Synthesis, radiofluorination and pharmacological evaluation of a fluoromethyl spirocyclic PET tracer for central σ1 receptors and comparison with fluoroalkyl homologs

Article abstract of DOI:10.1016/j.bmc.2011.11.002

The spirocyclic σ₁ receptor ligand 1 (1′-benzyl-3- (fluoromethyl)-3H-spiro[[2]benzofuran-1,4′-piperidine]) was prepared in four steps starting from methoxy derivative 5. Due to its high σ₁ affinity (K_i = 0.74 nM) and selec

Full text of DOI:10.1016/j.bmc.2011.11.002

Bioorganic & Medicinal Chemistry 20 (2012) 257–269
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Synthesis, radiofluorination and pharmacological evaluation of a fluoromethyl
spirocyclic PET tracer for central r1 receptors and comparison with fluoroalkyl
homologs
Aurélie Maisonial b,1, Eva Große Maestrup a,1, Christian Wiese a, Achim Hiller b, Dirk Schepmann a,
Steffen Fischer b, Winnie Deuther-Conrad b, Jörg Steinbach b, Peter Brust b, Bernhard Wünsch a,
⇑
a Institut für Pharmazeutische und Medizinische Chemie der Universität Münster, Hittorfstraße 58-62, D-48149 Münster, Germany
b Helmholtz-Zentrum Dresden-Rossendorf, Research Site Leipzig, Institute of Radiopharmacy, Division of Neuroradiopharmaceuticals, Permoserstraße-15, D-04318 Leipzig, Germany
a r t i c l e i n f o
a b s t r a c t
The spirocyclic r
1 receptor ligand 1 (10-benzyl-3-(fluoromethyl)-3H-spiro[[2]benzofuran-1,40-piperidine])
Article history:
Received 20 September 2011
Revised 31 October 2011
Accepted 2 November 2011
Available online 9 November 2011
was prepared in four steps starting from methoxy derivative 5. Due to its high r1 affinity (Ki = 0.74 nM) and
selectivity against several other relevant targets, 1 was investigated as 18F-labeled PET tracer and its biolog-
ical properties were compared with those of homologous fluoroalkyl derivatives 2–4. The fluoromethyl
derivative 1 was faster metabolized in vitro than homologs 2–4. In contrast to the radiosynthesis of
[
18F]2–4, the nucleophilic substitution of the tosylate 15 using the K[18F]F–K222–carbonate complex
Keywords:
required heating to 150 °C in DMSO to achieve high labeling efficiencies. Whereas radiometabolites of
18F]2–4 were not detected in vivo in the brain of mice, two radiometabolites of [18F]1 were found. Analysis
of ex vivo autoradiography images provided rather low target-to-nontarget ratio for [18F]1 compared with
18F]2–4. [18F]1 showed a fast uptake in the brain, which decreased continuously over time. The brain-to-
plasma ratio of the radiotracer [18F]1 was only exceeded by the fluoroethyl tracer [18F]2.
Ó 2011 Elsevier Ltd. All rights reserved.
r1 Receptor radioligands
Spirocyclic piperidines
Fluorine-18
Neuroimaging
Structure–affinity relationship
Fluoroalkyl homologs
[
[
1. Introduction
occurring hallucinogen N,N-dimethyltryptamine (DMT) is an
endogenous ligand for r1 receptors. Unlike the classical plasma-
membrane bound neurotransmitter receptors the r1 receptor re-
sides specifically at the endoplasmic reticulum (ER)-mitochondrion
interface called the mitochondrion-associated ER membrane
(MAM).9 It is a ligand-activated molecular chaperone that assists
the folding or unfolding and the assembly or disassembly of other
macromolecular structures and is proposed to act as inter-organelle
signaling modulator. The r1 receptor ligands, including DMT, at
concentrations close to their Ki values, are suggested to cause the
dissociation of r1 receptors from another ER chaperone, binding
The sigma (
r) receptor was first discovered about 30 years ago
as a new subtype of opioid receptors.1 At present it is known that
r
receptors possess specific drug selectivity patterns, differential
anatomical distribution, and unique properties which are different
from opioid and other known neurotransmitter and hormone
receptor families. Pharmacological data based on binding studies,
anatomical distribution, and biochemical features distinguish at
least two
r receptor subtypes (r1 and r2) which have recently
been characterized.2–6 The human r1 receptor is a unique protein
consisting of 223 amino acids, which was first cloned and function-
ally expressed by Kekuda et al.7
immunoglobulin protein (BiP), allowing
r1 receptors to chaperone
inositol 1,4,5-trisphosphate receptors (IP3Rs) at the MAM. Higher
concentrations of DMT are assumed to cause the translocation of
r1 receptors from the MAM to the plasma membrane, leading to
the inhibition of voltage-gated ion channels.9 Thus, r1 receptor
The
r1 receptor has long been considered to be an orphan recep-
tor. Recently, Fontanilla et al.8 have demonstrated that the naturally
ligands might shift the site of action of
the center of the cell to its periphery. In good agreement with this
model is a recent finding obtained using 1 receptor knockout mice
which identified the 1 receptor as a part of mechanisms modulat-
ing activity-induced sensitization in pain pathways.10 This and other
related findings provide evidence to consider selective 1 receptor
r1 receptor chaperones from
Abbreviations: DMT, N,N-dimetyltryptamine; MAM, mitochondrion-associated
ER membrane; ER, endoplasmatic reticulum; BiP, binding immunoglobulin protein;
IP3R, inositol 1,4,5-triphosphate receptor; PET, positron emission tomography;
EMP, emopamil-binding protein; VAChT, vesicular acetylcholine transporter; p.i.,
post injection; ID, injected dose.
r
r
r
⇑
Corresponding author. Tel.: +49 251 8333311; fax: +49 251 8332144.
E-mail address: wuensch@uni-muenster.de (B. Wünsch).
antagonists an innovative and alternative approach for treating
neuropathic pain.11
1
These authors contributed equally to this work.
0968-0896/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2011.11.002

258
A. Maisonial et al. / Bioorg. Med. Chem. 20 (2012) 257–269
fluvoxamine has also been reported to dramatically decrease delir-
F
ium rating scale scores of patients in intensive care units19 and it
has been hypothesized that selective serotonin reuptake inhibitors
such as fluvoxamine may reduce the risk of transition of people with
O
N
prodromal symptoms to schizophrenia.20 Accordingly,
r1 receptors
are considered as novel therapeutic targets for the pharmacotherapy
of depression, anxiety and various other brain diseases.21–25 Related
Bn
F
to this development it has been reviewed that r1 receptor imaging is
1
useful for studying the pathophysiology of neurological and psychi-
atric disorders and for evaluation of the pharmacodynamics of psy-
chiatric drugs.26
The only PET radioligand that was used for neuroimaging of r1
receptors in humans so far is [11C]SA4503. Studies were performed
in patients with Parkinson’s disease27 and Alzheimer’s disease.28
Recently some methodical aspects were in the main focus of the
human studies. Thus, combination with blood flow measurement,
a shortened protocol for the receptor quantification and improve-
ment in parametric imaging were proposed.29–31 Notably,
F
O
N
F
O
N
Bn
O
2
Bn
4
N
11C]SA4503 has high affinity not only to r1 receptors (Ki values
Bn
[
in the range of 4–14 nM32–34) but also to other targets such as
the vesicular acetylcholine transporter (VAChT, Ki = 50 nM34) and
the emopamil binding protein (EBP, Ki = 1.7 nM35) which may be
of high impact for the interpretation of neuroimaging findings.
A few years ago, we focused our research on the development of
3
Figure 1. Spirocyclic r1 ligands with different (fluoroalkyl) residues in 3-position.
Related to the concept described above is also the recent discov-
ery of heteromers consisting of r1 receptors and dopamine D1
receptors which provides a molecular explanation for D1 receptor-
mediated behavioral effects of cocaine.12 Beside drug addiction
there is strong evidence that r1 receptors play a major role in the
pathophysiology of depression. Behavioral models revealed that
ligands which bind to r1 receptors possess antidepressant-like
properties with fast onset of action.13,14 This is further supported
by a recent gene-based association analysis between r1 receptor
gene expression and major depressive disorder in the Japanese pop-
ulation,15 the development of a depressive-like phenotype in r1
receptor knockout mice16 and the consideration of r1 receptor-
mediated antidepressive effects of fluvoxamine.17,18 Interestingly,
alternative fluorinated
noninvasive in vivo PET imaging of central
r
1 receptor ligands, which could be used for
1 receptors. These stud-
r
ies led to an innovative compound class of spirocyclic piperidines. In
particular, ligands with a spiro[benzofuran-piperidine] scaffold met
two important requirements for a radiotracer by combining very
high affinity with suitable selectivity to r
1 receptors.36–43 Therefore,
this scaffold was used as a very promising starting point for further
structural modifications to develop a fluorinated PET tracer for
imaging of
r1 receptors in the central nervous system.
We have synthesized three new spirocyclic piperidines 2–4
(Fig. 1) bearing fluoroalkyl residues with different chain lengths
(n = 2–4) in position 3.43–47 These fluoroalkyl derivatives 2–4
OH
OCH3
O
CN
O
CO2Et
O
O
N
(a)
(b)
(c)
N
N
N
Bn
Bn
Bn
Bn
5
6
7
8
F
F
CH3
(d)
(e)
O
N
O
O
+
N
H
N
H
Bn
1
9
10
(f) or (g)
F
CH3
O
O
N
N
R
R
11,13
12,14
Scheme 1. Synthesis of spirocyclic
r1 receptor ligands with a (fluoromethyl) residue in 3-position. Reagents and conditions: (a) Me3SiCN, BF3ꢀOEt2, CH2Cl2, 20 min, ꢁ25 °C,
1 h, 3–5 °C, 87%. (b) H2SO4, EtOH, 19 h, reflux, 57%. (c) LiAlH4, THF, 30 min, ꢁ20 °C, 72%. (d) Et2NSF3 (DAST), CH2Cl2, 30 min, ꢁ78 °C, 19 h, rt, 70%. (e) NH4HCO2, Pd/C, CH3OH,
3 h, reflux, 92% (9+10). (f) FC6H4CH2Cl, CH3CN, K2CO3, 8 h, reflux, 23% (11), 10% (12). (g) PhCH(CH3)Br, CH3CN, K2CO3, 8 h, reflux, 8% (13), 11% (14).

