Design and synthesis of β-strand-fixed peptides inhibiting aggregation of amyloid β-protein

Article abstract of DOI:10.1016/j.bmc.2020.115676

Aggregation of 42-residue amyloid β-protein (Aβ42) can be prevented by β-sheet breaker peptides (BSBps) homologous to LVFFA residues, which are included in a β-sheet region of Aβ42 aggregates. To enhance the affinity of BSBps to the Aβ42 aggregates, we de

Full text of DOI:10.1016/j.bmc.2020.115676

Journal Pre-proofs
Design and synthesis of β-strand-fixed peptides inhibiting aggregation of
amyloid β-protein
Fumiya Tanaka, Kana Shibata, Yoko Monobe, Ken-ichi Akagi, Yuichi
Masuda
PII:
DOI:
Reference:
S0968-0896(20)30506-X
https://doi.org/10.1016/j.bmc.2020.115676
BMC 115676
To appear in:
Bioorganic & Medicinal Chemistry
Received Date:
Revised Date:
Accepted Date:
22 May 2020
13 July 2020
17 July 2020
Please cite this article as: Tanaka, F., Shibata, K., Monobe, Y., Akagi, K-i., Masuda, Y., Design and synthesis of
β-strand-fixed peptides inhibiting aggregation of amyloid β-protein, Bioorganic & Medicinal Chemistry (2020),
doi: https://doi.org/10.1016/j.bmc.2020.115676
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version
will undergo additional copyediting, typesetting and review before it is published in its final form, but we are
providing this version to give early visibility of the article. 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.
© 2020 Published by Elsevier Ltd.

Design and synthesis of -strand-fixed peptides inhibiting aggregation of amyloid
-protein
Fumiya Tanakaa, Kana Shibataa, Yoko Monobeb,1, Ken-ichi Akagib,2, and Yuichi Masudaa,*
aGraduate School of Bioresources, Mie University, 1577 Kurimamachiya-cho, Tsu, 514-8507,
Japan
bNational Institute of Biomedical Innovation, Health and Nutrition, Osaka 567-0085, Japan
*Corresponding author: Tel.: +81-59-231-9620; E-mail: masuda@bio.mie-u.ac.jp
Present Addresses
1Hanaichi UltraStructure Research Institute, Co., Ltd. 36 Teramae, Ida-cho-aza, Okazaki,
Aichi, 444-0076, Japan
2Environmental Metabolic Analysis Research Team, RIKEN Center for Sustainable Resource
Science, Yokohama, Kanagawa 230-0045, Japan
1

Abstract
Aggregation of 42-residue amyloid -protein (A42) can be prevented by -sheet breaker
peptides (BSBps) homologous to LVFFA residues, which are included in a -sheet region of
A42 aggregates. To enhance the affinity of BSBps to the A42 aggregates, we designed and
synthesized -strand-fixed peptides (BSFps) whose side chains were cross-linked by ring
closing metathesis. Conformation analysis verified that the designed peptides could be fixed
in -strand conformation. Among the synthesized pentapeptides, 1 and 12, whose side chains
of 2nd and 4th residues were cross-linked, significantly inhibited the aggregation of A42.
This suggested that -strand-fixation of BSBps could enhance their inhibitory activity against
the A42 aggregation. However, pentapeptides 1 and 12 had little effect on morphology of
A42 aggregates (fibrils) and neurotoxicity of A42 against SH-SY5Y cells.
Keywords:
Amyloid -protein, -sheet breaker peptide, -strand mimetics, conformation fixation,
conformation analysis
2

Abbreviations
AD, Alzheimer’s disease; A, amyloid -proteins; BSBp, -sheet breaker peptide; BBB,
blood-brain
barrier;
Tyr(Allyl),
O-allyl-tyrosine;
Gly(Allyl),
allylglycine;
BSFp, -strand-fixed peptides; HBTU, [Bis(dimethylamino)methylene]-1H-benzotriazolium
3-Oxide Hexafluorophosphate; DIEA, N,N-diisopropylethylamine; RCM, ring-closing
metathesis; DMSO, dimethyl sulfoxide; NOE, nuclear Overhauser effect; Th-T, thioflavin-T,
TEM,
transmission
electron
microscopy;
MTT,
3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; RP-HPLC, reversed-phase high
performance liquid chromatography; HFIP, 1,1,1,3,3,3-hexafluoro-2-propanol; PBS,
phosphate buffered saline
3

1. Introduction
Alzheimer’s disease (AD), the leading cause of dementia, is neuropathologically
characterized by appearance of senile plaques in the cerebral cortex.1 Senile plaques are
mainly composed of aggregated 40- and 42-mer amyloid -proteins, called A40 and A42,
3
respectively (Figure 1A).2, After excised from amyloid precursor protein, A forms
intermolecular -sheets to aggregate. Aggregation of A induces neurotoxicity,4 which is
thought to be the main cause of AD. Since A42 is far more aggregative than A40,5
inhibition of A42 aggregation is expected to be an effective treatment for AD.
The structure of the A42 aggregates (fibrils) has been intensively studied to elucidate the
7
aggregation mechanism.6, The A42 aggregates were revealed to form intermolecular
parallel -sheets, in which -strands (a ‘-strand’ is a stretch of amino acids whose peptide
backbones are almost fully extended) are connected with intermolecular hydrogen bonds.8-10
Hydrophobic residues at the positions 17−21 (LVFFA) of A42 are considered to play critical
roles in the aggregation by forming a core -sheet structure in the central region (Figure 1A).
In addition, intermolecular – stacking of two successive phenylalanines at the positions 19
and 20 might help the parallel alignment of -strands.11, 12 The aggregation of A42 can be
hampered in the presence of short peptides homologous to the ‘LVFFA’ residues (known as
-sheet breaker peptide, BSBp). Soto et al. first synthesized a series of BSBps and developed
a famous BSBp called iA5 (LPFFD).13, 14 So far, many researchers have developed BSBp
with good inhibitory activity against the A aggregation in vitro.15, 16 However, the peptide
backbones should be susceptible to proteolytic degradation and hard to cross the blood-brain
barrier (BBB). To overcome these problems, BSBps have been subjected to various chemical
18
modifications, such as N-methylation17,
and incorporation of unnatural amino acids
including D-isomers.19, 20 These modifications could improve their biological stability and
BBB permeability, showing a great potential of BSBp for becoming a drug for AD.
4

