Identification of human presequence protease (hPreP) agonists for the treatment of Alzheimer's disease

Article abstract of DOI:10.1016/j.ejmech.2014.02.046

Amyloid-β (Aβ), a neurotoxic peptide, is linked to the onset of Alzheimer's disease (AD). Increased Aβ content within neuronal cell mitochondria is a pathological feature in both human and mouse models with AD. This accumulation of Aβ within the mitochondrial landscape perpetuates increased free radical production and activation of the apoptotic pathway. Human Presequence Protease (hPreP) is responsible for the degradation of mitochondrial amyloid-β peptide in human neuronal cells, and is thus an attractive target to increase the proteolysis of Aβ. Therefore, it offers a potential target for Alzheimer's drug design, by identifying potential activators of hPreP. We applied structure-based drug design, combined with experimental methodologies to investigate the ability of various compounds to enhance hPreP proteolytic activity. Compounds 3c & 4c enhanced hPreP-mediated proteolysis of Aβ (1-42), pF₁β (2-54) and fluorogenic-substrate V. These results suggest that activation of hPreP by small benzimidazole derivatives provide a promising avenue for AD treatment.

Full text of DOI:10.1016/j.ejmech.2014.02.046

European Journal of Medicinal Chemistry 76 (2014) 506e516
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Identification of human presequence protease (hPreP) agonists for the
treatment of Alzheimer’s disease
*
Jhansi Rani Vangavaragu, Koteswara Rao Valasani, Xueqi Gan, Shirley ShiDu Yan
Department of Pharmacology and Toxicology, and Higuchi Bioscience Center, School of Pharmacy, University of Kansas, 2099 Constant Ave., Lawrence, KS
66047, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Amyloid-
content within neuronal cell mitochondria is a pathological feature in both human and mouse models
with AD. This accumulation of A within the mitochondrial landscape perpetuates increased free radical
production and activation of the apoptotic pathway. Human Presequence Protease (hPreP) is responsible
for the degradation of mitochondrial amyloid- peptide in human neuronal cells, and is thus an attractive
target to increase the proteolysis of A . Therefore, it offers a potential target for Alzheimer’s drug design,
b (Ab), a neurotoxic peptide, is linked to the onset of Alzheimer’s disease (AD). Increased Ab
Received 21 August 2013
Received in revised form
11 February 2014
Accepted 16 February 2014
Available online 18 February 2014
b
b
b
by identifying potential activators of hPreP. We applied structure-based drug design, combined with
experimental methodologies to investigate the ability of various compounds to enhance hPreP proteo-
lytic activity. Compounds 3c & 4c enhanced hPreP-mediated proteolysis of Ab (1e42), pF1b (2e54) and
fluorogenic-substrate V. These results suggest that activation of hPreP by small benzimidazole derivatives
provide a promising avenue for AD treatment.
Keywords:
Amyloid beta
Alzheimer’s disease
Enzyme activators
Benzimidazole derivatives
hPreP
Published by Elsevier Masson SAS.
1. Introduction
Subsequently, an increase in mitochondrial Ab intensifies its
interaction with mitochondrial protein including amyloid binding
alcohol dehydrogenase (ABAD) [14,21] and cyclophilin-D (CypD)
[16,22], exaggerating mitochondrial and neuronal stress. Thus,
Mitochondrial dysfunction is an early pathologic feature of
Alzheimer’s disease (AD). AD-affected brain shows abnormalities in
mitochondrial structure and function including hypometabolism,
increased production of reactive oxygen species (ROS) and
mitochondria provide a direct site for A
b
-induced mitochondrial
clearance
damage and reducing A accumulation or increasing A
b
b
decreased respiration potential [1e8]. Amyloid-
component of amyloid pathology contributing to the pathogenesis
of AD. Recent studies from our and other independent groups
b
(A
b
) is a major
in mitochondria may be a therapeutic strategy for prevention and
treatment of AD in particular during early stage.
Human Presequence protease (hPreP), a 114 kDa human zinc
metalloprotease 1, (hMP1) [23], is located in the mitochondrial
matrix [24] and belongs to the pitrilysin M16C family of peptidases
containing an inverted zinc-binding motif, HXXEH [25]. hPreP is an
ATP-independent protease that consists of 1037 amino acids
(AAH05025) and is encoded by the PITRM1 gene located on chro-
mosome 10 [23]. The presence of human PreP (hPreP) in mito-
chondrial matrix is confirmed by proteomics 17.31% amino acid
sequence [28] and is involved in the degradation of small-
unstructured peptides. Importantly, hPreP was found to be the
demonstrate accumulation of A
b in mitochondria in aged and AD
brain [6,9e17,51]. A enters mitochondria through the protein
b
translocase of the outer membrane (TOM) machinery [15]. The
receptor for advanced glycation end product (RAGE) also aids in
localization and transport of mitochondrial A
RAGE exhibit lowered mitochondrial A accumulation and are
protected from A -induced mitochondrial dysfunction [11]. The
degree of mitochondrial A accumulation correlates to mitochon-
drial and synaptic dysfunction as well as cognitive decline observed
in AD mouse models. Excess A disrupts normal mitochondrial
function thereby inducing oxidative stress, lowered ATP content,
and
rise in intercellular Ca2þ [14,16,18e20] concentration.
b. Neurons lacking
b
b
b
b
novel protease responsible for the degradation of mitochondrial A
b
in the human AD-affected brain [15,26,27].
Although it is a functional analog to insulin degrading enzyme
(IDE), unlike IDE, PreP cannot degrade insulin [24,28]. It does how-
a
ever degrade Ab, but the degradation of Ab peptides by recombinant
hPreP results in several fragments that are unique only to hPreP.
Despite the fact that the three dimensional structures of PreP [29]
* Corresponding author.
E-mail address: shidu@ku.edu (S.S. Yan).
http://dx.doi.org/10.1016/j.ejmech.2014.02.046
0223-5234/Published by Elsevier Masson SAS.

J.R. Vangavaragu et al. / European Journal of Medicinal Chemistry 76 (2014) 506e516
507
and IDE [30] are very similar, IDE harbors an exosite in the catalytic
chamber, which is hypothesized to unfold small proteins. The cor-
responding site is absent in PreP, thereby preventing the degrada-
tion ofsmall folded proteins. This makesPrePa bettercandidatethan
IDE for clear Ab since it does not degrade the important insulin
protein [24,28]. The available literature suggests that decreased PreP
activity in AD-affected human and mouse neuronal mitochondrial
that our benzimidazole derivatives could potentially reduce ROS
levels. A series of compounds were designed and tested to
determine their potential use in AD treatment. Each novel com-
pound was subjected to a structure based virtual screening
approach to identify their potential to bind with the hPreP active
site. The drug like properties of newly designed compounds was
predicted by QSAR and molecular docking studies. The QSAR
analysis helped to derive highly applicable models that allowed
for the modification of novel reactive molecules. Preliminary
structureeactivity relationship (SAR) studies indicated that
benzimidazole moiety is required for the activation of hPreP ac-
tivity. This is consistent with the benzimidazole derivative
described in the first examples of computer-aided exploration
into IDE regulators which can be used for the treatment of AD
[20]. The newly designed compounds proposed in this study
could serve as initiators for lead optimization studies, which
could ultimately lead towards the design of compounds for the
treatment of Alzheimer’s disease.
matrixes leads to Ab aggregation followed by apoptotic cell death.
