Design and synthesis of an irreversible dopamine-sparing cocaine antagonist

Article abstract of DOI:10.1016/S0968-0896(02)00244-4

Cocaine is a powerful reinforcer and stimulant that binds to specific recognition sites associated with monoamine transporters in the mammalian brain. The search for a functional antagonist to the addictive properties of cocaine has focused on the discovery of a molecule that can inhibit cocaine binding to the dopamine transporter (DAT) but continue to allow dopamine transport by the DAT. No such dopamine-sparing cocaine antagonist has been reported and it is becoming evident that dopamine-sparing antagonism of the pharmacological effects of cocaine by a classical antagonist may not be possible. Herein we present a new concept for the design of dopamine-sparing cocaine antagonists. A unique approach is utilized to deliver an inhibitor that binds irreversibly to the DAT, then cleaves and leaves behind a small fragment attached to the DAT that blocks access by cocaine but permits dopamine transport. The design of these compounds takes advantage of a cysteinyl sulfhydryl group in the DAT. This group is hypothesized to attack the incoming inhibitor and lead to selective inhibition of the cocaine binding site while sparing dopamine transport. This concept of a mechanism based irreversible dopamine-sparing cocaine antagonist has now been demonstrated to be viable and, as example, the unsaturated 6 showed inhibition of cocaine (63%) at the DAT after 24 h incubation, while at that point considerably less inhibition of dopamine is manifested (23%). In contrast, the epoxide 7 showed a greater inhibition of dopamine reuptake than cocaine binding at 24 h (68% versus 18%).

Full text of DOI:10.1016/S0968-0896(02)00244-4

Bioorganic & Medicinal Chemistry 10 (2002) 3583–3591
Design and Synthesis of an Irreversible Dopamine-Sparing
Cocaine Antagonisty
Peter C. Meltzer,a,* Shanghao Liu,a Heather S. Blanchette,a
Paul Blundella and Bertha K. Madrasb,{
aOrganix Inc., 240 Salem Street, Woburn, MA 01801, USA
bDepartment of Psychiatry, Harvard Medical School and New England Regional Primate Research Center,
Southborough, MA 01772, USA
Received 21 January 2002; accepted 8 May 2002
Abstract—Cocaine is a powerful reinforcer and stimulant that binds to specific recognition sites associated with monoamine
transporters in the mammalian brain. The search for a functional antagonist to the addictive properties of cocaine has focused on
the discovery of a molecule that can inhibit cocaine binding to the dopamine transporter (DAT) but continue to allow dopamine
transport by the DAT. No such dopamine-sparing cocaine antagonist has been reported and it is becoming evident that dopamine-
sparing antagonism of the pharmacological effects of cocaine by a classical antagonist may not be possible. Herein we present a new
concept for the design of dopamine-sparing cocaine antagonists. A unique approach is utilized to deliver an inhibitor that binds
irreversibly to the DAT, then cleaves and leaves behind a small fragment attached to the DAT that blocks access by cocaine but
permits dopamine transport. The design of these compounds takes advantage of a cysteinyl sulfhydryl group in the DAT. This
group is hypothesized to attack the incoming inhibitor and lead to selective inhibition of the cocaine binding site while sparing
dopamine transport. This concept of a mechanism based irreversible dopamine-sparing cocaine antagonist has now been demon-
strated to be viable and, as example, the unsaturated 6 showed inhibition of cocaine (63%) at the DAT after 24 h incubation, while
at that point considerably less inhibition of dopamine is manifested (23%). In contrast, the epoxide 7 showed a greater inhibition of
dopamine reuptake than cocaine binding at 24 h (68% versus 18%).
# 2002 Elsevier Science Ltd. All rights reserved.
Introduction
this increase in available DA then causes activation of
postsynaptic DA receptors. Excess stimulation of post-
synaptic DA receptors then leads to the well-known
sequellae of cocaine pharmacology. It has therefore
been a goal of research to discover a molecule that can
inhibit cocaine binding to the DAT but continue to
allow DA transport by the DAT. While the cocaine
inhibitor would still bind to the DAT, it would not
affect the rate of reuptake of DA by this transporter. In
this manner, the concentration of DA in the synapse
would remain at normal physiological levels.
The widespread abuse of cocaine has led to considerable
societal problems. It is a powerful reinforcer and stim-
ulant that binds to specific recognition sites associated
with monoamine transporters in the mammalian
brain.1À6 The search for a functional antagonist to the
addictive properties of cocaine has been the focus of
considerable research, and the dopamine transporter
hypothesis has guided much of this work.2,7,8 Accumu-
lating evidence supports the view that the addictive
properties of cocaine derive from inhibition of dopa-
mine reuptake by the presynaptic dopamine transporter
(DAT) (Fig. 1). Such inhibition leads to an increase of
dopamine (DA) concentration within the synapse, and
yA communication describing the biology of these compounds is being
submitted concurrently to Molecular Pharmacology.
*Corresponding author. Tel.: +1-781-932-4142; fax: +1-781-933-6695;
e-mail: meltzer@organixinc.com
{Corresponding author, biology.
Figure 1. Cocaine and dopamine.
0968-0896/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(02)00244-4

3584
P. C. Meltzer et al. / Bioorg. Med. Chem. 10 (2002) 3583–3591
The search for a cocaine antagonist, or cocaine repla-
cement, has been dominated by an exploration of six
main classes of compounds.9,10 These include the
tropanes,11À19 GBR analogues,20,21 methylphenidate
analogues,22,23 cocaine analogues,24À26 mazindol,27 and
diarylmethoxy tropanes.17,28,29 To date there has been
no discovery of a ‘dopamine-sparing cocaine antago-
nist’.30 Indeed, dopamine-sparing antagonism of the
pharmacological effects of cocaine by a classical
antagonist may not be possible. A consideration of the
difference between antagonism of a receptor versus
antagonism of a transporter, such as the DAT, is
instructive. A receptor agonist relies upon binding of the
agonist to the biological macromolecule. Once bound,
the agonist activates the receptor and this activation
leads to the biological and pharmacological sequellae. A
classical receptor antagonist binds to the receptor, but
cannot activate it. In analogy, DA activates post-
synaptic receptors by first binding and then activating
those receptors. The DA excess within the synapse is
removed by the DAT. Cocaine binds to the DAT and
inhibits this DA reuptake. Activation of postsynaptic
DA receptors caused by the inability to remove DA
from the synapse then leads to cocaine pharmacology.
Therefore, inhibition of DA reuptake may be a neces-
sary and sufficient cause of cocaine pharmacology.
Consequently a cocaine antagonist is not simply a
molecule that will inhibit cocaine binding at the DAT,
but more importantly, it will permit transport of
synaptic DA to the presynaptic neuron.
of the channel. Consequently, any ligand passing into
this channel is confronted by the side-chain face of the
relevant helices. Dopamine is conceived to pass into,
and through, this channel into the presynaptic neuron.
