See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/261409059

The Flavonoid Derivative 2-(4′ Benzyloxyphenyl)-3-hydroxy-chromen-4-one

Protects Against Aβ42-Induced Neurodegeneration in Transgenic Drosophila:
Insights from In Silico and In Viv...

Article  in  Neurotoxicity Research · April 2014

DOI: 10.1007/s12640-014-9466-z · Source: PubMed

CITATIONS READS

19 845

6 authors, including:

Sandeep Singh Ruchi Gaur

Indian Scientific Education and Technology Foundation, India Indian Institute of Technology Kanpur
16 PUBLICATIONS   99 CITATIONS    15 PUBLICATIONS   157 CITATIONS   

SEE PROFILE SEE PROFILE

Akhil Kumar Roshan Fatima

Central Institute of Medicinal and Aromatic Plants National Center for Biological Sciences (NCBS), Bangalore, India
22 PUBLICATIONS   117 CITATIONS    10 PUBLICATIONS   159 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Hmrhl enhancer RNA View project

Sensors and Actuators B: Chemical View project

All content following this page was uploaded by Srikrishna Saripella on 09 July 2014.

The user has requested enhancement of the downloaded file.

Neurotox Res
DOI 10.1007/s12640-014-9466-z
ORIGINAL ARTICLE
The Flavonoid Derivative 2-(40 Benzyloxyphenyl)-3-hydroxy-
chromen-4-one Protects Against Ab42-Induced Neurodegeneration
in Transgenic Drosophila: Insights from In Silico and In Vivo
Studies
Sandeep Kumar Singh • Ruchi Gaur •
Akhil Kumar • Roshan Fatima • Lallan Mishra •
Saripella Srikrishna
Received: 15 June 2013 / Revised: 15 March 2014 / Accepted: 19 March 2014
Ó Springer Science+Business Media New York 2014
Abstract In the pathogenesis of Alzheimer’s disease
(AD), it is well established that the self-association of Ab
peptides into amyloid fibrils and/or plaque like aggregates
causes neurotoxicity. As there is no cure for AD till date,
identification of specific compounds that either inhibit the
formation of Ab-fibrils or help in the dissolution of already
formed amyloid plaques makes an appealing therapeutic
and preventive strategy in the development of drugs. In the
present study, four synthetic flavonoid derivatives (1, 2, 3
and 4) were examined for docking studies with Amyloid
beta (PDB Code: 1IYT) and Amyloid fibril (PDB Code:
2BEG). Of these, compound 1 and 4 were found to be
potential inhibitors, as supported by computational
molecular docking studies with adequate pharmacokinetic
properties. Compound 1 was further tested in vivo in
transgenic AD model of Drosophila. The disease causing
Electronic supplementary material The online version of this
article (doi:10.1007/s12640-014-9466-z) contains supplementary
material, which is available to authorized users.
S. K. Singh  S. Srikrishna (&)
Department of Biochemistry, Faculty of Science, Banaras Hindu
University, Varanasi 221 005, Uttar Pradesh, India
e-mail: sskrishna2000@yahoo.com; skrishna@bhu.ac.in
R. Gaur  L. Mishra
Department of Chemistry, Faculty of Science, Banaras Hindu
University, Varanasi 221 005, Uttar Pradesh, India
A. Kumar
Department of Bioinformatics, Indian Institute of Information
Technology, Allahabad 211012, Uttar Pradesh, India
R. Fatima
Neurobiology Laboratory, National Center for Biological
Sciences, Tata Institute of Fundamental Research,
Bangalore 560 065, Karnataka, India
human Ab42 peptide was expressed in the compound eye
by driving UAS-Ab42 with ey-GAL4, which caused severe
degeneration in eye tissues ranging from loss of bristles,
ommatidial holes to severe ommatidial disruption as
revealed by digital camera imaging and scanning electron
microscopy. When the Ab42 expressing larvae were grown
in medium containing Compound 1, *70 % rescue of the
rough eye phenotype was observed at 75 and 100 lM
concentrations. This is further corroborated by significant
reduction in amyloid plaques in eye imaginal disks of
compound 1 treated larvae as revealed by immuno-confo-
cal imaging studies. Further, rescue of locomotor deficit
and improved life span in compound 1 treated Ab flies also
confirm the neuroprotective activity of this compound.
Thus, our results support the neuroprotective efficacy of
compound 1 in preventing Ab42-induced neurotoxicity
in vivo and identify it as a future therapeutic agent against
AD.
Keywords Alzheimer’s disease  Ab42  Flavones 
Neuroprotection  Molecular docking  Confocal imaging 
Drosophila
Introduction
Alzheimer’s disease (Selkoe 1991) is the most common
form of neurodegenerative diseases, affecting more than 35
million people worldwide (Huang and Mucke 2012). It
remains the main cause of dementia and therapeutic pro-
gress has been hampered due to lack of proper under-
standing about its pathogenesis (Parihar and Hemnani
2004). However, it is mainly characterized by the forma-
tion of extracellular senile plaques (Amyloid-b) and
intracellular neurofibrillary tangles (Selkoe 2001).
123

Amyloid-b peptides aggregate to form amyloid plaques in
the brain of individuals suffering from sporadic and
familial AD (Hardy and Higgins 1992). Both in vivo and
in vitro experiments suggest that small, soluble species of
Ab oligomers are more toxic than the mature amyloid
fibrils comprising insoluble amyloid plaques (Larson and
Lesné 2012). Ab is generated by the proteolytic processing
of amyloid precursor protein via the amyloidogenic path-
way (Zhang et al. 2012) by the action of proteases; b- and
c-secretases (Thinakaran and Koo 2008). The physico-
chemical characteristics of Ab depend on the exact
cleavage site by the c- secretase. The shorter fragment,
Ab40, generated by the action of b- and c-secretases is less
prone to aggregation, while Ab42 is more hydrophobic and
aggregates to form plaques in many familial AD cases
(Haass and Selkoe 2007). According to amyloid cascade
hypothesis, formation of Ab aggregates mainly induces the
sequential events ultimately leading to the neuronal cell
death (Tjernberg 1999; Pimplikar 2009). Hence, the current
focus of therapeutic strategies is to mainly prevent Ab42-
aggregation or dissolution of already formed Ab-
aggregates.
Many drugs are being developed for the treatment of AD
(Mangialasche et al. 2010), and several different thera-
peutic strategies have been proposed to suppress AD
pathology (Hardy and Higgins 1992). For example, mod-
ulation of the activity of b- or c- secretases (Petit et al.
2001), neutralization of Ab aggregates by specific anti-
bodies (Kayed and Glabe 2006), cholesterol lowering drugs
(Puglielli et al. 2003), administration of antioxidant and
anti-inflammatory compounds (Commenges et al. 2000),
and Cu2? and Zn2? chelating agents (Hua et al. 2011;
Singh et al. 2013) are thought to reduce AD risks. Con-
sequently, there is an intensive search for drugs that could
either prevent the formation of Ab aggregates or dissolve
the existing Ab aggregates (Hirohata et al. 2007; Hemming
et al. 2007; Costa et al. 2008; Scherzer-Attali et al. 2010;
Frydman-Marom et al. 2011). Further, efforts have been
made to develop therapeutic strategies by successful
screening of Ab fibril inhibitors, as well as b sheet dis-
solving compounds in vitro (Kim and Hecht 2006; Esteras-
Chopo et al. 2008; Lukiw 2012). Such candidates might be
promising therapeutic targets for the prevention of AD in
future. In addition, small molecules which can effectively
prevent formation of Ab fibrils and stabilize its nontoxic
conformations have been considered as possible therapeu-
tics for AD treatment as the self-aggregation of Ab-peptide
is main culprit for the pathogenesis of AD. Previous studies
based on curcumin have shown that the two terminal
phenyl groups, substituted aromatic end groups that par-
ticipate in hydrogen bonding, are essential structural fea-
tures for the efficient inhibition of Ab aggregation (Reinke
and Gestwicki 2007). Several other studies also suggested
123
Neurotox Res
that a flat and planar molecule with substituted aromatic
groups and two aromatic end groups can be a good ligand
for Ab and may be cytoprotective (Nakagami et al. 2002;
Yang et al. 2005; Reinke and Gestwicki 2007). The small
polyphenols, curcumin (Yang et al. 2005), epigallocatechin
gallate (EGCG) (Ehrnhoefer et al. 2008), and resveratrol
(Han et al. 2004) have been found to effectively reduce Ab
toxicity in vitro and in vivo. The in vivo therapeutic and
neuroprotective potential of flavonoid containing com-
pounds have been studied in Drosophila (Pfleger et al.
2010; Hong et al. 2011; Kim et al. 2011) as well as in other
model systems (Jang et al. 2010; Yen et al. 2012).
Flavonoids are a kind of plant secondary metabolites,
which comprise polyphenolic compounds in their structure.
Over 4,000 flavonoids have been identified in nature, many
of which are found in vegetables, fruits, cereals, and bev-
erages (Manach et al. 2004). Flavonoids and their deriva-
tives have already been shown to exhibit antifibrillogenic
and cytoprotective properties (Ono et al. 2003; Yang et al.
2005; Ono et al. 2006). There are at least two types of
flavonoids; Monoflavonoids are planar with two terminal
phenyl rings, which include myricetin, morin, quercetin,
kaempferol, epicatechin, and apigenin (AP), while bifl-
avonoids consist of two interconnected monoflavonoids.
Cytoprotective effects of some biflavonoids and its related
flavones have been previously reported (Kang et al. 2005;
Lee et al. 2008), but their antifibrillogenic activity remains
unknown. Polyphenolic flavonoids are beneficial for health
in humans, since they exhibit anti-cancer (De Sousa et al.
2007), anti-microbial (Cushnie and Lamb 2011), and neu-
roprotective properties in oxidative stress (Inanami et al.
1998; Luo et al. 2002; Bahia et al. 2008). An imbalance
between antioxidants and reactive oxygen species (ROS)
results in oxidative stress, leading to cellular damage. This
oxidative stress has been linked to aging, cancer, athero-
sclerosis, ischemic injury, inflammation as well as neuronal
degenerative diseases like Alzheimer’s and Parkinson’s.
Flavonoids shield the neurons against these adverse effects
of ROS by protecting them against oxidative stress, sup-
pressing neuroinflammation, and improving cognitive
functions (Rice-Evans et al. 1996; Heim et al. 2002).
Though flavonoids exert their potential antioxidant activity
in vitro, while in vivo conditions, majority of them get
metabolized at the stage of absorption, resulting in the
alteration of their reduction potential against ROS. Apart
from being beneficial for health, recent findings suggested
that flavonoids have greater importance specially in neu-
roprotection in Alzheimer’s and Parkinson’s diseases (i) by
the modulation of intracellular signaling cascades which
control neuronal survival, death and differentiation; (ii) by
affecting gene expression; and (iii) by interacting with
mitochondria (Schroeter et al. 2002; Spencer et al. 2003).
Flavonoids exhibit especially low toxicity and hence been
Neurotox Res
considered as leading anti-AD compounds (Williams and
Spencer 2012). In recent years, some research groups have
attempted to optimize flavonoids’ chemical structures in
order to develop new drugs to treat AD (Sheng et al. 2009;
Li et al. 2013). One research group from China has been
working mainly on the synthesis of flavonoids and the
development of ChE inhibitors for many years (Luo et al.
2011a, b). They recently screened out compounds which
contain an N,N-dimethylated flavonoid moiety with bio-
logical activity. These compounds inhibit Ab self-aggre-
gation and have significant antioxidant activity.
In view of this, in the present study, we investigated the
effect of 4 different flavonoid derivatives in Ab42-induced
toxicity and fibrillogenesis and compared their efficacies
by using in silico and in vivo approaches. In silico analysis
revealed that these flavonoid derivatives are interacting
with different amino acid moieties of Ab and apparently
interfere with Ab self-aggregation. Hence, in the light of
above background and search of the drug having better
therapeutic potential in vivo, we have synthesized these
few promising flavonoid compounds 1, 2, 3, and 4 (Fig. 1)
Fig. 1 Chemical structure of
various flavonoid derivates used
in this study, compound 1 =
2-(40 -benzyloxy-phenyl)-3-
hydroxy-chromen-4-one;
2 = 2-(4-hydroxy-cyclohexa- O
1,4-dienyl)-6, 8a-dihydro-
chromen-4-one; 3 = 1,4-bis-(3-
hydroxy-4-oxo-4H-chromen-2-
yl)-benzene); and compound OH
4 = 1,7 bis[40 (4-oxo- 4H-
O
chrom -2yl)-phenyl] pentane
1 2
O O
O O
to find out their neuroprotective activity against amyloid
beta.
In order to understand the binding properties of these
compounds with the Ab42, first we have tested their
interacting efficiency with the Amyloid b (PDB Code:
1IYT) and Amyloid fibril (PDB Code: 2BEG), separately
using Molecular docking program, and then we selected
compound 1 (De Meyer et al. 1991) that showed best
docking results for further in vivo experiments. As com-
pound 1 showed maximum interaction with Ab peptide, the
neuroprotective effect of compound 1 alone was tested
against Ab-induced neurodegeneration in Drosophila AD
model system. Drosophila melanogaster is being exten-
sively used as powerful in vivo genetic model system to
study neurodegenerative disorders like Alzheimer’s disease
and Parkinson’s disease as it shows similarities in about
75 % of known human disease causing genes, and more
than 50 % of fly protein sequences have mammalian
homologs. The fly is also being used to study mechanisms
underlying aging and oxidative stress, immunity, diabetes,
cancer, as well as drug abuse. The UAS/GAL4 transgenic
OH
O
O
O
O O
OH HO
O O
3
O O
123

