Syntheses and evaluation of fluorinated benzothiazole anilines as potential tracers for β-amyloid plaques in Alzheimer's disease

Article abstract of DOI:10.1016/j.jfluchem.2007.11.005

[¹¹C]2-(4′-(Methylamino)phenyl)-6-hydroxybenzothiazole ([¹¹C]PIB) is a most potential PET tracer for detecting the β-amyloid plaques in Alzheimer's disease. Here the syntheses of three fluorinated PIB, namely 2-(4′-(methylamino)phenyl)-6-fluoroethoxybenzothiazole (O-FEt-PIB), 2-(4′-(methylamino)phenyl)-6-fluoro-benzothiazole (F-N-Me) and 2-(4′-(dimethylamino)phenyl)-6-fluorobenzo-thiazole (F-N,N-Me), and the radiosynthesis of one corresponding ¹⁸F-labeled PIB compound, [¹⁸F]O-FEt-PIB, as well as their in vitro/in vivo biological characters were reported. The structures of the products were confirmed by IR, ¹H NMR, EI/ESI-MS, elemental analysis and HRMS techniques. The radiolabeled product was characterized by radio-TLC and radio-HPLC and purified by semi-preparative radio-HPLC. The suitable biological characters showed these tracers were potential to be developed as probes for detecting β-amyloid plaques in Alzheimer's disease.

Full text of DOI:10.1016/j.jfluchem.2007.11.005

Available online at www.sciencedirect.com
Journal of Fluorine Chemistry 129 (2008) 210–216
www.elsevier.com/locate/fluor
Syntheses and evaluation of fluorinated benzothiazole anilines as
potential tracers for b-amyloid plaques in Alzheimer’s disease
c
a
a
Ming-Qiang Zheng a,b, Duan-Zhi Yin a, , Jin-Ping Qiao , Lan Zhang , Yong-Xian Wang
*
a Research Center of Radiopharmaceuticals, Shanghai Institute of Applied Physics, Chinese Academy of Sciences, Shanghai 201800, China
b Graduate School of the Chinese Academy of Sciences, 19 YuQuan Road, Beijing 100049, China
c Hefei National Laboratory for Physical Sciences at Microscale and Department of Neurobiology, School of Life Sciences,
University of Science & Technology of China, Hefei, Anhui 230027, China
Received 5 October 2007; received in revised form 5 November 2007; accepted 5 November 2007
Available online 17 November 2007
Abstract
[11C]2-(40-(Methylamino)phenyl)-6-hydroxybenzothiazole ([11C]PIB) is a most potential PET tracer for detecting the b-amyloid plaques in
Alzheimer’s disease. Here the syntheses of three fluorinated PIB, namely 2-(40-(methylamino)phenyl)-6-fluoroethoxybenzothiazole (O-FEt-PIB),
2-(40-(methylamino)phenyl)-6-fluoro-benzothiazole (F-N-Me) and 2-(40-(dimethylamino)phenyl)-6-fluorobenzo-thiazole (F-N,N-Me), and the
radiosynthesis of one corresponding 18F-labeled PIB compound, [18F]O-FEt-PIB, as well as their in vitro/in vivo biological characters were
reported. The structures of the products were confirmed by IR, 1H NMR, EI/ESI-MS, elemental analysis and HRMS techniques. The radiolabeled
product was characterized by radio-TLC and radio-HPLC and purified by semi-preparative radio-HPLC. The suitable biological characters showed
these tracers were potential to be developed as probes for detecting b-amyloid plaques in Alzheimer’s disease.
# 2007 Elsevier B.V. All rights reserved.
Keywords: Alzheimer’s disease; F-N-Me; F-N,N-Me; [18F]O-FEt-PIB; PET
1. Introduction
fluorine at the 6-position of BTA will increase lipophilicity and
binding affinity to b-amyloid plaques. The syntheses of
fluorinated PIB analogues have been under investigation for
many years owing to their characters as potential fluorescence,
MRI or PET tracers for detecting amyloid in AD brain [6–8].
