Isomerization of (Z,Z) to (E,E)1-bromo-2,5-bis-(3- hydroxycarbonyl-4-hydroxy)-styrylbenzene in strong base: Probes for amyloid plaques in the brain

Article abstract of DOI:10.1021/jm010161t

In developing probes for detecting β-amyloid (Aβ) plaques in the brain of Alzheimer's disease (AD), we have synthesized 1-bromo-2,5-bis-(3-hydroxycarbonyl-4-hydroxy)styrylbenzene (5, BSB). Due to the presence of two double bonds, formation of four different isomers is possible. Four isomers, E,E-5, E,Z-5, Z,E-5, and Z,Z-5, were prepared. Surprisingly, all showed strong fluorescent labeling of Aβ plaques in the brain of postmortem brain sections of patients with confirmed AD. In vitro binding assay also showed that all four isomers of BSB (E,E-5, E,Z-5, Z,E-5, and Z,Z-5) displayed a similar high binding affinity inhibiting the binding of [¹²⁵I] E,E-6, 1-iodo-2,5-bis-(3-hydroxycarbonyl-4-methoxy)styrylbenzene (IMSB) to Aβ_1-40 aggregates. The inhibition constants (K_i) of EE-5, EZ-5, ZE-5, and Z,Z-5 were 0.11 ± 0.01, 0.19 ± 0.03, 0.27 ± 0.06, and 0.13 ± 0.02 nM, respectively. Due to the fact that geometric stability of these styrylbenzenes is unknown, and the conversion of Z,Z-5 to E,E-5 may occur automatically in the binding or labeling assaying conditions, we have investigated the kinetics of conversion of Z,Z-5 to E,E-5 by NMR in D₂O/NaOD at elevated temperatures (70, 95, and 115 °C). The activation energy was determined to be 14.15 kcal/mol. The results strongly suggest that the isomeric conversion at room temperature in aqueous buffer solution is unlikely. All of the styrylbenzene isomers clearly showed potential as useful tools for studying Aβ aggregates in the brain. The data suggest that, despite the rigidity of this series of styrylbenzenes, the binding sites on Aβ aggregates may have certain flexibility and the binding pockets could be adaptable for binding to other smaller ligands. Such information could be exploited to develop new ligands for detecting amyloid plaques in AD.

Full text of DOI:10.1021/jm010161t

2270
J . Med. Chem. 2001, 44, 2270-2275
Expedited Articles
Isom er iza tion of (Z,Z) to (E,E)1-Br om o-2,5-bis-(3-h yd r oxyca r bon yl-4-h yd r oxy)-
styr ylben zen e in Str on g Ba se: P r obes for Am yloid P la qu es in th e Br a in
Chi-Wan Lee,† Zhi-Ping Zhuang,† Mei-Ping Kung,† Karl Plo¨ssl,† Daniel Skovronsky,‡ Tamar Gur,‡
Catherine Hou,† J ohn Q. Trojanowski,‡ Virginia M.-Y. Lee,‡,§ and Hank F. Kung*,†,§
Departments of Radiology, Pathology and Laboratory Medicine, and Pharmacology, University of Pennsylvania,
3700 Market Street, Room 305, Philadelphia, Pennsylvania 19104
Received April 9, 2001
In developing probes for detecting â-amyloid (Aâ) plaques in the brain of Alzheimer’s disease
(AD), we have synthesized 1-bromo-2,5-bis-(3-hydroxycarbonyl-4-hydroxy)styrylbenzene (5,
BSB). Due to the presence of two double bonds, formation of four different isomers is possible.
Four isomers, E,E-5, E,Z-5, Z,E-5, and Z,Z-5, were prepared. Surprisingly, all showed strong
fluorescent labeling of Aâ plaques in the brain of postmortem brain sections of patients with
confirmed AD. In vitro binding assay also showed that all four isomers of BSB (E,E-5, E,Z-5,
Z,E-5, and Z,Z-5) displayed a similar high binding affinity inhibiting the binding of [125I]E,E-
6, 1-iodo-2,5-bis-(3-hydroxycarbonyl-4-methoxy)styrylbenzene (IMSB) to Aâ1-40 aggregates. The
inhibition constants (Ki) of E,E-5, E,Z-5, Z,E-5, and Z,Z-5 were 0.11 ( 0.01, 0.19 ( 0.03, 0.27
( 0.06, and 0.13 ( 0.02 nM, respectively. Due to the fact that geometric stability of these
styrylbenzenes is unknown, and the conversion of Z,Z-5 to E,E-5 may occur automatically in
the binding or labeling assaying conditions, we have investigated the kinetics of conversion of
Z,Z-5 to E,E-5 by NMR in D2O/NaOD at elevated temperatures (70, 95, and 115 °C). The
activation energy was determined to be 14.15 kcal/mol. The results strongly suggest that the
isomeric conversion at room temperature in aqueous buffer solution is unlikely. All of the
styrylbenzene isomers clearly showed potential as useful tools for studying Aâ aggregates in
the brain. The data suggest that, despite the rigidity of this series of styrylbenzenes, the binding
sites on Aâ aggregates may have certain flexibility and the binding pockets could be adaptable
for binding to other smaller ligands. Such information could be exploited to develop new ligands
for detecting amyloid plaques in AD.
