Synthesis and 11c-labelling of (E,E)-1-(3′,4′-dihydroxystyryl)-4-(3′-methoxy-4 ′-hydroxystyryl) benzene for PET imaging of amyloid deposits

Article abstract of DOI:10.1002/jlcr.585

Carboxylic acid derivatives of the amyloid-binding dye Congo red do not enter the brain well and are thus unable to serve as in vivo amyloid-imaging agents. A neutral amyloid probe, (E,E)-1-(3′,4′-dihydroxystyryl)-4-(3′-methoxy-4 ′-hydroxystyryl)benzene (3), devoid of any carboxylate groups has been designed and synthesized via a 12-step reaction sequence with a total yield of 30%. The unsymmetric compound 3 has also been labelled with C-11 via [¹¹C]methyl iodide ([¹¹C]CH₃I) methylation of a symmetric 4,4′-dimesyl protected precursor followed by deprotection. Preliminary evaluation indicated that compound 3 selectively stained plaques and neurofibrillary tangles in post-mortem AD brain, and exhibited good binding affinity (K_i = 38 ± 8 nM) for Aβ(1-40) fibrils in vitro. In vivo pharmacokinetic studies indicated that [¹¹C]3 exhibited higher brain uptake than its carboxylic acid analogs and good clearance from normal control mouse brain. [¹¹C]3 also exhibited specific in vivo binding to pancreatic amyloid deposits in the NOR-beta transgenic mouse model. These results justify further investigation of 3 and similar derivatives as surrogate markers for in vivo quantitation of amyloid deposits. Copyright

Full text of DOI:10.1002/jlcr.585

JOURNAL OF LABELLED COMPOUNDS AND RADIOPHARMACEUTICALS
J Label Compd Radiopharm 2002; 45: 647–664.
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jlcr.585
Research Article
11
Synthesis and C-labelling of (E,E)-1-(30,40-
dihydroxystyryl)-4-(30-methoxy-40-
hydroxystyryl) benzene for PET imaging of
amyloid depositsy
1
1
Yanming Wang1, , Chester A. Mathis , Guo-feng Huang ,
Daniel P. Holt1, Manik L. Debnath2 and William E. Klunk2
1 P.E.T. Facility, Department of Radiology, School of Medicine,
University of Pittsburgh, Pittsburgh, PA 15213, USA
*
2 Laboratory of Molecular Neuropharmacology, Department of Psychiatry,
University of Pittsburgh School of Medicine, Pittsburgh, PA 15213, USA
Summary
Carboxylic acid derivatives of the amyloid-binding dye Congo red do not enter
the brain well and are thus unable to serve as in vivo amyloid-imaging agents. A
neutral amyloid probe, (E,E)-1-(30,40-dihydroxystyryl)-4-(30-methoxy-40-hydro-
xystyryl)benzene (3), devoid of any carboxylate groups has been designed and
synthesized via a 12-step reaction sequence with a total yield of 30%. The
unsymmetric compound 3 has also been labelled with C-11 via [11C]methyl
iodide ([11C]CH3I) methylation of a symmetric 4,40-dimesyl protected
precursor followed by deprotection. Preliminary evaluation indicated that
compound 3 selectively stained plaques and neurofibrillary tangles in post-
mortem AD brain, and exhibited good binding affinity (Ki=38 ꢀ 8 nM) for
Ab(1–40) fibrils in vitro. In vivo pharmacokinetic studies indicated that [11C]3
exhibited higher brain uptake than its carboxylic acid analogs and good
clearance from normal control mouse brain. [11C]3 also exhibited specific
in vivo binding to pancreatic amyloid deposits in the NOR-beta transgenic
*Correspondence to: Y. Wang, P.E.T. Facility, Department of Radiology, School of Medicine,
University of Pittsburgh, Pittsburgh, PA 15213, USA. E-mail: wangy2@msx.upmc.edu
yDedicated to Professor Dr Dieter Seebach on the occasion of his 65th birthday and in recognition
of his many contributions to chemical science.
Copyright # 2002 John Wiley & Sons, Ltd.
Received 18 December 2002
Revised 23 January 2002
Accepted 25 February 2002

648
Y. WANG ET AL.
mouse model. These results justify further investigation of 3 and similar
derivatives as surrogate markers for in vivo quantitation of amyloid
deposits. Copyright # 2002 John Wiley & Sons, Ltd.
Key Words: Alzheimer’s disease; b-amyloid; 11C-labelling; brain uptake;
NOR-beta transgenic mouse model
Graphical abstract
Introduction
Progress has been made towards the development of new therapies
based upon the amyloid cascade hypothesis of Alzheimer’s disease
(AD).1 Optimal development of these anti-amyloid therapies,2,3 such as
b- and g-secretase inhibition and b-amyloid (Ab) immunization, awaits
methods to assess amyloid deposition in the living brain. A useful
biomarker of amyloid deposition in the brain could make possible
identification of individuals at risk for AD and facilitate the evaluation
of anti-amyloid therapies through molecular imaging techniques4 such
as positron emission tomography (PET) and single-photon emission
computed tomography (SPECT).
Towards this goal, efforts have been made to develop lipophilic
Congo red (CR) analogs that have high affinity for amyloid and are
capable of penetrating the blood–brain barrier (BBB).5 CR analogs,
such as Chrysamine G (CG) and X-34 (1,4-bis(30-carboxy-40-hydroxy-
styryl)benzene, 1) (Figure 1), are salicylic acid derivatives and exhibit
only marginal brain entry.6–11 This is likely due to the known
detrimental effects of aromatic carboxylic acids on brain entry.12 As a
strategy to increase the permeability across the BBB, this study focused
on the effects of replacing the acidic COOH groups of 1 with OH groups
on both binding to amyloid fibrils and brain permeability. Unlike
carboxylic acids, aromatic OH groups (pKa 8–11) are largely neutral at
physiologic pH, but could potentially participate in polar interactions
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2002; 45: 647–664

