Palladium-controlled β-selective glycosylation in the absence of the C(2)-ester participatory group

Article abstract of DOI:10.1021/jo802468p

The development of a new glycosylation method for the stereoselective synthesis of β-glycosides in the absence of the traditional C(2)-ester neighboring group effect is described. This process relies on the ability of the cationic palladium catalyst, Pd(P

Full text of DOI:10.1021/jo802468p

Palladium-Controlled ꢀ-Selective Glycosylation in the Absence of the
C(2)-Ester Participatory Group
Enoch A. Mensah, Joseph M. Azzarelli, and Hien M. Nguyen*
Department of Chemistry and Biochemistry, Montana State UniVersity, Bozeman, Montana 59717
hmnguyen@chemistry.montana.edu
ReceiVed NoVember 4, 2008
The development of a new glycosylation method for the stereoselective synthesis of ꢀ-glycosides in the
absence of the traditional C(2)-ester neighboring group effect is described. This process relies on the
ability of the cationic palladium catalyst, Pd(PhCN)2(OTf)2 generated in situ from Pd(PhCN)2Cl2 and
AgOTf, to direct ꢀ-selectivity. The new glycosylation reaction is highly ꢀ-selective and proceeds under
mild conditions with 1-2 mol % of catalyst loading. This ꢀ-glycosylation method has been applied to
a number of glucose donors with benzyl, allyl, and p-methoxybenzyl groups incorporated at the C(2)-
position as well as tribenzylated xylose and quinovose donors to prepare various disaccharides and
trisaccharides with good to excellent ꢀ-selectivity. Mechanistic studies suggest that the major operative
pathway is likely to proceed via a seven-membered ring intermediate, wherein the cationic palladium
complex coordinates to both the C(1)-imidate nitrogen and C(2)-oxygen of the trichloroacetimidate donor.
Formation of this seven-membered ring intermediate directs the selectivity, leading to the formation of
ꢀ-glycosides.
Introduction
competitive formation of ortho ester.4 Furthermore, if the
glycosylation is performed under basic conditions, the C(2)-
acyl functionality can migrate to both the C(1)-position of
glycosyl donors as well as the reactive sites of the nucleophilic
acceptors.5 To avoid such problems, Demchenko and co-workers
have recently reported the stereoselective synthesis of ꢀ-gly-
cosides utilizing C(2)-picolyl moiety as a novel neighboring
participatory group.6 Crich and co-workers have also reported
the use of a 3,4-O-bisacetal system as well as a trans-2,3-O-
carbonate group to influence ꢀ-glycosylation.7
On the surface, the stereoselective synthesis of ꢀ-glycosides
appears to be a straightforward task due to the effect of
traditional C(2)-neighboring group participation.1 Ester func-
tionalities are often employed as participatory groups at the C(2)-
positions of glycosyl donors. However, the reactivity of glycosyl
donors incorporating C(2)-ester functionality is significantly
decreased within many glycosylation methods.2 Thus, prolonged
reaction time is often required in order to achieve efficient
coupling.3 This popular approach can also suffer from the
(4) (a) Kunz, H.; Harreus, A. Liebigs Ann. Chem. 1982, 41. (b) Sato, S.;
Nunomura, S.; Nakano, T.; Ito, Y.; Ogawa, T. Tetrahedron Lett. 1988, 33, 4097.
(c) Seeberger, P. H.; Eckhardt, M.; Gutteridge, C. E.; Danishefsky, S. J. J. Am.
Chem. Soc. 1997, 119, 10064. (d) Crich, D.; Dai, Z.; Gastaldi, S. J. Org. Chem.
1999, 64, 5224–5229.
(5) Yu, H.; Ensley, H. E. Tetrahedron Lett. 2003, 4, 9363–9366.
(6) Smoot, J. T.; Pornsuriyasak, P.; Demchenko, A. V. Angew. Chem., Int.
Ed. 2005, 44, 7123–7126.
(1) (a) Nukada, T.; Berces, A.; Zgierski, M. Z.; Whitfield, D. M. J. Am.
Chem. Soc. 1998, 120, 13291–13295. (b) Barresi, F.; Hindsgaul, O. J. Carbohydr.
Chem. 1995, 14, 1043. (c) Schmidt, R. R.; Kinzy, W. AdV. Carbohydr. Chem.
Biochem. 1994, 50, 21–123. (d) Schmidt, R. R.; Michel, J. Angew. Chem., Int.
Ed. 1980, 19, 731–732.
(2) (a) Fraser-Reid, B.; Wu, Z.; Udodong, U. E.; Ottosson, H. J. Org. Chem.
1990, 55, 6068–6070. (b) Fraser-Reid, B.; Merritt, J. R.; Handlon, A. L.;
Andrews, C. W. Pure Appl. Chem. 1993, 65, 779–786.
(7) (a) Crich, D.; Subramanian, V.; Hutton, T. K. Tetrahedron 2007, 63,
5042–5049. (b) Crich, D.; Jayalath, P. J. Org. Chem. 2005, 70, 7252–7259.
(3) Toshima, K.; Tatsuta, K. Chem. ReV. 1993, 93, 1503–1531.