Ligands (e.g., cocaine) are conceived to pass into the
channel, bind across transmembrane domains and block
entry (binding) of any other potential ligands or substrates
(including DA). Blockade by ligand binding will thus
prevent binding of any competing ligand or substrate
(DA) whether or not the competing ligand or substrate
binds at the identical molecular acceptor site or not.
Therefore any molecule that can enter and bind to the
12-membered transmembrane channel and is not trans-
ported, may inhibit dopamine uptake. This implies that
it may not be possible to develop a dopamine-sparing
cocaine antagonist that is of molecular dimensions and
topology sufficient to obstruct physical passage of dopa-
mine itself. Therefore, notwithstanding the probability
that different ligands bind at different acceptor sites
within the DAT, each ligand that blocks DA reuptake
will be unable to function as a dopamine-sparing cocaine
antagonist. A new design of an antagonist is required.
Herein we present a different concept for dopamine-
sparing cocaine antagonists. It utilizes
a unique
approach to deliver a small non-tropane irreversible
DAT inhibitor that blocks DAT access by cocaine but
permits DA transport. These compounds are based on
the premise that to obtain cocaine antagonism without
dopamine inhibition, the bulk of the antagonist should be
removed from the binding site after an interaction that
renders the dopamine transporter unavailable for cocaine,
but readily available for dopamine transport. Thus selec-
tive and potent compounds bind to the dopamine
transporter but, once bound, interact covalently with a
proximate amino acid. The bulk of the parent guiding
tropane dissociates, leaving a small residue attached.
In recognition of this, much research has focused on the
design of DAT inhibitors that differentiate between
dopamine reuptake and cocaine inhibition.31 Deutsch
and Schweri32 have defined the discrimination ratio
(DR) to differentiate between dopamine reuptake and
cocaine inhibition. This is the ratio of the IC50 for the
test compound for inhibition of [3H]dopamine reuptake
divided by the corresponding IC50 of [3H]WIN 35,428
binding inhibition to the DAT. Since small differences
(DR <10) could be accounted for in terms of assay
difficulties, it was proposed that a ratio of greater than
10 may signal a potential cocaine antagonist. While the
discrimination ratio has been a useful tool, it is becoming
apparent that it does not necessarily indicate the potential
for pharmacological antagonism. Numerous compounds
that manifest a DR greater than 10 have been prepared
and none have proven to be dopamine-sparing cocaine
antagonists. We have prepared compounds that manifest
DRs of 10–137 but none have proven to be antagonists
of the in vivo pharmacological effects of cocaine.
Although a binding site for dopamine has been pro-
posed [(Asp79) in the transmembrane domain 1 (TMD 1)
and at two serine residues in the transmembrane
domain 7 (TMD 7)],30 it is still unclear where cocaine
and the tropane analogues bind on the DAT. SAR
studies have indicated that it is probable that the dis-
parate DAT ligands (cocaine, 3-aryltropanes, 8-oxa-3-
aryltropanes, 8-carba-3-aryltropanes, GBR compounds,
methyl phenidate, mazindol) occupy different acceptor
sites within the DAT. Consequently the design of an
irreversible mechanism-based cocaine antagonist of the
DAT has to rely on the probable availability of amine
(e.g., lysine) or sulfhydryl (e.g., cysteine) groups near
the cocaine acceptor site. This first generation of
mechanism-based cocaine antagonists is designed to
utilize available cysteine residues on the DAT.
The DAT has been cloned33À35 and the binding site for
dopamine described.30 The hDAT is comprised of 620
amino acids putatively arranged in 12 interconnected
helices. However the tertiary structure is unknown. In
1997 we proposed a working model for the DAT as a
transmembrane channel, comprised of the 12 surround-
ing helices, that provides an active channel through
which DA is transported intracellularly.13 Each helical
domain presents its polar (amide) linkage toward the
core of the helix. This places the amino acid side chain
on the outside of the helix facing the internal structure
There are 13 cysteine residues in the hDAT. Six of these
are located within the putative helices of the DAT while
seven are in the connecting amino acid chains.36 There-
fore, it is likely that cysteine residues are available in the
vicinity of the cocaine acceptor site on the DAT. Cao et
al.37 pointed out that the cocaine receptor has an active
nucleophilic thiol, and showed that N-ethylmaleimide,
an excellent sulfhydryl trap, inhibited [3H]dopamine

P. C. Meltzer et al. / Bioorg. Med. Chem. 10 (2002) 3583–3591
3585
uptake totally but [3H]cocaine binding only partially.
Results
Schweri showed that N-ethylmaleimide irreversibly
inhibits binding of methylphenidate to the stimulant
recognition site38 and Reith later showed39 that cocaine
and dopamine protect different sites on the DAT from
N-ethylmaleimide alkylation. Therefore not only are
cysteine residues important to DAT function, but it is
likely that different cysteines are proximate to the
dopamine and cocaine binding sites. Consequently, our
design of an irreversible antagonist takes advantage of a
reactive cysteinyl sulfhydryl group in the region of the
cocaine binding site. While cysteine can provide a
unique opportunity for design of mechanism-based
antagonists of the DAT, it also presents substantial
pharmacological challenges. This is because cysteine is
ubiquitous in proteins. As an example, glutathione,
present in large amount in vivo, may preempt the
delivery of a thiol sensitive ligand to the actual site of
interest. Consequently, we sought to design a molecule
that would be activated to thiol attack primarily within
the site of interest in a pseudo-intramolecular inter-
action with the available sulfhydryl moiety of cysteine.
While sulfhydryl nucleophile-accepting moieties such as
maleimides, a,b-unsaturated systems, and isothio-
cyanates, may be appropriate functionality for this
concept, these reactive functional groups may also be
expected to interact irreversibly with amino acid resi-
dues in unrelated biological structures. Consequently
careful control of the reactivity of these sulfhydryl
acceptors is required here. Interaction of the sulfhydryl
acceptor with the cysteine residue at the DAT should
preferably be triggered once it is localized within the
DAT acceptor site. We initially selected an epoxide as
the sulfhydryl nucleophile acceptor, but subsequently
found that a terminal double bond provided the oppor-
tunity via free radical initiated chemistry.
Chemistry
The synthetic route to the target compounds is pre-
sented in Scheme 1. Thus the 8-oxabicyclo[3.2.1]octene
313 served as precursor for the 3a-aryl target com-
pounds 7–10. Reduction of the 2,3-ene 3 with samarium
iodide provided both the 3a-aryl and 3b-aryl com-
pounds. The 3a-aryl compound 4 was used here since
3a-aryl compounds are more DAT selective than are the
3b-aryl analogues.40 Reduction of the ester with lithium
aluminum hydride in THF then provided the alcohol 5.
