N-[18F]fluoroethyl-4-piperidyl acetate ([18F]FEtP4A): A PET tracer for imaging brain acetylcholinesterase in vivo.

Article abstract of DOI:10.1016/S0968-0896(03)00177-9

N-[(18)F]Fluoroethyl-4-piperidyl acetate ([(18)F]FEtP4A) was synthesized and evaluated as a PET tracer for imaging brain acetylcholinesterase (AchE) in vivo. [(18)F]FEtP4A was previously prepared by reacting 4-piperidyl acetate (P4A) with 2-[(18)F]fluoroethyl bromide ([(18)F]FEtBr) at 130 degrees C for 30 min in 37% radiochemical yield using an automated synthetic system. In this work, [(18)F]FEtP4A was synthesized by reacting P4A with 2-[(18)F]fluoroethyl iodide ([(18)F]FEtI) or 2-[(18)F]fluoroethyl triflate ([(18)F]FEtOTf in improved radiochemical yields, compared with [(18)F]FEtBr under the corresponding condition. Ex vivo autoradiogram of rat brain and PET summation image of monkey brain after iv injection of [(18)F]FEtP4A displayed a high radioactivity in the striatum, a region with the highest AchE activity in the brain. Moreover, the distribution pattern of (18)F radioactivity was consistent with that of AchE in the brain: striatum>frontal cortex>cerebellum. In the rat and monkey plasma, two radioactive metabolites were detected. However, their presence might not preclude the imaging studies for AchE in the brain, because they were too hydrophilic to pass the blood-brain barrier and to enter the brain. In the rat brain, only [(18)F]fluoroethyl-4-piperidinol ([(18)F]FEtP4OH) was detected at 30 min postinjection. The hydrolytic [(18)F]FEtP4OH displayed a slow washout and a long retention in the monkey brain until the PET experiment (120 min). Although [(18)F]FEtP4A is a potential PET tracer for imaging AchE in vivo, its lower hydrolytic rate and lower specificity for AchE than those of [(11)C]MP4A may limit its usefulness for the quantitative measurement for AchE in the primate brain.

Full text of DOI:10.1016/S0968-0896(03)00177-9

Bioorganic & Medicinal Chemistry 11 (2003) 2519–2527
N-[18F]fluoroethyl-4-piperidyl Acetate ([18F]FEtP4A):
A PET Tracer for Imaging Brain Acetylcholinesterase In Vivo
Ming-Rong Zhang,a,b,* Kenji Furutsuka,a,b Jun Maeda,a,b Tatsuya Kikuchi,a,c
Takayo Kida,a,b Takashi Okauchi,a,b Toshiaki Iriea and Kazutoshi Suzukia
aDepartment of Medical Imaging, National Institute of Radiological Sciences, 4-9-1 Anagawa, Inage-ku, Chiba 263-8555, Japan
bSHI Accelerator Service Co. Ltd., 5-9-11, Kitashinagawa, Shinagawa-ku, Tokyo 141-8686, Japan
cDepartment of Pharmaceutical Sciences, Chiba University, 1-1-1 Yayoi, Inage-ku, Chiba 263-8522, Japan
Received 14 January 2003; accepted 11 March 2003
Abstract—N-[18F]Fluoroethyl-4-piperidyl acetate ([18F]FEtP4A) was synthesized and evaluated as a PET tracer for imaging brain
acetylcholinesterase (AchE) in vivo. [18F]FEtP4A was previously prepared by reacting 4-piperidyl acetate (P4A) with 2-[18F]fluoro-
ethyl bromide ([18F]FEtBr) at 130 ꢀC for 30 min in 37% radiochemical yield using an automated synthetic system. In this work,
[18F]FEtP4A was synthesized by reacting P4A with 2-[18F]fluoroethyl iodide ([18F]FEtI) or 2-[18F]fluoroethyl triflate ([18F]FEtOTf
in improved radiochemical yields, compared with [18F]FEtBr under the corresponding condition. Ex vivo autoradiogram of rat
brain and PET summation image of monkey brain after ivinjection of [ 18F]FEtP4A displayed a high radioactivity in the striatum, a
region with the highest AchE activity in the brain. Moreover, the distribution pattern of 18F radioactivity was consistent with that
of AchE in the brain: striatum>frontal cortex>cerebellum. In the rat and monkey plasma, two radioactive metabolites were
detected. However, their presence might not preclude the imaging studies for AchE in the brain, because they were too hydrophilic
to pass the blood–brain barrier and to enter the brain. In the rat brain, only [18F]fluoroethyl-4-piperidinol ([18F]FEtP4OH) was
detected at 30 min postinjection. The hydrolytic [18F]FEtP4OH displayed a slow washout and a long retention in the monkey brain
until the PET experiment (120 min). Although [18F]FEtP4A is a potential PET tracer for imaging AchE in vivo, its lower hydrolytic
rate and lower specificity for AchE than those of [11C]MP4A may limit its usefulness for the quantitative measurement for AchE in
the primate brain.
# 2003 Elsevier Science Ltd. All rights reserved.
Introduction
measuring the AchE activity and elucidating the rela-
tion between AchE and AD in the human brains.14ꢁ16
[11C]MP4A provided an index of regional AchE activity
and kinetic modeling techniques as shown in Scheme
1.14ꢁ16 After injecting this tracer into a subject, [11C]MP4A
penetrated across the blood–brain barrier (BBB) and
was then metabolized by AchE to N-[11C]methyl-4-
piperidinol ([11C]MP4OH), which resulted in a long-
term retention of radioactivity in the brain.