Considering the short half-life of carbon-11 (20.4 min), one
kind of fluorine-18 (109.6 min)-labeled PIB, [18F]PIB, was
synthesized and showed excellent in vitro and ex vivo
properties and might be as a viable lead candidate for in vivo
human studies [9]. Another fluorine-18-labeled PIB,
[18F]FBTA (Fig. 2), was successfully synthesized and its
characters were under investigation by autoradiography of
human AD brain sections and micro-PET in a transgenic mouse
[10]. Here we reported the convenient synthesis of two fluorine-
contained BTA compounds, 2-(40-(methylamino)phenyl)-6-
fluorobenzo-thiazole (F-N-Me) and 2-(40-(dimethylamino)phe-
nyl)-6-fluorobenzo-thiazole (F-N,N-Me). Both of them showed
higher binding affinity to b-amyloid in the AD human brain
homogenate than PIB [11]. Fluorescence staining of AD human
brain sections with F-N-Me and F-N,N-Me confirmed the
specific binding to b-amyloid. The more interesting finding was
that F-N-Me might also bind to tangles, which is similar to that
Alzheimer’s disease (AD) is a neurodegenerative disease,
with aggregates of b-amyloid (Ab) peptides and neurofibrillary
tangles in the brain. Positron emission tomography (PET) is a
very useful and powerful technology to study human brain. PET
imaging of AD has many aspects of significance for diagnostic
and therapeutic developments. Small molecules for detecting
the b-amyloid plaques and tangles as potential PET ligands
have been well developed [1]. Among them, benzothiazole
anilines (BTA; Fig. 1), such as [11C]PIB, are the most potential
tracers and have been used in clinical studies [2]. BTAs are
derivatives of thioflavin-T which is an amyloid-binding dye.
Thioflavin has been used as a pharmacophore for development
of amyloid-binding compounds [3,4]. It is well known that
introducing a fluorine atom to a compound will probably
change its biological activities and pharmacological properties
[5]. What is more, it is estimated that substitution of OH to
* Corresponding author. Fax: +86 21 59554696.
E-mail address: comingzeng@yahoo.com.cn (D.-Z. Yin).
0022-1139/$ – see front matter # 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.jfluchem.2007.11.005

M.-Q. Zheng et al. / Journal of Fluorine Chemistry 129 (2008) 210–216
211
(1) was prepared by base hydrolysis of the corresponding 2-
amino-6-fluro-benzothiazole, then further reacted with com-
mercially available p-(methylamino)-benzoic acid or (dimethy-
lamino)-benzaldehyde to form the secondary (F-N-Me) or
tertiary amine (F-N,N-Me) derivatives through different
reaction routes. Fluorine atom is an electron-withdrawing
group and this character is useful to increase the yield of
condensing reactions comparing to the electron-donating
group, for example, HO or EtO. Condition c is more convenient
than b, either because of the short reaction time or the simpler
procedure of the purification. But the raw compound p-
(methylamino)-benzaldehyde is not commercial and estimated
to be unstable.
Fig. 1. Structure of 6-substituted benzothiazole anilines (BTA) derivatives [4].
of FDDNP [12]. Most recently, Stephenson et al. reported their
excellent work about some fluoro-pegylated imaging agents as
potential PET tracers for Ab aggregates, which included the
syntheses of similar structures ([18F]5a–c) to [18F]O-FEt-PIB
and their distributions in normal mice [8]. Here we showed
another radiosynthesis method of [18F]O-FEt-PIB and its in
vivo distribution in normal rats was also described. Because
introducing a fluorine-18 atom into benzene ring directly
without adjacent electron-withdrawing groups is almost
impossible except the use of 18F2, [18F]F-N-Me and [18F]F-
N,N-Me are not synthesized and their application in MRI is
under investigation.
2.2. Synthesis of O-FEt-PIB and [18F]O-FEt-PIB
The preparation of O-FEt-PIB from 2-amimophenyl-6-
ethoxybenzothiazole through four steps is shown in Scheme 2.
The synthesis strategy was similar to that in ref. [4] with a
fluoroethylate of PIB at the last step. It was estimated that the
hydrogen at both NH and OH in PIB would be replaced by
fluoroethyl group. However, when we used condition d, O-FEt-
PIB was obtained as a main product. This interesting result
made it possible that [18F]fluoroethyltosylate ([18F]FEtOTs)
could selectively react with PIB in O-position with higher
labeling yield and simpler purification procedure in certain
conditions. [18F]FEtOTs and [18F]O-FEt-PIB were synthesized
2. Results and discussion
2.1. Synthesis of F-N-Me and F-N,N-Me
The synthesis is based on the following sequence of
reactions as shown in Scheme 1. 5-Fluoro o-aminothiophenol
Scheme 1. Synthesis of 2-(40-(methylamino)phenyl)-6-fluorobenzothiazole (F-N-Me) and 2-(40-(dimethylamino)phenyl)-6-fluorobenzothiazole (F-N,N-Me). (a)
Glycol, KOH, reflux for 48 h; (b) polyphosphoric acid. p-(Methylamino)-benzoic acid, 170 8C, 2 h; (c) Me2SO, p-(dimethylamino)-benzaldehyde, 170 8C, 20 min.
Fig. 2. Potential small molecular tracers for b-amyloid plaques [9,10].