Introduction of hPreP allows for the degradation of A
prove vital to the understanding of AD pathology.
b and could
The literature suggests that increased ROS production [31] is an
early indicator of AD pathogenesis and that increased ROS nega-
tively impacts the ability of hPreP to degrade A
five to twelve month-old mice [24], PreP proteolytic activity is
decreased with age and A -rich mitochondria and AD-affected
brain, indicating that there may be a link between overall Prep
activity, A accumulation, and age.
b. In a study, using
b
b
We built a three dimensional structural homologous model of
ꢀ
hPreP based on the 2.1A crystal structure of AtPreP [28], in which
two cysteines are in close proximity to each other allowing for the
possible formation of a disulfide bond under oxidizing conditions
inhibiting hPreP activity. A recent study demonstrated that hPreP
inactivation by H2O2 is not due to the formation of a disulfide
bridge, but is instead due to the methionine residues that are
readily oxidized to produce methionine sulfoxide ROS. Conserved
methionine residues play an important role in protecting proteins
from oxidative inactivation by scavenging oxidizing agents and
limiting the damage to catalytically essential residues [32]. This
data strongly supports the potential of hPreP to act as a target in the
development of AD treatments.
The goal of the present study was to develop small molecules
that regulate hPreP function to enhance degradation and clearance
of mitochondrial small unstructured peptides including Ab. We
utilized structure based virtual screening to identify novel com-
pounds that bind to the hPreP active site, with the goal of
enhancing proteolysis to specific substrates.
2.2. Synthesis
We synthesized a series of novel benzimidazole derivatives
(3aee & 4aee, shown in Scheme 1), with various aromatic alde-
hydes, 4-nitrobenzene-1,2-diamine (2) by mixing with trime-
thylsilyl chloride (TMSCL, 1 mol %) in water (5 mL). The reaction
mixture was stirred at room temperature for 8e12 h. TMSCl, an
extremely efficient catalyst, was used in the preparation of com-
pounds 4aee. All reactions were carried out following this general
procedure. Compounds 3aee were prepared through the reaction
of various aromatic aldehydes with 4-nitrobenzene-1,2-diamine in
trifluoroethanol (TFE) at room temperature. The chemical struc-
tures of the new compounds were confirmed using: IR, 1H NMR, 13
C
NMR spectral data and HRMS, the data for which are presented in
the Experimental Protocols section. We observed characteristic IR
absorption readings in regions at: 3475e3371 cmꢀ1 and 3342e
3330ꢀ1, for: NeH [36,37], OeH [38,39], respectively. In the 1H NMR
spectra of compounds 3aee & 4aee, the chemical shifts of aromatic
hydrogen’s on the phenyl ring appeared as doublets in region and
multiplets of 6.96e7.92 and 6.75e7.94 respectively [40,41]. The Oe
H hydrogen resonated as a broad singlet in region 11.59e10.97. In
13C NMR, we observed chemical shifts for compounds 3aee & 4aee
in expected regions.
2. Results and discussion
2.1. PreP activators design
It is also well known from the literature that benzimidazole
derivatives have antioxidant activity [33e35]. It is also possible
Scheme 1. Selective synthesis of imidazole scaffold and selected molecular targets of human PreP activators.

508
J.R. Vangavaragu et al. / European Journal of Medicinal Chemistry 76 (2014) 506e516
Table 1
QSAR descriptors defined for the title compounds.
Ligand Molecular weight Hydrogen bond Hydrogen bond Log P Molar refractivity Surface area Volume
Solvation energy Polarizability
(K cal/mol)
Total energy
(K cal/mol)
3
2
3
3
(daltans)
donors
acceptors
(Aꢂ
)
(Aꢂ
)
(Aꢂ
)
(Aꢂ
)
3a
3b
3c
3d
3e
4a
4b
4c
4d
4e
361.36
365.34
380.96
421.41
477.43
469.49
465.47
702.90
559.57
643.61
0
0
1
1
1
3
0
3
3
3
1
1
2
4
3
3
0
3
6
6
3.163 28.89
3.783 9.777
194.836
200.194
255.022
234.676
276.378
235.875
149.987
327.553
290.167
374.722
847.50
315.50
349.63
368.38
382.00
438.13
426.00
507.25
513.38
565.88
ꢀ229.942
ꢀ3.525
37.92
48.555
103.523
95.472
141.465
140.796
101.738
101.092
105.810
127.623
146.564
51.220
57.103
60.835
63.291
70.166
67.004
81.257
81.853
89.539
2.464 11.215
2.532 11.075
2.415 11.819
5.992 13.304
6.773 12.674
6.076 15.588
5.965 15.270
ꢀ190.670
ꢀ277.640
ꢀ379.375
ꢀ11.079
ꢀ6.174
ꢀ0.595
1.280
5.92
16.706
20.678
2.3. Quantitative structureeactivity relationship (QSAR) descriptors
polar pi-stacking interactions between Phe 123, Phe124, Try906,
Try921, with minimal H-bonding and Znþ2 coordination (Fig. 1).
The three dimensional structures of our compounds (3aee &
4aee) were constructed using the builder interface of the molec-
ular operating environment (MOE) program and subjected to en-
ergy minimization using the Merck molecular Force Field 94ꢁ
(MMFF94ꢁ) [42,43]. All items were set to default with an RMS
gradient of 0.01 kcal/mol and an RMS distance of 0.1 Aꢂ. We sub-
jected the resultant models to a systematic conformational search.
Using the above conformations, we studied QSAR descriptors,
meeting the Lipinski criteria for safe drugs [44,45]. Our designed
compounds have a molecular weight less than 500 Da with
exception of compounds 4c, 4d and 4e. Compound 3a had the
lowest molecular weight (361.35 Da), while compound 4c had the
highest molecular weight (702.90 Da). There were less than 5
numbers of hydrogen bond donors and less than 10 hydrogen bond
acceptors for all compounds. Log P values were below 5, indicating
that all are non-toxic to the host system. Molar refractivity was in-
between 40e150, the optimal range. Characterization of remaining
descriptors such as surface area, volume, hydration energy, polar-
izability and energy levels were also encouraging showing suitable
feature in binding domains of the targets and inhibitors, suggesting
the potential safety and efficacy of these targets (Table 1).
2.5. Pharmacophore model: PreP activators 5 feature
pharmacophore model [28,46,47]
Using the Pharmacophore five-feature model, shown in Fig. 3,
we selected 4 out of 10 compounds as ‘hit’s, namely compounds 3c,
3d, 4c and 4d. Compounds 3c and 4c exhibited strong binding
against hPreP. Based on the number of hits, the predictive capacity
of active hits is at least 50% (Fig. 2).
2.6. Evaluation of enzymatic activity of hPreP
Compounds with small molecular weights are necessary to
boost hPreP activity and provide therapeutic protection against AD.
The literature suggests that hPreP degrades mitochondrial target-
ing peptides that are cleaved off by mitochondrial processing
peptidases following import to mitochondria. PreP’s critical role in
the degradation of mitochondrial A
of mitochondrial A is directly linked to A
toxicity. Further, hPreP also degrades non-A
b
is significant as accumulation
-induced mitochondrial
substrates such as
b
b
b
pF1b, a 53 amino acid long mitochondrial presequence peptide, and
substrate V, a 9 amino acid fluorogenic peptide. Even though these
2.4. Molecular modeling [46,47]
Molecular modeling studies has often been proven to be a
powerful tool for rationalizing ligandereceptor interactions and for
making this information available to virtual screening techniques.
The ligand database that was developed from the total set of ten
compounds was docked into the specified binding domain of the
hPreP receptor. As we know that PreP is present in Arabidopsis
thaliana, and is responsible for the degradation of mitochondrial
presequences and elimination of chloroplast transit peptides
generated after organelle precursor protein import and processing
[48e50], we identified hPreP binding sites using the crystal struc-
ture of AtPreP (http://www.rcsb.org/pdb, PDB ID: 2FGE), with res-
idues: Asn136, Ala137, Phe138, Thr139, Ala140, Glu205, Try906,
Arg900, Gln222. A total of 30 conformations were generated for
each ligandereceptor complex and among them, the conformation
with least docking score was considered for further analysis. The
MOE interaction of all ligand molecules in the binding domain
cavity was analyzed. The ligand-receptor complexes were analyzed
Fig. 1. hPreP binding site for compounds 3c (yellow) and 4c (green). Shown are the
interacting residues within 5 Aꢂ that stabilize the binding of these two compounds.
