Mechanistic insights into the gold(I)-Catalyzed activation of glycosyl Ortho -Alkynylbenzoates for Glycosidation

Article abstract of DOI:10.1021/ja4064316

Anomerization, which involves cleavage and formation of the anomeric C-O bond, is of fundamental importance in the carbohydrate chemistry. Herein, the unexpected gold(I)-catalyzed anomerization of glycosyl ortho-alkynylbenzoates has been studied in detail. Especially, crossover experiments in the presence of an exogenous isochromen-4-yl gold(I) complex confirm that the anomerization proceeds via the exocleavage mechanism, involving (surprisingly) the addition of the isochromen-4-yl gold(I) complex onto a sugar oxocarbenium (or dioxolenium) and an elimination of LAu⁺ from the vinyl gold(I) complex. The inhibitory effect of the exogenous isochromen-4-yl gold(I) complex when in stoichiometric amount on the anomerization has guided us to disclose an isochromen-4-yl gem-gold(I) complex, which is inactive in catalysis but in equilibrium with the monogold(I) complex and the LAu⁺ catalyst. The proposed key intermediate in the anomerization, a transient glycosyloxypyrylium species, is successfully trapped via a cycloaddition reaction with n-butyl vinyl ether as a dienophile. S_N2-like substitution of the initially formed glycosyloxypyrylium intermediate has then been achieved to a large extent via charging with acceptors in an excess amount to lead to the corresponding glycosides in a stereoselective manner.

Full text of DOI:10.1021/ja4064316

Article
pubs.acs.org/JACS
Mechanistic Insights into the Gold(I)-Catalyzed Activation of Glycosyl
ortho-Alkynylbenzoates for Glycosidation
Yu Tang,†,§ Jiakun Li,‡,§ Yugen Zhu,† Yao Li,† and Biao Yu*,†
†State Key Laboratory of Bioorganic and Natural Products Chemistry, Shanghai Institute of Organic Chemistry, Chinese Academy of
Sciences, 345 Lingling Road, Shanghai 200032, China
‡Department of Chemistry, University of Science and Technology of China, 96 Jinzhai Road, Hefei, Anhui 230026, China
S
* Supporting Information
ABSTRACT: Anomerization, which involves cleavage and formation of
the anomeric C−O bond, is of fundamental importance in the
carbohydrate chemistry. Herein, the unexpected gold(I)-catalyzed
anomerization of glycosyl ortho-alkynylbenzoates has been studied in
detail. Especially, crossover experiments in the presence of an exogenous
isochromen-4-yl gold(I) complex confirm that the anomerization
proceeds via the exocleavage mechanism, involving (surprisingly) the
addition of the isochromen-4-yl gold(I) complex onto a sugar
oxocarbenium (or dioxolenium) and an elimination of LAu+ from the
vinyl gold(I) complex. The inhibitory effect of the exogenous
isochromen-4-yl gold(I) complex when in stoichiometric amount on
the anomerization has guided us to disclose an isochromen-4-yl gem-gold(I) complex, which is inactive in catalysis but in
equilibrium with the monogold(I) complex and the LAu+ catalyst. The proposed key intermediate in the anomerization, a
transient glycosyloxypyrylium species, is successfully trapped via a cycloaddition reaction with n-butyl vinyl ether as a dienophile.
SN2-like substitution of the initially formed glycosyloxypyrylium intermediate has then been achieved to a large extent via
charging with acceptors in an excess amount to lead to the corresponding glycosides in a stereoselective manner.
nucleophilic addition reaction with regeneration of the gold(I)
catalyst (step 3).9 The three-center two-electron gem-diaurated
complex C has recently been characterized experimen-
INTRODUCTION
■
Homogenous gold(I)-catalyzed organic reactions have attracted
tremendous attention in recent years.1 Most of these
transformations are initiated by the addition of heteroatom
nucleophiles onto LAu+ activated C−C π-bonds. A generic
mechanism for the gold(I)-catalyzed nucleophilic addition onto
alkynes is outlined in Figure 1. In that, the coordination of the
tally.8,10−19 Gagne,
́
Furstner, and co-workers proposed that
̈
the formation of the gem-diaurated species might compete with
protodeauration and hence have an impact on the catalytic
efficiency.10,11 Nevertheless, in the gold(I)-catalyzed intra-
molecular allene hydroalkoxylation reaction, the bis(gold)vinyl
species was found to be an off-cycle intermediate.8,17 The
occurrence of the gem-diaurated species and their roles in many
other gold(I)-catalyzed transformations remains to be disclo-
sed.19b,20
Recently, we developed a new glycosylation protocol with
glycosyl ortho-alkynylbenzoates as donors and a gold(I)
complex as catalyst (Figure 2).21 The activation mechanism
via an gold(I)-catalyzed intramolecular nucleophilic addition of
alkynes is unprecedented in the glycosylation reactions.22 Due
to the avoidance of strong acidic, nucleophilic, or electrophilic
species (compared to the classical glycosylation reactions), this
new protocol shows a broad applicability in the synthesis of
glycans and glycoconjugates.21e Especially, substrates vulnerable
to acidic conditions have been glycosylated effectively with this
method.23 This new reaction also opens a new window to look
into the mechanisms of both the gold(I) catalysis and the
Figure 1. A generic mechanism for the gold(I)-catalyzed nucleophilic
addition onto alkynes.
gold(I) cation to C−C triple bond is a reversible process (step
1), leading to the π-complex A as a transient intermediate.2
Addition of a nucleophile onto the π-complex provides
vinylgold(I) compound B. This key intermediate and the like
have been isolated and characterized from a number of the
gold(I)-catalyzed transformations since 2008.3,4 In a few of the
reactions the formation of the vinylgold(I) B (step 2) is also
found to be reversible.5−8 Protodeauration of B completes the
Received: June 25, 2013
Published: November 19, 2013
© 2013 American Chemical Society
18396
dx.doi.org/10.1021/ja4064316 | J. Am. Chem. Soc. 2013, 135, 18396−18405

Journal of the American Chemical Society
Article
Figure 2. Gold(I)-catalyzed glycosylation reaction with glycosyl ortho-alkynylbenzoates as donors, the present findings, and the alternative
endocleavage pathway for anomerization.
Figure 3. Major transformations of perbenzoyl glucopyranosyl ortho-hexynylbenzoates 1β and 1α in the presence of Ph3PAuOTf.
glycosylation reaction. In this respect, we have disclosed that
isochromen-4-yl gold(I) compound B1 is a stable intermediate
requiring H+ to regenerate the LAu+ catalyst,24 and the gold(I)
precatalysts, such as Ph3PAuOTf, undergo hydration easily to
form a series of the gold(I) oxo species.25a We have also
disclosed the first experimental evidence for the remote
participation of the 4-O-acyl group on the glycosidation of
glucopyranosyl donors26a and a concentration effect on the
remote participation of the 3-O-acyl group in the N-
glycosidation of 2-deoxyfuranosyl donors.26b
During experimentation with the gold(I)-catalyzed glyco-
sylation, we observed frequently the occurrence of anomeriza-
tion of the glycosyl ortho-alkynylbenzoates. The anomerization
of glycosides is well-known,27,28 which proceeds via two
alternative pathways with only a few exceptions.29 In the
presence of a protic or Lewis acid, glycosides undergo either a
endo- or exocyclic cleavage pathway.27,28 Either pathway if
catalyzed by a gold(I) cation would be unusual, with the former
(H↔I) involving (unlikely) an activation of the ring oxygen
(instead of the alkyne) by gold(I) cation and the latter an
nucleophilic addition of the isocoumarin (i.e., E+B1→D) and
an elimination of the vinylgold(I) complex to alkyne (D→A1).
This puzzle prompts us to make an intensive investigation of
the gold(I)-catalyzed anomerization of glycosyl ortho-alkynyl-
benzoates. Herein we report the interesting findings.
