Synthesis of imperatorin analogs and their evaluation as acetylcholinesterase and butyrylcholinesterase inhibitors

Article abstract of DOI:10.1002/ardp.201300259

In this study, we synthesized several imperatorin analogs using imperatorin and xanthotoxin as substrates. The anti-cholinesterase activities of all compounds were evaluated in in vitro experiments according to the modified Ellman's method. For each synthesized compound, IC₅₀ values for both enzymes were established. Galantamine hydrobromide was used as a positive control in the enzymatic experiments. All active compounds showed selectivity toward butyrylcholinesterase (BuChE) rather than acetylcholinesterase. The most active ones were 8-(3-methylbutoxy)-psoralen and 8-hexoxypsoralen with IC ₅₀ values for BuChE of around 16.5 and 16.4 μM, respectively. The results of our study may be considered as the beginning of a search for potential anti-Alzheimer's disease drugs based on the structure of natural furocoumarins. Several imperatorin analogs were synthesized using imperatorin and xanthotoxin as substrates. Their anti-cholinesterase activities were evaluated, and all active compounds showed selectivity toward butyrylcholinesterase (BuChE) rather than acetylcholinesterase. The most active compounds were 8-(3-methylbutoxy)-psoralen and 8-hexoxypsoralen with IC ₅₀ values for BuChE of around 16.5 and 16.4 μM, respectively.

Full text of DOI:10.1002/ardp.201300259

Arch. Pharm. Chem. Life Sci. 2013, 346, 775–782
775
Full Paper
Synthesis of Imperatorin Analogs and Their Evaluation as
Acetylcholinesterase and Butyrylcholinesterase Inhibitors
Sebastian Granica1, 2, Anna K. Kiss2, Małgorzata Jarończyk3, Jan K. Maurin3, 4
,
Aleksander P. Mazurek3, and Zbigniew Czarnocki1
1
Faculty of Chemistry, University of Warsaw, Warsaw, Poland
Faculty of Pharmacy, Department of Pharmacognosy and Molecular Basis of Phytotherapy, Medical University
of Warsaw, Warsaw, Poland
National Medicines Institute, Warsaw, Poland
National Centre for Nuclear Research, Otwock-Świerk, Poland
2
3
4
In this study, we synthesized several imperatorin analogs using imperatorin and xanthotoxin as
substrates. The anti-cholinesterase activities of all compounds were evaluated in in vitro experiments
according to the modified Ellman’s method. For each synthesized compounds, IC50 values for both
enzymes were established. Galantamine hydrobromide was used as a positive control in the enzymatic
experiments. All active compounds showed selectivity toward butyrylcholinesterase (BuChE) rather
than acetylcholinesterase. The most active ones were 8-(3-methylbutoxy)-psoralen and 8-hexoxypso-
laren with IC50 values for BuChE of around 16.5 and 16.4 mM, respectively. The results of our study may
be considered as the beginning of a search for potential anti-Alzheimer’s disease drugs based on the
structure of natural furocoumarins.
Keywords: Acetylcholinesterase / Alzheimer’s disease / Angelica archangelica / Butyrylcholinesterase /
Furocoumarins
Received: July 11, 2013; Revised: August 9, 2013; Accepted: August 12, 2013
DOI 10.1002/ardp.201300259
Introduction
bial property of psoralens was noted [4–6]. Later studies
have shown anti-inflammatory [7–12], antitumor [13],
antidiabetic [14], hepatoprotective [14, 15], and anticonvul-
sant [16–18] activity of these compounds. Imperatorin
enhances the effectivity of traditional antiepileptic drugs [16,
18]. Moreover, other studies have shown that psoralens can
modulate GABAA receptor function [19]. Recently, it has been
proved that linear furocoumarins such as xanthotoxin,
bergapten, imperatorin, isoimperatorin, and oxypeucedanin
inhibit acetylcholinesterase activity in vitro [20, 21]. It
suggests their possible use in prevention and treatment of
Alzheimer’s disease (AD). The use of acetylcholinesterase
inhibitors (AChEI) in AD increases the level of neural
acetylcholine (ACh) and improves cognitive functions [22].
Besides, acetylcholinesterase not only plays a crucial role in
AD because of its catalytic function but also is claimed to be
involved in the formation of b-amyloid in central nervous
system (CNS) [23]. Apart from AChEI, in the past few years
there has been growing interest in butyrylcholinesterase
(BuChE) and its role in AD. Some studies have shown that
BuChE is also responsible for a low level of ACh in CNS
Furocoumarins are naturally occurring compounds pro-
duced by a wide variety of plants. The chemical structure of
furocoumarins consists of furan ring fused with coumarin
(benzo-a-pyrone). The furan motif can be attached in a
different manner to produce several types of furocoumarins
but there are only two most common structure types. The
first one possesses an angelicin-type ring arrangement and
the second, even more popular group includes psolaren-type
furocoumarins often called linear furocoumarins. Some
psolarens, e.g., xanthotoxin and its derivatives, have been
used as photosensitizing compounds for the treatment of
psoriasis, vitiligo, and other skin infections [1–3]. In the
last decades, a great deal of research has been carried out on
the bioactivity of linear furocoumarins. Initially, antimicro-
Correspondence: Prof. Zbigniew Czarnocki, Faculty of Chemistry,
University of Warsaw, Pasteura 1, 02-093 Warsaw, Poland.
