The in vitro protective effects of the three novel nanomolar reversible inhibitors of human cholinesterases against irreversible inhibition by organophosphorous chemical warfare agents

Article abstract of DOI:10.1016/j.cbi.2019.06.027

Acetylcholinesterase (AChE) is an enzyme which terminates the cholinergic neurotransmission, by hydrolyzing acetylcholine at the nerve and nerve-muscle junctions. The reversible inhibition of AChE was suggested as the pre-treatment option of the intoxicat

Full text of DOI:10.1016/j.cbi.2019.06.027

Chemico-Biological Interactions 309 (2019) 108714
Contents lists available at ScienceDirect
Chemico-Biological Interactions
journal homepage: www.elsevier.com/locate/chembioint
The in vitro protective effects of the three novel nanomolar reversible
inhibitors of human cholinesterases against irreversible inhibition by
organophosphorous chemical warfare agents
a,*
b
c
a
Maja D. Vitorović-Todorović , Franz Worek , Andrej Perdih , Sonja Đ. Bauk ,
d
c,e
Tamara B. Vujatović , Ilija N. Cvijetić
a
b
c
d
e
Military Technical Institute, Ratka Resanovića 1, Belgrade, Serbia
Bundeswehr Institute of Pharmacology and Toxicology, Neuherbergstrasse 11, Munich, Germany
National Institute of Chemistry, Hajdrihova 19, Ljubljana, Slovenia
Faculty of Chemistry, University of Belgrade, Studentski Trg 12-16, Serbia
Innovation Center of The Faculty of Chemistry, University of Belgrade, Studentski Trg 12-16, Serbia
A R T I C L E I N F O
A B S T R A C T
Keywords:
Acetylcholinesterase (AChE) is an enzyme which terminates the cholinergic neurotransmission, by hydrolyzing
acetylcholine at the nerve and nerve-muscle junctions. The reversible inhibition of AChE was suggested as the
pre-treatment option of the intoxications caused by nerve agents. Based on our derived 3D-QSAR model for the
reversible AChE inhibitors, we designed and synthesized three novel compounds 8-10, joining the tacrine and
aroylacrylic acid phenylamide moieties, with a longer methylene chain to target two distinct, toplogically se-
parated anionic areas on the AChE. The targeted compounds exerted low nanomolar to subnanomolar potency
toward the E. eel and human AChE's as well as the human BChE and showed mixed inhibition type in kinetic
studies. All compounds were able to slow down the irreversible inhibition of the human AChE by several nerve
agents including tabun, soman and VX, with the estimated protective indices around 5, indicating a valuable
level of protection. Putative noncovalent interactions of the selected ligand 10 with AChE active site gorge were
finally explored by molecular dynamics simulation suggesting a formation of the salt bridge between the pro-
tonated linker amino group and the negatively charged Asp74 carboxylate side chain as a significant player for
the successful molecular recognition in line with the design strategy. The designed compounds may represent a
new class of promising leads for the development of more effective pre-treatment options.
Acetylcholinesterase
Irreversible inhibition
Nerve agents
Protection
Pre-treatment
Molecular dynamics simulations
1
. Introduction
Acetylcholinesterase (AChE, EC 3.1.1.7.) is a carboxylesterase
symptoms, an oxime, which is able, to some extent, to reactivate the
irreversibly inhibited AChE and to restore its activity and an antic-
onvulsant drug (diazepam or its pro-drug avizafone) which is able to
block nerve agent induced seizures. However, such treatment is not
fully effective under life-threatening conditions [1]. The reasons for the
ineffectiveness are diverse and multifaceted. Firstly, in order to be ef-
fective, the treatment has to be administered as soon as possible after
exposure has happened, which can be rather complicated in the bat-
tlefield conditions. In the past several decades, numerous oximes have
been synthesized and only two of them, pralidoxime (2-PAM) and
obidoxime, are licensed and commercially available. Oximes are con-
sidered to be insufficiently efficient in case of soman poisoning, mainly
due to rapid aging (i.e.dealkylation) of the nerve agent bound to the
enzyme which prevents its reactivation. Also, tabun-inhibited AChE is
highly resistant to reactivation [2,3]. Moreover, the two oximes, being
which terminates cholinergic neurotransmission, by hydrolyzing the
neurotransmitter acetyl-choline (ACh) in a synaptic cleft of nerve- and
nerve-muscle junctions. Organophosphonates (OP), such as nerve
agents used in terrorism and chemical warfare and organophosphates
which are widely used as pesticides rapidly inactivate AChE by binding
to the catalytic Ser203 leading to accumulation of ACh in the synaptic
cleft and causing muscarinic (miosis, hypersalivation, bradycardia,
diarrhea, bronchoconstriction), nicotinic (convulsions, muscle cramps
and muscle dysfunction) and central signs and symptoms of intoxica-
tion, ultimately causing death by respiratory failure.
The currently approved treatment for the OP intoxication consists of
anticholinergic drug, such as atropine, which relieves the muscarinic
*
Corresponding author.
E-mail address: mvitod@chem.bg.ac.rs (M.D. Vitorović-Todorović).
https://doi.org/10.1016/j.cbi.2019.06.027
Received 12 February 2019; Received in revised form 10 June 2019; Accepted 13 June 2019
Available online 19 June 2019
0009-2797/ © 2019 Elsevier B.V. All rights reserved.

M.D. Vitorović-Todorović, et al.
Chemico-Biological Interactions 309 (2019) 108714
a permanently positively charged species hardly cross blood-brain
barrier (BBB), and reactivate CNS AChE to a low extent. Another
drawback of 2-PAM is that a product of reactivation, phosphyl-oxime
can readily re-inhibit enzyme [4,5].
Given the drawbacks of the above described standard treatment, the
so-called ‘pre-treatment' option was proposed. The pre-treatment is
given to healthy individuals when there is a higher probability for
chemical attack (which represents dilemma by itself). The pre-treat-
ment usually consists of one or more following drugs: a pseudoirre-
versible, carbamate-type AChE inhibitor (pyridostigmine or physos-
tigmine) and an anti-cholinergic drug (caramiphen, aprophen or
scopolamine). The role of pseudoirreversible AChE inhibitor in the pre-
treatment mixture is to temporarily inhibit fraction of the enzyme and
protect it from irreversible inhibition by OP's. This might be especially
important in case of soman intoxications where reactivation of AChE is
impossible. One of the first carbamates tested for the pre-treatment is
pyridostigmine. Several studies on experimental animals showed that
pre-treatment with pyridostigmine or pyridostigmine/scopolamine
mixture improves efficacy of the antidotal treatment after exposure to
sarin or soman. However, this pre-treatment does not provide any
protection when given alone, i.e.without any antidotal post-treatment
Fig. 1. Outline of the design strategy of „dual binding”compounds 8–10:
Compounds join the tacrine unit which strongly binds to the anionic site (AS) of
AChE with the aroylacrylic acid phenylamide fragment which is a non-com-
petitive low micromolar inhibitor of the enzyme with a flexible linker.
and the second peripheral anionic site (PAS) which is located at the rim
of the active site gorge.
In this work, based on the literature data as well as the information
obtained by the analysis of the most important outcomes from our
QSAR models, we designed a set of dual binding compounds 8–10
which joined the tacrine and aroylacrylic acid amides fragments via an
eight methylene linker as depicted in Fig. 1, to target two distinct to-
plogically separated anionic areas on the AChE. Aroylacrylic acid amide
fragment was included, because we previously confirmed its low mi-
cromolar activity and non-competitive mode of binding [18,19], and
tacrine was included since it showed slightly better protective effects in
vivo against the paraoxone intoxications than pyridostigmine [16]. The
derived 3D-QSAR models suggested low nanomolar inhibitory activity
of the designed compounds toward AChE. We synthesized such dual
inhibitors, estimated their inhibitory activity toward AChE and other
available cholinesterases and evaluated their in vitro protective effects
against the irreversible inhibition of AChE by several nerve agents in-
cluding tabun, soman and VX. Finally, the putative non-covalent in-
teractions between the most active derivative and AChE were in-
vestigated by molecular dynamics to validate the design strategy. Such
compounds could provide more effective pre-treatment options against
nerve agents induced intoxications.
