Novel (bisarylmethoxy)butylpiperidine analogues as neurotransmitter transporter inhibitors with activity at dopamine receptor sites

Article abstract of DOI:10.1016/S0968-0896(02)00348-6

A series of (bisarylmethoxy)butylpiperidine derivatives was prepared and evaluated in vitro and in vivo to determine the structural requirements necessary for dual activity at the DAT and DA/5-HT receptor sites. These hybrid ligands, constructed by combining pharmacophores specific for the DAT and DA/5-HT receptors, could be useful drugs for treating cocaine addiction by assisting cocaine addicts in maintaining abstinence. The series was evaluated in vitro for DAT and DA/5-HT receptor activity and then selected compounds were tested in vivo for their effects on cocaine-induced hyperlocomotor activity (LMA). The majority of the new compounds demonstrated high to moderate affinity (4-191 nM) for the DAT with 4-hydroxy-4-phenylpiperidine analogues 14 and 15 possessing the greatest affinity. Compounds 15 and 22 exhibited the highest ratio of reuptake inhibition-to-binding (discrimination ratio, DR), 111 and 323, respectively. These derivatives had modest affinity and antagonistic activity for dopamine D₂/D₃ receptors. Compounds 9 and 15 (DR=0.9 and 111, respectively) stimulated locomotor activity, whereas the other compounds suppressed this response. All compounds tested except for 17 and 21 attenuated cocaine-induced hyperlocomotion.

Full text of DOI:10.1016/S0968-0896(02)00348-6

Bioorganic & Medicinal Chemistry 10 (2002) 4091–4102
Novel (Bisarylmethoxy)butylpiperidine Analogues as
Neurotransmitter Transporter Inhibitors with Activity at
Dopamine Receptor Sites
Sung-Woon Choi,a David R. Elmaleh,a Robert N. Hanson,b Timothy M. Shoupa
and Alan J. Fischmana,*
aDivision of Nuclear Medicine, Department of Radiology, Massachusetts General Hospital, Boston, MA 02114, USA
bDepartment of Chemistry, Northeastern University, Boston, MA 02115, USA
Received 7 January 2002; accepted 21 June 2002
Abstract—Aseries of (bisarylmethoxy)butylpiperidine derivatives was prepared and evaluated in vitro and in vivo to determine the
structural requirements necessary for dual activity at the DAT and DA/5-HT receptor sites. These hybrid ligands, constructed by
combining pharmacophores specific for the DAT and DA/5-HT receptors, could be useful drugs for treating cocaine addiction by
assisting cocaine addicts in maintaining abstinence. The series was evaluated in vitro for DAT and DA/5-HT receptor activity and
then selected compounds were tested in vivo for their effects on cocaine-induced hyperlocomotor activity (LMA). The majority of
the new compounds demonstrated high to moderate affinity (4–191 nM) for the DAT with 4-hydroxy-4-phenylpiperidine analogues
14 and 15 possessing the greatest affinity. Compounds 15 and 22 exhibited the highest ratio of reuptake inhibition-to-binding (dis-
crimination ratio, DR), 111 and 323, respectively. These derivatives had modest affinity and antagonistic activity for dopamine D2/
D3 receptors. Compounds 9 and 15 (DR=0.9 and 111, respectively) stimulated locomotor activity, whereas the other compounds
suppressed this response. All compounds tested except for 17 and 21 attenuated cocaine-induced hyperlocomotion.
# 2002 Elsevier Science Ltd. All rights reserved.
Introduction
drug development has focused on designing agents that
target the DAT.
Cocaine addiction represents one of the major public
health problems in American society today.1 Efforts
have been directed toward identifying and developing
new therapeutic agents that can mediate cocaine’s acute
behavior and reinforcing effects;2 however, at present
there are no effective long-term pharmacotherapy that
can lessen the abuse of cocaine.
Dopamine agonists and antagonists as medications for
cocaine abuse have been evaluated and extensively
reviewed.5 Although the behavioral effects of cocaine
are not consistently altered by selective dopamine
antagonists,6 many studies have reported their media-
tion of the reinforcing effects of cocaine.7 The dopamine
D3 receptor has been of particular interest because of its
relatively restricted localization within the limbic system
compared with the dopamine D2 receptor and due to its
role as a possible target for treatment of schizophrenia
and drug abuse.8 Arecent report of selective inhibition
of cocaine-seeking behavior by a partial dopamine D3
receptor agonist suggested that this receptor is an
important target for the development of medications for
cocaine abuse.9
Cocaine is pharmacologically a nonselective drug inter-
acting in the central nervous system at the dopamine
transporter (DAT), serotonin transporter (SERT), and
norepinephrine transporter (NET) sites by inhibiting
reuptake of the dopamine (DA), serotonin (5-HT), and
norephrine (NE) neurotransmitters, respectively. Sub-
stantial evidence indicates that the DAT is a critical
recognition site that contributes significantly to the
reinforcing properties of cocaine.3,4 Hence, therapeutic
Comparison of IC50 values for inhibition of [3H]DA
reuptake and Ki values for DAT ligand-binding affinity
(e.g., [125I]RTI-55) has been suggested as a method for
evaluating cocaine antagonists or partial agonists in
*Corresponding author. Tel.: +1-617-726-8353; fax: +1-617-726-
6165; e-mail: fischman@pet.mgh.harvard.edu
0968-0896/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(02)00348-6

4092
S.-W. Choi et al. / Bioorg. Med. Chem. 10 (2002) 4091–4102
vitro.10,11 In this context, important cocaine antagonists
would display high discrimination ratios (ratio of reup-
take inhibition-to-binding, DR) under comparable
experimental conditions.10 Although DR values from in
vitro assays have not always been predictive of in vivo
activities such as locomotor activation or depression,
drug discrimination or substitution for cocaine and
anti-cocaine activity, they can serve as an initial in vitro
screening process.11 While a pure cocaine antagonist at the
DAT may be of interest in preventing the reinforcing effects
of cocaine, a compound with some residual DAreuptake
inhibition may have greater potential for the treatment of
cocaine abuse. Such agents may serve to provide some relief
from cocaine craving and alleviate the withdrawal symp-
toms encountered during initial stages of abstinence.
albeit to a lesser extent compared to cocaine. Importantly,
when the dopamine D2-selective antagonist, spiperon,
was used to pretreat monkeys, the self-administration of
GBR 12909 was attenuated.18 In related studies, the
5-HT agonist, quipazine, reliably attenuated the beha-
vioral-stimulus effects of GBR 12909.20 Therefore,
assuming actions at the DAT are responsible for the
behavioral effects of GBR 12909, it is possible that
combining strong DAT binding properties and dopa-
mine antagonistic and/or 5-HT agonistic properties in
one molecule may offer a more effective treatment with
less abuse liability potential than GBR 12909 and fewer
side-effects observed from dopamine D2, D3 antagonist
monotherapy. Although our explanation is speculative
without previous neurobehavioral data from dual
activity ligands, such compounds should block the
binding of cocaine to the DAT with weak dopamine
reuptake inhibition. Such residual dopamine uptake
inhibiting properties will then be attenuated by activities
at the dopamine receptor site(s) as antagonist and/or 5-
HT agonist. Therefore, abuse liability that may arise
from the activity of dopamine reuptake inhibition can
be minimized.
Various DAT ligands have been proposed as potential
therapeutic agents for cocaine abuse including cocaine
analogues12 and disubstituted piperazine (GBR) analo-
gues.13,14 GBR analogues are particularly interesting
because their pharmacologic properties include slow
dissociation from the DAT, high transporter selectivity,
and low intrinsic activities.15 Oxygenated GBR analogues
have been evaluated as DAT-specific, potentially long
acting, anti-cocaine therapeutics for cocaine addiction.16
Although the behavioral profile of GBR 12909 (1, Figure
1) was shown to be different from that of cocaine, this
compound was reliably self-administered in nonhuman
primates and fully substituted for cocaine in drug dis-
crimination studies.17ꢀ20 Such studies indicate that GBR
12909 might have the potential for abuse in humans,
As part of our program to develop potent DAT inhibi-
tors, we have investigated numerous GBR analogues to
identify structural features that confer high binding
affinity for the DAT but weaker potency for inhibition
21,22
of DAreuptake.
By extending our studies to mea-
surements of DA/5-HT receptor activity, we have
anticipated that some agents might influence the dopa-
Figure 1.

