High affinity hydroxypiperidine analogues of 4-(2-benzhydryloxyethyl)-1-(4-fluorobenzyl)piperidine for the dopamine transporter: Stereospecific interactions in vitro and in vivo

Article abstract of DOI:10.1021/jm020275k

In our effort to develop high-affinity ligands for the dopamine transporter which might find potential use as cocaine medication, a polar hydroxy substituent was introduced into the piperidine ring of one of our disubstituted lead analogues derived from 1

Full text of DOI:10.1021/jm020275k

1220
J . Med. Chem. 2003, 46, 1220-1228
High Affin ity Hyd r oxyp ip er id in e An a logu es of
4-(2-Ben zh yd r yloxyeth yl)-1-(4-flu or oben zyl)p ip er id in e for th e Dop a m in e
Tr a n sp or ter : Ster eosp ecific In ter a ction s in Vitr o a n d in Vivo
Sujit K. Ghorai,† Charles Cook,‡ Matthew Davis,† Sylesh K. Venkataraman,† Clifford George,§
Patrick M. Beardsley,‡ Maarten E. A. Reith,| and Aloke K. Dutta*,†
Wayne State University, Department of Pharmaceutical Sciences, Detroit, Michigan 48202, Virginia Commonwealth University,
Department of Pharmacology and Toxicology, Richmond, Virginia 23298, Laboratory for the Structure of matter, Code 6030,
Naval Research Laboratory, 4555 Overlook Avenue, SW, Washington, DC 20735, and University of Illinois,
College of Medicine, Department of Biomedical and Therapeutic Sciences, Peoria, Illinois 61605
Received J une 27, 2002
In our effort to develop high-affinity ligands for the dopamine transporter which might find
potential use as cocaine medication, a polar hydroxy substituent was introduced into the
piperidine ring of one of our disubstituted lead analogues derived from 1-[2-(diphenylmethoxy)-
ethyl]-4-(3-phenylpropyl)piperazine (GBR 12935). Both cis- and trans-3-hydroxy derivatives
were synthesized and the racemic trans isomer, (()-5, was further resolved into two
enantiomers. Newly synthesized compounds were characterized for their binding affinity at
the dopamine, serotonin, and norepinephrine transporter systems in rat brain. The two
enantiomers (+)-5 and (-)-5 exhibited marked differential affinities at the dopamine transporter
with (+)-5 being 122-fold more potent than (-)-5 in inhibiting radiolabeled cocaine analogue
binding (IC50; 0.46 vs 56.7 nM) and 9-fold more active for inhibiting dopamine uptake (IC50;
4.05 vs 38.0 nM). Furthermore, the most active (+)-5 was 22-fold more potent at the dopamine
transporter compared to the standard GBR 12909. Absolute configuration of one of the
enantiomers was determined unambiguously by X-ray structural analysis. In in vivo locomotor
activity studies, the enantiomer (+)-5 and the racemic (()-5, but not (-)-5, exhibited stimulant
activity with a long duration of effect. All three compounds, (+)-5, (-)-5, and (()-5, within the
dose range tested, partially (50%) but incompletely (80%) produced cocaine-like responses in
mice trained to discriminate 10 mg/kg ip cocaine from vehicle. Compound (-)-5 was distinctive
in this regard in that, unlike (+)-5 and (()-5, it did not affect locomotor activity yet, but similar
to them, was able to engender (albeit incompletely) cocaine-like responses.
In tr od u ction
Since DAT has been implicated strongly in cocaine’s
reinforcing effect, many efforts have been directed
toward developing drugs for this transporter in an effort
to develop medication for cocaine addiction. Structurally
different classes of compounds have been developed for
DAT, and they are classified mainly into tropane,
benztropine, methylphenidate, mazindol, and GBR de-
rivatives.9,10 In our structure-activity relationship (SAR)
study with piperidine analogues of GBR 12935, we have
developed many potent and selective analogues for the
DAT molecule. We have looked at various aspects of
aromatic substitutions, influence of chain lengths, and
bioisosteric replacements with different isosteric aro-
matic moieties in these molecules.11-13 In this respect,
many analogues of conventional piperazine derivatives
of GBR 12935 and the benztropine molecule have been
made and characterized.14-16 A wealth of SAR data thus
generated has been very useful in the development of
an understanding of the mode of interaction of these
molecules with the DAT.
Cocaine is a powerful drug of abuse with strong
reinforcing activity. Addiction to cocaine is a major
problem in our society today which has greatly impacted
the nation in terms of its economy and securing law and
order.1 Furthermore, it has also contributed to the
spreading of HIV infection, as needle sharing is a
pervasive problem among drug abusers. At present no
effective medication is available for the treatment of
cocaine addiction, and there is an urgent need for the
development of an effective medication.2 Cocaine acts
on several sites in the brain which include all three
monoamine neurotransporter systems for dopamine,
serotonin, and norepinephrine.3 Mechanistically co-
caine’s reinforcing effect is believed to originate mainly
from its binding to the dopamine transporter (DAT).4-6
Additionally, recent experiments involving knockout
mice indicated that other systems, e.g., serotonin, might
contribute to some extent to the central effects of
cocaine.7,8
Hydroxy-functionalized GBR compounds have been
synthesized in the past and recently. Rice et al. inves-
tigated introduction of a hydroxy group in GBR 12935
derivatives. In their derivatives, hydroxyl group was
introduced on the phenyl propyl side chain and also in
the phenyl ring.15,17 In this regard, one of the earlier
* Corresponding author: Aloke K. Dutta, Ph.D., Department of
Pharmaceutical Sciences, Applebaum College of Pharmacy & Health
Sciences, Rm# 3128, Detroit, MI 48202, Tel: 1-313-577-1064, Fax:
1-313-577-2033, e-mail: adutta@ea.cphs.wayne.edu.
† Wayne State University.
‡ Virginia Commonwealth University.
§ Naval Research Laboratory.
| University of Illinois.
10.1021/jm020275k CCC: $25.00 © 2003 American Chemical Society
Published on Web 03/04/2003

High Affinity Ligands for the Dopamine Transporter
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 7 1221
F igu r e 1. Molecular structures of DAT blockers.
hydroxy derivatives developed from this group was
converted into a decanoate ester prodrug which could
attenuate self-administration of cocaine for an extended
period of time.18 In our own work with piperidine
analogues of GBR 12935, we have shown the introduc-
tion of hydroxy and methoxy groups in the pendant
phenyl ring of the N-substituted benzyl group resulted
in production of potent and selective derivatives.13
A
F igu r e 2. The molecular structure and numbering scheme
for one of the two molecules in the asymmetric unit of
compound (-)-5 with displacement ellipsoids drawn at the 30%
probability level.
suitable hydroxy compound, when developed from these
studies, can be converted into a lipophilic prodrug such
as decanoate ester. Such a prodrug will have the
potential to be a long-acting treatment agent in substi-
tution therapy for cocaine addiction.
of the olefinic double bond in 4 was carried out by a
borane reagent generated by reaction of sodium boro-
hydride with anhydrous boron trifluoride. The resulting
borane complex was hydrolyzed by sodium hydroxide
solution to yield racemic trans-hydroxy compound (()-5
in good yield. The trans structure of the product was
confirmed by both X-ray structure of (-)-5 (Figure 2)
In our SAR series of piperidine analogues so far,
introduction of a third substituent in the central pip-
eridine ring had not yet been explored. As the central
piperidine/piperazine ring could play an important role
in the interaction of these molecules with CNS recogni-
tion sites, we decided to explore the influence of
introduction of a polar group in the piperidine ring of
our previous lead compound 1. Polar groups such as
hydroxy or amino may give rise to additional interac-
tions through H-bonding or ionic interaction with the
DAT which can potentially change their functional
activity in addition to their binding interaction. More-
over, the presence of such functional group(s) will also
be helpful in making suitable long-acting prodrugs for
medication purpose.19 In this report, we are describing
the design, synthesis, and biological characterization of
hydroxy substituted piperidine analogues of 1 (Figure
1).