approach in Drosophila (Brand and Perrimon 1993) is an
excellent system to study the functions of desired genes/
gene products in tissue specific and spatio-temporal man-
ner. In order to check the expression of human Ab42 in the
Drosophila eyes, the UAS-Ab42 was driven under the
control of ey-Gal4 which led to dramatic reduction in eye
size, roughening of the eye, loss of bristles, and ommatidial
disruption, and in severe cases, there is a complete
degeneration of the eye. As in silico docking studies
showed a good interaction of compound 1 with Ab42, we
tested the neuroprotective efficacy of the compound 1
against Ab42 aggregation in Ab42 expressing eye tissues
in vivo. Interestingly, oral administration of compound 1 to
flies expressing Ab42 showed significant rescue of the eye
phenotypes, improved motor activity, and longevity, sug-
gesting that it could be a potential neuroprotective agent
against AD.
Results
In silico Docking Study of 1, 2, 3, and 4 Flavonoid
Derivatives with Protein Ab42 Amyloid and Amyloid
Fibrils
As b-amyloid and amyloid fibril do not have any particular
binding or active site, we used blind docking approach in
which large search space is created (search space covers
the whole protein) by docking algorithm to check all pos-
sible binding conformations. Docking of all flavonoid
derivatives with protein b-Amyloid (1IYT) and Amyloid
fibril (2BEG) was done. The result of molecular docking
between b-amyloid and amyloid fibril with all four com-
pounds is summarized in Table 1.
Docking between protein b-amyloid (1IYT) and com-
pound 4 showed MolDock Score of -88.48, Rerank Score of
-44.51, and Hydrogen bond Score -1.32. MolDock score of
compound 4 is the lowest among all tested compounds.
Docking of compound 4 with Amyloid fibril (2BEG), again
showed the second lowest MolDock score of -145.33, Re-
rank score of -105.92 but -0.45 hydrogen bond score.
Whereas in the case of docking of Protein b-Amyloid (1IYT)
with other three compounds, MolDock scores of compound 1
(-73.02) and compound 3 (-70.96) were found very close,
while compound 2 (-48.94) was not good as compared to 1
and 3. Hydrogen bond score in compound 1 (-5.01) was
much better than the compound 2 (0) and 3 (-3.55). Com-
pound 1 also showed very good binding affinity with 2BEG
-152.64 MolDock score, -122.28 Rerank score, and -1.55
hydrogen bond score, which was the best among all com-
pounds (see Table 1). Compounds 1 and 4 were found better
than 2 and 3 on the basis of hydrogen binding and docking
scores, so 1 and 4 were selected. We further moved to predict
123
Neurotox Res
the compatibility between 1 and 4 by checking pharmaco-
kinetics and ADMET properties of these compounds. We
found that compound 1 and 4 have better binding affinity
with 1IYT and 2BEG as revealed by docking result like
MolDock score, Hydrogen bond score, and other computa-
tional analysis. However, unfortunately compound 4 could
not be selected for further studies in vivo owing to solubility
and high molecular weight problems associated with it.
Pharmacokinetic Properties and Drug Likeness
of Compound 1 and 4
Many clinical trials are in progress to find suitable drug for
the treatment of AD including immunotherapy approaches
(Delrieu et al. 2012). Sometimes the drug trial leads to
failure during clinical trial phase due to the problems
associated with proper absorption, its distribution, metab-
olism, and toxicity. There is lot of time and monitory loss
due to these failures. So to avoid this problem, we pre-
dicted the Drug likeness and Pharmacokinetics properties
of the above mentioned two candidate drug molecules. For
drug likeness, the candidate molecules were predicted for
Lipinski rule of 5, CMC like rule, lead like rule MDDR like
rule, and WDI like rule (Table 2). ADME prediction
method involving Madin-Darby Canine Kidney Cell Per-
meability (MDCK), Caco-2cell permeability, Human
Intestinal Absorption (HIA), skin permeability, and blood–
brain barrier (BBB) penetration was also calculated for
these compounds by PreADMET web server (www.pre
admet.bmdrc.org), which is summarized in Table 3.
The results suggested that compound 1 is better than
compound 4 for the use of biological system or in vivo
study, as compound 4 violated the most of the rules which
lie under drug likeness and pharmacokinetics analysis
(ADME properties).
Compound 1 Shows High BBB Permeability
as Revealed by ADMET
ADMET prediction (Reddy et al. 2012) is very essential for
any compound which is going to be used as a drug in
future. In silico results revealed that compound 1 has high
blood brain permeability, and this is also supported in
terms of fast polar surface area (FPSA). The molecules
whose FPSA is greater than 140 Å exhibit poor perme-
ability. In case of blood brain barrier, the upper limit is
90 Å, and compound 1 shows 55.976 (Table 4), which is
optimal range and its optimal value for high permeability.
Aqueous solubility of compound 1 predicted ‘‘0’’ value
which indicates very good absorption. BBB level of a
compound varies from 0 to 4 (0—Very high, 1—High, 2—
Medium, 3—Low, 4—Undefined penetration level) as
indicated (Egan and Lauri 2002). Compounds that showed
 Table 1 MolDock scores (MDS), Rerank score (RS), and hydrogen bond score (HBS) observed after docking of 1IYT and 2BEG with the FLV compounds
Compounds Docking with 1IYT Docking with 2BEG
No. Structure MW MDS RS HBS MDS RS HBS
Neurotox Res