212
M.-Q. Zheng et al. / Journal of Fluorine Chemistry 129 (2008) 210–216
Scheme 2. Synthesis of 2-(40-(dimethylamino)phenyl)-6-fluoroethoxybenzothiazole (O-FEt-PIB). (a) Glycol, KOH, reflux for 48 h; (b) polyphosphoric acid, p-
(methylamino)-benzoic acid, 170 8C, 2 h; (c) 1 mol/L BBr3/CH2Cl2 solution, 16 h; (d) DMF, 2-fluoroethyl tosylate, potassium iodide, potassium carbonate, 80 8C, 20 h.
Scheme 3. Synthesis of [18F]O-FEt-PD3. (1) DMF, KI, K2CO3, 80 8C, 20 h, (2) anhydrous CH3CN, 110 8C, 20 min.
as shown in Scheme 3. The labeling yields were high (90 and
85%, respectively), but the total radiochemical yield of [18F]O-
FEt-PIB was only 5–10% (start from 18F anion, total synthetic
time of 95 min, decay not corrected). The reason was that only
one quarter of the reaction-available [18F]FEtOTs was obtained
via the purification by C18 Sep-Park cartridge. The most of the
lost was observed during the purification process through C18
Sep-Pak cartridge which was pre-saturated with water. The
aimed product was identified by radio-TLC [Rf = 0.16
(reference 0.18) in Vhexane/Vethyl acetate = 4:1]. Then the reaction
products were purified by semi-preparative HPLC column
(condition 1). The retention time of [18F]O-FEt-PIB was about
14 min and well separated from other chemical species (for
example, PIB, tR = 9.7 min, [18F]FEtOTs, tR = 11 min and by-
product, tR = 18 min). The analytical HPLC condition (condi-
tion 2) and radio-TLC with two different conditions were used
for quality control (RCP > 95%, tR = 18.2 min (reference
tR = 17.4 min); Rf = 0.66 in CHCl3 (reference Rf = 0.62) and
0.16 in V(hexane)/V(ethyl acetate) = 4:1 (reference Rf = 0.18)).
Specific activity estimated by comparing UV peak intensity of
purified [18F]O-FEt-PIB and corresponding radioactivity was
more than 740 TBq/mmol.
2.4. Fluorescent staining of senile plaques in human AD
brain sections
To demonstrate the specific binding to Ab plaques, the three
compounds and thioflavin-Twere used to stain paraffin sections
of postmortem AD brain (Fig. 4). Like thioflavin-T, the staining
of F-N-Me, F-N,N-Me or O-FEt-PIB was localized to Ab
plaques in the adjacent brain slices. The interesting results were
that F-N,N-Me could either bind to amyloid plaques or
cerebrovascular amyloid; F-N-Me could bind to both plaques
and tangles. No distinct tangle-binding of F-N,N-Me was found.
The identity of the plaques and tangles was confirmed by
staining serial sections of the AD brain slices.
2.5. Biodistribution of [18F]O-FEt-PIB in normal rats
Table 1 shows the brain, blood and bone entry of [18F]O-FEt-
PIB in normal Sprague–Dawley rats in terms of percent injected
2.3. In vitro binding assay in human AD brain homogenate
Compounds F-N-Me, F-N,N-Me and O-FEt-PIB competed
well with [3H]PIB for binding to AD brain homogenate (Fig. 3).
The Ki value for each compound was 0.20 (F-N-Me), 0.31 (F-
N,N-Me) and 0.17 nM (O-FEt-PIB). In the same condition, Ki
value of PIB was 1.59 nM. It confirmed the assumption that
substitution of OH to an electron withdrawing group at the 6-
position of BTA would increase its binding affinity to b-
amyloid.
Fig. 3. Detennination of specific binding of BTA’s derivatives in postmortem
AD brain liomogenate. Each data point came from the average of four
measurements and the bar represents the standard deviation (S.D.).

M.-Q. Zheng et al. / Journal of Fluorine Chemistry 129 (2008) 210–216
213
Fig. 4. Fluorescence micrographs of sections of AD brain (6 mm thick) stained with 1 mm F-N,N-Me (A), F-N-Me (B), O-FEt-PIB (C) and thioflavin-T (D). Section
(A) shows F-N,N-Me could either stain plaques (marked by red arrows) or cerebrovascular amyloid (hollow arrows). Section (B) shows F-N-Me could either stain
plaques (red arrows) or tangles (green arrows). Section (C and D) shows the corresponding plaques in adjacent AD brain slices. (For interpretation of the references to
color in this figure legend, the reader is referred to the web version of the article.)
dose per gram of organ (%ID-kg)/g. The brains were divided
into cortex, cerebrum (except cortex) and cerebellum. Organ
entry (except bone) was high and quickly at early time after
injection. The average (%ID-kg)/g value of [18F]O-FEt-PIB in
rat brains at 2 min p.i. was 0.5 and was good for a potential
neuroreceptor radioligands. Among three parts of the brain, the
uptake in cerebrum was a litter higher than that in cerebellum.