A. Maisonial et al. / Bioorg. Med. Chem. 20 (2012) 257–269
259
CH2F
3-CH
F
( )n
n =
4
O
N
Bn
3
2
1
Figure 2. Comparison of characteristic 1H NMR signals of the four homologous (fluoroalkyl) r1 ligands 1–4. 1: 5.40 ppm (dt, 3-CH), 4.56 ppm (ddd, CH2F), 4.61 ppm (ddd,
CH2F), J (H–C–F) = 47.3 Hz. 2: 5.32 ppm (dd, 3-CH), 4.63 ppm (dddd, CH2F), 4.75 ppm (dddd, CH2F), J (H–C–F) = 46.9 Hz. 3: 5.24 ppm (dd, 3-CH), 4.49 ppm (ddt, CH2F),
4.53 ppm (dddd, CH2F), J (H–C–F) = 47.2 Hz. 4: 5.21 ppm (dd, 3-CH), 4.46 ppm (dt, 2 H, CH2F), J (H–C–F) = 47.3 Hz.
The acetalic methoxy group of 5 was replaced with a cyano group
upon treatment with trimethylsilyl cyanide (Me3SiCN) and boron
trifluoride diethyl ether complex (BF3ꢀOEt2). The resulting nitrile
6 was converted into the ester 7 using ethanol and concentrated
sulfuric acid.38 Primary alcohol 8 was produced by reduction of 7
with lithium aluminum hydride (LiAlH4). Reaction of the primary
alcohol 8 with diethylaminosulfur trifluoride (DAST) provided the
fluorinated derivative 1 in 70% yield.
Since important metabolic pathways of this compound class are
hydroxylation in position 4 of the N-benzyl residue and N-deben-
Scheme 2. Synthesis of precursor 15 and radiosynthesis of [18F]1. Reagents and
zylation,36,42,43,45,47 we decided to investigate the corresponding
conditions: (a) TosCl, DMAP, NEt3, CH2Cl2, ꢁ25 °C, 30 min, then rt, 21 h, 88%. (b)
p-fluorobenzyl and 1-phenylethyl derivatives 11 and 13, which
should be metabolically more stable. In order to modify the residue
at the piperidine N-atom, the N-benzyl group of 1 was removed
hydrogenolytically with ammonium formate in the presence of
Pd/C.48 Purification by flash chromatography led to 92% of a mix-
ture, containing the desired fluoromethyl derivative 9 together
with the methyl derivative 10 (ratio 9:10 = 69:31). In addition to
the N-benzyl residue, the fluorine atom was cleaved off reductively
during the very mild transfer hydrogenolysis. An analogous reduc-
tive defluorination had not been observed for homologs 2–4 bear-
ing a 2-fluoroethyl,44 3-fluoropropyl,45 or 4-fluorobutyl46 residue
in position 3, respectively. The unexpected defluorination of 1
may originate from the b-fluoroether substructure, which is unique
among these four fluoroalkyl derivatives. The formation of an oxi-
ranium ion as reactive intermediate is postulated.
K[18F]F–K222–carbonate complex, DMF, 150 °C, 20 min, 72–86%.
possess very high r1 receptor affinity (Ki = 0.59–1.4 nM, see Table
1) and high selectivity over the r2 subtype (422- to 1331-fold).
These homologs differed in polarity, penetration into the central
nervous system and biotransformation during in vivo experiments.
These results prompted the biological characterization of the
smallest member of this homologous series, the fluoromethyl
derivative 1 (Fig. 1). Herein, we report on the synthesis as well as
radiosynthesis and the biological evaluation of the new ligand 1.
Additionally, the chemical, physical and biological properties of
all the four homologous fluoroalkyl derivatives 1–4 will be com-
paratively discussed.
2. Results and discussion
2.1. Synthesis
The mixture of fluoromethyl derivative 9 and methyl compound
10 was alkylated either with p-fluorobenzyl chloride or 1-bromo-
1-phenylethane to afford corresponding tertiary amines 11/12
and 13/14, which were separated by flash chromatography,
respectively.
The synthesis of the fluoromethyl substituted piperidine 1 is de-
picted in Scheme 1 and started with the methoxy derivative 5.39

260
A. Maisonial et al. / Bioorg. Med. Chem. 20 (2012) 257–269
Table 1
r1 and r2 receptor affinities of spirocyclic piperidines with substituted alkyl residues in 3-position
X
3
O
N
R
Compd
R
X
Ki SEM [nM] (n = 3)
r1r2 Selectivity
r1
r2
1
2
3
4
8
11
12
13
C6H5CH2
C6H5CH2
C6H5CH2
C6H5CH2
F
0.74 0.34
0.59 0.20
1.4 0.26
1.2 0.46
4.5 2.5
0.14 0.08
1.1 0.31
3.6 0.63
9.2 4.8
550
785
837
489
IC50 >1
89
925
IC50 >1
824
743
1331
620
422
>220
636
841
>275
90
CH2F
(CH2)2F
(CH2)3F
OH
F
H
F
H
OTos
C6H5CH2
l
M
FC6H4CH2
FC6H4CH2
C6H5CH(CH3)
C6H5CH(CH3)
C6H5CH2
9
l
M
M
14
15
2.9 0.08
6.3 1.6
5.7 2.2
IC50 >1
78 2.3
—
l
>345
12
—
Haloperidol
(+)-Pentazocine
1H NMR signals characteristic of the four homologous fluoroal-
kyl derivatives 1–4 are compared in Figure 2. With increasing dis-
tance between the proton in position 3 of the 2-benzofuran ring
and the electronegative fluorine atom, the chemical shift of this
signal is decreasing continuously (1: d = 5.40 ppm, 4:
d = 5.21 ppm, Fig. 2). Interestingly, for the derivatives 2–4 doublets
of doublets are observed for this proton, whilst in the shortest (flu-
oromethyl) derivative 1 an H–F-coupling over three bonds with a
coupling constant of 20.4 Hz renders this signal a doublet of trip-
lets. The signals of the CH2F groups are characterized by a very
large H–F-coupling (J ca. 47 Hz) over two bonds. Whereas the
CH2F protons of the (4-fluorobutyl) derivative 4 are magnetically
equivalent (long distance to the stereogenic center) these protons
are magnetically different in the shorter homologs 1–3, which
leads to ddd (1), dddd (2), or ddt (3) fine structures of the signals.
The tosylate precursor 15 chosen as starting material for radio-
labeling was synthesized from the alcohol 8, which was converted
into the tosylate 15 upon reaction with tosyl chloride in the pres-
ence of triethylamine and 4-(dimethylamino)pyridine (see
Scheme 2).
743). Altogether, the r1 receptor affinity and r1
the four homologous fluoroalkyl derivatives 1–4 are comparable.
/
r2 selectivity of
In addition to the fluoromethyl derivative 1, the receptor
r
affinities of the alcohol 8 and the tosylate precursor 15 were re-
corded. Compared with 1 the r1 affinity of the polar alcohol 8 is
sixfold reduced. However, compound 15 showed a rather high r1
affinity (Ki = 2.9 nM), which will be exploited together with the
strong alkylating activity of 15 for the development of a
r1 ligand,
which can be covalently bound to the r1 receptor protein.
While the introduction of a fluorine atom in p-position of the
N-benzyl residue (11) led to fivefold increase of the r1 receptor
affinity, the introduction of a methyl group in
N-benzyl residue (13) resulted in a fivefold decrease of
However, the very potent p-fluorobenzyl derivative 11 revealed a
rather high 2 affinity (Ki = 89 nM). Although the r1 2 selectivity
a
-position of the
r
1 affinity.
r
/r
is still acceptable, the p-fluorobenzyl derivative 11 was not consid-
ered for further development.
Replacement of the fluorine atom with a proton resulted in a
3- to 10-fold reduced r1 receptor affinity of the methylated
compounds 12 and 14. Obviously, the fluorine atom at the
methyl residue in position 3 is favorable for high r1 receptor
affinity.
2.2. Receptor affinity
In addition to the interactions with r1 and r2 receptors, the
affinities of the fluoroalkyl derivatives 1–4 towards related recep-
Competition experiments with radioligands were used for
determining
In the r1 assay homogenates of guinea pig brains were used as
receptor material and the
1 selective ligand [3H]-(+)-pentazocine
was employed as radioligand. Rat liver homogenates served as
source for 2 receptors in the 2 assay. Since a 2 selective radio-
r
receptor affinities of the spirocyclic compounds.
tors and proteins were investigated. Since potent NMDA and
receptor ligands are often very similar, the NMDA receptor affinity
was taken into account. Originally, the receptor has been classi-
fied as opioid receptor subtype. Therefore, the affinities towards
j, and d opioid receptors were determined. The interactions of the
r
r
r
l,
r
r
r
ligand is not commercially available, the nonselective radioligand
[3H]-1,3-di(o-tolyl)guanidine was employed in the presence of an
excess of nontritiated (+)-pentazocine, which selectively masks
r1 receptors.49,50
fluoroalkyl derivatives 2–4 with the vesicular acetylcholine trans-
porter (VAChT) and the emopamil-binding protein (EBP), which
is a vertebrate sterol isomerase, were also recorded, since some po-
tent
r1 ligands (e.g., SA4503) possess considerable affinity towards
The
r
1 and
r
2 receptor affinities of the fluorinated and methyl-
these target systems.
ated spirocyclic piperidines are summarized in Table 1. The r1
receptor affinity of the fluoromethyl derivative 1 (Ki = 0.74 nM) is
The results of these receptor binding studies are given in Table 2.
At a test compound concentration of 1 lM the binding of the radi-
in the same range as the
r
1 affinity of the homologous fluoroalkyl
2 receptor affinity test performed with 1
r2 selectivity (factor
oligand [3H]MK-801 towards the phencyclidine binding site was
not reduced. However, moderate affinities of the fluoroalkyl deriv-
atives 1–4 towards the opioid receptors were observed. The rather
derivatives 2–4.44–47 The
r
(Ki = 550 nM) resulted in an excellent r1
/

