Borinic Acid Catalyzed Stereo- and Regioselective Couplings of Glycosyl Methanesulfonates

Article abstract of DOI:10.1021/jacs.6b06943

In the presence of a diarylborinic acid catalyst, glycosyl methanesulfonates engage in regio- and stereoselective couplings with partially protected pyranoside and furanoside acceptors. The methanesulfonate donors are prepared in situ from glycosyl hemiacetals, and are coupled under mild, operationally simple conditions (amine base, organoboron catalyst, room temperature). The borinic acid catalyst not only influences site-selectivity via activation of 1,2- or 1,3-diol motifs, but also has a pronounced effect on the stereochemical outcome: 1,2-trans-linked disaccharides are obtained selectively in the absence of neighboring group participation. Reaction progress kinetic analysis was used to obtain insight into the mechanism of glycosylation, both in the presence of catalyst and in its absence, while rates of interconversion of methanesulfonate anomers were determined by NMR exchange spectroscopy (EXSY). Together, the results suggest that although the uncatalyzed and catalyzed reactions give rise to opposite stereochemical outcomes, both proceed by associative mechanisms.

Full text of DOI:10.1021/jacs.6b06943

Article
pubs.acs.org/JACS
Borinic Acid Catalyzed Stereo- and Regioselective Couplings of
Glycosyl Methanesulfonates
Kyan A. D’Angelo and Mark S. Taylor*
Department of Chemistry, University of Toronto, 80 St. George Street, Toronto, Ontario M5S 3H6, Canada
S
* Supporting Information
ABSTRACT: In the presence of a diarylborinic acid catalyst, glycosyl methanesulfonates engage in regio- and stereoselective
couplings with partially protected pyranoside and furanoside acceptors. The methanesulfonate donors are prepared in situ from
glycosyl hemiacetals, and are coupled under mild, operationally simple conditions (amine base, organoboron catalyst, room
temperature). The borinic acid catalyst not only influences site-selectivity via activation of 1,2- or 1,3-diol motifs, but also has a
pronounced effect on the stereochemical outcome: 1,2-trans-linked disaccharides are obtained selectively in the absence of
neighboring group participation. Reaction progress kinetic analysis was used to obtain insight into the mechanism of
glycosylation, both in the presence of catalyst and in its absence, while rates of interconversion of methanesulfonate anomers
were determined by NMR exchange spectroscopy (EXSY). Together, the results suggest that although the uncatalyzed and
catalyzed reactions give rise to opposite stereochemical outcomes, both proceed by associative mechanisms.
In 2011, our group showed that arylborinic acid (Ar2BOH)
derivatives accelerate regioselective glycosylations of acceptors
having 1,2- or 1,3-diol groups (Scheme 1).3a The proposed
INTRODUCTION
■
Control of regio- and stereoselectivity in glycosylation reactions
has been a long-standing challenge in carbohydrate chemistry.
Modern glycosylation methods rely heavily on protective groups
to achieve these two types of selectivity. Hydroxyl group
protection provides a means to enforce regiocontrol by blocking
undesired sites of reactivity. The ways that protective groups
influence stereoselectivity can be more subtle, and include
neighboring group participation as well as steric, inductive,
conformational, stereoelectronic and directing effects.1 In recent
years, considerable effort has been devoted to an alternative
approach involving the use of external reagents or catalysts to
achieve regio-2,3 and/or stereoselective glycosylations.4,5 Such
methods may offer greater efficiency or flexibility in the synthesis
of complex glycosides. Enzymes (naturally occurring glycosyl-
transferases and glycosyl hydrolases, as well as engineered
mutants) illustrate the high levels of glycosylation selectivity that
can be achieved under catalyst control, and are powerful tools for
laboratory synthesis.6 It remains to be established whether
comparable results can be achieved using synthetic glycosylation
catalysts, and whether such synthetic catalysts might offer
complementary properties relative to enzymes.
Scheme 1. Borinic Acid-Catalyzed Regioselective
Glycosylation of Partially Unprotected Acceptors
mechanism of activation involves the formation of a
tetracoordinate borinic ester that displays enhanced oxygen-
centered nucleophilicity relative to a free alcohol. Preliminary
kinetics experiments, as well as stereochemical outcomes
(formation of β-glucopyranosides or β-galactopyranosides from
α-configured halides, bearing either ester or ether protective
groups) were consistent with the acceleration of an associative
(SN2-type) pathway by the catalyst. In keeping with this idea, the
Received: July 5, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/jacs.6b06943
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
borinic acid was able to enhance the level of β-selectivity in
couplings of α-2-deoxyglycosyl halides.5f The borinic acid
activation concept has been extended to other classes of glycosyl
donors or equivalents, including pyranone-derived allylic
carbonates (in the presence of a Pd(0) cocatalyst)3c and glycal
epoxides.5h
from the corresponding free hemiacetals using benzenesulfonic
anhydride, dibutyl sulfoxide as a Lewis base catalyst and the
hindered Brønsted base 2,4,6-tri-tert-butylpyridine (TTBP) at
room temperature. Armed donors generally showed modest
levels of α-selectivity in couplings with primary and secondary
alcohols under these conditions. Bennett and co-workers
prepared α-2-deoxyglycosyl toluenesulfonates (tosylates) using
a strong base (potassium bis(trimethylsilyl)amide) and
toluenesulfonic anhydride or toluenesulfonylimidazole at low
temperature (−78 °C). These reacted with strong nucleophiles
(thiolates or alkoxides) via inversion of configuration to give β-2-
deoxyglycosides.
We were drawn in particular to glycosyl mesylates by two
reports detailing their rapid and efficient generation from free
hemiacetals, simply by treatment with methanesulfonic anhy-
dride (Ms2O) and an amine base.9d,e These conditions appeared
to be suitable for in situ formation of the mesylate, which could
then be added directly to a glycosyl acceptor in the presence of
the borinic acid catalyst. Reactions with simple acceptors
(methanol, isopropanol, cyclohexanol) reported in the 1970s
appear to be the only documented examples of O-glycosylation
using mesylate donors.9c−e
Development of Conditions for Borinic Acid-Catalyzed
Couplings of Glycosyl Mesylates. We selected the coupling
of 2,3,4,6-tetra-O-benzyl-D-glucopyranose (1a) and 6-O-silylated
α-mannopyranoside triol 2a to evaluate conditions for borinic
acid-catalyzed coupling of a mesylate generated in situ (Table 1).
