Remote Participation during Glycosylation Reactions of Galactose Building Blocks: Direct Evidence from Cryogenic Vibrational Spectroscopy

Article abstract of DOI:10.1002/anie.201916245

The stereoselective formation of 1,2-cis-glycosidic bonds is challenging. However, 1,2-cis-selectivity can be induced by remote participation of C4 or C6 ester groups. Reactions involving remote participation are believed to proceed via a key ionic intermediate, the glycosyl cation. Although mechanistic pathways were postulated many years ago, the structure of the reaction intermediates remained elusive owing to their short-lived nature. Herein, we unravel the structure of glycosyl cations involved in remote participation reactions via cryogenic vibrational spectroscopy and first principles theory. Acetyl groups at C4 ensure α-selective galactosylations by forming a covalent bond to the anomeric carbon in dioxolenium-type ions. Unexpectedly, also benzyl ether protecting groups can engage in remote participation and promote the stereoselective formation of 1,2-cis-glycosidic bonds.

Full text of DOI:10.1002/anie.201916245

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Direct Evidence for Remote Participation in Galactose Building
Blocks during Glycosylations Revealed by Cryogenic Vibrational
Spectroscopy
Authors: Mateusz Marianski, Eike Mucha, Kim Greis, Sooyeon Moon,
Alonso Pardo, Carla Kirschbaum, Daniel Thomas, Gerard
Meijer, Gert von Helden, Kerry Gilmore, Peter Seeberger, and
Kevin Pagel
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: Angew. Chem. Int. Ed. 10.1002/anie.201916245
Angew. Chem. 10.1002/ange.201916245
Link to VoR: http://dx.doi.org/10.1002/anie.201916245
http://dx.doi.org/10.1002/ange.201916245

10.1002/anie.201916245
Angewandte Chemie International Edition
Communication
Remote Participation during Glycosylation Reactions of Galac-
tose Building Blocks: Direct Evidence from Cryogenic Vibrational
Spectroscopy
Mateusz Marianski,*[a],[b] Eike Mucha, [c] Kim Greis, [c] Sooyeon Moon, [d],[e] Alonso Pardo, [d],[e] Carla
Kirschbaum, [c],[e] Daniel A. Thomas, [c] Gerard Meijer, [c] Gert von Helden, [c] Kerry Gilmore, [d] Peter H.
Seeberger, [d],[e] Kevin Pagel*[c],[e]
Abstract: The stereoselective formation of 1,2-cis-glycosidic bonds is
challenging. However, 1,2-cis-selectivity can be induced by remote
participation of C4 or C6 ester groups. Reactions involving remote
participation are believed to proceed via a key ionic intermediate, the
glycosyl cation. Although mechanistic pathways were postulated
many years ago, the structure of the reaction intermediates remained
elusive due to their short-lived nature. Here, we unravel the structure
of glycosyl cations involved in remote participation reactions via cryo-
genic vibrational spectroscopy and first principles theory. Acetyl
groups at C4 ensure α-selective galactosylations by forming a cova-
lent bond to the anomeric carbon in dioxolenium-type ions. Unexpect-
edly, also benzyl ether protecting groups can engage in remote par-
ticipation and promote the stereoselective formation of 1,2-cis-glyco-
sidic bonds.
attack from the cis-side, such as chiral auxiliaries,[4] intramolecu-
lar rearrangements,[5] or specific activators.[6] A more widely used
alternative involves the remote participation of acyl protecting
groups at the C4 or C6 position.[7] Similar to C2-participation, the
reaction mechanism is believed to proceed via a key ionic inter-
mediate, the glycosyl cation, where the trans-side of the anomeric
carbon is shielded by remote ester groups (Figure 1).[8] Remote
participation can, via a range of different possible intermediates,
induce a nucleophilic attack and lead to cis-glycosidic bonds, but
the reactive and short-lived nature of these intermediates has
greatly impeded their direct experimental characterization.
