ABP688, a novel selective and high affinity ligand for the labeling of mGlu5 receptors: Identification, in vitro pharmacology, pharmacokinetic and biodistribution studies

Article abstract of DOI:10.1016/j.bmc.2006.10.038

[¹¹C]ABP688 (2) has recently been demonstrated to be a useful PET tracer for in vivo imaging of the metabotropic glutamate receptors type 5 (mGluR5) in rodents. We describe here the identification and preclinical profiling of ABP688 and its tri

Full text of DOI:10.1016/j.bmc.2006.10.038

Bioorganic & Medicinal Chemistry 15 (2007) 903–914
ABP688, a novel selective and high affinity ligand for the labeling
of mGlu5 receptors: Identification, in vitro pharmacology,
pharmacokinetic and biodistribution studies
Samuel Hintermann,a Ivo Vranesic,a Hans Allgeier,a Armin Brulisauer,b Daniel Hoyer,a
¨
Michel Lemaire,b Thomas Moenius,b Stephan Urwyler,a Steven Whitebread,c
Fabrizio Gasparinia,* and Yves P. Aubersona
aNovartis Institutes for BioMedical Research, 4002 Basel, Switzerland
bNovartis Pharma AG, 4002 Basel, Switzerland
cNovartis Institutes for BioMedical Research, 02139 Cambridge, USA
Received 6 July 2006; accepted 19 October 2006
Available online 21 October 2006
Abstract—[11C]ABP688 (2) has recently been demonstrated to be a useful PET tracer for in vivo imaging of the metabotropic glu-
tamate receptors type 5 (mGluR5) in rodents. We describe here the identification and preclinical profiling of ABP688 and its tritiated
version [3H]ABP688, and show that its high affinity (Kd = 2 nM), selectivity, and pharmacokinetic properties fulfill all requirements
for development as a PET tracer for clinical imaging of the mGlu5 receptor.
ꢀ 2006 Elsevier Ltd. All rights reserved.
1. Introduction
and mGlu5) and II (mGlu2 and mGlu3) has facilitated
significant advances in the understanding of the roles
of these receptors in brain physiology and
pathophysiology.2
Based on amino-acid sequence, pharmacology, and sec-
ond messenger coupling, metabotropic glutamate recep-
tors (mGluRs) have been clustered into three groups (I–
III).1 Group I receptors (mGluR1 and mGluR5) are
coupled to the phospholipase C intracellular pathway,
Group II (mGluR2 and mGluR3) and Group III
(mGluR4, 6, 7, and 8) are negatively linked to adenylyl
cyclase.
The identification of MPEP (1, 2-methyl-6-(phenylethy-
nyl)-pyridine, Fig. 1), a selective, non-competitive, and
brain-penetrable mGlu5 receptor antagonist,3 allowed
the exploration of the therapeutic potential of mGlu5
receptor antagonism. Results from a variety of behav-
ioral studies revealed that—with the exception of the
benzodiazepines—mGlu5 receptor antagonists exhibit
the widest and most robust anxiolytic activity in pre-
clinical models seen to date.4–6 Upcoming clinical stud-
ies, involving the use of specific mGlu5 receptor antago-
nists, should soon indicate whether the preclinical
In contrast to the glutamate-gated ion channels
(NMDA, AMPA, and kainate receptors), which medi-
ate the glutamatergic neurotransmission in the CNS,
mGlu receptors belong to the ‘family 3’ G protein-cou-
pled receptors and have been shown to play a modulato-
ry role in the glutamatergic synaptic transmission, either
by regulating ion channel activity or by neurotransmit-
ter release. The recent discovery of small molecules that
selectively interact with receptors of Group I (mGlu1
N
N
N
O
Keywords: Metabotropic glutamate receptor subtype 5; [3H]ABP688;
ABP688; PET imaging.
1, MPEP
2, ABP688
*
Corresponding author. Tel.: +41 61 324 72 57; fax: +41 61 324 83
81; e-mail: fabrizio.gasparini@novartis.com
Figure 1. Structures of MPEP and ABP688.
0968-0896/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2006.10.038

904
S. Hintermann et al. / Bioorg. Med. Chem. 15 (2007) 903–914
anxiolytic-like effects translate into anxiolytic activity in
man.
mGlu5 receptor. Here, we report on the preclinical char-
acterization of ABP688 and its tritiated species
[3H]ABP688. This characterization allowed the selection
of this ligand for development as a PET tracer.16
As comprehensive as it might be, the preclinical profile
of a novel mGlu5 antagonist, as for other CNS drugs,
does not allow an accurate prediction of the adequate
drug dosing in humans. Positron emission tomography
(PET) with receptor specific ligands is an imaging meth-
od which has been proven useful for studying these
parameters in humans.7,8
2. Chemistry
2.1. Synthesis of ABP688 (2) and its labeling precursor 7
The synthesis of ABP688 (2) and of the precursor of
[3H]ABP688, oxime 7, started from commercially avail-
able 2-bromo-6-methyl-pyridine. A Sonogashira17 cross
coupling with 2-methyl-3-butin-2-ol yielded the tertiary
alcohol (3), which was easily and in good yields convert-
ed into the alkynyl-pyridine (4), by base-catalyzed ace-
tone elimination (Scheme 1). A second Sonogashira18
coupling with 3-bromo-cyclohex-2-enone (5)19 led
directly to ketone 6, which was then readily transformed
into the labeling precursor 7 (E/Z-mixture 1.6:1) by
base-catalyzed oxime formation with hydroxylamine
hydrochloride. Similarly, oxime formation with O-meth-
yl-hydroxylamine led to ABP688 (2) in excellent yield, as
a mixture of E/Z-isomers (ca. 1.5:1). The stereochemical
assignment of the ABP688 isomers was based upon a
ROESY20,21 peak between the olefinic H(2)-proton
and the methyl-group in the Z-isomer, and upon the
expected high field shift (123.1 ppm) in the 13C-reso-
nance of C(2) for the Z-isomer when compared to the
E-isomer (130.8 ppm). The assignment of the two iso-
mers of 7 was made by analogy with ABP688.
To date, a number of mGlu5 receptor antagonists have
been successfully used to label mGlu5 receptors in vi-
tro.9,10 However, the development of these ligands into
useful PET tracers has so far not met with success11,12
and only very recently a series of 11C and 18F-derivatives
of MTEP and MTEB (2-methyl-4-phenylethynyl-thia-
zole), allowing PET imaging in rhesus monkey, has been
described.13
In parallel to our discovery program aiming at the iden-
tification of mGlu5 receptor antagonists, we intended to
identify a suitable PET ligand to support the clinical
development of a pharmaceutical agent. Our target cri-
teria for a potential PET ligand were: (1) a logP below
4; (2) an easy radiolabeling in the last synthesis step; (3)
a high affinity (<10 nM) for mGluR5 and selectivity
(>100-fold) over other receptors; (4) a rapid brain up-
take and elimination; and (5) the absence of brain-pene-
trating radiolabeled metabolites. In order to allow for
displacement studies and for determination of receptor
occupancy by therapeutic drug candidates, any potential
tracer would have to be competitively and fully dis-
placed by drug candidates binding to the allosteric bind-
ing pocket at the mGlu5 receptor.
In summary, we could efficiently synthesize both
ABP688 (2) and the labeling precursor [3H]ABP688,
oxime 7, in four steps and excellent overall yields of
53%, starting from commercially available 2-bromo-6-
methyl-pyridine.
Initially, a series of PET candidates was synthesized
based on the prototypic antagonist MPEP. A 11C or
18F label was introduced in positions C-3 of the phenyl
ring or C-6 of the pyridine ring, which were shown to
tolerate modifications without compromising the affinity
for the allosteric binding site.14,15 Four of these initial
derivatives were tested in vivo in rat to determine their
brain uptake and distribution but failed to meet the tar-
get profile, by exhibiting a combination of a relatively
high lipophilicity, high non-specific binding, insufficient
affinity or limited metabolic stability.
2.2. Physicochemical properties and stability of ABP688
ABP688 is a low molecular weight molecule (mw = 240.3)
with measured log P and logD (pH 6.8) of 3.4, both one
unit above the calculated value, but still within an accept-
able range (<4). Its pKa is 3.8, and its solubility in water
(120 mg/L at pH 6.8) is largely sufficient for an iv formu-
lation. There was no decomposition of ABP688 or of 7 ob-
servable by HPLC after one week at 80 ꢁC as a powder or
in buffer suspension at pH 7.4. Precursor 7 is stable under
conditions used for radiolabeling with 3H (Scheme 2) and
11C (DMF, 5 min at 90 ꢁC).16
We subsequently revisited the library of compounds
which had been synthesized during the medicinal chem-
istry program for mGluR5 antagonists and we focused
on a series of derivatives with a cyclohexenone oxime in-
stead of the aromatic ring contained in MPEP. The
introduction of an oxime significantly decreased lipo-
philicity and improved water solubility, as compared
to the precursor series. The most promising candidate
identified was ABP688 (2), a high affinity and selective
mGluR5 antagonist with a calculated logP of 2.4
(Fig. 1).
2.3. Radiosynthesis of [3H]ABP688
Tritiated ABP688 was synthesized by reacting the pre-
formed potassium salt of oxime 7 with [3H]-methyl io-
dide at 120 ꢁC for 2.5 h (Scheme 2). The product was
isolated after purification by semi-preparative HPLC
(Nucleosil 100-3 C-18 HD) as an E/Z-mixture (ca. 4:1)
and stored as a solution in acetonitrile at ꢀ20 ꢁC.
Since this compound fulfilled all of our criteria in terms
of physicochemical properties, we decided to further ex-
plore its potential as a PET imaging ligand for the
By careful optimization of the labeling procedure,16 con-
sistent isomeric ratios of >10:1 in favor of the more po-
tent E-isomer could be obtained (data not shown).

