PET-compatible endothelin receptor radioligands: Synthesis and first in vitro and in vivo studies

Article abstract of DOI:10.1016/j.bmc.2009.08.058

The expression and function of endothelin (ET) receptors is abnormal in cardiovascular diseases, tumor progression, and tumor metastasis. In this study, we prepared two [¹⁸F]-fluorinated derivatives of the non-peptide ET_A receptor an

Full text of DOI:10.1016/j.bmc.2009.08.058

Bioorganic & Medicinal Chemistry 17 (2009) 7197–7208
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
PET-compatible endothelin receptor radioligands: Synthesis and first
in vitro and in vivo studies
a
a
Carsten Höltke a,b, , Marilyn P. Law , Stefan Wagner , Klaus Kopka a,c,d, Andreas Faust a,b
,
*
Hans-Jörg Breyholz a, Otmar Schober a, Christoph Bremer b,c, Burkhard Riemann a, Michael Schäfers c,d
a Department of Nuclear Medicine, Albert-Schweitzer-Str. 33, University Hospital Münster, D-48149 Münster, Germany
b Department of Clinical Radiology, Albert-Schweitzer-Str. 33, University Hospital Münster, D-48149 Münster, Germany
c Interdisciplinary Center for Clinical Research (IZKF Münster), University of Münster, D-48149 Münster, Germany
d European Institute of Molecular Imaging (EIMI), University of Münster, D-48149 Münster, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
The expression and function of endothelin (ET) receptors is abnormal in cardiovascular diseases, tumor
progression, and tumor metastasis. In this study, we prepared two [18F]-fluorinated derivatives of the
non-peptide ETA receptor antagonist PD 156707 and evaluated their ET receptor binding potencies. Ex
vivo as well as in vivo biodistribution studies in mice were performed, as well as the metabolism of
the radiotracer, which was examined by metabolite analysis in mice and rats. All tested derivatives of
PD 156707 exhibited potent in vitro pharmacological characteristics with Ki values comparable to that
of the lead compound. The biodistribution studies showed a high accumulation of the tracer in bile
and intestine. In vivo we were able to show that the visualization of the heart as a major target organ with
high ETAR expression is possible.
Received 10 June 2009
Revised 12 August 2009
Accepted 28 August 2009
Available online 2 September 2009
Keywords:
Endothelin axis
Radiofluorination
PET imaging
Biodistribution studies
Binding studies
Ó 2009 Elsevier Ltd. All rights reserved.
1. Introduction
In some examples of mouse models of human atherosclerosis
(e.g., apolipoprotein E-deficient mice)19–22 as well as in different
Endothelin (ET) was first described by Hickey et al.1 and subse-
quently isolated by Yanagisawa et al.2 as a 21-amino acid peptide
with vasoactive potential. Subsequent studies explored the role of
the endothelins in a manifold of vascular diseases, including pul-
monary arterial hypertension, atherosclerosis, and congestive
heart failure.3–8 In addition, the role of endothelin as an angiogenic
stimulus in a number of oncologic pathologies has recently been
described.9–13
Three isoforms of the endothelin peptides exist (ET-1, ET-2, and
ET-3), exerting their effects via two associated G-protein coupled
receptors (ETAR, ETBR). While ETAR is primarily located on vascular
smooth muscle cells and is responsible for vasoconstriction and
cell proliferation, ETBR is located on smooth muscle cells and vas-
cular endothelial cells, causing vasodilation by the release of nitric
oxide and prostacyclin and is responsible for the clearance of ET-1
from plasma in, for example, lung tissue.14–18 This system of the
three endothelin peptides and the two endothelin receptors is re-
ferred to as the endothelin axis.11
cases of cancer23–27 ET receptor density is upregulated. In this con-
text, a method for the visualization of ET receptor density in af-
fected tissue would be highly valuable for clinical diagnosis and
the evaluation of therapy. The method of choice for imaging recep-
tor distribution in vivo is positron emission tomography (PET). The
radiolabeling of ET receptor agonists and antagonists for in vivo
imaging studies using PET has been reported. [11C]L-753037, a
1,3-bisarylindane-2-carboxylic acid based non-selective ET recep-
tor ligand, has been used for the in vivo imaging of ET receptors
in a dog heart.28 Another representative of this compound class
(SB-209670) has been radiofluorinated for imaging ET receptor dis-
tribution in rats.29 The widely used ETA receptor selective ligand
Atrasentan (ABT 627) was radiomethylated and has been utilized
for the PET imaging of the ETA distribution in a baboon.30 Recently,
two different ETAR antagonists derived from BMS 207940 were
radiolabeled with [18F]fluoride and [11C]CH3I, respectively, and
used for in vivo PET imaging in a baboon.31 Interestingly, none of
these approaches has been applied to in vivo imaging of ET recep-
tors in humans so far. Therefore, there is still a strong motivation
for research toward ET receptor radioligands for clinical applica-
tions. PD 156707 (3-benzo[1,3]dioxol-5-yl-5-hydroxy-5-(4-metho-
xyphenyl)-4-(3,4,5-trimethoxybenzyl)-5H-furan-2-one) and related
butenolide derivatives are potent ET receptor antagonists with both
* Corresponding author at present address: Department of Nuclear Medicine,
Radiochemistry, Albert-Schweitzer-Str. 33, University Hospital Münster, 48149
Münster, Germany. Tel.: +49 251 8347362; fax: +49 251 8347363.
E-mail address: carsten.hoeltke@uni-muenster.de (C. Höltke).
0968-0896/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2009.08.058

7198
C. Höltke et al. / Bioorg. Med. Chem. 17 (2009) 7197–7208
or (2-fluoroethyl)-4-methylbenzene sulfonate yield the desired c-
H3CO
H3CO
OCH3
ketoesters 2 and 3 (Scheme 1) in ꢀ60% yield. 3,4,5-Trimethoxy-
benzaldehyde 4 is commercially available, aldehydes 5 and 6 were
synthesized starting from 3,4-dimethoxy-5-hydroxybenzaldehyde
and dibromoethane or (2-fluoroethyl)-4-methylbenzene sulfonate,
respectively (Scheme 2). Reactions were conducted in dimethyl-
formamide as solvent and with cesium carbonate as base. Product
yields were 65–73% in case of 5 and 69–93% for 6. The higher yields
of 6 were always obtained when dibromoethane was used in ex-
B
H3CO
HO
O
C
O
A
O
O
cess. The coupling of the benzaldehydes and the
c-ketoesters is
Figure 1. Structure of lead compound PD 156707.
realized by an aldol condensation, saponification and cyclization
sequence. This conversion requires refluxing with sodium methox-
ide in methanol for several hours followed by acidification of the
reaction mixture with acetic acid and further refluxing. The bute-
nolides 7–10 (Scheme 3) were obtained in 29–53% yield. Multiple
byproducts were detected by TLC. Derivatives 7 and 9 are the de-
sired non-radioactive reference compounds with 19F-fluoroethoxy
substitution. For the preparation of the radiolabeling precursors,
the two bromoethoxy-substituted butenolides 8 and 10 are heated
to reflux with an excess of silver tosylate in acetonitrile, yielding
the tosylates 11 and 12 in ꢀ80% yield. Conversions at 150 °C with
less silver tosylate and DMF as solvent were always less efficient
(data not shown).
high affinity and ETAR selectivity.32,33 In this study we describe the
synthesis and first in vivo evaluation of radiofluorinated derivatives
of this compound.
2. Results and discussion
The objective of this work was the modification of an existing
lead structure (PD 156707, Fig. 1) with the PET-compatible radio-
nuclide [18F]fluorine for the detection and quantification of ETA
receptors in PET examinations. We have chosen a one-step radiola-
beling procedure using [18F]K(Kryptofix 2.2.2)F and tosylate pre-
cursor compounds. Suitable positions for the placement of these
leaving groups are found on aromatic rings B and C of the lead
compound (Fig. 1), where a methoxy group can be replaced by
an adequately substituted ethoxy group. This approach leads to
two distinct radiofluorination precursors. The two non-radioactive
19F-substituted reference compounds were synthesized by similar
strategies.
2.2. Radiochemistry
The two precursor tosylates 11 and 12 were converted into the
[
[
18F]-substituted target compounds 18F]7 and 18F]9 using
[ [
18F]K(Kryptofix 2.2.2)F and potassium carbonate in acetonitrile
(Scheme 4). Compound [18F]7 was produced with a radiochemical
yield of 21.4 6.1% (noted as mean SD, decay-corrected, n = 12).
The synthesis was achieved with
98.1 0.8% in 71 2 min from end of radionuclide production
and a calculated specific radioactivity of 35–50 GBq/ mol at the
end of synthesis (EOS, n = 12). Besides the injection peak the UV-
trace of the HPLC quality control of the purified [18F]7 fraction
showed only an additional peak for the carrier, from which the spe-
cific radioactivity was calculated. Compound [18F]9 was produced
by the same procedure as [18F]7. Due to the higher in vitro potency
of 7 compared to 9 (see below) an optimization of the radiosynthe-
sis of [18F]9 was not performed. Thus, the number of experiments
a radiochemical purity of
2.1. Chemistry
l
Key intermediates of all four target compounds are the
esters 1(a–c), 2 and 3 and the benzaldehydes 4–6 (Schemes 1
and 2). The -ketoesters 1–3 were prepared following procedures
described by Patt et al.,33 where the preparation of 1a is illustrated.
Replacing the methoxy group with benzyl group yields
c-keto
c
a
c
c
-ketoester 1b, which enables subsequent hydrogenolysis to
-ketoester 1c.34 Re-substitution reactions with dibromoethane
O
O
Pd/C, H2
O
O
O
O
O
O
O
HO
OCH
OCH
3
3
1b
1c
O
Br(CH2)2Br or
TosO(CH2)2F
O
O
O
R O
1
OCH
3
2 R1 = -CH2CH2F
3 R1 = -CH2CH2Br
Scheme 1. Preparation of c-ketoesters 2 and 3 starting from benzyl protected 1b. Hydrogenation yields the phenoxy derivative 1c, which subsequently is converted to the
haloethoxy products 2 and 3.