PET IMAGING OF AMYLOID DEPOSITS
649
NH2
H2N
N
N
HO
HO
OH
N
N
Congo red
2
3
HO
HO
OH
NaO3S
SO3Na
N
N
OH
HO
N
N
OH
COOH
Chrysamine G
OCH3
HOOC
HO
HOOC
OH
COOH
HO
OCH3
1
4
HO
OH
Figure 1. Structures of amyloid-binding agents: Congo red (CR), Chrysamine G
(CG), (E,E)-1,4-bis(30-carboxy-40-hydroxystyryl)benzene 1, (E,E)-1,4-bis(30,40-
dihydroxystyryl)benzene 2, (E,E)-1-(30,40-dihydroxystyryl)-4-(30-methoxy-40-hy-
droxystyryl)benzene 3, and (E,E)-1-(30,40-dihydroxystyryl)-4-(30-hydroxy-40-
methoxystyryl)benzene 4
with Ab in a manner similar to the COOH groups. In addition, the
hydroxyl group can also be readily methylated with methyl iodide. With
our ultimate goal being development of an in vivo PET amyloid imaging
agent, we report here the design and synthesis of (E,E)-1-(30,40-
dihydroxystyryl)-4-(30-methoxy-40-hydroxystyryl)benzene (3), which
can be produced in a 11C-labelled form via methylation of an aromatic
OH group of the precursor (19). Preliminary evaluation indicated that 3
exhibited good permeability across the BBB and retained good affinity
for Ab fibrils in vitro.
Results and discussion
In order to examine the effect of replacing COOH groups of 1 with OH,
we initially studied 1,4-bis(30,40-dihydroxystyryl)benzene 2. Compared
to the dicarboxylic acid 1, compound 2 showed no significant loss of
affinity for Ab fibrils (Figure 2). Since our ultimate goal is the
development of a 11C-labelled compound for use with PET, we also
studied the two mono-methoxy analogs 3 and 4 (Figure 1). Mono-
methoxy analogs were studied because methylation of 2 with a tracer
amount of no-carrier-added [11C]CH3I resulted only in mono-methoxy
analogs for stoichiometric reasons. Preliminary studies indicated that
rodent brain uptake of the 30-O11CH3 analog (3) was ca. five-fold higher
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2002; 45: 647–664

650
Y. WANG ET AL.
Figure 2. (Left) Fluorescent microscopic image of post-mortem AD brain stained
with 3. Arrows indicate neurofibrillary tangles. The large, round, bright structures
are amyloid plaques. Bar=100 mm. (Right) Binding curves for the inhibition of
[3H]CG binding to Ab(1–40) fibrils by 1 (squares), 2 (circles) and 3 (triangles)
than the 40-OMe analog (4) at 2 min post-injection. Therefore, we
focused on the synthesis and 11C-labelling of 3, free from the 40-O11CH3
analog.
Although the synthesis of symmetric distyrylbenzene compounds has
been reported,13 synthesis of the unsymmetric variants has been rarely
documented.14 We thus developed a convergent synthetic approach to 3
based on Wittig–Horner coupling. The total synthesis involved 12 steps
with a total yield of 30% (Scheme 1). Beginning with methoxymethyl
(MOM) protection of 3,4-dihydroxybenzaldehyde 5b and 3-methoxy-4-
hydroxybenzaldehyde 5a, both MOM-protected aldehydes 6a and 6b
were reduced with sodium borohydride (NaBH4) to benzyl alcohols 7a
and 7b. Iodination of the resulting benzyl alcohols with iodine in the
presence of triphenyl phosphine (PPh3)/imidazole afforded the corre-
sponding benzyl iodides 8a and 8b. Subsequent Arbuzov reactions with
triethyl phosphate (P(OEt)3) yielded the benzylphosphonates 9a and 9b.
Compound 9a was first coupled with terephthaldehyde monoacetal 10
in the presence of sodium hydride (NaH) in good yield. After converting
the acetal 11 to the aldehyde, the monostyrylbenzaldehyde 12 was then
coupled with 3,4-diMOMO-benzylphosphonate 9b to give the MOM-
protected unsymmetric distyrylbenzene compound 13. This second
coupling proceeded well if potassium tert-butoxide was used as the base.
Cleavage of the MOM groups led to 3 in quantitative yield. NMR
analysis indicated that only the (trans, trans)-isomer of the product was
obtained (Scheme 1).
In addition, we developed a method for radiolabelling 3 selectively at
the 30-position via [11C]CH3I methylation of symmetric 19 (Scheme 2).
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2002; 45: 647–664

PET IMAGING OF AMYLOID DEPOSITS
651
CHO
OH
CHO
CH OH
2
MOMCl
K2CO3/acetone
NaBH4/EtOH
6a 96%
6b 93%
7a 91%
7b 94%
OR
OR
OR
OMOM
OMOM
5a R = Me
5b R = H
7a R = Me
7b R = MOM
6a R = Me
6b R = MOM
O
P(OEt)
2
CH I
2
I2/PPh3/Imidazole
P(OEt)3
8a 85%
8b 84%
9a 98%
9b 91%
OR
OR
OMOM
OMOM
8a R = Me
8b R = MOM
9a R = Me
9b R = MOM
CHO
MeO
MOMO
NaH/THF
91%
5% HCl/AcOEt
90%
+ 9a
CH(OEt)
2
CH(OEt)
2
11
10
MeO
9b/KOBut
MOMO
MOMO
OMOM
MOMO
80%
CHO
OMe
13
12
AcOH/H2O (3:1)
100%
HO
OH
OMe
HO
3
Scheme 1. Synthesis of the unsymmetric distyrylbenzene derivative 3
NaOCH3/DMSO
+
MeO
MOMO
CHO
MeO
MOMO
OMe
OMOM
(EtO) P
2
P(OEt)
83%
2
O
O
16
15
14
NaSEt/KOBut
63%
MsCl/Et3N
HO
OH
MsO
MOMO
OMs
OMOM
93%
18
MOMO
OMOM
17
19
1. [11C]CH3I
AcOH/H2O (3:1)
100%
MsO
OMs
HO
OH
CH
K2CO3/DMSO
2. KOBut/THF
11
O
[
11C]3
HO
OH
HO
3
Scheme 2. Synthesis of the symmetric precursor 19 and C-11-labelling of 3
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2002; 45: 647–664

