in this anhydrosugar adopts
the OE conformation.
be readily synthesized from building blocks 15-19 (Chart
1). Disaccharide 14 has previously been reported.18
Syn th esis of S-Lock ed Mon osa cch a r id e Sca ffold s
15 a n d 16. As shown in Chart 1, the two S-locked
scaffolds 15 and 16 are 1,5:3,6-dianhydroalditol deriva-
tives. The formation of 3,6-anhydrosugars from monosac-
charides via a two-step process involving installation of
a leaving group at C-6 followed by treatment with base
is well-known.20 We therefore postulated that it would
be possible to synthesize 15 and 16 in one step from the
corresponding 1,5-anhydroalditols, through the in situ
generation of an oxyphosphonium leaving group at C-6.
Indeed, the reaction of 2121 with triphenylphosphine and
iodine in pyridine at room temperature proceeded as
predicted (Scheme 1), yielding 1,5:3,6-dianhydro-D-ga-
lactitol (15). Due to the high polarity of 15, the most
efficient method of purification was to acetylate the
Syn th esis of Con for m a tion a lly Restr icted Sca f-
fold s. We envisioned that oligosaccharides 5-13 could
(18) Callam, C. S.; Gadikota, R. R.; Lowary, T. L. J . Org. Chem.
2001, 66, 4549.
(19) Marquez, V. E.; Siddiqui, M. A.; Ezzitouni, A.; Russ, P.; Wang,
J .; Wagner, R. W.; Matteucci, M. D. J . Med. Chem. 1996, 39, 3739
(20) Lewis, B. A.; Smith, F.; Stephen, A. M. Methods Carbohydr.
Chem. 1963, 2, 172.
(21) Bennek, J . A.; Gray, G. R. J . Org. Chem. 1987, 52, 892.

4154 J . Org. Chem., Vol. 67, No. 12, 2002
Houseknecht and Lowary
Sch em e 2a
F igu r e 5. Retrosynthetic analysis of 17.
hydroxyl groups, purify the resulting diacetate by chro-
matography, and then remove the acetate esters with
sodium methoxide. This three-step process afforded 15
in 63% yield from 21.
We initially had hoped that it would be possible to
access 16 from 15 by inversion of the axially oriented
alcohol at C-2 (see Chart 1 for atom numbers). Unfortu-
nately, our attempts to invert this stereocenter by several
variations of the Mitsunobu reaction failed. In all cases,
the rate of reaction was very slow and the regioselectivity
poor. Presumably, this lack of reactivity is due to the very
rigid nature of the molecule, which hinders the formation
of the transition state in the substitution reaction. We
therefore synthesized 16 by treatment of 1,5-anhydro-D-
talitol (22)22 with triphenylphosphine and iodine as was
done for the preparation of 15. Dianhydroalditol 16 was
a
Reagents and conditions: (a) OsO4, NMO, acetone/water (3:
1
1), 0 °C f rt, 85%; (b) m-CPBA, NaHCO3, CH2Cl2, rt, 33%; (c)
PCC, NaOAc, CH2Cl2, rt; (d) CH2dCHMgBr, THF, 0 °C, 45% (from
26).
obtained in 83% yield from 22. For both 15 and 16, H
NMR spectral data were consistent with these structures
adopting the conformations predicted by our computa-
tional investigations.
Syn th esis of a n N-Lock ed Mon osa cch a r id e Resi-
d u e. Our synthesis of 17 was considerably more involved
than the preparation of 15 and 16. The initial route we
designed was modeled after a previous synthesis of
nucleoside 20 (Chart 1).23 However, our desire to have
an R-methyl glycoside instead of a â-nucleoside neces-
sitated modification of the published route. The key
aspects of the synthesis are the addition of a hydroxy-
methyl group to the top face of the ring at C-3, conversion
of the hydroxyl group of this moiety into a leaving group,
and then displacement by O-2 (Figure 5). An alternate
strategy, involving displacement at C-2, was not pursued
because we anticipated elimination would be a major side
reaction in the displacement.
The primary challenge was the installation of the
hydroxymethyl group at C-3 with the correct stereochem-
istry. Our first attempt (Scheme 2) toward 17 started
with alkene 23.24 We anticipated that dihydroxylation of
this olefin would proceed from the bottom face of the ring
due to the steric demands of the isopropylidene protecting
group. However, although the reaction of 23 with OsO4
and N-methylmorpholine oxide (NMO) did afford a single
diol product in 85% yield, the structure was shown to be
24, indicating that the dihydroxylation had occurred cis
F igu r e 6. NOEs observed in 24, 25, and 27.
to the isopropylidene moiety. Determination of the struc-
ture of 24 was done through the measurement of the
NOEs involving the newly formed hydroxymethyl group
hydrogens. The presence (Figure 6) of an NOE between
the H6 atoms and H2, in addition to the absence of an
NOE between either H6 and the H5 atoms confirmed the
stereochemistry at C-3 in 24. Dihydroxylations with AD
mix R and AD mix â were attempted, but these reactions
also gave 24 as the only product. We also explored the
epoxidation of 23 with m-chloroperoxybenzoic acid as a
(22) Estevez, J . C.; Fairbanks, A. J .; Fleet, G. W. J . Tetrahedron
1998, 54, 13591.
(23) Christensen, N. K.; Petersen, M.; Nielsen, P.; J acobsen, J . P.;
Olsen, C. E.; Wengel, J . J . Am. Chem. Soc. 1998, 120, 5458.
(24) Prepared from 26 by oxidation and subsequent Wittig reaction
with CH2dPPh3. The NMR spectrum of the product was identical to
that previously described for the L-enantiomer: Doboszewski, B.;
Herdewijn, P. Tetrahedron 1996, 52, 1651.

Oligofuranoside Synthesis and Conformational Analysis
J . Org. Chem., Vol. 67, No. 12, 2002 4155
Sch em e 3a
Sch em e 4a
a
Reagents and conditions: (a) PCC, CH2Cl2, NaOAc, rt, or
TPAP, NMO, rt; (b) CH3PPh3Br, n-BuLi, THF, -78 °C f rt.
way to install the C-3 stereocenter. However, this reac-
tion proceeded in low yield and afforded only epoxide 25,
with the incorrect C-3 stereochemistry (see Figure 6 for
the NOEs).
On the basis of these results, we predicted that
oxidation of alcohol 2625 to the corresponding ketone
followed by addition of a vinyl group would provide a
product with the desired stereochemistry at C-3. Unfor-
tunately, reaction of vinylmagnesium bromide with the
ketone obtained upon oxidation of 26 produced only 27
in modest yield (see Figure 6 for the observed NOEs).
Our inability to set the C-3 stereocenter with the correct
stereochemistry starting from 23 forced us to explore
other routes.
We next investigated the use of ortho ester 28 (Scheme
3), which can be conveniently prepared in four steps from
D-arabinose.26 We postulated that conversion of 28 into
alkene 30 would provide a substrate that would neces-
sarily undergo dihydroxylation from the bottom face of
the ring. Although oxidation of 28 with either pyridinium
chlorochromate or TPAP/NMO was successful, the prod-
uct was isolated as the hydrate, and all attempts to
dehydrate this molecule to ketone 29 failed. We postulate
that the reluctance of this hydrate to lose water to
provide the corresponding ketone is due to the rigidity
of the molecule, which inhibits the necessary flattening
of the furanose ring in the ketone. As would be expected,
the subsequent conversion of this hydrate to 30, upon
reaction with methylenetriphenylphosphorane, produced
only trace amounts of the desired product. This route was
therefore abandoned.
a
Reagents and conditions: (a) NaOBn, BnOH, 120 °C, 83%; (b)
(t-Bu)2SiCl2, pyridine, CH2Cl2, -78 °C f 0 °C, 74%; (c) H2,
Pd(OH)2, CH3OH, rt, 94%; (d) PCC, NaOAc, CH2Cl2, rt; (e)
Cp2Ti(CH3)2), THF, 70 °C, 71% (from 34); OsO4, NMO, acetone/
water (3:1), 0 °C f rt.
conditions were unsuccessful due to the instability of the
di-tert-butylsilyl group to the basic conditions of the
reaction.
With an efficient route to 35 in place, the key dihy-
droxylation reaction could be attempted. Upon reaction
of olefin 35 with osmium tetroxide and NMO, only trace
amounts of the desired diol 36 were produced. Although
the product with the correct stereochemistry was formed,
the rates of conversion were exceedingly slow (3-10
days), and only low yields (5-15%) of the diol could be
isolated. The same results were obtained when the
reaction was carried out using potassium permangenate
or a stoichiometric amount of osmium tetroxide. Fortu-
nately, the ruthenium trichloride and sodium periodate
oxidation system reported by Hegedus and co-workers29
could be used to efficiently convert olefin 35 into 36 in
reasonable yield. Purification of the diol was complicated
by its tendency to sublime, and therefore, following
workup of the dihyroxylation reaction, the crude product
was immediately treated with toluenesulfonyl chloride
and pyridine, yielding 37 in 65% overall yield from 35
(Scheme 5).
Our third approach to 17 involved the use of the silyl
acetal 35 (Scheme 4) as a key intermediate. The prepara-
tion of 35 proceeded smoothly from epoxide 31.27 Opening
of the oxirane ring was achieved in 83% yield by heating
31 with sodium in benzyl alcohol at 120 °C. The diol in
32 was then protected as a di-tert-butylsilyl acetal,
providing 33 in 74% yield. Subsequent hydrogenolysis of
the benzyl ether (H2, Pd(OH)2) proceeded in 94% yield,
affording 34. Oxidation of alcohol 34 using buffered
pyridinium chlorochromate provided the corresponding
ketone, which was converted immediately to 35 upon
treatment with the Petasis reagent28 at 70 °C. The
product was produced in 74% overall yield from 34. Our
initial attempts to olefinate this ketone under Wittig
Having successfully synthesized a molecule with the
correct stereochemistry at C-3 and a leaving group at C-6,
all that remained was the formation of the oxetane ring.
This task required the preparation of a substrate in
which both OH-3 and OH-5 were protected, and we
(25) Va´zquez-Tato, M. P.; Seijas, J . A.; Fleet, G. W. J .; Mathews, C.
J .; Hemmings, P. R.; Brown, D. Tetrahedron 1995, 51, 959.
(26) Kochetkov, N. K.; Khorlin, A. Y.; Bochkov, A. F.; Yazlovetskii,
I. G. Isv. Akad. Nauk. SSSR, Ser. Khim. 1966, 11, 2030.
(27) Martin, M. G.; Ganem, B.; Rasmussen, J . R. Carbohydr. Res.
1983, 123, 332.
(28) Petasis, N. A.; Bzowej, E. I. J . Am. Chem. Soc. 1990, 112, 6392.
(29) Hegedus, L. S.; Geisler, L. J . Org. Chem. 2000, 65, 4200.

4156 J . Org. Chem., Vol. 67, No. 12, 2002
Houseknecht and Lowary
Sch em e 5a
a
Reagents and conditions: (a) RuCl3‚H2O, NaIO4, EtOAc, CH3CN, H2O, 0 °C; (b) TsCl, DABCO, CH2Cl2, 0 °C, 65% (from 35); (c) HF,
pyridine, THF, rt, 85%; (d) (Br(i-Pr)2Si)2O, pyridine, 0 °C f rt; (e) TBDMSCl, DMF, imidazole, 0 °C, 86%; (f) DBU, THF, 0 °C, 60%.
