Synthesis of an 18F-labelled high affinity β1- adrenoceptor PET radioligand based on ICI 89,406

Article abstract of DOI:10.1002/jlcr.1037

To date, some non-selective β-adrenoceptor (β-AR) positron emission tomography (PET) radioligands are in clinical use, but no PET radioligand for the selective imaging of cardiac β₁-ARs is clinically available. Therefore, the aim of this study

Full text of DOI:10.1002/jlcr.1037

JOURNAL OF LABELLED COMPOUNDS AND RADIOPHARMACEUTICALS
J Label Compd Radiopharm 2006; 49: 177–195.
Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jlcr.1037
Research Article
18
Synthesis of an F-labelled high affinity b1-
adrenoceptor PET radioligand based on ICI 89,406
Stefan Wagner1,*, Marilyn P. Law1, Burkhard Riemann1, Victor W. Pike2,
Hans-Jorg Breyholz1, Carsten Holtke1, Andreas Faust1,3, Christiane Renner1,
¨
¨
Otmar Schober1, Michael Schafers1,3 and Klaus Kopka1,3
¨
1 Department of Nuclear Medicine, University Hospital Munster, Albert-Schweitzer-Str.
¨
33, 48149 Munster, Germany
¨
2 Molecular Imaging Branch, National Institute of Mental Health, National Institutes of
Health, 10 Center Drive, Rm B3 C346A, MSC 1003, Bethesda, MD 20892-1003, USA
3 Interdisciplinary Center of Clinical Research (IZKF) Munster, Domagkstr. 3, 48149
¨
Munster, Germany
¨
Summary
To date, some non-selective b-adrenoceptor (b-AR) positron emission tomography
(PET) radioligands are in clinical use, but no PET radioligand for the selective imaging
of cardiac b1-ARs is clinically available. Therefore, the aim of this study was to develop
a potential high-affinity PET radioligand for the b1-subtype of ARs. Here, the synthesis
and in vitro evaluation of (S)- and (R)-N-[2-[3-(2-cyano-phenoxy)-2-hydroxy-propyla-
mino]-ethyl]-N0-[4-(2-fluoro-ethoxy)-phenyl]-urea (8a–b), derivatives of the well-charac-
terized b1-AR selective antagonist, ICI 89,406, are described. The (S)-isomer 8a shows
both higher b1-AR selectivity and b1-AR affinity than the (R)-enantiomer 8b
(selectivity: 40 800 vs 1580; affinity: KI1 ¼ 0:049 nM vs KI1 ¼ 0:297 nM). Therefore,
the 18F-labelled analogue 8e of compound 8a was synthesized. While the direct nucleo-
philic 18F-fluorination of the tosylate precursor 8d produced 8e in low radiochemical
yields (42.9% decay-corrected) and specific activities (43.5 GBq/mmol at the end of
synthesis (EOS), n ¼ 9) the alternative two-step synthesis of 8e from ethylene glycol
di-p-tosylate 9, [18F]fluoride ion and phenol precursor 8f gave satisfying results
(16.4 ꢀ 3.2% radiochemical yield (decay-corrected), 99.7 ꢀ 0.5% radiochemical
purity, 40 ꢀ 8 GBq/mmol specific activity at the EOS within 174 ꢀ 3 min from the
end of bombardment (EOB) (n ¼ 5)). Copyright # 2006 John Wiley & Sons, Ltd.
Key Words: ICI 89,406 derivative; b1-adrenoceptor selective ligand; PET radioligand
*Correspondence to: S. Wagner, Department of Nuclear Medicine, University Hospital Munster, Albert-
¨
Schweitzer-Str. 33, 48149 Munster, Germany. E-mail: stwagner@uni-muenster.de
¨
Contract/grant sponsor: Deutsche Forschungsgemeinschaft; contract/grant number: 656, A5
Contract/grant sponsor: Interdisciplinary center of Clinical Research Munster; contract/grant number: ZPG4b
¨
Contract/grant sponsor: Intramural Research Program of the National Institutes of Health
Copyright # 2006 John Wiley & Sons, Ltd.
Received 26 October 2005
Revised 17 November 2005
Accepted 17 November 2005

178
S. WAGNER ET AL.
Introduction
Postsynaptic b-adrenoceptors (b-ARs) are classified as rhodopsin/b2-AR-like
receptors that belong to one of three major subfamilies of the G-protein-
coupled-receptors (GPCRs).1 The b-AR family is subdivided into at least three
discrete subtypes, the b1-, b2-AR2 and the atypical b3-AR.3,4 Additionally, a
putative subtype has been identified in cardiac tissue, classified as the b4-AR.5
Within the left ventricle of the healthy human heart the b1- to b2-AR ratio is
approximately 3:1.6 In heart disease, both the b-AR density and the b1- to
b2-AR ratio may change. Several conditions, including hypertension, heart
failure, ischemia, hypertrophic and dilated cardiomyopathy (HCM, DCM)
are accompanied by a reduced b-AR density in the heart.7–12 In addition,
the failing human heart is often characterized by a selective reduction in
b1-adrenoceptors (b1-ARs) without change in b2-AR density.6
For this reason a non-invasive method for the visualization and quantifica-
tion of the b1-AR density rather than total b-AR density in the human heart is
of great interest in basic research and clinical application.13 PET is an unique
technique for this task, quantitatively and dynamically determining local
radioactivity concentrations in vivo from which receptor concentrations can
be calculated.14 Non-selective radioligands for the imaging of b-ARs with
positron emission tomography (PET), such as (S)-[11C]CGP 12177, whose
racemate was described by Delforge et al. in 1991,15 were used in PET studies
to quantify myocardial b-AR density in heart disease.11,16,17 Elsinga et al.
established (S)-[11C]CGP 12388, a straightforward accessible radioligand, for
non-selectively targeting b-ARs in vivo. In a recent study (S)-[11C]CGP 12388
was administered as an aerosol for PET visualization of pulmonary b-ARs in
healthy volunteers.18 The radiosynthesis of an 18F-labelled analogue of CGP
12388 has also been described.19,20 In contrast, the clinical application
of candidate b1-AR selective radioligands, such as (þ=ꢁ)-[11C]HX-CH 44,21
(S)-[11C]bisoprolol,22 or [11C]CGP 20712A23 and its (S)-enantiomer [11C]CGP
26505,24 is limited due to high non-specific binding, rapid metabolism or
a tissue uptake that does not reflect binding to b-ARs.21–24 In summary, no
b1-selective radioligand suitable for the non-invasive assessment of cardiac
b1-AR is clinically established for single photon emission computed
tomography (SPECT) or PET.
An early example of a b1-selective AR antagonist is the compound ICI 89,406
(Scheme 1). Membrane studies show that the (S)-enantiomer possess higher
affinity than the antipode, but even the application of the racemate produces
effective b1-AR blockade during exercise in patients with angina pectoris.30–32
ICI 89,406 was chosen as the lead structure for the development of new
b1-selective AR radiotracers for application in nuclear medical imaging. After
identification of several ICI 89,406 derivatives with high b1-AR binding
affinity33 two b1-AR radioligands labelled with 123I and 125I were synthesized
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 177–195

18F-LABELLED b1-ADRENOCEPTOR RADIOLIGAND BASED ON ICI 89,406
179
Scheme 1. Lead structure ICI 89,406 and its radiolabelled derivatives25–29
and evaluated in vitro and in vivo (I-ICI-H and I-ICI-COOH, Scheme 1).
Drawbacks of these radioiodinated compounds regarding binding specificity
and in vivo stability in the chosen animal models prevented further
evaluation.25,26 It seems that high non-specific binding and rapid deiodination
in vivo preclude the use of such iodinated ligands for SPECT or PET.
However, in vitro application of these radioligands seems possible, if
125I-labelled analogues are considered, Radioligands labelled with positron-
emitting 11C (t1=2 ¼ 20:4 min) or 18F (t1=2 ¼ 109:7 min) may be less susceptible
to rapid metabolism. An 11C-labelled ICI 89,406 derivative was recently intro-
duced by our group as a potential high-affinity b1-AR radioligand (ICI-OMe,
Scheme 1).27–29 At present this radioligand is being evaluated in vivo.
In a further step we aimed at the development of an 18F-labelled ICI 89,406
derivative due to the superior physical decay characteristics (longer half-life,
shorter average positron range) of 18F compared to 11C. The work presented
here includes the synthesis of (S)-N-[2-[3-(2-cyano-phenoxy)-2-hydroxy-
propylamino]-ethyl]-N0-[4-(2-[18F]fluoro-ethoxy)phenyl]-urea 8e as a putative
high affinity b1-AR PET radioligand as well as the in vitro evaluation of the
non-radioactive enantiomers 8a–b.
Results and discussion
The ureas 8a–d, that serve as precursor or reference compounds for the
18F-labelled target 8e, were synthesized in a seven- to nine-step sequence
(Schemes 2 and 3). The nucleophilic substitution at the homochiral glycidyl-3-
nitrobenzene sulphonates 2a and 2b with 2-cyano-phenol 1 under basic
conditions provided the corresponding oxiranes 3a and 3b in good yields
(85–92%). Comparison of the measured optical rotations with literature
values showed that the stereochemistry was retained; the absolute [a]D values
were within errors.34 The preparation of the second synthon, the hydro-
chlorides 7a–b, started with the aniline derivatives 4a–b that were prepared
in two steps similar to those previously described (steps not shown in
Scheme 2).35–38 Compounds 4a–b were converted into the corresponding
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 177–195

