Phenyl glycosides – Solid-state NMR, X-ray diffraction and conformational analysis using genetic algorithm

Article abstract of DOI:10.1016/j.chemphys.2018.12.001

The X-ray structures of 2,6-dimethylphenyl and phenyl 2,3,4,6-tetra-O-acetyl β-glucosides (1 and 3) and phenyl α-mannoside (6) were obtained. The independent part of the unit cell of the glycosides 1 and 6 was formed by one molecule, and for the glucoside 3, two molecules in the crystal cell were observed. In deacetylated glycosides 4 and 6 the crystal structure was established by a hydrogen bond network formed between the sugar hydroxyls and solvent molecules. The ¹³C CPMAS NMR spectra of aryl glycosides 1–6 were analysed. In the spectrum of 3, doubling of the C4 aryl signal was observed which confirmed the presence of two independent molecules in the solid sample. The GAAGS (Genetic Algorithm-Assisted Grid Search) method was used to determine the low-energy conformers of α-mannosides and β-glucosides. The orientation of the aryl pendant group was calculated using Molecular Mechanics (MMFF94) as well as Quantum Mechanics theory (DFT, B3LYP/6-31 + G(d,p)).

Full text of DOI:10.1016/j.chemphys.2018.12.001

Accepted Manuscript
Phenyl glycosides - Solid–State NMR, X-ray diffraction and conformational
analysis using genetic algorithm
Piotr Wałejko, Jarosław Bukowicki, Łukasz Dobrzycki, Katarzyna Paradowska
PII:
S0301-0104(18)30180-0
DOI:
https://doi.org/10.1016/j.chemphys.2018.12.001
CHEMPH 10254
Reference:
Chemical Physics
To appear in:
Received Date:
Revised Date:
Accepted Date:
26 February 2018
7 November 2018
2 December 2018
Please cite this article as: P. Wałejko, J. Bukowicki, L. Dobrzycki, K. Paradowska, Phenyl glycosides - Solid–State
NMR, X-ray diffraction and conformational analysis using genetic algorithm, Chemical Physics (2018), doi: https://
doi.org/10.1016/j.chemphys.2018.12.001
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

“Phenyl glycosides - Solid–State NMR, X-ray diffraction and conformational analysis
using genetic algorithm”
Piotr Wałejkoa*, Jarosław Bukowickib, Łukasz Dobrzyckic, Katarzyna Paradowskab
a/ Institute of Chemistry, University of Białystok, Ciołkowskiego 1K, 15-245 Białystok, Poland
b/ Medical University of Warsaw, Faculty of Pharmacy, Department of Physical Chemistry,
Banacha 1, 02-097 Warsaw, Poland
c/
University of Warsaw, Faculty of Chemistry, Czochralski Laboratory of Advanced Crystal
Engineering, Zwirki I Wigury 101, 02-089 Warsaw, Poland
*Corresponding author: Piotr Wałejko; University of Białystok, Institute of Chemistry,
Ciołkowskiego 1K, 15-245 Białystok, Poland; tel: +48 85 7388086; pwalejko@uwb.edu.pl,
Graphical abstract:
Highlights
- The structure of β-glucosides and -mannosides was studied by 13C CPMAS NMR.
- New X-ray structures of acetylated phenyl and 2,6-dimethylphenyl β-glucosides as well as
of phenyl -mannoside were described.
- The grid search method combined with a genetic algorithm was used to determine the low-
energy conformers of the analysed α- and β-glucosides.
- The influence of aryl aglycone rotation on the structure of glycosides was investigated.
- Solid-state NMR supported by calculation of the chemical shifts was used as a verification
method.
Keywords:
Phenyl glycosides, 13C CPMAS NMR, X-ray diffraction, genetic algorithm, grid search.
Abstract:
1

The X-ray structures of 2,6-dimethylphenyl and phenyl 2,3,4,6-tetra-O-acetyl β-glucosides
(1 and 3) and phenyl -mannoside (6) were obtained. The independent part of the unit cell
of the glycosides 1 and 6 was formed by one molecule, and for the glucoside 3, two
molecules in the crystal cell were observed. In deacetylated glycosides 4 and 6 the crystal
structure was established by a hydrogen bond network formed between the sugar hydroxyls
and solvent molecules. The 13C CPMAS NMR spectra of aryl glycosides 1-6 were analysed.
In the spectrum of 3, doubling of the C4 aryl signal was observed which confirmed the
presence of two independent molecules in the solid sample. The GAAGS (Genetic
Algorithm-Assisted Grid Search) method was used to determine the low-energy conformers
of α-mannosides and β-glucosides. The orientation of the aryl pendant group was calculated
using Molecular Mechanics (MMFF94) as well as Quantum Mechanics theory (DFT,
B3LYP/6-31+ G(d,p)).
1. Introduction
Phenyl glycosides are widely found in nature as secondary metabolites. [1,2] Many of them
are of pharmaceutical interest, e.g. flavonoids, tannins, lignans, etc. In general, the connection
of an aglycone to a sugar moiety confers the molecule higher polarity and amphiphilic
properties improving their pharmacokinetic properties and bioavailability in physiological
fluids. Glycosylation makes easier transport through physiological barriers, e.g. the blood-
brain or placental barrier. There are many literature reports indicating that the transformation
of biologically active compounds bearing phenolic functionality into the respective O-
glycosides increases their bioavailability. [3–6] One of the interesting examples of such a
transformation is the conversion of lipophilic -tocopherol (vitamin E) into amphiphilic O-
glycosidic derivatives. [7,8] Vitamin E glycoconjugates have been studied in the aspect of
their acidic and enzymatic hydrolysis in animal tissues. [9,10]
As a continuation of our earlier investigations concerning conformational analysis of aryl O-
galactosides [11], the following series of phenyl O-glycosides were obtained: 2,6-
dimethylphenyl 2,3,4,6-tetra-O-acetyl β-glucoside (1), 2,6-dimethylphenyl β-glucoside (2),
phenyl 2,3,4,6-tetra-O-acetyl β-glucoside (3), phenyl β-glucoside (4), phenyl 2,3,4,6-tetra-O-
acetyl -mannoside (5) and phenyl α-mannoside (6) (Fig. 1). One of the most important
aspects of the conformational analysis of aryl glycosides is the mutual orientation of the
pendant aryl group and the sugar part. The reported β-glucosides 1 – 4 and -mannosides 5
2

and 6 were taken as model compounds in conformational analysis of the aryl O-glycosides.
The main aim of this study was to investigate the rotation of a pendant aryl group in various
glycosides and its orientation in relation to the aglycone bulkiness and sugar moiety as well as
glycosidic bond geometry.
Fig. 1. Chemical structure of O-glycosides 1-6.
OAc
OH
O5'
OAc
2"
6
4'
6'
O
O
O
2
AcO
AcO
O1
HO
HO
AcO
AcO

3
5'
2'
1'
1
O
O
O
4
OAc
3'
OH
OAc

5
6"
1
2
3
 = O5'-C1'-O1-C1
 = C1'-O1-C1-C2
OH
OAc
AcO
OH
HO
O
O
O
HO
HO
AcO
AcO
HO
HO
O
OH
O
O
4
6
5
NMR spectroscopy in the solid phase (CPMAS) has become a powerful technique for
structural studies of all kinds of solids, including single crystals, crystalline powders,
amorphous powders and glasses [12–15]. Cross-polarization (CP) magic angle spinning
(MAS) NMR (13C CPMAS NMR) can provide interesting and quick information on the sugar
part structure in the solid sample. It may be used as a screening tool before undertaking more
complex X-ray diffraction for a single-crystal structural study.
13C NMR spectra in solid state and CDCl3 solution were recorded for all studied compounds.
The  parameter defined as  = solution - solid, was calculated for all carbons in the phenyl
ring and in the sugar skeleton. The differences in the  value resulted from varied shielding of
carbons in the conformationally flexible parts of molecules in a solution state and totally or
partially stiffened in the solid state. These differences provide information on conformational
flexibility but can also reflect changes in the intra- or intermolecular interactions, such as
hydrogen bonding. [16] For the flexible fragments, the higher  values were observed,
whereas for the more rigid parts these values did not exceed 0.5 ppm.
2. Experimental
3

