N-Oxide analogs of WAY-100635: New high affinity 5-HT1A receptor antagonists

Article abstract of DOI:10.1016/j.bmc.2004.10.033

WAY-100635 [N-(2-(1-(4-(2-methoxyphenyl)piperazinyl)ethyl))-N-(2-pyridinyl) cyclohexanecarboxamide] 1 and its O-desmethyl derivative DWAY 2 are well-known high affinity 5-HT_1A receptor antagonists, which when labeled with carbon-11 (β⁺

Full text of DOI:10.1016/j.bmc.2004.10.033

Bioorganic & Medicinal Chemistry 13 (2005) 883–893
N-Oxide analogs of WAY-100635: new high affinity 5-HT1A
receptor antagonists
Sandrine Marchais-Oberwinkler,a,* Bartek Nowicki,a Victor W. Pike,b
Christer Halldin,c Johan Sandell,c Yuan-Hwa Chou,c Balazs Gulyas,c
d
Lise T. Brennum, Lars Farde and Hakan V. Wikstro¨m
c
a
˚
aDepartment of Medicinal Chemistry, University Center for Pharmacy, University of Groningen, Antonius Deusinglaan 1,
NL-9713 AV Groningen, Netherlands
bPET Radiopharmaceutical Sciences, Molecular Imaging Branch, National Institute of Mental Health,
National Institutes of Health, Building 10, Room B3 C346A, 10 Center Drive, Bethesda, MD 20892-1003, USA
cKarolinska Institutet, Department of Clinical Neuroscience, Psychiatry Section, Karolinska Hospital, S-17176 Stockholm, Sweden
dDepartment of Molecular Pharmacology, H. Lundbeck A/S, Ottiliavej 9, DK-2500 Valby, Denmark
Received 18 November 2003; accepted 13 October 2004
Abstract—WAY-100635 [N-(2-(1-(4-(2-methoxyphenyl)piperazinyl)ethyl))-N-(2-pyridinyl)cyclohexanecarboxamide] 1 and its O-des-
methyl derivative DWAY 2 are well-known high affinity 5-HT1A receptor antagonists, which when labeled with carbon-11 (b+;
t1/2 = 20.4min) in the carbonyl group are effective radioligands for imaging brain 5-HT1A receptors with positron emission tomogra-
phy (PET). In a search for new 5-HT1A antagonists with different pharmacokinetic and metabolic properties, the pyridinyl N-oxide
moiety was incorporated into analogs of 1 and 2. NOWAY 3, in which the pyridinyl ring of 1 was oxidized to the pyridinyl N-oxide,
was prepared via nucleophilic substitution of 2-[4-(2-methoxyphenyl)piperazin-1-yl]ethylamine on 2-chloropyridine-N-oxide fol-
lowed by acylation with cyclohexanecarbonyl chloride. 6Cl-NOWAY 4, a more lipophilic (pyridinyl-6)-chloro derivative of 3,
was prepared by treating 1-(2-methoxyphenyl)-4-(2-(2-(6-bromo)aminopyridinyl-N-oxide)ethyl)piperazine with cyclohexanecarbon-
yl chloride for acylation and concomitant chloro for bromo substitution. NEWWAY 5, in which the 2-hydroxy-phenyl group of 2 is
replaced with a 2-pyridinyl N-oxide group with the intention of mimicking the topology of 2, was prepared in five steps from 2-(chlo-
roacetylamino)pyridine. N-Oxides 3–5 were found to be high affinity antagonists at 5-HT1A receptors, with 3 having the highest
affinity and a Ki value (0.22nM) comparable to that of 1 (0.17nM). By calculation the lipophilicity of 3 (LogP = 1.87) is lower than
that of 1 by 1.25 LogP units while TLC and reverse phase HPLC indicate that 3 has slightly lower lipophilicity than 1. On the basis
of these encouraging findings, the N-oxide 3 was selected for labeling with carbon-11 in its carbonyl group and for evaluation as a
radioligand with PET. After intravenous injection of [carbonyl-11C]3 into cynomolgus monkey there was very low uptake of radio-
activity into brain and no PET image of brain 5-HT1A receptors was obtained. Either 3 inadequately penetrates the blood–brain
barrier or it is excluded from brain by an active efflux mechanism. Rapid deacylation of 3 was not apparent in vivo; in
cynomolgus monkey plasma radioactive metabolites of [carbonyl-11C]3 appeared less rapidly than from the radioligands [carbo-
nyl-11C]1 and [carbonyl-11C]2, which are known to be primarily metabolized by deacylation. Ligand 3 may have value as a new phar-
macological tool, but not as a radioligand for brain imaging.
ꢀ 2004 Elsevier Ltd. All rights reserved.
1. Introduction
its O-desmethyl analog 2 are well-known high affinity
5-HT1A receptor antagonists1–6 (Chart 1). [carbon-
yl-11C]1 is the most popular radioligand for extensive
in vivo studies of 5-HT1A receptors in human brain with
positron emission tomography (PET),5,6 while [carbon-
yl-11C]2 is a radioligand with increasing popularity be-
cause of its enhanced brain entry.7,8 These
radioligands are giving new insight into the involvement
of these receptors in neuropsychiatric disorders
(e.g., depression, anxiety, and schizophrenia)5,6 and are
also useful for drug development.9 In human subjects
WAY-100635 1 [N-(2-(1-(4-(2-methoxyphenyl)piperazin-
yl)ethyl))-N-(2-pyridinyl)cyclohexanecarboxamide] and
Keywords: WAY-100635; 5-HT1A Receptor antagonist; High affinity;
Pyridinyl N-oxide; PET; Aromatic nucleophilic substitution.
*
Corresponding author at present address: Pharmazeutische und
Medizinische Chemie, Universita¨t des Saarlandes, Postfach 151150,
D-66041 Saarbrucken, Germany. Tel.: +49 681 3022484; fax: +49
¨
681 3024386; e-mail: s.marchais@mx.uni-saarland.de
0968-0896/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2004.10.033

884
S. Marchais-Oberwinkler et al. / Bioorg. Med. Chem. 13 (2005) 883–893
[carbonyl-11C]1 and [carbonyl-11C]2 are rather rapidly
metabolized by amide hydrolysis and are rapidly cleared
from plasma.8–10 These features detract from the suc-
cessful use of mathematical models based on the use
of an arterial input function for obtaining regional bind-
ing potentials for the radioligand in human brain.11–13
Therefore, we sought to develop new analogs of 1 and
2 that might retain their pharmacological profile, but
which might have more favorable pharmacokinetics
and metabolism.
to assess their behavior as ligands at the 5-HT1A recep-
tor and, if merited, to evaluate their potential as PET
radioligands. The target compounds (3–5) were simply
the pyridine-N-oxide of 1 (compound 3), the (pyridin-
yl-6)-chloro derivative of 3 (compound 4) and com-
pound 5 in which the 2-hydroxyphenyl group of 2 is
replaced with a 2-pyridinyl N-oxide group (compound
5) (Chart 1).
2. Results and discussion
2.1. Chemistry
Oxidation of a pyridinyl nitrogen to the N-oxide can
have various effects on drug-like behavior,14 often giving
drugs with good oral bioavailability,15,16 pharmacoki-
netic properties,15,16 solubility,15 and metabolic stabil-
ity.15,17 Affinity at the target receptor or binding site
may also be increased18,19 or reduced.20 Formation of
the N-oxide is a common route of phase 1 metabolism
for drugs containing pyridinyl moieties, and often these
metabolites may contribute to drug action.13,14 Thus, N-
oxide analogs of 1 and 2 might be interesting ligands for
the 5-HT1A receptor. We aimed to prepare such analogs,
Retrosynthetic analysis of the N-oxide 3 indicated that
the pyridine-N-oxide 12 might be a useful precursor
(Scheme 1). The syntheses of similar pyridine-N-oxide
derivatives via nucleophilic substitutions in 2-halo-pyr-
idine-N-oxides with the amine 14 and subsequent acyl-
ation with cyclohexanecarbonyl chloride have been
described.21
During acylation a side reaction might be expected be-
cause of the nucleophilicity of the oxygen in the N-oxide
moiety. For example, Deady and Stanborough22 found
that the N-oxide moieties in pyridin-2-amine-N-oxides
are more reactive than corresponding amines toward
acylation. In their study, O-acylation was followed by
a rapid intramolecular rearrangement, involving a fav-
ored five-center transition state to provide the N-acyl
compound.