S.-W. Choi et al. / Bioorg. Med. Chem. 10 (2002) 4091–4102
4093
minergic system in novel ways. In the current study, we
used a hybrid drug approach which combines pharma-
cophores specific for the DAT and DA/5-HT receptors.
as base and KI as catalyst. The final products were
converted into oxalate salts for elemental analysis and
biologic evaluation.
Structure–activity relationship (SAR) studies have
demonstrated that the (bisarylmethoxy)ethyl and diamine
moieties of GBR (1) are essential pharmacophores for
retaining favorable activities at the DAT.13,14,22 Also, at
least one nitrogen atom (2) attached to the benzyl moiety
is required.14,23 Previously, the rigid piperazine ring of
GBR (1) had been substituted with conformationally
flexible polymethylene diamines and amine-amide moi-
eties (3) without compromising biologic activities.22
Tolerance of the amide function at the N2 nitrogen
atom was interpreted as due to the possible presence of
strong hydrogen-bonding capabilities at binding site(s)
on the DAT protein. These acyclic analogues have pro-
vided a basis for the design of novel probes that
demonstrate selective and high affinity binding to the
DAT and weak potency for DA reuptake inhibition. In
order to develop compounds with a mixed pharmaco-
logic profile containing both DAT activity (binding/
reuptake inhibition) and dopamine antagonistic/5-HT
agonistic properties, we incorporated structural moieties
believed to induce dopamine D2/D3 antagonism such as
4-aryl-4-piperidinol found in bromperidol and haloper-
idol and 5-HT activity found in arylpiperazine.24,25
Compound selection was also based on H-bonding and
aryl functionality as elucidated in our previous findings
with amine–amide analogues.26 Ahmad et al.27 reported
benzhydroxyalkylpiperidine derivatives that were non-
selective reuptake inhibitors of DA, 5-HT, and nor-
epenephrine. As shown in the general structure 4,
several important aspects of SAR were identified,
including distance between the diaryl and piperidine
moieties (n, distance as shown by the bold line in 1–3)
and the nature of the substituents (R1–R4) on the
piperidine ring. Aspecific three-dimensional arrangement
of these groups (distance, basic nitrogen, hydrogen bond-
ing, and Ar function) was deduced to test the hypothesis
concerning which chemical groups (H-bonding, Ar
function, and distance) are important for a particular
biologic activity and what is the transporter-bound
conformation of these groups. The results of our work,
together with that of other investigators, led to the pre-
paration of several arylpiperidine derivatives. In this
study, we report the preparation and pharmacologic
evaluation of a series of (bisarylmethoxy)butylpiper-
idine derivatives having mixed activities at the DAT and
DA/5-HT receptor sites.
In vitro characterization
This research was designed to develop new agents with
higher DAT affinity and low potency for DA reuptake
inhibition by selective modification of the substituents
on a piperidine ring template structure (4). In order to
find new lead compound(s) for potential cocaine
antagonists, we focused on the introduction of new
moieties for hydrogen bonding and aromatic function.
Such modifications are presumed to be critical struc-
tural determinants for dual activity at DAT and DA/
5-HT receptor sites. Anew series of (bisarylmethoxy)
alkylpiperidine analogues was prepared and character-
ized. The analogues were evaluated for their ability to
displace [125I]RTI-55, a cocaine analogue (Ki values in
Table 1) as well as for potency of reuptake inhibition of
DA, 5-HT and NE (IC50 values in Table 2) with HEK
293 cells expressing cDNAfor human dopamine trans-
porter (hDAT), human serotonin transporter (hSERT)
or human norepinephrine transporter (hNET). In addi-
tion, some compounds were evaluated for their ability
to bind and function at DAand 5-HT receptor sites
(Table 3). Our initial targets were dopamine D2 and D3
receptors since the target compounds contain pharma-
cophores for dopamine receptors as well as 5-HT
receptors as non-selective antagonists (mostly dopamine
D2 and D3). However, it seems reasonable to test all
dopamine receptor subtypes to get a wider view of ligand
effects on dopaminergic and serotonergic systems, espe-
cially regarding the interpretation of locomotor activities.
The final products were submitted to the National Insti-
tute on Drug Abuse, Cocaine Treatment Development
Program (CTDP) for pharmacologic screening (Table 4).
Neurotransporter activities28
Substitution on the piperidine ring by –OH and –CONH2
groups (9–11) without an aromatic group produced
moderate binding activity (Ki=91–114 nM) with low
selectivity at the DAT (DR=0.6–1.3). The introduction
of an acetyl group and aromatic function at the 4-posi-
tion of the piperidine ring (13) resulted in a compound
that retained significant binding activity at the DAT
with a DR of 48. Substitution of –COCH3 with –CN
(12, non-H-bonding group) decreased activities drama-
tically. Further modification of the acetyl group to a
hydroxyl group (14) caused a marked increase in DAT
affinity with reduced selectivity at the SERT (6–3) and
decreased DR (48–15). When a large halogen atom (Br,
15) was introduced at the 4-position of the aromatic
ring, selectivity and potency of binding at the DAT were
improved (15 vs 14) and a significant increase in DR
(15–111) was achieved. Achange in alkyl chain length
(n=4–2, 15 vs 16) decreased activities at the DAT
(Ki=4–191 nM, DR=111–37). Substitution of the
bisarylmethoxy group with the phthalimide group (24 vs
14) caused loss of activities in some cases. Introduction
of the –CF3 group at the 3-position of the aromatic
group (14–17) resulted in weakly selective activities at
Results and Discussion
Chemistry
The preparation of the GBR analogues is depicted in
Scheme 1. The requisite starting materials 6 and 722
were prepared according to literature procedures with
or without modifications, and 8 was purchased. The
N-alkylated target compounds were prepared in good
yields (73–87%) by alkylation of the appropriate amine
with alkylating agents (6-–8) in the presence of K2CO3

4094
S.-W. Choi et al. / Bioorg. Med. Chem. 10 (2002) 4091–4102
Scheme 1. (a) concd H2SO4, PhCH3; (b) K2CO3, KI, DMF, 60–70 ꢁC.
DA/5-HT receptor activities29
the NET site (Ki=10 nM, IC50=780 nM). When an
aryl group was attached to the piperidine by a CH2
linkage (benzyl group, 18), selective binding at the DAT
was maintained but the selectivity of reuptake inhibition
was shifted to the NET site from the DAT site (14, 15 vs
18). When the aromatic group was attached directly to
the piperidine ring (19, 20) without an H-bonding
group, the binding was strongest at the SERT site
(Ki,=18 and 25 nM, respectively) while selective reup-
take inhibition of NE was maintained. Additional aro-
matic groups on the piperidine ring (20–21) were
detrimental and resulted in loss of activity at the DAT.
Compounds 22 and 23 contained both H-bonding (as
amide) and Ar groups but the activities at the transpor-
ters were different for each compound. Compound 22
represents the strongest binding ligand at the DAT
(Ki=1 nM) and has the highest DR value (323). From
this limited data set, it is apparent that the potency in
the DAT binding assay and reuptake inhibition are rela-
tively dependent on substitution and position on the
piperidine ring.
Compounds 14–17, 22 and 23 were evaluated for their
ability to bind and function at DAand 5-HT receptor
sites which is based on the fact that 4-hydroxy-4-arylpiper-
idines are the basic pharmacophore for activity at DA
receptor sites. The potency and selectivity of the selected
target compounds for DAand 5-HT receptors were
evaluated using ligand displacement assays (see Table 3
for radioligands and cell types used). D2 and D3 antag-
onistic activities were assessed in vitro by the ability of
the compounds to block [3H]thymidine incorporation
(inhibition of mitogenesis) induced by 10 nM quinpirole
in CHOp-D2 and D3 cells (Table 3). Most compounds
evaluated had weak to moderate affinity and antag-
onistic activity at D2 and/or D3 sites. (Bisarylmethoxy)-
butylpiperidines (4) are nonselective neurotransmitter
inhibitors,27 and bromperidol (5) is a potent nonselective
dopamine D2/D3 receptor antagonist.30 The combination
of both structural elements in 15 provided a compound
that was both a very potent DAT binder (DR value of
111) and a moderate dopamine D2/D3 receptor antagonist.

