Programmable one-pot oligosaccharide synthesis

Article abstract of DOI:10.1021/ja982232s

In an effort to develop a broadly applicable approach to the facile one- pot synthesis of oligosaccharides, the reactivity of a number of p- methylphenyl thioglycoside (STol) donors which are either fully protected or have one hydroxyl group exposed has b

Full text of DOI:10.1021/ja982232s

734
J. Am. Chem. Soc. 1999, 121, 734-753
Programmable One-Pot Oligosaccharide Synthesis
Zhiyuan Zhang, Ian R. Ollmann, Xin-Shan Ye, Ralf Wischnat, Timor Baasov,† and
Chi-Huey Wong*
Contribution from the Department of Chemistry and the Skaggs Institute for Chemical Biology,
The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, California 92037
ReceiVed June 26, 1998
Abstract: In an effort to develop a broadly applicable approach to the facile one-pot synthesis of
oligosaccharides, the reactivity of a number of p-methylphenyl thioglycoside (STol) donors which are either
fully protected or have one hydroxyl group exposed has been quantitatively determined by HPLC. We have
characterized and quantified the influence on reactivity of the structural effects of different monosaccharide
cores and different protecting groups on each glycoside donor. In addition, we have established a correlation
between glycosyl donor reactivity and the chemical shift of the anomeric proton by 1H NMR. Using the reactivity
data, we have created a database of thioglycosides as glycosyl donors and demonstrated its utility in the easy
and rapid one-pot assembly of various linear and branched oligosaccharide structures. In addition, we have
developed the first computer program, OptiMer, for use as a database search tool and guide for the selection
of building blocks for the one-pot assembly of a desired oligosaccharide or a library of individual
oligosaccharides.
Introduction
-withdrawing character of the leaving group within a given class
of glycosyl donors (e.g., thioglycosides). As a result, even
though identical chemistry is performed at each glycosidic
coupling step, with proper planning a high degree of sequence
selectivity can be achieved between competing donors of the
same class, eliminating the need for protecting group manipula-
tion between coupling steps. The synthetic approach is designed
such that the choice of protecting groups on sugar components,5a-c
or the combination of protecting groups and anomeric substit-
uent,5d,e will lead to a decrease in donor reactivity over the
course of the synthetic sequence. The most reactive donor is
used for the nonreducing end, and an unreactive donor is used
for the reducing end of the given oligosaccharide target.
Carbohydrates are ubiquitous in biological systems, involved
in such important functions1 as inflammation,2 immunological
response,2d,3a-c metastasis,3d and bacterial and viral infection.1a
While the synthesis of peptides and oligonucleotides was
automated decades ago, there are no such general synthetic
procedures available for the construction of complex oligosac-
charides.4 The necessity for regio- and stereocontrol in glycoside
bond-forming processes often leads to laborious synthetic
transformations, tremendous protecting group manipulations, and
tedious intermediate isolations which complicate the overall
synthetic process and decrease synthetic efficiency.
To facilitate the rapid synthesis of oligosaccharides, a new
chemoselective glycosylation strategy, the “one-pot sequential
glycosylation”, has recently been developed.5 This approach is
based on the observation that a large disparity between the
reactivities of different glycosyl donors can be achieved simply
by varying the protecting groups and the electron-donating or
Using these procedures, multiple coupling steps were suc-
cessfully performed by many laboratories to generate various
lengths of complex oligosaccharides.5,6 This methodology is,
however, not generally applicable because of the lack of precise
reactivity values of useful glycosyl donors and acceptors.7 A
particularly desirable goal in this research is to establish a
generally applicable method that will allow the rapid synthesis
of desired oligosaccharides from designed monomeric building
blocks in a programmable and predictive manner. Such a method
† On sabbatical leave from the Department of Chemistry, Technion, Haifa,
Israel 32000.
(1) (a) Varski, A. Glycobiology 1993, 3, 97. (b) Sharon, N.; Lis, H. Sci.
Am. 1993, 268 (1), 74.
(2) (a) Phillips, M. L.; Nudelman, E.; Gaeta, F. C. A.; Perez, M.; Singhal,
A. K.; Hakomori, S.; Paulson, J. C. Science 1990, 250, 1130. (b) Lasky, L.
A. Science 1992, 258, 964. (c) Giannis, A. Angew. Chem., Int. Ed. Engl.
1994, 33, 178. (d) Varski, A. Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 7390.
(e) Yuen, C.-T.; Bezouska, K.; O’brien, J.; Stoll, M.; Lemoine, R.; Lubineau,
A.; Kiso, M.; Hasegava, A.; Bockovich, N. J.; Nicolaou, K. C.; Feizi, T. J.
Biol. Chem. 1994, 269, 1595.
(3) (a) Ryan, C. A. Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 1. (b) Meldal,
M.; Mouritsen, S.; Bock, K. In Carbohydrate Antigens; Garegg, P. J.,
Lindberg, A. A., Eds.; ACS Symposium Series 519; American Chemical
Society: Washington, DC, 1993; p 19. (d) Feizi, T. Curr. Opin. Struct.
Biol. 1993, 3, 701.
(4) For recent review articles on oligosaccharide synthesis, see: (a)
Paulsen, H. Angew. Chem., Int. Ed. Engl. 1990, 29, 823-839. (b) Banoub,
J. Chem. ReV. 1992, 92, 1167-1195. (c) Toshima, K.; Tatsuta, K. Chem.
ReV. 1993, 93, 1503. (d) Schmidt, R. R.; Kinzy, W. AdV. Carbohydr. Chem.
Biochem. 1994, 50, 21. (d) Danishefsky, S. J.; Bilodeau, M. T. Angew.
Chem., Int. Ed. Engl. 1996, 35, 1380.
(5) (a) Fraser-Reid, B.; Udodong, U. E.; Wu, Z.; Ottosson, H.; Merritt,
J. R.; Rao, C. S.; Roberts, C. Synlett 1992, 927. (b) Raghavan, S.; Kahne,
D. J. Am. Chem. Soc. 1993, 115, 1580. (c) Yamada, H.; Harada, T.;
Miyazaki, H.; Takahashi, T. Tetrahedron Lett. 1994, 35, 3979. (d) Yamada,
H.; Harada, T.; Takahashi, T. J. Am. Chem. Soc. 1994, 116, 7919. (e)
Chenault, H. K.; Castro, A. Tetrahedron Lett. 1994, 35, 9145. (f) Ley, S.
V.; Priepke, H. W. M. Angew. Chem., Int. Ed. Engl. 1994, 33, 2292. (g)
Grice, P.; Ley, S. V.; Pietruszka, J.; Priepke, H. W. M.; Walther, E. P. E.
Synlett 1995, 781. (h) Geurtsen, R.; Holmes, D. S.; Boons, G.-J. J. Org.
Chem. 1997, 62, 8145. (i) Grice, P.; Ley, S. V.; Pietruszka, J.; Osborn, M.
I.; Henning, W. M.; Priepke, H. W. M.; Warriner, S. L. Chem. Eur. J.
1997, 431.
(6) Green, L.; Hinzen, B.; Ince, S. J.; Langer, P.; Ley, S. V.; Warriner,
S. L. Synlett 1998, 4, 440.
(7) For quantitative analysis of the glycosylation reactivity of several
glycosyl donors using NMR, see the leading work: Douglas, N. L.; Ley S.
V.; Lucking, U.; Warriner, S. L. J. Chem. Soc., Perkin Trans. 1 1998, 51.
10.1021/ja982232s CCC: $18.00 © 1999 American Chemical Society
Published on Web 01/14/1999

Programmable One-Pot Oligosaccharide Synthesis
J. Am. Chem. Soc., Vol. 121, No. 4, 1999 735
Scheme 1. Strategy for One-Pot Synthesis of Linear and Branched Oligosaccharidesa
a X represents the p-methylphenylthio group.
Scheme 2
can be used in the rapid assembly of complex oligosaccharides
and may be further developed toward automation.
for the use of several different donor types in a cooperative
fashion as part of an orthogonal coupling strategy.14 We chose
NIS-TfOH as the main promoter for this study. It is a
stoichiometric reagent and suitable for a variety of thioglyco-
sides.
As a first step toward this goal, we report here a general
procedure for the quantitative measurement of relative reac-
tivities of various glycosyl donors and acceptors using HPLC.
Employing this methodology, up to 50 different donor and
acceptor molecules, comprising six different monosaccharide
skeletons and 11 commonly used protecting groups, have been
evaluated. For each structure, a relative reactivity value (RRV)
is determined. With this relative reactivity database in hand,
we have developed a general computer program compatible with
Macintosh computers which can search the database to identify
optimal combinations of glycosyl building blocks. This strategy
enables the automated design of a rapid, one-pot synthetic
protocol for the synthesis of linear and branched oligosaccha-
rides (Scheme 1).
Our approach to the quantification of glycosyl donor reac-
tivities is based on earlier works by Sinnot15 and Fraser-Reid
et al.,16 which employ a competition reaction between two
donors to determine the relative rates of reactivity between them.
The overall competition reaction is thought to proceed as shown
in Scheme 2. Two glycosyl donors A1 and A2 are forced to
compete in a reversible fashion for a common electrophile (E+),
forming activated intermediates B1 and B2. These intermediates
may collapse to the corresponding transient oxocarbenium
cations C1 and C2, which then rapidly react with the acceptor
molecule to give the products. Assuming that the rate-determin-
ing step is the step B f C, the overall reaction can be described
by an apparent second-order rate constant, rate ) kobs[E+][A].
This is related to the first-order rate equation (rate ) k[C]) by
the relation, kobs ) k/K. Our measured rate constants, kobs, are
relative second-order rate constants, not absolute quantities.
Measuring the rates as relative values allows us to escape rate
constant dependence on the identity of the activator. To the
extent that B f C remains the rate-determining step and the
(8) For a recent review on the synthesis and application of thioglycosides,
see: Garegg, P. J. AdV. Carbohydr. Chem. Biochem. 1997, 52, 179.
(9) (a) Konradsson, P.; Udodong, U. E.; Fraser-Reid, B. Tetrahedron
Lett. 1990, 31, 4313. (b) Veeneman, G. H.; van Leuwen, S. H.; van Boom,
J. H. Tetrahedron Lett. 1990, 31, 1331. (c) Lonn, H. Carbohydr. Res. 1989,
135, 105. (d) Garegg, P. J. Acc. Chem. Res. 1992, 92, 1167. (e) Fugedi, P.;
Garegg, P. J. Carbohydr. Res. 1986, 149, C9. (f) Veeneman, G. H.; van
Boom, J. H. Tetrahedron Lett. 1990, 31, 275. (g) Burkart, M. D.; Zhang,
Z.; Hung, S.-C.; Wong, C.-H. J. Am. Chem. Soc. 1997, 119, 11743.
(10) Pinkernell, U.; Karst, U.; Cammann, K. Anal. Chem. 1994, 66, 2599.
(11) Kahne, D.; Walker, S.; Cheng, Y.; Van Engen, D. J. Am. Chem.
Soc. 1993, 115, 1580.
Results and Discussion
General Strategy. As a first step to this investigation, the
following questions were examined: (1) what is the best choice
of leaving group at the anomeric center of glycosyl donor, (2)
what promoter system will be most appropriate for the coupling
reaction, which is sufficiently general to handle a wide variety
of glycosyl donors and acceptors, and (3) what methodology
should be used to rapidly and accurately assess the reactivities
of different glycosyl donors?
Among the commonly used leaving groups, we chose
thioglycosides8 as “universal building blocks” because they are
stable under most reaction conditions frequently used for the
construction of building blocks. They can be activated by various
conditions such as NIS-TfOH,9a,b MeOTf,9c DMTST,9e IDCP,9f
Selectfluor,9g and other promoters.8 The characteristic high UV
absorbency of the p-methylphenylthio (STol) group (for ex-
ample, for p-methylphenyl mercaptan10 in MeOH at λ ) 256
nm, ꢀ ) 5 × 104 M-1 cm-1) facilitates the precise quantification
of the reactivities of glycosyl donors. In addition, thiogylcosides
with one hydroxy group as acceptor as well as donor are easily
prepared, and thioglycosides are readily converted to other
glycosyl donors commonly used in oligosaccharide synthesis,
including sulfoxides,11 fluorides,9g,12 and halides.13 Finally,
accessible alternative glycosylation chemistries may later allow
(12) (a) Mukayama, T.; Murai, Y.; Shoda, S. Chem. Lett. 1981, 431. (b)
Nicolaou, K. C.; Dolle, R. D.; Papahatjis, D. P.; Randall, J. L. J. Am. Chem.
Soc. 1984, 106, 4189.
(13) (a) Sato, S.; Mori, M.; Ito, Y.; Ogawa, T. Carbohydr. Res. 1986,
155, C6. (b) Anderson, F. O.; Fugedi, P.; Garegg, P. J.; Nashed, M.
Tetrahedron Lett. 1986, 27, 3919.
(14) Ito, Y.; Ogawa, T. J. Am. Chem. Soc. 1994, 116, 12073.
(15) Sinnott, M. L. Chem. ReV. 1990, 90, 1171.
(16) Fraser-Reid, B.; Wu, Z.; Udodong, U. E.; Ottosson, H. J. Org. Chem.
1990, 55, 6068.

736 J. Am. Chem. Soc., Vol. 121, No. 4, 1999
Table 1. Relative Reactivity Values of Thioglycosidesa
Zhang et al.
a This table was constructed on the basis of the manner shown in Chart 1. The relative reactivity value (RRV) of the most unreactive compound,
acetylated mannoside, was defined as 1.0. Compounds are numbered as the decreased reactivity trend, and the RRVs are shown in parentheses.
mechanism is the same, this should help ensure that the relative
Chart 1a
rates measured here will remain useful for many different
activators and perhaps other donor types as well. This analysis
is also true for glycosyl donors with neighboring group
participation, which occurs after the rate-determing step.
Construction of the General Reactivity Database. Table
1 was constructed by comparison of reactivities between the
various glycosyl donors and a small set of reference compounds
(7, 14, 33, and 48). The relative reactivities between the four
reference compounds were also examined. Based on these data,
the relative reactivities of all 50 compounds were quickly
determined through simple multiplication of the relative rate
constant ratios with the reference compounds. Since these values
are relative, they are normalized such that the least reactive
donor in the library has a reactivity of 1.0 (Chart 1).
The individual RRVs which were used to make the table were
determined by a direct competition assay. Unlike the previous
works,7 however, we did not directly obtain relative rate ratios
from ratios of product formation. While it is possible to directly
measure a relative first-order rate constant by a competition
assay between two donors using a large excess of reagents A
a The arrows indicate the deactivation direction. Numbers on the
arrows are the experimental relative reactivity ratios, and the numbers
over activator E+ (Scheme 2), such that the donor (A)
concentrations do not change appreciably over the course of
the reaction, these methods did not immediately lend themselves
to HPLC analysis in our case. Since we followed the reaction
by disappearance of the donor rather than appearance of product
(see Experimental Section), arranging the experiment so that
pseudo-first order kinetics are observed is relatively different.
in parentheses are the calculated RRVs.
To enhance the sensitivity of the experiment, we chose to
allow a significant portion of the donors to be consumed. 1:1:1
stoichiometries of A1:A2:E+ in the presence of 5 equiv of
acceptor (methanol) were used. The integrated second-order rate

Programmable One-Pot Oligosaccharide Synthesis
J. Am. Chem. Soc., Vol. 121, No. 4, 1999 737
Table 3. Competition Reactions between Perbenzylated and
Peracetylated Thioglycosides of Different Monosaccharidesa
a Relative reactivity ratios are shown in parentheses. The reference
donors are in a box. The RRVs shown were normalized based on 48
(A) and 1 (B).
Figure 1. Example of the relative reactivity measurement between a
thioglycoside (Dx) and the reference compound (Dref) by using HPLC
(see Experimental Section for details). The intensity of absorbance at
256 nm for each remaining donor was measured to determine the
relative reactivity (kx/kref). [D]t, donor concentration at time t; (A)t, donor
absorbance at time t.
applicable to other promoters as well. We have taken advantage
of this fact, employing DMTST in a one-pot oligosaccharide
synthesis (see One-Pot Synthesis of Oligosaccharides, below)
as a solution to the problem of byproduct participation in the
glycosylation reaction when NIS is used as an activator.
Structural Effects of Monosaccharides on the Anomeric
Glycosylation Reactivity. The completion of Table 1 allows
us to examine the influence of stereochemistry and identity of
different substituents on the anomeric reactivity of various types
of glycosyl donors. From the results in Table 3, it is clear that
the most reactive sugar is fucose, followed by galactose, glucose,
and mannose (Table 3A). Perbenzylated fucose 50 is 4.2-fold
more active than perbenzylated galactose 48, and the latter is
6.4-fold more active than perbenzylated glucose 43. The higher
reactivity of the fucoside is probably because the -CH3 group
is more highly electron donating than the -CH2OBn group. The
reactivity order of different sugar cores appears to remain very
similar when the experiments are performed with different types
of protecting groups. For example, when peracetylated galac-
tose 14 (Table 3B) was competed against peracetylated glucose
5, the ratio of 5.3 in favor of galactose 14 was obtained. The
higher reactivity of the galactoside than the glucoside has been
reported in the study of hydrolysis of glycosyl halides18 and
glycosides.19 It presumably arises due to the stereo and inductive
effects and possible involvement of the axial 4-O lone-pair
electrons in the stabilization of the oxocarbenium ion intermedi-
ate.19 Based on the results in Table 3, the reactivity order of
monosaccharides appears to be fucose > galactose > glucose
> mannose.
Table 2. Accuracy Measurement of Calculated Relative
Reactivitiesa
a The relative reactivity ratios obtained from experiment are shown
without parentheses. The calculated ratios are shown in parentheses.
The arrow indicates the direction of decreased reactivity.
equations from Scheme 2 give the relationship between relative
rate constant ratios and donor concentrations, as shown in eq
1. A typical HPLC analysis of relative reactivities in a competi-
tion experiment is shown in Figure 1.
kobs,1 ln([A1,t]/[A1,o])
)
(1)
kobs,2
ln([A2,t]/[A2,o])
The internal self-consistency of these measurements was
determined by examining a series of relative rate constants
(Table 2). Comparing the experimental data and calculated data
for each pair shows that the average error is about 10%. This
accuracy should be sufficient for the use of these data library
for guiding one-pot oligosaccharide syntheses.
Since amino sugars are widely found in natural systems and
provide very distinct electronic and structural features, we
wanted these important systems to be included in our investiga-
tion. Initially, we selected the most abundant glucosamine and
(17) A precise measurement with DMTST is difficult as its efficiency
for the activation is lower than that of NIS; normally 4 equiv of DMTST
is used for the glycosylation and the reaction is much slower than that of
NIS. We observed that 2 equiv could be enough to complete a reaction
with 1 equiv of thioglycoside within a certain time. For the competition
reactions, the conditions were exactly the same as those with NIS-TfOH
(see general procedure in Experimental Section), except that DMTST was
taken as 2 mole equiv relative to the donors (each 1 mole equiv) and the
reaction was continued for 3 h. This measurement can only be performed
between donors with small reactivity differences.
In addition, we were interested in how different promoters
will affect the reactivities of glycosyl donors. We measured the
relative reactivities between two pairs of donors, 14 vs 11 and
7 vs 5, with two promoter systems, NIS-TfOH and DMTST.
The observed RRVs were approximately the same in both
promoter systems: k14/k
) 1.4 (NIS-TfOH) and 1.4-2.0
11
(DMTST), and k7/k5 ) 2.0 for NIS-TfOH and 1.3-1.7 for
DMTST.17 These preliminary results suggest that different
promoters do not dramatically affect the relative reactivities of
glycosyl donors. Thus, the RRVs in Table 1 may also be
(18) Paulsen, H. Angew. Chem., Int. Ed. Engl. 1982, 21, 155.
(19) Miljkovic, M.; Yeagley, D.; Deslongchamps, P.; Dory, Y. L. J. Org.
Chem. 1997, 62, 7597.