Acceptor 2a is activated by diarylborinic acids at the 3-OH group,
but also shows a relatively high level of “intrinsic” selectivity for
reactions at this position. We anticipated that using an acceptor
showing this type of regiochemical bias would simplify the
analysis of product mixtures, allowing us to focus on the effect of
catalyst on the stereochemical outcome of the reaction. Addition
Although borinic acid-catalyzed glycosylations have already
been applied to facilitate the synthesis of complex, carbohydrate-
derived targets,3c,5h,7 there remains a need for regio- and
stereoselective methods that employ readily available glycosyl
donors and show high levels of functional group tolerance. In
both of these regards, the glycosyl bromide and chloride donors
employed in our first-generation catalytic method are less than
ideal. Their preparation requires relatively harsh reagents (e.g.,
HCl, HBr, BCl3), which may not be compatible with acid-labile
functional groups such as silyl ethers, acetals and glycosidic
linkages. Certain variants, including “armed”8 (ether-protected),
deoxygenated and furanosyl halides, are prone to hydrolysis
upon isolation and/or storage. The use of silver(I) oxide as halide
abstracting agent under our catalytic conditions created
mechanistic complexity and mass transfer problems due to its
low solubility in the reaction medium. We aimed to address these
limitations by identifying other glycosyl donors and activation
conditions that would be compatible with borinic acid activation
of glycosyl acceptors.
Here, we describe a method for regio- and stereoselective
couplings of diols and triols with glycosyl methanesulfonates
(mesylates), using a diarylborinic acid catalyst. The glycosyl
mesylates are generated in situ from the corresponding free
hemiacetals under mild conditions (methanesulfonic anhydride
(Ms2O), amine base), and their borinic acid-catalyzed reactions
with glycosyl acceptors take place readily at room temperature.
The borinic acid has a significant influence on the stereochemical
outcome, with reversals of α/β-selectivity relative to the
uncatalyzed reaction being observed in several instances. The
organoboron-catalyzed reaction enables the preparation of 1,2-
trans-configured linkages (e.g., β-glucopyranosides, β-galacto-
pyranosides, α-arabinofuranosides) in the absence of neighbor-
ing group participation, thus permitting the use of armed donors.
Detailed kinetic studies of both the catalyzed and uncatalyzed
reactions have been conducted. These suggest that although the
catalyzed and uncatalyzed pathways have different stereo-
chemical outcomes, both operate by associative (SN2-type)
mechanisms.
Table 1. Optimization of Catalytic Glycosylation of Acceptor
2a with Glycosyl Hemiacetal 1a
RESULTS AND DISCUSSION
■
Glycosyl sulfonates9 attracted our interest as potential electro-
philes for reactions with borinic acid-activated acceptors.
Previous work from our group indicated that donors requiring
activation by a Brønsted or Lewis acid (e.g., trichloroacetimi-
dates, thioglycosides, glycosyl phosphates) did not undergo
borinic acid-catalyzed glycosylation.3a These activators may have
interfered with the formation of the tetracoordinate borinate
intermediate, which is favored under basic conditions.10 The
ability of glycosyl sulfonates to react with alcohols in the absence
of additives,9a,b,g or under basic conditions,9c−f,h is thus an
important advantage. The trifluoromethanesulfonates (triflates)
have been studied thoroughly, and have been implicated as
intermediates in several important preparative methods.1a,11 The
chemistry of other classes of glycosyl sulfonates is less well
advanced, although important progress regarding arenesulfo-
nates has been achieved by the groups of Gin12 and Bennett.13
Gin found that glycosyl benzenesulfonates could be generated
equiv
Ms2O
equiv
2a
catalyst
(equiv)
conv
a
b
entry
base
(%)
α:β
1
2
iPr2NEt
iPr2NEt
iPr2NEt
iPr2NEt
iPr2NEt
Et3N
1.2
1.2
1.2
1.2
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.2
1.2
0.8
none
50
57
54
66
83
76
88
85
71
80
2.0:1
1:6.1
1:1.8
1:7.4
1:6.0
1:2.4
1:8.7
1:8.4
2.0:1
1:10
4a (0.1)
4b (0.2)
4c (0.2)
4c (0.2)
4c (0.2)
4c (0.2)
4c (0.1)
none
3
4
5
6
c
7
PMP
8
PMP
PMP
PMP
9
10
4c (0.08)
a
Conversion of 1a determined by HPLC analysis of the unpurified
b
reaction mixture. Determined by HPLC analysis of the unpurified
reaction mixture. PMP denotes 1,2,2,6,6-pentamethylpiperidine.
c
B
DOI: 10.1021/jacs.6b06943
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
of a solution of Ms2O to hemiacetal 1a and iPr2NEt in
dichloromethane resulted in a faint yellow color within minutes
of mixing at room temperature. Formation of the glycosyl
mesylate under these conditions was supported by nuclear
magnetic resonance spectroscopy (see below). Presumably, the
ability to generate a reactive sulfene electrophile from Ms2O and
an amine base contributes to the high rate of this trans-
formation.14 Acceptor 2a (either with or without a catalytic
amount of 4a, the anhydride of diphenylborinic acid) was then
added, and the resulting mixture stirred at room temperature
overnight. Analysis by high-pressure liquid chromatography
(HPLC) revealed similar levels of conversion of 1a for the
catalyzed and uncatalyzed reactions, with the formation of 1,3-
linked disaccharide 3a being observed in both cases, as
anticipated (entries 1 and 2). However, the stereochemical
outcomes of the uncatalyzed and catalyzed reactions clearly
differed, with the former showing a modest preference for the α-
glucopyranoside and the latter yielding the β-glucopyranoside as
the major product.
Scheme 2. Borinic Acid-Catalyzed Stereoselective
Glycosylation with Glycosyl Methanesulfonates
a
The solvent, base, organoboron catalyst and acceptor
concentration were varied with the goal of developing a high-
yielding, β-selective process. Dichloromethane provided superior
results to acetonitrile, toluene, tetrahydrofuran (THF) and
diethyl ether (Table S1, Supporting Information). Borinic ester
4b resulted in lower β-selectivity than 4a (entry 3). Since the
same catalyst−substrate complex would be generated from 4a
and 4b, this difference likely reflects a higher contribution from
the α-selective background pathway due to slow entry of
precatalyst 4b10 into the catalytic cycle (see below). An
improvement in β:α ratio was obtained using oxaboraanthra-
cene-derived catalyst 4c (entry 3).15 We have found that 4c
displays higher catalytic activity than 4b for several types of diol
functionalization reactions: the enhanced nucleophilicity of its
tetracoordinate borinates is apparently a result of incorporating
the borinic acid group into a 6-π electron system. Variation of the
tertiary amine base revealed that the more sterically encumbered
piperidine derivative 1,2,2,6,6-pentamethylpiperidine (PMP)
a
β:α ratios were determined by 1H NMR or HPLC analysis of
unpurified reaction mixtures. Yields are for isolation of the depicted
compound in isomerically pure form, unless otherwise indicated.