The low uptake and radioactivity in bone indicates that no or
little 18Fꢀ ion was generated while the metabolism. However,
the clearance rate of radioactivity from brain tissues was slow,
with a ratio value of about 2 comparing the radioactivity at 2–
30 min. The ratio of brain radioactivity from 2 to 60 min was
about 6 and this character was not as good as that of PIB (2–
30 min ratio, 12 [4]).
tracer in brain, only one time point was chosen to evaluate the
whole body distribution. It showed that the radioactivity in two
parts of the brain did not distinguish from each other, and more
radioactivities aggregated to cerebrum at 5 min (about 0.8) than
at 2 min (about 0.6). Radioactivities were found mostly in liver
and kidney, with 3.83 and 3.56, respectively. The muscle and
bone were found the least amounts, with 0.48 and 0.39,
individually. Comparing to the biodistribution of [18F]5a in
normal mice which was reported in ref. [8], the initial brain
Table 2
Biodistribntion of
[
18F]O-FEt-PIB in normal rats ((%ID-kg)/g, n = 5,
rats ꢁ S.D.)
5 min
Biodistribution studies in normal rats with 37 MBq [18F]O-
FEt-PIB at 5 min postinjection were performed (Table 2).
Considering the slow clearance rate and rapid uptake of the
Organ (body)
Blood
1.40 ꢁ 0.36
0.48 ꢁ 0.06
1.92 ꢁ 0.89
3.83 ꢁ 0.49
3.56 ꢁ 0.74
1.25 ꢁ 0.35
1.06 ꢁ 0.19
0.39 ꢁ 0.07
Muscle
Lung
Liver
Table 1
Kidney
Spleen
Heart
3B (brain, blood and bone) uptake and clearance of [18F]O-FEt-PIB in normals
rats at different time points after intravenous injection ((%ID-kg)/g, average of 3
rats ꢁ S.D.)
Bone
Organ
2 mm
30 mm
60 mm
120 mm
Organ (brain)
Cerebellum
Blood
1.79 ꢁ 0.21
0.21 ꢁ 0.13
0.50 ꢁ 0.28
0.43 ꢁ 0.25
0.64 ꢁ 0.32
0.46 ꢁ 0.15
0.10 ꢁ 0.04
0.28 ꢁ 0.16
0.18 ꢁ 0.04
0.32 ꢁ 0.22
0.20 ꢁ 0.05
0.09 ꢁ 0.02
0.08 ꢁ 0.03
0.06 ꢁ 0.02
0.09 ꢁ 0.04
0.12 ꢁ 0.05
0.11 ꢁ 0.03
0.06 ꢁ 0.03
0.04 ꢁ 0.01
0.06 ꢁ 0.02
0.46 ꢁ 0.04
0.43 ꢁ 0.09
0.43 ꢁ 0.21
0.81 ꢁ 0.20
0.86 ꢁ 0.27
Bone
Cortex (left)
Cerebel
Cortex
Cerebrum
Cortex (right)
Cerebrum (left)
Cerebrum (right)

214
M.-Q. Zheng et al. / Journal of Fluorine Chemistry 129 (2008) 210–216
uptake of [18F]O-FEt-PIB in rats was less than that in mice
(10.27% dose/g at 2 min postinjection), but the washout rate
(rat, 2–60 min, 6) was faster than that in mice (2–60 min, 2.6).
Another distinct difference existed in the bone uptake. The
bone uptake in mice increased with time which indicated that in
vivo defluorination was likely occurring. But in rats, no similar
phenomenon was found.
Radio-TLC analyse was performed using silica gel 60 GF-
254 plates (Merck, Darmstadt, Germany) on a Bioscan system
AR-2000 with Winscan software of Version 3.09. The
radioactivity measurements were done using an automatic
gamma counter (3-in. NaI (Tl) well crystal) coupled to a
multichannel analyzer (SN682B).
All commercial reagents and solvents were used without
further purification unless otherwise specified. 2-Amino-6-
fluorobenzothiazole, 2-amino-ethoxybenzothiazole, p-(methy-
lamino)-benzoic acid, p-(dimethylamino) benzaldehyde, 1,2-
bis(tosylate)ethane and 2-fluoroethyl tosylate were purchased
from Sigma–Aldrich Synthesis Ltd. (America) or Chemical
Reagent Companies in Shanghai (China).