S. Hintermann et al. / Bioorg. Med. Chem. 15 (2007) 903–914
905
Pd(Ph3P)2Cl2, CuI
NEt3, 55 ˚C, 2h
NaOH, benzene
reflux, 1h
N
CH3
OH
N
N
Br
OH
H3C
4, 78% (2 steps)
3
Br
O
Pd(Ph3P)2Cl2, CuI
NEt3, DMF, 55 ˚C, 1h
N
5
O
6, 75%
NH2OH x HCl
pyridine
rt, 17h
NH2OMe x HCl
pyridine
rt, 64h
N
N
N
N
OH
O
7, 91%
2, 90%
Scheme 1. Synthesis of ABP688 (2, 3-(6-methyl-pyridin-2-ylethynyl)-cyclohex-2-enone O-methyl-oxime) and the labeling precursor 3-(6-methyl-
pyridin-2-ylethynyl)-cyclohex-2-enone oxime (7).
1) [(H3C)3Si]2K, DMF, 20 ˚C, 30min
2) CT3I, toluene, 120˚C, 2.5h
3H
N
N
N
N
OH
O
3H
3H
7
[3H]-ABP688
Scheme 2. Radiosynthesis of [3H]ABP688: radiochemical purity: 99.4%, specific activity: 2.83 TBq/mmol (11.78 MBq/lg).
3. Pharmacological properties and evaluation as
potential PET tracer
IC50 value of 7.0 nM (95% CI: 6.1–8.1), or a correspond-
ing Ki of 3.5 (3.1, 4.0) nM. The Hill coefficient was
ꢀ0.92 0.10 ( 95% CI).
3.1. Potency and selectivity in functional assays
3.3. Receptor binding panel
ABP688 was originally shown to be a potent antagonist
of quisqualate-induced phosphoinositol (PI) accumula-
tion in L(tk-) cells expressing recombinant human
mGluR5 (hmGluR5), with an IC50 value of 2.4 nM
(95% CI: 0.5–12 nM) (Fig. 2A). In the same preparation,
ABP688 completely inhibited glutamate-induced calci-
um release with an IC50 value of 2.3 nM (95% CI: 2.1–
2.5) nM (data not shown) and had no effect, up to
10 lM, on ATP-induced PI accumulation via stimula-
tion of endogenously expressed purinergic receptors in
hmGluR5-expressing cells (Fig. 2A), or on quisqua-
late-induced PI accumulation in hmGluR1-expressing
cells. ABP688 had no effect on basal PI levels in hmG-
luR5- or hmGluR1-expressing cells (Fig. 2B).
The in vitro binding affinities of ABP688 for a variety of
CNS GPCR receptors, ion channels or transporters
were determined according to standard filtration meth-
ods.24 ABP688, at concentrations of 1 and 10 lM, did
not bind to any of the tested receptors or transporters,
confirming the high degree of selectivity of ABP688 (de-
tails are available as Supplementary information).
3.4. Binding characteristics of [3H]ABP688 in vitro
The characteristics of the binding of [3H]ABP688 to na-
tive mGlu5 receptor were determined in membranes pre-
pared from rat whole brain tissue. The observed
association and dissociation time courses were well de-
scribed by mono-exponential models, the respective rate
constants were kobs = 0.103 0.006 (n = 3) and
3.2. [3H]-M-MPEP displacement in membranes of hmG-
luR5-expressing cells
k
ꢀ1 = 0.075 0.006 minꢀ1(n = 5). The calculated associ-
The affinity of ABP688 for the allosteric binding site in
the transmembrane domain of mGluR522 was deter-
mined in a radioligand competition assay using [3H]-
M-MPEP.9 In membranes prepared from L(tk-) cells
expressing recombinant human mGluR5 (hmGluR5),23
ABP688 fully displaced the binding of [3H]-M-MPEP
in a concentration-dependent manner (Fig. 3) with an
ation rate constant k1 was 0.07 nMꢀ1 minꢀ1. From these
values, an equilibrium binding constant Kd of 1.07 nM
was calculated. The specific binding of [3H]ABP688
was saturable and adequately described by a single-
binding site model, with Bmax and Kd values of
0.87 0.096 pmol/mg
(n = 4) nM, respectively. (Fig. 4A–C).
protein
and
2.3 0.34