652
Y. WANG ET AL.
Because the precursor was in great excess relative to the tracer amount
of CH3I (ꢁ100:1), only monomethylation in the 30-position could be
detected. Similar to the synthesis of the cold standard, the precursor 19
was synthesized through Wittig–Horner reaction of the xylylene
diphosphonate 15 with 3-MOMO-4-methoxybenzaldehyde 14 in good
yield. The 4-methyl groups of 16 were then replaced with mesyl groups
to give 18. Cleavage of the 3-MOM groups gave the corresponding 19,
which was then methylated with [11C]CH3I. The C-11-labelled inter-
mediate was then treated with potassium tert-butoxide to cleave the
mesyl groups. After purification by high-pressure liquid chromatogra-
phy (HPLC), approximately 370–925 MBq (10–25 mCi) of [11C]3 was
routinely produced with a radiochemical purity of >95%, a chemical
purity of >80%, and specific activity >56 MBq/nmol at end of a
40 min synthesis (EOS). The radiochemical yield was 2.3% ꢀ 0.9 at
EOS based on [11C]CH3I (n=10), which was not optimized.
Like 1, compound 3 is fluorescent and stains amyloid plaques in AD
brain tissue sections (Figure 2). Compound 3 also stains the other
typical pathological feature found in post-mortem AD brain, the
neurofibrillary tangle.15 The predominant b-pleated sheet structure
common to amyloid plaques and neurofibrillary tangles has been
posited to mediate the binding of the Congo red class of compounds,
including 1. Compared to the absence of staining in age-matched
normal control brain (not shown), this indicates a selective binding of 3
to AD pathology. In addition, compound 3 competed with [3H]CG
binding to synthetic Ab(1–40) fibrils with an affinity (Ki=38 nM) similar
to that of CG (Ki=3 ꢀ 0.3 nM),
1
(Ki=6 ꢀ 0.3 nM),
2
(Ki=12 ꢀ 0.4 nM), and CR (Ki=48 ꢀ 6 nM) (Figure 2).
[11C]3 entered mouse brain in significant amounts, 2.25 ꢀ 0.98%
ID/g in whole brain (7.16 ꢀ 3.07% ID/g in whole blood) at 2 min after
intravenous (i.v.) injection. This value is consistent with levels achieved
by typical PET neuroimaging agents. At 30 min post-injection, the
concentration of radioactivity in mouse brain dropped to 0.71 ꢀ 0.33%
ID/g (2:30 min ratio=3.2:1), indicating good efflux of free and non-
specifically bound [11C]3 out of normal mouse brain.
The in vivo pharmacokinetic profile of [11C]3 was further evaluated in
the NOR-beta transgenic mouse pancreatic model of Ab amyloid
deposition.16 As shown in Figure 3, by 60 min post-[11C]3 injection, the
concentration of radioactivity in the amyloid-containing pancreas of the
transgene positive mice (Tg+, open circles) was found to be two-fold
higher than that in the pancreas of transgene negative littermates
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2002; 45: 647–664

PET IMAGING OF AMYLOID DEPOSITS
653
6
4
2
0
0
90
30
60
120
Time (min)
Figure 3. Time–radioactivity course of [11C]3 in the pancreas of transgene
positive amyloid-containing NOR-beta mice (open circles) and transgene negative
littermates (open square). Filled symbols at 60 min represent tissue levels of [11C]3
after pre-treatment with 40 mg/kg MeO-X-04 in transgene positive (filled circle)
or transgene negative (filled triangle) mice
(Tgꢂ, open squares). As a further test of whether the difference in
radioactivity concentrations between Tg+ and Tgꢂ pancreas was due to
saturable and specific binding of [11C]3 to Ab deposits, a competitive in
vivo binding experiment was carried out with (E,E)-2-methoxy-1,4-
bis(40-hydroxystyryl)benzene (MeO-X-04), another amyloid-binding
agent with high affinity (Ki=27 nM for Ab(1–40)).17 Mice were first
treated with MeO-X-04 (40 mg/kg) through intraperitoneal (i.p.)
injection 60 min prior to injection of [11C]3. At 60 min post-[11C]3
injection, the concentration of radioactivity in Tg+ mice pre-treated
with MeO-X-04 (filled circle) was the same as that in the untreated Tgꢂ
control mice (open square) (Figure 3). In addition, radioactivity levels in
Tgꢂ control pancreas were not affected by pre-treatment with MeO-X-
04 (filled triangle) (Figure 3). This suggests that the increased retention
of radioactivity in Tg+ pancreas was due to saturable and specific
binding of [11C]3 to Ab deposits.
In summary, we have synthesized and 11C-labelled compound 3,
which showed good permeability across the BBB in normal control
mice. Although the initial brain entry is less than that of recently
reported thioflavin-T derivatives,18,19 this compound represents a
distinct chemical class of amyloid-binding agents, which have different
binding properties compared to the thioflavin-T analogs. The in vivo
pharmacokinetic studies in the NOR-beta transgenic model indicated
that compound 3 may be useful in the study of peripheral amyloid
deposits.
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2002; 45: 647–664