738 J. Am. Chem. Soc., Vol. 121, No. 4, 1999
Zhang et al.
Table 4. Effect of C-2 Substitution on the Reactivity of
Table 5. Positional Effects of Benzoyl versus Hydroxyl Group on
Thiogalactoside Donors: RRVs for 4,6-Obenzylidene Derivativesa
the Reactivities of Thiogalactoside Donorsa
a Competition reactions were performed between perbenzylated
galactoside 7 and each of the other compounds in this table. RRVs
were calculated on the basis of 7. The substituents of interest are shown
in bold.
a Competition reactions were performed between 2-O-benzylgalac-
toside 29 as a reference and each of the other compounds in this table.
The RRVs were calculated on the basis of 18. The focal substituents
for the study are shown in bold.
core (Table 4), for example, we can conclude that the deactivat-
ing power of different electron-withdrawing groups is in the
order of -N3 > -OClAc > -NPhth > -OBz > -OBn.
Effects of the Electron-Withdrawing Group Position. The
reactivity perturbation arising from the exchange of one
protecting group for another was found to be highly dependent
on the position of the protecting group around the glycosyl core,
as well as the identity of that core. The first effect can be
demonstrated by comparing the reactivity of sugars with a
benzoyl protecting group versus those with a hydroxyl group
at the same position (Table 5). Removal of the benzoyl group
from the 4-O position causes the largest increase in the rate of
reaction. This was, however, not seen for other pyranoses, for
which the protecting group at the 2-position has been found to
be more important.7 This might be explained by the participation
of 4-oxygen of galactose in the stabilization of the putative
cationic transition state.18 The observed stronger effect of C3-
OH relative to that of C2-OH might be explained by its close
association with the substituent at the 4-position. In addition to
the destabilization of the presumed cationic transition state
through inductive effects, benzoylation of the 3-position may
also diminish the participation of the lone-pair electrons of C-4
oxygen in the stabilization of the cationic transition state. From
Table 5, the observed influence of the positions of the benzoate
group is in the order of 4 > 3 > 2 > 6. These results are
different from those observed for the mannose system, in which
the order of 2 > 6 > 4 > 3 was determined.7 In the latter case,
except for the neighboring 2-position, the proximity of the
benzoate to the ring oxygen was found to be more important
than its distance from the anomeric position.
galactosamine cores. The phthalyl group keeps the above trend
of the reactivity difference between galactose and glucose
structures (the reactivity of N-phthalyl galactose 11 is 6.1-fold
higher than that of N-phthalyl glucose 4); the unusually high
reactivity of trichloroethyl oxycarbonyl (Troc) group protected
amines was very surprising. For example, the N-Troc-protected
glucosamine 21 (Table 3B) is 33.7-fold more reactive than the
corresponding N-phthalyl-protected glucosamine 4. Furthermore,
glucosamine 21 becomes even more reactive (5.5-fold) than the
corresponding galactosamine 11, in which the amine is protected
with a phthalyl group. This observation20 is particularly
important in terms of regulating the relative reactivities of
aminoglycoside donors for the assembly of a target compound
with a 2-amino sugar either at (or close to) the nonreducing
end or at (or close to) the reducing end.
Effect of Protecting Groups on the Reactivity of Thiogly-
coside Donors. The effect of substituents on the reactivity at
the anomeric center has been long recognized.21 Glycosyl donors
with ester protecting groups undergo glycosylation reactions
much more slowly than do the corresponding donors with ether
protection. The reason for this phenomenon is that the electron-
withdrawing character of the ester group destabilizes the putative
cationic transition state, leading to glycoside formation.21c
Recently, Ley and co-workers have provided the first systematic
quantitation of such an effect by comparing the reactivity
differences between benzoate and benzyl groups on rhamnosyl
and mannosyl donors.7 In our hands, this effect seems to be
one of the most pervasive and consistent determinants of
glycosyl reactivity.
The results presented in Table 5 also provide, for the first
time to our knowledge, a direct reactivity comparison between
the protecting group and the hydroxyl group in the same
position. Such data are of particular importance for the design
of highly chemoselective one-pot glycosylation sequences
(Scheme 1), in which the acceptor molecule in one step becomes
the donor in the following step.5 Therefore, the reactivity
measurements of glycoside donors that also contain one free
hydroxyl group are crucial for the successful accomplishment
of such a challenging task. For this purpose, we have examined
and broadly quantified the influence of various protecting groups
and monosaccharide type on the reactivity of monohydroxy
glycosides as donors (Tables 5-7).
The data presented here allow detailed examination of the
activation or deactivation of a series of glycosyl core structures
in conjunction with a number of different protecting groups at
the 2-position. Within the specific confines of the galactosyl
(20) The elevated reactivity of N-Troc-protected glucosamines might be
the result of the electron-donating property of the N-trichloroethoxycarbonyl
group to stabilize the intermediate as shown in the following tentative
mechanism. Research is underway to test the proposal.
It should also be noted that the reactivity of 4,6-benzylidene-
protected galactose derivatives that have a free 3-OH group
(structures in Table 6A) is influenced by C-2 protecting groups
in the same manner as when the 3-OH was protected. Thus,
the deactivating effects of different groups at the C-2 position
For an accelerated acetolysis of 2-AcNH glucoside in acetic acid, see:
Cocker, D.; Sinnott, M. L. J. Chem. Soc., Perkin Trans. 2 1976, 618.
(21) (a) Feather, M. S.; Harris, J. F. J. Org. Chem. 1965, 30, 153. (b)
Capon, B. Chem. ReV. 1969, 69, 407. (d) Fraser-Reid, B.; Boctor, B. Can.
J. Chem. 1969, 47, 393. (e) Paulsen, H.; Richter, A.; Sinnwell, H.; Stenzel,
W. Carbohydr. Res. 1974, 38, 312.

Programmable One-Pot Oligosaccharide Synthesis
J. Am. Chem. Soc., Vol. 121, No. 4, 1999 739
Table 6. Effect of a Series of Different Protecting Groups on the
Fortunately, in the context of the synthetic methodology
developed here, the glycosylation reactions that have the most
impact on the anomeric reactivity of the acceptor are those that
are the most electron withdrawingsthose that tend to come at
the very end of the synthetic route. The largest deactivation
will occur on the last donor. Since the reducing end is typically
expected to be an unreactive glycoside such as a methyl
glycoside, there is no competing reaction at the step when it is
appended to the oligosaccharide chain, and so deactivation is
not expected to be a problem at that step. Thus, the worst
relevant reactivity shifts due to glycosylation will occur upon
glycosylation of the third residue from the reducing end by the
fourth residue from the reducing end. In cases of syntheses of
tetra-, penta-, or hexasaccharides, the fourth residue from the
reducing end is expected to be reasonably active. Therefore, its
deactivating effect is not likely to be large, even in the worst
of cases for targets of this size. However, it should be noted
that, in extended synthetic schemes, it would likely be prudent
to leave a larger reactivity window between donors at the last
steps of the reaction sequence to account for glycosylation-
associated deactivation.
Ability of Monohydroxy Thioglycosides To Serve as Donorsa
a The competition reactions were performed between benzylidene
galactoside 33 as a reference and each of the compounds. The RRVs
shown were normalized on the basis of 34 (A) and 8 (B). The
substituents of interest are shown in bold.
are in the same order (-OClAc > -OBz > -OBn) as was
observed from the 3-OClAc-protected galactoside derivatives
in Table 5. However, the magnitude of the deactiVating effects
is significantly dependent on the identity of the other protecting
groups on the molecule. For example, while benzoate pro-
tection at the 2-position caused the galactoside 24 to be 1.6
times less reactive than the benzyl group in the same position
(29, Table 4), the same comparison between benzoate and benzyl
systems in 3-hydroxyl-free galactosides 42 and 49 (Table 6A)
gave a factor of 11.2. A similar inconsistency in the deacti-
vation power is seen for the pair 2-OAcCl (20) and 2-OBn (29)
when the 3-position was protected (5.9-fold difference, Table
4) or free (62.5-fold difference between 34 and 49 in Table
6A). Thus, there are not consistent quantifiable reactivity trends
which lend themselves to a simple numerical reactivity model.
Though there exist consistent qualitative trends (e.g., -OBn is
less deactivating than -OBz), the degree of this influence
appears to be highly dependent upon the structure of the rest of
the molecule. As a result, we have taken the approach of directly
measuring the relative reactivity of each donor that enters our
library.
Effect of Glycosylation on Donor Reactivity. Glycosylation
of a free hydroxyl has a slight deactivating effect on the
anomeric reactivity of the acceptor. This is significant, as for
the second and later steps, the reactivity of the growing
oligosaccharide reducing end generally has not been explicitly
characterized. Fortunately, the deactivating effect of glyco-
sylation is typically small in most cases relevant to the synthesis.
This issue was examined on a series of different structures, and
the results are shown in Table 7. In general, the results show
that glycosylation of the hydroxyl function of a given mono-
hydroxy sugar causes a small decrease in reactivity. Glyco-
sylation by a perbenzylated sugar causes a decrease in reactivity
by a factor of 1.1-2.3, depending on the position of glyco-
sylation, while glycosylation by a perbenzoylated sugar caused
a more significant decrease in reactivity (a factor of 9.4 for 2
f 10). A somewhat larger deactivating effect from perbenzyl-
ated sugars when glycosylating the 2-position of the acceptor
was observed in one case (a factor of 4.5 for 45 f 39). This
may be the result of significant steric effect on the anomeric
center. It seems that the degree of deactivation introduced by
the glycosyl structure varies with the reactivities of the glycosyl
donors. Less reactive glycosyl donors are more strongly
deactivating once the glycosyl linkage is formed.
Correlation between Chemical Shift of Anomeric Proton
NMR and ln(RRVs). During the study of thioglycoside
reactivity, we have observed that the chemical shift of the
anomeric proton of the more active perbenzylated thioglycoside
48 is remarkably farther upfield than the less active perben-
zoylated analogue 7 (δ ) 4.58 ppm vs δ ) 4.98 ppm).
Interestingly, when we examined some compounds which have
the same protecting group (-OBz) at the C-2 position on the
same core structure (e.g., galactose), we observed a good linear
correlation between the chemical shifts of the anomeric protons
and the corresponding ln(RRVs) (Figure 2). Compounds with
different protecting groups at the C-2 position do not show a
good correlation, probably due to different shielding effects or
other unknown mechanisms. Since the stability of the glyco-
sylation transition state appears to be largely influenced by the
electron density of the anomeric center, the chemical shift of
the anomeric proton might be used to rapidly determine the
anomeric reactivities in certain cases.
Prediction of Optimal Donor Reactivities. Notwithstanding
the appearance of unwanted side reactions and other effects
which detract from the yield of all glycosylation reactions, the
highest yielding glycosylation steps will be those which have
the greatest difference in anomeric reactivity between the donor
and the acceptor. Thus, a high-yielding multistep procedure is
one which will employ the largest reactivity difference at each
step. As a result, those donors which are destined to react in
the first coupling step should be drawn from the pool of the
most reactive donors available, and those which are destined
to react at the last coupling step should be drawn from the pool
of least reactive available donors. The logarithmic nature of eq
1 implies that there is a strongly diminishing return in yield to
be gotten from increasing the ratio of relative anomeric
reactivities of the donor and acceptor at a given step. Thus,
reducing the ratio of relative rates for one step so as to increase
it for another will result in a sizable loss in overall yield, unless
the ratio in the first step was larger than that of the second to
begin with. Thus, it should be evident that, for optimal yield,
the ratio of reactivities between donor and acceptor should be
the same (and as large as possible) for each step. The principle
of constant rate ratios allows us to predict that the optimum
relative reactivity value (kx) to be used at each step x in a
sequence of N steps is given by eq 2. k1 is the relative reactivity

740 J. Am. Chem. Soc., Vol. 121, No. 4, 1999
Zhang et al.
Table 7. Effect of Sugar Substitution on the Reactivity of a Monohydroxy Thioglycosidea
a RRVs are given in parentheses. The arrows indicate pairs of compounds to be compared.
in-house database to identify the 100 best sets of available
characterized oligosaccharide donors (see Figure 2 as an
example). For large sequences, a Monte Carlo search method
is also available, which simply tests combinations at random.
The frequency with which each donor appears at each step in
the list of the best 100 donor sets is tabulated, and the user is
presented with a list of donors at each step which tend to give
the highest overall yields (Figure 3). The chemist is then free
to make an educated choice as to which donors to use on the
basis of availability, cost, and knowledge of the optimal
mathematical relationships between donor reactivities as detailed
above. We believe this program will become very useful when
the number of building blocks is expanded.
One-Pot Synthesis of Oligosaccharides. With this database
and the computer program in hand, we then performed the one-
pot synthesis of several tri- and tetrasaccharides (Schemes 3
and 4). The building blocks were added sequentially in order
of their decreasing RRVs. Scheme 3 illustrates the synthesis of
linear trisaccharides 52, 53, and 54. The overall yield of each
trisaccharide synthesis is around 60%, which corresponds to
about 80% yield of each coupling step. Trisaccharide 52 contains
both R- and â-glycosidic linkages and was prepared with the
donor 48 being active and the acceptor 19 being inactive. Indeed,
according to their RRVs, perbenzylated 48 is 588-fold more
reactive than the benzoylated 19. However, the syntheses of
trisaccharides 53 (Scheme 3b) and 54 (Scheme 3c) could not
be performed without quantitative information on the RRVs of
the appropriate donor and acceptor molecules. In the synthesis
of 53, we first employed the relatively activating N-Troc
protecting group for 2-aminoglycosides. The 2-N-Troc-protected
thioglucoside 21 is 4.4-fold more active than the benzoylated
galactoside 13. This difference in RRVs was sufficient to
perform the one-pot synthesis of trisaccharide 53 in the overall
yield of 56%.
Figure 2. Relation between the chemical shifts of the anomeric protons
of thioglycosides and their lnRRVs.
constant of the donor to be used in the first step. kN is the relative
reactivity constant of the donor to be used in the Nth step.
(x-1)/(N-1)
kN
kx ) k1
(2)
( )
k1
Computational Approach. Since exactly the right donor
reactivity as predicted by eq 2 is not always available for every
step, we generated a computer program that can search the donor
library for the best sets of available donors, allowing for the
programmed assembly of oligosaccharides. The program, named
OptiMer, interfaces with a database of donors (data from Table
1) stored using the FileMaker Pro 4.0 (FileMaker Inc.) database
program. The database stores the type of sugar core that it
contains, the location of any unprotected hydroxyls that may
be glycosylated, and the R- or â-directing nature of the
substituent at the 2-position, which is used to predict the product
stereochemistry at the anomeric carbon. OptiMer considers all
this information when examining sequence combinations. In
addition, the database also stores a picture of the compound,
its name, and a reference to the literature where the preparation
of the donor is described. Once given a target sequence (e.g.,
DGalR1,4DGlcNâ1,6DGalâ1,4DGlc) the program searches the
The synthesis of trisaccharide 54 (Scheme 3c) demonstrates
the sequential construction of two R-glycosidic bonds in a one-
pot manner. The thioglycosides used in this synthesis are 50
and 49, and the difference in their RRVs is only 3.6-fold. To
increase the selectivity of the coupling step, the donor 50 was
used with 1.5 equiv, and the reaction was performed with the
promoter DMTST. The results in Scheme 3 demonstrate that
the RRVs database can be sufficiently applied not only for NIS-
TfOH-promoted reactions but also for the DMTST-promoted

Programmable One-Pot Oligosaccharide Synthesis
J. Am. Chem. Soc., Vol. 121, No. 4, 1999 741
Figure 3. Picture of the OptiMer program in operation. In the background on the left is a FileMaker Pro database which contains the donor library.
Upon hitting the “Run Search” button, OptiMer launches and then presents the user with a dialogue box requesting the target oligosaccharide
sequence (foreground). The results appear in the remaining window as a list of donor sequence “hits” accompanied by optimum yield calculations,
and then a list of the frequency of donor appearance at each step from the top 100 hits (for use with large libraries). The chemist may opt to either
take one of the best sequences found or select from the list of best donors at each position to design his own sequence on the basis of knowledge
of donor availability or other desired properties.
glycosylations and may further be utilized for other promoters
to activate thioglycosides.
the succinimide accumulates from step to step, it may become
especially problematic for the later steps, resulting in a
significant decrease in the overall yield of the target glycosyl-
ation product. One way to avoid such problems is to use a
combination of two promoter systems: DMTST and NIS-TfOH.
Although DMTST is less active than NIS-TfOH and is not
suitable for unreactive donors, it does not produce a nucleophile
as a side product. It can be used at the initial steps of one-pot
syntheses, which often contain very reactive thioglycosides. In
later steps, which contain less reactive thioglycosides for which
DMTST is no longer appropriate, the reaction can be promoted
by NIS-TfOH, thereby diminishing the overall buildup of
succinimide. This protocol was successfully employed for the
assembly of the branched tetrasaccharide 57 (Scheme 4b). This
work highlights the importance of further investigation into
thioglycoside activators with nonnucleophilic byproducts.
Based on the orthogonal protection-deprotection strategy
developed in our laboratory for the synthesis of a highly
branched oligosaccharide library,22 we have also demonstrated
the possibility of combination of orthogonal protection-
deprotection strategy and the one-pot sequential strategy in the
synthesis of branched oligosaccharide. This combination pro-
To demonstrate the application of the relative reactivity donor
database, we constructed two different tetrasaccharides, com-
pounds 55 and 57 (Scheme 4), in a one-pot manner. In the
synthesis of 55, we used the reactivity differences of 104 (48
f 27) and 5.9 (27 f 13) for the first two glycosylation steps,
respectively. While these reactivity differences were sufficient
to synthesize the target tetrasaccharide 55 in an overall 40%
yield (about 74% yield for each coupling step), it should be
noted that the efficiency of this synthesis is much lower than is
predicted by the OptiMer search (82%, see Figure 2). Two main
reasons for this high discrepancy between the theoretical and
experimental results are immediately apparent. One reason is
that, at this stage of the development, the OptiMer program does
not consider the possibility of the occurrence of side reactions
(e.g., decomposition reactions) and the deactivation introduced
by the newly formed glycosyl bond. Thus, the reaction yields
reported by OptiMer should be considered yield optima,
analogous to a 100% yield in a normal reaction.
The second main reason relates to the promoter system NIS-
TfOH. When using NIS-TfOH, an equimolar amount of
succinimide is generated in the reaction mixture. Although the
nitrogen of succinimide is a poor nucleophile, in the case of
highly unreactive acceptors, it can compete effectively for the
remaining pool of activated glycosyl donors. This is most
important when the hydroxyl of the sugar acceptor is very
hindered (e.g., 4-OH of glucose, 4-OH of galactose) or has a
poor nucleophilicity (e.g., due to electron-withdrawing substit-
uents in the vicinal position). Indeed, in several cases of the
preparation of the disaccharide building blocks in Table 1, we
have isolated and characterized the glycosyl succinimide
products.23 Moreover, since in the one-pot sequential synthesis
(23) For example, in the synthesis of the lactoside 9 (see scheme below),
in addition to the target disaccharide 9 (65%), the succinimide product 58
(15%) was also isolated as a byproduct.
The low reactivity of the 4-OH of glucoside 10 (RRV ) 8.4), coupled with
an increased reactivity of the transient intermediate cation derived from
the initial reaction of galactoside 48 (RRV ) 1776) with iodonium ion,
could explain the formation of 57.
(22) Wong, C.-H.; Ye, X.-S.; Zhang, Z. J. Am. Chem. Soc. 1998, 120,
7137.

742 J. Am. Chem. Soc., Vol. 121, No. 4, 1999
Zhang et al.
Scheme 3. One-Pot Synthesis of Trisaccharides 52-54
Using Designed Building Blocks with Known RRVs (in
Parentheses)
Scheme 4. One-Pot Synthesis of Tetrasaccharides 55 and 56
Using Designed Building Blocks with Known RRVs (in
Parentheses)
is the use of an orthogonally protected monosaccharide, such
as galactoside 5922 and 35 (see Experimental Section). This
sugar was selectively deprotected at each position to generate
the corresponding monohydroxy sugars 25, 36, 44, and 45,
carrying a free hydroxyl group at the 3-, 6-, 4-, and 2-position,
respectively. Each of these compounds was glycosylated by the
donor with higher RRV, to generate the corresponding disac-
charides; for example, 25 f 26, 36 f 30, 44 f 41, and 45 f
39. The observed disaccharides each can be further deprotected
selectively at the remaining three positions of the orthogonally
protected sugar portion to generate 4 × 3 ) 12 monohydroxy
disaccharides (e.g., 32 was prepared from 41 by selective
deprotection of a 3-p-methoxybenzyl group). These structures
can then be subjected to the one-pot assembly of tetra- or
pentasaccharides, as we have demonstrated here in the synthesis
of 57 (Scheme 4b) and is generally illustrated in Scheme 1.
Such an approach, with the ultimate aid of the relative reactivity
database, forms a basis for the construction of linear and
branched oligosaccharide libraries, and approximately 30 fully
protected oligosaccharides have been prepared using this
reactivity-based one-pot strategy (Table 8). Complete depro-
tection and selective deprotection of the library will generate a
larger library. This work is in progress in our laboratory and
will be reported in due course.
vides an efficient method for rapid assembly of highly branched
oligosaccharide libraries, as shown in the construction of the
branched tetrasaccharide 57 (Scheme 4b). We have prepared
some orthogonally protected disaccharides thioglycoside ana-
logues and studied their reactivities (Table 1). In the synthesis
of the branched tetrasaccharide 57, we first performed the
reaction between 48 and 32 in the presence of NIS-TfOH and
found that the reaction proceeds with very fast consumption of
the donor to give exclusively the succinimide galactoside.23 All
the starting acceptor 32 could be recovered from this reaction
mixture. We assume that the elevated reactivity of the donor
48 coupled with high steric hindrance of the secondary hydroxyl
in 32 leads to a very efficient competition of succinimide to
afford the byproduct.
To solve this problem, the reaction between 48 and 32 was
performed with the use of DMTST as a coupling reagent
(Scheme 4b). In these conditions, the reaction proceeded
smoothly, and the product, without the isolation, was further
treated with acceptor 56 in the presence of NIS-TfOH to afford
57 in a one-pot, two-step procedure (57% yield). Further
investigation is underway in order to construct 57 from the
corresponding monosaccharides in a one-pot manner by using
a similar combination of NIS and DMTST. In addition, the
strategy employed for the synthesis of tetrasaccharide 57
(Scheme 4b) might be a general strategy for the preparation of
branched oligosaccharides library. The key of such an approach
Conclusion
Here we present the results of our first trial study that attempts
to bring the practice of oligosaccharide synthesis into a
systematic and rapid procedure. To achieve this goal, we pursued
an approach based on the chemoselective glycosylation of single
acceptors in the presence of other less reactive donors. To bring
this concept into predictable and reproducible practice, we have