E-mail: czarnoz@chem.uw.edu.pl
Fax: þ48 22 822 5996
ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

776
S. Granica et al.
Arch. Pharm. Chem. Life Sci. 2013, 346, 775–782
especially in the advanced phase of AD [24]. This is why the
usage of double AChE–BuChE inhibitors such as rivastigmine
is nowadays recommended in the treatment of AD.
were previously described in the literature [31–35]. Com-
pound 6 was reported as an anti-inflammatory agent but no
spectroscopic data were given [8]. To our knowledge,
compound 8 is a new substance. Some of the synthesized
compounds possess asymmetric centers that lead to the
formation of enantiomers. Since the reactions that we used
for the modification of heraclenin (9) were not stereoselective,
we obtained racemic mixtures in the case of compounds 10
and 11.
As previously reported, the presence of single furocoumar-
ins moiety is not enough for inhibitory activity toward
AChE and BuChE [21]. The most important for cholinesterase
inhibition is the structure of the side chain of the condensed
furocoumarin rings. We therefore synthesized several com-
pounds of different side chain structure. This part of
the molecule is responsible for the lipophilicity, which
may affect the biological profile. Consequently, we decided
to check if there is a simple correlation between the
structure and anti-AChE/BuChE activity for prepared
compounds.
AChE and BuChE are a dyad of enzymes occurring not only
in CNS but also in other tissues. Both enzymes catalyze
hydrolysis of selected ester bonds in organic compounds.
AChE is responsible for rapid degradation of neurotransmit-
ter – ACh in neurons. Although BuChE may exhibit the same
activity, it is also believed to have a profound toxicological
role as it hydrolyzes ester bonds in drugs and other
xenobiotics. AChE has characteristic narrow gorge leading
to its active site while BuChE gorge is wider [36, 37]. That is
probably why AChE can hydrolyze only small acyl esters in
contrary to BuChE, which can accommodate larger acyl
groups and hence larger substrates. Accordingly, we expected
that furocoumarins with longer and more branched side
chain would be more selective toward BuChE.
Archangelica angelica L., syn. Angelica officinalis HOFFM or
Angel’s herb, as it is commonly called, is a biennial plant from
Apiacae family, which is a popular vegetable and herb
cultivated especially in Europe and Asia for medical and
culinary purposes. In some countries extracts from Angelica
are used as an appetite stimulant, an antispasmodic, and a
remedy for gastrointestinal symptoms such as bloating, poor
digestion, eructation, and flatulence [25]. It is well known that
traditionally used roots of Angel’s herb, besides volatile oil
used for aromatization of liqueurs and vermouths [25], are
rich source of coumarins. After cultivation of A. angelica L. for
medical and culinary applications companies obtain vast
amounts of fruits and other aerial parts that become useless.
Some studies have proven that also fruits of Angelica contain
coumarins in high quantities [26]. In our previous study we
have shown that psoralens from A. angelica L. inhibit not only
AChE activity but BuChE as well [27]. Some compounds
isolated from Angel’s herb in our laboratory exhibited high
selectivity toward BuChE [27]. Considering this, we have
decided to use fruits of A. angelica L. as a rich source of linear
furocoumarins for chemical transformations and further
anti-AChE–BuChE activity studies.
Results and discussion
In the presented study, we found furocoumarins to be
selective BuChE inhibitors. Previous research has shown that
some psoralens may exhibit anti-AChE and anti-BuChE
activity but there are only few studies considering any
relationship between the structure and the effect on
inhibition of cholinesterase [20, 21, 27–30]. In these
publications, authors indicated that the presence of a side
chain at C-8 position of furocoumarin nucleus is crucial for
the inhibitory activity toward BuChE. Moreover, the C-5
position must not be occupied [27]. In the case of AChE, the
studies showed that furocoumarins should have elongated
side chain in the position C-5 or C-8 in order to achieve better
AChE inhibitory activity compared to the presence of single
–OCH3 group [30].
Initially, we isolated imperatorin (1), which was a necessary
substrate for further chemical transformations. Using fruits
of A. angelica L., which become useless after cultivation, we
obtained imperatorin albeit in rather inadequate amounts
and therefore we decided to use commercially available
xanthotoxin (2) as a substrate for further transformations.
This compound also occurs naturally in extracts from Angel’s
herb. The xanthotoxin was used only for the preparation of
xantotoxol (3) as a substrate for the formation of the
corresponding alkyl ethers. Most of the obtained compounds
On the basis of the results obtained from anti-AChE/BuChE
assays, we determined the IC50 values for majority of the
compounds based on the concentration–inhibition curves as
shown in Table 1.
