Experimental and in silico characterization of a biologically active inosose

Article abstract of DOI:10.1007/s11224-013-0221-5

Inositols have been recently reported to show a biological activity as inhibitors of both glycosidase and amyloid-β protein. After having harvested good crystals suitable for single crystal X-ray diffraction, we performed a comparison with the data inferred by means of a molecular dynamics simulation, based on the use of an appropriate Force Field coupled to the most performing charging scheme. This approach allowed a detailed analysis extended to ultra-fine details, such as atomic displacement parameters. It confirmed the good validity of a robust approach already tested by us in previous studies. A NMR analysis of the molecule in solution was also carried out, to compare the structural findings suggested by the X-ray analysis with the ones in solution and avoid confining them to the solid-state. In this framework, we investigated the above-mentioned inhibiting activity of a class of inososes, by means of a molecular docking investigation, which proved the suggested validity of the studied compound as inhibitor of the α-glucosidase.

Full text of DOI:10.1007/s11224-013-0221-5

Struct Chem (2013) 24:955–965
DOI 10.1007/s11224-013-0221-5
ORIGINAL RESEARCH
Experimental and in silico characterization of a biologically active
inosose
•
Venerando Pistara Giuseppe M. Lombardo
•
`
•
Antonio Rescifina Alessia Bacchi
•
•
Felicia D’Andrea Francesco Punzo
Received: 30 November 2012 / Accepted: 23 January 2013 / Published online: 8 February 2013
ꢀ Springer Science+Business Media New York 2013
Abstract Inositols have been recently reported to show a
biological activity as inhibitors of both glycosidase and
amyloid-b protein. After having harvested good crystals
suitable for single crystal X-ray diffraction, we performed a
comparison with the data inferred by means of a molecular
dynamics simulation, based on the use of an appropriate
Force Field coupled to the most performing charging
scheme. This approach allowed a detailed analysis exten-
ded to ultra-fine details, such as atomic displacement
parameters. It confirmed the good validity of a robust
approach already tested by us in previous studies. A NMR
analysis of the molecule in solution was also carried out, to
compare the structural findings suggested by the X-ray
analysis with the ones in solution and avoid confining them
to the solid-state. In this framework, we investigated the
above-mentioned inhibiting activity of a class of inososes,
by means of a molecular docking investigation, which
proved the suggested validity of the studied compound as
inhibitor of the a-glucosidase.
Keywords Inososes ꢀ Glycosidase inhibitors ꢀ Docking ꢀ
Molecular dynamics simulation ꢀ X-ray structural analysis ꢀ
Atomic displacement parameters
Introduction
Cyclitols are cycloalkanes with one hydroxyl group on
each ring atom and other closely related compounds con-
taining a C=C double bond in the ring such as amino, alkyl
or halogen substituents. Among these, inositols (hexa-
hydroxycyclohexanes) are an interesting group of carbo-
cyclic sugar analogs, which keep attracting a great deal of
attention due to the ubiquitous presence of their derivatives
in living cells and their implication in biological phe-
nomena [1–4]. Among them, myo-inositol is the most
abundant of the nine possible stereoisomers and it plays a
very important role in cellular signal transduction acting as
second messenger [1–4]. D-chiro-inositol is a drug candi-
date for the treatment of type 2 diabetes [5] and polycystic
ovary syndrome [6]. L-chiro-inositol is a potential anti-
diabetic compound [5]. Neo-inositol phosphates have been
isolated from both mammalian tissue [7] and parasitic
amoebas [8], while scyllo-inositol from human brain [9], as
well as from plants [10]. Epi-inositol recently has been
evaluated as a potential anti-depressant drug that could
interact with lithium ions and myo-inositol receptors in the
brain [11, 12].
Dedicated to Professor Aldo Domenicano on the occasion of his 75th
birthday.
`
V. Pistara ꢀ G. M. Lombardo ꢀ A. Rescifina (&) ꢀ F. Punzo (&)
Divisione Chimica, Dipartimento di Scienze del Farmaco,
`
Universita degli Studi di Catania, Viale Andrea Doria 6,
95125 Catania, Italy
e-mail: arescifina@unict.it
F. Punzo
e-mail: fpunzo@unict.it
A. Bacchi
Dipartimento di Chimica Generale ed Inorganica, Chimica
`
Analitica, Chimica Fisica, Universita degli Studi di Parma,
Parco Area delle Scienze 17/A, 43124 Parma, Italy
Even the corresponding ketones of inositols, the so
called inososes, have been found in small portions in
mammalian tissues [13] and in Streptomices griseus [14],
like the common myo-inosose (Fig. 1). Furthermore, these
F. D’Andrea
Dipartimento di Scienze Farmaceutiche, Universita degli
Studi di Pisa, Via Bonanno 33, 56126 Pisa, Italy
`
123

956
Struct Chem (2013) 24:955–965
yield), thus indicating that the solvent and/or the base
promoter play a specific role in the reaction stereochemical
outcome. At that time the structure of this second inosose
remained under investigation.
Nowadays, having obtained good crystals of 5, suitable
for X-ray diffraction, we report the complete character-
ization of inosose 5 by means of X-ray single crystal dif-
fraction and of both 5 and 6 with NMR. Moreover, we have
carried out a molecular dynamics (MD) simulation that,
combined and compared with the X-ray diffraction data of
the same condensed phase, was able to provide structural
information that are otherwise inaccessible [26–30].
Finally, as it was recently reported that several inositols
and their derivatives act as inhibitors of glycosidases, we
tried to determine whether compounds 5 and 6 and their
enantiomers, ent-5 and ent-6, are able to bind effectively
with the active site of the N-terminal catalytic subunit of
human maltase-glucoamylase (NtMGAM).
Results and discussion
Fig. 1 Chemical structures of cited cyclitols
X-ray crystallography
compounds are useful intermediates for cyclitol synthesis
in nature, as proposed in the Kiely’s biomimetic synthesis
of myo-inositol from D-xylo-hexos-5-ulose [15].
The molecular structure of 5 is reported in Fig. 2 along
with its labeling scheme. A ring puckering analysis—i.e.,
Cremer and Pople (C&P) analysis [31]—was performed on
the six-membered ring C1–C2–C3–C4–C5–C6 (see
Table 1 for ring nomenclature), actually the only suitable
ring for this analysis, being the other two aromatic and
therefore planar. The representation of the ring puckering
can be performed by means of a single amplitude-phase
In a recent study about the amyloid-b (Ab) protein [16],
a key mediator of Alzheimer’s disease pathology, the
effects of the inositol and inosose stereoisomers as poten-
tial therapeutic agent for amyloid disorders have been
reported. These compounds exhibit stereospecific inhibi-
tion of Ab aggregation in vitro and in vivo, with significant
decreases in Ab40 and Ab42 levels, vascular amyloid
levels, plaques size and area that were observed by the
authors.