906
S. Hintermann et al. / Bioorg. Med. Chem. 15 (2007) 903–914
A
B
+ 0.3 quis
+ 10 ATP
ABP688
+ 40 quis
ABP688
100
100
50
0
50
0
-9
-8
-7
-6
-5
-4
ctr
-11 -10 -9
-8
-7
-6
-5
-4
ctr
log([ABP688]/M)
log([ABP688]/M)
Figure 2. Effects of ABP688 on PI accumulation in cells expressing recombinant hmGluR5 (A) and hmGluR1 (B). (A) ABP688 fully inhibits
quisqualate-stimulated (filled circles, 0.3 lM), but not ATP-stimulated (empty triangles, 10 lM) PI accumulation. Applied alone, ABP688 has no
effect on basal PI accumulation (empty circles). Symbols offset to the left indicate respective normalized positive and negative controls SD. (B)
Absence of effects on quisqualate-stimulated (filled circles, 40 lM) PI accumulation and basal PI levels (empty circles) in hmGluR1-expressing cells.
All data sets shown represent percent control SD of n = 3 independent determinations, each performed in triplicate.
and the distribution of the labeled sites. In rat brain slic-
es, incubation with 5 nM (70 Ci/mmol) of [3H]ABP688
gave rise to a regionally differentiated binding pattern
(Fig. 6A, left panels). The binding was displaceable by
100
MPEP (10 lM), and the remaining (non-specific) bind-
ing was found to be nearly indistinguishable from back-
ground (Fig. 6).
50
High-to-medium specific binding was observed (in
descending rank order) in nucleus accumbens, caudate
putamen, hippocampus, olfactory tubercle, amygdala,
and occipital and frontal cortices; low-to-near back-
ground binding was found in thalamus, hypothalamus,
midbrain, pons-medulla, and cerebellum (Fig. 7). This
pattern is consistent with autoradiographic results ob-
tained with [3H]methoxy-PyEP,10 as well as mGluR5
expression patterns obtained with immunocytochemis-
try using mGluR5-specific antibodies25,26 or mGluR5
mRNA distribution as determined with in situ hybrid-
ization.27 Taken together, our present findings of a good
(qualitative) correlation between [3H]ABP688 binding
and mGluR5 expression, in combination with low
unspecific binding in native brain tissue, encouraged fur-
ther profiling in vivo.
0
-11
-10
-9
-8
-7
-6
-5
log([ABP688]/M)
Figure 3. Concentration dependence of [3H]-M-MPEP displacement
by ABP688 in membranes prepared from hmGluR5-expressing cells.
Data show percentage SD of specific binding of [3H]-M-MPEP
(2 nM). Non-specific binding was determined in the presence of 1 lM
unlabeled M-MPEP. Data were obtained from n = 2 independent
determinations, each performed in triplicate. Solid
(
dashed)
horizontal lines indicate normalized total and non-specific controls
( SD). Solid curve was obtained by fitting the logistic equation to the
pooled data set, best-fit parameters and respective 95% confidence
intervals are given in text.
In
a further series of experiments, we utilized
[3H]ABP688 to determine the affinities of ABP688,
MPEP, M-MPEP as well as CPCCOEt, an mGluR1
antagonist, to membrane preparations of hmGluR5-ex-
pressing cells and rat whole brain tissue. In both prepa-
rations, ABP688, MPEP, and M-MPEP fully displaced
[3H]ABP688 in a concentration-dependent manner with
Hill coefficients near negative unity, while CPCCOEt
had no appreciable effect on [3H]ABP688 binding up
to 10 lM (Fig. 5 and Table 1, for respective affinities).
The solvent, DMSO, had no effect on binding (not
shown). The near-equality of the respective binding
affinities of APB688, MPEP, and MPEP to hmGluR5
and rat brain membranes indicated that there are no
apparent species differences between recombinant hu-
man and native rat receptors in vitro.
3.6. Blood distribution, plasma protein binding, and
in vitro metabolism
The assessments were performed using [3H]ABP688 with
the objective of allowing a pharmacokinetic scaling from
animal to man.
3.6.1. In vitro blood distribution. [3H]ABP688 was stable
when incubated in plasma at 37 ꢁC and equilibrated rap-
idly between plasma and blood cells at 37 ꢁC. The blood
distribution was independent of concentration but
slightly species-dependent, as illustrated by the fraction
present in plasma of 76% in rat and 92% in human.
The blood to plasma concentration ratios were 0.74
and 0.57 in rat and human, respectively. The character-
istics of the blood distribution of [3H]ABP688, that is,
large fraction of the compound present in plasma com-
partment, independent of concentration, support the use
of plasma concentrations to estimate the input function
for PET studies.
3.5. Autoradiography in rat brain slices
In vitro autoradiography is an established method to
determine the binding characteristics of a radioligand

S. Hintermann et al. / Bioorg. Med. Chem. 15 (2007) 903–914
907
stability was assessed by calculation of the intrinsic
clearance CLint: the intrinsic clearance was medium to
high in rat and human (roughly 150 lL/min/mg). HPLC
analysis with radioactivity detection indicated oxidative
hydroxylation to be the main metabolic pathway, lead-
ing to formation of tritiated water and four mono-hy-
droxylated metabolites. The exact positions of the
hydroxy group could not be determined from the avail-
able LC–MS/MS data. No human-specific metabolites
were detected when compared to rat.
A
B
C
2000
1500
1000
500
0
3.7. Disposition of [3H]ABP688 in rat following an
intravenous microdose injection
0
0
0
20
40
60
80
100
Time (min)
The main objective of this pharmacokinetic study was to
establish the disposition of ABP688 in plasma and brain
of rats.
2000
1500
1000
500
0
3.7.1. Pharmacokinetics. After a 4 lg/kg [3H]ABP688
intravenous application, plasma concentrations declined
rapidly (Fig. 8, left panel). Compartmental analysis
according to a bi-exponential model resulted in plasma
half-lives of approximately 5 min (t1/2,1) and 1 h
(t1/2,2), respectively, where the fraction of AUC associ-
ated with the second half-life amounted to 59%. Since
[3H]ABP688 was stable when incubated in blood or
plasma at 37 ꢁC and only trace amounts of parent com-
pound were detected in excreta, its elimination occurred
almost exclusively by metabolism. Indeed, already
5 min. after administration, the fraction unchanged in
plasma had decreased to 34% of plasma radioactivity,
whereas the AUC of parent drug represented merely
1.4% of the AUC for total radioactivity. The blood
clearance value of ABP688 (60 mL/min) was close to
the cardiac output in rat indicating an important contri-
bution of extra-hepatic metabolism, possibly located in
the lung. The penetration of [3H]ABP688 into the brain
was rapid and extensive, and resulted in brain to plasma
ratios of up to 20. Total radioactivity and parent com-
pound concentrations in the brain were similar (Fig. 8,
right panel), indicating that the proportion of radiola-
beled metabolites entering the brain was negligible.
The parallel decline of brain and plasma levels, with ter-
minal half lives of 1.1 and 1.0 h, respectively, suggested
passive diffusion at the blood–brain barrier.
20
40
60
80
100
Time (min)
0.75
0.50
0.25
0.00
[
H]ABP688 [nM]
10
20
30
40
50
[3H]ABP688 [nM]
Figure 4. In vitro binding characteristics of [3H]ABP688 in rat brain
membranes. (A) and (B), association and dissociation kinetics of
0.4 nM [3H]ABP688 binding. A mono-exponential model was used to
derive apparent association and dissociation rate constants. The
corresponding best-fit curves are depicted by solid curves. The data
shown are from typical experiments, each performed in triplicate. (C)
Saturation binding (inset, total and non-specific binding). The means
of the kinetic or equilibrium binding parameters derived from several
experiments are given in the text.
3.7.2. Tissue distribution. The fraction of [3H]ABP688
escaping a rapid elimination was largely distributed into
tissues as indicated by the volume of distribution
(VSS = 11 L/kg). At 5 min post-dose the highest tissue
concentration of [3H]ABP688 (21 pmol/g) was observed
in the brain. At this time, the amount of parent drug left
in the body represented only 21% of the administered
dose, with a considerable fraction (0.8% of dose) present
in the brain (Fig. 9). The next highest [3H]ABP688 con-
centration occurred in fat tissue where the concentration
measured between 5 min and 2 h post-dose correspond-
ed to 6–10% of the administered dose.
3.6.2. Plasma protein binding. The fraction of
[3H]ABP688 bound to plasma proteins was independent
of concentration from 50 to 500,000 pg/mL, the average
unbound fraction in plasma was higher in rat (6.8%)
than in man (4.4%) in accordance with the stronger par-
titioning into blood cells.
No brain penetration of metabolites and no active trans-
ports of ABP688 from and to the brain were observed.
These rat results predict a favorable human PK profile
for ABP688 as PET ligand.
3.6.3. Metabolic stability. [3H]ABP688 was incubated
with liver microsomes from rat and man. The metabolic