654
Y. WANG ET AL.
Experimental section
General remarks
4-Hydroxy-3-methoxybenzaldehyde (5a), 3,4-dihydroxybenzaldehyde
(5b), terephthaldehyde mono-(diethyl acetal) (10), chloromethyl methyl
ether (MOMCl), triethyl phosphite, triphenyl phosphine, imidazole,
sodium hydride (powder, 95%), potassium tert-butoxide, sodium
borohydride, sodium thioethoxide, iodine were purchased from Aldrich
Chemical Company and used without further purification. Tetraethyl
1,4-xylylene diphosphonate (15) was obtained from TCI. Acetone
(solvent grade) was dried over potassium carbonate (K2CO3) at room
temperature (RT) for more than 24 h. K2CO3 was finely powdered
before use. Tetrahydrofuran (THF) was freshly distilled over sodium
metal and benzophenone.
The high-resolution mass spectra were taken under EI condition on a
double-focusing high-resolution mass spectrometer (AUTOSPEC of
Micromass Inc.). Samples were introduced by direct insertion probe.
The 1H-NMR spectra of all compounds were measured on a Bruker 300
and 500 MHz using TMS as internal reference and were in agreement
with the assigned structures. The thin layer chromatography (TLC) was
performed using Silica Gel 60 F254 from EM Sciences and detected
under UV lamp. Flash chromatography was performed on silica gel 60
(230–400 mesh; purchased from Mallinckrodt Company; column sizes:
54.0 cm ꢃ 2.5 cm, 36.5 cm ꢃ 2.5 cm or 39.2 cm ꢃ 2.2 cm) using hexanes
and ethyl acetate (EtOAc) as eluents.
3-Methoxy-4-methoxymethoxybenzaldehyde (6a).
4-Hydroxy-3-methoxybenzaldehyde (5a) (3.0 g, 20 mmol) was dissolved
in acetone (100 ml). To this solution, K2CO3 (15 g, 109 mmol) and
MOMCl (8.0 ml, 105 mmol) were added at RT. The reaction mixture
was allowed to stir at RT overnight (ꢁ16 h). The suspension was
filtered and the residue was washed with ethyl acetate (20 ml). After
evaporation of the solvent, the residue was purified by flash
chromatography on silica gel (70 g) eluted with ethyl acetate/hexanes
(1:4) to give 4.0 g (96%) of 6a as colorless oil with Rf 0.38 (hexanes/ethyl
acetate 2:1).
1H-NMR d(300 MHz, CDCl3): 3.48(3 H, s, OCH3), 3.90(3 H, s, CH3),
5.28(2 H, s, CH2), 7.72–7.40(3 H, arom), 9.82(1 H, s, CHO). HRMS: m/z
calcd for C10H12O4 (M+) 196.0736, found 196.0741.
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2002; 45: 647–664

PET IMAGING OF AMYLOID DEPOSITS
655
3,4-Dimethoxymethoxybenzaldehyde (6b)
This compound was synthesized in a manner analogous to 6a starting
with 5b to give 3.1 g (93%) of 6b as white solid with Rf 0.71 (hexanes/
ethyl acetate 1:1) and m.p. 40.5–45.58C.
1H-NMR d(300 MHz, CDCl3): 3.52(3 H, s, CH3), 3.52(3 H, s, CH3),
5.33(2 H, s, CH2), 5.30(2 H, s, CH2), 7.68–7.26(3 H, arom), 9.86(1 H, s,
CHO). HRMS: m/z calcd for C11H14O5 (M+) 226.0841, found
226.0842.
3-Methoxy-4-methoxymethoxybenzyl alcohol (7a)
3-Methoxy-4-methoxymethoxybenzaldehyde (6a) (4.0 g, 20 mmol) was
dissolved in anhydrous ethanol (EtOH) (100 ml). NaBH4 (0.4 g,
11 mmol) was added. The reaction mixture was allowed to stir at RT
for 10 min followed by the evaporation of solvent. The residue was
dissolved in water (100 ml) and extracted with EtOAc (50 ml ꢃ 3). The
organic layers were combined, dried over magnesium sulfate (MgSO4)
(10 g) and filtered. Evaporation of solvent gave 3.7 g (91%) of 7a as
colorless oil.
1H-NMR d(300 MHz, CDCl3): 3.48(3 H, s, OCH3), 3.84(3 H, s,
OCH3), 4.56(2 H, s, OCH2), 5.18(2 H, s, CH2), 6.80–7.10(3 H, arom),
9.86(1 H, s, CHO). HRMS: m/z calcd for C10H14O4(M+) 198.0892,
found 198.0888.
3,4-Dimethoxymethoxybenzyl alcohol (7b)
This compound was synthesized in a manner analogous to 7a starting
with 6b to give 2.8 g (94%) of 7b as colorless oil with Rf 0.41 (hexanes/
ethyl acetate 1:1).
1H-NMR d(300 MHz, CDCl3): 3.54(3 H, s, OCH3), 3.54(3 H, s,
OCH3), 4.63(2 H, s, CH2O), 5.25(2 H, s, CH2), 5.26(2 H, s, CH2),
6.99–7.21(3 H, arom). HRMS: m/z calcd for C11H16O5 (M+) 228.0998,
found 228.0996.
3-Methoxy-4-methoxymethoxybenzyl iodide (8a)
3-Methoxy-4-methoxymethoxybenzyl alcohol (7a) (3.7 g, 19 mmol) was
dissolved in THF (100 ml) followed by addition of PPh3 (5.4 g, 20 mmol)
and imidazole (1.5 g, 22 mmol). The reaction mixture was cooled with
ice-water bath. Iodine (5.2 g, 20 mmol) in THF (50 ml) was added over
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2002; 45: 647–664