[
6–9]. The major drawback of pyridostigmine lies in the fact that it is
permanently positively charged compound and it does not cross BBB, so
it is unable to protect AChE in CNS. Thus, pre-treatment with physos-
tigmine as an uncharged compound, able to cross BBB, was proposed
along with different anticholinergic drugs. In the past two decades,
several studies on experimental animals (Beagle dogs, rats and guinea
pigs) consistently and convincingly demonstrated that pre-treatment
with physostigmine when given in conjunction with anticholinergic
drug (atropine, aprophen, scopolamine, procyclidine, benactyzine, etc.)
via different administration routes (sustained release by mini-osmotic
pumps, i. v. single dose, or via transdermal route) is able to afford
considerable protection against multiple lethal doses of nerve agents
including sarin, soman and VX and that physostigmine is far more su-
perior as a pre-treatment comparing to pyridostigmine [8,10–13]. An-
other studies showed that huperzine A and galanthamine, the true re-
versible inhibitors of AChE, approved for the treatment of Alzheimer's
disease, are also able to exert protective effects against nerve agent
poisoning in experimental animals. Galanthamine was chosen for in vivo
testing not only due to its mere anticholinesterase activity but also due
to its additional advantages including BBB permeability, anticonvulsant
properties and ability to prevent neurodegeneration. Albuquerque et al.
showed that a single dose of galanthamine in combination with atro-
pine is able to protect guinea pigs from the exposure to lethal doses of
soman and sarin. Moreover, galanthamine was far more superior in
preventing lethality comparing to accepted pre-treatment, pyr-
idostigmine [14]. Similarly, it has been shown that huperzine A ad-
ministered as a single dose or s. c. via osmotic pumps protects experi-
mental animals (mice, non-human primates and guinea pigs) against
toxic signs and lethality of soman [15]. Petroianu et al. also investigated
and compared protection ability of eleven moderate to strong reversible
AChE inhibitors (pyridostigmine, physostigmine, ranitidine, tiapride,
tacrine, 7-metoxytacrine, amiloride, oxime K-27, metoclopramide and
methylene blue) against paraoxone exposure in Wistar rats [16]. Study
showed that the best protection from paraoxon-induced mortality was
observed after pre-treatment with physostigmine and the oxime K-27,
but pre-treatment with tacrine, pyridostigmine and ranitidine was also
significant. All other compounds were ineffective or even had un-
favorable effects.
2. Materials and methods
2.1. Chemistry
All chemicals were purchased from Sigma Aldrich or Merck, and
1
13
were used as received. The H and C NMR spectra were recorded in
CDCl on Varian Gemini 200/50 MHz or Bruker AVANCE 500/125 MHz
3
instruments. Chemical shifts are reported in parts per million (ppm)
relative to tetramethylsilane (TMS) as internal standard. Spin multi-
plicities are given as follows: s (singlet), d (doublet), t (triplet), m
(multiplet), or br (broad). The HR-ESI-MS spectra were recorded on
Agilent Technologies 6210-1210 TOF-LC-ESI-MS instrument in positive
mode. Samples were dissolved in MeOH. The detailed procedures for
the synthesis of aroylacrylic acid phenylamides are given elsewhere
[18,19].
Synthetic procedure for 9-chloro-1,2,3,4-tethydro-aminoacridine
(3): The 16 mL of POCl was carefully added to the mixture of an-
3
thranilic acid 1 (2.1 g, 15 mmol) and equimolar amount of cyclohex-
h
Recently we performed a 3D-QSAR analysis based on alignment
independent descriptors for the set of 110 structurally diverse AChE
inhibitors [17]. The study included the so-called ‘dual binding ligands’
or ‘dual inhibitors’ which consisted of two structural cyclic aromatic
fragments linked by a polymethylene chain. Such dual compounds were
envisioned to simultaneously bind to two distinct sites on the AChE
enzyme anionic site (AS), first one found deeper in the active site gorge
anone 2 in ice-bath. The mixture was heated under reflux for 2 , then
cooled at room temperature, and concentrated under reduced pressure
to give slurry. The residue was diluted with EtOAc, neutralized with ice-
cold aqueous K
2
CO , and washed with brine. The organic layer was
3
dried over anhydrous MgSO and concentrated in vacuum to provide
4
crude product, which was subsequently re-crystallized from acetone to
1
give a final compound. H NMR (200 MHz, CDCl
3
) δ: 1.83–1.96 (m, 4H,
2

M.D. Vitorović-Todorović, et al.
Chemico-Biological Interactions 309 (2019) 108714
cyclohexyl–CH
H, J = 6.35 Hz, cyclohexyl–CH
phenyl); 7.62 (t, 1H, J = 6,77 Hz, m-phenyl); 7.95 (t, 1H, J = 8.19 Hz,
2
–); 2.93 (t, 2H, J = 5.63 Hz, cyclohexyl–CH
2
–); 3.08 (t,
22.96; 23.55; 24.71; 26.82; 27.08; 29.26; 30.15; 31.69; 33.86; 34.21;
40.04; 48.37; 49.44; 59.45; 115.72; 119.24; 120.12; 122.82; 123.52;
124.04; 126.77; 128.27; 128.41; 128.52; 128.94; 134.12; 137.73; 147.31;
150.80; 155.21; 158.30; 171.90; 196.18. ESI-MS HR: 310.2046 (M +2),
Calc. 310.2045.
2
2
–); 7.47 (t, 1H, J = 7.06 Hz, m-
1
3
o-phenyl); 8.09 (t, 1H, J = 7.06 Hz, m-phenyl). C NMR (50 MHz,
CDCl ) δ: 12.42; 22.45; 27.28; 33.96; 123.46; 125.16; 126.17; 128.49;
3
1
29.05; 146.50; 159.26.
4-(3,4-Dimethylphenyl)-4-oxo-N-phenyl-2-[8-(1,2,3,4-tetra-
Synthetic procedure for N-(1,2,3,4-tetrahidroacridine-9-yl)-oc-
hydroacridin-9-ylamino)octyl-amino]-butanamide (10): C39
H
48
N
4
O ,
2
tane-1,8-diamine (4): The mixture of 3 (1 g, 4.61 mmol) and 1,8-dia-
reaction of (E)-4-(3,4-dimetylphenyl)-4-oxo-2-butenoic acid phenylamide
minooctane (1.99 g, 13.8 mmol) in 1-penthanol (5 mL) was refluxed for
(0.70 mmol) and equimolar amount of 4 gave 10 in quantitative yield as
h
1
1
8 at 160 °C. The mixture was cooled to room temperature and diluted
orange semi-solid. H NMR (500 MHz, CDCl
3
) δ: 1.25–1.37 (overlapped m,
with EtOAc (50 mL). The solution was washed with 10% NaOH aqueous
8H, linker–CH
1,90 (m, 4H, cyclohexyl–CH
2.69 (m, 4H, cyclohexyl–CH
1,2 = 8.70 Hz,
2
–); 1.50 (m, 2H, linker–CH
–); 2.29 (s, 3H, –CH
–); 3.06 (m, 4H, linker–CH
1,3 = 17.40 Hz, ABX); 3.47 (t, 2H, J = 7.15 Hz,
2
–); 1.64 (m, 2H, linker–CH
); 2.30 (s, 3H, –CH
–); 3.26 (dd, 1H,
2
–);
solution, twice with distilled water, and dried with anhydrous MgSO
4
.
2
3
3
);
After the filtration of the dried solution, the solvent was removed under
reduced pressure, and obtained semi-solid substance was purified by
2
2
J
J
silica gel column chromatography (CHCl
3
: MeOH: NH
4
OH = 7: 3:
amino–NH–); 3.62 (dd, 1H, J1,2 = 4.35 Hz, J1,3 = 17.39 Hz, ABX); 3.68
(dd, 1H, J1,2 = 4.35 Hz, J1,3 = 8.70 Hz, ABX); 7.09 (t, 1H, J = 7.14 Hz,
amido-p-phenyl); 7.13–7.25 (overlapped m, 3H, amido-m-phenyl, aroyl-m-
phenyl); 7.32 (m, 2H, amido-o-phenyl); 7.53 (t, 1H, J = 6.88 Hz, tacrine-
m-phenyl); 7.59 (d, 1H, J = 8.35 Hz, aroyl-o-phenyl); 7.70 (t, 1H,
J = 7.37 Hz, tacrine-m-phenyl); 7.74 (s, 1H, aroyl-o-phenyl); 7.91 (d, 1H,
1
0
.07). H NMR (200 MHz, CDCl
3
)
δ: 1.13–1.19 (overlapped m,
–); 1.47 (br, 2H, linker–CH –);
–); 2.52 (br, 2H, cyclohexyl–CH –); 2.92
–); 3.03 (br, 2H, amino–NH –); 3.30 (br, 2H,
–); 3.84 (br, 1H, amino–NH–); 7.18 (t, 1H, J = 7.14 Hz, m-
linker–CH
.74 (br, 2H, cyclohexyl–CH
br, 2H, linker–CH
linker–CH
2
–); 1.28 (br, 2H, linker–CH
2
2
1
2
2
(
2
2
2
phenyl); 7.38 (t, 1H, J = 7.52 Hz, m-phenyl); 7.78–7.82(overlapped m,
J = 8.35 Hz, tacrine-o-phenyl); 7.95 (d, 1H, J = 7.86 Hz, tacrine-o-
1
3
13
2
2
1
H, o-phenyl). C NMR (50 MHz, CDCl
3
) δ: 22.31; 22.58; 24.31; 26.36;
phenyl); 9.59 (s, 1H, amido-NH). C NMR (125 Hz, CDCl
3
) δ: 19.70;
8.80; 31.22; 32.66; 33.52; 41.39; 48.91; 115.26; 119.75; 122.42;
22.98; 127.67; 128.15; 146.99; 150.27; 157.87.
