Docking-based Design of Galantamine Derivatives with Dual-site Binding to Acetylcholinesterase

Article abstract of DOI:10.1002/minf.201600041

The enzyme acetylcholinesterase is a key target in the treatment of Alzheimer's disease because of its ability to hydrolyze acetylcholine via the catalytic binding site and to accelerate the aggregation of amyloid-β peptide via the peripheral anionic site (PAS). Using docking-based predictions, in the present study we design 20 novel galantamine derivatives with alkylamide spacers of different length ending with aromatic fragments. The galantamine moiety blocks the catalytic site, while the terminal aromatic fragments bind in PAS. The best predicted compounds are synthesized and tested for acetylcholinesterase inhibitory activity. The experimental results confirm the predictions and show that the heptylamide spacer is of optimal length to bridge the galantamine moiety bound in the catalytic site and the aromatic fragments interacting with PAS. Among the tested terminal aromatic fragments, the phenethyl substituent is the most suitable for binding in PAS.

Full text of DOI:10.1002/minf.201600041

Full Paper
www.molinf.com
DOI: 10.1002/minf.201600041
Docking-based Design of Galantamine Derivatives with
Dual-site Binding to Acetylcholinesterase
Georgi Stavrakov†,[a] Irena Philipova†,[b] Dimitrina Zheleva,[a] Mariyana Atanasova,[a] Spiro Konstantinov,[a] and
Irini Doytchinova*[a]
Abstract: The enzyme acetylcholinesterase is a key target
in the treatment of Alzheimer’s disease because of its abili-
ty to hydrolyze acetylcholine via the catalytic binding site
and to accelerate the aggregation of amyloid-b peptide via
the peripheral anionic site (PAS). Using docking-based pre-
dictions, in the present study we design 20 novel galanta-
mine derivatives with alkylamide spacers of different length
ending with aromatic fragments. The galantamine moiety
blocks the catalytic site, while the terminal aromatic frag-
ments bind in PAS. The best predicted compounds are syn-
thesized and tested for acetylcholinesterase inhibitory activ-
ity. The experimental results confirm the predictions and
show that the heptylamide spacer is of optimal length to
bridge the galantamine moiety bound in the catalytic site
and the aromatic fragments interacting with PAS. Among
the tested terminal aromatic fragments, the phenethyl sub-
stituent is the most suitable for binding in PAS.
Keywords: acetylcholinesterase inhibitors · galantamine · molecular docking · ChemPLP · dual-site binding · amyloid beta peptide
1 Introduction
The enzyme acetylcholinesterase (AChE) is a key target in
the treatment of Alzheimer’s disease (AD). It hydrolyses
acetylcholine (ACh), a neurotransmitter playing a main role
in memory and cognition. The AChE inhibition enhances
the levels of ACh and improves the cholinergic transmission
impaired in AD.[1] Additionally, AChE accelerates the aggre-
gation of amyloid-b peptide (Ab) and forms stable neuro-
toxic complexes with Ab.[2–4] The complexes induce Ab-de-
pendent deregulation of intracellular Ca2+ in hippocampal
neurons, mitochondrial dysfunction, neurite network dys-
trophia and apoptosis.[4]
AChE has several binding domains (sites): catalytic,
anionic, acylic, oxyanionic and peripheral anionic.[5–9] The
catalytic triad consists of three residues – Ser203, Glu334
and His447 (hAChE) – and is situated at the bottom of 20 ꢀ
deep and narrow binding gorge. The anionic domain binds
the quaternary trimethylammonium choline moiety of ACh
and is built by four aromatic residues: Trp86, Tyr130, Tyr337
and Phe338. Trp86 is involved in cation-p interactions with
the positively charged head of ACh. The acyl pocket deter-
mines the selective binding of ACh. It consists of two bulky
residues – Phe295 and Phe297 – which prevent the access
of larger choline esters. Gly121, Gly122 and Ala204 form
the oxyanion hole which hosts one molecule of structural
water. This molecule forms a dense hydrogen-bond net-
work between enzyme and substrate and stabilizes the
substrate tetrahedral transition state. The peripheral anionic
site (PAS) lies at the entrance to binding gorge and consists
of five residues: Tyr72, Asp74, Tyr124, Trp286 and Tyr341.
After the catalytic site, PAS is the second most important
drug target binding site of AChE. It modulates the catalysis
allosterically[10] and is implicated in non-cholinergic func-
tions as amyloid deposition,[11] cell adhesion and neurite
outgrowth.[12,13] The Ab peptide binds close to PAS and the
blockade of this domain prevents the AChE-induced Ab ag-
gregation.[11]
The multiple binding site attacks to AChE was initiated
by the development of bis(7)-tacrine analogs – acetylcholi-
nesterase inhibitors (AChEIs) designed to bind at both cata-
lytic site and PAS.[14] This finding prompted the develop-
ment of several series of dual- or multiple-site binding li-
gands with anti-AD activities.[15–23] Molecules with different
but synergistic mechanisms of action have been combined
in hybrid entities.[24–34] The multi-target design strategies
applied for the treatment of AD have been extensively re-
viewed recently.[35–38]
Galantamine (GAL) is an AChE inhibitor widely prescribed
as an anti-AD drug.[39–41] Additionally, GAL is an allosteric
[a] G. Stavrakov, D. Zheleva, M. Atanasova, S. Konstantinov,
I. Doytchinova
Faculty of Pharmacy, Medical University of Sofia, 2 Dunav str.,
1000 Sofia, Bulgaria
*e-mail: idoytchinova@pharmfac.net
[b] I. Philipova
Institute of Organic Chemistry with Centre of Phytochemistry
Bulgarian academy of Sciences, Acad. G. Bonchev str. 9, 1113
Sofia, Bulgaria
[†] Authors with equal contributions
Supporting Information for this article is available on the WWW
under www.molinf.com.
