An Alternative Reaction Course in O-Glycosidation with O-Glycosyl Trichloroacetimidates as Glycosyl Donors and Lewis Acidic Metal Salts as Catalyst: Acid-Base Catalysis with Gold Chloride-Glycosyl Acceptor Adducts

Article abstract of DOI:10.1021/jacs.5b07895

Gold(III) chloride as catalyst for O-glycosyl trichloroacetimidate activation revealed low affinity to the glycosyl donor but high affinity to the hydroxy group of the acceptor alcohol moiety, thus leading to catalyst-acceptor adduct formation. Charge sep

Full text of DOI:10.1021/jacs.5b07895

Subscriber access provided by CMU Libraries - http://library.cmich.edu
Article
An Alternative Reaction Course in O-Glycosidation with
O-Glycosyl Trichloroacetimidates as Glycosyl Donors
and Lewis Acidic Metal Salts as Catalyst: Acid-Base
Catalysis with Gold Chloride-Glycosyl Acceptor Adducts
Peng Peng, and Richard R. Schmidt
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.5b07895 • Publication Date (Web): 11 Sep 2015
Downloaded from http://pubs.acs.org on September 16, 2015
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the
course of their duties.

Page 1 of 8
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
An Alternative Reaction Course in O‐Glycosidation with O‐Glycosyl
Trichloroacetimidates as Glycosyl Donors and Lewis Acidic Metal
Salts as Catalyst: Acid‐Base Catalysis with Gold Chloride‐Glycosyl
Acceptor Adducts
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Peng Peng and Richard R. Schmidt*
Department of Chemistry, University of Konstanz, Germany
ABSTRACT: Gold(III) chloride as catalyst for O‐glycosyl trichloroacetimidate activation revealed low affinity to the glycosyl
donor but high affinity to the hydroxy group of the acceptor alcohol moiety, thus leading to catalyst‐acceptor adduct for‐
mation. Charge separation in this adduct, increasing the proton acidity and the oxygen nucleophilicity, permits donor ac‐
tivation and concomitant acceptor transfer in a hydrogen‐bond mediated S 2‐type transition state. Hence, the sequential
N
binding between acceptor and catalyst and then with the glycosyl donor enables self‐organization of an ordered transition‐
o
state. This way, with various acceptors even at temperatures below ‐60 C fast and high yielding glycosidations in high
anomeric selectivities were recorded, showing the power of this gold(III) chloride acid‐base catalysis. Alternative reaction
courses via hydrogen chloride or HAuCl activation or intermediate generation of glycosyl chloride as the real donor could
4
be excluded. With partially O‐protected acceptors, prone to bidentate ligation to gold(III) chloride, particularly high reac‐
tivities and anomeric selectivities were observed. Gold(I) chloride follows the same catalyst‐acceptor adduct driven acid‐
base catalysis reaction course.
Introduction
In accordance with this general reaction scheme, cata‐
lysts with high affinity to the glycosyl donor leaving group
For the activation of O‐glycosyl trichloroacetimidate gly‐
cosyl donors in the presence of glycosyl acceptors (for in‐
stance, hydroxy compounds) mainly catalytic amounts of
trimethylsilyl trifluromethanesulfonate (TMSOTf) or bo‐
(
as for instance TMSOTf or BF
3
∙Et O, respectively) exhibit
2
also disadvantages: As these catalysts generate the highly
reactive glycosyl cation intermediate irrespective of the
presence or absence of an acceptor, they permit competing
reactions that may lead to loss of the glycosyl donor prop‐
erties. To overcome this problem, recently a variation of
this reaction scheme was designed by choosing Lewis acid
catalysts (i) with low affinity to the glycosyl donor leaving
group, thus not directly generating glycosyl cations, how‐
ever (ii) with high affinity to the glycosyl acceptor, gener‐
ating in a rapid and reversible reaction an acceptor‐catalyst
adduct RO‐Cat‐H (Scheme 2).24 The ensuing (iii) increase
in proton acidity and (iv) increase of acceptor nucleophilic‐
ity in this adduct should provide the glycosyl donor activa‐
tion via proton transfer of the acidified acceptor proton to
the glycosyl donor and concomitantly to acceptor transfer
rontrifluoride etherate (BF
3
∙Et O), respectively, are em‐
2
ployed.1 There is good evidence that the activation of O‐
‐5
glycosyl trichloroacetimidates is effected by direct attack
of the Lewis acid or the Brønsted acid catalyst (Cat) at the
imidate nitrogen (Scheme 1). Thus, via an O‐glycosyl tri‐
chloroacetimidate‐catalyst transition state a glycosyl oxo‐
carbenium ion (short: glycosyl cation) intermediate and
trichloroacetamide (TCAA) as leaving group are generated.