180
S. WAGNER ET AL.
Scheme 2. Synthesis of the chiral b1-AR ligands 8a–e
phenylisocyanates 5a–b with trichloromethyl chloroformate (51–94%). The
addition reaction of 5a–b with N-mono-(t.-butyloxycarbonyl)ethylenediamine
yielded the ureas 6a–b (86–89%). After deprotection of 6a–b that proceeded
nearly quantitatively (94–96%) the target compounds 8a–c were formed from
the resulting hydrochlorides 7a–b plus oxiranes 3a–b via nucleophilic ring
opening under basic reaction conditions. The chemical yields were low and
ranged from 9 to 22% due to extensive purification procedures for the ureas
8a–c (Scheme 2). An attempt to improve the yield of the bromide 8c starting
from the free amine 7c, that was made from 7b and NaOH, and oxirane 3b
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 177–195

18F-LABELLED b1-ADRENOCEPTOR RADIOLIGAND BASED ON ICI 89,406
181
Scheme 3. Alternative synthesis of the chiral b1-AR ligand 8c
Table 1. Inhibition constants and calculated b1-AR selectivities of the ligands deter-
mined by a radioligand binding assay using mouse ventricular membrane preparations,
plus calculated ligand log P and log D values
Compound
8a
KI1 (nM)a
KI2 (nM)a
2000 ꢀ 220
470 ꢀ 180
890 ꢀ 480
149 ꢀ 27
41 ꢀ 3
b1-selectivityb
40 800 ꢀ 7700
1580 ꢀ 550
14 100 ꢀ 780
453 ꢀ 106
log Pc
1.44
1.44
1.98
0.83
1.57
log Dc
0.17
0.17
0.71
ꢁ0.55
0.13
0.049 ꢀ 0.008
0.297 ꢀ 0.066
0.063 ꢀ 0.036
0.328 ꢀ 0.046
0.28 ꢀ 0.12
8b
8c
8fd
ICI 89,406d
149 ꢀ 86
a Displacement of specifically bound non-selective b-AR ligand [125I]ICYP binding at b1- and b2-ARs
expressed in mouse ventricular membrane preparations as mean ꢀ SEM, n ¼ 3.
b The ratios of the low- over the high-affinity inhibition constants ðKI2=KI1Þ indicate the b1-selectivities of the
non-radioactive b1-AR ligands, noted as mean ꢀ SEM, n ¼ 3.
c log P values of the neutral form and log D values calculated by ACD/log D Suite (log D ¼ log P at
physiological pH 7.4 with consideration of charged species).
d from Reference27
.
did not succeed and provided, besides a low yield of urea 8c (11%), mainly
polymeric side products (Scheme 3). Finally, the bromide 8c was converted
with silver p-toluenesulphonate into the tosylate precursor 8d in 46% yield
(Scheme 2).
To assess the structure–activity relationships (SAR) between the ligands
and b1-ARs, competition studies using [125I]iodocyanopindolol ([125I]ICYP)
and mouse ventricular membrane preparations were performed. The high- and
low-affinity IC50 values for the b1- and b2-ARs of the non-radioactive
3-aryloxy-2-propanolamine derivatives 8a–c were calculated by non-linear
regression analysis of membrane bound radioactivity. From the IC50 values
the high- and low-affinity inhibition constants (KI1 for the b1-ARs and KI2 for
the b2-ARs) were obtained by the method of Cheng–Prusoff39 using the
previously determined KD value of [125I]ICYP (32.3 ꢀ 1.9 pM)33 (Table 1). The
b1-selectivities of the unlabelled compounds 8a–c are defined by the ratios of
the low- to high-affinity inhibition constants (KI2=KI1). To indicate the change
in lipophilicities caused by the chemical modifications of the lead compound
ICI 89,406 the calculated log P and log D values (ACD/log D Suite) of
compounds 8a–c are additionally listed in Table 1. The log P value of the non-
radioactive target compounds 8a and 8b (log P ¼ 1:44) is similar to the value
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 177–195

182
S. WAGNER ET AL.
for (S)-CGP 12177 (log P ¼ 1:81)40 whose 11C-labelled counterpart is an
effective radioligand for imaging cell surface b-ARs in the human myocardium
with PET.16,17 The compounds 8a–c are characterized both by high b1-AR
binding affinities and b1-AR selectivities. They possess b1-AR affinities in the
subnanomolar range (KI1 ¼ 0:04920:297 nM) and favour b1-AR binding over
b2-AR binding resulting in high b1-selectivities (1580–40 800). The enantio-
mers 8a and 8b display a normal binding behaviour for b1-ARs, since the
(S)-enantiomer 8a possesses a higher affinity for this receptor subtype than
the (R)-enantiomer 8b (KI1 ¼ 0:049 nM vs KI1 ¼ 0:297 nM). Up to 100-fold
higher b-AR affinities of (S)-enantiomers compared to the corresponding
(R)-enantiomers have been documented.41,42 A second feature of the enan-
tiomer pair 8a and 8b is the 26-fold higher b1-AR selectivity of 8a (40 800 vs
1580). The (S)-enantiomer of the bromo substituted ligand 8c shows similar
b1-AR affinity but nearly three times less b1-AR selectivity than the (S)-fluoro
counterpart 8a (KI1 ¼ 0:063 nM vs KI1 ¼ 0:049 nM, 14 100 vs 40 800).
Obviously a more sterically demanding halogen (covalent radius Br ¼
114 pm, covalent radius F ¼ 72 pm) at the 2-ethoxy position in 8c reduces
the b2-AR binding affinity less dramatically than the small-sized fluorine in 8a
resulting in a higher b1-AR selectivity for 8a. In summary, each ligand 8a–c
possesses higher b1-AR binding affinities as well as b1-AR selectivities than the
racemic lead compound ICI 89,406 (KI1 ¼ 0:28 nM, selectivity: 149).
As a result of these binding affinity and selectivity measurements, the more
potent (S)-enantiomer 8a (in comparison to its (R)-enantiomer 8b) was chosen
for conversion into a corresponding 18F-labelled radioligand 8e expected
to have high affinity and selectivity for the b1-AR subtype, for the use with
PET. Two synthetic routes for the target compound 8e were explored. The
first one started with the tosylate precursor 8d and [18F]K(Kryptofix 2.2.2)F.
Via nucleophilic substitution, the 18F-labelled 8e was obtained in one step
(Scheme 2). This approach is however characterized by low radiochemical
yields and specific activities (42.9% (decay-corrected), 43.5 GBq/mmol at the
end of synthesis (EOS), n ¼ 9) that could not be improved by varying the
reaction parameters (e.g. precursor concentration, reaction time and tempera-
ture). The specific activities may have been greatly underestimated due to
incomplete HPLC-resolution of the radioligand 8e from the non-radioactive
by-products of the labelling procedure. Due to the low radiochemical yield this
approach was not investigated further. The second approach (Scheme 4), a two-
step synthesis, started with ethylene glycol di-p-tosylate 9 and [18F]K(Kryptofix
2.2.2)F to yield 1-[18F]fluoro-2-(tosyloxy)ethane 10 (radiochemical yield of
42.8 ꢀ 2.8% (noted as mean ꢀ SD), decay-corrected, n ¼ 5). This well-known
labelling synthon was used to alkylate the phenol compound 8f under basic
reaction conditions. Compound 8f was prepared as previously described.27
The second step provided the desired 8e with a radiochemical yield of
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 177–195