A. Maisonial et al. / Bioorg. Med. Chem. 20 (2012) 257–269
261
Table 2
Receptor selectivity: affinity of the four homologous fluoroalkyl derivatives 1–4 towards related receptors and proteins
Compd
Ki [nM] (n = 3)
r1
r2
NMDA (
l
M)
l
j
d
VAChT (lM)
EBP
1
2
3
4
0.74
0.59
1.4
550
785
837
489
>1
>1
>1
>1
481
456
246
605
918
152
—
—
372
368
419
>1
>1
lM
1.4
>1
>1
211
455
233
209
lM
1.2
(CH2)n-F
O
N
Bn
1-4
n = 4
n = 3
n = 1
n = 2
Incubation time (min)
Figure 3. Comparison of the rate of biotransformation of the four fluoroalkyl derivatives 1–4. With exception of the curve of the 3-fluoropropyl compound 3 all data points
were recorded twice.
high d-opioid receptor affinity of the fluoromethyl (1, Ki = 152 nM)
and 3-fluoropropyl derivative 3 (Ki = 209 nM) are noteworthy. Nev-
2.4. Radiosynthesis
ertheless, the
r
1/d-selectivity is still high (205 for 1, 150 for 3).
The radiosynthesis of the radiotracer [18F]1 from the tosylate 15
(Scheme 2) was more difficult and differed from the syntheses of
the three other radiotracers [18F]2–4 of the fluoroalkyl series stud-
ied. Indeed, reaction of the precursor 15 with [18F]fluoride in ace-
tonitrile at 83 °C according to the reaction parameters previously
optimized for [18F]4 led to a labeling efficiency always below 16%
(Fig. 4A). This result is in sharp contrast to the data obtained with
the homologous fluoroalkyl derivatives [18F]2–4, which were pro-
duced with this method in very high labeling yields. To circumvent
this problem, other solvents such as N,N-dimethylformamide
(DMF) and dimethylsulfoxide (DMSO) allowing higher reaction
temperatures were investigated. With a constant precursor con-
centration of 2 mg/mL the best labeling efficiencies were obtained
with DMSO at 150 °C (57% after 10 min) (Fig. 4A). A few micro-
wave-assisted labeling experiments were conducted in DMF
(ꢂ75 W, up to 130 °C) which also resulted in good labeling efficien-
cies (ꢂ60%) but were less reproducible than the conventional heat-
ing protocol.
Further modifications of the concentration of the precursor re-
sulted in optimized conditions with 2.5–3 mg/mL of tosylate 15
in DMSO under conventional heating (150 °C, Fig. 4B) providing
labeling efficiencies of 65–70% after a reaction time of 10–
15 min. The crude product was then purified by isocratic semipre-
parative RP-HPLC, followed by solid phase extraction. The radio-
tracer [18F]1 was finally produced with radiochemical yields of
38–50%, radiochemical purities >99.1%, and high specific activities
Among this series of compounds, the 2-fluoroethyl derivative 2
with the highest r1 affinity shows the best selectivity against all
three opioid receptors (763 (r1/l), 630 (r1/j), >1000 (r1/d)).
The affinities of the fluoroalkyl derivatives 2–4 towards the VAChT
as well as the EBP were very low, indicating high selectivity against
these related proteins.
2.3. In vitro rate of degradation with rat liver microsomes
Since the fluoromethyl derivative 1 was promising for further
studies in vivo, its rate of degradation upon incubation with rat li-
ver microsomes was investigated. In Figure 3, the concentration of
parent compound 1 after definite time intervals is compared with
the concentration of homologous compounds 2–4.
For the investigation of the metabolic degradation six incuba-
tions were started at the same time and stopped after definite time
intervals to observe the decomposition. Directly after termination
of the metabolic process, the internal standard praziquantel was
added to quantify the remaining parent compounds by HPLC anal-
yses. A matrix calibration was carried out for all spirocyclic r1
ligands.
The rate of metabolic degradation in vitro correlates with the
length of the fluoroalkyl residue in position 3. Whereas com-
pounds 1 and 2 with short fluoroalkyl side chains were biotrans-
formed very fast, the metabolic degradation of the corresponding
homologs 3 and 4 was considerably slower. After an incubation
period of 30 min, approximately 20% of compound 1 and 13%
of compound 2 remained unchanged, but 38% and 45% of the
derivatives 3 and 4, respectively, were detected by HPLC analy-
ses. Therefore, regarding the rate of in vitro metabolic degrada-
tion longer fluoroalkyl residues seem to be favored over
shorter ones.
of 173–412 GBq/lmol (n = 6).
2.5. In vitro stability and lipophilicity
Next the in vitro stability of [18F]1 was investigated under phys-
iological conditions. The amount (in %) of the unchanged radio-
tracer after an incubation period of 2 h at 40 °C in 0.9% sodium

262
A. Maisonial et al. / Bioorg. Med. Chem. 20 (2012) 257–269
Figure 4. Labeling efficiency of [18F]1 using different solvents and temperatures (A) and different precursor concentrations (B).
Table 3
In vitro stability of 1–4 after an incubation period of 2 h in different buffer systems (n = 2)
[
18F]1 (%)
[
18F]2 (%)
[
18F]3 (%)
[
18F]4 (%)
0.9% NaCl solution, pH 7.2, 40 °C
Phosphate–saline-solution, pH 7.2, 40 °C
0.01 M TRIS–HCl, pH 7.4, 21 °C
96.9
96.5
100
97.1
98.4
97.6
>99
98.0
>98
97.9
94.7
95.4
Table 4
LogD values of the spirocyclic piperidines 1–4 (n = 3)
ACD/logD
n-Octanol/phosphate
n-Octanol/phosphate-
software pH 7.2
buffer pH 7.2
saline-solution pH 7.2
LogD [18F]1
LogD [18F]2
LogD [18F]3
LogD [18F]4
2.83
3.10
3.47
3.83
2.45 0.04
2.70 0.32
3.00 0.05
3.16 0.11
2.39 0.04
2.57 0.32
2.78 0.06
3.11 0.14
chloride solution (pH 7.2), phosphate–saline-solution (pH 7.2) or
0.01 M TRIS–HCl buffer (pH 7.4 at 21 °C) are summarized in Table 3
and compared with values previously determined for [18F]2–4.
According to radio-TLC and analytical radio-HPLC measurements,
the four radiotracers showed an excellent to sufficient stability un-
der the conditions tested. In particular, defluorination of the radio-
ligands was not observed.
after concentration were analyzed by analytical radio-HPLC and -
TLC. Generally, good agreement was achieved between these two
analytical methods. The recoveries of radioactivity from the brain
homogenates in acetonitrile extracts were moderate (75–80% at
30 min p.i. and 60–68% at 60 min p.i.).
The estimation of the lipophilicity of radiotracers developed for
neuroimaging is an important tool in predicting their blood–brain-
barrier permeability. In this study, logD values of [18F]1 were
determined by measuring the distribution of this radioligand be-
tween n-octanol and differently buffered aqueous solutions. As
shown in Table 4, experimental logD values determined by the
classical shake-flask method were comparable to the theoretical/
computed values or slightly lower. As expected, the lipophilicity
of the radioligands increased gradually with the length of the fluo-
roalkyl chain in position 3. Among this series of radiotracers, [18F]1
has the lowest logD value. However, the logD value of [18F]1 is still
in a promising range allowing the penetration of [18F]1 into the
brain during in vivo experiments.
2.6. In vivo stability in the brain
For the analysis of radiometabolites in the brain of CD-1 mice,
[
18F]1 was administered intravenously via the tail vein. At 30 and
Figure 5. Analytical RP-HPLC chromatogram of the acetonitrile extracts of brain
homogenates of CD-1 mice at 30 min p.i. of [18F]1 (Multospher 120 RP-18 AQ
60 min post injection (p.i.), brain samples were processed and two-
fold extracted with ice-cold acetonitrile. The amount (in %) of par-
ent radiotracer [18F]1 and radiometabolites in the isolated extracts
column, 250 ꢃ 4.6, 5
lm, 5% MeCN in 20 mM NH4OAcaq. over 50 min, 1 mL/min,
k = 254 nm).

A. Maisonial et al. / Bioorg. Med. Chem. 20 (2012) 257–269
263
Table 5
2.7. Ex vivo autoradiography
Amount of nonmetabolized radiotracers (in %) in the acetonitrile extracts of brain
homogenates at 30 and 60 min p.i. of [18F]1–4 in female CD-1 mice (n = 3)
In order to determine the regional distribution of [18F]1, an
ex vivo autoradiography of slices of CD-1 mouse brain were per-
formed. At 45 min after i.v. injection of [18F]1, localization of the
radioligand correlated with r1 receptor-rich regions such as the
area of the facial nucleus, the dorsolateral and lateral periaqueduc-
tal gray, the dorsomedial hypothalamic area, and the cerebellum
(Fig. 6). Moderate radiotracer accumulation was observed in tha-
lamic nuclei and the striatum and the lowest concentration of
[
18F]1
[
18F]2
[
18F]3
[
18F]4
30 min p.i.
60 min p.i.
90 1.5
79 1.0
97 1.2
98 0.5
96 1.5
98 1.4
96 1.0
95 2.7
[
18F]1 was detected in the anterior part of the olfactory bulb. Thus,
the latter region was chosen as nontarget tissue region. Analyses of
the images provided moderate target-to-nontarget tissue ratios for
[
18F]1 such as, for example, 1.34 for the facial nucleus. This value is
the lowest compared to those obtained for the derivatives [18F]2–4
(4.69, 1.97, and 2.83, respectively)44–46 at 45 min p.i.
2.8. Organ distribution in CD-1 mice
Table 6 summarizes the data obtained in biodistribution studies
in female CD-1 mice at 5, 30, 60, and 120 min after i.v. injection of
[
per gram of wet tissue (% ID/g) for all organs of interest. Interest-
ingly, a high uptake in the brain was observed at 5 min p.i.
Figure 6. Ex vivo autoradiographs of a sagittal slice of female CD-1 mouse brain at
45 min after i.v. injection of [18F]1 (24 MBq).
18F]1. Results are presented as percentage of the injected dose
(7.15 1.36% ID/g) but also in
r1 receptor expressing organs. Dur-
At 30 min p.i., 90% of the radioactivity in the brain extracts cor-
responded to the nonmetabolized radiotracer
ing the course of the study, the level of radioactivity detected in
these tissues decreased continuously.
[
18F]1 (Fig. 5,
tR = 44.4 min). However, two radiometabolites were also detected
(tR = 26.7 min, 4% and tR = 35.2 min, 6%). At 60 min p.i., nonmetab-
olized [18F]1 and the same radiometabolites were found with 79%,
10%, and 11%, respectively.
This observation is in strong contrast to the in vivo metabolism
of the three radiotracers [18F]2–4 previously investigated,44–46
where even at 60 min p.i., radiometabolites were not detected
and the amount of unchanged radioligand in the brain superna-
tants always exceeded 95% of total brain radioactivity (Table 5).
The current data indicate a slightly lower stability in vivo of the
fluoromethyl derivative [18F]1.
Generally, organ distribution of [18F]1 and its fluoroalkyl ana-
logs [18F]2–444–46 were comparable for most organs. High radioac-
tivity uptake in the bladder and urine pointed to renal elimination
as the major excretory pathway of all four radiotracers. The contin-
uous gastrointestinal accumulation of radioactivity was consistent
with an additional hepatobiliary elimination pattern.
Initially, defluorination of [18F]3 was hypothesized, which was
the first radiotracer characterized in this series.45 As illustrated in
Figure 7A, an increased uptake of radioactivity in the femur over
time was observed for all radiotracers [18F]2–444–46 except the flu-
oromethyl derivative [18F]1 with a direct correlation between
Table 6
Distribution of radioactivity in mice after i.v. injection of [18F]1 (300 kBq)
Organ
Radioactivity uptake (% ID/g wet weight)a
5 min p.i.
30 min p.i.
60 min p.i.
120 min p.i.
Blood
Plasma
Brain
Cerebellum
Heart
0.73 0.10
1.10 0.17
7.15 1.36
8.20 2.10
7.33 2.59
18.01 3.46
2.62 0.19
19.78 2.97
0.91 0.18
0.39 0.10
5.31 0.65
14.93 4.69
3.99 2.19
1.51 0.40
6.23 1.06
3.61 1.30
11.22 2.15
13.29 1.13
3.26 1.48
2.12 0.40
2.32 1.01
2.09 0.47
1.89 0.47
1.12 0.31
0.35 0.05
0.55 0.10
4.56 0.35
6.27 0.63
4.57 0.60
8.30 0.40
3.52 0.76
18.82 5.12
1.28 0.47
0.40 0.05
6.68 1.37
9.61 1.71
46.35 32.97
3.86 0.78
8.85 0.76
4.83 1.06
16.03 1.37
13.39 0.88
4.75 0.96
1.67 0.26
4.34 1.57
1.42 0.68
2.00 0.34
1.82 0.38
0.16 0.05
0.28 0.21
2.99 0.73
4.27 1.23
2.96 0.53
8.32 1.71
1.77 0.15
23.46 13.39
0.92 0.29
0.32 0.15
3.98 0.92
6.03 1.20
45.50 60.25
1.42 1.47
5.30 0.40
2.50 1.00
9.27 3.01
7.48 4.29
3.28 1.00
0.97 0.37
1.75 1.13
1.18 0.65
0.87 0.73
2.02 0.63
0.15 0.14
0.22 0.24
2.70 0.85
4.02 0.56
3.05 0.47
6.82 2.12
1.90 0.27
25.90 12.74
1.46 0.08
1.96 0.98
3.95 0.32
5.60 0.97
56.46 12.34
5.58 2.99
4.84 1.51
2.52 0.28
11.99 1.12
6.60 2.43
3.55 1.75
1.16 0.04
1.92 0.84
1.29 0.48
1.53 0.23
2.08 0.51
Lung
Stomach
Small intestine
Large intestine
Content of intestine
Liver
Kidney
Urine
Bladder
Spleen
Thymus
Pancreas
Adrenals
Gonads
Muscle
Femur
Femur (flushed)
Eye
Fat
a
Mean SD (n = 3).