2.1 Synthesis
Phenyl O-glycosides 1-6 were prepared according to our earlier published procedure starting
from penta-O-acetyl--D-glucose or penta-O-acetyl--D-mannose (modified Helferich
method) [17]. The mixture of phenol or 2,6-dimethylphenol (1 mmol), peracetylated sugar
(1.5 mmol) and BF3xEt2O (2.5 mmol) were stirred overnight under an inert atmosphere at r.t.
After aqueous work-up, the crude product was purified on Büchi Sepacore Easy Purification
System (MPLC), and subjected to spectroscopic analysis. The acetylated glucosides 1, 3 and 5
were deacetylated according to Herzig et al. (MeOH, KCN, r.t.). [18,19] The crude free
glycosides 2, 4 and 6 were purified by MPLC chromatography and subjected to spectroscopic
analysis.
All compounds 1-6 (Fig. 1) were obtained as white solids and gave 1H and 13C NMR spectra
in solution identical with those described in the literature: 1 [20,21], 2 [22], 3 [23–26],
4 [23,27,28], 5 [29] and 6 [30,31].
2.2 1H and 13C NMR spectra in solution and solid-state
Cross-polarisation magic angle spinning (CPMAS) 13C NMR spectra were recorded on a
Bruker DSX-400 spectrometer (100.16 MHz). Powder samples were spun at 10 kHz in a
4 mm ZrO2 rotor with a contact time of 2 or 4 ms, a repetition time of 6 s and a spectral width
of 20 kHz used for accumulation of 700-900 scans. Chemical shifts were calibrated indirectly
through glycine C=O signals recorded at 176.0 ppm relative to TMS. 1H-13C HSQC and DFQ
NMR correlation spectra for CDCl3 or CD3OD solutions were obtained using a Bruker 400
MHz spectrometer. Chemical shifts () are reported relative to TMS.
2.3 X-ray diffraction measurements
Suitable mono-crystals for the X-ray measurement were obtained by slow evaporation of
saturated methanolic solutions of the desired compounds. Single crystal X-ray diffraction
measurements for 1, 3 and 6 were performed on a Bruker D8 VENTURE diffractometer
equipped with a Photon 100 area detector, TRIUMPH monochromator and a MoKα fine focus
sealed tube (λ = 0.71073 Å) at a temperature of 100 K. The configurations of all investigated
compounds were known, thus it was not necessary to use copper radiation. The frames
collected with the Bruker APEX2 program [32] were integrated with the Bruker SAINT
software package [33] using a narrow-frame algorithm. Data were corrected for absorption
effects using the multi-scan method (SADABS). [34] The structures were solved and refined
using the Bruker SHELXTL software package. [35,36] The atomic scattering factors were
taken from the International Tables [37]. Molecular graphics were prepared using the
Diamond 3.2 program. [38]
4

All structural details and refinement parameters for 1, 3 and 6 are collected in Table 1. The
crystal structures were deposited in the Cambridge Crystallographic Data Centre under the
following numbers: CCDC 1451885-1451887.
Table 1. Crystal data and refinement parameters for single crystals of 1, 3 and 6.
Identification code
Chemical formula
M/ g mol-1
3 (CCDC 1451885)
C20H24O10
424.39
1 (CCDC 1451886)
6 (CCDC 1451887)
C22H28O10
C13H20O7
452.44
288.29
T/ K
100(2)
100(2)
100(2)
λ/ Å
0.71073
0.71073
0.71073
Crystal size/ mm
Crystal system
Space group
Unit cell dimensions
0.184×0.259×0.440
monoclinic
P21
0.134×0.180×0.425
orthorhombic
P212121
0.106×0.271×0.303
orthorhombic
P212121
a=5.7903(2) Å
b=38.6994(12) Å
c=9.5870(3) Å
β=107.3680(10)°
a=5.4950(2) Å
b=14.0554(4) Å
c=30.3088(9) Å
a=7.1642(3) Å
b=7.7570(3) Å
c=24.7538(9) Å
V/ Å3
2050.32(12)
4
2340.88(13)
4
1375.64(9)
4
Z
D/ g cm-3
μ/ mm-1
F(000)
θmin. θmax
Index ranges
1.375
1.284
1.392
0.111
0.102
0.113
896
960
616
3.06, 25.05°
2.48, 25.04°
2.75, 25.04°
-6≤h≤6.
-46≤k≤46
-11≤l≤11
-6≤h≤6.
-16≤k≤15
-36≤l≤36
-8≤h≤8
-9≤k≤9
-28≤l≤29
Reflections collected
Independent reflections
Data completeness
Absorption correction
Tmax. Tmin
35545
29509
17388
7216 [Rint=0.0241]
99.8%
4142 [Rint=0.0275]
99.9%
2435 [Rint=0.0258]
99.8%
multi-scan
multi-scan
0.986, 0.958
4142 / 0 / 295
multi-scan
0.988, 0.966
2435 / 0 / 202
0.980, 0.953
7216 / 19 / 695
Data / restraints /
parameters
Goodness-of-fit on F2
1.048
1.094
1.070
Final R indices
6708 data; I>2σ(I)
3870 data; I>2σ(I)
2367 data; I>2σ(I)
R1=0.0292, wR2=0.0672 R1=0.0278, wR2=0.0615 R1 = 0.0300, wR2=0.0734
all data all data all data
R1=0.0334, wR2=0.0691 R1=0.0316, wR2=0.0632 R1 = 0.0313, wR2=0.0741
0.178, -0.137 0.167, -0.137 0.188, -0.161
ρmax. ρmin / eÅ-3
2.4 Theoretical calculations
Determination of the most stable conformations of compounds 1-6 consisted in creating
potential energy maps with relation to values of the torsion angles around the glycosidic
linkage (φ and ψ) (φ = O1–C1’–O2–C1; ψ = C1’–O2–C1–C2). Examination of the potential
energy surface of compounds containing sugar moieties can be a real challenge due to
5

possible non-bonded interactions caused by a number of hydroxyl groups. Different relative
positions of these groups may affect the energy of the conformation. Taking into account all
combinations of orientation pendant groups in space would be impractical due to
computational workload needed even only assuming their staggered positions. Thus, in the
current method, the genetic algorithm searched for the most favourable spatial orientation of
pendant groups prior to optimisation of the conformation at every point on the potential
energy map.
Potential energy maps were created using the grid search method combined with a genetic
algorithm, namely the GAAGS (Genetic Algorithm-Assisted Grid Search) method. This
technique was successfully employed in the conformational analysis of phenyl galactosides
[11] as well as in other groups of compounds such as acetylated and deacetylated
disaccharides. [39,40] For this purpose, special software incorporating the GAAGS technique
was developed and adapted to cooperate with the TINKER 6.3 molecular mechanics package.
[41] This tool served to create potential energy maps using an MMFF94 molecular mechanics
force field [42,43] at default and elevated values of the dielectric constant: ε = 1.0 and ε = 4.0
(RMS < 0.001 kcal/mol). Such an approach was aimed at investigating whether the
weakening of non-bonded interactions via changing ε value had an impact on the optimal
arrangement of the pendant groups. Elevated value of the dielectric constant tried to mimic
the more polar surrounding of the molecule. In the case of native carbohydrate moieties, it
was primarily aimed at reducing the strength of the potential intramolecular hydrogen bonds.
The corresponding maps were prepared with Gnuplot software. [44]
For convenience, an additional subscript was added to the name of every conformer. It
contains the value of ε enclosed in parentheses in which a particular conformation was
obtained in the molecular mechanics calculations (e.g., 1a(4) denotes the best conformation of
1a found at ε = 4.0). For every compound, two disparate minima could be observed on the
potential energy map, regardless of the value of the dielectric constant used. In the case of 1
and 2, an additional third minimum could be localised. The resulting conformers (denoted as
a-b or a-c) differed mainly in values of torsion angles φ and ψ and the potential energy. The
selected best conformations were then subjected to further analysis at the DFT level of theory
using the B3LYP hybrid functional with the 6-31+G(d,p) basis set.
The chemical shifts were calculated for the low-energy conformers of 1-6 and the X-ray
molecular geometries of 1, 3, 4 and 6. The position of the hydrogen atoms was optimised for
the structure with frozen heavy atoms. The GIAO–DFT (gauge-independent atomic orbital)
6

approach was used with the same hybrid B3LYP functional and basis set. [45–48] Chemical
shifts were obtained using σ 13C (TMS) calculated at the same level of theory.
The DFT GIAO calculations were performed at the Interdisciplinary Centre for Mathematical
and Computational Modelling (ICM) at the University of Warsaw with the use of the
Gaussian 09 package. [49]
3. Results and Discussion
3.1 X-Ray analysis
a)
b)
c)
d)
Fig. 2. Atom numbering scheme and atomic displacement parameters at 50% probability level
for 1 a); molecule A in 3 b); molecule B in 3 c), and 6 d).
7