1650 J. Org. Chem. 2009, 74, 1650–1657
10.1021/jo802468p CCC: $40.75 2009 American Chemical Society
Published on Web 01/22/2009

StereoselectiVe Synthesis of ꢀ-Glycosides
SCHEME 1
TABLE 1. Pd(PhCN)2(OTf)2-Controlled ꢀ-Selective Glycosylationa
SCHEME 2
palladium, AgOTf,
temp,
°C
yield,b
%
entry
mol %
mol % additive
time
ꢀ:Rc
1
2
3
4
5
6
7
2
1
1
1
1
1
none
4
2
2
2
2
none
2
none
none
none
none
DTBP
none
none
25 15 min
25 15 min
98
96
83
87
85
<1
<1
1:1
1:1
1:1
10:1
10:1
On the other hand, ether protecting groups at the C(2)-position
of glycosyl donors have been explored in many glycosylation
methods (e.g., glycosyl trichloroacetimidate 1) because they
enhance the reactivity of glycosyl donors (Scheme 1). However,
since these C(2)-ether protected glycosyl donors appear to go
through an oxocarbenium intermediate, a mixture of R- and
ꢀ-glycoside products are often formed in the reaction. There
are few reports on the use of nitrile solvent to improve the
ꢀ-selectivity when a glycosyl donor bearing a nonparticipatory
group at the C(2)-position is employed in the glycosylation
reaction.8
0
30 min
-78 1 h
-78 1 h
-78 8 h
-78 5 h
a
All reactions were carried out in CH2Cl2 (0.15 M) with AgOTf and
Pd(PhCN)2Cl2. b Isolated yield. 1H NMR ratio.
c
SCHEME 3
Our strategy for ꢀ-glycosylation is to exploit the ability of
cationic palladium catalyst to direct ꢀ-glycosylation through
coordinating to both the imidate nitrogen and C(2)-oxygen of
glycosyl trichloroacetimidate donors (Scheme 1). We report
herein a novel method of cationic palladium controlled ꢀ-gly-
cosylation in the absence of the traditional C(2)-ester neighbor-
ing group effect. Due to mild reaction conditions with the
anomeric selectivity controlled by the nature of the cationic
palladium complex, this strategy represents a promising method
for constructing ꢀ-glycoside products.
Although the above conditions provide disaccharide 6 in good
yield with a good level of ꢀ-selectivity, the cationic palladium(II)
species, Pd(CH3CN)4(BF4)2, is relatively expensive, and at least
5 mol % of the catalyst is required for the reaction to go to
completion. Thus, an ongoing goal in our group has been the
development of stereoselective glycosylation reactions that
utilize more reactive palladium catalysts.11 Our approach to this
problem is to carefully look at the nature of counterions on
cationic Pd(II) complexes. It is known that a weakly coordinat-
ing counterion can increase catalyst activity.12 With this
hypothesis in mind, we investigated the use of cationic
Pd(PhCN)2(OTf)2, generated in situ from neutral Pd(PhCN)2Cl2
and AgOTf. We choose to use AgOTf because it is easier to
handle than other silver salts.
Coupling of nucleophilic acceptor 5 with trichloroacetimidate
4 in the presence of a preformed solution of 2 mol % of
Pd(PhCN)2Cl2 and 4 mol % of AgOTf at 25 °C provided the
desired disaccharide 6 in 98% yield as a 1:1 mixture of R- and
ꢀ-isomers (Table 1, entry 1). Decreasing the catalyst loading
to 1 mol % still maintained the yield. However, disaccharide 6
was still isolated as a mixture of R- and ꢀ-isomers (entry 2).
Results and Discussion
Initial studies were performed with 2,3,4,6-tetra-O-benzyl-
D-glycopyranosyl trichloroacetimidate 49 as the glycosyl donor
and 1,2:3,4-di-O-isopropylidene-D-galactopyranose 5 as the
nucleophilic acceptor.10 Upon treatment of both coupling
partners 4 and 5 with 5 mol % of commercially available
cationic Pd(II) species, Pd(CH3CN)4(BF4)2, in CH2Cl2 at 25 °C
for 3 h, the desired disaccharide 6 was isolated in 72% yield
with ꢀ:R ) 7:1 (Scheme 2). This result was encouraging because
it clearly showed that the cationic Pd(II) catalyst could direct
ꢀ-glycosylation with the C(2)-benzyl protected donor to provide
the desired glycoside with good stereoselectivity.
(8) (a) Vankar, Y. D.; Vankar, P. S.; Behrendt, M.; Schmidt, R. R.
Tetrahedron 1991, 47, 9985–9992. (b) Marra, A.; Esnault, J.; Veyrieres, A.;
Sinary, P. J. Am. Chem. Soc. 1992, 114, 6354–6360. (c) Crich, D.; Patel, M.
Carbohydr. Res. 2006, 341, 1467–1475.
(9) (a) Zhang, J.; Yergey, A.; Kowalak, J.; Kovac, P. Tetrahedron 1998, 54,
11783–11792. (b) Patil, V. J. Tetrahedron Lett. 1996, 37, 1481–1484. (c)
Nakanishi, N.; Nagafuchi, Y.; Koizumi, K. J. Carbohydr. Chem. 1994, 13, 981–
990. (d) Rathore, H.; Hashimoto, T.; Igarashi, K.; Nukaya, H.; Fullerton, D.
Tetrahedron 1985, 41, 5427–5438. (e) Schmidt, R. R.; Michel, J. Tetrahedron.
Lett. 1984, 25, 821–824.
(11) (a) Schuff, B. P.; Mercer, G. J.; Nguyen, H. M. Org. Lett. 2007, 9,
3173–3176. (b) Mercer, G. J; Yang, J.; McKay, M. J.; Nguyen, H. M. J. Am.
Chem. Soc. 2008, 130, 11210–11218.
(10) Yang, J.; Cooper-Vanosdell, C.; Mensah, E. A.; Nguyen, H. M. J. Org.
Chem. 2008, 73, 794–800.