b
Glycosylation reaction times were not optimized. Isolated as a
c
mixture of anomers. Mesylate formation was carried out at −25 °C.
improved both conversion and selectivity relative to Hunig’s
̈
Glycosylation was stirred overnight at −25 °C, then allowed to warm
d
base, while the less encumbered triethylamine had a negative
effect (entries 6 and 7). We note that the improvement obtained
using PMP rather than iPr2NEt was rather modest, and that the
latter is a viable alternative in situations where the higher cost of
PMP is an issue. A further increase in β-selectivity was obtained
by changing the concentration of the glycosyl acceptor: a trend of
higher β:α ratio at lower concentrations of 2a is evident (entries 8
and 10). This series of experiments also revealed that the catalyst
loading could be reduced to 10 mol % (relative to 2, the limiting
reagent under the conditions of entry 10) without deleterious
effect. A comparison of uncatalyzed and catalyzed reactions using
the optimized base (PMP) demonstrates the extent to which the
catalyst is able to “switch” the stereoselectivity of this
glycosidation (entries 8 and 9).
slowly to room temperature over 5 h. 4a (5 mol %) was used in place
of 4c.
regiocontrol were achieved for glycosyl acceptors having 1,2- or
1,3-diol motifs. Aryl and alkyl thioglycoside-derived glycosyl
acceptors were tolerated under these conditions. In the reaction
of the glucofuranoside acceptor to generate product 3i, a
preference for activation of the secondary 5-OH group over the
primary 6-OH group was observed (8:1 regioselectivity),
possibly via the 3,5-O-borinate. The isolation of the minor, 6-
O-glycosylated regioisomer as a mixture of stereoisomers (2:1
α:β) suggests that it arose from an uncatalyzed, background
reaction. For the synthesis of β-1−6-linked product 3j, the more
Lewis acidic diphenylborinic acid provided superior results to 4c.
Scheme 3 depicts substrate combinations that led to relatively
low levels of β-selectivity under the catalytic conditions. These
illustrate limitations of the methodology, but also provide
additional evidence for the catalyst’s ability to alter the
stereochemical course of reactions of glycosyl mesylate donors.
For example, the uncatalyzed reaction of 4,6-O-benzylidene-
protected galactopyranosyl donor showed a particularly high
level of α-selectivity (product 3k).16 The catalyst was not able to
fully overcome this intrinsic bias, leading to a 1.4:1 α:β mixture.
Similarly, the use of a 2-deoxyglucopyranosyl donor led to an α-
selective reaction in the absence of catalyst (5.1:1 α:β) and only a
Scope and Limitations. The optimized protocol was used
to construct several types of glycosidic linkages (Scheme 2). The
β:α ratios reported in Scheme 2 are based on analysis of
unpurified reaction mixtures by 1H NMR spectroscopy or
HPLC. The yields are for isolation of the major regio- and
stereoisomer in pure form by flash chromatography on silica gel,
unless otherwise indicated. High levels of 1,2-trans selectivity
were obtained for armed glucopyranose, galactopyranose and
arabinofuranose-derived donors. The ability to reliably generate
such linkages in the absence of neighboring group participation is
noteworthy. As observed in our previous studies, high levels of
C
DOI: 10.1021/jacs.6b06943
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Scheme 3. Catalyzed and Uncatalyzed Stereoselectivites for
Glycosylations of 2b with Selected Donors
Scheme 4. Generation and Borinic Acid-Catalyzed Reaction
of a Glycosyl Tosylate
a b
,
tosylate leaving groups, but the mesylate donor provides superior
levels of 1,2-trans selectivity in the presence of the catalyst
(compare Scheme 4 to Scheme 2, product 3c).
Studies of Glycosyl Mesylates by NMR Spectroscopy.
NMR spectroscopy was used to verify that glycosyl mesylates
were formed in the initial step of this protocol, and to evaluate
their stabilities and anomeric configurations. Treatment of 1a
with Ms2O and PMP in CDCl3 (in place of CH2Cl2, the solvent
for the glycosylation process) resulted in the quantitative
formation of mesylate 5a within 10 min at room temperature.
The hydrolysis of the mesylate generated under these conditions
has a half-life in excess of 24 h at room temperature in a capped
NMR tube. Both anomers of the glycosyl mesylate could be
observed, and a 10:1 α:β ratio was determined.
a
Conditions: Hemiacetal (1.25 equiv), PMP (4 equiv), Ms2O (1.88
equiv), <1 h; acceptor 2b (1 equiv), 4c (10 mol %), overnight. For 1h
and 1f, the glycosylations were stirred overnight at −25 °C, then
allowed to warm slowly to room temperature over 5 h. Reaction times
were the same for catalyzed and uncatalyzed reactions, and were not
b
1
optimized. α:β ratios were determined by H NMR spectroscopic
analysis of the crude reaction mixture. The diastereomeric
disaccharides were separated by flash chromatography on silica gel,
and the total yield (α + β) is reported. Product 3m was isolated as a
mixture of anomers.
c
In a similar way, mesylates 5b−5e were generated and
1
characterized by H and 13C NMR spectroscopy in CDCl3
(Scheme 5). With the exception of the 2-deoxy derivative, each
modestly β-selective reaction in the presence of 4c (1:2.7 α:β,
product 3l). Couplings of the pseudoenantiomeric galacto- and
manno-configured acceptors with the D-arabinofuranose-derived
mesylate donor appear to show a matching/mismatching effect
on stereoselectivity (products 3h (Scheme 2) and 3m): an
unusually high level of 1,2-cis-selectivity was observed for the
Araf/Man coupling in the absence of catalyst.17 Borinic acid
catalysis did not give synthetically useful results for couplings of
manno-configured glycosyl mesylates. The couplings of 2a with
2,3,4,6-tetra-O-benzyl-D-mannopyranose led to a complex
mixture of isomeric glycosides. The use of the mesylate derived
from 4,6-O-benzylidene-2,3-di-O-benzylmannopyranose re-
sulted in an α-selective 3-O-glycosylation of acceptor 2a in the
presence of the catalyst, but in low (approximately 25%) yield.