From among phenyl O-glycosides 1-6, X-ray data are available only for the glucoside 4
(refcode BONPAV) [50]. Thus, new X-ray structures of compounds 1, 3 and 6 were achieved.
The corresponding atomic displacement parameter plots, presented at 50% probability level,
are shown in Fig. 2a-d. The crystals of 1 and 6 had ordered structures; however, in 1
elongation of the atomic displacement parameters of terminal oxygen atoms in some acetyl
groups can be observed. Similarly to 3, in the structure of 1, no typical hydrogen bonds were
observed, although some weak interactions between C-H and the carbonyl groups occur. Only
for the glycoside 6 hydrogen bond interactions within non-acetylated OH group could be
expected. In addition, the unit cell contains a solvent molecule, and methanol acts as a
hydrogen bond donor and acceptor linking the two mannoside molecules.
The asymmetric unit cell of 3 contains two independent molecules denoted as 3A and 3B,
both disordered in the CH2OAc part (Fig. 2b and c). One acetyl group is also disordered in
molecule B. All of the disordered fragments can alternatively occupy two sites. During
structure refinement, three restraints were used for the anisotropic displacement parameters.
No puckering angles were calculated for the pyranose ring. The same approach was applied
for the glycosidic torsion angles, as these parameters would be too significantly affected by
disorder occurring in the considered fragment. Only weak dispersive interactions can be
observed in the structure as there are no strong hydrogen bond donors (the OH groups were
acetylated), although the phenyl rings of the neighbouring molecules are located in close
proximity and their relative orientation is not typical for π-π stacking. As stated above, all of
the samples had a known chemical configuration. Since a molybdenum X-ray source was
used for the non-heavy atom chiral structure, the results for the Flack parameters [51] for all
structures are non-meaningful. The puckering parameters [52], calculated for the X-ray
structures using the Platon package [53] combined with WinGX software [54] for 1 and 6
including the known dihydrate of 4, are collected in Table 2.
Table 2. Puckering parameters calculated for the X-ray structures.
Puckering Amplitude Q / Å Θ / º
4* 0.584 5.60
Φ / º
349.2
1
6
0.5883(19)
0.524(2)
11.75(19) 1.8(10)
6.7(2) 54(2)
*/ refcode BONPAV [55]
8

For the chair conformation of ideal cyclohexane amplitude Q equal to 0.63 Å, the Θ angle
should be equal to 0º whereas Φ can adopt any value. In the case of a pyranose ring, the
presence of the oxygen atom disturbed the amplitude due to a shortening of the O-C bond in
comparison to the C-C bond. The packing diagrams of the crystal structures of 1, 3 and 6 are
shown in Fig. 3.
Fig. 3. Packing diagram of 1 a), 3 b) and 6 c).
According to the data collected in Table 2, deviations from the ideal cyclohexane geometry
are rather small. It is interesting to note that the highest deviation of the ring from the chair
conformation was observed for the structure 1. In this case, due to acetylation of all the OH
groups, there are no hydrogen bonds. Thus, probably a weak interaction between the side
chains caused such distortion of the pyranose ring.
The XRD data for the analysed aryl O-glycosides indicated the influence of anomeric effects
[56,57] on carbon-oxygen bond length: O1-C1’ in the pyranosidic ring and the glucosidic
bond (C1’- OAr). [58,59] The endo-anomeric effect [60] is characteristic for compounds with
C-X-C-Y structural fragment (e.g. C5-O5-C1-O1). The effect concerns general observations
that the electron with-drawing aglycone favours axial over equatorial orientation.
9

Accordingly to the literature [56,61], the interaction (hyperconjugation) of an axially oriented
lone pair of non-bonding electrons at the endocyclic oxygen atom (O5) and the occupied
antibonding molecular orbital (*) of the C1-O1 bond is responsible for this effect. As a
result, shortening of the O5-C1 and elongation of the C1-O1 bonds of axially oriented
aglycone is observed. This fact is well known from X-ray data. [62] Apart from endo-
anomeric effect in the O-glycosides also exo-anomeric effect is present which determines the
spatial orientation of the aglycon in relation to the sugar ring. [60,61] It confers the opposite
direction than the endo-anomeric effect. The exo-anomeric effect consists on
hyperconjugation of lone pair of electrons of the glycosidic oxygen atom (O1) and
nonbonding orbital (*) of C1-O5 bond in both isomers. The occurrence and magnitude of
generalized anomeric effect (exo- and endo-) depend on the orientation of C5-O5-C1-O1
residue. In case of O-glycosides (e.g. methyl glucoside) the summarized anomeric effect is
responsible for almost identical O5’-C1’ bond length in both isomers, and significant
shortening of C1’-O1 bond due to prevailing β-isomer exo-anomeric effect. The glucosides 1,
3 and 4 bear the equatorially oriented aryl glycone and due to mentioned the earlier exo-
anomeric effect the glycosidic bond (C1’-O1) is shorter (1.393 - 1.395Å) compared to the
O5’-C1’ (1.404 - 1.435 Å). These regularities failed in case of structure 3A, although due to
some molecular disorder in the X-ray structure of 3 it can be omitted. In the investigated
mannoside 6 the aryl substituent is axially oriented and exo- and endo-anomeric effect can
occur and different regularities compare to the mentioned above glycosides were observed.
The bond lengths of C1’-O1 (1.419 Å) were longer than that of the corresponding O5’-C1’
(1.403 Å). (Table 3).
Table 3. Selected bond lengths, interatomic distances and glycosidic bond angles in 1, 3, 4,
and 6.
bond length / Å
torsion angle / °
nr
O5’-C1’
C1’-O1
O1-C1
φ = O5’-C1’-O1-C1
ψ = C1’-O1-C1-C2
1 1.421(2)
1.398(2) 1.407(2)
-69.0(2)
106.6(2)
1.404(7) (O5’B-C1’)
-83.0(4) (for O5’B)
-69.4(4) (for O5’D)
-76.9(5) (for O5’A)
-66.2(4) (for O5’C)
-82.55
3B
3A
1.395(3) 1.397(3)
1.393(3) 1.388(4)
139.7(2) (-45.18)
1.484(7) (O5’D-C1’)
1.36(1) (O5’A-C1’)
1.515(9) (O5’C-C1’)
10.8(4)
4 1.435
6 1.403(3)
1.393
1.389
14.32
-3.1(3)
1.420(3) 1.379(3)
70.8(2)
10

