Development of a new distyrylbenzene-derivative amyloid-β-aggregation and fibril formation inhibitor

Article abstract of DOI:10.1248/cpb.c12-00365

Several new amyloid-β (Aβ) aggregation inhibitors were synthesized according to our theory that a hydrophilic moiety could be attached to the Aβ-recognition unit for the purpose of preventing amyloid plaque formation. A distyrylbenzene-derivative, DSB(EEX

Full text of DOI:10.1248/cpb.c12-00365

1164
Regular Article
Chem. Pharm. Bull. 60(9) 1164–1170 (2012)
Vol. 60, No. 9
Development of a New Distyrylbenzene-Derivative Amyloid-β-
aggregation and Fibril Formation Inhibitor
,a
Hideharu Suzuki,* Akihito Ishigami,b Ayako Orimoto,a Akihiro Matsuyama,a Setsuko Handa,b
Naoki Maruyama,b Yuusaku Yokoyama,a Hiroaki Okuno,a and Masamichi Nakakoshia
b
a Faculty of Pharmaceutical Sciences, Toho University; Chiba 274–8510, Japan: and Molecular Regulation of Aging,
Tokyo Metropolitan Institute of Gerontology; Tokyo 173–0015, Japan.
Received April 22, 2012; accepted June 5, 2012
Several new amyloid-β (Aβ) aggregation inhibitors were synthesized according to our theory that a
hydrophilic moiety could be attached to the Aβ-recognition unit for the purpose of preventing amyloid
plaque formation. A distyrylbenzene-derivative, DSB(EEX)3, which consider the Aβ recognition unit (DSB,
1,4-distyrylbenzene) and expected to bind to amyloid fibrils (β-sheet structure), was combined with the hy-
drophilic aggregation disrupting element (EEX) (E, Glu; X, 2-(2-(2-aminoethoxy)ethoxy)acetic acid). This
DSB(EEX)3 compound, compared to several others synthesized similarly, was found to be the most active for
reducing Aβ toxicity toward IMR-32 human neuroblastoma cells. Moreover, its inhibition of Aβ-aggregation
or fibril formation was directly confirmed by transmission electron microscopy and atomic force microscopy.
These results suggest that the Aβ aggregation inhibitor DSB(EEX)3 disrupts clumps of Aβ protein and is a
likely candidate for drug development to treat Alzheimer’s disease.
Key words Alzheimer’s disease; amyloid-β; amyloid fibril; distyrylbenzene derivative
The cause of Alzheimer’s disease (AD) has been attributed acid), constructed by attaching the hydrophilic aggregation-
to an accumulation of amyloid-β (Aβ) peptides in the brain.1–6) disrupting element (DDX)3 to the Aβ recognition element
Aβ generally consists of 40–42 amino acids and is the major (KLVFF).21) Moreover, KLVFF(EEX)3 (E, Glu) was found to
component of senile plaques in the brains of AD patients.7–10) be an even more efficient inhibitor of Aβ aggregation.22)
The aggregation of Aβ causes extensive neurotoxicity that
We report here the new design, synthesis, and activities
leads to memory loss, cognitive decline, language impair- of strong Aβ-aggregation inhibitors formulated by attach-
ment, and dementia. Although several mechanisms have been ing a hydrophilic moiety to an Aβ-recognition compound.
proposed to explain the precise nature of the neuronal toxicity Thioflavin T and Congo red pigments are known to bind
exerted by Aβ peptides, that lethal activity is not yet fully specifically with amyloid fibrils (β-sheet structure) and
understood.11) Aβ peptide fibrils cause neuronal cell apoptosis, have been used extensively for detecting the presence of
resulting in breakage of the neuronal cell network in the brain Aβ-aggregates and measuring their rates of formation.23–25)
and the development of AD. The Aβ peptides are generated by Recently, the Congo red analogue ((E,E)-1-fluoro-2,5-bis(3-
the sequential action of β- and γ-secretase on amyloid precur- hydroxycarbonyl-4-hydroxy)styrylbenzene (FSB)), which is
sor protein (APP)12) in neuronal cells in the brain. Secretase a fluorinated 1,4-distyrylbenzene (DSB) derivative, was used
inhibitors may thus hold promise as a treatment for AD. as a fluorescent stain for Aβ plaques, and the 19F-derivative
Recently, several reports1–3) have described potent γ-secretase was also used for the detection of plaques by magnetic reso-
inhibitors that are active in animal models of Aβ deposition, nance imaging (MRI).26) For the structural design of the new
although amyloid fibrils remain in the brain.
inhibitors, we focused on this structure of FSB and selected
Numerous drug treatments have been shown to benefit AD DSB and stilbene (SB) skeletons as new molecular recognition
patients, but none of them truly cures the disease.13) However, moieties for amyloid fibrils (β-sheet structure). The DSB and
chemical agents that prevent the pathological oligomerization SB derivatives include the carboxylic acid and phenol groups
and fibrillation of Aβ have the potential for more effective on the edge of structures (Fig. 1). In this study, several new
treatment of AD.14) Recently, many compounds have been aggregation inhibitors composed of an Aβ recognition element
found and developed to block Aβ aggregation or neurotoxicity (DSB and SB derivatives) and the hydrophilic aggregation dis-
in vitro, including peptide and nonpeptide inhibitors of Aβ ag- rupting element (EEX) are examined for their ability to reduce
gregation.15–18)
Aβ toxicity toward IMR-32 human neuroblastoma cells.27–29)
We have focused on the concept that the inhibition of “Aβ Precise observation of Aβ-aggregation inhibition was achieved
aggregation or fibril formation ” offers an efficient option for by using transmission electron microscopy (TEM) and atomic
the treatment of AD. We previously reported that the KLVFF force microscopy (AFM).