As we expected, some of the compounds exhibited
selectivity toward BuChE. The most potent were derivatives
5 and 6, which contain simple saturated carbon chains. It may
be related to the optimal size of those molecules (they are too
large for AChE but, in the case of BuChE, they fit perfectly to
the active site of the enzyme). It is interesting to note that
compound 4 is less effective, although it has similar structure
to imperatorin. Inactivity of compound 8 is probably caused
by too long side chain that makes it impossible to get near the
active site of both enzymes. In the case of imperatorin
derivatives, it is important that compound 11 contains the
ester bond in its structure that should enhance the inhibitory
activity of both cholinesterases. Nevertheless, we did not
notice that the presence of an extra polar groups such as
ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

Arch. Pharm. Chem. Life Sci. 2013, 346, 775–782
Imperatorin Analogs as AChE and BuChe Inhibitors
777
Table 1. IC50 values for inhibition of AChE and BuChE by
synthesized furocoumarins.
both enzymes [38], that may explain the lower activity of
compound 4.
The docking studies indicated that compounds 5 and 6 can
form hydrogen bonds with Gly117 and with side chain of
Ser198 in the binding pocket of BuChE (Fig. 2). These H-bonds
are believed to contribute to the higher affinity of these
compounds toward BuChE than AChE, which may account for
their selectivity toward BuChE. For comparison, compounds 5
and 6 were located at the binding pocket of AChE next to large
aromatic amino acids that may cause steric interactions
(Fig. 3).
In the case of compound 8, the binding to AChE may be
impossible. The interactions of a long side chain of compound
8 with AChE are prevented due to the presence of aromatic
amino acids in the binding site: Phe288, Phe290, Phe330, and
AChE
BuChE
a)
b)
a)
IC50
%
inh
IC50
(mM)
at 50 mM
(mM)
Imperatorin (1)
Xanthotoxin (2)
Xanthotoxol (3)
>50
>50
>50
>50
>50
>50
>50
>50
NTc)
>50
>50
10.2 ꢀ 1.8 31.4 ꢀ 11.9
11.0 ꢀ 4.4 >50
NAd)
>50
8-Butoxypsoralen (4)
8-Hexoxypsoralen (5)
8-(3-Methylbutoxy)-psoralen (6)
8-Benzyloxypsoralen (7)
8-Decyloxypsoralen (8)
Heraclenin (9)
tert-O-Methylheraclenol (10)
tert-O-Methylheraclenol
acetate (11)
17.8 ꢀ 5.5 41.4 ꢀ 10.6
25.6 ꢀ 1.5 16.5 ꢀ 0.5
22.1 ꢀ 4.4 16.4 ꢀ 7.4
10.3 ꢀ 6.8 48.1 ꢀ 5.4
NAd)
NTc)
>50
NTc)
21.9 ꢀ 1.9 >50
29.1 ꢀ 3.4 50.4 ꢀ 12.3
Galanthamine hydrobromide
1.5 ꢀ 0.5 95.4 ꢀ 0.2
9.4 ꢀ 1.1
All values were calculated from three independent experiments
performed in triplicate.
a)
IC50 – values means ꢀ standard deviation.
b)
%
inh
– values means ꢀ standard deviation.
c)
NT – not tested.
NA – not active at 50 mM.
d)
–OCH3, –OH, or –OCOCH3 is crucial for a better selectivity
toward BuChE as our last results would suggest [27].
In order to investigate the binding mode for interaction of
the compounds with AChE and BuChE and to explain
different binding affinities, docking studies were performed.
The activity of compound 4 is lower than that of compound 1.
The binding of both compounds to AChE (Fig. 1) indicated that
p–p stacking interactions with Trp84 and Phe330 are
important. Additionally the double bond in prenyl group of
compound 1 serves as a specific p-hydrogen bond acceptor for
Figure 2. Binding interactions of compound 5 with selected
residues of BuChE (H-bonds are indicated as black dotted lines).
Figure 1. Binding interactions of compound 1 and 4 with selected
residues of AChE.
Figure 3. Binding interactions of compound 5 with selected
residues of AChE.
ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

778
S. Granica et al.
Arch. Pharm. Chem. Life Sci. 2013, 346, 775–782
Tyr334. Although the binding pocket in BuChE is larger than
that in AChE [39], the binding of compound 8 is not possible
because of the size of the molecule.
modifications, anhydrous aluminum chloride, anhydrous sodi-
um sulfate, anhydrous potassium carbonate, pyridine, acetic
anhydride, and chromatographic solvents (cyclohexane, ethyl
acetate, chloroform, isopropanol) were obtained from POCh
(Gliwice, Poland).
Conclusion
Plant material extraction and imperatorin (1) isolation
Dried and powdered fruits (200 g) were extracted with hexane
(3 ꢁ 700 mL) in a water bath at 70°C for 5 h each time. The
collected extracts were evaporated giving a residue of 22 g.
Hexane-soluble residue was subjected to silica gel column
chromatography and eluted with CHCl3 to obtain six main
fractions based on their TLC profile. Fraction 2 (15 g) was
rechromatographed on a silica gel column with a gradient of
hexane–CHCl3 from 97.5:2.5 to 85:15. One hundred ten fractions
were collected and pooled into 10 main fractions based on their
TLC profile. Imperatorin (270 mg) was isolated from fr. 6 by
recrystallization from petroleum ether, m.p. 93–95°C (lit. m.p.