908
S. Hintermann et al. / Bioorg. Med. Chem. 15 (2007) 903–914
A
B
100
50
0
100
50
0
-11 -10
-9
-8
-7
-6
-5
-4
-3
-11 -10
-9
-8
-7
-6
-5
-4
-3
log([cpd]/M)
Figure 5. Concentration dependence of [3H]ABP688 displacement by ABP688 (filled circles), M-MPEP (empty triangles), MPEP (filled triangles),
and CPCCOEt (empty diamonds) in membranes prepared from whole rat brain (A) or cells expressing recombinant human hmGluR5 (B). Data show
percentage SD of specific binding of 1.5 nM [3H]ABP688. Non-specific binding was defined in the presence of 10 lM M-MPEP. Solid ( dashed)
horizontal lines indicate normalized total and non-specific controls ( SD). Best-fit curves were obtained by fitting the logistic equation to the pooled
data sets, see Table 1 for respective best-fit parameters and their 95% confidence intervals.
Table 1. Binding affinities (in nM) of MPEP, M-MPEP, and CPCCOEt in membrane preparations of hmGluR5-expressing cells and rat whole brain
Membranes from L (tk-) cells expressing hmGluR5
N
Membranes from whole rat brain
N
Compound
M-MPEP
MPEP
Ki (95% CI) (nM)
10 (9.2, 11)
20 (17, 23)
n
3
3
7
3
Ki (95% CI) (nM)
11(9.8, 13)
19(16, 23)
n
3
3
7
3
ABP688
CPCCOEt
1.7 (1.5, 1.8)
>100,000
1.8(1.5, 2.2)
>10,000
4. Discussion
(Ki = 3.5 nM). The characterization of the binding of
the tritiated analog, [3H]ABP688, showed a highly spe-
cific binding to the human (hmGluR5-expressing cells)
or rat (whole brain) mGlu5 receptor. The binding was
found to be fully displaceable by MPEP and M-MPEP,
reversible, and with a very low level of non-specific bind-
ing. In membranes from whole rat brain, the saturation
binding experiments with [3H]ABP688 revealed a high
Bmax value of 0.87 0.1 pmol/mg and a Kd of 1.9 nM.
The profiling of ABP688 against a panel of CNS recep-
tors further confirmed an excellent degree of selectivity
for mGluR5.Autoradiography experiments in rat brain
slices demonstrated that [3H]ABP688 has a high specific
binding to brain regions such as the striatum, cortex,
and hippocampus with low or no binding to the cerebel-
lum or to mid-brain regions, in line with the known dis-
tribution of the mGlu5 receptor.
Initial efforts to identify a mGluR5-selective PET tracer
based on the structural framework of the selective
antagonist MPEP failed to produce a valid PET tracer.
It is only very recently that a first series of 11C and 18F-
derivatives of MTEP and MTEB (2-methyl-4-phenyle-
thynyl-thiazole) has been described, allowing in vivo
PET imaging in rhesus monkey.
The aim of our program was to identify a molecule with
an improved tracer profile compared to the MPEP ser-
ies, particularly a higher affinity, selectivity, and lower
lipophilicity. Targeted chemical modifications of the ori-
ginal structure allowed the identification of ABP688, a
derivative in which the aromatic ring of the MPEP series
is replaced by a functionalized cyclohexenone moiety.
These modifications allowed a significant improvement
of the physico-chemical properties such as a logP below
4 and a high water solubility (120 mg/L) compared to
the previously published ligands bearing a substituted
aromatic ring. As a result, ABP688 maintained an excel-
lent selectivity and affinity for the allosteric site of the
mGlu5 receptor while significantly improving physico-
chemical properties. The radiosynthesis was straightfor-
ward, with the label introduced in the last step of the
synthesis by alkylation of oxime 7 with methyl iodide
yielding either [3H]- or [11C]ABP68816 in high radio-
chemical yield and specific activity.
In rats, despite an extensive plasma protein binding,
[3H]ABP688 rapidly and extensively penetrated the
blood–brain barrier (BBB) with brain to plasma ratios
of up to 20. The maximal brain concentrations were
reached 5 min after iv injection. Importantly, despite
extensive metabolism, no metabolites which could inter-
fere with the imaging quality were detected in the brain.
In conclusion, the profile of ABP688 supports its selec-
tion for radiolabeling with [11C] and further develop-
ment as a PET tracer, to image mGlu5 receptors
in vivo in the brain. The expected clinical benefits of
such a tracer are multiple, from example receptor
expression studies in patients, to receptor occupancy
studies and clinical dose selection for drug candidates
acting at mGlu5 receptors.
In vitro, ABP688 was found to be a highly potent func-
tional inhibitor of the hmGlu5 receptor and to bind with
high affinity to the human mGlu5 receptor

S. Hintermann et al. / Bioorg. Med. Chem. 15 (2007) 903–914
909
5nM [3H]ABP688
total binding
10 µM MPEP
CA1
CA2
CA3
Cb
CPu
DG
GP
BL
PMCo
Lateral 4.60 mm
CA1
CA2
CA3
CPu
Tu
DG
Cb
GP Thal
CeM
PMCo
Lateral 3.40 mm
CA1
CA2
CA3 DG
Cb
CPu
Thal
Ac VP
SNr
Tu
Lateral 2.10 mm
1
Figure 6. Representative autoradiographs obtained with [3H]ABP688 in sagittal sections of rat brain slices taken at three different levels. Left panels,
total binding of 5 nM [3H]ABP688. Right panels, non-specific binding determined in the presence of 10 lM MPEP. Abbreviations: Ac, accumbens
nucleus; BL, basolateral nucleus of the amygdala; CA1, CA2, CA3, fields CA1-3 of Ammon’s horn; Cb, cerebellum; CeM, central nucleus of the
amygdala; CPu, caudate putamen; DG, dentate gyrus; GP, globus pallidus; PMCo, postero medial cortical amygdaloid nucleus; SNr, substantia
nigra reticulata; Thal, thalamic nuclei; Tu, olfactory tubercle; VP, ventral pallidum.
5. Experimental
400 MHz (Varian Inc., Palo Alto, USA) or a
DMX500 spectrometer (Bruker Biospin AG, Fa¨llanden,
Switzerland). All spectra were recorded in DMSO-d6
using tetramethylsilane as internal standard.
5.1. General section
5.1.1. Chemicals. Reagents and solvents were purchased
and used without further purification. Column chroma-
tography was performed with Merck Silica Gel 60
(0.040–0.063 mm). Thin-layer chromatography to mon-
itor reactions and to identify product-containing frac-
tions was carried out using glass plates pre-coated with
silica gel (60 F254) and was developed in a UV-chamber
(254 and/or 365 nm).
5.2. Chemistry
5.2.1. Preparation of the labeling precursor 7 and
ABP688 (2, Scheme 1)
5.2.1.1. 2-Methyl-4-(6-methyl-pyridin-2-yl)-but-3-yn-
2-ol (3). To a solution of 2-bromo-6-methyl-pyridine
(29.6 mL,
260 mmol)
and
2-methyl-3-butin-2-ol
(51 mL, 520 mmol, 2 equiv) in triethyl amine (720 mL)
were added copper(I) iodide (4.0 g, 20.8 mmol, 0.08
equiv) and bis(triphenylphosphine)palladium(II)dichlo-
ride (7.3 g, 10.4 mmol, 0.04 equiv), and the resulting
mixture was heated to 55 ꢁC for 2 h. Then, the reaction
mixture was cooled to room temperature (rt), poured
onto ethyl acetate (3 L), and washed with brine (500
5.1.2. Analytics. High-resolution mass spectra were ac-
quired on a 9.4T APEX III Fourier Transform mass
spectrometer (Bruker Daltonik GmbH, Bremen, Ger-
many) and mass spectra were obtained with a Micro-
mass Platform II spectrometer in positive ESI mode.
NMR spectra were measured at 300 K on a Varian