Residues are labeled with residue number. The Zn cation ligand coordinates with
residues Try906, Glu205, His104 and His108. Coordination distances range from 1.0 to
2.6 Aꢂ. Interestingly, the (2, 5-Bz) group on compound 3c appears to be involved with
this coordination forming a bipyramidal trigonal complex with its OH group. Although
the equivalent aromatic hydroxy for compound 4c is oriented in the same direction, it
is further away (3.6 Aꢂ) and forms a much weaker bond. (For interpretation of the
references to colour in this figure legend, the reader is referred to the web version of
this article.)
by both London DG free energy approximations and DE interaction
energies. Compounds 3c had the best docking score of ꢀ8.6 kcal/
mole, followed by docking scores for compound 4c (ꢀ8.3 kcal/
mole). Interestingly, compound 3c formed an arene-hydrogen bond
with Gly121 (Fig. 1). These two ligands had the best docking scores
with stable ligand interactions and are composed of primarily non-

J.R. Vangavaragu et al. / European Journal of Medicinal Chemistry 76 (2014) 506e516
509
peptides have slightly different properties (A
b
peptides are nega-
pF1b bands, indicate enhanced proteolytic activity of hPreP with
different concentrations of compounds 3c & 4c. Compounds 3c & 4c
increased hPreP-mediated pF1b degradation by 1.8 (45%) & 2.2 folds
(55%) respectively (Fig. 9B). EC50 values for 3c & 4c were measured
tively charged, and pF1b are positively charged), hPreP degrades
both with similar efficiency. To our knowledge, compound 3c & 4c
are the first drug-like compounds shown to stimulate hPreP-
mediated proteolysis of various substrates.
at 0.012 & 0.601 mM, respectively (Fig. 5B and 6B).
To study the regulatoryeffectof our target compounds (3aee&4ae
e) on the proteolytic activity of hPreP, we selected three substrates of
different lengths and properties for use inaninvitrodegradation assay:
Further, we performed kinetic studies using Substrate V, a flu-
orogenic peptide 9 amino acids long containing the fluorescent
group 7-methoxycoumarin and the quencher group 2, 4-
dinitrophenyl, thus, providing light emission upon cleavage of the
peptide bond between these two groups. Compounds, 3c & 4c
increased the initial rate of proteolysis activity of hPreP the most,
1) A
54); 3) and a fluorescent peptide known as substrate V. Activity assays
were initially set at a concentration of 100 M for all studied molecules
b (1e42); 2) the presequence of ATP synthase F1b subunit pF1b (2e
m
in order to identify the effect of those compounds on overall hPreP
activity. We quantified hPreP activity by monitoring the change in
fluorescence for proteolysis of fluorogenic-substrate V and monitored
specifically by 1.5 and 1.8 folds, respectively at 1 mM (Fig. 7A and
8A). Further, the EC50 values for 3c & 4c were 0.152 & 0.057 mM,
respectively (Fig. 7B and 8B).
the efficacy of degradation of A
for A with NuPAGE 12% Bis-tris gel assay.
Analysis of compounds showed that 3c & 4c increased proteo-
lytic activity of hPreP against biotin labeled A (1e42), significant
up to a 1 M concentration when compared to other analogs.
Compounds 3c & 4c effectively facilitated the degradation of A at
b
(1e42) & pF1b, using immunoblotting
Interestingly, condensation of 5-bromo-2-hydroxy benzalde-
hyde with nitro o-phenelene diamine plays a crucial role in
bioactivity of our compounds 3c & 4c. Introduction of a 5-bromo-2-
hydroxy group (3c into the benzimidazole ring) enhanced the
b
b
m
proteolytic activity of hPreP against biotin Ab (1e42), pF1b (2e54)
b
and substrate V. The same correlation was also observed in com-
pound 4c against these three aforementioned substrates. It is clear
that the presence of a bromo group on the aromatic ring is essential
for improving hydrophilic interactions between the compound and
hPreP. Substitution of the fluoro, methoxy & hydroxy in the aro-
matic aldehyde caused a substantial loss of proteolytic activity of
hPreP. These results suggest that the compounds with halogen
atoms (bromo) that underwent hydroxyl substitution on their
phenyl ring(s) (3c & 4c) were the most potent and selective acti-
vators for the proteolytic activity of hPreP. Our results clearly
indicate that benzimidazole analogs shed new light on the design
and understanding of selective proteolytic enzyme activators,
which will enable the synthesis of our scaffold into clinically useful
anti-Alzheimer’s drugs.
more efficient levels than hPreP alone (Fig. 3A and 4A). We
observed degenerated and dose dependent immunoreactive biotin
A
b
bands due to the enhanced proteolytic activity of hPreP at
various concentrations of compound 3c & 4c (Fig. 3C, D & 4C, D),
with no A immunoreactive bands at 100 M (Fig. 3D and 4D, lane
5), compared to hPreP alone. Analysis for the effect of compounds
b
m
on hPreP activity showed that 3c increased the hPreP-mediated A
degradation by 1.7 (42%) & 4c by 2.1 folds (54%) (Fig. 9A).
b
Next, we analyzed an EC50 value of each molecule that
enhanced hPreP activity using a dose response assay, where specific
activity of hPreP was measured with different concentrations of the
lead compounds and then the plot was fit to prism nonlinear
regression, a log (agonist) vs. normalized response equation.
Calculated EC50 values for 3c & 4c were 0.713 mM and 0.402 mM,
respectively (Fig. 3B and 4B).
3. Conclusion
In the case of non-A
compounds 3c & 4c showed significant proteolytic activity of hPreP
in degrading pF1b even at 1 M concentration. Significant changes
in pF1b degradation were observed with compounds 3c & 4c when
compared to hPreP alone (Fig. 5 and 6AA). In Figs. 5C and 6C the
presence of degenerated & dose dependent immunoreactive biotin
b substrates, hPreP also degraded pF1b and
In the present study, we designed novel benzimidazole de-
rivatives according to their binding modes inside the hPreP active
site. Amongst them, we selected compounds that obeyed the Lip-
inski’s “rule of five”. This is extensively used to screen for the drug-
like properties. Our novel compounds behaved like hPreP agonists,
with compounds 3c & 4c being most active. Introduction of a 5-
bromo-2-hydroxy benzaldehyde group increased potency result-
ing in the 4-bromo-2-(1-(5-bromo-2-hydroxybenzyl)-5-nitro-1H-
m
benzo[d]imidazol-2-yl)
phenol
&
2,20-((2-(5-bromo-2-
hydroxyphenyl)-5-nitro-1H-benzo[d]imidazole-1,3(2H) diyl)bis(-
methylene))bis(4-bromophenol), which are the most active com-
pounds (at 100
EC50 in the low
m
M) and seem to be good agonists of hPreP with
mM range.
Together with biological results (indicating that the synthesized
benzimidazole-based compounds possess agonist activity with
hPreP) and the results of the docking studies and regulating ac-
tivities, we conclude that compounds 3c & 4c are appropriate
scaffolds for the development of new hPreP regulators. Further
investigation of the effect of these benzimidazole derivatives on
hPreP agonist activity will move the development of new agents for
the treatment of AD forward towards translation to clinical use.
4. Materials and methods
Fig. 2. The predicted active hits for our 5-feature pharmacophore model. Spheres
represent the important regions in space for tight ligand binding. The two pink regions
show H-bonding, while orange shows 3 regions that characterize aromaticity. The 4
active hits are identified by their “bond stick” colors: 3c (orange), 3d (magenta), 4c
(turquoise) and 4d (white). (For interpretation of the references to colour in this figure
legend, the reader is referred to the web version of this article.)