Programmable One-Pot Oligosaccharide Synthesis
J. Am. Chem. Soc., Vol. 121, No. 4, 1999 743
Table 8. Sequences of Protected Oligosaccharides Prepared by
Programmed One-Pot Synthesis
first computer program, OptiMer, that searches through the
database of characterized donors and identifies sets of monomer
donors which are likely to give the best yield. This database
can be rapidly expanded to accommodate other mono- and
oligosaccharide systems. The further development of this new
synthesis strategy will surely impact the rapid access to the
molecular diversity of carbohydrate structures to facilitate the
discovery of carbohydrate ligands and receptors, which in turn
paves a way to a deeper understanding of their biological
functions.
Experimental Section
General Methods. Chemicals used were reagent grade and were
used as supplied except where noted. Solvents were either dried by
standard procedures or used as purchased. Unless otherwise stated, all
reactions were performed under an argon atmosphere. All thioglycosides
used in this research are p-methylphenyl thioglycosides. Compounds
3,24a 4,24a 5,24b 6,24b 7,24c 14,24b 43,24a and 5124d are known structures
and were prepared according to the published procedures. All glyco-
sylation experiments were performed by using molecular sieves, which
were flame-dried right before the reaction under high vacuum.
Analytical thin-layer chromatography was performed using silica gel
60 F254 glass plates (Merck); compound spots were visualized by UV
light (254 nm) and/or by staining with a yellow solution containing
Ce(NH4)2(NO3)6 (0.5 g) and (NH4)6Mo7O24‚4H2O (24.0 g) in 6% H2SO4
(500 mL). Flash column chromatography was performed on silica gel
60 Geduran (35-75 µm, EM Science). 1H NMR spectra were recorded
on a Bruker AMX-400 or a Bruker AMX-500 instrument. Low- and
high-resolution mass spectra were recorded under fast atom bombard-
ment (FAB) conditions. HPLC measurements were performed on a
Hitachi HPLC D-7000 system equipped with UV detector L7400, pump
L7100, and software D-7000. All the experiments were performed by
using a normal-phase column, 86-100-D5 (Microsorb-MV, Rainin
Instrument Co. Inc.) with the solvent system hexanes/EtOAc.
General Procedure for Competition Experiments. A mixture of
two thioglycoside donors (0.01 mmol of each, Dref is the reference donor
and Dx is any other donor molecule), absolute MeOH (0.05 mmol),
and molecular sieves (AW-300) in dichloromethane (1.0 mL) was stirred
at room temperature for 10 min. An aliquot of this mixture (30 µL)
was taken and separately injected (10 µL for each injection) into the
HPLC in order to determine the baseline separation conditions, and
also to measure a coefficient (R) between the absorption (A) and the
concentration of the donor molecule [D], R ) A/[D]. A solution of 0.5
M NIS in acetonitrile (20 µL, 0.01 mmol) was added to the above
mixture, followed by addition of a solution of 0.1 M TfOH (10 µL,
0.001 mmol), and the reaction was left at room temperature for 2 h.
The mixture was diluted with dichloromethane (2 mL), filtered, washed
with saturated aqueous sodium thiosulfate containing 10% sodium
hydrogen bicarbonate, dried (Na2SO4), and concentrated to dryness.
The residue was dissolved in dichloromethane (1.0 mL), and the
concentrations of the remaining donors ([Dx] and [Dref]) were measured
by HPLC using the same conditions (10 µL for each injection) as
determined for the mixture before the addition of reagents. An example
of such an experiment is given in Figure 3.
developed a general methodology for the easy, rapid, and precise
measurement of the relative reactivities of almost any glycosyl
donor by HPLC. We employed this methodology to quantify
the relative reactivities of a series of glycoside donors for use
as off-the-shelf reagents in the one-pot synthesis. In addition
to allowing us to plan a series of one-pot sequential syntheses
of linear and branched oligosaccharides, this database of donors
and reactivities also revealed a wealth of structural factors which
contribute to the reactivity of the anomeric position. As observed
by others, the electron-withdrawing contribution of same
protecting groups about the ring influences reactivity. We also
found the identity of the sugar, the position of protecting groups,
and other important factors contributing to anomeric reactivity.
During this study, we have demonstrated that, among the
monosaccharide and disaccharide structures tested, the fucose
core is the most reactive, followed by galactose, glucose, and
mannose. For 2-aminoglycosides, we show that their reactivity
is significantly influenced by the protecting group at the
nitrogen. The -NTroc group has a large activiting effect
compared to the -NPhth or azide group. By utilizing the Troc
and Phth groups as a versatile protecting groups for the amine
function, we have demonstrated that the change of only a single
protection on the amine can be sufficient to regulate the
reactivities either of a given monosaccharides or between
different monosaccharide structures. The degree of deactivation
at the anomeric center by benzoyl protection (as compared to
the free hydroxyl at that position) was found to be the greatest
at the 4-position on the galactose core, with the overall order
of 4 > 3 > 2 > 6.
Synthesis of Building Blocks. Scheme 5 illustrates the representative
synthesis of certain building blocks found in Table 1.
Because the absolute influence of a given protecting group
could not be simply accounted for, we found it necessary to
determine the reactivity of each monomer separately. For this
purpose, we defined the relative reactivity value (RRV) as a
relative measure of reactivity for a given monosaccharide or
disaccharide. By measuring the RRVs of up to 50 different
mono- and disaccharide structures, we were able to create the
first database of donor reactivities and demonstrated its useful-
ness for rapid one-pot assembly of various oligosaccharide
structures. In addition, we observed the linear correlation
between the glycosyl donor reactivity and the chemical shift of
General Procedure for Glycosylation. To a cold (-40 °C) mixture
of glycosyl donor (1.5 equiv), acceptor (1.0 equiv), NIS (1.6 equiv),
and MS was added CH2Cl2 by syringe, the mixture was stirred for 20
min, and then TfOH (0.015 equiv) was added. The mixture was stirred
at -20 °C for 0.5 h and then gradually raised to room temperature.
The reaction was monitored by TLC. After the reaction was finished,
solid Na2SO3, NaHCO3, and a few drops of water were added into the
reaction mixture, and the mixture was stirred for 5 min, diluted with
(24) (a) Balavoine, G.; Berteina, S.; Gret, A.; Fischer, J.; Lubineau, A.
J. Carbohydr. Chem. 1995, 14, 1217. (b) Magnusson, G. J. Org. Chem.
1977, 42, 913. (c) Burkart, M. D.; Zhang, Z.; Huang, S.-C.; Wong, C.-H.
J Am. Chem. Soc. 1997, 119, 11743. (d) DeNinno, M. P.; Etienne, J. B.;
Duplantier, K. C. Tetrahedron Lett. 1995, 36, 669.
1
the anomeric proton by H NMR. To aid in the planning of
high-yielding syntheses with available reagents, we report the

744 J. Am. Chem. Soc., Vol. 121, No. 4, 1999
Scheme 5. Preparation of Building Blocksa
Zhang et al.
a Conditions: (a) NaHCO3, MeOH/H2O; (b) BzCl, pyridine; (c) NH2NH2/AcOH; (d) NIS-TfOH, MS, CH2Cl2; (e) TFA, CH2Cl2; (f) HF/pyridine,
THF; (g) HO(CH2)5CO2Me, BF3/OEt2, CH2Cl2; (h) NaOMe, MeOH; (i) p-MeOBn(OMe)2, CAS, MeCN; (j) NaH, BnBr, DMF; (k) NaCNBH3-
CF3CO2H, DMF.
CH2Cl2, filtered, washed with saturated NaHCO3 and brine, dried over
Na2SO4 (or MgSO4), and concentrated. The residue was chromato-
graphed to give the named compounds.
(250 mL), washed with saturated aqueous NaHCO3 and brine, dried
over Na2SO4, filtered, and concentrated. The residue was chromato-
graphed (hexane/EtOAc, 4:1) to give 2 (3.06 g, 96%): [R]22 +37.8°
D
1
p-Methylphenyl 2,3,4,6-Tetra-O-acetyl-1-thio-r-D-mannopyran-
oside (1). To a mixture of D-mannose pentaacetate (16.0 g, 0.041 mol)
and p-thiocresol (7.6 g, 0.0615 mol) in CH2Cl2 (200 mL) was added
boron trifluoride diethyl etherate (6.8 mL, 0.0533 mol). The mixture
was stirred overnight. The reaction mixture was then diluted with
CH2Cl2 (200 mL) and washed with saturated NaHCO3 (2 × 120 mL)
and water (100 mL). The organic layer was dried over Na2SO4. The
solvent was removed, and the residue was purified by column
chromotography on silia gel (hexanes/EtOAc 3:1) to give product (16.2
g, 87%) as a syrup: 1H NMR (500 MHz, CDCl3) δ 7.38 (d, J ) 8.0
Hz, 2H), 7.12 (d, J ) 8.0 Hz, 2H), 5.49 (dd, 1H, J ) 2.5, 1.5 Hz),
5.42 (d, 1H, J ) 1.0 Hz), 5.30-5.35 (m, 2H), 4.54-4.58 (m, 1H),
4.30 (dd, 1H, J ) 12.5, 6.0 Hz), 4.11 (dd, 1H, J ) 12.0, 2.5 Hz), 2.33
(s, 3H), 2.15 (s, 3H), 2.08 (s, 3H), 2.06 (s, 3H), 2.02 (s, 3H); HRMS
(M + Cs) calcd for C21H26O9SCs 587.0352, found 587.0331.
p-Methylphenyl 2,3,6-Tri-O-benzoyl-4-O-(2,3,4,6-tetra-O-benzoyl-
â-D-galactopyranosyl)-â-D-glucopyranoside (2). Peracetylated lacto-
side 624b (2 g, 2.695 mmol) was dissolved in anhydrous methanol (25
mL), and a solution of 25% NaOMe in methanol (w/v, 0.5 mL) was
added dropwise. After being stirred at room temperature for 0.5 h, the
solution was neutralized with Amberlite HR-120 (H+), filtered, and
concentrated. The residue was dissolved in dry pyridine (10 mL) and
treated with excess benzoyl chloride (3 mL) and p-dimethylamino-
pyridine (100 mg) for 5 h at 50 °C. The mixture was evaporated to
dryness under reduced pressure, and the residue was diluted with CH2Cl2
(c, 0.5, CHCl3); H NMR (500 MHz, CDCl3) δ 2.23 (s, 3H), 3.69 (d,
2H, J ) 6.5 Hz), 3.88 (ddd, 1H, J ) 10.0, 4.5, 1.5 Hz), 3.93 (t, 1H, J
) 6.5 Hz, H4), 4.18 (t, 1H, J ) 9.5 Hz), 4.51 (dd, 1H, J ) 12.0, 5.0
Hz), 4.67 (dd, 1H, J ) 12.0, 1.5 Hz), 4.85 (d, 1H, J ) 10.0 Hz), 4.89
(d, J ) 8.0 Hz), 5.42 (dd, 1H, J ) 9.5, 1.5 Hz), 5.43 (dd, 1H, J )
10.0, 4.5 Hz), 5.74 (dd, 1H, J ) 10.0, 7.5 Hz), 5.76 (d, 1H, J ) 3.0
Hz), 5.82 (t, 1H, J ) 9.5 Hz), 6.90 (d, 2H, J ) 8.0 Hz), 7.11-7.61 (m,
24H), 7.73 (d, 2H, J ) 8.0 Hz), 7.9-8.0 (m, 12H); HRMS for
C68H56O17SCs (M + Cs) calcd 1309.2293, found 1309.2222.
p-Methylphenyl 3,6-Di-O-benzoyl-2-deoxy-2-phthalimido-1-thio-
â-D-glucopyranoside (8). p-Methylphenyl 2-deoxy-2-phthalimido-1-
thio-â-D-glucopyranoside24a (185 mg, 0.44 mmol) in CH2Cl2 (8 mL)
was treated with benzoyl chloride (203 µL, 1.76 mmol) and pyridine
(177 µL, 2.2 mmol) (see 2). The residue was purified by flash
chromatography (hexane/EtOAc, 3:1) to give the title compound (205
1
mg, 75%) as a white solid: [R]22 +107° (c, 0.1, CHCl3); H NMR
D
(400 MHz, CDCl3) δ 2.02 (s, 3H), 3.70 (s, 1H), 3.86 (t, 1H, J ) 9.4
Hz), 4.03 (m, 1H), 4.47 (t, 1H, J ) 10.4 Hz), 4.70 (dd, 1H, J ) 12.1,
5.1 Hz), 4.78 (dd, 1H, J ) 12.0, 4.0 Hz), 5.83 (d, 1H, J ) 10.4 Hz),
5.99 (dd, 1H, J ) 10.2, 9.1 Hz), 7.23-8.08 (m, 8 H); 13C NMR (400
MHz, CDCl3) δ 21.10, 53.44, 63.71, 70.01, 74.69, 78.21, 83.05, 123.54,
127.30, 128.29, 128.33, 128.63, 129.51, 129.69, 129.78, 129.83, 131.07,
131.43, 133.16, 133.40, 133.59, 134.13, 134.26, 138.29, 166.65, 166.87,
167.14, 167.89; HRMS for C35H29NO8SCs (M + Cs) calcd 756.0668,
found 756.0690.

Programmable One-Pot Oligosaccharide Synthesis
J. Am. Chem. Soc., Vol. 121, No. 4, 1999 745
p-Methylphenyl 2,3,6-Tri-O-Benzoyl-4-O-(2,3,4,6-tetra-O-benzyl-
â-D-galactopyranosyl)-â-D-glucopyranoside (9). The glycosylation of
donor 48 (151.2 mg, 0.234 mmol) and the acceptor 10 (100 mg, 0.167
mmol) was performed according to the general procedure described
above. Chromatography of the crude material gave the title compound
9 (121 mg, 65%) as white solid and succinimide derivative 58 (22 mg,
p-Methylphenyl 3,4,6-Tri-O-acetyl-2-deoxy-2-phthalimido-1-thio-
â-D-galactopyranoside (11). 1,3,4,6-Tetra-O-acetyl-2-deoxy-2-phthal-
imido-â-D-galactopyranose25 (1.7 g, 3.56 mmol) in CH2Cl2 (20 mL)
was treated with thiocresol (663 mg, 5.34 mmol) and BF3/Et2O (1.5
mL) according to the procedure described in the preparation of 1. After
workup, the crude product was crystallized from Et2O to give the title
compound 11 (1.62 g, 83%): mp 141-142 °C; [R]22D +45.4° (c, 0.5,
15% of the donor 48) as a byproduct. Data for 9: [R]22 +26.8° (c,
D
1
0.5, CHCl3); 1H NMR (500 MHz, CDCl3) δ 2.25 (s, 3H, Me of STol),
3.38 (m, 2H), 3.76 (dd, 1H, J ) 10.5, 4.0 Hz), 3.84 (dd, 1H, J ) 10.5,
2.5 Hz), 3.90-4.0 (m, 3H), 4.17 (t, 1H, J ) 9.5 Hz), 4.19 (d, 1H, J )
11.5 Hz), 4.23 (d, 1H, J ) 12.0 Hz), 4.28 (d, 1H, J ) 12.0 Hz), 4.41
(d, 1H, J ) 11.0 Hz), 4.56 (dd, 1H, J ) 12.0, 4.5 Hz), 4.59 (d, 2H, J
) 2.5 Hz), 4.77 (d, 1H, J ) 11.5 Hz), 4.87 (d, 1H, J ) 12.0 Hz), 4.94
(dd, 1H, J ) 12.0, 2.0 Hz), 4.99 (d, 2H, J ) 3.5 Hz), 5.36 (t, 1H, J )
10.0 Hz), 5.85 (t, 1H, J ) 7.5 Hz), 6.88 (d, 2H, J ) 8.0 Hz), 7.05-
7.52 (m, 35H), 7.63 (t, 1H, J ) 7.0 Hz), 7.92 (d, 2H, J ) 8.0 Hz),
7.95 (d, 2H, J ) 8.0 Hz), 8.02 (d, 2H, J ) 8.0 Hz); HRMS for
C68H64O13SCs (M + Cs) calcd 1253.3122, found 1253.3056. Anal.
Calcd for C68H64O13S: C, 72.94; H, 5.75; S, 2.86. Found: C, 72.94;
H, 5.60; S, 3.25. Data for 58: 1H NMR (500 MHz, CDCl3) δ 2.51-
2.64 (m, 4H), 3.42 (dd, 1H, J ) 9.5, 7.0 Hz), 3.49 (dd, 1H, J ) 9.5,
6.0 Hz), 4.03 (d, 1H, J ) 2.5 Hz), 4.36-4.49 (m, 5H), 4.55 (dd, 1H,
J ) 10.0, 3.0 Hz), 4.57 (d, 1H, J ) 9.0 Hz), 4.67 (d, 1H, J ) 11.5
Hz), 4.75 (d, 1H, J ) 11.5 Hz), 4.82 (d, 1H, J ) 11.5 Hz), 4.94 (d,
1H, J ) 11.5 Hz), 6.13 (d, 1H, J ) 7.5 Hz), 7.20-7.37 (m, 20H);
HRMS for C38H39O7NCs (M + Cs) calcd 754.1781, found 754. 1763.
CHCl3); H NMR (400 MHz, CDCl3) δ 1.97 (s, 3H), 2.05 (s, 3H),
2.08 (s, 3H), 2.31 (s, 3H), 4.09-4.24 (m, 3H), 4.62 (t, 1H, J ) 10.8
Hz), 5.49 (d, 1H, J ) 3.2 Hz), 5.64 (d, 1H, J ) 10.8 Hz), 5.80 (dd,
1H, J ) 10.8, 3.2 Hz),7.07 (d, 2H, J ) 8.1 Hz), 7.32 (d, 2H, J ) 8.1
Hz), 7.76-7.80 (m, 4H); 13C NMR (400 MHz, CDCl3) δ 20.52, 20.69,
21.14, 50.11, 61.63, 66.85, 68.81, 74.48, 84.35, 123.65, 129.62, 133.21,
134.40, 162.01, 163.10, 167.16, 167.90, 171.22; HRMS for C27H27-
NO9SCs (M + Cs) calcd 674.0461, found 674.0435.
p-Methylphenyl 3-O-Acetyl-6-O-benzoyl-2-deoxy-2-phthalimido-
1-thio-â-D-glucopyranoside (12). p-Methylphenyl 2-deoxy-2-phthal-
imido-1-thio-â-D-glucopyranoside24a (420 mg, 1.01 mmol) was treated
with camphorsulfonic acid (30 mg) and anisaldehyde dimethylacetale
(0.2 mL, 1.2 mmol) (see the preparation of 10). The residue in pyridine
(2 mL) was cooled to 0 °C, acetic anhydride (0.3 mL, 3 mmol) was
added, and the reaction was stirred for 1.5 h. After workup, the mixture
in HOAc/water (1:1, 20 mL) was heated at 60 °C for 30 min. After
cooling to room temperature, the reaction mixture was evaporated to
dryness. The residue was dissolved in CH2Cl2 (20 mL), washed with
saturated aqueous NaHCO3 and NH4Cl, and dried over MgSO4, and
the solvent was removed under reduced pressure. This material, without
further purification, was treated with pyridine (2 mL) and benzoyl
chloride (121 µL, 1.05 mmol) at 0 °C for 30 min and then evaporated
to dryness. The residue was purified by flash chromatography (hexane/
p-Methylphenyl 2,3,6-Tri-O-benzoyl-1-thio-â-D-glucopyranoside
(10). The title compound was prepared from peracetylated glucoside
524b by the following five-step synthesis:
(1) Compound 5 (2.2 g, 4.8 mmol) was treated with 1 M NaOMe in
methanol (0.5 mL) in anhydrous methanol (35 mL) by following the
procedure described in the preparation of 2.
EtOAc, 2:1) to give 12 (340 mg, 60% over four steps) as a white white
1
solid: [R]22 -8.6° (c, 0.5, CHCl3); H NMR (400 MHz, CDCl3) δ
D
1.88 (s, 3H), 2.24 (s, 3H), 3.51 (m, 1H), 3.71 (t, 1H, J ) 8.0 Hz), 3.96
(m, 1H), 4.27 (t, 1H, J ) 10.3 Hz), 4.63-4.75 (m, 2H), 5.76 (m, 1H),
5.80 (d, 1H, J ) 10.4 Hz), 6.94 (d, 1H, J ) 8.0 Hz), 7.32 (d, 2H, J )
8.1 Hz), 7.60-7.74 (m, 3H), 7.81-7.87 (m, 2H), 8.05-8.08 (?); 13C
NMR (400 MHz, CDCl3) δ 20.56, 21.06, 53.54, 63.64, 69.42, 74.07,
77.97, 82.72, 123.53, 127.13, 128.26, 128.33, 129.45, 129.60, 129.72,
129.78, 131.02, 131.46, 133.20, 133.59, 134.20, 134.37, 138.25, 166.67,
167.16, 167.90, 171.22; HRMS for C30H27NO8SClCs (M + Cs) calcd
694.0512, found 694.0537.
(2) The residue was dissolved in dry acetonitrile (20 mL), and
p-anisaldehyde dimethylacetal (1.3 g, 7 mmol) and camphorsulfonic
acid (100 mg) were added. After being stirred at room temperature for
2 h, the mixture was neutralized by addition of pyridine and then
evaporated to dryness.
(3) The residue was treated with excess benzoyl chloride (1.64 mL,
13.6 mmol) and 4-dimethylaminopyridine (0.5 g) in dry pyridine (10
mL) for 2 h at 50 °C (see 2). The residue was crystallized from ethyl
acetate and hexane to afford the 2,3-benzoyl-4,6-p-methoxybenzylydine
product as white crystals (2.4 g, 83% yield for three steps): mp 202-
203 °C; 1H NMR (500 MHz, CDCl3) δ 2.34 (s, 3H, Me of STol), 3.73
(ddd, 1H, J ) 14.0, 9.5, 4.5 Hz, H5), 3.75 (s, 3H, OMe), 3.83-3.88
(m, 2H), 4.43 (dd, 1H, J ) 10.5, 4.5 Hz), 4.96 (d, 1H, J ) 10.0 Hz,
H-1), 5.44 (t, 1H, J ) 9.5 Hz), 5.49 (s, 1H), 5.77 (t, 1H, J ) 9.5 Hz),
6.83 (d, 2H, J ) 9.0 Hz), 7.12 (d, 2H, J ) 8.0 Hz), 7.25-7.41 (m,
8H), 7.47 (t, 1H, J ) 7.5 Hz), 7.53 (t, 1H, J ) 7.5 Hz), 7.93 (d, 2H,
J ) 8.0 Hz), 7.97 (d, 2H, J ) 8.0 Hz); HRMS for C35H32O8SCs (M +
Cs): calcd 745.0872, found 745.0849.
p-Methylphenyl 2,3,4-Tri-O-benzoyl-1-thio-â-D-galactopyranoside
(13). Compound 16 (3.6 g, 5.06 mmol) was treated with the mixture
of CH3CN-(aq)HF (90:10, 100 mL) for 1 h. The solution was then
concentrated and the residue diluted with CH2Cl2 (300 mL), washed
with saturated NaHCO3 (2 × 150 mL), dried (Na2SO4), filtered, and
concentrated. The residue was purified by flash chromatography
(hexane/EtOAc, from 2:1 to 1:2) to give 13 (2.92 g, 96.8%): [R]22
D
+112.6° (c, 1, CHCl3); 1H NMR (400 MHz, CDCl3) δ 8.01-7.14
(19H), 5.83 (d, 1H, J ) 3.16 Hz, H-4), 5.77 (t, 1H, J ) 9.92 Hz, H-2),
5.67 (dd, 1H, J ) 3.16, 9.88 Hz, H-3), 4.968 (d, 1H, J ) 9.88 Hz,
H-1), 4.08 (t, 1H, J ) 6.68 Hz, H-5), 3,85 (dd, 1H, J ) 6.68, 11.88
Hz, H-6), 3.85 (dd, 1H, J ) 6.68, 11.88 Hz, H-6); 13C NMR (400
MHz, CDCl3) δ 166.47, 165.45, 165.13, 85.73, 77.76, 73.10, 68.91,
67.95, 60.6821.31; HRMS for C34H30O8Cs (M + Cs) calcd 731.0716,
found 731.0736.
(4) The above-mentioned benzylidene derivative (2.0 g, 3.27 mmol)
was suspended in 80% acetic acid/water (v/v, 25 mL), and the reaction
was stirred at 50 °C for 1 h. The mixture was evaporated to dryness.
The residue was dissolved in dry toluene (20 mL) and evaporated again
to dryness. This procedure was repeated three times.
p-Methylphenyl 3,4,6-Tri-O-benzoyl-1-thio-â-D-galactopyranoside
(15). A suspended mixture of p-methylphenyl 1-thio-â-D-galacto-
pyranoside26 (100 mg, 0.350 mmol) and Bu2SnO (266 mg, 1.04 mmol)
in toluene/benzene (1:1, 10 mL) was refluxed for 10 min. The mixture
was concentrated to a volume of 2 mL by distillation under Ar, and
then the hot solution (about 100 °C) was treated with benzoyl chloride
(130 µL, 1.12 mmol). After being stirred for 1 h at this temperature,
the reaction mixture was diluted with EtOAc (20 mL), cooled to room
temperature, filtered through Celite, and concentrated. Chromatography
(5) The crude residue was then dissolved in dichloromethane (20
mL), followed by addition of triethylamine (3 mL) and benzoyl chloride
(0.57 mL, 4.95 mmol). After 1 h at room temperature, the reaction
was quenched by addition of ethyl acetate and water, and the mixture
was evaporated to dryness. The product was crystallized from ethyl
acetate and hexane to afford 10 (1.5 g, 77%) as white crystals: mp
1
197 °C; [R]22 +56° (c, 0.5, CHCl3); H NMR (500 MHz, CDCl3) δ
D
2.28 (s, 3H), 3.46 (d, 1H, J ) 4.0 Hz), 3.84-3.86 (m, 2H), 4.72-4.76
(m, 2H), 4.89 (d, 1H, J ) 10.0 Hz, H-1), 5.37 (t, 1H, J ) 9.5 Hz),
5.47 (t, 1H, J ) 9.0 Hz), 6.96 (d, 2H, J ) 8.0 Hz), 7.31-7.40 (m,
6H), 7.47-7.54 (m, 4H), 7.63 (t, 1H, J ) 7.5 Hz), 7.93 (d, 2H, J )
8.0 Hz), 7.98 (d, 2H, J ) 8.0 Hz), 8.09 (d, 2H, J ) 8.0 Hz); HRMS
for C34H30O8SCs (M + Cs) calcd 731.0716, found 731.0689. Anal.
Calcd for C34H30O8S: C, 68.21; H, 5.05; S, 5.36. Found: C, 68.79; H,
5.06; S, 5.13.
of the residue (hexane/EtOAc, 4:1 to 2:1 to 1:1) furnished 15 (192
1
mg, 91.7%) as a white solid: [R]22 +15° (c, 0.5, CHCl3); H NMR
D
(25) Nilsson, U.; Ray, R. K.; Magnusson, G. Carbohydr. Res. 1990, 208,
260.
(26) Vargas-Berenguel, A.; Meldal, M.; Paulsen, H.; Jensen, K. J.; Bock,
K. J. Chem. Soc., Perkin Trans. 1 1994, 3287.