18F-LABELLED b1-ADRENOCEPTOR RADIOLIGAND BASED ON ICI 89,406
183
Scheme 4. Alternative synthesis of the chiral b1-AR radioligand 8e
38.3 ꢀ 5.5% (decay-corrected, n ¼ 5). The overall radiochemical yield of
8e for both steps was 16.4 ꢀ 3.2% (decay-corrected). The synthesis was
realized with a radiochemical purity of 99.7 ꢀ 0.5% in 174 ꢀ 3 min from
end of bombardment (EOB) and a calculated specific radioactivity of
40 ꢀ 8 GBq/mmol (30–50 GBq/mmol) at the EOS (n ¼ 5). In summary, the
two-step approach of 8e provided more satisfying results regarding the
radiochemical yield and specific radioactivity compared to the one-step
procedure and is now being used to evaluate this 18F-labelled compound in
animals with PET.
Experimental
General methods
All chemicals, reagents, and solvents for the synthesis of the compounds were
analytical grade and purchased from commercial sources. [125I]ICYP was
purchased from Perkin-Elmer.
The melting points (uncorrected) were determined on a Stuart Scientific
1
SMP3 capillary melting point apparatus. H-NMR and 13C-NMR spectra
were recorded on a Bruker ARX 300 and AMX 400 spectrometer, respectively.
Mass spectrometry was performed via a Varian MAT 212 (EI=70 eV)
spectrometer and a Bruker MALDI-TOF-MS Reflex IV (matrix: DHB). Exact
mass analyses were conducted on a Bruker MicroTof apparatus. ½aꢂD values
were determined on a Perkin-Elmer 341 polarimeter. Elemental analysis was
realized by a Vario EL III analyser. Radiosynthesis were partly carried out
using an automated PET tracer synthesizer (TRACERLab FxFDG Synthesizer;
GE Functional Imaging GmbH). The recorded data were processed by the
TRACERLab Fx software (GE Functional Imaging GmbH). Separation of
the radiosynthesized compounds was performed by gradient radio-HPLC
using a Knauer K-500 and a Latek P 402 pump, a Knauer K-2000
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 177–195

184
S. WAGNER ET AL.
UV-detector (wavelength 254 nm) and a Crismatec Na(Tl) Scintibloc 51 SP51
g-detector, a Nucleosil 100-5 C18 column (250 ꢃ 4.6 mm2) with a correspond-
ing precolumn (20 ꢃ 4.0 mm2) or a Nucleosil 100-10 C18 column (250 ꢃ
8 mm2). Sample injection was carried out using a Rheodyne injector block
(type 7125 incl. 200 ml or 1000 ml loop). The recorded data were processed by
the NINA version 4.9 software (GE Functional Imaging GmbH). The
radiochemical purities and the specific activities were acquired with a radio-
HPLC system composed of a Syknm S1021 pump, a Knauer K-2501 UV-
detector (wavelength 254 nm), a Crismatec Na(Tl) Scintibloc 51 SP51
g-detector, a Nucleosil 100-3 C18 column (200 ꢃ 3 mm2), a VICI injector
block (type C1 incl. 20 ml loop) and the NINA version 4.8, Rev. 4 software
(GE Functional Imaging GmbH).
Synthetic methods
Synthesis of oxiranes ð3a–bÞ27
General procedure. 2-Cyano-phenol 1 (2 eq.), (S)-glycidyl-3-nitrobenzene
sulphonate 2a or (R)-glycidyl-3-nitrobenzene sulphonate 2b (1 eq.) and
anhydrous K2CO3 (6 eq.) were refluxed in dry 2-butanone (3.8 ml/mmol
sulphonate) for 6–7 h and then stirred at RT for 16–72 h under an argon
atmosphere. The mixture was filtered and the filter cake was washed with
2-butanone. The combined filtrates were evaporated to dryness. The residue
was dissolved in water and CH2Cl2. The pH of the aqueous layer was adjusted
to be within the range 11–13 and the aqueous layer was extracted. The layers
were separated. The aqueous layer was twice extracted with CH2Cl2, the
combined organic layers were dried (Na2SO4) and the solvent evaporated off.
The residue was recrystallized from diisopropyl ether-CHCl3 (ꢄ3:2 v/v) at
ꢁ308C to provide 3a–b as colourless solids.
(S)-2-(2-cyano-phenoxymethyl)-oxirane ð3aÞ. Yield: 92%. Mp.: 868C; Lit.:
1
88–898C34. H-NMR (300 MHz, CDCl3): d [ppm]: 7.58–7.49 (m, 2H, HAryl),
7.10–7.01 (m, 2H, HAryl), 4.37 (dd, 2J ¼ 11:4 Hz, 3J ¼ 3:0 Hz, 1H, 1CH2), 4.12
2
3
(dd, J ¼ 11:4 Hz, J ¼ 5:4 Hz, 1H, 1CH2), 3.41–3.37 (m, 1H, CH), 2.93
2
3
(‘t’, J ¼ 4:5 Hz, 1H, 1CH2), 2.84 (dd, J ¼ 5:0 Hz, J ¼ 2:6 Hz, 1H, 1CH2).
13C-NMR (75.5 MHz, CDCl3): d [ppm]: 159.99, 134.35, 133.84, 121.42, 116.26,
112.80, 102.45, 69.47, 49.85, 44.54. ½aꢂ2D0 ¼ þ18:18 (c ¼ 1:0, EtOH); Lit.: ½aꢂ2D5
¼ þ17:78 (c ¼ 1:0, EtOH).34 Analytically calculated for C10H9NO2: C 68.56,
H 5.12, N 8.00. Found: C 68.80, H 4.96, N 8.38.
(R)-2-(2-cyano-phenoxymethyl)-oxirane ð3bÞ. Yield: 85%. Mp.: 85–868C.
1H-NMR (300 MHz, CDCl3): d [ppm]: 7.58–7.50 (m, 2H, HAryl), 7.09–7.01
(m, 2H, HAryl), 4.37 (dd, 2J ¼ 11:6 Hz, 3J ¼ 3:2 Hz, 1H, 1CH2), 4.12
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 177–195

