Design, synthesis and in vitro evaluation of bridgehead fluoromethyl analogs of N-{2-[4-(2-methoxyphenyl)piperazin-1-yl]ethyl}-N-(pyridin-2-yl) cyclohexanecarboxamide (WAY-100635) for the 5-HT1A receptor

Article abstract of DOI:10.1016/j.ejmech.2011.06.023

Fluorinated analogs that are related to the 5-hydroxytryptamine (5-HT _1A) antagonist, N-{2-[4-(2-methoxyphenyl)piperazin-1-yl]ethyl}-N- (pyridin-2-yl)cyclohexanecarboxamide (WAY-100635), have been synthesized and their binding affinity for the 5-HT_1A receptor and other neurotransmitter receptors (adrenoceptors, sigma receptors, and dopamine receptors), and serotonin transporters was examined in vitro. These ligands were designed to provide a possible potential positron emission tomography (PET) ligand with high metabolic stability. To this end, the cyclohexyl moiety in WAY-100635 and in O-desmethyl WAY-100635 was replaced by a bridge-fused ring (BFR) such as adamantyl, cubyl, bicyclo[2.2.2]octyl and bicyclo[2.2.1]heptyl to reduce the metabolic rate of the amide bond hydrolysis, while a fluoromethyl group was introduced on the other bridgehead of the BFR to prevent defluorination by HF elimination. All synthesized analogs displayed high affinity in the (sub)nanomolar range for the 5-HT_1A receptor, comparable to WAY-100635. In addition, 6b, 6c and 6d were reasonably selective to the 5-HT_1A receptor over the above mentioned receptors. In human hepatocytes, 6b showed a suitable metabolic stability. In conclusion, the obtained data provides a promising starting point for the synthesis of the corresponding ¹⁸F-labeled PET analogs.

Full text of DOI:10.1016/j.ejmech.2011.06.023

European Journal of Medicinal Chemistry 46 (2011) 5728e5735
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Design, synthesis and in vitro evaluation of bridgehead fluoromethyl analogs
of N-{2-[4-(2-methoxyphenyl)piperazin-1-yl]ethyl}-N-(pyridin-2-yl)
cyclohexanecarboxamide (WAY-100635) for the 5-HT1A receptor
,
a
a
b
a
Rana Al Hussainy a , Joost Verbeek , Dion van der Born , Jan Booij , J(Koos) D.M. Herscheid
*
a Department of Nuclear Medicine & PET Research, VU University Medical Center, PO Box 7057, 1007MB Amsterdam, The Netherlands
b Department of Nuclear Medicine, Academic Medical Center, University of Amsterdam, Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands
a r t i c l e i n f o
a b s t r a c t
Article history:
Fluorinated analogs that are related to the 5-hydroxytryptamine (5-HT1A) antagonist, N-{2-[4-(2-
methoxyphenyl)piperazin-1-yl]ethyl}-N-(pyridin-2-yl)cyclohexanecarboxamide (WAY-100635), have
been synthesized and their binding affinity for the 5-HT1A receptor and other neurotransmitter receptors
(adrenoceptors, sigma receptors, and dopamine receptors), and serotonin transporters was examined
in vitro. These ligands were designed to provide a possible potential positron emission tomography (PET)
ligand with high metabolic stability. To this end, the cyclohexyl moiety in WAY-100635 and in O-des-
methyl WAY-100635 was replaced by a bridge-fused ring (BFR) such as adamantyl, cubyl, bicyclo[2.2.2]
octyl and bicyclo[2.2.1]heptyl to reduce the metabolic rate of the amide bond hydrolysis, while a fluo-
romethyl group was introduced on the other bridgehead of the BFR to prevent defluorination by HF
elimination. All synthesized analogs displayed high affinity in the (sub)nanomolar range for the 5-HT1A
receptor, comparable to WAY-100635. In addition, 6b, 6c and 6d were reasonably selective to the 5-HT1A
receptor over the above mentioned receptors. In human hepatocytes, 6b showed a suitable metabolic
stability.
Received 18 March 2011
Received in revised form
10 June 2011
Accepted 16 June 2011
Available online 29 June 2011
Keywords:
5-HT1A receptor
WAY-100635
Bridge-fused ring
In conclusion, the obtained data provides a promising starting point for the synthesis of the corre-
sponding 18F-labeled PET analogs .
Ó 2011 Elsevier Masson SAS. All rights reserved.
1. Introduction
neurotransmitter receptors, re-uptake and ion channel sites,
however it showed also high affinity for a1-adrenoceptors and D4
1
Among many 5-HT1A receptor ligands, WAY-100635 was
developed as the first potent, “silent” and selective 5-HT1A receptor
antagonist. This compound has a binding selectivity of at least >10-
fold relative to other 5-HT receptor subtypes, major
receptors [1]. Later, WAY-100635 was labeled with carbon-11
([carbonyl-11C] WAY-100635) and evaluated to study in vivo
changes in the densities of 5-HT1A receptors in several neuropsy-
chiatric disorders, such as major depression and anxiety disorders
[2,3], Alzheimer’s disease [2,4,5] and schizophrenia [2,6] using PET.
Studies of the metabolic pathway of [carbonyl-11C] WAY-100635 in
humans revealed rapid metabolism to WAY-100634 due to the
hydrolysis of the amide bond, which interfere the PET measure-
ments (Fig. 1) [7e9]. In addition, a disadvantage of using carbon-11
(t1/2 ¼ 20 min) radiotracers is that these tracers can only be used
where both a cyclotron and a PET-camera are in close proximity due
to it short half-life. Zhuang and co-workers have shown in a series
of benzamido analogs of WAY-100635 a limited tolerance for
variations at the amino pyridine position in the native receptor [10].
Therefore, research has been directed toward WAY-100635 analogs
that were labeled with the longer lived fluorine-18 (t1/2 ¼ 110 min)
isotope with the label outside the WAY-100634 moiety [8]. [18F]
fluorinated analogs of WAY-100635 that contain an aromatic
* Corresponding author. Tel.: þ31 20 4449704; fax: þ31 20 4449121.
E-mail address: rhussainy@rnc.vu.nl (R. Al Hussainy).
1
Abbreviations: 5-HT, 5-hydroxytryptamine; BFR, bridge-fused ring; PET, posi-
tron emission tomography; D4, dopamine D4 receptor; [18F]MPPF, 4-[18F]fluoro-N-
{2-[1-(2-methoxyphenyl)piperazine-1-yl]ethyl}-N-(pyridine-2-yl)benzamide; [18F]
MeFBWAY,
ethyl}-N-(pyridine-2-yl)benzamide;
3-methyl-4-[18F]fluoro-N-{2-[1-(2-methoxyphenyl)piperazine-1-yl]
[
18F]FCWAY,
4-[18F]fluoro-N-2-[4-(2-
methoxyphenyl)piperazin-1-ylethyl]-N-(2-pyridyl) cyclohexanecarboxamide; [18F]
MeFWAY, N-{2-[4-(2-Methoxyphenyl)piperazinyl]ethyl}-N-(2-pyridyl)-N-(4-[18F]
fluoromethyl-cyclohexane)carboxamide; LG, leaving group; TBAF, tetrabutyl
ammonium fluoride; IC50, concentration producing 50% inhibition; Ki, apparent
equilibrium inhibition constant; KD, apparent equilibrium dissociation constant; tR,
retention time; Log D7.4, log base ten of the partition coefficient at pH ¼ 7.4; THF,
tetrahydrofuran; DIPA, diisopropylamine.
