Systematic Structural Characterization of Chitooligosaccharides Enabled by Automated Glycan Assembly

Article abstract of DOI:10.1002/chem.202005228

Chitin, a polymer composed of β(1–4)-linked N-acetyl-glucosamine monomers, and its partially deacetylated analogue chitosan, are abundant biopolymers with outstanding mechanical as well as elastic properties. Their degradation products, chitooligosacchari

Full text of DOI:10.1002/chem.202005228

Accepted Article
Accepted Article
Title: Systematic structural characterization of chitooligosaccharides
enabled by Automated Glycan Assembly
Authors: Theodore Tyrikos-Ergas, Vittorio Bordoni, Giulio Fittolani,
Manishkumar Chaube, Andrea Grafmüller, Peter H.
Seeberger, and Martina Delbianco
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.202005228
Link to VoR: https://doi.org/10.1002/chem.202005228
01/2020

10.1002/chem.202005228
Chemistry - A European Journal
COMMUNICATION
Systematic structural characterization of chitooligosaccharides
enabled by Automated Glycan Assembly
Theodore Tyrikos-Ergas,[a][b] Vittorio Bordoni,[a] Giulio Fittolani,[a][b] Manishkumar Chaube,[a] Andrea
Grafmüller,[c] Peter H. Seeberger[a][b] and Martina Delbianco*[a]
[a]
T. Tyrikos-Ergas, Dr. V. Bordoni, G. Fittolani, Dr. M. Chaube, Prof. Dr. P. H. Seeberger, Dr. M. Delbianco
Department of Biomolecular Systems
Max-Planck-Institute of Colloids and Interfaces
Am Mühlenberg 1, 14476 Potsdam, Germany
E-mail: martina.delbianco@mpikg.mpg.de
T. Tyrikos-Ergas, G. Fittolani, Prof. Dr. P. H. Seeberger
Department of Chemistry and Biochemistry
Freie Universität Berlin
[b]
[c]
Arnimallee 22, 14195 Berlin, Germany
Dr. A. Grafmüller
Department of Theory
Max-Planck-Institute of Colloids and Interfaces
Am Mühlenberg 1, 14476 Potsdam, Germany
Abstract: Chitin, a polymer composed of β(1-4)-linked N-acetyl-
glucosamine monomers, and its partially deacetylated analogue
chitosan, are abundant biopolymers with outstanding mechanical as
well as elastic properties. Their degradation products,
chitooligosaccharides (COS), can trigger the innate immune response
in humans and plants. Both material and biological properties are
dependent on polymer length, acetylation, as well as the pH. Without
well-defined samples, a complete molecular description of these
factors is still missing. Automated Glycan Assembly (AGA) enabled
rapid access to synthetic well-defined COS. Chitin-cellulose hybrid
oligomers were prepared as important tools for a systematic structural
analysis. Intramolecular interactions, identified by molecular
dynamics simulations and NMR analysis, underscore the importance
A detailed molecular description of chitin, chitosan, and COS
structure-function relation is missing, as most studies are performed
with ill-defined samples. Computationally, several all-atom models
have been applied to study the conformational space of COS,
showing that DP, FA, AP as well as pH strongly affect the
conformation and control aggregation.[10] Coarse grained (CG)
computational methods provide further insights on the COS
interactions in solution, aiming for a description of chitin and chitosan
polymers.[11] However, due to the intrinsic CG approximation,
chemical details are lost. In computational methods, the lack of
standards to validate the theoretical models remains the major
bottleneck, leading, in some cases, to contradictory theories.[11b] Well-
defined samples with controlled DP, FA, and AP are therefore
of the chitosan amino group for the stabilization of specific geometries. important targets to shine light on molecular conformations and
interaction mechanisms.
Polysaccharides are valuable biocompatible and recyclable materials.
Chitin, a polymer composed of N-acetylglucosamine repeating units,
is the second most abundant polysaccharide in Nature, after cellulose.
