Glycosidation-anomerisation reactions of 6,1-anhydroglucopyranuronic acid and anomerisation of β-D-glucopyranosiduronic acids promoted by SnCl 4

Article abstract of DOI:10.1002/chem.200601111

The reaction of silylated nucleophiles with 6,1-anhydroglucopyranuronic acid (glucuronic acid 6,1-lactones) catalysed by tin(IV) chloride provides 1,2-trans or 1,2-cis (deoxy)glycosides in a manner dependent on the donor structure. The α-glycoside was obt

Full text of DOI:10.1002/chem.200601111

DOI: 10.1002/chem.200601111
Glycosidation–Anomerisation Reactions of 6,1-Anhydroglucopyranuronic
Acid and Anomerisation of b-d-Glucopyranosiduronic Acids Promoted by
SnCl4
Colin OꢀBrien, Monika Polµkovµ, Nigel Pitt, Manuela Tosin, and Paul V. Murphy*[a]
Abstract: The reaction of silylated nu-
cleophiles with 6,1-anhydroglucopyra-
nuronic acid (glucuronic acid 6,1-lac-
tones) catalysed by tin(IV) chloride
provides 1,2-trans or 1,2-cis (deoxy)gly-
cosides in a manner dependent on the
donor structure. The a-glycoside was
obtained for reactions of the donor
with the 2-acyl group and 2-deoxydo-
nors, whereas the 2-deoxy-2-iodo donor
gave the b-glycoside. Experimental evi-
dence shows that when 1,2-cis-glyco-
side formation occurs, the anomerisa-
tion of initially formed 1,2-trans-glyco-
sides catalysed by SnCl4 is possible.
The anomerisation of b-d-glucopyrano-
siduronic acids was found to be faster,
in some cases, than anomerisation of
related b-d-glucopyranosiduronic acid
esters and b-d-glucopyranoside deriva-
tives and the rates are dependent on
the structure of the aglycon. Moreover,
the rates of anomerisation of b-d-glu-
copyranuronic acid derivatives can be
qualitatively correlated with rates of
hydrolysis of b-d-glucopyranosiduronic
acids. Mechanistic possibilities for the
reactions are considered.
Keywords: anomerisation · carbo-
hydrates · deoxyglycoside · glyco-
sides · Lewis acids
Introduction
tion being controlled by neighbouring-group participation.
The 6,1-anhydro derivative 4 is formed during the reaction
of 1 with SnCl4.[5] Both 4 and its 2-deoxy analogue 6 have
consequently been exploited to provide 1,2-cis-glycosides 5
and 7, respectively, when they are treated with silyl ether ac-
ceptors in the presence of SnCl4.[8] In contrast, the related 2-
deoxy-2-iodo donor 8 was found to give predominantly 1,2-
trans-glycosides 9 under similar conditions. These results are
summarised in Scheme 1. A mechanistic pathway accounting
for the results of these reactions proposed by us is shown in
Scheme 2 (pathway a). We suggested that the inversion of
configuration at C-1 of 12 occurs for donors in which X=H
or OAc without participation of the C-2 group when X=
OAc, anchimeric assistance by the carboxylate group[9] con-
tributing to the stereochemical outcome. The results ob-
served for 8 were explained by the iodine residue being a
better participator than the 2-acyl group (and the carboxyl-
ate group) and its glycosidation proceeding via 13 or 14.
Other conceivable pathways for the formation of 1,2-cis-gly-
cosides 16 from glucuronic acids involve the initial forma-
tion of b-glycoside intermediates 15 via 13 or 14 that subse-
quently rapidly anomerise, the anomerisation reaction oc-
curring via the reversible formation of 12 or by an inde-
pendent pathway. We initially discounted pathways involv-
ing b-glycosides[7] on the basis of the low degree of
anomerisation (<5% after 48 h) of the glycosyl azide 11e
when subjected to SnCl4. However, herein we provide the
Stereoselective glycoside synthesis is important because of
the biological and medical relevance of oligosaccharides,
glycoproteins, glycolipids,[1] and other carbohydrate deriva-
tives[2] and a range of strategies have been developed.[3,4]
The glycosyl azide 2 with a-configuration was found to be
the major product obtained from the SnCl4-promoted reac-
tion of the acid 1 with azidotrimethylsilane (Scheme 1).[5]
We did not expect this result for a number of reasons. First-
ly, this reaction, as described by Toth and co-workers[6] and
more recently by Lindhorst and co-workers,[7] gave the b-
azide 3. Secondly, the reaction of the glucuronic acid ester
derivative 10 under identical conditions to that described for
reactions of 1 gives the b-azide 11e, consistent with the fact
that 1,2-trans-glycosides are usually obtained from glycosyl
donors with 2-acyl protecting groups, glycoside bond forma-
[a] C. OꢀBrien, Dr. M. Polµkovµ, Dr. N. Pitt, Dr. M. Tosin,
Prof. P. V. Murphy
Centre for Synthesis and Chemical Biology
UCD School of Chemistry and Chemical Biology
Conway Institute of Biomolecular and Biomedical Research
University College Dublin, Dublin 4 (Ireland)
Fax : (+353)1-716-2501
E-mail: paul.v.murphy@ucd.ie
Supporting information for this article is available on the WWW
under http://www.chemistry.org or from the author.
902
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 902 – 909

FULL PAPER
gated. Reactions were either
carried out in NMR tubes, and
were directly monitored by
1H NMR
spectroscopy
(300 MHz), or were carried out
in flasks, with samples taken
periodically from the reaction
mixture so that product ratios
could be determined as a func-
tion of time. For reactions that
were carried out in flasks, ali-
quots were taken and worked
up by extraction of the carbox-
ylic acid product from organic
solvent into aqueous sodium
hydrogen carbonate. The water
was removed by freeze drying
and the resulting residue
(sodium salts of the a- and/or
b-glycosides) were analysed by
1H NMR spectroscopy in D2O.
For reactions in which the a-
G
azide 2 was obtained as the
major product, it was clear that
its formation resulted from the
anomerisation of an initially
formed b-glycoside. Thus, in
one experiment ([1]=0.05m in
CDCl3) after 45 min a 58:42
ratio of anomers (a/b) was
present, which changed to a
Scheme 1.
91:9 mixture favouring the a-
G
anomer after 180 min; the
ratios were determined by in-
tegration of the signals as-
signed to the pyranoside H-5
of the respective anomers after
1H NMR analysis [300 MHz,
CDCl3,
a-anomer:
d=
Scheme 2.
4.84 ppm (d, J=9.5 Hz, 1H);
b-anomer: d=4.28 ppm (d, J=
results of a more extensive study that demonstrates that the
anomerisation of b-glucopyranosiduronic acid derivatives
can lead to the formation of 1,2-cis-glycosides from 1 or 4.
The carboxyl group has a catalytic role, enhancing the rate
of anomerisation, which is also dependent on the nature of
the aglycon. Anomerisation can be correlated qualitatively
with the rates of hydrolysis of b-d-glucopyranosiduronic
acid derivatives and, therefore, we believe it likely that both
the anomerisation and the hydrolysis of b-glucopyranosidur-
onic acids could have mechanistically related pathways.