Reaction of the alcohol, under basic conditions with
selected acid chlorides and alkyl or alkenyl bromides,
provided the esters 6, 9 and 10 and ether 8 in good yield.
The epoxide 7 was obtained by oxidation of the unsat-
urated ester 6 with m-chloroperbenzoic acid in methyl-
ene chloride.
Biology
The affinities (IC50) of the compounds for the dopamine
and serotonin transporters were determined in competi-
tion studies using [3H]2b-carbomethoxy-3b-(4-fluoro-
phenyl)-8-methyl-8-azabicyclo[3.2.1]octane ([3H]WIN
35,428) to label the dopamine transporter and [3H]cita-
lopram to label the serotonin transporter. Binding data
for the compounds are presented in Table 1. Competi-
tion studies were conducted with a fixed concentration
of radioligand and a range of concentrations of the test
compound. All compounds inhibited [3H]WIN 35,428
and [3H]citalopram binding in a concentration-depen-
dent manner. The six compounds inhibit [3H]WIN
35,428 binding to the DAT with nM potency (17–75 nM)
and are reasonably selective versus the SERT (7-fold to
30-fold). In our assay of potential functional antagon-
Scheme 1. Synthesis of 2-substituted 8-oxabicyclo[3.2.1]octanes. Reagents and conditions: (i) Na(TMS)2N, Ph(Tf)2N, THF, À78 ꢀC; (ii) ArB(OH)2,
Pd2dba3, Na2CO3, LiCl; (iii) SmI2, methanol, À78 ꢀC; (iv) LAH, THF, 100%; (v) Et3N, RCOCl, 51%; (vi) THF, NaH, RBr, 30%; (vii) RCOCl,
Et3N, CH2Cl2, 12%; (viii) Et3N, RCOCl, CH2Cl2, 76%; (ix) mCPBA, CH2Cl2, 54%.

3586
P. C. Meltzer et al. / Bioorg. Med. Chem. 10 (2002) 3583–3591
ism, the hDAT was incubated with each of the com-
pounds 6–10 for a period of 24 h. The cells were then
exhaustively washed to remove non-covalently bound
ligand and binding of cocaine (20 nM) and dopamine
uptake (Vmax) were compared.
uptake at the DAT, may also block dopamine. There-
fore it can have the same, or similar pharmacological
effects as cocaine itself, that is, increase dopamine con-
centration in the synapse. There has been a significant
effort to discover compounds that bind potently and
selectively to the DAT but that bind allosterically in the
hope that such compounds would inhibit cocaine
uptake but not dopamine reuptake. To date this
approach to dopamine-sparing cocaine antagonists has
failed.
The controls selected for this study, WIN 35,428 and
(À)-cocaine did not bind irreversibly to the DAT since
inhibition of both DA uptake and cocaine binding after
24 h incubation was 90% or more in both cases (Table
1). Therefore, washout was essentially complete. The
compounds of this study may be expected to have simi-
lar lipophilicity and were therefore expected to be
washed out similarly. In contrast to WIN 35,428 and
(À)-cocaine, the sulfhydryl acceptor, N-ethylmaleimide,
did manifest irreversible binding within the same time
frame (35% inhibition of both functions). The epoxide 7
showed a greater inhibition of dopamine reuptake than
cocaine binding at 24 h (68% vs 18%). Compounds 6
[racemic and enantiomerically pure (1R)-6] significantly
inhibited cocaine binding to the DAT after 24 h (63 and
80%, respectively). However, inhibition of dopamine
was much reduced. Unsaturated ether 8 manifested a
preferred inhibition of cocaine binding to dopamine
uptake (80% vs 40%) and therefore bound irreversibly
and allowed substantial dopamine uptake. Saturated
ester 9 showed no inhibition of binding or uptake;
therefore no irreversible binding took place. In contrast,
saturated ester 10 showed similar and substantial inhi-
bition of DA uptake and cocaine binding after 24h
(60%). This latter result is surprising, and compound 10 is
currently under further investigation in our laboratories.
In contrast to this approach, the molecules presented
here are based on the premise that cocaine antagonism
without dopamine inhibition can be achieved if the bulk
of the antagonist is removed from the binding site after
an interaction that renders the site unavailable for
cocaine, but available for dopamine trafficking. To this
end we have designed molecules that manifest high
potency and selectivity for the DAT. They possess a
ligand that directs the agent to the binding site on the
DAT. The ligand is attached, via a cleavable tether, to
an acceptor moiety (barb) that can bind covalently to an
amino acid residue (e.g., thiol of cysteine) in the vicinity
of the binding site. The attack of the receptor site based
thiol is then followed by release of the ligand. The barb
remains to perturb, or block, the site locally to cocaine.
The barb should be sufficiently small so as not to block
‘incoming’ dopamine and should also not cause exten-
sive perturbation of the receptor site. This process is
depicted in Figure 2.
The selection of appropriate barb functionality is com-
plicated by the relative kinetics required in order for the
delivery, binding, reaction, release, and washout to
occur in sequential order. Specifically, the rate of inter-
action of the barb with a thiol must be slower than the
rate of delivery of the ligand to the site of action itself as
well as binding to that acceptor site. However, once
within the binding site, the attack upon the incoming
ligand should be efficient in order to avoid washout of
the ligand prior to covalent binding. The subsequent
cleavage of the ligand should be slower than the inter-
Discussion
To date, no cocaine antagonists have been reported,
although a significant effort has been directed toward
their discovery.9,10,18,26,28,31,41À43 The fundamental diffi-
culty arises from the fact that cocaine is a dopamine
antagonist at the DAT: it blocks dopamine reuptake.
Consequently, any other ligand that blocks cocaine
Table 1. Comparison of affinity of compounds for the dopamine transporter (DAT) and serotonin transporter (SERT) and the effects on dopamine
transport and cocaine bindinga
Compound
Dopamine
transporter
Serotonin
transporter
Selectivity
SERT/DAT
Inhibition of DA
uptake after 24 h
(%)
Inhibition of cocaine
binding after 24 h
(%)
IC50 (nM)
[3H]WIN 35,428
[3H]citalopram
WIN 35,428
(À)-Cocaineb
N-Ethylmaleimide
6, O-1893 (1R/S)
6, O-2185 (1R)
7, O-1834
8, O-2153
9, O-2102
10, O-2059
11
95
160
270
>100,000
593
326
15
2
2
12
>100,000
—
30
14
17
13
13
7
35
23
48
68
40
0
35
63
80
18
80
0
20
24
17
58
71
75
292
755
917
515
60
60
aEffects of 24-h pre-incubation of the test compounds on [3H]dopamine transport and [3H]cocaine (20 nM) bound in HEK-293 cells transfected with
the human dopamine transporter. The cells were extensively washed prior to conducting binding and transport assays. Errors generally do not
exceed 15% between replicate experiments. Highest doses tested were generally 10–100 mM. Results are expressed as % of control values and are the
means of 2–5 determinations, each conducted in triplicate.
bCocaine washes out completely within minutes.

