Synthesis and preliminary in vivo evaluation of 4-[18F]fluoro-N- {2-[4-(6-trifluoromethylpyridin-2-yl) piperazin-1-yl]ethyl}benzamide, a potential PET radioligand for the 5-HT1A receptor

Article abstract of DOI:10.1002/jlcr.837

4-Fluoro-N-{2-[4-(6-trifluoromethylpyridin-2-yl)piperazin-1-yl]ethyl} benzamide is a full 5-HT_1A agonist with high affinity (pK _i=9.3), selectivity and a clog P of 3.045. The corresponding PET radioligand 4-[¹⁸F]fluoro-N-{2-[4-(6-trifluoromethylpyridin-2-yl) piperazin-1-yl]ethyl}benzamide was synthesized by nucleophilic aromatic substitution on the nitro precursor. The fluorinating agent K[ ¹⁸F]F/Kryptofix 2.2.2 was both dried (9 min, 700 W) and incorporated in the precursor (5 min, 700 W) using a commercially available microwave oven. In a total synthesis time of 60 min, an overall radiochemical yield of 18% (SD=5, n=7, EOS) was obtained. Radiochemical purity was always higher than 99% and specific activity always higher than 81.4 GBq/μmol (2.2 Ci/μmol). Initial brain uptake in mice was 2.19% ID (5.47% ID/g, 2 min) but decreased rapidly (0.17% ID, 0.45% ID/g (60 min)). During the first 20 min p.i., radioactivity concentration of the brain was significantly higher than that of blood demonstrating good brain entry of the tracer. Copyright

Full text of DOI:10.1002/jlcr.837

JOURNAL OF LABELLED COMPOUNDS AND RADIOPHARMACEUTICALS
J Label Compd Radiopharm 2004; 47: 531–542.
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jlcr.837
Research Article
Synthesis and preliminary in vivo evaluation of 4-
[18F]fluoro-N-{2-[4-(6-trifluoromethylpyridin-2-yl)
piperazin-1-yl]ethyl}benzamide, a potential PET
radioligand for the 5-HT1A receptor
M. Vandecapelle1,*, F. Dumont1, F. De Vos2, K. Strijckmans3, D. Leysen4,
K. Audenaert5, R. A. Dierckx2 and G. Slegers1
1 Department of Radiopharmacy, Ghent University, Belgium
2 Division of Nuclear Medicine, Ghent University Hospital, Belgium
3 Department of Analytical Chemistry, Institute for Nuclear Sciences, Ghent University,
Belgium
4 Department of Medicinal Chemistry, Scientific Development Group, Organon NV, The
Netherlands
5 Department of Psychiatry and Medical Psychology, Ghent University Hospital,
Belgium
Summary
4-Fluoro-N-{2-[4-(6-trifluoromethylpyridin-2-yl)piperazin-1-yl]ethyl}benzamide is a
full 5-HT1A agonist with high affinity (pKi=9.3), selectivity and a c log P of 3.045.
The corresponding PET radioligand 4-[18F]fluoro-N-{2-[4-(6-trifluoromethylpyridin-
2-yl)piperazin-1-yl]ethyl}benzamide was synthesized by nucleophilic aromatic sub-
stitution on the nitro precursor. The fluorinating agent K[18F]F/Kryptofix 2.2.2 was
both dried (9 min, 700 W) and incorporated in the precursor (5 min, 700 W) using a
commercially available microwave oven. In a total synthesis time of 60 min, an overall
radiochemical yield of 18% (SD=5, n=7, EOS) was obtained. Radiochemical purity
was always higher than 99% and specific activity always higher than 81.4 GBq/mmol
(2.2 Ci/mmol). Initial brain uptake in mice was 2.19% ID (5.47% ID/g, 2 min) but
decreased rapidly (0.17% ID, 0.45% ID/g (60 min)). During the first 20 min p.i.,
radioactivity concentration of the brain was significantly higher than that of blood
demonstrating good brain entry of the tracer. Copyright # 2004 John Wiley & Sons,
Ltd.
Key Words: Org 13063; 5-HT1A agonist; 18F labelling; biodistribution study
*Correspondence to: M. Vandecapelle, Department of Radiopharmacy, Harelbekestraat 72, B-9000 Ghent,
Belgium. E-mail: marleen.vandecapelle@ugent.be
Copyright # 2004 John Wiley & Sons, Ltd.
Received 30 November 2003
Revised 25 April 2004
Accepted 30 April 2004