0223-5234/$ e see front matter Ó 2011 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2011.06.023

R. Al Hussainy et al. / European Journal of Medicinal Chemistry 46 (2011) 5728e5735
5729
N
N
H3C
N
HO
Metabolism
O
O
C
+
NH
C
O
N
N
N
N
O
WAY-100634
WAY-100635
F
O
R
R= H;MPPF
R= Me;MeFBWAY
F
O
O
FCWAY
CH2F
MeFWAY
Fig. 1. WAY-100635, WAY-100634 and WAY-100635 analogs; MPPF, MeFBWAY, FCWAY and MeFWAY.
moiety in the carboxamide part instead of the cyclohexyl moiety
such as 4-[18F]fluoro-N-{2-[1-(2-methoxyphenyl)piperazine-1-yl]
ethyl}-N-(pyridine-2-yl)benzamide ([18F]p-MPPF) [11e13] and 3-
methyl-4-[18F]fluoro-N-{2-[1-(2-methoxyphenyl)piperazine-1-yl]
receptor [19]. This study revealed that the bridge-fused rings (BFRs)
are well tolerated with respect to affinity for the 5-HT1A receptor.
Furthermore, it confirmed the finding of Wilson [20] that a more
bulky group reduces the hydrolysis of the amide bond.
ethyl}-N-(pyridine-2-yl)benzamide
([18F]MeFBWAY)
were
The aim of the present study was to synthesize analogs of WAY-
100635 and O-desmethyl WAY-100635 with a BFR moiety attached
to the carboxamide and with a fluoromethyl group on the other
bridgehead of the BFR (Fig. 2). Based on their chemical structure,
these compounds were expected to have a high affinity for the
target 5-HT1A receptor, at least comparable to that of WAY-100635.
The rapid in vivo hydrolysis of the amide bond is probably reduced
due to steric hindrance. Defluorination might be prevented because
fluorine is placed on a primary carbon and in such a position that
makes HF elimination chemically impossible. In addition, it is also
expected that these compounds will have a lipophilicity within the
range (Log D7.4 ¼ 2e3.5) [21], which is considered optimal for brain
penetration and low non-specific binding. An a-chiral structure is
maintained which avoid the often difficult separation of enantio-
mers during the chemical synthesis. Moreover, the labeling with
fluorine-18 can be performed in a last synthesis step when these
compounds are being used as radiopharmaceuticals. Herein, their
synthesis and in vitro binding affinity, selectivity and stability are
described.
prepared [11] (Fig. 1). [18F]p-MPPF showed ca. 4e6 times lower
in vivo binding potentials than [carbonyl-11C] WAY-100635 [13].
[
18F]MeFBWAY showed ca. 7 times lower in vitro binding affinity
than WAY-100635 [11]. Displacement experiments of [18F]p-MPPF
with WAY-100635 in the hippocampus of rats showed only 60% [11]
to 77% [12] brain uptake reduction and 80% of [18F]MeFBWAY [11].
[
18F]p-MPPF was rapidly metabolized in human subjects [13,14],
while [18F]MeFBWAY is less selective than [18F]p-MPPF. [18F]
MeFBWAY has a relatively high affinity for a1-adrenoceptors
(Ki ¼ 13.2 nM) compared to [18F]p-MPPF (Ki ¼ 151 nM) [11].
Alternatively, analogs with a saturated moiety such as [18F]-4-
fluoro-N-2-[4-(2-methoxyphenyl)piperazin-1-ylethyl]-N-(2-
pyridyl) cyclohexanecarboxamide, ([18F]FCWAY) were prepared
(Fig. 1).
The trans-isomer of [18F]FCWAY showed higher binding affinity
for the 5-HT1A receptor than [18F]p-MPPF and [18F]MeFBWAY, but
now, in addition to the hydrolysis of the amide bond, defluorination
was observed [11]. This led to a substantial skull uptake in human of
[
18F] amounting to twice that of average uptake in the whole brain
at 60 min post-injection, which hampers clinical applications
[15e17]. Placing fluorine on a primary carbon as in N-{2-[4-(2-
methoxyphenyl)piperazinyl]ethyl}-N-(2-pyridyl)-N-(4-[18F]fluo-
2. Chemistry
romethyl-cyclohexane)carboxamide
([18F]MeFWAY)
greatly
A general pathway for the synthesis of compounds 5aed is
outlined in Scheme 1. Compounds 1a and 1b were synthesized by
mono-saponification of the commercially available di-esters,
according to literature procedures of Eaton [22] and Frazer [23].
4-(Ethoxycarbonyl)bicyclo[2.2.2]octane-1-carboxylic acid (1c)2
reduced this in vivo instability [18].
Recently, we reported the synthesis of bridgehead iodinated
WAY-100635 analogs and the evaluation of their biological prop-
erties in vitro as potential SPECT tracers for imaging 5-HT1A
was
obtained
from
diethyl
2,5-dioxocyclohexane-1,4-
dicarboxylate in four steps as described by Boulerice and Frazer
[23e26]. Compound 1d was synthesized by mono-saponification of
dimethyl bicyclo[2.2.1]heptane-1,4-dicarboxylate, which was
synthesized according to a method of Della and Tsanaktsidis [27].
Compound 2 was synthesized starting from a reduction of the
appropriate acid 1 with borane dimethyl sulfide complex [28]. In
order to obtain the fluorinated compounds, both nucleophilic
substitution reactions and electrophilic substitution reactions were
F
F
F
F
R
R
R
R
O
O
O
O
R = WAY-100634, O-desmethyl WAY-100634
2
4-(Methoxycarbonyl)bicyclo[2.2.2]octane-1-carboxylic acid is nowadays
Fig. 2. Novel analogs of WAY-100635 and O-desmethyl WAY-100635.
commercially available, see Experimental section - Chemistry.

5730
R. Al Hussainy et al. / European Journal of Medicinal Chemistry 46 (2011) 5728e5735
i
iii
ii
O
OH
O
BFR
LG
O
BFR
O
BFR
O
BFR
F
R
R
R
R
O
O
O
OH
O
1
2
3
4
R = Me or Et
BFR = Bridged-fused ring
a:BFR = adamantane
b:BFR = cubane
c:BFR = bicyclo[2.2.2]octane
d:BFR = bicyclo[2.2.1]heptane
iv
N
OR
N
Cl
BFR
v
vi, vii
HO
BFR
N
BFR
N
O
F
O
F
O
F
5
BFR =
6a: R = Me, 7a: R = H
6b: R= Me, 7b: R= H
6c: R = Me, 7c: R = H
6d: R = Me, 7d: R = H
Scheme 1. General pathway for the synthesis of WAY-100635 and O-desmethyl WAY-100635 analogs : (i) Me2S.BH3, THF, 0 ꢁC; (ii) conversion to a LG by several methods (see
Discussion); (iii) nucleophilic fluorination (see Discussion); (iv) KOH, MeOH or EtOH, reflux. (v) thionyl chloride, MeCN, reflux; (vi) WAY-100634, NEt3, MeCN, RT.; (vii, for
O-desmethyl WAY-100634 analogs ) MeOH:NaHCO3:H2O (2:1:1), 50 ꢁC, 2 h.
studied carefully. Electrophilic fluorinations of compounds 2 using
(dimethylamino)sulfur trifluoride (DAST) [29] or Yarovenko [30]
reagent were unsuccessful. Maybe, due to the formation of an
intermediate cation, a rearrangement has taken place leading to
a ring enlargement. For a nucleophilic substitution reaction, the
primary alcohols (2aed) needed to be converted into a good
leaving group (LG). Of all tested leaving groups (LG ¼ tosylate, tri-
flate, iodide) the triflate gave the best results in the subsequent
fluorination step. The triflate compounds 3a, 3c and 3d were
obtained using triflic anhydride [31]. Fluorination with tetrabutyl
ammonium bifluoride (TBABF) [32], TBAF (1 M in THF), TBAF$xH2O
[33], KF [34], KF in crown ether [35] or AgF did not give the desired
products in a reasonable yield. The fluorination of 3a, 3c and 3d
(LG ¼ Triflate) with TBAF$(t-BuOH)4 complex gave 4a, 4c and 4d in
good yields [36,37]. Subsequent saponification gave the corre-
sponding derivatives 5 [21,22].
of 2b with tetrabromomethane and triphenylphoshine followed by
fluorination with tetrabutyl ammonium fluoride (TBAF, 1 M in THF).
This compound was purified byprep-HPLC insteadof normal column
chromatography to avoid rearrangement to the homocubyl deriva-
tives. The collected fractions were directlyconverted into the acid 5b.