(Lys-Leu-Val-Phe-Phe) sequence (Aβ16–20) in Aβ plays an
important role in aggregate/fibril formation.19,20) Consequently,
Experimental
we developed inhibitors containing the KLVFF sequence as
All melting points were determined using a micro melting
an Aβ recognition element and several hydrophilic amino acid point hot stage apparatus (Yanagimoto) and are uncorrected.
analogs as aggregation-disrupting elements. In 2002, we re- IR spectra were recorded on a JASCO FT/IR-300 spectropho-
1
ported the synthesis and Aβ-aggregation inhibitory activity of tometer. H-NMR (400MHz) and 13C-NMR (100MHz) spectra
KLVFF(DDX)3 (D, Asp; X, 2-(2-(2-aminoethoxy)ethoxy)acetic were recorded on an AL-400 spectrometer with tetramethyl-
silane as an internal reference. Mass spectra (MS) were mea-
sured on JEOL Automass System II, JMS D-300, and DX-303
The authors declare no conflict of interest.
*To whom correspondence should be addressed. e-mail: suzuki@phar.toho-u.ac.jp
© 2012 The Pharmaceutical Society of Japan

September 2012
1165
stirred for 3h at 180°C in an argon atmosphere. After cooling,
volatile substances were removed under reduced pressure. The
residual oil was purified by silica gel column chromatography
(CHCl3–MeOH=95:5) to yield compound 5 (2.19g, 94%) as a
pale yellow oil.
1H-NMR (CDCl3, 400MHz) δ: 1.25 (6H, m), 3.20 (2H, d,
JP–H=22Hz), 3.91 (3H, s), 4.02 (4H, m), 7.37 (2H, dd, J=8.8,
2.8Hz), 7.99 (2H, d, J=7.2Hz). 13C-NMR (CDCl3, 100MHz)
δ: 166.8, 137.1 (d, JP–C=9.1Hz), 129.8 (d, JP–C=5.8Hz), 128.8
(d, JP–C=5.7Hz), 127.3 (d, JP–C=3.0Hz), 62.2 (d, JP–C=6.6Hz),
52.0, 33.9, (d, JP–C=136.7Hz), 16.3 (d, JP–C=5.8Hz). MS (high
resolution (HR)) m/z 286 [M]+. IR (neat) 1721cm−1.
(E)-Methyl 4-(3-Carbo-tert-butoxy-4-(methoxymethoxy)-
styryl)benzoate (6) To a stirred solution of methyl 4-((di-
ethoxyphosphoryl)methyl)benzoate 5 (102.3mg, 0.36mmol) in
Fig. 1. Structure of Stilbene (SB) and 1,4-Distyrylbenzene (DSB)
spectrometers with direct inlet systems. Fast atom bombard- N,N-dimethylformamide (DMF) (5mL) at −20°C in an argon
ment mass spectra (FAB-MS) were measured with a JEOL atmosphere was added 1M tert-BuOK/tetrahydrofuran (THF)
JMS-600. For column chromatography, Silica gel 60 (70–230 (0.80mL, 0.80mmol), then stirred for 10min. To that mixture,
mesh ASTM, Merck, Darmstadt, Germany) was used. For a solution of tert-butyl 5-formyl-2-(methoxymethoxy)benzoate
TLC, Silica gel 60F254 (Merck) was used. The abbreviations 3 (99.0mg, 0.37mmol) in DMF (1.0mL) was added at −20°C,
used are as follows: s, singlet; d, doublet; dd, double doublet; then stirred for 15min at −20°C. The reaction mixture was
t, triplet; q, quartet; m, multiplet; br, broad; BP, base peak.
poured into excess aqueous saturated NH4Cl and extracted
Synthesis of Molecular Recognition Units 7 and 13. with AcOEt. The organic layer was washed with brine, dried
tert-Butyl 5-Formyl-2-hydroxybenzoate (2) To a stirred over anhydrous Na2SO4, and evaporated to dryness in vacuo.
solution of 5-formyl-2-hydroxybenzoic acid 1 (1.67g, 10mmol) The residual oil was purified by silica gel column chromatog-
and tert-butyl-2,2,2-trichloroacetimidate (7.20mL, 40mmol) raphy (AcOEt–hexane=1:10) to give the product titled com-
in CH2Cl2 (40mL) was added BF3·Et2O (200µL, 0.32mmol) pound 6 (92.2mg, 65%) as colorless prisms.
