N-2-(4-N-(4-[18F]Fluorobenzamido)phenyl)-propyl-2- propanesulphonamide: Synthesis and radiofluorination of a putative AMPA receptor ligand

Article abstract of DOI:10.1002/jlcr.1413

Arylpropylsulphonamides are in the focus of research as α-amino-3-hydroxy-5-methyl-4-isoxazolpropionic acid (AMPA) receptor ligands. A new fluorine-18-labelled potentiator of AMPA receptors was synthesized as a potential radiotracer for cerebral imaging w

Full text of DOI:10.1002/jlcr.1413

Journal of Labelled Compounds and Radiopharmaceuticals
J Label Compd Radiopharm 2007; 50: 1169–1175.
Published online 10 September 2007 in Wiley InterScience
(www.interscience.wiley.com). DOI: 10.1002/jlcr.1413
JLCR
Research Article
N-2-(4-N-(4-[18F]Fluorobenzamido)phenyl)-propyl-2-
propanesulphonamide: synthesis and radiofluorination
of a putative AMPA receptor ligand
UTE B. KRONENBERG, BIRTE DREWES*, WIEBKE SIHVER and HEINZ H. COENEN
¨
¨
¨
Institut fur Nuklearchemie, Forschungszentrum Julich GmbH, 52425 Julich, Germany
Received 15 November 2006; Revised 31 January 2007; Accepted 1 June 2007
Abstract: Arylpropylsulphonamides are in the focus of research as a-amino-3-hydroxy-5-methyl-4-isoxazolpropionic
acid (AMPA) receptor ligands. A new fluorine-18-labelled potentiator of AMPA receptors was synthesized as a
potential radiotracer for cerebral imaging with positron emission tomography. Using N-2-(4-N-(4-nitrobenzamido)-
phenyl)-propyl-2-propanesulphonamide (7) as labelling precursor for
a
Kryptofix 2.2.21/K2CO3-activated
nucleophilic radiofluorination, the putative AMPA receptor ligand N-2-(4-N-(4-[18F]fluorobenzamido)phenyl)-
propyl-2-propanesulphonamide [18F]8 was obtained in one step. Optimization of the reaction parameters time,
temperature, solvent and concentration gave a radiochemical yield of 38 ꢀ 8% at 1808C in dimethylsulphoxide
within 30-min reaction time. After a solid-phase extraction followed by a high-performance liquid chromatography
separation, the product could be obtained in radiochemical yields of 5 ꢀ 1.5%. Radiochemical purity was higher
than 95% and the specific activity amounted to 77 ꢀ 40 GBq/mmol. First in vitro assays with rat brain slices revealed
a high non-specific binding and a uniform distribution of [18F]8 not lending it for in vivo imaging purposes.
Copyright # 2007 John Wiley & Sons, Ltd.
Keywords: AMPA receptor ligand; n.c.a. radiofluorination; potentiator; fluorine-18; positron emission tomography
Introduction
compounds, namely quinoxalinediones and related
structures (e.g. NBQX 1) or 2,3-benzodiazepines (e.g.
GYKI 53655 2) (for structures see Figure 1).3
As the major excitatory neurotransmitter in the
mammalian central nervous system, glutamate acti-
vates a diversity of ionotropic and metabotropic recep-
tors. The ionotropic subspecies of these receptors are
named after the first identified selective agonists, which
were N-methyl-d-aspartate (NMDA), a-amino-3-hydro-
xy-5-methyl-4-isoxazolpropionic acid (AMPA) and kai-
nate.1 The AMPA receptor subtype is known to play an
important role in memory and learning,1 but also
seems to play a significant role in the aetiology of
disorders of the human brain, such as ischaemic
stroke or epilepsy.2
Further on there exist allosteric modulators, which
do not inhibit or excite the receptor directly. How-
ever, they are of great interest, as they amplify
the effects of agonists or antagonists on the receptor.