In tr od u ction
treatment, and we have synthesized and evaluated
potentially useful compounds based on small neutral
molecules.
Alzheimer’s disease (AD) is a neurodegenerative
disease of the brain characterized by dementia, cognitive
impairment, and memory loss. â-Amyloid (Aâ) peptides
are involved in the development and progression of AD.
The fibrillar aggregates of amyloid peptides, Aâ1-40 and
Aâ1-42, found in senile plaques and cerebrovascular
amyloid deposits in AD patients1 are major metabolic
products of amyloid precursor protein. Prevention and
reversal of Aâ plaque formation are being targeted as
treatments for this disease.2-8 Ligands exhibiting spe-
cific high binding affinity to Aâ aggregates may be
useful for diagnosis and treatment of AD.9 Since AD is
now becoming one of the most devastating illnesses
affecting a large number of older patients and their
families, there are urgent needs for better diagnosis and
Previously, ligands for staining amyloid aggregates
were based on highly conjugated dyes, such as Congo
Red (CR), Thioflavin T and S, Chrysamine G (CG), and
X-3410,11 (Chart 1). Thioflavins and CR have been used
in fluorescent staining of plaques and tangles in post-
mortem AD brain sections.12 A more abbreviated form
of CG, X-34 (Chart 1), has been reported as another
fluorescent dye for staining amyloid aggregates.13,14
Previously, a highly conjugated bromo derivative, E,E-5
(BSB), had been evaluated as a fluorescent probe for
AD.15 Several lines of evidence suggest that BSB is an
appropriate starting point for future efforts to generate
an imaging agent for monitoring AD. It labels Aâ
plaques in human AD brain sections with high sensitiv-
ity and specificity, and it crosses the blood-brain barrier
and labels numerous AD-like Aâ plaques throughout the
brain of transgenic mice following iv injection. Recently,
the synthesis of E,E-5 (BSB) and the corresponding
iodinated BSB were reported.21 This series of com-
pounds with strong fluorescent signals may have great
* Corresponding author address: Hank F. Kung, Ph.D., Depart-
ments of Radiology and Pharmacology, University of Pennsylvania,
3700 Market Street, Room 305, Philadelphia, PA 19104. Tel: (215)-
662-3096. Fax: (215)349-5035. E-mail: kunghf@sunmac.spect.upenn.edu.
† Department of Radiology.
‡ Department of Pathology and Laboratory Medicine.
§ Department of Pharmacology.
10.1021/jm010161t CCC: $20.00 © 2001 American Chemical Society
Published on Web 06/13/2001

Probes for Amyloid Plaques in the Brain
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 14 2271
Ch a r t 1. Chemical Structures of Congo Red (CR), Thioflavin T, Chrysamine G (CG) and X-34.
Sch em e 1. Synthesis of Styrylbenzenes
Ta ble 1. Excitation and Emission Wavelengths of Isomers of
BSB
potential as tools for staining Aâ aggregates by in vitro
labeling methods. On the basis of these exciting find-
ings, efforts were made in the synthesis of all four
possible isomers of E,E-5 (BSB) produced in the Wittig
reaction to investigate the influence of geometry on the
binding affinity toward Aâ plaques. We report herein
the synthesis and characterization of four isomers of 5:
E,E, Z,Z, Z,E, and E,Z. Isomeric conversion from Z,Z
to E,E and their binding affinities toward preformed Aâ
aggregates as well as fluorescent staining of Aâ plaques
in AD brain sections by these isomers were also evalu-
ated.
excitation [nm]
maximum
emission [nm]
maximum
compound
E,E-5 (BSB)
E,Z-5 (BSB)
Z,E-5 (BSB)
Z,Z-5 (BSB)
432a
418a
416a
394a
532b
502c
498c
498d
a
b
Emission set to 510 nm. Excitation set to 430 nm. c Excita-
d
tion set to 420 nm. Excitation set to 410 nm.
E,E-6 1-iodo-2,5-bis-(3-hydroxycarbonyl-4-methoxy)-
styrylbenzene (IMSB)21 binding to Aâ1-40 aggregates
(peptide concentrations 10-20 nM). Inhibition constants
(Ki) of E,E-5 (BSB), E,Z-5, Z,E-5, and Z,Z-5 were 0.11
( 0.01, 0.19 ( 0.03, 0.27 ( 0.06, and 0.13 ( 0.02 nM,
respectively, while CG and salicylic acid displayed Ki
values of 0.14 ( 0.04 and >1800 nM, respectively.
In brain sections of confirmed AD cases, Thioflavin
S, as expected, showed strong fluorescent staining of the
Aâ aggregates. Using the same concentration (0.05 mM),
all four isomers of 5 displayed excellent fluorescent
staining comparable or better than that observed with
Thioflavin S (Figure 1). The images of these isomers
substantiate the excellent binding affinity observed by
the in vitro binding assay using Aâ1-40 aggregates.