8.6 Hz, 1H)]. Reactions of 4 (0.12m in CD2Cl2) in the pres-
ence of SnCl4 (0.5 equiv) and TMSN3 (2.5 equiv) were car-
ried out in NMR tubes and H NMR analysis showed that
1
the lactone 4 was completely consumed after 30 min; the
mixture contained a major component (70%), which was a
4
b-glycoside with C1 conformation as evidenced by a signal
at d=4.00 ppm (d, J=9.0 Hz, 1H; H-5), and a minor com-
4
ponent (30%), which was an a-glycoside with C1 conforma-
tion [signal at d=4.26 (d, J=9.0 Hz, 1H; H-5)]. This reac-
tion was allowed to proceed and after 24 h the mixture con-
tained a 70:30 mixture of a- and b-glycosides; the a/b ratio
as a function of time for the reaction of 4 is shown in
Figure 1. The results of reactions carried out in NMR tubes
were confirmed by carrying out repeat experiments in flasks,
followed by normal reaction workup giving mixtures of
anomers 2 and 3 in >80% yield; the relative proportion of
Results and Discussion
Glycosidation reactions of 1 and 4 revisited: A series of re-
actions of both 1 and 4, catalysed by SnCl4, were re-investi-
Chem. Eur. J. 2007, 13, 902 – 909
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
903

P. V. Murphy et al.
analysis of residues obtained after workup of samples taken
during the reaction. When the reaction of 4 with PhOTMS
(2.5 equiv) catalysed by SnCl4 (0.5 equiv) was monitored by
1H NMR spectroscopy in CD2Cl2 at room temperature, the
b-glycoside was observed, but only at levels <10%. After
24 h a mixture containing the a-anomer 5d (52%), the un-
reacted donor 4 (45%) and the b-glycoside 20c (3%) was
obtained (a/b=95:5). When this reaction was carried out at
08C, mixtures that contained larger amounts of the b-
anomer were obtained. For example, after 6 h at 08C a
19:81 mixture of the a- and b-anomers was isolated; when
this reaction was allowed to proceed for 24 h at 08C a 45:55
mixture of the anomers, respectively, was obtained in 41%
isolated yield, with unreacted 4 accounting for the yield of
product. This series of experiments shows that anomerisa-
tion of an initially formed O- or N-glucopyranosiduronic
acid derivatives to the thermodynamically preferred a-glyco-
side is the major pathway accounting for 1,2-cis-glycoside
bond formation from either 1 or 4.
Figure 1. Relative ratio of a- and b-glycosides 2 (2/3) and 3 (3/2) in the
glycosidation reaction of 4 (0.12m), SnCl4 (0.5 equiv), TMSN3 (2.5 equiv)
in CD2Cl2 at RT as a function of time. The lactone 4 was completely con-
sumed after 30 min. The relative proportions of anomers was determined
1
by H NMR spectroscopy.
2 and 3 was dependent on the reaction time and consistent
with the time course diagram that was obtained for reac-
tions carried out in NMR tubes. The anomerisation reaction
was suppressed at lower temperature. When the reaction
was carried out at 08C, the complete consumption of 4 had
occurred after 35 min, resulting in a 13:87 mixture of a/b-
glycosides; this ratio did not alter significantly after 2 h (a/
b=17:83) at 08C. After allowing this reaction mixture to
attain room temperature and leaving for a further 22 h a
74:26 mixture of a- and b-anomers, favouring the a-anomer
was obtained.
The reactions of 4 in CDCl3 gave similar ratios of a- and
b-azides as those carried out in CD2Cl2, anomerisation again
being observed. The reaction of 4 with TMSN3 was fastest in
nitromethane, attaining equilibrium after 15 min and giving
an 80:20 mixture of a- and b-azides. The reaction of 4 with
the TMSO–threonine derivative leading to 5b was moni-
tored in a similar manner and the outcome is shown in
Figure 2. The rate of consumption of the donor 4 was slower
than in the presence of TMSN3 and only the a-glycoside
could be observed in this case, indicating that anomerisation
of the b- to the a-glycoside is too rapid to facilitate direct
observation of the b-anomer as an intermediate. The reac-
tion of 4 with CyOTMS catalysed by SnCl4 was carried out
at 08C for 22 h; evidence for the presence of the a-anomer
5a, but not for the b-anomer, was obtained by 1H NMR
Synthesis of substrates for anomerisation reactions: The gly-
cosidation reactions of 1 and 4 involved the use of silylated
nucleophiles, which could lead to the generation of silylated
intermediates that could be important to the catalytic pro-
cesses taking place. Thus a series of b-glycosides derived
from glucuronic acid and glucose were prepared so that
their potential to undergo anomerisation in the presence of
SnCl4 and in the absence of silylated nucleophiles could be
investigated. Such studies would also indicate whether the
anomerisation reactions of glucuronic acids and related
compounds would have potential in preparative chemistry.
In addition to 3,[10] 11d[11] and 11e,[12] the glucopyranosides
17a,[13] 17b[14] and 17c,[15] and the esters of glucopyranosidur-
onic acid 18b[16] and 19a[10] were obtained by literature pro-
cedures. The allyl esters 19b and 19c were synthesised from
the trichloroacetimidate donor 22 (Scheme 3). Thus 21[10]
was converted to 22 (36% over two steps) and subsequent
glycosidation using TMSOTf as promoter in dichlorome-
thane in the presence of cyclohexanol gave 19b (32%); the
reaction of 22 with phenol gave 19c (51%). The palladi-
um(0)-catalysed hydrolysis of the allyl esters of 19b and 19c
gave 20b and 20c, respectively. The disaccharide 20d was
Figure 2. Relative ratio of 4 (4/5b) and a-glycoside 5b (5b/4) in the gly-
cosidation reaction of 4 catalysed by SnCl4 (1.0 equiv) with threonine de-
rivative (2.0 equiv, CD2Cl2, RT) as a function of time. The relative pro-
portions of 4/5b were determined by 1H NMR spectroscopy. The b-
anomer could not be detected.
904
www.chemeurj.org
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 902 – 909

Glycoside Synthesis
FULL PAPER
anomerisation of the acid 3 proceeded less efficiently in
CDCl3; after 48 h there was an 82:16 mixture favouring the
b-anomer 3, the formation of the lactone 4 being also de-
tected (<3%); this changed to a 54:42:4 mixture of 3/2/4
after 12 days. The b-azides 17a–19a did not anomerise to
any detectable degree over a number of days.
The anomerisation reaction of the phenyl glucopyranosi-
duronic acid derivative 20c in both CD2Cl2 and CDCl3 at
room temperature proceeded more slowly than the azido
derivative, and much more slowly than in glycosidation reac-
tions of 4 with TMSOPh. After seven days in CDCl3 a mix-
ture favouring the a-glycoside 5d over 20c (48:38) resulted,
the formation of the lactone 4 being observed (<14% after
seven days) during the course of the reaction. After ten
days a 54:35:15 mixture of 5d/20c/4 was obtained, which did
not alter on leaving for a longer time. Phenyl glycosides
17c–19c did not undergo any anomerisation under identical
conditions to the reaction of 20c. The anomerisation of the
cyclohexyl glucuronide 20b was the fastest of those studied.