(12) (a) Reed, C. A. Acc. Chem. Res. 1998, 31, 133–139. (b) Mecking, S.
Coord. Chem. ReV. 2000, 203, 325–351.
J. Org. Chem. Vol. 74, No. 4, 2009 1651

Mensah et al.
SCHEME 4
TABLE 2. ꢀ-Selective Glycosylation with C(2)-Benzyl Glucose
Donorsa
SCHEME 5
Lowering the reaction temperature to 0 °C did not have any
effect on the outcome of the reaction (entry 3). Since the
glycosylation reaction was complete within 30 min at 0 and 25
°C, we decided to lower the reaction temperature further to -78
°C. We were surprised to find that the ꢀ-selectivity was
significantly increased at -78 °C (entry 4). In this event,
disaccharide 6 was isolated in 87% yield with ꢀ:R ) 10:1. The
increase in ꢀ-selectivity from 1:1 to 10:1 observed upon cooling
the reaction mixture from 0 to -78 °C suggests different
mechanisms for this glycosylation process. At 0 °C, the reaction
appears to go through an oxocarbenium intermediate, resulting
in a mixture of R- and ꢀ-isomers 6. In contrast, the reaction is
likely to go through an SN2 displacement at - 78 °C wherein
glycosyl acceptor 5 selectively approaches the ꢀ-face of
trichloroacetimidate 4, leading to the formation of ꢀ-glycoside
6.
a All reactions were performed in CH2Cl2 (0.15 M) with 1 mol % of
To determine if triflic acid, which may be generated from
Pd(PhCN)2(OTf)2, is the source of catalysis, the glycosylation
was performed in the presence of 2,6-di-tert-butylpyridine
(DTBP) (10 mol %) as an acid scavenger (Table 1, entry 5). In
this experiment, the desired disaccharide 6 was isolated in
comparable yield and selectivity to that of entry 4. A similar
glycosylation was attempted with neutral Pd(II) species,
Pd(PhCN)2Cl2 (entry 6); however, less than 1% conversion was
observed in the reaction. Since a stoichiometric amount of
AgOTf has been used to activate trichloroacetimidate,13 a control
experiment was performed to determine if trichloroacetimidate
4 was indeed activated by 2 mol % of AgOTf at -78 °C (entry
7). Less than 1% yield of disaccharide 6 was detected from
this experiment. We also investigated whether Pd(PhCN)2(OTf)2
could activate the ꢀ-isomer of trichloroacetimidate 4, and less
than 1% conversion was observed in the reaction at -78 °C.
Overall, Pd(PhCN)2(OTf)2 provided disaccharide 6 with
higher yield and ꢀ-selectivity than that of Pd(CH3CN)4(BF4)2.
In addition, the coupling process requires only 1 mol % of
c
Pd(PhCN)2Cl2 and 2 mol % of AgOTf for 1-3 h. b Isolated yield. 1H
NMR ratio.
Pd(CH3CN)4(BF4)2. Compared to other glycosylation procedures
utilizing trichloroacetimidate 4 as glycosyl donor, higher
ꢀ-selectivity and lower catalyst loading were observed with use
of Pd(PhCN)2(OTf)2 as the activator. For example, disaccharide
6 was obtained in 45% yield as a mixture of R- and ꢀ-isomers
with use of LiClO4 as a promoter.14 On the other hand, with
use of LiOTf as the catalyst, disaccharide 6 was obtained in
77% yield as a 1:1 mixture of R- and ꢀ-isomers.15 When a
moisture-stable activating reagent, Yb(OTf)3, was employed in
the glycosylation, disaccharide 6 was isolated in 90% as a 2:1
mixture of R- and ꢀ-isomers.16 With use of traditional Lewis
acid, TMSOTf, as an activator, disaccharide 6 was obtained in
71% yield as a 4:1 mixture of R- and ꢀ-isomers.17 Recently,
(14) Waldmann, H.; Bohm, G.; Schmid, U. Angew. Chem., Int. Ed. 1994,
33, 1944–1946.
(15) Lubineau, A.; Drouillat, B. J. Carbohydr. Chem. 1997, 16, 1179–1186.
(16) Adinolfi, M.; Iadonisi, A.; Ravida, A. Synlett 2006, 583–586.
(17) Meloncelli, P. J.; Williams, T. M.; Hartmann, P. E.; Stick, R. V.
Carbohydr. Res. 2007, 342, 1793–1804.
Pd(PhCN)2(OTf)2 in comparison to
5
mol
%
with
(13) Douglas, S. P.; Whitfield, D. M.; Krepinksy, J. J. J. Carbohydr. Chem.
1993, 12, 131–136.
1652 J. Org. Chem. Vol. 74, No. 4, 2009

StereoselectiVe Synthesis of ꢀ-Glycosides
TABLE 3. ꢀ-Selective Glycosylation with C(2)-Allyl Glucose
TABLE 4. ꢀ-Selective Glycosylation with C(2)-PMB-Glucose
Donorsa
Donora
a All reactions were performed in CH2Cl2 (0.15 M) with 2 mol % of
c
Pd(PhCN)2Cl2 and 4 mol % of AgOTf. b Isolated yield. 1H NMR ratio.
conditions were comparable to those under the Pd(PhCN)2(OTf)2
conditions, the glycosylation was performed with higher catalyst
loading (10 mol %) and temperature (-25 °C). Additionally,
acetonitrile was employed as the solvent to influence the
ꢀ-selectivity. On the other hand, when THF or a mixture of
THF and CH3CN (1:4) was employed as the solvent in the
glycosylation, disaccharide 6 was obtained as a 1:3 mixture of
R- and ꢀ-isomers.