˚
pair—q2 and /2, 0.019(2) A and 27(6)ꢁ, respectively—as
˚
well as a single puckering coordinate q3 (0.577(2) A) in
the case of six-membered rings, where the puckering
degrees of freedom are three. The same results can be
obtained using spherical coordinates, Q, h, and / as shown
in Ref. [32].
In the past years, starting from unsatutared derivatives
of galactopyranosides, some of us studied the synthesis of
1,5-dicarbonyl sugars [17–21], which are key intermediates
for the synthesis of high valued-added compounds such as
gabosines [22], cyclitols [23, 24], and azasugar [25]. Few
years ago, following the Kiely’s biomimetic synthesis [15],
was developed a useful synthetic procedure [23], that,
starting from methyl b-D-galactopyranoside 1, produced
the inosose derivative 3 via the diazabicycloundecene
(DBU)-mediated intramolecular aldol condensation of
l-arabino-aldohexos-5-ulose 2. Finally, the catalytic deb-
enzylation followed by the reduction of inosose gave rise to
epi-inositol 4 as described in Scheme 1 [23]. When the
aldol condensation was carried out with a catalytic amount
of NaOH in aqueous methanol, the formation of inosose 3
was not stereospecific and a second inosose, 5, was iden-
tified as minor product (about 2:1 ratio, 70 % overall
The conformation analysis allows the assignment of a
typical chair (C) conformation to the above-mentioned ring
[33].
The search for hydrogen bonds evidenced the existence
of a complex and vast network of interactions reported in
Table 2.
Based on distance criteria, whereby the donor…acceptor
distance is less the sum of the Van der Waals radii, both
intra- and intermolecular hydrogen bonds are present in the
structure, as well as strong (O–HꢀꢀꢀO) and weak (C–HꢀꢀꢀO)
ones, where the strength depends on the electronegativity
of the donor. The weak intermolecular C–HꢀꢀꢀO interac-
tions assist the packing in the formation of hydrogen
bonded chains along the a axis, already sustained by the
stronger O–HꢀꢀꢀO interactions (O3–HꢀꢀꢀO6 and O6–HꢀꢀꢀO3,
123

Struct Chem (2013) 24:955–965
957
Scheme 1 Synthetic pathway
for inososes 5 and 6. (i) Ref.
[17]; (ii) DBU, toluene, rt; (iii)
a H2–Pd(C), MeOH; b Ni/
Raney/H2, MeOH; (iv) NaOH,
MeOH/H2O
Fig. 2 Molecular structure and
labeling of 5. Thermal
displacement ellipsoids are
drawn at the 50 % probability
level while hydrogen atoms are
represented as spheres of
arbitrary radius
Table 1 Compact nomenclature for the center of gravity (Cg) of the
rings in 5
lipophilic CHꢀꢀꢀp interactions (Fig. 3). A characteristic
zigzag motif is evidenced. 5 Crystallized with a quite low
packing index [42]. Four small voids are present in the
structure although they are not suitable for a possible
inclusion of solvent molecules. In fact the total potential
5
Cg(1)
Cg(2)
Cg(3)
C1–C2–C3–C4–C5–C6
C8–C9–C10–C11–C12–C13
C15–C16–C17–C18–C19–C20
3
˚
solvent volume is 37.4 A which represents only the 2.1 %
of the total cell volume.
Molecular dynamics
Table 2). The ancillary role of weak CHꢀꢀꢀO hydrogen
bonds in supramolecular recognition is widely known both
in the crystal engineering and macromolecular crystallog-
raphy [34, 35]. Interestingly, there is evidence of two
bifurcated hydrogen bonds [36], involving O3–H30ꢀꢀꢀO6
and O5–H50ꢀꢀꢀO3, which provide a further aid to the
overall stabilization of the structure. In this framework, it
could not be neglected the role of the several non-covalent
p–p stacking and T-shaped interactions present and
reported in Table 3 [37–41].
The analysis of the internal molecular fluctuations, which are
directly comparable to those obtained from diffraction data,
provides a tool for validating the simulated trajectory [43].
The description of the various interactions between all the
atoms of the considered system, a prerequisite needed by any
correctly performed MD simulation, is usually done either
by the quantum mechanical (QM) scheme, as introduced by
Car and Parrinello [44], or by classical empirical force fields
(FF). As QM driven MD scheme are quite computationally
expensive, whenever QM effects can be neglected, empirical
FF are preferred [45]. In previous works we performed a
The crystal is organized in two-dimensional layers built
in the bc plane by strong hydrogen bonds, and then
assembled in the third direction, along the a axis, by weak
123

958
Struct Chem (2013) 24:955–965
Table 2 Hydrogen-bonds list
˚
Hydrogen-bond geometry (A, ꢁ)
D–HꢀꢀꢀA
D–H
HꢀꢀꢀA
DꢀꢀꢀA
D–HꢀꢀꢀA
H bond classification
O3–H30ꢀꢀꢀO4
O3–H30ꢀꢀꢀO6i
O5–H50ꢀꢀꢀO2ii
O5–H50ꢀꢀꢀO3ii
O6–H60ꢀꢀꢀO3ii
C6–H61ꢀꢀꢀO5iii
C7–H71ꢀꢀꢀO1
0.72 (3)
0.72 (3)
0.83 (4)
0.83 (4)
0.86 (4)
1.02 (3)
0.97 (3)
2.51 (3)
2.18 (3)
2.24 (4)
2.47 (4)
1.97 (4)
2.51 (2)
2.58 (3)
2.785 (3)
2.843 (2)
3.009 (2)
3.070 (2)
2.818 (2)
3.323 (3)
3.139 (3)
105 (2)
154 (3)
Intra and bifurcated
Bifurcated
155 (3)
Bifurcated
130 (3)
Bifurcated
170 (3)
136.4 (17)
116.9 (18)
Intra
Symmetry codes: (i) -1 - x, ‘ ? y, -‘ - z; (ii) 1 ? x, y, z; (iii) -2 - x, -‘ ? y, -‘ - z. Full geometry parameters are reported. D
identifies the donor, A the acceptor, and H the hydrogen atom involved. Standard uncertainties (s.u.s) are reported in parentheses for all distances
(see ‘‘Experimental’’ section for details). The last two entries are weak intramolecular hydrogen bonds [34]. H bond classification has been done
on the basis of Ref. [36]
the Discover module of the Materials Studio package of
Accelrys. Moreover, to validate the NPT (isothermal and
isobaric) MD simulations from these trajectories, we cal-
culated the anisotropic displacement parameters (adps)—or
temperature factors—to illustrate the atom’s thermal
motions relative orientation and compared them to the ones
inferred by means of the X-ray analysis.