5
N
Y
N
R2
6
X
1
N
R1
O
R3
Ligand
X
Y
R1
R2
R3
Acylation was studied on a model compound, 6-chloro-
N-propylpyridin-2-amine-N-oxide 15, to find acceptable
reaction conditions (Scheme 2). Acylation was first per-
formed with an excess of cyclohexanecarbonyl chloride
(50molequiv) in order to detect and identify an O-acyl
moiety. The N,O-diacyl compound 16 was isolated in
82% yield. One molar equivalent of acid chloride gave
a mixture of the N-acyl 17 and O-acyl 18 compounds,
the major product being 17. Reaction time slightly influ-
enced the ratio of the two products (i.e., 67% of 17 and
10% of 18 after 30min, and 50% of 17 and 25% of 18
after 22h).
WAY-100635
DWAY
NOWAY
6Cl-NOWAY
NEWWAY
trans-FCWAY
p-MPPF
1
2
3
4
5
6
7
8
9
N
N
C
C
C
C
N
C
C
C
C
C
C
OMe
OH
OMe
OMe
O
OMe
OMe
OMe
OMe
OMe
OMe
c.hexyl
-
-
-
c.hexyl
N=O
N=O
N
c.hexyl
6-Cl
-
-
-
6-F
6-Br
6-I
5-Br
c. hexyl
c.hexyl
N
trans-4-F-c.hexyl
4-F-phenyl
c.hexyl
N
6FPWAY
6BPWAY
6IPWAY
N
N
c.hexyl
10
11
N
c.hexyl
5BPWAY
N
c.hexyl
Chart 1. Structures of 5-HT1A receptor ligands.
N
N
N
ClOC
N
OMe
OMe
+
N
O
N
H
N
O
N
12
O
3 (NOWAY)
N
N
Cl
N
OMe
+
NH2
O
14
13
Scheme 1. Retrosynthetic analysis of 3 (NOWAY).

S. Marchais-Oberwinkler et al. / Bioorg. Med. Chem. 13 (2005) 883–893
885
+
Cl
O
N
O
N
Excess,
MeCN, RT, 18 h
O
Cl-
COCl
82%
Cl
N
N
H
16
O
15
+
Cl
+
N
N
H
Cl
N
O
N
O
1 Eq.
O
O
MeCN, RT, 30 min
Cl-
67%
10%
18
17
Scheme 2. Acylation of the N-oxide 15.
NOWAY 3 was prepared via the aromatic nucleophilic
substitution21 of 2-(4-(2-methoxyphenyl)piperazin-1-
yl)ethylamine 14 on 2-chloropyridine-N-oxide 13 to give
compound 12 in 30% yield, followed by acylation with
cyclohexanecarbonyl chloride for 55min to give about
a 3:1 mixture of 3 (N-acyl) and 19 (O-acyl) products in
moderate yield (3 plus 19, 53%) (Scheme 3); not all start-
ing material 12 was consumed.
COCl
4 (6Cl-NOWAY)
N
MeCN
N
OMe
82%
Br
N
O
N
H
0 oC for 20 min
and RT for 20 h
20
Scheme 4. Synthesis of 6Cl-NOWAY (4).
The two O-acyl compounds 18 (Scheme 2) and 19
(Scheme 3) could be isolated by TLC and studied with
The pyridinyl N-oxide group is considered isosteric with
certain substituted phenyl groups (e.g., nitro-phenyl),
and therefore replacement of a susbstituted phenyl
group with a pyridinyl-N-oxide group has been consid-
ered a strategy for exploring structure–activity relation-
ships in drug development.24,25 NEWWAY 5 was
designed to replace the (2-hydroxy)phenyl moiety of 2
by a 2-pyridinyl N-oxide moiety. The strategy for the
synthesis of 5 was to install the pyridine-N-oxide in
the last step to prevent any nucleophilic attack of the
N-oxide group during synthesis. A benzyl group was
chosen for protection of the piperazinyl moiety. The
synthesis (Scheme 5) started from 2-chloro-N-pyridin-
2-yl-acetamide 21.26 Nucleophilic substitution with N-
benzylpiperazine, reduction with lithium aluminum
hydride (LAH), and acylation with cyclohexanecarbonyl
chloride gave the benzyl intermediate 24 in 38% overall
1
IR but 19 was so unstable that no H NMR spectrum
could be recorded. The N-acyl compounds 17 (Scheme
2) and 3 (Scheme 3) were also isolated by TLC and were
stable when kept dry under an inert atmosphere.
The synthesis of 6Cl-NOWAY 4 was achieved similarly
by acylation of the 2-bromopyridine-N-oxide amine 2023
with cyclohexanecarbonyl chloride at 0ꢁC for 20min
and then at rt for 20h. Chlorinated ligand 4 was ob-
tained in 82% yield (Scheme 4). Different reaction condi-
tions were tried but none gave the expected bromo
analog. The bromine atom at the 6-position of the pyr-
idine-N-oxide was immediately exchanged by chloride
ion even in the presence of an excess of triethylamine
(TEA). Low temperature (0ꢁC) suppressed any forma-
tion of the O-acyl analog.
COCl
n-BuOH, TEA
N
120 oC, 20 h
+
N
OMe
3 (NOWAY) +
39%
12
13 + 14
N
N
H
O
MeCN,
O
RT, 55 min
Cl-
14%
19
Scheme 3. Acylation of the N-oxide 12; synthesis of NOWAY (3).

886
S. Marchais-Oberwinkler et al. / Bioorg. Med. Chem. 13 (2005) 883–893
Bn
NH
N Bn
O
O
N
Cl
N
MeCN, n-Bu4N+I-,TEA
Reflux, 6 h
N
N
H
N
N
H
87%
22
21
LAH, THF
Reflux, 20 h
COCl
Bn
Bn
N
N
N
N
N
N
N
N
H
CH2Cl2, DMAP, TEA
RT, 4d
O
52%
23
85%
24
EtOH
H2/Pd
RT, 24 h
NH
13
N
5 (NEWWAY)
N
N
n-BuOH, MeCN
140 0C, 26 h
20%
O
90%
25
Scheme 5. Synthesis of NEWWAY (5).
yield (three steps). Ligand 5 was obtained after removal
of the protecting benzyl group to give 25, and aromatic
nucleophilic substitution of the chloro group in the 2-
chloropyridine-N-oxide 13 with the piperazinyl group
of 25.
receptor for substitutions on the pyridinyl ring of 1
(Table 2). This tolerance has previously been noted for hal-
ogen substitutions for hydrogen in the 5- and 6-positions
(Table 2; compounds 8–11).27,28 Tolerance depends on
the nature of the halogen in the 6-position and, in the
case of bromine, distance from the pyridinyl nitrogen.
These variations may be due to steric and/or electronic
effects.
2.2. Pharmacology
Compounds 3–5 were assessed for their ability to inhibit
the binding of [3H]5-carboxamido-tryptamine ([3H]5-
CT) to human cloned 5-HT1A (h5-HT1A) receptors
in vitro (Table 1). NOWAY 3 displayed high affinity for
h5-HT1A receptors, similar to the high affinity of 1 and
2, while 6Cl-NOWAY 4 displayed 10-fold less affinity
and NEWWAY 5 almost 200-fold less affinity. The high
affinity of 3 shows the spatial tolerance of the 5-HT1A
The ability of each new ligand 3–5 to inhibit the produc-
tion of cAMP in the presence of forskolin was also as-
sessed on h5-HT1A receptors with serotonin (5-HT)
(1lM). Ligands 3–5 displayed complete antagonism in
vitro (no intrinsic efficacy).
