The disarming effect of the 4,6-acetal group on glycoside reactivity: Torsional or electronic?

Article abstract of DOI:10.1021/ja047578j

An evaluation of whether the well-known deactivating effect of a 4,6-acetal protection group on glycosyl transfer is caused by torsional or an electronic effect from fixation of the 6-OH in the tg conformation was made. Two conformationally locked probe m

Full text of DOI:10.1021/ja047578j

Published on Web 07/10/2004
The Disarming Effect of the 4,6-Acetal Group on Glycoside
Reactivity: Torsional or Electronic?
Henrik Helligsø Jensen, Lars Ulrik Nordstrøm, and Mikael Bols*
Contribution from the Department of Chemistry, UniVersity of Aarhus, 8000 Aarhus C, Denmark
Received April 27, 2004; E-mail: mb@chem.au.dk
Abstract: An evaluation of whether the well-known deactivating effect of a 4,6-acetal protection group on
glycosyl transfer is caused by torsional or an electronic effect from fixation of the 6-OH in the tg conformation
was made. Two conformationally locked probe molecules, 2,4-dinitrophenyl 4,8-anhydro-7-deoxy-2,3,6-
tri-O-methyl-â-D-glycero-D-gluco-octopyranoside (18R) and the L-glycero-D-gluco isomer (18S), were
prepared, and their rate of hydrolysis was compared to that of the flexible 2,4-dinitrophenyl 2,3,4,6-tetra-
O-methyl-â-D-glucopyranoside (21) and the locked 2,4-dinitrophenyl 4,6-O-methylidene-2,3-di-O-methyl-
â-D-glucopyranoside (26). The rate of hydrolysis at pH 6.5 was 21 > 18R > 18S > 26, which showed that
the deactivating effect of the 4,6-methylene group is partially torsional and partially electronic. A comparison
of the rate of acidic hydrolysis of the corresponding methyl R-glycosides likewise showed that the probe
molecules 17S and 17R hydrolyzed significantly slower than methyl tetra-O-methyl-glucoside 19, confirming
4
a deactivating effect of locking the saccharide in the C1 conformation. The experiments showed that the
hydroxymethyl rotamers deactivate the rate of glycoside hydrolysis in the order tg . gt > gg.
Introduction
The range of complicated oligosaccharides readily accessible
via chemical synthesis has within the past few years exploded
as a result of recent developments in carbohydrate synthesis
techniques. Among the most spectacular achievements are the
promising developments of solid-phase oligosaccharide syn-
thesis1 and the programmable one-pot glycosylation methods,2
the latter being an extensive advancement of the so-called
“armed-disarmed” principle.
The concept of armed and disarmed glycosyl donors refers
to the increased reactivity of benzylated over benzoylated
glycosyl donors, a phenomenon observed very early by Paulsen,3
and which originates from the greater electron-withdrawing
capability of ester blocking groups over ether blocking groups.
Fraser-Reid and co-workers were the first to name and exploit
the armed-disarmed principle, synthetically showing that donors
bearing ether protection groups (see 1, Figure 1) could be
activated chemoselectively over donors bearing ester groups4
(2, Figure 1) allowing one-pot synthesis of a trisaccharide via
the O-pentenyl method.5
Fraser-Reid also found that benzylidene protected donor 3
was less reactive than the benzylated donor 4 (Figure 1).6 A
similar order of reactivity is also found for other donors such
as mannosyl donors 5 and 6.2 Fraser-Reid suggested that the
Figure 1. Relative reactivity of various glycosyl donors. Data are from
refs 2, 4, 6, and 7.
trans fused protection group restricts the molecule from ring
flexibility, thereby making it increasingly difficult to reach a
half-chair transition state from a chair ground state.6 This
explanation has subsequently been invoked by Ley et al. to
account why donors such as 7 containing a vicinal diol
protection in the form of a CDA-type acetal was less reactive
than donor 8 (Figure 1).7,8 This modulation of reactivity by
(1) Plante, O. J.; Palmacci, E. R.; Seeberger, P. H. Science 2001, 291, 1523-
1527.
(2) Koeller, K. M.; Wong, C.-H. Chem. ReV. 2000, 100, 4465-4494.
(3) Paulsen, H. Angew. Chem., Int. Ed. Engl. 1982, 155-173.
(4) Mootoo, D. R.; Konradsson, P.; Udodong, U.; Fraser-Reid, B. J. Am. Chem.
Soc. 1988, 110, 5583-5584.
(5) Fraser-Reid, B.; Wu, Z.; Udodong, U. E.; Ottosson, H. J. Org. Chem. 1990,
55, 6068-6070.
6,9
cyclic protection groups is termed “torsional disarmament”
(6) Fraser-Reid, B.; Wu, Z.; Andrews, C. W.; Skowronski, E.; Bowen, J. P. J.
Am. Chem. Soc. 1991, 113, 1434-1435.
and has been exploited in synthesis to obtain anomeric selectiv-
9
10.1021/ja047578j CCC: $27.50 © 2004 American Chemical Society
J. AM. CHEM. SOC. 2004, 126, 9205-9213
9205

A R T I C L E S
Jensen et al.
Figure 2. Relative rates of acidic hydrolysis of methyl 4-C-methylglycopyranosides 9-12 (left, from ref 16). Charge-dipole interaction explanation for the
different electronic effects observed in amines (near right) and glycosyl transfer intermediates (far right, from ref 15).
Figure 3. Different reactivities associated with the different hydroxymethyl rotamers based on the anticipated electronic effect (left). Probe glycosides
synthesized in this work (right).
ity.10,11 While such torsional influence on glycosyl transfer is
well established, it is also clear that torsional effects in general
are quite minor as compared to the electronic effects of the
hydroxylic substituents.12
jugation has also been suggested to explain the effect; see note
14 of ref 15.)
Given these directional electronic effects observed for ring
hydroxyl groups, one realizes that the conformation of the
hydroxymethyl group of a hexopyranoside may be important
for the electronic effect from this group. Normally, the hy-
droxymethyl group will adopt three staggered conformations,
the tg, gt, and gg conformers, and the population of each will
depend on the structure of the monosaccharide (Figure 3).18,19
The gg and tg conformers have a spatial arrangement that is
comparable to having an axial and an equatorial 4-OH in a
monosaccharide in relation to O5, that is, galactose versus
glucose. Thus, using the charge dipole hypothesis, in the tg
conformer the C6-O6 bond acts as a dipole with the negative
terminus directed away from the electron-deficient center in the
transition state and it should be the least reactive, while in the
gg and gt conformers this dipole is perpendicular to the
developing positive charge and therefore more favorable.
Normally, the tg conformer is among the least abundant in a
flexible hexopyranoside.18,19 However, a 4,6-O-benzylidene
derivative is locked in this deactivating tg conformer, and a
disarming electronic effect may therefore be anticipated from
this group.
In a recent series of papers, we have investigated the
substituent effects of the carbohydrate hydroxylic substituents,13-16
and we find that the axial OH has a much reduced electron-
withdrawing power as compared to the equatorial OH. As
illustrated for the 4-position of hexopyranosides (Figure 2),
compound 9 having an axial OH undergoes acidic hydrolysis 6
times faster than the equatorial isomer 10. As reference, the
nonhydroxylated analogues 11 and 12 hydrolyze at compara-
tively similar rates.16 The exact cause of the effect remains
unclear, but it is presumably caused by differences in electro-
static effects17 and/or stereoelectronic effects. A difference in
charge-dipole interactions in the two systems has so far been
able to account for the observed effect (Figure 2). (Hypercon-
(7) Douglas, N. L.; Ley, S. V.; Lu¨cking, U.; Warriner, S. L. J. Chem. Soc.,
Perkin Trans. 1 1998, 51-66.
(8) Ley, S. V.; Baeschlin, D. K.; Dixon, D. J.; Foster, A. C.; Ince, S. J.; Priepke,
H. W. M.; Reynolds, D. J. Chem. ReV. 2001, 101, 53-80.
(9) Chong, P. Y.; Roush, W. R. Org. Lett. 2002, 4, 4523-4526.