The glycosyl cation reacts directly with the acceptor to the
glycoside, thus the anomeric selectivity is essentially deter‐
mined by steric and/or stereoelectronic effects of the gly‐
cosyl donor. Alternatively, the evolving glycosyl cation is
stabilized by anchimeric assistance, the solvent or by other
nucleophiles present in the reaction mixture; this way
these effectors influence the yield and particularly the ano‐
meric selectivity in the glycosidation step.
‐
from the negatively charged [RO‐Cat] moiety to the incip‐
ient glycosyl cation in an S 2‐type transition state. Thus, in
N
an hydrogen‐bond mediated self‐organizing intramolecu‐
lar acid‐base catalyzed reaction the glycoside should be
generated in high yield and generally high anomeric selec‐
tivity.2 This reaction design could be verified by us, for
For the O‐glycosyl trichloroacetimidate activation also
quite a few Lewis acidic metal salts were investigated and
the same reaction course was envisaged. Worth mention‐
4,25
instance, for PhBF
2
, Ph
2
BF or PhSiF , respectively, as cata‐
3
6‐11
12‐14
15
ing are Ni(II) and Pd(II) salts , ZnBr
2
, MgBr
2
16
∙Et
2
O in
lysts.24,26‐28
17
the presence and absence of a Brønsted acid, AuCl,
1
8
18
18
19
,20
As some of the metal salts employed for O‐glycosyl tri‐
chloroacetimidate activation are relatively weak Lewis ac‐
ids, the generally claimed direct activation of the
InCl
3
,
InBr
3
,
In(OTf)
3
,
Sm(OTf)
3
,
Yb(OTf)
3
21
22,23
Sn(OTf)
2
, AgOTf.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 8
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 1. Glycosyl cation generation and eventual concomitant intra‐/intermolecular stabilization and final
transformation into glycoside.
Scheme 2. Acceptor‐Catalyst adduct formation leads to acid‐base catalyzed glycosidation with a catalyst having
(i) low affinity to the donor but (ii) high affinity to the acceptor.
glycosyl donor by these catalysts is questionable. Rather a
high affinity to the acceptor hydroxy group is expected,
that favors the pathway outlined in Scheme 2. In order to
study this question, that has a major impact on product
formation, gold(III) chloride, with its strong tendency to
Reaction course of gold(III) chloride catalysis. In order
to cope with this task, first the interaction of gold(III) chlo‐
ride with glycosyl donor 1α (Scheme 3) in CDCl as solvent
3
was studied by NMR spectroscopy: Interaction of 1α with
o
0.1 equivalents of gold(III) chloride at ‐50 C was, if at all,
form dimers (2 AuCl
position four ligands in a square‐planar orientation,
seemed to be an ideal catalyst candidate: De‐dimerization
3
Au
2
Cl
6
) and in addition to
only very weak, as indicated by NMR shift differences (Fig‐
ure 1); 1α was also not decomposed (Table 1, entry 1). Only
by raising the temperature (entry 2) or by adding 1.0 equiv‐
alent of gold(III) chloride (entry 3) slow decomposition of
1α took place. However, interaction of gold(III) chloride
with an acceptor, as for instance isopropanol (A), was very
strong (Figure 2), thus demonstrating the formation of the
2
9
δ‐
of Au
2
Cl
6
by alcohols (HOR) should lead to an RO ‐AuCl
3
‐
δ+
H
adduct with strong charge separation for proton‐in‐
duced O‐glycosyl trichloroacetimidate activation and suf‐
ficient propensity to concomitant alkoxide transfer to the
incipient glycosyl cation (Scheme 2). Binding of the accep‐
tor in a square‐planar orientation will lead to a less hin‐
dered transition state than in a tetrahedral orientation of
the four ligands around the central atom (as for instance in
RO‐AuCl
3
‐H adduct (entry 4). Addition of 1α to this mix‐
o
37
ture at ‐60 C, (hence, employing the inverse procedure
for glycosidation) led practically to exclusive formation of
the β‐glucoside 2Aβ (Table 1, entry 5). Essentially the same
result was obtained with the standard glycosidation proce‐
dure, where the donor 1α and the acceptor A were dis‐
solved first in dichloromethane (DCM) and then 0.1 equiv‐
alents of gold(III) chloride were added to the reaction mix‐
2
the PhF B∙HOR or the Ph FB∙HOR adducts).