3. Conclusion
Three fluorine-contained BTA compounds, F-N-Me, F-N,N-
Me and O-FEt-PIB as well as [18F]O-FEt-PIB was successfully
synthesized. Binding assay showed that the binding affinity for
each compound was higher than that of PIB. Fluorescent
staining of senile plaques in human AD brain sections also
showed the specific binding of these compounds to b-amyloid
plaques. The interesting result was that F-N-Me could either
bind to b-amyloid plaques or tangles. Another two compounds,
F-N,N-Me and O-FEt-PIB, did not show this ability in our
experiments. [18F]O-FEt-PIB was synthesized via [18F]FEtOTs
as a fluoroethylated agent and the labeling yield was high
(85%). The total synthetic time was 95 min with a radio-
chemical yield of 5–10% (start from 18F anion, decay not
corrected), and the specific activity was estimated more than
740 TBq/mmol. It was able to penetrate the blood–brain barrier
and showed a high brain uptake ((ID%-kg)/g, 0.8 at 5 min p.i.)
in rat, but the washout rate was a little slow (2–60 min ratio, 6).
It was not observed that in vivo defluorination was occurring
because the bone uptake was minimal and not increased with
time. The in vivo evaluation of this tracer in AD model rats
using micro-PET is underway in our laboratory.
The animal experiments were performed in accordance
with the ‘‘Principles of Laboratory Animal Care’’ (NIH
Publication No. 86-23, revised 1985). Male and female
Sprague–Dawley rats, weighing 400–500 g at the beginning
of the experiment, were housed individually in a room
maintained at 23 8C with a 12-h light:12-h dark cycle for the
duration of the experiment.
4.2. Synthetic procedures
4.2.1. 2-Amino-5-fluorothiophenol (1)
The synthetic procedure was similar to that of 2-amino-5-
methoxythiophenol in ref. [4], with a yield of 50.1%. 1H NMR
(acetone-d6, 500 MHz), 6.81 (dd, J1 = 9 Hz, J2 = 2.5 Hz, 1H,
H–4C), 6.56 (t, J1 = 8.5 Hz, J2 = 2.5 Hz, 2H, H–6C and H–3C);
IR (KBr)n: 3427, 2560, 1615, 1487, 1406, 1289, 1200, 1141,
864, 812, 685, 462 cmꢀ1; MS (m/z) (%) calcd. for C6H6NSF
143, found 142 (100), 98 (37). Anal. calcd. for C6H6FNS: C
50.33, H 4.22, F 13.27, N 9.78, S 22.39; found C 50.51, H 4.82,
N 9.26, S 22.17.
4. Experimental
4.1. General experimental procedures
4.2.2. 2-(40-(Methylamino)phenyl)-6-fluorobenzothiazole
(F-N-Me)
1H NMR spectra was recorded on a Bruker AC-500
(500 MHz) instrument with Me4Si as internal standards in the
indicated solution described below. Chemical shifts were
reported in d values in parts per million (ppm) downfield from
Me4Si (d = 0) for proton NMR. Electron impact mass spectra
(EI-MS) were obtained on MicroMass GCT CA 055 spectro-
meters. Infrared spectra (IR) were recorded on Avatar 370 FTIR
(Thermo Nicolet) as KBr disc.
2-Amino-5-fluorothiophenol (0.72 g, 5.0 mmol) and 4-
methylaminobenzoic acid (0.80 g, 5.0 mmol) were mixed
together with PPA (polyphosphoric acid, ꢃ1 g) and heated
to 170 8C under N2 atmosphere with a mechanical mill for
1.5 h. The reaction mixture was cooled to room temperature
and poured into 10% K2CO3 solution. Ethyl acetate (100 mL)
was added to extract the organic compounds. The organic layer
was separated, and the aq. layer was extracted with another
60 mL ethyl acetate (three times). All the organic layers were
combined and washed with water. Evaporation of the solvent,
the residue was first purified by flash column (hexanes:ethyl
acetate = 4:1), then purified by recrystallization from THF/
water to give 0.28 g (21.6%) 2-(40-(methyl-amino)phenyl)-6-
fluorobenzo thiazole as yellow needle crystal. 1H NMR
(CDCl3, 500 MHz), 7.84 (s, 2H, H–20C and H–30C), 7.44
(dd, J1 = 9.0 Hz, J2 = 2.5 Hz, 2H, H–50C and H–60C), 7.10 (t,
J1 = 8.5 Hz, J2 = 2.5 Hz, 1H, H–5C), 6.68 (d, J = 8.5 Hz, 2H,
H–7C and H–4C), 2.81 (s, 3H, H–CH3); IR (KBr)n: 3312, 3273,
3289, 1605, 1564, 1480, 1450, 1276, 1250, 1062, 1049, 1001,
902, 854; MS (%): 258 (100). HRMS calcd. for C14H11FN2S
258.0627, found 258.0631.
High-performance liquid chromatography (HPLC) was
carried out on a Dionex system equipped with a P680 pump,
a PDA-100 photodiode array detector and a NaI (Tl)
scintillation detector. The columns used for identification
and purification were LiChrosorb C18, 10 mm, 300 mm
ꢂ 3.9 mm and 300 mm ꢂ 7.8 mm, respectively. The column
effluent passed through a UV (365 nm) detector and scintilla-
tion radioactivity detector in series. HPLC solvents consisted of
CH3OH (solvent A) and 0.1% CF3COOH/H2O (solvent B).