Formation of the acid chlorides of 5aed was achieved with
thionyl chloride in MeCN [7]. Acylation of WAY-100634 with the
appropriate acid chloride gave the four analogs 6aed in moderate
to high yields (59e97%). WAY-100634 (Fig. 1) was prepared
according to procedures described by Pike [7] and Cliffe [38]. For
the synthesis of O-desmethyl WAY-100634 the same route was
used. However, it was also prepared by demethylation of WAY-
100634 [39]. Upon reaction of the acid chlorides with O-des-
methyl WAY-100634 a competition was always observed between
N- and O-acylation. Therefore, this compound was first double
acylated after which the O-acyl bond was broken with base
(mixture of MeOH/saturated NaHCO3/H2O; 2:1:1) to afford the
analogs 7aed also in moderate to high yields (32e87%), Scheme 1.
All synthesized analogs were stable in solvents like dime-
thylsulfoxide, ethanol and water for at least several days. Even after
six months in cold (ꢀ20 ꢁC) ethanol no decomposition was detec-
ted, both by HPLC and 1H NMR analysis.
Compound 3b was totally unstable with all candidates of leaving
groups, except the bromide. According to COSY and NOSY NMR
spectra, this compound (3b) seems to rearrange to a homocubyl
product (Scheme 2). Therefore, 4b was synthesized by bromination
O
O
3. The lipophilicity of the investigated compounds
O
O
The lipophilicity (Log D7.4 values) of all compounds was calcu-
lated by using MarvinSketch (http://www.chemaxon.com). Data
are shown in Table 1. The Log D7.4 of the three selected compounds
6b, 6c, and 6d were determined in comparison to WAY-100635. The
Log D7.4 ꢂ S.D. values for 6b, 6c and 6d were 2.94 ꢂ 0.01, 3.37 ꢂ 0.05
and 2.70 ꢂ 0.01, respectively and 3.03 ꢂ 0.04 for WAY-100635.
RO
RO
Scheme 2. Rearrangement of 3b to a homocubyl compound. R ¼ Tosylate, Nosylate,
Triflate.

R. Al Hussainy et al. / European Journal of Medicinal Chemistry 46 (2011) 5728e5735
5731
Table 1
5. Metabolic stability
Calculated Log D7.4 value and affinity for the human 5-HT1A receptor of WAY-100635
analogs .
The metabolic stability study was performed using radiolabeled
[
18F]6b (see Supplementary data) and was compared with [18F]p-
N
MPPF [19]. The preliminary study showed that incubation of [18F]6b
with human hepatocytes demonstrated a better metabolic stability
than [18F]p-MPPF. Results are depicted in Table 3.
OR
N
N
BFR
N
O
F
6. Discussion
a
Compd
R
BFR
cLog D7.4 IC50
(nM)b
5-HT1A
Ki (nM)c
Similar to central nervous system drug discovery, PET radio-
tracers should be developed to have high affinity and selectivity for
the target protein. To reliably visualize the 5-HT1A receptor, which
exist in human brain at concentrations up to 100 fmol/mg protein
(equivalent to 10 pmol/mL or 10 nM), a radioligand should have
a KD value in the low or (sub)nanomolar range [5]. Almost without
exception all PET radiotracers for the 5-HT1A such us [18F]MPPF and
WAY-100635
CH3
e
4.08
4.48
2.04d
4.37
0.91 ꢂ 0.43 0.39
6a
6b
6c
6d
CH3 Adamantyl
CH3 Cubyl
CH3 Bicyclo[2.2.2]octyl
4.27 ꢂ 0.15 1.82
1.27 ꢂ 0.10 0.54
1.53 ꢂ 0.01 0.65
2.02 ꢂ 0.02 0.86
CH3 Bicyclo[2.2.1]heptyl 3.93
O-desmethyl
H
e
3.93
0.77 ꢂ 0.36 0.33
[
18F]FCWAY were sharing the same WAY-100634 moiety to guar-
WAYꢀ100635
7a
7b
7c
7d
H
H
H
H
Adamantyl
Cubyl
Bicyclo[2.2.2]octyl
4.33
1.89d
4.22
2.59 ꢂ 0.20 1.10
1.35 ꢂ 0.04 0.58
1.76 ꢂ 0.03 0.75
1.74 ꢂ 0.06 0.74
antee this high affinity as well as selectivity. For imaging purposes
the letter can be expressed as the ratio of Bmax/KD for the target
receptor over that of the other receptors. This ratio should be as
high as possible. Translated to in vitro this means in practice that
the Ki for the receptor of interest is preferably a factor of 100 lower
than for the others with a comparable density. For a low-density
non-target receptor this factor may of course be lower.
Bicyclo[2.2.1]heptyl 3.78
a
Calculated Log D7.4 using MarvinSketch (http://www.chemaxon.com).
Results are presented as concentration producing 50% inhibition (IC50) in nM
b
and the mean of three experiments per drug ꢂ SD, (n ¼ 3).
c
Ki values were calculated as Ki ¼ IC50/(1 þ C/KD), with C ¼ 0.5 nM, KD ¼ 1.48 nM
[40].
In this study, we were conscious to design and synthesize
fluorinated compounds (Fig. 2) that have a similar or even higher
affinity and a similar or even better selectivity than WAY-100635,
but we mainly focussed our attention on the metabolic stability
by using BFR groups. Defluorination is a major metabolic problem
for some 18F-labeled radiotracers. Unfortunately, there is no specific
rule to enhance the metabolic stability and to prevent HF elimi-
nation. Species (rats, monkeys and human) differences in the ability
to radiodefluorination exist for many radiotracers [21]. 18F-labeling
at an aromatic carbon atom within a phenyl or pyridinyl group
showed less to no defluorination such as in [18F]MPPF [12,14,21],
while 18F-labeling at an aliphatic carbon atom is often prone to
defluorination such as in [18F]FCWAY [17,21]. This is not unexpected
since a fluorine-carbon bond tends to an axial position due to
fluorine’s electronegativity, which in turn facilitates elimination.
For this reason, the corresponding cis-compound, in which fluorine
is always axial, defluorinates even much faster. It is interesting to
note that by placing fluorine on a primary carbon as in [18F]MeF-
WAY, the defluorination rate was greatly reduced [18].
d
Probably, these values were not correctly calculated, due to the presence of the
cubyl moiety.
4. Receptor binding in vitro
4.1. 5-HT1A receptor binding
Radioligand binding experiments were performed using the
5-HT1A agonist [3H]8-OH-DPATand a membrane suspension of cells
expressing the cloned human 5-HT1A receptor. Inhibition curves
were measured for the eight compounds and the concentration
producing 50% inhibition (IC50) for each were derived from non-
linear regression curve fitting and apparent equilibrium inhibi-
tion constant (Ki) values were calculated according to the
ChengePrusoff equation [40]. Data are shown in Table 1. All
compounds revealed affinity for the 5-HT1A receptor in the low
nanomolar range.
4.2. Receptor binding profile
A variety of bridgehead WAY-100635 derivatives (6aed and
7aed) with a fluoromethyl-BFR, such as -adamantyl, -cubyl,
-bicyclo[2.2.2]octyl and -bicyclo[2.2.1]heptyl, attached to the car-
boxamide were synthesized in moderate to good yields. The
attachment of a eCH2F moiety on the bridgehead carbon is
expected to hardly influence the biological activity, but has the
advantage that HF elimination is no longer possible. As also the
The selectivity for the 5-HT1A receptor over other relevant
receptors of three promising WAY-100635 analogs for PET, 6bed,
was determined by NIMH/Psychoactive Drug Screening Program
(PDSP). The results are shown in Table 2. A typical inhibition curve
is shown in Fig. 3.