One remaining question on the observed high binding
affinity for all isomers is the possibility that isomeriza-
tion of E,Z-5, Z,E-5, and Z,Z-5 to E,E-5 during the
assaying condition may occur automatically. As a con-
sequence, the high binding affinity and excellent stain-
ing seen in AD brain section may be from only one active
isomer, E,E-5 (BSB). By close examination of the
chemical structure, one may hypothesize that the sali-
cylic moiety on both ends of the molecule could promote
the conversion of the cis form to the trans form,
especially, when the phenol groups are ionized (Scheme
Resu lts a n d Discu ssion
For this particular series of styrylbenzenes, the
chemical synthesis of the isomers has never been
reported before. The key step in the synthesis of the
styrylbenzene derivatives was the Wittig reaction
(Scheme 1), by which aldehyde 2 and triphenylphos-
phonium salt 1 were condensed. During this condensa-
tion reaction, the E,E-3 isomer conveniently precipitated
from the reaction mixture in 11% yield. The other three
isomers, Z,Z-3, Z,E-3, and E,Z-3 were isolated by flash
chromatography from the remaining mother liquor in
24, 13, and 11% yield, respectively (Scheme 1).
All four isomers of 3 were converted to the corre-
sponding acid, BSB, 5 in two steps: demethylation with
BBr3 in CH2Cl2 followed by hydrolysis of ester groups
in EtOH in the presence of KOH. All four of these
compounds are strongly fluorescent, a property making
them useful for staining. The excitation and emission
wavelengths were between 394-431 nm and 489-532
nm, respectively (see Table 1).
Surprisingly, all four isomers of BSB (E,E-5, E,Z-5,
Z,E-5, and Z,Z-5) displayed strong inhibition on [125I]-

2272 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 14
Lee et al.
F igu r e 1. Fluorescent images of E,E-5, E,Z-5, Z,Z-5, and Thioflavin S using adjacent sections of postmortem AD brain. All of
the images were obtained under similar incubating conditions using a concentration of 0.05 mM. The vascular amyloid plaques
at the upper left-hand corner of each picture were used as markers to align the images.
Sch em e 2. Isomerization of Z,Z-5 to E,E-5
2). In general, it is assumed that the trans isomer, E,E-
5, is more stable than the cis isomer, Z,Z-5. Isomeriza-
tion will only happen from Z,Z-5 to E,E-5. It was
reported previously that a similar cis-trans isomeriza-
tion of diethylstilbestrol (DES) in basic condition was a
first-order reaction. The isomerization from Z,Z-5 to
E,E-5 most likely will go through a mechanism similar
to that of DES.16,17
To investigate this possibility we first dissolved Z,Z-5
and E,E-5 in D2O in the presence of a small drop of
NaOD to enhance the solubility. At room temperature
the NMR spectra of Z,Z-5 and E,E-5 in D2O showed
very little change for 6 h (data not shown). However,
after overnight at room temperature the sample for
Z,Z-5 did show some indication that indeed the conver-
sion to the E,E-5 isomer did happen.
To further confirm the isomerization reaction, a more
detailed kinetic study was performed. An NMR tube,
containing 11 mg of Z,Z-5, was dissolved in 0.1 mL of
1.4 M NaOD in D2O and followed by the addition of D2O
to bring the final volume to 0.6 mL (0.23 M). The NMR
tube was then placed in a heating block filled with sand
to control the temperature at either 70, 95, or 115 °C.
The changes can be readily quantified by measuring the
same NMR sample and observing the shifting of peaks
in the range of 5.9∼6.4 ppm (Figure 2). The aromatic
proton signals of the 1H NMR for Z,Z-5 exhibited
definite changes with time, although in nearly all cases
the NMR signals are difficult to interpret because of
overlapping and broadening. Fortunately, signals in the
olefinic-proton region (5.9∼6.4 ppm, box in Figure 2) are
unique for Z,Z-5 without any interference from other
signals. The reduction of peaks in this region was
proportional to the changes of Z,Z-5 to E,E-5 at various
time points. Eventually, Z,Z-5 was completely converted
to E,E-5, and the NMR spectrum of Z,Z-5 appeared
identical to that of authentic E,E-5 in the same solvent.
A representative sample of data is presented in Figure
2. The kinetics of this isomerization of BSB was evalu-
ated by calculating the fraction of isomerization (F),
which was calculated based on the following equation:
Ft ) (I0 - It)/(I0 - I∞)
I: integral ratio of olefinic peaks
(6.06∼5.91 ppm) to whole aromatic peaks
(7.35∼5.91 ppm)
It: integral ratio at time t
I0: integral ratio at time 0
I∞: integral ratio at time ∞
At time ∞, the integral of the regions representing
olefinic peaks of Z,Z-5 was very close to zero.
Therefore:
Ft ) (I0 - It)/(I0 )
A logarithmic plot of (1 - F) at different temperatures

Probes for Amyloid Plaques in the Brain
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 14 2273
F igu r e 3. (a) Effects of temperature (70, 95, and 115 °C) on
isomerization of Z,Z-5 to E,E-5. (b) Plot of reaction rates vs
1/T. By measuring the slope for the plot of ln k versus 1/T and
fitting the Arrhenius equation: -ln k ) Ea/(RT) + ln A (where
R is the gas constant and T is the temperature in kelvin). The
activation energy (Ea) for the isomerization of Z,Z-5 to E,E-5
was calculated to be 14.15 kcal/mol.