Postmortem studies in patients with Alzheimer’s disease
(AD) suggested that, the most profound change, com-
pared with normal subjects, is a progressive, inexorable
reduction in the activity of acetylcholinesterase (AchE)
in the neocortex and hippocampus.1ꢁ4 Positron emission
tomography (PET) has been used to measure the AchE
activity and to offer the means for longitudinal study of
AchE by imaging this enzyme in human brains with
normal aging and in different stages of AD.5ꢁ10 Using
acetylcholine as a lead compound, a series of positron
emitter-labeled acetylcholine analogues have been syn-
thesized and evaluated for this purpose.11ꢁ13 Of these
In PET chemistry, 18F-labeled radiotracers have several
advantages over 11C-labeled tracers. Due to the longer
half-life of 18F (b+; 96.7%, t1=2 ¼ 109:8 min), 18F is
convenient for long-time storage and long-distance
transportation, and multiple doses can be prepared in
one synthetic run. Moreover, due to the shorter posi-
tron range of 18F than that of 11C (650 KeV vs
960 KeV), 18F often gives a slightly higher quality image
with higher spatial resolution. Since PET measurement
for AchE activity often requires over 40 min, a 18F-
radiotracers,
N-[11C]methyl-4-piperidyl
([11C]MP4A, Scheme 1) was a successful PET tracer for
acetate
*Corresponding author. Tel.: +81-43-206-4041; fax: +81-43-206-
3261; e-mail: zhang@nirs.go.jp
0968-0896/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0968-0896(03)00177-9

2520
M.-R. Zhang et al. / Bioorg. Med. Chem. 11 (2003) 2519–2527
Scheme 1. Schematic diagram representing the biochemical rational of
the radioactive acetylcholine analogue method for mapping AchE
activity in the brain in vivo.
Scheme 2. Chemical synthesis and radiosynthesis: (a) CH3COCl, 25 ꢀC,
18 h; (b) 10% NH3/H2O, 0 ꢀC, 5 min; (c) (t-Boc)2O, NaOH, 25 ꢀC, 1 h;
(d) (CH3CO)2O, pyridine, 25 ꢀC, 6 h; (e) CF3COOH, 0 ꢀC, 1 h; (f)
FCH2CH2Br, i-Pr2NEt, NaI, 70 ꢀC, 5 h; (g) TfOCH2CH2Br, 130 ꢀC,
5 min; (h) P4A, 130 ꢀC, 30 min; (i) P4A, NaI, 130 ꢀC, 10 min; (j) NaOH,
95 ꢀC, 5 min; (k) AgOTf, 300 ꢀC, 10 min; (l) P4A, 25 ꢀC, 10 min, 25 ꢀC.
labeled tracer may be favorable over a 11C-labeled tra-
cer to gain more precise information about AchE activ-
ity. Based on this consideration, using [11C]MP4A as a
lead compound, we synthesized N-[18F]fluoroethyl-4-
piperidyl acetate ([18F]FEtP4A, Scheme 1) as a putative
tracer for AchE.17 First, comparing acetylcholine and
MP4A, FEtP4A with the N-fluoroethyl group was more
lipophilic and could readily enter the brain, which is a
prerequisite for a suitable PET tracer. Second, in the
development of acetylcholine analogues, it was found
that two methyl or ethyl groups were essential for accep-
tance by AchE, but the third methyl group could be
replaceable with a longer alkyl chain.18ꢁ20 Therefore,
introducing the fluoroethyl group for the methyl group
into the piperidine ring might not change its specificity for
AchE. Third, N-[18F]fluoroethyl-4-piperidinol ([18F]FEt-
P4OH), a hydrophilic metabolite resulting from the spe-
cific hydrolysis of [18F]FEtP4A by AchE, may be trapped
and retained at the sites of AchE in the brain (Scheme 1).
[18F]FEtBr/NaI and 2-[18F]fluoroethyl triflate21,22
([18F]FEtOTf); (2) imaging study for AchE in the rat
brain using ex vivo autoradiography and in the monkey
brain using PET; (3) analysis for metabolites of
[18F]FEtP4A in plasma and the brain of rats, and
plasma of monkey; (4) characterization of enzymatic
property in vitro.
Results and Discussion
Chemistry
4-Piperidinol acetate (P4A), the substrate for radio-
synthesis, was previously prepared by the acetylation of
4-piperidinol hydrochloride (Scheme 2).17 However, the
selective O-acetylation required a long time (18–25 h) at
25 ꢀC, because no base could be used for this reaction
where the N-acetylation occurred readily. Under this
limited condition, 4-piperidinol hydrochloride reacted
with acetyl chloride to give P4A only at a low yield of
12%. Recently, an improved chemical synthesis for P4A
was carried out by protecting the N atom in the piper-
idine ring with t-butoxycarbonyl (Boc) group, and then
reacting with acetic anhydride in the presence of pyri-
dine. After the O-acetylation finished perfectly, the Boc
group was removed with trifluoroacetic acid to give P4A
at a total yield of 71% starting from 4-piperidinol
hydrochloride (Scheme 2). This route provided a con-
venient and reliable method for preparing P4A.
In a previous study, we reported the radiosynthesis and
distribution of [18F]FEtP4A in the mouse brain.17
[18F]FEtP4A was synthesized by reacting piperidyl
4-acetate (P4A) with 2-[18F]fluoroethyl bromide
([18F]FEtBr) at 130 ꢀC for 30 min with a 37% radio-
chemical yield (corrected for decay, based on the total
[18F]FEtBr) (Scheme 2). Due to its high uptake into the
brain and distribution pattern corresponding to AchE
in the mouse brain, [18F]FEtP4A appeared to be ideally
suited for PET imaging of AchE. However, the rela-
tively low radiochemical yield and long synthetic time
for [18F]FEtP4A precluded further evaluation which
required a large amount of this tracer. Therefore, we
attempted to increase the fluoroethylation efficiency of
P4A for producing [18F]FEtP4A in a higher radio-
chemical yield, and then to evaluate this tracer by ima-
ging AchE in rat and monkey brains in vivo and
characterizing its enzymatic property.
We developed an automated system for synthesizing
18F-labeled compounds using [18F]FEtBr as a synthetic
precursor.17,23,24 Using this system, we prepared
[18F]FEtBr by heating [18F]F- with 2-bromoethyl triflate
We herein report (1) radiosynthesis of [18F]FEtP4A by
the respective fluoroethylation of P4A with [18F]FEtBr,

M.-R. Zhang et al. / Bioorg. Med. Chem. 11 (2003) 2519–2527
2521
(BrCH2CH2OTf) in a reproducible radiochemical yield
of 60–75% (corrected for the decay, based on total
[18F]F-), and synthesized several [18F]fluoroethylated
ligands including [18F]FEtP4A.23,24 However, the lower
reactivity of [18F]FEtBr with substrates containing
amine, phenol and other nucleophilic functional groups
than that of [11C]CH3I limited its wide application as a
fluoroalkylating agent. P4A reacted rapidly with
[11C]CH3I at 70 ꢀC for 3 min to give [11C]MP4A with a
high radiochemical yield of 60–85%. In contrast to
[11C]MP4A, P4A reacted with [18F]FEtBr to afford
[18F]FEtP4A only at a poor yield of <5% under the
same condition. In this time, to increase the reactivity of
[18F]FEtBr with P4A and to augment the radiochemical
yield of [18F]FEtP4A, [18F]FEtBr was converted into
more reactive (1) [18F]FEtI in situ by adding tiny
amounts of NaI to the fluoroethylation reaction mixture
in advance; (2) [18F]FEtOTf21 by transferring
[18F]FEtBr into a heated column22 containing sliver tri-
flate (Scheme 2).