20.00; 21.39; 22.70; 22.99; 24.71; 26.77; 27.09; 29.27; 30.16; 31.69;
33.87; 40.12; 48.38; 49.44; 59.50; 115.70; 119.23; 120.11; 122.79;
123.54; 124.03; 125.23; 125.88; 128.57; 128.94; 129.25; 129.91; 134.17;
137.05; 137.76; 143.25; 147.27; 150.82; 158.25; 171.95; 198.39. ESI-MS
HR: 605.3830 (M +1), Calc. 605.3856; 303.1967 (M+2), Calc. 303.1967.
General synthetic procedure for the target compounds 8–10: To a
mixture of aroyl-substituted (E)-4-aryl-4-oxo-2-butenoic acid phenyla-
mide (7 mmol) in chloroform (15 mL), equimolar amount of 4 and
1
5 mL of toluene were added and the resulting mixture was stirred at
h
room temperature for 24 . The solvent was removed under reduced
pressure, and obtained semi-solid substance was purified by silica gel
2.2. Biological studies
column chromatography (CHCl
3
: MeOH: NH
4
OH = 7: 3: 0.07).
2.2.1. Reversible inhibition of E. Eel AChE
4
-N-Diphenyl-4-oxo-2-[8-(1,2,3,4-tetrahydroacridin-9-yla-
The inhibition potency of the compounds 8–10 toward E. Eel AChE
was evaluated by Ellman procedure [20], using the type VI-S enzyme
(Sigma) and acetylthiocholine iodide (0.28 mM) as a substrate. Broad
range of concentrations, which produce 20–80% of enzyme activity
inhibition, were used for each compound. The reaction took place in the
final volume of 2 mL of 0.1 M potassium phosphate buffer, pH 8.0,
containing 0.03 units of AChE and 0.3 mM 5,5-dithio-bis(2-ni-
trobenzoic)acid (DTNB), used to produce yellow anion of 5-thio-2-ni-
trobenzoic acid in reaction with thiocholine released by AChE. Tested
compound was added to the enzyme solution and preincubated at 25 °C
for 15 min, followed by the addition of DTNB (0.1 mL) and substrate
(0.05 mL). Determination of inhibition curves were performed at least
in triplicate. One triplicate sample without test compound was always
present to yield 100% of AChE activity. The reaction was monitored for
0.5 min (absorbance was measured every 10 s), and the color produc-
tion was measured at 412 nm. The reaction rates were compared, and
the percent of inhibition, due to the presence of test compounds, was
calculated. IC values were obtained by fitting the data into dose-re-
mino)octylamino]butanamide (8): C37
H
44
N
4
O , reaction of (E)-4-
2
phenyl-4-oxo-2-butenoic acid phenylamide (0.70 mmol) and equimolar
1
amount of 4 gave 8 in quantitative yield as orange semi-solid. H NMR
(
(
200 MHz, CDCl
3
) δ: 1.01–1.20 (overlapped m, 8H, linker–CH –); 1.50
2
br, 2H, linker–CH
–); 2.90 (br, 4H, cyclohexyl–CH
– and ABX); 3.56–3.61 (m, 1H, ABX); 4.34 (m,
2
–); 1.72 (br, 2H, linker–CH
2
–); 2.51 (br, 4H,
cyclohexyl–CH
2
2
–); 3.33–3.40 (over-
lapped m, 7H, linker–CH
2
1
H, ABX); 4.34 (m, 1H, ABX); 4.56 (s, amino–NH–); 4.89 (s,
amino–NH–); 6.89–7.56 (overlapped m, tacrine-m-phenyl, amido-p-
phenyl, amido-m-phenyl, aroyl-p-phenyl and aroyl-m-phenyl);
7
.79–7.86 (overlapped m, 4H, aroyl-o-phenyl, amido-o-phenyl and ta-
13
crine-o-phenyl); 9.54 (s, 1H, amido-NH). C NMR (50 Hz, CDCl
3
) δ:
2
3
1
1
1
1.18; 22.23; 22.60; 24.35; 26.55; 26.84; 20.97; 29.90; 30.55; 31.37;
2.96; 41.57; 41.97; 43.05; 59.10; 61.38; 64.33; 67.15; 114.93; 119.13;
19.60; 122.03; 123.43; 124.34; 125.05; 126.67; 127.91; 128.78;
30.17; 135.99; 137.05; 137.56; 141.38; 142.14; 141.28; 151.10;
57.38; 170.15; 171.80; 196.49; 198.37. ESI-MS HR: 577.3529 (M +1),
5
0
Calc. 577.3537; 289.1807 (M+2), Calc. 289.1805.
sponse curves (inhibitor concentration vs. velocity of enzyme reaction),
4
-(4-Isopropylphenyl)-4-oxo-N-phenyl-2-[8-(1,2,3,4-tetra-
hydroacridine-9-ylamino)octyl-amino]-butanamide (9): C40
reaction of (E)-4-(4-isopropylphenyl)-4-oxo-2-butenoic acid phenylamide
0.70 mmol) and equimolar amount of 4 gave 9 in quantitative yield as
according to formula:
H
50
N
4
2
O ,
vi
v0
1
=
h
1 + ([I]/IC50)
(
1
orange semi-solid. H NMR (500 MHz, CDCl
3
) δ: 1.23–1.25 (overlapped m,
); 1.31–1.37 (overlapped m, 8H, linker
–); 1.64 (m, 2H, linker–CH –); 1.90 (m,
–); 2.69 (m, 4H, cyclohexyl–CH –); 2.95 (h, 1H,
–); 3.28
where v
i
and v are initial velocities of the enzyme reaction in the ab-
o
8
H, d, 2H, J = 7.09 Hz, i-PrCH
CH -); 1.57 (m, 2H, linker–CH
H, cyclohexyl–CH
3
2
sence and in presence of inhibitor, [I] is inhibitor concentration and h is
the Hill coefficient.
–
2
2
4
2
2
From dose-response experiments we were able to determine Hill
coefficient and to estimate the possible cooperative effects during the
process of inhibitor binding.
J
1,2 = 6.59 Hz, J1,3 = 13.78 Hz, i-PrCH); 3.05 (br, 2H, linker–CH
2
(
dd, 1H, J1,2 = 8.19 Hz, J1,3 = 17.38 Hz, ABX); 3.47 (t, 2H, J = 7.39 Hz,
amino-NH); 3.62 (dd, 1H, J1,2 = 3.20 Hz, J1,3 = 17.39 Hz, ABX); 3.70 (dd,
The inhibition reaction was also monitored continuously during
15 min after the initiation of the reaction in order to determine the time
interval which is needed to achieve equilibrium of the reversible in-
hibition reaction.
1
H, J1,2 = 3.64 Hz, J1,3 = 8.71 Hz, ABX); 7.09 (t, 1H, J = 7.47 Hz, amido-
p-phenyl); 7.29–7.34 (overlapped m, 6H, amido-m-phenyl, amido-o-phenyl
and tacrine-m-phenyl); 7.58 (d, 2H, J = 7,64 Hz, aroyl-m-phenyl); 7.91 (d,
2
H, J = 8.28 Hz, tacrine-o-phenyl); 7.95 (d, 2H, J = 8.28 Hz, aroyl-o-
For the estimation of the inhibition type, Lineweaver-Burk plots
were generated by using the fixed amount of acetylcholinesterase and
1
3
phenyl); 9.60 (s, 1H, amido-NH). C NMR (125 Hz, CDCl ) δ: 22.70;
3
3

M.D. Vitorović-Todorović, et al.
