Searching for improved mimetic peptides inhibitors preventing conformational transition of amyloid-β42 monomer

Article abstract of DOI:10.1016/j.bioorg.2018.08.018

A series of novel mimetic peptides were designed, synthesised and biologically evaluated as inhibitors of Aβ₄₂ aggregation. One of the synthesised peptidic compounds, termed compound 7 modulated Aβ₄₂ aggregation as demonstrated by thioflavin T fluorescence, acting also as an inhibitor of the cytotoxicity exerted by Aβ₄₂ aggregates. The early stage interaction between compound 7 and the Aβ₄₂ monomer was investigated by replica exchange molecular dynamics (REMD) simulations and docking studies. Our theoretical results revealed that compound 7 can elongate the helical conformation state of an early stage Aβ₄₂ monomer and it helps preventing the formation of β-sheet structures by interacting with key residues in the central hydrophobic cluster (CHC). This strategy where early “on-pathway” events are monitored by small molecules will help the development of new therapeutic strategies for Alzheimer's disease.

Full text of DOI:10.1016/j.bioorg.2018.08.018

Accepted Manuscript
fSearching for improved mimetic peptides inhibitors preventing conformational
transition of amyloid-β42 monomer
János Gera, Titanilla Szögi, Zsolt Bozsó, Livia Fülöp, Exequiel E. Barrera, Ana
M. Rodriguez, Luciana Méndez, Carina M.L. Delpiccolo, Ernesto G. Mata,
Federica Cioffi, Kerensa Broersen, Gabor Paragi, Ricardo D. Enriz
PII:
S0045-2068(18)30573-X
https://doi.org/10.1016/j.bioorg.2018.08.018
YBIOO 2475
DOI:
Reference:
Bioorganic Chemistry
To appear in:
Received Date:
Revised Date:
Accepted Date:
12 June 2018
10 August 2018
13 August 2018
Please cite this article as: J. Gera, T. Szögi, Z. Bozsó, L. Fülöp, E.E. Barrera, A.M. Rodriguez, L. Méndez, C.M.L.
Delpiccolo, E.G. Mata, F. Cioffi, K. Broersen, G. Paragi, R.D. Enriz, fSearching for improved mimetic peptides
inhibitors preventing conformational transition of amyloid-β42 monomer, Bioorganic Chemistry (2018), doi: https://
doi.org/10.1016/j.bioorg.2018.08.018
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Searching for improved mimetic peptides inhibitors preventing conformational
transition of amyloid-β42 monomer
János Gera a, Titanilla Szögia, Zsolt Bozsó a, Livia Fülöp g, Exequiel E. Barrera b,
Ana M. Rodriguez c,d, Luciana Méndez e, Carina M.L. Delpiccolo e, Ernesto G. Mata e, Federica
Cioffi f, Kerensa Broersen f, Gabor Paragi a,g , Ricardo D. Enriz b,c *
a Department of Medicinal Chemistry, University of Szeged, Dom ter 8, 6720 Szeged, Hungary
b
Biomolecular Simulations Group, Institut Pasteur de Montevideo, Mataojo 2020, 11400
Montevideo, Uruguay
c
Facultad de Química, Bioquímica y Farmacia, Universidad Nacional de San Luis, Chacabuco
915, 5700 San Luis, Argentina
d IMIBIO-SL (CONICET), Chacabuco 915, 5700 San Luis, Argentina
e
Instituto de Química Rosario (CONICET-UNR), Facultad de Ciencias Bioquímicas y
Farmacéuticas, Universidad Nacional de Rosario, Suipacha 531, 2000 Rosario, Argentina
f Nanobiophysics, Faculty of Science and Technology (TNW), Technical Medical Centre, University
of Twente, The Netherlands
g MTA-SZTE Biomimetic Systems Research Group, Dom ter 8, 6720 Szeged, Hungary
__________________________________________________________________________
* Corresponding author.
Phone: (54) 266-424689-int.153; e-mail: denriz@unsl.edu.ar;
1

Abstract
A series of novel mimetic peptides were designed, synthesised and biologically evaluated as
inhibitors of Aβ42 aggregation. One of the synthesised peptidic compounds, termed compound 7
modulated Aβ42 aggregation as demonstrated by thioflavin T fluorescence, acting also as an
inhibitor of the cytotoxicity exerted by Aβ42 aggregates. The early stage interaction between
compound 7 and the Aβ42 monomer was investigated by replica exchange molecular dynamics
(REMD) simulations and docking studies. Our theoretical results revealed that compound 7 can
elongate the helical conformation state of an early stage Aβ42 monomer and it helps preventing the
formation of β-sheet structures by interacting with key residues in the central hydrophobic cluster
(CHC). This strategy where early “on-pathway” events are monitored by small molecules will help
the development of new therapeutic strategies for Alzheimer’s disease.
Keywords: mimetic peptides; Alzheimer’s disease; amyloid β-peptide; Aβ aggregation; molecular
dynamics; molecular docking.
2

1. Introduction
Alzheimer’s disease (AD) is a complex, progressive, and irreversible neurological disorder
being the most prevalent neurodegenerative disease in humans [1]. By 2030, the number of people
with the disease is expected to rise to more than 70 million cases worldwide [2,3]. Unless there is a
breakthrough in treatment, nearly one in every 2–3 people over 85 years of age will attain AD.
Although there are some controversies [4], the most widely accepted theory regarding the
aetiology of AD is known as the “amyloid hypothesis” which features the amyloid -peptide (A)
as the central pathological agent. This hypothesis posits that pathology initiates because of an
imbalance in A production and/or clearance, which may result from altered expression or
processing of amyloid precursor protein (APP) or changes in Ametabolism [5]. Various forms of
A arise upon processing of APP, including peptides varying in length from 37 (A37) to 46 (A46)
amino acids. It is widely accepted that the more amyloidogenic A42 is most dominantly toxic and
accumulates in the brain of patients with AD [6].
In the last fifteen years, growing number of evidences suggest that the small oligomeric
aggregates of Aβ, rather than the fibrillar products are the most cytotoxic species in
neurodegenerative diseases [7,8]. Thus, strategies to prevent or destabilize the formation of Aβ
aggregates by interfering with Aβ oligomers, particularly those related to Aβ42 are workable
approaches in the field of drug discovery against AD. Such approaches may include the blockage of
protein–protein interactions that generate toxic Aβ aggregates [9] or the inhibition of
conformational transition between the native disordered conformation of Aβ and the aggregation-
prone β-sheet structures [10,11]. Insight into conformational transformation that trigger misfolding
and amyloid formation is useful in the design of compounds that target these transitions [12,13].
Broad range of anti-Aβ aggregation agents, peptidic and nonpeptidic inhibitors [14-16]
nanoparticles [17], small molecules [18-20] have been designed recently to target protein
misfolding.
In order to design new molecules that prevent Aβ42 aggregation more effectively, the
detailed underlying molecular mechanism of inhibition of Aβ42 aggregation should be elucidated.
Despite the significant progress in methodologies, experiments can give a limited picture about the
effect of small molecules in the early stage of the aggregation. Computational simulations,
especially molecular dynamics (MD) can provide atomic level description of conformational
transitions involved in peptide–ligand interactions. Recently, a large number of computational
studies have been reported to elucidate the fundamental molecular mechanism of unfolding of the
native conformation of Aβ42 and to characterize the inhibition mechanisms of current Aβ inhibitors
at atomic level [20].
3