910
S. Hintermann et al. / Bioorg. Med. Chem. 15 (2007) 903–914
5.2.1.2. 2-Ethynyl-6-methyl-pyridine (4). To a mixture
25
20
15
10
5
of crude 3 (65.8 g, <260 mmol) in benzene (760 mL) was
added freshly ground sodium hydroxide (8.9 g,
222 mmol, 0.85 equiv) and the mixture was heated to re-
flux for 1 h. Then, the mixture was cooled to rt, diluted
with methyl tert-butyl ether (MTBE, 1.5 L), and filtered.
The organic phases were washed with satd aq NaHCO3
(2· 400 mL), the aqueous phases re-extracted with
MTBE (0.5 L), the combined organic phases dried
(Na2SO4), and the solvent carefully removed under re-
duced pressure (100 mbar, 40 ꢁC bath temperature).
The dark-brown residue was purified by column chro-
matography (1 kg SiO2, eluent hexanes ! hexanes/ace-
tone 95:5 v/v), yielding 23.9 g (78% over two steps) of
the title compound as a brownish liquid.18
0
5.2.1.3. 3-(6-Methyl-pyridin-2-ylethynyl)-cyclohex-2-
enone (6). A mixture of 4 (17.0 g, 145 mmol), 5 (30.5 g,
174 mmol, 1.2 equiv), bis-(triphenylphosphine)-palladi-
um(II)dichloride (4.08 g, 5.8 mmol, 0.04 equiv), and
copper(I) iodide (2.21 g, 11.6 mmol, 0.08 equiv) in tri-
ethylamine (116 mL) and DMF (290 mL) was heated
to 55 ꢁC for 1 h. After completion of the reaction, the
mixture was diluted with ethyl acetate (3 L) and washed
with satd aq NaHCO3 (500 and 250 mL). The water
phase was extracted with ethyl acetate (1 L) and the
combined organic phases were dried (Na2SO4), filtered,
and concentrated in vacuo. The residue was purified
by column chromatography (1.2 kg SiO2, eluent hex-
ane/ethyl acetate 3:1 v/v) and the fractions containing
the title compound concentrated in vacuo to yield
23.1 g (75%) of the title compound as brownish crystals:
Figure 7. Quantitative analysis of the distribution of [3H]ABP688-
labeled sites in 12 regions of the rat brain. The data are expressed as
nCi/mg tissue SEM, data are from three animals and each region was
estimated at least nine times in total.
and 250 mL). The water phase was extracted with ethyl
acetate (1 L), and the combined organic phases dried
(Na2SO4), filtered, and concentrated in vacuo to yield
the crude title compound. It was then filtered over a sil-
ica gel pad (1 kg, eluent hexanes/acetone 4:1 v/v) to yield
65.8 g of a brownish oil, which was used for the next
1
step without further purification. H NMR (400 MHz):
1.48 (s, 6H, 2· Me), 2.45 (s, 3H, Me), 5.57 (s, 1H,
OH), 7.25 (app. t, J = 8.6 Hz, arom. H), 7.68 (t,
J = 7.8 Hz, 1H, arom. H); MS [M+H]+ 176.0.
Brain
A
Plasma
B
30
25
20
15
10
5
12
10
8
6
4
2
0
0
0
0.5
1
1.5
2
2.5
0
0.5
1
1.5
2
2.5
Time (h)
Time (h)
Figure 8. Plasma (A) and brain (B) concentrations of parent compound (squares) and total radioactivity (circles) after intravenous administration of
4 lg/kg [3H]ABP688. Value are means SD, n = 3.
50
40
30
20
10
0
Plasma
Lung
Heart
Liver
Kidney
Fat
Muscle
Skin
Brain
Figure 9. Tissue concentrations of radioactivity and parent drug. Mean concentrations ( SD, n = 3) of radioactivity (solid bars) and parent drug
(open bars) 5 min. after intravenous administration of 4 lg/kg [3H]ABP688.