We compared the results obtained using a glycosyl mesylate
donor to those using the corresponding tosylate, generated by
the reaction of the glycosyl hemiacetal with Ts2O in the presence
of catalytic dibutyl sulfoxide, according to Gin’s protocol (but
using PMP in place of the hindered pyridine derivative TTBP:
Scheme 4).12a Formation of the tosylate (albeit not in
quantitative yield, 17:1 α:β) was confirmed by 1H NMR
spectroscopy in CD2Cl2. In the presence of catalyst 4c,
disaccharide 3c was obtained as a 2.4:1 anomeric mixture
favoring the β-glycoside, whereas an α-selective reaction (5.1:1
α:β) took place in the absence of catalyst. The stereoselectivities
were determined by 1H NMR spectroscopy of unpurified
reaction mixtures. No effort was made to optimize the yields of
these reactions, which were lower than those obtained using the
glycosyl mesylate. In any case, the effect of the borinic acid on the
stereochemical outcome holds for both the mesylate and the
Scheme 5. Equilibrium Anomeric Ratios of Glycosyl
a b
,
Methanesulfonates Determined by 1H NMR
a
Anomeric ratios were determined by 1H NMR spectroscopy.
Temperatures at which NMR spectra were acquired.
b
mesylate was formed along with a detectable amount of the
minor anomer, allowing determination of the α:β ratios shown in
the Scheme. Whereas mesylates 5a−5d were sufficiently stable to
be characterized at room temperature, the 2-deoxypyranosyl and
furanosyl derivatives 5e and 5f were studied at lower temperature
(−25 °C) to prevent decomposition. (For the same reason, these
mesylates were generated and subjected to glycosyl acceptors at
−25 °C for the preparative reactions shown in Schemes 2 and 3.)
D
DOI: 10.1021/jacs.6b06943
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Variable temperature NMR revealed that decomposition of the
2-deoxy mesylate (via elimination to the glycal) took place
between −5 and 5 °C. Both the high (>20:1) α:β ratio and the
decomposition temperature of this compound are consistent
with Bennett and co-workers’ observations regarding the
corresponding 2-deoxy tosylate (onset of elimination of −5 °C
in THF-d8).13b
The interconversion of mesylate anomers 5b-α and 5b-β was
studied using NMR exchange spectroscopy (EXSY) in CDCl3 at
25 °C.18 1D selective EXSY buildup curves were constructed by
irradiating the signal corresponding to H-1 of 5b-α and
integrating that corresponding to H-1 of 5b-β, or vice versa,
for a variety of mixing times τm. The slopes of the initial, linear
regions of the plots of EXSY integral versus τm were used to
determine observed rate constants for the anomerization
reactions (Figure 1a). These were found to be (1.02 0.06) ×
10−2 s−1 (5b-β → 5b-α) and (9.3 1.3) × 10−4 s−1 (5b-α → 5b-
β). The sensitivity provided by a cryogenically cooled probe was
important for determining the rather low rates of exchange. The
ratio of rate constants is in good agreement with the equilibrium
ratio of anomers that was determined independently by 1H NMR
spectroscopy (Scheme 5).
hypothesis (Figure 1b): the linear relationship suggests first-
order kinetics in MsO−. Given that anomerization of activated
donors is an important process in stereoselective glycosidations,
further applications of the EXSY technique to related systems
may be of interest.
Reaction Progress Kinetic Analysis of Catalyzed and
1
Uncatalyzed Pathways. The H NMR spectroscopy studies
indicated that glycosyl mesylates 5a−5d could be generated
cleanly from 1a−1d, and that their decomposition reactions were
slow relative to nucleophilic substitution by a glycosyl acceptor
(either in the presence or in the absence of diarylborinic acid
catalyst). These features, as well as the ability to conduct the
coupling under homogeneous reaction conditions and without
additional activation of the donor (e.g., by a Brønsted acid,
electrophile or Lewis acid promoter), suggested the possibility of
using in situ analysis to study the kinetics of the glycosylation
reaction. Mechanistic understanding of nonenzymatic glyco-
sylation reactions has developed slowly relative to the discovery
of new protocols,19 and detailed kinetic studies remain
uncommon, especially for preparative glycosylation methods
(as opposed to hydrolysis or solvolysis reactions).20 Previously,
we determined initial rates of borinic acid-catalyzed couplings of
1
1a with acetobromoglucose by H NMR spectroscopy, taking
aliquots of reaction mixtures.3a The reactions showed first-order
kinetics in catalyst, glycosyl donor and glycosyl acceptor, and
apparent zero-order kinetics in Ag2O (presumably reflecting a
reaction at the surface of the insoluble promoter). We aimed to
conduct a more comprehensive study of the present method-
ology, using reaction progress kinetic analysis21 of glycosylations
conducted under synthetically relevant conditions.
The coupling of galactopyranosyl mesylate 5b and 6-O-TBS-
mannopyranosyl thioglycoside 2b was chosen for monitoring by
1
in situ H NMR spectroscopy (Scheme 6). On a 700 MHz
Scheme 6. Standard Conditions for In Situ Kinetic Studies
instrument equipped with a cryogenically cooled probe, the
spectrum was sufficiently well resolved to enable accurate
integration of signals corresponding to the two anomers of the
glycosyl mesylate, and those of the product glycoside. The
reaction was conducted under the standard conditions shown in
Scheme 2, but using CDCl3 as the NMR solvent with mesitylene
as a quantitative internal standard, collecting a spectrum every 3
min over the course of the experiment. Figure 2 shows the
concentrations of glycosyl donor (the sum of the concentrations
of the two mesylate anomers), glycosyl acceptor and products 3c-
β and 3c-α as a function of time. The α:β ratio of the glycosyl
mesylate 5b remained constant over the course of the
experiment. Donor concentration data for an experiment
conducted in the absence of borinic acid 4c are also shown.
The catalyzed reaction was complete in less than 2.5 h, whereas
the uncatalyzed reaction required more than 10 h. It should also
be noted that the acceptor and donor were consumed at different
rates under this protocol. Although the glycosyl mesylate was
present in excess at the outset of the reaction, unreacted acceptor
remained at its completion. Hydrolysis and other side reactions
Figure 1. (a) EXSY buildup curves (EXSY signal integration versus
■
◆
mixing time τm) for the formation of 5b-α (red ) and 5b-β (blue ) by
anomerization. The lines used to determine the apparent rate constants
are depicted on the graph. (b) Plot of observed rate constant for α → β
isomerization versus concentration of added Bu4N+MsO−.