1
and H NMR data analysis. Resolution of the racemic
trans compound was carried out by converting racemic
(()-5 into diastereoisomeric ester mixture 6 by treating
with optically active S-(-)-camphanic chloride in the
presence of a base. Separation of the diastereomeric
mixture was carried out by semipreparative HPLC, and
the respective optically pure hydroxy enantiomer was
liberated by treating each pure diastereomer with
potassium carbonate in methanol. One of the crystal-
lized pure enantiomers, (-)-5, underwent X-ray struc-
tural analysis to establish the absolute stereochemistry
of both the enantiomers.
Synthesis of cis-hydroxy compound 8 was achieved by
first oxidizing the hydroxy compound 5 to a keto
derivative 7 followed by reduction with sodium borohy-
dride. Thus, racemic hydroxy compound 5 was oxidized
under Swern oxidation condition to the keto compound
7 in good yield. Reduction of this keto compound 7 by
sodium borohydride in methanol produced a mixture of
trans-(()-5 and cis-(()-8 hydroxy molecules which were
separated by column chromatography. Finally, the
racemic trans-hydroxy (()-5 was converted into an ester
derivative 9.
The stereochemical assignment of the structures 5
and 8 was determined by 1-D and 2-D NMR (400 MHz)
experiments (see Supporting Information for detail) and
by X-ray crystal structure. In these compounds the
assumption that the benzhydrloxyethyl moiety would
assume the equatorial position was borne out by the
X-ray structure of one of the enantiomers of 5 and was
also confirmed by the 1H NMR data. The 1H NMR
spectra of compound 5 was taken in both CDCl3 and
Ch em istr y
Synthesis of our target compounds is shown in the
Schemes 1 and 2. We wanted to introduce a hydroxyl
functionality at the 3-position in the piperidine ring by
chemically modifying a double bond which was intro-
duced in the ring for this purpose. Our intention was
to synthesize both cis- and trans-hydroxy compounds
from this precursor. To generate a double bond in the
correct position of the piperidine ring, we planned to
exploit a well-known reaction involving reduction of a
quaternary pyridinium salt by sodium borohydride in
methanol to produce tetrahydropyridine intermediate
4.20 Thus, in Scheme 1, the intermediate 3 was first
made by heating 2-pyridin-4-ylethanol and 4-fluoroben-
zyl chloride together in methanol which was followed
by the reduction with sodium borohydride to provide the
tetrahydro intermediate 3. Conversion of the intermedi-
ate 3 into diphenylmethoxy derivative 4 was conducted
by following our published procedure.11 Hydroboration

1222 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 7
Ghorai et al.
Sch em e 1
Sch em e 2
CD3OD solvents. The spectra taken in CD3OD was
chosen for analysis as it gave clearer signals.
Finally, the absolute configuration of (-)-5 was de-
termined unambiguously from X-ray structural analysis
(Figure 2) which designates its stereocenters as (3S,4S).
Consequently, its enantiomer (+)-5 must have the
(3R,4R) configuration.
and synthesis of trans- and cis-hydroxy compounds 5
and 8. Biological studies of these novel compounds were
carried out to evaluate their binding affinity at the
dopamine, serotonin (SERT), and norepinephrine (NET)
transporter systems in the rat brain by measuring
competition for the binding of [3H]WIN 35 428, [3H]-
citalopram, and [3H]nisoxetine, respectively. Selected
compounds were also evaluated for their activity in
inhibiting the uptake of [3H]dopamine.
Resu lts a n d Discu ssion
Our effort to introduce a polar hydroxy group at the
3-position of the piperidine ring in one of our lead
piperidine analogues 1 (Figure 1) resulted in the design
Binding results demonstrated that introduction of a
hydroxy functionality in the piperidine ring of compound
1 resulted in the development of potent racemic trans

High Affinity Ligands for the Dopamine Transporter
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 7 1223
Ta ble 1. Affinity of Drugs at the Dopamine, Serotonin, and Norepinephrine Transporters in Rat Striatum and in Inhibition of
Dopamine (DA) Reuptake
DAT binding, IC50, nM,
[3H]WIN 35, 428a
SERT binding, IC50, nM,
[3H]citaloprama
NET binding, IC50, nM,
[3H]nisoxetinea
DAT uptake, IC50, nM,
[3H]DAa
compd
cocaine
GBR 12909
1
266 ( 37
10.6 ( 1.9
17.2 ( 4.7b
37.8 ( 3.4
4.14 ( 0.77
0.46 ( 0.05
56.7 ( 6.5
99 ( 18
737 ( 160
132 ( 0
1920 ( 230
1110 ( 120
2360 ( 500
3600 ( 270
1830 ( 80
7750 ( 1620
3310 ( 270
178000 ( 53000
3530 ( 550
496 ( 22
NTc
6.63 ( 0.43
2.48 ( 0.59
NT
3.22 ( 1.0
4.05 ( 0.73
38.0 ( 6.0
4
NT
(()-5
1030 ( 80
1880 ( 230
1550 ( 190
6300 ( 690
2150 ( 660
35800 ( 7700
(+)-R,R-5
(-)-S,S-5
7
(()-8
9
11.2 ( 1.0
583 ( 81
10.1 ( 0.7
NT
a
For binding, the DAT was labeled with [3H]WIN 35, 428, the SERT with [3H]citalopram and the NET with [3H]nisoxetine. For uptake
by DAT, [3H]DA accumulation was measured. Results are average ( SEM of three to eight independent experiments assayed in triplicate.
Concentrations of radioligands used in binding 4.0, 3.0, and 1.1 nM for DAT, SERT, and NET. For uptake experiment 50 nM of [3H]DA
b
was used. See reference no. 12. c Not tested.
Ta ble 2. Selectivity of Various Ligands for Their Activity at
compound (()-5. Clearly, the presence of a hydroxyl
Monoamine Transporters
functionality in compound 5 enhanced the interaction
SERT binding/
DAT binding
NET binding/
DAT binding
DAT uptake/
DAT binding
with the DAT compared with the parent molecule 1, as
racemic (()-5 displayed an approximately 3-fold lower
IC50 value. This result perhaps indicates generation of
additional binding interaction(s) mediated by the newly
introduced hydroxyl group, possibly via H-bonding. In
addition, compound (()-5 exhibited very high selectivity
for the DAT when its binding was compared with that
to the SERT and NET. When (()-5 was further resolved
into its enantiomers (Table 1), an appreciable separation
of activity was noted between the two enantiomers (+)-5
and (-)-5 with (+)-5 being much more potent than (-)-5
(0.46 nM vs 56.7 nM), consonant with stereospecific
interaction of these molecules with the recognition sites.