Glycans are biopolymers that consist of many different building
blocks and exhibit complex regio- and stereochemistry such that
their chemical synthesis has traditionally been very time consum-
ing. Although regiocontrol is readily exerted using orthogonal pro-
tecting group strategies,[1] stereocontrol during glycosidic bond
formation can be challenging. In many cases, custom-tailored ap-
proaches require empirical optimizations. For instance, 1,2-trans
glycosidic bonds are most reliably obtained using C2-participating
2-O-ester protecting groups (PGs)[2]. In these systems, the cleav-
age of a leaving group at the anomeric carbon promotes the for-
mation of a cyclic 1,2-cis dioxolenium intermediate that facilitates
a nucleophilic attack from the trans-side.[3] Stereocontrol during
the formation of 1,2-cis glycosides, however, requires shielding of
the trans-side. Various strategies are used to aid a nucleophilic
[a]
Prof. Dr. Mateusz Marianski
Department of Chemistry and Biochemistry
Hunter College
695 Park Ave, 10065, New York, NY
mmarians@hunter.cuny.edu
Prof. Dr. Mateusz Marianski
The Ph.D. Program in Chemistry
The Graduate Center of the City University of New York
365 5th Ave, New York, NY, 10016
Figure 1: Stereoselective formation of 1,2-cis-galactosides involving remote
participation of acetyl groups. While the reaction mechanism is believed to pro-
ceed via a glycosyl cation, the exact structure of the intermediate remains un-
clear. The intermediate ion structure could range from oxocarbenium-type to
hybrid and dioxolenium-type featuring a covalent bond to the anomeric carbon.
The exact intermediate structure impacts the stereochemistry of the newly
formed glycosidic bond. LG: leaving group; Nu: nucleophile.
[b]
[c]
A series of reactions (Figure 2) served as the basis to systemati-
cally investigate the effect of remote participation on 1,2-cis gly-
cosylations. The nucleophilic substitution of isopropanol on the
four galactose building blocks 2,3,4,6-tetra-O-benzyl-D-galacto-
pyranoside (Bn), 4-O-acetyl-2,3,6-tri-O-benzyl-D-galactopyra-
noside (4Ac), 6-O-acetyl-2,3,4-tri-O-benzyl-D-galactopyranoside
(6Ac) and 4,6-di-O-acetyl-2,3-di-O-benzyl-D-galactopyranoside
(4,6Ac) was carried out at five different temperatures between
‑50 oC and 30 oC under well-defined reaction conditions using
flow chemistry (see SI for details). These glycosylating agents fall
into two groups that either yield mainly α-glycosides (4Ac and
4,6Ac), or little α-glycosides (Bn and 6Ac). The difference in ste-
reoselectivity between the two groups is consistent for all meas-
Eike Mucha, Kim Greis, Carla Kirschbaum, Dr. Daniel A. Thomas,
Prof. Dr. Gerard Meijer, Prof. Dr. Gert von Helden, Prof. Dr. Kevin
Pagel
Fritz-Haber-Institut der Max-Planck-Gesellschaft,
Faradayweg 4-6, 14195, Berlin, Germany
Kevin.pagel@fu-berlin.de
[d]
[e]
Dr. Sooyeon Moon, Alonso Pardo, Dr. Kerry Gilmore, Prof. Dr. Peter
H. Seeberger
Max-Planck-Institut für Kolloid und Grenzflächenforschung, Am
Mühlenberg 1, 14476, Potsdam, Germany
Kim Greis, Dr. Sooyeon Moon, Alonso Pardo, Carla Kirschbaum,
Prof. Dr. Peter H. Seeberger, Prof. Dr. Kevin Pagel
Institut für Chemie und Biochemie, Freie Universität Berlin, Takus-
traße 3, 14195, Berlin, Germany
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/anie.201916245
Angewandte Chemie International Edition
Communication
ured temperatures and suggests that different reaction intermedi-
ates are responsible for the reaction outcome. If the reactions
would solely proceed via an SN2-mechanism, the β-glycoside
would be formed exclusively, because -acetimidate donors are
used. However, the relative abundance of α-products between
20% and 90% suggests dissociative SN1-mechanisms involving
glycosyl cations. All building blocks show a consistent increase of
α-selectivity with increasing temperature, which is in agreement
with the general assumption that a dissociative SN1-mechanism
is favored at higher temperatures due to a gain in entropy.[9]
cryogenic ion spectroscopy and first-principles theory to deter-
mine the exact structures of glycosyl cations of custom-tailored
model glycans carrying C2-participating protecting groups.[3a] The
group of Boltje et al., on the other hand, applied infrared multipho-
ton dissociation (IRMPD) to investigate structures of a series of
mannosyl donors.[3b] Recently, the same technique was applied
to investigate whether remote participation promotes high β-se-
lectivity of bicyclic 6,3-uronic acid lactones.[12] In this work, we use
more common, unconstrained glycosyl donors to unravel the
structure of key intermediates involved in remote participation[7].