In the search for new anti-aggregation agents, we have reported a series of mimetic peptides
with potent inhibitory activity against the formation of Aβ42 aggregates [21,22]. These compounds
were derived from a molecular modelling study using as molecular target a pentameric Aβ model
previously developed by our research group [23]. Among these compounds, DZK (Nα,Nε-Di-Z-L-
lysine hydroxysuccinimide ester, Fig. 1) presented the most potent inhibiting properties, therefore in
the present work this compound was taken as the starting structure in the search of new inhibitors.
Recently, we performed a theoretical study on DZK using QM/MM calculations and
QTAIM analysis using an Aβ42 monomeric model [24].
Fig. 1. DZK structure (core: in magenta, R1: in blue, R2: in red, R3: in green). Structure features of compounds 1-8
Following this work, we aimed to find new peptide-like structures that have an inhibitory
effect against the formation of Aβ42 aggregates based on combined theoretical-experimental study.
In a first step, taking compound DZK as a starting point, we designed and synthesised several
structurally related mimetic peptides. In the second stage, the anti-aggregation properties of these
new compounds were evaluated by spectroscopy and microscopy techniques, while their capacity to
inhibit Aβ42-induced toxicity was monitored in an in vitro viability assay. To have insights of the
atomic level interaction between the active compound and the monomeric Aβ42 peptide a molecular
modelling study was performed. This stage included docking calculations and enhanced sampling
MD simulations.
4

2. Materials and Methods
2.1. Chemistry
2.1.1. Synthesis
R2
O
R2
O
a) 30% piperidine
H
10% TFA
in DCM
O
H
N
in DMF
N
FmocHN
FmocHN
O
FmocHN
OH
O
b) EDC.HCl, HOBt, DIPEA
FmocNHCH(R2)CO2H
DCM:DMF (2:1)
For n=3: steps a) and b)
were repeated once
O
R1
O
R1
R1
AAn
AAn
= Wang resin
AA = amino acid
n=1,2,3
Scheme 1
The strategy employed for the synthesis of dipeptides (4, 6-8) and tripeptide (5) started with the
Fmoc-protected amino acids preloaded to Wang resin which were treated with 30% piperidine in
DMF to free the amino group (Scheme 1). The second or third amino acids were coupled using
standard solid-phase peptide synthesis conditions, hydroxybenzotriazole (HOBt) and 1-ethyl-3-(3-
dimethylaminopropyl)carbodiimide (EDC) [25,26]. The obtained immobilized peptides were
cleaved from the solid support by treatment with 10% trifluoroacetic acid in dichloromethane.
Formation of N-hydroxysuccinimide esters 2 and 3 were carried out using standard conditions
(N,N´-dicyclohexylcarbodiimide, N-hydroxysuccinimide). On the other hand, a lysine derivative of
1 was obtained by treatment of H-Lys(Z)OBn hydrochloride with benzyl chloroformate in the
presence of triethylamine and 4-N,N-dimethylaminopyridine.
2.1.2. General
Chemical reagents were purchased from commercial sources and used without further purification.
1
Solvents were analytical grade or were purified by standard procedures prior to use. H NMR
13
spectra were recorded on a Bruker Avance 300 spectrophotometer operating at 300 MHz. C NMR
spectra were recorded on the same apparatus at 75 MHz. Mass spectra were obtained on a LC-MS
Bruker micrOTOF-Q II spectrophotometer. Silica gel aluminium plates (Merck 60 F254) were used
for analytical TLC. Flash column chromatography was performed using Merck silica gel 60 (230–
400 mesh).
2.1.3. General procedure I. Solid phase coupling
The Fmoc-amino acid-Wang resin (300 mg) was suspended in a 30% piperidine solution in DMF
and stirred 50 minutes at r.t (room temperature). The resin was washed with CH2Cl2 (3 × 2 mL),
EtOAc (3 × 2 mL), MeOH (3 × 2 mL) and CH2Cl2 (2 mL), and finally dried in vacuum. The resin
5

was suspended in 5 mL of a DMF:DCM (2:1) mixture and Fmoc-amino acid (3 equiv.), 1-
hydroxybenzotriazole (3 equiv.), 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (3 equiv.) and
diisopropylethylamine (3 equiv.) were successively added. The solution was stirred at r.t. for 16 h.
After this, the resin was washed with DMF (3 × 2 mL), CH2Cl2 (3 × 2 mL), MeOH (3 × 2 mL) and
CH2Cl2 (2 mL), and finally dried in vacuum. Resin was treated with 5 mL of 10% TFA in CH2Cl2
for 1 h. The mixture was filtered and the filtrate was evaporated under reduced pressure to give the
crude product.
2.1.4. General procedure II. Formation of N-hydroxysuccinimide ester
The amino acid was dissolved in CH2Cl2 and N-hydroxysuccinimide (1.2 equiv.) was added. The
solution was cooled to 0°C and N,N´-dicyclohexylcarbodiimide (1.2 equiv.) dissolved in CH2Cl2
was added dropwise. The mixture was stirred overnight at r.t., filtered, and the filtrate was
evaporated under reduced pressure to give the crude product.
2.1.5. Z-Lys(Z)-OBn (1)
H-Lys(Z)OBn hydrochloride (100 mg, 0.24 mmol) was dissolved in anhydrous N,N-
dimethylformamide, trimethylamine (50 µL, 0.36 mmol, 1.5 equiv.), benzyl chloroformate (51 µL,
0.36 mmol, 1.5 equiv.) and 4-dimethylaminopyridine (cat.) were successively added. The reaction
mixture was stirred at 60°C overnight. Purification by silica gel column chromatography
(Hexane/EtOAc) gives 48 mg of product 1 (40% yield).
1H NMR (CDCl3, 300 MHz): δ (ppm) 7.35-7.28 (m, 15H), 5.17 (dd, J1= 17.8 Hz, J2= 12 Hz, 2H),
5.05 (m, 2H), 3.96 (m, 2H), 3.58 (t, J =6.1 Hz, 1H), 3.06 (m, 2H), 1.87 (m, 2H), 1.36-1.19 (m, 4H).
13C NMR (CDCl3, 75 MHz): δ (ppm) 169.6, 156.5, 136.6, 134.6, 131.7, 129.9, 129.1, 128.9, 128.8,
+
128.7, 128.4, 128.0, 67.8, 66.5, 58.6, 50.2, 40.2, 29.8, 29.1, 21.9. HRMS calcd. for C29H32N2NaO6
([M+Na]+, m/z): 527.21526, found: 527.21473.
2.1.6. Fmoc-Lys(Z)-OSu (2)
Starting from Fmoc-Lys(Z)-OH (114 mg, 0.22 mmol) and following general procedure II, we
obtained the crude material that was purified by silica gel column chromatography
(Hexane/EtOAc), providing 72.3 mg of product 2 (55% yield).
1H NMR (CDCl3, 300 MHz): δ (ppm) 7.75 (d, J= 7.3 Hz, 2H), 7.58 (d, J= 7.3 Hz, 2H), 7.38-7.25
(m, 9H), 5.57 (m, 1H), 5.08 (s, 2H), 5.05 (m, 1H), 4.73 (m, 1H), 4.43 (m, 2H), 4.22 (m, 1H), 3.21
(m, 2H), 2.76 (s, 4H), 1.97-1.90 (m, 2H), 1.51 (m, 4H). 13C NMR (CDCl3, 75 MHz): δ (ppm) 168.6,
168.2, 156.6, 155.6, 143.8, 143.6, 141.3, 136.6, 128.4, 128.0, 127.7, 127.1, 125.0, 119.9, 67.2, 66.6,
6