Although a variety of BSBps have been developed, their inhibitory activities were not
sufficient for practical use. In general, to enhance the affinity of the ligands to their target
proteins, it is necessary to consider enthalpic as well as entropic contribution. In case of
BSBp, conformation fixation of BSBp to a -strand conformation would be a good approach
for reducing the entropy loss upon binding. Arai et al. designed and synthesized cyclic BSBps
to identify the pharmacophores by stabilizing the active conformation.21, 22 Quite recently, Jha
et al. reported a promising example of a conformationally restricted BSBp.23 They developed
an octapeptide with a -hairpin structure stabilized by a β-turn-inducing segment (DPro-Gly
motif) and a disulfide cross-linkage. The β-hairpin-containing peptide was found to inhibit the
aggregation and neurotoxicity of Aβ42. This report demonstrated that the
conformation-fixation could be a powerful approach for enhancing the inhibitory activity of
BSBp.
The molecular size of BSBps should be as small as possible to keep the good BBB
permeability for AD drug. Although many methods for -strand-fixation of the peptide
backbone have been proposed,24, 25 a compact motif of a -strand is preferred for the design of
conformation-fixed BSBp. Abell et al. designed and synthesized a tripeptide, in which, the
side chains of the 1st and 3rd residues (O-allyl-tyrosine (Tyr(Allyl)) and allylglycine
(Gly(Allyl)), respectively) were cross-linked with (E)-alkene by ring-closing metathesis
(RCM) (Figure 1B).26, 27 X-ray crystal analysis of the synthesized peptide revealed its -strand
geometry (Figure 1C).27 Abell et al. also reported that -strand fixation of the peptides could
enhance their inhibitory activity against the calcium-activated cysteine protease calpain.26 It is
noteworthy that Abell’s motif can fix the peptide conformation in a -strand without
increasing its molecular size. Moreover, this motif includes an aromatic ring, which could
disrupt the intermolecular - stacking of the phenylalanine residues at the positions 19 and
20 of the A42. To enhance the inhibitory activity of BSBps against the A42 aggregation,
we designed and synthesized a series of -strand-fixed peptides (BSFps) whose side chains
are cross-linked.
5

(A)
1
17
21
40 42
DAEFRHDSGYEVHHQKLVFFAEDVGSNKGAIIGLMVGGVVIA
(B)
(C)
O
O
H
BocHN
N
N
OMe
H
O
O
Figure 1. (A) Amino acid sequence of A42. ‘LVFFA’ region is underlined. Chemical
structure (B) and X-ray crystal structure (C) of a -strand-fixed tripeptide developed by Abell
et al.27
2. Results
2.1. BSFps with Abell’s motif (1–4)
2.1.1. Design and synthesis of 1–4
A17–21 (LVFFA) pentapeptide was chosen as a template for designing the BSFps to target
the core -sheet structure in the central region of the A42 aggregates (Figure 1A). In a
general structural model of the β-strands, the side chains of neighboring residues face in the
opposite directions, resulting in the same direction of every second side chain. Hence, we
designed an A17–21-based pentapeptide 1 (Figure 2), in which, the side chains of 2nd and 4th
residues were cross-linked by RCM to mimic the -strand conformation using the Abell’s motif
(Figure 1B and 1C).26, 27 Since the incorporation of Asp and Pro into BSBp could enhance their
inhibitory activity,13, 14 2–4, bearing Asp and/or Pro residues at the termini, were also designed
(Figure 2).
6

Phe
Phe
Asp
Leu
Leu
O
O
Ala
O
O
O
O
OH
OH
H
H
H
H
N
N
OH
N
N
H2N
N
H
N
H2N
N
N
H
H
H
O
O
O
O
O
Tyr(Allyl-RCM)
Gly(Allyl-RCM)
Tyr(Allyl-RCM)
Gly(Allyl-RCM)
O
O
1
2
Phe
Phe
Asp
O
Pro
Pro
Ala
O
O
OH
OH
O
O
H
H
H
H
N
N
N
N
N
N
OH
N
H
N
N
H
N
H
O
H
O
H
H
O
O
O
O
Tyr(Allyl-RCM)
Gly(Allyl-RCM)
Tyr(Allyl-RCM)
Gly(Allyl-RCM)
O
O
3
4
Figure 2. Chemical structures of 1–4.
Our preliminary synthetic study found that the side-chain cross-linking of the peptides in
the early-stage lead to low yield because the cross-linked peptides are poorly soluble in any
solvent. Therefore, synthesis of 1–4 was conducted by solution-phase peptide elongation
followed by ring-closing metathesis of dienes (Scheme 1). The starting material 5 (Boc-Tyr
(Allyl)-OMe) was prepared as reported previously.28 After alkaline hydrolysis of the methyl
ester 5, the resultant carboxylic acid was coupled with H-Phe-OMe using
1-[Bis(dimethylamino)methylene]-1H-benzotriazolium 3-Oxide Hexafluorophosphate
(HBTU) and N,N-diisopropylethylamine (DIEA) to give dipeptide 6. Hydrolysis of the methyl
ester in 6 and the subsequent coupling with H-Gly(Allyl)-OMe provided tripeptide 7, which
can be used as a common synthetic intermediate for the syntheses of 1–4. After hydrolysis of
the methyl ester in 7, the resultant carboxylic acid was condensed with H-Ala-OMe and
H-Asp(OMe)-OMe to form tetrapeptide 8a and 8b, respectively. After removal of Boc groups
from 8a and 8b, the resultant amines were acylated with Boc-Leu-OH and Boc-Pro-OH to
give pentapeptides 9a–9d. RCM of dienes 9a–9d using Hoveyda-Grubbs’ 2nd generation
catalyst under high dilution condition resulted in 10a–10d in high yields. (E)-configurations
7