It is important to note that acetyl protecting groups at the C3 po-
sition are also hypothesized to participate during glycosylation re-
actions via dioxolenium intermediates. For brevity, however, the
investigation of C3-acetylated donors was omitted in this commu-
nication.
The experimental setup to investigate the structure of glycosyl
cations using vibrational spectroscopy in helium nanodroplets
was described in detail before.[13] Briefly, glycosyl cations were
generated by nano-electrospray ionization (nESI) and in-source
fragmentation of imidate or thioglycoside precursors (see SI). The
m/z-selected ions were accumulated inside a cryogenic (78 K)
hexapole ion trap, which thermalizes the trapped ions by buffer
gas cooling. Next, the ions were picked up by superfluid helium
nanodroplets with an average size of 105 helium atoms traversing
the ion trap. The helium environment inside the droplets ‘shock-
freezes’ the trapped ions to the droplet’s equilibrium temperature
of 0.4 K before they are irradiated with infrared radiation between
1000–1800 cm-1, produced by the Fritz Haber Institute Free-elec-
tron Laser (FHI-FEL[14]). The resonant absorption of multiple pho-
tons leads to the release of the bare intact ions from the droplet,
which is used as a messenger for photon absorption. Plotting the
ion count of released ions as a function of IR wavelength yields a
highly resolved IR signature of the investigated molecular ion.
To identify the structures responsible for the IR fingerprint, the
molecular ions’ conformational space was explored using an evo-
lutionary algorithm.[15] The dispersion-corrected PBE+vdWTS [16]
density-functional, implemented in FHI-aims,[17] was chosen as a
potential-energy function due to its excellent accuracy for carbo-
hydrates. Atomic connectivity is not constrained during the struc-
tural search which allows for a proton transfer or formation of new
bonds. This exhaustive conformational search included all rotata-
ble bonds and ring puckers and yielded around 300 unique con-
formations for each glycosyl cation. These ensembles included
various structural isomers, particularly those that feature a cova-
lent bond between the anomeric carbon and distinct protecting
groups. The structures were sorted according to the distance be-
tween the carbonyl oxygen of the acetyl group and the anomeric
carbon into dioxolenium ions (distance less than 2Å) and other
structures (distance greater than 2Å), (see SI). For each type, a
subset of all conformers within an energy window of 5 kcal mol-1
above the lowest-energy structure was further optimized and fre-
quencies calculated at the dispersion-corrected hybrid density-
functional theory (DFT) PBE0-D3/6-311+G(d,p) level.[18] For each
reoptimized structure, we computed RI-MP2-level single-point en-
ergies, extrapolated to the complete-basis set. The harmonic IR
spectra and free-energy corrections at 78 K were derived from the
Figure 2: Reactions of galactose building blocks to assess the impact of differ-
ent protecting group combinations (Bn=benzyl, Ac=acetyl) on the stereochemi-
cal outcome of the glycosylation reaction. The reactions were carried out at var-
ious temperatures using 2-propanol as a nucleophile and the reaction outcome
was determined by normal phase HPLC. With increasing temperature, the rela-
tive yield of α-glycosides increases, indicating that the mechanistic continuum
is effectively shifted towards SN1-type reactions. Building blocks that carry an
acetyl group at C4 (4,6Ac and 4Ac) consistently show a higher α-selectivity than
the building blocks carrying an acetyl group at C6 (6Ac) or carrying no acetyl
group (Bn).