Table 3 Selected p–p–stacking non-covalent interactions between
aromatic rings in 5
˚
˚
Centroid (I)ꢀꢀꢀcentroid (J) Distance (A) Angle (ꢁ) Slippage (A)
Cg (2)ꢀꢀꢀCg (2)i,ii
Cg (2)ꢀꢀꢀCg (2)iii
Cg (3)ꢀꢀꢀCg (3)i
Cg (3)ꢀꢀꢀCg (3)ii
Cg (3)ꢀꢀꢀCg (3)iv
5.060 (3)
5.222 (2)
5.061 (4)
5.059 (4)
5.357 (3)
0
89
0
3.622
–
3.500
3.499
–
0
In order to model the crystal structure without assuming
any particular symmetry, the original P212121 cell was
transformed in a triclinic P1 one with all the symmetry
related atoms, together with the cell parameters, to be
independent variables during a NPT ensemble simulation.
We proceeded by evaluating expanded structure models,
with a 5 9 2 9 1 supercell, i.e., built by replicating the
unit cells five times along a and two times along b to
withdraw border effects introduced by the short periodic
boundary conditions (PBC) of the single unit cell. The
starting models were based on the experimental cell and the
resulting MD trajectories embed cell parameters, which
now are fluctuating around the starting values without
88
Symmetry codes: (i) -1 ? x, y, z; (ii) 1 ? x, y, z; (iii) ‘ ? x, -‘ -
y, -1 - z; (iv) ‘ ? x, -‘ - y, -z. Distances and relative s.u.s are
˚
measured in A and angles in (ꢁ) between adjacent centroids (Cg)
thorough comparison of QM and empirical FF applied to the
study of small organic molecules [46, 47]. The reported
results evidenced how the chosen empiric FF best repro-
duced, even if compared to the QM ones, the ab initio
H-bonding and stacking stabilization energies.
This task was accomplished using the AMBER FF [48],
with the Cornell et al. parameterization, as implemented in
Fig. 3 Crystal packing of 5,
projected down the a axis.
Hydrogen bonds are reported as
a dashed line
123

Struct Chem (2013) 24:955–965
959
evidencing any particular phase transition, thus indicating
that the considered structure could be considered stable.
Furthermore, we tried to check whether the original
AMBER FF charges scheme adequately account for the
experimental structure. In order to perform this task, a step
was made toward the definition of a suitable atomic charge
distribution. In previous works only the charge equilibra-
tion (Qeq) scheme [46, 47], which bases its values on
atoms electronegativity differences and relative distances,
properly kept the MD simulation cell parameters with the
smallest differences with respect to the X-ray ones. How-
ever, as described in Fig. 4, the best results were achieved
by the direct use of the AMBER embedded setup. The
analysis of the above-mentioned figure allows the direct
qualitative evaluation of the closer correspondence of the
AMBER inferred adps (left) with respect to the ones cal-
culated with the Qeq charging scheme (right), as compared
to the ones inferred from the experimental X-ray data
(middle). Interestingly, not only the size of the adps could
be considered very similar, but also their preferential
elongation in a given direction is resembled.
not possible to claim that the calculated adps are system-
atically smaller than the X-ray inferred ones, as often
reported in literature, as their sizes and shapes are heavily
affected by the way the calculation is carried out—i.e.,
relaxing or not the symmetric constraints [49, 50].
Moreover, there is an expected and constant trend that
could be inferred by a visual inspection of the adps layout
obtained by the superimposition of the experimental data
(shadowed) to the simulated ones (solid) as reported in
Fig. 6: the ones belonging to the benzene rings are experi-
encing greater fluctuations around their equilibrium posi-
tions than the ones of the central nonaromatic ring, and
among them, the more external ones show this behavior even
in a more exasperated manner. This is clearly due to their
increasing distance from the center of mass of the molecule,
and this behavior is well-simulated by the computational
In Fig. 5 we reported a quantitative comparison among
the differently inferred adps values (AMBER, Qeq, X-ray
inferred), and the ones calculated basing our simulation on
the restrain of the symmetry equivalent fragments in the
unit cell, i.e., not relaxing but instead keeping the sym-
metry. It is evident that the latter values are quite smaller
than the other ones which, qualitatively show very similar
trends. An explanation for this behavior could be found by
considering the presence of symmetry constraints which
forbid the free and independent movement of each frag-
ment present in the unit cell, thus forcing symmetry-related
fragments to oscillate with opposite phases. The result of
this process is the overall dumping of the libration of each
single atom, evidenced by smaller adps, as reported in the
above-mentioned figure. In the light of this reasoning, it is
Fig. 5 Uequiv (Uiso for hydrogen atoms) trends comparison among
the X-ray experimental values and the MD ones, the latter calculated
with the standard AMBER setup, AMBER with Qeq charging scheme
and without relaxing the symmetry, respectively. Carbon, hydrogen,
and oxygen atoms, are reported in the upper, in the middle and in the
lower part of the figure, respectively
Fig. 4 Visual comparison among asymmetric units of 5 and their
relative adps, as inferred MD simulation based on the original
AMBER setup (left), by X-ray single crystal diffraction (middle) and
AMBER setup with Qeq charging scheme (right)
123

960
Struct Chem (2013) 24:955–965
Table 4 Calculated binding energies and inhibition constants to
catalytic site of NtMGAM for compounds 5 and 6, their enantiomers
and miglitol
Ligand
DGBind calcd. (kcal/mol)
Ki calcd. (lM)
5
-6.0
-6.3
-3.7
-3.6
-8.3
36.9
22.8
2,030
2,110
0.8b
ent-5
6
ent-6
Miglitola
a
With nitrogen azasugar protonated in equatorial
From Ref. [25]; exp. 1.0 from Ref. [57]
b
Fig. 6 Superimposition of the experimental crystal packing with the
model I MD averaged one. The experimental anisotropic thermal
factors are shown (shadowed) and compared with the MD calculated
ones (solid). The agreement in the anisotropic thermal factors is
extended to the whole MD calculated super-cell and not only confined
to the single unit fragment
results which are in very good agreement, once again, with
the X-ray experimental data. Moreover, the accordance in
the anisotropic thermal factors is extended to the whole MD
calculated super-cell and not only confined to the single unit
fragment. However, taking into account that it is well-known
that the overall X-ray low ability to univocally resolve
hydrogen atoms’ position, the MD simulation can produce a
good insight to their relative adps layout. As already com-
mented in Table 2, a complex hydrogen-bond network is
present; the adp analysis of the hydrogen atoms involved in
different type of hydrogen bonds—i.e., linear and bifurcated
ones—seems to confirm that the X-ray inferred values,
although not always trustable in principle, are in very good
agreement with the simulated ones, especially with the val-
ues resulted from the AMBER-based simulation (Figs. 5, 6).