C. Höltke et al. / Bioorg. Med. Chem. 17 (2009) 7197–7208
7199
was reduced (n = 3), just to prove the possibility for a radioconver-
sion of the tosylate precursor 12 to the radiolabeled product [18F]9.
OCH3
OH
OCH3
OR2
H3CO
H3CO
Br(CH2)2Br or
TosO(CH2)2F
2.3. Biology
In vitro determination of receptor affinity and selectivity: The
affinities of the prepared butenolide derivatives 7–12 toward
endothelin receptors were determined by competition binding
studies using [125I]ET-1 and mouse ventricular membrane prepara-
tions. The binding of the non-selective ET receptor ligand [125I]ET-1
to ventricular membranes was specific, saturable and of high affin-
ity. Values for the dissociation constant (KD = 208 2 pM) and
the maximum number of binding sites (Bmax = 300 3 fmol/mg
O
O
5 R2 = -CH2CH2F
6 R2 = -CH2CH2Br
Scheme 2. Synthesis of haloethoxy benzaldehydes 5 and 6 from commercially
available 3,4-dimethoxy-5-hydroxy benzaldehyde.
O
OCH3
OR2
H3CO
+
O
O
O
R1O
H3CO
OR2
OCH3
O
H3CO
R1O
HO
I) MeONa, MeOH
II) AcOH, MeOH
1a R1 = -CH3
2
3
R1 = -CH2CH2F
R1 = -CH2CH2Br
O
O
O
O
4
5
6
R2 = -CH3
R2 = -CH2CH2F
R2 = -CH2CH2Br
7
8
R1 = -CH3, R2 = -CH2CH2F
R1 = -CH3, R2 = -CH2CH2Br
AgOTos
AgOTos
9
R1 = -CH2CH2F, R 2 = -CH3
10 R1 = -CH2CH2Br, R2 = -CH3
11 R1 = -CH3, R2 = -CH2CH2OTos
12 R1 = -CH2CH2OTos, R 2 = -CH3
Scheme 3. Aldol condensation of c-ketoesters 1–3 with benzaldehydes 4–6 and subsequent cyclization, yielding the butenolides 7–10. The bromo derivatives 8 and 10 are
further converted to the radiofluorination precursor tosylates 11 and 12 by reaction with silver tosylate.
18F
H3CO
OTos
H3CO
H3CO
O
O
H3CO
H3CO
H3CO
HO
O
HO
O
O
O
O
O
O
O
11
[18F]7
18
[
]
F K(Kryptofix 2.2.2)F
CH3CN
H3CO
H3CO
H3CO
H3CO
OCH3
OCH3
O
O
TosO
18 F
HO
O
HO
O
O
O
O
O
O
O
12
[18F]9
Scheme 4. Radiofluorination reactions of the precursor tosylates with [18F]K(Kryptofix 2.2.2)F.

7200
C. Höltke et al. / Bioorg. Med. Chem. 17 (2009) 7197–7208
Table 1
Affinities of the prepared compounds toward endothelin receptors given as mean SEM (n = 3)
Structure
Weight
538.52
Affinities (nM) SEMa
Selectivity
218:1
Log Dcalc
OCH3
H3CO
O
F
H3CO
OH
Ki(ETA) = 1.09 0.60
Ki(ETB) = 240 16
3.51
7
O
O
O
O
OCH3
H3CO
O
Br
H3CO
OH
599.42
538.52
599.42
690.71
690.71
506.50
Ki(ETA) = 0.24 0.10
Ki(ETB) = 170 50
708:1
40:1
3:1
4.05
3.51
4.05
4.37
4.37
3.29
8
O
O
O
O
OCH3
H3CO
OCH3
O
OH
Ki(ETA) = 2.43 0.95
Ki(ETB) = 98 53
9
O
F
O
O
O
OCH3
H3CO
OCH3
O
OH
Ki(ETA) = 25 2.7
Ki(ETB) = 72 52
10
O
Br
O
O
O
OCH3
H3CO
O
OTos
H3CO
OH
Ki(ETA) = 1.26 0.81
Ki(ETB) = 1.26 0.81
1:1
11
O
O
O
O
OCH3
H3CO
OCH3
O
OH
Ki(ETA) = 0.36 0.15
Ki(ETB) = 96 51
267:1
12
O
Tos O
O
O
O
OCH3
H3CO
OCH3
H3CO
IC50(ETA) = 0.3b
IC50(ETB) = 780b
2600:1
OH
PD 156707
O
O
O
O
The IC50 values of the lead compound PD 156707 are given for comparison. Selectivities are given as Ki(ETB)/Ki(ETA) and IC50(ETB)/IC50(ETA) in case of PD 156707. Log D values
were calculated using the Log D Suite program by ACDLabs/ChemSketch 6.0 distribution (log D = log P at pH 7.4).
a
n = 3
b
IC50 values from Patt et al. J. Med. Chem. 1997, 40, 1063.

C. Höltke et al. / Bioorg. Med. Chem. 17 (2009) 7197–7208
7201
protein) were used for the calculations as previously determined.34
Different concentrations of the prepared compounds were incu-
bated with [125I]ET-1 and a fixed amount of mouse ventricular
membranes. The membrane bound radioactivity was evaluated
using a
c-scintillation counter. Non-linear regression analysis
showed that the data of the compounds fitted a two-site model.
This model assumes that the ligand binds to two receptor subtypes
with two different affinities, resulting in a high- and a low-affinity
inhibition constants Ki(high) and Ki(low). In the case of endothelin
receptors most ligands are known to be ETAR selective, so the high-
affinity constant describes the interaction of the ligand with the
ETAR subtype and the low-affinity constant that with the ETBR sub-
type. The high- and low-affinity inhibition constants Ki(ETA) and
Ki(ETB) are displayed in Table 1. For comparison, the values of
the lead compound PD 156707 are also given as published.33 All
prepared butenolide derivatives show a high affinity to ET recep-
tors with Ki values ranging from 0.24 to 240 nM. Except for precur-
sor tosylate 11 all compounds exhibit higher affinities to ETAR than
to ETBR with ratios for Ki(ETB)/Ki(ETA) from ꢀ3.0 for compound 10
to ꢀ700 for compound 8. Interestingly, the site of modification
(para-substituted phenyl vs 5-substituted dimethoxy-benzyl) is
not largely influencing the affinities. The target compounds with
fluoroethoxy substitution 7 and 9 show affinities to the ETA recep-
tor of 1.1 and 2.4 nM, respectively, proving the possibility to ex-
change an aromatic methoxy function with a 2-fluoroethoxy
group without profoundly decreasing the receptor affinity. How-
ever, the selectivities of the ligands suffer from this modification.
For this reason, we chose [18F]7 for in vivo and ex vivo experi-
ments. This ligand still displays a selectivity to the ETA receptor
of 220:1, compared to a ratio of only 40:1 for 9. Notably, compound
9, with the fluoroethoxy substitution on the phenyl-site, shows an
ETAR affinity 10 times that of the comparable bromoethoxy-substi-
tuted compound 10, while the affinity values are reversed when
looking at the benzyl-substituted compounds 7 and 8. Bromoeth-
oxy compound 8 is about five times more affine to ETAR than flu-
oroethoxy derivative 7 and also shows the highest selectivity
with Ki(ETB)/Ki(ETA) ꢀ 700.
Ex vivo biodistribution studies in mice: All animal studies were
performed with [18F]7. Upon intravenous administration of the
radiotracer, high levels of radioactivity were found in liver, lung,
bile, and duodenum (Fig. 2). Uptake into the brain was negligible,
indicating that the radiotracer is not passing the blood–brain-bar-
rier, and uptake indices for spleen and muscle were lower than that
for plasma. Uptake indices for other tissues were higher than that
for plasma at early times. Relatively low levels of radioactivity
were detected in urine but an exceptionally high concentration
of radioactivity was seen in bile. Time activity curves (Fig. 2) show
that in all tissues except duodenum highest levels of radioactivity
were detected at the first sampling time (2 min). Subsequently,
radioactivity cleared rapidly from tissues. The values for duode-
num indicate an increase in radioactivity between 2 and 5 min
after injection of [18F]7 followed by a slow loss, which may be re-
lated to the excretion of radioactivity into the bile. The radioactiv-
ity in the duodenal contents was assessed at 30 min by transferring
the contents directly into weighed vials. Radioactivity of the con-
tents was high (uptake index 33 7, n = 5) compared to that of
the duodenal wall (uptake index 9 2, n = 7).
Figure 2. Radioactivity in tissues of mice after intravenous injection of [18F]7. Each
symbol represents data for an individual mouse. The lines are biexponential fits to
data with the exception of duodenum, stomach, and bile which were fitted by a
peak equation (Weibull, 4 Parameter, SigmaPlot) for illustration. The top panel
shows results for myocardium (j), plasma (d), spleen (N), and muscle (.). The
dotted line shows the biexponential fit to plasma data and is included in all panels.
The second panel shows liver (j), lung (N), and kidney (.). The third panel shows
duodenum (j), stomach (N), and colon (.). The bottom panel shows data for bile
(j) and urine (N).
Predosing mice with the ETAR selective antagonist PD 156707
slightly reduced the uptake indices for all tissues except for plasma
and duodenum which showed no changes and stomach which
showed a small increase (Fig. 3). The differences were significant
for myocardium and muscle (t-test, p <0.05, n = 7 for each group),
but the overall retention of radioactivity in the majority of tissues
assessed was very low and, therefore, the differences between ‘to-
tal’ (no predose) and ‘non-specific’ (with PD 156707 predose) bind-
ing was small. An ETA receptor targeted conjugate with the
fluorescent dye Cyanine 5.5 based on the same lead structure PD
156707 has been shown to highly specifically bind to target recep-
tors in mice.35 Due to its high hydrophilicity, caused by the sulfo-
nate-substituted nature of the dye, it also showed a much more
favorable renal excretion profile. This leads to the assumption that
a modification of the radiotracer concerning higher hydrophilicity
is a possible route to a more applicable PET-tracer for the in vivo
imaging of endothelin receptors.