ꢁ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Mol. Inf. 0000, 00, 1 – 9
&1
&
ÞÞ
These are not the final page numbers!

Full Paper
www.molinf.com
modulator
of
nicotinic
acetylcholine
receptors
selected as flexible. These were Tyr72, Asp74, Trp86, Tyr124,
(nAChRs).[42–45] The stimulation of nAChRs increases the in-
tracellular Ca2+ levels and facilitates noradrenaline release;
both effects enhance the cognitive brain function.[46] The
treatment of rat microglia with GAL significantly enhanced
microglial Ab phagocytosis and facilitated Ab clearance in
brains of rodent AD models.[47] It is evident that GAL is
a multiple target drug and is a good scaffold for further de-
velopment. The rational design here is focused on the elon-
gation of N-side chain ending by a bulky substituent able
to interact with PAS. Several series of GAL derivatives with
dual-site binding to the enzyme have been prepared and
tested.[48–52] All of them showed good AChE inhibitory activ-
ities.
Molecular docking is a widely used structure-based
method for virtual screening and drug design, used solely
or in combination with 2D- and 3D-QSAR, high-throughput
screening and/or machine learning methods.[53–65] Recently,
we designed several galantamine derivatives with indole
moiety in the side chain and predicted their AChE binding
affinities by molecular docking simulations.[52] The four best
predicted compounds were synthesized and tested. All of
them showed affinities between 11 and 95 times higher
than the affinity of GAL. The novel derivatives had dual-site
binding to the enzyme – the GAL moiety binds to the cata-
lytic site and the indole moiety interacts with PAS.
Ser125, Trp286, Phe297, Tyr337, Phe338 and Tyr341. Each
run generated 10 docking poses. The poses were ranked
by two criteria: 1) highest fitness score and 2) rmsd (root
mean square deviation) lower than 1.5 ꢀ. Each docking run
was repeated three times. The average score of the highest
scoring poses was considered.
The molecular mass, logP, number of hydrogen bond
donors and acceptors in the molecules were calculated by
ACD/logD v.9.08 (Advanced Chemistry Development, Inc.).
The BBB permeability was predicted by the BBB Predictor
(http://www.cbligand.org/BBB/). It classifies whether a com-
pound can cross the blood-brain barrier (BBB+) or not
(BBB-). This predictor was built by applying SVM and LiCA-
BEDS[70] algorithms on four types of fingerprints (MACCS,
OpenBabel FP2, Molprint and PubChem) of 1593 reported
compounds.[71]
2.2 Synthesis
2.2.1 General
Reagents were commercial grade and used without further
purification. Thin layer chromatography (TLC) was per-
formed on aluminum sheets pre-coated with Merck Kiesel-
gel 60 F254 0.25 mm (Merck). Flash column chromatography
was carried out using Silica Gel 60 230–400 mesh (Fluka).
Commercially available solvents for reactions, TLC and
column chromatography were used after distillation (and
were dried when needed). Melting points were determined
in a capillary tube on SRS MPA100 OptiMelt (Sunnyvale, CA,
USA) automated melting point system (uncorrected). Opti-
cal rotation ([a]D20) were measured on JASCO P-2010 polar-
imeter. The NMR spectra were recorded on a Bruker Avance
In the present study, we design 20 novel GAL derivatives
with alkylamide spacers of different length ending with aro-
matic fragments. The aromatic fragments are selected as
suitable for binding to the aromatic residues in PAS be-
cause of their ability to take part in hydrophobic and p-p
interactions. The designed ligands are docked in rhAChE
and the best scored compounds are synthesized and
tested. Two of them show anti-AChE inhibitory activity in
the nano- and subnanomolar range.
1
II+ 600 (600.13 for H MHz and 150.92 MHz for 13C NMR)
spectrometer with TMS as internal standards for chemical
1
shifts (d, ppm). H and 13C NMR data are reported as fol-
2 Materials and Methods
lows: chemical shift, multiplicity (s=singlet, d=doublet,
t=triplet, q=quartet, br=broad, m=multiplet), coupling
constants (Hz), integration, identification. The assignment
2.1 Molecular Docking Protocol
1
The X-ray structure of rhAChE in complex with GAL (pdb
id: 4EY6, R=2.15 ꢀ)[66] was used as a template for docking
calculations by GOLD v. 5.1.[67] The molecular docking pro-
tocol was optimized previously in several steps using two
training sets.[52,68] At each step the correlation between the
docking scores and pIC50 (ÀlogIC50) values of the training
compounds was evaluated. The highest correlation was
achieved at the following settings: scoring function
ChemPLP,[69] flexible binding site, flexible ligand, radius of
the binding site 10 ꢀ and one water molecule (HOH846)
kept within the binding site (toggle on and spin). Accord-
ing to the X-ray structure of the complex GAL – rhAChE,[66]
only one water molecule was left in the binding gorge,
making a hydrogen bond network between GAL, Ser203,
Gly121 and Gly122. Ten amino acids from the binding site
situated in a close proximity to the binding ligands were
of the H and 13C NMR spectra was made on the basis of
1
DEPT, COSY and HSQC experiments. The H and 13C NMR
spectra of the synthesized compounds are given as Supple-
mentary data. LCÀMS analyses were performed using a Q
Exactive Plus Orbitrap Mass spectrometer (Thermo Fisher
Scientific, Bremen, Germany), equipped with an electro-
spray (ESI) probe. The spectra were recorded on a positive
mode using a MS Full Scan mode.