532
M. VANDECAPELLE ET AL.
Introduction
The neurotransmitter 5-hydroxytryptamine (5-HT, serotonin) plays an
important role in a variety of physiological, behavioral and cognitive functions
e.g. gastrointestinal motility, cardiovascular function, pain perception,
thermoregulation, sleep, mood and sexual behavior. Impairments of the
serotonergic functions have been implicated in the etiology of a variety of
mental disorders including psychotic disorders, mood and anxiety disorders
(obsessive compulsive disorder, panic disorder, phobia, etc.) schizophrenia,
sleep disorders and dementia.1–3
The early discovery of 8-OH-DPAT (8-hydroxy-2-(N,N-di-n-propylamino)-
tetraline) and buspirone as 5-HT1A (partial) agonists, makes this 5-HT
receptor subtype the most intensively studied. Clinically, it is meant to be
involved in the pathogenesis of several neuropsychiatric disorders e.g. anxiety
and depression, disorders in the regulation of aggression and substance abuse
disorders.3,4
The nuclear imaging modalities PET (positron emission tomography)
and SPECT (single photon emission computed tomography) offer the unique
opportunity to study in vivo and non-invasively, the 5-HT1A receptor, using
a selective radioligand. At the moment, only 5-HT1A receptor PET tracers
are available with the lead component [carbonyl-11C]WAY100635 (N-{2-[4-(2-
methoxyphenyl)piperazin-1-yl]ethyl}-N-(pyridin-2-yl)cyclohexane[11C]carbox-
amide, pKi=9.15) which is capable of delineating the 5-HT1A receptor in
human brain.6 [Carbonyl-11C]desmethyl-WAY100635 ([11C]DWAY, N-{2-
[4-(2-hydroxyphenyl)piperazin-1-yl]ethyl}-N-(pyridin-2-yl)cyclohexane[11C]-
carboxamide, IC50=1.4 nM7) seems to possess even better in vivo properties as
demonstrated by its significantly higher brain radioactivity signal providing
improved imaging statistics, advantages in biomathematic modeling and the
preclusion of [11C]-DWAY as a metabolite interfering with PET measure-
ments.8
Because of the better nuclear properties of the 18F isotope and the
more reliable radiosynthetic procedures, 18F labelled WAY100635
analogs were rapidly developed with [18F]MPPF ([18F]FBWAY, 4-[18F]fluo-
ro-N-{2-[4-(2-methoxyphenyl)piperazin-1-yl]ethyl}-N-(pyridin-2-yl)benzamide,
pKi=8.55) as most important example.9–11 The lower affinity of MPPF for the
5-HT1A receptor in comparison to WAY100635 could make this ligand sensitive
to endogenous serotonin concentration which could enable us to detect
differences in serotonin release.11
A completely different structure, based on 8-OH-DPAT, is found in the
antagonist [11C]NAD299 ((R)-3-N,N-dicyclobutylamino-8-fluoro-3,4-dihydro-
2H-1-benzopyran-5-[11C]carboxamide hydrogen (2R,3R)-tartrate monohy-
drate, pKi=9.212) that was shown to be able to delineate the 5-HT1A receptor
in non-human primates.12–14
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2004; 47: 531–542

A POTENTIAL PET RADIOLIGAND FOR THE 5-HT1A RECEPTOR
533
NH
X
N
N
N
O
CF
3
1.F
X=I
1.I
X=F
18F 1.18F
123I 1.123
I
X=
X=
Figure 1. Chemical structure of 1.F (Org 13063), 1.18F, 1.I and 1.123
I
Org 13063 or 4-fluoro-N-{2-[4-(6-trifluoromethyl-pyridin-2-yl)piperazin-1-
yl]ethyl}benzamide (1.F, Figure 1) is a full 5-HT1A agonist with high affinity
(pKi=9.3) and selectivity (pKi: 5-HT1B=5.4, 5-HT2A=6.27, 5-HT2C=6.07, 5-
HT6=5.53, 5-HT7=6.72, SERT55.1, D155.2, D2a=5.93, D3=7.05,
D4=8.57, DAT55.1, a1=6.5, a2A=5.62, a2B=5.5, a2C=5.34, NET55.0,
M155.2, M255.2, M355.18, M455.44, ORL-155.28, m=5.64, k=5.31,
d55.39).15 Lipophilicity was calculated to be 3.045.y Its iodine derivative (1.I,
Figure 1) was found to have lower in vitro affinity (pKi=8.2) without loss of
selectivity15 and was considerably more lipophilic (c log P=3.879y). Biodis-
tribution studies of its radioiodinated analogue (1.123I, Figure 1) were
performed in rats leading to a 0.22% ID in the brain at 5 min p.i. At none
of the selected time points, radioactivity concentration of brain was
significantly higher than that of blood. Furthermore, in rabbits, the
radioactivity distribution did not reflect the regional distribution of the 5-
HT1A receptor in the brain nor could it be blocked by 5-HT1A selective
ligands.16,17
The imaging of G-protein coupled receptors (GPCR) with agonist
radioligands has been very controversial. So far, no agonist or partial agonist
has yielded significant specific labelling of the 5-HT1A receptor in vivo
despite numerous efforts e.g. [11C]Org 13052,18 [11C]LY27460119 and 8{{3-[4-
(2-[11C]methoxyphenyl)piperazin-1-yl]-2-hydroxypropyl}oxy}thiochroman.20
[11C]Org 13052 (pKi=10) is a partial agonist with low intrinsic activity
(IA=0.2). Brain uptake of radioactivity in rat was rather low (0.8%
ID/g between 10 and 30 min), regional differences were minor and a spatially
uniform reduction in the amount of 11C was observed after treat-
ment of animals with 8-OH-DPAT.18,21 [11C]LY274601 (pKi=9.2) is a full
agonist with high blood-brain-barrier penetration in mice (6.8% ID, 30 min).
However hippocampus/cerebellum ratios were very poor ( ꢀ 1.2) and uptake
in target areas was hardly affected after pretreatment with 8-OH-
DPAT. The radioligand was metabolized quickly with a half life of
y Calculated with Biobyte c log P 4.0.
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2004; 47: 531–542