7202
C. Höltke et al. / Bioorg. Med. Chem. 17 (2009) 7197–7208
to [18F]fluoride ion in mouse and rat.37 The formation of [18F]fluo-
ride, however, is usually accompanied by an accumulation of
radioactivity in the skeleton, which could not be confirmed by
small-animal PET examinations (see below). The observed polar
radiometabolites of [18F]7 are therefore likely to be [18F]fluoroeth-
anol and its oxidation products.
Plasma
Muscle
]
*
Kidney
Colon
2.4. PET scans
Myocardium
Spleen
]
*
Although myocardial uptake of radioactivity after injection of
[
18F]7 was low, the heart was successfully visualized by PET
Stomach
Liver
in vivo (Fig. 5). In conformity with the ex vivo experiments time
activity curves for tissues, with the exception of the liver, showed
uptake of radioactivity followed by a rapid decrease during the first
2 min, mirroring organ perfusion by radioactivity in the blood. The
increase in blood activity between 10 and 20 min (Fig. 5) may re-
flect the re-entry of radioactivity from tissue to vascular compart-
ments or the spillover from liver or bile due to movement of the
animal. Myocardial radioactivity was found to be higher than that
in blood between 2 and 15 min (heart-to-blood ratio = 1.5–3.0),
but in brain, lung, spleen, muscle, and kidney it was similar to or
lower than that in blood. Liver radioactivity increased to a maxi-
mum at 3–4 min and then showed a gradual decrease over the next
hour. Similarly, radioactivity in the bile reached a maximum
(ꢁ6000 cps mlꢂ1) at 3–4 min and then decreased, whereas urine
levels (200 cps mlꢂ1) showed very little change over 5–60 min
(data not shown).
Lung
Duodenum
0
1
2
3 5 10 15
Uptake index
Figure 3. Effect of the ETA receptor antagonist PD 156707 on radioactivity in tissues
of mice after intravenous injection of [18F]7. Water for injection (seven mice) or PD
156707 at 1.0 l
mol/kg (seven mice) was injected 10 min before [18F]7 (<1.0 nmol/
kg). Tissue radioactivity (mean uptake index with standard errors) at 30 min after
[
18F]7 is shown. *Significant difference (t-test, p <0.05).
Water for injection;
PD 156707.
Metabolism: Preliminary studies were carried out in which rat
3. Conclusion
plasma was incubated with ꢀ19 MBq of [18F]7 at 37 °C for 1, 2.5,
20, and 60 min. Plasma protein binding of the tracer amounted
to only 4.6–6.7% of applied radioactivity. HPLC analysis of the
supernatant detected only parent [18F]7, confirming the plasma
stability of the compound. Figure 4A shows results obtained for a
single rat after administration of 40 MBq of [18F]7 where serial
blood samples (2.5–60 min) were taken. Only parent [18F]7 was
detected at 2.5 min after injection. Very low amounts of radioactiv-
ity were extracted at 10 min after injection but again only parent
We successfully established the synthesis of two novel ETA
receptor affine radioligands by semiautomatic one-step radiofluo-
rination procedures. The ligand ([18F]7) was tested in biodistribu-
tion studies in rats and mice ex vivo and in vivo. Although
myocardial uptake of [18F]7 as an indicator of ETA receptor binding
was low, a significant reduction in tracer uptake of target tissue
could be demonstrated by predosing experiments with the ETA
receptor ligand PD 156707. Also, we were able to visualize myocar-
dial tracer uptake by small-animal PET experiments in mice. How-
ever, the very low uptake of [18F]7 into target tissue makes it
difficult to determine specificity of the tracer in vivo. The rapid
metabolism into polar radiometabolites and a major hepatobiliary
route of excretion are the major drawbacks for that matter. There-
fore, the improvement of stability and enhancement of hydrophi-
licity are among the major strategies in future synthetic work.
Disease models with locally upregulated expression of endothelin
receptors, on the other hand, may provide a useful platform to
more closely examine the binding potency of the designed tracer
in vivo.
[
18F]7 was clearly observed, although there was possibly a polar
metabolite with tR = 3.6 min. At 20 min parent [18F]7 was detected
but there were at least two polar radioactive metabolites (3.6 and
4.4 min). Plasma radioactivity at 60 min was below detection lim-
its on the used HPLC system. Radioactive metabolites (3.6, 4.5, and
6.2 min), however, were observed in urine at 60 min.
Figure 4B shows HPLC analysis of plasma and urine samples ta-
ken at 5 and 20 min after intravenous injection of [18F]7 into mice.
For each time point blood from two mice was combined to give
1 ml aliquots for preparation of plasma. The urine sample was ta-
ken from one mouse and injected directly onto the HPLC system.
Five minutes after injection parent [18F]7 was detected (elution
time 26.0 min) in plasma but not in urine. Two more polar radioac-
tive metabolites were detectable at 17.1 min in plasma and at
3.0 min in urine. At 20 min parent [18F]7 was still present in plas-
ma but could not be detected in urine. One major polar radiome-
tabolite was detectable in plasma at 3.6 min, accompanied by
two faintly detectable signals at 11.7 and 17.1 min. In urine three
polar metabolites could be detected at 3.2, 4.1, and 5.9 min.
These results indicate that in mice and rats the radiotracer
4. Experimental
4.1. General
All chemicals, reagents and solvents for the synthesis of the
compounds were analytical grade and purchased from commercial
sources. PD 156707 and 2-(benzo-[1,3]-dioxol-5-yl)-1-(4-benzyl-
oxyphenyl)-4-oxobutyric acid methyl ester (1b) were synthesized
as reported.34,33 (2-Fluoroethyl)-4-methylbenzene sulfonate was
synthesized as described by Parenty et al.38 Melting points (uncor-
rected) were determined on a Stuart Scientific SMP3 capillary
melting point apparatus. 1H and 13C NMR spectra were recorded
on a Bruker ARX 300, and AMX 400 spectrometer, respectively.
Mass spectrometry was performed using a Varian MAT 212
(EI = 70 eV) spectrometer and a Bruker MALDI-TOF-MS Reflex IV
[
18F]7 is rapidly converted to polar radiometabolites which are
detectable in plasma at early times after iv injection and at later
times (>5 min) accumulate in urine. A possible metabolic pathway
is described by Zoghbi et al. They proposed the defluoroethylation
of a radiofluorinated dopamine transporter ligand, resulting in the
formation of 2-[18F]fluoroethanol and, in turn, its oxidation prod-
ucts
[ [
18F]fluoroacetaldehyde and 18F]fluoroacetate.36 Further-
more, Sykes et al. described that [18F]fluoroacetate is metabolized