semi-preparative HPLC system. It was of paramount
importance in the synthesis of [18F]FEtP4A to remove
all traces of the unreacted substrate P4A, because P4A
might have a certain specificity to AchE and obscure the
imaging studies. Complete separation of [18F]FEtP4A
(tR=9.8–11.2 min) from P4A (tR=5.2–7.5 min) could be
accomplished under the present conditions as shown in
the experimental. Using radio-TLC, the identity for
[18F]FEtP4A was confirmed by the observation of co-
elution with authentic non-radioactive FEtP4A. The
radiochemical purity of [18F]FEtP4A was determined to
be 97.2Æ1.8% for all formulations with an analytical
reverse HPLC system. Since FEtP4A displays no special
absorption on UV spectrum (200–400 nm), the specific
activity of [18F]FEtP4A at the end of synthesis was
estimated to be 70–110 GBq/mmol from that of
[18F]FEtBr prepared in other synthetic runs. In the final
product solution (10 mL), the contamination of P4A
was quantified by LC/MS to be 0.6–1.4 mg, which met
the specification of a radiopharmaceutical product pro-
vided for clinical use in our facility. The stability of
[18F]FEtP4A in its prepared form was evaluated by
monitoring its radiochemical purity using the same
analytic HPLC. The radiochemical purity remained
>95% after maintenance of the [18F]FEtP4A prepara-
tion at 25 ꢀC for 4 h at the radioactivity concentration of
37–65 MBq/mL. On the other hand, N-[18F]fluoroethyl-
4-piperidinol ([18F]FEtP4OH), a possible radioactive
metabolite of [18F]FEtP4A, was prepared by hydrolyz-
ing the parent compound with 0.1 N NaOH at >98%
radiochemical yield (Scheme 2).
Table 1 shows the radiochemical yields and synthetic
results of [18F]FEtP4A under various reaction condi-
tions. The reaction efficiency of P4A with [18F]FEtBr/
NaI or [18F]FEtOTf was significantly augmented, com-
pared with [18F]FEtBr. When NaI (1 mg) was added to
the reaction mixture in advance, P4A reacted with
[18F]FEtBr/NaI at 130 ꢀC for 10 min to give
[18F]FEtP4A at a 74% radiochemical yield, whereas,
without NaI, the reaction of P4A with [18F]FEtBr
afforded [18F]FEtP4A only at an 18% yield under the
same conditions. On the other hand, [18F]FEtOTf dis-
played a higher reactivity with P4A than [18F]FEtBr as
well as [18F]FEtBr/NaI. [18F]FEtOTf was synthesized
by flowing [18F]FEtBr into a heated (300 ꢀC) column
containing silver triflate (AgOTf) with helium gas (30–
50 mL/min). After trapping [18F]FEtOTf into the reac-
tion mixture for 10 min at 25 ꢀC, 71% of the total
radioactivity was found to represent [18F]FEtP4A, and
29% was [18F]FEtBr (Table 1). Since P4A did not react
with [18F]FEtBr at 25 ꢀC, the conversion efficiency of
[18F]FEtBr into [18F]FEtOTf was at least 71%.
Evaluation
Using [11C]MP4A as a lead compound, we synthesized
[18F]FEtP4A as a putative PET imaging agent for AchE
in vivo.17 On the basis of structural similarity, it was
anticipated that [18F]FEtP4A would mimic the dis-
tribution and metabolism of [11C]MP4A, resulting in a
similar acceptance by AchE. The preliminary study on
mice displayed that the initial uptake of [18F]FEtP4A
into the mouse brain reached 8% injected dose/g tissue,
and its radioactivity distribution was consistent with
the distribution pattern of AchE activity in the brain:
striatum>cerebral cortex>cerebellum.17 Moreover,
After the [18F]fluoroethylation reaction was terminated,
the radioactive mixture was purified using a reverse
Table 1. Radiosynthesis results of [18F]FEtP4A under various conditions
Reagent
Reaction temperature
(ꢀC)
Reaction time
(min)
Radiochemical yielda
(%)
Purification yieldb
(%)
Synthesis timec
(min)
25
70
130
130
130
25
10
10
10
20
30
10
10
10
20
30
10
10
No reaction
8
18
32
56
3
29
74
86
90
71
67
[
18F]FCH2CH2Br
39
64
50
78
59
48
70
[
[
18F]FCH2CH2Br/NaI
18F]FEtOTf
130
130
130
25
70
aRadiochemical yield as determined by analytic HPLC from a sample withdrawn from the reaction mixture.
bDecay corrected radiochemical yield (%) after HPLC purification based on the total [18F]FEtBr.
cFrom EOB, including the recovery of [18F]F- from the target.

2522
M.-R. Zhang et al. / Bioorg. Med. Chem. 11 (2003) 2519–2527
Figure 1. A: Ex vivo autoradiogram of [18F]FEtP4A in the striatal section of the rat brain at 30 min postinjection (20 MBq). B: Ex vivo autoradio-
graphic localizations of 18F radioactivity in the sagittal sections of rat brains (n=3) at 30 min postinjection of [18F]FEtP4A. The regional uptake of
18F was expressed as photo-stimulated-luminescence (PSL) values per square millimeter of brain section (PSL/mm2).
[18F]FEtP4A was rapidly hydrolyzed to [18F]FEtP4OH
in the mouse brain after injection.17 Thus, [18F]FEtP4A
appeared to be a promising PET tracer for AchE in
vivo. These results prompted us to evaluate the tracer in
rats and monkey by imaging AchE using ex vivo auto-
radiography and PET, and characterizing its enzymatic
property.
from all structures until 10 min. At 10 min postinjection,
the 18F radioactivity in the striatum was greater than
2-fold those in the frontal cortex, thalamus and cere-
bellum. From then, a slow decline of radioactivity was
seen from the striatum, from which the concentration of
18F at the end of the PET scan (120 min) was reduced to
half the 10-min value. On the other hand, the radio-
activities in the frontal cortex, thalamus and cerebellum
remained on plateau levels until the end of PET scan.
