An Empirical Understanding of the Glycosylation Reaction

Article abstract of DOI:10.1021/jacs.8b04525

Reliable glycosylation reactions that allow for the stereo- and regioselective installation of glycosidic linkages are paramount to the chemical synthesis of glycan chains. The stereoselectivity of glycosylations is exceedingly difficult to control due to the reaction's high degree of sensitivity and its shifting, simultaneous mechanistic pathways that are controlled by variables of unknown degree of influence, dominance, or interdependency. An automated platform was devised to quickly, reproducibly, and systematically screen glycosylations and thereby address this fundamental problem. Thirteen variables were investigated in as isolated a manner as possible, to identify and quantify inherent preferences of electrophilic glycosylating agents (glycosyl donors) and nucleophiles (glycosyl acceptors). Ways to enhance, suppress, or even override these preferences using judicious environmental conditions were discovered. Glycosylations involving two specific partners can be tuned to produce either 11:1 selectivity of one stereoisomer or 9:1 of the other by merely changing the reaction conditions.

Full text of DOI:10.1021/jacs.8b04525

Subscriber access provided by Technical University of Munich University Library
Article
An Empirical Understanding of the Glycosylation Reaction
Sourav Chatterjee, Sooyeon Moon, Felix Hentschel, Kerry Gilmore, and Peter H Seeberger
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.8b04525 • Publication Date (Web): 20 Aug 2018
Downloaded from http://pubs.acs.org on August 20, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 18
Journal of the American Chemical Society
1
2
3
4
5
An Empirical Understanding of the Glycosylation Reaction
6
7
8
Sourav Chatterjee,[a]† Sooyeon Moon, [a,b]† Felix Hentschel, [a] Kerry Gilmore,*[a] Peter H.
Seeberger*[a,b]
9
a
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Department of Biomolecular Systems, Max-Planck-Institute of Colloids and Interfaces, Am
Mühlenberg 1, 14476 Potsdam, Germany
b Freie Universität Berlin, Institute of Chemistry and Biochemistry, Arnimallee 22, 14195 Berlin,
Germany
Abstract:
Reliable glycosylation reactions that allow for the stereoꢀ and regioselective installation of
glycosidic linkages are paramount to the chemical synthesis of glycan chains. The
stereoselectivity of glycosylations is exceedingly difficult to control due to the reaction’s high
degree of sensitivity and its shifting, simultaneous mechanistic pathways that are controlled by
variables of unknown degree of influence, dominance, or interdependency. An automated
platform was devised to quickly, reproducibly, and systematically screen glycosylations and
thereby address this fundamental problem. Thirteen variables were investigated in as isolated a
manner as possible, to identify and quantify inherent preferences of electrophilic glycosylating
agents (donors) and nucleophiles (glycosyl acceptors). Ways to enhance, suppress, or even
override these preferences using judicious environmental conditions were discovered.
Glycosylations involving two specific partners can be tuned to produce either 11:1 selectivity of
one stereoisomer or 9:1 of the other by merely changing the reaction conditions.
Introduction
Polysaccharides are a dominant class of biopolymers that play central roles in structure,
energy storage, and biological functions.1 The chemical assembly of polysaccharides is
conceptually simple as it relies on only a single type of bond forming reaction: the coupling of
the electrophilic glycosylating agent (“glycosyl donor”) with a nucleophile (“glycosyl acceptor”).
Polysaccharides are more complex in their composition and structure than oligonucleotides or
proteins. As such, the synthesis of glycans is very sensitive and relies on many diverse building
blocks accounting for variants of hydroxyl substitution and stereochemistry on the pyran/furan
core as well as multiple sites of connectivity for the creation of nonꢀlinear and branched
structures. The number of possible structures is further significantly increased by the fact that
each glycosidic bond generates a stereogenic center where two diastereomers (alpha or beta) can
be formed (Figure 1).
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 18
1
2
3
4
5
Figure 1: Three types of structural parameters contribute to the inherent complexity of glycans.
Figure adapted from ref 2.
6
7
8
9
The reaction mechanism of glycosidic bond formation is complex, often making
controlled, selective bond formation elusive (Figure 2).3 The initial key step of the reaction is the
cleavage of the bond between the anomeric carbon and the anomeric leaving group upon reaction
with the activator.4 Thereby, a cationic5 intermediate is formed that can be engaged by the
acceptor in a true SN1ꢀtype mechanism. The bondꢀforming reaction by attack of a nucleophile can
either result in the formation of the more thermodynamically stable αꢀproduct6 or via the
potentially more accessible “top face” of the donorꢀintermediate to give the βꢀproduct.7 However,
the carbocation can also form a close ion pair or a series of highly reactive glycoside
intermediates from the reaction of part of the activator (conjugate base or byproduct).8 In these
situations, the outcome of the glycosylation may resemble a SN2ꢀtype reaction.9 These
intermediates – whose identity, relative ratios, kinetics, and favorabilities are essential to
understand the mechanism – are often extremely difficult or even impossible to ascertain.10
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2: Glycosidic bond formation entails complex mechanistic pathways.
This is compounded by a lack of understanding regarding the underlying factors which
have the potential to control the stereochemical outcome, both those dictated by the coupling
partners (permanent) and the conditions chosen for the transformation (environmental, Figure 3).
The donor11 contains five stereocenters, four of which can hyperconjugatively influence the
activation as well as the stability and conformations of intermediates.12,13 Each of these four
positions are also generally protected by groups capable of directly6,14 or indirectly influencing
the reaction15 by donating (“arm”) or withdrawing (“disarm”) electrons from the glycosylating
agent,16 or by restricting conformational freedom via bridged structures to influence reactivity17
and stereochemical outcome.18 The C1 leaving group chosen requires activation to induce
cleavage, creating neutral or charged byproducts.19 With respect to the acceptor, the
nucleophilicities of primary and secondary alcohols (the vast majority of glycosylations are CꢀO
couplings) can depend on the orientation (axial/equatorial) of the alcohol, the adjacent protecting
groups,20 and by structural changes induced by conformational locking21 or electronic
modification.22 The sterics surrounding the nucleophile also will influence the rate of reaction
and thus potentially the stereochemical outcome.
ACS Paragon Plus Environment

Page 3 of 18
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3: A selection of the permanent and environmental factors of glycosylation reactions,
with the specific examples examined in this study, as well as how these factors potentially
influence the stereoselectivity.
The mechanistic path these coupling partners follow, as well as the intermediates which
are formed, will be strongly influenced by the reaction environment. The physical properties of a
solvent can promote bond cleavage, stabilize intermediates, and influence reaction pathways. The
activator bears potential influencing factors resulting from interactions with the conjugate base,
either via contactꢀion pairs or covalent intermediates such as αꢀtriflates.10a,23 The temperature will
impact the stability of intermediates, the reaction pathways followed, and product composition.
Finally, traditional reaction parameters such as concentration,24 stoichiometry,25 reaction time,
and mixing26 can have an impact on the efficiency of the reaction as well as the resulting
stereoselectivity.