534
M. VANDECAPELLE ET AL.
15 min.19 8{{3-[4-(2-[11C]methoxyphenyl)piperazin-1-yl]-2-hydroxypropyl}ox-
y}thiochroman (pKi=8.3) is a partial agonist and was in vivo evaluated in rat
and cat. Results showed low accumulation in the brain dominated by non-
specific binding.20
It has been postulated that agonists only bind to receptors that are coupled
to a G-protein (high affinity state) which represent only a variable fraction
of the total receptor amount available for binding of antagonists. Agonist
radioligands require therefore a higher affinity (smaller Kd) than antagonists
to achieve equal binding potential (Bmax/Kd) in receptor-containing areas
of the brain. In addition, the binding of agonists is generally transient,
since agonist binding converts the receptor into a G-protein dissociated
conformational state that usually has low affinity for the agonist. Nevertheless,
an effective agonist radioligand for brain 5-HT1A receptors would be a
useful tool for the selective investigation of the functional GPCR since
it is likely that, in certain diseases, only the high affinity state GPCR is
altered.21,22
We hereby report the synthesis and in vivo evaluation in mice of 4-
[18F]fluoro-N-{2-[4-(6-trifluoromethylpyridin-2-yl)piperazin-1-yl]ethyl}benza-
mide (1.18F, Figure 1).
Results and discussion
The precursor 1.NO2 was synthesized (Figure 2) through a four-step synthesis
starting from 2-chloro-6-trifluoromethylpyridine (2). Coupling of this pyridine
derivative (2) with piperazine, followed by a Gabriel synthesis, gave the
primary ethylamine (5). The latter was coupled with the appropriate
benzoylchloride to give the nitro (1.NO2) derivative of Org 13063. Overall
O
N
(i)
(ii)
Cl
N
NH
N
N
N
N
N
O
4
2
3
CF
3
CF
3
CF3
NH2
NH
O
(iii)
(iv)
NO2
N
N
N
N
N
N
5
1.NO2
CF3
CF
3
Figure 2. Synthesis of the precursor compound (i) piperazine, reflux, 22 h (ii) 2-
Br-ethylphtaalimide, Na2CO3, 808C, 3 h (iii) 1. hydrazine hydrate, reflux, 3 h; 2.
HCl, reflux, 4 h (iv) 4-NO2-benzoylchloride, TEA, RT, 3 h
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2004; 47: 531–542