1 344 -73.02 -61.22 -5.01 -152.64 -122.28 -1.55

OH

O
2 OH 238 -48.94 -17.44 0 -86.61 -50.02 -0.32

O
3 400 -70.96 -53.48 -3.55 -132.83 -74.66 -0.0
O O

OH HO

O O
4 O O 544 -88.48 -44.51 -1.32 -145.33 -105.92 -0.45

O O

O O

MW molecular weight

123
 Neurotox Res

Table 2 Lipinski rule of 5, lead like rule, CMC like rule, MDDR like Intestinal absorption is defined as a percentage absorbed
rule, and WDI like rule for compound 1 and compound 4 rather than as a ratio of concentrations. According to
Compound 1 Compound 4 Egan and Lauri (2002), ADMET predicts the HIA after
oral administration. A well-absorbed compound is one
CMC like rule Qualified Not qualified
that is absorbed at least 90 % into the bloodstream in
Lead like rule Violated Violated humans. The ADME properties of compound 1 calcu-
Lipinski’s ‘‘Rule of five’’ Suitable Violated lated for anti-amyloid beta are shown in Fig. 2, which
WDI like rule In 90 % cutoff Out of 90 % cutoff show the ADMET_PSA_ 2D (polar surface area) versus
MDDR Like Mid-structure Mid-structure the ADMET_AlogP98 (the logarithm of the partition
WDI World Drug Index coefficient between n-octanol and water). The ellipses
in the figure define the regions where well-absorbed
Table 3 Depicting various pharmacokinetics properties predicted by
compounds are expected to be located. For intestinal
PreADMET server absorption, 95 and 99 % of well-absorbed compounds
are expected to fall within the ellipses colored in red
Absorption Compound 1 Compound 4 Range
and green, respectively. Similarly, for overcoming the
Human intestinal 96.23 97.90 Well-absorbed BBB, 95 and 99 % of well-absorbed compounds are
absorption (HIA, compounds expected to fall within the ellipses colored with
%) (70–100 %)
magenta and aqua, respectively. Compound 1 was found
In vitro Caco-2 cell 53.12 54.34 Middle
permeability (nm/s) permeability inside all four ellipses, therefore, satisfying the condi-
(4–70) tions for absorption by the intestines as well as BBB
In vitro MDCK cell 18.43 1.40 Low capacity.
permeability (nm/ permeability
sec) (if less than
25) Interaction of Compound 1 with Valine, Lysine,
Distribution Glutamic Acid and Phenylalanine Amino Acids of Ab
In vitro plasma 92.76 99.96 Chemicals Peptide
protein binding strongly
(%) bound (if Compound 1 (Fig. 3a) was further tested in vivo using
more than
90 %) Drosophila eye model system of AD. This computational
In vivo blood brain 0.15 0.099 Middle hypothesis proved correct in light of our experimental data
barrier absorption to from in vivo assays performed with Drosophila model
penetration CNS system. The molecular docking between compound 1 and
(c.brain/c.blood) (2.0–0.1)
1IYT gave insight regarding interaction of compound 1
with b-amyloid and halts the self-association of b-amyloid.
Detail analysis of docking result showed hydrogen bond
interaction between compound 1 and amino acid residues
Table 4 ADMET prediction of compound 1
of b-amyloid.
Properties Compound 1 Good–bad Amino acid Gln15 involved in the formation of two
ADMET_BBB_level 1 0–4: 1 = high
hydrogen bonds with oxygen at position 4, OH group at
ADMET_absorption_level 0 0–3: 3 = good
position 3, and Val12 with OH group at position 3 of
compound 1 (Fig. 3b). Also Lys16 forms two hydrogen
ADMET_solubility_level 2 0–6:2 = low
bonds with the OH group at position 3 and oxygen atom at
ADMET_PPB_level 2 0–2: 2 C 95 %
the 40 position of the ligand. Aromatic cyclic structure is
ADMET_PSA_2D 55.976 \140
well fitted in the hydrophobic pocket created by amino acid
ADMET_BBB_Level = 1 Phe19 and Phe20 and forms strong hydrophobic interac-
tion. Val12, Gln15, and Lys16 are the major residues that
are involved in Hydrogen bonding interactions with com-
penetration to BBB can be used as CNS drugs. CNS active pound 1 by different energy score of -24.93, -15.34, and
drugs can cross BBB by several mechanisms, including -27.33, respectively. Phe19 and Phe20 also showed good
passive diffusion. This assumption formed the basis for energy score of -11.28 and -6.98 which confirmed the
developing earlier BBB prediction models (Garg and hydrophobic interaction with compound 1. Ser8, Gln11,
Verma 2005; Refsgaard et al. 2005). His13, Gly9, and Asp23 also play some important role
The HIA level of a compound varies from 0 to 3 (0— during interaction by adding some value to energy score,
Good absorption, 1—Moderate, 2—Low, 3—Very low). and the detail values are given in Table 5. These amino

123
Neurotox Res

Fig. 2 ADMET plot.

ADMET_PSA_2D (polar
surface area) versus
ADMET_AlogP98 (the
logarithm of the partition
coefficient between n-octanol
and water). ADME absorption,
distribution metabolism
excretion properties

Fig. 3 a Showing the 2D

structure of compound 1 and
numbering of atoms and
b showing predicted binding A
mode of the compound 1 with
protein b-Amyloid (PDB 1IYT)
and interacting amino acids
residue (Glu11, Val12, Gln15,
Lys16, Phe19, and Phe20).
H-bonds between Amino acid
and compound 1 are represented
as green dashed lines (Color
figure online)

123
 Neurotox Res

Table 5 Energy scores of Amino acid residue (AA) of amyloid beta which are contributing the energy score (ES)
different amino acid residues of
b-amyloid (1IYT) interacting AA Lys16 Val12 Gln15 Phe19 Ser8 Glu11 Phe20 His13 Gly9 Asp23
with compound 1
ES -27.33 -24.93 -15.34 -11.28 -9.41 -7.09 -6.98 -0.98 -0.64 -0.62

Fig. 4 Channel formation in Ab-fibril where compound 1 is inter- interacting with amino acid residues of A, B, C, D, and E chains of
acting. 3D structure of Amyloid fibril (PDB ID: 2BEG), showing the A-beta fibril (a). Surface view of interaction of Ab-fibril with
space between b-sheet structure of A-beta fribril where compound 1 compound 1 through channel formation (b)

acids are core residue of the Ab-peptide which directly
interacts with the compound 1 through hydrogen bonding.
2BEG is fibril structure made up of a chain of beta
amyloid aligned side by side to form b-sheet like structure
(A, B, C, D, E) in Fig. 4a. Compound 1 fits into the pocket
by forming a channel surrounded by Leu17, Val18, and
Phe19 residues of all the chains of Ab fibril (A–E)
(Fig. 4a). Channel formation is clearly seen by surface
view of Ab-fibril interaction with compound 1 (Fig. 4b).
Docking study between 2BEG and compound 1 revealed

123
Neurotox Res

Fig. 5 Figure shows predicted

binding mode of the compound
1 with protein b-Amyloid fibril
(2BEG) and interacting amino
acids residue. H-bonds between
Amino acid and compound 1 are
represented as green dashed
lines. A, B, C, D, and E
represent beta amyloid chains of
Amyloid Fibrils. Red-, blue-,
and gray-colored balls are
representing oxygen, nitrogen
and carbon atoms, respectively
(Color figure online)

that it forms hydrogen bonds with Val18 and Phe19 resi-
dues of amyloid fibril. The Val 18 of chain C and D forms
hydrogen bonds with OH and oxo group of compound 1
with H-bond distance of 3.40 and 2.49 Å, respectively.
Whereas Phe19 of chain D forms hydrogen bond with OH
group of compound 1 with H-bond distance of 3.01 Å (as
shown in Fig. 5). In other way, the active cavity where the
amino acid residues of amyloid b fibril interacts with
compound 1 and form hydrogen bond with different amino
acid moiety is shown in (Fig. S1, Supporting Information).
Concentration Dependent Effect of Compound 1
on Eye Phenotypes
To determine the neuroprotective effect of compound 1 on
Ab toxicity, we cultured the Ab42 expressing larvae till
adult stage in the media containing compound 1 at three
different concentrations (50, 75, and 100 lM); larvae fed
on medium without compound 1 were treated as control.
The Ab42 expressing flies grown on normal medium
showed mild to severe eye phenotypes (Fig. 6). Mild
phenotypes were characterized by partial ommatidial dis-
ruption with disorganized bristles, while severe eye phe-
notypes show highly disrupted ommatidia with complete
loss of bristles. Moreover, severe eye phenotype is pre-
dominant over mild phenotype in Ab42 expressing flies.
Oral administration of compound 1 dramatically rescues
the Ab42-induced eye phenotypes. The mean percent of
flies showing rescue in eye phenotypes increases with
increasing concentration of compound 1, from 50 to
100 lM, as summarized in Fig. 6.
Scanning Electron Microscopic Examination of Eye
Phenotype
SEM examination of eyes of Ab42 expressing flies revealed
ultra structural details of ommatidia and bristles at various
stages of degeneration and their rescue by the treatment
with compound 1. While wild type and ey-Gal4 flies
exhibit typical pattern of synchronized ommatidial
arrangement in their compound eyes (Fig. 7a, b, respec-
tively), severe eye degeneration with reduced eye size is
clearly evident in Ab42 expressing eye tissues (Fig. 7c).
High magnification images showed severe degeneration of
ommatidia and complete loss of bristles in such eyes
(Fig. 7h). Rescue in eye phenotype was found in flies
treated with 75 and 100 lM concentrations of compound 1
(Fig. 7d, e), which was clearly evident in high resolution
images of SEM (Fig. 7i, j).
Compound 1 Rescues Locomotor Deficits of Ab
Expressing Flies
In order to assess the locomotor activity of Ab expressing
flies grown in normal media versus compound 1 treated
food media, we have cultured elav-Gal4 driven Ab flies in