4.1. Chemistry
4.1.1. General
All reagents were commercially available and used without
further purification. Melting points were determined in open

510
J.R. Vangavaragu et al. / European Journal of Medicinal Chemistry 76 (2014) 506e516
Fig. 3. Effect of compound 3c on A
b
degradation. (A) Densitometry of A
b
immunoreactive bands is shown by analysis using NIH ImageJ software. Determination of proteolytic
activity of hPreP showing degradation of biotin-A
to immunoblotting with ExtrAvidin peroxidase conjugated IgG and detection with ECL to reveal immunoreactive biotin A
42) in the presence of compound 3c at concentrations ranging from 100 to 0.01
b
(1e42), pure hPreP protein was incubated with A
b
(1e42) and various concentrations of compound 3c, which was then subjected
. (B) Measurement of increases in hPreP activity for A (1e
M. The panels C and D showed the representative gel images.
b
b
m
capillary tubes using a Laboratory Devices Mel-Temp apparatus and
are uncorrected. 1H and 13C NMR spectra were recorded in d6-
DMSO on a Bruker DRX-500 spectrometer operating at 500 MHz,
and 125 MHz, respectively, and calibrated to the solvent peak.
Abbreviations used for the split patterns of proton NMR signals are:
singlet (s), doublet (d), triplet (t), quartet (q), quintet (qui), multi-
plet (m) and broad signal (br). High-resolution mass spectrometry
(HRMS) was recorded on a LCT Premier Spectrometer.
4.1.4. 1-(4-Fluorobenzyl)-2-(4-fluorophenyl)-5-nitro-1H-benzo[d]
imidazole (3b)
Yellow solid, 88% yield; mp: 201e203 ꢂC; Rf 0.32; (hexane: ethyl
acetate, 1:1 v/v). IR: 3076, 3055, 2948, 2765, 1897, 1610, 1523, 1483,
1329, 1265, 1172, 1110, 954, 823, 748 cmꢀ1 1H NMR (DMSO-d6)Ô
;
8.63e8.61 (m, 1H), 8.22e8.18 (s, 1H), 7.86e7.79 (m, 3H) 7.45e7.40
(t, 2H), 7.15e7.02 (m, 4H), 5.76e5.68 (m, 2H). 13C NMR (DMSO-d6)Ô
164.3(d, J ¼ 7.5 Hz), 162.4, 160.4(d, J ¼ 3.75 Hz), 157.2, 156.1, 146.9,
143.2, 142.9, 141.6, 140.2, 135.3, 132.4(d, J ¼ 3.75 Hz), 132.2(d,
J ¼ 2.5 Hz),131.7,128.3(d, J ¼ 8.75 Hz),119.6,118.4,118.0,116.1,115.9,
115.7, 115.6, 115.4, 111.7, 108.1, 47.2 (d, J ¼ 15 Hz). HRMS cald for
4.1.2. General procedure for the preparation of compounds (3ae3e)
A mixture of aromatic aldehyde (2 mmol), 4-nitrobenzene-1,2-
diamine (1 mmol, 2 equiv), and TFE (1 mL) was stirred magneti-
cally at room temperature for 1 h. The progress of the reaction was
monitored using TLC (hexane: ethyl acetate, 1:1 v/v). After
completion of the reaction, the reaction mixture was dissolved in
EtOAc (3 mL), adsorbed on silica gel (0.5 g, 230ꢀ400 mesh), and
concentrated under rotary vacuum evaporation. The resultant solid
mass was charged onto a flash chromatography column and eluted
with hexane-EtOAc (85:15) to afford the title compounds.
C
20H14F2N3O2 (M þ H) 366.1052; found 366.1054 (TOF MS ESþ).
4.1.5. 4-Bromo-2-(1-(5-bromo-2-hydroxybenzyl)-5-nitro-1H-
benzo[d]imidazol-2-yl) phenol (3c)
White crystals, 78% yield; mp: 226e228 ꢂC; Rf 0.32; (hexane:
ethyl acetate, 1:1 v/v). IR: 3475, 3342, 3055, 2989, 2893, 2304, 1903,
1629, 1610, 1579, 1488, 1434, 1348, 1222, 1176, 977, 896, 748 cmꢀ1
;
1H NMR (DMSO-d6)Ô 11.60 (s, 2H), 8.92 (s, 2H), 8.12 (d, 1H) 7.96e
7.93 (m, 2H), 7.56e7.53 (m, 2H), 6.96 (d, J ¼ 10.0 Hz, 2H), 6.81e6.78
(m, 2H). 13C NMR (DMSO-d6)Ô 159.1, 158.3, 150.3, 136.0, 135.5,
133.6, 132.5, 124.4, 122.6, 199.0, 144.1, 113.1, 110.4. HRMS cald for
4.1.3. 4-(1-(4-Hydroxybenzyl)-5-nitro-1H-benzo[d]imidazol-2-yl)
phenol (3a)
C
20H13Br2N3O4 (M þ H); 517.9351, found 517.9352 (TOF MS ESþ).
Light color solid, 77% yield; mp: 185e186 ꢂC; Rf 0.22; (hexane:
ethyl acetate, 1:1 v/v). IR: 3234, 3050, 2902, 2864, 2362, 1712, 1668,
1529, 1431, 1350, 1257, 1091, 1031, 748 cmꢀ1; 1H NMR (DMSO-d6)Ô
10.15 (s, 1H), 9.46 (s, 1H), 8.45 (d, J ¼ 3.75 Hz, 1H), 8.15e8.13 (m, 1H)
7.84 (d, J ¼ 3.75 Hz, 1H), 7.67e7.63 (m, 2H), 6.94e6.91 (m, 2H),
6.69e6.66 (m, 2H), 5.61 (s, 2H). 13C NMR (DMSO-d6)Ô 159.6, 158.6,
156.8, 147.6, 142.6, 135.4, 130.9, 127.5, 126.4, 119.6, 119.0, 117.8, 115.7
(J ¼ 15.0 Hz), 47.6. HRMS cald for C20H16N3O54 (M þ H) 362.1132;
found 362.1141 (TOF MS ESþ).
4.1.6. Methyl 2-hydroxy-5-(1-(4-hydroxy-3-(methoxycarbonyl)
benzyl)-5-nitro-1H-benzo[d]imidazol-2-yl)benzoate (3e)
Yellow solid, 86% yield; mp: 265e267 ꢂC; Rf 0.20; (hexane: ethyl
acetate, 1:1 v/v). IR: 3371, 3055, 2950, 2767, 1903, 1627, 1612, 1487,
1265, 1176, 977, 896, 817, 738 cmꢀ1; 1H NMR (DMSO-d6)Ô 11.06 (s,
2H), 8.88e8.68(m, 2H), 8.45e8.38 (m, 5H), 8.15e8.13 (m, 1H) 7.78
(d, J ¼ 5.0 Hz, 2H), 7.24 (d, J ¼ 10.0 Hz, 2H), 3.96 (s, 6H). 13C NMR

J.R. Vangavaragu et al. / European Journal of Medicinal Chemistry 76 (2014) 506e516
511
Fig. 4. Effect of compound 4c on A
showing degradation of biotin-A
noblotting with ExtrAvidin peroxidase conjugated IgG and detection with ECL to reveal immunoreactive biotin A
presence of compound 4c at concentrations ranging from 100 to 0.01 M. The panel C and D showed the representative gel images.
b
degradation. (A) Densitometry of A
b
immunoreactive bands is shown using NIH ImageJ software. Determination of proteolytic activity of hPreP
(1e42) and various concentrations of compound 4c, which was then subjected to immu-
. (B) Measurement of increased hPreP activity for A (1e42) in the
b
(1e42), pure hPreP protein was incubated with A
b
b
b
m
(DMSO-d6)Ô 167.8, 162.0, 154.0, 143.1, 141.2, 137.1, 133.8, 130.0,
118.7, 118.5, 114.7, 114.4, 111.1, 52.6. HRMS cald for C24H20N3O8
(M þ H) 478.1250; found 478.1259 (TOF MS ESþ).