Carbohydrate chemistry has thus remained a conceptually simple but technically very
demanding field, as optimization is required for every new reaction with strict maintenance of all
reaction parameters. Additionally, the lack of understanding of what influence these factors have
on stereochemical outcomes has left the selection of initial reaction parameters – such as the
choice of anomeric leaving group, protecting groups, solvent, temperature, and activating agent –
often more of a matter of personal preference due to experience than knowledgeꢀbased decision
making, leaving the field as much an art as a science.
In an effort to demystify glycosylation reactions and guide synthetic chemists towards
optimal reaction conditions without prior reaction optimization, general guidelines regarding the
selection of appropriate reaction conditions based on the intrinsic preferences of the coupling
partners are urgently needed. However, before these can be determined, each of the factors
influencing glcosylation reactions must first be identified and categorized before interrogatory
experiments are devised and performed in a controlled, reproducible environment in as isolated a
manner as possible. This reproducibility can be achieved using minimal amounts of material in
microreactor flow chemistry setups.27 Herein, we describe the development and utilization of an
automated microreactor/HPLC platform capable of varying reactant ratios, temperature, and
reaction time autonomously. This instrument was used for the rapid and reproducible isolation
and interrogation of variables influencing the stereoselectivity, providing an unprecedented,
systematic, and quantifiable view of glysosidic bond formation from 270 experiments. The
identification of, and factors providing control over, specific intermediates of the glycosylation
reaction is outside the scope of this work.
Instrument and Experimental Design
The heart of the reaction optimizer platform is a 78 ꢀL silicon microreactor comprised of
a separate addition port for donor, acceptor, and activator.28 Each of these lines is driven by two
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 18
1
2
3
4
5
6
7
8
9
syringe pumps for reagent and solvent, allowing for both line purging and changes in
concentration in any of the reagents (Figure 4). The combined solution passes through a mixing
zone before entering the reaction zone. A quench is added prior to the reaction mixture exiting
the chip to ensure the reaction is only occurring at the desired temperature. Once steady state is
reached after ten reactor volumes, a sample is automatically removed via a 1 ꢀL HPLC injection
loop. The temperature in the chip is maintained reliably within a range from ꢀ55 to +70 °C within
±0.1 °C by a surrounding aluminum block cooled or heated using a thermostat.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4: Automated glycosylation instrument consisting of three sections: multipart reaction
section, HPLC analysis, and automation. For complete system details, see supporting
information.
With automated system in hand, a model system was chosen to investigate the factors
influencing the stereoselectivity of glycosylations (Figure 3). Three pyran cores – glucose,
mannose, and galactose – were chosen as donors to probe the effects of the C2 and C4 positions,
respectively. All hydroxyl groups, except at the anomeric position, were protected as nonꢀ
participatory benzyl ethers. Three leaving groups – trichloroacetimidate, ethylthioether, and n-
dibutylphosphate – were investigated. However, due to the similar stereochemical outcomes
exhibited (vide infra), the more reactive glycosyl trichloroacetimidates were utilized primarily,
activated with triflic acid rather than trimethylsilyltriflate (TMSOTf) to limit the potential
reaction pathways. Methanol, ethanol, 2,2ꢀdifluoroethanol, 2,2,2ꢀtrifluoroethanol, isopropanol,
and tertꢀbutanol served as model acceptors to explore a range of steric and electronic effects.29
The temperatures investigated were limited by the reactivity of the donor and the physical
properties of the solvents, kept under the respective boiling points.
The general glycosylation procedure required that the requisite benzyl protected donors be
dried under high vacuum overnight prior to use and that all solvents, activators, and acceptors
contained less than 3 ppm water. Glass syringes containing solutions of the donor (50ꢀ110 mM),
acceptor (60ꢀ110 mM), and the activator (22.6ꢀ120 mM) feed lines were diluted as desired via
accompanying syringe pumps prior to mixing in the reactor. The reaction temperature, reagent
concentration and stoichiometry, as well as the residence time were all set with the aid of the
developed software. After reaction completion and inline quenching with pyridine, the reaction
products were monitored by an automated injection into the online HPLC.
The results discussed below are based on those among the 270 glycosylation reactions
that exhibited yields greater than 60% product (for a complete list, see the supporting
information). The results of these investigations are discussed as isolated factors with respect to
the change in temperature, starting with the permanent factors, followed by environmental.
ACS Paragon Plus Environment

Page 5 of 18
Journal of the American Chemical Society
1
2
3
4
Results and Discussion:
5
6
7
8
Permanent: Donor Leaving Group (C1): Over the past century, a host of different leaving groups
and corresponding activators to induce anomeric cleavage have been introduced as part of the
quest for mild, selective, and highꢀyielding glycosylation reactions.19 To explore the effects these
different groups have on the stereochemical outcome of glycosylation, a fully benzylated
glucosyl αꢀtrichloroacetimidate (Schmidt donor), glucosyl βꢀethanethio ether, or glucosyl αꢀn-
dibutylphosphate were reacted with isopropanol in DCM at temperatures ranging from ꢀ50 →
+30 °C. The activation conditions for each leaving group were chosen to minimize differences in
the conjugate bases/byproducts present in the solution, keeping triflate present in all cases. All
other variables were kept constant.
Glycosylating agents with the three leaving groups gave nearly identical stereochemical
outcomes under these conditions and differed mainly in conversion and yield at low temperatures
(Figure 5). Glycosylation yields for thioglycosides dropped rapidly from 88% at 10 °C to 45% at
–10 °C. A similar situation is observed for the phosphate (60% at ꢀ10 °C to 30% at ꢀ30 °C). At
these lower conversions, a slight increase (< 7%) in βꢀselectivity is observed as compared to the
trichloroacetimidate donor (see SI).
Due to the similar stereochemical outcomes but increased reactivity range, the remaining
studies were performed using the glycosyltrichloroacetimidates. Unexpectedly, stereoselectivity
and temperature are related in a nearly linear fashion in DCM for many of the coupling partners
examined. This linearity is lost, however, in certain environmental conditions (vide infra). Over
the range examined (–50 °C to +30 °C), the selectivity of the glucose/isopropanol coupling
ranges from favoring the βꢀproduct (73%) to the αꢀproduct (39%). This corresponds to a
temperature sensitivity (the slope of the plotted data) of 0.41% per °C. These values serve as a
comparison benchmark for all other variables examined herein. While the stereochemistry of
some trichloroacetimidate donors, when reacted with TMSOTf, has been shown to have an
influence on the stereochemical outcome of glycosylations,30 under our standard conditions in
DCM, no difference was observed between the αꢀ and βꢀglucose donor (Figure S17 in SI).31
Implication: The different leaving groups investigated, under the conditions explored,
have little influence on the stereochemical outcome of glycosylations. The relative stabilities and
desired reaction temperature are thus the key considerations when selecting the leaving group.