123
 Neurotox Res

Food Supplemented with

A Normal food B Compound 1 (50 μM)
75
75
Normal Normal

% of flies ± SD
Mild Mild
% of flies ± SD

50 Severe
50 Severe

25
25

0
0 Normal Mild Severe
Normal Mild Severe H32.12
Eye Phenotype of UAS-A β /ey-GAL4
Eye Phenotype of UAS-Aβ H32.12 /ey-GAL4

Food supplemented with Food Supplemented with

C Compound 1 (75 μM) D Compound 1 (100 μM)
75 75
Normal
Normal
Mild
% of flies ± SD

Mild

% of flies ± SD
50 Severe
50 Severe

25
25

0
Normal Mild Severe 0
Normal Mild Severe
Eye Phenotype of UAS-Aβ H32.12 /ey-GAL4
Eye Phenotype of UAS-A β H32.12 /ey-GAL4

Fig. 6 Histogram showing mean percentage of UAS-Ab42/ey-GAL4 and 100 lM (d) of compound 1. p value is \0.0001. The number of
flies showing normal, mild, and severe eye phenotypes, cultured in flies (n = 100) for a, b, c, and d on y axis expressed as % of flies
normal food (a), and food supplemented with 50 lM (b), 75 lM (c), against eye phenotype in each case

both media. Ab42 expressing flies grown on normal food
media showed severe locomotor dysfunction, while same
flies grown on food supplemented with compound 1 with
75 and 100 lM showed very good recovery (Fig. 8).
Control flies (Gal4-elavc155) show about 90–95 %
climbing behavior. In contrast flies, expressing Ab42 shows
rapid and much earlier decline in climbing behavior from
day 5 to day 10; only 10–20 % of flies were able to climb.
But there was about 65–70 % recovery in climbing ability
of flies fed on compound 1 supplemented food medium
(Fig. 8).
Compound 1 Improves Life Span of Alzheimer’s Flies
The effect of compound 1 on fly expressing the Ab42
peptide in fly neural brain tissue was assessed. We per-
formed longevity assays on these flies expressing Ab42
peptides and found that expression of the Ab42 peptide
caused a marked reduction in survival as compared with
flies without a transgene. Survival assays were repeated
three times in each category of flies, and the Kaplan–Meier
analysis was used to analyze the significance of the data. A
quantitative measure of the effects of compound 1 was
provided by determining the longevity of flies expressing
transgene driven by the pan-neuronal elav-GAL4. Flies
expressing Ab42 peptide reared in normal food having a
significantly shorter life span (Fig. 9, red line) compared
with controls (Gal4-elavc155) (Fig. 9, blue line). Mean
survival 23 versus 47 days, n = 100 for both, p [ 0.001. In
other way, the increased life span was observed in case of
compound 1 supplemented Ab42 expressing flies of 75
(Fig. 9, green line) and 100 lM (Fig. 9, purple line)
compared with flies reared in normal food. Mean survival
of 34, 35 versus 23 days, n = 100 for all groups,
p [ 0.001. Results show a significant difference between
flies expressing the Ab42 transgene grown on regular
medium versus medium supplemented with compound 1
(p [ 0.001).

123
Neurotox Res
Fig. 7 Scanning electron micrographs showing eye degeneration and
rescue in compound 1 treated Ab42 flies. Upper panel showing the
reduction in size of eye in UAS- Ab42/ey-Gal4 in normal food (c) as
compare to the wild type (a) and ey-Gal4 (b) as controls. Rescue in
Fig. 8 Graph showing the effect of compound 1 on the motor activity
(climbing ability) of Ab42 expressing flies. Four categories (elavC155
Gal4, Ab flies grown in normal medium, Ab flies grown in 75 and
100 lM) of flies each having six vials with ten flies in each were
observed for their climbing behavior in each case. Results show the
percent of flies climbing up to the top of the vials in 18 s, during
course of 5–10 days. p \ 0.0001
Compound 1 Rescues in Vivo Amyloid b Aggregation
in Eye Imaginal Tissues
In order to show the inhibitory effect of compound 1 on Ab
aggregation in situ, immunofluorescence staining of Ab42
the eye phenotype was observed after the treatment with 75 lM
(d) and 100 lM (e) of compound 1. Magnified images are shown in
lower panel (f–j)
was carried out using anti-b-amyloid specific antibody in
intact eye imaginal disks of Ab expressing late third instar
larvae with and without compound 1 treatment. Eye
imaginal tissues of Ab expressing larvae clearly showed b-
amyloid plaques (green in Fig. 10). Phalloidin staining that
specifically detects F-actin is also done in the same tissues
to see cortical acting organization both in treated as well as
untreated tissues (red in Fig. 10). Confocal imaging clearly
revealed the b-amyloid plaques formation in the omma-
tidial region of eye imaginal disk tissues (green foci in I, J,
L of Fig. 10), while F-actin also highly disorganized in
same tissues (red foci in J, K, L of Fig. 10). The compound
1 treated eye imaginal disks did not show any foci corre-
sponding to b-amyloid plaques (green in panel A), and
intact cortical organization of F-actin is clearly evident in
compound 1 treated eye tissues (red in panel B). The
uniform ubiquitous pattern of staining seen in compound 1
treated imaginal eye tissues also suggests that the expres-
sion of Ab42 is happening, but the presence of compound 1
apparently prevents aggregation of these monomeric pep-
tides (compare Fig. 10a with e). Cortical actin organization
plays an important role for ommatidial differentiation
during eye development. In Ab42 expressing flies, acting is
highly disorganized as revealed by Phalloiding staining
(compare Fig. 10b with f). These immuno-confocal imag-
ing results clearly reflect that the rescues of eye phenotypes
are indeed due to prevention of Amyloid b aggregation by
123

Fig. 9 Compound 1
supplementation extended the
average life span of Ab42
expressing flies. Gal4-elavc155
flies (blue square) and Ab42
driven flies fed on the normal
food (red diamond),
experimental diets
supplemented with 75 lM
(green triangle) and 100 lM
(purple line) compound 1 at
29 °C. The significance of the
difference between survival
curves (n = 100 flies per group)
was analyzed using the Kaplan–
Meier log-rank statistical test
(*p \ 0.001) (Color figure
online)
compound 1 in eye imaginal tissues during larval to pupal
developmental stages of the eye.
Discussion
Flavonoid derivatives belong to the category of polyphe-
nolic compounds having potential cytoprotective activity
as well as antioxidant property (Heim et al. 2002; Reinke
and Gestwicki 2007). The monoflavonoids reduce Ab
toxicity by abrogating the activation of caspase-3 and the
release of cytochrome c (Wang et al. 2001). The cytopro-
tective effect of monoflavonoids is also linked with its
antifibrillogenic and antioxidative activity. While biflavo-
noids have also been shown to reduce Ab toxicity and
oxidative stress (Shin et al. 2006; Lee et al. 2008), but the
underlying molecular mechanism by which these flavo-
noids attenuate Ab fibrillogenesis and toxicity remains
largely unexplored. Thus, the present study was aimed to
understand the mode of interaction between flavonoid
derivatives and amyloid b peptides through in silico and
in vivo analyses.
Docking studies revealed the possible mode of interac-
tion of amino acid residues of b-amyloid (1IYT) and
amyloid beta fibril (2BEG) with flavonoids through
hydrogen bonding. The compound 1 in the active site was
compact with the ladder-like structure as Val12, Gln15,
and Lys16 that are mainly involved in the hydrogen bond
formation at distances 3.31 Å (between O of Val12 and
3-OH group of compound 1), 2.78 Å (between N of Gln15
123
Neurotox Res
and 3-OH group of compound 1), 3.05 Å (between N of
Gln15 and 4-C=O group of compound 1), 3.04 Å (between
N of Lys16 and 3-OH group of compound 1), and 3.16 Å
(between N of Lys16 and O of compound 1 at 40 position)
with compound 1 as shown in Fig. 11.
Among the above amino acids, Lys16 showed better
energy score (-27.33) followed by Val12 (-24.93), Gln15
(-15.34), and so on (Table 5), while other amino acid
residues like Phe19 and Phe20 are also involved in
hydrophobic interaction between b-amyloid and compound
1, which is also very important. The docking study in case
of b-amyloid with compound 1 suggests that these amino
acid residues (Val12, Gln15, Lys16, Glu11 and Phe19) are
mainly responsible for interaction with compound 1. It is
important to note here that the Lys-16 appears to be a key
factor in promoting Ab toxicity, which is in agreement
with Sinha et al. 2012, who demonstrated the reduced
cytotoxicity of Ab when Lys-16 was substituted by Alanine
residue. On the other hand, as shown in Fig. 4, the amino
acid residues Leu17, Val18, and Phe19 interacting with
compound 1 through a channel formation. Among all res-
idues, Val18 and Phe19 are mainly involved in hydrogen
bonding (Fig. 5). The other residues interacting with the
compound 1 are Phe19D, Phe19B, Leu17D, Leu17A, and
Leu17C with the energy score of -26.7765, -18.7381,
-17.1096, -12.7118, and -11.1426, respectively (Table 6).
These residues are critically involved in the interaction of
five chains of b-amyloid, which play a major role in sta-
bility of amyloid fibril through b-sheet formation. Specific
interaction of compound 1 with these amino acids could,
Neurotox Res
Fig. 10 Confocal images are showing b-amyloid (green), F-actin
(red), and DNA (blue) in Ab42 expressing eye imaginal disks of late
third instar larvae (a–l). Note the presence of typical b-amyloid
plaque aggregates in ey-Gal4 driven Ab42 imaginal disks (green foci
pointed by white arrows in i, j and l) and the absence of these
aggregates in compound 1 treated eye imaginal disk (green in a, d).
therefore, potentially hinder the formation of stable fibril
structure.
Usually, there are two molecular frameworks for the
binding of small molecules to amyloid fibrils. The first one
is relatively well-defined site-specific binders that form
networks of interactions with sequence motifs. The second
one, far less well defined at this point, aromatic compounds
that bind to tube-like cavities between b-sheets. As
reported by Landau et al. (2011), Orange-G is bound
internally to the steric zipper of KLVFFA residues (16–21)
of Ab. While Leu17, Val18, Phe19, and Phe20 form the
Pannels i–l are magnified views of e–h, respectively. Actin organi-
zation is also highly disrupted in Ab expressing tissues (red in f, h, j,
k, and l). Yellow arrows pointing toward cortical actin organization
(panel B) and disorganized actin foci (panel f). Scale bar represents
50 lm (Color figure online)
hydrophobic core and provide 80 % hydrophobic interac-
tion with the orange-G. Compound 1 appears to follow the
similar nature of hydrophobic interaction (Fig. 4a). Landau
et al. also concluded that various polyphenolic compounds
have propensity to bind with amyloid-forming sequences
because of the formation of a cylindrical, partially apolar
cavity between the pairs of b-sheets forming the fibers.
Drugs like benzodiazepines and anesthetics bind through
cavity formation, explaining some of the altered pharma-
cokinetic properties and increased sensitivity detected in
elderly (Landau et al. 2011). Further evidence of
123
 Neurotox Res