18F-LABELLED b1-ADRENOCEPTOR RADIOLIGAND BASED ON ICI 89,406
185
2
3
(dd, J ¼ 11:6 Hz, J ¼ 5:3 Hz, 1H, 1CH2), 3.40–3.37 (m, 1H, CH), 2.93 (dd,
2J ¼ 4:8 Hz, J ¼ 4:2 Hz, 1H, 1CH2), 2.84 (dd, J ¼ 5:0 Hz, J ¼ 2:4 Hz, 1H,
1CH2). 13C-NMR (75.5 MHz, CDCl3): d [ppm]: 160.01, 134.35, 133.87, 121.44,
116.26, 112.79, 102.50, 69.49, 49.85, 44.55. ½aꢂ2D0 ¼ ꢁ15:88 (c ¼ 1:0, EtOH);
Lit.: ½aꢂ2D5 ¼ ꢁ16:08 (c ¼ 1:0, EtOH).34 Analytically calculated for C10H9NO2:
C 68.56, H 5.12, N 8.00. Found: C 68.93, H 5.07, N 8.16.
3
2
3
Synthesis of the isocyanate derivatives ð5a–bÞ
General procedure. A solution of 4-(2-fluoro-ethoxy)-aniline 4a, prepared
from 4-acetamido-phenol and toluene-4-sulphonic acid-(2-fluoro-ethyl ester)
in two steps similar to those previously described,35,36 or 4-(2-bromo-ethoxy)-
aniline 4b, prepared from 4-nitro-phenol and 1,2-dibromoethane in two steps
as previously described,37,38 in dry ethyl acetate (0.6–1.0 mmol/ml) was slowly
added to a solution of 1 eq. trichloromethyl chloroformate in dry ethyl acetate
(0.6 mmol/ml) at RT. The mixture was refluxed for 2 h and evaporated to
dryness. The residue was distilled in a Kugelrohr to provide the phenyliso-
cyanates 5a–b as colourless oils.
4-(2-fluoro-ethoxy)-phenylisocyanate ð5aÞ. Yield: 94%. Bp. (1.2 mbar): 41658C.
1
3
4
H-NMR (300 MHz, CDCl3): d [ppm]: 7.01 (dd, J ¼ 9:0 Hz, J ¼ 0:9 Hz, 2H,
3
4
2
HAryl), 6.86 (dd, J ¼ 9:0 Hz, J ¼ 0:9 Hz, 2H, HAryl), 4.73 (dt, J ¼ 47:4 Hz,
3J ¼ 4:2 Hz, 2H, FCH2), 4.18 (dt, J ¼ 27:6 Hz, J ¼ 4:2 Hz, 2H, CH2O).
3
3
13C-NMR (75.5 MHz, CDCl3): d [ppm]: 156.30, 126.66, 125.71, 115.65, 81.82
(d, J ¼ 170:4 Hz), 67.56 (d, J ¼ 19:9 Hz). 19F-NMR (282 MHz, DMSO-d6):
d [ppm]: ꢁ223.91. Analytically calculated for C9H8FNO2: C 59.67, H 4.45, N
7.73. Found: C 59.55, H 4.20, N 7.65.
1
2
4-(2-bromo-ethoxy)-phenylisocyanate ð5bÞ. Yield: 51%. Bp. (1.4 mbar): 160–
1708C. 1H-NMR (400 MHz, CDCl3): d [ppm]: 7.01 (d, 3J ¼ 9:0 Hz, 2H, HAryl),
6.84 (d, 3J ¼ 9:0 Hz, 2H, HAryl), 4.26 (t, 3J ¼ 6:6 Hz, 2H, CH2O), 3.62
(t, 3J ¼ 6:6 Hz, 2H, BrCH2). 13C-NMR (100.6 MHz, CDCl3): d [ppm]: 155.91,
126.79, 125.73, 115.65, 68.29, 28.84. MS (EI): m/z (intensity %): 243 (M*
,
+
100), 241 (M*+, 94), 135 (39), 134 (47), 109 (74), 107 (78), 106 (18). Ana-
lytically calculated for C9H8BrNO2: C 44.66, H 3.33, N 5.79. Found:
C 44.63, H 3.18, N 5.85.
Synthesis of the [2-[3-(4-(2-halogeno-ethoxy)-phenyl)-ureido]-ethyl]-carbamic
acid t.-butyl ester derivatives ð6a–bÞ
General procedure. 4-(2-Fluoro-ethoxy)-phenylisocyanate 5a or 4-(2-bromo-
ethoxy)-phenylisocyanate 5b in dry diethyl ether (0.3–0.7 mmol/ml) was added
slowly to 1 eq. N-mono-(t.-butyloxycarbonyl)ethylenediamine in dry diethyl
ether (1.3–1.4 mmol/ml) at 08C. The mixture was stirred for further 30 min at
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 177–195

186
S. WAGNER ET AL.
08C and cooled to ꢁ308C for 1–14 h. After filtration the filter cake was washed
with diethyl ether and dried in vacuo to provide [2-[3-(4-(2-halogeno-ethoxy)-
phenyl)-ureido]-ethyl]-carbamic acid t.-butyl ester 6a–b as colourless solids.
[2-[3-(4-(2-fluoro-ethoxy)-phenyl)-ureido]-ethyl]-carbamic acid t.-butyl ester
ð6aÞ. Yield: 86%. Mp.: 163–1648C. 1H-NMR (300 MHz, DMSO-d6): d [ppm]:
3
4
8.29 (s, broad, 1H, NH), 7.28 (dt, J ¼ 9:0 Hz, J ¼ 2:7 Hz, 2H, HAryl), 6.83
3
4
(dt, J ¼ 9:3 Hz, J ¼ 2:8 Hz, 2H, HAryl), 6.78 (s, broad, 1H, NH), 6.03
(t, 3J ¼ 5:6 Hz, 1H, NH), 4.69 (dt, 2J ¼ 47:7 Hz, 3J ¼ 4:0 Hz, 2H, FCH2), 4.14
3
3
3
(dt, J ¼ 30:0 Hz, J ¼ 3:9 Hz, 2H, CH2O), 3.11 (q, J ¼ 5:8 Hz, 2H, CH2),
2.99 (q, J ¼ 5:9 Hz, 2H, CH2), 1.37 (s, 9H, C(CH3)3). 13C-NMR (75.5 MHz,
3
DMSO-d6): d [ppm]: 155.84, 155.62, 152.90, 134.20, 119.56, 114.82, 82.34
(d, 1J ¼ 166:6 Hz), 77.75, 67.43 (d, 2J ¼ 19:1 Hz), 40.69, 39.22, 28.36.
19F-NMR (282 MHz, DMSO-d6): d [ppm]: ꢁ217.45. MS (MALDI-TOF):
m/z: 364 (M+Na)+. Analytically calculated for C16H24FN3O4: C 56.29,
H 7.09, N 12.31. Found: C 56.12, H 6.99, N 12.41.
[2-[3-(4-(2-bromo-ethoxy)-phenyl)-ureido]-ethyl]-carbamic acid t.-butyl ester
ð6bÞ. Yield: 89%. Mp.: 126–1278C. 1H-NMR (300 MHz, DMSO-d6): d [ppm]:
3
3
8.31 (s, 1H, NH), 7.27 (d, J ¼ 8:7 Hz, 2H, HAryl), 6.83 (d, J ¼ 9:0 Hz, 2H,
HAryl), 6.77 (s, broad, 1H, NH), 6.04 (t, 3J ¼ 5:3 Hz, 1H, NH), 4.23
(t, 3J ¼ 5:3 Hz, 2H, CH2O), 3.74 (t, 3J ¼ 5:6 Hz, 2H, BrCH2), 3.10
3
3
(q, J ¼ 6:0 Hz, 2H, CH2), 2.99 (q, J ¼ 5:8 Hz, 2H, CH2), 1.37 (s, 9H,
C(CH3)3). 13C-NMR (75.5 MHz, DMSO-d6): d [ppm]: 155.65, 155.61, 152.60,
134.35, 119.51, 115.06, 77.74, 68.20, 40.66, 39.21, 31.66, 28.34. MS (EI): m/z
(intensity %): 403 (M*+, 2), 401 (M*+, 94), 243 (100), 241 (100), 135 (28), 134
(32), 109 (67), 108 (51), 107 (69). Analytically calculated for C16H24BrN3O4:
C 47.77, H 6.01, N 10.45. Found: C 47.82, H 6.13, N 10.53.
Synthesis of the N-(2-amino-ethyl)-N0-[4-(2-halogeno-ethoxy)-phenyl]-urea
hydrochloride derivatives ð7a–bÞ
General procedure. [2-[3-(4-(2-Fluoro-ethoxy)-phenyl)-ureido]-ethyl]-carbamic
acid t.-butyl ester 6a or [2-[3-(4-(2-bromo-ethoxy)-phenyl)-ureido]-ethyl]-
carbamic acid t.-butyl ester 6b was dissolved in a c. HCl-MeOH mixture
(1:1 v/v, 0.6–0.7 mmol/ml). The solvents were evaporated in vacuo within 1 h.
The residue was treated with 25 ml anhydrous acetone and the solvent was
removed in vacuo. This procedure was repeated three times to provide the
hydrochlorides 7a–b as colourless solids that were dried in vacuo.
N-(2-amino-ethyl)-N0-[4-(2-fluoro-ethoxy)-phenyl]-urea hydrochloride ð7aÞ. Yield:
96%. Mp.: 2048C decomposition. 1H-NMR (300 MHz, DMSO-d6): d [ppm]: 8.91
1
+
3
4
(s, H, NH), 8.09 (s, broad, 3H, NH3 ), 7.31 (dt, J ¼ 9:3 Hz, J ¼ 2:9 Hz,
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 177–195