at room temperature in an argon atmosphere, then the mixture
mp 86–88°C. 1H-NMR (CDCl3, 400MHz) δ: 1.62 (9H,
was stirred for 1h at room temperature. The reaction mixture s), 3.53 (3H, s), 3.92 (3H, s), 5.26 (2H, s), 7.03 (1H, d,
was poured into excess aqueous saturated NaHCO3 with ice J=16.0Hz), 7.16 (1H, d, J=16.0Hz), 7.18 (2H, d, J=8.0Hz),
cooling, then the mixture was extracted with chloroform. The 7.53 (1H, d, J=16.0Hz), 7.57 (1H, dd, J=9.0, 3.4Hz), 7.85
organic layer was washed with brine, dried over anhydrous (1H, d, J=2.4Hz), 8.02 (2H, d, J=8.0Hz). 13C-NMR (CDCl3,
Na2SO4, and evaporated to dryness in vacuo. The residual oil 100MHz) δ: 166.7, 165.3, 156.0, 141.6, 130.4, 130.2, 129.9,
was purified by silica gel column chromatography (AcOEt– 129.7, 129.3, 128.7, 126.8, 126.1, 123.8, 116.4, 94.8, 56.2, 51.9,
hexane=1:20) to give the compound titled 2 (1.48g, 66%) as 28.2. MS (FAB) m/z 399 [M+H]+, 421 [M+Na]+. IR (KBr)
a pale yellow oil.
1714cm−1. EA Calcd for C23H26O6: C, 69.33: H, 6.58. Found:
1H-NMR (CDCl3, 400MHz) δ: 1.62 (9H, s), 7.05 (1H, C, 69.02: H, 6.60.
d, J=8.4Hz), 7.94 (1H, dd, J=8.4, 2.4Hz), 8.27 (1H, d,
(E)-4-(3-(tert-Butoxycarbonyl)-4-(methoxymethoxy)-
J=2.4Hz), 9.86 (1H, s) ,11.69 (1H, s). 13C-NMR (CDCl3, styryl)benzoic Acid (7) To a stirred solution of (E)-methyl
100MHz) δ: 190.0, 169.1, 166.7, 135.3, 133.8, 128.3, 118.6, 4-(3-carbo-tert-butoxy-4-(methoxymethoxy)styryl)benzoate 6
114.0, 84.2, 28.1. MS (FAB) m/z 223 [M+H]+, 245 [M+Na]+. (103.7mg, 0.25mmol) in MeOH (10mL) at room temperature
IR (neat) 1694cm−1.
was added 1M KOH (2.5mL, 2.5mmol), then the mixture was
tert-Butyl 5-Formyl-2-(methoxymethoxy)benzoate (3) stirred for 22h at room temperature in an argon atmosphere.
To a stirred suspension of tert-butyl 5-formyl-2-hydroxyben- The reaction mixture was carefully neutralized with aqueous
zoate 2 (1.47g, 6.6mmol) and K2CO3 (1.84g, 13mmol) in ace- 1M HCl with ice cooling, then MeOH was removed under re-
tone (40mL) at room temperature in an argon atmosphere was duced pressure. The residual aqueous layer was extracted with
added dropwise chloromethylmethylether (750µL, 10mmol), AcOEt, washed with brine, dried over anhydrous Na2SO4, and
then the mixture was stirred for 1h at room temperature. The evaporated to dryness in vacuo to give the product titled com-
reaction mixture was poured into ice water and extracted with pound 7 (61.4mg, 63%) as colorless prisms.
AcOEt. The organic layer was washed with aqueous 10%
mp 180–183°C. 1H-NMR (CDCl3, 400MHz) δ: 1.62 (9H,
NaOH, then brine, dried over anhydrous Na2SO4, and evapo- s), 3.53 (3H, s), 5.26 (2H, s), 7.18 (2H, d, J=8.0Hz), 7.19 (1H,
rated to dryness in vacuo to give the title compound 3 as pale d, J=16.0Hz), 7.57 (1H, d, J=16.0Hz), 7.57 (1H, dd, J=8.8,
yellow oil (1.68g, 95%).
2.4Hz), 7.58 (1H, d, J=8.4Hz), 7.85 (1H, d, J=2.4Hz), 8.09
1H-NMR (CDCl3, 400MHz) δ: 1.57 (9H, s), 3.51 (3H, s), (2H, d, J=8.4Hz). 13C-NMR (CDCl3, 100MHz) δ: 171.4,
5.31 (2H, s), 7.28 (1H, d, J=8.8Hz), 7.92 (1H, dd, J=8.8, 165.4, 156.2, 142.6, 130.7, 130.6, 130.3, 129.5, 127.9, 126.8,
2.0Hz), 8.18 (1H, d, J=2.0Hz), 9.90 (1H, s). 13C-NMR (CDCl3, 126.3, 123.8, 116.5, 94.9, 81.5, 56.3, 28.3. MS (FAB) m/z 385
100MHz) δ: 190.3, 164.4, 160.8, 133.5, 133.4, 129.8, 123.7, [M+H]+, 407 [M+Na]+. EA Calcd for C22H24O6: C,68.74: H,
115.6, 94.5, 81.9, 56.5, 28.2.