Sch em e 6a
a
Reagents and conditions: (a) RuCl3‚H2O, NaIO4, EtOAc, CH3CN, H2O, 0 °C; (b) PhCH(OCH3)2, p-TsOH, CH2Cl2, rt f 40 °C, 60%
(from 35); (c) HF, pyridine, THF, rt, 86%; (d) TBDMSCl, DMF, imidazole, 0 °C, 87%; (e) NBS, BaCO3, CCl4, reflux; (f) K2CO3, DMF, rt,
47% (from 44); (g) n-Bu4NF, THF, 0 °C, then NaOCH3, CH3OH, rt, 91%.
initially explored protecting the tertiary hydroxyl group
in 37. However, attempts to protect OH-3 with either an
acetate or a benzoate group failed, undoubtedly due to
the hindered nature of this alcohol. We next explored the
possibility of protecting both OH-3 and OH-5 as a
siloxane. To this end, the di-tert-butylsilyl acetal was
cleaved in 85% yield with HF in pyridine, and the
resulting unstable triol 38 was reacted with 1,3-dibromo-
1,1,3,3-tetraisopropyldisiloxane30 in pyridine. Unfortu-
nately, none of the desired product 39 could be isolated.
Faced with this failure, and given the difficulty in
protecting the 3-hydroxyl group in 37, we explored an
alternate approach. We postulated that if OH-5 in 38
were protected, a large nonnucleophilic base might
differentiate between the OH-2 and OH-3 in favor of the
former, thus allowing the preparation of the desired
oxetane. To investigate this possibility, the primary
hydroxyl group in 38 was selectively protected, providing
40 in 86% yield. Several reaction conditions were screened
to effect oxetane formation (DBU in THF or DMF, sodium
tert-butoxide in THF, NaH in DMF), but all provided
exclusively epoxide 41. These results forced us to again
reevaluate the route to 17.
The final, and successful, route to 17 is shown in
Scheme 6. Alkene 35 was treated with ruthenium trichlo-
ride and sodium periodate, and the resulting diol was
then protected as a benzylidene acetal. Acetal 42 was
produced in 60% overall yield from 35 as a 1:1 mixture
of diastereomers. Also recovered, in 12% yield, was olefin
34 due to incomplete reaction during the dihydroxylation
step. The silyl acetal in 42 was then cleaved by HF in
pyridine, affording diol 43 in 86% yield. Selective protec-
tion of the primary hydroxyl group in 43 was achieved
upon reaction with tert-butylchlorodimethylsilane and
imidazole, providing 44 (87% yield). The benzylidene
acetal was then opened via reaction with N-bromosuc-
cinimide and barium carbonate in refluxing carbon
tetrachloride,31 which produced bromide 45. In this
manner, a protecting group on O-3 and a leaving group
at C-6 were installed in a single step. This task had
eluded us in our previous approaches. Upon its formation,
(30) Otmar, M.; Rosenberg, I.; Masoj´ıdkova´, M.; Holy, A. Collect.
Czech. Chem. Commun. 1993, 58, 2159.
(31) Hanessian, S. Carbohydr. Res. 1966, 1, 86.

Oligofuranoside Synthesis and Conformational Analysis
J . Org. Chem., Vol. 67, No. 12, 2002 4157
Sch em e 7a
Sch em e 8a
a
Reagents and conditions: (a) N-iodosuccinimide, silver triflate,
CH2Cl2, 0 °C; (b) NaOCH3, CH3OH, rt, 68% (two steps).
a
Reagents and conditions: (a) N-iodosuccinimide, silver triflate,
CH2Cl2, 0 °C; (b) NaOCH3, CH3OH, rt; (c) Ac2O, pyridine, rt; (d)
NaOCH3, CH3OH, rt, 20% 5 and 20% 6 (four steps).
spectra of products arising from exhaustive acetylation
of 5 and 6. In the case of 5, peracetylation clearly resulted
in a downfield shift of H-2; a similar shift of H-4 was
seen in the peracetate derivative of 6.
compound 45 was eluted through a short plug of silica
gel and immediately treated with base. The use of NaH
and DBU proved unsuccessful as rapid debenzoylation
occurred, leading to epoxide formation. However, potas-
sium carbonate in DMF was found to efficiently promote
formation of the desired 2,3-oxetane. Under these condi-
tions, no epoxide byproducts were observed during the
closing of the four-membered ring, and compound 46
could be isolated in 47% overall yield from 44. Removal
of the protecting groups by treatment of 46 with n-Bu4-
NF and then sodium methoxide provided 17 in 91% yield.
Syn th esis of Oligosa cch a r id es. Once 15-17 had
been synthesized, the assembly of the oligosaccharides
was carried out with minimal difficulty. The preparation
of 5 and 6 is illustrated in Scheme 7. Glycosylation of 15
with 1 equiv of 1818 using N-iodosuccinimide and silver
triflate promotion yielded a mixture of 47 and 48, which
were not separable by chromatography. The mixture of
disaccharides was then debenzoylated and then acety-
lated to provide two separable products. After purifica-
tion, each was deprotected, affording 5 and 6 in 20% yield
each from 15. Although the yields are rather modest, we
viewed this approach as better than a strategy involving
monoprotection of diol 15 followed by glycosylation of the
resulting alcohols.
As outlined in Scheme 8, the preparation of trisaccha-
ride 9 was also straightforward. Reaction of 15 with 2
equiv of 18 yielded trisaccharide 49. Purification of the
product was complicated by the presence of the glycosyl-
succinimide adduct 50,33 which had chromatographic
properties essentially identical to those of 49. Therefore,
the crude product was treated with sodium methoxide
in methanol to provide 9, which could easily be separated
from the deacylated derivative of 50. Trisaccharide 9 was
isolated in 68% overall yield from 15.
The conversion of 16 into 7, 8, and 10 (Scheme 9) was
achieved in a manner similar to that of the synthesis of
5, 6, and 9. Glycosylation of 16 with a limiting amount
of 18 provided a mixture of disaccharides 51 and 52 in
addition to trisaccharide 53. These oligosaccharides were
separated by chromatography and immediately depro-
tected. After removal of the benzoyl groups, 7, 8, and 10
were obtained in 6%, 34%, and 17% yields, respectively.
The structures of 7 and 8 were differentiated as described
for 5 and 6 (see Figure S1 in the Supporting Information).
In the synthesis of oligosaccharide analogues from 17,
we anticipated that glycosylation of the tertiary hydroxyl
group would pose a particular challenge. We therefore
chose to use sulfoxide 1934 (Chart 1) as the glycosyl donor.
Glycosyl sulfoxides are highly reactive glycosylating
agents, which have often been used to provide good yields
of glycosides from hindered alcohols.35 As illustrated in
Scheme 10, reaction of 17 with 2.5 equiv of 19 and triflic
anhydride in the presence of di-tert-butylmethylpyridine
provided both disaccharide 54 and trisaccharide 55.
Following their separation by chromatography, each
The structures of the glycosylation products were
differentiated through the use one- and two-dimensional
1
1H/1H and H/13C correlation NMR spectroscopy. In the
13C NMR spectrum of 5, the signal arising from C-4 was
shifted downfield relative to that of 15 (76.4 ppm in 5 vs
70.7 ppm in 15) as would be expected32 (see Chart 1 for
atom numbers). A similar trend was observed for the C-2
resonance in the 13C NMR spectrum of 6 (76.6 ppm in 6
vs 69.5 ppm in 15). Further support for the structures of
1
(33) McCarren, P. R.; Lowary, T. L. Unpublished results.
(34) Prepared by reaction of 18 with m-CPBA; see the Experimental
Section for details.
these disaccharides could be obtained from the H NMR
(32) Duus, J . Ø.; Gottfredsen, C. H.; Bock, K. Chem. Rev. 2000, 100,
4589.
(35) Kahne, D.; Walker, S.; Cheng, Y.; Van Engen, D. J . Am. Chem.
Soc. 1989, 111, 6881.

4158 J . Org. Chem., Vol. 67, No. 12, 2002
Houseknecht and Lowary
Sch em e 9a
a
Reagents and conditions: (a) N-iodosuccinimide, silver triflate, CH2Cl2, 0 °C; (b) NaOCH3, CH3OH, rt, 6% (from 16); (c) NaOCH3,
CH3OH, rt, 34% (from 16); (d) NaOCH3, CH3OH, rt, 17% (from 16).
Sch em e 10a
Sch em e 11a
a
Reagents and conditions: (a) BzCl, DMAP, pyridine, 0 °C, 87%;
(b) 19, Tf2O, DTBMP, CH2Cl2, -78 °C f rt, (c) NaOCH3, CH3OH,
rt; (d) Ac2O, pyridine, DMAP, rt; (e) NaOCH3, CH3OH, rt, 9% (from
56).
yield from 17. Although the yield of the product is very
low, we were nevertheless able to synthesize sufficient
quantities for the required NMR studies.
NMR In vestiga tion s. With oligosaccharides 5-14 in
1
hand, we used H NMR spectroscopy in combination with
PSEUROT analysis to investigate the conformers popu-
lated by the rings in these molecules with no conforma-
tional constraints (residues B and C). Measurement of
a
Reagents and conditions: (a) 19, Tf2O, DTBMP, CH2Cl2, -78
1
the vicinal H-1H coupling constants required for these
°C f rt; (b) NaOCH3, CH3OH, rt, 37% (from 17); (c) NaOCH3,
CH3OH, rt, 29% (from 17).
calculations was possible in all cases except 11. In this
disaccharide analogue, spectral overlap prohibited the
3
oligosaccharide was debenzoylated to provide 12 (37%)
measurement of the appropriate J H,H. The results of the
and 13 (29%). The structure of 12 was determined by 13
C
PSEUROT analyses of rings B and C in 5-14 are
provided in Table 1. For purposes of comparison, the
conformers in 2-4 are also included. These analyses
required that the R-arabinofuranosyl residues in 9, 10,
and 13 be differentiated. For 9 and 13, this could be done
NMR spectroscopy as described above for 5 and 6. The
hydroxymethyl group carbon in 17 resonates at 59.9 ppm,
whereas in the disaccharide this signal appears at 65.6
ppm.
The remaining disaccharide target 11 was prepared as
outlined in Scheme 11. Protection of the primary hydroxyl
group was achieved by reaction of 17 with benzoyl
chloride in pyridine to provide 56 in 87% yield. As
expected, glycosylation of this alcohol was difficult;
reaction of 56 with an excess of 19 proceeded slowly and
produced only small amounts of 57. Following deprotec-
tion, the 3-linked disaccharide 11 was obtained in 9%
through the use of 13C and H NMR chemical shifts (see
1
Figure S1 in the Supporting Information). This was not
possible for 10, and therefore the conformer populations
presented in Table 1 for residues B and C in this molecule
could be reversed.