Figure 1 shows the autoradiographic localizations of
18F-radioactivity in the rat brain at 30 min postinjection
of [18F]FEtP4A. As can be seen in the sagittal section, a
significant radioactivity was observed in the striatum
(Fig. 1a,b), while a modest uptake was present in the
frontal cortex and thalamus, and the lowest uptake was
in the cerebellum (Fig. 1b). The radioactivity in the
striatum was approximately 3-fold or 2-fold higher than
that in the frontal cortex and thalamus or cerebellum
(Fig. 1b).
In imaging studies of the brain, the presence of meta-
bolites in the plasmas of experimental animal might
preclude the evaluation of a PET tracer if the metabo-
lites entered the brains and were retained/bound at the
Figure 2 shows a PET image of the monkey brain
acquired from 0 to 120 min after [18F]FEtP4A injection
(88 MBq/1.5 mL). This image displays evident accumu-
lation of radioactivity in the striatum compared with the
other regions (frontal cortex, thalamus and cerebellum),
which was similar with the autoradiogram of the rat
brain.
Figure
3 shows time–activity curves (TACs) of
[18F]FEtP4A in the striatum, frontal cortex, thalamus
and cerebellum after injection of [18F]FEtP4A (88 MBq)
into the monkey. Immediately following ivadministra-
tion, the radioactivity began to accumulate in the brain
regions. The initial uptake of [18F]FEtP4A into the
monkey brain was high, reflecting a high penetration of
the tracer across the monkey BBB consistent with its
partition coefficient of 0.78,17 which is favorable for the
permeability of the tracer across the BBB. The radio-
activity in all regions examined reached a peak by 3–
5 min postinjection, and then showed a rapid washout
Figure 2. PET summation image of midstriatial/midthalamic level of
the monkey brain acquired between 0 and 120 min after [18]FEtP4A
injection (88 MBq).

M.-R. Zhang et al. / Bioorg. Med. Chem. 11 (2003) 2519–2527
2523
Figure 3. Time–activity curves (TACs) in the striatum, thalamus,
frontal cortex and cerebellum of the monkey brain after ivinjection of
18F]FEtP4A (88 MBq).
[
Figure 4. Percent conversion of [18F]FEtP4A to metabolites in plasma
at several time points after iv injection of [18F]FEtP4A (50 MBq) into
the monkey. The metabolites and unchanged [18F]FEtP4A were ana-
lyzed by HPLC for CH3CN extracts from plasma and brain prepared
as described in the experimental.
target sites. Therefore, it is necessary to determine whe-
ther the radioactive metabolites in the brain were con-
tributing to the signal observed in vivo. Chemical
analyses of the radioactive metabolite(s) of [18F]FEtP4A
in the plasma and brain were carried out using radio-
TLC and HPLC. After ivadministration of
[18F]FEtP4A (20 MBq) into rats (n=3) at 30 min, the
plasma and brain homogenate were respectively
obtained and immediately deproteinized with CH3CN.
Extraction25 of the radioactivity in the plasma in addi-
tion to the brain into CH3CN proved to be efficient
(>70%). In the HPLC analysis for plasma samples,
three radioactive components were observed. Using
radio-TLC, two of these fractions were respectively
identified to be unchanged [18F]FEtP4A (2% of total
radioactivity) and the hydrolysate [18F]FEtP4OH
(77%). In addition to the two fractions, an unknown
product (21%) that was too polar to be retained on the
column under the present conditions was also detected.
By treating this product with 10% Na2CO3 solution, it
was entirely changed to [18F]FEtP4OH, suggesting it
was glucuronidated [18F]FEtP4OH, the most common
product of phase II metabolism.26 On the other hand,
only [18F]FEtP4OH was detected in the homogenate
sample of rat brains at 30 min postinjection.
that the exchange of two metabolites between the blood
and brain was small in the rat and monkey, although
[18F]FEtP4A was extensively metabolized in the plasma.
Thus, the presence of metabolites in the blood did not
obscure the signals acquired from the ex vivo auto-
radiogram of the rat brain and PET image of the mon-
key brain.
Since (1) the 18F radioactivity of showed a slow washout
from the monkey brain and displayed a relatively long-
term retention in the brain as shown in the TAC of
[18F]FEtPA for the monkey; (2) only [18F]FEtP4OH
was detected in the rat homogenates, the autoradiogram
and PET images obtained in these studies reflected the
distribution of AchE in the brains. However, whether
[18F]FEtP4A could be used for the quantitative
measurement of AchE activity in a primate brain
depends upon its specificity for AchE and the hydrolytic
rate by AchE in the brain. Therefore, the enzymatic
hydrolysis rate and specificity to AchE of [18F]FEtP4A
were measured and compared with [11C]MP4A using rat
brain homogenates. The hydrolysis rates and specificity
measurements of [18F]FEtP4A and [11C]MP4A are
shown in Table 2. [18F]FEtP4A displayed a 14-fold
slower rate of hydrolysis in the rat brain homogenate in
vitro than [11C]MP4A. The specificity of [18F]FEtP4A
for AchE was reduced to 71%, compared with that
(96%) of [11C]MP4A. The reason of the differences
between the enzymatic hydrolysis rates and specificities
of two tracers were due to the discrepancy of N-methyl
and N-fluoroethyl groups. Introduction of the bulky
and electronegative fluoroethyl group into the piperidyl
acetate molecule decreased the binding efficiency to the
AchE, compared with the methyl group. Since the
approach for measuring AchE activity in vivo demands
complete hydrolysis of used PET tracer by this enzyme
within designated scan times, the slower hydrolytic rate
of [18F]FEtP4A limited its usefulness for the quantita-
tive determination for AchE in the primate brain.
Moreover, the relatively low specificity of [18F]FEtP4A
After ivinjection of [ 18F]FEtP4A into the monkey,
[18F]FEtP4A in the monkey plasma decreased rapidly
from 100 to 45% at 10 min (Fig. 4). From then, the
amount of unchanged [18F]FEtP4A continued to
decrease, while those of [18F]FEtP4OH and the same
product determined in the rat plasma increased slowly
during the entire PET experiment. The fraction corres-
ponding to [18F]FEtP4A in the plasma was 25% at
20 min, 17% at 30 min, and 16% at 60 min of the total
radioactivity. At the end of the PET scan (120 min),
[18F]FEtP4A was entirely transformed to [18F]FEt-
P4OH and the putative glucuronidated [18F]FEtP4OH.