C. Höltke et al. / Bioorg. Med. Chem. 17 (2009) 7197–7208
7203
A
Rat
B
Mouse
300
200
100
0
plasma 2.5 min
160
120
80
40
0
plasma 5 min
1500
1000
500
0
0
10
20
30
40
1500
1000
500
0
0
10
20
30
40
0
10
20
30
40
0
10
20
30
40
40
30
plasma 10 min
25
20
urine 5 min
20
1500
1000
500
0
15
5000
0
10
20
30
40
0
10
20
30
40
2500
0
0
10
20
30
40
0
10
20
30
40
80
60
plasma 20 min
30
25
plasma 20 min
40
20
20
15
1000
1500
1000
500
0
0
10
20
30
40
0
10
20
30
40
500
0
0
10
20
30
40
0
10
20
30
40
urine 20 min
urine 60 min
100
75
50
25
0
80
60
40
20
0
10
20
30
40
0
10
20
30
40
4000
2000
0
4000
2000
50
0
0
10
20
30
40
0
10
20
30
40
Elution time (min)
Elution time (min)
Figure 4. Rat (left) and murine (right) radio- and UV–HPLC-traces of plasma and urine samples taken at the denoted time points after tail vein injection of [18F]7. The vertical
line at ꢀ26 min indicates the elution time of the reference compound 7, which has been co-injected for identification.
(matrix: DHB). Elemental analysis was realized by a Vario EL III
analyzer. Separation of the radiosynthesized compounds, analyses
of radiochemical yields and radiochemical purities were performed
by gradient radio-RP-HPLC using a Knauer system with two K-1800
pumps, an S-2500 UV detector (Herbert Knauer GmbH, Berlin, Ger-
many) and a RP-HPLC Nucleosil 100-5 C18 column (250 mm ꢃ
4.2. Chemistry
4.2.1. 2-(1,3-Benzodioxol-5-yl)-1-(4-hydroxyphenyl)-4-
oxobutyric acid methyl ester 1c
An amount of 12.5 g (30.0 mmol) 2-(1,3-benzodioxol-5-yl)-1-
(4-benzyloxyphenyl)-4-oxobutyric acid methyl ester (1b) is sus-
pended in 200 ml of a 1:1 mixture of ethanol and ethyl acetate.
To this is added 150 mg of hydrogenation catalyst (Pd/C 10%,
FLUKA) and the mixture is stirred under a hydrogen atmosphere for
48 h. The catalyst is filtered off and the solvent is evaporated in va-
cuo. The remaining solid is recrystallized from methanol, giving
8.13 g of colorless crystals (24.8 mmol, 83%), mp 155 °C. 1H NMR
(CDCl3) d = 9.78 (br, 1H), 7.83 (d, 2H, J = 8.3 Hz), 6.85 (m, 3H),
6.75 (m, 2H), 5.90 (s, 2H), 4.35 (dd, 1H, J = 3.8 Hz, J = 10.1 Hz),
4.6 mm) coupled to a raytest GabiStar c-detector (raytest Isotopen-
messgeraete GmbH, Straubenhardt, Germany). The conditions used
are described in detail below. The recorded data were processed by
the ChromGate HPLC software (Knauer). Radioactivity of tissue was
determined using an automated
c
-counter (Wallac Wizard 300,
Perkin–Elmer Life Sciences, Boston, USA). All animal experiments
were conducted in accordance with local institutional guidelines
for the care and use of laboratory animals.

7204
C. Höltke et al. / Bioorg. Med. Chem. 17 (2009) 7197–7208
Figure 5. Small-animal PET scan of a wild type mouse (2 h pi) after tail vein injection of 6.67 MBq of [18F]7. High signal intensity is found mainly in liver and intestine.
Although uptake of radioactivity in the myocardium was low the heart could be visualized (values: cps mlꢂ1). Regions of interest were drawn in the left ventricle (blood s),
round the heart (d), in the lung (N), and in the liver (j). Symbols indicate values (cps mlꢂ1) for each time frame, lines indicate empirical fits to the data.
3.77 (dd, 1H, J = 10.1 Hz, J = 17.8 Hz), 3.64 (s, 3H), 3.18 (dd, 1H,
J = 3.8 Hz, J = 17.9 Hz) ppm. 13C NMR d = 195.0, 173.1, 161.7,
147.0, 146.0, 131.2, 129.6, 127.3, 120.2, 114.6, 107.5, 107.2,
100.2, 51.1, 45.1, 41.3 ppm. MS: m/z = 296 [MꢂMeOH]+, 328 [M]+.
Anal. Calcd for C18H16O6: C, 65.85; H, 4.91. Found: C, 65.98; H, 5.04.
4.19 (dd, 1H, J = 4.1 Hz, J = 10.2 Hz), 3.82 (dd, 1H, J = 10.2 Hz,
J = 17.9 Hz), 3.68 (s, 3H), 3.65 (t, 2H, J = 6.1 Hz), 3.20 (dd, 1H,
J = 4.1 Hz, J = 17.9 Hz). 13C NMR (CDCl3) d = 196.0, 173.9, 162.1,
148.0, 147.0, 132.1, 130.4, 130.2, 121.1, 114.4, 108.5, 108.2,
101.2, 67.9, 52.3, 46.0, 42.5, 28.6. MS (EI): m/z = 459 [M+Na]+
(
81Br), 457 [M+Na]+ 79Br), 437 [M+H]+ 81Br), 435 [M+H]+ 79Br),
( ( (
4.2.2. 2-(1,3-Benzodioxol-5-yl)-1-[4-(2-fluoroethoxy)phenyl]-4-
oxobutyric acid methyl ester 2
282, 234. Anal. Calcd for C20H19BrO6: C, 55.19; H, 4.40. Found: C,
55.20; H, 4.43.
A suspension of 1.64 g (5.0 mmol) 2-(1,3-benzodioxol-5-yl)-1-
(4-hydroxyphenyl)-4-oxobutyric acid methyl ester (1c), 1.20 g
(5.5 mmol) (2-fluoroethyl)-4-methylbenzene sulfonate and 3.25 g
(10.0 mmol) cesium carbonate in 50 ml of DMF are warmed to
80 °C for 1 h. The mixture is poured into 100 ml of water, neutral-
ized with acetic acid and extracted with dichloromethane. The
combined organic layers are washed with water and brine and
dried over magnesium sulfate. After filtration the solvent is re-
moved and the crude residue purified by silica gel column chroma-
tography (cyclohexane/ethyl acetate 3/2, Rf = 0.54). Yield 1.20 g
(2.93 mmol, 59%), mp 124 °C. 1H NMR (CDCl3) d = 7.93 (d, 2H,
J = 8.8 Hz), 6.93 (d, 2H, J = 8.8 Hz), 6.84–6.72 (m, 3H), 5.92 (s, 2H),
4.75 (dm, 2H, J = 47 Hz), 4.24 (dm, 2H, J = 28 Hz), 4.17 (dd, 1H,
J = 3.6 Hz, J = 10.1 Hz), 3.81 (dd, 1H, J = 10.1 Hz, J = 18.1 Hz), 3.67
(s, 3H), 3.18 (dd, 1H, J = 3.6 Hz, J = 18.1 Hz). 13C NMR (CDCl3)
d = 196.0, 174.0, 162.4, 147.9, 146.9, 132.0, 130.4, 129.9, 121.1,
114.2, 108.5, 108.1, 101.1, 81.5 (d, J = 171 Hz), 67.1 (d, J = 20 Hz),
52.3, 46.0, 42.5. 19F NMR (CDCl3) d = ꢂ224, MS (EI): m/z = 397
[M+Na]+. Anal. Calcd for C20H19FO6: C, 64.17; H, 5.12. Found: C,
63.98; H, 5.21.
4.2.4. 3-(2-Fluoroethoxy)-4,5-dimethoxybenzaldehyde 5
A mixture of 1.00 g (5.00 mmol) 3,4-dimethoxy-5-hydroxy-
benzaldehyde, 1.79 g (5.5 mmol) cesium carbonate and 1.20 g
(5.5 mmol) (2-fluoroethyl)-4-methylbenzene sulfonate in 50 ml
of DMF is heated to 80 °C for 4 h. The mixture is poured into
150 ml of diluted, ice-cold hydrogen chloride solution which is
then extracted thrice with ethyl acetate. The combined organic lay-
ers are washed with brine, dried over magnesium sulfate and
filtered. After removal of the solvent the residue is chromato-
graphed on silica gel (petroleum ether/ethyl acetate 2/1,
Rf = 0.41) yielding 810 mg (3.55 mmol, 65%) of the product as a
clear oil, which solidifies after 1 day at rt, mp 44 °C. 1H NMR
(CDCl3) d = 9.85 (s, 1H), 7.16 (d, 1H, J = 1.6 Hz), 7.13 (d, 1H,
J = 1.6 Hz), 4.77 (dm, 2H, J = 47 Hz), 4.33 (dm, 2H, J = 28 Hz),
3.96 (s, 3H), 3.93 (s, 3H). 13C NMR (CDCl3) d = 190.6, 154.0,
152.5, 142.0, 131.7, 109.1, 107.3, 81.8 (d, J = 171 Hz), 68.7 (d,
J = 21 Hz), 61.1, 56.4. 19F NMR (CDCl3) d = ꢂ224.0. MS (EI): m/
z = 228 M+. Anal. Calcd for C11H13FO4: C, 57.89; H, 5.74. Found:
C, 57.74; H, 5.56.
4.2.3. 2-(1,3-Benzodioxol-5-yl)-1-[4-(2-bromoethoxy)phenyl]-
4-oxobutyric acid methyl ester 3
4.2.5. 3-(2-Bromoethoxy)-4,5-dimethoxybenzaldehyde 6
A mixture of 1.04 g (5.70 mmol) 3,4-dimethoxy-5-hydroxy-
benzaldehyde, 9.30 g (28.5 mmol) cesium carbonate and 10.7 g
(4.9 ml, 57.0 mmol) 1,2-dibromoethane in 50 ml of DMF is stirred
at rt for 3 h. The mixture is poured into 150 ml of aqueous satu-
rated ammonium chloride solution which is then extracted thrice
with dichloromethane. The combined organic layers are washed
with brine, dried over magnesium sulfate and filtered. After re-
moval of the solvent the residue is chromatographed on silica gel
(cyclohexane/ethyl acetate 2/1, Rf = 0.34) yielding 1.14 g
(3.94 mmol, 69%) of the product as a clear yellowish oil. 1H NMR
(CDCl3) d = 9.84 (s, 1H), 7.13 (d, 1H, J = 1.6 Hz), 7.08 (d, 1H,
J = 1.6 Hz), 4.37 (t, 2H, J = 6.1 Hz), 3.95 (s, 3H), 3.90 (s, 3H), 3.67
(t, 2H, J = 6.1 Hz). 13C NMR (CDCl3) d = 190.6, 153.9, 152.0, 144.3,
A solution of 7.79 g (23.7 mmol) 2-(1,3-benzodioxol-5-yl)-1-(4-
hydroxyphenyl)-4-oxobutyric acid methyl ester (1c), 30.5 g
(96.6 mmol) cesium carbonate and 45.1 g (240 mmol, 20.1 ml)
1,2-dibromoethane in 150 ml of DMF are stirred at rt for 20 h.
The mixture is poured into 250 ml of water which is then extracted
thrice with dichloromethane. The combined organic layers are
washed with brine, dried over magnesium sulfate and filtered.
After removal of the solvent the residue is chromatographed on sil-
ica gel (cyclohexane/ethyl acetate 7/3, Rf = 0.48) yielding 6.60 g
(15.2 mmol, 64%) of the product as colorless crystals, mp 112 °C.
1H NMR (CDCl3) d = 7.94 (d, 2H, J = 9.0 Hz), 6.93 (d, 2H,
J = 9.0 Hz), 6.85–6.74 (m, 3H), 5.93 (s, 2H), 4.34 (t, 2H, J = 6.1 Hz),