264
A. Maisonial et al. / Bioorg. Med. Chem. 20 (2012) 257–269
Figure 7. [18F]1–4 uptake in femur and bone marrow of CD-1 mice. Evidence for lack of defluorination. Values are means SD (% ID/g).
Figure 8. Brain uptake in CD-1 mice and comparison of the brain-to-plasma ratios for [18F]1–4. Values are means SD (% ID/g).
lipophilicity and amount of uptake at 120 min p.i. Therefore, the
highly lipophilic bone marrow was isolated from the femur during
the biodistribution studies of [18F]2 and [18F]4, and accumulation
of radioactivity in this tissue was identified as the reason for the ob-
served increase in femur uptake. The initially assumed defluorina-
tion of [18F]3 can be excluded (Fig. 7B). In the present study the
uptake of radioactivity in the flushed femur was nearly constant
over time and lend support to limited or no defluorination of [18F]1.
As expected, the time-activity data on brain permeation of [18F]1–4
clearly showed that the early brain uptake (5 min p.i.) is inversely re-
lated to the lipophilicity of the radiotracers (Fig. 8A). However, at
30 min p.i. only [18F]2 showed an increased brain uptake compared
to 5 min p.i. This result indicates specific accumulation of this radio-
tracer which is most likely related to its highest target affinity within
the series under investigation. Accordingly, the brain-to-plasma ratio
was considerably higher for [18F]2 than for the other three 18F-labeled
spirocyclic piperidine derivatives (Fig. 8B).
of 1 with rat liver microsomes was very fast. Moreover, in vivo
two radiometabolites of [18F]1 were detected in the mouse brain.
This observation is in sharp contrast to the findings obtained with
the radiotracers [18F]2–4, which do not form radiometabolites in
the brain. Whereas [18F]1 provided the lowest target-to-nontarget
ratio in ex vivo brain autoradiography experiments, the highest ra-
tio was found for the fluoroethyl derivative [18F]2. Although the
early brain uptake (5 min after injection) was inversely related to
the lipophilicity of the four homologous radiotracers [18F]1–4,
the fluoroethyl radiotracer [18F]2 showed the highest accumulation
in the brain uptake over a longer period of time, which is presum-
ably due to its very high
r1 receptor affinity. The good penetration
of [18F]1 into the central nervous system is also reflected by the
high brain/plasma ratio, which is only exceeded by the fluoroethyl
derivative [18F]2. However, because of the presence of radiometab-
olites in the brain the fluoromethyl derivative 1 is not suitable for
neuroimaging of
r1 receptors. The fluoroethyl derivative 2 has the
best properties among the homologous radiotracers [18F]1–4 and,
therefore, will be further developed for clinical studies.
3. Conclusion
Herein are described the synthesis and biological characteriza-
tion of the fluoromethyl derivative 1 which represents the shortest
4. Experimental
4.1. Chemistry
homolog of fluoroalkyl substituted spirocyclic
r ligands 1–4. The
fluoromethyl derivative showed very high r1 affinity
1
(Ki = 0.74 nM) and selectivity against several other related targets.
Due to the branched b-position, an efficient radiosynthesis of [18F]1
was only possible upon heating in DMSO at 150 °C the tosylate 15
in the presence of the K[18F]F–K222–carbonate complex. Compared
with the fluoroalkyl homologs 2–4, the in vitro biotransformation
4.1.1. General
Unless otherwise noted, moisture sensitive reactions were con-
ducted under dry nitrogen. THF was dried with sodium/benzophe-
none and was freshly distilled before use. Thin layer
chromatography (tlc): silica gel 60 F254 plates (Merck). Flash