2.2.2 General Procedure for Synthesis of Building Blocks
To a solution of corresponding acid (1 equiv) in CH2Cl2 was
added N-[3-(dimethylamino) propyl]-N-ethylcarbodiimide
(1 equiv), 1-Hydroxybenzotriazole (1 equiv) and the appro-
priate amine (1 equiv). The mixture was stirred at r.t. and
the product formation was monitored by TLC. The reaction
ꢁ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Mol. Inf. 0000, 00, 1 – 9
&2
&
ÞÞ
These are not the final page numbers!

Full Paper
www.molinf.com
was quenched with water, and the product was extracted
with CH2Cl2. The combined organic extracts were dried
over Na2SO4 and the solvent was evaporated under re-
duced pressure. The amides were purified by flash-column
chromatography.
tration were used and data were processed using the
GraphPad Prism 5.0 software.2 Heading 1st Order.
3 Results and Discussion
2.2.3 General Procedure for Preparation of Galantamine
Derivatives
3.1 Design of GAL Derivatives with Dual-site Binding
Five series (1–5) with alkylamide spacers and different aro-
matic fragments were designed (Figure 1). Each series con-
sisted of four members with different length of the alkyla-
mide spacers. Members a contained a butylamide linker,
members b – a pentylamide, members c – a hexylamide,
and members d – a heptylamide. The aromatic fragment in
the compounds of series 1 was phenyl, in series 2 – benzyl,
in series 3 – phenethyl, in series 4 – phenylglycine methyl
ester, and in series 5 – phenylalanine methyl ester.
Appropriate bromo-amide (1.1 equiv) and anhydrous K2CO3
(3 equiv) were added to a solution of norgalanthamine
(1 equiv) in anhydrous acetonitrile (2 mL) under argon at-
mosphere. After stirring at 608C for 24 h, the solvent was
removed in vacuo. The residue was directly subjected to
purification by flash column chromatography on silica gel
(CH2Cl2/MeOH/NH4OH) to give desired products.
2.3 Assessment of AChE Inhibitory Activity
AChE activity was assayed as described by Ellman et al.[72]
with some modifications.[73] Fifty mL of Electrophorus electri-
cus AChE (Sigma-Aldrich) in buffer phosphate (pH 7.6) and
50 mL of the tested compounds (0.2–300 mM in methanol)
were dissolved in 700 mL in the same buffer. The mixtures
were incubated for 30 minutes at room temperature before
the addition of 100 mL of the substrate solution (0.5 M
DTNB, 0.6 mM ATCI in buffer, pH 7.6). The absorbance was
read in Shimadzu spectrophotometer at 405 nm after three
minutes. Enzyme activity was calculated as a percentage
compared to an assay using a buffer without any inhibitor
using nonlinear regression. IC50 values are meansÆSD of
three individual determinations each performed in tripli-
cate.
3.2 Molecular Docking of the Designed Derivatives and
AChE Inhibitory Activity Prediction
The 20 newly designed compounds were docked into the
rhAChE (pdb code: 4EY6)[66] by GOLD v. 5.1[67] using a previ-
ously optimized docking protocol[52,68] as described in Meth-
ods. The only difference in the present protocol was the
scoring function used. Here, we used the novel ChemPLP
function in GOLD which is 5 times faster than GoldScore
and recent validation tests have shown it to be generally
more effective than the other scoring functions for both
2.4 Assessment of Neurotoxicity on NEURO-2A Cells
Murine neuroblastoma NEURO-2A cells (German collection
DSMZ in Braunschweig, Germany) were cultivated under
standard conditions: complete medium (90% DMEM, 10%
heat inactivated FBS and 1x non-essential aminoacids);
378C and 5% CO2 in fully humidified atmosphere. The cell
line was kept in the logarithmic growth phase by splitting
1:4 once a week using trypsin/EDTA. About 30% of the
cells grow like neuronal cells. For the experimental evalua-
tion of the cytotoxicicty NEURO-2A cells were plated in 96-
well flat bottomed cell culture plates at the recommended
density of 1ꢂ106 cells/25 cm2. Twenty four hours later cells
were treated with various concentrations of the investiga-
tional compounds. After 72 h incubation a MTT dye reduc-
tion assay was performed.[74] Briefly, at the end of incuba-
tion a MTT stock solution (10 mg/ml in PBS) was added
(10 ml/well). Plates were further incubated at 378C for 4 h.
Next, the formazan crystals were dissolved by the addition
of 110 ml/well 5% formic acid in 2-propanol (v/v). Absorp-
tion was measured at 580 nm wave length on an automat-
ed ELISA reader Labexim LMR1. At least 6 wells per concen-
Figure 1. GAL derivatives with dual-site binding, designed in the
present study. The underlined structures were synthesized and
tested for AChE inhibitory activity.
ꢁ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Mol. Inf. 0000, 00, 1 – 9
&3
&
ÞÞ
These are not the final page numbers!