+
52.1, 47.1, 40.2, 31.8, 29.1, 25.5, 21.6. HRMS calcd. for C33H33N3NaO8 ([M+Na]+, m/z):
622.21599; found: 622.21553.
2.1.7. Z-Lys(Boc)-OSu (3)
Starting from (Z)-Lys(Boc)dicyclohexylamonium salt and following general procedure II, we
obtained the crude material that was purified by silica gel column chromatography
(Hexane/EtOAc), providing 75 mg of product 3 (98% yield).
1HNMR (CDCl3, 300 MHz): δ (ppm) 7.34 (m, 5H), 5.55 (m, 1H), 5.12 (s, 2H), 4.70 (m, 2H), 3.10
13
(m, 2H), 2.82 (s, 4H), 2.03-1.86 (m, 2H), 1.50 (m, 4H), 1.40 (s, 9H). CNMR (CDCl3, 75 MHz): δ
(ppm) 168.6, 168.2, 156.2, 155.6, 135.9, 128.5, 128.2, 79.1, 67.3, 52.2, 39.6, 31.8, 29.4, 28.4, 25.5,
+
21.7. HRMS calcd. for C23H31N3NaO8 ([M+Na]+, m/z): 500.20034, found: 500.19978.
2.1.8. Fmoc-Lys(Fmoc)-Phe-OH (4)
Starting from Fmoc-Phe-Wang resin (300 mg, 0.21 mmol, sust: 0.7 mmol/g) and following general
procedure I, using Fmoc-Lys(Fmoc)-OH (372.1 mg, 0.63 mmol), we obtained the crude material
that was purified by silica gel column chromatography (CH2Cl2/MeOH, 95:5, drops of AcOH)
providing 96.9 mg of compound 4 (65% yield).
1H NMR ((CD3)2SO, 300 MHz): δ (ppm) 8.05 (d, J = 8.1 Hz, 1H), 7.88-7.85 (m, 4H), 7.70-7.65 (m,
4H), 7.41-7.14 (m, 13H), 4.46-4.38 (m, 1H), 4.28-4.15 (m, 6H), 4.00-3.93 (m, 1H), 3.16 (m, 1H),
3.07-2.85 (m, 4H), 1.53-1.21 (m, 6H), 13C NMR ((CD3)2SO, 75 MHz): δ (ppm) 172.7, 171.9, 156.0,
155.8, 143.8, 143.7, 140.6, 137.3, 129.0, 128.0, 127.5, 127.0, 126.3, 125.2, 125.0, 120.0, 65.5, 65.1,
54.4, 53.2, 48.5, 46.7, 46.6, 36.6, 31.6, 29.0, 22.7.
2.1.9. Fmoc-Lys(Fmoc)-Ala-Gly-OH (5)
Starting from Fmoc-Gly-Wang resin (300 mg, 0.18 mmol, sust.: 0.6 mmol/g) and following the
general procedure I, using Fmoc-Ala-OH (168 mg, 0.54 mmol) and Fmoc-Lys-Fmoc (318mg, 0.54
mmol), we obtained the crude material that was purified by silica gel column chromatography
(CH2Cl2/MeOH, 95:5, drops of AcOH) providing 21.4 mg of compound 5 (20% yield).
1H NMR ((CD3)2SO, 300 MHz): δ (ppm) 8.08 (m, 1H), 7.97 (m, 1H), 7.87 (m, 3H), 7.72-7.64 (m,
3H), 7.47-7.28 (m, 8H), 4.26-4.18 (m, 5H), 3.96 (m, 1H), 3.75-3.69 (m, 1H), 2.95 (m, 2H), 1.62-
1.21 (m, 3H), 1.20 (d, J= 7.4 Hz, 3H). 13C NMR ((CD3)2SO, 75 MHz): δ (ppm) 172.8, 172.1, 171.5,
156.5, 144.3, 144.2, 141.1, 128.0, 127.5, 125.7, 125.6, 120.5, 66.04, 65.6, 49.0, 48.3, 47.2, 47.1,
32.0, 29.5, 23.3, 18.8.
2.1.10. Fmoc-Phe-Gly-OH (6)
7

Starting from Fmoc-Gly-Wang resin (300 mg, 0.21 mmol, sust: 0.70 mmol/g) and following general
procedure I, using Fmoc-PheOH (244 mg, 0.63 mmol), we obtained the crude material that was
purified by silica gel column chromatography (CH2Cl2/MeOH, 95:5, drops of AcOH) providing
16.2 mg of compound 6.
1HNMR ((CD3)2CO, 300 MHz): δ (ppm) 7.39 (d, J = 7.3 Hz, 2H), 7.35-7.16 (m, 11H), 4.59-4.52
(m, 1H), 4.29-4.12 (m, 3H), 4.01 (d, J = 5.7 Hz, 2H), 3.27 (dd, J1= 14.0 Hz, J2= 4.2 Hz, 1H), 2.96
13
(dd, J1 = 14 Hz, J2 = 9.6 Hz, 1H); CNMR ((CD3)2CO, 75 MHz): δ (ppm) 172.7, 172.2, 171.3,
156.8, 144.9, 142.0, 138.6, 130.2, 129.0, 128.4, 127.9, 127.2, 126.2, 126.1, 120.7, 67.2, 57.1, 47.8,
+
41.5, 38.8. HRMS calcd. for C26H23N2Na2O5 [(M-H)+2Na]+, m/z: 489.1397; found: 489.1378.
2.1.11. Fmoc-Asp-Tyr-OH (7)
Starting from Fmoc-Tyr(t-Bu)-Wang resin (300 mg, 0.21 mmol, sust: 0.70 mmol/g) and following
general procedure I, using Fmoc-Asp(OtBu)OH (300 mg, 0.63 mmol), we obtained the crude
material that was purified by silica gel column chromatography (CH2Cl2/MeOH, 95:5, drops of
AcOH) providing 16.2 mg of compound 7.
1H NMR ((CD3)2CO, 300 MHz): δ (ppm) 7.84 (d, J = 7.3 Hz, 2H), 7.69 (d, J =7.3 Hz, 2H), 7.42-
7.28 (m, 4H), 7.07 (d, J = 8.3 Hz, 2H), 6.92 (d, J = 6 Hz, 1H), 6.72 (d, J =8.3 Hz, 2H), 4.69- 4.57
(m, 2H), 4.32-4.21 (m, 3H), 3.12-2.70 (m, 5H). 13C NMR ((CD3)2CO, 75 MHz): δ (ppm) 172.7,
172.5, 171.3, 157.1, 145.0, 142.0, 131.3, 128.5, 128.0, 126.2, 120.8, 116.0, 67.6, 54.5, 54.5, 52.4,
+
47.9, 37.1, 36.5. HRMS calcd. for C28H25N2Na2O8 [(M-H)+2Na]+, m/z: 563.1401; found:
563.1383.
2.1.12. Fmoc-Glu-Tyr-OH (8)
Starting from Fmoc-Tyr(t-Bu)-Wang resin (285 mg, 0.19 mmol, sust: 0.7 mmol/g) and following
general procedure I, using Fmoc-Glu(OtBu)-OH (270 mg, 0.57 mmol). TFA re-treatment of the raw
reaction was necessary to complete removal of the tert-butyl group. We obtained the crude material
that was purified by silica gel column chromatography (CH2Cl2/MeOH, 95:5, drops of AcOH)
providing 66.9 mg of pure compound 8 (68% yield).
1H NMR ((CD3)2CO, 300 MHz): δ (ppm) 7.83 (d, J = 7.4 Hz, 2H), 7.69 (m, 2H), 7.41-7.28 m, 4H),
7.07 (d, J = 8.3 Hz, 2H), 6.72 (d, J= 8.5 Hz, 2H), 4.68 (m, 1H), 4.33-4.20 (m, 4H), 3.14-2.94 (m,
13
2H), 2.43 (t, J = 7.4 Hz, 2H), 2.17-1.88 (m, 2H); C NMR ((CD3)2CO, 75 MHz): δ (ppm) 174.3,
172.9, 172.1, 157.1, 157.0, 145.1, 144.9, 142.0, 131.2, 128.5, 128.4, 127.9, 126.2, 120.7, 116.0,
67.3, 55.4, 55.0, 54.4, 48.0, 37.1, 28.5.
2.2. Biophysical assays
8