P. C. Meltzer et al. / Bioorg. Med. Chem. 10 (2002) 3583–3591
3587
Figure 2. The approaching antagonist binds and leaves the barb behind, covalently bound to the receptor.
action with the cysteine in order to avoid cleavage of the
molecule prior to delivery, binding, and covalent
attachment at the active site. However, this cleavage,
once within the DAT acceptor site, should be complete,
with release of the tropane moiety. The washout rate of
the released tropane moiety will then determine the
onset time of the dopamine sparing cocaine antagonist
as well as its efficacy. The duration of action of the
successful cocaine antagonist will then be determined by
the relative turnover rate of the DAT itself. These con-
siderations have guided our selection of a suitable barb
and ligand.
While we had clearly outlined the conceptual approach
toward the design of these ‘Trojan Horse’ molecules, it
is interesting that the molecules that have now become
most exciting from this work were obtained serendipi-
tously! We feel it instructive to describe the events that
have led to this development in order to communicate
not only the excitement of science, but also the some-
time reality! An immediate synthetic precursor to the
epoxide 7 was the ene 6. While we initially had little
expectation for 6 since reaction with a nucleophile is
unprecedented, we submit all advanced intermediates
for biological evaluation and 6 was similarly included.
We were therefore incredulous when this compound
showed attributes of
a dopamine-sparing cocaine
The ligand
antagonist! In fact, the racemic compound 6 manifested
a 23% inhibition of dopamine uptake compared with a
robust 63% inhibition of cocaine binding after incu-
bation for 24 h and extensive attempts to wash out any
non-bound compounds. The enantiomerically pure
(1R)-6 inhibited cocaine binding by 80% and dopamine
uptake by 48% after 24 h incubation. The (1S)-6 ana-
logue was anticipated to manifest weaker binding to
the DAT and has therefore not been explored at this
time.
The ligand serves as a ‘honing’ device to guide the
molecule specifically and selectively to the dopamine
transporter. The 8-oxabicyclo[3.2.1]octane family has
provided an array of DAT binding agents.13 Herein we
used 2b-carbomethoxy-3a-(30,40-dichlorophenyl)-8-oxa-
bicyclo[3.2.1]octane, a potent (DAT IC50=2.34 nM)
and reasonably selective (SERT IC50=31 nM) com-
pound, for this first generation of irreversible dopamine-
sparing cocaine antagonists.
We immediately explored the possibility that the term-
inal double bond in 6 acts as a sulfhydryl acceptor.
Indeed, when 6 was stirred with ethanethiol (EtSH) in
THF with either TsOH or with acetic acid at room
temperature for 75 h, about 50% of it was consumed
(Scheme 2). The product of addition, 11, as well as the
alcohol 5, were obtained. Since 11 results from anti-
Markownikoff addition of EtSH to the terminal double
bond in 6, the addition likely occurred via a free radical
mechanism with O2 (air) in the solvent as initiator.
Indeed, when the above reactions were conducted under
an N2 atmosphere in an oxygen free solvent, no addition
or cleavage occurred. The addition could also be facili-
tated by the free radical initiator AIBN. Thus, when 6
was stirred with diprotected cysteine 12 in acetonitrile in
the presence of AIBN, 13 was obtained in 50% yield.
Compound 5 was not detected in the crude product; in
this case there was no acid in the reaction mixture to
catalyze cleavage.
The tether
The C2-position of the 8-oxabicyclo[3.2.1]octanes pro-
vides an excellent opportunity for introduction of our
tether. An ester function was initially selected as the
cleavable link.
The barb
Our initial design of a barb was based upon the premise
that nucleophilic attack by a cysteinyl sulfhydryl group
on an epoxide (possibly catalyzed by aspartic acid)
would cause a cascade of reactions to form a five-mem-
bered lactone attached to the cysteine. We had antici-
pated acceleration of this reaction in the pseudo-
intramolecular situation in which the ligand is tightly
held in close proximity to the nucleophile. Therefore, the
first barb selected was an epoxide. However, the epoxide
7 manifested a greater inhibition of dopamine reuptake
than of cocaine binding. Apparently compound 7 does
not guide the ligand to the cocaine binding site, but
prefers binding to the dopamine site, much as observed
by Cao37 in his experiments with N-ethylmaleimide.
Notwithstanding, the idea of a ‘honing device’ had suc-
ceeded, albeit to the undesired site.
The fact that 6 reacts with a thiol supports the possibi-
lity that the sulfhydryl group of a cysteine in the DAT
could also add to the terminal double bond in 6 under
appropriate conditions such as the presence of oxygen
or the presence of an hydroxyl radical. Furthermore, we

3588
P. C. Meltzer et al. / Bioorg. Med. Chem. 10 (2002) 3583–3591
Scheme 2. Anti-Markownikoff addition of a thiol to 6a. Reagents and conditions: (i) EtSH, THF, TsOH or CH3COOH, 75 h, 22 ꢀC, 12%; (ii) 12,
AIBN, CH3CN, 4 h, 50%.
demonstrated that the ester in 6 is relatively stable prior
to the radical attack by the thiol, and are therefore
hopeful that this will be the case in vivo as well.
Future work will be focused on the optimization of lead
compounds and the generalization of the concept to
other templates. Finally, the mechanism by which
unsaturated compounds are attacked by sulfhydryl
radicals is well documented in the literature. Notwith-
standing, at this time we cannot be confident that this
mechanism pertains in the biological assays nor in vivo.
We are currently investigating the biological fate of
these compounds in an effort to better understand their
mechanism of action.
To evaluate this further we prepared the unsaturated
ether 8 and the saturated esters 9 and 10. Compound 8
was anticipated to bind irreversibly by radical chemistry
but not to cleave. Surprisingly, it also proved to be a
dopamine-sparing antagonist (Table 1). Compound 9,
which clearly cannot react with a sulfhydryl radical does
not bind irreversibly as evidenced by the complete
washout at the 24 h period. Compound 10, in stark
contrast, inhibits both dopamine uptake and cocaine
binding equally notwithstanding the absence of a term-
inal double bond. These results are currently under
investigation in our laboratories.
Experimental
NMR spectra were recorded in CDCl3 on a JEOL
300 NMR spectrometer operating at 300.53 MHz for
1H, and 75.58 MHz for 13C. Tetramethylsilane (TMS)
was used as internal standard. Melting points are
uncorrected and were measured on a Gallenkamp melt-
ing point apparatus. Thin layer chromatography (TLC)
was carried out on Baker Si250F plates. Visualization
was accomplished with either UV exposure or treatment
with phosphomolybdic acid (PMA). Flash chromato-
graphy was carried out on Baker Silica Gel 40 mM or by
radial chromatography on a Chromatotron. Elemental
analyses were performed by Atlantic Microlab, Atlanta,
GA. All reactions were conducted under an inert (N2)
atmosphere. Coupling constants (J ) are reported in Hz.