(10) Crich, D.; Sun, S. Tetrahedron 1998, 54, 8321-8348.
(11) (a) Crich, D.; Smith, M. J. Am. Chem. Soc. 2002, 124, 8867-8869. (b)
Nukada, T.; Berces, A.; Whitfield, D. M. Carbohydr. Res. 2002, 337, 765-
774.
We became interested in the origin of this disarming effect
during the study of the enhanced reactivity of galactosides over
glucosides using probes 9 and 10.16 First, the experiments with
models 9 and 10 (Figure 2) having bulky methyl groups in the
4-position showed that steric effect had little influence on
(12) Dean, K. E. S.; Kirby, A. J.; Komarov, I. V. J. Chem. Soc., Perkin Trans.
2 2002, 337-341.
(13) Jensen, H. H.; Lyngbye, L.; Jensen, A.; Bols, M. Chem.-Eur. J. 2002, 8,
1218-1226.
(14) Bols, M.; Liang, X.; Jensen, H. H. J. Org. Chem. 2002, 67, 8970-8974.
(15) Jensen, H. H.; Lyngbye, L.; Bols, M. Angew. Chem., Int. Ed. 2001, 40,
3447-3449.
(16) Jensen, H. H.; Bols, M. Org. Lett. 2003, 5, 3201-3203.
(17) Woods, R. J.; Andrews, C. W.; Bowen, J. P. J. Am. Chem. Soc. 1992, 114,
859-864.
(18) Bock, K.; Duus, J. Ø. J. Carbohydr. Chem. 1994, 13, 513-543.
(19) Rockwell, G. D.; Grindley, T. B. J. Am. Chem. Soc. 1998, 120, 10953-
10963.
9
9206 J. AM. CHEM. SOC. VOL. 126, NO. 30, 2004

Effect of 4,6-Acetal Group on Glycoside Reactivity
A R T I C L E S
reactivity, indicating that facilitation of bond rotation had a
minor influence on reactivity in this part of the molecule.
Second, a small unexpected difference in reactivity between
probes 11 and 12 was observed (Figure 2), which could be
explained by different rotameric populations around the C5/C6
bond, as was supported by the differences in the NMR coupling
constants of 11 and 12. As the size of the deactivating effect
observed from having a 4,6-benzylidene in pentenyl glycoside
reactions was unclear, a 2-fold decrease in reaction times as
measured by TLC,6 it became questionable whether the observed
deactivation was actually caused by a torsional influence or a
larger electron-withdrawing effect from O6 and, furthermore,
desirable to know how large was actually the deactivating effect
of 4,6-benzylidene protection.
To solve this problem, the models 18R and 18S were
conceived (Figure 3). These molecules have a torsional restric-
tion similar to that of a 4,6-benzylidene group, but without
having O6 in the deactivating tg conformer. Comparing the
hydrolysis of these molecules with torsionally free analogue 21
and 4,6-methylidene derivative 26, which has both torsional
restriction and tg conformation but otherwise similar substitu-
ents, would resolve whether electronic or torsional effects are
the cause of the benzylidene disarming effect. The dinitrophenyl
glycosides allow us to determine first-order rate constants for
spontaneous hydrolysis by the method of Withers’ and co-
workers.20 We, hence, address a problem about reactions
typically conducted on protected glycosides in dichloromethane
at low temperatures using Lewis acid promotion by conducting
hydrolysis on quite unlike systems. Even though we expect
mechanisms in terms of leaving group activation, charge
buildups, timing of bond fission and fusion, and solvation to
be largely different in the two systems, there are several
examples of different carbohydrates behaving remarkably
similar. Thus, the order of reactivity in both glycosylation
reactions carried out in DCM, acidic hydrolysis of methyl
glycosides, and spontaneous hydrolysis of dinitrophenyl gly-
cosides has been found to be fucoside > galactoside >
glucoside.2,20,21 Additionally, the ratios of reactivity have been
found to be largely constant.
procedure employing PhBCl2/Et3SiH26 resulted in a mixture of
regioisomers that was inseparable by chromatography. Oxidation
of the alcohol 14 into the corresponding aldehyde was carried
out with Dess-Martin periodinane reagent.36 It was found to
be essential to carry on with the crude aldehyde in the following
allylation reaction without extensive chromatographic purifica-
tion. Allylation was carried out either by the Sakurai reaction
using allyltrimethylsilane/TiCl4 or by the Grignard reaction
giving, after methylation of the resulting secondary alcohol, the
stereoisomers 15R/S in 75% or 67% yield, respectively. Both
reactions favored the S-isomer with the Sakurai reaction being
the more selective (selectivity 93:7 versus 63:37 for Grignard),
all in accordance with what is known from related cases.27
Because in this particular case both isomers were required
and because they could be separated chromatographically, we
employed the Sakurai reaction to produce 15S and the Grignard
reaction to produce 15R. The diastereoselectivity of the Sakurai
reaction can be rationalized by considering the chelated structure
displayed in Scheme 1, which shows the si-face to be more
accessible.
The alkene 15R was now subjected to ozonolysis, reduction
with NaBH4, and mesylation to give octuloside-8-mesylate 16R
in 75% yield. Hydrogenolysis with palladium catalysis of the
4-O-benzyl group followed by cyclization with NaH/DMF and
NaI as nucleophilic catalyst smoothly gave bicyclic octuloside
17R in 86% yield. After acetolytic cleavage of the methyl
glycoside and Zemple´n deacetylation, a 1-OH sugar was
obtained that after reaction with 2,4-dinitrofluorobenzene gave
the â-dinitrophenylglycoside 18R in 53% yield.
Through a similar sequence of reactions, 15S was converted
to 18S. Ozonolysis, reduction, and mesylation gave 16S in 74%
yield. After hydrogenolysis and cyclization, 16S gave 89% yield
of 17S. This methyl glycoside was similarly by acetolysis,
deacetylation, and arylation converted to dinitrophenyl glycoside
18S in 52% yield.
The configuration of C-6 of 17R/S and 18R/S was assigned
1
on the basis of the width of the H NMR multiplet originating
from H7ax, which was easily identified in each of the spectra
by being the most upfield shifted proton. This proton couples
to four other protons, and the width of this multiplet is the sum
of the coupling constants. Because, in the S isomers, it will have
three large couplings (one geminal and two diaxial) while the
R isomers only have two large couplings (one geminal and one
diaxial coupling), there is a significant difference in width. The
width of the H7ax was for the S isomers ∑J(17S) 42.0 Hz and
∑J(18S) 42.4 Hz, and for the R-isomers ∑J(17R) 31.4 Hz and
∑J(18R) 34.4 Hz and thus was evident.
The reference compound 21 was prepared by perchloric acid-
catalyzed acetolysis of the known permethylated R-glucoside
19,28 giving 1-O-acetyl glucoside 20 as an anomeric mixture
favoring the R-isomer in 81% combined yield (Scheme 2). After
deacetylation and reaction with Sanger’s reagent, the â-dini-
trophenyl glucoside 21 was obtained in 40% yield.
In the present paper, we report the synthesis of and hydrolysis
studies with these models and find that the deactivation
associated with having a 4,6-acetal protection is indeed con-
siderable and caused by a roughly equal mixture of torsional
and electronic effects.
Results and Discussion
Synthesis. The synthesis of 18R and 18S was carried out as
outlined in Scheme 1. Known benzylidene derivative 1322 was
regioselectively opened using BH3-THF in CH2Cl2 with
scandium triflate catalysis,23 giving the 6-ol 1424 exclusively
in excellent yield. This reaction was particularly satisfactory
because the standard LAH/AlCl3 protocol25 as well as a
A somewhat more elaborate synthetic route was required for
the preparation of the methylene derivative 26. Attempting
formation of the methylene acetal via Lewis/Brønsted acid-
catalyzed reactions29 on methyl 2,3-di-O-methyl-R-D-glucopy-
(20) Namchuk, M. N.; McCarter, J. D.; Becalski, A.; Andrews, T.; Withers, S.