Fig. 11 The interaction

between compound 1 and amino
acid residues of Ab forming
H-bond with distance in Å
between the oxygen/nitrogens
of Lys 16, Gln 15 amino acids,
and oxygen of Val 12

Table 6 Energy scores of AA ES AA ES AA ES AA ES

different amino acid residues of
amyloid fibrils (2BEG) Phe19A -5.82973 Val18A -3.88863 Leu17A -12.7118 Val40A 0
interacting with compound 1
Phe19B -18.7381 Val18B -7.67205 Leu17B -8.07806 Val40B -2.35734
Phe19C -9.84869 Val18C -9.81247 Leu17C -11.1426 Val40C -4.04689
Phe19D -26.7765 Val18D -5.67099 Leu17D -17.1096 Val40D -4.88641
Phe19E -7.21843 Val18E -1.00168 Leu17E -6.88319 Val40E -3.23453

interaction of hydrophobic residues of Amyloid fibrils with
flavonoids has been shown by (Choi et al. 2008) and
(Lemkul and Bevan 2010) using myricetin and morin
respectively that inter into the hydrophobic core and
destabilize the amyloid fibril beta. The space between
U-shaped conformations of Amyloid fibril provides suit-
able binding site for a flat-shaped myricetin molecule
(Rochet and Lansbury 2000).
In silico analysis also further predicted the possible
hydrogen bonding interaction between the other ligands
and protein. Compound 2 forms hydrogen bonds through
its hydroxyl group and oxo linkage with Ser8 and Gln15
residues, respectively (Fig. S2, Supporting Information).
But after evaluation using software by optimizing and
protein hydrogen position, the net hydrogen bond energy
score is zero. Compound 3 forms three hydrogen bonds
through its carbonyl group with residue Glu22 and Asp23
of b-amyloid (Fig. S3, Supporting Information), while
compound 4 forms only one hydrogen bond between its
oxo group and Gln15 amino acid residue of b-amyloid
(Fig. S4, Supporting Information).
The above in silico studies were further supported by the
in vivo neuroprotective activity of compound 1 against
Ab42 expressing transgenic Drosophila. The Ab42
expressing flies administered with compound 1 showed
tremendous rescue in eye degeneration phenotypes as
evident by increased number of flies with normal eyes
(Fig. 6). The scanning electron microscopy revealed com-
plete ommatidial degeneration of the eyes in Ab42
expressing flies, while significant rescue of ommatidial
degeneration was evident in compound 1 treated one
(Fig. 7). It is worth mentioning here that earlier studies
reported a pattern of increased roughness in eyes due to the
disruption of bristles in Ab42 expressing flies when driven
by GMR-Gal4 (Hong et al. 2011). However, in our case,
when driven with ey-Gal4, we found different degrees of
eye degenerations ranging from rough eyes (mild) to
severely degenerated eyes. The in vivo neuroprotective
activity was reflected in the form of rescue in severe eye
phenotypes, and the degree of rescue increases from 50 to
100 lM concentrations of compound 1. Of the above
concentrations, 100 lM was very effective against Ab42

123
Neurotox Res
induced toxicity in vivo, since it rescues about 70 % of the
neurodegenerative eye phenotypes (Fig. 6). We have also
tested at 150 lM concentration, but it was toxic to larval as
well as adult stages (Data not shown). So it is determined
that 100 lM is the best suitable dose for in vivo treatment.
Further confirmation of the in vivo neuroprotective activity
of compound 1 was done by observing improved locomotor
activity (Fig. 8) and increased longevity of Ab expressing
flies driven by pan-neuronal elav-GAL4 (Fig. 9) supple-
mented with 100 lM concentration of compound 1.
As per earlier reports, crude extracts derived from
plants showed some positive effects on the life span and
neuroprotective activity in Drosophila model system (Kim
et al. 2011). In other studies, natural herbal extracts
showed improvement in the eye phenotypes in Drosophila
model of taupathy (Pfleger et al. 2010), which could
possibly be due to the presence of flavonoid derivatives in
such extracts. Hong et al. (2011) showed the significant
neuroprotective effect of Chinese medicine having crude
extract of 15 herbs in Drosophila model of Alzheimer’s
disease. They evaluated the neuroprotective effect of the
Chinese medicine and suggested that this herbal cocktail
somehow protects the cells from Ab42-induced cytotoxic-
ity and helps proper neuronal functioning and develop-
ment. But in all these cases of natural products, no specific
flavone has been isolated or suggested to have neuropro-
tective effect per se. However, in the same context, our
results strongly suggest that the compound 1 has a better
neuroprotective effect as evident by Immuno-confocal
imaging studies (Fig. 10) exhibiting rescue in formation of
amyloid beta plaques in the eye imaginal tissues of third
instar larvae thereby rescuing neurodegenerative pheno-
types in adult eyes as revealed by scanning electron
microscopy (Fig. 7).
Over all our studies demonstrate that the usage of
compound 1, which was selected through the docking
studies, has advantages over other herbal compounds and
can fulfill the criteria of an ideal drug. Moreover, com-
pound 1 has the potential to cross the blood brain barrier
which is the key feature required for formulation of a
suitable drug. Results from our studies, therefore, clearly
indicate that the rescue in abnormal eye phenotypes of AD
was indeed due to inhibition of Amyloid aggregates by
compound 1 in the eye imaginal tissues during the transi-
tion period between late third instar larvae to pupal
development, which is the critical developmental period
where ommatidial differentiation into adult eyes takes
place. Further molecular studies would be required to
identify the exact mode of binding of compound 1 with the
Ab42 peptide, and how the flavones bring about the con-
formational changes in Ab42 fibrils to prevent its aggre-
gation, which may open the field for the development of a
novel class of therapeutics against Alzheimer’s disease.
Conclusion
In conclusion, four different flavonoid derivatives were
designed, synthesized, and subjected to docking studies
with Ab-peptide and Ab-fibril. Among all compound 1,
which showed best activity toward inhibition of self-med-
iated Ab42 aggregation, was selected for in vivo study.
Overall, compound 1 displayed strong anti-Ab activity by
rescuing eye degeneration phenotypes, improvement in
motor function, and increased longevity in Drosophila
model of Amyloidogenesis. Thus, this type of flavonoid
derivatives might provide a useful template for the devel-
opment of new anti-AD agents with neuroprotective
activity.
Materials and Methods
Starting materials were purchased from Sigma–Aldrich and
used without further purification. The Infrared, UV–Vis,
and luminescence spectra were recorded on VARIAN 3100
FTIR, JASCO V-630, and Perkin-Elmer LS-45 spectro-
photometer, respectively. However, 1H NMR spectra were
recorded on JEOL AL 300 MHz spectrometer using TMS
as internal reference. A schematic presentation of the
synthesis of different compounds used in the present study
is depicted in Scheme 1 (Fig. 12).
Preparation of Compounds 1, 2 and 3
2-(40 Benzyloxyphenyl)-3-hydroxy-chromen-4-one (com-
pound 1), 2-(40 -hydroxy-phenyl)-chromen-4-one (com-
pound 2), and 1,4-bis-(3-hydroxy-4-oxo-4H-chromen-2-
yl)-benzene (compound 3) were synthesized and charac-
terized using reported method (De Meyer et al. 1991;
Cabrera et al. 2007; Svechkarev et al. 2012).
Preparation of Compound 4 (1,7 Bis[40 (4-oxo-4H-
chrom-2yl)-phenyl]pentane)
B (0.476 g, 2 mmol) and 1, 5 di bromo pentane (0.220 g
1 mmol) were dissolved in 10 mL DMF in the presence of
K2CO3 (0.525 g). The reaction mixture was refluxed with
stirring for 24 h. The reaction mixture was then diluted
with distilled water (25 mL), and organic layer was
extracted with chloroform. The extract was washed with
brine and dried over anhydrous magnesium sulfate. The
solvent was removed and recrystallized by methanol and
ethanol. Yield: 0.320 g 58 %, M.P. 164–165 °C, IR (mmax/
cm-1): 2939(m) t(CH, Ph), 1643 (CO), 1254 (O-Ph), 1H
NMR (CDCl3, d ppm): 8.23 (d, J = 7.2 Hz, 2H, –Ar), 7.89
(d, 8.4 Hz, 4H, –Ar) 7.68 (d, J = 7.8 Hz, 2H, Ar), 7.55 (d,
J = Hz, 2H, Ar), 7.43 (t, 2H, Ar) 7.50 (m, 2H; Ar), 7.82(s,
123
 Neurotox Res