RESULTS AND DISCUSSION
■
Transformation of Perbenzoyl Glucopyranosyl ortho-
Hexynylbenzoates (1β/1α) in the Presence of
Ph3PAuOTf. A commonly used donor, perbenzoyl glucopyr-
anosyl ortho-hexynylbenzoate (1β/1α)21 was first examined for
its transformations under the typical gold(I)-catalyzed glyco-
sylation reaction conditions (0.05 equiv Ph3PAuOTf, CH2Cl2,
rt) in the absence of an acceptor (Figure 3). Starting with either
1β or 1α, a mixture of 1α/1β resulted, in addition, lactol 3,
glycal 4, and isocoumarin 5 were isolated. NMR monitoring of
the reaction processes (0.015 equiv Ph3PAuOTf, CDCl3, 25
°C) revealed that the hydrolysis product 3 was formed within 5
min; the anomerization reached at a equilibrium (1α/1β =
∼13:1) at about 1 h; the elimination reaction (leading to 4)
completed slowly in ∼6 h; and isocoumarin 5 was formed in a
similar rate as that of the formation of 4 and was obtained in
nearly quantitative yield at last (Figure S4). The yield of the
hydrolysis product (3) was dependent on the amount of
moisture occurring in the reaction; exclusion of moisture (with
freshly dried molecular sieves) could avoid largely the
hydrolysis.
An HPLC approach was then set up to analyze accurately the
processes of these gold(I)-catalyzed transformations. Thus, a
small amount of the reaction mixture (∼10 μL) was taken by
syringe into a solution of acetonitrile containing (p-MeOPh)3P
18397
dx.doi.org/10.1021/ja4064316 | J. Am. Chem. Soc. 2013, 135, 18396−18405

Journal of the American Chemical Society
Article
(for quenching the reaction)30 at intervals, and the resulting
solution was directly subjected to HPLC analysis.31 The yield
and relative ratio of 1β and 1α together with the yields of glycal
4 and isocoumarin 5 as a function of time could then be
recorded. To examine factors (including the gold(I) catalyst
and its amount and the reaction solvent and temperature) that
affect the reaction rate and equilibrium position, reactions (all
started with 1β at 0.10 mmol scale) under varied conditions
were analyzed (Figure 4).
shift of the equilibrium position from 1α/1β = 24:1 to 16.5:1
(Figure 4A, line d). When the solvent was changed from
CH2Cl2 to toluene, the anomerization also speeded up
considerably, reaching the equilibrium at ∼1.5 h with the
position of the equilibrium being shifted to 1α/1β = 9.5:1
(Figure 4A, line e). These results show clearly that the rate of
the anomerization is sensitive to all the reaction parameters,
while the position of the equilibrium is only affected by the
reaction solvent and temperature but not by the nature of the
gold(I) catalyst and its loading, indicating a thermodynamically
controlled process for the anomerization.
Surprisingly, when replacing Ph3PAuOTf with Ph3PAuNTf2
as the catalyst, the anomerization of 1β (under otherwise
identical conditions) was found to be much slower, with the
ratio of 1α/1β reaching only 1:4.9 after 10 h (Figure 4A, line f).
Additionally, the elimination product (4) was not detected. The
dramatic difference in the catalytic property of Ph3PAuOTf and
Ph3PAuNTf2 in a few of other reactions has been disclosed.32a
This might be attributed to the fact that −NTf2 is a much
−
stronger coordinating anion than OTf so that dissociation of
Ph3PAuNTf2 is more difficult than that of Ph3PAuOTf to
provide the catalytic species [Ph3PAu]+.25,32b The conjugated
−
−
acids of OTf and NTf2 generated in situ might also play a
role in the reaction33,34 and cause the disparate rates of the
present anomerization (HOTf is a much stronger acid than
HNTf2).35 In fact, in the presence of 2,6-di-tert-butylpyridine
(0.15 equiv) as a H+ scavenger, the anomerization (of 1β)
under Ph3PAuOTf (0.03 equiv) still proceeded much faster
than under Ph3PAuNTf2, nevertheless, both came to an early
stop with all the Au(I) rested at isochromen-4-yl gold(I) 2
(Figure S3). Recharging a second portion of Ph3PAuOTf
brought up a second round of anomerization. The stop of
anomerization is because of depletion of the oxocarbenium E1
(which undergoes decomposition to give glycal 4) and the
Au(I) catalyst (which rests as vinyl complex 2). Without the
base, the H+ generated from the decomposition of E1 would
release the LAu+ catalyst from vinyl complex 2 and thus ensure
anomerization of the remaining 1β to completion.
Another reason accounting for the disparate rate of
anomerization catalyzed by Ph3PAuOTf and Ph3PAuNTf2
could be a disparate involvement of −OTf and −NTf2 in
association with the oxocarbenium E1 and a reactivity disparity
of the resultant complexes (i.e., contact ion pairs, solvent-
separated ion pairs, and covalent species) for anomeriza-
tion.36,37 It should also be noted that a sugar dioxolenium
intermediate (not shown) developed from E1 (via participation
of the 2-O-benzoyl group) was believed to prevail over E1 in
the glycosylation reaction.38 The involvement of the
dioxolenium intermediate in the present anomerization (to
afford the β anomer) seemed minimal in the presence of −OTf.
However, this pathway could be competitive in the presence of
−NTf2, leading to the apparent slow rate of anomerization.
Thus, glycosyl ortho-alkynylbenzoates devoid of a 2-O-
participating group were examined next.
Comparison on the Anomerization of ‘Armed’ and
‘Disarmed’ Glycosyl ortho-Alkynylbenzoates Catalyzed
by Ph3PAuOTf/Ph3PAuNTf2. Protecting groups on the
glycosyl donors have a big impact on their reactivity in
glycosidation. Thus, a donor, ‘armed’ with electron-donating
protecting groups, could be activated for glycosidation
selectively in the presence of another donor bearing the same
leaving group but ‘disarmed’ with electron-withdrawing
protecting groups.39 Such a disparity of donor reactivity was
Figure 4. Kinetics of the transformations of glycosyl ortho-
hexynylbenzoate 1β in the presence of R3PAuOTf (or Ph3PAuNTf2)
under varied conditions. (A) The percentage of 1α/1α + 1β; (B) the
yield of 1α, shown as a function of time. All the experiments were
performed at 0.10 mmol scale in 0.55 mL solvent under conditions in
contrast to Ph3PAuOTf (0.03 equiv)/CH2Cl2/0 °C (line b).
The anomerization proceeded smoothly, reaching equili-
brium with 1α/1β = 24:1 at ∼2.5 h, under the action of
Ph3PAuOTf (0.03 equiv) in CH2Cl2 at 0 °C (Figure 4A, line
b). As the percentage of 1α increased, the rate of the
anomerization decreased gradually. It was noted that the yield
of 1α decreased gradually after the equilibrium was reached
(Figure 4B, line b), in accordance with the increased yield of
the elimination product (Figure S2). When the loading of
Ph3PAuOTf was decreased to 0.015 equiv, the anomerization
reached equilibrium at ∼6.0 h, with the equilibrium position
being unchanged (Figure 4A, line a). Switching the ligand in
Ph3PAuOTf to the sterically hindered (o-Tol)3P led to a much
slower rate for the anomerization (Figure 4A, line c),
nevertheless, the position of the equilibrium remained
unchanged. Raising the reaction temperature from 0 to 23 °C
resulted in a big increase of the anomerization rate and a slight
18398
dx.doi.org/10.1021/ja4064316 | J. Am. Chem. Soc. 2013, 135, 18396−18405

Journal of the American Chemical Society
Article
examined in the present gold(I)-catalyzed anomerization,
employing 2,3,4,6-tetra-O-benzyl-D-glucopyranosyl ortho-hexy-
nylbenzoate (6) as a ‘armed’ donor and 3,4,6-tri-O-acetyl-2-
azido-2-deoxy-D-glucopyranosyl ortho-hexynylbenzoate (7) as a
‘disarmed’ counterpart, which are devoid of the possibility of
neighboring group participation (cf., 1). The anomerization
processes starting from 6β or 7β were monitored by HPLC (as
previously described for donor 1β) comparatively under
identical conditions (0.10 mmol donor in 0.55 mL CH2Cl2,
0.03 equiv Ph3PAuOTf/Ph3PAuNTf2, 0 °C) (Figure 5). The
‘armed’ donor 6β underwent anomerization extremely fast,
reaching the equilibrium at only ∼3.5 min with 6α/6β = 11.8:1.