of 10a–10d were confirmed by the characteristic coupling constants (>15 Hz for trans) of the
corresponding two vinylic protons. Few (Z)-isomers were detected in all the RCM products.
In Abell’s synthesis of -strand-fixed tripeptide (Figure 1B), 10mol% Grubbs’ 2nd generation
catalyst was used for RCM of the dienes in 1,1,2-trichloroethane under thermal or microwave
heating.26, 27 They also reported that the addition of chlorodicyclohexylborane as Lewis acid
markedly improved the yield of the RCM.27 In contrast, we succeeded in high-yield RCM of
the dienes 9a–9d using 5mol% of Hoveyda-Grubbs’ 2nd generation catalyst without additive
in dichloromethane under reflux condition. Finally, hydrolysis of the methyl esters and
subsequent removal of the Boc groups from 10a–10d provided the desired 1–4.
O
1) NaOH aq
MeOH/dioxane
rt, 1 h
1) NaOH aq
MeOH
rt, 1 h
O
O
O
BocHN
H
N
OMe
O
BocHN
BocHN
OMe
N
H
OMe
N
H
O
O
2) H-Gly(Allyl)-OMe
HBTU, DIEA
CH2Cl2
2) H-Phe-OMe
HBTU, DIEA
CH2Cl2
O
O
rt, 2 h
rt, 2 h
93% for 2 steps
99% for 2 steps
5
6
7
1) NaOH aq
MeOH/dioxane
rt, 2 h
O
O
Y
X
H
O
O
Y
H
H
N
1) 3 M HCl/dioxane
rt, 2–6 h
BocHN
N
OMe
N
OMe
N
H
N
BocRN
N
H
N
H
H
O
O
O
O
O
2) Boc-Xaa-OH
HBTU, DIEA
CH2Cl2
2) H-Yaa-OMe
HBTU, DIEA
CH2Cl2
rt, 2–4 h
O
O
rt, 2–3 h
9a
: X=Leu (R=H), Y=Ala, 93%
: X=Leu (R=H), Y=Asp(OMe), 85%
: X=Pro, Y=Ala, 76%
8a
8b
: Y=Ala, 98%
: Y=Asp(OMe), 97%
9b
9c
9d
: X=Pro, Y=Asp(OMe), 82%
X
O
O
Y
X
O
O
Y
Hoveyda-Grubbs
2nd cat.
1) NaOH aq
MeOH/dioxane
rt, 2–5 h
H
H
H
H
N
N
OMe
N
N
OH
BocRN
N
N
H
RHN
N
N
(5mol%)
H
H
H
O
O
O
O
O
O
CH2Cl2 (10 mM)
reflux, 21–43 h
2) 3 M HCl/dioxane
rt, 2–7 h
O
O
10a
10b
10c
10d
1
2
3
4
: X=Leu (R=H), Y=Ala, 98%
: X=Leu (R=H), Y=Asp(OMe), 86%
: X=Pro, Y=Ala, quant
: X=Leu (R=H), Y=Ala, 40%
: X=Leu (R=H), Y=Asp, 21%
: X=Pro, Y=Ala, 18%
: X=Pro, Y=Asp(OMe), 86%
: X=Pro, Y=Asp, 48%
Scheme 1. Synthesis of 1–4
8

2.1.2. Conformational analysis of 1–4
To verify whether the conformations of 1–4 were fixed to -strand, their 3D structures
were analyzed by NMR. NMR experiments for structural analysis were performed using
dimethyl sulfoxide (DMSO)-d6 as the solvent because all the BSFps were soluble in it. Since
2D NOESY experiments did not indicate any nuclear Overhauser effect (NOE) that can be
used for distance information, dihedral angle information based on J-coupling constants
between vicinal protons was used for conformation analysis. It has been reported that the
amide proton-C proton coupling constants (3JHN) in the peptides and proteins correlate well
with their torsion angles .29, 30 Since dihedral angles  of the peptides and proteins are
3
dependent on the torsion angles , JHN is known to correlate with the local conformation;
3JHN value is typically 8–10 Hz for a β-strand conformation, 6–8 Hz for a random coil, and
3
less than 6 Hz for an -helix. JHN values of 1–4 were measured by 1D and 2D NMR
experiments, which are summarized in Table 1. It is noteworthy that all the 3JHN values at the
4th residues (Gly(Allyl-RCM)) were considerably larger than 8 Hz, suggesting the formation
3
of -strand conformation at this position of 1–4. On the other hand, JHN values at the 2nd,
3rd, and 5th residues were 6–8 Hz, indicating that these positions might not be fully fixed to
the -strand conformation.
3D structures of 1–4 were estimated by the molecular mechanics calculation on
CONFLEX 8 program.31-33 Molecular modelling was performed using the distance geometry
followed by a conformational search based on the reservoir-filling algorithm with dihedral
angle constraints derived from the NMR data. The -angles at the 4th residues of 1–4 were set
3
at around 180°, which constrain the position in a -strand conformation. Since the JHN
values of 6–8 Hz can be observed in any conformations, the -angles at the 2nd, 3rd, and 5th
residues were not constrained for the calculations. The search limit was set at 3 kcal/mol
9

above the global energy minimum. For energy minimization, we applied MMFF94s force
field with a dielectric constant value of 46.45 (the dielectric constant for DMSO at 25 °C).
The above calculations yielded stable conformations of 1–4 as shown in Figure 3. The 
and  angles of the lowest energy conformers (Figures 3A, C, E, and G) are summarized in
Table S1. It is well known that the amino acid with torsion angles in the range of ‒180° <  <
‒45° and 45° <  < 225° are considered to be in the -sheet region.34, 35 According to this
empirical rule, the  and  angles of 1–4 (Table S1) suggested that all the residues form a
-strand conformation. The similarity of the macrocyclic frames in the superpositions of the
top-10 stable conformers (Fig. 3B, D, F, and H) supported their rigidity.
Table 1. 3JHN (Hz) values for 1–4 in DMSO-d6 at 25 °C.
Compound
Residue No.
Residue
3JHN (Hz)
N.D.a
7.1
1
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
Leu
Tyr(Allyl-RCM)b
Phe
6.8
Gly(Allyl-RCM)b
9.2
Ala
7.2
N.D. a
2
3
4
Leu
Tyr(Allyl-RCM) b
7.3
Phe
6.7
Gly(Allyl-RCM) b
9.2
Asp
6.0
N.D. a
Pro
Tyr(Allyl-RCM) b
7.1
Phe
6.9
Gly(Allyl-RCM) b
9.1
Ala
7.0
N.D. a
Pro
Tyr(Allyl-RCM) b
7.0
Phe
7.0
Gly(Allyl-RCM) b
Asp
9.0
7.9
aThe 3JHN values for the 1st residues could not be determined.
bThe side chains of the residues were cross-linked by RCM (see Figure 2).
10

Lowest energy conformer
Top-10 stable conformers
1
2
3
4
A
B
C
D
E
F
G
H
Figure 3. 3D structural models of 1 (A, B), 2 (C, D), 3 (E, F), and 4 (G, H) proposed by the
molecular mechanics calculations with restrained conformational search by the distance
geometry method. The lowest energy conformers (A, C, E, and G) and superposition of the
top-10 stable conformers (B, D, F, and H) of 1–4 are shown. The  and  angles of the lowest
energy conformers (Figures 3A, C, E, and G) are summarized in Table S1.
11