Fig. 12 Scheme of synthesis of

compounds 1, 2, 3, and 4 O

OH
O
2-(4'-Benzyloxy-phenyl)-3-hydroxy-chromen-4-one

1 NaOH/H2O2

O
OH OH
NaOH / HCl
O
+
CH3 r.t.
OHC
O O
3-(4'-benzyloxyphenyl)-1-(2-hydroxylphenyl)-prop-2-en-1-one

I2 / DMSO
Reflux

OH O

O AcOH /HCl O
"

O O
2-(4'-Hydroxy-phenyl)-chromen-4-one 2-(4'-Benzyloxy-phenyl)-chromen-4-one
2
Br-(CH2)5-Br

DMF / K2CO3

O O
4'
O O

O O
1,7 Bis[ 4'-(4-oxo-4H-chrom-2yl)-phenyl]pentane

CHO
HO 50% NaOH
+
HCl
HO O OH
CHO O O
1-(2-Hydroxy-phenyl)-3-{4-[3-(2-hydroxy-phenyl)-3-oxo-
propenyl]-phenyl}-propenone

30%H2O2 /NaOH

O O

OH HO
O O
1,4-bis-(3-hydroxy-4-oxo-4H-chromen-2-yl)-benzene)

123
Neurotox Res
4H, Ar),7.03 (d, J = 7.4 Hz, 4H, Ar) 6.74 (s, 2H, CH–CO),
4.11 (t, 4H, CH2), 1.96 (t, 4H, CH2), 1.59 (m, 2H, CH2);
UV–Vis (DMF, 10-4 M): kmax (nm) (emax 9 104
M-1 cm-1) 260 (0.516), 316 (1.22). kemission 389 nm at
kexcitation 316 nm FAB-MS: m/z 544 (M?).
Preparation of Ligand
We modeled the 3D optimized structures of given flavo-
noid derivates with molecular modeling software package
ACD chem. sketch (http://www.acdlabs.com) and saved
into .mol file. These flavonoid derivatives then converted
into 3D (pdb) format by using Open Bable (O’Boyle et al.
2011). Bonds, bond orders, hybridization, charges, etc.,
were assigned to these flavonoid derived compounds by
default parameter using Molegro Virtual Docker software
(MVD) (http://www.preadmet.bmdrc.org).
Preparation of Proteins
The structures of Amyloid beta (PDB Code: 1IYT) and
Amyloid fibril (PDB Code: 2BEG) were downloaded from
the Protein Data Bank (PDB). 1IYT and 2BEG both are the
3D NMR Solution structure of the b-amyloid peptide and
Amyloid fibril, respectively. With 1IYT having total 10
conformations, and for the docking studies, chain A is
used. 2BEG is fibril structure composed of five chains of
beta amyloid namely A, B, C, D, and E. All chains in
2BEG are intact and used in the docking studies.
Docking Studies
MVD (Molegro Virtual Docker software) was used to
calculate the interaction energies between ligands and
macromolecular systems from the 3D structures of the
protein and ligands. The algorithm used was the MolDock
Score, an adaptation of the Differential Evolution (DE)
algorithm (Thomsen and Christensen 2006). In previous
docking studies from different literature, MVD H yielded
higher docking accuracy than other docking programs; the
accuracies were MVD, 87 %; Glide, 82 %; Surflex, 75 %
and FlexX, 58 % (Thomsen and Christensen 2006).
In the molecular docking of protein ligand, we used
MolDock scoring function and Moldock SE search algo-
rithm for the optimal solution. Other parameters like
number of run, max Iterations, and max Population size
were set as default. This molecular docking protocol gen-
erate five best predicted pose for each flavonoid derived
compound with Moldock score, Rerank score, and
Hydrogen bond score. The docked conformations or pose
with the minimum MolDock score values is the optimal
pose. The docked conformation further analyzed on the
basis of the Re-Rank score function. It is believed that the
Re-Ranking score function is generally more reliable than
the MolDock score function at selecting the best solution
among multiple solutions derived from the same ligand
(http://preadmet.bmdrc.org/).
During docking, MVD returns multiple poses (maxi-
mum 300 for each run) representing different potential
binding modes. In MVD, poses found during the docking
run were automatically clustered through MVD default
clustering algorithm using the RMSD threshold (1.00)
criteria, and only the lowest energy representative pose
from each cluster was returned when the docking run is
completed. Poses which were generated further sorted by
rerank score. The Rerank score uses a weighted combina-
tion of the terms used by the MolDock score mixed with a
few addition terms (the Rerank Score includes the Steric
(by LJ12-6), which is Lennard–Jones approximations to the
steric energy. The reranking score function is determining
the best pose among several poses originating from the
same ligand. In the docking run, best five poses have the
similar mode of binding, and all are in same cavity. To
validate the predicted pose, we used another docking
software Autodock4.2. The lowest energy conformation,
i.e., best pose predicted by autodock, shows similar binding
mode with Ab and Ab fibrils (Figs. S5a, b, S6; Tables S1
and S2, Supporting Information).
Drosophila Strains and Fly Husbandry
The Drosophila strains used were, wild type (Oregon R),
ey-GAL4/CyO, which direct the transgene expression
specifically in eye tissue, were obtained from the Bloom-
ington Drosophila Stock Center at Indiana University and
w; UAS-AbH32.12/CyO with Ab42 transgene (Finelli et al.
2004) under the control of UAS was generously gifted by
Dr. M. Konsolaki (Department of Genetics, Rutgers, The
state University of New Jersey, USA). Flies were main-
tained at 24 ± 1 °C on standard cornmeal agar medium.
Generation of Ab42 Expressing Flies
The UAS/GAL4 system (Brand and Perrimon 1993) was
used for the over-expression of UAS-Ab42 transgene in
Drosophila using ey-Gal4 specific for eye tissues. The ey-
GAL4/CyO male flies were crossed to virgin females of
UAS-Ab42/CyO. From F1 progeny, the UAS-Ab42
expressing flies (UAS-Ab42/eyGal4) were selected for the
AD phenotypic observations.
Drug Treatment
The UAS-Ab42/ey-GAL4 larvae were cultured in com-
pound 1 supplemented food media at 50, 75, and 100 lM
concentrations in 0.1 % DMSO to see the neuroprotective
123