The ‘disarmed’ donor 7β (so as the ‘disarmed’ donor 1β)
underwent anomerization in a much slower rate, with the
equilibrium being reached at ∼3 h (7α/7β = 8.1:1). The yields
of 7α increased gradually before the equilibrium and remained
unchanged afterward. This indicates that donor 7 is resistant to
elimination under catalytic amount of Ph3PAuOTf.
Compared to the reaction with Ph3PAuOTf as catalyst, the
anomerization of the ‘armed’ donor 6β in the presence of
Ph3PAuNTf2 under otherwise identical conditions was found to
proceed at a much slower rate, reaching equilibrium after ∼30
min, nevertheless, the position of the equilibrium remained
essentially unchanged (6α/6β = 11.3:1; Figure 5). In this case,
the possibility of neighboring participation (as in the previous
anomerization of 1β) is ruled out, therefore, we could conclude
confidently that Ph3PAuOTf is indeed a stronger catalyst than
Ph3PAuNTf2 for the present anomerization reaction. In
addition, in the presence of Ph3PAuOTf the yields of 6α
decreased sharply after the equilibrium due to formation of the
elimination product, in contrast, 6α stayed nearly intact in the
presence of Ph3PAuNTf2.
Anomerization Involves a Reversible C−Au Bond
Formation/Cleavage. The mechanism of the present gold-
(I)-catalyzed anomerization was puzzling. If it proceeded (and
most likely it did) via an exocleavage pathway (Figure 2), a
nucleophilic addition of the isochromen-4-yl gold(I) complex
B124,40 onto sugar oxocarbenium E as well as an elimination of
the LAu+ from vinyl gold(I) intermediate D (to give back
alkyne A1) should take place. To prove this process,
anomerization in the presence of an exogenous isochromen-
4-yl gold(I) complex was examined (Figure 6). The
corresponding scrambling glycosyl ortho-alkynylbenzoates
were constantly isolated, confirming therefore unambiguously
this mechanistic proposal.
Unexpectedly, the anomerization in the presence of an
exogenous isochromen-4-yl gold(I) complex (so that the
concentration of one reactant is greatly increased) was found
to be not faster but much slower compared to the previous
reaction without the additional isochromen-4-yl gold(I)
complex. Some comparable results are depicted in Figure 6,
in that all the reactions were carried out under similar
conditions (30 mg donor in 0.8 mL CH2Cl2, 4 Å MS, 0 °C),
and the yields and α/β ratio were determined accurately by
HPLC analysis.31 A 0.1 equiv of Ph3PAuOTf/Ph3PAuNTf2 was
used instead of the 0.03 equiv in the previous anomerization
reactions to enable the present anomerization (in the presence
of an equal equivalent of the isochromen-4-yl-gold(I) derivative
2)24 proceed in an appreciable rate.
Thus, treatment of 3,4,6-tri-O-acetyl-2-azido-2-deoxy-β-D-
glucopyranosyl ortho-cyclopropylethynylbenzoate (8β) and
gold(I) complex 2 with 0.1 equiv Ph3PAuOTf for 2 h led to
the scrambling product 7 in 10% yield (α/β = 1:1.8), with 8β
remaining largely intact (80%, α/β = 1:110). Replacing
Ph3PAuOTf with Ph3PAuNTf2, 7 was also indentified, although
in only 2% yield (α/β = 2.5:1); the yield of 7 could be
increased to 15% (α/β = 4.7:1) when the reaction was carried
out at 26 °C for 4 h.31 The scrambling product 7 was not
detected when the gold(I) complex 2 was replaced with
isocoumarin 5. In the absence of Ph3PAuOTf or Ph3PAuNTf2,
no reaction took place. These results support strongly the
proposed exocleavage pathway. It was noted that the
anomerization was greatly inhibited by the additional vinyl
gold(I) complex 2, leading to the large recovery of the starting
β-anomer. In comparison, in the absence of 2, the
anomerization of 8β reached at α/β = 2.0:1 (cf., α/β = 1:110
in the presence of 2) with Ph3PAuOTf and at α/β = 1:54.6 (cf.,
Figure 5. Anomerization of glycosyl ortho-alkynylbenzoates 6 and 7
(as well as 1) catalyzed by Ph3PAuOTf/Ph3PAuNTf2. (A) The
percentage of the α-anomers; (B) the yield of the α-anomers, shown as
a function of time.
18399
dx.doi.org/10.1021/ja4064316 | J. Am. Chem. Soc. 2013, 135, 18396−18405

Journal of the American Chemical Society
Article
Figure 6. The gold(I)-catalyzed anomerization of glycosyl ortho-cyclopropylethynylbenzoates (8β-10β) in the presence of an exogenous
isochromen-4-yl gold(I) complex 2. The hydrolysis product turned out to be the major byproduct in those cases where the recovery yield was low.
α/β = 1:165.7 in the presence of 2) with Ph3PAuNTf2 as the
catalyst.
catalyzed process. Thus, a solution of the 2-azido-glucosyl
ortho-hexynylbenzoate 7α, which proceeded anomerization at a
slow rate without formation of glycal in the previous
experiments, was monitored by NMR measurement in the
presence of 1.1 equiv of Ph3PAuNTf2 in CDCl3 at rt. 31P NMR
revealed formation of a new gold(I) species with a small singlet
at 37.1 ppm, which disappeared after consumption of the
substrate (Figure S6).41,42 This new species corresponded to
the appearance of a triplet at 3.05 ppm in the 1H NMR spectra.
The same species was also detected during the anomerization
of 7β under similar conditions (Figure S5). MALDI/TOF MS
analysis of the reaction mixture revealed a weak peak at m/z =
1119.2057, in accordance to the gem-diaurated complex 11.
The putative gem-diaurated complex 11 was then synthesized
and fully characterized (Figure 7). Thus, dissolving an equal
molar mixture of complex 2 and Ph3PAuNTf2 in CDCl3
resulted in a nearly quantitative formation of the gem-diaurated
The anomerization reaction of the ‘armed’ 9β proceeded
rather smoothly in the presence of 1.0 equiv isochromen-4-yl-
gold(I) complex 2 catalyzed by either Ph3PAuOTf or
Ph3PAuNTf2. The scrambling product 6 was formed in high
yield in 0.5 h and predominantly in its α anomer (51%, α/β =
6.5:1 with Ph3PAuOTf and 60%, α/β = 14.6:1 with
Ph3PAuNTf2), with the recovery 9 also existed predominantly
in its α anomer (α/β = 7.3:1 with Ph3PAuOTf and α/β = 2.3:1
with Ph3PAuNTf2).
The anomerization reaction of the ‘disarmed’ 10β in the
presence of 1.0 equiv gold(I) complex 2 led to 37% yield of the
scrambling product 1 (α/β = 1:21.7) under the action of
Ph3PAuOTf and to 28% yield of 1 (α/β = 1:99) under
Ph3PAuNTf2 in 1 h. Interestingly, both the scrambling product
1 and the recovery 10 were predominantly in their β anomers;
in fact, with Ph3PAuNTf2 as the catalyst, 10α was not detected
at all. These results imply that the isochromen-4-yl gold(I)
complexes add onto the sugar dioxolenium intermediate
predominantly rather than onto the sugar oxocarbenium E.