A. Maisonial et al. / Bioorg. Med. Chem. 20 (2012) 257–269
265
chromatography (fc): silica gel 60, 40–64
l
m (Merck); parentheses
1H, ArCHO), 7.14–7.16 (br d, J = 8.0 Hz, 1H, arom. H), 7.23–7.40
(m, 8H, arom. H). Purity (HPLC): 95.6%, tR = 17.72 min.
include: diameter of the column, eluent, fraction size, Rf value. Melt-
ing point: Melting point apparatus SMP 3 (Stuart Scientific), uncor-
rected. MS: MAT GCQ (Thermo-Finnigan); EI = electron impact;
Thermo Finnigan LCQÒ ion trap mass spectrometer with an electro-
spray ionization (ESI) interface. IR: IR spectrophotometer 480Plus
FT-ATR-IR (Jasco). 1H NMR (400 MHz), 13C NMR (100 MHz): Mer-
cury-400BB spectrometer (Varian); d in ppm related to tetramethyl-
silane; coupling constants are given with 0.5 Hz resolution. HPLC:
Merck Hitachi Equipment; UV detector: L-7400; autosampler: L-
7200; pump: L-7100; degasser: L-7614; Method 1: column: LiChro-
4.1.4. (10-Benzyl-3H-spiro[[2]benzofuran-1,40-piperidin]-3-
yl)methanol (8)38
Under N2 the ester 7 (127 mg, 0.36 mmol) was dissolved in THF
(6 mL) and the solution was cooled down to ꢁ20 °C. A solution of
LiAlH4 (1 M in THF, 0.74 mL, 0.74 mmol) was added slowly and
the mixture was stirred for 30 min at ꢁ20 °C. After addition of a
saturated solution of NaCl, it was filtered and the filtrate was ex-
tracted with CH2Cl2 (3ꢃ). The organic layer was dried (Na2SO4),
concentrated in vacuo and the residue was purified by fc (2 cm,
cyclohexane/ethyl acetate 5:5, 10 mL, Rf 0.17). Colorless oil, yield
80.5 mg (72%). C20H23NO2 (309.4). MS (EI): m/z = 309 [M], 218
~
spherÒ 60 RP-select B (5
injection volume: 5.0 L; detection at k = 210 nm; solvents: A:
lm), 250–4 mm; flow rate: 1.00 mL/min;
l
water with 0.05% (v/v) trifluoroacetic acid; B: acetonitrile with
0.05% (v/v) trifluoroacetic acid: gradient elution: (A%): 0–4 min:
90%, 4 min: 90%, 4–29 min: gradient from 90% to 0%, 29–31 min:
0%, 31–31.5 min: gradient from 0% to 90%, 31.5–40 min: 90%. The
purity of all test compounds was greater than 95%, which was deter-
mined by the given HPLC method.
[M–CH2Ph], 91 [PhCH2]. IR:
m
(cmꢁ1) = 3386 (w, O–H), 2917 (C–
H), 1050 (C–O), 754 (C–H, 1,2-disubst. arom.), 699 (C–H, monos-
ubst. arom.). 1H NMR (CDCl3): d (ppm) = 1.73 (ddd, J = 13.8/5.3/
2.7 Hz, 2H, N(CH2CH2)2), 1.96 (td, J = 13.0/4.4 Hz, 1H, N(CH2CH2)2),
2.16 (td, J = 13.2/4.2 Hz, 1H, N(CH2CH2)2), 2.49 (br t, J = 11.9 Hz, 2H,
N(CH2CH2)2), 2.82–2.90 (m, 2H, N(CH2CH2)2), 3.61 (s, 2H, NCH2Ph),
3.75 (dd, J = 11.6/5.5 Hz, 1H, CH2OH), 3.94 (dd, J = 11.6/3.3 Hz, 1H,
CH2OH), 5.29 (br t, J = 4.4 Hz, 1H, ArCHO), 7.15–7.18 (m, 2H, arom.
H), 7.26–7.38 (m, 7H, arom. H). A signal for the OH-proton is not
seen in the spectrum. Purity (HPLC): 95.6%, tR = 13.28 min.
4.1.2. 10-Benzyl-3H-spiro[[2]benzofuran-1,40-piperidine]-3-
carbonitrile (6)38
Under N2 compound 5 (504 mg, 1.64 mmol) was dissolved in
CH2Cl2 (17 mL) and the solution was cooled to ꢁ25 °C. Then, Me3-
SiCN (1.22 mL, 9.7 mmol) and BF3ꢀEt2O (0.25 mL, 1.94 mmol) were
added and the mixture was stirred for 20 min at ꢁ25 °C and for 1 h
at 3–5 °C. Methanol (2 mL) and 2 M NaOH (pH 9–10) were added,
the layers were separated and the organic layer was extracted with
CH2Cl2 (4ꢃ). The combined organic layers were dried (Na2SO4),
concentrated in vacuo and the residue was purified by fc (3 cm,
cyclohexane/ethyl acetate 7:3, 20 mL, Rf 0.22). Colorless oil, yield
432 mg (87%). C20H20N2O (304.4). MS (EI): m/z = 304 [M], 227
~
4.1.5. 10-Benzyl-3-(fluoromethyl)-3H-spiro[[2]benzofuran-1,40-
piperidine] (1, WMS-1850)
Under N2 CH2Cl2 (10 mL) was cooled to ꢁ78 °C. Then, diethyl-
aminsulfur trifluoride (DAST, 0.2 mL, 1.62 mmol) and 5 min later
a solution of alcohol 8 (224.7 mg, 0.73 mmol) in CH2Cl2 (5 mL)
were added slowly. After stirring for 30 min at ꢁ78 °C the mixture
was warmed to rt and stirred for 19 h. Then an additional amount
[M–Ph], 213 [M–CH2Ph], 91 [PhCH2]. IR:
m
(cmꢁ1) = 2944 (C–H),
of DAST (30
l
L, 0.2 mmol) was added at ꢁ40 °C and the mixture
1044 (C–O), 754 (C-H, 1,2-disubst. arom.), 699 (C–H, monosubst.
arom.). 1H NMR (CDCl3): d (ppm) = 1.71 (ddd, J = 13.8/5.1/2.6 Hz,
1H, N(CH2CH2)2), 1.92–2.11 (m, 3H, N(CH2CH2)2), 2.41 (br td,
J = 12.0/2.9 Hz, 1H, N(CH2CH2)2), 2.50 (td, J = 11.6/3.7 Hz, 1H,
N(CH2CH2)2), 2.82–2.89 (m, 2H, N(CH2CH2)2), 3.59 (s, 2H, NCH2Ph),
5.86 (s, 1H, ArCHO), 7.19–7.21 (m, 1H, arom. H), 7.25–7.44 (m, 8H,
arom. H). 13C NMR (CDCl3): d (ppm) = 37.1 (1C, N(CH2CH2)2), 37.5
(1C, N(CH2CH2)2), 49.9 (1C, N(CH2CH2)2), 50.1 (1C, N(CH2CH2)2),
63.6 (1C, NCH2Ph), 69.1 (1C, ArCHCN), 88.3 (1C, ArCO), 118.7 (1C,
CN), 121.6 (1C, arom. CH), 122.2 (1C, arom. CH), 127.3 (1C, arom.
CH), 128.5 (1C, arom. CH), 129.0 (1C, arom. CH), 129.5 (1C, arom.
CH), 130.0 (1C, arom. CH), 134.0 (1C, arom. C), 138.6 (1C, arom.
C), 145.7 (1C, arom. C). Purity (HPLC): 99.6%, tR = 16.48 min.
was stirred for 5 h at rt. Under cooling with ice 2 M NaOH was
added (pH >10). The layers were separated and the aqueous layer
was extracted with CH2Cl2 (4ꢃ). The combined organic layers were
dried (Na2SO4), the solvent was removed in vacuo and the residue
was purified by fc (2 cm, cyclohexane/ethyl acetate 7:3, 10 mL,
Rf (cyclohexane/ethyl acetate 5:5) 0.40). Pale yellow oil, yield
158.7 mg (70%). C20H22FNO (311.4). MS (EI): m/z = 311 [M], 234
~
[M–Ph], 220 [M–CH2Ph], 91 [PhCH2]. IR:
m
(cmꢁ1) = 2941 (C–H),
1604 (arom. C@C), 1052 (C–O), 756 (C–H, 1,2-disubst. arom.),
740, 698 (C–H, monosubst. arom.). 1H NMR (CDCl3):
d
(ppm) = 1.73 (ddd, J = 13.8/5.6/2.8 Hz, 2H, N(CH2CH2)2), 1.98 (td,
J = 12.7/3.5 Hz, 1H, N(CH2CH2)2), 2.07 (td, J = 12.8/3.9 Hz, 1H,
N(CH2CH2)2), 2.47 (br t, J = 10.9 Hz, 1H, N(CH2CH2)2), 2.49 (br t,
J = 10.9 Hz, 1H, N(CH2CH2)2), 2.84 (br d, J = 11.1 Hz, 2H,
N(CH2CH2)2), 3.59 (s, 2H, NCH2Ph), 4.56 (ddd, J = 47.3/9.7/4.9 Hz,
1H, CHCH2F), 4.61 (ddd, J = 47.3/9.7/3.9 Hz, 1H, CHCH2F), 5.40 (dt,
J = 20.4/4.2 Hz, 1H, ArCHO), 7.16 (br d, J = 7.2 Hz, 1H, arom. H),
7.22 (br d, J = 6,7 Hz, 1H, arom. H), 7.25–7.38 (m, 7H, arom. H).
13C NMR (CDCl3): d (ppm) = 37.7 (1C, N(CH2CH2)2), 38.4 (1C,
N(CH2CH2)2), 50.1 (1C, N(CH2CH2)2), 50.2 (1C, N(CH2CH2)2), 63.5
(1C, NCH2Ph), 80.4 (d, J = 21.1 Hz, 1C, CHCH2F), 85.2 (1C, ArCO),
85.4 (d, J = 174.9 Hz, 1C, CHCH2F), 121.3 (1C, arom. CH), 121.8
(1C, arom. CH), 127.2 (1C, arom. CH), 128.0 (1C, arom. CH), 128.4
(2C, arom. CH), 128.5 (1C, arom. CH), 129.5 (2C, arom. CH), 137.1
(d, J = 5.9 Hz, 1C, arom. C), 138.5 (1C, arom. C), 146.6 (1C, arom.
C). Purity (HPLC): 98.3%, tR = 17.35 min. Elemental analysis: calcd:
C, 77.14; H, 7.12; N, 4.50; found: C, 76.55; H, 7.07; N, 4.34.
4.1.3. Ethyl 10-benzyl-3H-spiro[[2]benzofuran-1,40-piperidine]-
3-carboxylate (7)38
The nitrile 6 (211 mg, 0.69 mmol) was dissolved in EtOH. H2SO4
concd. (ꢂ1.0 g) and three drops of H2O were added slowly. The
mixture was heated to reflux for 19 h. Then 2 M NaOH was added
under cooling with ice (pH 8–9). After addition of a saturated solu-
tion of NaCl the layers were separated, and the aqueous layer was
extracted with CH2Cl2 (4ꢃ). The combined organic layers were
dried (Na2SO4), the solvent was removed in vacuo and the residue
was purified by fc (2 cm, cyclohexane/ethyl acetate 7:3, 10 mL, Rf
(cyclohexane/ethyl acetate 5:5) 0.43). Pale yellow oil, yield
138.4 mg (57%). C22H25NO3 (351.4). MS (EI): m/z = 351 [M], 260
~
[M–CH2Ph], 91 [PhCH2]. IR:
m
(cmꢁ1) = 2943 (C–H), 1755, (C@O),
1075 (C–O), 755 (C–H, 1,2-disubst. arom.), 700 (C–H, monosubst.
arom.). 1H NMR (CDCl3): d (ppm) = 1.28 (t, J = 7.1 Hz, 3H, OCH2CH3)
1.73 (ddd, J = 13.5/5.0/2.6 Hz, 1H, N(CH2CH2)2), 1.93–2.18 (m, 3H,
N(CH2CH2)2), 2.48 (td, J = 12.3/2.7 Hz, 1H, N(CH2CH2)2), 2.58 (td,
J = 11.3/2.7 Hz, 1H, N(CH2CH2)2), 2.82–2.90 (m, 2H, N(CH2CH2)2),
3.59 (s, 2H, NCH2Ph), 4.20 (q, J = 7.1 Hz, 2H, OCH2CH3), 5.66 (s,
4.1.6. 3-(Fluoromethyl)-3H-spiro[[2]benzofuran-1,40-piperidine]
(9) and 3-methyl-3H-spiro[[2]benzofuran-1,40-piperidine] (10)
Pd/C (32 mg, 10% (m/m)) and dried ammonium formate
(140 mg, 2.2 mmol) were added to a solution of 1 (127.6 mg,
0.41 mmol) in CH3OH (10 mL). Under N2 the mixture was heated