The reaction pathway leading to the formation of α-products is
believed to proceed through a short-lived and reactive glycosyl
cation intermediate. The underlying reaction mechanism is diffi-
cult to study using classical condensed-phase techniques such as
NMR. In recent studies, various glycosyl cations were stabilized
in superacids and studied via NMR spectroscopy. However, in the
superacid medium, the protecting groups are protonated, such
that results cannot be directly translated to classical reaction con-
ditions.[10] In another approach, side products were isolated that
provided strong, but indirect, evidence for remote participation by
acetyl esters at the C4-position of glycosyl donors.[8c, 8d] Alterna-
tively, mass spectrometry-based techniques can be used to iso-
late and characterize the glycosylation reaction intermediates in
the gas phase.[3a, 3b, 11] Previously, we applied a combination of
This article is protected by copyright. All rights reserved.

10.1002/anie.201916245
Angewandte Chemie International Edition
Communication
frequency analysis.[19] Ring puckers were assigned using Cremer-
Pople coordinates.[20]
assignment. The region between 1200 cm-1 and 1800 cm-1, on the
other hand, features characteristic modes, such as C=O-stretch,
C–O–C-stretch and O–C–O-stretch vibrations. The exact position
of these bands is strongly dependent on the type of interaction
and therefore enables the structural identification. Because all hy-
droxy groups are protected, the spectral region around 3000 cm-
The IR spectra of the glycosyl cations show several absorption
bands between 1000 cm-1 and 1800 cm-1, which can be grouped
into two main regions. Below 1200 cm-1, complex C–O and C–C
stretching modes dominate the spectrum. These mostly coupled
bands, however, are not sufficient for an unambiguous structural
1 does not carry any analytical bands.
Figure 3: Infrared spectra of the glycosyl cations generated from a) the 4,6Ac and b) the 4Ac precursor. Blue traces are experimental
IR spectra while grey inverted spectra are calculated corresponding to the low energy conformers shown below in a simplified repre-
sentation. Numbers in brackets indicate free energies in kcal mol-1. Full structures are shown in the SI. The highlighted absorption
bands indicate vibrations from free acetyl groups (purple) and participating acetyl groups in dioxolenium-type structures (green). The
light green band indicates a second dioxolenium-type conformer D’. For both building blocks, the majority of ions adopt dioxolenium
structures that exhibit a covalent bond between the C4-acetyl group and the anomeric carbon (highlighted with a red dot). Structure E
shows a distorted 1C4 ring pucker, indicated by a star (*).
The 4,6Ac building block results in the highest selectivity for α-
glycosidic bond formation. The infrared spectrum of the corre-
sponding glycosyl cation (Figure 3a) displays six well-resolved
absorption bands. The low-energy structure A of this species is
predicted to adopt a 1S5 ring pucker featuring a covalent bond be-
tween the carbonyl oxygen of the C4-acetyl group and the ano-
meric carbon (1.52 Å). The characteristic bands associated with
this bridging dioxolenium motif are the symmetric and antisym-
metric O-C-O-stretch modes predicted at 1463 cm-1 and 1566 cm-
1, both in very good agreement with the experiment. Two addi-
tional characteristic bands for the non-interacting C6-acetyl group
are predicted at 1217 cm-1 and 1772 cm‑1, in line with the experi-
mental absorption bands. The alternative structure B, where the
dioxolenium motif originates from a covalent bond between the
C6-acetyl and the anomeric carbon, has a free-energy of around
5 kcal mol-1 larger with respect to structure A. Predicted symmet-
ric and antisymmetric O-C-O-stretch modes in structure B at 1459
cm-1 and 1592 cm-1 agree less well with the experimental spec-
trum. Furthermore, a free-energy difference of more than 14 kcal
mol-1, and a poor spectral match eliminates oxocarbenium-type
structure C with two non-participating acetyl groups from consid-
eration. In summary, the IR signature of the 4,6Ac galactosyl do-
nor provides direct evidence that the C4-acetyl group participates
through a dioxolenium-type intermediate that ensures α-selectiv-
ity, whereas participation via the C6-acetyl group is not observed.