The glycosyl trichloroacetimidates are generally activated
with strong and moisture sensitive Lewis acids such as
BF3 ·OEt2,19 TMSOTf,20 TBSOTf,21 Tf2O,22 and ZnBr2;23
BF3 ·OEt2 has been among the widely used Lewis acids.24
Glycosylation of galactosyl nucleophile 5 with trichloroace-
timidate 4 was performed with 1 mol % of BF3 ·OEt2 at -78
°C so that we can determine the reactivity and selectivity of
BF3 · OEt2 compared to that of cationic Pd(II) species,
(19) (a) Schmidt, R. R.; Michel, J. J. Carbohydr. Chem. 1985, 4, 141–169.
(b) Zimmermann, P.; Bommer, R.; Ba¨r, T.; Schmidt, R. R. J. Carbohydr. Chem.
1988, 7, 435–452.
(20) (a) Schmidt, R. R.; Behrendt, M.; Toepfer, A. Synlett 1990, 694–696.
(b) Schaubach, R.; Hemberger, J.; Kinzy, W. Liebigs Ann. Chem. 1991, 607–
614. (c) Paulsen, H.; Wilkens, R.; Reck, F.; Brockhausen, I. Liebigs Ann. Chem.
1992, 1303–1313. (d) Kulkarni, S. S.; Hung, S.-C. Lett. Org. Chem. 2005, 2,
670–677.
(21) Haase, W. C.; Seeberger, P. H. Curr. Org. Chem. 2000, 4, 481–511.
(22) Dobarro-Rodriguez, A.; Trumtel, M.; Wessel, H. P. J. Carbohydr. Chem.
1992, 11, 255–263.
(23) Ubran, F. J.; Moore, B. S.; Breitenbach, R. Tetrahedron Lett. 1990, 31,
4421–4424.
(24) (a) Halcomb, R. L.; Boyer, S. H.; Danishefsky, S. J. Angew. Chem.,
Int. Ed. 1992, 31, 338–340. (b) Mayer, T. G.; Kratzer, B.; Schmidt, R. R. Angew.
Chem., Int. Ed. 1994, 33, 2177–2181. (c) Kim, Y.-J.; Wang, P.; Navarro-
Villalobos, M.; Rohde, B. D.; Derryberry, J.; Gin, D. Y. J. Am. Chem. Soc.
2006, 128, 11906–11915.
a All reactions were performed in CH2Cl2 (0.15 M) with 1 mol % of
c
Pd(PhCN)2Cl2 and 2 mol % of AgOTf. b Isolated yield. 1H NMR ratio.
Iadonisi and co-workers reported the use of Sm(OTf)3 to activate
glycosyl trichloroacetimidate 4.18 Although the yield and
ꢀ-selectivity of the desired product 6 under the Sm(OTf)3
(18) Adinolfi, M.; Barone, G.; Guariniello, L.; Iadonisi, A. Tetrahedron. Lett.
2000, 41, 9005–9008.
J. Org. Chem. Vol. 74, No. 4, 2009 1653

Mensah et al.
TABLE 5. Selective Glycosylation with Benzylated Xylose and
Quinovose Donorsa
Under standard cationic palladium conditions, disaccharides
12-15 were isolated in good yields with excellent ꢀ-selectivity
(Table 2). With use of dihydrocholesterol 8 as a glycosyl
acceptor, the glycosylation reaction was performed at -45 °C
as the reaction solidified at -78 °C (entry 2). We also tried to
compare our cationic palladium results with other Lewis acid
results.1b Apart from glycosyl acceptor 8, most of these acceptors
have not been investigated with perbenzylated trichloroacetimi-
date 4 under Lewis acid conditions. In the case of glycosyl
nucleophile 8, Pd(PhCN)2(OTf)2 provided disaccharide 19 with
higher yield and selectivity than that of BF3 ·OEt2 (Table 2, entry
3).15 In addition, the reaction required lower catalyst loading
of Pd(PhCN)2(OTf)2 in comparison to BF3 ·OEt2. Furthermore,
the higher temperature reaction (-40 to 25 °C) was required
with use of BF3 ·OEt2 as an activating reagent.
Our next step was to determine how other ether protecting
groups at the C(2)-position of glycosyl donors would affect the
ꢀ-selectivity. Accordingly, the C(2)-allyl donors were examined
with a variety of nucleophilic acceptors 7-11. As shown in
Table 3, the desired disaccharides 16-26 were also obtained
in good yields with good to excellent ꢀ-selectivity. These results
are encouraging because they suggest that cationic palladium
catalyst selectively coordinates to the C(1)-imidate nitrogen in
the presence of the allyl protecting group.
The results obtained in Tables 2 and 3 suggest that the
C(2)-ether protecting groups (benzyl and allyl) in conjunction
with the cationic palladium catalyst play important roles in
controlling the ꢀ-selectivity at the newly formed glycosidic
bonds. We also investigated the ability of the p-methoxyb-
nezyl group to direct ꢀ-glycosylation since this protecting
group has been extensively employed in glycosylation
reaction with no or little effect on reactivity and selectivity
in comparison to the benzyl group. The p-methoxybenzyl
group can be orthogonally removed in the presence of the
benzyl protecting group. Accordingly, selective coupling of
2,3,4,6-tetra-O-p-methoxybenzyl-D-glucopyranosyl trichlo-
roacetimidate 27 with an array of nucleophilic acceptors
provided the desired disaccharides 28-31 in good ꢀ-selectiv-
ity (Table 4). In these reactions, 2 mol % of the cationic
palladium catalyst, Pd(PhCN)2(OTf)2, was required for the
coupling to go to completion at -78 °C. Overall, the reaction
of the p-methoxylbenzyl-protected donor 27 is less selective
than that of the benzyl-protected donor 4.
a All reactions (except entries 2 and 4) were carried out in CH2Cl2
(0.15 M) with AgOTf and Pd(PhCN)2Cl2 at -78 °C. b Isolated yield.
c
1H NMR ratio.