Chitin serves mainly structural roles in the exoskeleton of crustaceans,
insects, and fungal cell wall.[1] Its (partially) deacetylated counterpart
– chitosan – is easily obtained via hydrolytic deacetylation of chitin
and, due to its higher water solubility and easy functionalization, is
used for industrial applications such as coating material, ingredient in
cosmetics, and pharmaceutical excipient.[2] Chitin and chitosan are
commonly used to produce fibers, particles, and composites with
exceptional biological and mechanical properties.[3] Degree of
polymerization (DP) and fraction of acetylation (FA) offer the
opportunity to tune the stiffness, solubility, and transparency of the
resulting materials.[4]
Chitin degradation produces chitooligosaccharides (COS). These
short oligomers are known to trigger an innate immune response in
humans[5] and antifungal defense mechanisms in plants[6]. The DP of
COS is crucial for the biological response, as size-dependent
recognition was observed in plant chitin receptors as well as in toll-
like receptors (TLR2).[7] It has been suggested that the acetylation
pattern (AP) of COS modulates the biological activity[8] and may
Figure 1. AGA of COS and cellulose-chitin hybrids. Isolated yields after AGA
explain the existence of sequence-specific chitosan hydrolases in
most organisms.[9]
and post-AGA (deprotection and purifications) are reported. *Yield obtained
when AGA was performed using BB1a.
1
This article is protected by copyright. All rights reserved.

10.1002/chem.202005228
Chemistry - A European Journal
COMMUNICATION
COS commonly obtained by partial degradation of polymeric chitin
and chitosan require extensive purification and typically exist as
mixtures.[6a] Chemical or enzymatic N-(de)acetylation are common
manipulations, but in most cases yield ill-defined products with varying
DP, FA, and AP.[12] Sequence-specificity or regio-selectivity may be
achieved enzymatically,[13] but only few of the required enzymes are
available. To date, no general method to produce all possible patterns
exists.[8] Alternatively, well-defined but simple COS can be prepared
by chemical synthesis using orthogonal protecting groups[14] and
glycosylation conditions.[15] Conventional synthetic approaches are
too laborious to prepare a large collection of well-defined COS
required for systematic structural studies. Solid phase based
automated techniques offer the ideal solution for the quick production
of large series of related compounds.[16] Still, their scope is limited by
problems with sequences that form rigid tertiary structures, such as
cellulose and chitin oligomers.[16b]
glycosylation is ensured by C(2) anchimeric assistance by the
protecting group.[18] Two glucosamine BBs are designed for the
introduction of either N-acetyl (N) or free (K) glucosamine units. The
amino group in BB1 is equipped with the trichloroacetyl (TCA) group,
whereas BB2 bears a carboxybenzyl (Cbz) group. The glucose unit
(A), required for the chitin-cellulose oligomers, is installed using BB3.
The desired sequence is assembled on solid support, following
iterative cycles of glycosylation and deprotection (see SI). The
protected oligomer is released from the solid support upon cleavage
of the UV-labile linker 4.[19] Hydrogenolysis removes all the benzyl
ether (Bn) protecting groups and allows for the concomitant TCA
reduction and Cbz cleavage, affording the desired COS with defined
AP. The hybrid structures require basic hydrolysis of the benzoate
esters (Bz), prior to hydrogenolysis.
AGA using thioglycoside BB1a, suffered from low yields and
significant amounts of deletion sequences, particularly when multiple
NN linkages had to be incorporated. This could be ascribed to the
poor reactivity of the glucosamine acceptor[20] and the potential
Here, we report the Automated Glycan Assembly (AGA) of a collection
of hexasaccharides, including well-defined COS as well as hybrid
chitin-cellulose oligomers. These unnatural analogues are designed
to explore the importance of the amino groups in COS.
Monosaccharide building blocks (BBs) are iteratively combined on a
solid support, granting full-control over length and acetylation
pattern.[17] The conformational space of the synthetic oligomers and
recurrent intramolecular interactions are studied systematically using
molecular dynamics (MD) simulations and NMR experiments.