2
Gold(III) chloride mediated activation of glycosyl donors
having alkyne moiety containing leaving groups has been
30‐35
reported.
Very recently, Vankar et al. also reported on
o
ture at ‐70 C (entry 6). However, addition of the catalyst
the activation of O‐glycosyl trichloroacetimidates by
gold(III) chloride in the presence of phenylacetylene.
Some annotations on their results will be discussed below.
o
36
to the donor 1α at ‐70 C and thereafter adding the acceptor
A led only to a very sluggish reaction (entry 7) that could
be accelerated by raising the temperature; yet, this resulted
Results and Discussion
ACS Paragon Plus Environment

Page 3 of 8
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Table 1. Interaction and reactions between glycosyl donor 1α, acceptor A and gold(III) chloride as catalysta.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
aEntries 1‐4 were carried out in CDCl3 (in the NMR spectrometer), entries 5‐15 were carried out in DCM.
in partial decomposition of 1α. Presumably, addition of the
catalyst to the donor leads to formation of clusters
[
AuCl
3
∙(1α) , n = 1, 2, 3…] that can only slowly be penetrated
n
by the acceptor molecules, this way slowing down the re‐
action rate. Cluster formation between catalyst and accep‐
tor has been previously observed in the “inverse procedure”,
37
however in this case donor decomposition is obviously
no problem and the mobility of the cluster constituents
seems to be high, therefore there is only a small effect on
the reaction rate.3
8
Besides the proposed gold(III) chloride mediated acid‐
base catalysis of this reaction (Scheme 2), alternative
routes for product formation are available that had to be
investigated (Scheme 3). (a) Reaction of 1α and A under
gold(III) chloride catalysis leads only to minor if any glyco‐
syl chloride 3α, β formation ( Table 1, entries 5‐7). Yet, as
an alternative to gold(III) catalysis, acid catalysis of this re‐
action by hydrogen chloride (HCl), eventually generated
1
Figure 1. H NMR of donor 1α and a mixture of donor 1α
3
9
from the reaction of gold(III) chloride with the acceptor,
had to be investigated. Therefore, 0.1 equivalents
o
and gold(III) chloride (0.1 equivalents) in CDCl at ‐50 C.
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 8
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
further supports the proposed acid‐base catalysis of the
glycosidation reaction between 1α and A by gold(III) chlo‐
ride.
As the gold(III) chloride catalyzed reaction is sensitive to
the presence of impurities that are capable of binding com‐
petitively to gold(III), it was of interest to study the influ‐
ence of solvents with ligand properties. To this end, the in‐
fluence of acetonitrile was studied. As expected, this mol‐
ecule lowered the glycosidation rate, however not the final
glycosidation result (entry 12).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
In this context also the influence of phenylacetylene was
36
studied because of the high alkynophilicity of gold(III).
o
Application of the inverse glycosidation procedure at ‐70 C
in the presence of phenylacetylene led to total inhibition
of the glycosidation reaction; this was even so at room tem‐
perature (entries 13, 14; D was selected as acceptor). How‐
ever, the reaction proceeded nicely with the standard gly‐
1
Figure 2. H NMR of Isopropanol and a mixture of isopro‐
panol and gold(III) chloride in CDCl at room temperature.