According to the literature, in the CPMAS three predominant rotamers of the CH2OH or
CH2OAc residue in glycone give signals: the carbon C-6’ (gauche-gauche: 62.0 ppm, gauche-
trans: 62.7 - 64.5 ppm and trans-gauche: 66.0 ppm). [63] (Fig.S1) On the basis of the above-
mentioned literature data, experimental chemical shifts (13C CPMAS) and available X-ray
structures (1, 3, 4 and 6) one can state the presence of gauche-gauche rotamers in solids for
glycosides 1, 2 and 4 and gauche-trans for 3, 5 and 6.
3.2. Potential energy surfaces, DFT calculations and solid-state NMR
The potential energy maps of compounds 1-6 were obtained by systematic rotation of the
torsion angles around the glycosidic linkage (φ and ψ) (φ = O1–C1’–O2–C1; ψ = C1’–O2–
C1–C2). In case of the φ angle, a full 360° rotation was performed from -180º to +180º,
whereas in the case of the ψ angle the rotation within the range of -60° to +120° was taken
into account due to the symmetry of the phenyl and 2,6-dimethyl phenyl substituents. Every
torsion angle was changed by 20°.
The MMFF94 potential energy maps generated for ε = 4.0 are presented in Fig. 4 (for the map
calculated with ε = 1.0, see Supplementary Inf. Fig. S3).
The torsion angles and relative values of potential energy for the preliminary low-energy
conformations of 1–6 found by GAAGS method are presented in Table S1. These values refer
to conformations subjected to fully relaxed optimisation with molecular mechanics (MMFF94
force field, RMS < 0.0001 kcal/mol at appropriate value of ε).
The final low-energy conformations were found by DFT optimisation (B3LYP hybrid
functional with 6-31+G(d,p) basis set) of the best conformations determined at molecular
mechanics level listed in Table S1.
In the case of minima regions on the potential energy map for 2, 4 and 5, elevation of the
dielectric constant at the MMFF94 level of calculations didn’t lead to the significantly
different optimal arrangement of the pendant groups. Slight differences in geometries after
molecular mechanics calculations caused by weakening of the non-bonded interactions
disappeared after DFT optimization. Thus, results of all DFT calculations for the best
conformations of 2, 4, and 5 were reported using notation without subscript indicating which
ε value was used. For example, a pair of conformers 4b(1) and 4b(4), which differed in φ/ψ
values of about 7.1°/13.5° after MMFF94 calculations (see Table S1), led to the identical
conformation denoted as 4b after DFT optimization (see Table 4). However, this was not the
case for the minima of 1, 3 and 6.
11

The corresponding values for low-energy conformations of 1–6 and the X-ray data of 1, 3, 4
and 6 obtained by the DFT calculation were collected in Table 4.
Table 4. Glycosidic torsion angles for low-energy conformations of 1–6 found by GAAGS
optimised with the DFT (B3LYP/6-31+G(d,p) method and for the X-ray structures of 1, 3, 4
and 6 (the low-energy conformers of the analysed glycosides 1-6 were marked in italics; for a
graphical presentation, see Supplementary Inf. Fig. S2.).
Conformations
1a(1)
1a(4)
1b(1)
1b(4)
1c(1)
1c(4)
1 (XRD)
2a
2b
Φ / °
-71.5
-73.9
-121.9
-118.3
54.4
107.1
-69.0
-69.1
-112.8
111.5
-82.4
-82.0
70.5
Ψ / °
103.7
107.4
57.8
ΔE / kcal/mol
2.33
0.00
4.21
3.16
7.60
5.26
--
0.00
3.23
4.62
2.19
0.00
4.74
3.02
--
50.3
102.5
113.1
106.6
107.9
63.9
105.5
23.7
21.7
2c
3a(1)
3a(4)
3b(1)
-48.6
-37.6
10.8
3b(4)
70.0
3 (XRD_A)
3 (XRD_C)
3 (XRD_B)
3 (XRD_D)
4a
4b
4 (XRD)
5a
5b
6a(1)
6a(4)
6b(1)
-76.9
-66.2
-83.0
-69.4
-79.8
73.4
-82.6
76.2
133.4
73.7
10.8
--
--
--
-45.2*
-45.2*
13.1
-43.7
14.3
-18.7
26.7
-11.2
-12.1
15.6
0.00
3.65
--
0.00
1.91
0.27
0.00
2.84
2.35
--
74.9
140.6
137.5
70.8
6b(4)
6 (XRD)
19.5
-3.1
* Due to the symmetry of the phenyl substituent, values of ψ torsion angle are given in the range -60° - +120° in
order to facilitate comparison with theoretical values.
12

Fig. 4. Potential energy surfaces obtained using the MMFF94 force field at ε = 4.0 for
compounds 1-6 using the GAAGS method. The contour lines are drawn relative to the lowest
energy value on the map (ΔE = 0). Minima found after relaxed optimisation with molecular
mechanics and DFT methods are marked with + and ▲, respectively ( denotes the location
of the XRD structure).
1 (ε = 4.0)
2 (ε = 4.0)
3 (ε = 4.0)
4 (ε = 4.0)
5 (ε = 4.0)
6 (ε = 4.0)
For glucoside 1, three low-energy conformations denoted as 1a-c were located. The lowest 1a
with φ/ψ = -71°/104° (1a(1)) or -74°/107° (1a(4)) and the second one near φ/ψ = -122-118°/50-
58° (1b(1) and 1b(4)), whereas the third minimum 1c was located near φ/ψ = 54°/102° (1c(1))
and 107°/113° (1c(4)) (Table 4).
Differences between structures 1c(1) and 1c(4) in the φ values were mainly due to the
orientation of the acetate group at the C2′ atom. In all three conformations, 1a-c(4), this group
was oriented in such a manner that the carbonyl eclipsed the hydrogen at the corresponding
13

sugar ring carbon atom. During calculations (ε = 1) the acetate group at C2′ was oriented in
the opposite manner. Similarly, the varied orientation of the acetoxymethyl group was
observed between structures 1b(1) and 1b(4). For ε = 1, the oxygen atom O6′ was found in a g-
g conformation, while the g-t position turned out to be the most optimal.
For the remaining minima pairs, e.g. 1a(1) and 1a(4), as well as 1c(1) and 1c(4), the most
optimal arrangement, was g-t for the former pair and g-g for the latter, regardless of used
dielectric constant value.
In the case of 2, the three minima were located near φ/ψ = -69°/107° (2a), -112°/64° (2b) and
111°/105° (2c) (Table 4). All conformations of 2 were characterised by a primary hydroxyl
group positioned in the g-t orientation, and the secondary hydroxyl groups were oriented
counter-clockwise regardless of which dielectric constant value was used. Scanning the φ/ψ
space of 3 yielded minima near φ/ψ = -82°/24° 3a(1) and 70°/-49° 3b(1) when the calculations
were run at a default dielectric constant. Calculations conducted at ε = 4 revealed minima near
φ/ψ = -82°/22° (3a(4)) and 70°/-38° (3b(4)) (Table 4). The slight difference in ψ values was a
result of the different orientation of the secondary acetate group at C2′; the situation is
analogous to 1. In all of the theoretical conformations of 3, the oxygen atom O6′ was found in
the g-t conformation. Also, in this case, the DFT calculations confirmed that the conformation
of the pendant groups obtained at ε = 4 in the MMFF94 calculation, turned out to be more
energetically favourable. The closest structure to the present in the XRD with respect to the
values of the glycosidic torsion angles was 3a(4). Both conformations were characterised by a
practically identical arrangement of the pendant groups. Due to the presence of two
independent molecules in the XRD structure of 3 an occupancy by disordered fragments two
alternatives sites. The four conformations of 3 (denoted as XRD_A-D) were obtained to
preserve the symmetry of the spatial group for the DFT calculation. The analyzed conformers
were different in angle φ and ψ values for A: -76.9° and 10.8°; B: -83.0° and 45.2°; C: -66.2°
and 10.8°; and D: -69.4° and 45.2°, respectively (Table 4). The calculated chemical shifts for
conformers 3XRD A-D are presented in Table S5.
The location of the minima regions on the φ/ψ surface for 4 is practically the same as those
for the acetylated glucoside 3. Minimum 4a was found in the vicinity of φ/ψ = -80°/13°,
whereas 4b was near 73°/-44° (Table 4). All optimal conformations of 4 had secondary
hydroxyl groups oriented in a counter-clockwise manner, whereas the free hydroxymethyl
group was positioned in the g-t conformation. The minimum 4a turned out to be the closest to
the conformation found in the crystal structure. The difference in the φ torsion angle was 2.8°,
14