656
Y. WANG ET AL.
20 min. After being stirred for an additional 30 min, the reaction was
quenched by addition of water (0.5 ml) and the solvent was evaporated.
The residue was purified by flash chromatography on silica gel (90 g),
eluted with hexanes/ethyl acetate (85:15) to give 4.9 g (85%) of 8a as
white solid with Rf 0.23 (hexanes/ethyl acetate 4:1).
1H-NMR d(300 MHz, CDCl3): 3.50(3 H, s, CH3O), 3.89(3 H, s,
CH3O), 4.46(2 H, s, CH2I), 5.22(2 H, s, CH2O), 7.00–7.19(3 H, m,
arom.).
3,4-Dimethoxymethoxybenzyl iodide (8b)
This compound was synthesized in a manner analogous to 8a starting
with 7b to give 3.2 g (84%) of 8b as colorless oil with Rf 0.46 (hexanes/
ethyl acetate 3:1).
1H-NMR d(300 MHz, CDCl3): 3.51(3 H, s, CH3O), 3.53(3 H, s,
CH3O), 4.44(2 H, s, CH2I), 5.22(2 H, s, CH2O), 5.23(2 H, s, CH2O).
7.00–7.19(3 H, m, arom.).
Diethyl 3-methoxy-4-methoxymethoxybenzylphosphonate (9a)
3-Methoxy-4-methoxymethoxybenzyl iodide (8a) (4.8 g, 15 mmol) was
dissolved in triethyl phosphite (10 ml) in a vial (20 ml). The vial was
sealed and heated at 1408C overnight (ꢁ16 h). The volatiles and excess
triethyl phosphite were removed in a kugelrohr oven (1008C) under high
vacuum (ꢁ1 mmHg) to give 4.8 g (98%) of 9a as colorless oil.
1H-NMR d(300 MHz, CDCl3): 1.23(6 H, t, J=7.1 Hz, 2CH3),
3.06(2 H, d, J=21.3 Hz, PCH2), 3.47(3 H, s, CH3O), 3.84(3 H, s,
CH3), 3.99(4 H, m, J=7.17 Hz, 2CH2OP), 5.17(2 H, s, CH2O),
6.74–7.06(3 H, m, arom.). HRMS: m/z calcd for C14H123O6P (M+)
318.1232, found 318.1246.
Diethyl 3,4-dimethoxymethoxybenzylphosphonate (9b)
This compound was synthesized in a manner analogous to 9a starting
with 8b to give 3.0 g (91%) of 9b as colorless oil.
1H-NMR d(300 MHz, CDCl3): 1.22(6 H, t, J=7.1 Hz, 2CH3),
3.04(2 H, d, J=21.3 Hz, PCH2), 3.47(6 H, s, 2CH3O), 3.99(4 H, m,
J=7.26 Hz, 2CH2OP), 5.17(2 H, s, CH2O), 5.18(2 H, s, CH2O).
6.80–7.10(3 H, m, arom.). HRMS: m/z calcd for C15H25O7P (M+)
348.1337, found 348.1344.
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2002; 45: 647–664

PET IMAGING OF AMYLOID DEPOSITS
657
(E)-4-diethoxymethyl-30-methoxy-40-methoxymethoxystilbene (11)
Diethyl 3-methoxy-4-methoxymethoxybenzylphosphonate (9a) (2.0 g,
6.3 mmol) and terephthaldehyde mono-(diethyl acetal) (10) (1.3 g,
6.3 mmol) were dissolved in THF (150 ml) followed by addition of
sodium hydride (0.5 g, 19 mmol). The reaction was refluxed for 3 h and
then cooled to RT. The excess sodium hydride was hydrolyzed with
water (0.5 ml) and the solvent was evaporated. The residue was
dissolved in water (100 ml) and extracted with ethyl acetate
(3 ꢃ 30 ml). The combined extracts were dried over MgSO4 (5 g),
filtered, and evaporated to give an oil which was purified by flash
chromatography on silica gel (60 g). Elution with hexanes/ethyl acetate
(85:15) gave 2.1 g (91% ) of 7 as a light yellow oil with Rf 0.40 (hexanes/
ethyl acetate 3:1).
1H-NMR d(300 MHz, CDCl3): 1.25(6 H, t, J=6.72 Hz, CH3),
3.52(3 H, s, CH3O), 3.54–3.66(4 H, m, 2CH2O), 5.23(2 H, s, CH2O),
5.52(1 H, s, CHOEt), 6.90–7.55(9 H, m, 9CH).
(E)-4-(30-methoxy-40-methoxymethoxystyryl)benzaldehye (12)
(E)-4-diethoxymethyl-30-methoxy-40-methoxymethoxystilbene (11) (3.0 g,
8.0 mmol) was dissolved in ethyl acetate (50 ml), and 5% aqueous HCl
solution (25 ml) was added. The reaction mixture was allowed to stir at
RT for 30 min. The organic layer was separated and the aqueous layer
was extracted with ethyl acetate (30 ml ꢃ 3). The combined organic
layers were dried over MgSO4 (15 g) and evaporated to give 2.1 g (90%)
of 12 as yellow solid with Rf 0.39 (hexanes/ethyl acetate 2:1).
1H-NMR d(300 MHz, CDCl3): 3.53(3 H, s, CH3O), 3.95(3 H, s,
CH3O), 5.26(2 H, s, CH2O), 6.94–7.86(9 H, m, 9CH), 9.97(1 H, s, CHO).
HRMS: m/z calcd for C18H118O4 (M+) 298.1205, found 298.1200.
(E,E)-1-(30,40-dimethoxymethoxystyryl)-4-(30-methoxy-40-methoxy-
methoxystyryl)benzene (13)
4-(30-Methoxy-40-methoxymethoxystyryl)-benzaldehye
(12)
(1.8 g,
6.2 mmol) and diethyl 3,4-dimethoxymethoxybenzyl phosphonate (9b)
(2.2 g, 6.2 mmol) were dissolved in THF (100 ml). After addition of
potassium tert-butoxide (2.1 g, 19 mmol), the reaction mixture was
refluxed for 30 min and then cooled to RT. The solvent was evaporated
and the residue was dissolved in water (150 ml) and extracted with ethyl
acetate (6 ꢃ 50 ml). The combined extracts were dried over MgSO4 and
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2002; 45: 647–664