2.3. Lipophilicity
Lipophilicity is a key predictor of plasma protein bind-
ing, blood–brain barrier (BBB) penetration, and the
level of brain nonspecific binding of a PET radioligand
in vivo.29 The lipophilicities of ligands 1–5 were esti-
mated by computation of LogP and LogD from their
structures (Table 1). The values for the N-oxides de-
crease in the order 4, 3, 5 and are all lower than the cor-
responding values for 1 and 2. The calculated LogP
values for 3 and 5 are, respectively, close to and within
the desirable range (2.5–3.5) for a compromise between
good BBB penetration and low nonspecific binding in
brain.29 Chromatographic mobility, especially in reverse
phase HPLC, may provide a good experimental index of
lipophilicity. In TLC (silica gel; CH2Cl2–MeOH; 9:1 v/v)
Table 1. In vitro binding affinities and calculated LogP and LogD
values for ligands 1–5
Ligand
Affinity Ki
(nM)
LogP
(Pallas)
LogDa
(Pallas)
WAY-100635 1
DWAY 2
NOWAY 3
0.17
3.28
2.77
1.87
2.69
1.25
2.54
2.03
1.14
1.95
0.85
n.d.b
0.22
6Cl-NOWAY 4
NEWWAY 5
1.6 (n = 6)
33 (n = 5)
a Log of overall equilibrium distribution between octanol and phos-
phate buffer at pH7.4, uncorrected for ionization.
b Not determined. IC50 = 2.1 0.52nM and pA2 = 8.0;3 IC50
1.4nM.4 Cf. Ligand 1: IC50 = 0.37 0.02nM.3
=

S. Marchais-Oberwinkler et al. / Bioorg. Med. Chem. 13 (2005) 883–893
887
Table 2. Effect of substituents at the pyridine ring of 1 on affinity at h5-HT1A receptors and on maximal brain uptake in cynomolgus monkey brain
Ligand
Pyridinyl ring substituent(s)
Ki (nM)
Maximal brain uptakea (% ID)
Reference
WAY-100635 1
NOWAY 3
—
1-O
0.029
0.028
0.93
5.0 at 2min
0.65 at 1min
n.m.
This work, 21
This work
This work
21,27
6Cl-NOWAY 4
6FPWAY 8
6BPWAY 9
6-Cl, 1-O
6-F
10.0
8.0 at 10min
5.2 at 5min
6-Br
6-I
4.30
10.00
2.80
21,27
28
21,27
6IPWAY 10
5BPWAY 11
5-Br
n.m.
n.m. = not measured. Data are uncorrected for radioactivity in blood.
a Measured with intravenously administered [carbonyl-11C]labeled ligand with PET.
analysis, ligands 1, 3, and 4 had Rf values of 0.72, 0.5,
and 0.5, respectively, so indicating 3 and 4 to be some-
what less lipophilic than 1, in accord with the calculated
LogP/LogD values. When aqueous ammonia was
added to the mobile phase or when TLC was performed
on an aluminum oxide layer, ligands 1 and 3 showed the
same mobility, indicating the same lipophilicity. Ligand
3 showed only slightly lower shorter retention time
(6.3min) than ligand 1 under several reverse phase con-
ditions (e.g., on l-Bondapak C-18 column eluted with
MeCN–0.01M H3PO4, 35:65 v/v). Thus, by considera-
tion of calculated LogP/LogD values and chromato-
graphic mobility, ligands 3 and 4 appear overall to
have only somewhat lower lipophilicity than 1. Com-
pound 5 was calculated to have substantially lower
LogP and LogD values than 1 or 2 and was not assessed
chromatographically.
brain 5-HT1A receptors. Hence, we set out to evaluate
[carbonyl-11C]3 as a potential PET radioligand in cyno-
molgus monkey. Ligand 3 was successfully labeled in its
carbonyl position with carbon-11 in 80% decay-cor-
rected radiochemical yield according to a previously de-
11
scribed C-acylation technique,30,31 but using amine 12
as precursor.
2.4. PET experiments
[carbonyl-11C]3 was used in two separate PET experi-
ments in a single cynomolgus monkey, one in which
the radioligand was given alone (baseline experiment)
and one in which the radioligand was given after treat-
ment of the monkey with 1 at a dose known to block
monkey brain5-HT1A receptors7,32 (pre-treatment
experiment).
The prima facie properties of ligand 3 [high affinity,
antagonist action (i.e., no efficacy), moderate lipophilic-
ity, low molecular weight, and a site for radiolabeling]
are very similar to those of 1 and 2, and they suggested
that 3, when labeled with carbon-11 (b+; t1/2 = 20.4min)
might be an effective radioligand for PET imaging of
No image of brain 5-HT1A receptors was obtained after
intravenous injection of [carbonyl-11C]3 in either PET
experiment due to a very low whole brain uptake of
radioactivity (Fig. 1). This was about 0.54% of injected
dose at about 2.5min, steadily declining to about 0.32%
of injected dose at 90min in the baseline experiment,
Figure 1. Horizontal (panels A and D), sagittal (panels B and E), and coronal (panels C and F) PET images of cynomolgus monkey head after
intravenous injection of NCA [carbonyl-11C]NOWAY in the baseline experiment (panels A–C) and pre-treatment experiment (D–F). Images are from
emission data summed from 9min after injection.

888
S. Marchais-Oberwinkler et al. / Bioorg. Med. Chem. 13 (2005) 883–893
and about 0.59% of injected dose at about 2.5min,
steadily declining to about 0.15% of injected dose at
90min in the pre-treatment experiment (Fig. 2). Time–
radioactivity curves for cerebellum were very similar in
each experiment. The higher whole brain radioactivity
(as % of injected dose) between 25 and 90min in the
baseline experiment compared to that in the pre-treat-
ment experiment may indicate some generalized but very
low level of 5-HT1A receptor-specific binding. It should
be noted that these data are uncorrected for blood radio-
activity content. For comparison [carbonyl-11C]1 and
some halo analogs maximally enter cynomolgus monkey
brain to a strikingly greater extent, namely about 5–8%
of injected doses at 2.5–8min after injection7,10 (Table
2). The somewhat lower lipophilicity of 3, compared
to 1, may not account alone for the poor ability of 3
to enter monkey brain.
109 itself.38 For compounds related to Lu 25-109, LogD
did not predict BBB penetration, indicating the possible
role of transporter proteins in uptake and/or elimination
from brain. These reports and our findings suggest that
Pgp and/or other transporter proteins (pumps) may gen-
erally exclude N-oxides from brain, perhaps as part of a
Ôdrug efflux-metabolism allianceÕ.39 The possibility that
3, by virtue of its structural similarity to 1 and 7 and
its N-oxide moiety, is a very good substrate for Pgp or
other transporter was not however investigated further
in this study.
In the baseline and pre-treatment PET experiments, the
proportions of radioactivity in monkey plasma repre-
sented by parent radioligand were, respectively, 80%
and 94% at 5min, 55% and 55% at 30min, and 46%
and 46% at 45min. [carbonyl-11C]1 and [carbonyl-11C]2
are each primarily metabolized by deacylation (amide
,
Ligand 1 and the structurally related ligand, p-MPPF
(N-(2-(1-(4-(2-methoxyphenyl)piperazinyl)ethyl))-N-(2-
pyridinyl)p-fluorophenylcarboxamide, 7) are known to
hydrolysis),8,10 and this is the expected route of metab-
olism for [carbonyl-11C]3. The HPLC analysis of the
radioactive metabolites of [carbonyl-11C]3 is consistent
with this expectation, since only more polar radioactive
metabolites were observed. Comparison of the limited
data from the two experiments with [carbonyl-11C]3
and the averaged data published for [carbonyl-11C]1
from six monkeys,10 suggests that [carbonyl-11C]3 is less
rapidly metabolized than [carbonyl-11C]1 in cynomolgus
monkey.
be substrates for rat brain P-glycoprotein (Pgp),33
a
pump which can exclude certain compounds (substrates)
from brain. Indeed, 7 has significantly lower brain up-
take than 1 because it is so much better a substrate for
elimination from brain by Pgp.34 Thus, it may reason-
ably be expected that 3 is also a substrate for brain
Pgp, based on its very close structural similarity to 1
and 7.