Methyl 4-((Diethoxyphosphoryl)methyl)benzoate (5)
mixture of methyl 4-(bromomethyl)benzoate
6.29. Found C, 68.45: H, 6.38.
Methyl 2,5-Bis((diethoxyphosphoryl)methyl)benzoate
(1.87g, (11) A mixture of methyl 2,5-bis(bromomethyl)benzoate30)
A
4
8.18mmol) and triethylphosphite (1.40mL, 8.16mmol) was 10 (1.63g, 5.06mmol) and triethylphosphite (1.80mL,

1166
Vol. 60, No. 9
10mmol) was stirred for 4h at 180°C under an argon atmo- 3.46 (3H, s), 3.45 (3H, s), 1.55 (9H, s), 1.53 (9H, s). 13C-NMR
sphere. After cooling, volatile substances were removed under (CDCl3, 100MHz) δ: 165.4, 165.4, 156.0, 155.9, 138.5, 136.3,
reduced pressure. The residual oil was purified by silica gel 131.1, 130.6, 130.3, 130.1, 130.0, 130.0, 129.6, 129.2, 128.3,
column chromatography (CHCl3–MeOH=95:5) to give the 127.2, 126.5, 126.4, 123.8, 123.5, 116.6, 116.6, 95.0, 81.4, 81.3,
product titled compound 2 (2.04g, 93%) as a pale yellow oil.
60.4, 56.3, 56.2, 29.7, 28.3, 28.2. MS (FAB) m/z 647 [M+H]+,
1H-NMR (CDCl3, 400MHz) δ: 1.22 (12H, m), 3.13 (2H, 669 [M+Na]+. IR (KBr); 3423, 2977, 1604, 1572, 1368cm−1.
d, JP–H=20.8Hz), 3.75 (2H, d, JP–H=22.8Hz), 3.90 (3H, EA Calcd for C37H42O10 C, 68.71: H, 6.55. Found C, 68.62, H,
s), 4.00 (8H, m), 7.32 (1H, dd, J=8.0, 2.0Hz), 7.38 (1H, d, 6.59
J=8.0Hz), 7.83 (1H, s). 13C-NMR (CDCl3, 100MHz) δ: 167.7,
Synthesis of the New Aβ-Aggregation Inhibitor Peptides
133.2 (dd, JP–C=6.2, 3.3Hz), 132.4 (dd, JP–C=6.2, 2.9Hz), 9a–c and 15a–c Using a Peptide Synthesizer Peptides 9a–
132.1 (dd, JP–C=7.0, 3.3Hz), 132.0 (dd, JP–C=9.9, 3.7Hz), c, 15a–d were synthesized on TGS RAM resin (30–70µmol,
130.6 (dd, JP–C=9.1, 4.1Hz), 130.2 (dd, JP–C=6.2, 2.9Hz), Shimadzu Co., Kyoto, Japan) using an Fmoc solid-phase syn-
62.2 (d, JP–C=6.6Hz), 62.0 (d, J=P–C6.6Hz), 52.1, 33.2 (d, thesis method on an automated peptide synthesizer (Shimadzu
JP–C=136.6Hz), 30.8 (d, JP–C=134.6Hz), 16.3 (d, JP–C=5.8Hz), PSSM-8 Peptide Synthesizer Simultaneous Multiple). The mo-
16.2 (d, JP–C=6.2Hz). MS (electron ionization (EI)) m/z 436 lecular recognition unit 7 or 13, Fmoc-Glu(OtBu) (hydrophilic
[M]+. IR (neat) 1721cm−1.
domain), and Fmoc-AEEA (9-fluorenylmethoxycarbonyl-8-
Methyl 2,5-Bis((E)-3-(tert-butoxycarbonyl)-4-(methoxy- amino-3,6-dioxaoctanoic acid, linker domain) were coupled
methoxy)styryl)benzoate (12) To a stirred solution of meth- sequentially. Benzotriazole-1-yl-oxy-tris-pyrrolidino-phospho-
yl 2,5-bis((diethoxyphosphoryl)methyl)benzoate 11 (395.0mg, nium hexafluorophosphate (PyBOP), N-hydroxybenzotriazole
0.905mmol) in THF (5mL) at −78°C in an argon atmosphere (HOBt), and N-methylmorpholine (NMM) were used for the
was added 1M tert-BuOK/THF (1.20mL, 1.20mmol), then the peptide coupling reaction with 1.0, 1.0, and 1.5eq, respec-
mixture was stirred for 5min. To the mixture, a solution of tively, based on the amino acids. Fmoc amino acids and alkyl
tert-butyl 5-formyl-2-(methoxymethoxy)benzoate 5 (489.5mg, chain were used in 7 and 5 excess equivalents, respectively.