Nine hexasaccharides were assembled by AGA employing four
differentially protected thioglycoside or glycosyl phosphate
monosaccharide BBs (Figure 1). A fluorenylmethoxycarbonyl (Fmoc)
temporary protecting group masks the C(4) hydroxyl group that, upon
cleavage, allows for chain elongation. Stereochemical control during
formation of side-products during activation of the thioglycoside[21]
.
Glycosyl phosphate BB1b performed significantly better as indicated
by an increase in yield of N6 from 8% to 34%. Therefore, all COS were
synthesized using glycosyl phosphate building blocks. Nine oligomers
including the chitin oligomer N6, four COS with different acetylation
degree and patterns, and four hybrid chitin-cellulose analogues were
assembled and were found to be highly water soluble (Figure 1).
The synthetic oligomers were modelled using MD simulations,
employing a modified version of the GLYCAM06 carbohydrate force
field.[22] The partially deacetylated COS and the hybrid cellulose-chitin
oligomers were compared to the reference chitin oligomer N6. Amino
substituted structures were simulated with neutral NH2 (K) as well as
b)
a)
0.35 nm
0.19 nm
Ψ < 0
H-bond
Ψ > 0
No H-bond
NNNNNN-OH
Ψ > 0
NK+NNK+N-OH
NANNAN-OH
NKNNKN-OH
c)
d)
Ψ < 0
Ψ > 0
Ψ < 0
Figure 2. Definition of the glycosidic bond torsion angles (Ψ and Φ) (a) and representative snapshots indicating the presence (Ψ < 0) or absence (Ψ > 0) of
the conventional O(5)∙∙∙OH(3) H-bond (b). Analysis of the Ramachandran plots obtained by MD simulations for a single molecule (c) and for a concentrated
system with 25 molecules (d). Negative degrees of Ψ (red circles) indicate the presence of the conventional O(5)∙∙∙OH(3) H-bond, whereas the increased
distance between these two residues is reflected by positive Ψ (blue circles). Deviation from the main conformations are highlighted with arrows.
2
This article is protected by copyright. All rights reserved.

10.1002/chem.202005228
Chemistry - A European Journal
COMMUNICATION
a)
b)
c)
NKNNKN-OH
NK+NNK+N-OH
65 % gg | 32 % gt | 3% tg
63 % gg | 27 % gt | 10 % tg
d)
pH 8.5
pH 4.0
3JH5H6R
3JH5H6S
5.1 Hz
2.3 Hz
4.9 Hz
2.6 Hz
gg
gt
tg
59 %
41 %
0 %
59 %
38 %
4 %
Figure 3. Definition of the ω torsion angle (a) and ω distribution obtained by MD simulations for the NKNNKN analogue with different protonation (b).
Representative snapshots indicating the presence of an intramolecular H-bond between the NH3+(2) and the OH(6) of the following residue (R+1) (c). NMR
analysis of the 3JH5H6 coupling constants measured for the model dimer 5 at a different pH (d).
with protonated NH3 (K+), as representative models of COS at
different pH. All the modified analogues result in more flexible
structures when compared to N6. Ramachandran plots are used to
compare changes on the glycosidic bond torsion angles (Ψ, Φ)
(Figure 2). No significant differences are observed for Φ, stabilized by
the exo-anomeric effect.[23] The high population of Ψ at negative
degrees (Figure 2c, red circle) is related to the presence of the
conventional O(5)∙∙∙OH(3) hydrogen bond (Figure 2b), which rigidifies
the chitin structure (N6).[16b] All modified oligomers show an increased
population at positive Ψ (Figure 2c, blue circles), albeit with different
intensity. Major disruption of the conventional H-bond is observed for
the charged COS (e.g. (NK+N)2).
of A3N3 does not resemble the XRD profiles of the natural analogs of
cellulose (A6) or chitin (N6).