S.-W. Choi et al. / Bioorg. Med. Chem. 10 (2002) 4091–4102
4095
Table 1. Binding affinities and selectivities of bisarylmethoxybutylpiperidine derivatives at the DA, 5-HT, and NE transporters labeled with
125I]RTI-55
[
Compound
Binding (KiꢂSD [nM])
Ratios
DAT
SERT
NET
SERT/DAT
NET/DAT
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
114ꢂ27a
91ꢂ26
423ꢂ98a
437ꢂ66
132ꢂ26
311ꢂ58
126ꢂ45
17.9 ꢂ8.8
38ꢂ10
610ꢂ130a
611 ꢂ76
578ꢂ52
3.7
4.8
1.4
2.3
5.8
2.8
9.7
3.1
3.6
1.5
0.3
0.4
8
5.4
6.7
6.1
2.7
5.3
8.4
64.1
6.5
0.6
4.3
3.9
5.1
12.3
16.8
2.9
1
94ꢂ42
135ꢂ34
21.7ꢂ8.0
6.4ꢂ1.2
3.9ꢂ1.8
191ꢂ78
15.8ꢂ6.1
41ꢂ12
368 ꢂ14
114ꢂ4.1
54ꢂ16
250ꢂ110
1240ꢂ310
10ꢂ1.9
590ꢂ180
57ꢂ16
62ꢂ27
175ꢂ47
66ꢂ20
18.4ꢂ4.1
25.3ꢂ7.2
3000ꢂ1100
3.8ꢂ1.9
192ꢂ95
1590ꢂ43
186ꢂ30
419ꢂ12
343ꢂ31
402ꢂ62
260ꢂ9.4
217ꢂ23
496ꢂ37
260ꢂ41
73ꢂ33
371ꢂ35
375ꢂ90
1.32ꢂ0.49
125ꢂ22
>10 mM
27ꢂ8
4600ꢂ1800
22.2ꢂ3.6
356ꢂ36
2.9
1.5
<0.2
6.9
>10 mM
163ꢂ39
GBR 12909
Cocaineb
Cocainec
Cocained
Cocainee
Cocainef
Cocaineg
6.0
915ꢂ214
258ꢂ23
573ꢂ54
350ꢂ45
271ꢂ65
621ꢂ45
600ꢂ216
1740ꢂ180
2040ꢂ240
1610ꢂ300
1730ꢂ280
1400ꢂ320
aResults are averageꢂSEM of three independent experiments assayed in triplicate.
bCocaine as reference for GBR 12909.
cCocaine as reference for 9, 10, 14, 16, 22.
dCocaine as reference for 11, 24.
eCocaine as reference for 12, 17, 19.
fCocaine as reference for 13–15.
gCocaine as reference for 18, 20, 21, 23.
Table 2. DA, 5-HT and NE reuptake inhibition and ratios of reuptake inhibition to binding of (bisarylmethoxy)butylpiperidine/piperazine deri-
vatives at the DAtransporters
Compound
Reuptake inhibition (IC50ꢂSD [nM])
Discrimination ratios
[3H]DA[
3H]5-HT
[3H]NE
[3H]DAreuptake
inhibition/DAT binding
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
101ꢂ11a
114ꢂ28
56ꢂ16
314ꢂ71a
860ꢂ180
203ꢂ78
143ꢂ50a
310ꢂ110
158ꢂ60
393ꢂ30
2250ꢂ950
179ꢂ47
1870ꢂ670
>10 mM
780ꢂ310
30.7ꢂ8.3
1410ꢂ300
159ꢂ21
>10 mM
132ꢂ11
64ꢂ19
0.9
1.3
0.6
3.1
47.5
15.0
110.5
37.1
63.3
4.5
26.5
6.0
8.3
323
27.9
1
415ꢂ87
1030ꢂ330
96ꢂ11
2030ꢂ190
>10 mM
136ꢂ41
431ꢂ88
7080ꢂ510
1000ꢂ160
184ꢂ47
1750ꢂ730
440ꢂ120
3100ꢂ1400
426ꢂ54
3490ꢂ220
>10 mM
246ꢂ142
152ꢂ24
276ꢂ22
237ꢂ41
417ꢂ35
278ꢂ53
3070ꢂ830
>10 mM
3800ꢂ1300
570ꢂ110
2100ꢂ400
1700ꢂ390
3070ꢂ740
570ꢂ210
111ꢂ31
361ꢂ68
>10 mM
532ꢂ183
445ꢂ205
264ꢂ57
190ꢂ38
239ꢂ26
209ꢂ36
GBR 12909
Cocaineb
Cocainec
Cocained
Cocainee
Cocainef
584ꢂ237
425ꢂ114
301ꢂ53
9.1
348ꢂ66
405ꢂ39
189ꢂ31
aResults are averageꢂSEM of three independent experiments assayed in triplicate.
bCocaine as reference for GBR 12909.
cCocaine as reference for 9, 10, 14, 16, 22.
dCocaine as reference for 11, 24.
eCocaine as reference for 12, 17, 19.
fCocaine as reference for 13–15.

4096
S.-W. Choi et al. / Bioorg. Med. Chem. 10 (2002) 4091–4102
Table 3. Receptor binding profile and effects of selected target compounds 14–17, 22, and 23
Receptor
Affinity (KiꢂSD [nM])a
Antagonist activity (IC50 [nM])b
14
15
16
17
22
23
D1
D2
2460ꢂ38.1c
22.6ꢂ0.5a
136ꢂ1.3b
239ꢂ121a
52.0ꢂ4.8b
1400ꢂ247
566ꢂ45.5
5870ꢂ4140
2470ꢂ46.6
3050ꢂ1440
27.6ꢂ0.8
172ꢂ10.3
51.5ꢂ14.6
42.3ꢂ13.6
143ꢂ5.1
174ꢂ50.2
>10 mM
639ꢂ107
4300ꢂ411
ntd
1570ꢂ523
37.4ꢂ5.0
70.1ꢂ13.0
65.2ꢂ20.9
85.9ꢂ30.6
1130ꢂ37.5
621ꢂ83.3
6390ꢂ1760
nt
1640ꢂ527
17.0ꢂ6.17
48.7ꢂ2.67
15.8ꢂ1.08
87.4ꢂ17.0
857ꢂ123
580ꢂ58.1
3100ꢂ235
nt
1890ꢂ763
18.8ꢂ4.32
67.8ꢂ1.22
33.0ꢂ13.8
77.3ꢂ6.26
682ꢂ77.7
241ꢂ64.7
>10 mM
nt
90.6ꢂ40.3
56.8ꢂ34.1
91.3ꢂ28.8
2870ꢂ693
580ꢂ44.4
5790ꢂ419
5270ꢂ594
D3
5-HT1A
5-HT2A
5-HT2C
5-HT3
aBinding assay receptors/radioligands: D1 (human cloned receptors in LHD1 cells/[3H]SCH 23,390); D2 and D3 (human cloned receptors in CHOp
cells/[3H]YM-09151–2); 5-HT1A (human cloned receptors in HA7 cells/[3H]8-OH-DPAT); 5-HT2A (rat receptors in NIH-3T3-GF6 cells/[3H]ketan-
serin); 5-HT2C (rat receptors in NIH-3T3-Pf cells/[3H]mesulergine); 5-HT3 (rat/mouse hybrid receptors in NG108–15 cells/[3H]GR65630).
bD2, D3 antagonist assays: CHOp cells (human receptor), [3H]thymidine incorporation (inhibition of mitogenesis), quinpirole as internal standard
(ED50 range 6.5–57 nM for D2 receptor, ED50 range 2.8–25 nM for D3 receptor).
cAll values represent the mean of at least two determinations.
dnt=Not tested.
Thus, the 4-hydroxy-4-phenylpiperidine moiety pro-
vides additional binding to the DAT. Analogues 14–16
resemble the antipsychotic bromperidol and were found
to be moderate affinity DAantagonists. Addition of
aryl and hydroxyl groups at the 4-position of the piper-
idine ring provided a desirable combination of DAT
and DAreceptor activity. The 4-aryl-4-piperidinol
analogues 14 and 15 demonstrated very weak binding at
the D1 and 5-HT receptor sites. Analogues 17, 22, and
23, each containing a partial structure of a DAantagonist
(17, trifluperidol; 22, pimozide; 23, spiperone), also
showed moderate activity at the D2/D3 receptor sites.
Most compounds had selective and weak activity at the
5-HT2A receptor (Ki=241–639 nM). As expected, the
bulky bisaryl group was not well tolerated in terms of
DAactivities.
both DAT and DA receptor sites. Maximal stimulatory
effect on spontaneous LMAby 9 and 15 is only 40–80%
of that achieved with cocaine as shown by the ratio of
ME/CME in Table 4. Attenuation of cocaine-induced
LMA(20 mg/kg cocaine) was demonstrated by most of
the compounds. Only compounds 17 and 21 failed to
attenuate the hyperlocomotion induced by cocaine.
Compound 24 showed very weak receptor activities
(2.4 mM for D2, 1.8 mM for D3, data not shown) and
weak 5-HT reuptake inhibition (361 nM, Table 2). No
activity at the DAT and NET sites was identified;
however, hyper LMAcaused by cocaine was atte-
nuated. Correlations between DR values and ED50 or
AD50 were not found, although other factors such as
potency as D2/D3 antagonists and pharmacokinetics
(uptake rate into brain) might be critical for the
potency in LMAstudies.
In vivo locomotor activity
Based on their in vitro activities at the DAT, dopamine
D2/D3 receptors and DR values, compounds 15–17
(high DR) and 9, 18, 21 (low DR) were selected for
preliminary behavioral screening which involved testing
alone and in combination with cocaine for their effect
on locomotor activity (LMA) in mice. Likewise, the
effects of the dopaminergic antagonists (14–17, and 23)
on cocaine-induced hyperactivity were studied. Com-
pound 22 was not studied even though it showed a high
DR value at the DAT and SERT because NE reuptake
inhibition was strong.31 As summarized in Table 4,
compounds 9 and 15 produced a significant increase in
Table 4. Effect of compounds 9, 13, 15–18, 21, 23 and 24 on loco-
motor activity in mouse
Compound
ED50 (mg/kg)a
ID50 (mg/kg)b
ME/CMEc
AD50
(mg/kg)d
9
13
15
16
17
18
21
23
6.6/Sa
170/Ib
0.78
0.38
77.6
109
15.8
2.42/Sa,e
14.1/Ib
95.7/Ib
50.7/Ib
132/Ib
13.8
No attenuation
16.4
No attenuation
45.7
40.1
49.2/Ib
82.8/Ib
6.24/Sa
spontaneous LMAwith low ED compared with GBR
50
24
GBR 12909
12909. Significant suppression of spontaneous LMA
was demonstrated by all the compounds except 9 and
15. Compound 15 produced a biphasic effect on LMA;
activity was increased with low doses and decreased
with high doses. Biphasic effects on spontaneous LMA
by dopamine D2/D3 dopaminergic antagonistic ligands
have been previously demonstrated with other com-
pounds.32 Since this ligand possesses activities at the
DAT and is a moderate dopamine D2/D3 antagonist, its
effects on LMAare probably the result of activities at
1.22
Biphasicf
aED50=dose producing 1/2 maximal stimulant activity, S=stimula-
tion.
bID50=dose-producing 1/2 maximal depressant activity, where max-
imal depression=0 counts/30 min, I=inhibition.
cMaximal effect (ME)/cocaine maximal effect (CME).
dAD50=dose attenuating cocaine-induced stimulation by 50%,
ligand–cocaine interaction study.
eLigand study, see text for details; in general low doses produce sti-
mulation and high doses inhibition.
f10 mg/kg enhanced and 50 mg/kg depressed 30–60 min postinjection.