Chemico-Biological Interactions 309 (2019) 108714
varying amounts of the substrate (0.097–0.582 mM), in the absence and
in the presence of different inhibitor concentrations. The re-plots of the
slopes and intercepts of the double reciprocal plots against inhibitor
concentrations gave the inhibitor constants (Ki1 and Ki2, for the binding
to free enzyme and enzyme substrate complex correspondingly) as the
intercepts on the x-axis.
with human AChE in the presence of the substrate ATCh: 10 μL native
AChE (the estimated AChE concentration in the cuvette was ~20 pM)
and 5 μL diluted OP (8 different concentrations; tabun: 150–500 nM,
soman: 10–70 nM, VX: 20–130 nM)) were added to pre-heated cuvettes,
containing 3000 μL phosphate buffer (0.1 M, pH 7.4), 100 μL DTNB and
50 μL ATCh (final volume 3165 μL). ATCh hydrolysis was continuously
monitored for 5 min. The inhibition kinetics were determined in the
absence or presence of compounds 8 (20 and 100 nM), 9 (30 and
100 nM) and 10 (1 and 5 nM). The recorded curves were analyzed by
non-linear and linear regression analysis according to Forsberg and Puu
2
.2.2. Reversible inhibition of native HuAChE and HuBChE
Acetylthiocholine iodide (ATCh), S-butyrylthiocholine iodide
(
BTCh), 5,5′-dithio-bis-2-nitrobenzoic acid (DTNB) and isolated human
butyrylcholinesterase (BChE) were purchased from Sigma Aldrich
in order to calculate the second order inhibition rate constants k
[23,24]. The protective index (PI) was calculated by dividing the k in
the absence by the k in the presence of compounds 8, 9 or 10.
i
(
Taufkirchen, Germany). All other chemicals were purchased from
i
Merck Eurolab GmbH (Darmstadt, Germany) at the purest grade
available. Processing of experimental data was performed by linear and
non-linear regression analysis using line and curve fitting programs
provided by GraphPad Prism 5.0 (GraphPad Software, San Diego, Calif.
USA). All experiments were carried out at a minimum of n = 2 and data
were expressed as mean ± SD.
i
2.2.4. Reactivation of sarin-inhibited AChE by HI-6 in the absence or
presence of compounds 8–10
HI-6 (1-[[[4-(aminocarbonyl)-pyridinio]-methoxy]methyl]-2-[(hy-
droxyimino)methyl]pyridinium dichloride monohydrate) was provided
by Dr. Clement (Defence Research Establishment Suffield, Ralston,
Alberta, Canada). Sarin-inhibited human AChE was incubated with
10 μM HI-6 in the absence or presence of compounds 8 (20 and
100 nM), 9 (30 and 100 nM) and 10 (1 and 5 nM). Aliquots were taken
after 2–60 min to determine AChE activities which were referred to the
activity of native AChE and the % reactivation was calculated. The
Haemoglobin-free human erythrocyte membranes (‘ghosts’) were
prepared from human whole blood and served as source of human er-
ythrocyte AChE [21]. Aliquots of erythrocyte ghosts were adjusted to
the original whole blood AChE activity and stored at −80 °C. Prior to
use, aliquots were thawed gently and homogenized on ice using a So-
noplus HD 2070 ultrasonic homogenator (Bandelin electronic, Berlin,
Germany), three times for 5 s with 30-s intervals to obtain a homo-
geneous matrix for kinetic analysis.
pseudo-first order reactivation rate constant k was calculated by non-
obs
linear regression analysis.
A modified Ellman procedure was used [20,22] to determine the
AChE and BChE activity spectrophotometrically (Cary 50 Bio UV/
Visible Spectrophotometer) at 412 nm using polystyrol cuvettes with
2.3. Molecular modelling study
0
.45 mMATCh (AChE) and 1 mMBTCh (BChE) as substrate and 0.3 mM
The initial coordinates of AChE enzyme were taken from the PDB
DTNB as chromogen in 0.1 M phosphate buffer (pH 7.4). All measure-
ments were performed at least in duplicate at 37 °C and pH 7.4. The
IC50 was measured at a fixed substrate concentration using different
compound concentrations (0.1–1000 μM). In brief, 10 μL erythrocyte
ghosts or BChE and 3.2 μL compound were added to a cuvette prefilled
with phosphate buffer and DTNB. After adding ATCh or BTCh, the
enzyme activity was continuously monitored for up to 3 min and IC50
values were calculated from semi-logarithmic dose-response curves of
the ligand concentration versus AChE/BChE activity.
database (entry 4EY4) [25]. The missing residues were added using
SwissPDB web server [26]. All co-crystallized ligands, ions and water
molecules were removed. The 3D conformation of ligand 10 used for
molecular docking was generated using MMFF94 force field [27] and
reoptimized using semiempirical PM7 method [28]. The protonation
state of ligand and protein were set at pH 7.4, as predicted by PROPKA
[29]. The binding site was defined as all residues in the 13 Å sphere
around Tyr340, comprising both deep-pocket active site residue Trp86
and peripheral site residue Trp286, as well as other residues important
for binding of AChE inhibitors. AutoDockVina [30] and PLANTS [31]
were used for docking the ligand into the AChE active site. Auto-
DockVina exhaustiveness parameter was set to 250 while other para-
meters were default. Vega ZZ 3.1.0 [32] was used as GUI for all docking
calculations.
The time dependence of inhibition was determined by incubating
human AChE and BChE with compounds 8, 9 or 10 (500 nM) and the
AChE/BChE activity was determined after 1, 7 and 15 min. The rever-
sibility of inhibition of human AChE and BChE was tested by incubating
enzyme solutions with appropriate compound concentrations (AChE:
3
00 nM compounds 8 and 9, 10 nM compound 10; BChE: 100 nM
The best docking solution of compound 10 from the AutoDockVina
calculation was further used for the molecular dynamics simulation
using NAMD MD engine. CHARMM-GUI [33] was utilized for the pre-
paration and solvation of the protein-ligand complex, adding an octa-
hedral cluster of TIP3 water molecules at 10 Å from the edge. The
solvated system was neutralized by placing the 7 potassium ions using
Monte Carlo method. CHARMM36 m [34] force field parameters were
used, while the atomic types and partial charges for ligand were as-
signed using CHARMM General Force Field (CGenFF) [35](see Fig. S1
for assigned ligand 10 parameters). The final system consisted of 48334
atoms and the necessary NAMD simulation files were generated using
CHARMM-GUI. All MD calculations were performed on the GPU cluster
at the National Institute of Chemistry, Ljubljana.
compounds 8 and 9, 50 nM compound 10) for 5 min, followed by ex-
tensive dilution (300 fold) and determination of AChE/BChE activity.
All enzyme activities were referred to untreated control activities.
2
.2.3. Tabun, soman and VX inhibition kinetics with human AChE in the
absence and presence of compounds 8–10
The organophosphorus compounds (OP) tabun (O-ethyl N,N-di-
methylphosphoroamidocyanidate), soman (O-pinacolyl methylpho-
sphonofluoridate) and VX (O-ethyl S-2-diisopropylamino-ethylmethyl
1
31
phosphonothiolate) (> 95% by GC-MS, H NMR and P NMR) were
made available by the German Ministry of Defence. OP stock solutions
(
0.1% v/v) were prepared in acetonitrile, stored at 20 °C and appro-
priately diluted in distilled water just before use. Stock and working
solutions of compounds 8–10 were prepared in methanol.
NAMD 2.11 engine [36] was used for the MD calculations using the
following settings. All 1–3 and modified 1–4 interactions were ac-
counted by specifying “exclude scaled 1-4” keyword. Periodic boundary
conditions were specified based on the size of the full solvated system.
Initial minimization of the system was performed to remove the steric
bumps followed by a 500 ps MD equilibration with a 2 fs step using the
Nose-Hoover Langevin piston pressure of 1 atm at the 303.15 K. Elec-
trostatics was treated with the Particle mesh Ewald (PME) algorithm,
with the cut-off distance for the non-bonded interactions set to 12 Å,
The effect of the tested compounds on the time-dependent inhibi-
tion of human AChE by tabun, soman and VX was tested by mixing the
native human AChE with compounds 8 (100 nM), 9 (100 nM) and 10
(
(
5 nM), immediately followed by the addition of tabun (50 nM), soman
10 nM) or VX (10 nM). After 1–20 min, aliquots were taken and the
residual AChE activity was determined (n = 2).
The inhibition kinetics of tabun, soman and VX was determined
4

M.D. Vitorović-Todorović, et al.
Chemico-Biological Interactions 309 (2019) 108714
switch distance at 10 Å, and storing all pairs within 16 Å (pairlist-
dist = 16). After equilibration, 100 ns of production through un-
constrained, Langevin MD simulation was performed using a 2 fs in-
tegration step.
HuAChE and HuBChE inhibition are provided in Fig. 3. All three com-
pounds are highly potent low nanomolar inhibitors of all three choli-
nesterases. The more potent inhibitors are compounds which have no
substituents on the aroyl-phenyl ring (8) or have small substituents
such as methyl groups on the aroyl-phenyl ring (10). Compound 9, with
a more voluminous isopropyl group has slightly higher IC value to-
For the MD trajectory analysis, 1000 equidistant frames were ex-
tracted from the trajectory, and aligned using the RMSD Trajectory Tool
in VMD [37]. Interaction energy analysis was performed using NAMD
Interaction energy tool available in VMD. The RMSD, RMSF, gyration
radius, distances and other geometric parameters were explored using
VMD and VEGA ZZ.