Pd(PhCN)2(OTf)2 (Scheme 3). In this event, less than 5%
conversion was observed even after the reaction had been stirred
for 8 h.
The outcome of the Lewis acid experiment (Scheme 3)
suggests that there is no turnover with use of BF3 ·OEt2 as a
catalyst. In this glycosylation reaction, trichloroacetamide is
generated as the byproduct (Scheme 4). Since BF3 ·OEt2 is quite
oxophilic, it preferentially coordinates to the carbonyl oxygen
of trichloroacetamide. Thus, this byproduct inhibition prohibits
BF3 ·OEt2 from turnover. On the other hand, trichloroacetamide
is not a good ligand for Pd(PhCN)2(OTf)2, thereby resulting in
high turnover of the palladium catalyst.
Encouraged by the results obtained with D-glucose donors,
we decided to explore the feasibility of our cationic palladium
method with tribenzylated D-xylose trichloroacetimidate
donor 32 and D-quinovose trichloroacetimidate donor 33
(Table 5). These substrates are the carbohydrate moieties
found within biologically important oligosaccharide natural
To assess the scope of this new glycosylation method to direct
the stereoselective formation of ꢀ-glycosidic linkages in the
absence of the traditional C(2)-ester participatory group, a
number of primary, secondary, and tertiary hydroxyls of
carbohydrate acceptors (Scheme 5) were investigated with
perbenzylated trichloroacetimidate donor 4.
SCHEME 6
1654 J. Org. Chem. Vol. 74, No. 4, 2009

StereoselectiVe Synthesis of ꢀ-Glycosides
FIGURE 1. Proposed cationic Pd(II) directed ꢀ-stereoselective glycosylation.
SCHEME 7
SCHEME 8
products.25 Since these substrates lack the protected C(6)-
hydroxyl functionality, we are not sure how this would affect
the ꢀ-selectivity. Gratifyingly, it was found that both
trichloroacetimidates 32 and 33 were able to couple with a
variety of primary and hindered alcohols to provide the
desired disaccharides 34-38 in good yields with moderate
to good ꢀ-selectivity (Table 5).
Many glycosylation methods work well to make disaccharides
but potentially break down with oligosaccharides. Therefore, it
is worth investigating the scope of the cationic palladium
controlled ꢀ-glycosylation in the context of trisaccharide
synthesis (Scheme 6). In a [1+2] approach, glycosylation of
disaccharide nucleophile 399 with trichloroacetimidate donor 4
under standard cationic palladium conditions provided the
corresponding trisaccharide 40 in 71% yield with excellent
ꢀ-selectivity (Scheme 6). This result illustrates the efficiency
of this cationic palladium methodology, and it does so with the
formation of 40 in excellent anomeric selectivity. In addition,
the coupling was performed at -78 °C with only 2 mol % of
Pd(PhCN)2(OTf)2 as the activating reagent.
A proposed mechanism for cationic palladium controlled
ꢀ-glycosylation in the absence of the traditional C(2)-ester
participatory group is outlined in Figure 1. It is hypothesized
that the reaction might operate via a seven-membered-ring
intermediate such as 41, wherein the cationic Pd(II) species
coordinates to both C(1)-imidate nitrogen and C(2)-oxygen of
trichloroacetimidate donor 2 (pathway a). Formation of this
seven-membered-ring intermediate directs the selectivity, leading
to the formation of ꢀ-glycoside 3. However, the possibility of
the glycosylation reaction occurring through a simple SN2
displacement without internal chelation cannot be ruled out
(pathway b).
to the C(2)-oxygen of glycosyl donor 43. Accordingly, treatment
of both coupling partners 5 and 43 under the standard cationic
palladium(II) conditions provided the corresponding disaccha-
ride 44 in 54% yield with ꢀ:R ) 3:1 (Scheme 7). Thus, evidence
suggests that the operative pathway is likely to be the seven-
membered-ring intermediate (Figure 1, pathway a), in which
the initial step of activation involves the coordination of Pd(II)
species to both C(1)-imidate nitrogen and C(2)-oxygen of
glycosyl donor.
A second experiment was carried out employing C(2)-deoxy
glycosyl donor 45 (Scheme 8). Since there is no oxygen
functionality at the C(2)-position of 46, formation of a seven-
membered-ring intermediate such as 41 (Figure 1, pathway a)
cannot occur. Thus, if the seven-membered-ring pathway is
operative, disaccharide 46 will be formed as a mixture of R-
and ꢀ-isomers. On the other hand, if the simple SN2 pathway
without internal chelation is operative (Figure 1, pathway b),
ꢀ-isomer of 46 will be formed as a major product. Accordingly,
glycosylation of 5 with 45 provided disaccharide 46 as a mixture
of R- and ꢀ-isomers (ꢀ:R ) 2:1) (Scheme 8). The outcome of
this experiment suggests that the SN2 pathway without internal
chelation is unlikely to be the major operative pathway under
cationic palladium conditions.