The agreement is evident both in the trends and in each
atom’s value. The inspection of the adps layout, for what
concerns hydrogen atoms, can be source of other cues. In
fact, a preferential direction of elongation of each adp can be
considered sign of a directional constrain: the comparison
between the experimental and computed data can therefore
highlight potential details hidden to the X-ray analysis. At
variance with previous works where the differences between
calculated and experimental adp values evidenced an
underestimation of the weight ascribed by X-ray analysis to
the hydrogen bonding [46, 47], the present analysis was not
able to spot any substantial difference. This is confirmed by
the analysis of Fig. 6, where adps belonging to H30 and
H50—the two hydrogen atoms involved in a bifurcated
hydrogen bond—as well as the one regarding H60—to
mention only the intermolecular hydrogen bonds—show no
appreciable differences even by a direct superimposition of
the calculated and experimental adp values.
Fig. 7 a-D-Glucosidase inhibitors
It is worth analyzing even the good agreement between
the computed and experimental bond distances. However,
among many very similar calculated and experimental
distances, there is actually an overall tendency to overes-
timate the bonds involving hydrogen atoms when com-
paring the simulated data to the experimental ones. This is
probably due to the lack of precision in the hydrogen
determination—i.e., their position—embedded in the X-ray
measurement and not to a MD artifact. In fact, as specified
in the experimental section, by following a quite standard
operating procedure, hydrogen atoms attached to oxygens,
were located in a difference Fourier map and refined iso-
tropically while the other ones bound to carbon were
refined using a riding model. This is another evidence of
how the comparison with the simulated structure allows an
assessed comment on some features of the X-ray one.
NMR studies
The 1D and 2D (NOE, COSY and HETCOR) NMR
experiments are in agreement with the X-ray results,
confirming the absolute stereochemistry inferred by the
solid-state. For what concerns the conformation of the
cyclohexanone ring, NMR results showed that there is a
little distortion as compared to that evidenced by the solid-
state analysis. As a consequence, in the inosose 5 the H2
proton appears as a double doublet (4.31 ppm) with a large
Jax/ax (9.5 Hz) coupling with the adjacent H3 and a small
Jax/ax (1.5 Hz) with H6 proton, whereas H4 proton
123

Struct Chem (2013) 24:955–965
961
Molecular docking
Recently it has been demonstrated that several inositols and
their derivatives act as inhibitors of glycosidases [51–56].
It is therefore plausible that the derivatives 5 and 6 may
present a similar activity. Thus, we conducted a series of
studies of molecular docking to determine whether the
subject compounds and their enantiomers, ent-5 and ent-6,
are able to bind effectively with the active site of the
N-terminal catalytic subunit of human maltase-glucoamy-
lase (NtMGAM). The 3D X-ray crystal structure of NtM-
GAM domain–miglitol complex [57], was retrieved from
the Protein Data Bank [58] and processed, as previously
reported [25], following a tested protocol [59].
The obtained results, reported in Table 4, show that
derivative ent-5 has the best activity among the studied
compounds with an in silico inhibition constant of
22.8 lM, whereas its enantiomer is slightly less efficient.
The unprotected compounds 6 are one hundreds of times
less active of miglitol, chosen as reference, and then
practically useless as drugs. On the contrary, compounds 5
are only about twenty times less active.
Regarding the structure–activity relationships, it is evi-
dent that the relative simplification of compounds 6 respect
to the miglitol (Fig. 7) and, especially, the absence of a
protonable atom, at physiological pH, such the nitrogen
one, that engage an hydrogen bond with the Asp443 in the
catalytic residue, essential for the inhibitory activity [60],
preclude a strong interaction with the enzyme. Vice versa,
the presence of the two benzyl groups allow a more ade-
quate filling of the catalytic pocket establishing, at the
same time, an extensive hydrophobic interactions network
(Fig. 8). Moreover, the benzyl moiety at O4 takes the place
of the miglitol, in the catalytic residue, constraining the
inosose ring to occupy the ?1 site, similarly to what
happens with the second glycosidic ring of the MDL
73945, one of the leader a-D-glucosidase inhibitors derived
from miglitol (Fig. 7).
Finally, the major activity of compound ent-5 with
respect to 5 is due to more extended hydrophobic interac-
tions, as shown in the lower part of Fig. 8.
Fig. 8 2D Schematic view of hydrogen bonds and hydrophobic
interactions for 5 (upper) and its overlap with ent-5 (down) docked in
NtMGAM, prepared with LigPlot? [61]. The circles indicate common
interactions
Conclusions
(3.93 ppm) shows a small trans di-equatorial coupling
(3.8 Hz) with the H5 one (4.22 ppm) and a cis-axial/
equatorial disposition with the H3 (3.93 ppm, 4.0 Hz).
Although these coupling constants agree with the absolute
configuration, in solution they are consistent with a small
variation in the dihedral angles, especially for the axial
benzylic substituent at C4, which appears as in a quasi-
axial conformation, producing a distortion of the classic
chair conformation.
In this study, we carried out a thorough characterization of
5 not only for what concerns the solid-state but also in
solution. The X-ray single crystal structure was the starting
point for a detailed comparison, performed by means of a
molecular dynamics simulation, that not only confirmed the
X-ray inferred experimental data, but also allowed a pre-
cise analysis of other structural features which are not
directly accessible by the traditional experimental
123

962
Struct Chem (2013) 24:955–965
approach. This was done carrying out a qualitative and
quantitative adp analysis, extended to the hydrogen atoms,
i.e., one of the known Achille’s heel in the X-ray structural
determination. It allowed a direct validation of the pro-
posed interpretation of what evidenced by X-rays validat-
ing, at difference with previous works the landscape
already depicted by the experimental analysis [46, 47].
Moreover, this study confirmed that the magnitude of the
calculated adps is strongly affected by the way their sim-
ulation is performed i.e., among other things, whether
symmetry constraints are released or not. On the other
hand, the NMR analysis confirmed, although with a slight
variation due to the greater flexibility experienced by the
molecule in solution, the peculiarities highlighted by the
solid-state analysis. The so inferred structural evidences
were joined with the evaluation of the inhibiting activity of
5, which showed good potentialities, as checked by his
docking capabilities.
resulted mixture was stirred for 2 h. At the end of the
reaction the solvent was removed at reduced pressure and
the residue was extracted with CH2Cl2 (3 9 5 mL). The
combined organic layers were dried over Na2SO4, filtered
and evaporated at reduced pressure. The resulting crude
was purified by flash chromatography (cyclohexane/ethyl
acetate, 6:4), affording inososes 3 and 5 in a 2:1 ratio
(70 % overall yield).
2D-2,4-Di-O-benzyl-(2,3,6/4,5)-pentahydroxycyclohex-
anone 5: yield 24 %, 26 mg; mp 169–171 ꢁC, colorless
25
crystals from acetonitrile; ½aꢁD = -51 (c = 0.31, CHCl3).