C. Höltke et al. / Bioorg. Med. Chem. 17 (2009) 7197–7208
7205
131.6, 108.9, 107.2, 69.2, 61.1, 56.2, 28.9. MS (EI): m/z = 290 M+
are added. The mixture is refluxed for 12 h under argon. Acetic acid
(ꢀ0.8 ml) is added and the mixture is further refluxed for 5 h. The
reaction mixture is poured onto ice and extracted with ethyl ace-
tate. The combined organic phases are washed with saturated
bicarbonate solution and brine and dried over magnesium sulfate.
After filtration and removal of the solvent the residue is chromato-
graphed on silica gel (petroleum ether/ethyl acetate 1/1, Rf = 0.26)
to yield 250 mg (0.46 mmol, 29%) of the product as a light yellow
powder, mp 196 °C (decomp.). 1H NMR (CDCl3) d = 7.39 (d, 2H,
J = 8.9 Hz), 6.94 (dd, 1H, J = 8.1 Hz, J = 1.7 Hz), 6.91 (d, 1H,
J = 1.7 Hz), 6.85 (d, 2H, J = 8.9 Hz), 6.79 (d, 1H, J = 8.1 Hz), 6.00 (s,
2H), 5.95 (s, 2H), 4.88 (s, br, 1H), 4.73 (dm, 2H, J = 47 Hz), 4.17
(dm, 2H, J = 28 Hz), 3.73 (s, 3H), 3.68 (s, 2H), 3.63 (s, 3H). 13C
NMR (CDCl3) d = 171.2, 160.3, 159.2, 152.9, 148.2, 147.7, 136.7,
131.8, 129.6, 128.2, 127.5, 123.1, 123.0, 114.5, 109.4, 108.4,
106.0, 105.6, 101.3, 81.7 (d, J = 171 Hz), 67.3 (d, J = 20 Hz), 60.7,
55.9, 32.1. 19F NMR (CDCl3) d = ꢂ224.0. MS (EI): m/z = 561
[M+Na]+, 521 [MꢂH2O]+. Anal. Calcd for C29H27FO9: C, 64.68; H,
5.05. Found: C, 64.44; H, 4.88.
(
81Br), 288 M+
(
79Br), 181 [MꢂCH2CH2Br]+, 109 [CH2CH281Br]+,
107 [CH2CH279Br]+. Anal. Calcd for C11H13BrO4: C, 45.70; H, 4.53.
Found: C, 45.57; H, 4.70.
4.2.6. 3-Benzo[1,3]dioxol-5-yl-4-[3-(2-fluoroethoxy)-4,5-
dimethoxybenzyl]-5-hydroxy-5-(4-methoxyphenyl)-5H-furan-
2-one 7
To a suspension of 1.21 g (3.54 mmol) of 2-(1,3-benzodioxol-5-
yl)-1-(4-methoxyphenyl)-4-oxobutyric acid methyl ester 1a and
195 mg (3.60 mmol) sodium methylate in 40 ml methanol
810 mg (3.55 mmol) of 3-(2-fluoroethoxy)-4,5-dimethoxybenzal-
dehyde 5 are added. The mixture is refluxed for 12 h under argon.
Acetic acid (ꢀ1.6 ml) is added and the mixture is further refluxed
for additional 5 h. The reaction mixture is poured onto ice and ex-
tracted with ethyl acetate. The combined organic phases are
washed with saturated bicarbonate solution and brine and dried
over magnesium sulfate. After filtration and removal of the solvent
the residue is chromatographed on silica gel (petroleum ether/
ethyl acetate 1/1) to yield 820 mg (1.52 mmol, 43%) of the product
as a light yellow powder, mp 66 °C (decomp.). 1H NMR (CDCl3)
d = 7.37 (d, 2H, J = 8.9 Hz), 6.91 (dd, 1H, J = 8.1 Hz, J = 1.7 Hz), 6.88
(d, 1H, J = 1.7 Hz), 6.82 (d, 2H, J = 8.9 Hz), 6.78 (d, 1H, J = 8.1 Hz),
6.02 (d, 1H, J = 1.8 Hz), 6.00 (d, 1H, J = 1.8 Hz), 5.95 (s, 2H), 4.66
(dm, 2H, J = 47 Hz), 4.60 (s, br, 1H), 3.99 (dm, 2H, J = 28 Hz), 3.78
(s, 3H), 3.74 (s, 3H), 3.65 (d, 2H, J = 2.8 Hz), 3.62 (s, 3H). 13C NMR
(CDCl3) d = 171.1, 160.5, 160.4, 153.2, 151.8, 148.2, 147.8, 137.5,
131.9, 128.8, 128.3, 127.5, 123.2, 123.0, 109.5, 108.5, 108.3,
106.9, 105.7, 101.4, 81.9 (d, J = 171 Hz), 68.6 (d, J = 21 Hz), 60.8,
56.0, 55.4, 32.0. 19F NMR (CDCl3) d = ꢂ224.0. MS (EI): m/z = 561
[M+Na]+, 521 [MꢂH2O]+. Anal. Calcd for C29H27FO9: C, 64.68; H,
5.05. Found: C, 64.48; H, 4.93.
4.2.9. 3-Benzo[1,3]dioxol-5-yl-5-hydroxy-5-[4-(2-bromoethoxy)-
phenyl]-4-(3,4,5-trimethoxybenzyl)-5H-furan-2-one 10
To a suspension of 3.00 g (6.89 mmol) of 2-(1,3-benzodioxol-5-
yl)-1-[4-(2-bromoethoxy)phenyl]-4-oxobutyric acid methyl ester
3 and 432 mg (8.00 mmol) sodium methylate in 50 ml methanol
1.37 g (7.00 mmol) of 3,4,5-trimethoxybenzaldehyde (4, FLUKA)
are added. The mixture is refluxed for 24 h under argon. Acetic acid
(ꢀ3.0 ml) is added and the mixture is refluxed for further 6 h. The
reaction mixture is poured onto ice and extracted with ethyl ace-
tate. The combined organic phases are washed with saturated
bicarbonate solution and brine and dried over magnesium sulfate.
After filtration and removal of the solvent the crude residue is
chromatographed on silica gel (cyclohexane/ethyl acetate 2/1,
Rf = 0.26) to yield 1.70 g (2.80 mmol, 41%) of the product as a white
powder, mp 178 °C. 1H NMR (CDCl3 + 10% DMSO-d6) d = 7.37 (d,
2H, J = 9.0 Hz), 6.94–6.88 (m, 2H), 6.85 (d, 2H, J = 9.0 Hz), 6.81 (d,
1H, J = 7.9 Hz), 5.98 (s, 2H), 5.92 (s, 2H), 4.29 (t, 2H, J = 6.3 Hz),
3.76 (d, 1H, J = 15.9 Hz), 3.68 (t, 2H, J = 6.3 Hz), 3.64 (s, 3H), 3.59
(s, 6H), 3.51 (d, 1H, J = 15.9 Hz). The signal of the proton of the hy-
droxy group is not detected. 13C NMR (CDCl3 + 10% DMSO-d6)
d = 170.0, 160.3, 157.2, 151.1, 146.5, 146.2, 134.6, 130.7, 128.8,
126.5, 126.1, 122.2, 121.8, 113.0, 108.1, 106.9, 104.7, 104.6,
100.0, 66.6, 59.2, 54.4, 30.8, 28.6. MS (EI): m/z = 623 [M+Na]+
4.2.7. 3-Benzo[1,3]dioxol-5-yl-4-[3-(2-bromoethoxy)-4,5-
dimethoxybenzyl]-5-hydroxy-5-(4-methoxyphenyl)-5H-furan-
2-one 8
A suspension of 1.50 g (5.18 mmol) of 3-(2-bromoethoxy)-4,5-
dimethoxybenzaldehyde 6, 286 mg (5.30 mmol) sodium methylate
and 1.78 g (5.20 mmol) of 2-(1,3-benzodioxol-5-yl)-1-(4-methoxy-
phenyl)-4-oxobutyric acid methyl ester 1a in 40 ml of methanol is
refluxed for 40 h under argon. Acetic acid (ꢀ3.0 ml) is added and
the mixture is refluxed for further 8 h. The reaction mixture is
poured onto ice and extracted with ethyl acetate. The combined or-
ganic phases are washed with saturated bicarbonate solution and
brine and dried over magnesium sulfate. After filtration and re-
moval of the solvent the crude residue is chromatographed on sil-
ica gel (cyclohexane/ethyl acetate 3/1, Rf = 0.29) to yield 1.60 g
(2.67 mmol, 52%) of the product as a white powder, mp 64 °C. 1H
NMR (CDCl3) d = 7.35 (d, 2H, J = 9.0 Hz), 6.92–6.86 (m, 2H), 6.82
(d, 2H, J = 9.0 Hz), 6.77 (d, 1H, J = 7.6 Hz), 5.98 (s, 2H), 6.01 (m,
1H), 5.97 (m, 1H), 5.94 (s, 2H), 4.30 (s, br, 1H), 4.04 (m, 2H), 3.77
(s, 3H), 3.74 (s, 3H), 3.63 (d, 2H, J = 11.4 Hz), 3.60 (s, 3H), 3.54 (t,
2H, J = 6.0 Hz). 13C NMR (CDCl3) d = 170.9, 160.4, 160.1, 153.1,
151.3, 148.1, 147.7, 137.3, 131.8, 128.6, 128.2, 127.3, 123.1,
122.8, 113.9, 109.3, 108.4, 108.1, 106.8, 105.5, 101.3, 69.0, 60.9,
(
79Br), 621 [M+Na]+ 81Br). Anal. Calcd for C29H27BrO9: C, 58.11;
(
H, 4.54. Found: C, 58.28; H, 4.61.
4.2.10. 2-(5-{[4-(Benzo[1,3]dioxol-5-yl)-2-hydroxy-2-(4-methoxy-
phenyl)-5-oxo-2,5-dihydrofuran-3-yl]methyl}-2,3-dimethoxy-
phenoxy)ethyl 4-methylbenzenesulfonate 11
A mixture of 1.45 g (2.42 mmol) of the bromide 3-benzo[1,3]-
dioxol-5-yl-4-[3-(2-bromoethoxy)-4,5-dimethoxybenzyl]-5-hydroxy-
5-(4-methoxyphenyl)-5H-furan-2-one 8 and 6.75 g (24.0 mmol)
silver-4-methylbenzene sulfonate in 100 ml of acetonitrile is heated
to reflux for 17 h. After filtration to remove the silver bromide and
evaporation of the solvent acetone is added. The precipitate (excess
silver-4-methylbenzenesulfonate) is removed by filtration and the
solvent is evaporated. Purification of the crude product by silica
gel column chromatography (cyclohexane/ethyl acetate 3/2,
Rf = 0.21) yields the product as a white foam (1.30 g, 1.88 mmol,
78%), mp 64 °C (decomp.). 1H NMR (CDCl3) d = 7.76 (d, 2H,
J = 8.2 Hz), 7.34 (d, 2H, J = 8.7 Hz), 7.30 (d, 2H, J = 8.2 Hz), 6.88–
6.83 (m, 2H), 6.80 (d, 2H, J = 8.7 Hz), 6.74 (d, 1H, J = 7.9 Hz), 5.98
(m, 1H), 5.95 (m, 1H), 5.92 (s, 2H), 4.68 (s, br, 1H), 4.23 (t, 2H,
J = 4.7 Hz), 3.94 (m, 2H), 3.75 (s, 3H), 3.63 (s, 3H), 3.60–3.56 (m,
2H), 3.58 (s, 3H), 2.40 (s, 3H). 13C NMR (CDCl3) d = 171.2, 160.4,
153.1, 151.4, 148.2, 147.7, 145.1, 137.3, 132.8, 131.9, 130.0, 128.7,
55.9, 55.3, 31.8, 29.1. MS (EI): m/z = 623 [M+Na]+ 79Br), 621
(
[M+Na]+
(
81Br), 583 [MꢂOH]+
(
81Br), 581 [MꢂOH]+
(
79Br).
Anal. Calcd for C29H27BrO9: C, 58.11; H, 4.54. Found: C, 58.25; H,
4.64.
4.2.8. 3-Benzo[1,3]dioxol-5-yl-5-[4-(2-fluoroethoxy)phenyl]-5-
hydroxy-4-(3,4,5-trimethoxybenzyl)-5H-furan-2-one 9
To a suspension of 585 mg (1.61 mmol) of 2-(1,3-benzodioxol-
5-yl)-1-[4-(2-fluoroethoxy)phenyl]-4-oxobutyric acid methyl ester
2 and 91 mg (1.68 mmol) sodium methylate in 20 ml methanol
320 mg (1.63 mmol) of 3,4,5-trimethoxybenzaldehyde (4, FLUKA)