Table 2
Affinity (Ki value) of WAY-100635 analogs 6b, 6c and 6d for selected receptors.a,b
Compd
5-HT1A
5-HT2B
5-HT7
5-HT1D
a1A-A
a1B-A
a1D-A
D4
WAY-100635
0.6
0.2
1.1
0.9
24c
99
79
88
>10000c
132
>10000c
122
20c
16
74
36
322c
176
110
38
5c
27
73
19
16c
71
84
92
6b
6c
6d
77
97
361
92
a
The primary testing of the compounds on a number of receptors, (serotonin receptor subtypes: 5-HT1E, 5-HT2A, 5-HT2C, 5-HT3, 5-HT6; adrenoceptor subtypes: a2A
1R, 2R) showed <50% inhibition at 10 M. On other receptors (serotonin receptor subtypes: 5-HT1B, 5-HT1D, 5-HT5A; adrenoceptor
b3 and dopamine receptor subtype D2) where the compounds showed >50% inhibition at 10 M the Ki values were at least a factor of 100 greater than
those for the 5-HT1A receptor.
, a2B;
serotonin transporter sigma receptor:
subtypes: a2C b1 b2
s
s
m
,
,
,
m
b
Results are presented as apparent equilibrium inhibition constant (Ki) in nM and are the mean of three experiments.
Data taken from NIMH-PDSP [http://pdsp.med.unc.edu/pdsp.php].
c

5732
R. Al Hussainy et al. / European Journal of Medicinal Chemistry 46 (2011) 5728e5735
Screening Program (PDSP). The PDSP labs have used the same
human (HEK-293 EBNA)-cells but other practical conditions, which
explains the differences in Ki values between Table 1 and Table 2. For
in vivo application, compound 6c might show a better selectivity
than WAY-100635 because of its lower binding affinities for a1A
-
adrenoceptors receptor and the D4 receptor. Compounds 6b and 6d
might also show a better in vivo selectivity than WAY-100635
because of their lower binding affinities for the D4 receptor. On the
other hand, 6b, 6c and 6d show higher affinity for the subreceptors
5-HT7 and 5-HT1D than WAY-100635 itself, albeit to be over 30e200
times lower than the affinity for the 5-HT1A receptor.
A metabolic stability study with human hepatocytes was per-
formed using radiolabeled [18F]6b and compared with that of [18F]
p-MPPF. This preliminary study showed that at 15 min of incuba-
tion >73% of the parent compound [18F]6b was still present,
whereas under the same conditions [18F]MPPF was more rapidly
metabolized as only 45% was left. This better metabolic stability of
[
18F]6b compared to [18F]MPPF must for a large part be due to the
expected decrease in rate of the amide hydrolysis. In vivo studies
with the radiolabeled compounds of 6 are planned to give a deci-
sive answer.
7. Conclusions
Fig. 3. The inhibition curves of 6b and WAY-100635 using the 5-HT1A receptor agonist
[3H]8-OH-DPAT.
Eight novel fluoromethyl-BFR analogs of WAY-100635 were
synthesized successfully. They all bind to the 5-HT1A receptor with
high affinity (Ki values in the (sub)nanomolar range). The binding
affinity and selectivity of the three selected analogs 6b, 6c and 6d
and preliminary stability of 6b make them worthwhile for further
in vivo investigation as potential 5-HT1A receptor binding ligands
and promising 18F-labeled PET tracers.
carboxamide function is at a bridgehead carbon, the proposed
molecules have the advantage that no enantiomeric mixtures are
formed and that the amide bond is more stable toward in vivo
hydrolysis. Hence, the novel analogs are aliphatic, a-chiral and
expected to be stable toward HF elimination.
The eight fluoromethyl WAY-100635 analogs have been tested
in an in vitro radioligand binding assay to determine their affinity
for 5-HT1A receptors. All compounds showed a similar high affinity
for the 5-HT1A receptor as that of the parent compounds WAY-
100635 and O-desmethyl WAY-100635. The presence of
a hydroxyl group instead of the methoxy group in the aromatic
moiety does not lead to a substantial better in vitro binding to the 5-
HT1A receptor or a real decrease in cLog D7.4. Since on the other
hand, this will complicate the radiolabeling with 18F (the hydroxyl
group must be protected before and deprotected after the fluori-
nation step), we excluded compounds 7 for further extensive
evaluation.
Of the other fluoromethyl-BFR analogs of WAY-100635,
compound 6a was found to have the lowest affinity for the 5-
HT1A receptor and the highest lipophilicity (cLog D7.4 ¼ 4.48),
confirmed by a tR on HPLC (see Experimental) which was more than
twice that of the other compounds, which might be too high for
in vivo application. Therefore, we further focused our attention on
the BFR analogs of WAY-100635, 6bed. The Log D7.4 of 6bed and of
WAY-100635 for comparison was determined experimentally. The
lipophilicity of all three compounds was comparable to that of
WAY-100635. These measured Log D7.4 values differ from the
calculated values, but are now within the optimal range (2e3.5).
Furthermore, it confirmed our assumption that the calculating
software is not capable for handling a cubane structure.
8. Experimental section
8.1. Chemistry
The solvents were dried according to standard procedures.
Reactions involving moisture sensitive compounds were performed
under an anhydrous atmosphere of dry argon, unless indicated
otherwise. Commercially available chemicals were used without
further purification. Reactions were monitored by using thin-layer
chromatography (TLC) on silica-coated plastic sheets (Merck silica
gel 60 F254) with the indicated eluent. The compounds were visu-
alized by UV light (254 nm), I2, Ceric Ammonium Molybdate (CAM),
Bromocrysol or potassium permanganate staining, followed by
charring at 130 ꢁC. Flash chromatography refers to purification using
the indicated eluent and Acros silica gel (0.030e0.075 mm). All other
reagents were purchased from Fluka, SigmaeAldrich, and Acros.
Nuclear magnetic resonance spectra (1H NMR and 13C NMR) were
determined in the indicated solvent using a Bruker AC 200 MHz
(200.13 MHz and 50.32 MHz, respectively), a Bruker Avance
250 MHz (250.13 and 62.90 MHz, respectively) or a Bruker MSL
400 MHz (400.13 and 100.61 MHz, respectively). Chemical shifts (d)
are given in ppm downfield from tetramethylsilane (1H, 13C) and
coupling constants J in Hz. Mass spectrawere recorded on a Finnigan
MAT-90 mass spectrometer. Products were detected at
All tested compounds have a purity of at least 95% according to NMR
and HPLC analysis. Analytical HPLC was performed on a C18 e 10
column (kromasil, 4.6 mm ꢃ 250 mm), using MeOH/H2O/DIPA (65/
35/0.02) as eluent and a flow of 1 mL minꢀ1
l
¼ 254 nm.
The selectivity of 6bed for the 5-HT1A receptor over other rele-
vant receptors was determined at NIMH/Psychoactive Drug
mm
.
Table 3
Stability of [18F]6b and [18F]MPPF in a medium with human hepatocytes.
8.1.1. General procedure for the synthesis of compounds 6a, 6b, 6c
and 6d
In a flame dried flask under argon atmosphere, compound 5 was
dissolved in MeCN and thionyl chloride (1 equivalent) was added.
t ¼ 0 min
99.6%
99.2%
t ¼ 15 min
73.2%
45.2%
t ¼ 60 min
30.7%
22.7%
t ¼ 150 min
17.4%
8.4%
[
[
18F]6b
18F]MPPF

R. Al Hussainy et al. / European Journal of Medicinal Chemistry 46 (2011) 5728e5735
5733
The reaction mixture was refluxed at 70 ꢁC in MeCN for 45 min to
1 h and the excess of thionyl chloride was totally evaporated at
room temperature (because of volatility of the acid chlorides)
under reduced pressure, then co-evaporated with MeCN. After-
ward, WAY-100634 (1 equivalent) in MeCN and NEt3 (2 equiva-
lents) were added to the corresponding acid chloride and stirred at
room temperature for 1 h. The solvent was evaporated and the
residue was dissolved in water and extracted with CH2Cl2. The
organic phase was collected, dried over anhydrous Na2SO4 and
evaporated to dryness.