All the compounds were dissolved in DMSO prior to experiments at a concentration of 200 mM.
From this stock, serial dilutions, ranging from 0.75 to 200 M were prepared. However, the final
DMSO concentration in the analyzed samples was less than 1%.
2.2.1. Thioflavin T fluorescence assay
Amyloid aggregation was measured by a Thioflavin-T (ThT) fluorescence assay, a common
technique which allows to monitor fibril formation [27]. Aβ42 was dissolved using a previously
published solubilisation procedure using HFIP, DMSO and desalting column separation [28]. Aβ42
concentration was adjusted to 25 μM using PBS buffer, pH 7.4 and a final concentration of 12 μM
ThT in a 96-well plate. Fluorescence intensity was measured at 37 °C using an automated well-plate
reader (TECAN Infinite 200 PRO) at an excitation wavelength of 450 nm and emission detection
from 480 to 600 nm. Measurements were performed as independent triplicates. Recorded values
were averaged and background measurements (buffer containing 12 μM ThT and compound) were
subtracted. Statistical significance of the results was established by P-values using two-tailed t-tests
(GraphPad Software).
2.2.2. Cytotoxicity assay
SH-SY5Y cells were grown in DMEM/F12 medium supplemented with 10% FBS, 1%
penicillin/streptomycin and 1% nonessential amino acids (Gibco). Cells were seeded in a 96-well
plate at 25,000 cells/well and maintained in phenol-red free DMEM/F12 (L-Glutamine, 15mM
HEPES) supplemented with 1% penicillin/streptomycin and incubated at 5% CO2. Samples
containing 25 μM Aβ42, in absence or presence of 50 or 0.75 μM of compound 7, were pre-
incubated at r.t. for 2 h and added to the cells. As a positive control an 8% SDS solution was
included. The plates were incubated at 37°C for 48 h, followed by addition of CellTiter-Blue®
Reagent (20 μl/well) and incubation for 4 h. The fluorescence intensity (excitation wavelength 560
nm and emission wavelength 590nm) was measured using a TECAN Infinite 200 PRO fluorescence
platereader. The medium background values were subtracted from the values obtained in
experimental wells.
2.2.3. Transmission electron microscopy
An isoform of Aβ42 was synthesised by Bozso et. al, as described earlier [29]. The peptide was
dissolved in 100 mM NaOH (pH=11) to a concentration of 1 mg/ml. The stock solution was
sonicated for 3 min then incubated for 2 h at r.t., aliquoted and snap-frozen for further use. Prior to
the experiments, stock aliquots were diluted with 20 mM PBS (pH=7.4) to a final concentration of
25 µM. To the peptide solutions, either DMSO (vehicle) or compound 7 dissolved in DMSO was
9

o
added at two final concentrations, 25 µM or 100 µM. The samples were incubated for 37 C for 0
min, 24 h and 168 h, and following incubation, 10 μl aliquots were placed on formvar carbon 400-
mesh copper grids (Electron Microscopy Sciences, Washington, PA, USA). Grids were stained with
2% (w/v) uranyl acetate. Images were taken on a JEOL JEM-1400 transmission electron
microscope (JEOL Ltd., Japan) operating at 120 kV. Images were captured at magnifications of x
25,000 and 40,000 and analysed with a SightX Viewer Software (EM-15300SXV Image Edit
Software, JEOL Ltd., Tokyo, Japan).
2.3. Molecular modelling
PDB entry 1IYT was selected as starting monomeric conformation of Aβ42. This conformation was
determined by NMR measurements in apolar environment [30], being characterized by two helices
between the 8-25 and the 28-39 residues. The starting structure was a highly helical conformation of
the monomer in order to start the simulation from a close state of the cleavage, which according to
the literature [31] can occur in an α-helical state of the Aβ section in the Amyloid Precursor Protein.
Compound 7 was optimized first at molecular mechanical level, with PRCG (Polak-Ribiere
Conjugate Gradient) method [32] using Macromodel [33] from the Schrodinger suit and it was
further optimized with Gaussian09 [34] at HF/6-31g level of theory. The atomic charges were taken
from the quantum calculation and attached to the structure parameters with the Antechamber [35]
program. Other force constant parameters of compound 7 were based on the GAFF parameter set
[36].
2.3.1. Simulation details
Sampling conformational space of monomeric Aβ42 in the presence and absence of compound 7,
Replica Exchange Molecular Dynamics (REMD) simulations were carried out in explicit solvent
with the GROMACS 5.1 package using the AMBER99SB-ILDN force field for the Aβ42 peptide. A
dodecahedron box with 1.2 nm distance between the box and the solute was taken and solvated with
explicit TIP3P water molecules. The system was neutralized with Na+ ions and further Na+Cl- ion
pairs were added to mimic the physiological (0.15 M) salt concentration. Both systems (Aβ42 and
Aβ42 + ligand) were energy minimized with 50000 steepest-descent steps. After minimization the
system was heated up and 200 ps long NPT and NVT equilibrated at 315 K. For REMD
simulations, 48 replicas were taken in the range of 315K and 400K using the temperature
distribution provided by the web server of D. Van der Spoel [37]. Each replica was 250 ns long and
temperature coupling was applied using velocity rescaling with a stochastic term [38] (0.1 ps time
constant), as well as the isotropic Parrinello-Rahman barostat [39] with 0.5 ps time constant. P-
10

LINCS algorithm [40] was selected for hydrogen atom connection constrains and in the electrostatic
interaction the Particle Mesh Ewald (PME) method [41] was applied.
2.3.2. Ensemble analysis
Two time periods were selected for analysis from the lowest temperature (315 K) replicas: the first
50 ns and the last 150 ns period. The recently published Dihedral-based Segment Identification and
Classification method (DISICL) [42] was applied to calculate the secondary structure distribution
for each residue. The contact maps are based on the probability of contacts over the selected
periods. Two residues were considered to be in contact if their centre of mass distance is equal or
lower than 0.4 nm.
2.3.3. Docking
The pure Aβ42 monomer simulation was clustered with single linkage method taking the last 150 ns
from the full trajectory. Considering clusters with more than 100 cluster members, the average
middle structures were selected as representants of the trajectory for docking calculations.
Compound 7 was docked onto these representants using the Glide software [43]. First, blind
dockings were performed with the single precision (SP) method, where the inner box size was
adjusted to 40-40-40 Angstrom (Å) while the outer box size was set to 76-76-76 Å. Refining the
blind docking results, the top 5 positions were selected, and redocked using the extra precision (XP)
as well as the induced fit docking (IFD) methods [44]. XP-dockings were performed with 20-20-20
and 35-35-35 Å inner and outer box size, respectively. All the other docking parameters were kept
at its default values during SP or XP docking calculations, as well as the IFD using standard
protocol with default settings. All interaction energies are characterized by the Glide-score values
[43] which is a widely accepted scoring function for computational binding affinity prediction. The
Glide score takes into account many aspects of a ligand-protein interaction just like hydrogen
bonds, rotatable bond penalty, and contributions from protein-ligand coulomb-vdW energies, as
well as a consequence of the hydrophobic environment.
3. Results and Discussion
3.1. Searching new inhibitors preventing conformational transition of Aβ42 monomer
In search for new inhibitors with improved potency to modulate Aβ42 aggregation, our
previously reported compound DZK (Fig. 1) was used as initial structure from which further
optimizations were explored. The general structure of this compound consists of four parts: the core
11