3
3
6
cosidation procedure, i.e. first the donor, acceptor and
phenylacetylene were dissolved at room temperature and
then gold(III) chloride was added (entry 15). Therefore,
NMR studies were performed at room temperature, that
showed between gold(III) chloride and phenylacetylene
complex formation. Following the inverse procedure, addi‐
tion of isopropanol to this mixture leads to a new complex
that can also be obtained by adding first isopropanol and
then phenylacetylene to gold(III) chloride (see S.I.) How‐
ever, this new complex is unable to activate O‐glycosyl tri‐
chloroacetimidate glycosyl donor 1α (Table 1, entries 13, 14).
Hence, it seems that under the standard glycosidation pro‐
cedure, formation of this inactive catalyst‐phenylacety‐
lene‐acceptor complex is slower than the direct activation
of glycosyl donor 1α by the catalyst. (Table 1, entry 15).
Applications to glycoside synthesis. The application of
this gold(III) chloride acid‐base catalysis to various accep‐
tors (Table 2, A‐H) having one unprotected hydroxy group
and 1α as donor led to excellent glycosidation yields and
generally to high β‐selectivities, particularly with reactive
acceptors (Table 2, 2A‐H). Worth mentioning are the gly‐
cosidation results obtained with 2‐O‐ and 4‐O‐unprotected
acceptors E and G, respectively, exhibiting the high reac‐
tivity of these hydroxy groups that for steric reasons were
found less reactive (particularly the 4‐hydroxy group) in
Scheme 3. Potential alternative reaction pathways. (R
CHMe
of HCl (Et
=
2
)
2
O adduct) were added to a mixture of donor 1α
and acceptor A; however, only a trace of a 2Aα,β‐mixture
and some glycosyl chloride 3α were formed (Table 1, entry
). Thus, it could be concluded that HCl is not a decisive
8
catalyst for the observed fast β‐glycoside formation. This
was also supported by the fact that higher amounts of HCl
led to formation of a product mixture of 2Aα,β and 3α (en‐
try 9). (b) Alternatively, the HCl‐addition product of
gold(III) chloride i.e. 0.1 equivalents of HAuCl
tigated as catalyst (entry 10). As expected, beside glycoside
Aβ some glycosyl chloride 3α was now obtained. Hence,
after consumption of some HCl from HAuCl subsequently
the above discussed intramolecular gold(III) chloride–me‐
diated acid‐base catalysis is effective in this reaction. Yet,
these studies also exhibit that any compound in the reac‐
tion mixture with affinity to gold(III) chloride will inhibit
the glycosidation, as found for substrates and/or reagents
containing impurities. (c) Finally, the question remains is
glycosyl chloride 3α,β, generated as intermediate and then
the PhBF
2
or Ph
BF catalyzed reactions24 (Scheme 2).
2
4
, was inves‐
Similar results were obtained with 2,3,4,6‐tetra‐O‐ben‐
zyl‐α‐D‐galactopyranosyl trichloroacetimidate 4α as glyco‐
syl donor. With acceptor A and B the β‐glycosides 5Aβ and
2
4
5Bβ, respectively, were generated (Table 2, entries 9, 10).
In accordance with the general reaction scheme the corre‐
sponding β‐D‐galactopyranosyl trichloroacetimidate 4β af‐
forded with acceptor A mainly the α‐glycoside 5Aα (entry
11). However, neighboring group participation, as in 2‐O‐
benzoyl protected α‐D‐mannopyranosyl trichloroacetimi‐
date 6α as glycosyl donor, overrides the formation of the
S 2‐type arrangement between donor and catalyst‐accep‐
N
attacked by gold(III) chloride providing the glycosyl cation
and stable tetrachloroaurate (AuCl
nor? To this end, glycosyl chloride 3α was activated by 0.1
equivalents of gold(III) chloride in the presence of acceptor
A; however, no reaction was observed (entry 11). This result
tor adduct and leads to exclusive formation of the 1,2‐trans
product, i.e. the α‐glycoside 7Aα (entry 12).
‐
4
), the real glycosyl do‐
ACS Paragon Plus Environment

Page 5 of 8
Journal of the American Chemical Society
1
2
3
Table 2. Gold (III) chloride catalyzed glycosidations
with Trichloroacetimidates donors and various
mono‐O‐unprotected acceptors.
amount of released HCl had practically no effect on the re‐
action result (see Table 1, entry 8). Thus, from 4,6‐O‐un‐
protected glucopyranoside I the β‐(1‐6)‐linked disaccha‐
ride 2Iβ and from 3,4‐O‐unprotected α‐D‐galactopyra‐
noside J the β‐(1‐3)‐linked disaccharide 2Jβ were practi‐
cally exclusively obtained (entries 1, 2).