From these data, the major conclusion that can be
drawn is that locking ring A in 2-4 into either the E3 or
OE conformation has essentially no effect upon the

Oligofuranoside Synthesis and Conformational Analysis
J . Org. Chem., Vol. 67, No. 12, 2002 4159
Ta ble 1. P SEUROT An a lysis of Resid u es B a n d C in 2-14a
2
3
4
5
6
7
8
9
10
11
12
13
14
ringb
B
68
C
B
C
C
B
C
B
B
C
Bg
66
E4
73
185
2T3
27
Cg
64
E4
71
184
2T3
29
Ch
B
B
C
B
PNc (deg)
66
E4
71
185
2T3
29
65
E4
67
185
2T3
33
65
E4
69
185
2T3
31
62
E4
72
184
2T3
28
65
E4
74
185
2T3
26
63
E4
72
184
2T3
28
66
E4
73
185
2T3
27
65
E4
76
185
2T3
24
65
E4
67
185
2T3
33
NDi 66
61
E4
73
184
2T3
27
62
E4
69
184
2T3
31
68
E4
69
185
2T3
31
N conformerd E4
ND
ND
ND
ND
ND
E4
70
185
2T3
30
e
XN (%)
71
PSc (deg)
S conformerd
XSf (%)
185
2T3
29
RMS (Hz)
0.21 0.12 0.13 0.10 0.16 0.14 0.07 0.04 0.12 0.34 0.04 0.13 ND
0.14 0.08 0.35 0.11
a
b
Calculated using a constant tm ) 39°. See Figure 3 for the assignment of ring letters. c P ) pseudorotational phase angle and is
defined in ref 4. See Figure 1 for conformer definitions. e Population of the N conformer. f Population of the S conformer. Results could
be reversed. Analysis could not be due to spectral overlap that prohibited the measurement of coupling constants. Not determined.
d
g
h
i
(15 mL). Iodine (1.38 g, 5.4 mmol) and triphenylphosphine
(1.42 g, 5.4 mmol) were added, and the reaction mixture was
stirred for 3 h before CH3OH was added and the solvents were
evaporated. Chromatography (CH2Cl2/CH3OH, 20:1) yielded
the diol contaminated with iodine. This oil was dissolved in
pyridine (10 mL) and acetic anhydride (10 mL), and the
solution was stirred for 12 h. The reaction mixture was then
cooled to 0 °C, CH3OH was added, and then the reaction
mixture was diluted with CH2Cl2. The resulting solution was
washed with 0.1 M HCl and brine and then dried. Evaporation
of the solvent and chromatography (hexanes/EtOAc, 3:1)
yielded the pure diacetylated product. This solid was dissolved
in CH3OH (5 mL), 1 M sodium methoxide (5 drops) was added,
and the reaction mixture was stirred for 12 h. The solution
was then neutralized with Amberlite IR 120 (H+) resin and
filtered. Evaporation of the solvent gave 15 as a white solid
(248 mg, 63%): Rf 0.23 (CH2Cl2/CH3OH, 6:1); [R]D +43.2 (c 1.3,
CH3OH); 1H NMR (800 MHz, D2O, δ) 4.36 (d, 1 H, J ) 1.9
Hz), 4.32 (dd, 1 H, J ) 1.9, 1.9 Hz), 4.25 (d, 1 H, J ) 5.5 Hz),
4.15 (d, 1 H, J ) 10.8 Hz), 3.96 (dd, 1 H, J ) 3.2, 10.8 Hz),
3.93 (dd, 1 H, J ) 2.8, 5.5 Hz), 3.79 (dd, 1 H, J ) 2.9, 13.3
Hz), 3.51 (d, 1 H, J ) 13.3 Hz); 13C NMR (100.6 MHz, D2O, δ)
80.8, 78.6, 70.6, 69.4, 68.1, 65.2; HRMS (EI) m/z calcd for
C6H10O4Na+ 169.0471, found 169.0478.
conformers populated by rings B and C. Similar to those
in the parent structures, these residues in 5-14 adopt
approximately 1:1 ratios of 3T4/E4 (N) and E1 (S) conform-
ers. The results presented in Table 1 are in contrast to
the previous investigations on oligonucleotides. In those
studies17 it was demonstrated that a nucleoside contain-
ing a conformationally locked or biased furanose ring does
influence the conformations of adjacent rings in the
oligonucleotide. Our results suggest that the conforma-
tional effects seen in those systems depend on the
presence of the nucleotide base. However, it is also
possible that in larger oligosaccharides, possessing more
than one conformationally restricted residue, the trans-
mission of conformational information may be more
significant. In particular, one could imagine that such
effects might be larger in cyclic arabinofuranosyl oli-
gosaccharides (e.g., cyclodextrin-like species) that contain
conformationally locked residues.
Con clu sion s
In conclusion, described here is the synthesis of a panel
of arabinofuranosyl oligosaccharide analogues (5-13) in
which one ring is locked into either the E3 or OE
conformation. Subsequent analysis of these glycans, as
well as disaccharide 14, by 1H NMR spectroscopy has
demonstrated that in these oligosaccharides the locked
residue does not exert any detectable influence upon the
conformers populated by adjacent conformationally un-
restricted furanose rings. We are currently testing 5-14
as substrates for mycobacterial arabinosyltransferases.
1,5:3,6-Dian h ydr o-D-talitol (16). 1,5-Anhydro-D-talitol (22;22
99 mg, 0.60 mmol) was dissolved in pyridine (2 mL). To this
solution were added iodine (316 mg, 1.21 mmol) and triphen-
ylphosphine (304 mg, 1.21 mmol). After 4 h, CH3OH was added
and the solution concentrated. Chromatography (CH2Cl2/CH3-
OH, 20:1) yielded the desired diol contaminated with iodine.
The crude product was therefore dissolved in pyridine (5 mL)
and acetic anhydride (5 mL), and the solution was stirred for
12 h, at which point it was cooled to 0 °C and CH3OH was
added. The reaction mixture was diluted with CH2Cl2 and
washed with 0.1 M HCl and brine and the organic layer dried.
Evaporation of the solvent and chromatography of the result-
ing residue (hexanes/EtOAc, 3:1) yielded the pure, fully
acetylated product. This solid was dissolved in CH3OH (5 mL),
1 M sodium methoxide (5 drops) was added, and the reaction
mixture was stirred for 12 h. The solution was neutralized with
Amberlite IR 120 (H+) resin and filtered. Evaporation of the
solvent gave 16 as a white solid (73 mg, 83%): Rf 0.28 (CH2-
Cl2/CH3OH, 6:1); [R]D +51.9 (c 0.7, CH3OH); 1H NMR (800
MHz, D2O, δ) 4.27 (dd, 1 H, J ) 2.4, 2.4 Hz), 4.17 (s, 1 H),
3.99 (d, 1 H, J ) 10.8 Hz), 3.95 (dd, 1 H, J ) 3.0, 10.8 Hz),
3.87 (d, 1 H, J ) 1.8 Hz), 3.80 (dd, 1 H, J ) 6.7, 11.1 Hz), 3.76
Exp er im en ta l Section
Gen er a l Meth od s. Solvents were distilled from the ap-
propriate drying agents before use. Unless stated otherwise,
all reactions were carried out under a positive pressure of
argon and were monitored by TLC on silica gel 60 F254 (0.25
mm). Spots were detected under UV light, by charring with
10% H2SO4 in ethanol, or by charring with anisaldehyde in
ethanol. Solvents were evaporated under reduced pressure and
below 40 °C (bath). Organic solutions of crude products were
dried over anhydrous Na2SO4. Column chromatography was
performed on silica gel 60 (40-60 mM). The ratio between
silica gel and crude product ranged from 100:1 to 50:1 (w/w).
Optical rotations were measured at 21 ( 2 °C. 1H NMR spectra
were recorded at 400, 500, or 800 MHz, and chemical shifts
are referenced to either TMS (δ 0.0, CDCl3) or external dioxane
(δ 3.75, D2O). 13C NMR spectra were recorded at 100 or 125
MHz, and 13C chemical shifts are referenced to CDCl3 (δ 77.00,
CDCl3) or external dioxane (δ 68.11, D2O). Elemental analyses
were performed by Atlantic Microlab Inc., Norcross, GA.
Electrospray mass spectra were recorded on samples sus-
pended in THF or CH3OH.
(dd, 1 H, J ) 1.2, 6.7 Hz), 3.23 (dd, 1 H, J ) 9.3, 11.0 Hz); 13
C
NMR (150.9 MHz, D2O, δ) 83.7, 77.4, 73.0, 68.7, 67.3, 63.3;
HRMS (EI) m/z calcd for C12H20O8Na+ (dimer) 315.1050, found
315.1044.
5-O-ter t-Bu tyld ip h en ylsilyl-3-C-h yd r oxym eth yl-1,2-O-
isop r op ylid en e-â-D-lyxofu r a n osid e (24). Olefin 23 (33.0
mg, 0.078 mmol) and 4-methylmorpholine N-oxide (15.7 mg,
0.117 mmol) were dissolved in acetone and water (3:1) and
the solution cooled to 0 °C. To this solution was added OsO4
(0.2 mg, 0.001 mmol), and the reaction mixture was stirred
for 12 h at 0 °C. The solution was then stirred for an additional
1 d at room temperature before a saturated aqueous solution
of sodium sulfite and CH2Cl2 were added. The organic layer
1,5:3,6-Dia n h yd r o-D-ga la ctitol (15). 1,5-Anhydro-D-ga-
lactitol (21;21 444 mg, 2.71 mmol) was dissolved in pyridine

4160 J . Org. Chem., Vol. 67, No. 12, 2002
Houseknecht and Lowary
was then separated, washed with water and brine, and dried.
Chromatography (hexanes/EtOAc, 2:1) yielded diol 24 (30 mg,
85%): Rf 0.42 (hexanes/EtOAc, 1:1); [R]D +10.4 (c 1.5, CHCl3);
1H NMR (500 MHz, CDCl3, δ) 7.727.67 (m, 4 H), 7.49-7.31
(m, 6 H), 5.92 (d, 1 H, J ) 3.8 Hz), 4.44 (d, 1 H, J ) 3.8 Hz),
4.29-4.24 (m, 2 H), 4.06 (dd, 1 H, J ) 10.5, 10.5 Hz), 3.91 (dd,
1 H, J ) 9.6, 12.2 Hz), 3.74 (dd, 1 H, J ) 3.9, 10.5 Hz), 3.22 (s,
1 H), 2.92 (dd, 1 H, J ) 4.5, 9.6 Hz), 1.26 (s, 3 H), 1.23 (s, 3
H), 1.13 (s, 9 H); 13C NMR (125.8 MHz, CDCl3, δ) 135.5, 135.4,
131.9, 130.1, 128.0, 112.5, 106.0, 88.8, 85.4, 82.4, 77.3, 77.0,
76.7, 64.1, 62.6, 26.8, 26.3, 25.5, 19.1; HRMS (EI) m/z calcd
for C25H34O6SiNa+ 481.2017, found 481.2032.
110.0, 84.7, 83.8, 78.7, 72.2, 62.0, 55.1; HRMS (EI) m/z calcd
for C13H18O5Na+ 277.1046, found 277.1031.
Meth yl 3-O-Ben zyl-2,5-bis-O-d i-ter t-bu tylsilyl-r-D-a r a -
bin ofu r a n osid e (33). Diol 32 (126 mg, 0.049 mmol) was
dissolved in CH2Cl2 and the solution cooled to -78 °C. Freshly
distilled pyridine (120 µL, 1.47 mmol) and then di-tert-
butylsilyl bis(trifluoromethanesulfonate) (220 µL, 0.59 mmol)
were added to this solution via syringe. The reaction mixture
was warmed to 0 °C over 4 h before CH3OH was added.