The strongest influencing factor associated with the leaving group is the means of activation,
particularly the conjugate base that is generated (vide infra). Coordination effects of the different
neutral species formed upon decomposition of activated donor cannot be discounted, but
appeared to be minimal under the conditions that were investigated.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 18
1
2
3
4
5
BnO
BnO
HO
O
BnO
BnO
Activator: Variable
O
BnO
BnO
+
O
Tres: 45 seconds
Temp: Variable
Solvent: DCM
BnO
LG
OBn
1 equiv.
1.2 equiv.
LG:
Variable
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5: Comparison of stereoselectivities for glycosylations of glucose, bearing one of three
leaving groups, with iPrOH as acceptor in DCM. For full experimental details, see SI. Figure
code: Glucose (▲); Trichloroacetimidate (blue) with TfOH (0.2 equiv); ethyl thioether (red) with
TfOH (0.2 equiv) and Nꢀiodosuccinamide (1.2 equiv); nꢀbutylphosphate (green) with TMSOTf
(1.2 equiv).
Donor Stereochemistry (C2 and C4): Donor activity and intermediate stability can both be
influenced by throughꢀbond or throughꢀspace hyperconjugation12 of the ether groups of the pyran
core.32 With nonꢀparticipating groups, the influence of the C2 position has been shown to be
significant in conformationally locked33 and unlocked glucose/mannose derivatives,34 as well as
for less common derivatives such as glucoseꢀ/mannoseamine and the C2 fluorinated derivatives.35
Comparing the coupling of isopropanol with the αꢀglucosyl and mannosyl
trichloroacetimidates in DCM, a significant (34%) decrease in temperature sensitivity is observed
when the C2 benzyl ether is axial (mannose, Figure 6). The less sensitive mannose slightly favors
the formation of the αꢀproduct, with α:β ratios ranging from 48:52 (ꢀ50 °C) to 61:39 (30 °C).
Monosaccharides differing with respect to the C4 position, galactose (axial C4 ether) and
glucose (equatorial C4 ether), exhibit similar temperature sensitivities (Tsens = 0.43%/°C)
although galactose is 1.13 times more likely to give the βꢀproduct (9% more βꢀproduct formed)
than glucose, ranging from 81% βꢀselectivity at ꢀ50 °C to 49% at +30 °C (Figure 6).
Implication: Glycosylating agents display inherent preferences concerning mechanistic pathways
and stereoselectivity. Glucose favors βꢀproduct formation, particularly at low temperatures,
exhibiting a moderate degree of temperature sensitivity and follows a more SN2ꢀlike reaction
pathway. The preference for βꢀproduct formation is increased when the C4ꢀether is axial
(galactose), and in DCM exhibits a similar SN2ꢀlike mechanism to glucose. The C2 position is
significantly more influential, and mannose has an inherent preference for the formation of αꢀ
product, proceeding via a more SN1ꢀlike pathway. These inherent preferences can be enhanced or
overridden by the other reaction variables (vide infra).
ACS Paragon Plus Environment

Page 7 of 18
Journal of the American Chemical Society
1
2
3
4
OBn
OBn
O
BnO
BnO
OH
TfOH (0.2 equiv.)
O
O
BnO
BnO
+
BnO
Tres: 45 seconds
Temp: Variable
Solvent: DCM
O
NH
CCl3
OBn
5
6
7
8
1.2 equiv.
1 equiv.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 6: Comparison of the stereochemical outcome of three benzylated glycosyl
trichloroacetimidates reacting with isopropanol and TfOH. For full experimental details, see SI.
Figure code: Glucose (▲); Galactose (■); Mannose (●); DCM (blue).
Acceptor Sterics and Electronics: To probe the factors that potentially influence the acceptor, a
series of simple alcohols differing in both sterics36 and electronics (methanol, ethanol,
isopropanol, tertꢀbutanol, 2,2ꢀdifluoroethanol, 2,2,2ꢀtrifluoroethanol) were reacted with the
glucosyl, galactosyl, and mannosyl trichloroacetmidate donors in DCM (Figure 7). Beginning
with the “neutral” glucoside and the electronically “active” alcohols, a clear trend is observed in
the series tꢀBuOH/iPrOH/EtOH, with average stepwise increases of 8.2% and 6.7% in the βꢀ
selectivity at a given temperature, respectively, while a near identical rate of change with respect
to temperature is maintained (Figure 7a). However, the least sterically encumbered alcohol,
methanol, exhibits a 38% decrease in average temperature sensitivity as compared to ethanol (ꢀ
0.3 %β/°C vs ꢀ0.48 %β/°C) and selectivities ranging from 71%β at ꢀ50 °C to 47%β at 30 °C.
The “active” acceptor series all exhibit higher βꢀselectivities with galactose compared
with glucose, as was observed with isopropanol (Figure 7b). While the βꢀselectivity increases in
the t-BuOH to EtOH series (4.2% and 4.8%), the temperature sensitivity is not constant as it is in
the glucose series. With increasing sterics, the rate of change with respect to temperature
decreases, from ꢀ0.51 %β/°C (EtOH) to ꢀ0.4 %β/°C (iPrOH) to ꢀ0.27 %β/°C (tꢀBuOH). The
coupling of galactosyl donor and methanol is more temperature sensitive than the rest of the
series, with the βꢀselectivity ranging from 86% at ꢀ50 °C to 41% at 30 °C (Tsens = ꢀ0.56%β/°C).
The coupling of mannose with isopropanol was much less temperature sensitive than for
glucose and galactose (Figure 6). The stereoselectivity of mannose couplings is also much less
sensitive to substitution of the acceptor (Figure 7c). Methanol, ethanol, and isopropanol all
behave similarly with average selectivities ranging from ꢀ51% β at ꢀ50 °C to 37% at 30 °C
(average Tsens = ꢀ0.17%β/°C) and only with the increased t-BuOH sterics was a deviation
observed. t-BuOH behaves similar to the other acceptors below 10 °C, however with a stepwise
decrease of ~5% in βꢀselectivity. Above 10 °C, a rapid change is observed, reaching 95%
selectivity for the αꢀproduct at 30 °C.
Turning to the electronics of the acceptor, the insertion of electronꢀwithdrawing fluorines
beta to the nucleophilic oxygen has a pronounced effect on the stereoselectivity, favoring the
formation of the αꢀproduct (Figure 7d). When galactose37 was coupled to 2,2ꢀdifluoroethanol, the
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 18
1
2
3
4
5
6
temperature sensitivity drops 70% compared to ethanol (Tsens = ꢀ0.15 vs ꢀ0.51%β/°C) and more αꢀ
product is formed overall, with β product formed ranging from 52% ꢀ50 °C to 41% at 30 °C.
Mannose, which already favors the formation of the αꢀproduct, exhibits complete αꢀselectivity
when coupled with 2,2,2ꢀtrifluoroethanol.