We suggest that irreversible inhibition of the cocaine
(but not dopamine) binding site by 6 on the DAT may
occur as follows. The first step is non-covalent binding
of the ligand 6 to the acceptor site on the DAT. Since
the ligand is now tightly bound, this is followed by
pseudo-intramolecular attack by the sulfhydryl radical
of a proximal cysteine, upon the double bond, to pro-
vide the covalently bound DAT–ligand complex. This
complex then undergoes cleavage of the ester and relea-
ses 2-hydroxmethyloxabicyclo[3.2.1]octane for washout
from the acceptor site and the DAT itself. The remain-
ing barb on the binding site now serves to differentially
inhibit dopamine uptake and cocaine binding. This
explanation of the experimentally observed antagonism
is entirely speculative at this time. It remains quite pos-
sible that these molecules effect conformational change
of the DAT in such a way as to differentiate dopamine
uptake from inhibition of cocaine binding.
[3ꢀ-(3,4-dichlorophenyl)-8-oxabicyclo[3.2.1]oct-2ꢁ-yl]
methanol (5). Lithium aluminum hydride (LAH)
(3.64 g, 96 mmol) was cooled to 0ꢀC in anhydrous THF
(40 mL). 2b-Carbomethoxy-3a-(3,4-dichlorophenyl)-8-
oxabicyclo[3.2.1] octane, 413 (8.0 g, 25.4 mmol) was dis-
solved in THF (60 mL) and added dropwise to the stir-
red reaction mixture. The reaction was warmed to 22 ꢀC
and stirred for 18.5 h. The slurry was cooled to 0 ꢀC and
the excess LAH was slowly decomposed by addition of
water (25 mL) and refluxing for 20 min. The mixture
was filtered and the solid collected was rinsed with
ether. The filtrate was dried (MgSO4), filtered and con-
densed in vacuo to a viscous clear oil (7.3 g, ca. 100%).
A portion of the clear oil was purified by crystallization
(methylene chloride/hexanes) yielding colorless cubic
Conclusion
Herein we have demonstrated that the concept of a
mechanism based irreversible dopamine-sparing cocaine
antagonist is viable: 6 shows inhibition of cocaine (63%)
at the DAT up to 24 h while at that point only 23%
inhibition of dopamine transport (Vmax) is manifested.

P. C. Meltzer et al. / Bioorg. Med. Chem. 10 (2002) 3583–3591
3589
crystals. Mp 75–76 ꢀC; 1H NMR d 7.36 (d, 1H), 7.31 (d,
1H), 7.06 (dd, 1H), 4.44 (dt, 2H, J=2.8, 8.3 Hz), 3.53
(dt, 2H, J=2.8, 6.4 Hz), 2.61 (m, 1H), 2.36 (m, 1H),
2.22–2.14 (m, 1H), 2.11–1.90 (m, 1H), 1.75–1.62 (m,
3H), 1.47 (dt, 1H, J=4.6, 5.2 Hz), 1.33 (ddd, 1H,
J=2.8, 11.0, 13.8 Hz); 13C NMR d 144.8, 132.2, 130.2,
130.0, 129.6, 127.3, 74.6, 71.9, 64.6, 51.0, 37.6, 35.2,
31.9, 30.8.
was cooled and water (40 mL) was added. This was
extracted with CH2Cl2 (2 ꢁ 50 mL). The extracts were
dried (Na2SO4), combined and evaporated. The residue
was purified by flash column chromatography yielding a
light yellow oil (74 mg, 30%). Rf 0.32 (30% ethyl ace-
1
tate/hexanes); H NMR d 7.35 (d, 1H), 7.29 (d, 1H),
7.04 (dd, 1H), 5.85–5.72 (m, 1H), 5.03–4.92 (m, 2H),
4.45–4.36 (m, 2H), 3.39–3.28 (m, 2H), 3.26–3.16 (m,
2H), 2.59–2.49 (m, 1H), 2.37–2.29 (m, 1H), 2.18–2.03
(m, 3H), 2.00–1.92 (m, 1H), 1.74–1.58 (m, 5H), 1.32–
1.22 (m, 1H); 13C NMR d 145.0, 138.3, 132.3, 130.2, 130.0,
129.8, 127.4, 114.7, 74.8, 72.7, 71.7, 70.5, 49.4, 38.3, 35.8,
32.2, 31.0, 30.3, 28.7. Anal. (C19H24Cl2O2) C, H, Cl.
Pent-4-enoic acid-3ꢀ-(3,4-dichlorophenyl)-8-oxabicyclo
[3.2.1]oct-2ꢁ-ylmethyl ester (6). The crude alcohol 5
(5.44 g, 18.9 mmol) was stirred with dry Et3N (5 mL) in
anhydrous CH2Cl2 (75 mL) under an N2 atmosphere.
Pentenoyl chloride (2.6 mL, 23.7 mmol) was added
slowly via syringe. The reaction mixture was stirred for
15.5 h, filtered through a pad of silica and concentrated
under vacuum to yield an orange oil. The crude product
was purified by radial chromatography (6 mm, 20%
ether/hexanes) to give a yellow oil (6.71 g, 96%). Rf 0.63
(20% ethyl acetate/hexanes). A portion of the crude oil
(3.71 g, 10.0 mmol) was crystallized from hexanes to
Pentanoic acid-3ꢀ-(3,4-dichlorophenyl)-8-oxabicyclo[3.2.1]
oct-2ꢁ-ylmethyl ester (9). The crude alcohol 5 (0.70 g,
2.43 mmol) was dissolved in dry CH2Cl2 (20 mL) and
treated with Et3N (1.0 mL, 7.2 mmol) and valeryl chlor-
ide (0.6 mL, 5.1 mmol). The reaction solution was stir-
red under an N2 atmosphere at 22 ꢀC for 16 h. It was
then diluted with CH2Cl2 (50 mL) and extracted with
saturated NaHCO3 solution (50 mL) and brine (50 mL).