2.1.3. Effect of 1–4 on A42 aggregation
Inhibitory activities of 1–4 against A42 aggregation were evaluated by using thioflavin-T
(Th-T), which detects the A42 aggregates through fluorescence.36 For comparison, the
activities of linear peptides such as LVFFA and LPFFD were also analyzed. As shown in
Figure 4A, only 1 among all the peptides significantly reduced the relative Th-T fluorescence
caused by A42 aggregation. We confirmed that 1 exhibited dose-dependent inhibition of
Th-T fluorescence by A42 aggregation (Figure S1). These data indicated the potent
inhibition of A42 aggregation by 1. On the other hand, the Th-T fluorescence was only slightly
decreased by LVFFA and LPFFD. These data suggested that the cross-linking of the side chains
could enhance the inhibitory activity of BSBps against the A42 aggregation. Since the inhibition
of the Th-T fluorescence by 2–4 was weaker than that of 1, introduction of Pro and/or Asp residue
into the termini of BSFps would decrease the inhibitory activities.
To analyze the effect of BSFps on the fibril formation of A42, we observed the morphology
of the A42 aggregate co-incubated with each peptide for 48 h using transmission electron
microscopy (TEM) (Figure 5). Since all the aggregates were found to form fibril structures, 1–4
would not change the pathway of amyloid fibril formation.
2.1.4. Effect of 1–4 on A42-induced neurotoxicity
Inhibitory activities of 1–4 against A42-induced neurotoxicity were evaluated by MTT
(3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay using human
neuroblastoma SH-SY5Y cells. After 48-h incubation of the cells with A42 in the presence
or absence of each compound, the viabilities of the resultant cells were evaluated by MTT
assay. The cell viabilities with 1–4, despite the 10-fold addition, were almost same as that
without compound, indicating that not all the peptides prevented the A42-induced toxicity
significantly (Figure 4B and S2).
12

6000
5000
4000
3000
2000
1000
0
(A)
(B)
120
100
80
A42 only
A42 + 1
A42 + 2
60
**, ##
A42 + 3
40
A42 + 4
20
A42 + LVFFA
A42 + LPFFD
0
A42
−
−
+
+
+
+
+
+
+
0
10
20
30
40
50
Compound
−
+ 1 + 2 + 3 + 4 + LV + LP
Incubation time (h)
FFA FFD
Figure 4. (A) Effect of 1–4 on the A42 aggregation evaluated by Th-T assay (excitation:
430 nm, emission: 510 nm). A42 (10 M) was co-incubated with or without each compound
(40 M) in PBS (pH 7.4) at 37 °C. Data are presented as the mean ± SD (n = 4). ** p < 0.01
vs. A42 only, ## p < 0.01 vs. A42+LVFFA by Tukey-Kramer test. (B) Effect of 1–4 on
A42-induced toxicity in SH-SY5Y cells. After 48-h incubation of the cells with A42 (1
M) in the presence of each compound (10 M), the viabilities of the resultant cells were
evaluated by MTT assay. Data are presented as the mean ± SD (n = 3).
(A) A42 alone
(B) A42 + 1
(C) A42 + 2
(D) A42 + 3
(E) A42 + 4
(F) A42 + LVFFA
(G) A42 + LPFFD
Figure 5. TEM images of A42 aggregates prepared in the presence of BSFps 1–4. The
aggregates were prepared by incubation of A42 (10 M) with each compound (40 M) in
PBS (pH 7.4) at 37 ºC for 48 h. Scale bar = 100 nm.
13

2.2. BSFps with various cross-linking motifs (11–13)
2.2.1. Design and synthesis of 11–13
The study of 1–4 indicated that cross-linking of the side chains in BSFp pentapeptides
could mimic the -strand conformation and enhance their inhibitory activity against the A42
aggregation. To find the optimal way of cross-linking, we designed and synthesized 11–13
whose cross-linking patterns were different from that of 1–4 (Figure 6). In these
pentapeptides, the aromatic rings were located at the side chains of 3rd and 4th residues,
mimicking the two successive phenylalanines of the LVFFA sequence. It should be noted that
the positions of Tyr(Allyl-RCM) and Gly(Allyl-RCM) in 11 and 12 are opposite to those in
Abell’s motif (1–4 and 13). 11–13 were synthesized in the synthetic routes similar to those of
1–4 (Schemes S1–S3).
Phe
Tyr(Allyl-RCM)
Tyr(Allyl-RCM)
O
O
Gly(Allyl-RCM)
Gly(Allyl-RCM)
Leu
Ala
Leu
Ala
O
O
H
H
O
O
O
O
N
N
OH
H
H
N
H
H
H2N
N
H
N
N
OH
N
N
OH
O
H
H2N
N
H
N
H2N
N
N
O
O
O
H
H
H
O
O
O
O
O
Val
Val
Phe
Gly(Allyl-RCM)
Phe
O
Tyr(Allyl-RCM)
11
12
13
Figure 6. Chemical structures of 11–13
14

2.2.2. Conformational analysis of 11–13
3
NMR analyses of 11–13 in DMSO-d6 were performed and the obtained JHN values are
3
shown in Table 2. The JHN values at 2nd and 3rd residues of 11 and those at 3rd and 4th
residues of 12 were considerably larger than 8 Hz, suggesting the formation of -strand at
these positions.29, 30 3JHN value at 5th residue of 13 was 9.0 Hz, which is as large as at 4th
residues of 1–4 (Table 1).
Molecular modelling of 11–13 was conducted in the same manner as those of 1–4. The
3
-angles with their JHN values larger than 8.5 Hz were set at around 180°. The obtained
structural models are described in Figure 7, and  and  angles of the lowest energy
conformers (Figures 7A, C, and E) are summarized in Table S1. According to the empirical
rule,34, 35 the  and  angles in the macrocyclic frames suggested that the residues are
included in a -strand conformation. Macrocyclic frames of the top-10 stable conformers
were similar to each other, whereas two residues outside the macrocycles were quite different
among the conformers (Figures 7B, D, and F).
Table 2. 3JHN (Hz) values for 11–13 in DMSO-d6 at 25 °C.
Compound
Residue No.
Residue
Gly(Allyl-RCM)b
3JHN (Hz)
N.D.a
9.0
11
1
2
3
4
5
1
2
3
4
5
1
2
3
4
5
Val
Tyr(Allyl-RCM)b
9.7
Phe
8.0
Ala
7.2
N.D. a
12
13
Leu
Gly(Allyl-RCM) b
8.0
Phe
8.6
Tyr(Allyl-RCM)b
9.7
Ala
6.9
N.D. a
Leu
Val
8.7
Tyr(Allyl-RCM) b
6.8
Phe
6.5
Gly(Allyl-RCM) b
9.0
aThe 3JHN values for the 1st residues could not be determined.
bThe side chains of the residues were cross-linked by RCM (see Figure 5).
15