Since [18F]FEtP4OH (partition coefficient: 0.06)17 in
addition to the water-soluble product displayed high
hydrophilicities, they could not cross the BBB and enter
the brain. The very low uptake (<0.6% dose/g) of
[18F]FEtP4OH into the mouse brain was also deter-
mined in the mouse brain.17 These findings suggested

2524
M.-R. Zhang et al. / Bioorg. Med. Chem. 11 (2003) 2519–2527
Table 2. Enzymatic hydrolysis rate and% AchE specificity of
18F]FEtP4A and [11C]MP4A in rat brain homogenate
25 ꢀC for 1 h. After quenching CHCl3 with this mixture,
the organic layer was washed with water, 27% ammonia
water and saturated NaCl solution. After drying the
organic layer with Na2SO4, the solvent was removed to
afford a yellowish oil.
[
Hydrolysis ratea
(min/g/mL)
Specificity
for AchEb (%)
[
[
18F]FEtP4A
11C]MP4A
1.345
0.094
95.8
71.3
This oil was dissolved in a solution of pyridine (3 mL)
and acetic anhydride (1 mL, 10.6 mmol), and the reac-
tion mixture was stirred at 25 ꢀC for 6 h. After quench-
ing the reaction mixture with AcOEt, the organic layer
was washed with water, 27% ammonia water and satu-
rated NaCl solution. Drying the AcOEt layer over
Na2SO4, and removing the solvent gave a colorless
reactant. The reactant dissolved in CH2Cl2 (20 mL) was
treated with trifluoroacetic acid (1.5 mL) at 0 ꢀC, con-
tinuously stirred for 1 h at 0 ꢀC and evaporated to dry-
ness. The residue was dissolved in 1% ammonia water
and extracted with CH2Cl2. After drying the organic
layer with Na2SO4, the solvent was removed to give a
crude oil. Column chromatograph of the crude product
on silica gel and elution with CHCl3/MeOH/27%
ammonia water (9/1/0.01) provided P4A (590 mg, 57%)
as a slightly yellowish crystal. 1H NMR (CDCl3) d:
4.71–4.76 (1H, m), 2.27–2.51 (4H, m), 1.98 (3H, s),
1.58–1.88 (4H, m). IR (Nujol): 1720cmꢁ1. FAB-MS (m/z)
calcd for C9H16FNO2 (M++1): 144; found: 144.
aFirst-order hydrolysis rate constant (fraction/min) for rat brain
homogenate concentration of 1 g/mL, n=3).
bBW284c51 (30 mM) was used as an inhibitor of AchE.
for AchE may reduce the accuracy of data if this tracer
was used to measure quantitatively the AchE activity in
the brain.
Conclusion
[18F]FEtP4A was synthesized by the respective reaction
of P4A with [18F]FEtBr, [18F]FEtBr/NaI and
[18F]FEtOTf. The order of the [18F]fluoroethylation
efficiency
was
[18F]FEtOTf
>[18F]FEtBr/NaI
>[18F]FEtBr. The ex vivo autoradiography and PET
imaging study demonstrated that [18F]FEtP4A was a
potential PET tracer for AchE in the rat and monkey.
However, the lower hydrolytic rate and lower specificity
for AchE of [18F]FEtP4A may limit its usefulness for
the quantitative measurement for AchE in the primate
brain.
N-Fluoroethyl-4-piperidyl acetate (FEtP4A). A mixture
of P4A (1.0 g, 7.0 mmol), 1-bromo-2-fluoroethane
(1.1 mL, 14.8 mmol), NaI (2.2 g, 14.7 mmol) and
i-Pr2NEt (1.5 mL, 17.2 mmol) in DMF (12.0 mL) was
heated at 70 ꢀC for 5 h. After cooling the reaction mix-
ture, the mixture was quenched with AcOEt and washed
with water, and saturated NaCl solution. After the
organic layer was dried over Na2SO4, the solvent was
removed to give a residue. Column chromatography of
the residue on silica gel with CHCl3 gave FEtP4A
Experimental
General
1H NMR spectra were recorded on a JNM-GX-270
spectrometer (JEOL, Tokyo) with tetramethylsilane as
an internal standard. All chemical shifts (d) were repor-
ted in parts per million (ppm) downfield from the stan-
dard. FAB-MS was obtained on a JEOL NMS-SX102
spectrometer (JEOL, Tokyo). Column chromatography
was performed using Merck Kieselgel gel 60 F254 (70–
230 mesh). 18-Fluorine (18F) was produced by the 18O
(p, n) 18F nuclear reaction using a CYPRIS HM-18
cyclotron (Sumitomo Heavy Industry, Tokyo). If not
otherwise stated, radioactivity was determined with an
IGC-3R Curiemeter (Aloka, Tokyo). HPLC was per-
formed using a JASCO HPLC system (JASCO, Tokyo):
effluent radioactivity was monitored using a NaI (Tl)
scintillation detector system. All chemical reagents with
the highest grade commercially available were pur-
chased from Aldrich Chem. (Milwaukee, WI) and
Wako Pure Chem. Ind. (Osaka). The animal experi-
ments were carried out according to the recommenda-
tions of the committee for the care and use of
laboratory animals, National Institute of Radiological
Sciences (NIRS).
1
(990 mg, 74.8%) as a colorless oil. H NMR (CDCl3)
d: 4.71 (1H, m), 4.49 (2H, dt, JH,F=40.1 Hz,
JH,H=4.8 Hz), 2.70 (2H, m), 2.58 (2H, dt,
JH,F=27.6 Hz, JH,H=4.8 Hz), 2.28 (2H, m), 1.98 (3H,
s), 1.58–1.88 (4H, m). IR (neat): 1730 cmꢁ1. FAB-MS
(m/z) calcd for C9H16FNO2(M++1): 190; found: 190.
Anal. calcd for C9H16FNO2: C 57.14, H, 8.47, N 7.41;
found: C 57.31, H 8.40, N 7.53.
N-[18F]Fluoroethyl-4-piperidinyl acetate([18F]FEtP4A).
(1) Production of [18F]F-. [18F]F- was produced by the
18O (p,n) 18F reaction on 10–20 atom% H128O using
18 MeV protons (14.2 MeV on target) from the cyclo-
tron and separated from [18O]H2O using Dowex 1-X8
anion exchange resin in an irradiating room. The [18F]F-
was eluted from the resin with aqueous potassium car-
bonate (3.3 mg/0.3 mL) into a glass vial containing
CH3CN (1.5 mL)/Kryptofix 2.2.2. (25 mg) and trans-
ferred into a reaction vessel in a hot cell.