1
mmax (KBr) 3471, 3450, 3382, 1725, 1624 cm-1. H NMR
(500 MHz, CDCl3, 27 ꢁC): d 7.25–7.47 (m, 10H, phenyl H),
4.77–4.68 (AB system, 2H, J = 11.5 Hz, CH2Ph), 4.76–
4.48 (AB system, 2H, J = 11.5 Hz, CH2Ph), 4.49 (dd, 1H,
J = 1.5, 4.0 Hz, H6), 4.31 (dd, 1H, J = 1.5, 9.5 Hz, H2),
4.22 (t, 1H, J = 3.8 Hz, H5), 3.93 (dd, 1H, J = 4.0, 9.5 Hz,
H3), 3.92 (dd, 1H, J = 3.8, 4.0, H4); 13C NMR (125 MHz,
CDCl3, 27 ꢁC): d 207.5 (C=O), 139.3 (phenyl C), 138.8
(phenyl C), 129.3, 129.2, 129.0, 128.8 and 128.7 (phenyl
CH), 83.7 (C2), 79.6 (C4), 74.9 (C6), 74.4 (CH2Ph), 73.7
(CH2Ph), 73.4 (C3), 71.9 (C5). Anal. calcd. for C20H22O6: C
67.03, H 6.29 %; found C 67.19, H 6.31 %.
Experimental
Melting points were determined with a Kofler hot-stage
apparatus and are uncorrected. Optical rotations were mea-
sured on a Perkin-Elmer 241 polarimeter at 20 2 ꢁC. All
reactions were followed by TLC on a Kieselgel 60 F254 with
detection by UV light and/or with ethanolic 10 % sulfuric
acid. Kieselgel 60 was used for column and flash chroma-
tography (E. Merck, 70–230 and 230–400 mesh, respec-
tively). 1H NMR spectra were recorded with a Varian VnmrJ
instrument at 500 MHz in the stated solvent (Me4Si was used
as the internal standard). 13C NMR spectra were recorded at
50 MHz. Assignments were made, when possible, with the
aid of DEPT experiments, for comparison with values for
known compounds. High-resolution mass spectra were
recorded on aVG ZAB-2SEdouble focusing magnetic sector
mass spectrometer operating at 70 eV. The hydrogenation
were performed with a Parr apparatus. Solvents were dried
by distillation according to standard procedure [62], and
Debenzylation of 5
To a solution of compound 5 (26 mg 0.07 mmol) in MeOH
(2 mL) was added a spatula tip of 10 % Pd/C and the
suspension was set shaken on a Parr apparatus under
hydrogen (90 psi) for 24 h. The reaction mixture was fil-
tered through Celite and the filtrate was evaporated at
reduced pressure, affording the pure debenzylated com-
pound 6.
2D-2,3,6/4,5)-Pentahydroxycyclohexanone 6: yield 95 %
11.8 mg; colorless crystals from water/2-propanol; mp
25
195–197 ꢁC; ½aꢁD = –68 (c = 0.5, H2O). mmax (KBr) 3665,
2917, 1734, 1563 cm-1. 1H NMR (200 MHz, D2O, 27 ꢁC):
d 4.62 (dd, 1H, J = 1.3, 3.6 Hz, H2), 4.39 (dd, 1H,
J = 1.3, 10.2 Hz, H6), 4.16 (dd, 1H, J = 3.6, 3.9 Hz, H3),
4.05 (dd, 1H, J = 3.3, 3.9 Hz, H4), 3.73 (dd, 1H, J = 3.3,
10.2 Hz, H5); 13C NMR (50 MHz, D2O, 27 ꢁC): d 208.0
(C=O), 75.2, 73.2, 73.1, 72.9, 70.3. Anal. calcd. for
C6H10O6: C 40.63, H 5.66 %; found C 40.69, H 5.64 %.
˚
stored over 4 A molecular sieves activated for at least 24 h at
400 ꢁC. Na2SO4 was used as drying agent for solutions.
Good quality crystals were obtained by dissolving 5 in a
test tube with acetonitrile and adding few drops of water.
The solution was sealed and stored at 4 ꢁC. After several
weeks some prismatic colorless crystals were harvested.
Methods
Chemistry
X-ray crystallography
Synthesis of inososes 3 and 5
5
Shows
a
P212121 orthorhombic unit cell with
˚
˚
˚
a = 5.0600(4) A, b = 12.6478(9) A, c = 28.567(2) A, R
being 5.09 %. Although Friedel pairs were not merged, the
high s.u. value associated to the—meaningless—Flack
To a solution of EtOH/H2O 9:1 (5 mL) containing the
dicarbonyl compound 2 (107.5 mg, 0.3 mmol), was added
a 0.005 M NaOH hydroalcoholic solution (2 mL) and the
123

Struct Chem (2013) 24:955–965
963
parameter (-0.7 and 12, respectively), coupled withthe use of
Mo Ka, suggested that experimental data did not support the
determination of the absolute configuration of the molecule.
This issue has been ascribed to twinning effects that may alter
the intensities distribution. However, it was assigned by
comparison with a reference molecule of known absolute
configuration.
what concerns the electrostatic interactions, the dielectric
value e was set to 1.0, and the Ewald summation method
˚
(accuracy = 0.0001 kcal/mol and update width = 5.00 A)
was used. Finally, a scale factor of 1.3 was applied for the
torsion interaction.
Protocols The MD simulations started from the energy-
minimized structures. The PBC with NPT ensemble
(P = 0.0 GPa and T = 293 K) were used for all the runs. In
order to allow the cell to change both shape and volume, we
used the Parrinello pressure control method [69], whereas
Berendsen’s thermostat with the default decay constant
(0.01 ps) was used to control the temperature [70]. Tran-
sients of 3.5 ns were collected, after a 500 ps equilibration
period, with 0.001 ps integration time and a sampling
interval of 200 time-steps. The energy-optimized structures
were obtained with the smart minimizer of the Discover
An unusual C atoms thermal displacement parameter
(Ueq) ratio, confirmed by the MD inferred data, is the result
of the comparison between carbons having a very different
mobility. This is evidenced by the above or below average
U
eq values of C17, C12, and C19 on one hand and C8, C15,
and C18 on the other. The failure in the Hirshfeld rigid-
bond test difference [63] for both C18–C19 bonds is due to
some disorder still present in the structure after modeling
and, as it is associated with the s.u.; it could be the result of
over refinement. The search for possible solvent accessible
voids was performed using the VOID algorithm setting a
-1
module, satisfying a gradient of 0.1 kcal mol-1
.
˚
A
Modeling The structural models submitted to the simu-
lations were generated on the basis of the single crystal X-ray
structure (unit cells and atomic fractional coordinates) as
reported in the cif file available as supplementary material.
The periodic structural model was built by considering in the
single unit cell the fractional parameters of each atom and
allowing their P212121 symmetry related images to be
developed. This model was then converted to P1 by
removing the symmetry constrains, in order to transform all
the atomic positions into independent sites (four indepen-
˚
˚
grid of 0.20 A and a probe radius of 1.20 A [64]. The H
atoms were all located in a difference map, but those
attached to aromatic carbon atoms were repositioned geo-
metrically with fixed bond lengths and angles to regularize
˚
their geometry (C–H in the range 0.93–0.98 A) and Uiso
(H) (in the range 1.2–1.5 times Ueq of the parent atom),
riding on their carrier atoms. All the other hydrogen atoms
were refined isotropically without restraints. The data
collection was performed at 293 K, and the strategy used
aimed to achieve a complete data set to 2h = 55ꢁ. Some
higher angle data were collected in the process and these
have been included in the refinement.