The so-called AMPAkines, a series of benzamides (for
example, Aniracetam 3 and CX 516 4, see Figure 1),
showed promising effects as positive modulators
(also called potentiators) and have lately been of
great interest for potential use as drugs for memory
enhancement.
The labelling of such an AMPAkine with fluorine-18
could lead to a new class of radioligands that would
enable a first look at the AMPA receptor system in vivo.
With tritiated LY395153 (5, see Figure 1), Zarrinmayeh
et al. described a molecule with a benzamide structure
and sulphonamide functionality, which showed pro-
mising results during in vitro tests using rat cortical
membranes or recombinant human GluR4 recep-
tors.4,5 The in vitro binding in rat cortex, but also at
Whereas agonists for this subclass of receptors
mainly base on the structure of AMPA itself and derived
compounds, antagonists with a significant affinity for
the AMPA receptor consist mainly of two classes of
*Correspondence to: Birte Drewes, Institut fu¨r Nuklearchemie,
Forschungszentrum Ju¨lich GmbH, 52425 Ju¨lich, Germany.
E-mail: b.drewes@fz-juelich.de
Copyright # 2007 John Wiley & Sons, Ltd.

1170 U. B. KRONENBERG ET AL.
recombinant human GluR4 receptors, revealed
a
Results and discussion
reasonable specific binding. This molecule was the
origin for the synthesis of a labelled potentiator that
could be a potential radiotracer for cerebral imaging.
Since the benzamide and sulphonamide moieties
are the major binding parts of the molecule to the
receptor, it could be hoped that the potency of the
potentiator is not affected by substitution of a hydrogen
atom with a fluorine atom on the aromatic ring. Since
the aminocarbonyl moiety as electron-withdrawing
group enables nucleophilic labelling in para-position
of the arene,6 with fluorine-18 under activation with
Kryptofix 2.2.21, this was attempted according to
Scheme 1.7
Preparation of precursor and standard
The authentic compound N-2-(4-N-(4–fluorobenzami-
do)phenyl)propyl-2-propanesulphonamide (8) and the
corresponding nitro derivative 7 to be used as labelling
precursor were synthesized following the procedure by
Zarrinmayeh et al.4 with comparable results. One
exception was the formation of the intermediate N-2-
(4-(N0,N0-dibenzylamino)phenyl)propyl-2-propanesul-
phonamide, which was obtained in significantly lower
yields of 21% instead of 63% as reported.4 This is
probably due to utilization of triethylamine as given in
the instructions for synthesis in the supporting in-
formation provided in Zarrinmayeh et al.,4 instead of
diaza(1,3)bicyclo[5.4.0]undecane (DBU) as originally
suggested.4
O
O
H
The intermediate N-2-(4-aminophenyl)propyl-2-pro-
pansulphonamide (6) was synthesized in five steps
from 4-aminobenzyl cyanide. Products 7 and 8 were
obtained through reaction of compound 6 with 4-
nitrobenzoyl chloride and 4-fluorobenzoyl chloride,
respectively (see Scheme 2). Yields were 70 and 90%
for compounds 7 and 8, respectively.
H2N
N
O
O
N
S
O2
MeO
O2N
N
H
3
1
O
N
N
HN
O
O
O
N
N
Purification of compounds 7 and 8 was not possible
with (flash-)column chromatography or preparative
thin-layer chromatography (TLC). Several eluent mix-
tures were applied using TLC and two of them, which
showed promising results, were utilized for (flash-)
column chromatography and preparative TLC. But
neither ethyl acetate/n-hexane (50/50; v/v) nor acet-
one/n-hexane (50/50; v/v) gave a significant raise in
purity. Best results were acquired through recrystalli-
zation from acetone/n-hexane, which raised the purity
up to ꢁ98% (determined by high-performance liquid
N
4
2
H
N
S
O
O2
H2N
N
H
5
Figure
1
Examples of AMPA antagonists and allosteric
modulators.