Only a few studies have examined the uptake of N-oxi-
des into rodent or primate brain directly. The brain up-
takes of clozapine and its N-oxide metabolites have been
examined in mice.35 At 10min after i.p. administration,
clozapine has 24-fold higher brain uptake than the N-
oxide. A further study has shown that clozapine N-oxide
is unable to cross the BBB in rat,36 though N-oxidation
is the main route of clozapine metabolism in rat brain.37
The N-oxide of Lu 25-109 (5-ethyl-2H-tetrazol-5-yl-
1,2,3,6-tetrahydro-methylpyridine), a M1 muscarinic
agonist, does not cross the BBB in rat, unlike Lu 25-
3. Conclusions
Three new high affinity antagonists of the h5-HT1A
receptor 3–5 were discovered. Among these, the N-oxide
3 is notable in having lipophilicity and high affinity that
are similar to those of the well known antagonists, 1 and
2. Ligand 3 however does not enter brain to any great
extent in monkey and when labeled with carbon-11 gives
no PET image of the distribution of brain 5-HT1A recep-
tors. Ligand 3 is possibly a good substrate for Pgp or
another transporter protein that actively excludes it
from brain. [carbonyl-11C]3 appears to have a slower
rate of appearance of radioactive metabolites in plasma
than [carbonyl-11C]1 and [carbonyl-11C]2. Ligand 3 may
have potential value as a new pharmacological tool but
not as a radioligand for brain imaging. In ligand 4 the
6-chloro group increased lipophilicity but reduced affi-
nity for the 5-HT1A receptor by almost 7-fold; this
shows the tolerance of the 5-HT1A receptor for substit-
uents in both the 1- and 6-positions of the pyridinyl ring
of 1. Ligand 5 has almost 200-fold lower affinity for the
5-HT1A receptor than 2. Hence, the N-2-pyridinyl moi-
ety is not an especially good bioisosteric mimic for the
1.00
0.75
0.50
0.25
From this study, the possibility that [carbonyl-11C]1 is metabolized in
vivo via rapid oxidation to its N-oxide, followed by intramolecular
trans-O-acylation and O-deacylation might be considered. However,
after the injection of [carbonyl-11C]1 into cynomolgus monkey, no
radioactive N-oxide has been specifically detected in plasma with
radio-HPLC.10 Also careful investigations of other radioligands that
are closely related in structure to 1 (e.g., trans-FCWAY; 6) have not
revealed any pyridinyl-N-oxides forming as metabolites in human or
liver microsomes.40,41
0.00
0
30
60
90
Time after injection (min)
Figure 2. Time courses for radioactivity (% injected dose) in monkey
whole brain and cerebellum after injection of NCA [carbon-
yl-11C]NOWAY: d cerebellum in baseline experiment;
ꢀ
cerebellum
in pre-treatment experiment; j whole brain in baseline experiment; Ã
whole brain in pre-treatment experiment.

S. Marchais-Oberwinkler et al. / Bioorg. Med. Chem. 13 (2005) 883–893
889
2-hydroxyphenyl group in DWAY 2 or the N-(2-meth-
oxy)phenyl groups in WAY 1, which are recognized as
being important structural features for 5-HT1A receptor
binding.42
yl]ethylamine 14 (4.38g; 18.6mmol) in n-BuOH
(80mL). The mixture was heated at 120ꢁC for 20h
and then the reaction stopped by evaporation of solvent.
Column chromatography (silica gel; CH2Cl2–MeOH,
95:5 v/v) provided pure 12 (1.83g; 30%). The hydrochlo-
ride salt was then prepared by adding HCl (1M in Et2O;
4equiv) and recrystallized from absolute EtOH. IR
4. Experimental
4.1. General
(neat): 1625, 1569, 1498, 1241 (N!O) cmꢁ1
;
1H
NMR (300MHz, CDCl3): 8.05 (d, J = 6Hz, 1H), 7.13
(t, J = 8Hz, 1H), 7.05 (br s, 1H), 6.78–6.94 (m, 4H),
6.49–6.57 (m, 2H), 3.80 (s, 3H), 3.31–3.37 (m, 2H),
3.00–3.07 (m, 4H), 2.68–2.74 (m, 6H); 13C NMR
(50MHz, CDCl3): 139.60, 136.00, 127.00, 121.50,
119.50, 116.80, 109.90, 109.55, 104.35, 54.95, 53.85,
51.80, 48.95, 37.50; MS (EI): 328 [M]+; Anal. Calcd
for C18H24N4O2: C, 65.83; H, 7.37; N, 17.06. Found:
C, 65.39; H, 7.35; N, 16.70.
4.1.1. Materials. Chemical names follow IUPAC
nomenclature. WAY-100635 1 was prepared as de-
scribed previously.43 2-[4-(2-Methoxyphenyl)piperazin-
1-yl]ethylamine 14,44 2-chloropyridine-N-oxide 13,45
1-(2-methoxyphenyl)-4-(2-(2-(6-bromo)aminopyridinyl-
N-oxide)ethyl)piperazine 20,21 and 2-chloro-N-pyridin-
2-yl-acetamide 2126 were prepared as reported. All other
chemicals were obtained commercially (Aldrich, Acros
or Fluka) and used without purification.
4.2.2. General acylation procedure (on 12, 15, and 20).
Cyclohexanecarbonyl chloride (1molequiv) was added
to a solution of amine (1molequiv) in MeCN under
N2. The mixture was stirred at rt or at 0ꢁC (ice-water
bath) and then the reaction quenched with aq Na2CO3
solution (10% w/v). Product was extracted with CH2Cl2,
dried over MgSO4, filtered, and concentrated.
4.1.2. Analytical, spectroscopic, and chromatographic
methods. Melting points were determined on a digital
melting point apparatus (Electrothermal) and are
uncorrected.
Elemental analyses were performed by the Microanalyt-
ical Laboratory of the University of Groningen and
were within 0.4% of the theoretical values.
4.2.2.1.
N-(2-(1-(4-(2-Methoxyphenyl)piperazinyl)-
ethyl))-N-(2-pyridinyl-N-oxide)cyclohexanecarboxamide,
NOWAY (3). Acylation was performed on 12 (33mg;
99lmol) at rt. Preparative TLC (0.2mm layer;
CH2Cl2–MeOH, 9:1 v/v) gave pure 3 (17mg, 39%). IR
IR spectra were obtained on an ATI-Mattson
spectrometer.
(neat): 1673 (amide), 1496, 1451, 1240 (N!O) cmꢁ1
;
1H NMR spectra were obtained in CDCl3 on a Varian
Gemini (300 or 200MHz) instrument at 300K. 13C
NMR spectra were obtained in CDCl3 on a Varian Ge-
mini (50MHz) instrument at 300K. Chemical shifts are
reported in d values (ppm), by reference to the hydro-
genated residues of deuteriated solvent as internal stand-
1H NMR (200MHz, CDCl3): 8.26–8.30 (m, 1H),
7.51–7.56 (m, 1H), 7.24–7.31 (m, 2H), 6.81–7.03 (m,
4H), 4.13 (m, 1H), 3.83 (s, 3H), 2.99 (br s, 4H), 2.65
(br s, 6H), 1.03–1.94 (m, 12H); 13C NMR (50MHz,
CDCl3): 175.20, 150.65, 146.40, 139.40, 138.70,
125.70, 124.20, 123.40, 121.55, 119.40, 116.55, 109.70,
54.45, 53.85, 51.60, 48.70, 41.55, 41.25, 24.05; MS
(ES): 439 [M+H]+.