The most active (+)-5 was 38-fold more active than the
parent 1 (0.46 vs 17.2 nM).
compound
GBR 12909
1
4
12
110
29
47
0.62
0.14
(()-5
570
7800
32
250
4100
27
64
190
61
1.7
8.7
0.67
(+)-R,R-5
(-)-S,S-5
7
78
(()-8
9
300
300
0.90
5, (+)-5, and (-)-5 to in vivo locomotor and drug
discrimination studies.
Effects of Coca in e, (()-5, (-)-5, a n d (+)-5 on Tota l
Dista n ce Tr a veled . Figure 3 shows the time-course
effects of cocaine, (()-5, (-)-5, and (+)-5 during the first
60 min of testing as well as the summed effects across
the remaining 3 h. Doses of 10 and 30 mg/kg cocaine
produced peak effects on total distance traveled during
the first 10 min following administration with near-
vehicle levels of activity obtained by 60 min (Figure 3A).
When the data were summed across the remaining 3 h,
cocaine did not significantly increase total distance
traveled (F(3,27) ) 2.52, P >0.05). During the first 60
min, doses of 30 and 56 mg/kg (()-5 produced peak
levels of activity between 20 and 30 min postadminis-
tration and significantly increased distance traveled
across the remaining 3 h (F(3,28) ) 7.76, P e 0.05)
(Figure 3B). (-)-5, up to a dose of 100 mg/kg, had little
effect on total distance traveled during the first 60 min
or across the remaining three h of the experimental
sessions (F(3,28) ) 1.88, P > 0.05) (Figure 3C). Doses
of 30 and 56 mg/kg of (+)-5 produced increases in total
distance traveled with peak effects occurring within the
first 20 min (Figure 3D). Locomotor activity effects of
56 mg/kg of (+)-5 decreased to, and remained near,
vehicle levels following 20 min postinjection, but activity
levels following 30 mg/kg remained elevated throughout
this period. Activity levels following administration of
56 mg/kg of (+)-5, which were near-vehicle levels from
20 to 60 min, increased to levels similar to that of 30
mg/kg such that when the data were summed over the
remaining 3 h of the test session both doses produced
significant increases in total distance traveled (F(4,35)
) P e 0.05). As a comparison, a dose of 30 mg/kg cocaine
results in producing peak levels of activity (approxi-
We next wanted to evaluate the importance of con-
tribution of the hydroxyl functionality in (()-5 in its
high-affinity binding to the DAT. Thus, we converted
the hydroxy group into its ester as in 9 and also oxidized
it to the keto compound 7. The keto compound 7 showed
moderate activity while the ester derivative 9 exhibited
relatively weak binding activity at the DAT (99 nM and
583 nM, respectively). In addition, compound 9 became
essentially inactive at the SERT and NET. These results
show the importance of a hydroxyl group for high
affinity binding of compound (()-5 to the DAT probably
due to H-bonding with the receptor. In consonance, this
high potency was lost when the hydroxyl group was
converted into an ester group as in 9.
Next we evaluated another stereoisomeric form of
trans-(()-5, compound cis-(()-8. The results showed cis
derivative (()-8 to be three times less potent at DAT
than trans compound (()-5 (11.4 vs 4.1 nM), illustrating
the importance of stereochemistry of the hydroxyl
moiety at the 3-position of the piperidine ring in binding
to DAT; thus, the equatorial orientation of the hydroxyl
group in the trans isomer was more favored for interac-
tion than the axial orientation in the cis compound.
Selected potent compounds were tested for their
potency in inhibiting dopamine uptake by DAT. Uptake
activity of most compounds corresponded very well with
their binding activity except for the compound (+)-5
(IC50 of 4.05 nM for uptake inhibition vs 0.464 nM for
binding inhibition). Following characterization of in
vitro transporter activities, we subjected compounds (()-

1224 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 7
Ghorai et al.
F igu r e 3. Effects of cocaine, (()-5, (-)-5, and (+)-5 on total distance traveled (cm) across the first 60 min of the test session as
well as the effects of cocaine, (()-5, (-)-5, and (+)-5 when the data were summed across the remaining 3 h of the test session (10
min intervals 7-24). * Indicates significant differences compared to vehicle control based on Dunnett’s posthoc tests.
effects (63%) obtained at a dose of 56 mg/kg resulting
in an ED50 value of 27.66 (11.13-68.72). Two mice failed
to respond at the 56 mg/kg dose of (()-5, and the group
mean rate of responding was reduced to levels similar
to that produced by 30 mg/kg cocaine. (+)-5 produced
maximal cocaine lever responding (67%) at a dose of 30
mg/kg resulting in an ED50 value of 13.59 (4.90-37.71).
At a dose of 30 mg/kg of (+)-5, rate of responding was
suppressed with one mouse not responding. (-)-5 pro-
duced dose-dependent increases in cocaine lever press-
F igu r e 4. Effects of cocaine, (()-5, (-)-5, and (+)-5 on total
ing. At a dose of 56 mg/kg, which resulted in 50%
distance traveled (cm) when the data were summed across the
cocaine lever pressing, one of four rats failed to respond.
entire 4 h test session. * Indicates significant differences
At a dose of 100 mg/kg only three mice were tested, and
compared to vehicle control based on Dunnett’s posthoc tests.
this dose produced 99% cocaine lever pressing in the
mately 2800 cm in total distance traveled) at 10 and 20
min postadministration, with total distance traveled
values returning to near vehicle control levels by ap-
proximately 2 h.13
one mouse that responded. Because only one mouse
responded at a dose of 100 mg/kg, an ED50 value was
not calculated.
Con clu sion
Figure 4 shows the effects of cocaine, (()-5, (-)-5, and
(+)-5 at each dose when averaged over the entire 4-h
test session. There was a main effect of dose for cocaine
(F(3,27) ) 12.87, P e 0.05) such that a dose of 30 mg/
kg increased total distance traveled relative to vehicle
control. There was a main effect of dose for (()-5 (F(3,-
28) ) 13.55, P e 0.05) and (+)-5 (F(4,35) ) 10.49, P e
0.05) such that doses of 30 and 56 mg/kg increased total
distance traveled relative to vehicle control. At the doses
tested, (-)-5 failed to alter total distance traveled
relative to vehicle control (F(3,28) ) 1.17, P > 0.05).
Effects of Coca in e, (()-5, (-)-5, (+)-5 in Coca in e-
Discr im in a tin g Mice. As shown in Figure 5, cocaine
produced dose-dependent increases in cocaine lever
responding with an ED50 value (( 95% C.I.) of 3.35 mg/
kg (2.76-4.07). At the highest dose of cocaine tested (30
mg/kg) rate of responding was decreased by over 50%
relative to saline levels. (()-5 produced dose-dependent
increases in cocaine-lever responding with maximal
In this report we describe development of 3-hydroxy-
substituted piperidine derivatives of GBR 12935 which
exhibited low to subnanomolar activity for the DAT.
Trans isomers displayed somewhat more activity com-
pared to cis isomers. Enantiomers of the trans isomer
showed differential activity with R,R-(+)-5 exhibiting
much higher activity than the S,S-(-)-5 isomer. Separa-
tion of activity between the two enantiomers was very
high, but such differential activity is not unprec-
edented.15 This may be due to complex interaction of
this racemic mixture with the recognition sites of the
transporter molecule causing reduction in affinity of the
most active enantiomer. A 9-fold separation between
cocaine analogue binding and dopamine uptake inhibi-
tory activity was obtained for R,R-(+)-5. As in the in
vitro dopamine uptake assays, R,R-(+)-5 was more
potent than S,S-(-)-5 in the in vivo behavioral assays,
and it was more potent than, or similar to, racemic (()-

High Affinity Ligands for the Dopamine Transporter
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 7 1225
was either CDCl3 or CD3OD as indicated. TMS was used as
an internal standard. Elemental analyses were performed by
Atlantic Microlab, Inc. and were within ( 0.4% of the theoreti-
cal value. Optical rotation were recorded on a Perkin-Elmer
241 polarimeter.
