Lanthanide-assisted NMR evaluation of a dynamic ensemble of oligosaccharide conformations

Article abstract of DOI:10.1039/c2cc30353a

A novel methodology is presented for evaluating a dynamic ensemble of oligosaccharide conformations by lanthanide-assisted NMR spectroscopy combined with molecular dynamics (MD) simulations. The results obtained using the GM3 trisaccharide demonstrated th

Full text of DOI:10.1039/c2cc30353a

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2012, 48, 4752–4754
www.rsc.org/chemcomm
COMMUNICATION
Lanthanide-assisted NMR evaluation of a dynamic ensemble of
oligosaccharide conformationsw
ab
abc
Sayoko Yamamoto,z Ying Zhang,z Takumi Yamaguchi,abc Tomoshi Kamedad and
Koichi Kato*abc
Received 16th January 2012, Accepted 19th March 2012
DOI: 10.1039/c2cc30353a
A novel methodology is presented for evaluating a dynamic
ensemble of oligosaccharide conformations by lanthanide-assisted
NMR spectroscopy combined with molecular dynamics (MD)
simulations. The results obtained using the GM3 trisaccharide
demonstrated that pseudocontact shift measurements offer a
valuable experimental tool for the validation of MD simulations
of highly flexible biomolecules.
low-proton density, a low number of independent bond vectors,
and smaller magnitudes of these parameter values in oligo-
saccharides often preclude the achievement of precision and
accuracy in measuring these NMR parameters.
Another potentially useful tool involves the introduction of
paramagnetic lanthanide ions with an anisotropic magnetic
susceptibility tensor (Dw tensor), which induces pseudocontact
shifts (PCSs) as chemical shift changes related to the spatial
position of the observed nuclei with respect to the para-
magnetic center.15–17 Despite the relative ease of their quanti-
tative observation, the application of PCSs to carbohydrate
conformational analyses has been quite limited probably
because of the lack of a conventional method to attach
lanthanide ions to oligosaccharides. Recently, we and other
groups have reported protocols for the lanthanide tagging of
disaccharides to enable their NMR conformational analyses.18–20
Conformational flexibility is an important property of bio-
logical molecules functioning in living systems, as best
exemplified by oligosaccharides, which possess significant
degrees of motional freedom, thereby exhibiting conformational
adaptability upon interacting with various target molecules in
molecular recognition events.1,2 Therefore, atomic descrip-
tions of dynamic oligosaccharide conformations are important
not only for understanding the quantitative energetics of
carbohydrate–protein and carbohydrate–carbohydrate inter-
actions but also for designing drugs targeting these interacting
systems. However, although recent advancements in computa-
tional approaches have enabled large-scale molecular dynamics
(MD) simulations of oligosaccharides in solution with improved
force fields,3–6 the experimental methodology to evaluate their
conformational dynamics has not yet been fully developed.
NMR spectroscopy has immense potential to deal with such
dynamics issues in various ranges of spatial and time
scales.7–10 In conformational analyses of flexible biomolecules,
NMR data should be interpreted as a weighted average of two
or more conformational states. Such analyses have been
attempted in interpretations of NOE and residual dipolar
coupling (RDC) data of oligosaccharides,11–14 although
In particular, Erdelyi et al. demonstrated that the PCS data
´
for lactose could be effectively reproduced by a population-
optimized combination of selected low-energy conformers.19
In view of this situation, we herein attempt to combine the
lanthanide-assisted NMR method with MD simulations for
the evaluation of dynamic conformational ensembles of highly
flexible oligosaccharides that exhibit shallow and broad energy
minima in their conformational space.
We used the trisaccharide of the GM3 ganglioside aN-
eu5Ac-(2–3)-bGal-(1–4)-bGlc as a model system because this
trisaccharide is the common core structure shared among
gangliosides. They are glycosphingolipids having a sialyl
oligosaccharide, which act as targets for various patho-
logically relevant proteins.21–23 There have been several
reports describing NMR and computational analyses of this
trisaccharide conformation.24–28 The key to lanthanide tagging
is to introduce a metal-chelating unit into the reducing end of
the oligosaccharide, which can form a stable complex with a
a Graduate School of Pharmaceutical Sciences, Nagoya City
University, 3-1 Tanabe-dori, Mizuho-ku, Nagoya, Japan
b Institute for Molecular Science and Okazaki Institute for Integrative
Bioscience, 5-1 Higashiyama, Myodaiji, Okazaki, Japan.