Lowest energy conformer
Top-10 stable conformers
11
A
B
12
C
D
13
E
F
Figure 7. 3D structural models of 11 (A, B), 12 (C, D), and 13 (E, F) proposed by the
molecular mechanics calculations with restrained conformational search by the distance
geometry method. The lowest energy conformers (A, C, and E) and superposition of top-10
stable conformers (B, D, and F) of 11–13 are shown. The  and  angles of the lowest energy
conformers are summarized in Table S1.
16

2.2.3. Effect of 11–13 on A42 aggregation
Inhibition of A42 aggregation by 11–13 was evaluated using Th-T assay. 12 significantly
decreased the relative Th-T fluorescence induced by the A42 aggregation (Figure 8A) and
the inhibitory activity was found to be dose-dependent (Figure S1), suggesting the potent
inhibition of A42 aggregation by 12. However, the inhibitory activity of 12 was slightly weaker
than that of 1. In addition, incubation of 12 alone in PBS caused a slight increase of Th-T
fluorescence (Figure S1), indicating a small amount of self-aggregation of 12. On the other hand,
11 and 13 did not reduce the Th-T fluorescence, suggesting that these peptides could not inhibit
the A42 aggregation. TEM analysis of the A42 aggregate co-incubated with each peptide
suggested the formation of fibril structures in all the aggregates (Figure 9).
2.2.4. Effect of 11–13 on A42-induced neurotoxicity
Effects of 11–13 on A42-induced neurotoxicity were evaluated by MTT assay using
SH-SY5Y cells. Since the cell viabilities in the presence of each compound were almost same
as that without compound, 11–13 would not prevent the A42-induced toxicity (Figure 8B).
We confirmed that 12 alone exhibited significant toxicity against the cells (Figure S2). This
cytotoxicity might be caused by the self-aggregation of 12 observed in the Th-T assay (Figure
S1).
17

6000
5000
4000
3000
2000
1000
0
(A)
(B)
120
100
80
A42 only
A42+ 11
A42 + 12
A42 + 13
A42 + 1
60
*
40
20
**
0
A42 + LVFFA
A42
−
−
+
+
+
+
+
+
Compound
−
+ 11 + 12 + 13 + 1 + LV
0
10
20
30
40
50
FFA
Incubation time (h)
Figure 8. (A) Effect of 11–13 on A42 aggregation evaluated by Th-T assay (excitation: 430
nm, emission: 510 nm). A42 (10 M) was co-incubated with or without each compound (40
M) in PBS (pH 7.4) at 37 °C. Data are presented as the mean ± SD (n = 4). ** p < 0.01 vs.
A42 only. (B) Effect of 11–13 on A42-induced toxicity in SH-SY5Y cells. After 48-h
incubation of the cells with A42 (1 M) in the presence of each compound (10 M), the
viabilities of the resultant cells were evaluated by MTT assay. Data are presented as the mean
± SD (n = 3).
(A) A42 + 11
(B) A42 + 12
(C) A42 + 13
Figure 9. TEM images of A42 aggregates prepared in the presence of BSFps 11–13. The
aggregates were prepared by incubation of A42 (10 M) with each compound (40 M) in
PBS (pH 7.4) at 37 ºC for 48 h. Scale bar = 100 nm.
18

3. Discussion
3.1. Verification of -strand-fixation by cross-linking of side chains
In
this
study,
we
synthesized
such
two
types
and
of
BSFps:
and
‘Tyr(Allyl-RCM)-Xaa-Gly(Allyl-RCM)’
as
1–4
13,
‘Gly(Allyl-RCM)-Xaa-Tyr(Allyl-RCM)’ such as 11 and 12. As shown in Tables 1 and 2,
3JHN values for Gly(Allyl-RCM) in ‘Tyr(Allyl-RCM)-Xaa-Gly(Allyl-RCM)’ type and those
for Tyr(Allyl-RCM) in ‘Gly(Allyl-RCM)-Xaa-Tyr(Allyl-RCM)’ type were much larger than
8.0 Hz. This indicated that the designed motifs could fix the residues in the -strand
conformation. It is noteworthy that the 3JHN values of Xaa in the
‘Gly(Allyl-RCM)-Xaa-Tyr(Allyl-RCM)’ type (11 and 12) were considerably large. The motif
might strongly fix the conformation of this residue in -strand.
Our conformational analysis proposed 3D structural models of the synthesized peptides
(Figures 3 and 7). 3D structures of the macrocyclic frames of 1–4 were quite similar to that in
the X-ray crystal structure of Abell’s tripeptide (Figure 1C).27 This supports the validity of
our methodology for conformational analysis. The  and  angles of the lowest energy
structures (Table S1) indicated that three residues in the macrocycle (2nd to 4th residues in 1–
4 and 12, 1st to 3rd residues in 11, and 3rd to 5th residues in 13) should form the -strand
conformation. Although the side chains and two successive residues were variable, the
macrocyclic frames of top-10 stable conformers were similar in each compound. These data
strongly supported that the cross-linking of the side chains in every second residue is an
effective way to fix the peptide conformation in a -strand.
3.2. Structure-activity relationships of BSFp
1 and 12 inhibited the aggregation of A42, whereas 11 and 13 did not (Figures 4A and
8A). As described above, the whole conformation of 1 and 12 were fixed to the -strand, but
19