At least one equivalent of an ammonium mesylate salt is
present in solutions of glycosyl mesylates generated from
hemiacetals using Ms2O and PMP. We estimate that the
concentration of MsO− in solutions of glycosyl mesylate
generated under the conditions shown in Scheme 5 is 65 mM.
Nucleophilic substitution by MsO− is thus a plausible
anomerization pathway. The dependence of the α → β
anomerization rate on the concentration of added tetrabuty-
lammonium mesylate (Bu4N+MsO−) is consistent with this
E
DOI: 10.1021/jacs.6b06943
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
donor concentration implies first-order kinetics in the glycosyl
mesylate.
The data that were used to determine the kinetic orders in
catalyst and base are shown in Figures 3b and 3c, respectively.
The overlay of rate/[catalyst] versus [donor] graphs for
reactions conducted at different catalyst loadings (10 vs 15 mol
%) reflect first-order kinetics in borinic acid 4c. Figure 3c shows
concentration of product versus time data for two experiments:
one conducted with the standard concentration of PMP (3.2
equiv relative to the starting glycosyl hemiacetal) and one with a
higher concentration (4.8 equiv). Since roughly 1.5 equiv of PMP
are consumed in the reaction of the hemiacetal with Ms2O, this
corresponds to roughly a doubling of the concentration of “free”
PMP in the glycosylation step. The data are presented as
concentration of product versus time because glycosyl donor
decomposition and side reactions were more evident when the
higher concentration of base was used. These issues complicated
interpretation of the data, but the similar initial regions of the
graphs suggest roughly zero-order kinetics in PMP.
◆
●
Figure 2. Graph of the concentrations of 5b (black ), 2b (red ), 3c-β
▲
(blue ) and 3c-α (green +) versus time for the standard conditions
shown in Scheme 3. Concentration of 5b versus time is also plotted for
the uncatalyzed reaction (grey ). The starting concentration of 5b was
different for the uncatalyzed reaction, and so the graph has been
◆
timeshifted for comparison purposes.
Reaction progress kinetic analysis provides a powerful way to
assess catalyst decomposition or product inhibition using “same
excess” experiments. These experiments rely on the assumption
that the difference in concentration of two reactants (the
“excess”) remains constant over time if they react in a 1:1
stoichiometry. Unfortunately, side reactions of the glycosyl
mesylate (e.g., hydrolysis, formation of isomeric glycosides or
bis-glycosylated products: see above) caused a sufficient change
in excess over time to compromise the results of this type of
experiment. (The “different excess” experiments of the type
shown in Figure 3 were possible because concentrations of
reactants and products were determined independently using an
internal standard.) However, informative results were obtained
from an experiment in which tetrabutylammonium mesylate
(100 mM) was added to the borinic acid-catalyzed glycosylation.
A decrease in rate was observed (Supporting Information, Figure
S4), along with a decrease in β-selectivity from 10:1 to
approximately 3:1 β:α. Either a common-ion effect22 or
inhibition of the borinic acid by anion binding could be
responsible for this behavior. The second explanation seems
more reasonable, given (i) the observation of changes in the 1H
NMR spectrum of 4c upon addition of Bu4N+MsO−; (ii) the
known ability of boronic and borinic acids to act as anion
receptors;23 and (iii) the low likelihood of a dissociative
of the donor are responsible for its accelerated consumption
relative to the acceptor.
The concentration versus time data for borinic acid-catalyzed
formation of β anomer 3c-β were transformed into rate versus
time data by fitting to a differential equation (see the Supporting
Information). These were used to construct “graphical rate
equations” of rate versus glycosyl donor concentration. The
concentration of the β-glycoside (rather than the sum of
concentrations of the two product anomers) was used in this
analysis based on the hypothesis that the α-anomer arises
primarily from the uncatalyzed background pathway (see below).
Figure 3a shows graphs of rate versus glycosyl mesylate
concentration for reactions conducted at the same starting
concentration of donor and catalyst loading, but with three
different acceptor concentrations (30 mM, 40 mM and 60 mM).
Overlay of the plots corresponding to 40 mM and 60 mM
acceptor concentration is evident, indicating zero-order (or
apparent zero-order) kinetics in acceptor. The rate profile at 30
mM acceptor concentration overlays with the two others at the
early stages of the reaction, but a decrease in rate and departure
from linearity are evident at lower donor concentrations. This
behavior is discussed in more detail below. Given the zero-order
kinetics in glycosyl acceptor, the linear dependence of rate on
Figure 3. Kinetics of the glycosylation reaction of acceptor 2b with donor 5b, catalyzed by borinic acid 4c. (a) Rate of 3c-β product formation versus
concentration of donor 5b ([5b-α] + [5b-β]) for experiments carried out at different acceptor concentrations. (b) Rate of 3c-β product formation,
normalized by catalyst concentration, versus concentration of 5b ([5b-α] + [5b-β]) for experiments carried out at different catalyst concentrations. (c)
Concentration of 3c-β versus time for experiments carried out at two concentrations of PMP. The graph for the reaction at the lower PMP concentration
has been timeshifted for comparison purposes.
F
DOI: 10.1021/jacs.6b06943
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 4. Kinetics of the uncatalyzed glycosylation reaction of acceptor 2b with donor 5b. (a) Rates of product formation (3c-α and 3c-β) versus
concentration of donor 5b ([5b-α] + [5b-β]) for experiments carried out at different acceptor concentrations. (b) Rates of product formation (3c-α and
3c-β), normalized by acceptor concentration, versus concentration of donor 5b ([5b-α] + [5b-β]). (c) Concentration of product (3c-α and 3c-β) versus
time for experiments carried out at two concentrations of PMP.
mechanism for the borinic acid-catalyzed glycosylation (see
below). The decreased β-selectivity is due to the increased
contribution of the uncatalyzed pathway, which is not inhibited
by Bu4N+MsO−, as discussed later in this section.