Because of the overwhelming presence of the exogenous
isochromen-4-yl gold(I) complex 2, the anomerization of the
starting β-anomer (requiring addition with the transient
endogenous isochromen-4-yl gold(I) complex) becomes even
more negligible. Additionally, the reaction pathway via the
dioxolenium intermediate is even more favorable under the
catalysis of Ph3PAuNTf2 than under Ph3PAuOTf.
1
complex 11, as shown by H NMR measurement. After many
attempts, single crystals of complex 11 suitable for X-ray
diffraction analysis were obtained by slow diffusion of n-hexane
into a CH2Cl2-toluene solution of 11 at 0 °C.43
The distance between the two gold centers in complex 11
[2.7315(9) Å] shows a strong aurophilic interaction, with the
Au1−C1−Au2 angle of 78.6(5)°, consistent with those of the
previously reported gem-diaurated complexes.42 A comparison
between the crystal structures of gem-diaurated complex 11 and
its monoaurated precursor 224 provides additional structural
information. The Au−C distances of 2.181(15) and 2.131(15)
Å in complex 11 are slightly longer than that found in the
monoaurated precursor 2, at 2.068(7) Å. Remarkably, the C−
Au−P angles in complex 11 are not equal, with C1−Au1−P1
170.7(4)° and C1−Au2−P2 177.5(4)°, smaller than in the
linear structure (180°), whereas in complex 2, the angle is
Characterization of a gem-Diaurated Species (11). The
inhibitory effect of the additional isochromen-4-yl gold(I)
complex (i.e., 2) on the gold(I)-catalyzed anomerization
reaction was unexpected. This promoted us to investigate the
additional role of this key intermediate in the present gold(I)-
18400
dx.doi.org/10.1021/ja4064316 | J. Am. Chem. Soc. 2013, 135, 18396−18405

Journal of the American Chemical Society
Article
In solution (CDCl3), the 31P NMR spectra of complex 11
showed a broad singlet at 37.1 ppm, indicating the occurrence
1
of a dynamic equilibrium (vide infra). The H NMR signals of
the isocoumarin ring in 11 shifted downfield compared to those
in complex 2. The allylic H10 showed a characteristic triplet at
3.05 ppm, which is downfielded by 0.23 ppm from its position
in vinyl gold(I) precursor 2 (2.82 ppm), whereas the methyl
protons H13 shifted upfield from 0.89 ppm (in complex 2) to
0.76 ppm. Similar trends were observed in the 13C NMR
spectra, where the allylic carbon C10 shifted downfield from
38.17 ppm (in complex 2) to 41.21 ppm, whereas the carbonyl
carbon C8 and the methyl carbon C13 shifted upfield from
165.13 and 14.10 ppm (in complex 2) to 160.77 and 13.75
ppm, respectively.
The dynamics of complex 11 in solution was then
investigated by NMR spectroscopy. Experimentally, portions
of Ph3PAuNTf2 were added gradually to a CDCl3 solution of
Figure 7. Preparation of gem-diaurated complex 11 and its ORTEP
diagram with 50% probability ellipsoids (hydrogen atoms are omitted
for clarity). Key bond lengths (Å) and angles (°): Au1−Au2
2.7315(9), Au1−C1 2.181(15), Au2−C1 2.131(15), Au1−P1
2.264(4), Au2−P2 2.277(4), C1−C9 1.37(3), C9−C10 1.51(4),
O1−C9 1.36(3), O1−C8 1.41(3), O2−C8 1.19(2), C1−Au1−P1
170.7(4), C1−Au2−P2 177.5(4), Au2−C1−Au1 78.6(5), C9−C1−
C2 114.5(17).
1
complex 2, and the H and 31P NMR spectra were recorded
after each addition. The 31P NMR signal of 2 at 45.1 ppm
decayed, while that of 11 at 37.1 ppm enhanced along with
each addition of Ph3PAuNTf2. The signal of 2 disappeared after
more than 1 equiv of Ph3PAuNTf2 was added, while the signal
of Ph3PAuNTf2 at 30.7 ppm started to appear. Coalesced
1
signals were observed in the H NMR spectra before <1 equiv
176.0(2)°. The diauration also results in a change of bond
lengths in the isocoumarin ring, with a lengthening of C1−C9
and O1−C8 distances from 1.351(13) to 1.37(3) Å and
1.366(15) to 1.41(3) Å, and a shortening of O1−C9 and O2−
C8 distances from 1.409(11) to 1.36(3) Å and 1.222(11) to
1.19(2) Å on going from monoaurated complex 2 to gem-
diaurated complex 11. This observation indicates clearly a
partial charge delocalization over the isocoumarin ring in gem-
diaurated complex 11, which contains an oxygen atom at the
adjacent position, analogous to the complexes reported by
of Ph3PAuNTf2 was added, in that the allylic H10 exhibited a
broad singlet between the H10 signal in complex 2 (2.82 ppm)
and in complex 11 (3.05 ppm, Figure S9A). In the 13C NMR
spectra, each carbon of the n-butyl group showed only one
signal in the presence of substoichiometric amount of
Ph3PAuNTf2, especially the allylic C10 and the adjacent C11
showed board yet weak signals. These results prove again the
existence of a dynamic equilibrium of the gem-diaurated
complex and its monoaurated precursor in solution (Figure
S9B), that has been described by Nesmeyanov et al.,44
Schmidbaur and Furstner et al.11,19,42d The extent of charge
́
Schmidbaur et al.,45 and recently Gagne et al.10 Interestingly,
̈
delocalization in complex 11, however, is intermediate between
addition of more than 1 equiv of Ph3PAuNTf2 into complex 2
led to a further slightly downfield shift of the H10 signal from
3.050 ppm (at 1.0 equiv) to 3.063 ppm (at 1.91 equiv). This
result suggests a partial dissociation of complex 11 in solution
(Figure S8).46
Schmidbaur complex 12 and Furstner complex 13, with the
̈
former shows bond length change of about ∼2 pm, while the
latter shows ∼10 pm in bond length change.
Trapping of a Glycosyloxypyrylium Intermediate. The
key intermediate, which is the glycosyloxypyrylium gold(I)
complex D (Figure 2), in the present anomerization reaction
was left undetected in the NMR measurement, owing
conceivably to its short lifetime and low concentration in the
Figure 8. A proposed transformation of the glycosyloxypyrylium intermediate D with a vinyl ether.
18401
dx.doi.org/10.1021/ja4064316 | J. Am. Chem. Soc. 2013, 135, 18396−18405

Journal of the American Chemical Society
Article
reaction mixture. To prove the occurrence of this transient
species, we sought to trap it via a chemical transformation.
Iwasawa et al. have reported a Pt(II)-catalyzed condensation of
ortho-alkynylbenzoates with vinyl ethers to provide 1-acyl-4-
alkyoxynaphthalenes (K), in that a [3 + 2]-cycloaddition of the
alkyloxypyrylium salts (analogous to D) with vinyl ethers
(followed by 1,2-alkyl-migration and subsequent eliminations)
was involved (Figure 8).47,48 However, a similar reaction of a
glycosyl ortho-alkynylbenzoate with a vinyl ether would be
complicated by easy degradation of the glycosyloxypyrylium
intermediate D (Figures 2 and 3), including the elimination and
hydrolysis reactions (to give the corresponding glycal and
lactol, respectively).
17 and 18 representing good and poor nucleophiles,
respectively, we examined the stereochemistry outcomes of
their glycosylation reactions (Figure 10).