1
ard (CDCl3, d = 7.24ppm in H NMR and d = 77ppm
in 13C NMR). Signals are described as s, d, t, dd, m,
br for singlet, doublet, triplet, double doublet, multiplet,
and broad, respectively.
4.2.2.2. 1-(2-Methoxyphenyl)-4-(2-(2-(1-cyclohexane-
acyloxypyridinium)amino)ethyl)piperazine chloride (19).
Acylation was performed on 12 (33mg; 99lmol) at 0ꢁC.
Preparative TLC (0.2mm layer; CH2Cl2–MeOH, 9:1
v/v) gave pure 19 (6mg; 14% yield). IR (neat): 1730
(NOCOC6H11), 1458cmꢁ1; MS (ES): 439 [M+H]+.
Mass spectra (ES) were recorded on a Sciex API 3000
instrument or on a Unicam 610/Automass 150 GC–
MS system (EI). HRMS data were obtained by the
Microanalytical Department of the University of Gron-
ingen on a Jeol MS Route JMS600H instrument.
4.2.2.3. N-(6-Chloro-1-cyclohexaneacyloxypyridini-
um-2-yl)-N-propylcyclohexanecarboxamide
Analytical thin layer chromatography (TLC) was per-
formed on silica gel layers (60 PF 254 pre-coated alum-
inum sheets, 0.2mm layer; Merck) or aluminum oxide
layers (60 F 254 neutral precoated aluminum sheets,
0.2mm layer; Merck). Preparative TLC was performed
on silica gel (60 PF 254 pre-coated glass plate; 1 or
2mm layer; Merck).
chloride
(16). Acylation was performed on 15 (20mg; 0.1mmol).
Preparative TLC (0.2mm layer; CH2Cl2–MeOH, 95:5
v/v) gave pure 16 (39mg, 82%). IR (neat): 1705 (NO-
1
COC6H11) cmꢁ1; H NMR (300MHz, CDCl3): d 7.46–
7.49 (dd, J = 3Hz, J = 7Hz, 1H), 7.15–7.21 (m, 2H),
2.22–2.32 (m, 4H), 1.10–1.90 (m, 22H), 0.82 (t,
J = 7Hz, 3H).
4.2. Chemistry
4.2.2.4.
N-(6-Chlorooxidopyridin-2-yl)-N-propylcy-
4.2.1.
1-(2-Methoxyphenyl)-4-(2-(2-aminopyridinyl-N-
clohexanecarboxamide (17). Acylation was performed
on 15 (20mg; 0.1mmol) at rt. Preparative TLC
(0.2mm layer; CH2Cl2–MeOH, 95:5 v/v) gave pure 17
(21.5mg; 67%). IR (neat): 1670 (amide), 1272 (N!O)
oxide)ethyl)piperazine (12). 2-Chloropyridine-N-oxide
13 (2.41g; 18.6mmol) and TEA (2.59mL; 18.6mmol)
were added to 2-[4-(2-methoxyphenyl)piperazin-1-

890
S. Marchais-Oberwinkler et al. / Bioorg. Med. Chem. 13 (2005) 883–893
cmꢁ1; 1H NMR (300MHz, CDCl3): d 7.46 (dd, J = 3Hz,
J = 7Hz, 1H), 7.11–7.21 (m, 2H), 1.36–1.85 (m, 11H),
0.901.19 (m, 4H), 0.82 (t, J = 8Hz, 3H); MS (EI): 296–
298 [M]+.
The organic layers were dried over Na2SO4, filtered, and
concentrated. Preparative TLC (1mm layer; CH2Cl2–
MeOH, 9:1 v/v) gave pure 23 (105mg; 52%). IR (neat):
3368, 1602, 1494, 1444cmꢁ1 1H NMR (300MHz,
;
CDCl3): 8.01 (d, J = 4Hz, 1H), 7.34 (t, J = 7Hz, 1H),
7.26–7.17 (m, 5H), 6.49 (t, J = 6Hz, 1H), 6.32 (d,
J = 9Hz, 1H), 5.10 (br s, 1H), 3.45 (s, 2H), 3.28 (m,
2H), 2.56 (t, J = 5Hz, 2H), 2.44 (m, 8H); 13C NMR
(75MHz, CDCl3): 156.25, 145.60, 135.50, 134.80,
126.70, 125.70, 124.55, 110.15, 104.55, 60.50, 54.15,
50.45, 50.35, 36.00; MS (EI): 202, 189, 146, 119.
4.2.2.5. (6-Chloro-1-cyclohexaneacyloxypyridinium-2-
yl)-N-propyl chloride (18). Acylation was performed on
15 (20mg; 0.1mmol) at rt. Preparative TLC (0.2mm
layer; CH2Cl2–MeOH, 95:5 v/v) gave pure 18 (3.3mg;
10%). IR (neat): 3284 (amine), 1707 (NOCOC6H11),
1
1621, 1559cmꢁ1; H NMR (300MHz, CDCl3): d 7.06–
7.21 (m, 2H), 6.68 (d, J = 8Hz, 1H), 6.46 (d, J = 8Hz,
1H), 3.16–3.22 (m, 2H), 2.21–2.28 (m, 1H), 1.86 (d,
J = 13Hz, 2H), 1.59–1.70 (m, 5H), 1.17–1.44 (m, 5H),
0.96 (t, J = 7Hz, 3H).
4.2.5. N-(2-(1-(4-(Benzyl)piperazinyl)ethyl))-N-(2-pyridi-
nyl)cyclohexanecarboxamide (24). Cyclohexanecarbonyl
chloride (29lL; 43.2mg; 0.29mmol; 1.2equiv), TEA
(51lL; 37mg; 0.37mg; 1.5equiv), and a catalytic
amount of 4-(dimethylamino)pyridine were added suc-
cessively to [2-(4-benzyl-piperazin-1-yl)-ethyl]-pyridin-
2-yl-amine 23 (73mg; 0.25mmol) in CH2Cl2 (5mL) at
0ꢁC. The mixture was stirred under N2 at 0ꢁC for
15min and then at rt for 4days. Reaction was stopped
by addition of H2O and Na2CO3. The mixture was ex-
tracted with dichloromethane and the organic layers
dried over Na2SO4, filtered, and concentrated. Prepara-
tive TLC (three plates, 0.2mm layers; CH2Cl2–MeOH;
95:5 v/v) gave pure 24 (85mg; 85%). IR (neat): 1655,
4.2.2.6.
N-(2-(1-(4-(2-Methoxyphenyl)piperazinyl)-
ethyl))-N-(2-(6-chloro)pyridinyl-N-oxide)cyclohexanecar-
boxamide, 6Cl-NOWAY (4). Acylation was performed
on 20 (65mg; 0.16mmol). Preparative TLC (1mm layer;
CH2Cl2–MeOH, 90:10 v/v) gave pure 4 (62mg; 82%). IR
(neat): 1678 (amide), 1597, 1500, 1240 (N!O) cmꢁ1
;
1H NMR (300MHz, CDCl3): d 7.43 (m, 2H), 7.12 (m,
1H), 6.95–6.76 (m, 4H), 4.00 (s, 1H), 3.76 (s, 3H),
2.92 (br s, 4H), 2.62 (br s, 6H), 1.83–0.96 (m, 12H);
13C NMR (50MHz, CDCl3): d 175.30, 150.65, 147.90,
141.20, 139.30, 124.45, 123.35, 122.80, 121.65,
119.45, 116.60, 109.70, 54.50, 53.85, 51.55, 48.85, 48.50,
41.70, 41.25, 28.55, 27.60, 24.30, 24.05; MS (ES): 473
[M+H]+.