S.-W. Choi et al. / Bioorg. Med. Chem. 10 (2002) 4091–4102
4097
It must be noted that ID50 is similar to ED50 (LMA
increase or decrease) and a high ID50 value means low
intrinsic activities of LMAinhibition for the compound
alone, whether as a weak DAreceptor antagonist or
with other unknown activities. But if a compound also
has a low AD50, this means a strong counter action
against cocaine at low dosage. Such activities can be
explained with other in vitro activities and with other
pharmacophysical properties of the ligand. High ID50
and low AD50 values (discrimination and self-adminis-
tration) are currently important criteria for the selection
of ligands for in vivo testing by NIDA.
or alumina precoated plates (Whatman, AL SIL G/UV or
J.T. Baker, Baker-flex, SILICAGEL IB-F) containing
fluorescent indicator (2ꢃ8 cm). Column chromato-
graphy was performed on silica gel (Baker, 40 mm Flash
chromatography). Fractions were analyzed using TLC
and compounds were visualized with ninhydrin (0.5 g/
100 mL methanol) or iodine vapor for primary and
secondary amine(s). Free bases were dissolved in ethyl
acetate and/or diethyl ether, filtered, and precipitated by
addition of a solution of oxalic acid. The resulting solids
were collected by filtration and recrystallized.
Synthetic procedures
Biphasic effects on spontaneous LMAby D /D3 dopa-
2
minergic antagonistic ligands have been previously
demonstrated with other compounds. Activities at the
D2/D3 autoreceptors might explain some of possible
underlying mechanisms of biphasic activities. Since
these ligands possess activities at the DAT and are
moderate inhibitors of DAreuptake, their effects on
LMAare probably the result of activities at both DAT
and DAreceptor sites. The biphasic activity as stimu-
lant or inhibitant cannot be explained with only in vitro
receptor activities and DAT activities because the other
sites also can modulate dopaminergic activities. However,
the time course LMAstudies will clearly demonstrate
whether such compounds have beneficial long-term
effects or not which also can explain pharmacokinetic
factors.
1-[bis(4-Fluorophenyl)methoxy]-4-chlorobutane (6).
A
mixture of 4-chloro-1-butanol (6.51 g, 60 mmol), 1 mL
concentrated sulfuric acid, and 4,40-difluorobenzhydrol
(2.2 g, 10 mmol) in 50 mL toluene was heated at reflux for
12 h. The reaction mixture was cooled, washed succes-
sively with saturated sodium bicarbonate solution (50
mL) and water (50 mL). The organic layer was dried
over magnesium sulfate, filtered, and the solvent was
removed by rotary evaporation. The resultant brown oil
was purified on a silica gel column. Elution with 2%
ethyl acetate–hexane afforded 2.33 g (7.5 mmol, 75%)
1
of the product as a colorless oil. H NMR (CDCl3): d
1.71–1.80 (2H, m), 1.84–1.93 (2H, m), 3.43 (2H, t,
J=6.0), 3.52 (2H, t, J=6.4), 5.30 (1H, s) 6.95–7.02 (4H,
m), 7.24–7.29 (4H, m).
1-[bis(4-Fluorophenyl)methoxy]-2-bromoethane (7). As
described for 6, 7 was prepared from 2-bromo-1-ethanol
and 4,40-difluorobenzhydrol.
Conclusions
The introduction of hydroxyl and/or aromatic sub-
stituents to the piperidine ring of template structure 4
produced retention of transporter affinity and DA
reuptake inhibition. Such modification resulted in DA
receptor affinity and selectivity when compared with
5-HT receptors. These data give insight into possible
structural similarities between ligand-binding sites on
the DAT and DA receptors. The current studies have
provided additional information on pharmacophores
that are responsible for selective binding and potency of
reuptake inhibition at the DAT and on activities at the
DA/5-HT receptor sites. Although correlation between
in vitro DR values and in vivo potency in LMAactivities
was not observed, these new analogues inconjuction
with the results of the LMAstudies may provide a basis
for the design of other ligands with strong binding to
the DAT, low potency inhibition of DA reuptake, and
binding to DA/5-HT receptor sites.
N-(4-Bromobutyl)-phthalimide (8). This compound was
purchased from Sigma Chemical Company.
1-[4-[bis(4-Fluorophenyl)methoxy]butyl]-4-hydroxypiper-
idine (9). 4-Hydroxypiperidine (1.2 g, 5 mmol) was dis-
solved in 20 mL anhydrous DMF and stirred with
K2CO3 powder (2.76 g, 20 mmol) and KI (100 mg, 0.6
mmol) for 0.5 h. To this turbid solution, 6 (1.55 g, 5
mmol) in 10 mL DMF was added dropwise. The reaction
mixture was stirred for 72 h at 60–70 ꢁC. The turbid
reaction mixture was poured into 200 mL ethyl acetate,
washed with saturated sodium chloride solution (60
mLꢃ3), dried over magnesium sulfate, and evaporated
to dryness. The crude oil was applied to a silica gel column
for purification. Elution with CHCl3/MeOH (98:2)
afforded the desired alkylated product 9 as a colorless
oil (87% yield). 1H NMR (CDCl3): d 1.54–1.66 (6H, m),
1.86–1.89 (2H, m), 2.09 (2H, t, J=10.0), 2.31 (3H, t,
J=7.5 overlap with br s), 2.74–2.76 (2H, m), 3.42 (2H,
t, J=6.3), 3.66 (1H, heptet, J=4.5), 5.28 (1H, s), 6.98–
7.02 (4H, m), 7.25–7.29 (4H, m); mp (oxalate salt): 117–
120 ꢁC.
Experimental
Melting points were determined with a Thomas-Hoover
melting point apparatus and are uncorrected. Elemental
analyses, performed by Atlantic Microlabs, Atlanta,
1-[4-[bis(4-Fluorophenyl)methoxy]butyl]-3-hydroxypiper-
idine (10). As described for 9, 10 was prepared from 6
and 3-hydroxypiperidine as a colorless oil (87% yield).
1H NMR (CDCl3): d 1.48–1.67 (7H, m), 1.76–1.79 (1H,
m), 2.26 (1H, br s), 2.33 (2H, t, J=7.3), 2.41–2.45 (4H,
m), 3.42 (2H, t, J=6.3), 3.80 (1H, br s), 5.28 (1H, s),
1
GA, were within 0.4% of calculated values. H NMR
spectra were recorded on a Varian XL-500 spectro-
meter. In general, CDCl3 was used for the free bases
and DMSO-d6 for salts. Thin layer chromatography
(TLC) was performed on 250-mm-thick silica gel plates