two residues outside the macrocycle were variable in 11 and 13 (Figures 3 and 7). The
positions of cross-linking in the peptides should be critical for -strand-fixation, leading to a
high affinity to the -sheet in the A42 aggregates. Incorporation of Pro and/or Asp residues
in the terminus did not enhance but reduce the inhibitory activity against the A42
aggregation as shown in case of 2–4 (Figures 4A). Since LPFFD also did not prevent the
aggregation, introduction of Pro and/or Asp residue might not enhance the inhibitory activity
in Th-T assay. Murvai et al. indicated that LPFFD might interrupt the assembly of
protofilaments.37 This might be a reason why LPFFD would not alter the aggregation kinetics
of A42 in Th-T assay because binding of Th-T to the protofilament could also increase its
fluorescence.
Although 1 and 12 inhibited the A42 aggregation in Th-T assay (Figure 4A), these
peptides could not prevent the A-induced neurotoxicity in SH-SY5Y cells (Figure 4B). This
might be due to their little effect on the aggregation pathway of A42. Soluble A aggregates
formed in the pathway of amyloid fibrils are known to be toxic to neuronal cells.38-40 The
aggregation pathway could be changed into non-toxic by some aggregation inhibitors.41, 42
Cyclic BSBp developed by Arai et al.21,22 and Jha et al.23 also prevented the formation of
fibrillar aggregates, resulting in the reduction of A-induced neuronal death. In contrast, our
BSFp 1 and 12 could slow the aggregation kinetic (Figure 4A and 8A), but had little effect on
the morphologies of the A42 fibrils (Figure 5 and 9). This might be a reason why 1 and 12
could not alleviate the neurotoxicity of A42 in SH-SY5Y cells. The difference of the
experimental condition between Th-T assay and cell-based assay should be also considered.
The content in the cell culture such as protein and cell membrane might interact with BSFp,
leading to fewer interactions between BSFp and A42 molecules.
20

4. Conclusion
In this study, we designed and synthesized a new type of BSBp whose side chains were
cross-linked by RCM to fix the conformation in a -strand. According to the conformation
analysis, cross-linking of every second side chains could fix the conformation of peptides in
the -strands. Among the synthesized pentapeptides, 1 and 12, whose side chains at the 2nd
and 4th residues were cross-linked, inhibited the aggregation of A42 in Th-T assay,
suggesting that the -strand conformation fixation of BSBp could improve their inhibitory
activity against the A42 aggregation. However, 1 and 12 could not change the aggregation
state of A42 into a nontoxic one.
5. Materials and methods
5.1. Molecular modeling based on NMR Data
General techniques of NMR experiments were described in Supplementary Material.
1
1
3JHN values in Tables 1 and 2 were determined by H-1D along with H-1H J-resolved 2D
and DQF-COSY experiments.
Molecular modeling was performed on the CONFLEX 8 program31-33 by the distance
geometry method. For energy minimization, we utilized MMFF94s force field with a
dielectric constant value of 46.45 (DMSO at 25 °C). Conformational search was performed
using the reservoir-filling algorithm with dihedral angle constraints derived from the NMR
data (Tables 1 and 2); the -angles with their 3JHN values larger than 8.5 Hz were set at 180°
with force constant of 1 kcal•mol‒1•rad‒2. The search limit was set at 3 kcal/mol above the
global energy minimum. Atomic coordinates and calculated potential energies of the
lowest-energy structures were described in Supplementary material.
21

5.2 Preparation of A42
A42 peptide was kindly provided by Prof. Kazuhiro Irie at Graduate School of
Agriculture in Kyoto University. Synthetic procedure of the A42 peptide was described
elsewhere.43, 44
A42 peptide for all the assays was pre-treated with 1,1,1,3,3,3-hexafluoro-2-propanol
(HFIP) to dissociate aggregation seeds, which were unavoidably formed during purification
and/or storage. A42 peptide was dissolved in HFIP at 1 mg/mL. After being incubated at
room temperature for 30 min, the solution was sonicated for 5 min, and dried in vacuo using a
centrifugal evaporator. The resultant film (HFIP-treated A42) was used for all the assays.
5.3. Thioflavin-T (Th-T) fluorescence assay
To 534 L of phosphate buffered saline (PBS: 50 mM sodium phosphate with 100 mM
NaCl, pH 7.12) was added 6 L of each compound solution (4 mM in DMSO), then 60 L of
A42 solution (100 M in 0.1% NH4OH). The final solution included 10 M of A42, 40
M of each compound, and 1% DMSO in PBS (pH 7.4). After being incubated at 37 °C for 0,
4, 8, 24, 48 h, the suspension was vortexed for 5 second and 15 L of the aliquot was
collected in a 96-well black plate. The plate was stored at ‒80 °C until measurement.
To each well of the plate was added 235 µL of 5.0 M Th-T in 50 mM Gly-NaOH (pH
8.5), followed by the measurement of fluorescence using a 430 nm (FWHM 10 nm) a
excitation filter and a 510 nm (FWHM 40 nm) emission filter on a microplate reader (TriSter
LB941, Berthold Technologies). The background of the buffer without A42 was subtracted
from recorded values.
5.4. Transmission electron microscopy (TEM)
22

Morphology of the A42 aggregates co-incubated with each compound were observed
by a TEM (Figures 5 and 9). The incubation conditions were the same as that for preparing
the samples for the Th-T assay. A solution of A42 (10 M) and each compound (40 M) in
PBS (containing 1% DMSO, pH 7.4) was incubated at 37 °C for 48 h. After centrifugation,
the supernatant was removed from the pellets, and the aggregates were suspended in distilled
water by gentle vortex mixing. The suspensions were applied to a 200-mesh formvar-coated
copper grid (Nissin EM, Tokyo, Japan) and allowed to stand for 5 min before being
negatively stained with 2% uranylacetate. The fibrils were examined with the H-7650
electronmicroscope (Hitachi).
5.5. Cell culture
Human neuroblastoma SH-SY5Y cells were obtained from the American Type Culture
Collection (ATCC®, CRL-2266™). They were cultured in Dulbecco’s Modified Eagle
Medium/F-12 Nutrient Mixture (Gibco) containing 10% (v/v) fetal bovine serum (Biowest),
100 U/mL penicillin, and 100 g/mL streptomycin at 37 °C under 5% CO2 atmosphere.
5.6. MTT assay
Near-confluent cultures of cells were plated at 2.0 × 104 cells/200 L/well fresh culture
medium in 96-well plates, and pre-incubated at 37 °C under 5% CO2 for 24 h. To the cell
culture was added 1 L of compound solution (2.2 mM in DMSO) and 20 L of A42
solution (11 M in 0.1% NH4OH) and. The final culture included 10 M of each compound,
1 M of A42, 0.5% DMSO, and 0.01% NH4OH. After incubation at 37 °C under 5% CO2
for 48 h, 20 µL of MTT solution (5 mg/mL in water) was added to each well, followed by
incubation at 37 °C for 4 h. After discarding the medium, 100 µL of DMSO was added to
each well. The optical density of the cell lysate was read at 545 nm (after background
23