746 J. Am. Chem. Soc., Vol. 121, No. 4, 1999
Zhang et al.
(250 MHz, CDCl3) δ 5.91 (d, 1H, J ) 3.1 Hz, H-4), 5.43 (dd, 1H, J
) 3.3, 9.7 Hz, H-3), 4.720 (d, 1H, J ) 9.6 Hz, H-1), 4.29 (bt, 1H, J
) 6.5 Hz, H-5), 4.62 (dd, 1H, J ) 6.6 Hz, H-6) 4.37 (dd, 1H, J ) 5.4,
11.5 Hz, H-6′), 4,03 (t, 1H, J ) 9.6, 11.5 Hz, H-2), 2.38 (s, 3H, Me);
13C NMR (270 MHz, CDCl3) δ 165.98, 165.90, 165.29, 88.26, 74.97,
74.47, 68.59, 67.34, 62.45, 21.28; HRMS for C40H44O8SiCs (M + Cs)
calcd 845.1581, found 845.1609.
1H, J ) 3.4, 0.8 Hz), 4.39 (dd, 1H, J ) 12.4, 1.7 Hz), 4.45 (d, 1H, J
) 9.8 Hz), 4.84 (dd, 1H, J ) 10.8, 3.4 Hz), 5.47 (s, 1H), 7.05-7.07
(m, 2H), 7.37-7.40 (m, 5H), 7.60-7.62 (m, 2H); 13C NMR (400 MHz,
CDCl3) δ 21.25, 40.45, 58.08, 69.11, 69.38, 72.27, 76.68, 85.21, 100.92,
125.77, 126.42, 128.16, 129.28, 129.58, 134.72, 137.32, 138.35, 166.83;
HRMS for C22H22N3O5SClNa (M + Na) calcd 498.0866, found
498.0882.
p-Methylphenyl 2,3,4-Tri-O-benzoyl-6-O-tert-butyldimethylsilyl-
1-thio-â-D-galactopyranoside (16). A solution of p-methylphenyl
1-thio-â-D-galactopyranoside26 (4.0 g, 14.0 mmol) in DMF (10 mL)
was treated with imidazole (1.4 g, 20.6 mmol) and tert-butyldimethyl-
silyl chloride (2.26 g, 14.5 mmol) at 0 °C for 12 h. The mixture was
evaporated under reduced pressure, and the residue was purified by
flash chromatography (hexane/EtOAc, from 4:1 to 1:1) to give the
corresponding 6-O-tert-butyldimethylsilyl ether (5.3 g, 96%) as a syrup.
Part of this product (4.0 g, 10.0 mmol) was treated with pyridine (50
mL), an excess of benzoyl chloride (8 mL), and 4-dimethylamino-
pyridine (100 mg) at 50 °C for 1 h. After workup, the residue was
purified by chromatography on a silica gel column (hexane/EtOAc,
p-Methylphenyl 2,4,6-Tri-O-benzoyl-1-thio-â-D-galactopyranoside
(19). Compound 13 (300 mg, 0.502 mmol) was dissolved in the mixture
of pyridine/H2O (4:1, 3 mL) to which AgF (30 mg) was added, and
the mixture was stirred at 50 °C for 20 min. The reaction mixture was
concentrated under reduced pressure and the residue was chromatog-
raphied (hexane/EtOAc, 2:1) to give 19 (96 mg, 32%) as white
1
powder: [R]22 -10.4° (c, 0.5, CHCl3); H NMR (400 MHz, CDCl3)
D
δ 8.11-7.02 (19H), 5.77 (d, 1H, J ) 3.36 Hz, H-4), 5.26 (t, 1H, J )
9.64 Hz, H-2), 4.879 (d, 1H, J ) 9.88 Hz, H-1), 4.58 (dd, 1H, J )
7.24, 11.56 Hz, H-6), 4.45 (dd, 1H, J ) 5.44, 11.60 Hz, H-6), 3.85
(bt, 1H, J ) 6.44 Hz, H-5), 4.15 (dq, 1H, J ) 3.56, 6.28, 9.64 Hz,
H-3), 2.77 (d, 1H, J ) 6.24 Hz, OH), 2.34 (s, 3H, Me); HRMS for
C34H30O8SCs (M + Cs) calcd 731.0716, found 731.0737. Anal. Calcd
for C34H30O8S: C, 68.21; H, 5.05; S, 5.36. Found: C, 67.65; H, 4.96;
S, 5.45.
4:1) to afford the title compound 16 (6.96 g 97.7%) as a syrup: [R]22
D
1
+102° (c, 0.8, CHCl3); H NMR (400 MHz, CDCl3) δ 5.93 (d, 1H, J
) 3.2, H-4), 5.66 (t, 1H, J ) 9.84 Hz, H-2), 5.56 (dd, 1H, J ) 3.24,
9.96 Hz, H-3), 4.935 (d, 1H, J ) 9.80 Hz, H-1), 4.05 (bt, 1H, J ) 7.08
Hz, H-5), 3,85 (dd, 1H, J ) 5.92, 10.0 Hz, H-6), 3.85 (dd, 1H, J )
7.52, 10.0 Hz, H-6); 13C NMR (400 MHz, CDCl3) δ 165.52, 165.21,
165.09, 86.09, 73.32, 68.12, 67.95, 60.96, 25.77, 25.73, 21.34, -5.60,
-5.70; HRMS for C40H44O8SiCs (M + Cs) calcd 845.1581, found
845.1609.
p-Methylphenyl 4,6-O-Benzylidene-2,3-di-O-chloroacetyl-1-thio-
â-D-galactopyranoside (20). p-Methylphenyl 1-thio-â-D-galactopyrano-
side26 (3.5 g, 12.24 mmol) was suspended in CH3CN (30 mL) and
treated with benzaldehyde dimethylacetal (9.35 mL, 18.36 mmol) and
camphorsulfonic acid (100 mg). The mixture was stirred at room
temperature until all the starting material had been consumed (2 h).
After evaporation of the solvent, the crude product was crystallized
from MeOH to afford the corresponding 4,6-O-benzylidene derivative
(3.8 g, 83%). Part of this product (200 mg, 0.535 mmol) was dissolved
in CH2Cl2 (5 mL) and treated with pyridine (300 µL) and chloroacetyl
chloride (153 µL, 1.91 mmol). After usual workup, the crude product
was purified by chromatography (hexane/EtAc, 1:1) to give the title
p-Methylphenyl 4,6-O-Benzylidene-2-deoxy-2-phthalimido-1-thio-
â-D-glucopyranoside (17). p-Methylphenyl 2-deoxy-2-phthalimido-1-
thio-â-D-glucopyranoside24a (820 mg, 1.98 mmol) was treated with
benzaldehyde dimethylacetate (385 µL, 2.56 mmol) and camphorsul-
fonic acid (50 mg) in dry CH3CN (20 mL) according to the procedure
described for the preparation of 10. The residue was crystallized from
EtOH (-25 °C) to give 17 (780 mg, 79%) as white crystals: mp 121-
compound 20 (276 mg, 98%) as a white solid: [R]22 -1° (c, 0.5,
D
1
122 °C; [R]22 +34° (c, 0.5, CHCl3); H NMR (400 MHz, CDCl3) δ
CHCl3); 1H NMR (500 MHz, CDCl3) δ 7.49 (d, 2H, J ) 8.0 Hz), 7.09
(d, 2H, J ) 8.0 Hz), 5.45 (s, 1H), 5.24 (t, 1H, J ) 10.0 Hz), 5.08 (dd,
1H, J ) 3.5, 12.4 Hz), 4.68 (d, 1H, J ) 10.0 Hz), 4.4 (m, 2H), 4.11-
4.01 (m, 6H), 3.62 (s, 1H), 2.35 (s, 3H); HRMS for C24H24Cl2O7SCs
(M + Cs) calcd 658.9674, found 658.9692.
D
7.75-7.06 (13H), 5.624 (d, 1H, J ) 10.52 Hz, H-1), 5.60 (dd, 1H, J
) 2.96, 9.48 Hz, H-3), 4.39 (dd, 1H, J ) 4.76, 10.48 Hz, H-6), 4.30
(t, 1H, J ) 10.16 Hz, H-2), 3.81 (t, 1H, J ) 10.16 Hz, H-6), 3.67 (dd,
1H, J ) 4.88, 9.80 Hz, H-5), 3.58 (t, 1H, J ) 9.16 Hz, H-4), 2.29 (s,
3H, Me); 13C NMR (400 MHz, CDCl3) δ 101.93, 84.43, 81.88, 70.24,
69.71, 68.55, 55.57, 21.12; HRMS for C28H25NO6SCs (M + Cs) calcd
636.0457, found 636.0475.
p-Methylphenyl 2-Azido-4,6-O-benzylidene-3-O-chloroacetyl-2-
dexoy-1-thio-â-D-galactopyranoside (18). The title compound was
prepared from p-methylphenyl 2-dexoy-2-azido-1-thio-â-D-galacto-
pyranoside27 (527 mg, 1.70 mmol), benzaldehyde dimethylacetate (280
µL, 1.87 mmol), and camphorsulfonic acid (30 mg, 0.15 mmol) in dry
CH3CN (15 mL) as described above. Flash chromatography of the crude
residue (hexane/EtOAc, 2:1) gave the corresponding 4,6-O-benzylidene
derivative 18a (584 mg, 86%) as a white solid: 1H NMR (400 MHz,
CDCl3) δ 2.34 (s, 3H), 3.39 (s, 1H), 3.49 (t, 1H, J ) 7.6 Hz), 3.57 (m,
1H), 3.94 (dd, 1H, J ) 10.0, 1.2 Hz), 4.05 (d, 1H, J ) 3.5 Hz), 4.31
(d, 1H, J ) 9.5 Hz), 4.32 (m, 1H), 7.32 (d, 2H, J ) 8.1 Hz), 7.60-
7.74 (m, 5H), 7.81-7.87 (m, 2H); 13C NMR (400 MHz, CDCl3) δ
21.10, 61.70, 69.0, 69.5, 72.8, 79.2, 84.8, 101.1, 126.2, 126.5, 128.1,
129.2, 129.6, 134.5, 137.3, 138.5; HRMS for C20H21N3O4SNa (M +
Na) calcd 422.1150, found 422.1134.
p-Methylphenyl 3,4,6-Tri-O-acetyl-2-deoxy-2-(2,2,2-trichloro-
ethoxylcarbonylamino)-1-thio-â-D-glucopyranoside (21). The title
compound was prepared from 1,3,4,6-tetra-O-acetyl-2-deoxy-2-(2,2,2-
trichloroethoxylcarbonylamino)-â-D-glucopyranose28 (11.8 g, 22.5 mmol),
thiocresol (3.07 g, 24.75 mmol), and BF3/Et2O (10 mL) in dry CH2Cl2
(60 mL) by following the procedure described above (see 1). The crude
product was crystallized from ether to give 21 (11.8 g, 89%) as white
crystals: mp 175-176 °C; [R]22D +3° (c, 0.5, CHCl3); 1H NMR (400
MHz, CDCl3) δ 2.00 (s, 3H), 2.01 (s, 3H), 2.09 (s, 3H), 2.35 (s, 3H),
3.71-3.73 (m, 1H), 4.11-4.25 (m, 2H), 4.73 (d, 1H, J ) 12.0 Hz),
4.79 (d, 1H, J ) 10.8 Hz), 5.01 (t, 1H, J ) 9.8 Hz), 5.26 (d, 1H, J )
9.8 Hz), 5.31 (d, 1H, J ) 9.1 Hz), 7.11 (d, 2H, J ) 8.0 Hz), 7.41 (d,
2H, J ) 8.0 Hz); 13C NMR (400 MHz, CDCl3); δ 20.56, 20.61, 20.73,
21.15, 54.93, 62.25, 68.44, 73.14, 74.47, 75.70, 86.65, 95.36, 127.84,
129.71, 133.62, 138.70, 153.85, 169.45, 170.61; HRMS for C22H26-
NO9SCl3Na (M + Na) calcd 610.0266, found 610.0286. Anal. Calcd
for C22H26NO9SCl3: C, 45.03; H, 4.47; N, 2.39. Found: C, 45.46; H,
4.26; N, 2.30.
The above-mentioned 4,6-O-benzylidene derivative (212 mg, 0.53
mmol) was dissolved in dry pyridine (5 mL) and cooled to 0 °C. The
mixture was treated with chloroacetyl chloride (47.8 µL, 0.6 mmol),
and the reaction progress was monitored by TLC. After the mixture
was stirred for 20 min at room temperature, the solvent was removed
by evaporation under a high vacuum, and the residue was purified on
p-Methylphenyl 2,3,6-Tri-O-benzoyl-1-thio-â-D-galactopyranoside
(22). Compound 33 (200 mg, 0.342 mmol) was dissolved in a mixture
of TFA/CH2Cl2 (1:1, 5 mL) and stirred for 1 h. The reaction mixture
was diluted with CH2Cl2 (50 mL), washed with saturated NaHCO3,
dried over Na2SO4, filtered, and concentrated. The residue was further
treated with BzCl (56 µL, 0.479 mmol) and pyridine. After general
workup, the residue was chromatographed (hexanes/AcOEt, 3:2) to give
22 (153 mg, 85%): 1H NMR (500 MHz, CDCl3) δ 8.16-6.95 (19H),
5.80 (t, 1H, J ) 10.0 Hz, H-2), 5.38 (dd, 1H, J ) 3.0, 10.0 Hz, H-3),
4.920 (d, 1H, J ) 10.0 Hz, H-1), 4.66 (m, 2H, H-6), 4.40 (d, 1H, J )
a silica gel column (hexane/EtOAc, 2:1) to afford 18 (247 mg, 98%)
1
as a white solid: [R]22 -28° (c, 0.5, CHCl3); H NMR (400 MHz,
D
CDCl3) δ 2.34 (s, 3H), 3.56 (m, 1H), 3.81 (t, 1H, J ) 10.1 Hz), 4.00
(dd, 1H, J ) 12.4, 1.6 Hz), 4.08 (ABq, 2H, J ) 12.2 Hz), 4.36 (dd,
(27) Luening, B.; Norberg, T.; Tejbrant, J. Glycoconjugate J. 1989, 6,
5-19.
(28) Boullanger, P.; Jouineau, M.; Bouammal, B.; Lafont, D.; Descotes,
G. Carbohydr. Res. 1990, 202, 151.