F igu r e 2. 1H NMR spectra of a selected region of Z,Z-5 (0.02
mmol) in D2O/NaOD (0.23 M). The solution in the NMR test
tube was heated at 95 °C, and the spectra were obtained at
various time points after heating. The NMR spectrum of a
solution of authentic E,E-5 (BSB) (0.02 mmol) in D2O/NaOD
(0.23 M) is shown on the top.
these two fixed negative charges. Since all four possible
isomers of BSB (E,E-5, E,Z-5, Z,E-5, and Z,Z-5) showed
a similar binding affinity toward the aggregates, and
the relative distance between the two negative charges
is unlikely to be the same for all four of the isomers,
conceptual alternatives for the high binding affinity of
these compounds to the â-sheet structure may be
necessary. The nature of the interactions between these
small ligands with the macromolecular Aâ aggregates
remains to be explored. Such information will be
extremely useful in designing new ligands as probes for
Aâ aggregates.
In conclusion, synthesis of novel styrylbenzenes as
ligands for binding Aâ1-40 aggregates is reported. Four
possible isomers of 1-bromo-2,5-bis-(3-hydroxycarbonyl-
4-hydroxy)styrylbenzene (BSB), 5, were prepared and
elucidated. The kinetics of conversion of Z,Z-5 to E,E-5
was evaluated by NMR under strong basic conditions
(D2O/NaOD) at elevated temperatures. The activation
energy was determined to be 14.15 kcal/mol. It is likely
that no conversion of Z,Z-5 to E,E-5 would happen
during binding assays at room temperature. On the
basis of their exquisite binding affinity and selectivity
to Aâ aggregates, they are extremely useful tools for
studying the aggregates in vitro by binding assay and
fluorescent staining of the brain sections of AD patients.
vs the time t gave linear curves, suggesting a first-order
reaction under basic conditions:
Z,Z-5 f E,E-5
and
ln(1 - F)t ) -kt + ln(1 - F)0 (see Figure 3a)
The rate constants k at various temperatures can be
determined from measuring the slopes of the curves.
The activation energy Ea can then be calculated by
fitting the Arrhenius equation where R is the gas
constant and T is the temperature in kelvin.
Ea 1
R T
ln(k) ) -
+ ln(A) (see Figure 3b)
( )
A plot of ln(k) vs 1/T showed a straight line (R2
)
0.9912), and Ea was determined to be 14.15 kcal/mol.
On the basis of this result it is reasonable to assume
that at room temperature the kinetics of isomerization
from Z,Z-5 to E,E-5 will be very slow. Therefore, it is
likely that the in vitro binding data and staining of AD
brain sections in Figure 2 are due to the isomers as
indicated. It was reported previously that highly con-
jugated fluorescent probes, such as CG and its related
derivatives, such as X-34, exhibited binding predomi-
nantly to the â-sheet structure. Due to the relatively
rigid structure of these probes, the distance between the
two carboxylic groups is about 18 Å.10,11 It was suggested
that the binding affinity was largely determined by
Exp er im en ta l Section
All reagents used in the synthesis were commercial products
and were used without further purification unless otherwise
indicated. Anhydrous Na2SO4 was used as a drying agent.
Flash column chromatography was performed on 230-400
mesh silica gel. 1H NMR spectra were obtained on Bruker

2274 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 14
Lee et al.
spectrometers (Bruker DPX 200 and AMX 500). Chemical
shifts are reported as δ values with respect to residual protons
in CDCl3 unless otherwise mentioned. Coupling constants are
reported in hertz. Kinetic experiments were maintained at
constant temperature by a heating block (Thermolyne, Type
17600) equipped with multiple wells (the stability of the
heating block is (0.7 °C).
ArH), 7.04 (d, J ) 16.3 Hz, 1H, olefinic), 7.33 (d, J ) 16.5 Hz,
1H, olefinic), 7.40 (d, J ) 8.7 Hz, 1H, ArH), 7.60-7.73 (m, 4H,
ArH), 7.97 (s, 2H, ArH), 10.81 (s, 1H, OH), 10.83 (s, 1H, OH);
MS m/z 509 (M+ + H).
Z,Z-4: 1H NMR (200 MHz, CDCl3) δ 3.82 (s, 3H, -COOCH3),
3.83 (s, 3H, -COOCH3), 6.36 (d, J ) 12.1 Hz, 1H, olefinic),
6.45 (d, J ) 12.1 Hz, 1H, olefinic), 6.46 (d, J ) 12.1 Hz, 1H,
olefinic), 6.51 (d, J ) 12.1 Hz, 1H, olefinic), 6.68 (d, J ) 8.7
Hz, 1H, ArH), 6.74 (d, J ) 8.7 Hz, 1H, ArH), 6.90 (d, J ) 8.1
Hz, 1H, ArH), 6.97 (d, J ) 8.1 Hz, 1H, ArH), 7.14 (d, d, J )
8.7, 2.3 Hz, 1H, ArH), 7.25 (d, d, J ) 8.7, 2.3 Hz, 1H, ArH),
7.46 (s, 1H, ArH), 7.60 (d, J ) 2.2 Hz, 1H, ArH), 7.69 (d, J )
2.2 Hz, 1H, ArH), 10.72 (s, 1H, OH), 10.74 (s, 1H, OH).
Z,E-4: 1H NMR (200 MHz, CDCl3) δ 3.98 (s, 3H, -COOCH3),
3.99(s, 3H, -COOCH3), 6.44 (d, J ) 12.5 Hz, 1H, olefinic), 6.54-
(d, J ) 12.5 Hz, 1H, olefinic), 6.82-7.03 (m, 4H, ArH), 7.16-
7.50 (m, 3H, ArH), 7.60-7.77 (m, 3H, ArH), 7.95 (d, J ) 2.2
Hz, 1H, ArH), 10.77 (s, 1H, OH), 10.83 (s, 1H, OH).