18F-LABELLED b1-ADRENOCEPTOR RADIOLIGAND BASED ON ICI 89,406
187
3
4
2H, HAryl), 6.83 (dt, J ¼ 9:0 Hz, J ¼ 2:8 Hz, 2H, HAryl), 6.64 (s, broad, 1H,
2
3
3
NH), 4.68 (dt, J ¼ 47:7 Hz, J ¼ 4:0 Hz, 2H, FCH2), 4.13 (dt, J ¼ 30:3 Hz,
3J ¼ 3:9 Hz, 2H, CH2O), 3.31 (‘d’, J ¼ 4:5 Hz, 2H, CH2), 2.86 (q, J ¼ 5:8 Hz,
3
2H, CH2). 13C-NMR (75.5 MHz, DMSO-d6): d [ppm]: 156.18, 152.99,
134.07, 119.63, 114.82, 82.40 (d, 1J ¼ 166:7 Hz), 67.42 (d, 2J ¼ 19:3 Hz),
39.53, 37.34. 19F-NMR (282 MHz, DMSO-d6): d [ppm]: ꢁ217.41. MS (MAL-
DI-TOF): m/z: 264 (M-HCl+Na)+, 242 (M-HCl+H)+. Analytically calcu-
lated for C11H17ClFN3O2: C 47.57, H 6.17, N 15.13. Found: C 47.22, H 6.01,
N 15.09.
N-(2-amino-ethyl)-N0-[4-(2-bromo-ethoxy)-phenyl]-urea hydrochloride ð7bÞ. Yield:
1
94%. Mp.: 1768C decomposition. H-NMR (400 MHz, DMSO-d6): d [ppm]: 8.85
+
3
(s, 1H, NH), 7.99 (s, broad, 3H, NH3 ), 7.31 (d, J ¼ 8:8 Hz, 2H, HAryl), 6.82
3
(d, J ¼ 7:6 Hz, 2H, HAryl), 6.55 (s, broad, 1H, NH), 4.22–3.73 (m, 4H, CH2O
and BrCH2), 3.31–2.85 (m, 4H, 2CH2). 13C-NMR (100.6 MHz, DMSO-d6):
d [ppm]: 156.12, 152.74, 134.18, 119.65, 115.04, 68.22, 43.37, 39.55, 37.33.
MS (MALDI-TOF): m/z: 326 (M-HCl+Na)+, 324 (M-HCl+Na)+, 304
(M-HCl+H)+, 302 (M-HCl+H)+. Analytically calculated for C11H17BrClN3O2:
C 39.02, H 5.06, N 12.41. Found: C 38.87, H 5.17, N 12.56.
N-(2-amino-ethyl)-N0-[4-(2-bromo-ethoxy)-phenyl]-urea ð7cÞ. 14.00 g (41.3 mmol)
N-(2-Amino-ethyl)-N0-[4-(2-bromo-ethoxy)-phenyl]-urea hydrochloride 7b were
dissolved in 150 ml water. The pH value was adjusted with 1 N NaOH to 9.0
and 200 ml CH2Cl2 were added. The aqueous layer was extracted and the two-
phase mixture was filtered. The crude product on the filter was washed with 50 ml
water and 50 ml CH2Cl2 and dried in vacuo over P2O5 to yield 7c as a colourless
1
solid. Yield: 82%. Mp.: 74–758C. H-NMR (300 MHz, DMSO-d6): d [ppm]:
7.33–7.26 (m, 2H, HAryl), 6.86–6.80 (m, 2H, HAryl), 6.33 (t, 3J ¼ 5:4 Hz, 1H, NH),
6.18 (s, broad, 1H, NH), 4.24–3.73 (m, 4H, CH2O and BrCH2), 3.31–2.84 (m, 4H,
2CH2), 2.72 (t, J ¼ 5:2 Hz, 2H, NH2). 13C-NMR (100.6 MHz, DMSO-d6):
3
d [ppm]: 156.05, 152.80, 134.06, 119.75, 115.05, 68.20, 47.88, 39.81, 37.70.
MS (MALDI-TOF): m/z: 304 (M+H)+, 302 (M+H)+. Analytically calculated
for C11H16BrN3O2 ꢅ 0.5 H2O: C 42.46, H 5.51, N 13.50. Found: C 42.81, H 5.48,
N 13.72.
Synthesis of the N-aryl-N0-[2-[3-aryloxy-2-hydroxy-propylamino]-ethyl]-urea
derivatives ð8a–cÞ
General procedure of Scheme 2 ð8a–cÞ. 1 eq. 3a or 3b, an equimolar amount
N-(2-amino-ethyl)-N0-[4-(2-halogeno-ethoxy)-phenyl]-urea hydrochloride 7a
or 7b and 1.05 eq. of 10 N NaOH were heated in n-propanol (1.9–2.8 ml/
mmol oxirane) and water (0.14–2.4 ml/mmol oxirane) up to 908C for 3.5 h.
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 177–195

188
S. WAGNER ET AL.
Purification procedure A (8a–b). The mixture was stored for crystallization at
þ48C, the product was filtered off, washed with water (100 ml) and diethyl
ether (100 ml), and was purified by silica gel chromatography (ethyl acetate–
MeOH 4:1). The product fraction was evaporated and recrystallized from
MeOH.
Purification procedure B (8c). Water was added (4.2 ml/mmol urea), the
mixture was extracted with ethyl acetate three times, the combined organic
layers were dried over Na2SO4, and the solvent was evaporated. The crude
product was purified by silica gel chromatography (ethyl acetate–MeOH 4:1).
The urea derivatives 8a–c were obtained as colourless solids.
Procedure of Scheme 3 (8c). 9.50 g (31.4 mmol) N-(2-Amino-ethyl)-N0-[4-(2-
bromo-ethoxy)-phenyl]-urea 7c were dissolved in 500 ml dry MeCN.
Then 5.50 g (31.4 mmol) (S)-2-(2-cyano-phenoxymethyl)-oxirane 3a were
added. The solution was refluxed for 10 h. In the meantime, the mixture
was decanted from the insoluble polymeric products three times. After 15 h
stirring at RT it was refluxed for 5 h. The solution was evaporated at RT
and the crude product was purified by silica gel chromatography (ethyl
acetate–MeOH 4:1).
(S)-N-[2-[3-(2-cyano-phenoxy)-2-hydroxy-propylamino]-ethyl]-N0-[4-(2-fluoro-
1
ethoxy)-phenyl]-urea ð8aÞ. Yield: 9%. Mp.: 142–1438C. H-NMR (300 MHz,
DMSO-d6): d [ppm]: 8.32 (s, 1H, NH), 7.69 (dd, 3J ¼ 7:7 Hz, 4J ¼ 1:7 Hz, 1H,
H
Aryl), 7.61 (ddd, 3J ¼ 8:9 Hz, 3J ¼ 7:2 Hz, 4J ¼ 1:5 Hz, 1H, HAryl), 7.30–7.23
(m, 3H, HAryl), 7.06 (dt, 3J ¼ 7:7 Hz, 4J ¼ 0:8 Hz, 1H, HAryl), 6.82
(d, 3J ¼ 9:0 Hz, 2H, HAryl), 6.04 (t, 3J ¼ 5:6 Hz, 1H, NH), 5.01
2
3
(s, broad, 1H, OH), 4.69 (dt, J ¼ 47:8 Hz, J ¼ 4:0 Hz, 2H, FCH2), 4.20–
3.90 (m, 5H, 2OCH2 and CH), 3.14 (q, J ¼ 5:8 Hz, 2H, CH2), 2.76–2.60 (m,
3
4H, 2CH2), 1.93 (s, broad, 1H, NH). 13C-NMR (75.5 MHz, DMSO-d6): d
[ppm]: 160.51, 155.65, 152.83, 135.13, 134.32, 133.77, 121.12, 119.41, 116.55,
114.83, 113.36, 100.81, 82.37 (d, 1J ¼ 166:6 Hz), 71.77, 68.08, 67.42
(d, 2J ¼ 19:3 Hz), 52.12, 49.50, 31.45. 19F-NMR (282 MHz, DMSO-d6):
d [ppm]: ꢁ217.36. MS (MALDI-TOF): m/z: 439 (M+Na)+, 417 (M+H)+.
Analytically calculated for C21H25FN4O4: C 60.57, H 6.05, N 13.45. Found:
C 60.22, H 5.94, N 13.35.
(R)-N-[2-[3-(2-cyano-phenoxy)-2-hydroxy-propylamino]-ethyl]-N0-[4-(2-fluoro-
1
ethoxy)-phenyl]-urea ð8bÞ. Yield: 10%. Mp.: 140–1418C. H-NMR (300 MHz,
3
4
DMSO-d6): d [ppm]: 8.32 (s, 1H, NH), 7.69 (dd, J ¼ 7:5 Hz, J ¼ 1:8 Hz, 1H,
3
3
4
H
Aryl), 7.61 (ddd, J ¼ 8:6 Hz, J ¼ 7:4 Hz, J ¼ 1:4 Hz, 1H, HAryl), 7.30–7.23
(m, 3H, HAryl), 7.06 (t, 3J ¼ 7:5 Hz, 1H, HAryl), 6.83 (dt, 3J ¼ 8:7 Hz,
4J ¼ 2:7 Hz, 2H, HAryl), 6.04 (t, J ¼ 5:4 Hz, 1H, NH), 5.00 (s, broad, 1H,
3
OH), 4.69 (dt, 2J ¼ 47:8 Hz, 3J ¼ 4:0 Hz, 2H, FCH2), 4.20–3.90 (m, 5H, 2OCH2
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 177–195