The IR spectrum of 4Ac (Figure 3b) displays characteristic bands
between 1450 cm-1 and 1600 cm-1, similar to those observed for
the 4,6Ac cation. The calculated vibrations at 1465 cm-1 and 1568
cm-1 for dioxolenium structure D featuring a 1S5 ring pucker agree
well with the experiment. The lowest free-energy structure is more
stable by 0.7 kcal mol-1, but shows a slightly worse agreement to
the experimental data (see SI). The presence of multiple bands
around 1500 cm-1 indicates that a structurally very similar dioxole-
nium ion coexists (ΔF=2.0 kcal mol-1, D’). The IR spectrum fur-
thermore contains two additional low intensity bands at 1220 cm-
1 and 1780 cm‑1, indicating that a fraction of glycosyl cations adopt
other structures. These bands can be assigned to the oxonium
structure E that exhibits an interaction between C6-OBn and the
anomeric carbon. This structure has a free-energy of around
4 kcal mol-1 larger, relative to the lowest-energy structure, and the
vibrations of the free acetyl group at 1227 cm‑1 and 1763 cm‑1
align well with the experimentally resolved bands. Observing this
mode of remote participation is particularly surprising, because
synthetic chemists consider benzyl protecting groups as non-par-
ticipating. The most stable oxocarbenium structure F has a much
higher free energy (ΔF=11.9 kcal mol-1) and the predicted vibra-
tions agree less well with the experiment. The spectroscopic data
This article is protected by copyright. All rights reserved.

10.1002/anie.201916245
Angewandte Chemie International Edition
Communication
provides evidence that the 4Ac cation predominantly adopts an
α-selective dioxolenium structure, similar to the one observed for
4,6Ac.
Figure 4: Infrared spectra of the glycosyl cations generated from a) the 6Ac and b) the Bn precursors. The blue traces show the
experimental IR spectra and the grey inverted spectra are calculated spectra corresponding to the low energy conformers shown below
in a simplified representation. Numbers in brackets indicate free energies in kcal mol-1. The complete structures are shown in the SI.
The highlighted absorption bands indicate vibrations from free acetyl groups (purple) and participating acetyl groups in dioxolenium-
type structures (green). Interestingly, both structures show non-classical remote participation of benzyl groups at C4 or C6, leading to
α-selective oxonium intermediates. Structure H shows a heavily distorted ring pucker, closest to a 5S1 and 5H4 conformation, indicated
by a star (*).
The glycosyl donors 6Ac and Bn consistently produce a low to
medium abundance of α-products, suggesting that the underlying
intermediates are structurally different from those observed for
4,6Ac and 4Ac. The IR spectrum of the glycosyl cation corre-
sponding to 6Ac (Figure 4a) displays a variety of absorption
bands above 1200 cm-1. The lowest free-energy structure G
shows a dioxolenium motif with a 1C4 ring pucker and is predicted
to have two characteristic modes at 1457 cm-1 and 1593 cm-1. Alt-
hough matching vibrations can be found in the experimental spec-
trum, the higher energy band shows very low intensity. Another
low free-energy structure is oxonium structure H, which is charac-
terized by a covalent bond between the C4-oxygen and the ano-
meric carbon in a 1,4B ring-pucker. This mode of remote participa-
tion was recently reported by Boltje et al. after observing a 1,4-
anhydro-3,6-lactone as a side product during the stereoselective
synthesis of 1,2-cis mannosides. This side product strongly sug-
gests the presence of a glycosyl oxonium ion with remote partici-
pation of the oxygen atom of the C4 benzyl group.[11b] The pre-
dicted vibrations of the free acetyl group at 1220 cm-1 and 1762
cm-1 match the two prominent bands in the experimental spectrum.