In less than 10 min in CDCl3 the complete anomerisation
occurred of 20b to give a >95:5 mixture of the a-anomer
5a and b-anomer 20b. The half-time (t1/2) for this reaction
as determined by NMR spectroscopy was <5 min. The half
time for the corresponding reaction in CH2Cl2, determined
by monitoring the change in specific rotation that occurs on
the conversion of b-anomer to the a-anomer, was approxi-
mately 4 min. The t1/2 is the time required for the reaction to
have proceeded to 50% of the final equilibrium. The anom-
erisation of the cyclohexyl glucopyranoside 17b gave, after
24 h, a 92:8 mixture of the a-glycosides 25 and 17b (quanti-
tative yield); the t1/2 for this anomerisation reaction was
>300 min. Anomerisation of the methyl ester 18b gave a
93:7 mixture of a-glycosides 26 and 18b after 190 min
(quantitative yield); the t1/2 for the anomerisation of 18b
was <20 min. Thus the anomerisation of acid 20b is more
than four times faster than that of its methyl ester, and at
least sixty times faster than that of the per-O-acetylated glu-
coside, based on the estimated t1/2 values. The anomerisation
of the disaccharide 20d did not proceed using SnCl4 in di-
chloromethane or in nitromethane. Partial formation of the
a-anomer (36%) was observed by reaction of 20d with
TiCl4 in nitromethane after 24 h. The products from anomer-
isation of the disaccharide were characterised as the methyl
esters 27 and 28 (Scheme 5).
Scheme 3.
Scheme 4.
prepared from the b-d-glucopyranoside 23 (Scheme 4).[17]
The oxidation of 23 using TEMPO[12a] followed by allylation
and subsequent acetylation gave 24. The palladium(0)-cata-
lysed removal of the allyl ester protecting group gave
20d.[18]
Anomerisation of b-d-glucopyranosides and b-d-glucopyra-
nosiduronic acid derivatives: The reactions of a series of b-
d-glucopyranosides and b-d-glucopyranosiduronic acid de-
rivatives (0.08–0.1m), catalysed by SnCl4 (0.5 equiv), were
carried out and monitored as described above for the gylco-
sidation reactions of 1 and 4 in the absence of silylated nu-
cleophiles. The relative proportions of 2 and 3 as a function
of time from the reaction of b-azide 3 (0.09m) is shown in
Figure 3; anomerisation was slower than in the glycosida-
tion–anomerisation reaction of 4 in CD2Cl2 with TMSN3,
but gave a similar ratio of 2 and 3 (71:29) after 48 h. The
Mechanistic discussion: Anomerisation of initially formed b-
glycosides is the major pathway accounting for the forma-
tion of a-glycosides from reactions of 4. Although the condi-
tions for the anomerisation of the b-d-glucopyranuronic acid
derivatives do not mimic those of the glycosidation–anomer-
isation reaction, there is sufficient evidence that related
mechanisms operate. Glycosidation–anomerisation reactions
catalysed by SnCl4 have been described[19–21] and the combi-
nation of SnCl4 and a carboxylic acid can provide powerful
inter- or intramolecular catalysis for anomerisation,[22] the
latter which was rationalised on the basis that the interac-
tion of a carboxylic acid and SnCl4 provides super acid con-
Figure 3. Relative ratio 2 (2/3) and 3 (3/2) present in the anomerisation
reaction of 3 (0.09m) catalysed by SnCl4 (0.5 equiv) in CD2Cl2 at RT, as a
function of time. The percentage of anomers was determined by
1H NMR spectroscopy.
Chem. Eur. J. 2007, 13, 902 – 909
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
905

P. V. Murphy et al.
as the anomerisation of 20b is faster than of 17b. The rates
of hydrolysis of b-d-glucopyranosiduronic acids also de-
pends on the electron affinity of the aglycon,[23] the hydroly-
sis of the cyclohexyl b-d-glucopyranosiduronic acid being
approximately 30 times faster than that of the corresponding
phenyl glycoside, for example. These rates also correlate
with anomerisations, which were faster for cyclohexyl glyco-
side 20b than for the phenyl glycoside 20c. Timell and co-
workers suggest that glucopyranosiduronic acids[23] are hy-
drolysed by a different mechanism to glucopyranosides[24]
and it has been proposed that intramolecular general acid
catalysis[25] through C1 O1 bond cleavage (exocyclic cleav-
À
age) accounts for the rate enhancement observed for some
glucuronic acid derivatives. This could involve an intermedi-
2
ate that adopts a SO conformation (29), which would give
an oxycarbenium ion (30) with a 2,5B conformation[26]
(Scheme 6, pathway a), or involve an intermediate that
1
adopts a C4 conformation (31) giving an oxycarbenium ion
with a half chair conformation (32; Scheme 6, pathway b).
In either case the carboxylic acid group is sufficiently proxi-
mate to the aglycon oxygen atom to facilitate intramolecular
proton transfer. Related intramolecular general acid cataly-
sis processes could be used to explain anomerisation
(Scheme 6, pathways a and b). Alternatively, the cleavage of
À
C1 O5 bond (endocyclic bond cleavage, Scheme 6, path-
way c)[27] leading to 35 cannot be discounted when discussing
the acid-catalysed hydrolysis and/or the SnCl4-catalysed
anomerisation of glucopyranosiduronic acids; this process
can also explain those instances where rates are faster for
glucopyranosiduronic acids than glucopyranosides. An at-
tractive feature of proposing endocyclic bond cleavage is
Scheme 5.
ditions, leading to enhanced protonation of the pyranose
oxygen and subsequent cleavage of the C1 O5 bond (endo-
À
cyclic cleavage) giving an open chair intermediate (related
to 35 in Scheme 6). The rates of anomerisation described
herein have the order glucopyranosiduronic acid>ester of
glucopyranosiduronic acid>glucopyranoside, again demon-
strating the role that a carboxylic acid group (and even a
carboxylate ester) has in enhancing the catalytic efficiency
of SnCl4. There are cases in which the rates of acid-catalysed
hydrolysis of some b-d-glucopyranosiduronic acid deriva-
tives are faster than corresponding b-d-glucopyranosides,[23]
the hydrolysis of cyclohexyl b-d-glucopyranosiduronic acid
being about five times faster than cyclohexyl b-d-glucopyra-
noside at 608C, for example. This trend is apparent herein,
4
that general acid catalysis occurs from the ground state C1
conformation.
Efforts to trap intermediates such as 30, 32 or 35 by addi-
tion of nucleophiles during the course of the anomerisation
reactions, in order to distinguish between the possible path-
ways, were not successful. The anomerisation of 17c in pres-
ence of SnCl4 have been described; a trans-glycosidation re-
action leading to anomerisation was found to occur during
the reaction of 17c in presence of o-cresol in benzene, sug-
gesting that anomerisation involves exocyclic bond cleavage
for this glucoside.[28] In our hands the anomerisation of 17c
Scheme 6.