E-mail: kkatonmr@ims.ac.jp; Fax: +81-564-59-5224;
Tel: +81-564-59-5225
paramagnetic lanthanide ion.
A phenylenediamine-based
lanthanide-chelating tag was newly designed to improve the
rigidity of the tag, which is crucially important for the accurate
interpretation of PCS data in terms of carbohydrate dynamics.
This novel tag was successfully attached to the trisaccharide,
as shown in Scheme 1. The reducing terminus of the sugar
moiety was selectively aminated in good yield; subsequently,
it was connected to the tag through an amide linkage. By
1H NMR titration experiments, it was confirmed that the
c Department of Functional Molecular Science, The Graduate
University for Advanced Studies, 5-1 Higashiyama, Myodaiji,
Okazaki, Japan
d Computational Biology Research Center, National Institute of
Advanced Industrial Science and Technology, 2-4-7 Aomi, Koto,
Tokyo, Japan
w Electronic supplementary information (ESI) available: Experimental
and simulation details, NMR data. See DOI: 10.1039/c2cc30353a
z These authors contributed equally to this work.
c
4752 Chem. Commun., 2012, 48, 4752–4754
This journal is The Royal Society of Chemistry 2012

View Article Online
Scheme 1 Introduction of the lanthanide-chelating tag into the GM3 trisaccharide.
lanthanide ions were selectively bound to the tag moiety of the
trisaccharide, giving rise to a stable 1 : 1 complex.
Table 1 Values of PCSs (ppm) derived from the Tm3+ ion
Neu5Ac
Gal
Glc
1
The PCSs were measured as the differences between the H
and 13C chemical shifts of the compound chelated to the
paramagnetic ion and those observed with the diamagnetic
ion in their 1H–13C HSQC spectra (Fig. 1 and Table S1, ESIw).
In this comparison, most of the CH groups exhibited PCS
values sufficiently large for a quantitative conformational
analysis, as summarized in Table 1. Significantly larger PCS
values were observed for the H8 of Neu5Ac and its proximal
atoms, suggesting a contribution of some bent conformations
of this trisaccharide, which is consistent with a previously
reported NMR study.26 These experimentally determined PCS
values were compared with those calculated from a conforma-
tional ensemble of the GM3 trisaccharide.
Dd13
Dd H
1
Dd13
Dd H
1
Dd13
C
Dd H
1
C
C
1
2
3
4
5
6
7
8
9
—
—
—
—
ꢁ0.62 ꢁ0.56
ꢁ0.45 ꢁ0.42
ꢁ2.83 ꢁ2.51
ꢁ1.87 ꢁ1.73
ꢁ1.22 ꢁ1.12
ꢁ1.12 ꢁ0.98
ꢁ1.45 ꢁ1.35
ꢁ0.16 ꢁ0.17/ꢁ0.20 ꢁ0.34 ꢁ0.30
ꢁ0.19 ꢁ0.14
ꢁ0.14 ꢁ0.17
ꢁ0.19 ꢁ0.15
ꢁ0.20 ꢁ0.15
ꢁ0.23 ꢁ0.25
ꢁ0.20 ꢁ0.21/ꢁ0.17
ꢁ0.31 ꢁ0.23
ꢁ0.36 ꢁ0.34
ꢁ0.27 ꢁ0.23/ꢁ0.23 ꢁ1.12 ꢁ1.04/ꢁ0.98
—
—
—
—
—
—
—
—
—
—
—
—
extracted at equal intervals to create an ensemble model,
which involved harmonic motions in a local minimum as well
as transitions from one low-energy region to another in the
energy landscape (Fig. 2). The coordinate of the average
position of the paramagnetic center with respect to the inner-
most Glc residue was defined from additional MD calculations
of the tag moiety (see ESIw).
To generate atomic coordinates of the conformationally
fluctuating trisaccharide, 10 MD simulations were performed
in explicit water for 12 ns with the GLYCAM_06 force field.4
Consistent with the previously reported calculation, torsion
angle pairs of this trisaccharide show high flexibility especially
around the glycosidic linkage between the Neu5Ac and Gal
residues (Fig. 2).