Evaporation of the solvent and chromatography (hexanes/
EtOAc, 20:1) yielded 33 (144 mg, 74%): Rf 0.63 (hexanes/
1
EtOAc, 4:1); [R]D +102.0 (c 1.0, CHCl3); H NMR (400 MHz,
3,6-An h yd r o-5-O-ter t-bu tyld ip h en ylsilyl-3-C-h yd r oxy-
m eth yl-1,2-O-isopr opyliden e-â-D-lyxofu r an oside (25). Ole-
fin 23 (100 mg, 0.236 mmol) was dissolved in CH2Cl2, and a
saturated aqueous solution of NaHCO3 (1 mL) was added.
m-Chloroperoxybenzoic acid (233 mg, 0.944 mmol) was added,
and the solution was stirred for 2 d. The reaction mixture was
then diluted with CH2Cl2 and the organic layer washed with
a saturated aqueous solution of sodium sulfite, then water,
and brine. The organic layer was dried and concentrated;
chromatography (hexanes/EtOAc, 10:1) yielded epoxide 25 (34
mg, 33%): Rf 0.51 (hexanes/EtOAc, 6:1); [R]D +6.7 (c 0.7,
CHCl3); 1H NMR (500 MHz, CDCl3, δ) 7.70-7.68 (m, 4 H),
7.45-7.43 (m, 6 H), 6.00 (d, 1 H, J ) 4.1 Hz), 4.39 (d, 1 H, J
) 4.1 Hz), 4.07 (dd, 1 H, J ) 4.9, 8.6 Hz), 4.00 (dd, 1 H, J )
8.8, 10.1 Hz), 3.82 (dd, 1 H, J ) 4.9, 10.1 Hz), 3.45 (d, 1 H, J
) 4.8 Hz), 3.21 (d, 1 H, J ) 4.8 Hz), 1.41 (s, 3H), 1.33 (s, 3H),
1.09 (s, 9H); 13C NMR (125.8 MHz, CDCl3, δ) 135.6, 135.5,
134.8, 133.1, 132.8, 129.8, 129.7, 129.6, 127.8, 127.8, 127.7,
112.7, 105.1, 84.3, 82.8, 64.7, 64.3, 46.2, 26.8, 26.7, 26.6, 26.0,
19.1; HRMS (EI) m/z calcd for C25H32O5SiNa+ 463.1911, found
463.1932.
CDCl3, δ) 7.34-7.26 (m, 5 H), 4.96 (s, 1 H), 4.69 (d, 1 H, J )
12.6 Hz), 4.58 (d, 1 H, J ) 12.5 Hz), 4.43 (s, 1 H), 4.42 (s, 1 H),
4.10-4.07 (m, 2 H), 3.80 (dd, 1 H, J ) 12.4,2.2 Hz), 3.44 (s, 3
H), 0.98 (s, 9 H), 0.84 (s, 9 H); 13C NMR (100.6 MHz, CDCl3,
δ) 137.5, 128.5, 128.0, 127.9, 110.1, 86.2, 81.8, 79.1, 71.9, 66.9,
55.5, 28.0, 27.5, 21.6, 21.2. Anal. Calcd for C21H34O5Si: C,
63.92; H, 8.69. Found: C, 63.58; H, 8.71.
Meth yl 2,5-Bis-O-d i-ter t-bu tylsilyl-r-D-a r a bin ofu r a n o-
sid e (34). Benzyl ether 33 (853 mg, 2.16 mmol) was dissolved
in CH3OH, and Pd(OH)2 (133 mg, 0.22 mmol) was added. The
flask was then flushed with H2, and the reaction mixture was
stirred for 3 d before being filtered through Celite. The filtrate
was evaporated, and the resulting residue was chromato-
graphed (hexanes/EtOAc, 10:1), yielding pure 34 as a clear oil
(618 mg, 94%): Rf 0.53 (hexanes/EtOAc, 4:1); [R]D +84.6 (c 2.6,
CHCl3); 1H NMR (400 MHz, CDCl3, δ) 4.96 (s, 1 H), 4.34-
4.26 (m, 3 H), 4.11 (dd, 1 H, J ) 12.6, 1.2 Hz), 4.05 (dd, 1 H,
J ) 12.6, 2.2 Hz), 3.42 (s, 3 H), 3.39 (d, 1 H, J ) 12.2 Hz), 0.99
(s, 18 H); 13C NMR (100.6 MHz, CDCl3, δ) 109.0, 89.1, 80.5,
76.2, 66.2, 55.0, 28.0, 27.6, 21.7, 21.3. Anal. Calcd for C14H28O5-
Si: C, 55.23; H, 9.27. Found: C, 55.21; H, 9.37.
5-O-ter t-Bu tyld ip h en ylsilyl-1,2-O-isop r op ylid en e-3-C-
vin yl-â-D-lyxofu r a n osid e (27). Alcohol 26 (100 mg, 0.23
mmol) was added to a mixture of NaOAc (115 mg, 1.40 mmol),
PCC (152 mg, 0.70 mmol), and crushed 4 Å molecular sieves
in CH2Cl2. The solution was stirred for 4 h before diethyl ether
and hexanes were added. The reaction mixture was then
filtered through silica gel and eluted with diethyl ether.
Evaporation of the solvent yielded the crude ketone, which was
dissolved in THF, and the solution was cooled to 0 °C. A
solution of 1 M vinylmagnesium bromide (0.51 mL, 0.51 mmo1)
in THF was then added slowly, and the reaction mixture was
stirred for 12 h before CH3OH was added. The reaction mixture
was diluted with CH2Cl2 and then washed with water and
brine before being dried and evaporated. Chromatography
(hexanes/EtOAc, 10:1) yielded the pure olefin 27 (48 mg,
45%): Rf 0.62 (hexanes/EtOAc, 4:1); [R]D +5.3 (c 1.6, CHCl3);
1H NMR (500 MHz, CDCl3, δ) 7.75-7.73 (m, 4 H), 7.43-7.42
(m, 6 H), 5.91 (dd, 1 H, J ) 10.7, 17.2 Hz), 5.80 (d, 1 H, J )
4.1 Hz), 5.51 (dd, 1 H, J ) 1.2, 17.2 Hz), 5.24 (dd, 1 H, J )
1.1, 10.7 Hz), 4.41 (d, 1 H, J ) 4.1 Hz), 4.14 (dd, 1 H, J ) 5.2,
10.5 Hz), 3.98 (dd, 1 H, J ) 6.1, 6.1 Hz), 3.93 (dd, 1 H, J )
6.2, 10.5 Hz), 3.41 (s, 1 H), 1.54 (s, 3 H), 1.42 (s, 3 H), 1.10 (s,
9 H); 13C NMR (125.8 MHz, CDCl3, δ) 139.6, 135.7, 135.6,
133.3, 133.1, 129.7, 127.7, 127.7, 114.7, 114.6, 104.6, 85.4, 85.4,
78.0, 63.0, 27.0, 26.9, 26.8, 19.1; HRMS (EI) m/z calcd for
Meth yl 3-Deoxy-2,5-bis-O-d i-ter t-bu tylsilyl-3-C-m eth -
ylen e-r-D-ar abin ofu r an oside (35). PCC (833 mg, 3.84 mmol),
sodium acetate (629 mg, 7.68 mmol), and crushed 4 Å molec-
ular sieves were stirred in CH2Cl2 for 30 min. To this solution
was added dropwise a solution of alcohol 34 (388 mg, 1.28
mmol) dissolved in CH2Cl2. After the addition was complete
the solution was stirred for 8 h before hexanes and ether were
added. After being stirred for an additional 1 h, the reaction
mixture was filtered through a column of silica gel and the
column washed with ether. The solvents were evaporated to
yield the desired ketone as a light brown oil. A stir bar and a
0.5 M solution of Petasis reagent (15.4 mL, 7.68 mmol) in THF
were then added, and the reaction mixture was heated at
reflux for 13 h. The solution was then cooled to room temper-
ature, and petroleum ether was added. The resultant suspen-
sion was filtered, and the filtrate was concentrated. Chroma-
tography (hexanes/EtOAc, 20:1) yielded olefin 35 as a clear
oil (273 mg, 71%): Rf 0.54 (hexanes/EtOAc, 6:1); [R]D +107.8
(c 1.7, CHCl3); 1H NMR (400 MHz, CDCl3, δ) 5.48 (d, 1 H, J )
1.6 Hz), 5.31 (d, 1 H, J ) 1.3 Hz), 4.89 (s, 1 H), 4.63 (dd, 1 H,
J ) 1.6, 1.6 Hz), 4.40 (s, 1 H), 4.22 (dd, 1 H, J ) 12.0, 1.5 Hz),
3.90 (dd, 1 H, J ) 12.0, 1.9 Hz), 3.37 (s, 3 H), 1.01 (s, 9 H),
0.92 (s, 9 H); 13C NMR (100.6 MHz, CDCl3, δ) 146.0, 113.7,
107.8, 81.3, 78.7, 77.2, 68.6, 54.8, 28.0, 27.9, 21.5, 21.0. Anal.
Calcd for C15H28O4Si: C, 59.96; H, 9.39. Found: C, 60.12; H,
9.25.
C
26H34O5SiNa+ 477.2068, found 477.2048.
Meth yl 3-O-Ben zyl-r-D-a r a bin ofu r a n osid e (32). Benzyl
Meth yl 2,5-Bis-O-d i-ter t-bu tylsilyl-3-C-h yd r oxym eth yl-
6-O-tolu en esu lfon yl-r-D-a r a bin ofu r a n osid e (37). Olefin 35
(25 mg, 0.83 mmol) was dissolved in EtOAc (8 mL) and
acetonitrile (8 mL) and the solution cooled to 0 °C. To this
stirred solution was added a mixture of RuCl3‚H2O (17 mg,
0.08 mmol) and sodium periodate (267 mg, 1.25 mmol) in water
(4 mL). After 30 min the reaction mixture was poured into a
saturated aqueous solution of Na2S2O3 and the resulting
mixture stirred for 30 min. The layers were separated, and
then the aqueous layer was extracted with EtOAc. The
combined organic layers were washed with water and brine
before being dried with MgSO4. Evaporation yielded the crude
diol, which was dissolved in CH2Cl2, and the solution was
cooled to 0 °C. DABCO (372 mg, 3.32 mmol) and then TsCl
(476 mg, 2.5 mmol) were added, and the reaction mixture was
stirred for 2 h. The reaction mixture was filtered through
alcohol (15 mL) was heated at 120 °C while sodium (2.8 g,
123.3 mmol) was added in small portions. The suspension was
stirred until the sodium was completely dissolved. Epoxide 31
(6.0 g, 41.1 mmol) was added to the solution in one portion,
and the reaction mixture was stirred for 3 h. The solution was
then neutralized with acetic acid, and silica gel was added to
absorb the entire reaction mixture. The resulting solid was
transferred to the top of a column of silica gel, and the column
was eluted (hexanes/EtOAc, 20:1 f 10:1 f 1:2) to yield diol
32 as a colorless oil (8.71 g, 83%): Rf 0.51 (hexanes/EtOAc,
1
1:4); [R]D +109.3 (c 1.9, CH3OH); H NMR (400 MHz, CDCl3,
δ) 7.35-7.28 (m, 5 H), 4.87 (s, 1 H), 4.70 (d, 1 H, J ) 12.2 Hz),
4.56 (d, 1 H, J ) 12.2 Hz), 4.20-4.17 (m, 1 H), 4.16 (s, 1 H),
3.84-3.80 (m, 2 H), 3.56 (dd, 1 H, J ) 2.5, 11.9 Hz), 3.39 (s, 3
H); 13C NMR (100.6 MHz, CDCl3, δ) 137.6, 128.5, 127.9, 127.9,

Oligofuranoside Synthesis and Conformational Analysis
J . Org. Chem., Vol. 67, No. 12, 2002 4161
Celite, and the Celite was rinsed with CH2Cl2. The filtrate was
washed with water and brine, then dried, and evaporated.