7
8
9
Implication: A fine balance exists concerning the influential dominance between the inherent
preferences of the donor and those of the acceptor. For donors favoring SN2ꢀlike pathways such
as glucose and galactose, changes in acceptor nucleophilicity have a pronounced effect on the
observed stereoselectivity, with stronger nucleophiles favoring the βꢀproduct at low temperatures
with good levels of temperature sensitivity. The influence of the C2 position, favoring αꢀproduct
formation via an SN1ꢀtype mechanism, overrides subtle differences in the nucleophile, and only
major changes in nucleophilicity result in significant modifications to stereoselectivity. With a
large decrease in the nucleophilicity (CF3CH2OH vs CH3CH2OH), the inherent favorability of
mannose (αꢀformation), coupled with that of the lowꢀnucleophilic acceptor, results exclusively in
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
the αꢀproduct.38
OBn
OBn
BnO
BnO
OBn
BnO
BnO
OBn
O
O
BnO
BnO
Acceptor: TfOH (0.2 equiv.)
O
Acceptor: TfOH (0.2 equiv.)
O
BnO
BnO
+
+
Variable
Variable
O
O
BnO
Tres: 45 seconds
BnO
Tres: 45 seconds
R3
R1 R2
R3
OBn
R1 R2
O
NH
CCl3
O
NH
CCl3
OBn
Temp: Variable
Solvent: DCM
Temp: Variable
Solvent: DCM
1.2 equiv.
1 equiv.
1.2 equiv.
1 equiv.
ACS Paragon Plus Environment

Page 9 of 18
Journal of the American Chemical Society
1
2
3
4
5
BnO
BnO
OBn
O
BnO
R1
BnO
OBn
O
R1
HO
BnO
BnO
O
BnO
BnO
Activator: TfOH
Acceptor: TfOH (0.2 equiv.)
O
R1
R2
BnO
BnO
BnO
BnO
+
+
R1
Variable
O
O
Tres: 45 seconds
Temp: Variable
Solvent: DCM
Tres: 45 seconds
R3
R1 R2
BnO
O
NH
CCl3
O
CCl3
1 equiv.
R2
Temp: Variable
Solvent: DCM
OBn
1.2 equiv.
1 equiv.
1.2 equiv.
NH
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 7: Comparison of different activators with three benzylated glycosyl
trichloroacetimidates. For full experimental details, see SI. Figure code: Glucose (▲); Galactose
(■); Mannose (●); MeOH (blue); EtOH (red); iPrOH (green); t-BuOH (orange); CF2HCH2OH
(light purple, R1 = F, R2= H); CF3CH2OH (dark purple, R1 = R2= F).
Environmental: The glycosylation conditions can enhance, diminish, or even override the
intrinsic selectivities of a given donor/acceptor pair. Five factors were determined to strongly
influence the reaction: temperature, stoichiometry, activator, presence of water, and the solvent.
The other factors examined exhibited only minor influence (see SI for details).
Temperature: The strongest influence on α/β selectivity is temperature. At low temperatures, βꢀ
product formation is favored, presumably due to an increased proportion of SN2ꢀlike pathways
from activated donors, favorable intermediate conformations, or species such as αꢀtriflates. The
temperature influence can be overridden when the SN1 pathway becomes dominant.
Implication: The degree of variance observed in the stereoselective outcome of a glycosylation as
a function of temperature28 emphasizes the need for precise temperature control. Temperature
variation provides the most straightforward means of modifying the stereoselectivity. The clear
temperature dependency breaks down in other solvents, presenting distinct mechanistic zones
(vide infra).
Stoichiometry: Reagent stoichiometry should matter for processes that involve SN2ꢀlike
mechanisms. Often, the donor is used in excess in a glycosylation39 to ensure high yields when
poor nucleophiles are used or when other side reactions consume the activated donor. However,
additional equivalents of acceptor can influence not only yield, but also selectivity.25
To investigate the effect of acceptor stoichiometry on stereoselective outcome, the
perbenzylated galactosyl trichloroacetimidate was reacted with the poorly nucleophilic 2,2ꢀ
difluoroethanol. With only one equivalent acceptor, the process is temperature independent,
favoring αꢀformation (73%). At 30 °C, little change is observed when five or ten equivalents
acceptor are added. At lower temperatures, however, increased amounts of acceptor result in an
increased amount of the βꢀproduct formed (at a rate of 0.15 and 0.22%β/°C) resulting in increases
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 18
1
2
3
4
5
6
7
8
9
from 30% βꢀproduct (1 equiv.) to 42% and 50% βꢀdiastereomer for one and ten equivalents,
respectively at ꢀ50 °C (Figure 8, left).
This effect is not dominant. While the stoichiometry of the acceptor has been shown to
affect the selectivity in competition experiments with conformationally locked mannose donors,29
the perbenzylated mannosyl trichloroacetimidate exhibited little difference with one and five
equivalents of isopropanol in toluene, a 4.4% increase in βꢀproduct at elevated temperatures
(Figure 8b).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Implication: For donors with weaker inherent preferences such as glucose and galactose that
favor the SN2 pathway, the formation of βꢀproduct can be enhanced by increasing the equivalents
of the acceptor. Marked increases can be seen even for poor nucleophiles, and larger differences
are expected for stronger nucleophiles: a 26% increase in the βꢀselectivity is observed when ten
equivalents of methanol were reacted with galactosyl trichloroacetimidate at 30 °C as compared
to the 0.8 equivalents used in our standard experiments (see SI). However, the strongly
influencing axial C2 group in mannose overrides this factor, resulting in little difference between
one and five equivalents of isopropanol over a wide temperature range.
BnO
BnO
OBn
OBn
O
F
BnO
OBn
BnO
BnO
O
HO
OBn
O
BnO
BnO
HO
TfOH (0.2 equiv.)
TfOH (0.2 equiv.)
BnO
O
F
F
BnO
BnO
+
+
O
O
BnO
Tres: 45 seconds
Temp: Variable
Tres: 45 seconds
Temp: Variable
O
NH
CCl3
O
NH
CCl3
Equiv:
Variable
F
Equiv:
Variable
OBn
1.2 equiv.
1.2 equiv.
DCM
Toluene
Solvent:
Solvent:
Figure 8: Comparison of acceptor stoichiometry in the reaction of galactosyl trichloroacetimidate
in DCM and the mannosyl trichloroacetimidate in toluene. For full experimental details, see SI.
Figure code: Galactose (■); Mannose (●); Donor:Acceptor 1:1 (blue); 1:5 (red); 1:10 (green). For
mannose, the ratio of Donor:Acceptor in the blue line is 1:0.8.
Activator: The activator has multiple potential roles in a glycosylation: the native species
activates the leaving group, while the conjugate base stabilizes charged intermediates and may
even trap these intermediates by reversible covalent bond formation. To investigate what
influence the activator, and in particular the conjugate base of the activator, exerts on
stereoselectivity, a range of substoichiometric acids were reacted with perbenzylated mannose
and glucose donors using tertꢀbutanol as acceptor in DCM. When triflic acid (blue)40 is
exchanged for fluorosulfuric (green) or methanesulfuric acid (orange), a loss in temperature
sensitivity is observed with mannose, giving an approximate 3:2 ratio of α:β (Figure 9a).
Alternatively, when the less electronꢀrich triflimide (Tf2NH) is used, a significant change is
ACS Paragon Plus Environment

Page 11 of 18
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
observed. Here, temperature sensitivity is observed at lower temperatures (approx. ꢀ10 °C vs. +10
°C with TfOH) and quickly approaches near complete αꢀselectivity above +10 °C. This dataset
represents the most rapid change of stereoselectivity with respect to temperature of all conditions
examined. Interestingly, this finding differs from previous observations for the conformationally
locked mannosyl trichloroacetimidate, which favored βꢀproduct formation using Tf2NH as
activator.41
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In glucose the situation is different. With TfOH, tertꢀbutanol exhibits poor selectivity
ranging from 60% βꢀproduct at ꢀ50 °C to 40% at 30 °C. When the activator is changed to Tf2NH,
the reaction is almost completely βꢀselective (95%) at ꢀ50 °C and is nearly 1.5 times more
sensitive to temperature (Figure 9b).
Implication: Through modification of the nucleophilicity of the conjugate base of the acceptor,
the inherent stereoselective preference of the donor can be enhanced. Thus, by changing from
TfOH to Tf2NH – both equally capable of activating the trichloroacetimidate donor – the
mannose donor gives more of the αꢀproduct and breaks way from the SN1ꢀlike pathway at lower
temperatures. For glucose, the same change in acceptor sees a large increase in βꢀproduct
formation and increased SN2ꢀlike behavior. This factor appears to have a high degree of
dominance in glycosylation reactions.
BnO
BnO
OBn
O
BnO
BnO
Activator: Variable
(0.2 equiv.)
Activator: Variable
(0.2 equiv.)
O
OBn
O
BnO
BnO
BnO
BnO
OH
OH
O
BnO
BnO
BnO
BnO
+
+
O
O
BnO
Tres: 45 seconds
Temp: Variable
Solvent: DCM
Tres: 45 seconds
Temp: Variable
Solvent: DCM
O
NH
CCl3
O
NH
CCl3
OBn
1.2 equiv.
1 equiv.
1.2 equiv.
1 equiv.
Figure 9: a) Comparison of different activators on the coupling of mannosyl trichloroacetimidate
with t-BuOH. b) Tf2NH enhances the inherent stereoselective favorabilities of donors. For full
experimental details, see SI. Figure code: Glucose (▲); Mannose (●); Tf2NH (red); TfOH (blue);
MsOH (orange); FSO3H (green).
Presence of Water: Water is detrimental to glycosylation reactions in organic solvents, as it
efficiently traps the activated donor and/or intermediates to give the hydrolyzed donor, which
cannot be reactivated under standard glycosylation conditions. Here, all solvents were dried to
less than 3 ppm water (see SI) to minimize this unproductive pathway.
However, it was found that even substoichiometric amounts of water can have a
significant influence not on the yield, but rather on the stereoselectivity of glycosylations. Under
the standard “anhydrous” conditions, the coupling of mannose and tertꢀbutanol exhibits low
temperature sensitivity until 10 °C, when a rapid change is observed until almost complete αꢀ
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 18
1
2
3
4
5
6
7
8
9
selectivity is achieved at 30 °C (Figure 10). However, in the presence of 0.25 equivalents of
water, the amount present in the solid glycosylating agent prior to the extensive drying procedure,
a complete loss of temperature sensitivity is observed, and the reaction follows a SN1ꢀlike path
over the entire temperature range. No difference was observed in the coupling of glucose and
isopropanol under “anhydrous” conditions or in the presence of 0.25 equivalents of water. The
yields for the two respective coupling conditions for mannose and glucose couplings were
similar.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Implication: High percentages of water present in the glycosylation medium results in the
competitive trapping of intermediates to give the hydrated donor byproduct. However, in
substoichiometric amounts, water promotes the SN1 pathway without significant increases in
byproduct formation. As such, donor/acceptor pairs which favor SN1ꢀlike processes, such as
mannose or electronꢀdeficient acceptors, will be affected by small amounts of water in the
medium, while coupling partners favoring SN2ꢀlike pathways will be less affected.
OBn
OBn
TfOH (0.2 equiv.)
dry/H2O (0.25 equiv.)
O
BnO
BnO
OH
O
BnO
BnO
R
+
O
BnO
Tres: 45 seconds
Temp: Variable
Solvent: DCM
R
O
NH
CCl3
OBn
1 equiv.
R = H, CH3
1.2 equiv.
Figure 10: Comparison of stereoselectivities of glucosyl trichloroacetimidate with
i
PrOH and
mannosyl trichloroacetimidate with ꢀBuOH under “anhydrous” conditions and in the presence of
t
substoichiometric amounts of water. For full experimental details, see SI. Figure code: Glucose
(▲); Mannose (●); DCM (blue); DCM + 0.25 equiv. H2O (light blue, dashed line).
Solvent: The choice of solvent used for glycosylation reactions greatly influences the
stereochemical reaction outcome,42 which can result from solvent coordination or conformer and
counterion distribution.43 A general observation is that αꢀlinkages form preferentially in ether
solvents, while βꢀlinkages are formed in acetonitrile. To investigate how the different solvents
enhance, suppress, or override the other factors influencing glycosylation stereoselectivity, four
solvents were examined: dichloromethane (DCM), acetonitrile (ACN), toluene (Tol), and methyl
tertꢀbutyl ether (MTBE).
Glucose, which bears no inherent influencing groups, is strongly affected by the solvent
choice (Figure 11, top). Compared to the “standard” solvent DCM, a significant increase in the
formation of βꢀproduct is observed with acetonitrile when using isopropanol as acceptor, ranging
from 90%β at ꢀ30 °C to 58%β at 70 °C. The coupling also exhibits a 27% decrease in temperature
ACS Paragon Plus Environment

Page 13 of 18
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
sensitivity versus DCM. While toluene gives a very similar range of selectivities to DCM (81%
→ 38% vs 73% → 39%), two distinct mechanistic pathways appear to be occurring above and
below 10 °C, with SN2ꢀlike favored at lower temperatures and SN1ꢀlike at higher temperatures.
The largest change in stereoselectivity is observed with MTBE. Here, a relatively temperatureꢀ
insensitive process greatly favors αꢀproduct formation (15%β at ꢀ50 °C → 18%β at +50 °C). The
average selectivity across all solvents and temperatures examined for the coupling of
perbenzylated glucose donor and isopropanol using TfOH as activator is nearly equal mixture of
diastereomers (52:48 α:β).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Compared to glucose, the axial C4ꢀOꢀbenzyl ether of galactose increases the inherent
preference for the formation of the βꢀdiastereomer via a SN2ꢀlike pathway, as is reflected across
all solvents examined (Figure 11, middle). Again, acetonitrile enhances this preference, giving
higher βꢀselectivities with decreased temperature sensitivities (50% decrease compared to DCM).
In toluene, the coupling of galactose and isopropanol gives a broad range of selectivities, from
90%β at ꢀ50 °C to only 31%β at 70 °C. Similar to glucose, a selectivity plateau is observed in
toluene as well, albeit at higher temperatures (> 50 °C). While the αꢀproduct is formed
predominantly also in MTBE, the change is not as pronounced as with glucose, and the system
exhibits a similar temperature sensitivity as observed in DCM. The average selectivity across all
solvents and temperatures examined for the coupling of perbenzylated galactose donor and
isopropanol using TfOH as activator is 43:57 (α:β), 1.2 times more selective for the βꢀproduct
compared to glucose.
Mannose has a strong inherent preference to form the αꢀproduct, and this effect is
enhanced in all solvents examined compared to DCM. Near stepwise increases in αꢀproduct
selectivity by 10% are seen in the DCM/toluene/MTBE progression. Similarly low temperature
sensitivities are observed with these solvents, with a slight increase in αꢀselectivity observed in
toluene at higher temperatures (>50 °C). Acetonitrile, while acting to enhance the inherent
preference of the donor, appears to proceed via two distinct mechanisms based on temperature –
similar to what is observed with glucose/galactose in toluene. At lower temperatures, the
selectivity is constant at 35% βꢀproduct formation. Above 10 °C, a rapid increase in temperature
sensitivity (1.1%/°C) is observed as compared to the low temperature data, reaching near
complete αꢀselectivity above 30 °C. The average selectivity across all solvents and temperatures
examined for the coupling of perbenzylated mannose donor and isopropanol using TfOH as
activator is 73:27 (α:β), 1.4 times more selective for the αꢀproduct compared to glucose.
Implication: The solvent has a strong effect on the resulting α/βꢀselectivity of glycosylations.
Acetonitrile enhances the inherent donor preferences, providing higher proportions of the βꢀ
product for glucose/galactose and αꢀproduct for mannose. Toluene also enhances this
favorability, though to a lesser extent. Methyl tertꢀbutyl ether enhances the formation of the αꢀ
product for all donors. Glucose is particularly susceptible; however the major diastereomer is the
αꢀproduct for galactose at temperatures above ꢀ40 °C. Mannose is less strongly influenced, but
does exhibit increased proportions of the αꢀdiastereomer in MTBE.
Certain solvents also appear able to exhibit distinct mechanisms within certain
temperature zones, as evidenced by the sometimes sharp inflection points in the plotted data. In
acetonitrile, the coupling of mannose and isopropanol undergoes a shift between 0 °C and 20 °C,
rapidly changing from SN1ꢀlike to SN2ꢀlike to give near complete selectivity for the αꢀproduct. In
toluene, glucose exhibits a similar mechanistic shift from SN2ꢀlike to SN1ꢀlike at 20 °C and
galactose at 50 °C. Interestingly, the SN1ꢀlike plateau of stereoselectivity occurs in the same
region for all three donors (65:35 α:β, Figure 11d) in the two different solvents, potentially
providing a general value for couplings proceeding via this mechanistic path.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 18
1
2
3
4
OBn
OBn
O
TfOH
BnO
BnO
OH
O
(0.2 equiv.)
BnO
BnO
+
O
Tres: 45 seconds
Temp: Variable
Solvent: Variable
BnO
O
NH
CCl3
OBn
5
6
7
8
1.2 equiv.
1 equiv.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 11: Comparison of four different solvents for the coupling of the three model donors with
iPrOH using TfOH as activator. For full experimental details, see SI. Figure code: Glucose (▲);
Galactose (■); Mannose (●); DCM (blue); Toluene (red); ACN (green); MTBE (orange).
Discussion:
The multitude of factors that influence the stereoselectivity of glycosylations has
traditionally been challenging to identify and disentangle. Reproducibility issues due to the
sensitivity of glycosylations to environmental factors compounded the problem. As such, while
some aspects of glycosylations have been studied in great depth, the interrelationship of these
factors and their degree of influence has remained unclear. Here, we provide the most complete
picture to date of what these influencing factors are, what their affect is, and a direct comparison
of how the degrees of influence of these factors rank with respect to one another (Figure 12).
The most important lesson this work teaches us is that the donor and acceptor coupling
partners possess inherent preferences for the formation of either the αꢀ or βꢀstereoisomer, and that
these preferences can be enhanced, diminished, or overridden by environmental variables. The
first, and most important, of the variables dictated by coupling partners is the stereochemistry of
the C2 position of the donor, acting as a mechanistic divergent point in the coupling. Mannose,
bearing a C2 axial ether, has a strong inherent preference for the formation of the αꢀproduct. The
ACS Paragon Plus Environment

Page 15 of 18
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
degree of mannose’s αꢀselectivity can be enhanced in a number of ways, most notably by choice
of solvent and temperature. Solvents capable of stabilizing cationic intermediates such as the πꢀ
system of toluene or the lone pairs of MTBE or ACN increase the amount of alpha product
formed, as does an increase in temperature, though to a lesser extent. The next most influential
factor for mannose couplings is the activator, where the nature of the activator’s conjugate base
can change the selectivity at higher temperatures. Least influential appears to the acceptor, where
minor differences in sterics and electronics appear to have a correspondingly minor impact on
selectivity. The exception is poorly nucleophilic acceptors such as fluorinated alcohols, whose
strong electron withdrawing groups give highly αꢀselective mannosylations. Finally, the mannose
donor is highly sensitive to water, with subꢀstoichiometric amounts of water being sufficient to
negate the environmental effects described above.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 12: Degree of influence of environmental variables and permanent factors on the
stereoselective outcome of glycosylation.
Donors lacking an axial C2ꢀether, such as glucose and galactose, have an inherent
preference for the formation of the βꢀstereoisomer. However, this selectivity is more easily
influenced than what is observed for the mannose branch. The glucose/galactose class of
glycosylations is strongly influenced by the temperature, favoring the βꢀproduct at lower
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 16 of 18
1
2
3
4
5
6
7
8
9
temperatures and a weak preference for alpha at higher temperatures. The next most influential
factor is solvent, where nonꢀhalogen lone pair electrons play a key role: the nucleophilic
acetonitrile enhancing βꢀisomer formation, while the coordinating MTBE enhancing the
formation of the αꢀisomer. Similar to mannose, the activator enhances the inherent selectivity
preference of the donor, here the βꢀproduct. The degree of influence of the stereochemistry of the
C4 position is similar to that of the acceptor’s nucleophilicity. When the C4 ether is axial
(galactose), a moderate increase in the βꢀselectivity is observed, as is the case with stepwise
increases in the nucleophilicity of the acceptor. Poor nucleophiles, such as fluorinated alcohols,
are poorly selective, weakly favoring the αꢀstereoisomer. Finally, glucose/galactose
glycosylations can be slightly more βꢀselective with increased equivalents of acceptor. This
branch of couplings appears to be significantly less sensitive to water than mannosylations.
While the glycosylation coupling partners, in particular the donor, dictate the favored
stereochemical outcome, these inherent preferences can be manipulated by judicious choice of
environmental conditions. The selectivity is predominantly determined by the set of intermediates
formed following decomposition of the activated donor. The nature, ratio, and stability of these
undetermined structures is influenced both by the stereoelectronics of the donor and, primarily,
by the temperature, solvent, and conjugate base of the acceptor. The environmental conditions
can also be utilized to influence the mechanistic pathway followed, increasing either the SN1 or
SN2ꢀcharacter of the glycosylation. Glucose is the donor most susceptible to manipulation of the
stereochemical outcome, followed by galactose, and mannose.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Based on this work, it should be possible to identify conditions to maximize the
proportion of either the αꢀ or βꢀproduct for a given coupling pair. The coupling of the benzylated
glucosyl αꢀtrichloroacetimidate with isopropanol serves to showcase this ability (Figure 13). In
DCM using Tf2NH, 11:1 selectivity favoring the βꢀdiastereomer is observed. When the reaction
is run in MTBE with TfOH as activator at ꢀ30 °C, the selectivity is reversed to give a 9:1 ratio
favoring the αꢀdiastereomer.
Figure 13: Tunable stereoselective outcome ratios of a given glycosylation coupling pair based
on variation of environmental conditions.
In conclusion, an in depthꢀanalysis of factors that influence the outcome of glycosylation
reactions was possible thanks to an instrument that enables the systematic execution of reactions
quickly, reliably, and reproducibly. The results of these automated flow experiments were used to
identify and interrogate both permanent and environmental variables that influence the
stereoselectivity. For the first time, the extent that these variables affect the outcome of
glycosylations was quantified, and directly compared. As a result, an empirical understanding of
how to control the stereochemical outcome of a glycosylation was gained. Further studies
quantifying the degree of dominance of these interdependent variables, exploring the competing
mechanistic pathways, and the modelling of these complex systems are currently underway.
Acknowledgements:
ACS Paragon Plus Environment

Page 17 of 18
Journal of the American Chemical Society
1
2
3
4
5
6
The authors would like to thank generous financial support provided by the Max Planck Society
and DFG InCHeM (FOR 2177). We are also sincerely thankful to Dr. FeiꢀFei Xu, Dr. Marilda P.
Lisboa, Dr. Ju Yuel Baek, Eva Settels, and Olaf Niemeyer (all MPIKG) for their suggestions and
support.
7
8
9
Corresponding Authors:
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
kerry.gilmore@mpikg.mpg.de
peter.seeberger@mpikg.mpg.de
† These authors contributed equally to this work
Supporting Information:
Detailed experimental procedures, complete datasets, additional graphs and control studies,
details regarding automation and instrumentation, as well as characterization data. This
information is available free of charge at http://pubs.acs.org.
References:
1 (a) A. Varki, Glycobiology, 1993, 3, 97ꢀ130. (b) R. A. Dwek, Chem. Rev., 1996, 96, 683ꢀ720.
2 Hofmann, J.; Hahm, H. S.; Seeberger, P. H.; Pagel, K. Nature 2015, 526, 241ꢀ244.
3 a) Crich, D. Acc. Chem. Res. 2010, 43, 1144ꢀ1153; b) Bohé, L.; Crich, D. Carbohyd. Res. 2015, 403, 48ꢀ59.
4 Ranade, S. C.; Demchenko, A. V. J. Carbohyd. Chem. 2013, 32, 1ꢀ43.
5 Hosoya, T.; Takano, T.; Kosma, P.; Rosenau, T. J. Org. Chem. 2014, 79, 7889ꢀ7894.
6 Tvaroska, I.; Bleha, T.; Adv. Carbohydr. Chem. Biochem. 1989, 47, 45ꢀ123.
7 Cumpstey, I. Org. Biomol. Chem. 2012, 10, 2503ꢀ2508.
8 Frihed, T. G.; Bols, M.; Pedersen, C. M. Chem. Rev. 2015, 115, 4963ꢀ5013.
9 Park, J.; Kawatkar, S.; Kim, J.ꢀH.; Boons, G.ꢀJ. Org. Lett. 2007, 9, 1959ꢀ1962.
10 Reports discuss the observation of αꢀtriflates and cationic intermediates under extreme conditions, see: a) Crich, D.
J. Org. Chem. 2011, 76, 9193ꢀ9209; b) Martin, A.; Arda, A.; Désiré, J.; MartinꢀMingot, A.; Probst, N.; Sinaÿ, P.;
JiménezꢀBarbero, J.; Thibaudeau, S.; Blériot, Y. Nature Chem. 2016, 8, 186ꢀ191.
11 Satoh, H.; Manabe, S. Chem. Soc. Rev. 2013, 42, 4297ꢀ4309.
12 a) Nigudkar, S. S.; Demchenko, A. V. Chem. Sci. 2015, 6, 2687ꢀ2704; b) Garcia, A.; Otte, D. A. L.; Salamant, W.
A.; Sanzone, J. R.; Woerpel, K. A. Angew. Chem. Int. Ed. 2015, 54, 3061ꢀ3064; c) Lourenço, E. C.; Ventura, M. R.
Tetrahedron 2013, 69, 7090ꢀ7097.
13 a) Alabugin, I. V.; Gilmore, K. M.; Peterson, P. W. WIREs Comp. Mol. Sci. 2011, 1, 109ꢀ141; b) Alabugin, I. V.
Stereoelectronic Effects: A Bridge Between Structure and Reactivity. (2016) John Wiley & Sons.
14 Transient protecting groups can also be used to gain stereoselectivity through sterically blocking one face, see:
Hoang, K. L. M.; He, J.ꢀx.; Báti, G.; ChanꢀPark, M. B.; Liu, X.ꢀW. Nature Commun. 2017, 8, 1146.
15 a) Kalikanda, J.; Li, Z. J. Org. Chem. 2011, 76, 5207ꢀ5218; b) Baek, J. Y.; Lee, B.ꢀY.; Jo, M. G.; Kim, K. S. J.
Am. Chem. Soc. 2009, 131, 17705ꢀ17713.
16 FraserꢀReid, B.; Wu, Z.; Udodong, U. E.; Ottosson, H. J. Org. Chem.1990, 55, 6068ꢀ6070.
17 Jensen, H. H.; Nordstrom, L. U.; Bols, M. J. Am. Chem. Soc. 2004, 126, 9205ꢀ0213.
18 a) Huang, M.; Garrett, G. E.; Birlirakis, N.; Bohé, L.; Pratt, D. A.; Crich, D. Nat. Chem. 2012, 4, 663ꢀ667; b)
Crich, D.; Cai, W. J. Org. Chem. 1999, 64, 4926ꢀ4930; c) Crich, D.; Chandrasekera, N. S. Angew. Chem. Int. Ed.
2004, 43, 5386ꢀ5389.
19 Trichloracetimidates: a) Schmidt, R. R.; Michel, J. Angew. Chem. Int. Ed. 1980, 19, 731ꢀ733; thioglycosides: b)
Pozsgay, V.; Jennings, H. J. Tetrahedron Letters 1987, 28, 1375ꢀ1376; c) Lemieux, R. U. Can. J. Chem. 1951, 29,
1079ꢀ1091; d) Garegg, P. J. Adv. Carbohyd. Chem. Bi. 1997, 52, 179ꢀ205; phosphates/phosphites: e) Hashimoto, S.;
Honda, T.; Ikegami, S. J. Chem Soc., Chem. Commun. 1989, 685ꢀ687; f) Martin, T. J.; Schmidt, R. R. Tetrahedron
Letters 1992, 33, 6123ꢀ6126; g) Martin, T. J.; Brescello, R.; Toepfer, A.; Schmidt, R. R. Glyconjugate Journal 1993,
10, 16ꢀ25; h) Kondo, H.; Ichikawa, Y.; Wong, C.ꢀH. J. Am. Chem. Soc. 1992, 114, 8748ꢀ8750; i) Sim, M. M.;
Kondo, H.; Wong, C.ꢀH. J. Am. Chem. Soc. 1993, 115, 2260ꢀ2267; Overviews: j) Demchenko, A. V. Curr. Org.
Chem. 2003, 7, 35ꢀ79; k) Demchenko, A. V. Synlett 2003, 1225ꢀ1240.
20 a) Paulsen, H. Angew. Chem. Int. Ed. 1982, 21, 155ꢀ173; b) Sinay, P. Pure Appl. Chem. 1978, 50, 1437ꢀ1452.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 18 of 18
1
2
3
4
5
6
7
8
9
21 Orgueira, H. A.; Bartolozzi, A.; Schell, P.; Seeberger, P. H. Angew. Chem. Int. Ed. 2002, 41, 2128ꢀ2131.
22 (a) van der Vorm, S.; Hansen, T.; Overkleeft, H. S.; van der Marel, A.; Codée, J. D. C. Chem. Sci. 2017, 8, 1867ꢀ
1875; (b) Schumann, B.; Parameswarappa, S. G.; Lisboa, M. P.; Kottari, N.; Guidetti, F.; Pereira, C. L.; Seeberger, P.
H. Angew. Chem. Int. Ed. 2016, 55, 14431ꢀ14434.
23 Hosoya, T.; Kosma, P.; Rosenau, T. Carbohyd. Res. 2015, 401, 127ꢀ131.
24 Kononov, L. O.; Malysheva, N. N.; Orlova, A. V.; Zinin, A. I.; Laptinskaya, T. V.; Kononova, E. G.; Kolotyrkina,
N. G. Eur. J. Org. Chem. 2012, 1926ꢀ1934.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
25 a) Huang, M.; Retailleau, P.; Bohé, L.; Crich, D. J. Am. Chem. Soc. 2012, 134, 14746ꢀ14749; b) Adero, P. O.;
Furukawa, T.; Huang, M.; Mukherjee, D.; Retailleau, P.; Bohé, L; Crich, D. J. Am. Chem. Soc. 2015, 137, 10336ꢀ
10345.
26 Tanaka, K.; Mori, Y.; Fukase, K. J. Carbohdr. Chem. 2009, 28, 1ꢀ11.
27 Plutschack, M. B.; Pieber, B.; Gilmore, K.; Seeberger, P. H., Chem. Rev. 2017, 117, 11796ꢀ11893.
28 Ratner, D. M.; Murphy, E. R.; Jhunjhunwala, M.; Snyder, D. A.; Jensen, K. F.; Seeberger, P. H. Chem. Commun.
2005, 578ꢀ580.
29 a) van der Vorm, S.; van Hengst, J. M. A.; Bakker, M.; Overkleeft, H. S.; van der Marel, G. A.; Codée, J. D. C.
Angew. Chem. Int. Ed. 2018, DOI: 10.1002/anie.201802899; b) van der Vorm, S.; Overkleeft, H. S.; van der Marel,
G.; D. C. Codée, J. J. Org. Chem. 2017, 82, 4793ꢀ4811.
30 (a) Durantie, E.; Bucher, C.; Gilmour, R. Chem. Eur. J. 2012, 18, 8208; (b) Rencurosi, A.; Lay, L.; Russo, G.;
Caneva, E.; Poletti, L. J. Org. Chem. 2005, 70, 7765.
31 Further studies exploring the unique reactivies of the αꢀ and βꢀdonor will be discussed in a subsequent publication.
32 a) Yang, M. T.; Woerpel, K. A. J. Org. Chem. 2009, 74, 545ꢀ553; b) Chamberland, S.; Ziller, J. W.; Woerpel, K.
A. J. Am. Chem. Soc. 2005, 127, 5322ꢀ5323.
33 Crich, D.; Vinogradova, O. J. Org. Chem. 2006, 71, 8473ꢀ8480.
34 Lucero, C. G.; Woerpel, K. A. J. Org. Chem. 2006, 71, 2641ꢀ2647.
35 Bucher, C.; Gilmour, R. Angew. Chem. Int. Ed. 2010, 49, 8724ꢀ8728.
36 Kalikanda, J.; Li, Z. Carbohydr. Res. 2011, 346, 2380ꢀ2383.
37 α/βꢀproducts of perbenylated glucose coupled to either 2,2,2ꢀtrifluoroethanol or 2,2ꢀdifluoroethanol could not be
separated using the HPLC methods developed.
38 A similar observation has been made for 2ꢀdeoxyglycosyl donors, see: Beaver, M. G.; Woerpel, K. A. J. Org.
Chem. 2010, 75, 1107ꢀ1118.
39 The concentration of the donor has been shown to influence the temperature of activation, the reaction time, and
the yield, see: Kononov, L. O.; Fedina, K. G.; Orlova, A. V.; Kondakov, N. N.; Abronina, P. I.; Podvalnyy, N. M.;
Chizhov, A. O. Carbohyd. Res. 2017, 437, 28ꢀ35.
40 For a discussion about the nucleophilicity of the trifluoromethanesulfonate anion, see: Dhakal, B.; Bohé, L.; Crich,
D. J. Org. Chem. 2017, 82, 9263ꢀ9269.
41 Kowalska, K.; Pedersen, C. M. Chem. Commun. 2017, 53, 2040ꢀ2043.
42 Kafle, A.; Liu, J.; Cui, L. Can. J. Chem. 2016, 94, 894ꢀ901.
43 Satoh, H.; Hansen, H. S.; Manabe, S.; van Gunstersen, W. F.; Hünenberger, P. H. J. Chem. Theory Comput. 2010,
6, 1783ꢀ1797.
Table of Contents Graphic:
ACS Paragon Plus Environment