Programmable One-Pot Oligosaccharide Synthesis
J. Am. Chem. Soc., Vol. 121, No. 4, 1999 747
2.50 Hz, H-4), 4.12 (bt, 1H, J ) 6.0 Hz, H-5), 2.27 (s, 3H, -Me); 13
NMR (500 MHz, CDCl3) δ 166.4, 165.8, 165.3, 87.1, 76.2, 67.9, 67.6,
63.4, 21.1; HRMS (M + Cs) calcd for C34H30O8SCs 731.0716, found
731.0736.
C
mg) and 43 (80.1 mg) as described in the general glycosylation
procedure to yield 93.1 mg (90%): 1H NMR (500 MHz, CDCl3) δ
7.98 (dd, J ) 8.0, 1.0 Hz, 2H), 7.63-7.69 (m, 4H), 7.16-7.45 (m,
29H), 7.00 (d, 2H, J ) 8.0 Hz), 6.79 (dd, 2H, J ) 8.0, 2.0 Hz), 5.66
(d, 1H, J ) 3.0 Hz), 5.46 (t, 1H, J ) 9.0 Hz), 5.16 (d, 1H, J ) 3.5
Hz), 4.81 (d, 1H, J ) 11.0 Hz), 4.69 (d, 1H, J ) 12.0 Hz), 4.59 (d,
2H, J ) 11.0 Hz), 4.54 (d, 1H, J ) 11.5 Hz), 4.46 (d, 1H, J ) 12.0
Hz), 4.36 (d, 1H, J ) 12.0 Hz), 4.20 (d, 1H, J ) 11.0 Hz), 4.05-4.07
(m, 1H), 3.80 (dd, 1H, J ) 9.5, 6.0 Hz), 3.72 (t, 1H, J ) 6.0 Hz),
3.63-3.69 (m, 4H), 3.39-3.44 (m, 2H), 3.35 (t, 1H, J ) 9.5 Hz), 3.30
(dd, 1H, J ) 10.5, 4.0 Hz), 2.39-2.45 (m, 1H), 2.23-2.33 (m, 6H),
1.95 (s, 3H), 1.06 (s, 9H); 13C NMR (125 MHz, CDCl3) δ 206.1, 171.6,
165.0, 138.6, 138.5, 138.4, 137.9, 137.9, 135.6, 135.6, 133.1, 133.0,
132.9, 129.8, 129.7, 129.7, 129.6, 129.5, 129.1, 128.6, 128.3, 128.2,
128.2, 127.9, 127.8, 127.7, 127.7, 127.6, 127.5, 127.4, 127.3, 127.3,
127.1, 93.32, 87.08, 81.41, 79.31, 77.96, 77.38, 75.45, 74.28, 73.91,
73.31, 73.17, 70.54, 68.26, 65.31, 62.22, 37.84, 29.66, 29.50, 27.98,
26.76, 21.09, 19.13; HRMS (M + Cs) calcd for C75H80O13SSiCs
1381.4143, found 1381.4077.
p-Methylphenyl 4,6-O-Benzylidene-2-deoxy-2-phthalimido-1-thio-
â-D-galactopyranoside (23a). p-Methylphenyl 2-deoxy-2-phthalimido-
1-thio-â-D-galactopyranoside (1.5 g, 3.91 mmol) in dry CH2Cl2 (15
mL) was treated with camphorsulfonic acid (30 mg, 0.15 mmol) and
benzaldehyde dimethylacetal (0.59 mL, 3.94 mmol). Flash chroma-
tography (hexane/AcOEt, 1:2) of the residue gave 23a (1.83 g, 93%)
as a white solid: 1H NMR (500 MHz, CDCl3) δ 2.35 (s, 3H), 3.70 (m,
1H), 4.06 (dd, 1H, J ) 12.0, 2.0 Hz), 4.12 (dd, 1H, J ) 14.5, 7.0 Hz),
4.28 (m, 1H), 4.39-4.51 (m, 3H), 5.56 (s, 1H), 5.61 (d, 1H, J ) 10
Hz, H-1), 7.07 (d, 2H, J ) 8.0 Hz), 7.39-7.48 (m, 7H), 7.72-7.87
(m, 4H); HRMS (M + Cs) calcd for C28H25O6NSCs 636.0457, found
636.0435.
p-Methylphenyl 4,6-O-Benzylidene-3-O-chloroacetyl-2dexoy-2-N-
phthalimido-1-thio-â-D-galactopyranoside (23). Compound 23a (580
mg, 1.15 mmol) was treated with chloroacetyl chloride (238 µL, 3.45
mmol) and pyridine (5 mL) at 0 °C according to the procedure described
above (see the preparation of 18). The residue was purified by flash
chromatography (hexane/EtOAc, 2:1) to give 23 (647 mg 97%) as a
white solid: [R]22D +10° (c, 0.5, CHCl3); 1H NMR (500 MHz, CDCl3)
δ 2.34 (s, 3H), 3.75 (s, 1H), 3.82-3.90 (ABq, 2H, J ) 15 Hz), 4.07
(dd, 1H, J ) 11.5, 1.5 Hz), 4.40 (dd, 1H, J ) 12.5, 1.5 Hz), 4.51 (d,
1H, J ) 3.5 Hz), 4.77 (t, 1H, J ) 10.5), 5.52 (s, 1H), 5.67 (d, 1H, J
) 10.4 Hz), 5.76-5.79 (dd, 1H, J ) 11.0, 3.5 Hz), 7.04 (d, 2H, J )
8.0 Hz), 7.38-7.46 (m, 7H), 7.74-7.77 (m, 2H), 7.84-7.89 (m, 2H);
13C NMR (500 MHz, CDCl3) δ 21.22, 40.47, 49.33, 69.63, 71.74, 72.57,
82.50, 101.03, 123.49, 123.68, 126.53, 126.57, 126.72, 128.15, 129.18,
129.53, 131.27, 131.46, 134.15, 134.28, 134.32, 137.41, 138.40, 166.79,
166.87, 168.30; HRMS (M + Cs) calcd for C30H26NO7SClCs 712.0173,
found 712.0196.
p-Methylphenyl 3,6-Di-O-benzoyl-2-deoxy-2-(2,2,2-trichloro-
ethoxylcarbonylamino)-1-thio-â-D-glucopyranoside (27). Compound
21 (2.0 g, 3.41 mmol) in MeOH (10 mL) was treated with a catalytic
amount of 1 M NaOMe for 5 h, and the solution was then neutralized
with IR-120 (H+), filtered, and concentrated. To a solution of the crude
product (610 mg, 1.324 mmol), pyridine (700 µL), and DMAP (50
mg) in CH2Cl2 (20 mL) was added BzCl (338 µL, 2.916 mmol). The
mixture was refluxed for 5 h under Ar and then concentrated and
coevaporated twice with toluene, diluted with CH2Cl2 (100 mL), washed
with 10% aqueous H2SO4, brine, and saturated aqueous NaHCO3, dried
over Na2SO4, filtered, and concentrated. Column chromatography
(hexane/AcOEt, 2:1 to 1:2) of the residue gave 27 (495 mg, 56%) as
a white solid; [R]22 +28.4° (c, 0.5, CHCl3); 1H NMR (500 MHz,
D
CDCl3) δ 5.50 (d, 1H, J ) 9.5 Hz, H-2), 5.41 (t, 1H, J ) 9.5 Hz, H-3),
4.844 (d, 1H, J ) 10.0 Hz, H-1), 4.74-4.64 (m, 3H), 4.55 (d, 1H, J )
12.0 Hz, OCH2CCl3), 3.90 (q, 1H, J ) 10.0 Hz, H-2), 3.79 (m, 2H,
H-5, H-4), 3.42 (d, 1H, J ) 5.0 Hz, OH), 2.27 (s, 3H, Me); HRMS (M
+ Cs) calcd for C30H28O8Cl3NSCs 799.9655, found 799.9680.
p-Methylphenyl 2-O-Benzoyl-3-O-(2,3,4,6-tetra-O-benzyl-r-D-glu-
copyranosyl)-4-O-levulinyl-1-thio-â-D-galactopyranoside (28). Com-
pound 28 was prepared from 26 (76.0 mg) as described for the
preparation of 36, yielding 48.3 mg (79%): 1H NMR (400 MHz,
CDCl3) δ 7.99 (d, J ) 7.5 Hz, 2H), 7.43 (t, J ) 7.4 Hz, 1H), 7.36 (d,
J ) 8.0 Hz, 2H), 7.20-7.31 (m, 20H), 7.09 (d, J ) 8.1 Hz, 2H), 6.84
(dd, J ) 7.5, 2.0 Hz, 2H), 5.48-5.52 (m, 2H), 4.98 (d, J ) 3.4 Hz,
1H), 4.84 (d, J ) 10.9 Hz, 1H), 4.74 (d, J ) 9.9 Hz, 1H), 4.66 (d, J
) 10.9 Hz, 1H), 4.61 (s, 2H), 4.58 (d, J ) 11.4 Hz, 1H), 4.42 (d, J )
12.0 Hz, 1H), 4.27 (d, J ) 12.0 Hz, 1H), 4.22 (d, J ) 11.3 Hz, 1H),
4.04-4.08 (m, 1H), 3.68-3.79 (m, 3H), 3.57-3.63 (m, 2H), 3.38-
3.44 (m, 2H), 3.15-3.23 (m, 2H), 2.64 (bs, 1H), 2.26-2.53 (m, 7H),
2.09 (s, 3H); 13C NMR (100 MHz, CDCl3) δ 206.73, 173.55, 165.07,
138.58, 138.44, 138.40, 138.23, 137.77, 133.13, 133.08, 129.77, 129.56,
128.73, 128.29, 128.25, 127.96, 127.80, 127.71, 127.60, 127.51, 127.43,
127.24, 127.14, 94.32, 87.00, 81.46, 78.96, 77.28, 75.44, 74.76, 74.24,
73.31, 73.15, 70.66, 69.04, 69.03, 67.90, 66.20, 60.21, 37.72, 29.63,
27.82, 21.11; HRMS (M + Cs) calcd for C59H62O13SCs 1143.2965,
found 1143.2920.
p-Methylphenyl 2-O-Benzoyl-4,6-O-benzylidene-3-O-chloroacetyl-
1-thio-â-D-galactopyranoside (24). Compound 42 (80 mg, 0.167
mmol) was treated with ClAcCl (27 µL, 0.335 mmol) and pyridine
(100 µL) in CH2Cl2 (2 mL). Chromatography (hexanes/AcOEt, 1:1) of
the residue gave 24 (87 mg, 94%) as a white powder: 1H NMR (500
MHz, CDCl3) δ 8.05-7.04 (14H), 5.54 (t, 1H, J ) 10.0 Hz, H-2),
5.49 (s, 1H, ArCH), 5.22 (dd, 1H, J ) 3.5, 10.0 Hz, H-3), 4.828 (d,
1H, J ) 10.0 Hz, H-1), 4.45 (d, 1H, J ) 3.5 Hz, H-4), 4.43 (d, 1H, J
) 11.0 Hz, H-6), 4.06 (d, 1H, J ) 11.0 Hz, H-6), 3.98 and 3.89 (dAB
,
1H each, J ) 15.0 Hz, ClCH2), 3.67 (bs, 1H, H-5), 2.34 (s, 3H, Me);
13C NMR (500 MHz, CDCl3) δ 167.2, 164.8, 101.2, 85.2, 75.0, 73.3,
69.6, 69.1,67.2, 40.6, 21.05; HRMS (M + Cs) calcd for C29H27O7-
ClCs 687.0220, found 687.0235.
p-Methylphenyl 2-O-Benzoyl-6-O-tert-butyldiphenylsilyl-4-O-
levulinyl-1-thio-â-D-galactopyranoside (25). To a stirred solution of
35 (1.1 g, 1.3 mmol) in CH2Cl2 (45 mL) was added trifluoroacetic
acid (8 mL) at -20 °C. After the solution was stirred for 20 min at
this temperature, methanol (8 mL) and CH2Cl2 (80 mL) were then added
to the reaction mixture, and the mixture was washed with saturated
NaHCO3 (3 × 30 mL) and brine (30 mL) and then dried over Na2SO4.
After removal of the solvent, the residue was purified by chromatog-
raphy on a silica gel column (hexanes/EtOAc, 2.5:1) to yield 25 (547
mg, 58%) as white solids: 1H NMR (500 MHz, CDCl3) δ 8.08 (d, J )
7.5 Hz, 2H), 7.66 (t, J ) 7.5 Hz, 4H), 7.59 (t, 1H, J ) 7.5 Hz), 7.37-
7.48 (m, 8H), 7.33 (d, 2H, J ) 8.0 Hz), 7.02 (d, 2H, J ) 8.5 Hz), 5.58
(d, 1H, J ) 3.5 Hz), 5.13 (t, 1H, J ) 10.0 Hz), 4.76 (d, 1H, J ) 10.0
Hz), 3.95 (dt, 1H, J ) 3.0, 8.5 Hz), 3.76-3.82 (m, 2H), 3.66-3.72
(m, 1H), 3.01 (d, 1H, J ) 8.0 Hz), 2.67-2.79 (m, 2H), 2.42-2.56 (m,
2H), 2.30 (s, 3H), 2.16 (s, 3H), 1.05 (s, 9H); 13C NMR (125 MHz,
CDCl3) δ 207.7, 172.0, 166.3, 138.1, 135.6, 135.6, 133.2, 133.1, 133.1,
133.0, 130.0, 129.8, 129.8, 129.7, 129.6, 128.9, 128.4, 127.7, 86.7,
77.83, 73.18, 71.58, 70.36, 61.97, 38.41, 29.68, 28.17, 26.75, 21.13,
19.11; HRMS (M + Cs) calcd for C41H46O8SSiCs 859.1737, found
859.1704. Anal. Calcd for C41H46O8SSi: C, 67.74; H, 6.38; S, 4.41.
Found: C, 68.10; H, 6.46; S, 4.91.
p-Methylphenyl 2-O-Benzyl-4,6-O-benzylidene-3-O-chloroacetyl-
1-thio-â-D-galactopyranoside (29). Compound 49 (200 mg, 0.431
mmol) was treated with chloroacetyl chloride (69 µL, 0.862 mmol)
and pyridine (2.0 mL). The residue was purified by chromatography
(hexanes/AcOEt, 1:1) to give 29 (229 mg, 98%) as a syrup: [R]22
D
+27.4° (c, 0.5, CHCl3); 1H NMR (400 MHz, CDCl3) δ 7.63-7.04
(14H), 5.48 (s, 1H, ArCH), 5.01 (dd, 1H, J ) 3.36, 9.64 Hz, H-3),
4.81 and 4.50 (d, 1H each, J ) 10.96 Hz, ArCH2), 4.642 (d, 1H, J )
9.5 Hz, H-1), 4.40 (dd, 1H, J ) 1.5, 12.5 Hz, H-6), 4.39-4.36 (m,
2H), 4.00-3.79 (m, 4H), 3.50 (bs, 1H, H-5), 2.36 (d, 1H, J ) 11.0
Hz, OH), 2.35 (s, 3H, Me); 13C NMR (500 MHz, CDCl3) δ 166.78,
100.75, 86.35, 77.03, 75.11, 73.60, 73.31, 69.07, 68.96, 40.50, 21.02;
HRMS (M + Cs) calcd for C29H29O6ClSCs 673.0428, found 673.0457.
Anal. Calcd for C29H29O6ClS: C, 64.38; H, 5.40. Found: C, 64.18; H,
5.32.
p-Methylphenyl 2-O-Benzoyl-3-O-(2,3,4,6-tetra-O-benzyl-r-D-glu-
copyranosyl)-6-O-tert-butyldiphenylsilyl-4-O-levulinyl-1-thio-â-D-
galactopyranoside (26). Compound 26 was prepared from 25 (60.0

748 J. Am. Chem. Soc., Vol. 121, No. 4, 1999
Zhang et al.
p-Methylphenyl 2-O-Benzoyl-6-O-(2,3,4,6-tetra-O-benzyl-â-D-
galactopyranosyl)-4-O-levulinyl-3-O-p-methoxybenzyl-1-thio-â-D-
galactopyranoside (30). Compound 30 was prepared from 36 (49.0
mg) and 48 (78.1 mg) as described in the general glycosylation
procedure (16.2 mg, 18%): 1H NMR (500 MHz, CDCl3) δ 7.97 (d, J
) 8.0 Hz, 2H), 7.60 (t, J ) 7.5 Hz, 1H), 7.46 (t, J ) 7.5 Hz, 2H),
7.19-7.36 (m, 22H), 6.95 (t, J ) 8.5 Hz, 4H), 6.57 (d, J ) 9.0 Hz,
2H), 5.50 (d, J ) 3.0 Hz, 1H), 5.32 (t, J ) 10.0 Hz, 1H), 4.94 (d, J )
12.5 Hz, 1H), 4.92 (d, J ) 11.5 Hz, 1H), 4.68-4.76 (m, 4H), 4.63 (d,
J ) 11.5 Hz, 1H), 4.46 (d, J ) 12.5 Hz, 1H), 4.45 (d, J ) 12.0 Hz,
1H), 4.41 (d, J ) 7.5 Hz, 1H), 4.40 (d, J ) 12.0 Hz, 1H), 4.21 (d, J
) 12.5 Hz, 1H), 3.76-3.92 (m, 5H), 3.70 (s, 3H), 3.50-3.60 (m, 5H),
2.57-2.75 (m, 4H), 2.20 (s, 3H), 2.13 (s, 3H); 13C NMR (125 MHz,
CDCl3) δ 206.3, 171.9, 165.1, 159.1, 138.8, 138.6, 138.5, 137.8, 137.6,
133.0, 132.8, 132.3, 129.9, 129.9, 129.6, 129.6, 129.5, 129.5, 129.3,
129.2, 128.4, 128.3, 128.3, 128.3, 128.2, 127.9, 127.8, 127.5, 127.4,
113.6, 104.0, 86.64, 82.06, 79.49, 76.82, 76.67, 75.08, 74.56, 73.49,
73.40, 73.24, 73.09, 70.42, 69.42, 68.57, 68.48, 67.05, 55.10, 38.14,
29.68, 28.13, 21.00; HRMS (M + Cs) calcd for C67H70O14SCs
1263.3541, found 1263.3604.
p-Methylphenyl 2-O-(2,3,4-Tri-O-benzyl-r-L-fucopyranosyl)-6-
O-tert-butyldiphenylsilyl-4-O-levulinyl-1-thio-â-D-galactopyrano-
side (31). Compound 31 was prepared from 39 (100 mg) as described
for the preparation of 25, yielding 89 mg (99%) of a syrup: 1H NMR
(500 MHz, CDCl3) δ 7.64 (t, J ) 7.5 Hz, 4H), 7.22-7.43 (m, 23H),
7.00 (d, J ) 8.0 Hz, 2H), 5.50 (d, J ) 3.0 Hz, 1H), 4.97 (d, J ) 3.0
Hz, 1H), 4.96 (d, J ) 12.0 Hz, 1H), 4.86 (d, J ) 12.0 Hz, 1H), 4.78
(d, J ) 12.0 Hz, 1H), 4.75 (s, 2H), 4.66 (d, J ) 12.0 Hz, 1H), 4.64 (d,
J ) 2.5 Hz, 1H), 4.52 (d, J ) 9.5 Hz, 1H), 4.28 (q, J ) 6.5 Hz, 1H),
4.04-4.09 (m, 1H), 3.97 (dd, J ) 10.0, 2.5 Hz, 1H), 3.83 (dt, J ) 9.0,
2.5 Hz, 1H), 3.63-3.75 (m, 4H), 3.57 (t, J ) 9.0 Hz, 1H), 2.58-2.70
(m, 2H), 2.51 (t, J ) 7.0 Hz, 2H), 2.29 (s, 3H), 2.12 (s, 3H), 1.20 (d,
J ) 6.5 Hz, 3H), 1.03 (s, 9H); 13C NMR (125 MHz, CDCl3) δ 206.3,
171.8, 138.5, 138.4, 137.4, 137.3, 135.6, 133.2, 133.1, 132.3, 129.9,
129.7, 129.6, 129.5, 128.6, 128.4, 128.3, 128.2, 128.0, 127.7, 127.7,
127.6, 127.4, 101.2, 86.6, 79.97, 79.80, 77.76, 77.38, 75.83, 74.73,
74.66, 74.17, 72.52, 69.44, 67.33, 62.08, 38.14, 29.78, 28.05, 26.74,
21.07, 19.10, 16.43; HRMS (M + Cs) calcd for C61H70O11SSiCs
1171.3462, found 1171.3423.
p-Methylphenyl 2-O-Benzoyl-4-O-(2,3,4-tri-O-benzyl-r-L-fuco-
pyranosyl)-6-O-tert-butyldiphenylsilyl-1-thio-â-D-galactopyranoside
(32). Compound 32 was prepared from 41 (48 mg) as described for
the preparation of 25, yielding 26.3 mg (61%): 1H NMR (500 MHz,
CDCl3) δ 8.12 (dd, J ) 8.0, 1.0 Hz, 2H), 7.65-7.68 (m, 4H), 7.57 (tt,
J ) 7.5, 1.0 Hz, 1H), 7.16-7.49 (m, 25H), 6.90 (d, J ) 8.0 Hz, 2H),
5.33 (t, J ) 10.0 Hz, 1H), 5.02 (d, J ) 11.5 Hz, 1H), 4.93 (t, J ) 12.5
Hz, 2H), 4.84 (d, J ) 11.5 Hz, 1H), 4.76 (d, J ) 12.0 Hz, 1H), 4.70
(d, J ) 3.5 Hz, 1H), 4.67 (d, J ) 12.0 Hz, 1H), 4.66 (d, J ) 10.0 Hz,
1H), 4.65 (d, J ) 11.5 Hz, 1H), 4.00 (dd, J ) 10.5, 3.5 Hz, 1H), 3.95
(d, J ) 3.0 Hz, 1H), 3.73-3.84 (m, 4H), 3.56-3.62 (m, 2H), 3.47 (d,
J ) 1.5 Hz, 1H), 2.19 (s, 3H), 1.01 (s, 9H), 0.79 (d, J ) 6.5 Hz, 3H);
13C NMR (125 MHz, CDCl3) δ 165.5, 138.6, 138.4, 137.5, 137.2, 135.6,
133.7, 133.2, 132.8, 130.4, 129.9, 129.7, 129.2, 128.8, 128.5, 128.4,
128.4, 128.4, 128.2, 128.1, 127.9, 127.7, 127.6, 102.0, 85.7, 80.98,
79.70, 79.25, 77.38, 76.35, 74.67, 74.36, 74.19, 73.14, 71.84, 67.30,
62.86, 29.65, 26.78, 21.10, 16.21; HRMS (M + Cs) calcd for C63H68O10-
SSiCs 1177.3357, found 1177.3397.
p-Methylphenyl 2,3-Di-O-benzoyl-4,6-O-benzylidene-1-thio-â-D-
galactopyranoside (33). p-Methylphenyl 4,6-O-benzylidene-1-thio-â-
D-galactopyranoside (1.0 g, 2.67 mmol) was treated with BzCl (2 mL)/
pyridine (3 mL) in CH2Cl2 (10 mL). Chromatography (hexane/AcOEt,
1:1) of the residue gave 33 (1.48 g, 95%): [R]22D +34° (c, 0.5, CHCl3);
1H NMR (500 MHz, CDCl3) δ 5.76 (t, 1H, J ) 10.0 Hz, H-2), 5.49 (s,
1H, ArCH), 5.34 (dd, 1H, J ) 3.5, 10.0 Hz, H-3), 4.902 (d, 1H, J )
10.0 Hz, H-1), 4.57 (d, 1H, J ) 3.0 Hz, H-4), 4.44, 4.08 (dd, 1H each,
J ) 1.5, 12.5 Hz, H-6), 3.75 (s, 1H, H-5), 2.34 (s, 3H, -Me); HRMS
(M + Cs) calcd for C34H30O7SCs 715.0767, found 715.0792. Anal.
Calcd for C34H30O7S: C, 70.09; H, 5.19; S, 5.50. Found: C, 70.05; H,
5.02; S, 6.06.
benzylidene-1-thio-â-D-galactopyranoside (2.5 g, 6.68 mmol), DCC
(1.65 g, 8.0 mmol), and DMAP (200 mg) in CH2Cl2 (40 mL) was added
levulinic acid (770 µL, 7.35 mmoL). The mixture was stirred at room
temperature for 3 h, and then pyridine (2.5 mL) and chloroacetyl
chloride (537 µL, 13.36 mmol) were added. After being stirred
overnight, the mixture was diluted with CH2Cl2 (150 mL), washed with
water, 10% H2SO4, water, and saturated aqueous NaHCO3, dried over
Na2SO4, filtered, and concentrated. The residue was chromatographed
(hexane/AcOEt, 2:1 to 1:1) to give p-methylphenyl 4,6-O-benzylidene-
2-O-chloroacetyl-3-O-levulinyl-1-thio-â-D-galactopyranoside (1.32 g,
36%), which was treated with NH2NH2 (1 mL)/AcOH (2 mL) in THF/
MeOH (10:1, 10 mL). The residue was concentrated, diluted with
CH2Cl2 (150 mL), washed with saturated aqueous NaHCO3, dried
(Na2SO4), filtered, and concentrated. Chromatography of the residue
gave 34 (1.03 g, 95%) as a white solid: [R]22D -17.6° (c, 0.5, CHCl3);
1H NMR (400 MHz, CDCl3) δ 7.50-7.10 (9H), 5.47 (s, 1H, ArCH),
5.01 (t, 1H, J ) 9.64 Hz, H-2), 4.579 (d, 1H, J ) 9.76 Hz, H-1), 4.36
(dd, 1H, J ) 1.48, 12.48 Hz, H-6), 4.17 (dd, 1H, J ) 1.0, 3.64 Hz,
H-4), 4.14 (s, 2H, CH2Cl), 3.73 (bs, 1H, H-3), 3.51 (d, 1H, J ) 1.08
Hz, H-5), 2.66 (bs, 1H, OH), 2.35 (s, 3H, Me); 13C NMR (500 MHz,
CDCl3) δ 166.41, 138.53, 137.20, 101.39, 84.29, 75.45, 72.32, 71.76,
69.75, 68.98, 40.84, 21.21; HRMS (M + Cs) calcd for C22H23O6ClSCs
473.0802, found 473.0817.
p-Methylphenyl 2-O-Benzoyl-6-O-tert-butyldiphenylsilyl-4-O-
levulinyl-3-O-p-methoxybenzyl-1-thio-â-D-galactopyranoside (35).
Compound 45 (1.97 g, 2.65 mmol) was treated with pyridine (50 mL)
and benzoyl chloride (1.2 mL, 10.6 mmol). The residue was subjected
to a column chromatography on silica gel (hexanes/EtOAc, 4:1) to give
35 (2.24 g, 100%) as a syrup: 1H NMR (500 MHz, CDCl3) δ 7.98 (d,
J ) 8.0 Hz, 2H), 7.65-7.68 (m, 4H), 7.60 (t, J ) 7.5 Hz, 1H), 7.37-
7.48 (m, 8H), 7.32 (d, J ) 8.0 Hz, 2H), 7.01 (d, J ) 9.0 Hz, 2H), 6.99
(d, J ) 8.0 Hz, 2H), 6.59-6.62 (m, 2H), 5.66 (d, J ) 3.0 Hz, 1H),
5.31 (t, J ) 10.0 Hz, 1H), 4.68 (d, J ) 10.0 Hz, 1H), 4.54 (d, J ) 12.5
Hz, 1H), 4.33 (d, J ) 12.5 Hz, 1H), 3.80-3.85 (m, 1H), 3.68-3.73
(m, 5H), 3.62 (dd, J ) 9.5, 3.0 Hz, 1H), 2.55-2.73 (m, 4H), 2.27 (s,
3H), 2.13 (s, 3H), 1.07 (s, 9H); 13C NMR (125 MHz, CDCl3) δ 206.1,
171.8, 165.1, 159.1, 137.9, 135.6, 135.6, 133.1, 133.0, 132.9, 129.9,
129.9, 129.8, 129.8, 129.6, 129.5, 129.3, 129.2, 128.3, 127.8, 127.7,
113.5, 87.0, 77.67, 76.87, 70.35, 69.62, 66.16, 61.99, 55.11, 38.18,
29.75, 28.18, 26.74, 21.08, 19.12; HRMS (M + Cs) calcd for C49H54O9-
SSiCs 979.2312, found 979.2349.
p-Methylphenyl 2-O-Benzoyl-4-O-levulinyl-3-O-p-methoxybenzyl-
1-thio-â-D-galactopyranoside (36). A mixture of 35 (330 mg, 0.39
mmol), HF/pyridine complex (3.5 mL), acetic acid (5.0 mL), and dry
THF (25 mL) was stirred for 11 h under argon. The reaction mixture
was diluted with EtOAc (150 mL) and then washed with saturated
aqueous NaHCO3 (3 × 30 mL) and brine (30 mL), dried over Na2SO4,
filtered, and concentrated under reduced pressure. The residue was
purified by chromatography on a silica gel column (hexanes/EtOAc,
1:1.5) to yield 36 (237 mg, 100%) as a syrup: 1H NMR (500 MHz,
CDCl3) δ 7.99 (dd, J ) 8.5, 1.0 Hz, 2H), 7.61 (t, J ) 7.0, 1.0 Hz, 1H),
7.47 (t, J ) 8.0 Hz, 2H), 7.32-7.34 (m, 2H), 7.06 (d, J ) 8.0 Hz,
2H), 7.00 (d, J ) 9.0 Hz, 2H), 6.59-6.62 (m, 2H), 5.53 (d, J ) 3.0
Hz, 1H), 5.38 (t, J ) 10.0 Hz, 1H), 4.70 (d, J ) 10.0 Hz, 1H), 4.53 (d,
J ) 12.0 Hz, 1H), 4.31 (d, J ) 12.5 Hz, 1H), 3.77-3.80 (m, 1H), 3.71
(s, 3H), 3.62-3.70 (m, 2H), 2.82-2.89 (m, 1H), 2.67-2.74 (m, 2H),
2.57-2.64 (m, 2H), 2.30 (s, 3H), 2.19 (s, 3H), 2.04 (d, J ) 0.5 Hz,
1H); 13C NMR (125 MHz, CDCl3) δ 206.9, 173.1, 165.1, 159.2, 138.1,
133.1, 129.9, 129.5, 129.1, 128.8, 128.2, 113.6, 86.86, 77.40, 76.64,
70.37, 69.67, 66.62, 60.51, 55.08, 38.00, 29.70, 28.01, 21.02; HRMS
(M + Cs) calcd for C33H36O9SCs 741.1134, found 741.1155.
p-Methylphenyl 3-O-Benzyl-4,6-O-benzylidene-2-O-levulinyl-1-
thio-â-D-galactopyranoside (37). p-Methylphenyl 3-O-benzyl-4,6-O-
benzylidene-1-thio-â-D-galactopyranoside (300 mg, 0.646 mmol) in
CH2Cl2 (5 mL) was treated with levulinic acid (120 µL, 0.97 mmol),
DCC (200 mg), and DMAP (50 mg) according to the procedure
described for the preparation of 34. The residue was purified by column
chromatography (hexanes/AcOEt, 1:1) to give 37 (345 mg, 95%) as a
white solid: [R]22D -2.6° (c, 0.5, CHCl3); 1H NMR (500 MHz, CDCl3)
δ 7.49-7.03 (m, 14 H), 5.42 (s, 1H), 5.26 (t, 1H, J ) 10.0 Hz, H-2),
4.65, 4.61 (d, each 1H, J ) 12.5 Hz), 4.573 (d, 1H, J ) 9.5 Hz, H-1),
p-Methylphenyl 4,6-O-Benzylidene-2-O-chloroacetyl-1-thio-â-D-
galactopyranoside (34). To a solution of the p-methylphenyl 4,6-O-