H
N
H
N
O
S
O
S
O2
O2
K+[2.2.2] 18F-
N
N
H
H
18F]8
7
[
18F
O2N
Scheme 1 Radiosynthesis of N-2-(4-N-(4-[18F]fluorobenzamido)phenyl)-propyl-2-propanesulphonamide [18F]8.
O
H
H
N
Cl
N
O
S
S
O2
O2
R
N
NEt3, CH2Cl2
H2N
H
7 R=NO2
8 R=F
6
R
Scheme 2 Synthesis of N-2-(4-N-(4-nitrobenzamido)phenyl)propyl-2-propanesulphonamide (7) and N-2-(4-N-(4-fluorobenzami-
do)phenyl)propyl-2-propanesulphonamide (8).
Copyright # 2007 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2007; 50: 1169–1175
DOI: 10.1002.jlcr

SYNTHESIS AND RADIOFLUORINATION OF A PUTATIVE AMPA RECEPTOR LIGAND 1171
chromatography, HPLC). To achieve higher grades of
purity, a preparative HPLC was deployed. Thereby
599% of pure compounds 7 and 8 could be isolated.
Since precursor 7 was not easily available in high-
grade purity, labelling reactions were performed using
the ꢁ98% pure product. Unless other numbers are
given, all labelling experiments were repeated three
times and average values are given.
Optimization of labelling conditions
After the production of
[
18F]fluoride through the
18O(p,n)18F nuclear reaction on enriched [18O]H2O,
the fluoride was azeotropically dried in the presence of
Kryptofix 2.2.21 and K2CO3 and then utilized in a
nucleophilic reaction8 with the nitroprecursor 7 to give
n.c.a. N-2-(4-N-(4-[18F]fluorobenzamido)phenyl)-pro-
pyl-2-propanesulphonamide ([18F]8).
Figure 2 Radiochemical yields for the [18F]fluoride-for-nitro
exchange on N-2-(4-N-(4-nitrobenzamido)phenyl)propyl-2-
propanesulphonamide at three different temperatures in
DMSO at a concentration of 0.025 mol/l.
No noteworthy formation of radioactive by-products
occurred until 30 min of reaction time. However, yields
of by-products varied between 2 and 10% after 60 min
depending on the reaction conditions.
In order to optimize the conditions of the radio-
fluorination reaction, different solvents, temperatures
and concentrations were used. Optimization started
with the variation of solvents at selected reaction
temperatures and times. Since the precursor is not
sufficiently soluble in acetonitrile, N,N–dimethylforma-
mide (DMF), dimethylacetamide (DMA) and dimethyl-
sulphoxide (DMSO) were tested. Results with DMF and
DMA were rather unsatisfactory. Even though the
temperature was risen from 130 to 1508C, radio-
chemical yields did not exceed 5% after 60 min and
several by-products occurred in radiochemical yields
as high as those of the desired product [18F]8. DMSO
proved to be the most suitable solvent for the aromatic
radiofluorination of 7. As shown in Figure 2, very high
temperatures and extended reaction times were neces-
sary in order to achieve satisfactory radiochemical
yields. Radio–TLC control showed that precursor 7 and
the labelled product [18F]8 were stable under these
conditions, and that within 30 min by-products oc-
curred only with small radiochemical yields. Therefore,
work was continued using these conditions.
Figure 3 Radiochemical yields for the [18F]fluoride-for-nitro
exchange on N-2-(4-N-(4-nitrobenzamido)phenyl)propyl-2-
propanesulphonamide at different concentrations of precur-
sor in DMSO at 1808C.
The concentration of precursor 7 was varied between
0.015 and 0.049 mol/l. Figure 3 illustrates that radio-
chemical yields rose with higher concentrations as
expected; however, significantly lower yields were
tions in the reaction solution. A concentration of
0.035 mol/l showed highest yields (see Figure 3). Thus,
in summary, the optimized reaction conditions corre-
spond to a concentration of 0.035 mol/l of precursor in
DMSO at a temperature of 1808C with an equimolar
amount of Kryptofix1. Radiochemical yields are
38 ꢀ 8% after 30-min reaction time (n ¼ 10) and raise
up to 63 ꢀ 5% after 60-min reaction time (n ¼ 5) (see
Figure 4).
observed at
a high concentration of 0.049 mol/l.