7206
C. Höltke et al. / Bioorg. Med. Chem. 17 (2009) 7197–7208
128.3, 127.9, 127.5, 123.2, 122.9, 114.0, 109.5, 108.4, 108.3, 107.0,
105.8, 101.4, 68.3, 66.8, 60.8, 56.0, 55.4, 32.0, 21.6. The signal of
one quaternary carbon atom is not detected. MS (EI): m/z = 713
[M+Na]+, 673 [MꢂOH]+. Anal. Calcd for C36H34O12S: C, 62.60; H,
4.96. Found: C, 62.28; H, 4.84.
(v/v); B, acetonitrile/TFA 1000/1 (v/v); elution gradient: 70% A for
4 min, A from 70% to 5% in 38 min, 5% A for 2 min, from 5% to
70% A in 1 min) resulting in [18F]7 with a radiochemical yield of
21.4 6.1% (mean SD, decay-corrected, n = 12) and a radiochem-
ical purity >98% (retention time tR = 27.0 min). The time of synthe-
sis and purification was ꢀ70 min from the end of radionuclide
production. The determined specific radioactivity was 35–
4.2.11. 2-{4-[4-(Benzo[1,3]dioxol-5-yl)-2-hydroxy-5-oxo-3-(3,4,5-
trimethoxybenzyl)-2,5-dihydro-furan-2-yl]phenoxy}ethyl 4-
methylbenzenesulfonate 12
50 GBq/
lmol at the end of synthesis. The specific activity of
[
18F]7 was estimated by comparing the peak area of the UV-trace
A mixture of 800 mg (1.33 mmol) of the bromide 3-benzo[1,3]-
dioxol-5-yl-5-hydroxy-5-[4-(2-bromoethoxy)phenyl]-4-(3,4,5-tri-
methoxybenzyl)-5H-furan-2-one 10 and 3.9 g (13.0 mmol)
silver-4-methylbenzene sulfonate in 80 ml of acetonitrile is heated
to reflux for 17 h. After filtration to remove the silver bromide and
evaporation of the solvent acetone is added. The precipitate (excess
silver-4-methylbenzenesulfonate) is removed by filtration and the
solvent is evaporated. Purification of the crude product by silica gel
column chromatography (cyclohexane/ethyl acetate 3/2, Rf = 0.18)
yields the product as a white foam (750 mg, 1.09 mmol, 82%), mp
68 °C (decomp.). 1H NMR (CDCl3) d = 7.78 (d, 2H, J = 8.0 Hz), 7.32
(dd, 1H, J = 8.8 Hz, J = 2.5 Hz), 6.92–6.86 (m, 2H), 6.75 (d, 2H,
J = 8.0 Hz), 6.79 (d, 2H, J = 8.8 Hz), 5.94 (s, 2H), 5.92 (s, 2H), 4.31
(dd, 2H, J = 3.6 Hz, J = 5.4 Hz), 4.08 (dd, 2H, J = 3.6 Hz, J = 5.4 Hz),
3.67 (s, 3H), 3.62 (s, 2H), 3.57 (s, 6H), 2.42 (s, 3H). The proton signal
of the hydroxy group is not detected. 13C NMR (CDCl3) d = 171.2,
160.6, 158.7, 152.7, 148.0, 147.6, 145.1, 136.3, 132.6, 131.8,
129.9, 129.6, 128.0, 127.9, 127.5, 123.1, 122.9, 114.4, 109.4,
108.4, 105.7, 101.3, 67.9, 65.5, 60.7, 55.8, 32.0, 21.6. The signal of
one quaternary carbon atom is not detected. MS (EI): m/z = 713
[M+Na]+. Anal. Calcd for C36H34O12S: C, 62.60; H, 4.96. Found: C,
62.24; H, 5.01.
of purified [18F]7 with a standard curve of known concentrations
of reference compound 7 realized with the above mentioned RP-
HPLC system and the described conditions. The chemical identity
of [18F]7 was proved by co-injection and co-elution of [18F]7 and
non-radioactive counterpart 7 on the mentioned HPLC system.
For animal studies, the product fraction of the HPLC purification
procedure was collected, evaporated to dryness and reconstituted
in 10% ethanol/water for injection (v/v). Derivative [18F]9 was pre-
pared and purified by the same reaction conditions as stated above
for [18F]7. The radiochemical yield was ꢀ12% (decay-corrected,
n = 3) and the radiochemical purity was >98% (retention time
tR = 29.2 min). The time of synthesis and purification was
ꢀ70 min from the end of radionuclide production. The specific
radioactivity was not determined. The chemical identity of [18F]9
was proved by co-injection and co-elution of [18F]9 and non-radio-
active counterpart 9 on the mentioned HPLC system.
4.4. Biology
In vitro determination of receptor affinities: Microsomes were
prepared by homogenizing ventricles from DBA mice at 4 °C for
90 s in 1 ml of buffer A (10 mM EDTA, 10 mM HEPES, 0.1 mM ben-
zamidine, pH 7.4), using a Polytron PT 3000 (Kinematica, Lucerne,
Switzerland). Homogenates were centrifuged at 45,000g for 15 min
at 4 °C. The pellets were resuspended in 1 ml of buffer B (1 mM
EDTA, 10 mM HEPES, 0.1 mM benzamidine, pH 7.4) and recentri-
fuged at 45,000g for 15 min at 4 °C. The pellets were resuspended
in 1 ml of buffer B and centrifuged at 10,000g for 10 min at 4 °C.
The supernatants were recentrifuged at 45,000g for 15 min at
4 °C. The pellets, partially enriched membranes, were resuspended
in buffer C (50 mM Tris–HCl, 5 mM MgCl2, pH 7.4), and stored fro-
zen at ꢂ80 °C. For competition binding studies, the prepared mem-
branes were resuspended in buffer D (10 mM Tris–HCl, 154 mM
4.3. Radiochemistry
The production of no-carrier-added aqueous [18F]fluoride ion
was achieved on a CTI RDS111e 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
9.4 GBq of [18F]fluoride ion at the end of radionuclide production
for currents of 27 lA and irradiation times of 20 min. The radiosyn-
thesis of 3-benzo[1,3]dioxol-5-yl-4-[3-(2-[18F]fluoroethoxy)-4,5-
dimethoxybenzyl]-5-hydroxy-5-(4-methoxyphenyl)-5H-furan-2-
one
[
18F]7, and 3-benzo[1,3]dioxol-5-yl-5-[4-(2-[18F]fluoroeth-
NaCl, 0.1 mM ascorbic acid, pH 7.4) at 0 °C. An amount of 10 lg
oxy)phenyl]-5-hydroxy-4-(3,4,5-trimethoxy-benzyl)-5H-furan-2-
one [18F]9 was accomplished using a computer controlled TRACER-
lab FXFDG Synthesizer (GE Healthcare Germany, München,
Germany). The batch of aqueous [18F]fluoride ion was passed
of membranes were incubated with 40 pM of [125I]ET-1 (Perkin–
Elmer Live Sciences Inc., Billerica, MA, USA) and with varying
concentrations (10 pM–100 lM) of ligands 7–12 at 37 °C for 4 h.
Reactions were stopped by filtering onto Whatman GF/B filters
and washed with 0.9% NaCl at 4 °C. The membrane bound radioac-
through an anion exchange resin (Sep-PakÒ Light Waters Accell
TM
Plus QMA cartridge, preconditioned with 5 ml 1.0 M K2CO3 and
tivity was determined in a c-scintillation counter. Competition
10 ml water). [18F]Fluoride ion was eluted from the resin with a
binding curves were analyzed by nonlinear regression analysis
using the XMGRACE program (Linux software). The high- and
low-affinity IC50 values were converted into the high- and low-
affinity inhibition constants (Ki(ETA)and Ki(ETB)) by the method
of Cheng–Prusoff39 using the previously determined KD value of
mixture of 40
800
ll 1 M K2CO3, 200 ll water for injection, and
l
l DNA-grade acetonitrile containing 18 mg KryptofixÒ 2.2.2.
Subsequently, the aqueous [18F]K(Kryptofix 2.2.2)F solution was
carefully evaporated to dryness in vacuo. Compound [18F]7 was
prepared by treating 1.7 mg (2.98
l
mol) tosylate precursor 11 with
[
125I]ET-1 (208 2 pM).34
the dried [18F]K(Kryptofix 2.2.2)F residue in 1 ml DNA-grade aceto-
nitrile at 84 °C 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-PakÒ Light C18 cartridge (preconditioned
with 10 ml of ethanol and 10 ml of water for injection). The
cartridge was washed with additional 10 ml water for injection,
followed by elution of raw [18F]7 with 0.8 ml of acetonitrile
Biodistribution studies: Adult C57BL/6 mice (male and female, 25–
30 g) were anesthetized by isoflurane/N2O/O2 and one lateral tail
was canulated (27 G VenofixA needle with 20 cm polythene catheter
tubing, od 1 mm, id 0.38 mm). Animalswere allowedto recoverfrom
the anesthesia for ꢁ1 h and during the studies they were conscious
but under light restraint. The radiofluorinated ETA ligand [18F]7
(0.2–2.0 MBq at injection, 0.3–1.0 nmol/kg) was injected as a bolus
(DNA-grade). An amount of 200
l
l of the solution was fractionized
(2.0
PD 156707 at 1.0
n = 7) as an iv bolus (1.0
18F]7. Aliquots of each injectate were diluted in saline and
l
l/g, 100
l
l saline flush) via the tail vein. Some animals received
mol/kg (n = 7) or vehicle (water for injection,
l/g, 100 l saline flush) 10 min before
using a gradient radio-HPLC procedure (conditions: k = 254 nm;
flow = 1.5 ml/min; column: Nucleosil 100-5 C18 (250 ꢃ 4.6 mm2);
eluents: A, water for injection/trifluoroacetic acid (TFA) 1000/1
l
l
l
[