Programmable One-Pot Oligosaccharide Synthesis
J. Am. Chem. Soc., Vol. 121, No. 4, 1999 749
4.33 (d, 1H, J ) 12.0 Hz), 4.14 (d, 1H, J ) 3.5 Hz), 3.97 (d, 1H, J )
12.0 Hz), 3.60 (dd, 1H, J ) 10.0, 3.5 Hz), 3.41 (s, 1H), 2.77, 2.62 (m,
2H each), 2.32, 2.19 (s, 3H each); 13C NMR (125 MHz, CDCl3) δ
206.4, 171.7, 138.1, 138.0, 137.6, 101.2, 85.3, 78.4, 73.2, 71.2, 69.9,
69.2, 68.6, 37.9, 29.9, 28.1, 21.2; HRMS (M + Cs) calcd for C32H34O7-
SCs 695.1080, found 695.1054.
p-Methylphenyl 2,3-Di-O-acetyl-6-O-benzyl-1-thio-â-D-galacto-
pyranoside (38). p-Methylphenyl 4,6-O-benzylidene-1-thio-â-D-
galactopyranoside (200 mg, 0.437 mmol) was treated with Ac2O/
pyridine (1:2, 3 mL). Flash chromatography (hexanes/AcOEt, 1:1) of
the residue gave p-methylphenyl 2,3-di-O-acetyl-4,6-O-benzylidene-
1-thio-â-D-galactopyranoside (358 mg, 97%). One part of the product
(200 mg, 0.437 mmol) was mixed with 3-Å molecular sieves (300 mg)
and NaCNBH3 (155 mg, 2.19 mmol) in THF (2 mL), and 1 M HCl/
Et2O solution was added until no more gas evolved. The mixture was
diluted with CH2Cl2 (20 mL), filtered through Celite, washed with brine,
dried (Na2SO4), filtered, and concentrated. Chromatography (hexanes/
p-Methylphenyl 2-O-Benzoyl-4-O-(2,3,4-tri-O-benzyl-r-L-fuco-
pyranosyl)-6-O-tert-butyldiphenylsilyl-3-O-p-methoxybenzyl-1-thio-â-D-
galactopyranoside (41). Compound 41 was prepared from 44 (311
mg) and 50 (337 mg) as described for the preparation of 39, yielding
356 mg (74%): 1H NMR (500 MHz, CDCl3) δ 8.02 (dd, J ) 8.0, 1.0
Hz, 2H), 7.69 (dd, J ) 8.0, 1.0 Hz, 2H), 7.63 (dd, J ) 8.0, 1.5 Hz,
2H), 7.59 (t, J ) 7.5 Hz, 1H), 7.17-7.48 (m, 25H), 6.98 (t, J ) 8.5
Hz, 4H), 6.63 (d, J ) 8.5 Hz, 2H), 5.76 (t, J ) 9.5 Hz, 1H), 5.57 (d,
J ) 4.0 Hz, 1H), 4.92 (d, J ) 12.0 Hz, 1H), 4.87 (d, J ) 11.5 Hz,
1H), 4.77 (d, J ) 12.0 Hz, 2H), 4.67 (d, J ) 12.0 Hz, 1H), 4.56 (d, J
) 11.5 Hz, 1H), 4.49 (d, J ) 11.5 Hz, 1H), 4.44 (d, J ) 12.5 Hz, 1H),
4.39 (d, J ) 11.5 Hz, 1H), 4.21 (d, J ) 2.0 Hz, 1H), 3.96 (dd, J )
11.0, 7.5 Hz, 1H), 3.89 (dd, J ) 10.5, 4.0 Hz, 1H), 3.63-3.77 (m,
7H), 3.44 (q, J ) 6.5 Hz, 1H), 3.34 (d, J ) 2.0 Hz, 1H), 2.21 (s, 3H),
1.07 (s, 9H), 0.74 (d, J ) 6.5 Hz, 3H); 13C NMR (125 MHz, CDCl3)
δ 164.9, 159.2, 139.4, 139.0, 138.6, 137.0, 135.7, 135.5, 133.4, 133.0,
132.9, 131.6, 130.9, 130.3, 129.8, 129.7, 129.6, 129.5, 129.3, 128.6,
128.3, 128.2, 128.1, 128.1, 127.9, 127.7, 127.5, 127.5, 127.2, 126.8,
113.7, 96.6, 87.8, 81.27, 80.23, 78.32, 77.51, 77.21, 75.12, 74.50, 73.61,
71.83, 71.46, 70.32, 69.22, 66.63, 64.76, 55.15, 26.84, 21.04, 16.54;
HRMS (M + Cs) calcd for C71H76O11SSiCs 1297.3932, found
1297.3995.
p-Methylphenyl 2-O-Benzoyl-4,6-O-benzylidene-1-thio-â-D-ga-
lactopyranoside (42). p-Methylphenyl 4,6-O-benzylidene-1-thio-â-D-
galactopyranoside (300 mg, 0.802 mmol) in CH2Cl2 (10 mL) was treated
with DCC (238 mg, 1.16 mmol), DMAP (50 mg), and levulinic acid
(100 µL, 0.963 mmol) as described for the preparation of 34. The
residue was treated with pyridine (500 µL) and BzCl (400 µL) under
reflux conditions. After workup, the residue was chromatographed
(hexanes/AcOEt, 2:1 to 1:1) to give p-methylphenyl 2-O-benzoyl-4,6-
O-benzylidene-3-O-levulinyl-1-thio-â-D-galactopyranoside (226 mg,
49%), which was treated with NH2NH2 (100 µL)/AcOH (200 µL) in
THF/MeOH (10:1, 5 mL). The residue was concentrated, diluted with
CH2Cl2 (150 mL), washed with saturated aqueous NaHCO3, dried
(Na2SO4), filtered, and concentrated. Chromatography of the residue
gave 42 (185 mg, 99%): 1H NMR (500 MHz, CDCl3) δ 7.50-7.10
(14H), 5.52 (s, 1H, ArCH), 5.20 (t, 1H, J ) 9.5 Hz, H-2), 4.760 (d,
1H, J ) 10.0 Hz, H-1), 4.40 (dd, 1H, J ) 1.5, 12.5 Hz, H-6), 4.24 (bd,
1H, J ) 3.5 Hz, H-4), 4.04 (dd, 1H, J ) 1.5, 12.5 Hz, H-6), 3.87 (ddd,
1H, J ) 3.0, 9.5 Hz, H-3) 3.59 (bs, 1H, H-5), 2.62 (d, 1H, J ) 11.0
Hz, OH), 2.35 (s, 3H, Me); 13C NMR (500 MHz, CDCl3) δ 165.96,
101.44, 84.80, 75.60, 72.90, 70.67, 69.88, 69.10, 21.21; HRMS (M +
Na) calcd for C27H26O6SNa 501.1348, found 501.1364.
p-Methylphenyl 2-O-benzoyl-6-O-tert-butyldiphenylsilyl-3-O-p-
methoxybenzyl-1-thio-â-D-galactopyranoside (44). To a stirred solu-
tion of 35 (420 mg, 0.496 mmol) in THF/MeOH (10:1 v/v, 22 mL)
was added 1 M NH2NH2/AcOH (1:2.5 v/v) in THF/MeOH (5:1 v/v)
(5 mL). The reaction mixture was stirred for 4 h and then evaporated
to dryness. The residue was diluted with EtOAc (80 mL), washed with
saturated aqueous NaHCO3 and brine, and dried over Na2SO4. The dried
organic layer was concentrated under reduced pressure, and the residue
was purified by chromatography on a silica gel column (hexanes/EtOAc,
4:1) to give 44 (343 mg, 92%) as solids: 1H NMR (500 MHz, CDCl3)
δ 8.01 (dd, J ) 8.0, 1.0 Hz, 2H), 7.68-7.71 (m, 4H), 7.60 (t, J ) 7.0,
1.5 Hz, 1H), 7.36-7.48 (m, 8H), 7.33 (d, J ) 8.0 Hz, 2H), 7.07 (d, J
) 8.5 Hz, 2H), 6.99 (d, J ) 8.0 Hz, 2H), 6.67 (d, J ) 9.0 Hz, 2H),
5.43 (t, J ) 10.0 Hz, 1H), 4.67 (d, J ) 10.0 Hz, 1H), 4.57 (d, J ) 12.0
Hz, 1H), 4.45 (d, J ) 12.0 Hz, 1H), 4.12 (bs, 1H), 3.94-4.00 (m, 2H),
3.71 (s, 3H), 3.61 (dd, J ) 9.0, 3.0 Hz, 1H), 3.56 (t, J ) 6.0 Hz, 1H),
2.54 (d, J ) 1.0 Hz, 1H), 2.27 (s, 3H), 1.07 (s, 9H); 13C NMR (125
MHz, CDCl3) δ 165.2, 159.3, 137.7, 135.6, 135.6, 133.2, 133.2, 133.0,
132.6, 130.1, 129.9, 129.7, 129.5, 129.5, 129.3, 128.3, 127.7, 113.8,
87.06, 79.02, 78.75, 71.01, 69.85, 66.30, 63.15, 55.12, 26.81, 21.08,
19.19; HRMS (M + Cs) calcd for C44H48O7SSiCs 881.1944, found
881.1976.
AcOEt, 1:1) of the residue gave 38 (184 mg, 92%) as a white solid:
1
[R]22 +7° (c, 0.5, CHCl3); H NMR (500 MHz, CDCl3) δ 41-7.05
D
(9H), 5.27 (t, 1H, J ) 9.5 Hz, H-2), 4.95 (dd, 1H, J ) 3.0, 9.5 Hz,
H-3), 4.625 (d, 1H, J ) 10.0 Hz, H-1), 4.55 (ABq, 2H, J ) 12.0 Hz,
ArCH2), 4.16 (bs, 1H, H-4), 3.77 (m, 2H, H-6), 3.70 (t, 1H, J ) 5.0
Hz, H-5), 2.74 (d, 1H, J ) 3.5 Hz, OH), 2.30, 2.08, 2.06 (s, 3H each,
Me); HRMS (M + Na) calcd for C24H28O7SNa 483.1453, found
483.1465.
p-Methylphenyl 2-O-(2,3,4-Tri-O-benzyl-r-L-fucopyranosyl)-6-
O-tert-butyldiphenylsilyl-4-O-levulinyl-3-O-p-methoxybenzyl-1-thio-
â-D-galactopyranoside (39). To a mixture of 45 (300 mg, 0.404 mmol),
50 (327 mg, 0.606 mmol), and 4-Å molecular sieves (800 mg) in
CH2Cl2 (8 mL) was added NIS (143 mg, 0.606 mmol) under argon.
After the mixture was stirred for 20 min at -20 °C, TfOH (65 µL, 0.3
M in Et2O) was added at -20 °C, and the reaction mixture was stirred
for 20 min. Et3N (1.0 mL) was then added. The reaction mixture was
filtered through Celite, and the filtrate was diluted with CH2Cl2. The
solution was washed with saturated aqueous Na2S2O5, saturated
NaHCO3, and brine, dried over Na2SO4, and concentrated. The residue
was purified by column chromatography on silica gel (hexanes/EtOAc,
4:1) to give 39 (343 mg, 73%): 1H NMR (500 MHz, CDCl3) δ 7.60-
7.63 (m, 4H), 7.19-7.43 (m, 23H), 7.06 (d, J ) 9.0 Hz, 2H), 7.01 (d,
J ) 8.0 Hz, 2H), 6.76-6.80 (m, 2H), 5.76 (d, J ) 4.0 Hz, 1H), 5.63
(d, J ) 3.0 Hz, 1H), 4.94 (d, J ) 11.5 Hz, 1H), 4.79 (d, J ) 11.5 Hz,
1H), 4.74 (d, J ) 10.0 Hz, 1H), 4.72 (d, J ) 11.5 Hz, 1H), 4.58-4.69
(m, 5H), 4.26 (d, J ) 10.5 Hz, 1H), 3.97-4.07 (m, 3H), 3.76 (s, 3H),
3.71-3.75 (m, 3H), 3.59-3.65 (m, 2H), 2.49-2.56 (m, 4H), 2.29 (s,
3H), 2.07 (s, 3H), 1.14 (d, J ) 6.5 Hz, 3H), 1.02 (s, 9H); 13C NMR
(100 MHz, CDCl3) δ 171.73, 158.91, 138.76, 138.61, 138.43, 137.23,
135.61, 135.55, 133.17, 132.94, 131.56, 130.42, 129.76, 129.67, 129.58,
128.68, 128.35, 128.29, 128.14, 128.10, 127.70, 127.67, 127.51, 127.40,
127.33, 113.60, 97.57, 87.52, 82.57, 79.47, 77.71, 77.10, 75.55, 74.68,
73.14, 72.81, 71.53, 70.25, 67.37, 65.95, 62.00, 55.22, 38.13, 29.74,
28.08, 26.72, 21.04, 19.06, 16.58; HRMS (M + Cs) calcd for C69H78O12-
SSiCs 1291.4038, found 1291.4100.
p-Methylphenyl 2-O-(2,3,4-Tri-O-benzyl-r-L-fucopyranosyl)-4-
O-levulinyl-3-O-p-methoxybenzyl-1-thio-â-D-galactopyranoside (40).
Compound 40 was prepared from 39 (86.0 mg) as described for the
preparation of 36, yielding 62.6 mg (92%): 1H NMR (500 MHz,
CDCl3) δ 7.18-7.37 (m, 17H), 7.09 (d, J ) 8.0 Hz, 2H), 7.04 (d, J )
8.5 Hz, 2H), 6.78 (d, J ) 8.5 Hz, 2H), 5.76 (d, J ) 4.0 Hz, 1H), 5.45
(d, J ) 3.5 Hz, 1H), 4.94 (d, J ) 11.5 Hz, 1H), 4.80 (d, J ) 11.5 Hz,
1H), 4.76 (d, J ) 10.0 Hz, 1H), 4.73 (d, J ) 11.5 Hz, 1H), 4.58-4.68
(m, 4H), 4.55 (d, J ) 10.5 Hz, 1H), 4.27 (d, J ) 10.5 Hz, 1H), 4.05-
4.09 (m, 3H), 4.00 (dd, J ) 10.0, 2.5 Hz, 1H), 3.82 (dd, J ) 9.0, 3.5
Hz, 1H), 3.77 (s, 3H), 3.73 (bs, 1H), 3.67 (dd, J ) 11.5, 6.5 Hz, 1H),
3.59 (t, J ) 6.5 Hz, 1H), 3.51 (dd, J ) 11.0, 6.5 Hz, 1H), 2.59-2.75
(m, 3H), 2.50-2.56 (m, 1H), 2.32 (s, 3H), 2.13 (s, 3H), 1.17 (d, J )
6.5 Hz, 3H); 13C NMR (125 MHz, CDCl3) δ 206.4, 173.2, 159.0, 138.7,
138.6, 138.3, 137.6, 132.0, 129.8, 129.6, 129.5, 128.5, 128.3, 128.2,
128.1, 128.1, 127.7, 127.5, 127.4, 127.4, 113.6, 97.6, 87.08, 82.10,
79.42, 77.66, 76.75, 75.51, 74.68, 73.15, 72.82, 71.79, 70.13, 67.40,
66.41, 60.53, 55.20, 37.93, 29.70, 27.92, 21.04, 16.58; HRMS (M +
Cs) calcd for C53H60O12SCs 1053.2860, found 1053.2830.
p-Methylphenyl 6-O-tert-Butyldiphenylsilyl-4-O-levulinyl-3-O-p-
methoxybenzyl-1-thio-â-D-galactopyranoside (45). A mixture of
p-methylphenyl 2-O-chloroacetyl-3-O-p-methoxybenzyl-4-O-levulinyl-
6-O-tert-butyldiphenylsilyl-1-thio-â-D-galactopyranoside22 (550 mg,
0.672 mmol), saturated NaHCO3 in MeOH/H2O (5:1 v/v) solution (50
mL), and THF (12 mL) was stirred for 6 h at 60 °C. The solvent was