[3H]WIN 35,428 (86.0 Ci/mmol), [3H]nisoxetine (80.0 Ci/
mmol), and [3H]dopamine (48.2 Ci/mmol) were obtained from
Dupont-New England Nuclear (Boston, MA). [3H]Citalopram
(85.0 Ci/mmol) was from Amersham Pharmacia Biotech Inc.
(Piscataway, NJ ). WIN 35,428 naphthalene sulfonate was
purchased from Research Biochemicals, Inc. (Natick, MA). (-)-
Cocaine HCl was obtained from the National Institute on Drug
Abuse. GBR 12909 dihydrochloride (1-[2-[bis(4-fluorophenyl)-
methoxy]ethyl]-4-[3-phenylpropyl]piperazine) was purchased
from Sigma-Aldrich (St. Louis, MO).
2-[1-(4-F lu or oben zyl)-1,2,3,6-tetr a h yd r op yr id in -4-yl]-
eth a n ol (3). 2-Pyridin-4-ylethanol 2 (3.0 g, 25 mmol) and
4-fluorobenzyl chloride (3.89 g, 26.8 mmol) were dissolved in
15 mL of dry methanol and refluxed for 24 h. Methanol was
removed in vacuo and dried for 2 h under high vacuum. The
residue was dissolved in 50 mL of dry methanol and cooled in
an ice bath. NaBH4 (1.57 g, 40.0 mmol) was then added very
slowly portionwise over a period of 1.5 h, and the solution was
gradually brought to room temperature. The reaction was
quenched with water after stirring for 4 h, and the methanol
was removed in vacuo. The residual product was dissolved in
ethyl acetate, washed with water, dried (Na2SO4), and con-
centrated. The crude product was purified by column chroma-
tography over silica gel (Hex/EtOAc/MeOH ) 20/10/2.5) to
produce 3: 2.87 g (50.0% yield) as a colorless oil.
F igu r e 5. Effects of cocaine (10 min pretreatment interval
(PT); n ) 11), (()-5 (20 min PT); n ) 7), (+)-5 (20 min PT; n
) 7), and (-)-5 (20 min PT, n ) 7) on the percentage of cocaine-
lever responding (top panel) and rate of responding (bottom
panel) in mice trained to discriminate 10 mg/kg cocaine from
saline. At a dose of 56 mg/kg of (-)-5 only five mice were tested
as two mice had died from nondrug related causes. At a dose
of 100 mg/kg (-)-5, only three mice were tested and two of
those failed to respond. Therefore, the % cocaine lever re-
sponding data point for 100 mg/kg (-)-5 represents the data
from one mouse and the % control responding represents the
mean data from three mice. Additional mice were not tested
as convulsions were observed in the two mice that did not
respond. The data points above “S” indicate the mean percent-
age of cocaine-lever responding and rate of responding follow-
ing the administration of saline.
1H NMR (CDCl3, 400 MHz): δ 2.09 (2H, brm, H-3), 2.23 (2H,
t, J ) 5.6 Hz, CH2CH2OH), 2.54 (2H, t, J ) 5.6 Hz, H-2), 2.93
(2H, brm, H-6), 3.51 (2H, s, CH2Ar) 3.66 (2H, t, J ) 5.6 Hz,
CH2OH), 5.46 (1H, brm, CHd), 6.98 (2H, t, J ) 9.2 Hz, ArH,
ortho to F), 7.25-7.29 (2H, m, ArH).
4-(2-Ben zh yd r yloxyet h yl)-1-(4-flu or ob en zyl)-1,2,3,6-
tetr a h yd r op yr id in e (4). Benzhydrol (4.04 g, 22.0 mmol),
2-[1-(4-fluorobenzyl)-1,2,3,6-tetrahydropyridin-4-yl]ethanol 3
(2.87 g, 12.2 mmol), and p-toluenesulfonic acid (2.90 g, 15.3
mmol) were mixed together in 160 mL of toluene, and the
solution was heated to reflux under azeotropic distillation
conditions for 3 h under nitrogen. Toluene was removed in
vacuo, and the residue was partitioned between ether and
saturated NaHCO3 solution. The ether layer was separated,
and the aqueous layer was further extracted thrice with ether.
Combined organic layers were dried (Na2SO4) and concen-
trated to give the crude product which was chromatographed
over silica gel (EtOAc/hexane ) 1/5) to give 4, 2.26 g (46%
yield) as a colorless oil.
5. In locomotor activity studies, compounds R,R-(+)-5
and (()-5 stimulated activity during the 4-h test session
and for hours longer compared to cocaine. All three
compounds, (+)-5, (-)-5, and (()-5, within the dose
range tested partially (50%) but incompletely (80%)
produced cocaine-like responses in mice trained to
discriminate 10 mg/kg ip cocaine from vehicle. Com-
pound (-)-5 was distinctive in this regard in that, unlike
(+)-5 and (()-5, it did not affect locomotor activity yet,
but similar to them, was able to engender (albeit
incompletely) cocaine-like responding. The inactivity of
(-)-5 in locomotor activity tests is unlikely attributable
to poor entry into the CNS because it reduced response
rates and produced (incompletely) cocaine-like respond-
ing at similar doses to (()-5 during cocaine discrimina-
tion tests. Thus, a close correlation exists between in
vitro and in vivo activities of the racemic hydroxy
compound and its enantiomers.
1H NMR (CDCl3, 400 MHz): δ 2.12 (2H, bs, H-3), 2.37 (2H,
t, J ) 6.4 Hz, CH2CH2O), 2.55 (2H, t, J ) 5.6 Hz, H-2), 2.95
(2H, bs, H-6), 3.53-3.59 (4H, m, CH2Ar and CH2O), 5.38 (1H,
s, CH(Ph)2), 5.45 (1H, s, CHd), 7.02 (2H, t, J ) 8.8 Hz, ArH,
ortho to F), 7.24-7.38 (12H, m, ArH).
Ra cem ic tr a n s-4-(2-Ben zh yd r yloxyeth yl)-1-(4-flu or o-
ben zyl)p ip er id in -3-ol (5). Into a stirred solution of NaBH4
(0.57 g, 15 mmol) in 150 mL of dry THF at 0 °C under N2 was
added dropwise 48% w/w BF3-ether complex (2.00 mL, 15.9
mmol). The cooling bath was removed, and the solution was
allowed to stir for 1 h at room temperature (RT). The mixture
was then cooled in an ice bath. Into the cooled solution was
added dropwise a solution of 4-(2-benzhydryloxyethyl)-1-(4-
fluorobenzyl)-1,2,3,6-tetrahydropyridine 4 (3.00 g, 7.48 mmol)
dissolved in THF (15 mL). The solution was brought back to
room temperature and stirred for an additional 2 h. The
solution was again cooled to 0 °C and H2O (8.5 mL), EtOH
(8.5 mL), and 3 N NaOH solution(6 mL) were added followed
by dropwise addition of 30% H2O2 (4 mL). The reaction mixture
was stirred at 55 °C overnight and then cooled to RT, and THF
was removed in vacuo. The product was partitioned between
water and ethyl acetate, and the organic layer was collected.