G. J. Am. Chem. Soc. 2000, 122, 1270-1277.
(21) Bu¨low, A.; Meyer, A.; Olszewski, T.; Bols, M. Eur. J. Org. Chem. 2004,
2, 323-329.
(22) Zeller, S. G.; D’Ambra, A. J.; Rice, M. J.; Gray, G. R. Carbohydr. Res.
1988, 182, 53-62.
(23) Wang, C.-C.; Luo, S.-Y.; Shie, C.-R.; Hung, S.-C. Org. Lett. 2002, 4, 847-
849.
(24) Guindon, Y.; Girard, Y.; Berthiaume, S. Y.; Gorys, V.; Lemieux, R.;
Yoakim, C. Can. J. Chem. 1990, 897-902.
(25) Liptak, A.; Goendoer, A. Acta Chim. Hung. 1983, 114, 309-322.
(26) Sakagami, M.; Hamana, H. Tetrahedron Lett. 2000, 41, 5547-5551.
(27) Danishefsky, S. J.; DeNinno, M. P.; Phillips, G. B.; Zelle, R. E. Tetrahedron
1986, 42, 2809-2819.
(28) West, E. S.; Holden, R. F. J. Am. Chem. Soc. 1934, 56, 930-932.
9
J. AM. CHEM. SOC. VOL. 126, NO. 30, 2004 9207

A R T I C L E S
Jensen et al.
Scheme 1. Synthesis of 18R and 18S
ranoside resulted in the undesired di-MOM-protected product.
Turning to a procedure using POCl3 in hot DMSO proved
successful.30 The relatively stable methylene acetal, however,
was found to undergo cleavage under the subsequent acid-
catalyzed acetolysis reaction. A different strategy going through
a thioglucoside was therefore investigated.
Accordingly, the known glucose pentaacetate 22 (Scheme 2)
was converted to the thioglycoside with p-thiocresol and ZnCl2
using a known procedure.31 The crude tetraacetate was then
deacetylated and benzylidene protected using dimethoxytoluene/
TsOH in DMF. Methylation of the remaining two alcohols with
MeI/NaH and hydrolysis of the benzylidene group with metha-
nolic HCl gave the diol 23 in high overall yield. Attempting
methylidene acetal formation using the POCl3/DMSO procedure
was counterproductive because the thioglycoside could not
tolerate the harsh reaction conditions. We therefore employed
Fleet’s procedure for the formation of methylidene acetals32
using methylene bromide, NaOH, and tetrabutylammonium
iodide in dioxane, which provided 24 in 34% yield. The
thioglycoside was converted to the 1-acetate by means of
N-iodosuccinimide/acetic acid,33 giving 25 in 89% yield.
Deacetylation followed by reaction with Sanger’s reagent in the
presence of DABCO then led to 26 in 34% yield.
Hydrolysis Experiments. The rate of hydrolysis was deter-
mined under two different sets of conditions. The rate of acidic
hydrolysis of methyl glycosides 17R, 17S, and 19 was deter-
mined in 2.56 M HClO4 at 82 °C by following the change in
optical rotation (Table 1). The reaction was followed to
completion, and from the plot of (R - Rend/Ro - Rend) versus
(29) Gras, J.-L.; Nouguier, R.; Mchich, M. Tetrahedron Lett. 1987, 28, 6601-
6604.
(30) Guiso, M.; Procaccio, C.; Fizzano, M. R.; Piccioni, F. Tetrahedron Lett.
1997, 38, 4291-4294.
(32) Fleet, G. W. J.; Shing, T. K. M. J. Chem. Soc., Chem. Commun. 1984, 13,
835-837.
(31) Shimidate, T.; Chiba, S.; Inouye, K.; Iino, T.; Hosoyama, Y. Bull. Chem.
Soc. Jpn. 1982, 11, 3552-3554.
(33) Veeneman, G. H.; van Leeuwen, S. H.; van Boom, J. H. Tetrahedron Lett.
1990, 31, 1331-1334.
9
9208 J. AM. CHEM. SOC. VOL. 126, NO. 30, 2004

Effect of 4,6-Acetal Group on Glycoside Reactivity
A R T I C L E S
Scheme 2. Synthesis of 21 and 26
Table 1. First-Order Rate Constants for the Acidic Hydrolysis of
Methyl Glycosides
for glycosides bearing OMe, the pH independence of the
hydrolysis was checked, by performing the hydrolysis of 18S
at pH 5.7 and 7.9 with no significant influence on the rate.
The significant difference between these two reactions is that
in the acidic hydrolysis of methyl glycosides a mechanism
involving endocyclic protonation is possible, which would
generate a rather different transition state. Recent work indicates,
however, that in acidic hydrolysis of R-glycopyranosides
endocyclic cleavage is negligible.35 This work likewise shows
no significant rate difference between the two cases.
For both reactions, the data clearly show that a torsional
disarming effect exists. Methyl glycoside 17R hydrolyzes 4
times slower than 19, while 17S hydrolyzes 6 times slower than
19 (Table 1). Similarly, the dinitrophenyl glycosides 18R and
18S hydrolyze 4 and 6 times slower than 21, respectively
(Table 2).
Also, the experiments show a clear electronic effect from
locking of the hydroxymethyl group in the tg conformation.
Thus, the least reactive dinitrophenyl glycoside is the methylene
derivative 26. While 26 is about 14 times less reactive than 21
and thus significantly disarmed as compared to 21, only about
a factor of 4 is caused by torsional disarming, and the remaining
by electronic effects. It is also noteworthy that the locked gt
isomers 17S and 18S are consistently 1.5 times less reactive
than the gg isomers. This observation cannot be explained by
the charge-dipole hypothesis as the charge-dipole interaction
between C6-O6 and a charged O5 or C1 is identical or virtually
identical for the gt and gg isomers. It is possible that hyper-
conjugation can be invoked to explain this difference, but due
to the limited data it cannot be done with confidence. It is in
any case interesting that it is the equatorial isomer that is the
least reactive of the two, which is similar to what is seen for
OH substituents in the pyranoside ring.
t (see Supporting Information) the first-order rate constant was
successfully determined by fitting the data to an exponential
decay curve. The reactions were repeated 2-3 times, and the
average rate constants are shown in Table 1.
The rates of spontaneous hydrolysis of dinitrophenyl glyco-
sides 18R, 18S, 21, and 26 were determined in a buffer at pH
6.5, 0.4 M KCl containing 25% dioxane to secure full solubility.
The progress of the reaction was followed by measuring the
change in absorbance at 400 nm (Table 2). Initial velocities
were determined at variant substrate concentration and used to
check that the reaction was first order and to calculate first-
order rate constants. On the basis of values at five different
temperatures, Arrhenius plots were constructed (see Supporting
Information), and activation energies, entropy, and rate constants
at 37 °C were calculated (Table 2). It was ensured that the
methylene acetal of 26 was stable under the conditions by
checking the product of the hydrolysis by NMR.
Cocker and Sinnott34 have previously reported that hydrolysis
of dinitrophenyl â-D-galactopyranoside is pH independent
between 1.6 and 8.4. To check whether this was also the case
(34) Cocker, D.; Sinnott, M. L. J. Chem. Soc., Perkin Trans. 2 1975, 1391-
1393.
(35) Liras, J. L.; Lynch, V. M.; Anslyn, E. V. J. Am. Chem. Soc. 1997, 119,
8191-8200.
9
J. AM. CHEM. SOC. VOL. 126, NO. 30, 2004 9209

A R T I C L E S
Jensen et al.