The kinetic orders for the coupling of 2b and 5b catalyzed by
borinic acid 4c are summarized in Figure 5. First-order kinetics in
catalyst and glycosyl donor, along with apparent zero-order
kinetics in glycosyl acceptor, could arise from acceleration of an
associative process by catalyst binding to the glycosyl acceptor,
with the apparent zero-order dependence resulting from
saturation kinetics. The other possibility is a dissociative (SN1-
like) mechanism involving Lewis acid activation of the glycosyl
mesylate by the catalyst. The former proposal is consistent with
our mechanistic data for other borinic acid-catalyzed trans-
formations,3a,10 and with the ability of borinic acids to activate
diols toward reactions with electrophiles other than glycosyl
donors. Further support for the glycosyl acceptor activation
hypothesis was obtained by replacing PMP with the weaker base
sym-collidine. Borinic acid−diol complexation was not observed
in the presence of the latter, and neither rate acceleration by 4c
Figure 5. Summary of kinetic orders determined by reaction progress
kinetic analysis for the catalyzed and uncatalyzed reactions.
nor a catalyst-induced change in stereoselectivity was observed
for glycosylations using this base. Additional evidence for
saturation kinetics can be drawn from the rate profile of the
reaction conducted at 30 mM acceptor concentration (Figure
3a). At the point where the data deviate from those at the higher
acceptor loadings, the acceptor concentration has reached
roughly 6 mM, which is on par with the catalyst concentration
(4 mM). A decrease in rate and deviation from an overall first-
order rate law is expected at these relative concentrations (and
beyond), as saturation kinetics would no longer apply.
We conducted a similar analysis of the kinetics of the
uncatalyzed reaction of 2b with 5c (Figures 4a−4c). This
reaction gives an α:β ratio of 2.6:1, and the rates of formation of
the α- and β-anomers of product were analyzed separately. The
rate law was found to be the same for the pathways leading to the
two anomers of product: first-order kinetics in acceptor 2b, first-
order in donor 5c, and roughly zero-order in PMP (Figure 5).
Inhibition by Bu4N+MsO− was not observed (Supporting
Information, Figure S5). Both the observation of a kinetic
dependence on acceptor concentration and the lack of a
common ion effect suggest that the uncatalyzed glycosylation,
like the catalyzed process, proceeds by an associative mechanism.
Having obtained this information regarding the kinetics of the
uncatalyzed reaction, we returned to the data for the borinic acid-
catalyzed process, examining the rate of formation of the minor
anomer 3c-α. It proved difficult to accurately determine the
concentration of the small amount of α-configured product
formed under the conditions of Scheme 6. However, conducting
the experiment with higher concentrations of acceptor resulted
in a lower level of β-selectivity, thus permitting analysis of the
concentration data for the minor α-anomer. Whereas the process
leading to the β-configured product showed apparent zero-order
(saturation) kinetics in acceptor, the formation of 3c-α followed
first-order kinetics in 2b (Figure S6). This observation suggests
that much of the α-product formed in the presence of 4c arises
from a competing, uncatalyzed pathway, and that the intrinsic β-
selectivity of the borinic acid-catalyzed process is very high for
this donor−acceptor combination. It provides a rationale for the
optimization data from Table 1 showing increased β-selectivity
upon decreasing the quantity of acceptor 2b. It also suggested
that reducing the concentration of “free” 2b in the reaction
mixture by slow addition of the acceptor would improve the β-
selectivity of the borinic acid-catalyzed process. Indeed, when a
solution of 2b was added to mesylate 5b, catalyst and PMP over 3
h by syringe pump, an improvement in stereoselectivity was
obtained (β:α 14:1, versus 8:1 β:α without using the slow
addition protocol). No attempt was made to further optimize the
addition rate for this process. This result suggests that slow
addition may be a useful approach for improving stereoselectivity
in challenging couplings of mesylate donors.
G
DOI: 10.1021/jacs.6b06943
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
CONCLUSIONS
Scheme 7. Mechanistic Proposal for Stereochemical
Outcomes of Uncatalyzed and Catalyzed Glycosylation of
Mesylate Donor 5b
■
In summary, glycosyl mesylate donors have been employed in
borinic acid-catalyzed couplings with diol-containing acceptors,
enabling the formation of a variety of regio- and stereodefined
linkages. From a preparative standpoint, the catalyst-controlled
method represents an operationally simple protocol for
generating 1,2-trans-configured linkages from armed donors.
The glycosyl mesylates are generated efficiently from the
corresponding hemiacetals, and can be characterized in solution.
They react with glycosyl acceptors under mild conditions
(hindered amine base, CH2Cl2, room temperature), but the
diarylborinic acid catalyst gives significant levels of rate
acceleration for couplings of acceptors having 1,2- or 1,3-diol
moieties. Our “second-generation” oxaboraanthracene-derived
borinic acid, which gives rise to highly nucleophilic borinic ester
adducts, has proved to be especially useful in this context. We
anticipate that these conditions will prove useful for
regioselective glycosylations of more complex substrates.
A noteworthy aspect of this study is the influence of borinic
acid catalyst 4c on the stereoselectivity of reactions of glycosyl
mesylates. In several instances (Table 1, Scheme 3), there is a
clear reversal in stereochemical outcome for the reaction carried
out in the presence of the catalyst versus in its absence.
Influencing the stereoselectivity of glycosylations by catalytic
activation of the acceptor is an emerging concept, and one that
may offer new ways to approach the formation of challenging
classes of linkages.5f−h The kinetics experiments described here
provide insight into how such a switch in stereoselectivity can
arise.
Both the uncatalyzed and catalyzed glycosylations show
kinetics consistent with associative mechanisms. The rates of
MsO−-catalyzed anomerization of glycosyl mesylate 5b
determined by EXSY indicate that the predominant product,
both in the presence and the absence of catalyst, can plausibly be
generated by inversion of configuration of the corresponding
mesylate. The general approach of using EXSY to determine
anomerization rates in combination with analysis of reaction
kinetics may have broader utility for studying the mechanisms of
glycosylations (e.g., for other classes of glycosyl sulfonates). In
the present case, the modest preference for the α-configured
glycoside product in the absence of catalyst appears to reflect the
higher reactivity of the minor β-mesylate, consistent with the
Curtin−Hammett principle. This corresponds to the mechanism
proposed by Lemieux and co-workers for halide-catalyzed 1,2-cis-
selective couplings of glycosyl halides.5a In the presence of the
catalyst, the displacement of the α-mesylate is accelerated,
leading to the selective formation of the β-glycoside. The
selectivity for acceleration of the pathway leading to the β-
anomer appears to be high, since most of the minor α-anomer (in
the case of product 3c) arises from the uncatalyzed “background”
pathway. Kinetic isotope effect experiments and computation
would be needed to elucidate the transition state structural
effects that lead to these differences in behavior between
unactivated alcohols and borinate-activated acceptors. However,
a reasonable preliminary hypothesis is that the sterically hindered
nature of the tetracoordinate borinic ester nucleophile results in
selective acceleration of the 1,2-trans-selective pathway. This idea
is consistent with the scope and limitations depicted in Scheme 2
and 3. The overall mechanistic proposal is summarized in
Scheme 7.
the borinic acid-catalyzed process. The observation of a lower
kinetic order in glycosyl acceptor for the catalyzed versus the
uncatalyzed reaction (apparent zero-order (saturation) versus
first-order kinetics), led to the use of a slow addition protocol to
maximize β-selectivity in the catalytic process. This study has
yielded insight into the reactivity of glycosyl sulfonates, and may
be a useful starting point for the development of other
stereoselective glycosylation methods that take advantage of
catalytic glycosyl acceptor activation.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/jacs.6b06943.