A single Dw tensor was determined for the conformational
ensemble by inspection of the experimentally obtained PCSs
with the assumption that every conformer contributes equally
to the PCSs. The anisotropy values of the Dw tensors Dwax and
Dwrh, derived from the Tm3+ ion for the ensemble, were
estimated to be 8.1 ꢀ 10ꢁ23 m3 and 3.5 ꢀ 10ꢁ23 m3, respectively.
The results for the back-calculated PCSs were in excellent
agreement with the experimental data: the Q valuey was 0.05
(Fig. 3(a)). Similarly, a low Q (0.06) was obtained using Tb3+
as a lanthanide probe (Fig. 3(b)). Such low Q values could be
obtained neither by employing most single conformers nor
by using a combination of selected low-energy conformers
(Table S2, ESIw). Furthermore, a conformational ensemble
derived from only one trajectory (12 ns), which does not include
a low-populated conformational cluster in Gal–Glc linkage
(C E 1801), gave a compromised Q value (Fig. S1, ESIw).
These results indicate that minor conformers significantly
contribute to the observed PCS values. The minor ‘‘anti’’
conformations about the Gal–Glc linkage were barely detected
in the MD simulation restrained by rotating frame Overhauser
effect data that were obtained using 13C-enriched GM3
trisaccharide.27 These results demonstrated the utility of our
All MD runs were combined after excluding the first 2 ns of
trajectories. Subsequently, 2000 trisaccharide conformers were
Fig. 1 1H–13C HSQC spectra of the sugar tagged with Tm3+ (magenta)
and La3+ (blue). Chemical shift differences of the anomeric CH
groups are indicated by arrows.
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 4752–4754 4753

View Article Online
an experimental tool for validating MD simulations of not only
oligosaccharides but also other flexible biomacromolecules such
as intrinsically disordered proteins.
We thank Dr Christian Griesinger (Max-Planck Institute
for Biophysical Chemistry) for useful discussion. This work
was supported by KAKENHI (20107004 and 21370050) of
MEXT and CREST of JST. A part of computational calculation
was performed at Research Center for Computational Science,
Okazaki, Japan.
Notes and references
y Q = rms(Ddcalc ꢁ Ddobs)/rms(Ddobs). Ddcalc is given by following equation;
P
N
i=1
Ddcalc
=
(pi1/12pr3i [Dwax(3cos2Wiꢁ1) + 3/2Dwrh(sin2Wi cos2ji)]),
where pi is the population of each structure (set to 0.0005), N is the
number of each conformers, and (ri, yi, ji) defines the position vector
for conformer i of the nucleus in polar coordinates with respect to the
metal center and principal axis of a Dw tensor.
1 M. R. Wormald, A. J. Petrescu, Y. L. Pao, A. Glithero, T. Elliott
and R. A. Dwek, Chem. Rev., 2002, 102, 371–386.
2 Y. Kamiya, M. Yagi-Utsumi, H. Yagi and K. Kato, Curr. Pharm.
Des., 2011, 17, 1672–1684.
3 E. Fadda and R. J. Woods, Drug Discovery Today, 2010, 15,
596–609.
´
4 K. N. Kirschner, A. B. Yongye, S. M. Tschampel, J. Gonzalez-
Outeirino, C. R. Daniels, B. L. Foley and R. J. Woods, J. Comput.
Chem., 2008, 29, 622–655.
5 R. D. Lins and P. H. Hunenberger, J. Comput. Chem., 2005, 26,
1400–1412.
¨
6 D. Kony, W. Damm, S. Stoll and W. Van Gunsteren, J. Comput.
Chem., 2002, 23, 1416–1429.
7 G. Lipari and A. Szabo, J. Am. Chem. Soc., 1982, 104, 4546–4559.
8 D. D. Boehr, D. McElheny, H. J. Dyson and P. E. Wright, Science,
2006, 313, 1638.
9 O. F. Lange, N. A. Lakomek, C. Fares, G. F. Schroder,
¨
K. F. Walter, S. Becker, J. Meiler, H. Grubmuller, C. Griesinger
¨
and B. L. de Groot, Science, 2008, 320, 1471–1475.