S. Hintermann et al. / Bioorg. Med. Chem. 15 (2007) 903–914
911
1
mp: 57–59 ꢁC. H NMR (400 MHz): 1.96–2.04 (m, 2H,
J = 7.7 Hz, arom. H), 7.27 (m, 1H, arom. H), 7.55 (m,
1H, arom. H). 1H NMR (500 MHz, Z-isomer): 1.88
(m, 2H, CH2), 2.40 (dt, 2H, J = 1.5 and 5.9 Hz, CH2),
2.44 (m, 2H, CH2), 2.56 (s, 3H, Me), 3.88 (s, 3H,
OMe), 7.10 (d, 1H, J = 7.7 Hz, arom. H), 7.16 (t,
J = 1.7 Hz, 1H, @CH–), 7.27 (m, 1H, arom. H), 7.54
(m, 2H, arom. H). 13C NMR (125 MHz, E-isomer):
20.8 (t), 22.1 (t), 24.6 (q), 29.4 (t), 62.0 (q), 89.5 (s),
92.1 (s), 122.7 (d), 124.5 (d), 127.4 (s), 130.8 (d), 136.3
(d), 142.5 (s), 155.4 (s), 159.0 (s). 13C NMR (125 MHz,
Z-isomer): 22.3 (t), 24.6 (q), 27.5 (t), 30.4 (t), 61.6 (q),
89.3 (s), 93.0 (s), 122.9 (d), 123.1 (d), 124.6 (d), 130.8
(s), 136.3 (d), 142.2 (s), 151.8 (s), 159.1 (s); HR-MS
[M+H]+ 241.13344 (calcd 241.13354).
CH2), 2.41 (t, J = 6.6 Hz, 2H, CH2), 2.50 (s, 3H, Me),
2.54–2.56 (m, 2H, CH2), 6.27 (br d, J = 1.2 Hz, 1H,
@CH–), 7.35 (d, J = 7.8 Hz, arom. H), 7.48 (3,
J = 7.8 Hz, 1H, arom. H), 7.78 (t, J = 7.6 Hz, 1H, arom.
H); MS [M+H]+ 212.0.
5.2.1.4. 3-(6-Methyl-pyridin-2-ylethynyl)-cyclohex-2-
enone oxime (7). A solution of 6 (200 mg, 0.95 mmol)
and hydroxylamine hydrochloride (132 mg, 1.9 mmol,
2 equiv) in pyridine (9.5 mL) was stirred for 17 h at rt.
After completion of the reaction, the solvent was re-
moved in vacuo. The remaining pyridine was removed
by dissolving the residue in toluene and re-evaporation
(2· 100 mL). The resulting oil was dissolved in ethyl ace-
tate (600 mL) and washed with satd aq NaHCO3 (150
and 100 mL). The water phase was extracted with ethyl
acetate (200 mL), and the combined organic phases
dried (Na2SO4), filtered, and concentrated in vacuo.
The residue was purified by column chromatography
(250 g SiO2, eluent toluene/acetone 6:1 v/v) and the frac-
tions containing the desired compound concentrated to
yield 196 mg (91%, 1.9:1 mixture of E/Z-isomers) of
the title compound as light yellow crystals; mp 166–
5.2.1.6. [3H]ABP688 (Scheme 2). Tritiated ABP688
was synthesized by forming the potassium salt of 7
(0.45 mg, 2 lmol) in anhydrous DMF with KHMDS
(2 lmol, 0.5 M solution in toluene), followed by reaction
with [3H]-methyl iodide (0.66 lmol, solution in toluene)
at 120 ꢁC for 150 min. The product was obtained after
purification by semi-preparative HPLC (Nucleosil 100-
3 C-18 HD, 125 · 3 mm, water/acetonitrile 1:1 to 5:95
over 15 min., 1 mL/min, 40 ꢁC) as an E/Z-mixture (ca.
4:1, retention times = 14.3 and 15.8 min). Radiochemi-
cal purity of the E/Z-mixture was 99.4%, specific activity
2.83 TBq/mmol (11.78 MBq/lg).
1
168 ꢁC. H NMR (500 MHz, E-isomer): 1.73–1.85 (m,
2H, CH2), 2.40 (t, 2H, J = 5.7 Hz, CH2), 2.57 (s, 3H,
Me), 2.64 (t, J = 6.6 Hz, 2H, CH2), 6.76 (br s, 1H,
@CH–), 7.09 (d, 1H, J = 8.0 Hz, arom. H), 7.27 (m,
1H, arom. H), 7.55 (m, 1H, arom. H), 10.28 (s, 1H,
5.3. Pharmacology
1
NOH). H NMR (500 MHz, Z-isomer): 1.89 (m, 2H,
CH2), 2.44 (t, 2H, J = 6.6 Hz, CH2), 2.45 (t, 2H,
J = 5.9 Hz, CH2), 2.59 (s, 3H, Me), 7.11 (d, 1H,
J = 8.1 Hz, arom. H), 7.29 (m, 1H, arom H), 7.38 (br
s, 1H, @CH–), 7.56 (m, 1H, arom. H), 10.02 (s, 1H,
NOH). 13C NMR (125 MHz, E-isomer): 20.8 (t), 21.6
(t), 24.3 (q), 29.2 (t), 90.2 (s), 91.6 (s), 122.8 (d), 124.3
(d), 124.3 (d), 126.8 (s), 136.6 (d), 142.3 (s), 155.9 (s),
159.0 (s). 13C NMR (125 MHz, Z-isomer): 22.3 (t),
24.3 (q), 27.6 (t), 30.5 (t), 90.1 (s), 92.4 (s), 123.0 (d),
123.2 (d), 124.6 (d), 130.1 (s), 136.6 (d), 142.1 (s),
152.1 (s), 159.1 (s); HR-MS [M+H]+ 227.11789 (calcd
227.11789).
5.3.1. Cell lines. The generation and the pharmacologi-
cal characterization of the CHO and L(tk-) cell lines
expressing recombinant human mGluR1 (hmGluR1)
or human mGluR5 (hmGluR5), respectively, was de-
scribed previously.23,30 Cell culture was performed
essentially as described by Flor et al.30 Briefly, cells were
cultured in glutamate-free medium composed of
DMEM lacking L-glutamate (without phenol red, Gib-
co, #11880-028) containing a reduced concentration of
2 mM L-glutamine (Gibco, # 25030-024), supplemented
with 0.046 mg/mL proline (Sigma, # P0380), 10% dia-
lyzed fetal calf serum (Gibco, # 26300-061), and
50 mg/mL geneticin (G-418 sulfate, Gibco, #11811-
031). Expression of hmGlu5a receptors in L(tk-)
cells was induced by 1 lM dexamethasone (Sigma,
# D-2915).
5.2.1.5. 3-(6-Methyl-pyridin-2-ylethynyl)-cyclohex-2-
enone O-methyl-oxime (ABP688, 2). A solution of 6
(1.06 g, 5 mmol) and O-methyl-hydroxylamine hydro-
chloride (460 mg, 5.5 mmol, 1.1 equiv) in pyridine
(50 mL) was stirred for 64 h at rt. After completion of
the reaction, the solvent was evaporated in vacuo. The
remaining pyridine was removed by dissolving the resi-
due in toluene and re-evaporation (2· 20 mL). The res-
idue was then taken up in ethyl acetate (150 mL) and
washed with satd aq NaHCO3 (2· 25 mL), the aqueous
phases were re-extracted with ethyl acetate (50 mL), and
the combined organic phases dried (Na2SO4) and con-
centrated in vacuo. Purification by flash chromatogra-
phy (80 g SiO2, eluent toluene/ethyl acetate 9:1 v/v)
yielded 1.08 g (90%, 1.5:1 mixture of E/Z-isomers) of
the title compound as a pale yellow oil which solidified
upon standing. 1H NMR (500 MHz, E-isomer): 1.79
(m, 2H, CH2), 2.40 (t, 2H, J = 6.4 Hz, CH2), 2.54 (t,
J = 6.6 Hz, 2H, CH2), 2.56 (s, 3H, Me), 3.92 (s, 3H,
OMe), 6.56 (d, 1H, J = 1.5 Hz, @CH–), 7.09 (d, 1H,
5.3.2. Phosphoinositol accumulation assay. Measure-
ments of phosphoinositide hydrolysis in L-hmGlu5a or
CHO-hmGlu1b cells were carried out by determination
of inositol monophosphate accumulation in the presence
of lithium essentially as published previously.28,29 De-
tails are available in the Supplementary Information.
5.3.3. Membrane preparation. Membranes of hmGluR5-
expressing cells were custom prepared by CMT (Cell
and Molecular Technologies, Phillipsburg, NJ, USA),
according to a previously established protocol.9
Rat brain membranes were obtained from ABS (Analyt-
ical Biological Services, Wilmington, DE, USA). All cell
membranes used for the receptor binding panel were ob-
tained commercially from Perkin-Elmer (Boston, MA,