A POTENTIAL PET RADIOLIGAND FOR THE 5-HT1A RECEPTOR
535
chemical yield of this precursor synthesis was 8% and resulted in chemically
pure, brown crystals.
4-[18F]Fluoro-N-{2-[4-(6-trifluoromethylpyridin-2-yl)piperazin-1-yl]ethyl}-
benzamide (1.18F) was synthesized by nucleophilic aromatic substitution on
1.NO2 with the fluorinating agent K[18F]F/Kryptofix 2.2.2 (K2.2.2) (Figure 3).
Both drying and incorporation of [18F]Fꢁ were initially performed by
conventional heating. The use of aluminium heating blocks for the azeotropic
distillation process (3 ꢂ 1 ml CH3CN, 3 ꢂ 10 min, 1108C, He flow) and the
nucleophilic labelling reaction (30 min, 1508C) rendered low radiochemical
yields (8% (SD=3, n=3)) and long production times (105 min). Therefore, the
possible benefit of microwave heating was examined. The evaporation process
of the azeotropic distillations originally was difficult to control (bumping)
resulting in a loss of radioactivity of up to 30%. Optimization of the reaction
vial position in the microwave oven, power and time of heating and flow rate
of the He gas was performed taking into account that the first two parameters
(vial position and microwave power) were to be compatible with the
subsequent labelling reaction. This resulted in the preparation of K[18F]F/
K2.2.2 in a vial placed against the door of the microwave oven working at
700 W within 9 min (PHe=0.2 bar). Under these circumstances, the subsequent
radiosynthesis only took 5 min giving a radiochemical yield of 42% (SD=3,
n=7).
The 18F labelling of a molecule containing a trifluoromethyl group could
pose problems because of the potential isotopic fluoride exchange. The
possible incorporation of 18F in the CF3 group of 1.NO2 was examined by
performing a radiosynthesis using N-{2-[4-(6-trifluoromethylpyridin-2-yl)pi-
perazin-1-yl]ethyl}benzamide as precursor in stead of 1.NO2 under the exact
same labelling conditions as described above. This radiosynthesis did not
NH
N
N
NO2
N
O
1.NO2
CF3
K[18
F]F/K2.2.2
5 min, 700 W
NH
O
18F
N
N
N
1.18F
CF3
Figure 3. Radiosynthesis of 1.18F
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2004; 47: 531–542

536
M. VANDECAPELLE ET AL.
result in any labelled products hereby excluding the possibility of labelling
the CF3 moiety under our labelling conditions. The exchange of the
NO2 group of the precursor 1.NO2 for a 19F originating from the large excess
of precursor CF3 cannot be excluded but cannot be too extensive seeing
the relatively good specific activity of the produced tracer (SA=81.4 GBq/
mmol).
The use of DMSO as reaction medium makes it impossible to directly inject
the crude reaction mixture onto the HPLC system. Classically, this problem is
solved by performing a solid phase extraction redissolving precursor and
tracer compound in a HPLC compatible solvent. We solved this problem
simply by diluting the reaction mixture with cold, acidic H2O. This H2O was
acidified (pH=2.2) to prevent precipitation of the precursor and cooled in an
ice bath to prevent degradation of the tracer under acidic aqueous
circumstances. This dilution-cooling step did not introduce an extra operation
in the synthesis protocol and was less time consuming and less expensive than
the traditional solid phase extraction.
The large excess of precursor compared to the fluorinating agent K[18F]F/
K2.2.2 is used to drive this type of reaction to completion. However,
this seriously hampered the subsequent HPLC separation of precursor and
tracer compound. Ideally, the tracer should elute before its precursor to
avoid contamination of the tracer with small amounts of precursor.
Several eluentia were tested (NaOAc/EtOH, KH2PO4/EtOH, NaOAc/
CH3CN, KH2PO4/THF) on a Symmetry Prep column but only the
use of the tertiary eluent NaOAc(0.05 M, pH=5)/MeOH/THF (55/28/17) at
5 ml/min combined speed, good resolution and elution of 1.18F (tR=29 min)
before 1.NO2 (tR=35 min). This semi-preparative HPLC set up
made radiochemical purity higher than 99%. Specific activity was
always higher than 81.4 GBq/mmol (2.2 Ci/mmol). In a typical produc-
tion, 4.07 GBq (110 mCi) of injectable tracer solution was produced
starting from 22.2 GBq (600 mCi) within 60 min (overall radiochemical
yield=18% (SD=5, n=7, EOS)). Stability tests showed that the tracer
remained more than 98% radiochemically pure during the first 24 h after
production.
Biodistribution studies were performed in mice leading to a 2.19% ID
(5.47% ID/g) in the brain at 2 min p.i. decreasing rapidly afterwards (0.17%
ID, 0.45% ID/g (60 min)). During the first 20 min p.i., the radioactivity
concentration (% ID/g) of the brain was significantly higher than that of
blood (Table 1). Brain to blood ratios turned out to be much better for 1.18F
than for 1.123I, probably because of its lower lipophilicity (c log P
(1.18F)=3.045 versus c log P (1.123I)=3.879). Further experiments are in
progress to investigate the selective and specific binding of the tracer to the 5-
HT1A receptor in the brain.
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2004; 47: 531–542