whereas in the case of ψ the difference was 1.2°; however, in the crystal structure, the free
hydroxymethyl group was found in the g-g conformation.
In the case of 5, the minima were located near φ/ψ = 76°/-19° (5a) and 133°/27° (5b (see
Table 4). In all conformations, secondary acetate groups were positioned in such a manner
that the carbonyl group eclipsed the hydrogen at the corresponding pyranose ring carbon atom
whereas the primary acetate group was found in the g-t conformation.
Analysis of non-acetylated mannoside 6 revealed minima located near φ/ψ = 75°/-12° (6a(4)),
or 137°/19° (6b(4)). In all of the obtained conformations, secondary hydroxyl groups were
oriented in a clockwise manner. Structures 6a(1) and 6a(4) differed mainly in the arrangement
of the hydroxymethyl group. In 6a(1) this group was in the t-g position, and in 6a(4) it was in
g-g position. An analogous difference was observed between the remaining minima pair, i.e.
6b(1) and 6b(4). It should be noted that in the crystallographic structure this group was found in
a t-g conformation. This arrangement may result from the possibility of the stabilising
hydrogen bond between the hydrogen atom of a methanol molecule and the oxygen of the
hydroxyl group at the C6'. The closest conformation to that present in the crystal structure
with regard to values of φ and ψ torsion angles was 6a(4).
13C CPMAS NMR measurements supported by DFT calculation has become a crucial
technique for structural studies of all kinds of solid powders, [12,64,65] particularly when a
13
collection of X-ray data was not possible. The C NMR spectra were recorded for all of the
studied glycosides 1-6 in solid phase and solutions (1, 3 and 5 in CDCl3, and 2, 4 and 6 in
CD3OD). The chemical shift values () are presented in Table 5 for the acetylated 1, 3, and 5
and for the deacetylated glycosides 2, 4 and 6 in Table 6. The selected 13C CPMAS spectra of
the glycosides 1, 3 and 6 are illustrated in Fig. 5. For the spectra of other analysed
compounds, see Supplementary Inf. Fig. S4.
Table 5. 13C NMR chemical shifts () for acetylated glycosides 1, 3 and 5 in solution and solid
state (the differences of  = solution – solid are in parentheses for values above 1.0 ppm).
carbon
1
3
5
solution
(CDCl3)
solid
solution
(CDCl3)
solid
solution
(CDCl3)
solid
C-1’ 101.3
C-2’ 71.6
--
101.8
73.2 (-1.6)
--
99.1
71.1
--
96.3 (2.8)
72.2 (1.1)
71.4
95.7
69.4
--
95.3
68.4 (1.0)
--
15

C-3’ 72.8
C-4’ 68.3
C-5’ 71.4
C-6’ 61.5
73. 7
68.8
71.9
68.3
72.7
73.4 (-1.5)
69.2
74.2 (-1.5)
62.2 (gt)
156.3
118.2 (-1.3)*
129.2*
123.0
68.9
65.9
69.1
62. 1
155.6
116.5
129.6
123.0
--
67.0 (1.9)
66.0
69.3
63.8 (gt) (-1.7)
155.0
116.0
130.5
122.2
--
130.5
73.2 (-1.8)
60.5 (gg) (1.0) 61.9
152.7
133.6 (-2.2)*
129.6*
124.0
C-1
C-2
C-3
C-4
152.6
131.4
128.8
124.9
--
156.8
116.9
129.5
123.3
--
--
121.9 (1.4)
129.2*
118.2 (1.3)*
C-5
C-6
128.8
131.4
128.9*
128.9 (2.5)*
129.5
116.9
129.6
116.5
116.0
C=O 170.4
170.2
169.2
CH3 20.6
20.5
20.4
C2” 16.7
C6” 16.7
171.3
170.8
170.5 (1.5)
22.1 (-1.5)
21.0
--
15.9
14.7 (2.0)
170.5
170.2
169.3
20.6
20.5
20.5
--
171.5 (-1.0)
170.1
--
170.5
169.9
169.7
20.8
20.7
20.6
--
171.4
170.0
--
21.5
20.2
--
23.0 (-2.4)
21.4
20.4
--
--
--
--
--
--
* / arguable assignment done by DFT calculation.
Table 6. 13C NMR chemical shifts () for deacetylated glycosides 2, 4 and 6 in solution and
solid state (the differences of  = solution –  solid) are in parentheses for values above 1.0 ppm).
* / arguable assignment done by DFT calculation.
carbon
2
4
6
solution solid
(CD3OD)
solution solid
(CD3OD)
solution solid
(CD3OD)
105.5
75.9
77.9
71.9
78.1
63.0
--
103.3 (2.2)
75.3
101.9
74.5
77.6
71.0
77.7
98.7 (3.2)
99.3
71.3
71.8
67.6
74.1
96.3 (3.0)
C-1’
C-2’
C-3’
C-4’
C-5’
73.0 (1.5)
72.9 (4.7)
66.7 (4.3)
76.3 (1.4)
59.9 (gg) (2.4) 62.0
--
69.8 (1.5)
69.8 (2.0)
68.9 (-1.3)
73.3
63.3 (gt) (-1.3)
--
75.3 (2.6)
68.8 (3.1)
77.1 (1.0)
61.3 (gg) (1.7) 62.3
59.5 (gg) (3.5) --
C-6’
--
154.9
134.0
129.8
152.4 (2.5)
133.4*
129.8*
158.6
117.5
130.1
156.5 (2.1)
115.6 (1.9)
129.1 (1.0)
157.1
117.1
130.0
156.8
112.3 (4.8)*
130.4
C-1
C-2
C-3
C-4
C-5
C-6
C2”
C6”
125.4
129.8
134.0
17.5
125.0
123.2
130.1
117.5
--
121.6 (1.6)
129.1 (1.0)
115.6 (1.9)
--
122.8
130.0
117.1
--
122.0
130.4
118.7(-1.6)*
--
--
128.2 (1.6)*
128.2 (5.8)*
16.5 (1.0)
15.3 (2.2)
17.5
--
--
--
16

Fig. 5 13C CPMAS NMR spectra of selected phenyl glycosides 1 a), 3 b) and 6 c).
Full assignment of chemical shifts in solution was done with the aid of 2D NMR techniques
13
(DFQ, HSQC and HMBC). The C NMR spectra of glycosides 1-6 in solution showed the
common signal attributed to aryl carbons C3 and C5 and one to C2 and C6. In the glycosides
3-6, the signal from C3/5 was deshielded in comparison to that of C4 and C2/6 (in the
sequence C1>C3/5>C4>C2/6), whereas in the glucosides 1 and 2 signal C2/6 was more
deshielded (in sequence C2/6>C3/5>C4). The carbon signals in the 13C CPMAS NMR spectra
of 3, 4 and 5 were assigned by comparison with those obtained for the solution. The tertiary
signals of C2 and C6 in the glucosides 1 and 2 were identified on the basis of dipolar
dephasing experiments (Fig. 5a).
The correlation between the experimental chemical shift values (CPMAS) and the calculated
chemical shifts (, GIAO/DFT B3LYP/6-31+G(d,p)) for all of the cases gave a correlation
coefficient (R2) above 0.98. It must be noted that all theoretical calculations of NMR chemical
17