906
www.chemeurj.org
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 902 – 909

Glycoside Synthesis
FULL PAPER
did not occur in the presence of SnCl4 in dichloromethane
or chloroform and attempts to carry out trans-glycosidation
reactions of 17c (in the presence of TMSOCy as the nucleo-
phile) were not successful. Efforts to carry out trans-glycosi-
dation of the rapidly anomerising 20b in the presence of
excess TMSN3 provided only the a-anomer 5a after 5 min
and <5% of the a-azide 2 after one week, consistent with
endocyclic bond cleavage in this case.
That the anomerisation of the glucuronic acid ester deriv-
ative 18b is faster than the glucopyranoside derivative 17b
can be explained by the SnCl4 coordinating to both the C-5
carbonyl group and the pyranose oxygen atom, leading to a
Experimental Section
General experimental conditions: Optical rotations were determined
with a Perkin–Elmer 241 or 343 model polarimeter at the sodium D line
at 238C. The NMR spectra were recorded with Varian Inova 300 and
500 MHzspectrometers. Chemical shifts are reported relative to internal
Me4Si in CDCl3 (d=0.0 ppm) or HOD for D2O (d=4.63 ppm) or
CD2HOD (d=3.36 ppm) for 1H and CDCl3 (d=77.0 ppm) or CD3OD
(d=47.7 ppm) for 13C. 1H NMR signals were assigned with the aid of
COSY. 13C NMR signals were assigned with the aid of DEPT. Coupling
constants are reported in Hz. The IR spectra were recorded with a Matt-
son Galaxy Series FTIR 3000 using either thin film between NaCl plates
or KBr discs, as specified. Melting points were measured on a Gallen-
kamp melting-point apparatus. Elemental analysis was performed on an
Exeter Analytical CE440 elemental analyser. Low- and high-resolution
mass spectra were measured at the University of York (UK) or with a
Micromass LCT KC420 or Micromass Quattro instrument. Thin-layer
chromatography (TLC) was performed on aluminium sheets precoated
with silica gel 60 (HF254, E. Merck) and spots visualised by UV and char-
ring with H2SO4-EtOH (1:20). Flash column chromatography was carried
out with silica gel 60 (0.040–0.630 mm, E. Merck). Chromatography sol-
vents used were EtOAc (Riedel-deHaen), petroleum ether (b.p. 40–608C,
BDH laboratory supplies), cyclohexane and MeOH (Sigma Aldrich). Tol-
uene (Sigma–Aldrich), acetonitrile (Sigma–Aldrich) and CH2Cl2 (Riedel-
deHaen) were freshly distilled from calcium hydride. THF was freshly
distilled from Na/Benzophenone. MeOH was distilled from Mg. Nitrome-
thane was used as obtained from Sigma–Aldrich.
2,3,4-Tri-O-acetyl-1-(2,2,2-trichloroethanimidate)-a-d-glucopyranuronic
acid, 2-propenyl ester (22): Compound 21 (5 g, 12.4 mmol) was dissolved
in THF (35 mL), benzylamine (2.71 mL, 24.8 mmol) was added and the
mixture was stirred overnight. The excess solvent was removed and the
residue was purified by chromatography (EtOAc/cyclohexane, 1:3) to
give the intermediate hemiacetal as an oil (2.53 g, 57%). Rf =0.30
five-membered ring chelate 36 that would facilitate endocy-
clic bond cleavage. Such a chelated intermediate 37 could
form from 13 or 14, accounting for anomerisation in reac-
tions of lactone 4. The reduced rate of anomerisation for the
glycosidation reactions with the 2-deoxy-2-iodo derivative 8
can be explained by participation of the iodo group leading
1
(EtOAc/cyclohexane, 1:1); H NMR (300 MHz, CDCl3): d=5.83 (tdd, J=
6.0, 10.3, 16.7 Hz, 1H; CH=CH2), 5.51 (d, J1,2 =4.4 Hz, 1H; H-1), 5.51 (t,
À
to 38 which slows down the required C1 C2 bond rotation,
J
2,3 =10.0 Hz, J3,4 =9.8 Hz, 1H; H-3), 5.30 (dd, J=13.7, 20.5 Hz, 2H;
the implication being that anchimeric assistance from the 2-
iodo group is more effective than from an 2-acetoxy group
or the C5-carboxyl group. The formation of a-glycosides
from 2-deoxydonor 6 could proceed by inversion of configu-
ration or by means of the anomerisation of initially formed
b-glycosides.
CH=CH2), 5.13 (t, J3,4 =9.7 Hz, J4,5 =9.9 Hz, 1H; H-4), 4.85 (dd, J1,2
=
3.5 Hz, J2,3 =10.1 Hz, 1H; H-2), 4.57 (m, 3H; OCH2CH=CH2 and H-5
overlapping), 2.04, 1.99, 1.97 ppm (3”s, 3”3H; 3”COCH3); 13C NMR
(75 MHz, CDCl3): d=170.2, 170.1, 169.7 (3”C, 3”COCH3), 167.9 (C,
COOAll), 130.9 (CH, CH=CH2), 119.7 (CH2, CH=CH2), 90.1 (CH, C-1),
70.7, 69.4, 69.1, 67.8 (4”CH, C-2–5), 66.7 (CH2, OCH2CH=CH2), 20.6,
20.6 ppm (3”CH3 overlapping, 3”COCH3); ES-HRMS: m/z calcd:
359.0978; found: 359.0959 [MÀH]À. This intermediate (1 g, 2.78 mmol)
was dissolved in dry dichloromethane (20 mL) under a N2 atmosphere,
trichloroacetonitrile (1.4 mL, 13.9 mmol) was added and the mixture was
Conclusion
stirred at 08C (15 min). 1,8-Diazabicyclo[5.4.0]undec-7-ene (DBU) was
A
then added (~0.2 mL) and stirring continued (4 h). The excess solvent
was removed and the residue was purified by flash chromatography
(EtOAc/cyclohexane, 1:2) to give compound 22 as an oil (874 mg, 63%).
Rf =0.56 (EtOAc/cyclohexane, 1:1); H NMR (300 MHz, CDCl3): d=8.77
(s, 1H; NH), 6.65 (d, J1,2 =3.7 Hz, 1H; H-1), 5.88 (tdd, J=6.0, 10.5,
16.4 Hz, 1H; CH=CH2), 5.63 (t, J2,3 =9.8 Hz, J3,4 =9.9 Hz, 1H; H-3), 5.30
The reaction of silylated acceptors with 2,3,4-tri-O-acetyl-
6,1-anhydroglucopyranosiduronic acids catalysed by tin(IV)
chloride provides 1,2-cis-glycosides; this result is accounted
for by the anomerisation of initially formed 1,2-trans-glyco-
sides. A rational study of the SnCl4-catalysed anomerisation
of b-glycoside derivatives shows that the presence of the C-5
carboxylic acid group enhances the efficiency of anomerisa-
tion. Moreover, the rates of anomerisation of b-d-glucopyra-
nuronic acid derivatives can be qualitatively correlated with
rates of hydrolysis of b-d-glucopyranosiduronic acids. Thus
mechanisms for the anomerisation of b-glycoside intermedi-
ates generated in the reaction of compound 4, like those of
anomerisation of b-d-glucopyranuronic acids and the hy-
drolysis of b-d-glucopyranosiduronic acids, can be consid-
ered in the context of general (Lewis) acid catalysis. The
synthetic potential of the anomerisation reaction is being
further investigated.