(amino acid Lys) and three substituents located at its α-carboxylic (R1), α-amine (R2) and ε-amine
group (R3). The strategy chosen for the design of the eight new structures was based on the
modification of natural amino acids with aromatic or hydrophobic substituents hypothesised to
interact with the central hydrophobic cluster (CHC) of the Aβ42 peptide. Compounds 1, 2 and 3
presented minor variations compared to DZK by maintaining the Lys core and introducing new
substituents in R1=Z (compound 1); R2=Fmoc (compound 2); and R3=Boc (compound 3).
Compounds 4 and 5 not only presented modified substitution patterns with R2=R3=Fmoc but also
extra residues at R1 like Phe (compound 4) and Ala-Gly (compound 5). Finally, compounds 6, 7
and 8 presented major variations losing the characteristic Lys core. Compound 6 core was formed
by the dipeptide Phe-Gly; compound 7 by Asp-Tyr; and compound 8 by Glu-Tyr. These last three
structures were modified in their N-terminal moieties with the Fmoc group as substituent.
To evaluate the inhibition of Aβ42 aggregation, a ThT study was performed incubating Aβ42
peptide for seven days at 37°C in the absence and presence of this new series of compounds. Of the
eight studied mimetic peptides only compound 7 showed activity being considerably more potent at
inhibiting ThT-positive Aβ42 aggregation compared to the previously reported DZK (Fig.2) [21].
Namely, DZK 100 M was needed to reduce by 50% the ThT positive aggregates while compound
7 reached similar inhibitory activity at 3 M concentrations.
Fig. 2. Aβ42 amyloid aggregation is affected in a dose-dependent manner by compound 7. Dose-response amyloid fibril
formation of 25 μM Aβ42 incubated in the presence of compound at 37° C for 7 days was monitored using ThT
12

fluorescence intensity at 485 nm. Values represent results of three independent replicates. Statistical significance of the
results was established by P-values using paired two-tailed t-tests. Statistical significance levels were * P < 0.05, **P <
0.005 and ***P < 0.0005.
To obtain further insight into the effect of compound 7 on the Aβ42 aggregation kinetics a ThT
assay was performed monitoring fluorescence intensity at different times of incubation. Similar to
the 7-day incubation shown in Fig. 2, after 1.5 hrs ThT-positive Aβ42 formation was diminished in a
dose-dependent manner suggesting its potential interaction with Aβ42 monomers or early
aggregation species (Fig. S1, Supplementary material).
The inhibitory effect of compound 7 over Aβ42 aggregates was further established using a
cytotoxicity assay and TEM analysis. A cell viability assay, Cell Titer-Blue, demonstrated that 25
μM (based on monomeric concentration) of oligomeric Aβ42 induced significant loss of viability in
an SHSY-5Y neuroblastoma cell line. Co-incubation of 25 μM Aβ42 oligomers with compound 7 at
a concentration as low as 0.75 μM resulted in complete prevention of Aβ42-mediated cell toxicity.
Moreover, the compound itself was found to be non-toxic at concentrations up to 200 μM compared
to SDS 8% which was used as a control for cell death (Fig. 3).
Fig. 3. Cell viability assay using the CellTiter-Blue reagent with SHSY-5Y cells exposed to 25 μM Aβ42 in presence or
absence of compound 7. Controls include 8% μM solution of sodium dodecyl sulfate, full-length Aβ42, compound and
13

cells in buffer. The results are represented as mean values ± standard error of the mean. Values represent results of
three independent replicates. Significant p-values are indicated by: *P < 0.05, **P < 0.01 and ***P < 0.001.
Complementary, we conducted an in vitro fibril formation experiment in order to assess the
effect of compound 7 on the aggregation process of Aβ42 by TEM. Compound 7 was dissolved in
DMSO and added in two different concentrations of 25 and 100 μM to a solution of 25 μM freshly-
prepared Aβ42 oligomers. The peptide solutions were incubated at 37 ºC, and aliquots were
analysed by TEM at three time points: 0 min, 24 h and 168 h. Fig. 4 shows representative images of
the Aβ42 oligomers incubated with a concentration of 25 μM compound 7. Images were analysed
according to the main morphological characteristics of the formed aggregates on randomly selected
images (Table 1). According to the results of the analysis, average diameters of the aggregates did
not change with time considerably. Lengths of fibrillary aggregates varied in a wide range, but the
semi-quantitative evaluation allows a rough estimation for the size distribution. According to this,
we can conclude that after 24 h, the length of the fibrils fell in a lower size range when treated with
compound 7 (Fig. 4D) than without it (Fig. 4C). This difference in the lengths between the two
samples was later partly equalized, as large, mature fibrils formed in both samples after 168 h.
These results indicate that although compound 7 can not hinder the fibril formation, it influences
the kinetics of the aggregation.
0 min
24 h
48 h
Diameter
Length
Diameter
Length
Diameter
Length
Aβ42
22-85 nm
60-842 nm
55-760 nm
8.32.4 nm
7.61.3 nm
8.91.8 nm
oligomers
Aβ42 +
28-62 nm
29-240 nm
24-450 nm
6.71.2 nm
6.81.2 nm
7.21.9 nm
Compound 7
Table 1. Average diameters of the aggregates and length intervals of the fibrillar forms
14

Fig. 4. Representative transmission electron microscopy images of Aβ42 incubated either in the presence of DMSO or
25 µM compound 7. Images were taken at three time points. 0 min: Aβ42 with DMSO (A) or 25 µM compound 7 (B).
24 h: Aβ42 with DMSO (C) or 25 µM compound 7 (D). 168 h: Aβ42 with DMSO (E) or 25 µM compound 7 (F). Scale
bar: 500 nm.
3.2. Molecular Modelling
In order to have an atomistic insight into the experimentally hard to realize early stage effect
of the most promising ligand, computational investigations were carried out regarding the
interaction of compound 7 and the monomeric Aβ42. MD analysis and docking studies were applied
for the computationally derived Aβ42 peptide ensemble, which represent the conformational space
immediately after the enzymatic cleavage as well as for a more general state of the Aβ42 monomer.
Considering the MD studies, two parts of the trajectories were examined using replicas at
the lowest temperature (315 K). The simulations were started from a highly helical experimental
conformation of the peptide which mimicked an initial stage after cleavage and the first 50 ns were
selected to see how the presence or the absence of the ligand affected the helical content. Secondly,
15

the last 150 ns parts were considered from the same replicas to examine the effect of the ligand on
the conformation space of the Aβ42 monomer characterized by a higher disordered content.
Docking calculations followed the MD simulations and molecular targets were taken as
cluster representatives of the Aβ42 monomer derived from the last 150ns part of the replica at the
lowest temperature. These calculations allowed us to examine the ligand binding energies
concerning the Aβ42 monomer in a broader conformational scenario.
3.3.1. Maintenance of the original helical content
As was outlined in our previous articles [21,24], a possible strategy to decrease the toxicity
of the Aβ42 peptide is to avoid its transition to a β-hairpin conformation, which is related to the
formation of toxic oligomers. This goal can be achieved by stabilizing the early stage conformation
of the Aβ42 monomer characterized by its high α-helical or random coiled content. Therefore, we
employed the model from the PBD entry 1IYT as our starting configuration, which has been
determined in an apolar environment and is characterized by two α-helical regions (residues 8-25
and 28-39). Considering the first 50 ns from the REMD simulations we found that compound 7
inhibited the loss of helical structure especially on the C-terminal region compared to the pure Aβ42
monomer simulation (see Fig. 5).
Fig. 5. The α-helical content of the Aβ42 peptide in the presence (red) and the abscense (blue) of compound 7 taking the
first 50ns of the trajectories at the lowest temperature.
16