Full Paper
www.molinf.com
pose prediction and virtual screening.[56] The pKa values of
the
Lipinski’s rule is fulfilled for the novel compounds with the
only exception of molecular mass which is slightly higher
than 500 for some of the compounds. The BBB permeabili-
ty was predicted by two algorithms (SVM and LiCABEDS70)
applied on four types of fingerprints and eight different
predictions were generated. Series 1a–d, 2a–d and 3a–d
were predicted as BBB permeable (BBB+) by all eight
methods, while series 4a–d and 5a–d were predicted as
BBB non-permeable (BBBÀ) by some of the methods
(Table 1). The presence of methyl ester group might affect
negatively the BBB permeability of compounds.
As series 3 and 5 were among the highest scored com-
pounds without and with methyl carboxylate fragment, re-
spectively, they were selected in full for synthesis. One or
two compounds were selected randomly from the other
three series.
novel compounds range from 7.52 to 7.93, i.e. at physio-
logical pH part of the molecules are protonated. The X-ray
structures of the complexes between Torpedo californica
AChE and three GAL derivatives show that both axial and
equatorial orientations are possible to bind inside the
gorge.[49,50] We modeled four sets of compounds for dock-
ing: equatorial protonated, equatorial non-protonated, axial
protonated and axial non-protonated. The average scores
of the highest scoring poses from 3 runs with 10 poses
each of all docked compounds from each set were as fol-
lows: 103.24, 105.92, 101.72 and 103.02, respectively. The
scores were very close and intercorrelated (data not
shown). No robust conclusion about the preferred orienta-
tion of the alkylamide spacers could be drawn. The scores
of the equatorial protonated set are given in Table 1.
The ChemPLP scores reveal a clear relationship: score in-
creases when the length of alkylamide spacer increases.
This relationship was expected because longer chains
mean more interactions along the AChE gorge. The molec-
ular weights vary from 434.53 (1a) to 562.70 (5d) (Table 1).
The logP values are between 2.80 (2a) and 4.24 (3d). All
compounds contain 2 hydrogen bond donors. The com-
pounds with and without methyl carboxylate fragment
have 8 and 6 hydrogen-bond acceptors, respectively. The
3.3 Synthesis of the Best Predicted GAL Derivatives
The target molecules were synthesised in two steps: prepa-
ration of bromo-amides and their subsequent reaction with
N-demethylated galantamine.
The first step was performed by reacting the commercial-
ly available 4-bromobutanoic acid, 5-bromopentanoic acid,
6-bromohexanoic acid, or 7-bromoheptanoic acid with ani-
line, benzylamine, phenethylamine, phenylglycine methyl
Table 1. ChemPLP scores, experimental IC50 and pIC50 values, physicochemical properties, predicted BBB permeability and neurotoxicity of
the newly designed GAL derivatives. The underlined compounds were synthesized and tested for AChE inhibitory activity.
Comp. ChemPLP IC50 mM (exp)
Times more active pIC50
Mw
logP H-bond
donors
H-bond ac- BBB permeabili- Neuro toxicity
score
EeAChE*
than GAL
(exp)
ceptors
ty****
IC50 mM
1a
1b
1c
94.45
97.67
98.38
434.53 3.14
448.55 3.44
462.58 3.67
2
2
2
2
2
2
2
2
2
2
2
2
6
6
6
6
6
6
6
6
6
6
6
6
8
8
8
8
8
8
8
8
8
8
8
8
1d 100.87
2a
2b
2c 101.61
2d 104.84
3a 100.55
3b 101.79
3c 106.09
3d 109.94
0.0169Æ0.0054
0.0308Æ0.0025
63
35
7.772 476.61 4.20
448.55 2.80
7.512 462.58 3.10
476.61 3.33
490.63 3.86
462.58 3.18
7.512 476.61 3.48
7.679 490.63 3.71
9.097 504.66 4.24
>50
93.69
99.82
>50
0.0308Æ0.0026
0.0210Æ0.0016
0.0008Æ
35
51
1338
>50
>50
>50
0.00003
4a 103.16
4b 105.50
4c 107.52
4d 111.86
506.59 2.99
520.62 3.29
2
2
2
2
2
2
2
2
1
8
8
8
8
8
8
8
8
4
6
5
7
7
7
6
7
7
8
0.0527Æ0.0025
0.0211Æ0.0027
0.0958Æ0.0017
0.0264Æ0.0023
0.0246Æ0.0016
20
51
11
40
43
7.278 534.64 3.52
7.677 548.67 4.05
7.019 520.62 3.14
7.578 534.64 3.44
7.609 548.67 3.67
8.974 562.70 4.20
5.971 287.35** 1.75
>50
>50
>50
>50
>50
>50
>50
5a
97.53
5b 104.38
5c 108.24
5d 116.95
0.0011Æ0.0007 1008
GAL
HBr
r***
74.56
1.070Æ0.0738
1
0.828
* Recalculated at GAL IC50 =1.07 mM. ** Molecular mass of GAL base. *** Correlation coefficient between ChemPLP scores and pIC50 values.
**** Predictions made by BBB Predictor (http://www.cbligand.org/BBB/). The integer shows the number of models giving positive predic-
tion (BBB+).
ꢁ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Mol. Inf. 0000, 00, 1 – 9
&4
&
ÞÞ
These are not the final page numbers!

Full Paper
www.molinf.com
tassium carbonate as a base led to the set of target com-
pounds 1–5 (Scheme 2).