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 3. Gold(III) chloride catalyzed glycosidation
with partly O‐protected acceptors.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
aEntries 1‐6 were carried out with 1.0 equivalent of donor,
1
.5 equivalents of acceptor and 0.15 equivalents of AuCl ;
3
b
Small amounts (<8 %) of β‐(1‐4) linked disaccharide 2Iβ’,
c
2
Jβ’, 2Kβ’, 2Lβ’ and 2Nβ’ were also detected; Trisaccha‐
ride 2Mβ’ was obtained in 16% yield.
OBn HO OTBDPS
O
O
BnO
BnO
OMp
O
OBn OBn
O
O
BnO
BnO
2M
'
OBn
Surprisingly, with the closely related β‐D‐galactopyra‐
noside K as acceptor besides the β‐(1‐3)‐linked disaccha‐
ride 2Kβ as main product also the β‐(1‐4)‐linked disaccha‐
ride 2Kβ’ was found as minor product (entry 3). Though
acceptors K and L are pseudo‐enantiomeric, similar regio‐
and stereoselectivity was observed for acceptor L leading
to 2Lβ and 2Lβ’ (entry 4). Hence, there is, if at all, only a
minor effect of the acceptor stereochemistry on the glyco‐
sidation result. The 2,3,4‐O‐unprotected acceptor M af‐
forded, as expected, the β‐(1‐3)‐linked disaccharide 2Mβ as
main product. However, as minor product β‐(1‐2)‐/β‐(1‐3)‐
linked trisaccharide 2Mβ’ was also obtained (entry 5).
Linkage to the 4‐position was not reported, as it is disfa‐
vored due to the presence of the bulky tert‐butyldiphen‐
ylsilyl (TBDPS) 6‐O protecting group, thus exhibiting how
to control the regioselectivity in these reactions. This was
also found for the reaction of acceptor N where mainly 2Nβ
was obtained (entry 6).
a
b
Donor : acceptor ratio 1 : 1.5; Under strictly anhydrous
conditions the same result was obtained without addition
c
d
of molecular sieves. Donor : acceptor ratio 1.5 : 1; The re‐
o
e
actions were performed at – 60 C. β/α ratio was obtained
from the H‐NMR spectra.
1
In total agreement with the proposed reaction scheme,
partly protected acceptors, that act as bidentate ligands
were more reactive and led to higher yields, and even
higher β‐selectivities than mono‐O‐unprotected acceptors
in these gold(III) chloride catalyzed glycosidation reac‐
tions (Table 3, reactions with acceptors I‐N). Obviously,
this is due to binding of the acceptor via two oxygens to
gold(III) under release of one molecule of HCl. However,
as only 0.1 equivalents of gold(III) chloride were added, the
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 8
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
The glycosidation rate difference between one O‐unpro‐
gold(III) chloride (Tables 3 and 4, compare entries 3). This
tected and di‐O‐unprotected bidentate ligands was estab‐
lished through a competition experiment between struc‐
turally closely related acceptors H and K under the general
glycosidation conditions (see Table 2, 3). As expected, with
excess H and K (1.5 equiv each) from the total disaccharide
yield of 81% a 1:6 ratio of 2Hβ vs 2Kβ + 2Kβ’ was obtained.
Hence, the initial glycosidation rate difference is about 1:10
in favor of diol K (Scheme 4). Contrary to the results with
result is either due to expansion of the linear two‐coordi‐
nate gold(I) complexation to a three‐coordinate, trigonal‐
planar complexation42 or alternatively, due to dispropor‐
tionation of gold(I) chloride to gold(0) and gold(III) chlo‐
ride.4
3
Table 4. Gold(I) chloride catalyzed glycosidations.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
OBn
AuCl
- 60 o
C
OBn
O
O
BnO
BnO
gold(III) chloride, use of PhBF
2
, Ph
2
BF and PhSiF
3
as acid‐
+
ROH
BnO
NH
BnO O
1
BnO
O
OR
base catalysts led to a decrease in reaction rate and yield
with di‐O‐unprotected bidentate ligands.2 Thus, further
support is provided that gold(III) chloride is available to
bidentate ligation and to lower steric constraints in the gly‐
cosidation transition state in a square‐planar ligand orien‐
tation compared with a tetrahedral ligand orientation of
the acceptor oxygen(s).