A POTENTIAL PET RADIOLIGAND FOR THE 5-HT1A RECEPTOR
537
Table 1. Average %ID/g tissue and standard deviations as a function of time (min)
Time (min)
Tissue
Blood
1
2
5
10
20
40
60
120
180
3.293
0.649
2.546
0.355
1.715
0.112
1.515
0.324
1.189
0.152
0.599
0.126
0.538
0.201
0.363
0.007
0.333
0.061
SD
SD
Brain
6.766
1.529
5.468
1.018
3.944
1.059
2.900
0.603
1.533
0.054
0.561
0.172
0.452
0.154
0.328
0.111
0.266
0.087
Experimental
Material and methods
Org 13063 was obtained from Organon NV (The Netherlands). Chemicals
were purchased from Sigma-Aldrich-Fluka (Belgium), Fluorochem (United
Kingdom) or Merck (Belgium) and were of the highest available purity. They
were used without further purification. He gas (Alphagaz He1) was purchased
(
from Air Liquid (Belgium) and was dried over molecular sieves (3 A). H2O was
deionized with a Milli-Q water system (Millipore). [18O]H2O (95+%) was
purchased from Campro Scientific (The Netherlands).
Chemical reactions were monitored by TLC using coated silica gel plates
(Polygram SIL G/UV254 } Machery-Nagel) and mobile phases as
mentioned in the text. The different spots were visualized using either UV
light and/or ninhydrine (0.2% in EtOH). Preparative TLC was performed
using coated silica glass plates (SIL G-200 UV254 (2 mm) } Machery-Nagel)
and mobile phases as mentioned. Column chromatography was performed
.
with silica gel (50–200 mm, Sud Chemie) and mobile phases as mentioned in
the text.
1H NMR spectra were obtained on a 500 MHz spectrometer (Bruker,
Accuspec 3000). Chemical shifts are reported in ppm units downfield from
internal (CH3)4Si. Electrospray Ionisation Mass Spectrometry (ESI MS) was
performed on a Micromass ZMD (Waters, Belgium). Exact mass measure-
ments were performed on a quadrupole-time of flight mass spectrometer (Q-
Tof-2, Micromass, UK) equipped with a standard electrospray ionization
(ESI) interface (Rega Institute, Belgium).
For microwave heating, a commercially available oven (Samsung M935,
1000 W) was used. Borosilicate vials (Maxi-vial, 10 ml, Alltech, Belgium)
equipped with screw caps and TFE/silicone liners were used in the microwave
experiments.
Purification of the tracer was performed by semi-preparative HPLC using a
Symmetry Prep C18 column (Waters, 300 ꢂ 7.8 mm, 7 mm) and NaOAc
(0.05 M, pH=5)/MeOH/THF: 55/28/17 at a flow rate of 5 ml/min. Analytical
HPLC was performed using a Symmetry C18 column (Waters, 150 ꢂ 4.6 mm,
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2004; 47: 531–542

538
M. VANDECAPELLE ET AL.
5 mm) and NaOAc (0.05 M, pH=4.5)/MeOH/THF: 62/23/18. All HPLC runs
were carried out making use of a Waters 510 pump, a UV/VIS detector at
l=254 nm (PU 4110 UV/VIS } Philips) and a NaI(Tl) detector (Bicron
Frisktech1, probe 1 ꢂ 1 in). Radiochromatograms were recorded on a two
channel integrator (C-R5A Chromatopac, Shimadzu). C18 Sep-Pak1
cartridges were purchased from Waters Corporation.
Radioactivity measurements were performed on a single channel NaI(Tl)
detector (Canberra Packard, type 2007P, 3 ꢂ 3 in) or on a Cobra B5005 auto
gamma (Canberra Packard).
Synthesis of 1-(6-trifluoromethylpyridin-2-yl)piperazine (3)
Piperazine (2.871 g, 33 mmol, 99%) was dissolved in EtOH (20 ml, pa) and
brought to reflux. A solution of 2-chloro-6-trifluoromethylpyridine (2, 1.997 g,
11 mmol) in EtOH (10 ml, pa) was added slowly. The stirred reaction mixture
was refluxed and monitored with TLC (CH2Cl2/CH3OH/HOAc: 50/10/3).
After 22 h, the reaction was stopped, the solvent was evaporated and H2O
(100 ml) was added. 1-(6-Trifluoromethylpyridin-2-yl)-piperazine (3) was
extracted from the H2O with EtOAc (2 ꢂ 100 ml). The organic layer was
washed with H2O (2 ꢂ 100 ml) and evaporated. To the resulting brown oil,
H2O (20 ml) was added and stirred for 3 h. 3 was extracted with CH2Cl2
(2 ꢂ 50 ml). The organic layer was dried with Na2SO4 (anhydrous) and
evaporated to give a yellow oil. Chemical yield was 85%. 1H NMR (360 MHz,
CDCl3): 1.80 (1H, s, NH), 2.96 (4H, t), 3.59 (4H, t), 6.75 (1H, d), 6.92 (1H, d),
7.56 (1H, t); ESI MS 232.1 [MH]+.
Synthesis of 2-{2-[4-(6-trifluoromethylpyridin-2-yl)piperazin-1-yl]ethyl}isoin-
dole-1,3-dione (4)
A stirred mixture of 3 (5.077 g, 21.959 mmol) and Na2CO3 (2.327 g,
21.959 mmol) in DMF (30 ml, pa) was heated to 808C. 2-Bromo-ethylphtha-
limide (5.873 g, 21.959 mmol, 95%) was added while the temperature was kept
at 808C. The reaction was monitored with TLC (CH2Cl2/CH3OH/NH3: 90/16/
21) and stopped after 3 h by adding cold H2O (300 ml). 2-{2-[4-(6-
Trifluoromethylpyridin-2-yl)piperazin-1-yl]ethyl}isoindole-1,3-dione (4) was
extracted from the aqueous phase with EtOAc (2 ꢂ 100 ml). The organic layer
was washed with H2O (2 ꢂ 200 ml) and acidified with HCl by means of a Kipp
device. The hydrochloride salt of 4 was filtered off and washed with EtOAc.
The precipitation was dissolved in H2O (100 ml) which subsequently was made
basic (pH=14). 4 was extracted with CH2Cl2 (2 ꢂ 100 ml). The organic layer
was dried with Na2SO4 (anhydrous) and evaporated. The formed, white
precipitation was redissolved in warm CH3OH. Cooling this solution, formed
1
4 as white crystals. Chemical yield was 65%. H NMR (360 MHz, CDCl3):
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2004; 47: 531–542

A POTENTIAL PET RADIOLIGAND FOR THE 5-HT1A RECEPTOR
539
2.62 (4H, t), 2.69 (2H, t), 3.53 (4H, t), 3.87 (2H, t), 6.74 (1H, d), 6.92 (1H, d),
1
7.55 (1H, t), 7.72 (2H, d), 7.85 (2H, d); H NMR (300 MHz, d6-DMSO): 2.51
(4H, t), 2.58 (2H,t), 3.45 (4H, t), 3.74 (2H, t), 7.00 (1H, d), 7.07 (1H, d), 7.70
(1H, t), 7.80-7.88 (4H, m); ESI MS 405.2 [MH]+.
Synthesis of 2-[4-(6-trifluoromethylpyridin-2-yl)piperazin-1-yl]ethylamine (5)
To a stirred solution of 4 (8.661 g, 21.417 mmol) in EtOH (28 ml, pa) at reflux,
was slowly added hydrazine hydrate (1.049 ml, 21.417 mmol, 99%). The
reaction was monitored with TLC (CH2Cl2/CH3OH/NH3: 90/16/21). After
3 h, H2O (13 ml) and HCl (6.5 ml, 37%) were added. After another 4 h of
reflux, the formed solid was filtered off and washed with EtOH/H2O/HCl
(37%): 28/13/6.5. EtOH was evaporated and the pH of the remaining aqueous
fraction was adjusted to 13 with NaOH (30 ml, 20%). 2-[4-(6-Trifluoro-
methylpyridin-2-yl)piperazin-1-yl]ethylamine (5) was extracted with CH2Cl2
(2 ꢂ 100 ml). The combined organic layers were dried with Na2SO4 (anhy-
drous) and evaporated. A yellow precipitation was formed. Further
purification was performed by column chromatography (CH2Cl2/CH3OH/
1
NH3: 90/16/21). Chemical yield was 26%. H NMR (360 MHz, CDCl3): 1.78
(2H, s, NH), 2.49 (2H, t), 2.55 (4H, t), 2.85 (2H, t), 3.61 (4H, t), 6.76 (1H, d),
6.93 (1H, d), 7.56 (1H, t); ESI MS: 275.2 [MH]+, 258.3 [MH-NH2]+.
Synthesis of 4-NO2-N-{2-[4-(6-trifluoromethylpyridin-2-yl)piperazin-1-yl]
ethyl} benzamide (1.NO2)
Compound 5 (274.3 mg, 1 mmol) was dissolved in CH2Cl2 (2 ml). After
addition of TEA (221 ml, 1.586 mmol), the reaction mixture was cooled to 08C.
4-NO2-benzoylchloride (213.2 mg, 1.126 mmol, 98%) in CH2Cl2 (2 ml) was
added dropwise. The reaction was monitored with TLC (CH2Cl2/CH3OH/
NH3: 90/16/21). This reaction mixture was stirred for 3 h at ambient
temperature. The organic layer was washed with H2O, dried with Na2SO4
(anhydrous) and evaporated. Purification was performed by column chroma-
1
tography (CH2Cl2/CH3OH/NH3: 90/16/21). Chemical yield was 60%. H
NMR (360 MHz, CDCl3): 2.65 (4H, t), 2.70 (2H, t), 3.60 (6H, m), 6.80 (1H, d),
6.95 (2H, d), 7.60 (1H, t), 7.95 (2H, d), 8.30 (2H, d); ESI-MS [M+H]+: 424;
Exact mass (ESI-MS) [M+H]+: found 424.1596, calculated 424.1589.
Production of the fluorinating agent K[18F]F/K2.2.2
No carrier added [18F]Fꢁ was produced with 18 MeV protons (5 mA, 120 min)
via the 18O(p,n)18F reaction on enriched [18O]H2O. Radioactivity was trapped
on an anion exchange resin (Dowex 1 ꢂ 8, 200–400 mesh) and eluted with a
K2CO3 (5.4 mg, 39 mmol)/K2.2.2 (22 mg, 58 mmol) aqueous solution (350 ml,
mQ H2O). After 3 successive azeotropic distillations with CH3CN (3 ꢂ 1 ml,
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2004; 47: 531–542

540
M. VANDECAPELLE ET AL.
dry) in the microwave oven (3 ꢂ 3 min, 700 W) under continuous He flow
(PHe=0.2 bar, flow rate=50 ml/min), the fluorinating agent K[18F]F/K2.2.2
could be used for labelling.
Radiosynthesis and purification of 4-[18F]fluoro-N-{2-[4-(6-trifluoromethylpyr-
idin-2-yl)piperazin-1-yl]ethyl}benzamide (1.18F)
Compound 1.NO2 (10.0 mg, 24 mmol) in DMSO (0.9 ml, dry) was added to the
reaction vial containing the fluorinating agent K[18F]F/K2.2.2 and reacted in a
microwave oven for 5 min at 700 W. The reaction mixture was subsequently
added to ice cold H2O (1.1 ml, 1.68 M HOAc, pH=2.2) and loaded onto the
loop (2 ml) of the semi-preparative HPLC. The eluted tracer fraction was
diluted 5 times with H2O and loaded onto a C18 Sep-Pak1 cartridge
(previously activated with EtOH (5 ml) and H2O (10 ml)). The C18 Sep-Pak1
cartridge was rinsed with H2O (10 ml). Subsequently, the tracer was eluted
with EtOH (1 ml) and diluted with H2O (9 ml).
Quality control
Analysis of the tracer solution was performed by analytical HPLC (see
Material and Methods Section). Stability testing was carried out by reinjection
onto the same analytical HPLC set up.
Animal studies
All animal studies were carried out in compliance with the Animal Ethical
Committee of the Ghent University and the Belgian laws. NMRI mice
(m=20–25 g, male) were used which had ad libitum access to food and water
before and during the experiments.
Biodistribution studies in mice
Mice (n=3 ꢂ 9) were injected directly in the tail vein with 1.18F ( ꢀ 37 kBq
(1 mCi), ꢀ 200 ml H2O/EtOH, max. 10% EtOH) and sacrificed by decapitation
at set time points p.i. (1, 2, 5, 10, 20, 40, 60, 120 and 180 min). All animals were
dissected; blood and urine samples were collected. The dissected tissues were
blotted to remove adhering blood and weighed in tarred counting tubes. The
present radioactivity was measured on a Cobra B5005 auto gamma counter.
All values were corrected for decay and background. The injected dose (ID)
was calculated by determining the radioactivity per gram injection solution on
the same counting device and weighing the injection syringe, before and after
injection.
Results are expressed as percentage of the injected dose per gram tissue (%
ID/g) or as percentage of the injected dose (% ID). Averages and standard
deviations (SD) were calculated. The t-test (two sided, paired, n=3, a=0.05)
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2004; 47: 531–542

A POTENTIAL PET RADIOLIGAND FOR THE 5-HT1A RECEPTOR
541
was used for determining significant differences between results of different
organs.
Conclusions
4-[18F]Fluoro-N-{2-[4-(6-trifluoromethylpyridin-2-yl]piperazin-1-yl]ethyl}ben-
zamide was synthesized by nucleophilic aromatic substitution on the
nitro precursor in a microwave oven (5 min, 700 W) in sufficient
overall radiochemical yield (18% (EOS)), radiochemical purity (>99%)
and specific activity (>81.4 GBq/mmol (2.2 Ci/mmol)) to start in vivo
evaluation.
Biodistribution studies in mice were performed leading to 2.19% ID (5.47%
ID/g) in the brain at 2 min p.i. During the first 20 min, radioactivity
concentration of brain was significantly higher than that of blood. Brain to
blood ratios turned out to be much better for 1.18F than for 1.123I, probably
because of its lower lipophilicity. Although the affinity of 1.18F is 10 times
higher than that of 1.123I, only little retention of radioactivity could be
demonstrated in the mouse brain. Further experiments are in progress to
investigate the selectivity and the specificity of the binding of the tracer in the
brain.
Acknowledgements
The authors wish to thank Prof. H. Van Praag for his contribution in the
realization of the cooperation with Organon NV, Oss. We would also like to
thank J. Sambre (Ghent University Hospital, Institute for Nuclear Sciences)
for his help with the production of fluorine-18 and J. Rozenski for the
performance of the exact mass measurements.
References
1. Borne RF. Drugs Top 1994; 138: 108–117.
.
2. Ogren SO, Misane I. Nucl Med Biol 1998; 25: 747–749.
3. Saxena PR. Pharmacol Ther 1995; 66: 339–368.
4. De Vry J. Psychopharmacologia 1995; 121: 1–26.
5. Wilson AA, Inaba T, Fischer N, Dixon LM, Nobrega J, DaSilva JN, Houle S.
Nucl Med Biol 1998; 25: 769–776.
6. Pike VW, McCarron JA, Lammertsma AA, Osman S, Hume SP, Sargent PA,
Bench CJ, Cliffe IA, Fletcher A, Grasby PM. Eur J Pharm 1996; 301: R5–R7.
7. Pike VW, Halldin C, McCarron JA, Lundkvist C, Hirani E, Olsson H, Hume SP,
.
Karlsson P, Osman S, Swahn CG, Hall H, Wikstrom H, Mensonidas M, Poole
KG, Farde L. Eur J Nucl Med 1998; 25: 338–346.
!
8. Andree B, Halldin C, Pike VW, Gunn RN, Olsson H, Farde L. J Nucl Med 2002;
43: 292–303.
9. Passchier J, van Waarde A, Pieterman RM, Elsinga PH, Hendrikse HN, Pruim J,
Willemsen AT, Vaalburg W. J Nucl Med 1999; 40(5)(Suppl.): 29.
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2004; 47: 531–542

542
M. VANDECAPELLE ET AL.
10. Plenevaux A, Lemaire C, Salmon E, Laureys S, Fuchs S, Maquet P, Degueldre C,
Aerts J, Brihaye C, Damhaut P, Biver F, Goldman S, Le Bars D, Ginovart N,
Cinotti L, Pujol JF, Luxen A. J Label Compd Radiopharm 1999; 42(Suppl. 1):
S60–S62.
11. Sanabria-Bohorquez SM, Biver F, Damhaut P, Wikler D, Veraart C, Goldman S.
Eur J Nucl Med 2002; 29: 76–81.
12. Halldin C, Sandell J, Chou YH, Hall H, Thorberg SO, Werner T, Schwahn CG,
Farde L. J Nucl Med 1999; 40(5)(Suppl.): 38.
13. Sandell J, Halldin C, Chou YH, Schwahn G, Thorberg SO, Sohn D, Werner T,
Farde L. J Label Compd Radiopharm 1999; 42(Suppl. 1): S675–S677.
14. Sandell J, Halldin C, Chou YH, Swahn CG, Thorberg SO, Farde L. Nucl Med
Biol 2002; 29: 39–45.
15. G. Walker, Organon NV, Newhouse. Personal Communication.
16. Vandecapelle M, De Vos F, Dumont F, Audenaert K, Leysen D, Dierckx RA,
Slegers G. J Label Compd Radiopharm 2001; 44: 73–88.
17. Vandecapelle M, De Vos F, Vermeirsch H, De Ley G, Audenaert K, Leysen D,
Dierckx RA, Slegers G. Nucl Med Biol 2001; 28: 639–643.
18. Visser GM, van Waarde A, Medema J, Vaalburg W. J Label Compd Radiopharm
1995; 37: 280–282.
19. Suehiro M, Underwood M, Arango V, Wang TS, Kassir S, Bakalian M, Yatabe
T, Pratap M, Van Heertum RL, Mann JJ. Life Sci 1998; 63: 1533–1542.
20. Zimmer L, Fournet G, Joseph B, Guillaumet G, Le Bars D. Nucl Med Biol 2003;
30: 541–546.
21. Passchier J, van Waarde A. Eur J Nucl Med 2001; 28: 113–129.
.
22. Pike VW, Halldin C, Wikstrom HV. Radioligands for the study of brain 5-HT1A
receptors in vivo. In Progress in Medicinal Chemistry, King FD, Oxford AW (eds).
Elsevier Science: Oxford, 2001; 189–247.
Copyright # 2004 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2004; 47: 531–542