+
A closer look at the atomistic model shows a significant percentage of
tg rotamers (orientation of the C6 side chain) for the charged COS
(Figure 3a-b).[10a] MD suggests the formation of an intramolecular
+
hydrogen bond between the NH3 (2) and the OH(6) of the following
residue (R+1) (Figure 3c). This happens regardless of the acetylation
of the R+1 unit or the position of the two residues (K+ and R+1) in the
entire molecule. MD confirmed that this interaction is present also in
the KN disaccharide (Figure S12). Therefore, to reduce the
complexity of the hexamers that suffer from chemical shift degeneracy,
NMR measurements were performed on the model dimer 5. Selective
1D HOHAHA-NMR experiments[24] were performed to simplify the
Long MD atomistic simulations (1 μs production run) of concentrated
experiments (25 molecules) were performed, to resemble a crowded
environment. A remarkable increase in the population at negative Ψ
is detected for N6 and for all the modified uncharged structures (Figure
2d, red circles). Stacking reduces the conformational freedom,
favoring the formation of the O(5)∙∙∙OH(3) hydrogen bond as well as
inter-molecular H-bonds (Figure S13). Deviations from the main
population are observed for the modified compounds (Figure 2d,
arrows), suggesting a higher conformational freedom and a less
regular packing than N6. The low aggregation tendency of the ionic
COS (NK+N)2 is confirmed by the similarity of the Ramachandran plots
obtained for the single molecule and the concentrated experiments.
In agreement with the computational model, powder XRD analysis
shows sharp peaks for N6, indicating the tendency to pack with a
regular architecture (Figure S6). All the modified compounds have
amorphous XRD profiles, except from A3N3 that shows a sharp peak
at 21.3°, suggesting that the di-block hybrid maintain the ability to
pack in an ordered fashion (Figure S6). Interestingly, the XRD profile
3
spectra even further (Figure S2-5). JH5H6 coupling constants were
measured at different pH, as these values can be converted to
rotamer percentage using empirical equations (Figure 3d and Table
S1).[25] A small percentage of tg rotamers is detected at pH 4, when
the amine is fully protonated. No tg rotamers are detected at basic
pH (amine not protonated), in agreement with the predictions.
Although NMR data indicate the existence of a small tg population at
acidic pH, the calculated fraction of tg is significantly lower than the
predicted value (Figure S12). Different simulation conditions using i)
a different water model (tip3p), ii) N3 angle parameters derived in the
context of GAG molecules,[26] iii) increased ionic strength, and iv)
reduced Lennard-Jones interactions of the nitrogen atoms (consistent
with the changes introduced in the GLYCAMOSMO,r14 force field) did not
lead to major changes in the simulation results (Figure S12). This
overrepresentation of the observed H-bonds trend demonstrates the
need of further optimization of the dihedral potentials, especially in the
presence of ionic moieties (e.g. amines).
3
This article is protected by copyright. All rights reserved.

10.1002/chem.202005228
Chemistry - A European Journal
COMMUNICATION
In conclusion, AGA is a powerful tool to produce oligosaccharides that
are essentials for systematic structural analysis. Five COS were
assembled with full-control over length as well as acetylation degree
and pattern. Four unnatural hybrid chitin-cellulose oligomers were
prepared to study the importance of the amino group in chitosan.
Single molecule as well as concentrated MD simulations showed that
all COS analogues have more conformational freedom than the fully
N-acetylated hexamer N6, resulting in amorphous aggregation upon
drying. The hybrid compounds showed a similar conformational
behavior as the neutral partially acetylated COS. Amine protonation
results in intramolecular interactions, detected by NMR, that stabilize
new geometries. This finding stresses the importance of the pattern
of de-acetylation of COS, because these interactions exist only in the
presence of a deacetylated GlcN unit (K+). Therefore, knowing the
position of the deacetylated residue in the polymer is essential for the
description of the COS conformation. This observation is particularly
relevant to clarify molecular mechanisms of chitosan-protein
interactions, as glycoside hydrolases binding is affected by the
orientation of the C6 side chain.[27]
[8]
S. Basa, M. Nampally, T. Honorato, S. N. Das, A. R. Podile, N.
E. El Gueddari, B. M. Moerschbacher, J. Am. Chem. Soc. 2020,
142, 1975-1986.