4098
S.-W. Choi et al. / Bioorg. Med. Chem. 10 (2002) 4091–4102
1
6.99–7.03 (4H, m), 7.26–7.29 (4H, m); mp (oxalate salt):
106–109 ꢁC.
4-piperidinol as a colorless oil (81% yield). H NMR
(CDCl3): d 1.56–1.67 (4H, m), 1.71 (2H, d, J=12.0),
2.13 (2H, t t, J=13.1, 4.0), 2.36–2.45 (4H, m), 2.79 (2H,
br d, J=11.0), 3.42 (2H, t, J=6.3), 5.29 (1H, s), 6.97–
7.01 (4H, m), 7.26–7.30 (4H, m), 7.42 (1H, t, J=7.8),
7.48 (1H, d, J=7.0), 7.67 (1H, d, J=8.0), 7.83 (1H, s);
mp (oxalate salt): 148–150 ꢁC.
1-[4-[bis(4-Fluorophenyl)methoxy]butyl]-3-piperidine carb-
oxamide (11). As described for compound 9, 11 was
prepared from 6 and nipecotamide (Sigma Chemical
Company) as a colorless oil (78% yield). 1H NMR
(CDCl3): d 1.48–1.67 (7H, m), 1.76–1.79 (1H, m), 2.26
(1H, br s), 2.33 (2H, t, J=7.3), 2.41–2.45 (4H, m), 3.42
(2H, t, J=6.3), 3.80 (1H, br s), 5.28 (1H, s), 6.99–7.03 (4H,
m), 7.26–7.29 (4H, m); mp (oxalate salt): 143–146 ꢁC.
4-Benzyl-1-[4-[bis(4-fluorophenyl)methoxy]butyl]-4-piper-
idinol (18). As described for 9, 18 was prepared from 6
and 4-benzyl-4-hydroxypiperidine as a colorless oil
1
(73% yield). H NMR (CDCl3): d 1.21 (1H, br s), 1.51
1-[4-[bis(4-Fluorophenyl)methoxy]butyl]-4-phenylpiper-
idine-4-carbonitrile (12). As described for compound 9,
12 was prepared from 6 and 4-cyano-4-phenylpiperidine
(2H, br d, J=12.5), 1.56–1.66 (4H, m), 1.74 (2H, t d,
J=12.9, 3.8), 2.26 (2H, br t, J=11.0), 2.35 (2H, t,
J=7.3), 2.65 (2H, br d, J=11.5), 2.75 (2H, s), 3.42 (2H, t,
J=5.8), 5.28 (1H, s), 6.97–7.02 (4H, m), 7.19–7.21 (2H,
m), 7.23–7.32 (7H, m); mp (oxalate salt): 136–138 ꢁC.
1
hydrochloride as a colorless oil (83% yield). H NMR
(CDCl3): d 1.60–1.70 (4H, m), 2.05–2.12 (4H, m), 2.42–
2.48 (4H, m), 2.99 (2H, br d, J=12.5), 3.44 (2H, t,
J=6.5), 5.29 (1H, s), 6.96–7.00 (4H, m), 7.27–7.30 (5H,
m), 7.35–7.38 (2H, m), 7.48–7.50 (2H, m); mp (oxalate
salt): 176–178 ꢁC.
2-[4-[bis(4-Fluorophenyl)methoxy]butyl]-1,2,3,4-tetrahydro-
isoquinoline (19). As described for 9, 19 was prepared
from 8 and 1,2,3,4-tetrahydroisoquinoline as a slightly
1
yellow oil (79% yield). H NMR (CDCl3): d 1.68–1.71
1-[4-[bis(4-Fluorophenyl)methoxy]butyl]-4-acetyl-4-phenyl-
piperidine (13). As described for 9, 13 was prepared from
6 and 4-acetyl-4-phenylpiperidine hydrochloride as a
colorless oil (73% yield). 1H NMR (CDCl3): d 1.56–1.68
(4H, m), 1.90 (3H, s), 2.06 (2H, t d, J=11.8, 2.7), 2.21
(2H, br t, J=10.8), 2.30 (2H, t, J=7.1), 2.46 (2H, br d,
J=11.7), 2.71 (2H, br d, J=10.5), 3.42 (2H, t, J=5.9),
5.28 (1H, s), 6.96–7.02 (4H, m), 7.25–7.37 (9H, m); mp
(oxalate salt): 138–140 ꢁC.
(4H, m), 2.50 (2H, br t, J=7.0), 2.69 (2H, t, J=5.8),
2.88 (2H, t, J=5.8), 3.45 (2H, br t, J=5.5), 3.60 (2H, s),
5.28 (1H, s), 6.97–7.00 (5H, m), 7.06–7.11 (3H, m),
7.26–7.28 (4H, m); mp (oxalate salt): 135–137 ꢁC.
2-[4-[bis(4-Fluorophenyl)methoxy]butyl]-6,7-dimethoxy-
1,2,3,4-tetrahydroisoquinoline (20). As described for 9,
20 was prepared from 8 and 6,7-dimethoxy-1,2,3,4-tetra-
hydroisoquinoline hydrochloride as a colorless oil (81%
yield). 1H NMR (CDCl3): d 1.69–1.71 (4H, m), 2.50 (2H,
br t, J=6.8), 2.69 (2H, t, J=5.8), 2.81 (2H, t, J=5.8), 3.45
(2H, t, J=5.8), 3.53 (2H, s), 3.83 (3H, s), 3.84 (3H, s), 5.30
(1H, s), 6.50 (1H, s), 6.59 (1H, s), 6.98–7.02 (4H, m), 7.26–
7.30 (4H, m); mp (oxalate salt): 88–102ꢁC.
1-[4-[bis(4-Fluorophenyl)methoxy]butyl]-4-phenylpiper-
idin-4-ol (14). As described for 9, 14 was prepared from
6 and 4-phenyl-4-piperidinol as a colorless oil (79%
yield). 1H NMR (CDCl3): d 1.58–1.73 (6H, m), 2.13
(2H, t d, J=13.0, 4.5), 2.37 (2H, t, J=7.2), 2.43 (2H, br
d, J=11.8), 2.76 (2H, br d, J=11.1), 3.42 (2H, t,
J=5.6), 5.28 (1H, s), 6.96–7.03 (4H, m), 7.22–7.34 (7H,
m), 7.49 (2H, d, J=7.9); mp (oxalate salt): 168–170 ꢁC.
2-[4-[bis(4-Fluorophenyl)methoxy]butyl]-6,7-dimethoxy-
1-phenyl-1,2,3,4-tetrahydroisoquinoline (21). As described
for 9, 21 was prepared from 6 and 6,7-dimethoxy-1-
phenyl-1,2,3,4-tetrahydroisoquinoline hydrochloride as a
colorless oil (75% yield). 1H NMR (CDCl3): d 1.48–1.63
(4H, m), 2.28–2.33 (1H, m), 2.49–2.56 (2H, m), 2.75 (1H, d
t, J=16.5, 4.7), 2.93–2.99 (1H, m), 3.14 (1H, d t, J=12.0,
5.3), 3.27–3.34 (2H, m), 3.58 (3H, s), 3.84 (3H, s), 4.46
(1H, s), 5.20 (1H, s), 6.16 (1H, s), 6.60 (1H, s), 6.96–7.00
(4H, m), 7.20–7.27 (9H, m); mp (oxalate salt): 115–118 ꢁC.
1-[4-[bis(4-Fluorophenyl)methoxy]butyl]-4-(4-bromophenyl)-
4-piperidinol (15). As described for 9, 15 was prepared
from 6 and 4-(4-bromophenyl)-4-piperidinol as a colorless
oil (80% yield). 1H NMR (CDCl3): d 1.60–1.73 (6H, m),
2.12 (2H, t d, J=12.8, 2.9), 2.39–2.45 (4H, m), 2.80 (2H,
br d, J=10.9), 3.43 (2H, t, J=5.4), 5.29 (1H, s), 6.97–
7.03 (4H, m), 7.26–7.34 (4H, m), 7.37 (2H, d, J=8.7),
7.46 (2H, d, J=8.4); mp (oxalate salt): 156–158 ꢁC.
1-[1-[4-[bis(4-Fluorophenyl)methoxy]butyl]piperidin-4-yl]-
1,3-dihydrobenzoimidazol-2-one (22). As described for 9, 22
was prepared from 6 and 4-(2-keto-1-benzimidazolinyl)-
piperidine as a colorless oil (80% yield). 1H NMR
(CDCl3): d 1.64–1.70 (4H, m), 1.82 (2H, br d, J=10.0),
2.13 (2H, t, J=11.5), 2.42 (2H, t, J=7.0), 2.50 (2H, q d,
J=12.7, 3.4), 3.10 (2H, br d, J=11.0), 3.45 (2H, t,
J=5.8), 4.40 (1H, t t, J=12.3, 4.0), 5.30 (1H, s), 6.98–
7.04 (6H, m), 7.12–7.13 (1H, m), 7.53–7.30 (5H, m),
11.09 (1H, br s); mp (oxalate salt): 225–227 ꢁC.
1-[4-[bis(4-Fluorophenyl)methoxy]ethyl]-4-(4-bromophenyl)-
4-piperidinol (16). As described for 9, 16 was prepared
from 7 and 4-(4-bromophenyl)-4-piperidinol as a color-
less oil (76% yield). 1H NMR (CDCl3): d 1.67 (2H, br d,
J=12.0), 2.07 (2H, t d, J=12.1, 4.0), 2.14 (1H, br s), 2.52
(2H, t d, J=11.8, 2.5), 2.71 (2H, t, J=5.8), 2.78 (2H, br
d, J=12.0), 3.57 (2H, t, J=6.0), 5.32 (1H, s), 6.97–7.02
(4H, m), 7.26–7.29 (4H, m), 7.35 (2H, d, J=8.5), 7.44
(2H, d, J=8.5); mp (oxalate salt): 148–151 ꢁC.
8-[4-[bis(4-Fluorophenyl)methoxy]butyl]-1-phenyl-1,3,8-
triazaspiro-[4,5]decan-4-one (23). As described for 9, 23
was prepared from 6 and 1-phenyl-1,3,8-triazaspiro-
[4,5]decan-4-one as a colorless oil (79% yield). 1H NMR
1-[4-[bis(4-Fluorophenyl)methoxy]butyl]-4-(3-trifluoro-
methylphenyl)-4-piperidinol (17). As described for 9, 17
was prepared from 6 and 4-[3-(trifluoromethyl)phenyl]-