Table 2. First-Order Rate Constants, Energies of Activation, and Activation Entropies for the Solvolysis of Dinitrophenyl Glycosides
C4, C5, C6), 61.1, 61.0, 60.7, 59.5 (OCH3). HRMS(ES): calcd for
Experimental Section
C16H22N2O10Na, 425.1172; found, 425.1170.
1
General. Assignment of NMR spectra is based on H-1H COSY
and DEPT-135 techniques. Column chromatography was carried out
on silica gel.
p-Tolyl 2,3-Di-O-methyl-â-D-thioglucopyranoside (23). â-Glu-
cose pentaacetate (22) was added to a round-bottomed flask with
p-thiocresol and heated to 165 °C, after which ZnCl2 was added
according to the procedure described in the literature.31 After the
reaction was complete, the resulting mixture was left to cool to room
temperature, after which it was treated with ethanol. This resulted in a
precipitate that was pure enough for further reaction.
2,4-Dinitrophenyl 2,3,4,6-Tetra-O-methyl-â-D-glucopyranoside
(21). Methyl glucopyranoside 19 (156 mg, 0.62 mmol) was dissolved
in acetic anhydride (1.8 mL). To this solution was added concentrated
aqueous HClO4 (70%, 5 µL) at room temperature. After 10 min, the
reaction was quenched with saturated aqueous NaHCO3 (8 mL) and
left to stir for 2 h, after which solid NaHCO3 was added until the
solution was found to be neutral. The aqueous solution was subsequently
extracted with AcOEt (5 × 8 mL), after which the combined organic
extracts were dried (MgSO4) and concentrated. The remaining oil
underwent column chromatography (AcOEt/pentane 1:2). This resulted
in a mixture of anomers 20 in a combined yield of 141 mg (81%).
NMR (CDCl3) δC (major anomer, R): 169.3 (C(O)CH3), 89.5 (C1),
83.1, 80.9, 78.7, 72.6, 70.8 (C2, C3, C4, C5, C6), 60.8, 60.6, 59.2,
59.0 (OCH3), 21.0 (COCH3). (minor anomer, â): 169.3 (C(O)CH3),
93.9 (C1), 86.5, 82.5, 78.7, 75.2, 70.5 (C2, C3, C4, C5, C6), 60.5,
60.4, 59.2, 59.0 (OCH3), 21.1 (COCH3).
The crude acetyl thioglucoside (2.14 g, 4.7 mmol) was dissolved in
methanol (50 mL) containing a catalytic amount of NaOCH3. This was
left to stir until TLC analysis indicated no further development. A
minimum of concentrated aqueous HCl was added to the methanolic
mixture until the solution was slightly acidic. The resulting mixture
was then evaporated to dryness with toluene under reduced pressure.
The resulting off-white solid was dissolved in dry DMF (15 mL). R,R-
Dimethoxytoluene (1.77 mL, 11.8 mmol) and p-toluenesulfonic acid
(134 mg, 0.71 mmol) were added to this solution, which was heated to
65 °C. Vacuum was applied via a water aspirator to the reaction setup
with 30 min intervals to remove methanol. After 4 h, the reaction was
cooled to room temperature and quenched with an aqueous saturated
NaHCO3 solution (15 mL), after which AcOEt (20 mL) was added.
The phases were separated, and the aqueous phase was extracted with
AcOEt (2 × 20 mL), after which the combined organic phases were
washed with water (3 × 20 mL), dried (MgSO4), and concentrated
under reduced pressure. The resulting oil was dissolved in dry DMF
(12 mL), after which NaH (617 mg, 14 mmol) and methyl iodide (0.73
mL, 12 mmol) were added. This solution was stirred at room
temperature for 90 min before water (15 mL) was carefully added. After
extraction with AcOEt (3 × 15 mL), the combined organic phases were
washed with water (3 × 15 mL), dried (MgSO4), and concentrated.
The remaining oil was stirred in methanol (25 mL) to which had been
added concentrated aqueous HCl (0.3 mL). This mixture was left
overnight at room temperature after which the reaction mixture was
neutralized with solid NaHCO3 and concentrated under reduced
pressure. The resulting oil underwent column chromatography in Et2O
(Rf 0.24). This gave 1.38 g (93%) of the desired diol 23. [R]RTD -36°
(c 1, CHCl3). NMR (CDCl3) δH: 7.41 (d, 2H, Jortho 8.0 Hz, ArH), 7.12
(d, 2H, ArH), 4.53 (d, 1H, J1,2 10.0 Hz, H1), 3.89 (dd, 1H, J5,6a, 3.4
Hz, J6a,6b 11.8 Hz, H6a), 3.75 (dd, 1H, J5,6b 5.4 Hz, H6b), 3.67 (s, 3H,
OCH3), 3.63 (s, 3H, OCH3), 3.47 (t, 1H, J 9.4 Hz, H4), 3.33 (ddd, 1H,
H5), 3.17 (t, 1H, H3), 3.05 (t, 1H, H2), 2.33 (s, 3H, ArCH3). δC: 138.1,
132.6, 129.9, 129.5 (Ar), 88.0, 87.3 (C1, C5), 82.7 (C3), 79.1 (C2),
A mixture of acetyl glucosides 20 (428 mg, 1.54 mmol) was
dissolved in methanol to which had been added a small lump of metallic
sodium. This solution was stirred at room temperature until TLC
analysis (silica, AcOEt) indicated reaction completion. A lump of dry
ice was then added to the solution, after which the solvent was removed
under reduced pressure. Toluene was added to the flask and subse-
quently removed by evaporation under reduced pressure. The resulting
substance was dissolved in dry DMF (12 mL), after which DABCO
(587 mg, 5.2 mmol) and 2,4-dinitrofluorobenzene (231 µL, 1.9 mmol)
were added. This solution was stirred for 5 h at room temperature before
the mixture was added to a flask containing AcOEt and H2O (20 mL
of each). The aqueous phase was extracted with AcOEt (2 × 15 mL),
and the combined organic layers were dried (MgSO4). After removal
of the organic solvent, the resulting oil underwent column chromatog-
raphy (eluent: AcOEt/pentane 3:7, Rf 0.19), which gave 247 mg (40%)
of the desired product (21) as a colorless solid. Recrystallization proved
possible in Et2O/pentane. [R]RT -53° (c 1, CHCl3). NMR (CDCl3)
D
δH: 8.73 (d, 1H, Jmeta 2.8 Hz, ArH), 8.41 (dd, 1H, Jortho 9.2 Hz, ArH),
7.35 (d, 1H, ArH), 5.04 (d, 1H, J1,2 7.8 Hz, H1), 3.67 (s, 3H, OCH3),
3.64 (s, 3H, OCH3), 3.63-3.66 (m, 1H, H6a), 3.56 (s, 3H, OCH3),
3.48-3.58 (m, 2H, H6b, H5), 3.38 (t, 1H, H2), 3.37 (s, 3H, OCH3),
3.28 (t, 1H, H3), 3.23 (t, 1H, H4). δC: 154.7, 141.4, 139.5, 128.9,
121.6, 117.3 (Ar), 100.9 (C1), 86.3, 83.1, 78.8, 75.6, 71.0 (C2, C3,
9
9210 J. AM. CHEM. SOC. VOL. 126, NO. 30, 2004

Effect of 4,6-Acetal Group on Glycoside Reactivity
A R T I C L E S
70.1 (C4), 62.6 (C6), 61.3, 60.8 (OCH3), 21.3 (ArCH3). HRMS(ES):
calcd for C15H22O5SNa, 337.1086; found, 337.1088.
mg, 0.15 mmol) were added at room temperature under an atmosphere
of nitrogen. This mixture was stirred for 6 h, after which Et3N (0.14
mL, 1 mmol) and methanol (18 mL) were added. The solvents were
removed under reduced pressure, and the remaining substance was put
on a short column of silica gel and eluted (Et2O). This gave 323 mg
(98%) of the known desired primary alcohol 14. This experimental
procedure was taken from the literature.23 The 1H NMR spectrum was
identical to that previously reported.24 NMR (CDCl3) δC: 138.2, 128.3,
127.9, 127.7 (Ar), 97.4 (C1), 83.5, 81.9, 77.3, 74.7, 70.7, 61.5 (C2,
C3, C4, C5, C6, PhCH2O), 60.9, 58.8, 55.0 (OCH3).
p-Tolyl 2,3-Di-O-methyl-4,6-methylidene-â-D-thioglucopyranoside
(24). Diol 23 (592 mg, 1.9 mmol) was dissolved in CH2Br2 (3.9 mL)
and dioxane (1.95 mL). To this solution were added 50% aqueous
NaOH (3.9 mL) and TBAI (70 mg, 0.19 mmol). After this mixture
was vigorously stirred for 6 h, Et2O (15 mL) was added. The aqueous
phase was then extracted with ether (2 × 15 mL), after which the
combined organic phases were washed with water (25 mL), dried
(MgSO4), and concentrated. The remaining oil underwent column
chromatography in Et2O/pentane 1:2 (Rf 0.58). This gave 212 mg (34%)
of the desired formaldehyde acetal 24 as a colorless solid. Mp (uncorr.)