The above experiments allow us to establish the possible
mechanism of cationic palladium(II) catalyst directed ꢀ-selective
glycosylation. The results obtained from C(2)-TIPS donor 43
(Scheme 7) and C(2)-deoxy donor 45 (Scheme 8) support that
the seven-membered-ring pathway, wherein the palladium(II)
species coordinates to both the C(1)-imidate nitrogen and C(2)-
oxygen of glycosyl trichloroacetimidate donor, is likely to be
the major operative pathway.
To differentiate between the two glycosylation reaction
pathways, two control experiments were performed. In the first
experiment, galactose acceptor 5 was coupled with C(2)-
triisopropylsilyl protected donor 43 (Scheme 7). If the seven-
membered-ring intermediate such as 41 (Figure 1) is the
operative pathway, it is anticipated that disaccharide 44 will be
formed as a mixture of R- and ꢀ-isomers because the bulky
TIPS group prevents the palladium catalyst from coordinating
Conclusions
In summary, a novel method for stereoselective synthesis of
ꢀ-glycosides in the absence of the traditional C(2)-ester
neighboring group participation has been developed. This
approach relies on the nature of the cationic palladium(II)
complex, Pd(PhCN)2(OTf)2, to control the ꢀ-selectivity. This
new ꢀ-glycosylation method is applicable to a variety of glucose
donors incorporating benzyl, allyl, and p-methoxybenzyl ether
protecting groups at the C(2)-position as well as tribenzylated
xylose and quinovose donors. The reaction is highly ꢀ-selective
and proceeds under low temperature (-78 °C) with low catalyst
(25) (a) Pereda-Miranda, R.; Kaatz, G. W.; Gibbons, J. J. Nat. Prod. 2006,
69, 406–409. (b) Lee, K.-H.; Morris-Natschke, S. L. Pure. Appl. Chem. 1999,
71, 1045–1051. (c) Sun, I.-C.; Kashiwada, Y.; Morris-Natschke, S. L.; Lee, K.-
H. Curr. Med. Chem. 2003, 3, 155–169.
J. Org. Chem. Vol. 74, No. 4, 2009 1655

Mensah et al.
128.9, 128.7, 128.5, 128.4, 128.3, 128.2, 128.1, 128.0, 127.9,
127.8, 127.7, 127.69, 127.6, 127.5, 109.1, 108.3, 103.8, 101.3,
96.1, 84.6, 82.0, 77.6, 75.7, 74.9, 74.8, 74.7, 74.0, 73.4, 73.1,
71.8, 70.8, 70.5, 70.4, 70.2, 68.7, 68.6, 68.5, 67.3, 25.8, 25.6,
24.8, 24.1; IR (film, cm-1) ν 3063, 3030, 2986, 2922, 1733,
1452, 1281, 1260, 1093, 1069, 1027, 709; J(13CH) ) 166 Hz
(94.4 ppm); HRMS (ESI) calcd for C73H76O19Na (M + Na)
1279.48730, found 1279.47682.
loading (1-2 mol %). The efficiency of cationic palladium
conditions has been applied to the synthesis of a number of
disaccharides and trisaccharides with moderate to excellent
ꢀ-selectivity. Mechanistic studies suggest that the reaction may
operate via a seven-membered-ring pathway, in which the
cationic Pd(II) species coordinates to both the C(1)-imidate
nitrogen and C(2)-oxygen of glycosyl trichloroacetimidate
donors. Formation of this seven-membered-ring intermediate
directs the anomeric selectivity, leading to the formation of
ꢀ-glycosides.
Preparation of Disaccharide 44. A 10 mL oven-dried Schlenk
flask was charged with 3,4,6-tribenzyl-2-triisopropylsilyl-O-D-
glucopyranosyl trichloroacetimidate 43 (60 mg, 0.080 mmol, 1
equiv), 1,2:3,4-di-O-isopropylidene-D-galactopyranose 5 (27 mg,
0.10 mmol, 1.3 equiv), and CH2Cl2 (0.6 mL). The resulting solution
was cooled to -78 °C, and a preformed solution of Pd(PhCN)2Cl2
(0.31 mg, 0.0006 mmol, 1 mol %) and AgOTf (0.41 mg, 0.0016
mmol, 2 mol %) in CH2Cl2 (0.2 mL) was then added. The reaction
mixture was stirred at -78 °C for 5 h, diluted with benzene (1