J ¼ 48 Hz, 2H, CH2F), 6.78e7.08 (m, 4H, 4ꢃ CH), 7.20e7.40 (m, 2H,
2ꢃ CH), 7.75 (dt, J ¼ 7.4 Hz, 1H, CH), 8.52 (dd, J ¼ 4.9 Hz, 1H, CH). 13C
NMR (50.32 MHz, CDCl3):
d 31.42, 31.50, 34.15, 44.77, 44.85, 46.98,
47.29, 50.36, 53.24, 55.16, 55.48, 55.69, 84.92, 88,53, 110.98, 117.93,
120.75, 122.72, 122.89, 123.56, 137.97, 141.05, 148.98, 152.03, 155.53,
175.42. HRMS (EI) m/z calcd for C27H35FN4O2; 466.2744; found:
467.2811 (M þ H), tR: 16.6 min.
8.1.2. General procedure for synthesis of compounds 7a, 7b, 7c and
7d
The acid chlorides of 5aed were synthesized as mentioned
above. Then O-desmethyl WAY-100634 (0.5 equivalent) and NEt3
(2 equivalents) were dissolved in MeCN and added to the corre-
sponding acid chloride. This new reaction mixture was stirred at
room temperature for 30 min. Then the solvent was evaporated, the
residue was dissolved in a mixture of MeOH/saturated NaHCO3/
H2O (2/1/1) and heated at 50 ꢁC for 2 h, unless stated otherwise. The
solution was evaporated and the residue was dissolved in water,
extracted with EtOAc, unless noted otherwise, dried over Na2SO4,
filtered and evaporated to dryness.
8.1.1.1. 4-(Fluoromethyl)-N-(2-(4-(2-methoxyphenyl)piperazin-1-yl)
ethyl)-N-(pyridin-2-yl)adamantane-1-carboxamide
(6a). Starting
with 5a (50 mg, 0.24 mmol) and WAY-100634 (73.8 mg, 0.24 mmol)
gave 116 mg (95%) of 6a as colorless glass after column chroma-
tography (EtOAc/NEt3, 1/0.01). 1H NMR (400.13 MHz, CDCl3):
d
1.28e1.44 (m, 4H, 2ꢃ CH2), 1.48e1.55 (m, 4H, 2ꢃ CH2), 1.67 (q,
J ¼ 12 Hz, 4H, 2ꢃ CH2), 1.95 (s, 2H, 2ꢃ CH), 2.64 (m, 6H, N(CH2)3),
2.90e3.10 (m, 4H, N(CH2)2), 3.80e3.86 (m, 5H, CH3, NCH2), 3.84 (d,
J ¼ 49 Hz, 2H, CH2F), 6.81e7.01 (m, 4H, 4ꢃ CH), 7.23e7.37 (m, 2H,
2ꢃ CH), 7.75 (dt, J ¼ 7.6 Hz, 1H, CH), 8.50 (d, J ¼ 4.9 Hz, 1H, CH). 13C
NMR (100.61 MHz, CDCl3):
d
28.07, 34.62, 34.79, 35.74, 37.02, 37.06,
8.1.2.1. 4-(Fluoromethyl)-N-(2-(4-(2-hydroxyphenyl)piperazin-1-yl)
39.34, 40.56, 40.61, 44.27, 49.06, 50.61, 53.51, 55.33, 55.76, 91.33,
93.05, 111.25, 118.08, 120.94, 122.79, 123.13, 126.11, 138.23, 141.35,
148.85,152.24,156.88,178.13. HRMS (EI) m/z calcd for C30H39FN4O2;
506.3057; found: 507.3127 (M þ H), tR: 48.8 min.
ethyl)-N-(pyridin-2-yl)adamantane-1-carboxamide
(7a). Starting
with 5a (50 mg, 0.24 mmol) and O-desmethyl WAY-100634 (35 mg,
0.12 mmol) in MeCN (4 mL) gave after column chromatography
(EtOAc/Hexane, 5/2) 47 mg (80%) of 7a as colorless glass. 1H NMR
(400.13 MHz, CDCl3):
d
1.30e1.80 (m, 12H, 6ꢃ CH2), 2.09 (s, 2H, 2ꢃ
8.1.1.2. 4-(Fluoromethyl)-N-(2-(4-(2-methoxyphenyl)piperazin-1-yl)
ethyl)-N-(pyridin-2-yl)cubane-1-carboxamide (6b). Starting with
5b (300 mg, 1.09 mmol) and WAY-100634 (346 mg, 1.11 mmol)
gave 441 mg (84%) of 6b as colorless glass after column chroma-
tography (EtOAc/NEt3, 1/0.01). 1H NMR (200.13 MHz, CDCl3):
CH), 2.60e2.72 (m, 6H, N(CH2)3), 2.79e2.89 (m, 4H, N(CH2)2),
3.83e3.90 (m, 2H, N(CH2)), 3.86 (d, J ¼ 50 Hz, 2H, CH2F), 6.72e6.89
(m, 2H, 2ꢃ CH), 6.94e7.08 (m, 2H, 2ꢃ C), 7.17e7.25 (m, 2H, 2ꢃ CH),
7.70 (dt, J ¼ 7.6 Hz, 1H, CH), 8.44 (d, J ¼ 4.9 Hz, 1H, CH). 13C NMR
(62.90 MHz, CDCl3):
d 28.10, 34.61, 34.89, 35.91, 37.21, 37.28, 39.39,
d
2.46e3.20 (m, 10H, N(CH2)3 and N(CH2)2), 3.51e3.85 (m, 9H, 6ꢃ
40.62, 40.70, 44.33, 49.10, 52.54, 53.98, 55.73, 90.86, 93.59, 114.04,
119.99, 121.34, 122.85, 123.02, 126.39, 138.20, 139.02, 148.95, 151.52,
156.90, 178.27. HRMS (EI) m/z calcd for C29H37FN4O2; 492.2901;
found: 493.297 (M þ H), tR: 24.9 min.
CH and CH3), 3.90e4.12 (m, 2H, NCH2), 4.40 (d, J ¼ 48 Hz, 2H,
CH2F), 6.70e7.05 (m, 4H, 4ꢃ CH), 7.15e7.30 (m, 2H, 2ꢃ CH), 7.60
(dt, J ¼ 7.6 Hz, 1H, CH), 8.50 (dd, J ¼ 4.9 Hz, 1H, CH). 13C NMR
(50.32 MHz, CDCl3):
d 43.63, 43.83, 44.98, 47.34, 50.42, 53.18,
54.55, 54.96, 55.14, 56.01, 59.34, 81.13, 84.41, 110.96, 117.87, 119.93,
120.72, 121.61, 122.65, 137.96, 141.09, 148.79, 152.01, 155.23, 171.49.
HRMS (EI) m/z calcd for C28H31FN4O2; 474.2431; found: 475.2505
(M þ H), tR: 15.8 min.
8.1.2.2. 4-(Fluoromethyl)-N-(2-(4-(2-hydroxyphenyl)piperazin-1-yl)
ethyl)-N-(pyridin-2-yl)cubane-1-carboxamide (7b). Starting with 5b
(50 mg, 0.27 mmol) and O-desmethyl WAY-100634 (41 mg,
0.14 mmol) in MeCN (4 mL) gave after column chromatography
(EtOAc/MeOH,1/0.1 to only EtOAc) 38 mg (59%) of 7b as white solid.