activity of this compound on AD eye phenotypes. For
control, the UAS-Ab42/ey-GAL4 larvae were grown in
normal food medium containing 0.1 % DMSO. The UAS-
Ab42/ey-Gal4 flies showing non-curly wings were selected
from F1 generation and observed for eye degeneration
phenotypes under stereo binocular microscope. Data on eye
phenotypes were collected from both normal as well as
drug supplemented media cases in triplicates (n = 100 for
each set).
Scanning Electron Microscopy (SEM) of Drosophila
Compound Eye
For the SEM analysis, we followed the method of Tanya
Wolff (Wolff 2011) with minor modifications. About 8–10
flies of each group were etherized, fixed in 1.5 ml of fix-
ative (0.1 M PBS, 25 % Glutaraldehyde, 1–2 drops of
0.2 % Tween20 and dH20), and kept at 4 °C for overnight
on a shaker followed by washing once with 1XPBS and
twice with distilled water for 15 min each. After removing
water, dehydration of tissues was done by treating them
gradually with 25, 50, 75, and 100 % alcohol for about 3 h.
Finally tissues were washed with 100 % alcohol three
times for 15 min each followed by CPD (Critical point
drying). After CPD, the tissues were coated with gold, and
serial focal plane images of eye tissues were captured by
SEM (Hitachi S-3400N).
Climbing Assay for Motor Activity
This assay has been performed to test the locomotor
activity of Ab42 expressing flies. Test vials of 10-cm length
containing each of the four classes {elavC155, Ab-driven in
normal food, compound 1 (75 and 100 lM) treated Ab
flies} were taken for the climbing assay experiment. Ten
flies were placed in each plastic vial and gently tapped to
the bottom. The number of flies at the top of the vial was
counted after 18 s. The percentage of flies climbed up to
the top of the vial was calculated over time. Six indepen-
dent test repeats have been performed for each class. The
climbing assay was repeated three times. Statistical ana-
lysis was done by using Prism3, one-way ANOVA.
Longevity Assay
The measurement of longevity was conducted as follows.
In ten flies, each expressing Ab42 peptide and Gal4-
elavc155 was cultured at 29 °C. Ten vials have been taken
for each group (n = 100). Gal4-elavc155 flies were taken
as control and reared in normal food, whereas flies
expressing Ab42 were reared in normal as well as com-
pound 1 treated food. Viable transgenic (Ab driven) and
control (Gal4-elavc155) flies were counted every day.
123
Neurotox Res
Experiment was repeated for three times. Differences in
survival were analyzed using the Kaplan–Meier analysis.
Immunohistochemistry and Confocal Microscopy
Eye imaginal disk tissues of late third instar larvae grown
in normal food as well as compound 1 treated food were
dissected out in 19 phosphate buffer saline (PBS) and then
fixed in 4 % w/v paraformaldehyde for 15 min at room
temperature, washed twice with 19 PBS for 5 min each
then incubated with Anti-b-Amyloid (22–35) primary
antibody of 1:200 dilution (Product No. A3356, Sigma) for
overnight at 4 °C. Tissues were washed twice with 1X PBS
and incubated with secondary antibody (Anti rabbit-FITC)
for 3 h at room temperature. Following incubation, tissues
were washed with 19 PBS twice for 5 min each and
incubated with Phalloidin (Product No. 77418, Sigma) for
45 min, washed twice with 1X PBS for 10 min, and
counter stained with HOECHST (Hoechst #33258, MP
Biomedicals) solution for 5 min. Finally tissues were
washed with 1X PBS and mounted in DABCO (Product
No. 157609, MP Biomedicals). All fluorescence imaging
was done using LSM 510 META laser scanning confocal
microscope.
Acknowledgments We sincerely thank Dr. Mary Konsolaki,
Department of Genetics, Rutgers, The State University of New Jersey,
USA for kindly providing us UAS-Ab42 flies. We are also thankful to
Indian Council of Medical Research (ICMR) for providing Senior
Research Fellowship (SRF) to SKS.
Conflict of interest The authors declare that they have no conflict
of interest.
References
Bahia PK, Rattray M, Williams RJ (2008) Dietary flavonoid (-)
epicatechin stimulates phosphatidylinositol 3-kinase-dependent
anti-oxidant response element activity and up-regulates gluta-
thione in cortical astrocytes. J Neurochem 106:2194–2204.
doi:10.1111/j.1471-4159.2008.05542.x
Brand AH, Perrimon N (1993) Targeted gene expression as a means
of altering cell fates and generating dominant phenotypes.
Development 118:401–415
Cabrera M, Simoens M, Falchi G et al (2007) Synthetic chalcones,
flavanones, and flavones as antitumoral agents: biological
evaluation and structure–activity relationships. Bioorg Med
Chem 15:3356–3367
Choi Y, Kim TD, Paik SR et al (2008) Molecular simulations for anti-
amyloidogenic effect of flavonoid myricetin exerted against
Alzheimer’ s b -amyloid fibrils formation. Bull Korean Chem
Soc 29:1505–1509
Commenges D, Scotet V, Renaud S et al (2000) Intake of flavonoids
and risk of dementia. Eur J Epidemiol 16:357–363
Costa R, Ferreira-da-Silva F, Saraiva MJ, Cardoso I (2008) Trans-
thyretin protects against A-beta peptide toxicity by proteolytic
cleavage of the peptide: a mechanism sensitive to the Kunitz
Neurotox Res
protease inhibitor. PLoS ONE 3:e2899. doi:10.1371/journal.
pone.0002899
Cushnie TPT, Lamb AJ (2011) Recent advances in understanding the
antibacterial properties of flavonoids. Int J Antimicrob Agents
38:99–107
De Meyer N, Haemers A, Mishra L et al (1991) 40 -Hydroxy-3-
methoxyflavones with potent antipicornavirus activity. J Med
Chem 34:736–746
De Sousa RRR, Queiroz KCS, Souza ACS et al (2007) Phosphopro-
tein levels, MAPK activities and NFkappaB expression are
affected by fisetin. J Enzyme Inhib Med Chem 22:439–444.
doi:10.1080/14756360601162063
Delrieu J, Ousset PJ, Caillaud C, Vellas B (2012) Clinical trials in
Alzheimer’s disease: immunotherapy approaches. J Neurochem
120 Suppl :186–193. doi:10.1111/j.1471-4159.2011.07458.x
Egan WJ, Lauri G (2002) Prediction of intestinal permeability. Adv Drug
Deliv Rev 54:273–289. doi:10.1016/S0169-409X(02)00004-2
Ehrnhoefer DE, Bieschke J, Boeddrich A et al (2008) EGCG redirects
amyloidogenic polypeptides into unstructured, off-pathway olig-
omers. Nat Struct Mol Biol 15:558–566. doi:10.1038/nsmb.1437
Esteras-Chopo A, Pastor MT, Serrano L, López de la Paz M (2008)
New strategy for the generation of specific d-peptide amyloid
inhibitors. J Mol Biol 377:1372–1381
Finelli A, Kelkar A, Song H-J et al (2004) A model for studying
Alzheimer’s Abeta42-induced toxicity in Drosophila melanogaster.
Mol Cell Neurosci 26:365–375. doi:10.1016/j.mcn.2004.03.001
Frydman-Marom A, Levin A, Farfara D et al (2011) Orally admin-
istrated cinnamon extract reduces b-amyloid oligomerization and
corrects cognitive impairment in Alzheimer’s disease animal
models. PLoS ONE 6:e16564. doi:10.1371/journal.pone.0016564
Garg P, Verma J (2005) In silico prediction of blood brain barrier
permeability: an Artificial Neural Network model. J Chem Inf
Model 46:289–297. doi:10.1021/ci050303i
Haass C, Selkoe DJ (2007) Soluble protein oligomers in neurode-
generation: lessons from the Alzheimer’s amyloid beta-peptide.
Nat Rev Mol Cell Biol 8:101–112. doi:10.1038/nrm2101
Han Y-S, Zheng W-H, Bastianetto S et al (2004) Neuroprotective
effects of resveratrol against beta-amyloid-induced neurotoxicity
in rat hippocampal neurons: involvement of protein kinase C. Br
J Pharmacol 141:997–1005. doi:10.1038/sj.bjp.0705688
Hardy JA, Higgins GA (1992) Alzheimer’s disease: the amyloid
cascade hypothesis. Science 256:184–185
Heim KE, Tagliaferro AR, Bobilya DJ (2002) Flavonoid antioxidants:
chemistry, metabolism and structure–activity relationships.
J Nutr Biochem 13:572–584
Hemming ML, Patterson M, Reske-Nielsen C et al (2007) Reducing
amyloid plaque burden via ex vivo gene delivery of an Abeta-
degrading protease: a novel therapeutic approach to Alzheimer
disease. PLoS Med 4:e262. doi:10.1371/journal.pmed.0040262
Hirohata M, Hasegawa K, Tsutsumi-Yasuhara S et al (2007) The anti-
amyloidogenic effect is exerted against Alzheimer’s beta-amy-
loid fibrils in vitro by preferential and reversible binding of
flavonoids to the amyloid fibril structure. Biochemistry
46:1888–1899. doi:10.1021/bi061540x
Hong YK, Park SH, Lee S et al (2011) Neuroprotective effect of
SuHeXiang Wan in Drosophila models of Alzheimer’s disease.
J Ethnopharmacol 134:1028–1032
Hua H, Münter L, Harmeier A et al (2011) Toxicity of Alzheimer’s
disease-associated Ab peptide is ameliorated in a Drosophila
model by tight control of zinc and copper availability. Biol
Chem 392:919–926. doi:10.1515/BC.2011.084
Huang Y, Mucke L (2012) Alzheimer mechanisms and therapeutic
strategies. Cell 148:1204–1222. doi:10.1016/j.cell.2012.02.040
Inanami O, Watanabe Y, Syuto B et al (1998) Oral administration of
(-)catechin protects against ischemia-reperfusion-induced neu-
ronal death in the gerbil. Free Radic Res 29:359–365
Jang S, Dilger RN, Johnson RW (2010) Luteolin inhibits microglia
and alters hippocampal-dependent spatial working memory in
aged mice. J Nutr 140:1892–1898. doi:10.3945/jn.110.123273
Kang SS, Lee JY, Choi YK et al (2005) Neuroprotective effects of
naturally occurring biflavonoids. Bioorg Med Chem Lett
15:3588–3591
Kayed R, Glabe CG (2006) Conformation-dependent anti-amyloid
oligomer antibodies. Methods Enzymol 413:326–344. doi:10.
1016/S0076-6879(06)13017-7
Kim W, Hecht MH (2006) Generic hydrophobic residues are
sufficient to promote aggregation of the Alzheimer’s Abeta42
peptide. Proc Natl Acad Sci USA 103:15824–15829. doi:10.
1073/pnas.0605629103
Kim S-I, Jung J-W, Ahn Y-J et al (2011) Drosophila as a model
system for studying lifespan and neuroprotective activities of
plant-derived compounds. J Asia Pac Entomol 14:509–517
Landau M, Sawaya MR, Faull KF et al (2011) Towards a
pharmacophore for amyloid. PLoS Biol 9:e1001080. doi:10.
1371/journal.pbio.1001080
Larson ME, Lesné SE (2012) Soluble Ab oligomer production and
toxicity. J Neurochem 120 Suppl:125–39. doi: 10.1111/j.1471-
4159.2011.07478.x
Lee C-W, Choi H-J, Kim H-S et al (2008) Biflavonoids isolated from
Selaginella tamariscina regulate the expression of matrix
metalloproteinase in human skin fibroblasts. Bioorg Med Chem
16:732–738
Lemkul JA, Bevan DR (2010) Destabilizing Alzheimer’s Abeta(42)
protofibrils with morin: mechanistic insights from molecular
dynamics simulations. Biochemistry 49:3935–3946. doi:10.
1021/bi1000855
Li R-S, Wang X-B, Hu X-J, Kong L-Y (2013) Design, synthesis and
evaluation of flavonoid derivatives as potential multifunctional
acetylcholinesterase inhibitors against Alzheimer’s disease.
Bioorg Med Chem Lett 23:2636–2641
Lukiw WJ (2012) Amyloid beta (Ab) peptide modulators and other
current treatment strategies for Alzheimer’s disease (AD).
Expert Opin Emerg Drugs. doi: 10.1517/14728214.2012.672559
Luo Y, Smith JV, Paramasivam V et al (2002) Inhibition of amyloid-
beta aggregation and caspase-3 activation by the Ginkgo biloba
extract EGb761. Proc Natl Acad Sci USA 99:12197–12202.
doi:10.1073/pnas.182425199
Luo W, Li Y-P, He Y et al (2011a) Design, synthesis and evaluation
of novel tacrine-multialkoxybenzene hybrids as dual inhibitors
for cholinesterases and amyloid beta aggregation. Bioorg Med
Chem 19:763–770
Luo W, Li Y-P, He Y et al (2011b) Synthesis and evaluation of
heterobivalent tacrine derivatives as potential multi-functional
anti-Alzheimer agents. Eur J Med Chem 46:2609–2616
Manach C, Scalbert A, Morand C et al (2004) Polyphenols: food
sources and bioavailability. Am J Clin Nutr 79:727–747
Mangialasche F, Solomon A, Winblad B et al (2010) Alzheimer’ s
disease : clinical trials and drug development. Lancet Neurol
9:702–716. doi:10.1016/S1474-4422(10)70119-8
Nakagami Y, Nishimura S, Murasugi T et al (2002) A novel beta-
sheet breaker, RS-0406, reverses amyloid beta-induced cytotox-
icity and impairment of long-term potentiation in vitro. Br J
Pharmacol 137:676–682. doi:10.1038/sj.bjp.0704911
O’Boyle NM, Banck M, James CA et al (2011) Open Babel: an open
chemical toolbox. J Cheminform 3:33. doi:10.1186/1758-2946-3-33
Ono K, Yoshiike Y, Takashima A et al (2003) Potent anti-
amyloidogenic and fibril-destabilizing effects of polyphenols
in vitro: implications for the prevention and therapeutics of
Alzheimer’s disease. J Neurochem 87:172–181. doi:10.1046/j.
1471-4159.2003.01976.x
Ono K, Naiki H, Yamada M (2006) The development of preventives
and therapeutics for Alzheimer’s disease that inhibit the
123