18F-LABELLED b1-ADRENOCEPTOR RADIOLIGAND BASED ON ICI 89,406
189
3
and CH), 3.14 (q, J ¼ 5:8 Hz, 2H, CH2), 2.75–2.60 (m, 4H, 2CH2), 1.91
(s, broad, 1H, NH). 13C-NMR (75.5 MHz, DMSO-d6): d [ppm]: 160.48, 155.60,
152.79, 135.09, 134.30, 133.74, 121.09, 119.39, 116.52, 114.81, 113.32, 100.80,
1
2
82.34 (d, J ¼ 165:5 Hz), 71.73, 68.06, 67.42 (d, J ¼ 19:3 Hz), 52.09, 49.48,
32.42. 19F-NMR (282 MHz, DMSO-d6): d [ppm]: ꢁ217.59. MS (MALDI-TOF):
m/z: 439 (M+Na)+, 417 (M+H)+. Analytically calculated for C21H25FN4O4:
C 60.57, H 6.05, N 13.45. Found: C 60.42, H 6.31, N 13.33.
(S)-N-[4-(2-bromo-ethoxy)-phenyl]-N0-[2-[3-(2-cyano-phenoxy)-2-hydroxy-
propylamino]-ethyl]-urea ð8cÞ. Yield: 22% (Scheme 2); 11% (Scheme 3). Mp.:
1
97–988C. H-NMR (300 MHz, DMSO-d6): d [ppm]: 8.33 (s, 1H, NH), 7.69
3
4
(dd, J ¼ 7:7 Hz, J ¼ 1:7 Hz, 1H, HAryl), 7.65–7.59 (m, 1H, HAryl), 7.26
(‘t’ 3J ¼ 9:0 Hz, 3H, HAryl), 7.06 (t, 3J ¼ 7:5 Hz, 1H, HAryl), 6.82
3
3
(d, J ¼ 9:0 Hz, 2H, HAryl), 6.04 (t, J ¼ 5:4 Hz, 1H, NH), 5.02 (s, broad,
3
1H, OH), 4.23 (t, J ¼ 5:4 Hz, 2H, CH2), 4.18–3.88 (m, 3H, CH2CH), 3.74
3
3
(t, J ¼ 5:4 Hz, 2H, CH2), 3.15 (q, J ¼ 5:8 Hz, 2H, CH2), 2.77–2.62 (m, 4H,
2CH2). 13C-NMR (75.5 MHz, DMSO-d6): d [ppm]: 160.47, 155.61, 152.53,
135.10, 134.46, 133.75, 121.11, 119.40, 116.53, 115.07, 113.33, 100.81, 71.22,
68.20, 67.96, 52.02, 49.34, 39.30, 31.68. MS (MALDI-TOF): m/z: 517
(M+K)+, 515 (M+K)+, 501 (M+Na)+, 499 (M+Na)+, 479 (M+H)+,
477 (M+H)+. Analytically calculated for C21 H25BrN4O4: C 52.84, H 5.28, N
11.74. Found: C 53.13, H 5.25, N 11.58.
Toluene-4-sulphonic acid 2-[4-(3-{2-[3-(2-cyano-phenoxy)-2-hydroxy-propyla-
mino]-ethyl}-ureido)-phenoxy]-ethyl ester ð8dÞ. Under an Ar-atmosphere
600 mg (1.26 mmol) (S)-N-[4-(2-bromo-ethoxy)-phenyl]-N0-[2-[3-(2-cyano-phe-
noxy)-2-hydroxy-propylamino]-ethyl]-urea 8c and 2.98 g (6.28 mmol) silver
p-toluenesulphonate were refluxed in 80 ml dry MeCN for 70 h. The mixture
was filtered and the filtrate was evaporated at RT. Then 30 ml of acetone
was added, the mixture was stirred for 5 min, filtered and the filtrate
was evaporated at RT. The crude product was purified by two silica gel
chromatographies (ethyl acetate–MeOH 4:1) and 8d was obtained as a pale
yellow oil that solidified at 48C. Yield: 46%. Mp.: 2528C decomposition.
1
3
H-NMR (400 MHz, CDCl3): d [ppm]: 7.76 (d, J ¼ 6:0 Hz, 2H, HAryl), 7.68
(s, broad, 1H, NH), 7.46–7.41 (m, 2H, HAryl), 7.31 (d, 3J ¼ 6:0 Hz, 2H, HAryl),
7.16 (d, 3J ¼ 6:6 Hz, 2H, HAryl), 6.95 (t, 3J ¼ 5:7 Hz, 1H, HAryl), 6.85
3
3
(d, J ¼ 6:6 Hz, 1H, HAryl), 6.59 (d, J ¼ 6:9 Hz, 2H, HAryl), 6.08 (s, broad,
1H, NH), 4.26–3.98 (m, 7H, 3OCH2 and CH), 3.39–3.34 (m, 2H, CH2), 2.95–
2.84 (m, 4H, 2CH2), 2.40 (s, 3H, CH3). 13C-NMR (100.6 MHz, CDCl3):
d [ppm]: 160.35, 157.11, 153.96, 145.04, 134.72, 133.55, 129.94, 129.16, 128.01,
125.84, 121.74, 121.32, 116.76, 115.13, 112.63, 101.73, 68.37, 66.98, 65.93,
51.32, 49.59, 31.97, 29.73, 21.67. MS (ESI-EM): m/z: 569.2055 (M+H)+
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 177–195

190
S. WAGNER ET AL.
Calculated for C28H32N4O7SH 569.2064; m/z: 591.1872 (M+Na)+ Calculated
for C28H32N4O7SNa 591.1884.
Radiochemical experiments
Production of [18F]fluoride ion and synthesis of [18F]K(Kryptofix 2.2.2)F.
No-carrier-added aqueous [18F]fluoride ion was produced on a CTI-RDS-111
cyclotron by irradiation of a 1.2 ml water target using 10 MeV proton beams
on 97.0% enriched [18O]water by the 18O(p,n)18F nuclear reaction. A typical
batch was 7.4 GBq of [18F]fluoride ion at EOB for currents of 32 mA and
irradiation time of 12 min. To recover the [18O]water, the batch of aqueous
[18F]fluoride ion was passed through an anion exchange resin
(Sep-Pak1 Light Waters AccellTM Plus QMA cartridge, preconditioned with
5 ml 1 M K2CO3 and 10 ml water). [18F]Fluoride ion was eluted from the resin
with a mixture of 15 ml 1 M K2CO3, 200 ml water for injection, and 800 ml
DNA-grade MeCN containing 15 mg Kryptofix1 2.2.2. Subsequently, the
aqueous [18F]K(Kryptofix 2.2.2)F solution was carefully evaporated to
dryness in vacuo.
Radiosynthesis of (S)-N-[2-[3-(2-cyano-phenoxy)-2-hydroxy-propylamino]-
ethyl]-N0-[4-(2-[18F]fluoro-ethoxy)phenyl]-urea ð8eÞ via direct nucleophilic
fluorination (one-step procedure). Compound 8e was prepared by treating
1.7 mg (2.98 mmol) tosylate precursor 8d with the carefully dried [18F]K(Kryp-
tofix 2.2.2)F residue in 1 ml DNA-grade MeCN at 848C for 5 min. After
cooling to RT the crude reaction mixture was diluted with 10 ml water for
injection and passed through a Waters Sep-Pak1 Light C18 cartridge. The
cartridge was washed with additional 10 ml water for injection, dried in
a He-flow for 3 min, followed by elution of the raw 8e with 1.5 ml MeOH. The
solution was evaporated to a volume of 0.2 ml and fractionized using a
gradient radio-HPLC procedure (conditions: l ¼ 254 nm; flow = 2 ml/min;
eluents: A = MeCN-H2O-TFA, 950/50/1, B = MeCN-H2O-TFA, 50/950/1;
column: Nucleosil 100-5 C18 (250 ꢃ 4.6 mm2) with corresponding precolumn
(20 ꢃ 4.0 mm2); time programme: eluent B from 92 to 30% in 45 min, halt
5 min, from 30 to 92% in 8 min) resulting in 8e with a radiochemical yield of
2.9% (decay-corrected) and a radiochemical purity >99% (retention time
Rt ¼ 35:0 min). The time of synthesis and purification was 91 min from the
EOB. The determined specific radioactivity was 3.5 GBq/mmol at the EOS. The
specific activity of 8e was estimated by comparing the peak area of the UV
channel of purified 8e with a standard curve of known concentrations of
reference compound 8a realized with a RP-HPLC system (conditions:
l ¼ 254 nm; flow = 0.3 ml/min; eluent: MeCN-H2O-TFA, 700/300/1; column:
Nucleosil 100-3 C18 (200 ꢃ 3 mm2)). Chemical identity of 8e was proved by
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 177–195