Syn th esis of 1-Br om o-2,5-bis-(3-m eth oxylca r bon yl-4-
m eth oxy)styr ylben zen e (3). To a mixture of 5-formyl-2-
methoxy-benzoic acid methyl ester 2 (1.16 g, 6.0 mmol,
prepared from 5-formylsalicylic acid18) and phosphonium salt
1 (2.6 g, 3.0 mmol, prepared from 2-bromo-1,4-p-xylene19) in
MeOH (12 mL) was added a solution of sodium methoxide (1.3
mL, 25 wt % in MeOH) dropwise at 0 °C in an ice bath. The
mixture was stirred at room temperature overnight. The solid
formed was filtered and washed with methanol to give E,E-3
(175 mg 10.9%): 1H NMR (500 MHz, CDCl3) δ 3.907 (s, 3H,
CH3O-), 3.909 (s, 3H, CH3O-), 3.915 (s, 3H, CH3O-), 3.919 (s,
3H, CH3O-), 6.91 (d, J ) 16.3 Hz, 1H, olefinic), 6.97(d, J ) 8.5
Hz, 1H, ArH), 6.98 (d, J ) 8.5 Hz, 1H, ArH), 6.99 (d, J ) 16.4
Hz, 1H, olefinic), 7.03 (d, J ) 16.3 Hz, 1H, olefinic), 7.34 (d, J
) 16.2 Hz, 1H, olefinic), 7.40 (d, J ) 8.2 Hz, 1H, ArH), 7.59
(d,d, J ) 8.5, 2.1 Hz, 1H, ArH), 7.60 (d, J ) 8.3 Hz, 1H, ArH),
7.65 (d,d, J ) 8.7, 2.2 Hz, 1H, ArH), 7.69 (s,1H, ArH), 7.94 (s,
1H, ArH), 7.95 (s, 1H, ArH); MS m/z 559 (MNa+). HRMS m/z
Calcd for C28H26BrO6(MH+): 537.0913. Found: 537.0963.
The filtrate was neutralized with HCl (10%). MeOH was
removed, and water was added to the residue. The mixture
was extracted with CH2Cl2. The organic phase was concen-
trated to give crude product. Flash chromatography (Hex:
EtOAc, 8:1) gave (Z,Z)-1-bromo-2,5-bis-(3-methoxylcarbonyl-
4-methoxy)styrylbenzene Z,Z-3, Z,E-3, and E,Z-3.
Z,Z-3 (385 mg, 24.0%): 1H NMR (500 MHz, CDCl3) δ 3.80
(s, 3H, CH3O-), 3.81 (s, 3H, CH3O-), 3.83 (s, 3H, CH3O-), 3.85
(s, 3H, CH3O-), 6.42 (d, J ) 12.2 Hz, 1H, olefinic), 6.51 (d, J )
12.2 Hz, 1H, olefinic), 6.52 (d, J ) 12.1 Hz, 1H, olefinic), 6.57
(d, J ) 12.0 Hz, 1H, olefinic), 6.75 (d, J ) 8.7 Hz, 1H, ArH),
6.80 (d, J ) 8.7 Hz, 1H, ArH), 6.94 (d,d, J ) 8.1, 1.2 Hz, 1H,
ArH), 7.01 (d, J ) 8.0 Hz, 1H, ArH), 7.21 (d,d, J ) 8.7, 2.3 Hz,
1H, ArH), 7.31 (d,d, J ) 8.6, 2.2 Hz, 1H, ArH), 7.50 (d, J ) 1.3
Hz, 1H, ArH), 7.60 (d, J ) 2.3 Hz, 1H, ArH), 7.68 (d, J ) 2.2
Hz, 1H, ArH); MS m/z 554 (M+ + NH4).
E, Z-4: 1H NMR (200 MHz, CDCl3): δ 3.98 (s, 3H,
-COOCH3), 3.99(s, 3H, -COOCH3), 6.57 (d, J ) 12.5 Hz, 1H,
olefinic), 6.59 (d, J ) 12.5 Hz, 1H, olefinic), 6.80-7.03 (m, 5H,
ArH), 7.16-7.62 (m, 2H, ArH), 7.60-7.74 (m, 3H, ArH), 7.96
(d, J ) 2.2 Hz, 1H, ArH), 10.73 (s, 1H, OH), 10.80 (s, 1H, OH).