Thus, the 2-azido-glucopyranosyl ortho-hexynylbenzoate 7,
which was previously found to be resistant to elimination, was
selected as the substrate and 4 Å MS (300% w/w) was added in
the reaction to prevent hydrolysis by exclusion of moisture
(Figure 9). Experimentally, a toluene solution of the pure
Figure 9. Trapping of the glycosyloxypyrylium intermediate D1 via
cycloaddition with vinyl ether 14 and cascade transformations.
anomer 7β and vinyl ether 14 (2 equiv) was dried by 4 Å MS
(300% w/w) for 1 h, and then Ph3PAuNTf2 (0.1 equiv) was
added. As shown on TLC, the anomerization proceeded as
under the previous conditions, however, the desired naph-
thalene glycoside 15 was not detected within 4 h. A second
portion of vinyl ether (8 equiv) was then added, gratifyingly,
the desired product 15 was then detected and finally isolated in
12% yield (α/β = 5.6:1, with 7 being recovered in 42% yield
with α/β = 9.3:1). Further experiments showed that the yield of
15 was dependent on the level of anomerization at the time of
the addition of vinyl ether 14. Thus, addition of vinyl ether after
0.6 h at the time the 7α/7β ratio reached 2.7:1 led to only 2%
of 15 (α/β = 5:1) with 40% recovery of 7 (α/β = 5.3:1). In
comparison, addition of the vinyl ether after 4 h when the
anomerization reached equilibrium resulted in 20% of 15 (α/β
= 10:1) with 27% recovery of 7 (α/β = 11.4:1). By addition of
vinyl ether 14 (40 equiv) into an equilibrated anomerization
mixture of 7 in four equal portions at intervals of 1 h, we
managed to isolate 15 in 34% yield (α/β = 13:1). These results
confirm the occurrence of the glycosyloxypyrylium intermedi-
ate (i.e., D) in the anomerization process and indicate that the
α-glycosyloxypyrylium intermediate occur (and react with vinyl
ether) favorably compared to its β-counterpart.
SN2-Like Glycosylation under Forced Conditions. Now
that the activation of a glycosyl ortho-alkynylbenzoate with
LAu+ does lead to the glycosyloxypyrylium intermediate (i.e.,
D), a stereoselective glycosidation via SN2-like substitution of
this intermediate should be possible.22,49 Nevertheless, it is
required that the glycosidation takes place on the glycosylox-
ypyrylium intermediate or a contact ion pair before it falls apart
to the solvent-separated oxocarbenium species (i.e., E).22,36,49
With a pair of the easily available glycosyl ortho-hexynylben-
zoate anomers 16β and 16α as donors and two sugar alcohols
Figure 10. SN2-like glycosidation of ortho-alkynylbenzoate donor 16α/
β driven by a large excess (10 equiv) of the acceptors. (a) Isolated
yield. (b) The α/β ratio was determined by H NMR measurement.
1
(c) 0.2 equiv of Ph3PAuNTf2 was used. (d) 44% 16 was recovered.
The glycosidation of 16β with 17 (1.0 equiv) under usual
conditions (0.1 equiv Ph3PAuNTf2, 4 Å MS, CH2Cl2, 0 °C) led
to the coupled disaccharide 19 in good yield and poor
stereoselectivity (α/β = 3.7:1; entry 1). This α/β ratio is
indicative of a SN1 glycosidation (and the present anomeriza-
tion as well via oxocarbenium E) which favors formation of the
α anomer due to the anomeric effect.50 Simply increasing the
amount of acceptor 17 to 10.0 equiv resulted in a remarkable
increase of the α/β ratio of the disaccharide (α/β = 18:1; entry
2).51 It is known that a SN2 reaction is favored at lower reaction
temperature and less polar solvent. Indeed, the condensation of
16β and 17 at −20 °C led to further increase of the α-selectivity
(α/β = 24:1; entry 3), while at 27 °C led to decrease of the α-
selectivity (α/β = 13:1; entry 4); and the reaction with Et2O as
solvent led to a considerable decrease of the α-selectivity (α/β
= 7.9:1; entry 5). Under the fixed reaction conditions (0.1 equiv
Ph3PAuNTf2, 4 Å MS, CH2Cl2, 0 °C), the glycosidation of 16β
with the poorly nucleophilic glucose-4-OH derivative 18 led to
similar results as with 17 as the acceptor. Thus, in the presence
of 1 equiv of 18, the coupled disaccharide 20 was obtained in a
α/β ratio of 3.7:1 (entry 6); and the α/β ratio of 20 was
increased remarkably to a high 7.0:1 when 10.0 equiv of 18 was
charged (entry 7).
The SN2-like glycosidation of 16α to give the corresponding
β-glycoside would be much more difficult, because the
corresponding α-glycosyloxypyrylium is more stable than its
18402
dx.doi.org/10.1021/ja4064316 | J. Am. Chem. Soc. 2013, 135, 18396−18405

Journal of the American Chemical Society
Article
β-counterpart and thus reluctant to undergo substitution. The
condensation of 16α with 17 (1.0 equiv) under usual
conditions led to 19 in poor stereoselectivity (α/β = 1.3:1;
entry 8). This outcome implies that a certain extent of SN2
glycosidation does occur, if compared to the higher α selectivity
(α/β = 3.7:1) resulted from the reaction of 16β under identical
conditions (entry 1). Increasing the amount of acceptor 17 to
10 equiv, the β-selective glycosidation was achieved (α/β =
1:3.7; entry 9). Under more favorable conditions for a SN2
reaction (with 0.2 equiv of the precatalyst Ph3PAuNTf2 in a less
polar solvent CH2Cl2/n-hexane (1:4) and at a lower temper-
ature of −20 °C), the condensation of 16β and 17 furnished
disaccharide 19 in an excellent β-selectivity (α/β = 1:14; entry
10). Under this optimized set of conditions, the glycosidation
of 16α with the poorly nucleophilic sugar alcohol 18, however,
led to the α-product predominantly even in the presence of 10
equiv of the acceptor. Nevertheless, the increase of the amount
of acceptor 18 (from 1.0 to 10.0 equiv) did led to a remarkable
increase of the β-product (from α only to α/β = 3.8:1; entries
11 and 12).
It should be noted that all these intermediates and reversible
steps would be involved in the glycosylation reaction when an
acceptor is added. Therefore, understanding this anomerization
process shall help to find solutions to address the problems in
glycosylation reactions. As a preliminary effort, we attempted at
SN2-like glycosylation via intercepting the initially formed
glycosyloxypyrylium intermediate (D). With a good nucleo-
phile and in an excess amount (10 equiv), high α-selective
glycosylation from the β-glycosyl ortho-alkynylbenzoate or high
β-selective glycosylation from its α-counterpart has been
showcased, although the scope awaits as a subject of future
research.
ASSOCIATED CONTENT
■
S
* Supporting Information
Experimental details and characterization data, and H and 13C
1
NMR spectra of all new compounds. This material is available
free of charge via the Internet at http://pubs.acs.org
AUTHOR INFORMATION
Corresponding Author
byu@mail.sioc.ac.cn
■
CONCLUSION
■
Author Contributions
The gold(I)-catalyzed glycosylation reaction with ortho-
alkynylbenzoates as donors has recently been proven to be a
versatile method for the synthesis of various glycosidic linkages,
including remarkably those vulnerable to acidic conditions.