mL), and purified by silica gel flash chromatography (6/1, hexane/
ethyl acetate) to give the desired disaccharide 44 (37 mg, 54%,
ꢀ:R ) 3:1) as a pale yellow oil. Rf 0.36 (hexanes/ethyl acetate,
6/1); 1H NMR (CDCl3, 500 MHz) δ 7.34-7.26 (m, 9H), 7.22-7.20
(m, 4H), 7.07 (t, J ) 3.5 Hz, 1H), 7.02-7.01 (m, 1H), 5.49 (d, J
) 5 Hz, 1H), 5.48 (d, J ) 5 Hz, 0.3H), 4.93 (d, J ) 11.5 Hz, 1H),
4.87 (d, J ) 11 Hz, 1H), 4.83-4.80 (m, 1H), 4.74-4.69 (m, 1H),
4.65-4.62 (m, 1H), 4.59-4.58 (m, 1H), 4.55-4.52 (m, 1H),
4.49-4.45 (m, 1H), 4.42 (d, J ) 10.5 Hz, 1H), 4.29-4.25 (m,
2H), 4.11 (dd, J ) 10, 5 Hz, 0.3H), 3.99 (t, J ) 7 Hz, 1H), 3.89
(d, J ) 3 Hz, 0.3H), 3.88 (d, J ) 3.5 Hz, 1H), 3.85 (d, J ) 8.5 Hz,
1H), 3.83-3.80 (m, 1H), 3.77-3.74 (m, 2H), 3.71-3.69 (m, 1H),
3.67-3.60 (m, 3H), 3.56 (dd, J ) 11, 7.5 Hz, 0.3H), 3.51 (t, J )
8.5 Hz, 0.3H), 3.44 (d, J ) 10.5 Hz, 0.3H), 1.51 (d, J ) 16.5 Hz,
3H), 1.40 (s, 3H), 1.32-1.29 (m, 6H), 1.21-1.16 (m, 1H),
1.06-1.03 (m, 20H); 13C NMR (CDCl3, 125 MHz) δ 139.3, 139.0,
138.4, 138.2, 138.0, 128.3, 128.2, 128.1, 128.08, 128.03, 127.95,
127.83, 127.8, 127.64, 127.6, 127.5, 127.2, 127.1, 109.2, 109.1,
108.6, 108.4, 103.7, 99.8, 96.32, 96.25, 86.6, 82.8, 78.2, 78.0, 75.6,
75.4, 75.2, 74.9, 74.8, 74.7, 73.9, 73.6, 73.5, 71.5, 70.9, 70.7, 70.5,
70.47, 70.2, 68.8, 68.5, 67.1, 66.74, 66.66, 26.1, 26.0, 25.99, 24.98,
24.6, 24.4, 18.2, 18.11, 18.06, 13.1, 12.7; IR (film, cm-1) ν 2939,
2868, 1498, 1457, 1377, 1252, 1214, 1160, 1005, 917, 883, 735,
692; HRMS (ESI) calcd for C48H68O11Si Na (M + Na) 871.44231,
found 871.44197.
Experimental Section
Representative experimental procedures are listed here. Full
experimental details and spectral data for all new compounds can
be found within the Supporting Information.
Preparation of Disaccharide 6. A 10 mL oven-dried Schlenk
flask was charged with 2,3,4,6-tetra-O-benzyl-D-glucopyranosyl
trichloroacetimidate donor 4 (102 mg, 0.15 mmol, 1 equiv), 1,2:
3,4-di-O-isopropylidene-D-galactopyranose 5 (51 mg, 0.195
mmol, 1.3 equiv), and CH2Cl2 (0.8 mL). The resulting solution
was cooled to -78 °C, and
a preformed solution of
Pd(PhCN)2Cl2 (0.58 mg, 0.0015 mmol, 1 mol %) and AgOTf
(0.77 mg, 0.003 mmol, 2 mol %) in CH2Cl2 (0.2 mL) was then
added. The reaction mixture was stirred at -78 °C for 1 h,
diluted with benzene (1 mL), and purified by silica gel flash
chromatography (2/1, hexane/ethyl acetate) to give the desired
disaccharide 6 (103 mg, 87%, ꢀ:R ) 10:1) as a pale yellow oil.
Rf 0.36 (hexanes/ethyl acetate, 4/1); 1H NMR (CDCl3, 500 MHz)
δ 7.44-7.14 (m, 20H), 5.58 (d, J ) 5.0 Hz, 1H), 5.07 (d, J )
11 Hz, 1H), 4.97 (d, J ) 11 Hz, 1H), 4.82 (d, J ) 11 Hz, 1H),
4.79 (d, J ) 11 Hz, 1H), 4.73 (d, J ) 11 Hz, 1H), 4.65-4.58
(m, 2H), 4.55 (s, 1H), 4.51 (d, J ) 11 Hz, 1H), 4.47 (d, J ) 7.5
Hz, 1H), 4.33-4.31 (m, 1H), 4.25 (d, J ) 8 Hz, 1H), 4.17 (dd,
J ) 10.5, 3.5 Hz, 1H), 4.1 (m, 1H), 3.83-3.68 (m, 3H),
3.67-3.58 (m, 2H), 3.49-3.45 (m, 2H), 1.51 (s, 3H), 1.46 (s,
3H), 1.32 (2s, 6H);. 1H NMR matches with the literature report,25
13C NMR (CDCl3, 125 MHz) δ 138.7, 138.2, 128.7, 128.4, 128.2,
128.0, 127.9, 127.7, 127.6, 127.55, 127.49, 109.4, 108.6, 104.4,
96.4, 84.6, 81.6, 77.7, 75.7, 75.0, 74.8, 74.4, 73.5, 71.5,
70.8, 70.5, 69.7, 68.8, 67.4, 26.1, 26.0, 25.1, 24.5; 13C NMR
matches with the literature report;25 J(13CH) ) 156 Hz (104.4
ppm), 179 Hz (96.4 ppm); IR (film, cm-1) ν 2902, 1454, 1381,
1255, 1211.