Chromatography (hexanes/EtOAc, 10:1) yielded alcohol 37 as
a white solid (264 mg, 65%): Rf 0.48 (hexanes/EtOAc, 2:1); [R]D
+25.5 (c 1.5, CHCl3); 1H NMR (500 MHz, CDCl3, δ) 7.82-7.80
(m, 2 H), 7.34-7.32 (m, 2 H), 5.01 (s, 1 H), 4.59-4.56 (m, 2
H), 4.43 (dd, 1 H, J ) 10.1, 1.2 Hz), 4.29 (dd, 1 H, J ) 1.6, 1.6
Hz), 4.22 (s, 1 H), 4.13 (dd, 1 H, J ) 13.6, 1.8 Hz), 3.90 (dd, 1
H, J ) 13.6, 1.7 Hz), 3.42 (s, 3 H), 2.43 (s, 3 H), 1.03 (s, 9 H),
0.95 (s, 9 H); 13C NMR (125.8 MHz, CDCl3, δ) 144.7, 133.0,
129.7, 128.1, 108.1, 89.3, 89.2, 80.9, 80.4, 69.5, 64.8, 55.3, 29.1,
29.0, 28.0, 27.8, 22.5, 21.6, 21.3; HRMS (EI) m/z calcd for
1H NMR (500 MHz, CDCl3, δ) 7.53-7.52 (m, 2 H), 7.43-7.41
(m, 3 H), 6.07 (s, 0.5 H), 5.95 (s, 0.5 H), 5.11 (s, 1 H), 4.63-
4.60 (m, 1 H), 4.50-4.48 (m, 1 H), 4.41-4.39 (m, 1 H), 4.36 (s,
0.5 H), 4.31 (dd, 0.5 H, J ) 14.6, 1.2 Hz), 4.24 (dd, 0.5 H, J )
14.6, 1.2 Hz), 4.17 (d, 0.5 H, J ) 9.5 Hz), 4.06-4.00 (m, 1 H),
3.51 (s, 1.5 H), 3.48 (s, 1.5 H), 1.15 (s, 9 H), 1.08 (s, 4.5 H),
1.05 (s, 4.5 H); 13C NMR (125.8 MHz, CDCl3, δ) 137.6, 136.9,
129.3, 129.1, 128.3, 128.2, 126.7, 126.4, 108.8, 108.2, 104.1,
103.9, 88.9, 88.3, 88.0, 87.1, 81.3, 69.0, 66.9, 65.9, 65.8, 56.0,
55.6, 29.2, 28.2, 22.8, 22.7, 21.3, 21.2; HRMS (EI) m/z calcd
for C22H34O6SiNa+ 445.2017, found 445.2049.
Met h yl 3,6-Di-O-(b en zylid en e a cet a l)-3-C-h yd r oxy-
m eth yl-r-D-a r a bin ofu r a n osid e (43). Compound 42 (340 mg,
0.80 mmol) was dissolved in THF (20 mL) and the solution
transferred to a plastic syringe, with the tip sealed. To the
syringe were added a stir bar and 1.2 N HF-pyridine (2.0 mL,
2.4 mmol). The syringe was sealed, and the solution was
stirred for 4 h before the solvent was evaporated. Chromatog-
raphy (hexanes/EtOAc, 1:2) yielded 43 (195 mg, 86%) as a
white foam: Rf 0.49 (hexanes/EtOAc, 1:4); [R]D +104.3 (c 1.4,
CHCl3); 1H NMR (500 MHz, CDCl3, δ) 7.53-7.52 (m, 2 H),
7.43-7.41 (m, 3 H), 6.01 (s, 0.5 H), 5.94 (s, 0.5 H), 4.59 (d, 0.5
H, J ) 9.6 Hz), 4.51 (dd, 0.5 H, J ) 2.3, 2.3 Hz), 4.47-4.45
(m, 1 H), 4.17 (d, 0.5 H, J ) 9.7 Hz), 4.13-4.09 (m, 1 H), 4.05
(d, 0.5 H, J ) 9.7 Hz), 4.00-3.94 (m, 1 H), 3.84-3.68 (m, 2 H),
3.48 (s, 3 H), 2.67-2.63 (m, 1 H); 13C NMR (125.8 MHz, CDCl3,
δ) 137.4, 137.0, 129.4, 129.3, 128.4, 128.3, 126.7, 126.5, 109.5,
109.2, 104.5, 104.0, 88.9, 88.7, 85.6, 85.0, 79.0, 78.3, 66.7, 66.5,
61.2, 61.0, 55.7, 55.5; HRMS (EI) m/z calcd for C14H18O6Na+
305.0996, found 305.1001.
Meth yl 3,6-Di-O-(ben zylid en e a ceta l)-5-O-ter t-bu tyl-
d ip h e n ylsilyl-3-C-h yd r oxym e t h yl-r-D-a r a b in ofu r a n o-
sid e (44). Diol 43 (17 mg, 0.06 mmol) was dissolved in DMF
and the solution cooled to 0 °C. Imidazole (10 mg, 0.15 mmol)
and then TBDMSCl (14 mg, 0.09 mmol) were added, and the
reaction mixture was stirred for 4 h before water and then
ether were added. The layers were separated, and the aqueous
layer was extracted with ether (3×). The combined organic
extracts were then washed with water (3×) and brine before
being dried. Chromatography (hexanes/EtOAc, 4:1) yielded 44
(20 mg, 87%) as a clear syrup: Rf 0.6 (hexanes/EtOAc, 2:1);
[R]D +88.7 (c 1.5, CHCl3); 1H NMR (400 MHz, CDCl3, δ) 7.47-
7.45 (m, 2 H), 7.36-7.34 (m, 3 H), 5.96 (s, 0.5 H), 5.89 (s, 0.5
H), 4.94-4.91 (m, 1 H), 4.57-4.42 (m, 2 H), 4.05-3.73 (m, 5
H), 3.43 (s, 3 H), 0.93 (s, 9 H), 0.14 (s, 6 H); 13C NMR (100.6
MHz, CDCl3, δ) 137.5, 137.1, 129.3, 129.2, 128.3, 128.2, 126.7,
126.6, 109.8, 109.5, 104.6, 104.0, 88.9, 88.8, 86.3, 85.8, 78.3,
77.7, 77.3, 66.8, 66.7, 62.7, 62.5, 55.7, 55.5, 25.7, 18.2, -5.7;
HRMS (EI) m/z calcd for C20H32O6SiNa+ 419.1860, found
419.1843.
Meth yl 2,6-An h yd r o-3-O-ben zoyl-5-O-ter t-bu tyld ip h en -
ylsilyl-3-C-h yd r oxym eth yl-r-D-a r a bin ofu r a n osid e (46).
Alcohol 44 (31 mg, 0.08 mmol) was dissolved in dry CCl4 (5
mL), and barium carbonate (102 mg) and N-bromosuccinimide
(17 mg, 0.10 mmol) were added. The mixture was then heated
at reflux for 3 h before being filtered through Celite. The
filtrate was concentrated, and the resulting residue was
purified by chromatography (10:1, hexanes/EtOAc), yielding
bromide 45, which was immediately dissolved in DMF. Potas-
sium carbonate (83 mg, 0.6 mmol) was then added, and the
reaction mixture was stirred for 36 h. Water and ether were
then added, the layers were separated, and the ether layer
was washed several times with water and then brine. The
combined organic extracts were concentrated, and the result-
ing residue was purified by chromatography (hexanes/EtOAc,
10:1), yielding oxetane 46 (15 mg, 47%): Rf 0.52 (hexanes/
EtOAc, 4:1); [R]D +81.6 (c 1.3, CHCl3); 1H NMR (500 MHz,
CDCl3, δ) 8.05-8.04 (m, 2 H), 7.61-7.58 (m, 1 H), 7.47-7.44
(m, 2 H), 5.26 (s, 1 H), 5.02 (d, 1 H, J ) 8.0 Hz), 5.01 (s, 1 H),
4.87 (d, 1 H, J ) 7.7 Hz), 4.19-4.16 (m, 2 H), 4.11 (dd, 1 H, J
) 12.0, 7.6 Hz), 3.44 (s, 3 H), 0.88 (s, 9 H), 0.09 (s, 6 H); 13C
NMR (125.8 MHz, CDCl3, δ) 165.0, 133.6, 129.9, 129.1, 128.5,
128.4, 105.4, 91.3, 82.6, 81.3, 74.8, 62.4, 54.6, 25.8, 18.3, -5.3
(×2); HRMS (EI) m/z calcd for C20H30O6SiNa+ 417.1704, found
417.1724.
C
22H36O5SSiNa+ 511.1792, found 511.1812.
Meth y1 5-O-ter t-Bu tyldiph en ylsilyl-3-C-h ydr oxym eth yl-
6-O-tolu en esu lfon yl-r-D-a r a bin ofu r a n osid e (40). Alcohol
37 (225 mg, 0.46 mmol) was dissolved in THF (15 mL) and
the solution transferred to a plastic syringe, with the tip
sealed. A stir bar and 1.2 M HF-pyridine (1.2 mL, 1.4 mmol)
were added, and the reaction mixture was stirred for 4 h.
Evaporation of the solvent and chromatography (hexanes/
EtOAc, 1:3) yielded 38 (135 mg, 85%).
Triol 38 (112 mg, 0.32 mmol) was immediately dissolved in
DMF and the solution cooled to 0 °C. Imidazole (55 mg, 0.81
mmol) and then TBDMSCl (74 mg, 0.49 mmol) were added,
and the reaction mixture was stirred for 3 h. The solution was
diluted with water and ether, and the layers were separated.
The aqueous layer was extracted with ether (3×), and the
combined organic extracts were washed with water and brine
before being dried and evaporated. Chromatography (hexanes/
EtOAc, 3:1) yielded diol 40 (127 mg, 86%): Rf 0.42 (hexanes/
EtOAc, 1:1); [R]D +38.2 (c 2.3, CHCl3); 1H NMR (500 MHz,
CDCl3, δ) 7.86-7.85 (m, 2 H), 7.38-7.36 (m, 2 H), 4.91 (s, 1
H), 4.42 (d, 1 H, J ) 10.2 Hz), 4.33 (d, 1 H, J ) 10.2 Hz), 4.27
(dd, 1 H, J ) 2.4 Hz), 3.86 (s, 1 H), 3.82 (d, 2 H, J ) 2.4 Hz),
3.44 (s, 3 H), 2.48 (s, 3 H), 0.90 (s, 9 H), 0.13 (s, 6 H); 13C NMR
(125.8 MHz, CDCl3, δ) 144.8, 132.6, 129.8, 128.1, 109.2, 88.4,
80.8, 78.3, 69.9, 61.9, 55.2, 25.6 (×2), 21.6, 18.2, -5.8 (×2);
HRMS (EI) m/z calcd for C20H34O8SSiNa+ 485.1636, found
485.1645.