˚
dent unit formulas in a unit cell with a = 5.060 A;
˚
˚
b = 12.6478 A; c = 28.567 A; a = b = c = 90.0ꢁ). The
next modeling step was that of replicating this cell five times
along the a axis and two times b axis, attaining in this way a
The following software have been used for data treat-
ment and collection: SAINT-Plus [65] and SADABS [66].
Mercury was used for molecular graphics [67].
˚
super-cell of dimensions a = 25.300 A; b = 25.2956 A;
˚
˚
c = 28.567 A; a = b = c = 90.0ꢁ with 40 independent
units. No holonomic constraints have been used.
Atomic displacement parameters
Molecular docking
For the detailed and complete description of the calculation
of adps see Ref. [47].
Computational docking was carried out applying the
Lamarckian genetic algorithm implemented in AutoDock
4.2.3 [71]. For finedocking weusedthefollowing parameters:
Molecular dynamics
˚
grid spacing = 0.260 A; number of runs = 100; npts = 70
Force field settings The Materials Studio 4.4 package
(Accelrys Software Inc., San Diego, CA, USA) was used
(CINECA) to perform the MM and MD calculations through
the Discover program using the implemented AMBER FF.
The charges for each atom were either those implemented in
the FF’s file or those obtained through the charge equili-
bration method which were then scaled up by a factor of
1.414 (H2) [68]. The non-bond (NB) settings were different
for the coulomb and the VdW interactions. For the vdW ones
an atom-based summation method (cutoff = 15.50, spline
70 70 centered on 6a, ga_num_evals = 25,000,000;
ga_pop_size = 150andga_num_generations = 27,000. The
graphical user interface AutoDockTools (1.5.6rc1, R45) [72]
was used for establishing the Autogrid points as well as
visualization of docked ligand-nucleic acid structures.
Supplementary information available
912822 is the CCDC number of the archived CIF file con-
taining the supplementary crystallographic data refined and
studied in this paper. These data are available free of charge
˚
width = 5.00 and buffer width = 2.00 A) with a long range
˚
energy correction (tail correction = 15.50 A) was used. For
123

964
Struct Chem (2013) 24:955–965
`
19. Catelani G, D’Andrea F, Guazzelli L, Pistara V (2009) Toward
from the Cambridge Crystallographic Data Centre via
http://www.ccdc.cam.ac.uk/data_request/cif or contacting
the CCDC, 12 Union Road, Cambridge CB21EZ, UK.
the synthesis of fine chemicals from lactose: preparation of D-xylo
and L-lyxo-aldohexos-5-ulose derivatives. Carbohydr Res
344(6):717–724. doi:10.1016/j.carres.2009.01.014
20. Corsaro A, Catelani G, D’Andrea F, Fisichella S, Mariani M,
`
Pistara V (2003) A new route for the chemical valorisation of
lactose. Environ Sci Pollut Res 10(5):325–328. doi:10.1065/
espr2001.12.104.2
References
`
21. Pistara V, Chiacchio MA, Corsaro A (2010) Towards the syn-
thesis of new dideoxy d-dicarbonyl heptoses. Carbohydr Res
345(10):1482–1485. doi:10.1016/j.carres.2010.03.013
1. Berridge MJ, Irvine RF (1989) Inositol phosphates and cell sig-
naling. Nature 341(6239):197–205. doi:10.1038/341197a0
2. Billington DC (1993) Inositol phosphates: chemical synthesis and
biological significance. VCH Verlagsgeselschaft; VCH Publish-
ers, Weinheim
3. Sasaki K, Taylor IEP (1986) Myo-inositol synthesis from [1-H-3]
glucose in Phaseolus vulgaris L during early stages of germina-
tion. Plant Physiol 81(2):493–496. doi:10.1104/Pp.81.2.493
4. Schmalz H-G, Wirth T (2003) Organic synthesis highlights.
Wiley, Weinheim (Great Britain)
5. Larner J (2002) D-chiro-inositol—its functional role in insulin
action and its deficit in insulin resistance. Int J Exp Diabetes Res
3(1):47–60
6. Iuorno MJ, Jakubowicz DJ, Baillargeon JP, Dillon P, Gunn RD,
Allan G, Nestler JE (2002) Effects of D-chiro-inositol in lean
women with the polycystic ovary syndrome. Endocr Pract
8(6):417–423
`
22. Corsaro A, Pistara V, Catelani G, D’Andrea F, Adamo R, Chi-
acchio MA (2006) A new method for the synthesis of carba-sugar
enones (gabosines) using a mercury(II)-mediated opening of 4,
5-cyclopropanated pyranosides as the key-step. Tetrahedron Lett
47(37):6591–6594. doi:10.1016/j.tetlet.2006.07.023
`
23. Pistara V, Barili PL, Catelani G, Corsaro A, D’Andrea F, Fis-
ichella S (2000) A new highly diastereoselective synthesis of epi-
inositol from D-galactose. Tetrahedron Lett 41(17):3253–3256.
doi:10.1016/S0040-4039(00)00360-9
`
24. Catelani G, Corsaro A, D’Andrea F, Mariani M, Pistara V (2002)
Intramolecular aldol cyclization of l-lyxo-hexos-5-ulose deriva-
tives: a new diastereoselective synthesis of d-chiro-inositol.
Bioorg Med Chem Lett 12(22):3313–3315. doi:10.1016/
S0960-894x(02)00692-3
`
25. Pistara V, Rescifina A, Punzo F, Greco G, Barbera V, Corsaro A
(2011) Design, synthesis, molecular docking and crystal structure
prediction of new azasugar analogues of a-glucosidase inhibitors.