HPLC gradient: 0–1 min 70% B, 1–25 min 20%, 25–40 min
20%. The flow rates of purification and analysis were 3 mL/min
(condition 1, with column 300 mm ꢂ 7.8 mm) and 1 mL/min
(condition 2, with column 300 mm ꢂ 3.9 mm), separately.

M.-Q. Zheng et al. / Journal of Fluorine Chemistry 129 (2008) 210–216
215
4.2.3. 2-(40-(Dimethylamino)phenyl)-6-fluorobenzothiazole
(F-N,N-Me)
4.2.6. Synthesis of 2-18F-fluoroethyl tosylate
([18F]FEtOTs)
A
mixture of 2-amino-5-fluorothiophenol (760 mg,
[18F]Fluoride, which produced by a cyclotron using 18O(p,
n)18F reaction, was passed through a Sep-Pak Light QMA
cartridge as an aq. solution in [18O]-enriched water. The
cartridge was dried by airflow, and the 18F activity was eluted
with 1 mL of Kryptofix 222/K2CO3 solution (10 mg of K222 and
2 mg of K2CO3 in CH3CN/H2O, v/v = 3.54/0.46). The solvent
was then removed at 100 8C and the residue was azeotropically
dried with 400 mL of CH3CN three times under the nitrogen
stream. A solution of 1,2-bis(tosylate)ethane (6–8 mg) in
anhydrous CH3CN (400 mL) was added to the reaction vessel
which contained the dried 18F fluoride. The mixture was heated
at 110 8C for 10 min and the products were analyzed on radio-
TLC. The Rf value was identical with that of the reference
compound 2-fluoroethyl tosylate.
5.3 mmol) and 4-(dimethylamino) benzaldehyde (799 mg,
5.4 mmol) in Me2SO (5 mL) was heated to 170 8C for
20 min. The reaction mixture was cooled to room temperature
and poured into water. The organic component was extracted
with ethyl acetate (2 ꢂ 30 mL). The combined organic layers
were washed with water and dried over MgSO4. Evaporation of
the solvent, the residue was then purified by flash column
(hexanes:ethyl acetate = 4:1) to give 373 mg (26%) yellow
1
solid. H NMR (acetone-d6, 500 MHz), d: 7.88 (d, J = 8.9 Hz,
2H, H–20C and H–30C), 7.50 (dd, J1 = 8.5 Hz, J2 = 2.5 Hz, 2H,
H–50C and H–60C), 7.04 (dd, J1 = 9.0 Hz, J2 = 2.5 Hz, 8.9,
2.5 Hz, 1H, H–5C), 6.80 (s, 2H, H–7C and H–4C), 3.05 (d,
J = 9.0 Hz, 6H, H–CH3); IR (KBr)n: 3447, 2972, 2917, 1607,
1489, 1455, 1365, 1222, 1166, 1065, 936, 819, 593 cmꢀ1; MS
(%): 272 (100), 256 (13). HRMS calcd. for C15H23FN2S
272.0783, found 272.0781.
4.2.7. Synthesis of [18F]O-FEt-PIB
After condensing the solution of [18F]FEtOTs to about
20 mL, 2 mL of water was added and the mixture was pushed
through a pre-saturated C18 Sep-Pak cartridge. The cartridge
was dried under a nitrogen stream, and then rinsed with
2 ꢂ 1 mL ethyl acetate. The combined organic solution was
collected to a reaction vessel and dried under a nitrogen stream
with gentle heating (about 60 8C). PIB (3 mg) in 200 mL
Me2SO, 1 mol/L K2CO3 solution (20 mL) and CH3CN
(400 mL) were added to the vessel and heated for 20 min at
110 8C. The products were analyzed by radio-TLC and radio-
HPLC, and then purified using semi-preparative HPLC Column
(condition 1). Collected the elution from 13 to 15 min and dried
at 60 8C under a nitrogen stream.
4.2.4. 2-(40-(Methylamino)phenyl)-6-hydroxybenzothiazole
(PIB compound)
The synthetic process was similar to that in ref. [4], with a
yield of 60% as yellow solid. 1H NMR (CDCl3, 500 MHz), 7.81
(d, J = 8.5 Hz, 2H), 7.70 (d, J = 9.0 Hz, 1H), 7.36 (s, 1H), 6.98
(d, J = 9.3 Hz, 1H), 7.0 (d, J = 8.6 Hz, 2H), 2.87 (d, J = 5.5 Hz,
3H); IR (KBr)n: 3395, 2870, 2786, 2665, 1609, 1577, 1494,
1428, 1340, 1279, 1241, 1180, 818 cmꢀ1; MS (m/z) (%) calcd.
for C14H12N2OS 256, found 256 (100), 241 (11).