266
A. Maisonial et al. / Bioorg. Med. Chem. 20 (2012) 257–269
to reflux for 3 h. It was filtered over CeliteÒ, the filtrate was con-
centrated in vacuo and the reside was purified by fc (0.5 cm, ethyl
acetate/MeOH/NH3 9:1:0.2, 5 mL, Rf 0.06). Pale yellow oil, yield
83.1 mg (92%). According to the 1H NMR spectrum and the HPLC
analysis, the isolated product contained 9 and 10 in the ratio
69:31. The products were separated only after N-alkylation. 9:
(0.01%), 5 mL). In addition to the pure samples 28 mg (ca. 59%) of
the mixture was isolated.
13 (Rf (cyclohexane/ethyl acetate 7:3+NH3 (0.01%)) 0.29): color-
less oil, yield 4.2 mg (7.6%). C21H24FNO (325.4). MS (EI): m/z = 325
~
[M], 310 [M–CH3], 248 [M–Ph], 91 [CH2Ph]. IR:
m
(cmꢁ1) = 2938 (C–
H), 1601 (arom. C@C), 1052 (C–O), 756 (C–H, 1,2-disubst. arom.),
699 (C–H, monosubst. arom.). 1H NMR (CDCl3): d (ppm) = 1.42 (d,
J = 6.3 Hz, 3H, NCHArCH3), 1.66 (ddd, J = 13.5/4.7/2.5 Hz, 1H,
N(CH2CH2)2), 1.78 (ddd, J = 13.5/6.0/2.9 Hz, 1H, N(CH2CH2)2),
1.84–2.13 (m, 2H, N(CH2CH2)2), 2.31–2.50 (m, 2H, N(CH2CH2)2),
2.72 (d broad, J = 11.3 Hz, 1H, N(CH2CH2)2), 3.04 (d broad,
J = 9.8 Hz, 1H, N(CH2CH2)2), 3.45–3.52 (m, 1H, NCH(CH3)Ar),
4.45–4.66 (m, 2H, CHCH2F), 5.35 (dt, J = 20.4/4.7 Hz, 0.5H, ArCHO),
5.36 (dt, J = 20.4/4.7 Hz, 0.5H, ArCHO), 7.16–7.38 (m, 9H, arom. H).
Purity (HPLC): 95.5%, tR = 17.92 min.
14 (Rf (cyclohexane/ethyl acetate 7:3+NH3 (0.01%)) 0.37): Color-
less oil, yield 5.8 mg (11.2%). C21H25NO (307.4). MS (EI): m/z = 307
[M], 292 [M–CH3], 230 [M–Ph], 202 [M–CH3CHPh], 91 [CH2Ph]. IR:
m
(cmꢁ1) = 2969 (C–H), 1601 (arom. C@C), 1071 (C–O), 754 (C–H,
~
C
13H16FNO (221.3); 10: C13H17NO (203.3). MS (ESI): m/z = 222
(cmꢁ1) = 3298 (N–H), 2940
~
m
[MH (9)], 204 [MH (12)]. IR (9+10):
(C–H), 1604 (arom. C@C), 1072 (C–O), 754 (C–H, 1,2-disubst.
arom.). 1H NMR (CDCl3): d (ppm) = 1.50 (d, J = 6.4 Hz, 3 ꢃ 0.31H,
CHCH3), 1.68–2.04 (m, 4H, N(CH2CH2)2), 3.00–3.17 (m, 4H,
N(CH2CH2)2), 4.57 (ddd, J = 47.3/9.7/4.9 Hz, 0.69H, CHCH2F), 4.61
(ddd, J = 47.7/9.7/3.8 Hz, 0.69H, CHCH2F), 5.29 (q, J = 6.4 Hz,
0.31H, ArCHO), 5.41 (dt, J = 20.3/4.3 Hz, 0.69H, ArCHO), 7.10–7.18
(m, 1H, arom. H), 7.21–7.36 (m, 3H, arom. H). A signal for the
NH-proton is not seen in the spectrum. Purity (HPLC): 9: 66.6%,
tR = 12.01 min; 10: 30.2%, tR = 13.05 min.
4.1.7. 10-(4-Fluorobenzyl)-3-(fluoromethyl)-3H-spiro[[2]benzo-
furan-1,40-piperidine] (11) and 10-(4-fluorobenzyl)-3-methyl-
3H-spiro[[2]benzofuran-1,40-piperidine] (12)
A mixture of the secondary amines 9/10 (37.4 mg, 0.17 mmol),
4-fluorobenzyl chloride (26 lL, 0.22 mmol), K2CO3 (0.118 mg,
0.85 mmol) and CH3CN (5 mL) was heated to reflux for 8 h and sub-
sequently stirred at rt for 15 h. The mixture was filtered over Cel-
iteÒ, the solvent was removed in vacuo and the residue was
purified by fc (0.7 cm, cyclohexane/ethyl acetate 8:2, 5 mL). In
addition to the pure samples 24 mg (ca. 27%) of a mixture was
isolated.
1,2-disubst. arom.), 700 (C–H, monosubst. arom.). 1H NMR (CDCl3):
d (ppm) = 1.42 (d, J = 6.5 Hz, 3H, NCHArCH3), 1.46 (t, J = 6.3 Hz, 3H,
ArCHCH3) 1.64 (ddd, J = 13.5/4.3/2.8 Hz, 1H, N(CH2CH2)2), 1.76
(ddd, J = 13.5/4.8/2.8 Hz, 1H, N(CH2CH2)2), 1.72–1.82 (m, 0.5H,
N(CH2CH2)2), 1.91 (td, J = 13.1/4.6 Hz, 0.5H, N(CH2CH2)2), 2.02 (td,
J = 13.3/3.9 Hz, 0.5H, N(CH2CH2)2), 2.14 (td, J = 12.9/4.3 Hz, 0.5H,
N(CH2CH2)2), 2.28–2.50 (m, 2H, N(CH2CH2)2), 2.63–2.72 (m, 1H,
N(CH2CH2)2), 2.98–3.08 (m, 1H, N(CH2CH2)2), 3.48 (q, J = 6.5 Hz,
1H, NCH(CH3)Ar), 5.25 (m, 1H, ArCHO), 7.09–7.14 (m, 2H, arom.
H), 7.22–7.27 (m, 2H, arom. H), 7.30–7.37 (m, 5H, arom. H). Purity
(HPLC): 98.3%, tR = 18.44 min.
11 (Rf (cyclohexane/ethyl acetate 5:5) 0.40): colorless oil, yield
19.6 mg (22%). C20H22F2NO (329.4). MS (EI): m/z = 329 [M], 234
~
[M–PhF], 220 [M–CH2PhF], 109 [PhCH2F]. IR:
m
(cmꢁ1) = 2941 (C–
4.1.9. [(10-Benzyl-3H-spiro[[2]benzofuran-1,40-piperidin]-3-yl)-
methyl] tosylate (15)
H), 1603 (arom. C@C), 1053 (C–O), 828 (C–H, 1,4-disubst. arom.),
755 (C–H, 1,2-disubsst. arom.). 1H NMR (CDCl3): d (ppm) = 1.74
(ddd, J = 13.9/5.8/2.8 Hz, 2H, N(CH2CH2)2), 1.97 (td, J = 13.1/
4.5 Hz, 1H, N(CH2CH2)2), 2.06 (td, J = 13.1/4.4 Hz, 1H, N(CH2CH2)2),
2.45 (td, J = 11.8/2.5 Hz, 1H, N(CH2CH2)2), 2.48 (td, J = 11.8/2.5 Hz,
1H, N(CH2CH2)2), 2.81 (br d, J = 11.2 Hz, 2H, N(CH2CH2)2), 3.55 (s,
2H, NCH2Ph), 4.56 (ddd, J = 47.3/9.7/4.9 Hz, 1H, CHCH2F), 4.61
(ddd, J = 47.7/9.7/3.8 Hz, 1H, CHCH2F), 5.40 (dt, J = 20.3/4.3 Hz,
1H, ArCHO), 7.01 (t, J = 8.7 Hz, 2H, arom. H), 7.16–7.18 (m, 1H,
arom. H), 7.21–7.23 (m, 1H, arom. H), 7.28–7.35 (m, 4H, arom.
H). Purity (HPLC): 95.1%, tR = 17.74 min.
Under N2 the alcohol 8 (97 mg, 0.31 mmol), 4-(dimethyl-
amino)pyridine (DMAP, 9.7 mg, 0.08 mmol) and NEt3 (0.2 mL,
1.44 mmol) were dissolved in CH2Cl2 (8 mL) and the solution was
cooled down to ꢁ25 °C. A solution of p-toluenesulfonyl chloride
(120 mg, 0.63 mmol) in CH2Cl2 (2 mL) was added and the mixture
was stirred for 30 min at ꢁ25 °C and for 21 h at rt. Diluted NaOH
was added, the layers were separated and the aqueous layer was
extracted with CH2Cl2 (4ꢃ). The combined organic layers were
dried (Na2SO4), concentrated in vacuo and the residue was purified
by fc (1.7 cm, cyclohexane/ethyl acetate 7:3, 10 mL, Rf (cyclohex-
ane/ethyl acetate 5:5) 0.35.
12 (Rf (cyclohexane/ethyl acetate 5:5) 0.44): colorless oil, yield
8.0 mg (10%). C20H22FNO (311.4). MS (EI): m/z = 311 [M], 216
(cmꢁ1) = 2939
~
m
Colorless oil, yield 127.4 mg (88%). C27H29NO4S (463.6). MS (EI):
m/z = 463 [M], 372 [M–CH2Ph], 308 [M–SO2PhCH3], 91 [PhCH2]. IR:
(cmꢁ1) = 2925 (C–H), 1598 (arom. C@C), 1361, 1175 (O2S@O),
~
m
[M–PhF], 202 [M–CH2PhF], 109 [F PhCH2]. IR:
(C–H), 1602 (arom. C@C), 1071 (C–O), 829 (C–H, 1,4-disubst.
arom.), 754 (C-H, 1,2-disubst. arom.). 1H NMR (CDCl3):
d
813 (C–H, 1,4-disubst. arom.), 756 (C–H, 1,2-disubst. arom.), 740,
698 (C–H, monosubst. arom.). 1H NMR (CDCl3): d (ppm) = 1.56
(ddd, J = 13.6/5.2/2.5 Hz, 1H, N(CH2CH2)2), 1.65 (ddd, J = 13.4/5.2/
2.5 Hz, 1H, N(CH2CH2)2), 1.92 (td, J = 12.9/4.5 Hz, 1H, N(CH2CH2)2),
1.98 (td, J = 13.1/4.5 Hz, 1H, N(CH2CH2)2), 2.30 (td, J = 12.7/2.5 Hz,
1H, N(CH2CH2)2), 2.38 (td, J = 12.5/2.7 Hz, 1H, N(CH2CH2)2), 2.42
(s, 3H, ArCH3), 2.73 (d broad, J = 11.2 Hz, 1H, N(CH2CH2)2), 2.79 (d
broad, J = 11.3 Hz, 1H, N(CH2CH2)2), 3.56 (s, 2H, NCH2Ph), 4.15
(dd, J = 10.1/5.2 Hz, 1H, CH2OTos), 4.21 (dd, J = 10.1/4.5 Hz, 1H,
CH2OTos), 5.35 (t broad, J = 4.8 Hz, 1H, ArCHO), 7.12 (br t,
J = 7.1 Hz, 2H, arom. H), 7.22–7.37 (m, 9H, arom. H), 7.73 (br d,
J = 8.3 Hz, 2H, arom. H). Purity (HPLC): 99.2%, tR = 19.39 min.
(ppm) = 1.49 (d, J = 6.4 Hz, 3H, CHCH3), 1.72 (ddd, J = 13.8/5.4/
2.7 Hz, 2H, N(CH2CH2)2), 1.88 (td, J = 13.0/4.0 Hz, 1H, N(CH2CH2)2),
2.10 (td, J = 12.9/4.1 Hz, 1H, N(CH2CH2)2), 2.44 (td, J = 14.7/2.3 Hz,
1H, N(CH2CH2)2), 2.47 (td, J = 14.7/2.3 Hz, 1H, N(CH2CH2)2), 2.76–
2.82 (m, 2H, N(CH2CH2)2), 3.55 (s, 2H, NCH2Ph), 5.28 (q,
J = 6.3 Hz, 1H, ArCHO), 7.01 (t, J = 7.7 Hz, 2H, arom. H), 7.11–7.15
(m, 2H, arom. H), 7.24–7.28 (m, 2H, arom. H), 7.30–7.34 (m, 2H,
arom. H). Purity (HPLC): 97.6%, tR = 17.92 min.
4.1.8. 3-(Fluoromethyl)-10-(1-phenylethyl)-3H-spiro[[2]benzo-
furan-1,40-piperidine] (13) and 3-methyl-10-(1-phenylethyl)-3H-
spiro[[2]benzofuran-1,40-piperidine] (14)
A mixture of the secondary amines 9/10 (37.4 mg, 0.17 mmol),
4.2. Receptor binding studies
1-bromo-1-phenylethane (30 lL, 0.22 mmol), K2CO3 (0.117 mg,
0.85 mmol) and CH3CN (5 mL) was heated to reflux for 8 h and sub-
sequently stirred at rt for 15 h. The mixture was filtered over Cel-
iteÒ, the solvent was removed in vacuo and the residue was
purified by fc (0.7 cm, cyclohexane/ethyl acetate 7:3+NH3
Membrane preparations and r1 and r2 assays were performed
as described in Ref. 36,38–41 The protein concentration was deter-
mined according to the method of Bradford51 using bovine serum
albumin as standard. The r1 assay was performed with the