1585, 1466cmꢁ1 1H NMR (200MHz, CDCl3): 8.48–
;
8.52 (m, 1H), 7.78–7.69 (m, 1H), 7.30–7.18 (m, 7H),
3.93 (t, J = 7Hz, 2H), 3.47 (s, 2H), 2.57–2.16 (m,
11H), 1.71–1.43 (m, 7H), 1.25–1.02 (m, 3H); 13C NMR
(50MHz, CDCl3): 174.75, 147.70, 136.55, 136.45,
127.70, 126.65, 125.50, 120.75, 120.65, 61.50, 54.55,
51.50, 43.70, 40.85, 28.00, 24.10; MS (ES): 407
[M+H]+; HRMS (EI): Calcd for C27H36N4O:
406.2732, obsd: 406.2718.
4.2.3.
2-(4-Benzyl-piperazin-1-yl)-N-pyridin-2-yl-acet-
amide (22). N-Benzylpiperazine (5.10g; 29mmol), TEA
(12.1mL; 8.78g; 87mmol; 3equiv) and n-Bu4NI
(535mg; 1.45mmol; 0.05equiv) were added to 2-
chloro-N-(pyridin-20-yl)-acetamide 21 (5.42g; 31.9
mmol; 1.10equiv) in MeCN (250mL). The mixture
was refluxed for 6h and the reaction stopped by evapo-
ration of the solvent. The residue was dissolved in
CH2Cl2. The organic layer was successively washed with
H2O, aq Na2CO3 (10% w/v), and brine. The organic
layer was dried over Na2SO4, filtered, and concentrated.
Purification by column chromatography (silica gel;
CH2Cl2–MeOH, 98:2 v/v) provided pure 22 (8.61g;
4.2.6.
N-(2-(1-Piperazinyl)ethyl)-N-(2-pyridinyl)cyclo-
hexanecarboxamide (25). Palladium on charcoal
(10mg; 10%) was added to N-(2-(1-(4-(benzyl)piperazin-
yl)ethyl))-N-(2-pyridinyl)cyclohexanecarboxamide
24
(40mg; 98lmol) in absolute EtOH (6mL). A hydrogen
balloon was attached to the flask and the air purged
by successive evacuation and hydrogen fills. The mixture
was stirred at rt for 24h. Reaction was stopped by filtra-
tion over Celite, which was then rinsed with ethanol.
Preparative TLC (silica gel; CH2Cl2–MeOH; 90:10 v/v)
gave pure 25 (28mg; 90% yield). IR (neat): 2928, 1655,
1
87%). IR (neat): 3309, 1698, 1574, 1508, 1432cmꢁ1; H
NMR (300MHz, CDCl3): 9.54 (br s, 1H), 8.27 (d,
J = 4Hz, 1H), 8.19 (d, J = 8Hz, 1H), 7.66 (dt,
J = 2Hz, J = 9Hz, 1H), 7.28–7.19 (m, 5H), 6.99 (m,
1H), 3.49 (s, 2H), 3.11 (s, 2H), 2.60 (s, 4H), 2.52 (s,
4H); 13C NMR (50MHz, CDCl3): 168.00, 146.50,
136.80, 127.65, 126.80, 125.65, 118.35, 112.40, 61.40,
60.65, 52.10, 51.50; MS (EI): 310 [M]+. HRMS (EI):
Calcd for C18H22N4O: 310.1793, obsd: 310.1790; Anal.
Calcd for C18H22N4O: C, 69.65; H, 7.14; N, 18.05.
Found: C, 69.41; H, 7.15; N, 17.92.
1584, 1466, 1437cmꢁ1 1H NMR (200MHz, CDCl3):
;
8.52 (dd, J = 2Hz, J = 6Hz, 1H), 7.76 (dt, J = 2Hz,
J = 8Hz, 1H), 7.22–7.27 (m, 2H), 4.75 (br s, 1H), 4.30
(t, J = 6Hz, 2H), 2.84 (m, 4H), 2.49 (m, 6H), 2.21 (m,
1H), 1.93–0.84 (m, 10H); 13C NMR (50MHz, CDCl3):
174.90, 154.35, 147.80, 136.65, 120.75, 120.65, 95.45,
54.95, 51.80, 43.85, 43.40, 40.85, 28.00, 24.05; MS
(ES): 317 [M+H]+.
4.2.4. N-[2-(4-Benzyl-piperazin-1-yl)-ethyl]-pyridin-2-yl-
amine (23). LAH (50mg; 1.36mmol; 2equiv) was added
to 2-(4-benzyl-piperazin-1-yl)-N-pyridin-2-yl-acetamide
22 (210mg; 0.68mmol) in THF (25mL). The mixture
was refluxed for 20h under N2. Reaction was stopped
by addition of H2O and NaOH (4M). The mixture was
then refluxed for 10min and extracted with ethyl acetate.
4.2.7.
N-(2-(1-(4-(2-Pyridinyl-N-oxide)piperazinyl)-
ethyl))-N-(2-pyridinyl)cyclohexanecarboxamide NEW-
WAY (5). 2-Chloropyridine-N-oxide 13 (21mg;
0.16mmol) and TEA (33lL; 24mg; 0.24mmol;
1.5equiv) were added to N-(2-(1-piperazinyl)ethyl)-N-
(2-pyridinyl)cyclohexanecarboxamide 25
0.16mmol) in n-BuOH (3mL) plus CH3CN (3mL).
(50mg;

S. Marchais-Oberwinkler et al. / Bioorg. Med. Chem. 13 (2005) 883–893
891
The mixture was heated at 140ꢁC for 26h. Reaction was
then stopped by evaporation of the solvent. Preparative
TLC (four plates, 0.2mm layers; CH2Cl2–MeOH, 90:10
v/v) gave pure 5 (12.6mg; 20%). IR (neat): 1653 (amide),
ammonium formate–TEA (570:430:3 by vol at 6mL/
min and formulated by evaporation of mobile phase
and dissolution is saline for injection.
1584, 1500, 1437, 1242 (N!O) cmꢁ1
;
1H NMR
4.7. Analysis of [carbonyl-11C]3
(300MHz, CDCl3): 8.47 (d, J = 4Hz, 1H), 8.08 (d,
J = 6Hz, 1H), 7.70 (dt, J = 2Hz, J = 7Hz, 1H), 7.21–
7.11 (m, 3H), 6.80–6.74 (m, 2H), 3.93 (t, J = 7Hz,
2H), 3.28 (br s, 4H), 2.62–2.53 (m, 6H), 2.17 (m, 1H),
1.88 (br s, 1H), 1.70–0.79 (m, 9H); 13C NMR
(50MHz, CDCl3): 147.85, 139.15, 136.65, 125.70,
120.65, 116.20, 113.35, 54.60, 51.10, 46.10, 43.50,
40.90, 28.05, 24.10; MS (ES): 410 [M+H]+.
The radiochemical purity of [carbonyl-11C]2 was deter-
mined on a sample of the final formulation with HPLC
on a l-Bondapak-C18 column (300 · 3.9mm o.d.; parti-
cle size, 10lm; waters) eluted with MeCN–0.01M phos-
phoric acid (30:70 v/v) at 3mL/min with eluate
monitored for radioactivity (Beckman b-flow detector)
and absorbance at 270nm. [carbonyl-11C]3 (retention
time, 6.3min) was identified by co-injection with authen-
tic 3.
4.3. In vitro receptor binding assays
Membranes from HeLa cells expressing h5-HT1A recep-
tors are incubated in tris buffer (50mM; pH7.7) with
[3H]5-CT (4.5nM) for 15min at rt in the presence of
the test compound. Nonspecific binding was determined
by adding methiothepine (10lM). Incubation was ter-
minated by rapid filtration through unifilter GF/B filters
using a Tomtec cell harvester. Filters were counted in a
Wallac Scintillation counter and data were processed
using Prism (GraphPad).