1
(m, 3H; CH=CH2 and H-4 overlapping), 5.15 (dd, J1,2 =3.7 Hz, J2,3
9.8 Hz, 1H; H-2), 4.61 (dd, J=4.4, 5.3 Hz, 2H; OCH2CH=CH2), 4.52 (d,
4,5 =10.3 Hz, 1H; H-5), 2.05, 2.03, 2.02 ppm (3”s, 3”3H; 3”COCH3);
=
J
13C NMR (75 MHz, CDCl3): d=169.7, 169.6, 169.3 (3”C, 3”COCH3),
166.4 (C, COOAll), 160.4 (CH3, C=NH), 130.9 (CH, CH=CH2), 119.6
(CH2, CH=CH2), 92.5 (CH, C-1), 90.4 (C, CCl3), 70.4, 69.3, 69.0, 68.9 (4”
CH, C2–5), 66.8 (CH2, OCH2CH=CH2), 20.5, 20.5, 20.3 ppm (3”CH3, 3”
COCH3); ES-HRMS: m/z calcd: 526.0050; found: 526.0072 [M+Na]+.
Cyclohexyl 2,3,4-tri-O-acetyl-b-d-glucopyranosiduronic acid, 2-propenyl
ester (19b): Anhydrous dichloromethane (30 mL) was added to the tri-
chloroacetimidate 22 (1.0 g, 1.99 mmol) and cyclohexanol (253 mL,
2.39 mmol) and 4 Š molecular sieves under a N2 atmosphere. TMSOTf
(36 mL, 0.199 mmol) was added and reaction mixture stirred at RT for
40 min. Solid NaHCO3 (400 mg) was then added and the stirring contin-
ued for a further 20 min. The mixture was filtered through Celite, the sol-
Chem. Eur. J. 2007, 13, 902 – 909
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
907

P. V. Murphy et al.
vent was removed and the residue was purified by flash chromatography
(EtOAc/cyclohexane, 1:4) to give compound 19b (282 mg, 32%).
1H NMR (300 MHz, CDCl3): d=5.85 (tdd, J=5.9, 10.4, 16.3 Hz, 1H;
CH=CH2), 5.29 (m, 4H; CH=CH2, H-2 and H-3 overlapping), 4.99 (t,
layer dried (Na2SO4) and filtered, and the solvent removed to give com-
pound 20c (221 mg, 81%). 1H NMR (300 MHz, CDCl3): d=9.04 (brs,
1H; COOH), 7.30 (d, J=8.2 Hz, 2H; Ar H), 7.07 (t, J=7.7 Hz, 1H; Ar
H), 7.00 (d, J=8.3 Hz, 2H; Ar H), 5.35 (m, 3H; H-2, H-3, H-4 overlap-
ping), 5.16 (d, J1,2 =7.2 Hz, 1H; H-1), 4.22 (d, J4,5 =9.5 Hz, 1H; H-5),
2.06, 2.04, 2.04 ppm (3”s, 3”3H; 3”COCH3); 13C NMR (75 MHz,
CDCl3): d=170.3 (C, COOH), 170.0, 169.7, 169.4 (3”C, 3”COCH3),
156.6 (C, Ar-C), 129.6 (CH, Ar-C), 123.6 (CH, Ar-C), 117.1 (CH, Ar-C),
99.1 (CH, C-1), 71.9, 70.9, 68.9 (3”CH, C-2–5), 20.5 ppm (3”CH3, over-
lapping, 3”COCH3); ES-HRMS: m/z calcd: 395.0978; found 395.0963
[MÀH]À
1,6-Anhydro-2-azido-3-O-benzyl-2-deoxy-4-O-(2,3,4-tri-O-acetyl-6-(2-
propenyl)-b-d- glucopyranuronosyl)-b-d-glucopyranose (24): A solution
of sodium hypochlorite (15 mL, 19.38 mmol (1.3m, Aldrich)) was added
dropwise to a solution of 1,6-anhydro-2-azido-3-O-benzyl-2-deoxy-4-O-
(b-d-glucopyranosyl)-b-d-glucopyranose 23 (1.0 g, 2.28 mmol), KBr
J
3,4 =9.5 Hz, J4,5 =10.0 Hz, 1H; H-4), 4.76 (d, J1,2 =7.7 Hz, 1H; H-1), 4.66
(d, J=5.9 Hz, 2H; OCH2CH=CH2), 4.32 (d, J4,5 =10.0 Hz, 1H; H-5), 3.59
(dt, J=3.7, 8.5 Hz, 1H; cyclohexyl CH), 2.06, 2.04, 2.02 (3”s, 3”3H; 3”
COCH3), 1.50 ppm (m, 10H; cyclohexyl CH2); 13C NMR (75 MHz,
CDCl3): d=170.1, 169.7, 169.6 (3”C, 3”COCH3), 166.0 (C, COOAll),
131.3 (CH, CH=CH2), 118.0 (CH2, CH=CH2), 98.8 (CH, C-1), 77.0, 72.2,
71.7, 71.4 (4”CH, C-2–5), 69.7 (CH, cyclohexyl CH), 66.0 (CH2,
OCH2CH=CH2), 32.8, 31.2, 25.2, 23.2, 23.1 (5”CH2, cyclohexyl CH2),
19.2, 19.2, 19.1 ppm (3”CH3, 3”COCH3); ES-HRMS: m/z calcd:
465.1737; found 465.1742 [M+Na]+.
Cyclohexyl 2,3,4-tri-O-acetyl-b-d-glucopyranosiduronic acid (20b): The
propenyl ester 19b (250 mg, 0.56 mmol) was dissolved in dry MeCN
(4 mL) under an atmosphere of N2 and cooled to 08C. [Pd
A
(28.0 mg,
0.228 mmol),
and
2,2,6,6-tetramethylpiperdine-1-oxide
0.056 mmol) and pyrrolidine (55 mL, 0.616 mmol) were added and mix-
ture was stirred at RT for 1 h. The mixture was then filtered through
Celite, the solvent was removed, and the residue was dissolved in EtOAc
and washed with water. The aqueous layer was then acidified to pH 2
and extracted with EtOAc, dried (Na2SO4), and filtered and the solvent
removed to yield compound 20b (133 mg, 59%). 1H NMR (300 MHz,
CDCl3): d=5.25 (m, 2H; H-3 and H-4 overlapping), 4.98 (t, J2,3 =8.6 Hz,
(TEMPO, 36.0 mg, 0.228 mmol), in satd NaHCO3/THF (10:1.5, 58 mL)
which was stirring at 08C. The reaction was allowed to attain RT, and
monitored by TLC (Rf =0.4, iPrOH/MeNO2/H2O 10:9:2). After 30 min
the THF was removed under reduced pressure at 308C and the remain-
ing solution was lyophilised to give the intermediate as a white solid. The
residue (1 g, 2.2 mmol) was suspended in anhydrous DMF (20 mL) and
stirred in the presence of allyl iodide (265 mL, 2.86 mmol) for 12 h at RT.