10 C. Tang, C. D. Schwieters and G. M. Clore, Nature, 2007, 449,
1078–1082.
Fig. 2 Conformational ensemble of the GM3 trisaccharide. (a) Snap-
shots of the GM3 trisaccharide from a simulated trajectory super-
imposed on the ring atoms of the Gal residue. All hydrogen atoms are
omitted. Torsion angle density maps of MD trajectories of (b) the
Neu5Ac–Gal linkage and (c) the Gal–Glc linkage. Scattered plots of
torsion angles of (d) the Neu5Ac–Gal linkage and (e) the Gal–Glc
linkage of the ensemble for PCS analysis. The NMR definitions of F
and C were used, namely, for the Neu5Ac–Gal linkage, F =
C1–C2–O03–C03 and C = C2–O03–C03–H03 and for the Gal–Glc
linkage, F = H1–C1–O04–C04 and C = C1–O04–C04–H04.
11 J. Landstrom and G. Widmalm, Carbohydr. Res., 2010, 345,
330–333.
¨
12 S. Ganguly, J. Xia, C. Margulis, L. Stanwyck and C. A. Bush,
Biopolymers, 2011, 95, 39–50.
13 J. Xia, C. J. Margulis and D. A. Case, J. Am. Chem. Soc., 2011,
133, 15252–15255.
14 J. P. M. Lommerse, J. J. M. van Rooijen, L. M. J. Kroon-Batenburg,
J. P. Kamerling and J. F. G. Vliegenthart, Carbohydr. Res., 2002, 337,
2279–2299.
15 I. Bertini, C. Luchinat and G. Parigi, Prog. Nucl. Magn. Reson.
Spectrosc., 2002, 40, 249–273.
16 P. H. Keizers and M. Ubbink, Prog. Nucl. Magn. Reson. Spectrosc.,
2011, 58, 88–96.
17 G. Otting, Annu. Rev. Biophys., 2010, 39, 387–405.
18 S. Yamamoto, T. Yamaguchi, M. Erde
K. Kato, Chem.–Eur. J., 2011, 17, 9280–9282.
19 M. Erdelyi, E. d’Auvergne, A. Navarro-Vazquez, A. Leonov and
C. Griesinger, Chem.–Eur. J., 2011, 17, 9368–9376.
20 A. Mallagaray, A. Canales, G. Domınguez, J. Jime
and J. Perez-Castells, Chem. Commun., 2011, 47, 7179–7181.
´
lyi, C. Griesinger and
´
´
Fig. 3 Correlations between the experimentally observed PCS values
with (a) Tm3+ and (b) Tb3+ and back-calculated PCS values.
´
´
nez-Barbero
´
21 T. Ariga, M. P. McDonald and K. Y. Robert, J. Lipid Res., 2008,
49, 1157–1175.
22 J. Fantini and N. Yahi, Expert Rev. Mol. Med., 2010, 12, e27.
23 E. Posse de Chaves and S. Sipione, FEBS Lett., 2010, 584,
1748–1759.
24 H. C. Siebert, G. Reuter, R. Schauer, C. W. von der Lieth and
J. Dabrowski, Biochemistry, 1992, 31, 6962–6971.
25 Y. Aubin, Y. Ito, J. C. Paulson and J. H. Prestegard, Biochemistry,
1993, 32, 13405–13413.
26 G. R. Kiddle and S. W. Homans, FEBS Lett., 1998, 436, 128–130.
27 M. J. Milton, R. Harris, M. A. Probert, R. A. Field and
S. W. Homans, Glycobiology, 1998, 8, 147–153.
lanthanide-assisted NMR method in conjunction with MD
simulations in the evaluation of dynamic conformational
ensembles of highly flexible oligosaccharides, considering their
minor conformers in a systematic manner.
In summary, we successfully acquired PCSs derived from
the GM3 trisaccharide by the lanthanide-tagging method and
interpreted the PCS data by inspecting a vast conformational
ensemble of this flexible trisaccharide generated from MD
simulations. The lanthanide-assisted NMR approach serves as
28 M. L. DeMarco and R. J. Woods, Glycobiology, 2009, 19, 344–355.
c
4754 Chem. Commun., 2012, 48, 4752–4754
This journal is The Royal Society of Chemistry 2012