1483, 1344, 1265, 1174, 1110, 896, 748 cmꢀ1
8.68e8.59 (m, 2H), 8.28e8.22 (s, 2H), 7.70e7.59 (m, 5H) 7.48e7.41
(m, 4H), 7.16e7.01(m, 4H), 5.80e5.71 (m, 4H). 13C NMR (DMSO-d6)
;
1H NMR (DMSO-d6)
d
d
165.2, 163.2, 161.1(d, J ¼ 3.75 Hz), 158.1, 158.1, 142.1, 140.5, 139.8,
138.8, 138.2, 136.3, 131.4(d, J ¼ 3.75 Hz), 131.1, 129.2, 127.3, 120.6,
119.3, 118.2, 116.8, 115.9, 115.2, 114.2, 113.2, 110.1, 107.1, 48.3 (d,
J ¼ 15 Hz). HRMS cald for C27H21F3N3O2 (M þ H) 476.1586; found
476.1591 (TOF MS ESþ).
4.1.7. General procedure for the preparation of compounds (4ae4e)
To
a stirred solution aromatic aldehyde (3 mmol), 4-
nitrobenzene-1,2-diamine (1 mmol, 2 equiv), was added in the
presence of trimethylsilyl chloride dissolved in water (5 mL). The
reaction mixture was stirred for 8e12 h at room temperature. The
progress of the reaction was monitored by TLC (hexane: ethyl-
acetate1:1). After completion of the reaction, a solid was formed.
The solid was filtered off, washed with water, and dried. It was
purified by silica gel column chromatography eluting with hexane:
ethyl acetate (95:5) mixture to afford the title compounds.
4.1.10. 2,20-((2-(5-bromo-2-hydroxyphenyl)-5-nitro-1H-benzo[d]
imidazole-1,3(2H) diyl)bis(methylene))bis(4-bromophenol) (4c)
White crystals, 82% yield; mp: 256e257 ꢂC; Rf 0.28; (hexane:
ethyl acetate, 1:1 v/v). IR: 3330, 3236, 2927, 2883, 1924, 1714, 1668,
1568, 1525, 1346, 1253, 1091, 1051, 881, 738, 698 cmꢀ1 1H NMR
;
(DMSO-d6)Ô 11.59 (s, 2H), 8.94 (s, 2H), 8.12 (d, 2H) 7.96e7.94 (m,
4H), 7.56e7.54 (m, 2H), 6.96 (d, J ¼ 10.0 Hz, 2H), 6.80e6.78 (m, 6H).
13C NMR (DMSO-d6)Ô 159.1, 158.3, 150.3, 136.0, 135.5, 133.6, 132.5,
124.4, 122.6, 199.0, 144.1, 113.1, 110.4. HRMS cald for C27H20Br3N3O5
(M þ H) 702.8953; found 702.8922 (TOF MS ESþ).
4.1.8. 4,40-((2-(4-Hydroxyphenyl)-5-nitro-1H-benzo[d]imidazole-
1,3(2H)-diyl)bis(methylene)) diphenol (4a)
Light orange color solid; mp: 224e226 ꢂC; Rf 0.24; (hexane:
ethyl acetate, 1:1 v/v). IR: 3234, 3103, 2902, 2796, 2362, 1712, 1668,
1529, 1350, 1091, 1031, 881, 738 cmꢀ1 1H NMR (DMSO-d6)
; d 10.23
(s, 1H), 9.45 (s, 2H), 8.48 (d, J ¼ 3.75 Hz, 2H), 8.15e8.13 (m, 2H)
4.1.11. Dimethyl 5,50-((2-(4-hydroxy-3-(methoxycarbonyl)phenyl)-
5-nitro-1H-benzo[d]imidazole-1,3(2H)-diyl)bis(methylene))bis(2-
hydroxybenzoate) (4e)
7.68e7.62 (m, 4H), 7.01e7.06 (m, 3H), 6.82e6.89 (m, 3H), 6.71e6.69
(m, 2H), 5.56 (s, 4H). 13C NMR (DMSO-d6)
d 160.3,159.2,158.0,148.1,
142.8,136.1,131.2,128.6,127.5,120.0,119.4,118.5,116.8 (J ¼ 15.0 Hz),
White crystals, 72% yield; mp: 286e288 ꢂC; Rf 0.18; (hexane:
ethyl acetate, 1:1 v/v). IR: 3055, 2989, 2950, 1629, 1579, 1587, 1534,
1487,1222, 1176, 977, 817, 750 cmꢀ1; 1H NMR (DMSO-d6)Ô 10.97 (s,
3H), 8.81 (d, J ¼ 5.0 Hz, 3H), 8.41 (m, 2H), 8.32e8.30 (m, 1H), 8.15e
8.13 (m,1H) 8.00 (d, J ¼ 5.0 Hz, 3H), 7.92 (d, J ¼ 5.0 Hz,1H), 7.14e7.11
(m, 3H), 6.77e6.75 (m, 4H), 3.94 (s, 9H). 13C NMR (DMSO-d6)Ô
168.5, 162.2, 158.6, 150.9, 135.9, 135.0, 134.2, 132.4, 127.8, 124.1,
48.1. HRMS cald for
470.1761(TOF MS ESþ).
C
27H24N3O5 (M
þ
H) 470.1716; found
4.1.9. 1,3-Bis(4-fluorobenzyl)-2-(4-fluorophenyl)-5-nitro-2,3-
dihydro-1H-benzo[d]imidazole (4b)
Yellow solid, 92% yield; mp: 196e198 ꢂC; Rf 0.30; (hexane: ethyl
acetate, 1:1 v/v). IR: 3099, 3055, 2927, 2761, 1895, 1730, 1610, 1579,

512
J.R. Vangavaragu et al. / European Journal of Medicinal Chemistry 76 (2014) 506e516
Fig. 5. Effect of compound 3c on pF1
b
degradation. (A) Densitometry of pF1
, pure hPreP protein was incubated with pF1
and detection with ECL to reveal immunoreactive biotin A . (B) Measurement of increased hPreP activity for pF1
ranging from 100 to 0.01 M. The panel C demonstrated the representative gel image.
b
immunoreactive bands is shown using NIH ImageJ software. Determination of proteolytic activity of
and various concentrations of compound 3c, which was then subjected to immunoblotting with
in the presence of compound 3c at concentrations
hPreP showing degradation of pF1
antibody to F1
b
b
b
b
b
m
118.0, 113.8, 113.0, 112.5, 52.6. HRMS cald for C33H30N3O11 (M þ H)
[Edelsbrunner et al. “Measuring Proteins and Voids in Proteins”.
Proceedings from the 28th International Conference on Systems
Science, 256e264 (1995)], incorporated into the Chemical
Computing Group’s the molecular modeling software. This algo-
rithm calculates and displays potential regions of tight atomic
packing on a protein surface.
644.1880; found 644.1852 (TOF MS ESþ).
4.1.12. Preparation of human PreP homology model [28]
A Blast (NCBI, www.ncbi.nih.gov) database search of the PDB
Databank revealed the optimal starting template structure for hu-
man PreP as a zinc metalloprotease (PDB:http://www.rcsb.org/pdb,
PDB ID: 2FGE). Sequence homology construction of the human
PMP1 enzyme sequence based on 2FGE structure was carried out
using a MOE protein homology algorithm. Hydrogens were added
to the resulting 3D structure based on a pH of 7 and a salt con-
centration of 0.1 M. pH of 7. The implicit born solvation model and a
non-bonded cut off value of 10e12 were incorporated in the
MMFF94ꢁ force field. The complete structure was energy mini-
mized to a gradient cut off value of 0.05. Molecular dynamic sim-
ulations were carried out at a constant temperature of 300 K for a
heating time of 10 pico seconds. Simulations were carried out for 10
nano seconds. The time step was considered as 0.001pico seconds
and the temperature relaxation time was set to 0.2 pico seconds.