anhydrous DCM
4Å MS
BnO
4
CCl3
2
ratioc
entry
acceptor
product
OBn
yield
/
1a
O
OH
BnO
BnO
84%
A
F
BnO
2A
Ph
O
Ph
2b
O
O
O
BnO
BnO
BnO
O
O
Borinic acid derivatives, accessible to bidentate ligation
with the acceptors, have been studied with glycosyl halides
as glycosyl donors and their activation by equivalent
O
O
78% : =3:1
OMe
HO
BnO
BnO
OMe
OBn 2F
HO OBn
O
OBn HO OBn
4
0,41
amounts of silver oxide.
However, as the glycosidation
3a
O
O
BnO
OMp BnO
81%
HO
O
OMp
procedure is non catalytic, the high yielding anomeric se‐
lectivities are essentially based on neighboring group par‐
ticipation and high regioselectivities are only obtained
with the help of bulky protecting groups, the results are
not really comparable with the present work, that employs
powerful gold(III) chloride acid‐base catalysis for the con‐
comitant activation of glycosyl donor and acceptor without
resorting to archimeric assistance for anomeric selectivity
control.
BnO
OBn
BnO
K
2K
a
c
b
Donor : Acceptor ratio 1:1.5; Donor : Acceptor ratio 1.5:1;
1
β/α ratio was obtained from the H‐NMR spectra.
Conclusion
Lewis acids with low affinity to the glycosyl donor leav‐
ing group but with high affinity to the glycosyl acceptor are
expected to form catalyst‐acceptor adducts that ‐ appropri‐
ately selected ‐ permit glycosyl donor activation and con‐
comitant acceptor transfer to the incipient glycosyl cation
in an S 2‐type intramolecular fashion. This novel concep‐
N
tual approach to glycosidations that is facilitated by glyco‐
syl donors accessible to activation by catalysis, as the
1
,3
highly reactive O‐glycosyl trichloroacetimidate donors,
works particularly well with gold(III) chloride as catalyst.
This Lewis acid turned out to be a powerful catalyst by
binding firstly to the alcohol generating an adduct that re‐
acts with the donor to the glycoside via acid‐base catalysis,
i.e. via intramolecular dual catalysis. The superior reactiv‐
ity of this system even at low temperatures exhibits the
power of this approach and permits the use of a broad ac‐
ceptor range, resulting in high glycoside yields and ano‐
meric selectivities. The access of gold(III) chloride to bi‐
dentate ligation of diols and the square‐planar arrange‐
ment of the ligands turns out to be a further advantage of
this catalyst system that seems to hold many more sur‐
prises, as the studies with competing ligands for gold(III)
and the extension of this catalysis to gold(I) chloride dis‐
play.
Scheme 4. Competitive reaction of 1α between accep‐
tor K and H catalyzed by gold(III) chloride.
Gold(I) chloride as catalyst. The results obtained with
gold(III) chloride raise the question, do other Lewis acidic
metal salts employed for O‐glycosyl trichloroacetimidate
activation follow the same reaction course? Due to the
close relationship with gold(III) chloride we looked briefly
into the behavior of gold(I) chloride, for which good glyco‐
sidation results with varying anomeric selectivities were
17
obtained at room temperature. We noticed that even 1.0
equivalent of gold(I) chloride is unable to directly activate
and this way decompose O‐glycosyl trichloroacetimidate
1
α. However, with acceptor A complex formation was ob‐
served; following addition of donor 1α, furnished glycoside
Experimental Section
o
2
Aβ even at ‐60 C in very good yield (Table 4, entry 1). 3‐
General procedure for gold(III) catalyzed glyco‐
sidation To a solution of gold (III) chloride (4.4 μmol) and
O‐Unprotected acceptor F afforded under these conditions
disaccharide 2Fα, β in 78% yield in a 1:3 ratio (entry 2); this
way, the reaction rate difference became obvious, with
gold(III) chloride being the more active catalyst. Yet, gold
I) chloride acts also as diol accepting entity, as it led with
acceptor K practically to the same result, as obtained with
4
Å molecular sieves (200 mg)* in 2 mL of anhydrous DCM
was added acceptor A (66 μmol) at room temperature. Af‐
ter cooling down the reaction mixture to – 70 C, highly
o
(
pure donor 1α (44 μmol), dissolved in 1 mL of anhydrous
ACS Paragon Plus Environment

Page 7 of 8
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(11) McConnell, M. S.; Mensah, E. A.; Nguyen, H. M. Carbohydr.