Upon completion of the reaction as determined by TLC (EtOAc/MeOH,
2:1), acetic anhydride (5 mL) and 4-dimethylaminopyridine (30 mg,
0.246 mmol) were added and the stirring was continued for a further 12 h
at RT. The mixture was diluted with water and extracted with EtOAc.
The combined organic extracts were dried (Na2SO4) and filtered, the sol-
vent was removed and the residue purified by chromatography to give
compound 24 (858 mg, 63%). 1H NMR (300 MHz, CDCl3): d=7.24 (m,
5H; Ar-H), 5.81 (tdd, J=6.0, 10.4, 16.6 Hz, 1H; CH=CH2), 5.38 (s, 1H;
J
1,2 =8.3 Hz, 1H; H-2), 4.69 (d, J1,2 =8.3 Hz, 1H; H-1), 4.12 (d, J4,5 =
9.4 Hz, 1H; H-5), 3.59 (dt, J=4.0, 8.6 Hz, 1H; cyclohexyl CH), 2.04,
2.04, 2.02 (3”s, 3”3H; 3”COCH3), 1.50 ppm (m, 10H; cyclohexyl CH2);
13C NMR (75 MHz, CDCl3): d=170.3 (C, COOH), 170.1, 169.7, 169.3
(3”C, 3”COCH3), 99.0 (CH, C-1), 78.0 (CH, cyclohexyl CH), 72.1, 71.9,
71.3, 69.2 (4”CH, C-2–5), 33.0, 31.3, 25.3, 23.5, 23.4 (5”CH2, cyclohexyl
CH), 20.6, 20.5 ppm (3”CH3, 3”COCH3); ES-HRMS: m/z calcd:
401.1448; found 401.1445 [MÀH]À.
H-1*), 5.22 (m, 4H; CH=CH2, H-3, H-4 overlapping), 4.95 (dd, J1,2
=
Phenyl 2,3,4-tri-O-acetyl-b-d-glucopyranosiduronic acid, 2-propenyl ester
(19c): Trichloroacetimidate 22 (1.0 g, 1.99 mmol) and pre-dried 4 Š mo-
lecular sieves were stirred in anhydrous dichloromethane (30 mL) under
a N2 atmosphere for 1 h. Phenol (225 mg, 2.39 mmol) was added to this
solution and stirring was continued for 1 h; then TMSOTf (36 mL,
0.199 mmol) was added and the mixture was stirred for 40 min at RT.
Solid NaHCO3 (400 mg) was then added and stirring continued for an-
other 20 min and the mixture filtered through Celite. The solvent was re-
moved and the residue purified by flash chromatography (EtOAc/cyclo-
hexane, 1:4) to give compound 19c (442 mg, 51%). 1H NMR (300 MHz,
CDCl3): d=7.30 (d, J=7.7 Hz, 2H; Ar H), 7.08 (t, J=7.4 Hz, 1H; Ar
H), 7.01 (d, J=8.6 Hz, 2H; Ar H), 5.86 (tdd, J=5.9, 10.7, 16.6 Hz, 1H;
CH=CH2), 5.30 (overlapping signals, 5H; CH=CH2, H-2, H-3, H-4), 5.14
(d, J1,2 =6.8 Hz, 1H; H-1), 4.60 (d, J=5.8 Hz, 2H; OCH2CH=CH2), 4.20
(d, J4,5 =8.9 Hz, 1H; H-5), 2.06, 2.04, 2.02 ppm (3”s, 3”3H; 3”COCH3);
13C NMR (75 MHz, CDCl3): d=170.1, 169.2, 169.2 (3”C, 3”COCH3),
166.0 (C, COOAll), 156.7 (C, Ar-C), 131.0 (CH, CH=CH2), 129.6 (CH,
Ar-C), 123.5 (CH, Ar-C), 119.3 (CH2, CH=CH2), 117.2 (CH, Ar-C), 99.3
(CH, C-1), 72.6, 72.0, 71.0, 69.1 (4”CH, C-2–5), 66.6 (CH2, OCH2CH=
CH2), 20.6 (CH3, COCH3), 20.5 ppm (2”CH3, overlapping, 2”COCH3);
ES-HRMS: m/z calcd: 459.1288; found: 459.1288 [M+Na]+. Alternative-
ly, the propenyl ester 21 (0.15 g, 0.25 mmol) was dissolved in dry CH2Cl2
(4 mL) under N2 and TMSOPh (0.15 g, 0.62 mmol, 2.5 equiv) and SnCl4
(22 mL, 0.16 mmol, 0.5 equiv) were then added. The reaction mixture was
stirred for 4 h at RT and then satd NaHCO3 (10 mL) and CH2Cl2 (6 mL)
were added and stirring continued for a further 30 min. Filtration through
Celite was followed by separation of the layers and the organic layer was
dried (Na2SO4) and the solvent removed. Chromatography of the residue
gave compound 19c (58%).
7.7 Hz, J2,3 =9.2 Hz, 1H; H-2), 4.75 (d, J1,2 =7.7 Hz, 1H; H-1), 4.51 (m,
5H; CH2Ph, OCH2CH=CH2, H-5* overlapping), 3.99 (m, 2H; H-6a*, H-
5 overlapping), 3.77 (s, 1H; H-6b*), 3.70 (m, 2H; H-3*, H-4* overlap-
ping), 3.12 (s, 1H; H-2*), 1.96, 1.93, 1.93 ppm (3”s, 3”3H; 3”COCH3);
13C NMR (75 MHz, CDCl3): d=170.1, 169.2, 169.1 (3”C, 3”COCH3),
166.1 (C, COOAll), 137.3 (C, Ar-C), 131.0 (CH, CH=CH2), 128.4 (CH,
Ar-C), 127.9 (C, Ar-C), 125.9 (CH, Ar-C), 119.5 (CH2, CH=CH2), 100.7,
98.9 (2”CH, 2”C-1), 76.7, 75.3, 73.3, 72.5, 72.0, 71.0, 69.1 (7”CH), 68.9
(CH2, CH2-Ar), 66.6 (CH2, OCH2CH=CH2), 64.9 (CH2, C-6*), 59.4 (CH),
20.5, 20.5 ppm (3”CH3, 3”COCH3); ES-HRMS: m/z calcd: 642.1911;
found 642.1928 [M+Na]+.