96–98°C [40]).
In conclusion, the presented study indicates that furocou-
marins can be considered as selective BuChE inhibitors. Their
activity depends on the structure of the molecule and can be
modified by a simple chemical transformation. Therefore,
plants containing significant amounts of furocoumarins
should be considered as a rich source of substrates for
chemical transformations leading to new and more effective
compounds. Molecular docking studies were carried out to
further investigate the binding modes of studied compounds
with both cholinesterases.
During the isolation process we obtained other less pure
fractions of imperatorin, but suitable for chemical modification.
Total amount of imperatorin was 990 mg (85–95% HPLC).
Experimental
General procedure
1H and 13C NMR spectra were recorded on Varian Unityplus-200
spectrometer. MS analyses were performed using LCT (TOF) and
Amazon SL Brucker apparatus. The X-ray data for a crystal
structure were collected using Oxford Diffraction Xcalibur R k-axis
diffractometer equipped with CCD detector. Data were collected at
room temperature using the monochromated Cu Ka radiation.
Thin layer chromatography (TLC) was carried out on silica gel F254
plates (Merck, Darmstadt, Germany). Silica gel 60 (230–400 mesh)
for column chromatography was purchased from Merck.
Synthetic approach
We used xanthotoxin supplied by Acros Organics or imperatorin
(1) isolated in our laboratory as starting material. Xanthotoxin (2)
was first demethylated with AlCl3 to obtain 8-hydroxypsoralen
(xanthotoxol) (3) according to the method previously described
[41] with our modifications. It was then used for the preparation
of the corresponding alkyl ethers using the method described
in literature [34, 41–43] (Scheme 1). The epoxidation of imperatorin
was carried out using MCPBA [35]. The epoxide ring was opened
with oxalic acid to be next acetylated in Ac2O/pyridine system
(Scheme 2) [32].
Chemicals
Acetylcholinesterase (AChE) (EC 3.1.1.7. type VI-S from Electric
Eel), BuChE (EC 3.1.1.8. from horse serum), acetylthiocholine
iodide (ATCI), S-butyrylthiocholine chloride (BTCCl), 5,50-dithio-
bis(2-nitrobenzoic acid) (DTNB), and dimethylsulfoxide (DMSO)
were obtained from Sigma–Aldrich (St. Louis, MO, USA). Tris–HCl
was obtained from Merck. Xanthotoxin was obtained from
Acros Organics (New Jersey, USA). Alkyl bromides used for
Preparation of 8-hydroxypsoralen (9-hydroxy-7H-
furo[3,2g]chromen-7-one) (3)
To the solution of xanthotoxin (500 mg, 2.31 mmol) in dichloro-
methane (25 mL) anhydrous aluminum chloride (6 g, 45 mmol)
was added. The mixture was stirred for 24 h at room temperature.
Then 20 mL of water was added in small portions followed by the
Scheme 1. Chemical transformations of xanthotoxin.
ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

Arch. Pharm. Chem. Life Sci. 2013, 346, 775–782
Imperatorin Analogs as AChE and BuChe Inhibitors
779
Scheme 2. Chemical transformations of imperatorin.
addition of 5 mL of concentrated hydrochloric acid. After cooling,
the mixture was extracted three times with 25 mL portions of
ethyl acetate (AcOEt). Organic layers were dried over anhydrous
sodium sulfate. After evaporation under reduced pressure, yellow
powder was obtained (315 mg, 1.55 mmol, 67%), m.p. 247–250°C
(lit. m.p. 247°C [44]).
8-(3-Methylbutoxy)-psoralen (9-(3-methylbutoxy)-7H-furo-
[3,2g]chromen-7-one) (6): An off-white powder; m.p. 67–
72°C; yield 35%; 1H NMR (200 MHz, CDCl3) d 7.77 (d, J ¼ 9.6 Hz, 1H,
H-4), 7.69 (d, J ¼ 2.3 Hz, 1H, H-11), 7.36 (s, 1H, H-5), 6.82 (d,
J ¼ 2.2 Hz, 1H, H-12), 6.37 (d, J ¼ 9.6 Hz, 1H, H-3), 4.52 (t, J ¼ 6.7 Hz,
2H, 13-CH2), 2.03–1.85 (m, 1H, 15-CH), 1.77 (q, J ¼ 6.8 Hz, 2H, 14-
CH2), 1.01 (s, 3H, 16-CH3), 0.98 (s, 3H, 17-CH3). 13C NMR (50 MHz,
CDCl3) d 160.8 C-2, 148.4 C-7, 146.8 C-11, 144.5 C-4, 143.7 C-9, 132.3
C-8, 126.2 C-6, 116.7 C-10, 114.9 C-5, 113.1 C-3, 106.9 C-12, 72.8 C-
13, 39.0 C-14, 29.9 C-15, 25.0 C-16, 22.8 C-17. In the literature, no
spectral data or melting point given [8].