Water layer was extracted with ethyl acetate. Combined
organic layers were dried (Na2SO4) and concentrated to give
Exp er im en ta l Section
Analytical silica gel-coated TLC plates (Silica Gel 60 F254
)
were purchased from EM Science and were visualized with
UV light or by treatment with phosphomolybdic acid (PMA).
Flash chromatography was carried out on Baker Silica Gel 40
mM. 1H NMR spectra were routinely obtained on GE-300
MHz and Varian 400 MHz FT NMR. The NMR solvent used

1226 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 7
Ghorai et al.
a crude product which was chromatographed over silica gel
(EtOAc/hexane ) 3/1) to give 5 as a viscous liquid: 2.6 g (83%
yield).
1H NMR (CDCl3, 400 MHz): δ 1.34-1.39 (2H, m, H-4,
H-5ax), 1.59-1.66 (2H, m, H-5eq, CH2CH2O), 1.83 (1H, t, J )
10.4 Hz, H-2ax), 1.89-1.95 (2H, m, CH2CH2O, H-6ax), 2.55
(1H, brs, OH), 2.72 (1H, d, J ) 9.9 Hz, H-6eq), 2.98 (1H, dd, J
) 10.4, 4.0 Hz, H-2eq), 3.42-3.63 (5H, m, H-3ax, CH2O, CH2-
Ar), 5.37 (1H, s, CH(Ph)2), 6.98 (2H, t, J ) 9.3 Hz, ArH, ortho
to F), 7.22-7.35 (12H, m, ArH).
Syn t h esis of 4-(2-Ben zh yd r yloxyet h yl)-1-(4-flu or o-
ben zyl)p ip er id in -3-on e (7). To a stirred solution of oxalyl
chloride 0.50 mL (5.72 mmol) in dry CH2Cl2 (10 mL) at -70
°C was added dropwise dimethyl sulfoxide 0.81 mL (11.45
mmol) in CH2Cl2 (3 mL). The mixture was stirred for 15 min,
and a solution of racemic trans-4-(2-Benzhydryloxyethyl)-1-
(4-fluorobenzyl)-piperidin-3-ol 5 (0.6 g, 1.43 mmol) dissolved
in CH2Cl2 (4 mL) was added dropwise. After stirring for 15
min, triethylamine 3.21 mL (22.91 mmol) was added, and the
reaction mixture was stirred for 10 min and then allowed to
warm to room temperature. Water was added, and the aqueous
layer was extracted twice with CH2Cl2. The combined organic
phase was washed with brine, dried (Na2SO4), and concen-
trated. The crude product was purified by column chromatog-
raphy over silica gel (EtOAc/hexane ) 1/2) to produce 7: 0.29
g (48% yield) as a colorless oil.
1H NMR (CDCl3, 400 MHz): δ 1.47-1.63 (2H, m, H-5, CH2-
CH2O), 2.01-2.07 (1H, m, H-4), 2.22-2.30 (1H, m, CH2CH2O),
2.38-2.44 (1H, m, H-6), 2.47-2.55 (1H, m, H-5), 2.75 (1H, d,
J ) 12.8 Hz, H-2). 2.90 (1H, bd, J ) 11.2 Hz, H-6), 3.19 (1H,
d, J ) 12.8 Hz, H-2), 3.46-3.56 (4H, m, CH2O, CH2Ar), 5.30
(1H, s, CH(Ph)2), 6.99 (2H, t, J ) 8.8 Hz, ArH, ortho to F),
7.20-7.39 (12, m, ArH). The free base was converted into its
oxalate salt. Elemental analysis calculated for (C27H28NO2F‚
(COOH)2) C, H, N.
1H NMR in (CD3OD/D2O, 400 MHz): δ 1.18-1.24 (1H, m,
H-5ax), 1.32-1.41 (2H, m, H-4, CH2CH2O), 1.70-1.76 (1H, m,
H-5eq), 1.79 (1H, t, J ) 10.8 Hz, H-2ax), 1.92 (1H, t, J ) 10.4
Hz, H-6ax), 2.14-2.21 (1H, m, CH2CH2O), 2.77 (1H, bd, J )
10.4 Hz, H-6eq), 2.95 (1H, ddd, J ) 10.8, 3.2, 1.6 Hz, H-2eq),
3.22-3.32 (1H, m, H-3ax), 3.43-3.55 (4H, m, CH2O, CH2Ar),
5.36 (1H, s, CH(Ph)2), 7.03(2H, t, J ) 9.2 Hz, ArH, ortho to
F), 7.19-7.23 (2H, m, ArH), 7.25-7.34 (10H, m, ArH). The
free base was converted into its oxalate salt, mp 166-172 °C.
Analysis calculated for (C27H30FNO2‚(COOH)2‚0.3H2O) C, H,
N.
4,7,7-Tr im et h yl-3-oxo-2-oxa -b icyclo[2.2.1]h ep t a n e-1-
ca r b oxylic Acid 4-(2- Ben zh yd r yloxyet h yl)-1-(4-flu o-
r oben zyl)p ip er id in -3-yl Ester (6). Into a stirred solution of
4-(2-benzhydryloxyethyl)-1-(4-fluorobenzyl)piperidin-3-ol 5 (3.0
g, 7.2 mmol), Et3N (1.40 g, 14.4 mmol), and DMAP (10 mg) in
75 mL of dry CH2Cl2 under nitrogen at 0 °C was added a
solution of (1S)-(-)-camphanic chloride (2.30 g, 10.8 mmol).
The temperature was allowed to rise to RT, and the mixture
was stirred for 3 h. The reaction was quenched with water,
and the product was partitioned between organic and aqueous
layers. The organic layer was collected, and the aqueous layer
was extracted with additional dichloromethane. The combined
organic phase was dried (Na2SO4) and concentrated. The crude
product was purified by column chromatography over silica
gel (EtOAc/hexane ) 1/3) to produce 6: (3.6 g, 84% yield) as a
colorless semisolid.
R a cem ic
cis-4-(2-Ben zh yd r yloxyet h yl)-1-(4-flu or o-
ben zyl)p ip er id in -3-ol (8). To a stirred solution of 4-(2-
benzhydryloxyethyl)-1-(4-fluorobenzyl)piperidin-3-one 7 (0.16
g, 0.38 mmol) in 4 mL of dry methanol at 0 °C was added
NaBH4 portionwise and allowed to stir for 1 h. After quench-
ing, methanol was removed under vacuo. The residue was
dissolved in ethyl acetate and washed with water. The water
layer was extracted with ethyl acetate twice. The combined
organic phase was dried over Na2SO4 and concentrated. The
crude product was purified by chromatography over silica gel
(Hex/EtOAc/Et3N ) 80/40/1.2) to give a mixture of 5 and 8:
0.11 g (66% yield). Eluting first: racemic cis-4-(2-benzhydryl-
oxyethyl)-1-(4-fluorobenzyl)piperidin-3-ol 8: 0.03 g, (26.4%
yield).