To elucidate the contact between the protein and the ligand we calculated their centre of
mass distances (the resulting contact distribution is shown in Fig. S2, Supplementary material). It
was found that the largest number of contacts between the ligand and the monomer peptide was
found within the helical regions in particular the one comprised between the residues 16 to 20. As
Fig. 5 presents, the helical content also increased partly in this region between K16LV18 residues. It
is worth to note that the hydrophobic interaction between the β-strand K16LVFFA21 and
N27KGAIIG33 regions has important role in the formation of the β-hairpin conformation.
3.3.2. Secondary structure elements progression
To have a general picture about the effect of compound 7 to the monomeric state of Aβ42
peptide, we compared secondary structure elements between the ligand+monomer and the free
monomer cases using the last 150 ns of the replicas with the lowest temperature. Difference
between the probability of α-helix (red), 310-helix (blue) and β-strand (green) for each residue is
presented in Fig. 6.
Fig. 6. Difference between the probability of secondary structure elements adopted by the Aβ42 peptide in the presence
and absence of compound 7 taking the last 150ns of the trajectories at the lowest temperature. Colour code: α-helix
(red), 310-helix (blue) and β-strand (green).
17

We found that in the presence of compound 7 the β-strand content was significantly
diminished at K16LVFFA21 (CHC), S26N27 and A30II32 segments. These residues are mainly
involved in the formation of hairpin structures in the Aβ42 toxic aggregates [45,46]. This difference
observed with the free ligand system had as counterpart increased 310-helix propensities between
S8GYEYH13 and CHC region. We also observed higher α-helical distribution between
N27KGAIIG33 residues. These results are consistent with theoretical [47] and experimental [48]
studies which posit that the preservation of these helical regions can stop oligomerization at the
dimer stage by impeding unfolding and β-sheet formation.
3.3.3. Contacts between Aβ42 monomer and compound 7 during REMD simulations
Following the original principle of the design we dissect compound 7 into three fragments
namely Fmoc, Asp and Tyr-OH. These segments presented different interactions with the Aβ42
monomer i.e. hydrophobic (Fmoc, Tyr-OH), ionic (Asp) or hydrogen-bond (Tyr-OH, peptide-
bonds). Then, we calculated the centre of mass distances between each residues of the protein and
the three fragments, where the critical distance for a contact was defined as less than or equal to 0.4
nm. Fig. 7 shows the contact distribution per residue between the Aβ42 monomer and the three
major parts of the ligand. The highly aromatic Fmoc group often forms contacts with residues 12-19
and 28-38 providing a putative explanation for the inhibitory capacity of compound 7 on the
conformational transition of the Aβ42 monomer.
18

Fig. 7. Distribution of the contacts between compound 7 and the Aβ42 monomer in the last 150ns of the trajectory taking
the lowest temperature.
3.3.4. Estimating potential binding modes
To have a more depicted description of putative binding modes between compound 7 and
the Aβ42 monomer different docking calculations were performed. Here, we considered the
dynamical character of the peptide by generating multiple docking targets. The targets were derived
by clusterization of the last 150 ns of the lowest temperature Aβ42 monomer trajectory taken from
the previous REMD study. According to this, 4602 clusters were generated, of which only the ones
with more than 100 elements were selected. This way 21 clusters representants were selected to
perform blind docking with the quickest SP method. The best docking scores of each target are
reported in Table S1 (Supplementary material)
To refine our blind docking results, we selected 5 complexes with the best SP docking
values and performed XP and IFD calculations. Refining docking was restricted to a 20 Å
environment of the poses provided by the blind docking SP calculations. All Glide score values (SP,
XP and IFD) are shown in Table 2. As can be observed, score rankings were not maintained when
using the different docking approaches.
Table 2. Single precision, extra precision and induced fit docking scores.
Nevertheless, the analysis of the highest and lowest binding energies provided hints about
relevant interactions between compound 7 and the Aβ42 monomer. For example, in the case of
cluster 787 both SP and IFD showed the higher scorings among the five analysed systems while XP
method presented the weakest interaction with the peptide. This main difference could be explained
considering how in the first two methods compound 7 formed π-stacking interactions between the
Fmoc group and aromatic residues His14 and Tyr10 (SP and IFD methods, respectively) while in the
XP case this interaction was lost leaving the hydrophobic group exposed to the solvent and so
reducing its binding affinity to the Aβ42 peptide. Other preferred binding modes, e.g. the one
obtained in cluster 3354 using the IFD method showed the same peculiarity having π-stacking
19

interactions with Tyr10. Another important feature in this particular binding mode were two salt-
bridges formed between Lys28 and Arg5 with the COO- groups of 7 (Fig. 8).
Overall, the close interaction of compound 7 burying its hydrophobic group in the CHC of
Aβ42 regardless of the peptide conformation coincides with the results observed in the REMD
trajectory where the contact distribution of the ligand was located in this region of the Aβ42
monomer. This interaction is in line with the capacity of the ligand to inhibit the conformational
transition in the L17VFFA21 region towards higher contents of β-strand structures.
Fig. 8. 2-D diagrams of 7-Aβ42 complexes for cluster 787 employing different docking methods: SP (A), IFD (B), XP
(C) and cluster 3354 employing IFD (D).
4. Conclusions
In a previous work we reported a mimetic peptide, DZK, possessing a potent inhibitory
effect on Aβ42 aggregate formations [21]. Here, we investigated experimentally a new series of
mimetic peptides based on the DZK molecule completed with detailed theoretical analysis for a
20

selected case. Among the tested molecules, compound 7 possessed a higher inhibitory potency at
ThT-positive Aβ42 aggregation than the original DZK molecule. This compound was selected for a
more extensive physico-chemical evaluation including a ThT assay monitoring fluorescence
intensity at different times and TEM studies, as well as a cell viability assay to obtain further
insights into the prevention of Aβ42 -mediated cell toxicity.
Moreover, different computational modelling techniques were employed to identify the
interactions of compound 7 with early stage Aβ42 monomer and look for possible explanations of
how this mimetic peptide affects the conformational transition of the peptide. These simulations
enabled us to obtain a picture for the experimentally hard to attain, very early stage interaction with
the ligand at atomic level. REMD simulations showed that compound 7 diminished the formation of
aggregation-prone structures by maintaining helical content which were reproduced by our ligand-
free control simulations. It was also found that the β-strand structure was significantly diminished at
CHC region (residues K16LVFFA21). Regarding helical structures, the α-helical distribution was
higher at N27KGAIIG33 residues and 310-helix propensities were increased at N-terminal and CHC
regions in the presence of compound 7. These secondary structure results are in good agreement
with experimental [46] and theoretical [47] studies where it had been indicated that the loss of
helical content can be observed during the early stage of toxic species formation.
Molecular docking studies revealed that compound 7 interacted with the Aβ42 monomer
through π-stacking interactions between its Fmoc group and the aromatic residues His14 or Tyr10.
Compound 7 buries its hydrophobic Fmoc group in the CHC leading to the dominance of helical
conformation in this region. This result corroborates the REMD simulations where the contact
distribution of the ligand is also enhanced at the CHC region of the Aβ42 monomer.
In conclusion, our experimental and theoretical work demonstrated that compound 7
modulates the early steps of the amyloidogenic process also inhibiting the cytotoxicity linked to
Aβ42 aggregation with full viability recovery of cultured cells under submicromolar concentrations
of the ligand.
Acknowledgments
Grants from Universidad Nacional de San Luis (UNSL), partially supported this work. This
research was also supported by Agencia Nacional de Promoción Científica y Tecnológica (PICT
2015-1769). L.M.; C.M.L.D and E.G.M. thank CONICET, ANPCyT and UNR for support. E.E.B.
is beneficiary of a postdoctoral fellowship of CONICET (Consejo Nacional de Investigaciones
Científicas y Técnicas, Argentina). Some of the authors (J.G., T.Sz., L.F. and G.P.) acknowledge
21

financial support from the Hungarian government (GINOP-2.3.2-15-2016-00034) and from the
Hungarian Scientific Research Fund (OTKA) K111862.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in the online version, at http:/
22