Unfortunately, all attempts to prepare the galantamine
derivative 3a were unsuccessful. The problem appeared
due to the unstable starting bromo-amide 8a under the re-
action conditions. The low yield of 5a was explained also
with the low stability of its precursor 10a when heated
under basic conditions. All other substances were isolated
in moderate to good yields after flash column chromatog-
raphy and were fully characterized. Details about the syn-
thesized compounds are given in Supporting Information
1
1. The H and 13C NMR spectra of the synthesized com-
pounds are given in Supporting Information 2.
3.4 Assessment of AChE Inhibitory Activity and
Neurotoxicity
The inhibitory potential of the novel GAL derivatives
against Electrophorus electricus (electric eel) AChE was
tested according to the methodology developed by Ellman
et al.[72] with some modifications.[73] The UniProt alignment
of rhAChE (P22303) and EeAChe (O42275) shows that all 17
amino acids forming the binding gorges are identical. The
aligned structures are given as Supporting Information 3.
On this basis, we use the predictions made on rhAChE to
assess experimentally the anti-AChE activity of novel com-
pounds on EeAChE. GAL was used as a positive control and
the enzyme activity was calculated at IC50 =1.07 mM for
GAL. The IC50 values (mM) of the novel compounds are
shown in Table 1. The relationship between the ChemPLP
scores and the length of linkers was confirmed by the ex-
perimental IC50 values. The IC50 values decrease gradually
from members a to members d in both 3 and 5 series.
Compounds 3d and 5d demonstrated the highest acetyl-
cholinesterase inhibitory activities among the novel deriva-
tives, i.e. 0.8 nM and 1.1 nM, respectively. They are 1338
and 1008 times, respectively, more active than GAL. The
rest compounds showed AChE inhibitory activities between
9 and 63 times higher than this of GAL (IC50 =1.070 mM).
Good correlation (r=0.828) was found between the
ChemPLP scores and the pIC50 (ÀlogIC50) values of the
novel GAL derivatives. All compounds were non-toxic on
NEURO-2A cells. The IC50 values exceeded 50 mM as estimat-
ed by MTT-dye reduction assay.
Scheme 1. Synthesis of the bromo-amide intermediates.
ester, or phenylalanine methyl ester. The chosen combina-
tions were performed, using the reagents for peptide syn-
thesis 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC)
and 1-Hydroxybenzotriazole hydrate (HOBT), resulting into
the desired bromo-amides 6–10 (Scheme 1).
Amides 8a and 10a were unstable under the reaction
conditions and storage at room temperature. Hence, they
were prepared by shortening the reaction times, and after
quick purification on silica gel were stored at À108C.
The galantamine was demethylated following a known
procedure.[75] The reaction of norgalantamine with the ini-
tially prepared bromides 6–10 in dry acetonitrile, and po-
The highest scored poses of 3d and 5d are visualized in
Figure 2. They are superposed with AChE – Ab complex,
derived previously by RosettaDock.[68,76] It is seen that both
novel GAL derivatives fill well the binding gorge. The ab-
sence of methyl carboxylate fragment in 3d allows the
phenyl ring to adopt a conformation sterically hindered for
Ab, while the presence of this group in 5d prevents from
adopting such conformation. Additionally, the terminal
phenyl ring in 3d takes part in a sandwich-type p-p interac-
tion with Tyr72 and His287 (Figure 3). In this sense, the GAL
derivatives with flexible heptylamide spacer ending with
Scheme 2. Synthesis of the target compounds.
ꢁ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Mol. Inf. 0000, 00, 1 – 9
&5
&
ÞÞ
These are not the final page numbers!

Full Paper
www.molinf.com
proximity to the aromatic fragment affects unfavourably
the binding conformation.
Conflict of Interest
None declared.
Supporting Information
Supporting Information 1: Data for the newly synthesized
compounds in the present study.
Supporting Information 2: H and 13C NMR spectra of the
synthesized compounds 1–5.
Supporting Information 3: Multiple sequence alignment
of hAChE (UniProt P22303) and EeAChE (UniProt O42275)
using CLUSTAL O (1.2.1).
1
Figure 2. The highest scored poses of 3d (magenta) and 5d (blue)
superposed with AChE – Ab complex. Ab is given as yellow ribbon.
The solvent accessible surface of AChE is given in grey.
Acknowledgements
This work was supported by the Medical Science Council,
Medical University, Sofia (Grants 1-S/2014 and 8-S/2015).
We are thankful to the Operational Programme “Devepop-
ment of the Competitiveness of the Bulgarian Economy”,
European Regional Development Fund (Contract
No BG161PO003-1.2.04-0007-C0001) for the JASCO P-2010
polarimeter. Authors are thankful for the excellent technical
assistance of Mrs. T. Athanassova, who performed the cell
culture experiments.
References
[1] C. J. Lasner, J. M. J. Lee, Neurophathol. Exp. Neurol. 1998, 57,
719–731.
[2] N. C. Inestrosa, A. Alvarez, C. A. Perez, R. D. Moreno, M.
Vicente, C. Linker, O. I. Casanueva, C. Soto, J. Garrido, Neuron
1996, 16, 881–891.
Figure 3. The terminal phenyl ring in 3d (blue) takes part in a sand-
wich-type p-p interaction with Tyr72 and His287. The solvent ac-
cessible surface of AChE is given in grey.
[3] M. P. Mattson, Nature 2004, 430, 631–639.
phenethyl fragment are perspective AChE inhibitors with
dual-site binding.
[4] M. C. Dinamarca, J. P. Sagal, R. A. Quintanilla, J. A. Godoy, M. S.