Met h yl 3,6-An h yd r o-5-O-ter t-b u t yld ip h en ylsilyl-3-C-
h yd r oxym eth yl-r-D-a r a bin ofu r a n osid e (41). Diol 40 (33
mg, 0.07 mmol) was dissolved in THF and the solution cooled
to 0 °C. DBU (12 mL, 0.08 mmol) was added slowly, and the
reaction mixture was stirred for 4 h. Evaporation of the solvent
and chromatography (hexanes/EtOAc, 6:1) yielded epoxide 41
(12 mg, 60%): Rf 0.34 (hexanes/EtOAc, 3:1); [R]D +104.2 (c 1.1,
CHCl3); 1H NMR (500 MHz, CDCl3, δ) 4.97 (s, 1 H), 4.09 (d, 1
H, J ) 11.6 Hz), 4.06 (dd, 1 H, J ) 1.6, 1.6 Hz), 3.94 (dd, 1 H,
J ) 11.3, 1.9 Hz), 3.62 (d, 1 H, J ) 11.6 Hz), 3.58 (dd, 1 H, J
) 11.3, 1.5 Hz), 3.48 (s, 3 H), 3.20 (d, 1 H, J ) 4.4 Hz), 2.91 (d,
1 H, J ) 4.4 Hz), 0.96 (s, 9 H), 0.17 (s, 6 H); 13C NMR (125.8
MHz, CDCl3, δ) 108.5, 80.0, 75.5, 65.1, 63.0, 54.8, 46.0, 25.8,
18.3, -5.7; HRMS (EI) m/z calcd for C13H26O5SiNa+ 313.1442,
found 313.1436.
Meth yl 3,6-Di-O-(ben zylid en e a ceta l)-2,5-bis-O-d i-ter t-
bu tylsilyl-3-C-h ydr oxym eth yl-r-D-ar abin ofu r an oside (42).
Olefin 35 (67 mg, 2.23 mmol) was dissolved in EtOAc (20 mL)
and acetonitrile (20 mL) and the solution cooled to 0 °C. To
this stirred solution was added a solution of RuCl3‚H2O (45
mg, 0.22 mmol) and sodium periodate (716 mg, 3.35 mmol) in
water (10 mL). After 20 min the reaction mixture was poured
into a saturated aqueous solution of Na2S2O3 and the resulting
mixture stirred for 30 min. The aqueous layer was separated
and then extracted with EtOAc. The combined organic layers
were washed with water and brine before being dried with
MgSO4. Evaporation yielded the crude diol, which was then
dissolved in CH2Cl2. Benzylidene dimethyl acetal (0.66 mL,
4.46 mmol) and p-TsOH (catalytic) were added, and the
solution was stirred for 6 h, before being heated at 40 °C, with
stirring for 12 h. The reaction mixture was cooled, and a
saturated aqueous solution of NaHCO3 was added followed by
CH2Cl2. The organic layer was then washed with water and
brine, dried, and evaporated. Chromatography (hexanes/
EtOAc, 40:1) yielded 42 (569 mg, 60%) as a 1:1 diastereomeric
mixture: Rf 0.4 (hexanes/EtOAc, 6:1); [R]D +49.8 (c 4.6, CHCl3);

4162 J . Org. Chem., Vol. 67, No. 12, 2002
Houseknecht and Lowary
Meth yl 2,6-An h yd r o-3-C-h yd r oxym eth yl-r-D-a r a bin o-
fu r a n osid e (17). Compound 46 (62 mg, 0.157 mmol) was
dissolved in THF (5 mL) and the solution cooled to 0 °C. A
solution of 1 M TBAF in THF (0.47 mL, 0.47 mmol) was added,
and the reaction mixture was stirred for 2 h. A solution of 1
M sodium methoxide in CH3OH (1 mL) was then added, and
the solution was stirred for another 2 h. The reaction mixture
was then neutralized with acetic acid, and the solvent was
evaporated. Chromatography (CH2Cl2/CH3OH, 20:1) yielded
diol 17 (25 mg, 91%) as a white solid: Rf 0.26 (CH2Cl2/CH3-
OH, 12:1); [R]D +108.1 (c 0.6, CH3OH); 1H NMR (500 MHz,
D2O, δ) 5.00 (s, 1 H), 4.92 (s, 1 H), 4.76 (d, 1 H, J ) 7.8 Hz),
4.63 (dd, 1 H, J ) 7.8, 1.1 Hz), 4.00 (ddd, 1 H, J ) 7.6, 3.7, 0.9
Hz), 3.91 (dd, 1 H, J ) 12.0, 7.7 Hz), 3.82 (dd, 1 H, J ) 12.0,
3.8 Hz), 3.39 (s, 3 H); 13C NMR (125.8 MHz, D2O, δ) 105.5,
93.9, 81.3, 78.5, 77.8, 59.9, 54.7; HRMS (EI) m/z calcd for
C7H12O5Na+ 199.0577, found 199.0595.
2,3,5-Tr i-O-ben zoyl-r-D-a r a bin ofu r a n osyl p-Tolyl-(R/S)
Su lfoxid e (19). Thioglycoside 18 (1.00 g, 1.76 mmol) was
dissolved in CH2Cl2 and cooled to 0 °C. m-Chloroperoxybenzoic
acid (407 mg, 2.37 mmol) was added, and the reaction mixture
was allowed to warm slowly to room temperature. A saturated
aqueous solution of Na2S2O3 was added, and the reaction
mixture was diluted with CH2Cl2. After the reaction mixture
was stirred for 30 min, the layers were separated, and the
organic layer was washed with water and brine. Chromatog-
raphy (hexanes/EtOAc, 4:1) yielded sulfoxide 19 in a diaster-
eomeric ratio of 7:3 (962 mg, 94%) as a clear oil: Rf 0.4
(hexanes/EtOAc, 2:1); [R]D -10.4 (c 1.4, CHCl3); 1H NMR (500
MHz, CDCl3, δ) 8.18-7.23 (m, 19 H), 6.42 (dd, 0.7 H, J ) 3.4
Hz), 6.09 (dd, 0.3 H, J ) 3.0 Hz), 5.85 (dd, 0.3 H, J ) 5.8, 3.2
Hz), 5.79 (dd, 0.7 H, J ) 5.3, 3.5 Hz), 5.08 (d, 0.7 H, J ) 3.3
Hz), 5.04 (dd, 0.3 H, J ) 9.5, 5.4 Hz), 4.99 (d, 0.3 H, J ) 2.8
Hz), 4.86 (dd, 0.7 H, J ) 9.5, 5.2 Hz), 4.77-4.72 (m, 1 H), 4.68-
4.64 (m, 1 H), 2.36 (s, 1 H), 2.20 (s, 2 H). 13C NMR (125.8 MHz,
CDCl3, δ) 166.0, 165.9, 165.6, 165.4, 164.5, 142.4, 142.1, 136.5,
133.6, 133.4, 133.1, 130.1, 130.0, 130.0, 129.9, 129.9, 129.7,
129.6, 129.5, 128.8, 128.5, 128.5, 128.3, 128.3, 128.3, 125.2,
124.2, 100.3, 100.0, 83.9, 83.5, 78.9, 77.6, 75.0, 63.5, 63.3, 21.4,
21.1; HRMS (EI) m/z calcd for C33H28O8SNa+ 607.1397, found
607.1351.
(d, 1 H, J ) 5.5 Hz), 4.16 (d, 1 H, J ) 10.8 Hz), 3.98 (ddd, 1 H,
J ) 3.2, 6.1, 6.1 Hz), 3.97 (dd, 1 H, J ) 1.9, 3.7 Hz), 3.94-3.92
(m, 2 H), 3.84 (dd, 1 H, J ) 3.7, 6.4 Hz), 3.78 (dd, 1 H, J ) 2.8,
13.3 Hz), 3.74 (dd, 1 H, J ) 3.2, 12.3 Hz), 3.62 (dd, 1 H, J )
5.8, 12.3 Hz), 3.51 (d, 1 H, J ) 13.4 Hz);
l3C NMR (150.9 MHz,
D2O, δ) 106.9, 84.2, 81.8, 80.2, 76.8, 76.4, 69.6, 68.6, 65.5, 61.6;
HRMS (EI) m/z calcd for C11H18O8Na+ 301.0894, found 301.902.
Data for 6: Rf 0.55 (CH2Cl2/CH3OH, 3:1); [R]D +118.7 (c 0.6,
l
CH3OH); H NMR (800 MHz, D2O, δ) 5.02 (d, 1 H, J ) 1.7
Hz), 4.42 (d, 1 H, J ) 5.5 Hz), 4.39 (d, 1 H, J ) 1.9 Hz), 4.33
(dd, 1 H, J ) 2.5 Hz), 4.15 (d, 1 H, J ) 10.9 Hz), 4.05 (d, 1 H,
J ) 1.7 Hz), 4.01 (dd, 1 H, J ) 1.7, 3.6 Hz), 3.98 (dd, 1 H, J )
3.1, 10.8 Hz), 3.88 (dd, 1 H, J ) 2.7, 5.5 Hz), 3.87 (dd, 1 H, J
) 3.5, 6.5 Hz), 3.78 (dd, 1 H, J ) 2.8, 13.5 Hz), 3.76 (dd, 1 H,
J ) 3.3, 12.4 Hz), 3.66 (d, 1 H, J ) 13.4 Hz), 3.64 (dd, 1 H, J
) 6.0, 12.4 Hz); l3C NMR (150.9 MHz, D2O, δ) 107.9, 84.3, 81.9,
80.4, 78.7, 77.0, 76.6, 71.2, 68.5, 63.5, 61.7; HRMS (EI) m/z
calcd for C11Hl8O8Na+ 301.0894, found 301.0878.
r-D-Ar a b in ofu r a n osyl-(lf2)-[r-D-a r a b in ofu r a n osyl-
(1f4)]-1,5:3,6-d ia n h yd r o-D-ga la ctitol (9). Diol 15 (50 mg,
0.342 mmol), thioglycoside 18 (486 mg, 0.856 mmol), and a
stir bar were dried overnight with P2O5 in vacuo in a round-
bottom flask. The flask was cooled to 0 °C, and then 5:1 CH2-
Cl2/acetonitrile (6 mL) was added, followed by crushed 4 Å
molecular sieves. The mixture was stirred for 20 min before
NIS (241 mg, 1.07 mmol) and AgOTf (37 mg, 0.143 mmo1) were
added. After the mixture was stirred at 0 °C for 5 min,
triethylamine (5 drops) was added, and the reaction mixture
was filtered through Celite and then diluted with CH2Cl2. The
organic layer was then washed in succession with a saturated
aqueous solution of Na2S2O3, water, and brine, before being
dried and concentrated. Chromatography (hexanes/EtOAc, 3:1)
yielded the crude product (346 mg, 98%) contaminated with
50: Rf 0.47 (hexanes/EtOAc, 2:1).