A different, energetically very similar oxocarbenium structure I
features two bands of the free acetyl group at 1222 cm-1 and 1749
cm-1, with an additional distinct vibration associated to the oxo-
bands at 1336 cm-1, 1412 cm-1 and 1464 cm-1. An expanded con-
formational search, that included structural rearrangements, such
as a proton shift towards an acetyl or benzyl-group did not identify
the origin of these bands (see Supporting Information). Either yet
another (unknown) structure is present, or the harmonic approxi-
mation cannot resolve the position and intensity of anharmonic
absorption bands involving a shared proton present in some struc-
tures. Nevertheless, the lack of diagnostic bands for the dioxole-
nium-motif and the low to medium abundance of α‑products in the
test reactions suggest that the α-selective dioxolenium-type ion G
is a minor reaction channel that is suppressed by oxonium struc-
ture H and oxocarbenium structure I.
The low abundance of α-glycosides for 6Ac is comparable to the
fully benzylated building block Bn. The IR spectrum of the corre-
sponding glycosyl cation (Figure 4b) does not feature any signifi-
cant absorption bands above 1200 cm-1. Low-energy structure K,
however, is expected to show a strong diagnostic band at 1550
cm-1 that is associated to the oxocarbenium motif. A better match
for the experimental spectrum is found for oxonium structure J
that is characterized by a covalent bond between the anomeric
carbon and the C6-oxygen in a 4C1 ring pucker. Interestingly, the
remote participation via the C6 benzyl group renders this interme-
diate energetically more stable than structure K by 4.6 kcal mol-1.
Although glycosylation reactions through this intermediate should
be highly α-selective, the test reactions yield of α-product is below
50%. This finding indicates that the gain in energy due to entropy
promoting an SN1-type mechanism through structure J is weak-
ened by an enthalpic penalty, which effectively shifts the glycosyl-
ation reaction to the SN2-end of the mechanistic continuum. It has
+
carbenium [C1=O5 ] stretch at 1567 cm-1 that can be found in the
experimental spectrum. In this structure, the oxocarbenium motif
is stabilized by a long-range interaction (2.8 Å) with the C6-oxy-
gen. Although both structures H and I are energetically less fa-
vored by around 4 kcal mol-1, the match between theory and ex-
periment shows that either one or both of these intermediates are
present. None of these structures, however, explain the three
This article is protected by copyright. All rights reserved.

10.1002/anie.201916245
Angewandte Chemie International Edition
Communication
[5]
[6]
[7]
[8]
K.-H. Jung, M. Müller, R. R. Schmidt, Chemical Reviews 2000,
100, 4423-4442.
S. S. Nigudkar, A. V. Demchenko, Chemical Science 2015, 6,
to be noted, however, that the strong nucleophile used in the re-
actions generally leads to more SN2-type reactions[21]. In oligosac-
charide synthesis, the nucleophiles (glycosyl acceptors) are usu-
ally weaker and highly α-selective reactions for this building block
(α:ß-ratio of 14:1) have been reported.[7] This suggests that the α-
selectivity of this building block is based on remote participation
of the C6-benzyl group.
2687-2704.
H. S. Hahm, M. Hurevich, P. H. Seeberger, Nat. Commun. 2016, 7,
12482.
a) L. Bohé, D. Crich, Carbohydr. Res. 2015, 403, 48-59; b) A. V.
Demchenko, E. Rousson, G.-J. Boons, Tetrahedron Lett. 1999, 40,
6523-6526; c) Y. Ma, G. Lian, Y. Li, B. Yu, Chemical
Communications 2011, 47, 7515; d) D. Yao, Y. Liu, S. Yan, Y. Li,
C. Hu, N. Ding, Chemical Communications 2017, 53, 2986-2989.
a) P. O. Adero, H. Amarasekara, P. Wen, L. Bohé, D. Crich, Chem.