Arrazola, N. C. Inestrosa, Mol. Neurodegener. 2010, 5, 4.
[5] M. Harel, I. Schalk, L. Ehret-Sabatier, F. Bouet, M. Goeldner, C.
Hirth, P. H. Axelsen, I. Silman, J. L. Sussman, Proc. Natl. Acad.
Sci. USA 1993, 90, 9031–9035.
[6] A. Ordentlich, D. Barak, C. Kronman, Y. Flashner, M. Leitner, Y.
Segall, N. Ariel, S. Cohen, B. Velan, A. J. Shafferman, Biol. Chem.
1993, 268, 17083–17095.
[7] A. Ordentlich, D. Barak, C. Kronman, N. Ariel, Y. Segall, B. Velan,
A. Shafferman, J. Biol. Chem. 1998, 273, 19509–19517.
[8] Z. Radic, N. A. Pickering, D. C. Vellom, S. Camp, P. Taylor, Bio-
chemistry 1993, 32, 12074–12084.
[9] J. L. Sussman, M. Harel, F. Frolow, C. Oefner, A. Goldman, L.
Toker, I. Silman, Science 1991, 253, 872–879.
[10] R. J. Kitz, L. M. Braswell, S. Ginsburg, Mol. Pharmacol. 1970, 6,
108–121.
4 Conclusions
The molecular docking-based predictions are a useful tool
in the design of novel GAL derivatives with dual-site bind-
ing fragments. The synthesized and tested novel com-
pounds confirm the predictions. The heptylamide spacer is
long enough to bridge the GAL moiety bound in the cata-
lytic site and the aromatic fragments interacting with PAS.
Among the tested terminal aromatic fragments, the phe-
nethyl substituent is the most suitable for binding in PAS.
The presence of a methyl carboxylate group in a close
[11] G. V. De Ferrari, M. A. Canales, I. Shin, L. M. Weiner, I. Silman,
N. C. Inestrosa, Biochemistry 2001, 40, 10447–10457.
ꢁ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Mol. Inf. 0000, 00, 1 – 9
&6
&
ÞÞ
These are not the final page numbers!

Full Paper
[12] G. Johnson, S. W. Moore, Biochem. Biophys. Res. Commun.
www.molinf.com
Hrabinova, O. Soukup, D. Jun, K. Kuca, Molecules 2015, 20,
22084–22101.
2004, 319, 448–455.
[13] G. Johnson, S. W. Moore, Curr. Pharm. Des. 2006, 12, 217–225.
[14] Y.-P. Pang, P. Quiram, T. Jelacic, F. Hong, S. J. Brimijoin, Biol.
Chem. 1996, 271, 23646–23649.
[15] L. Piazzi, A. Rampa, A. Bisi, S. Gobbi, F. Belluti, A. Cavalli, M.
Bartolini, V. Andrisano, P. Valenti, M. Recanatini, J. Med. Chem.
2003, 46, 2279–2282.
[34] W. J. Geldenhuys, C. J. Van der Schyt, Curr. Med. Chem. 2013,
20, 1662–1672.
[35] E. Viayna, R. Sabate, D. Munoz-Torrero, Curr. Topics Med. Chem.
2013, 13, 1820–1842.
[36] M. Rosini, E. Simoni, A. Minarini, C. Melchiorre, Neurochem.
Res. 2014, 39, 1914–1923.
[16] A. Cavalli, M. L. Bolognesi, S. Capsoni, V. Andrisano, M. Bartoli-
ni, E. Margotti, A. Cattaneo, M. Recanatini, C. Melchiorre,
Angew. Chem., Int. Ed. 2007, 46, 3689–3692.
[17] M. L. Bolognesi, M. Rosini, V. Andrisano, M. Bartolini, A. Minari-
ni, V. Tumiatti, C. Melchiorre, Curr. Pharm. Des. 2009, 15, 601–
613.
[37] M. Singh, M. Kaur, N. Chadha, O. Silakari, Mol. Divers. 2016, 20,
271–297.
[38] W. Parys, Alzheimer’s Reports 1998, 53, S19–S20.
[39] M. R. Farlow, Clin. Ther. 2001, 23, A13–A24.
[40] J. Marco-Contelles, C. Rodrꢄguez, M. C. Carreiras, M. Villarroya,
A. G. Garcꢄa, Chem. Rev. 2006, 106, 116–133.
[18] M. L. Bolognesi, A. Cavalli, C. Melchiorre, Neurotherapeutics
2009, 6, 152–162.
[19] M. L. Bolognesi, G. Chiriano, M. Bartolini, F. Mancini, G. Botte-
goni, V. Maestri, S. Czvitkovich, M. Windisch, A. Cavalli, A. Min-
arini, M. Rosini, V. Tumiatti, V. Andrisano, C. Melchiorre, J. Med.
Chem. 2011, 54, 8299–8304.
[20] V. Capurro, P. Busquet, J. P. Lopes, R. Bertorelli, G. Tarozzo,
M. L. Bolognesi, D. Piomelli, A. Reggiani, A. Cavalli, PLoS One
2013, 8, e56870.
[21] M. Hernꢃndez-Rodrꢄguez, J. Correa-Basurto, F. Martꢄnez-Ramos,
I. I. Padilla-Martꢄnez, C. G. Benꢄtez-Cardoza, E. Mera-Jimꢅnez,
M. C. Rosales-Hernꢃndez, J. Alzheimer’s Dis. 2014, 41, 1073–
1085.