4.2.5. 2-(40-(Methylamino)phenyl)-6-
fluoroethoxybenzothiazole (O-FEt-PIB)
4.2.8. In vitro binding assays with AD brain homogenates
The AD human brain homogenates 3.21 ꢁ 0.04 mg in the
final mixture, 100 mL) were obtained through National Institute
of Mental Health, National Institute of Health as an aqueous
solutions protected in dry ice, which were homogenated again
before use. The procedure was similar to that described in ref.
[13] with some modification. The binding was assayed in
12 mm ꢂ 75 mm borosilicate glass tubes for saturation studies,
with the mixture contained 100 mL homogenates and 10–
350 mL [3H]PIB (4.00 ꢂ 10ꢀ2 mCi/100 mL; 350, 250, 200,
150, 100, 80, 60, 40, 20, 10 mL) and adding PBS (pH 7.4) to a
final volume of 1 mL. Nonspecific binding was defined in the
presence of 50 mL PIB (1 ꢂ 10ꢀ3mol/L in PBS) in five
different tubes. For the competition binding, 10ꢀ4 to 10ꢀ9 mol/
L compounds and 4.00 ꢂ 10ꢀ2 mCi [3H]PIB were used for the
studies. The mixture was incubated at 37 8C for 2 h. The
radioactivity was counted next day in the scintillation counter
(Beckman) with 57% counting efficiency. Under the assay
conditions, the specifically bound fraction was less than 15% of
the total radioactivity.
2-(40-(Methylamino)phenyl)-6-hydroxybenzothiazole (PIB,
103 mg, 0.4 mmol), anhydrous potassium carbonate (55.4 mg,
0.4 mmol) and potassium iodide (61 mg, 0.37 mmol) were
dissolved in DMF (10 mL), mixed with a magnetic stirer and
heated to 80 8C. 2-Fluoroethyl tosylate (148 mg, 0.6 mmol)
which dissolved in another 30 mL DMF was dropped into the
mixture and then reacted for 20 h at 80 8C. Stopped the
reaction, filtered the solvent while hot, evaporated the filtrate
under reduced pressure and added water to remove the
inorganic salt. Then the residue was filtered, rinsed with water,
rinsed with ethanol and finally gave a brown solid. Purification
by flash column (silica gel, hexanes:ethyl acetate = 4:1) to give
21 mg yellow solid. Analytical HPLC: condition 2,
1
tR = 17.4 min, purity 98% H NMR (CDCl3, 500 MHz), 8.21
(d, J = 8.5 Hz, 2H, H–ArCH2), 8.12 (d, J = 9.0 Hz, 1H, H–5C),
7.38 (d, J = 8.5 Hz, 1H, H–50C), 7.34 (d, J = 9.3 Hz, 1H, H–
60C), 7.18 (dd, J1 = 8.6 Hz, J2 = 2.5 Hz, 2H, H–20C and H–30C),
6.80 (s, 2H, H–7C and H–4C), 4.75 (dd, J1 = 9.0 Hz,
J2 = 5.5 Hz, 2H, H–FCH2), 3.40 (s, 3H, H–CH3); IR (KBr)n:
3443, 3265, 2958, 2924, 1744, 1720, 1609, 1561, 1543, 1489,
1460, 1446, 1333, 1261, 1227, 1208, 1181, 1083, 1021, 964,
940, 876, 802 cmꢀ1; MS (m/z) (%) calcd. for C16H15FN2OS
302, found 302 (42), 255 (100), 227 (16), 95 (23). HRMS calcd.
for C16H15FN2OS 302.0889, found 302.0887.
4.2.9. Fluorescent staining of senile plaques in human AD
brain sections
Postmortem brain tissues from AD cases were obtained
through Hefei National Laboratory for Physical Sciences at

216
M.-Q. Zheng et al. / Journal of Fluorine Chemistry 129 (2008) 210–216
Microscale and Department of Neurobiology, School of Life
Sciences, University of Science & Technology of China, (from
the Netherlands Brain Bank (Coordinator Dr I. Huitinga )). 6 mm
thick serial sections of paraffin embedded blocks from the
hippocampus were used for staining. Tissue was processed for
staining and the staining protocols were following that in ref. [4].
Quenched tissue sections were taken from PBS (pH 7.4) into a
solution of 1 mM F-N-Me, F-N,N-Me, thioflavin-Tor O-FEt-PIB
in PBS for 30 min. Fluorescent sections were viewed using a
BX50 fluoromicroscope with an M-3204C CCD camera
(Olympus, Tokyo) equipped with a B-filter/FITC (excites
360–460 nm).