1H NMR (CDCl3, 400 MHz): δ 1.43-1.49 (2H, m, H-5),
1.57-1.65 (2H, m, H-4, CH2CH2O-), 1.77-1.84 (1H, m, CH2-
CH2O), 1.92-1.99 (1H, m, H-6ax), 2.12 (1H, bd, J ) 11.2 Hz,
H-2ax), 2.62 (1H, brs, OH), 2.78 (1H, dd, J ) 10.8, 2.4 Hz,
H-6eq), 2.92 (1H, ddd, J ) 11.2, 3.2, 1.6 Hz H-2eq), 3.45-3.56
(4H, m, CH2O, CH2Ar), 3.63 (1H, brm, H-3eq), 5.32 (1H, s, CH-
(Ph)2), 6.99 (2H, t, J ) 9.2 Hz, ArH, ortho to F), 7.21-7.33
(12H, m, ArH). Free base was converted into its oxalate salt,
mp 152-155 °C. Analysis calculated for (C27H30FNO2‚
(COOH)2).0.7 H2O) C, H, N.
Eluting second: racemic trans-5: 0.08 g (73.6% yield).
R a cem ic tr a n s-3-P r op ion yl-4-[2-(d ip h en ylm et h oxy)-
eth yl]-1-[(4-flu or op h en yl)m eth yl]p ip er id in e (9). To a so-
lution of 3-hydroxy-4-[2-(diphenylmethoxy)ethyl]-1-[(4-fluo-
rophenyl)methyl]piperidine (()-5 (0.06 g, 0.15 mmol) and
triethylamine (0.02 g, 0.23 mmol) in CH2Cl2 (2 mL) at 0 °C
was added dropwise propionyl chloride (0.02 g, 0.19 mmol),
and the reaction was allowed to warm to RT. After 1 h, the
same quantities of triethylamine and propionyl chloride were
added. One hour later water was added to quench the reaction,
and the organic layer was separated, dried (Na2SO4), and
evaporated. The residue was chromatographed on silica gel
to yield the title compound as the free base 0.05 g (77% yield).
1H NMR (CDCl3) δ 1.1 (3H, t, J ) 8.1 Hz), 1.26-1.95 (7H,
m), 2.26 (2H, q, J ) 7.2 and 7.5 Hz), 2.71 (1H, d, J ) 11.7 Hz),
3.02 (1H, d, J ) 10.5 Hz), 3.41-3.52 (4H, m), 4.67 (1H, t, J )
9.6 Hz), 5.31 (1H, s), 6.98 (2H, t, J ) 9.3 Hz), 7.2-7.4 (12H,
m). The free base was converted into its oxalate salt. Elemental
analysis calculated for (C30H34NO3F‚(COOH)2‚0.5H2O) C, H,
N.
1H NMR (CDCl3, 400 MHz): δ 0.94 (3H, s, CH3), 1.04 (3H,
s, CH3), 1.11 (3H, s, CH3), 1.25-1.34 (1H, m, H-5ax), 1.43-
1.48 (1H, m, CH2CH2O), 1.63-1.81 (3H, m, H-5eq, H-4, Camp-
CH2), 1.85-2.05 (5H, m), 2.33-2.40 (1H, m, Camp-CH2,), 2.70
(1H, d, J ) 10.8 Hz, H-6eq), 2.99-3.06 (1H, m, H-2eq), 3.39-
3.53 (4H, m, CH2-O, CH2Ar), 4.48-4.86 (1H, brm, H-3), 5.29-
5.40 (1H, s, CH(Ph)2), 6.99 (2H, t, J ) 9.2 Hz, ArH, ortho to
F), 7.20-7.50 (12H, m, ArH).
Separation of diastereoisomers 6: The diastereoisomers
were separated by semipreparative HPLC using a normal
phase column (Nova-Pack Silica 6 µm). The mobile phase used
was 0.4% 2-propanol in hexane with a flow rate of 18 mL/min.
The two fractions were eluted with retention time of 28.64 min
for (+)-isomer and 31.36 min for (-)-isomer. Final purity of
the separated diastereoisomers was checked by an analytical
normal phase column (Nova-Pack Silica 60 Å 4 µm) using the
same mobile phase with a flow rate of 1 mL/min. Pure
diastereomers were eluted at 9.56 and 10.73 min, respectively.
Syn th esis of (+)-5. The camphanic ester fraction I (0.41
g, 0.67 mmol) from HPLC separation of diastereomeric mixture
of 6 was hydrolyzed with 50 mg of K2CO3 in methanol (5 mL)
at room temprature for 12 h. Methanol was removed in vacuo,
and the residue was partitioned between ethyl acetate and
water. The organic layer was separated, and the water layer
was extracted twice with ethyl acetate. Combined organic
layers were dried over Na2SO4 and concentrated. The crude
product was chromatographed over silica gel (EtOAc/hexane
) 3/1) to give (+)5, 0.3 g (76% yield). Optical rotation [R]25
)
D
+19.8 (c 1, MeOH). The free base was converted to its oxalate
salt, mp 167-171 °C. Analysis calculated for (C27H30FNO2‚
(COOH)2) C, H, N.
Syn th esis of (-)-5. Fraction II (0.58 g, 0.96 mmol) from
HPLC was treated with 75 mg of K2CO3 in MeOH to produce
(-)-5, 0.3 g (76% yield), as mentioned above. [R]25D ) -20.3 (c
1, MeOH). The free base was converted to the corresponding
Tr a n sp or ter Bin d in g Assa ys. The affinity of test com-
pounds for the rat DAT, SERT, and NET was assessed by
measuring inhibition of binding of [3H]WIN 35,428, [3H]-
citalopram, and [3H]nisoxetine, respectively, exactly as de-
scribed by us previously.13 Briefly, rat striatum was the source
oxalate salt. mp 168-172 °C Analysis calculated for (C27H30
FNO2‚(COOH)2) C, H, N.
-

High Affinity Ligands for the Dopamine Transporter
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 7 1227
for DAT, and cerebral cortex for SERT and NET. Final [Na+]
was 30 mM for DAT and SERT assays, and 225 mM for NET
assays. All binding assays were conducted at 0-4°C, for a
period of 2 h for [3H]WIN 35,428 and [3H]citalopram binding,
and 3 h for [3H]nisoxetine binding. Nonspecific binding of [3H]-
WIN 35,428 and [3H]citalopram binding was defined with 100
µM cocaine, and that of [3H]nisoxetine binding with 1 µM
desipramine. Test compounds were dissolved in dimethyl
sulfoxide (DMSO) and diluted out in 10% (v/v) DMSO. Addi-
tions from the latter stocks resulted in a final concentration
of DMSO of 0.5%, which by itself did not interfere with
radioligand binding. At least five triplicate concentrations of
each test compound were studied, spaced evenly around the
IC50 value. Radioligand concentrations in DAT, SERT, and
NET assays were 4.0, 3.0, and 1.1 nM, respectively, as
compared with observed Kd values of 21, 3.2, and 2.2 nM.
Dop a m in e Tr a n sp or ter Up ta k e Assa ys. Uptake of [3H]-
DA into rat striatal synaptosomes was measured exactly as
described by us previously.13 Briefly, rat striatal P2 membrane
fractions were incubated with test drug for 8 min followed by
the additional presence of [3H]DA for 4 min at 25 °C.
Nonspecific uptake was defined with 100 µM cocaine. Con-
struction of inhibition curves and dissolution of test compounds
were as described above. The concentration of [3H]DA was 50
nM compared with an observed Km of uptake of 0.2 µM.
Locom otor Activity. Su bjects. Adult male Swiss Webster
mice (Harlan Sprague Dawley, Inc., Indianapolis, IN) weighing
30-35 g were used. Mice were housed five per cage with
continuous access to food and water and were allowed to
acclimate to the vivarium environment one week prior to the
start of any experiment. The mice were housed in an AALAC-
accredited animal facility with a controlled temperature (22-
24 °C) on a 12 h light-dark cycle. All testing occurred during
the light component.