[41] E. F. Pereira, S. Reinhardt-Maelicke, A. Schrattenholz, A. Mae-
licke, E. X. Albuquerque, J. Pharmacol. Exp. Ther. 1993, 265,
1474–1491.
[42] A. Storch, A. Schrattenholz, J. C. Cooper, E. M. Abdel Ghani, O.
Gutbrod, K. H. Weber, S. Reinhardt, C. Lobron, B. Hermsen, V.
Soskic, E. F. R. Pereira, E. X. Albuquerque, C. Methfessel, A.
Maelicke, Eur. J. Pharm. 1995, 290, 207–219.
[43] A. Schrattenholz, E. F. Pereira, U. Roth, K. H. Weber, E. X. Albu-
querque, A. Maelicke, Mol. Pharmacol. 1996, 49, 1–6.
[44] E. X. Albuquerque, M. Alkondon, E. F. Pereira, C. F. Castro, A.
Schrattenholz, C. T. Barbosa, R. Bonfante-Canarcas, Y. Aracava,
H. M. Eisenberg, A. Maelicke, J. Pharmacol. Exp. Ther. 1997,
280, 1117–1136.
´
[22] N. Guzior, M. Bajda, M. Skrok, K. Kurpiewska, K. Lewinski, B.
[45] F. A. Dajas-Bailador, K. Heimala, S. Wonnacott, Mol. Pharmacol.
2003, 64, 1217–1226.
ˇ
Brus, A. Pislar, J. Kos, S. Gobec, B. Malawska, Eur. J. Med. Chem.
2015, 92, 738–749.
[23] P. Meena, V. Nemaysh, M. Khatri, A. Manral, P. M. Luthra, M.
Tiwari, Bioorg. Med. Chem. 2015, 23, 1135–1148.
[24] P. Camps, X. Formosa, C. Galdeano, D. Munoz-Torrero, L. Ram-
irez, E. Gomez, N. Isambert, R. Lavilla, A; Badia, M. V. Clos, M.
Bartolini, F. Mancini, V. Andrisano, M. P. Arce, M. I. Rodrigues-
Franco, O. Huertas, T. Dafni, F. J. Luque, J. Med. Chem. 2009,
52, 5365–5379.
[46] K. Takata, Y. Kitamura, M. Saeki, M. Terada, S. Kagitani, R. Kita-
mura, Y. Fujikawa, A. Maelicke, H. Tomimoto, T. Taniguchi, S.
Shimohama, J. Biol. Chem. 2010, 285, 40180–40191.
[47] A. Mary, D. Z. Renko, C. Guillou, C. Thal, Bioorg. Med. Chem.
1998, 6, 1835–1850.
[48] C. Guillou, A. Mary, D. Z. Renko, E. Gras, C. Thal, Bioorg. Med.
Chem. Lett. 2000, 10, 637–639.
[49] H. M. Greenblatt, C. Guillou, D. Guenard, A. Argaman, S. Botti,
B. Badet, C. Thal, I. Silman, J. L. Sussman, J. Am. Chem. Soc.
2004, 126, 15405–15411.
[25] L. Yu, R. Cao, W. Yi, Q. Yan, Z. Chen, L. Ma, W. Peng, H. Song,
Bioorg. Med. Chem. Lett. 2010, 20, 3254–3258.
[26] S. Rizzo, M. Bartolini, L. Ceccarini, L. Piazzi, S. Gobbi, A. Cavalli,
M. Recanatini, V. Andrisano, A. Rampa, Bioorg. Med. Chem.
2010, 18, 1749–1760.
[27] R. Leon, J. Marco-Contelles, Curr. Med. Chem. 2011, 18, 552–
576.
[50] C. Bartolucci, L. A. Haller, U. Jordis, G. Fels, D. Lamba, J. Med.
Chem. 2010, 53, 745–751.
[51] E. Simoni, S. Daniele, G. Bottegoni, D. Pizzirani, M. L. Trincavelli,
L. Goldoni, G. Tarozzo, A. Reggiani, C. Martini, D. Piomelli, C.
Melchiorre, M. Rosini, A. Cavalli, J. Med. Chem. 2012, 55, 9708–
9721.
[52] M. Atanasova, G. Stavrakov, I. Philipova, D. Zheleva, N. Yorda-
nov, I. Doytchinova, Bioorg. Med. Chem. 2015, 23, 5382–5389.
[53] J. J. Jansen, E. J. Martin, Curr. Opin. Chem. Biol. 2004, 8, 359–
364.
[54] A. N. Jain, Curr. Opin. Drug Discov. Devel. 2004, 7, 396–403.
[55] K. V. Kishan, Curr. Protein Pept. Sci. 2007, 8, 376–380.
[56] B. O. Villoutreix, N. Renaut, D. Lagorce, O. Sperandio, M.
Montes, M. A. Miteva, Curr. Protein Pept. Sci. 2007, 8, 381–411.
[57] Y. Fukunishi, Comb. Chem. High Troughput Screen 2009, 12,
397–408.
[58] M. E. Sobhia, R. Singh, P. Kare, S. Chavan, Expert Opin. Drug
Discov. 2010, 5, 543–557.
[59] J. Kirchmair, S. Distinto, K. R. Liedl, P. Markt, J. M. Rollinger, D.
Schuster, G. M. Spitzer, G. Wolber, Infect Disord Drug Targets
2011, 11, 64–93.