subtraction at 630 nm) using a microplate reader (CHROMATE4300, Awareness Technology
Inc.). The absorbance obtained by adding the vehicle was taken as 100%.
Acknowledgments
This research was supported by Sugiyama Houkoukai (grant in 2020). The authors are
grateful to Prof. Kazuhiro Irie, Assoc. Prof. Kazuma Murakami, and Dr. Yumi Irie at
Graduate School of Agriculture in Kyoto University for kindly providing A42 peptides. The
authors greatly thank Prof. Minoru Inagaki at Graduate School of Bioresources in Mie
University for his generous support in the laboratory.
Supplementary Material
Figures S1−S2, Table S1, Schemes S1−S4, synthetic procedures (1−4, 11−13, LVFFA,
LPFFD), copies of 1H and 13C NMR spectra for synthetic compounds, and atomic coordinates
and calculated potential energies of the lowest-energy structures. This material is available via
the Internet.
References and notes
1. Selkoe DJ, Hardy J. The amyloid hypothesis of Alzheimer's disease at 25 years.
EMBO Mol Med. Jun 2016;8(6):595-608. doi:10.15252/emmm.201606210
2. Glenner GG, Wong CW. Alzheimer's disease: initial report of the purification and
characterization of a novel cerebrovascular amyloid protein. Biochem Biophys Res Commun.
May 16 1984;120(3):885-90. doi:10.1016/s0006-291x(84)80190-4
3. Masters CL, Simms G, Weinman NA, Multhaup G, McDonald BL, Beyreuther K.
Amyloid plaque core protein in Alzheimer disease and Down syndrome. Proc Natl Acad Sci
U S A. Jun 1985;82(12):4245-9. doi:10.1073/pnas.82.12.4245
24

4. Crouch PJ, Harding SM, White AR, Camakaris J, Bush AI, Masters CL. Mechanisms
of A beta mediated neurodegeneration in Alzheimer's disease. Int J Biochem Cell Biol.
2008;40(2):181-98. doi:10.1016/j.biocel.2007.07.013
5. Jarrett JT, Berger EP, Lansbury PT, Jr. The carboxy terminus of the beta amyloid
protein is critical for the seeding of amyloid formation: implications for the pathogenesis of
Alzheimer's disease. Biochemistry. May 11 1993;32(18):4693-7. doi:10.1021/bi00069a001
6. Meier BH, Riek R, Bockmann A. Emerging Structural Understanding of Amyloid
Fibrils by Solid-State NMR. Trends Biochem Sci. Oct 2017;42(10):777-787.
doi:10.1016/j.tibs.2017.08.001
7. Loquet A, El Mammeri N, Stanek J, et al. 3D structure determination of amyloid
fibrils using solid-state NMR spectroscopy. Methods. Apr 1 2018;138-139:26-38.
doi:10.1016/j.ymeth.2018.03.014
8. Serpell LC. Alzheimer's amyloid fibrils: structure and assembly. Biochim Biophys
Acta. Jul 26 2000;1502(1):16-30. doi:10.1016/s0925-4439(00)00029-6
9. Antzutkin ON, Leapman RD, Balbach JJ, Tycko R. Supramolecular structural
constraints on Alzheimer's beta-amyloid fibrils from electron microscopy and solid-state
nuclear
magnetic
resonance.
Biochemistry.
Dec
24
2002;41(51):15436-50.
doi:10.1021/bi0204185
10. Tycko R. Insights into the amyloid folding problem from solid-state NMR.
Biochemistry. Mar 25 2003;42(11):3151-9. doi:10.1021/bi027378p
11. Gazit E. A possible role for pi-stacking in the self-assembly of amyloid fibrils. Faseb
j. Jan 2002;16(1):77-83. doi:10.1096/fj.01-0442hyp
12. Porat Y, Abramowitz A, Gazit E. Inhibition of amyloid fibril formation by
polyphenols: structural similarity and aromatic interactions as a common inhibition
25

mechanism.
Chem
Biol
Drug
Des.
Jan
2006;67(1):27-37.
doi:10.1111/j.1747-0285.2005.00318.x
13. Soto C, Kindy MS, Baumann M, Frangione B. Inhibition of Alzheimer's amyloidosis
by peptides that prevent beta-sheet conformation. Biochem Biophys Res Commun. Sep 24
1996;226(3):672-80. doi:10.1006/bbrc.1996.1413
14. Soto C, Sigurdsson EM, Morelli L, Kumar RA, Castano EM, Frangione B.
Beta-sheet breaker peptides inhibit fibrillogenesis in a rat brain model of amyloidosis:
implications for Alzheimer's therapy. Nat Med. Jul 1998;4(7):822-6. doi:10.1038/nm0798-822
15. Tjernberg LO, Naslund J, Lindqvist F, et al. Arrest of beta-amyloid fibril formation
by
a
pentapeptide ligand.
J
Biol Chem. Apr 12 1996;271(15):8545-8.
doi:10.1074/jbc.271.15.8545
16. Austen BM, Paleologou KE, Ali SA, Qureshi MM, Allsop D, El-Agnaf OM.
Designing peptide inhibitors for oligomerization and toxicity of Alzheimer's beta-amyloid
peptide. Biochemistry. Feb 19 2008;47(7):1984-92. doi:10.1021/bi701415b
17. Adessi C, Frossard MJ, Boissard C, et al. Pharmacological profiles of peptide drug
candidates for the treatment of Alzheimer's disease. J Biol Chem. Apr 18
2003;278(16):13905-11. doi:10.1074/jbc.M211976200
18. Cruz M, Tusell JM, Grillo-Bosch D, et al. Inhibition of beta-amyloid toxicity by
short peptides containing N-methyl amino acids. J Pept Res. Mar 2004;63(3):324-8.
doi:10.1111/j.1399-3011.2004.00156.x
19. Chalifour RJ, McLaughlin RW, Lavoie L, et al. Stereoselective interactions of
peptide inhibitors with the beta-amyloid peptide.
2003;278(37):34874-81. doi:10.1074/jbc.M212694200
J
Biol Chem. Sep 12
26