4-Piperidyl acetate (P4A). 4-Hydroxypiperidine hydro-
chloride (1.0 g, 7.3 mmol) was dissolved in distilled
water (2 mL). To this solution was added 0.5 N NaOH
solution (5 mL) and then di-t-butyl dicarbonate (1.8 g,
8.2 mmol) was added dropwise with vigorous stirring at
(2) Radiosynthesis of [18F]FEtP4A via [18F]FEtBr.
After the [18F]F- was dried to remove H2O and CH3CN
from the reaction vessel, 2-bromoethyl triflate31
(BrCH2CH2OTf) (8 mL) in o-dichlorobenzene (500 mL)
was added into the radioactive solution. Then, the

M.-R. Zhang et al. / Bioorg. Med. Chem. 11 (2003) 2519–2527
2525
[18F]FEtBr that resulted in this vessel was distilled under
a helium flow (90–100 mL/min) at 130 ꢀC for 2 min and
bubbled into another vessel containing a solution of
anhydrous DMF (300 mL) and P4A (0.8 mg) at ꢁ20 ꢀC.
After maximum radioactivity was reached into the
solution, the reaction mixture was warmed to 130 ꢀC
and maintained for 30 min. The content was diluted
with an HPLC mobile phase (500 mL) and applied onto
a semi-preparative column (10 mm IDÂ250 mm, Mega-
pak SIL C18) attached to the JASCO HPLC system.
Elution with 5 mM CH3COONH4 (pH=4.9)/CH3CN
(9/1) at a flow rate of 4 mL/min gave a radioactive
fraction corresponding to pure [18F]FEtP4A (tR=9.8–
11.2 min). The fraction was collected in a rotary eva-
porator and evaporated to dryness at 90 ꢀC under
reduced pressure. The residue was re-dissolved in 10 mL
of saline and passed through a Millipore filter (GS,
0.22 mm) to provide an injectable solution of
[18F]FEtP4A for animal experiments. All the above
procedures were carried out automatically by the use of
a newly developed system.23,24 Total synthesis time was
73Æ5 min (n=6) from the end of bombardment. At the
end of the synthesis, 210Æ90 MBq (n=6) of
[18F]FEtP4A were obtained after 10–15 min proton
bombardment at a beam current of 15 mA.
(Rf=0.68) or FEtP4OH (Rf=0.42) which was visualized
by heating the TLC plate sprayed with ninhydrin solu-
tion.
Radiochemical purity determinations. Radiochemical
purity of [18F]FEtP4A or [18F]FEtP4OH was assayed by
analytical HPLC (column: Finepak Sil C18-10, JASCO,
4.6 mm ID Â 250 mm). The mobile phase was 50 mM
CH3COONH4
(pH=3.1)/CH3CN
ð3=7Þ
for
[18F]FEtP4A and 50 mM CH3COONH4 (pH=3.1)/
CH3CN ð1=9Þ for [18F]FEtP4OH, respectively. The tR
was 6.8 min for [18F]FEtP4A and 5.6 min for [18F]FEt-
P4OH at a flow rate of 2 mL/min, respectively.
Specific activity determination. Since FEtP4A displays
no special absorption on a UV spectrum (200–400 nm),
the specific activity of [18F]FEtP4A was estimated from
that of [18F]FEtBr in the other synthetic run. The spe-
cific activity of [18F]FEtBr was determined by compar-
ison of UV absorbance of [18F]FEtBr at 210 nm with
those of known concentrations of FEtBr, which was
obtained by HPLC analysis with CAPCELL PAK C18
column (4.6 mm ID Â 250 mm). The tR for [18F]FEtBr
was 4.5 min at a flow rate of 2.0 mL/min using a mobile
phase of CH3CN/H2O (30/70).
Determination of P4A in the final product solution with
LC-MS. After the radioactivity decayed, the final pro-
duct solution of [18F]FEtP4A (1 mL) was concentrated
to 50 mL for analysis. The LC-MS system consisted of a
Hewelett-Packard HPLC system (SERIES 1100, Hewe-
lett-Packard Japan, Tokyo) and a Finepak Sil C18-10
column (10 mm, 4.6 mm ID Â 250 mm; JASCO, Tokyo)
eluted with 5 mM CH3COONH4 (pH=4.9)/CH3CN
ð8=2Þ at 0.75 mL/min. The effluent (5 mL) from the col-
umn was led into a Frit FAB ion source of the mass
spectrometer (JEOL JMS-SX102A spectrometer; JEOL,
Tokyo). The mass of P4A in the final solution was esti-
mated by comparison of LC-MS detector response to
m/z of P4A with the injected authentic standard.
(3) Radiosynthesis of [18F]FEtP4A via [18F]FEtBr/NaI.
The [18F]FEtBr was trapped into a solution of anhy-
drous DMF (300 mL), NaI (1 mg) and P4A (0.8 mg) at
ꢁ20 ꢀC. After the reaction was finished, the radioactive
mixture was treated as described above.
(4) Radiosynthesis of [18F]FEtP4A via [18F]FetOTf.
The [18F]FEtBr was passed through a heated steel col-
umn (3 mm IDÂ40 mm, oven temperature: 300 ꢀC) con-
taining AgOTf (100 mg) impregnated graphite carbon
(300 mg) under helium gas flow (50 mL/min).21,22 After
trapping [18F]FEtOTf into a solution of anhydrous
DMF (300 mL) and P4A (0.8 mg) at 25 ꢀC for 10 min,
the reaction mixture was treated as described above.
Ex vivo autoradiography study in rat. Male ddy rats
(n=3) were sacrificed 30 min after ivinjection of
[18F]FEtP4A (20 MBq). The brain tissue was removed
and frozen in a freezer at ꢁ80 ꢀC at 3 min, and the fro-
zen tissue was cut into 20 mm thick horizontal sections.
Each section was thawed on a cover glass, exposed to
HCl vapor, dried on a hot plate for 10 s and placed in
contact with a phosphor imaging plate (BAS-SR 127,
Fujiphoto Film) for 60 min. The radioactivity distribu-
tion on the plate was analyzed using a FUJI BAS 3000
bioimaging analyzer (Fujiphoto Film). Regions of
interest (ROIs) on the sections were placed on the
striatum, thalamus, frontal cortex and cerebellum. The
regional uptake of 18F was expressed as photo-stimu-
lated-luminescence (PSL) values per square millimeter
of the brain section (PSL/mm2).