Acknowledgements
The authors will thank for the Amersham Kexing
Pharmaceuticals Co. Ltd. for supplying us No-carrier-added
[18F]Fꢀ solution. This project was supported by the Science
Foundation of Shanghai (No. 06DZ19506) and National
Natural Science Foundation of China (No. 20601028).
References
[1] L.-S. Cai, R.B. Innis, V.W. Pike, Curr. Med. Chem. 14 (2007) 19–52.
[2] A. Nordberg, Lancet Neurol. 3 (2004) 519–527.
[3] C.A. Mathis, D.P. Holt, Y. Wang, G.F. Huang, M.L. Debnath, W.E. Klunk,
Soc. Nucl. Med. (2002) (Abstract No. 605).
4.2.10. Biodistribution in normal rats of [18F]O-FEt-PIB
3B uptake and clearance studies were performed in male or
female Sprague–Dawley rats, weighing 400–500 g (ꢃ3 months
old). The animal studies were performed according to the
practice for the care and use of animals. Under pentabarbiturate
anesthesia, each animal was weighed and received approxi-
mately 7.4 MBq (200 mCi) of [18F]O-FEt-PIB through a lateral
tail vein. The animals were killed by cardiac excision following
cardiac puncture to obtain arterial blood samples and
decapitation to obtain brain and crania at 2, 30, 60 and
120 min postinjection. The brains were rapidly excised and
divided into cerebellum, cortex and cerebrum. All the samples
were weighed and counted, and the counts were decay-
corrected to the time of injection relative to 18F standards
(prepared from the injection solution). The tissue radioactivity
was determined as percentage injected dose per gram tissue
normalized to body weight (in kg) or (%ID-kg)/g.
[4] C.A. Mathis, Y. Wang, D.P. Holt, G.F. Huang, M.L. Debnath, W.E. Klunk,
J. Med. Chem. 46 (2003) 2740–2754.
[5] H.J. Borhm, D. Banner, S. Bendels, M. Kansy, B. Kuhn, K. Murler, U.
Obst-Sander, M. Stahl, Chem. Biol. Chem. 5 (2004) 637–643.
[6] B.J. Bacskai, G.A. Hickey, J. Skoch, S.T. Kajdasz, Y. Wang, G.-F. Huang,
C.A. Mathis, W.E. Klunk, B.T. Hyman, PNAS 100 (2003) 12462–
12467.
[7] K. Noda-Saita, A. Yoneyama, Y. Shitaka, Y. Hirai, K. Terai, J. Wu, T.
Takeda, K. Hyoda, N. Osakabe, T. Yamaguchi, M. Okada, Neuroscience
138 (2006) 1205–1213.
[8] K.A. Stephenson, R. Chandra, Z.-P. Zhuang, C. Hou, S. Oya, M.-P. Kung,
H.F. Kung, Bioconjugate Chem. 18 (2007) 238–246.
[9] N.S. Mason, W.E. Klunk, M. Debnath, N. Flatt, G. Huang, L. Shao, C.A.
Mathis, J. Label. Compd. Radiopharm. 50 (Suppl. 1) (2007) S87.
[10] U. Berndt, C. Stanetty, M. Berger, H. Kvaternik, T. Wolf, C. Kuntner, T.
Wanek, P. Angelberger, C. Noe, J. Label. Compd. Radiopharm. 50 (Suppl.
1) (2007) S86.
[11] M.-Q. Zheng, D.-Z. Yin, L. Zhang, D.-F. Cheng, Y.-X. Wang, H.-C. Cai, J.
Label. Compd. Radiopharm. 50 (Suppl. 1) (2007) S401.
[12] E.D. Agdeppa, V. Kepe, J. Liu, S. Flores-Torres, N. Satyamurthy, A.
Petric, G.M. Cole, G.W. Small, S.-C. Huang, J.R. Barrio, J. Neurosci. 21
(2001) 1–5.
For the biodistribution studies, each animal received
approximately 37 MBq (1 mCi) of [18F]O-FEt-PIB in
100 mL of 0.9% NaCl solution through a lateral tail vein
under pentabarbiturate anesthesia. Rats (n = 5) were sacrificed
by cervical dislocation. At 5 min postinjection, organs of
interest were removed, weighed and counted. The data were
acquired as the same method described above.
[13] L.S. Cai, F.T. Chin, V.W. Pike, H. Toyama, J.S. Liow, S.S. Zoghbi, K.
Modell, E. Briard, H.U. Shetty, K. Sinclair, S. Donohue, D. Tipre, M.P.
Kung, C. Dagostin, D.A. Widdowson, M. Green, W. Gao, M.M. Herman,
M. Ichise, R.B. Innis, J. Med. Chem. 47 (2004) 2208–2218.