1,6-Anhydro-2-azido-3-O-benzyl-2-deoxy-4-O-(2,3,4-tri-O-acetyl-b-d-glu-
copyranuronosyl)-b-d-glucopyranose (20d): Propenyl ester 24 (250 mg,
0.404 mmol) was dissolved in anhydr MeCN (5 mL) and the mixture was
cooled to 08C. [Pd(Ph3)4] (48 mg, 0.041 mmol) and pyrrolidine (38 mL,
R
0.451 mmol) were then added and stirring was continued (1 h). The mix-
ture was then filtered through Celite, the solvent removed and the resi-
due was taken up in EtOAc, which was then washed with water. The
aqueous layer was then acidified (pH 2–3) and washed with EtOAc. The
organic layer was dried (Na2SO4) and filtered and the solvent removed to
give compound 20d (206 mg, 88%). 1H NMR (300 MHz, CDCl3): d=
7.32 (m, 5H; Ar-H), 5.46 (s, 1H; H-1*), 5.28 (m, 2H; H-3, H-4 overlap-
ping), 5.01 (t, J1,2 =7.6 Hz, J2,3 =9.3 Hz, 1H; H-2), 4.79 (d, J1,2 =7.6 Hz,
1H; H-1), 4.60 (m, 3H; CH2Ph, H-5* overlapping), 4.09 (m, 2H; H-6a*,
H-5 overlapping), 4.02 (d, J4,5 =9.6 Hz, 1H; H-5), 3.83 (s, 1H; H-6b*),
3.76 (brs, 2H; H-3*, H-4*), 3.20 (s, 1H; H-2*), 2.04, 2.03, 2.01 ppm (3”s,
3”3H; 3”COCH3); 13C NMR (75 MHz, CDCl3): d=170.2, 169.7, 169.6
(3”C, 3”COCH3), 169.3 (C, COOH), 137.3 (C, Ar-C), 128.5 (CH, Ar-
C), 128.0 (C, Ar-C), 127.8 (CH, Ar-C), 100.6, 98.9 (2”CH, 2”C-1), 77.1,
75.5, 73.4 (2”CH), 72.6 (CH2, C-6*), 71.1, 69.0 (2”CH), 65.0 (CH2, CH2-
Ar), 63.2, 59.4 (2”CH), 23.7, 20.5 ppm (3”CH3, 3”COCH3); ES-
HRMS: m/z calcd: 578.1622; found 578.1631 [MÀH]À.
Phenyl 2,3,4-tri-O-acetyl-b-d-glucopyranosiduronic acid (20c): Propenyl
ester 19c (300 mg, 0.69 mmol) was dissolved in dry MeCN (5 mL) under
N2 and cooled to 08C. [Pd(Ph3)4] (80 mg, 0.069 mmol) and pyrrolidine
A
(65 mL, 0.76 mmol) were added and the mixture was stirred for 40 min.
The mixture was then filtered through Celite and excess solvent re-
moved. The residue was taken up in EtOAc and washed with water. The
aqueous layer was acidified to pH 2, washed with EtOAc, the organic
1,6-Anhydro-2-azido-3-O-benzyl-2-deoxy-4-O-(2,3,4-tri-O-acetyl-6-
methyl-a/b-d-glucopyranuronosyl)-b-d-glucopyranose (27 and 28): TiCl4
(11 mL, 0.1 mmol) was added to
a
stirred solution of 20d (58 mg,
908
www.chemeurj.org
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 902 – 909

Glycoside Synthesis
FULL PAPER
0.1 mmol) in nitromethane (3 mL) under a N2 atmosphere and stirring
was continued (48 h). Saturated NaHCO3 was added and the mixture was
stirred for a further 20 min. The mixture was filtered through Celite and
the layers separated. The aqueous layer was then acidified (pH 2–3), ex-
tracted with EtOAc and dried (Na2SO4) and the solvent removed to yield
1,6-anhydro-2-azido-3-O-benzyl-2-deoxy-4-O-(2,3,4-tri-O-acetyl-a/b-d-
glucopyranuronosyl)-b-d-glucopyranose[17] (quantitative). This mixture
(60 mg, 0.104 mmol) was dissolved in anhydrous DMF, solid NaHCO3
(14 mg, 0.156 mmol) was added and the stirring continued for 20 min. Io-
domethane (10 mL, 0.135 mmol) was then added and the reaction mixture
left to stir for 12 h and was then filtered through Celite, the solvent was
removed and the residue was taken up in EtOAc and was subsequently
washed with water, dried (Na2SO4) and then filtered. The excess solvent
was removed and the residue was purified by flash chromatography
(40 mg, 64%, a/b, 34:66); ES-LRMS: m/z calcd: 616.0 ; found 616.1
[M+Na]+; elemental analysis calcd (%) for C26H31N3O13: C 52.61, H 5.26,
N 7.08; found C 52.75, H 5.30, N 6.58.
[8] M. Polµkovµ, N. Pitt, M. Tosin, P. V. Murphy, Angew. Chem. 2004,
116, 2572–2575; Angew. Chem. Int. Ed. 2004, 43, 2518–2521.
[9] Anchimeric assistance by the carboxylic acid group or anion in cat-
alysis of glucuronic acid derivatives has been suggested previously.
M. D. Saunder, T. E. Imell, Carbohydr. Res. 1968, 6, 12–16.
[10] M. Tosin, C. OꢀBrien, G. M. Fitzpatrick, H. Müller-Bunz, W. K.
Glass, P. V. Murphy, J. Org. Chem. 2005, 70, 4096–4106.
[11] B. Fischer, A. Nudelman, M. Ruse, J. Herzig, H. E. Gottlieb, J. Org.
Chem. 1984, 49, 4988–4993.
[12] a) Z. Gyçrgydeµk, J. Thiem, Carbohydr. Res. 1995, 268, 85–92; b) E.
Graf von Roedern, E. Lohof, G. Hessler, M. Hoffmann, H. Kessler,
J. Am. Chem. Soc. 1996, 118, 10156–10167.
[13] F. D. Tropper, F. O. Andersson, S. Braun, R. Roy, Synthesis 1992,
618–620.
[14] N. Ikemoto, O. Kyung Kim, L.-C. Lo, V. Satyanarayana, M. Chang,
K. Nakanishi, Tetrahedron Lett. 1992, 33, 4295–4298.
[15] E. Smits, J. B. N. F. Engberts, R. M. McKellog, H. A. Von Doren, J.
Chem. Soc. Perkin Trans. 1 1996, 2873–2877.
1
Data for b-anomer: H NMR (500 MHz, CDCl3): d=7.32 (m, 5H; Ar-H),
[16] J. R. Ferguson, J. R. Harding, D. A. Killick, K. W. Lumbard, F.
Scheinman, A. V. Stachulski, J. Chem. Soc. Perkin Trans. 1 2001,
3037–3041.
[17] A. OꢀBrien, C. Lynch, K. M. OꢀBoyle, P. V. Murphy, Carbohydr. Res.
2004, 339, 2343–2354.
[18] The procedure used was adapted from that described previously.
See H. Kunz, H. Waldmann, Angew. Chem. 1984, 96, 49–50; Angew.
Chem. Int. Ed. Engl. 1984, 23, 71–72.