General procedure for preparation of O-alkylated
xanthotoxol derivatives
A mixture of xanthotoxol (100 mg, 0.49 mmol), anhydrous K2CO3
(215 mg, 1.15 mmol), alkyl or benzyl bromide (1.43 mmol),
and acetone (7.5 mL) was stirred and refluxed for 24 h. Then
30 mL of water was added and the mixture was extracted three
times with 25 mL portions of ethyl acetate (AcOEt). Organic layers
were dried over anhydrous sodium sulfate. Removal of solvent
yielded yellowish oil or solid, which was purified by column
chromatography on a silica gel, eluting with mixtures of
cyclohexane:ethyl acetate to give pure corresponding 8-O-alkyl
ethers of xanthotoxol 4–8.
8-Benzyloxypsoralen (9-benzyloxy-7H-furo[3,2g]chromen-
7-one) (7): An off-white powder; m.p. 118–124°C (lit. m.p. 121–
122°C [31]); yield 8%.
8-n-Decyloxypsoralen(9-n-decyloxy-7H-furo[3,2g]chro-
men-7-one) (8): Colorless oil; yield 32%; IR (KBr): n cmꢂ1 3112,
2917, 1720, 1585, 1401, 1145 1H NMR (200 MHz, CDCl3) d 7.77 (d,
J ¼ 9.8 Hz, 1H, H-4), 7.69 (d, J ¼ 2.2 Hz, 1H, H-11), 7.35 (s, 1H, H-5),
6.82 (d, J ¼ 2.2 Hz, 1H, H-12), 6.37 (d, J ¼ 9.4 Hz, 1H, H-3), 4.49 (t,
J ¼ 6.7 Hz, 2H, 13-CH2), 1.87 (quin, J ¼ 7.1 Hz, 2H, 14-CH2), 1.69–
1.48 (m, 6H, 3 ꢁ CH2), 1.46–1.18 (m, 8H, 4 ꢁ CH2), 0.91–0.85 (m,
3H, 22-CH3). 13C NMR (50 MHz, CDCl3) d 160.8 C-2, 148.4 C-7, 146.8
C-11, 144.5 C-4, 143.7 C-9, 132.3 C-8, 126.1 C-6, 116.7 C-10, 114.9
C-5, 113.1 C-3, 106.9 C-12, 74.4 C-13, 32.1 C-14, 30.3 C-15, 29.8 C-16,
29.6 C-17, 29.5 C-18, 27.1 C-19, 25.9 C-20, 22.9 C-21, 14.3 C-22.
8-Butoxypsoralen (9-butoxy-7H-furo[3,2g]chromen-7-one)
(4): An off-white powder; m.p. 79–81°C (lit. m.p. 83°C [45]);
yield 30%.
8-Hexoxypsoralen (9-hexoxy-7H-furo[3,2g]chromen-7-one)
(5): An off-white powder; m.p. 54–56°C (lit. m.p. 55°C [34]);
yield 35%.
ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

780
S. Granica et al.
Arch. Pharm. Chem. Life Sci. 2013, 346, 775–782
LR-ESI-MSþ: 365 [MþNa]þ, 397 [MþMeOHþNa]þ; HR-ESI-MSþ:
365.1726 [MþNa]þ (calculated for C21H26O4Na – 365.1729).
approach [48]. The ligand molecules were fully flexible and the
protein was represented by grid interaction potentials.
Preparation of heraclenin (9-(20,30-epoxy-30-
Crystal structures
Monocrystals of (5) were obtained by slow evaporation of solvents
(hexanes/ethyl acetate). The structure was solved using direct
methods from SHELXS93 program [51]. After location of most
non-hydrogen atoms forming condensed rings system from
initial E-maps, further heavy atoms were found in subsequent
difference-Fourier syntheses using SHELXL93 software [51]. The
structure was refined using full-matrix least-squares procedure
from the same software. Hydrogen atoms were then added using
standard geometrical criteria. Crystal structure is triclinic from
the centrosymmetric P-1 space group. In the independent part of
the unit cell, there are two molecules of different conformation of
the substituent at C8 atom (see Fig. 4).
methylbutoxy)-7H-furo[3,2-g]-[1]benzopyran-7-one) (9)
The mixture of imperatorin (406 mg, 1.5 mmol), chloroform
(25 mL), and m-chloroperbenzoic acid (725 mg, 3.6 mmol) was
stirred at room temperature for 2 h. Then the mixture was poured
into saturated solution of NaHCO3 (100 mL) and extracted three
times with 25 mL portions of ethyl acetate (AcOEt). Organic layers
were dried over anhydrous sodium sulfate. Removal of solvent
yielded a brown solid, which was purified by column chroma-
tography on a silica gel, eluting with mixtures of cyclohexane:
ethyl acetate from 100:0 to 80:20 v/v. After purification, a pale
yellow powder was obtained (287 mg, 1.0 mmol, 67%), m.p. 108–
111°C (lit. m.p. 112–114°C [46]).
Molecules are partly disordered at the hydrocarbon unit, which
was expected for flexible fragments. The doubling of the number
of molecules in the asymmetric part of the unit cell doubles the
number of the refined parameters. Poor quality of the crystal
taken for the experiment, large number of parameters to be
refined, together with a partial disorder of the substituents
resulted in a relatively high agreement factor R1. Nevertheless, all
obtained geometrical parameters of the model are acceptable.