5
0
ward both electric eel and human AChE. It has been proven in a number
of previous studies that dual AChE inhibitors bind in the way in which
the tacrine substructure is oriented toward the bottom of the active site
gorge, interacting with Trp86 and Tyr337 AS residues, while the other,
usually aromatic and polycyclic fragment of the molecule is oriented
toward the entrance of the active site gorge and interacts with amino
acid residues that belong to PAS namely Trp286, among others.
The finding that compound 9 exerts slightly lower inhibition po-
tency, compared to compounds 8 and 10 was unexpected, since it is
well known that PAS area of the enzyme is wide and can accommodate
a variety of highly voluminous molecular fragments. However, it is
possible that aroylacrylic acid amide fragment of 9, bearing a 4-i-Pr
substituent, experience some steric hindrance upon binding to PAS and
therfore tacrine moiety is unable to fully reach an AS of AChE, which
results in slightly lower potency.
To further provide a more comprehensive outlook of the observed
dynamical interaction pattern between the simulated ligand 10 and the
AChE binding site, 1000 exported MD frames were used in the dynamic
pharmacophore analysis using DynophoreApp framework developed in
the group of Prof. Gerhard Wolber with default settings [38,39]. Dy-
nophores were subsequently visualized within the LigandScout pro-
gram [40]. These calculations yielding a dynophore model and corre-
sponding output files were performed on the computers of the
Molecular Design Lab at the Freie Universität Berlin, Germany and
subsequently visualized in LigandScout.
Subsequently, we tested the velocity of reaching equilibrium for the
reversible inhibition, by following the residual enzyme activity in
longer period of time (15 min) after the initiation of the inhibition re-
action. The equilibrium is reached almost instantly and residual enzyme
activity was constant during the remaining reaction time (results not
shown). The inhibition of HuAChE and HuBChE was also fully reversible
as tested by the dilution experiments. Therefore, we can state that three
synthesized novel compounds (8, 9 and 10) represent true, fast and
highly potent reversible inhibitors of both AChE and BChE.The in-
spection of the Hill's coefficients (Table 1), showed that all three
compounds have the value of coefficients around 1.0, which indicates
almost no cooperativity in the ligand binding to the four subunits of the
E. eel AChE tetramer.
3
. Results and discussion
3.1. Chemistry
Synthetic path to compounds 8–10 is given in Fig. 2. The Nie-
mentowski reaction between 2-aminobenzoic (anthranilic) acid and
cyclohexanone proceeded smoothly to give 3. The presence of 9-chloro
substituent on the molecule enabled further functionalization of the
compounds. The nucleophilic aromatic substitution of 9-chloro sub-
stituent of 3, by using triple molar amount of linker (1,8-diami-
nooctane) gave exclusively monomeric product 4. Michael's addition of
4
on correspondingly substituted aroylacrylic acid amide gave target
1
compounds 8–10. Synthesized compounds were characterized by
H
1
3
We also determined the inhibition type for the compounds 9 and 10
by following the enzyme reaction at several different substrate con-
centrations, for the two fixed inhibitor concentrations. The inhibition
constants and types are shown in Table 2, along with the corresponding
double-reciprocal Lineweaver-Burk plots in Fig. 4. From the inspection
of the Lineweaver-Burk plots compounds 9 and 10 are noncompetitive
inhibitors and bind both to the free enzyme and the enzyme inter-
mediate formed during substrate hydrolysis, which is evident from the
intersection of the lines in the upper left quadrant.
and C NMR spectroscopy and by high resolution mass spectrometry
1
13
with electrospray ionization (ESI-HR-MS). The H and C NMR spectra
of 8–10 are given in Supplementary material (Figs. S2–S7).
3.2. Biological studies
3
.2.1. Inhibition potency of 8–10 toward EeAChE, HuAChE and HuBChE
Inhibition potency of compounds 8–10 determined toward EeAChE,
HuAChE and HuBChE is given in Table 1 and dose-response curves for
3
.2.2. Tabun, soman and VX inhibition kinetics with the human AChE in the
absence and presence of compounds 8–10. Protective effects of 8-10 against
irreversible inhibition by nerve agents
We examined the inhibition kinetics with the three nerve agents
tabun, soman and VX in the absence and in the presence of compounds
8
–10, in order to calculate the protective index (PI) of reversible in-
hibitors against irreversible inhibition by the nerve agents. Two con-
centrations of 8–10 were used and the results are given in Fig. 5 (left
panel).
The experiments revealed that under the experimental conditions
applied for the determination of the second order inhibition rate con-
stants the compounds were able to protect AChE from inhibition by
nerve agents partially at high compound concentrations, i.e. 100 nM of
compounds 8 and 9 and 5 nM of compound 10, resulting in a PI of up to
6
(Fig. 5 left panel). At lower compound concentrations (i.e. 20 nM
compound 8, 30 nM compound 9, 1 nM compound 10), being slightly
above the respective IC50 values, the effect was only marginal. It was
also interesting to note that different compounds, although quite si-
milar in structure and the mode of reversible inhibition, provided dif-
ferent protection against different organophosphates.
Fig. 2. Synthetic path to compounds 8 (R
= 3,4-diMe).
1
= H) 9 (R = 4-i-Pr) and 10
1
The protective index (PI) was calculated by dividing the second
(
R
1
order rate constant k
i
in the absence by the k in the presence of
i
5

M.D. Vitorović-Todorović, et al.
Chemico-Biological Interactions 309 (2019) 108714
Table 1
Inhibition potency of compounds 8–10 toward EeAChE, HuAChE and HuBChE.
No.
EeAChE
HuAChE
HuBChE
R1
IC50
±
SEM (nM)
Hill's coeff.
IC50
±
SD (nM)
IC50 ± SD (nM)
8
9
-H
7.76 ± 0.34
15.61 ± 1.12
5.24 ± 0.41
1.09 ± 0.08
1.05 ± 0.08
1.24 ± 0.16
18.7 ± 0.2
27.1 ± 0.1
0.66 ± 0.01
4.4 ± 0.1
8.0 ± 0.01
2.0 ± 0.01
-4-i-Pr
−3,4-diMe
1
0
Fig. 3. Concentration-dependent inhibition of the human AChE and BChE by compounds 8–10. Fourteen different compound concentrations ranging from 0.1 to
000 μM were tested. Data were presented as % AChE activity and expressed as mean values ± SD of two independent experiments.
1
Table 2
inhibitors (galanthamine and huperzine A) with known ADME-Tox
properties. However, in vitro data for the protective indices for these
compounds are not published.
The inhibition constants for the binding to free enzyme (Ki1) and enzyme in-
termediate formed during substrate hydrolysis (Ki2) for the 9 and 10, and
corresponding inhibition types.
Various studies investigated reversible AChE inhibitors in vitro, but
for those compounds in vivo data on animal models are scarce. For in-
stance, Radić et al. [41] investigated the effects of several peripheral
site reversible AChE inhibitors (gallamine, D-tubocurarine, propidium,
atropine, coumarine derivatives) on rates of irreversible AChE inhibi-
tion by several organophosphate compounds, including paraoxon, ha-
loxon, DDVP and neutral and cationic alkylmethylphosphonylthioates.
Interestingly, most of the peripheral site reversible inhibitors actually
increased the rates of the irreversible inhibition for about 6-fold in
submilimolar concentrations while at higher concentration decreased
the constants of irreversible inhibition but in the best cases to the level
of the constants without reversible inhibitor present. Therefore, neither
of these peripheral site ligands affords any protection in vitro. Petroianu
et al. [42–44] investigated the effects of several moderately potent re-
versible AChE inhibitors including metoclopramide, ranitidine and L-
lactate on the rates of irreversible inhibition by organophosphate pes-
ticides paraoxon and mipafox. The protective effects were estimated in
the terms of IC50 shift for paraoxon in the presence and in the absence
of AChE reversible inhibitors, which are not comparable to our data for
PI. Although these reversible inhibitors provided some level of in vitro
Comp
Ki1 (nM)
K
i2 (nM)
Inh. Type
9
7.45
4.46
9.26
9.18
noncompetitive
noncompetitive
1
0
compounds 8, 9 or 10. According to the experimental protocol, the k
i
values were calculated from the inhibition curves recorded for 5 min.
Hence, it was tempting to test whether the test compounds are able to
reduce the inhibition of human AChE by the nerve agents over a pro-
longed time (Fig. 5 right panel). In fact, compounds 8 and 9 (100 nM)
were able to slow down the inhibition by tabun (50 nM), soman
(
10 nM) and VX (10 nM) and to preserve a small portion of AChE from
the irreversible inhibition after 20 min. In contrast, compound 10
(
5 nM) failed to preserve AChE from phosphylation.