Rev. 2018, 118, 8242-8284; b) S. Chatterjee, S. Moon, F.
Hentschel, K. Gilmore, P. H. Seeberger, Journal of the American
Chemical Society 2018, 140, 11942-11953.
a) A. Martin, A. Arda, J. Désiré, A. Martin-Mingot, N. Probst, P.
Sinaÿ, J. Jiménez-Barbero, S. Thibaudeau, Y. Blériot, Nat. Chem.
2015, 8, 186; b) L. Lebedel, A. Arda, A. Martin, J. Desire, A.
Mingot, M. Aufiero, N. Aiguabella Font, R. Gilmour, J. Jimenez-
Barbero, Y. Bleriot, S. Thibaudeau, Angew. Chem. Int. Ed. 2019,
58, 13758-13762.
a) C. Denekamp, Y. Sandlers, J. Mass Spectrom. 2005, 40, 1055-
1063; b) H. Elferink, R. A. Mensink, W. W. A. Castelijns, O.
Jansen, J. P. J. Bruekers, J. Martens, J. Oomens, A. M. Rijs, T. J.
Boltje, Angew. Chem. Int. Ed., in press, 10.1002/anie.201902507.
H. Elferink, R. A. Mensink, W. W. A. Castelijns, O. Jansen, J. P. J.
Bruekers, J. Martens, J. Oomens, A. M. Rijs, T. J. Boltje, Angew.
Chem. Int. Ed. 2019, 58, 8746-8751.
a) A. I. Gonzalez Florez, E. Mucha, D. S. Ahn, S. Gewinner, W.
Schollkopf, K. Pagel, G. von Helden, Angew. Chem. Int. Ed. 2016,
55, 3295-3299; b) E. Mucha, A. I. Gonzalez Florez, M. Marianski,
D. A. Thomas, W. Hoffmann, W. B. Struwe, H. S. Hahm, S.
Gewinner, W. Schollkopf, P. H. Seeberger, G. von Helden, K.
Pagel, Angew. Chem. Int. Ed. 2017, 56, 11248-11251.
W. G. Schöllkopf, S.; Junkes, H.; Paarmann, A.; von Helden, G.;
Bluem, H. P.; Todd, A. M. M., Proc. SPIE Int. Soc. Opt. Eng. 2015,
95121L.
a) A. Supady, V. Blum, C. Baldauf, J. Chem. Inf. Model. 2015, 55,
2338-2348; b) M. Marianski, A. Supady, T. Ingram, M. Schneider,
C. Baldauf, J. Chem. Theory Comput. 2016, 12, 6157-6168.
a) J. P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 1996, 77,
3865-3868; b) A. Tkatchenko, M. Scheffler, Phys. Rev. Lett. 2009,
102, 073005.
V. Blum, R. Gehrke, F. Hanke, P. Havu, V. Havu, X. Ren, K.
Reuter, M. Scheffler, Comput. Phys. Commun. 2009, 180, 2175-
2196.
a) C. Adamo, V. Barone, J. Chem. Phys. 1999, 110, 6158-6170; b)
S. Grimme, J. Antony, S. Ehrlich, H. Krieg, J. Chem. Phys. 2010,
132, 154104.
a) F. Weigend, M. Häser, H. Patzelt, R. Ahlrichs, Chem. Phys. Lett.
1998, 294, 143-152; b) S. Kossmann, F. Neese, Chem. Phys. Lett.
2009, 481, 240-243; c) F. Neese, E. F. Valeev, J. Chem. Theory
Comput. 2011, 7, 33-43; d) F. Neese, WIREs Comput. Mol. Sci.
2012, 2, 73-78.