(6-R) Methyl 6-C-Allyl-4-O-benzyl-2,3,6-tri-O-methyl-r-D-glu-
copyranoside and (6-S) Methyl 6-C-Allyl-4-O-benzyl-2,3,6-tri-O-
methyl-r-D-glucopyranoside (15R and 15S). Sakurai alkylation: To
a stirred solution of primary alcohol 14 (438 mg, 1.4 mmol) in CH2Cl2
(56 mL) was added freshly prepared Dess-Martin periodinane (893
mg, 2.1 mmol).36 This mixture was allowed to react for 45 min at room
temperature before Et2O (40 mL) and a saturated aqueous solution of
NaHCO3/Na2S2O3 (50 mL) were added. Additional stirring for 2 h
before phase separation and extraction of the aqueous layer with
CH2Cl2 (2 × 25 mL), drying (MgSO4), and evaporation resulted in an
oil that was directly carried on further.
102-104 °C. [R]RT -39° (c 1, CHCl3). NMR (CDCl3) δH: 7.40 (d,
D
2H, Jortho 8.0 Hz, ArH), 7.11 (d, 2H, ArH), 5.03 (d, 1H, Jgem 6.0 Hz,
acetalHeq), 4.58 (d, 1H, acetalHax), 4.52 (d, 1H, J1,2 9.6 Hz, H1), 4.19
(dd, 1H, J5,6eq 4.8 Hz, J6eq,6ax 10.4 Hz, H6eq), 3.62 (s, 3H, OCH3), 3.62
(s, 3H, OCH3), 3.45 (t, 1H, H6ax), 3.35 (t, 1H, H3), 3.25-3.31 (m,
1H, H5), 3.22 (t, 1H, H4), 3.05 (dd, 1H, J2,3 8.6 Hz, H2), 2.33 (s, 3H,
ArCH3). δC: 138.2, 133.1, 129.8, 129.2 (Ar), 93.6 (acetalC), 88.3 (C1),
84.8, 82.6, 80.7, 70.5 (C2, C3, C4, C5), 68.5 (C6), 61.2, 61.0 (OCH3),
21.2 (ArCH3). HRMS(ES): calcd for C16H22O5SNa, 349.1086; found,
349.1079.
The crude product from the Dess-Martin procedure was dissolved
in dry CH2Cl2 (70 mL) and cooled to -78 °C, after which TiCl4 (232
µL, 2.1 mmol) and allyltrimethylsilane (280 µL, 1.75 mmol) were
added. This mixture was kept at -78 °C for 2.5 h before the reaction
was quenched by adding a saturated aqueous solution of NaHCO3 (50
mL) and hereafter allowed to warm to room temperature. The organic
and aqueous phases were separated, and subsequent extraction with
CH2Cl2 (2 × 30 mL), drying of the combined organic phases (MgSO4),
and concentration resulted in an oil that was used for further reaction.
Acetyl 2,3-Di-O-methyl-4,6-methylidene-r/â-D-glucopyranoside
(25). To a stirred solution of thioglucoside 24 (200 mg, 0.61 mmol) in
dry dichloroethane (4 mL) were added NIS (152 mg, 0.68 mmol) and
dry glacial acetic acid (106 µL, 1.8 mmol). The mixture was heated to
70 °C for 3 days before it was quenched at room temperature with 5%
aqueous solution of Na2S2O3 (8 mL). The aqueous phase was addition-
ally extracted with CH2Cl2 (2 × 8 mL), after which the combined
organic phases were dried (MgSO4) and concentrated. The remaining
oil underwent column chromatography (Et2O/pentane: first 1:3, then
1:2). This gave a mixture of anomeric acetates (25) in a yield of 143
mg (89%). NMR (CDCl3) δC (major isomer, R): 169.4 (C(O)CH3),
93.8 (acetalC), 89.7 (C1), 81.1, 80.9, 79.5, 68.5, 64.8 (C2, C3, C4,
C5, C6), 61.0, 59.3 (OCH3), 21.0 (C(O)CH3). (minor anomer, â): 169.0
(C(O)CH3), 94.1, 93.6 (C1, acetalC), 82.7, 82.6, 80.6, 68.3, 66.8 (C2,
C3, C4, C5, C6), 60.8, 59.3 (OCH3), 21.0 (C(O)CH3). HRMS(ES):
calcd for C11H18O7Na, 285.0950; found, 285.0950.
The crude product from the Sakurai reaction was dissolved in dry
DMF (3 mL) to which was added NaH (177 mg, 4.2 mmol) and
iodomethane (174 µL, 2.8 mmol). This mixture was stirred for 45 min
at room temperature before it was carefully added to a flask containing
AcOEt/H2O (15 mL of each). The layers were then separated, and the
aqueous phase was extracted with AcOEt (3 × 10 mL). The combined
organic phases were washed with H2O (3 × 10 mL), dried (MgSO4),
and concentrated under reduced pressure. This resulted in an oil
containing two diastereoisomers which by 13C NMR was estimated to
be a 93:7 mixture, 15S being the major product. These diastereoisomers
could be separated by column chromatography (Et2O/pentane: first 1:2,
then 1:1) to give 15R and 15S, which both appeared as colorless oils,
in a combined yield of 382 mg (75% over three steps). Diastereoisomer
15R was found to be the least polar of the two products.
2,4-Dinitrophenyl 2,3-Di-O-methyl-4,6-methylidene-â-D-glucopy-
ranoside (26). A mixture of anomeric acetates 25 (143 mg, 0.55 mmol)
was dissolved in methanol (2 mL) to which had been added a catalytic
amount of metallic sodium. This solution was stirred until TLC
monitoring (Et2O/pentane 1:1) demonstrated full conversion. The
solvent was then removed under reduced pressure. The remaining
substance was dissolved in dry DMF (3.2 mL), after which DABCO
(208 mg, 1.9 mmol) and 2,4-dinitrofluorobenzene (82 µL, 0.66 mmol)
were added. This reaction mixture was stirred at room temperature for
5 h, after which it was poured into a flask containing water and AcOEt
(10 mL of each). The aqueous layer was then extracted with AcOEt (2
× 10 mL), after which the combined organic phases were washed with
water (2 × 10 mL), dried over MgSO4, and concentrated under reduced
pressure. This resulted in an oil that underwent column chromatography
(CH2Cl2/AcOEt: first 19:1, then 9:1), which gave the desired glucoside
26 (77 mg, 34%) as a solid that could be recrystallized in Et2O/hexane.