Gen er a l P r oced u r e for 1-Br om o-2,5-bis-(3-h yd r oxy-
ca r bon yl-4-h yd r oxy)styr ylben zen e (5). To a solution of
E,E-4 (534 mg, 1.0 mmol) in EtOH (60 mL) was added KOH
(1.2 g) in solid form. The mixture was stirred under reflux for
5 h. The cold mixture was acidified with HCl (10% solution)
and extracted with mixed solvent (CH2Cl2:MeOH, 9:1). The
organic phase was dried over Na2SO4 and filtered. The filtrate
was concentrated to give 480 mg of E,E-5 (95%): 1H NMR (200
MHz, MeOD) δ 6.96 (d,d, J ) 3.4, 8.7 Hz, 2H, ArH), 6.99 (d, J
) 16.3 Hz, 1H, olefinic), 7.12 (d, J ) 16.3 Hz, 1H, olefinic),
7.17 (d, J ) 16.3 Hz, 1H, olefinic), 7.34 (d, J ) 16.3 Hz, 1H,
olefinic), 7.56 (d, J ) 8.4 Hz, 1H, ArH), 7.73-7.80 (, m, 4H
ArH), 8.02 (t, J ) 2.6 Hz, 2H, ArH); MS m/z 480 (M+). Anal.
Calcd for C24H17BrO6‚0.5 H2O: C, 58.79; H, 3.70. Found: C,
58.89, H, 3.34.
1
Z,Z-5: H NMR (200 MHz, MeOD) δ 6.48 (d, J ) 12.1 Hz,
1H, olefinic), 6.49 (d, J ) 12.1 Hz, 1H, olefinic), 6.58 (d, J )
12.1 Hz, 1H, olefinic), 6.61 (d, J ) 12.1 Hz, 1H, olefinic), 6.73
(d, J ) 8.6 Hz, 1H, ArH), 6.77 (d, J ) 8.6 Hz, 1H, ArH), 7.01
(d, J ) 8.1 Hz, 1H, ArH), 7.06 (d, J ) 8.1 Hz, 1H, ArH), 7.23
(d, d, J ) 8.6, 2.3 Hz, 1H, ArH), 7.33 (d, d, J ) 8.6, 2.3 Hz,
1H, ArH), 7.51 (s, 1H, ArH), 7.68 (d, J ) 2.3 Hz, 1H, ArH),
Z,E-3 (204 mg, 12.7%): 1H NMR (200 MHz, CDCl3) δ 3.85
(s, 3H, CH3O-), 3.89 (s, 3H, CH3O-), 3.92 (s, 3H, CH3O-), 3.93
(s, 3H, CH3O-), 6.45 (d, J ) 12.2 Hz, 1H, olefinic), 6.56 (d, J )
12.1 Hz, 1H, olefinic), 6.83 (d, J ) 8.7 Hz, 1H, ArH), 6.96 (d,
J ) 16.5 Hz, 1H, olefinic), 6.98 (d, J ) 8.1 Hz, 1H, ArH), 7.15
(d, J ) 8.1 Hz, 1H, ArH), 7.33 (d, J ) 16.1 Hz, 1H, olefinic),
7.35 (d,d, J ) 8.6, 2.3 Hz, 1H, ArH), 7.48 (d, J ) 8.1 Hz, 1H,
ArH), 7.49 (d, J ) 1.1 Hz, 1H, ArH), 7.64 (d,d, J ) 8.9, 2.3 Hz,
1H, ArH), 7.72 (d, J ) 2.2 Hz, 1H, ArH), 7.95 (d, J ) 2.3 Hz,
1H, ArH); MS m/z 554 (M+ + NH4).
7.78 (d, J ) 2.3 Hz, 1H, ArH). HRMS m/z Calcd for C24H17
-
BrO6 (MH+): 480.0208. Found: 480.0201.
Z,E-5: 1H NMR (200 MHz, MeOD) δ 6.48 (d, J ) 12.1 Hz,
1H, olefinic), 6.54 (d, J ) 12.1 Hz, 1H, olefinic), 6.93-7.38 (m,
4H, ArH), 7.47-7.79 (m, 5H, ArH), 8.01 (s, 2H, ArH).
E,Z-5: 1H NMR (200 MHz, MeOD): δ 6.52 (d, J ) 12.1 Hz,
1H, olefinic), 6.63 (d, J ) 12.1 Hz, 1H, olefinic), 6.91-7.03 (m,
3H, ArH), 7.08-7.37 (m, 3H, ArH), 7.58-7.18 (m, 3H, ArH),
8.0 (s, 2H, ArH).
E,Z-3 (176 mg, 11.0%): 1H NMR (200 MHz, CDCl3) δ 3.83
(s, 3H, CH3O-), 3.85 (s, 3H, CH3O-), 3.92 (s, 3H, CH3O-), 3.93
(s, 3H, CH3O-), 6.56 (d, J ) 11.9 Hz, 1H, olefinic), 6.62(d, J )
12.2 Hz, 1H, olefinic), 6.77 (d, J ) 8.7 Hz, 1H, ArH), 6.89 (d,
J ) 16.2 Hz, 1H, olefinic), 6.98 (d, J ) 8.4 Hz, 1H, ArH), 7.04
(d, J ) 16.0 Hz, 1H, olefinic),7.15 (d, J ) 8.1 Hz, 1H, ArH),
7.22 (d,d, J ) 8.5, 1.5 Hz, 1H, ArH), 7.24 (d,d, J ) 8.5, 2.4 Hz,
1H, ArH), 7.58 (d,d, J ) 8.7, 2.3 Hz, 1H, ArH), 7.66 (d, J ) 2.3
Hz, 1H, ArH), 7.73 (s, 1H, ArH), 7.95 (d, J ) 2.3 Hz, 1H, ArH);
MS m/z 554 (M+ + NH4).