References
[1] I. Melnikova, Therapies for Alzheimer’s disease, Nat. Rev. Drug Discovery 6 (2007) 341-352.
[2] H.W. Querfurth, F.M. LaFerla, Alzheimer’s disease, N. Engl. J. Med. 362 (2010) 329-344.
[3] 2012 Alzheimer’s disease facts and figures, Alzheimer's & Dementia 8 (2012) 131-168.
[4] J. Hardy, D.J. Selkoe, The amyloid hypothesis of Alzheimer’s disease: progress and problems
on the road to therapeutics, Science 297 (2002) 353-356.
[5] A.S. Cohen, L.A. Jones, Amyloid and amyloidosis, Curr. Opin. Rheumatol. 3 (1991) 125-138.
[6] M.A. Findeis, The role of amyloid beta peptide 42 in Alzheimer’s disease, Pharmacol. Ther. 116
(2007) 266-286.
[7] D.M. Walsh, D.J. Selkoe, Aβ oligomers - A decade of discovery, J. Neurochem. 101 (2007)
1172-1184.
[8] E. Hayden, D. Teplow, Amyloid β-protein oligomers and Alzheimer’s disease, Alzheimer's Res.
Ther. 5 (2013) 60.
[9] J.E. Gestwicki, G.R. Crabtree, I.A. Graef, Harnessing chaperons to generate small-molecule
inhibitors of amyloid β aggregation, Science 306 (2004) 865-869.
[10] A.J. Doig, Peptide inhibitors of β-amyloid aggregation. Curr. Opin. Drug Discovery Dev. 10
(2007) 533-539.
[11] C. Yang, X. Zhu, J. Li, R. Shi, Exploration of the mechanism for LPFFD inhibiting the
formation of β-sheet conformation of Aβ(1–42) in water, J. Mol. Model. 16 (2010) 813–821.
[12] Y. Cote, G.G. Maisuradze, P. Delarue, H.A. Scheraga, P. Senet, New insights into protein
(Un)folding dynamics, J. Phys. Chem. Lett. 6 (2015) 1082–1086.
[13] R.P. Bywater, Protein folding: A problem with multiple solutions, J. Biomol. Struct. Dyn. 31
(2013) 351-362.
[14] D. Goyal, S. Shuaib, S. Mann, B. Goyal, Rationally designed peptides and peptidomimetics as
inhibitors of amyloid-β (Aβ) aggregation: Potential therapeutics of Alzheimer’s disease. ACS
Comb. Sci. 19 (2017) 55–80.
[15] D. Galimberti, E. Scarpini, Emerging amyloid disease-modifying drugs for Alzheimer’s
disease, Expert Opin. Emerging Drugs 21 (2016) 5-7.
[16] Z.-L. Zhou, Y. Ho, H.-L. Liu, P. Elumalai, W.-H. Chen, Computer-aided discovery of novel
non-peptide inhibitors against amyloid-beta (Aβ) peptide aggregation for treating Alzheimer’s
disease, Mol. Simul. 41 (2015) 622-632.
[17] L. Bellucci, A. Ardèvol, M. Parrinello, H. Lutz, H. Lu, T. Weidner, S. Corni, The interaction
with gold suppresses fiber-like conformations of the amyloid β(16–22) peptide, Nanoscale 8 (2016)
8737-8748.
23

[18] G.S. da Silva, M. Figueirό, C.F. Tormena, F. Coelho, W.P. Almeida, Effects of novel
acylhydrazones derived from 4-quinolone on the acetylcholinesterase activity and Aβ42 peptide
fibrils formation, J. Enzyme Inhib. Med. Chem. 1 (2016) 1-7.
[19] W. Xu, X.-B. Wang, Z.-M. Wang, J.J. Wu, F. Li, J. Wang, L.-Y. Kong, Synthesis and
evaluation of donepezil–ferulic acid hybrids as multi-target-directed ligands against Alzheimer’s
disease, Med. Chem. Commun. 7 (2016) 990-998.
[20] J. Nasica-Labouze, P.H. Nguyen, F. Sterpone, O. Berthoumieu, N. Buchete, S. Cotꢀ, A. De
Simone, A.J. Doig, P. Faller, A. Garcia, A. Laio, M. Suan Li, S. Melchionna, N. Mousseau, Y. Mu,
A. Paravastu, S. Pasquali, D.J. Rosenman, B. Strodel, B. Tarus, J.H. Viles, T. Zhang, C. Wang, P.
Derreumaux, Amyloid β Protein and Alzheimer’s Disease: When Computer Simulations
Complement Experimental Studies, Chem. Rev. 115(2015) 3518-3563.
[21] E.E. Barrera Guisasola, S.A. Andujar, E. Hubin, K. Broersen, I.M. Kraan, L. Méndez, C.M.L.
Delpiccolo, M.F. Masman, A.M. Rodríguez, R.D. Enriz, New mimetic peptides inhibitors of Αβ
aggregation. Molecular guidance for rational drug design, Eur. J. Med. Chem. 95 (2015) 136-152.
[22] E.E. Barrera Guisasola, L.J. Gutierrez, S.A. Andujar, E. Angelina, A.M. Rodríguez, R.D.
Enriz, Pentameric models as alternative molecular targets for the design of new antiaggregant
agents, Curr. Protein Pept. Sci. 17 (2016) 156-168.
[23] M.F. Masman, U.L.M. Eisel, I.G. Csizmadia, B. Penke, R.D. Enriz, S.J. Marrink, P.G.M.
Luiten, In silico study of full-length amyloid 1-42 tri- and penta-oligomers in solution, J. Phys.
Chem. B. 113 (2009) 11710-11719.
[24] E.E. Barrera Guisasola, L.J. Gutiérrez, R.E. Salcedo, F.M. Garibotto, S.A. Andujar, R.D.
Enriz, A.M. Rodríguez, Conformational transition of Aβ42 inhibited by a mimetic peptide. A
molecular modeling study using QM/MM calculations and QTAIM analysis, Comput. Theor.
Chem. 1080 (2016) 56-65.
[25] F. Albericio, Solid-phase synthesis: A practical guide, CRC Press, 2000.
[26] N.L. Benoiton, Chemistry of peptide synthesis, Taylor & Francis, London, 2005.
[27] M. Biancalana, S. Koide, Molecular mechanism of thioflavin-T binding to amyloid fibrils.
Biochim, Biophys. Acta 1804 (2010) 1405–1412.
[28] K. Broersen, W. Jonckheere, J. Rozenski, A. Vandersteen, K. Pauwels, A. Pastore, F.
Rousseau, J. Schymkowitz, A standardized and biocompatible preparation of aggregate-free
amyloid beta peptide for biophysical and biological studies of Alzheimer’s disease, Protein Eng.
Des. Sel. 9 (2011) 743-750.
[29] Z. Bozso, B. Penke, D. Simon, I. Laczko, G. Juhasz, V. Szegedi, A. Kasza, K. Soos, A.
Hetenyi, E. Weber, H. Tohati, M. Csete, M. Zarandi, L. Fulop, Controlled in situ preparation of A
24