C. Höltke et al. / Bioorg. Med. Chem. 17 (2009) 7197–7208
7207
measured to determine the radioactivity injected into each animal.
At selected times after injection of the radioligand, animals were
killed by intravenous injection of sodium pentobarbitone (Euthatal)
at 200 mg/kg and tissues were rapidly removed. Tissue samples
were blotted and transferred to weighed vials for reweighing and
measurement of radioactivity using an automated gamma counter
(Wallac Wizard 300, Perkin–Elmer Life Sciences, Boston, USA). Radio-
activity was expressed as an uptake index, defined as
animal PET scanner.40 C57BL/6 mice (female, 22ꢂ23 g) were anes-
thetized by inhalation (isoflurane 2%, oxygen 0.5 l minꢂ1) for inser-
tion of catheters into tail veins and subsequent PET scanning. Two
mice were positioned in parallel in the scanner and 30 s after start
of the acquisition the radioligand [18F]7 (6.6 MBq in 100
ll, 100 ll
saline flush) was injected into each mouse simultaneously. List
mode data were acquired for 2 h and reconstructed into 10 min
time frames using an iterative reconstruction algorithm.41 PET
images were analyzed using in-house software programs in MAT-
LAB (The MathWorks Company, Natick, MA) and C programming
languages.42 For in vivo biodistribution measurements ROIs were
drawn over the respective organs in five different coronal planes
and analyzed as above.
Tissue radioactivity ðcpmÞ=Tissue wet weight ðgÞ
Uptake index ¼
Radioactivity injected ðcpmÞ=Body weight ðgÞ
Metabolism: Rat plasma. One male Wistar rat (Charles River Lab-
oratories, Sulzfeld, Germany, 350 g) was anesthetized by isoflurane/
N2O/O2 and blood (10 ml) was removed by cardiac puncture. Plasma
was separated by centrifugation (10,000g for 4 min) and stored at
ꢂ20 °C overnight. Samples (0.5 ml) were incubated with 19 MBq of
the radioligand [18F]7 at 37 °C using an Eppendorf Thermomixer
comfort (Eppendorf AG, Hamburg, Germany). At different time points
(1, 2.5, 20, and 60 min) proteins were precipitated by addition of ice-
Acknowledgements
This study was in part supported by the Deutsche Forschungs-
gemeinschaft (DFG), project SCHA 758-5-1, DFG Sonderfor-
schungsbereich SFB 656, Münster, Germany (subprojects A1 and
A4), the Interdisziplinäres Zentrum für Klinische Forschung (IZKF),
Münster, Germany (core unit SmAP) and the Bundesministerium
für Bildung und Forschung (BMBF), Germany (MoBiMed subproject
01EZ0807). The authors wish to thank Sandra Schröer, Anne
Kanzog, and Christine Bätza for technical support.
cold acetonitrile (700 ll) and centrifugation at 10,000g for 4 min.
The supernatant was separated and the radioactivity of the protein
pellet and the supernatant was measured (Ionization chamber
637620, MED Nuklearmedizin-Technik GmbH, Dresden, Germany)
to estimate protein binding. A sample of each supernatant (200
was analyzed by HPLC (conditions see above).
ll)
References and notes
Mouse: Adult C57BL/6 mice (male and female, 25–30 g) were
prepared in accordance with the chapter biodistribution (see
1. Hickey, K. A.; Rubanyi, G.; Paul, R. J.; Highsmith, R. F. Am. J. Physiol. 1985, 248,
C550.
above). [18F]7 was injected as a bolus (100
ll) via the tail vein. Ali-
quots of each injectate were diluted in saline and measured to
determine the radioactivity injected into each animal. Animals
were sacrificed at 5 or 20 min after injection of the radiotracer
by intravenous injection of sodium pentobarbitone (Euthatal) at
200 mg/kg. Blood was taken immediately by cardiac puncture
and centrifuged at 10,000g for 4 min (Eppendorf centrifuge
5415D, Eppendorf AG, Hamburg, Germany) to separate plasma.
Plasma from two mice was pooled for analysis by HPLC. Ice-cold
acetonitrile (0.7 ml) was added to the pooled plasma (0.7–1.0 ml)
and the precipitated proteins were removed by centrifugation
2. Yanagisawa, M.; Kurihara, H.; Kimura, S.; Tomobe, Y.; Kobayashi, M.; Mitsui, Y.;
Yazaki, Y.; Goto, K.; Masaki, T. Nature 1988, 332, 411.
3. Bousette, N.; Giaid, A. Can. J. Physiol. Pharmacol. 2003, 81, 578.
4. Giaid, A. Chest 1998, 114, 208S.
5. Giaid, A.; Yanagisawa, M.; Langleben, D.; Michel, R. P.; Levy, R.; Shennib, H.;
Kimura, S.; Masaki, T.; Duguid, W. P.; Stewart, D. J. N. Eng. J. Med. 1993, 328,
1732.
6. Saito, Y.; Nakao, K.; Mukoyama, M.; Imura, H. N. Eng. J. Med. 1990, 322, 205.
7. Teerlink, J. R.; Loffler, B. M.; Hess, P.; Maire, J. P.; Clozel, M.; Clozel, J. P.
Circulation 1994, 90, 2510.
8. Yoshibayashi, M.; Nishioka, K.; Nakao, K.; Saito, Y.; Matsumura, M.; Ueda, T.;
Temma, S.; Shirakami, G.; Imura, H.; Mikawa, H. Circulation 1991, 84, 2280.
9. Bagnato, A.; Natali, P. G. Cancer Treat. Res. 2004, 119, 293.
10. Bagnato, A.; Natali, P. G. J. Transl. Med. 2004, 2, 16.
11. Nelson, J.; Bagnato, A.; Battistini, B.; Nisen, P. Nat. Rev. Cancer 2003, 3, 110.
12. Rosano, L.; Spinella, F.; Di Castro, V.; Nicotra, M. R.; Albini, A.; Natali, P. G.;
Bagnato, A. Am. J. Pathol. 2003, 163, 753.
13. Smollich, M.; Wülfing, P. Curr. Vasc. Pharmacol. 2007, 5, 239.
14. Dupuis, J.; Goresky, C. A.; Fournier, A. J. Appl. Physiol. 1996, 81, 1510.
15. Dupuis, J.; Stewart, D. J.; Cernacek, P.; Gosselin, G. Circulation 1996, 94, 1578.
16. Fukuroda, T.; Fujikawa, T.; Ozaki, S.; Ishikawa, K.; Yano, M.; Nishikibe, M.
Biochem. Biophys. Res. Commun. 1994, 199, 1461.
17. Langleben, D.; Dupuis, J.; Hirsch, A.; Giovinazzo, M.; Langleben, I.; Khoury, J.;
Ruel, N.; Caron, A. Chest 2005, 128, 622S.
18. Polikepahad, S.; Moore, R. M.; Venugopal, C. S. Res. Commun. Mol. Pathol.
Pharmacol. 2006, 119, 3.
19. Barton, M.; Haudenschild, C. C.; d’Uscio, L. V.; Shaw, S.; Munter, K.; Lüscher, T.
F. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 14367.
20. d’Uscio, L. V.; Barton, M.; Shaw, S.; Lüscher, T. F. Cardiovasc. Res. 2002, 53, 487.
21. Iwasa, S.; Fan, J.; Miyauchi, T.; Watanabe, T. Pathobiology 2001, 69, 1.
22. Maguire, J. J.; Wiley, K. E.; Kuc, R. E.; Stoneman, V. E.; Bennett, M. R.; Davenport,
A. P. Exp. Biol. Med. (Maywood) 2006, 231, 806.
23. Alanen, K.; Deng, D. X.; Chakrabarti, S. Histopathology 2000, 36, 161.
24. Asham, E.; Shankar, A.; Loizidou, M.; Fredericks, S.; Miller, K.; Boulos, P. B.;
Burnstock, G.; Taylor, I. Br. J. Cancer 2001, 85, 1759.
25. Bagnato, A.; Salani, D.; Di Castro, V.; Wu-Wong, J. R.; Tecce, R.; Nicotra, M. R.;
Venuti, A.; Natali, P. G. Cancer Res. 1999, 59, 720.
26. Nelson, J. B.; Chan-Tack, K.; Hedican, S. P.; Magnuson, S. R.; Opgenorth, T. J.;
Bova, G. S.; Simons, J. W. Cancer Res. 1996, 56, 663.
27. Wülfing, P.; Diallo, R.; Kersting, C.; Wülfing, C.; Poremba, C.; Rody, A.; Greb, R.
R.; Bocker, W.; Kiesel, L. Clin. Cancer Res. 2003, 9, 4125.
(10,000g for 4 min).
A
sample of the resulting supernatant
g) and analyzed
(200 l) was spiked with non-radioactive 7 (10
l
l
by HPLC (conditions see above). In addition, urine samples were ta-
ken at the given time points from one mouse each and analyzed by
HPLC (conditions see above).
Rat. One male Wistar rat (Charles River Laboratories, Sulzfeld,
Germany, 400 g) was used for the experiment. The animal was
anesthetized by isoflurane/N2O/O2 for insertion of catheters (see
above) into the ventral tail artery and one lateral tail vein and were
allowed to recover from the anesthesia for ꢀ2 h. An amount of
40 MBq of [18F]7 was injected as a bolus via the tail vein. Four
blood samples (600–800 ll) were collected from the tail artery
catheter at 2.5, 10, 20, and 60 min after injection. Cell free plasma
was obtained by centrifugation at 10,000g for 4 min. Ice-cold ace-
tonitrile (0.7 ml) was added to the plasma and the precipitated
proteins were removed by centrifugation (10,000g for 4 min). A
sample of the resulting supernatant (200
ll) was spiked with
non-radioactive 7 (10 g) and analyzed by HPLC (conditions see
l
above). At 60 min the animal was killed by intravenous injection
of sodium pentobarbitone (Euthatal) at 200 mg/kg. A urine sample
was taken by puncture of the bladder and was directly injected
into the HPLC for analysis.
28. Aleksic, S.; Szabo, Z.; Scheffel, U.; Ravert, H. T.; Mathews, W. B.; Kerenyi, L.;
Rauseo, P. A.; Gibson, R. E.; Burns, H. D.; Dannals, R. F. J. Nucl. Med. 2001, 42,
1274.
29. Johnstrøm, P.; Fryer, T. D.; Richards, H. K.; Barret, O.; Clark, J. C.; Ohlstein, E. H.;
Pickard, J. D.; Davenport, A. P. J. Cardiovasc. Pharmacol. 2004, 44, S34.
30. Mathews, W. B.; Zober, T. G.; Ravert, H. T.; Scheffel, U.; Hilton, J.; Sleep, D.;
Dannals, R. F.; Szabo, Z. Nucl. Med. Biol. 2006, 33, 15.
4.5. PET scans
Uptake of radioactivity after intravenous injection of ETA ligand
[
18F]7 was visualized using the high-resolution quadHIDAC small-