Because preliminary experiments showed higher yields
when using equimolar amounts of Kryptofix and
precursor, all experiments were performed under these
conditions. The lower yields at high concentrations
may be caused by the resulting high overall concentra-
Copyright # 2007 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2007; 50: 1169–1175
DOI: 10.1002.jlcr

1172 U. B. KRONENBERG ET AL.
a rate of 90%, it was not expected that the potentiator 8
would block the [3H]AMPA binding; however, it did not
potentiate the [3H]AMPA binding on frozen brain slices
while similar compounds like LY395153 (5, see Figure
1) potentiated AMPA receptor-mediated inward cur-
rents in cerebellar Purkinje neurons with an EC50 of
300 nmol at room temperature.4
Image scanning after the [18F]8 assay displayed a
uniform binding distribution, no influence of unla-
belled CNQX and only 30–40% blocking power by
macroscopic amounts of 8 on the binding. This leads
to the assumption that the radioligand binding repre-
sents a remarkable part of non-specific binding, which
was higher than that found with [3H]5 in rat cortex (10–
15%) or in recombinant human GluR4 receptors (20–
25%).4,5
Figure
4 Time dependence of the [
18F]fluoride-for-nitro
exchange on N-2-(4-N-(4-nitrobenzamido)phenyl)propyl-2-
propanesulphonamide at a concentration of 0.035 mol/l of
precursor in DMSO at 1808C.
Experimental
General
All chemicals were purchased from Aldrich (Taufkirch-
en, Germany), Merck (Darmstadt, Germany), Fluka
(Buchs, Suisse), KMF (St. Augustin, Germany) or Acros
Organics (Geel, Belgium) and were used without
further purification.
Preparative experiments
In order to save time during the preparation of the
desired product, the reaction was stopped with water
after 30 min. For isolation, a solid-phase extraction was
utilized, followed by semi-preparative reversed-phase
HPLC purification. The obtained product fraction of
eluent was passed though a Waters Sep-Pak1 Plus C18
cartridge and the retained product was eluted thereof
with ethanol.
For TLC, pre-coated plates of silica gel 60 (Polygram
SIL G/UV254 40 ꢂ 80 mm) were purchased from Ma-
cherey-Nagel (Germany). They were analysed at a
wavelength of 254 nm. The eluent used was acetone/
n-hexane (40/60; v/v). Radioactivity on TLC plates was
measured with an Instant ImagerTM (Packard).
Radioactive labelling with the objective of isolating
the desired product was performed with 5 mg
(12.5 mmol) of labelling precursor.
Mass spectra were measured on a Thermoquest
Automass Multi III Mass Spectrometer. 1H-, 13C- and
19F-NMR spectra were recorded on a Bruker DPX
Avance 200 using the signal of the appropriate
standard as reference. DMSO-d6 was used as solvent.
All chemical shifts are given in ppm. Elemental analysis
was done at the Central Institute of Analysis of the
The specific activity of product [18F]8 was measured
by comparing the area of the detected UV-peak to a
standard curve relating mass to UV-absorbance. If no
UV-peak could be detected, the detection limit was
used to determine the lower threshold of the specific
activity. The detection limit of the HPLC system used
for this purpose for compound 8 is at an amount of
1.75 pmol. Under these conditions, the specific activity
was 577 ꢀ 40 GBq/mmol (n ¼ 3).
¨
Forschungszentrum Julich GmbH with a Leco CHNOS-
932 Microelement analyser. Melting points were deter-
¨
mined with a Buchi B-540 melting point apparatus and
are uncorrected.
Analytical HPLC was performed on a system consist-
ing of a Knauer WellChrom Mini-Star K500 pump, a
Rheodyne 7125 injector block, a Merck/Hitachi L4000
UV/Vis photometer with the wavelength set to 254 nm
and a NaI(Tl) well-type scintillation detector (EG&G
Ortec, Modell 925 Scint Amplifier and Bias Supply).