Ap p a r a tu s P r oced u r e a n d An a lyses. The locomotor
activity chambers and procedure have been described in detail
elsewhere.13 Briefly, four commercially obtained, automated
activity monitoring devices each enclosed in sound- and light-
attenuating chambers were used (AccuScan Instruments, Inc.,
Columbus, OH). The interior of each device was divided into
two separate 20 × 20 × 30-cm arenas, permitting the inde-
pendent and simultaneous measurement of two mice. Sixteen
photobeam sensors were spaced 2.5 cm apart along the walls
of the chamber. Nonhabituated mice were administered a dose
of a test compound or vehicle and immediately placed into
automated activity chambers where distance traveled (cm) was
recorded for 24, 10 min periods. A one-factor (dose) ANOVA
was conducted on each drug and its respective vehicle followed
by Dunnett’s posthoc tests when the overall ANOVA was
significant. The alpha level for all comparisons was set at 0.05.
pressing the opposite lever until reliable responding was
obtained under FR20 conditions. Discrimination training
occurred during daily (Mon-Fri) 15 min experimental ses-
sions. Mice were injected with cocaine (10 mg/kg) or saline ip
10 min prior to the start of the session start. The cocaine- and
saline-associated levers were counterbalanced across the mice.
Responses on the injection-appropriate lever resulted in
delivery of the sweetened milk solution. Responses on the
injection-inappropriate lever resulted in resetting the response
requirement on the correct lever. A schedule was used to
determine which injection was administered, with the restric-
tion that the same injection was not given on more than two
consecutive sessions, and over 30 training sessions the number
of saline and cocaine injections were approximately equal.
Testing commenced when (1) a mouse completed the first
fixed-ratio (FFR) on the correct lever on at least 8 or 10
consecutive days; and (2) at least 80% of the total responses
were made on the correct lever during those eight sessions.
Tests were conducted on Tuesdays and Fridays provided that
the mouse completed the FFR on the correct lever during the
most recent cocaine and saline training sessions, otherwise, a
training session was administered. During test days, respond-
ing on either lever was reinforced with milk.
An a lyses. The percentage of responses on the cocaine-lever
was calculated for each mouse during training and test
sessions by dividing the number of lever presses emitted upon
the cocaine-lever by the total number of lever presses emitted
on both levers, and then this quotient was multiplied by 100.
Additionally, rate of responding was calculated for each mouse
by dividing the total number of responses emitted on both
levers by 900 s. Individual cocaine-lever responding percent-
ages and responses per second were then averaged (( SEM).
If a mouse failed to complete a FFR, then its data were
excluded from calculations of mean cocaine-lever responding,
but were included for mean response rate calculations. ED50
values (95% confidence limit) (mg/kg) were calculated for
percent cocaine-lever responding using a sigmoidal dose-
response (variable slope) curve fitting procedure (GraphPad
Prism; San Diego, CA). A log transformation on dose was used.
Dr u gs. Cocaine (U.S. National Institute Drug Abuse) was
dissolved in 0.09% sterile saline. Compounds (()-5, (+)-5, and
(-)-5 were dissolved in 20% w/v hydroxypropyl-â-cyclodextrin
(Cavitron 82003, Cerestar USA, Inc., Hammond, IN). All drugs
were administered by the ip route in a volume of 10 mL/kg.
During drug discrimination testing the following pretreatment
intervals were used: cocaine: 10 min, (()-5: 20 min, (+)-5:
10 min and (-)-5: 10 min.
Sin gle-Cr yst a l X-r a y Diffr a ct ion An a lysis of (-)-5.
C27H31FN+O2‚1/2(C4H4O62-)‚H2O ‚0.106(CH3OH), FW ) 516.49,
orthorhombic space group P212121,
a
)
9.265(1),
b
)
)
18.010(1), c ) 32.932(1) Å, V ) 5495.1(4) Å3, Z ) 8, rcalcd
Dr u g Discr im in a tion . Su bjects. Adult male Swiss Web-
ster mice (Harlan Sprague Dawley, Inc., Indianapolis, IN) were
used. Mice were individually housed and maintained on a 12
h light-dark cycle with continuous access to water. Training
and testing occurred during the light component. Mice were
maintained at 35 ( 5 g by supplemental post-session feedings
of laboratory chow (Harlan Tekled, Madison, WI).
Ap p a r a tu s a n d P r oced u r e. The drug discrimination
chambers and procedure have been described in detail else-
where.13 Briefly, eight standard, light- and sound-attenuated
mouse operant conditioning chambers were used (Med Associ-
ates, Inc., St. Albans, VT; model ENV-307A). Each chamber
was equipped with two response levers separated by a trough
into which a 0.01 mL dipper cup could be presented. A house
light was centered at the top of the front panel and three cue
lights were located above each lever. Control of lights, dipper
presentations, and recording of lever presses were accom-
plished by a microcomputer system (Med-PC software, Med
Associates, Inc., St. Albans, VT).
1.249 mg mm-3, l(Cu KR) ) 1.54178 Å, m ) 0.752 mm-1
F(000) ) 2203, T ) 293 K.
,
A clear colorless 0.72 × 0.16 × 0.015 mm crystal grown from
MeOH nitromethane and water was used for data collection
with a Bruker SMART21 6K CCD detector on a Platform
goniometer. The Rigaku rotating Cu anode source was equipped
with incident beam Gobel mirrors. Lattice parameters were
determined using SAINT21 from 5078 reflections within 5.37<
2q < 127.38. Data were collected to 2q ) 133.7°. A set of 25423
reflections was collected in the ω scan mode. There were 9345
unique reflections. Corrections were applied for Lorentz,
polarization, and absorption effects. The structure was solved
with SHELXTL22 and refined with the aid of the SHELX97
system of programs. The full-matrix least-squares refinement
on F2 used eight restraints and varied 706 parameters: atom
coordinates and anisotropic thermal parameters for all non-H
atoms except the lower occupancy solvate atoms. H atoms were
included using a riding model [coordinate shifts of C applied
to attached H atoms, C-H distances set to 0.96 to 0.93 Å, H
angles idealized, Uiso(H) were set to 1.2 to 1.5 Ueq(C). Final
residuals were R1 ) 0.070 for the 6300 observed data with Fo
> 4s(Fo) and 0.93 for all data. Final difference Fourier
excursions of 0.37 and -0.30 eÅ-3. The asymmetric unit
Mice were initially trained to press one of the two levers at
a fixed-ratio 1 (FR1) schedule of reinforcement in which each
lever press resulted in a 0.01 mL delivery of sweetened
condensed milk. The response requirement was gradually
increased to FR20. Subsequently, the mice were reinforced for

1228 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 7
Ghorai et al.
(10) Singh, S. Chemistry, design, and structure-activity relationship
of cocaine antagonists. Chem. Rev. 2000, 100, 925-1024.
(11) Dutta, A. K.; Xu, C.; Reith, M. E. A. Structure-Activity
Relationship Studies of Novel 4-[2-[Bis(4-fluorophenyl)methoxy]-
ethyl]-1-(3-phenylpropyl)piperidine Analogues: Synthesis and
Biological Evaluation at the Dopamine and Serotonin Trans-
porter Sites. J . Med. Chem. 1996, 39, 749-756.