Based on the current available literature data, it is difficult to esti-
mate whether the achieved protective effects would be significant in
vivo. The most abundant in vivo data are those on animal models which
investigated protective effects of FDA approved AChE reversible
Fig. 4. Left: Lineweaver-Burk plot of AChE (0.03
U) in the absence and in the presence of dif-
ferent concentration of 9: blue squares - no in-
hibitor, red dots – 6.3 nM and green triangles –
1
2.6 nM. Right: Lineweaver-Burk plot of AChE
(
0.03 U) in the absence and in the presence of
different concentration of 10: blue squares - no
inhibitor, red dots – 3.97 nM and green triangles
–
7.94 nM.
6

M.D. Vitorović-Todorović, et al.
Chemico-Biological Interactions 309 (2019) 108714
Fig. 5. Left panel: Protective index (PI) of com-
pounds 8–10 for the inhibition of human AChE
by tabun, soman and VX. The inhibition kinetics
of tabun (150–500 nM), soman (10–70 nM) and
VX (20–130 nM) was determined with human
AChE in the presence of the substrate ATCh and
the absence and presence of compounds 8–10.
The protective index (PI) was calculated by di-
viding the k
i
in the absence by the k in the
i
presence of compounds 8, 9 or 10. Yellow col-
umns: 20 nM compound 8, 30 nM compound 9,
1
1
1
nM compound 10; red hatched columns:
00 nM compounds 8 and 9, 5 nM of compound
0. Right panel: Effect of compounds 8 (yellow
squares; 100 nM), 9 (green dots; 100 nM) and 10
(
blue romboids; 5 nM) on the time dependent
inhibition of human AChE by tabun (50 nM),
soman (10 nM) and VX (10 nM).
protection, only ranitidine was tested in vivo [45]. The 30 min admin-
istration of ranitidine before exposure to paraoxon increased survival of
male Wistar rats, but the effects were far better with pyridostigmine.
Taking all of above into consideration, it might be concluded that
compounds 8–10 showed moderate level of protection of AChE against
irreversible inhibition by nerve agents and might be a promising leads
for the development of more effective protective agents.
interaction between the ligand structural motifs and AChE residues was
calculated. Compound 10 was simulated with the protonated distal
linker –NH- group which is near the aroylphenylamide fragment of the
molecule, as estimated by the PROPKA program. The proximal linker
–NH- group, as well as the tacrine pyridine nitrogen, were estimated as
being unprotonated and thus were modelled as uncharged.
The gyration radius of the whole protein was found to be in the
range between 20.9544 and 23.2942 with an average value of
22.8296 ± 0.2636 and remained stable during the MD simulation,
indicating a stable folding of the protein structure (Fig. S8, Supple-
mentary material). After the initial conformational change observed
during the first 10 ns of production stage, the protein's conformation
was stable as indicated by RMSD of protein's backbone (Fig. S9, Sup-
plementary material). The movie, obtained from the trajectory showing
a representative behavior of the protein ligand-system is also provided
as Supplementary material, Video 1.
The effects of compounds 8–10 on the sarin-inhibited AChE re-
activation by oxime HI-6 were also tested (results not shown). All three
compounds had neither beneficial nor detrimental effects on enzyme
reactivation. The reason for the lack of the effect of impairment i.e.
decrease in the reactivation reaction is probably due to a high con-
centration of HI-6 (10 μM) and low concentration of compounds 8, 9
and 10 that were used in the assay.
3.3. Molecular modelling of the inhibitor 10-AChE molecular recognition
Supplementary video related to this article can be found at https://
doi.org/10.1016/j.cbi.2019.06.027
After firmly establishing the AChE inhibition of the synthesized
The exported frames of the MD trajectory were then analyzed using
the dynamic pharmacophore-Dynophore approach [38,39]. This pro-
vided a broader, more general overview opposed to the usual MD
analysis which is typically directed to analyzing selected distances of
the observed intermolecular interaction pattern during the MD trajec-
tory. With dynophores we were able to evaluate the importance, oc-
currence and spatial evolution of the chemical interactions in terms of
derivatives, molecular dynamics simulations (MD) provided further
valuable insights into the possible molecular recognition between the
most active derivative, compound 10 and AChE active site gorge re-
sidues. Thus, the selected docked binding mode of compound 10 was
subsequently simulated in the 100 ns long MD simulation, using NAMD
suite of programs. Subsequently, 1000 equidistant MD frames were
exported and analyzed, and the percentage of the occurrence of each
7

M.D. Vitorović-Todorović, et al.
Chemico-Biological Interactions 309 (2019) 108714
Fig. 6. (a) The observed intermolecular interactions between the compound 10 (grey sticks) and AChE. Ligand molecular surface is indicated. The carbon atoms of
the important interacting residues are colored blue, while electrostatically bonded Asp74 is colored as red sticks; (b) Schematic representation of the results of the
dynophore analysis for binding of the protonated ligand 10 to the AChE. The percentage of occurrence of each pharmacophore feature (H-hydrophobic, HBD-
hydrogen bond donating, and PI- positive ionizable) is shown, along with theAChE active site residues forming each interaction; (c) Determined dynophore model
showing the interaction pharmacophore features of compound 10 throughout MD simulation. Hydrophobic interactions are shown as yellow, HBD as green and PI as
red spheres.
pharmacophore elements associated with a certain interaction while
concurrently identifying the involved interaction residues [46]. The
most important observed intermolecular interactions between the
compound 10 and AChE residues, as a result of dynophore analysis are
given in Fig. 6.
from the Trp286 indole ring toward the hydrophobic pocket formed by
Val365, Ala343 and Val294 residues. Such observed behaviours con-
firmed the importance of the postdocking analysis of the suggested
binding modes to provide deeper insight in the molecular recognition
[47]. Hydrophobic interactions between these residues and 3,4-di-
methylphenyl ring were stable during the whole trajectory with the
occurrence above 80%. The ionic interaction, the salt bridge between
the protonated linker –NH- group and negatively charged Asp74 car-
boxylate side chain was also identified. Although Asp74 is part of the
flexible “Ω-loop” of AChE, this interaction is present during the whole
simulation with the high occurrence of 99.6% (Figs. 6b and 7a).
However, two temporary shifts (increase) in the distance have been
At the beginning of the MD simulation, the obtained ligand 10
docked conformation stretched between the two main sites in the AChE
gorge: the anionic site (AS), found almost at the bottom of the gorge
and comprising Trp86 residue, and the peripheral anionic site (PAS),
found at the rim of the gorge and comprising Trp286 residue. At this
starting docking position, tacrine moiety and the phenylamide ring of
the aroylacrylic acid phenylamide moiety were found near the Trp86
indole ring. However, during the first 5 ns of the simulation, the tacrine
fragment of the ligand protruded more deeply into the active site of the
enzyme and occupied the hydrophobic pocket formed by Ile451,
Val132, Met85 and Leu130 residues (Fig. 6a and b).
th
nd
observed around 17 and 92 ns of simulation.
We also determined the interaction energy between the simulated
ligand 10 and AChE and the energy graph is shown in Fig. 7a. The
overall interaction energy remained stable with the electrostatic com-
ponent predominantly connected with the Asp74 - charged amino
linker having a bigger contribution that the van der Waals interactions.
Furthermore, the sharp drop in electrostatic energy nicely geome-
trically corresponds to the temporary shift of Asp74 away from the
positively charged linker –NH- group. It also confirms that this ionic
interaction indeed provides a significant contribution to the total en-
ergy of the complex formation. The non-polar interaction contribution
was still considerable owing to the many hydrophobic interactions
observed between the compound 10 and the surrounding residues
(Fig. 6b). These interactions remained stable throughout the trajectory.
The dynophore analysis revealed that hydrophobic interactions of
the tacrine phenyl ring with Met85 with the occurrence of 80.5% and
Val132 with occurrence 59.7% are the most prominent and most stable
during the simulation. Next, the proximal linker –NH- group forms a
strong hydrogen bond with the backbone oxygen of Met85 (Fig. 7c) and
this interaction remained stable and present during the whole trajec-
tory. Interestingly, no hydrophobic interactions were observed for the
polymethylene chain linker.
When analyzing the distal part of the molecule, i.e. aroylacrylic acid
phenylamide part, several important interactions were detected.
Phenylamide ring is situated near the anionic site of the enzyme and
forms hydrophobic interactions most of the MD simulation (76.6%)
with several residues including: Ala343, Trp86, Tyr77, Leu 76 and
Thr75. The stacking interaction between phenylamide ring and Trp86
indole ring was also observed (Fig. 7b). Dimethylaroyl ring of the distal
part of the molecule, in the first 5 ns of the simulations shifted away
4
. Conclusions
In the present study, by utilizing previously derived 3D-QSAR model
we designed and synthesized three novel AChE reversible “dual
binding” inhibitors 8-10 joining the tacrine and aroylacrylic acid
8

M.D. Vitorović-Todorović, et al.