Herein we studied in details the gold(I)-catalyzed trans-
formation of ortho-alkynylbenzoates in the absence of acceptor,
that is disclosed to be mainly the anomerization. An
exocleavage mechanism, as shown in Figure 2, has been proved
for the anomerization. Thus, LAu+ coordinated to the C−C
triple bond installed in the glycosyl ortho-alkynylbenzoate (A1,
either α or β anomer); the activated triple bond induced an
intramolecular nucleophilic addition of the benzoyl oxygen,
leading to the 1-glycosyloxy-isochromenylium-4-gold(I) com-
plex D (which was proven by tripping via a cycloaddition with a
vinyl ether); the resulting glycosyloxypyrylium fall apart quickly
to give sugar oxocarbenium E and isochromen-4-gold(I)
complex B1 (which has been characterized previously and
prepared readily); significantly, E and B1 underwent SN1-type
nucleophilic addition to provide the glycosyloxypyrylium D as a
mixture of the α and β anomers (that was proven by crossover
experiments with an exogenous B1 congener); and the vinyl
gold(I) complex D then underwent elimination to give the
alkyne derivative A1. Through these reversible steps (glycosyl
ortho-alkynylbenzoate + LAu+ ↔ A1 ↔ D ↔ E + B1),
anomerization reached finally to an equilibrium. In addition, the
isochromen-4-yl gem-gold(I) complex C1 was characterized
and found to be in equilibrium with the vinyl gold(I) complex
B1 (B1 + LAu+ ↔ C1). These two gold(I) complexes were
proved to be inactive species in the catalysis; the gem-gold(I)
complex C1 could release readily the catalytic LAu+ and
monogold(I) B1, while B1 required H+ to release LAu+ and
isocoumarin G.
§These authors contributed equally.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work is financially supported by the Ministry of Sciences
and Technology of China (2012ZX09502-002) and the
National Natural Science Foundation of China (20932009
and 20921091). We thank Dr. Xuebing Leng for X-ray
crystallographic analysis.
REFERENCES
■
(1) For selected reviews, see: (a) Rudolph, M.; Hashmi, A. S. K.
Chem. Soc. Rev. 2012, 41, 2448−2462. (b) Gaillard, S.; Cazin, C. S. J.;
Nolan, S. P. Acc. Chem. Res. 2012, 45, 778−787. (c) Boorman, T. C.;
Larrosa, I. Chem. Soc. Rev. 2011, 40, 1910−1925. (d) Krause, N.;
Winter, C. Chem. Rev. 2011, 111, 1994−2009. (e) Corma, A.; Leyva-
́
Perez, A.; Sabater, M. J. Chem. Rev. 2011, 111, 1657−1712. (f) Bandini,
M. Chem. Soc. Rev. 2011, 40, 1358−1367. (g) Rudolph, M.; Hashmi, A.
S. K. Chem. Commun. 2011, 47, 6536−6544. (h) Furstner, A. Chem.
̈
Soc. Rev. 2009, 38, 3208−3221. (i) Li, Z.; Brouwer, C.; He, C. Chem.
Rev. 2008, 108, 3239−3265. (j) Widenhoefer, R. A. Chem.Eur. J.
2008, 14, 5382−5391. (k) Gorin, D. J.; Sherry, B. D.; Toste, F. D.
Chem. Rev. 2008, 108, 3351−3378. (l) Hashmi, A. S. K.; Rudolph, M.
́ ́
Chem. Soc. Rev. 2008, 37, 1766−1775. (m) Jimenez-Nunez, E.;
̃
Echavarren, A. M. Chem. Rev. 2008, 108, 3326−3350. (n) Arcadi, A.
Chem. Rev. 2008, 108, 3266−3325. (o) Hashmi, A. S. K. Chem. Rev.
2007, 107, 3180−3211. (p) Gorin, D. J.; Toste, F. D. Nature 2007,
446, 395−403. (q) Furstner, A.; Davies, P. W. Angew. Chem., Int. Ed.
2007, 46, 3410−3449. (r) Hashmi, A. S. K.; Hutchings, G. J. Angew.
Chem., Int. Ed. 2006, 45, 7896−7936.
(2) (a) Schmidbaur, H.; Schier, A. Organometallics 2010, 29, 2−23.
(b) Raubenheimer, H. G.; Schmidbaur, H. S. Afr. J. Sci. 2011, 107, 31−
44. (c) Brooner, R. E. M.; Widenhoefer, R. A. Angew. Chem., Int. Ed.
2013, 52, 11714−11724.
(3) (a) Liu, L. -P.; Xu, B.; Mashuta, M. S.; Hammond, G. B. J. Am.
Chem. Soc. 2008, 130, 17642−17643. (b) Liu, L.-P.; Hammond, G. B.
Chem. Asian J. 2009, 4, 1230−1236.
(4) For reviews on the characterized monogold complexes
intermediates, see: (a) Liu, L. P.; Hammond, G. B. Chem. Soc. Rev.
2012, 41, 3129−3139. (b) Hashmi, A. S. K. Angew. Chem., Int. Ed.
2010, 49, 5232−5241.
̈
Additionally, Ph3PAuOTf has been disclosed to be a stronger
precatalyst than Ph3PAuNTf2 in the present anomerization
reaction, and the weaker Ph3PAuNTf2 facilitates the addition of
isochromen-4-yl gold(I) complex B1 onto the sugar
dioxolenium intermediate (if a neighboring participating acyl
group was equipped at the sugar moiety) instead of the sugar
oxocarbenium E, leading to an apparent slow rate of the
anomerization.
18403
dx.doi.org/10.1021/ja4064316 | J. Am. Chem. Soc. 2013, 135, 18396−18405

Journal of the American Chemical Society
Article
(5) Akana, J. A.; Bhattacharyya, K. X.; Muller, P.; Sadighi, J. P. J. Am.
(26) (a) Ma, Y.; Lian, G.; Li, Y.; Yu, B. Chem. Commun. 2011, 47,
7515−7517. (b) Yang, F.; Zhu, Y.; Yu, B. Chem. Commun. 2012, 48,
7097−7099.
̈
Chem. Soc. 2007, 129, 7736−7737.
(6) (a) Shi, Y.; Roth, K. E.; Ramgren, S. D.; Blum, S. A. J. Am. Chem.
Soc. 2009, 131, 18022−18023. (b) Roth, K. E.; Blum, S. A.
Organometallics 2010, 29, 1712−1716.
(27) For selected earlier reports on sugar anomerization, see:
(a) Lemieux, R. U.; Brice, C. Can. J. Chem. 1952, 30, 295−310.
(b) Lemieux, R. U.; Hayami, J. I. Can. J. Chem. 1965, 43, 2162−2173.
(c) Bonner, W. A. J. Am. Chem. Soc. 1951, 73, 2659−2666. (d) Bonner,
W. A. J. Am. Chem. Soc. 1961, 83, 962−965. (e) Capon, B. Chem. Rev.
1969, 69, 407−498.
(7) (a) LaLonde, R. L.; Brenzovich, W. E.; Benitez, D.; Tkatchouk,
E.; Kelley, K.; Goddard, W. A.; Toste, F. D. Chem. Sci. 2010, 1, 226−
233. (b) de Haro, T.; Nevado, C. Angew. Chem., Int. Ed. 2011, 50,
906−910.
(28) For recent reports on sugar anomerization, see: (a) Sharma, I.;
(8) Brown, T. J.; Weber, D.; Gagne,
Chem. Soc. 2012, 134, 9134−9137.
(9) Hashmi, A. S. K. Catal. Today 2007, 122, 211−214.
(10) (a) Weber, D.; Tarselli, M. A.; Gagne, M. R. Angew. Chem., Int.
Ed. 2009, 48, 5733−5736. (b) Weber, D.; Gagne, M. R. Chem. Sci.
́
M. R.; Widenhoefer, R. A. J. Am.
Bohe,
́
L.; Crich, D. Carbohydr. Res. 2012, 357, 126−131. (b) Satoh, H.;
Manabe, S.; Ito, Y.; Luthi, H. P.; Laino, T.; Hutter, J. J. Am. Chem. Soc.
̈
2011, 133, 5610−5619. (c) Vidadala, S. R.; Pimpalpalle, T. M.; Linker,
T.; Hotha, S. Eur. J. Org. Chem. 2011, 2426−2430. (d) Manabe, S.;
Ishii, K.; Satoh, H.; Ito, Y. Tetrahedron 2011, 67, 9966−9974.