Rf (CH2Cl2/AcOEt 9:1) 0.32. [R]RTD -86° (c 1, CHCl3). NMR (CDCl3)
δH: 8.74 (d, 1H, Jmeta 2.8 Hz, ArH), 8.41 (dd, 1H, Jortho 9.4 Hz, ArH),
7.32 (d, 1H, ArH), 5.18 (d, 1H, J1-2 5.6 Hz, H1), 5.08 (d, 1H, Jgem 6.4
Hz, acetalHeq), 4.62 (d, 1H, acetalHax), 4.21 (dd, 1H, J6ax-5 9.8 Hz,
15S. [R]RT 142° (c 1, CHCl3). Rf (Et2O/pentane 1:1) 0.21. NMR
D
(CDCl3) δΗ: 7.26-7.36 (m, 5H, ArH), 5.74-5.85 (m, 1H, CH2d
CHCH2), 5.13 (d, 1H, Jtrans 16.8 Hz, CHaHbdCHCH2), 5.07 (d, 1H,
Jcis, 10.0 Hz, CHaHbdCHCH2), 4.93 (d, 1H, Jgem, 11.2 Hz, PhCHaHbO),
4.85 (d, 1H, J1,2 3.6 Hz, H1), 4.61 (d, 1H, PhCHaHbO), 3.55-3.64 (m,
4H, H3, H4, H5, H6), 3.63 (s, 3H, OCH3), 3.50 (s, 3H, OCH3), 3.40
(s, 3H, OCH3), 3.35 (s, 3H, OCH3), 3.26 (dd, 1H, J2,3 8.8 Hz, H2),
2.52-2.59 (m, 1H, CH2dCHCHaHb), 2.37-2.44 (m, 1H, CH2d
CHCHaHb). δC: 138.7 (Ar), 134.7 (CH2dCHCH2), 128.6, 127.9, 127.8
(Ar), 117.6 (CH2dCHCH2), 97.9 (C1), 84.0, 82.1, 77.5, 77.4 (C2, C3,
C4, C5), 75.0 (PhCH2O), 70.8 (C6), 61.0, 59.0, 57.9, 55.6 (OCH3),
33.8 (CH2dCHCH2). HRMS(ES): calcd for C20H30O6Na, 389.1940;
found, 389.1938.
15R. [R]RT 136° (c 1, CHCl3). Rf (Et2O/pentane 1:1) 0.39. NMR
D
J6ax-6eq 15.8 Hz, H6A), 3.65 (s, 3H, OCH3), 3.64 (s, 3H, OCH3), 3.38-
(CDCl3) δΗ: 7.18-7.27 (m, 5H, ArH), 5.67-5.77 (m, 1H, CH2d
3.50 (m, 5H, H2, H3, H4, H5, H6B). δC: 154.3, 141.7, 139.8, 128.9,
121.8, 117.2 (Ar), 101.2 (C1), 93.8 (acetalC), 83.3, 82.5, 80.1, 67.0
(C2, C3, C4, C5), 68.2 (C6), 61.5, 61.1 (OCH3). HRMS(ES): calcd
for C15H18N2O10Na, 409.0859; found, 409.0860.
CHCH2), 4.93-4.99 (m, 2H, CH2dCHCH2), 4.79 (d, 1H, Jgem
,
PhCHaHbO), 4.74 (d, 1H, J1,2 3.6 Hz, H1), 4.55 (d, 1H, PhCHaHbO),
3.78 (dd, 1H, J5,6 1.4 Hz, J4,5 10.2 Hz, H5), 3.56 (s, 3H, OCH3), 3.55
(t, 1H, J 9.0 Hz, H3), 3.44 (s, 3H, OCH3), 3.33-3.38 (m, 1H, H6),
Methyl 4-O-Benzyl-2,3-di-O-methyl-r-D-glucopyranoside (14).
Acetal 13 (314 mg, 1.0 mmol) was dissolved in dry CH2Cl2 (9 mL) to
which BH3-THF complex (1 M, 5 mL, 5.0 mmol) and Sc(OTf)3 (75
(36) Boeckman, R. K., Jr.; Shao, P.; Mullins, J. J. Org. Synth. 2000, 77, 141-
152.
9
J. AM. CHEM. SOC. VOL. 126, NO. 30, 2004 9211

A R T I C L E S
Jensen et al.
3.33 (s, 3H, OCH3), 3.31 (s, 3H, OCH3), 3.27 (dd, 1H, H4), 3.11 (dd,
1H, H2), 2.25-2.33 (m, 1H, CH2dCHCHaHb), 2.06-2.13 (m, 1H,
CH2dCHCHaHb). δC: 138.4 (Ar), 136.0 (CH2dCHCH2), 128.6, 128.2,
127.9 (Ar), 116.7 (CH2dCHCH2), 97.3 (C1), 84.4, 82.3, 80.5, 78.2
(C2, C3, C4, C5), 74.7 (PhCH2O), 69.8 (C6), 61.2, 59.1, 58.0, 55.1
(OCH3), 34.2 (CH2dCHCH2). HRMS(ES): calcd for C20H30O6Na,
389.1940; found, 389.1930.
same series of chemical manipulations and subsequent work-up
procedures as described for 15S. This gave 375 mg (75%) of mesylate
16R, which appeared as a colorless oil. [R]RT 125° (c 1, CHCl3). Rf
D
(Et2O) 0.31. NMR (CDCl3) δΗ: 7.17-7.29 (m, 5H, ArH), 4.78 (d,
1H, Jgem 11.2 Hz, PhCHaHbO), 4.72 (d, 1H, J1,2 3.6 Hz, H1), 4.52 (d,
1H, PhCHaHbO), 4.21 (dd, 2H, J7a,8 4.8 Hz, J7b,8 7.6 Hz, H8), 3.82 (d,
1H, J4,5 10.0 Hz, H5), 3.56 (s, 3H, OCH3), 3.50 (m, 2H, H3, H6), 3.43
(s, 3H, OCH3), 3.32 (s, 3H, OCH3), 3.31 (s, 3H, OCH3), 3.15 (t, 1H,
H4), 3.09 (dd, 1H, J2,3 9.6, H2), 2.83 (s, 3H, CH3SO3), 1.84-1.92 (m,
1H, H7a), 1.65-1.74 (m, 1H, H7b). δC: 138.1, 128.7, 128.3, 128.1
(Ar), 97.4 (C1), 84.3, 82.2, 77.8, 75.9, 69.0 (C2, C3, C4, C5, C6),
74.8 (PhCH2O), 67.4 (C8), 61.2, 59.1, 57.9, 55.2 (OCH3), 37.2 (CH3-
SO3), 29.3. HRMS(ES): calcd for C20H32O9SNa, 471.1665; found,
471.1682.
Grignard Alkylation. Primary alcohol 14 (796 mg, 2.6 mmol) was
reacted with Dess-Martin periodinane (1.62 g, 3.8 mmol) in CH2Cl2
(98 mL) as was previously described. The crude product from this
reaction was dissolved in dry Et2O (11 mL), after which a solution of
allylmagnesium bromide was carefully added at 0 °C. The Grignard
solution was made by adding allyl bromide (2.2 mL, 26 mmol) to Mg-
turnings (930 mg, 38 mmol) in dry Et2O (22 mL) at such a rate that
the solution was kept at a gentle reflux. After the addition was
completed, the mixture was heated to reflux for an additional 2 h before
it was used for further reaction.
Methyl 4,8-Anhydro-7-deoxy-2,3,6-tri-O-methyl-â-L-glycero-D-
gluco-octopyranoside (17S). Mesylate 16S (325 mg, 0.73 mmol) was
dissolved in methanol (6 mL) and AcOEt (0.5 mL). Palladium on
charcoal (10%, 150 mg) was added to the solution together with two
drops of concentrated aqueous hydrochloric acid before hydrogen was
applied (balloon, 1 atm). After 2 h, the solution was filtered through a
bed of Celite and evaporated to dryness with toluene. The remaining
oil was dissolved in dry DMF (3 mL) to which were added NaH (380
mg, 0.87 mmol) and NaI (54 mg, 0.36 mmol) at room temperature.