8.1.1.3. 4-(Fluoromethyl)-N-(2-(4-(2-methoxyphenyl)piperazin-1-yl)
ethyl)-N-(pyridin-2-yl)bicyclo[2.2.2]octane-1-carboxamide
1H NMR (400.13 MHz, CDCl3):
d 2.50e2.90 (m, 10H, N(CH2)3 and
N(CH2)2), 3.60e3.89 (m, 6H, 6ꢃ CH), 4.05e4.12 (m, 2H, NCH2), 4.46
(d, J ¼ 4.9 Hz, 2H, CH2F), 6.80e7.30 (m, 6H, 6ꢃ CH), 7.81 (dt,
J ¼ 7.6 Hz, 1H, CH), 8.54 (d, J ¼ 4.9 Hz, 1H, CH). 13C NMR
(6c). Starting with 5c (25 mg, 0.13 mmol) and WAY-100634 (42 mg,
0.13 mmol) gave 38 mg (61%) of 6c after column chromatography
(EtOAc/MeOH, 1/0.07). 1H NMR (400.13 MHz, CDCl3):
d
1.23e1.40
(100.61 MHz, CDCl3): d 43.91, 44.01, 45.17, 47.58, 52.53, 53.81, 54.91,
(m, 6H, 3ꢃ CH2), 1.58e1.73 (m, 6H, 3ꢃ CH3), 2.60e2.70 (m, 6H,
N(CH2)3), 3.03 (bs, 4H, N(CH2)2), 3.82e3.91 (m, 5H, CH3 and NCH2),
3.93 (d, J ¼ 49 Hz, 2H, CH2F), 6.83e7.02 (m, 4H, 4ꢃ CH), 7.24e7.32
(m, 2H, 2ꢃ CH), 7.77 (dt, J ¼ 7.6 Hz, 1H, CH), 8.52 (d, J ¼ 4.9 Hz, 1H,
55.11, 56.16, 59.58, 82.15, 83.79, 114.04, 119.96, 121.28, 121.75,
126.37, 138.11, 138.98, 140.46, 149.07, 151.49, 155.48, 171.75. HRMS
(EI) m/z calcd for C27H29FN4O2; 460.2275; found: 461.2344 (M þ H),
tR: 11.6 min.
CH). 13C NMR (100.61 MHz, CDCl3):
d 26.95, 27.00, 28.60, 32.28,
32.46, 42.44, 48.88, 50.58 53.48, 55.35, 55.71, 89.85, 91.54, 111.27,
118.11, 120.96, 122.66, 122.82, 123.44, 138.20, 141.36, 149.00, 152.27,
156.88, 178.05. HRMS (EI) m/z calcd for C28H37FN4O2; 480.2901;
found: 481.2967 (M þ H), tR: 23.6 min.
8.1.2.3. 4-(Fluoromethyl)-N-(2-(4-(2-hydroxyphenyl)piperazin-1-yl)
ethyl)-N-(pyridin-2-yl)bicyclo[2.2.2]octane-1-carboxamide
(7c). Starting with 5c (50 mg, 0.27 mmol) and O-desmethyl WAY-
100634 (35 mg, 0.13 mmol) in MeCN (4 mL) gave after column
chromatography (EtOAc) 20 mg (33%) of 7c as colorless glass. 1H
8.1.1.4. 4-(Fluoromethyl)-N-(2-(4-(2-methoxyphenyl)piperazin-1-yl)
ethyl)-N-(pyridin-2-yl)bicyclo[2.2.1]heptane-1-carboxamide
(6d). Starting with 5d (50 mg, 0.29 mmol) and WAY-100634
(90.48 mg, 0.29 mmol) gave 88 mg (65%) of 6c after column chro-
matography (EtOAc/MeOH, 1/0.07). 1H NMR (250.13 MHz, CDCl3):
NMR (400.13 MHz, CDCl3):
d
1.20e1.38 (m, 6H, 3ꢃ CH2), 1.59e1.74
(m, 6H, 3ꢃ CH2), 2.49e2.72 (m, 6H, N(CH2)3), 2.74e2.90 (m, 4H,
N(CH2)2), 3.83e3.90 (m, 2H, CH2), 4.00 (d, J ¼ 49 Hz, 2H, CH2F),
6.79e7.17 (m, 4H, 4ꢃ CH), 7.23e7.31 (m, 2H, 2ꢃ CH), 7.77 (dt,
J ¼ 7.6 Hz, 1H, CH), 8.50 (d, J ¼ 4.9 Hz, 1H, CH). 13C NMR
d
1.10e1.40 (m, 4H, 2ꢃ CH2), 1.40e1.70 (m, 4H, 2ꢃ CH2), 1.78e1.98
(100.61 MHz, CDCl3): d 26.95, 26.99, 28.62, 32.29, 32.47, 42.47,
(m, 2H, CH2), 2.50e2.75 (m, 6H, N(CH2)3), 2.80e3.10 (m, 4H,
N(CH2)2), 3.88 (s, 3H, CH3), 3.90e3.40 (m, 2H, NCH2), 4.33 (d,
48.94, 52.32, 53.95, 55.65, 89.83, 91.53, 114.02, 119.97, 121.34,
122.71, 123.31, 126.42, 138.26, 138.96, 149.08, 151.51, 156.86, 178.06.

5734
R. Al Hussainy et al. / European Journal of Medicinal Chemistry 46 (2011) 5728e5735
HRMS (EI) m/z calcd for C27H35FN4O2; 466.2744; found: 467.2817
(M þ H), tR: 16.5 min.
Radioactivity was quantified (vide supra). IC50 values were derived
from non-linear regression curve fitting using computer program
Graphpad PrismÒ (version 4.00) and Ki values were calculated
according to ChengePrusoff equation using the measured KD of
[3H]8-OH-DPAT radioligand.
8.1.2.4. 4-(Fluoromethyl)-N-(2-(4-(2-hydroxyphenyl)piperazin-1-yl)
ethyl)-N-(pyridin-2-yl)bicyclo[2.2.1]heptane-1-carboxamide
(7d). Starting with 5d (50 mg, 0.19 mmol) and O-desmethyl WAY-
100634 (28 mg, 0.09 mmol) in MeCN (5 mL) and subsequent
hydrolysis in a mixture of MeOH/saturated NaHCO3/H2O (4/1/1),
the solvent was evaporated and the residue was extracted with
CH2Cl2 (3 ꢃ 1 mL). Column chromatography (CH2Cl2/MeOH, 1/
0.025) afforded 7d (20 mg, 49%). 1H NMR (250.13 MHz, CDCl3):
8.2.2. Receptor binding profile
The receptor binding profile of compound 6b, 6c and 6d was
investigated by radioligand binding by NIMH/psychoactive Drug
Screening Program (PDSP). For assay conditions and assay protocol
see http://pdsp.med.unc.edu/pdspw/binding.php.
d
1.08e1.35 (m, 4H, 2ꢃ CH2), 1.40e1.66 (m, 4H, 2ꢃ CH2), 1.75e2.00
(m, 2H, CH2), 2.48e2.90 (m, 10H, N(CH2)3 and N(CH2)2), 3.83e4.00
(m, 2H, NCH2), 4.34 (d, J ¼ 48 Hz, 2H, CH2F), 6.77e7.33 (m, 6H, 6ꢃ
CH), 7.79 (dt, J ¼ 7.6 Hz,1H, CH), 8.54 (d, J ¼ 4.9 Hz,1H, CH). 13C NMR
8.3. Lipophilicity of the investigated compounds
The lipophilicity of the investigated compounds was deter-
mined by measuring the partition of the compounds between
1-octanol and water (buffer), and is expressed as the log D7.4 value.
The investigated compounds were dissolved in 1-octanol (3 mL) in
a concentration of 1 mg/mL. The phosphate buffer (25 mM,
pH ¼ 7.4) was made of 1.160 g of NaH2PO4 and 2.175 g of NaHPO4 in
1 L water. The HPLC eluent of a mixture of methanol: phosphate
buffer (80:20) was used for further dilutions.
(100.61 MHz, CDCl3):
d 31.56, 31.59, 34.32, 44.92, 44.95, 47.25,
47.40, 53.73, 55.38, 55.80, 59.05, 86.00, 87.32, 114.12, 120.07, 121.40,
123.19, 123.51, 126.57, 138.31, 141.09, 149.21, 151.36, 155.53, 175.48.
HRMS (EI) m/z calcd for C26H33FN4O2; 452.2588; found: 453.2658
(M þ H), tR: 11.7 min.