18F-LABELLED b1-ADRENOCEPTOR RADIOLIGAND BASED ON ICI 89,406
191
coinjection and coelution of 8e and non-radioactive counterpart 8a on the
mentioned HPLC system.
Radiosynthesis of (S)-N-[2-[3-(2-cyano-phenoxy)-2-hydroxy-propylamino]-
ethyl]-N0-[4-(2-[18F]fluoro-ethoxy)phenyl]-urea ð8eÞ via 1-[18F]fluoro-2-(to-
syloxy)ethane ð10Þ (two-step procedure). 4.0 mg (10.8 mmol) Ethylene glycol
di-p-tosylate 9 and the carefully dried [18F]K(Kryptofix 2.2.2)F residue were
heated at 848C in 1 ml DNA-grade MeCN for 4 min. The mixture was cooled
to RT, diluted with 10ml water for injection and passed through a Waters Sep-
Pak1 Light C18 cartridge. The cartridge was washed with additional 10 ml
water for injection and eluted with 1.0 ml DNA-grade MeCN to yield raw 10.
The solution was purified using a gradient radio-HPLC procedure (conditions:
l ¼ 254 nm; flow = 5 ml/min; eluents: A = MeCN-H2O-TFA, 950/50/1, B =
MeCN-H2O-TFA, 50/950/1; column: Nucleosil 100-10 C18 (250ꢃ 8 mm2); time
programme: eluent B from 92 to 30% in 20 min, halt 5 min, from 30 to 92% in
5 min). The product fraction of 10 (retention time Rt ¼ 19:4 min) was
evaporated to a volume of 2 ml, diluted with 10 ml water for injection and
passed through a Waters Sep-Pak1 Light C18 cartridge. After washing the
cartridge with additional 5 ml water for injection it was dried in a He-flow for
10 min, followed by elution of 10 with 0.4 ml DMF that was tempered to 1258C
before elution. Compound 10 was obtained with a radiochemical yield of
42.8ꢀ 2.8% (decay-corrected, n ¼ 5). 4.0 mg (10.8 mmol) Compound 8f,
prepared as previously described,27 in 0.1 ml DMF and 16.4ml 1 N NaOH
were added to the eluate of 10 (comment: the solution of 8f, NaOH in DMF
was prepared 3 h before reaction start and turned brown after 2 h). The reaction
mixture was stirred at 1258C for 15 min, diluted with 0.5 ml water for injection
and purified with a gradient radio-HPLC system (conditions: l ¼ 254 nm; flow
= 5 ml/min; eluents: A = MeCN-H2O-TFA, 950/50/1, B = MeCN-H2O-
TFA, 50/950/1; column: Nucleosil 100-10 C18 (250ꢃ 8 mm2); time programme:
eluent B from 92 to 30% in 45 min, halt 5 min, from 30 to 92% in 5 min). After
isolation of 8e (retention time Rt ¼ 28:4 min) the fraction was evaporated to a
volume of 2 ml, diluted with 12 ml water for injection and passed through a
Waters Sep-Pak1 Light C18 cartridge. The cartridge was washed with 5 ml
water for injection and eluted with 1 ml EtOH, that was tempered to 608C
before elution. For in vivo investigations the EtOH solution can be diluted with
an appropriate amount of 0.9% NaCl or water for injection, respectively. In
this second step the target compound 8e was obtained with a radiochemical
yield of 38.3 ꢀ 5.5% (decay-corrected, n ¼ 5). The overall radiochemical yield
of 8e for both steps was 16.4ꢀ 3.2% (decay-corrected). Target compound 8e
was prepared with a radiochemical purity of 99.7 ꢀ 0.5% in 174 ꢀ 3 min from
EOB. The determined specific radioactivity was 40ꢀ 8 GBq/mmol at the EOS
(n ¼ 5) (see under one-step procedure).
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 177–195

192
S. WAGNER ET AL.
Radioligand binding assay
Microsomes were prepared by homogenizing heart ventricles from DBA mice
at 48C for 90 s in buffer A (1 ml) containing 10 mM EDTA, 10 mM HEPES
and 0.1 mM benzamidine (pH 7.4), using a Polytron PT 3000 (Kinematica,
Lucerne, Switzerland). Homogenates were centrifuged at 45 000 ꢃ gmax for
15 min at 48C. The pellets were resuspended again in buffer B (1 ml) containing
1 mM EDTA, 10 mM HEPES and 0.1 mM benzamidine (pH 7.4) and
recentrifuged at 45 000 ꢃ gmax for 15 min at 48C. The pellets were resuspended
in buffer B (1 ml) and centrifuged at 10 000 ꢃ gmax for 10 min at 48C. The
supernatants were recentrifuged at 45 000 ꢃ gmax for 15 min at 48C. The pellets,
partially enriched membranes, were resuspended in buffer C (50 mM
Tris ꢅ HCl, 5 mM MgCl2 (pH 7.4)), and stored frozen at ꢁ808C. For
competition binding studies, the prepared membranes were resuspended in
buffer D (10 mM Tris ꢅ HCl, 154 mM NaCl, 0.1 mM ascorbic acid, pH 7.4).
15 mg of membranes were incubated with a constant concentration of
[
125I]ICYP (80 pM) and with varying concentrations (1 pM–100 mM) of
compounds 8a–c (Scheme 2). Reactions were conducted at 378C for 60 min.
Reactions were stopped by filtering onto Whatman GF/B filters and washed
with water for injection. The membrane bound radioactivity was determined
in a g-scintillation counter. Competition binding curves were analysed by non-
linear regression analysis as previously described43–45 using the XMGRACE
programme (Linux software). The data of the ligands 8a–c fitted a two-site
model significantly better than a one-site model (F ¼ 2:15, p50:05). The
resulting high- and low-affinity IC50 values of the non-radioactive 3-aryloxy-2-
propanolamine derivatives 8a–c were converted into the high- and low-affinity
inhibition constants (KI1 for the b1-ARs and KI2 for the b2-ARs) by the
method of Cheng–Prusoff39 using the experimentally determined KD value of
[
125I]ICYP (32.3 ꢀ 1.9 pM).33 The ratios of the low- to high-affinity inhibition
constants (KI2/KI1) yield the b1-selectivities of the unlabelled derivatives 8a–c
(Table 1). These compounds show a significantly higher affinity to b1- than to
b2-ARs. Additionally, the calculated log P and log D values (ACD/log D Suite)
for compounds 8a–c are listed in Table 1 to emphasise the changes of the
lipophilicities caused by the chemical modifications of the lead compound,
ICI 89,406.
Conclusion
The synthesis, radiosynthesis and in vitro pharmacology of (S)- and (R)-N-[2-
[3-(2-cyano-phenoxy)-2-hydroxy-propylamino]-ethyl]-N0-[4-(2-fluoro-ethoxy)-
phenyl]-urea (8a and 8b, respectively), and the 18F-labelled (S)-enantiomer
(8e), a potential new high-affinity b1-AR selective PET radioligand, are
reported. The non-radioactive counterpart 8a of the radiolabelled target
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 177–195