Position, velocity and acceleration were saved every 0.5 pico
seconds.
4.1.14. Molecular docking [24,25]
A conformational set was generated for each of our ligand
molecules using the LowModeMD Search algorithm within the MOE
program. This involves a short molecular dynamics simulation us-
ing velocities with low kinetic energy on high-frequency vibra-
tional modes. An iteration limit of 5000 (maximum number of
attempts to generate a new conformation) and a minimization limit
of 250 (maximum number of minimization steps) were set to
control the number of conformations generated. Each generated
conformation was docked into the specified binding domain of the
human PreP receptor. We analyzed interaction of all ligand mole-
cules in the binding domain cavity using ligand interaction study in
the MOE software. The ligand-receptor complexes were analyzed
using the sum of the following two scoring functions: (i) London
5G free energy approximations and (ii) interaction energies, 5E.
Ligands that bound with the lowest docking score were considered
for further analysis.
4.1.13. Prediction of binding site for ligands
The binding site of hPreP was elucidated through the crystal
structure of AtPreP (PDB:http://www.rcsb.org/pdb, PDB ID: 2FGE).
In our case, we used a proximity radius of r ꢃ 10 Aꢂ. A confirmatory
approach for determining the binding site of these novel drug-like
candidates involved use of Alpha Site Finder methodology
4.1.15. Pharmacophore model
A pharmacophore defines features as well as locations of
important binding interactions between
a ligand and its

J.R. Vangavaragu et al. / European Journal of Medicinal Chemistry 76 (2014) 506e516
513
Fig. 6. Effect of compound 4c on pF1
hPreP showing degradation of pF1 , pure hPreP protein was incubated with pF1
antibody to F1 and detection with ECL to reveal immunoreactive biotin A . (B) Measurement of increases in hPreP activity for pF1
centrations ranging from 100 to 0.01 M. The panel C demonstrated the representative gel image.
b
degradation. (A) Densitometry of pF1
b
immunoreactive bands is shown using NIH ImageJ software. Determination of proteolytic activity of
and various concentrations of compound 4c, which were then subjected to immunoblotting with
in the presence of compound 4c at con-
b
b
b
b
b
m
receptor. Our five-feature pharmacophore model was con-
structed by overlapping the top two ligand candidates (com-
pounds 3c and 4c) that had the strongest binding affinities. The
Unified scheme within MOE was used to define the unique
features (2 H-bond [Don/Acc] and 3 aromatic centers [Aro]). The
locations of these features were determined by inspection of
strong interactions of these two compounds at the PreP active
site.
4.2. Degradation assays
For the analysis of the novel compounds as chemical regulators
for hPreP activity, we performed degradation assays with biotin-
labeled A
degradation buffer (20 mM HEPES-KOH pH 8.0) with 10 mM MgCl2,
g of purified hPreP and 0.02 g of a substrate at various con-
centrations (0.00001e100 M). Compounds were incubated for
b (1e42) and F1b (2e54) [35] presequence peptides in
1
m
m
m
Fig. 7. Kinetics of degradation of fluorogenic Substrate V by hPreP in the presence of compound-3c at 1
and emission wavelength set at 320 nm and 405 nm, respectively. (B) Measurement of increases in hPreP activity for Substrate V degradation in the presence of 3c at concentrations
ranging from 0.00001 to 5 M.
mM. (A) The change in fluorescence quenching was measured with excitation
m

514
J.R. Vangavaragu et al. / European Journal of Medicinal Chemistry 76 (2014) 506e516
Fig. 8. Kinetics of degradation of the fluorogenic Substrate V by hPreP in the presence of compound-4c at 1
excitation and emission wavelength set at 320 nm and 405 nm, respectively. (B) Measurement of increases in hPreP activity for Substrate V degradation in the presence of 4c at
concentrations ranging from 0.00001 to 5 M.
mM. (A) The change in fluorescence quenching was measured with
m
2.5 h at 37 ꢂC. Reactions were stopped by the addition of 4ꢁ sample
buffer, and then analyzed on NuPAGE 12% Bis-Tris gel (Invitrogen,
CA), and run in 1ꢁ MES buffer. Proteins were electrophoretically
transferred to nitrocellulose membrane HybondTM (Amersham
1 h. Immunoblotting was performed with ExtraAvidin Peroxidase
Conjugate 1:3000 (Sigma) and detection by ECL.
4.3. Degradation of substrate V by hPrep
Bioscience) for 1 h at 100 V. For pF1
lulose membrane was blocked overnight in 5% milk-PBS followed
by incubation with pF1 antibody (1:2000) and detected using
horseradish peroxidase-conjugated anti-rabbit secondary antibody
(1:2500) and enhanced chemiluminescence (ECL, GE Healthcare).
b identification, the nitrocel-
To determine the efficiency of these novel compounds as
chemical regulators for hPreP activity, we used a fluorescence assay
with fluorogenic substrate V (7-methoxycoumarin-4-yl-acetyl-
NPPGFSAFK-2, 4-dinitrophenyl, R&D Systems) and measured the
kinetics of proteolysis. Reaction was carried out in the presence of
b
For analyzing the degradation of biotin labeled A
LC-A 1-40, biotin-LC-A 1-42), the nitrocellulose membrane was
dried overnight at 25 ꢂC followed by blocking with 2% milk-PBS for
b (1e42) (biotin-
b
b
1
mg hPreP in 20 mM HEPES, pH 8.0 with 10 mM MgCl2 mixed with
0.1
m
g Substrate V and various concentrations (0.00001e1 M) of
m
Fig. 9. The change in magnitude of hPreP activity for three substrates. (A) Compound 3c & 4c showed folds increase respectively in the magnitude of hPreP activity for A
Densitometry of A (1e42) immunoreactive bands is shown. (B) Compound 3c & 4c show folds increase, respectively, in the magnitude of hPreP activity for pF1 (2e54). (C)
Compound 3c & 4c show folds increase, respectively, in the magnitude of hPreP activity for substrate V.
b (1e42).
b
b

J.R. Vangavaragu et al. / European Journal of Medicinal Chemistry 76 (2014) 506e516
515
neurons: implications for free radical generation and oxidative damage in
disease progression, Human Molecular Genetics 15 (2006) 1437e1449.
[14] J.W. Lustbader, M. Cirilli, C. Lin, H.W. Xu, K. Takuma, N. Wang, C. Caspersen,
X. Chen, S. Pollak, M. Chaney, F. Trinchese, S. Liu, F. Gunn-Moore, L.F. Lue,
D.G. Walker, P. Kuppusamy, Z.L. Zewier, O. Arancio, D. Stern, S.S. Yan, H. Wu,
ABAD directly links Abeta to mitochondrial toxicity in Alzheimer’s disease,
Science 304 (2004) 448e452.
[15] C.A. Hansson Petersen, N. Alikhani, H. Behbahani, B. Wiehager, P.F. Pavlov,
I. Alafuzoff, V. Leinonen, A. Ito, B. Winblad, E. Glaser, M. Ankarcrona, The
amyloid beta-peptide is imported into mitochondria via the TOM import
machinery and localized to mitochondrial cristae, Proceedings of the National
Academy of Sciences 105 (2008) 13145e13150.
our compounds in a final volume of 250 ml. The hydrolysis of sub-
strate V was measured for 10 min using a fluorometer (SpectraMax
Gemini) with excitation and emission wavelengths set at 320 and
405 nm, respectively. Results are shown as the Substrate V degra-
dation rate and averaged over three independent experiments
(Fig. 7).