912
S. Hintermann et al. / Bioorg. Med. Chem. 15 (2007) 903–914
USA) or Euroscreen (Gosselies, Belgium). Details are
available in the Supplementary information.
and food (Ecosan, Eberle Nafag AG, Gossau,
Schweiz), ad libitum.
5.3.4. [3H]-M-MPEP competition assays. Radioligand
displacement assays utilizing [3H]-M-MPEP were car-
ried out essentially as described previously.22 Details
are available in the Supplementary information.
Animals were killed by decapitation, the brains were re-
moved and immediately frozen on dry ice and kept at
ꢀ80 ꢁC. Sagittal brain slices (10 lm) for ex vivo receptor
autoradiography were cut from frozen brains with a
microtome-cryostat and thaw-mounted on silane-coated
microscope slides. Receptor autoradiography was per-
formed according to the following procedure: after
30 min of pre-incubation at room temperature in KRH
buffer (Krebs–Ringer Hepes buffer) containing 20 mM
Hepes (2-[4-(2-Hydroxyethyl)-1-piperazinyl] ethanesulf-
onic acid), pH 7.4, 118 mM NaCl, 4.8 mM KCl,
2.5 mM CaCl2, 1.2 mM MgSO4, and 10 mM NaOH,
the sections were incubated for 15 min at room temper-
ature in KRH buffer, supplemented with 0.05 mg/mL
BSA (bovine serum albumin) and 5 nM [3H]ABP688
(70 Ci/mmol). Non-specific binding was determined in
a set of adjacent slides by incubation in the presence
of 10 lM MPEP. The washing of labeled sections was
carried out as follows: three 20-min washes in the ice-
cold KRH buffer (without ligand), 20 s in ice-cold 10
times diluted KRH buffer, and a brief dipping in ice-cold
distilled water to remove salts. Finally, the sections were
dried under a stream of cold air. Autoradiograms were
generated by apposing the labeled tissues to BioMax
MR Films (Eastman Kodak Company, Rochester,
New York 14650) at 4 ꢁC for 6 weeks. Sections were
finally counterstained with 0.5% Cresyl violet and nuclei
localized according to Paxinos and Watson.31 Data
from binding were analyzed by optic densitometry of
BioMax MR Films using the computerized image anal-
ysis system (MCID, Imaging Research, St. Catherines,
Ontario, Canada). For a given labeled region, the optic
density (OD) corresponding to the total binding and
non-specific binding was measured.
5.3.5. [3H]ABP688 binding assays. Reactions were per-
formed in 96-well microtiter plates in a final assay vol-
ume of 200 ll per well. The assay mixtures contained
membranes, suspended in assay buffer composed of
Na–Hepes, (30 mM), NaCl (110 mM), MgCl2
(1.2 mM), KCl (5 mM), CaCl2Æ2 H2O (2.5 mM) at pH
8.0 [3H]ABP688, and other reagents were added at the
concentrations given below. Samples were incubated at
37 ꢁC (duration given below) and then filtered through
glass-fiber filters (Unifilter-96 GF/C plate) utilizing a
96-well plate filtration unit (Inotech). The filters were
rinsed five times with cold assay buffer and dried before
the addition of a scintillation fluid (Ultimate Gold XR,
Packard; 100 ll per well). The assay plates were then
shaken for 2–3 h and subsequently counted in a liquid
scintillation counter (Wallac 1450 Microbeta or Packard
Topcount NXT). Individual determinations were per-
formed in triplicate.
5.3.5.1. Kinetic studies. To assess the time course of
association, mixtures of [3H]ABP688 (0.4 nM) and
membranes (ca. 20 lg/well, hmGluR5 40 lg/well) were
incubated at 37 ꢁC from 1 to 90 min prior to filtration.
The time course of dissociation was assessed in mem-
branes previously equilibrated with [3H]ABP688
(0.4 nM) for 60 min, after which 10 lM M-MPEP was
added 1–90 min prior to filtration. A mono-exponential
model was used to derive the observed association (kobs
)
and dissociation (kꢀ1) rate constants. The association
rate (k1) was calculated according to k1 = (kobs ꢀ kꢀ1)/
[L], where [L] was the added radioligand concentration.
5.4. DMPK studies
5.3.5.2. Saturation binding and displacement assays.
Saturation binding was assessed by adding various con-
centrations of [3H]ABP688 (0.2–50 nM) to the mem-
branes (rat 40 lg/well, hmGluR5 15 lg/well). Samples
were left to equilibrate at 37 ꢁC for 60 min prior to filtra-
tion. Non-specific binding of [3H]ABP688 was defined as
the radioactivity of samples incubated in the presence of
an excess of M-MPEP (10 lM). Samples for displace-
5.4.1. In vitro blood distribution and protein binding.
Fresh heparinized human and rat blood was used for
blood distribution studies and to obtain plasma.
5.4.1.1. Blood distribution. Rat and human blood was
spiked with [3H]ABP688 to 5–500,000 pg/mL; blood
cells and plasma were separated by centrifugation
(1500g, 10 min, 37 ꢁC). Blood distribution was measured
at 37 ꢁC with total radioactivity being quantified in both
plasma and blood; hematocrite values were determined
using microhematocrite capillaries (13,000g, 5 min,
n = 3). The fraction of compound in plasma (fP) was cal-
culated as fP (%) = (Cp/Cb) · (1-H) · 100, where Cb is
the concentration in blood, Cp is the concentration in
plasma, and H is the hematocrite value.
ment studies contained
a fixed concentration of
[3H]ABP688 (1.5 nM) and other ligands at desired final
concentrations. The data from the kinetic and equilibri-
um binding experiments were analyzed by the appropri-
ate non-linear curve fitting procedures using GraphPad
Prism 3.03 software package (GraphPad Software Inc.,
San Diego, CA). Inhibition constants were calculated
according to the Cheng–Prussoff relation.
5.4.1.2. Plasma protein binding. Plasma was spiked
with [3H]ABP688 to 50–500,000 pg/mL and submitted
to ultrafiltration at 37 ꢁC with a Centrifreeꢂ device
(Amicon Inc., Beverly, MA, USA, molecular cut-off of
30 kDa). The total radioactivity was determined in the
ultrafiltrate by liquid scintillation (Cu, concentration
of unbound compound) and in the sample introduced
5.3.6. Brain slice autoradiography. Male Sprague–Daw-
ley rats (RA238, Iffa Credo, France; 190–220 g) were
individually housed in macrolon Type II cages. The
animal room was temperature-controlled and
equipped with artificial illumination (6:00–18:00 h,
lights on). The animals had free access to water