formation of beta-amyloid fibrils (fAbeta), as well as destabilize
preformed fAbeta. Curr Pharm Des 12:4357–4375
Parihar MS, Hemnani T (2004) Alzheimer’s disease pathogenesis and
therapeutic interventions. J Clin Neurosci 11:456–467. doi:10.
1016/j.jocn.2003.12.007
Petit A, Bihel F, Alvès da Costa C et al (2001) New protease
inhibitors prevent gamma-secretase-mediated production of
Abeta40/42 without affecting Notch cleavage. Nat Cell Biol
3:507–511. doi:10.1038/35074581
Pfleger CM, Wang J, Friedman L, et al. (2010) Grape-seed
polyphenolic extract improves the eye phenotype in a Drosoph-
ila model of tauopathy. Int J Alzheimers Dis. doi: 10.4061/2010/
576357
Pimplikar SW (2009) Reassessing the amyloid cascade hypothesis of
Alzheimer’s disease. Int J Biochem Cell Biol 41:1261–1268.
doi:10.1016/j.biocel.2008.12.015
Puglielli L, Tanzi RE, Kovacs DM (2003) Alzheimer’s disease: the
cholesterol connection. Nat Neurosci 6:345–351. doi:10.1038/
nn0403-345
Reddy GD, Pavan Kumar KNV, Duganath N, Divya RAK (2012)
ADMET, Docking studies and binding energy calculations o f
some Novel ACE—inhibitors for the treatment of Diabetic
Nephropathy. Int J Drug Dev Res 4:268–282
Refsgaard HHF, Jensen BF, Brockhoff PB et al (2005) In silico
prediction of membrane permeability from calculated molecular
parameters. J Med Chem 48:805–811. doi:10.1021/jm049661n
Reinke AA, Gestwicki JE (2007) Structure–activity relationships of
amyloid beta-aggregation inhibitors based on curcumin: influ-
ence of linker length and flexibility. Chem Biol Drug Des
70:206–215. doi:10.1111/j.1747-0285.2007.00557.x
Rice-Evans CA, Miller NJ, Paganga G (1996) Structure-antioxidant
activity relationships of flavonoids and phenolic acids. Free
Radic Biol Med 20:933–956
Rochet JC, Lansbury PT (2000) Amyloid fibrillogenesis: themes and
variations. Curr Opin Struct Biol 10:60–68
Scherzer-Attali R, Pellarin R, Convertino M et al (2010) Complete
phenotypic recovery of an Alzheimer’s disease model by a
quinone-tryptophan hybrid aggregation inhibitor. PLoS ONE
5:e11101. doi:10.1371/journal.pone.0011101
Schroeter H, Boyd C, Spencer JPE et al (2002) MAPK signaling in
neurodegeneration: influences of flavonoids and of nitric oxide.
Neurobiol Aging 23:861–880
Selkoe DJ (1991) The molecular pathology of Alzheimer’s disease.
Neuron 6:487–498
Selkoe DJ (2001) Alzheimer’s disease: genes, proteins, and therapy.
Physiol Rev 81:741–766
Sheng R, Lin X, Zhang J et al (2009) Design, synthesis and evaluation
of flavonoid derivatives as potent AChE inhibitors. Bioorg Med
Chem 17:6692–6698
Shin DH, Bae YC, Kim-Han JS et al (2006) Polyphenol amentoflav-
one affords neuroprotection against neonatal hypoxic-ischemic
123
View publication stats
Neurotox Res
brain damage via multiple mechanisms. J Neurochem
96:561–572. doi:10.1111/j.1471-4159.2005.03582.x
Singh SK, Sinha P, Mishra L, Srikrishna S (2013) Neuroprotective
role of a novel copper chelator against Ab 42 induced
neurotoxicity. Int J Alzheimers Dis 2013:567128. doi:10.1155/
2013/567128
Sinha S, Lopes DHJ, Bitan G (2012) A key role for lysine residues in
amyloid b-protein folding, assembly, and toxicity. ACS Chem
Neurosci 3:473–481. doi:10.1021/cn3000247
Spencer JPE, Rice-Evans C, Williams RJ (2003) Modulation of pro-
survival Akt/protein kinase B and ERK1/2 signaling cascades by
quercetin and its in vivo metabolites underlie their action on
neuronal viability. J Biol Chem 278:34783–34793. doi:10.1074/
jbc.M305063200
Svechkarev DA, Doroshenko AO, Kolodezny DY (2012)
art%3A10.2478%2Fs11532-011-0127-2.pdf. Cent Eur J Chem
205–215
Thinakaran G, Koo EH (2008) Amyloid precursor protein trafficking,
processing, and function. J Biol Chem 283:29615–29619. doi:10.
1074/jbc.R800019200
Thomsen R, Christensen MH (2006) MolDock: a new technique for
high-accuracy molecular docking. J Med Chem 49:3315–3321.
doi:10.1021/jm051197e
Tjernberg LO (1999) A Molecular model of Alzheimer amyloid beta -
peptide fibril formation. J Biol Chem 274:12619–12625. doi:10.
1074/jbc.274.18.12619
Wang CN, Chi CW, Lin YL et al (2001) The neuroprotective effects
of phytoestrogens on amyloid beta protein-induced toxicity are
mediated by abrogating the activation of caspase cascade in rat
cortical neurons. J Biol Chem 276:5287–5295. doi:10.1074/jbc.
M006406200
Williams RJ, Spencer JPE (2012) Flavonoids, cognition, and demen-
tia: actions, mechanisms, and potential therapeutic utility for
Alzheimer disease. Free Radic Biol Med 52:35–45
Wolff T (2011) Preparation of Drosophila eye specimens for scanning
electron microscopy. Cold Spring Harb Protoc 2011:1383–1385.
doi:10.1101/pdb.prot066506
Yang F, Lim GP, Begum AN et al (2005) Curcumin inhibits
formation of amyloid beta oligomers and fibrils, binds plaques,
and reduces amyloid in vivo. J Biol Chem 280:5892–5901.
doi:10.1074/jbc.M404751200
Yen T-L, Hsu C-K, Lu W-J et al (2012) Neuroprotective effects of
xanthohumol, a prenylated flavonoid from hops (Humulus
lupulus), in ischemic stroke of rats. J Agric Food Chem
60:1937–1944. doi:10.1021/jf204909p
Zhang H, Ma Q, Zhang Y, Xu H (2012) Proteolytic processing of
Alzheimer’s b-amyloid precursor protein. J Neurochem 120
Suppl:9–21. doi:10.1111/j.1471-4159.2011.07519.x