7208
C. Höltke et al. / Bioorg. Med. Chem. 17 (2009) 7197–7208
31. Mathews, W. B.; Murugesan, N.; Xia, J.; Scheffel, U.; Hilton, J.; Ravert, H. T.;
Dannals, R. F.; Szabo, Z. J. Nucl. Med. 2008, 49, 1529.
32. Doherty, A. M.; Patt, W. C.; Edmunds, J. J.; Berryman, K. A.; Reisdorph, B. R.;
Plummer, M. S.; Shahripour, A.; Lee, C.; Cheng, X. M.; Walker, D. M., et al J. Med.
Chem. 1995, 38, 1259.
33. Patt, W. C.; Edmunds, J. J.; Repine, J. T.; Berryman, K. A.; Reisdorph, B. R.; Lee, C.;
Plummer, M. S.; Shahripour, A.; Haleen, S. J.; Keiser, J. A.; Flynn, M. A.; Welch, K.
M.; Reynolds, E. E.; Rubin, R.; Tobias, B.; Hallak, H.; Doherty, A. M. J. Med. Chem.
1997, 40, 1063.
36. Zoghbi, S. S.; Shetty, H. U.; Ichise, M.; Fujita, M.; Imaizumi, M.; Liow, J. S.; Shah,
J.; Musachio, J. L.; Pike, V. W.; Innis, R. B. J. Nucl. Med. 2006, 47, 520.
37. Sykes, T. R.; Ruth, T. J.; Adam, M. J. Int. J. Rad. Appl. Instrum. B 1986, 13,
497.
38. Parenty, A. D.; Smith, L. V.; Pickering, A. L.; Long, D. L.; Cronin, L. J. Org. Chem.
2004, 69, 5934.
39. Cheng, Y.; Prusoff, W. H. Biochem. Pharmacol. 1973, 22, 3099.
40. Schäfers, K. P.; Reader, A. J.; Kriens, M.; Knoess, C.; Schober, O.; Schäfers, M. J.
Nucl. Med. 2005, 46, 996.
34. Höltke, C.; Law, M. P.; Wagner, S.; Breyholz, H. J.; Kopka, K.; Bremer, C.; Levkau,
B.; Schober, O.; Schäfers, M. Bioorg. Med. Chem. 2006, 14, 1910.
35. Höltke, C.; Waldeck, J.; Kopka, K.; Heindel, W.; Schober, O.; Schäfers, M.;
Bremer, C. Mol. Imaging 2009, 8, 27.
41. Reader, A. J.; Ally, S.; Bakatselos, F.; Manavaki, R.; Walledge, R. J.; Jeavons, A. P.;
Julyan, P. J.; Zhao, S.; Hastings, D. L.; Zweit, J. IEEE Trans. Nucl. Sci. 2002, 49, 693.
42. Biedenstein, S.; Schäfers, M.; Stegger, L.; Kuwert, T.; Schober, O. Eur. J. Nucl.
Med. 1999, 26, 201.