shifts were done for isolated molecules and they don’t take into account the influence of
molecular packing. The best fitting was observed for low-energy conformations of the
acetylated glycosides 1, 3 and 5 (R2>0.997). Minima 1a(4), 3a(4) and 5a were characterized by
the lowest values of Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE). In
deprotected compounds 2, 4 and 6 the fitting was poorer (R2 = 0.991-0.995). The linear
regression (R2) was above 0.993 except 4 (BONPAV), for which it was 0.98.
The  parameter (defined as solution-solid), calculated for all carbons in glycosides 1-6,
provides information about the conformationally, flexible parts of the molecules. The
significant  values were observed in the pyranose ring as well as in the pendant aryl ring of
glycosides 1-6. In acetylated derivatives 1, 3 and 5, greater  values (over 1 ppm) were
detected for aromatic carbons due to the hampered rotation of the aryl substituent in the solid
phase (e.g. -2.2 ppm for C2 and 2.5 ppm for C6 in 1). The  values of the other carbon atoms
were significantly lower (< 1 ppm) when compared to those found for C2 and C6 (see Table
5).
In deacetylated glycosides 2, 4 and 6, significant  values were observed due to restricted
rotation of the aryl ring as well as the presence of intermolecular hydrogen bonding (e.g. 4.7
ppm for C3’ and 4.3 ppm for C4’ in 4) (Table 6). The difference in chemical shift values for
ortho aryl carbons C2 and C6 ( = C2 - C6) in the glucoside 1 reached 4.7 ppm (see Table 5
CPMAS value) The calculated  values for the low-energy conformers of 1a-c(1) and 1a-c(4)
are summarised in Table S2.
Taking into consideration the influence of aromatic ring rotation on the carbon chemical shift,
the chemical shifts () were calculated for conformers 1a(4) upon rotation around the  angle
(from -60° to 120°). The corresponding values are included in Supplementary Information
Table S3. It was found that the statistic dispersion (max - min) in the sugar part was the largest
for the carbon atom neighbouring to the rotated pendant group. It amounted to 15.33 ppm for
the anomeric atom C1' and less than 1.8 ppm for the remaining sugar atoms (see data
collected in Table S3). The separation of resonances of C2 and C6 in the CPMAS spectra of 1
was interpreted as a result of hampering of the aryl group rotation. The XDR data of 1 proved
that in solid phase the aryl ring is locked in conformation with angles φ = -69.0° and ψ =
106.6° with the intramolecular distances O1-C6 and O1-C2 3.743Å and 3.230Å, respectively.
The distance between endocyclic ring oxygen (O5’) and aromatic carbons C2 and C6 found
for XRD structures of 1, 3A, 3B and 6 are shown in Fig. 6.
18

a)
b)
c)
d)
Fig. 6. X-ray structures of phenyl O-glycosides with the indicated distance in Å (dotted lines):
1 a), 3A b), 3B c), 6 d).
It is worth noting that the aryl substituent rotation should be easier in the glucoside 3 (lack of
flanking methyl groups) or in glycosides with axial aryl residue (mannosides 5 and 6).
Furthermore, the liberation of the hydroxyl group (deacetylation of glycosides 1, 3 and 5)
enables the formation of hydrogen bonding, which has a significant influence on the
deacetylated glycosides 2, 4 and 6 signal shape in 13C CPMAS NMR.
The CPMAS spectra of 2 and 4 showed a significant signals broadening with half-width
achieving 130 - 150 Hz for 2 and almost twice higher for 4 (180 - 250 Hz) (for spectra 2, and
4, see Supporting Inf. Fig. S4). The significant  values for glycosides 2 and 4 were
established for almost all carbons signal in the sugar moiety Table 6. The XRD data of 4
19

(refcode BONPAV) indicated that the deacetylated glucoside 4 co-crystalized with two
molecules of water and that all hydroxyl groups are involved in hydrogen bonding (Fig. S5).
From the other side, the CPMAS spectrum of 6 showed an additional signal coming from co-
crystalized methanol molecules (CH3OH, singlet at 49.0 ppm) (Fig. 5c). Significant  values
for 6 were calculated for the signals of C6 deshielded,  = -1.6 ppm) and C2 carbons
(shielded,  = 4.8 ppm) while in solution it was one signal at 117.1 ppm (Table 6).
The X-ray structure of 6 as co-crystals with methanol is presented in Fig. 6d. According to the
data for 6, the aryl ring is oriented almost perpendicularly to the averaged plane of the
carbohydrate moiety. The bond angles φ (O1–C1’–O2–C1) and ψ (C1’–O2–C1-C2) in the
analysed structure were found to be 70.8 and -3.1°, respectively.
Conclusions
The O-glycosides 1-6 were studied by means of 13C CPMAS NMR the highest  values
(defined as  = solution - solid) were observed for the ortho aryl carbons (C2 and C6) due to
hampered rotation of the aryl substituent in the solid phase. The effect was recognized as a
dynamic disorder of so-called “flip-flop” of the aromatic ring. The new X-ray structures of
aryl glycosides (glucosides 1 and 3 and mannoside 6) were obtained. Crystalline 1 and 6 had
ordered structures with one molecule in the independent part of the cell while the asymmetric
part of the crystal structure of 3 contained two molecules disordered in the sugar residue
(CH2OAc). The most deformed ring was observed in the glucoside 1 which possess a bulky
aryl aglycone. Due to steric hindrance, the aromatic ring is significantly twisted out and the
aryl carbons (C2 and C6) are in a different distance from the endocyclic ring oxygen atom
(O1). The most stable conformations of compounds 1-6 were established using the grid search
method combined with the genetic algorithm, namely the GAAGS (Genetic Algorithm-
Assisted Grid Search) method. In the first attempt, the resulting conformations were
optimized using an MMFF94 molecular mechanics force field at two values of dielectric
constant (ε = 1.0 and 4.0). The best conformations obtained in such a way were subjected to
further analysis using the B3LYP hybrid functional with the 6-31+G(d,p) basis set. The
potential energy maps of compounds 1-6 were obtained by gradual changes of the torsion
angles around the glycosidic linkage (φ and ψ) (φ = O1–C1’–O2–C1; ψ = C1’–O2–C1–C2).
Generally, two disparate minima were observed for compounds 1-6. Furthermore, a third
minimum was localised for 1 and 2. The selected low-energy conformations varied mainly in
20

the torsion angle values (φ and ψ). The correlation between experimental chemical shifts
(CPMAS) and calculated chemical shifts (, GIAO/DFT B3LYP/6-31+G(d,p)) gave a good
correlation coefficient (R2) in the range of 0.98-0.99 in all cases.
Acknowledgements
Financial support from the University of Białystok (project no. BST-124) is gratefully
acknowledged.
The X-ray structures were determined at the Advanced Crystal Engineering Laboratory
(aceLAB), established via generous support from the Polish Ministry of Science and Higher
Education (grant no. 614/FNiTP/115/2011), at the Chemistry Department of the University of
Warsaw.
The DFT GIAO calculations using Gaussian-09 were performed in the Interdisciplinary
Centre for Mathematical and Computational Modelling (ICM) at the University of Warsaw
under the computational grant no. G14-6.
References:
[1] Harborne J.B, Phenolic glycosides and their natural distribution. In: Harborne J.B. (ed), Biochemistry of
phenolic compounds. Academic Press, London, New York, 1964, pp. 124-170;, (n.d.).
[2] S.M. Hopkinson, The chemistry and biochemistry of phenolic glycosides, Quarterly Reviews, Chemical
Society. 23 (1969) 98. doi:10.1039/qr9692300098.
[3] K.J. Jensen, O-Glycosylations under neutral or basic conditions, Journal of the Chemical Society, Perkin
Transactions 1. (2002) 2219–2233. doi:10.1039/b110071h.
21