4.8. PET experiments with NCA [carbonyl-11C]3 in
cynomolgus monkey
The study was approved by the Animal Ethics Commit-
tee of Northern Stockholm. A male cynomolgus mon-
key (6.53kg) was anaesthetized by repeated
intramuscular injection of ketamine-xyalazine [Keta-
larꢂ; 1–2mg/kg per h)–(Rompunꢂ; 1mg/kg per h)]
and then positioned in a Siemens ECAT EXACT HR
PET camera (resolution: 3.8mm full width half maxi-
mum) so that the transaxial imaging planes of the head
were parallel to the cantomeatal line. The monkey head
was fixed in position during the PET scans.47 Monkey
body temperature was maintained with a thermostati-
cally controlled heating pad. NCA [carbonyl-11C]3
(51MBq) was injected intravenously into the left sural
vein and regional cerebral radioactivity uptake was
measured in 3-D mode for up to 90min and corrected
for physical decay. Data were obtained from 47 sections
with a separation of 3.3mm. The whole brain contour
was defined in situ according to an atlas of cryosected
cynomolgus monkey head.46 Radioactivity was calcu-
lated from the sequence of time frames, corrected for
physical decay and plotted versus time. The radioactiv-
ity concentration in the ROI for the whole brain was
multiplied by the brain volume (estimated to be
ꢂ65mL) to calculate the percent of injected radioactiv-
ity in whole brain, The calculated value for radioactivity
in the brain, as percent of injected dose, was then
divided by the radioactivity injected and multi-
4.4. Efficacy assays
HeLa cells expressing h5-HT1A receptor (Bmax = 50
fmol/mg protein) were washed and then incubated for
20min with the test compound 3–5 at concentrations
up to 10lM and a fixed concentration of 5-HT
(1lM). Incubation was terminated by aspiration and
cAMP was extracted with ice-cold HCl. The amount
of cAMP in each sample was measured in a Flashplate
(RIA-based system). Data were processed using Prism
(GraphPad).
4.5. Lipophilicity
The LogP and LogD values of 1–5 were calculated from
the Pallas 3.0 program (CompuDrug International Inc.,
San Francisco, CA, USA). Lipophilicity was also
assessed with TLC (silica gel; CH2Cl2–MeOH, 9:1 v/v)
for 1, 3, and 4 and also with reverse phase HPLC
analysis on a l-Bondapak C-18 column (300 · 3.9mm
o.d.; particle size 10lm) eluted at mL/min with
MeCN–0.01M H3PO4 (65:35 v/v) at 3mL/min for 1
and 3.
plied by 100. In
a
second experiment, NCA
[carbonyl-11C]3 (51MBq) was injected intravenously at
10min after injection of 1 (0.5mg/kg; i.v.) into the same
monkey.
4.6. Preparation of [carbonyl-11C]3
4.9. Measurement of unchanged [carbonyl-11C]3 and
radioactive metabolites in monkey plasma
NCA [carbonyl-11C]cyclohexanecarbonyl chloride was
produced from cyclotron-produced [11C]carbon dioxide,
exactly as described previously.46,47 The preparation of
[carbonyl-11C]3 followed a procedure analogous to that
used formerly to produce [carbonyl-11C]1 and other ana-
logs.30,31 Thus, the N-oxide 12 (9.6mg; 29lmol) was
treated with NCA [carbonyl-11C]cyclohexanecarbonyl
chloride in tetrahydrofuran (500lL) containing TEA
(40lL) for 4min and then separated by reverse phase
HPLC on a l-Bondapak column (300 · 3.9mm o.d.;
10lm particle size; waters) eluted with MeOH–0.1M
The percentage of radioactivity in plasma as unchanged
radioligand was determined by an HPLC method that
has been shown to be effective for several other PET
radioligands.47,48 After intravenous injection of [carbon-
yl-11C]3 into monkey, arterial blood samples were taken
at 5, 30, and 45min during the PET scan. Each blood
sample was centrifuged at 2000g for 1min. The supern-
atant plasma (0.5mL) was deproteinized with acetonit-
rile (0.7mL) that had been pre-spiked with reference 3.

892
S. Marchais-Oberwinkler et al. / Bioorg. Med. Chem. 13 (2005) 883–893
B.; Wong, B.; Lyle, E. A.; Stranieri, M. T.; Cook, J. J.;
McMasters, D. R.; Pellicore, J. M.; Pal, S.; Wallace, A. A.;
Clayton, F. C.; Bohn, D.; Welsh, D. C.; Lynch, J. J., Jr.;
Yan, Y.; Chen, Z.; Kuo, L.; Gardell, S. J.; Shafer, J. A.;
Vacca, J. P. Bioorg. Med. Chem. Lett. 2003, 13, 1353–
1357.
The radioactivity of this mixture was measured in a well
counter and a portion (1mL) analyzed by injection onto
a gradient HPLC system, a l-Bondapak-C18 column
(300 · 7.8mm o.d.; 10lm particle size) eluted at
6.0mL/min with acetonitrile–0.01M phosphoric
acid.47,48 Over 98% of the radioactivity in the blood
sample was recovered in the deproteinized plasma taken
for analysis.
17. Mallams, A. K.; Rossman, R. R.; Doll, R. J.; Girijaval-
labhan, V. M.; Ganguly, A. K.; Petrin, J.; Wang, L.;
Patton, R.; Bishop, W. R.; Carr, D. M.; Kirschmeier, P.;
Catino, J. J.; Bryant, M. S.; Chen, K.-J.; Korfmacher, W.
A.; Nardo, C.; Wang, S.; Nomeir, A. A.; Lin, C.-C.; Li, Z.;
Chen, J.; Lee, S.; Dell, J.; Lipari, P.; Malkowski, M.;
Yaremko, B.; King, I.; Liu, M. J. Med. Chem. 1998, 41,
877–893.
18. Davidsen, S. K.; Summers, J. B.; Sweeny, D. J.; Holms, J.
H.; Albert, D. H.; Carrera, G. M., Jr. Bioorg. Med. Chem.
Lett. 1995, 5, 2909–2912.
19. Harmat, N. J. S.; Giorgi, R.; Bonaccorsi, F.; Cerbai, G.;
Colombani, S. M.; Renzetti, A. R.; Cirillo, R.; Subissi, A.;
Alagono, G.; Ghio, C.; Arcamone, F.; Giachetti, A.;
Paleari, F.; Salimbeni, A. J. Med. Chem. 1995, 38, 2925–
2937.
Acknowledgements
The authors are grateful to the Human Science Frontier
Program Organization for major grant support for this
project [RG0235/1998-B] and to the COST B3 and
B12 programs for facilitating collaborations.
References and notes
1. Cliffe, I. A. 206th National Meeting of the American
Chemical Society, August 22–27, 1993, Chicago, IL.
Abstr. Medi, 30.
2. Fletcher, A.; Bill, D. J.; Cliffe, I. A.; Forster, E. A.; Reilly,
Y. T. Br. J. Pharmacol. 1994, 112, 91P.
20. Mitrenga, M.; Hartmann, R. W. Eur. J. Med. Chem. 1995,
30, 241–244.
21. Marchais, S.; Nowicki, B.; Wikstro¨m, H. V.; Brennum, L.
T.; Halldin, C.; Pike, V. W. Bioorg. Med. Chem. 2001, 9,
695–702.
3. Brightwell, C. I.; Kelly, M. G. 1995, WO95/33743; PCT/
GB95/01263.
22. Deady, L. W.; Stanborough, M. S. Aust. J. Chem. 1982,
35, 1841–1849.
4. Mensonides-Harsema, M.; Liao, Y.; Bo¨ttcher, H.; Barto-
zyzyk, G. D.; Greiner, H. E.; Harting, J.; de Boer, P.;
Wikstro¨m, H. V. J. Med. Chem. 2000, 43, 432–439.
5. Pike, V. W.; Halldin, C.; Wikstro¨m, H. V. Prog. Med.
Chem. 2001, 38, 189–247, and references cited therein.
6. Passchier, J.; van Waarde, A. Eur. J. Nucl. Med. 2000, 28,
113–129, and references cited therein.