(e) Malik, S.; Shah, K. J.; Kartha, K. P. R. Carbohydr. Res. 2010, 345,
867−871. (f) Pilgrim, W.; Murphy, P. V. J. Org. Chem. 2010, 75,
́
́
2013, 4, 335−338.
(11) Seidel, G.; Lehmann, C. W.; Furstner, A. Angew. Chem., Int. Ed.
̈
2010, 49, 8466−8470.
6747−6755. (g) Forsman, J. J.; Warna, J.; Murzin, D. Y.; Leino, R.
̈
̊
(12) Hooper, T. N.; Green, M.; Russell, C. A. Chem. Commun. 2010,
46, 2313−2315.
Carbohydr. Res. 2009, 344, 1102−1109. (h) Olsson, J. D. M.; Eriksson,
L.; Lahmann, M.; Oscarson, S. J. Org. Chem. 2008, 73, 7181−7188.
(29) 2,4-Dinitrophenyl glycosides undergo anomerization in the
presence of a base via a nucleophilic aromatic substitution mechanism.
(a) Lindberg, B. Acta Chem. Scand. 1950, 4, 49−51. (b) Berven, L. A.;
Dolphin, D. H.; Withers, S. G. J. Am. Chem. Soc. 1988, 110, 4864−
4866. (c) Berven, L. A.; Dolphin, D. H.; Withers, S. G. Can. J. Chem.
1990, 68, 1859−1866.
(13) Himmelspach, A.; Finze, M.; Raub, S. Angew. Chem., Int. Ed.
2011, 50, 2628−2631.
́
(14) Weber, D.; Jones, T. D.; Adduci, L. L.; Gagne, M. R. Angew.
Chem., Int. Ed. 2012, 51, 2452−2456.
(15) (a) Hashmi, A. S. K.; Braun, I.; Nosel, P.; Schadlich, J.; Wieteck,
̈
̈
M.; Rudolph, M.; Rominger, F. Angew. Chem., Int. Ed. 2012, 51, 4456−
4460. (b) Hashmi, A. S. K.; Wieteck, M.; Braun, I.; Nosel, P.;
̈
(30) It was found that the transformations were completely inhibited
in the presence of an excess amount of a phosphine ligand (e.g, PPh3),
due to formation of the inactive (Ph3P)2Au+ species.
Jongbloed, L.; Rudolph, M.; Rominger, F. Adv. Synth. Catal. 2012, 354,
555−562. (c) Hashmi, A. S. K.; Braun, I.; Rudolph, M.; Rominger, F.
Organometallics 2012, 31, 644−661.
(16) Heckler, J. E.; Zeller, M.; Hunter, A. D.; Gray, T. G. Angew.
Chem., Int. Ed. 2012, 51, 5924−5928.
(31) See Supporting Information for details.
(32) (a) Wang, D.; Cai, R.; Sharma, S.; Jirak, J.; Thummanapelli, S.
K.; Akhmedov, N. G.; Zhang, H.; Liu, X.; Petersen, J. L.; Shi, X. J. Am.
Chem. Soc. 2012, 134, 9012−9019. (b) Antoniotti, S.; Dalla, V.;
(17) Gom
́ ́
ez-Suarez, A.; Dupuy, S.; Slawin, A. M. Z.; Nolan, S. P.
Angew. Chem., Int. Ed. 2013, 52, 938−942.
(18) Roithova,
Angew. Chem., Int. Ed. 2012, 51, 8378−8382.
(19) (a) Zhdanko, A.; Maier, M. E. Chem.Eur. J. 2013, 19, 3932−
3942. (b) Zhdanko, A.; Maier, M. E. Organometallics 2013, 32, 2000−
2006.
Dunach, E. Angew. Chem., Int. Ed. 2010, 49, 7860−7888.
̃
̌
́ ́ ̌ ́ ́ ̌ ́
J.; Jankova, S.; Jasíkova, L.; Vana, J.; Hybelbauerova, S.
(33) Although the present anomerization does not invoke H+
theoretically, in practice, H+ is generated from the adventitious H2O
and the oxocarbenium intermediates via hydrolysis and elimination (to
give 3 and 4, respectively, Figure 3). Meanwhile, the resultant H+ is
consumed in part via protodeauration (to give isocoumarin 5).
(34) Dang, T. T.; Boeck, F.; Hintermann, L. J. Org. Chem. 2011, 76,
9353−9361.
(20) Gom
8156−8159.
́ ́
ez-Suarez, A.; Nolan, S. P. Angew. Chem., Int. Ed. 2012, 51,
(21) (a) Li, Y.; Yang, Y.; Yu, B. Tetrahedron Lett. 2008, 49, 3604−
3608. (b) Li, Y.; Yang, X.; Liu, Y.; Zhu, C.; Yang, Y.; Yu, B. Chem.
Eur. J. 2010, 16, 1871−1882. (c) Li, Y.; Yu, B. Chem. Commun. 2010,
46, 6060−6062. (d) Yang, Y.; Li, Y.; Yu, B. J. Am. Chem. Soc. 2009,
131, 12076−12077. (e) Yu, B.; Sun, J.; Yang, X. Acc. Chem. Res. 2012,
45, 1227−1236.
(35) (a) Foropoulos, J.; DesMarteau, D. D. Inorg. Chem. 1984, 23,
3720−3723. (b) Mathieu, B.; Ghosez, L. Tetrahedron 2002, 58, 8219−
8226.
́
(36) (a) Bohe, L.; Crich, D. C. R. Chimie 2011, 14, 3−16. (b) Aubry,
S.; Sasaki, K.; Sharma, I.; Crich, D. Top. Curr. Chem. 2011, 301, 141−
188.
(22) For selected recent reviews on the mechanism of glycosylation
reactions, see: (a) Ranade, S. C.; Demchenko, A. V. J. Carbohydr.
Chem. 2013, 32, 1−43. (b) Mydock, L. K.; Demchenko, A. V. Org.
Biomol. Chem. 2010, 8, 497−510. (c) Crich, D. Acc. Chem. Res. 2010,
43, 1144−1153. (d) Walvoort, M. T. C.; Dinkelaar, J.; van den Bos, L.
(37) Numerous glycosyl triflates have been documented,36b while
only one glycosyl triflimide has so far been implied by NMR signals,
see: Yamago, S.; Yamada, T.; Maruyama, T.; Yoshida, J.-i. Angew.
Chem., Int. Ed. 2004, 43, 2145−2148. The SN2 pathway for
glycosidation of the glycosyl triflates has been ruled out, see: Crich,
D.; Chandrasekera, N. S. Angew. Chem., Int. Ed. 2004, 43, 5386−5389.
́
J.; Lodder, G.; Overkleeft, H. S.; Codee, J. D. C.; van der Marel, G. A.
Carbohydr. Res. 2010, 345, 1252−1263. (e) Zhu, X.; Schmidt, R. R.
Angew. Chem., Int. Ed. 2009, 48, 1900−1934. (f) Crich, D. J. Org.
Chem. 2011, 76, 9193−9209.
́
(38) (a) Nukada, T.; Berces, A.; Zgierski, M. Z.; Whitfield, D. M. J.
Am. Chem. Soc. 1998, 120, 13291−13295. (b) Berces, A.; Enright, G.;
Nukada, T.; Whitfield, D. M. J. Am. Chem. Soc. 2001, 123, 5460−5464.
(39) For the concept of ‘armed’/‘disarmed’ donors, see: (a) Mootoo,
D. R.; Konradsson, P.; Udodong, U.; Fraser-Reid, B. J. Am. Chem. Soc.
1988, 110, 5583−5584. (b) Fraser-Reid, B.; Udodong, U.; Wu, Z. F.;
Ottosson, H.; Merritt, J. R.; Rao, C. S.; Roberts, C.; Madsen, R. Synlett
1992, 927−942.