750 J. Am. Chem. Soc., Vol. 121, No. 4, 1999
Zhang et al.
then removed under reduced pressure, and the residue was diluted with
EtOAc (150 mL) and water (30 mL). The organic layer was separated,
and the aqueous layer was extracted with EtOAc (2 × 25 mL). The
organic fractions were combined, washed with brine, and dried over
Na2SO4. The dried organic layer was concentrated, and the residue was
purified by chromatography on a silica gel column (hexanes/EtOAc,
3:1) to yield 45 (498 mg, 100%) as a syrup: 1H NMR (500 MHz,
CDCl3) δ 7.63-7.66 (m, 4H), 7.36-7.45 (m, 8H), 7.26 (d, J ) 8.5
Hz, 2H), 7.04 (d, J ) 8.0 Hz, 2H), 6.87 (d, J ) 8.5 Hz, 2H), 5.63 (d,
J ) 3.0 Hz, 1H), 4.71 (d, J ) 11.0 Hz, 1H), 4.46 (d, J ) 9.5 Hz, 1H),
4.40 (d, J ) 11.0 Hz, 1H), 3.79 (s, 3H), 3.78 (dd, J ) 12.5, 9.0 Hz,
1H), 3.64-3.68 (m, 2H), 3.62 (dt, J ) 1.5, 9.5 Hz, 1H), 3.44 (dd, J )
9.0, 3.0 Hz, 1H), 2.50-2.67 (m, 4H), 2.42 (d, J ) 1.5 Hz, 1H), 2.30
(s, 3H), 2.12 (s, 3H), 1.05 (s, 9H); 13C NMR (125 MHz, CDCl3) δ
205.9, 171.7, 159.4, 138.1, 135.6, 135.6, 133.1, 133.0, 129.9, 129.8,
129.8, 129.6, 129.4, 128.2, 127.7, 113.9, 88.4, 79.83, 77.58, 71.17,
68.38, 66.21, 61.89, 55.25, 38.06, 29.79, 28.02, 26.74, 21.10, 19.10;
HRMS (M + Cs) calcd for C42H50O8SSiCs 875.2050, found 875.2082.
p-Methylphenyl 2,3-Di-O-benzyl-4,6-O-benzylidene-1-thio-â-D-
galactopyranoside (46) and p-Methylphenyl 2-O-Benzyl-4,6-O-
benzylidene-1-thio-â-D-galactopyranoside (49). To a mixture of
p-methylphenyl 4,6-O-benzylidene-1-thio-â-D-galactopyranoside (2.0
g, 5.35 mmol) and BnBr (1.08 mL, 6.42 mmol) in a mixture of CH2Cl2
(75 mL) and 1.0 M aqueous NaOH (27 mL) was added Bu4N/HSO4
(367 mg, 1.07 mmol). The mixture was stirred vigorously for 20 h
under reflux and then diluted with CH2Cl2 (100 mL), washed with brine,
dried over Na2SO4, filtered, and concentrated. Chromatography (hex-
p-Methylphenyl 2,3,4,6-Tetra-O-benzyl-1-thio-â-D-galactopyran-
oside (48). p-Methylphenyl 1-thio-â-D-galactopyranoside (1.0 g, 3.5
mmol) was dissolved in dry DMF (15 mL) and cooled to -10 °C. At
this temperature, NaH (840 mg, 35 mmol) was added, and the mixture
was allowed to warm to room temperature and stirred for an additional
10 min. Then BnBr (4.1 mL, 35 mmol) was added, and the mixture
was stirred for 5 h. At 0 °C, 10 mL of H2O was added, and the mixture
was extracted with EtOAc (3 × 50 mL). The organic layer was dried
over MgSO4 and evaporated to dryness. Flash chromatography using
hexanes/EtOAc (5:1) gave the title compound as a white solid (2.01 g,
1
89%): [R]22 -1.2° (c, 0.5, CHCl3); H NMR (400 MHz, CDCl3) δ
D
2.27 (s, 3H), 3.56-3.65 (m, 4H), 3.89 (t, 1H, J ) 9.4 Hz) 3.97 (s,
1H), 4.42 (ABq, 2H, J ) 11.6 Hz), 4.59-4.60 (m, 3H), 4.70 (m, 3H),
4.73 (ABq, 2H, J ) 10.1 Hz), 4.95 (d, 1H, J ) 11.5 Hz), 6.98 (d, 2H,
J ) 8.6 Hz), 7.25-7.47 (m, 22H); 13C NMR (400 MHz, CDCl3) δ
21.04, 68.68, 72.63, 73.48, 74.34, 75.54, 77.16, 77.24, 84.13, 87.95,
127.36, 127.49, 127.60, 127.65, 127.71, 127.78, 127.86, 128.09, 128.26,
128.35, 129.49, 132.10; HRMS (M + Cs) calcd for C41H42O5SCs
779.1807, found 779.1782.
p-Methylphenyl 2,3,4-Tri-O-benzyl-1-thio-â-L-fucopyranoside (50).
p-Methylphenyl 2,3,4-tri-O-acetyl-1-thio-â-L-fucopyranoside9 (2.5 g,
6.31 mmol) was treated with NaOMe/MeOH. The product was further
treated with 85% NaH (463 mg, 22.7 mmol) and BnBr (5.8 mL, 37.86
mmol) in DMF (25 mL). After the reaction was finished, hexane (30
mL) and water (300 mL) were added into the vigorously stirred solution.
The solid product was filtered and washed with water (2 × 5 mL) and
hexanes/AcOEt (15:1, 2 × 5 mL). The crude product was suspended
in a hot (70 °C) mixture of AcOEt/hexanes (10:1), stirred for 10 min,
cooled to room temperature, and filtered to give pure 50 (2.6 g, 76%)
anes/AcOEt, 5:1 to 1:1) of the residue gave dibenzylated 46 (170 mg,
1
6%) as a syrup: [R]22 -23° (c, 0.5, CHCl3); H NMR (500 MHz,
D
as a solid: [R]22 +8° (c, 0.5, CHCl3); HRMS (M + Cs) calcd for
D
CDCl3) δ 5.48 (s, 1H, ArCH), 4.72 (m, 4H, ArCH2), 4.577 (d, 1H, J
) 9.50 Hz, H-1), 4.37 (dd, 1H, J ) 1.5, 12.0 Hz, H-6), 4.14 (d, 1H, J
) 3.5 Hz, H-4), 3.98 (dd, 1H, J ) 1.5, 12.5 Hz, H-6), 3.85 (t, 1H, J
) 9.5 Hz, H-2), 3.63 (dd, 1H, J ) 3.5, 9.5 Hz, H-3), 3.40 (bs, 1H,
H-5), 3.51 (d, 1H, J ) 1.08 Hz, H-5), 2.31 (s, 3H, Me); 13C NMR
(500 MHz, CDCl3) δ 138.5, 138.1, 137.9, 137.6, 133.4, 101.3, 86.5,
81.4, 75.4, 75.3, 73.6, 71.8, 69.7, 69.4, 21.1. Data for p-methylphenyl
3-O-benzyl-4,6-O-benzylidene-1-thio-â-D-galactopyranoside (1.10 g,
44.7%): 1H NMR (500 MHz, CDCl3) δ 5.39 (s, 1H, ArCH), 4.68 (m,
2H, ArCH2), 4.435 (d, 1H, J ) 9.5 Hz, H-1), 4.30 (dd, 1H, J ) 1.0,
12.5 Hz, H-6), 4.09 (d, 1H, J ) 3.5 Hz, H-4), 3.92 (dd, 1H, J ) 1.5,
12.5 Hz, H-6), 3.85 (t, 1H, J ) 9.5 Hz, H-2), 3.47 (dd, 1H, J ) 3.5,
9.5 Hz, H-3), 3.38 (bs, 1H, H-5), 2.31 (s, 3H, Me); 13C NMR (500
MHz, CDCl3) δ 101.0, 87.0, 80.1, 73.2, 71.5, 69.9, 69.3, 67.0, 21.2;
HRMS (M + Na) calcd for C27H28O5SCs 597.0712, found 597.0728.
49 (780 mg, 31.4%) was formed as syrup: 1H NMR (500 MHz, CDCl3)
δ 5.51 (s, 1H, ArCH), 4.77 and 4.67 (d, 1H each, ArCH2), 4.541 (d,
1H, J ) 9.9 Hz, H-1), 4.34 (bd, 1H, J ) 12.0 Hz, H-6), 4.12 (d, 1H,
J ) 3.5 Hz, H-4), 3.97 (bd, 1H, J ) 12.5 Hz, H-6), 3.60 (t, 1H, J )
9.0 Hz, H-2), 3.75 (bs, 1H, H-3), 3.40 (bs, 1H, H-5), 2.60 (bs, 1H,
OH), 2.33 (s, 3H, Me); 13C NMR (500 MHz, CDCl3) δ 101.3, 86.3,
77.0, 75.7, 75.2, 74.2, 69.6, 69.2, 21.1; HRMS (M + Cs) calcd for
C27H28O5SCs 597.0712, found 597.0727.
C34H36O4SCs 672.1290, found 672.1287.
Methyl 2,3-Di-O-acetyl-6-O-benzyl-4-O-[2,4,6-tri-O-benzoyl-3-O-
(2,3,4,6-tetra-O-benzyl-r-D-galactopyranosyl)-â-D-galactopyranosyl]-
r-D-glucopyranoside (52). A mixture of 48 (37.5 mg, 0.052 mmol),
19 (24 mg, 0.039 mmol), and MS-AW-300 (400 mg) in CH2Cl2 (2
mL) was stirred at 0 °C for 15 min, and then NIS (13.5 mg, 0.057
mmol) was added, followed by 0.5 M TfOH in Et2O (10 µL, 0.005
mmol). After 30 min, a solution of 51 (36 mg, 0.080 mmol) in CH2Cl2
(0.5 mL) and MS AW-300 (100 mg) was added, followed by NIS (13.5
mg, 0.057 mmol). The mixture was stirred for 2 h at room temperature,
and then solid Na2S2O3, NaHCO3, and two drops of H2O were added,
and the mixture was stirred for 5 min, diluted with CH2Cl2 (25 mL),
filtered through Celite, washed with brine, dried (Na2SO4), filtered,
and concentrated. Chromatography (hexanes/AcOEt, 2:1 to 1:1) of the
residue gave 52 (35.9 mg, 64%) as a syrup: 1H NMR (400 MHz,
CDCl3) δ 5.80 (d, 1H, J ) 3.1 Hz), 5.56 (dd, 1H, J ) 8.4, 10.1 Hz),
5.48 (d, 1H, J ) 9.8 Hz), 5.09 (d, 1H, J ) 3.2 Hz), 4.86 (d, 1H, J )
3.6 Hz), 4.79 (dd, 1H, J ) 3.7, 10.4 Hz), 4.69 (d, 1H, J ) 11.4 Hz),
4.60 (d, 1H, J ) 8.1 Hz), 4.58 (d, 1H, J ) 12.0 Hz), 3.95 (m, 2H),
3.85 (dd, 1H, J ) 3.2 Hz), 3.81-3.74 (m, 2H), 3.69 (m, 2H), 3.28 (s,
3H), 2.05, 1.96 (s, 3H each); 13C NMR (500 MHz, CDCl3) δ 170.3,
169.7, 166.1, 165.8, 164.3, 243.6, 243.3, 243.0, 138.8, 138.6, 138.2,
138.1, 100.7, 96.9, 95.6, 78.8, 75.1, 75.0, 74.5, 74.3, 73.4, 73.3, 73.1,
72.5, 71.5, 71.4, 71.3, 70.0, 69.9, 69.7, 68.9, 67.6, 86.6, 62.3, 20.9,
20.7; HRMS (M + Cs) calcd for C79H80O21Cs 1497.4246, found
1497.4362.
Methyl 2,3-Di-O-acetyl-6-O-benzyl-4-O-[2,3,4-tri-O-benzoyl-6-O-
(3,4,6-tri-O-acetyl-2-deoxy-2-(2,2,2,-trichloroethoxylcarbonylamino)-
â-D-glucopyranosyl)-â-D-galactopyranosyl]-r-D-glucopyranoside (53).
A mixture of 21 (70 mg, 0.12 mmol), 13 (48 mg, 0.08 mmol), and
MS-AW-300 (400 mg) in CH2Cl2 (2 mL) was stirred at 0 °C for 15
min, and then NIS (30.5 mg, 0.132 mmol) was added, followed by a
0.1 M solution of TfOH (5 µL) in Et2O. After 15 min, a solution of 51
(72 mg, 0.16 mmol) in CH2Cl2 (0.5 mL) and MS AW-300 (100 mg)
was added, followed by NIS (33 mg, 0.13 mmol). The mixture was
stirred for 1 h, followed by workup as described in the preparation of
52. Chromatography (hexanes/AcOEt, 3:2 to 1:1) of the residue gave
53 (62.5 mg, 60%) as a white solid: 1H NMR (500 MHz, CDCl3) δ
5.80 (d, 1H, J ) 3.1 Hz), 5.77 (d, 1H, J ) 8.2 Hz), 5.56 (dd, 1H, J )
8.2, 10.0 Hz), 5.42 (t, 1H, J ) 10.0 Hz), 5.34 (bt, 1H, J ) 9.4 Hz),
5.25 (dd, 1H, J ) 3.2, 10.3 Hz), 5.0 (m, 3H), 4.87 (d, 1H, J ) 3.5 Hz),
p-Methylphenyl 2-O-(2,3,4-Tri-O-benzyl-r-L-fucopyranosyl)-6-
O-tert-butyldiphenylsilyl-3-O-p-methoxybenzyl-1-thio-â-D-galacto-
pyranoside (47). Compound 47 was prepared from 39 (73.0 mg) as
described for the preparation of 44, yielding 57.3 mg (86%): 1H NMR
(500 MHz, CDCl3) δ 7.63-7.65 (m, 4H), 7.22-7.41 (m, 23H), 7.13
(d, J ) 8.5 Hz, 2H), 7.00 (d, J ) 8.0 Hz, 2H), 6.80 (d, J ) 8.5 Hz,
2H), 5.75 (d, J ) 3.5 Hz, 1H), 4.95 (d, J ) 12.0 Hz, 1H), 4.80 (d, J
) 11.5 Hz, 1H), 4.65-4.77 (m, 5H), 4.60 (q, J ) 6.5 Hz, 1H), 4.53
(d, J ) 11.5 Hz, 1H), 4.43 (d, J ) 11.5 Hz, 1H), 4.01-4.12 (m, 4H),
3.83-3.89 (m, 2H), 3.76 (s, 3H), 3.67-3.73 (m, 2H), 3.45 (t, J ) 6.0
Hz, 1H), 2.36 (bs, 1H), 2.28 (s, 3H), 1.14 (d, J ) 6.0 Hz, 3H), 1.02 (s,
9H); 13C NMR (125 MHz, CDCl3) δ 159.2, 138.8, 138.6, 138.5, 137.0,
135.6, 135.5, 135.5, 133.2, 133.1, 131.4, 130.7, 129.7, 129.6, 129.5,
129.3, 128.7, 128.4, 128.3, 128.1, 127.8, 127.7, 127.5, 127.4, 113.9,
97.6, 87.34, 84.24, 79.59, 78.00, 77.66, 75.70, 74.66, 73.37, 72.78,
71.55, 70.38, 67.31, 65.54, 63.04, 55.19, 29.66, 26.76, 21.04, 16.55;
HRMS (M + Cs) calcd for C64H72O10SSiCs 1193.3670, found
1193.3613.