The organic phase was condensed and purified by radial
chromatography (2 mm plate, 10% ethyl acetate/hex-
anes) to give a clear, colorless oil (0.70 g, 78%). The oil
was crystallized from hexanes to yield colorless crystals
(0.11 g, 12%). Rf 0.50 (20% ethyl acetate/hexanes). Mp
yield colorless plates (1.91 g, 51%). Mp 49–50 ꢀC; H
1
NMR d 7.34 (d, 1H), 7.29 (d, 1H), 7.04 (dd, 1H), 5.89–
5.72 (m, 1H), 5.01 (ddd, 2H, J=1.6, 8.8, 17.3 Hz), 4.45
(ddd, 1H, J=2.48, 6.33, 8.8 Hz), 4.26 (d, 1H,
J=7.7 Hz), 3.95 (d, 2H, J=6.1 Hz), 2.61 (m, 1H), 2.40–
2.30 (m, 4H), 2.22–1.91 (m, 2H), 1.84–1.80 (m, 1H),
1.73–1.63 (m, 2H), 1.30 (ddd, 1H, J=2.5, 11.3, 13.8 Hz);
13C NMR d 173.0, 144.2, 136.52, 132.4, 130.4, 130.3,
129.7, 127.3, 115.5, 74.2, 66.3, 47.9, 38.1, 36.0, 33.3, 32.0,
30.9, 28.7. Anal. (C19H22O3Cl2) C, H.
26 ꢀC; H NMR d 7.36 (d, 1H), 7.29 (d, 1H), 7.06 (dd,
1
1H), 4.44 (ddd, 1H, J=2.5, 6.6, 8.8 Hz), 4.27 (d, 1H,
J=7.7 Hz), 3.95 (d, 2H, J=6.1 Hz), 2.59 (dt, 1H, J=4.0,
6.9 Hz), 2.39–2.29 (m, 1H), 2.24–1.46 (m, 9H), 1.36–1.22
(m, 3H), 0.89 (t, 3H, J=7.2 Hz); 13C NMR d 173.7,
144.2, 132.3, 130.3, 130.2, 129.6, 127.2, 74.2, 71.7, 66.1,
47.87, 38.1, 36.0, 33.8, 31.9, 30.8, 26.8, 22.1, 13.6. Anal.
(C19H24Cl2O3) C, H.
3-Oxiranyl-propionic acid 3ꢀ-(3,4-dichlorophenyl)-8-oxa-
bicyclo[3.2.1]oct-2ꢁ-ylmethyl ester (7). Alkene 6 (3.4 g,
9.2 mmol) in dry CH2Cl2 (100 mL) was treated with
mCPBA and stirred at 22 ꢀC for 19 h. Excess mCPBA
was quenched with Na2S2O3 (3 g) and stirred with water
(100 mL). The two phases were separated, and the organic
phase was washed with saturated NaHCO3 (2 ꢁ 100 mL).
The aqueous phase was back extracted with methylene
chloride, then the combined organic layers were dried
(MgSO4), filtered and condensed in vacuo to a yellow oil
(4g). The crude oil was purified by radial chromato-
graphy (6 mm, 20–30% ethyl acetate/hexanes) to give a
pale-yellow oil (1.90 g, 54%). Rf 0.22 (30% ethyl ace-
3-Methylbutanoic acid-3ꢀ-(3,4-dichlorophenyl)-8-oxabi-
cyclo[3.2.1]oct-2ꢁ-ylmethyl ester (10). The alcohol 5
(1.0 g, 3.48 mmol) was dissolved in dry CH2Cl2 (20 mL)
and treated with Et3N (1.0 mL, 7.2 mmol) and i-valeryl
chloride (0.52 mL, 4.29 mmol). The reaction solution
was stirred under N2 at 22 ꢀC for 16 h. It was then dilu-
ted with CH2Cl2 (50 mL) and extracted with saturated
NaHCO3 solution (50 mL) and brine (50 mL). The
organic phase was condensed to a yellow oil, which was
purified by radial chromatography (6 mm plate, 10–
20% ethyl acetate/hexanes) to give a clear, colorless oil
1
tate/hexanes); H NMR d 7.37 (d, 1H), 7.28 (d, 1H),
7.05 (dd, 1H), 4.45 (ddd, 1H, J=2.5, 6.3, 8.8 Hz), 4.25
(d, 1H, J=7.43 Hz), 3.95 (d, 2H, J=6.0 Hz), 2.94, (ddd,
1H, J=3.0, 6.9, 9.6 Hz), 2.75 (dd, 1H, J=4.7, 4.1 Hz),
2.59 (dt, 1H, J=6.9, 10.6 Hz), 2.49 (dd, 1H, J=4.7,
5.0 Hz), 2.39–2.30 (m, 3H), 2.20–2.08 (m, 1H), 2.00–1.75
(m, 3H), 1.76–1.60 (m, 3H), 1.29 (ddd, 1H, J=2.5, 11.3,
13.8 Hz); 13C NMR d 172.8, 144.2, 132.4, 130.4, 130.4,
129.7, 127.3, 74.2, 71.8, 66.5, 51.1, 47.9, 46.9, 38.1, 36.0,
32.0, 30.9, 30.2, 27.4. Anal. (C19H22O4Cl2) C, H.
1
(0.98 g, 76%). Rf 0.40 (20% ethyl acetate/hexanes); H
NMR d 7.37 (d, 1H), 7.29 (d, 1H), 7.06 (dd, 1H), 4.44
(ddd, 1H, J=2.5, 6.3, 8.8 Hz), 4.27 (d, 1H, J=7.7 Hz),
3.92 (d, 2H, J=6.1 Hz), 2.62 (dt, 1H, J=6.9, 11.0 Hz),
2.40–2.30 (m, 1H), 2.20–1.80 (m, 4H), 1.80–1.75 (m,
1H), 1.75–1.58 (m, 2H), 1.3–1.20 (m, 1H), 0.91 (d, 6H,
J=6.3); 13C NMR d 173.0, 144.1, 132.3, 130.3, 130.2,
129.7, 127.3, 74.2, 71.7, 66.0, 48.0, 43.2, 38.1, 35.9, 31.9,
30.9, 25.5, 22.3. Anal. (C19H24O3Cl2) C, H, Cl.
3ꢀ-(3,4-Dichlorophenyl)-2ꢁ-pent-4-enyloxymethyl-8-oxa-
bicyclo[3.2.1]octane (8). The alcohol
5
(200 mg,
5-Ethylsulfanyl-pentanoic acid 3-(3,4-dichlorophenyl)-8-
oxabicyclo[3.2.1]oct-2-ylmethyl ester (11). Compound 6
(100 mg, 0.26 mmol, 1.0 equiv) was mixed with TsOH
(110 mg, 0.58 mmol, 2.2 equiv) in THF (4 mL). EtSH
(1.2 mL, excess) was added. The resulting mixture was
stirred at 22 ꢀC for 75 h. Volatiles were removed under a
0.69 mmol) and 5-bromo-1-pentene (114 mg. 0.76 mmol)
were mixed in anhydrous THF (20 mL) at room tem-
perature under N2. Sodium hydride (113 mg, 60% in
mineral oil, 2.82 mmol) was added. The resulting solu-
tion was heated at reflux for 3.5 h. The reaction solution

3590
P. C. Meltzer et al. / Bioorg. Med. Chem. 10 (2002) 3583–3591
stream of N2. The residue was dissolved in CH2Cl2
(5 mL) and washed with dilute NaOH (10 mL). The
CH2Cl2 layer was separated and the aqueous layer
extracted twice with CH2Cl2 (5 mL each). The CH2Cl2
solutions were dried over Na2SO4, combined and eva-
EtOH that was added to the drug solution was set aside
to add to the plates. The 24-well plates were removed
from the incubator and 200 mL of novel drug or control
buffer was added to each well to a total volume of 1 mL.