■
S
Experimental procedures, characterization data, additional
optimization data and kinetic profiles, EXSY buildup
curves and exchange rate constants, sample code for fitting
kinetic data, and NMR spectra for all new compounds.
(PDF)
AUTHOR INFORMATION
Corresponding Author
*mtaylor@chem.utoronto.ca
■
The kinetics experiments have not only provided mechanistic
insight, but have also pointed toward practical improvements to
Notes
The authors declare no competing financial interest.
H
DOI: 10.1021/jacs.6b06943
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
1762. (g) Srivastava, V. K.; Schuerch, C. J. Org. Chem. 1981, 46, 1121−
1126. (h) Szeja, W. Synthesis 1988, 1988, 223−224.
(10) Lee, D.; Williamson, C. L.; Chan, L.; Taylor, M. S. J. Am. Chem.
Soc. 2012, 134, 8260−8267.
(11) (a) Garcia, B. A.; Poole, J. L.; Gin, D. Y. J. Am. Chem. Soc. 1997,
119, 7597−7598. (b) Nguyen, H. M.; Chen, Y.; Duron, S. G.; Gin, D. Y.
J. Am. Chem. Soc. 2001, 123, 8766−8772. (c) Crich, D.; Smith, M. J. Am.
Chem. Soc. 2001, 123, 9015−9020. (d) Nokami, T.; Shibuya, A.;
Tsuyama, H.; Suga, S.; Bowers, A. A.; Crich, D.; Yoshida, J.-i. J. Am.
Chem. Soc. 2007, 129, 10922−10928. (e) Crich, D. J. Carbohydr. Chem.
2002, 21, 663−686. (f) Crich, D. Acc. Chem. Res. 2010, 43, 1144−1153.
(g) Frihed, T. G.; Bols, M.; Pedersen, C. M. Chem. Rev. 2015, 115,
4963−5013.
(12) (a) Boebel, T. A.; Gin, D. Y. Angew. Chem., Int. Ed. 2003, 42,
5874−5877. (b) Boebel, T. A.; Gin, D. Y. J. Org. Chem. 2005, 70, 5818−
5826.
(13) (a) Issa, J. P.; Steliotes, E.; Bennett, C. S. Org. Lett. 2013, 15,
4170−4173. (b) Issa, J. P.; Bennett, C. S. J. Am. Chem. Soc. 2014, 136,
5740−5744.
(14) Truce, W. E.; Campbell, R. W. J. Am. Chem. Soc. 1966, 88, 3599−
3604.
ACKNOWLEDGMENTS
■
This work was funded by NSERC (Discovery Grants and Canada
Research Chairs Programs), the Canada Foundation for
Innovation (projects #17545 and #19119) and the Ontario
Ministry for Research and Innovation. Dr. Darcy Burns, Dr.
Sergiy Nokhrin and Dr. Jack Sheng (University of Toronto) are
acknowledged for helpful discussions and technical assistance
with the NMR studies.
REFERENCES
■
(1) For examples of these types of effects: (a) Crich, D.; Sun, S. J. Am.
Chem. Soc. 1997, 119, 11217−11223. (b) Kim, J.-H.; Yang, H.; Park, J.;
Boons, G.-J. J. Am. Chem. Soc. 2005, 127, 12090−12097. (c) Okada, Y.;
Mukae, T.; Okajima, K.; Taira, M.; Fujita, M.; Yamada, H. Org. Lett.
2007, 9, 1573−1576. (d) Garcia, A.; Otte, D. A. L.; Salamant, W. A.;
Sanzone, J. R.; Woerpel, K. A. Angew. Chem., Int. Ed. 2015, 54, 3061−
3064. (e) Kayastha, A. K.; Jia, X. G.; Yasomanee, J. P.; Demchenko, A. V.
Org. Lett. 2015, 17, 4448−4451.
(2) (a) Auge,
́
C.; Veyrier
̀
es, A. J. Chem. Soc., Perkin Trans. 1 1979,
́
(15) Dimitrijevic, E.; Taylor, M. S. Chem. Sci. 2013, 4, 3298−3303.
1825−1832. (b) Paulsen, H.; Hadamczyk, D.; Kutschker, W.; Bunsch, A.
̈
(16) Crich, D.; de la Mora, M.; Vinod, A. U. J. Org. Chem. 2003, 68,
8142−8148.
Liebigs Ann. Chem. 1985, 1985, 129−141. (c) Cruzado, C.; Bernabe, M.;
Martin-Lomas, M. Carbohydr. Res. 1990, 203, 296−301. (d) Garregg, P.
J.; Maloisel, J.-L.; Oscarson, S. Synthesis 1995, 1995, 409−414.
(e) Oshima, K.; Aoyama, Y. J. Am. Chem. Soc. 1999, 121, 2315−2316.
(f) Kaji, E.; Harita, N. Tetrahedron Lett. 2000, 41, 53−56. (g) Kaji, E.;
Shibayama, K.; In, K. Tetrahedron Lett. 2003, 44, 4881−4885. (h) Wu,
B.; Li, M.; O’Doherty, G. A. Org. Lett. 2010, 12, 5466−5469. (i) Maggi,
A.; Madsen, R. Eur. J. Org. Chem. 2013, 2013, 2683−2691. (j) Pelletier,
G.; Zwicker, A.; Allen, C. L.; Schepartz, A.; Miller, S. J. J. Am. Chem. Soc.
2016, 138, 3175−3182.
(17) Islam, M.; Gayatri, G.; Hotha, S. J. Org. Chem. 2015, 80, 7937−
7945.
(18) (a) Perrin, C. L.; Dwyer, T. J. Chem. Rev. 1990, 90, 935−967.
(b) Kumar, A.; Wagner, G.; Ernst, R. R.; Wuethrich, K. J. Am. Chem. Soc.
1981, 103, 3654−3658. (c) Haindl, M. H.; Hioe, J.; Gschwind, R. J. Am.
Chem. Soc. 2015, 137, 12835−12842.
(19) (a) Horenstein, N. A. Adv. Phys. Org. Chem. 2006, 41, 275−314.