L. Grifoll-Romero, S. Pascual, H. Aragunde, X. Biarnés, A.
Planas, Polymers 2018, 10, 352.
a) S. Skovstrup, S. G. Hansen, T. Skrydstrup, B. Schiøtt,
Biomacromolecules 2010, 11, 3196-3207; b) E. F. Franca, R. D.
Lins, L. C. G. Freitas, T. P. Straatsma, J. Chem. Theory Comput.
2008, 4, 2141-2149; c) A. Almond, J. K. Sheehan, Glycobiology
2003, 13, 255-264; d) M. T. McDonnell, D. A. Greeley, K. M.
Kit, D. J. Keffer, J. Phys. Chem. B 2016, 120, 8997-9010.
a) Z. Yu, D. Lau, Journal of Molecular Modeling 2015, 21, 128;
b) L. Tsereteli, A. Grafmüller, PLoS One 2017, 12, e0180938; c)
J. D. Schneible, A. Singhal, R. L. Lilova, C. K. Hall, A.
Grafmüller, S. Menegatti, Biomacromolecules 2019, 20, 3126-
3141.
[9]
[10]
[11]
[12]
[13]
M. X. Weinhold, J. C. M. Sauvageau, J. Kumirska, J. Thöming,
Carbohydr. Polym. 2009, 78, 678-684.
a) L. Hembach, S. Cord-Landwehr, B. M. Moerschbacher, Sci.
Rep. 2017, 7, 17692; b) S. N. Hamer, S. Cord-Landwehr, X.
Biarnés, A. Planas, H. Waegeman, B. M. Moerschbacher, S.
Kolkenbrock, Sci. Rep. 2015, 5, 8716.
[14]
R. Enugala, L. C. R. Carvalho, M. J. Dias Pires, M. M. B.
Marques, Chem. – Asian J. 2012, 7, 2482-2501.
[15]
[16]
Y. Yang, B. Yu, Tetrahedron 2014, 70, 1023-1046.
a) M. Guberman, P. H. Seeberger, J. Am. Chem. Soc. 2019, 141,
5581-5592; b) M. Delbianco, A. Kononov, A. Poveda, Y. Yu, T.
Diercks, J. Jiménez-Barbero, P. H. Seeberger, J. Am. Chem. Soc.
2018, 140, 5421-5426.
a) Y. Yu, T. Tyrikos-Ergas, Y. Zhu, G. Fittolani, V. Bordoni, A.
Singhal, R. J. Fair, A. Grafmüller, P. H. Seeberger, M. Delbianco,
Angew. Chem., Int. Ed. 2019, 58, 1433-7851; b) Y. Zhu, T.
Tyrikos-Ergas, K. Schiefelbein, A. Grafmüller, P. H. Seeberger,
M. Delbianco, Org. Biomol. Chem. 2020, 18, 1349-1353; c) M.
Panza, S. G. Pistorio, K. J. Stine, A. V. Demchenko, Chem. Rev.
2018, 118, 8105-8150.
Acknowledgements
We thank the Max-Planck Society, the Minerva Fast Track
Program, and the MPG-FhG Cooperation Project Glyco3Dysplay,
for generous financial support.
[17]
Keywords: carbohydrates • chitin • chitosan • MD simulations •
structure-property correlations
[18]
[19]
D. Crich, Acc. Chem. Res. 2010, 43, 1144-1153.
[1]
[2]
[3]
K. Kurita, Mar Biotechnol (NY) 2006, 8, 203-226.
M. J. P. i. p. s. Rinaudo, 2006, 31, 603-632.
a) S. Gim, Y. Zhu, P. H. Seeberger, M. Delbianco, Wiley
K. Le Mai Hoang, A. Pardo-Vargas, Y. Zhu, Y. Yu, M. Loria, M.
Delbianco, P. H. Seeberger, J. Am. Chem. Soc. 2019, 141, 9079-
9086.