Plates were then placed in the incubator and incubated
for appropriate times periods (1 h, 24 h). At the end of
the preincubation, cell plates were removed from the
incubator and media aspirated by suction. Each well
was rinsed 5 times and a final rinse with buffer (pH 7.4
at 25 ꢀC) was made for 1–2 min in preparation for the
assay.
1
porated. The crude product was analyzed by H NMR
which showed 52% of 6 was consumed. Column
chromatographic separation of the crude product gave
three major compounds: 6 (recovered starting material,
25 mg, 25%), 11 (addition product, 14 mg, 12%) and 5
(11 mg, 15%). 11: 1H NMR d 7.36 (d, 1H), 7.28 (d, 1H),
7.04 (dd, 1H), 4.48–4.42 (m, 1H), 4.25 (d, 1H), 3.94 (d,
2H), 2.61–2.55 (m, 1H), 2.51 (t, 4H), 2.40–2.30 (m, 1H),
2.23 (t, 2H), 2.17–2.06 (m, 1H), 2.01–1.92 (m, 1H), 1.87–
1.78 (m, 1H), 1.74–1.52 (m, 5H), 1.34–1.25 (m, 2H), 1.25
(t, 3H); MS (CI, m/z): 431 [MÀH]À.
Cocaine binding assay. Serial dilutions of (À)-cocaine,
mazindol (10 mM), [3H]cocaine (20 nM) were made. The
buffer was aspirated, rinsed from the plate and
[3H]cocaine (200 mL) was added to each well, followed
by (À)-cocaine (200 mL) of the diluted stock, to a final
total volume of 600 mL. Plates were incubated for 1 h,
the drug solution was aspirated, each well was rinsed
once with 1 mL ice-cold cell buffer (pH 7.4 at 4 ꢀC) and
SDS was added to each well. After removing the SDS
scintillation fluid was added and radioactivity was mea-
sured by liquid scintillation spectrometry.
5-(2-tert)-Butoxycarbonylamino-2-ethoxycarbonlyethyl-
sulfanyl)-pentanoic acid 3-(3,4-dichlorophenyl)-8-oxabi-
cyclo[3.2.1]oct-2-ylmethyl ester (13). Compound
6
(56 mg, 0.15 mmol, 1.0 equiv), 12 (73 mg, 0.30 mmol,
2.0 equiv) and AIBN (4.0 mg, 0.024 mmol, 0.16 equiv)
were mixed in a round bottom flask under N2 for
10 min. CH3CN (2.0 mL), which had been degassed with
N2 for 10 min, was added. The resulting solution was
stirred and bubbled with N2 for 10 min and then heated
to 50–65 ꢀC for 4 h. CH3CN was removed and the crude
Dopamine uptake assay. Dopamine and mazindol were
prepared in assay buffer (pH 7.4 at 37 ꢀC). Serial dilu-
tions of dopamine combined with 20 nM [3H]dopamine,
and mazindol were prepared to measure non-specific
binding. Room temperature assay buffer was added to
each well and buffer and mazindol were added to
measure non-specific binding. In the dark, serially dilu-
ted [3H]dopamine was added, incubation proceeded for
10 min, the incubation medium was aspirated and each
well rinsed with assay buffer. SDS was added to each
well following aspiration of last rinse, removed, scintil-
lation fluid added and radioactivity measured with
liquid scintillation spectrometry.
1
product was analyzed by H NMR, which showed 60%
of 6 was consumed. Column chromatographic separa-
1
tion of this crude product gave 47 mg (50%) of 13. H
NMR d 7.36 (d, 1H), 7.28 (d, 1H), 7.04 (dd, 1H), 5.35
(d, 1H), 4.52–4.42 (m, 2H), 4.21 (q, 3H), 3.94 (d, 2H),
2.95 (d, 2H), 2.61–2.55 (m, 1H), 2.51 (t, 2H), 2.40–2.30 (m,
1H), 2.23 (t, 2H), 2.17–2.06 (m, 1H), 2.01–1.92 (m, 1H),
1.87–1.78 (m, 1H), 1.74–1.52 (m, 5H), 1.52 (s, 9H), 1.34–
1.25 (m, 2H), 1.25 (t, 3H); MS (CI, m/z): 616 [MÀH]À.
Cell plating procedure. HEK-293 cells stably transfected
with the hDAT plasmid were grown at 5% CO2 in a
37 ꢀC water-jacketed incubator until ready to plate.
Media of dishes with 90% confluency was aspirated off
and cells were rinsed once with 6 mL cold PBS (4 ꢀC, pH
7.4). The rinse was aspirated and cells were harvested in
10 mL PBS and centrifuged for 5 min at 2000 rpm. The
cells were then resuspended in an appropriate amount
of media and the suspension was stirred. Tissue culture
plates coated with poly-d-lysine were prepared by add-
ing 500 mL of media to each well and then 300 mL of cell
suspension. Each well contained 800 mL of media and
cells. Each plate was shaken vigorously to distribute cells
evenly in wells and care was taken to keep the media in
the wells. Cells were grown overnight in the incubator
and were about 40–50% confluent the next day.
Acknowledgement
The authors would like to thank the National Institute
on Drug Abuse for financial support (DA11542, DA7–
8081 DA11558, DA06303, DA00304 and RR00168).
References and Notes
1. Schoemaker, H.; Pimoule, C.; Arbilla, S.; Scatton, B.;
Javoy-Agid, F.; Langer, S. Z. Naunyn-Schmiedeberg’s Arch.
Pharmacol. 1985, 329, 227.
2. Kennedy, L. T.; Hanbauer, I. J. Neurochem. 1983, 34, 1137.
3. Reith, M. E. A.; Meisler, B. E.; Sershen, H.; Lajtha, A.
Biochem. Pharmacol. 1986, 35, 1123.
4. Madras, B. K.; Fahey, M. A.; Bergman, J.; Canfield, D. R.;
Spealman, R. D. J. Pharmacol. Exp. Ther. 1989, 251, 131.
5. Canfield, D. R.; Spealman, R. D.; Kaufman, M. J.;
Madras, B. K. Synapse 1990, 6, 189.
6. Kaufman, M. J.; Spealman, R. D.; Madras, B. K. Synapse
1991, 9, 177.
7. Ritz, M. C.; Lamb, R. J.; Goldberg, S. R.; Kuhar, M. J.
Science 1987, 237, 1219.