(23) (a) Yu, J.; Sun, J.; Niu, Y.; Li, R.; Liao, J.; Zhang, F.; Yu, B.
Chem. Sci. 2013, 4, 3899−3905. (b) Yu, J.; Sun, J.; Yu, B. Org. Lett.
2012, 14, 4022−4025. (c) Zhang, J.; Shi, H.; Ma, Y.; Yu, B. Chem.
Commun. 2012, 48, 8679−8681. (d) Zhang, Q.; Sun, J.; Zhu, Y.;
Zhang, F.; Yu, B. Angew. Chem., Int. Ed. 2011, 50, 4933−4936. (e) Li,
Y.; Sun, J.; Yu, B. Org. Lett. 2011, 13, 5508−5511. (f) Ma, Y.; Li, Z.;
Shi, H.; Zhang, J.; Yu, B. J. Org. Chem. 2011, 76, 9748−9756. (g) Yang,
W.; Sun, J.; Lu, W.; Li, Y.; Shan, L.; Han, W.; Zhang, W. -D.; Yu, B. J.
Org. Chem. 2010, 75, 6879−6888.
(40) The decomposition of isochromen-4-yl gold(I) complex B1 into
ortho-alkynylbenzoic acid (which could then involve in anomerization)
and LAu+ was not detected.
(41) Under the action of 1.1 equiv of Ph3PAuNTf2, 7 was consumed
within 2 h; the acetyl transfer product 1,3,4,6-tetra-O-acetyl-2-azido-2-
deoxy-α-D-glucose was isolated in 36% yield.
(24) Zhu, Y.; Yu, B. Angew. Chem., Int. Ed. 2011, 50, 8329−8332.
(25) (a) Tang, Y.; Yu, B. RSC Adv. 2012, 2, 12686−12689.
(b) Zhdanko, A.; Strobele, M.; Maier, M. E. Chem.Eur. J. 2012, 18,
̈
(42) For crystal structures of the gem-diaurated complexes, see:
(a) Nesmeyanov, A. N.; Perevalova, E. G.; Grandberg, K. L.;
14732−14744.
18404
dx.doi.org/10.1021/ja4064316 | J. Am. Chem. Soc. 2013, 135, 18396−18405

Journal of the American Chemical Society
Article
Lemenovskii, D. A.; Baukova, T. V.; Afanassova, O. B. J. Organomet.
Chem. 1974, 65, 131−144. (b) Andrianov, V. G.; Struchkov, Y. T.;
Rossinskaya, E. R. J. Struct. Chem. 1974, 65−72. (c) Uson, R.; Laguna,
A.; Fernandez, E. J.; Mendia, A.; Jones, P. G. J. Organomet. Chem.
1988, 350, 129−138. (d) Porter, K. A.; Schier, A.; Schmidbaur, H.
Organometallics 2003, 22, 4922−4927. (e) Osawa, M.; Hoshino, M.;
Hashizume, D. J. Chem. Soc., Dalton Trans. 2008, 2248−2252 see also
ref 10b, 11, 16, 17, and 19.
(43) CCDC 943326 contains the supplementary crystallographic
data for compound 11. The data can be obtained free of charge from
the Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
(44) (a) Nesmeyanov, A. N.; Perevalova, E. G.; Afanasov, A. B.;
Grandberg, K. I. Bull. Acad. Sci. USSR, Div. Chem. Sci. (Engl. Transl.)
1978, 27, 973−976. (b) Nesmeyanov, A. N.; Perevalova, E. G.;
Ovchinnkov, M. V.; Snakin, Y. Y.; Grandberg, K. I. Bull. Acad. Sci.
USSR, Div. Chem. Sci. (Engl. Transl.) 1978, 27, 1695−1697.
(45) (a) Schmidbaur, H.; Inoguchi, Y. Chem. Ber 1980, 113, 1646−
1653. (b) Schmidbaur, H.; Schier, A. Chem. Soc. Rev. 2012, 41, 370−
412.
́
(46) In comparison, Gagne et al. found that addition of more than
one equivalent of AuPPh3NTf2 to a CD2Cl2 solution of (p-MeO-
C6H4)AuPPh3 led to no further change of the chemical shift; see ref
14.
(47) Kusama, H.; Funami, H.; Takaya, J.; Iwasawa, N. Org. Lett. 2004,
6, 605−608.
(48) In relevant condensations of ortho-alkynylbenzaldehye/ketone
with alkyne/alkene to provide naphthyl ketone, a [4 + 2] cycloaddition
mechanism of benzo[c]pyrylium intermidates is proposed, see:
(a) Asao, N.; Takahashi, K.; Lee, S.; Kasahara, T.; Yamamoto, Y. J.
Am. Chem. Soc. 2002, 124, 12650−12651. (b) Asao, N.; Nogami, T.;
Lee, S.; Yamamoto, Y. J. Am. Chem. Soc. 2003, 125, 10921−10925.
(c) Asao, N.; Sato, K.; Menggenbateer, M. N.; Yamamoto, Y. J. Org.
Chem. 2005, 70, 3682−3685. (d) Sato, K.; Asao, N.; Yamamoto, Y. J.
Org. Chem. 2005, 70, 8977−8981. (e) Asao, N.; Sato, K. Org. Lett.
2006, 8, 5361−5363. (f) Asao, N. Synlett 2006, 1645−1656.
(49) For selected references on SN2-like glycosidation, see:
(a) Lemieux, R. U.; Hendriks, K. B.; Stick, R. V.; James, K. J. Am.
Chem. Soc. 1975, 97, 4056−4062. (b) Banait, N. S.; Jencks, W. P. J.
Am. Chem. Soc. 1991, 113, 7951−7958. (c) Callam, C. S.; Gadikota, R.
R.; Krein, D. M.; Lowary, T. L. J. Am. Chem. Soc. 2003, 125, 13112−
13119. (d) El-Badri, M. H.; Willenbring, D.; Tantillo, D. J.; Gervay-
Hague, J. J. Org. Chem. 2007, 72, 4663−4672. (e) Krumper, J. R.;
Salamant, W. A.; Woerpel, K. A. J. Org. Chem. 2009, 74, 8039−8050.
(f) Walvoort, M. T. C.; Lodder, G.; Mazurek, J.; Overkleeft, H. S.;
Codee, J. D. C.; van der Marel, G. A. J. Am. Chem. Soc. 2009, 131,
́
12080−12081. (g) Prevost, M.; St-Jean, O.; Guindon, Y. J. Am. Chem.
Soc. 2010, 132, 12433−12439. (h) Kumar, A.; Kumar, V.; Dere, R. T.;
Schmidt, R. R. Org. Lett. 2011, 13, 3612−3615. (i) Fascione, M. A.;
Kilner, C. A.; Leach, A. G.; Turnbull, W. B. Chem.Eur. J. 2012, 18,
321−333. (j) Gouliaras, C.; Lee, D.; Chan, L.; Taylor, M. S. J. Am.
Chem. Soc. 2011, 133, 13926−13929. (k) Huang, M.; Retailleau, P.;
́
Bohe, L.; Crich, D. J. Am. Chem. Soc. 2012, 134, 14746−14749.
(50) (a) Deslongchamps, P. Pure App. Chem. 1993, 65, 1161−1178.
(b) Juaristi, E.; Cuevas, G. Tetrahedron 1992, 48, 5019−5087.
(c) Kirby, A. J. The Anomeric Effect and Related Stereoelectronic Effects
at Oxygen; Springer-Verlag: Berlin, 1983.
(51) Hanessian, S.; Lou, B. Chem. Rev. 2000, 100, 4443−4463.
18405
dx.doi.org/10.1021/ja4064316 | J. Am. Chem. Soc. 2013, 135, 18396−18405