Res. 2013, 381, 146‐152.
12) Yang, J.; Cooper‐Vanosdell, C.; Mensah, E. A.; Nguyen, H. M.
J. Org. Chem. 2008, 73, 794‐800.
13) McKay, M. J.; Naab, B. D.; Mercer, G. J.; Nguyen, H. M. J. Org.
Chem. 2009, 74, 4705‐4711.
14) Mensah, E. A.; Azzarelli, J. M.; Nguyen, H. M. J. Org. Chem.
009, 74, 1650‐1657.
DCM, was slowly added into the reaction. The reaction was
further stirred for 30 min at this temperature. After the
TLC analysis showed the completion of the reaction, the
reaction was quenched by addition of triethylamine and di‐
luted with 50 mL of DCM. Then the precipitate was filtered
off through a pad of Celite. The organic layer was washed
(
(
(
2
with NaHCO
3
(aq.) (10 mL) and brine (10 mL), dried over
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Na SO , filtered, and concentrated. The residue was puri‐
2
4
(15) Urban, F. J.; Moore, B. S.; Breitenbach, R. Tetrahedron Lett.
1990, 31, 4421‐4424.
fied by column chromatography (hexane/acetone or hex‐
ane/ethyl acetate) on silica gel.
(16) Gould, N. D.; Liana Allen, C.; Nam, B. C.; Schepartz, A.; Miller,
S. J. Carbohydr. Res. 2013, 382, 36‐42.
*When substrates, reagents and solvents are particularly
carefully dried, molecular sieves are not required in order
to obtain the same results. (see Table 2, entry 1).
(
(
17) Goetze, S.; Fitzner, R.; Kunz, H. Synlett 2009, 3346‐3348.
18) Mattson, A. L.; Michel, A. K.; Cloninger, M. J. Carbohydr. Res.
2012, 347, 142‐146.
(19) Adinolfi, M.; Barone, G.; Guariniello, L.; Iadonisi, A. Tetrahe‐
dron Lett. 2000, 41, 9005‐9008.
ASSOCIATED CONTENT
(
20) Adinolfi, M.; Barone, G.; Iadonisi, A.; Mangoni, L.; Schiat‐
tarella, M. Tetrahedron Lett. 2001, 42, 5967‐5969.
21) Bartek, J.; Müller, R.; Kosma, P. Carbohydr. Res. 1998, 308,
59‐273.
22) Douglas, S. P.; Whitfield, D. M.; Krepinsky, J. J. J. Carbohydr.
Supporting Information
(
2
(
1
13
Full experimental details and H and C NMR spectra for
all new compounds.The Supporting Information is availa‐
ble free of charge on the ACS Publication web‐
site.http://pubs.acs.org.”
Chem. 1993, 12, 131‐136.
(23) Wei, G.; Gu, G.; Du, Y. J. Carbohydr. Chem. 2003, 22, 385‐393.
(
2
(
24) Kumar, A.; Kumar, V.; Dere, R. T.; Schmidt, R. R. Org. Lett.
011, 13, 3612‐3615.
25) Geng, Y.; Kumar, A.; Faidallah, H. M.; Albar, H. A.; Mhkalid,
I. A.; Schmidt, R. R. Angew. Chem. Int. Ed. 2013, 52, 10089‐10092.
26) Kumar, A.; Geng, Y.; Schmidt, R. R. Adv. Synth. Catal. 2012,
54, 1489‐1499.
AUTHOR INFORMATION
(
3
Corresponding Author
richard.schmidt@uni‐konstanz.de
(27) Kumar, A.; Schmidt, R. R. Eur. J. Org. Chem. 2012, 2715‐2719.