20. Francioso A, Punzi P, Boffi A, et al. beta-sheet interfering molecules acting against
beta-amyloid aggregation and fibrillogenesis. Bioorg Med Chem. Apr 15 2015;23(8):1671-83.
doi:10.1016/j.bmc.2015.02.041
21. Arai T, Araya T, Sasaki D, et al. Rational design and identification of a non-peptidic
aggregation inhibitor of amyloid-beta based on a pharmacophore motif obtained from
cyclo[-Lys-Leu-Val-Phe-Phe-]. Angew Chem Int Ed Engl. Jul 28 2014;53(31):8236-9.
doi:10.1002/anie.201405109
22. Arai T, Sasaki D, Araya T, Sato T, Sohma Y, Kanai M. A cyclic KLVFF-derived
peptide aggregation inhibitor induces the formation of less-toxic off-pathway amyloid-beta
oligomers. Chembiochem. Nov 24 2014;15(17):2577-83. doi:10.1002/cbic.201402430
23. Jha A, Kumar MG, Gopi HN, Paknikar KM. Inhibition of beta-Amyloid Aggregation
through a Designed beta-Hairpin Peptide. Langmuir. Jan 30 2018;34(4):1591-1600.
doi:10.1021/acs.langmuir.7b03617
24. Loughlin WA, Tyndall JD, Glenn MP, Fairlie DP. Beta-strand mimetics. Chem Rev.
Dec 2004;104(12):6085-117. doi:10.1021/cr040648k
25. Loughlin WA, Tyndall JD, Glenn MP, Hill TA, Fairlie DP. Update 1 of: Beta-strand
mimetics. Chem Rev. Jun 9 2010;110(6):Pr32-69. doi:10.1021/cr900395y
26. Abell AD, Jones MA, Coxon JM, et al. Molecular modeling, synthesis, and
biological evaluation of macrocyclic calpain inhibitors. Angew Chem Int Ed Engl.
2009;48(8):1455-8. doi:10.1002/anie.200805014
27. Abell AD, Alexander NA, Aitken SG, et al. Synthesis of Macrocyclic β-Strand
Templates by Ring Closing Metathesis. The Journal of Organic Chemistry. 2009/06/05
2009;74(11):4354-4356. doi:10.1021/jo802723w
27

28. Roosenburg S, Laverman P, Joosten L, et al. Stabilized (111)in-labeled sCCK8
analogues for targeting CCK2-receptor positive tumors: synthesis and evaluation. Bioconjug
Chem. Apr 21 2010;21(4):663-70. doi:10.1021/bc900465y
29. Pardi A, Billeter M, Wuthrich K. Calibration of the angular dependence of the amide
proton-C alpha proton coupling constants, 3JHN alpha, in a globular protein. Use of 3JHN
alpha for identification of helical secondary structure. J Mol Biol. Dec 15
1984;180(3):741-51. doi:10.1016/0022-2836(84)90035-4
30. Smith LJ, Bolin KA, Schwalbe H, MacArthur MW, Thornton JM, Dobson CM.
Analysis of main chain torsion angles in proteins: prediction of NMR coupling constants for
native and random coil conformations. J Mol Biol. Jan 26 1996;255(3):494-506.
doi:10.1006/jmbi.1996.0041
31. Goto H, Osawa E. Corner flapping: a simple and fast algorithm for exhaustive
generation of ring conformations. Journal of the American Chemical Society. 1989/11/01
1989;111(24):8950-8951. doi:10.1021/ja00206a046
32. Gotō H, Ōsawa E. An efficient algorithm for searching low-energy conformers of
cyclic and acyclic molecules. 10.1039/P29930000187. Journal of the Chemical Society,
Perkin Transactions 2. 1993;(2):187-198. doi:10.1039/P29930000187
33. CONFLEX 8. Goto, H. Obata, S. Nakayama, N. Ohta, K. CONFLEX Corporation,
Tokyo, Japan, 2017.
34. Ramachandran GN, Ramakrishnan C, Sasisekharan V. Stereochemistry of
polypeptide chain configurations. Journal of Molecular Biology. 1963/07/01/
1963;7(1):95-99. doi:https://doi.org/10.1016/S0022-2836(63)80023-6
35. Hovmoller S, Zhou T, Ohlson T. Conformations of amino acids in proteins. Acta
Crystallogr D Biol Crystallogr. May 2002;58(Pt 5):768-76. doi:10.1107/s0907444902003359
28

36. Naiki H, Gejyo F. [20] Kinetic analysis of amyloid fibril formation. Methods in
Enzymology. Academic Press; 1999:305-318.
37. Murvai Ü, Soós K, Penke B, Kellermayer MSZ. Effect of the beta-sheet-breaker
peptide LPFFD on oriented network of amyloid β25-35 fibrils. Journal of Molecular
Recognition. 2011/05/01 2011;24(3):453-460. doi:10.1002/jmr.1113
38. Nichols MR, Colvin BA, Hood EA, Paranjape GS, Osborn DC, Terrill-Usery SE.
Biophysical comparison of soluble amyloid-beta(1-42) protofibrils, oligomers, and
protofilaments. Biochemistry. Apr 7 2015;54(13):2193-204. doi:10.1021/bi500957g
39. Walsh DM, Hartley DM, Kusumoto Y, et al. Amyloid beta-protein fibrillogenesis.
Structure and biological activity of protofibrillar intermediates. J Biol Chem. Sep 3
1999;274(36):25945-52. doi:10.1074/jbc.274.36.25945
40. Shankar GM, Li S, Mehta TH, et al. Amyloid-beta protein dimers isolated directly
from Alzheimer's brains impair synaptic plasticity and memory. Nat Med. Aug
2008;14(8):837-42. doi:10.1038/nm1782
41. Ladiwala AR, Lin JC, Bale SS, et al. Resveratrol selectively remodels soluble
oligomers and fibrils of amyloid Abeta into off-pathway conformers. J Biol Chem. Jul 30
2010;285(31):24228-37. doi:10.1074/jbc.M110.133108
42. Ehrnhoefer DE, Bieschke J, Boeddrich A, et al. EGCG redirects amyloidogenic
polypeptides into unstructured, off-pathway oligomers. Nat Struct Mol Biol. Jun
2008;15(6):558-66. doi:10.1038/nsmb.1437
43. Irie K, Oie K, Nakahara A, et al. Molecular Basis for Protein Kinase C
Isozyme-Selective Binding:ꢀ The Synthesis, Folding, and Phorbol Ester Binding of the
Cysteine-Rich Domains of All Protein Kinase C Isozymes. Journal of the American Chemical
Society. 1998/09/01 1998;120(36):9159-9167. doi:10.1021/ja981087f
29