Gen er a l P r oced u r e for 1-Br om o-2,5-bis-(3-m eth oxy-
ca r bon yl-4-h yd r oxy)styr ylben zen e (4). To a solution of
E,E-3 (730 mg, 1.36 mmol) in CH2Cl2 (250 mL) was added BBr3
(10.5 mL, 1 M solution in hexane) dropwise at -78 °C in a
dry ice-acetone bath. The mixture was allowed to warm to
room temperature. Water was added while the reaction
mixture was cooled at 0 °C in an ice bath. The mixture was
extracted with CH2Cl2. The organic phase was dried over Na2-
SO4 and filtered. The filtrate was concentrated to afford 636
mg of E,E-4 (92%): 1H NMR (200 MHz, CDCl3) δ 3.99 (s, 6H,
CH3O-, -COOCH3), 6.99 (d, J ) 16.3 Hz, 1H, olefinic), 6.98
(d, J ) 16.5 Hz, 1H, olefinic), 7.01 (d, d, J ) 8.3, 2.6 Hz, 2H,
Bin d in g Assa ys Usin g Aggr ega ted P ep tid e F ibr ils in
Solu tion . A solid form of peptide Aâ1-40 was purchased from
Bachem (King of Prussia, PA). Aggregation of peptides was
carried out by gently dissolving the peptide (0.5 mg/mL) in a
buffer solution (pH 7.4) containing 10 mM sodium phosphate
and 1 mM EDTA. The solutions were incubated at 37 °C for
36-42 h with gentle and constant shaking. Binding studies
were carried out in 12 × 75 mm borosilicate glass tubes
according to the procedure described10 with some modifica-
tions. For inhibition studies, 1 mL of the reaction mixture
contained 40 µL of inhibitors (10-5-10-10 M in 10% EtOH) and
0.05 nM radiotracer in 40% EtOH. Nonspecific binding was
defined in the presence of 800 nM of CG. The mixture was
incubated at room temperature for 3 h, and the bound and
the free radioactivity were separated by vacuum filtration
through Whatman GF/B filters using a Brandel M-24R cell
harvester followed by 2 × 3 mL washes of 10% ethanol at room
temperature. Filters containing the bound I-125 ligand were
counted in a γ counter (Packard 5000) with 70% counting
efficiency. The results of inhibition experiments were subjected

Probes for Amyloid Plaques in the Brain
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 14 2275
20
(4) Skovronsky, D. M.; Lee, V. M. Beta-secretase revealed: starting
gate for race to novel therapies for Alzheimer’s disease. Trends
Pharmacol. Sci. 2000, 21, 161-163.
to nonlinear regression analysis using software EBDA by
which Ki values were calculated.
F lu or escen t Sta in in g. Brains of Alzheimer’s disease (AD)
patients (n ) 3) were obtained at autopsy, and neuropatho-
logical diagnosis was confirmed by current criteria. We per-
formed a variety of studies using 6 µm thick serial sections of
paraffin-embedded blocks of hippocampus and cortex from
brains fixed in 70% ethanol (EtOH)/150 mM NaCl. For BSB
staining, sections were immersed in 0.05 mM of E,E-5 (BSB),
E,Z-5, or Z,Z-5, in 50% EtOH for 30 min. Sections were quickly
differentiated in saturated Li2CO3 and rinsed in 50% EtOH
before examination by fluorescent microscopy. For staining
with Thioflavin-S (TF-S), serial sections of brain tissue were
immersed in 10% neutral buffered formalin (NBF) for 1 h,
followed by bleaching of lipofuscin in 0.05% KMnO4 and
destaining in 0.2% K2S2O5/0.2% oxalic acid prior to staining
with TF-S for 3 min (0.0125% 3 TF-S in 40% EtOH). Sections
were quickly differentiated in 50% EtOH prior to examination
by fluorescent microscopy with a UV filter set (wide excitation
filter at 360 nm and long pass emission filter at 420 nm). Data
for Z,E-5 are not shown in Figure 2; it stains the brain section
as well as the other three isomers at 0.05 mM.
Isom er iza tion of Z,Z-5 to E,E-5 a t Differ en t Tem p er -
a tu r es. Variable time NMR spectra of Z,Z-5 at 70, 95, and
115 °C were measured in order to determine the activation
energy for the isomerization (Figures 2 and 3). An NMR tube,
containing 11 mg of Z,Z-5, which was dissolved in 0.1 mL of
1.4 M NaOD in D2O, followed by adding D2O to bring final
volume to 0.6 mL (0.23 M), was placed in a heating block filled
with sand. The temperature was maintained constant through-
out the experiment. Separate NMR samples were prepared for
each temperature. At various time points, the NMR spectra
were acquired, the integral of each isomer was calculated, and
the data were plotted (Figure 3a,b).