658
Y. WANG ET AL.
filtered. The solvent was evaporated to give a yellow solid which
was purified by flash chromatography on silica gel (80 g), eluted
with hexanes/ethyl acetate (2:1) to give 2.4 g (80%) of 13 as a
yellow solid with Rf 0.41 (hexanes/ethyl acetate 1:1) and m.p. 103.0–
106.08C.
1H-NMR d(300 MHz, CDCl3): 3.55(6 H, s, 2CH3O), 3.58(3 H, s,
CH3O), 3.96(3 H, s, CH3O), 5.26(4 H, s, 2CH2O), 5.31(2 H, s, CH2O),
6.90–7.50(14 H, m, 14 CH). HRMS: m/z calcd for C29H32O7 (M+)
492.2148, found 492.2130.
(E,E)-1-(30,40-dihydroxystyryl)-4-(30-methoxy-40-hydroxystyryl)
benzene (3)
(E,E)-1-(30,40-dimethoxymethoxystyryl)-4-(30-methoxy-40-methoxy-
methoxystyryl)benzene (13) (0.46 g, 0.85 mmol) was suspended in acetic
acid (AcOH)/H2O (3:1, 80 ml) and heated at 1008C for 1 h. After cooling
to RT, the reaction mixture was poured into 800 ml of water. The
aqueous solution was extracted with ethyl acetate (100 ml ꢃ 5).
The combined extracts were dried over MgSO4 (30 g), filtered, and
evaporated. The residue was dried under high vacuum (ꢁ1 mm Hg)
overnight (ꢁ16 h) to give 0.34 g of 3 as yellow solid with quanti-
tative yield with Rf 0.15 (hexanes/ethyl acetate 1:1) and m.p. 1708C
(dec.).
1H-NMR d(300 MHz,
acetone-d6):
3.91(3 H,
s,
CH3O),
6.81–7.57(14 H, m, 14CH). HRMS: m/z calcd for C23H20O4 (M+)
360.1362, found 360.1357.
3-Methoxymethoxy-4-methoxybenzaldehyde (14)
To a solution of 3-hydroxy-4-methoxy-benzaldehyde (9.5 g, 63 mmol) in
175 ml of methylene chloride (CH2Cl2) was added dimethoxymethane
(50 ml, 0.56 mol) followed by the addition of p-toluenesulfonic acid
(73 mg). The reaction mixture was refluxed overnight while water
generated as the reaction proceeded was removed using a soxhlet
(
extractor containing 27 g of 3 A molecular sieves. After the completion
of the reaction, 0.5 ml of triethylamine was added to the dark, black–
purple solution, which was then washed three times with 1 N NaOH and
two times with 100 ml of water. The organic layer was dried over
MgSO4 and evaporated to give 2.0 g (16% yield) of 3-methoxymethoxy-
4-methoxybenzaldehyde.
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2002; 45: 647–664

PET IMAGING OF AMYLOID DEPOSITS
659
1H-NMR (500 MHz, DMSO-d6) 7.63(dd, J1=8.14 Hz, J2=2.03, 1 H,
5-H); 7.54(d, J=2.03 Hz, 1 H, 2-H); 7.23(d, J=8.14, 1 H, 6-H); 5.24(s,
2 H, CH2); 3.89(s, 3 H, CH3O). 3.40(s, 3 H, CH3O).
(E, E)-1,4-bis(30-methoxymethoxy-40-methoxystyryl)benzene (16)
To a solution of tetraethyl 1,4-xylylene diphosphonate (15) (0.76 g
2.0 mmol) in 3 ml of dried dimethyl sulfate (DMSO) was added sodium
methoxide solution in methanol (1.1 ml, 5 mmol) followed by addition
of a solution of 3-methoxymethoxy-4-methoxybenzaldehyde (14)
(0.78 g, 4.0 mmol) in 3 ml of dried DMSO. The reaction mixture was
then stirred overnight at RT. It was then poured into 200 ml of water.
After adjusting the pH to 5.1, the water layer was extracted with 200 ml
of ethyl acetate. The organic layer was then washed with water
(3 ꢃ 100 ml) and dried over MgSO4. After evaporation of the solvent,
the residue was recrystallized by dissolving in a minimal amount of
THF followed by adding hexanes until the pure product precipitated
giving 0.76 g (83% yield) of 16.
1H-NMR(500 MHz, DMSO-d6): 7.57(s, 4 H, C6H4); 7.34(d,
J=2.03 Hz, 2 H, 20-H), 7.22(dd, J1=8.65 Hz, J2=2.03 Hz, 2 H, 50-H);
7.20(d, J=16.28 Hz, 2 H, ethylene); 7.08(d, J=16.28 Hz, 2 H, ethylene);
7.02(d, J=8.65 Hz, 2 H, 60-H); 5.21(s, 4 H, CH2); 3.80(s, 6 H, 40-OCH3);
3.43(s, 6 H, OCH3).
(E, E)-1,4-bis(30-methoxymethoxy-40-hydroxystyryl)benzene (17)
To a solution of the (E, E)-1,4-bis(30-methoxymethoxy-40-methoxystyr-
yl)benzene (16) (0.40 g, 0.87 mmol) in 20 ml of dried N,N-dimethyl
formamide (DMF) was added potassium tert-butoxide (0.98 g,
8.7 mmol) and sodium thioethoxide (NaSEt) (0.74 g, 8.7 mmol). The
dark solution was then refluxed for 140 min. The reaction was complete
according to HPLC analysis. The reaction mixture was poured into
200 ml of water (pH 3.8) and extracted with 200 ml of ethyl acetate. The
organic layer was then washed with water (3 ꢃ 100 ml) and dried over
MgSO4. The residue was then purified by dissolving in minimal amount
of ethyl acetate followed by adding hexanes until the pure product
precipitated giving 0.21 g (63% yield) of 17.
1H-NMR(500 MHz, DMSO-d6): 7.53(s, 4 H, C6H4); 7.28(d,
J=2.03 Hz, 2 H, 20-H), 7.14(d, J=16.28 Hz, 2 H, ethylene); 7.10(dd,
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2002; 45: 647–664