Programmable One-Pot Oligosaccharide Synthesis
J. Am. Chem. Soc., Vol. 121, No. 4, 1999 751
4.75 (d, 1H, J ) 12.5 Hz), 4.70 (d, 1H, J ) 8.3 Hz), 4.38 (d, 1H, J )
12.5 Hz), 3.31 (s, 3H), 2.07, 2.06 (s, 3H each), 2.01 (s, 9H); 13C NMR
(500 MHz, CDCl3) δ 170.6, 170.4, 169.6, 169.5, 165.4, 165.3, 164.6,
133.6, 133.3, 133.2, 100.7, 100.1, 97.1, 74.3, 74.2, 73.6, 72.5, 72.0,
71.8, 71.6, 70.8, 70.7, 70.0, 69.6, 68.6, 68.2, 67.6, 67.1, 61.8, 56.2,
55.3, 31.6, 29.7, 29.6, 22.63, 21.22, 20.8, 20.6; HRMS (M + Cs) calcd
for C60H64O25NSiCs 1436.1887, found 1436.1824.
was added, and the mixture was stirred for 10 min. The reaction mixture
was poured into ice-water (200 mL) and extracted with EtOAc (3 ×
150 mL). The combined organic liquid was dried over Na2SO4. The
solvent was removed and the residue subjected to column chromatog-
raphy on silica gel (hexanes/EtOAc, 3:1) to give 5-methoxycarbonyl-
pentyl 3-O-benzyl-2-deoxy-4,6-O-p-methoxybenzylidene-2-phthalimido-
â-D-glucopyranoside (5.56 g, 54%) as a white solid: 1H NMR (500
MHz, CDCl3) δ 7.66-7.85 (m, 4H), 7.45 (d, J ) 8.5 Hz, 2H), 7.00
(dd, J ) 8.0, 1.5 Hz, 2H), 6.86-6.94 (m, 5H), 5.58 (s, 1H), 5.18 (d,
J ) 8.5 Hz, 1H), 4.78 (d, J ) 12.5 Hz, 1H), 4.49 (d, J ) 12.5 Hz,
1H), 4.40 (dd, J ) 10.5, 9.0 Hz, 1H), 4.38 (dd, J ) 10.5, 5.5 Hz, 1H),
4.20 (dd, J ) 10.5, 8.5 Hz, 1H), 3.74-3.86 (m, 6H), 3.58-3.67 (m,
4H), 3.38 (dt, J ) 9.5, 6.5 Hz, 1H), 1.92-2.03 (m, 2H), 1.33-1.44
(m, 4H), 1.03-1.12 (m, 2H); 13C NMR (125 MHz, CDCl3) δ 173.8,
160.0, 137.9, 133.9, 131.6, 129.8, 128.0, 127.3, 123.3, 113.6, 101.3,
98.8, 83.05, 74.58, 74.02, 69.59, 68.73, 66.10, 55.81, 55.27, 51.44,
33.67, 28.91, 25.24, 24.35; HRMS (M + Cs) calcd for C36H39O10NCs
778.1628, found 778.1652.3). The product (2.7 g, 4.19 mmol) from
step 2 was mixed with NaCNBH3 (1.38 g, 21 mmol), DMF (35 mL),
and 3-Å molecular sieves (5.0 g), and a solution of trifluoroacetic acid
(3.2 mL, 42 mmol) in DMF (25 mL) was added dropwise at 0 °C. The
reaction mixture was stirred for 1.5 h at 0 °C and then 24 h at room
temperature. After addition of further trifluoroacetic acid (3.0 mL) in
DMF (20 mL), the reaction mixture was stirred for another 24 h at
room temperature. The mixture was diluted with EtOAc (50 mL) and
MeOH (5 mL) and filtered through Celite, and the cake was washed
with EtOAc (150 mL). The combined filtrates were washed with cold
water (200 mL), cold saturated NaHCO3 (200 mL), and brine (2 ×
100 mL) and dried over Na2SO4. The solvent was removed, and the
residue was purified by column chromatography on silica gel (hexanes/
EtOAc, 2:1) to give 56 (2.45 g, 90%) as an oil: 1H NMR (500 MHz,
CDCl3) δ 7.68-7.81 (m, 4H), 7.28 (d, J ) 8.5 Hz, 2H), 7.04-7.06
(m, 2H), 6.92-6.96 (m, 3H), 6.89 (d, J ) 8.5 Hz, 2H), 5.12 (d, J )
8.0 Hz, 1H), 4.75 (d, J ) 12.0 Hz, 1H), 4.57 (d, J ) 11.5 Hz, 1H),
4.53 (d, J ) 12.5 Hz, 1H), 4.51 (d, J ) 11.5 Hz, 1H), 4.21 (dd, J )
11.0, 8.5 Hz, 1H), 4.13 (dd, J ) 11.0, 8.5 Hz, 1H), 3.81 (s, 3H), 3.73-
3.82 (m, 4H), 3.60-3.64 (m, 1H), 3.59 (s, 3H), 3.35 (dt, J ) 9.5, 6.5
Hz, 1H), 3.07 (d, J ) 2.0 Hz, 1H), 1.91-2.02 (m, 2H), 1.32-1.45 (m,
4H), 1.01-1.12 (m, 2H); 13C NMR (125 MHz, CDCl3) δ 173.8, 159.3,
138.2, 133.7, 129.6, 129.4, 128.1, 127.8, 127.3, 113.9, 98.2, 78.55,
74.66, 74.20, 73.36, 73.36, 70.45, 69.18, 55.31, 55.20, 51.31, 33.60,
28.87, 25.17, 24.35; HRMS (M + Cs) calcd for C36H41O10NCs
780.1785, found 780.1759.
Methyl 2,3-Di-O-acetyl-6-O-benzyl-4-O-[2-O-benzyl-4,6-O-ben-
zylidene-3-O-(2,3,4-tri-O-benzyl-r-L-fucopyranosyl)-r-D-galacto-
pyranosyl]-r-D-glucopyranoside (54). To a cold (0 °C) solution of
50 (32 mg, 0.6 mmol), 49 (19 mg, 0.04 mmol), 2,6-di-tert-butyl-4-
methylpyridine (25 mg, 0.12 mmol), and MS-AW-300 (300 mg) in
CH2Cl2 (2 mL) was added freshly prepared 1 M DMTST in CH2Cl2
(120 µL, 0.12 mmol). After 1 h, a solution of methyl 2,3-di-O-acetyl-
6-O-benzyl-R-D-glucopyranoside 51 (36 mg, 0.08 mmol) in CH2Cl2
(0.5 mL) and MS AW-300 (100 mg) was added, followed by addition
of 1 M DMTST (120 µL, 0.12 mmol). The mixture was stirred for 2
h and then diluted with CH2Cl2 (25 mL), filtered through Celite, washed
with saturated aqueous NaHCO3, dried (Na2SO4), filtered, and con-
centrated. The residue was chromatographed (hexanes/AcOEt, 3:1 to
1:1) to give 54 (28.4 mg, 63%) as a syrup: 1H NMR (500 MHz, CDCl3)
δ 5.53 (t, 1H, J ) 9.5 Hz), 5.33 (s, 1H), 5.32 (d, 1H, J ) 3.0 Hz), 5.14
(d, 1H, J ) 3.5 Hz), 4.92-4.49 (m, 12 H), 4.28 (dd, 1H, J ) 3.0, 10.0
Hz), 4.10 (t, 1H, J ) 9.5 Hz), 3.87 (dd, 1H, J ) 3.0, 10.0 Hz), 3.72
(bt, 1H, J ) 12.5 Hz), 3.39 (s, 3H), 2.07, 2.06 (s, 3H each), 0.97 (d,
1H, J ) 6.5 Hz); HRMS (M + Cs) calcd for C65H72O17Cs 1257.3824,
found 1257.3870.
Methyl 2,3-Di-O-acetyl-6-O-benzyl-4-O-{2,3,4-tri-O-benzoyl-6-O-
[3,6-di-O-benzoyl-2-deoxy-2-(2,2,2,-trichloroethoxylcarbonylamino)-
4-O-(2,3,4,6-tetra-O-benzyl-â-D-galactopyranosyl)-â-D-glucopyranosyl]-
â-D-galactopyranosyl}-r-D-glucopyranoside (55). A mixture of 48
(73 mg, 0.112 mmol), 27 (50 mg, 0.075 mmol), and MS-AW-300 (400
mg) in CH2Cl2 (3 mL) was stirred at -25 °C for 20 min, and then NIS
(27.8 mg, 0.118 mmol) was added, followed by 1 M TfOH (10 µL,
0.001 mmol). After 20 min, 13 (68.8 mg, 0.112 mmol), MS AW-300
(100 mg), and NIS (18.6 mg, 0.079 mmol) were added. The mixture
was stirred at 0 °C for 1 h, and then a solution of 51 (250 µL of 125
mg/0.5 mL, 0.150 mmol) and NIS (40 mg, 0.169 mmol) was added,
stirred for 50 min, and worked up as described in the preparation of
52. Chromatography (hexanes/AcOEt, 3:2 to 1:1) of the residue gave
55 (56 mg, 39.2%) as a syrup: 1H NMR (500 MHz, CDCl3) δ 5.80 (d,
1H, J ) 3.1 Hz), 5.59 (t, 1H, J ) 9.5 Hz), 5.52 (dd, 1H, J ) 8.2, 10.0
Hz), 5.41 (t, 1H, J ) 9.1 Hz), 5.24 (dd, 1H, J ) 3.2, 10.4 Hz), 5.03 (d,
1H, J ) 3.4 Hz), 5.25 (dd, 1H, J ) 3.2, 10.3 Hz), 4.94 (dd, 1H, J )
3.5, 10.0 Hz), 4.67 (d, 1H, J ) 7.9 Hz), 3.29 (s, 3H), 2.07, 2.02 (s, 3H
each), 2.01 (s, 9H); HRMS (M + Cs) calcd for C102H102O29NCl3Cs
2042.4657, found 2042.
5-Methoxycarbonylpentyl 3-O-Benzyl-2-deoxy-6-O-p-methoxy-
benzyl-2-phthalimido-â-D-glucopyranoside (56). The title compound
was prepared via a three-step procedure: (1) 61b (16.0 g, 36.6 mmol)
in dry acetonitrile (200 mL) was treated with anisaldehyde dimethyl
acetyl (9.3 mL, 54.9 mmol) and 10-camphorsulfonic acid (173.5 mg)
as described for the preparation of 10. The residue was subjected to
column chromatography on silica gel (hexanes/EtOAc, 2:1) to give
5-methoxycarbonylpentyl 2-deoxy-4,6-O-p-methoxybenzylidene-2-
phthalimido-â-D-glucopyranoside (16.4 g, 81%) as solids: 1H NMR
(400 MHz, CDCl3) δ 7.84-7.87 (m, 2H), 7.71-7.75 (m, 2H), 7.42 (d,
J ) 8.7 Hz, 2H), 6.89 (d, J ) 8.8 Hz, 2H), 5.52 (s, 1H), 5.24 (d, J )
8.5 Hz, 1H), 4.57-4.63 (m, 1H), 4.36 (dd, J ) 10.2, 4.3 Hz, 1H), 4.22
(dd, J ) 10.5, 8.5 Hz, 1H), 3.78-3.85 (m, 5H), 3.55-3.66 (m, 5H),
3.37-3.45 (m, 1H), 2.64 (d, J ) 3.5 Hz, 1H), 1.93-2.07 (m, 2H),
1.32-1.51 (m, 4H), 1.05-1.15 (m, 2H); 13CNMR (100 MHz, CDCl3)
δ 173.86, 160.24, 134.13, 131.56, 129.40, 127.59, 123.42, 113.69,
101.83, 98.81, 82.19, 69.66, 68.62, 68.58, 66.09, 56.53, 55.27, 51.39,
33.67, 28.91, 25.24, 24.34; HRMS (M + Cs) calcd for C29H33O10NCs
688.1159, found 688.1188.
5-Methoxycarbonylpentyl 4-O-[3-O-(2,3,4,6-Tetra-O-benzyl-r-D-
galactopyranosyl)-4-O-(2,3,4-tri-O-benzyl-r-L-fucopyranosyl)-2-O-
benzoyl-6-O-tert-butyldiphenylsilyl-â-D-galactopyranosyl]-3-O-benzyl-
2-deoxy-6-O-p-methoxybenzyl-2-phthalimido-â-D-glucopyran-
oside (57). To a solution of the disaccharide 32 (15.0 mg, 0.01437
mmol) and the fucoside 50 (13.9 mg, 0.02155 mmol) in CH2Cl2 (2
mL) was added 4-Å molecular sieves (300 mg), and the mixture was
stirred for 20 min at room temperature under argon. A freshly prepared
solution of DMTST from methyl disulfide (8.2 mg, 86.2 µmol) and
methyl trifluoromethanesulfonate (14.3 mg, 86.2 µmol) in CH2Cl2 (0.5
mL) was added to the above stirred mixture at 0 °C, and the progress
of the reaction was followed by TLC. The reaction mixture was stirred
at 0 °C for 30 min and at room temperature for 1 h, and then a solution
of 62 (27.9 mg, 0.04311 mmol) in CH2Cl2 (0.5 mL), NIS (5.1 mg,
0.02155 mmol), and TfOH (25 µL, 0.3 M etheral solution) were added.
The reaction mixture was stirred for 30 min at room temperature and
then quenched by adding Et3N (0.8 mL), diluted with CH2Cl2, and
filtered. The organic phase was successively washed with saturated
aqueous Na2S2O3, NaHCO3, and brine, dried over Na2SO4, and
concentrated. The residue was purified by column chromatography on
silica gel (hexanes/EtOAc, 3:1 to 2:1) to give product 57 (17.2 mg,
57%) as a thick oil: 1H NMR (500 MHz, CDCl3) δ 8.00 (d, J ) 8.0
Hz, 2H), 7.08-7.77 (m, 46H), 6.96-6.97 (m, 4H), 6.81-6.85 (m, 6H),
6.76 (d, J ) 6.0 Hz, 2H), 6.64 (t, J ) 7.5 Hz, 1H), 6.55 (t, J ) 7.5 Hz,
2H), 5.98 (d, J ) 3.5 Hz, 1H), 5.90 (dd, J ) 10.0, 7.5 Hz, 1H), 5.28
(d, J ) 3.5 Hz, 1H), 4.94 (d, J ) 8.0 Hz, 1H), 4.88 (d, J ) 12.5 Hz,
1H), 4.83 (d, J ) 7.5 Hz, 1H), 4.81 (d, J ) 11.5 Hz, 1H), 4.78 (d, J
) 11.5 Hz, 1H), 4.64-4.71 (m, 2H), 4.59 (d, J ) 11.0 Hz, 1H), 4.55
(2) To a solution of the product (8.85 g, 15.9 mmol) from step (1)
in dry DMF (40 mL) was added NaH (60%, 0.95 g, 23.8 mmol) in
portions. After the mixture was stirred for 10 min, benzyl bromide (2.9
mL, 23.8 mmol) and Bu4NI (3.5 g, 9.5 mmol) were added at 0 °C. The
reaction mixture was stirred for 3 h at 0 °C, and then MeOH (8 mL)

752 J. Am. Chem. Soc., Vol. 121, No. 4, 1999
Zhang et al.
(d, J ) 12.5 Hz, 1H), 4.54 (d, J ) 12.0 Hz, 1H), 4.45 (d, J ) 11.0 Hz,
2H), 4.43 (d, J ) 12.0 Hz, 1H), 4.27-4.36 (m, 6H), 4.17 (d, J ) 12.0
Hz, 1H), 3.98-4.09 (m, 5H), 3.93 (d, J ) 12.0 Hz, 1H), 3.85-3.90
(m, 2H), 3.73-3.80 (m, 5H), 3.57-3.68 (m, 5H), 3.56 (s, 3H), 3.51
(q, J ) 6.5 Hz, 1H), 3.36-3.42 (m, 3H), 3.19-3.27 (m, 4H), 1.84-
1.96 (m, 2H), 1.25-1.35 (m, 4H), 1.10 (s, 9H), 0.95-1.03 (m, 2H),
0.90 (d, J ) 6.5 Hz, 3H); 13C NMR (100 MHz, CDCl3) δ 164.43,
159.11, 139.45, 139.42, 139.17, 138.78, 138.66, 138.51, 138.17, 137.88,
135.69, 135.48, 133.53, 133.47, 133.00, 130.40, 129.82, 129.74, 129.33,
128.70, 128.53, 128.45, 128.38, 128.33, 128.24, 128.19, 128.15, 128.00,
127.87, 127.74, 127.69, 127.66, 127.51, 127.48, 127.22, 126.84, 126.58,
113.71, 100.44, 98.15, 94.78, 94.29, 78.72, 78.28, 77.22, 77.17, 76.51,
75.59, 74.97, 74.83, 74.69, 74.65, 74.50, 74.22, 73.44, 73.34, 73.03,
72.84, 72.07, 71.46, 69.75, 69.06, 68.97, 67.90, 66.64, 64.92, 63.76,
55.74, 55.24, 51.35, 33.69, 29.70, 28.83, 27.02, 25.27, 24.37, 22.69,
19.25, 16.84, 14.13; HRMS (M + Cs) calcd for C126H135O25NSiCs
2222.8147, found 2222.8294.
∂A1 k1 ∂A2
)
k
[A1]
2 [A2]
This is then integrated over the bounds of the experiment:
∂A
k1
k2
∂A
[A2]t
2
[A1]t
1
)
∫
∫
[A1]0
[A2]0
[A1]
[A2]
k
ln[A1]t - ln[A1]o ) 1(ln[A2]t - ln[A2]o)
k2
Finally,
k1 ln[A1]t - [A1]o ln[A1]t/[A1]o
)
)
(1)
k2
ln[A2]t - [A2]o ln[A2]t/[A2]o
(ii) The Yield Equation. Assumptions: All reacting starting material
5-Methoxycarbonylpentyl 3,4,6-Tri-O-acetyl-2-deoxy-2-phthal-
imido-â-D-glucopyranoside (61a). To a mixture of 6025 (30.0 g, 62.89
mmol) and methyl 6-hydroxylhexate (13.8 g, 94.3 mmol) in CH2Cl2
(550 mL) was added BF3/OEt2 (120 mL, 943 mmol) at 0 °C. After the
reaction mixture was stirred overnight, water (15 mL) was added to
the mixture. The mixture was diluted with CH2Cl2 (400 mL), washed
with saturated NaHCO3 (3 × 200 mL) and brine (200 mL), dried over
Na2SO4, filtered, and concentrated. The residue was purified by column
chromatography on silica gel (hexanes/EtOAc, 2.5:1) to give 61a (26.0
g, 73%) as a solid: 1H NMR (500 MHz, CDCl3) δ 7.84-7.89 (m,
2H), 7.73-7.77 (m, 2H), 5.79 (dd, J ) 10.5, 9.0 Hz, 1H), 5.36 (d, J )
8.5 Hz, 1H), 5.18 (dd, J ) 10.0, 9.5 Hz, 1H), 4.34 (dd, J ) 12.0, 4.5
Hz, 1H), 4.31 (dd, J ) 10.5, 8.5 Hz, 1H), 4.18 (dd, J ) 12.5, 2.5 Hz,
1H), 3.81-3.89 (m, 2H), 3.61 (s, 3H), 3.44 (dt, J ) 10.0, 6.5 Hz, 1H),
2.12 (s, 3H), 2.04 (s, 3H), 1.97-2.07 (m, 2H), 1.87 (s, 3H), 1.37-
1.51 (m, 4H), 1.04-1.18 (m, 2H); 13C NMR (125 MHz, CDCl3) δ
173.8, 170.7, 170.2, 169.5, 134.3, 131.3, 123.6, 98.10, 71.79, 70.74,
69.73, 68.97, 62.00, 54.59, 51.38, 33.67, 28.84, 25.20, 24.37, 20.75,
20.57, 20.44; HRMS (M + Na) calcd for C27H33O12NNa 586.1900,
found 586.1920.
goes to product:
[A1]t
[A1]o
[P]t [A2]t
[Q]t
) 1 -
) 1 -
[A1]o [A2]o
[A2]o
The two donors are present in equimolar concentrations:
[A1]o ) [A2]o
The only detractor from yield is competition from the undesired donor:
[P]t
[Q]t
yield )
) 1 -
[A1]o
[A2]o
Borrowing eq 1 for the relationship between rate constant ratios and
consumption of starting material, we can quickly use it to predict the
production of the two products P (desired) and Q (undesired).
k1 ln[A1]t/[A1]o ln(1 - [P]t/[A1]o)
k2
)
)
(2)
ln[A2]t/[A2]o
ln(1 - [Q]t/[A2]o)
5-Methoxycarbonylpentyl 2-Deoxy-2-phthalimido-â-D-glucopyran-
oside (61b). Compound 61a (26.0 g, 46.18 mmol) was treated with
NaOMe/MeOH (25 wt %, 2.7 mL) in MeOH (200 mL) at room
temperature. After 6 h, the mixture was neutralized (IRC-50 resin, weak
acid) and concentrated to give product 61b (20.0 g, 99%) as a
semisolid: 1H NMR (500 MHz, CDCl3) δ 7.78-7.82 (m, 2H), 7.68-
7.72 (m, 2H), 5.16 (d, J ) 8.5 Hz, 1H), 4.63 (d, J ) 4.5 Hz, 1H),
4.25-4.31 (m, 2H), 4.06 (dt, J ) 2.0, 8.5 Hz, 1H), 3.88-3.89 (m,
2H), 3.76 (dt, J ) 10.0, 6.5 Hz, 1H), 3.65-3.71 (m, 1H), 3.59 (s, 3H),
3.48 (t, J ) 6.5 Hz, 1H), 3.43 (dt, J ) 9.5, 3.5 Hz, 1H), 3.38 (dt, J )
10.0, 6.5 Hz, 1H), 1.90-2.02 (m, 2H), 1.31-1.43 (m, 4H), 1.00-1.09
(m, 2H); 13C NMR (125 MHz, CDCl3) δ 174.0, 168.4, 134.0, 131.6,
123.4, 98.31, 75.53, 71.53, 71.20, 69.43, 61.63, 56.66, 51.38, 33.65,
28.84, 25.18, 24.34; HRMS (M + Na) calcd for C21H27O9NNa
460.1584, found 460.1598.
Rearranging, we find that
ln((1 - [Q]t/[A2]o)k /k ) ) ln(1 - [P]t/[A1]o)
1
2
(1 - [Q]t/[A2]o)k /k ) 1 - [P]t/[A1]o
1
2
Finally, employing the simple yield relations listed in the assumptions
above, this becomes
(yield)k /k ) 1 - yield
(3)
1
2
This is not directly solvable for yield. However, a computer can
find the solution through a number of approaches involving iterative
solution. The rest of us can simply refer to Figure 4 for the solution to
this equation, which is possible to draw because eq 3 may be solved
for k1/k2.
Mathematical Model. So as to be able to analyze the problem of
designing optimal syntheses of oligosaccharides with a computer, it
was necessary to quantify the relationship between yield and relative
rate constants for donors in the database. A short treatment for this
problem is shown below. It gives the mathematical relationship between
yield under optimal conditions (no side reactions, 100% reaction with
promoter, etc.) and the relative rate ratio between the two donors being
considered for each individual step.
Optimer. To search the library of donors for the combination of
donors which produces the best possible yield, a computer program,
OptiMer, was constructed using C/C++ using Metrowerks Code-
Warrior Professional Release 3 to run on Macintosh PowerPC comput-
ers to identify sets of available donors which best satisfy the yield
calculation shown in eq 2 above. The program interfaces with a database
of donors stored using the FileMaker Pro 4.0. When the user signals
his readiness to run a search, the C++ program loads the database
and searches it to identify the 100 best sets of available characterized
oligosaccharide donors. For large sequences, a Monte Carlo search
method is also available, which simply picks combinations at random.
The frequency with which each donor appears at each step in the list
of the best 100 donor sets is tabulated, and the user is presented with
a list of donors at each step which tend to give the highest overall
yields. The user is then free to make an educated choice as to which
donors to use on the basis of availability, cost, and donor reactivity.
Relationship between Relative Rates and Substrate Consumption.
Assumptions: Each competing reaction can be described by a second-
order rate equation:
∂A
∂t
∂A
∂t
1 ) -k1[A1][E]
2 ) -k2[A2][E]
(i) Integrated Second-Order Rate Equation. Solving for [E] for
one of them and substituting into the other, one obtains

Programmable One-Pot Oligosaccharide Synthesis
J. Am. Chem. Soc., Vol. 121, No. 4, 1999 753
normal set of optimizations reduces the number of sequences needed
to be scanned from 320 billion to approximately 70 million (a factor
of 4500). (The intelligent search further accelerates it by a factor of 3
or 4.) Given that current desktop computers manage about 200 million
CPU clock cycles per second, searching the full sequence space even
at one CPU cycle per sequence would require 15 min of search time.
(The real number of cycles per sequence is orders of magnitude larger
than that.)
OptiMer calculates the stepwise yield by minimization of an error
function which is based on the yield equation (eq 3). This error function
should have its minimum for values of yield and k1/k2 which best satisfy
eq 3, and it should always be positive, with a minimum of zero for 0
< yield < 1 (100%).
2
error ) (yieldk /k + yield - 1)
1
2
An initial guess as to the correct yield value was obtained using a
simpler stepwise yield equation which assumes first-order kinetics [yield
) k1/(k1 + k2)]. This value was optimized by picking two yield values
which are somewhat higher and lower than the original value and
evaluating the error function for all three yield values. This provided
three points (yield vs error for the given k1/k2) with which to calculate
a parabolic approximation of the error function (error ) A + B(yield)
+ C(yield2)). From this, a better estimate for the yield can be directly
obtained, based on the minimum of the parabolic function (minimum
) -B/2C). This was performed in an iterative fashion, in the process
narrowing the bracketing outlying points and optimizing the yield value
until the yield value had converged to stepwise changes of less than
0.00001%. The yield values so obtained were checked against eq 3
(which can be solved for k1/k2) and were found to deviate from the
expected value by no more than (0.001% over the range 10-6 < k1/k2
< 106. Past this range, the yield is expected to be extremely close to
0% (k1/k2 < 10-6) and 100% (k1/k2 > 106). Errors outside the tested
range are expected to influence the overall yield calculation insignifi-
cantly, since the yield behaves asymptotically in these areas. The
stepwise yields are multiplied together to obtain an overall calculated
yield for the set of donors; if the overall yield is large enough to be
included in the list of top hits, the set of donors is stored for later
output to the user.
Figure 4. Relationship between expected yield and the rate ratios
between the donor and acceptor.
One potential limitation of this approach stems from the extremely
dense information capacity inherent in the carbohydrate code.29 This
manifests itself as an extremely rapid increase in the computational
burden with increasing library size and increasing number of monomer
units in the oligomer. As a result, the following optimizations were
made to the search logic: The program only checks donors for a
reaction step which have a donor reactivity less than or equal to the
reactivity of the donor used in the previous step. As a result, the program
will return no result in which the reactivities of the donors do not
decrease (or stay the same) over the course of the multistep reaction.
In addition, the program (when not in Monte Carlo mode) checks the
sequences in order of decreasing reactivity for the donor used in step
one. If the program finds that a particular donor at step one yields no
“hits” for all combinations of donors for steps which follow it, the
program will not examine less reactive step one donors. This can be
done because a more reactive step one donor will always give a better
estimated yield than a slower one, provided the reactivities of the other
donors are kept constant. For large libraries which are well represented
for all values for most donors, there is an intelligent search option which
further restricts the search at each step to the range of values around
the expected optimal reactivity.
Acknowledgment. We gratefully acknowledged a fellowship
(NATO-Postgraduiertenstipendium) granted by the DAAD to
R.W. We thank Dr. Shuichi Takayama for helpful discussions
and Dr. L. Taylor Ollmann for valuable and insightful discus-
sions regarding the handling of eq 3.
For a hexasaccharide search using a library with 200 random donors
with reactivities ranging from 1 to 10 000 available for each step, the
(29) Laine, R. A. Pure Appl. Chem. 1997, 69, 1867.
JA982232S