8.2. In vitro receptor binding
The HPLC samples were made as follows: the same amounts of
the buffer and the 1-octanol stock (250 mL) were added to each
other in a mixing vial and the mixture was shaken for 1 min. After
resting for 30 min, a sample was taken from the 1-octanol layer
Materials: [3H]8-OH-DPAT (154.2 Ci/mmol, 5.71 TBq/mmol),
membrane preparations of HEK-293 EBNA-cells expressing the
cloned human serotonin 5-HT1A receptor (HEK-293 EBNA-cells, 1
unit ¼ 2.5
mL), the 96-well microplates and an UniFilter-96 GF/C
(50
Another 50
with 450 L of the HPLC eluent. This mixture was again diluted in
the same previous manner and 50 L of this sample was injected
into the HPLC. A 50 L sample of the buffer layer was also injected
mL) and diluted with 450
mL of the HPLC eluent in a new vial.
filter plate were obtained from Perkin Elmer Life Sciences (Well-
seley, MA). Tris-(hydroxymethyl)-aminomethane was purchased
from AppliChem (Darmstadt, Germany). HCl and MgSO4 were
purchased from SigmaeAldrich (Zwijndrecht, The Netherlands).
The filtration was carried out with a MicroBeta FilterMate-96
Harvester (PerkinElmer) and the radioactivity was quantified
using a Wallac MicroBeta TriLux counter (PerkinElmer).
mL was taken from this new mixture and diluted again
m
m
m
into the HPLC. This whole procedure was repeated five times.
As a reference, the partition of WAY-100635 between 1-octanol
and water (buffer) was measured in a similar way. The log D7.4 was
calculated according to the formula: The log D7.4
¼
10Log(Aoct
/
8.2.1. 5-HT1A receptor binding
Abuffer), with Aoct ¼ average concentration of the five octanol
samples and Abuffer ¼ average concentration of the five buffer
samples.
Binding experiments were performed using the 5-HT1A agonist
[3H]8-OH-DPAT and a cell membrane suspension containing the
human 5-HT1A receptor. A quality control of the radioligand and the
membrane preparation was done by performing radioligand satu-
ration binding experiments. The membrane suspension was diluted
(factor 1:40) in binding buffer (50 mM TriseHCl, pH 7.4, 5 mM
MgSO4) and incubated in the dark on 96-well microplates for 1 h at
37 ꢁC with the following radioligand concentrations: 0.25, 0.5, 0.75,
1.00, 2.00, 2.50, 3.00, 4.00, 5.00, 6.00, 8.00, and 10.00 nM in
8.4. Stability toward human hepatocytes
Approximately 4 million of cryopreserved human liver cells (In
Vitro Technologies, Leipzig, Germany) were thawed rapidly and
suspended in RPMI 1640 supplemented with 10% fetal bovine
serum and 1% penicillin/streptomycin. Cells were maintained at 5%
CO2 humidified atmosphere and at 37 ꢁC, and washed with phos-
phate buffered saline (PBS). Viability was determined by trypan
blue exclusion. After centrifugation, cells were washed and dis-
solved in the same medium as above. The viable cell concentration
was adjusted to 1.0 million per mL. To 1 mL of this cell suspension
a volume of 200
mL. Non-specific binding was measured at each
radioligand concentration in the presence of 10
mM WAY-100635.
Incubation was rapidly terminated by filtration over an UniFilter-
96 GF/C filter plate presoaked in 0.3% polyethyleneimine (PEI),
followed by three rapid washes with ice-cold washing buffer
(50 mM TriseHCl, pH 7.4, 5 mM MgSO4) using a MicroBeta
FilterMate-96 Harvester. Filter plates were dried for 2 h in an oven
75
incubation for 15, 60 and 150 min at 37 ꢁC, a 200
taken, added to 200 L of MeOH, sonicated and centrifuged. The
m
L of [18F]6b (50 MBq) or [18F]MPPF (50 MBq) was added. After
L sample was
m
at 54 ꢁC. Radioactivity was quantified with 10
mL of liquid scintil-
m
lation fluid using a Wallac MicroBeta TriLux counter. The measured
KD value for the 5-HT1A receptor in this membrane preparation was
comparable to the one provided by the manufacturer (1.48 nM and
1.86 nM, respectively).
composition of the supernatant was determined using HPLC
combined with online radioactivity detection.
For the competitive binding experiments, an assay volume of
Acknowledgment
200
membrane suspension, 50
ligand dissolved in binding buffer, 10
increasing concentrations (range 10ꢀ11 to 10ꢀ6 M) and 40
binding buffer. Incubation in the 96-well microplates was carried
out in the dark for 2 h at 37 ꢁC. Incubation was rapidly terminated
(vide supra). Filter plates were dried overnight in an oven at 54 ꢁC.
m
L per well was used containing 100
L of 2 nM of [3H]8-OH-DPAT radio-
L of each novel analog in
L of the
mL of the diluted cell
m
The authors would like to acknowledge the Dutch Technology
Foundation STW for the financial support of the project VBC.7048,
BV Cyclotron (VU University) for the supply of fluorine-18, Covidien
Mallinckrodt Int. and Philips for financial support and Dr. Bryan
Roth of the NIMH-PDSP program for performing the receptor
profiling of compounds 6b, 6c and 6d. The authors also would like
m
m

R. Al Hussainy et al. / European Journal of Medicinal Chemistry 46 (2011) 5728e5735
5735
[18] N. Saigal, R. Pichik, B. Easwaramoorthy, D. Collins, B.T. Christian, B. Shi,
T.K. Narayanan, S.G. Potkin, J. Mukherjee, Synthesis and biological evaluation
of novel serotonin 5-HT1A receptor radioligand, 18F-labeled mefway, in
rodents and imaging by PET in nonhuman primate, J. Nucl. Med. 47 (2006)
1697e1706.
[19] R. Al Hussainy, J. Verbeek, D. van der Born, A.H. Braker, J.E. Leysen, R.J. Knol,
J. Booij, J.D.M. Herscheid, Design, synthesis, radiolabeling, and in vitro and
in vivo evaluation of bridgehead iodinated analogues of N-{2-[4-(2-
methoxyphenyl)piperazin-1-yl]ethyl}-N-(pyridin-2-yl)cyclohexanecarboxa-
mide (WAY-100635) as potential SPECT ligands for the 5-HT1A receptor,
J. Med. Chem. 54 (2011) 3480e3491.
[20] A.A. Wilson, T. Inaba, N. Fischer, L.M. Dixon, J. Nobrega, J.N. DaSilva, S. Houle,
Analogues of WAY 100635 as potential imaging agents for 5-HT1A receptors:
syntheses, radiosyntheses, and in vitro and in vivo evaluation, Nucl. Med. Biol.
25 (1998) 769e776.
[21] V.W. Pike, PET radiotracers: crossing the bloodebrain barrier and surviving
metabolism, Trends Pharmacol. Sci. 30 (2009) 431e440.
[22] P.E. Eaton, N. Nordari, J. Tsanaktsidis, S.P. Upadhyaya, Barton decarboxylation
of cubane-1,4-dicarboxylic acid: optimized procedures for cubanecarboxylic
acid and cubane, Synthesis (1995) 501e502.
[23] J.D. Roberts, W.T. Moreland, W. Frazer, Synthesis of some 4-substituted
bicyclo[2.2.2]octane-1-carboxylic acids, J. Am. Chem. Soc. 75 (1953) 637e640.
[24] L.G. Humber, G. Myers, L. Hawkins, C. Schmidt, M. Boulerice, Agents affecting
lipid metabolism XIII. The synthesis of 1,4-disubstituted bicyclo[2.2.2]octane
derivatives, Can. J. Chem. 42 (1964) 2852e2860.
to acknowledge Prof. Josée E. Leysen for advice regarding the
in vitro 5-HT1A receptor binding affinity measurements.
Appendix. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.ejmech.2011.06.023.
References
[1] B.R. Chemel, B.L. Roth, B. Armbruster, V.J. Watts, D.E. Nichols, WAY-100635 is
a potent dopamine D4 receptor agonist, Psychopharmacology 188 (2006)
244e251.
[2] J. Passchier, A. Van Waarde, Visualisation of serotonin-1A (5-HT1A) receptors
in the central nervous system. Eur, J. Nucl. Med. 28 (2001) 113e129.
[3] S. Matsubara, R.C. Arora, H.Y. Meltzer, Serotonergic measures in suicide brain e
5-HT1A binding-sites in frontal cortex of suicide victims, J. Neural Transm. 85
(1991) 181e194.