The crude trisaccharide was dissolved in CH3OH (5 mL),
and 1 M sodium methoxide (5 drops) was added. The solution
was stirred for 14 h, then neutralized with Amberlite IR 120
(H+) resin, filtered, and concentrated. Chromatography (CH2-
Cl2/CH3OH, 3:1) yielded the pure trisaccharide 9 (93.7 mg,
68%): Rf 0.38 (CH2Cl2/CH3OH, 3:1); [R]D +141.1 (c 1.3, CH3-
1
OH); H NMR (800 MHz, DzO, δ) 5.09 (d, 1 H, J ) 1.9 Hz),
4-O-(r-D-Ar a bin ofu r a n osyl)-1,5:3,6-d ia n h yd r o-D-ga la c-
titol (5) a n d 2-O-(r-D-Ar a bin ofu r a n osyl)-1,5:3,6-d ia n h y-
d r o-D-ga la ctitol (6). Diol 15 (50 mg, 0.342 mmol), thioglyco-
side 18 (233 mg, 0.41 mmol), and a stir bar were dried
overnight with P2O5 in vacuo in a round-bottom flask. The
flask was cooled to 0 °C, and then 10:3 CH2Cl2/acetonitrile (6.5
mL) was added. Crushed 4 Å molecular sieves were added,
and the mixture was stirred for 20 min before NIS (115 mg,
0.51 mmol) and AgOTf (17 mg, 0.07 mmol) were added. After
the mixture was stirred for 5 min, triethylamine (5 drops) was
added, and the reaction mixture was then filtered through
Celite and diluted with CH2Cl2. The organic layer was washed
successively with a saturated aqueous solution of Na2S2O3,
water, and brine, before being dried and concentrated. Chro-
matography (hexanes/EtOAc, 1:1) yielded an inseparable
mixture of the two disaccharides. This oil was then dissolved
in CH3OH (5 mL), 1 M sodium methoxide (5 drops) was added,
and the solution was stirred for 14 h. The solution was then
neutralized with Amberlite IR 120 (H+) resin and filtered and
the solvent evaporated. The deprotected disaccharides were
still inseparable by chromatography, so the mixture was
dissolved in pyridine (1 mL) and acetic anhydride (1 mL). The
reaction mixture was stirred for 6 h and cooled to 0 °C, and
then CH3OH was added. The solvent was evaporated, and the
products were separated by chromatography (hexanes/EtOAc,
3:2). Each pure, fully acetylated disaccharide was dissolved,
separately, in CH3OH (5 mL), and 1 M sodium methoxide (3
drops) was added. After being stirred for 2 h, the solution was
neutralized with Amberlite IR 120 (H+) resin, filtered, and
concentrated. Each disaccharide was obtained in 20% yield (18
mg) from 15.
5.00 (d, 1 H, J ) 1.7 Hz), 4.55 (d, 1 H, J ) 5.6 Hz), 4.49 (dd,
1 H, J ) 1.8, 2.2 Hz), 4.43 (d, 1 H, J ) 1.8 Hz), 4.16 (d, 1 H,
J ) 10.8 Hz), 4.04 (dd, 1 H, J ) 1.7, 3.5 Hz), 4.03 (ddd, 1 H,
J ) 3.4, 5.8, 5.8 Hz), 3.99-3.97 (m, 2 H), 3.95 (dd, 1 H, J )
3.2, 10.9 Hz), 3.88-3.85 (m, 3 H), 3.78 (dd, 1 H, J ) 2.8, 13.4
Hz), 3.73 (dd, 1 H, J ) 2.4, 12.4 Hz), 3.73 (dd, 1 H, J ) 2.3,
12.4 Hz), 3.66 (d, 1 H, J ) 13.4 Hz), 3.63 (d, 1 H, J ) 12.4 Hz),
3.61 (d, 1 H, J ) 12.4 Hz); 13C NMR (150.9 MHz, D2O, δ) 108.1,
106.8, 84.5, 84.2, 81.8, 79.7, 77.0, 76.8, 76.7, 68.8, 63.9, 61.6,
61.5; HRMS (EI) m/z calcd for C16H26O12Na+ 433.1322, found
433.1324.
4-O-(r-D-Ar a b in ofu r a n osyl)-1,5:3,6-d ia n h yd r o-D-t a li-
tol (7), 2-O-(r-D-Ar a bin ofu r a n osyl)-1,5:3,6-d ia n h yd r o-D-
ta litol (8), a n d r-D-a r a bin ofu r a n osyl-(lf2)-[r-D-a r a bin o-
fu r a n osyl-(1f4)]-1,5:3,6-d ia n h yd r o-D-ta litol (10). Diol 16
(169 mg, 1.16 mmol), thiog1ycoside 18 (723 mg, 1.27 mmol),
and a stir bar were dried overnight with P2O5 in vacuo in a
round-bottom flask. The flask was cooled to 0 °C, and then
5:2 CH2Cl2/acetonitrile (14 mL) was added, followed by crushed
4 Å molecular sieves. The mixture was stirred for 20 min before
NIS (115 mg, 0.51 mmol) and AgOTf (17 mg, 0.07 mmol) were
added. After the mixture was stirred at 0 °C for 10 min,
triethylamine (5 drops) was added, and the reaction mixture
was then filtered through Celite and diluted with CH2Cl2. The
organic layer was washed successively with
a saturated
aqueous solution of NaZS2O3, water, and brine, before being
dried and concentrated. The three possible products were
separated by chromatography (hexanes/EtOAc, 2:1), dissolved
separately in CH3OH, and then treated with 1 M sodium
methoxide (5 drops). After being stirred for 2 h, each reaction
was then neutralized by addition of acetic acid. Chromatog-
raphy (CH2Cl2/CH3OH, 6:1) yielded the trisaccharide analogue
(82 mg, 17%) and the 2-linked disaccharide analogue (109 mg,
Data for 5: Rf 0.55 (CH2Cl2/CH3OH, 3:1); [R]D ) +137.5 (c
0.6, CH3OH); lH NMR (800 MHz, D2O, δ) 5.10 (d, 1 H, J ) 1.9
Hz), 4.48 (dd, 1 H, J ) 2.1 Hz), 4.40 (d, 1 H,J ) 1.8 Hz), 4.37

Oligofuranoside Synthesis and Conformational Analysis
J . Org. Chem., Vol. 67, No. 12, 2002 4163
34%). The 4-linked disaccharide analogue was also isolated,
but only in 6% yield.
12.3, 3.3 Hz), 3.74 (dd, 1 H, J ) 12.4, 3.2 Hz), 3.64 (dd, 1 H, J
) 12.4, 5. 7 Hz), 3.63 (dd, 1 H, J ) 12.5, 5.6 Hz), 3.36 (s, 3 H);
13C NMR (125.8 MHz, D2O, δ) 107.8, 105.3, 105.1, 90.8, 84.4,
84.3, 82.9, 81.8, 81.4, 79.0, 76.9, 76.2, 76.1, 65.8, 61.5, 61.5,
54.9; HRMS (EI) m/z calcd for C17H28O13Na+ 463.1422, found
463.1452.
Data for 7: Rf 0.49 (CH2Cl2/CH3OH, 3:1); [R]D +102.0 (c 0.8,
CH3OH); 1H NMR (800 MHz, D2O, δ) 5.10 (d, 1 H, J ) 1.7
Hz), 4.48 (dd, 1 H, J ) 2.3, 2.3 Hz), 4.33 (s, 1 H), 4.03 (d, 1 H,
J ) 10.8 Hz), 4.00 (ddd, 1 H, J ) 6.1, 6.1, 3.2 Hz), 3.99 (dd, 1
H, J ) 3.7, 1.8 Hz), 3.96 (d, 1 H, J ) 1.8 Hz), 3.95 (dd, 1 H, J
) 10.9, 3.1 Hz), 3.86-3.81 (m,3 H), 3.75 (dd, 1 H, J ) 12.4,
3.1 Hz), 3.63 (dd, 1 H, J ) 12.4, 5.8 Hz), 3.27 (dd, 1 H, J )
10.0, 8.3 Hz); 13C NMR (125.8 MHz, D2O, δ) 106.9, 84.3, 83.0,
81.9, 78.4, 77.0, 75.7, 69.1, 67.4, 63.7, 61.7; HRMS (EI) m/z
calcd for C11H18O8Na+ 301.0894, found 301.0921.
Data for 8: Rf 0.49 (CH2Cl2/CH3OH, 3:1); [R]D +161.8 (c 0.7,
CH3OH); 1H NMR (800 MHz, D2O, δ) 5.03 (d, 1 H, J ) 1.5
Hz), 4.37 (s, 1 H), 4.29 (dd, 1 H, J ) 2.3, 2.3 Hz), 4.01 (d, 1 H,
J ) 10.8 Hz), 3.99 (dd, 1 H, J ) 3.5, 1.6 Hz), 3.96 (dd, 1 H, J
) 10.8, 3.0 Hz), 3.92 (dd, 1 H, J ) 11.5, 6.7 Hz), 3.90-3.89
(m, 2 H), 3.81 (dd, 1 H, J ) 6.3, 3.5 Hz), 3.79 (ddd, 1 H, J )
9.5, 6.9, 1.1 Hz), 3.71 (dd, 1 H, J ) 12.4, 3.2 Hz), 3.58 (dd, 1
H, J ) 12.4, 5.9 Hz), 3.33 (dd, 1 H, J ) 11.4, 9.6 Hz); 13C NMR
(125.8 MHz, D2O, δ) 106.9, 84.2, 81.6, 81.6, 77.6, 76.9, 73.7,
73.0, 68.8, 62.6, 61.5; HRMS (EI) m/z calcd for C11H18O8Na+
301.0894, found 301.0922.
Data for 10: Rf 0.83 (CH2Cl2/CH3OH, 3:1); [R]D +140.0 (c
2.0, CH3OH); 1H NMR (800 MHz, D2O, δ) 5.09 (d, 1 H, J ) 1.7
Hz), 5.03 (d, 1 H, J ) 1.5 Hz), 4.51 (s, 1 H), 4.48 (dd, 1 H, J )
2.3, 2.3 Hz), 4.04 (d, 1 H, J ) 10.8 Hz), 4.00-3.97 (m, 4 H),
3.96-3.93 (m, 2 H), 3.89 (ddd, 1 H, J ) 6.1, 6.1, 3.2 Hz), 3.85-
3.83 (m, 2 H), 3.81 (dd, 1 H, J ) 6.3, 3.5 Hz), 3.74 (dd, 1 H, J
) 12.4, 3.1 Hz), 3.71 (dd, 1 H, J ) 12.3, 3.2 Hz), 3.61 (dd, 1 H,
J ) 12.3, 5.9 Hz), 3.58 (dd, 1 H, J ) 12.3, 5.9 Hz), 3.35 (dd, 1
H, J ) 11.4, 9.6 Hz); 13C NMR (125.8 MHz, D2O, δ) 107.0,
106.8, 84.2, 84.2, 81.7, 81.6, 80.9, 78.2, 76.9, 76.8, 75.6, 73.6,
69.1, 62.9, 61.5, 61.5; HRMS (EI) m/z calcd for C16H26O12Na+
433.1316, found 433.1278.
Meth yl r-D-Ar a bin ofu r a n osyl-(lf5)-2,6-a n h yd r o-3-C-
h yd r oxym eth yl-r-D-a r a bin ofu r a n osid e (12) a n d Meth yl
r-D-Ar abin ofu r an osyl-(lf3)-[r-D-ar abin ofu r an osyl-(lf5)]-
2,6-a n h yd r o-3-C-h yd r oxym e t h yl-r-D -a r a b in ofu r a n o-
sid e (13). Sulfoxide 19 (91 mg, 0.16 mmol) and crushed 4 Å
molecular sieves were dried overnight in vacuo in a round-
bottom flask. Di-tert-butylmethylpyridine (64 mg, 0.31 mmol)
was then added, and the mixture was dried for another 20
min. This mixture was then dissolved in CH2Cl2 (10 mL) and
the solution cooled to -78 °C. Triflic anhydride (32 mL, 0.19
mmol) was then added and the solution stirred for 5 min. Diol
17 (11 mg, 0.06 mmol) was then added as a solution in CH3-
CN (5 drops) and CH2Cl2 (3 mL). The solution was then stirred
for 5 h and allowed to warm slowly to room temperature, before
a saturated aqueous solution of NaHCO3 was added. The
reaction mixture was then diluted with CH2Cl2. The layers
were separated, and the organic layer was dried and concen-
trated to provide a residue that was chromatographed (hex-
anes/EtOAc, 3:1) to afford the separated products. These two
products were dissolved, separately, in CH3OH (5 mL) and
treated with 1 M sodium methoxide (5 drops) and the solutions
stirred for 4 h. Neutralization of the reaction mixture with
acetic acid (2 drops), concentration, and then chromatography
(CH2Cl2/CH3OH, 10:1) yielded the desired trisaccharide 13 (8
mg, 29%) as well as the 5-linked disaccharide 12 (7 mg, 37%).