D. Crich, V. Dudkin, J. Am. Chem. Soc. 2001, 123, 6819-6825.
C.-W. Chang, M.-H. Lin, C.-H. Wu, T.-Y. Chiang, C.-C. Wang,
J. Org. Chem. 2020.
a) R. J. Woods, Chem. Rev. 2018, 118, 8005-8024; b) X. Xiong,
Z. Chen, B. P. Cossins, Z. Xu, Q. Shao, K. Ding, W. Zhu, J. Shi,
Carbohydr. Res. 2015, 401, 73-81; c) J. r. Sauter, A. Grafmꢀller,
J. Chem. Theory Comput. 2015, 11, 1765-1774; d) M. Marianski,
A. Supady, T. Ingram, M. Schneider, C. Baldauf, J. Chem. Theory
Comput. 2016, 12, 6157-6168.
Interdisciplinary
Reviews:
Nanomedicine
and
[20]
[21]
Nanobiotechnology 2019, 0, e1558; b) H. Merzendorfer, E.
Cohen, Extracellular Sugar-Based Biopolymers Matrices 2019,
12, 541-624.
J. Wattjes, S. Sreekumar, C. Richter, S. Cord-Landwehr, R. Singh,
N. E. El Gueddari, B. M. Moerschbacher, Reactive and
Functional Polymers 2020, 151, 104583.
S. J. Van Dyken, R. M. Locksley, J Allergy Clin Immunol 2018,
142, 364-369.
a) S. Naqvi, B. M. Moerschbacher, Crit Rev Biotechnol 2017, 37,
11-25; b) L. P. Erwig, N. A. R. Gow, Nat. Rev. Microbiol. 2016,
14, 163-176; c) K. Li, R. Xing, S. Liu, P. Li, J. Agric. Food Chem.
2020.
K. Fuchs, Y. Cardona Gloria, O.-O. Wolz, F. Herster, L. Sharma,
C. A. Dillen, C. Täumer, S. Dickhöfer, Z. Bittner, T.-M. Dang, A.
Singh, D. Haischer, M. A. Schlöffel, K. J. Koymans, T.
Sanmuganantham, M. Krach, T. Roger, D. Le Roy, N. A.
Schilling, F. Frauhammer, L. S. Miller, T. Nürnberger, S.
LeibundGut-Landmann, A. A. Gust, B. Macek, M. Frank, C.
Gouttefangeas, C. S. Dela Cruz, D. Hartl, A. N. Weber, EMBO
Rep. 2018, 19, e46065.
[22]
[4]
[5]
[6]
[23]
H. Booth, K. A. Khedhair, J. Chem. Soc., Chem. Commun. 1985,
467-468.
[24]
[25]
D. G. Davis, A. Bax, J. Am. Chem. Soc. 1985, 107, 7197-7198.
a) H. Amarasekara, S. Dharuman, T. Kato, D. Crich, J. Org.
Chem. 2018, 83, 881-897; b) M. G. Pirrone, T. Matsushita, A.
Vasella, D. Crich, Carbohydr. Res. 2020, 491, 107984.
A. Singh, M. B. Tessier, K. Pederson, X. Wang, A. P. Venot, G.-
J. Boons, J. H. Prestegard, R. J. Woods, Can. J. Chem. 2016, 94,
927-935.
[7]
[26]
[27]
J. C. K. Quirke, D. Crich, J. Am. Chem. Soc. 2020, 142, 16965-
16973.
4
This article is protected by copyright. All rights reserved.

10.1002/chem.202005228
Chemistry - A European Journal
COMMUNICATION
Entry for the Table of Contents
Automated Glycan Assembly was employed to prepare a collection of chitooligosaccharides with full control over length, fraction and
pattern of acetylation. Systematic structural studies, based on molecular dynamics simulations and NMR experiments, revealed
intramolecular interactions responsible for the stabilization of particular conformations.
5
This article is protected by copyright. All rights reserved.