A. Maisonial et al. / Bioorg. Med. Chem. 20 (2012) 257–269
267
radioligand [3H]-(+)-pentazocine (22 Ci/mmol; Perkin Elmer). The
thawed membrane preparation (about 75 g of the protein) was
spray (ESI) interface. The ion spray voltage was 3 kV in positive
mode at a sheath gas flow of 80 arbitrary units. Temperature of
the capillary was set to 200 °C and the capillary voltage to 4 V. A
l
incubated with various concentrations of test compounds, 2 nM
[3H]-(+)-pentazocine, and buffer (50 mM TRIS, pH 7.4) in a total
20
l
L volume of the prepared incubation solution was injected
volume of 200
nated by rapid filtration through the presoaked filtermats by using
the cell harvester. After washing each well five times with 300
lL for 180 min at 37 °C. The incubation was termi-
onto a LiChrospherÒ RP Select B 5
lm (250 ꢃ 4 mm) column
(Merck, Germany) at a flow rate of 1.0 mL/min. The mobile phase
was composed of (A) 15% acetonitrile in water and (B) 60% aceto-
nitrile in water. 0.05% trifluoroacetic acid was added to both com-
ponents. The following gradient was applied (A%): 0 min: 100%,
20 min: 0%, 23 min: 0%, 24 min: 100%, and 30 min: 100%.
In addition to the MS spectra the UV absorption at 210 nm was
recorded.
lL
of water, the filtermats were dried at 95 °C. Subsequently, the solid
scintillator was put on the filtermat and melted at 95 °C. After
5 min, the solid scintillator was allowed to solidify at rt. The bound
radioactivity trapped on the filters was counted in the scintillation
analyzer. The nonspecific binding was determined with 10 lM
unlabeled (+)-pentazocine. The Kd-value of the radioligand [3H]-
(+)-pentazocine is 2.9 nM.52
4.4. Radiochemistry
4.3. In vitro biotransformation
4.4.1. General
No-carrier-added [18F]fluoride (half-life: 109.8 minutes) was
produced via the [18O(p, n)18F] nuclear reaction by irradiation of
a [18O]water target (>97%-enriched, 2 mL) on an General Electric
PETtrace cyclotron (16.5 MeV proton beam). All radiotracers were
shown by radio-TLC or analytical radio-HPLC to be identical to
the authentic nonradioactive material and to be free of significant
chemical and radiochemical impurities.
4.3.1. Preparation of rat liver microsomes
Frozen rat livers from male Wistar rats were thawn in phos-
phate buffer pH 7.4 with 0.25 M sucrose and 5 mM EDTA, cut into
small pieces and homogenized with a Potter (Elvehjem Potter, B.
Braun Biotech International). The homogenization was carried
out at 4 °C. The resulting suspension was centrifuged at 10,000g
for 15 min at 4 °C. Fat was removed by cellulose. The pellet was
resuspended in buffer, the mixture was centrifuged again and
afterwards both supernatants were combined. This suspension
was transferred into an ultracentrifuge for 60 min at 100,000g
and 4 °C. The resulting supernatant (cytosol) was discarded, the
pellet washed carefully with buffer, resuspended and the centrifu-
gation repeated. Finally the supernatant was removed, the pellet
resuspended in a small amount of phosphate buffer pH 7.4 and
the microsome suspension was stored at ꢁ80 °C. The protein con-
centration was determined according to the method of Bradford3
using bovine serum albumin as standard.
4.4.2. [18F]10-Benzyl-3-(fluoromethyl)-3H-spiro[[2]benzofuran-
1,40-piperidine] ([18F]1)
The aqueous [18F]fluoride solution was azeotropically dried
using acetonitrile in the presence of potassium carbonate
(1.8 mg) and K222 (Kryptofix, 11.2 mg) for [18F]fluoride activation.
No-carrier-added aliphatic nucleophilic substitution was per-
formed by reacting tosylate 15 (2.5 mg) in anhydrous DMSO
(1 mL) with the dried K[18F]F–K222–carbonate complex at 150 °C
under stirring for 20 min. After cooling to room temperature, the
labeling efficiencies obtained were controlled by radio-TLC analy-
sis (between 67% and 86%, n = 4, SiO2, ethyl acetate/petroleum
ether/ammonia, 5/5/0.1, v/v/v, Rf = 0.74). The crude reaction mix-
ture was diluted with water (3 mL) and directly applied on a
semi-preparative radio-HPLC system for purification consisting of
a S1021 pump (SYKAM Chromatographie), UV detector (Well-
ChromK-2001, KNAUER), NaI(Tl)-counter, data acquisition by an
automated system (NINA, Nuclear Interface) on a Multospher 120
RP-18 AQ column with precolumn (150 mm ꢃ 10 mm and
4.3.2. Incubation of 1 (WMS-1850) with rat liver microsomes
The incubation was carried out in phosphate buffer pH 7.4 at rt
in a circular shaker (IKA vibrax VXR) and contained rat liver micro-
somes (1.5 mg/mL protein), 0.86 mM MgCl2 and 2.6 mM NADPH/
H+. The concentration of 1 was 260
lM in a final volume of
0.9 mL. Usually, the incubation was stopped after 90 min by addi-
tion of cold acetonitrile (ꢁ20 °C). The samples were stored for
30 min at ꢁ20 °C to complete protein precipitation. After thawing,
the samples were centrifuged (10,000g). The supernatant was dec-
50 mm ꢃ 10 mm, respectively, particle size 5
lm, Chromatogra-
phie Service) using the following conditions: isocratic elution,
anted, filtered with a 0.45 lm (pore size) syringe filter made from
regenerated cellulose and finally analyzed.
55% acetonitrile in aqueous 20 mM ammonium acetate, flow
rate = 2 mLꢀminꢁ1
,
k = 254 nm. The collected fractions ([18F]1,
tR = 37.4 min) were combined, diluted with water (40–45 mL),
passed through a Sep-PakÒ Plus C-18 cartridge (Waters Corpora-
tion), and washed with water (2 ꢃ 5 mL). The radiotracer [18F]1
was eluted using methanol (2 mL). The solvent was carefully evap-
orated under argon and [18F]1 was dissolved in 0.9% sodium chlo-
ride isotonic solution containing 5% of ethanol. Radiochemical
yields were obtained between 38 and 50% (n = 3). Radio-TLC and
analytic radio-HPLC analyses (Multospher 120 RP-18 AQ column,
4.3.3. Degradation of 1 (WMS-1850) during 90 min
Seven incubations of 1 were carried out in phosphate buffer pH
7.4 at rt in a circular shaker containing rat liver microsomes
(1.5 mg/mL protein), 0.86 mM MgCl2 and 2.6 mM NADPH/H+. The
concentration of 1 was 320
the first hour every 10 min one incubation was stopped by addition
of cold acetonitrile (ꢁ20 °C). 200 L of a praziquantel solution
(0.6 mg/mL) were added as internal standard resulting in a final
concentration of 220 M in a total volume of 1.1 mL. After
lM in a total volume of 0.9 mL. During
l
250 mm ꢃ 4.6 mm, particle size 5
lm, Chromatographie Service,
l
5% MeCN in aqueous 20 mM ammonium acetate for 5 min and gra-
dient to 80% MeCN in aqueous 20 mM ammonium acetate over
50 min, flow rate = 1 mLꢀminꢁ1, k = 254 nm) revealed a radiochem-
ical purity of 99.4 0.3% (n = 6) and a specific activity usually high-
90 min the last incubation was stopped and analyzed. The calibra-
tion was carried out with the same matrix except NADPH/H+. All
calibration standards were treated in the same way (90 min on
the shaker, protein precipitation with acetonitrile, centrifugation,
etc.).
er than 375 GBq/lmol.
4.5. In vitro stability of [18F]1
4.3.4. HPLC (ESI)–MS
The HPLC–MS system consisted of a Waters AllianceÒ 2690 sep-
arations module, a Waters 2487 dual k absorbance detector and a
Thermo Finnigan LCQÒ ion trap mass spectrometer with an electro
The radiotracer [18F]1 was incubated at 40 °C for 120 minutes in
either 0.9% sodium chloride solution, phosphate-saline-solution
(0.1 M sodium phosphate, 0.15 M sodium chloride, pH 7.2) or