4.4. Statistical analysis
[16] H. Du, L. Guo, F. Fang, D. Chen, A.A. Sosunov, G.M. McKhann, Y. Yan, C. Wang,
H. Zhang, J.D. Molkentin, F.J. Gunn-Moore, J.P. Vonsattel, O. Arancio, J.X. Chen,
S.D. Yan, Cyclophilin D deficiency attenuates mitochondrial and neuronal
perturbation and ameliorates learning and memory in Alzheimer’s disease,
Nature Medicine 14 (2008) 1097e1105.
[17] N. Dragicevic, M. Mamcarz, Y. Zhu, R. Buzzeo, J. Tan, G.W. Arendash,
P.C. Bradshaw, Mitochondrial amyloid-beta levels are associated with the
extent of mitochondrial dysfunction in different brain regions and the degree
of cognitive impairment in Alzheimer’s transgenic mice, Journal of Alz-
heimer’s Disease 20 (Suppl. 2) (2010) S535eS550.
Statistical analyses were performed using STATVIEW software.
One-way ANOVA was used for repeated measures followed by
Bonferroni/Dunn Protected Least Significant Difference analysis for
post-hoc comparisons. Results are expressed as mean ꢄ Standard
Error Mean (SEM). Significance was set at p < 0.05.
Acknowledgments
[18] D.A. Butterfield, K. Hensley, M. Harris, M. Mattson, J. Carney, beta-Amyloid
peptide free radical fragments initiate synaptosomal lipoperoxidation in a
sequence-specific fashion: implications to Alzheimer’s disease, Biochemical
and Biophysical Research Communications 200 (1994) 710e715.
[19] S.M. Cardoso, I. Santana, R.H. Swerdlow, C.R. Oliveira, Mitochondria
dysfunction of Alzheimer’s disease cybrids enhances Abeta toxicity, Journal of
Neurochemistry 89 (2004) 1417e1426.
[20] B. Cakir, O. Dagliyan, E. Dagyildiz, I. Baris, I.H. Kavakli, S. Kizilel, M. Turkay,
Structure based discovery of small molecules to regulate the activity of human
insulin degrading enzyme, Plos One 7 (2012) e31787.
This study was supported by grant awards (R37AG037319, and
RO1NS65482) from the National Institute on Aging and the Na-
tional Institute of Neurological Disease and Stroke.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.ejmech.2014.02.046.
[21] K. Takuma, J. Yao, J. Huang, H. Xu, X. Chen, J. Luddy, A.C. Trillat, D.M. Stern,
O. Arancio, S.S. Yan, ABAD enhances Abeta-induced cell stress via mitochon-
drial dysfunction, The FASEB Journal 19 (2005) 597e598.
[22] H. Du, L. Guo, W. Zhang, M. Rydzewska, S. Yan, Cyclophilin D deficiency im-
proves mitochondrial function and learning/memory in aging Alzheimer
disease mouse model, Neurobiology of Aging 32 (2011) 398e406.
[23] N. Mzhavia, Y.L. Berman, Y. Qian, L. Yan, L.A. Devi, Cloning, expression, and
characterization of human metalloprotease 1: a novel member of the pitri-
lysin family of metalloendoproteases, DNA and Cell Biology 18 (1999) 369e
380.
[24] A. Falkevall, N. Alikhani, S. Bhushan, P.F. Pavlov, K. Busch, K.A. Johnson,
T. Eneqvist, L. Tjernberg, M. Ankarcrona, E. Glaser, Degradation of the amyloid
beta-protein by the novel mitochondrial peptidasome, PreP, The Journal of
Biological Chemistry 281 (2006) 29096e29104.
References
[1] I. Maurer, S. Zierz, H.J. Moller, A selective defect of cytochrome c oxidase is
present in brain of Alzheimer disease patients, Neurobiology of Aging 21
(2000) 455e462.
[2] J.P. Sheehan, R.H. Swerdlow, S.W. Miller, R.E. Davis, J.K. Parks, W.D. Parker,
J.B. Tuttle, Calcium homeostasis and reactive oxygen species production in
cells transformed by mitochondria from individuals with sporadic Alzheimer’s
disease, The Journal of Neuroscience: The Official Journal of the Society for
Neuroscience 17 (1997) 4612e4622.
[25] A.B. Becker, R.A. Roth, An unusual active site identified in a family of zinc
metalloendopeptidases, Proceedings of the National Academy of Sciences of
the United States of America 89 (1992) 3835e3839.
[26] S.W. Taylor, E. Fahy, B. Zhang, G.M. Glenn, D.E. Warnock, S. Wiley,
A.N. Murphy, S.P. Gaucher, R.A. Capaldi, B.W. Gibson, S.S. Ghosh, Character-
ization of the human heart mitochondrial proteome, Nature Biotechnology 21
(2003) 281e286.
[27] N. Alikhani, M. Ankarcrona, E. Glaser, Mitochondria and Alzheimer’s disease:
amyloid-beta peptide uptake and degradation by the presequence protease,
hPreP, Journal of Bioenergetics and Biomembranes 41 (2009) 447e451.
[28] K.A. Johnson, S. Bhushan, A. Stahl, B.M. Hallberg, A. Frohn, E. Glaser,
T. Eneqvist, The closed structure of presequence protease PreP forms a unique
10,000 Angstroms3 chamber for proteolysis, The EMBO Journal 25 (2006)
1977e1986.
[3] R.H. Swerdlow, S.M. Khan, A “mitochondrial cascade hypothesis” for sporadic
Alzheimer’s disease, Medical Hypotheses 63 (2004) 8e20.
[4] H. Du, S.S. Yan, Mitochondrial medicine for neurodegenerative diseases, The
International Journal of Biochemistry & Cell Biology 42 (2010) 560e572.
[5] M.T. Lin, M.F. Beal, Alzheimer’s APP mangles mitochondria, Nature Medicine
12 (2006) 1241e1243.
[6] A. Eckert, S. Hauptmann, I. Scherping, V. Rhein, F. Muller-Spahn, J. Gotz,
W.E. Muller, Soluble beta-amyloid leads to mitochondrial defects in amyloid
precursor protein and tau transgenic mice, Neuro-Degenerative Diseases 5
(2008) 157e159.
[7] J. Yao, R.W. Irwin, L. Zhao, J. Nilsen, R.T. Hamilton, R.D. Brinton, Mitochondrial
bioenergetic deficit precedes Alzheimer’s pathology in female mouse model of
Alzheimer’s disease, Proceedings of the National Academy of Sciences of the
United States of America 106 (2009) 14670e14675.
[29] M. Maduke, D. Roise, Import of a mitochondrial presequence into protein-free
phospholipid vesicles, Science 260 (1993) 364e367.
[30] Y. Shen, A. Joachimiak, M.R. Rosner, W.J. Tang, Structures of human insulin-
degrading enzyme reveal a new substrate recognition mechanism, Nature
443 (2006) 870e874.
[8] C. Caspersen, N. Wang, J. Yao, A. Sosunov, X. Chen, J.W. Lustbader, H.W. Xu,
D. Stern, G. McKhann, S.D. Yan, Mitochondrial Abeta: a potential focal point for
neuronal metabolic dysfunction in Alzheimer’s disease, The FASEB Journal:
Official Publication of the Federation of American Societies for Experimental
Biology 19 (2005) 2040e2041.
[31] D.A. Butterfield, Amyloid beta-peptide (1-42)-induced oxidative stress and
neurotoxicity: implications for neurodegeneration in Alzheimer’s disease
brain. A review, Free Radical Research 36 (2002) 1307e1313.