Molecules in the crystal structure form two well-defined areas:
hydrophilic (polar) formed by planar ring fragments connected
via weak C–H ꢃ ꢃ ꢃ O hydrogen bonds into planes parallel to (1 0 0)
plane; and hydrophobic (lipophilic) formed by alkyl substituents
(see Fig. 5). The consecutive, translation related layer of molecules
in the b-direction is turned with its hydrophilic part toward
hydrophilic area of preceding layer of molecules. Such structure
resembles that of liposome or biological membranes.
Preparation of tert-O-methylheraclenol (9-(30-hydroxy-20-
methoxy-30-methylbutoxy)-7H-furo[3,2g]chromen-7-one)
(10)
Heraclenin (200 mg, 0.7 mmol) and oxalic acid (80 mg, 0.6 mmol)
in methanol (20 mL) were stirred and refluxed for 2 h. Removal of
solvent yielded a brownish solid, which was purified by column
chromatography on a silica gel, eluting with mixtures of
cyclohexane:ethyl acetate from 95:5 to 85:15 v/v. After purifica-
tion, a yellow solid was obtained (197 mg, 0.62 mmol, 89%), m.p.
75–80°C (lit. m.p. 84–85°C [33]).
Preparation of tert-O-methylheraclenol acetate
(9-(30-acetoxy-20-methoxy-30-methylbutoxy)-7H-furo[3,2g]
chromen-7-one) (11)
tert-O-Methylheraclenol (50 mg, 0.16 mmol) was kept overnight
in pyridine (3 mL) containing acetic anhydride (10 drops).
Pyridine was then removed under reduced pressure and the
residue was washed twice with toluene and 5% solution of citric
acid in water. Organic layers were dried over anhydrous sodium
sulfate. Removal of the solvent yielded yellow oil (35 mg,
0.1 mmol, 62%).
AChE inhibitory assay in 96-microtiter-well plates
The AChE assay in 96-microtiter-well plates used in this paper was
based on Ellman reaction [52] with further modifications. The
1H NMR (200 MHz, CDCl3) d 7.75 (d, J ¼ 9.6 Hz, 1H, H-4), 7.68 (d,
J ¼ 2.2 Hz, 1H, H-11), 7.37 (s, 1H, H-5), 6.82 (d, J ¼ 2.3 Hz, 1H, H-12),
6.36 (d, J ¼ 9.6 Hz, 1H, H-3), 5.42 (dd, J ¼ 8.7, 2.6 Hz, 1H, 13-CH2),
5.00 (dd, J ¼ 11.1, 2.6 Hz, 1H, 13-CH2), 4.42 (dd, J ¼ 11.1, 8.7 Hz, 1H,
14-CH), 3.26 (s, 3H, 18-CH3), 2.16 (s, 3H, 19-CH3), 1.27 (s, 3H, 16-
CH3), 1.23 (s, 3H, 17-CH3). 13C NMR (50 MHz, CDCl3) d 170.8 C-19,
160.2 C-2, 148.1 C-7, 146.6 C-11, 144.2 C-4, 143.5 C-9, 131.5 C-8,
125.9 C-6, 116.5 C-10, 114.9 C-5, 113.5 C-3, 106.8 C-12, 75.9 C-15,
75.5 C-14, 72.5 C-13, 49.9 C-18, 22.6 C-20, 21.2 C-16, 21.1 C-17. In
the literature, no full NMR data were presented [47].
Molecular modeling
The docking study was performed using ICM suite of programs
(Internal Coordinate Mechanics) [48]. 3D structures of the ligands
(1–11) were built using ICM sketcher module. The crystal
structures of AChE (PDB-id: 1ACJ) [49] and BuChE (PDB-id:
1P0I) [50] retrieved from the Protein Data Bank were used as
the initial 3D structure. Then, the water molecules and co-
crystallized ligand were removed from the PDB structures. The
ICMPocketFinder was used to identify binding site. The ligands
were docked using a regular rigid receptor-flexible ligand docking
Figure 4. Structure of independent part of the unit cell.
ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

Arch. Pharm. Chem. Life Sci. 2013, 346, 775–782
Imperatorin Analogs as AChE and BuChe Inhibitors
781
[5] Y. S. Kwon, A. Kobayashi, S. I. Kajiyama, K. Kawazu,
H. Kanzaki, C. M. Kim, Phytochemistry 1997, 44, 887–889.
[6] Y. Tada, Y. Shikishima, Y. Takaishi, H. Shibata, T. Higuti,
G. Honda, M. Ito, Y. Takeda, O. K. Kodzhimatov,
O. Ashurmetov, Y. Ohmoto, Phytochemistry 2002, 59, 649–654.
[7] A. N. García-Argáez, T. O. Ramírez Apan, H. P. Delgado,
G. Velázquez, M. Martínez-Vázquez, Planta Med. 2000, 66,
279–281.