Semi-preparative HPLC was performed on a different
system than the one mentioned above; it consisted of a
Merck/Hitachi L6000 pump, a Rheodyne 7125 injector
block, a Merck/Hitachi L4000 UV/Vis Photometer with
the wavelength set to 254 nm and a NaI(Tl) well-type
In vitro binding assays
Autoradiography of rat brain slices in the presence of 8
and [3H]AMPA revealed the typical AMPA receptor-
binding distribution9 with high binding in the hippo-
campus, followed by cortex striatum and cerebellum,
while low binding was detected in the thalamus.
Whereas the receptor antagonist 6-cyano-7-nitroqui-
noxaline-2,3-dione (CNQX) blocked the [3H]AMPA up to
Copyright # 2007 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2007; 50: 1169–1175
DOI: 10.1002.jlcr

SYNTHESIS AND RADIOFLUORINATION OF A PUTATIVE AMPA RECEPTOR LIGAND 1173
scintillation detector (EG&G Ortec, Modell 925 Scint
C–H); 124.36 (2 ꢂ Aryl C–H); 121.43 (2 ꢂ Aryl C–H);
52.19 (C–H); 50.36 (CH2); 40.44 (C–H); 20.04 (CH3);
17.17 (2 ꢂ CH3); mass spectrum (m/e): 405.9 (Mþ,
100%), 811 (2 ꢂ Mþ, 35%); analysis calculated for
Amplifier and Bias Supply).
Organic syntheses of precursor and standard
C
19H23N3O5S: C, 56.28%; H, 5.72%; N, 10.36%; O,
N-2-(4-Aminophenyl)propyl-2-propansulphonamide (6)
was synthesized with slight modifications according to
the procedure of preparation described by Zarrinmayeh
et al.4 in five steps starting from 4-aminobenzyl
cyanide. In order to protect the amino group, the
4-aminobenzyl cyanide was derivatized with benzyl
bromide which gave 68% of the protected product. The
reagent was then methylated in the a-position with
methyl iodide resulting in 96% of 2-(4-(N,N-dibenzyla-
mino)phenyl)propionitrile. This was quantitatively con-
verted into 2-(4-(N,N-dibenzylamino)phenyl)propylamine
utilizing a borane dimethyl sulphide complex. Reaction
with isopropanesulphonyl chloride gave 21% of the
corresponding N-2-(4-(N0,N0-dibenzylamino)phenyl)pro-
pyl-2-propanesulphonamide, which was deprotected
under standard conditions with hydrogen and palladium
on activated carbon to give 96% of compound 6.
19.73%; S, 7.91%; found: C, 55.10%; H, 5.74%; N,
10.30%; O, 22.40%; S, 7.79%.
Compound 8: Rf: 0.49; m.p.: 195–1988C; 1H NMR
(200.13 MHz, d6-DMSO), d (ppm): 10.23 (s, 1H, N–H);
8.05 (t, 2H); 7.70 (d, 2H); 7.38 (t, 2H); 7.24 (d, 2H); 7.06
(t, 1H, N–H); 3.09 (m, 1H); 2.85 (m, 2H), 1.17 (m, 9H);
13C NMR (50.32 MHz, d6-DMSO), d (ppm): 165.11 (C–O,
C–F); 140.69 (C); 138.19 (C); 132.27 (C); 131.22
(2 ꢂ Aryl C–H); 128.21 (2 ꢂ Aryl C–H); 121.31 (2 ꢂ Aryl
C–H); 116.18 (2 ꢂ Aryl C–H); 52.15 (C–H); 50.38 (CH2);
40.42 (C–H); 20.07 (CH3); 17.17 (2 ꢂ CH3); 19F NMR
(188.28 MHz, d6-DMSO),
spectrum (m/e): 379.2 (Mþ, 100%), 757.4 (2 ꢂ Mþ,
20%); analysis calculated for 19H23FN2O3S: C,
d (ppm): ꢃ109.38; mass
C
60.30%; H, 6.13%; N, 7.40%; O, 12.68%; S, 8.47%;
found: C, 59.30%; H, 6.41%; N, 7.45%; O, 14.30%; S,
8.28%.
Production of [18F]fluoride
N-2-(4-N-(4-Nitrobenzamido)phenyl)propyl-2-pro-
panesulphonamide (7) and N-2-(4-N-(4-fluoroben-
zamido)phenyl)propyl-2-propanesulphonamide (8)
[
18F]Fluoride was produced routinely at the BC 1710
(JSW) cyclotron at the Institute of Nuclear Chemistry of
A solution of 6 (0.79 or 0.68 mmol, respectively) in 5 ml
dichloromethane was treated with triethylamine
(1.5 eq) and 4-nitrobenzoyl chloride (1.5 eq) or 4-
fluorobenzoyl chloride (1.5 eq), respectively. The mix-
ture was stirred under argon at ambient temperature
until no residual educt could be detected via TLC.
Ten millilitre of water was added to stop the reaction
and the product was extracted with ether (3 ꢂ 10 ml).
The combined organic phases were washed with brine
(50 ml) and dried over Na2SO4. The solvent was
removed in vacuo and the remaining solid was recrys-
tallized from acetone/n-hexane to give 288 mg
(0.71 mmol; 90%; purity: 98% by HPLC) of light yellow
crystals of 7 and 180 mg (0.48 mmol, 70%; purity: 98%
by HPLC) of light pink crystals of 8, respectively.
Compounds 7 and 8 were purified using the semi-
preparative HPLC system utilizing a Nucleosil 100-7
C18 (250 ꢂ 20 mm) column under isocratic conditions
with an eluent mixture of acetonitrile/water (50/50; v/
v) at a flow rate of 8 ml/min.
¨
the Forschungszentrum Julich GmbH using the
18O(p,n)18F reaction on nuclear-enriched [18O]H2O.
The produced [18F]fluoride was purified through elec-
trostatic adsorption on a Sigradur1-Anode and follow-
ing desorption into pentadistilled water.10
Labelling synthesis
General
The labelling synthesis was carried out in a 2.5 ml
Wheaton reaction vial that was closed with a silicone
septum and screw cap. The vial was equipped with two
steel needles, one of which lead to an oil-sealed vacuum
pump, while the other led to the argon supply. The vial
was heated to about 808C under argon stream using an
oil bath, while the pressure was reduced to about
800 mbar. Depending on the amount of precursor to be
used, between 5 and 10 mg (13.28–26.56 mmol) of
Kryptofix 2.2.21 and 6.5–13 ml (6.5–13 mmol) of a 1 M
solution of potassium carbonate in water were mixed
with aqueous [18F]fluoride (between 80 and 120 MBq).
The solution was then mixed with 1 ml of acetonitrile
and transferred into the Wheaton vial. Under the
aforementioned pressure and temperature conditions,
the solvent was evaporated until dryness and the
residue was dissolved with 1 ml of acetonitrile. This
Compound 7: Rf : 0.46; m.p.: 182–1838C; 1H NMR
(200.13 MHz, d6-DMSO), d (ppm): 10.56 (s, N–H); 8.26
(t, 2H); 8.21 (m, 2H); 7.74 (d, 2H); 7.26 (d, 2H); 7.06 (t,
N–H); 3.11 (m, 3H); 2.87 (d, 1H); 1.22 (m, 9H); 13C NMR
(50.32 MHz, d6-DMSO),
d (ppm): 164.54 (C¼¼O);
149.98 (C–NO2); 141.50 (C); 141.18 (C); 137.84 (C);
131.39 (Aryl C–H); 130.06 (Aryl C–H); 128.29 (2 ꢂ Aryl
Copyright # 2007 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2007; 50: 1169–1175
DOI: 10.1002.jlcr

1174 U. B. KRONENBERG ET AL.
step was repeated twice in order to azeotropically
remove residual water. Then total vacuum was applied
for 5 min and the pressure afterwards set to ambient
with argon.
semi-preparative HPLC column. Using a reversed-
phase column (Kromasil 100-10 C18 250 ꢂ 10 mm)
with an eluent system consisting of tetrahydrofuran/
water (40/60; v/v) at a flow rate of 5 ml/min gave the
98% pure product at a k0-value of 8.1 (purity was
determined by analytical HPLC).
Optimization conditions
The two obtained product fractions were unified,
diluted with 5 ml water and passed through a Waters
Sep-Pak1 Plus C18 cartridge (first washed with 3 ml
ethanol and afterwards rinsed with 5 ml water) to
perform a solvent exchange. After washing the car-
tridge with 5 ml water and drying it under a moderate
argon stream for 2 min, the desired product was eluted
with 500 ml of ethanol. Radiochemical yield amounted
to 5 ꢀ 1.5% after a total reaction time of 120 min. The
product was 95% radiochemically pure.
The vial was transferred to a second oil bath with a
temperature of 130–1808C. Between 3 and 10 mg
(24.66 mmol) of precursor 7 was dissolved in 0.5–
1.6 ml of solvent (DMF, DMA or DMSO) and transferred
via a syringe into the reaction vessel.
The progress of the radioactive synthesis was con-
trolled with radio-TLC and HPLC. For this purpose,
after defined time intervals, aliquots of 10 ml were taken
with a Hamilton syringe, put into an Eppendorf vial
and then diluted with 30 ml of dichloromethane. As an
eluent for radio-TLC, acetone/n-hexane (40/60; v/v)
was used. On each silica TLC plate besides the labelled
product, a run was performed using the non-labelled
compound 8 to compare the Rf-values of the other
spots. Under these conditions the labelled product
For determination of the specific activity with analy-
tical HPLC, the detection limit of the analytical HPLC
system for the authentic compound 8 was used, since
no UV-peak was measurable. Based on the standard
curve this limit was reached at
1.75 pmol.
a threshold of
[
18F]8 had a Rf-value of 0.49, while fluoride remained at
An aliquot of 50 ml was taken from the ethanol
solution. The aliquot was transferred to a 1.5 ml
Eppendorf vial and its radioactivity was measured.
Relating to the above given threshold, the specific
activity was 577 ꢀ 40 GBq/mmol (n ¼ 3).
the starting point.
In order to analyse the course of the reaction, a
dedicated analytical HPLC system was established. A
mixture of 45/55 acetonitrile/water (v/v) as eluent and
a flow rate of 2 ml/min in combination with a Kromasil
100-5 C18 (250 ꢂ 4.6 mm) column offered best separa-
tion conditions.
Preliminary in vitro assays
Preparative runs
For autoradiography, frozen brains of Wistar rats were
cut horizontal into 20 mm thick sections at ꢃ188C (Leica
AG Microsystems, Germany), mounted onto gelatine-
coated object glasses (LO-Laboroptik GmbH) and
stored at ꢃ808C until use.
After azeotropically drying the aqueous [18F]fluoride
(between 80 and 160 MBq) in the presence of 5 mg of
Kryptofix and 6.5 ml of a 1 M solution of potassium
carbonate in water, the vial was transferred to a second
oil bath with a temperature of 1808C. For the pre-
parative synthesis of [18F]8 only 5 mg (12.33 mmol) of
precursor was used. Five milligram of precursor was
dissolved in 0.35 ml of DMSO and transferred to the
reaction vial via a syringe.
For ligand-binding assays, the brain sections were
thawed and dried at room temperature. After
pre-incubation for 15 min, at 48C in 50 mM Tris–HCl
containing 0.5 mM CaCl2, the cryosections were
additionally incubated with 50 mM KSCN and about
2.3 nM [3H]AMPA (Perkin Elmer) or about 6 nM [18F]8
in the buffer for 45 min at 48C. To determine the
non-specific binding, adjacent sections were incubated
in the presence of up to 15 mM 8 or CNQX (AMPA
antagonist). After this treatment, the sections were
washed twice for 15 s ([3H]AMPA) or for 1 min ([18F]8)
in buffer without KSCN at 48C, dipped in deionized
water, dried and exposed for four days ([3H]AMPA)
or 60 min ([18F]8) to a phosphor image plate (Fuji).
A laser phosphor imager (BAS 5000, Fuji) controlled
with software by the vendor (Version 3.14, Raytest,
Germany) scanned the images.
After 30 min of reaction time, the reaction was
stopped by adding 4.5 ml of water. The obtained
aqueous solution was then slowly passed through a
Phenomenex Strata1 C18-E cartridge (conditioned
with 3 ml ethanol and washed with 5 ml water) to
remove residual fluorine-18 and Kryptofix 2.2.21. The
cartridge was then washed with 5 ml water and dried in
a moderate argon stream for 2 min. After elution of the
product with 500 ml of acetonitrile, less than 5% of the
radioactivity was left on the cartridge. The solution was
injected into the semi-preparative HPLC system in two
fractions of 250 ml, in order to avoid overcharging the
Copyright # 2007 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2007; 50: 1169–1175
DOI: 10.1002.jlcr

SYNTHESIS AND RADIOFLUORINATION OF A PUTATIVE AMPA RECEPTOR LIGAND 1175
Conclusion
4. Zarrinmayeh H, Bleakman D, Gates MR, Yu H,
Zimmerman DM, Ornstein PL, McKennon T, Arnold
A
Kryptofix 2.2.21/K2CO3-activated nucleophilic
MB, Wheeler WJ, Skolnick P. J Med Chem 2001;
44: 302–304. DOI: 10.1021/jm000462n
radiofluorination reaction for the AMPA potentiator N-
2-(4-N-(4-[18F]fluorobenzamido)-phenyl)-propyl-2-pro-
panesulphonamide ([18F]8) was developed and its
radiochemical yield was optimized. It was produced
reliably in radiochemical yields of 5 ꢀ 1.5% within
120 min and a specific activity of 577 ꢀ 40 GBq/mmol.
From the first in vitro assays it can be concluded that
5. Linden AM, Yu H, Zarrinmayeh H, Wheeler WJ,
Skolnick P. Neuropharmacology 2001; 40:
1010–1018.
6. Vandecapelle M, Dumont F, De Vos F, Strijckmans K,
Leysen D, Audenaert K, Dierckx RA, Slegers G.
J Label Compd Radiopharm 2004; 47: 531–542.
DOI: 10.1002/jlcr.837
[
18F]8, due to its high non-specific binding and the
uniform binding distribution, appears to be a rather
unsuitable ligand to image AMPA potentiator-binding
sites.
7. Coenen HH. No-carrier-added 18F-chemistry of
radiopharmaceuticals. In Synthesis and Applica-
tions of Isotopically Labelled Compounds, Baillie TA
(ed.). Elsevier: Amsterdam, 1989; 443–448.
¨
¨
8. Coenen HH, Klatte B, Knochel A, Schuller M,
REFERENCES
¨
Stocklin G. J Label Compd Radiopharm 1986; 23:
1. Dingledine R, Borges K, Bowie D, Traynelis SF.
Pharmacol Rev 1999; 51: 7–61.
455–466.
9. Nielsen EO, Cha JH, Honore T, Penney JB, Young
AB. Eur J Pharmacol 1988; 157: 197–203. DOI:
10.1016/0014-2999(88)90383-4
2. Bliss TVP, Collingridge GL. Nature 1993; 361:
31–39. DOI: 10.1038/361031a0
3. Gitto R, Barreca ML, De Luca L, Chimirri A. Expert
Opin Ther Patents 2004; 14: 1199–1213.
10. Hamacher K, Blessing G. J Label Compd Radio-
pharm 1995; 37: 739.
Copyright # 2007 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2007; 50: 1169–1175
DOI: 10.1002.jlcr