1.96mmol) in THF (2.0mL) was added at −78°C, then stirred The coupling reactions were carried out for 30min. De-protec-
for 30min at −78°C. The reaction mixture was poured into tion and cleavage of the resin were accomplished with a cleav-
excess aqueous saturated NH4Cl and extracted with AcOEt. age cocktail (trifluoroacetic acid (TFA)/H2O=10/90) for 15min
The organic layer was washed with brine, dried over anhy- at room temperature; the resin was removed by filtration, and
drous Na2SO4, and evaporated to dryness in vacuo. The re- the targeted peptides 9a (yield 48%), 9b (yield 56%), 9c (yield
sidual oil was purified by silica gel column chromatography 64%), 15a (yield 43%), 15b (yield 72%) and 15c (yield 50%)
(AcOEt–hexane=1:10) to give the product titled compound 12 were precipitated by adding a large amount of cold ether. The
(374.9mg, 63%) as pale yellow oil.
overall yields of the novel inhibitors were calculated based on
1H-NMR (CDCl3, 400MHz) δ: 1.55 (9H, s), 1.55 (9H, s), the molar ratio of the amino group on TGS-RAM resin. HPLC
3.46 (6H, s), 3.89 (3H, s), 5.19 (4H, s), 6.94 (1H, d, J=16.4Hz), analysis of the products showed essentially one peak, and
6.95 (1H, d, J=16.4Hz), 7.05 (1H, d, J=16.4Hz), 7.10 (1H, d, matrix-assisted laser desorption ionization time-of-flight mass
J=8.8Hz), 7.10 (1H, d, J=8.8Hz), 7.49 (1H, dd, J=8.4, 2.0Hz), spectrometry (MALDI-TOF-MS), ABI Voyager-DETM STR,
7.56 (1H, dd, J=8.8, 2.4Hz), 7.56 (1H, dd, J=8.8, 2.4Hz), with 3-hydroxypicolinic acid used as the matrix), showed an
7.64 (1H, d, J=8.4Hz), 7.76 (1H, d, J=2.4Hz), 7.78 (1H, d, exact mass of m/z 723 [M+Na]+ for 9a, 1141 [M+Na]+ for
J=2.4Hz), 7.80 (1H, d, J=16.4Hz), 7.96 (1H, d, J=2.0Hz). 9b, 1516 [M+Na]+ for 9c, 862[M]+ for 15a, 1292 [M+K]+ for
13C-NMR (CDCl3, 100MHz) δ: 167.8, 165.4, 165.4, 155.9, 15b, and 1677 [M+Na]+ for 15c.
155.9, 137.8, 136.2, 131.1, 130.6, 130.3, 129.9, 129.8, 129.5,
Cell Culture IMR-32 human neuroblastoma cells27–29)
129.2, 128.9, 128.7, 128.2, 127.1, 126.6, 126.3, 123.8, 123.6, were obtained from the Health Science Research Resources
116.6, 116.6, 95.0, 81.4, 81.3, 60.4, 56.3, 56.3, 52.2, 28.3. MS Bank (Osaka, Japan) and cultured in minimum essential me-
(FAB) m/z 661 [M+H]+, 683 [M+Na]+.
dium (MEM) supplemented with 10% fetal calf serum. Cells
2,5-Bis((E)-3-(t-butoxycarbonyl)-4-(methoxymethoxy)- were grown in 75cm2 plastic dishes at 37°C in 5% CO2 and
styryl)benzoic Acid (13) To a stirred solution of methyl air, and subcultured weekly.
2,5-bis((E)-3-(t-butoxycarbonyl)-4-(methoxymethoxy)styryl)-
Neurotoxicity Assay IMR-32 cells (1.5×105 cells/mL)
benzoate 12 (284mg, 0.4mmol) in MeOH (50mL) at room were inoculated in a 96-well flat-bottom microplate (Asahi
temperature was added aqueous 1M KOH (4.3mL, 4.3mmol), Techno Glass, Chiba, Japan) and cultured for two days at
then the mixture was stirred at room temperature for 6d in an 37°C in 5% CO2 and air. Aβ1–40 (30µM) and various con-
argon atmosphere. The reaction mixture was carefully neutral- centrations of Aβ aggregation inhibitor were added to each
ized with 1M HCl with ice cooling, then MeOH was removed well and incubated for two days. Cell viability was measured
under reduced pressure. The residual aqueous layer was ex- using a Cell Proliferation Kit II (XTT) (Roche Diagnostics,
tracted with AcOEt, washed with brine, dried over anhydrous Mannheim, Germany) according to the manufacturer’s in-
Na2SO4, and evaporated to dryness in vacuo to give the prod- structions. Briefly, XTT labeling mixture (final XTT concen-
uct titled compound 13 (254mg, 92%) as yellow prisms.
tration: 0.3mg/mL) was added to each well (50µL/well) and
1
mp 58–60°C. H-NMR (CDCl3, 400MHz) δ: 8.12 (1H, d, incubated for 4 to 6h. Absorbance at 405nm was measured
J=1.6Hz), 7.92 (1H, d, J=16.4Hz), 7.78 (1H, d, J=1.2Hz), 7.78 using an enzyme-linked immunosorbent assay (ELISA) Plate
(1H, d, J=1.2Hz), 7.61 (1H, d, J=8.8Hz), 7.56 (1H, dd, J=8.8, Reader (Emax, Molecular Devices, Sunnyvale, CA, U.S.A.).
2.4Hz), 7.49 (1H, dd, J=8.8, 2.4Hz), 7.11 (1H, d, J=8.8Hz),
TEM Samples of Aβ1–40 (100µg/mL) and Aβ1–40
7.11 (1H, d, J=8.8Hz), 7.07 (1H, d, J=16.4Hz), 6.96 (1H, d, (100µg/mL) plus DSB(EEX)3 15c (moleq) in aq. AcOH (pH
J=16.4Hz), 6.96 (1H, d, J=16.4Hz), 5.19 (2H, s), 5.18 (2H, s), 3.3) were incubated at 25°C for 6h. These samples were

September 2012
1167
Chart 1. Synthesis of 1,4-Distyrylbenzene Derivatives
observed using a JEM-2100F TEM (JEOL Co., Ltd., Tokyo,
Japan) at an accelerating voltage of 80kV. Negative staining
was achieved using 0.3% phosphotungstic acid solution on
polyvinyl formal membrane and carbon-coated copper grids
(400 mesh).