660
Y. WANG ET AL.
J1=8.65 Hz, J2=2.03, 2 H, 50-H); 7.00(d, J=16.28 Hz, 2 H, ethylene);
6.83(d, J=8.65 Hz, 2 H, 60-H); 5.20 (s, 4 H, CH2); 3.44(s, 6 H, OCH3).
(E, E)-1,4-bis(30-methoxymethoxy-40-methanesulfonyloxystyryl)-
benzene (18)
To a solution of (E, E)-1,4-bis(30-methoxymethoxy-40-hydroxystyryl)-
benzene (17) (0.18 g, 0.40 mmol) in 30 ml of dried methylene chloride
was added triethylamine (1.7 ml, 12 mmol) and mesyl chloride (0.62 ml,
8.0 mmol). The reaction mixture was stirred at RT for 22 h. After
complete conversion as monitored by HPLC, the reaction mixture was
poured into 100 ml of 0.1 N HCl and extracted with methylene chloride.
The organic layer was washed with water (3 ꢃ 20 ml) and dried over
MgSO4. The residue was then purified by dissolving in minimal amount
of THF followed by adding hexanes until the pure product precipitated
giving 0.19 g (83% yield) of 18.
1H-NMR(500 MHz, CDCl3): 7.52(s, 4 H, C6H4); 7.39(d, J=2.03 Hz,
2 H, 20-H), 7.31(d, J=8.65 Hz, 2 H, 60-H); 7.19(dd, J1=8.65 Hz,
J2=2.03, 2 H, 50-H); 7.07(s, 4 H, ethylene); 5.30(s, 4 H, CH2); 3.55
(s, 6 H, OCH3); 3.22(s, 6 H, SCH3).
(E, E)-1,4-bis(30-OH-40-methanesulfonyloxystyryl)benzene (19)
To cleave the MOM groups, the mesylated intermediate 18 (50 mg,
90 mmol) was added to 3 ml of glacial acetic acid in a 5-ml vial. After
adding 1 ml of water, the reaction mixture was heated at 1108C for
85 min. It was then poured into 100 ml of water (pH 6.4). The product
was then extracted with 100 ml of ethyl acetate. The organic layer was
washed with water (3 ꢃ 100 ml) and dried over MgSO4. The crude
product was then purified dissolving in minimal amount of THF
followed by adding hexanes until the pure product precipitated giving
39 mg (83% yield) of 19.
1H-NMR(500 MHz, DMSO): 7.63(s, 4 H, C6H4); 7.24(d, J=16.28 Hz,
2 H, ethylene), 7.23(d, J=8.65 Hz, 2 H, 60-H); 7.19(d, J=2.03 Hz, 2 H,
20-H); 7.16(d, J=16.28 Hz, 2 H, ethylene); 7.13(dd, J1=8.65 Hz,
J2=2.03, 2 H, 50-H); 3.36(s, 6 H, SCH3).
(E,E)-1,4-bis(30,4,-dihydroxystyryl)benzene (2)
To a solution of tetraethyl 1,4-xylylene diphosphonate (0.19 g,
0.5 mmol) in 5.0 ml of DMF was added potassium tert-butoxide
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2002; 45: 647–664

PET IMAGING OF AMYLOID DEPOSITS
661
(0.56 g, 5.0 mmol) followed by the addition of 3,4-di(methoxymethoxy)
benzaldehyde (0.16 g, 1.0 mmol) in 2.0 ml of DMF. After stirring
overnight at RT, NaSEt (2.0 g, 24 mmol) and 20 ml of DMF were
added. The reaction mixture was refluxed overnight and poured into
200 ml of water. After adjusting the pH to 6.7 with diluted HCl solution,
the water layer was extracted with 200 ml of ethyl acetate. The organic
layers were washed with water (3 ꢃ 100 ml) and dried over MgSO4. The
residue was recrystallized in THF and hexanes to give 0.13 g (77%) of 2.
Binding studies with synthetic Ab (1-40) fibrils.18 The appropriate
concentrations of cold inhibitors to be tested were combined with
[3H]CG (specific activity 35 Ci/mmol) in a volume of 0.95 ml of binding
buffer (150 mM Tris–HCl, pH 7.0 containing 20% ethanol). The assay
was begun by addition of 50 ml of 100 nM Ab stock to achieve a final
concentration of 0.30 nM [3H]CG, 5.0 nM Ab fibrils, and the appro-
priate concentration of test compound. After incubation for 60 min at
RT, the binding mixture was filtered through a Whatman GF/B glass
filter via a Brandel M-24R cell harvester (Gaithersburg, MD) and
rapidly washed twice with 3.0 ml binding buffer. The filters were placed
in Cytoscint-ES (ICN Biomedical, Irvine, CA) and immediately
counted. Complete (100%) inhibition of binding was defined as the
number of counts displaced by 10 mM unlabelled CG. All assays were
done at least in triplicate.
Tissue staining. Tissue autofluorescence was quenched and staining was
accomplished by the procedure used for 1 staining described in detail by
Styren et al.20 Briefly, deparaffinized, quenched tissue sections were
taken from PBS into a 100 mM solution of 3 in 40% ethanol (adjusted to
pH 10 with 0.1 N NaOH) for 10 min. The sections were then dipped
briefly five times into tap water before differentiation in 0.2% NaOH in
80% ethanol for 2 min. The sections were then placed in tap water for
10 min prior to coverslipping with Fluoromount-G (Electron Micro-
scopy Sciences, Fort Washington, PA). Fluorescent sections were
examined using an Olympus Vanox AH-RFL-LB fluorescence micro-
scope and were optimally viewed with a V-filter set (excites 400–410 nm,
dichroic mirror DM455, 455 nm longpass filter).
Radiolabelling. The [11C]methylation of 19 was performed via the
reaction of no-carrier-added [11C]CH3I with 1.0 mg 19 in the presence of
1–2 mg K2CO3 in 0.4 ml of DMSO at 658C for 5 min. After the reaction,
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2002; 45: 647–664