[4] M. Jacobsson, J. Malmberg, U. Ellervik, Aromatic O-glycosylation, Carbohydrate Research. 341 (2006)
1266–1281. doi:10.1016/j.carres.2006.04.004.
[5] K. Ishihara, H. Hamada, T. Hirata, N. Nakajima, Biotransformation using plant cultured cells, Journal of
Molecular Catalysis B: Enzymatic. 23 (2003) 145–170. doi:10.1016/S1381-1177(03)00080-8.
[6] K. Shimoda, Y. Kondo, K. Abe, H. Hamada, H. Hamada, Formation of water-soluble vitamin derivatives
from lipophilic vitamins by cultured plant cells, Tetrahedron Letters. 47 (2006) 2695–2698.
doi:10.1016/j.tetlet.2006.02.082.
[7] M. Lahmann, J. Thiem, Synthesis of α-tocopheryl oligosaccharides, Carbohydrate Research. 299 (1997)
23–31. doi:10.1016/S0008-6215(96)00331-X.
[8] S. Witkowski, P. Wałejko, Synthesis of a-Tocopheryl Glycosides, Zeitschrift Fur Naturforschung B. 56b
(2001) 411–415.
[9] S. Witkowski, P. Wałejko, M. Knaś, J. Maj, D. Dudzik, J. Marciniak, A.Z. Wilczewska, K. Zwierz, The
cleavage of vitamin E galactoside in the rat tissue homogenates, Il Farmaco. 59 (2004) 669–671.
doi:10.1016/j.farmac.2004.04.007.
[10] M. Knaś, P. Wałejko, J. Maj, A. Hryniewicka, S. Witkowski, M. Borzym-Kluczyk, D. Dudzik, K. Zwierz,
Decomposition of α-Tocopheryl Glycosides in Rat Tissues, Toxicology Mechanisms and Methods. 18
(2008) 491–496. doi:10.1080/15376510802164519.
[11] P. Wałejko, K. Paradowska, J. Bukowicki, B.S.P. S.Witkowski, I. Wawer, Phenyl galactopyranosides –
13C CPMAS NMR and conformational analysis using genetic algorithm, Chemical Physics. 457 (2015)
43–50. doi:10.1016/j.chemphys.2015.05.015.
[12] I. Wawer, Witkowski, Analysis of Solid State 13C NMR Spectra of Biologically Active Compounds,
Current Organic Chemistry. 5 (2001) 987–999. doi:10.2174/1385272013374905.
[13] D.D. Laws, H.-M.L. Bitter, A. Jerschow, Solid-State NMR Spectroscopic Methods in Chemistry,
Angewandte Chemie International Edition. 41 (2002) 3096–3129. doi:10.1002/1521-
3773(20020902)41:17<3096::AID-ANIE3096>3.0.CO;2-X.
[14] C. Dybowski, S. Bai, S.V. Bramer, Solid-State Nuclear Magnetic Resonance, Analytical Chemistry. 74
(2002) 2713–2718. doi:10.1021/ac0201981.
[15] J.C. Lindon, O.P. Beckonert, E. Holmes, J.K. Nicholson, High-resolution magic angle spinning NMR
spectroscopy: Application to biomedical studies, Progress in Nuclear Magnetic Resonance Spectroscopy.
55 (2009) 79–100. doi:10.1016/j.pnmrs.2008.11.004.
[16] H.-R. Tang, P.S. Belton, S.C. Davies, D.L. Hughes, Solid state NMR and X-ray diffraction studies of α-d-
galacturonic acid monohydrate, Carbohydrate Research. 330 (2001) 391–399. doi:10.1016/S0008-
6215(00)00307-4.
[17] B. Helferich, E. Schmitz-Hillebrecht, Berichte. 66 (1933) 378–383.
[18] J. Herzig, A. Nudelman, H.E. Gottlieb, B. Fischer, Studies in sugar chemistry. 2. A simple method for O-
deacylation of polyacylated sugars, J. Org. Chem. 51 (1986) 727–730. doi:10.1021/jo00355a026.
[19] J. Herzig, A. Nudelman, Regioselective heterogeneous O-deacylation of polyacylated sugars,
Carbohydrate Research. 153 (1986) 162–167. doi:10.1016/S0008-6215(00)90208-8.
[20] M. Yamaguchi, A. Horiguchi, A. Fukuda, T. Minami, Novel synthesis of aryl 2,3,4,6-tetra-O-acetyl-D-
glucopyranosides, Journal of the Chemical Society, Perkin Transactions 1. (1990) 1079.
doi:10.1039/p19900001079.
[21] A.E. Pavlov, V.M. Sokolov, V.I. Zakharov, [No Title], Russian Journal of General Chemistry. 71 (2001)
1811–1814. doi:10.1023/A:1013971214679.
[22] R.N. Desai, L.F. Blackwell, TEMPO-Mediated Regiospecific Oxidation of Glucosides to Glucuronides,
Synlett. (2003) 1981–1984. doi:10.1055/s-2003-42048.
[23] E.J. Cocinero, D.P. Gamblin, B.G. Davis, J.P. Simons, The Building Blocks of Cellulose: The Intrinsic
Conformational Structures of Cellobiose, Its Epimer, Lactose, and Their Singly Hydrated Complexes,
Journal of the American Chemical Society. 131 (2009) 11117–11123. doi:10.1021/ja903322w.
[24] C. O’Brien, M. Poláková, N. Pitt, M. Tosin, P.V. Murphy, Glycosidation–Anomerisation Reactions of 6,1-
Anhydroglucopyranuronic Acid and Anomerisation of β-D-Glucopyranosiduronic Acids Promoted by
SnCl4, Chemistry - A European Journal. 13 (2007) 902–909. doi:10.1002/chem.200601111.
[25] V.M. Sokolov, V.I. Zakharov, E.P. Studentsov, Stereoselectivity of Reactions at the Glycoside Center of
Carbohydrates: VII. Synthesis of Aryl &alpha;- and -&beta;-D-Glucopyranosides by Helferich, Catalyzed
by Boron Trifluoride Etherate, Russian Journal of General Chemistry. 72 (2002) 806–811.
[26] E. Smits, J.B.F.N. Engberts, R.M. Kellogg, H.A. van Doren, Reliable method for the synthesis of arylꢀ?-D-
glucopyranosides, using boron trifluoride?diethyl ether as catalyst, Journal of the Chemical Society,
Perkin Transactions 1. (1996) 2873. doi:10.1039/p19960002873.
[27] A.F. Hartog, R. Wever, Substrate Engineering and its Synthetic Utility in the Sulfation of Primary
Aliphatic Alcohol Groups by a Bacterial Arylsulfotransferase, Advanced Synthesis & Catalysis. 357
(2015) 2629–2632. doi:10.1002/adsc.201500482.
22