The mixture was stirred for 1 h before it was quenched with methanol
(0.5 mL) and evaporated to dryness. The remaining substance underwent
column chromatography (AcOEt/pentane: first 1:3, then 2:1), which
resulted in 170 mg (89%) of the cyclized product 17S that appeared as
After two-thirds of the freshly prepared Grignard solution had been
added to the crude aldehyde, the reaction mixture was left to stir for 3
h at 0 °C before it was quenched by carefully pouring the solution
onto aqueous HCl (1.5 M, 20 mL). Extraction of the aqueous phase
with CH2Cl2 (2 × 20 mL), drying (MgSO4), and concentration gave a
crude oil that was further reacted as previously described in DMF (6
mL) with NaH (248 mg, 5.7 mmol) and iodomethane (265 µL, 4.3
mmol). Chromatography in Et2O/pentane resulted in a combined yield
of 626 mg (67%) of two compounds, which were found to be
spectroscopically identical to those produced via the Sakurai pathway.
The ratio of diastereoisomers was estimated from 13C NMR to be a
37/63 in favor of 15S.
a colorless oil. [R]RT 115° (c 1, CH3OH). Rf (AcOEt) 0.48. NMR
D
(CDCl3) δΗ: 4.87 (d, 1H, J1,2 3.6 Hz, H1), 4.04 (ddd, 1H, J7eq,8eq 1.3
Hz, J7ax,8eq 5.3 Hz, J8eq,8ax 11.9 Hz, H8eq), 3.61 (s, 3H, OCH3), 3.52 (s,
3H, OCH3), 3.41-3.54 (m, 3H, H3, H5, H8ax), 3.47 (s, 3H, OCH3),
3.45 (s, 3H, OCH3), 3.30 (ddd, J6,7eq 5.0 Hz, J5,6 9.0 Hz, J6,7ax 10.8 Hz,
H6), 3.24 (dd, J2,3 9.4 Hz, H2), 2.98 (t, 1H, H4), 2.04-2.11 (m, 1H,
H7eq), 1.58 (ddt, 1H, H7ax). δC: 97.6 (C1), 81.8, 80.7, 80.0, 78.6,
71.8 (C2, C3, C4, C5, C6), 66.3 (C8), 61.2, 59.1, 57.5, 55.2 (OCH3),
31.2 (C7). HRMS(ES): calcd for C12H22O6Na, 285.1314; found,
285.1315.
Methyl 4-O-Benzyl-7-deoxy-2,3-di-O-methyl-8-O-methylsulfonyl-
â-L-glycero-D-gluco-octopyranoside (16S). At -78 °C, ozone was
bobbled through a solution of olefin 15S (993 mg, 2.7 mmol) in
methanol/CH2Cl2 (2:1, 250 mL) until it turned persistently blue. Pure
O2 followed by N2 was bobbled through the reaction mixture (5 min)
to which then was added NaBH4 (2.06 g, 54 mmol). This solution was
stirred and allowed to reach room temperature. After 90 min, an
additional amount of NaBH4 (0.52 g, 13.5 mmol) was added and
allowed to react for further 1 h before most of the solvent was removed
under reduced pressure. To the remaining solution were added AcOEt
(30 mL) and hydrochloric acid (1 M) until the aqueous phase was found
to be neutral. The aqueous phase was extracted with AcOEt (3 × 20
mL) before the combined organic phases were washed with brine, dried
(MgSO4), and concentrated. The remaining crude oil was dissolved in
dry pyridine (3 mL). At room temperature was then added methane-
sulfonyl chloride (262 µL, 3.4 mmol). The reaction mixture was left
to stir for 90 min before it was poured into a flask containing AcOEt/
H2O (20 mL of each). The layers were separated before the aqueous
phase was extracted with AcOEt (3 × 15 mL), and the combined
organic phases washed with diluted HCl (1 M, 3 × 15 mL), dried
(MgSO4), and removed under reduced pressure. The remaining oil was
put on a short column of silica gel and eluted (Et2O). This resulted in
901 mg (74%) of the desired mesylate, which appeared as a colorless
oil. [R]RTD 82° (c 1, CHCl3). Rf (Et2O) 0.19. NMR (CDCl3) δΗ: 7.29-
7.37 (m, 5H, ArH), 4.97 (d, 1H, Jgem 11.2 Hz, PhCHaHbO), 4.86 (d,
1H, J1,2 3.6 Hz, H1), 4.63 (d, 1H, PhCHaHbO), 4.37 (t, 2H, J 6.6 Hz,
H8), 3.76 (t, 1H, J 6.0 Hz, H5), 3.64 (s, 3H, OCH3), 3.52-3.63 (m,
3H, H3, H4, H6), 3.51 (s, 3H, OCH3), 3.41 (s, 3H, OCH3), 3.40 (s,
3H, OCH3), 3.27 (dd, 1H, J2,3 8.8 Hz, H2), 2.96 (s, 3H, CH3SO3), 2.15-
2.23 (m, 1H, H7a), 2.03-2.13 (m, 1H, H7b). δC: 138.5, 128.6, 127.9
(Ar), 97.9 (C1), 84.0, 82.1, 77.4, 74.3, 72.0 (C2, C3, C4, C5, C6),
74.8 (PhCH2O), 66.9 (C8), 61.1, 59.1, 58.8, 55.7 (OCH3), 37.6 (CH3-
SO3), 30.4 (C7). HRMS(ES): calcd for C20H32O9SNa, 471.1665; found,
471.1679.
Methyl 4,8-Anhydro-7-deoxy-2,3,6-tri-O-methyl-r-D-glycero-D-
gluco-octopyranoside (17R). Mesylate 16R (375 mg, 0.84 mmol) was
dissolved in AcOEt/methanol and underwent a series of chemical
manipulations and work-up procedures identical to that described for
17S. This gave 188 mg (86%) of the desired product 17R. [R]RTD 89°
(c 1, CHCl3). Rf (AcOEt/pentane 1:1) 0.16. NMR (CDCl3) δH: 4.86 (d,
1H, J1,2 3.2 Hz, H1), 3.41-3.75 (m, 6H, H3,H4, H5, H6, H8ax, H8eq),
3.57 (s, 3H, OCH3), 3.47 (s, 3H, OCH3), 3.40 (s, 3H, OCH3), 3.39 (s,
3H, OCH3), 3.25 (dd, 1H, J2,3 9.2 Hz, H2), 1.95 (m, 1H, H7eq), 1.68
(dddd, 1H, J7ax,7eq 14.4 Hz, J7ax,8ax 9.0 Hz, J7ax,8eq 6.0 Hz, J7ax,6 2.0 Hz,
H7ax). δC: 98.3 (C1), 81.6, 81.0, 75.1, 74.2, 69.3 (C2, C3, C4, C5,
C6), 62.2 (C8), 61.0, 59.0, 57.1, 55.4 (OCH3), 28.7 (C7). HRMS(ES):
calcd for C12H22O6Na, 285.1314; found, 285.1319.