S.-W. Choi et al. / Bioorg. Med. Chem. 10 (2002) 4091–4102
4099
(CDCl3): d 1.59–1.70 (4H, m), 1.72 (2H, br d, J=13.5),
2.43 (2H, t, J=7.5), 2.66 (2H, t d, J=12.7, 5.5), 2.75–
2.84 (4H, m), 3.44 (2H, t, J=6.3), 4.72 (2H, s), 5.29 (1H,
s), 6.85 (1H, t, J=7.3), 6.91 (2H, d, J=8.0), 6.97–7.02
(4H, m), 7.25–7.30 (6H, m), 8.08 (1H, s); mp (oxalate
salt): 106–110 ꢁC.
50 mM Tris, containing 120 mM NaCl, 5 mM KCl, 1.5
mM CaCl2, 4 mM MgCl2, and 1 mM EDTA(pH 7.4).
Then 0.8 mL cell homogenate (0.05 and 0.01 mg protein/
well for D2 and D3, respectively) was added to wells
containing 100 mL test drug or buffer and 100 mL
[3H]YM-09151–2 (0.21 nM final concentration). Chlor-
promazine at 1 mM was used to determine nonspecific
binding. The plates were incubated at 25 ꢁC for 60 min.
The incubation was terminated by rapid filtration
through Whatman GF/B filter paper soaked in 0.1%
polyethylenimine (PEI) on a Brandel cell harvester. The
filters were washed three times with ice-cold 50 mM
Tris–HCl (pH 7.7) and soaked overnight in scintillation
cocktail before counting for 2 min on a Beckman LS
6000 scintillation counter.
2-[4-[4-[4-Bromophenyl]-4-hydroxypiperidin-1-yl]-butyl]-
isoindole-1,3-dione (24). As described for 9, 24 was pre-
pared from 8 and 4-(4-bromophenyl)-4-piperidinol as a
1
white solid (80% yield). H NMR (DMSO-d6): d 1.47
(2H, p, J=7.3), 1.55 (2H, br d, J=12.5), 1.64 (2H, p,
J=7.3), 1.85 (2H, t d, J=12.8, 3.8), 2.31–2.36 (4H, m),
2.61 (2H, br d, J=10.5), 3.42 (1H, br s), 3.61 (2H, t,
J=7.0), 7.42 (2H, d, J=9.0), 7.48 (2H, d, J=9.0), 7.83–
7.88 (4H, m); mp (oxalate salt): 223–225 ꢁC.
Binding assay for the 5-HT1A receptor. HA7 cells
(human receptor) were grown and harvested as above.
The pellet was homogenized in 50 mM Tris–HCl (pH 7.7)
by polytron, centrifuged at 27,000g, and resuspended at
10 mg protein/mL in the same buffer. The homogenate
was then stored at ꢀ70ꢁC in 1-mL aliquots. When needed,
the thawed cells were washed once and resuspended at 10
mg protein/80 mL in 25 mM Tris–HCl, containing 100
mM ascorbic acid and 10 mM nialamide (pH 7.4). The
assay was performed in triplicate in 96-well plates. To
each well, 100 mL [3H]8-OH-DPAT (0.5 nM final con-
centration), 100 mL test compound or buffer, and 0.8
mL cell homogenate (0.1 mg protein/well) were added
by a Tomtec Quadra 96. Dihydroergotamine at 1 mM
was used to determine nonspecific binding. The plates
were incubated at 25 ꢁC for 60 min. The incubation was
terminated by rapid filtration through glass-fiber filter
paper on a Tomtec cell harvester. The filters were
washed four times with ice-cold 50 mM Tris–HCl (pH
7.7), dried overnight, and mixed with 10 mL scintillation
cocktail before counting for 2 min on a Wallac Betaplate
1205 liquid scintillation counter.
Biologic methods
In vitro binding and reuptake inhibition assays. In vitro
assays for [125I]RTI-55-binding inhibition and [3H]neuro-
transmitter-uptake inhibition using HEK 293 cells
expressing recombinant biogenic amine transporters
were carried out according to the procedures described
previously.22,28
In vitro receptor binding and functional biochemical assays.29
Binding assay for the D1 receptor LHD1 cells (human
receptor) were grown to confluence on 100ꢃ20 mm
plates in DMEM containing 10% fetal calf serum,
0.05% penicillin-streptomycin, and 400 mg/mL G418
sulfate. The cells were scraped from the plates and centri-
fuged at 500g for 5 min. The pellet was homogenized in
50 mM Tris–HCl (pH 7.7) by polytron, centrifuged at
27,000g, and resuspended at 10 mg protein/mL in the
same buffer. The homogenate was then stored at ꢀ70 ꢁC
in 1-mL aliquots. When needed, the thawed cells were
washed once and resuspended at 5 mg protein/80 mL
50 mM Tris–HCl, containing 120 mM NaCl, 5 mM
KCl, 2 mM CaCl2, and 1 mM MgCl2 (pH 7.4). The
assay was performed in triplicate in 96-well plates. To
wells containing 100 mL test drug or buffer and 100 mL
[3H]SCH 23,390 (0.18 nM final concentration), 0.8 mL
cell homogenate (0.05 mg protein/well) was added, and
the plates were incubated at 25 ꢁC for 60 min. SCH
23,390 at 1 mM was used to determine nonspecific
binding.
Binding assay for the 5-HT2A receptor. NIH-3T3-GF6
cells (rat receptor) were grown as above. On the day of
the experiment, the cells were thawed, resuspended in 50
mM Tris–HCl, and centrifuged at 27,000g for 12 min.
The pellet was resuspended at 1 mg protein/80 mL 25
mM Tris–HCl (pH 7.7) and 0.8 mL cell homogenate
(0.01 mg protein/well) was added to wells containing
100 mL test drug or buffer and 100 mL [3H]ketanserin
(0.40 nM final concentration). Ketanserin at 1 mM was
used to determine nonspecific binding. The plates were
incubated at 25 ꢁC for 60 min and processed as above
for the 5-HT1A receptor.
Binding assay for the D2 and D3 receptors. CHOp cells
(human receptors) were grown to confluence on 100ꢃ20
mm plates in a minimum essential medium (aMEM),
containing 10% fetal calf serum, 0.05% penicillin–
streptomycin, and 600 mg/mL G418 sulfate. The cells
were scraped from the plates and centrifuged at 500g for
5 min. The pellet was homogenized by polytron in 50
mM Tris–HCl (pH 7.7) and centrifuged at 27,000g for
12 min. The pellet was resuspended in 50 mM Tris to
contain D2 at 5 mg protein/mL or D3 at 1 mg protein/
mL and stored at ꢀ70 ꢁC in 1-mL aliquots. On the day
of the experiment, CHOp-D2 or CHOp-D3 cells were
thawed, resuspended in 50 mM Tris, and centrifuged at
27,000g for 12 min. The pellet was resuspended at 5 mg
protein/80 mL for D2 and 1 mg protein/80 mL for D3 in
Binding assay for the 5-HT2C receptor. NIH-3T3-Pf
cells (rat receptor) were grown as above. The final pellet
was resuspended at 3 mg protein/80 mL 50 mM Tris–
HCl, containing 4 mM CaCl2, 10 mM pargyline, and
0.1% ascorbic acid (pH 7.7). Wells with 100 mL test
drug or buffer, 100 mL [3H]mesulergine (0.4 nM final
concentration), and 0.8 mL cell homogenate (0.03 mg
protein/well) were incubated at 25 ꢁC for 60 min.
Mesulergine at 10 mM was used to determine non-
specific binding. After incubation, the plates were
processed as above for the 5-HT1A receptor.