(28) Kumar, A.; Schmidt, R. R. In Glycoscience: Biology and Medi‐
cine; Taniguchi, N.; Endo, T.; Hart, G. W.; Seeberger, P. H.; Wong,
C.‐H.; Eds.; Springer Japan: 2014, p 295‐303.
Notes
The authors declare no competing financial interests.
(
29) Clark, E. S.; Templeton, D. H.; MacGillavry, C. H. Acta Crys‐
tallographica 1958, 11, 284‐288.
30) Rao, B. V.; Manmode, S.; Hotha, S. J. Org. Chem. 2015, 80,
(
ACKNOWLEDGMENT
1499‐1505.
We are grateful to the University of Konstanz for support of
this work.
(31) Hotha, S.; Kashyap, S. J. Am. Chem. Soc. 2006, 128, 9620‐9621.
(32) Li, Y.; Yang, X.; Liu, Y.; Zhu, C.; Yang, Y.; Yu, B. Chem. Eur. J.
2
(
1
(
010, 16, 1871‐1882.
33) Tang, Y.; Li, J.; Zhu, Y.; Li, Y.; Yu, B. J. Am. Chem. Soc. 2013,
35, 18396‐18405.
34) Yu, B.; Sun, J.; Yang, X. Acc. Chem. Res. 2012, 45, 1227‐1236.
REFERENCES
(1) Schmidt, R. R. Angew. Chem. Int. Ed. Engl. 1986, 25, 212‐235.
(2) Handbook of Chemical Glycosylation: Advances in Stereoselec‐
tivity and Therapeutic Relevance; Demchenko, A. V., Ed.; Wiley‐
VCH: Weinheim, 2008.
(35) Adhikari, S.; Baryal, K. N.; Zhu, D.; Li, X.; Zhu, J. ACS Catal.
2013, 3, 57‐60.
(36) Roy, R.; Palanivel, A. K.; Mallick, A.; Vankar, Y. D. Eur. J. Org.
Chem. 2015, 4000‐4005.
(37) Schmidt, R. R.; Toepfer, A. Tetrahedron Lett. 1991, 32, 3353‐
3356.
(38) The importance of the inhomogeneity of solutions on glyco‐
sidation results has been recently also discussed by: Kononov, L.
O. RSC Adv. 2015, 5, 46718‐46734.
(39) Wabnitz, T. C.; Yu, J.‐Q.; Spencer, J. B. Chem. Eur. J. 2004, 10,
484‐493.
(40)Oshima, K.; Aoyama, Y. J. Am. Chem. Soc. 1999, 121, 2315‐2316.
(41) Gouliaras, C.; Lee, D.; Chan, L.; Taylor, M. S. J. Am. Chem. Soc.
2011, 133, 13926‐13929.
(42) Gimeno, M. C.; Laguna, A. Chem. Rev. 1997, 97, 511‐522.
(43) Modern Gold Catalyzed Synthesis; Hashmi, A. S. K.; Toste, F.
D. Eds.; Wiley‐VCH: Weinheim, 2012.
(
1
(
8
3) Zhu, X.; Schmidt, R. R. Angew. Chem. Int. Ed. 2009, 48, 1900‐
934.
4) Mydock, L. K.; Demchenko, A. V. Org. & Biomol. Chem. 2010,
, 497‐510.
(5) Nigudkar, S. S.; Demchenko, A. V. Chem. Sci. 2015, 6, 2687‐
2704.
(
8
(
1
6) Mensah, E. A.; Nguyen, H. M. J. Am. Chem. Soc. 2009, 131, 8778‐
780.
7) Mensah, E. A.; Yu, F.; Nguyen, H. M. J. Am. Chem. Soc. 2010,
32, 14288‐14302.
8) Yu, F.; Nguyen, H. M. J. Org. Chem. 2012, 77, 7330‐7343.
(
(9) Yu, F.; McConnell, M. S.; Nguyen, H. M. Org. Lett. 2015, 17,
2018‐2021.
(
4
10) McConnell, M. S.; Yu, F.; Nguyen, H. M. Chem. Commun. 2013,
9, 4313‐4315.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 8
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
SYNOPSIS TOC
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Insert Table of Contents artwork here
8
ACS Paragon Plus Environment