Chemico-Biological Interactions 309 (2019) 108714
Fig. 7. Selected geometric and energetic parameters of the MD simulation: (a) Interaction energy between 10 and AChE during the 100 ns MD simulation.
Electrostatic component of the total energy correlates with the distance between Asp74 and positively ionized N atom of linker. The average distance between Asp74
and 10 is 3.05 ± 0.81 Å. (b) Observed stacking interaction between aniline ring of 10 and indole ring of Trp86. (c) Hydrogen bond interaction between the linker
NH proximal to tacrine moiety and backbone oxygen from Met85. The average distance is 2.50 ± 0.55 Å.
phenylamide structural fragments with a flexible linker. In our design
strategy these two moieties targeted two distinct AChE interaction sites
which, although being the most potent subnanomolar inhibitor, pro-
vided only a marginal protection. Based on the results obtained and
current available literature data, it still remains an open question
whether this level of protection observed in vitro would be significant
also in vivo. However, taking into consideration that there are some
reversible inhibitors which actually increase the rates of AChE phos-
phylation [41] and that even moderately effective AChE reversible in-
hibitors such as ranitidine [16] improve survival of the experimental
animals, it may be concluded that compounds 8–10 might be a pro-
mising leads in development of more effective protective agents.
The 100 ns long MD simulation initiated to study the compound 10
proposed binding mode in the AChE revealed that compound's moieties
engage in several stable hydrophobic interactions, and both main
moieties provide stable interactions with the targeted AS and PAS sites.
Furthermore, a hydrogen bond between the amino group next to the
-
anionic site deep inside the active and the peripheral anionic site lo-
cated outside of the main binding gorge on the enzyme. All compounds
displayed potent low nanomolar to subnanomolar inhibition potency
toward several cholinesterase's EeAChE, HuAChE and HuBChE which
was in accordance with the developed 3D-QSAR model.
Compounds 8-10 were found to bind reversibly and non-competi-
tively to AChE and showed moderate level of protection of HuAChE
against the irreversible inhibition by nerve agents tabun, soman and
VX, although at higher concentrations in respect to their IC50 values.
Compounds showed different levels of protection in relation to different
nerve agents tested with compound 10 having the best protection
against tabun irreversible inhibition, while compounds 8 and 9 were far
superior against VX irreversible inhibition, comparing to compound 10,
9

M.D. Vitorović-Todorović, et al.
Chemico-Biological Interactions 309 (2019) 108714
tacrine moiety and Met85 side chain oxygen and the ionic interaction
between the protonated linker amino group and negatively charged
Asp74 carboxylate side chainalso play a role in molecular recognition.
The later ionic interaction provides one of the core contributions to the
total interaction energy. The designed compounds thus represent pro-
mising leads for further development of more effective pre-treatment
options against nerve agents.
[12] Y. Meshulam, G. Cohen, S. Chapman, D. Alkalai, A. Levy, Prophylaxis against or-
ganophosphate poisoning by sustained release of scopolamine and physostigmine,
J. Appl. Toxicol. An Int. J. 21 (2001) S75–S78.
[
13] W.-S. Kim, Y. Cho, J.-C. Kim, Z.-Z. Huang, S.-H. Park, E.-K. Choi, S. Shin, S.-Y. Nam,
J.-K. Kang, S.-Y. Hwang, Protection by a transdermal patch containing physos-
tigmine and procyclidine of soman poisoning in dogs, Eur. J. Pharmacol. 525
(
2005) 135–142.
[
14] E.X. Albuquerque, E.F.R. Pereira, Y. Aracava, W.P. Fawcett, M. Oliveira,
W.R. Randall, T.A. Hamilton, R.K. Kan, J.A. Romano, M. Adler, Effective counter-
measure against poisoning by organophosphorus insecticides and nerve agents,
Proc. Natl. Acad. Sci. U.S.A. 103 (2006) 13220–13225.
Conflicts of interest
[
[
15] G. Lallement, V. Baille, D. Baubichon, P. Carpentier, J.-M. Collombet, P. Filliat,
A. Foquin, E. Four, C. Masqueliez, G. Testylier, Review of the value of huperzine as
pretreatment of organophosphate poisoning, Neurotoxicology 23 (2002) 1–5.
16] G.A. Petroianu, S.M. Nurulain, M. Shafiullah, M.Y. Hasan, K. Kuča, D.E. Lorke,
Usefulness of administration of non‐organophosphate cholinesterase inhibitors be-
fore acute exposure to organophosphates: assessment using paraoxon, J. Appl.
Toxicol. 33 (2013) 894–900.
The authors declare that they have no conflict of interest.
Ethical standards
[
[
17] M.D. Vitorović-Todorović, I.N. Cvijetić, I.O. Juranić, B.J. Drakulić, The 3D-QSAR
study of 110 diverse, dual binding, acetylcholinesterase inhibitors based on align-
ment independent descriptors (GRIND-2). The effects of conformation on predictive
power and interpretability of the models, J. Mol. Graph. Model. 38 (2012) 194–210.
18] M.D. Vitorović-Todorović, I.O. Juranić, L.M. Mandić, B.J. Drakulić, 4-Aryl-4-oxo-N-
phenyl-2-aminylbutyramides as acetyl-and butyrylcholinesterase inhibitors.
Preparation, anticholinesterase activity, docking study, and 3D structure–activity
relationship based on molecular interaction fields, Bioorg. Med. Chem. 18 (2010)
The manuscript does not contain clinical studies or patient data.
Acknowledgements
This work was supported by the Serbian Ministry of Education
Science and Technological Development (grant OI172035), the Serbian
Ministry of Defence, and by the German Ministry of Defence. AP was
supported by the Ministry of Higher Education, Science and Technology
of the Republic of Slovenia through Grant P1-0012. The authors are
grateful to G. Duman for expert technical assistance and prof. Gerhard
Wolber for providing us access to dynophore calculations at the Freie
Universität Berlin, Germany.
1
181–1193.
[
19] M.D. Vitorović-Todorović, C. Koukoulitsa, I.O. Juranić, L.M. Mandić, B.J. Drakulić,
Structural modifications of 4-aryl-4-oxo-2-aminylbutanamides and their acetyl-and
butyrylcholinesterase inhibitory activity. Investigation of AChE–ligand interactions
by docking calculations and molecular dynamics simulations, Eur. J. Med. Chem. 81
(
2014) 158–175.
[
[
20] G.L. Ellman, K.D. Courtney, V. Andres Jr., R.M. Featherstone, A new and rapid
colorimetric determination of acetylcholinesterase activity, Biochem. Pharmacol. 7
(
1961) 88–95.
21] F. Worek, T. Wille, M. Koller, H. Thiermann, Reactivation kinetics of a series of
related bispyridinium oximes with organophosphate-inhibited human acet-
ylcholinesterase—structure–activity relationships, Biochem. Pharmacol. 83 (2012)
1700–1706.
Transparency document
Transparency document related to this article can be found online at
https://doi.org/10.1016/j.cbi.2019.06.027.
[
22] F. Worek, U. Mast, D. Kiderlen, C. Diepold, P. Eyer, Improved determination of
acetylcholinesterase activity in human whole blood, Clin. Chim. Acta 288 (1999)
7
3–90.
Appendix A. Supplementary data
[23] Å. Forsberg, G. Puu, Kinetics for the inhibition of acetylcholinesterase from the
electric eel by some organophosphates and carbamates, Eur. J. Biochem. 140
(
1984) 153–156.
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.cbi.2019.06.027.
[
[
[
24] N. Aurbek, H. Thiermann, L. Szinicz, P. Eyer, F. Worek, Analysis of inhibition, re-
activation and aging kinetics of highly toxic organophosphorus compounds with
human and pig acetylcholinesterase, Toxicology 224 (2006) 91–99.
25] J. Cheung, M.J. Rudolph, F. Burshteyn, M.S. Cassidy, E.N. Gary, J. Love,
M.C. Franklin, J.J. Height, Structures of human acetylcholinesterase in complex
with pharmacologically important ligands, J. Med. Chem. 55 (2012) 10282–10286.
26] A. Waterhouse, M. Bertoni, S. Bienert, G. Studer, G. Tauriello, R. Gumienny,
F.T. Heer, T.A.P. de Beer, C. Rempfer, L. Bordoli, R. Lepore, T. Schwede, SWISS-
MODEL: homology modelling of protein structures and complexes, Nucleic Acids
Res. 46 (2018) W296–W303.
References
[
1] A. Levy, G. Cohen, E. Gilat, R. Duvdevani, N. Allon, S. Shapira, S. Dachir, E. Grauer,
Y. Meshulam, I. Rabinovitz, Therapeutic versus prophylactic treatment strategies
against nerve-agent induced brain injuries, Proc. Med. Def. Biosci. Rev. US-
AMRMC, Balt. (2000) 280–290.