[32] P.F. Teixeira, C.M. Pinho, R.M. Branca, J. Lehtio, R.L. Levine, E. Glaser, In vitro
oxidative inactivation of human presequence protease (hPreP), Free Radical
Biology & Medicine 53 (2012) 2188e2195.
[33] S.H. Nile, B. Kumar, S.W. Park, In vitro evaluation of selected benzimidazole
derivatives as an antioxidant and xanthine oxidase inhibitors, Chemical
Biology & Drug Design 82 (2013) 290e295.
[34] G. Ayhan-Kilcigil, C. Kus, E.D. Ozdamar, B. Can-Eke, M. Iscan, Synthesis and
antioxidant capacities of some new benzimidazole derivatives, Archiv der
Pharmazie 340 (2007) 607e611.
[35] H. Gurer-Orhan, H. Orhan, S. Suzen, M.O. Puskullu, E. Buyukbingol, Synthesis
and evaluation of in vitro antioxidant capacities of some benzimidazole de-
rivatives, Journal of Enzyme Inhibition and Medicinal Chemistry 21 (2006)
241e247.
[36] V. Koteswara Rao, A. Janardhan Rao, S. Subba Reddy, C. Naga Raju,
P. Visweswara Rao, S.K. Ghosh, Synthesis, spectral characterization and bio-
logical evaluation of phosphorylated derivatives of galanthamine, European
Journal of Medicinal Chemistry 45 (2010) 203e209.
[9] J. Yao, R.W. Irwin, L. Zhao, J. Nilsen, R.T. Hamilton, R.D. Brinton, Mitochondrial
bioenergetic deficit precedes Alzheimer’s pathology in female mouse model of
Alzheimer’s disease, Proceedings of the National Academy of Sciences 106
(2009) 14670e14675.
[10] J. Yao, H. Du, S. Yan, F. Fang, C. Wang, L.F. Lue, L. Guo, D. Chen, D.M. Stern,
F.J. Gunn Moore, J. Xi Chen, O. Arancio, S.S. Yan, Inhibition of amyloid-beta
(Abeta) peptide-binding alcohol dehydrogenase-Abeta interaction reduces
Abeta accumulation and improves mitochondrial function in a mouse model
of Alzheimer’s disease, The Journal of Neuroscience 31 (2011) 2313e2320.
[11] K. Takuma, F. Fang, W. Zhang, S. Yan, E. Fukuzaki, H. Du, A. Sosunov,
G. McKhann, Y. Funatsu, N. Nakamichi, T. Nagai, H. Mizoguchi, D. Ibi, O. Hori,
S. Ogawa, D.M. Stern, K. Yamada, S.S. Yan, RAGE-mediated signaling con-
tributes to intraneuronal transport of amyloid-beta and neuronal dysfunction,
Proceedings of the National Academy of Sciences of the United States of
America 106 (2009) 20021e20026.
[12] P.F. Pavlov, C.H. Petersen, E. Glaser, M. Ankarcrona, Mitochondrial accumu-
lation of APP and Abeta: significance for Alzheimer disease pathogenesis,
Journal of Cellular and Molecular Medicine 13 (2009) 4137e4145.
[13] M. Manczak, T.S. Anekonda, E. Henson, B.S. Park, J. Quinn, P.H. Reddy, Mito-
chondria are a direct site of A beta accumulation in Alzheimer’s disease

516
J.R. Vangavaragu et al. / European Journal of Medicinal Chemistry 76 (2014) 506e516
[37] V.K. Rao, S.S. Reddy, B.S. Krishna, C.S. Reddy, N.P. Reddy, T.C.M. Reddy,
C.N. Raju, S.K. Ghosh, Design, synthesis and anti Colon Cancer activity eval-
uation of phosphorylated derivatives of Lamivudine (3TC), Letters in Drug
Design & Discovery 8 (2011) 59e64.
[45] C.A. Lipinski, F. Lombardo, B.W. Dominy, P.J. Feeney, Experimental and
computational approaches to estimate solubility and permeability in drug
discovery and development settings, Advanced Drug Delivery Reviews 46
(2001) 3e26.
[38] V. Jhansi Rani, A. Raghavendra, P. Kishore, Y. Nanda Kumar, K. Hema Kumar,
K. Jagadeeswarareddy, Synthesis and biological activity evaluation of cytidine-
5’-deoxy-5-fluoro-N-[(alkoxy/aryloxy)] carbonyl-cyclic 2’,3’-carbonates, Eu-
ropean Journal of Medicinal Chemistry 54 (2012) 690e696.
[46] K.R. Valasani, M.O. Chaney, V.W. Day, S. Shidu Yan, Acetylcholinesterase in-
hibitors: structure based design, synthesis, pharmacophore modeling and
virtual screening, Journal of Chemical Information and Modeling 53 (2013)
2033e2046.
[39] V.J. Rani, R. Aminedi, K. Polireddy, K. Jagadeeswarareddy, Synthesis and
spectral characterization of new bis(2-(pyrimidin-2-yl)ethoxy)alkanes and
their pharmacological activity, Archiv der Pharmazie 345 (2012) 663e669.
[40] V.K. Rao, B.H. Babu, K.R. Babu, D. Srinivasulu, C.N. Raju, Ecofriendly synthesis
of tetrahydropyrimidine derivatives in aqueous medium under ultrasonic
irradiation, Synthetic Communications 42 (2012) 3368e3376.
[41] V.K. Rao, S.S. Reddy, B.S. Krishna, K.R.M. Naidu, C.N. Raju, S.K. Ghosh, Synthesis
of Schiff’s bases in aqueous medium: a green alternative approach with
effective mass yield and high reaction rates, Green Chemistry Letters and
Reviews 3 (2010) 217e223.
[42] T.A. Halgren, Merck molecular force field .1. Basis, form, scope, parameteri-
zation, and performance of MMFF94, Journal of Computational Chemistry 17
(1996) 490e519.
[43] T.A. Halgren, R.B. Nachbar, Merck molecular force field .4. Conformational
energies and geometries for MMFF94, Journal of Computational Chemistry 17
(1996) 587e615.
[47] K.R. Valasani, G. Hu, M.O. Chaney, S.S. Yan, Structure-based design and syn-
thesis of benzothiazole phosphonate analogues with inhibitors of human
ABAD-Abeta for treatment of Alzheimer’s disease, Chemical Biology & Drug
Design 81 (2013) 238e249.
[48] P. Moberg, A. Stahl, S. Bhushan, S.J. Wright, A. Eriksson, B.D. Bruce, E. Glaser,
Characterization of a novel zinc metalloprotease involved in degrading tar-
geting peptides in mitochondria and chloroplasts, The Plant Journal: for Cell
and Molecular Biology 36 (2003) 616e628.
[49] A. Stahl, P. Moberg, J. Ytterberg, O. Panfilov, H. Brockenhuus Von Lowenhielm,
F. Nilsson, E. Glaser, Isolation and identification of a novel mitochondrial
metalloprotease (PreP) that degrades targeting presequences in plants, The
Journal of Biological Chemistry 277 (2002) 41931e41939.
[50] A. Stahl, P.F. Pavlov, C. Szigyarto, E. Glaser, Rapid degradation of the prese-
quence of the f1beta precursor of the ATP synthase inside mitochondria, The
Biochemical Journal 349 (Pt 3) (2000) 703e707.
[51] V.K. Rao, E.A. Carlson, S.S. Yan, Mitochondrial permeability transition pore is a
potential drug target for neurodegeneration, Biochimica et Biophysica Acta
(BBA) e Molecular Basis of Disease (2013 Sep 18), pii: S0925-4439(13)00281-0.
[44] C.A. Lipinski, F. Lombardo, B.W. Dominy, P.J. Feeney, Experimental and
computational approaches to estimate solubility and permeability in drug
discovery and development settings, Advanced Drug Delivery Reviews 23
(1997) 3e25.