beta(1-42) oligomers from the isopeptide "iso-A beta(1-42)", physicochemical and biological
characterization, Peptides 31 (2010) 248-256.
[30] O. Crescenzi, S. Tomaselli, R. Guerrini, S. Salvadori, A.M. D'Ursi, P.A. Temussi, D. Picone,
Solution structure of the Alzheimer amyloid β-peptide (1–42) in an apolar microenvironment, Eur.
J. Biochem. 269 (2002) 5642-5648.
[31] S.S Sisodia, Beta-amyloid precursor protein cleavage by a membrane-bound protease, PNAS
89 (1992) 6075-6079.
[32] E. Polak, G. Ribiere, Note sur la Convergence de Methods de Directions Conjugres, Revue
Francaise Informat. Recherche Operationnelle 16 (1969) 35–43.
[33] 2017-3, S. R. MacroModel, New York, 2017.
[34] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman, G.
Scalmani, V. Barone, G.A. Petersson, H. Nakatsuji, X. Li, M. Caricato, A. Marenich, J. Bloino,
B.G. Janesko, R. Gomperts, B. Mennucci, H.P. Hratchian, J.V. Ortiz, A.F. Izmaylov, J.L.
Sonnenberg, D. Williams-Young, F. Ding, F. Lipparini, F. Egidi, J. Goings, B. Peng, A. Petrone, T.
Henderson, D. Ranasinghe, V.G. Zakrzewski, J. Gao, N. Rega, G. Zheng, W. Liang, M. Hada, M.
Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai,
T. Vreven, K. Throssell, J.A. Montgomery, Jr., J.E. Peralta, F. Ogliaro, M. Bearpark, J.J. Heyd, E.
Brothers, K.N. Kudin, V.N. Staroverov, T. Keith, R. Kobayashi, J. Normand, K. Raghavachari, A.
Rendell, J.C. Burant, S.S. Iyengar, J. Tomasi, M. Cossi, J.M. Millam, M. Klene, C. Adamo, R.
Cammi, J.W. Ochterski, R.L. Martin, K. Morokuma, O. Farkas, J.B. Foresman, and D.J. Fox
Gaussian 09, Revision A.02, Gaussian, Inc.: Wallingford CT, 2016.
[35] J. Wang, W. Wang, P.A. Kollman, D.A. Case, Automatic atom type and bond type perception
in molecular mechanical calculations, J. Mol. Graph. Model. 25 (2006) 247-260.
[36] J. Wang, R.M. Wolf, J.W. Caldwell, P.A. Kollman, D.A. Case, Development and testing of a
general amber force field, J. Computat. Chem. 25 (2004) 1157-1174.
[37] A. Patriksson, D. van der Spoel, A temperature predictor for parallel tempering simulations,
Phys. Chem. Chem. Phys. 10 (2008) 2073-2077.
[38] G. Bussi, M. Parrinello, Accurate sampling using Langevin dynamics, Phys. Rev. E 75 (2007)
056707.
[39] M. Parrinello, A. Rahman, Polymorphic transitions in single crystals: A new molecular
dynamics method, J. Appl. Phys. 52 (1981) 7182-7190.
[40] B. Hess, P-LINCS:ꢁ A parallel linear constraint solver for molecular simulation, J. Chem.
Theory Comput. 4 (2008) 116-122.
[41] T. Darden, D. York, L. Pedersen, Particle mesh Ewald: An N⋅log(N) method for Ewald sums
in large systems, J. Chem. Phys. 98 (1993) 10089-10092.
25

[42] G. Nagy, C.Oostenbrink, Dihedral-based segment identification and classification of
biopolymers I: Proteins, J. Chem. Inf. Model. 54 (2014) 266-277.
[43] R.A. Friesner, R.B. Murphy, M.P. Repasky, L.L. Frye, J.R. Greenwood, T.A. Halgren, P.C.
Sanschagrin, D.T. Mainz, Extra precision glide:ꢁ docking and scoring incorporating a model of
hydrophobic enclosure for protein−ligand complexes, J. Med. Chem. 49 (2006) 6177-6196.
[44] R. Farid, T. Day, R.A. Friesner, R.A. Pearlstein, New insights about HERG blockade obtained
from protein modeling, potential energy mapping, and docking studies, Bioorg. Med. Chem. 14
(2006) 3160-3173.
[45] W.P. Esler, E.R. Stimson, J.R. Ghilardi, Y.A. Lu, A.M. Felix, H. V Vinters, P.W. Mantyh, J.P.
Lee, J.E. Maggio, Point substitution in the central hydrophobic cluster of a human beta-amyloid
congener disrupts peptide folding and abolishes plaque competence, Biochemistry 35 (1996)
13914–13921.
[46] N.S. de Groot, F.X. Aviles, J. Vendrell, S. Ventura, Mutagenesis of the central hydrophobic
cluster in Abeta42 Alzheimer’s peptide. Side-chain properties correlate with aggregation
propensities, FEBS J. 273 (2006) 658–668.
[47] A. V Rojas, A. Liwo, H.A. Scheraga, A study of the α-helical intermediate preceding the
aggregation of the amino-terminal fragment of the β amyloid peptide (Aβ(1-28)), J. Phys. Chem. B
115 (2011) 12978–12983.
[48] A. Kapurniotu, A. Buck, M. Weber, A. Schmauder, T. Hirsch, J. Bernhagen, M. Tatarek-
Nossol, Conformational Restriction via Cyclization in β-Amyloid Peptide Aβ(1-28) Leads to an
Inhibitor of Aβ(1-28) Amyloidogenesis and Cytotoxicity, Chem. Biol. 10 (2003) 149–159.
26

Figure captions
Fig. 1. DZK structure (core: in magenta, R1: in blue, R2: in red, R3: in green). Structure features of
compounds 1-8
Fig. 2. Aβ42 amyloid aggregation is affected in a dose-dependent manner by compound 7. Dose-
response amyloid fibril formation of 25 μM Aβ42 incubated in the presence of compound at 37° C
for 7 days was monitored using ThT fluorescence intensity at 485 nm. Values represent results of
three independent replicates. Statistical significance of the results was established by P-values using
paired two-tailed t-tests. Statistical significance levels were * P < 0.05, **P < 0.005 and ***P <
0.0005.
Fig. 3. Cell viability assay using the CellTiter-Blue reagent with SHSY-5Y cells exposed to 25 μM
Aβ42 in presence or absence of compound 7. Controls include 8% μM solution of sodium dodecyl
sulfate, full-length Aβ42, compound and cells in buffer. The results are represented as mean values ±
standard error of the mean. Values represent results of three independent replicates. Significant p-
values are indicated by: *P < 0.05, **P < 0.01 and ***P < 0.001.
Fig. 4. Representative transmission electron microscopy images of Aβ42 incubated either in the
presence of DMSO or 25 µM compound 7. Images were taken at three time points. 0 min: Aβ42
with DMSO (A) or 25 µM compound 7 (B). 24 h: Aβ42 with DMSO (C) or 25 µM compound 7 (D).
168 h: Aβ42 with DMSO (E) or 25 µM compound 7 (F). Scale bar: 500 nm.
Fig. 5. The α-helical content of the Aβ42 peptide in the presence (red) and the abscense (blue) of
compound 7 taking the first 50ns of the trajectories at the lowest temperature.
Fig. 6. Difference between the probability of secondary structure elements adopted by the
Aβ42 peptide in the presence and absence of compound 7 taking the last 150ns of the trajectories at
the lowest temperature. Colour code: α-helix (red), 310-helix (blue) and β-strand (green).
Fig. 7. Distribution of the contacts between compound 7 and the Aβ42 monomer in the last 150ns of
the trajectory taking the lowest temperature.
Fig. 8. 2-D diagrams of 7-Aβ42 complexes for cluster 787 employing different docking methods: SP
(A), IFD (B), XP (C) and cluster 3354 employing IFD (D).
27

Highlights:
A novel mimetic peptide that inhibits Aβ cytotoxicity at submicromolar concentration is reported
Antiaggregant activities were measured using different biophysical assays
Complex formation and its dynamic behaviour was analysed using docking and enhanced sampling
MD
Docking posits compound 7 interacts with the CHC region implicated in Aβ oligomerization
The mimetic peptide would act through the inhibition of the conformational transition of Aβ42
Graphical abstract
28