[
2] P. Eyer, D. Kiderlen, V. Meischner, L. Szinicz, H. Thiermann, F. Worek, The current
status of oximes in the treatment of OP poisoning—comparing two regimes. Rome:
european Association of Poisons Centres and Clinical Toxicologists, J. Toxicol. Clin.
Toxicol. 41 (2003) 441–443 May 2003. Abstract.
[27] T.A. Halgren, MMFF VI. MMFF94s option for energy minimization studies, J.
Comput. Chem. 20 (1999) 720–729.
[28] J.J.P. Stewart, Optimization of parameters for semiempirical methods VI: more
modifications to the NDDO approximations and re-optimization of parameters, J.
Mol. Model. 19 (2013) 1–32.
[
[
3] F. Worek, H. Thiermann, L. Szinicz, P. Eyer, Kinetic analysis of interactions between
human acetylcholinesterase, structurally different organophosphorus compounds
and oximes, Biochem. Pharmacol. 68 (2004) 2237–2248.
[29] T.J. Dolinsky, P. Czodrowski, H. Li, J.E. Nielsen, J.H. Jensen, G. Klebe, N.A. Baker,
PDB2PQR: expanding and upgrading automated preparation of biomolecular
structures for molecular simulations, Nucleic Acids Res. 35 (2007) W522–W525.
4] Y. Ashani, A.K. Bhattacharjee, H. Leader, A. Saxena, B.P. Doctor, Inhibition of
cholinesterases with cationic phosphonyl oximes highlights distinctive properties of
the charged pyridine groups of quaternary oxime reactivators, Biochem. Pharmacol.
[
30] O. Trott, A.J. Olson, AutoDock Vina, Improving the speed and accuracy of docking
with a new scoring function, efficient optimization, and multithreading, J. Comput.
Chem. 31 (2010) 455–461.
6
6 (2003) 191–202.
[
[
[
[
[
5] D. Kiderlen, P. Eyer, F. Worek, Formation and disposition of diethylphosphoryl-
obidoxime, a potent anticholinesterase that is hydrolyzed by human paraoxonase
[31] O. Korb, T. Stützle, T.E. Exner, PLANTS: application of ant colony optimization to
structure-based drug design, in: M. Dorigo, L.M. Gambardella, M. Birattari,
A. Martinoli, R. Poli, T. Stützle (Eds.), Ant Colony Optim. Swarm Intell. Springer
Berlin Heidelberg, Berlin, Heidelberg, 2006, pp. 247–258.
(
PON1), Biochem. Pharmacol. 69 (2005) 1853–1867.
6] W.K. Berry, D.R. Davies, The use of carbamates and atropine in the protection of
animals against poisoning by 1, 2, 2-trimethylpropyl methylphosphonofluoridate,
Biochem. Pharmacol. 19 (1970) 927–934.
[
32] A. Pedretti, L. Villa, G. Vistoli, VEGA – an open platform to develop chemo-bio-
informatics applications, using plug-in architecture and script programming, J.
Comput. Aided Mol. Des. 18 (2004) 167–173.
7] P. Dirnhuber, M.C. French, D.M. Green, L. Leadbeater, J.A. Stratton, The protection
of primates against soman poisoning by pretreatment with pyridostigmine, J.
Pharm. Pharmacol. 31 (1979) 295–299.
[33] S. Jo, T. Kim, V.G. Iyer, W. Im, CHARMM-GUI: a web-based graphical user interface
for CHARMM, J. Comput. Chem. 29 (2008) 1859–1865.
[
34] J. Huang, S. Rauscher, G. Nawrocki, T. Ran, M. Feig, B.L. de Groot, H. Grubmüller,
A.D. MacKerell Jr., CHARMM36m: an improved force field for folded and in-
trinsically disordered proteins, Nat. Methods 14 (2016) 71–73.
8] L.W. Harris, D.L. Stitcher, W.C. Heyl, The effects of pretreatments with carbamates,
atropine and mecamylamine on survival and on soman-induced alterations in rat
and rabbit brain acetylcholine, Life Sci. 26 (1980) 1885–1891.
9] L. Leadbeater, R.H. Inns, J.M. Rylands, Treatment of poisoning by soman, Fundam.
Appl. Toxicol. 5 (1985) S225–S231.
[35] K. Vanommeslaeghe, E. Hatcher, C. Acharya, S. Kundu, S. Zhong, J. Shim, E. Darian,
O. Guvench, P. Lopes, I. Vorobyov, A.D. Mackerell, CHARMM general force field: a
force field for drug-like molecules compatible with the CHARMM all-atom additive
biological force fields, J. Comput. Chem. 31 (2009) 671–690.
[
[
10] W.J. Lennox, L.W. Harris, D.R. Anderson, R.P. Solana, M.L. Murrow, J.V. Wade,
Successful pretreatment/therapy of soman, sarin and VX intoxication, Drug Chem.
Toxicol. 15 (1992) 271–283.
[
36] J.C. Phillips, R. Braun, W. Wang, J. Gumbart, E. Tajkhorshid, E. Villa, C. Chipot,
R.D. Skeel, L. Kalé, K. Schulten, Scalable molecular dynamics with NAMD, J.
Comput. Chem. 26 (2005) 1781–1802.
11] Y. Meshulam, R. Davidovici, A. Wengier, A. Levy, Prophylactic transdermal treat-
ment with physostigmine and scopolamine against soman intoxication in
Guinea‐pigs, J. Appl. Toxicol. 15 (1995) 263–266.
[37] W. Humphrey, A. Dalke, K. Schulten, VMD: visual molecular dynamics, J. Mol.
10

M.D. Vitorović-Todorović, et al.
Chemico-Biological Interactions 309 (2019) 108714
Graph. 14 (1996) 33–38.
[43] G. Petroianu, K. Arafat, M. Kosanovic, A. Saleh, V. Camasamudram, M.Y. Hasan, In
vitro protection of red blood cell acetylcholinesterase by metoclopramide from
inhibition by organophosphates (paraoxon and mipafox), J. Appl. Toxicol. An Int. J.
23 (2003) 447–451.
[
[
38] M. Bermudez, C. Rakers, G. Wolber, Structural characteristics of the allosteric
binding site represent a key to subtype selective modulators of muscarinic acet-
ylcholine receptors, Mol. Inform. 34 (2015) 526–530.
39] A. Bock, M. Bermudez, F. Krebs, C. Matera, B. Chirinda, D. Sydow, C. Dallanoce,
U. Holzgrabe, M. De Amici, M.J. Lohse, G. Wolber, K. Mohr, Ligand binding en-
sembles determine graded agonist efficacies at a G protein-coupled receptor, J. Biol.
Chem. 291 (2016) 16375–16389.
[44] G.A. Petroianu, K. Arafat, A. Schmitt, M.Y. Hasan, Weak inhibitors protect choli-
nesterases from strong inhibitors (paraoxon): in vitro effect of ranitidine, J. Appl.
Toxicol. An Int. J. 25 (2005) 60–67.
[45] G.A. Petroianu, M.Y. Hasan, S.M. Nurulain, M. Shafiullah, R. Sheen, N. Nagelkerke,
Ranitidine in acute high‐dose organophosphate exposure in rats: effect of the
time‐point of administration and comparison with pyridostigmine, Basic Clin.
Pharmacol. Toxicol. 99 (2006) 312–316.
[
[
[
40] G. Wolber, T. Langer, LigandScout: 3-D pharmacophores derived from protein-
bound ligands and their use as virtual screening filters, J. Chem. Inf. Model. 45
(
2005) 160–169.
41] Z. Radić, P. Taylor, The influence of peripheral site ligands on the reaction of
symmetric and chiral organophosphates with wildtype and mutant acet-
ylcholinesterases, Chem. Biol. Interact. 119 (1999) 111–117.
[46] B. Nizami, D. Sydow, G. Wolber, B. Honarparvar, Molecular insight on the binding
of NNRTI to K103N mutated HIV-1 RT: molecular dynamics simulations and dy-
namic pharmacophore analysis, Mol. Biosyst. 12 (2016) 3385–3395.
[47] J. Mortier, C. Rakers, M. Bermudez, M.S. Murgueitio, S. Riniker, G. Wolber, The
impact of molecular dynamics on drug design: applications for the characterization
of ligand-macromolecule complexes, Drug Discov. Today 20 (2015) 686–702.
42] G. Petroianu, U. Beha, C. Roth, W. Bergler, R. Rüfer, L‐lactate protects in vitro
acetylcholinesterase (AChE) from inhibition by paraoxon (E 600), J. Appl. Toxicol.
An Int. J. 20 (2000) 249–257.
11