18F-LABELLED b1-ADRENOCEPTOR RADIOLIGAND BASED ON ICI 89,406
193
compound 8e displays a high b-AR binding potency as well as a high b1-AR
selectivity in murine myocardial membranes and a moderate lipophilicity
comparable with the known non-selective PET radioligand (S)-[11C]CGP
12177. The radiosynthesis of 8e was achieved in two alternative routes. The
two-step synthesis of 8e with 1-[18F]fluoro-2-(tosyloxy)ethane 10 as inter-
mediate shaped up as the superior route compared to the one-step procedure
starting with the tosylate precursor 8d. The two-step synthesis route yielded
the target compound 8e with reasonable radiochemical yields and specific
activities. The formulation of the preparations is suitable for preclinical in vivo
studies using small animal PET.
Acknowledgements
This work was partly supported by grants from the Deutsche Forschungsge-
meinschaft (DFG), Sonderforschungsbereich 656 MoBil Munster, Germany
¨
(project A5) and the Interdisciplinary Center of Clinical Research (IZKF)
Munster, Germany (project ZPG4b). Dr V.W. Pike is supported by the Intra-
¨
mural Research Program of the National Institutes of Health (the National
Institute of Mental Health). The authors gratefully thank Ms Monika Trub,
Ms Sandra Schroer and Mr Sven Fatum for technical assistance.
¨
References
1. Gether U. Endocr Rev 2000; 21: 90–113.
2. Lands AM, Arnold A, McAuliff JP, Luduena FP, Brown TG. Nature 1967; 214:
597–598.
3. Arch JR, Ainsworth AT, Cawthorne MA, Piercy V, Sennitt MV, Thody VE,
Wilson C, Wilson S. Nature 1984; 309: 163–165.
4. Bond RA, Clarke DE. Br J Pharmacol 1988; 95: 723–734.
5. Sarsero D, Molenaar P, Kaumann AJ, Freestone NS. Br J Pharmacol 1999; 128:
1445–1460.
6. Brodde OE, Michel MC. Pharmacol Rev 1999; 51: 651–690.
7. Castellano M, Bohm M. Hypertension 1997; 29: 715–722.
¨
8. Khamssi M, Brodde OE. J Cardiovasc Pharmacol 1990; 16(Suppl 5): S133–S137.
9. Brodde OE, Zerkowski HR, Doetsch N, Motomura S, Khamssi M, Michel MC.
J Am Coll Cardiol 1989; 14: 323–331.
10. Anthonio RL, Brodde OE, van Veldhuisen DJ, Scholtens E, Crijns HJ, van Gilst
WH. Int J Cardiol 2000; 72: 137–141.
11. Schafers M, Dutka D, Rhodes CG, Lammertsma AA, Hermansen F, Schober O,
¨
Camici PG. Circ Res 1998; 82: 57–62.
12. Yamada S, Ohkura T, Uchida S, Inabe K, Iwatani Y, Kimura R, Hoshino T,
Kaburagi T. Life Sci 1996; 58: 1737–1744.
13. Riemann B, Schafers M, Law MP, Wichter T, Schober O. Nuklearmedizin 2003;
¨
42: 4–9.
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 177–195

194
S. WAGNER ET AL.
14. Phelps ME, Mazziotta J, Schelbert HR (eds). Positron Emission Tomography and
Autoradiography: Principles and Applications for the Brain and Heart. Raven
Press: New York, 1986.
15. Delforge J, Syrota A, Lancon JP, Nakajima K, Loch C, Janier M, Vallois JM,
Cayla J, Crouzel C. J Nucl Med 1991; 32: 739–748.
16. Schafers M, Lerch H, Wichter T, Rhodes CG, Lammertsma AA, Borggrefe M,
¨
Hermansen F, Schober O, Breithardt G, Camici PG. J Am Coll Cardiol 1998; 32:
181–186.
17. Wichter T, Schafers M, Rhodes CG, Borggrefe M, Lerch H, Lammertsma AA,
¨
Hermansen F, Schober O, Breithardt G, Camici PG. Circulation 2000; 101:
1552–1558.
18. Van Waarde A, Maas B, Doze P, Slart RH, Frijlink HW, Vaalburg W, Elsinga PH.
Chest 2005; 128: 3020–3027.
19. Elsinga PH, van Waarde A, Jaeggi KA, Schreiber G, Heldoorn M, Vaalburg W.
J Med Chem 1997; 40: 3829–3835.
20. Elsinga PH, Doze P, van Waarde A, Pieterman RM, Blanksma PK, Willemsen
AT, Vaalburg W. Eur J Pharmacol 2001; 433: 173–176.
21. Valette H, Dolle F, Guenther I, Demphel S, Rasetti C, Hinnen F, Fuseau C,
Crouzel C. Nucl Med Biol 1999; 26: 105–109.
22. Soloviev DV, Matarrese M, Moresco RM, Todde S, Buonasera TA, Sudati F,
Simonelli P, Magni F, Colombo D, Carpinelli A, Kienle MG, Fazio F.
Neurochem Int 2001; 38: 169–180.
23. Elsinga PH, van Waarde A, Visser GM, Vaalburg W. Nucl Med Biol 1994; 21:
211–217.
24. van Waarde A, Meeder JG, Blanksma PK, Bouwer J, Visser GM, Elsinga PJ,
Paans AM, Vaalburg W, Lie KI. Int J Appl Instrum B 1992; 19: 711–718.
25. Riemann B, Law MP, Kopka K, Wagner S, Luthra SK, Pike VW, Neumann J,
Kirchhefer U, Schmitz W, Schober O, Schafers M. Nuklearmedizin 2003; 42:
¨
173–180.
26. Wagner S, Kopka K, Law MP, Riemann B, Pike VW, Schober O, Schafers M.
¨
Bioorg Med Chem 2004; 12: 4117–4132.
27. Wagner S, Law MP, Riemann B, Pike VW, Breyholz HJ, Holtke C, Faust A,
¨
Schober O, Schafers M, Kopka K. J Label Compd Radiopharm 2005; 48: 721–733.
¨
28. Kopka K, Law MP, Engelhardt S, Riemann B, Pike VW, Schober O, Schaefers
M, Wagner S. J Label Compd Radiopharm 2005; 48: S267.
29. Kopka K, Law MP, Breyholz HJ, Faust A, Holtke C, Riemann B, Schober O,
¨
Schafers M, Wagner S. Curr Med Chem 2005; 12: 2057–2074.
¨
30. Imperial Chemical Industries Limited, London (UK). Patent CH 605666 1978;
DE 2458908 1975. Chem Abstr 1976; 84: 43599.
31. Majid PA, Schreuder JE, de Feyter PJ, Roos JP. J Cardiovasc Pharmacol 1980;
2: 435–444.
32. Svendsen TL, Hartling O, Trap-Jensen J. Eur J Clin Pharmacol 1979; 15: 223–228.
33. Kopka K, Wagner S, Riemann B, Law MP, Puke C, Luthra SK, Pike VW,
Wichter T, Schmitz W, Schober O, Schafers M. Bioorg Med Chem 2003; 11:
¨
3513–3527.
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 177–195

18F-LABELLED b1-ADRENOCEPTOR RADIOLIGAND BASED ON ICI 89,406
195
34. Nicola M, Depaoli A, Inglesi M. Gazz Chim Ital 1990; 120: 393–396.
35. Clinton RO, Laskowski SC. J Am Chem Soc 1952; 74: 2226–2237.
36. Villa S, Barlocco D, Cignarella G, Papp GJ, Balati B, Takacs J, Varro A,
Borosy A, Keseru K, Matyus P. Eur J Med Chem Chim Ther 2001; 36: 495–506.
37. Wilson WC, Adams R. J Am Chem Soc 1923; 45: 539–540.
38. Marquet J, Cayon E, Martin X, Casado F, Gallardo I. J Org Chem 1995; 60:
3814–3825.
39. Cheng Y, Prusoff WH. Biochem Pharmacol 1973; 22: 3099–3108.
40. Pike VW, Law MP, Osman S, Davenport RJ, Rimordi O, Giardina D, Camici PG.
Pharm Acta Helv 2000; 74: 191–200.
41. Mutschler E. b-Adrenozeptor-Antagonisten. Arzneimittelwirkungen. Wissenschaf-
tliche Verlagsgesellschaft mbH: Stuttgart, 1996; 291–292.
42. Mehvar R, Brocks DR. J Pharm Pharmaceut Sci 2001; 4: 185–200.
43. Nanoff C, Freissmuth M, Schutz W. Naunyn Schmiedeberg’s Arch Pharmacol
¨
1987; 336: 519–525.
44. Engel G, Hoyer D, Berthold R, Wagner H. Naunyn Schmiedeberg’s Arch
Pharmacol 1981; 317: 277–285.
45. DeLean A, Hancock AA, Lefkowitz RJ. Mol Pharmacol 1982; 21: 5–16.
Copyright # 2006 John Wiley & Sons, Ltd.
J Label Compd Radiopharm 2006; 49: 177–195