AFM The cantilever used was a NanoWorld Pointprobe®
NCHR probe. Aβ aggregation performed in the absence and in
the presence of the disruptor at a molar ratio of 1:1 was im-
aged using a modular AFM.
Sample 1) Ten microliter samples of Aβ1–40 (100µg/mL)
and Aβ1–40 (100µg/mL) plus DSB(EEX)3 15c (moleq) in aq.
AcOH (pH 3.3) were incubated at 25°C for 6h.
Sample 2) After incubation of 10µL Aβ1–40 (100µg/mL)
for 6h, DSB(EEX)3 15c (moleq) in aq. AcOH (pH 3.3) was
added, followed by incubation for 6h. These solutions were
deposited on freshly cleaved mica. The maximal scanning
area was 5.0µm×5.0µm, or 100nm×100nm for samples with
heights less than 500nm. The AFM used was a JSPM-5200
(JEOL Co., Ltd.).
Results and Discussion
Fig. 2. Cytotoxic Activity of Aβ1–40 towards IMR-32 Human Neuro-
Chemistry To transform the DSB or SB moieties into a
new Aβ-aggregation inhibitor, we synthesized carboxylic acid
derivatives of DSB to generate compound 13, and of SB to
generate compound 7. To use an Fmoc peptide synthesizer for
derivatization of the final peptide analogues, the carboxylic
acid and phenolic hydrogens required protection with acid-
blastoma Cells
Cells (1.5×105 cells/mL) were inoculated in a 96-well flat-bottom microplate
and cultured for two days at 37°C in 5% CO2 and air. Various concentrations
of Aβ1–40 were added to each well and incubated for two days. Cell viability
was measured using a Cell Proliferation Kit II (XTT). Values are expressed as
means S.D. (n=5), Williams test *p<0.025.
sensitive and piperidine-resistant protecting groups. We select- diethyl benzylphosphonate 5, 11. On the other hand, com-
ed tert-butyl ester and methoxymethyl ether groups to protect mercially available 5-formylsalicylic acid 1 was protected as a
both styryl groups. To allow the selective deprotection of the t-butyl ester and methoxymethyl ether. The protected aldehyde
two carboxylic acids, the carboxylic acid on the B-ring was 3 was reacted with mono-phosphonate 5 or di-phosphonate 11
synthesized as a methyl ester. For construction of the trans under Horner–Emmons conditions to form SB 6 and DSB 12,
C=C double bond, the Horner–Emmons reaction was used respectively. The final yield of DSB 12 formed by coupling 11
for coupling the two or three aromatic components. Commer- and 3 was very sensitive to the reaction conditions, as Chart
cially available methyl 4-(bromomethyl)benzoate 4, or known 1 portrays.
methyl 2,5-bis(bromomethyl)benzoate 10 prepared from
The methyl esters of 6 and 12 were hydrolyzed under alka-
carboxylic acid, were reacted with triethyl phosphite to form line conditions to form the desired carboxylic acids 7 and 13

1168
Vol. 60, No. 9
Fig. 3. Survival Rates of IMR-32 Neuroblastoma Cells Combined with 9a, 9b, and 9c in the Presence of Aβ1–40
Cells (1.5×105 cells/mL) were cultured for two days and then Aβ1–40 (30µM) and various concentrations of 9a, 9b and 9c were added to each well and incubated for 2d.
Cell viability was measured using a Cell Proliferation Kit II (XTT). Values are expressed as means S.D. (n=5), Williams test *p<0.025.
Fig. 4. Survival Rates of IMR-32 Neuroblastoma Cells Combined with 15a, 15b and 15c in the Presence of Aβ1–40
Cells (1.5×105 cells/mL) were cultured for two days and then Aβ1–40 (30µM) and various concentrations of 15a, 15b, and 15c were added to each well and incubated for
two days. Cell viability was measured using a Cell Proliferation Kit II (XTT). Values are expressed as means S.D. (n=5), Williams test *p<0.025.
(Chart 1). Compounds 7 and 13 could be used as fragments on
the Fmoc peptide-synthesizer, allowing synthesis of EEX 9a
and 15a, (EEX)2 9b and 15b, (EEX)3 9c and 15c, derivatives
of SB and DSB (Fig. 1). Upon final deprotection and cleav-
age of the peptides from resin, the methoxymethyl group
was cleanly deprotected by 90% aqueous trifluoroacetic acid
(TFA).
Aβ1–40 HCl salt was purchased from the Peptide Insti-
tute, Inc. (Osaka, Japan).
Toxicity Assay Using IMR-32 Human Neuroblastoma
Cells We evaluated the ability of these compounds to pro-
tect against neuronal cell death from exposure to Aβ1–40 by
using IMR-32 cells, which are known to produce APP, secrete
Aβ28) and have neurotoxicity by Aβ.27–29) As shown in Fig.