N-[18F]Fluoroethyl-4-piperidinol ([18F]FEtP4OH).
A
mixture of [18F]FEtP4A (37 MBq) in 0.1 N aqueous
NaOH solution (0.5 mL) was heated at 95 ꢀC for 5 min.
After adjusting the pH of the reaction solution to 6–7,
the reaction solution containing [18F]FEtP4OH with
96% radiochemical purity was used for identification
and analysis of metabolites.
Identity confirmation. A solution of [18F]FEtP4A or
[18F]FEtP4OH (2–3 mL) containing non-radioactive
FEtP4A or FEtP4OH (1 mg/5 mL CH3CN) was spotted
on a silica gel TLC plate, which was then developed
with CHCl3/CH3OH/27% ammonia water (4/1/0.01) as
a mobile phase. The TLC plate was air-dried and placed
in contact with a phosphor imaging plate (BAS-SR 127,
Fujiphoto Film, Tokyo) for 10 min, and the radio-
activity distribution on the plate was analyzed using a
FUJI BAS 3000 bioimaging analyzer (Fujiphoto Film).
[18F]FEtP4A or [18F]FEtP4OH was identified by com-
paring the Rf with the non-radioactive FEtP4A
PET study in monkey. PET scan was performed using a
high-resolution SHR-7700 PET camera (Hamamatsu
Photonics, Hamamatsu, Japan) designed for laboratory
animals, which provides with 31 transaxial slices 3.6 mm
(center-to-center) apart and a 33.1 cm field of view. A

2526
M.-R. Zhang et al. / Bioorg. Med. Chem. 11 (2003) 2519–2527
male rhesus monkey (macaca mulatta) weighing about
5 kg was repeatedly anesthetized with ketamine (Keta-
lar1, 10 mg/kg/h, im) every h through the session. After
transmission scans for attenuation correction were per-
formed for 1 h using a 74 MBq 68Ge-68Ga source, a
dynamic emission scan in 3-D acquisition mode was
performed for 90 min (2 min  5 scans, 4 min  10
scans, 10 min  4 scans). All emission scan images were
reconstructed with a 4-mm Colsher filter, and circular
regions of interest (ROIs) with a 5-mm diameter were
placed over the striatum, frontal cortex, thalamus and
cerebellum using image analysis software.27 A solution
of [18F]FEtP4A (88 MBq) was injected ivinto the mon-
key, and time-sequential tomographic scanning was
performed on a transverse section of the brain for
120 min. The time–activity curves (TAC) in ROIs were
obtained for each scan of the brain.
blood samples (1 mL) were collected at periods of 2, 10,
30, 60, 90 and 120 min. All samples were centrifuged at
15,000 rpm for 1 min at 4 ꢀC to separate plasma, which
(250 mL) was collected in a test tube containing CH3CN
(0.5 mL). The tube was vortexed for 15 s and centrifuged
at 15,000 rpm for 1 min for deproteinization. The
extraction efficiency of radioactivity into the CH3CN
ranged from 78 to 94% of the total radioactivity in the
plasma. The radioactive fractions in these plasma sam-
ples were determined using HPLC.
Characterization of enzymatic property usingrat brain
homogenate in vitro. Brain tissues from male ddy rats
(n=3) were combined, weighed and homogenized in
0.1 M phosphate buffer (pH 7.4) with a glass-teflon
homogenizer, and the homogenate was diluted to the
final concentration of 35 mg/mL. Two hundreds mL of
the homogenate and 100 mL of either the buffer alone or
1,5-bis(4-allydimethylammoniumphenyl)pentane-3-one
dibromide (BW284c51, a specific inhibitor of AchE,
120 mM) solution were added to a reaction tube and
preincubated for 5 min at 25 ꢀC. The reaction was initi-
ated by adding [18F]FEtP4A (5 MBq) or [11C]MP4A
(20 MBq) in 100 mL of phosphate buffer. At designated
time intervals, 200 mL of ethanol was added to each tube
to stop the reaction. Three mL of the solution was then
applied to a silical gel TLC plate. The TLC plates were
developed and the radioactivities corresponding
[18F]FEtP4A, [11C]MP4A and their metabolites in the
TLC plates were determined according to the proce-
dures as shown in the item of identity confirmation. The
enzymatic hydrolysis rate of [18F]FEtP4A or
[11C]MP4A was calculated as: K ¼ ꢁLnðA2=A2Þ=
ðT2 ꢁ T1Þ=C, where A1 and A2 represent the radio-
activity of [18F]FEtP4A or [11C]MP4A at times T1 and
T2, respectively, and C represents the concentration (g/
mL) of the brain tissue in the reaction mixture. The
AchE specificity of [18F]FEtP4A or [11C]MP4A in rat
brain homogenate was calculate as: specificity
ð%Þ ¼ 100 Â ðK ꢁ K1Þ=K, where K1 represents the rate
constant in the brain homogenate added with BW284c51.
Metabolite analysis in rat plasma and brain homogenate.
After ivadministration of [ 18F]FEtP4A (20 MBq/
200 mL) into rats (ddy, n=3), these animals were sacri-
ficed by cervical dislocation at 30 min postinjection.
Blood (0.7–1.0 mL) and whole brains were respectively
removed from these rats quickly. The blood sample was
centrifuged at 15,000 rpm for 1 min at 4 ꢀC to separate
plasma, which (250 mL) was collected in a test tube
containing CH3CN (500 mL) and a solution of non-
radioactive FEtP4A and FEtP4OH (10 mL, 0.5 mg/mL
CH3CN). After the tube was vortexed for 15 s and cen-
trifuged at 15,000 rpm for 2 min, the CH3CN super-
natant was collected. The extraction efficiency of
radioactivity into the supernatant ranged from 71 to
89% of the total radioactivity in the plasma. On the
other hand, the cerebral cortex, striatum, thalamus and
cerebellum were dissected from the rat brain and
homogenized together in an ice-cooled CH3OH/CHCl3
(1=1; 1.5 mL) solution containing non-radioactive
FEtP4A and FEtP4OH (0.5 mg/mL CH3CN). The
homogenate was centrifuged at 15,000 rpm for 1 min
and the supernatant was collected. The recovery of
radioactivity into the CH3CN supernatant was 78–87%
based on the total radioactivity in the brain homo-
genate.