Figure 5. The crystal packing of compound 5. The molecules are
bonded into layers by weak C–H ꢃ ꢃ ꢃ O hydrogen bonds shown as the
blue dashed lines.
[8] S. Bal-Tembe, D. D. Joshi, A. D. Lakdawala, Indian J. Chem. B
1996, 35, 518–519.
[9] N. Márquez, R. Sancho, M. Ballero, P. Bremner,
G. Appendino, B. L. Fiebich, M. Heinrich, E. Muñoz, Planta
Med. 2004, 70, 1016–1021.
protocol started by adding 50 mL of 50 mM Tris–HCl buffer (pH
7.8) and 25 mL of 3 mM ATCI in water to 125 mL of 0.6 mM DTNB in
50 mM Tris–HCl buffer (pH 7.8). Next 25 mL of tested furocoumar-
ins solution in buffer containing DMSO (to obtain a final
concentration in the assay of 10, 20, and 50 mM) or galantamine
hydrobromide solution in 50 mM Tris–HCl buffer (pH 7.8) (as a
positive control) or buffer containing DMSO (as a negative
control, DMSO final concentration <0.05%) was added. Changes
in absorbance were measured nine times during 5-min period to
assess the spontaneous hydrolysis of the substrate (at l ¼ 405 nm
in 37°C, Biotek Synergy 4). Then, 25 mL of AChE (0.30 U/mL in
50 mM Tris–HCl buffer, pH 7.8) was added, and immediately after
that, the plates were shaken for 3 s and the measurement was
repeated under the same conditions as described above. The
activity of tested furocoumarins was calculated as inhibition
percentage (%inh) of AChE in relation to maximum activity
(negative control) after subtracting in all cases spontaneous
hydrolysis of the substrate as shown in Eq. (1)
[10] H. S. Ban, S. S. Lim, K. Suzuki, S. H. Jung, S. Lee, Y. S. Lee,
K. H. Shin, K. Ohuchi, Planta Med. 2003, 69, 408–412.
[11] P. Zhou, Y. Takaishi, H. Duan, B. Chen, G. Honda, M. Itoh,
Y. Takeda, O. K. Kodzhimatov, K. H. Lee, Phytochemistry 2000,
53, 689–697.
[12] G. Roos, J. Waiblinger, S. Zschocke, J. H. Liu, I. Klaiber,
W. Kraus, R. Bauer, Pharm. Pharmacol. Lett. 1997, 7, 157–160.
[13] Y. K. Kim, S. K. Young, Y. R. Shi, Phytother. Res. 2007, 21, 288–
290.
[14] A. C. Adebajo, E. O. Iwalewa, E. M. Obuotor, G. F. Ibikunle,
N. O. Omisore, C. O. Adewunmi, O. O. Obaparusi, M. Klaes,
G. E. Adetogun, T. J. Schmidt, E. J. Verspohl, J. Ethnopharmacol.
2009, 122, 10–19.
[15] H. Oh, H. S. Lee, T. Kim, K. Y. Chai, H. T. Chung, T. O. Kwon,
J. Y. Jun, O. S. Jeong, Y. C. Kim, Y. G. Yun, Planta Med. 2002, 68,
463–464.
ꢀ
ꢁ
[16] J. J. Luszczki, E. Wojda, M. Andres-Mach, W. Cisowski,
M. Glensk, K. Glowniak, S. J. Czuczwar, Epilepsy Res. 2009, 85,
293–299.
Asample
%
¼
1 ꢂ
ꢁ 100%
ð1Þ
inh
Acontrol
[17] S. Y. Choi, E. M. Ahn, M. C. Song, D. W. Kim, J. H. Kang,
O. S. Kwon, T. C. Kang, N. I. Baek, Phytother. Res. 2005, 19, 839–
845.
BuChE inhibitory assay in 96-microtiter-well plates
BuChE inhibitory activity of furocoumarins was measured using
protocol similar to the one described for AChE activity. The
difference was that 25 mL of 3 mM BTCCl in water was added as a
substrate and 25 mL of BuChE (0.15 U/mL in 50 mM Tris–HCl
buffer, pH 7.8) was used to start enzymatic reaction. Galantamine
hydrobromide solution in 50 mM Tris–HCl buffer (pH 7.8) was
used as a positive control.
[18] J. J. Luszczki, K. Glowniak, S. J. Czuczwar, Eur. J. Pharmacol.
2007, 574, 133–139.
[19] J. Zaugg, E. Eickmeier, D. C. Rueda, S. Hering, M. Hamburger,
Fitoterapia 2011, 82, 434–440.
[20] K. K. Dae, P. L. Jong, H. Y. Jae, O. E. Dong, S. E. Jae, H. L. Kang,
Arch. Pharm. Res. 2002, 25, 856–859.
All compounds used for biochemical assays were purified up to
95% (HPLC).
[21] A. Fallarero, P. Oinonen, S. Gupta, P. Blom, A. Galkin,
C. G. Mohan, P. M. Vuorela, Pharmacol. Res. 2008, 58, 215–221.
[22] J. Vetulani, Psychogeriatria Polska 2008, 5, 1–13.