(b) Crich, D. J. Org. Chem. 2011, 76, 9193−9209. (c) Krumper, J. R.;
Salamant, W. A.; Woerpel, K. A. Org. Lett. 2008, 10, 4907−4910.
(d) Wurst, J. M.; Liu, G.; Tan, D. S. J. Am. Chem. Soc. 2011, 133, 7916−
(3) (a) Gouliaras, C.; Lee, D.; Chan, L.; Taylor, M. S. J. Am. Chem. Soc.
2011, 133, 13926−13929. (b) Muramatsu, W.; Yoshimatsu, H. Adv.
Synth. Catal. 2013, 355, 2518−2524. (c) Bajaj, S. O.; Sharif, E. U.;
Akhmedov, N. G.; O’Doherty, G. A. Chem. Sci. 2014, 5, 2230−2234.
(4) (a) Mulani, S. K.; Hung, W.-C.; Ingle, A. B.; Shiau, K.-S.; Mong, K.-
K. T. Org. Biomol. Chem. 2014, 12, 1184−1197 and references cited
therein. (b) Mukaiyama, T.; Matsubara, K. Chem. Lett. 1992, 21, 1041−
1044. (c) Fukase, K.; Hasuoka, A.; Kinoshita, I.; Aoki, Y.; Kusumoto, S.
Tetrahedron 1995, 51, 4923−4932.
́
7925. (e) Huang, M.; Garrett, G. E.; Birlirakis, N.; Bohe, L.; Pratt, D. A.;
Crich, D. Nat. Chem. 2012, 4, 663−667. (f) Chan, J.; Tang, A.; Bennet,
A. J. J. Am. Chem. Soc. 2012, 134, 1212−1220. (g) Huang, M.; Retailleau,
́
P.; Bohe, L.; Crich, D. J. Am. Chem. Soc. 2012, 134, 14746−14749.
(h) Chan, J.; Sannikova, N.; Tang, A.; Bennet, A. J. J. Am. Chem. Soc.
2014, 136, 12225−12228. (i) Bohe, L.; Crich, D. Carbohydr. Res. 2015,
́
403, 48−59.
(20) (a) Wulff, G.; Rohle, G. Chem. Ber. 1972, 105, 1122−1132.
̈
(5) (a) Lemieux, R. U.; Hendricks, K. B.; Stick, R. V.; James, K. J. Am.
Chem. Soc. 1975, 97, 4056−4062. (b) Reisman, S. E.; Doyle, A. G.;
Jacobsen, E. N. J. Am. Chem. Soc. 2008, 130, 7198−7199. (c) Mensah, E.
A.; Nguyen, H. M. J. Am. Chem. Soc. 2009, 131, 8778−8780. (d) Cox, D.
J.; Smith, M. D.; Fairbanks, A. J. Org. Lett. 2010, 12, 1452−1455.
(e) Geng, Y.; Kumar, A.; Faidallah, H. M.; Albar, H. A.; Mhkalid, I. A.;
Schmidt, R. R. Angew. Chem., Int. Ed. 2013, 52, 10089−10092. (f) Beale,
T. M.; Moon, P. J.; Taylor, M. S. Org. Lett. 2014, 16, 3604−3607.
(g) Peng, P.; Schmidt, R. R. J. Am. Chem. Soc. 2015, 137, 12653−12659.
(h) Tanaka, M.; Nashida, J.; Takahashi, D.; Toshima, K. Org. Lett. 2016,
18, 2288−2291.
(b) Bowden, T.; Garegg, P. J.; Maloisel, J.-L.; Konradsson, P. Isr. J. Chem.
2001, 40, 271−277. (c) Fang, T.; Gu, Y.; Huang, W.; Boons, G.-J. J. Am.
Chem. Soc. 2016, 138, 3002−3011.
(21) (a) Blackmond, D. G. Angew. Chem., Int. Ed. 2005, 44, 4302−
4320. (b) Blackmond, D. G. J. Am. Chem. Soc. 2015, 137, 10852−10866.
(22) Bateman, L. C.; Church, M. G.; Hughes, E. D.; Ingold, C. K.;
Taher, N. A. J. Chem. Soc. 1940, 979−1011.
(23) (a) Kubo, Y.; Kobayashi, A.; Ishida, T.; Misawa, Y.; James, T. D.
Chem. Commun. 2005, 2846−2848. (b) Wan, W.-M.; Chen, F.; Jakle, F.
̈
Angew. Chem., Int. Ed. 2014, 53, 8934−8938. (c) Mar
́
tinez-Aguirre, M.
A.; Yatsimirsky, A. K. J. Org. Chem. 2015, 80, 4985−4993.
́
(6) (a) Breton, C.; Snajdrova, L.; Jeanneau, C.; Koca, J.; Imberty, A.
Glycobiology 2006, 16, 29−37R. (b) Hancock, S. M.; Vaughan, M. D.;
Withers, S. G. Curr. Opin. Chem. Biol. 2006, 10, 509−519. (c) Schmaltz,
R. M.; Hanson, S. R.; Wong, C.-H. Chem. Rev. 2011, 111, 4259−4307.
(7) (a) Beale, T. M.; Taylor, M. S. Org. Lett. 2013, 15, 1358−1361.
(b) Mancini, R. S.; McClary, C. A.; Anthonipillai, S.; Taylor, M. S. J. Org.
Chem. 2015, 80, 8501−8510. (c) Li, S.-P.; Hsiao, W.-C.; Yu, C.-C.;
Chien, W.-T.; Lin, H.-J.; Huang, L.-D.; Lin, C.-H.; Wu, W.-L.; Wu, S.-H.;
Lin, C.-C. Adv. Synth. Catal. 2014, 356, 3199−3213.
(8) Mootoo, D. R.; Konradsson, P.; Ududong, U.; Fraser-Reid, B. J.
Am. Chem. Soc. 1988, 110, 5583−5584.
(9) (a) Kronzer, F. J.; Schuerch, C. Carbohydr. Res. 1973, 27, 379−390.
(b) Eby, R.; Schuerch, C. Carbohydr. Res. 1974, 34, 79−90.
(c) Machinami, T.; Suami, T. Chem. Lett. 1974, 1177−1180.
(d) Koto, S.; Hamada, Y.; Zen, S. Chem. Lett. 1975, 587−588.
(e) Leroux, J.; Perlin, A. S. Carbohydr. Res. 1978, 67, 163−178. (f) Koto,
S.; Sato, T.; Morishima, N.; Zen, S. Bull. Chem. Soc. Jpn. 1980, 53, 1761−
I
DOI: 10.1021/jacs.6b06943
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX