Synthesis of β-Glucosides with 3-O-Picoloyl-Protected Glycosyl Donors in the Presence of Excess Triflic Acid: A Mechanistic Study

Article abstract of DOI:10.1002/chem.201905277

Our previous study showed that picoloylated donors are capable of providing excellent facial stereoselectivity through the H-bond-mediated aglycone delivery (HAD) pathway. Presented herein is a detailed mechanistic study of stereoselective glycosylation with 3-O-picoloylated glucosyl donors. While reactions of glycosyl donors equipped with the 3-O-benzoyl group are typically non-stereoselective because these reactions proceed via the oxacarbenium intermediate, 3-O-picoloylated donors are capable of providing enhanced, but somewhat relaxed, β-stereoselectivity by the HAD pathway. In an attempt to refine this reaction, we noticed that glycosylations are highly β-stereoselective in the presence of NIS and stoichiometric TfOH. The HAD pathway is highly unlikely because the picoloyl nitrogen is protonated under these reaction conditions. The protonation and glycosylation were studied by low-temperature NMR, and the intermediacy of the glycosyl triflate has been observed. This article is dedicated to broadening the scope of this reaction in application to a variety of substrates and targets.

Full text of DOI:10.1002/chem.201905277

A Journal of
Accepted Article
Title: Synthesis of β-glucosides with 3-O-picoloyl-protected glycosyl
donors in the presence of excess triflic acid: a mechanistic study
Authors: Michael Mannino and Alexei V. Demchenko
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201905277
Link to VoR: http://dx.doi.org/10.1002/chem.201905277
Supported by

10.1002/chem.201905277
Chemistry - A European Journal
Synthesis of β-glucosides with 3-O-picoloyl-protected glycosyl donors in the presence of
excess triflic acid: a mechanistic study
Michael P. Mannino and Alexei V. Demchenko*
Department of Chemistry and Biochemistry, University of Missouri – St. Louis
One University Boulevard, St. Louis, Missouri 63121, USA
e-mail: demchenkoa@umsl.edu
Abstract: Our previous study showed that picoloylated donors are capable of providing excellent
facial stereoselectivity via the H-bond-mediated aglycone delivery (HAD) pathway. Presented
herein is a detailed mechanistic study of stereoselective glycosylation with 3-O-picoloylated
glucosyl donors. While reactions of glycosyl donors equipped with the 3-O-benzoyl group are
typically non-stereoselective because these reactions proceed via the oxacarbenium intermediate,
3-O-picoloylated donors are capable of providing enhanced, but somewhat relaxed, β-
stereoselectivity via the HAD pathway. In attempt of refining this reaction, we noticed that
glycosylations are highly β-stereoselective in the presence of NIS and stoichiometric TfOH. The
HAD pathway is highly unlikely because the picoloyl nitrogen is protonated under these reaction
conditions. The protonation and glycosylation were studied by low-temperature NMR, and the
intermediacy of glycosyl triflate has been observed. The subsequent article in this issue is
dedicated to broadening the scope of this reaction in application to a variety of substrates and
targets.
Keywords: glycosylation, synthesis, oligosaccharides, stereoselectivity, mechanism
1
This article is protected by copyright. All rights reserved.

10.1002/chem.201905277
Chemistry - A European Journal
Introduction
Carbohydrates or glycans are a class of biomolecules that are ubiquitously found in all
living organisms. The importance of these molecules is connected to their involvement in a wide
variety of biological functions.[1] Determining a relationship between the structure and function of
these glycans not only helps to elucidate biological processes, but also opens the door for
carbohydrate-based therapeutic-agent and diagnostic-platform development. Obtaining pure
glycans from natural sources to study their biological function and/or utilize them for medicinal
purposes can be challenging due to their high complexity and heterogeneity.[2] The availability of
pure glycans of interest is essential for innovations in all areas of glycosciences, and chemical
synthesis can be an efficient mode for the production of structurally defined glycans, making it a
powerful tool.
Chemical glycosylation is the cornerstone reaction of carbohydrate chemistry. This
reaction links together the glycosyl donor and acceptor through the formation of a glycosidic bond.
Due to their inherit SN1-type nature, achieving stereoselectivity in these reactions is a constant
struggle. To alleviate this problem, many direct and indirect methods that attempt to control the
preferential formation of either 1,2-cis or 1,2-trans glycosides have been developed.[3] One of these
methods is the Hydrogen-bond-mediated Aglycone Delivery (HAD) reaction introduced by our
lab in 2012.[4] In accordance with this reaction, the glycosyl donor is functionalized with a
protecting group capable of acting as a hydrogen bond acceptor at a remote position of the sugar
(C-3, 4 or 6). Several dedicated protecting groups have been developed,[4-5] among which the
picoloyl ester (Pico) is the most common. The advantage of this substituent lies in its facile
installation and chemoselective removal using zinc(II) acetate or copper(II) acetate,[6] making it
very appealing for iterative oligosaccharide synthesis.[7] The Pico group forms a hydrogen bond
2
This article is protected by copyright. All rights reserved.

10.1002/chem.201905277
Chemistry - A European Journal
with the free hydroxyl of the glycosyl acceptor, resulting in the formation of a hydrogen-bonded
complex. The orientation of this complex determines the stereoselectivity afforded as the acceptor
is “delivered” to the anomeric carbon in a syn-fashion with respect to the Pico group (Scheme 1).
In a majority of applications, picolinyl ether (Pic) provided comparable results in terms of
stereoselectivities and yields.[4] Presented herein is a dedicated investigation of 3-O-picoloylated
(3-Pico) glucosyl donors. Glycosidation of these substrates is practically unexplored, limiting the
scope and applicability of the HAD method.
Scheme 1. Hydrogen-bond-mediated Aglycone Delivery (HAD)
Results and Discussion
In our original study of the HAD, we developed reaction conditions that led to excellent
stereoselectivity of glycosylations with a variety of substrates.[4] With the primary focus on
glycosyl donors of the D-gluco series, several compounds were investigated, and the performance
of 4-Pico donor 1 was particularly impressive.[8] Reaction of this donor with acceptor 2[9]
exclusively provided α-linked product 3 due to the downward orientation of the C-4 substituent in
the D-gluco series. To achieve this stereoselectivity in a good yield (73%), the reaction was
3
This article is protected by copyright. All rights reserved.

10.1002/chem.201905277
Chemistry - A European Journal
performed in the presence of dimethyl(methylthio)sulfonium triflate (DMTST) at low
concentration (5 mM, entry 1, Table 1).[4] Over the course of the original study, we came to a
realization that other important factors are essential for high stereoselectivity. For example, it was
determined that donor preactivation leads to complete loss of stereoselectivity. We also observed
that these reactions cannot be “pushed” by increasing the amount of the electrophilic promoter,
such as DMTST or N-iodosuccinimide (NIS)/TfOH, because both of these attempts led to a rapid
decline of stereoselectivity. These observations highlighted the necessity for the formation of the
H-bonded donor-acceptor complex for the HAD pathway to take place.
Equally impressive was performance of 6-Pico donor 4 that was found to be capable of
providing complete stereoselectivity in reaction with glycosyl acceptor 2.[10] In this case, complete
β-stereoselectivity was obtained due to the upward orientation of the substituent at C-6. To achieve
the stereoselective formation of disaccharide 5 in a good yield of 78% the reaction was performed
under essentially the same reaction conditions, low excess of DMTST and using high dilution (5
mM, entry 2).[10]
Despite the successful application of glycosyl donors equipped with 4- or 6-Pico group,
the HAD glycosylation using 3-Pico glycosyl donors remained practically unexplored.
Glycosidation of 3-Pic ether protected glycosyl donor 6 has been reported, and although
disaccharide 7 was formed in a good yield of 85% and with a respectable β-stereoselectivity (α/β
= 1/15.6, entry 3),[4] further experimentation to enhance this promising result has not been pursued.
To broaden the scope of the HAD methodology, we synthesized 3-Pico protected donor 8 and
investigated its glycosidation with acceptor 2. Although the corresponding disaccharide 9 was
obtained in an excellent yield of 90%, to our surprise, the stereoselectivity dropped dramatically
(α/β = 1/2.6, entry 4). Interestingly, our previous studies with donor 10, a D-manno analogue of
4
This article is protected by copyright. All rights reserved.

10.1002/chem.201905277
Chemistry - A European Journal
donor 8, gave us very strong preference for the formation of β-configured products.[7b]
Mannosylation conducted at room temperature and under common concentration (50 mM)
produced β-disaccharide 11 in 92% yield (α/β = 1/7.0, entry 5).
Even higher β-manno selectivity was achieved with NIS/ catalytic TfOH promoter system
leading to disaccharide 11 in 89% yield (α/β = 1/8.0, entry 6).[7b] Given this improvement, we
decided to investigate the effectiveness of this promoter system for the activation of glucosyl donor
8. Indeed, the activation of donor 8 under essentially the same conditions as those developed for
the HAD synthesis of mannosides[7b] led to the formation of disaccharide 9 with an improved
stereoselectivity in comparison to that previously achieved with DMTST (α/β = 1/2.6, entry 4).
Thus, reactions using low (5 mM) or regular (50 mM) concentrations afforded disaccharide 9 in
81% (α/β = 1/4.3, entry 7) and 90% (α/β = 1/3.3, entry 8), respectively. Encouraged by this initial
improvement, we began a focused study to further enhance the stereoselectivity of these
glycosylations. Interestingly, recent work by De Meo et al. introduced new reaction conditions
that led to a significant increase in selectivity with 4-Pico functionalized sialyl donors.[11] The new
conditions were made by modifying the NIS/TfOH promoter system, specifically altering the
amount of TfOH keeping the amount of NIS constant (2.0 equiv with respect to the donor).
Traditionally used in catalytic amounts (0.2 equiv with respect to the donor), TfOH was increased
to equimolar amounts in De Meo’s reactions. These modified conditions brought about a dramatic
change in stereoselectivity from α/β = 3/1 to α/β = 14/1, and have since been applied to other
sialylations.[12] Schmidt, Sun et al also obtained excellent stereoselectivity with 1-Pic-protected
sialyl donors.[13] In application to our substrates, when glycosidation of donor 8 was performed in
the presence of NIS and TfOH (2.0 equiv each) disaccharide 9 was obtained in a good yield of
79% and a dramatically enhanced stereoselectivity (α/β = 1/16.6, entry 9).
5
This article is protected by copyright. All rights reserved.

10.1002/chem.201905277
Chemistry - A European Journal
Table 1. Performance of glycosyl donors in the HAD reaction can differ
Promoter (equiv),
conditions, time
Entry
Donor (conc.)
Product, yield, α/β ratio
DMTST (2)
4Å MS, 1,2-DCE
-30 oC  rt, 4 h
1
1 (5 mM)
3, 73%, > 25/1
5, 78%, < 1/25
7, 85%, 1/15.6
9, 90%, 1/2.6
DMTST (2)
4Å MS, 1,2-DCE
-30 oC  rt, 2 h
2
3
4 (5 mM)
DMTST (2)
4Å MS, 1,2-DCE
-30 oC  rt, 3 h
6 (5 mM)
DMTST (2)
4Å MS, 1,2-DCE
-30 oC  rt, 3 h
4
5
8 (5 mM)
DMTST (2)
4Å MS, 1,2-DCE
rt, 30 min
10 (50 mM)
11, 92%, 1/7.0
11, 89%, 1/8.0
NIS (3)/TfOH (0.3)
4Å MS, 1,2-DCE
rt, 20 min
NIS (3)/TfOH (0.3)
4Å MS, 1,2-DCE
-30 oC  rt, 3.5 h
6
7
10 (50 mM)
8 (5 mM)
9, 81%, 1/4.3
6
This article is protected by copyright. All rights reserved.

10.1002/chem.201905277
Chemistry - A European Journal
NIS (2)/TfOH (0.2)
4Å MS, 1,2-DCE
-30 oC  rt, 24 h
NIS (2)/TfOH (2)
4Å MS, 1,2-DCE
-30 oC, 30 min
NIS (1.2)/TfOH (2.5)
4Å MS, 1,2-DCE
-30 oC, 30 min
8
8 (50 mM)
8 (50 mM)
8 (50 mM)
9, 90%, 1/3.3
9, 79%, 1/16.6
9, 85%, 1/23
9
10
NIS (1.2)/TfOH (2.5)
4Å MS, 1,2-DCE
-30 oC, 30 min
11
12 (50 mM)
13, 86%, 1/1.0
Given this impressive enhancement in stereoselectivity, we decided to further optimize the
reaction conditions. Through the course of this investigation, we found that the stereoselectivity in
these reactions was proportional to the amount of TfOH in relation to NIS. After careful
refinement, 1.2 equiv of NIS and 2.5 equiv of TfOH were found to be the optimal amounts of
promoters. These reaction conditions provided the best combination of reaction rates,
stereoselectivities, and yields. Thus, glycosylation of acceptor 2 with donor 8 produced
disaccharide 9 in 85% yield and practically complete β-stereoselectivity (α/β = 1/23, entry 10).
The complete details of the study dedicated to refining the reaction conditions are presented as a
part of the supporting information (SI).
Considering the large amount of protic acid present in the reaction medium, and the fact
that excess TfOH should disrupt the HAD pathway,[4] we endeavored a dedicated investigation of
the mechanistic forces behind this enhancement of stereoselectivity. First, a possibility of post-
glycosylational anomerization of the glycosidic bond in disaccharide 9 under these highly acidic
reactions was ruled out by performing a dedicated experimentation. This was performed by placing
disaccharide 9 under the reaction conditions used for glycosylations, which resulted in no changes
in the anomeric configuration (see the SI for further details). Second, to understand whether the
7
This article is protected by copyright. All rights reserved.

10.1002/chem.201905277
Chemistry - A European Journal
high stereoselectivity obtained under these modified reaction conditions was due to the presence
of the Pico group, we prepared 3-O-benzoylated donor 12 lacking the Pico substituent. When 3-
Bz donor 12 was glycosidated with acceptor 2 under the same reaction conditions, the
corresponding disaccharide 13 was rapidly produced in 86% yield. This reaction, however, was
completely non-stereoselective (α/β = 1/1.0, entry 11). These results imply that the 3-Pico group,
combined with these newly optimized activation conditions, are both required factors to achieve
excellent stereoselectivity. However, there could be other electron-withdrawing protecting groups
capable of providing a similar effect.
In this context, De Meo et al.[11] proposed that protonation of the Pico nitrogen might be
taking place in the presence of excess triflic acid. Very recently, Schmidt, Sun et al also noted a
possibility of protonation of Pic ether groups, which was supported by DFT calculations.[14]
Considering the highly acidic medium of the reaction mixture due to a full equivalent excess of
TfOH in respect to the iodonium conjugate base, succinimide anion from NIS, along with the basic
nature of the Pico nitrogen, our working hypothesis was that the reaction proceeds via the initial
protonation of glycosyl donor 8 with TfOH. Consequently, this led us to a belief that glycosidation
of 3-Pico derivative 8 under the newly established reaction conditions follows a different
mechanistic pathway than that broadly known for conventional glycosidation of glycosyl donors
lacking the Pico group. As shown in Scheme 2, conventional glycosidation of donor 12 begins by
the activation of the ethylthio leaving group via the interaction with the iodonium ion liberated by
a reaction of NIS with TfOH. The anomeric C—S bond of the resulting glycosyl sulfonium ion
intermediate then dissociates leading to the formation of the oxacarbenium ion. The subsequent
nucleophilic attack of glycosyl acceptor 2 then takes place in a non-stereoselective manner leading
to product 13 that is commonly formed as a mixture of diastereomers.
8
This article is protected by copyright. All rights reserved.

10.1002/chem.201905277
Chemistry - A European Journal
We believe that glycosidation of 3-Pico derivative 8 under the newly developed reaction
conditions follows a different mechanistic pathway. The excess TfOH serves dual purpose: to form
the reactive iodonium cation by reacting with NIS and to protonate the Pico group in donor 8
(Scheme 3). From these initial steps in the reaction of donor 8 onwards, the activation of the
ethylthio leaving group via the interaction with the iodonium ion follows a different pathway.
Scheme 2. A conventional glycosidation of thioglycosides in the presence of NIS/TfOH
It is possible that sulfonium ion formation would be disfavored by the electron withdrawing
effect of the positively charged protonated 3-Pico substituent. While the iodonium ion helps the
leaving group departure, the triflate counterion approaches the anomeric center. These concomitant
interactions result in the formation of the covalently bound anomeric triflate A in a concerted
fashion (Scheme 3). As a result, the mechanistic step wherein formation of the oxacarbenium ion
intermediate, which would have been destabilized by the protonated Pico substituent, is bypassed.
The subsequent nucleophilic attack of acceptor 2 then also follows the concerted displacement
pathway that avoids the ionization at the anomeric center leading to the overall retention of the
anomeric configuration in product 9 with respect to the starting material. It is also possible, that
9
This article is protected by copyright. All rights reserved.

10.1002/chem.201905277
Chemistry - A European Journal
the developed reaction conditions may favor rapid equilibria proceeding via the intermediacy of
close ion pairs instead.
Scheme 3. Proposed pathway for the reaction of 3-Pico glycosyl donor 8 with acceptor 2 in the
presence of NIS and excess TfOH
To gain better understanding of this interesting reaction, we decided to conduct a dedicated
mechanistic investigation. Glycosyl triflates have been widely studied in the field of synthetic
carbohydrate chemistry.[15] First reported by Kronzer and Schuerch in 1973,[16] these highly
reactive species are susceptible to SN2-like attacks from glycosyl acceptors with the contact ion
pair of α-triflate intermediates, affording excellent stereoselectivity.[17] Glycosyl triflates have
been formed from a variety of different donor substrates using a handful of promoter systems,[15]
including NIS/TfOH.[18] These are highly reactive intermediates, and their stabilities can be
accessed by recording the temperature at which they decompose in variable temperature NMR
(VT-NMR) experiments.[15] The stability of glycosyl triflates is often a function of specific
structural features of the corresponding donor that favors their formation over the charged, and
10
This article is protected by copyright. All rights reserved.

10.1002/chem.201905277
Chemistry - A European Journal
conformationally altered oxacarbenium ion. These structural features include bicyclic protecting
groups that hinder the conformational flip of the oxacarbenium ion,[19] or electron withdrawing
protecting groups, which destabilize the positively charged oxacarbenium ion. In the latter
example, it has been shown that both the strength of the electron withdrawing group[20] and its
location on the ring[21] influence glycosyl triflate stability. Importantly, the stabilities of these
intermediates have been found to be directly related to the stereoselectivity obtained.[20-22] In this
context, excellent stereoselectivity of sialylations in the study by Schmidt, Sun et al was also
attributed to the electron-withdrawing effect of the protonated 1-Pic group.[14]
Although not as thoroughly investigated as their mannosyl counterparts, several examples
of glucosyl triflates have been reported, but the stereoselectivity obtained with these intermediates
can be poor due to their instability.[15] All previously reported glucosyl triflates that produce high
β-stereoselectivity in the absence of a participating group at C-2 contain bicyclic functionality.
These examples include 2,3-O-carbonate[23] and 3,4-O-bisacetal[24] protecting groups. Unlike with
mannosyl substrates, electron withdrawing protecting groups at remote positions have not been
reported for the stabilization of glucosyl triflates.
In application to our substrates, the protonated 3-Pico donor 8 forms a stable five-
membered ring between the nitrogen and the adjacent carbonyl oxygen of the Pico group (Scheme
3). The protonated Pico group will be capable of strong electron withdrawal, comparable to that
of other charged nitrogen moieties [N+(Me)3 σp = 0.82, N≡N+ σp = 1.91],[25] and its proximity to
the anomeric center, may inhibit the formation of the positively charged glycosylation
intermediates, such as sulfonium and oxacarbenium ion. This affect causes the preferential
formation of the glycosyl triflate intermediate, which is subsequently displaced by the glycosyl
acceptor in a SN2-like fashion providing excellent β-stereoselectivity.
11
This article is protected by copyright. All rights reserved.

10.1002/chem.201905277
Chemistry - A European Journal
To investigate the anticipated protonation of donor 8 with TfOH we conducted
spectroscopic studies dedicated to monitoring possible chemical shift changes taking place in 1H
NMR spectra. These experiments were performed with thioglycosides carrying different
substituents at C-3 with and without TfOH. The first series of experiments were conducted with
3-O-benzylated donor 14[26] and 3-O-benzoylated donor 12. As evident from Figure 1, no changes
in chemical shifts between spectra recorded with and without TfOH were observed for these
compounds. This result suggests that TfOH neither interacts with compounds 12 and 14 nor acts
as an NMR shift reagent.
Figure 1. Studying the effect of protonation on a series of glycosyl donors
Conversely, we noticed a substantial shift in location of several proton signals in 3-Pico
donor 8 in the presence of TfOH confirming the protonation (Figure 1). The major changes
occurred with the aromatic, benzylic, and H-3 signals suggesting protonation of the Pico group.
12
This article is protected by copyright. All rights reserved.

10.1002/chem.201905277
Chemistry - A European Journal
Thus, the pyridine ring protons exhibited a downfield shift, whereas other aromatic protons and
the H-3 signal were partially shifted upfield.
Having acquired proof of the protonation of donor 8 in the presence of excess TfOH, we
turned our attention to obtaining the evidence of the proposed glucosyl triflate intermediate
existence. The anomeric proton of glycosyl triflates is typically found between 5.95-6.94 ppm.[15]
However, in the presence of TfOH benzylated 3-Pico donor 8 produces overcrowded NMR
spectrum in the region of interest due to an upfield shift of aromatic protons that appear between
6.50 and 7.50 ppm (Figure 1). Due to this concern, and inspired by studies by Crich,[19a] the
corresponding methylated 3-Pico donor 15 was synthesized for this spectroscopic investigation.
Methylated 3-Pico donor 15 provided similar results for glycosylation of acceptor 2 (vide infra,
Table 2), but produced much simpler spectra in the presence of TfOH with no signals between 5.5-
8.0 ppm (Figure 2). The activation of glycosyl donor 15 in the absence of a glycosyl acceptor and
in the presence of only minimal amounts of promoters NIS/TfOH (1.0/2.0 equiv) was analyzed by
NMR. Glycosyl donor 15 was dissolved in CDCl3, the solution was cooled to -50 °C. NIS (1.0
equiv) and TfOH (2.0 equiv) were added, the resulting solution was then immediately transferred
to the standard NMR tube and cooled down further using liquid nitrogen. The reaction mixture
was then analyzed by NMR, and the proton spectrum recorded at this timepoint revealed the
presence of two species (Figure 2). The major of the two species was identified as the protonated
glycosyl donor 15, with characteristic signals at δ 5.4 ppm (H-3) and 4.4 ppm (H-1). The minor
species contains a doublet peak resonating at 6.9 ppm.
At that point, we could not yet rule out the presence of other possible reactive intermediates
that could have been formed in the presence of NIS. Spectroscopic evidence for both glycosyl
iodides[15] and succinimides[27] have previously been reported. While glycosyl succinimide was
13
This article is protected by copyright. All rights reserved.

10.1002/chem.201905277
Chemistry - A European Journal
shown to resonate at 5.5 ppm, glycosyl iodides have been known to reside further downfield.[15]
To exclude a possibility that the observed intermediate is a glycosyl iodide, the NMR experiment
was repeated using N-bromosuccinimide (NBS) instead of NIS. This experiment also showed the
complete consumption of donor 15, but in this case, signals corresponding to three different species
were detected in the area of interest. Among these, the suspected glycosyl triflate signal at 6.9 ppm
was also present signifying that it is not a glycosyl iodide. To account for other species observed
during the NMR experiment we also obtained methyl glycoside 16, hemiacetal 17, and bromide
18. Spectra of these compounds in the presence of TfOH have been recorded and are shown in
Figure 2. This allowed us to identify bromide 18 in the reaction mixture with its anomeric proton
resonating at 6.7 ppm (Figure 2). We do not yet have the proof of the third anomeric signal
resonating at 7.7 ppm, which could be a highly elusive β-bromide.[28]
Analysis of the COSY spectrum of the reaction mixture produced from donor 15 in the
presence of NIS/TfOH determines that the H-1 signal of the glycosyl triflate intermediate A at 6.9
ppm is coupled to H-2 at 4.3 ppm, which in turn is coupled to H-3 at 5.5 ppm (Figure 3, see also
the SI for further detail). The H-3 signal has a similar downfield chemical shift to that of the other
H-3 protons adjacent to Pico. Furthermore, the signal at 6.9 ppm integrates to the same value as a
Pico proton signal found at 9.7 ppm. All of these aspects suggest that the signal appearing at 6.9
ppm is indeed the anomeric proton of a substrate functionalized with Pico at the C-3 position and
a strong electron withdrawing group at the anomeric carbon.
14
This article is protected by copyright. All rights reserved.

10.1002/chem.201905277
Chemistry - A European Journal
Figure 2. Low temperature reaction monitoring and identification of the key intermediates
To confirm this species as a glycosylation reaction intermediate, methanol (2.0 equiv) was
added as soon as the species has been identified. The signals corresponding to triflate A disappear,
and the emergence of the corresponding methyl glycoside 16 signals, such as H-1 at 4.3 ppm are
formed. Interestingly, the signals of the protonated donor 15 remain after methanol addition. These
results confirm that the species resonating at 6.9 ppm is indeed triflate intermediate A, and that the
protonated donor is unable to form product directly. Both of these observations support the
hypothesis of the glycosyl triflate formation as a result of 3-Pico protonation.
15
This article is protected by copyright. All rights reserved.

10.1002/chem.201905277
Chemistry - A European Journal
Figure 3. NMR monitoring experiment with donor 15 and MeOH
In further investigation of changes taking place in the NMR spectra recorded in the
presence of TfOH, we noticed that the H-3 signal in protonated donor 8 experienced an upfield
shift (see Figure 1). This was peculiar as one would anticipate a downfield shift, if any, for the
proton adjacent to a charged electron-withdrawing moiety. For comparison, a mild downfield shift
of the methyl group was observed upon protonation of methyl picolinate with TfOH (see the SI).
In this context, Mong et. al. reported that coordination of the Pico group with TMS does not alter
the chemical shift of the adjacent sugar proton.[29] This suggests that the unexpected upfield shift
of the H-3 signal in compound 8 is not due to the Pico group protonation, but rather arises from
other structural features. In fact, practically no shift of the H-3 signal was noted in the spectrum of
protonated methylated donor 15 (see Figure 2). These observations led to a hypothesis that the
protonated pyridine ring acts as a powerful π-acceptor, which causes it to participate in π-π or
cation-π interactions with neighboring benzyl substituents.[30]
16
This article is protected by copyright. All rights reserved.

10.1002/chem.201905277
Chemistry - A European Journal
To support the viability of this hypothesis, we obtained a series of partially methylated 3-
Pico derivatives 20, 22, 24, 26, 28, and 30 depicted in Table 2. This series of compounds were
analyzed by NMR with and without TfOH showing that the upfield shift of the H-3 proton is
diminished upon exchanging benzyl groups with methyl groups (Table 2). No significant changes
in both the chemical shift and splitting of other sugar or aglycone protons took place. All of the
methylated donors exhibited similar yields (78-96%) and complete β-stereoselectivity in reactions
with acceptor 2 for the synthesis of the corresponding disaccharides 21, 23, 25, 27, 29, and 31
under the optimized promoter conditions (Table 2). This result indicates that the proposed π-π
stacking interactions with benzyl groups do not influence the stereoselectivity (see the SI for
complete details of this study).
Table 2. Investigation of partially methylated glycosyl donors
Donor (H-3 ∆δ upon
Conditions,
yield, α/β ratio
Entry
1
Product
protonation)
A: 89%, 1/3.0
B: 85%, < 1/25
8 (-0.3 ppm)
9
A: 84%, 1/6.0
B: 80%, < 1/25
2
15 (-0.06 ppm)
19
17
This article is protected by copyright. All rights reserved.

10.1002/chem.201905277
Chemistry - A European Journal
A: 55%, 1/2.6
B: 96%, < 1/25
3
4
5
6
7
8
20 (-0.23 ppm)
22 (-0.16 ppm)
24 (-0.21 ppm)
26 (-0.08 ppm)
28 (-0.18 ppm)
30 (-0.2 ppm)
21
23
25
27
29
31
A: 92%, 1/3.1
B: 85%, 1/23
A: 85%, 1/6.8
B: 84%, < 1/25
A: 77%, 1/10
B: 78%, < 1/25
A: 92%, 1/5.0
B: 73%, < 1/25
A: 81%, 1/4.0
B: 78%, < 1/25
In conclusion, excellent β-selectivity for the glycosylation with 3-Pico glucosyl donors,
without the use of participating group at C-2, was observed. This method utilizes the optimized
NIS/excess TfOH promoter system introduced by De Meo in application to 4-Pico protected sialyl
donors[11] and further optimized here in application to 3-Pico glucosyl donors. Protonation of the
Pico group was determined to be crucial in this reaction. Optimal results were obtained when a
full additional equivalent of TfOH was used in respect to NIS. The protonation of the Pico group
18
This article is protected by copyright. All rights reserved.

10.1002/chem.201905277
Chemistry - A European Journal
changes its properties transforming it into a very powerful electron-withdrawing group, which that
could have deep mechanistic implications that may trigger new exciting applications in synthetic
chemistry. NMR experiments showed the presence of an intermediate derived from the glycosyl
donor with an anomeric signal resonating at 6.9 ppm. We acquired spectroscopic evidence
consistent with this intermediate being a glycosyl triflate. High stability of this glucosyl triflate is
predicated on the electron withdrawing potency of the protonated Pico group, which is a novel
electronically deactivating (or disarming) method of stabilizing glucosyl triflates. Further
investigation of this reaction using different reaction conditions, promoter systems, and protecting
groups is described in the subsequent article in this issue.[31]
Experimental
General. Column chromatography was performed on silica gel 60 (70-230 mesh), reactions were
monitored by TLC on Kieselgel 60 F254. The compounds were detected by examination under
UV light and by charring with 10% sulfuric acid in methanol. Solvents were removed under
reduced pressure at <40 °C. CH2Cl2 and ClCH2CH2Cl (1,2-DCE) were distilled from CaH2 directly
prior to application. Molecular sieves (3 or 4 Å) used for reactions were crushed and activated in
vacuo at 390 °C for an initial 8 h and then for 2-3 h at 390 °C directly prior to application. Optical
1
rotation was measured at ‘Jasco P-2000’ polarimeter. Unless noted otherwise, H NMR spectra
were recorded in CDCl3 at 300 or 600 MHz, 13C{H} NMR spectra were recorded in CDCl3 at 75
MHz. High resolution mass spectrometry (HRMS) was carried out on ESI-TOF mass
spectrometer.
19
This article is protected by copyright. All rights reserved.

10.1002/chem.201905277
Chemistry - A European Journal
Synthesis of key glycosyl donors
Ethyl 2,4,6-tri-O-benzyl-3-O-picoloyl-1-thio-β-D-glucopyranoside (8).
1-Ethyl-3-(3-
dimethylaminopropyl) carbodiimide (0.155 g, 0.80 mmol), 4-dimethylaminopyridine (0.01 g, 0.08
mmol), and picolinic acid (0.10 g, 0.80 mmol) were added to a solution of ethyl 2,4,6-tri-O-benzyl-
1-thio-β-D-glucopyranoside[32] (0.20 g, 0.40 mmol) in CH2Cl2 (10 mL), and the resulting mixture
was stirred under argon for 30 min at rt. The reaction mixture was then diluted with CH2Cl2 (~ 00
mL) and washed with sat. aq. NaHCO3 (30 mL) and water (2 x 40 mL). The organic phase was
separated, dried with magnesium sulfate, and concentrated under reduced pressure. The residue
was purified by column chromatography on silica gel (ethyl acetate – toluene gradient elution) to
give the title compound as a white amorphous solid in 93% yield. The analytical data for 8: Rf =
21
0.6 (ethyl acetate/toluene, 2/3, v/v); [α]D +3.8 (c 1.1, CHCl3); 1H NMR: δ, 1.39 (t, 3H, J = 7.4
Hz, SCH2CH3), 2.84 (m, 2H, SCH2CH3), 3.63 (m, 1H, J5,6a = 3.7 Hz, H-5), 3.72 (dd, 1H, J2,3 = 9.4
Hz, H-2), 3.81 (m, 2H, H-6a, 6b), 4.00 (dd, 1H, J4,5 = 9.6 Hz, H-4), 4.52-4.71 (m, 6H, H-1, 5 ×
CHPh), 4.89 (dd, 1H, 2J = 10.9 Hz, CHPh), 5.66 (dd, 1H, J3,4 = 9.8 Hz, H-3), 7.08-7.41 (m, 15H,
aromatic), 7.53 (m, 1H, Pico-H), 7.86 (m, 1H, Pico-H), 8.08 (d, 1H, J = 7.8 Hz, Pico-H), 8.80 (d,
1H, J = 4.8 Hz, Pico-H) ppm; 13C{H} NMR: δ, 15.2, 25.1, 68.8, 73.5, 74.6, 75.1, 76.1, 79.1 (ⅹ2),
79.5, 85.1, 125.6, 127.0, 127.6, 127.7 (ⅹ2), 127.8 (ⅹ2), 128.1 (ⅹ2), 128.2 (ⅹ2), 128.3 (ⅹ4), 128.4
(ⅹ2), 137.0, 137.6, 137.7, 138.2, 147.9, 149.8, 164.4 ppm; ESI-TOF [M+Na]+ calcd for
C35H37NNaO6S 622.2239, found 622.2243.
Ethyl 3-O-benzoyl-2,4,6-tri-O-benzyl-1-thio-β-D-glucopyranoside (12). Benzoyl chloride
(0.02 mL, 0.18 mmol) was added to a solution of ethyl 2,4,6-tri-O-benzyl-1-thio-β-D-
glucopyranoside[32] (0.06 g, 0.12 mmol) in pyridine (10 mL) and the resulting mixture was stirred
20
This article is protected by copyright. All rights reserved.

10.1002/chem.201905277
Chemistry - A European Journal
under argon for 3 h at 80 °C. The reaction mixture was then diluted with CH2Cl2 (~100 mL) and
washed with 1 N aq. HCl (3 x 20 mL) and water (2 x 40 mL). The organic phase was separated,
dried with magnesium sulfate, filtered, and concentrated under reduced pressure. The residue was
purified by column chromatography on silica gel (ethyl acetate – hexane gradient elution) to give
the title compound as a white amorphous solid in 96% yield. The analytical data for 12: Rf = 0.65
(ethyl acetate/hexane, 2/3, v/v); [α]D21 +19.2 (c 1.2, CHCl3); 1H NMR: δ, 1.40 (t, 3H, J = 7.4 Hz,
SCH2CH3), 2.85 (m, 2H, SCH2CH3), 3.60 (dd, 1H, J2,3 = 9.5 Hz, H-2), 3.60 (m, 1H, H-5), 3.76-
3.81 (m, 2H, H-6a, 6b), 3.86 (dd, 1H, J4,5 = 9.4 Hz, H-4), 4.52-4.64 (m, 4H, 2 x CH2Ph), 4.62 (d,
2
2
1H, J1,2 = 6.6 Hz, H-1), 4.70 (dd, 1H, J = 12.1 Hz, CHPh), 4.84 (dd, 1H, J = 10.8 Hz, CHPh),
5.60 (dd, 1H, J3,4 = 9.2 Hz, H-3), 7.05-8.02 (m, 20H, aromatic) ppm; 13C{H} NMR: δ, 15.1, 25.1,
63.8, 75.2, 75.6, 76.0, 77.2, 77.8, 81.7, 85.1, 86.6, 127.8, 127.9 (ⅹ2), 128.0 (ⅹ2), 128.1 (ⅹ2),
128.3 (ⅹ2), 128.4 (ⅹ2), 128.5 (ⅹ6), 129.7 (ⅹ2), 129.9, 133.1, 137.5, 137.8, 138.2, 166.2 ppm;
ESI-TOF [M+Na]+ calcd for C36H38NaO6S 621.2287, found 621.2291.
Ethyl 2,4,6-tri-O-methyl-3-O-picoloyl-1-thio-β-D-glucopyranoside (15) was prepared as a
white amorphous solid from ethyl 2,4,6-tri-O-methyl-1-thio-β-D-glucopyranoside (see the SI for
the synthesis) in 97% yield using the same protocol as that described for the synthesis of compound
26
8. The analytical data for 15: Rf = 0.65 (ethyl acetate/dichloromethane, 1/1, v/v); [α]D -24.6 (c
1.0, CHCl3); 1H NMR: δ, 1.36 (t, 3H, J = 7.5 Hz, SCH2CH3), 2.81 (m, 2H, SCH2CH3), 3.41 (dd,
1H, J2,3 = 9.4 Hz, H-2), 3.45, 3.45, 3.52 (3 s, 9H, 3 x OCH3), 3.52 (m, 1H, H-5), 3.62-3.68 (m, 2H,
J6a,6b = 11.1 Hz, H-4, 6a), 3.72 (dd, 1H, H-6b), 4.51 (d, 1H, J1,2 = 9.3 Hz, H-1), 5.51 (dd, 1H, J3,4
= 9.2 Hz, H-3), 7.57 (m, 1H, Pico-H), 7.93 (m, 1H, Pico-H), 8.28 (d, 1H, J = 7.8 Hz, Pico-H), 8.86
(d, 1H, J = 5.5 Hz, Pico-H) ppm; 13C{H} NMR: δ, 14.9, 25.0, 59.4, 60.2, 60.5, 71.2, 77.5, 78.8,
21
This article is protected by copyright. All rights reserved.

10.1002/chem.201905277
Chemistry - A European Journal
79.2, 81.0, 84.7, 125.6, 127.1, 137.13, 147.9, 150.0, 164.5 ppm; ESI-TOF [M+Na]+ calcd for
C17H25NNaO6S 394.1300, found 394.1301.
A typical protocol for reaction monitoring by low-temperature NMR.
Glycosyl donor 15 (24.6 mg, 0.066 mmol) was dried in vacuo for 1 h at rt, then dissolved in CDCl3
(1.5 mL), freshly activated molecular sieves (4 Å) were added, and the resulting mixture was
stirred under argon for 1 h at rt. The mixture was cooled to -50 °C, NIS (1.0 equiv) and TfOH (2.0
equiv) were added, and the resulting mixture was stirred for 10 min at rt. The mixture was then
cooled to -60 °C, about 1 mL of the mixture was transferred to an NMR tube, and the latter was
placed in liq. nitrogen. The NMR spectrometer was preshimed using a standard sample containing
15 (24.6 mg, 0.066 mmol) in CDCl3 (1.0 mL). The NMR tube containing frozen reaction mixture
was then inserted into the NMR spectrometer and analyzed repeatedly (1H NMR and COSY) for
10 min. Methanol (5.0 μL, 2.0 equiv) was then added and the analysis was continued as needed.
Synthesis of disaccharides
General procedure for glycosylation. Glycosyl donor (0.05 mmol) and glycosyl acceptor 2[9]
(0.038 mmol) were dried in vacuo for 1 h at rt. Molecular sieves (4 Å, 60 mg) and freshly distilled
1,2-dichloroethane or dichloromethane (1.0 mL) were added for 50 mM reaction and the resulting
mixture was stirred under argon for 1 h at rt. Molecular sieves (4 Å, 150 mg) and freshly distilled
1,2-dichloroethane or dichloromethane (10 mL) were added for 5 mM reaction and the resulting
mixture was stirred under argon for 1 h at rt. The mixture was cooled (-30 °C or -50 °C), and the
promoter was added. The external cooling was removed, and the reaction mixture was allowed to
warm gradually to rt for reactions longer than 2 h. Upon completion (see the time listed in Tables),
22
This article is protected by copyright. All rights reserved.

10.1002/chem.201905277
Chemistry - A European Journal
the solid was filtered off through a pad of Celite and rinsed successively with dichloromethane.
The combined filtrate (~ 30 mL) was washed with either sat. aq. NaHCO3 (10 mL) or 10 % NaS2O3
(10 mL), for DMTST or NIS/TfOH promoted reactions, respectively), and water (2 x 10 mL). The
organic phase was separated, dried with magnesium sulfate, and concentrated under reduced
pressure. The residue was purified by column chromatography on silica gel (ethyl acetate –
hexanes gradient elution) to afford the corresponding disaccharide. Diastereomeric ratios were
determined by comparison of the integral intensities of relevant signals in 1H NMR spectra.
Methyl O-(2,4,6-tri-O-benzyl-3-O-picoloyl-β-D-glucopyranosyl)-(16)-2,3,4-tri-O-benzyl-α-
D-glucopyranoside (9) was obtained as a white amorphous solid from glycosyl donor 8 and
acceptor 2[9b] following general glycosylation procedure in 85% yield (0.032 g, 0.032 mmol, α/β
= 1/23). Analytical data for 9: Rf = 0.45 (ethyl acetate/toluene, 2/3, v/v); 1H NMR: δ, 3.40 (s, 3H,
OCH3), 3.55-3.60 (m, 3H, H-2, 4, 5’), 3.66 (dd, 1H, J2’,3’ = 9.6 Hz, H-2’), 3.75 (dd, 1H, J6a,6b
10.9 Hz, H-6a), 3.80 (m, 2H, H-6a’, 6b’), 3.89 (m, 1H, J5,6a = 5.1 Hz, H-5), 3.95 (dd, 1H, J4’,5’
=
=
9.6 Hz, H-4’), 4.06 (dd, 1H, J3,4 = 9.1 Hz, H-3), 4.23 (dd, 1H, H-6b), 4.51 (d, 1H, J1’,2’ = 7.7 Hz,
H-1’), 4.51-5.05 (m, 12H, 6 × CH2Ph), 4.66 (d, 1H, J1,2 = 3.1 Hz, H-1), 5.61 (dd, 1H, J3’,4’ = 9.5
Hz, H-3’), 7.00-7.42 (m, 30H, aromatic), 7.52 (m, 1H, Pico-H), 7.84 (m, 1H, Pico-H), 8.02 (d, 1H,
J = 7.8 Hz, Pico-H), 8.80 (d, 1H, J = 3.9 Hz, Pico-H) ppm; 13C{H} NMR: δ, 55.1, 59.4, 60.1, 60.8,
68.5, 70.0, 71.2, 73.4, 74.7, 75.0, 75.9, 77.2, 77.7, 78.0, 79.8, 81.1, 82.1, 98.0, 103.6, 125.6, 126.5,
127.0, 127.7, 127.8, 127.9 (×3), 128.1 (×4), 128.4 (×4), 137.0, 138.1, 138.2, 138.6, 147.9, 149.4,
150.0, 164.6 ppm; ESI-TOF [M+Na]+ calcd for C61H63NNaO12 1024.4248, found 1024.4241.
23
This article is protected by copyright. All rights reserved.

10.1002/chem.201905277
Chemistry - A European Journal
Methyl O-(3-O-benzoyl-2,4,6-tri-O-benzyl-α/β-D-glucopyranosyl)-(16)-2,3,4-tri-O-benzyl-
α-D-glucopyranoside (13) was obtained as a white amorphous solid from glycosyl donor 12 and
acceptor 2 following general glycosylation procedure in 86% yield (0.033 g, 0.033 mmol, α/β =
1
1/1.0). Selected analytical data for α-13: Rf = 0.55 (ethyl acetate/hexane, 2/3, v/v); H NMR: δ,
3.38 (s, 3H, OCH3), 5.04 (d, 1H, J1’,2’ = 3.5 Hz, H-1’) 5.79 (dd, 1H, J3’,4’ = 9.6 Hz, H-3’) ppm;
13C{H} NMR: δ, 96.9, 98.0 ppm. Selected analytical data for β-13: Rf = 0.55 (ethyl acetate/hexane,
2/3, v/v); 1H NMR: δ, 3.35 (s, 3H, O CH3), 4.16 (m, 1H, H-5), 5.48 (dd, 1H, J3’,4’ = 9.4 Hz, H-3’)
13
ppm; C{H} NMR: δ, 98.1, 103.8 ppm; ESI-TOF [M+Na]+ calcd for C62H64NaO12 1023.4295,
found 1023.4307.
Methyl O-(2,4,6-tri-O-methyl-3-O-picoloyl-β-D-glucopyranosyl)-(16)-2,3,4-tri-O-benzyl-α-
D-glucopyranoside (19) was obtained as a white solid from glycosyl donor 15 and acceptor 2
following general glycosylation procedure in 80% yield (0.026 g, 0.033 mmol, α/β < 1/25).
Analytical data for 19: Rf = 0.35 (ethyl acetate/dichloromethane, 1/1, v/v); 1H NMR: δ, 3.39 (dd,
1H, J2’,3’ = 9.7 Hz, H-2’), 3.42, 3.43, 3.44 (3 s, 9H, OCH3), 3.47-3.59 (m, 4H, H-2, 4, 4’, 5’), 3.65-
3.71 (m, 2H, H-6a’, 6b’), 3.70 (dd, 1H, J6a,6b = 10.9 Hz, H-6a), 3.87 (m, 1H, H-5), 4.04 (dd, 1H,
J3,4 = 9.3 Hz, H-3), 4.20 (dd, 1H, H-6b), 4.30 (d, 1H, J1’,2’ = 7.7 Hz, H-1’), 4.65 (d, 1H, J1,2 = 3.6
Hz, H-1), 4.67 (d, 1H, 2J = 11.3 Hz, CHPh), 4.70 (d, 1H, 2J = 12.0 Hz, CHPh), 4.84 (d, 1H, 2J =
12.1 Hz, CHPh), 4.88 (d, 1H, 2J = 10.7 Hz, CHPh), 4.98 (d, 1H, 2J = 10.9 Hz, CHPh), 5.02 (d, 1H,
2J = 10.8 Hz, CHPh), 5.44 (dd, 1H, J3’,4’ = 9.4 Hz, H-3’), 7.31-7.42 (m, 15H, aromatic), 7.56 (m,
1H, Pico-H), 7.92 (m, 1H, Pico-H), 8.24 (d, 1H, J = 7.9 Hz, Pico-H), 8.86 (d, 1H, J = 6.5 Hz, Pico-
H) ppm; 13C{H} NMR: δ, 55.1, 59.4, 60.1, 60.8, 68.5, 70.0, 71.2, 73.4, 74.7, 75.0, 75.9, 77.2, 77.5,
77.7, 79.8, 81.1, 82.1, 98.0, 103.6, 125.6, 127.0, 127.7, 127.8, 127.9 (×3), 128.1 (×4), 128.4 (×2),
24
This article is protected by copyright. All rights reserved.