In summary, we provide direct evidence for remote participation
in galactose building blocks routinely used in oligosaccharide syn-
theses. α-Selectivity is achieved by a C4 acetyl group, promoting
energetically preferred dioxolenium ions with a covalent bond be-
tween the carbonyl oxygen and the anomeric carbon, as shown
for building blocks 4Ac and 4,6Ac. Such a dioxolenium structure
is energetically less favored for acetyl groups at the C6-position,
leading to the formation of oxocarbenium or oxonium structures
observed for building block 6Ac. The presence of oxonium struc-
tures is particularly surprising, because they involve the remote
participation of benzyl protecting groups that promote the for-
mation of α-glycosides. For the fully benzylated Bn building block,
remote participation via the C6-benzyl group is energetically pre-
ferred and leads to an α-selective oxonium intermediate. Our ob-
servations provide a structural basis for the different modes of re-
mote participation, which are essential during the stereoselective
formation of 1,2-cis-glycosidic bonds. By removing the influence
of solvent and counter ions, the intrinsic stereoselectivity of glyco-
syl cations can be studied as a basis to design glycosyl donors by
tuning the electronic properties of participating groups.
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
Acknowledgements
The authors gratefully acknowledge generous funding by the
Max-Plank-Society and the great expertise of Dr. Wieland
Schöllkopf and Sandy Gewinner for providing numerous coherent
photons produced by the FHI-FEL. DAT acknowledges support
from the Alexander von Humboldt Foundation and SM acknowl-
edges support from the DFG InCheM (FOR 2177). MM acknowl-
edges The Professional Staff Congress and The City University
of New York and for a PSC-CUNY Award. KG is grateful to the
“Fondation Félix Chomé” for the “Bourse Chomé-Bastian” schol-
arship. The present project is supported by the National Research
Fund, Luxembourg.
[20]
[21]
a) D. Cremer, J. A. Pople, J. Am. Chem. Soc. 1975, 97, 1354-
1358; b) A. D. Hill, P. J. Reilly, J. Chem. Inf. Model. 2007, 47,
1031-1035.
S. van der Vorm, T. Hansen, H. S. Overkleeft, G. A. van der Marel,
J. D. C. Codée, Chem. Sci. 2017, 8, 1867-1875.
Keywords: Glycosylation • Spectroscopy • Mass Spectrometry •
Glycosyl Cations • Reaction Intermediates
[1]
[2]
[3]
Glycochemical Synthesis, John Wiley & Sons, Inc., Hoboken, NJ,
USA, 2016.
a) O. J. Plante, E. R. Palmacci, P. H. Seeberger, Science 2001,
291, 1523-1527; b) B. Capon, S. P. McManus, 1976.
a) E. Mucha, M. Marianski, F.-F. Xu, D. A. Thomas, G. Meijer, G.
von Helden, P. H. Seeberger, K. Pagel, Nat. Commun. 2018, 9,
4174; b) H. Elferink, M. E. Severijnen, J. Martens, R. A. Mensink,
G. Berden, J. Oomens, F. P. J. T. Rutjes, A. M. Rijs, T. J. Boltje, J.
Am. Chem. Soc. 2018, 140, 6034-6038; c) D. Crich, Z. Dai, S.
Gastaldi, The Journal of Organic Chemistry 1999, 64, 5224-5229.
a) J.-H. Kim, H. Yang, G.-J. Boons, Angewandte Chemie
International Edition 2005, 44, 947-949; b) M. A. Fascione, W. B.
Turnbull, Beilstein Journal of Organic Chemistry 2010, 6.
[4]
This article is protected by copyright. All rights reserved.

10.1002/anie.201916245
Angewandte Chemie International Edition
Communication
COMMUNICATION
Author(s), Corresponding Author(s)*
Page No. – Page No.
Title
We unravel the structure of glycosyl cations involved in remote participation using
cryogenic ion IR spectroscopy and density-functional theory. We demonstrate that
acetyl groups at C4 promote the formation of α-product by forming a covalent bond
to the anomeric carbon in dioxolenium type ions.
This article is protected by copyright. All rights reserved.