An aliquot of the supernatant (100–500 mL) of plasma
or brain homogenate was injected into the HPLC with a
highly sensitive positron detector for radioactivity. The
radioactivity in the fraction was analyzed using a
Megapak SIL C18 column (10 mm IDÂ250 mm) and an
eluent of 5 mM CH3COONH4 (pH=4.9)/CH3CN 9=1
at a flow rate of 4 mL/min. The tR for the unchanged
[18F]FEtP4A was 15.9 min, whereas that for [18F]FEt-
P4OH or an unknown product was 6.2 or 1.8 min. The
percent ratio of each radioactive fraction to the total
radioactivity on the HPLC chromatogram was calcu-
lated as %=(peak area for the fraction /total peak area)
Â100. Moreover, these fractions were respectively col-
lected and identified using the radio-TLC described
Acknowledgements
The authors thank the staff of the Cyclotron Section
and Radiopharmaceuticals Section of the National
Institute of Radiological Sciences (NIRS) for their sup-
port in the operation of the cyclotron and the produc-
tion of radioisotopes.
References and Notes
1. Davis, P.; Maloney, A. Lancet ii 1976, 1403.
2. Perry, E.; Perry, R.; Gary, B.; Bernard, E. Lancet i 1977,
189.
above: Rf=0.68;
Rf=0.42; the unknown product, Rf <0.1.
[18F]FEtP4A,
[18F]FEtP4OH,
3. Bierer, L.; Haroutunian, V.; Gabriel, S.; Knott, P.; Carlin,
L.; Purohit, D.; Perl, D.; Schmeidler, J.; Kanof, P.; Davis, K.
J. Neurochem. 1995, 64, 749.
Metabolite analysis in monkey plasma. After ivinjection
of [18F]FEtP4A (50 MBq) into the monkey, arterial

M.-R. Zhang et al. / Bioorg. Med. Chem. 11 (2003) 2519–2527
2527
4. Whitehouse, P.; Price, D.; Clark, A.; Coyle, J.; DeLong, M.
Ann. Neuro. 1981, 10, 122.
Suhara, T.; Sudo, Y.; Suzuki, K.; Irie, T. Eur. J. Nucl. Med.
1999, 28, 135.
5. Iyo, M.; Namba, H.; Fukushi, K.; Shinotoh, H.; Nakat-
suka, S.; Suhara, T.; Sudo, Y.; Suzuki, K.; Irie, T. Lancet
1997, 349, 1805.
6. Kilbourn, M.; Sherman, P.; Snyder, S. Nucl. Med. Biol.
2000, 26, 543.
16. Namba, H.; Irie, T.; Fukushi, K.; Iyo, M. Brain Res. 1994,
667, 278.
17. Zhang, M.-R.; Tsuchiyama, A.; Haradahira, T.; Yoshida,
Y.; Irie, T.; Suzuki, K. Nucl. Med. Biol. 2002, 29, 463.
18. Krupka, M. Biochem. 1966, 5, 1988.
7. Koeppe, R.; Frey, K.; Synder, S.; Meyer, P.; Kilbourn, M.;
Kuhl, D. J. Cereb. Blood. Flow 1999, 19, 1150.
8. Namba, H.; Iyo, M.; Shinotoh, H.; Nakatsuka, S.;
Fukushi, K.; Irie, T. Lancet 1998, 351, 881.
19. Deves, R.; Krupka, R. Biochim. Biophys. Acta 1979, 557,
469.
20. Clary, G.; Tsai, C.; Guynn, R. Arch. Biochem. Biophys.
1987, 254, 214.
9. Bencherif, B.; Musachio, J.; Villalobos, A.; Hilton, J.;
Ravert, H.; Mathews, W.; Dannals, R.; Williams, S.; Frost, J.
J. Nucl. Med. 1999, 40, 272.
10. Musachio, J.; Flesher, J.; Scheffel, U.; Rauseo, P.; Hilton,
J.; Mathews, W.; Ravert, H.; Dannals, R.; Frost, J. Nucl.
Med. Biol. 2002, 29, 547.
21. Mulholland, G.; Mock, B.; Zheng, Q. H.; Vavrck, M. J.
Labelled Compd. Radiopharm. 1999, 42, s318.
22. Jewett, D. Appl. Radiat. Isot. 1992, 43, 1383.
23. Zhang, M.-R.; Tsuchiyama, A.; Haradahira, T.; Yoshida,
Y.; Irie, T.; Suzuki, K. J. Labelled Compd. Radiopharm. 2001,
44 (suppl. 1), S883.
11. Irie, T.; Fukushi, K.; Akimoto, Y.; Tamagami, H.;
Nozaki, T. Nucl. Med. Biol. 1994, 21, 801.
12. Kilbourn, M.; Snyder, S.; Sherman, P.; Kuhl, D. Synapse
1996, 22, 123.
13. Irie, T.; Fukushi, K.; Namba, H.; Iyo, M.; Tamagami, H.;
Nakatsuka, S.;Akimoto, Y.; Ikota, N. J. Nucl. Med. 1996, 37, 649.
14. Shinotoh, H.; Namba, H.; Yamaguchi, M.; Fukushi, K.;
Nagatsuka, S.; Iyo, M.; Hattori, T.; Tannada, S.; Irie, T. Ann.
Neurol. 1999, 46, 62.
24. Zhang, M.-R.; Tsuchiyama, A.; Haradahira, T.; Yoshida,
Y.; Furutsuka, T.; Suzuki, K. Appl. Radiat. Isot. 2002, 57, 335.
25. Zhang, M.-R.; Haradahira, T.; Maeda, J.; Okauchi, T.;
Kawabe, K.; Kida, T.; Obayashi, S.; Suzuki, K.; Suhara, T.
Nucl. Med. Biol. 2002, 29, 469.
26. Ding, Y.-S.; Fowler, J.; Dewey, S.; Wolf, A.; Logan, J.;
Gatley, J.; Volkow, N.; Shea, C.; Taylor, D. J. Nucl. Med.
1993, 34, 246.
27. Okauchi, T.; Suhara, T.; Maeda, J.; Kawabe, K.; Obaya-
shi, S.; Suzuki, K. Synapse 2001, 41, 87.
15. Namba, H.; Iyo, M.; Shinotoh, H.; Nagatsuka, S.;