662
Y. WANG ET AL.
0.3 ml of potassium tert-butoxide (1.0 M in THF) was added and the
mixture was heated for an additional 3 min at 658C. The reaction
mixture was diluted with 18 ml of 0.1 M ammonium formate and eluted
through a Waters C8 Sep Pak Plus. The Sep Pak was subsequently
eluted with 0.8 ml of acetonitrile followed by 0.5 ml of 0.1 M ammonium
formate into a V-vial. The solution was purified by semi-preparative
HPLC (Phenomenex Prodigy ODS Prep 10 mm 250 ꢃ 10 mm eluted with
50/50 acetonitrile/TEA buffer [TEA=50 mM triethylammonium phos-
phate, pH 7.2]) at 5 ml/min. The fraction containing [11C]3 (k0 3.5) was
diluted with 30 ml of 18.2 MO-cm deionized water and eluted through a
C8 Sep Pak Plus. The Sep Pak was washed with 10 ml of deionized
water and the product was eluted with 1.0 ml of absolute EtOH into a
sterile vial containing 9 ml saline. Quality control was performed on the
final formulation by analytical HPLC using a Waters 996 photodiode
array detector set at 369 nm with Bioscan Radio-HPLC detector. [11C]3
was found to co-elute with an authentic sample of 3 (k0=3.6) on a
Phenomenex Prodigy ODS (3) 5 mm 250 ꢃ 4.6 mm column eluted with
50/50 acetonitrile/TEA buffer [TEA=50 mM triethylammonium phos-
phate, pH 7.2] at 2 ml/min.
Brain uptake studies in mice. Female Swiss–Webster mice (n=3–5 at
each time point) weighing 20–30 g were injected with 5–30 mCi of [11C]3
(specific activity ꢁ37 MBq/nmol) contained in 0.1 ml of isotonic saline
solution (containing ꢁ5% ethanol from the Sep Pak elution) into a
lateral tail vein. The mice were anesthetized at 2 or 30 min post-injection
and killed by cardiac excision following cardiac puncture to obtain a
terminal arterial blood sample. The brains were rapidly removed and
dissected into cerebellum and remaining whole brain (including brain
stem) fractions. Radioactivity content of brain and blood samples was
assayed using a gamma well-counter (Packard Instruments Model 5003,
Meridan CT), and the counts were decay corrected to the time of
injection relative to C-11 standards prepared from the injection solution
to determine the percent injected dose (% ID) in the samples. The
samples were weighed to determine the percent injected dose per gram
tissue (% ID/g).
NOR-beta mouse studies. NOR-beta mice (Tg positive and Tg negative:
n=3–5 at each time point) weighing 20–30 g were injected with 5–30 mCi
of [11C]3 (specific activity ꢁ37 MBq/nmol) contained in 0.1 ml of
isotonic saline solution (containing ꢁ5% ethanol from the Sep Pak
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2002; 45: 647–664

PET IMAGING OF AMYLOID DEPOSITS
663
elution) into a lateral tail vein. The mice were anesthetized at 2, 30, 60,
and 120 min post-injection and killed by cardiac excision following
cardiac puncture to obtain a terminal arterial blood sample. The mice
were dissected and the pancreas was rapidly removed. Radioactivity
content of the pancreas and blood samples was assayed using a gamma
well-counter and the % ID/g was determined by the method described
in the above procedure.
NOR-beta mouse blocking studies. NOR-beta mice (Tg positive and Tg
negative n=3) weighing 20–30 g were treated with an i.p. injection of
40 mg/kg unlabelled MeO-X-0417 in sterile water (buffered to pH 8.0
using sodium bicarbonate for solubility). At 60 min post-treatment, the
mice were injected with 50–100 mCi of [11C]3 (specific activity ꢁ37 MBq/
nmol) contained in 0.1 ml of isotonic saline solution (containing ꢁ5%
ethanol from the Sep Pak elution) into a lateral tail vein. The mice were
anesthetized at 60 min post-injection and killed by cardiac excision
following cardiac puncture to obtain a terminal arterial blood sample.
Radioactivity content of the pancreas and blood samples was assayed
using a gamma well-counter and the % ID/g was determined by the
method described in the above procedure.
Acknowledgements
The NOR-beta transgenic mice were kindly made available to us by
Aventis Pharmaceuticals, Inc. This work was supported in part by
grants from the Alzheimer’s Association (IIRG-95-076) and Aventis
Pharmaceuticals, Inc (WEK).
References
1. Hardy JA, Higgins GA. Science 1992; 256: 184–185.
2. Schenk DB, Seubert P, Lieberburg I, Wallace J. Arch Neurol 2000; 57:
934–936.
3. Nunan J, Small DH. FEBS Lett 2000; 483: 6–10.
4. Selkoe DJ. Nat Biotechnol 2000; 18: 823–824.
5. Klunk WE. Neurobiol Aging 1998; 19: 145–147.
6. Klunk WE, Debnath ML, Pettegrew JW. Neurobiol Aging 1995; 16: 541–
548.
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2002; 45: 647–664

664
Y. WANG ET AL.
7. Mathis CA, Mahmood K, Debnath ML, Klunk WE. J Label Compd
Radiopharm 1997; 40: 94–95.
8. Zhen W, Han H, Anguiano M, Lemere CA, Cho CG, Lansbury Jr PT. J
Med Chem 1999; 42: 2805–2815.
9. Dezutter NA, Dom RJ, de Groot TJ, Bormans GM, Verbruggen AM. Eur
J Nucl Med 1999; 26: 1392–1399.
10. Skovronsky DM, Zhang B, Kung MP, Kung HF, Trojanowski JQ, Lee
VM. Proc Natl Acad Sci USA 2000; 97: 7609–7614.
11. Zhuang ZP, Kung MP, Hou C, Skovronsky DM, Gur TL, Plossl K,
Trojanowski JQ, Lee VM, Kung HF. J Med Chem 2001; 44: 1905–1914.
12. Salminen T, Pulli A, Taskinen J. J Pharm Biomed Anal 1997; 15: 469–477.
13. Meier H. Angew Chem Int Ed Engl 1992; 31: 1399–1540.
14. Zerban G, Meier H. Z Naturforsch 1993; 48b: 171–184.
15. Mann DM. Ann Med 1989; 21: 133–136.
16. Shoji T, Sato M, Sasaki A, Matsubara E, Igeta Y, Kanai M, et al.
Neurobiol Aging 1998; 19: S59–63.
17. Klunk WE, Bacskai BJ, Mathis CA, Kajdasz ST, McLellan ME, Frosch
MP, Debnath ML, Holt DP, Wang Y, Hyman BT. J Neurophath Exp
Neurol 2002; in press.
18. Klunk WE, Wang Y, Huang GF, Debnath ML, Holt DP, Mathis CA. Life
Sci 2001; 69: 1471–1484.
19. Mathis CA, Bacskai BJ, Kajdasz ST, McLellan ME, Frosch MP, Hyman
BT, Holt DP, Wang Y, Huang GF, Debnath ML, Klunk WE. Bioorg Med
Chem Lett 2002; 12: 295–298.
20. Styren SD, Hamilton RL, Styren GC, Klunk WE. J Histochem Cytochem
2000; 48: 1223–1232.
Copyright # 2002 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2002; 45: 647–664