Statistics
[23] N. C. Inestrosa, A. Alvarez, C. A. Pérez, R. D. Moreno,
M. Vicente, C. Linker, O. I. Casanueva, C. Soto, J. Garrido,
Neuron 1996, 16, 881–891.
The results were expressed as a mean ꢀ standard deviation (SD) of
three independent experiments performed in triplicate. IC50
values for examined compounds were established basing on
concentration–inhibition curves.
[24] J. Vetulani, Roczn. Psychogeriatr. 2003, 6, 1–12.
[25] M. Wichtl, Herbal Drugs and Phytopharmaceuticals, third ed.,
Medpharm Scientific Publishers, Stuttgart 2004.
References
[26] K. Glowniak, A. Gawron, B. Kwietniewska, Ann. Univ. Mariae
Curie Sklodowska D Med. 1976, 31, 349–354.
[1] M. Couperus, Calif. Med. 1954, 81, 402–406.
[27] N. Wszelaki, K. Paradowska, M. K. Jamróz, S. Granica,
[2] H. Tronnier, R. Loehning, Diagnostik 1975, 8, 392–394.
[3] L. Kittler, G. Lober, Hygiene þ Medizin. 1993, 18, 541–547.
[4] M. A. Loutfy, A. Khatibie, Pharmazie 1983, 38, 137–138.
A. K. Kiss, J. Agric. Food Chem. 2011, 59, 9186–9193.
[28] S. Y. Kang, K. Y. Lee, S. H. Sung, M. J. Park, Y. C. Kim, J. Nat. Prod.
2001, 64, 683–685.
ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com

782
S. Granica et al.
Arch. Pharm. Chem. Life Sci. 2013, 346, 775–782
[29] A. Urbain, A. Marston, K. Hostettmann, Pharm. Biol. 2005, 43,
647–650.
[41] V. Lartillot, A. Risler, Z. Andriamialisoa, M. Giraud, T. Sá.
E. Melos, L. Michel, R. Santus, Photochem. Photobiol. 2003, 78,
623–632.
[30] S. Di Giovanni, A. Borloz, A. Urbain, A. Marston,
K. Hostettmann, P. A. Carrupt, M. Reist, Eur. J. Pharm. Sci.
2008, 33, 109–119.
[42] E. C. Row, S. A. Brown, A. V. Stachulski, M. S. Lennard, Bioorg.
Med. Chem. 2006, 14, 3865–3871.
[31] S. K. Banerjee, B. D. Gupta, K. Singh, J. Chem. Soc. Chem.
Commun. 1982, 14, 815–816.
[43] M. A. Loutfy, Pharmazie 1981, 36, 166.
[44] S. Harkar, T. K. Razdan, E. S. Waight, Phytochemistry 1984, 23,
419–426.
[32] M. Bandopadhyay, S. B. Malik, T. R. Seshadri, Indian J. Chem.
1973, 11, 530–531.
[45] A. Schönberg, A. Sina, J. Am. Chem. Soc. 1950, 72, 4826–
[33] M. Bandopadhyay, S. B. Malik, T. R. Seshadri, Indian J. Chem.
1973, 11, 410–412.
4828.
[46] M. M. Abou-Elzahab, W. Adam, R. Saba-Moller, Liebigs Ann.
Chem. 1992, 731–733.
[34] T. M. El-Gogary, E. M. El-Gendy, Spectrochim. Acta A 2003, 59,
2635–2644.
[47] M. Bandopadhyay, T. R. Seshadri, Indian J. Chem. 1970, 8, 855–
[35] E. M. Elgendy, A. A. Ramadan, W. Fadaly, M. Hammouda, Boll.
Chim. Farm. 2002, 141, 188–191.
856.
[48] R. Abagyan, M. Totrov, D. Kuznetsov, J. Comput. Chem. 1994,
15, 488–506.
[36] H. Soreq, S. Seidman, Nat. Rev. Neurosci. 2001, 2, 294–302.
[37] S. Darvesh, D. A. Hopkins, C. Geula, Nat. Rev. Neurosci. 2003, 4,
131–138.
[49] M. Harel, D. M. Quinn, H. K. Nair, I. Silman, J. L. Sussman,
J. Am. Chem. Soc. 1996, 118, 2340–2346.
[38] E. Rivera-Becerril, P. Joseph-Nathan, V. M. Pérez-Álvarez,
[50] Y. Nicolet, O. Lockridge, P. Masson, J. C. Fontecilla-Camps,
M. S. Morales-Ríos, J. Med. Chem. 2008, 7, 5271–5284.
F. Nachon, J. Biol. Chem. 2003, 278, 41141–41147.
[39] D. C. Vellom, Biochemistry 1993, 32, 12–17.
[51] G. M. Sheldrick, Acta Crystallogr. A 2007, 64, 112–122.
[40] E. H. Oertli, R. C. Beier, G. W. Ivie, L. D. Rowe, Phytochemistry
1984, 23, 439–441.
[52] G. L. Ellman, K. D. Courtney, V. Andres, Jr., R. M. Feather-
Stone, Biochem. Pharmacol. 1961, 7, 88–95.
ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.archpharm.com