(5) Moore, C. L.; Leatherwood, D. D.; Diehl, T. S.; Selkoe, D. J .;
Wolfe, M. S. Difluoro ketone peptidomimetics suggest a large
S1 pocket for Alzheimer’s gamma-secretase: implications for
inhibitor design. J . Med. Chem. 2000, 43, 3434-3442.
(6) Masel, J .; J ansen, V. A. A. Designing drugs to stop the formation
of prion aggregates and other amyloids. Biophys. Chem. 2000,
88, 47-59.
(7) Findeis, M. A. Approaches to discovery and characterization of
inhibitors of amyloid beta-peptide polymerization. Biochim.
Biophys. Acta 2000, 1502, 76-84.
(8) Thorsett, E. D.; Latimer, L. H. Therapeutic approaches to
Alzheimer’s disease. Curr. Opin. Chem. Biol. 2000, 4, 377-382.
(9) Selkoe, D. J . Imaging Alzheimer’s amyloid. Nat. Biotechnol.
2000, 18, 823-824.
(10) Klunk, W. E.; Debnath, M. L.; Pettegrew, J . W. Small-molecule
beta-amyloid probes which distinguish homogenates of Alzhe-
imer’s and control brains. Biol. Psychiatry 1994, 35, 627.
(11) Klunk, W. E.; Debnath, M. L.; Koros, A. M.; Pettegrew, J . W.
Chrysamine-G, a lipophilic analogue of Congo red, inhibits Aâ-
induced toxicity in PC12 cells. Life Sci. 1998, 63, 1807-1814.
(12) Elhaddaoui, A.; Pigorsch, E.; Delacourte, A.; Turrell, S. Competi-
tion of congo red and thioflavin S binding to amyloid sites in
Alzheimer’s diseased tissue. Biospectroscopy 1995, 1, 351-356.
(13) Styren, S. D.; Hamilton, R. L.; Styren, G. C.; Klunk, W. E. X-34,
a fluorescent derivative of Congo Red: a novel histochemical
stain for Alzheimer’s disease pathology. J . Histochem. Cytochem.
2000, 48, 1223-1232.
(14) Link, C. D.; J ohnson, C. J .; Fonte, V.; Paupard, M.-C.; Hall, D.
H.; et al. Visualization of fibrillar amyloid deposits in living,
transgenic Caenorhabditis elegans animals using the sensitive
amyloid dye, X-34. Neurobiol. Aging 2001, 22, 217-226.
(15) Skovronsky, D.; Zhang, B.; Kung, M.-P.; Kung, H. F.; Trojan-
owski, J . Q.; et al. In vivo detection of amyloid plaques in a
mouse model of Alzheimer’s disease. Proc. Natl. Acad. Sci. U.S.A.
2000, 97, 7609-7614.
(16) Winkler, V. W.; Nyman, M. A.; Egan, R. S. Diethylstilbestrol
cis-trans isomerization and estrogen activity of diethylstil-
bestrol isomers. Steroids 1971, 17, 197-207.
(17) White, W. A.; Ludwig, N. H. Isomerization of R,R′- diethylstil-
bestrol, isolation and characterization of the cis isomer. Biochim.
Biophys. Acta 1971, 19, 388-390.
(18) Filler, R.; Lin, S.; Zhang, Z. Synthesis of fluorovinylsalicylic acids
and their derivatives. J . Fluorine Chem. 1995, 74, 69-76.
(19) Katz, T. J .; Sudhaka, R. A.; Teasley, M. F.; Gilbert, A. M.; Geiger,
W. E. et al. Synthesis and properties of optically active helical
metallocene oligomers. J . Am. Chem. Soc. 1993, 115, 3182-3198.
(20) Munson, P. J .; Rodbard, D. LIGAND: a versatile computerized
approach for characterization of ligand binding system. Anal.
Biochem. 1980, 107, 220-239.
(21) Zhuang, Z. P.; Kung, M. P.; Hou, C.; Skovronsky, D.; Gur, T. L.;
Plo¨ssl, K.; Trojanowski, J . Q.; Lee, V. M.-Y.; Kung, H. F.
Radioiodinated styrylbenzenes and thioflavins as probes for
amyloid aggregates. J . Med. Chem. 2001, 44, 1905-1914.
Ack n ow led gm en t. This work was supported by
grants awarded from the National Institutes of Health
(NS 18509, H.F.K.; PO1 AG-11542, V.M.Y.L.), Institute
for the Study of Aging Research (M.P.K), and Medical
Scientist Training Program Predoctoral Fellowship from
the National Institutes of Health (D.M.S.).
Refer en ces
(1) Na¨slund, J .; Haroutunian, V.; Mohs, R.; Davis, K. L.; Davies,
P.; et al. Correlation Between Elevated Levels of Amyloid [beta]-
Peptide in the Brain and Cognitive Decline. J AMA 2000, 283,
1571-1577.
(2) Selkoe, D. J . The origins of Alzheimer’s disease: A is for amyloid.
J AMA, J . Am. Med. Assoc. 2000, 283, 1615-1617.
(3) Wolfe, M. S.; Citron, M.; Diehl, T. S.; Xia, W.; Donkor, I. O.; et
al. A substrate-based difluoro ketone selectively inhibits Alzhe-
imer’s γ-secretase activity. J . Med. Chem. 1998, 41, 6-9.
J M010161T