Eur J Org Chem 36:7278–7287. doi:10.1002/ejoc.201100832
26. Alberti G, Lombardo GM, Pappalardo GC, Vivani R (2002)
Modeling and analysis of the X-ray powder diffraction structure
of gamma-zirconium phosphates pillared with butyl chains
through molecular dynamics simulations. Chem Mater 14(1):
295–303. doi:10.1021/Cm011156g
27. Caminiti R, Gleria M, Lipkowitz KB, Lombardo GM, Pappalardo
GC (1997) Molecular dynamics simulations combined with large
angle X-ray scattering technique for the determination of the
structure, conformation, and conformational dynamics of poly-
phosphazenes in amorphous phase: study of poly[di(4-methyl-
phenoxy)phosphazene]. J Am Chem Soc 119(9):2196–2204
28. Carriedo GA, Alonso FJG, Alvarez JLG, Lombardo GM, Pap-
palardo GC, Punzo F (2004) Molecular dynamics (MD) simula-
tions and large-angle X-ray scattering (LAXS) studies of
the solid-state structure and assembly of isotactic (R)-poly
(2,20-dioxy-1,10-binaphthyl-)phosphazene in the bulk state and in
the cast film. Chem Eur J 10(15):3775–3782. doi:10.1002/
chem.200400051
7. Sherman WR, Goodwin SL, Gunnell KD (1971) Neo-Inositol in
mammalian tissues. Identification, measurement, and enzymatic
synthesis from mannose 6-phosphate. Biochemistry (Mosc)
10(19):3491–3499
8. Martin JB, Laussmann T, Bakker-Grunwald T, Vogel G, Klein G
(2000) neo-Inositol polyphosphates in the amoeba Entamoeba
histolytica. J Biol Chem 275(14):10134–10140
9. Shapiro J, Belmaker RH, Biegon A, Seker A, Agam G (2000)
Scyllo-inositol in post-mortem brain of bipolar, unipolar and
schizophrenic patients. J Neural Transm 107(5):603–607
10. Ichimura K, Kohata K, Yamaguchi Y, Douzono M, Ikeda H,
Koketsu M (2000) Identification of L-inositol and scyllitol and
their distribution in various organs in chrysanthemum. Biosci
Biotechnol Biochem 64(4):865–868
11. Shaldubina A, Ju S, Vaden DL, Ding D, Belmaker RH, Green-
berg ML (2002) Epi-inositol regulates expression of the yeast
INO1 gene encoding inositol-1-P synthase. Mol Psychiatry
7(2):174–180. doi:10.1038/sj.mp.4000965
12. Williams RS, Cheng L, Mudge AW, Harwood AJ (2002) A
common mechanism of action for three mood-stabilizing drugs.
Nature 417(6886):292–295. doi:10.1038/417292a
29. Lombardo GM, Pappalardo GC (2008) Thermal effects on mixed
metal (Zn/Al) layered double hydroxides: direct modeling of the
X-ray powder diffraction line shape through molecular dynam-
ics simulations. Chem Mater 20(17):5585–5592. doi:10.1021/
Cm801053d
30. Lombardo GM, Pappalardo GC, Punzo F, Costantino F,
Costantino U, Sisani M (2005) A novel integrated X-ray powder
diffraction (XRPD) and molecular dynamics (MD) approach for
modelling mixed-metal (Zn, Al) layered double hydroxides
(LDHs). Eur J Inorg Chem 24:5026–5034. doi:10.1002/ejic.
200500666
13. Sherman WR, Stewart MA, Kurien MM, Goodwin SL (1968) The
measurement of myo-inositol, myo-inosose-2 and scyllo-inositol
in mammalian tissues. Biochim Biophys Acta 158(2):197–205
14. Horner WH, Thaker IH (1968) The metabolism of scyllo-inositol
in Streptomyces griseus. Biochim Biophys Acta 165(2):306–308
15. Kiely DE, Fletcher HG Jr (1969) Cyclization of D-xylo-hexos-
5-ulose, a chemical synthesis of scyllo-and myo-inositols from
D-glucose. J Org Chem 34(5):1386–1390
16. Nitz M, Fenili D, Darabie AA, Wu L, Cousins JE, McLaurin J
(2008) Modulation of amyloid-b aggregation and toxicity by
inosose stereoisomers. FEBS J 275(8):1663–1674. doi:10.1111/j.
1742-4658.2008.06321.x
31. Cremer D, Pople JA (1975) General definition of ring puckering
coordinates.
ja00839a011
J Am Chem Soc 97:1354–1358. doi:10.1021/
17. Barili PL, Berti G, Catelani G, Dandrea F (1992) New synthetic
pathways to 5-C-alkoxypyranosides and to hexos-5-ulose deriv-
atives. Gazz Chim Ital 122(4):135–142
32. Alvarez JLG, Amato ME, Lombardo GM, Carriedo GA, Punzo F
(2010) Self-organization by chiral recognition based on ad hoc
chiral pockets in cyclotriphosphazenes with binaphthoxy and
biphenoxy substituents: an X-ray, NMR and computational study.
Eur J Inorg Chem 28:4483–4491. doi:10.1002/ejic.201000586
`
18. Catelani G, Corsaro A, D’Andrea F, Mariani M, Pistara V,
Vittorino E (2003) Convenient preparation of L-arabino-hexos-
5-ulose derivatives from lactose. Carbohydr Res 338(22):
2349–2358. doi:10.1016/j.carres.2003.08.001
123

Struct Chem (2013) 24:955–965
965
33. Evans DG, Boeyens JCA (1989) Conformational-analysis of ring
pucker. Acta Crystallogr B 45:581–590. doi:10.1107/S01087
68189008190
34. Desiraju GR, Steiner T (1999) The weak hydrogen bond in
structural chemistry and biology. international union of crystal-
lography monographs on crystallography, vol 9. Oxford Univer-
sity Press, Oxford
glycosidase inhibitors. Bioorg Med Chem Lett 14(5):1209–1212.
doi:10.1016/j.bmcl.2003.12.050
53. Painter GF, Eldridge PJ, Falshaw A (2004) Syntheses of tetra-
hydroxyazepanes from chiro-inositols and their evaluation
as glycosidase inhibitors. Bioorg Med Chem 12(1):225–232.
doi:10.1016/j.bmc.2003.10.003
54. Paul BJ, Willis J, Martinot TA, Ghiviriga I, Abboud KA, Hud-
licky T (2002) Synthesis, structure, and biological evaluation of
novel N- and O-linked diinositols. J Am Chem Soc 124(35):
10416–10426. doi:10.1021/Ja0205378
55. Sun M, Wang T, Shu X (2009) The use of inositol derivative or
salts thereof in the manufacture of medicaments as glycosidase
inhibitors or medicaments for treating diabetes. WO200915
5753(A1), 30 Dec 2009
56. Sureshan KM, Ikeda K, Asano N, Watanabe Y (2008) Efficient
syntheses of optically pure chiro- and allo-inositol derivatives,
azidocyclitols and aminocyclitols from myo-inositol. Tetrahedron
64(18):4072–4080. doi:10.1016/j.tet.2008.02.032
57. Sim L, Jayakanthan K, Mohan S, Nasi R, Johnston BD, Pinto
BM, Rose DR (2010) New Glucosidase inhibitors from an
ayurvedic herbal treatment for type 2 diabetes: structures and
inhibition of human intestinal maltase-glucoamylase with com-
pounds from Salacia reticulata. Biochemistry (Mosc) 49(3):
443–451. doi:10.1021/Bi9016457
35. Steiner T (1996) C–HꢀꢀꢀO hydrogen bonding in crystals. Crys-
tallogr Rev 6:1–51. doi:10.1080/08893119608035394
36. Jeffrey GA, Maluszynska H, Mitra J (1985) Hydrogen-bonding in
nucleosides and nucleotides. Int J Biol Macromol 7(6):336–348.