S. Hintermann et al. / Bioorg. Med. Chem. 15 (2007) 903–914
913
into the reservoir before ultrafiltration (Cp, total plasma
concentration of compound). The unbound (fu) fraction
in plasma was calculated as fu (%) = Cu/Cp · 100.
Wyss, A. Buck, and P. A. Schubiger (Center for Radio-
pharmaceutical Science of ETH, PSI and University
Hospital Zurich) for constructive discussions, G. Bilbe,
¨
K. McAllister for support of the project.
5.4.2. In vitro biotransformation and metabolites
5.4.2.1. Liver microsomes. Liver microsomal pools
were obtained from Gentest (Woburn, MA, USA, male
Sprague–Dawley rat H501 lot 11, mixed gender Cauca-
sian human H161 lot 24). [3H]ABP688 (0.92 lmol/L)
was incubated for 40 min at 37 ꢁC with liver microsomes
(0.2 mg protein/mL) in 0.1 M sodium phosphate buffer,
pH 7.4, containing 4 mM UDPGA (Sigma, St. Louis,
MO, USA), 1 mM b-NADPH (Sigma), 5 mM MgCl2
(Merck), and 12 lg/mL. Individual samples were ana-
lyzed on-line by HPLC with radioactivity detection for
depletion of unchanged compound as well as formation
of metabolites; intrinsic clearance was calculated for
each species from the kinetic data of parent compound.
Supplementary data
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/
j.bmc.2006.10.038.
References and notes
1. Pin, J. P.; Duvoisin, R. Neuropharmacology 1995, 34, 1.
2. Spooren, W.; Ballard, T.; Gasparini, F.; Amalric, M.;
Mutel, V.; Schreiber, R. Behav. Pharmacol. 2003, 14, 257.
3. Gasparini, F.; Lingenhohl, K.; Stoehr, N.; Flor, P. J.;
Heinrich, M.; Vranesic, I.; Biollaz, M.; Allgeier, H.;
Heckendorn, R.; Urwyler, S.; Varney, M. A.; Johnson,
E. C.; Hess, S. D.; Rao, S. P.; Sacaan, A. I.; Santori, E.
M.; Velicelebi, G.; Kuhn, R. Neuropharmacology 1999, 38,
1493.
4. Spooren, W. P. J. M.; Vassout, A.; Neijt, H. C.; Kuhn, R.;
Gasparini, F.; Roux, S.; Porsolt, R. D.; Gentsch, C. J.
Pharmacol. Exp. Ther. 2000, 295, 1267.
5. Tatarczynska, E.; Klodzinska, A.; Chojnacka-Wojcik, E.;
Palucha, A.; Gasparini, F.; Kuhn, R.; Pilc, A. Br. J.
Pharmacol. 2001, 132, 1423.
5.4.3. Disposition of [3H]ABP688 in rat. Male Wistar
rats, 220–250 g (Charles River), were used. The injection
solution consisted of 28 lg [3H]ABP688 dissolved in
7.0 mL EtOH/glucose 5% (1:99 v/v). The dose was
injected (1 mL/kg) into the surgically exposed femoral
vein under slight isoflurane anesthesia.
The rats (n = 3 per sampling time) were sacrificed by iso-
flurane inhalation, heparinized blood was collected and
centrifuged to obtain plasma; tissues (lung, heart, liver,
kidney, fat, muscle, skin, and brain) were dissected.
6. Spooren, W.; Gasparini, F. Drug News Perspect. 2004, 17,
251.
The radioactivity of all samples was measured by liquid
scintillation counting (Tri-Carb liquid scintillation spec-
trometer Model A2200, Packard). The concentration of
unchanged [3H]ABP688 in plasma and tissue homoge-
nate was determined by LC-RID with 5 lg of non-la-
beled ABP688 as internal standard: [3H]ABP688 was
separated from metabolites and endogenous compounds
on a LC-18 column (Supelcosil 5 lm, 4.6 · 150 mm,
40 ꢁC, 10 mM NH4OAc–acetonitrile (45:55 v/v),
1.0 mL/min, UV-detection, k = 312 nm).The peak corre-
sponding to [3H]ABP688 was collected and radioactivity
quantified. Finally, the concentration of [3H]ABP688
was calculated from the ratio of radioactivity in the elu-
ate fraction and the area of the UV peak of non-radio-
labeled ABP688.
7. Passchier, J.; Gee, A.; Willemsen, A.; Vaalburg, W.; van
Waarde, A. Methods 2002, 27, 278.
8. Talbot, P. S.; Laruelle, M. Eur. Neuropsychopharmacol.
2002, 12, 503.
9. Gasparini, F.; Andres, H.; Flor, P. J.; Heinrich, M.;
Inderbitzin, W.; Lingenhoehl, K.; Mueller, H.; Munk, V.
C.; Omilusik, K.; Sierlin, C.; Stoehr, N.; Vranesic, I.;
Kuhn, R. Bioorg. Med. Chem. Lett. 2002, 12, 407.
10. Patel, S.; Krause, S. M.; Hamill, T.; Chaudhary, A.;
Burns, H. D.; Gibson, R. A. Life Sci. 2003, 73, 371.
11. Hamill, T. G.; Seiders, T. J.; Krause, S.; Ryan, C.;
Sanabria, S.; Gibson, R. E.; Patel, S.; Cosford, N.; Roppe,
J.; Yang, J.; King, C.; Hargreaves, R. J.; Burns, H. D. J.
Labelled Compd. Radiopharm. 2003, 46, S184.
12. Yu, M.; Tueckmantel, W.; Wang, X.; Zhu, A.; Kozi-
kowski, A. P.; Brownell, A.-L. Nucl. Med. Biol. 2005,
32, 631.
13. Hamill, T. G.; Krause, S.; Ryan, C.; Bonnefous, C.;
Govek, S.; Seiders, T. J.; Cosford, N. D. P.; Roppe, J.;
Kamenecka, T.; Patel, S.; Gibson, R. E.; Sanabria, S.;
Riffel, K.; Eng, W.; King, C.; Yang, X.; Green, M. D.;
O’Malley, S. S.; Hargreaves, R.; Burns, H. D. Synapse
2005, 56, 205.
14. Ametamey, S. M.; Kessler, L.; Honer, M.; Auberson, Y.;
Gasparini, F.; Schubiger, P. A. J. Labelled Compd.
Radiopharm. 2003, 46, S188.
15. Kokic, M.; Honer, M.; Ametamey, S. M.; Gasparini, F.;
Andres, H.; Bischoff, S.; Flor, P. J.; Heinrich, M.;
Vranesic, I.; Spooren, W.; Kuhn, R.; Schubiger, P. A. J.
Labelled Compd. Radiopharm. 2001, 44, S231.
16. Ametamey, S. M.; Kessler, L. J.; Honer, M.; Wyss, M. T.;
Buck, A.; Hintermann, S.; Auberson, Y. P.; Gasparini, F.;
Schubiger, P. A. J. Nucl. Med. 2006, 47, 698.
Acknowledgments
The authors thank H. P. Muller, W. Inderbitzin, R. Fel-
¨
ber (Chemistry); D. Fehlmann, E. Shuepbach (autoradi-
ography); P. Guntz (animal PK); R. Masero, R.
Lehmann (Radiochemistry) C. Stierlin, N. Reymann
(in vitro pharmacology), J. Hamon, Y. Gezahegen, T.
Luethi, L. Sampson, S. Staub, and S. Vidal (cross-selec-
tivity panel)for excellent technical support, E. Francotte
for stability testing and HPLC analysis, Analytics
Department (G. Bovermann, L. Oberer) for assignment
of isomers, K. Kaupmann, J. Heid, S.Vognsen, and D.
Monna for [3H]ABP688 binding assays, G. Bruin and
M. Weiss for providing in vitro metabolism and protein
binding data, S. Ametamey, L. Kessler, M. Honer, M.
17. Sashida, H.; Kato, M.; Tsuchiya, T. Chem. Pharm. Bull.
1988, 36, 3826–3832.

914
S. Hintermann et al. / Bioorg. Med. Chem. 15 (2007) 903–914
18. Sakamoto, T.; Nagata, H.; Kondo, Y.; Sato, K.; Yama-
naka, H. Chem. Pharm. Bull. 1984, 32, 4866.
19. Mewshaw, R. E. Tetrahedron Lett. 1989, 30, 3753.
20. Bax, A.; Davis, D. G. J. Magn. Reson. 1985, 63, 207.
21. Hwang, T. L.; Shaka, A. J. J. Am. Chem. Soc. 1992, 114,
3157.
22. Pagano, A.; Ruegg, D.; Litschig, S.; Stoehr, N.; Stierlin,
C.; Heinrich, M.; Floersheim, P.; Prezeau, L.; Carroll, F.;
Pin, J.-P.; Cambria, A.; Vranesic, I.; Flor, P. J.; Gasparini,
F.; Kuhn, R. J. Biol. Chem. 2000, 275, 33750.
23. Daggett, L. P.; Sacaan, A. I.; Akong, M.; Rao, S.
P.; Hess, S. D.; Liaw, C.; Urrutia, A.; Jachec, C.;
Ellis, S. B.; Dreessen, J.; Knoepfel, T.; Landwehr-
meyer, G. B.; Testa, C. M.; Young, A. B.; Varney,
M.; Johnson, E. C.; Velicelebi, G. Neuropharmacology
1995, 34, 871.
25. Shigemoto, R.; Nomura, S.; Ohishi, H.; Sugihara, H.;
Nakanishi, S.; Mizuno, N. Neurosci. Lett. 1993, 163, 53.
26. Romano, C.; van den Pol, A. N.; O’Malley, K. L. J.
Comp. Neurol. 1996, 367, 403.
27. Romano, C.; Sesma, M. A.; McDonald, C. T.; O’Malley,
K.; Van den Pol, A. N.; Olney, J. W. J. Comp. Neurol.
1995, 355, 455.
28. Lin, F. F.; Varney, M.; Sacaan, A. I.; Jachec, C.; Daggett,
L. P.; Rao, S.; Flor, P.; Kuhn, R.; Kerner, J. A.;
Standaert, D.; Young, A. B.; Velicelebi, G. Neuropharma-
cology 1997, 36, 917.
29. Litschig, S.; Gasparini, F.; Rueegg, D.; Stoehr, N.; Flor,
´
P. J.; Vranesic, I.; Prezeau, L.; Pin, J. P.; Thomsen, C.;
Kuhnm, R. Mol. Pharmacol. 1999, 55, 453.
30. Flor, P. J.; Lindauer, K.; Puttner, I.; Ruegg, D.; Lukic, S.;
Knopfel, T.; Kuhn, R. Eur. J. Neurosci. 1995, 7, 622.
31. Paxinos, G.; Watson, C. The Rat Brain in Stereotaxic
Coordinates, 2nd ed.; Academic Press: San Diego, 1986.
24. Halme, M.; Sjoholm, B.; Savola, J.-M.; Scheinin, M.
Biochim. Biophys. Acta 1995, 1266, 207.