4100
S.-W. Choi et al. / Bioorg. Med. Chem. 10 (2002) 4091–4102
Binding assay for the 5-HT3 receptor. NG108–15 cells
(rat/mouse hybrid) were grown to confluence on
100ꢃ20 mm plates in DMEM with HAT supplement
and 10% fetal calf serum. The cells were washed from
the plates, centrifuged, homogenized as described
above, and stored at ꢀ70 ꢁC in aliquots of 15 plates/4 mL.
When needed, the thawed cells were washed once and
resuspended at 15 plates/20 mL 25 mM Tris–HCl (pH
7.7). The assay was performed by adding 50 mL test
drug or buffer, 50 mL [3H]GR65630 (1.6 nM final con-
centration), and 0.4 mL cell homogenate (0.13 mg protein/
tube) to each tube. The tubes were then incubated at
25ꢁC for 45 min. Zacopride at 1 mM was used to deter-
mine nonspecific binding. The incubation was terminated
by rapid filtration through Whatman GF/B filter paper
soaked in 0.1% polyethylenimine (PEI) on a Brandel
cell harvester. The filters were washed three times with
ice-cold 50 mM Tris–HCl (pH 7.7) and soaked over-
night in scintillation cocktail before counting for 2 min
on a Beckman LS 6000 scintillation counter.
mice were injected with either 0.9% saline or 20 mg/kg
cocaine intraperitoneally and placed in the Digiscan
apparatus for a 1-h session.
Maximal effects of cocaine and stimulant test compound
(independent studies of cocaine and test compound). One
30-min time period in which maximal effects of cocaine/
test compound were evident at most doses was selected.
The mean (ꢂSEM) maximal stimulant activity (total
counts in the 30-min period divided by 3) was plotted
versus dose. Alog 10 transformation of the 30-min period
average counts for individual subjects was performed to
homogenize variances for subsequent analyses. ANOVA
was used and each dose of cocaine/test compound (spe-
cified a priori) was contrasted to saline (the vehicle) to
determine significant (P<0.05) dose effects. The 30-min
period average counts across subjects was fit to a function
of log10 dose using least-squares curve-fitting analyses
(i.e., TableCurve software from Jandel). The maximum
effect from the resultant dose–response curve was esti-
mated. Atest compound maximal effect/cocaine max-
imum effect (ME/CME) ratio was calculated and the
ED50 (dose that produces 1/2 maximal stimulant activity)
was determined from a linear regression analysis of the
ascending portion (up to the dose that produced a
maximal effect) of the curve. The mean maximal effect
for cocaine was calculated for each mouse.
D2 and D3 functional mitogenesis assay. To measure D2
and D3 inhibition of quinpirole stimulation of mitogen-
esis (antagonist assay), CHOp cells (human receptor)
were seeded in 96-well plates at 5000 cells/well. The cells
were incubated at 37 ꢁC in aMEM with 10% FBS,
0.05% penicillin–streptomycin, and 200 mg/mL G418
sulfate. After 48 h, the wells were rinsed twice with 100
mL serum-free a-MEM and incubated for 24 h at 37 ꢁC
in serum-free a-MEM. The medium was then removed
and replaced with 90 mL serum-free a-MEM and 10 mL
drug in sterile water plus quinpirole. After another 24 h
incubation at 37 ꢁC, 0.25 mCi [3H]thymidine was added
to each well. The cells were incubated for 2 h at 37 ꢁC.
Then, 10 mL/well 10X trypsin (trypsin–EDTAsolution:
5 g trypsin in 20 mL) was added to release the cells, and
the plates were filtered using a Tomtec cell harvester.
The filters were washed four times with deionized water,
dried overnight, and mixed with 10 mL scintillation
cocktail before counting for 2 min on a Wallac Betaplate
1205 liquid scintillation counter. Quinpirole was run on
every plate as an internal standard.
Maximal effects of depressant test compound (studies of
test compound alone). A30-min time period was selected
in which cocaine (20 mg/kg) produced its maximal
effects as determined from the studies with cocaine
alone. The mean (ꢂSEM) activity (total counts in the
30-min period divided by 3) was plotted versus dose. A
log10 transformation of the 30-min period average
counts for individual subjects was performed to homo-
genize variances for subsequent analyses. ANOVA was
used and each dose of the test compound (specified a
priori) was contrasted to the vehicle to determine sig-
nificant (P<0.05) dose effects. Alinear least-squares
regression analysis was conducted and the 30-min period
average counts across subjects were regressed over the
descending potion of the curve against the log10 dose of
the test compound. The ID50 (dose that produces 1/2
maximal depressant activity where maximum depression=
0) was calculated from the linear regression analysis.
Locomotor activity22
This study was conducted using 16 Digiscan locomotor
activity testing chambers (40.5ꢃ40.5ꢃ30.5 cm). Panels
of infrared beams (16 beams) and corresponding photo-
detectors were located in the horizontal direction along
the sides of each activity chamber. Separate groups of
eight nonhabituated male Swiss–Webster mice were
injected via the intraperitoneal route with either vehicle
(methylcellulose, saline, or distilled water) or compound
(3, 10, 30, or 100 mg/kg), 20 min prior to locomotor
activity testing. Just prior to placement in the appara-
tus, all mice received a saline injection intraperitoneally.
In all studies, horizontal activity (interruption of 1
photocell beam) was measured for 1 h within 10-min
periods. Testing was conducted with one mouse per
activity chamber. For the cocaine/compound interac-
tion study, 20 min following administration of vehicle
or compound intraperitoneally (3, 10, 30, or 100 mg/
kg), groups of eight nonhabituated male Swiss–Webster
Maximal effects of test compound–cocaine interaction
studies. A30-min time period was selected in which
cocaine (20 mg/kg) produced its maximal activity as
determined from the studies with cocaine alone. The
mean (ꢂSEM) maximal activity (total counts in the
30-min period divided by 3) for vehicle, vehicle pre-
treatment plus cocaine (20 mg/kg), and test compound
pretreatment plus cocaine (20 mg/kg) were plotted in a
histogram. Alog transformation of the 30-min period
10
average counts for individual subjects was performed to
homogenize variances for subsequent analyses.
ANOVA was used and vehicle and each dose of the test
compound plus cocaine to cocaine alone (specified a
priori) were contrasted to determine significant
(P<0.05) dose effects. Alinear least-squares regression
analysis was performed and the 30-min period average

S.-W. Choi et al. / Bioorg. Med. Chem. 10 (2002) 4091–4102
4101
counts across subjects were regressed over the descend-
ing portion of the curve against the log10 dose of the test
compound. The AD50 (dose that attenuates cocaine-
induced stimulation by 50%) was calculated from the
linear regression analysis.
2-[4-[bis(4-Fluorophenyl)methoxy]butyl]-6,7-dimethoxy-
1-phenyl-1,2,3,4- tetrahydroisoquinoline (21). Anal.
calcd for C36H37NO7F2: C, 68.23; H, 5.89; N, 2.21.
Found: C, 67.69; H, 6.02; N, 2.14.
1-[1-[4-[bis(4-Fluorophenyl)methoxy]butyl]piperidin-4-yl]-
1,3-dihydrobenzoimidazol-2-one (22). Anal. calcd for
C31H33N3O6F2: C, 64.02; H, 5.72; N, 7.22. Found: C,
63.93; H, 5.67; N, 7.19.
Elemental Analysis
1-[4-[bis(4-Fluorophenyl)methoxy]butyl]-4-hydroxypiper-
idine (9). Anal. calcd for C24H31NO7F2: C, 59.62; H,
6.46; N, 2.90. Found: C, 60.12; H, 6.08; N, 2.84.
8-[4-[bis(4-Fluorophenyl)methoxy]butyl]-1-phenyl-1,3,8-
triazaspiro.[4,5]decan-4-one (23) Anal. calcd for
C32H35N3O6F2: C, 64.53; H, 5.92; N, 7.05. Found: C,
63.75; H, 6.01; N, 6.86.
1-[4-[bis(4-Fluorophenyl)methoxy]butyl]-3-hydroxypiper-
idine (10). Anal. calcd for C24H29NO6F2: C, 61.93; H,
6.28; N, 3.01. Found: C, 61.84; H, 6.23; N, 2.99.
2-[4-[4-[4-Bromophenyl]-4-hydroxypiperidin-1-yl]-butyl]-
isoindole-1,3-dione (24). Anal. calcd for C25H27BrN2O7:
C, 54.85; H, 4.97; N, 5.12. Found: C, 54.33; H, 4.49; N,
5.24.
1-[4-[bis(4-Fluorophenyl)methoxy]butyl]-3-piperidine carb-
oxamide (11). Anal. calcd for C25H32N2O7F2: C, 58.82;
H, 6.32; N, 5.49. Found: C, 59.02; H, 5.96; N, 5.29.
Acknowledgements
1-[4-[bis(4-Fluorophenyl)methoxy]butyl]-4-phenylpiper-
idine-4-carbonitrile (12). Anal. calcd for (C31H32N2O5F2):
C, 67.62; H, 5.86; N, 5.09. Found: C, 67.53; H, 5.91; N,
5.05.
The authors are grateful to the Benedict Cassen Post-
doctoral Fellowship program sponsored by the Society
of Nuclear Medicine. We are thankful for the service
provided by the NIDA-supported Cocaine Treatment
Discovery Program (CTDP) Medication Development
Division (MDD) and for in vivo data from Robert J.
Mitkus and Jonathan L. Katz of the Psychobiology
Section. In addition we are grateful for the NIDA
Intramural Research Program for the in vitro data from
Aaron Janowsky, PhD, Oregon Health Sciences Uni-
versity, Portland, OR, USA.
1-[4-[bis(4-Fluorophenyl)methoxy]butyl]-4-acetyl-4-phenyl-
piperidine (13). Anal. calcd for C32H35NO6F2: C, 67.71;
H, 6.22; N, 2.47. Found: C, 67.86; H, 6.32; N, 2.50.
1-[4-[bis(4-Fluorophenyl)methoxy]butyl]-4-phenylpiper-
idin-4-ol (14). Anal. calcd for C30H33NO6F2: C, 66.53;
H, 6.14; N, 2.59. Found: C, 66.66; H, 6.25; N, 2.59.
1-[4-[bis(4-Fluorophenyl)methoxy]butyl]-4-(4-bromophenyl)-
4-piperidinol (15). Anal. calcd for C30H32BrNO6F2: C,
58.07; H, 5.20; N, 2.26. Found: C, 58.09; H, 5.28; N,
2.24.
References and Notes
1. (a) Das, G. J. Clin. Pharmacol. 1993, 33, 296. (b) Sub-
stance Abuse and Mental Health Services Administration
National Household Survey on Drug Abuse: Main Findings
1997; Department of Health and Human Services: Washing-
ton, D.C., 1998. (c) McCance, E. F. NIDA Res. Monogr. 1997,
175, 36.
1-[4-[bis(4-Fluorophenyl)methoxy]ethyl]-4-(4-bromophenyl)-
4-piperidinol (16). Anal. calcd for C28H28BrNO6F2: C,
56.77; H, 4.76; N, 2.36. Found: C, 56.00; H, 4.96; N,
2.43.
2. For reviews, see (a) Carroll, F. I.; Lewin, A. H.; Boja, J. W.;
Kuhar, M. J. J. Med. Chem. 1992 35 969, and references
therein (b) Carroll, F. I.; Howell, L. L.; Kuhar, M. J. J. Med.
Chem. 1999, 42, 2721. (c) Howell, L. L; Wilcox, K. M. Pre-
spectives in Pharmacology 2001, 28, 1.
1-[4-[bis(4-Fluorophenyl)methoxy]butyl]-4-(3-trifluoro-
methylphenyl)-4-piperidinol (17). Anal. calcd for
C31H32NO6F5: C, 61.08; H, 5.29; N, 2.30. Found: C,
60.96; H, 5.31; N, 2.26.
3. (a) Ritz, M. C.; Lamb, R. J.; Goldberg, S. R.; Kuhar, M. J.
Science 1987, 237, 1219. (b) Kuhar, M. J.; Ritz, M. C.; Boja,
J. W. Trends Neurosci. 1991, 14, 299. (c) Giros, B.; Jaber, M.;
Jones, S. R.; Wightman, R. M.; Caron, M. G. Nature 1996,
379, 606.
4-Benzyl-1-[4-[bis(4-fluorophenyl)methoxy]butyl]-4-piper-
idinol (18). Anal. calcd for C32H42NO8F2: C, 61.62; H,
6.95; N, 2.25. Found: C, 61.42; H, 5.82; N, 2.14.
4. (a) Reith, M. E. A.; Sershen, H.; Lajtha, A. Life Sci. 1980,
27, 1055. (b) Reith, M. E. A.; Meisler, B. E.; Sershen, H.;
Lajtha, A. Brain Res. 1985, 342, 145.
5. (a) Witkin, J. M. Neurosci. Biobehav. Rev. 1994, 18, 121. (b)
Mello, N. K.; Negus, S. S. Neuropsychopharmacology 1996,
14, 375.
2-[4-[bis(4-Fluorophenyl)methoxy]butyl]-1,2,3,4-tetrahydro-
isoquinoline (19). Anal. calcd for C28H29NO5F2: C,
67.59; H, 5.88; N, 2.82. Found: C, 67.54; H, 5.86; N,
2.79.
2-[4-[bis(4-Fluorophenyl)methoxy]butyl]-6,7-dimethoxy-
1,2,3,4- tetrahydroisoquinoline (20). Anal. calcd for
C30H33NO7F2: C, 64.62; H, 5.97; N, 2.51. Found: C,
64.96; H, 5.88; N, 2.24.
6. (a) Spealman, R. D. Psychopharmacology 1990, 101, 142.
(b) Witkin, J. M.; Schindler, C. W.; Tella, S. R.; Goldberg,
S. R. Psychopharmacology 1991, 104, 425.
7. (a) Corrigall, W. A.; Coen, K. M. Pharmacol. Biochem.