Preparation of Disaccharide 46. A 10 mL oven-dried Schlenk
flask was charged with 2-deoxy-3,4,6-tri- benzyl-O-D-glucopy-
ranosyl trichloroacetimidate 45 (75 mg, 0.129 mmol, 1 equiv),
1,2:3,4-di-O-isopropylidene-D-galactopyranose 5 (44 mg, 0.168
mmol, 1.3 equiv), and CH2Cl2 (1 mL). The resulting solution
Preparation of Trisaccharide 40. A 10 mL oven-dried Schlenk
flask was charged with 2,3,4,6-tetra-O-benzyl-D-glucopyranosyl
trichloroacetimidate donor 4 (56 mg, 0.082 mmol, 1 equiv),
disaccharide acceptor 39 (66 mg, 0.090 mmol, 1.1 equiv), and
CH2Cl2 (0.6 mL). The resulting solution was cooled to -78 °C,
and a preformed solution of Pd(PhCN)2Cl2 (0.62 mg, 0.0016
mmol, 2 mol %) and AgOTf (0.84 mg, 0.0032 mmol, 4 mol %)
in CH2Cl2 (0.2 mL) was then added. The reaction mixture was
stirred at -78 °C for 5 h, diluted with benzene (1 mL), and
purified by silica gel flash chromatography (2/1, hexane/ethyl
acetate) to give the desired disaccharide 40 (73 mg, 71%, ꢀ:R
was cooled to -78 °C, and
a preformed solution of
Pd(PhCN)2Cl2 (0.50 mg, 0.00129 mmol, 1 mol %) and AgOTf
(0.66 mg, 0.00258 mmol, 2 mol %) in CH2Cl2 (0.2 mL) was
then added. The reaction mixture was stirred at -78 °C for 2 h,
diluted with benzene (1 mL), and purified by silica gel flash
chromatography (4/1, hexane/ethyl acetate) to give the desired
disaccharide 46 (83 mg, 95%, ꢀ:R ) 3:1) as a pale yellow oil.
Rf 0.32 (hexanes/ethyl acetate, 4/1); 1H NMR (CDCl3, 500 MHz)
δ 7.32-7.27 (m, 14), 7.18 (d, J ) 4.5 Hz), 5.54 (d, J ) 21 Hz,
1H), 5.53 (d, J ) 19.5, 0.3H), 5.03 (s, 0.3H), 4.90-4.87 (m,
2H), 4.69-4.58 (m, 6H), 4.54-4.50 (m, 5H), 4.31 (s, 2H), 4.21
(d, J ) 6.5 Hz, 2H), 4.09 (d, J ) 11.5 Hz, 1H), 4.00-3.94 (m,
2H), 3.79 (t, J ) 9 Hz, 1H), 3.72-3.65 (m, 6H), 3.54 (t, J ) 9
Hz, 1H), 3.39 (d, J ) 8 Hz, 1H), 2.48 (d, J ) 11 Hz, 1H), 2.33
(d, J ) 13 Hz, 0.3H), 1.74 (d, J ) 12.5 Hz, 0.3H), 1.65 (dd, J
) 23, 12 Hz, 1H), 1.53 (s, 3H), 1.51 (s, 1H), 1.43 (s, 5H), 1.32
(s, 9H); 13C NMR (CDCl3, 125 MHz) δ 138.7, 138.5, 138.4,
138.3, 138.2, 138.1, 138, 128.4, 128.4, 128.0, 127.9, 127.8,
127.7, 127.6, 127.5, 109.4, 109.3, 108.7, 108.6, 100.5, 97.3, 96.3,
79.3, 78.2, 77.9, 77.5, 75.0, 74.9, 71.8, 71.5, 71.2, 71.0, 70.9,
70.7, 70.6, 70.4, 69.1, 68.9, 68.7, 67.7, 65.7, 65.4, 36.5, 35.4,
26.2, 26.1, 26.0, 25.0, 24.9, 24.6, 24.4; IR (film, cm-1) ν 3061,
1
) 12:1) as a pale yellow oil. H NMR (CDCl3, 500 MHz) δ
7.94 (d, J ) 7.0 Hz, 2H), 7.90 (d, J ) 7.0 Hz, 2H), 7.79 (d, J
) 8.0 Hz, 2H), 7.48-7.12 (m, 29H), 5.85 (t, J ) 9.5 Hz, 1H),
5.48 (t, J ) 8.5 Hz, 1H), 5.41 (t, J ) 9.5 Hz, 1H), 5.36 (d, J )
4.5 Hz, 1H), 4.98 (d, J ) 11.0 Hz, 1H), 4.91 (d, J ) 8.5 Hz,
1H), 4.88 (d, J ) 8.5 Hz, 1H), 4.77 (d, J ) 11.0 Hz, 1H), 4.73
(d, J ) 10.5 Hz, 1H), 4.68 (d, J ) 11.0 Hz, 1H), 4.53 (d, J )
12.5 Hz, 1H), 4.47 (d, J ) 7.0 Hz, 1H), 4.41 (d, J ) 12.0 Hz,
1H), 4.37 (d, J ) 8.0 Hz, 1H), 4.16-4.04 (m, 3H), 3.97-3.91
(m, 2H), 3.87-3.72 (m, 4H), 3.62-3.51 (m, 5H), 3.39 (t, J )
7.5 Hz, 2 H), 1.30 (s, 3H), 1.15 (s, 3H), 1.14 (s, 3H), 1.12 (s,
3H); 13C NMR (CDCl3, 125 MHz) δ 165.7, 165.4, 165.2, 138.6,
138.5, 138.1, 133.4, 133.3, 133.1, 133.0, 130.1, 129.8, 129.7,
1656 J. Org. Chem. Vol. 74, No. 4, 2009

StereoselectiVe Synthesis of ꢀ-Glycosides
3031, 2988, 2939, 2900, 2863, 1603, 1496, 1454, 1381, 1255,
1209, 1068; HRMS (ESI) calcd for C39H48O10Na (M + Na)
699.31397, found 699.31658.
Supporting Information Available: Experimental proce-
dures and 1H NMR and 13C NMR spectra of all new compounds.
This material is available free of charge via the Internet at
http://pubs.acs.org.
Acknowledgment. We thank Montana State University for
financial support. This research is supported by the National
Science Foundation (CHEM-0809200).
JO802468P
J. Org. Chem. Vol. 74, No. 4, 2009 1657