Fig. 5. Images of Aβ1–40 Fibrils by TEM
2, amounts of Aβ1–40 above 15µM produced significant cy-
totoxic activity; however, at 30µM the cell survival rate was
(a) Fibrils of Aβ1–40. (b) Aβ1–40 and 15c. Staining by phosphotangustate solu-
tion. Scale bars represent 500nm.
only 35.4 4.5%. Thus, 30µM Aβ1–40 was clearly toxic to
neuronal cells and became the concentration used for further with SB derivatives 9a–c (addition of 1.5–30µM) were de-
experiments.
termined using neurotoxicity assays. As shown in Fig. 3, all
The survival rates of IMR-32 cells following incubation (EEX) moieties of SB derivatives affected the survival rate

September 2012
1169
Fig. 6. Images of Aβ1–40 Fibrils and Fragments by AFM
(a) Fibrils of Aβ1–40 incubated in aq. AcOH (pH 3.3) at 25°C for 6h. (b) Fibril fragments of Aβ1–40 incubated with 15c in aq. AcOH (pH 3.3) at 25°C for 6h. Scale
bars represent 500nm.
of IMR-32 cells, with 9c being the most potent promoter of to a solution of Aβ fibrils and incubating the mixture for 6h,
survival, and 9a and 9b exhibiting lower levels of protection. fibrils were no longer visible in TEM images (Fig. 5b).
That is, 1.5 and 3µM of compounds 9a and 9b suppressed the
Finally, we added 15c to an Aβ fibril solution and observed
cytotoxic activity of Aβ1–40 only slightly, although at the the sample by AFM after 6h. The addition of 15c resulted in
higher concentration of 30µM, 9a and 9b evoked a level of cy- fragmentation of the fibrils (Fig. 6b), but the identical solution
totoxic activity for Aβ1–40 that reached the level of statistical lacking 15c had no evidence of fragmentation (Fig. 6a). Thus,
significance. However, 15 and 30µM of compound 9c signifi- 15c must have fragmented the Aβ fibrils, thus providing a bio-
cantly suppressed the cytotoxic activity of Aβ1–40 yielding logical activity that protected the IMR-32 cells from undergo-
the best survival rates.
ing Aβ-induced cytotoxicity.
The survival rates of IMR-32 cells following incubation
with DSB derivatives 15a–c determined by neurotoxicity as-
says produced the results shown in Fig. 4. Less than 3µM 15a
Conclusion
In conclusion, compound 15c effectively reduced the toxic-
and 15b suppressed the cytotoxicity of Aβ1–40 slightly; con- ity of Aβ towards IMR-32 cells, although 15c itself was not
versely, cytotoxicity in the presence of these compounds was cytotoxic. Instead, 15c apparently inhibited Aβ aggregation
significantly enhanced at 30 and 60µM. However, treatment or fibril formation. When 15c was present, no nuclei or fibrils
with 1.5–60µM 15c showed significant differences compared of Aβ were visible by TEM, and that outcome was verified by
with the control (0µmol) group. Since 15c was effective at a AFM, which showed that, upon the addition of 15c, fragmen-
lower dosage such as 1.5 and 3µM than 9c in the neurotoxicity tation of Aβ fibrils indeed occurred. These results strongly
assay, the DSB skeleton was considered more effective than suggest that the Aβ-aggregation inhibitor, compound 15c, dis-
the SB skeleton as a basic structure for an Aβ-induced cyto- rupts clumps of Aβ protein and is a likely candidate for drug
toxic inhibitor.
development to treat AD.
The EEX, (EEX)2 and (EEX)3 structures (9a–c, 15a–c)
showed different biological activities. Although SB and DSB
Acknowledgements We are grateful to the “Open Re-
moieties themselves were poorly soluble in water, as is pre- search Center” project for a private university matching sub-
dictable from their chemical structure, attaching an EEX sidy and a Grant-in-Aid for Scientific Research from the Min-
component, a known provider of hydrophilic amino acid (E, istry of Education, Culture, Sports, Science and Technology
Glu), should increase that property. Such an increase of water of Japan. We thank Ms. P. Minick for the excellent English
solubility from the addition of multiple EEX structures, i.e., editorial assistance.
(EEX)2 and (EEX)3, to SB and DSB should similarly enhance
the related biological activities of these fibril formation inhibi-
tors. Compound 15c at a low concentration of 3µM had supe-
rior activity when compared to the same concentration of 9c.
Thus, 15c was the most promising compound identified from
the IMR-32 survival rate assay.
Imaging of Fibril Formation and Fragmentation of Aβ
Fibrils by TEM and AFM Next, we used TEM and AFM
to observe the inhibition of fibril formation and the fragmenta-
tion of Aβ fibrils of 15c. These Aβ1–40 fibrils were 20–30nm
wide and composed of β-sheet structures of Aβ protein. Al-
though aggregation is known to trigger fibril formation, the
aggregation and fibril formation process of Aβ was not visible
by TEM. Therefore, we used the final product to view these
fibrils, which TEM clearly revealed in Aβ solutions to which
no inhibitor was added (Fig. 5a). In contrast, after adding 15c
References
1) Haass C., De Strooper B., Science, 286, 916–919 (1999).