[28] J. Jung, S. Park, S. Hong, M.W. Ha, H. Park, Y. Park, H.-J. Lee, Y. Park, Synthesis of gold nanoparticles
with glycosides: synthetic trends based on the structures of glycones and aglycones, Carbohydrate
Research. 386 (2014) 57–61. doi:10.1016/j.carres.2013.12.012.
[29] T. Yamanoi, I. Yamazaki, The catalytic synthesis of aryl O-glycosides using triaryloxyboranes,
Tetrahedron Letters. 42 (2001) 4009–4011. doi:10.1016/S0040-4039(01)00644-X.
[30] T. Klein, D. Abgottspon, M. Wittwer, S. Rabbani, J. Herold, X. Jiang, S. Kleeb, C. Lüthi, M. Scharenberg,
J. Bezençon, E. Gubler, L. Pang, M. Smiesko, B. Cutting, O. Schwardt, B. Ernst, FimH Antagonists for
the Oral Treatment of Urinary Tract Infections: From Design and Synthesis to in Vitro and in Vivo
Evaluation, Journal of Medicinal Chemistry. 53 (2010) 8627–8641. doi:10.1021/jm101011y.
[31] S. Kyosaka, S. Murata, M. Tanaka, Rates of acidic and alkaline hydrolysis of substituted phenyl
.ALPHA.- and .BETA.-D-mannopyranosides., CHEMICAL & PHARMACEUTICAL BULLETIN. 31
(1983) 3902–3905. doi:10.1248/cpb.31.3902.
[32] APEX2,. Bruker AXS Inc., Madison, Wisconsin, USA, 2013, n.d.
[33] SAINT,. Bruker AXS Inc., Madison, Wisconsin, USA, 2013., n.d.
[34] SADABS,. Bruker AXS Inc., Madison, Wisconsin, USA, 2012., n.d.
[35] G.M. Sheldrick, Phase annealing in SHELX-90: direct methods for larger structures, Acta
Crystallographica Section A Foundations of Crystallography. 46 (1990) 467–473.
doi:10.1107/S0108767390000277.
[36] G.M. Sheldrick, A short history of SHELX, Acta Crystallographica Section A Foundations of
Crystallography. 64 (2008) 112–122. doi:10.1107/S0108767307043930.
[37] Wilson, A. J. C. (1992). International Tables for Crystallography, Vol. C, pp. 583-584. Dordrecht: Kluwer
Academic Publishers, n.d.
[38] Diamond – Crystal and Molecular Structure Visualization Crystal Impact – K. Brandenburg & H. Putz
GbR, Rathausgasse 30, D-53111 Bonn, 2012., n.d.
[39] J. Bukowicki, A. Wawer, K. Paradowska, Conformational Analysis of Gentiobiose Using Genetic
Algorithm Search and GIAO DFT Calculations with 13 C CPMAS NMR as a Verification Method, Journal
of Carbohydrate Chemistry. 34 (2015) 145–162. doi:10.1080/07328303.2015.1016230.
[40] T. Gubica, J. Bukowicki, D.K. Stępień, A. Ostrowski, D.M. Pisklak, M.K. Cyrański, Solid-state structure
of methyl 2,4,6-tri-O-acetyl-3-O-(2,3,4,6-tetra-O-acetyl-β-d-glucopyranosyl)-β-d-galactopyranoside and
methyl 3,4,6-tri-O-acetyl-2-O-(2,3,4,6-tetra-O-acetyl-β-d-glucopyranosyl)-β-d-galactopyranoside, Journal
of Molecular Structure. 1037 (2013) 49–56. doi:10.1016/j.molstruc.2012.12.027.
[41] J.W. Ponder, Tinker Software Tools for Molecular Design. Version 6.0.2012, n.d.
[42] C.A. Stortz, Comparative performance of MM3(92) and two TINKER MM3 versions for the modeling of
carbohydrates, J Comput Chem. 26 (2005) 471–483. doi:10.1002/jcc.20185.
[43] A.G. Gerbst, A.A. Grachev, A.S. Shashkov, N.E. Nifantiev, Computation techniques in the conformational
analysis of carbohydrates, Russ J Bioorg Chem. 33 (2007) 24–37. doi:10.1134/S1068162007010037.
[44] T. Williams, C. Kelley, et al., Gnuplot 4.6: an interactive plotting program, (2014)., n.d.
[45] A.D. Becke, Density‐functional thermochemistry. III. The role of exact exchange, The Journal of
Chemical Physics. 98 (1993) 5648–5652. doi:10.1063/1.464913.
[46] C. Lee, W. Yang, R.G. Parr, Development of the Colle-Salvetti correlation-energy formula into a
functional of the electron density, Phys. Rev. B. 37 (1988) 785–789. doi:10.1103/PhysRevB.37.785.
[47] G.A. Petersson, M.A. Al‐Laham, A complete basis set model chemistry. II. Open‐shell systems and the
total energies of the first‐row atoms, The Journal of Chemical Physics. 94 (1991) 6081–6090.
doi:10.1063/1.460447.
[48] G.A. Petersson, A. Bennett, T.G. Tensfeldt, M.A. Al‐Laham, W.A. Shirley, J. Mantzaris, A complete basis
set model chemistry. I. The total energies of closed‐shell atoms and hydrides of the first‐row elements,
The Journal of Chemical Physics. 89 (1988) 2193–2218. doi:10.1063/1.455064.
[49] Gaussian 09, Revision D.01, Gaussian, Inc., M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A.
Robb, J.R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G.A. Petersson, H. Nakatsuji, M. Caricato,
X. Li, H.P. Hratchian, A.F. Izmaylov, J. Bloino, G. Zheng, J.L. Sonnenberg, M. Hada, M. Ehara, K.
Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J.A.
Montgomery, Jr., J.E. Peralta, F. Ogliaro, M. Bearpark, J.J. Heyd, E. Brothers, K.N. Kudin, V.N.
Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J.C. Burant, S.S. Iyengar, J. Tomasi,
M. Cossi, N. Rega, J.M. Millam, M. Klene, J.E. Knox, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R.
Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, R.L. Martin,
K. Morokuma, V.G. Zakrzewski, G.A. Voth, P. Salvador, J.J. Dannenberg, S. Dapprich, A.D. Daniels, Ö.
Farkas, J.B. Foresman, J.V. Ortiz, J. Cioslowski, and D.J. Fox, Wallingford CT, (2009), n.d.
[50] P.G. Jones, G.M. Sheldrick, W. Clegg, A.J. Kirby, R. Glenn, Crystal structure of phenyl- β - D -
glucopyranoside dihydrate 1, Zeitschrift Für Kristallographie. 160 (1982) 269–274.
doi:10.1524/zkri.1982.160.3-4.269.
23

[51] H.D. Flack, On enantiomorph-polarity estimation, Acta Crystallographica Section A Foundations of
Crystallography. 39 (1983) 876–881. doi:10.1107/S0108767383001762.
[52] D. Cremer, J.A. Pople, General definition of ring puckering coordinates, Journal of the American
Chemical Society. 97 (1975) 1354–1358. doi:10.1021/ja00839a011.
[53] A.L. Spek, Structure validation in chemical crystallography, Acta Crystallographica Section D Biological
Crystallography. 65 (2009) 148–155. doi:10.1107/S090744490804362X.
[54] L.J. Farrugia, WinGX and ORTEP for Windowsꢀ: an update, Journal of Applied Crystallography. 45 (2012)
849–854. doi:10.1107/S0021889812029111.
[55] P.G. Jones, G.M. Sheldrick, W. Clegg, A.J. Kirby, R. Glenn, Z. Kristallogr., 160 (1982) 269–274.
[56] R.U. Lemieux, S. Koto, D. Voisin, The Exo-Anomeric Effect, in: W.A. Szarek, D. Horton (Eds.),
Anomeric Effect, AMERICAN CHEMICAL SOCIETY, WASHINGTON, D. C., 1979: pp. 17–29.
http://pubs.acs.org/doi/abs/10.1021/bk-1979-0087.ch002 (accessed January 25, 2017).
[57] E. Juaristi, G. Cuevas, Recent studies of the anomeric effect, Tetrahedron. 48 (1992) 5019–5087.
doi:10.1016/S0040-4020(01)90118-8.
[58] The Anomeric Effect, CRC Press. (1994). https://www.crcpress.com/The-Anomeric-Effect/Juaristi-
Cuevas/p/book/9780849389412 (accessed January 25, 2017).
[59] A. Demchenko, 1,2-cis O-Glycosylation: Methods, Strategies, Principles, Current Organic Chemistry. 7
(2003) 35–79. doi:10.2174/1385272033373175.
[60] R.U. Lemieux, A.A. Pavia, J.C. Martin, K.A. Watanabe, Solvation effects on conformational equilibria.
Studies related to the conformational properties of 2-methoxytetrahydropyran and related methyl
glycopyranosides, Canadian Journal of Chemistry. 47 (1969) 4427–4439. doi:10.1139/v69-731.
[61] E.R. Alonso, I. Peña, C. Cabezas, J.L. Alonso, Structural Expression of Exo-Anomeric Effect, The Journal
of Physical Chemistry Letters. 7 (2016) 845–850. doi:10.1021/acs.jpclett.6b00028.
[62] C. Altona, C. Knobler, C. Romers, The conformation of non-aromatic ring compounds. VII. Crystal
structure of trans -2,5-dichloro-1,4-dioxane at 125°C, Acta Crystallographica. 16 (1963) 1217–1225.
doi:10.1107/S0365110X63003200.
[63] R.P. Veregin, C.A. Fyfe, R.H. Marchessault, M.G. Taylor, Correlation of 13C chemical shifts with
torsional angles from high-resolution, 13C-C.P.-M.A.S. N.M.R. studies of crystalline cyclomalto-
oligosaccharide complexes, and their relation to the structures of the starch polymorphs, Carbohydrate
Research. 160 (1987) 41–56. doi:10.1016/0008-6215(87)80302-6.
[64] S. Witkowski, K. Paradowska, I. Wawer, 13C CP/MAS NMR studies of vitamin E model compounds,
Magnetic Resonance in Chemistry. 42 (2004) 863–869. doi:10.1002/mrc.1472.
[65] S. Olejniczak, M.J. Potrzebowski, Solid state NMR studies and density functional theory (DFT)
calculations of conformers of quercetin, Org. Biomol. Chem. 2 (2004) 2315–2322. doi:10.1039/b406861k.
24