(12) Dutta, A. K.; Coffey, L. L.; Reith, M. E. A. Highly selective, novel
analogues of 4-[2-(diphenylmethoxy)ethyl]-1-benzylpiperidine for
the dopamine transporter: Effect of different aromatic substitu-
tions on their affinity and selectivity. J . Med. Chem. 1997, 40,
35-43.
(13) Dutta, A. K.; Davis, M. C.; Fei, X.-S.; Beardsley, P. M.; Cook, C.
D.; Reith, M. E. A. Expansion of Structure-Activity Studies in
piperidine analogues of 1-[2-(diphenylmethoxy)ethyl]-4-(3-phe-
nylpropyl)piperazine (GBR 12935) compounds by altering sub-
stitutions in the N-benzyl moiety and behavioral pharmacology
of selected molecules. J . Med. Chem. 2002, 45, 654-662.
(14) Newman, A. H.; Allen, A. C.; Izenwasser, S.; Katz, J . L. Novel
3R-(diphenylmethoxy)tropane analogues: potent dopamine up-
take inhibitors without cocaine-like behavioral profiles. J . Med.
Chem. 1994, 37, 2258-2261.
contains two molecules of the title compound,. an l-(+)-tartrate
anion, two molecules of water and a methanol solvate present
at a partial occupancy of 0.216. The chirality of the title
compound was based on the known chirality of the L-(+)-
tartrate. The primary differences between the two molecules
in the asymmetric unit are the orientations of the (4-fluo-
rophenyl)methyl and the diphenylmethoxyethane moiety with
respect to the piperidine ring where the respective torsion
angles about the bond to the ring varied by 110.0 and 54.2°,
respectively. Tables of coordinates, bond distances and bond
angles, and anisotropic thermal parameters have been depos-
ited with the Crystallographic Data Centre, Cambridge, CB2,
1EW, England.
Ack n ow led gm en t. This work was supported by the
National Institute on Drug Abuse, Grant No. DA 12449
(A.K.D.). We thank Mrs. J anet Berfield and Li J uan
Wang for their help with the binding and uptake assays.
(15) Hsin, L.-W.; Dersch, C. M.; Baumann, M. H.; Stafford, D.; Glowa,
J . R.; Rothman, R. B.; J acobson, A. E.; Rice, K. C. Development
of long-acting dopamine transporter ligands as potential cocaine-
abuse therapeutic agents. Chiral hydroxyl-containing derivatives
of 1-[2-[bis(4-fluorophenyl)methoxy]ehtyl]-4-(3-phenylpropyl)pip-
erazine and 1-[2-(diphenylmethoxy)ethyl]-4-(3-phenylpropyl)-
piperazine. J . Med. Chem. 2002, 45, 1321-1329.
(16) Agoston, G. E.; Wu, J . H.; Izenwasser, S.; George, C.; Katz, J .;
Kline, R. H.; Newman, A. H. Novel N-substituted 3R-[bis(4′-
fluorophenyl)methoxy]tropane analogues: selective ligands for
the dopamine transporter. J . Med. Chem. 1997, 40, 4329-4339.
(17) Lewis, D. B., Matecka, D., Zhang, Y., Hsin, L.-W., Dersch, C.
M., Stafford, D., Glowa, J . R., Rothman, R. B., Rice, K. C.
Oxygenated analogues of 1-[2-(diphenylmethoxy)ethyl]- and 1-[2-
[bis(4-fluorophenyl)methoxy]ethyl]-4-(3-phenylpropyl)pipera-
zines (GBR 12935) and GBR 12909) as potential extended-action
cocaine-abuse therapeutic agents. J . Med. Chem. 1999, 42,
5029-5042.
(18) Glowa, J . R., Fantegrossi, W. E., Lewis, D. B., Matecka, D., Rice,
K. C., Rothman, R. B. Sustained decrease in cocaine-maintained
responding in rhesus monkeys with 1-[2-[bis(4-fluorophenyl)-
methoxy]ethyl]4-(3-hydroxy-3-phenylpropyl)piperazinyl decanoate,
a long-acting ester derivative of GBR 12909. J . Med. Chem. 1996,
39, 4689-4691.
(19) Gorelick, D. A.; The rate hypothesis and agonist substitution
approaches to cocaine abuse treatment. Adv. Pharmacol. (NY)
1998, 42, 995-997.
Su p p or tin g In for m a tion Ava ila ble: Crystal structure
data. This material is available free of charge via the Internet
at http://pubs.acs.org.
Refer en ces
(1) (a) Substance Abuse and Mental Health Services Administration
National Household Survey on Drug Abuse: Main Findings
1997; Department of Health and Human Services: Washington,
DC, 1998. (b) Mathias, R. Cocaine, marijuana and heroin abuse
up, methamphetamine abuse down. NIDA Notes 2000, 15 (3),
4-5.
(2) Carroll, F. I.; Howell, L. L.; Kuhar, M. J . Pharmacotherapies
for Treatment of Cocaine Abuse: Preclinical Aspects. J . Med.
Chem. 1999, 42, 2721-2736.
(3) Ritz, M. C.; Cone, E. J .; Kuhar, M. J . Cocaine inhibition of ligand
binding at dopamine, norpinephrine and serotonin transport-
ers: A structure-activity study. Life Sci. 1990, 46, 635-645.
(4) Ritz, M. C.; Lamb, R. J .; Goldberg, R.; Kuhar, M. J . Cocaine
receptors on dopamine transporters are related to self-admin-
stration of cocaine. Science 1987, 237, 1219-1223.
(5) Wilcox, K. M.; Paul, I. A.; Woolverton, W. L. Comparison between
dopamine transporter affinity and self-administration potency
of local anesthetics in rhesus monkeys. Eur. J . Pharmacol. 1999,
367, 175-181.
(6) Kuhar, M. J .; Ritz, M. C.; Boja, J . W. The dopamine hypothesis
of the reinforcing properties of cocaine. Trends. Neurosci. 1991,
14, 299-302.
(7) Rocha, B. A., Fumagalli, F., Gainetdinov, R. R., J ones, S., Ator,
R., Giros, B., Miller G. W. and Caron, M. Cocaine self-
administration in dopamine transporter knockout mice. Nature
Neurosci. 1998, 1, 132-137.
(8) Walsh, S. L.; Cunningham, K. A. Serotonergic mechanisms
involved in the discriminative stimulus, reinforcing and subjec-
tive effects of cocaine. Psychopharmacology 1997, 130, 41-58.
(9) Carrol, F. I.; Lewin, A. H.; Kuhar, M. J . Dopamine transporter
uptake blockers: Structure activity relationships. In Neuro-
transmitter transporters: Structure and function; Reith, M. E.
A., Eds.; Human Press: Totowa, NJ , 1997; pp 263-295.
(20) Efange, S. M. N.; Michelson, R. H.; Remmel, R. P.; Boudreau,
R. J .; Dutta, A. K.; Freshler, A. Flexible N-methyl-4-phenyl-
1,2,3,6-tetrahydropyridine analogues: Synthesis and monoamine
oxidase catalyzed bioactivation. J . Med. Chem. 1990, 33, 3133-
3138.
(21) Bruker 1995 SMART and SAINT Data Collection and Reduction
Software for the SMART system; Bruker-AXS: Madison, WI.
(22) Sheldrick, G. M. 1997. SHELXTL Version 5.1, Bruker Analytical
X-ray Instruments: Madison, WI.
J M020275K