268
A. Maisonial et al. / Bioorg. Med. Chem. 20 (2012) 257–269
0.01 M TRIS–HCl (pH 7.4 at 21 °C). Aliquots were taken from the
crude mixtures at selected times points during the experiments
and analyzed by radio-TLC (SiO2, ethyl acetate/petroleum ether/
ammonia, 5/5/0.1, v/v/v) and gradient analytical radio-HPLC
(Multospher 120 RP-18 AQ column, 250 mm ꢃ 4.6 mm, particles
References
1. Martin, W. R.; Eades, C. G.; Thompson, J. A.; Huppler, R. E.; Gilbert, P. E. J.
Pharmacol. Exp. Ther. 1976, 197, 517.
2. Hayashi, T.; Su, T. P. CNS Drugs 2004, 18, 269.
3. Hayashi, T.; Su, T. P. Curr. Neuropharmacol. 2005, 3, 267.
4. Kitaichi, K.; Chabot, J. G.; Moebius, F. F.; Flandorfer, A.; Glossmann, H.; Quirion,
R. J. Chem. Neuroanat. 2000, 20, 375.
5. Bowen, W. D.; Hellewell, S. B.; McGarry, K. A. Eur. J. Pharmacol. 1989, 163, 309.
6. Quirion, R.; Bowen, W. D.; Itzhak, Y.; Junien, J. L.; Musacchio, J. M.; Rothman, R.
B.; Su, T. P.; Tam, S. W.; Taylor, D. P. Trends Pharmacol. Sci. 1992, 13, 85.
7. Kekuda, R.; Prasad, P. D.; Fei, Y. J.; Leibach, F. H.; Ganapathy, V. Biochem.
Biophys. Res. Commun. 1996, 229, 553.
size 5 lm, Chromatographie Service, 5% MeCN in aqueous 20 mM
ammonium acetate for 5 min and gradient to 80% MeCN in aque-
ous 20 mM ammonium acetate over 50 min, flow rate = 1 -
mLꢀminꢁ1, k = 254 nm).
4.6. Determination of in vivo stability in brain
8. Fontanilla, D.; Johannessen, M.; Hajipour, A. R.; Cozzi, N. V.; Jackson, M. B.;
Ruoho, A. E. Science 2009, 323, 934.
The radiotracer [18F]1 (191 MBq) in 200
lL isotonic solution
9. Su, T. P.; Hayashi, T.; Maurice, T.; Buch, S.; Ruoho, A. E. Trends Pharmacol. Sci.
2010, 31, 557.
were injected via the tail vein in CD-1 mice. At 30 or 60 min p.i.
animals were sacrificed. After isolation, whole brains (n = 2 per
time) were homogenized in 1–2 v/w 50 mM TRIS–HCl, pH 7.4 at
4 °C, in a borosilicate glass cylinder by 10 strokes of a PTFE plunge
at a speed of 1000ꢁ1 using a Potter S Homogenizer (B. Braun,
Germany). The homogenates were extracted using ice-cold aceto-
nitrile according a standard protocol.46 After concentration, the
supernatants were analyzed by gradient analytical radio-HPLC
(Multospher 120 RP-18 AQ column, 250 mm ꢃ 4.6 mm, particle
10. de la Puente, B.; Nadal, X.; Portillo-Salido, E.; Sanchez-Arroyos, R.; Ovalle, S.;
Palacios, G.; Muro, A.; Romero, L.; Entrena, J. M.; Baeyens, J. M.; Lopez-Garcia, J.
A.; Maldonado, R.; Zamanillo, D.; Vela, J. M. Pain 2009, 145, 294.
11. Diaz, J. L.; Zamanillo, D.; Corbera, J.; Baeyens, J. M.; Maldonado, R.; Pericas, M.
A.; Vela, J. M.; Torrens, A. Cent. Nerv. Syst. Agents Med. Chem. 2009, 9, 172.
12. Navarro, G.; Moreno, E.; Aymerich, M.; Marcellino, D.; McCormick, P. J.; Mallol,
J.; Cortes, A.; Casado, V.; Canela, E. I.; Ortiz, J.; Fuxe, K.; Lluis, C.; Ferre, S.;
Franco, R. Proc. Natl. Acad. Sci. U.S.A. 2010, 107, 18676.
13. Hayashi, T.; Su, T. P. Expert Opin. Ther. Targets 2008, 12, 45.
14. Lucas, G.; Rymar, V. V.; Sadikot, A. F.; Debonnel, G. Int. J. Neuropsychopharmacol.
2008, 11, 485.
15. Kishi, T.; Yoshimura, R.; Okochi, T.; Fukuo, Y.; Kitajima, T.; Okumura, T.;
Tsunoka, T.; Kawashima, K.; Yamanouchi, Y.; Kinoshita, Y.; Umene-Nakano, W.;
Naitoh, H.; Nakamura, J.; Ozaki, N.; Iwata, N. Neuropharmacology 2010, 58,
1168.
size 5 lm, Chromatographie Service, 5% MeCN in aqueous 20 mM
ammonium acetate for 5 min and gradient to 80% MeCN in aque-
ous 20 mM ammonium acetate over 50 min, flow rate = 1 -
16. Sabino, V.; Cottone, P.; Parylak, S. L.; Steardo, L.; Zorrilla, E. P. Behav. Brain Res.
2009, 198, 472.
mLꢀminꢁ1
, k = 254 nm) and radio-TLC (SiO2, ethyl acetate/
petroleum ether/ammonia, 5/5/0.1, v/v/v, Rf = 0.74). Spots on
radio-TLC plates were analyzed by radioluminescence recording
(BAS-1800 II Bioimaging Analyzer, Fuji Film) followed by evalua-
tion with the AIDA 2.31 software (Raystest).
17. Hindmarch, I.; Hashimoto, K. Hum. Psychopharmacol. 2010, 25, 193.
18. Furuse, T.; Hashimoto, K. Ann. Gen. Psychiatry 2009, 8, 26.
19. Furuse, T.; Hashimoto, K. Ann. Gen. Psychiatry 2010, 9, 18.
20. Hashimoto, K. CNS Neurol. Disord. Drug Targets 2009, 8, 470.
21. Paschos, K. A.; Veletza, S.; Chatzaki, E. CNS Drugs 2009, 23, 755.
22. Kulkarni, S. K.; Dhir, A. Expert Rev. Neurother. 2009, 9, 1021.
23. Maurice, T.; Su, T. P. Pharmacol. Ther. 2009, 124, 195.
24. Fishback, J. A.; Robson, M. J.; Xu, Y. T.; Matsumoto, R. R. Pharmacol. Ther. 2010,
127, 271.
25. Cobos, E. J.; Entrena, J. M.; Nieto, F. R.; Cendan, C. M.; Del Pozo, E. Curr.
Neuropharmacol. 2008, 6, 344.
26. Toyohara, J.; Sakata, M.; Ishiwata, K. Cent. Nerv. Syst. Agents Med. Chem. 2009, 9,
190.
27. Mishina, M.; Ishiwata, K.; Ishii, K.; Kitamura, S.; Kimura, Y.; Kawamura, K.; Oda,
K.; Sasaki, T.; Sakayori, O.; Hamamoto, M.; Kobayashi, S.; Katayama, Y. Acta
Neurol. Scand. 2005, 112, 103.
28. Mishina, M.; Ohyama, M.; Ishii, K.; Kitamura, S.; Kimura, Y.; Oda, K.;
Kawamura, K.; Sasaki, T.; Kobayashi, S.; Katayama, Y.; Ishiwata, K. Ann. Nucl.
Med. 2008, 22, 151.
4.7. Ex vivo autoradiography study
The radiotracer [18F]1 (24 MBq) in 200
lL isotonic solution were
injected via the tail vein into conscious female CD-1 mouse. At
45 min p.i., the animal was sacrificed. After removing the brain,
the hemispheres were frozen in isopentane (ꢁ35 °C). Then, consec-
utive 12 lm thick saggital sections were sliced on a cryostat micro-
tome (Microm), thaw-mounted onto microscope slides, dried with
a stream of cold air and exposed to 18F-sensitive storage phosphor
screens (Fuji Film). The image plates were analyzed using a BAS-
1800II system bioimaging analyzer (Fuji Film). Scan data were
visualized and processed by computer assisted microdensitometry
(Aida 2.31, Raytest). Brain structures were confirmed by Nissl
staining of the exposed sections.
29. Ishiwata, K.; Ishii, K.; Kimura, Y.; Kawamura, K.; Oda, K.; Sasaki, T.; Sakata, M.;
Senda, M. Ann. Nucl. Med. 2008, 22, 411.
30. Sakata, M.; Kimura, Y.; Naganawa, M.; Ishikawa, M.; Oda, K.; Ishii, K.;
Hashimoto, K.; Chihara, K.; Ishiwata, K. Ann. Nucl. Med. 2008, 22, 143.
31. Shidahara, M.; Seki, C.; Naganawa, M.; Sakata, M.; Ishikawa, M.; Ito, H.; Kanno,
I.; Ishiwata, K.; Kimura, Y. Ann. Nucl. Med. 2009, 23, 163.
32. Matsuno, K.; Nakazawa, M.; Okamoto, K.; Kawashima, Y.; Mita, S. Eur. J.
Pharmacol. 1996, 306, 271.
33. Lever, J. R.; Gustafson, J. L.; Xu, R.; Allmon, R. L.; Lever, S. Z. Synapse 2006, 59,
350.
34. Shiba, K.; Ogawa, K.; Ishiwata, K.; Yajima, K.; Mori, H. Bioorg. Med. Chem. 2006,
14, 2620.
35. Berardi, F.; Ferorelli, S.; Colabufo, N. A.; Leopoldo, M.; Perrone, R.; Tortorella, V.
Bioorg. Med. Chem. 2001, 9, 1325.
36. Große Maestrup, E.; Wiese, C.; Schepmann, D.; Hiller, A.; Fischer, S.;
Scheunemann, M.; Brust, P.; Wünsch, B. Bioorg. Med. Chem. 2009, 17, 3630.
37. Jasper, A.; Schepmann, D.; Lehmkuhl, K.; Vela, J. M.; Buschmann, H.; Holenz, J.;
Wünsch, B. Eur. J. Med. Chem. 2009, 44, 4306.
38. Maier, C. A.; Wünsch, B. J. Med. Chem. 2002, 45, 4923.
39. Maier, C. A.; Wünsch, B. J. Med. Chem. 2002, 45, 438.
40. Oberdorf, C.; Schepmann, D.; Vela, J. M.; Diaz, J. L.; Holenz, J.; Wünsch, B. J. Med.
Chem. 2008, 51, 6531.
41. Schläger, T.; Schepmann, D.; Wurthwein, E. U.; Wünsch, B. Bioorg. Med. Chem.
2008, 16, 2992.
4.8. Organ distribution study
About 300 kBq of [18F]1 in 200
lL of 0.9% sodium chloride solu-
tion containing 5% of ethanol were injected via the tail vein into
conscious female CD-1 mice. At 5, 30, 60, and 120 min p.i., 3–6 ani-
mals per time points were anesthetized. Blood and urine samples
were collected and the animals were euthanized by luxation of
the cervical spine. Organs of interest were dissected. All samples
were weighed, and radioactivity was measured in a calibrated
c
-counter (WallacWIZARD, Perkin-Elmer). The percentage of the
injected dose per gram of wet tissue (% ID/g) was calculated by
dividing the tissue counts by the initial dose counts and then by
the weight of the tissue sample. Data presented correspond to
means values with standard deviation (SD).
42. Wiese, C.; Große Maestrup, E.; Schepmann, D.; Vela, J. M.; Holenz, J.;
Buschmann, H.; Wünsch, B. J. Pharm. Pharmacol. 2009, 61, 631.
43. Wiese, C.; Große Maestrup, E.; Schepmann, D.; Grimme, S.; Humpf, H. U.; Brust,
P.; Wünsch, B. Chirality 2011, 23, 148.
Acknowledgment
44. Fischer, S.; Wiese, C.; Große Maestrup, E.; Hiller, A.; Deuther-Conrad, W.;
Scheunemann, M.; Schepmann, D.; Steinbach, J.; Wünsch, B.; Brust, P. Eur. J.
Nucl. Med. Mol. Imaging 2011, 38, 540.
This work was supported by the Deutsche Forschungsgemeins-
chaft, which is gratefully acknowledged.

A. Maisonial et al. / Bioorg. Med. Chem. 20 (2012) 257–269
269
45. Große Maestrup, E.; Fischer, S.; Wiese, C.; Schepmann, D.; Hiller, A.; Deuther-
48. Ram, S.; Spicer, L. D. Tetrahedron Lett. 1987, 28, 515.
Conrad, W.; Steinbach, J.; Wünsch, B.; Brust, P. J. Med. Chem. 2009, 52, 6062.
46. Maisonial, A.; Große Maestrup, E.; Fischer, S.; Hiller, A.; Scheunemann, M.;
Wiese, C.; Schepmann, D.; Steinbach, J.; Deuther-Conrad, W.; Wünsch, B.;
Brust, P. Chem. Med. Chem. 2011, 6, 1401.
49. Wirt, U.; Schepmann, D.; Wünsch, B. Eur. J. Org. Chem. 2007, 462.
50. Holl, R.; Jung, B.; Schepmann, D.; Humpf, H.-U.; Grünert, R.; Bednarski, P. J.;
Englberger, W.; Wünsch, B. Chem. Med. Chem 2009, 4, 2111.
51. Bradford, M. M. Anal. Biochem. 1976, 72, 248.
47. Große Maestrup, E.; Wiese, C.; Schepmann, D.; Brust, P.; Wünsch, B. Bioorg.
Med. Chem. 2011, 19, 393.
52. DeHaven-Hudkins, D. L.; Fleissner, L. C.; Ford-Rice, F. Y. Eur. J. Pharmacol. 1992,
722, 371.