4102
S.-W. Choi et al. / Bioorg. Med. Chem. 10 (2002) 4091–4102
Behav. 1991, 39, 799. (b) Britton, D. R.; Curzon, P.; Mac-
Kenzie, R. G.; Kebabian, J. W.; Williams, J. E. G.; Kerkman,
D. Pharmacol. Biochem. Behav. 1991, 39, 911.
8. Shafer, R. A.; Levant, B. Psychopharmacology 1998, 135, 1.
9. Pilla, M.; Perachon, S.; Sautel, F.; Garrido, F.; Mann, A.;
Wermuth, C. G.; Schwartz, J.-C.; Everitt, B. J.; Sokoloff, P.
Nature 1999, 400, 371.
10. Rothman, R. B.; Becketts, K. M.; Radesca, L. R.; de
Costa, B. R.; Rice, K. C.; Carroll, F. I.; Dersch, C. M. Life
Sci. 1993, 53, PL.
11. Deutsch, H. M.; Collard, D. M.; Zhang, L.; Burnham,
K. S.; Deshpande, A. K.; Holtzman, S. G.; Schweri, M. M. J.
Med. Chem. 1999, 42, 882.
12. (a) Simoni, D.; Stoelwinder, J.; Kozikowski, A. P.; John-
son, K. M.; Bergmann, J. S.; Ball, R. G. J. Med. Chem. 1993,
36, 3975. (b) Nader, M. A.; Grant, K. A.; Davies, H. M. L.;
Mach, R. H.; Childers, S. R. J. Pharmacol. Exp. Ther. 1997,
280, 541.
13. (a) Matecka, D.; Rothman, R. B.; Radesca, L.; de Costa,
B. R.; Dersch, C. M.; Partilla, J. S.; Pert, A.; Glowa, J. R.;
Wojnicki, F. H. E.; Rice, K. C. J. Med. Chem. 1996, 39, 4704.
(b) Matecka, D.; Lewis, D.; Rothman, R. B.; Dersch, C. M.;
Wojnicki, F. H. E.; Glowa, J. R.; DeVries, A. C.; Pert, A.;
Rice, K. C. J. Med. Chem. 1997, 40, 705. (c) Zhang, Y. Med.
Chem. Res. 1998, 8, 66.
19. Katz, J. L.; Izenwasser, S.; Terry, P. Psychopharmacology
2000, 148, 90.
20. Howell, L. L.; Czoty, P. W.; Byrd, L. D. Psychopharma-
cology 1997, 131, 40.
21. Deutsch, H. M.; Schweri, M. M. Life Sci. 1994, 55, PL.
22. (a) Hanson, R. N.; Choi, S.-W.; Elmaleh, D. R.; Fisch-
man, A. J. Bioorg. Med. Chem. Lett. 1997, 7, 2559. (b) Choi,
S.-W.; Elmaleh, D. R.; Hanson, R. N.; Fischman, A. J. J.
Med. Chem. 1999, 42, 3647.
23. Dutta, A. K.; Meltzer, P. C.; Madras, B. K. Med. Chem.
Res. 1993, 3, 209.
24. Taverne, T.; Diouf, O.; Depreux, P.; Poupaert, J. H.;
Lesieur, D.; Guardiola-Lemaitre, B.; Renard, P.; Rettori, M-
C.; Caignard, D-H.; Pfeiffer, B. J. Med. Chem. 1998, 41, 2010.
25. Caliendo, G.; Fiorino, F.; Grieco, P.; Perissutti, E.; San-
tagada, V.; Severino, B.; Bruni, G.; Romeo, M. R. Bioorg.
Med. Chem. 2000, 8, 533.
26. Results from an arylpiperazine series are currently under
manuscript preparation.
27. Ahmad, Y. E.; Maillet, P.; Laurent, E.; Talab, A.; Teste,
J. F.; Cedat, M. J.; Fiez-Vandal, P. Y.; Dokhan, R.; Ollivier,
R. Eur. J. Med. Chem. 1997, 32, 205.
28. Eshleman, A. J.; Carmolli, M.; Cumbay, M.; Martens,
C. R.; Neve, K. A.; Janowsky, A. J. Pharmacol. Exp. Ther.
1999, 289, 877.
14. (a) Madras, B. K.; Reith, M. E. A.; Meltzer, P. C.; Dutta,
A . K.Eur. J. Pharmacol. 1994, 267, 167. (b) Dutta, A. K.;
Coffey, L. L.; Reith, M. E. A. J. Med. Chem. 1997, 40, 35.
15. (a) van der Zee, P.; Koger, H. S.; Gootjes, J.; Hespe, W.
Eur. J. Med. Chem. 1980, 15, 363. (b) Rothman, R. B.; Mele,
A.; Reid, A. A.; Akunne, H. C.; Greig, N.; Thurkauf, A.; de
Costa, B. R.; Rice, K. C.; Pert, A. Pharmacol. Biochem. Behav.
1991, 40, 387. (c) Baumann, M. H.; Char, G. U.; de Costa,
B. R.; Rice, K. C.; Rothman, R. B. J. Pharmacol. Exp. Ther.
1994, 271, 1216. (d) Tella, S. R.; Ladenheim, B.; Andrews,
A. M.; Goldberg, S. R.; Cadet, J. L. J. Neurosci. 1996, 16, 7416.
16. Lewis, D. B.; Matecka, D.; Zhang, Y.; Hsin, L.-W.;
Dersch, C. M.; Stafford, D.; Glowa, J. R.; Rothman, R. B.;
Rice, K. C. J. Med. Chem. 1999, 42, 5029.
29. (a) In vitro pharmacological assays for receptors by
NIDA/MDD (NIDA contract NO1DA-7–8072, Contract
title: ‘Receptor Activity Testing for Medication Discovery’. (b)
Chio, C. L.; Lajiness, M. E.; Huff, R. M. Mol. Pharmacol.
1994, 45, 51.
30. Dubinsky, B.; McGuire, J. L.; Niemegeers, C. J. E.; Jans-
sen, P. A. J.; Weintraub, H. S.; McKenzie, B. E. Psycho-
pharmacology 1982, 78, 1.
31. After testing compounds 18 and 23 which are also potent
NE reuptake inhibitor, NIDAindicated that they are looking
for ligands that posses a high DR value and a high selectivity
but with a low NE reuptake inhibition. Also, compound 22
contained a partial structure of a strong dopamine D2 recep-
tor antagonist, and thus, metabolism might generate a
N-dealkylation product, which has freezing activities that can
be correlated to the EPS of conventional antipsychotics.
17. Bergman, J.; Madras, B. K.; Johnson, S. E.; Spealman,
R. D. J. Pharmacol. Exp. Ther. 1989, 251, 150.
18. Howell, L. L.; Byrd, L. D. J. Pharmacol. Exp. Ther. 1991,
258, 178.
˜
32. Manzanedo, C.; Aguilar, M. A.; Minarro, J. Psycho-
pharmacology 1999, 143, 82.