23. Adams, R.; Miyano, S. J. Am. Chem. Soc. 1954, 76, 2785–
2786.
24. Gignarella, G.; Vianello, P.; Berti, F.; Rossoni, G. Eur. J.
Med. Chem. 1996, 31, 359–364.
25. Cugola, A.; Donati, D.; Guarneri, M.; Micheli, F.; Missio,
A.; Pecuiosa, A.; Reggiani, A.; Zanirato, V. Bioorg. Med.
Chem. Lett. 1996, 6, 2749–2754.
7. Pike, V. W.; Halldin, C.; McCarron, J. A.; Lundkvist, C.;
Hirani, E.; Olsson, H.; Hume, S. P.; Karlsson, P.; Osman,
S.; Swahn, C.-G.; Hall, H.; Wikstro¨m, H.; Mensonidas,
M.; Poole, K. G.; Farde, L. Eur. J. Nucl. Med. 1998, 25,
338–346.
26. Zhuang, Z.-P.; Kung, M.-P.; Kung, H. F. J. Med. Chem.
1994, 37, 1406–1407.
27. Sandell, J.; Halldin, C.; Pike, V. W.; Chou, Y.-H.; Varna¨s,
K.; Hall, H.; Marchais, S.; Nowicki, B.; Wikstro¨m, H.;
Swahn, C.-G.; Farde, L. Nucl. Med. Biol. 2001, 28, 177–
185.
28. Sandell, J.; Pike, V. W.; Hall, H.; Marchais, S.; Wikstro¨m,
H. V.; Halldin, C. J. Label. Compd. Radiopharm. 2001,
44(Suppl. 1), 173–175.
29. Pike, V. W. J. Psychopharmacol. 1993, 7, 139–158.
30. Pike, V. W.; McCarron, J. A.; Hume, S. P.; Ashworth, S.;
Opacka-Jufry, J.; Osman, S.; Lammertsma, A. A.; Poole,
K. G.; Fletcher, A.; White, A. C.; Cliffe, I. A. Med. Chem.
Res. 1995, 5, 208–277.
31. Hall, H.; Lundkvist, C.; Halldin, C.; Farde, L.; Pike, V.
W.; McCarron, J. A.; Fletcher, A.; Cliffe, I. A.; Barf, T.;
Wikstro¨m, H.; Sedvall, G. Brain Res. 1997, 745, 96–
108.
32. Farde, L.; Ginovart, N.; Ito, H.; Lundkvist, C.; Pike, V.
W.; McCarron, J. A.; Halldin, C. Psychopharmacology
1997, 133, 196–202.
33. Passchier, J.; van Waarde, A.; Doze, P.; Elsinga, P. H.;
Vaalburg, W. Eur. J. Pharmacol. 2000, 407, 273–
280.
34. Passchier, J. PhD Thesis, University of Groningen (Neth-
erlands), 2001.
´
8. Andree, B.; Olsson, J.; Halldin, C.; Pike, V. W.; Farde, L.
J. Nucl. Med. 2002, 43, 292–303.
´
9. Andree, B.; Halldin, C.; Thorberg, S.-O.; Sandell, J.;
Farde, L. Nucl. Med. Biol. 2000, 27, 515–521.
10. Osman, S.; Lundkvist, C.; Pike, V. W.; Halldin, C.;
McCarron, J. A.; Swahn, C.-G.; Ginovart, N.; Luthra, S.
K.; Bench, C. J.; Grasby, P. M.; Cliffe, I. A.; Fletcher, A.;
Farde, L. Nucl. Med. Biol. 1998, 25, 215–223.
11. Gunn, R. N.; Sargent, P. A.; Bench, C. J.; Rabiner, E. A.;
Osman, S.; Pike, V. W.; Hume, S. P.; Grasby, P. M.;
Lammertsma, A. A. NeuroImage 1998, 8, 426–
440.
12. Farde, L.; Ito, H.; Swahn, C.-G.; Pike, V. W.; Halldin, C.
J. Nucl. Med. 1998, 39, 1965–1971.
13. Parsey, R. V.; Slifstein, M.; Hwang, D.-R.; Abi-Dharg-
ham, A.; Simpson, N.; Mawlawi, O.; Guo, N. N.; van
Heertum, R.; Mann, J. J.; Laruelle, M. J. Cerebr. Blood
Flow Metab. 2000, 20, 1111–1133.
14. Bickel, M. H. Pharmacol. Rev. 1969, 21, 325–355.
15. Bell, I. M.; Erb, J. M.; Freidinger, R. M.; Gallicchio, S.
N.; Guare, J. P.; Guidotti, M. T.; Halpin, R. A.; Hobbs,
D. W.; Homnick, C. F.; Kuo, M. S.; Lis, E. V.; Mathre, D.
J.; Michelson, S. R.; Pawluczyk, J. M.; Pettibone, D. J.;
Reiss, D. R.; Vickers, S.; Williams, P. D.; Woyden, C. J. J.
Med. Chem. 1998, 41, 2146–2163.
16. Burgey, C. S.; Robinson, K. A.; Lyle, T. A.; Natermet, P.
G.; Selnick, H. G.; Isaacs, R. C. A.; Lewis, S. D.; Lucas,
B. J.; Krueger, J. A.; Singh, R.; Miller-Stein, C.; White, R.
35. Bender, D.; Holschbach, M.; Sto¨cklin, G. Nucl. Med. Biol.
1994, 21, 921–925.
36. Baldessarini, R. J.; Centorrino, F.; Flood, J. G.; Volpicelli,
S. A.; Hustonlyons, D.; Cohen, B. M. Neuropsychophar-
macology 1993, 9, 117–124.
37. Fang, J. Eur. J. Drug Metab. Pharmacokinetics 2000, 25,
109–114.

S. Marchais-Oberwinkler et al. / Bioorg. Med. Chem. 13 (2005) 883–893
893
38. Sveigaard, H. H.; Dalgaard, L. Pharm. Res. 2000, 17, 70–
76.
39. Benet, L. Z.; Cummins, C. L. Adv. Drug Delivery Rev.
2001, 50, S3–S11.
40. Ma, Y.; Lang, L.; Kiesewetter, D. O.; Eckelman, W. C. J.
Chromatogr. B 2002, 780, 99–110.
44. Dukat, M.; Abdel-Rahman, A. A.; Ismaiel, A. M.; Ingher,
S.; Teitler, M.; Gyermek, L.; Glennon, R. A. J. Med.
Chem. 1996, 39, 4017–4026.
45. Wolfe, S.; Yang, K.; Weimberg, N.; Shi, Z.; Hsieh, Y.-H.;
Sharma, R. D.; Ro, S.; Kim, C.-K. Chem. Eur. J. 1998, 4,
886–902.
41. Ma, Y.; Lang, L.; Kiesewetter, D. O.; Jagoda, E.;
Sassaman, M. B.; Der, M.; Eckelman, W. C. J. Chroma-
togr. B 2001, 755, 47–56.
46. Karlsson, P.; Farde, L.; Halldin, C.; Swahn, C.-G.;
Sedvall, G.; Foged, C.; Hansen, K. T.; Skrumsager, B.
Psychopharmacology 1993, 113, 149–156.
´
42. Lopez-Rodriguez, M. L.; Ayala, D.; Benhamu, B.; Mor-
cillo, M. J.; Viso, A. Curr. Med. Chem. 2002, 9, 443–
469.
43. Cliffe, I. A.; Mansell, H. L. European Patent 0512 755 A2,
1993.
´
47. Halldin, C.; Swahn, C.-G.; Farde, L.; Sedvall, G. In PET
for Drug Development and Evaluation; Comar, D., Ed.;
Kluwer Academic: Dordrecht; 1995, pp 55–65.
48. Swahn, C.-G.; Halldin, C.; Farde, L.; Sedvall, G. Human
Psychopharmacol. 1994, 9, 25–31.