Data for 12: Rf 0.41 (CH2Cl2/ CH3OH, 6:1); [R]D +133.3 (c
0.3, CH3OH); lH NMR (800 MHz, D2O, δ) 4.99 (d, 1 H, J ) 1.5
Hz), 4.98 (s, 1 H), 4.88 (s, 1 H), 4.79 (d, 1 H, J ) 7.9 Hz), 4.61
(dd, 1 H, J ) 7.9, 1.0 Hz), 4.08 (ddd, 1 H, J ) 7.1, 3.5, 1.0 Hz),
4.05 (dd, 1 H, J ) 3.3, 1.7 Hz), 4.05-4.01 (m, 2 H), 3.88 (dd,
1 H, J ) 6.1, 3.3 Hz), 3.76-3.74 (m, 2 H), 3.64 (dd, 1 H, J )
12.3, 5.8 Hz), 3.35 (s, 3H); 13C NMR (125.8 MHz, D2O, δ) 107.9,
105.7, 93.6, 84.4, 81.3, 79.4, 78.6, 78.0, 76.9, 65.6, 61.5, 54.9;
HRMS (EI) m/z calcd for C12H20O9Na+ 331.1000, found 331.1002.
Data for 13: Rf 0.36 (CH2Cl2/CH3OH, 3:1); [R]D +166.6 (c
0.5, CH3OH); lH NMR (800 MHz, D2O, δ) 5.22 (d, 1 H, J ) 2.1
Hz), 5.15 (s, 1 H), 5.00 (s, 1 H), 4.99 (d, 1 H, J ) 1.6 Hz), 4.80
(dd, 1 H, J ) 8.2, 1.2 Hz), 4.76 (d, 1 H, J ) 8.0 Hz), 4.21 (ddd,
1 H, J ) 7.1, 3.0, 1.0 Hz), 4.07-4.04 (m, 4 H), 4.02 (ddd, 1 H,
J ) 5.8, 5.8, 3.3 Hz), 3.90-3.86 (m, 3 H), 3.75 (dd, 1 H, J )
Meth yl 2,6-An h yd r o-5-O-ben zoyl-3-C-h yd r oxym eth yl-
r-D-a r a bin ofu r a n osid e (56). Diol 17 (16 mg, 0.09 mmol) was
dissolved in pyridine (1 mL) and the solution cooled to 0 °C.
Benzoyl chloride (0.12 mL, 1.02 mmol) was then added, and
the reaction mixture was stirred for 12 h. DMAP (catalytic)
was then added, and the solution was stirred for a further 4 h
before CH3OH was added. The reaction mixture was diluted
with CH2Cl2 and then washed successively with a saturated
aqueous solution of NaHCO3, water, and brine. Chromatog-
raphy (hexanes/EtOAc, 2:1) yielded monobenzoate 56 as a
white solid (23 mg, 87%): Rf 0.34 (hexanes/EtOAc, 1:1); [R]D
+72.0 (c 0.8, CHCl3); 1H NMR (500 MHz, CDCl3, δ) 8.10-8.08
(m, 2 H), 7.64-7.61 (m, 1 H), 7.51-7.48 (m, 2 H), 5.00 (s, 1
H), 4.91-4.90 (m, 2 H), 4.71-4.67 (m, 3 H), 4.26 (ddd, 1 H, J
) 5.7, 5.7, 0.8 Hz), 3.46 (s, 3 H), 3.31 (s, 1 H); 13C NMR (125.8
MHz, CDCl3, δ) 166.8, 133.4, 129.7, 129.4, 128.5, 106.0, 93.2,
79.8, 78.7, 77.5, 62.7, 54.7; HRMS (EI) m/z calcd for C14H16O6-
Na+ 303.0839, found 303.0853.
Met h yl r-D-a r a b in ofu r a n osyl-(lf3)-2,6-a n h yd r o-3-C-
h yd r oxym eth yl-r-D-a r a bin ofu r a n osid e (11). The sulfoxide
donor 19 (79 mg, 0.14 mmol) and crushed 4 Å molecular sieves
were dried overnight in vacuo in a round-bottom flask. Di-
tert-butylmethylpyridine (56 mg, 0.27 mmol) was then added
and the mixture dried for another 20 min. This mixture was
then dissolved in CH2Cl2 (10 mL) and the solution cooled to
-78 °C. Triflic anhydride (28 mL, 0.16 mmol) was then added,
and the reaction mixture was stirred for 5 min. Benzoate 56
(19 mg, 0.07 mmol) was then added as a solution in CH2Cl2 (3
mL). The reaction mixture was then stirred 5 h and allowed
to warm slowly to room temperature. Addition of a saturated
aqueous solution of NaHCO3, dilution with CH2Cl2, separation
of the layers, drying of the organic layer, and chromatography
(hexanes/EtOAc, 6:1) yielded a mixture of products. These
products were dissolved in CH3OH (5 mL), 1 M sodium
methoxide (5 drops) was added, and the reaction mixture was
stirred for 12 h. Neutralization with acetic acid (2 drops) and
chromatography (CH2Cl2/CH3OH, 10:1) yielded the crude
3-linked disaccharide. The crude disaccharide was dissolved
in pyridine (1 mL) and acetic anhydride (1 mL), and dimethyl-
aminopyridine (catalytic) was added. The reaction mixture was
stirred for 6 h before CH3OH was added, and the solution was
concentrated. Chromatography (hexanes/EtOAc, 2:1), followed
by Zemplen deacylation, and chromatography (CH2Cl2/CH3-
OH, 10:1) provided the pure disaccharide 11 in poor yield (2
mg, 9%): Rf 0.18 (CH2Cl2/CH3OH, 6: 1); [R]D +153.6 (c 0.1,
1
CH3OH); H NMR (800 MHz, D2O, δ) 5.23 (s, 1 H), 5.16 (s, 1
H), 4.98 (s, 1 H), 4.78 (d, 1 H, J ) 7.9 Hz), 4.69 (d, 1 H, J )
7.9 Hz), 4.08 (dd, 1 H, J ) 5.5, 5.5 Hz), 4.07-4.04 (m, 2 H),
3.89-3.88 (m, 3 H), 3.74 (dd, 1 H, J ) 12.4, 3.0 Hz), 3.62 (dd,
1 H, J ) 12.5, 5.8 Hz), 3.35 (s, 3 H); 13C NMR (125.8 MHz,
D2O, δ) 105.1, 105.1, 90.9, 84.4, 82.9, 81.8, 80.6, 76.2, 76.0,
61.5, 60.1, 54.7; HRMS (EI) m/z calcd for
C
12H20O9Na+
331.1000, found 331.0981.
Com p u ta tion a l In vestiga tion s. The SPMC search pro-
tocol available in MacroModel Version 6.536 was used to
generate an initial family of 1000 structures of 15-17. Each
conformer was then minimized in the gas phase using the
AMBER* force field. The conformers within 5 kcal/mol of their
respective global minimum (at the AMBER* level of theory)
were then optimized at the Hartree-Fock level of theory37 with
the 6-31G* basis set. The Cartesian coordinates for the
optimized conformers are given in the Supporting Information.
Single-point energy calculations on the HF conformers were
carried out at the B3LYP38 level of theory with the 6-31+G**
(36) (a) MacroModel V6.5: Mohamadi, F.; Richards, N. G. J .; Guida,
W. C.; Liskamp, R.; Lipton, M.; Caulfield, C.; Chang, G.; Hendrickson,
T.; Still, W. C. J . Comput. Chem. 1990, 11, 440. (b) Goodman, J .; Still,
W. C. J . Comput. Chem. 1991, 12, 1110.

4164 J . Org. Chem., Vol. 67, No. 12, 2002
Houseknecht and Lowary
basis set; these energies can be found in the Supporting
Information (Table S1). All HF and DFT calculations were
conducted using Gaussian 98.39
Ai, Bi) provided in that program for the R-D-arabinofuranosyl
ring (see example 7 in the PSEUROT manual). In the calcula-
tions the puckering amplitude, τm, was kept constant at 39°;
this value corresponds to the puckering observed in the crystal
structure of 1.41 This approach has previously been used
successfully in these calculations in other systems.42
NMR a n d P SEUROT Ca lcu la tion s. The 3J H,H values used
in all PSEUROT calculations were measured from 1D 1H NMR
spectra recorded on samples at 15-20 mM concentration in
0.6 mL of D2O (pH 6.0) at 300 K. These data are included in
Table S2 in the Supporting Information. Where necessary,
assignments were confirmed by 2D 1H-1H correlation spec-
troscopy (COSY, TOCSY) and by simulation of the spectra by
NMRSim.40 All PSEUROT 6.2 calculations were done using
the default parameters (e.g., electronegativities, phase angles,
Ack n ow led gm en t. This work was supported by the
National Science Foundation (Grant CHE-9875163) and
the Ohio Supercomputing Center. J .B.H. is supported
as a graduate research fellow by an NIH Training Grant
for Chemistry at the Biology Interface. We thank
Christopher S. Callam and Dr. Christopher M. Hadad
for helpful discussions.
(37) Hehre, W. J .; Radom, L.; Schleyer, P. v. R.; Pople, J . A. Ab initio
Molecular Orbital Theory; Wiley-Interscience: New York, 1986; see
also references therein.
Su p p or t in g In for m a t ion Ava ila b le: 1H and 13C NMR
spectra of all new compounds, Cartesian coordinates for the
HF/6-31G*-optimized geometries of 15-17 (global minima), a
table containing relative energies, a table of 3J H,H values used
(38) (a)Becke, A. D. Phys. Rev. A 1988, 38, 3098. (b) Becke, A. D. J .
Chem. Phys. 1993, 98, 5648. (c) Lee, C.; Yang, W.; Parr, R. G. Phys.
Rev. B 1988, 37, 785.
(39) Frisch, M. J .; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J . R.; Zakrzewski, V. G.; Montgomery, J . A.,
J r.; Stratmann, R. E.; Burant, J . C.; Dapprich, S.; Millam, J . M.;
Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J .;
Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo,
C.; Clifford, S.; Ochterski, J .; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J . B.; Cioslowski, J .; Ortiz, J . V.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.;
Fox, D. J .; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; J ohnson, B.; Chen, W.;
Wong, M. W.; Andres, J . L.; Gonzalez, C.; Head-Gordon, M.; Replogle,
E. S.; Pople, J . A. Gaussian 98, Revision A.7; Gaussian, Inc.: Pitts-
burgh, PA, 1998.
1
in PSEUROT calculations, and a figure with 13C and H NMR
chemical shifts used in determining the structures of 5-14.
This material is available free of charge via the Internet at
http://pubs.acs.org.
J O011127P
(41) Evdokimov, A. G.; Kalb, A. J .; Koetzle, T. F.; Klooster, W. T.;
Martin, J . M. L. J . Phys. Chem. A 1999, 103, 744.
(42) (a) Church, T. J .; Carmichael, I.;Serianni, A. S. J . Am. Chem.
Soc. 1997, 119, 8946. (b) Hoffmann, R. A.; van Wijk, J .; Leeflang, B.
R.; Kamerling, J . P.; Altona, C.; Vliegenthart, J . F. G. J . Am. Chem.
Soc. 1992, 114, 3710.
(40) NMRSim Version 2.5, Bruker Analytik GmbH, Silberstreifen,
D-76287 Rheinstetten, Germany.