doi:10.1016/0141-8130(85)90048-0
37. Dinadayalane TC, Leszczynski J (2009) Geometries and stabili-
ties of various configurations of benzene dimer: details of
novel V-shaped structure revealed. Struct Chem 20(1):11–20.
doi:10.1007/s11224-009-9411-6
38. DiStasio RA, von Helden G, Steele RP, Head-Gordon M (2007)
On the T-shaped structures of the benzene dimer. Chem Phys Lett
437(4–6):277–283. doi:10.1016/j.cplett.2007.02.034
39. Fraga ARL, Destri GL, Forte G, Rescifina A, Punzo F (2010)
Could N-(diethylcarbamothioyl)benzamide be a good ionophore
for sensor membranes? J Mol Struct 981(1–3):86–92
40. Grimme S (2008) Do special noncovalent pi–pi stacking inter-
actions really exist? Angew Chem Int Ed 47(18):3430–3434.
doi:10.1002/anie.200705157
58. http://www.rcsb.org/PDBcode 3L4W
41. Sato T, Tsuneda T, Hirao K (2005) A density-functional study on
pi–aromatic interaction: benzene dimer and naphthalene dimer.
J Chem Phys 123(10):104307. doi:10.1063/1.2011396
42. Kitaˇıgorodskiˇı AI (1973) Molecular crystals and molecules.
Physical chemistry, a series of monographs, vol 29. Academic
Press, New York
43. Burden CJ, Oakley AJ (2007) Anisotropic atomic motions in high-
resolution protein crystallography molecular dynamics simula-
tions. Phys Biol 4(2):79–90. doi:10.1088/1478-3975/4/2/002
44. Car R, Parrinello M (1985) Unified approach for molecular-
dynamics and density-functional theory. Phys Rev Lett 55(22):
2471–2474
59. Mazue F, Colin D, Gobbo J, Wegner M, Rescifina A, Spatafora C,
Fasseur D, Delmas D, Meunier P, Tringali C, Latruffe N (2010)
Structural determinants of resveratrol for cell proliferation inhibi-
tion potency: experimental and docking studies of new analogs. Eur
J Med Chem 45(7):2972–2980. doi:10.1016/j.ejmech.2010.03.024
60. de Melo EB, Gomes AD, Carvalho I (2006) Alpha- and beta-glu-
cosidase inhibitors: chemical structure and biological activity. Tet-
rahedron 62(44):10277–10302. doi:10.1016/j.tet.2006.08.055
61. Wallace AC, Laskowski RA, Thornton JM (1995) Ligplot: a
program to generate schematic diagrams of protein ligand inter-
actions. Protein Eng 8(2):127–134. doi:10.1093/protein/8.2.127
62. Perrin DD, Armarego WLF, Perrin DR (1980) Purification of
laboratory chemicals, 2nd edn. Pergamon Press, Oxford
63. Hirshfeld FL (1976) Can X-ray data distinguish bonding effects
45. Hobza P, Sponer J (1999) Structure, energetics, and dynamics of
the nucleic acid base pairs: nonempirical ab initio calculations.
Chem Rev 99(11):3247–3276
from vibrational smearing? Acta Crystallogr
doi:10.1107/s0567739476000533
a 2:239–244.
46. Lombardo GM, Portalone G, Chiacchio U, Rescifina A, Punzo F
(2012) Potassium caffeate/caffeic acid co-crystal: the rat race between
the catecholic and carboxylic moieties in an atypical co-crystal.
Dalton Trans 41(47):14337–14344. doi:10.1039/c2dt31092a
47. Lombardo GM, Portalone G, Colapietro M, Rescifina A, Punzo F
(2011) From the X-rays to a reliable ‘‘low cost’’ computational
structure of caffeic acid: dFT, MP2, HF and integrated molecular
dynamics-X-ray diffraction approach to condensed phases. J Mol
Struct 994(1–3):87–96. doi:10.1016/j.molstruc.2011.03.001
48. Cornell WD, Cieplak P, Bayly CI, Gould IR, Merz KM, Ferguson
DM, Spellmeyer DC, Fox T, Caldwell JW, Kollman PA (1995) A
2nd generation force-field for the simulation of proteins, nucleic-
acids, and organic-molecules. J Am Chem Soc 117(19):5179–5197
49. Reilly AM, Habershon S, Morrison CA, Rankin DW (2010)
Simulating thermal motion in crystalline phase-I ammonia.
J Chem Phys 132(13):134511. doi:10.1063/1.3387952
50. Reilly AM, Wann DA, Morrison CA, Rankin DWH (2007)
Experimental equilibrium crystal structures: molecular dynamics
as a probe for atomic probability density functions. Chem Phys
Lett 448(1–3):61–64. doi:10.1016/j.cplett.2007.09.073
51. Falshaw A, Hart JB, Tyler PC (2000) New syntheses of 1D-and
1L–1, 2-anhydro-myo-inositol and assessment of their glycosi-
dase inhibitory activities. Carbohydr Res 329(2):301–308.
doi:10.1016/S0008-6215(00)00192-0
64. Vandersluis P, Spek AL (1990) Bypass: an effective method for
the refinement of crystal-structures containing disordered solvent
regions. Acta Crystallogr A 46:194–201. doi:10.1107/S01087673
89011189
65. Bruker (2007) SAINT-Plus. Bruker AXS Inc, Madison
66. Bruker (2001) SADABS. Bruker AXS Inc, Madison
67. Macrae CF, Edgington PR, McCabe P, Pidcock E, Shields GP,
Taylor R, Towler M, van De Streek J (2006) Mercury: visuali-
zation and analysis of crystal structures. J Appl Crystallogr
39:453–457. doi:10.1107/S002188980600731x
68. Rappe AK, Goddard WA (1991) Charge equilibration for
molecular-dynamics simulations. J Phys Chem 95(8):3358–3363.
doi:10.1021/J100161a070
69. Parrinello M, Rahman A (1982) Strain fluctuations and elastic
constants. J Chem Phys 76:2662–2666
70. Berendsen HJC, Postma JPM, van Gunsteren WF, DiNola A,
Haak JR (1984) Molecular dynamics with coupling to an external
bath. J Chem Phys 81:3684–3690
71. Morris GM, Goodsell DS, Halliday RS, Huey R, Hart WE, Belew
RK, Olson AJ (1998) Automated docking using a Lamarckian
genetic algorithm and an empirical binding free energy function.
J Comput Chem 19(14):1639–1662. doi:10.1002/(Sici)1096-987x
(19981115)19:14\1639:Aid-Jcc10[3.0.Co;2-B
52. Freeman S, Hudlicky T (2004) New oligomers of conduritol-F
and muco-inositol, synthesis and biological evaluation as
72. Sanner MF (1999) Python: a programming language for software
integration and development. J Mol Graph Model 17(1):57–61
123