10.1002/chem.201905277
Chemistry - A European Journal
128.5 (×3), 137.0, 138.1, 138.2, 138.6, 147.9, 150.0, 164.6 ppm; ESI-TOF [M+Na]+ calcd for
C43H51NNaO12 796.3309, found 796.3310.
Methyl O-(2,4-di-O-benzyl-6-O-methyl-3-O-picoloyl-β-D-glucopyranosyl)-(16)-2,3,4-tri-
O-benzyl-α-D-glucopyranoside (21) was obtained as a clear syrup from glycosyl donor 20 and
acceptor 2 following general glycosylation procedure in 96% yield (0.031 g, 0.036 mmol, α/β <
1/25). Analytical data for 21: Rf = 0.40 (ethyl acetate/toluene, 2/3, v/v); 1H NMR: δ, 3.39, 3.43 (2
s, 6H, 2 x OCH3), 3.48-3.59 (m, 4H, H-2, 2’, 4, 5’), 3.63-3.69 (m, 2H, 6a’, 6b’), 3.73 (dd, 1H, J6a,6b
= 11.0 Hz, H-6a), 3.89 (m, 1H, H-5), 3.91 (dd, 1H, J4’,5’ = 9.5 Hz, H-4’), 4.05 (dd, 1H, J3,4 = 9.3
Hz, H-3), 4.25 (dd, 1H, H-6b), 4.51 (d, 1H, J1’,2’ = 7.8 Hz, H-1’), 4.56-4.66 (m, 4H, 2 ⅹCH2Ph),
4.62 (d, 1H, J1,2 = 3.6 Hz, H-1), 4.71 (d, 1H, 2J = 10.7 Hz, CHPh), 4.82-4.87 (m, 4H, 2 ⅹCH2Ph),
5.03 (d, 1H, 2J = 10.8 Hz, CHPh), 5.60 (dd, 1H, J3’,4’ = 9.5 Hz, H-3’), 7.02-7.42 (m, 25H, aromatic),
7.52 (m, 1H, Pico-H), 7.84 (m, 1H, Pico-H), 8.00 (d, 1H, J = 7.8 Hz, Pico-H), 8.80 (d, 1H, J = 5.6
Hz, Pico-H) ppm; 13C{H} NMR: δ, 55.2, 59.3, 68.5, 70.0, 70.9, 73.3, 74.3, 74.5, 74.7, 75.0, 75.7,
75.8, 77.2, 77.4, 78.1, 79.7, 82.0, 97.9, 103.8, 125.5, 127.3, 127.7 (×2), 127.8, 127.9 (×3), 128.0
(×6), 128.1 (×2), 128.2 (×2), 128.4 (×4), 128.5 (×2), 136.9, 137.6, 137.9, 138.1, 138.2, 138.7,
147.6, 149.7, 164.3 ppm; ESI-TOF [M+Na]+ calcd for C55H59NNaO12 948.3935, found 948.3943.
Methyl O-(2,6-di-O-benzyl-4-O-methyl-3-O-picoloyl-β-D-glucopyranosyl)-(16)-2,3,4-tri-
O-benzyl-α-D-glucopyranoside (23) was obtained as a clear syrup from glycosyl donor 22 and
acceptor 2 following general glycosylation procedure in 85% yield (0.030 g, 0.032 mmol, α/β =
1/23). Analytical data for 23: Rf = 0.40 (ethyl acetate/toluene, 2/3, v/v); 1H NMR: δ, 3.39, 3.39 (2
s, 6H, 2 x OCH3), 3.49-3.68 (m, 5H, H-2, 2’, 4, 4’, 5’), 3.76 (dd, 1H, J6a,6b = 11.0 Hz, H-6a), 3.78
25
This article is protected by copyright. All rights reserved.

10.1002/chem.201905277
Chemistry - A European Journal
(m, 2H, H-6a’, 6b’), 3.90 (m, 1H, H-5), 4.05 (dd, 1H, J3,4 = 9.3 Hz, H-3), 4.23 (dd, 1H, H-6b), 4.47
(d, 1H, J1’,2’ = 7.8 Hz, H-1’), 4.57 (d, 1H, 2J = 11.1 Hz, CHPh), 4.61 (d, 1H, 2J = 12.1 Hz, CHPh),
4.67 (d, 1H, J1,2 = 4.4 Hz, H-1), 4.66-4.69 (m, 2H, CH2Ph), 4.72 (d, 1H, 2J = 12.0 Hz, CHPh), 4.80-
4.89 (m, 4H, 2 ⅹCH2Ph), 5.02 (d, 1H, 2J = 10.8 Hz, CHPh), 5.52 (dd, 1H, J3’,4’ = 9.4 Hz, H-3’),
6.98-7.39 (m, 25H, aromatic), 7.53 (m, 1H, Pico-H), 7.87 (m, 1H, Pico-H), 8.10 (d, 1H, J = 7.7
Hz, Pico-H), 8.83 (d, 1H, J = 5.5 Hz, Pico-H) ppm; 13C{H} NMR: δ, 55.1, 60.1, 68.7 (×2), 70.0,
73.4 (x2), 74.4, 74.8, 75.0, 75.7, 77.2, 77.9, 78.9, 79.7, 82.0, 98.0, 103.8, 125.5, 126.9, 127.3,
127.6 (×4), 127.7, 127.8 (×3), 127.9 (×2), 128.0 (×5), 128.2 (×2), 128.3 (×2), 128.4 (×4), 128.5
(×2), 137.0, 137.9, 138.1, 138.2 (×2), 138.7, 147.8, 149.8, 164.3 ppm; ESI-TOF [M+Na]+ calcd
for C55H59NNaO12 948.3935, found 948.3943.
Methyl O-(4,6-di-O-benzyl-2-O-methyl-3-O-picoloyl-β-D-glucopyranosyl)-(16)-2,3,4-tri-
O-benzyl-α-D-glucopyranoside (25) was obtained as a clear oil from glycosyl donor 24 and
acceptor 2 following general glycosylation procedure in 84% yield (0.029 g, 0.032 mmol, α/β <
1/25). Analytical data for 25: Rf = 0.35 (ethyl acetate/toluene, 2/3, v/v); 1H NMR: δ, 3.40 (s, 3H,
OCH3), 3.43 (dd, 1H, J2’,3’ = 7.2 Hz, H-2’), 3.53 (s, 3H, OCH3), 3.55-3.61 (m, 3H, H-2, 4, 5’), 3.78
(dd, 1H, J6a,6b = 10.9 Hz, H-6a), 3.80 (m, 2H, H-6a’, 6b’), 3.89 (m, 1H, H-5), 3.94 (dd, 1H, J4’,5’
=
9.5 Hz, H-4’), 4.06 (dd, 1H, J3,4 = 9.3 Hz, H-3), 4.21 (dd, 1H, H-6b), 4.46 (d, 1H, J1’,2’ = 7.7 Hz,
H-1’), 4.53-4.63 (m, 3H, 3 ⅹCHPh), 4.67 (d, 1H, J1,2 = 4.5 Hz, H-1), 4.66-4.73 (m, 3H, 3 ⅹ
CHPh), 4.85 (d, 1H, 2J = 12.0 Hz, CHPh), 4.88 (d, 1H, 2J = 10.7 Hz, CHPh), 4.96 (d, 1H, 2J = 10.9
2
Hz, CHPh), 5.03 (d, 1H, J = 10.8 Hz, CHPh), 5.54 (dd, 1H, J3’,4’ = 9.5 Hz, H-3’), 7.07-7.40 (m,
25H, aromatic), 7.54 (m, 1H, Pico-H), 7.89 (m, 1H, Pico-H), 8.16 (d, 1H, J = 7.9 Hz, Pico-H),
8.84 (d, 1H, J = 6.3 Hz, Pico-H) ppm; 13C{H} NMR: δ, 55.2, 60.9, 68.6 (×2), 69.9, 73.4 (×2), 74.5,
26
This article is protected by copyright. All rights reserved.