2,4-Dinitrophenyl 4,8-Anhydro-7-deoxy-2,3,6-tri-O-methyl-r-L-
glycero-D-gluco-octopyranoside (18S). Concentrated aqueous HClO4
(70%, 20 µL) was added to acetic anhydride (7.8 mL) at room
temperature. A portion of this (2.8 mL) was poured onto methyl
glycoside 17S (255 mg, 0.97 mmol). This solution was stirred for 15
min before the reaction was quenched with a saturated aqueous NaHCO3
solution (15 mL). Stirring for 2 h at room temperature before
neutralization with solid NaHCO3, extraction with AcOEt (3 × 20 mL),
drying (MgSO4), and concentration resulted in a crude oil that was
sufficiently pure for further reaction. The mixture of crude acetates
was stirred in methanol (8 mL) to which had been added a tiny piece
of metallic sodium until TLC analysis (AcOEt) indicated consumption
of all starting material. A small lump of dry ice was then added to the
reaction flask before all solvent was carefully removed under reduced
pressure. The remaining material was dissolved in dry DMF (5.8 mL)
before 2,4-dinitrofluorobenzene (147 µL, 1.2 mmol) and DABCO (371
Methyl 4-O-Benzyl-7-deoxy-2,3,6-tri-O-methyl-8-O-methylsulfo-
nyl-r-D-glycero-D-gluco-octopyranoside (16R). Olefin 15R (405 mg,
1.1 mmol) dissolved in methanol/CH2Cl2 (2:1, 100 mL) underwent the
9
9212 J. AM. CHEM. SOC. VOL. 126, NO. 30, 2004

Effect of 4,6-Acetal Group on Glycoside Reactivity
A R T I C L E S
mg, 3.3 mmol) were added at room temperature. After being stirred
for 5 h, the reaction mixture was poured into a flask containing H2O
and AcOEt (30 mL of each). After extraction of the aqueous phase
with AcOEt (3 × 20 mL), the combined organic extracts were washed
with H2O (3 × 20 mL), dried (MgSO4), and concentrated. The
remaining oil underwent column chromatography (eluent: AcOEt/
pentane, 3:7, Rf 0.11), yielding 210 mg (52%) of the desired aryl
glycoside 18S as a colorless solid. Recrystallization proved possible
Procedure for Determining the Rate of Hydrolysis of Dinitro-
phenyl Glycopyranoside. A solution (2.4 mM) of the dinitrophenyl
glycopyranoside (21, 18R, 18S, or 26) in 1,4-dioxane was prepared,
as well as a phosphate buffer (pH 6.5, 25 mM) containing 0.4 M KCl.
These were preheated to the desired temperature, and the following
volumes were transferred into several cuvettes: buffer 1.500 mL and
glycoside solution 0.500 mL (for the reference sample was used 0.500
mL of 1,4-dioxane). The solution was mixed thoroughly on a vortex
mixer. After a 5 min delay (to ensure temperature equilibration), the
rate of hydrolysis was measured spectrophotometrically at 400 nm over
a time span of 3 h. The hydrolysis of each compound at each
temperature was carried out 5-7 times to get reliable results. Each
measurement was carried out at four different substrate concentrations
(0.6, 0.3, 0.15, and 0.075 mM).
in hexane/Et2O. [R]RT -112° (c 1, CHCl3). NMR (CDCl3) δΗ: 8.76
D
(d, 1H, Jmeta 2.8 Hz, ArH), 8.42 (dd, 1H, Jortho 9.2 Hz, ArH), 7.44 (d,
1H, ArH), 5.08 (d, 1H, J1,2 7.2 Hz, H1), 4.04-4.08 (m, 1H, H8eq),
3.65 (s, 6H, OCH3), 3.47 (dt, J7eq,8ax 1.9 Hz, J 12.4 Hz, H8ax), 3.32-
3.45 (m, 2H, H2, H6), 3.39 (s, 3H, OCH3), 3.33, 3.21, 3.12 (t, J 9.0
Hz, H3, H4, H5), 2.11-2.15 (m, 1H, H7eq), 1.53-1.64 (m, 1H, H7ax).
δC: 154.6, 141.5, 139.5, 129.0, 121.8, 117.4 (Ar), 101.2 (C1), 83.5,
83.1, 78.8, 78.3, 76.9 (C2, C3, C4, C5, C6), 66.2 (C8), 61.5, 61.3,
58.2 (OCH3), 31.2 (C7). HRMS(ES): calcd for C17H22N2O10Na,
437.1172; found, 437.1169.
The rate constant was calculated from the slope of a plot of
absorbance versus time. Arrhenius plots were produced from the rate
constants for five different temperatures, and the activation energy, Ea,
was calculated from the slope and the activation entropy, ∆Sq, was
found at the y-axis intercept.
2,4-Dinitrophenyl 4,8-Anhydro-7-deoxy-2,3,6-tri-O-methyl-â-D-
glycero-D-gluco-octopyranoside (18R). Methyl glycoside 17R (116 mg,
0.44 mmol) was stirred for 15 min in acetic anhydride containing HClO4
(1.3 mL) prepared as described for the synthesis of 18S. Work-up
procedures and further chemical manipulations were carried out as in
the synthesis of 18S. This eventually resulted in 98 mg (53%) of the
Conclusion
A series of probe molecules for the delineation of the effects
of a 4,6-acetal on glycoside reactivity were efficiently prepared
through multistep synthetic paths. Kinetic results from acid-
catalyzed glycoside hydrolysis of three methyl glycosides and
spontaneous dinitrophenyl glycoside hydrolysis confirmed the
existence of a disarming torsional effect. It was furthermore
shown that electronic effects also play a substantial role in the
disarming effect of the 4,6-acetal group. It is therefore concluded
that the deactivating effect of a 4,6-benzylidene group is not
exclusively “torsional disarmament” but also to a roughly equal
extent an electronic effect associated with locking the hy-
droxymethyl group in the tg conformation. A small electronic
difference between the gg and the gt isomers 18R and 18S was
also observed, with the latter being the less reactive glycoside.
The results show once more the very crucial influence electronic
effects have on carbohydrate reactivity.
desired aryl glycoside 18R. [R]RT -49° (c 1, CHCl3). Rf (AcOEt/
D
pentane 1:1) 0.22. NMR (CDCl3) δΗ: 8.74 (d, 1H, Jmeta 2.8 Hz, ArH),
8.41 (dd, 1H, Jortho, 9.6 Hz, ArH), 7.35 (d, 1H, ArH), 5.12 (d, 1H, J1,2
7.2 Hz, H1), 3.71-3.79 (m, 4H, H4, H6, H8ax, H8eq), 3.64 (s, 3H,
OCH3), 3.63 (s, 3H, OCH3), 3.44 (dd, 1H, J2,3 8.6 Hz, H2), 3.38 (dd,
1H, J5,6 2.8 Hz, J4,5 9.6 Hz, H5), 3.36 (s, 3H, OCH3), 3.28 (t, 1H, H3),
1.96-2.00 (m, 1H, H7eq), 1.69-1.77 (m, 1H, H7ax). δC: 154.6, 141.5,
139.8, 128.7, 121.9, 117.3 (Ar), 101.5 (C1), 83.9, 82.8, 74.5, 73.9, 73.9
(C2, C3, C4, C5, C6), 62.2 (C8), 61.3, 60.9, 59.9 (OCH3), 29.4 (C7).
HRMS(ES): calcd for C17H22N2O10Na, 437.1172; found, 437.1173.
Procedure for Determining the Rate of Glycoside Hydrolysis in
Acidic Media. A solution (26 mM) of glycoside (19, 17R, or 17S) in
perchloric acid (2.56 M by titration with NaOH(ai)) was prepared. This
was added to a cuvette preheated to the desired temperature (82 °C).
The polarimeter was reset whereupon the cuvette was positioned in
the apparatus. The optical rotation was measured as a function of time
at sodium’s D-line. The measurements went on until the rotational value
had stabilized (20-166 h). At this point, the R(inf) was determined.
The solution of hydrolyzed glycoside underwent MS(ES) analysis,
which only revealed the presence of the hemiacetal corresponding to
the glycoside starting material. The same procedure was then repeated
2-3 times for each of the three glycosides.
Acknowledgment. This paper is dedicated to the memory of
Professor Christian Pedersen. We thank the Lundbeck founda-
tion, SNF, and the European commission for financial support.
Supporting Information Available: Progress curves for the
acidic hydrolysis of methyl glycosides and Arrhenius plot for
the solvolysis of dinitrophenyl glycosides. This material is
available free of charge via the Internet at http://pubs.acs.org.
The measurements for each determination were fitted as A(t) ) (R-
(t) - R(inf))/(R(0) - R(inf)) to a curve of the type y(t) ) yo + a‚exp-
(-bt). Good fits were obtained in every case (R2 ) 0.9994-0.9999).
The rate constant was obtained as the value of the constant b in the
equation.
JA047578J
9
J. AM. CHEM. SOC. VOL. 126, NO. 30, 2004 9213