[4] V. Kepe, J.R. Barrio, S. Cheng Huang, L. Ercoli, P. Siddarth, K. Shoghi Jadid,
G.M. Cole, N. Satyamurthy, J.L. Cummings, G.W. Small, M.E. Phelps, . Serotonin
1A receptors in the living brain of Alzheimer’s disease patients, PNAS 103
(2006) 702e707.
[25] R. Mozingo, H. Adkins, L. Richards, Catalyst, Raney Nickel, W-2, Org. Syn. Coll.
3 (1955) 181 21 (1941) 15.
[26] C.N. Sukenik, S.S. Wang, K. Kumar, Synthesis, characterization, and chemistry
of bridgehead-functionalized bicyclo[2.2.2]octanes: reactions at neopentyl
sites, J. Org. Chem. 49 (1984) 665e670.
[5] V.W. Pike, C. Halldin, H.V. Wikström, Radioligands for the study of the brain
5-HT1A receptors in vivo, Prog. Med. Chem. 38 (2001) 189e247.
[6] F. Yasuno, T. Suhara, T. Ichimiya, A. Takano, T. Ando, Y. Okubo, Decreased
5-HT1A receptor binding in amygdala of schizophrenia, Biol. Psychiatry 55
(2004) 439e444.
[27] E.W. Della, J. Tsanaktsidis, Synthesis of bridgehead-bridgehead substituted
bicycloalkanes, Aust. J. Chem. 38 (1985) 1705e1718.
[28] R. Priefer, P.G. Farrell, D.N. Harpp, Effective synthetic routes to cubylcarbinol
derivatives, Synthesis (2002) 2671e2673.
[7] V.W. Pike, J.A. McCarron, S.P. Hume, S. Ashworth, J. Opacka-Juffry, S. Osman,
A.A. Lammertsma, K.G. Poole, A. Fletcher, A.C. White, I.A. Cliffe, Pre-clinical
development of a radioligand for studies of central 5-HT-1A receptors in vivo
[
11C]-WAY 100635, Med. Chem. Res. 5 (1994) 208e227.
[29] W.J. Middleton, New fluorinating reagents. Dialkylaminosulfur fluorides,
J. Org. Chem. 40 (1975) 574e578.
[30] M. Aleksandrov, G.I. Danilenko, L.M. Yagupol’skii, Fluorine derivatives of
adamantane, J. Org. Chem. USSR 9 (1973) 976e977.
[31] J. Christopher Mc Williams, F.J. Fleitz, N. Zheng, D. Armstrong, Preparation
of n-butyl-4-chlorophenyl sulfide, Org. Syn. Coll. 10 (2004) 147 79 (2002)
43.
[32] P. Bosch, F. Camps, E. Chamorro, Trtrabutylammonuim bifluoride: a versatile
and efficient fluorinating agent, Tetrahedron Lett. 28 (1987) 4733e4736.
[8] I.A. Cliffe, A retrospect on the discovery of WAY 100635 and the prospect for
improved 5-HT1A receptor PET radioligands, Nucl. Med. Biol. 27 (2000) 441e447.
[9] V.W. Pike, C. Halldin, H. Wikström, S. Marchais, J.A. McCarron, J. Sandell,
B. Nowicki, C.-G. Swahn, S. Osman, S.P. Hume, M. Constantinou, B. Andree,
L. Farde, Radioligands for the study of brain 5-HT1A receptors
in vivoedevelopment of some new analogues of WAY, Nucl. Med. Biol. 27
(2000) 449e455.
[10] Z. Zhuang, M. Kung, H.F. Kung, Synthesis and evaluation of 4-(20-methox-
yphenyl)-1-[20-[N-(20-pyridinyl)-p-iodobenzamido]ethyl]piperazine
(p-
[33] C.R.
Davis,
D.C.
Swenson,
D.J.
Burton,
Tetramethyl
1,1,4,4-
MPPI): a new iodinated 5-HT1A ligand, J. Med. Chem. 37 (1994) 1406e1407.
[11] L. Lang, E. Jagoda, B. Schmall, B.K. Vuong, H. Richard Adams, D.L. Nelson,
R.E. Carson, W.C. Eckelman, Development of fluorine-18-labeled 5-HT1A
antagonist, J. Med. Chem. 42 (1999) 1576e1582.
cyclohexanetetacarboxylate: preparation and conversion to key precursors
of fluorinated, stereochemically defined cyclohexanes, J. Org. Chem. 58 (1993)
6843e6850.
[34] P.A. Grieco, T.R. Vedananda, C(12)-substituted prostaglandins. An enantio-
[12] A. Plenevaux, D. Weissmann, J. Aerts, C. Lemaire, C. Brihaye, C. Degueldre,
D. Le Bars, D. Comar, J. Pujo, A. Luxen, Tissue distribution, autoradiography
and metabolism of MPPF a new 5-HT1A antagonist for PET: an in vivo study in
rat, J. Neurochem. 75 (2000) 803e811.
specific total synthesis of (þ)-12-(Fluoromethyl)prostaglandin F2 methyl
a
ester, J. Org. Chem. 48 (1983) 3497e3502.
[35] A. Ricci, A. Degl’innocenti, M. Fiorenza, M. Taddei, M.A. Spartera,
D.R.M. Walton, Fluoride ion induced reactions of organosilanes with electro-
philes, Tetrahedron Lett. 23 (1982) 577e578.
[36] D.W. Kim, H.J. Jeong, S.T. Lim, M.H. Sohn, Tetrabutylammonium tetra-butanol
coordinated fluoride as a facile fluoride source, Angew. Chem. Int. Ed. 47
(2008) 8404e8406.
[37] D.W. Kim, H.J. Jeong, S.T. Lim, M.H. Sohn, J.A. Katzenellenbogen, D.Y. Chi, Facile
nucleophilic fluorination reactions using tert-alcohol as reaction medium:
significantly enhanced reactivity of alkali metal fluorides and improved
selectivity, J. Org. Chem. 73 (2008) 957e962.
[38] I.A. Cliffe, H.L. Mansell. European Patent nr. 0 512 755 A2, 1993.
[39] D. Lednicer, M.F. Grostic, Selective demethylation of deoxyanision. Mass
spectra of the products, J. Org. Chem. 32 (1967) 3251e3253.
[13] J. Passchier, P.H. Elsinga, Quantitative imaging of 5-HT1A receptor binding in
healthy volunteers with [18F]MPPF, Nucl. Med. Biol. 27 (2000) 473e476.
[14] L. Zimmer, N. Aznavour, [18F]MPPF as a tool for the in vivo imaging of 5-HT1A
receptors in animal and human brain, Neuropharmacology 52 (2007) 695e707.
[15] L. Lang, E. Jagoda, Y. Ma, M.B. Sassaman, W.C. Eckelman, Synthesis and in vivo
biodistribution of
[
18F] labeled 3-cis-3-trans,4-cis-, and 4-trans-fluo-
rocyclohexane derivatives of WAY 100635, Bioorg. Med. Chem. 14 (2006)
3737e3748.
[16] R.E. Carson, Y. Wu, L. Lang, Y. Ma, M.G. Der, P. Herscovitch, W.C. Eckelman,
Brain uptake of the acid metabolites of F-18-labeled WAY 100635 analogs,
J. Cereb. Blood Flow Metab. 23 (2003) 249e260.
[17] Y. Hoon Ryu, J. Liow, S. Zoghbi, M. Fujita, J. Collins, D. Tipre, J. Sangare, J. Hong,
V.W. Pike, R.B. Innis, Disulfiram inhibits defluorination of 18F- FCWAY, reduces
bone radioactivity, and enhances visualization of radioligand binding to
serotonin 5-HT1A receptors in human brain, J. Nucl. Med. 48 (2007)
1154e1161.
[40] Y.C. Cheng, W.H. Prusoff, Relationship between the inhibition constant
(Ki) and the concentration of inhibition which causes 50 percent inhibi-
tion (IC50
) of an enzymatic reaction, Biochem. Pharmacol. 22 (1973)
3099e3108.