8. Kuhar, M. J.; Ritz, M. C.; Boja, J. W. Trends Neurosci.
1991, 14, 299.
9. Newman, A. Med. Chem. Res. 1998, 8, 1.
Pre-incubation and rinse procedure. Novel compounds
were dissolved, if necessary in an appropriate con-
centration of EtOH and then at a concentration of
1 mM in 37 ꢀC assay buffer consisting of Tris (5 mM),
Hepes (8.5 mM), NaCl (120 mM), KCl (5.4 mM), CaCl2
(1.2 mM), MgSO4 (1.2 mM), glucose (5 mM), and Tro-
polone (1 mM) at pH 7.4. The stock solution of 1 mM
was diluted to 5–25 mM concentration. Cell buffer (pH
7.4 at 37 ꢀC) containing the same final percentage of

P. C. Meltzer et al. / Bioorg. Med. Chem. 10 (2002) 3583–3591
3591
10. Singh, S. Chem. Rev. 2000, 100, 925.
27. Houlihan, W. J.; Boja, J. W.; Kopajtic, T. A.; Kuhar,
M. J.; Degrado, S. J.; Toledo, L. Med. Chem. Res. 1998, 8, 77.
28. Meltzer, P. C.; Liang, A. Y.; Madras, B. K. J. Med. Chem.
1994, 37, 2001.
11. Carroll, F. I.; Mascarella, S. W.; Kuzemko, M. A.; Gao,
Y.; Abraham, P.; Lewin, A. H.; Boja, J. W.; Kuhar, M. J.
J. Med. Chem. 1994, 37, 2865.
12. Xu, L.; Kelkar, S. V.; Lomenzo, S. A.; Izenwasser, S.; Katz,
J. L.; Kline, R. H.; Trudell, M. L. J. Med. Chem. 1997, 40, 858.
13. Meltzer, P. C.; Liang, A. Y.; Blundell, P.; Gonzalez,
M. D.; Chen, Z.; George, C.; Madras, B. K. J. Med. Chem.
1997, 40, 2661.
29. Meltzer, P. C.; Liang, A. Y.; Madras, B. K. J. Med. Chem.
1996, 39, 371.
30. Kitayama, S.; Shimada, S.; Xu, H.; Markham, L.; Dono-
van, D. H.; Uhl, G. R. Proc. Natl. Acad. Sci. U.S.A. 1993, 89,
7782.
14. Carroll, F. I.; Howell, L. L.; Kuhar, M. J. J. Med. Chem.
1999, 42, 2721.
31. 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, 267.
32. Deutsch, H. M.; Collard, D. M.; Zhang, L.; Burnham,
K. S.; Deshpande, A. K.; Holtzman, S. G.; Schewri, M. M.
J. Med. Chem. 1999, 42, 882.
15. Meltzer, P. C.; Blundell, P.; Chen, Z.; Yong, Y.; Madras,
B. K. Bioorg. Med. Chem. Lett. 1999, 9, 857.
16. Meltzer, P. C.; Blundell, P.; Huang, H.; Liu, S.; Yong,
Y. F.; Madras, B. K. Bioorg. Med. Chem. 2000, 8, 581.
17. Newman, A. H.; Robarge, M. J.; Howard, I. M.; Witt-
kopp, S. L.; George, C.; Kopajtic, T.; Izenwasserr, S.; Katz,
J . L.J. Med. Chem. 2001, 44, 633.
18. Meltzer, P. C.; B, W.; Chen, Z.; Blundell, P.; Jayaraman,
M.; Gonzalez, M. D.; George, C.; Madras, B. K. J. Med.
Chem. 2001, 44, 2619.
19. Davies, H. M. L.; Gilliatt, V.; Kuhn, L. A.; Saikali, E.;
Ren, P.; Hammond, P. S.; Sexton, T.; Childers, S. R. J. Med.
Chem. 2001, 44, 1509.
20. van der Zee, P.; Koger, H. S.; Gootjes, J.; Hespe, W. Eur.
J. Med. Chem. 1980, 15, 363.
21. Dutta, A. K.; Meltzer, P. C.; Madras, B. K. Med. Chem.
Res. 1993, 3, 209.
22. Deutsch, H. M.; Shi, Q.; Gruszecka-Kowalik, E.; Schweri,
M. M. J. Med. Chem. 1996, 39, 1201.
23. Deutsch, H. M. Med. Chem. Res. 1998, 8, 91.
24. Carroll, F. I.; Coleman, M. L.; Lewin, A. H. J. Org.
Chem. 1982, 47, 13.
25. Carroll, F. I.; Lewin, A. H.; Abraham, P.; Parham, K.;
Boja, J. W.; Kuhar, M. J. J. Med. Chem. 1991, 34, 883.
26. Kozikowski, A. P.; Roberti, M.; Johnson, K. M.; Berg-
mann, J. S.; Ball, R. G. Bioorg. Med. Chem. Lett. 1993, 3,
1327.
33. Kilty, J. E.; Lorang, D.; Amara, S. G. Science 1991, 254,
578.
34. Shimada, S.; Kitayama, S.; Lin, C.-L.; Patel, A.; Nanthaku-
mar, E.; Gregor, P.; Kuhar, M.; Uhl, G. Science 1991, 254, 576.
35. Giros, B.; el Mestikawy, S.; Godinot, N.; Zheng, K.; Han,
H.; Yang-Feng, T.; Caron, M. G. Mol. Pharmacol. 1992, 42,
383.
36. Pristupa, Z. B.; Wilson, J. M.; Hoffman, B. J.; Kish, S. J.;
Niznik, H. B. Mol. Pharmacol. 1994, 45, 125.
37. Cao, C. J.; Young, M. M.; Wong, J. B.; Mahran, L. G.;
Eldefrawi, M. E. Membr. Biochem. 1989, 8, 207.
38. Schweri, M. M. Neuropharmacology 1990, 29, 901.
39. Xu, C.; Coffey, L. L.; Reith, M. E. A. Naunyn-Schmiede-
berg’s Arch. Pharmacol. 1997, 355, 64.
40. Meltzer, P. C.; Blundell, P.; Madras, B. K. Med. Chem.
Res. 1998, 8, 12.
41. Rothman, R. B. Life Sci. 1990, 46, 17.
42. Rothman, R. B.; Mele, A.; Reid, A. A.; Akunne, H. C.;
Greig, N.; Thurkauf, A.; deCosta, B. R.; Rice, K. C.; Pert, A.
Pharmacol. Biochem. Behav. 1991, 40, 387.
43. Carroll, F. I.; Kotian, P.; Dehghani, A.; Gray, J. L.;
Kuzemko, M. A.; Parham, K. A.; Abraham, P.; Lewin, A. H.;
Boja, J. W.; Kuhar, M. J. J. Med. Chem. 1995, 38, 379.