[19] In the presence of a catalyst generated from SnCl4 and a silver salt,
various 2-amino-2-deoxy-a-d-glucopyranosides or galactopyrano-
sides can be synthesised with high stereoselectivities through glycosi-
dation–anomerisation starting from 1,3,4,6-tetra-O-acetyl-2-deoxy-
2(2,2,2-trichloroethoxycarbonylamino)-b-d-glucopyranose or galac-
topyranose and alkyl trimethylsilyl ethers, respectively. K. Matsu-
bara, T. Mukaiyama, Chem. Lett. 1993, 2145–2148.
[20] Mixtures of anomers of glucuronides have been obtained from estra-
diol derivatives in the presence of SnCl4 in CH2Cl2. H. E. Hadd, W.
Slikker, Jr., D. W. Miller, E. D. Helton, W. L. Duax, P. D. Strong,
D. C. Swenson, J. Steroid Biochem. 1983, 18, 81–87.
[21] The reaction of 8-ethoxycarbonyloctanol by cellobiose, lactose, and
maltose octaacetates in the presence of SnCl4 gives mixtures of a-
and b-glycosides, the a-glycosides being formed through anomerisa-
tion reactions. J. Banoub, D. R. Bundle, Can. J. Chem. 1979, 57,
2085–2090.
5.46 (s, 1H; H-1*), 5.28 (m, 2H; H-3, H-4 overlapping), 5.01 (t, J1,2
7.8 Hz, J2,3 =7.8 Hz, 1H; H-2), 4.79 (d, J1,2 =7.7 Hz, 1H; H-1), 4.64 (d,
J=5.1 Hz, 2H; CH2Ph), 4.60 (dd, J=5.1, 11.1 Hz, 1H; H-5*), 4.09 (d,
=
J
6a*,5 =7.2 Hz, 2H; H-6a*), 4.02 (d, J4,5 =9.6 Hz, 1H; H-5), 3.83 (s, 1H;
H-6b*), 3.76 (brs, 2H; H-3*, H-4*), 3.75 (s, 3H; OCH3), 3.20 (s, 1H; H-
2*), 2.04, 2.03, 2.01 ppm (3”s, 3”3H; 3”COCH3); 13C NMR (125 MHz,
CDCl3): d=170.2, 169.7, 169.6 (3”C, 3”COCH3), 169.3 (C, COOH),
137.3 (C, Ar-C), 128.5 (CH, Ar-C), 128.0 (C, Ar-C), 127.8 (CH, Ar-C),
100.6 (CH, C-1*), 98.9 (CH, C-1), 77.1, 75.5, 73.4 (3”CH), 72.6 (CH2, C-
6*), 71.1, 69.0 (2”CH), 65.0 (CH2, CH2-Ar), 63.2, 59.4 (2”CH), 53.2
(CH3, OMe), 20.6, 20.4, 20.4 ppm (3”CH3, 3”COCH3); IR (KBr): n˜ =
3457, 2918, 2107, 1758, 1437, 1377, 1255, 1046, 631 cmÀ1
Selected data for a-anomer: 1H NMR (500 MHz, CDCl3): d=7.32 (m,
.
5H; Ar-H), 5.59 (t, J3,4 =9.7, 1H; H-3), 5.58 (s, 1H; H-1*), 5.23 (d, J1,2
3.9, 1H; H-1), 5.19 (t, J3,4 =9.6 Hz, J4,5 =9.8 Hz, 1H; H-4), 4.86 (dd, J1,2
=
=
3.9 Hz, J2,3 =10.2 Hz, 1H; H-2), 4.71 (d, J=5.8 Hz, 2H; CH2Ph), 4.57 (m,
2H; H-5*, H-5 overlapping), 4.13 (d, J6a*,5 =7.4 Hz, 1H; H-6a*), 3.73 (s,
3H; OCH3), 3.59 (s, 1H; H-4*), 3.52 (s, 1H; H-3*), 3.16 (s, 1H; H-2*),
2.04, 2.03, 2.01 ppm (3”s, 3”3H; 3”COCH3); 13C NMR (125 MHz,
CDCl3): d=136.7 (C, Ar-C), 128.7 (CH, Ar-C), 128.3 (C, Ar-C), 127.7
(CH, Ar-C), 100.8 (CH, C-1*), 97.0 (CH, C-1), 75.0 (CH2, CH2-Ar), 72.4,
70.4, 69.4, 69.2 (4”CH, C-2–5), 53.5 ppm (CH3, OMe).
[22] The anomerisation of isopropyl 2,3,4,6-tetra-O-acetyl-b-d-glucopyra-
noside catalysed by SnCl4 had a half time of about 1400 min; in the
presence of one equivalent of acetic acid the half time was approxi-
mately 14 min. The anomerisation of isopropyl 3,4,6-tri-O-acetyl-2-
O-(2-carboxypropanoyl)-b-d-glucopyranoside catalysed by SnCl4
had a half time of roughly 10 min. R. U. Lemieux, O. Hindsgaul,
Carbohydr. Res. 1980, 82, 195–206.
[23] a) T. E. Timell, W. Enterman, F. Spencer, E. J. Soltes, Can. J. Chem.
1965, 43, 2296–2298; b) M. D. Saunders, T. E. Timell, Carbohydr.
Res. 1967, 5, 453–460; c) M. D. Saunders, T. E. Timell, Carbohydr.
Res. 1968, 6, 12–17.
Acknowledgements
This work was supported by Enterprise Ireland (SC2000/182, SC2002/
225), IRCSET (Ph.D. Scholarship to M.T.) and Science Foundation Ire-
land (04/BR/C0192).
[1] A. Varki, Glycobiology 1993, 3, 97–130.
[2] F. Schweizer, O. Hindsgaul, Curr. Opin. Chem. Biol. 1999, 3, 291–
298.
[24] H. HelligsØ Jensen, M. Bols, Acc. Chem. Res. 2006, 39, 259–265.
[25] B. Capon, Tetrahedron Lett. 1963, 4, 911–913.
[3] R. R. Schmidt, Angew. Chem. 1986, 98, 213; Angew. Chem. Int. Ed.
Engl. 1986, 25, 212–235.
[4] For a review on 1,2-cis-glycoside synthesis see A. V. Demchenko,
Synlett 2003, 1225–1240.
[26] The hydrolysis of glycosides proceeding via an oxycarbenium ion
with a boat conformation cannot be discounted. See Y. Blꢂriot, S. K.
Vadivel, A. J. Herrera, I. R. Greig, A. J. Kirby, P. Sinaꢃ, Tetrahedron
2005, 60, 6813–6828.
[5] M. Tosin, P. V. Murphy, Org. Lett. 2002, 4, 3675–3678.
[6] J. P. Malkinson, R. A. Falconer, I. Toth, J. Org. Chem. 2000, 65,
5249–5252.
[27] J. Janson, B. Lindberg, Acta Chem. Scand. 1960, 14, 877–881.
[28] T. D. Audichya, T. R. Ingle, J. L. Bose, J. L. Indian J. Chem. Sect B
1976, 14, 369–370.
[7] N. Roeckendorf, T. K. Lindhorst, J. Org. Chem. 2004, 69, 4441–
4445.
Received: July 31, 2006
Published online: November 6, 2006
Chem. Eur. J. 2007, 13, 902 – 909
ꢁ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
909