10.1002/chem.201905277
Chemistry - A European Journal
74.9, 75.0, 75.9, 77.2, 77.7, 77.9, 79.7, 81.2, 82.2, 98.0, 103.6, 125.6, 127.0, 127.6 (×2), 127.7
(×3), 127.8, 127.9 (×2), 128.0 (×3), 128.1 (×3), 128.2 (×2), 128.3 (×2), 128.4 (×2), 128.5 (×4),
137.0, 137.6, 138.1 (×2), 138.2, 138.6, 147.8, 149.9, 164.5 ppm; ESI-TOF [M+Na]+ calcd for
C55H59NNaO12 948.3935, found 948.3944.
Methyl O-(6-O-benzyl-2,4-di-O-methyl-3-O-picoloyl-β-D-glucopyranosyl)-(16)-2,3,4-tri-
O-benzyl-α-D-glucopyranoside (27) was obtained as a clear syrup from glycosyl donor 26 and
acceptor 2 following general glycosylation procedure in 78% yield (0.025 g, 0.030 mmol, α/β <
1/25). Analytical data for 27: Rf = 0.50 (ethyl acetate/dichloromethane, 1/1, v/v); 1H NMR: δ, 3.40
(dd, 1H, J2’,3’ = 9.3 Hz, H-2’), 3.40, 3.43, 3.52 (3 s, 9H, 3 x OCH3), 3.54-3.60 (m, 3H, H-2, 4, 5’),
3.65 (dd, 1H, J4’,5’ = 9.6 Hz, H-4’), 3.78 (dd, 1H, J6a,6b = 10.9 Hz, H-6a), 3.79 (m, 2H, H-6a’, 6b’),
3.88 (m, 1H, H-5), 4.05 (dd, 1H, J3,4 = 9.2 Hz, H-3), 4.22 (dd, 1H, H-6b), 4.43 (d, 1H, J1’,2’ = 7.8
Hz, H-1’), 4.60 (d, 1H, 2J = 12.1 Hz, CHPh), 4.67 (d, 1H, J1,2 = 2.2 Hz, H-1), 4.67-4.72 (m, 3H, 3
ⅹCHPh), 4.85 (d, 1H, 2J = 12.1 Hz, CHPh), 4.89 (d, 1H, 2J = 11.1 Hz, CHPh), 4.95 (dd, 1H, 2J =
10.9 Hz, CHPh), 5.03 (dd, 1H, 2J = 10.8 Hz, CHPh), 5.46 (dd, 1H, J3’,4’ = 9.5 Hz, H-3’), 7.30-7.42
(m, 20H, aromatic), 7.55 (m, 1H, Pico-H), 7.92 (m, 1H, Pico-H), 8.26 (d, 1H, J = 7.8 Hz, Pico-H),
13
8.66 (d, 1H, J = 5.6 Hz, Pico-H) ppm; C{H} NMR: δ, 55.2, 60.1, 60.9, 68.5, 69.9, 73.4 (×2),
74.8, 75.0, 75.9, 77.2, 77.6, 77.8, 77.9, 79.7, 81.1, 82.1, 98.0, 103.6, 125.6, 127.0, 127.5, 127.6
(×2), 127.7, 127.8 (×3), 127.9, 128.1 (×4), 128.3 (×2), 128.4 (×2), 128.5 (×4), 137.1, 138.1, 138.2,
138.6, 147.9, 150.0, 164.6 ppm; ESI-TOF [M+Na]+ calcd for C49H55NNaO12 872.3622, found
872.3628.
27
This article is protected by copyright. All rights reserved.

10.1002/chem.201905277
Chemistry - A European Journal
Methyl O-(4-O-benzyl-2,6-di-O-methyl-3-O-picoloyl-β-D-glucopyranosyl)-(16)-2,3,4-tri-
O-benzyl-α-D-glucopyranoside (29) was obtained as a clear syrup from glycosyl donor 28 and
acceptor 2 following general glycosylation procedure in 73% yield (0.023 g, 0.028 mmol, α/β <
1/25). Analytical data for 29: Rf = 0.60 (ethyl acetate/dichloromethane, 1/1, v/v); 1H NMR: δ, 3.39
(dd, 1H, J2’,3’ = 9.6 Hz, H-2’), 3.42, 3.43, 3.50 (3 s, 9H, 3 x OCH3), 3.51-3.59 (m, 3H, H-2, 4, 5’),
3.68 (m, 2H, H-6a’, 6b’), 3.76 (dd, 1H, J6a,6b = 10.9 Hz, H-6a), 3.87 (m, 1H, H-5), 3.90 (dd, 1H,
J3,4 = 9.6 Hz, H-3), 4.05 (dd, 1H, J4’,5’ = 9.4 Hz, H-4’), 4.22 (dd, 1H, H-6b), 4.46 (d, 1H, J1’,2’ = 7.7
Hz, H-1’), 4.61-4.69 (m, 3H, 3 ⅹCHPh), 4.66 (d, 1H, J1,2 = 3.7 Hz, H-1), 4.71 (d, 1H, 2J = 10.2
Hz, CHPh), 4.85 (d, 1H, 2J = 12.1 Hz, CHPh), 4.87 (d, 1H, 2J = 10.7 Hz, CHPh), 4.95 (d, 1H, 2J =
10.9 Hz, CHPh), 5.03 (d, 1H, 2J = 10.8 Hz, CHPh), 5.53 (dd, 1H, J3’,4’ = 9.5 Hz, H-3’), 7.18-7.42
(m, 20H, aromatic), 7.53 (m, 1H, Pico-H), 7.88 (m, 1H, Pico-H), 8.15 (d, 1H, 3J = 7.8 Hz, Pico-
3
13
H), 8.84 (d, 1H, J = 6.4 Hz, Pico-H) ppm; C{H} NMR: δ, 55.1, 59.4, 60.8, 68.5, 69.8, 71.0,
73.4, 74.5, 74.7, 75.0, 75.8, 75.9, 77.2, 77.7, 78.0, 79.7, 81.2, 82.1, 79.9, 103.6, 125.5, 126.9, 127.7
(×2), 127.8, 127.9 (×3), 128.0 (×2), 128.1 (×4), 128.2 (×2), 128.4 (×2), 128.5 (×3), 137.0, 137.7,
138.1, 138.2, 138.6, 147.8, 149.9, 164.5 ppm; ESI-TOF [M+Na]+ calcd for C49H55NNaO12
872.3612, found 872.3622.
Methyl O-(2-O-benzyl-4,6-di-O-methyl-3-O-picoloyl-β-D-glucopyranosyl)-(16)-2,3,4-tri-
O-benzyl-α-D-glucopyranoside (31) was obtained as a clear oil from glycosyl donor 30 and
acceptor 2 following general glycosylation procedure in 78% yield (α/β < 1/25). Analytical data
for 31: Rf = 0.55 (ethyl acetate/dichloromethane, 1/1, v/v); 1H NMR: δ, 3.38, 3.43, 3.47 (3 s, 9H,
OCH3), 3.48-3.67 (m, 7H, H-2, 2’, 4, 4’, 5’, 6a’, 6b’), 3.68 (dd, 1H, J6a,6b = 10.9 Hz, H-6a), 3.86
(m, 1H, H-5), 4.04 (dd, 1H, J3,4 = 9.3 Hz, H-3), 4.24 (dd, 1H, H-6b), 4.47 (d, 1H, J1’,2’ = 7.7 Hz,
28
This article is protected by copyright. All rights reserved.

10.1002/chem.201905277
Chemistry - A European Journal
2
H-1’), 4.55-4.65 (m, 2H, CH2Ph), 4.65 (d, 1H, J1,2 = 3.1 Hz, H-1), 4.70 (d, 1H, J = 12.1 Hz,
CHPh), 4.81-4.88 (m, 4H, 2 ⅹCH2Ph), 5.03 (d, 1H, 2J = 10.9 Hz, CHPh), 5.51 (dd, 1H, J3’,4’ = 9.4
Hz, H-3’), 7.02-7.41 (m, 20H, aromatic), 7.51 (m, 1H, Pico-H), 7.87 (m, 1H, Pico-H), 8.08 (d, 1H,
J = 7.8 Hz, Pico-H), 8.83 (d, 1H, J = 5.6 Hz, Pico-H) ppm; 13C{H} NMR: δ, 55.1, 59.4, 60.1, 68.6,
70.1, 73.3, 74.3, 74.7, 75.0, 75.7, 77.2, 78.1, 79.0, 79.8, 82.0, 98.0, 103.8, 125.4, 126.9, 127.3,
127.6, 127.8 (×2), 27.9 (×3), 128.0 (×4), 128.1 (×2), 128.4 (×4), 128.5 (×2), 136.9, 137.9, 138.1,
138.2, 138.7, 147.7, 149.9, 164.4 ppm; ESI-TOF [M+Na]+ calcd for C49H55NNaO12 872.3612,
found 872.3627.
ASSOCIATED CONTENT
Supporting Information
Additional experimental data and NMR spectra for all new compounds. This material is available
free of charge via the Internet at
AUTHOR INFORMATION
Corresponding Author
* Department of Chemistry and Biochemistry, University of Missouri – St. Louis, One University
Boulevard, St. Louis, Missouri 63121, USA; demchenkoa@umsl.edu
Author Contributions
The manuscript was written through contributions of all authors. / All authors have given approval
to the final version of the manuscript.
29
This article is protected by copyright. All rights reserved.