[60] L. Scotti, F. J. Bezerra Mendonca Jr., D. R. Magalhaes Moreira,
M. S. da Silva, I. R. Pitta, M. T. Scotti, Curr. Top. Med. Chem.
2012, 12, 2785–2809.
[28] M. I. Fernꢃndez-Bachiller, C. Pꢅrez, L. Monjas, J. Rademann,
M. I. Rodrꢄguez-Franco, J. Med. Chem. 2012, 55, 1303–1317.
[29] E. Simoni, S. Daniele, G. Bottegoni, D. Pizzirani, M. L. Trincavelli,
L. Goldoni, G. Tarozzo, A. Reggiani, C. Martini, D. Piomelli, C.
Melchiorre, M. Rosini, A. Cavalli, J. Med. Chem. 2012, 55, 9708–
9721.
[30] E. Nepovimova, E. Uliassi, J. Korabechy, L. E. Pena-Altamira, S.
Samez, A. Pesaresi, G. E. Garcia, M. Bartolini, V. Andrisano, C.
Bergamini, R. Fato, D. Lamba, M. Roberti, K. Kuca, B. Monti,
M. L. Bolognesi, J. Med. Chem. 2014, 57, 8576–8589.
[31] O. M. Bautista-Aguilera, G. Esteban, M. Chioua, K. Nikolic, D.
Agbaba, I. Moraleda, I. Iriepa, E. Soriano, A. Samadi, M. Unzeta,
J. Marco-Contelles, Drug Des. Devel. Ther. 2014, 8, 1893–1910.
[32] L. Wang, G. Esteban, M. Ojima, O. M. Bautista-Aguilera, T. Ino-
kuchi, I. Moraleda, I. Iriepa, A. Samadi, M. B. Youdim, A.
Romero, E. Soriano, R. Herrero, A. P. Fernꢃndez, M. Ricardo-
Martꢄnez, J. Marco-Contelles, M. Unzeta, Eur. J. Med. Chem.
2014, 80, 543–561.
[33] J. Korabecny, M. Andrs, E. Nepovimova, R. Dolezal, K. Babkova,
A. Horova, D. Malinak, E. Mezeiova, L. Gorecki, V. Sepsova, M.
[61] A. K. Debnath, Methods Mol. Biol. 2013, 993, 185–204.
ꢁ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Mol. Inf. 0000, 00, 1 – 9
&7
&
ÞÞ
These are not the final page numbers!

Full Paper
[62] M. Safavi, M. Baeeri, M. Abdollahi, Expert Opin. Drug Discov.
2013, 8, 733–751.
[63] V. Kumar, S. Chandra, M. I. Siddiqi, Curr. Pharm. Des. 2014, 20,
3488–3499.
[64] H. Tomioka, Curr. Pharm. Des. 2014, 20, 4305–4306.
[65] N. I. Siddiqi, M. I. Siddiqi, Curr. Pharm. Des. 2014, 20, 4418–
4426.
[66] J. Cheung, M. J. Rudolph, F. Burshteyn, M. S. Cassidy, E. N. Gary,
J. Love, M. C. Franklin, J. J. Height, J. Med. Chem. 2012, 55,
10282–10286.
[67] G. Jones, P. Willett, R. C. Glen, A. R. Leach, R. J. Taylor, Mol. Biol.
1997, 267, 727–748.
[68] M. Atanasova, N. Yordanov, I. Dimitrov, S. Berkov, I. Doytchino-
va, Mol. Inf. 2015, 34, 394–403.
www.molinf.com
[71] Y. H. Zhao, M. H. Abraham, A. Ibrahim, P. V. Fish, S. Cole, M. L.
Lewis, M. J. de Groot, D. P. Reynolds, J. Chem. Inf. Model. 2007,
47, 170–175.
[72] G. L. Ellman, K. D. Courtney, V. Andreas Jr., R. M. Featherstone,
Biochem. Pharmacol. 1961, 7, 88–95.
[73] J. Ortiz, S. Berkov, N. Pigni, C. Theoduloz, G. Roitman, A. Tapia,
J. Bastida, G. E. Feresin, Molecules 2012, 17, 13473–13482.
[74] G. Momekov, D. Ferdinandov, A. Bakalova, M. Zaharieva, S.
Konstantinov, M. Karaivanova, Arch. Toxicol. 2006, 80, 555–
560.
[75] A. Mary, D. Z. Renko, C. Guillou, C. Thal, Tetrahedron Lett. 1997,
38, 5151–5152.
[76] S. Lyskov, J. J. Gray, Nucleic Acids Research 2008, 36 (Web
Server Issue), W233–W238.
[69] O. Korb, T. Stꢆtzle, T. E. Exner, Swarm. Intell. 2007,1, 115–134.
[70] X. Q. Xie, J. Z. Chen, E. Billings, Proteins 2003, S3, 307–319.
Received: March 20, 2016
Accepted: May 17, 2016
Published online: && &&, 0000
ꢁ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Mol. Inf. 0000, 00, 1 – 9
&8
&
ÞÞ
These are not the final page numbers!

Full Paper
www.molinf.com
FULL PAPERS
G. Stavrakov, I. Philipova, D. Zheleva,
M. Atanasova, S. Konstantinov,
I. Doytchinova*
&& – &&
Docking-based Design of
Galantamine Derivatives with Dual-
site Binding to Acetylcholinesterase
ꢁ 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Mol. Inf. 0000, 00, 1 – 9
&9
&
ÞÞ
These are not the final page numbers!