2) Geula C., Wu C. K., Saroff D., Lorenzo A., Yuan M., Yankner B.
A., Nat. Med., 4, 827–831 (1998).
3) Austen B. M., Paleologou K. E., Ali S. A., Qureshi M. M., Allsop
D., El-Agnaf O. M., Biochemistry, 47, 1984–1992 (2008).
4) Liu D., Xu Y., Feng Y., Liu H., Shen X., Chen K., Ma J., Jiang H.,
Biochemistry, 45, 10963–10972 (2006).
5) Hasegawa T., Ukai W., Jo D. G., Xu X., Mattson M. P., Nakagawa
M., Araki W., Saito T., Yamada T., J. Neurosci. Res., 80, 869–876
(2005).
6) Billings L. M., Oddo S., Green K. N., McGaugh J. L., LaFerla F.
M., Neuron, 45, 675–688 (2005).
7) Iwatsubo T., Odaka A., Suzuki N., Mizusawa H., Nukina N., Ihara
Y., Neuron, 13, 45–53 (1994).
8) Suzuki N., Cheung T. T., Cai X. D., Odaka A., Otvos L. Jr., Eckman

1170
Vol. 60, No. 9
C., Golde T. E., Younkin S. G., Science, 264, 1336–1340 (1994).
1–6 (2003)
9) Dahlgren K. N., Manelli A. M., Stine W. B. Jr., Baker L. K., Krafft 21) Watanabe K., Nakamura K., Akikusa S., Okada T., Kodaka M.,
G. A., LaDu M. J., J. Biol. Chem., 277, 32046–32053 (2002).
10) Selkoe D. J., Nature, 399. A23–A31 (1999).
Konakahara T., Okuno H., Biochem. Biophys. Res. Commun., 290,
121–124 (2002).
11) Klein W. L., Krafft G. A., Finch C. E., Trends Neurosci., 24, 219– 22) Okuno H., Mori K., Okada T., Yokoyama Y., Suzuki H., Chem.
224 (2001).
Biol. Drug Des., 69, 356–361 (2007).
12) Selkoe D. J., Physiol. Rev., 81, 741–766 (2001).
23) Klunk W. E., Pettegrew J. W., Abraham D. J., The Journal of Histo-
chemistry and Cytochemistry, 37, 1293–1297 (1989)
24) Conway K. A., Lee S. J., Rochet J. C., Ding T. T., Williamson R.
E., Lansbury P. T. Jr., Proc. Natl. Acad. Sci. U.S.A., 97, 571–576
(2000).
13) Bolognesi M. L., Banzi R., Bartolini M., Cavalli A., Tarozzi A.,
Andrisano V., Minarini A., Rosini M., Tumiatti V., Bergamini C.,
Fato R., Lenaz G., Hrelia P., Cattaneo A., Recanatini M., Melchi-
orre C., J. Med. Chem., 50, 4882–4897 (2007).
14) Lansbury P. T. Jr., Curr. Opin. Chem. Biol., 1, 260–267 (1997).
15) Talaga P., Mini Rev. Med. Chem., 1, 175–186 (2001).
16) Estrada L. D., Soto C., Curr. Top. Med. Chem., 7, 115–126 (2007).
17) Bose P. P., Chatterjee U., Hubatsch I., Artursson P., Govender T.,
Kruger H. G., Bergh M., Johansson J., Arvidsson P. I., Bioorg. Med.
Chem., 18, 5896–5902 (2010).
18) Soto-Ortega D. D., Murphy B. P., Gonzalez-Velasquez F. J., Wil-
son K. A., Xie F., Wang Q., Moss M. A., Bioorg. Med. Chem., 19,
2596–2602 (2011).
25) Findeis M. A., Musso G. M., Arico-Muendel C. C., Benjamin H. W.,
Hundal A. M., Lee J. J., Chin J., Kelley M., Wakefield J., Hayward
N. J., Molineaux S. M., Biochemistry, 38, 6791–6800 (1999).
26) Higuchi M., Iwata N., Matsuba Y., Sato K., Sasamoto K., Saido T.
C., Nat. Neurosci., 8, 527–533 (2005).
27) Presper K. A., Basu M., Basu S., Proc. Natl. Acad. Sci. U.S.A., 75,
289–293 (1978).
28) Asami-Odaka A., Ishibashi Y., Kikuchi T., Kitada C., Suzuki N.,
Biochemistry, 34, 10272–10278 (1995).
19) Watanabe K., Segawa T., Nakamura K., Kodaka M., Konakahara T.,
Okuno H., The Journal of Peptide Research, 58, 342–346 (2001)
29) Kandimalla R. J., Wani W. Y., Binukumar B. K., Gill K. D., J.
Biomed. Sci., 19, 2 (2012).
20) Akikusa S., Watanabe K. I., Horikawa E., Nakamura K., Kodaka 30) Baell J. B., Gable R. W., Harvey A. J., Toovey N., Herzog T., Hän-
M., Okuno H., Konakahara T., The Journal of Peptide Research, 61, sel W., Wulff H., J. Med. Chem., 47, 2326–2336 (2004).